CONTENTS Submersion of Dugaldia and Plummera in Hymenoxys (Asteraceae: Heliantheae: Gaillardiinae) Mark W. Bierner ] Bulbiferous Aconitum (Ranunculaceae) of the western United States Donald E. Brink, Jennifer A. Woods and Kingsley R. Stern ? On the identity of two taxa of Berberis (Berberidaceae) from Tibet Tarig Husain, Bhaskar Datt and R.R. Rao Hemerocallis hakuunensis (Liliaceae) in Korea Soon Suk Kang and Myong Gi Chung 20 An undescribed Saccharum (Poaceae: Andropogoneae) from Jammu and Kashmir, northwest Himalaya, India S. Rajeshwari, R.R. Rao and A. Garg 8 Population and site characteristics of a recently discovered disjunct populatior of Croton amensis (Euphorbiaceae) Gregory H. Aplet, Richard D. Laven, Maurya B. Falkner and Robert B. Shaw By Notes on Eriochloa weberbaueri (Poaceae: Paniceae) Robert D. Webster, Paul M. Peterson and Robert B. Shaw Dilunas Taxonomy of the Sida rhombifolia (Malvaceae) complex in India _V.V. Sivarajan and A.K. Pradeep ee Flora vascular de la Sierra de Ia José A. Villarreal Q. 109 Vascular plant type specimens int with a brief history of the herbariun Michael O. Moore and David E. Gié 139 (continued on back cover) CONTRIBUTIONS ¢ VOLUME 16 NUMBER 1 AUGUST, 1994 CONTRIBUTIONS TO BOTANY FOUNDED BY LLOYD H. SHINNERS 1962 & Wm. F. Mahler Publisher 1971-1992 Director Emeritus ® S.H. Sohmer Director Barney L. Lipscomb Editor Botanical Research Institute of Texas, Inc. 509 Pecan Street Fort Worth, Texas 76102-4060 817 332-4441 / 817 332-4112 FAX John W. Thieret Associate Editor Northern Kentucky University Highland Heights, Kentucky 41076 Guidelines for contributors are available upon request and on the inside back cover of the last issue of each volume. Subscription per year: $20. Individual, $30. U.S. Institutions, $40. Outside U.S.; numbers issued twice a year = © SIDA, CONTRIBUTIONS TO BOTANY, Volume 16, Number 1, pages 1-222 Copyright 1994 Botanical Research Institute of Texas, Inc. ISSN 0036-1488 SUBMERSION OF DUGALDIA AND PLUMMERA IN HYMENOXYS (ASTERACEAE: HELIANTHEAE: GAILLARDIINAE) MARK W. BIERNER The Marie Selby Botanical Gardens 811 South Palm Avenue Sarasota, FL 34236, U.S.A. ABSTRACT Dugaldia Cass. and Plummera A. Gray are herein recognized as subgenera of Hymenoxys Cass. Morphologic, cytologic, and chemical data are all consistent with this treatment. The appropriate new combinations and a new name are presented. RESUMEN Se reconocen Digaldia Cass. y Plummera A. Gray como subgeneros de Hymenoxys Cass. Los datos morfoldicos, citol6gicos y quimi Se presentan combinacions nuevas y un nombre nuevo. INTRODUCTION Generic level taxonomy in subtribe Gaillardiinae has long been problem- atic. Depending on which taxa are included in or excluded from the subtribe and how the included taxa are delimited into genera, subtribe Gaillardiinae might contain as few as two genera (Gaillardia Foug. and Helenium L.) or more than 25 genera. With regard to Dugaldia Cass. and Plummera A. Gray, questions have arisen as to whether they should continue to be recognized as genera or submerged within Hymenoxys Cass. When I resurrected Dugaldia (Bierner 1974), those taxa were being treated as congeneric with Heleninm by most workers (e.g., Gray 1874; Standley 1940). It has been clear for some time, however, not only that they are morphologically and chemically very different from taxa in Helenivm, but that they are morphologically and chemically very similar to taxa in Hymenoxys (e.g., Bierner 1978). Likewise, the taxa included in Plummera have been recognized for quite some time as being very similar to taxa in Hymenoxys (Bierner 1978), and it had even been suggested earlier (Turner et al. 1973) that Plammera be combined with Hymenoxys. Consideration of these questions in light of recent micromolecular chemical studies (Spring et al. 1994) and chloroplast DNA restriction site analyses (Bierner and Jansen unpublished) has led me to the conclusion that indeed Dugaldia and Plummera should be combined with Hymenoxys and Sipa 16(1): 1 — 8. 1994 2 SIDA16(1) 1994 recognized as subgenera. For purposes of this discussion, Hymenoxys is considered to include taxa that at times have been segregated into the genera Macdougalia A. Heller, Phileozera Buckley, Picradenia Hook., and Rydbergia Greene, but not taxa that can be segregated as Tetraneuris Greene (see Table 1). DISCUSSION Morphology The taxa of Dugaldia are similar morphologically to Hymenoxys brandegei (Porter ex A. Gray) K.L. Parker, H. grandiflora (Torr. & A. Gray) K.L. Parker, and H. insignis (A. Gray ex S. Watson) Cockerell, which are members of Hymenoxys subgenus Rydbergia (Table 1). In particular, the involucral bracts of these taxa are organized into two to four subequal series, and the outer bracts are usually densely lanate toward their bases. The leaves of the Dugaldia taxa tend to be entire, while those of the Hymenoxys taxa tend to be divided; however, Dugaldia integrifolia (Xunth) Cass. often has upper and