Journal of the Botanical Research Institute of Texas J. Bot. Res. Inst. Texas ISSN 1934-5259 Fort Worth, Texas 76107-3400, USA 817-332-7432; 817-332-4112 fax Electronic mail: barney@brit.org; jbrit@brit.org History and Dedication 1962— Lloyd H.Shinners (left), a member of the Southern Methodist University (SMU) faculty and a prolific researcher and writer, published the first issues of S/do, Contributions to Botany (now J. Bot. Res. Inst. Texas) 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. 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. Bot. 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. Felix Llamas (Universidad of Le6r Editorial Board Harold Keller (BRIT) Robert J.O'Kennon (BRIT) Richard Rabeler (MICH) Subscription Prices (2013) $48. Personal (Individual/Family) $38. Student (with verification) $115. USA (Institutional) $125. Outside USA (Institutional) Back issues upon request. Email: orders@brit.org www.britpress.org Composition rhorngraphics, Plano, Texas; rlrcf Cover Illustration Electronically tinted botanical ill of Liatris aestivalis originally user BRITs anniversary poster 2001. Summer gayfeather flowers mic Jul-Aug(-Sep) and is endemic t Oklahoma and Texas. Sida 19:768. 2001. Press Table of Contents SYSTEMATICS A new Sisyrinchium (Iridaceae) from cedar glades in northern Alabama Bruce A. Sorrie, Wesley M. Knapp, L. Dwayne Estes, and Daniel D. Spaulding Polemonium elusum (Polemoniaceae), a new species from east central Ida Joshua J. Irwin, Rebecca Stubbs, and Ronald L. Hartman M ' S ind James L. Reveal Systematics of Lindleya (Rosaceae: Maloideae) James Henrickson Illustrations and studies in neotropical Orchidaceae. 5. The I (Pleurothallidinae) with three new species from Costa Rica Diego BogarIn, Christina M. Smith, and Daniel JimEnez SSOURI BOTANICAL mi DFK 0 4 7ni2 “9 GARDEN LIBRARY 341 Helen Kennedy of the Colombian A a new species of Gesneriaceae from the Cordillera Central [ohn L. Clark and Laura Clavijo \ new variety of Dedieuxia cacui Joseph H. Kirkbride, Jr. and Piero G. Delprete A previously unrecognized species of Senegalia (Fabaceae) from northeastern Brazil David S. Seigler, John E. Ebinger, and Petala Gomes Ribeiro 391 397 403 Cytotypic variation in Phlox pilosa ssp. pilosa (Polemoniaceae) at the western edge of its range in the central United States Lindsey Worcester, Mark H. Mayfield, and Carolyn J. Ferguson A new combination and a new species in Combretum (Combretaceae) from India K.A. Sujana, M.K. Ratheesh Narayanan, and N. Anil Kumar i, Mexico. I. Convolvulus, Cressa 443 453 Richard S. Felger, Daniel F. Austin, Thomas R. Van Devender, J. JesOs Sanchez-Escalante, and Mihai Costea id J. Jesus SAnchez-Escalante ~. Austin, Thomas R. Van Devender, PALEOBOTANY 459 529 551 557 A new fossil species of Colpothrinax (Arecaceae) from Mid-Tertiary Mexican amber Kenton L. Chambers, George 0. Poinar, Jr., and Alex E. Brown Additional fossils in Dominican amber give evidence of anther abortion in Mid-Tertiary Trichilia (Meliaceae) Kenton L. Chambers and George O. Poinar ,Jr. DEVELOPMENT AND STRUCTURE Limitations to natural production of Lophophora williamsii (Cactaceae) 11. Effects of repeated harvesting at two-year intervals in a South Texas population Martin Terry, Keeper Trout, Bennie Williams, Teodoso Herrera, and Norma Fowler HERBARIUM METHODS AND TECHNIQUES A cost-effective method for constructing magnetic fumicels for herbarium cabinets FLORISTICS, ECOLOGY, AND CONSERVATION Nuevos registros de Poaceas para el norte de Mexico Y. Herrera Arrieta, C.A. Silva Salas, L. Ruacho Gonzalez y O. Rosales Carrillo Distribution actual y potencial de Taxus globosa (Taxaceae) en Mexico Mario A. GarcIa-Aranda, CEsar CantO-Ayala, Eduardo Estrada-CastillOn, Marisela Pando-Moreno y Antonio Moreno-Talamantes and A. nigrum (Viscaceae) in Mexico Robert L. Mathiasen, Shawn C. Kenaley, and Brian P. Reif Taxonomic history, rediscovery, and assessment of threat status of Streblus ilicifolius (Moraceae) from India Bikarma Singh, Arun Chettri, Dibyendu Adhikari, and Saroj K. Barik Three noteworthy additions to the Alabama flora Alvin R. Diamond and Brian R. Keener A first spontaneous record of Actinidia chinensis var. deliciosa (Actinidiaceae) in the United States flora Brett E. Serves, David H. Mason, and Troy L. Bray Pallenis maritima (Asteraceae) of salt-tolerant ornamental plants Richard E. Riefner, Jr. and Werner Greuter Vascular flora and plant communities of Dead Horse Knob (Rucker’s Knob), Madison County, Kentucky Ralph L. Thompson, Derick B. Poindexter, and J. Richard Abbott Vascular plants of the Ya 2 in Grenada County, Mississippi Michael Wayne Morre and John R. MacDonald ity, Pennsylvania New Pennsylvania county occurrences for Beilis perennis. Hibiscus moscl Lamiutn maculatum, and Robinia hispida Jerky G. Chmielewski and David Krayesky 561 567 579 583 599 615 617 621 653 681 707 A NEW SISYRINCHIUM (IRIDACEAE) FROM CEDAR GLADES IN NORTHERN ALABAMA Bruce A. Sorrie NC Natural Heritage Program and University of North Carolina Herbarium North Carolina Botanical Garden Chapel Hill, North Carolina 27599-3280, U.S. A. nruce.sorrie@ncdenr.gov L. Dwayne Estes Department of Biology & Center for Field Biology Austin Peay State University Clarksville, Tennessee 37044, USA. Wesley M. Knapp Maryland Department of Natural Resources Wildlife and Heritage Service P.O. Box 68, Wye Mills, Maryland 21679, U.S.A wknapp@dnr.state.md.us Daniel D. Spaulding Anniston Museum of Natural History P.O. Box 1587 Anniston, Alabama 36202, U.S.A. RESUMEN Sisyrinchium L. is a taxonomically difficult genus with 37 species in North America north of Mexico. Of these, only Sisyrinchium albidum Raf. and S. capillare Bicknell have paired inflorescences (Cholewa & Henderson 2002). Each of the inflorescences is subtended by two bracts or spathes, which terminate the unbranched stems. Although this combination of characters is rare in North American Sisyrinchium, a third entity shares these character states and is the subject of this paper. We propose Sisyrinchium calciphilum Sorrie for this spe- cies, which is endemic to limestone glades in northwestern Alabama. Plants: perennial, cespitose, 20-42 cm tall. Stems simple, clearly but narrowly winged, 0.7-1.3(-1.5) mm wide, each wing wider than stem core, glabrous, margins entire, stem base purple to pinkish brown. Leaves: gla- brous, bases not persistent as fibrous tufts. Inflorescences: paired one sided cymes, i.e., rhiphidia, each sub- tended by a pair of spathes, these subtended by a bractlike leaf 41-82 mm long that often obscures inner inflo- rescence; bractlike leaf and spathes purple tinged, spiculate, spathe keels denticulate, outer spathe of outer pair 12-18 mm, spathes of outer inflorescences averaging 4.4 mm longer than spathes of inner inflorescences, spathe margins hyaline, translucent to purple. Flowers: tepals blue with yellow bases, 7-9 mm long, apex weakly emarginate, aristate; filaments connate most of length, egandular; ovary green and glandular-hairy. Capsules: fully mature capsules not seen; immature capsules more-or-less globose, 4-5 mm long and wide, pale green, pedicels spreading to ascending, glabrous or with very sparse glandular hairs. Seeds: mature seeds 324 Journal of the Botanical Research Institute of Texas 6(2) Sorrie et al., A new Sisyrinchium from Alabama 325 27 Apr 1993, R. Whetstone, D. Spaulding, J. Ballard, & T. Dobson 16436 (JSU); N of Courtland, just S of Wheeler Station Sporting Clays, recent- ' 4 May 1996, R. Whetstone &S.Hruska 17457 (JSU); Prairie Grove Glades Preserve, abundant on thin clay e, 19 Apr 2003, R. Krai 93898 (AMAL, APSC); 2 mi SE of Speake on Alabama 157, clearing in limestone woods, 1968, R. Krai 30489 (GA); 1.5 mi SW of Flat Rock, limestone cedar glade, tepals dark purple, 17 Apr 2006, B. Keener & D. Spaulding 2817 (AMAL, UNA); Bankhead National Forest, off FSR-264, limestone woods, 17 Apr 2005, D. Spaulding & B. Keener 12252 (AUA.UNA). CR-55 near Emmanuel Road, disturbed limestone glade, 13 Apr 2003, D. Spaulding 11723 «d E of Massey, 16 Apr 2011, B. Sorrie, D. Estes, & W. Knapp 12746 (DUKE, GA, NCSC, n slope in dryish woods, 16 Apr 2011, B. Sorrie, D. Estes, & W. Knapp 12747 (NCU, NY). Sisyrinchium albidum was the first of the blue-eyed-grasses with paired inflorescences to be named (Rafinesque 1832). Much later, Bicknell split out S. capillare with its wiry, barely winged stems (Bicknell 1899). Since there appeared to be overlapping characters between these two species, some authors declined to recognize S. capil- lare (e.g., Radford et al. 1968) or confounded their habitats and ranges (Cholewa & Henderson 2002). Recent herbarium and field work by the authors show that S. capillare occurs strictly in the Atlantic Coastal Plain Physiographic Province whereas S. albidum occurs primarily in the Piedmont, Interior Low Plateau, and Inte- rior Highlands Physiographic Provinces, northward to the Great Lakes states; and with populations southward to the Gulf Coastal Plain (Fig. 6). In addition, there are strong morphological differences, notably the slender and nearly wingless stem of S. capillare (vs. clearly winged in S. albidum), stem with dense fibrous remains of leaves (vs glabrous to only moderately dense), nearly equal spathes (vs. very unequal), and short outermost spathe (vs. long) (Table 1). Note that Table 1 divides S. albidum into two groups, east and west of the Appala- chian Mountains. These groups appear to differ from each other, notably in stem width and the difference be- tween inner and outer spathe length, but there is much overlap. At this time we are reluctant to recognize any infraspecific taxa and treat S. albidum as a widespread, variable species. Specimens of S. calciphilum are quite different from S. capillare, where they had been placed due to their slender stems. Stems of S. calciphilum are definitely (albeit narrowly) winged, the fibrous remains of leaf bases are absent, the two outer spathes exceed the two inner by a mean of 4.4 mm, spathes and leaflike bracts are strongly purple tinged and spiculate. Corolla color of S. calciphilum is medium blue, unlike the pale blue to whitish color of S. capillare (Figs. 2 & 3). Moreover, S. calciphilum is restricted to high pH limestone glades, whereas S. capillare to acidic, fire-maintained longleaf pine savannas (Table 1). calciphilum is less strikingly distinct. While a single morphological charac- em populations of S. albidum, the suite of characters distinguishes 5. calci- philum (Table 1). Critical differences are: stem width of S. calciphilum is similar to that of many plants from east of the Appalachians, but only half as wide as sympatric plants from west of the Appalachians. Stem margins of S. calciphilum are smooth, but denticulate (often strongly so) in S. albidum. The difference between the two in- um blue, unlike the pale blue to whitish color of S. albidum (Figs. 2 & 4). Coupled with the restricted range and habitat preference, these morphological differences are significant at the species level. DISTRIBUTION AND HABITAT : counties in northwestern Alabama (Fig. 5), a region underlair .. Heller) Bameby, Forestiera ligustri Journal of the Botanical Research Institute of Texas 6(2) j Sisyrinchium from Alabama 327 Journal of the Botanical Research Institute of Texas 6(2) Table 1. Co U 9 2-)1.5-2.5(-2.8) 330 Journal of the Botanical Research Institute of Texas 6 ( 2 ) BOOK REVIEW Noel H. Holmgren, Patricia K. Holmgren, James L. Reveal, and Collaborators. 2012. Intermountain Flora, Vascular Plants of the Intermountain West, U.S.A. Volume Two, Part A: Subclasses Magnoliidae- CaryophylUdae. (ISBN: 978-0-89327-520-4, hbk.). New York Botanical Garden Press, 2900 Southern Blvd., Bronx, New York 10458-5126, U.S.A. (Orders: http://www.nybgpress.org, 718-817-8721, fax 718- 817-8842). $150.00, 731 pp, color frontispiece, grayscale epilogue, line drawings throughout, Th n x Volume 2A, the Intermountain Flora Team brings to a successful close a decades-long, j multigenerational effort to document the plant life of the colder drylands occupying the large region in the western United States more or less bounded by the Sierra Nevada to the west, the Rocky Mountains to the east, j and stretching from central Idaho south to northwestern Arizona. Originally conceived by Bassett Maguire j around 1940, the ensuing research became a driving force in the careers of Arthur Holmgren and Arthur Cron- | quist, as well as the botanists who authored this final volume. Publication of the six volumes in eight big books (most floristicians have difficulty estimating space requirements for their works) took forty years, but time alone is a poor estimate of the scope of the exploration and research that went into the project. The full series j treats 3,847 species in 898 genera. For those who do not own these wonderful books, the publisher is cur- ; rently offering the whole set for $520 (a $640 value). There also are rumors of a planned ninth volume with a Volume 2A of the work, which includes the beginning of the dicots in the Cronquistian classification -j system, treats 147 genera, 611 species, and 301 additional infrataxa in 31 families, including such regionally j diverse important groups as the Papaveraceae, Ranunculaceae, Polygonaceae, Caryophyllaceae, Chenopodia- ceae, Amaranthaceae, Nyctaginaceae, Montiaceae, and Cactaceae. The presentation is identical to that in ear- lier volumes, with indented keys and detailed descriptions followed by range/ecology statements and very j useful critical notes on taxonomic and other issues (along with copious literature citations). The treatments j also include extensive synonyms, with complete citations of types. The numerous plates of line drawings (by ; several artists), which cover varying numbers of taxa per genus, are uniformly excellent in composition and ; detail. The drawings of Cactaceae are particularly beautiful. A useful addendum summaries the 6 new combi- J nations, 34 new typifications, and 1 new cytological report included in the volume. Throughout, the contents are encyclopedic and the treatment of any family or genus provides a marvelous introduction to the taxonomy, nomenclature, ecology, and uses for that group. There is little to criticize in this enormous work. Perhaps some of the more recent volumes are a bit pricey, but sadly that has become the nature of the printed page. The lack of a comprehensive key to dicot families (such keys exist for the remaining major plant groups in other volumes) might be addressed in the proposed follow-up volume, as might a more detailed taxonomic summary of the plants in the series or a discussion of plants of conservation concern in the region (a 70-year study surely can offer a unique perspective on regional patterns of plant endangerment). However, such potential additions in no way diminish the present utility of the volumes. Between the burnt-orange covers of these volumes lies a treasure-trove of information and in- sights on the vascular plants unequaled for most other regions of the country. The authors, past and present, are to be congratulated on their persistence in seeing this project through to successful completion . — George Yatskievych, Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166-0299, USA. POLEMONIUM ELUSUM (POLEMONIACEAE), A NEW SPECIES FROM EAST CENTRAL IDAHO, U.S.A. Ronald L. Hartman mm Perennial herbs, caespitose, glabrous to pubescent with trichomes sparse to moderately dense, minute, flaccid to erect, often glandular trichomes, not odiferous; branches subterranean, 3-15. Stems 20-50 cm tall. Leaves mostly basal and the bases sheathing to scattered along stem, 5-20, 10-30 cm long proximally to 2 cm distally, 1.5-3(-4) wide; petioles 2— 7(— 11) cm long; pinnae 12-24, mostly 2-parted, alternate or paired and thus appear- ing whorled, narrowly to broadly elliptic or spatulate, occasionally obovate, 4-28 mm long, 1-3.5 mm wide. Inflorescences a thyrse, often diffuse; branches 2-5, 5-11 cm long, strict; flowers 15-60, congested distally; pedicels 5-6 mm long. Flowers 5-merous; perianth regular; calyx fused at base, campanulate; lobes lanceo- late, green to purplish in part, 4.4-8 mm long, diverging at a 35-40 degree angle; connective membrane in bud, flower, and fruits translucent, whitish, oblong to spatulate, 0.4-0.7 mm wide; sinuses rounded; corolla Ught blue to white, campanulate, 10-13 mm long, tube 4-5.2 mm long; lobes 5-7.8 mm long, spatulate to obovate; apex rounded, minutely apiculate; stamens usually exserted; filaments attached 2-3 mm above base, at which point minutely and densely tomentose for 0.8-1 mm; anthers basifixed, narrowly to broadly sagittate, broadly elliptic to orbicular following dehiscence, 1.4-1.8 mm long; ovary greenish, broadly ovoid, 2 mm long, with 5 raised nerves at least in part; placentation axile, chambers 3; styles 6.5-8 mm long; branches flattened, 2.5-3.2 mm long; stigma glandular over adaxial surface. Fruit globose, tricarpellate, 4.5-5 mm long, split for 60-70 percent of length. Seeds brown, trigonous, the abaxial surface rounded, the edges acute, 24—2.8 mm long; Etymology.— The epithet was chosen as this species has eluded discovery for well over a century, this despite sporadic botanical forays to the region. Distribution and ecology. — Polemonium elusum is known from the mountains and foothills surrounding the town of Ellis, in east central Idaho (Fig. 2). Populations occur from 1440 to 2560 m in elevation, where vegetation transitions from sagebrush and mountain mahogany to Douglas-fir woodland, depending on slope and aspect. Within this range of vegetation, it was found in a variety of microhabitats, thus the species appears to be an ecological generalist. Microhabitats in which plants are found include the margins of talus fields, dry Douglas-fir woodland, outer riparian margins, and shaded rock outcrops. Plants occur in stable but loose, coarse textured colluvial soils. All populations were found on various geologic units of the Challis volcanics group (Zientek et al. 2005) and were most abundant from lower Cow Creek southward across the adjacent un- The initial discovery was made by potential habitat for Cryptantha salmon (Rydb.) Blank, var. salmonensis P £ found growing under more mesic conditi 1-June 2010. This locality was surveyed as Payson, Astragalus amblytropis Barneby, and Oxytropis besseyi demies was encountered, but Polemonium elusum was Id be expected for the aforementioned taxa. Further ivn that populations of P. pulcherrimum var. pulcherrimum surround the distribution of P. cies have several similarities, including overlapping elevation ranges, dry habitats, and a e of the corolla and the inflorescence. Additionally, P viscosum has been observed as low as p canyons above the Salmon River to the north and south of P. elusum. However, these con- Journal of the Botanical Research Institute of Texas 6(2) Phenology . — Flowering begins in late-May at more xeric, low elevation sites, continuing through mid-July at higher, cooler sites; fruiting from mid-June through August. Conservation .— Perhaps the greatest threat to the existence of Polemonium elusion is the small total popu- >tal about 1,300 plants, occupying eight km . Other probable threats include herbivory by ungulates and habitat invasion by exotic weeds. Interestingly, the population at Gerry Gulch occurs at a location where the vegetation is significantly impacted by cattle. Her- bivory by white-tailed deer has been shown to be a significant factor for population survival in Polemonium vanbruntiae Britton (Bermingham 2010). If herbivory by cattle and wildlife affects the survival of P. elusum, then the maintenance of ungulate populations below certain levels may be required. Other aspects of the biology of this taxon are unknown. However, Polemonium vanbruntiae is self compat- i species of Polemo ible (Hill et al. 2008). Because reproductive and life history traits influence the survivorship of plant popula- tions, further research on P. elusum is necessary in order to develop a more effective conservation strategy. Molecular analysis. — Sequence data from 28 taxa were analyzed. The 46 vouchers used in this phyloge- netic study are listed in Appendix 1. Twenty-six of these were accessions from GenBank. Leptosiphon, Linanthus, and Phlox were used as outgroups because they have been proposed as genera closely related to Polemonium (Porter 1997; Johnson et al. 2008; Hankamp 2011). Field collections of fresh plant material were preserved in silica gel and voucher specimens are deposited in the herbarium at San Francisco State University (Table 1). DNA extracted from herbarium sheets follow the same process. Extraction protocols are described in DNeasy (2004). The 1TS1, 5.8S, and ITS2 regions were amplified using primers 1TS4 and 1TSLEU (White et al. 1990). Both regions were duplicated using standard Polymerase Chain Reaction (PCR) in 25 pi volume reactions. Excess nucleotides and primers remaining in the samples from the PCR were removed using ExoSAP-it kit. An Applied Biosystems 3100 Genetic Analyzer was used for the capillary electrophoresis of all samples. Fragments were sequenced using BigDye following the manufacturer’s protocols. Precipitation was with EDTA/Ethanol/Sodium acetate, and the remaining cycle sequence products were resuspended in Hi-Di before being denatured. Fragments were visualized using an Applied Biosystems PRISM 3100 Genetic Analyzer. Run modules were conducted using liquid polymer POP-6 or POP-7. Sequence files were base-called using Sequenc- ing Analysis 5.1 (Applied Biosystems 2003), and the forward and reverse reads were formed into a consensus sequence using Sequencher 4.8 (Gene Codes Corporation 2007). The consensus sequence contig was loaded into MacClade 4.08X (Maddison & Maddison 2005) to visually confirm the coherence of the bases, and into ClustalX Version 2 (Larkin etal. 2007) for a complete alignment. The nexus file was analyzed in PAUP4.0al 12 (Swofford 2002), Mr. Bayes 3.1.2 (Ronquist & Huelsenbeck 2003), and GARLI 0.951-GUI (Zwickl 2006). This ITS data set included 46 samples with a total of 706 characters, 163 of these were variable. Of the variable characters, 61 were parsimony informative. Bootstrap analyses were performed in PAUP 4.0all2 (Swofford 2002) using a heurist search with 100 repetitions and number of trees increasing by 100. Trees were also ana- lyzed using maximum likelihood and Bayesian analysis. For statistical selection of the best fit model, jModelTest (Posada 2008) was used. The -InL using jModelT- est was 2412.83. The results indicated GTR+I+G, and these parameters were employed. Bootstrapping was tested with GARLI and Bayesian analysis provided numbers for branch support. DISCUSSION Molecular analysis of the ITS region of Polemonium weakly suggests P. elusum is sister to R mexicanum, P. pauci- florum, P. grandiflorum, and P cameum (Fig. 3). This is the best inference from the maximum likelihood tree but it is not statistically supported by either bootstrap (56%) or Bayesian (0.84) values. A few conclusions can be drawn from these results. First, the four species that are most closely related to P. elusum are all included in Grant’s (1959) sect. Polemonium and P. elusum also fits into this section. Worley et al. (2009) further divided Grant’s sect. Polemonium into three species complexes, and except for P. mexicanum, the other three species are all included in the P. pauciflorum complex. This species complex includes plants that are erect to decumbent. The inflorescences are panicle-like cymes or the flowers are arranged in groups of one to three. Polemonium elusum meets these criteria. Second, under Wherry’s (1942) classification, P. pauciflorum , P. grandiflorum, and P cameum all fall under the large flowered section Eupolemonium. Once more, P. elusui These species all have overlapping elevation ranges and low ir North America. Although there is not much statistical support for this grouping, morphometric features help validate the placement of Polemonium elusum with this group. Polemonium elusum is congruent with the rest of the genus in its campanulate corolla, habit, compound leaves, stamen attachment, pubescent filaments, and brown seeds. This species notably differs in the calyx The leaf- Journal of the Botanical Research Institute of Te ACKNOWLEDGMENTS We would like to thank Teresa Prendusi and Bruce Smith for their support of the floristic inventory, thus mak- ing this discovery possible. Lucinda Haggas coordinated our logistical needs during summer field work. Bob Patterson and Dieter Wilken provided helpful suggestions regarding the manuscript. Wendy Irwin graciously provided Figure 1. Applied Biosystems. 2003. Sequencing Analysis* version 5.1 . httpy/www.appliedbiosystems.com. Bermingham, L.H. 2010. Deer herbivory and habitat type influence long-term population dynamics of a rare wetland plant. PI. Ecol. 210:359-378. DNeasy, Q. 2004. Plant DNA extraction handbook, http://www.qiagen.com. Gene Codes Corporation. 2007. Sequencher* version 4.8 sequence analysis software. Ann Arbor, Ml. http://www.gene- codes.com. Grant, V. 1959. Natural history of the Phlox family, vol. 1, systematic botany. The Hague: Martinus Nijhoff. Hankamp, P.Z. 201 1 . Molecular systematics of Leptosiphon (Polemoniaceae). M.S. thesis. San Francisco State University, CA. Hartman, R.L. 1992. The Rocky Mountain Herbarium, associated floristics inventory, and the Flora of the Rocky Moun- tains project. J. Idaho Acad. Sci. 28:22-43. Hartman, R.L. and B.E. Nelson. 201 1 . General information for floristics proposals. [The Boiler Plate]. http:/www.rmh.uwyo. edu. Hartman, R.H., B.E. Nelson, and B. Legler. 2009. Rocky Mountain H< Hill, LM., A.K. Brody, and C.L Tedesco. 2008. Mating strategies < Polemonium vanbruntiae. Acta Oecol. 33:314-323 Johnson, LA., L.M. Chan, T.L. Weese, L.D. Busby, and S. McMurry. 2 strong inference of higher phylogenetic relationships in th Kesonie, D. (Scott) and R.L. Hartman. 201 1. A floristic inventory of Grand Teton National Park, Pinyon Peak Highlands, and vicinity, Wyoming, U.S.A. J. Bot. Res. Inst. Texas 5:357-388. Kuhn, B, B.E. Nelson, and R.L. Hartman. 2011. A floristic inventory of the Cimarron National Grassland (Kansas) and the Comanche National Grassland (Colorado). J. Bot. Res. Inst. Texas 5:753-772. Larkin, M A, G. Blackshields, N.P Brown, R. Chenna, P.A. McGettigan, H. R Thompson, TJ. Gibson, and D.G. Higgins. 2007. Clustal W and Clustal X version 2.0. B Maddison, D.R. and W.P. Maddison. 2005. MacClade 4: Analysis of phylogeny and char Sinauer Associates, Sunderland, MA. :r DNA sequences. Aliso Posada, D. 2008. jModelTest: phylogenetic model averaging. Molec. Biol. Evol. 25:1253-1256. Reie, B, J. Larson, B. Jacobs, B.E. Nelson, and R.L. Hartman. 2009. Floristic studies in north central New Mexico, U.SATheTusas Mountains and the Jemez Mountains. J. Bot Res. Inst. Texas 3:921-961 . RONQU.ST, F. AND J.P. Huelsenbeck. 2003. MRBAYES 3: Bayesian phylogenetic inference under mixed models 19:1572-1574. Swofpord, D.L. 2002. PAUP*. Phylogenetic analysis using parsimony (*and Other Methods). Version 4. Sin; d direct sequencing of fungal ribosomal RNA genes White, TJ.,T.D. Bruns, S.B. L for phylogenetics. In: Innis, M A, D.H. Gelfand, JJ. Sininsky, and TJ.White.'eds. PCR ^oio^ols^’guide tom* hods'and applications. Academic Press Inc., San Diego, CA. Pp. 31 5-322. Wherry, E.T. 1942. The genus Polerru Worley, A.C., H. Ghazvini, and D.W. Schemske. 2009. / length polymorphism (AFLP) markers. Syst. Bot. 34:149-161. Zientek, M.L, P.D. Derkey, a i datasets NEOTYPIFICATION OF AMORPHA ROEMERIANA (FABACEAE: AMORPHEAE) Shannon C.K. Straub 1 and James L. Reveal L.H. Bailey Hortorium, Department of Plant Biology, 4 12 Mann Library Building Cornell University, Ithaca, New York 14853, USA. ss463@cornell.edu; jlr326@cornell.edu Se designa un neotipo (WR Carr et cd 27f RESUMEN In the process of preparing a treatment of Amorpha L. for a forthcoming volume of Flora of North America, we attempted to locate original material of Amorpha roemeriana Scheele, a relatively rare species found primarily on the Edwards Plateau of Bandera, Bexar, Blanco, Comal, Gillespie, Hays, Kendall, Kerr, Kinney, Medina, Travis, and Uvalde counties of central Texas (Turner et al. 2003: 306), with disjunct populations in Coahuila, Mexico, that scarcely differ from the Texas populations. The type was collected by Carl Ferdinand von Roemer (1818-1891) “In margine rivulorum prope, Austin” during his 1845-1846 visit to Texas where he concentrated on geology (Simonds 1902). Upon his return to Europe, Roemer wrote several books and articles on his visit parleying these activities eventually into a professorship in geology at the University of Breslau. Roemer’s plant specimens were given to George Heinrich Adolf Scheele (1808-1864) who accounted for several new Texas species (Scheele 1848) gathered by Roemer and, independently, by Ferdinand Jacob Lindheimer (1801-1879), including A. roemeriana which was reported to have been collected, in flower, in April of 1846. Roemer (1849: 429) himself stated that the plant was found “Bei Austin am Bachrande,” and the “Bluthen sehr wohlriechend, violet.” Scheele’s types were housed at Berlin, but as noted by Wilbur (1975), no one has reported, in the literature at least, the existence of any original material of Amorpha roemeriana. We attempted to locate a specimen at the Roemer-und Pelizaeus-Museum in Hildesheim, Germany, which at one time contained specimens gathered by Ferdinand Roemer (the Museum is named for his brother, Hermann Roemer [1816-1894]), only to be in- formed that the plant collection was sent to Berlin in the early 1900s. Nonetheless, there is no record at the Museum that there was a specimen of Amorpha from Austin, Texas, gathered in 1846. The detailed description written by Scheele provides the characters necessary to distinguish Amorpha roemeriana from A.fruticosa, the most widespread species of the genus and the only one co-occurring with A. roemeriana. The inclusion of the swollen mucro, glandular punctuate leaflets, petiolules bearing purplish glands, and glandular vexillum in the description clearly point to A. roemeriana as that name is applied cur- rently (Schneider 1907; Rydberg 1919; Palmer 1931; Wilbur 1975; Isley 1998), and contrast with the long, slen- der mucro, commonly eglandular leaflets, eglandular or inconspicuously glandular petiolules, and generally eglandular vexillum of A. fruticosa. Confirming the assumption that the name as currently applied is correct could be only accomplished by examining type specimen material. Being unable to locate any original mate- rial, we are here designating a neotype to maintain the current concept of A. roemeriana, which we believe to be consistent with the original description. With the able assistance of William R. Carr of the Nature Conser- vancy of Texas and Dr. Thomas Wendt of the University of Texas, to whom we are grateful, a suitable collection was obtained. 340 irpha roemeriana Scheele, Linnaea 21:461. 1848. Type UNITED STATES. Texas. Hays Co.: Ashe juniper woodland on extremely shallow stony clay loam on top of N- to WY-facing bluff of Fredericksburg Limestone ca. 40-50 ft above the S bank of ir mi W of the junction of State Route 150 and Ranch Road 2770 near M e 150 and Ranch Road 3237 at Hays City, at N30°00'37.5", W097°58'01.r Kf.WR. Carr, B.Johnso ACKNOWLEDGMENTS We wish to thank the curator of TEX for loan of material that was essential to completing this work and our reviewers, Billie L. Turner and Stanley L. Welsh. REFERENCES Isley, D. 1 998. Native and naturalized Leguminosae (Fabaceae) of the United States. Monte L. Bean Life Science Museum, Brigham Young University, Provo, Utah. Palmer, E J. 1 931 . Conspectus of the genus Amorpha. J. Arnold Arbor. 1 2:1 57-1 97. Roemer, C.F. von. 1 849. Texas. Mit besonderer rucksicht auf deutsche auswanderung und die physischen verha'ltnisse des landes nach eigener beobachtung geschildert von Ferdinand Roemer. Mit einem naturwissenschaftlichen anhange und einertopographisch-geognostischen karte von Texas. Adolph Marcus, Bonn. Rydberg, P.A. 1919. Amorpha. N. Amer. FI. 24(1):26-34. Scheele, A. 1 848. Beitrage zur Flor von Texas. Linnaea 21 :453-472, 576-602, 747-768. Schneider, C.K. 1 907. Conspectus generis Amorphae. Bot. Gaz. 43:297-307. Simonds, F.W. 1902. Dr. Ferdinand von Roemer, the father of the geology of Texas; his life and work. Amer. Geol. 29: 131-140. Turner, B.L., H. Nichols, G. Denny, and O. Doron. 2003. Atlas of the vascular plants of Texas, vol. 1 : Dicots, vol. 2: Ferns, gym- nosperms, monocots. Sida, Bot. Misc. 24. Wilbur, R.L. 1 975. A revision of the North American i genus Amorpha (Leguminosae-Psoraleae). Rhodora 77:337-409. SYSTEMATICS OF LINDLEYA (ROSACEAE: MALOIDEAE) James Henrickson Plant Resources Center University of Texas, Austin, Texas 78712, U.SA. ABSTRACT Chihuahua, south to central San Luis Potosi and southwestern Tamaulipas; in Hidalgo-Queritaro; and Puebla-Oaxaca with outlying collec- is discussed along with a call for further modifications of the classification of the Rosaceae. RESUMEN INTRODUCTION The monotypic genus Lindleya (Rosaceae) consists of sclerophyllous, evergreen shrubs with conspicuous white flowers and distinctive, woody, 5-carpelled capsular fruit. It is native to the arid and semi-arid chaparral or matorral of Mexico, with populations known from Puebla-Oaxaca, Hidalgo-Queretaro and the mountains of the Sierra Madre Oriental and the Chihuahuan Desert from San Luis Potosf, Tamaulipas, Nuevo Le6n, Coa- huila and eastern parts of Zacatecas, Durango and Chihuahua. The genus has a complex but interesting taxonomic history that is reviewed below. Data on vegetative and reproductive features are presented and illustrated along with a formal taxonomic treatment and distribution maps. The most interesting aspect of Lindleya is its place in the phylogeny of the Rosaceae. With its five-car- peled capsular fruit, its placement within the family has been problematic. MATERIALS AND METHODS The study is based on empirical observations from field-collected and herbarium material as well as extensive field studies. Herbarium specimens were borrowed from A, GH, MO, MICH, NY, TEX-LL and US. Additional data was obtained from collections from ARIZ, ASU, CAS, DS, ENCB, MEXU, RSA, POM during visits to those herbaria. Anatomical studies incorporated standard paraffin techniques (Johansen 1940) for production of serial sections of leaves and flowers. All plant measurements were made from dried material unless otherwise 342 TAXONOMIC HISTORY Taxonomic History.— The genus Lindleya was named by Kunth (1824) in the sixth volume of Humboldt and Bonpland’s Nova Genera et Species Plantarum [5 Jan 1824-pertinent dates from International Plant Name Index (wwwipnixHg) and Taxonomic Literature II (Stafleu & Cowan 1976-1988)] with a complete description. An illustration of the sole species, Lindleya mespiloides Kunth, was published in Apr 1824 (fig. 562). The genus was named for the then young John Lindley (1799-1865), British botanist, student of the Rosaceae and Orchida- ceae, who went on to be professor of botany at the University College London (1829-1860). However, prior to this, the name Lindleya had been used as a generic name several times. The first use of Lindleya as a genus name was by Nees von Essenbeck (1821), in a paper (21 May 1821) determining specimens from the Brazilian collections of Spix and Martinus. His Lindleya was a genus in the Theaceae, which Index Kewensis states is based on a species Lindleyafruticosa Nees ( =Laplacea semiserrata Cambess.). Nees’ Lindleya, however, is a homotypic synonym of a slightly older Wilkstroemia Schrader (5 May 1821) and thus is a superflu- ous name. Kunth in Humboldt, Bonpland loc. cit. (25 Feb. 1822) recognized this genus as Laplacea Kunth. Kunth himself had previously used the generic name Lindleya twice. His first use was on plates of two species that were published in volume five of Nova Genera et Species Plantarum. According to Taxonomic Lit- erature II, the fascicle of plates bearing the name Lindleya (plates 479-480) were published on 25 Feb. 1822, while the text, (vol. 5: 361-367) was not published until 24 March 1823, however, in the text, the species were placed in the genus Casearia (Flacortiaceae). The use of generic name Lindleya on the plates has been consid- ered as a nomen nudum by Index Kewensis and Index Nominum Genericorum database and l.P.N.I. But the International Code of Botanical Nomenclature (McNeil et al. 2006) Art. 42.3 considers an illustration with analysis (i.e., with separate figures showing details to aid with identification) as being acceptable in place of a written description or diagnosis for a species description. But Art. 42.1, 42.2 notes that this can apply only if the genus is monotypic. As there were two species illustrated, the use of Lindleya here can be considered as a nomen Kunth (1822) again u le and Tiliaceae (20 Apr, J m. 46D) between Theo- j broma Linn, and Guazuma Plum., in Sectio [subfamily] II “Buttneriaceae verae” of Buttneriaceae (once Stercu- , liaceae, now Malvaceae s.l.). This is purely a listing of a known or proposed name and again is a nomen inval, j without description given, implied or cited. Rydberg (1908a) in his Notes on Rosaceae preceding his North America Flora treatment (Rydberg 1908b) listed the previous use of Lindleya by Kunth and Nees (as noted in Index Kewensis) and offered a new generic name Lindleyella for the Rosaceaous genus, and described a second species L. s chiedeana Rydb., characterized by more obovate leaf blades and longer, more obovate petals. In 1940 Fedde, for some reason, rejected the ear- lier Lindleyella of Rydberg (1908), in favor of a later described Lindleyella Schlechter (1914), a genus of Orchida- ceae with five species that were named from 1914 to 1924, and substituted a new name, Neolindleyella Fedde for Rydberg’s genus. As to why Fedde would choose a younger name ( Lindleyella of Schlechter 1914) against an older name ( Lindleyella of Rydberg 1908) is unknown as it goes against the rules of priority, which he so strongly espoused. Fedde’s work on botanical nomenclature had become such a disruption on nomenclatural stability that the provision for conservation of generic names was enacted during the Vienna Congress in 1905 (Briquet 1906). In 1930, at the fifth International Botanical Congress (Briquet 1935), the rule concerning homonyms was altered. Prior to that time, a later homonym could be accepted as a valid name if the earlier name had become a synonym and was not being used. The fifth congress maintained strict nomenclatural priority and rejected later homonyms with the understanding that all well known generic homonyms, as far as possible, should be To this end, a systematic search was made to validate later homonyms via generic conservation (Rehder et al. 1935). The task was assigned to various botanists each responsible for genera based on alphabetical group- ings. Genera beginning with the letters L through P were assigned to Dr. Rudolf Mansfield of the Botanical Henrickson, Systematic of Lindleya Museum of Berlin. Through his work, Lindleya Nees was rejected in favor of Laplacea H.B.K. (nom. cons.). Kunth’s use of Lindleya as a replacement name for Casearia was considered a nomen nudum, and Lindleya of Kunth in Humboldt and Bonpland (1824) was conserved against Lindleyella Rydb. (Green 1940). Thus the cor- rect name for the genus, under the present rules, following that round of conservation, would be Lindleya H.B.K., or following those who object to the H.B.K. designation, Lindleya Kunth in Humboldt & Bonpland (nom. cons.) or just Lindleya Kunth following ICBN 2006, Art. 46, ex. 9, note 1 (McNeill et al. 2006). Of interest, in 1858, J. Agardh, in his Theoria Systematis Plantarum, placed Lindleya in its own family Lindleyaceae J. Agardh. In the same paper, he also recognized Cercocarpaceae J. Agardh, Coleogynaceae J. Agardh, and many other new families of dicotyledoneae. >RPHOLOGY AND ANATOMY » few-stemmed shrub (Fig. 1A). In more mesic habitats, the main s an obovoid-spreading shrub to 2-4(-5) m tall with the uppermost branches either ascending or arching outward. In dry habitats the plants form more rounded, densely and di- varicately branched shrubs to 1.3 m tall and wide. These arid-land shrubs may be misshapen due to browsing by goats, or somewhat thorny due to the placement of dried lateral stems. Stems . — Stems are heteroblastic. Young long-shoot stems are initially maroon, glabrous, with internodes 7-15 mm long. They develop a single band of vascular tissue and dense strands of primary phloem fibers. The cortex and pith cells both develop thickened walls. Secondary growth initiates soon after stem development. Short-shoots develop from the nodes. As they mature they may develop into variously branched short-shoot systems to 1— 5(— T) cm long with intemodes 0.4-1.5(-5) mm long. The epidermis of young stems is soon re- placed by a light gray periderm. Older branches have a smooth gray periderm marked by distinctive horizontal lenticles (Fig. ID). Stipules . — Paired yellow-green stipules occur at each long- and short-shoot node at the margins of the canaliculate petioles. They are typically acicular, 0.4-0.9(-2.2) mm long and are bordered on two sides by well- spaced or crowded, distinct, reddish-maroon multicellular glands to 0.2-0.3 mm long (Fig. 3A). When shorter, which decrease in size acropetally, make the stipules appear more deltate. When longer, icular with well spaced marginal glands. The stipules mark the separation of the broadened leaf base from the canaliculate petiole and occur just below the zone of leaf abscission. The petioles mm long 2.5 mm wide to mesophylls 55 mm long and 17 mm owly oblanceolate, narrowly obovate to obovate. The apices range from narrowly acute, obtuse, rounded, to retuse often with an acuminate to mucronate tip. Any one plant may have some leaves with acute tips, other leaves with rounded tips. They are cuneate at the base with the margins forming decurrent narrow wings along the short, canaliculate, yellowish petioles. The margins are crenate; the crenations are gland tipped, with rounded distal and proximal margins. The crenations are usu- ally uniform in distribution, but are more widely separated along the lower blade margins. The leaf blades are glabrous throughout. The marginal dark red-maroon glands are compressed ovoid, sessile, the marginal ones ascending towards the tip, the distal ones erect, 210-290(-350) pm long, to 100-120(-160) pm wide at the broad base. The gland’s marginal cells are anticlinally elongated. The glands fall from older leaves. The dorsi-ventral leaf blades are generally ascending, with the adaxial surface slightly concave (Fig. IB). The blades range from 220-360(-430) pm in trans-section thickness and are 350-400(-430) pm thick at the midvein. The midvein is raised on both the adaxial and abaxial surface proximally, but only on the abaxial surface in the mid section and distally, except in the thickest blades where the midvein is not raised (Fig. 2A). narrow external cell with rather dense cytoplasm and a much larger internal water-storage cell. The outermost wall of the upper epidermal cell is about 5.5-7 pm thick, with the cuticle 4-5.6 pm in thickness. The abaxial Journal of the Botanical Research Institute of Texas 6(2) epidermal and guard cells and the structure is difficult to sort out. The epidermal cells are larger and have a smaller external cell, but the guard cells do not. The adaxial leaf mesophyll consists of 2(-3) layers of crowded palisade cells. A palisade-type cell orientation may or may not occur in the abaxial mesophyll but cells are not as dense. The primary, secondary, and usually tertiary veins have vascular cambia. A mass of support cells develops adaxial to the phloem. In some leaves these consist entirely of collenchyma cells, in others collen- chyma mixed with fibers, and in the primary and secondary veins consist of masses of lignified fibers. Usually :curs both above and below the primary-secondary-tertiary veins rntinue to the fourth and fifth order veins (seen throughout the a collenchymous times bundle sheath e 345 Fig. 2. Leaf structure of Lindleya specimens from Coahuila-Nuevo Leon (B-H), Hidalgo (l-L), and Puebla-Oaxaa (M-Q). B. Henridaon 22056 (TEX)-NL, C Henrickson 6225 (TEX)-Coah. D. Johnston etaL 1 1465 (LL)-Dgo. L Steward 294 (OH)-Coah. F. Steward 378 (GH)-Coah. €. Henridaon 22062 (Tex)-LL. Stanford etai 626 (GH)-Tamp. I. Gonzalez 21443 (F)-Qto. J. Gonzalez 2401 (LL)-Hgo. K. Gold 324 (TEX)-Hgo. L Moore 2481 (GH)-Hgo. M. Salinas 6 Solis 3238 (TEX)-0ax. N. Tenorio 6882 (TEX)-Pue. 0. Dorado 8, Salinas sj). (Tex.)-Pue. P. Tenorio 18341 (TEX)-0ax. Q. Tenorio Gonzdlez 18341 (TEX)-0ax. B-S. Leaf clearings. R. McVaugh 10346 (US). S. Gonzalez 1443. (MEXU). Sale in A = 0.3 mm, in B-Q = 20 mm, in R-S = 1mm. Journal of the Botanical Research Institute of Texas 6(2) states of Queretaro, Hidalgo, Puebla, and Oaxaca). The mesophyll has scattered cuboidal crystals that are more common in bundle-sheath extensions. Venation, as seen in cleared leaves is pinnate, weakly brochidodromous to semicraspedodromus (Ellis et al. 2009). The primary vein is straight, symmetrical, and moderately thickened (Fig. 2R-S). Secondary veins are irregularly alternate, extending from the primary vein at angles of 20-40 degrees. In narrow leaves they tend to arch slightly towards the tip hut do not extend to the leaf tip. The secondary veins exhibit moderate branching towards the margins, often forming through connections with adjacent secondary veins one or more series of arches below the margins, with the crenations served by tertiary or smaller veinlets. Intersec- ondary veins are composite and muched branched, connecting to the lower portions of adjacent secondary veins. Tertiary veins form random reticulations, mostly at right angles to the secondaries; they are percurrent and forked. There are 5 orders of vein branching; the areoles are well developed, random, 4-5 angled, the smallest veinlets are 1-3 times branched. Venation of small leaves differs in that the primary and secondary veins are much thicker than the tertiary and smaller veinlets, and secondary veins may not connect with super- adjacent veins. Stomata are restricted to the lower (abaxial) leaf surface and range in number from 96-175 per mm 2 in small leaves ca. 13 mm long, and 140-175 per mm 2 in larger leaves to 27 mm long. Stomata do not occur below the veinlets with bundle sheath extensions. The stomata are usually subtended by four subsidiary cells. Subsid- Inflorescences . — Flowers are usually solitary terminating new growth of the season (Fig. 1C), but occa- sionally, with vigorous new growth, more than one flower will develop, with the subtending flower(s) develop- ing from lateral shoot(s) from a subterminal node in the new growth of the same season. In such conditions the flowers will appear to form simple corymbs or three-flowered cymes (Fig. IE). The uppermost, reduced leaves, that form with the season’s growth are typically crowded below the flower and extend onto the pedicel. The uppermost cauline leaves are typically only 8-18 mm long, petiolate with slender to deltate, gland-margined stipules. They have canaliculate petioles and gland-margined, glabrous blades (Fig. 3B). The uppermost leaves are reduced to linear to lanceolate bracts to 3-5 mm long, 0.5-1.2 mm wide, without stipules but with glands continuing from the petioles to the blade margins. True pedicels, when present, may be up to 2 mm long. The terminal flowers disrupt terminal growth of a shoot, continued growth occurs through development of sub- :rminate in thick-walled, coriaceous, obconic, hypanthia topped with 5 sepals, 5 petals and about 20 stamens (Figs. 3C, 5A). The ovaries are not attached to the hypanthia except at their sessile bases. The free portion of the hypanthia is about 0.7-1 mm in thickness. The outer hypanthial surface is light green, glabrous and shiny. The inner surface is yellowish-green in color, nectariferous, some- times vertically ribbed (the ribs reflecting the position of stamen traces), with a somewhat expanded inner rim subtending the filaments. Sepals . — The hypanthia terminate in five coriaceous, light greenish, broad-based, ovate to oblong-ovate, lance-ovate, usually acuminate, imbricate sepals (Figs. IE, 3B). Where the sepal margins overlap, underlying sepal margins are membranous to 0.5(-0.9) mm wide and are sometimes cordate at the broad sepal base. The external or overtopping sepal margins are usually not membranous but usually have distinctive reddish glands similar to those found on leaf margins. The sepal tips usually terminate with a reddish gland. The sepals may be glabrous throughout the abaxial surface or variously villous distally. The inner surfaces typically are gla- brous in the lower half but distinctly villous distally and along the distal margins, with slender, crinkly, white hairs to 0.6(-0.9) mm long. There are some differences in sepal size throughout the range of the species; plants from Oaxaca-Puebla tend to have shorter sepals, but this is not consistent as similar short sepals occur in small flowers throughout its northern range. Sepal size usually reflects overall flower size, and flower size can vary with environmental conditions. The sepals usually persist on the rim of the hypanthium as ascending or re- flexed structures (Fig. 3G). Petals .— The five broadly obovate, broadly clawed petals are borne equally along the hypanthium rim al- ternate to the sepals. The petals are spirally arranged in bud and are oblique distally with the portion of the 347 by dry hypanthium and sepals. Note the style bases have been split apart. H. Post-dehiscent fruit with some sepals removed showing dehisced carpels. I. Longitudinal view of mature fruit with carpel interior exposed showing seed. J. Seed. K. Embryo. Magnifications as indicated. A-E ( Henridaon 22 1 12, TEX); G-K [Henridaon 22222, TEX). petal that develops under adjacent petals being much larger than the portion exposed in the bud (Figs. IF, In most petals the right half of the petal is interior in bud and largest. But this will vary from flower to flow< one plant, with occasional flowers having the left half of the petal interior and larger than the right half, petals are white, waxy appearing, well veined, thickened at the base, and thinning towards the somet crinkled margins. They spread outward at anthesis (Fig. IF) and tardily abscise after anthesis. The mid often terminates in a single gland When clearing the petals, the thick basal portion of petal often stains dark— Atubvecia . — Flowers typically have 20 stamens borne at the inner rim of the hypanthia in one series or occasionally in two weak series with occasional smaller stamens born inside larger outer series of stamens (Fig. 3D). The inner and outer stamens usually develop opposite the petals with the interior stamen having a shorter incurved filament. The white filaments are acicular, very broad at the base, tapering distally. The fila- ments are attached to the mid-portion of the versatile anthers (Figs. 3D). Anthers are introrse in bud, 1.5-4 mm long, oblong, apiculate at the tip, with V anther sacs, and the medial septa betwe< The gyn. icated basal lobes. Anthers are light yellow, longicidal, with 4 inther sacs appears to be secretory (glandular) as it is dull red- dish in color (in dried anthers) and stains strongly in microscope slides. Within the buds, the outer, larger stamens are erect with straight anthers, while the inner stamens have inflexed filaments as their anthers de- velop further within the crowded hypanthial cup. These inner anthers are typically bent below the attachment to the filaments (Figs. 3D). They usually retain this shape at anthesis. ound ovary is 5 carpeled, ovoid, glabrous, slightly 5-angled in transverse section, the receptacle apex (Figs. 3C, 5A) and the carpels are united ventrally up to the level of ovule insertion forming a thick-walled compound ovary with axile placentation (Fig. 6D). The carpels are laterally united from the dorsal edge to near the ventral margin. The carpels, however, are not connate centrally where a five-lobed opening extends to the base of the ovary (Fig. 4A). Sterling (1966) noted that this is a carpel fusion pattern similar to that found in the Maloideae. He also noted that occasionally the carpels will not be fused ventrally. Each carpel contains two apical, pendent, collateral ovules, about 0.7-1 mm long at- tached at the inner tip of the locule and receiving a downward-oriented vein from the adjacent ventral trace (Fig. 5A). The short, thick funiculus appears to function as an obtruator. The ovules have a broad outer integu- ment, a thinner inner integument, and the embryo sac is quite large with a crassinucleate nucellus. The carpels are free from the inner hypanthium walls except at the very base. The five styles are separate, terminal, slender, cylindrical and are obliquely expanded at the stigmatic tip (Fig. 3E). A split develops across the oblique tip, exposing papillate interior tissue. Dried styles may persist on the fruit, or they may break off above the bases (Fig. 3F-I). Floral vascularization. The pedicel contains a single cylinder of vascular bundles (Fig. 5E-1). At the base of the hypanthium, 10 (occasionally more) traces separate from the central traces and extend up the hy- panthium (Fig. 5E-3-4). These produce an irregular series of lateral traces that extend horizontally into the hypanthial tissue. The hypanthial traces branch in the mid toupper hypanthium to produce additional stamen traces (Fig. 5D). The traces opposite the sepals typically divide to form 2 stamen traces that then pass parallel branch near the mid-hypanthium to form additional stamen traces, or one or moresumen tr^may^eparatt from the trace near the top of the hypanthium. In addition the trace opposite the petal divides near the top of the hypanthium to produce two lateral sepal traces, one to each of the two adjacent sepals (Fig. 5B, D). At the run oi the hypanthium, the 5 sepals then each receive one medial trace and two lateral traces derived from adjacent petal traces. The lateral traces each divide int. ree separate traces at the base of the sepal, ore parallel traces that extend up through the sepal base and branch above (Fig. 5B). .single basal trace that quickly divides into five to seven traces. These continue to 3 the petal in a pattern as shown in Fig. 5C. The 20 stamens each receive separate from the original 10 hypanthial traces at some point in the mid to The petals each receive branch and anastomose further single traces that, as noted abov upper hypanthium. .p . ^' e ^ entra ^ tratesret ^ a ' n | n S af . er the initia! hypanthia! traces diverge, form into five central packets (Fig. cf, r i maSS ’ c Ve dor5al ‘ races div “8 e opposite the sepal traces receiving tissue from two adjacent vascu ar tissue (Fig 5E 6). These leave behind five dense masses of vascular tissue that becomes the W ™er ventral carpel traces (Fig 5E-6). The dorsal paces continue upwatd akmg the ovary periphery. They grve off senes of branch traces to the ovary wall, but disposition of these late J traces PcLLd by the demtely staining, tannin-containing cells of the developing ovary wall. The ventral vascular tissue forms ten As noted by Sterling (1966), no wing traces diverge from the ventral traces at this time one .SToTon^nh b gl ° bOSe ' 5CarpeUed ' W °° d * loculiddal -P-Ksubtended in the lower I of upper portion of two ovules from B. All from Henrickson 22 1 18, TEX (Galeana area, Nuevo Leon, Mexico). Bar At maturity, the carpels dehisce loculicidally directly through the ventral traces, splitting the ventral traces and the base of the style through a suture that is visible even in the developing ovary wall (Fig. 6A). The dehis- cence splits the style bases in half, and continues onto the distal portion of the abaxial fruit surface (Fig. 3G- H). The inner lateral walls of the carpels are smooth and cartilaginous (as in an apple), lined with a single layer of macrosclereids 100-120 pm thick and subtended by a thick layer of brachysclereids each 35-65 pm in diam- eter with lignified walls 11-22 pm thick (Fig. 6E). The highly lignified tissue extends 1.5 mm in radial thick- ness in the triangular segments between the locules. Only the outermost 0.5 mm of the abaxial-most portion of the triangular carpel segments is not lignified. Seeds.— Two compressed, half-ovoid, dull-brown seeds are produced in each locule. The seed shape con- forms to the locule space, being straight along the ventral edge and rounded on the dorsal edge, with one flat- tened surface (where contacting the adjacent seed) with the outer surface convex (Fig. 3J, 6G). The seed coat is crystals (Fig. 6F). The embryo is oriented with the hypocotyl superior to (i.e., above) the cotyledons. The em- bryo, occupies about 70 percent of the total seed length, leaving a thin wing, 0.2-0.5 mm wide, at the dorsal margins. The embryo consists of two compressed, oblong cotyledons and a smaller, obovoid hypocotyl (Fig. 3K). Endosperm is absent at maturity. The seeds are wind dispersed. Upon germination the cotyledons form the first seed leaves of the seedling. TAXONOMIC TREATMENT LINDLEYA KunthNov. Gen. Sp. [H.B.K.] 6:240 (ed. qto.); 188 (ed. foL). 1824 (nom. cons.), nonUmfleyaKunihNov. Evergreen, multistemmed shrubs; periderm gray, smooth. Stems heteroblastic, tending to form shortened axil- lary spurs in arid conditions. Leaves simple, alternate; leaf bases short, the stipules acicular to debate, maroon, sometimes with marginal glands; leaf blades narrow to broadly oblanceolate to obovate, ovate, acute to round- ed, sometimes retuse at tip, cuneate with the margins decurrent on short petiole, the margins closely crenate, the marginal teeth terminating in distinct glands, the blade coriaceous, shiny green, glabrous on both surfaces. 3S0 Journal of the Botanical Research Institute of Texas 6(2) Henrickson, Systematics of Lindleya 351 major veins with bundle sheath extensions, venation brochidodromous to semicrospedodromus. Flowers complete, perigynous, usually solitary and terminal on long and short shoots, these sometimes forming termi- nal corymbs when the shoots aggregated or on long-shoot stems; bracts reduced, linear, gland-margined, borne on upper pedicel; hypanthia obconic, hemispherical in fruit, green, glabrous outside, nectariferous and yellow-green inside, coriaceous; sepals 5, imbricate, debate to ovate, acute to acuminate, glabrous outside, vil- lous near tip inside, somewhat coriaceous except along overtopped margins, spreading, persisting on fruit; petals 5, borne at the rim of the hypanthium, obliquely obovate, asymmetrical, white, spreading at anthesis, tardily deciduous, aromatic; stamens usually 20; filaments subulate, broadened at the base, borne at the inner rim of the hypanthium in a single (rarely two) series; anthers lanceolate, large, versatile, longicidal, yellowish; ovary; styles terminal, distinct, the stigmas terminal, oblique; ovules 2 per carpel, collateral, apically attached, pendent, the micropyles superior, the funiculi thickened, the tissue serving as an obtruator. Fruits spheroidal, Journal of the Botanical Research Institute of Texas 6(2) woody, loculicidally dehiscent capsules, the lower third surrounded by the coriaceous, persisting hypanthi- um, the carpels dehiscent distally along ventral and dorsal sutures, with thick, inwardly cartilaginous valves; seeds 2 per locule, compressed, narrowly winged abaxially, brown, the seed coat thin, the embryo with oblong- ovate, flattened cotyledons, the hypocotyl small, superior; endosperm absent. With one species. Lindleya mespiloides Kunth, No 22(3):259. 1908. Type: MEXICO. Hn ENE of Actopan, Hgo.), May (holoti lotype: NY!). Distinguished by Rydberg as ien. Sp. [H.B.K.] 6:240. 182! ;o: Inter La Puente de la Madre de D specimen of L. mespiloides with very large features. Other specimens from smaller leaves. As noted al N W of that of L. mespiloides. As it is distinguished only by quant Erect, much-branched, evergreen shrubs to small trees l-3(-5) m tall, in dry habitats forming small, tightly, divaricately branched shrubs with many short shoots, in mesic habitats forming erect-ascending, moderately branched, tall shrubs; stems heteroblastic; long-shoot branches with intemodes 7-15 mm long, glabrous, ini- tially maroon, developing a close smooth, gray periderm; short-shoot branches l-5(-7) cm long, variously branched, with intemodes 0.4-1.5(-5) mm long. Leaves with petioles l-2(-4) mm long; stipules 0.4-2.2 mm long, acicular to deltate, maroon, when longer more attenuate and bearing marginal glands; leaf blades nar- rowly oblanceolate, oblanceolate, spatulate, obovate, sometimes elliptical-oblanceolate, rarely somewhat ovate, (3.5-)ll-32(-55) mm long, (1.6-)4-13(-21) mm wide, acute, acute-acuminate, rounded to emarginate at the tip, narrowly cuneate with the margins forming wings above the short petiole at base, the margins closely crenate to crenulate with 5-10(-15) teeth per cm of margin, the teeth each terminating in a maroon, conical gland 0.1-0.3 mm long, the blades coriaceous, glabrous throughout, shiny, dark green, slightly concave above, more yellow-green beneath, the midvein yellowish and raised on both surfaces. Flowers terminal, soli- tary on short, leafy shoots, rarely in terminal 3-flowered racemes, the subtending leaves reduced, the upper- i to linear, gland-margined bract(s) to 5 mm long; pedicels 1-2 mm long, glabrous, 4-5 mm wide (to 7 mm wide pressed), green, glabrous outside, yellowish-green, „ , m : the ovar y- the distal rim ^ghtly expanded; sepals 5, ± imbricate, deltate to ovate, ! ally^Uhr^t^^no^s mm W ' de ’ *“'* “* COriaCeOUS ^ 8rKn med '' glabrous at the base but densely villous n^r t^e tipinsitk^ind ^tenTiUous^dlL^hrou^ouu^ffwrgins* tfw ] s With conical glands along the margins, the thin inner margins somewhat cordate at the urity; petals broadly obovate to ob- bowers W-)22-40 mm tot,! diameter], obliquely rounded to emarginate at the dp, broadly cuneate above a very short and broad claw at the base, firm, thickish, waxy, white, spreading, emitting a street aroma remain- t™””" -tfter anthesis, eventually deciduous, stamens usually 20; filaments linear-subulate, dis- I n long, to 0.5-0.8 mm broad at the expanded bases, borne in the inner rim of the hypan- u e with the shortest laments occurring inside a larger filament opposite the a ^ 2 ^ 3 5( - 4> r m “ ^ p - co " I itue, reddish to usually dark yellow; ovary two-thirds + most leaf(s) often ri ectariferous i base, the sepals spreading in flower and persisting around the fru --orbicular, obliquely asymmetrical distally, contorted in bud wit! :t, (2-)3.5-7(-8) n n, of variable length and s date, the basal lobes often spreading or folded - TXT Z " a, h “ Rnded t,ps - Fruils of ' TOod * ovoid-spheroidal, long-persistent capsules, 7-10 adlZienUn^dis, Tft T° Unded by ““ I”*** ‘W—W™. the carpelsLh locuB- y ’ 8 ’ 8 ark red to maro °n, drying dark brown, glabrous; ovules 2 per locule, pendent; Henrickson, Systematics of Lindleya 353 seeds two per locule, oblong-ovate in outline, flattened, 4.3-6 mm long, 1.8-2.6 mm wide, narrowly winged on the curved abaxial margin, the adaxial margin straight; cotyledons ovate, endosperm absent, radical superior. The species is characterized by its shrub to small tree growth habit with smaller plants of more xeric habitats often developing short-shoot branches and a somewhat thorny aspect, by its smooth gray bark, by its small to moderately large shiny green, mostly oblanceolate, acute to round-tipped leaves with gland-tipped finely crenulate margins, by its large terminal, sweetly aromatic flowers with a thickened obconic hypanthium that bears 5 glabrous sepals with thin villous-ciliate margins, 5 large, white, obliquely obovate to somewhat orbicular petals, and ± 20 stamens with subulate filaments and large versatile anthers, and by its superior, 5-carpelled ovary with 2 suspended ovules per locule, and 5 separate styles. The fruit are globose, woody, locu- licidally dehiscent capsules with each locule producing two flattened seeds, each with a short wing along its outer margin. The flowers are conspicuous and remain on the plant after the anthers have shed their pollen creating a conspicuous floral display. The fruits are often long persistent, allowing recognition of the species in the field. The species has three regions of distribution (Fig. 7): northern Oaxaca and adjacent Puebla, where it oc- curs oak-pine woodlands, chaparral and thorn scrub in association with species of Malacomeles, Vauquelinia, Comarostaphylos, Rhus, Quercus, Juniperus, Pinus, Acacia, Leucanea, Beaucamea, and Yucca etc., from 1900- 2400 meters elevation; in Hidalgo and adjacent northern Queretaro, where it again occurs from pinyon-oak- juniper woodland, chaparral association with many of the same genera. Its largest distribution is in the Sierra Madre Oriental and Chihuahuan Desert region from southwestern Tamaulipas, southern Nuevo Leon, north- ern San Luis Potosi, northern Zacatecas, northeastern Durango, the southern half of Coahuila and adjacent southeastern Chihuahua where it occurs in pinyon-juniper woodland, chaparral and canyons in desert scrub often in limestone, rarely gypseous or sandstone substrates in association with species of Rhus, Garrya, Ceano- thus, Fraxinus, Cercocarpus, Gochnatia, Vauquelinia, Berberis, Arctostaphylos, Quercus, Arbutus, Juniperus, Agave, Yucca, Dasylirion, Ungnadia, Leucophyllum, Condalia, Mimosa, Mortonia, Foresteria, Acacia, and Pinus from 1100-2700 m elevation. Flowering typically occurs in May but may occur from March to September usually following rains, with occasional flowering occurring throughout the year. Throughout the wide range of the species some notable variation occurs. Most apparent is the variation in leaf size. Most specimens from Hidalgo and Oaxaca have moderate large leaves 18-55 mm long, 8-18 mm wide. In contrast, most specimens from Coahuila and Nuevo Le6n have smaller leaves, 10-26 mm long, 3.5-8 mm wide. The separation is by no means complete, as some specimens from both Hidalgo and Oaxaca also have small leaves. Leaf size corresponds to habitat, as plants in dry exposed areas have very small leaves that contrast with larger leaves of plants of nearby less-arid, shaded sites. As noted above, plants from Hidalgo, Queretaro, Puebla and Oaxaca have bundle-sheath extensions extending to the 4th and 5th order of vein branching in contrast to only the 1st, 2nd, and in part to the 3rd order of branching in more northern range. Specimens observed from Puebla and Oaxaca also tend to have shorter sepals measuring 3.0-5.2 mm long. In contrast, sepals from Hidalgo and northward specimens tend to be larger, (3.0-)4.5-7(-9) mm long, but there is considerable overlap, with sepal length corresponding to flower size and being variable even on a single specimen. Flowers range in total diameter (petal tip to petal tip) from 17 to 40 mm. Fruits also vary in size, with occasional specimens having fruits much smaller or larger than average. PHYLOGENETIC RELATIONSHIPS The placement of Lindleya within the Rosaceae has been problematic. Numerous subfamilial and tribal classi- fications of the Rosaceae have been proposed (Focke 1888; Hutchinson 1964; Schulze-Menz 1964; Cronquist 1981; Takhtajan 1987 1997, 2009; Kalkman 1988, 2004; Thome 1983, 1992) etc. and most all retain four sub- families, (some older treatments have five subfamilies with the inclusion of the Chrysobalanoideae, or six with the Neuradoideae) that corresponded well to fruit types: the Spiraeoideae, with follicles, (rarely achenes — Ho- lodiscus, or capsules — Vauquelinia, Lindleya)- Rosoideae with achenes (rarely druplets — Rubus); the Amygda- loideae with drupes ( Prunus etc.), and the Maloideae (Pomoideae) with pomes ( Malus etc.). Within these sub- families, the treatments differed in the placement of genera within tribes. See summary in Potter et al. (2007). Journal of the Botanical Research Institute of Texas 6(2) There has always existed a problem group of six genera with woody fruit and winged seeds some with follicles, others with capsules, consisting of: Quillaja, (plant polygamo-dioecious, fruit of five radiating woody follicles each with many (10-16) seeds in two series per carpel— northern South America); Kageneckia (dioe- cious, fruit with five separate follicles with many seeds in two series per carpel— western South America); Vauquelinia and Lindleya, (dry capsules with two-winged seeds per carpel— Mexico); Exochorda (plant polyg' amo-dioecious; fruit with five, rather compressed woody follicles— Eurasian) and Lyonothamnus (two follicles ing x=9, the Amygdaloideae x=8, the Rosoideae x=7, 9 (rarely 8) and the Maloideae x=17. The high chromosome number in Maloideae caused Sax (1931, 1932, 1933) and later Stebbins (1950, 1958) and many others (see Phipps et al. 1991; Rohrer et al. 1991, Cronquist 1981) to promote the view that the Maloids arose via paleo-allopolyploidy from x=9 Spiraeoideae and x=8 Amygdaloideae ancestors or from within the Spiraeoideae (Gladkova 1972). In 1976 Goldblatt contributed chromosome numbers of several of the problem genera noted above. Of these Exochorda was found to be x=8, Kageneckia and Lindleya x=17; Vauquelinia x=15; and Lyonothamnus x=27 and Quillaja x=14. The high numbers in these taxa again gave clues that their relationship may lie with the Maloideae, and Lindleya and Vauquelinia were subsequently transferred to the Maloideae (Pyroideae) by Thorne (1983) and Takhtajan (1987). The 5-carpelled, dry-fruited Exochorda (x=8) was considered by Goldb- latt (1976) to belonging to the x=8 Prunoideae, which has been supported by molecular data (Morgan et al. 1994). Kageneckia and Quillaja, which have similar appearing fruits, remained in subfamily Quillajeoideae (Thorne 1983; Takhtajan 1987). Floral morphology Data . — Sterling (1966), on the basis of ovary morphology, considered the Quillajeae (containing Exochorda, Kageneckia, Lindleya, Quillaja, and Vauquelinia) to have sharp differences in the gynoe- cium structure and number and orientation of ovules. He noted that Lindleya had ovaries with complete lateral intercarpellary fusion of a type characteristic of the Maloideae, but with minimal hypanthium fusion. He also noted that the carpels of Vauquelinia were also maloid in nature though fused only basally and ventrally while being separate laterally. Data from Rusts . — Savile (1979) summarized the use of fungus-host relationships in plant phylogeny. He notes that cedar-apple rusts Gymnosporangium occur throughout Pomoideae (Maloideae) with species also known from Myricaceae, Hydrangeaceae and from two genera of Spiraeoideae ( Vauquelinia and Porteranthus — now Gillenia ). He cites Gymnosporangium vauqueliniae Long and Goodding (1939) on Vauquelinia califomica from Arizona and interestingly, Gymnosporangium externum Arth. & F. Kern in Arth., on Gillenia in eastern United States, which had been known since 1903. The on-line “Fungal Database” also gives references of Gym- ditional collections of G. externum on both species of Gillenia in the eastern United States (http://nt.ars.grin. gov/fungaldatabases). Gymnosporangium has not been reported on Lindleya. r Data . — If we stop here and look at the data accumulated by the 1980s, we number as members of the then recognized Maloideae, and has some floral features in common with the Maloideae. We also have data from Cedar-apple rusts linking the re- lated Vauquelinia with the Maloids. But confusingly, the rusts also occurred in another Spiraeoid, namely Gil- lenia. But as there was evidence that indicated that Vauquelinia and Lindleya were related to Maloids, there was no data indicating whether they were derived from the fleshy-fruited Maloids, or if they were basal to the fleshy- fruited Maloids. When we monographed Vauquelinia (Hess & Henrickson 1978), we had no evidence pertain- ing to the relative placement of Vauquelinia to the fleshy-fruited Maloids, so in that paper we concentrated just on species relationships. In the 1991 symposium on the Evolution in the Maloideae (Rosaceae) published in Systematic Botany, Vauquelina and Lindleya was not mentioned nor included in data sets (Phipps et al. 1991; Robertson et al. 1991). But by the mid 1990s, molecular data began to provide answers and laid the ground- work for a new phylogeny of the Rosaceae. ar data presented by Morgan et al. (1994) from chloroplast rbcL sequences vidence for a needed subfamilial rearrangement of the Rosaceae. Their paper divided the family along the same lines as cytological data. Their data, however, showed that the old Rosoideae was polyphyletic, that thex=7 genera ( Filipendula, Fallugia, Geum, Waldsteinia, Potentilla, Fragaria, Agrimonia, Rosa, Rubus and the x=8 Alchemilla) formed the core of the Rosoideae and the x=9 genera formed four distinct Journal of the Botanical Research Institute of Texas 6(2) groups: the Neviusia, Rhodotypos group; the Cercocarpus, Purshia, Lyonothamnus group; and the core Spiraeoi- deae with basal Spiraea-Aruncus-Holodiscus group, separate from a derived Physocarpus-Neillia group, and an- other Sorbaria group associated with Chamaebatiaria including the achene-bearing Adenostoma. Exochorda fell into the x=8 Amygdaloideae. Their data excluded the South American Quillaja from the Rosaceae. Of significance to this paper, their data showed Kageneckia, Undleya and Vauquelinia were basal to the remainder of the x=17 Maloideae, i.e., their data showed Vauquelinia and Lindleya, with capsular fruits, and Kageneckia with follicle-like fruits, as remnants of a clade that have given rise to the core Maloideae. That is, they were basal to the Maloids, not dry-fruited derivatives of fleshy-fruited Maloids. Campbell et al. (1995), using the internal transcribed spacers (ITS) of the nuclear ribosomal DNA region, studied phylogenetic relationships within the Maloideae. Their study, however, concluded that the Maloideae was not monophyletic. They also noted that Vauquelinia forms a well supported clade with fleshy-fruited Erio- botrya and Raphiolepis that is the sister group to the remainder of the Maloideae. These data, however, were badly skewed, as their “ Vauquelinia ” sample was actually Raphiolepis collected at the University of Arizona campus by a graduate student, vouchered by a specimen collected twenty years earlier in 1975. Their data set did not include material of Lindleya, Kageneckia, and as it turned out, also did not contain Vauquelinia. Data presented by Evans et al. (2000) on the “Granule-Bound Starch Synthase 1” gene (GBSSI) provided further insights into the phylogeny of Rosaceae. The portion of this nuclear gene used (near the 5’ end) consists of seven complete, short exons, and parts of two other exons alternating with non-coding introns. While all other diploid families in which this gene has been used, have only one GBSSI sequence, all diploid Rosaceae have two distinct sequences (designated as GBSSI-1 and GBSSI-2) that differ in the length of, or presence or absence of, the introns between particular exons providing evidence of Rosaceae monophyly. Species of Maloi- deae, with their higher chromosome number, have two copies or loci of each sequence, each of which have differences in their base-pair sequences that are designated GBSSI-1A and IB, and GBSSI-2A and 2B. So there are six different sequences or loci, the GBSSI-1 and -2 in the diploid non-maloids studied, and GBSSI-1A, -IB, -2A, and -2B occurring the Maloideae. The sequence data, using only exon base pairs, showed that various sampled collections of Kageneckia had GBSSI-1A, -2A and -2B loci; and Vauquelinia had GBSSI-1A, -IB, -2A and -2B loci as in members of the Maloideae. Their results again showed that Kageneckia and Vauquelinia were A later report by Evans and Campbell (2002) used CBSSI gene to investigate the origin of the x=17 Maloi- deae. Their phylogenetic analysis of some 42 genera showed that GBSSI-1 and -2 alleles of Prunus (Amygdaloi- deae) were not closely associated with the Maloideae, but rather the sequences from the genus Gillenia (a her- baceous, x=9 Spiraeoid, with compound leaves and 5 separate ovaries each with 4-6 ovules that form folicular fruits with non-winged seeds) were strongly associated with, and basal to, sequences of Vauquelinia, Lindleya and Kageneckia at the base of the Maloideae clade. Gillenia has a GBSSI-1 locus that shares distinct intron dele- tions and additions with the GBSSI-1B loci of the Maloideae and the GBSSI-2 intron shared a distinct base pair substitution with the GBSSI-2B loci of Vauquelinia, Kageneckia and Lindleya and core Maloideae placing the diploid Gillenia at the base of the Maloideae. Thus Gillenia would appear to be an extant survivor of a lineage ancestral to the Maloids. But what is the other parent? Their survey of the Amygdaloideae has found no potential parent. The other parent has either not been sampled or is long extinct. They conclude that the other parent could have been another x=9 Spiraeoid that, in forming a hybrid via amphiploidy, could double the chromosomes to x=18, (2n=36) and this could be reduced to x=17 (2n=34) via aneuploidy (Evans & Campbell 2002). But whatever the other parent would be, there is no reason to expect its lineage to be extant today. But definitely one of the parental lineages has a sur- viving member, that being Gillenia, a genus of two species native to the eastern United States. The most recent molecular study of relationships within Rosaceae was presented by Potter et al. (2007) that investigated the relationships of 88 genera using nucleotide sequence data from six nuclear and four chlo- roplast regions. Their paper resulted in a complete infrafamilial rearrangement recognizing three clades as subfamilies. The basal subfamily Rosoideae consists of herbs, shrubs rarely trees, that lack cyanogenic glyco- sides and sorbitol. They have alternate, usually compound, stipulate leaves; ovaries are usually numerous (rarely 1), separate, free from the hypanthium, some borne in a spiral arrangement on expanded receptacles, and the fruit are indehiscent [x=7(-8) — including Agrimonia, Filipendula , Fragaria, Geum, Potentilla, Rosa, Rubus etc.]. The subfamily Dryadoideae clade consists of shrubs, subshrubs, with cyanogenic glycosides and traces of sorbitol and a tendency to have nitrogen fixing symbionts. The leaves are simple or compound, stipu- late; ovaries are 1 or many; fruits are achenes, (x = 9) including Cercocarpus, Chamaebatia, Dryas and Purshia (including Cowania). The Dryadoideae is sister to a highly diverse Spiraeoideae 1 clade consisting of mostly shrubs, small trees with some cyanogenic glycosides and strong sorbitol presence. Leaves are usually simple, alternate, the stipules persistent (deciduous in Prunus); ovaries number 1-5, mostly separate, radially oriented, usually free from the hypanthium, and fruits ranging from achenes, drupes, to pomes (x= 8, 9, 15, 17). It in- cludes what was in the subfamilies Spiraeoideae ( Spiraea , Holodiscus, Petrophyton etc.), Amygdaloideae (Prunus), and Maloideae (Amelanchier, Crataegus, Malus, Pyrus etc.) of previous classifications and several Within the subfamily Spiraeoideae, the Maloid clade was designated in Potter et al. (2007) as the super- tribe Pyrodae [the subfamily name Pyroideae Burnett (1835) named for Pyrus, the pear, has priority over name Maloideae Weber (1964)]. Basal in the supertribe is the genus Gillenia (x=9), that Evans and Campbell (2002) showed is a surviving genus of a lineage that gave rise to the Maloids. The supertribe has a single tribe, Pyreae, containing the three genera Lindleya, Kageneckia (x=17), and Vauquelinia (n=15) — its chromosome number the product of further aneuploidy and the tribe Pyreae has one subtribe, Pyrinae, that includes all the core maloids with apple-like fruits (pomes). But I see a problem with this classification. There remains no rank to distin- guish the variation in the core Pyrinae (the maloids) as investigated by Lo and Donoghue (2012). And that is the subject of another paper. Lindleya vs. Vauquelinia . — Both Lindleya and Vauquelinia (Hess & Henrickson 1987) are moderately large, evergreen shrubs, distributed from northern, east-central to southern Mexico, with Lindleya ranging from northern Oaxaca to western Coahuila and the three species of Vauquelinia occurring from central Oaxaca to northern Baja California Norte, south-central Arizona and trans-Pecos Texas. Both genera occur in arid to semiarid scrublands, often limited to more mesic niches on north-facing slopes and along drainages or on rocky habitats where their roots can reach deeper moisture. Both have coriaceous leaves with well developed, fibrous bundle-sheath extensions with leaves of Vauquelinia being much larger and usually than those of Lindleya. The smaller flowers of Vauquelinia are arranged in distinct well-branched compound corymbs (but see Evans & Dickinson 1999, who consider the inflorescences to be determinate, alternately branched dichasia with lateral pleiocasia similar to those found in some Crataegus, sensu Weberling 1989), not mostly solitary as in Lindleya. As in Lindleya, their leathery hypanthia bear five, ± thick, persistent sepals, five, white, ovate pet- als, (18-)20 stamens with tapered filaments, yellowish, introrse anthers and the sessile 5-carpelled ovary is mostly free from the hypanthium and topped with five separate styles, each somewhat compressed distally with broad stigmas. But unlike Lindleya, sepal margins of Vauquelinia do not contain the multicellular glands; anthers and petals are much smaller; ovaries are villous; and carpels are connate only along the inner (adaxial or ventral) margins — the outer (lateral and abaxial) margins are free (see Hess and Henrickson 1987). In Vau- quelinia, each carpel has two basal-attached ovules (not apically attached as in Lindleya); the ovules, as in Lindleya, have two integuments, and the upper portion of the ovule develops into a wing. In Vauquelinia, ma- ture fruits are more ovoid to oblong-ovoid (not ovoid-globose) in shape, and the fruit body is distinctly five lobed in cross section with five radial incisions, villous to strigose (not glabrous), but as in Lindleya, each carpel Journal of the Botanical Research Institute of Texas 6(2) is loculicidally dehiscent all across the ventral and distal portion of the dorsal sutures, splitting the persisting style base to shed the winged seeds. The embryos in Vauquelinia are about half the total seed length, are ascend- j ing, with basal hypocotyls (not two thirds the seed length with apical hypocotyls as in Lindleya). In both genera the seed coat is thin, brown, and endosperm is lacking. They also differ in chromosome number, with Vau- j quelinia n=15 and Lindleya n= 17. The South American (Chile, Peru, Brazil) Kageneckia (n= 17) is quite distinct. The 3(-4) species are dioe- j cious, ± large shrubs, vegetatively quite similar to some species of Vauquelinia in having thick, coarse leaves and occurring in dry scrublands. They have flowers ± 20-40 mm in diameter with attenuate sepals, moder- ately large, oblong-ovate white petals, 15-20 stamens along the hypanthial rim in male flowers but with re- duced staminodia, and 5 separate vertical ovaries in female flowers, with terminal short styles and 2 rows of j ovules (10-12 total) along the ventral traces. In fruit, the separate carpels expand abaxially and radiate out- ward to enclose the seed wings and the fruiting ovaries open along both the ventral and dorsal sutures to dis- perse the 10-12 winged seeds. Spjut (1994) refers to the fruit of Vauquelinia as a coccetum “a multiple fruit with dehiscent fruitlets” j with each carpel having a separate style-stigma. But the distinction between the fruits of Vauquelinia and Lind- leya is only in the amount of lateral connation of the 5 carpels, being restricted to near the interior (ventral or adaxial) region in Vauquelinia but throughout the lateral surfaces in Lindleya. In overall structure and function, they are both loculicidal capsules. In contrast in Kageneckia. the carpels are completely separate, and Spjut (1994) would designate the fruit as a follicetum (i.e., a cluster of follicles) but while dehiscence occurs primar- ily along the dorsal suture, it continues onto the ventral suture, and his definition removes it from the follicle category, making it fit Spjut’s definition of a coccetum as in Vauquelinia. In molecular phytogenies that include Vauquelinia, Lindleya and Kageneckia [Potter et al. (2007); Camp- bell et al. (2007); Lo & Donoghue (2012)], Lindleya and Kageneckia are most often associated and sister to Vauquelinia and the rest of the pome-bearing Pyrinae. But also see Campbell 2007 for analysis of separate GBSSI genes. APPENDIX 1 rra de Pampas, W of Hacienda El Berrendo, 27°20’N, I04°43'W, 25 Aug 1972, Chians J mmit Sierra de Chupaderos, 27°12’N, 104°43'W, 5300 ft, 2 Oct 1973, Henrickson 13776 J (TEX). Coahuila: E slope Sierra Almargre, 4800 ft, 5 May 1973, Gentry 6- Engard 23219 (CAS, US); Vicinity of Santa Elena Mines, E foothills j of Sierra de las Cruces, 30 May 1941, Stewart 378 (F, GH, LL); Sierra Mojada, S of Esmeralda, 27°16'N, 103°4rW, 1 Sep 1972, Chiang etal. j 9086n (LL); 23 (air) mi NW of Las Delicias, Valley N of Sierra de las Delicias, 26°23‘N, 102°52’W, 4800 ft, Henrickson 6124 (TEX); ± 29 (air) : mi WNW of Cuatro Cienegas, N slope of Sierra de la Madera, 7.8 (rd) mi W of Rancho Cerro de la Madera, Cafidn Desiderio, 27°08'N, 102°30'W, 12 Aug 1976, Henrickson & Prigge 15310 (TEX); Cerro San Pedro, near N.L. line, 2 mi E of Ejido Presa de San Javier, 24°44’N, 100°46'W, 2200 m, 21 Aug 1974, Wendt & Lott 608 (TEX); Sierra dejimulco, 8 km NE Estacion Otto, 27 Sep 1972, Chiang et al. 9552h (LL); ? Sien-a de Parras, 5500 ft, Shreve & Tinkham 9859 (GH); 10 mi W Saltillo, near Las Barrancas, 3 May 1959, Correll & Johnston 21400 (GH, NY, J , 4 Sep 1938, Shreve 8581 (US); Concepcion del Oro, 2500-2700 m, 18-19 Jul | atecas-Coahuila state line, 24°44'N, 101°10'W, 1990 m, 29 Mar 1973 Johnston et al. 10491a (LL). San Luis PotosE 15.1 id. mi N of Zac-SLP, hwy 49 on rd to Charcas, near Cerro Tecalote, 22°30'N, 101°09'W, 2100 m, 23 Sep 1978, Henrickson & Lee 17553 (TEX); Lajoya, 4 km NW de Ventura, Mcpio Villa Hidalgo, 1900 m, 11 Jan 1955, Rzedowski 5705 (TEX); 20 | ss & Byrne 4710 (F); E de Nunez, km 84 t, Rzedowski 5549 (ENCB). Nuevo Leon: Mts near Monterrey, Jul 1933; Mud- j E of San Rafael, 25°03'N, 100°25 r W, 22 > jn 1934, j «,US);13kmalESan 22 Sep 1974, ordetal 626 (DS, GH, NY); 34 km 80 km NE of Querttaro, above 1. 7425b (GH). d to Pinal de A moles, 2700 m, 24 Apr 1, Pennell 17419 (GH, h Henrickson, Systematic of Lindleya 2850 (DS, TEX); Cerro San Miguel, 14 km NNE de Actopan, 2300 m, 6 May 1965, C lan, road to Zacualtipan, 1600-1800 m, 24 Mar 1947, Moore 2481 (GH). Puebla: Los Naranjos, May 1908, Purpus 3234 (F, GH, NY); Mcpio Caltepec, Cerro El Mirador al SW de Coatepec, 19 Apr 1985, Tenorio8822 (TEX); 8kmalNE Acatepec, 16 May 1981, Chiongrtol. 1959 (TEX). Oaxaca: 3 km al SW de Tamazulapan sobre carr. Tamazulapin-Chilapa de Diaz, 15 May 1982, Rico et al. 332 (F); Cerro sobre el camino de Teposcolula a San Andres Lagunas, 10 May 1981, Cedillo et al. 770 (CAL, F); 3 km al S de Santiago Teotongo por la Terraceria a San Pedro Nopala, 17°45'N, 97°33W, 12 May 1986, Salinas &■ Solis F3238 (TEX); 3 km SW de Magdalena Jicotlin a Santiago Teotongo, 97°29'N, ACKNOWLEDGMENTS I thank the curators for the herbaria mentioned for loans and Thomas Wendt for reading over an earlier version of this paper, the reviewers of the article, which has been awaiting publication for more than a decade. Figures 3 and 5 were drawn by Bobbi Angell. Timothy A. Dickson (TRT) and Joseph R. Rohrer (UWEC) gave the manuscript a critical review. REFERENCES Agardh, J.G. 1858.Theoria systematis plantarum. Apud. C.W.K. Gleerup, Lund. Briquet, J. 1906. Regies internationales pour la nomenclature botanique principalement des plantes vasculaires. Gustav Fischer, Jena. Briquet, J. 1 935. International rules of botanical nomenclature, adopted by the International Botanical Congresses of Vi- enna 1 905, and Brussels, 1 91 0, revised by the International Botanical Congress of Cambridge, 1 930. G. Fischer, Jena. Campbell, C.S., MJ. Donoghue, B.G. Baldwin, and M.F. Wojciechowski. 1995. Phylogenetic relationships in Maloideae (Rosaceae): evidence from sequencies of the internal transcribed spaces of nuclear ribosomal DNA and its congru- Campbell, C.S., R.C. Evans, D.R. Morgan, TA Dickinson, and M.P. Arsenault. 2007. Phylogeny of subtribe Pyrinae (formerly the Maloideae, Rosaceae): limited resolution of a complex evolutionary history. PI. Syst Evol. 266:1 19-145. Cronquist, A 1 981 . An integrated system of classification of flowering plants. Columbia Univ. Press. Elus, B., D.C. Daly, LJ. Hickey, K.R. Johnson, J.D. Mitchell, P. Wilf, and S.L Wing. 2009. Manual of leaf architecture. Comstock Publishing Associates (Cornell University Press) in association with The New York Botanical Garden Press, Ithaca NY. Evans, R.C. andT.A. Dickinson. 1999. Floral ontogeny and morphology in subfamily Spiraeoideae Endl. (Rosaceae). Int. J. PI. Sci. 160381-1012. Evans, R.C., LA. Alice, C.S. Campbell, EA Kellogg, andT.A. Dickinson. 2000. The Granule-Bound Starch Synthase (GBSSI) gene in the Rosaceae: multiple loci and phylogenetic utility. Molec. Phylogen. Evol. 1 7:388-400. Evans, R.C. and C.S. Campbell. 2002. The origin of the apple subfamily (Rosaceae: Maloideae) is clarified by DNA sequence data from duplicated GBSSI Genes. Amer. J. Bot. 89:1478-1484. Fedde, F. 1940. Nomenklatorische Notizen. 1. Repert. Spec. Nov. Regni Veget. 48:1 1. Focke, W.0. 1888. Rosaceae. In: A. Engler and K. Prantl, eds. Die natiirlichen pflanzenfamilien 3, 3:1-61. Engelmann, Leipzig. Gladkova, V.N. 1 972. On the origin of the subfamily Maloideae. Bot. Zurnal SSSR. 57:42-49 (in Russian). Goldblatt, P. 1976. Cytotaxonomic studies in the tribe Quillajeae (Rosaceae). Ann. Missouri Bot. Gard. 63:200-206. Green, M.L. 1940 XII. Additional nomina generica conservanda (Pteridophyta and Phanerogamae). Kew Bull. 1940: 81-134. Gunner, S.T., S.M. Stuber, M.R. Carvalho, G.S. Lim, and J.L Reveal. 201 1. (2038) Proposal to conserved Malaceae, nom. cons, against Amygdalaceae, nom. cons. (Magnoliophyta), a 'superconservation' proposal. Taxon 60:1 776-1 778. Hess, W. and J. Henrickson. 1 987. A taxonomic revision of Vauquelinia (Rosaceae). Sida 1 2:1 01-1 63. Hutchinson, J. 1964. The genera of flowering plants. Vol 1. Dicotyledons. Claredon Press, Oxford. Johansen, D.A. 1940. Plant microtechnique. McGraw-Hill Book Co. New York. Long, W.H. and LN. Goodding 1 939. Two new species of rusts. Mycologia 31 :670-673. Kalkman, C. 1988. The phylogeny of the Rosaceae. Bot. J. Linnean Soc. 98:37-59. Kalkman, C. 2004. Rosaceae. In: K. Kubitzki, ed.The families and genera of vascular plants. 6:343-386. Kunth, C.S. 1823. Casearia Jacq. In: C. Humboldt and A. Bonpland. Nova Genera et Species Plantarum 5:361-367, [tabs. 479, 480— published 25 Feb 1822]. Kunth, C.S. 1822. Malvaceae, Biittneriaceae, Tiliaceae. 1-10 p. Vulgavit, Paris. Kunth, C.S. 1824. Lindleya. In: C. Humboldt and A. Bonpland. Nova Genera et Species Plantarum 6: 239-241, tab. 562. 360 Journal of the Botanical Research Institute of Texas 6(2) Lo, E.Y.Y. and MJ. Donoghue. 201 2. Expanded phylogenetic and dating analysis of the apples and their relatives (Pyreae, Rosaceae). Molec. Phylogen. Evol. 63:230-243. Morgan, D.R., D.E. Soltis, and K.R. Robertson. 1 994. Systematic and evolutionary implications of rbcL sequence variation in Rosaceae. Amer. J. Bot. 81 :890-903. McNeill, J., F.R. Barrie, H.M. Burdet, V. Demoulin, D.L. Hawksworth, K. Marhold, D.H. Nicolson, J. Prado, P.C. Silva, J.E. Skog, J.H. Wiersema, and N J. Turland (eds & Compilers). 2006. International code of botanical nomenclature (Vienna Code) adopted by the Seventeenth International Botanical Congress Vienna, Austria, July 2005. Gantner, Ruggell. (Regnum Veg. McNeill J. and N J. Turland. 201 1 . Article 1 6. In: Synopsis of proposals on Botanical Nomi review of the proposals concerning the International Code of Botanical Nomendatu Nees von Essenbeck,T.F.L. 1821 . Beschlufs der Nachrichten iiber de bieden brasilischen Reisenden, die herren Doctoren v. Spix und v. Martius. Flora 4:289-304. Phipps, J.B., K.R. Robertson, J.R. Roher, and P.G. Smith. 1991. Origins and evolution of Subfam. Maloideae (Rosaceae). Syst Bot. 16:303-332. Potter, D.,T. Eriksson, R.C. Evans, S.-H. Oh, J.E.E. Smedmark, D.R. Morgan, M. Kerr, K.R. Robinson, M. Arsen, CS. Campbell. 2007. Phytogeny and classification of Rosaceae. PI. Syst. Evol. 266:5-43. Rehder, A., C.A. Weatherby, R. Mansfeld, and M.L. Green. 1935. XXVI Conservation of later generic 1935341-544. ilts,TA Dickinson, and i Maloideae (Rosaceae). Syst. Bot. e among fruits of Maloideae (Rosa Robertson, K.R., J.B. Phipps, J.R. Roher, and P.G. Smith. 1991. A synopsis of g 16:376-394. Rohrer, J.R., K.R. Robertson, and J.B. Phipps. 1991. Varial 78:1617-1635. Rydberg, P.A. 1 908a. Notes on Rosaceae II. Bull.Torrey Bot. Club 36:397-407. Rydberg, P.A. 1908b. Tribe 4. Quillajeae. North Amer. Flora 22:259-261. Saville, D.B. 0. 1 979. Fungi as aids in higher plant classification. Bot. Rev. 45:377-503. Sax, K. 1 93 1 . The origin and relationships of the Pomoideae. J. Arnold Arbor. 1 2:3-22. Sax, K. 1 932. Chromosome relationships in the Pomoideae. J. Arnold Arbor. 1 3:363-367. Sax, K. 1933. The origin of the Pomoideae. Proc. Amer. Soc Hort. Sci. 30:147-150. Schlechter, R. 1 91 4. Die Orchideen 6:414. Schulze-Menz, G.K. 1964. Fam. Rosaceae. In: A. Engler, Syllabus der Pflanzenfamilien, II Band. Gebruder Borntraeger, ; Berlin. Pp. 209-218. v York Bot. Gard. 70:1-1 82. Stebbins, G.L 1958. On the hybrid origin of the Angiosperms. Evolution 12:267-270. J. Bot 53:951-960. Takhtajan, A. 1987. System of Magnolic Takhtajan, A. 1997. Diversity and class Takhtajan, A. 2009. Flowering plants, e I in the Rosaceae IX. Spiraeoideae; Quillajeae, Sorbarieae. A v. Press. New York. Thorne, R.F. 1992. An updated phylogenic < Weberling, F. 1989. Morphology of flowers c ssifi cation of the flowering plants. Aliso 13:365-389. i inflorescences. Cambridge Univ. Press, Cambridge. ILLUSTRATIONS AND STUDIES IN NEOTROPICAL ORCHIDACEAE. 5. THE LEPANTHES OVALIS GROUP (PLEUROTHALLIDINAE) WITH THREE NEW SPECIES FROM COSTA RICA 1 Diego Bogarfn Christina M. Smith Daniel Jimenez Journal of the Botanical Research Institute of Texas 6(2) Bogarin 2010); L. schizocardia Luer (Pupulin et al. 2010). Two other groups are under review: L. guatemalensis Schltr. (Pupulin & Bogarin in prep.) and L. minutilabia Ames & C. Schweinf. (Smith et al. in prep.). In this paper, we studied a group of species related to the Jamaican endemic Lepanthes ovalis (Sw.) Fawc. & Rendle, the first species of Lepanthes to be described and which was originally placed in Epidendrum L. (Swartz 1788). The other species of this group are L. erinacea Rchb.f., one of the first species described in Costa Rica (Reichenbach 1855), L. pan Luer & Dalstrom from Ecuador (Luer 1996), L. seegeri Luer from Panama (Luer 1987), L. trichocaulis Luer & R. Escobar (Luer 1988) and L. viridis Pupulin & Bogarin, recently published from Costa Rica (Pupulin & Bogarin 2011). All members share similar plant morphology, having ramicauls with long-ciliate sheaths with markedly dilated, long-ciliate ostia. The leaves are subcoriaceous, always light green without purple stripes or spots. The congested racemes arise under or above the leaves and are shorter than the leaves (rarely larger in some specimens of L. erinacea). The sepals are ovate, acute to obtuse, pubescent or glabrous. The upper lobes of the petals are wider and longer than the lower lobes. The lip is triangular, cor- date or subcordate, with the sides strongly folded and embracing the column at sides (rarely expanded), the blades are flattened with cuneate connectives, the apex is cleft leaving a narrow space between the apices with a pubescent appendix (Luer 1987; Luer 1996; Pupulin & Bogarin 2011). At the moment the group range in- ^ eludes Costa Rica, Panama, Colombia, Ecuador and Jamaica. We add three species proposed here as new to science. A key to the species of the group is provided. Additional material is provided online at www.epiden- E COSTA RICA. CArrAGO.Jin Plant epiphytic, caespitose, pendent herb, up to 15 cm long. Roots slender, flexuous, up to 1 mm in diameter. Ramicaul more or less descending, up to 10.5 cm, enclosed by 11-21 ciliate, lepanthiform sheaths, lightly cili- ate especially on new growth; the ostia markedly dilated, ciliate. Leaves coriaceous, green, elliptic, subob- ; ovate, acuminate with a short apiculus, 3.0-4.0 x 1.0-1.5 cm; cuneate base narrowing into a petiole ca. 3.5 mm long. Inflorescence racemose, distichous, successively flowered, borne above the leaf, shorter than the leaves, up to 3 cm, peduncle 1.5-2 cm long, rachis 0.5-1.0 cm. Floral bracts 1 mm long, ciliate. Pedicel 1.5 mm long, persistent. Ovary up to 1 mm long. Flowers light orange, yellow, red-orange and scarlet; sepals light orange; petals yellow with red-orange and scarlet; lip scarlet; column red. Dorsal sepal ovate, acute, connate to the lateral sepal for about 1 mm, 3.2 x 2.9 mm. Lateral sepals ovate to elliptic, acute, connate for about 1 mm, 2.5 x 2.5 mm. Petals essentially glabrous, transversely bilobed, entire, 0.8 x 2.4 mm; the upper lobes broadly un- cinate, rounded, overlapping, the apex broadly rounded; the lower lobes ovate, falcate, with apex broadly rounded, slightly smaller than the upper lobes. Lip bilaminate, adnate to the column, 1.8 x 1.3 mm expanded; blades narrowly oblong, pubescent, with narrowly rounded ends, falcate; connectives broadly cuneate, up to 1 mm long, perpendicular to the column and not leaving it exposed; body broadly oblong, connate to the base of the column; appendix narrowly oblong, pubescent. Column cylindric, 1 mm long; anther apical; stigma ven- tral. Pollinia two, ovoid. Anther cap cucullate. Distribution.— Only known from Costa Rica. Habitat and ecology. -Epiphyte in secondary premontane wet forest on the Atlantic watershed of Cordik lera de Talamanca, between 1000 and 1200 m of elevation. Bogarin et al.. New species of Lepanthes from Costa Rica 363 Drawn by CM. Smith from £ Pupulin 7477 (JBL-Spfrit). 368 Journal of the Botanical Research Institute of Texas 6(2) i. E) Up, spread. F) Pollinarium Bogarin et al„ New species of Lepanthes from Costa Rica apex. Dorsal sepal broadly ovate, obtuse or subacute, connate to the lateral sepal for about 1.3 mm, 3.3 mm x 2.8 mm. Lateral sepals ovate to elliptic, obtuse, connate for about 1.4 mm, 2.9 mm x 2.7 mm. Petals minutely pubescent, transversely bilobed, entire, 1.2 x 3.2 mm; the upper lobes oblong-obovate, truncate, not overlap- ping, the apex broadly rounded; the lower lobes oblong, obtuse, subfalcate, the apex broadly rounded, smaller than the upper lobes, lip bilobate, spreading, adnate to the column, 1.2 mm x 1.5 mm, expanded; the blades ovate, with rounded ends, falcate; the connectives cuneate, up to 0.5 mm long, perpendicular to the column leaving the appenndix exposed; the body oblong, connate to the base of the column; the appendix oblong, pubescent. Column cylindric, 1 mm long; the anther apical; the stigma ventral. Pollinia two, ovoid. Anther Distribution. — Only known from the southern Pacific region of Costa Rica. Habitat and ecology. — Epiphytic in tropical wet forest, premontane belt transition at 800 m of elevation along the Pacific watershed of Cordillera de Talamanca in southern Costa Rica. Etymology. — From the Latin expansus “spread” and labium “lip,” in reference to the spreading lip with the lobes not touching each other, leaving the appendix exposed. Phenology. — Plants have been recorded in flower in April. Discussion. — It is similar to L. erinacea but differs mainly by the pendent plants (vs. erect) with ramicauls with sparsely ciliate sheaths (vs. heavily ciliate), the petals with the lobes oblong, subsimilar (vs. the upper flabellate, the lower minute, oblique), the lower lobe yellow (vs. red), the lip spreading with the lobes separated and not touching each other, leaving the appendix exposed (vs. the apices of the blades approximate, hiding the appendix) (Fig.l). Other differences are summarized in Table 1. k Lepanthes ovalis (Sw.) Fawc. & Rendle, FI. Jamaica 1:71. 1910 . 1 As noted by Pupulin (2009), there is no way to assume that any of the specimens at BM is the holotype. The species should be lectotypified after a more detailed study of the available material; drawing of type, W). For a complete description and illustration see Luer (2009). Distribution. — Only known from Jamaica. Habitat and ecology. — Epiphytic in cloud forest along the Blue Mountains. Discussion. — Plants are recognized by the erect to suberect arching habit, the inflorescences developed above the leaf, the upper lobe of petals suborbicular, rounded, yellow with red-purple base, the lower lobe yel- low, pubescent and the inconspicuous appendix. 5. Lepanthes pan Luer & Dalstrom, Monogr. Syst. Bot. Mis UADOR. Imbabura: Los Cedros Reserve, wet forest near Rfo Los Cedi Ingram-FerrcU 2200 (holotype: MO). For a complete description and i Bot. Gard. 61:133-134, f. 178. 1996. Type ec- 1200-1300 m, 24 Mar 1996, 5. Dal strOm, S. Ingram & K. ■ation see Luer (1996). Distribution. — Only known from Ecuador. Habitat and ecology. — Epiphytic in wet forest at around 1200-1300 m of elevation. Discussion. — It is distinguished by the erect plants with inflorescences developed behind the leaf, the lateral sepals acute, pubescent, the upper lobe of petals with a lobule at the inner apical comer, the lower lobe long-ciliate, pubescent, the apex of the lip is shallowly cleft with apiculate apices in apposition with a small triangular appendix. 6. Lepanthes seegeri Luer, Orchidee (Hamburg) 38:60. 1987. Type: PANAMA. CHnuqut: epiphytic in forest above Boquete, alt. 1600-1700 m, Apr 1985, collected by H. Seeger 678A, flowered in cultivation at Heidelberg University 16 Oct. 1985, C. Luer 11424 (holotype: MO). For a complete description and illustration see Luer (1987). Distribution. — Only known from Panama. 370 Journal of the Botanical Research Institute of Texas 6(2) Habitat and ecology.— Epiphytic in cloud forest at around 1600-1700 m of elevation along the Cordillera de Talamanca in western Panama. Discussion .— It is recognized by the erect plants with inflorescences developed behind the leaf, the lateral sepals obtuse, glabrous the upper lobe of petals with a lobule at the inner apical corner, the lower lobe long- ciliate, pubescent and the lip with a long, slender, descending, pubescent appendix that protrudes from a shal- nsis Bogarin & D. Jimenez, sp. nov. (Figs. ID, 5). Type: COSTARICA. SanJose.- Dota, Sa i and Naranjillo de Tarrazu, 9°35'33.35"N 83°58'26.74"W, 1615 m, epiphytic, legit Daniel JimEnez, f Plant epiphytic, caespitose, erect herb, up to 14 cm tall. Roots slender, flexuous, up to 1 mm in diameter. Ramicauls erect to suberect, 3.5-9.0 cm long, enclosed by 7-10 ciliate, lepanthiform sheaths, ciliate especially ; on new growth; ostia markedly dilated, ciliate. Leaves coriaceous, green, elliptic to oblong, acute to acuminate Inflorescence racemose, distichous, successively flowered, beneath (rarely above) the leaf, shorter than the leaves, up to 2.5 cm long; peduncle 1.3 cm long; rachis 1.2 cm. Floral bracts 1 mm long, ciliate. Pedicel 2 mm long, persistent. Ovary up to 1 mm long. Flowers sepals light orange, petals and lip with the bases scarlet-red and the apices yellowish-orange, the column pinkish-purplish. Dorsal sepal broadly ovate, acute, glabrous, connate to the lateral sepals for about 1.2 mm, 3.7 x 3.6 mm. Lateral sepals broadly ovate, acute, glabrous, con- \ nate for about 1 mm, 3.9 x 2.5 mm. Petals transversely bilobed, 1.8 x 4.2 mm, slightly ciliate along the margins; 1 upper lobes oblong-trapeziform, overlapping, apex broadly rounded or truncate; lower lobes smaller than the upper lobes, oblong to narrowly triangular, oblique, apex rounded. Up bilobate, adnate to the column, 1.5 x 2.2 mm expanded; blades ovate, glabrous with ciliate, rounded apices, subfalcate; connectives cuneate, up toi 0.7 mm long, perpendicular to the column and not leaving it exposed; body oblong, connate to the base of the column; the appendix oblong, shortly pubescent. Column cylindric, 1.2 cm long; anther apical; stigma subapH Distribution .— Only known from the central-southern Pacific region of Costa Rica. Habitat and ecology.— Epiphytic in premontane and lower montane rain forest along the Pacific watershed of Cordillera de Talamanca. Etymology.— From the locality of Tarrazu in San Jose province. Tarrazu comes from an indigenous Hu- etar word of unknown meaning. Phenology.— Plants were recorded in flower from February to March. Discussion. It is similar to L. seegeri but differs mainly by the glabrous petals and the lip glabrous with ciliate apices (vs. long ciliate-pubescent), the upper lobe of petals truncate, without lobules at the inner apical corner (vs. oblong, rounded, with lobule at the inner apical comer), the petals and lip basally red, tinged with yellow-orange at apex (vs. completely red without stains), the blades of the lip touching each other above the column (vs. slightly expanded and not touching each other), the appendix shorter, a small triangular lobule, not extending far from the apex of the column (vs. appendix longer, descending, extending far from the apex of the column) (Fig.l). Other differences are summarized in Table 1. 8. Lepanthes trichocaulis Luer & R. Escobar, Orquideologia 17:224-226. ] & J. Posada at Colomborquldeas 26 Mar 1984, R. Escobar 3245 (holotyfe: MO). For a cc 1. Type: COLOMBIA. A Distribution . — Only known from Colombia. Habitat and ecology.— Epiphytic around 1820 m of elevation (Luer 1988). Discussion.— It is recognized by the erect plants, the inflorescences developed behind the leaf, the upper j 372 Journal of the Botanical Research Institute of Texas 6(2) lobe of petals oblong, rounded and not overlapping each other, the lower lobe longer or as long as the blades of the lip a couple of longer hairs in place of an appendix. 9 Lepanthes viridis Pupulin & Bogarin, Lindleyana in Orchids, Mag. Amer. Orch. 5 COSTA RICA. Turk, alba: Tuis, Cien Manzanas, 9°51'25"N 83=3347 AN ca. 3,1 17 ft (950 m), along a mi. premontane wet forest, secondary vegetation and remnants of primary, 26 May 2003, F. Pupulin 4801, G. Gerlach, P Kindlmann, H. Ledn-Patz, S. Pugh -Jones & E. Serrano (holxjtype: JBL). For a complete des Distribution . — Only known from Costa Rica. Habitat and ecology.— Epiphytic in premontane wet forest on the Caribbean slopes of Cordillera de Tala- Discussion.— It is superficially similar to L. pan, from which it mainly differs by the distinctly obovate leaves (vs. orbicular and shortly acuminate), the yellow flowers with bright green petals (vs. peach and red), the lateral sepals shorter than the dorsal sepal (vs. equal in length), the densely villose petals (vs. upper lobe pubes- cent, lower lobe pubescent on the outer half), with the upper lobe rounded (vs. truncate), and the apices of the lateral lobes of the lip acute (vs. obtuse). KEY TO THE SPECIES OF THE LEPANTHES OVALIS GROUP ;rthan the lobes of the lip, apex orange-yellow, lip red L. dikoensis Bogarin & C.M. Sm. (Costa Rica) is Pupulin & Bogarin (Costa Rica) " ACKNOWLEDGMENTS We acknowledge the scientific services of Ministerio del Ambiente, Energia y Telecomunicaciones de Costa | Rica (MINAET) and Sistema Nacional de Areas de Conservation (SINAC) for issuing the Scientific Passports under which wild specimens treated in this study were collected. The curators and staff at CR, 1NB and US| herbaria for granting access to the collections and for the facilities provided. Franco Pupulin, Jose F. Morales, Lisa Thoerle, and one anonymous reviewer, provided useful discussions and comments that improved the manuscript. We thank Adam Karremans for helping with fieldwork and literature. William Chacon of Orquideario Happy Garden, Coto Brus, Puntarenas kindly provided plant material and data. The present pa- per is part of the Project 814-A0-052, “Flora Costaricensis: taxonomia y filogenia de la subtribu Pleurothallidi- nae (Orchidaceae) en Costa Rica,” sponsored by the Vice-Presidency of Research, University of Costa Rica. J Bogarin et al.. New species of Lepanthes from Costa Rica 373 REFERENCES Barringer, K. 1 986. Typification of Schlechter's Costa Rican Orchidaceae. I. Types collected by A Brenes. Fieldiana, Bot. Bogarin, D. and M. Fernandez. 2010. Lepanthes arenasiana (Pleurothallidinae: Orchidaceae), a new species from Costa Rica. Lankesteriana 9:487-489. Luer, C.A. 1987. Vier neue Arten im Subtribus Pleurothallidinae: Lepanthes hubeinii, seegeri, rauhii und Platystele rauhii. Orchidee 38:58-62. Luer, CA 1988. Species of the genus Lepanthes from Colombia. Series 2-3. Orquideologla 1 7:145-230. Luer, C.A. 1992. New species in Lepanthes Sw. (Orchidaceae). Lindleyana 7:100-1 18. Luer, CA 1996. leones Pleurothallidinarum XIV. Systematics of Draconanthes, Lepanthes subgenus Marsipanthes, and subgenus Lepanthes of Ecuador (Orchidaceae). Part Three: The genus Lepanthes subgenus Lepanthes in Ecuador. Monogr. Syst. Bot. Missouri Bot. Gard. 61 :1-255. Luer, C.A. 2009. leones Pleurothallidinarum XXX. Lepanthes of Jamaica: Systematics of Stelis: Stelis of Ecuador. Part four Addenda, systematics of Masdevallia: new species of Lepanthes from Ecuador, and miscellaneous new combinations. Monogr. Syst. Bot. Missouri Bot. Gard. 1 15:1-265. Pupuun, F. 2009. Pupulin, F.Typi Swartziani Orchidacearum Indiae Occidentalis in Herbario Vindobonense conservandi. Ann. Naturhist. Mus. Wien 1 10B:21 3-247. Pupuun, F. 2010. Orchidaceae werckleanae: typification of Costa Rican orchid species described from collections by K. Werckle. Bot. J. Linnean Soc. 163:111-154. Pupuun, F. and D. Bogarin. 2010. Illustrations and studies in Neotropical Orchidaceae— The Lepanthes jimenezii group (Pleurothallidinae). Harvard Pap. Bot. 15:11 1-1 21. Pupuun, F. and D. Bogarin. 2011. Two new Lepanthes from Costa Rica. Lindleyana in Orchids, Mag. Amer. Orch. Soc. 80(3):1 78—1 81 . Pupuun, F., D. Bogarin, and C.M. Smto. 2010. Two new species of Lepanthes from Costa Rica close to L schizocardia (Orchi- daceae: Pleurothallidinae). Lankesteriana 9:423-430. Reichenbach, H.G. 1855. Symbolae Orchidaceae. Bonplandia (Hanover) 3(15-16)212-227. Reichenbach, H.G. 1 858. Lepanthes Sw. Xenia Orch. 1:151—1 52, 1 54. Tab. 49. Schlechter, R. 1923. Beitrage zur Orchideenkunde von Zentralamerika, II. Additamenta ad Orchideologiam Costaricen- sem. Repert. Sp. Nov. Reg. Veg. Beih. 19:3-307. Swartz, 0. 1788. Nova genera et species plantarum; seu, Prodromus descriptionum vegetabilium, maximam partem in- cognitorum quae sub itinere in Indiam Occidentalem annis 1 783-87 / digessit, Olof Swartz, M.D. - Stockholm, Upsala and Aboa: Bibliopolid Acad. M. Swederi. 374 BOOK REVIEWS Christian Ziegler. Introduction by Michael Pollan. 2011. Deceptive Beauties: The World of Wild Orchids. (ISBN-13: 978-0-226-98297-7, hbk.). The University of Chicago Press, Chicago, Illinois 60637, U.S.Af | (Orders: www.sinauer.com). $45.00, 183 pp., color photos throughout, 9 3 A" x 9 3 /4". The book jacket describes the author as “a biologist turned photographer specializing in tropical natural his- ] tory .” That is obviously a true statement— but it doesn’t go far enough. This is a person who greatly enjoys what I he’s doing and it is clearly reflected in the beauty, the layout, and the genuine “feel” — and love — of orchids. | Michael Pollan’s introduction, “Sex Among the Orchids,” definitely catches the reader’s attention with his opening statement: “We animals don’t give plants nearly enough credit. “ His following discourse is not only | fascinating and treated with fact, history, and humor, the reader is easily fully engaged — and can’t put the book The photography is outstanding. Ziegler’s accompanying chapters are well written, explanatory, and pro- vide exceptional background, diversity, descriptions of various habitats, atmospheric changes, uses, concerns, biological changes that influence the habitat environments, and, of course, effects of human behavior as well— Expect to take time to browse and enjoy this volume. It is really well done, and it makes one very aware of potential challenges in the upcoming years. — Helen Jeude, Volunteer and Assistant Editor, Botanical Research Institute of Texas, 1700 University Dr., Fort Worth, Texas 76107-3400, U.SA. Carol Grace with a foreword by Eric Lamont. 2012. Spring Wildflowers of the Northeast: A Natural History. 2012. (ISBN: 978-0-691-14466-5, hbk.). Princeton University Press, 41 William Street, Princeton, New Jersey 08540, U.S.A. (Orders: http://press.princeton.edu/titles/9668.html, 609-258-4884, 609-258- 6305 fax). $29.95, 290 pp., 512 color illus., T/i” x 10". What a welcome addition to wildflower literature! This is the ideal book for readers of all levels of expertise, from the professional botanist to the armchair naturalist. Princeton University Press affirms the value of Carol Gracie’s passion and attention to detail, her botanical research and travel, her years of teaching, her insightful writing, and 500 revealing photographs. The book first appears to be a gorgeous coffee table book, but then, I opened those first pages, and found myself dropping down the rabbit hole into Carol Gracie’s extraordinary botanical world rarely experienced by most people. Her teaching years at the New York Botanical Gardens, her intelligence and curiosity come alive to show the identification and natural history of a select group of wood- land wildflowers from baneberries to violets. She thinks of the question before you ask it. She touches on com- plicated topics of climate change, ethnobotany, horticulture, medicine and more, without being tedious. In addition, she refers to Indian lore, botanical literature and art. Carol Grade explains the reason for taxonomic name changes and adds an impressive list of references, which is testament to the extensive research to prepare this work. I have heard that some times when we look, we don’t see. Carol Gracie teaches us to see. I salute this gifted plantswoman for focusing on the beauty and intricacies of northeastern wildflowers, adding a valuable contribution to our knowledge of the flora of the United States.— Jane Austin Bruckner, graduate of Rutgers Uni- versity, New Jersey, a Registered Nurse, a Certified Master Gardener and BRIT Volunteer, 1700 University Dr., Fort Worth Texas 76107-3400, USA. CALATHEA BASIFWRA (MARANTACEAE), ANEW SPECIES ENDEMIC TO PANAMA Helen Kennedy Department of Botany and Plant Science University of California Riverside Riverside, California 92521, U.S.A ganders@mail.ubc.ca RESUMEN In the Flora of Panama (Woodson & Schery 1945), a total of 23 species of Marantaceae were listed with 14 in the genus Calathea. By 1972, Dressier (1972: 184) reported 35 species. Later, Kennedy (1976: 312-313) noted a total of 49 species and predicted an eventural increase to 60 or 70 because of the species richness found along the Rio Guanche(Col6nProv.) and the El Llano-Carti road (Panama Prov. and San Bias Prov.) and general inac- cessibility of those areas at the time. More recently, Kennedy (2012: 49) reported 63 species of Marantaceae from Panama. Currently 67 species (ca. 191% increase from the 1945 Flora of Panama treatment) are recog- nized with 49 in the genus Calathea. Eighteen species are recognized as endemic, including the one described herein plus two as yet undescribed taxa. Additional collecting in Costa Rica, but especially in adjacent Colom- bia, will reduce this number. TAXONOMIC TREATMENT Calathea basiflora H. Kenn., sp. nov. (Fig. 1) Type: PANAMA. San Bias. Rio PlayOn Chico, Campamento Nebba Dummat, Filo del Sureste 200 120 m 09°14'N 78°15'W, lOJul 1994, H. Herrera 1661 (holotype: PMA; isotypes: MO 6315278, PMA 75534). Calathea basiflora differs from C vmcunda in the wider leaf blades, 17.5-21.5 vs. 4-8 cm, the obtuse vs. acute to acuminate bract apex; from both C. verecunda and C. cleisthantha by the presence of 1-2 claviculate bracteoles vs. all bracteoles membranous; and from C. rhizanthouks by the larger leaf blades, 42-46 x 17.5-21.5 vs. 17-30 x 9.1-16.5 cm. two inflorescences per shoot vs. a single inflorescence, and the ratto of peduncle length to inflorescence length less than 1.2 vs. 3 or more. Plants rhizomatous, perennial, herbs, 40-60 cm; cataphylls herbaceous, narrowly ovate, minutely pilose, hairs ca. 0.2 mm, 3.5-5.5 on shoots bearing an inflorescence. Leaves all basal, 1-2 per shoot; leaf sheath not auriculate, appressed pilose, hairs more dense along margins of wings and toward base, base and intemode below, sericeous, ca. 40 cm; petiole with a distinct groove adaxially, scattered minutely pilose, hairs ca. 0.1 mm, 2.5-3 cm; pulvinus minutely tomentose adaxially, sides and abaxial portion glabrous, 1-1.5 cm; leaf blade soft, J78 Journal of the Botanical Research Institute of Texas 6(2) Etymology . — The specific epithet, basiflora, refers to the inflorescences borne directly from the rhizome at the base of the plant, well below the leaves. ACKNOWLEDGMENTS The Missouri Botanical Garden provided support for my accommodations while working in the MO herbari- um (organized, thanks to Olga Martha Montiel). The travel expenses for the trip to MO were provided by Fred Ganders. I am very grateful to Gerrit Davidse, Jim Solomon and Mary Merello for their help in the MO her- barium, to Barry Hammel, Isabel Perez and Mireya Correa for help with the Spanish resumen and to Heraclio Herrera for collecting this species. I thank Mireya Correa and an anonymous reviewer for helpful suggestions and corrections. The type scan was provided by the Missouri Botanical Garden. Dressler, R.L 1972. Terrestrial plants of Panama. Bull. Biol. Soc. Wash. 2:179-186. Kennedy, H. 1976. Notes on Central American Marantaceae II. New species from Panama and Costa Rica. Bot. Not 128312-322. Kennedy, H. 2012. Calathea rhizanthoides and C. peregrina (Marantaceae), new species endemic to Panama. J. Bot. Res. Inst. Texas 6:49-54. Woodson, R.E. Jr. and R.W. Shery. 1 945. Marantaceae. In: Flora of Panama. Ann. Missouri Bot. Gard. 32:81 -1 05. TWO NEW DISTICHOUS-BRACTED CALATHEA (MARANTACEAE) SPECIES FROM CENTRAL AMERICA Journal of the Botanical Research Institute of Texas 6(2) 2.2-2.36:1 in cauline leaves, ca. 17 lateral veins per 3 cm and 15-19 cross-veinlets per 5 mm (both measured at midpoint of sides of blade), adaxial surface green, glabrous, midrib minutely pubescent in basal half, hairs 0.1 mm, abaxial surface nearly glabrous, minute hairs to 0.2 mm along some lateral veins close to midrib, midrib below minutely tomentose along sides, hairs 0.1-0.2 mm. Inflorescences 1-2 per shoot, first one terminal, peduncle glabrous, 22-36 cm. Bracts 14-16, herbaceous, distichous, overlapping at base when live, rachis usu- ally not visible, broadly ovate in basalmost, transverse broadly elliptic in subsequent ones to broadly elliptic in uppermost, apex rounded with acumen in basalmost one, rounded to broadly obtuse in upper ones, margin straight (not recurved), 4-5.3 x 3.6-5.3 cm, each bract subtending 3 or more flower pairs, abaxial surface of bracts green, glabrous, adaxial surface glabrous; bicarinate prophyll membranous, rectangular-elliptic, apex rounded to subtruncate, glabrous, 3-3.2 x 1.0-1.5 cm, 0.9-1. 1 cm wide, carina to carina; secondary bract mem- branous, elliptic, apex rounded to minutely tridentate, glabrous, 2.6-3 x (0.85-) 1.25-1.5 cm; bracteole 1 per flower pair, medial, membranous, 1.1-1.35 x 0.15-0.2 cm. Flowers open spontaneously, pale yellow (fide Croat 16835). Sepals membranous, narrowly obovate, obtuse, apical margin incurved, glabrous, 18-20 x 3-4 mm. Corolla tube glabrous, 26-32 mm; corolla lobes subequal, elliptic, obtuse to 90°, glabrous, 12-14 x 3.5-5 mm. Staminodes 3; outer staminode spathulate, 9-10 x 6-8 mm; callose staminode rounded apically, ca. 11.5 mm; cucullate staminode 6-8 mm; stamen with lateral, petaloid appendage, 1 mm wide, anther 2.5 mm; ovary smooth, glabrous, ca. 2.5 x 1.5 mm. Capsule unknown. Distribution and habitat. — Calathea ravenii is known only from the type collection near Puerto Obaldia in wet tropical forest habitat. Considering the proximity of Puerto Obaldia to the border with Colombia, it is highly likely it will be eventually found there as well. It is apparently uncommon, as in two days collecting in July, 2005, in the vicinity of Puerto Olbalda, we failed to find any plants of it, even sterile ones, though another, previously unknown, undescribed species was found. Discussion . — Calathea ravenii belongs to Calathea G. Meyer section Calathea, having the characteristic habit of several basal leaves and a cauline leaf subtending the 1-several distichous-bracted inflorescences. It is distinguished from other Panamanian species of Calathea with distichous, complanate, bracts by the elliptic leaf blades, length 2.2-2.36 x width, the greenish, glabrous bracts with non-recurved apical margins, 4-5.3 x 3.6-53 cm, the single membranous bracteole per flower pair, and pale yellow flowers. A comparison of five of the distichous-bracted Panamanian species is given in Table 1 (Kennedy 2011: 202). It is readily distinguished from the other three distichous-bracted species not mentioned in the table, C. lasiostachya, C. caesariata and C. similis H. Kenn. by its glabrous, green bracts and the glabrous adaxial surface of the leaf blade. Etymology .— The specific epithet ravenii, is in honor of Dr. Peter Raven, director emeritus of the Missouri Botanical Garden. It is with deep gratitude that I have the privilege of naming a species in his honor as he was instrumental in my choosing the Marantaceae initially as an OTS field project and encouraging me to continue this for my PhD research, besides indirectly introducing me to my partner. I am glad to finally be able to thus acknowledge the debt of gratitude I owe for his help throughout my career, and that of numerous other botan- ists, in this mai Plants perennial, rhizomatous, caulescent, herbs, 2 or more m high. Leaves several, basal, and 1 cauline above an elongate stem intemode; leaf sheath not auriculate, green, appressed tomentose, hairs to 1 mm, very dense- ly tomentose to sericeous at very base in cauline leaves, 16-25 cm in cauline leaf; petiole green, appressed to- mentose in basal %, subglabrous to glabrous apically 47-60 cm in cauline leaf; pulvinus glabrous except for row of minutely tomentose hairs adaxially, hairs ca. 0.2 mm, glabrous near junction with petiole, 5-7 cm in 382 Journal of the Botanical Research Institute of Texas 6(2) Fig. 2. Calatheaoscariana H. Kenn. Holotype (A Vetiz&M. Veliz 94.3926, BIGU). I 384 Journal of the Botanical Research Institute of Texas 6(2) oscariana (Vdiz 99.7115, MEXU) and C. sclerobracteata ( Vdiz 99.7115B, BIGU), growing together under cultiva- tion, are the notes: “Planta cultivada dendro de cafetales o areas con banano, sus hojas son empleadas para envolver la masa de los tamales.” Likewise, such cultivation is noted on the type specimen, “Planta cultivada utilizada para la elaboration de tamales (Envoltorio)” and similar on Vdiz & Vdiz 94.3983B (BIGU). Interest- ingly, such use is not restricted to the campo, as I was served a tamale wrapped in the leaves of C. oscariana at the buffet in my hotel in Guatemala City. The portion of leaf was saved and pressed for later comparison. Etymology. — The specific epithet, oscariana, was recommended by Mario Veliz (BIGU), who collected the type and provided photos and scans of the new species, as a dedication to his friend Senor Oscar Archila Euler (+), “ la persona que la colectb y la tiene actualmente cultivada en su casa. . . .” ACKNOWLEDGMENTS I am especially grateful to Andrew Sanders of the UCR herbarium for allowing me the space to store and study all the Mesoamericana loan material and for taxonomic discussions. I thank the following for help and use of the herbarium facilities: Gerrit and Jeany Davidse (MO), Mario Veliz and Jorge Vargas (BIGU), Marie S. Cerma- kova (USCG), Mario Sousa, Gloria Andrade and Gerardo Salazar (MEXU), Christine Niezgoda (F) and Car- men Galdames (SCZ). I gratefully acknowledge the considerable help of Mario Veliz in providing color photos of the inflorescences of both C. oscariana and C. sclerobractea as well as photographs of specimens for com- parison. I am most grateful to Tom Croat for making the one and only (known) collection of C. ravenii. Teresa Salvato provided accommodations and transport for my work at UCR. I am deeply indebted to Fred Ganders for personally funding the cost of the herbarium visits and to the Missouri Botanical Garden for providing support for my accommodations while working in the MO herbarium (organized, thanks to Olga Martha Montiel). I thank Barry Hammel and an anonymous reviewer for help with the Spanish resumen and Gordon McPherson and the anonymous reviewer for their helpful suggestions and corrections. Thanks to the curators of F, MEXU, MO, PMA, and SCZ for the loan of their specimens. The type scan of C. ravenii was provided by the Missouri Botanical Garden and the scan of C. oscariana, by Mario Veliz (BIGU). REFERENCES Kennedy, H. 201 1 . Three new distichous-bracted species of Calathea (Marantaceae) from Panama. Novon 21 :201 -21 1 . Standley, P.C. and JA. Steyermark. 1952. Marantaceae. In: P.C. Standley and J.A. Steyermark, eds. Flora of Guatemala. Field Mus. Nat. Hist. Bot. Ser. 24:207-221. COLUMNEA ANTENNIFERA , A NEW SPECIES OF GESNERIACEAE FROM THE CORDILLERA CENTRAL OF THE COLOMBIAN ANDES Department of Biological Sciences Box 870345 The University of Alabama Tuscaloosa, Alabama 35487, U.S.A. Laura Clavijo Box 870345 The University of Alabama Tuscaloosa, Alabama 35487, U.S. lauriclav@gmail.com Key Words: Columnea, Colombia, Episcieae, G INTRODUCTION The genus Columnea L. is primarily epiphytic and belongs to the New World subfamily Gesnerioideae and tribe Episcieae. Columnea ranges from Mexico south to Bolivia and is most diverse in the northern Andes of Colom- bia and Ecuador. With over 200 species, Columnea is the largest genus in the subfamily Gesnerioideae (Burtt & Wiehler 1995; Weber 2004; Skog & Boggan 2006). The genus is distinguished from other closely related genera by an indehiscent berry instead of a fleshy bivalved capsule. Columnea antennifera J.L. Clark & Clavijo was discovered during a 2012 research expedition to the Co- lombian department of Antioquia in the Cordillera Central of the northern Andes. A remarkable character of Columnea antennifera is the presence of five elongate appendages near the corolla sinuses. The presence of co- rolla appendages and where they appear has been discussed in numerous artificial classifications of groups now recognized as Columnea, such as the section Ortholoma Benth. and the genus Trichantha Hook. (Morton 1963, 1971; Morley 1976; Smith 1994). Corolla appendages have not been thoroughly evaluated in a phyloge- netic context and most likely this character is convergent within Columnea. Columnea antennifera J.L. Clark & Clavijo, sp. nov. (Fig. 1) Tyre: COLOMBIA. Antioquia: Municipio Valdivia, Cordillera Central, road Ventanas to Briceilo, before the quebrada El Oro, 07°05'20*N, 75°29’20"W, 1802 m, 19 May 2012 01), JX. Clark, J. Anderson, L. Clavijo, M. Mazo & D. Suescun 13036 (holotype: COL; isoiytcs: BRIT, HUA, MO, NY, UNA, US). Obligate scandent epiphytic climber; stems elongate and horizontal, 2-3 m long, suffrutescent, glabrescent below, sparsely pilose above. Leaves opposite, strongly anisophyllous in a pair; larger leaf with petioles terete, 4-10 mm long, blade coriaceous when dry, elliptic to oblong, 3-12 x 1.3-3.4 cm, base rounded to oblique. 387 sometimes asymmetrical, apex acute, margin entire, adaxially shiny green, abaxially light green, sparsely pi- lose on upper surface and densely pilose on lower surface; smaller leaf greatly reduced relal nearly sessile, orbicular to ovate, 1-2.7 x 0.5-1.5 cm, base rounded to cordate, apex acute, m faces and vestiture similar to larger leaf. Flowers solitary and erect; pedicels 1.3-2 cm long, red, pilose; calyx 2-3.5 cm long, uniformly bright red, inside and outside sparsely pilose, inside lanate at the base, lobes 5, erect at anthesis, each lobe tightly appressed to adjacent lobe and folded lengthwise, ovate, apex broadly acuminate margin entire, 4 lobes nearly equal 0.7-1 x 1.2-1.5 cm, fused at the base for 1.5-2.6 cm, 5 th lobe (dorsal) slight- ly smaller, fused at the base for 1.1-1.5 cm; corolla tubular, posture erect in calyx, 3.0-4.5 x 0.6-1.0 cm; outside uniformly bright red and tomentose, internally glabrate, lobes bright yellow, appressed, 5-6 x 2-2.5 ovate; corolla appendages present in immature and mature flowers, located in each sinus alt* lobes, 14-20 mm long, pilose, bright yellow with or without a dark spot at the base; stamens 4, didynamous included; filaments 12-15 mm long, coiled after anthesis, connate and adnate to the base of the corolla tube, glabrous; anthers connate, longer than broad, 2-2.5 x 1-1.5 mm, dehiscing by longitudinal slits; staminode not seen; nectary a dorsal gland, glabrous; ovary superior, lanate, ca. 5 x 4 mm, style ca. 30 mm long, glabrous, stigma included and capitate. Fruits not seen. Columned antennifera is morphologically similar to C. dissimilis C.V. Morton (Fig. 2, A-D). These two spe- cies are easily differentiated by the elongate corolla appendages in Columned antennifera (Fig. 1) in contrast to the relatively short corolla appendages in C. dissimilis (Fig. 2A). The corolla appendages are developed in im- mature flowers of Columned dntenniferd (Fig. 1C). In contrast, the corolla appendages in Columned dissimilis are either absent or significantly reduced when the flowers are immature (Fig. 2B). Another species that has co- rolla appendages and is morphologically similar to C. dntenniferd is C.filomentosd (Figs. 2E, F). These two spe- cies are readily differentiated by the uniformly red corolla in Columnedfildmentosd (Fig. 2E) in contrast to the red corolla with bright yellow lobes in C. dntenniferd (Fig. 1). Vegetatively these two species are differentiated by the isophyllous leaf arrangement in Columnedfildmentosd in contrast to the strongly anisophyllous C. dnten- niferd. A third species that has corolla appendages is Columned rosed (C.V. Morton) C.V. Morton (Fig. 3E). How- ever, the calyx lobes of Columned rosed are deeply serrate to fimbriate (Fig. 3E) in contrast to the entire calyx margins of C. dntenniferd (Fig. 1). Distribution and habitat. — Columned antennifera is known from the northern Cordillera Central of the Colombian Andes in the department of Antioquia from montane forests (1800 m). Herbarium collections of Columned antennifera were not seen during recent visits to the National University of Colombia (COL) or the University of Antioquia (HUA), but an additional population was observed and photographed near the type locality between Yarumal and Ventanas during a 1996 field expedition by Gunter Gerlach from the Munich Botanical Garden (Botanischer Garten Munchen-Nymphenburg). Etymology. — The specific epithet, antennifera, refers to the resemblance of the elongate appendages at the apex of the corolla tube to insect antennae. Classification. — Columned antennifera appears to belong to section Ortholoma Benth. because of the pres- ence of corolla appendages and an obligate epiphytic habit. However, the traditional sectional classification of Columned is artificial and arbitrary. As an example, the section Ortholoma has been recognized at the generic level as T richantha Hook, by previous authors (Morton 1963; Wiehler 1973, 1975). A monographic revision of Trichantha by Morton (1963) was followed by another paper by the same author (Morton 1971) with a reduc- tion of all species recognized as Trichantha to Columned. The type species for Trichantha is Columned minor (Hook.) Hanst. and is characterized by the presence of appendages at the sinuses of the corolla (Fig. 3C, D). It is important to note that corolla appendages are not a unifying character for section Ortholoma or genus Trichantha. The type species for section Ortholoma is Columnea anisophylla DC., which lacks corolla append- ages as do many other species that have been assigned to this section. The traditional sectional classification of Columnea has been shown to be artificial because many sections do not represent monophyletic lineages (Smith 1994; Smith & Sytsma 1994; Clark et al. 2006). A revised sectional classification system based on mo- lecular sequence data is currently a collaborative research focus by numerous authors (e.g., James Smith, John L. Clark, Lacie Schulte and others). Journal of the Botanical Research Institute of Texas 6(2) 389 Fw - 3 Variation in corolla appendages present in Columnea. Aft B. (oiumnea coronata Amaya, LE. Skog ft LP. Kvist C ft 0. Columnea minor (Hook.) Hanst. E. Columnea rosea (CM. Morton) CV. Morton, f, Columnea mem (Wiehler) L.E. Skog & LP. Kvist (Photos A & B from/L Clark etal. 12990; C from J .L Clark etaL 10870; D from J.L Clark etal. 9647; E from J. Betancur 12394; F from J.L Clark etal. 7140). Corolla appendages vs. corolla lobes. — The presence of corolla appendages is widespread in Columnea and this character is often not accurately distinguished from corolla lobes. Various taxa from different sections have corolla appendages and this character is probably convergent within Columnea. For example, Columnea fi- lifera (Fig. 3F) has been assigned to section Collandra Lem. (Kvist & Skog 1993) and genus Dalbergaria Tussac. (Wiehler 1992; Kvist & Skog 2004) because of subsessile leaves, dorsiventral shoots, and a facultative epiphyt- ic habit. A recently described species, Columnea coronata Amaya, L.E. Skog & L. P. Kvist, was assigned to sec- tion Collandra (Amaya et al. 2004), where the specific epithet refers to a “corona” at the apex of the corolla tube. More accurately, the “corona” in Columnea coronata is homologous to reduced corolla lobes like those found in Columnea antennifera. Thus, what Amaya et al. (2004) referred to as “petals” in the description of Columnea coronata are actually appendages that appear petaloid (Fig. 3A) and what was described in Amaya et al. (2004) as the “corona” is homologous to reduced corolla lobes. Another species that has reduced corolla lobes and pet- aloid appendages is Columnea filifera (Fig. 3F). The petaloid appendages in Columnea filifera (Fig. 3F) are visible in the field and in photographs, but the corolla lobes are only visible with a hand lens or microscope. Phyloge- netic studies on the evolution of corolla appendages and their presence in numerous lineages of Columnea will play an important role in understanding their function, homology, and role in plant-pollinator interactions. ACKNOWLEDGMENTS This study was supported by funds from the National Science Foundation (DEB-841958 and DEB-0949169). We thank Christian FeuiUet (US) and William R. Anderson (MICH) for help in selecting the specific epithet; Alain Chautems (G), Laurence E. Skog (US), and Jeremy Keene (BHO) for providing helpful reviews of the manuscript. Our 2012 research expedition to Colombia was a tremendous success because of logistical sup- port from Alvaro Idarraga (HUA), Felipe Cardona (HUA), Julio Betancur (COL), Alvaro Cogollo (JAUM), and Diego Suescun (JAUM). We thank Gunter Gerlach from the Munich Botanical Garden (Botanischer Garten Munchen-Nymphenburg) for sharing his observations and images of Columnea antennifera. We gratefully ac- knowledge Norris H. Williams (University of Florida) for his carefully curated slide collection that included field images of voucher specimens (e g., Columnea filamentosa- Fig. 2 E, F) from a 1972 research expedition to Colombia with Hans Wiehler. We thank Julio Betancur from the Universidad Nacional de Colombia (COL) for providing images of Columnea rosea (Fig. 3E). w species of Columnea (Gesm Amaya, M., LE. Skog, and L.P. Kvist. 2004. Novae Gesneriaceae Neotropicarum XII: four n ceae) section Collandra from Colombia. Edinburgh J. Bot. 60:41 5-424. Burtt, B.L. and H. Wiehler. 1 995. Classification of the family Gesneriaceae. Gesneriana 1 *1-4 Cl* J.L.PS. Herendeen, L.E. Sk*. » EA 2-mer. 2006. Phytogene, ic relationships and generic boundaries in the Eph ^'^aasn^n’^n' 10 " ^ “"** SK,k>m ** 9 *" us Contr. Natt Morton, C.V. 1963. A revision of Trichantha (Gesneriaceae). Contr. U.S. Natl. Herb. 38:1-27. Morton, C.V. 1 971 . A reduction of Trichantha to Columnea (Gesneriaceae). Phytologia 22:223-224 Skog, LE. andJ.K. Boggan. 2006. A new classification of the Western Hemisphere Gesneriaceae. Gesneriads 56:12-17. ^4:1^9 yS,ema,ICS of Columnea SMion and section Srygnomhe (Gesneriaceae). Syst. Bot. Monogr. ,M4 ' MOleCUlK and m ° rPh0l09y: COn9nK "“ <* d “ (Gesneriaceae). PI. Syst familiM ^ r era ° f vascuia ' pian,s - f,mem 7:63 1 58 H mdC■ Bot Res. Inst Texas 6(2): 397 - 401 . 2 398 Fk. 1 . Senegaliapaganuccii Seigler, Ebinger, & Ribeiro. k. Flowering branch. B F. Flower at anthesis. G. Gynoecium showing detail of ovary. H. Prickles. I. Vegetative bud detail. J. Fruit. K. Detail of innt H-l from RM. Harley etal. 26439 (HUEFS); J-L from >1.4. Conceipoetal. 1925 (HUEFS). vein from the base, base oblique and truncate on one side, margins ciliate, apex obtuse, midvein submarginal Inflorescence a densely 40-75-flowered cylindrical spike, 9-15 x 20-45 mm, 1 to 3 in the leaf axils; peduncles 6-22 long, 0.4-0.6 mm thick, pubescent; receptacle not enlarged; involucre absent; floral bracts spatulate. 8-15i AACCM-MCGCGCGTCCMCTGGACG. . . . . . GCGCCCGAGGCCT- SCCMCM? ? ? TGCACCGCACGGAA Fk. 2. ITS sequence data for Senegalia paganuccii. Paraphyllidia " 1X0 1 ' Jlna estraaa Serrana e Mocambmho, 5 Sep 1974, M. Magalhaes & M.B. Ferreira 5021 (IPA). HAUL Sao Raimundo INonare- entre Sao Raimundo Nonato e Anisio de Abreu, caatinga, 17 Nov 1981 (fl.), M.R. DeVArco & E. Nunes 2214 (IPA, TEFB). Bentham (1842) in his discussion accompanying the description of Acacia velutina DC., first mentioned A monacantha Willd. and concluded that if it were a true Acacieae, it would belong to this tribe. Under A. velutina, Bentham (1842) cited Blanchet 2772 from Bahia and Pohl s.n. from Brazil, as possibly belonging to this species. However, our studies show that the type of Acacia velutina DC. (BRAZIL. RIO DE JANEIRO: [holotype: G, (photos F, MO, SI); isotype: G]), is distinct both from Blanchet 2772 (“habitat ad Utinga in deserto fluminis S. Francisco provinciae Bahiensis”) and the Pohl collection (“inter Praia et Bom Jardim Provinciae Minas Gerais”). Later, Bentham (1876) considered Blanchet 2772 (and another collection, apparently from the same location (Blanchet 3772), the Pohl collection and a Lindberg coUection (“ad Santos in provinceae Sao Paulo”) to represent Acacia monacantha Willd. Additionally, he considered Acacia velutina Benth. (1842) to be a synonym of A. monacantha Willd. Today these species are regarded as distinct, namely, Senegalia monacantha (syn. Aca- tina ), and the Blanchet and Pohl collections i da monacantha ) and Sengalia velutina (syn. Acatia velut to our new species, 5. paganuctii. The descriptions of Bentham (1875, 1876) and Lewis (1987) for Acada monacantha and the description and illustrations (p. 201, HI, H2, and H3) of Senegalia monacantha of de Queiroz (2009) all appear to be based on Blanchet 2772 and similar collections and not on the original type material of Acacia monacantha Willd. We The authors wish to thank several colleagues for advice concerning questions of nomenclature and general taxonomic advice, in particular, K.N. Gandhi. We wish to acknowledge support by the National Science Foun- dation (NSF DEB 04-15803), by the American Philosophical Society (1992) and by the CNPq for financial sup- port to PGR, on her master’s degree in the Postgraduate Program in Botany at the Universidade Estadual de Feira de Santana, Bahia, Brazil; also to the curators of the herbaria that were visited or provided loans of speci- mens for our study (ALCB, BHCB, CEN, CEPE C, F, G, H, HRB, HST, HUEFS, IBGE, IPA, K, MBM, MO, NY, RB, R, SP, SPF, TEX, UB). We thank Bruce R. Maslin and an anonymous reviewer for helpful comments. REFERENCES Bentham, G. 1842. Notes on Mimoseae, with a synopsis of species, London J. Bot. 1:318-392, 494-528. Bentham, G. 1875. Revision of the suborder Mimoseae. Trans. Linnaean Soc London 30:335-664. Bentham, G. 1876. Leguminosae. In: C.F.P. von Martius, Flora Brasiliensis 15(1 &2). Munich and Leipzig. 1-527 pp., 1-138 Queiroz, LP., de, 2009. Leguminosas da Caatinga. Universidade de Feira de Santana, Feira de Santana. Lewis, G.P. 1 987. Legumes of Bahia. Royal Botanical Gardens, Kew, England. appears to be most closely related to S, tenuifolia (L.) Britton & Rose. The characters used to distinguish Senega- 402 Journal of the Botanical Research Institute of Texas 6{2) BOOK REVIEW WelbyR. Smith. Illustrations by Vera Ming Wong and Bobbi Angell. 2012. Native Orchids of Minnesota. (ISBN- 13: 978-0-8166-7823-5, hbk.). University of Minnesota Press, 111 Third Ave. South, Suite 290, Minne- apolis, Minnesota 55041, U.S.A. (Orders: www.upress.umn.edu; 612-627-1980 fax; 612-627-1970 phone). $34.95, 288 pp„ 93 b/w illustrations, 174 color plates, 53 maps, 7" x 10". Nineteen years after publishing Orchids of Minnesota, the author has published an updated, comprehensive, and beautifully illustrated new volume. Native Orchids of Minnesota. His Table of Contents, map of the counties in Minnesota where orchids are most likely to be found, and Preface provide an excellent “starting platform" for the Introduction and his gentle guidance into the descriptions and basic understanding of Minnesota’s or- chids. After providing straightforward answers to the usual questions flower lovers ask, he provides a thorough and easily useful “Key to the Genera of Orchids Found in Minnesota.” An unusual but very helpful addition to the keys is the author’s careful inclusion of two or three simple line drawings of the described plant part at the end of each leg of the key. The reader has an immediate and accurate vision of where and what to look for as (s) he examines the plant. Each genus is introduced on the left page and accompanied by a photo on the right page. The following pages provide thorough descriptions, explanations, photos, and illustrations of the individual species in each genus. Most orchid fanciers will find themselves totally enchanted by the “reading and admiring” process. Obviously, this book would be incredibly useful to have with you in the field — not only for identifications but also to have use of the accompanying maps of Minnesota counties, showing where each species is most likely to be found. (This is such a beautiful book, I think I would need to buy two copies: one to take into the field for quick and accurate identification and one to keep safely at home in an easily available, beautiful “pristine condi- tion!”).— Helen Jeude, Volunteer and Assistant Editor, Botanical Research Institute of Texas, 1700 University Dr., Fort Worth, Texas 76107-3400, U.SA. TAXONOMY OF LANTANA SECT. LANTANA (VERBENACEAE): II. TAXONOMIC REVISION Roger W. Sanders Bryan College It 7802 721 Bryan Drive Dayton, Tennessee 37321, U.S.A. RESUMEN Due to a long history of cultivation, hybridization, and invasiveness, the taxonomy of Lantana L. sect. Lantana resists partitioning into easily identified species (see Sanders 2006 for review). While some workers might prefer the convenience of recognizing a single highly variable species, Lantana camara L., previous biosyste- matic studies (Sanders 1987a, 1987b, 1987c, 1989) have shown the presence of morphologically discrete dip- loid taxa having coherent ecological and geographic ranges where they appear to have speciated in situ. These studies have been corroborated by a recent molecular analysis of the taxa in Florida (Maschinski et al. 2010). The present study attempts to delineate the indigenous taxa of sect. Lantana, even in the face of rampant hy- bridization due to human-induced ecological disturbance and the failure of odd polyploidy as a breeding bar- rier in this group. This second paper in the series builds on the first (Sanders 2006), which detailed the typifica- tion of species of sect. Lantana. As suggested by Sanders (2006), the identity of individual specimens constitut- ing the hybrid plexus found growing outside cultivation today cannot be unraveled by morphology alone, and it may be recalcitrant even to molecular genome analysis. Thus, variation encompassing the indigenous species now connected by hybrids may appear to be constituted more of adaptive peaks rather than bell curves sur- rounded by discontinuity. Furthermore, no phylogenetic analysis has been attempted here because the out- group relationships of sect. Lantana are not understood, and the significant trichome and inflorescence char- acters are homoplastic with regard to potential outgroups. The rank of series is established only for grouping species of presumed origin by divergence from the an- cestor of sect. Lantana. Species of interseries hybrid origin are not placed into series and are listed separately. These species are presumed to have originated by natural selection acting on the variable pool of original hy- brids resulting in one or a few closely similar phenotypes and, thus, may not be strictly intermediate to the Journal of the Botanical Research Institute of Texas 6(2) parental species. The surviving phenotype has become self-propagating and has attained a geographic range exceeding the original area of sympatry. Spontaneous and cultivated hybrids that have received Latin names but do not behave biologically as species are in a third separate list. 1 view varietas as the least inclusive taxonomic rank composed of a minimum of one breeding population | (as inferred from available ecological data) having geographic coherence in a limited part of the species range and imperfect discontinuity from similar, geographically adjacent taxa within the species. Subspecies is used either to group varieties or to recognize a taxon within a species with geographic coherence over an extensive geographic range (e.g., usually several islands or subcontinental areas) and having imperfect discontinuity orl minor differences from similar, geographically adjacent subspecies. One impetus to employ subspecies in Lon - 1 tana has been to avoid instability in infraspecific names that could be caused by the subsequent discovery of i poorly known varietal names, of which there are many. MORPHOLOGICAL TAXONOMIC CRITERIA AND ANALYTICAL CAVEATS Caveats for identification and descriptions are given in italics. Prickles . — Whereas a majority of species either lack prickles or bear only small weak straight or recurved | prickles, pronouncedly stout recurved prickles are inconsistently present (varying among herbarium collec- tions and field populations) in the remaining species, notably Lantana camara subsp. aculeata, L. hirsuta subsp. 1 amazonica, L. horrida, L. nivea, L. planaltensis, L strigocamara, L. urticoides, and L. viscosa. The tendency to 1 produce prickles appears more pronounced in hybrids than in most indigenous species. Trichames . — The form of trichomes on the abaxial leaf surfaces (filiform vs. setiform vs. strigiform) and their length are highly correlated with ecological and geographic coherence of indigenous taxa and provides | can be nearly twice as long as those on the remaining tissue. Therefore measurements in the key and descriptions for I adaxial hairs are taken between the secondary veins. Likewise, hairs on the nodal lines of the stems are often about twice as long as other hairs along the stem and are excluded from measurements in the key and descriptions. Filiform hairs and setae are both erect from the base with the setae differing primarily by greater length and stouter, and a more conical proximal portion. Both types may be somewhat flexuous, arching or curly 1 Strigae are stiff conical hairs that are geniculately bent in the proximal quarter or third with the remain- ing distal portion directed antrorsely. On the adaxial leaf surfaces, the antrorse portion is more or less ascend- 1 ing and arching. The broadened base emerges from a buttressing ring of epidermal cells that form a pustulate base. Especially in Ser. Strigosae or taxa of hybrid origin with genes of its species, the strigae often are decidu- 1 ous leaving the pustulate bases as rough points. In some species the bases enlarge with age and become vitre- J ous (clear or white). In taxa and hybrids with strigae on the abaxial surface, the strigae lack the buttress base, arising directly from the epidermis and the antrorse portion is held more or less parallel to the epidermis. In 1 some Lantana nivea, they are so short that the antrorse portion is not well developed, appearing as a short point 1 angling upwards. In recognizing the abaxial strigae, one must also be aware that filiform or setiform hairs that are crushed against the surface during pressing can be mistaken for strigae, which occur consistently over the pertinent j The co-occurrence on the abaxial surfaces of filiform hairs or setae with stn brid nature or heritage of the specimen at hand. The presence of stipitate glands on the twigs, peduncles, petioles, and ei several taxa, noteably Lantana camara subsp. portoricensis, L. horrida subsp. 2 crantha, L. paraensis, and L. planaltensis. While glands are consistent in L. le indicat 1 of the hy- ■af-blades is variable within ii and subsp. tiliifolia , L. mi- 1 the basis of glands. Therefor it presence of glandular ha iara is simply variable in tl nages, with the widest poi alf, making it difficul shape. Bases of the blades in almost all cases abruptly taper to a narrow wing onto the petiole distally. Leaf- blades are considered to be triplinerved (as opposed to pinninerved) if the basal pair (or two pairs) of secondary veins among them. Nigrescence refers to a distinct blackening of mature leaves occurring during drying for preservation, not normal senescence. While such blackening is diagnostic for certain species, newly emerging leaves can blacken in most species. Leaves atypical for size, shape, bases, apices, and venation are present on most plants. Leaf shape, size, and vestiture traits are measured only on fully developed, non-senescent leaves. Measurements for marginal teeth are taken mid-margin, avoiding the reduced teeth near the base and apices of the blades. Inflorescences.— The basic structure of the inflorescences has been discussed in detail (Sanders 2001). Peduncle length in the key and descriptions are given for fully opened inflorescences and infructescences. Bracts generally decrease gradually in length and width from the proximal to the distal series that seem- ingly spiral up the receptacle. The distal (inner) bracts are about 2-4 mm long and about 0.3-1 mm wide in most species and, thus, are not detailed in the descriptions. The exceptions appear to be diagnostic in Ser. Spi- catae and a few taxa in Ser. Lantana in which almost all the bracts are the same dimensions. In some taxa there is an abrupt diminution from the proximal (basal or outer) two or three series of bracts to the more distal series. Shape and size of the proximal bracts appear to be consistent and diagnostic, with the exception of one or rarely two subfoliaceous bracts that develop sporatically in almost any taxon; hence, these atypical bracts are exclud- ingfrom the measurements. Flowers . — Corolla color has been discussed by Sanders (2001, 2006). It often changes from bud to early opened flowers to late flowers to fading flowers, especially in plants that produce both yellow to carmin pig- ments and purplish pigments. The throat is often not only different but changes during flower maturation. This developmental variation is often further complicated by intraspecific variability. Unfortunately, detailed information is usually lacking from collection labels. Corolla shape is nearly uniform in the group, but size appears to be consistent within taxa when measured from fresh material. However, dried corollas are often shrivelled and difficult to measure; in the descriptions, “fresh” size has been extrapolated from dried specimens. Cytology. — Chromosome numbers are not given in the descriptions because those for only a few taxa are known (See Sanders 1987a, 1987b, 1989). Phenology . — Flowering time is not given in the descriptions because any species can flower anytime dur- ing the year whenever moisture is available. This is true even of species l seasons when the species are grown in frost-free areas. TAXONOMIC TREATMENT See Sanders (2001) and Sanders (2006) for further characterization of Lantana and comparison of sect. Lantana with other sections. Also, see Sanders (2006) for details of species typifications, which are supplemented here only as needed. Please note that, in the type and other specimen citations, the abbreviation “di” refers to a digital image made available online or as a courtesy by the cited herbarium. Many thousands of specimens representing this group are in herbaria awaiting identification. For this study only a small sample, primarily from major U.S. institutions, has been selected for annotation and citation here as these specimens will be most easily available for consultation by other professionals. Even so, annotations made during quick visits to her- baria may differ than those cited herein as a result of reflection and more careful study of digital images that I made or were sent to me. Selection of specimens was to establish only distribution limits of the species, as well as document as many hybrid combinations as possible, thus, resulting in a falsely apparent predominance of hybrids in some cases. To assist those attempting to identify specimens of sect. Lantana, a richly illustrated interactive key (in which vernacular names are also discussed and provided) has been made available online (Offutt & Sanders, 2012). Lantana L. sect. Lantana Shrubs or rarely treelets, erect to decumbent or subscandent, height (or length) (0.1-)0.5-3 m (to 4 or even 6 m m subscandent, especially aggressively naturalized forms); the internodes usually less than to almost twice as Journal of the Botanical Research Institute of Texas 6(2) long as leaves (mostly twice to thrice as long in L. splendens), with or without weak to stout, conical to recurved prickles; vestiture antrorsely strigose to puberulent, pilose, setose, glabrescent, or stipitate-glandular and thus markedly viscid, the trichomes of twigs, peduncles, and petioles often noticeably longer and stiffer than those on remaining herbage. Leaves opposite or sporadically ternate, petiolate, simple; blades usually ovate or lan- ceolate to elliptic, usually hardly to moderately rugose, i.e., puckered between tertiary veins (strongly so in some L. horrida and usually bullate in L. leonardiorum, which is puckered between the secondary veins), usu- ally longitudinally flat or somewhat undulate (incurved in L. depressa ); base attenuate to cordate; apex attenu- ate, acuminate, acute, obtuse, or occasionally rounded; margin usually finely serrate-crenate but coarsely so in L. urticoides and some L. hirsuta and L. kingii or subentire in some L. cujabensis, flat to revolute, usually green (often purple-tinged in L. kingii); adaxial surface strigose (strigae typically ascending distally to antrorsely bent, ± appressed in L. kingii, flaccid and strongly appressed in L. hodgei), strigose-villous, setose-villous, or j nearly glabrous; abaxial surface strigose, pilose, puberulent, setose, or glabrescent, with the veins green to pale brown or sometimes nigrescent or occasionally tinged with purple (frequently purplish in L. kingii). Inflores- cences pedunculate, capituliform spikes, one (or sporadically two in several species) per subtending leaf; pe- duncles about a third to twice the length of leaves (up to four times in some L. horrida); axis (common recepta- cle) ellipsoid or fusiform, spongy; bracts nearly always subtending a flower, linear triangular or linear lanceolate to oblong, elliptic, or spatulate, ± appressed to spreading or reflexed. Flowers in several series, two to three series in anthesis at a time, zygomorphic; corolla salverform with inflexed tube and four unequal lobes, pig- ments either yellow to reddish or pink to purple or admixtures of both (in hybrids or taxa of hybrid origin) or lacking. Drupes usually blue-black (but sporadically described by collectors as dark violet-purple), usually with a metallic iridescence; pulp watery-mealy; endocarp turbinate-obpyriform with an inflated commissure and external circumferential ridge below the seed chambers. x=ll. 407 409 broadly ovate or oblong-deltate to elliptic-lanceolate, (l-)3-8(-16) cm long, the length (0.9-)1.5-2.5 x width, usually not nigrescent, papery, pinninerved; base subcordate, truncate, rounded or broadly cuneate, usually very briefly, narrowly cuneate onto petiole at very base; apex acute to acuminate, occasionally attenuate or rounded; marginal teeth 6-35(-50) per side, rounded to acute, spreading to appressed, sometimes with tips recurved, with sinuses 0.2-2 mm deep; adaxial surface dull, antrorsely strigillose to strigose-pilose or with stipitate glands mixed in, the hairs occurring on veins and intervening tissue, thin canopy of hairs only 0.2-0.5 mm (occasional hairs 0.7 mm in subsp. aculeata ) with understory of shorter hairs not well developed, 10-90(- strigae ca. 0.1-0.2(-0.3) mm in diam.; abaxial surface duller green than adaxial surface, moderately densely (occasionally sparsely) pilose, the hairs on all veins and intervening tissue, 0.2-0.5 mm, all about same length, (10-)40-250/sq. mm. Inflorescences remaining hemispheric; peduncles 0.3-2 x leaf length. Proximal bracts above middle, with 3 veins from the base, appressed or spreading, deciduous after flowering; apex attenuate to rounded; indument pilose to strigillose, sometimes stipitate-glandular, somewhat or not ciliate, the longest hairs £ 0.5 mm. Corolla yellow to or aging reddish orange (infused with pink or purple in subsp. aculeata), rarely white; corolla tube 4-12 mm. Distribution and habitat. — Mexico, Central America, West Indies, and northern South America; cultivat- ed and escaped pantropically, especially in Australia; disturbance openings in tropical evergreen and decidu- ous forest, open pine forest, thorn shrubland, savanna; 0-2000 m. Lantana urticifolia Mill., Card. Diet. ed. 8, Lantana 5. 