middle leaves that are deeply three-toothed, and Hymenoxys brandegei some- times has all of its leaves entire. The P/ammera taxa are very similar morphologically to taxa in several of the Hymenoxys subgenera (Table 1). These similarities include involucral bracts that are in two unequal series with the outer bracts united, and leaves that are divided into linear segments. The disc florets of Hymenoxys and Dugaldia, however, are hermaphroditic, while those of P/ammera are functionally staminate. Cytology Except for reports of dysploidy in Hymenoxys odorata DC. (e.g., Sanderson and Strother 1973) and H. texana (J.M. Coult. & Rose) Cockerell (Strother and Brown 1988) of subgenus Phileozera, the taxa of Hymenoxys, Dugaldia, and P/ummera consistently have the same chromosome number, » = 15 (e.g., Beaman and Turner 1962; Bierner 1974; Sanderson 1973; Speese and Baldwin 1952; Strother 1966; Turner et al. 1973; Table 1). No polyploidy has been reported in any of these taxa, but polyploidy has been reported in some taxa of Tetraneuris Johnston and Bonde 1969; Parker 1970; Speese and Baldwin 1952; Strother 1966, 1972). Flavonoid, Monoterpene, and Sesquiterpene Lactone Chemistry Hymenoxys, Dugaldia, and Plummera are extremely similar chemically (Table 1). All of them produce similar or identical 6-methoxy flavone aglycones, flavonol aglycones, flavonol glycosides, and 3-O-acety! flavonol glycosides (Bierner 1974, 1978, 1994, unpublished; Sanderson 1975; Wagner et al. 1971, 1972a, 1972b). Taxa of Tetranenris, however, have been found to produce 6, 8-dimethoxy flavone aglycones rather than 6-methoxy TABLE 1. Comparison of Dugaldia and Plummera with Hymenoxys and Tetraneuris. TAXON MORPHOLOGY CYTOLOGY FLAVONOIDS AND MONOTERPENES SESQUITERPENE LACTONES Hymenoxys Annuals n=15 6-Methoxy flavone aglycones Guaianolides subg. Hymenoxys Bracts in 2 unequal series Flavonol aglycones Pseudoguaianolides Outer bracts united Flavonol glycosides Seco-pseudoguaianolides Outer bracts not lanate Flavonol 3-O-acetyl glycosides Plants caulescent cee dissec c flowers ie Hymenoxys Annuals n= 3,8, 6-Methoxy flavone aglycones Guaianolides subg. Phileozera* Bracts in 2 unequal series LL 12; Flavonol aglycones Bee vemusi cok des Outer bracts unitec 14, 15 Flavonol glycosides Seco-pseudoguaianolides Outer bracts not lanate avonol 3-O-acetyl glycosides present or absent (H. texana) Plants caulescent Moneiemren e glycosides present Leaves dissected or entir only in H. texana to toothed (H. texana) Disc flowers hermaphroditic Hymenoxys Biennials and perennials n=15 6-Methoxy flavone eee Guaianolides subg. coi Bracts in 2 unequal series Flavonol aglyc Pseudoguaianolides Outer bracts uni Flavonol ee Sic ide es Seco- des oe uter bracts not lanate Flavonol 3-O-acety! glycosides Plants caulescent Leaves dissec Disc flowers pea Hymen Perennial n=15 6-Methoxy flavone aglycones Siege des subg. Mes eae Bracts in 2 unequal series fre Plants caulescent Leaves entire Disc flowers hermaphroditic Flavonol aglycones Flavonol glycosides Flavonol 3-O-acetyl glycosides eudoguaianolides oie ieee anita sAxouswAP] UI viaWwUN][g pure vipresng “wANUAIg Tas_e 1. Comparison of Dugaldia and Plummera with Hymenoxys and Tetraneuris. (continued) TAXON MORPHOLOGY CYTOLOGY FLAVONOIDS AND MONOTERPENES SESQUITERPENE LACTONES Hymenoxys Perennial n= 15 6-Methoxy ae aglycones Guaianolides subg. Rydbergia Bracts in 3 ve series lavonol aglycones Pseudoguaianolides Outer beacts free Flavonol ee Seco-pseudoguaianolides Outer bracts lanate Flavonol 3-O-acetyl glycosides Leaves usually dissected Disc flowers hermaphroditic Dugaldia Perennials n=15 6-Methoxy flavone aglycones Bracts in 2 to 4 subequal Series Flavonol aglycones Outer bracts free or unite Flavonol glycosides Outer bracts lanate Flavonol 3-O-acety! glycosides Plants caulescent Leaves usually entire Disc flowers hermaphroditic Guaianolides Pseudoguaianolides Seco-pseudoguaianolides Guaianolides Plammera Perennials n=15 6-Methoxy flavone oo Bracts in 2 unequal series Flavonol aglycone Pseudoguaianolides Outer bracts united Flavonol cones Seco-pseudoguaianolides Outer bracts not lanate Flavonol 3-O-acetyl glycosides Plants caulescent Leaves dissected Disc flowers staminate Tetraneuris Annuals and perennials n= 14,15, 6, 8-dimethoxy — aglycones Guaianolides Bracts in 2 Saul series 28, 30, Flavonol aglyc Pseudoguaianolides Outer bracts free 45 Flavonol glycosides Outer bracts not lanate Monoterpene glycosides Plants caulescent or ieee Leaves usually entir Disc flowers lea ier ee *H i i ] li lt Phileozera with H. odorata and H th id inl he t it. However, it is anomalous es L ith regard to its leaves (entire to toothed), chromosome number (7 = 3 and 8), and chemistry (monoterpene glycosides present and seco- n thi a eid absent). It has not been examined for flavonoids. OGL (1)OTVAIS BIERNER, Dugaldia and Plummera in Hymenoxys 5 flavone aglycones (Bierner unpublished; Thomas and Mabry 1967, 1968a, 1968b), and