1768. Lanu (misapplied to L. camara subsp. aculeata). Type: MEXICO. > un s.n.. Herb. Sloan 6:84 (lectotype: BM- id Virgin Is 412 Journal of the Botanical Research Institute of Texas 6(2) Stems usually without prickles or with few weak, straight ones; twigs, peduncles and often petioles moder- ately to densely covered with usually ascending, soft to somewhat stiff, curled or straight hairs, the hairs 0.1- ; 0.5(-0.7) mm, mostly ca. 0.3 mm. Leaf-blades broadly ovate to oblong-deltate to elliptic lanceolate, widest usually in or near proximal third, sometimes near middle, (1.5-)3-7(-9) cm long, the length (1.1— )1.3— 2 x| width; marginal teeth 20-35(-50) per side, rounded, obtuse, or acute, often appressed, with sinuses 0.3-0.8(-| 1) mm deep; adaxial surface antrorsely strigillose to strigose-pilose, the hairs mostly about 0.3 mm or less. Peduncles 0.5-1.2 x leaf length. Bract series gradually reduced in size; proximal bracts linear-oblong, oblan- ceolate-oblong, linear-lanceolate, or linear-triangular, 4-8 mm long, 0.5-1.5 mm wide, widest near the base or the outermost series sometimes widest above middle (if 4 mm or less long, then widest near the base); apex acute to attenuate. Corolla yellow to or aging reddish orange; corolla tube 7-12 mm; corolla limb 6-9 mm in a subsp. glandulosissima (Hayek) R.W. Sanders, comb. & stat. i epert. Spec. Nov. Regm Ytg. 2 IM l'H'o Tm: MEXICO. Jalisi .. Tequila 2 Jul ilO!, NY!, P[2,di!]). Stems usually without prickles or with few weak, straight to recurved ones; twigs, peduncles and often peti- oles densely covered with stipitate glands or also with eglandular filiform hairs mixed in, the hairs (and glands) 0.1-0.5 mm, mostly 0.2-0.3 mm. Leaf-blades broadly ovate or broadly elliptic to oblong-lanceoate or elliptic-lanceolate, widest usually near proximal third or middle, (l-)4-10(-16) cm long, the length (1.2-)l-jB 2.1 x width; marginal teeth 10-30(-45) per side, usually rounded or obtuse, usually spreading, with sinuses (°.4-)0,7-1.5(-2) mm deep; adaxial surface mixed antrorsely strigillose to strigose-pilose and stipitate-glandu- lar, the hairs 0.1-0.5 mm. Peduncles 0.5-1.8 x leaf length (often almost doubling in length in fruit). Bract se- nes gradually reduced in size; proximal bracts oblanceolate-oblong (rarely obovate) to triangular-oblong or ! Sanders, Taxonomy of Lantana sect. Lantana 413 linear-lanceolate, (2.5-)4-8 mm long, 0.8-1.7(-2) mm wide, widest above or near the middle or near the base; apex acute to attenuate, often rounded at very tip. Corolla yellow to or aging reddish orange, rarely white; co- rolla tube (5-)7-12 mm; corolla limb 6-9 mm in diam. Distribution and habitat . — Mexico (northwestern, central, and southern) and Central America to north- ern Colombia and Venezuela; open pine-oak forest, thorn and tropical deciduous shrubland and woodland, and savanna, especially in disturbance openings; 0-2000 m. Lantana camara subsp. glandulosissima differs from subsp. camara only in the strong development of stipitate glands in place of filiform hairs on twigs, peduncles, petioles, and leaf surfaces and in the longer bracts and corollas. Because the development of glandular hairs is variable within several other taxa in sect. Lantana, parapatric or narrowly sympatric in the vicinity of Veracruz (as evidenced by the intermediate or hybrid specimen, Gilly et al. 75, MSC), perhaps due to human activity. Although no specimens of subsp. glandulosis- sima from Veracruz came to my attention, interspecific hybrids (see section below) further evidence its pres- ence there. Furthermore, at least in Bocas del Toro Prov., Panama, subsp. glandulosissima intergrades with subsp. moritziana ( Peterson &Annable 868 , MO). Lantana camara subsp. glandulosissima is broadly sympatric with L. horrida. The two “pass the test of sympatry” (Stebbins 1966, p. 95-96) despite occasional hybrids that are probably limited to disturbed areas. I take this as evidence that L. camara and L. horrida are distinct. On the other hand, if one considered the differ- ences in length of the adaxial leaf-surface trichomes an inadequate species criterion and submerged L. horrida 414 Journal of the Botanical Research Institute of Texas 6(2) If. Lantana camara subsp. aculeata (L.) R.W. Sanders, Sida 22:394. 2006. Bask Lantana sanguined Medik., Hist. & Commentat. Acad. Elect. Sci. Theod.-Palat. 3. Phys. 229. 1775. Camara aculeata (L.) Kuntze [var. subinermis Kuntze] f. sanguined (Medik.) Kuntze, Revis. Gen. PI. 2:503. 1891. nom. illeg. (see synonyms below). Lantana aculeata L r. Stirp. Chap. Allerton. 1C Ecole Bot., ed. 3 (Cat. PI. I n. illeg. Type: None selected. Stems usually with stout, recurved prickles, often abundant; twigs, peduncles and often petioles moderately covered with antrorse to ascending or retrorse, curled or straight hairs or also stipitate glands, the hairs 0.1-0 7 f mm. Leaf-blades broadly ovate or oblong-deltate to elliptic lanceolate, widest usually in or near proximal third, sometimes near middle, 3-9 cm long, the length (l.l-)1.3-2 x width; marginal teeth 10-30(-45) per side, usually acute or obtuse, sometimes rounded, usually spreading, with sinuses 0.5-2 mm deep; adaxial surface antrorsely strigillose to strigose-pilose, the hairs 0.1 -0.5 mm (occasional ones to 0.7 mm). Peduncles 0.5-1.2 x leaf length. Bract series gradually reduced in size; proximal bracts linear-oblong, oblanceolate-ob- long, linear-lanceolate, or linar-triangular, 4-8(-10) mm long, 0.5-1.5 mm wide, widest near the base or proximal third, sometimes the outermost one or two slightly broader above middle; apex usually attenuate. Corolla yellow to or aging red-orange and usually infused with purple or opening pink aging to deep reddish purple; corolla tube (5-)7-12 mm; corolla limb 6-10 mm in diam. Distribution and habitat . — Historically cultivated worldwide and escaped pantropically, especially com- mon in Africa and Australia; disturbance openings in tropical evegreen, deciduous, and thorn forest and sa- vanna; 0-2000 m. Selected specimens examined: AUSTRALIA. Queensland: Day 8 (BRIT). KENYA. Taita Taveta: Wakanene et al. 383 (MO). ZAIRE. Haut- Katanga: Fabri 60415 (MO). Presumed hybrids with: 9i-cvx20. L. Callowiana Hybrid Group cultivars (L. depressa-tetraploid x strigocamara). AUSTRALIA Queensland: Day 64 (BRIT); Riding 76 (BRIT); Robazza & McAndrew 17 (BRIT). 12b. L. nivea subsp. mutabilis. AUSTRALIA. New South Wales: Day 38 (BRIT); Day 42 (BRIT); Day 71 (BRIT). Queensland: Hannan-Jones 29 (BRIT); McAndrew 48 (BRIT); McAndrew 81 (BRIT); McAndrew 84 (BRIT). CHINA. Guangdong: Deng Lang 10459 (BRITldi]). RHODESIA. Guruve: Nyariri 167 (MO). TANZANIA. Kiliman- jaro: Mlongwa et al. 459 (BRIT). Tanga: Mwongofea & Kayombo 113 (MO). U.S.A. Hawai i. Oahu: Degener 11467, identification uncertain (SMU). North Carolina. Forsyth Co.: cult,. Schallert 1352 (SMU). 20. L strigocamara. AUSTRALIA. Queensland: Day 69 (BRIT). TANZA- Shrubs erect, rounded, lax, or trailing, dense to open; stems 0.5-3 m; branches ascending and normally sev- eral to decumbent and few; twigs, peduncles and often petioles puberulent, setose, glabrescent, or stipita te ' glandular, the hairs 0.1-1.5 mm. Leaf-blades broadly ovate to rotund, elliptic or lanceolate-deltate, (0.5-)l-9(" ; Sanders, Taxonomy of Lantana sect. Lantana 12) cm long, the length 1-2.2 x width, not nigrescent, papery, pinninerved; base usually rounded to truncate, sometimes broadly cuneate or cordate, briefly narrowly cuneate onto petiole at very base; apex acuminate, acute, obtuse, or rounded; marginal teeth (4-)6-25(-45) per side, acute, obtuse, or rounded, spreading or ap- pressed, then sometimes with tips recurved, with sinuses 0.2-2.5 mm deep; adaxial surface dull, antrorsely dense canopy of hairs 0.6-0.8 (-1) mm with understory of hairs 0.2-0.5 mm, (5-)10-50(-150)/sq. mm, not noticeably vitreous-pustulate, the circular bases of the strigae ca. 0.1-0.2 mm in diam.; abaxial surface dull green, moderatly densely (occasionally sparsely) pilose, the hairs on all veins and intervening tissue, 0.3-0.5 mm, all about same length except for a few scattered arching hairs 0.7-1 mm on the midrib or secondary veins, 10-200/sq. mm. Inflorescences remaining hemispheric; peduncles (0.5-)0.8-4 x leaf length (usually about equalling to almost twice when mature). Proximal bracts lanceolate-triangular, lanceolate-linear or narrowly elliptic, narrowly oblanceolate, narrowly oblong to oblanceolate-spatulate, 2-12 mm long, 0.5-3 mm wide, widest in proximal, middle, or distal third, with 3(-5) veins from the base, appressed or spreading, persisting or not; apex acute, attenuate or obtuse to rounded; indument strigose-pilose or setose, often sessile- or stipitate- glandular, ciliate or not, the longest hairs mostly 0.3-1 mm. Corolla yellow to or aging reddish orange; corolla tube 4-12 mm. Distribution and habitat .— Mexico and West Indies to subtropical South America; tropical savanna with gallery forest, montane humid, pine, or dry forest, and disturbed successional woodland, shrubland and grass- land; 0-2500 m. Journal of the Botanical Research Institute of Texas 6(2) 45:296. 1980. Type: CUBA. Isle of Pines: Siguanea^ 21 May 1910 Jennings 458 (holotype: NY!). Shrubs erect, rounded, or lax, dense to open, the central axis ± developed, branches ascending or clambering and several; twigs, peduncles and often petioles moderately setose, rarely with stipitate glands mixed among the eglandular hairs. Leaf-blades ovate to broadly ovate or broadly elliptic, (l-)3-9 cm long, moderately to weakly rugose, puckered between tertiary veins; apex acute to acuminate, occasionally obtuse or rounded; marginal teeth (4-)10-25(-35) per side, with sinuses 0.5-2 mm deep; adaxial surface antrorsely strigose-velu- tinous, the hairs 10-50/sq. mm. Peduncles (0.5-)0.8-2 x leaf length (usually about equalling to almost twice when mature). Bract series gradually reduced in size and width; proximal bracts lanceolate, lanceolate-linear, narrowly elliptic or rarely narrowly oblanceolate, 5-12 mm long, 0.5-1. 5 (rare outermost one subfoliar to 2.5) mm wide, widest in proximal third (often near base), sometimes near middle or distal third; apex attenuate; indument setose or pilose, ciliate or not; distal bracts 3-5(-8) mm long. Corolla yellow aging yellowish or red- dish orange; corolla tube 7-12 mm; corolla limb 6-10 mm in diam. Distribution and habitat . — Mexico (northwest, central, southern), Central America (Guatamala to central Panama), Cuba; cultivated and escaped in Old World tropics; littoral and thorn shrubland, open pine-oak and deciduous montane forest and woodland; disturbance openings in tropical evergreen, sclerophyll, and decidu- ous forest and woodland; tropical savanna; 0-2500 m. e Cooper 5892 (US). San Jose: Sidney 42, also x L. nivea subsp. mutabilis (F); Tonduz 3377 (US); Tonduz 7 035 (LL[di]). GUATEMALA. Aha Vera Paz: Turckheim 39 (US). MEXICO. Chiapas: Ton 7048 (TEXIdi)). Jalisco: Gregory & Eiten 209 (SMU). Nayarit: Waterfall 16328, also x L. kingii (SMU[di]). Veracruz: Nee et al. 25132 subsp. glandulosissimal (MO). 5. L. insularis. CUBA. Santiago de Cuba: Britton et al 12624 (NY); Ekman 7 972 (NY); Havard 125 (NY). 6.%J scabrida. COSTA RICA. Gnanacaste: Tonduz 13630 (LL[di]). Limon: fiminez 1903 (NY). CUBA. Havana: Ledn 1744 (NY). Matanzas: Brit- ton et al 2 34 (NY). Santiago de Cuba: Britton et al. 12898 (NY). HONDURAS. Comayagua: Wilson 478 (NY). PANAMA. Bocas del Toro: Peterson & Annable 7269 (MO). Codfc Gonz0kz23 (MO). Coldn: Blum & Dwyer 2119 (MO); Mill er & Miller 908 (MO). Panama: Ebinger29 camara). AUSTRALIA. New South Wales: Riding 77 (BRIT). Queensland: Hannan-Jones 73 (BRIT). 10. L. kingii. MEXICO. Hidalgo: Carney 31 (BRlT[di]). Oaxaca; King 1178 (NY). Sinaloa: Gentry 7133 (NY). Sonora: Frye & Frye 2308 (NY). 12b. L. nivea subsp. mutabilis (L. horrida subsp. uncertain, could also be L. horrida subsp. tMfolia). AUSTRALIA. New South Wales: Day 70 (BRIT); Day 72 (BRIT). RWANDA. Butare: D’Arcy 8700 (MO). 20. L. strigocamara. AUSTRALIA. Queensland: Hannan-Jones 35 (BRIT). See also taxa la, Id, le 2b. Lantana horrida subsp. zanonii (R.W. Sanders) R.W. Sanders, comb. nov. Basionym: Lantana urticifolia Mill, subsp. Shrubs erect, rounded, lax, or trailing, dense to open, the central axis well-developed to abortive, branches ascending to arching and several or decumbent and few; twigs, peduncles and/or petioles densely setose and stipitate-glandular. Leaf-blades ovate, ovate-elliptic, trullate, ovate-deltate, or broadly ovate, or smallest ones subrotund, (0.5-)l-4(-6) cm long, moderately to prominently rugose, puckered between tertiary and/or sec- ondary veins, apex acute to rounded; marginal teeth 6-20 per side, with sinuses 0.2-2.5 mm deep; adaxial surface antrorsely strigillose to strigose-velutinous, viscidly stipitate-glandular or not, the hairs 10— 150/sq- ; mm. Peduncles 1-4 x leaf length (usually about 2 when mature). Bract series gradually reduced in size and width or all similar; proximal bracts lanceolate, lanceolate-linear, narrowly oblong or oblong-obovate to spatu- late, 2.5-10 mm long, 0.5-1.5 mm wide, widest in proximal, middle, or distal third; apex rounded, obtuse or acute; indument strigose-pilose or setose, often sessile- or stipitate-glandular, ciliate or not; distal bracts 2-4 mm long. Corolla yellow and aging yellow, orange or orange-red; corolla tube 4-10 mm; corolla limb 4-8 mm Sanders, Taxonomy of Lantana sect. Lantana Distribution and habitat . — Eastern Cuba, Jamaica, Hispaniola, Puerto Rico, Virgin Islands and northern Lesser Antilles; brushland and open tropical deciduous to semi-evergreen woodland or open pine woodland on rocky (often calcareous) slopes; 0-600 m. When I originally described L. urticifolia subsp. zanonii, it initially appeared to differ from L. arida var. s argentii in leaf and bract shape and some vestiture traits (Sanders 1989). However, careful examination for the present study failed to produce consistent distinctions or geographic correlations. If L. subcordata had not proved to be partially continuous with L. horrida var. sargentii, there would have been no need to recognize varieties within the subspecies. However, there is overlap in bract shapes, and one gathering (Dominican Re- public. Santiago: Liogier 13272, LL, NY), otherwise identical to L. subcordata, is an erect shrub as in var. sargen- tii. Recognition of only two varieties within Lantana horrida subsp. zanonii has resulted in the lack of a nomi- nate variety because autonyms exist only for infraspecific taxa that include the type of the species (ICBN Art. 26, Note 1, McNeill et al. 2007). Because, u sargentii” has priority at the varietal level, I am not free to publish the name L. horrida var. zanonii for the variety that includes the type of L. horrida subsp. zanonii (ICBN Recom- mendation 26A, Examplel). 2b.i. Lantana horrida subsp. zanonii var. sargentii (Moldenke) R.W. Sanders, comb. & stat. nov. Basionym: Lan- Shrubs erect or rounded and open to dense, the central axis well-developed, branches ascending to arching and several; twigs, peduncles and/or petioles densely setose and stipitate-glandular. Leaf-blades ovate, ovate- elliptic, trullate, ovate-deltate, or broadly ovate, or smallest ones subrotund, (0.5-)2-4(-6) cm long, moder- ately to strongly rugose, puckered between tertiary and/or secondary veins; apex acute to rounded; marginal teeth 6-20 per side, with sinuses (0.5-)l-2.5 mm deep; adaxial surface with hairs 10-50/sq. mm. Peduncles 1-3 x leaf length (usually about 2 when mature). Bract series gradually reduced in size and width; proximal bracts lanceolate, lanceolate-linear, elliptic-lanceolate, narrowly oblong, oblong-oblanceolate, or spatulate, 5-10 mm long, widest in proximal, middle, or distal third; apex rounded, obtuse or acute; distal bracts 2-4 mm long. Corolla yellow aging orange or orange-red; corolla tube 5-10 mm. Distribution and habitat .— Eastern Cuba, Jamaica, Hispaniola, Puerto Rico, Virgin Islands and northern Lesser Antilles; brushland and open tropical deciduous to semi-evergreen woodland or open pine woodland on rocky (often calcareous) slopes; 0-600 m. See description and illustration in Sanders (1989). Selected specimens examined: CUBA. Guantanamo: Britton 2168 (NY). Santiago de Cuba: Britton et al. 12622 (NY). DOMINICAN RE- PUBLIC . Azua: M. /.ii a ,1/ I ,sl ,X\ II \ M 1 1 IVdernalcs: !. ' I >•%<• > > ' Ul«> Peravia: / ,oMI\|, AN Kl PI Bl K near \imi.i>y \t„. Quezaltenanjjo: Skutch 1347 (NY, Sanders, Taxonomy of Lantana sect. Lantana & Hernandez 596 (NY); Ortega et al. 19650 (NY); Sharp 44197 (TENN); Vasquez 488 (NY). U.S.A. Alabama. Mobile Co.: cult., Mohr s.n. (US). i. GUATEMALA. Retalhuleu: M axon et al. 3523 (US). MEXICO. Chiapas: Martinez 8949 x L. kingii (BRIT[dil). Quintana Roo: Cabrera & Cabrera 4046 (TEX[dil). PANAMA. Bocas il. 898 (MO). Chiriqui: Churchill & Churchill 6136 (MO); Lewis et al 616 (MO); Tyson 897 (MO). Kuna Yala (San Bias): McDonagh etal. 167 (MO). Panama: Knapp et al. 3298 (MO). 9i-cvx20. L. CaUowiana Hybrid Group cultivars (L. depressa-tetraploid x siri B oiamara) TANZANIA. Kilimanjaro: Ml, , UJ I'l. l I 10. I . kingii. MEXK O. Oaxaca: c : n-iiXalicrnaiively x / scabrida or L. CaUowiana Hybrid Group cv. (TEX[dil). Puebla: Arsine s.n. 3 Nov 1908 (US). San Luis Potosfc Davis 24 4 (NY); Krai 2 4812 (VDB); Parry L. hirsuta subsp. amazcmica). AUSTRALIA. Queensland: Day 56 (BRIT); Day 87 )UTH AFRICA. Limpopo: Day 6 (BRIT). 18. L. urticoides. MEXICO. Nuevo subsp. uncertain, could also be L. hirsuta subsp. amazonica). AUSTRALIA. )Sp. nov. (Fig. 1). Type: BR/ IBGE.NY!). Leaf-blades broadly ovate-elliptic to ovate, oblong-ovate, or elliptic; apex acuminate (abruptly contracted to triangular tip (5-)10-15 mm long), acute or rarely obtuse or rounded (triangular tip lacking); marginal teeth (20-)25-35 per side, with sinuses 0.3-1.5 mm deep; adaxial surface with the setae (3-)10-40/sq. mm; abaxial surface with the setae erect and usually arching or sinuate, 10-80/sq. mm. Peduncles 2-5(-8) cm, 0.3-0.5(-l) x leaf length. Proximal bracts with longest hairs 0.3-1 mm. Distribution and habitat . — Brazil (eastern Amazon Basin and Planalto), Paraguay, northern Argentina, Bolivia, Peru, Ecuador, and Colombia to western Venezuela; cultivated and escaped in Old World tropics; openings in tropical semi-evergreen forest and montane evergreen forest on poor soils, tropical savanna with gallery forest and areas of dense woodland, shrubland, and grassland; 0-2000(-3000) m. Schauer (1847, 1851 [t. 42]) recognized Lantana hirsuta subsp. amazonica as a species but misapplied the name L. mista L. to it (see hybrid synonymy lfx4). ularis Moldenke, Caribl >;16. 1940. Type: JAMAICA: [1 Shrubs erect, apparently pyramidal and open; stems 1-3 m; branches ascending and several; twigs, peduncles and often petioles spasely setose but viscid with conspicuous, dense stalked glands, the setae 1.2-2 mm, stipi- late glands, 0.3-0.7 mm. Leaf-blades ovate-triangular to ovate-elliptic or lanceolate, 4-10 cm long, the length Journal of the Botanical Research Institute of Texas 6(2) 423 a and Cuba; openings or disturbed areas in cloud c Presumed hybrids with: 6. L. scabiida. CUBA. Sancti Spiritus: Alain/ C. Lantana sect. Lantana series Strigosae R.W. Sanders, ser. Adaxial leaf-surfaces strigose, strigillose, or scabrid, the hail abaxial leaf-surfaces strigose, usually sparsely so, the hairs o innervated tissue, strigiform, 0.03-1 mm. Inflorescences an bose in fruit (continuing to initiate flowers somewhat in L. n ciduous in lower portion leaving it more or less bare w >tent bases) about ] of flowers/fruits active). c Sagua de Tanamo, 8 Apr 1987, Mendez 3373 (holotype: HIPC[di!]). Shrubs rounded and open, lax or subscandent; stems 0.5-6 m; branches ascending to clambering, often few; twigs, peduncles and often petioles thinly to moderately strigose or setose, the hairs 0.2-1 mm. Leaf-blades ovate to elliptic-lanceolate or lance-oblong, (2-)5-15 cm long, the length (1.5-)1.7-2.5(-3) x width, nigrescent, papery to subcoriaceous, triplinerved; base attenuately to abruptly tapering onto petiole from middle or just below middle; apex acuminate, acute, obtuse, occasionally rounded; marginal teeth (12-)17-25(-35) per side, obtuse, rounded, or acute, spreading to appressed, then sometimes with tips recurved, with sinuses 0.5-1.5 mm deep; adaxial surface lustrous, thinly strigose or scabrous, the hairs occurring on veins and center of are- oles, 0.1-0.6(-1.2) mm (longest usually 0.2-0.4 mm, except in “scandens” morph where 0.6-1.2 mm), (4-)6- 12/sq. mm, sometimes with conspicuous vitreous or whitened pustulate bases 0.1-0.3 mm in diam.; abaxial surface whitish or pale green but not glaucous, antrorsely strigose-scabrous, with the strigae scattered on veins and veinlets, (0.03-)0.1-0.5(-1) mm (longest mostly 0.3-0.5 mm, except in “scandens” morph where 0.5-1 mm), (4-)6-12(-25)/mm sq. (“scandens” morph 0-5)/sq. mm). Inflorescences remaining hemispheric; pe- duncles 0.5-1 x leaf length. Proximal bracts narrowly lanceolate, lance-elliptic or -oblong (including those with slight constriction in proximal third; occasionally 1 or 2 outermost bracts subfoliar or narrowly spatu- late), 4-8 mm long, 1—2 mm wide, widest at very base, in proximal third, or in middle third (then often equally wide at base), with 3 veins from the base, appressed or spreading, usually deciduous after flowering; apex at- tenuate or acuminate; indument sparsely strigose, somewhat or not ciliate, the longest hairs 0.3— 0.5(— 1) mm. Corolla yellow to or aging reddish orange; corolla tube 7-12 mm. Distribution and habitat .— West Indies (Cuba, Jamaica, possibly Hispaniola, Puerto Rico, possibly the Virgin Islands, and the northern Lesser Antilles), eastern coastal Mexico (Tamaulipas southward) with local- ized disjunction in southwestern Michoacan (long-haired, scandent morph). Central America, and Caribbean coast and slopes of Colombia, Venezuela, and the Guianas; disturbance openings, savanna and man-made grassland in tropical dry to humid forest; 0-1600 m. Lantana scabrida is replaced in upland and western Mexico by L. kingii, but the two are apparently sympat- ric in Tamaulipas. Field work is needed to determine the status of the isolated scandent morph in Michoacan. Sanders (1987b) reported the chromosome number (2n = 44) of Lantana scabrida in the Luquillo Mountains of Puerto Rico ( Sanders 1510 ) as L. camara due to misapplication of the name and confusion with L. strigocamara. Sanders, Taxonomy of Lantana sect. Lantana tensely lustrous, strigose-glabrescent but smooth, the hairs occurring on veins (scattered) and center of are- oies, 0.2-0.4(-0.8) mm, 0-2/sq. mm, flaccid and strongly appressed to surface (unique in this species), often deciduous not pustulate based; abaxial surface whitish or gray-green (but not glaucous), weakly strigose and nearly glabrous, with the strigae scattered on veins and veinlets, 0.1-0.6 mm, 0-4/sq. mm. Inflorescences re- maining hemispheric; peduncles Vfe-VS x leaf length. Proximal bracts oblanceolate to narrowly oblong to nar- rowly lanceolate, (4-)6-10 mm long, 1-1.7 mm wide, widest in proximal third to distal third, with 3 veins from the base, appressed or spreading, deciduous after flowering; apex attenuate; indument glabrescent, not cihate, the longest hairs 5 0.5 mm. Corolla yellow to or aging reddish orange; corolla tube 7-10 mm. Distribution and habitat.-Ce ntral Lesser Antilles (Dominica, Martinique, probably Saint Lucia); sunny slopes in borders and openings of montane rainforest; 400-1000 m. See discussion in Sanders (1987c). Sheeted specimens examined: LESSER ANTILLES. Dominica: Lee 9 (NY); Lloyd 201 (NY); Smith 10216 (SMU). Martinique: Bailey & Bailey ANTILLES. Dominica: Hill 23959, altera 9. Lantana depressa Small, Bui v York Bot. Gard. 3:436. 1905. 1 Shrubs trailing to erect, dense to ± open; stems 0.1-3 m; branche several to numerous; twigs, peduncles and often petioles thinly setose or strigose-s Leaf-blades ovate-elliptic to elliptic, l-6(-8) cm long, the length 1.7-2.3(-3) x width, induplicately curved at maturity (unique to this species), nigrescent, papery, triplinerved; base obtuse or acute, tapering onto petiole from middle or just below middle; apex abruptly tapered, obtuse or acute; marginal teeth 3-15 per side, rounded to acute, often appressed, with sinuses 0.5-1.5 mm deep; adaxial surface lustrous, antrorsely strigil- lose to strigose the hairs occurring on veins and center of areoles, 0.1-0.7(-l) mm, 2-8/sq.mm, not noticeably vitreous-pustulate, the circular bases of the strigae ca. 0.1-03 mm in diam.; abaxial surface slightly lighter or duller green, antrorsely strigose-scabrous, with the strigae scattered on veins and veinlets, longest ones 0.5-1 mm, 0.5-8/sq. mm. Inflorescences remaining hemispheric; peduncles 0.7-2 x leaf length. Proximal bracts elliptic-lanceolate, 4-7 mm long, 0.5-1.5 mm wide, widest at middle or just below, with3veins from the base, appressed or spreading, persisting and reflexed from base in fruit; apex acute to attenuate, indument strigose, ciliate or not, the longest hairs 0.1-1 mm. Corolla yellow aging to a dark yellow or dull, pale orange; corolla tube 5-11 mm. Distribution and habitat .— Peninsular Florida; limestone pinelands, wet prairies, and dunes; 0-50 m. See more complete discussion in Sanders (1987a) and Maschinski et al. (2010) and illustrations in Sandeis (1987a). i. Lantana depressa var. depressa ttoon. Plant Buyer’s Guide, ed. 6, 1 ssa Small var. depressa (see Sanders 2001, [ V. Long, Rhodora 72:34. 1970. Type U.S.A. Florida. E 426 Shrubs low mounded, dense, 0. 1-0.3 (spreading to 1) m, the central axis abortive or hardly developed; branch- es prostrate or decumbent, twigs, peduncles and often petioles with hairs 0.5-1 .5 mm. Leaf-blades 1-3 cm long (to 4 or 5 cm in some cultivars). Corolla with tube 5-9 mm; corolla limb 5-8 mm in diam. Distribution and habitat.— Peninsular Florida (Miami Ridge); cultivated and escaped in tropics and sub- tropics worldwide; limestone pinelands; 0-25 m. Because of its drought tolerance, compact habit, and profuse flowering, Lantana depressa var. depressa has been cultivated widely since the 1950s. A tetrploid cultivar and L. strigocamara apparently are the parents of the currently popular Callowiana Hybrid Group cultivars, which have the floral colors of L. strigocamara and are cultivated worldwide and escaped pantropically (see Sanders 2001, specimen citations below, and 9i-cvx20 in the section on hybrid synonymy). Selected specimens examined: AUSTRALIA. Queensland: cult., McAndrew MJH-647 (BRIT). U.S.A. Florida. Dade Co.: Deam 60894 (NY); Demaree 10208 (SMU); Krai 53943 (VDB); Krai 53964 (VDB); Krai 66236 (NY, VDB); Krai 70742 (VDB); Small et at. 3482 (NY); Traverse 646 e: Moldenke 797 (NY), i: Moldenke 19862 (SMU). GUATEMA1 (BRIT). MEXICO. Jalisco: Jimenez 357 (NY). Sinaloa: Hutchison 2544 (NY). Veracrnz: Day 15 (BRIT). TANZANIA. Kilimanjaro: Mlangwa 352, possibly with genes of L. hirsute (BRIT[di]). Tanga Sallu 248 (BRIT). TRINIDAD & TOB, U.S.A. Florida. De Soto Co.: Krai 5 7309 (VDB). Lee Co.: Krai 11998 , alternatively a backcross in Co.: Carter & Carter 13218 (FTG). Louisiana. Iberia Par.: Webb 4764 (VDB). Massachusetts: ci Co.: cult., Dunn 13828 (BRIT). South Carolina. Beaufort Co.: Leonard & Radford 2743 p.p. (VDB). Tennessee. Davidson Co.: cult., Hackney s.n. 10 Oct 1990 (VDB). Texas. Blanco Co.: cult., Sanders 5470 (BRIT). Dallas Co.: cult., Shinners 8526 (SMU); cult., Skinners 29844 , probably ‘Cream Carpet’ (SMU); cult., Wansbrough 251 (SMU). Galveston Co.: cult.. Traverse 2515 (BRIT). Harris Co.: c Traverse 2401 (BRIT). Hidalgo Co.: cult., Green/ield5(BRIT). Montg Presumed further hybrids of I9i-cvx20] with: 10. L. kingii. MEXICO. Guerrero: Guerra 36 (LL[di]). I AUSTRALIA. Queensland: Day 86 (BRIT); Horrocks 55 (BRIT). TRINIDAD & TOBAGO. 5 planaltensis. BRAZIL. Bahia: Ddbereiner & Tokarma 1490 (LL[di]). PARAGUAY. Central: Perez 890 (MO). 1 Anderson Co.: cult., Couch 73 (SMU). Cameron Co.: Traverse 1030 (BRIT). 20. 1 (SMUjdi]). Leon Co.: Bratcher 57, alternatively x L. depressa var. depressa (BRIT); (VDB). Seminole Co.: SchaUert378 (BRITldi]). See also taxa la, lb, Id, le, If, 2a, 4i a and 20 and section on hybrid synonymy: 9i-cvx20. | 9ii. Lantana depressa var. floridana (Moldenke) R.W. Sanders, Syst. Bot. 12:55. 1987. Basionym: Lantai Shrubs rounded, open, 0.4-1 m, the central axis ± developed, but not prominent; branches arching or ascend- ing; twigs, peduncles and often petioles with hairs 0.5-1 mm. Leaf-blades 3-6(-8) cm long. Corolla with tube 7-11 mm, corolla limb 8-10 mm in diam. Distribution and habitat. — Peninsular Florida; Atlantic barrier dunes and interior sand ridges, stabilized and relictual dunes; 0-50 m. Selected specimens examined: U.S.A. Florida. Brevard Co.: Moldenke 2 18 (NY). Martin Co.: Krai 15385 (VDB); Krai 57137 (NY, VDB). N* Beach Co.: Small 2134 (NY). St. Johns Co.: Morton 4508 (SMU); Small 2313 (NY). Volusia Co.: Ray et al. 10817 (SMU). Geogia: cult., Bol** s.n. (NY). Presumed hybrids with: 20. L. strigocamara. U.S.A. Florida. Duval Co.: Curtiss 1968 (NY[2]). Lake Co.: Small 8666 (NY[21). Vd* ria Co.: Krai 18406 (VDB); Whetstone 145 18 (VDB). See also section on hybrid synonymy: 9iix20. 9iii. Lantana depressa var. sanibelensis R.W. Sanders, Syst. Bot. 12:55. 1987. Type: u.SA. Florida. Lee Co.: Sanibd* land, 11 May 1954, Cooley 2674 (holoiype. GH!; isotypes: USF[2]!). Shrubs erect and pyramidal, proximally dense, distally open, nent; proximal branches decumbent, distal ones ascending; i 1-1.5 mm. Leaf-blades 3-6 cm long. Corolla with tube 7-10 v Distribution and habitat.— Peninsular Florida; wet limesto barrier islands; 0-25 m. -3 m, the central axis well-developed, pro®*- vigs, peduncles and often petioles with hairs 427 ock 268 (NY). Presumed hybrids with: 20. L. strigc Brumbach 8182 (NY); Brumbach 8283 (NY, U ; Brumbach 9058 (VDB). f 7737 (FTG, NY); Sheehan s n. 7 Mar 1919 (NY). I 10. Lantana kingii Moldenke, Phytologia 8:161. 1962. Type: MEXICO. Oaxaca: Isthmus of Tehuantepec, Niltepec, I7Jul 1959, IQng 1775 (holotwe: TEX!; isotype: US!). Shrubs erect or rounded, open; stems 0.5-2(-3) m; branches ascending and several; twigs, peduncles and often petioles glabrescent with scattered antrorse hairs, the hairs ca. 03-0.7 mm. Leaf-blades ovate, ovate-elliptic or ovate-triangular (rarely, especially if less than 2 cm long, obovate to rotund), 1-8 cm long, the length (0.8-)1.2- attenuately tapering onto petiole from middle or just below middle to abruptly contracted and broadly cuneate, sometimes forming a short narrow petiolar wing; apex acute or acuminate (or abruptly to broadly rounded); marginal teeth (3-)6-15(-25) per side, obtuse or rounded, spreading to appressed, then sometimes with tips recurved, with sinuses 1-2.5 mm deep; adaxial surface lustrous, scabrous, the hairs occurring on veins and intervening tissue (sometimes just center of areoles), 0.1-0.5 mm, 3-7(-12)/sq. mm, mostly deciduous leaving the noticeably vitreous-pustulate circular bases, these 0.2-0.5 mm in diam.; abaxial surface whitish or pale green but not glaucous, antrorsely strigose-scabrous to nearly glabrous, with the strigae scattered on veins and veinlets, 0.1-0.7 (longest ones usually 0.3-0.4) mm, l-8(-12)/sq. mm. Inflorescences remaining hemispheric; peduncles about 0.5 to 1.5 x leaf length. Proximal bracts spatulate, oblanceolate, or oblong-oblanceolate, oc- casionally oblong-elliptic to broadly elliptic, (3.5-)5-10 mm long, 1.2-3 mm wide, widest in distal half (often just above middle), occasionally at or just below middle, with 3 veins from the base, appressed or spreading, persisting and reflexed from base in fruit; apex obtuse to acute; indument strigose, hardly ciliate, the longest hairs 5 0.5 mm. Corolla yellow to orange aging orange to orange-red; corolla tube 6-10 mm. Distribution and habitat. — Mexico (central highlands to eastern slopes of the Sierra Madre Oriental, Pa- cific slope, and Isthmus of Tehuantepec) to northern Central America (only to central Guatemala?); thorn for- est and schrubland; 0-2000 m. Selected specimens examined: MEXICO. Coahuila: Johnston 9325 (LL[di]); Henrickson 18926 (NY); Waterfall 16661 (SMU). Colima: Grego- ry & Eiten 334 (BRIT). Guerttero: Flyr 623 (SMU). Guerrero: Mocktord & Rowell 2790 (SMU). Michoacan: Turner 2024 (BRIT, SMU). Neuvo Le6tt: Garcia 10 (SMU); Pennell 16860 (NY); Waterfall 13187 (SMU); Waterfall 15312 (SMU[di]). Oaxaca: King 1328 (NY); King 1464 (NY); King 1598 (NY); Purpus 7306 (NY). Puebla: Chiang et al. F-2610 (TEX); Davis 211 (NY); Martinez 21705 (TEXldi]). Sinaloa: Gentry 11454 (LL); Palmer 1511 (NY); Rose et al. 13366 (NY). Sonora: Wiggins & Rollins 138 (NY). Tamaulipas: Smith Mex. 94 (LLldil). Presumed hybrids with: 2a/4a. L. horrida subsp. horrida or L. hirsuta subsp. hirsuta. MEXICO. Hidalgo: Cawrey3 (BRIT). 18. L. urticoides. MEXICO. Coahuila: Henrickson 11352 (LL[dil); Muller 3069 (LL(dil); Reveal et al. 2604 (NY): Wehbe 052 ( TEX[di] ): Wy nd & Mueller 88 (NY). Neuvo Le6n. Frye & Frye 2447 (NY); Meyer & Rogers 2686 (NY); Pringle 11670 (SMU). Tamaulipas: Meyer & Rogers 2499 (NY); Stanford ctal. 2302A (NY). 20. L. strigocamara. MEXICO. Tamaulipas: Krai 24799, alternatively x L. scabrida (VDB). Seealso taxa le, 2a, 4a, and 9i (as cv x 20) and section on hybrid synonymy: lexlO, Iex2axl0, lexl0/20, 2axl0, 4axl0, 10xla/le?, and 10x2a/4a? 11. Lantana ovatifolia Britton, Bull. New York Bot. Gard. 4:123. 1905. Type: BAHAMA ARCHIPELAGO. Grand Bahama: Feb 1905, Britton & Millspaugh 2450 (holotype: NY!; botype: F!). Shrubs lax and trailing, sparse; stems 0.3-1 m; branches prostrate, few; twigs, peduncles and often petioles moderately strigose, the hairs 0.2-1.5 mm. Leaf-blades ovate to ovate-elliptic, 2-6 cm long, the length 1.2-1.7 * width, not nigrescent, papery to subcoriaceous, pinninerved; base rounded to tapering onto petiole mostly from proximal third; apex acute to obtuse or rounded; marginal teeth 8-18 per side, rounded to acute, often appressed, with sinuses 0.7-1.5 mm deep; adaxial surface lustrous, antrorsely strigose to scabrous (due to loss of deciduous longer hairs), the hairs occurring on veins and intervening tissue (sometimes just center of are- oles), 0.2-1 mm, 2-7/sq. mm, noticeably vitreous-pustulate, the circular bases 0.3-0.5 mm in diam.; abaxial surface slightly lighter or duller green than adaxial surface, antrorsely strigose-scabrous to nearly glabrous, with the strigae scattered on veins and veinlets, 0.2-1 (longest ones usually 0.4-0.8) mm, 2-7/sq. mm. Inflo- rescences remaining hemispheric; peduncles about 1—1.5 x leaf length. Proximal bracts spatulate or oblan- ceolate, 6-12 mm long, 1-3 mm wide, widest in distal half, with 3 veins from the base, appressed or spreading, 428 Journal of the Botanical Research Institute of Texas 6(2) persisting and reflexed from base in fruit; apex obtuse to acute; indument strigose, hardly ciliate, the longest hairs £ 0.6 mm. Corolla yellow aging to a dark yellow or yellow-orange; corolla tube 6-10 mm. Distribution and habitat.— Northern Bahama Archipelago; limestone pinelands with open or low shrubby understory; 0-25 m. In certain respects Lantana ovatifolia is rather similar to L. kingii. However, as a narrow endemic with a narrow range of variation, it is best kept as a distinct species. This is further supported by its distribution, which is oddly disjunct from that of L. kingii. See Sanders (1987a) for detailed discussion and illustration. Selected specimens examined: BAHAMA ARCHIPELAGO. Grand Bahama: Brace 3686 (NY); Correll & Popcnoe 45402 (FTG, NY, SMU); Cone II &■ Krai 42892 (FTG, VDB); Correll & Krai 42946 (FTG, VDB). Shrubs lax, rounded and open, sometimes forming treelets; stems 0.5-4 m; branches ascending or divaricate, several to numerous; twigs, peduncles and often petioles thinly to moderately strigillose, setulose, or scabridu- lous, the hairs 0.03-0.4(-0.7) mm. Leaf-blades ovate, lanceolate or elliptic, 3-12 cm long, the length some- times of those subtending inflorescences distinctly reduced (unique to this species), (1.5-)1.7-3(-3.6) x width, nigrescent, membranous to papery, triplinerved; base attenuately tapering onto petiole from widest point or acute; marginal teeth (13-)18-40 per side, obtuse, rounded, or acute, usually appressed, sometimes spreading at tip, with sinuses 0.2-1 mm deep; adaxial surface lustrous, antrorsely strigose to scabrous or scabridulous, the hairs occurring on veins and intervening tissue, 0.03-0.5 mm, 10-60/sq. mm, often with conspicuous vitreous or whitened pustulate bases 0.1-0.3 mm diam.; abaxial surface slightly lighter or duller green than adaxial surface, antrorsely strigillose to strigose-scabrous, with the strigae scattered on veins and veinlets and sometimes on intervening tissue, 0.03-0.4(-0.7) mm, 10-60/sq. mm. Inflorescences sometimes 2 per leaf axil, hemispheric but receptacle often elongating by slight separation or prolonged initiation of nodes and be- coming naked below the hemispheric flower cluster at apex; peduncles 0.5-1.2 x leaf length. Proximal bracts narrowly triangular, lanceolate or linear-lanceolate, occasionally oblanceolate, 2.5-7(-10) mm long, 0.6-13 mm wide, widest at very base, proximal third, or occasionally distal third, with 3 veins from the base, ap- pressed or spreading, usually deciduous after flowering; apex acute, attenuate, or subulate; indument strigil- lose-scabridulous, hardly ciliate, the longest hairs 5 0.5 mm. Corolla white aging white, pale pink or light blue, or opening pink, cream or yellow aging cream, yellow or orange infused with purple, pale yellow throat usu- ally developed and fading with age; corolla tube 7-12 mm. Distribution and habitat. — Eastern and southeastern Brazil; cultivated world-wide, escaped pantropically understory and disturbance openings and man-made grasslands in tropical humid forest, occasionally in dry forest; 0-1500 m. Sanders, Taxonomy of Lantana sect. Lantana Leaf-blades with the indument of the adaxial surface composed of strigae and rough points, 0.03-0.3 mm; indument of the abaxial surface composed only of strigae 0.03-0.3 mm, never with short, straight fililform prolonged initiation of nodes and becoming naked below the hemispheric flower cluster at apex. Corolla white aging white, bluish or pale pink, or pink aging light purple. Distribution and habitat . — Eastern and southeastern Brazil; cultivated world-wide, sometimes escaped pantropically; understory and disturbance openings and man-made grasslands in tropical humid forest, oc- casionally in dry forest; 0-1500 m. Many of the native collections have narrowly elliptic leaf-blades. However, other native collections vary toward having the more ovate or lanceolate blades typically seen in cultivated plants of the species (e.g., the type specimens of Lantana nivea and L. triplinervia). There is a tendency, especially in the collections from Ba- hia, for a marked reduction in size of leaves subtending the inflorescences, resulting in a corymb-like arrange- Selected specimens examined: AUSTRALIA. Queensland: McAndrew 32 (BRIT). BRAZIL. Bahia: Mori et al. 10266 (LL); Sieber s.n. 1826 (BR[di]); Silva58360 (NY, US). Distrito Federal: Duarte & Pereira 4740 (US). Minas Gerais: Hatschbach 31331 (VS); Irwin 2112 (NY, US). Rio de Janeiro: Araujo 3034 (NY); Brack 24153 (NY); Carauta et al. (Herb. 18132) (LL); Nee 3389 (US); Pereira et al. 4383 (NY). MEXICO. Distrito Federal: cult,. Bonpland s.n. (P[2di]). SRI LANKA. North Central: cult., Moldenke et al 28233 (US). (NY); Harley 21658 (NY); Harley 21927 (NY); Hatschbach 46372 (LL[dil); Irwin et al. 27777 (LL[di]). 17. L. planaltensis. BRAZIL. Bahia: Harley et al. 21617 (US). Minas Gerais: Anderson 9142 (F, US); Mexia 5436 (F, NY). Rio de Janeiro: Segadas-Vianna et al. 593 (SMU). 20. L. 12b. Lantana nivea subsp. mutabilis (Hook.) R.W. Sanders, Sida 22:395. 2006. Bason™: Lantana nivea Vent. var. Leaf-blades with the indument of the adaxial surface composed of strigae, 0.05-0.5 mm; indument of the ab- axial surface composed of strigae, 0.05-0.7 (longest ones mostly 0.4-0.5) mm, often with short, straight filil- form hairs to 0.3 mm mixed in but not dominant or codominant. Inflorescences remaining hemispheric; re- ceptacle rarely elongating and becoming naked below. Corolla opening creamy white or pink, aging cream, yellow or orange infused with purple or opening yellow aging purple. Distribution and habitat. — Probably of cultivated hybrid origin, cultivated world-wide and escaped pan- tropically (especially in Australia), but also collections from southeastern Brazil apparently of natural hybrid origin; understory and disturbance openings and man-made grasslands in tropical humid forest, occasionally in dry forest; 0-1500 m. Selected specimens examined: AUSTRALIA. Queensland: Day 53 (BRIT). BRAZIL. Goias: Irwin et aL 25204 (NY). Rio de Janeiro: Carauta & Araujo 2244 (LL[di]); G6es 64 (NY); Sarahyba 80 (NY); Segadas-Vianna 4053 (SMU). BURUNDI. Bujumbura: Lewalla 5717 (MO). CO- LOMBIA. Caldas: cult., Ldpez-Palacios 4023 (LL). DOMINICAN REPUBLIC. La Vega: Ososki & Rodriguez 249 (BRIT). GOLD COAST. Knmari: D arko 598 (MO). TRINIDAD & TOBAGO. Smith Tr. 2 (LL[dil); Smith Tr. 24 (LL[diJ). MEXICO. Tabasco: Taylor & Taylor 12574 (BRIT). PARAGUAY. Canendiyu: Zardini 48550 (MO). SIERRA LEONE. Freetown: Johnston s.n. 3J Jan 1882 (MO); Morton SL185 (MO). y 19 (BRIT). SRI LANKA. Central Moldenke et al 28159 (US); A i. Dade Co.: cult., Avery 1867 (NY). tania: Crosby &Doore 162 (SMU). 13. L. cuja i: Huge 36 5 (LL[diJ). 17. L. planaltensis. BR ,n hybrid synonymy; ldxl2b, 2cxl2b, and l: 13. LantanacujabensissJu.au Prodr [A. P. dc Candolle! 11 1847 i i u M ( ing syntypes: BRAZIL. ;Rio Negro, Martius s.n. (M[di!l). PERU: Poeppig 148 5 (G[di!]). (Urban 1906)], Riedel s.n. (holotype: LE, n.v.). Lantana tenuifolia Rusby, Phytologia 1:74. 1934. Type: BOLIVIA. [La Paz:) Bopi River, 11 Sep 1921, Rusby 653 (holotype: NY!). Lantana cujabens is Schauer var. parvijolia Moldenke, Phytologia 9:186. 1963. Type: PERU. La Libertad: Otuzco, 1 Jul 1951, Angulo & Rusby 715 (holotype: NY!). Shrubs rounded and ± open, lax, or subscandent; branches ascending or clambering, several to few, occasion- ally herbaceous; twigs, peduncles and often petioles glabrescent to thinly setose or scabrous, the hairs (0.1- J )0.3-0.5(-l) mm. Leaf-blades ovate, lanceolate, elliptic, or narrowly oblong, (3-)4-12(-16) cm long, the length 1.3-2.5(-4) x width, nigrescent or not, papery to subcoriaceous, triplinerved to pinninerved; base rounded or truncate and abruptly tapered onto petiole or cuneate and often forming an attenuate wing; apex acute, abruptly acuminate, or attenutate; margin with the teeth (15-)25-40 per side, rounded to acute, often appressed to strongly so and barely discemable, the sinuses 0.1-1.5 mm deep; adaxial surface dull to occasion- ally lustrous, antrorsely strigillose to strigose-pubescent, the hairs occurring on veins and intervening tissue, 0.1-0.4(-0.7) mm, 3-20/sq mm, not noticeably vitreous-pustulate, the circular bases of the strigae ca. 0.1-02 J mm in diam; abaxial surface slightly lighter or duller green than adaxial surface, sometimes whitish green but not glaucous, antrorsely strigose-scabrous to nearly glabrous, with the strigae scattered on veins and veinlets, 0.1-0.6 mm, 3-KVsq mm. Inflorescences with peduncles 0.5- 2 x leaf length. Proximal bracts lanceolate, oblong, or ovate-elliptic, often subfoliaceous, (4-)6-20 mm long, 2-8 mm wide, widest in proximal to middle third, occasionally distal third, with 5-7 veins from the base, appressed to spreading, persisting and recurved or reflexed in fruit; apex acute, briefly acuminate, or obtuse, often rounded at very tip; indument thinly stri- gose, usually not distinedy ciliate, with longest hairs 0.1-0.6(-l) mm. Corolla yellow, orange, or light red aging reddish orange to bright red (or occasionally intense reddish purple); corolla tube 7-12 mm. Distribution and habitat. — Brazil (westernmost Amazonia, southern and central Planalto), the Guianas, Venezuela, Colombia, Ecuador, Peru, Bolivia and Paraguay; understory and disturbance openings in tropical humid forest, occasionally in dry forest or hard-pan savannas; 100-3000 m. Lantana cujabensis is variable with respect to leaf-blade shape and width (broadly ovate to narrowly ellip- tic-oblong) and marginal serration, outer bract length and width, and altitude preference. Apparently the type specimens of L. cujabensis and L. riedeliana represent the broad-leaved, toothed vs. narrow-leaved, subentire extremes, respectively; that of L. tenuifolia is intermediate. Different specimens exhibit all possible combina- tions, which do not correlate with geography, therefore, no infraspecific taxa are justifiable based on the sam- ple studied. t eetdl& Selected specimens examined: BOLIVIA. Cochabamba: Jaramillo etal. 1212 (MO); Ritter 1644 (MO); Steinbach 644 (F, N 1 1927 (MO). Santa Cruz: Guillen &Roca 3334 (F); Steinbach347 (NY, SMU); Steinbach 794 (SMU). BRAZIL. Acre: A I i: Maas &• Maas 273 (MO). Bahia: dos Santos & Barreto 65 (LL[di]). Teixeim et al. 331 (M w k Diaz 3478 (MO). Meta: Betancur 1336 (M0* !. 2877 (F); Holm-Nielsen etaL 3016 (ft Napo: Abbott 15637 (MO); Campos 135 (F); Cento 278 (MO); Croat & Hannon 93505 (MO); Ponce & Ght a 320 (MO). PERU. Amazonas: Casin' 1. 18983 (MO); Lewis et al. 18100 (MO). Cusco: Huamantupa et al. 4024 (MO). Macurdy 1007 (F, M0* (MO); Santos et il. 18843 (MO>, Castro et 2(F).L k Fosberg 29017 (F); Fosberg 29104 (F); Gentry et al. 15599 (F); R Sanders, Taxonomy of Lantana sect. Lantan 431 Z 745 (MO); Gudnchez 2132 (MO); Uesner 4021 (MO); Liesner 7454 (MO). Guarico: Ramirez 2027 (NY). Merida: Ldpez-Palados 2695 (LL). Presumed hybrids with: 15. L. micrantha. BOLIVIA. Chuquisaca: Jimenez & Flores Til (MO). 17. L. planaltensis. BRAZIL. Minas Gerais: Gentry et al. 49588 (MO), identification uncertain. PARAGUAY. Itapua: Perez 179 (MO). 19. L. paraensis. SURINAME: Si- paliwini: Miller etal. 9367 (MO). VENEZUELA. Bolivar. Sanoja 2343 (MO). See also taxa Id, 2c, 4b, 9i (as cv.) and 12b and section on hy- brid synonymy: 2cxl3, 4bxl3, and 15x13/20? D. Lantana sect. Lantana series Spicatae R.W. Sanders, ser. nov. Type: Lantana viscosa Pohl ex Schauer Adaxial leaf surfaces strigose-villosulous to setose-villous, the hairs up to 2.5 mm; abaxial leaf surfaces setose or pilose, often densely so, the hairs occurring on veins and non-innervated tissue, setiform or filiform, 0.2-2.0 mm. Inflorescences initially hemispheric becoming short-cylindric by prolonged initiation of flowers or elon- gation of internodes. 14. Lantana viscosa P Shrubs erect or rounded and open to lax and subscandent; stems 1-3 m; branches ascending and numerous to clambering and few; twigs, peduncles and often petioles sparsely setose but viscid with dense understory of conspicuous stalked glands mixed with short hairs, the setae 1-2 mm, the stipitate glands and short hairs, ca. 0.5 mm. Leaf-blades broadly ovate, ovate, or ovate-elliptic, 2-8 cm long, the length 1.4-1.8(-2.2) x width, not nigrescent, papery, pinninerved; base usually rounded to truncate, sometimes broadly cuneate or subcordate, briefly narrowly cuneate onto petiole at very base; apex usually abruptly acuminate, sometimes acute; mar- ginal teeth (10-)20-35 per side, acute to rounded, spreading to ascending, rarely appressed with tips recurved, with sinuses (0.3-)0.6-1.5 mm deep; adaxial surface dull, setose to villous, the hairs occurring on veins and intervening tissue, longer ones 1-1.5 mm or more (shorter ones ±. 0.5 mm, often mixed glandular and eglandu- lar), 30-70/sq. mm, not noticeably vitreous-pustulate, the circular bases of the hairs ca. 0.1-0.2 mm in diam.; abaxial surface slightly lighter or duller green than adaxial surface, setose to villous, the hairs on all veins and intervening tissue, longer ones 1-1.5 mm or more (understory hairs 0.3-0.8 mm, these often mixed glandular and eglandular), 50-150/sq. mm. Inflorescences occasionally 2 per leaf axil, becoming short-cylindric by prolonged initiation of flowers; peduncles 0.5-1.3 x leaf length. Proximal bracts lanceolate or elliptic to ovate- elliptic, (2.5-)4-7 mm long, 1.5-3 mm wide, widest in proximal or middle third, with 3 veins from the base, ± spreading, persisting (proximally ± cupped around enlarging fruit) and becoming reflexed (± distally only) in fruit; apex abruptly acuminate with prolonged tip to caudate; indument setose-pilose, usually ciliate, often also stipitate-glandular, the longest hairs 0.7-1.5 mm. Corolla reddish purple to pale pink, often with white or yel- low throat, occasionally white; corolla tube 5-10 mm. Distribution and habitat.— Venezuela, Brazil (northern, eastern, and southern), Paraguay, Bolivia, and Peru, possibly also Ecuador, Colombia, and the Guianas; widely distributed but infrequent; openings in tropi- cal evergreen forest, tropical savanna with gallery forest, and areas of dense woodland, shrubland, and grass- land; 100-1200 m. Selected specimens examined: BRAZIL. Distrito Federal: Heringer et cd 7087 (NY). Pari: Strudwick & Sobel 4236 (LL[di], NY). Pernam- buco: Figueiredo et al 18 (US). PARAGUAY. Amambay: Zardini & Guerrero (NY). VENEZUELA. Bolivar: Steyermarh cl al. 115523 (NY). Presumed hybrids with: 15. L. micrantha. PARAGUAY. Cordillera: Merries & Degen 5522 (MO). See also taxa 2c, 4b, and 12a and section on hybrid synonymy: 12axl4. 15. Lantana micrantha Briq., Annuaire Conserv. Jard. Bot. Geneve 7-8:299. 1904. Type PARAGUAY: AscunriOn, Apr 1874, Bdlansa 1039 (holotype: G(dU]; isotypes: Kldi!], P[2,di!]). Shrubs erect or rounded, open; stems 0.5-2 m; branches ascending and several; twigs, peduncles and often petioles moderately to densely setulose, setose, pilose, or also stipitate-glandular, the hairs usually 0.5-1.2 mm (these sometimes lacking), mixed with shorter glandular and eglandular hairs about 0.2-0.3 mm. Leaf-blades Sanders, Taxonomy of Lantana sect. Lantana tween the more or less sympatric taxa L. camara subsp. camara and L. s plendens, probably prior to European colonization or perhaps human habitation. Moreover, the range of variation is narrower than expected for a hybrid swarm, and the distribution extends beyond the area of sympatry. Selected specimens examined: BAHAMA ARCHIPELAGO. Cat Island: Britton & Millspaugh 5763 (NY); Correll 46083 (FTG, NY); Correll 46098 (FTG, SMU). Eleuthera: Lewis 7232 pp. (FTG). Grand Caicos: Gillis 12317 (LL). Great Exuma: Correll & Correll 42298 (FTG, NY); oles moderately to densely puberulent or setulose, sometimes with stipitate glands intermixed, the hairs mostly 0.1-0.4 mm, occasionally up to 0.8, rarely to 1 mm. Leaf-blades ovate, lanceolate or elliptic, 2-10 cm long, the length (1.4-)l,7-2.7 x width, ± nigrescent, papery, triplinerved; base attenuate onto petiole from wid- est point or rounded and abruptly narrowed to an often attenuate or cuneate petiolar wing; apex usually acu- minate, sometimes acute; marginal teeth (14-)20-35(-50) per side, rounded, obtuse, or acute, spreading to appressed, then sometimes tip recurved, with sinuses 0.3-1.2 mm deep; adaxial surface dull to somewhat lustrous, antrorsely strigillose to strigose-pilose, the hairs occurring on veins and intervening tissue, forming a thin canopy of hairs only 0.3-0.5(-0.8) mm with understory of shorter hairs often well developed, (2-)20- 80(-200)/sq. mm, sometimes vitreous-pustulate, the circular bases of the strigae ca. 0.1-0.3 m in diam.; abax- ial surface slightly lighter or duller green than adaxial surface, moderately densely to sparsely pilose, if some strigiform hairs mixed in, then filiform hairs dominating, the hairs on all veins and intervening tissue, 0.05- 0.5 mm, all about same length (or those on areoles evenly much shorter), occasionally a few along midrib to 0.8 mm, 20-150/sq. mm. Inflorescences occasionally 2 per leaf axil, remaining hemispheric; peduncles 0.2-0.7(- 1.3) x leaf length. Proximal bracts linear-, narrowly elliptic-, or lanceolate-oblong (occasionally 1 or 2 outer- most bracts subfoliar or narrowly spatulate and distinctly longer), 4-7 mm long, 0.5-1. 5 mm wide, widest in proximal or middle third, with 3 veins from the base, appressed or spreading, usually deciduous after flower- ing; apex acute to attenuate, often rounded at very tip; indument pilose to strigillose, hardly ciliate, the longest hairs 0.2-0.3(-0.5) mm. Corolla opening yellow or white with yellow throat, aging to dark yellow, orange, or red, or opening white becoming infused with pink, blue, or purple, or opening pink, aging pink, purple, or white with yellow throat (those opening with yellowish pigments becoming infused with purple also expect- ed); corolla tube 7-12 mm. Distribution and habitat — The Planalto of eastern and southern Brazil, northeastern Argentina, and east- ern Paraguay; openings in tropical semi-evergreen forest, tropical savanna with gallery forest or areas of dense woodland, shrubland, and grassland; 0-1000 m. Lantana planaltensis to have arisen by hybridization between L. horrida subsp. tiliifolia and L. nivea subsp. nivea. Furthermore its geographic distribution far exceeds the zone of contact of the probable parental species, verifying its status as an independent species. The new name is required because L. hispida Kunth (= L. horri- da) already exists. Most of the plants annotated by H.N. Moldenke and me as “ Lantana triplinervia " are in- cluded here. Selected specimens examined: ARGENTINA. Bnenos Aires: Cabrera 7020 (SMU, VBD); Krapovickas 2 891 (SMU) Misiones: Ekman 1985 (F); Schwarz 3635 (SMU); Schwarz 4074 (F); Zuloaga et al. 6633 (MO). BRAZIL. Distrito Federal: Heringer 13834 (NY); Heringer et al. 4253 (NY). Golds: Anderson 9479 (NY). Mato Grosso do Sol: Hatschbach 49116 (LLjdi]); Salvador 3094 (US). Minas Gerais: Hatschbach 25966 (US), Hatschbach 46673 (NY); Mexia 5436 p.p. (US). Parana Hatschbach 11224 (F); Hatschbach 16038 (VBD); Hatschbach 241 54 (US); Hatsch- bach 41549 (NY); Wasum 2498 (BRIT); Winder 001 (BRIT). Peranamlmco: Silva etal.82 (MO, US). Rio de Janeiro: Carauta 3430 (LL[di]). Rio Grande do Sol: M acedo 5507 (NY); Sehnen 9562b (US); Wasum et al. 1425 (US); Winder 006 (BRIT). Santa Catarina: Reitz & Klein 1778 (F); ir rounded, open; st< often petioles thinly to densely si ovate or ovate-deltate to rotund, ( so, membranous to papery, pinnii marginal teeth 5-15 per side, acut 1.5 (longest o lairs 0.1-1.8 mm, the longest 0.8-1.8 mm. Leaf-blades broadly long, the length 1-1.5 x width, not nigrescent or only somewhat e rounded, truncate, or cordate; apex rounded to abruptly acute; , spreading, with sinuses (l-)1.5-5 mm deep; adaxial surface dull /illous, the hairs occurring on veins and intervening tissue, 0.1- f usually 0.7-1.5) mm, (2-)5-20/sq. mm, noticeably vitreous-pustulate or not, the circular bases of the hairs ca. 0.1-0.5 mm in diam.; abaxial surface slightly lighter or duller green than adaxial surface, thinly to moderately densely setose or setulose to pilose, the hairs on most veins and some intervening tissue, longest ones 1.5-2 mm on proximal portions of major veins, those increasingly distal gradually reduced (near margin ca. 0.7 mm), those on intervening tissue mostly 0.2-0.5 mm, 5-20/sq. mm. Inflorescences remaining hemispheric; peduncles 0.8-2.3 x leaf length (usually nearly twice when mature). Proximal bracts narrowly oblanceolate or spatulate to elliptic-oblong, (5-)7-12 mm long, (l-)1.5-3 (rare outermost one subfoliar to 5) mm wide, widest in distal half or near middle, with 3 veins from the base, appressed or spreading, persisting and reflexed from base in fruit; apex obtuse or acute; indument strigose, ciliate or not, the longest hairs 0.3-0.7 | mm. Corolla opening yellow, aging to red-orange; corolla tube 7-12 mm mm. Distribution and habitat . — Central and southern Texas, Mexico (Coahuila, Nuevo Leon, Tamaulipas), ap- parently cultivated and naturalized across the southwestern and southeastern United States from northern Texas to California and to Florida and North Carolina; open woodlands, brushland, thickets, and grasslands on calcareous clays or sandy soils; 0-1000 m. Lantana urticoides likely originated by hybridization between L. hirsuta subsp. hirsuta and L. kingii, hav- ing developed greater frost tolerance and a more northerly distribution than either parental species. Selected specimens examined: MEXICO. Tamaulipas: Dominguez & McCart 8231 (SMU).U.S.A. Texas. Aransas Co.: Uzzell 51 (US). Comal (SMU); San Patricio Co.: Jones 83 (SMU). Somervell Co.: Helm s.n. 9 May 1948 (SMU). Starr Co.: Garza et al. 8470 (SMU). Travis Co.: Han- sen 26 (VDB); Lundell & Lundell 8928 (SMU). Uvalde Co.: Dickey 70 (SMU). Willacy Co.: Lundell & Lundell 8751 (SMU). Presumed hybrids with: 20. L. strigocamara. MEXICO. Coahuila: Hazard s.n. May 1883 (US). Nuevo Leon: Rodriguez 62 (SMU); Dodge 100 (NY). Tamaulipas: Berland ier s.n. 1836 (NY). U S A. Aiabama. Baldwin Co.: Krai 39530 (NY, VDB). Crenshaw Co.: Diamond 11455 (BRIT[di]). Florida. Citrus Co.: Krai 4542 (SMU). Marion Co.: Slaughter 13954 (BRIT[di], SMU). Monroe Co.: Krai 53896 (VDB). RIT(di], NY). South Carolina. Orangeburg Co.: Leonard et al. 5001 (VDB). Texas. Blanco Co.: Sanders 5143 (SMU). C » 6244 (US). Dallas Co.: cult. Niblack 50 (SMU). Fayette Co.: Krai 68519 (VDB). Galveston Co.: Waller 2579 (US). Sai MU). Tarrant Co.: Krai 91937 (VDB); Whitehouse 16027 (SMU). Val Verde Ca: Spjut & Marin 15152 (BRIT). See also Sanders, Taxonomy of Lantana sect. Lantana 435 e FRENCH GUIANA: Grand-Santi, 26 Aug 1961, Schndl ;: BRAZIL. ParA: 0° 55’S, 54° 26’W, 23 Jul 1981, Strudwick Shrubs rounded and ± open, lax, or subscandent; stems 0.5-5 m; branches ascending and several or clamber- ing and few, occasionally herbaceous; twigs, peduncles and often petioles moderately setose or pilose, often with stipitate glands mixed in, or occasionally glabrescent, the hairs (0.1-)0.3-0.6(-l) mm. Leaf-blades broadly to narrowly ovate or ovate-elliptic, 3-9 cm long, the length 1.4-2.2 x width, nigrescent or not, papery, usually pinninerved; base rounded to subcordate and abruptly tapered onto petiole or cuneate; apex acumi- nate, sometimes with a prolonged narrow tip, or acute; marginal teeth 15-35 per side, rounded or acute, often appressed, then sometimes with tips recurved, with sinuses 0.5-1.5 mm deep; adaxial surface dull to occasion- ally lustrous, antrorsely strigose or strigose-setose, the hairs occurring on veins and intervening tissue, 0.1- 0.6(-1.2) mm, 3-20/sq. mm, usually not noticeably vitreous-pustulate, the circular bases of the hairs ca. 0.1- 0.2 mm in diam.; abaxial surface slightly lighter or duller green than adaxial surface, sometimes whitish green but not glaucous, antrorsely strigose to spreading-setose, with the hairs frequent on veins and veinlets, 0.1-0.7 mm, 3-10/sq. mm. Inflorescences often 2 per leaf axil, becoming short-cylindric (resembling spikes of Carex lupulina ) by prolonged initiation of flowers; peduncles 0.5-l(-2) x leaf length. Proximal bracts lanceolate to lanceolate-elliptic, (4-)6-8(-10) mm long, (1.5-)2-3 mm wide, widest in proximal third to just below middle, with 5-7 veins from the base, spreading, persisting and recurved or reflexed (± distally) in fruit; apex acumi- nate with prolonged tip; indument setose or strigose to glabrescent, usually distinctly ciliate, the longest hairs (0.3-)0.5-l mm. Corolla rose-pink or white with pale yellow throat and aging pinkish purple, rarely yellow to red-oranage; corolla tube 7-12 mm. Distribution and habitat . — Brazil (central and eastern Amazonian), the Guianas, Venezuela (Amazonian) Colombia (Amazonian),and Bolivia (Amazonian); disturbance openings, savannas and man-made grasslands in tropical humid forest; 50-300 m. Lantana paraensis exhibits characters of both L. cujabensis and L. viscosa , suggesting that it arose from natural hybrids of the two. Its species status is suggested by greater consistency in the expression of its traits than expected of a hybrid swarm and the much wider distribution than the area of sympatry of the proposed Selected specimens examined: BOLIVIA. Santa Crnz: Guillen &Roca 3523 1 Croat 62252 (MO); Tsugaru. & Sano B-598 (MO). Para: Ginzberger 822 (F); P et al. 664 (F, NY). COLOMBIA. Vichada: Davidse5196 (MO). FRENCH G (MO). SURINAME. Paramaribo: Florschutz&Florschutz 1693 (SMU). Sipal *6>Black27459 (LLldi]). Amazonas: 20. Lantana strigocamara R.W. Sanders, Sida 22:392. 2006. Type: U.S.A. Florida. Dade Co.: Coral Gables, 23 Sep 1981, Sanders 1450 (holotype: FTG!; isotype: NY!). Shrubs erect or rounded, open; stems 0.3-3 m; branches ascending, several to numerous; twigs, peduncles and often petioles thinly to moderately strigose, setose, or pilose, the hairs 0.1-1.2(-1.5) mm, the longest mostly 0.5-1 mm. Leaf-blades ovate to broadly ovate, (2-)5-10 cm long, the length 1-1.7 x width, usually not nigres- cent, papery, pinninerved; base rounded, truncate, or cordate, shortly and narrowly cuneate onto petiole at very base; apex usually acuminate; marginal teeth 15-40 per side, rounded to acute, often appressed, some- times spreading at tip, with sinuses 0.5-1.5 mm deep; adaxial surface usually dull, antrorsely strigose or stri- gose-setose, the hairs occurring on veins and intervening tissue (sometimes just center of areoles), 0.2-1.2 (longest mostly 0.5-0.8) mm, 1-12/sq. mm, not noticeably vitreous-pustulate, the circular bases of the strigae ca- 0.1-0.2 mm in diam.; abaxial surface slightly lighter or duller green than adaxial surface, antrorsely stri- gose-scabrous, with the strigae scattered to moderately dense on veins and veinlets, 0.1-0.6 (longest ones usu- ally 0.4-0.6) mm (sometimes accompanied by scattered short [mosdy <: 0.3 mm] erect filiform hairs along Sanders, Taxonomy of Lantana sect. Lantana . horrida subsp. horrida 160. 1978. Type: MEXICO. Durango: 1 Iex2axl0. Lantana camara subsp. glandulosissima x L. horrida subsp. horrida x L. kingii Lantana urticoides Hayek f. aculeata Moldenke, Phytologia 49:182. 1981. Type: MEXICO. Puebla: TehuacSn, 4 Aug 1966 Mex-28 (holotype: MEXU[di!]). Lantana mdu ma ioldei 2:2 1982.1 PAN \M \ I on Sherman, 14 ]un 1923, M axon& xlO. Lantana camara subsp. glandulosissima x L. kingii Lantana glandulosissima Hayek f. parvifolia Moldenke, Phytologia 49:182. 1< 7933 (holotype: MEXU[di!]). e: MEXICO. San! lexl0/20. Lantana camara subsp. glandulosissima x L. kingii or L. strigocam los: Yantepec, 14 Aug 1950, Wyatt 45 (holotype: MEXU[di!]). Ifx2. Lantana camara subsp. aculeata x L. horrida (probably subsp. horrida) / iimwii.i niiKnlulivL I W'cieel. I'lijsioer. SaUk. I l.tndl I 4(> r’l. I \i i > nil < ireilsuald llr.i ( i.iril . > lfx4. Lantana camara subsp. aculeata x L. hirsuta Lantana mista L., Syst. Nat., ed. 12. 2:417. 1767. Camara aculeata (L.) Kuntze [var. subinermis Kuntze Lantana albopurpurea Desf., Tabl. Ecole Bot., ed. 3 (Cat. PI. Horti Paris) 392. 1829. Type: cult., Hort. Paris, s. coll, (le lfxC. Lantana camara subsp. aculeata x L. sp. Ser. Strigosae 2xC. Lantana horrida (probably subsp. horrida) x L. sp. Ser. Strigosae 2ax4a. Lantana horrida subsp. horrida x L. hirsuta subsp. hirsuta Lantana poly acantha Schauer, Prodr. [A.P. de Candolle] 11:597. 18' 17481, F!]; lectotype, here designated: P, barcode P00713484[di!]) Lantana horrida Kunth f. inermis Moldenke, Phytologia 52:130.1982. Type MEXICO: Yucatan, Gamier 808 (holotype US!). iaxlO. Lantana horrida subsp. horrida x L. kingii dez424 (holotype: MEXU[di!]). 2ax20. Lantana horrida subsp. horrida x L. strigocamara V 1948, Moldenke & Moldenke 19861 (holotype: N L. SAo Paulo: Muji-Gaugu, 31 Jul 1! 2cxl2b. Lantana horrida subsp. tiliifolia x L. nivea subsp. mutabilis ^Wona camara L. f. glandulosa R. Fern., Bol Soc. Brot. sir. 2, 61:132. 1988. Type: ANGOLA, t (holotype COl, n.v.; isotypes: BM|dil], K(dil], USU p.p.). 438 Journal of the Botanical Research I ? of Texas 6(1) 2cxl3. Lant 2cxl7. Lantana horrida subsp. tiliifolia x L. planaltensis 2185 (holotype: NY!). Alternatively nivea subsp. mutabilis could possibly be the second pare 2/4x? Complex hybrid involving Lantana horrida or L. hir 2. Type: MEXICO. Nuevo Le6n: Cafion Dient Lantana horrida xinth f. bracteosa Moldenke, PhytologiT 52:231. 1982. Type: MEXKX). Puebla: Puebla, 15 Sep 1910, Arsdne 5426 (how- 4bxl3. Lantana hirsuta subsp. amazonica x L. cujabensis Lantana cujabensis Schauer var. hispida Moldenke, Phytologia 46:58.1980. ' 4bx 15. Lantana hirsuta subsp. amazonica x I ... micrantha 9i-cvx20. L. Callowiana Hybrid Group cultivars (derived from tetraploid cv. L. depressa var. depressa x L. strigocamara) viana Monrovia Nursery, Monrovia Nursery Catalog 1952-1953:44. 1952. nom. illeg. Type: none. ra L. var. nana Moldenke, Phytologia 28:402. 1974. Lantana camara L. f. nana (Moldenke) Moldenke, Phytologia 45:296 e: U.S.A. New Yokx: cult.. New York. Bot. Gard., 14 Oct 1941, Moldenke & Moldenke 11903 (holotype: NY!). This is either an r an independent cultivation of a wild-collected hybrid betw Moldenke & Moldenke 29885 (holotype: LL, n.v.; isotype: !). Lantana Callowiana Hybrid Group ‘Cre n L. depressant. id, 20 May 197V Sanders (2001) argued that the parents of the Callowiana Hybrid Group were Lantana strigocamara and L. de- pressa var. depressa rather than L. strigocamara and L. montevidensis (as claimed by Monrovia Nursery, see How- ard 1969) based on character intermediacy and chromosome number incompatability of the latter combina- tion. However, one likely hybrid of L. montevidensis with L. strigocamara was seen for this study (see taxon 20), but it is very different in character details from the Callowiana Hybrid Group cvs, as well as appears to be sterik- 9iix20. Lantana depressa var. floridana x L. strigocamara lOxla/le? Lantana kingii hybrid (x L. c antana kingii hybrid (x L. horrida subsp. horrida or L. hirsuta subsp. hirsuta^ 10x2a/4a? I 439 12x6/7? Lantana nivea x L. scabrida o 12axl4. Lantana nivea subsp. nivea x L. viscosa Lantana pohliana Schauer, Prodr. [A.P. de Candolle] 11:601. IE BRAZIL. GoiAs: “Inter Pirapora et Jenipapa,” Pohl 3088 (I strayed [MacbrideNeg. 17480 at BRIT!, F!, GH!], F [fragm The collections known to me combine the reduced leaves subtending inflorescences typical of some Lantana nivea subsp. nivea (see comments, taxon 12a) and the stipitate glands, bracts, and elongating receptacles of L. viscosa. They are geographically restricted near the type locality of L. pohliana, which is the area of sympatry of the two species. For a narrowly endemic taxon, they exhibit a pronounced inconsistency in structure, length, and density of trichomes compared to other natural taxa. These data suggest that the collections represent in- dependent spontaneous hybrids, or at most, an unstable hybrid swarm. Thus, L. pohliana is not recognized as 12bxl7. L. nivea subsp. mutabili ;e) Moldenke, Phytologia 28:403. 1974. Type: BRAZIL. Bahia: Cocos I? L. micrantha x L. cujabensis or L. strigocamara 2:468.1948. Type: ARGENTINA. Chaco: Colonu 18x20. L. urticoides x L. strigocama Journey Mex., t.5 (top, facing p. 410). IS EXCLUDED AND DUBIOUS NAMES Lantana asperata Hort. ex Vis., Orto Bot. Padova 142. 1842. nom. nud. Type: unknown. Lantana hahiensis Turcz., Bull. Soc. Imp. Naturalistes Moscou 36:206. 1863. Type: BRAZIL. Bahia: Salzrnann s.n. (holotype: KW[di!]). Species of Lantana sect. Callioreas. (Misapplied to L. planaltensis by Moldenke in sched.) c Lantana Camara L. var. rosea Mattoon, Plant Buyer’s Guide, ed. 6. 167. 1958. nom. nud. Lanlam Camara L. f. r«eu(Mattoon) Moldenke, Phytologia 45:296. 1980. nom. illeg. Type: none. Lantana camara L. var. rubra Mattoon, Plant Buyer’s Guide, ed. 6. 167. 1958. nom. nud. Lantana camara rubra (Mattoon) Moldenke, Phytologia 45:296. 1980. nom. illeg. Type: none. Lantana cujabensis Schauer var. punctata Moldenke, Phytologia 2:411. 1948. - Lantana lopez-palacn Moldenke, Phytologia 27:359. 1973. Type: COLOMBIA. Antioquia: Ceja, 1 Nov 1947, Barkley et al. 1536 (houttype: MEDEL,n.v.). Species of Lantana sect. Callioreas. Lantana hispida Kunth f. alba Moldenke, Phytologia 9:99. 1963. Type: GUATEMALA. PetEn: Ttka rums, jun I960, Contreras 1056 (holotype: LL!). Aff. L. hirta Grah. of sect. Callioreas. «— hispida Kunth f. parvt/dia Moldenke, Phytologia 52:130.1982. Tvm: HONDURAS. M^ Ct^ad Univereitaria, 26 May 1978, Romero 71 (uoutrYm: MOD. Aff. L. velutina M. Martens & Galeom of sect Lwa^m !ZL Briq. var. beck,, Moldenke, Phytologia 50:13. 1981. Tvrr: BOLIVIA. Bern: BallivMn, 12 Apr 1981, Beck 5339 (holotype: LL!). Aff. Lfucata Lindl. of sect. Callioreas. Lantana multicolor Lem., FI. Serresjard. Eur. 3:239. 1847. nom. dub. Type: unknown. Lantana notha Moldenke, Phytologia 1:422. 1940. Type: MEXICO. StNAUts: Fuerte, 27 Mar 1910, Rose et al. 13573 (holotype: NY!). Aff. L. hirta Grah. of sect. Callioreas. bantam pulchra Larrahaga, Escritos DSntaso Antonio Larranaga 1:406. 1922 [Pub. Inst. Geog Uragl. nom. dub. Type: Destroyed. Lantana purpurea (Jacq.) Benth. & Hook!., Gen. Pi. [Benth. & Hooker f.] 2(2):1142. 1876. nom. illeg. (non Hor- nem.) Species of Lippia or Lantana sect. Callioreas. Lantana riedeliana Schauer var. pubescens Moldenke, Phytologia 19:435. 1970. Type: BRAZIL: Rio de Janeiro, Pabst 9310 (holotype: LL!). Aff. L.fucata Lindl. of sect. Callioreas. Lantana rosea Raf., Sylva Tellur. 83. 1838. Type: unknown. Probably a species of Lantana sect. Callioreas. EPITHET INDEX (USING TAXON AND HYBRID SYNONYMY CODES’, EDN=EXCLUDED AND DUBIOUS NAMES) aculeata:la, If, 12a, 12b, Iex2ax6, lfx4, 2ax20, 2/4x?, 10xla/le?. aculeatissima:lex2a. aculeifera:ldx2c. alba:12a, EDN. albiflora:le, 19, lexl0/20, 2cxl7, 9i-cvx20. albopurpurea:lfx4. amazonica:4b amethystina: 1 2b. antidotalis:lx2xl2. antillana:2. arida:lb, 2b, 2bi. armatarld, 12b, 19, ldx6, 2cx4b, 4bxl5. asperata:EDN. aurea:9i. bahamensis:9ii, 16, laxl6, 9i-cvx20. bahiensis:EDN. bartramii:9iix20. beckii:EDN. bracteosa:4axl0. brittonii:6. caffertyi:lax6. callowiana:9i-cvx20. camara:l, la-le, 2a, 2b, 7, 12a, 12b, 14, lax6, ldx2c, ldxl2b, lexl0/20, lfx4, 2xC, 2ax20, 2cxl2b, 9i-cvx20, 10xla/le?, 12x6/7?, 12axl4, EDN. canescens:laxl6. coccinea:lf. crenulata:6. crocea:la, 6. cujabensis:13, 19, 4bxl3, EDN. cummingiana:2c. depressa:9, 9i-9iii. eitenorum:2cx4b. flava:le, 6, 2xC. floridana:9ii, 9iix20. foetida:2c formosa:la. g!andulosa:2c, 2cxl2b. glandulosissima:le, 2bi, lex2a, lex4a, lexlO. glutinosa:2c, 2cxl3. grandiflora:2a. grandis:lex4a. guatemalensis:6. guianensis:19. hirsuta:4, 4a, 4b. hirta:4axl0. hispida:2a, 17, 4bxl3, 10x2a/4a?, EDN. hispidula:18. hodgei:8. horrida:2, 2a-2c, 18, lex2a, 2ax4a, 4axl0. hybrida:2/4x? incarnata:12b. inermis:2ax4a. insularis:5. kingii:10. latibracteata:18. leonardio- rum:3. longibracteata:4bxl2a. lopez-palacii:EDN. macrantha:2a. macrophylla:2axl0 micrantha:15 2cx4b, 4bxl5, 15x13/20?, EDN. microphylla:lex2a. minasensis:12a, 17, 4bxl2a, 12bxl7. mista:lf, If* 4 moldenkei:lc. montevidensis:9i. morii:12a. moritziana:ld, le, ldx2c, lexl0/20. multicolor:EDN. mul- tiflora: 12x6/7?. mutabilis:lf, 12b, 20, lfx2. nana:9i-cvx20. niveailf, 12, 12a, 12b. normalis:lf, 12b. notha:EDN. obtusifolia:la orientalis:2c. ovatifolia:9i, 11. paraensis:19. parviflora:2a. parvifolia:le 9i, 13, lexlO, 10xla/le?, EDN. planaltensis:17. pohliana:12axl4. polyacantha:2ax4a. portoricensis:lb puberulenta: 1 2bx 17. pubescens:EDN. pu!chra:EDN. punctata:EDN. purpurea:lfxC, EDN. redinata:9i riedeliana: 13, EDN. robusta:4bxl2a. rosea:EDN. rubella:2ax20. rubello-flavescens:ldxl2b. rubra:18x20, EDN. ragosa:2cxl3. sandersii:6 sanguinea:lf. sanibelensis:9iii sargentii:2bi. scabra:ldx2c. scabrida:6 scabrifolia:13. scandens:6. scorta:4a. splendens:7. strigocamara:20. suaveolens:lf. subcordata:2bii subinermis:la, If, 12a, 12b, lfx4. tenuifolia:13. temata:2a, 4axl0, 10x2a/4a? temifolia:ld. tiliifolia:2c, ldx2c, 2cxl7. triplinervia:12a, 12b, 17, 4bxl2a, 12bxl7. urticifolia:la, lb, lc, 2b. urticoides:18, Iex2ax6, 2axl0. varia:lf. variagatarlf. velutina:ldx6 violacea:15xl3/20? viscosa:14. vulgaris:la. weberbaueri:2cx4b zanonii:2b, 2bi, 2bii. ACKNOWLEDGMENTS I thank the following herbaria for searches, loans, digitization of specimens, and access to collections: A, AAU BH, BM, BM-SL, BR, BRIT, B, B-WILLD, C, DWC, E, F, FI, FTG, G, G-DC, GH, GOET, HIPC, HOH, JE, K, KW, LASCA, LINN, LIV, LL, M, MANCH, MEXU, MO, MPU, MSC, MVFQ, NY, OXF, P, P-HBK, PAD, PH, PL RB, & SMU, STU, TENN; TEX, UC, UPRRP, UPS, US, VDB, W, WECO, WIS, WLU, WS, WTU, WU. Special thanks go to the curators of BRIT and TEX for extensive digitization and of TENN for hosting loans. Financial support was provided by the Appalachian College Association and Queensland Department of Primary Industries- Constructive comments by Michael Nee and an anonymous reviewer are appreciated. REFERENCES Howard, R.A. 1969. A checklist of cultivar names used in the genus Lantana. Arnoldia 29:73-109. Maschinski, J E. Sirkin, AND J. Fant. 201 0. Using genetic and morophological analysis to distinguish endangered taxa fro*" their hybrids with the cultivated exotic pest plant Lantana strigocamara (syn: Lantana camara). Conservation Gen* 11:1607-1621. 146). Offutt, ICE. and R.W. Sanders. 201 2. Identification guide to the Lantana camara complex: an interactive, multi-access key. Bryan College, Dayton, TN. Accessed November 23, 2012 at http://www.bryancore.org/sliks/ Rotman, A.D. and M.E. Mulgura de Romero. 2010. Novedades nomenclaturales en los generos Lippia y Lantana Sanders, R.W. 1 987a. Identity of Lantana depressa and L ovatifolia in Florida and the Bahamas. Syst. Bot. 1 2:44-60. Sanders, R.W. 1987b. Taxonomic significance of chromosome observations of Caribbean species of Lantana (Verbenaceae). Amer. J. Bot. 74:914-920. Sanders, R.W. 1987c. A new species of Lantana (Verbenaceae) from Dominica, Lesser Antilles. J. Arnold Arbor. 68: 343-348. Sanders, R.W. 1989. Lantana sect. Camara (Verbenaceae) in Hispaniola: novelties and notes. Moscosoa 5:202-215. Sanders, R.W. 2001 . The genera of Verbenaceae in the southeastern United States. Havard Pap. Bot. 5:303-358. Sanders, R.W. 2006. Taxonomy of Lantana sect. Lantana (Verbenaceae): I. Correct application of Lantana camara and Santos Silva, T.R. 2001. Lectotypifications and neotypifications in Lantana and Lippia (Verbenaceae). Taxon 50: 1115-1118. Schauer, J. C. 1 847. Verbenaceae. Prodr. [A.P. de Candolle] 1 1 : 522-700. Schauer, J. C. 1851 . Lantana. FI. Bras. [Martius] 9: 251-266. Stebbins, G.L 1966. Processes of organic evolution. Prentice-Hall, Englewood Cliffs, NJ. Urban, 1. 1906.Collectores: Riedel, Ludwig. FI. Bras. [Martius] 1(1):89-91. Journal of the Botanical Research Institute ofTexas6(2) BOOK REVIEW Craig Pittman. 2012. The Scent of Scandal: Greed, Betrayal, and the World’s Most Beautiful Orchid. (ISBN: 978-0-8130-3974-9, hbk.). University Press of Florida, 15 Northwest 15th St., Gainesville, Florida 32611- 2079, U.S.A. (Orders: http://www.upf.com). $24.95, 299 pp., b/w photos, 6" x 9 M . After carefully reading this book, I turned to the back cover, and read the small heading: “TRUE CRIME/GAR- DEN1NG.” The first review stated: “FANTASTIC. If I did not know most of the main players I would have thought the author had a vivid and twisted imagination.” — Paul Martin Brown, author of Wild Orchids of Florida ” — (and also a number of other orchid books and papers.) 