none has been found to produce 3-O-acety] flavonol glycosides. With the exception of Hymenoxys texana (tentatively placed in subgenus Phileozera), monoterpene glycosides have not be detected in Hymenoxys, Dugaldia, or Plummera (Spring et al. 1994). Monoterpene glycosides have been found, however, in all of the Tetraneuris taxa (Spring et al. 1994). Hymenoxys (except for H. texana), Dugaldia, and Plummera all produce similar or identical guaianolides, pseudoguaianolides, and seco-pseudogua- ianolides (Bohlmann et al. 1985; Hill et al. 1977; Ivie et al. 1976; Romo de Vivar et al. 1987; Seaman 1982; Spring et al. 1994). Tetraneuris taxa again are somewhat different; they produce guaianolides and pseudoguaianolides, but seco-pseudoguaianolides have not yet been found (Seaman 1982; Spring et al. 1994). Chloroplast DNA Recently, I worked with Robert K. Jansen at the University of Texas at Austin on chloroplast DNA restriction site analyses of many of these taxa (this work is being prepared for separate publication). Subgenera Hymenoxys and Rydbergia were not represented, but one species of Dugaldia (D. hoopesti [A. Gray} Rydb.) and both species of P/ammera were included. In the phylogenetic analysis of the data, the Dugaldia and Plummera taxa were very strongly supported as being in the same branch of the chloroplast DNA phylogenetic tree with the taxa of Hymenoxys subgenera Macdougalia, Phileozera, and Picradenia, while the taxa of Tetraneuris were separated with strong confidence into a separate branch. TAXONOMY Hymenoxys Cass., Dict. Sci. Nat. 55:278. 1828. Type ne s: Hymenopappus anthemoides Juss., Ann. Mus. Natl. Hist. Nat. {Paris} 2:426. 1 Hymenoxys subgenus Dugaldia (Cass.) Bierner, stat. nov. Basionym: Dugaldia Cass., Dict. Sci. Nat. 55:270. 1828. Type Species: Actinea integrifolia Kunth. Oxylepis Benth., Pl. Hartw. 87. 1841. Type Species: Oxylepis danata Benth, Hymenoxys integrifolia (Kunth) Bierner, comb. nov. Basionym: Actinea hess Kunth, Nov. Gen. et Sp. 4:297.t. 410. 1820. Dugaldia a Cass., Dict. Sci. Nat. 55:270. 1828. Cephalophora ee h) Steud., ed. a 1(3):328. 1840. Helenium integrifolia {sic} (Kunth) Benth. & ee fe leek: ath Centr. Amer. Bot. 2:227. 881. ee ens Lee’ (Kunth) Kuntze, vis. Gen. Pl. 1:342. 1891. Type: aie Hipaico: “...inter Omitlan et Serro del a alt. 1400 hex.” Humboldt & Bonpland s.n. ne PD. Oxylepis lanata Benth., Pl. Hartw. 87. 1841. Heleninm lanatum (Benth.) A. Gray, Proc. Amer. Acad. Arts 9:205. 1874. Type: GUATEMALA: “In summo monte Cumbre de Argueta dicto,” Hartweg 593 (HOLOTYPE: K!; tsoryPE: P!). 6 SIDA16(1) 1994 Hymenoxys hoopesii(A. Gray) Bierner, comb. nov. Basiony: Heleninm hoopesii A. Gray, Proc. Acad. Nat. Sci. Philadelphia 1863:65. 1864. Heleniastrum hoopesii . Gray) Kuntze, Revis. Gen. Pl. 1:342. 1891. Dugaldia hoopesti (A. Gray) Rydb., Mem. New York Bot. Gard. 1:425. 1900. Typr: UNITED STATES. COLORADO: * ae Park and west of Pike’s Peak,” Ha// & Harbour 272 (ioLotyPE: GH!; isoryprs: MO!, NY ald: Hymenoxys pinetorum (Standl.) Bierner, comb. nov. Basionym: Heleninm pinetorum Standl., Field oa Pub. Bot. 22: fe 1940. Diugaldia pinetorum (Standl.) Bie Bicone 26:3 1974. Type: CO. Nurvo LEON: “Sierra Madre Oriental, ascent of see Pores by north vere about 20 mi NE of Galeana, abundant in upper pine forest, alt. 3390 meters.” 26 Jul 1934, C.H. G M.T. Mueller 1258 (HOLOTYPE: F!). The species of Hymenoxys subgenus Dugaldia are well defined and have already been treated taxonomically (Bierner 1974). Hymenoxys subgenus Plummera (A. Gray) Bierner, stat. nov. Basionya: Plummera A. Gray, Proc. Amer. Acad. Arts 17:215. 1882. Type Species: Plammera floribunda A. Gray. Hymenoxys microcephala Bierner, nom. nov. BasionyM: Plammera floribunda A. sray, Proc. Amer. Acad. Arts 17:215. 1882. non Hymenoxys PN es (A. Gray) Cockerell, Bull. Torrey Bot. Club 31:485. 1904. Typz: UNITED STATES. Arizona: Cochise Co.: “Apache Pass, Chirricahua [sic} yer Sep 1881, Lemmon & Lemmon 352 (HoLoTyPE: GH!; tsorypes: BM!, F-207616!, GH!, K [photo at F- 1645644!], NDG-061757!, PH-two sheets!, US-47542!; PROBABLE ISOTYPES: 313722!, MO-3726424!, NDG-061758!, NY!). Hymenoxys ambigens (S.F. Blake) Bierner, comb. nov. Basionym: Plammera ambigens S.F. Blake, J. Wash. Acad. Sci. 19:276. 1929. Type: UNITED STATES. ARIZONA: Graham Co.: “Fairly common on lower slopes of Mt. Graham, ca 22 Jul ai Gas Harrison & Kearney 4395 (HovotyPe: US-1436073!, aphote at NMC}; tsor The two taxa - Hymenoxys subgenus Plammera could be treated as varieties of a single species, as suggested by Turner et al. (1973). I agree that morphologic differences between the two are slight, but the taxa appear to be geographically isolated from one another without a zone of contact and intergradation. Furthermore, Spring et al. (1994) found several differences between them with regard to sesquiterpene lactone substituents and side chains. For now, the conservative approach of continuing to recognize them as distinct species seems more appropriate. ACKNOWLEDGMENTS I chank the curators of the various herbaria from which type specimens were borrowed. This work was supported in part by a National Science Foundation Research Opportunity Award supplement to grant BSR-9200707 (Robert K. Jansen). BIERNER, Dugaldia and Plummera in Hymenoxys ij REFERENCES BeaMaN, J.H. and B.L. Turner. 