1 was delighted. AND I had had about the same reaction as he did. I had worked with Paul when 1 was Sr. Technical Editor for the Flora of North America j North of Mexico project and had edited some of his contributions to the Orchidaceae. I also edited some of his j other orchid manuscripts separately. I have tremendous respect for his contributions — AND his knowledge of orchids. Long, long ago, I had married the son of a well-known African Violet (and companion plants) family. After ; college and a stint in the Air Force, we moved back to Illinois and setded down in a nearby small town. We im- mediately got involved with the plants (Gesneriaceae and Orchidaceae, especially) — AND my mother and I were quickly introduced to the monthly African Violet Club. (My father lovingly considered us all “a bunch of delightful screwballs”) People who REALLY love orchids, African violets, and companion plants really do tend to get very in- volved, and often competition can get pretty nasty. Of course, there are local shows, state shows, regional, and national shows. Ribbons are nice to earn, but tempers can really skyrocket at times. The more involved one is — and the more competition gets heated— well, it can wreak havoc, even at the local level. When you get in- volved enough to go to the state shows and the national shows, most people are really committed to doing their best. However, while competition for new varieties is always a big challenge, when it comes to finding beautiful and truly new species in native soils in other countries, it can become truly illegal, involved, and potentially unlawful to bring in plants collected in other countries and illegally slipped through customs. The Scent of Scandal is an actual, very carefully documented account of a rare and unusually beautiful or- chid that got into the country, was recognized as an extremely new and beautiful specimen, caused all kinds of interest, all kinds of trouble, was illegally named, and becomes an absolutely true, carefully documented case history. Craig Pittman has done an incredible job of researching all strands of information, documentation, actions (good or bad), timing, and ultimately providing 242 pages of page-turning fascination, 43 pages of fine print, and a bibliography, tracing the path of one extremely beautiful, very rare, illegally named orchid but Stitt bound to the name given it according to legal practice. This is a page turner— with real people, real emotions, good intentions, devious actions, careless deci- sions, and a very beautiful plant— legally or illegally “officially named.” It prompts numerous concerns all the way: legalities, illegal actions, some knowingly, others perhaps not so knowingly. And every bit is carefully researched and documented by the very capable author . — Helen Jeude, Volunteer and Assistant Editor, Botanic d Research Institute of Texas, 1700 University Dr., Fort Worth, Texas 76107-3400, U.S.A. CYTOTYPIC VARIATION IN PHLOX P1LOSA SSP. PILOSA (POLEMONIACEAE) AT THE WESTERN EDGE OF ITS RANGE IN THE CENTRAL UNITED STATES Lindsey Worcester, Mark H. Mayfield, and Carolyn J. Ferguson Herbarium and Division of Biology Kansas State University Manhattan, Kansas 66506-4901, U.SA ferg@ksu.edu RESUMEN Polyploidy is frequent in plants and is considered an important factor in plant evolution (Otto & Whitton 2000; Adams & Wendel 2005; Soltis et al. 2009; Soltis et al. 2010). Autopolyploidy, or genome doubling within a species, can lead to intraspecific ploidy level (cytotypic) variation (although some autopolyploids alterna- tively meet criteria for species recognition; see Soltis et al. 2007). Such cytotypic variation within species may be underappreciated, as systematists have sometimes made assumptions about ploidy levels for entire taxa based on one or few chromosome counts. Advances in flow cytometry methods enable rapid assessment of genome size and inference of ploidy level (Dole&l & BartoS 2005), and recent studies have demonstrated in- triguing cytotypic variation in some species (e.g., Balao et al. 2009; Cires et al. 2009; Whittemore & Olsen 2011). Documentation of cytotypic variation is a critical first step toward understanding ecological, genetic, and taxonomic consequences of autopolyploidy in particular groups. Phlox pilosa L. (Polemoniaceae) is a showy, upright, perennial species ranging throughout most of the eastern United States, from the east coast to the edge of the Great Plains and Central Texas (Wherry 1955; Great Plains Flora Association 1986; Gleason & Cronquist 1991). Phlox pilosa exhibits noteworthy morphological variation across its range, and several subspecies are usually recognized (Wherry 1955; Levin & Smith 1965; Levin 1966; Locklear 2011; see also Ferguson 1998). Phlox pilosa ssp. pilosa occurs in prairies and woodland openings across most of the general range of P. pilosa as a whole, with westernmost populations occurring in north central Texas and south central Oklahoma. Most published chromosome counts for P. pilosa are diploid, and represent the wide-ranging P. pilosa ssp. pilosa (e.g., Kelly & Wahl 1928; Flory 1931 [P argillacea Clute & Ferris, a taxonomic synonym], 1934; Meyer 1944; Levin & Smith 1965; Levin 1966, 1968; Smith & Levin 1967; see also Levin & Schaal 1970; Levy & Levin 1974). However, there has long been a suggestion of polyploidy near the western edge of the range of P. pilosa ssp. pilosa. Smith and Levin (1967) reported a tetraploid chromo- some count for material of P. pilosa ssp. pilosa from a mile south of Sachse, Texas (Dallas County, in north central Texas) and one tetraploid count for material of P. pilosa ssp. pilosa from the state of Arkansas (precise locality unknown). This has led to some question regarding ploidy level and possible cytotypic variation in P. Phlox pilosa is one of the most thoroughly studied species of the genus Phlox (due particularly to the extenr sive systematic and ecological research conducted by D.A. Levin and colleagues in the 1960s and 1970s; e.g., citations listed above), and further work on P. pilosa advances the utility of Phlox as a study system. Phlox is a genus of ca. 60 species of perennial herbs occurring mostly in North America (Wherry 1955; Locklear 2011; Ferguson et al. in prep. [FNA vol. 15]). The base chromosome number for Phlox is x=7 (Flory 1934), and diploid, tetraploid and hexaploid chromosome counts have been reported (see Kelly & Wahl 1928; Flory 1931, 1934, 1937, 1948; Meyer 1944; Levin 1964, 1966, 1968; Levin & Smith 1965; Eater 1967; Smith & Levin 1967; Love 1971; Strakosh 2004; Fehlberg & Ferguson 2012 and in press). Furthermore, polyploidy has been implicated as a factor contributing to complicated patterns of phylogeny, including incongruence between nrDNA and cpDNA phylogenies for samples of P. pilosa and close relatives (Ferguson & Jansen 2002; see also Ferguson et al. 1999). Study of cytotypic variation within P. pilosa ssp. pilosa can thus be placed within an extensive broad- In this study, patterns of cytotypic variation in P pilosa ssp. pilosa were explored at the western edge of its range, from north central Texas to eastern Oklahoma, western Arkansas, and southeastern Kansas. Flow cy- tometry was conducted chiefly on fresh field-collected leaf material, as well as previously preserved material for some populations (silica gel-dried, frozen, or herbarium material). Meiotic chromosome counts were per- formed on bud material from several populations to enable ploidy level inferences based on genome size (Dole&l et al. 2007). MATERIALS AND METHODS Field collection. — Material was collected from over 50 populations of P. pilosa ssp. pilosa ranging from the gen- eral area of the earlier reported tetraploid count in Texas (Smith & Levin 1967) and north along the western edge of the taxon range as far as southeastern Kansas. Fresh leaf material from an individual plant from each population was reserved for flow cytometry. For some populations, leaves from multiple individuals were col- lected in silica gel to enable later testing for cytotypic variation within populations. For populations at an early stage of flowering, bud material for chromosome counts was collected (and preserved in a solution of 3 parts 95% ethanol: 1 part glacial acetic acid, with subsequent transfer to 70% ethanol). Voucher specimens fromaB populations were prepared and deposited in the Kansas State University Herbarium (KSC). This study was further augmented with material from several populations previously collected (material that had been col- lected in silica gel or fresh-frozen and stored at -70°C, and one recent herbarium specimen). Flow cytometry. — DNA content per cell was assessed for individuals using flow cytometry. At least one individual was sampled per population, with additional within-population sampling for 11 populations. Leaf material (0.04-0.10 grams) was chopped with a fresh razor blade in 2 mL of chopping buffer specified by Da- vison et al. (2007; modified from Bino et al. 1993). The resulting slurry was filtered through 30 pm nylon mesh (Small Parts, Inc.), followed by vortexing and centrifugation (500 RCF for 7 minutes). The supernatant was decanted off of the pellet, or if a pellet did not form, approximately half of the liquid was decanted. Propidiu® iodide staining solution (BioSure; 700 pL) was added to each sample to fluorescently stain the DNA; and 2 pf of chicken erythrocyte nuclei (CEN) singlets (BioSure) were added to each sample as an internal standard Samples were vortexed and stored on ice for 1-2 hours prior to processing on a Becton Dickinson FACS-Cafr bur flow cytometer at the Flow Cytometry Lab of the College of Veterinary Medicine, Kansas State University The flow cytometer quantifies the amount of DNA by measuring the fluorescence per nucleus. The result' Worcester et al., Cytotypic variation in Phlox pilosa ssp. pilosa ing histogram for each sample was analyzed using CellQuest software (Becton Dickinson), and a coefficient of variation (CV) was obtained. To ensure highest quality data, we aimed to retain samples with measurements based on ca. 10,000 nuclei per sample and having CV values <5% (see Galbraith et al. 1997; Dole£el & Bartos 2005). Picogram (pg) amounts were then calculated using the equation from Dolezel and BartoS (2005). Chromosome counts. — Chromosome counts were conducted using a modified version of B.L. Turner’s squash technique for meiotic cells (Jones & Luchsinger 1986). These counts enabled linking of picogram DNA amounts to chromosome numbers for P. pilosa ssp. pilosa, and thus inference of ploidy levels based on flow cytometry data. Flow cytometry data were retained for 46 populations: these samples had 10,005-11,205 nuclei scored and CV values from 1.91-4.98% (Table 1). Flow cytometry worked well for the Phlox material; most runs yielded high quality data, although some samples did not meet criteria for inclusion (see Materials and Methods; due to lower numbers of events, high CV values, or, occasionally, poorly defined peaks). Based on these data (Table 1), cytotypic variation in P. pilosa ssp. pilosa was inferred, with 13 diploid populations, 30 tetraploid populations, and three unusual populations with higher DNA content (discussed below). DNA content per nucleus for the diploid and tetraploid populations averaged 10.66 pg (range 9.22-13.36 pg) and 21.90 pg (range 18.27-24.81 pg), respectively (based on measurements of all individuals; Table 1). The three remaining populations had an average DNA content per nucleus of 26.55 pg (range 23.59-29.46 pg; Table 1). For 11 populations, additional individuals (2-9) were analyzed. No cases of intrapopulational cytotypic variation were inferred (Table 1). However, measured within population variation in picogram values ranged from 0.26 pg (in diploid population 16) to 4.13 pg (in tetraploid population 17; Table 1). Chromosome counts were obtained, including a diploid count (n= 7) and a tetraploid count (n=14; Fig. la, lb; Table 1). Cells undergoing meiosis were also observed for two of the three populations that exhibited high genome size values (populations 12 and 13). Cells from some buds of population 12 showed clear tetraploidy (n=14), while those from other buds suggested additional chromosomes or chromosomal fragments (although clear counts could not be obtained). Cells from buds of population 13 appeared to have more chromosomes than the tetraploid material (n=17+), although the chromosomes could not be sufficiently spread apart. While it is possible that higher level polyploids are present, these results may also be due to the presence of B chromo- somes (supemumary chromosomes; see Smith & Levin 1967 for reports of B chromosomes in P. pilosa; see also Meyer 1944 for notation of “fragments” in meiotic figures of some Phlox taxa). Mapping the cytotypic data for sampled populations revealed a generally east-west pattern (Fig. 2), with diploid populations occurring in the eastern part of the sampling range, tetraploid populations in the western part, and the three populations with unusually high DNA content in southeastern Oklahoma (Fig. 2). DISCUSSION This study demonstrates that P. pilosa ssp. pilosa occurs as diploid and tetraploid populations at the western e dge of its range from the southeastern Great Plains (sensu Great Plains Flora Association 1986) and south into northern Texas. This variation does not correspond to any previously recognized infraspecific taxa in P pilosa, and no morphological differences were noticed among cytotypes (future work will investigate micromorphol- ogy of these populations). Interestingly, two subspecies of P. pilosa occurring further west in Texas (onto the Edwards Plateau) are known to be tetraploid (P. pilosa ssp. latisepala Wherry and P. pilosa ssp. riparia Wherry; see Levin 1966, 1968; Smith & Levin 1967). Some workers have recognized these Central Texas taxa at the specific level, partly due to the polyploid condition: P aspera E.E. Nelson and P. villosissima (A. Gray) White- house (Levin 1968; Levin & Schaal 1970; Levy 1973); or, as a single species under P villosissima (Turner 1998; Locklear 2011). Our documentation of tetraploid populations of P. pilosa ssp. pilosa in northern Texas (and northward) indicates that there are no ploidy level differences precluding intergradation with the Central Texas taxa. Investigation of population level morphological, genetic, and cytotypic variation are warranted is (in P pilosa ssp. latisepala and P. within P pilosa thus occurs in the region sampled in the present study 0 northern Texas; in P. pilosa ssp. pilosa), west and south through Central Te pilosa ssp. riparia). Ploid, three populations in southeastern Oklahoma had unusually high DNA content per nucleus, and com- pound light microscopy of meiosing cells suggested presence of chromosomes (or chromosome fragments) beyond the tetraploid number. Further cytological study of these populations will be necessary to definitively document ploidy. Accessory, B chromosomes have been documented in P pilosa ssp. pilosa (Smith & Levin 1967 > »n populations from central and northern Alabama, northeastern Arkansas, and northwestern Indiana), and may well explain the variation detected here. Although perhaps less likely, higher ploidy levels are also Journal of the Botanical Research Institute of Texas 6(2) Worcester et al., Cytotypic variation in >. pilosa 449 known in Phlox. For example, Fehlberg & Ferguson (2012; in press) have found diploid, tetraploid and hexa- ploid cytotypes in R amabilis Brand and P woodhousei (A. Gray) E.E. Nelson (with evidence from chromosome counts, flow cytometry, and population genetic data); Eater (1967) obtained diploid, tetraploid and hexaploid chromosome counts for P nana Nutt, (following current taxon recognition; e.g., Wilken & Porter 2005); and a hexaploid count is known for P. andicola E.E. Nelson (Love 1971). Tetraploid individuals of P. pilosa ssp. pilosa have a genome size approximately double that of diploids (averaging 21.90 pg [n=58] vs. 10.66 pg [n=19]; populations with unusually high genome size values averaged 26.55 pg [n=6]). Ranges in genome size within a ploidy level (e.g., 9.22-13.36 pg for individual diploids) were somewhat greater than expected. For example, Fehlberg and Ferguson (2012) found picogram ranges within a ploidy level of less than three picograms for R amabilis and P. woodhousei. However, their sampling on a per cytotype basis within a taxon was more limited (with a maximum of eight individuals for a single cytotype P. woodhousei; Fehlberg & Ferguson 2012). In the present study, there were no clear geographical patterns to genome size variation within cytotypes: in some cases, geographically proximate populations of the same ploidy level differed greatly (e.g., DNA content for two Wagoner County, Oklahoma, tetraploid populations was 21.60 vs. 24.15 pg), and measurements from individuals within populations differed (with the extreme being a tetraploid population with a range in values slightly over four picograms). Variation in genome size observed within a ploidy level for P pilosa ssp. pilosa may reflect, at least in part, an error level for flow cytometry mea- surements in this study; it may reflect true variation in genome size within cytotypes, and this could be due in part to differential presence of B chromosomes (discussed above; see Smith & Levin 1967); or, cytotypic varia- tion may actually be more complex than inferred here (i.e, some additional cytotypes such as triploids and hexaploids may be present). Spatial patterns of cytotypic variation in P. pilosa ssp. pilosa show that tetraploids generally occur west of diploid populations, potentially under more xeric conditions. While diploid and tetraploid populations were detected in close proximity (within eight miles), we did not detect mixed cytotype populations (though, with- in-population sampling was limited). Cytotypes are somewhat separated by geographic features, particularly river systems: for example, the Neosho River in southeastern Kansas separates sampled diploids to the east and tetraploids to the west. Into central Oklahoma (where most of our sampling was conducted), tetraploid popula- tions tend to occur at slightly lower altitudes relative to diploids and west of major river systems, but detailed study of potential ecological correlates of cytotypic variation in the area will require additional sampling. Documentation of intraspecfic cytotypic variation in P. pilosa ssp. pilosa at the western edge of its range is intriguing, and sets the stage for further study exploring ecological and cryptic morphological correlates of this variation. This work advances study of systematics of Phlox by adding to our understanding of cytotypic variation. Current work in our lab seeks to synthesize chromosome count and flow cytometry data for Phlox in a revised taxonomic context. In a broader context, cytotypic variation may in general be underappreciated m plants: for certain, systematists should not assume inference of ploidy level for an entire taxon based ononeor few chromosome counts alone. Broader surveys (made easier through techniques such as flow cytoi detailed studies of cytotypic patterns promise to improve our gen sity in plants. tnding of this aspect of biodiver- ACKNOWLEDGMENTS We gratefully acknowledge support from the following programs for LWs undergraduate research: NSF Un- dergraduate Research Mentoring Grant URM-1041199, the KSU McNair Scholars Program, NSF-ESPCoR (EPS-0903806), and the State of Kansas through Kansas Technology Enterprise Corporation; and from the KSU Division of Biology. We thank Richard Noyes and Alan Prather for valuable reviewer comments, and Shannon Fehlberg and the Ferguson lab group for valuable comments on an earlier version of the manuscript. We thank Theresa Melhem Susan Rolfsmeier, Wanda Worcester, and Bethany Wright for assistance with held work; Mike Herman for use of his camera-equipped microscope; and Molly Bernstein, Bernard Friebe, Theresa 450 Journal of the Botanical Research Institute of Texas 6(2) Melhem, Susan Rolfsmeier, Spencer Tomb, Bethany Wright, and the KSU Flow Cytometry Facility (School of Veterinary Medicine) for technical assistance. Online specimen database resources are gratefully acknowl- edged: R.L. McGregor Herbarium (KANU), Oklahoma Vascular Plants Database (particularly Oklahoma State University Herbarium [OKLA] and Robert Bebb Herbarium [OKL]), and Flora of Texas database (LLTEXk This is publication 12-452-J of the Kansas Agricultural Experiment Station. l.Bot. 95:99-1 10. ersity of California, Santa REFERENCES Adams, K.L. and J.F. Wendel. 2005. Polyploidy and genome evolution in plants. Curr. Opin. PI. Biol. 8:135-141. Balao, F., R. Casimiro-Soriguer, M. Talavera, J. Herrera, and S. Talavera. 2009. Distribution and diversity of cytotypes in Dian- thus broteri as evidenced by genome size variations. Ann. Bot. 104:965-973. Bino, RJ., S. Lanteri, H.A. Verhoeven, and H.L Kraak. 1993. Flow cytometric determination of nuclear replication stages in seed tissues. Ann. Bot. 722:181-187. Gres, L, C. Cuesta, E.L Peredo, M.A. Revilla, and JA.F. Prieto. 2009. Genome size variation and morphological differentiation Evol. 281:1 93-208. Davison, J., A.Tyagi, and L. Comai. 2007. Large-scale polymorphism of heterochromatic repeats in the DNA of Arabidopsis thaliana. B. M. C. PI. Biol. 7:44. DoleZel, J. and J. Bartos. 2005. Plant DNA flow cytometry and estimation in n Dolezel, J., J. Greilhuber, and J. Suda. 2007. F Wiley-VCH, Weinheim, Germany. Eater, J.W. 1 967. A systematic study of subsection Nanae of the genus Phlox. M A T1 Barbara. Fehlberg, S.D. and CJ. Ferguson. 2012. Intraspecific cytotypic variation and complicated genetic structure in the Phlox amabilis-P. woodhousei (Polemoniaceae) complex. Amer. J. Bot. 99:865-874. Fehlberg, S.D. and CJ. Ferguson. In press. Intraspecific cytotype variation and conservation: an example from Phlox (Polemoniaceae). S.E. Meyers, tech ed. Southwestern rare and endangered plants: Proceedings of the fifth confer- ence; March 16-20, 2009; Salt Lake City, Utah. Proceedings RMRS-P-XX. U.S. Department of Agriculture, Forest Ser- vice, Rocky Mountain Research Station, Fort Collins. Ferguson, CJ. 1998. Molecular systematics of eastern Phlox L. (Polemoniaceae). Ph.D. dissertation, University of Texas, Austin. Ferguson, CJ. and R.K. Jansen. 2002. A chloroplast DNA phytogeny of eastern Phlox (Polemoniaceae): implications of con- gruence and incongruence with the ITS phytogeny. Amer. J. Bot. 89:1 324-1 335. Ferguson, C J., F. Kramer, and R.K. Jansen. 1 999. Relationships of eastern North American Phlox (Polemoniaceae) based on ITS sequence data. Syst. Bot. 24:616-631. Flory, W.S., Jr. 1931. Chromosome numbers in Phlox. The Amer. Naturalist 65:473-476. Fiory, W.S., Jr. 1 934. A cytological study on the genus Phlox. Cytologia 6:1 -1 8. Flory, W.S., Jr. 1937. Chromosome numbers in the Polemoniaceae. Cytologia Fujii Jubilaei:1 71-180. Flory, W.S., Jr. 1 948. The chromosomes of a tetraploid Phlox from the Chisos Mountains. Proc. West Virginia Acad. Sd. d ploidy in higher plants. Curt scular plants of northeastern United States , 26:85. Galbraith, D.W., G.M. Lambert, Protoc. Cytom. 7.6.1-7.6.22. Gleason, HA and A. Cronquist. 1991. Manual of v« ond edition. New York Botanical Garden, New Great Plains Flora Association. 1986. Flora of the Great Plains. University Press of Kansas, Lawrence. Jones, S.B., Jr. and A.E. Luchsinger. 1986. Plant systematics, second edition. McGraw-Hill Book Company, New York. Kelly, J.P. and H.A. Wahl. 1928. Genetics of the genus Phlox. Bull. Pennsylvania State College Agric. Exp. Sta. 230:18. Levin, D.A. 1 964. Variation and evolution in Phlox subsection Divaricatae. Ph.D. dissertation, University of Illinois, Urbana. Levin, D.A. 1966. The Phlox pilosa complex: crossing and chromosome relationships. Brittonia 18:142-162. Levin, D.A. 1 968. The genome constitutions of eastern North American Phlox amphiploids. Evolution 22:61 2-632. Levin, D.A. and B.A. Schaal. 1970. Reticulate evolution in Phlox as seen tl 57:977-987. Levin, DA and D.M. Smith. 1965. An enigmatic Phlox from Illinois. Brittonia 7:254-266. Worcester et at., Cytotypic variation in Phlox pilosa ssp. pilosa 451 Levy, M. 1973. Novel flavonoids and reticulate evolution in Phlox. Ph.D. dissertation, Yale University, New Haven. J. Bot 61:1 56-1 67. Locklear, J.H. 201 1 . Phlox: a natural history and gardener's guide. Timber Press, Portland. Love, A. 1971 . IOPB Chromosome number reports XXXI. Taxon 20:157-160. Meyer, J.R. 1 944. Chromosome studies of Phlox. Genetics 29:1 99-21 6. Ono, S.P. and J. Whitton. 2000. Polyploid incidence and evolution. Ann. Rev. Genet. 34:401 -437. Smith, D.M. and D.A. Levin. 1 967. Karyotypes of eastern North American Phlox. Amer. J. Bot. 54:324-334. Soltis, D.E., VA Albert, J. Leebens-Mack, C.D. Bell, A.H. Paterson, C. Zheng, D. Sankoff, C.W. dePamphilis, P.K. Wall, and P.S. Soltis. 2009. Polyploidy and angiosperm diversification. Amer. J. Bot. 96:336-348. Souis, D.E., RJ.A. Buggs, J J. Doyle, and P.S. Soltis. 201 0. What we still don't know about polyploidy. Taxon 59:1 387-1 403. Soltis, D.E., P.S. Soltis, D.W. Schemske, J.f. Hancock, J.N. Thompson, B.C. Husband, and W.S. Judd. 2007. Autopolyploidy in angio- sperms: have we grossly underestimated the number of species? Taxon 56:1 3-30. Strakosh, S.C. 2004. Systematic studies in Phlox (Polemoniaceae) with a focus on P dolichantha, P. superba, P. stansburyi, and P. grayi. M.S. thesis, Kansas State University, Manhattan. Turner, B.L 1998. Atlas of the Texas species of Phlox (Polemoniaceae). Phytologia 85:309-326. Wherry, E.T. 1955. The genus Phlox. Morris Arbor. Monogr. 3:1-174. Whittemore, A.T. and R.T. Olsen. 2011. Ulmus americana (Ulmaceae) is a polyploid complex. Amer. J. Bot. 98:754-760. Wilken, D.H. AND J.M. Pokter. 2005. Vascular plants of Arizona: Polemoniaceae. Canotia 1:1-37. Journal of the Botanical Research Institute of Texas 6(2) BOOK REVIEW David Bramwell and Juu CaujapE-Castells (eds.). 2011. The Biology of Island Floras. (ISBN 978-0-531-11808- j 8, hbk.). Cambridge University Press, 100 Brook Hill Drive, West Nyack, New York 10994-2133, U.S.A. (Orders: www.cambridge.org; phone 1-800-872-7423, fax 845-353-4141), $120.00, 536 pp„ weight 2.65 lbs., 72 b/w illus., 36 tables, 9 3 A" x 6%". It is estimated that the biodiversity of 53 archipelagos includes between 50,000 and 52,000 plant endemic species woHd- | wide that are highly threatened, and 20,000 of these species are in critical danger of extinction as part of fragile ecosystems. I This suggests the urgency and importance of continued study of island biodiversity and conservation highlighted in this I book. The 21 separate chapters and 48 authors discuss the evolution, diversity, and conservation of island vascular plants. I Some examples of chapters that emphasize specific island groups are: Origin and evolution of Hawaiian endemics new patterns revealed by molecular phylogenetic studies; Origin and evolution of Galapagos endemic vascular plants; The I plants of the Caribbean islands: a review of the biogeography, diversity and conservation of a storm-battered biodiversity J hotspot; The biogeography of Madagascar palms; Evolution and biogeography of the flora of the Socotra archipelago (Ye- j men); Biogeography and conservation of the flora of New Caledonia; Phytogeography and relationships of the Pitcairn Is- lands flora; Chromosomes and evolution in New Zealand endemic angiosperms and gymnosperms; Jesters, red queens, boomerangs and surfers: a molecular outlook on the diversity of the Canarian endemic flora; and Dispersal, diversity and j The prestigious list of 48 contributors are represented by Juli Caujape-Castells, David Bramwell, Daniel J. Crawford, j GregoryJ. Anderson, Gabriel Bemardello, Paula Posadas, Jorge V. Crisci, Liliana Katinas, Sterling C. Keeley, Vicky A. Funk, Alan Tye, Javier Francisco-Ortega, Michael Maunder, Melissa Abdo, Rosalina Berazain, Colin Clubbe, Francisco Jimenez, j Angela Leiva, Eugenio Santiago-Valentin, John Dransfield, Mijoro Racotoarinivo, Lisa M. Banfield, Kay Van Damme, An- thony G. Miller, Steve Waldren, Naomi Kingston, Brian G. Murray, Peter J. de Lange, M. Dolores Ued6, Per O. Karis, Manu- el B. Crespo, Michael F. Fay, Mark W. Chase, Alain Vanderpoorten, Ben Laenen, Rosalina Gabriel, Juana M. Gonzalez- Mancebo, FredJ. Rumsey, Mark A. Carine, Michael Kiehn, Ole Hamann, Stuart Cable, Jennifer L. Trusty, Herbert C. Kesler, Jorge Rodriguez, Sara Oldfield, and Vernon H. Heywood. The Socotra archipelago in the Indian Ocean off the coast of Yemen currently counts 835 as endemic vascular plant species. Some of the tree species are unique plant growth forms found nowhere else in the world. The most outstanding example is the so-called bottle tree ( Adenium obesum subsp. sokotranum (Apocynaceae)), a succulent that stores water in the swollen trunk. Also, Dorstenia gigas (Moraceae) has a similar appearance and is far larger than any other species of its genus, suggesting these may be examples of island gigantism. The island has been described as the most alien-looking place on the planet earth. Additionally, the dracoid growth forms (e.g., Dracaena cinnabari (Ruscaceae)) — which have a single trunk with ramified branches coming off at one point on the trunk giving an umbrella-shaped crown — occur in 15 unrelatedspe- cies, suggesting that this growth form has survival value in this landscape. Exploration, collections, and research on island biology have been supported largely by botanic gardens as part of Botanic Gardens Conservation International, BDCI. These activities are highlighted in a chapter titled “Botanic gardens and the conservation of island floras.’’ This network of people and organizations operates to support the multidisciplinary con- servation objectives to protect and understand insular endemic floras. The features of this book update recent research ad- vances including molecular biology, genetic diversity, population dynamics, invasiveness, and colonization events that in- clude human intervention among others as they apply to island biology and conservation. An abstract for each chapter that briefly summarizes the salient points represented in the topical headings would have been a valuable addition. Each chapter is a stand-alone thematic topic that has supporting references. This book should be on the bookshelf of every biologist who teaches plant evolution, plant systematics, ecology, and conservation as well as de- cision-makers and organizations who seek to preserve biodiversity.— Harold W. Keller, Research Associate, Botanical Re- search Institute of Texas, 1700 University Drive, Fort Worth, Texas 76102-4025, USA. I.BotRes. Inst Texas 6(2): 452. 2 A NEW COMBINATION AND A NEW SPECIES IN COMBRETUM (COMBRETACEAE) FROM INDIA K.A. Sujana, M.K. Ratheesh Narayanan 1 , and N. Anil Kumar 1 Central Botanical Laboratory, Botanical Survey of India AJCB Indian Botanic Garden, Botanical Garden P. 0., Howrah, West Bengal - 71 1 103, INDIA Author for correspondence: sujanacabc@yahoo.com lalabaricum (Bedd.) Sujana RESUMEN INTRODUCTION The genus Quisqualis L. (Combretaceae) comprises ca. 16 species of lianas from the Old World tropics (Exell & Stace 1966). Recent research, however, has shown that the supposed differences with Combretum Loefl. cannot be maintained on the basis of molecular evidence (Jongkind 1991; Stace 2007; Maurin et al. 2010). Quisqualis dates from 1762, Combretum from 1758, and the latter name therefore has priority. Gangopadhyay and Chakrabarty (1997) recognized two species of Quisqualis from India, viz., Q. indica L. and Q. malabarica Bedd. The latter is here transferred to Combretum and a new combination, viz., Combretum malabaricum (Bedd.) Su- jana, Ratheesh, & Anil is proposed. Both species have been recorded for Kerala (Nayar et al. 2006; Sasidharan 2011). During botanical exploration as part of a project on “Investigation of rare, endemic and threatened woody climbing plants of the Western Ghats,” one of us (Sujana) collected specimens with recurved petals from the Aralam Wildlife Sanctuary in 2008 which did not match with any known species of Combretum or Quisqualis and is therefore described as a new species here. The species, although morphologically similar to the Combre- tum malabaricum (Bedd.) Sujana, Ratheesh, & Anil, comb, nov., differs from it in some characters; these data are given in Table 1. Combretum malabaricum (Bedd.) Sujana, Ratheesh, & Anil, comb. nov. (Fig. 1). Quisqualis malabarica Bedd., icon. PL Ind. Or. 1:33, 1. 155. 1874. Type: INDIA. Kerala: Carcoor ghat Wayanad, dev. ca. 450 m, Beddome s.n. (lectotype: MH - Acc. No. 20405; isolectotype: BM). Combretum recurvatum Sujana, Ratheesh, & Anil, sp. nov. (Figs. 2, 3). Type: India. Kerala. Kannur district: Aralam Wildlife Sanctuary, ca. 800 m, 28 Feb 2008, Su jana 0509 (holotype: CAL; isotypes: CAU). A liana, 15-20 m tall. Stems 15-25 cm in diameter; bark light brown with vertical striations, spines 1.5-2.5 cm lon g, at the base of the branchlets; young branchlets puberulous, purplish, terete. Leaves opposite, ovate to oblon g, symmetric, 4-15 x 4-7 cm, membranous, glabrous above, lead-coloured when dry, brown beneath, domatia as tufts of brown hairs in the axils of the secondary nerves, base rounded, margins entire to undulate **• Res. Inst Texas 6(2): 453 - 458. 2012 Sujana et al., A new s 455 Flt - 2 ‘ ^return recurvatum Sujana, Ratheesh, & Anil. A. Habit, B. Inflorescence, C Flowers - dose up, D. Fruit Journal of the Botanical Research Institute of Texas 6(2) 457 458 with tawny hairs, apex shortly acuminate or acute, lateral veins 5-8 pairs, slender, adaxially faint, abaxially prominent, arcuate, tertiary nerves adaxially faint, sometimes abaxially prominent, scalariform, marginal ul- timate veins recurved to form loops; petioles 1-1.2 cm long, sulcate, densely pubescent. Inflorescence termi- nal, spicate, 18-26-flowered; peduncles slender, purplish, puberulous, 7-11 cm long; bracteoles linear-lanceo- late or triangular, ca. 5 x 1 mm, tomentose, deciduous. Flowers 5-merous, 1.5-1.7 x 1.5-1.6 cm; pedicels 0-1 mm long, puberulous. Calyx tube pale green, narrowly infundibuliform, ridged, 0.9-1 cm long, puberulous; teeth triangular, ca. 3 x 2 mm, ciliolate, apex recurved. Petals white, turning pink with age, ovate or oblong, 13-14 x 6.8-7 mm, recurved with 3 distinct nerves, glabrous, apex obtuse. Stamens 5 + 5, inserted in the throat of the calyx tube; filaments ca. 1 mm long, white; anthers yellow, basifixed, ca. 1 x 0.7 mm. Ovary 4-5 x 1-1.2 mm, tawny puberulous, ridged, 1-celled with 3 ovules; upper free portion of style 1-2 mm long, yellowish white, slender, flattened; stigma capitate, creamy white. Fruits chocolate brown, chartaceous, 2-2.3 x 1.8-2.2 cm, villous when young, glabrous when mature, horizontally faintly striate, widely ellipsoid; stalk ca. 1 mm long, wings 5, thin, 10-12 mm broad. Seed solitary, fusiform, ca. 13 x 0.5 cm. Funicle ca. 6 mm long, curved; testa brown, membranous. Distribution. — Combretum recurvatum is so far known only from Aralam Wildlife Sanctuary, Kannur district of Kerala, India (Fig. 4). Flowering. — -January-March; Fruiting: February-May. Habitat. — Rarely seen in the semi-evergreen forests of the Aralam Wildlife Sanctuary, Kerala (Fig. 3), at an altitude of 800 m above sea level and is found associated with Anamirta cocculus (L.) Wight & Am., Baccrn- rea courtallensis (Wight) Mull.-Arg., Combretum latifolium Blume, C. razianum K.G. Bhat, Sterculia foetida L., and Vateria indica L. Additional specimens examined: INDIA. KERALA. Kannur district: Aralam Wildlife Sanctuary, alt. 800 m above sea level, 26 Mar 2009, Sujana 0714; 6 May 2009, Sujana 0786; 2 April 2010, Sujana 0833 and 6 Apr 2011, Sujana 0913 (Herbari- um of M.S. Swaminathan Research Foundation, Wayanad). The research grant provided by Sir Dorabji Tata Trust, Mumbai, is gratefully acknowledged. Sincere thanks are expressed to J. F. Veldkamp (L) for valuable remarks and comments. We very much thank Clive Stace (LTR) for confirming the novelty; Carel Jongkind (WAG) and Brent Berger (WIS) for their advice on the Combretum/ Quisqualis relationships; and P. Lakhshminarasimhan (BSI) for reviewing the paper. Thanks to K.M. Manudev, former Research Fellow of MSSRF for the drawings. REFERENCES Exell, A.W. AND CA Stace. 1 966. Revision of the Combretaceae. Bol. Soc. Brot. II 40:5-25. Gangopadhyay, M. and T. Chakrabarty. 1997. The family Combretaceae of Indian subcontinent. J. Econ. Taxon. Bot. 21: 1 Jongkind, C.C.H. 1991. Novitates Gabonenses, 6. Some critical observations on Combretum versus Quisqualis (Combre- taceae) and description of two new species of Combretum. Bull. Mus. Natl. Hist. Nat. Sect. B, Adansonia 12:275-280. hase, M. Jordaan, and M. van der Bank. 2010. Phylogenetic relationships of Combretaceae inferred from jstid DNA sequence data: implications for generic classification. Bot. J. Linn. Soc. 1 62:453-476. egam, N. Mohanan, and G. Rajkumar. 2006. Flowering plants of Kerala— a handbook. Tropical Botanic Sasidharan, N. 2011. Flowering plants of Kerala - Ver 2.0 (CD-ROM). Kerala Forest Research Institute, Peechi, Thrissur, Kerala, India. Stace, C.A. 2007. Combretaceae. In K. Kubitzki, ed. The families and genera of vascular plants. Springer-Verlag, Berlin, Heidelberg. 9:67-82. CONVOLVULACEAE OF SONORA, MEXICO. 1. CONVOLVULUS , CRESSA, DICHONDRA, EVOLVULUS, 1POMOEA, JACQUEMONTIA, MERREMIA, AND OPERCULINA Richard S. Felger Herbarium, University of Arizona P.O. Box 210036, Tucson, Arizona 85721 and Sky Island Alliance, Tucson rfelger@ag.arizona.edu Thomas R. Van Devender Daniel F. Austin Arizona-Sonora Desert Museum fi 021 N. Kinney Road, Tucson, Arizona 85743, U.S. A. and Herbarium, University of Arizona, Tucson J. Jesus Sanchez-Escalante Universidad de Sonora Dept, de Investigaciones CientificasyTecnoldgicas Rosales y Ninos Heroes, Centro Hermosillo, Son, 83000, MEXICO jsanchez@guayacan.uson.mx Mihai Costea Dept, of Biology Wilfrid Laurier University 75 University Avenue W Waterloo, ON, N2L 3C5, CANADA mcostea@wlu.ca RESUMEN INTRODUCTION The Convolvulaceae constitute a large and diverse assemblage of 58 genera and about 1800 species worldwide (Staples 2011). This monophyletic family has greatest diversity in tropical and subtropical regions worldwide and does not generally occur in higher latitudes. The sweet potato ( Ipomoea batatas) and water spinach (I. aquatica) are the only major crop plant in the family. In addition to species with horticultural value (e.g., vari- ous morning glories), there are significant crop weeds (e.g., certain species of Convolvulus, Cuscuta, and Ipomoea) and many medicinal uses among Sonoran people and worldwide. This publication is the first taxo- nomic account of all the known Convolvulaceae (convolvs) native or naturalized in the state of Sonora, Mexico (Fig. 1). We include approximately 84 species and 2 infraspecific taxa in 9 genera for the state. Due to the large size of the article, we divided it into two parts: the first part includes all the Convolvulaceae genera minus Cus- cuta, while the second part provides a floristic/taxonomic treatment for the latter genus (Costea et al. 2012a). This introduction includes the 9 genera. Although a pohtical border is not necessarily an ecological or biological boundary, the borders of Sonora are to varying degrees biologically and logistically significant. The western boundary (except the extreme northwestern comer) is the Gulf of California. The east boundary mostly coincides with the continental di- vide. The north boundary marks a division between the better-known continental flora of Arizona and the relatively less-known flora of northern Sonora. The southern border with the state of Sinaloa border separates a continuous flora but is far enough south to include the northern climatological limits of the New World trop- ics (in east-central Sonora at 28-29°N). Sonora encompasses 185,934 km 2 (Molina-Freaner & Van Devender 2010) and is the second largest state in Mexico after Chihuahua. The region is topographically and biologically diverse (Molina-Freaner & Van Devender 2010). Three major river systems, the Rio Colorado, Rio Yaqui, and Rio Mayo, and several minor riv- ers systems course through the state and empty into the Gulf of California. The eastern margin of the state is comprised of the Sierra Madre Occidental and numerous north-south trending Sky Island ranges forming the Madrean Archipelago. The highest elevation is 2625 m in the Sierra de los Ajos in northeastern Sonora. The Sierra San Luis in the extreme northeastern comer is nearly as high, and another high peak, further south on the Chihuahua border, near Mesa Tres Rios may be equally high. Numerous other ranges, generally decreasing in peak elevations westward, spread across the rest of the state, interspersed with broad valleys and expansive plains. Sonora includes the northern limits of tropical and subtropical biota as well as some of the most arid des- ert regions of North America. Tropical species follows the lowland tributaries in the Rio Yaqui drainage system northward in northeastern Sonora. Total annual precipitation decreases from south to north and east to west, and increases with elevation. Rainfall is largely bi-seasonal with summer and winter-spring r ainy seasons. Precipitation increases from west to east: for example from the Rio Colorado River to northeastern Sonora/New Mexico, and with elevation and from north to south. Total annual precipitation varies from less than 40 mm in the extreme northwest of the tr the delta of the Rio Colorado to about 1000 mm in southeast and east-central Sonora (e.g., Y£cora at and probably considerably more farther north, such as the Tres Rios-Sierra Huachinera region (Brito- et al. 2010; Felger et al. 2001; Martinez-Yrizar et al. 2010). man populations in Sonora remained sparse and major roads were relatively few until the mid-twenti- 1500 ir Felger etal., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico eth century (e.g., Stoleson et al. 2005). Much diversity and richness of natural habitat remain, but like else- where, assault* on the environment are escalating (e.g., Castellanos-Villegas et al. 2010). Since the latter part of the twentieth century large areas of the state have been established for conservation purposes (e.g., Burquez and Martinez -Yrizar 2007, Felger and Broyles 2007). Major works including ethnobotanical information on convolvs in Sonora include Felger and Moser (1985), Gentry (1942, 1963), Yetman and Felger (2002), and Yetman and Van Devender (2002). For summary information see species accounts for Ipomoea ancisa, I. arborescens, I. bracteata , and I. pedicellaris. While soil moisture is the principal limiting factor in this generally arid region, freezing temperatures limit the northern distributions of the more tropical or subtropical species (e.g., Brito-Castillo et al. 2010; Shieve 1951; Turner et al. 1995; also see the species accounts for Ipomoea arborescens ). Freezing weather is in- frequent across most of Sonora, especially in the southern part of the state and at low to moderate elevations, and many habitats are essentially frost-free, but freezing temperatures become more frequent and severe north- ward and at higher elevations (Brito et al. 2010). The total flora of the state includes about 3700 taxa (Van Devender et al. 2010 and new records). Within this rich flora, the Convolvulaceae is the seventh largest family, following the Asteraceae, Poaceae, Fabaceae, Euphorbiaceae (Steinman & Felger 1997), Malvaceae (s.l.), Cactaceae (Paredes et al. 2000), and Cyperaceae (Van Devender et al. 2010). With 42 taxa, Ipomoea is the fourth largest vascular plant genus in the state and the largest genus in the family worldwide. Among the documented convolv species in Sonora, we include species that occur in Arizona and New Mexico close to the northern boundary and are expected to occur in Sonora. Growth forms. — Most convolvs throughout the world are scandent annuals or herbaceous and woody perennials. Three native Sonoran Ipomoea are trees or large shrubs: I. arborescens, I. chilopsidis, and I. seaania. The non-native I. cornea subsp. fistulosa is a shrub while I. bracteata and some Jacquemontia may become some- what shrubby. Ipomoea bracteata and I. longiflora form large tuberous roots, as do I. capillacea and I. plummerae, while Merremia palmeri, Operculina pinnatifida , and O. pteripes probably have large tuberous roots because their closest relatives have them. The majority of Sonoran convolvs are vining or have twining stems. The liana growth form, characteristic of tropical regions, occurs among Ipomoea bracteata, I. pedicellaris, several Jacquemontia, Merremia quinquefo- lia, and Operculina pteripes. All occur in tropical deciduous forest in Sonora and some also range into thom- scrub and oak woodland. Cuscuta spp. are always annual when parasitic on annual hosts; however, when para- sitizing on perennial woody hosts, some species such as C. americana, C. corymbosa var. grandiflora, and C. tinctoria are often perennial, regenerating every year from haustorial tissues left inside the stems of the host (Costea & Tardif 2006). Endemism and Rare/Endangered Taxa.— Although many species may be locally rare and known from few collections in Sonora, most of these are common elsewhere. Only a small number of Sonoran convolvs are rare and/or endangered throughout their range. Ipomoea seaania, the only convolv endemic to the state of Sonora, is known only from the vicinity ot the type locality and seems to be globally restricted to a relatively small area. Among Cuscuta, C. dentatasquamata is known in Sonora only from the type collection; C. solina is known from one locality in Sonora but is common elsewhere, and the only known Sonoran C. tinctoria specimen is a parasite on a cultivated tree and is not native. Tkere are two records for Dichondra brachypoda in Sonora, both in mountains in the northeastern part of the state. Evolvulus prostratus is known from two collections in Sonora but is widespread elsewhere in Mexico. Ipomoea alba, with two records, is common elsewhere and may or may not be native in Sonora. Ipomoea ancisa is a relatively narrow endemic in mountains in eastern Sonora and western Chihuahua but is locally common. Jacquemontia abutiloides is widespread in Baja California (none) and Gulf of California islands and approaches Sonora on Isla Tiburon, although a thorough taxonomic investigation may render it a synonym of a mainland s P ec ' es - Merremia cissoides, with a single Sonora collection, is a cosmopolitan species. Non-natives. — About eight convolvs found in Sonora are not native to the state. Convolvulus Owens ts and 1 xl-wunthu are widespread weeds. Ipomoea cornea subsp. fistulosa, I. triloba, and Merremia disjecta are prob- 463 ably of Caribbean origin and I. batatas, the sweet pototo, is of tropical American origin. The origin of I. hep- taphylla is not known although it is probably native to the Old World. Ipomoea batatas and I. camea subsp.fis- tulosa are cultivated and are sometimes encountered outside of cultivation. Cuscuta campestris and C. indecora are weedy and probably have been introduced, and C. tinctoria probably is also not native to the flora. Diversity. — In comparison with neighboring areas to the north and west (Arizona, the two Baja Califor- nia states, California, and New Mexico), the convolvs are quite diverse in Sonora. This is due to the more tropi- cal affinities of much of the Sonoran flora as well as habitat diversity. As one moves further south in Mexico and Central America, the convolv diversity continues to increase. A comparison of the convolv diversity in Sonora with adjacent and comparable areas in Mexico and the southwestern United States is shown in Table 1. Vegetation of Sonora and the Convolvs The major habitats or vegetation types in Sonora include mangroves and coastal vegetation, tropical deciduous forest, thomscrub, desertscrub (both Sonoran and Chihuahuan), grassland, oak woodland, pine-oak forest, and mixed conifer forest (Fig. 2, Table 2). The Sonoran convolvs are distributed as follows: Chihuahuan desert (11), Sonoran desert (33), coastal thomscrub (30), foothills thomscrub (31), tropical deciduous forest (41), grassland (13), oak woodland (34), pine-oak forest, 27), mixed conifer forest (1), salt scrub (1), mangroves (1). These vegetation regions are briefly mentioned below, with some examples of characteristic convolvs. For dis- cussions of the vegetation of Sonora the reader is referred to Brown (1982), Felger et al. (2001), Gentry (1942), Martin et al. (1998), Martinez-Yrizar et al. (2010), Rzedowski (1978), and Shreve (1951). Coastal vegetation. — Mangroves occur sporadically along the coastal fringe of the southern two-thirds of the state, bordering salt scrub. Cressa truxillensis occurs along beaches and extends into tidal marshes (lo- cally called esteros ) of salt scrub and sometimes at the inland border of mangroves, as well as inland in some agricultural areas. Ipomoea imperati and I. pes-caprae are found along beaches. northeastern Sonora and adjacent southeastern Arizona at elevations below about 1430 m. This landlocked desert covers much of north-central Mexico between the Sierra Madre Occidental and Sierra Madre Oriental extends into adjacent inland areas of the southwestern United States. Hard freezes may occur in the Chi- diich accounts for the absence of columnar cacti and reduced convolv diversity. Rainfall mostly occurs during the summer. The Sonoran portion of the Chihuahuan desert is bordered by grassland and oak woodland. As with the Sonoran desert, there is open ground, and shrubs predominate. The substrate often consists of limestone and alkaline soils. A total of 12 convolv species are recorded for the relatively small area of the Chihuahuan desert in Sonora (one occurs adjacent in Arizona) and all except Ipomoea cardiophylla also occur in adjacent vegetation. Sonoran desert.— The Sonoran desert, defined and described elegantly by Forrest Shreve (1951), covers roughly the northwestern two-thirds of Sonora. Shreve divided the Sonoran desert into seven geographic veg- etation zones, five of which occur in Sonora, although Shreve’s Foothills of Sonora is reclassified as thomscrub (Felger & Lowe 1976), leaving the Plains of Sonora, portions of the Arizona Upland, the Sonora portions of the Crural Gulf Coast, and the Lower Colorado Valley. Within the span of environments in the Sonora portion of the Sonoran desert there is great variation in vegetation cover and structure and regional plant diversity. Thir- ty-two convolv species are documented for the Sonora portion of the Sonoran desert. Ray6n 33, Rosario , San Luis Ri'o Colorado 1, San Miguel de Horcasitas 42, San Pedro de >, Tubutama 13, Ures 43, Villa Hidalgo 30, Villa a !. 1 ss»ss ] !h!i! 3 2 i i.ciflt ,-Jll toslilil .»*Ss!if I» Felger et al., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico ^Unlike thornscrub there is much open ground. Most of the perennial Sonoran desert plants have evolved ° m ,ro P ical or subtropical relatives that are found today in thornscrub and tropical deciduous forest. Many of 1 Perennials are sensitive to winter freezing, their northern limits fixed by an invisible line of freezing tem- peratures and drought. The desert is essentially frost-free at its southern limits, such as near Guaymas, and the ***** and Oration of freezing increases northward. Similarly, drought is more severe northward and west- (Felger 2000; Turner et al. 1995). show a strong seasonality linked with monsoonal rains and for the most part are highly frost-sensitive. Unlike TDF, natural thornscrub generally does not form 100% perennial coverage. The stature of the vegetation is generally lower and the dominant species tend to have smaller leaves than those of the TDF. Thornscrub in Sonora was described as Sinaloan Thornscrub by Brown (1982) and Thom Forest by Gentry (1942). Two thornscrub formations can be discerned in Sonora: the coastal thornscrub (CTS) of southwestern Sonora (Felger and Lowe 1976; Friedman 1996; Martin et al. 1998) and the interior, foothills thornscrub (FTS) along the east side of the Sonoran desert and at higher elevations within the Sonoran desert. We document 30 convolv species in CTS in Sonora and 31 in FTS. Coastal thornscrub is on the coastal plain from Empalme-Guaymas southward. Foothills thornscrub is on inland, often rocky slopes. In southern Sonora, FTS is below tropical deciduous forest in elevation. In cen- tral Sonora, it is the transitional vegetation between the Plains of Sonora subdivision of the Sonoran desert on the west and oak woodland in Sky Island mountain ranges and the Sierra Madre Occidental to the east. In the north FTS is replaced by desert grassland as winters become colder and periodic fires become ecological pro- cesses. The northern limits of FTS in Sonora are at about 30°11'N in the Rio Sonora Valley and 30°26'N on the Rio Bavispe at the southern end of the Sierra El Tigre. FTS does not reach Arizona, but the distributions of a number of FTS species extend into southern Arizona in desert grassland or oak woodland. Coastal thornscrub extends southward on the coastal plain from the southern margin of the Sonoran desert in the vicinity of Guaymas into coastal northwestern Sinaloa. Southward and eastward this vegetation type merges with TDF. Much of the coastal thornscrub has been converted to large-scale modem agriculture. Foothills thornscrub. — This vegetation type is essentially synonymous with Shreve’s (1951) Foothills of Sonora subdivision of the Sonoran desert (Felger and Lowe 1976; Felger et al. 2001). FTS is shrubby or semi' arborescent vegetation with a nearly closed canopy of small trees and large shrubs. FTS extends northward along the eastern side of the Sonoran desert, becoming narrower in geographic and elevational range towards its northern limits. Northward it gives out at about the vicinity of Arizpe, where rise in elevations and winter freezing as well as drier conditions become pronounced. Tropical deciduous forest. — The northern arm of TDF, sweeping northward from the tropics, ends in the mountains of eastern Sonora and southwestern Chihuahua. Sonoran TDF is sandwiched between thorn- scrub to the west at lower elevations and oak woodland eastward at higher elevations. Northward, along the east side of the Sonoran desert, TDF merges into foothills thornscrub (Felger et al. 2001; Martlnez-Yrizaretd. 2010). There is a long dry season interrupted by a short but intense rainy season generally from mid-June to October. Summers are long and hot and winters short and mild. Freezing weather is rare and most Sonoran TDF species are highly frost-sensitive. Awesome and seemingly sudden transformation to luxuriant tropical green occurs with the onset of the summer monsoon. TDF, with 44 documented species, supports a greater convolv species richness than any other vegetation in the state. The tree morning glory, lpomoea arborescens is a common and conspicuous component, as are others such as lpomoea bracteata with its spectacular pink in- florescences. lpomoea muricata, I. quamoclit , Merremia cissoides (with one record), M. dissecta, and M. quinqut folia are more or less restricted to TDF. Grasslands.— The southwestern extension of the Great Plains grassland biome in the mid-continent is in the southwestern United States and northwestern Mexico. The best developed regional grasslands with mo* rainfall and colder winter temperatures termed Plains Grassland are restricted to the Animas and San Rafael Valleys along the Arizona-Sonora border. Most other areas are termed desert grassland (McClaran & Van Devender 1995), with warmer, drier climates. Dominant or common species in the regional desert grassland have fluctuated four times during the last 4,000 years— between bunch grasses during wetter periods and shrubs such as mesquites (P rosopis glandulosa, P. velutina ) and others during drier periods (Van Devender 1995). Today desert grassland controlled by human disturbance related to cattle grazing is widespread in val- ley lowlands below oak woodland from northeastern Sonora west to the Sdsabe area southeast of the Baboqui- van Mountains and south to the Cananea-Fronteras area within 100 km of the Arizona border. In southeastern Arizona and northeastern Sonora, valley bottom desert grassland forms a mosaic with Chihuahuan Desert on rocky limestone slopes. To the west, Sonoran desert borders desert grassland at its lower elevations. To the south, desert grassland is replaced by foothills thomscruh below oak woodland, al- though there are local areas of grassland still present. In addition, there are open grassy areas within oak woodland and pine-oak forest in the higher Sky Island mountain ranges and the Sierra Madre Occidental. As freezes decrease southward, foothills thornscrub replaces desert grassland. Only 13 convolvs are documented from grasslands in Sonora. Characteristic species include Ipomoea longifolia, a large herbaceous perennial often conspicuous sprawling across expanses of short grasses, and Evolvulus seriteus, a small, silvery-leaved herba- ceous perennial. Oak woodland. — Oak woodland is widely distributed at elevations above desert, grassland, thornscrub, and tropical deciduous forest, but below pine-oak woodland. In Sonora, 34 convolv species occur in OW. Across Sonora, and northern Mexico, the species composition and density of oak woodland changes both with elevation and from south to north. Although these oak zones have been called Madrean Evergreen Woodland (Brown 1982), many Sonoran oaks and associated species are deciduous during the late spring drought, and their biggest flush of new foliage occurs with the renewal of summer rains. In these regions "autumn color” occurs in late spring as the air and ground desiccate and temperatures soar. Extensive areas in northeast and north-central Sonora are dominated by open woodland of Quercus emo- ryi. This species, with Q. oblongifolia and Q. arizonica, are among the most common low-elevation oaks in the northern part of the state. At lower elevations the oaks border desert grassland, foothills thornscrub, and des- ertscrub. There is sometimes a broad ecotone between oak woodland and grassland where the oak trees be- come widely spaced and grasses predominate. Oak woodland sometimes occurs on acidic, hydrothermally altered soils within tropical deciduous forest where the ecotone between the two plant communities is often only a few meters wide. Oak woodland in southeastern Sonora, called Oak Forest by Gentry (1942), shows considerable tropical affinity. The lower limits border tropical deciduous forest, and the boundaries are often remarkably well de- fined, apparently maintained by fire. Fire, however, is not an ecological process in desertscrub, thornscrub, or tropical deciduous forest. Across mountains in the Rio Mayo and Rio Fuerte drainages, low fires creeping al- most harmlessly through dry grasses and forbs among the leafless oaks used to be a common sight in May and June. These fires destroy small TDF trees and shrubs but not the perennial grasses, forbs, and oaks. Many of the oaks in east-central and southeastern Sonora and nearby southwestern Chihuahua are tropical montane oaks. Ipomoea chilopsidts is the most spectacular and unique convolv among the diverse 34 convolv species in oak woodland. Pine-oak forest.— There are numerous montane islands of pine-oak forest in the mountains of eastern Sonora. However, pine-oak forest is more extensive east of Sonora in Chihuahua along the east side of the con- tinental divide. In comparison, on the western slope of the Sierra Madre Occidental the climate is generally somewhat wetter, with presumably milder winter temperatures, resulting in a more diverse flora with more tropical-derived pines such as Pirns engelmannii, P. h errerae, P. oocarpa, and a number of tropical-montane oaks such as Q. tarahumam. The pine-oak forest has been included within the concept of Madrean Evergreen Woodland and Madrean Montane Conifer Forest (Brown 1982; Martin et al. 1998). Towards southeastern Sonora the pine-oak woodland is floristically and structurally more like Mexican pine-oak woodland than the temperate pine-oak woodland to the north. Pine-oak forest, where the pines form the overstory while the oaks generally form an understory, is continuous with oak woodland at lower elevations. Among the 27 convolv species in POF, Cuscuta dentatasquama (the type collection is the only Sonora record), Ipomoea ampullacea (one ttC0rd > see the species accounts), and Ipomoea decasperma (one record) appear to be restricted to this habitat. Mixed conifer forest.— This zone is restricted to limited areas on the several highest mountain tops in northeastern Sonora and very limited areas near Yecora, in the upper Rio Mayo drainage. Three conifers, Abies, Pinus ’ especially p strobiformis, and Pseudotsuga , define this vegetation. Evolvulus wtundifolius is the only con- volv recorded in mixed conifer forest in Sonora. 470 Convolvulaceae Jussieu - Morning glory Family is, subshrubs, shrubs, or trees, some species with milky sap. Rootsto< simple, entire to pinnately lobed or pecti Twining herbs, ltanj tuberous, otherwise fibrous. Leaves a] cies palmately compound; stipules absent. Inflorescences solitary in leaf axils or in racemose or paniculate cymes, some dichasial basally and monochasial above. Flowers small and inconspicuous to large and showy, but usually wilting quickly after opening (mostly within 4-5 hours) except Cuscuta and often excepting plants flowering during cooler weather, bisexual (or unisexual in som< irregular. Sepals 5, distinct, imbricate, equal or unequal, persistei alous, tubular, funnelform, campanulate, urceolate, or salverform, 5-lobed, 5-toothed, or ± entire, with plicae (areas folded in bud) and interplicae (unfolded in bud), usually induplicate and often also convolute in bud. Nectary disc annular or cup-shaped, sometimes 5-lobed, occasionally absent. Stamens 5, distinct; filaments inserted on the corolla tube base alternate with corolla lobes; anthers dithecal, usually linear or oblong, ex- trorse or introrse. Ovary superior, 2-4(-6)-carpellate, usually with as many cells, placentation basal or basal- axile, ovules 2 (4-6) per cell, or ovary 1-celled and ovules 4, these erect, anatropous; style 1, filiform, simple or bifid, or sometimes with 2 distinct styles; stigmas capitate or bilobed, or, when stigmas 2, then linear, ellipsoid, or globose. Fruits capsular, dehiscent by valves, transversely or irregularly, or indehiscent and baccate or nut- like. Seeds l-4(-6; to 10 in Ipomoea decaspema), often fewer than ovules, glabrous or pubescent, endosperm mated 1880 (Staples 2011), cosmopolit; a 9, species about 84 in Sonora. SPECIES ACCOUNTS infraspecific taxa. The months or seasons noted refer to the recorded times of reproduction. Flowering and fruiting usually overlap broadly, and therefore we generally do not distinguish separate flowering and fruiting times. Many species that are reproductive at various seasons do so facultatively, mostly depending on soil moisture and temperature. There is, however, a marked tendency for members of the family to have flowering initiated by short day length and thus they usually flower in the autumn. Representative specimens are cited in the last paragraph of each species or infraspecific taxon account We include the Sonora municipios (mpio.), as of 2011, to help place the collection localities (Fig. 1). We also in- clude records from Gulf of California islands nearest the Sonora coast (Islas Tibur6n, Alcatraz in Bahia Kino, Datil, San Esteban, and San Pedro Nolasco; see Felger and Wilder 2012, Felger et al. 2011). We have seen all specimens (except those specified as “not seen”) and use “!” for all specimens or images that we have seen, ex- cept in Cuscuta where “!” is used for the types seen. All Cuscuta specimens cited have been seen by Costea. All other convolvs have been seen by Austin and/or Felger. Unless otherwise indicated, specimens cited are deposited at the University of Arizona Herbarium (AR1Z); specimens in other herbaria are indicated by the abbreviations given in Thiers (2011). Most of our col- lections are duplicated in the herbaria of USON, MEXU, and SD and other regional and international colfcc- len is at AR1Z, we generally do not cite duplicates at other herbaria. When a specimen lacks* collection n r, it is identified by the date if nple: Ezeurra 9 Nov 1982. We gei abridge label information, but provide enough that one can find the specimen at a herbarium or search addi- tional information in a data base, especially SEINet (Southwest Environmental Information Network 2012) j and MABA (Madrean Archipelago Biodiversity Assessment 2012). Usually only the first collector’s name's listed. Elevations and reproductive times (flowers and fruiting) are mostly from herbarium label data and I specific only for Sonora. Coordinates for specimens cited are often available in SEINet (Southwest Environ- mental Information Network 2012), however these might not have come from the collectors’ labels. Many co- ordinates were added/determined by students entering information by looking on Google Earth, or other maj* | and some may be inaccurate (especially for specimens from Mexico). Coordinates for many specimens cited ; are not repeated here to save space. North America is defined here as Mexico northward, excluding Cen0» Felger et al., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico America. Hundreds of photographs that illustrate diagnostic details of morphology for the majority of species, are provided on a companion website — Convolvulaceae (morning glories) of Sonora, Mexico, which is hosted at ARIZ and WLU (Costea et al. 2012b). Plants not native to flora area are marked with an asterisk (*). TYPIFICATIONS Special attention has been given to study of type specimens for the taxa included, not only to assure the correct names but also for proper identification. These data are included because there have been errors in citing types in the literature. Types have been checked by consultation of protologues and corrected where necessary. Type sheets have been studied at various herbaria or as loans, and also from online databases at various herbaria, some through JSTOR Plants (2011). Moreover, because of the information now available, it is possible to note many more duplicates of type collections than previously known. The information provided here gives a more complete picture of the available resources for each taxon than formerly obtainable. Problems and errors in typification are discussed in several cases. Some taxa have never had lectotypes selected. In several cases, we designate lectotypes to establish proper use of names and concepts. In other cas- es, various problems with former selections of types have been located and these are discussed and resolved except in the case of E, alsinoides var. angustifolia. Lectotypes are selected for nine species and new information and clarifications are provided for others: Convolvulus palmatus (in Merremia dissecta). Convolvulus pennatus (in Ipomoea quamoclit), Cuscuta corymbosa var. grandiflora, lpomoea ancisa, Ipomoea alata (in Opcrculina pteripes), I. alatipes (in Operculina pteripes), I. decasperma, l. pedicellaris, Merremia dissecta, Operculina pinnati- fida, 0. pteripes, and O. roseana (in Operculii sBenth., Pi. Hartweg. 16. 1839. Type: MEXICO. Nuevo Leon: 1837. Hart weg 98 (m des A. Gray. Syn. Fl N. Amer. 2(1):216. 1878. Type: U.S.A. T Convolvulus inconus sensu Kearney & Peebles, not Vahl (see also Staples et al. 2006). Perennial herbs with branched, prostrate or decumbent steins arising from a taproot; densely hairy. Leaves variable, ovate-elliptic to triangular-lanceolate or narrowly oblong with projecting basal lobes, blades most often deeply indented basally, 1-7 cm long, 0.2-4 cm wide, densely hairy on both surfaces with loosely ap- pressed indumentum, margins toothed or lobed or both, rarely entire; petioles 0.25-5 ci usually 1-flowered, less often 2 or 3 and cymose, on peduncles 0.5-10.5 c late, 1.5-3 mm long or sometimes scale-like, hairy like the leaves. Flowers often solitary, on short pedicels 5-24 mm long. Sepals oblong to ovate, 6-12 mm long, 3-6 mm wide, obtuse to weakly retuse apically, ap- pressed sericeous, the margins membranaceous, subcordate with age. Corollas (1.5—) 2.5-3 cm long, cam- panulate, white or pink to pale lavender, at times with a reddish center, sericeous on the petal lobes. Capsules ± globose, 7-8 mm wide, glabrous. Seeds 1-4, 4-4.5 mm long, black, granulate, glabrous, n = 12. Sonora . — Northeastern part of the state in Chihuahuan Desert, grassland, and oak-mesquite scrub; often in disturbed sites; 1200-1600 m. Flowering April-December. General distribution. — California, Arizona, New Mexico, Texas; Chihuahua, Coahuila, Guanajuato, Hi- dalgo, Edo. Mexico, Michoacan, Nuevo Leon, Oaxaca, Puebla, QuerCtaro, San L Mna2009-1747\ Mpio Cans CRESSA L., Sp. PI, 223. 1753. [Greek, based on feris or kriti , “from Crete,” a Cretan woman; apparently not ety- mologically related to cress or cressa (Cruciferae) of Germanic derivation]. Alkali weed Perennial herbs or subshrubs, gray appressed pilose to sericeous, usually much-branched, the stems erect to decumbent. Leaves sessile or short-petiolate, entire, small or scale-like. Inflorescences axillary, 1-flowered. Flowers appearing sessile or on short peduncles (at least some, e.g„ C. truxillensis), bracteate, in spicate to head-like clusters at tips of branchlets, bracteoles unequal in length. Sepals ovate to obovate, imbricate. Corol- las salverform, the limb 5-lobed, the lobes mostly ovate, imbricate, spreading to reflexed. Stamens and styles exserted; filaments filiform; pollen 3-colpate. Ovary 2-locular, 4-ovulate; styles 2, distinct to the base; stigmas capitate. Fruit capsular, ovoid, unilocular, 2-4-valved, usually 1-seeded. Seeds glabrous, smooth and shining to reticulate, dark brown. Species 4, two in the Americas, one in Asia, and one in Australia. Selected reference .— Austin (2000b). Prodr. 9:440. 1845 ed.) 3:119. 1818 [1819]; also folio ed. 3:93. 1819. Type: PERU: ioto F!, isotype: F!). Cressa cretica var. truxillensis (Kunth) Choisy in DC., 7-July3,l 474 s gray, appressed pilose to silvery-sericeous, not twining, usually much-branched, at first erect, becoming decumbent to spreading, often 8-25 cm long in open, sunny habitats, or often with weak slender stems to 75 cm long when growing through and branching over the tops of other salt-marsh halo- phytes; stems dying back during adverse times to thickened rhizomes and/or rootstocks often 8-15+ cm below the surface; lower stems often semiwoody. Leaves on main branches often larger than those on branchlets, with pubescence like the stems, subsessile or with petioles 0.5-2 mm long; blades mostly 3-10 mm long, 1-4 older leaves sometimes thick and succulent in hypersaline habitats. Inflorescences of solitary flowers, axil- lary, usually concentrated in the upper leaf axils and appearing almost spicate. Flowers on stalks (peduncles and pedicels) 2-6 mm long, the pedicel much reduced and more slender than the peduncle, the bracteoles ovate to ovate-lanceolate, acute, unequal, 2-3 mm long, mostly 1 mm wide. Sepals ± equal or the inner slightly longer; outer sepals obovate, elliptic, 3-4 mm long, 2.5-3 mm wide, obtuse or acute, pubescent; inner sepals obovate, 3-4.5 mm long, 2-3 mm wide, acute, with scarious margins, appressed sericeous only at the apex. Corollas white, becoming scarious when dry and somewhat persistent, salverform, 5-6.5 mm long, the tube 3-3.5 mm long, the limb 5-lobed. Corolla lobes obtuse to ± acute, about as long as the tube, somewhat pubes- cent on the outer surface, becoming reflexed with age. Stamens exserted, 4-6 mm long, usually equal, the fila- ments basally pubescent with glandular indumentum; anthers red and becoming purple with age, 1-1.5 mm long, oblong with the base cordate to ± bilobed. Ovary ovoid, unilocular to ± bilocular, apically somewhat hir- sute; styles pure white, unequal, 3-5 mm long; stigmas pure white, capitate, smooth. Capsules 5-6 mm long, ovoid, shiny brown, surrounded at least basally by the calyx, apically hirsute, unilocular. Seeds usually 1, ovoid, 3-4 mm long, brown, glabrous. 2n = 28. Sonora . — Often common in tidally wet saline mud and sandy soils among saltscrub and margins of man- groves, esteros, bays, and low-lying coastal soils from the Rio Colorado delta to the Sinaloa border. In north- western Sonora sometimes in small inland playas but near the coast, and as an agricultural weed south of San Luis, especially in fine-textured silty-clay alkaline soils. Sonoran desert and coastal thornscrub: 0-20 ffl. Flowering March-December. In contrast to populations in northwestern Sonora, this species has apparently become relatively rare, at least in recent years, in nearby southwestern Arizona (Austin 1992). In the early 1990s Cressa was a common weed in sandy soil of seawater-irrigated experimental plots at the Environmental Research Laboratory adja- cent to Estero Morua at Puerto Penasco. Cressa had been evaluated earlier as a potential halophytic seed crop in these plots. Substantial seed crops were obtained, but no economic value was found and further evaluation was terminated. This is probably the first report of a weed in seawater-irrigated agriculture (Felger 2000). General distribution.— California, Arizona, Nevada, New Mexico, Texas, Utah; Baja California (norte) and Sur, Chiapas, Chihuahua, Coahuila, Colima, Jalisco, Nuevo Le6n, Oaxaca, San Luis Potosi, Sinaloa, Revilla- gigedo Islands; Ecuador, Peru, Chile, Argentina. San Luis R.C.: Colorado River, opposite mouth of Hardy River, 29 6 Oct 1985, Felger 92-989. SONORAN ISLANDS: Alcatraz, Bahia Miguel, E side of island, 23 Nov 2006, Wilder 06-366! (See Felger a >r 1990, some older leaves succulent, Felger 91-40. Mpio ns 2842 (CAS!); Ci£nega de Santa Clara, 5 km S of Rilhto, 17, Felger 07-167!; Tiburdn, base and N side of Punta San DICHONDRAJ.R. i(m ' Ml >' ■ Ho., kv hillside llcrmo,ilk> 220m.2^W'N. I Id >4 W. 23 Nov |UW & Kuh.uJ; 3754 (MO!); Sierra Seri, 550 m, 2 Feb 1969, Felger 181301 Mpio Hutabampo: Camahuiroa, 23 Nov 1993, Friedman 330-93 ! Mpio La Cob- b ocher 1891 Mpio Onavas: Onavas, 1 1 Oct 1986, Rea 1208! Mpio San Javier: Cerro Verde to San Javier, Martin & Ferguson 10 Mar 1 990! Mp*> foothills thomscrub, 1 Sep 2009, Van Devender 2009-649 (USON!). ep 1996, Van Devender 96-357! Mpio Ures: 6 mi N of Ures, 20 Sep 1! a, 23 Sep 1977, Goldberg 77-170!; Agua Amarilla, 1000 m, Martin 14 * Suffrutescent herbs, densely appressed pilose to almost woolly tomentose throughout. Stems few to many, arising from a woody base, erect to ascending or decumbent, 10-30(-45) cm long. Leaves lanceolate to linear- j lanceolate, 10-25(-35) mm long, 2.5-6(-14) mm wide, gradually decreasing in size toward apex, the upper I leaves linear, acute or obtuse apically, attenuate basally; petioles absent or short, sparsely to densely pilose on | both surfaces, with strongly and loosely appressed, soft, short, grayish trichomes. Inflorescences cymose, 1-3 flowered on slender peduncles usually as long as or longer than the leaves. Flowers on pedicels 3-4(-8) n®> long, reflexed in fruit; bracteoles linear-subulate, 1.5-3 mm long. Sepals equal, lanceolate, acuminate, 3-33 | mm long. Corollas rotate to broadly campanulate, blue or blue with white stripes, (10-)12-22 mm wide. Fib- | ments inserted near the corolla base, 1.5-2 times as long as the linear anthers. Ovary globular, glabrous. Cap- sules globular, 3.5-4 mm long, reflexed, glabrous. Seeds 2-4, 1-1.25 mm long, tan to brown, glabrous. Sonora.— Widespread in eastern and central Sonora; Sonoran desert (Plains of Sonora), grassland includ- ing mesquite grassland, foothills thomscrub, tropical deciduous forest, oak woodland, and pine-oak M* natural and disturbed sites, often in rocky habitats; 150-ca. 1200 m. Flowering mostly August-December. | General distribution. — Arizona, New Mexico; Chihuahua; disjunct in Argentina. This species is easier to recognize in living material than on some herbarium specimens. Some have#' ognized two varieties that are not distinct. Selected reference.— Austin (1990b). Mpio Agua Prieta: Rancho Nuevo, Caj6n Bonito, 2May 1976, Mason 3213b! Mpio Alamos: Mocuzari, La Cruz, 17°13'N, 109 °° 5 . 5 ’W.^^ ( Guayabo (upper) crossing of the Rio Cuchujaqui 3 km NE of Sabinito Sur and 15 km (airltaT) ESE of Alamos, near 27°00'N, 108°4^ I Felgeretal., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico 479 Evolvulus filipes Mart., Flora 24 (2, Beibl.):100. 1841. tofthes Annual herbs, the stems erect to ascending, delicate, generally sparsely pubescent. Leaves linear or narrowly lanceolate, l-2.5(-3) cm long, usually 2-5 mm wide, sessile or ± sessile or sometimes short-petioled on larger leaves, cuneate basally, acute apically, sparsely pilose on the upper surface, slightly more pubescent on the lower surface. Inflorescences cymose to solitary, on peduncles usually 1-2.5 cm long or slightly longer. Flow- ers l(-3) on pedicels to 2.5 mm long, short-pilose; bracteoles lanceolate to subulate, 1-2 mm long. Sepals lan- ceolate 2-2.5 mm long, glabrous or pubescent and ciliate. Corollas pale blue or white, 3-5 mm long, rotate, with 5 sericeous bands outside, filaments ca. 2 mm long. Ovary globose, glabrous. Capsules globose to ovoid, 3-4 mm long, reflexed, glabrous, brown. Seeds 1-4, smooth, dark brown to b Sonora.— Several specimens identified to this species are from the eastei ciduous forest and oak woodland. General distribution.— C oahuila, Durango, Michoacan, Sinaloa, Veracruz; Mesoamerica; South America. This species is weakly differentiated from E. alsinoides. In South America they appear to be distinctive, while in Mexico they are not always separable. The Sonora specimens have the appearance of first season plants of E. alsinoides. Perhaps those called E. filipes in part (or all) of the range are nothing more than depau- perate or first season E. alsinoides, but there are insufficient data to confirm or deny that suspicion and we re- luctantly retain them as separate until more detailed studies are made. n E of Alamos, 2r0'20"N, 108°4710"W, 250 m, Martin 16 Mar 1989 (FTG-FAU!). Mpio Cucurpe: ), Toolin 453-C (2 sheets, AR1Z 220685!, ARIZ 221490)! Schult., Syst. Veg. 6:198. 1820, New name E.-USA Missouri: Rapid River, “On the banks of the Missour., v F .< tly destroyed during the Second World War, although van Ooststro fwlwdus oeophlu Greene, Leafl. Hot. Observ. Crit. 1(U):151. 1905. TnK USA New Mexico: Sima [Co.l: S e^oftheBladt Rang®. * miW of Hillsboro, dry hills, 5500 ft, 16 Aug 1904, Metcalfe 1228 (u-cronre designated here: US!; isotypes: MO. NMC. in Suffrutescent herbs, the stems several, erect to ascending, 10-15 cm long, densely spreading-pilose with an indumentum of ferrugineous, brown, fulvous or gray color. Leaves linear-oblong, narrow-1; row-oblanceolate or rarely oblong, 8-20 mm long, 1.5-5 mm wide, attenuate basally, acute tc densely pilose on both surfaces; petioles short or absent. Inflorescences solitary, m axils fength of stem; peduncles short or absent, flowers on pedicels 3-4 mm long, becoming reflexed in fruit; brac- teoles subulate, 1-4 mm long. Sepals lanceolate to narrowly lanceolate, long-acuminate, 4-5 mm long, sprea - ing villose. Corollas rotate to broadly campanulate, 8-12 mm wide, subentire, purple o Journal of the Botanical Research Institute of Texas 6(2} long, oblong, basally auriculate; filaments twice as long as the anthers. Ovary subglobose, glabrous. Capsules ovoid, about as long as sepals, reflexed, glabrous. Seeds (1) 2, brown, smooth. This species grows in the mountains of southeastern Arizona near the Mexican border (Austin 1991, 1998b) and probably will be found in northeastern Sonora in oak woodland and pine-oak forest; 820-2450 m. Flowering April-September. General distribution. — Arizona, New Mexico, and Texas to Montana, North Dakota, Illinois, Arkansas, disjunct in Tennessee; Chihuahua, Coahuila. i B.L. Rob., Proc. Amer. Arts 29:320. 1894. i Y!; isolectotypes: GOET!, K!, PH!). MEXICO, t :n). MEXICO: Valley of Mexico at Santa Ffi, 5 Jul 1865-66, Bourgeau 323 (synty Perennial herbs; stems few or several from a woody perpendicular root, prostrate, 10-20 cm long, sericeo-vil- lose, with fulvous, grayish or whitish trichomes, glabrescent. Leaves distichous, more or less imbricate, at right angles to the stems or somewhat reflexed, broadly-ovate to orbicular or sometimes broader than long, 9-16 mm long, 7-16 mm wide, apically rounded to somewhat emarginate, basally truncate to cordate or rounded, covered with appressed silky-villous, light-brown to grayish trichomes below, green and glabrous long. Flowers 1(2) on pedicels 2-3 mm long, appressed-villose; bracteoles oblong to linear-oblong, to 3.5 mm long. Sepals ovate-oblong to ovate, 3-4.5 mm long, acute, appressed-villose. Corollas white, outer surfaces of corollas in bud and midpetaline conspicuously tawny sericeous-hirsute, rotate to broadly funnelform, tube short, the limb 10-12 mm wide. Anthers 1-2 mm long, filaments 3-4 times as long as the anthers. Capsules globose, 4-valved, 3-5 mm long, glabrous. Seeds 1-4, brown, glabrous. Sonora. — Pine-oak forest in east-central Sonora; 1550 m. Flowering at least in September. General distribution. — Chihuahua, Distrito Federal, Durango, Hidalgo, Jalisco, Michoacan, Nayarit, Nuevo Leon, Oaxaca, Pubela, San Luis Potosl, Veracruz. Mpio Yecora: Ydcora, 28°23%'N, 108°54Vi'W, 1550 m, open pine forest, occasional, 7 Sep 1995, Fishbein 2481 !; Yecora, near baseball fieldand cemetery, 28°22'25"N, 108°56’W, 1540 m, common on clay flats, 5 Sep 1996, Reina-G. 96-377 (FTG-FAU!). Evolvulus rotundifolius (S. Watson) Hallier f., Bot. Jahrb. Syst. 16:530. 1893. Type: MEXICO. Chihuahua: damp places, Perennial herbs; stems several from a woody root, prostrate, 10-15 cm long, sericeo-villose, glabrescent to- ward the base. Leaves distichous but not imbricate, at right angles to the stem, ovate to elliptic, sometimes broad-ovate, 13-22 mm long, 8-12 mm wide, apically obtuse to somewhat emarginate, basally rounded to cordate, appressed sericeo-villose below, glabrous above, the petioles 1-2 mm long. Inflorescences axillary, sessile, and 1-flowered. Flowers sessile; bracteoles linear-oblong, to 6 mm long. Sepals ovate-oblong to oblong, 4-5 mm long, acute, appressed-villose. Corollas blue with a white throat and midpetaline stripes, rotate to broadly funnelform, the tube short, the limb 10-13 mm wide. Filaments 2-4 mm long; anthers 1-2 mm long Capsules globose, 4-5 mm long, 4-valved, glabrous. Seeds 1-4, brown, glabrous. Sonora.— East-central and southeastern Sonora near the Chihuahua border; arroyos and open areas in oak woodland, pine-oak forest, and mixed conifer forest (with Abies durangensis, etc.); 1600-2100 m. Flower- ing August-September; General distribution. — Chihuahua, and Durango; reported but not verified for Aguascalientes and Queretaro. !I Aug 1991 ! Mpio Yecora: Mesa El Campaner -FTG!); El Kipor (Quipur), Arroyo El KIpor, p sahuaribo), 27°19'N, 108°40'W, 1600 m, 21 Aug 1991, i el Salto, 2040 m, arroyo margin in pine-oak forest will !; N end of Mesa Campanero at head of Bart e-oak forest, 28°24'25"N, 108°35'40"W, 1 i in tropical deciduous forest at about 500 central Sonora in oak woodland, ca. 940 m, and southe; m. Flowering September-November. General distribution . — Sinaloa southward and perhaps in all Mexican states; southeastern United States; Mesoamerica; Colombia, Venezuela, Guyanas, Ecuador, Peru, Brasil, Argentina; Caribbean. Often cultivated; native to tropical America. Flowering September-May. Apparently first recorded in the New World by Oviedo (1526). This plant was probably first carried around the world for its medicinal seeds. Later it was spread for the nocturnal, fragrant flowers. Determining the region of nativity of this species within the New World is particularly difficult since it was carried from at least Cuba around the world in the early 1500s by the Spanish and probably the Portuguese. However, the center of diversity is tropical North America (McDonald 1993a). Typically, the plants are associated with wet- lands, the seeds being distributed by water. Seeds arrive on the coasts of the British Isles with regularity, and surprisingly a few of them are viable and germinate. The occurrence in southern Sonora may be part of the native range, or may result from cultivated plants that have escaped. Perennial herbs from a tuberous root, the stems woody below, retrorsely hispid. Leaves 8-10 cm long, almost as wide, cordate to broadly ovate, entire or 3-lobed, remotely appressed sericeous. Inflorescences dichasial, axillary. Flowers 1-4 on peduncles 10-12 mm long. Sepals 2.5-4 cm long, ovate, apically acuminate to obtuse, and becoming spatulate-attenuate with age, the outer ones appressed sericeous, the inner ones glabrous on the margins, sericeous on the dorsal regions. Corollas funnelform, white, 3-4 cm long, pubescent on the outer surface. Capsules 8-10 mm long and wide, almost globose, brown, glabrous. Seeds 1-4. Sonora . — Known in the state from a single record in pine-oak forest, 1220 m. Flowering in September. General distribution.— Sinaloa, Guerrero, Edo. Mexico, especially tropical deciduous forests. The occur- rence of this species in pine-oak woodland in Sonora seems to be an anomaly as compared with its general distribution. Y«ora: Rio Maycoba at MEX 16 (20.5 km W of Maycoba, 28.6 km E of Yecora), 28°22'15"N, 108°45'30” W, common, to 2 m in shrubs. Ipomoea ancisa House. Ann. N.Y. Acad. Sci. 18:187. 1908. im MEXIC O Chihuahua: 22-24 Aug I8!) ivnu.lvnlio u, r ,IUcus Kunth. Nov Gen. Sp. PI. 3:97.1819. Ipomoea murieata Cav., Icon. 5:52. 1799, not Jacquin (1798). Type: MEXICO. Guanajuato: Oct, Cavan Hies i.n. (holotwe: MA!; isotype: MA!). Ipomoea muricatisepala Matuda, Ann. Inst. Biol. Mex. 34:124. 1964. New name for I. murieata Cav. Perennial herbs with erect to ascending glabrous stems, from an underground elongate tuber. Leaves with blades sessile, incised and appearing compound, the segments usually 5, filiform, 5-15 mm long. Inflores- cences axillary, solitary. Flowers on peduncles 5-7 mm long, the pedicels 3-6 mm long, reflexed in fruit. 