1962. Chromosome numbers in Mexican and Guatemalan Compositae. Rhodora 64:27 1-276. Brenner, M.W. 1974. A systematic study of Dugaldia (Compositae). Brittonia 26:385-392. «1978. Flavonoids of P/wmmera: taxonomic implications at the generic level within the Gaillardiinae. Biochem. Syst. Ecol. 6:293-295. 994. eer raenentee: Pctlinaigenin from Dapaldia pinetorum (Standl.) Bierner. Biochem. Syst. and Ecol. 22:109-11 BoHLMANN, E., L.N. Misra and J. Jakupovic. 1985. Seco-helenanolides from Dagaldia salle Phytochemistry 24:1378—1380. GRAY 874. Notes on —— and characters of certain genera and species. Proc. ee ek Arts 9:187—218. Hu, D.W., H.L. Kin, C.L. oan and B.J. Camp. 1977. J. Agric. Food Chem. 25:1304— 1307, Iviz, G.W., D.A. WirzeL, W. Herz, R.P. SHARMA and A.E. JOHNSON. 1976. Isolation of Hymenovin from Hymenoxys richardsonii (pingue) and Dugaldia hoopesit (orange sneeze- weed). J. Agric. Food Chem. 24:681-—682 JoHNston, M.A. and E.K. Bonpe. 1969. In IOPB chromosome reports XXII. Taxon 18:433-442. ParKER, K.F. 1970. Two new taxa in Texas ines! Sale 20:192. Romo pe Vivar, A., G. DetGapo, M. SoriANO-GarcCIA, R.A. Toscano, A. RUBEN, E. HUERTA and R.G. Reza-GarbDuno. 1987. meer ee na a population of Helenium integrifolium. J. | Prod. 50:284—286. SANDERSON, S.C. 1973. In IOPB chromosome reports XL. Taxon 22:285—-291. a A systematic study of North American and South American a hae species in Pcie (Asteraceae). Ph.D. dissertation, University of Texas at Aus fee ROTHER. 1973. The origin of aneuploidy in Hymenoxys i Nature New Biol. 242:220-221. SEAMAN, E.C. 1982. Sesquiterpene lactones as taxonomic characters in the Asteraceae. Bot. ev. 48:121-595 SpEESE, B.M. and J.T. Batpwin. 1952. Chromosomes of Hymenoxys. Amer. J. Bot. 39:685— 688. SPRING, O., B. ZirreReLt-Harp, M.W. BierNeR and T.J. Masry. 1994. Chemistry - glandular trichomes in Hymenoxys and related genera. Biochem. Syst. and Ecol. 22:17 1 STANDLEY, P.C. 1940. Studies of American plants—X. Publ. Field Mus. Nat. Hist., Bor. Ser. 2127. STROTHER, J.L. 1966. Chromosome numbers in Hymenoxys (Compositae). Southw. Naturalist 11:223-227 «1972. Chromosome studies in western North American Compositae. Amer. J. Bot. 59:242-247. and L.E. Brown. 1988. Dysploidy in Hymenoxys texana (Compositae). Amer. J. Bor. 75:1097-1098. Tuomas, M.B.a rae a Masry. 1967. Isolation, structure | is of | in.anew flavone from Hymenoxys scaposa. J. Org. Chem. 32:3254—3256. and 1968a. Isolation, synthesis, and structure of scaposin, a new flavone from Hymenoxys scaposa DC. Tetrahedron 24:3675-3679. es and «12908: Flavonoid constituents of Hymenoxys scaposa. Phytochemistry 7:787—790. 8 SIDAL6(1) 1994 Turner, B.L., A.M. PoweLt and T.J. Watson. 1973. Chromosome numbers in Mexican Asteraceae. Amer. J. Bor. Pi WAGNER, H., M.A. IvENGAR, E. MICHAHELLES and W. Herz. 1971. Quercetin- 3-(O- acetyl)- B-D- glucopyranosi in ns floribunda and Heleninum hoopesii. Phytochemistry 10:2547-2548 L. HORHAMMER Soma Herz. 197 2a. Flavonol-3-glycosides in Plammera ambigens. Phytochemistry a 197 2b. Flavonol-3-glyco- sides in eight Hymenoxys species. eee ote 11:3087—3088. BULBIFEROUS ACONITUM (RANUNCULACEAE) OF THE WESTERN UNITED STATES DONALD E. BRINK and JENNIFER A. WOODS P.O. Box 671327 Chugiak, AK 99567-1327, U.S.A. KINGSLEY R. STERN Department of Biological Sciences California State University Chico Chico, CA 95929, U.S.A. ABSTRACT Aconitum columbianum Nutt. is a large and camera! variable complex of popula- tions extending from southern Canada to northern Mexico, and from California to New York. Bulbiferous populations have a restricted een occurring in the Cascade Range from northern to southern Oregon, in the Klamath Mountains of northern California, and witha small, disjunct group in the California Sierra Nevada south of Lake Tahoe. These bulbiferous aconites are probably the most distinctive group within the A. columbianum complex, and they are treated herein as a subspecies RESUMEN Aconitum columbianum Nutt. es un complejo grande y variable morfolégicamente de poblaciones que se extienden de ce el sur oi Catads hasta el notte os México, y desde emai. Nueva York. I | $n restringida, encontrandose en la Cordillera de la Cascada desde el norte al sur de Oregén, en las Mentatas Klamath del norte de California, y un pequefio grupo disyunto en al sur del Lago Tahoe en la Sierra Nevada de California. Estos acénitos bulbiferos son probablemente el grupo mds distintivo dentro del complejo A. columbinum, y aquf son tratados como una subespecie. Key worpbs: Aconitum columbianum, bulbifery, taxonomy, distribution. INTRODUCTION Aconitum columbianum Nutt. is a polymorphic complex of populations occurring in the mountains of western North America from British Colum- bia, Canada to Chihuahua, Mexico and from California to South Dakota. Disjunct, outlying populations of A. co/wmbianum occur also in the Driftless area of lowaand Wisconsin (Brink 1982), and in Ohio and New York (Brink & Woods submitted for publication; Htis 1965). Previous investigations indicated extreme differences among regional groups of populations, par- ticularly in floral morphology and other correlated characters. This variation was continuous, however, with no clearly defined breaks, if populations throughout the range of the species were considered in aggregate (Brink SIDA 16(1): 9-15. 