5e- Pak ± equal, 5-6 mm long, the outer one 2 mm wide, the inner ones 3 mm wide, acute to obtuse and mucronate apically, the outer sepals elliptic to oblong, the inner sepals ovate, muricate-tuberculate at least on the midvem. Corolla, 3-4 cm long, funnelform, lavender .0 reddish purple, .he limb 2-2,5 cm wide, glabrous Xapsules * globose, glabrous, 4-5 mm wide, apiculum 1 mm long. Seeds 1-4, 3 mm long, ovoid, black to dark brown, shortly erect hispid. Sonora. — Oak woodland and pine-oak forest in eastern Sonora; 1100-2100 (to 2500 m in nearby SW Chihuahua). Flowering August-October. General distribution.— Arizona, New Mexico, Texas; Baja California (norte) and Sur, Chihuahua, Coa- buila, and southward to Puebla; Central America; South America. See comments under I. murieata and also I. plummerae for potentially confusing species. V. AUmos; Sierra . A r 7n All „ 1Q o 4 steinmonn 94-741; Vallecito, 3 km SW of Santa Barbara, 1100 m, Martin 4 Oct 1990 . i, 1372-1585 m, south 6 Aug 2003. K amp 445 (COCHISE!). Mpio *“ anJos * fountains, 3 Aug 1893, Mearm 1628 (US, not seen, cited by McDonald 1995). M P .o Nacoxar. de ' eras ' 20 Aug 1941, While 4059 (Gil specimen not relocated, cited by McDonald 1995). Mpio Yecora: Arroyo Otro "e. Van Devender 95-840!; Mesa de los Corona ' *'20 m , 28°23'49”N, 108°54’48"W, 7 S ^upanero, 28°19’30"N. 109°01'40'W, 2100 n ^ ^Barranca El Salto, pine-oak forest, 2 j 1995, flowers ] flowers pink, 1 Sep 2000, Van Devende 00 m, 6 Sep 1996, Van Devender 96-396! Felgeretal., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico 493 Ipomoca chilopsidis Standi, Publ. Field. Mus. purple throat, singularly of the high and arid crags, 1 er. 17:206- 1937. Type: MEXICO. Cmh n high, few and irregularly branched, flo' 2391 (hOlotyfe: F!; isotypes: AR1Z 76041!, 2 Multiple-stem small trees and large shrubs 2-5 m tall, the stems broadly ridged on drying, glabrous. Leaves mostly l0-17.5(-20) cm long, 0.5-1.3 cm wide, with 18-30 pairs of secondary veins, entire, linear, apically acute, basally cuneate, glabrous; petiole 0.5-1.2 cm long. Inflorescences monochasial, terminal on reduced branches. Flowers l(-3) on peduncles 0.4-2.2 cm long, glabrous except inner surfaces of sepals. Sepals 12-16 mm long, 7-9 mm wide, ovate, about equal or the outer ones slightly shorter than the inner ones, apically ob- tuse to acute, the inner surfaces short-pubescent. Corollas 8-9.5 cm long, 8-9 cm wide, funnelform, white with a purple throat. Capsules 18-22 mm long, conic, brown, glabrous. Seeds often 4, 10-15 mm long, oblong, brown, woolly with long trichomes on the ventral margins. Sonora .— High and arid rock ridges and on indurated ash in woodland, often with Acacia pennatula, Dodonaeaviscosa, and Quercus chihuahuensis ; 1000-1300 m. Flowering documented May-November. General distribution . — Oak woodland in southeastern Sonora and southwestern Chihuahua; 1000-1800 m. The leaf shape of this shrub makes it unique among the Ipomoea in the Americas. As the species name indicates, the leaves resemble those of Chilopsis linearis (Bignoniaceae). loom, 24 Aug 1992, SU tensis, 1100 m, large wl t, 27 D 58 , 20"N, 109°06'30"W, ridge, ca. 5000 ft, Qjun 1963, Friger 8078b!; Barranca de de Batopilas, 5 mi S of Quirare, rocky :n CF39 (UCR!); La Bufa, canyon Rio Batopilas, short tree forest, barancan oak forest, 3300- d new settlement, 10 Aug 1971, Bye 1977 (UCR!); Sierra Madre Occidental between Creel and ior near mouth of creek called Arroyo San Fernando, 3330-4180 ft. 27 Jun 1982, Siplivinsfey Bound. 149. 1859. type: l Herbs, annual, from a slender taproot; stems erect at first, in age trailing or twining at the tips, 2-3 m long, glabrous. Leaves with blades ± sessile or on petioles 1-3 cm long, deeply palmately divided with the lateral divisions two-cleft (pedatisect), the segments 5-9, linear or linear-lanceolate, 7-25 mm long, glabrous. Inflo- rescences mostly solitary, axillary. Flowers on peduncles l-3(-7) cm long, the pedicels 15-25 mm long, erect in fruit. Sepals slightly unequal, the outer ones 3-5 mm long, 1-2 mm wide, the inner ones 4-6 mm long, 2-3 ™n wide, oblong-lanceolate, acute, mucronulate, scarious margined, at least the inner slightly rugose along the veins. Corollas 10-12 mm long 5-10(14) mm wide, pale lavender to pink, usually with a white throat. Capsule* ± globose to ellipsoid-globose, 4-5 mm wide, with a 1-2 mm caducous apiculum, tan, glabrous. Seeds 3 or 4, ■ i long, ovoid, black, glabrous, land, coastal Sonora.— Southern and eastern margins of the Sonoran desert, Chihuahuan desert, gra and foothi U thomscrub, tropical deciduous forest, oak woodland, and pine-oak forest; 50-1550+ m. Flowering August-November. Cncrd distribution. — Arizona, New Mexico, Texas, Baja California Sur, Chihuahua, Coahuila, Nuevo U6n ’ Tar nauli pas south to Chiapas; introduced into South America. This widespread species is self-fertilizing (McDonald et al. 2011); indicators are the annual habit and the inconspicuous small flowers. 495 Felger et al., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico mi S of Esqueda, 22 Jul I960, Felger 4045! Mpio Guaymas: Canon Nacapule, 18 Nov 1985, Felger 85-1310! Mpio Hennosillo: Cafldn Las Chivas, Sierra Libre, al Sur de la ciudad de Hermosillo, por la Carr. Fed. 15, 28.583°N, 110.96528°W, 8 Oct 1997, Burquez 97-501 (USON!). Sep 2006, Boyle 7840! Mpio Yecora: Near Los Vallecitos, 14.7 km E ofYecora on MEX 16, pine-oak forest, 28°22'38"N, 108°50'03"W, 1470 m, 10 Sep 1996, Van Devender 96-521 ! i. Type: MEXICO. Hidalgo: rocky hills. Lens Static 1905. Pringle 10034 U Perennials from a large, thickened root, the steins 1-1.5 m long, twining, finely white-pubescent, becoming glabrous with age. Leaves 3-5 cm long, 3-4 cm wide, entire or trilobate, or almost 5-lobed, white-pilose, some- what sagittate to cordate, the apex acute, mucronulate; petioles 2-3.5 cm long, white-pilose. Inflorescences with axillary peduncles 3-20 cm long, white-pilose. Flowers solitary on slender pedicles 8-10 mm long, pi- lose; bracts 2-4 mm long, linear, pilose. Sepals 10-15 mm long, 7-8 mm wide at the base, herbaceous, more or less equal, pubescent, triangular-ovate, acute (not attenuate), white-pilose. Corollas blue to bluish, 5 cm long, 4-5 cm wide, the tube white. Capsules about as long as the calyx, glabrous, 5-valved, 5-locular, often 10-seed- ed, the 5 valves often bifid; pericarp chartaceo-ligneus, brownish. Seeds 1-potentially 10, 4 mm long, discoid, black, with short erect trichomes. Sonora.— East-central part of the state in pine-oak forest, 1300 m. Flowering August-September. General distribution.— Hidalgo, Michoacan, Edo. Mexico, Sinaloa, Zacatecas; reported from Chiapas (Nel- W49, NY and GH, neither seen). This perennial species is similar to and might be confused with annual I. purpurea. The sepals of I. deca- sperma lack the setose trichomes of I. purpurea ; leaves of I purpurea, while variable, are not sagittate as in I. decasperma. Mpio Yecora: Tributary of Arroyo de Pilar magenta ’ 2 2000, Reina-G. 2 000-622, 2000-627 (USON!). Ipomoea hederacea Jacq., Collect. Bot. 1:124, pi. 36, 1787 [title page 1786 1^' “^ IERICAS ” (ac ( “g to the Code) since Jacquin did not mention the Dillenius plate. Ipotnoea desertorum.House, Ann. New York Acad. Sci. 18:203. 1906. Type: U.S.A. Arizona: Thurber 29 (holotype: NY!, Ipomoea hederacea var. integriuscula A. Gray, Syn. FI. N. Amer. ed. 2, 2:433. 1886. Type: U.S.A: East Florida, Curtiss (h T *°MPiUO, TRQMPILLO AZUL, TMM1LLO MORADO ; IVY-LEAF MORNING-GLORY; HEHE QUVJAM ‘PLANT THAT-CURLS-AROUND-.T,’ mm] W K spun (like a top)’ (Seri) Annual herbs, the stems twining, often 2-3+ m long, densely to sparsely pubescent throughout. Leaves with ^des ovate to * orbicular, (2-) 3.5-13 cm long and about as wide, entire to 3- or 5-lobed, basally cordate, the a P ica % acute to acuminate, pubescent; petioles to 12 cm long, rarely longer. Inflorescences cymose. Flowers l-3(-6), on peduncles 5-10 cm long, the pedicels 3-7 mm long, erect in fruit; bracts foliaceous elhp- lc to lanceolate, 5-8 mm long. Sepals ± equal, 12-25 mm long, 4-5 mm wide, herbaceous, lanceolate, abrupt- 497 Hobs, annual, the stems twining or prostrate, 2-8 m long, glabrous. Leaves petiolate, 1-4 cm long and wide, 5-foliolate, the lobules lanceolate to linear-lanceolate, sessile, apically acute, glabrous. Inflorescences axillary and terminal, flowers usually solitary, with filiform peduncles as long as or longer than the petioles, often spirally twirled. Sepals ± equal, 5-7 mm long, ovate to oval, apically obtuse to rounded, glabrous. Corollas 1.8-3 cm long, funnelform, glabrous, rosy (or sometimes white) with the interior of the tube red-violet. Cap- sules 8-10 mm long, globose, brown, glabrous. Seeds 1-4, 4-6 mm long, ovoid, brown, finely pubescent. Sonora .— Coastal thomscrub of the coastal plain in southwest part of the state, apparently not common. Margins of temporarily flooded or swampy places including roadside; below ca. 50 m. flowering September- October. General distribution.— Texas, SE USA; Mesoamerica; Argentina, Brasil, Ecuador, Paraguay, Peru; pantropical. Although no one knows where this species is native, it shares no obvious relatives in the New World. It may well be an Old World species and perhaps is related to I. cairica, which is probably native to Africa. Ipomoea heptaphylla has been confused at least with I cairica and is morphologically more similar to that spe- cies and others like it in Africa than those elsewhere. Roxburgh (1824) pointed out that the plant appeared unexpectedly in the nursery at the botanical garden in Calcutta, but he did not think it native, at least in his region. Verdcourt (1961: 11) thought that the plant climbed with the spiraled peduncles. Although Austin has this plant few times, no note was made of the peduncles behaving like tendrils. Perhaps someone examin- ing living plants in the field will be able to clarify the role of the spiraled structures. ^om^ C b l,,: 17 011 MCX HWy ^ ^ °^ unct * on ° f road t0 (= edge of P°ttUi Oct 1985, Mger 85-12641 Mpio Navojoa: Navojoa, water edge of river back-water, limb and tube white, throat purple, 27 Oct 1939, a imperati (Vahl) Griseb., Cat. Pi. Cub. 203. 1866. Type: imperato, Hist. >) (lectotyfe!). Convolvulus imperati Vahl, Symb. Bot. 1:17-18. 1 J24. 1763. Type: based on Convolvulus Joliis obtusis, Plu Convolvulus stoloniferus Cyrillo, PL Rar. Neap. 1:14. 1 5, 1788. Type: based on ‘iw Plate (lectotype!). Herbs > Perennial, the stems prostrate, rooting at the nodes and often underground, reaching 5 m or more in ‘ en gth. Leaves 1.5-8 cm long sometimes linear to ovate or oblong, lobed or not lobed, the size and shape van- l-T — a- in th<« morning. Se- ** Ul5 ^ long, unequal, oblong, the outer two sepals shorter, acute to obtuse, glabroi Felgeretal., 499 loba, with larger flowers, is usually allogamous (outcrossing; McDonald et al. 2011). The honeybee, introduced from the Old World, carries pollen from one species to the other and is the only insect known to move between the two. However, certain other pollinators, when present, will visit I. lacunosa and the hybrid. Presumably the hybrid is moved, as it has been around the world, as a contaminant in seeds of cultivated plants. , 1200 ft, 28 Oct 1939, Gentry 4767!; Rancho San Pedro, E entry to El Caj6n -26!; Alamos, in streambed, 390 m, flowers pink, 2 Feb 1992, Van Devender imin Hill: 24 km S of Rancho El Seri, 8.2 km S of Rancho El Carrizo (16.9 le, 29°58'58"N, 111°15 , 24"W, 735 m, 22 Aug 2007, Van Devender 2007-890! 1, Friedman 438-94 ! Mpio Navojoa: Navojoa, w river back-water, 27 Oct 1939, Gentry 4747! Mpio Onavas: Onava 1986, Rea 1210! Mpio San Lois R.C.: 5 km S of San Luis, roadside morning, 6 Oct 1985, Felger 85-1032 ! Mpio Yecora: Santa Ana 2 28°22'40" N, 109°08' W, 950 m, Otts, Martin, et al. 10 Oct 1988. Ipornoea longifolia Benth., PI. Hartw. 16. 1839. Type: MEXICO. Zacatecas: Hartwcg 97 (iiolotype: K!; isotypes: NY!, P!>; other sheets with species #97 (species numbers?) are from Chihuahua (K!) and Leon (BR!, LD!). Large perennial herbs forming large tuberous roots, somewhat fleshy, rhizomatous, glabrous, the stems long, trailing, decumbent, or ascending. Leaves with blades linear to oblong-lanceolate, often 10-12(-20) cm long, 2-4 cm wide, entire, glabrous; petiole 0.5-1.7 cm long. Inflorescences mostly solitary. Flowers on peduncles 3.5-4(-ll) cm long, the pedicels 2-3.5 cm long, mostly erect in fruit. Sepals ovate, the outer ones 12— 14(— 17) mm long, 6-7 mm wide, the inner ones 15-20 mm long, 7-8 mm wide, coriaceous, glabrous. Corollas funnel- form, 7-10 cm long, the limb whitish to cream-white, 7-8 cm wide, the throat purple or deep reddish purple. Capsules ovoid, 14-16 mm wide, with an apiculum 2-3 mm long. Seeds 1-3(4), 10-11 mm long, ovoid, brown, with long pilose indumentum on the margins and near the apex. Sonora. — North-central and northeast part of the state in grassland, oak woodland, foothills thomscrub, and tropical deciduous forest; 975-1850 m. Flowering April-September. General distribution . — Southeastern Arizona and northern Mexico in Aguascalientes, Chihuahua, Du- The flowers are open in the evening and are pollinated by moths, although there have been reports of bees also visiting (Austin 1986). Perennial herbs, roots tuberous, the stems erect, scandent, or less often twining, 4-50 cm long, mostly branched from the base, glabrous. Leaves variable, the first ones usually elliptic, rhombic, or narrowly lanceo- kte, 1.5-5 cm long, 3-20 mm wide, occasionally lobed, the lobes to 6 mm long, 5 mm wide, margins entire or Regularly dentate, apex acute mucronulate the base attenuate, the distal leaves sometimes grading into pal- •^tisect laminas with 3-5 segments, about equal or unequal, filiform, linear, or lanceolate, 7-42 mm long, 1-6 mm Wide - the outer segments often shorter than the inner ones, margins entire, apex obtuse or acute, the base attenuate, glabrous; petioles almost absent or 2-10 mm long. Inflorescences monochasial axillary cymes. ,1-3 ml nn 502 Sonora .— Central and mostly southeastern part of the state in tropical deciduous forest, ca. 150-790 a. Flowering at least August-November. General distribution . — Texas and Arkansas to Florida; Mexico including Baja California Sur, Chihuahua, uila, Guerrero, Jalisco, Edo. Mexico, Sinaloa; Argentina, Ecuador, Peru, Venezuela. Native to Mexico and ttive in the southern United States and South America, and also naturalized in many parts of the Old d. This species has been widely spread through the southeastern United States as a contaminant in soy- seeds (Gunn 1969a, 1972). It continues to be dispersed and established in new areas. Confusion between names for I. capillacea and I. muricata began in the 1800s and a new layer of confusion was added by Gunn (1969a, 1969b, 1972). What these authors failed to realize, or at least point out, in addition was clarified by Staples et al. (2005). Ipomoea capillacea is a moming-flowering herb with corollas 3-4 cm long while I. muricata is an evening- and night-flowering vine with corollas 3-7.5 cm long. The leaves are also nota- bly different: I. capillacea has nearly sessile highly dissected leaf blades with linear segments and I. muricata has petiolate entire to 3- or 5-lobed blades. The current residual confusion involves Gunn’s conclusions so the earlier problems will not be addressed. The underlying problem is that Gunn (1969b) used the wrong date for the names involved, rejecting I. muricata (L.) Jacq. (1798) and using I. turbinata Lag. (1816). Staples et al. (2005) Ipomoea nil (L.) Roth, Catal.Bot. 1:36. I/ 1 -)/. Type: Dillenius, Hort. Eltham. 1:96 i 80 1 91 1732(lec totype!). PhariritisnilO- Herbs, annual, herbaceous, the stems twining or prostrate, to 3-4 m long, hispid, the trichomes yellow. Leave* ovate to ± orbicular, 5-15 cm long, 2-14 cm wide, entire to 3- or 5-lobed, the apices acuminate, hirsute. Inflo- rescences cymose, dichasial. Flowers 1-4. Sepals 1.5-2.5(-3) cm long, ± equal, linear-lanceolate, basally nar- rowly ovate, the base densely hispid with yellow trichomes, hispid to strigose on the upper parts, or distally glabrous. Corollas (2-)3-6 cm long, funnelform, purplish to blue, at times white or red, the tube white or yel- low within, glabrous. Capsules 8-12 mm long, rounded-depressed, 3-locular, brown, glabrous. Seeds l-4(-®- 4-9 mm long, dark brown to black, densely short-pubescent. 2 n = 30. Sonora .— Near the margins of the Sonoran desert, coastal and foothill thomscrub, and tropical deciduous forest, in disturbed as well as natural habitats; 50-ca. 1400 m. Growing and flowering mostly following war® weather rains, especially August-October. General distribution .— Mexico (probably most or all states); Mesoamerica; Argentina, Bolivia, Brasil, Co- lombia, Guyanas, Ecuador, Venezuela; Caribbean; cultivated and introduced into the Old World. The seeds have been utilized as a laxative (Austin 2000c). There are three species of somewhat simi#' Felgeret al, Convolvulaceae (excluding Cuscuta) of Sonora, Mexico S03 appearing morning glories that are often confused. One is I. nil, characterized by long, straight, subulate sepals, in tropical regions worldwide. The second is I. hederacea, also characterized by long sepals, but curving and with an ovate base and abruptly attenuate apex, in temperate regions by latitude or higher elevations in tropical latitudes. The third member is I. purpurea, which has short fat sepals and is worldwide. Ipomoea nil and I. hederacea can be hybridized with difficulty (Yoneda and Takenaka 1981). Historically they were geographically isolated. Superficially I. nil seems similar to I. hederacea, but detailed morphological and molecular genetics show that they are not as close as they may seem. Furthermore, I. nil seems closer to I. eriocalyx Mart, of South America than to I. hederacea (Austin et al. 2001). Selected references.— Austin (2000c), Austin et al. (2001), McDonald et al. (2011). nth) G. Don, Gen. Hist. 4:275. 1838. Tm: VENEZUELA: Humbol is parasiticus Kunth, Nov. Gen. Sp. (quarto ed.) 3:103. 1818 [1819]. Proc. Amer. Acad. Arts 21:319-320. 1894. Type: MEXICO. Jalisco: Tequila, 1 Herbs, annual, the stems twining, herbaceous, to 2-7 m long, with fleshy tubercles on the older parts, or smooth, glabrous or glabrate. Leaves 5-10 cm long, 6-9 cm wide, entire, cordate to broadly cordate, apically acuminate to rounded and cuspidate, lightly pubescent or glabrous above, mostly glabrous below. Inflores- cences usually simple cymes, less often compound-cymose. Flowers 2-10. Sepals 4-6 mm long, more or less e qual, the outer ones ovate to ovate-lanceolate, more or less acute and cuspidate, with small appressed-puber- u, ent indumentum. Corollas 2.5-3 cm long, funnelform, purple, sericeous on the interplicae. Capsules 10-12 mm lon g> ovoid to ovoid-globose, reflexed, glabrous. Seeds 1-4, 6-7 mm long, narrowly ellipsoidal, brown, Wlth sma11 Puberulent indumentum or glabrous. Sonora.— Eastern Sonora in Chihuahuan desert, foothills thomscrub, tropical deciduous forest, and ri- P-Jrtan in oak woodland; 260-1290 m. Flowering November to February. Although not known from the Unit- e States, its occurrence at Agua Prieta suggests that it might be found in nearby Arizona. General distribution .— Baja California Sur, Chiapas, Chihuahua, Guanajuato, Guerrero, Jalisco, Edo. Mexi- c °. Michoacan, Morelos, Nayarit, Oaxaca, Sinaloa, Veracruz; Mesoamerica; Brazil, Colombia, Peru, Venezuela. These Pknts contain an ecdysone that was considered at one time as a potential insecticide. Subsequently, or similar compounds were found in I. nil and became commercially known as “kaladana” (Austin 504 Journal of the Botanical Research Institute of Texas 6(2) . Type: MEXICO. Chiapas: wooded slope 9 km N of Tuxda GutiCrrez along MANTELA DE MARIA, TROMPILLO; JtCURE ‘TWISTED YARN’ (Mayo) Lianas, perennial, the stems 2-10 m long, twining with a woody base, often winged, the younger parts herba- ceous, smooth to somewhat winged, glabrous or pubescent. Leaves 3.5-14 cm long, 3-12 cm wide, ovate, often broadly ovate, entire or 3- or 5-lobed, chartaceous, basally cordate, apically acuminate, glabrous or pubescent on both surfaces at least near the base. Inflorescences cymose, axillary. Flowers (2-)5-15(-35). Sepals un- equal; the 2 outer sepals 4-7 mm long, ovate-lanceolate, broadly elliptic to oblong, acute, muricate or more of- ten with wings on the lower portion; inner sepals 5-8.5 mm long, obtuse to obtuse-mucronate, coriaceous, the margins scarious, glabrous or pubescent only toward the apex. Corollas 5-9 cm long, funnelform, purple, slightly pubescent at the apices of the petals. Capsules 10-19 mm long, ovoid, brown, apiculate, glabrous. Seeds 1-4, 6-7 mm long, ellipsoid, brown, minutely puberulent or glabrescent. Sonora . — Central and southern parts of the state in coastal and foothills thomscrub and tropical decidu- ous forest; ca. 50-550 m. Flowering September to November. The seeds, ground, roasted, and boiled, were used by the Guarijlos as a purgative (Gentry 1942). The Guarijlos also used a morning glory, possibly this spe- cies, as follows: “If a woman does not wish to have a child, she will sometimes eat the seeds of trompillo, grind- ing up the seeds, mixing the gruel into water, and drinking it” (Yetman & Felger 2002:193). The Mayos con- sider the herbage of this fast-growing summer vine to be valuable forage for cattle and goats (Yetman and Van Devender 2002). General distribution. — Chiapas, Chihuahua, Colima, Guerrero, Hidalgo, Jalisco, Michoacan, Morelos, Nayarit, Oaxaca, Puebla, San Luis Potosl, Sinaloa, Tamaulipas, Veracruz; Mesoamerica. Selected reference .— Austin (1997). Mpk> Alamos: San Bernardo, 24 Aug 1935, Gentry 1616 ; Rio Mayo Raft Trip, confluence of Rio Mayo with San Ignacio, 27°55'N, 108°4TW, Jenkins & Rondeau 26 Sep 1991!; 23.3 mi by road W of Alamos, Soule & Krizman 2 8 Aug 1964!; Arroyo el Mentidero at El Chinal road, 113 b» S of Alamos, 26°54 I 45"N, 108°55'05" W, 240 m, 20 Sep 1993, Van Devender 93-855!; Alamos, 28 Sep 1991, Van Devender 91-755 !; Gueybatnpo. E edge of Arroyo Tojibampo, 26°42'30"N, 109°16’W, 50 m, 22 Sep 1994, Van Devender 94-713 !; 0.3 km E of Tqjibampo (S side of the Sierra* Alamos), 26°48'50"N, 108°58'W, 240 m, 21 Sep 1994, Von Devender 94-659!; Arroyo Huirotal, Rancho Las Uvalamas, E slopes of Sierra* Alamos, 550 m, vine 6-8 m in trees, flowers purple, 13 Sep 1994, Van Devender 94-564 (AR1Z!, ASU!). Mpio Huatabampo: 1 km SEofCer- rillos, 9.5 km SE Melchor Ocampo, 40 m, flowers purple, 21 Sep 1994, Friedman 347-94 (ASU!). Mpio La Colorada: 7 mi NE of Colorado, between Colorado and Mazatan, climbing vine to 8 or 10 m, corolla rich purple, 6 Sep 1941, Wiggins & Rollins 323 (CAS!): 3.5 km SE * ' from Navajoa-Alamos, sandy bottomlands, thorn forest, seeds used as purgative, M antela de Maria, 5 Nov 1939, Gentry 4880 ! Felger et al., Convolvulaceae (excluding Cuscuta) of Sonora, K Ipomoea pes-caprae (L.) R. Br. in Tuckey subsp. I isiliensis (L.) Ooststr., Blumea 3:533. 1940. i Batatilla, cmmisTATE de playa; bejuco de playa; beach morning glory, railroad vine, g Herbs, perennial, the stems mostly prostrate on beaches, rarely twining, fleshy, to 10 m long (usually much shorter in Sonora), glabrous, and with milky sap. Leaves 3-10 cm long, 5-10 cm wide, ovate to reniform, ba- sally rounded, truncate to cordate, apically normally emarginate, the blade with 2 glands near the base. Inflo- rescences axillary, monochasial and/or dichasial. Flowers 1-5. Sepals 5-11 mm long, equal or unequal, ellip- tic, ovate-elongate to orbicular, glabrous. Corollas 5-6 cm long, funnelform, pinkish or lavender, the throat darker within, glabrous. Capsules 1.5-2.2 cm long, rounded, straw-colored or brown with purplish patches, glabrous. Seeds 1-4, 8-9 mm long, rounded, densely brown-tomentose. 2 n » 30, 60. Sonora . — Cultivated on beaches from the vicinity of Bahia Kino southward and sometimes weakly estab- lished, and perhaps native or at least established on beaches along coastal thomscrub in extreme southwestern Sonora; near sea level. Flowering at least during the warmer months. General distribution . — This species occurs on beaches and coastal dunes worldwide; subsp. brasiliensis is widespread in the New World tropical shores. Texas; Baja California (norte) and Sur, Chiapas, Guerrero, Jalis- co, Michoacan, Oaxaca, Sinaloa, Tamaulipas, Veracruz; Mesoamerica; Brasil, Colombia, Ecuador, Guyanas, Pern, Venezuela; Caribbean. Subspecies pes-caprae occurs in a narrow range in the Indian Ocean. A close rela- tive, I. asarifolia (Desrousseaux) Roemer & Schultes, has been introduced into wetlands in the American trop- ics (Austin 2005). Selected reference.— St. John (1970). it, 26.6917°N, 109.587°W, o a (by air) S of Us Bocas, 26.65°N, I09.331°W, s< Perennial herbs with globose, tuberous roots, the stems erect to procumbent, ascending, not twining or only slightly at tips, glabrous. Leaves 1-3 cm long, orbicular in outline, palmately and pedately lobed (rarely cune- ate-obovate in which case the apex is laciniate-dentate), basally cordate, the lobes acute, glabrous; petioles 2-5 mm ^ng Inflorescences solitary. Flowers on peduncles 1.5-2.5 cm long, erect or reflexed in fruit; bracts ca- duceus. Sepals unequal; outer sepals 5-8 mm long, 2-3 mm wide, oblong, obtuse to acute, mucronate, muri- ate at least along the midrib; inner sepals 7-9(-10) mm long, 3-4 mm wide, broadly ovate, acute to acuminate, muricate on the midrib or glabrous. Corollas funnelform, 2.5-3.1 cm long, purple, glabrous, the limb 1.8-2.2 cm wide. Capsules ± globose, 5-6 mm wide, with an apiculum to 5 mm long. Seeds 1-4, 2-2.5 mm long, ovoid, black to dark brown, finely tomentose. Sonora.— At least in southern Sonora, coastal thomscrub and in pine-oak forest in southeastern Sonora near the Chihuahua border. Also southeastern Arizona near the border and expected in nearby Sonora and Perhaps elsewhere in mountains in eastern Sonora. Open rocky slopes. 20-above 1250 m. Flowering April- October; General distribution. — Arizona, New Mexico, Texas; Chihuahua, Coahuila, Distrito Federal, Durango, ^go, Jalisco, Edo. Mexico, Michoacan, Puebla, Veracruz, Zacatecas; Argentina, Bolivia, Peru. McDonald " 5) rec °gnized several widespread varieties, with the Sonoran population being var. plummerae. Journal of the Botanical Research Institute of Texas 6(2) Although McDonald (1995) recognized var. cmeifolia as distinct, DFA has seen individual populations in southern Arizona with both “var. cuneifolia” and “var. plummerae ” growing intermixed. We suspect that the two “varieties” are nothing more than variants in leaf shapes. This species is easily confused with I. capillacea. Leaf segments are <1 mm wide in I. capillacea, and >1 mm wide in 1. plummerae. Also, I. capillacea is an erect herb, while I. plummerae is a prostrate or rarely twining Selected reference. — McDonald (1995). Mpio Alamos: Cliffs 5 km W of Chiribo, Sierra Saguaribo, 1400 m, pine-oak woods, Martin 24 Aug 1993. Mpio Yecora: Yecora, 28°22'25"N, 108°56'W, 1540 m, 6 Sep 1995, Rema-G. 96-373. CHIHUAHUA. Mpio Buenaventura: Rio Santa Maria, 10 km SW of Buena Ventura [Buena- ventura], 2100 m, 14 Aug 1989, Jenkins et al. 89-311. Ipotnoea pubescens Lam., Tabl. Encycl. 1:4 5. 1791 [1793]. Type: AMERICA, c 1707. Type MEXICO: horto Regio, 17 York Acad. Sci. 18:196. 1908. Type: U 7. Ortega (lectotype MA! , Perennial herbs from a large, oblong root; stems twining, hirsute with retrorse trichomes. Leaves with blades 2-8 cm long, 2-9 cm wide, ovate, nearly entire with the margins sinuate, or 3- or 5-lobed and palmate, basally cordate, the lobes elliptic to ovate, hirsute with antrorse trichomes, at times sericeous, apically acute to obtuse, mucronate; petioles 2-5 cm long. Inflorescences 1- or 2-flowered. Flowers on peduncles 15-18 mm long; pedicels 3-10 mm long, erect in fruit; bracts 5-12 mm long, subulate, acuminate. Sepals unequal; outer sepals, 9-21 mm long, 5-11 mm wide, ovate, basally truncate, acuminate; middle sepals 9-19 mm long, 3-8 mm wide, asymmetrical, ovate, acuminate; inner sepals 9-20 mm long, 2-4 mm wide, ovate-lanceolate, antrorsely his- pid, at times sericeous. Corollas 5.5-8 cm long, funnelform, blue to violet with a white throat, glabrous, the limb 6-7 cm wide. Capsules 10-12 mm wide, ± globose, surrounded by the sepals. Seeds (1-) 3-6, 5-6 mm long, ovoid, brown to black, densely hairy with velvety trichomes. Sonora.— Mountains in eastern Sonora. It grows in south-central and southeastern Arizona near the bor- der and is expected in nearby Sonora. Oak woodland and pine-oak forests, especially rocky areas and near streambeds; ca. 1200-2000 m. Flowering August-September. General distribution. — Arizona, New Mexico, Texas; Chihuahua, Durango, Hidalgo, Michoacan, Quere- taro, San Luis Potosi; disjunct to South America. The Arizona plants have larger flower than those listed from South America by O’Donell (1959a) and there is perhaps more than one taxon involved. McDonald et al. (2011) have placed all the large-flowered plants of northern Mexico and the SW United States in I. lindheimeri. We cannot agree because the sepals are distinct within the populations that we have studied in Sonora and Arizona. McDonald (pers. comm., June 2012) says that he has found what appear to be hybrid swarms between the typical I. lindheimeri with narrow lanceolate- linear, cuneate sepals and plants of I. pubescens with broadly ovate, basally truncate sepals. We have not seen such populations and maintain them as separate species until future studies can be made of their relationships Selected reference.— Austin (1991). s: Sierra Saguaribo, Chir d, rare, one patch along s 1 24 Aug 19931; Sierra Sahuaribo, ca. 2 km (by air) NE of La Vinateria, 27°17'30 , N, Ipio San Felipe de Jesus: Cajbn Infiemo, northern Sierra Aconchi, 4480-5120 ft •row part of canyon, 18 Sep 1982, Reichenbacher 11391 Mpio Yecora: ReS "‘^ ;. 2000-561 (USONt). Ipomoea purpurea (L.) Roth, Bot. Abh. Beobs Ipomoea hirsutula Jacq. f. U.84,f.97,1732(tECTorrf* ype. Pharbitis htspida Cho^f’ r. diversifolia (Lindl.) Journal of the Botanical Research Institute of Texas 6(2) Ipomoea scopulorum Brandegee, 1 (holotype: UC!, photo FTG-FAU!). : 5:169. 1903. ^MEXICO. B*a California [Sur]: Cape F Perennial twining herbs, with tuberous roots?, 1-2 m long, sparsely to densely pubescent throughout with simple and stellate trichomes. Leaves 6-8 cm long, 4-6 cm wide, cordate-ovate, entire, stellate-pubescent above and below; petioles 2-5 cm long. Inflorescences cymose or solitary, axillary. Flowers on peduncles 2-9 cm long, the pedicels 2-3 cm long, erect in fruit; bracts 2-3 mm long, subulate, caducous. Sepals 7-10 mm long, 4-6 mm wide, unequal, broadly oblong to broadly ovate, smooth, rounded to obtuse, the upper margins some- what scarious, glabrous or less often sericeous on the outside of the outer two sepals. Corollas 6-8 cm long, 6-7 cm wide, funnelform, white with a purple throat or less often purplish throughout, glabrous. Capsules ovoid, 10-14 mm long, 8-10 mm wide. Seeds 1-4, 6-8 mm long, ovoid, brown, silky, with white or tawny trichomes 8-10 mm long on the margins, otherwise glabrous. Sonora . — Southern to north-central parts of the state; Sonoran desert, coastal and foothills thornscrub, and tropical deciduous forest; 150-725 m. Flowering August-September. General distribution . — Baja California Sur, Sinaloa. Ipomoea seaania Felger & D.F. Austin, Sida 21:1296. 2005. Type: MEXICO. Sonora: Mui 1 km north of Bahia San Carlos on old road to Bahia Algodones, 27 Feb 1985, Felger & Devini MEXU!, MO!, NY! See label information below). Openly-branched shrubs 1-4 m. tall, with multiple woody stems from the base, the upper twigs sometime 5 sinuous or moderately spiraling, sometimes becoming extremely slender. Herbage largely glabrous or glabrate except newest growth densely to sparsely short-pilose, the trichomes mostly spreading. Leaves drought de ciduous, lanceolate to ovate, long shoot leaves often 2-8 cm long, 1.5-2 cm wide, the blades lanceolate to ovate, with 6— 8(— 10) lateral pairs of primary veins, the base obtuse to subtruncate, the apex obtuse to emarginate, tk midrib often ending in a short mucrone, glabrous; petioles 8-15(-20) mm long, with a pair of glands, usua^ conspicuous, at the junction of petiole and blade, seen on the lower leaf surface. Spur branch leaves often 4^ cm long, to 4-11 mm wide, linear to linear lanceolate with 6-10 lateral pairs of primary veins, the base obtu# to subtruncate, the apex obtuse or blunt, or sometimes shallowly emarginate, the midrid often ending io 1 Felgeretal., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico 509 short mucrone; petioles to 2-9(-13.5) mm long. Inflorescences of 1 or 2 (3) flowers, appearing solitary but cymose on short-shoots 2-5 mm long, these sometimes with a few small leaves; bracts 5-8 mm long, quickly deciduous, broadly oblong with an obtuse tip; peduncles very short, to 5 mm long, the pedicels 8-22 mm long. Sepals 12-17 mm long, 6-8 mm wide, broadly lanceolate to mostly ovate, puberulous to villous, the inner surfaces generally more densely hairy than the outer surfaces, the trichomes white, appressed to mostly as- cending, and curly to straight. Inner (adaxial) sepals obtuse, the surfaces with trichomes 0.15-0.6 mm long; outer (abaxial) 2 sepals acute, slightly narrower and more sparsely pubescent than the inner 3, the trichomes 0.1-0.5 mm long, the sepal margins scarious and glabrous or glabrate. Corollas 4-6 cm long and 7-8 cm wide, showy, funnelform, glabrous, white with yellowish interplicae and a maroon band at inside base of the tube. Stamens 5, with 4 filaments 25-26 mm. long, the fifth stamen 23-24 mm long, filaments pubescent on the basal 4 mm; anthers 6-7 mm, long oblong, sagittitate; pollen spheroidal, spinulose. Ovary 3 mm long, gla- brous; style 37-38 mm long, glabrous; stigma 2-globose. Sonora . — Near the southern margin of the Sonoran desert where it is endemic to hills and mountains west of Bahia San Carlos, generally on rocky ridges, steep colluvium and rhyolite slopes of canyon sides and cliff base; near sea level-100 m. Although locally common, the San Carlos population is threatened by tourist de- velopment (Felger 1999). There are no other records for this unusual shrub. Flowering January-April. Ipomoea sescossiana Baill., Bull. Mens. Soc. Linn. Par. 1:385. 1883. Type MEXICO. San Luis Potosc Sescosse s.n. (holo- Perennial erect herbs, with tuberous roots, the stems shrubby, spreading, glabrous. Leaves 2-4 cm long, 0.5- 1-5 cm wide, orbicular in outline, irregularly pinnately divided into 5-9 obtuse-tipped divisions, the lobes linear to filiform, entire or rarely toothed, glabrous; petioles 3-15 mm long. Inflorescences mostly of solitary flowers, at times 2 or 3 flowers. Rowers on peduncles 0.8-3.5(-8) cm long, the pedicels 0.6-1.5 cm long, ac- crescent and erect in fruit; bracts 2-3 mm long, lanceolate. Sepals 6-ll(-13) mm long, 3-6(-8) mm wide, subequal, oblong-ovate, smooth, obtuse-rounded to emarginate or cuspidate, the margins scarious. Corollas 6-10 cm long, funnelform, rose-purple to purple, the tube whitish, glabrous, the limb 6-8 cm wide. Capsules 14-16 mm long, 10-14 mm wide, ovoid to almost globose. Seeds 1-4, 8-9 mm long, compressed ovoid, black, finely appressed tomentose. Although not documented for Sonora, it occurs in Chihuahua near high elevations in east-central Sonora and * likel Y to turn up in the highlands of easternmost Sonora. General distribution.— Chihuahua, Coahuila, Durango, San Luis Potosi, Zacatecas; 1000-2400 m. Flow- erin gJune-October. This species is an erect herb with bee-flowers. This xenogamous species is sister to I. ancisa and I. stans (McDonald 2001; Ana Rita Simoes, pers. comm. 23 Nov 2011). CHIHUAHUA. Mpio Moris: La Cieneguita, Rfo Mayo Upper Sonoran, 10 Sep 36, Gentry 26481 [ca. 35 km E of Sonora]; Mpio Madera: 510 Journal of the Botanical Research Institute of Texas 6 ( 2 ) Chuichupa, Aug 1936, Horde LeSueur 850\ [ca. 25 km E of Sonora] . Mpio Temosachi: N W of Yepdmera on hwy 180, 2100 m, flowers mostly pale lavender, shrub Jenkins, Martin, & Moore 21 Jul 1986\ Ipomoea tenuiloba Torn, Rep. U.S. Mex. Bound., Bot. 148-149. 1859. Type: U.S.A. Texas: hills and rocky places near pedately 5-9 lobed, the lobes linear to lanceolate, 0.5-6.5 mm wid Perennial herbs, with tuberou long, 3-10 cm wide, orbicular : entire, glabrous; petioles 2-38 1 mm long, sessile or the pedicels to 8 mm long, ate. Sepals unequal; outer sepals 5-11.5 mm lor rved in fruit; bracts 1-3 mm long, 1 -3 mm wide, oblong-lanceolate, muricate along the midrib nate, smooth, the margins scarious, glabrous. Corollas 4.5-10 cm long, funnelform or salverform, completely white or white with pale rose to purple limb, glabrous, the limb 3-3.6 cm wide. Capsules 4-8 mm wide, ± globose to broadly ovoid, with an apiculum 4-5 mm long. Seeds 1-4, 3.5-5 mm long, ovoid, black to dark brown, finely appressed tomentose. This delicate twiner has moth flowers that McDonald (2011) considered xenogamous. Sonora . — Documented in Sonora with four specimens from mountains in the eastern part of the state, in hills and rocky sites, oak woodland and pine-oak forest; ca. 1280-1950 m. Flowering August-September. General distribution. — Arizona, New Mexico, Texas; Chihuahua. Two varieties are recognizable: Variety tenuiloba with 5-7 leaf divisions to 1.2 mm wide, a mostly white salverform corolla 6.5-10 cm long and the free portions of the filaments 8-11 mm long. Variety lemmoni (A. Gray) Yatsk. & Mason with 7-9 leaf divisions to 6.5 mm wide, a funnelform corolla 3.5-6.5 cm long, with a white tube and rose to purple limb and the free portion of the filaments 14-19 mm long. In some areas these varieties intergrade, and a specimen from northeast Sonora (White 3474) was considered an intermediate by Yatskievych and Mason (1984). The species is nocturnal and those studied in southern Arizona opened about 1:00 a.m. and often closed before dawn (Austin 1991). Mpio Alamos: Sahuaribo and vicinity, 27°19'N, 108°34'W, 1550 m, Martin 23 Aug 19921 Mpio Bavispe: Sierra el Tigre, Las Tierritas deB 28.38°N, 108.8258°W, 1410 m, very common herbaceous perennial in dense open at 9:30 a.m., 2 Sep 2000, Van Devender 2000-640 (USON!); Yficora, 28.3636°N, 108.9228°W, 1540 m, grassland, locally conn .,5 Sep 2001, V fectively selected the Thurber 977 Annual herbs, with a fibrous root system; stems slender and procumbent, twining at tips, glabrous. L*® 1 ** with blades pedately 5 or 7 parted, 1.5-3 cm long and wide, basally cordate, the segments linear, apically acute, glabrous to remotely setose; petioles to 2 cm long. Inflorescences of solitary flowers or cymose. Flowers 1 ° r 2, the peduncles slender, equaling or exceeding the leaves, the pedicels (10-)15-18(-20) mm long, erect in fruit; bracteoles 2-2.5 mm long, subulate. Sepals unequal; outer sepals 8-9 mm long, 2-3 mm wide, lanceolate; a wide, attenuate-acuminate, glabrous or hirsute at least on the three main ns, otherwise membranaceous. Corollas 2.5-3.2 cm long, funnelform, purple (oc- casionally white), glabrous, the limb 3.2-4.6 cm wide. Capsules 3-5 mm wide, ellipsoid to ovoid depending on the number of seeds, with an apiculum 5 mm long or longer. Seeds 1-4, 5-6 mm long, ovoid, black and gray mottled, minutely appressed tomentose. Sonora . — Statewide except the northwest; Sonoran and Chihuahuan deserts, foothills thomscrub, tropi- cal deciduous forest, mesquite-grassland, and oak woodland; near sea level-ca. 2000 m. Flowering January andJune-November. General distribution .— The species, with 3 allopatric varieties, is widespread in deserts and tropical de- dduous forests of southwestern United States to El Salvador. Variety leptotoma , the northernmost one, occurs in southeastern Arizona, southwestern Chihuahua, and most of Sonora and Sinaloa, and Baja California Sur. This species is self-incompatible and bee pollinated (McDonald 1995; McDonald et al. 2011). Selected reference .— McDonald (1995). MpioAgoaPrieta: Colonia Morelos, 2600 ft, 5 Oct 1941, White 4619! Mpio Alamos: Alamos, 16 Oct 1936, Gentry 2928!; 3.7 km N of Guiro- coba, 450m, abundant annual infield, flowers white, 23 Sep 1994, Van Devender 94-759!; El Guayabo (upper) crossing of the Rio Cuchujaqui 3kmNE of Sabinito Sur and 15 km (airline) ESE of Alamos, near 27°00'N, 108°47’W, 350 m, tropical deciduous forest, 12 Oct 1992, Sanders white flowers among abundant purple-flowered plants on disturbed roadside, flowers open 3:45 p.m., foothills thomscrub, 16 Sep 2000 1934, Shreve 6794!; Rancho el Carrizo, 2400 ft, Tomelson 7 Aug 1968\\ 2 km S of Los Chinos, 23 km N of El Oasis, foothills thomscrub, 665 m, flowers lavender, drying magenta, open 2:00 p.m., 6 Jan 2001, Van Devender 2001-14! Mpio La Colorada: 3 mi E of La Colorada, 1300 ft, 12 Jan 1981, Fischer 6869!; 4 km W of Cobachi (dirt road between paved roads to Tecoripa Cnrarpe: Rancho Agua Fria, flowers white, Van Devender 6 Sep 197 6!; 4 mi SW of Agua Ft Gunpas: 5 km ESE ofjecori, road to Moctezuma, 26 Oct 1984, Felger 84-380! Mpio Fronteras: 2 nal purple flowers, 4200 ft, 7 Sep 1960, Felger 4048! Mpio Guaymas: Arroyo Las Pirinolas, roat Joside Robinson, 28.1112°N, 111.036667°W, 100 m, Felger 01-7421 Mpio Hermosillo: La Puerca, 10 a.m, 4 Sep 2000, Van Devender 2 000-671! Mpio Moctezu ith white throat, open 8:00 am, 14 Aug 2006, R eina-G. 2006-468! Mp! DV-NW of San Jose de Masiaca, 26.78170°N, 109.295°W, 200 m, 19 Jan 1995, Friedman .an N of Teachive, SW end of Cerro 1 nscmb. 26°48.1'N, 109°14AV7, 100 m, 28 Sep 1' 96-582! Mpio Opodepe: 5.6 mi on Mex Hwy 15 S of Be 154-95 (ASU!); 2 51 3206°N, 110.367°W, 1555 m, grassland with isolated , 30°23'38"N, 110°3r43"W, Meling I Oct 1986 (USON!). Mpio :s, 29°25'30"N, 110°23'21"W, Meling 19 Sep 19 &(US0N!). Ipomoea thurberi A. Gray, Syn. Fl. N. Amer. 2:212. 1878. Type: U.S.A. Arizona: Thurber 966 (ho^typh: GH!). Ipomoea gentry! Sundl., Field Mus. Nat. Hist. 22:46. 1940. Type: MEXICO. Chihuahua: Gentry 2497 (holotype: F!; isotype: MO!). Glabrous perennial twining herbs from an elongate, tuberous root, the stems trailing. Leaves with blades 1-5 on long, 2-6.5 cm wide, sagittate to pedately 5 or 7 lobed, the lobes divergent, lanceolate, linear to oblong ^tate, sparsely strigose; petioles 0.9-2 cm long. Inflorescences solitary. Flowers on peduncles 3 1,1111 lon g, the pedicels 7-8 mm long, erect in fruit; bracts ca. 1 mm long, scale-like. Sepals 12-15 mm long, 3 mmurirb, . . ,, -u obscurelv warty at base or not. wide, ± equal, lanceolate, « HI | -orollas 5-8 cm long, funnelform-salverform, white with a ous-caudate, obscurely warl 7 mb and green tube (drying purple), opening 512 in the evening, glabrous, the limb 5-6.5 cm wide. Capsules 6-7 mm wide, ± globose to ovoid, with an apieu- lum 4-5 mm long. Seeds 3-6, 3.5-4 mm long, ovoid, black to dark brown, finely tomentose. Sonora . — Mountains of eastern Sonora in grassland, oak woodland, and pine-oak forest; at least 1150- 1600 m. Flowering Juiy-September. The flowers open at dusk when they are visited by hawkmoths, perhaps Hyles lineata and probably others. Also near the Sonora border in Cochise and Santa Cruz Counties in Arizona Herbs, annual, the stems prostrate and twining, 1-3 m long, glabrous or occasionally sparsely pubescent, the indument concentrated on the nodes. Leaves 2-8 cm long, 2-7 cm wide, broadly ovate to orbicular, entire, coarsely dentate to deeply 3-7-lobed, basally cordate, the basal lobes rounded to angular or lobed, apically acute to obtuse, both surfaces glabrous or sparsely pilose. Inflorescences axillary, cymose. Flowers 1— 3(-12) with peduncles variable in length from shorter to longer than the petioles, glabrous, angular, minutely ver- ruculose at least toward the apex. Sepals 6-7 mm long, ± equal, the outer sepals oblong to narrowly elliptic- oblong, obtuse to acute, mucronulate-caudate, glabrous or sparsely pubescent without, conspicuously cihate, mostly glabrous otherwise. Corollas 1-2 cm long, funnelform, lavender, glabrous. Capsules 5-6 mm wide, ± globose, brown, bristly pubescent. Seeds 1-4, 2.5-3.2 mm long, globose, dark brown, glabrous. 2 n = 30, 60. Sonora .— A weed carried, at least partly, as a contaminant in rice and others seeds. Sonoran desert, coastal thomscrub and tropical deciduous forest, often in disturbed sites; 0-950 m. Flowering all year. General distribution. — California; Campeche, Chiapas, Guerrero, Jalisco, Edo. Mexico, Nayarit, Nuevo Le6n, Oaxaca, Sinaloa, Tabasco, Tamaulipas, Veracruz, Yucatan; Mesoamerica; Argentina, Bolivia, Brasil, Co- lombia, Ecuador, Paraguay, Peru, Venzuela; Caribbean. Felgeretal., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico Vines, herbs or suffrutescent shrubs; stems herbaceous toward tips, procumbent to twining (except some- times in ]. agrestis ), perennials or occasionally annuals, glabrous or hairy. Leaves chartaceous to herbaceous, mosdy cordate, glabrous or hairy usually with stellate trichomes, entire or variously repand, dentate, or lobate, petiolate. Inflorescences in scorpioid cymes, head-like cymes, umbelliform, or flowers solitary. Flowers on pedicels 5-30 mm long, the bracts small and linear or lanceolate or large and foliose. Sepals equal or unequal, variable in shape, hairy or glabrous. Corollas blue, lilac, or white (red in one West Indian species), subrotate, campanulate, or funnelform, deeply lobed, dentate or almost entire, glabrous or hairy. Stamens and styles in- cluded (in our species). Pollen 3-colpate, 12- or 15-rugate. Ovary 2-locular, 4-ovulate, glabrous or hairy; styles 1, filiform; 2 stigmas ellipsoid or oblong and flattened. Fruits capsular, 2-celled, with 4 or 8 valves, globose to subglobose. Seeds 1-4, glabrous or pilose, or tuberculate, or winged. About 80-100 species in the Americas and several in Australia. Selected references .— Austin (2006), Robertson (1971). Jacquemontia abutiloides Benth., Bot. Voy. Sulphur 34-35. 1844. Type: MEXICO. Baja Cauforn* [Sur]: Bay of Magdalena, Hinds s.n. (holotype: K!; photo MO!). Scrambling and twining perennial vines, the stems generally woody near the base and slender above growing into and overtopping shrubs reaching 2 (3) m long; younger stems and herbage with 3-branched stellate hairs, mostly crowded but varying from densly overlapping to sometimes moderately dispersed when vigorously growing following ample rain, the hairs nearly white when young and golden-brown with age. Leaves ovate to broadly ovate, to 8 cm long and 4.5 cm wide, usually Vi to% that size, apically obtuse, acute or rarely retuse, mucronate, acuminate or cuspidate, basally cordate with a deep and wide sinus, pubescent; petiolate. Inflores- cences axillary, cymose, rarely solitary, the dichasia 1-2 times compound, the peduncles 2.5-8(-13.5) cm lon g. Flowers on pedicels 1-6 mm long, the bracts linear, to 12 mm long. Sepals of different sizes, the outer 2 ovate to narrowly ovate with attenuate apices, 7-11 mm long, 3-6 mm wide, the middle sepal narrowly ovate, attenuate apically, sometimes slightly falcate, the inner 2 sepals ovate or broadly ovate, 5-7 mm long, 2.5-3 mm ^de, apically attenuate, the bases narrowed with a stipe ca. 1 mm long; all 5 sepals pubescent on the outer surfaces and to a lesser extent on the inner surfaces, enlarging slightly in fruit. Corollas (1.8-) 2-3.5 cm broad when open, broadly campanulate to rotate, blue. Stamens ± equal or unequal, the anthers 1.5-2.5 mm long, ^ary ovoid, 1.5-2 mm long, the styles 6-7 mm long, longer than the stamens. Capsules 5-6 mm ong broadly ovoid, usually opening by 8 segments, partly enclosed by sepals. Seeds 1-4, 2.7-3.5 mm long, l.B-2.3 ““a wide, trigonous, minutely areolate and ruminate. j J . . , ^ region. — Sonoran desen on Ida Tiburdn from near sea level-a. least 490 m; widespread m the b 'en mountain mass (Sierra Kunkaak) and its eastern bajada to the south shore of the island, especially a o g Was ^ les and canyons, andalsoon desert plains and rocky slopes (Felger et al. 201 2). Flowering with sufficient -tat a. various seasons, flowers recorded October-April Not known from mainland Sonora, whteh Seems unusual since seemingly similar habitat to that on the island occurs on die adjacent Sonora mainland * heic Fel s er et al. (2012) have searched for it. 515 Jacqnemontia pentanthos (Jacq.) G. Don This species has not been verified for Sonora. While it is common in eastern Mexico, it is either absent or rare in western Mexico. According to Robertson (1971) J. pentanthos is at the center of a group of species that in- cludes J. abutiloides.J. albidaj. eastwoodianaj. polyanthaj. pringlei, and five others. This group of taxa ranges from Arizona to Central America. We tentatively accept the distinctions made by Robertson (1971) in these segregates (except for J. albida), although the relationships of these taxa generally remains confused. These taxa are often misidentified or difficult to identify on herbarium specimens, while living plants are often con- siderably different. Within this alliance J. polyantha and J. pentanthos are closely related (see J. polyantha ). Jacquemontia albida Wiggins & Rollins, Contr. Dudley Herb. 3:277. 1943. Type: MEXICO. Sonora: [Mpio Hermosillo] Along an arroyo 1 mile NE of El Zapo, between Hermosillo and Tastiota, 4Sep 1941, Wiggins & Rollins 273 (holotype: DS at CAS!; isotypes: ARIZ!, GH!, LL!, MICH!, MO!, NY!, RSA!,US!). Paneya (Guarijio, Gentry 1081) Perennial vines, sparsely to densely pubescent with 3-armed trichomes. Stems twining to several meters long or procumbent, herbaceous or becoming woody near the base. Leaves: Blades 2.5-11.5 cm long, 1.5-6.5 cm wide, ovate, broadly ovate, to almost circular; basally cordate to less often truncate; apically highly variable— pand to undulate. Petioles 0.2-6.5 cm long. Inflorescences axillary, loosely cymose, of simple or often com- pound dichasia; peduncles 5-13 cm long and often longer than the leaves. Flowers on pedicels 0.5-2 cm long, the bracts linear, often 5-10 mm long. Sepals unequal to more or less equal; the outer two sepals 6-6.5 mm long, 3.5 mm wide, elliptic, broadly elliptic, or ovate to broadly ovate, the bases narrowed, the the apices acute or acuminate, or shortly attenuate; the inner sepals smaller (narrower and shorter), glabrous or sparsely to densely stellate. Corollas 1-2.5 cm long, funnelform to campanulate, white, glabrous. Stamens unequal, 5-12 mm long; anthers 1-2 mm long. Ovary 1.5 mm long, ovoid to cylindrical, glabrous; styles 6-11 mm long. Cap- sules 4-5 mm long, ovoid, partly enclosed by the sepals. Seeds 1-4, 2.7-3 mm long, rotund, brown, minutely Sonora.— From near Hermosillo and the east-central part of the state southward and e; desert, coastal thomscrub, and tropical deciduous forest; near sea level-770 m. Flowering a March, September-December. General distribution— Widespread in Mexico including the Sierra Madre Occidental i Oriental, and Sierra Madre del Sur, and at least in Chihuahua, Guerrero, Edo. M< Tamaulipas, Veracruz; from desert margins to oak and pine-oak zones Robertson (1971: 133) reported that J. albida is closely related to J. polyantha, “which has outer sepals that are broadly ovate or ovate and pubescent instead of elliptic and glabrous.” He knewj. albida only from the vi- cinity of Hermosillo. Many additional specimens available to us from a wide range indicate that differences between J. albida and J. polyantha are too few and do not distinguish them. As Robertson suspected, we con- clude that they are best treated as a single species. Robertson (1971: 168-169) also pointed out that, “Both J. polyantha and J. pentanthos are very closely re- lated.” Primary differences between J. polyantha and J. pentanthos are in inflorescences, sepals, and flower color. Jacquemontia pentanthos has compact cymes and usually rhomboidal outer sepals with long attenuate apices; J. Polyantha tends to have open cymes and elliptic to ovate outer sepals and acute or acuminate to short-attenuate apices. Moreover, J. pentanthos has blue flowers and those of J. polyantha are usually white. 516 Perennial vines, the stems twining, with 4-6-armed stellate trichomes, rarely glabrate. Leaves broadly ovate to ovate, 2-6.5 cm long, 1.5-4.8 cm wide, basally shallowly cordate to truncate, apically acute or less often re- tuse to obtuse, sometimes mucronate. Inflorescences axillary, loosely cymose or solitary. Flowers 1-7 on pe- duncles 1-1 1 cm long, the pedicels 2-10 mm long, erect to nodding in fruit; bracts linear, to 6 mm long. Sepals subequal, the outer ovate to broadly ovate, the middle sepal falcate, the inner narrowly ovate, 5.5-9 mm long, the apices acute, pubescent without and to a lesser degree within. Corollas 1.4-2.7 cm long, funnelform, laven- der to white, glabrous. Stamens 5-10 mm long, included, anthers 1.5-2 mm long. Ovary 1-1.5 mm long, ovoid, 2-locular, glabrous; styles 7-9 mm long. Capsules 5-6 mm long broadly ovoid, enclosed by accrescent sepals. Seeds 1-4, 2-2.5(-3) mm long, trigonous, brown, minutely areolate and ruminate. Sonora .— Widespread except the northwestern corner of the state; Chihuahuan and Sonoran deserts, coastal and foothills thomscrub, and tropical deciduous forest. Near sea level-1000+ m. Flowering (March) August-December. General distribution. — Pima, Yuma and Cochise Cos, Arizona; southwestern Chihuahua to northwestern Sinaloa. Felger et al., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico 517 Lianas or herbaceous annual or perennial vines (also flowering in first season or perhaps annuals), the stems twining or prostrate. Leaves simple or palmately compound, ovate, hastate, sagittate, variably palmately or pinnately divided, the segments 3-9, ovate-linear, the margins entire or sinuate-serrulate, sessile or if present the petiole slender, cylindrical, occasionally sulcate, or rarely scale-like. Inflorescences axillary, in dichasia or monochasia, sometime umbellate. Flowers 1-40, mostly diurnal. Peduncles usually similar to the stems and petioles or reduced or absent, the pedicels usually shorter than the peduncle, smooth, striate, or notably five- angled, slender, stout to distinctly clavate. Bracts glabrous or pilose, usually two, prominent and foliaceous to reduced or scale-like or absent, usually caducous or fugaceous, rarely persisting in fruit, the bracteoles, when present, similar to the bracts. Sepals 5, persistent, imbricate, ovate-lanceolate, herbaceous, membranaceous- coriaceous, glabrous, pilose-appressed, or hirsute, the margins entire, the apex acute to obtuse, attenuate, or rarely emarginate. Corollas campanulate to funnelform, the tube widening gradually or abruptly, the limb more or less entire, the lobes 5-10, white, yellow, or rose-purple, usually glabrous. Stamens 5, included, in- serted at the base of the corolla tube, erect, glabrous or pubescent with glandular trichomes; anthers helicoid- contorted; pollen 3-colpate, or polycoplate (Austin et al. 2008; Ferguson et al. 1977). Style 1, white, glabrous, the stigma 2-globose or superficially 4-globose. Fruits capsular, globose to more or less conical, brown to straw-colored or gray or transparent when dry, entire or 4-lobed, the locules 2-4, the valves usually 4, glabrous, dehiscence valvular or irregular. Seeds 1-4, brown or black (straw-colored in M. aegyptia ), rounded or 1-3-an- gled, glabrous, puberulent, or tomentose. About 100 species, largely in tropical zones; 27 species known from the Americas. Selected references — Austin (1979, 1995), Gunn (1977), O’Donell (1941). Ana Rita Simdes (atBM, working on the Merremieae, pers. comm. 2011). 11862], Stems to 6 m long, herbaceous at least above, with appressed trichomes. Leaves palmately compound, the leaflets 5, 1.5-6 cm long, 0.7-1.4 cm wide, ovate, the base decurrent, the margins entire, undulate, dentate or dentate-sinuate, the apex acuminate, membranaceous, glabrous or appressed-pubescent, strigose or hirsute on both surfaces. Petioles 3-5 cm long, hirsute-pilose to glabrous and/or glandular. Inflorescences of monocasia 519 Felgeretal., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico (USONt). Mpio Baviaccora: Suaqui along SON 89, 29 0 41'46''N, 1 10°09'18"W, 680 m, solitary vine to 3 m in tree away from houses, flowers Merremia palmeri (S. Watson) Hallier f., Jahrb. Hamburg. Wiss. Anst. Beih. 3:38. 1899. Type: Mexico. Sonora: Trompiuo, huirote (these are general names; trompillo for any morning glory and huirote in Mexico for vine) Large, robust perennial vines, glabrous, with milky sap, the stems to 8 m long; growing and flowering during hot weather, often covering trees and shrubs, and leafless and dying back severely in dry seasons. Leaves broadly ovate to circular in outline, palmately compound; leaflets 5, lanceolate 5-14 cm long, 0.6-2.5 mm wide (the median one largest), basally attenuate, apically truncate and mucronate, the margins entire or nearly so; petioles 1.5-2.5 cm long. Inflorescences solitary, axillary. Flowers nocturnal (opening around dusk and quickly wilting with the early-morning heat), on peduncles greatly exceeding the petioles, the pedicles 2.2-4 cm long, the bracts deltoid-ovate and apically caudate, 4-5 mm long, 1-1.5 mm wide. Sepals 2.6-3.5 cm long, in fruit and becoming pock-marked with glands on the inner surfaces. Corollas 5.5-7.3 cm long, 4-6(-8) cm wide, white, salverform to funnelform-salverform, glabrous. Capsules brown, globose to ovoid, 1.5 cm long and wide, subtended and partly enclosed by the accrescent sepals. Seeds 2-4, 9-11 mm long, 6-9 mm wide, black, wedge-shaped, puberulent when fresh, glabrescent. Endemic to Sonora and Sinaloa: Sonoran desert, coastal and foothills thomscrub, and tropical deciduous forest. Northern Sonora from the vicinity of Trincheras, Benjamin Hill, and Presa Angostura southward to Sinaloa; 10-800 m. Flowering recorded (January) March, May-October. This species is self-incompatible. The nocturnal flowers are pollinated by hawk moths, but excess pollina- tor visits can lead to a decline in female reproductive success as a result of flower damage or pollen interference (Willmott & Burquez 1996). Alberto Burquez (personal communication 6 October 2001) writes that “the flow- ers emit a faint, sweet fragrance after opening that lasts through the night. Hawkmonths use it as a long-dis- tance cue. When close, they use eyesight and proboscis.” In contrast, pollination ecologist Robert A. Raguso described the fragrance of flowers remaining open in the morning as being disagreeable (specimen label for Raguso RAR 98-162). The closely related species M. platyphylla (Femald) O’Doneli appears to have bat-pollinated flowers. M. palmeri is also related to the Mexican M. tuberosa (L.) Rendle and M. discoidesperma (Donn. Sm.) O’Donell (Austin 1998c). m, 6 Nov 1982, Starr 192'. Mpio Sahuaripa: Mountains 6.7 mi Wof Sahuaripa, Gates 9 Sep 1959'. Mplo Soyopa: NE side of Rio Yaqui bridge Perennial or sometimes annual herbs, twining or prostrate, glabrous or sparsely hirsute, the stems herbaceous toward the apex, woody toward the base, to 5 m long. Leaves 1.5-5 cm long, 2.5-7 cm wide, palmately com- pound, circular in outline, the leaflets 5, elliptic or lanceolate to oblanceolate, the margins sinuate-serrulate, the segments more or less sessile, apically and basally acute to acuminate, glabrous. Petioles 2-9 cm long, gla- brous or with a few scattered patent trichomes. Inflorescences of monochasia or dichasia. Flowers 1-10, diur- nal. Sepals unequal, the outer ones 3-5 mm long, the inner ones 4-7 mm long, oblong, ovate to elliptic, coria- ceous to chartaceous, the margins scarious or not, obtuse, mucronulate, glabrous. Corollas 1. 5-2.5 cm long, campanulate, cream to white, glabrous. Capsules 5-8 mm long, more or less globose, straw-colored to brown, glabrous, the sepals partly surrounding and covering the fruit. Seeds 1-4, 3-5 mm long, straw-colored or black, pubescent with short, brown trichomes. Sonora. — Southeast and east-central part of the state in tropical deciduous forest , often in disturbed sitesand riparian habitats near rivers and along arroyos; 120-460 m. Flowering March-May and September-October. General distribution. — Florida; Baja California Sur, Chiapas, SW Chihuahua, Colima, Guerrero, Jalisco, Edo. Mexico, Michoacin, Nayarit, Oaxaca, San Luis Potosi, Sinaloa, Tamaulipas, Veracruz; Mesoamerica; Co- lombia, Venezuela, Ecuador, Peru, Guayanas, Brasil; Antilles. OPERCULINA Silva Manso, Enum. Subst. Braz. 16. 1836. [From Latin operculum, a lid or cover, and -inus, pertaining to, referring to the covering of the capsule.] Lianas or small twining herbs, the stems prostrate or climbing, to 6 m long, smooth or striate, glabrous. Leaves ovate, broadly ovate, pinnatly or palmately lobed or compound, the segments 5 or 7 or entire, glabrous; petioles and pedicels sometimes winged, mostly glabrous. Inflorescences in axillary monocasia. Flowers few or solitary, often with foliose bracts. Sepals equal or unequal, enlarging in the fruit and becoming coriaceous, sometimes irregularly dentate on the margins, glabrous. Corollas broadly campanulate, funnelform or salver- form, white, yellow, or reddish to salmon, the interplicae pilose, the plicae glabrous. Stamens included (ex- serted in O. pteripes); anthers twisted when fully mature; pollen 3-colpate. Ovary glabrous, bilocular, each locule 2-lobed; style included (exserted in O. pteripes), filiform; stigma of 2 globose lobes. Fruits dehiscent, the upper part separating by a circumscissile epicarp, the upper part more or less fleshy and separating from the lower segment and from the endocarp, 2-locular. Seeds 1-4, ovoid to ovate, glabrous or pubescent. Species about 15; 10 known from only the Old World. Selected reference. — Staples and Austin (1981). Gaujma, pata be GALLO (Friedman 1996); tansy-leaf lid-pod (USDA). Perennial herbs, the stems twining or prostrate, to 6 m long, sometimes angular, glabrous. Leaves 2-12 cm long, 2-11 cm wide, ovate in outline, pinnatisect to palmately compound, the segments 5-9, linear, lanceolate toelliptic-obovate, apically obtuse-acuminate, basally truncate or auriculate, with few trichomes on upper and lower surfaces, the margins entire. Inflorescences of monocasia. Flowers 1-3, on winged peduncles (at least in the upper part), glabrous. Sepals 1.1— 1.6 cm long, equal, straw-colored to slightly rosy during anthesis, ovate to obovate, accrescent in fruit, membranaceous, glabrous. Corollas 3.4-5.3 cm long, campanulate, white, the limb more or less entire, widening gradually, the interplicae sericeous. Capsules 1.2-2 cm long, transparent, brown, glabrous. Seeds 1-4, 5-7 mm long, ellipsoidal, black, glabrous. Sonora .— Arroyos in coastal thomscrub in the far southwestern part of the state and foothills thomscrub in the central part of the state; 10-730 m. Probably flowering May-October General distribution— Texas; Guerrero, Edo. Mexico, Michoacan, Morelos, Nuevo Leon, Oaxaca, San Luis Potosi, Sinaloa, Tamaulipas, Veracruz; Mesoamerica. Through much of its range this species is known as queibra-platos and is considered such a drastic laxative that even handling the plants will cause dishes to break at home (Alcorn 1984). In some areas it is an important medicinal plant. °perculina pteripes (G. Don) O’Donell, Lilloa 23:435, t. 6, 1950. Type: ECUADOR: Guayaquil, Ruiz & Pawn mil 1778-88 (printed label), 1800 (typed label) (holotype: MA!; isotype: F!). Cahmyction pteripes G. Don, Gen. Hist 4:264. 1838. 522 Journal of the Botanical Research Institute of Texas 6 ( 2 ) Campanula chocolate Perennial herbs or lianas, glabrous, the stems twining, pendulous, or prostrate, reaching 5 m long, becoming woody toward the base, striate to angular, glabrous. Leaves 3-17 cm long, 2-8 cm wide, simple, entire, ovate to broadly-ovate, apically acute to acuminate, attenuate, mucronate, basally cordate to almost truncate; petioles to about half as long as the blades. Inflorescences monocasial to dicasial. Rowers (1)2-12, on peduncles with 3 wings in the central part 0.6-3.5 mm wide, becoming attenuate toward both ends. Sepals more or less equal, the outer ones 2.2-2.6 mm long, ovate to ovate-elliptic, the apex acute to obutse, the inner ones 2.3-2.5 cm long, ovate, obtuse, glabrous or more often pubescent near the base, surrounding the base of the corolla. Corol- las 4-7 cm long, salverform, the limb more or less entire, broadening abruptly, reddish or red-orange to salm- on, tomentulose on the tube and the interpliae. Capsules 1.5-2 cm long, transparent, brown, glabrous. Seeds 1-4, 7-9 mm long, ovoid to ellipsoid, black, glabrous. Sonora — Foothills thornscrub, tropical deciduous forest, and oak woodland in the southeastern and east-central part of the state; 240-1250 m. Flowering July-October. General distribution. — Chiapas, southwestern Chihuahua, Colima, Distrito Federal, Edo. Mexico, Guer- rero, Jalisco, Michoacan, Morelos, Nayarit, Oaxaca, Sinaloa; Mesoamerica; Colombia, Venezuela, Ecuador, Peru. The flowers of this vine have the shape and color of bird flowers and are visited by hummingbirds. There is an incredible variation in colors of the flowers even within the limited range available. No study of pollina- tion or color variation has been made beyond casual observations. However, the various names given to it, and the many illustrations from the 1800s onward show the fascination that Europeans had with the flowers. Mpio Alamos: Arroyo el Mentidero at El Chinal road, 11.3 km S of Alamos, 26°54'45”N, 108 o 55'05"W, 240 m, 20 Sep 1993, flowers red. Van Devender 93-843!; Rio Mayo Raft trip, 2 km beyond Palmarito on bank of Rio Mayo, 27°53'N, 108°48'W, 26 Sep 1991, Rondeau & Jenkins 91- 180 (FTG-FAU!); 1.3 km S of Guirocoba Road, 3.3 km S of Alamos on road to El CKinal, Sinaloa, 26°59'30"N, 108°55'25"W, 340 m, 11 Oct ACKNOWLEDGMENTS Over the years many friends and colleagues have provided information and assistance, and/or accompanied us in the field. In this regard, we thank Thomas Bowen, Kathy Bunnell, Alberto Burquez-Montijo, Mark A. Dim- mitt, Exequiel Ezcurra, Ana Luisa Rosa Figueroa-Carranza, Lloyd Findley, Mark Fishbein, Francisco Molina- Freaner, Juan Pablo Gallo-Reynoso, Pedro Garcillan, Edward Erik Gilbert, Powell B. “Gill” Gillenwater, lllji® Henrickson, Cathy Moser Marlett, the late Paul S. Martin, Angelina Martinez-Yrizar, J. Andrew McDonald, William (Bill) Risner, the late Alexander Russell, Jean Russell, Andrew C. Sanders, Silke Schneider, Barbara Straub, Raymond Marriner Turner, Benjamin T. Wilder, and Michael F. Wilson. The staff at several herbaria have been especially helpful of our multiple requests and have provided sig- nificant assistance: Bradley (Brad) Lome Boyle, Benjamin Daniel Brandt, W. Eugene Hall, Sarah Hunkins, Philip D. Jenkins, and Michelle (Shelley) McMahon (ARIZ); Comision Nacional de Areas Naturales Protegidas (CONANP); Prescott College Kino Bay field station; Jon P. Rebman, John F. Sanborn, and especially Judy Ann Gibson who provided innumerable search results and other data (SD); Herbarium of the University of Sonora (USON); and Herbarium of Centro de Investigaciones Biologicas del Noroeste (HCIB). People at many herbaria provided generous assistance. In this regard we especial thank: Anne Barber an Elizabeth Makings (ASU), Ana Rita Simoes (BM), Ria DAversa and Debra Track (CAS), Wendy Caye Hodgson 523 and Andrew Michael Salywon (DES), Jacqueline Kallunki (NY), Layne Huiet (DUKE), Brett Jestrow (FTG), Laurent Gautie (G), Brian Franzone, Melinda Peters, Julie Shapiro, and Emily Wood (Harvard University Her- baria), Kenneth R. Robertson (ILLS), Anita F. Cholewa (MIN), James C. Solomon, Cynthia Strickland (MO); Richard Spellenberg (NMC), Tom Zanoni (NY), Ronald L. Hartman and Burrell E. Nelson (RM), Sula E. Van- derplank (RSA); George Staples (SING); J. Andrew McDonald (PAUH), Andrew S. Doran (UC); Andrew C. Sanders (UCR); Maria Teresa Buril (UFP), and Larry Hufford and Mare Nazaire (WS). Ana Lilia Reina-Guerrero translated the abstract for the resumen and Pedro Garcillan drafted the map. Two reviewers, J. Andrew McDonald and Javier Ortega provided constructive comments improving the manu- From Tom Van Devender: I thank my wife Ana Lilia Reina-Guerrero for 16 years of botanical adventures in her native Sonora. I thank Father William Trauba, Capuchin Franciscan Missionary, for sharing our field work in the Yecora area. The late Paul S. Martin inspired us and a talented group of botanists in the Rio Mayo area of southern Sonora. The following herbaria have provided Cuscuta plant material: AAU, ALTA, ARIZ, ASU, B, BAB, BOL, BRIT, CANB, CAS, CEN, CHR, CHSC, CIIDIR, CIM1, CTES, DAO, F, G, GH, H, HUFU, IAC, IEB, IND, J, JEPS, LL, LP, LPB, LPS, K, MEL, MERL, MEXU, MICH, MO, NMC, NY, OAC, OKLA, OSC, OXF, PACA, PRE, QCNE, QFA, P, PACA, RB, RSA, SAM, S, SD, SGO, SI, SPF, TEX, TRT, TRTE, UA, UB, UBC, UCR, UCT, UNB, UNM, UPRRP, UPS, US, USAS, WTU and XAL. Cuscuta research was supported by a Natural Sciences and Engineer- ing Research Council of Canada Discovery grant to Costea (327013-06 and 327013-12). Felger acknowledges support from the Wallace Research Foundation and the David and Lucile Packard Foundation. Botanical col- lections were made under Mexican Federal collecting permits including NOM-126-SEMARNAT-2000 with the generous assistance of Exequiel Ezcurra. Abel, W.E. and D.F. Austin. 1981. Introgressive hybridization between Ipomoea trichocarpa and /. lacunosa. Bull. Torrey Bot. Club 108:231-239. Aicorn, J.B. 1984. Huastec Mayan Ethnobotany. University Texas Press, Austin. Aored, K.W. 201 1 . Flora Neomexicana I: annotated checklist, 2nd edition, http://www.lulu.com Austin, D.F. 1978. The Ipomoea batatas complex-1. Taxonomy. Bull. Torrey Bot. Club 105:1 14-1 29. Ausun, D.F. 1979. Studies of the Florida Convolvulaceae— II. Merremia. Florida Scientist 42:216-222. Austin, D.F. 1986. Moth pollinated Ipomoea longifolia (Convolvulaceae). Desert PI. 8(1): 15-16. Austin, D.F. 1990a. Annotated checklist of New Mexican Convolvulaceae. Sida 14:273-286. Austin, D.F. 1990b. Comments on southwestern United States Evolvulus and Ipomoea (Convolvulaceae). Madrono 37:124-132. Austin, D.F. 1 991 . Annotated checklist of Arizona Convolvulaceae. Sida 1 4:443-457. Ausun, D.F. 1992. Rare Convolvulaceae in the southwestern United States. Ann. Missouri Bot. Gard. 79:8-16. Aus ™. D.F 1995. Merremia discoidesperma (Convolvulaceae) seeds as medicines in Mexico. Econ. Bot. 49:330-332. Austin, D.F. 1 997. Dissolution of Ipomoea ser. Anisomerae (Convolvulaceae). Bull. Torrey Bot. Club 1 24:1 40-1 59. Austin, D.F. 1998a. The indiscriminate vector: human distribution of Dichondra micrantha (Convolvulaceae). Econ. Bot. 52:88-106. Awin, D.F. 1998b. [Vascular plants of Arizona] Convolvulaceae, morning glory family. J. Arizona-Nevada Acad. Sci. 3061-83. AufflN ' D. F. 1998c. Xixicamatic or wood rose ( Merremia tuberosa , Convolvulaceae): origins and dispersal. Econ. Bot 52:412-422. *** D F - 2 °00a. Bindweed (Convolvulus arvensis, Convolvulaceae) in North America-From medicine to menace. J. lur rey Bot. Club 1 27:172-1 77. . USnN ' DF- 2 °00b. A revision of Cressa L. (Convolvulaceae). Bot. J. Linn. Soc 1 33:27-39. a ust H OF. 2000c The search for"kaladana"(/pomoea, Convolvulaceae). Econ. Bot. 54:1 14-118. u % D.F. 2001. Convolvulaceae In: D. Stevens, ed. Flora de Nicaragua vol. 85 (1), Missouri Bot Lo ^is. Pp. 653-679. al Garden Press, St. 524 Austin, D.F. 2005. The enigma of salsa da rua ( Ipomoea asarifolia, Convolvulaceae). Ethnobotany (Silver Jubilee Issue) 17(1-2):41-48. Austin, D.F. 2006 Noteworthy distributions and additions in southwestern Convolvulaceae. Canotia 2:79-106. Austin, D.F. 2007. Merremia dissecta (Convolvulaceae)— a condiment, medicine, ornamental, and weed. Econ. Bot Austin, D.F. 2012. Convolvulaceae (in part). In: Jepson Flora Project. 2012 (v. 1.0). Jepson eFIora, http://ucjeps.berkeley. edu/IJM.html (viewed 25 March 2012). Austin, D.F., R.K. Brummitt, K.R. Robertson, C. Allen, and A. Krings. In prep. Convolvulaceae. In: Flora of North American North of Mexico, vol. 14. Oxford University Press, New York. Austin, D.F., R.S. Felger, and T.R. Van Devender. 2005. Nomenclature of Ipomoea arborescens (Convolvulaceae) in Sonora, Mexico. Sida 21:1 283-1 292. Austin, D.F. and R.K. Jansson. 1 988. Range extension of Ipomoea turbinata Lag. (Convolvulaceae) to southern Florida. Florida Scientist 51(3-4):182, 183. Austin, D.F., K. Kitajima, Y. Yoneda, and L. Qian. 2001 . A putative tropical American plant, Ipomoea nil (Convolvulaceae), in Pre-Columbian Japanese art. Econ. Bot. 55:515-527. Austin, D.F., G.W. Staples, and M. Carine. 2008 onward. Convolvulaceae pollen atlas, http://ag.arizona.edu/herbarium/as- soc/projects/convolv/Convolvulaceae_Pollen_Atlas.htm (viewed 8 November 201 1). Beliz, T. 2001. Cuscutaceae. In: W.D. Stevens, C. Ulloa Ulloa, A. Pool, and O.M. Montriel. Flora de Nicaragua, introduced, Gimnospermas y Angiospermas.Tomo I. Missouri Botanical Garden Press, St. Louis. Pp. 719-720. Bohac, J.R., D.F. Austin, and A. Jones. 1 993. Discovery of wild tetraploid sweetpotatoes. Econ. Bot. 47:1 93-201 . Bohac, J.R., P.D. Dukes, and D.F. Austin. 1 995. Sweet potato. In: J. Smartt and N.W. Simmonds, eds. Evolution of crop plants, ed. 2. Longman Group, Ltd, London. Pp. 57-62. Brito-Castillo, L, M.A. Crimmins, and S.C. DIaz C. 2010. Clima. In: F. Molina-Freaner and T.R. Van Devender, eds. Diversidad biologica del estado de Sonora. Universidad Nacional Autdnoma de Mexico, Mexico D.F. Pp. 73-96. Brown, D.E. (ed.). 1982. Biotic communities of the American Southwest - United States and Mexico. Desert PI. 4:3-341. Bcirquez, A., and A. Martinez-YrIzar, A. 2007. Conservation and landscape transformation in northwestern Mexico. In: RS. Felger and B. Broyles, eds. Dry borders: great natural reserves of the Sonoran desert. University of Utah Press, Salt Lake City. Pp. 537-547. Carranza, E. 2008. Convolvulaceae II. Flora del Bajio y de regiones adyacentes. Fasciculo 155, Instituto de Ecologia AC, Centro Regional del Bajio, Patzcuaro, Michoadin, Mexico. Castellanos-Villegas, A.E., LC. Bravo, G.W. Koch, Lllano, D. Lopez, R. Mendez. J.C. Rodriguez, R. Romo, T.D. Sisk, and G. Yanes- In: F. Molina-Freaner and T.R. Van Devender, eds. Diversidad biologica del estado de Sonora. Universidad Nacional Autonoma de Mexico, Mexico D.F. Pp. 157-186. Costea, M., D.F. Austin, R.S. Felger, T.R. Van Devender, and J J. Sanchez-Escalante. 201 2b. Convolvulaceae (morning glories) of Sonora. University of Arizona Herbarium (http://ag.arizona.edu/herbarium/node/80) and Wilfrid Laurier Herbarium (httpV/www.wlu.ca/page.php?grp_id=2147&p=2061 2). Costea, M., R.S. Felger, D.F., Austin, T.R.Van Devender, and J J. SAnchez-Escalante. 201 2a. Convolvulaceae of Sonora, Mexico. ft Cuscuta. J. Bot. Res. Inst. Texas 6:527-548. Costea, M. and G.L Nesom. In prep. Cuscuta (Convolvulaceae). In: Flora of North American North of Mexico, vol. 14. Oxford University Press, New York. Costea, M. and FJ.Tardif. 2006. The biology of Canadian weeds. Cuscuta campestris, C. gronovii, C. umbrosa, C. epithymw n and C. epilinum. Can. J. Plant Sci. 86:293-31 6. Costea, M. and S. Stefanowc. 201 2. Cuscuta. In: B.G. Baldwin, D.H. Goldman, DJ. Keil, R. Patterson, and TJ. Rosatti, eds.The Jepson manual, vascular plants of California, 2nd Ed. University of California Press, Berkeley. Pp. 659-662. DeTheis, A. 1810. Glossaire de botanique ou Dictionnaire etymologique de tous les noms et termes relatifs a cette sci- ence. G. Dufour, Paris. Elmore, C.D. 1 986. Mode of reproduction and inheritance of leaf shape in Ipomoea hederacea. Weed Sci. 34:391-395. Felger, R.S. 1999. The flora of Cafion del Nacapule: a desert-bounded tropical canyon near Guaymas, Sonora, Mexi® Proc. San Diego Soc. Nat. Hist. 35:1-42. Felger, R.S. 2000. Flora of the Gran Desierto and Rio Colorado of northwestern Mexico. University of Arizona Pre» 525 Felger, R.S., and B. Broyles (eds.). 2007. Dry borders: great natural reserves of the Sonoran Desert. University of Utah Press, Salt Lake City. Felger, R.S., M.B. Johnson, and M.F. Wilson. 2001 . Trees of Sonora, Mexico. Oxford University Press, New York. Felger, R.S. and C.H. Lowe. 1 976. The Island and coastal vegetation and flora of the Gulf of California, Mexico. Contrib. Sci. Nat. Hist. Mus. Los Angeles County, Contr. Sci. 285:1-59. Felger, R.S. and M.B. Moser. 1985. People of the desert and sea: ethnobotany of the Seri Indians. University of Arizona Press, Tucson. Felger R.S. and B.T. Wilder with H. Romero-Morales. (forthcoming, 201 2). Plant life of a desert archipelago: flora of the So- noran Islands in the Gulf of California. University of Arizona Press, Tucson. Felger, R.S., B.T. Wilder, and J.P. Gallo-Reynoso. 2011. Floristic diversity and long-term vegetation dynamics of Isla San Pedro Nolasco, Gulf of California, Mexico. Proc. San Diego Soc. Nat. Hist. 43:1 -42. Ferguson, I. K., B. Verdcourt, and M. M. Poole. 1977. Pollen morphology in the genera Merremia and Operculina (Convolvulaceae) and its taxonomic significance. Kew Bull. 31:763-773, pi. 30-33. Friedman, S.L. 1996. Vegetation and flora of the Coastal Plains of the Rfo Mayo Region, Southern Sonora, Mexico. MS thesis, Arizona State University, Tempe. Gentry, H.S. 1942. Rio Mayo plants: a study of the flora and vegetation of the valley of the Rio Mayo, Sonora. Carnegia Institution of Washington publ. 527. Gentry, H.S. 1 963. The Warihio Indians of Sonora-Chihuahua: an ethnographic survey. Anthr. Pap. 35. Bureau Amer. Eth- nol. Bull. 1 86. Smithsonian Institution. Government Printing Office, Washington, D.C. Gunn, C.R. 1969a. Seeds of the United States noxious and common weeds in the Convolvulaceae, excluding the genus Cuscuta. Proc. Assoc. Official Seed Analysts North America 59:101-1 15. Gunn, C.R. 1 969b. History and taxonomy of the purple moonflower, Ipomoea tubinata Lagasca y Segura. Proceedings of the Association of Official Seed Analysts of North America 59:1 1 6-1 23. Gunn,C.R. 1972. Moonflowers, Ipomoea section Calonyction, in temperate North America. Brittonia 24:150-168. Gunn, C.R. 1 977. Merremia discoidesperma (Donn.-Sm.) O'Donell: its taxonomy and capacity of its seeds to drift in ocean currents. Econ. Bot 31:237-252. Hammel, B.E. 2010. Convolvulaceae, pp. 72-126. In: B.E. Hammel, M.H. Grayum, C. Herrera, and N. Zamora, eds. Dico- tiledoneas (Clusiaceae-Gunneraceae). Manual de plantas de Costa Rica, vol. V. Monog. Syst. Bot. Missou 119:1-970. House, H.D. 1908. The North American species of the genus Ipomoea. Ann. New York Acad. Sci. 18:181-263. Johnston, M.C. 1 963. The geography of the five Texas species of Dichondra (Convolvulaceae). Wrightia 1 :252-253. Madrean Archipelago Biodiversity Assessment (MABA). 2012 http://www.madrean.org/rn ' 201 2 and earlier). Manitz, H. 1 983. Zur nomenklatur einiger Convolvulaceae und Cuscutaceae. I. Feddes Repert. 94:1 73-182. Martin, P.S., D. Yetman, M. Fishbein, P. Jenkins, T.R. Van Devender, and R.K. Wilson. 1998. Gentry's Rio Mayo plants. University of Arizona Press, Tucson. Martinez-YrIzar, A., R.S. Felger, and A. BOrquez. 2010. Los ecosistemas terrestres de Sonora: un diverso capital natural. In: F. Molina-Freaner and T.R. Van Devender, eds. Diversidad biologica del estado de Sonora. Universidad Nacional Autbnoma de Mexico, Mexico D.F. Pp. 129-156. Mason, C.T., R.K. Van Devender, and G.D. Starr. 1 986. Notes on the flora of Arizona VIII. Desert PI. 8:38-40. McClaran, M.P. and T.R. Van Devender (eds.). 1 995. The desert grassland. University of Arizona Press, Tucson. McDonald, JA 1 982. Biosystematics of the Ipomoea tricolor complex (Convolvulaceae), Ph.D. dissertation, University of Texas, Austin. JA 1991. Origin ami diversity of Mexican Convolvulaceae. An. Inst Biol. Univ. Mac. Anton. Mfe. SArie Bo- tunica 62:6-82. McDonald, J a 1 993a. A new species of Ipomoea (Convolvulaceae) from Costa Rica and notes on the circumscription of section Calonyction (Choisy) Griseb. Harvard Pap. Bot. 4:53-56. McDonald, JA. 1 993b. Convolvulaceae I. Flora de Veracruz 73:1-99. McDonald, J.A. 1 994. Convolvulaceae II. Flora de Veracruz 77:1-133. McDonald, J.A. 1 995. Revision of Ipomoea section Leptocaulis (Convolvulaceae). Harvard Pap. Bot. 6:97-1 22. ).Lundellia 4:76-93. J A, DA Hansen, J~R. Md^xun B.B.'srMesoc Ml L A phylogenetic assessment of breeding systems and floral morphology of North American Ipomoea (Convolvulaceae). J. Bot. Res. Inst. Texas 5:1 59-1 77. ri Bot. Gard. i/ (viewed 20 April . Diversidad biologica del < O'Donell, C.A. 1941. Revision de las especis americanas de Merremia. Lilloa 6:467-554. O'Donell, C.A. 1957. Convolvulaceas Chilenas. Bol. Soc. Argent. Bot. 6(3-4):143-184. O'Donell, C.A. 1959a. Convolvulaceas argentinas. Lilloa 29:87-343, pi. I-V. O'Donell, C.A. 1 959b. Las especies Americanas de Ipomoea L. sect. Quamoclit (Moench) Griseb. Lilloa 29:1 9-86. OED onune. 