1994 10 SIDA16(1) 1994 1975, 1980, 1981, 1982; Brink & deWet 1980). Bulbiferous populations were somewhat different in this regard since they appeared to be consistently distinct from all other A. colwmbianum. The purpose of this investigation was to determine the geographical distribution of these bulbiferous aconites, and to review the taxonomy of this group. MATERIALS AND METHODS Aconitum populations of the United States were compared in field, herbarium, laboratory, and uniform nursery investigations. Geographical distribution of bulbiferous populations was determined by field investiga- tion and by mapping locations from herbarium specimens (CAS, CHSC, CLM, ILL, JEPS, OSC, RM, UC, US, UTC, WILLU, WS). Aconitum plants from bulbiferous and non-bulbiferous populations were transplanted and grown in growth chambers and greenhouses to study bulbil production under controlled conditions. RESULTS In field investigations we found no intergradation between bulbiferous and non-bulbiferous Aconitum columbianum (however, see Hitchcock et al. 1964; Abrams 1944). The populations that we have seen were either bulbiferous, wich all plants of sufficient size and maturity producing conspicuous bulbils in che leaf axils, or they were completely non-bulbifer- ous with no bulbil producing plants. In growth chambers and greenhouses, only plants from bulbiferous populations produced bulbils. Bulbils were illustrated in Brink (1980) and Hickman (1993). Bulbils are an effective means of vegetative reproduction. In natural populations they fall to the ground late in the season where they sprout vigorously, giving rise to new plants. A single leaf axil may produce one or more bulbils. Bulbil production usually is greatest at about the middle of the aerial stem, and diminishes toward the top and bottom of the plant. Most bulbiferous plants bear both bulbils and flowers. On a small percentage of plants, however, bulbifery completely supplants flower production, and bulbils are produced to near the apex of the terminal inflorescence, in place of flowers. A more in-depth description of bulbifery can be found in Brink (1975). Bulbil production should not be confused with the production of one to several small daughter tubers at the first few nodes above the parent tuber, usually below ground. This occurs on a small percentage of the plants in all bulbiferous and non-bulbiferous populations. Bulbiferous Aconitum columbianum have a restricted distribution (Fig. 1), with one group of populations occurring in the Sierra Nevada mountains of California south of Lake Tahoe in El Dorado, Amador and Alpine counties; BRINK, WOODS AND STERN, Bulbiferous Aconitum @ = BULBIFEROUS O=NON-BULBIFEROUS FIG. 1. Distribution of bulbiferous Aconitum columbianum. 12 SIDA1G6(1) 1994 and another group beginning ca. 350 miles away in northern California in the Klamath Mountains of Siskiyou and Del Norte counties. Bulbiferous populations extend from southern to northern Oregon in the Cascade range in Josephine, Jackson, Klamath, Douglas, Lane, Deschutes, Linn, Jefferson, Marion, Clackamas, Wasco, and Hood River counties. Munz and Keck (1968) were uncertain whether the two disjunct groups of bulbiferous Aconitum in California (Fig. 1) should be treated as separate taxa, We can discern no consistent morphological features separating these two groups (Brink 1975). DISCUSSION We have decided to treat bulbiferous Aconitum columbianum at the subspecific level, due in part to the nature of the variation within the species as a whole. The A. co/umbianum complex is extremely polymorphic (Brink 1980, 1982; Brink & de Wet 1980). The existence of geographical races led toa proliferation of published names at the specific, subspecific and varietal levels. This nomenclatural proliferation was often the result of regional studies, which did not consider patterns of variation within the entire Aconitum columbianum complex. Geographical races that appear quite dis- tinct within a regional context may be indistinguishable from other races occurring hundreds or even thousands of miles away. Robust and diminutive races that are regionally distinct are “connected” by intermediate races if the entire range of the species is considered (Brink 1980, 1981, 1982). We have treated Aconitum columbianum as one, large, polymorphic, intergrading complex. Only one group of populations is consistently distinct from all others. These are the bulbiferous aconites, which occur in California and Oregon. We have chosen to distinguish this group as a subspecies because it is geographically and morphologically distinct. Rec- ognition of these bulbiferous aconites at the species level would not seem to be warranted, especially since we have not accorded formal taxonomic recognition to regional variants that are exceedingly diverse, but in charac- ters that intergrade throughout the range of the complex (Brink 1982). TAXONOMIC TREATMENT Aconitum columbianum Nutt. in Torrey and Gray, Fl. N. Amer. 1:34. Tuber toca. 6 X 1.5 cm; “parent” tuber produces 1 (rarely 2) “daughter” tubers with a connective so short the tubers are essentially contiguous. Stem erect and stout to twining and reclining, 2-30 dm long. Leaves 5-15 cm broad, deeply 3—5 (7) cleft, lobes variously incised and toothed. Inflores- cence racemic, terminal, or terminal and axillary. Flowers typically deep BRINK, WOODS AND STERN Bulbiferous Aconitum 13 purple to blue, occasionally white, yellowish, or blue-cinged; 18-50 mm high, helmet 11-34 mm high, pendent sepals 6B—16 mm lon KEY TO SUBSPECIES la. Without bulbils in leaf axils or inflorescence subsp. columbianum 1b. With conspicuous bulbils in leaf axils and/or inflorescence .............. subsp. viviparum la. Aconitum columbianum subsp. columbianum Aconitum arizonicum Greene; A. bakeri Greene; A. cheirophyllum Greene; A. columbianum er? Harrington; A. columbianum var. lutescens M. E. Jones; A. ipl var. ochroleucum A. Nelson; A. columbianum subsp. pallidum Piper; A. divaricatum Rydberg; A. haes te Greene; A. glaberrimum Rydb erg; Pe gracilentum Greene; A. helleri Greene; A. infectum Greene; A. insigne Greene; A. le:bergii Greene; A. eee A. Nelson, A. macilentum Greene: A. sai Greene NG ee ore) A hg oe A. porrectum Seen . ramosum A, gt lson; A. arcnians eene; A. subcaesium Greene; A. tenue Rydberg; A. tricorne Greene; A. uncinatum subsp. noveboracense (A. Gray) Hardin; A. vestitum ee Spring-fed bogs, seep areas, meadows, along streams, and in other wet areas in the mountains of western North America from British Columbia, Canada to Chihuahua, Mexico at elevations of ca. 900—4,000 m. Brit. Col., Wash., Ore., Calif., Nev., Mont., Idaho, Utah, Ariz., S. Dak., Wyo., Colo., and N. Mex. Disjunct, outlying, populations occur at lower elevations (as low as 300 m), in lowa, Wisconsin, Ohio and New York. Chromosome Number: z = 8 [Wiens & Halleck 1962, Cates 1968], 2 = 16 {Longacre 1942, Kawano 1965, Crawford & Gardner 1974, Brink 1975}, 2n = 18 [Crawford & Gardner 1974}. 1b. Aconitum columbianum subsp. viviparum (Greene) Brink aan nov. BasionyM: A. viviparum Greene, Feddes Repert. Spec. Nov. Regni V 2. 1909. Type: OREGON: swamps near Crater Lake, southern Oregon, Cusick 2972 (LecrotyPe, designated herein: WS!; isovecrotyPe: UC!). Aconitum bulbiferum at Fl. NW Amer. 1:25.1897. NonA. ee Saas ers. Acon. 55 al me _ Nels. & Macbr., Bot. Ga LOL cli va, Macbride) C.L. Hicicode: a ae Pacie 7 1964. OREGON: ee eastern slope Cascade Mountains near Mt. Hood, jon a . hanseni Greene, Feddes Repert. Spec. Nov. Regni Veg. 7:3. 1909. CALIFORNIA. A Amador Co.: Hope Valley, 1892, George Hansen. [Note: Hope Valley is actually in Alpine County. } A. ak Greene, Feddes Repert. Spec. Nov. Regni Veg. 7:2. 9. CALIFOR- : S side Mt. Shasta, northern California, Jul 1897, H.E. Brow Spring-fed bogs, seep areas, meadows, along streams, and in other wet areas at elevations of ca. 900—2,500 m in the Cascade Range from northern to southern Oregon, in the Klamath Mountains of northern California, and 14 SIDA16(1) 1994 with a small disjunct group of populations in the California Sierra Nevada south of Lake Tahoe. These bulbiferous aconites were first distinguished as Aconitum bulbiferum Howell in 1897, but this name is rejected asa later homonym of A. bulbiferum Reichenbach, which is part of A. variegatum L., a European aconite. Three specific names were published simultaneously in 1909 by E.L. Greene for bulbiferous Aconitum. One of these names (A. viviparum), was chosen to serve as the basionym for the new subspecific combination, which refers to all bulbiferous Aconitum columbianum. The lectotype was chosen from among the syntypes of the basionym. Bulbiferous populations in California have small flowers and shallow nectaries (Brink 1975), and we suspected that this would be the case for all bulbiferous Aconitum columbianum. We found, however, that bulbiferous aconites in the Mount Hood region of Oregon, at the northern extreme o the range of bulbiferous populations, have relatively large flowers and deep nectaries. Flower color in A. columbianum is typically deep purple to blue; however, color variants include completely white, yellowish, cream-colored, or blue- tinged at