201 2. Oxford English dictionary. Oxford University Press, Oxford, http://www.oed.com. Oviedo y Valdes, C.F. de. 1 526 [1 969]. De la natural hystoria de las Indias. University of North Carolina Press, Chapel Hill. Paredes Aguilar, R, T.R. Van Devender, and R.S. Felger, with G.P. Nabhan and A.L Reina Guerrero. 2000. Las Cactciceas de Sonora: su diversidad, uso y conservation. Arizona-Sonora Desert Museum Pennington, C.W. 1969. The Tepehuan of Chihuahua, their material cult Robertson, K.R. 1971. A revision of the genus Jacquemontia (Convolvuli Indies. Ph.D. dissertation, Washington University, St. Louis. Roxburgh, W. [1824] 2006. Roxburgh's flora Indica. http://apps.kew.org/floraindica/hi Rzedowski, J. 1978. Vegetacion de Mexico. Editorial Limusa, Mexico, D.F. Sa'ad, F. 1967. The Convolvulus species of the Canary Islands, the Mediterranean Region .do (viewed 29 Dec 201 1) Meded. Bot. Mus. Herb. Rijks Univ. Utrecht 281 :1 -288. Shreve, F. 1951. Vegetation of the Sonoran Desert. Carnegie Institution of Washington Publication 591:1-192. Southwest Environmental Information Network (SEInet). 2012. http://swbiodiversity.org/seinet/index.php (viewed 20 March 2012 and earlier). Standley, P.C. 1920-1926. Convolvulaceae. In: Trees and shrubs of Mexico, vol. 2. Contr. U.S. Natl. Herb. 23:1 194-1208. Standley, P.C. and LO. Wiluams. 1 970. Convolvulaceae. Flora of Guatemala. Fieldiana Bot. 24(9):4-85. Staples, G.W. (ed). 201 1 . Convolvulaceae unlimited, http://convolvulaceae.myspecies.info/ (viewed 7 October 2011). Staples, G.W. and D.F. Austin. 1 981 . Changes in West Indian Operculina (Convolvulaceae). Brittonia 33:591 -596. Staples, G.W., D.F. Austin, and J A McDonald. 2006. (1 724) Proposal to reject Ipomoea glaucifolia L. (Convolvulaceae). Taxon 55:535-536. Staples, G.W., D.F. Austin, and R. Simao-Bianchini. 2012. Disposition of the names published by A. Peter in Convolvulaceae. Taxon 61:671-679. Staples, G.W., J.H. Wiersema, N.A. Chambers, and D.F. Austin. 2005. The restoration of Ipomoea muricata (L.) Jacq. (Convolvulaceae). Taxon 54:1 075-1 079. Steinmann, V.W. and R.S. Felger. 1997. The Euphorbiaceae of Sonora, Mexico. Aliso 16:1-71. St. John, H. 1970. Classification and distribution of the Ipomoea pes-caprae group (Convolvulaceae). Bot. Jahrb. Syst 89:563-583. Stoleson, S.H., R.S. Felger, G. Ceballos, C. Raish, M.F. Wilson, and A. BOrquez. 2005. Recent history of natural resource use and population growth in northern Mexico. In: J-.LE. Cartron, G. Ceballos, and R.S. Felger, eds. Biodiversity, ecosystems and conservation in northern Mexico. Oxford University Press, New York. Pp. 52-86. Tharp, B.C. and M.C. Johnston. 1961. Recharacterization of Dichondra (Convolvulaceae) and a revision of the North Ameri- can species. Brittonia 13:346-360. Thiers, B. 201 1 . NYBG.org: Index Herbariorum: a global directory of public herbaria and associated staff. sciweb.nybd- org/science2/lndexHerbariorum.asp (viewed 30 October 201 1 ). Turner, R.M., J.E. Bowers, andT.L. Burgess. 1995. Sonoran Desert plants, an ecological atlas. University of Arizona Press. (an Devender, T.R. 1 995. Desert grassland history: changing climates, evolution, biogeography, and community dynam- ics. In: M.P. McClaran and T.R. Van Devender, eds. The desert grassland. University of Arizona Press, Tucson. Pp. 68-9* 'an Devender, T.R., R.S. Felger, F. Mouna-Freaner, M. Fishbein, J J. Sanchez-Escalante, and A.L Reina-Guerrero. 2010. Biodiversidad de plantas vasculares. In: F. Molina-Freaner and T.R. Van Devender, eds. Diversidad biologica del estado de Sonora. Universidad Nacional Autbnoma de Mexico, Mexico D.F. Pp. 229-262. an Ooststroom, SJ. 1934. A monograph of the genus Evolvulus. Meded. Bot. Mus. Herb. Rijks Univ. Utrecht. 14:1-267. 'erdcourt, B. 1 961 . Notes from the East African Herbarium: XII. Kew Bull. 1 5:1-18. Yard, D.B. 1968. Contributions to the flora of Florida-3. Evolvulus. Castanea 33:76-79. Amer.J. Bot. 83:1050-1056. Felgeretal., Convolvulaceae (excluding Cuscuta) of Sonora, Mexico Yatskievych, G. and C. Mason. 1984. A taxonomic study of Ipomoea tenuiloba Torrey (Convolvulaceae), with notes lated species. Madrono 31:102-108. Yetman, D. and R.S. Felger. 2002. Ethnoflora of the Guarijios. In: D. Yetman, The Guarijfos of the Sierra Madre, people of northwestern Mexico. University of New Mexico Press. Albuquerque. Pp. 1 74-230. Yetman, D. andT.R. Van Devender. 2002. Mayo ethnobotany: land, history, and traditional knowledge in northwest University of California Press, Berkeley. 528 BOOK REVIEW Emily Monosson. 2012. Evolution in a Toxic World: How Life Responds to Chemical Threats. (ISBN-13: 978- 1-59726-976-6, hbk.). Island Press, 1718 Connecticut Ave., NW, Suite 300, Washington, DC 20009, U.S.A. (Orders: http://islandpress.org/index.html). $35.00 hbk., $29.95 pbk., 240 pp., 10 figures, notes, “The best way to envisage the situation is as follows: the environment presents challenges to living species, to which the latter may respond by adaptive genetic changes.” — Theodosius Dobzhansky. . . . And this is exactly what Emily Monosson discusses throughout the ten chapters of Evolution in a Toxic World: How Life Responds to Chemiccd Threats. This 222-page book is a thought-provoking summary of an im- portant but often ignored subject matter, environmental toxicology. The introduction (Chapter 1) is a fascinat- ing overview of the subject matter. Monosson opens her introduction with two powerful sentences: “All of life is chemical. But not all chemicals are compatible with life.” Think about that contrast! The author gives an overview of the four sections of the book: 1) Nature’s Toxicants (everything is a poison in the right dose), 2) Evolutionary History of Toxicology, 3) Toxic Evolution in Action, and 4) Looking Forward by Looking Back. Chapter 2 — Shining a Light on Earth’s Oldest Toxic Threat, looks at UVR, a highly energetic and destruc- tive force to be reckoned with. We all know metals like arsenic and the secondary plant metabolite strychnine are poisonous, but Monosson throws at the reader chemical toxicants that most of us don’t think about, like oxygen. Monosoon points out, “we cannot live without it, yet every day we struggle to coexist with this highly reactive and potentially toxic chemical.” Oxygen is discussed in detail in Chapter 3 — When Life Gives you Oxygen, Respire. Chapter 5 looks at the many toxic metals that life deals with. Chapter 6 discusses chemical warfare ... the battle to protect and to survive. The combatants? Plants and animals. Sounds like a sci-fi movie, and it very well could be. Maybe people would watch and then take notice of the thousands of chemicals in our environment. If you like war, then Chapter 6 is a captivating account of this historical battle and challenge to stay alive; it’s my favorite chapter in the book. This is not to say that the rest of the book is not interesting. Quite the contrary. Chapter 7 — Sensing Chemicals is enticing, Chapter 8 — Coordinated Defense is inviting, Chap- ter 9— Toxic Evolution is captivating, and Chapter 10 is all about toxic overload: “How will life’s toxic defense mechanisms respond to industrial age chemicals?” And yes, there are many. “In 2009, the Chemical Abstracts Service, which catalogs and tracks all known chemicals, announced the registration of its fifty-millionth ‘novel’ chemical — the last ten million chemicals having been registered over the preceding nine months. There are plenty more novel chemicals to be found.” In the conclusion, Monosson writes that this toxic world we live in is challenging to each and every human. We do live in a sea of toxic chemicals. “Life on Earth is now subject to a virtual onslaught of chemicals associated in one way or another with human activity. We are a society built on chemicals, and there is no turning back ” Monosson says that we must strive to better understand how chemicals affect wildlife and human health. “We have to do so. There is no higher ground, no corner on earth where life can escape the influence of toxic chemicals. The choice must not be to ‘evolve or die.’”— Barney Lipscomb, Botanical Research Institute of Texas, 1700 University Dr., Fort Worth, Texas 76107-3400, USA ). Bot. Res. Inst. Texas 6(2): 528. 2012 CONVOLVULACEAE OF SONORA, MEXICO. II: CUSCUTA Mihai Costea Richard S. Felger Department of Biology Wilfrid Laurier University 75 University Avenue W Waterloo, ON, N2L 3C5, CANADA Daniel F. Austin Arizona-Sonora Desert Museum 2021 N. Kinney Road, Tucson, Arizona 85743, U.SA. and Herbarium, University of Arizona P.O. Box 21 0036, Tucson, Arizona 8572 1, U.SA. Herbarium, University of Arizona P.O. Box 210036, Tucson, Arizona 85721, U.S.. , and Sky Island Alliance, P.O. Box41165 Tucson, Arizona 85717, U.S.A. Thomas R. Van Devender Sky Island Alliance, P.O. Box 41 165 Tucson, Arizona 85717 and Herbarium University of Arizona P.O. Box 21 0036, Tucson, Arizona 85721, U.SA VanDevender@skyislandalliance.org J. Jesus Sanchez-Escalante Dept, de Investigaciones Cientificas y Tecnoldgicas Rosales y Ninos Heroes, Centro Hermosillo, Son, 83000, MEXICO RESUMEN INTRODUCTION ^ is the second part of a comprehensive study regarding the Convolvulaceae of Sonora (Felger et aL 2012). ® the firs t Part, we included a general introduction, information about the growth forms, cases of endemism/ f s P«te, species diversity, and types of vegetation for all the genera/species. The first part also provided Wentihcation keys, and extensive taxonomic and floristic data for all the Convolvulaceae genera except Cus- 530 Journal of the Botanical Research Institute of Texas 6(2) cuta (dodder). Since the diversity of dodder species and their association with certain vegetation types were presented together with the other Convolvulaceae (Felger et al. 2012), this article concludes the monographic study of this botanical family in Sonora with a taxonomic and floristic account of the genus Cuscuta. Cuscuta is nearly cosmopolitan and comprises over 200 species (Costea 2007-onwards). Although some dodders are agricultural pests (Costea & Tardif 2006), more numerous Cuscuta species require conservation measures (Costea & Stefanovic 2009). Roughly 75% of species are native to the Americas and about 30% to Mexico and southern USA (Stefanovic et al. 2007). We have documented 21 species of Cuscuta in the flora of Sonora, a number that is higher or comparable to that of the neighboring areas to the north and west (Arizona, the two Baja California states, California, and New Mexico; see table 1, Felger et al. 2012). Interestingly, while the species diversity of the other Convolvulaceae genera increases towards southern Mexico and Central America, dodder species richness declines sharply in these geographical areas (Felger et al. 2012). The 21 spe- cies that occur in Sonora belong to nine of the 15 clades of subgenus Grammica, the most complex infrageneric taxon of Cuscuta (Stefanovic et al. 2007). The best represented in Sonora is the C. umbellata clade (clade “L”) with seven of the nine North American species growing in Sonora (C. desmouliniana, C. legitima, C. leptantha, C. odontolepis, C. polyanthemos, C. tuberculata, and C. umbellata). This strongly suggests that Sonora is part of the genetic center of origin for this clade, which has a complicated evolutionary history shaped by reticulate evolution (Costea & Stefanovic 2010). The other eight clades of subg. Grammica are represented in Sonora by one species (C. salina, clade “A”; C. tinctoria, clade “G”; C. corymbosa var. grandiflora, clade “J”, C. indecora, clade “M”; C. campestris, clade “B”; C. vandevenderi, clade “N”), two species (C. azteca and C. chinensis var. applanata, Clade “H”; C. americana and C. macrocephala, clade “D, or three species (C. boldinghii, C. costaricensis, andC. erosa, clade “K”). CUSCUTA L.,Sp. PI. 124. 1753. [Based on the Aramaic and Hebrew triradical root of the verb K-S-Y PH Kaph, ® Shin, ’ Yodh), XT. Common names.— Dodder; fideo Herbaceous vines. Stems filiform, yellow or orange, trailing or dextrorsely twining and attached to the host by numerous small haustoria, glabrous. Leaves reduced to minute, alternate scales. Inflorescences monochazial clusters further grouped in cymose inflorescences that are often confluent. Flowers 4-5-merous, small, always ± fleshy when fresh, thick or membranous-thin when dry, white, white-cream, sometimes yellowish or red- dish. Conic-cylindrical papillae present or absent on the pedicels, perianth and ovary/capsule; laticifers visible or not in the calyx, corolla, ovary/capsules, translucent, white, yellow or orange, isolated or arranged in rows especially in the midveins of the calyx and corolla lobes, round, ovoid or elongated. Calyx gamosepalous; lobes basally overlapping or not, sometimes with multicellular projections (C. chinensis var. applanata, C. boldinghii)- Corolla gamopetalous with lobes imbricate in bud, sometimes with a subapical cusp or horn-like multicellular appendage (C. boldinghii, C. costaricensis, C. erosa). Stamens alternating with the corolla lobes. Pollen 3-colpate (sometimes 4- or 5-colpate in the same anther), tectum imperforate to reticulate. Infrastaminal scales com- monly present, scale-like appendages dentate or fringed, bridged and adnate to the corolla tube base, forming a corona alternating with the corolla lobes. Ovary superior, 2-locular, each locule with 2 anatropous ovules. Styles 2, terminal, distinct or united, equal or unequal. Stigmas spherical to linear (only distinct unequal styles with spherical stigma in the species from Sonora). Fruits capsules, indehiscent (sometimes opening irregu- larly between the styles) or circumscissile by a ± regular line near the base. Seeds 1-4 per capsule, 3-angled or dorsoventrally compressed; endosperm nuclear; embryo uniformly slender, 1-3-coiled, without cotyledons, consisting mostly from the hypocotyl; seed coat alveolate when dry and papillate when hydrated (rarely not alveolate/papillate with cells ± rectangular, puzzle-like arranged). Cuscuta campestris and C. indecora are weeds, the former subcosmopolitan, the latter widespread in North and South America. Both species have been likely introduced to Sonora with contaminated legume seeds (e g - alfalfa; Costea & Tardif 2006). Another species, C. tinctoria, may also have been introduced to Sonora on the cultivated tree Schinus terebinthifolia. Although none of the Cuscuta spp. are endemic to Sonora, many occur only here and in adjacent geographical areas (Table 1, Felger et al. 2012). The diversity of Cuscuta species in Sonora is most similar to that of the Baja California Peninsula and Arizona (12 and 11, respectively common species), followed by New Mexico (9 common species), while only three of the Sonoran species occur in California (Table 1, Felger et al. 2012). Selected references . — Stefanovid et al. (2007), Welsh et al. (2010), Wright et al. (2011, 2012), Yuncker (1932, 1965). Identification of most Cuscuta spp. is a lengthy process because rehydration of flowers, dissection, and examination under a microscope are usually necessary. Measurements of floral parts were done on rehydrated herbarium material. Length of flowers was measured from the base of calyx to the tip of straightened corolla lobes. The texture of flowers and the color of calyx were noted on dry herbarium material. Observation of pa- pillae and laticifers requires magnifications of at least 100 x. Examination of seed surface requires magnifica- tions of at least 150 x. In describing the stem, the following categories based on stem diameter were used (Yuncker 1921): “slender” with the diameter of 0.35-0.4 mm, “medium” with the diameter of 0.4-0.6 mm, and “coarse” when diameter is greater than 0.6 mm. The geographical distribution, both in Sonora and Mexico, is The host range is also based on herbarium labels as well as observations by Richard Felger; hosts ob- served in other geographical areas are included when they are present in Sonora. SEM images of the flowers for some species are provided to help identification. The vouchers are indicated (“SEM”) in the lists of typical collections examined. Pictures were taken with the scanning electron micro- scopes Hitachi S-570 and LEO 1530 FE-SEM at 15 KV. Samples were coated with 30 nm gold using an Emitech K 550 sputter coater. Numerous images for all the species, including the types, with details of dissected flowers are available from Digital Atlas of Cuscuta (Costea 2007-onwards). All specimens cited have been seen by Costea and are at ARIZ unless otherwise indicated. For citation of herbarium specimens see Felger et al. (2012). Plants not native to flora area are marked with an asterisk (*). 532 -0.5 x 0.2-0.3 1 . nil 537 desert grassland, 31°11'23"N, 110°36’22"W, 1340 m, on Ambrosia confertiflora, 17 Aug 2001, Von Devender 2001-710 (ARIZ, USON, WLU) [SEMI. Cuscuta corymbosa Ruiz & Pav. var. grandiflora Engelm., Trans. Acad. Sci. St. Louis 1:483. 1859; Syst. Ar- rang. Sp. Cuscuta 33. 1859. (Engelmann published the same name in both publications, the same year). Type: COLOMBIA, [from Popaydn?]: Humboldt 2002, ex herb Witldenow 3157 (lectoiyfe here designated: B!; fragment at MO!). En- . Sulphur 35. 1844. Type MEXICO. Baja California Sur: Magdalena Bay, Bentham s.n. (K!). ay. Sulphur 138. 1844. Type: MEXICO. Guerrero: Acapulco, Bentham s.n. (K!). Stems medium, orange. Inflorescences dense, corymbiform or umbellate, often confluent. Pedicels 2-6 mm long. Bracts 1 at the base of clusters, absent at the base of pedicels, 0.5-1 mm long, ovate lanceolate, margins entire, apex acute. Flowers 5-merous (Fig. le), 4.5-6.5(-7) mm long, membranous, white when fresh, creamy brownish when dried, papillae absent. Laticifers barely visible in the calyx and corolla lobes, isolated, ovoid to elongate. Calyx 2-2.5 mm long, straw-yellow to brownish, membranous, finely reticulate, not shiny, cylindric campanulate, V2- 3 /4 as long as the corolla tube, divided ca. l A its length; tube 1.1-2 mm long; lobes 0.4-0.75 mm tong, not overlapping or only slightly so, rounded, not carinate, margins entire. Corolla 4-6 mm long; tube 3-5 nun long, cylindric, becoming dilated in the lower W; lobes 1-1.5 mm long, initially erect, later spreading, VS-Va of the corolla length, ovate, mar gins entire, apex obtuse to rounded, ± incurved. Stamens included, shorter than the corolla lobes; filaments 0.1-0.3 mm long; anthers 0.5-0.6 x 0.3-0.4 mm, subround to elliptic. Infra- staminal scales 1.5-2 mm long, Vi-% as long as the corolla tube, bridged at 0.2-0.35 mm, forming oblong ridges with fringed margins, fimbriae 0.4-0.15 mm long. Styles 2.4-4.2 mm long, much longer than the ovary, evenly filiform. Capsules circumscissile, 2-2.9 x 2.2-2.6 mm, globose to slightly depressed, not thickened or risen around the small interstylar aperture, translucent, surrounded by the withered corolla. Seeds 2-4 per capsule, 1.1— 1.3 x 0.7-0.9 mm, slightly angled, broadly elliptic, sometimes with a longitudinal groove on the ventral face; seed coat cells alveolate/papillate. Sonora.— Locally in Gulf Coast of the Sonoran Desert in a large, ecologically pristine canyon in the Guay- mas region opposite Isla San Pedro Nolasco and in foothills thornscrub in east-central Sonora; also on Islas San Pedro Nolasco and San Esteban; ca. 5-450 m. Flowering December-April. Probably more widespread in the state. Parasitic especially on Colubrina viridis and Vaseyanthus insularis (Felger et al. 2011); alsc urtfornicum, Acalypha califomica, Cottsia linearis (Janusia linearis ), n ka, and Poaceae. General distribution.— One of the most common species in Mexico, where it is sometimes weedy. Baja California Sur, Chiapas, Colima, Durango, Edo. Mexico, Guerrero, Guanajuato, Hidalgo, Jalisco, Michoacan, Morelos, Nayarit, Sinaloa, Tamaulipas, Veracruz; also in Central and South America. Selected references — Felger and Wilder (2011, 2012), Yuncker (1932). M P» Guaymas: Caft6n las Barajitas Sierra el Aguaje, 28°03'27"N, lll o 09’27"W, 90 m, 19 Feb 1995, Felger 95-208. Mpio Soyopa: 0.6 km N 16 on road to T6nichi just E of Rfo Yaqui, 28.5758°N, 109.5505°W, 270 m, foothills thornscrub, on Janusia linearis, 7 Jan 2001. Van I96« ,der2001 ' 16 SONORAN ELANDS. San Esteban: N side of island, o S ?' Fdger 1755 °- St eep N slope of NE peak, 28°42'N, 112 0 35'W, 450 m, o ® Nolasco: NE side of Island, 18 Jan 1965, Felger 12082 [S 5, Felger 12082; l' i. TorreyBot. Club 18:227. 1932. i e: MEXICO. Durango: Santiago Papasqui; Journal of the Botanical Research Institute of Texas 6(2) The type locality is in pine-oak forest and the host plant is Bouvardia temifolia. In addition to the type collec- tion, C. dentatasquamata is recorded from a canyon in southern Arizona mountains in oak woodland and should be sought elsewhere with Bouvardia in southern Arizona and northern Sonora mountains. General distribution. — Arizona; Chihuahua, and Sonora; apparently very rare. Selected reference. — Yuncker (1932). S Biol. Monogr. 6(2-3):40-41. 1931. Type: MEXICO. SONORA: hills near Altar, UZ!, ASU1, Gl, GH!, IND, MEXU!, MO!, NY!, US!). iol. Monogr. 6(2-3):41. 1921. Regarding Pringle 105, Yuncker (1921, p 41) raen- Following two distinguishable varieties (“typical” and “attenuiloba”). rr. 6(2-3):43. 1921. Type: MEXICO. Sonora: Guaymas, 22 Feb 1904, Palmer 1209 Stems slender, yellow-orange. Inflorescences loose, umbellate, often confluent. Pedicels 1-5 mm long. Bracts 1 at the base of clusters and 0-1 at the base of pedicels, 0.6-1 mm long, ovate-lanceolate, margins entire, apex acute. Flowers 5-merous (Fig. 3d), 2-3 mm long, membranous, white when fresh, creamy-white when dried. Papillae usually present on the pedicels, calyx, abaxial and adaxial epidermis of corolla lobes, and sometimes on the ovary/capsule. Laticifers not visible or hardly so in the midveins of corolla lobes, elongate. Calyx 0.6-1.2 mm long, brownish-yellow, ± reticulate or shiny, campanulate, equaling or somewhat longer than the corolla tube, divided VS-Vi the length; tube 0.25-0.5 mm long; lobes 0.5-0.76 mm long, not overlapping, triangular- ovate to lanceolate, weakly to distinctly carinate, with small dome-like multicellular projections on the mid- veins, margins irregular, ± revolute at the base and forming angled sinuses (especially when lobes are triangu- lar ovate), apex acute to acuminate. Corolla 1.5-2.9 mm long; tube narrow-campanulate, 0.8-1.5 mm long lobes 1-1.5 mm long, initially erect, later spreading or reflexed, slightly longer than the tube, lanceolate, mar- gins entire sometimes involute upon drying and appearing very narrow, apex acute, ± incurved. Stamens short-exserted, shorter than corolla lobes; filaments 0.4-0.7 mm long; anthers 0.4-0.6 x 0.2-0.3 mm, ovate to oblong. Infrastaminal scales 0.6-1 mm long, ca. % of the corolla tube, bridged at 0.1-0.2 mm, oblong to spath- ulate, short-fringed, fimbriae 0.05-0.15 mm long. Styles 1.2-2.1 mm long, longer than the ovary, evenly fili- form. Capsules circumscissile, 1.5-2 x 0.9-1.7 mm, globose, to globose-depressed, slightly thickened and risen, or with a few protuberances around the inconspicuous interstylar aperture, translucent, capped by the withered corolla (Fig. 3e). Seeds 2-4 per capsule, 0.75-0.9 x 0.7-0.8 mm, angled, subrotund to broadly elliptic; seed coat cells alveolate/papillate. Sonora. — Common in the Sonoran Desert, especially on sandy flats, valley floors, and bajadas, and in coastal and foothills thomscrub; near sea level-300 m. Recorded on herbaceous hosts including Amaranthus watsonii, Boerhavia (including B. coccinea, B. triquetra). Euphorbia subgenus Chamaesyce (especially E. poly- carpa), Pectis (including P. coulteri , P. papposa), and occasionally on Tumamoca macdougalii. General distribution. — Baja California (norte) and Sur, Sinaloa. Selected reference. — Costea and Stefanovic (2010). Mpio Caborca: 35.2 km W of Caborca on road to Desemboque, desertscrub on sandy flats with Lama and Olneya, 30°44'35"N, 112°26'32’W, 63 m, 16 Jan 2002, Van Devender 2002-23 (WLU). Mpio Guaymas: 6 mi NW of Guaymas, 28 Feb 1933, Shreve 6134; Old road to Algodone. San Carlos, 27°57'42"N, 111«03’43"W, ca. 35 m, 26 Dec 2000, Reina-G. 2000-917 (CAS, US, WLU) Mpio Hermosillo: 5 mi by road E of Bahia Kino, 19 Oct 1963, Felger 9046. Mpio Huatabampo: 6.25 km E of Camahuiroa, 1.9 km W of Tierra y Libertad, 26°33'00"N, 109°12’45’W, 23 tn 20 |.,n f 2H-*.s Mpio Pitiquilo: at coast on N side of headland ca 10 ,h „f PcsomhocpK |sm hyuM-l 22 Man Spellenberg 4943 (NMC). Mpio Soyopa: Arroyo Los Garambullos, 1.5 km E of T6nichi, 28°34'10"N, 109°33'00"W, 180 m, 15 Sep 1998> V* Devender 98-1 120 (ARIZ, WLU); 4 Sepl996, Von Devender 96-360 (ARIZ, NMC, WLU) [SEM]. SONORAN ISLANDS. DAtik NW side of land, 20 Dec 1966, Felger 15313A. Tiburon: SE side of Agua Dulce Valley, ca. 12 mi S from Tecomate, 28°57'20"N, 112°24.5 , W, ca. 280®^ 26 Sep 2008, Felger 08-120. poy rpa, qu Cuscuta erosa Yunck., Illinois Biol. Monogr. 6(2-3):26. 1921. Type: MEXICO. Sonora: 1869, Palmer s.n. (holotype: US!) Cap- 543 ment bases, bridged at 0.5-0.7 mm, obovate to spathulate, rounded, uniformly dense, long fringed, fimbriae 0.2-0.4 mm long. Styles 1-2.5 mm long, ca. as long as the ovary, evenly filiform. Capsules indehiscent, 1.9-4(- 5) x 2-3.5 mm, globose to subglobose, thickened and risen around the medium large interstylar aperture, semi-translucent, surrounded or capped by the withered corolla. Seeds 2-4 per capsule, 1.42-1.86 x 1.25-1.6 mm, shape heterogeneous on the same plant: dorsoventrally compressed to weakly angled, broadly elliptic to transversely oblique, seed coat cells variable: alveolate/papillate, polygonal (not alveolate/papillose), and puz- zle-like arranged, or both kinds present on the same seed. In * 30. Sonora.— Sonoran Desert and coastal thornscrub; ca. 10-100 m. Flowering August-November(-March). Apparently spreading as a weed of alfalfa (with contaminated seeds); other hosts include Baccharis, Chenopo- dium, Helianthus, Heterotheca, Ipomoea, Pluchea, Polygonum, Rhynchosia, and Tephrosia. General distribution. — throughout most of the USA; Aguascalientes, Chihuahua, Coahuila, Edo. Mexico, Jalisco, Michoacan, Nuevo Leon, Puebla, Queretaro, Sinaloa, Tabasco, Tamaulipas, Veracruz, Zacatecas, Yuca- tan; West Indies; Central America; South America. As in the case of C. campestris, C. indecora may be more widely distributed in Mexico than current herbarium data suggest. Selected reference. — Costea et al. (2006b). ir Garcia 9 Aug 2010 Cm- a 2010-01574,” WLU). Mpio Hus Van Devender 92-1 101. and 1 '> M.ir log \ p, u „ J, , ■. t-,522 ■ \RIZ U.UR ARI7 l ( >.2 km Wot lurr.i v l.ik-i Lid mi \ ro.ul (<’ c .tm.i- hmroa, 26°33'50”N > 109°12'50"W. 24 Nov 1093, Van Devender 93-1283 (ARIZ, ASU, CAS, TEX, UC.UCR, USON). Mpio Etchojoa: Etchojoa, Febrero, weed in alfalfa, Ley 21 Aug 2010 (“muestra 2010-01980,” WLU). Mpio San Ingnacio: Ejido San Ignacio Rio Muerto, weed in alfalfa, Sdanr-Garda 9 Aug 2010 (“muestra 2010-01573,” WLU). Cuscuta legitima Costea & Stefanov., Taxon 59:1795. 2010. Type: MEXICO. Sonora: NW side of Rio Yaqui at MEX 15 near r. reflexa (J M. Coult.) , 1889, Nealley 338 (holotype: US!; iso Stans slender, yellow-orange. Inflorescences dense to loose, umbelliform, confluent. Pedicels 2-10 mm long, tacts 1 at the base of clusters and 0-1 at the base of pedicels, 2.0-3.6 mm long, broadly triangular-ovate, mar- gins entire, apex acuminate. Flowers 5-merous (Fig. 3f), 4.0-5.5(-6.0) mm long, membranous, white when fresh, creamy-white when dried. Papillae absent. Laticifers evident in the bracts, calyx, corolla, tips of infras- taminal scale fimbriae, and ovary, isolated, ovoid. Calyx 2.5-3.2 mm long, straw-yellow, finely reticulate, slightly shiny, campanulate, longer than corolla tube, divided ca. % the length, tube 0.6-1.0 mm long, lobes l5 ~ 22 mm long, not basally overlapping, ovate-lanceolate, not carinate, margins entire, apex acuminate. Corolla 3.8-5.2(-5.6) mm long, tube 1.6-2.1 mm long, campanulate, lobes 1.8-3.0 mm long, initially erect, fcr reflexed, longer than the tube, linear-lanceolate, margins entire, apex acuminate, straight. Stamens ex- j^ ned ’ starter than the lobes, anthers 0.50-0.70 x 0.24-0.36 mm, elliptic to oblong, filaments 0.6-1.0 mm 0ng hrfrastaminal scales 1.8-2.2 mm long, equaling or slightly longer than the tube, bridged at 0.2-0.4 mm, ^thulate to obovate, uniformly dense-fringed, fimbriae 0.2-0.5 mm long. Styles 0.9-2.5 mm, longer than the ° Var y- eve nly filiform. Capsules circumscissile, 2-3 x 1-2 mm, depressed, irregularly thickened and slightly 'lj Cn arou nd the inconspicuous interstylar aperture, translucent, surrounded or capped by the withered cor- Seeds 2-4 per capsule, 0.9-1.2 x 0.8-0.9 mm, broadly elliptic to subround. Sonora. Chihuahuan and Sonoran Deserts, coastal and foothills thornscrub, tropical deciduous forest; 200 m - lowering August-November. Common but not weedy; hosts herbaceous, including Allionia in- , Boerhavia, Chamaesaracha, Evolvulus, Kallstroemia, Salsola, Solanum, Tidestromia la- „.n, and Tribulus terrestris. General distribution. Arizona Kansas New Mexico, Texas; Baja California (norte). Chihuahua, Coa- tala, Tamaulinas 544 Selected reference . — Costea and Stefanovic (2010). Mpio Agna Prieta: S edge of Agua Prieta on Mex 17, Chihuahuan desertscrub, 31°18’21"N, 109°34'55"W, 1204 m, 13 Sep 2006, Van Devender h, ii, m. • II 'AH Mpio Manios: t .> pi i ,i hu.t -..i 2oBT W N I0S V. W In) m i=> Vp Van Dcvntdei ••’.-J/.MiARIZ UC, UCR). Mpio Cajeme: Ciudad Obregdn, 29 Sep 1933, Gentry 272 (ARIZ, MICH); Cerro La Antena, 1 km N of Microondas La Cabana, 27°27'45"N, 109°46'20"W, 200 m, Sinaloan thomscrub, 19 Sep 1994, Van Devender 94-603 (ARIZ, ASU, MEXU, UC, USON); NWsideof Rio Yaqui at MEX 15 near Esperanza, ca. 9 km N of Ciudad Obregdn, 27°35'45"N, 109°56’W, ca. 40 m, 10 Sep 1994, Van Devender 94-458 (ARIZ, ASU, MEXU, UC, UCR) [SEM], Mpio Hermosillo: New Year’s Mine, 20 mi S of Hermosillo, Jones 28 Oct 1926 (MO); 27 mi W of Hermosillo, on road to Kino Bay, 28 Aug 1941, Wiggins & Rollins 1 33 (ARIZ, CAS, DS). Mpio Navojoa: Sanjose de Masiaca; 26°45'N. 109 B 50"W 70 m 22 Sep 1994, Van Devender 94-710 , 94-711 (ARIZ, MEXU, UC). Mpio Puerto Penasco: Pinacate Region, MacDougal Crater, 8 Sep 1964, Fel- Pinacate Region, 155 m, 8 Dec 1970, Felger 20035 - 1 km SWW Papago Tanks, 28 Sep 1964, Felger 10608 ; Rancho Grijalva (Rancho Guadalupe Victoria), 32°00'35"N, 113°34'25"W, 225 m, Ezcurra9Nov 1982. Mpio Soyopa: Tdnichi, 28°35'55"N, 109°33'50"W, 200 m, 17 Aug 2006, Van Devender 2006-627 (ARIZ, NMC, WLU); Arroyo Las Tinajas below ruins of Toledo smelter, near Loma Maderista, 3.5 km S of TOnichi, W (MEXU, WLU), Reina-G. 2006-606 (WLU), Reina-G. 2006-612 (WLU). a leptantha Engelm., Tra :r544 (holotype: NY!; ts Stems slender, yellow-orange. Inflorescences loose, umbellate, confluent. Pedicels (l-)2-7 mm long. Bracts 1 at the base of clusters and 0-1 at the base of pedicels, 0.75-1 mm long, triangular ovate, margins entire, apex acute. Flowers 4-merous (Fig. 3h), 3.5-4.5(-5) mm long, membranous, white when fresh, creamy-white when dried, papillae usually present on the pedicels and perianth. Laticifers not visible. Calyx 1.5-1.8 mm long, straw-yellow, not reticulate or shiny, campanulate, V1-V2 of the corolla tube, divided ca. Vi the length, the tube 0.5-0.8 mm long, lobes 0.8-1 mm long, not basally overlapping, triangular-ovate, not carinate, margins entire, apex acute. Corolla 3-4 mm long, tube cylindric, 1.5-2.5 mm long, lobes 1.5-2 mm long, initially erect, later spreading or reflexed, as long as the tube, lanceolate, margins entire often involute upon drying and corolla lobes appearing narrow, apex acute ± cucullate. Stamens short-exserted, shorter than corolla lobes, anthers 0.4-0.6 x 0.35-0.45 mm, subround to broadly elliptic, filaments 0.3-0.6 mm long. Infrastaminal scales 1.3-2J mm long, ca. Vi of the corolla tube, bridged at 0.4-0.8 mm, oblong, uniformly short-fringed, fimbriae 0.05-0.15 mm long. Styles 1.2-2. 1 mm long, longer than the ovary, evenly filiform. Capsules circumscissile, 1-5-2 * 1.6-1.9 mm, globose, slightly thicken and risen or with a few protuberances around the inconspicuous mter- stylar aperture, translucent, capped by the withered corolla. Seeds 2-4 per capsule, 0.75-0.9 x 0.7-0.8 mm, angled, subrotund to broadly elliptic, seed coat cells alveolate/papillate. Sonora. — Sonoran Desert; 5-150 m. Flowering December-May. Parasitic on Euphorbia subgenus Charnae- syce, especially E. polycarpa. General distribution. — Texas, New Mexico; Baja California (norte) and Sur, and Sinaloa. Selected reference. — Costea and Stefanovtf (2010). Mpio Hermosillo: 1 .5 mi E of Santa Rosa, 15 Feb 1965, Felger 12575; 4 mi by road NW of Rancho Noche Buena at ca. 0.5 mi E of crest of “Sm Pass,” Sierra Seri, 14 May 1966, Felger 14035 ; Roadside 3.7 mi S of Punta Chueca, 13 Apr 1980, Bowers 1966; Playa Esthela, just N of Bahia Kino, 28°52’28"N, 112°0r20"W, 50 m, 31 Dec 2000, Von Devender 2000-933 (WLU) [SEM] . Mpio Pitiquito: 5.9 mi S of Desemboque RioS» Ignacio, 14 Apr 1968, Felger 17762; 19.8 mi S of Desemboque Rio San Ignacio, 14 May 1966, Felger 14080; ca. 1 mi E of 19 mi by roadSd Desemboque, vie. 29°20'N, 11214^, 18 Feb 1968, Felger 17205. SONORAN ISLANDS. Tiburon: SW Central Valley, Felger 17342 ; 1 inland at Zozni Cmiipla, at base and N side of Punta San Miguel, 23 Nov 2006, Wilder 06-368; Canyon at base of CapxOlim, 24 Nov Wilder 06-381. Cuscuta macrocephala W. Schaffn. ex Yunck., Illinois Biol. Monogr. 6(2-3):36. 1921. Type: MEXICO. Sinau* !®> date], Schaffners.n. (uoi.otypi : NY!). Stems orange, coarse. Inflorescences dense, paniculiform-glomerulate. Pedicels 0.3-3.2 mm long. Bracts 1 i[ the base of clusters and 0-1 at the base of pedicels, 1.5-3 mm long, ovate, margins entire, apex obtuse. Flo** 1 * 5-merous (Fig. lb), 5-6.5 mm 1 ca. % of its length, tube 1.3-2 1 per capsule, 1.4-1.9 x 1-1.3 mm, a (osteaetal.. 547 x 1.25-1.43 mm, ± visible through the pericarp, dorsoventraily compressed, broadly elliptic to subround; sur- face of seed coat epidermis alveolate when dried and papillate when hydrated. Sonora. — The species is apparently localized at Sonoyta (400 m) and Quitobaquito (at the Arizona- Sonora border, 335 m). Flowering April-May. Growing on Suaeda moquinii. General distribution. — Arizona, California, Nevada, New Mexico, Utah, Texas; Baja California (norte) on herbaceoushosts (e.g. , species ofFrankenia, Salsola, Suaeda, Wislizenia) from inland salt flats, marshes, and ponds. Selected references. — Costea et al. (2006c, 2009), Felger (2000). I Type: MEXICO. Oaxaca; 1827, Pedicels 0.5-2.6 mm long. Bracts 1 at the base of clusters, usually absent at the base of pedicels or flowers, 1.5-3 mm long, oblong to oblong lanceolate, acute to obtuse, margins entire. Flowers 5-merous, 4-5.2 mm long, thick, white when fresh, reddish-brownish when dried. Papillae absent. Laticifers visible in the calyx, corolla, isolated or in rows, ovoid to elongated. Calyx 2-3 mm long, reddish-brownish, more or less reticulate, ± glossy, campanulate, equaling corolla tube, divided VS-*A the length, tube 0.5-1 mm long, lobes 1.6-2.2 mm long, broadly overlapping, round to broader than long or occasionally broadly elliptic, not carinate or with multicellular protuberances on the midveins, margins entire, apex rounded. Corolla 3.5-5 mm long, tube 2.3-3 mm long, campanulate, lobes 1.5-2.5 mm long, initially erect, later reflexed, equaling or shorter than the lube, oblong-ovate, overlapping, margin entire, apex rounded, straight. Stamens exerted, shorter than corolla lobes, anthers 0.7— 1.1 mm long, oblong-elliptic, filaments 0.8-1.2 mm long. Infrastaminal scales 2.5—3 mm long, equaling corolla tube, bridged at 0.8-1.2 mm, oblong to ovate, uniformly dense-fringed, fimbriae 0.2-0.5 mm long. Styles 1.2-2.1(-3) mm long, longer than the ovary, thick, but uniform. Capsules circumscissile, 1-5-3 x 1.8-2.5 mm, globose to depressed-globose, not thickened and/or risen around the small interstylar aperture, translucent, capped by the withered corolla. Seeds (2-)4 per capsule, 1.5-2 x 12-1.9 mm, angled or slightly dorsoventraily compressed, elliptic-oblong to subround, seed coat cells alveolate/papillate or wrin- Sonora.— This species may be a sporadic introduction. It was collected in northern Sonora in 1994 from a cultivated Schinus terebinthifolia tree. On many subsequent visits, however, it was not found again. Ornamental trees and shrubs grown in Sonora are often brought from nurseries in Guadalajara, Jalisco, which points to a Potential source for new introductions. Sonoran Desert. Flowering December-January. General distribution.— Juncker (1932, 1965) mentioned that C. tinctoria is common throughout Mexico to Guatemala. This species is part of the largest and most complicated taxonomically clade in Mexico (clade “G”; Stefanovic et al. 2007) and the delimitation of species and their distribution require more study. Selected reference.— Yuncker (1932). 10 °8(ariz, mexu, uson, WLU) ISEM], 5153 gd Guscuta tuberculata Brandegee, Univ. Calif. Publ. Bot 3 38Q 1909 Type MEXICO. Lower California [Baja California S ‘ ems fiIif °rm, yellow-orange. Inflorescences loose, umbelliform or racemiform, confluent. Pedicels 2-3(-5) !" m lon & Bracts 1 at the base of clusters, usually absent at the base of peduncles, 0.5-0.75 mm long, ovate- nce °late, margins entire, apex acute. Flowers 5-merous (Fig. 3i), 2.5-4 mm long, membranous, white- when fresh, creamy when dried. Papillae present especially at the base of the corolla tube. Laticifers ^visible in the corolla, isolated, ovoid to elongated. Calyx 0.5-1.5 mm long, yellow, not or finely reticulate, ^Pulate-angular, Vi-Vi as long as the corolla tube, divided almost to the base, tube 0.2-0.5 mm long, fr-1-3 mm long, not basally overlapping, triangular to lanceolate, carinate and/or with multicellular pro- Costea et al., Convolvulaceae of Sonora, Mexico: Cuscuta General distribution. — Arizona, Colorado, Ne' Federal, Guanajuato, Guerrero, Edo. Mexico, Hidalgo, Jalis rftaro, San Luis Potosi, Tamaulipas, Veracruz; West Ind Selected reference . — Costea and Stefanovic (2010). i, Durango, Distrito on, Oaxaca, Puebla, Que- and South America. Cuscuta vandevenderi Costea & St< :r. Pedicels 0.7-6 ir *s corymbiform cymes arranged in dense globose inflorescences, 1-3 cm in diame- lg. Bracts 1 at the base of clusters, 0.5-0.75 mm long, triangular-ovate, margins entire . Flowers (4-) 5-merous, 2-2.6 mm long, membranous, white turning cream yellow- ish when fresh, creamy-light brown when dried. Papillae absent. Laticifers prominent in the calyx and corolla articulated or isolated, rectangular, ovoid to elongated. Calyx 0.9-1.6 mm long, yellow-gray, not shiny, cam panulate, divided y h- l h mm, tube 0.4-0.8 mm long, lobes 0.5-1.2 mm long, not overlapping to overlapping triangular to triangular-lanceolate, carinate, margins ± entire to serrulate, apex acute, acuminate to obtuse Corolla 1.5— 2.1 mm long, tube campanulate, 0.9-1.3 mm long, lobes 0.6-1.3 mm long, erect to slightly spread ing, triangular, margins entire to irregular, apex obtuse to acute. Stamens equaling to longer than corolla lobes anthers 0.3-0.5 x 0.3-0.4 mm, subround to broadly elliptic, filaments 0.6-1.5 mm long. Infrastaminal scales 1-1.2 mm long, equaling corolla tube, bridged at 0.4-0.5 mm, oblong-ovate to truncate, densely fringed, fim- briae 01-0.2 mm long. Styles 0.8-1.2 mm long, longer than the ovary, uniformly filiform. Capsules indehis- cent , 1-8-2.4 x 0.8— 1.2 mm globose— depressed to globose— obovoid, slightly thickened but not risen around the relatively large interstylar aperture; persistent corolla surrounding the base of capsules. Seeds 3-4 per tapsule, 0.9-1.1 x 1-1.2 mm, subrotund, seed coat cells alveolate/papillate. Cuscuta vandevenderi resembles C. gracillima from which it differs through the smaller flowers and inde- hiscent capsules surrounded by persistent corollas, and the often serrate calyx and corolla lobes. From C. del- tottko, which is the closest related species, it can be separated by the indehiscent capsules and denser inflores- cences (Costea et al. 2008). Sonora.— Southeast and central part of the state in tropical deciduous forest, oak woodland, and pine-oak forests (sometimes in openings); 350-1550 m. Flowering September-December. Parasitizing various herbs, e &,Ayenia, Chamaecrista, Cosmos, Euphorbia, Sida, and Evolvulus. General distribution .— Sonora and Baja California Sur. Selected reference .— Costea et al. (2008). ^ Ahmos: Sierra Tecurahui, 1200-1500 m, 26-28 Oct 1961, Gentry 19423 (US); 3.9 km above Rancho El Palmarito, 23.9 km, E-NE of AUmos ’ 27 °03'04"N, 108°45'51"W, 516 m, 1 Oct 2006, Van Devender 2006-983 (WLU); El Guayabo Crossing of Rio Cuchujaqui, 14 Km (by ^E-SEof Alamos, 27°00’05"N, 108°47'08"'W, 370 m, 2 3, Steinmann 93-349 (ASU). Mpio Mazatan: Sierra de Mazatfn, E-SEofYecora, 2 ^ see Felger et al. (2012) for full acknowledgments.The following herbaria have provided Cuscuta plant ^ nal: AA U ALTA, AR1Z, ASU B BAB BOL BRIT, CANB, CAS, CEN, CHR, CHSC, CIIDIR, CICY, CIMI, rf- DA 0. F, G, GH, H, HUFU, IAC, IEb] IND, j, JEPS, LL, LP, LPB, LPS, K, MEL, MERL, MEXU, MICH, MO, C ’ ***• ° A Q OKLA, OSC, OXF, PACA, PRE, QCNE, QFA, P, PACA, RB, RSA, SAM, S, SD, SGO, SI, SPF, 550 TEX, TRT, TRTE, UA, UB, UBC, UCR, UCT, UNB, UNM, UPRRP, UPS, US, USAS, WTU and XAL. Cuscuta research was supported by a Natural Sciences and Engineering Research Council of Canada Discovery grant to Costea (327013-06 and 327013-12). We thank J. Andrew McDonald and Javier Ortega for providing helpful Austin, D.F. 1982. 165.Convolvulaceae. In: G.W. Harling and B.B. Sparre, eds. Flora of Ecuador, vol. 15. University of Gote- borgand Swedish Museum of Natural history, Goteborg and Stockholm. Pp. 1-98. Costea, M. 2007-onwards. Digital atlas of Cuscuta (Convolvulaceae). Wilfrid Laurier University, Ontario, Canada, https:// www.wlu.ca/page.php?grp_id=2147&p=8968&pv=1 (viewed 16 June 2011). Costea, M. and FJ. Tardif. 2004. Cuscuta (Convolvulaceae) — the strength of weakness: a history of its name, uses and parasitism concept during ancient and medieval times. Sida 21:369-378. Costea, M. and FJ. Tardif. 2006. The biology of Canadian weeds. Cuscuta campestris, C. gronovii, C. umbrosa, C. epithymum and C. epilinum. Canad. J. PI. Sci. 86:293-31 6. Costea, M., and S. StefanoviC. 2009. Cuscuta jepsonii (Convolvulaceae), an invasive weed or an extinct endemic? Amer. J. Costea, M., and S. StefanoviC. 2010. Evolul evidence of extensive hybridization from discordant nuclear and plastid phylogenies. Taxon 59:1 783-1800. Costea, M., G.L. Nesom, and S. StefanoviC. 2006a. Taxonomy of Cuscuta pentagona complex. Sida 22:151-175. Costea, M., G.L. Nesom, and S. StefanoviC. 2006b. Taxonomy of Cuscuta indecora complex. Sida 22:1 76-1 95. Costea, M., G.L. Nesom, and S. StefanoviC. 2006c. Taxonomy of Cuscuta californica-salina complex. Sida 22:197-207. Costea, M., F. Aiston, and S. StefanoviC. 2008. Species delimitation, phylogenetic relationships and two new species in the ex (Convolvulaceae). Botany 86:670-681 . 3 S. StefanoviC. 2009. Untangling the systematics of salt marsh dodders: Cuscuta pacifka a i Una. Syst. Bot. 34:787-795. Costea, M., I. Spence, and S. StefanoviC. 2011a. Cuscuta chinensis species complex evidence for long-distance dispersal and one new species. Org. Divers. Evol. 1 1:373-386. Costea, M., I.R. Garcia, and S. StefanoviC. 201 1 b. 'Horned' dodders: phylogenetic relationships and two new species within Cuscuta chapalana complex (Convolvulaceae). Botany 89:71 5-730. Felger, R.S. 2000. Flora of the Gran Desierto and Rio Colorado of northwestern Mexico. University of Arizona Press. Felger, R.S., D.F. Austin, T.R. Van Devender, J J. SAnchez-Escalante, and M. Costea. 201 2. Convolvulaceae of Sonora, Mexico. L Convolvulus, Cressa, Dichondra, Evolvulus, Ipomoea, Jacquemontia, Merremia, and Operculina. J. Bot. Res. Inst. Texas 6:459-527. Felger, R.S., B.T. Wilder, and J.P. Gallo-Reynoso. 201 1 . Floristic diversity and long-term vegetation dynamics of Isla San Pedro Nolasco, Gulf of California, Mexico. Proc. San Diego Soc. Nat. Hist. 43:1 -42. StefanoviC, S., M. Kuzmina, and M. Costea. 2007. Delimitation of major lineages within Cuscuta subg. Grammica using pbs- tid and nuclear DNA sequences. Amer. J. Bot. 94:568-589. Welsh, M., S. StefanoviC, and M. Costea. 2010. Pollen evolution and its taxonomic significance in Cuscuta. PI. Syst Evol 285:83-101. Wright, MAR., M. Welsh, and M. Costea. 2011. Diversity and evolution of gynoecium in Cuscuta (dodders, Convolvulaceae) in relation to their reproductive biology: two styles are better than one. PI. Syst. Evol. 296:51 -76. Wright, M.A.R., M.D. ianni, and M. Costea. 201 2. Diversity and evolution of pollen and ovule production in Cuscuta (dod- ders, Convolvulaceae) in relation to floral morphology. PI. Syst. Evol. 2:369-389. Yuncker, T.G. 1921. Revision of the North American and West Indian species of Cuscuta. Illinois Biol. Monogr. 691-231 Yuncker, T.G. 1932. The genus Cuscuta. Mem. Torrey Bot. Club 18:113-331. Yuncker, T.G. 1965. Cuscuta. North American Flora, ser. 2, 4:1-51. TREPTOSTEMON (LAURACEAE), A NEW GENUS OF FOSSIL FLOWER FROM MID-TERTIARY DOMINICAN AMBER Kenton L. Chambers George 0. Poinar, Jr. Andre S. Chanderbali Department of Botany and Plant Pathology Oregon State University Corvallis, Oregon 9733 1, U.S.A. chamberk@science.oregonstate.edu Department of Zoology Oregon State University Corvallis, Oregon 97331, U.SJ Department of Biology University of Florida Gainesville, Florida 3261 1, U.S.A ABSTRACT be assigned to any modem gem RESUMEN INTRODUCTION Amber deposits of the Cordillera Septentrional, Dominican Republic, continue to yield interesting fossil angio- sperm flowers from the low-elevation tropical forests characteristic of Mid-Tertiary Caribbean vegetation (Poinar & Poinar 1999). We have recently described fossils assignable to Licania (Chrysobalanaceae) (Poinar etal. 2008a, revised by Chambers & Poinar 2010), Persea (Lauraceae) (Chambers et al. 2011a), Thchilia (Meliaceae) (Chambers et al. 2011b), Swietenia (Meliaceae) (Chambers & Poinar 2012), and Trochanthera (pos- sibly Moraceae) (Poinar et al. 2008b). A second flower of Lauraceae is described in the present paper. It has several well-marked features, including 6 fertile stamens in whorls 1 and 2, whose anthers have 2 pairs of pores tlwt °P en extrorsely. The stamens of whorl 3 display 2 upright valves but are possibly staminodial, since their terminal portion is dissimilar in size and shape from the fertile anthers. The anthers are closely adjacent and are on short filaments. The epidermis of the anthers and staminodes is densely covered with minute trichomes. No PisfiHode or whorl 4 staminodes are visible. Because of the extrorse dehiscence of the 6 anthers of whorls 1 3nd 2 ’ the fossil cannot easily be accommodated in any modem genus of Lauraceae. We therefore propose to Separate 11 as the new genus and species Treptostemon domingensis. MATERIALS AND METHODS AH the amber fossils referred to above came from mines in the Cordillera Septentrional, between Puerto Plata ®nd Santiago, Dominican Republic. Dating methods applied to the deposits are thus far equivocal. An age of 30 ®ybp was assigned by Cepek in Schlee (1999) based on fossil coccoliths, and one of 20-15 mybp was ermined by Iturralde-Vinent and MacPhee (1996) based on foraminifera. The amber is found in turbiditic 552 sandstones of the Upper Eocene to Lower Miocene Mamey Group (Draper et al. 1994). Animals and plants of the forest vegetation present at the time of amber deposition were described by Poinar and Poinar (1999), the original resin having been a product of the arborescent genus Hymenaea (Fabaceae). In Poinar and Poinar (op. ciL), the present flower was inadvertently assigned to the genus Nectandra (Pg. 21, Fig. 10). DESCRIPTION Treptostemon K.L. Chambers, Poinar, & A.S. Chanderbali, gen. nov. Type Species: Treptostemon domingensis K.L. Chan- Diagnosis.— Flower staminate, radially symmetrical (Fig. 1), receptacle flat, pedicel remnant strigose, tepals6, in 2 whorls of 3, separate, approximately equal, spreading, lightly strigose on both surfaces, margins involute, glabrous (Fig. 2), stamens 9, in whorls of 3, at least the outer 2 whorls fertile, anthers minutely puberulent, those of whorls 1 and 2 with 4 pores arranged as 2 pairs, one above the other, all extrorse, opening by apical valves, stamens of whorl 3 functional or staminodial, their distal portion dissimilar in size and shape from the fertile anthers (Fig. 2), 2 or 4 small pores probably present, the distal pair opening extrorsely by upturned valves, glands of whorl 3 stamens small, whorl 4 staminodes and pistillode not evident. Pistillate flower un- Etymology. — From Greek “treptos,” turned, and “stemon,” stamen. Description.— Tepals lanceolate, acute, laterally spreading, 3.1 to 4.0 mm long, 1.1 to 1.6 mm wide (Fig. 1), an- thers of whorls 1 and 2 0.9 to 1.0 mm wide, ca. 1,0 mm long, with short filaments, pores extrorse, circular to oblong, ca. 0.3 mm in diameter, valve remnants visible in a few cases (Fig. 2), connective forming an adaxial ridge, stamens of whorl 3 club-shaped, distal portion cylindrical, 0.7 mm in diameter, pores extrorse, the distal pair defined by small, upturned valves (Fig. 2), otherwise not observable, glands of whorl 3 stamens 0.19 mm wide (only 1 observed). Etymology. — From source of amber in Domin DISCUSSION Examination of the fossil is best done from above, as in Figures 1 and 2. In a lateral view, obtained with diffi- culty, the outer anthers display all 4 extrorse pores, although in apical view, the lower pair may be barely visible (Fig. 2). The filaments of the outer stamens are quite short and curve at the tip, so that the anthers face dorso- laterally. The positioning of the stamens does not allow observation of pores on whorl 3, and their number is uncertain. The upraised valves of the distal pores (Fig. 2) are in an extrorse position. Despite the reduced size and cylindrical shape of these anthers, it is uncertain whether the whorl 3 stamens are fertile or staminodial. A staminodial condition is possible, considering the report by Kubitzki and Kurz (1984) that in staminate flow- ers of 3 dioecious species of Ocotea they studied, “[tjhe staminodes often possess valves that open after the wilting of the stigma” (i.e., stigma of the pistillode). Only one gland of the whorl 3 stamens is in view (Fig. 2) On some anthers, small, black air bubbles that have settled among the surface trichomes may be confused for pores, but these are artifacts. The extrorse dehiscence in all androecial whorls of Treptostemon is unique in Lauraceae, although in Pleurothyrium there are 9 tetrasporangiate, latrorse anthers which may look almost extrorse (J. Rohwer, pers. comm.). The shortness of the stamens of Treptostemon and their close positioning in the flower are features similar to Aniba and Aiouea (Kubitzki and Renner 1982), among other genera, although a close relationship to these 2 taxa, which are hermaphrodite, with upright tepals and only 2 pores per anther, is not likely. Furthermore, the usual laurad introrse dehiscence of whorls 1 and 2 anthers occurs in the 2 genera. Stamens with short filaments also occur in the New World dioecious genera Rhodostemonodaphne (MadriMn 2004) and Ocotea (the Chambers et al., Treptostemon, a new f 553 v of complete flower. Scale bar = 1 .4 mm. with bisexual or unisexual flowers). Both have 4-pored, introrse anthers, but in Khodostemonodaphne the pores form a more or less horizontal row (Rohwer 1993; van der Werff 1991 , Fig. 2E), unlike the pores of Treptostemon anthers. The latter two genera are similar in lacking staminodes of whorl 4 but differ in their receptacle, which 15 narrow ly tubular in Rhodostemonodaphne (Rohwer op. cit.). The closest relative among genera with 4-pored *«heis is probably Ocotea, a species-rich and variable taxon characterized by Rohwer (1993, p. 382) as “the dustbin of the Perseae.” Dioecious species such as O. pyramidata (Allen 1945) may have anthers 1.0 mm long, Wuh the filament only 0.5 mm. (As an aside, although pollen is shed inwards by these anthers, it is also released outwards from the 2.1 5 mm-long stamens of whorl 3). According to Rohwer (op. cit.) the whorl 4 staminodes of Ocot ea are absent in unisexual flowers, the receptacle varies from flat to deeply tubular, and the pistillode in flower s may be present or absent. The principal difference between Ocotea and Treptostemon, therefore, is 1 at the former has anthers of whorls 1 and 2 dehiscing introrsely. Pollen had recently been discharged from the Treptostemon flower before it became imme 555 in, as shown by a cloud of tiny pollen grains belc would have had to release pollen from the stami visiting insects for transfer to pistillate flowers on another tree. It is likely that dioecy, nectar ] extrorse anther dehiscence were adaptive features of the species’ pollination syndrome, but w moved in time to reconstruct other details of its floral biology. A brief review of reports on poll present-day Lauraceae may be of interest, nonetheless, in conjunction with a discussion of avail in the insect fauna already known from Dominican amber (Michener & Poinar 1997; Poinar & Poinar 1999). A report on pollinators identified from modern tropical rain forests in Costa Rice (Bawa et al. 1985) lists only one member of Lauraceae, an unidentified species of Ocotea. It is cited as hermaphroditic and a member of the subcanopy, its major pollinators being beetles. Whether the flowers offer both nectar and pollen as food rewards is not mentioned. However, in her generic description of Ocotea, Allen (1945) states that stamens of the inner series always bear two sessile or stipitate glands, presupposing the presence of a nectar reward for visitors. In a study of dichogamy and dioecy in Neotropical Lauraceae, Kubitzki and Kurz (1984) included ob- servations of pollinators for 3 dioecious species of Ocotea. Two species were seen to be visited by brown bees, about 1 cm in length, while small diptera, wasps and moths sucking nectar were occasionally encountered. Larger flies were observed visiting flowers of O. guianensis and O. opifera but could not be caught for identifica- tion. A more detailed study of pollination in a single dioecious species, Laurus azorica of Macaronesia, was published by Forfang and Olesen (1998). In a sample of over 200 trees, flower visitors included at least 11 dif- ferent insect species. These were: Hymenoptera-2 species of Halictidae, 1 of Apidae, and 1 of Ichneumonidae; Diptera-1 species of Musicidae, 1 of Syrphidae, 1 of Tachinidae, and 1 of Bibionidae; Lepidoptera-1 species of Nymphalidae; Coleoptera-1 species of Nitidulidae; Hemiptera-1 species of Pentatomidae. Only Halictidae and Tachinidae were common, comprising 97% of the visits. Bees are also considered to be the principle pollinators of the avocado, Persea americana, even though flies and other insects also visit the flowers (Free 1993). Generalized, radially symmetrical flowers of Lauraceae, with spreading perianth and exposed stamens and pistil, are open to visits by many different groups of anthophilous insects. A potential pollinator might develop a specific attraction to the odor of pollen or nectar in such a flower and become faithful to one or a few species (Faegri & van der Pijl 1979). However, it is unlikely that an unspecialized flower like that of Treptoste- mon - even with its modified form of pollen presentation atypical of the family, would have been closely adapted to a single kind of pollinator. Since their origination in the Early Cretaceous (Danforth & Poinar 2011), bees, which rear their young on pollen, have been among the most dependable pollinators of angiosperms. Based on the above reports indicating that bees are frequent visitors of present-day Lauraceae, it is likely that this group of insects also visited and pollinated T. domingensis in the Tertiary forests of Hispaniola. The bee fauna of Do- toinican amber includes representatives of families Apidae, Andrenidae, Colletidae, and Halictidae (Michener & Poinar 1997). Of these, the most common bee in Dominican amber is the small, stingless Proplebia domini- ““ (Apidae, tribe Meliponini), which could well have been significant in the pollination system of Treptoste- mon - D espite their presence in the fossil record, stingless bees do not occur in Hispaniola today. Their disap- pearance as keystone species could have led to the elimination of many plant species, including Treptostemon. ACKNOWLEDGMENTS We thank Jens G. Rohwer and Santiago Madrinan for their helpful review comments. REFERENCES ^ CA 1945. Studies of the Lauraceae, VI. Preliminary survey of the Mexican and Central American species. J. Arnold Arbor. 26:280-434. Bawa ' K - s - s -H. Bullock, D.R. Perry, R.E. Coville, and M.H. Grayum. 1 985. Reproductive biology of tropical lowland rain forest bees- "• Pollination systems. Amer. J. Bot. 72:346-356. T*®*’ K.L and G.O. Poinar, Jr. 201 0. The Dominican amber fossil Lasiambix (Fabaceae: Caesalpimoideae?) is a Ucama hrysobalanaceae). J. Bot. Res. Inst. Texas 4:21 7-21 8. 556 K.L., G.O. Poinar, Jr., and A.E. Brown. 201 la. A fossil flower oiPersea (Laura tes. Inst. Texas 5:457-462. », K.L, G.O. Poinar, Jr., and A.E. Brown. 201 1 b. Two fossil flowei nst. Texas 5:463-468. », K.L. AND G.O. Poinar, Jr. 2012. A Mid-Tertiary fossil flower of Swietenia i Danforth, B.N. and G.O. Poinar, Jr. 201 1 . Morphology, classification, and antiquity of Mellittosphex burmensis (Apoidea: Melittosphecidae) and implications for early bee evolution. J. Paleontol. 85:882-891 . and J.F. Lewis. 1 994. Hispaniola. In: S. Donovan and T.A. Jackson, eds. Caribbean geology: an introduc- sity of the West Indies Publishers' Association, Kingston, Jamaica. Pp. 129-150. der Pul. 1979. The principles of pollination ecology. Ed. 3. Pergamon Press, Oxford. Forfang, A-S. and J.M. Olesen. 1 998. Male-biased sex ratio and promiscuous pollination in the dioecious island tree Laurns azorica (Lauraceae). PI. Syst. Evol. 21 2:1 43-1 57. Free, J.B. 1993. Insect pollination of crops. Ed. 2. Academic Press, New York. 1850-1852. Kubitzki, K. and H. Kurz. 1984. Synchronized dichogamy and dioecy in Neotropical Laurai Kubitzki, K. and S. Renner. 1982. Lauraceae I (Aniba and Aiouea). FI. Neotropica 31:1-124. MadriNAn, S. 2004. R ’aceae). FI. Neotropica 92:1-102. Michener, C.D. and G.O. Poinar, Jr. 1997. The known bee fauna of Dominican amber. J. 1. Syst. Evol. 147:253-266 s Entomol. Soc 69, suppL 1996:353-361. Poinar, G.O. Jr. and R. Poinar. 1 999. The amber forest. Princeton University Press, Princeton, NJ. Poinar, G.O. Jr., K.L. Chambers, and A.E. Brown. 2008a. Lasiambix dominicensis gen. and sp. nov., a eudicot flower in Domini- can amber showing affinities with Fabaceae subfamily Caesalpinioideae. J. Bot. Res. Inst. Texas 2:463-471. Poinar, G.O. Jr., K.L Chambers, and A.E. Brown. 2008b. Trochanthera lepidota gen. and sp. nov., a fossil angiosperm inflores- cence in Dominican amber. J. Bot. Res. Inst. Texas 2:1 167-1173. Rohwer, J.G. 1 993. Lauraceae. In: K. Kubitzki, J.G. Rohwer, and V. Bittrich, eds. The families and genera of vascular plants II. Springer-Verlag, Berlin, Germany. Pp. 366-391 . Schlee, D. 1999. Das Bernstein-Kabinett. Stuttgarter Beitr. Naturk. Ser. C, 28. VAN DER Werff, H. 1 991 . A key to the genera of Lauraceae in the New World. Ann. Missouri Bot. Gard. 78:377-387. A NEW FOSSIL SPECIES OF COLPOTHRINAX (ARECACEAE) FROM MID-TERTIARY MEXICAN AMBER Kenton L. Chambers George 0. Poinar, Jr. Alex E. Brown Department of Botany and Plant Pathology Oregon State University Corvallis, Oregon 97331, U.S.A. :e.oregonstate.edu Department of Zoology Oregon State University Corvallis, Oregon 97331, U.S.A. 629 Euclid Ave. Berkeley, California 94708, U.S.A. RESUMEN INTRODUCTION A new fossil species of palms, described here as Colpothrinax chiapensis, is represented by 4 flowers embedded m Mid-Tertiary amber from a mine in the state of Chiapas, Mexico. The floral morphology of the species is similar in general features to the modern genus Colpothrinax, whose three species are found in Cuba and Cen- tral America, from Guatemala to Panama (Evans 2001). The pistil, in particular, is like that of this genus, with 3 separate but apposite carpels and an elongate style, which appears to be fused and grooved rather than com- posed of 3 separate units. The style tip is tapering and lacks an enlarged stigma. The basally connate petals, forming a short tube surrounding the ovary, and the almost completely united sepals with short, triangular lobes, provide further similarities. Principal differences are in the androecium of C. chiapensis, in which the longer, linear-subulate filaments are widened only near the base and slightly united into a staminal tube. The anthers are large, dorsifixed, and latrorse in dehiscence, as described for Colpothrinax (Dransfield et al. 2008). The probable age of the fossils, 22.5-26 Ma, is commensurate with an estimated mean stem age for the genus of 28-56 Ma (Bacon et al 2012). The species’ Mexican origin gives evidence for an evolutionary radiation of Col- pothrinax in the Caribbean region, as suggested by Bacon et al. (op. cit.) for Copemicia and certain other genera s . Specimen originated from an amber mine in the unojovel area in Chiapas, Mexico. Maps of the area i Was P rod «ced by Hymenaea mexicana (Poinar & Brc Lorthem mountain ranges (Chiapas Highlands) of the e given in Poinar (1992). Amber from Chiapas, which m 2002), occurs in lignitic beds among sequences of Journal of the Botanical Research Institute of Texas 6(2) primarily marine calcareous sandstones and silts. The particular amber under study is associated with the Balumtun Sandstone of Early Miocene and the La Quinta Formation of Late Oligocene, with radiometric ages from 22.5-26 Ma (Berggren & Van Couvering 1974). The amber is secondarily deposited in these marine for- mations and may be somewhat older than the above dates. The original vegetation was probably a wet to moist tropical to subtropical forest. Colpothrinax chiapensis K.L. Chambers, Poinar, & A.E. Bi deposited in the Poinar amber collection maintained at Oregon Stat 2, catalogue number Sd-9-181B, Figs. 3-4, c nov. (Figs. 1-4). Type: MEXICO. Chiapas: amber 1 y, Corvallis, Oregon 97331, U.S.A.; paratypes: see Fig Flowers hermaphrodite, ca. 6 mm in diameter when petals spread, calyx cupulate, glabrous, 0.4 mm (Fig. 4), sepals united, with 3 free, triangular, 0.4-1.1 mm lobes, petals 3, glabrous, ovate-lanceolate, ca. 3.3 mm, spreading or reflexed, with numerous parallel veins and no raised mid-nerve adaxially (Fig. 1), shallowly grooved abaxially (Fig. 2), tip boat-shaped (Fig. 1), stamens 6, exserted, glabrous, filaments subulate, 1.0-1.2 mm, widened at base, the connate portion short (Fig. 3), attached at the mouth of the corolla tube, anthers ca. 2.3 mm, dorsifixed, locules 4, dehiscence latrorse (Fig. 2), gynoecium glabrous, 3-carpellate, ovary loosely enclosed in corolla tube (Fig. 1), carpels free, apposite, style appearing united, ca. 1.2 mm, 3-grooved, tip acute, stigma not enlarged (Fig. 2). Etymology. — from the fossils’ origin in Chiapas, Mexico. r, the species’ pres- DISCUSSION The fossils differ from other genera of Trachycarpeae in such features as the enlarged anthers, persistent petals (cf. Pritchardia), cupulate calyx (cf. Brahea, Acoelorrhaphe), long filaments (cf. Copemicia), and presence of a staminal tube (cf. Serenod) (see Dransfield et al. 2008, pp. 272—285, for illustrations). The orientation of the flowers does not allow measurement of the depth of the corolla tube. In other respects, the fossils can be matched with excellent descriptions and floral illustrations of Colpothrinax in Uhl and Dransfield (1987), Dransfield and Uhl (1998), Dransfield et al. (2008), and Evans (2001). Whether the proposed new species falls within the full morphological variation of any of the extant taxa is uncertain, but we believe that the androe- cium, in particular, sets it apart from Colpothrinax wrightii (Bailey 1940) and C. cookii (Read 1969), and large petals and long, subulate, scarcely united filaments are adequately distinct from C. aphanopetala (Evans 2001). The published treatments of the 3 modern species include much information about stem and leaf mor- phology, inflorescence, pollen (esp. Dransfield et al. 1990), habitats, and associated vegetation. Vegetativi inflorescence morphology is, of course, unknown for C. chiapensis, as is the pollen; howevr ervation in resin from Hymenaea mexicana is evidence of moist tropical or subtropical forest vegetation as u* preferred habitat. . The phylogenetic relationships of the genera of tribe Trachycarpeae (Dransfield et al. 2008 [formerly istoneae, Dransfield et al. 2005]) have begun to be clarified by recent molecular studies utilizing nuclear plastid DNA (Asmussen et al. 2006; Roncal et al. 2008; Baker et al. 2009; Bacon et al. 2012). In the first 3 pape* cited, Colpothrinax is in a clade that includes several other genera with which it had previously been assoCiat ^ including Pritchardia, Washingtonia, Serenoa, Acoelorraphe, Brahea, Lmstona, and Chamaerops. However, in ^ report by Bacon et al. (2012), involving matK, ndhF, tmD-tmT and 3 nuclear loci, Colpothrinax is alone m highly supported clade that is sister to 2 large clades containing, inter alia, Brahea, Chamaerops, Serenoa, lorraphe, and Lmstona, and is well separated from Washingtonia and Pritchardia. The subtribal classification Trachycarpeae is presently in flux, it appears, as perhaps is true also in other large families now under int molecular phylogenetic study. The new intrafamilial relationships of palms revealed by molecular research have been used to in J eSt ^ additional questions beyond taxonomy alone. These include the origin and global diversificatic rain forests (Couvreur et al. 2011) and the effects of Miocene dispersal on island radiations ii 559 560 Journal of the Botanical Research Institute of Texas 6(2) tailed attention by Bacon et al. (2012). Of interest here is the authors’ estimation of the age of crown and stem nodes for the clades that they resolve within this tribe. The clade comprising Colpothrinax alone, mentioned above, is assigned a mean stem age of 28.56 Ma (dates for upper and lower age estimates were not well sup- ported in posterior probabilities analysis). The mean crown age for the genus is 7.87 Ma, with upper and lower estimates of 14.44 and 2.74 Ma. The age range assigned to the Mexican amber (Berggren & Van Couvering 1974) of 26-22.5 Ma is therefore in good agreement with a stem position for C. chiapensis. It is reasonable that crown species radiation would have occurred later, in the Central American and Caribbean area. This is one of the regions of interest to Bacon et al. (op. cit.), which they specifically mention as characteristic of the evolution of this genus, as well as of Rhapidophyllum, Brahea, Washingtonia, and Caribbean Copemida (their Fig. 5). ACKNOWLEDGMENTS Careful review comments by John Dransfield and Scott Zona are much appreciated. We thank Deborah Car- roll, Valley Library, Oregon State University, for help with bibliographic material. REFERENCES Asmussen, C.B., J. Dransfield, V. Deickmann, A.S. Barford, J.C. Pintaud, and WJ. Baker. 2006. A new subfamily classification of the palm family (Arecaceae): evidence from plastid DNA. Bot. J. Linn. Soc. 151:1 5-38. Bacon, C.D., WJ. Baker, and M.P. Simmons. 2012. Miocene dispersal drives island radiation in the palm tribe Trachycarpeae (Arecaceae). Syst. Biol. Advanced Access, pp. 1-17. Bailey, L.H. 1940. The problem of Colpothrinax. Gentes Herb. 4:256-260. Baker, WJ., V. Savolainen, C.B. Asmussen-Lange, M.W. Chase, J. Dransfkld, F. Forest, M.M. Harley, N.W. Uhl, and M. Wilkinson. 2009. Complete generic-level phylogenetic analysis of palms (Arecaceae) with comparison of supertree and super- matrix approaches. Syst. Biol. 58:240-256. Berggren, W.A. and J AH. Van Couvering. 1974. The late Neogene. Palaeogeogr. Palaeodimatol. Palaeoecol. 16:1-216. Couvreur, T.L.P., F. Forest, and WJ. Baker. 201 1. Origins and global diversification patterns of tropical rain forests: infer- ences from a complete genus-level phylogeny of palms. BMC Biol. 9:44. Dransfield, J., I.K. Ferguson, and N.W. Uhl. 1990. The coryphoid palms: patterns of variation and evolution. Ann. Missouri Bot. Gard. 77:802-81 5. Dransfield, J. and N.W. Uhl. 1 998. Palmae. In: K. Kubitzki, ed.The families and genera of vascular plants. IV. Springer Verlag, Berlin. Pp. 306-389. Dransfield, J., N.W. Uhl, C.B. Asmussen, WJ. Baker, M.M. Harley, and C.E. Lewis. 2005. A new phylogenetic classification ottne palm family, Arecaceae. Kew Bull. 60:559-569. Dransfield, J., N.W. Uhl, C.B. Asmussen, WJ. Baker, M.M. Harley, and C.E. Lewis. 2008. Genera palmarum— the evolution and classification of palms. Royal Botanic Gardens, Kew, Richmond, UK. Evans, RJ. 2001 . Monograph of Colpothrinax. Palms 45:1 77-1 95. Poinar, G.O. Jr. 1992. Life in amber. Stanford Univ. Press, Stanford, CA. Poinar, G.O. Jr. 2002. Fossil palm flowers in Dominican and Mexican amber. Bot. J. Linn. Soc. 1 38:57-61 . Poinar, G.O. Jr. and A.E. Brown. 2002. Hymenaea mexicana sp. nov. (Leguminosae: Caesalpinioideae) from Mexican am indicates Old World connections. Bot. J. Linn. Soc. 1 39:1 25-1 32. Read, R.W. 1 969. Colpothrinax cookii — a new species from Central America. Principes 1 3:1 3-22. Roncal, J., S. Zona, and C.E. Lewis. 2008. Molecular phylogenetic studies of Caribbean palms (Arecaceae) and their relat ships to biogeography and conservation. Bot. Rev. 74:78-102. ADDITIONAL FOSSILS IN DOMINICAN AMBER GIVE EVIDENCE OF ANTHER ABORTION IN MID-TERTIARY TRICHILIA (MELIACEAE) Kenton L. Chambers George O. Poinar, Jr. 562 Journal of the Botanical Research Institute of Texas 6(2) MATERIALS AND METHODS The fossils originated from mines in the Cordillera Septentrional of the Dominican Republic, between the cit- ies of Puerto Plata and Santiago, island of Hispaniola. Two differing ages have been proposed for this amber deposit the older being 45-30 Ma, based on fossil coccoliths (Cepek in Schlee 1999) and the younger being 20-14 Ma, based on foraminifera (Iturralde-Vinent & McPhee 1996). Most of the amber is secondarily depos- ited in turbiditic sandstones of the Upper Eocene to Lower Miocene Mamey Group (Draper et al. 1994). The specimens described here were found during a search of the Poinar amber collection for additional flowers of Meliaceae that might relate to those already described. They are designated as numbers D-9-27C, D-9-27D and D-9-27E in this collection. A newly described species of Swietenia, which incidentally had been illustrated earlier (Poinar & Poinar 1999), is also representative of fossil Meliaceae flowers in these amber deposits from the Dominican Mid-Tertiary tropical forest (Chambers & Poinar 2012). RESULTS The first illustrated flower, designated A (Figs. 1, 2) resembles Trichilia glaesaria in the shape and pubescence of its calyx and corolla (cf. Fig. 2 in Chambers et al. 2011). Its lobed, strigillose calyx differs from that of T. an- tique which was described as glabrous and rotate. In apical view (Fig. 1), the ovary of flower A is well devel- oped, 1 mm wide, strigillose, and situated at the base of the staminal tube. Four anthers can be seen, showing different degrees of extreme abortion. Three have knob-like terminal structures, which resemble the one much reduced anther seen in A. antiqua (op. cit.. Fig. 5). A short filament is present on one of these (upper left arrow, Fig. 1), while the fourth remnant is a tiny filament with a mere suggestion of an anther (upper right arrow). Anthers are absent at other potential sites on the irregularly lobed, partly obscured staminal tube. On the evi- dence from this flower, the filament stubs of T. antiqua shown in Fig. 3 of Chambers et al. (op. cit.) might be reinterpreted as lacking anthers from the beginning, rather than having them removed earlier by an herbivo- In the flower designated B (Figs. 3, 4), the adaxial and abaxial sides of the petals are densely covered with shiny droplets of a glassy deposit, which obscures the natural pubescence of these surfaces. The calyx, barely in view, also has this deposit. In lateral view (Fig. 3), anthers with much reduced locules are visible in alterMt- ing higher and lower notches between acute lobes on the rim of the staminal tube. The apical view of is flower (Fig. 4) shows approximately 10 anthers, which are also covered with the glassy deposit. The stigma ol the pistil is visible, but the ovary is too deeply placed to be characterized as to its pubescence or size. The third flower, C, is illustrated in lateral view in Figure 5. Its calyx is not in view, and its petals a« mostly too darkly stained to show the necessary details for a complete description. However, on the petal at left (arrow), the abaxial surface can be seen to be strigillose, as was the abaxial petal surface in T. glaesaria (Chambers et al. 2011, Figs. 1, 2), and the petal at the right appears to have the papillate adaxial surface charac- teristic of flower A. At least 8 much reduced anthers are present on the rim of the staminal tube, in alternating higher and lower positions between acute lobes as in flower B, above, and in T. glaesaria (op. cit.. Fig. D- locules of these anthers appear to be more reduced than in flower B. In an apical view of flower C (not shown, the ovary is glabrous, unlike that of flower A. It is interesting that in two extant Trichilia species, the ovary «s glabrous before fertilization but develops pubescence afterward, as does the fruit (Pennington et al. 198k P- 11). The intense staining of flower C does not allow further details to be observed. Flowers A and C i e greatly in anther development; however, as noted by an anonymous reviewer, “(t)he range of anther reduction between the 3 flowers described here could easily be found in a single modem day species.” DISCUSSION The flowers under consideration here were not discovered until our earlier paper had been published. fossils are linked through their having been excavated from amber mines in the same ocean-deposite Tertiary strata, now uplifted in the Cordillera Central of Hispaniola. As discussed by Poinar and Poinar (1^ ^ ^ Chambers and Poinar, Fossil ( 563 possible that the Trichilia fossils described here do not represent contemporaneous species. What they illus- trate is that Trichilia taxa in these forests were diverse in features such as perianth pubescence, pistil pubes- cence, and especially the nature of anther reduction and loss in pistillate flowers. However, lacking evidence relating to staminate flowers or to differences in the vegetative parts and inflorescence structure of the sampled taxa, we are limited as to any broad taxonomic conclusions that might be drawn from the observed variability in pistillate flowers. It is important, nevertheless, that these fossils be placed on record as part of the history of floral evolution in the genus, to be considered in possible future discussions of Meliaceae phylogeny. The loss and extreme reduction of anthers in flower A, and probably also in Trichilia antiqua , goes beyond the description by Pennington et al. (1981) of androecia in functionally female flowers. These authors simply ^te (p. 26) that in Trichilia the “antherodes [are] narrower than anthers, not dehiscing, without pollen ” In view of the review comments quoted above, this may be a generalization that omits a complete description of anther abortion and loss in modern species. That two of the Dominican fossils display such androecial reduc- tion is evidence of its occurrence earlier in Trichilia floral evolution than might have been expected. As discussed by Pennington et al. (1981), Trichilia is today the largest genus of Meliaceae in the New World, with ca. 85 species in lowland tropical America as well as ca. 14 species in Africa and 2 in the Indo- Malesian region. Its floral diversity, particularly in the androecium, well exceeds our sample of amber fossils (Pennington et al. op. cit., illustrations on pages 36-226). Respecting the age of Trichilia, Muellner et al. (2006) ^ evidence from known Meliaceae fossils, combined with chloroplast rbcL data, to estimate divergence times » the family. The DNA cladogram presented by these authors includes a monophyletic clade of Trichilia and 1 1 other genera, which is assigned an origin in the Oligocene. Apart from Trichilia, members of this clade are to- limited to the Old World tropics, from Africa to Madagascar, India, lnto-China, Malesia, and Austroasia. Muellner et al. (2006) propose a West Gondwanan Cretaceous origin for the family, followed by dispersal **°ss Eurasia and between Eurasia and North America over the Beringian and North Atlantic land bridges. Eurther movement from North to South America occurred via island hopping and/or direct land connections uunng the Tertiary ( see also Muellner et al. 2010, with a dense sampling of Cedreleae). As noted previously ^Chambers et al. 2011), Pennington et al. (1981) recognized only two sections in Trichilia, Sect. Trichilia and T* Moschoxyhim C. DC. The fossils discussed here are best placed in the emended Sect. Moschoxylum, the defimn g trai ts of which are the valvate petals and completely united filaments. 565 ACKNOWLEDGMENTS We thank Andrea Muellner and 2 anonymous reviewers for their helpful comments, which improved the con- tent and clarity of our presentation. REFERENCES Chambe «. K.L, G.O. Poinar, Jr., and A.E. Brown. 201 1 . Two fossil flowers of Trichilia (Meliaceae) in Dominican amber. J. Bot Res. Inst. Texas 5:463-468. K.L. and G.O. Poinar, Jr. 2012. A Mid-Tertiary fossil flower of Swietenia (Meliaceae) in Dominican amber. J. Bot. Res. Inst. Texas 6:1 23-1 27. Drapb '. G, P. Mann, and J.F. Lewis. 1 994. Hispaniola. In: S. Donovan and T.A. Jackson, eds. Caribbean geology: an introduc- tion. The University of the West Indies Publishers’ Association, Kingston, Jamaica. Pp. 129-150. itURRAiDE-ViNENT, M.A. and R.D.E. Macphee. 1966. Age and paleogeographic origin of Dominican amber. Science 273: 1850-1852. Mueuner ' A.N., V. Savolainen, R. Samuel, and M.W. Chase. 2006. The mahogany family 'out-of-Africa": divergence estimation, global biogeographic patterns inferred from plastid rbcL sequences, extant and fossil distributi diversity. Molec. Phylogen. Evol. 40:236-250. AJU.D. Pennington, A.B. Koecke, and S.S. Renner. 2010. Biogeography of Cedrela (Meliaceae, Sapindales) in Ce tral and South America. Amer. J. Bot. 97:51 1-518. p INGT0N ' ID - B.T. Styles, and D.A.H. Taylor. 1981. Meliaceae. FI. Neotropica 28:1 -470. G -°” jR - R- Poinar. 1 999. The amber forest. Princeton University Press, Princeton, NJ. *** D - 1999- Das Bernstein-Kabinett. Stuttgarter Beitr. Naturk. Ser. C, 28. 566 BOOK REVIEW Scott Calhoun. 