the edges of the sepals. Color variants occur occasionally within blue-flowered populations, but there are also white-flowered populations, and groups of populations. For example, a group of white-flowered, non- bulbiferous populations occurs in northeast Oregon and southeast Washing- ton. Consequently, it is not surprising that there are also white-flowered bulbiferous populations, which occur in Alpine and Amador counties of California (see Representative Specimens, below). Chromosome number: 27 = 18, 19, 20 [Brink 1975}. Sane Specimens: Aconitum columbianum subsp. viviparum: CALIFORNIA. Alpine Co.: Hope Valley, 29 Aug 1974, Brink & Mayer s.n. (CHSC) [flowers white, blue- tinged at odgest. eae: Cork . Silver Lake Dam, 14 Aug 1982, McNeal 2750 (OSC) {[white-flowered]. Del Norte Co.: Creek, 6 Aug 1938, Keck 4793 (DS). El Dorado Co.: Lily Lake, 12-14 Aug 1978, ae oe (CHSC); Echo Summit, 29 Aug 1974, Brink & Mayer s.n. (CHSC). Siskiyou Co.: Red Rock Creek, 26 Jul 1974, Brink & Brink s.n. (CHSC); Kangaroo Lake and vicinity, 24 Jul 1974, Brink s.n. (CHSC). OREGON Clackamas Co.: Clackamas Lake, 24 Jul 1927, Peck 15852 (DS). Deschutes Co.: Tumalo Creek, 12 Aug 1906, Whited 3162 (OSC). Douglas Co.: Diamond Lake, 17 Jul 1924, Applegate 4134 (OSC). Hood River Co.: Barlow Pass, Mount Hood, 31 Jul 1934, Thompson 11208 (US). Jackson Co.: Woodruff Meadows, ee 1925, Pendleton s.n. (OSC). Jefferson Co.: Camp Sherman, 27 A 1962, Schoth 1929 (OSC). Josephine Co.: Lake Creek, Siskiyou Mountains, 18 Jul con Whittaker SS217 (/S). Klamath Co.: Pole Bridge, Crater Lake Park, 31 Jul 1922, Applegate 3413 (UC). Lane Co.: Gold Lake, 1 Sep 1962, Dennis 2433 (UTC). Linn Co.: Monument Peak, 9 Aug 1947, Aller s.n. (OSC). Marion Co.: Olallie Meadow, 9 Aug 1978, Dawn 63 (OSC). Wasco Co.: Barlow Road, 1 Sep 1964, French 2604 (OSC). BRINK, WOODS AND STERN Bulbiferous Aconitum 15 ACKNOWLEDGMENTS We gratefully acknowledge anonymous reviewers, J. Bissell, L. M. Mayer, J.M.J. deWet, R.A. Schlising, the herbaria from which loans were obtained (JEPS, OSC, UC, WILLU, WS), and the herbaria that were visited. We appreciate receiving loans through the herbarium of the Cleveland Museum of Natural History (CLM). REFERENCES ABRAMS, L. 1944. Illustrated flora of the Pacific States. Vol. 2. Stanford Univ. Press, Stanford. Brink, D.E. 1975. Aconitum (Ranunculaceae) in California. Master’s thesis. Calif. State Univ., Chico. «1980. Reproduction and variation in oo (Ranunculaceae), with emphasis on oni populations. Amer. J. Bot. 67:263—27 . Biosystematic and ecological oe on ve Aconitum (R eas of the United States. Ph.D. diss. Univ. of Illinois, Urban 1982. Tuberous Aconitum (Ranunculaceae) of the continental ca States: coesholog cal variation, taxonomy and disjunction. Bull. Torrey Bot. Club 109:13—23. and J.M.J. De Wer. 1980. eee variation in nectar production in Aconitum columbianum (Ranunculaceae). Oecologia 47:160-163. and J.A. Woops. Aconitum (Ranunculaceae). Flora of North America. (submitted for oe Cates, R. 1968. ieee ae study of the genus Aconitum (Ranunculaceae) in the ntermountain ee Master's thesis. Utah State Univ., Logan. Crawrorb, D.J. and R.C. Garpner 1974. Documented plant chromosome numbers 1974:2. Sida eee 94. HIckMaN, J.C. (Ed.) 1993. The Jepson manual: Higher plants of California. University of California Press, pa Hitcucock, C.L.A., een M. Ownsey and J.W. THompson. 1964. Vascular plants of the Pacific Morehwest, Vol. 2. Univ. of Washington Press, Seattle. Iris, H.H. 1965. The genus Gentianopsis (Gentianaceae): Transfers and phytogeographic comments. Sida 2:129-154. Kawano, S. 1965. oe of pectinase and cellulase in an orcein squash method. Bot. Mag. (Tokyo) i Loncacre, D.J. 1 a oo of Aconitum noveboracense and A. uncinatum. ull. ‘Torrey a i 69:235-239. Munz, P.A. and D.D. Keck. 2 A California flora and supplement. Univ. of California Press, Berkeley. Wrens, D. and D.K. HaLLeck. 1962. Chromosome numbers in Rocky Mountain plants. 1. Bot. Not. 115:455—464. 16 SIDA16(1) 1994 BOOK NOTICES Kinpscuer, Ketty. 1992. Medicinal Wild Plants of the Prairie. An Ethnobotanical Guide. (ISBN 0-7006-0527-4, pbk.) University Press of Kansas, 2501 West 15th Street, Lawrence, KS 66049-3904. $25.00 (hbk), $9.95 (pbk). 340 pp. The first part of this well-done, authoritative book considers 43 species in detail: common names, Indian names, See! name, ah eer habia, pee used, Indian use, Anglo folk use, medical history, all of these for every plant). A full-page line ee illustrates each of these species. The second part, not illustrated, has brief accounts of 60 species less important as medicine. Range maps are given for all the book’s species. Closing the volume are a glossary, 15 pages of references (ca. 240 entries), and an index. Data in the book are mainly historical; only a few of the species have been studied recently to learn what biologically active compounds they contain.—John W. Thieret. W AsOWSKI, SALLY and ANDy W AsowskI. 