2012. The Gardener’s Guide to Cactus: The 100 Best Paddles, Barrels, Columns, and Globes. (ISBN-13: 978-1-609469-200-6, pbk.). Timber Press, Inc., The Haseltine Building, 133 S.W. Second Avenue, Suite 450, Portland, Oregon 97204-3527, U.S.A. (Orders: www.timberpress.com, 800- 327-5680). $24.95, 227 pp„ color throughout, 8” x 9”. Thinking of going to the Desert Botanical Garden in Phoenix, Arizona? If so, now is the time to read Scott Calhoun’s The Gardner’s Guide to Cactus: The 100 Best Paddles, Barrels, Columns, and Globes. The juxtaposition of Dale Chihuly’s innovative glass sculptures with cactus and succulents was a stroke of genius. Likewise is Mr. Calhoun’s new cactus guidebook. Succulent plant expert Scott Calhoun has selected 100 of the best cactus available and shows how they can make striking additions to desert gardens or serve as dazzling container specimens. His writing style is one which offers a clean, helpful, well-presented text with reliable design suggestions. Mr. Calhoun’s stunning photograph on the book cover clearly demonstrates to the book publishing world and its gardening readership how easily cactus appreciation can be for everyone— not just specialists. He is the quintessential model of cactus authorship. It is a great book for beginners and expert gardeners. Paddles, barrels, columns, and globes add new meaning to the cactus gardening neophyte! One look at the Tuna Colorado cactus in its purple high-winter coloration photograph at the Desert Botanical Garden sur- rounded by a purple Chihuly glass sculpture suggests Mr. Calhoun’s new book is set apart from normal cactus guidebooks. First of all, it is a total delight to look at! As an effort to simplify the issues related to plurals, Mr. Calhoun has chosen to call the plants cactus. He describes cactus conservation and defines Crassulacean Acid Metabolism (CAM). A glance at the Introduction and Table of Contents describes planting and care, cactus planted in the ground and containers, tools of the prickly trade, and the art of showing cactus. Mr. Calhoun not only describes how one plants and cares for cactus; he gives tips for staging cactus at shows sponsored by the Cactus and Succulent Society of America (CSSA). The 100 plants are grouped as follows: low and mounding, barrels and globes, paddles and rods, and columns. Each entry has a colorful photograph, scientific and common names, habitat, mature size, hardiness, flowering season, cultivation, design suggestions, and, yes, culinary value! In addition, the book has extensive cross-referenced index, bibliography, plant hardiness tables, cactus pests, cactus for special purposes, and selected cactus nurseries. One can tell after reading The Gardner’s Guide to Cactus: The 100 Best Paddles, Barrels, Columns, and Globes, cactus sagacity can be great fun and rewarding. When is your first CSSA entry to demonstrate what you have learned from Mr. Calhoun ?— Kay M. Stansbery, Ph.D., Library Volunteer, Botanical Res 76107-3400, U.SA. e of Texas, 1700 1 y Dr., Fort V I. Bot. R«. Inst Texas 6(2): 566. 2012 LIMITATIONS TO NATURAL PRODUCTION OF LOPHOPHORA WILLIAMSU (CACTACEAE) II. EFFECTS OF REPEATED HARVESTING AT TWO-YEAR INTERVALS IN A SOUTH TEXAS POPULATION Martin Terry KeeperTrout Bennie Williams Sul Ross State University Alpine, Texas 79832, U.SA. Cactus Conservation Institute P.O.Box 561 Alpine, Texas 79831, U.S.A. Cactus Conservation Institute P.O.Box561 Alpine, Texas 79831, U.S.A. Teodoso Herrera Norma Fowler P.O. Box 460346 San Antonio, Texas 78246, U.S.A The University of Texas at Austin Department of Integrative Biology C0930 1 University Station Austin, Texas 78712, U.SA. RESUMEN 568 Journal of the Botanical Research Institute of Texas 6(2) Lophophora williamsii (Lem. ex Salm-Dyck) J.M. Coult. (Cactaceae), known as peyote both in Spanish and in English, is a small cactus (rarely exceeding 10 cm in diameter) of northeastern Mexico and adjacent border areas of Texas. The aerial crowns of plants are approximately hemispherical in shape. Some plants are caespi- tose; i.e., they have multiple crowns arising from a single rootstock. The literature on the biology of this plant up to the mid-1990s is summarized by Anderson (1996), who first suggested that the species might be endan- gered by overharvesting (Anderson 1995). There is active commercial trade in the harvested crowns of peyote, which are collected and sold by li- censed distributors to the Native American Church (NAC) for religious use as protected by U.S. law. There is substantial concern that the rate of harvest of peyote from wild populations is not sustainable. Anecdotal re- ports by members of the NAC include descriptions of the decline or decimation of natural populations and a decrease in both the availability and the quality of peyote being offered for sale in the regulated peyote market (TH, pers. obs.). A number of papers in the scientific literature have described the decline of peyote in its native habitat, apparently due to overharvesting (Anderson 1995; Trout 1999; Terry & Mauseth 2006; Powell etal. 2008; Terry 2008a, b,c; Terry et al. 2011). Despite such reports involving both Texas and Mexican populations, the species is not (yet) considered in danger of extinction (NatureServe 2012; Fitz Maurice and Fitz Maurice 2009), except in Texas, where NatureServe determined it to be in the S4 (imperiled) category. The work of Terry et al. (2011) was the first experimental investigation of the effects of harvesting on peyote plants in situ. In that paper we reported the effects that were detectable two years after the initial harvest. The present report focuses on effects detectable four years after the initial harvest. MATERIALS AND METHODS The study site was described in Terry et al. (2011). Because of the multi-year duration of the ongoing study and the complexity of the study design, it is appropriate to provide a clear, detailed description of what was done to which plants, and when. At the start of the study, in March 2008, 100 L. williamsii plants that appeared not to have been previously harvested were individually numbered and tagged along a transect through the population. The number of crowns on each plant was counted and the horizontal diameter of each crown was measured. Fifty of these plants that were single-crowned were then harvested (i.e., the crown of each plant was cut off transversely at ground level and removed), and the other 50 plants (most but not all of which were single-crowned) were left unharvested as controls. The harvested crown of each plant in the harvested group was weighed, to determine the harvested fresh biomass obtained from each of these “virgin” plants. At the end of the second year of the study, in March 2010, all surviving plants from the original groups of 50 harvested and 50 control plants were located, the number of crowns on each plant was counted, and the diameter of each crown was measured. Then the 43 surviving plants in the harvested group were divided into two subgroups: 20 multiple-harvest plants and 23 single-harvest plants. All regrowth crowns were harvested from the 20 multiple-harvest plants, leaving these now twice-harvested plants without crowns (and thus with- out photosynthetic tissue) for the second time in two years. Reharvest at two-year intervals is typical in current commercial harvest (MT, pers. obs.). The harvested crown(s) of each multiple-harvest plant were weighed to obtain harvested fresh biomass at a second harvest. A comparison of harvested biomass between the 2008 and 2010 harvests was reported by Terry et al. (2011). The single-harvest plants were not reharvested, and the sur- viving plants of the 50 original control plants continued to serve as unharvested controls. At the end of the fourth year of the study, in March 2012, all surviving plants were again located, counted, and measured. In addition, all new regrowth crowns were again harvested and weighed from the 16 surviving plants in the multiple-harvest treatment. In summary, control plants have never been harvested, single-harvest All stalls lalyses were done with SAS 9.1 ( :, Cary, NC, USA). Teny et al., Harvesting effects on wild populations of Lophophora in Texas 569 RESULTS Survival. — Of the 100 plants of the initial (2008) census, 4 (2 control, 2 harvested) were dug up by feral hogs and were therefore dropped from all further analyses, leaving 96 plants. Of these 96 plants, 6 (1 control, 5 harvested) died before the second (2010) census. Ninety plants were still alive in 2010 (census 2): 47 control plants and 43 plants that had been harvested in 2008. Of these 43 surviving plants that had experienced one harvest, 23 were assigned to the single-harvest treatment and 20 were assigned to the multiple-harvest treat- Of the 47 control plants alive in 2010, 45 were still alive at census 3 in 2012. Nineteen of the 23 single- harvest plants (83%) and 16 of the 20 multiple-harvest plants (80%) were still alive in 2012. By 2012 the sur- vival rate of control plants from census 1 through census 3 was significantly higher than the survival rate of harvested plants over the same interval (94% [45/48] versus 73% [35/48], x 2 = 8.65, P = .0033; Fig. 1). Harvested mass .— Weights of all the crowns of a plant were summed to calculate harvested fresh mass per plant. Mass per crown was calculated for each plant by dividing its total mass by its crown number; these val- ues were then averaged for statistical analysis and for Figure 2. Average harvested mass per plant decreased from census to census: 44% between the first and second censuses and 32% between the second and third censuses (Fig. 2, solid line). The differences between harvests were significantly different from zero (paired t- tests: harvest 1 vs harvest 2: 14.0 g average difference, N = 20, t = 6.73, P < 0.0001; harvest 2 vs harvest 3: 6.9 g average difference, N = 16, t = 4.24, P = 0.0007). Plants initially responded to harvesting by increasing the average number of crowns per plant (Fig. 2, dashed line), although this increase was not sufficient to counterbalance the decrease in mass per crown (Fig.2, dashed and dotted line). After the second harvest, both the average number of crowns per plant and the average mass per crown decreased. Volume.— The above-ground volume of each plant was estimated by first estimating the volume of each trown as a hemisphere from its measured diameter: estimated volume = % n (diameter/2) 3 . The estimated volumes of all the crowns on the plant were then summed to estimate total plant above- ground volume. The estimated volume of each plant in 2012 was very closely correlated with its harvested fresh mass in 2012 (Fig. 3). Volumes were log-transformed before analysis of covariance (ANCOVA) to improve normality of the re- siduals. Volume at census 2 (2010) was used as a covariate. Treatments did not differ significantly in their sfopes: the slope of the relationship between log-transformed volume in 2010 (x-axis) and log-transformed volume in 2012 (y-axis) was the same for each treatment. Therefore the final ANCOVA model assumed equal Note that equality of slopes in a model fitted to log-transformed data does not imply that slopes will be near w hen untransformed data are graphed on a linear scale (e.g., Fig. 4). Estimated plant volume at census 3 (2012) was closely related to estimated plant volume at census 2; 74% the variation in the former was explained by variation in the latter amount. Treatment accounted for an ad- r°nal 10% of the variation among plants at census 3. The effects of the single-harvest treatment did not differ ,llose of the control (Scheffe contrast, Fj 75 = 1.33, P * 0.25), but each of these treatments differed signifi- cantly fro m the multiple-harvest treatment (Scheffe contrasts; control versus multiple-harvest: F 1J5 = 29.20, P < 0001; single-harvest versus multiple-harvest, F, 75 = 41.46, P < 0.0001; Fig. 4). In other words, the surviving ^-harvest plants were growing (on a logarithmic scale) about as fast as the surviving control plants be- , een 2010 and 2012, but they began the interval with much smaller sizes than the control plants. In contrast, mu ltiple-harvest plants were decreasing in size. For example, the final ANCOVA model predicts that a COntro1 P lant with a volume of 10.0 cm 3 , the average size of all 90 surviving plants at census 2, would have to 21.9 cm 3 and a single-harvest plant of the same size would have grown to 26.3 cm 3 (not significantly 1 erent from 21.9 cm 3 ), but a multiple-harvest plant with a volume of 10.0 cm 3 would have decreased slightly nsi ze, to 9.66 cm 3 . Regional harvesting trends in South Texas .— Annual peyote sales data covering the years 1986-2011 (Texas Anient of Public Safety, unpublished data) are presented in Figure 5. Although these figures do not in- 570 Journal of the Botanical Research Institute of Texas 6(2) census s. Solid line: control plants; dotted line: harvested plants. elude all sales of peyote (Terry et al. 2011), it is reasonable to assume that the number of buttons sold in the regulated trade is positively correlated with the total number of buttons harvested in the region of South Texas known as “the Peyote Gardens.” In 2011, the DPS-regulated peyote sales totaled slightly over 1.4 million but- tons, continuing the generally downward trend which such sales have followed since 1997. It is noteworthy that prior to the current decline there was a decrease in numbers of buttons sold during the late 1980s that appears to have corresponded to the historical decline in the available harvest of mature plants, followed in the early to mid-1990s by a marked increase in numbers of buttons sold when the proliferation of small regro buttons began to be harvested to meet the needs of the NAC. Anecdotal accounts from NAC meetings during the period of temporary increase in numbers of buttons noted the prevalence of fresh buttons as small as dira® (TH & KT, pers. obs.). The number of buttons sold in 2011 was the lowest for any year in the last quarter o a century. As the annual number of buttons sold has declined steadily since 1997, the price has shown a ma increase; the price per button is roughly equal to total sales (in U.S. dollars) divided by the number of buttons sold. DISCUSSION Effects of harvesting on plant survival and growth The negative effects of harvesting on survival may be delayed. The initial harvest did not significantly reduce survival during the first two years after the harvest (2008 to 2010; Terry et al. 2011), but its effects were higWy significant by 2012 (73% survival to the four-year time point in 2012 among plants harvested in 2008, vers® 94% among control plants). Any delayed effects of the second (2010) harvest on survival were not yet evide» in 2012. sub- file effects of precipitation may also be delayed. The six months preceding the 2010 census received su era 9 e fresh above-ground mass per harvested plant, number of crowns per plant, and average fresh mass pc wrage fresh mass per crown, an average was calculated for all cr r _ Ptots were used to calculate the values in this graph. Vertical bars: 1 standard error. n at each date. To calculate stantially more rain (32.4 cm October-March precipitation) than the six months periods preceding the other censuses (6.0 cm and 15.4 cm October-March precipitation preceding the 2008 and 2012 censuses, re- U.S. Department of Agriculture 2012), but the average size of control plants declined in the first in- terval and incre ased in the second (Fig. 4). However, it may be that in wetter years peyote experiences more “^Petition from other plants that have responded rapidly to the increased soil moisture. Harvesting also sig- m Cantl >’’ and strongly, affected plant growth rate and therefore plant size (Fig. 4). Each harvest reduced plant tom ratCS ThC 2008 harvest reduced the average growth rate of all harvested plants (Terry et al. 2011). The harvest of the multiple-harvest plants significantly reduced their growth rate below that of the single- est plants (harvested only in 2008) as well as below the growth rate of the never-harvested control plants / & 4). While the single-harvest plants and the controls had about the same growth rate between 2010 and -the single-harvest plants were so much smaller in 2010 (due to the 2008 harvest) that they remained . smaller than the control plants in 2012. Meanwhile, the multiple-harvest plants continued to decline in tween 2010 and 2012. Plant size and plant survival are usually highly correlated (Harper 1977), so we 1*^0 ^ continuing excess mortality of the multiple-harvest plants, infec . re are P r °bably several reasons why harvesting reduces growth rate, size and survival rate. Microbial out ft, 0 " ° f Ae ° PCn Wound crated by the act of harvesting the crown of a plant, for example, cannot be ruled Mon mectlanism that appears to be an inevitable consequence of harvesting is that of exhaustion due to n ged deprivation of solar energy. The crown, being the only aerial organ of the peyote plant, is the plant’s Journal of the Botanical Research Institute of Texas 6(2) only site of photosynthesis. Without photosynthesis, the plant cannot use solar energy to create and store car- bohydrates, and nutrient, carbon, and water uptake are greatly reduced. The crown is also the part of the plant that is always harvested for ceremonial use. When the crown is harvested— thereby becoming a button in the peyote trade — the plant’s ability to photosynthesize is ipso facto reduced to zero. The harvested plant then uses stored energy, nutrients, and water to regrow its above-ground biomass. If reharvesting occurs before the p has had time to rebuild its stored reserves from photosynthesis in its regrown above-ground tissue, it w come successively smaller at each harvest and eventually die. The reduced size and growth rate, and increase mortality, of harvested plants strongly support the hypothesis that a two-year cycle of harvesting of this spe- cies is too frequent for plant recovery. A sustainable frequency of harvesting would be low enough to allow a plant to fully regrow and to fully rebuild its supply of stored resources between harvests. We hope eventua } to be able to determine the maximum sustainable harvesting frequency. Whatever that frequency may be, our As is true of most plant species, the removal of the apical meristem (part of the harvested crown of pep te), in addition to stimulating regrowth, probably also de-represses axillary meristems, resulting in the forma- tion of multiple crowns. In the absence of the continual secretion by the apical meristem of the hormone t normally suppresses lateral branching (presumably auxin, based on Mauseth and Halperin 1975), one <* more) of the axillary meristems in the areoles on the subterranean portion of the stem is de-repressed and gins to form a new crown at the apex of a lateral branch which emerges from the subterranean stem and g roWS toward the surface of the ground. This phenomenon accounts for harvested plants of this species having ® crowns than unharvested plants (Fig. 2). It is a temporary phenomenon, however, because eventually a P la ^ s stored resources are exhausted by too frequent harvesting, causing the number of crowns per plant to e 573 census single-harvest multiple-harvest plants ► average - control plants * average - single-harvest plants * average - multiple-harvest plants buttons sold (millions) 574 annual numbers of buttons sold (in millions) •••&•• total sales (in thousands of US $) Fig. 5. Annual peyote sales by licensed distributors in South Texas from 1986 through 2011. crease. This stage appeared to be reached at the second harvest: the number of crowns per plant increased after the first harvest but declined after the second. The negative impact of harvesting will be greater if portions of the subterranean stem are also removed To avoid such damage to the harvested plants, in this study we used only best harvesting practices, viz., cutting the crown at its base, parallel to the surface of the ground. However, commercial harvesting practices may re- move a substantial portion of the subterranean stem along with the crown (Terry & Mauseth 2006). Such re- moval of subterranean stem tissue reduces the number of areoles available to initiate lateral branching, reduc- ing the number of new crowns that can be formed. In addition it removes even more resources from the plant, reducing the amount available for regrowth. Changes in harvest yield over time As a result of the effects of harvesting on plant size, the yield per plant of harvested biomass decreased after each biennial harvest, first by 44% and then by 32% (Fig. 2, solid line). If one includes mortality in these calcu- lations, the decrease in harvest yield is even more marked: the third harvest produced only 25% of the bioma* that the first harvest did. For example, if we had begun with 100 plants, the first harvest would have yield* 1 3125 g (i.e., 100 plants x 31.25 g/plant), the second harvest would have yielded 1547 g (100 plants x 0.895 sur- vival rate of harvested plants 2008-2010 x 17.27 g/plant), and the third harvest would have yielded 769 g (10° plants x 0.895 x 0.800 survival rate 2010-2012 x 10.73 g/plant), declines of about 50% per harvest. The absolute (as opposed to relative) decline was smaller in the second two-year period, but only because it began fro® 4 Terry etal., Harvesting effects on wild populations of Lophophora in Texas 575 lower baseline. These are exactly the effects on harvest yield to be expected if harvesting is occurring too fre- quently for plants to regrow and to rebuild their stored resources. The fact that harvested mass was even more closely correlated with estimated volume at census 3 (Fig. 3) than at census 2 (Fig. 3 in Terry et al. 201 1) may seem surprising in that the crowns of most adult peyote plants do not appear to have a true hemispherical shape, but rather the shape of half of an oblate sphere. However, many of the younger plants — and especially young regrowth crowns — do indeed have vertically extended crowns, and all peyote plants tend to expand vertically in response to rain (MT, pers. obs.), which would tend to balance out the more flattened shape of the adults and the flattening effect of drought, over time. Peyote shows many of the hallmarks of a classic case of unsustainable harvesting of a wild resource. First, the decline in total harvest combined with an increase in price/unit is characteristic of overharvested wild species (cf. Fig. 1 in Schippman et al. 2002). A declining number of wild plants is a likely explanation for the failure of the harvest to increase in response to the increase in unit price (because a declining population causes de- there are still individuals to be harvested (Hilbom & Walters 1992; Thurstan et al. 2010). Second, there are anecdotal reports of declining unit (button) size (TH, pers. obs.). Declining body size is another classic indica- tor of overharvesting (Stergiou 2002; Berkeley et al. 2004; Genner et al. 2010). Third, there are anecdotal re- ports of declining quality of the harvested buttons (TH, pers. obs.). Fourth, the harvesting frequency (every other year) shown to be unsustainable by the present study is typical. Finally, our results may underestimate impacts of harvesting, as our harvests may have been less damaging to individual plants than a commercial harvest, due to the care taken in the harvests of this study. As far as we are aware, this study is the first well documented case of overharvesting of a cactus species (but see Jimenez-Sierra and Eguiarte 2010, in which browsing was also involved). It is also one of a limited number of well documented cases of overharvesting of non-timber plant species in general. Most well docu- mented cases of overharvesting of wild resources involve marine and freshwater animal species (Jackson et al. 2001; Allen et al. 2005; Genner et al. 2010). There are detailed reports of overharvesting of many tree species (e g., Schwartz et al. 2002; Schulze et al. 2008). There are some detailed reports of overharvesting of herbaceous plant species, of which ginseng ( Patiax quinquefolius) is perhaps the best documented (Nantel et al. 1996; Mc- Graw 2001; Case et al. 2007; McGraw et al. 2010). However, many hundreds (at least) of other plant species are threatened by overharvesting, especially plant species harvested for medicinal uses (Schippman et al. 2002), for lumber (Oldfield et al. 1998), or for collectors (Oldfield 1997). The regulatory panorama At the moment there are only two major interested parties with any standing in the discussion about the fate of P^ote in its natural habitat: (1) the Native American Church (NAC), whose right to consume peyote for reli- cs purposes is protected by legislation such as the American Indian Religious Freedom Act (AIRFA), and (2) the Drug Enforcement Administration (DEA), which is obligated by the Controlled Substances Act to regulate ^ use and distribution of peyote by and for the NAC, and to prevent the diversion of peyote to non-authorized Persons. Neither of these parties is speaking very audibly about regulatory solutions to mitigate the deteriorat- ln g state of the wild peyote populations. This is unfortunate, as the problem has a feasible solution, namely the regulated cultivation of peyote by and for the NAC, which would reduce the harvesting pressure on the wild Populations (as in, e.g., Kay et al. 2011). Furthermore, this solution is technically within reach (Chandra et al. 311(1 oulturally acceptable (TH, pers. obs.). The barrier to bringing this solution to fruition is essentially a regulatory one. Cultivation of L. williamsii is anticipated in the American Indian Religious Freedom Act (as tended 1994), which “. . .does not prohibit such reasonable regulation and registration by the Drug Enforce- Administration of those persons who cultivate. . .peyote. . ..” But to date no interested party (e.g., the NAC ° orth America) has petitioned the DEA to promulgate any such “reasonable regulation” spelling out the QClaik for SUc h registration. Pending such action, cultivation of peyote, though not illegal, lacks the needed te ** ulator y framework to provide legal certainty and protection for NAC members who would prefer to produce populations of peyote continue to produce steadily decreasing yields, as demonstrated in this study and in the regulated peyote market. Under the current system— which can accurately be described as “management by extirpation” — at some point the conservation crisis will become so critical that the U.S. Fish and Wildlife Ser- vice will be obligated by the terms of the Endangered Species Act to evaluate the conservation status of the species Lophophora williamsii. At that point the regulatory situation will become substantially more complex. If a regulatory stalemate then ensues, the NAC’s options may broaden (or narrow) to include the Supreme Court and/or Congress as sources of relief. ACKNOWLEDGMENTS We are most grateful to C.W. Hellen Ranches, Ltd. - La Mota Division - Charles W. (Bill) Hellen, Managing Partner, for providing access to his ranch and good company. We also thank the younger generations of the Herrera and Terry families for their help with the hard labor involved in the logistics of conducting the study. Garry Stephens kindly provided USDA Field Office climate data for Hebbronville, Texas. Essential funding for the study was generously provided by Libbie and Jerald Mize, the Alvin A. and Roberta T. Klein Foundation, Sul Ross State University (in the form of a Research Enhancement grant), and all the donors supporting the scientific work of the Cactus Conservation Institute, Inc. We appreciate the helpful comments of Michael Pow- Allen, J.D., Ft Abell, Z.E.B. Hogan, C. Revenga, B.W. Taylor, R.L. Welcomme, and K. Winemiller. 2005. Overfishing of inland waters. Bioscience 55:1041-1051. Anderson, E.F. 1995. The "Peyote Gardens' of South Texas: A conservation crisis? Cact. Succ J. (U.S.) 67:67-73. Anderson, E.F. 1996. Peyote: The divine cactus. University of Arizona Press, Tucson. Berkeley, SA., MA Hixon, R J. Larson, and M.S. Love. 2004. Fisheries sustainability via protection of age structure and spatial distribution offish populations. Fisheries 29:23-32. Case, M.A., K.M. Funn, J. Jancaitis, A. Alley, and A. Paxton. 2007. Declining abundance of American ginseng ( Panaxquinque- folius) documented by herbarium specimens. Biol. Conservation 134:22-30. Chandra, B„ L Palni, and S. Nandi. 2006. Propagation and conservation of Picrorhiza kurrooa Royle ex Benth.: an endan- gered Himalayan medicinal herb of high commercial value. Biodivers. & Conservation 1 5:2325-2338. Fnz Maurice, B. and W.A. Fitz Maurice. 2009. Lophophora williamsii. In: IUCN, ed. 201 2. IUCN Red List of Threatened Species Version 201 2.1 . . Accessed: 28 June 201 2. Genner, M J., D.W. Sims, AJ. Southward, G.C. Budd, P. Masterson, M. McHugh, P. Rendle, EJ. Southall, VJ. Wearmouth, and SJ. Hawkins. 2010. Body size-dependent responses of a marine fish assemblage to climate change and fishing over a century-long scale. Global Change Biol. 16:51 7-527. Harper, J. 1977. Population biology of plants. Academic Press, London. e, dynamics, and uncertainty. Chapman Jackson, J.B.C., M.X. Kirby, W.H, Berger, K A Bjorndal, L.W. Botsforo, B J. Bourque, R.H. Bradbury, R. Cooke, J. Erlandson, J A Es®> T.P. Hughes, S. Kidwell, C.B. Lange, H.S. Unihan, J.M. Pandolfi, C.H. Peterson, R.S. Steneck, M J. Tegner, and R-R. Warner- 2001- Historical overfishing and the recent collapse of coastal ecosystems. Science 293:629-637. JimEnez-Sierra, C. and L Egularte. 2010. Candy barrel cactus ( Echinocactus platyacanthus Link & Otto): a traditional P^ Kay, J. r A. Strader, V. Murphy, L. Nghiem-Phu, M. Calonje, and M.P. Griffith. 201 1 . Palma Corcho: a case study in botanic garde*’ conservation horticulture and economics. HortTechnol. 21:474-481. Mauseth, J.D. and W. Halperin. 1 975. Hormonal control of organogenesis in Opuntia polyacantha. Amer. J. Bot. 62:869-877. Terry et al„ Harvesting effects on wild populations of Lophophora in Texas S77 McGkaw, J.B., S. Souther, and A.E, Lubbers. 201 0. Rates of harvest and compliance with regulations in nat of American ginseng (Panax quinquefolius L.). Nat. Areas J. 30:202-210. Nantel, P, D. Gagnon, and A. Nault. 1996. Population viability analysis of American ginseng and wild l< stochastic environments. Conservation Biol. 10:608-621. NatureServe. 2012. NatureServe Explorer: An online encyclopedia of life [web application]. Version 1 Arlington, Virginia. Available at http://www.natureserve.org/explorer (Accessed: July 3, 201 2.) Outfield, S. (ed). 1997. Cactus and succulent plants— Status survey and conservation action plan. IUCN/SSC Cactus and Succulent Specialist Group. IUCN, Gland, Switzerland and Cambridge, UK. OoFfiD, S., C. Lusty, and A. MacKinven. 1 998. The world list of threatened trees. World Conservation Press, Cambridge, UK. Powell, A.M., J.F. Weedin, and S.A. Powell. 2008 . Cacti of Texas: a field guide. Texas Tech University Press, Lubbock. Schippmann, U„ DJ. Leaman, and A.B. Cunningham. 2002. Impact of cultivation and gathering of medicinal plants on biodiver- sity: global trends and issues. In: Departmental Working Group on Biological Diversity for Food and Agriculture, FAO (ed). Biodiversity and the Ecosystem Approach in Agriculture, Forestry and Fisheries. Satellite event on the occasion of the Ninth Regular Session of the Commission on Genetic Resources for Food and Agriculture. Rome, 1 2-1 3 Octo- ber 2002. FAO, Rome. Available at http://www.fao.org/DOCREP/005/AA010E/AA010E00.HTM Schulze, M, J. Grogan, C. Uhle, M. Lenttni, and E. Vidal 2008. Evaluating ip# (Tabebuio, Bignoniaceae) logging in Amazonia: sustainable management or catalyst for forest degradation? Biol. Conservation 141 :2071 -2085. Schwaru, M.W., T.M. Caro, and T. Banda-Sakala. 2002. Assessing the sustainability of harvest of Pterocarpus angolensis in Rukwa Region, Tanzania. Forest Ecol. Managem. 170:259-269. Stehgiou, K.1. 2002. Overfishing, tropicalization of fish stocks, uncertainty and ecosystem management: resharpening Ockham's razor. Fish. Res. 55:1-9. Terry, M. 2008a. Stalking the wild Lophophora. Part 1 . Chihuahua and Coahuila. Caet. Succ. J. (US) 80:1 81 -1 86. Terry, M. 2008b. Stalking the wild Lophophora. Part 2. Zacatecas, San Luis Potosi, Nuevo Leon, and Tamaulipas. Cact Succ. J.(U.S.) 80:222-228. Terry, M. 2008c. Stalking the Wild Lophophora. Part 3. San Luis Potosf (central), Quer#taro, and Mexico City. Cact. Succ. J. (US.) 80:310-31 7. Terry, M. and J.D. Mauseth. 2006. Root-shoot anatomy and post-harvest vegetative clonal developnr williamsii (Cactaceae: Cacteae): implications for conservation. Sida 22:565-592. Terry, M, K. Trout, B. Wiluams, T. Herrera, and N. Fowl£r. 201 1 . Limitations to natural production of Lophophora willi (Cactaceae) I. Regrowth and survivorship two years post harvest in a South Texas population. J. Bot. Res. Inst 5:661-675. Texas Department of Public Safety. Peyote sales data. Unpublished. "f™"' R - H - s - Brockington, and C.M. Roberts. 2010. The effects of 118 years of industrial fishing on UK bottom fisheries. Nature Commun. 1 : 15 . Jout, K. 1999. Sacred cacti. Second Edition. Better Days Publishing, Austin. ■5. Department of Agriculture. 2012. USDA Field Office Climate Data for WETS Station at Hebbronville, Texas. 578 BOOK REVIEW Ellen Sousa, Foreword by William Culuna. 2011. The Green Garden: A New England Guide to Planning, Planting, and Maintaining the Eco-Friendly Habitat Garden. (ISBN-13: 978-1-59373-091-8, pbk). Bunker Hill Publishing, 285 River Road, Piermont, New Hampshire 03779, U.S.A. (Orders: www.bun- kerhillpublishing.com). $34.95, 224 pp., color throughout, 7" x 10". A growing number of people are becoming more agricurious. They are interested in how things are grown, land stewardship, animal husbandry, and soundscape ecology. Thus, planning an eco-friendly habitat garden and landscape is like becoming a creative gourmet cook orchestrating a perfect dinner party with tried and tested recipes. Ellen Sousa’s The Green Garden: A New Eng- land Guide to Planning, Planting, and Maintaining the Eco-Friendly Habitat Garden is the ideal manual/reference book to use if one wants to establish a notable green garden, particularly for those wanting to stop the deteriora- tion of the natural world around us. Ms. Sousa presents new concepts for gardening that allow the gardener to work with nature to develop an area’s natural ability to maintain its own health and sustainability. Her cogent writing style is a true testimonial that she is a cognoscente of natural habitat gardening. She and her husband live on a farm landscaped as a Certified Wildlife Habitat and Monarch Waystation. Renowned native plant au- thor William Cullina’s Foreword in the book is a validation that Ms. Sousa is the real deal! Even though the title suggests it is only about the New England area of the country, it is not. Remember, the fight for establishing eco-friendly habitats is a universal challenge. Fifty-plus contributing photographers fill this book with amazing wildlife photos and native plants. It is not only a wonderful primer for those just learning about natural habitat gardening but also a valuable educa- tional resource for seasoned gardeners seeking to increase their eco-friendly habitat gardening knowledge, A quick glance at the Table of Contents denotes the principles Ms. Sousa cites are applicable from forests to seasides, from lots to farmlands and, yes, to what Victorian gardeners called a “moon garden” of fragrant tese are for gardeners who are only able to enjoy their garden during the evening In addition, the book includes a guide of the best plants for the region. The plants are grouped by plant type in alphabetical order by genera. Common names are listed after each plant genus name. A key of numer- ous icons is included. Thoreau wrote “the world is but a canvas for your imagination” — when are you going to begin developing your own Eden?— Kay M. Stansbery, Ph.D., Library Volunteer, Botanical Research Institute of Texas, 1700 Univer- sity Dr, Fort Worth, Texas 76107-3400, U.S.A. J. Bot Res. Inst. Texas 6(2): 57S. 2012 A COST-EFFECTIVE METHOD FOR CONSTRUCTING MAGNETIC FUMICELS FOR HERBARIUM CABINETS Richard Carter Director, Herbarium (VSC), Biology Department Valdosta State University Valdosta, Georgia 31698-0015, U.S.A. ABSTRACT RESUMEN INTRODUCTION Substantial literature exists on the importance and efficacy of various pest repellants in herbaria and their health risks (e.g.. Hall 1988; Strang 1999). However, methods for containing and securing repellants such as naphthalene and paradichlorobenzene (PDB) in herbarium cabinets have received little attention. A variety of methods are used for containment in herbaria, e.g., cloth bags with drawstrings, paper envelopes, small paper bags, small cardboard boxes, or small compartments in cabinet doors, and some entomologists use fumicels constructed from small cardboard boxes with wire mesh tops (Anonymous 2012). Because naphthalene and PDB vapors are heavier than air and tend to settle at the bottom of the cabinet, the repellant is normally placed 011 t0 P uppermost folder of specimens in the uppermost pigeonhole of each column of the herbarium ^net. However, there are several annoying aspects of this widely used method. When specimens are re- trieved from the uppermost pigeonholes, the packets of repellant fall out, often striking the individual remov- mg the specimens, or the packets are inadvertently pushed toward the back of the pigeonhole where they slip °w of sight. Retrieving the “lost” packets usually requires use of a ladder or step-stool, which is time-consum- wg and potentially hazardous. I devised, tested, and recommend placement of solid repellant in fumicels constructed from acid-free 8“Ssine envelopes suspended from the metal ceilings of herbarium cabinets by small neodymium magnets ed to small rectangles cut from herbarium paper. This method uses archival materials, is cost effective, a «d— except for magnets and glassine envelopes — employs materials commonly available in the herbarium. MATERIALS AND METHODS United States, neodymium magnets and glassine envelopes are normally sold by English units; thus, x 8 9 UnitS are P rilT tarily employed here. Paper magnet-holders were made from 2 Vi inch x 3V4 inch [6.4 cm Thec Cm ] P a P er rectangles cut from 0.12 caliper archival herbarium paper using a guillotine type paper cutter. ‘^terofeach rectangle was marked usinga wooden pencil, and a V4 inch x Vie inch [6.4 mm x 1.6 mm] N42 ^ 1Um (NdFeB) disc magnet was affixed at the center with a small dot of water-base herbarium glue. The ^ ve was allowed to dry, and the paper magnet-holder was placed in a 2)4 inch x 3)4 inch [7 cm x 9.5 cm] matehT’ glassirie envelope with a single opening along one 2)4 inch end and a foldable end flap. Approxi- Og of naphthalene (three Enoz® moth-balls) was placed in the packet against the surface of the paper Vnolder opposite the magnet. The envelope was closed, and a 1V+ inch stainless steel paper-clip was 3.75 inches placed across one corner to secure the flap. See Figs. 1, 2 and 3. Alternatively, the loose flaps can be taped closed, but this would most likely prevent reuse of envelopes. Glassine paper is relatively non-porous and ini' pervious. Therefore, 1 punctured the lower surface of each gWw envelope about 20 times with a steel sewing needle to increase its permeability to naphthalene. Magnets.— % inch x Vis inch [6.4 mm x 1.6 mm] Grade N42 Neodymium (NdFeB) Disc Magnet, 3x nickel plated, pull force 2.5 lbs: MAGNETman® via amazon.com®. http://www.amazon.com/shops/A2WOM254L' CFKNF Glassine envelopes.— Vh inch x Vh inch [7 cm x 9.5 cm] acid-free, glassine envelope with foldable end Ihp and single opening along 2Y4 inch end: Uline®, 12575 Uline Drive, Pleasant Prairie, W1 53158, USA. h«F RESULTS AND DISCUSSION The major expense was the cost of the neodymium magnets ($0.24 each, including shipping; pack of 1°°** $12.99, excluding shipping), the acid-free, glassine envelopes ($0.07 each, including shipping; box of lOOOfor jft-OO, excluding shipping), and stainless steel paper-clips ($0.07 each, including shipping; box of 500 for 9.95, excluding shipping). The cost of the other, commonly used materials was negligible. Thus, I equipped rtwrium cabinets, each with two fumicels, at a cost of less than $0.40 per fumicel. Although 1 recommend siting the fumicels to ceilings of cabinets, they also seem to work well on cabinet doors, which might be a consideration when the uppermost pigeonholes are packed with specimens. j. ^ ve t0 l hcir size, neodymium magnets are very strong, and they are readily available from several on- vendors. Marketed as permanent, they are subject to chipping if allowed to snap together. Although I used ^ 1 10g of repellant per fumicel, these small magnets are sufficiently strong to suspend at least 20g of k Kpellant (Le., six moth-balls and the maximum capacity of the fumicel envelope) securely, and they can ^easily detached from the herbarium cabinet. In addition to being dangerous, larger, stronger neodymium C al^ CtS 310 m ° re ex P ens * ve and would require considerable effort to detach from the metal surface of the Caching them would most likely result in tearing both glassine envelope and paper mag- «ver magnet £ rade and size recommended herein seem ideal for this particular application. How- iUm ma S nets c ome in a variety of sizes and shapes, and heavier containers could easily be at- e Wlth larger, stronger magnets. This article should not be construed to be a recommendation by either the author or this journal th at ^ g h n ^ hthalene ’ PDB » or other pest repellants in herbaria. Rather, its purpose is to provide information * beneficial t0 th °se who find it necessary to use repellants or other solid materials such as desic- 3 Patt ofthe overall program for herbarium pest management. ACKNOWLEDGMENTS ^ assistant Zachary J. Buning is gratefully acknowledged. The constructive criticism of Wendy B. Zorn- 5g2 Journal of the Botanical Research Institute of Texas 6(2) lefer (GA), Kent D. Perkins (FLAS), and Melanie Link-Perez (AASU) improved this paper. This work was sup- ported in part by NSF DBI-1054366 (J R- Carter, PI). Anonymous. 2012. Australian Entomological Supplies Pty. Ltd. http://www.entosupplies.com.au (18 January 2012). Hail, A.V. 1988. Pest control in herbaria. Taxon 37:885-907. Stoang, TJ.K. 1999. A healthy dose of the past: a future direction in herbarium pest control? In: DA. Metsger and S.C. Byers, eds. Managing the modern herbarium, an interdisciplinary approach. Society for the Preservation of Natural History Collections (SPNHC). Washington, D.C. Chapter 3:59-80. NUEVOS REGISTROS DE POACEAS PARA EL NORTE DE MEXICO Y. Herrera Arrieta 1 , C.A. Silva Salas, L. Ruacho Gonzalez y 0. Rosales Carrillo Instituto Politicnico Nacional, CIIDIR Unidad Durango-COFAA Sigma 1 1 9, Fracc. 20 de Noviembre II, Durango, Dgo. 34220, MEXICO ABSTRACT & Columbus, is ex Durante el desarrollo de estudios de biodiversidad (floristica y citologica), cuyo objetivo principal fue realizar lesisde Maestrfa en el IPN CIIDIR Durango, sobresale el hallazgo de localidades no registradas con anteriori- para dos especies de gramineas (familia Poaceae), familia reconocida por su importancia ecologica y gran calidad alimenticia de muchas de sus especies (ejemplo cereales, forrajes, cana de azucar). Se encontro que la •ocalidad de colectas recientes en la vegetation de cimas del cerro Huehuento, San Dimas, Durango expande el area de distribucion conocida de Muhlenbergiafiliculmis Vasey (taxon de distribucion boreal conocido de las nontafias del W de Estados Unidos), resultando ser este e 1 primer registro para Mexico. As! tambien, colectas recientes en la Laguna de Salitrillo, Zacatecas expanden el area de distribucion de Distichlis eludens (Soderstr. &H.F. Decker) H.L. Bell & Columbus, taxon escaso y endemico de suelos halofilos en terrenos inundables del “°ne-centro de Mexico, cuya distribucion conocida se restringia a dos localidades disyuntas: Humedal de ga, Durango y Salinas de Hidalgo, San Luis Potosi, siendo la nueva localidad un punto intermedio entre las Baades an, ertormeme conocidal. METODOS esludia r°n ejemplares de los taxa incluidos, provenientes de colectas recientes, realizadas en las areas de n^ribucion registradas en el reporte presente. La determination de la identidad de las especies se llevb a cabo rzando claves y descripciones de diversas obras floristicas para Norteamerica (Hitchcock & Chase 1951) y ^“nas regiones de Mexico (McVaugh 1983; Herrera 1998; Herrera 2001; Herrera y Pamanes 2006; Herrera y RESULTADOS ^identifiearon ejemplares con el nombre de Muhlenbergia filiculmis de la vegetacibn de alta montana en el W range y de Distichlis eludens proveniente de la laguna del Saladillo, Zacatecas. Ambas localidades am- n a distribucion conocida de estas especies. ** 3ra ^ x * co y P ara e * f stado de Durango n r g»a filiculmis Vasey es una especie que habita en lugares templados frios de los 1 ** s - Inst Teas 6,2): 583 _ 5g$ j 01 2 bosques de 584 Journal of the Botanical Research Institute of Texas 6(2) conlferas de las montanas del NW de Estados Unidos, entre los 2500 y 3300 m de altitud, con distribution conocida en Arizona, California, Colorado, Nuevo Mexico, Utah, Washington y Wyoming (Herrera, 1998). Es una especie perenne pero de porte pequeno, de 10 a 35 cm de alto, por lo que pudo haber pasado desapercibida a los ojos de colectores anteriores, sumado a ello la caracteristica de que es apetecida por los herbivorosyse encuentra generalmente ramoneada. Es una especie del complejo aparentemente monofiletico de Muhlenberg montana (Nutt.) Hitchc., propuesto por Herrera (1998), cuya caracteristica morfologica principal es tenerla segunda gluma 3-nervada; siendo ahora reconocido como un linaje natural (subgenero Clomena), con base a el andlisis filogenetico del ADN (Peterson y col. 2010). Las colectas de material botanico en la vegetation de alta montana de la Sierra Madre Occidental son es- casas, especialmente en el municipio de San Dimas, Durango, donde lo inaccesible del terreno hace quesean pocos los colectores que se internan en sus caminos para realizar estudios floristicos. La localidad del ejemplar de M.filiculmis [Mexico. Durango. Municipio San Dimas: Cima del cerro Hue- huento, 24°04’3rN, 105°44’24”W, 3262 m, M. Gonzdlez 4288A (CIIDIR)] es nueva para Mexico y se constituye como el extremo septentrional de distribution para esta especie, a un limite altitudinal sobre los 3200 m, misma que sale del limite sur conocido hasta ahora, cerca del paralelo 32° N en Arizona y Nuevo Mexico hasta cerca de 24° N correspondientes a la nueva localidad en San Dimas, Dgo. Podria sospecharse la existenciade otras localidades en que posiblemente se encuentre creciendo esta especie, sin embargo dichas localidades debieran ser en las altas montanas de la Sierra Madre Occidental, a fin de compensar las diferencias de latitud con altitud que permite se presenten los requerimientos climaticos del habitat de la especie. Las localidades de alta montana en Mexico son escasas. Challenger (1998) marca 18 picos altos con vegetation alpina o subalpina, 10 de ellos en el Eje Volcanico Central, 7 en la Sierra Madre Oriental y 1 en la Sierra Madre Occidental (el cerro Mohinora). Sin embargo en la Sierra Madre Occidental, se encuentran otros picos que quizes no Uegan a tener vegetacidn alpina tipica, pero que presentan elementos de vegetacion alpina por la alti- tud que alcanzan, como es el caso de los cerros siguientes: Cerro Mohinora (3,293 m), mpio. Guadalupe y Calvo, Chih.; Cerro Barajas (3,280 m), mpio. Guanacevi, Dgo.; Cerro Huehuento (3,233 m), mpio. San Dimas, Dgo., donde se encontrd M. filiculmis para este registro; Cerro Gordo (3,328 m), mpio. Pueblo Nuevo, Dgo.; Cerro de las Antenas (3,221 m), mpio. Canatlan, Dgo. Ruacho (2011), inicio muestreos y dejo marcas enlas localidades de algunas de estas montanas (en aquellas donde la seguridad actual lo permitio), con el fin de re- alizar monitoreo para ver cambios posteriores en la vegetacion de la Sierra Madre Occidental. II. Nuevo registro para el Estado de Zacatecas El gdnero Distichlis, fue propuesto por Rafinesque en 1819 (L6pez y col. 2009). Se le conoce vulgarmente como “pasto salado” ya que habita en suelos naturalmente salitrosos o salinizados artificialmente, desde elevadas montaiias hasta la orilla del mar, se destaca en la clasificacion sintaxonomica, debido a la relation tan estrecha que tiene con los ambientes en los que habita (Pelliza y col. 2005). Generalmente, estas plantas permanecenen estado verde durante el periodo de sequia, al tiempo que el resto de los pastos forrajeros en el sitio no estan disponibles, por esta razon el ganado lo consume (Estrada-Hemandez y col. 2005). Un nivel elevado de salinidad en el suelo aumenta el punto de marchitamiento de las plantas como conse- cuencia del incremento de la presion osmotica y de los efectos toxicos de las sales (Proyecto SoCo 2009). U salinidad puede llegar a ser un factor importante en la reproduction de las especies de Distichlis, lo que hace que su estudio sea de singular importancia (Proyecto SoCo 2009). Distichlis esta formado por un pequeno grupo de plantas herbaceas, dioicas, perennes y rizomatosas, con espiguillas multifloras y hojas conspicuamente disticas, pertenecen a la familia Poaceae, subfamilia Chloridoi- deae, tribu Cynodonteae y subtribu Monanthochloineae (Lopez y col. 2009). La taxonomia de este grupo, fue definida recientemente por Bell y Columbus (2008) con base en caracteres de la topologia nuclear y del cloro- plasto, adicionales a los caracteres morfolbgicos y anatomicos del grupo. Concluyendo que los generos MoruW- thochloc y Rcederochloa estan anidados dentro de un Distichlis parafiletico, conformado por unas siete especi* distribuidas en sitios salinos del Hemisferio Occidental y Australia (Bell y Columbus 2008, Barkworth y BeU 2011), cinco de esas especies se encuentran en Mexico. Distichlis eludens (Soderstr. & H.F. Decker) H.L. Bell & Columbus (antes Reederochloa eludens Soderstr. & H.F. Decker) (Fig. 1) es una planta perenne, dioica, estolonifera que se-desarrolla en zonas de alta concen- tration de sales en pastizales halofilos o en claros de matorral xerofilo y requiere de altos niveles de agua para su crecimiento. Se encuentra en cualquier epoca del ano aunque este marchita y es por ello que el ganado la consume. Es una planta que en ocasiones es dificil de encontrar por el tamaiio tan pequeno que presenta y por elgrado de ramoneo que el ganado ejerce sobre ella, asi tambien en otras ocasiones, se puede Uegar a observar facilmente ya que la densidad de individuos que forman una colonia puede ser hasta de 10 m continuos. La especie se consideraba endemica de las dos localidades disyuntas, en los Estados de Durango y San Luis Potosi, d ejemplar recientemente colectado [Mexico. Zacatecas: orilla SE de la Laguna el Saladillo, 22°38’38.7 , 'N, 102°02'24.9"W, Silva, Carrillo y Herrera 29 (CIID1R)], amplia el area de distribution conocida para este taxbn alEstado de Zacatecas, region intermedia a las de distribution conocida. El estudio citologico en poblaciones de D. eludens realizado por Silva (2011), proportion un ideograma del cariotipo de esta especie, confirma el numero cromosomico (x=9) y la aneuploidla (2n=4x+2=38) reporta- dos anteriormente, con resultados novedosos adicionales que se reportaran en un artlculo en proceso. CONCLUSION U revi si6n y colecta de material botanico con fines de estudios de biodiversidad floristica, citologica, ecologica 0 de otra indole, siguen aportando information novedosa y relevante para el conocimiento de nuestra flora ®exicana. Se dan a conocer a la comunidad botanica localidades nuevas de dos importantes gramlneas que se rep °rtan por primera vez: Muhlenbergiafiliculmis nuevo registro para Mexico y Distichlis eludens nuevo registro P® 3 el Estado de Zacatecas. AGRADEC1MIENTOS %adecemos a la Secretaria de Investigation y Posgrado del 1PN por el financiamiento recibido en los proyec- ^ S ' ^E-20110017 “Citologia de Gramineas y Ciperaceas de importancia economica del n-c de Mexico y S1P2 °1 10681 “Fitodiversidad y Ecosistemas de la Sierra Madre Occidental”. Asi tambien agradecemos al CONACyT por las becas de estudios de posgrado a la 2 a (VU: 331664) y 4° coautores. Por ultimo agradecemos a J J. Ortiz y un revisor anonimo por las observations y sugerencias al manuscrito del presente reporte. 586 Barkworth, M.E. and H.L. Bell. 201 1 . 1 7.04 Distichlis Raf. http://herbarium.usu.edu/treatments/Distichlis-1 Bell, H.L. and J.T. Columbus. 2008. Proposal for an expanded Distichlis (Poaceae: Chloridoideae): support morphological, and anatomical characters. Syst. Bot. 33:536-551. Challenger, A. 1998. Utilizacton y Conservacton de los Ecosistemas Terrestres de Mexico, pasado, presente y futuro. CONABIO (Comiston Nacional para el Conocimiento y Uso de la Biodiversidad)-UNAM Universidad Autonoma de M6xico)-Agrupaci6n Sierra Madre, S.C., Mexico. Estrada-HernAndez, A., E. Troyo-Dieguez., J.L GarcIa-HernAndez, H. HernAndez-Contreras, B. Murillo-Amador, y R. LOpez-Aguiuul 2005. Potencial forrajero del pasto salado Distichlis spicata (L.) Greene en ecosistemas costeros de Baja California Sur, Mexico porel m&odo de"componentes principales.'Tec. Pecuaria Mexico 43(1):1 3—25. Herrera A., Y. 1 998. A reviston of the Muhlenbergia montana (Nutt.) Hitchc. complex (Poaceae: Chloridoideae). Brittonia 50:23-50. Herrera A., Y. 2001 . Las Gramfneas de Durango. Institute Politecnico Nacional (IPN)-CONABIO, Mexico. Herrera A., Y y D.S. PSmanes G. 2006. Guia de pastes para el ganadero del Estado de Durango. IPN-COCyTED-Fundac6n Produce, Durango, A.C. Hitchcock, A.S. y A. Chase. 1951. Manual of the grasses of the United States. 2nd ed. USDA Misc. Publ. 200. Herrera A., Y„ P.M. Peterson, y A. Cort£s 0. 201 0. Gramfneas de Zacatecas, Mexico. Sida, Bot. Misc. 32:1 -239. Lopez S., Ma.M., S.D. Koch, M. Flores-Cruz y E.M. Engleman. 2009. Anatomfa comparada de la lamina foliar del genera Disti- chlis (Poaceae). Acta Bot. Mexicana 89:1-23. McVaugh, R. 1983. Gramineae. In: Anderson, W.R., ed. Flora Novo-Galiciana Vol 14. University of Michigan Press, Ann Peluza, A., L. Borrelu y G. Bonvasuto. 2005. El pasto Salado ( Distichlis spp.) en la Patagonia: una forrajera adaptada a la aridezy a la salinidad. Facultad de Ciencias Agropecuarias. Rev. Ci. Agropecuaria 9(2):1 19-131. Peterson, P.M., K. Romaschenko, and G. Johnson. 2010. A phytogeny and classification of the Muhlenbergiinae (Poaceae: Chloridoideae: Cynodonteae) based on plastid and nuclear DNA sequences. Amer. J . Bot. 97:1532-1554. Proyecto SoCo. 2009. Agricultura sostenible y conservation de los suelos, procesos de degradation del suelo. Ficha informativa n°4: Salinidad y Sodificacton. Ruacho G., L 201 1. El elemento alpino en la vegetation de cimas de la Sierra Madre Occidental. Tesis de Maestrfa, IPN CIIDIR Dgo., Mexico. &lva S., C.A. 201 1 . Citologfa de especies de gramfneas en el norte-centro de Mexico. Tesis de Maestrfa, IPN CIIDIR Dgo, distribucion actual y potencial DE TAXUS GLOBOSA (TAXACEAE) EN MEXICO Mario A. Garda-Aranda, Cesar Cantu-Ayala, Eduardo Estrada-Castillon, Marisela Pando-Moreno y Antonio Moreno-Talamantes Facultadde Ciencias Forestales Universidad Autdnoma de Nuevo Ledn Km 145 Carr. Nacional, Apartado Postal 41 C.P. 67700, Unares, N. L, MEXICO RESUMEN Clave Taxus globosa, Modelacion, Habitat, MaxEnt ABSTRACT NOM-059-SEM ARNAT-20 1 0 , a modeling distribution procedure of maximum entropy or MaxEnt, based on 40 occurrences of field records REM1B-CONABIO records. Two different scenarios of geographical and environmental coverages— climatic and topographic— were Proved statistically. The BIOCL1M variables scenario with topography data show a more accurate model in the range of 0.9 to 1.0 probability roth a 53% coincidence; it was statistically significant at (AUC=0.985). Slope variable indicates an environmental variable with more influ- ““ in potential modeling, followed by Biol4 (Precipitation of Driest Month). In agreement with the MaxEnt analysis with 19B10CLIM and 3 topographic variables, the main plant communities in the modeling area (where Taxus globosa is distributed in Mexico) are pine-oak for- ^osk-pine forest, pine forest, and cloud mesic forest. Taxus globosa is potentially distributed in Hidalgo, Nuevo Leon, Puebla, Tamaulipas, aodVm k prediction de la distribution potencial de especies raras o en riesgo de extintion es relevante para llevar a un se guimiento y conservation, ya que proporciona un punto de partida en practicas de recuperacion de suscondiciones ambientales o del establecimiento de areas de repoblacion artificial (Leal 2009). Taxus globosa « una conifera incluida en la NOM-059-SEMARNAT-20I0 en la categorla de sujeta a protection especial (SEMARNAT 2010). La information de registros de especies raras como Taxus globosa, en bases de datos o Publicaciones, es escasa, aun lo es mas sobre las areas de distribucion local o regional (Zamudio 1992; Garcia y ^Uo 2000; Contreras y Luna 2001; Zavala 2002; Zavala et al. 2001; Bonilla y Canchola 2008). Existen diver- ** metodos para el modelado de las areas de distribucion potencial (Phillips et al. 2006; Ward 2007; Phillips y Dud* 2008; Hernandez et al. 2008; Garcia 2008), algunos de los cuales se han aplicado en estudios en plantas en categoria de riesgo de extintion o de distribucion restringida (Engler et al. 2004), ademas de un estudio sobre el modelado del nicho de Taxus globosa con analisis de los efectos del cambio en uso de la tierra y conservation en Mtirico (Contreras etal. 2010). H proposito del estudio fue modelar el area de distribucion potencial de Taxus globosa y conocer la m- de enables bioclimaticas y topograficas en la precision del modelado a traves de una comparacion. i el objetiv aportar information para planes c MATERIALES Y MfiTODOS Se visitaron 40 sitios donde crece Taxus globosa en recorridos de campo ei Nuevo Leon y Tamaulipas, en el noreste de Mexico, en estos sitios se tom: cuales fueron incorporadas a una base de datos. En estos sitios se colectaron ejemplares botanicos Estrada et al, con numeros de colecta 11804, 15993, 16129, 16299, 16373, 16383, 16507, 16545, 16820, 16828, 19155, 19436, 2000 los cuales fueron depositadosen el herbario CFNL. Una serie de 39 registros (coordenadas geograficas) de Taxus globosa se obtuvieron de la consulta a la base de datos de la Red Mundial de Information sobre Biodiversidad REMIB, CONABIO, http# www.conabio.gob.mx/remib/doctos/remib_esp.html. Los 40 registros de campo se sumaron a los 39 registros de REMIB integrando una base de 79 registros totales utilizados en la modelacion de la distribution potenciaL Posteriormente del uso de esta base de datos, se utilizaron 42 registros de colecta adicionales obtenidos de visi- tas de campo posteriores con el propbsito de realizar pruebas de validation de los modelos por porcentaje de Del sitio web de WorldClim (Hijmans et al. 2005; http://www.worldclim.org/bioclim) se obtuvieron cu- biertas geograficas de variables ambientales de 1960-1990 en formato raster de un km 2 de resolution. La base de datos BIOCL1M utilizada en el modelado de distribution consta de 19 cubiertas (Cuadro 1), la cual fue combinada en una segunda modelacion con un grupo de tres variables topograficas altitud, exposition y pen- diente (Cuadro 1) que fueron calculadas a partir de la cubierta ‘altitude’ de WorldClim (Hijmans et al. 2005; http://www.worldclim.org/bioclim), todas estas fueron recortadas en ArcGIS version 9.2 a los limites de 118°20'3.42 n a 86°25'3.42" Oeste y 34° 01'0.08" a 12° 5530.76" Norte, correspondiente al territorio de Mexico; las capas se transformaron en ArcGIS version 9.2 a formato ASCII Grid. Se utilizb el modelado Maximum Entropy Distribution o MaxEnt, el cual es el mejor metodo (Kumar y Stohlgren 2009), es de los que procesan bases de datos con pocos registros (Hernandez et al. 2006; Pearson et al. 2007). Es un mtiodo basado en un protocolo que estima la probabilidad de ocurrencia de las especies basa- do en requerimientos ambientales que genera una estimation de probabilidad de presencia de la especie am valores de 0 y 1, donde 0 se considera como la minima y 1 para la maxima probabilidad (Phillips et al. 2006), para realizar el analisis solo se requiere de datos de presencia de la especie (no ausencia) y cubiertas geograficas de variables ambientales de las areas de analisis (continuas o categoricas). El software MaxEnt versibn 3.3.2 es de uso libre y esti disponible enhttp://www.cs.princeton.edu/~schapire/maxent/. La base de datos compilada de 79 registros de Taxus globosa y los cuatro grupos de variables ambientales fueron utilizadas para la modelacion de un escenario climatico actual (periodo de 1950 a 2000). El programs generb intigenes logisticas acumulativas y analizo a un maximo de 500 aproximaciones sucesivas (iteraciones). Los dos modelos generados se sometieron a las pruebas curva de respuesta (analisis de omision/comision y sensitividad ROC AUC=Area Under Curve) y una prueba Jackknife para medir el efecto de cada variable u0- lizada en la generacibn de los modelos. Para realizar esto se dividio la base de datos de 79 registros en un subgrupo de 40 registros para el modelado de habitat potential y otro de 39 para pruebas estadisticas de vaK- Con una base de 42 registros adicionales de Taxus globosa, a traves del modulo Point Analyst 10 pan Arcview, se registrb el valor de los dos modelos de distribution potential generados en formato raster los cualti presentan valores de probabilidad de habitat de 0.1 a 1.0, de manera que solo se contabilizaron los registroscon valores entre 0.8 y 1 .0 para expresarlos en porcentaje de ocurrencias >80 % de probabilidad de tener un habitat potential para Taxus globosa. Se realizb un analisis de distributibn de la superficie actual y potential (entre los pixeles de 0.8 a 1.0 pR>* babilidad) generado en el mejor modelo de acuerdo al uso actual de la tierra y vegetation (INEGI 2005). A* mismo, se determino su area de distributibn actual y potential en Mexico. Garcia et al.. Distribution de Taxus globosa en Mexico 0*«»1.Ustadodelasvariablesambientales BIOCLIM 2 y topografi 589 RESULTADOS Y DISCUSlON de distribution potencial para Taxus globosa en Mexico se muestra en la Figura 1 el cual considera ^ 19 cubiertas geograficas BIOCLIM sin la de topografia. Este mapa tiene un alto nivel de prediccion. El del area bajo la curva ROC (AUC = 0.987) (Cuadro 2). En contraste el mapa del modelo de habitat poten- Pwa T. globosa considerando las 19 cubiertas geograficas BIOCLIM y las tres cubiertas geograficas de topo- 2 X genero una tasa de exito mayor en su prediccion con un valor de area bajo la curva ROC (AUC - M , ) (Cuadr ° 2), estos valores son menores al obtenido por el modelo (AUC = 0.997) generado por Contreras tt *l (2010). ilidacion de MaxEnt (Cuadro 2), los dos modelos generados tuv Rirv-T Vaiiables 9 ue tuvieron mayor ponderacion en la generacion de los modelos, fueron: para el modelo tem Peratura maxima del mes mas calido (Bio05) 39.4%, isotermalidad [P2A>7] [*100] (Bio03) ’ Pecipitacion del mes mas seco (Biol4) 15.3% y temperatura minima del mes mas frio (Bio06) 10.3%; 590 Journal of the Botanical Research Institute of Texas 6(2) las de mayor probabilldad. para el modelo BiOCLIM con topografla: Pendiente (Topo03) 39.8%, precipitacion del mes mis seco (Biol4) 25.5%, elevacibn (TopoOl) 14.0% y temperatura minima del mes mas frio (Bio06) 7,5%, muestra que la pen* diente del terreno tiene relevancia en uno de los dos modelados, seguida de la precipitacion en el mes mas seco y elevacibn. Esto significa que el segundo modelo produjo areas mas especificas que discriminan zonas p® topografla (Fig. 3, Cuadro 1), Los patrones de distribucion biogeografica estan cambiando en respuesta a recientes alteraciones del cfr ma, como lo establece un indice que mide la velocidad de cambio de la temperatura (km afio) y la velocidad cambio varia en los ecosistemas (Loarie et aL 2009). En relacibn a la influencia de la topografla, dicho tadis establece que el efecto topografko, influye de manera importante en la velocidad de cambio de la temper®" 1 *’ siendo menor (0.11 km ano) en el ecosistema con bosques de coniferas tropicales y subtropicales, dondest distribuyen las comunidades vegetales con presencia de Taxus globosa. Ademas, el tiempo de residencia Ok®" po que permanece un ecosistema en una regibn) de este es mayor (63.6 anos) respecto a otros ecosistemas analizados (Loraie et al. 2009). La Figura 4 muestra la distribucion de los registros usados en el modelado de habitat potencial con la '^ liable Topo03 (pendiente), donde se observa que los datos se concentran entre los 89.99 y los 90.00 grades pendiente, lo cual indica que la especie prefiere crecer en los canones. En la Figura 5 se muestra la distribuc*® de los registros en la variable Biol4 (precipitacion del mes mas seco), los datos se concentran entre 8.0 y 32®*® de precipitacibn (mes mSs seco del ano), lo que indica zonas con periodos de sequla. La Figura 6, muesttf distribucibn de los registros para la variable Topo02 (elevacibn), en la que se observa una preferencia por aones etitre 1,200 y 3,000 m, ubicando la condition del habitat en zonas elevadas. En la Figura 7, se muestra la nbucidn de los registros en la variable Bio06 (temperatura minima del mes mds frlo), los datos se ubican entre 24 y 33 °C, lo que denota una preferencia por sitios templados, En el analisis de distribucion por deciles de probabilidad (entre 0.1 y 1.0) de presencia de distribution P^ial de Toxus globosa en Mexico para lo dos modelos, se observo que los uitimos dos deciles (0.8-0.9 y 1 °) re P re sentan las areas donde sc advierte mayor probabilidad de encontrar a la especie y coincide con la 20113 nticleo del area potential de distribucion de la especie (Figs. 1 y 2). Earaelpcnultimodecilde probabilidad (0.8-0.9) la superficie calculada para la distribution potencial fue en el modelo BlOCLIM con 9,299.38 km 2 que representa un 0.47% del territorio de Mexico, mientras qUCenel mod elo Biocilm+Topografia genero una superficie de 8,581 km 2 que representan el 0.43% del territo- Porelc. **** nna suj ^CLIM+Topografia genero una superficie de 1,804 km 2 (0.09% del territorio mexicano). Es probable que la cam CnCia Cn tendencia Sca resultado del efecto de la pendiente y elevaciOn ya que los registros tornados en Po estan ubicados en zonas de elevation mayores a 1,200-3,800 m y con pendiente pronunciada (Fig. 4). 1 grupo de 42 registros adicionales de Taxus globosa se realizO una prueba de validation de s generados, contabilizando los registros que coinridieran geograficamente con las zonas de los e marcaron valores de probabilidad entre los deciles 0.8-0.9 y 0.9-1.0, estos se expresaron en valo- Journal of the Botanical Research Institute of Texas 6(2) ft. s. Mapa de la distnbucidn de 79 registros de Taxus globosa en reladdn con la variable ambiental Bio14 (precipitation promedio del mes mis seco). res porcentuales de coincidencia con los modelos. Para el modelo BIOCLIM se obtuvo una precision general de 52% en el deed 0.8-0.9 y 2% en el deed 0.9-1 .0) mientras que para el modelo BIOCLIM+Topografia se obtuvo una precision de 55% (36% en el decil 0.8-0.9 y 19% en el deed 0.9-1.0), Considerando el total de valo- probabilidad de los pixeles, el modelo BIOCLIM tiene mayor precision con 83%, mientras que si consi- ^ ran,os Flores del decil de mayor probabilidad de ocurrencia de habitat potencial para Taxus globosa, el mo- *° B I0CLIM+ Topografia tiene la mayor precision con 55% (Cuadro 4). U P meba muestra que el modelo de BIOCLIM+Topografia tiene una mayor eficiencia en el Ultimo percen- " de ma r or probabilidad de ubicar Taxus globosa de 0.9-1.0, mientras que el modelo BIOCLIM es mA s preciso * wpercentil de menor probabilidad de ubicacion de Taxus globosa 0.8-0.9 (Fig. 8), en contraparte Contreras C,a ' ^ 01 °) obtuvieron un 94 5% de eficacia en la prediction para su modelo; esta diferencia se establece ya que ^subdividio el modelo en deciles y fue considerada la totalidad de la superficie del modelo, la eficiencia de “todelos considerando las coincidencias de los registros de la prueba de eficiencia son del 90% para el mo- un ° y 81 % p ara m0{ j e l 0 2 ( a partir de una probabilidad del 0.6 al 1.0. t ^ fealizo un analisis de la superficie del habitat potencial (0.8-1.0) para T. globosa, de acuerdo al uso de la y Ve g e taci6n (INEGI 2005), al cruzarla information se analizO la distribution de los 10,385 km 2 en rela- ° n a 'os tipos de vegetation y uso actual de la tierra. De este analisis se obtuvo que la comunidad con mas P^dicie del area de distribution potencial es el bosque de pino-encino con 2,230 km 2 , (21.48%), el bosque de "O-Pino eon 1,638.5 km 2 (15.78%), el bosque de pino con 1,244.75 km 2 (11.99%). Existe un Area abierta a la 594 agriculture de temporal de 1,031.53 km 2 (9,93%), el bosque mesofilo de montana con 1,027.12 km 2 (9.89%), d bosque de encino con 10,016 km 2 (9.78%), ver Cuadro 5. En un analisis de distribucion posterior de la superficie generada de habitat potencial (0.8-1.0) para I globosa de acuerdo al modelo BIOCLIM+Topografia, los estados con mayor superficie potencial son: Hidalgo, Nuevo Leon, Veracruz, Puebla y Tamaulipas. En el percentil de mayor probabilidad (0.9-1.0) los estados de Veracruz (474,5 km 2 ), Puebla (401.3 km 2 ), Hidalgo (329.9 km 2 ) y Nuevo Leon (310.1 km 2 ) son los que tienen mayor superficie. En el percentil medio (0.8-0.9) los estados de Hidalgo (1,935.95 km 2 ), Nuevo LeOn (1,775 km 2 ), Veracruz (1,342 km 2 ), Puebla (1,258 km 2 ) y Tamaulipas (1,173 km 2 ) son los que tienen mas superficie (Cuadro 6). Los modelos desarrollados en MaxEnt para obtener distribucion potencial para Taxus globosa en Mexico, advi- erten un area muy reducida, con una superficie entre los 9,650 km 2 (BIOCLIM) y los 10,385 km (BIOCLIM+Topografia) que representan entre el 0.49% al 0.52% del territorio nacional. El modelo BIOCLIM (19 variables climaticas) resulta tener el mejor porcentaje de precision global enel percentil de probabilidad (0.8-0.9) de presencia de habitat potencial para Taxus globosa en Mexico. Para el percentil de mayor probabilidad de presencia de la especie (0.9-1 .0), el modelo BIOCLIM+Topogr 3 " 2 resulto ser mas preciso con el 55% en la prueba de validacion. La variable de pendiente es de mayor influence en el modelado de habitat potencial, seguida por Biol4 (precipitation del mes mas seco). II /IIIIIIIII! I Garcia et al.. Distribution deTaxus globosa en Mexico 595 1,843,243.20 56,857.47 23.614.17 13.427.17 10,439.67 8,558.69 8,093.88 1,828,028.46 62,255.25 27,856.23 16,507.72 It, 496.47 9,701.42 La ubicacion del habitat potencial de Taxus globosa en zonas de montana correspondientes al bosque de atferas " ’ • 1 im bio de temperatura por efecto global, asf como una [iempo de residencia de este habitat L °S tipos de vegetation que el habitat potencial de los ultimos dos percentiles de probabilidad de encino-pino, bosque de pino y bosque mesofilo de montafla, Garda etal., Distribution de Taxus globosa en Mexico 597 Gmm» 5. Distribution de la superfide generada de habitat potential para Taxus globosa (0.8-1 .0) con base en ei mapa uso del suelo y vegetation (INEGI, 2005). Gmdro6. Superficies (km 2 ) de habitat potential para Taxus globosa (km 2 ) de los principles estados en Mexico en los percentiles 0.8-0.9 y 0.9-1 .0 de probabilidad scgun el modelado BIOCLIM+Topografia. odstiendo para el 2005 una area abierta a la agricultura de 1,027 km 2 que representa casi un 10% de este habi- tat potencial en Mexico. De acuerdo al analisis MaxEnt utilizando las 19 cubiertas geograficas BIOCLIM y tres topograficas, los estados de Hidalgo, Nuevo Leon, Veracruz, Puebla y Tamaulipas poseen mayor superficie donde se puede dis- tnbuir Taxus globosa en Mexico. REFERENCIAS Bomua ' L -L y P.B. Canchola G. 2008. Fitogeografia de Taxus globosa en la regitin centro-oriente de Mtixico.Tesis licencia- tura - Divisitin de Ciencias Forestales, Universidad Autonoma Chapingo.Texcoco, Edo. Mexico. CoNm£RAS < M.R. e |. Luna. 2001 . Presencia de Taxus globosa Schltdl. (Taxaceae) en el estado de Chiapas, Mexico. PolibofcS- nka 1251-55. M.R., |. Luna y CA Rios. 2010. Distribution de Taxus globosa (Taxaceae) en Mexico: Modelos ecolbgicos de nkho, efectos del cambio del uso de suelo y conservation. Revista Chilena Hist. Nat. 83:421-433. kLoNG, E.R., D.M. DeLong y D.L. Clarke. 1988. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44:837-845. ***' R- A. Guisan y L. Rechsteiner. 2004. An improved approach for predicting the distribution of rare and endangered species from occurrence and pseudo-absence data. J. Appl. Ecol. 41 :236-274. M R - 2008. Modelos predictivos de riqueza de diversidad vegetal, comparacion y optimization de mbtodos de Modelado ecologico. Ph. D.Tesis. Facultad de ciencias biologicas, Departamento de Biologia Vegetal, Universidad Complutense de Madrid. Espana. 598 Garcia, S.F. v P.L Castillo. 2000. Aspectos ecolbgicos de Taxus globosa Schltdl. en las Mesas de San Isidro, Municipiode Rioverde, San Luis Potosi. BIOTAM 1 1(3):1 1-18. Hernandez, PA, C.H. Graham, LL.Master y D.L Albert. 2008. The effect of sample size and species characteristics on perfor- mance of different species distribution modeling methods. Ecography 29:773-785. Humans, RJ., S.E. Cameron, J.L. Parra, P.G. Jones y A. Jarvis. 2005. Very high resolution interpolated climate surfaces for global landa areas. Int. J. Climatol. 25:1 965-1 978. Instttuto Nacional de EstadIstica, GeografIa e InformAtica. 2005. Mapa uso del suelo y vegetacibn, Mexico [CD-ROM]. Kumar, S. y TJ. Stohlgren. 2009. MaxEnt modelling for predicting suitable habitat for threatened and endangered tree Canacomyrica monticola in New Caledonia. J. Ecol. Nat. Environm. 1 (4):094-098. Leal, N.0. 2009. Aspectos y patrones de distribucibn de Pinus martinezii Larsen en la cuenca del lago de Cuitzeo, Mexico: Un modelo de distribucibn potencial aplicado a la conservacibn.Tesis Posgrado en Geografi'a, UNAM. Mexico, D.F. Loraie, R.S., B.D. Phiup, H. Hamilton, G.P. Asner, C.B. Field y D.D. Ackerly 2009. The velocity of climate change. Nature 462:1052-1057. Pearson, R.G., CJ. Nakamura, y A.T. Peterson. 2007. Predicting species distributions from small numbers of occurrence records: a test case using cryptic geckos in Madagascar. J. Biogeogr. 34:102-1 17. Phillips, SJ., R.P. Anderson y R.E. Schapire. 2006. Maximum entropy modeling of species geogra; Model. 190:231-259. Phillips, SJ. y M. DudIk. 2008. Modeling of species distributions with MaxEnt: new extensions and a comprehensive evalu- ation. Ecography 31:161-175. Semarnat [Secretary del Medio Ambiente y Recursos Naturales]. 2010. Norma oficial mexicana NOM-059-SEMARNAT-2010que determina las especies nativas de Mexico de flora y fauna silvestres-categorias de riesgo y especificaciones para su inclusibn, exclusion o cambio-lista de especies en riesgo. Diario oficial de la federacion, 2 a seccibn, 30 de Diciembre de 2010. Ward, D.F. 2007. Modelling the potencial geographic distribution of invasive ant species in New Zeland. Biol. Invasions 9:723-735. Zamudio, S. 1992. Familia Taxaceae. En: Flora del Bajfo y de regiones adyacentes. Institute de Ecologia. PStzcuaro, Mi- choacan, Mbxico. Fascfculo 9. Zavala, C.F. 2002. Anblisis demogrbfico preliminar de Taxus globosa Schltdl. en el Parque Nacional El Chico, Hidalgo, Mbxico. I: Poblacibn de adultos y algunas caracteristicas del hbbitat. Ciencia 8(2):169-1 74. Zavala, C.F., M. Soto H. yT. Rodriguez G. 2001. El romerillo (Taxus globosa Schlecht.): biologfa, dificultades y perspectivas de su uso. Revista Chapingo, Ser. Hort. 7:77-94. DISTRIBUTION AND MORPHOLOGICAL CHARACTERISTICS OF ARCEUTHOBIUM HONDURENSE AND A. NIGRUM (VISCACEAE) IN MEXICO Robert L. Mathiasen School of Forestry Northern Arizona University Flagstaff, Arizona 86011 U.S.A. Robert.Mathiasen@nau.edu Shawn C. Kenaley Department of Plant Pathology and Plant-Microbe Biology Cornell University Ithaca, New York 14853 U.S.A. sck26@cornell.edu Brian P. Reif School of Forestry Northern Arizona University Flagstaff, Arizona 860 1 1 U.S.A. Brian.Reif@nau.edu ABSTRACT RESUMEN Tfe genus Arceuthobium (Santalales: Viscaceae) consists of 42 species (Hawksworth & Wiens 1996) that are aenal parasites of Pinaceae or Cupressaceae. Many of the species, commonly known as dwarf mistletoes, are cognized as serious forest pathogens (Hawksworth & Wiens 1996; Mathiasen et al. 2008). Morphological characters consistent throughout the genus include small flowers produced on male and female plants, leaves reduced to squamate scales, and morphologically similar bi-colored fruits (Hawksworth & Wiens 1996). The 8enus ^ long been considered a taxonomically difficult group because of the extreme morphological reduc- ’'° n dialed with the parasitic habit and the morphological similarities between species (Hawksworth & lens 1996 )- Factors that complicate classification and identification include a large amount of variation in morphology and geographic distributions as well as flowering periods that occasionally overlap. Two species ' 1 exem Phfy the problems associated with field identification of morphologically similar dwarf mistletoes ^ in southern and central Mexico: Arceuthobium hondurense Hawksw. & Wiens (Honduran dwarf mistle- 106 3ndA - "igrum Hawksw. & Wiens (black dwarf mistletoe). Journal of the Botanical Research Institute of Texas 6(2) Arceuihobium hondurense was originally described from central Honduras (Hawksworth & Wiens 1970) and thought to be extremely rare, even on the verge of extinction due to rapid and extensive harvesting of its pine hosts (Hawksworth & Wiens 1972). However, Honduran dwarf mistletoe was later found to occur from northern Nicaragua, through much of Honduras into Chiapas, Mexico and as far north as central Oaxaca (Ma- thiasen et al. 2001; Mathiasen et al. 2002a; Mathiasen et al. 2003; Mathiasen et al. 2006; Mathiasen & Melgar Wiens 1977; Mathiasen et al. 2003). Another dwarf mistletoe, originally described as A. hawfesworthii Wiens and C.G. Shaw (Hawksworth’s dwarf mistletoe), was recombined as a subspecies of A. hondurense (Mathiasen 2007). Therefore, A. hondurense presently consists of two subspecies: A. hondurense subsp. hondurense which occurs from Nicaragua north to Oaxaca, Mexico (Mathiasen et al. 2010) and A. hondurense subsp. hawksworthu (Wiens & C.G. Shaw) Mathiasen which is primarily distributed in the Mountain Pine Ridge area of Belize (Mathiasen 2007), but also has been reported from central Honduras (Mathiasen et al. 2002b). Because plants of Arceuthobium hondurense subsp. hondurense (hereafter referred to as A. hondurense) art similar in size and color to those of A. nigrum, determining the geographic distribution of these species in southern Mexico has been difficult (Hawksworth & Wiens 1989, 1996; Mathiasen et al. 2001, 2002a, 2003, 2010). Although A. nigrum was thought to be distributed from northern Durango, Mexico into southern Mexi- co (Hawksworth & Wiens 1996), it is now thought to be distributed only as far south as central Mexico (Ma- thiasen et al. 2010). Further, both A. hondurense and A. nigrum produce red flowers that bloom in the fall. We have collected additional morphological data for A. hondurense and A. nigrum since 1998. Here we report our findings and discuss the distribution of these dwarf mistletoes in Mexico based on our field observations and morphological measurements. Because prior studies (Mathiasen et al. 2003; Nickrent et al. 2004) have success- fully used ribosomal DNA (rDNA) sequence information to discriminate between A. hondurense and A ni- grum, we conducted additional analyses of the internal transcribed spacer (ITS) region for several populations of both species, particularly populations in central Mexico where these species may be sympatric (Nickrent et al. 2004). In addition, because A. vaginatum (Willd.) Presl subsp. vaginatum (Mexican dwarf mistletoe) is also morphologically similar to A. hondurense as well as A. nigrum, and often confused with them, we have provided information on how to distinguish these species from A. vaginatum. The primary objective of this study, how- ever, was to provide additional data on how to discriminate A. hondurense from A. nigrum — and vice versa- and in so doing, better determine their geographic and host ranges. Morphological data and ITS sequences for A. vaginatum were taken from Hawksworth and Wiens (1996) and obtained from GenBank, respectively. MATERIALS AND METHODS Morphology and Phenology To compare morphological characters we sampled 16 populations of Arceuthobium hondurense (two fro® Oaxaca, Mexico, one from Nicaragua, and 13 from previous work by Mathiasen (2007)) and 14 populations A. nigrum from throughout its geographic range (Fig. 1). Plants were measured from the type locality for both mistletoe species (Hawksworth & Wiens 1965, 1970, 1977) (Fig. 1; locations 6 and 23). From each population, 10-20 male and 10-20 female plants were collected and the dominant shoot from each infection was used for morphological measurements. Characters measured were those used by Hawksworth and Wiens (1996) far taxonomic classification of Arceuthobium: height, basal diameter, third internode length and width, and color of male and female plants; mature fruit length, width, and color; seed length, width and color; length and width of staminate spikes; staminate flower diameters for 3- and 4-merous flowers; length and width of stamina* flower petals; and, anther diameter and anther distance from the petal tip. Plants were measured within 2 hours after collection using a digital caliper and a Bausch and Lomb 7X hand lens equipped with a micromeW Staminate spike and flower measurements were made during the peak of anthesis and fruit and seed measure- ments were made during the peak of seed dispersal. One-way analysis of variance (ANOVA) was used to exam- ine the variance in the above characters for A. hondurense and A. nigrum and significant differences between means were determined using a Tukey’s honestly significant difference (HSD) post-hoc test (a = 0.05). All s» tistical analyses were performed using JMP 8.0.2 software (SAS Institute, Cary, NC). 601 Because the times of flowering and seed dispersal for Arceuthobium hondurense and A. nigrum are poorly known (Hawksworth & Wiens 1996), additional observations of the phenology of these taxa were made dur- ing the spring and fall of 1999, 2003, 2005, 2007, 2008, and 2010 as well as during the early spring of 2011. DNA Extraction and ITS Sequencing ^nples of DNA were obtained from five and six specimens, each representing a geographically separate popu- !ati °n, of Arceuthobium hondurense and A. nigrum , respectively. Locality and voucher number for each speci- "•en (bold print) are presented in Fig. 1. For each specimen, total DNA was extracted using the DNeasy™ Plant MlruKit (Qiagen, Valencia, CA) according to the manufacturer’s instructions. DNA purity and concentration Were quantified for each sample using a NanoDrop ND-1000 (Thermo Fischer Scientific, Wilmington, DE). ^'kngth, ITS sequences (comprising ITS1, 5.8S rDNA gene, and ITS2) were PCR-amplified using the primer P® 1 IBS 1830for and 26S 40rev (Nickrent et al. 2004). PCR amplifications were carried out in 25 pL reaction "jhuures containing 12.5 pL of 2X AmpliTaq Gold® Master Mix (Applied Biosystems, Foster City, CA), 0.5 pL ofea ch20pM primer, 11.25 pL nuclease-free water, and -2-18 ng (0.25 pL of 8-78 ng/pL) of genomic DNA. 602 PCRs were performed in an Eppendorf Mastercycler® pro thermal cycler (Eppendorf, Westbury, NY) with the following cycling parameters: initial hold for 6 min. at 95°C; 5 cycles at 94°C for 30s, 55°C for 30s, and 72°C for 1 min.; 33 cycles at 94°C for 30s, 48°C for 30s, and 72°C for 1 min.; and, a final extension step of 72°C for 10 min. Blank reactions (i.e., minus genomic DNA) were run concomitantly to check for contamination of the reagents. The size of each PCR product (bp) was checked separately by ultraviolet fluorescence after 1.2% agarose gel electrophoresis in 0.5x TAE buffer and staining with GelRed™ (Phenix Research Products, Candler, NQ. Amplification products were purified directly from reactions using ExoSAP-lT (0.4 pL per pL of reaction prod- uct; USB Inc., Cleveland, OH) and normalized to 130 ng per sequencing reaction. Sequencing was carried out using a BigDye terminators DNA sequencing kit (Applied Biosystems), ABI 3730 DNA sequencer, and the above forward and reverse primers. PCR products were sequenced in both directions. Sequences were proof- read and assembled in CodonCode Aligner (CodonCode Corporation, Dedham, MA). Boundaries to the 5- and 3'-region of ITS1 and ITS2, respectively, were previously identified by Nickrent et al. (1994). ITS sequences for A. hondurense (n=5) and A. nigrum (n=6) produced in this study were deposited in GenBank. Phylogenetic Analysis ITS sequences for Arceuthobium hondurense and A. nigrum obtained in this study and from GenBank (A. hondu- rense AY2888263 and A. nigrum AY288271) as well as A. vaginatum subsp. vaginatum (AY288286 and AY288287) and A. douglasii Engelmann (L25687; outgroup) were included in the dataset. Sequences were aligned using ClustalX ver. 2 (Larkin et al. 2007) and visually edited as necessary in CodonCodon Aligner. Maximum Likelihood (ML) trees were constructed using PAUP* 4.0bl0 (Swofford 2003). The DNA substitu- tion model TIM2 and the parameter estimates for tree reconstruction were determined using the Akaike Infor- mation Criterion (AIC; Akaike 1974) as implemented in jModelTest 0.1.1 (Posada 2008). All nucleotides were included in the phylogenetic analysis; gaps were treated as missing characters. Heuristic searches were per- formed with 200 replicates of random sequence addition and tree bisection-reconnection (TBR) branch swap- ping. Branch support was evaluated using 1000 bootstrap replicates and 10 random additions of sequences per pseudo-replicate. Inter- and intraspecific genetic distances were also examined using Kimura’s two-parameter model (K2P; Kimura 1980) for base substitution as implemented in PAUP*. Bayesian analysis was also performed using MrBayes 3.1.2 (Huelsenbeck & Ronquist 2001). The best-fit model for DNA substitution was determined as described previously; however, Hasegawa-Kishino-Yano (HKY; Hasegawa et al. 1985) model and parameter estimates were determined using the Bayesian Information Crite- rion (BIC; Schwarz 1978). One cold and three heated Markov chain(s) were run, and samples were taken every 100 generations over 5.0 x 10 6 generations. The potential scale reduction factor (PSRF) for each of the model parameters was > 1.0 when the program was terminated. Stationarity was accessed by examining the average standard deviations of split frequencies and likelihood values. Bum-in value (10%) was determined using Tracer vl.5 (Rambaut & Drummond 2009). The r emaining trees were used to calculate a 50% majority rule consensus tree and to determine the posterior probabilities. RESULTS AND DISCUSSION Our measurements of Arceuthobium hondurense indicate it forms larger plants than previously reported by Hawksworth and Wiens (1970, 1996); they reported plant heights averaged approximately 14 cm with a ma»- mum height of 21 cm, but we measured plants (male and female combined) that averaged 22 cm. We found some male plants in Chiapas, Mexico that were over 65 cm in height (Table 1). The discrepancy in maximum heights is probably related to Hawksworth and Wiens only measuring specimens from central Honduras where plants were generally smaller than in southern Mexico (Mathiasen et al. 1999; Mathiasen 2007). A* mean basal diameter of dominant shoots was the same as reported by Hawksworth and Wiens (1996); ap proximately 5 mm. However, we measured some shoots with basal diameters of nearly 13 mm, while Hawk' sworth and Wiens (1996) only reported a maximum of 9 mm for this character. The means and ranges for m^ of the remaining morphological characters were similar to those previously reported (Hawksworth & W*® 5 Previously, Mathiasen et al. (2003) listed the pines parasitized by Arceuthobium hondurense in Central America and Chiapas, Mexico. Our observations support their findings, except that A. hondurense also parasit- ized Pirns teocote Schiede ex Schlechtendal & Chamisso in central Oaxaca. Although we have only observed A. hondurense infecting this pine in one location north of Ixtlan (Fig. 1, location 13), the level of infection (> 90%) indicated Pinus teocote was a principal host. This population was misidentified as A. vaginatum subsp. vaginatum by Hawksworth and Wiens (1996), as our morphological and ITS analyses indicated this population was indeed A. hondurense. Mistletoe plants on P. teocote at this location were dark brown to black, similar in color to A. vaginatum, but male plants flowered in the fall producing dark red flowers. Large male plants at this locality also had swollen nodes and parasitized P tecunumanii Eguiluz et J. P. Perry, a principal host of A. hon- durense elsewhere in southern Mexico (Mathiasen et ai. 2003). In addition, A. hondurense was found parasitiz- ing P. lawsonii Roezl ex Gordon & Glendining in central and northern Oaxaca. Although Hawksworth and Wiens (1977, 1996) reported A. hondurense (but identified as A. nigrum ) parasitizing Pinus oaxacana Mirovand P. patula Schiede ex Schlechtendal & Chamisso in Chiapas, we did not observe these host-mistletoe combina- tions during our field work in southern Mexico. Arceuthobium nigrum is morphologically very similar to A. hondurense. Both species produced relatively large, dark brown to black plants on their pine hosts (Hawksworth & Wiens 1996). Male and female plants of A ni- grum averaged nearly 25 and 18 cm in height, respectively, but were not significantly larger than A. hondurense (Table 1). It is difficult to compare our results for plant heights with those of Hawksworth and Wiens (1996) because they only provided a range of heights for A. nigrum (15-35 cm, maximum = 45 cm). However, the larg- est plant we measured for A. nigrum was a male over 53 cm tall from Puebla, Mexico. We also found that the basal diameter of shoots averaged two-mm larger (7.6 mm) than that previously reported (5 mm, Hawksworth & Wiens 1996) with a maximum basal diameter nearly twice that described by Hawksworth and Wiens. Mea- surements of the third intemode widths also indicated that A. nigrum produced thicker plants (about 5 mm) than what Hawksworth and Wiens reported (about 4 mm). The means for the basal diameters of male (7 mm) and female plants (7.8 mm) and third intemode widths of A. nigrum were significantly greater than those for A hondurense (Table 1). Furthermore, the mean length of the third internode of female plants of A. nigrum (16.5 mm) was significantly longer than those of A. hondurense (13.8 mm) but not the mean length of the third inter- node of male plants. The only flower characteristics Hawksworth and Wiens reported for Arceuthobium nigrum was the diam- eter of 3-merous flowers (3.5 mm), which was sUghtly larger than the mean diameter for the 3-merous flowers we measured (3.2 mm). Our observations indicated that A. nigrum commonly produced 4-merous flowers also, thus, we sampled these flowers and found they averaged nearly 5 mm in diameter (Table 1). Collectively, the mean diameters of 3 and 4-merous flowers of A. nigrum were significantly larger than those of A. hondurense. Petal sizes were also relatively large for A. nigrum when compared to other dwarf mistletoes (Hawksworth & Wiens 1996). We found petals longer than 2 mm and nearly as wide, both significantly larger than those for A hondurense (Table 1). Another key characteristic of A. nigrum flowers that was consistent throughout its geo- graphic range, was that the adaxial surface of its petals was dark red. While this characteristic was easily ob- served, and is similar to flowers for A. hondurense, a review of the literature (Hawksworth & Wiens 1989, 1996) on A. nigrum revealed that the petal color, as a diagnostic character, had never been mentioned before, The mean f fruits 8.8 rr of fruits we Arceuthobium nigrum were remarkably glaucous and large compared to other dwarf mistletoes it length was nearly 7 mm, which is what Hawksworth and Wiens reported. However, we found in length compared to 9.0 mm by Hawksworth and Wiens (1996). In contrast, the average width amined was larger (4.1 mm) than that reported by Hawksworth and Wiens (3.5 mm). We lat were shorter, but wider on average than those examined by Hawksworth and Wiens (1996) • fruit length and width of A. nigrum were significantly larger than those of A. hondurense, ho* i length and width of seeds were similar for both species (Table 1). 605 The phenology of A. nigrum requires additional observations as we were unable to confirm the incidence of two flowering periods — one in March-April and one in September-October— as reported previously by Hawksworth and Wiens (1989, 1996). We examined male plants of A. nigrum at several localities in mid- to late-March during 2003, 2005, and 2007 as well as in early April 2011 and never observed open flowers. Fur- thermore, staminate flowers did not appear to be approaching anthesis in late March or early April. Our field observations, however, indicated that it flowers beginning in mid-September and continued into November in Durango, Mexico, while in central Mexico, A. nigrum flowers in late-September and continued into at least January. Peak flowering was in early October in Durango; however, the peak flowering period in central Mex- ico is still poorly understood. Seed dispersal initiated in early September and peaked in mid-October continu- ing to mid-November in Durango and elsewhere in central Mexico. Our field observations of pines infected by Arceuthobium nigrum in Mexico did not reveal any additional hosts. The principal hosts of A. nigrum in Durango were clearly Pinus leiophylla Schiede ex Schlechtendal & Chamisso, P. lumholtzii B. L. Robinson & Femald, P. teocote, and P chihuahuana Engelmann as reported by Hawksworth and Wiens (1996). This mistletoe also been reported to rarely infect P. arizonica Engelmann and P. cooperi Blanco in northern Mexico (Hawksworth & Wiens 1996), but we have not observed it on these infre- quent hosts. In central Mexico its principal host was P teocote. Pinus patula Schlechtendal & Chamisso was a secondary host at several locations in Hidalgo and Puebla. While we agree with the classification of P. pseu- dostrobus Lindley as an occasional host of A. nigrum (Hawksworth & Wiens 1996), we were unable to verify whether P montezumae A. B. Lambert was also an occasional host. Moreover, we did not observe any P mont- ezumae at the location where Hawksworth and Wiens reported an infestation of A. nigrum in Hidalgo nor in Veracruz where we found large, mistletoe-free P. montezumae growing near P teocote severely-infected with A. nigrum. Therefore, the susceptibility of P. montezumae to A. nigrum needs further study. Although Hawksworth and Wiens (1989, 1996) reported that both P. lawsonn and P. oaxacana Mirov were principal hosts of A. nigrum , this host susceptibility classification was based on infection of these pines by A. hondurense in Oaxaca and Chiapas, respectively (Mathiasen et al. 2003). DNA Analyses DNA sequence analysis demonstrated that Arceuthobium hondurense, A. nigrum, and A. vaginatum occurred in three well supported clades (Fig. 2). All samples identified morphologically as A. hondurense and A. nigrum yielded a 627 and 623 bp fragment, respectively, consisting of the 3'end of the 18S (4 bp), complete ITS1-5.8S- HS2 sequence (604 and 600 bp), and the 5'end of the 26S (19 bp). Four of five sequences for A. hondurense were identical (mean K2P value= 0.0017); however, RLM 98107 differed by an A/G nucleotide change at positions 22 and 40 in 1TS1. Similarly, ITS sequences for A. nigrum were nearly identical (mean K2P value= 0.0018); except for A/T nucleotide changes at position 508 in ITS2. The alignment for phylogenetic analyses consisted of 649 characters including those of A. douglasii and A. vaginatum. Of these characters, 574 were constant, 50 were tree supported three distinct clades (Fig. 2) with bootstrap values >98% and posterior probability values equal to 1.00, respectively. Each plant identified according to morphometric data as A. hondurense or A. nigrum formed a distinct clade with either A. hondurense (RLM 0136, Nickrent et al. 2004) or A. nigrum (DLN 2019, Nic krent et al. 2004). Sequences of A. nigrum differed from those of A. hondurense and A. vaginatum by ap- proximately 43 nucleotides (mean nucleotide difference = 43.07, mean K2P value= 0.0775). Likewise, the mean number of nucleotide changes between A. hondurense and A. vaginatum was 13.0 (mean K2P value= 0.0224). Nickrent etal. (2004) reported thatA. hondurensemay occur in Veracruz based on molecular data (GenBank accession no. L25693; voucher DLN 2018), but our results did not support this. Plants collected south of Sierra de A A Ua (RLM 1083), the approximate location where Nickrent collected DLN 2018 (D. Nickrent, pers. comm.), Were morphologically similar to those of A. nigrum. As suspected, the ITS sequences generated from RLM 1083 identical to L25693 (data not shown). However, in a separate phylogenetic analysis (data not shown), the se collections/sequences were unrelated to A. nigrum (AY288271) as well as GenBank accessions of A. du- r ®«go«eHawksw. & Wiens, A. gUlu Hawksw. & Wiens, A. hondurense, and A. vaginatum. Nickrent previously A. nigrum in Mexico 607 identified DLN 2018 (L25693) as either A. vaginatum (Nickrent et al. 1994) or A. hondurense (Nickrent et al. 2004). The species identity of this mistletoe (RLM 1083 and DLN 2018), therefore, remains unresolved and requires further study. SUMMARY Although Arceuthobium hondurense, A. nigrum, and A. vaginatum are morphologically similar and often diffi- cult to distinguish from each other in situ, there are several diagnostic characteristics that can be used to iden- tify t h e m in central Mexico where they may be sympatric. Our results support the classification of these taxa as distinct species and the principal morphological and physiological characters that can be used to distin- guish these species are summarized in Table 2. While the overall height of male and female plants and their color cannot be used to easily separate these species, A. hondurense is a more slender plant than both A. nigrum and A. vaginatum. It also has swollen, rounded nodes, particularly near the base of older plants. This character- istic is most evident on older, male plants. In contrast, A. nigrum and A. vaginatum lack swollen, rounded nodes near the base of plants. Another key characteristic of A. hondurense that separates it from the other dwarf mistletoes is the width of its staminate spikes. While the length of staminate spikes often is too variable to be of any diagnostic value, the width of the staminate spikes of A. hondurese are thinner (mean 1.7 mm) compared to those of A. nigrum (2.9 mm) andA. vaginatum (2.0 mm). Furthermore, the staminate spikes of A. nigrum and A. hondurense gener- ally do not form secondary branches, while those of A. vaginatum typically do. Arceuthobium hondurense primarily forms 3-merous flowers, and occasionally 4-merous flowers, but A. nigrum andA. vaginatum commonly form both 3- and 4-merous flowers. Although the adaxial surface of petals of male flowers for A. hondurense andA. nigrum is distinctively dark red, the 3-merous flowers of A. hondurense are smaller (2.5 mm) on average than those of A. nigrum (3.5 mm). The color of male flower petals of A. vagina- turn, however, is dark brown to green. While the fruits of both A. hondurense and A. nigrum are usually mark- edly glaucous, the fruits of A. nigrum are larger than those of A. hondurense as well as A. vaginatum. Addition- ally, A. hondurense and A. nigrum primarily flower from late August through September and October, but A. vaginatum flowers from March through April (Hawksworth & Wiens 1965, 1996). Additional observations of A. nigrum are still necessary to determine if it flowers in the spring as reported by Hawksworth and Wiens (1989, 1996). Our observations of A. nigrum over multiple seasons and years in Durango, Mexico, do not sup- port a spring flowering period for A. nigrum. Furthermore, our analyses confirm that these species can be readily distinguished using ITS-rDNA sequences as previously demonstrated by Nickrent et al. (1994, 2004). The host specificity of these mistletoes may help separate them, depending on the locality in Mexico. In Arango, Arceuthobium nigrum and A. vaginatum both parasitize P. teocote, but P. teocote is less susceptible to A. vaginatum (a secondary host) (Hawksworth & Wiens 1996). Moreover, A. vaginatum does not parasitize P Iftophylla, P lumholtzii, nor P. chihuahuana, which are all highly susceptible to A. nigrum. In central Mexico, the Principal host of A. nigrum is P. teocote, but P. patula is also infected by A. vaginatum there. Now that A. hondu- rense h as been discovered severely infecting P teocote in Oaxaca, Mexico, infection of this pine cannot be used 10 distinguish A. nigrum from A. hondurense, since these mistletoes both flower in the fall, have red flowers, and are similar in size and color. The width of internodes and staminate spikes, therefore, are likely the best char- acters for distinguishing between them. The size of 3-merous flowers, petals, and fruits will also assist in dis- tln guishing A. nigrum (larger flowers and fruits) from A. hondurense (Table 2). Based on our field observations and measurements of plant characteristics of the dwarf mistletoes in **thern Mexico, we do not agree that Arceuthobium nigrum or A. vaginatum occur in Oaxaca or Chiapas, hese species are primarily distributed along the Central Volcanic Cordillera of central Mexico and north into Durango. Arceuthobium vaginatum extends as far north as central Chihuahua in the Sierra Madre Occidental an d as far north as southern Coahuila in the Sierra Madre Oriental (Hawksworth & Wiens 1996). However, the graphic distribution of A. nigrum is centered on the eastern side of the Central Cordillera and extends north rato Durango (Fig. 1). Arceuthobium vaginatum is sympatric with A. nigrum in central Mexico (Hawksworth & 608 Journal of the Botanical Research Institute of Texas6(2) Wiens 1996) and since it also extends as far north as Chihuahua, it is probably sympatric with A. nigrum in Durango. Honduran dwarf mistletoe, which was once thought to be near extinction (Hawksworth & Wiens 1972), is now known to be distributed from northern Nicaragua to northern Oaxaca, Mexico (Fig. 1). Our surveys in 2010 confirmed thatArceuthobiumhondurense occurs in northern Oaxaca (Fig. 1, location 16), so we now know this species occurs almost to Veracruz and Puebla in central Mexico. Moreover, it has only been found in widely-scattered, small populations throughout its geographic range and therefore, should not be considereda common parasite of its pine hosts. It should also be noted that the reports of A. nigrum in Guatemala and 0 Salvador (Hawksworth & Wiens 1977, 1989, 1996) should be considered as reports of A. hondurense. In addi- tion, sinf-p our results demonstrated that at least one of the populations of A. vaginatum from north-central Oaxaca was misidentified by Hawksworth and Wiens (1996) and is indeed A. hondurense (Fig. 1, location 13), we suspect the southern distribution of A. vaginatum only extends into Puebla and not Oaxaca. Further inves- tigations, however, are warranted to assess whether A. vaginatum occurs in Oaxaca as several collections of this species have been made in the Sierra Juarez near Ixtlan. While all of these were classified as A. vaginatum by Hawksworth and Wiens (1996, page 370), we suspect these collections represent additional populations of A hondurense in Oaxaca, but this needs to be confirmed. ACKNOWLEDGMENTS The field assistance provided by Carolyn Daugherty and earlier reviews of the manuscript by Carolyn Daugh- erty, Dan Nickrent, and Job Kuijt are greatly appreciated. We also appreciate the help of Gustavo Perez, Socorro Gonzalez-Elizondo, and Juan Tun Garrido with reviews and Spanish translation for the Resumen. REFERENCES Akaike, H. 1974. A new look at the statistical model identification. IEEE Transactions on Automatic Control AC-19: 716-723. Hawksworth, F.G. and D. Wiens. 1 965. Arceuthobium in Mexico. Brittonia 1 7:21 3-238. Hawksworth, F.G. and D. Wiens. 1970. New taxa and nomendatural changes in Arceuthobium (Viscaceae). Brittonia 22:265-269. Hawksworth, F.G. and D. Wiens. 1 972. Biology and classification of dwarf mistletoes {Arceuthobium ). Agriculture Handbook 401, USDA Forest Service, Washington, DC. Hawksworth, F.G. and D. Wiens. 1 977. Arceuthobium in Mexico: additions and range extensions. Brittonia 29:41 1 -41 8. Hawksworth, F.G. and D. Wiens. 1989. Two new species, nomendatural changes, and range extensions in Mexican Arceu- thobium (Viscaceae). Phytol. 66:5-1 1 . Hawksworth, F.G. and D. Wiens. 1996. Dwarf mistletoes: biology, pathology, and systematics. Agriculture Handbook 709, USDA Forest Service, Washington, DC. of nucleotide sequences. Molec. Evol. 16:1 1 1-120. Larkin, M A, G. Biackshieids, N.P. Brown, R. Chenna, PA McGettigan, H. McWiluam,TJ. Gibson, and D.G. Higgins. 2007. ClustalW and Clustal X version 2.0. Bioinform. 23:2947-2948. Hasegawa, M„ H. Kishino, andT. Yano. 1 985. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. Molec. Evol. 22:1 60-1 74. Huelsenbeck, J.P. AND F. Ronquist. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinform. Appl. Note 17: 754-755. Mathiasen, R.L. 2007. A new combination for Hawksworth's dwarf mistletoe (Viscaceae). Novon 1 7:21 7-221 . Mathiasen, R.L, C.M. Daugherty, and V. Guerra de ia Cruz. 2010. Muerdagos enanos {Arceuthobium spp.) en el sur de Mexico: Distribucion, hospederos y cambios en la nomendatura. In: Proceedings of the XV Simposio Nacional de Parasitoto- gia Forestal, November 18-20, 2009, Oaxaca, Mexico. Pp. 160-166. Mathiasen, R.L., B. Howell, and J. Melgar. 2002b. First report of Arceuthobium hawksworthii in Honduras. PI. Dis. 86:81 5- Mathiasen, R.L, J. Melgar, J. Beatty, C. Parks, D. L Nickrent, S. Slsnie, C. Daugherty, B. Howell and G. Garnett. 2003. New distri- butions and hosts for mistletoes parasitizing pines in southern Mexico and Central America. Madrono 50:115-121- . 2006. First report of Arceuthobium hondurense in Department El Paraiso, Honduras. PI. Dis. > C. Daugherty. 2002a. First report of Arceuthobium hondurense in Oaxaca, Mexico. PI. Dis. Mathiasen, R.L., D.L Nickrent, D.C. Shaw, and D.M. Watson. 2008. Mistletoes: pathology, systematics, ecology, and manage- ment. PI. Dis. 92:988-1 006. Mathiasen, R.L., D. Nickrent, C. Parks, J. Beatty, and S. Sesnie. 2001 . First report of Arceuthobium hondurense in Mexico. PI. Dis. Mathiasen, R.L., C.G. Parks, B.W. Geils, and J.S. Beatty. 1 999. Notes on the distribution, host range, plant size, phenology, and sex ratio of two rare dwarf mistletoes from Central America: Arceuthobium hawksworthii and A. hondurense. Phytol. 84:154-164. Mathiasen, R.L., A. Sediles, and S. Sesnie. 2006. First report of Arceuthobium hondurense and Struthanthus deppeanus in Nicaragua. PI. Dis. 90:1 458. Nickrent, D.L., M.A. Garc(a, M.P. Martin, and R.L Mathiasen. 2004. A phytogeny of all species of Arceuthobium (Viscaceae) using nuclear and chloroplast DNA sequences. Amer. J. Bot. 91 :1 25-1 38. Nickrent, D.L., K.P. Schuette, and E.M. Starr. 1994. A molecular phytogeny of Arceuthobium (Viscaceae) based on nuclear Posada, D. 2008. jModelTest: Phylogenetic modellveraging. Molec. Biol. Evol. 25:1 253-1 256. Rambaut, A. and J. Drummond. 2009. Tracer 1 .5, 2009. MCMC Trace File Analyser. http://tree.bio.ed.ac.uk/software/tracer/. Schwarz, G. 1978. Estimating the dimension of a model. Ann. Stat. 6:461-464. Swofford, D.L 2003. PAUP*. Phylogenetic analysis using parsimony (*and other methods). Version 4. Sinauer Associates, Sunderland, Massachusetts, USA. 610 BOOK REVIEW Alex George. 2011. A Banksia Album: Two Hundred Years of Botanical Art. (ISBN-13: 978-0-642-27739-8, pbk. flexbound). National Library of Australia, Canberra ACT 2600, Australia. (Orders: www.nla.gov. au/). $34.95, 128 pp., color throughout, 9 l A” x 11W. This volume not only comes with a beautiful eye-catching cover, it captures the reader’s attention with its in- troductory paragraph. “The banksia illustrations chosen for this book are the work of botanical artists who are represented in the collections of the National Library of Australia. The images cover a period of nearly 240 years, beginning in 1770 with the art of Sydney Parkinson, who sailed on HM Bark Endeavour under Lieutenant James Cook, and ending in 2007 with two prints by Celia Rosser of Banskia rosserae, a species named in her honour. This intro- duction discusses the artists whose works appear in this publication.” The author provides a quick three page overview, entitled Biology and Ecology, also with three smaller pictures. This is extremely helpful for those of us who have never been fortunate enough to visit Australia. He also thoughtfully provides definitions of Type specimens, Species, Subspecies, and Variety. Occasionally, he adds additional helpful insights and comments Get comfortable, and be prepared to get lost in fascination of these species . — Helen Jeude, Volunteer and Assistant Editor, Botanical Research Institute of Texas, 1700 University Dr., Fort Worth, Texas 76107-3400, USA J. Bot Res. Inst Texas 6(2): 610. 2012 TAXONOMIC HISTORY, REDISCOVERY, AND ASSESSMENT OF THREAT STATUS OF STREBLUS 1LICIFOLIUS (MORACEAE) FROM INDIA Bikarma Singh, Arun Chettri, Dibyendu Adhikari, and Saroj K. Barik 1 Centre for Advanced Studies in Botany i - Eastern Hill University, Shillong 793 022, INDIA ABSTRACT e cit6 por primera vez en 1914 de las Lushai Hills, actus to de S. ihafohus — despues de ui INTRODUCTION The genus Streblus Lour, comprises 25 species (Mabberley 2008; Berg et al. 2006) distributed mostly in tropical and subtropical Asia (Haridasan & Rao 1987). In China, the genus is represented by seven species viz., S. tax- oides (Roth) Kurz, S. tonkinensis (Dubard & Eberhardt) Corner, S. asper Lour., S. indicus (Bureau) Comer, S. tsylmcus (Thwaites) Kurz, S. ilicifolius (S. Vidal) Comer and S. macrophyllus Blume (Fu et al. 2003). There are four Streblus species in India viz., 5. asper, S. ilicifolius, S. zeylanicus, and S. indicus (Haridasan & Rao 1987). Only two species viz., S. asper and 5. ilicifolius were reported from north-eastern India by Kanjilal and Bor (1940). While S. ilicifolius was restricted only to one site in Lushai Hills, S. asper was reported to have more common occurrence in the former undivided province of Assam (Kanjilal & Bor 1940). Haridasan and Rao (1987) also reported the occurrence of S. asper from Baghmara in Garo Hills of Meghalaya. Streblus ilicifolius (S.Vidal) Corner has not been collected and/or reported by any worker since its first re- port by U. Kanjilal in 1914 (ASSAM herbarium/Accession no. 28476, 28477 & 28478) from Lushai Hills in northeastern India under the name Balanostreblus ilicifolia Kurz. These accessions were renamed by G.K. Upadhaya in 2008 as Streblus ilicifolius (S. Vidal) Comer. The species has a very restricted occurrence, and is found only on calcareous habitat. The occurrence of S. ilicifolius in Meghalaya has been recorded and is re- ported for the first time, thereby extending the known distribution of the species in South-East Asia. MATERIALS AND METHODS ^ring the floristic exploration of Nongtrai and its adjoining areas in Sheila, Meghalaya, specimens of Streblus “Kifolius (Fig. 1A) were collected, processed, and housed in the ASSAM herbarium at Botanical Survey of In- ■ Eastern Regional Circle, Shillong, Meghalaya. During the survey, the habitat conditions, associated spe- n, extending to S. Portugal along the on rocky shores and sea cliffs mostly Keywords: Pflllenis maritima (=Asteriscus maritimus), California, halophytes, non-native plants, c INTRODUCTION Pallenis maritima (L.) Greuter was not reported for California in The Jepson Manual (Keil 2012a; Jepson Flora Project 2012), the Consortium of California Herbaria (2012), or other publications that address non-native spe- des growing outside of cultivation in California (Hrusa et al. 2002; DiTomaso & Healy 2007; Dean et al. 2008; Roberts 2008). Pallenis maritima also was not included in the Flora of North America treatment of the Inuleae, “or has it been reported at all, outside of cultivation, for North America (Preston 2006; USDA, NRCS 2012). In this paper, we provide the first documented records of P. maritima for California and North America, w be f e it grows spontaneously outside of cultivation on rocky shores and sea cliffs in coastal Orange County, southern California. We provide voucher documentation, and review horticultural uses, seashore ecology, and a PParent mode of introduction into the State. We also review the taxonomy and nomenclature of the species ^ ,ts Placement within the Asteriscus alliance, in view of the fact that it is still often referred to as Asteriscus B 6(2): 621 -629.2012 lilli 622 , Riefner 11-86 (RSA); City of Dana Point, rocky shore between Mussel Cove and Salt Creek Beach, 33° 29' 2.5 m-5 m, subshrubs and seedlings in rock crevices at the edge of an ephemeral calcareous-saline seep c on the coastal strand, 25 May 2012, Riefner 12-393 (RSA). Patterns maritima has long been known as Asteriscus maritimus (L.) Less., a name that has still some usage at present. As any given species can only bear one correct name once its taxonomic placement and limits are ac- cepted, an overview of the historical background may be helpful to understand the disparity. That background is fairly intricate: the genus in which P. maritima has been placed through time, or to which it is currently as- signed, has been given no fewer than eight different names: Asteriscus Mill., Athalmum Kuntze, Bubonium Hill, Buphthalmum L„ Nauplius Cass., Odontospermum Sch. Bip., Patterns Cass., and Saulcya Michon. For more de- tails, see Greuter (1997). Patterns maritima belongs to a group of 15 Mediterranean and Macaronesian species, together forming a natural unit defined by morphology and chromosome number that has been called the Asteriscus alliance (An- derberg 1994; Francisco-Ortega et al. 1999; Goertzen et al. 2002). These species have been monographed by Wiklund (1983, 1985, 1987). Most are local, little known endemics of North Africa and the Atlantic Islands and need not be considered here. The three most widespread, however, were known to botanists even m pre-lin- naean times. They are the upright, annual, non-pungent Asteriscus aquaticus (L.) Less., the annual to biennial Pallenis spinosa (L.) Cass., characterised by pungent involucral bracts, and the perennial P. maritima, which is discussed in this paper. These three species together formed the original genus Asteriscus, first described and named by Toumefort (1700). The name, he explained, means “little star” and refers to the star-like appearance of the “calyx” (invohi- crum) surrounding the “flower” (capitulum). The remarkably natural Toumefortian concept was maintained by Miller (1754) when he validated the generic name, but was lost when Linnaeus (1737) included Asteriscus in his own artificial new genus Buphthalmum (the name means “ox eye”). Linnaeus (1753) redefined Buphthal- mum considerably, but did not succeed in making it more natural. Of the 11 Buphthalmum species he descnbed and named over the years only two are left today, the remainder are currently placed in no fewer than seven different genera (Jarvis 2012). Greuter (1997) summarized the post-Linnaean taxonomy, which is outlined here. Most 19 th century au- thors essentially reverted to the Toumefortian generic concept of Asteriscus, but split it in two based on the obvious character of pungent vs. unarmed involucral bracts. That is, they left Pallenis maritima and Asteriscus aquaticus together, but separated Pallenis spinosa. The first to do so was Cassini (1822), who named the spiny plants Pallenis (after the borough Pallini E. of Athens— Fournier 1934-1940) and the unarmed ones Nauplius (perhaps, by analogy, after the Greek harbour, Nauplion). Subsequent authors accepted the division but too up the name Asteriscus for one of the genera— unfortunately not always for the same one. The problem was that Asteriscus had not been typified. A type was not designated until very late (Jeffrey 1982), and the monograp Wiklund (1985) refused to accept Jeffrey’s choice of A. aquaticus. Meanwhile Briquet (in Briquet & CavilUer 1917) had discovered profound micromorphological differ- ences between the two unarmed species, placing them at least as far apart from each other as from their spiny relative. By consequence, he placed each of the three in a genus of its own. Wiklund (1985, 1987) went one step further, uniting the superficially dissimilar P. maritima and P. spinosa in one genus (her mis-typified Aste^ iscus”) while leaving Asteriscus aquaticus in the other (which she named Nauplius). Apart from her splitting one deviating species as a monotypic genus Ighermia (Wiklund 1985), which has gained little acceptance, taxonomic conclusions, with the nomenclature rectified, has been generally accepted (e.g., Greuter 2003, > Greuter & Raab-Straube 2006-2009). That treatment, including the repatriation of Ighermia to Asteriscus, received firm support by molecular studies (Goertzen et al. 2002). ._. Back then, to the initial question: which name, Pallenis maritima or Asteriscus maritimus, is correct, answer, as so often: it depends. Accordingly, if one follows the concepts of Toumefort (1700), and wis ^ recognize a single genus comprising the whole Asteriscus alhance, then A. maritimus is correct— a sri ^ cally tenable if somewhat outmoded choice. However, if one honors the taxonomic progress made during last 3+ centuries and maintain two genera, then P. maritima is the preferred choice. Pallenis maritima is a cushion-forming subshrub (a low, spreading woody perennial) that grows up to 40 on tall and about 1 mwide. It is known from the Western Mediterranean region, extending to S. Portugal along the Atlantic Coast, with casual occurrences in Great Britain and N. France (Wiklund 1985). Pallenis maritima grows on sand dunes, cliffs, and rocky shores primarily along the coastline, and it frequently occurs within the salt-spray zone (Beckett 1993; Mucina 1997; NATURA 2003; Estrada et al. 201 1). It is also found m dry grass s ofNorth Africa and Spain, and grows on limestone, clayey soils, including marl, and sandstone (Wiklund 1985). In spring through early summer, P. maritima produces abundant golden-yellow flower heads up to 4 cm wide, with showy, finely toothed ray florets about as long as or longer than the involucre. The leaves and rnvo- hcral bracts of P maritima are narrowly obovate-spathulate, sparsely to densely villous, glandular, and some- what folded. The receptacular scales contain oxalate crystals, and the achenes are irregularly four-sided to subcylindrical and densely strigose-hispid (see Wiklund 1985 for a detailed description). Pallenis mantxma is ^own by the common names of gold coin, Mediterranean beach daisy, and sea daisy. A photograph is pro- ved as Figure 1. Horticultural Use Poflenis maritima exhibits a high tolerance to severe water stress and highly saline irrigation water and there- I°re is useful for revegetation programs and xeroscape gardens (Rodriguez et al. 2005). Owing to diese proper- t * es ' h has been widely cultivated in Europe (Brickell 1996; Walters & Yeo 2000; Vogl-Lukasser & Vogl 2004). Htlfenis maritima is a!so cultivated in California, where it is marketed by the nursery trade as a drought-tolerant indy soils and coastal gardens (Brenzel 2007; California Gardens 2012; San Marcos Growers 2012). s posted on Dave’s Garden (2012) webpage confirm P. maritima is a hardy, drought-tolerant plant for 1 southern California coastal gardens; Dave’s Garden is an on-line forum where people share garden- and information about growing or purchasing ornamental plants. Casual observations by the senior licate P. maritima is often planted in street-side gardens in the coastal communities of Laguna Beach ort Beach, southern California. dwide, escaped garden plants are one of the primary sources of non-native plant introductions, and is no exception (Mack 2000; Reichard & White 2001; Bell et al. 2007). Although the mode of intro- a species can be difficult to identify, documenting whether or not the plant is intentionally cultivated ommercially can be significant (Dehnen-Schmutz et al. 2005; Dean et al. 2008). Because P. maritima to be cultivated in our area, it likely escaped from residential gardens and commercial landscape to become naturalized in seashore habitats along the southern California coast. Salt Tolerance and Specialized Seashore Habitats Halophytes grow naturally in saline environments and have evolved various mechanisms to cope with salinity where other plants cannot thrive (Breckle 2002; Parida & Das 2005). Tolerance to highly saline soils, exposure to salt spray, and periodic inundation are critical factors affecting vegetation and the establishment of non-na- tive species in many habitats of coastal California (Barbour & De Jong 1977; Kuhn & Zedler 1997; Grewellet al. 2007; Pickart & Barbour 2007). Therefore, knowing whether or not a non-native plant is tolerant of saline conditions can have important implications regarding its potential invasive spread and the habitats it might Pallenis maritima was not identified as a halophyte in the databases of Aronson (1989), Menzel and Lieth (2003), or Yensen (2012). Experimental data, however, indicate it is tolerant of high salinity levels, with electri- cal conductivity measurements ranging from 1.25 to 12.61 dS/m-1 during 150 days of treatment (Rodriguez et al. 2005). For Aronson (1989), the minimum criterion required for a plant to be classified as a halophyte is a salinity level with an electrical conductivity of at least 7-8 dS/m-1 during significant portions or all of the plant’s life cycle. More recently, however, halophyte classification has become more demanding, and the threshold of salinity tolerance for a plant to qualify as halophyte has been raised to 20 dS/m-1 (~ 200 mM Nad) (Flowers & Colmer 2008). For comparison, the salt concentration of seawater is approximately equivalent© 50 dS/m-1 (~ 500 mM NaCl). See Grigore et al. (2010) for a review of the history and evolving concepts that Rodriguez et al. (2005) also reported that salt-treated P. r ulate high Na+ and d- mechanism and osmotic adjustment that maintains leaf tugor in response to salt stress. The preferential accu- mulation of either Na+ and/or Cl- may account, in part, for salt tolerance (Nieman et al. 1988; Gibbs etal. 1989; Boursier & Lauchli 1990). Accordingly, P. maritima is a salt-tolerant species and should be considered forindu- sion in future revisions of the world-wide halophyte database. In Orange County, southern California, P. maritima grows on rocky shores or sea cliffs, often within the salt-spray zone and in habitats periodically inundated during storm surge. At one location, it also grows at the edge of an ephemeral calcareous-saline seep. The known distribution of P. maritima in southern California is depicted in Figure 2. Laboratory analysis using saturated extracts of soil samples taken within the root zone of P maritima at each naturalized population indicate the substrate is slightly to moderately alkaline (7.5-8.1 pH), slightly to strongly saline (6.5-34.7 dS/m-1), and slightly to highly calcareous. In addition, the micro-habitat of the* rocky seashore environments is barren or covered with sparse vegetation, often with documented halophyte 5 such as Cakile maritima Scop., Carpobrotus chUensis (Molina) N.E. Br, Distichlis spicata (L.) Greene, Extripk* califomica (Moq.) E.H. Zacharias (syn., Atriplex califomica Moq.), and Limonium perezii (Stapf) Hubb. There- fore, P. maritima functions as a halophyte in coastal southern California, and the habitats it occupies here are similar to its seashore environment along the Mediterranean Sea. Riefrier and Greuter, F 625 There is always some concern about the possible invasiveness of new non-native plants that become estab- lis hed in natural ecosystems. Pallenis maritima is cited as a “casual alien or weed” for Great Britain and is listed ** 3 >tential weed introduction” in Australia, but is not known to be invasive anywhere around the world (GCW 2012; ISSG 2012; PIER 2012; Randall 2007, 2012; Weeds Australia 2012). Wiklund (1985) noted that P. maritima plants kept in isolation produce viable seed, pointing to autoga- ®° Us 0r apomictic reproduction. Additionally, in 2003 a gardener in Huntington Beach, coastal Orange Count y, California, posted a comment on-line indicating that P. maritima may become slightly invasive (Dave’s Garden 2012). Accordingly, P. maritima possesses reproductive and physiological traits that enable it to escape cultivation and become established in seashore environments. We do not know if it will become a pest plant in North America, but it is expected to expand its range and establish elsewhere in coastal southern Califor nia Ornamental horticulture promotes invasions in many ways, and the tradition of selecting and distributing showy species purely for aesthetic reasons remains an important pathway of dispersal (Dehnen-Schmutz & Touza 2008). Aesthetically-pleasing and easy-to-grow plants often become popular with gardeners, and thereby frequently become available for sale in large numbers of nurseries and in expanding horticultural mar- kets. Market availability drives prices, so the more often a species is offered for sale, purchased, and planted, the greater its chance to escape cultivation and find suitable natural habitats (Lockwood et al. 2005; Dehnen- Schmutz & Touza 2008). In addition, biological characteristics that make a plant interesting and easy to grow for gardeners, such as climatic suitability, long blooming period, quick propagation, hardiness, and drought- tolerance, may also enhance the successful establishment of escaped plants in new regions and start an invasion process (Mack 2000; Dehnen-Schmutz et al. 2005; Anderson et al. 2006). We believe salt-tolerance in orna- mentals, especially those selected and marketed for coastal gardens, contributes to their ability to successfully escape cultivation and become established in unintended habitats. Pallenis maritima is one of a growing number of salt-tolerant, hardy, and attractive ornamental plants that have naturalized recently in southern California. Others include Lagunariapatersonia (Andrews) G. Don (Mal- vaceae), Plecostachys serpyllifolia (PJ. Bergius) Hilliard & B.L. Burtt (Asteraceae), and several species of Umo- nium (Plumbaginaceae) (Hill 2012; Keil 2012b; Preston 2012). Plecostachys serpyllifolia and several Limonium taxa are highly invasive species (Riefner & Nesom 2009; BAEDN 2012). Halophytes are receiving substantial attention from botanists and agriculturists as potential fodder crops, food plants, biofuel sources, turf grasses, ornamentals, sand dune stabilizers, and redeemers of salt-affected land, among others (Glenn et al. 1999; Barrett-Lennard et al. 2003; Rozema & Flowers 2008; Yensen 2008). As greater numbers of salt-tolerant plants become cultivated, and as the horticultural trade continues to meet consumers’ demands for hardy and attractive plants for coastal gardens, we can expect to find other non-native halophytes naturalized along the California coast. ACKNOWLEDGMENTS Edward P. Glenn (University of Arizona) and David J. Keil (California Polytechnic State University, San Luis Obispo) provided helpful comments that greatly improved the manuscript. We also greatly appreciate the ef- forts of Harvey Brenneise and Irene Holiman (Library of Rancho Santa Ana Botanic Garden) for assistance with document retrieval, and to Gam Wallace (Wallace Laboratories) for helpful review of saline and calcare- ous substrate data. ± Office of Arid Land Studies, University of 291^ 1 ^ TribC lnUleae ' ln: Asteraceae ' cladistics and classification. K. Bremer, ed. Timber Press, Portland. Pp- Anderson, N.O., S.M. Galatowitsch, and N. Gomez. 2006. Selection strategies to reduce the invasive potential in introduced plants. Euphytica 148:203-216. Aronson, J.A. 1989. Haloph: a data base of salt tolerant plan Arizona, Tucson. Barbour, M.G. and T.M. De Jong. 1977. Response of West Coast beach taxa to salt spray seawater inundation, and soil salinity. Bull. Torrey Bot. Club 1 04: 29-34. Barrett-Lennard, E.G., C.V. Malcom, and A. Bathgate. 2003. Saltland pastures in Australia-a practical guide, ed. 2. Land, Water and Wool, Canberra, Australia. Bay Area Early Detection Network (BAEDN) 2012. Algerian sea lavender: 2010-2011 report. Available at: http://*** baedn.org/index.php?option=com_content&view=artide&id=1%3Awelcome-to-the-bay-area-early-detection- network<emid=2 [accessed June 2012]. 627 Beckett, E 1 993. Illustrated flora of Mallorca. Ed. Moll. Palma de Mallorca. feu, CE, J.M. DiTomaso, and C. When. 2007. Invasive plants pest notes. University of California Agriculture and Natural Resources Publication 741 39. University of California Statewide IPM Program, University of California, Davis. Boorsier, P. and A. LAuchli. 1990. Growth responses and mineral nutrient relations of salt-stressed sorghum. Crop Sci. 30:1226-1233. Breckle, S.W. 2002. Salinity, halophytes and salt affected natural ecosystems. In: A. Lauchli and U. Luttge, eds. Salinity: environments-plants-molecules. Kluwer, Dordrecht. Pp. 53-77. Brenzel, K.N. (ed.). 2007. Western garden book, ed. 8. Sunset Books, Sunset Publishing Corporation, Menlo Park, CA. Brickell, C. (ed.). 1 996. Royal Horticultural Society A-Z encyclopedia of garden plants, ed. 3. Dorling Kirtdersley, London. Briquet, J. and F. Caviluer. 1 91 7. Compositae (suite) [Achillea-Calendula]. In: E. Burnat, ed. Flore des Alpes Maritimes, 6. H. Georg, Geneve & Basel. Pp. 171-316. California Gardens. 2012. Drought tolerant plants for a xeric garden. Available at:httpy/www.californiagardens.com/ Lists/Drought_Tolerant_List.htm [accessed January 2012]. Cassini, H. 1822. Inures, Inuleae. In: Cuvier, G.L.C.F.D., ed. Dictionnaire des sciences naturelles dans lequel on traite methodiquement des differents etres de la nature, 23. Levrault, Strasbourg, Paris. Pp. 559-583. Consortium of California Herbaria. 201 2. Pallenis maritima, Asteriscus maritimus, Nauplius maritimus, and Odontospermum maritimum. Available: http://ucjeps.berkeley.edu/consortium/ [accessed December 201 1 -April 2012]. Dave's Garden. 2012. PlantFiles: Asteriscus maritimus. Available at: http://davesgarden.com/guides/pf/go/60531/ [ac- cessed February 2012]. Dean, E, G.F. Hrusa, G. Leppig, A. Sanders, and B. Ehtter. 2008. Catalogue of nonnative vascular plants occurring spontane- ously in California beyond those addressed in the Jepson manual-Part II. Madrono 55:93-1 12. Dehnen-Schmutz, K., J. Touza, C. Perrings, and M. Williamson. 2005. Ornamental horticultural trade as pathway for invasion. BCPC Proceedings Introduction and Spread of Invasive Species. Pp. 7-12. Dehnen Schmutz, K. and J.Touza. 2008. Plant invasions and ornamental horticulture: pathway, propagule pressure and the legal framework. In: JATeixeira da Silva, ed. Floriculture, ornamental and plant biotechnology: advances and topical issues. Global Science Books, Isleworth. Pp 15-21. DiTomaso, J.M. and E.A. Healy. 2007. Weeds of California and other western states, Vol. 1 , Aizoaceae-Fabaceae. U.C. Agri- culture and Natural Resources Publication 3488, Oakland, CA. Estrada, B., J. Palenzuela, JosE-Miguel Barea, J.M. Ruiz-Lozano, G.A. Da Silva, and F. Oehl. 2011. Diversispora clara (Glomeromycetes)-a new species from saline dunes in the Natural Park Cabo de Gata (Spain). Mycotaxon 1 1 8:73-81 . Flowers, TJ. andT.D. Colmer. 2008. Salinity tolerance in halophytes. New Phytol. 1 79:945-963. FouRNIEfl , p - 1934-1940. Les quatre flores de la France, Corse comprise. Lechevalier, Paris. Franosco-Ortega, J., L.R. Goertzen, A. Santos-Guerra, A. Benabid, and R.K. Jansen. 1 999. Molecular systematics of the Asteriscus alliance (Asteraceae: Inuleae) I: evidence from the internal transcribed spacer of the nuclear ribosomal DNA. Syst Bot. 24:249-266. &B8S ' J -' M - Dra cup, H. Greenway, and J.A. McComb. 1989. Effects of high NaCI on growth, turgor and internal solutes of tobacco callus. J. Plant Physiol. 1 34:61-69. Glenn, E.P., J j. b R0W n, and E J. Blumwald. 1 999. Salt tolerance and crop potential of halophytes. Crit. Rev. PI. Sci. 1 8:227-255. Gu3Bal Comp endium of Weeds (GCW). 2012. Asteriscus maritimus/Pallenis maritima (Asteraceae). Available: http://www. he aror 9 /gcw/species/asteriscus_maritimus/ [accessed January 201 2]. plzen, L.R., J. Franosco-Ortega, A. Santos-Guerra, J.P. Mower, C.R. Under, and R.K. Jansen. 2002. Molecular systematics of the Asteriscus alliance (Asteraceae: Inuleae) II: combined nuclear and chloroplast data. Syst. Bot. 27:81 5-823. Reuter, w. 1 997. Save Asteriscus, sink Nauplius (Compositae). FI - Medit. 7:41 -48. ulep-W. 2003. The Euro+Med treatment of Gnaphalieae and Inuleae (Compositae)-generic concepts and required ”ew names. Willdenowia 33:239-244. 2007 - Med-Checkiist. A critical inventory of vascular plants of the circum-mediterranean countries. 2. Dicoty- ^edones (Compositae). OPTIMA Secretariat, Palermo, Geneve, Berlin. ' AN0 E. von Raab-Straube (ed.). 2006-2009. Compositae. Euro+Med Plantbase - the information resource for uro-Mediterranean plant diversity. Available: http://ww2.bgbm.org/EuroPlusMed/PTaxonDetail.asp?Nameld=141 PTR efFk=7000000/ [accessed June 2012]. L * Bj < J.C Callaway, and W.R. Ferren, Jr. 2007. Estuarine wetlands. In: M.G. Barbour, T. Keeler-Wolf, and A.A. Schoen- 124^ TeiTeStrial ve 9 etation California, ed. 3. University of California Press, Berkeley, Los Angeles, London. Pp. 628 Grigore, M.N., C.Toma, and M. Boscaiu. 201 0. Dealing with halophytes: an old problem, the same continuous exciting chal- lenge. Ann. Alexandru loan Cuza Univ. Ia$i, N.S., Sect. 2, Veg. Biol. 56:21-32. Hill, S.R. 2012. Lagunaria. In: B.G. Baldwin, D.H. Goldman, DJ. Keil, R. Patterson, TJ. Rosatti, and D.H. Wilken, eds.The Jepson manual: vascular plants of California, 2 nd ed. University of California Press, Berkeley. Pp. 882. Hrusa, F., B. Ertter, A. Sanders, G. Leppig, and E. Dean. 2002. Catalogue of non-native vascular plants occurring spontane- ously in California beyond those addressed in the Jepson manual-Part I. Madrono 46:61 -98. Invasive Species Specialist Group (ISSG). 2012. Global Invasive Species Database. Available: http://www.issg.org/database/ welcome/ [accessed January 201 2]. Jarvis, C.E. 2012. The Linnaean plant name typification project. Available: http://www.nhm.ac.uk/research-curation/ research/projects/linnaean-typification/ [accessed June 2012]. Jeffrey, C. 1 982. Generic names of Compositae IV. Additions; corrections; Ar-Az. Compositae Newsletter 1 3:28-38. Jepson Flora Project. 201 2 (v. 1 .0). Jepson eFIora, Asteriscus maritimus/Pallenis maritima. Available: http://ucjeps.berkeley. edu/IJM.html [accessed June 2012]. Keil, DJ. 2012a. Asteraceae. In: B.G. Baldwin, D.H. Goldman, DJ. Keil, R. Patterson, TJ. Rosatti, and D.H. Wilken, eds. The Jepson manual: vascular plants of California, 2 nd ed. University of California Press, Berkeley. Pp. 21 1-433. Keil, DJ. 2012b. Plecostachys. In: B.G. Baldwin, D.H. Goldman, DJ. Keil, R. Patterson, TJ. Rosatti, and D.H. Wilken, eds. The Jepson manual: vascular plants of California, 2 nd ed. University of California Press, Berkeley. P. 401 . Kuhn, N.L. and J.B. Zedler. 1 997. Differential effects of salinity and soil saturation on native and exotic plants of a coastal salt marsh. Estuaries 20:391-403. Linnaeus, C. 1737. Hortus cliffortianus, plantas exhibens quas in hortis tarn vivis quam siccis, Hartecampi in Hollandia, coluit vir nobilissimus & generosissimus. Georgius Clifford. Linnaeus, C. 1753. Species plantarum, exhibentes plantas rite cognitas, ad generarelatas, cum differentiis specifics, nominibus trivialibus, synonymis selectis, locis natalibus,secundum systema sexuale digestas. Salvius, Stockholm. Lockwood, J.L., P. Cassey, andT. Blackburn. 2005.The role of propagule pressure in explaining species invasions. Trends Ecd Evol. 20:223-228. Mack, R.N. 2000. Cultivation fosters plant naturalization by reducing environmental stochasticity. Biol. Invas. 2:1 1 1-122. Menzel, U. and H. Lieth. 2003. Halophyte database version 2.0. In: H. Lieth and M. Mochtchenko, eds. Cash crop halophytes: recent studies. Kluwer Academic Publishers, Dordrecht. Pp. 221-250. Miller, P. 1754. The gardeners’ dictionary ... abridged from the last folio edition, ed. 4. Rivington, London. NATURA 2000 (NATURA). 2003. Interpretation manual of European Union habitats. EUR 25, European Commission DG Environment, Nature and Biodiversity. Nieman, R.H., RA Clark, D. Pap, G. Ogata, and E.V. Mass. 1 988. Effect of salt stress on adenine and uridine nucleotide pools, sugars and acid soluble phosphate in shoots of pepper and sunflower. J. Exp. Bot. 39:301 -309. Pacific Island Ecosystems at Risk (PIER). 2012. Plant threats to Pacific ecosystems, Asteriscus maritimus/Pallenis maritima (Asteraceae). Institute of Pacific Islands, Forestry. Available: http://www.hear.org/pier/ [accessed January 2012]. Parida, A.K. and A.B. Das. 2005. Salt tolerance and salinity effects on plants: a review. Ecotoxicol. Environm. Safety 60:324-349. Pickart, AJ. and MJ. Barbour. 2007. Beach and dune. In: M.G. Barbour, T. Keeler-Wolf, and A.A. Schoenherr, eds. Terrestrial vegetation of California, ed. 3. University of California Press, Berkeley, Los Angeles, London. Pp. 155-179. Preston, R.E. 2006. Dittrichia, Inula, and Pulicaria (Inuleae). In: Flora of North America Editorial Committee, eds. Flora of North America North of Mexico, Vol. 1 9, Magnoliophyta: Asteridae, Part 6: Asteraceae, Part 1 . Oxford University Press. New York and Oxford. Pp. 471-474. Preston, R.E. 2012. Limonium. In: B.G. Baldwin, D.H. Goldman, DJ. Keil, R. Patterson, TJ. Rosatti, and D.H. Wilken, eds.The Jepson manual: vascular plants of California, 2 nd ed. University of California Press, Berkeley. P. 1035. Randall, R. 2007. The introduced flora of Australia and its weed status. CRC for Australian Weed ment of Agriculture and Food, Western Australia. Randall, R. 2012. Rod Randall’s big weed list: Asteraceae. The Global Invasive Species Team. / sive.org/gist/biglist.html [accessed January 2012]. White. 2001. Horticulture as a pathway of invasive plant introductions ir e51:103-113. R.L, Jr. and G. Nesom. 2009. Plecostachys serpyllifolia (Asteraceae) naturalized in California. Phytologia 91 541-564- Riefnerand Greuter, Pallenis r 629 Roberts, F.M., Jr. 2008. The vascular plants of Orange County, California: an annotated checklist. F.M. Roberts Publica- tions, Encinitas, CA. on growth and leaf water rela Rozema, J. and T. Flowers. 2008. Crc San Marcos Growers. 2012. Asteri asp?plant_id=3343 [accessed nufto, and M J. SAnchez-Blanco. 2005. Effects of NaCI salinity and water stress is of Asteriscus maritimus plants. Environm. Exp. Bot. 53:1 1 3-1 23. for a salinized world. Science 322:1478-1480. s maritimus. Available: http://www.smgrowers.com/products/plants/plantdisplay. uary 2012]. United States Department of Agriculture, Natural Resource Conservation Service (USDA, NRCS). 2012. The PLANTS Database. National Plant Data Center, Baton Rouge, LA. Available: http://plants.usda.gov [accessed January 2012]. Vogi-Lukasser, B. and C.R. Vogl 2004. Ethnobotanical research in homegardens of small farmers in the Alpine Region of Osttirol (Austria): an example of bridges built and building bridges. Ethnobotany Res. & Applic. 2:1 1 1 -1 37. Walters, S.M. and P.F. Yeo (eds.). 2000. The European garden flora, Vol. VI: Loganiaceae to Compositae, Dicotyledons: Part IV. Cambridge University Press, Cambridge. Weeds Australia. 201 2. Australian Weeds Committee National Initiative. Asteriscus maritimus, Pallenis maritima. Available: http://www.weeds.org.au/ [accessed January 201 2]. Wikujnd, A. 1983. Ighermia, a new genus of the Asteraceae-lnuleae. Nord. J. Bot. 3:443-446. Wiklund, A. 1 985. The genus Asteriscus (Asteraceae-lnuleae). Nord. J. Bot. 5:299-314. Wiklund, A. 1 987. The genus Nauplius (Asteraceae-lnuleae). Nord. J. Bot.7:1 -23. Yensen, N.P. 2008. Halophyte uses for the twenty-first century. In: M.A. Khan and DJ. Weber, eds. Ecophysiology of high salinity tolerant plants. Dordrecht, Springer. Pp. 367-3%. Yensen, N.P. 2012. Halophyte database: salt-tolerant plants and their uses. USDA-ARS, U.S. Salinity Laboratory, Riverside, California. Available: http://www.ussl.ars.usda.gov/pls/caliche/halophyte.query [accessed April 2012]. Journal of the Botanical Research Institute of Texas 6(2) BOOK NOTICES Adam Cross. 2012. Aldrovanda. The Waterwheel Plant. (ISBN-13: 978-1-9087870-4-0, hbk.). Redfem Natural History Productions, 61 Lake Drive, Hamworthy, Poole, Dorset BH15 4LR, England, UK. (Orders: www. redfemnaturalhistory.com, sales@redfemnaturalhistory.com, inside UK 01202 686585, outside UK +44 1202 686585). £34.99, 248 pp., 91 figures (b/w and color), 6" x 9". From the Publisher: The first comprehensive monograph of Aldrovanda, the rare and extraordinary carnivorous “Waterwheel Plant.” This incredible carnivore, known for trapping aquatic prey between jaw-like lobes that dramatically snap shut when triggered, is related to the famous Venus Flytrap and employs one of the fastest movement responses known in the plant kingdom. Not only arthropods are caught in its deadly jaws, but even small vertebrates including tadpoles and fish fry. Representing over four years of study and research, this long-awaited work presents a pioneering and uniquely detailed treatment of all aspects of the botanical history, ecology, evolutionary history, distribution and cultivation of this novel plant. The text also addresses the pressing need for integrated conservation and Review forthcoming! Robert Lee Riffle, Paul Craft, and Scott Zona. 2012. The Encyclopedia of Cultivated Palms, Second Edition. (ISBN-13: 978-1-609469-205-1, hbk.). Timber Press, Inc., The Haseltine Building, 133 S.W. Second Ave- nue, Suite 450, Portland, Oregon 97204-3527, U.S.A. (Orders: www.timberpress.com, 800-327-5680). $59.95, 517 pp., 950 color plattes, 8 Vi" x 11". From the Publisher: “This essential reference, now completely revised and updated, describes more than 900 species of palms that can be grown in climates ranging from tropical to temperate. With new species, current taxonomy, plant characteristics, cultural requirements such as cold hardiness and water needs, and uses in the landscape as well as Indoors, this is the definitive resource on this economically and horticulturally important plant group.” “ Winner of an American Horticultural Society Book Award in 2004, it also provides history, ethnobotany, “The magnificent new edition includes an expanded discussion of palm health as well as information on gardens with important palm collections that are open to the public. Generously illustrated with more than 730 new photos for a total of nearly 1000, this volume i< Palm lovers have surely discovered this fantastic palm book by now. But this book is not just for pa experts; gardeners, botanists, and taxonomists will also surely find this a useful reference. Aside from the cen- ter piece of 950 colorful plates, the section on Palm Descriptions A to Z is filled with botany, taxonomy, horticultural information. Each genus/species has a phonetic pronunciation guide. I like this and you will toft With this, you too, can roll these Latin names off your tongue in no time and sound like an expert. VASCULAR FLORA AND PLANT COMMUNITIES OF DEAD HOR (RUCKER’S KNOB), MADISON COUNTY, KENTUCKY Derick B. Poindex 632 Journal of the Botanical Research Institute of Texas 6(2) This survey is the first comprehensive flora from an individual peak in the Knobs Region of south-central Kentucky and provides additional information to augment the rich natural history of southern Madison County (e.g., Grossman & Pittillo 1962; Jones & Thompson 1986; Wade & Thompson 1990; Thompson & Fleming 2004; Thompson 2005; Thompson 2008). Furthermore, DHK serves as an area of historic and cul- tural importance due to a small private Revolutionary War cemetery at the summit and the remnants of an old clay pit area at the westernmost foot slope, which was used in the past for the brick and tile industry of Berea College. The objectives of our descriptive study were to: 1) document the vascular plants growing at DHK through collection of representative voucher specimens for deposit in the Berea College Herbarium (BEREA); 2) discuss the present effects of invasive species and livestock disturbance on the native flora and forest vegetation, 3) designate plant communities and describe plant habitats with characteristic associated species; and 4) increase our understanding of vascular plant diversity in preparation of a future publication of the Madison County flora. Additionally, all of these objectives provide a reference baseline for future comparative studies and poten- tial land management decisions. History Pullins-Rucker Cemetery — Dead Horse Knob on topographic maps by Weir (1967) was historically known as Rucker’s Knob to acknowledge two pioneer families who initially colonized and farmed the immediate sur rounding land. A unique feature at the knob summit is an abandoned small private cemetery, the Pulh° s ' 633 Rucker Cemetery. Descendants of the Pullins and Rucker families intermarried and used the graveyard as a burial ground for their relatives. The cemetery dates from Revolutionary War times, with the interment of Loftus Pullins, Jr. (1764-1841), a Revolutionary War soldier with the Virginia Militia who served in the regi- ment of Colonel Sampson Matthew under General George Washington at the Battle of Yorktown in 1781. After his Revolutionary War service, Loftus Pullins, Jr. received a small pension in Kentucky, and he moved to south- ern Madison County with his family in 1787 (E.T. Pullins, pers. comm. 2010). His grandson, Alva Pullins, Jr., married Alma Rucker, a descendant of Jeremiah Rucker, Jr. and Susan Ann Morton, owners of the land that encompassed Rucker’s Knob in the 1860s. One of their sons, John Morton Rucker, farmed the land until it was sold to Joseph and Sallie Coyle in 1888. On September 27, 1898, Berea College acquired “the Jeremiah Rucker Place” of 15.8 ha from the Coyle family including the knob, which then became known as “Rucker’s Knob” (S. Wilson, pers. comm. 2011). Among the few markers in the Pullins-Rucker Cemetery, only Loftus Pullins, Jr. (1764-1841), Celia Pullins (1775-1841), Samuel Pullins (1807-1832), James Pullins (1794-1854), and Susan A. (Morton) Rucker (1815- 1855), are identified. The Daughters of the American Revolution placed an additional marble commemorative marker on the Loftus Pullins grave site on November 8, 1981 (E.T. Pullins, pers. comm. 2010). Brick and Tile Yard . — Another significant historic feature of Dead Horse Knob was the presence of gray clay from weathered New Albany Shale in the vicinity of the western foot slope. This clay was excavated from a group of shallow pits that were used for brick and tile making by Berea College workers in the early 1900s. These obliterated clay cavities range from 1.0-2.0 m in depth and now he in the concave flatlands contiguous to the western foot slope of Rucker’s Knob (Foerste 1906). During the 19th century, Berea College building campaigns needed great quantities of bricks for college buildings and as a labor industry for many male students (Boyce 2006). In 1901, Berea College created a large Brick and Tile Yard adjacent to the western side of Rucker’s Knob where bricks were hand-made by students. In 1902, a brick machine and kiln increased the operation to a high capacity of several thousand bricks per day. By 1906, because of higher production needs, the brickyard industry created up to 25,000 bricks per day from four kilns with excess bricks shipped to other regional cities by a switch line of the Louisville & Nashville Railroad. The Berea College Brick and Tile Yard operation at Rucker’s Knob was closed down in January 1911 (Boyce 2006). Among college buildings constructed with the red bricks from the clay pits at Rucker’s Knob, were the Edwards Building (Men’s Industrial Building/College Square) in 1902-1903, the Phelps Stokes Cha- Pd (Main Chapel) during 1903-1906, and the Frost Building (originally Carnegie Library) in 1904-1907 (Boyce 2006). Physiography Fenneman (1938) classified the physiographic region comprising Dead Horse Knob as the Kentucky Knobs e gion within the Interior Low Plateaus Physiographic Province. Braun (1950) described the southern Ken- tUck y Kn °bs Border Area of the Interior Low Plateaus from Berea to Frenchburg to be included within the Hill Sttbon of the Norman Uplands. Quarterman and Powell (1978) designated this hilly region as the Knobstone ^rpment of the Interior Low Plateaus. Keys et al. (1995) classified the ecological unit, which would include bHK, into the Eastern Knobs Transitional Subsection of the Interior Low Plateau, Highland Rim Section of the “stem Broadleaf Forest Province. Based the most recent classification per Woods et al. (2002), the knobs of southern Madison County Phvc^ted Cntirely within the 400 million-year-old Knobs-Norman Upland Ecoregion of the Interior Plateau jT ,0 graphic Province. These Uplands are interspersed among the western and eastern Outer Bluegrass region, the southern Hills of the Bluegrass Ecoregion, the northeastern Cumberland Plateau Ecoregion, m e Nor them Forested Plateau Escarpment of the Western Alleghany Plateau (Woods et al. 2002). The JSged terrain of the Knobs is characterized by colluvial and residual shale, limestone, and sandstone slopes higher ridgetops often capped by vertical conglomeratic sandstone cliffs inclining down steep V- hiUs and rid ges into broad U-shaped valleys. These landscapes are subsequently drained by numerous led mterraiUent first or der and flowing second order streams. In the Knobs Region, vast open agriculture P**ure lands adjoin and surround the moderately hilly terrain (Woods et al. 2002). Geology The Knobs-Norman Uplands are underlain by Silurian, Devonian, Mississippian, and Pennsylvanian-aged sedimentary bedrock (Woods et al. 2002). At DHK, the bedrock is entirely underlain by the New Albany Shale Formation of the Middle and Upper Devonian System (Weir 1967). To the north and west of DHK, the Knobs- Norman Uplands intergrade into the Outer Bluegrass Ecoregion. The Outer Bluegrass is characterized by broad rolling hills consisting of Upper Ordovician limestone bedrock on open, flatland topography dissected by shallow, narrow streams (Woods et al. 2002). The New Albany shale of southern Madison County consists of carbonaceous, black shale about 24 to 30 m thick containing sparsely crystalline pyrite and concretions of fine-grained calcitic and iron-rich or phos- phatic material. When weathered, this black shale is typically brownish-black to yellowish-brown often with a few thin seams (2.5-4.0 cm)of yellowish-green shales, which locally form abundant chips and plates on outcrops (Campbell 1946). Megafossils are scarce and are mainly comprised of small brachiopods, fish plates, plant parts, and spores (Campbell 1946). In the vicinity of DHK, basal content is evident where shale rests concordantly above the Middle Devonian Boyle Dolomite. Although the DHK summit is only 312 m above sea level, it is unique in being the only solitary, isolated knob to be separate from other higher elevated peaks of the Berea College Forest (BCF). Within the adjacent BCF in southern Madison County, higher knob summits range from Welch Mountain (383 m). West Pinnacle or Barton Knob (453 m). East Pinnacle (458 m). Robe Mountain (465 m), and Pinnacle Knob (487 m) to Bear Mountain (504 m), the highest point in Madison County (Weir 1967; Weir et al. 1971). Within the city of Berea proper on the Devonian black shale Berea Ridge, elevations vary from 267 m at Silver Creek to 316 m on the Berea College campus to 324 m in the Dogwood Heights subdivision (Thompson et al. 2008). Forest Vegetation Deciduous forest vegetation of southern Madison County mainly consists of Oak-Hickory Forest (Kuchler 1964; Woods et al. 2002; Thompson 2008). Braun (1950) classified some forest vegetation of more separated hills in the Knobs Region as examples of Oak-Hickory communities within the Western Mesophytic Forest Region, a transitional mosaic of Oak-Hickory Forest and Mixed Mesophytic Forest. Evans (1991) character- ized vegetation in the Knobs Region as Acidic Sub-Xeric Forest based on topographic moisture, slope aspect, and dominant forest vegetation consisting of Oak-Hickory components with a poorly-developed understory and sparse herbaceous cover. In the oil-bearing Devonian black shale Knobs Region, Wharton (1945) described five upland forest types: oak, oak-pine, chestnut oak-scarlet oak, white oak, and mixed mesophytic, while Muller and McComb (1986) reported white oak, chestnut oak, scarlet oak, and mesophytic hardwoods forest types in their study of upland forests of the Knobs Region. Vegetation studies in the Knobs Region have emphasized the correlation of forest types with soils, site moisture characters, slope position, topographic aspect, and physiognomy over time (Wharton 1945; Braun 1950; Fedders 1983; Muller & McComb 1986; Woods et al. 2002; Thompson 2008). Dry-Mesic Oak-Hickory Forest is the major vegetation, albeit altered, of DHK from the knob summit down to interspersed planted pines stands of upper slopes, middle slopes, lower slopes, and foot slopes at the boundary, with the perimeter being grassland pasture habitat. As mapped by the Web Soil Survey (Soil Survey Staff 2012) from Newton et al. (1973), the 2.6 ha of DHK is comprised of four major forest soil series (Fig. 2). The knob residual and colluvial soils belong to the Coyler- Weikert-Captina Association. These series are characteristically shallow, moderately well-drained, acid in m* aetton (4.5-5.0 pH) from the summit down all steep slope aspects to foot slopes (Newton et al. 1973). Coyler shaley silt loam (CoF) comprises the shallow soils of 0.8 ha (of the total 2.6 ha DHK site), for 310* of the area on the summit and upper western-trending slopes. Coyler soils are clayey-skeletal residuum weath- ered from New Albany Shale on steep 12 to 50 percent slopes. This soil is very strongly acidic in reaction, exces- sively well-drained, channery silt clay loam from 0 to 51.0 cm to the shale bedrock. The soil profile of* 1 * surface horizon is 0 to 12.7 cm deep, brown, friable, fine granular channery shaley silt loam with small Hat* mimm ^ fragments; the B horizon varies from 12.7-38.1 cm deep, yellowish-brown, friable to firm, very heavy channery silty clay with 35 percent black shale fragments; the C horizon is 38.1-50.8 in depth, brown heavy silty clay with 75 percent highly weathered black shale layers down to the unweathered hard, brittle, black shale bedrock of the R horizon (Newton et al. 1973; Fig. 2). The highly eroded Coyler shaley silty clay loam (CsF3) is the principal soil of another 0.6 ha, for 21.8% of the area - on southern and eastern upper and middle 12 to 50 percent slopes. Coyler shaley silty clay loams are different in composition from CoF in having the top brown surface layer lost through erosion and a surface fryer of exposed yellowish-brown silty clay loam present (Newton et al. 1973; Fig. 2). Whitley silt loam soils (WhC) comprise the deep soils of a circular, crescent-shaped band of 0.5 ha, for * 68% of die site, on the southern and western middle slopes, lower slopes, and foot slopes. Whitley silt loam is “ne-silty colluvium derived from weathered New Albany Shale on 6 to 12 percent slopes. This weathered soil Bver y strongly acid, deep, well-drained, silt loam to gravelly silty clay loam from 0 to 122 cm to shale bedrock. J^soil profile is an A horizon from 0-23 cm deep, dark yellowish-brown, friable granular silt loam; the B finnzon ranges from 23-91.4 cm deep, brown to yellowish-brown, moderate to fine and medium blocky heavy r* loam w «fr a few, small black concretions; and the C horizon ranges is 91.4-122 cm in depth of yellowish- ° Wn % day loam with fine, distinct light, brownish gray clay mottles with an abrupt smooth boundary. R horizon beyond 122 cm consists of highly fissured black shale bedrock (Newton et al. 1973; Fig. 2). Whitley silt loam (WhD) soils are in colluvial areas totaling 0.7 ha, for 29.8% of the area, at the base of 12 J? ^ rctm foot slopes at the northern and eastern aspects and part of the southern mid-slope. The WhD soil k simil ar to the WhC soils but differs in a 7.6 cm thick yellowish-brown silt loam surface layer inter- m *ith cl ayey subsoil material (Newton et al. 1973; Fig. 2). The continental climate of Kentucky has warm to hot summers, mild to cold winter temperatures, and gener- ally abundant moisture during all seasons (Trewartha & Horn 1980). Kentucky climatic data for 1971-2000 were obtained from the Berea College Weather Station-150619 (Midwest Climate Center 2011). Mean length of the growing season is 194 days. Median first frost is October 22 and the median last frost occurs April 10. Mean annual precipitation is 120.2 cm with the highest in May at 13.4 cm and the lowest in October at 7.8 cm. Mean annual snowfall approximates 29.7 cm. Mean annual temperature of Berea is 13.6°C with January the coldest month at 1.5°C and August the warmest month at 24.3°C (Midwest Climate Center 2011). METHODS AND MATERIALS Vascular plants at DHK were collected from 26 field trips during the growing seasons of 2010 (11 trips), 2011 (10 trips) and spring 2012 (5 trips). A complete voucher specimen set is deposited at the Berea College Her- barium (BEREA) and a partial duplicate set is filed at the University of North Carolina-Chapel Hill Herbarium (NCU). In the annotated species list, the following data are assigned for each taxon: origin (native vs. exotic), invasive plant pest status, Madison County distribution records, relative abundance, habitat(s) of occurrence, and collection numbers). Gleason and Cronquist (1991), Jones (2005), and Weakley (2011) were the primary manuals consulted for taxon identification. Madison County records were determined based on Campbell and Medley (2012), Clark and Weckman (2008), USDA, NRCS (2012), and from herbarium searches of BEREA and Eastern Kentucky University Herbarium (EKY). A relative abundance value was assessed for each taxon utilizing the definitions of Thompson (2007). Each taxon was assigned a single comprehensive value inclusive for all plant community Nomenclature follows Weakley (2011), with two exceptions mentioned later. Assigned exotic status was also based on Weakley (2011), except where origin was dubious and the species are considered invasive in Kentucky (e.g., Chenopodium album, Dysphania ambrosioides). Invasive pest plant status is from the current list from the Kentucky Exotic Plant Pest Council (KY-EPPC 2012). Plant communities were delineated through field reconnaissance and sampling data of characteristic as- sociated and/or dominant species in conjunction with physical factors (general topography, slope aspect, moisture regimes, soil types, geology) and anthropogenic influences (livestock disturbance, invasive plant impact, farming practices). Diameter at Breast Height (DBH) measurements were made on selected canopy trees with Haglof Swedish Mantax Black Calipers® and transverse core samples for age determination of the largest pine species usings Haglof Swedish Increment Borer®. RESULTS AND DISCUSSION Taxonomic Summary The vascular flora of Dead Horse Knob consists of 291 species (this total includes infraspecific taxa) in 191 genera from 67 families (Table 1). The flora includes two Monilophytes (0.69%), five Gymnosperms (1.72%), and 284 Angiosperms (97.59%), comprised of 71 Monocots (24.40%) and 213 “Dicots” (73.20%), the latter in- cluding eudicots and the paraphyletic grade of “basal angiosperms.” The six largest families in species are the Poaceae (43), Asteraceae (39), Cyperaceae (17), Brassicaceae (15), Fabaceae (15), and Rosaceae (13). Carexis the largest genus with 15 taxa, followed by Quercus with 8 taxa. The 116 exotic taxa (39.86% of the total flora) art all Angiosperms (27 Monocots and 89 “Dicots”). PySek (1998) reported that the Poaceae, Asteraceae, Fabaceae, and Brassicaceae are the four families that accounted for the greatest percentage of exotic species in local flora* from 26 regions in a worldwide survey, and indeed, the DHK flora follows the same trend, as the four families with the most naturalized species are Poaceae (24), Asteraceae (13), Brassicaceae (12), and Fabaceae (ID- state-listed rare, endangered, or threatened species according to the Kentucky State Nature Preserves Commi*' sion list (KSNPC 2010) are present.