1992. Requiem for a Lawnmower. (ISBN 0-87833-811-X, hbk.) Taylor Publishing Company, 1550 West Mockingbird Lane, Dallas, TX 75235. $15.95 (hbk). 182 pp. This book, written “with Andy Wasowski,” is a series of essays about gardening with native plants. The book has a bias toward the Southwest, but the principles presented can be cial eae anywhere. Kinds of plants to use; why and how to use them, attracting mingbirds; our lost native herbs; sae goldenrod s (that they cause hayfever is a ee “unfounded charge”), trash trees (e.g., mimosa, Russian-olive, tree-of-heaven, Chinese tallowtree, white mulberry) to get rid of; creating habitats; desert gardens; a warning about misinformation in many gardening books; and even bladderworts (Utricu- laria), “a wonderful addition toa natua earthen water garden”—these and many other subjects are interestingly v t. Sixteen of the many plants discussed are illustrated with watercolors. The es ee an index, is well worth reading, even by non- gardeners.—John W. Thieret KiRKPATRICK, ZOE MERRIMAN. 1992. Wildflowers of the Western Plains. (ISBN 0-292-79061-9, hbk.) University of Texas Press, Box 7819, Austin, TX 78713-7819. $24.95. 240 pp. Tveren, JOHN and Goria TVETEN. 1993. Wildflowers of Houston. (ISBN 0-89263-319-0, pbk.) Rice University Press, P.O. Box 1892, Hous- ton, TX 77251. $18.50 (pbk), $29.85 (hbk). 309 pp. To paraphrase Ecclesiastes: of making many wildflower books there is no end. These two are fine examples of this genre. toa ers of the Western Plains illustrates about 180 species a alphabetically by family; its text includes, for each, detailed description of the plant, flower, and fruit; notes on range, oe remarks. The composites occupy nearly 1/4 of the book, attesting to the prevalance of ae of that family among at least the more showy flowers of the region. Wildflowers of Houston includes about 200 species grouped by flower color, briefer descriptions than Western Plains, and remarks. Both volumes are jon splendidly illustrated with color photographs of high 1 ne Both have a bibliography anc a detailed index; Western Plains has a glossary. One can do naught but recommend these books highly.—Johna W. Thieret. ON THE IDENTITY OF TWO TAXA OF BERBERIS (BERBERIDACEAE) FROM TIBET TARIQ HUSAIN, BHASKAR DATT and R.R. RAO Taxonomy & Herbarium Division National Botanical Research Institute Lucknow-226001, INDIA ABSTRACT The identity of two taxa of Berberis, B. replicata var. dispar and B. griffithiana var. pallida is discussed. RESUMEN Se discute la identidad de dos taxa del género Berberis, B. replicata var. dispar y B. griffithiana var. pallida. While examining type specimens of Berberis on loan from the British Museum (BM), the authors came across two sheets from Tibet, both annotated by C.M. Hu as Berberis griffithiana Schneid. var. pallida (Hk.f. & Th.) Chamberlain & Hu and were included in the systematic treatment of Berberis by Chamberlain & Hu (1985). Critical examination of these two type sheets revealed considerable differences between them. One sheet (Ludlow & Sherriff 1274) was annotated by Ahrendt as B. replicata W.W. Sm. var. dispar Ahrendt (Fig. 1), and the other sheet (Ludlow, Sherriff & Elliot 72518) annotated by Ahrendt as B. atrocarpa Schneid. var. trimensis Ahrendt (Fig. 2). Chamberlain & Hu (1985) separated the two species, Berberis replicata and B. griffithiana as below: la. Leaves distinctly white below, dull green above, enervate, strongly revolute; shoots slender, terete B. replicata 1b. Leaves glaucous below, noe strongly LE volute, with lateral veins visible and reticulate; shoots + angled B. griffithiana Further examination of L. & S. 1274 revealed all characters of Berberis replicata whereas L. S. & E. 12518 clearly revealed the characters of B. griffithiana. Even Ahrendt (1961) had kept B. replicata var. dispar ander B. griffithiana without citing this particular type specimen under this taxon. Further, the study of floral characters reveals that the apices of the stamens are truncate with long connectives in L. 8S. & E. 12518 while such character is not noticed in L. & S. 1274. The gland on the petals of L. S. & E. 12518 are ovoid, 1.25 mm long, with rounded apices, whereas in L. & S. 1274 the glands on the petals are distinctly ovate, much longer, and with acute apices. SIDA 16(1): 17 — 21. 1994 18 SIDA16(1) 1994 Ce en ee, ee om mB, nl, G2 , te FIG.1. Holotype of Berberis replicata var. dispar (Ludlow & Sherriff 1274, BM). These characters clearly support the separation of these two taxa as distinct. The evidence suggests that Chamberlain & Hu (1985) were correct in reducing B. taronensis vat. trimensis toasynonym of B. griffithiana var. pallida, but that B. replicata var. dispar is closer to var. replicata than it is to B eriffithiana. HUSAIN, DATT AND RAO, Berberis 19 FLORA OPTI 12518 | Sf. 8 as eee 2 oe Pee Ree | BBs eteee ot TYPP SPECIMEN FIG.2. Berberis griffithiana vat. pallida (Ludlow, Sherriff & Elliotut 12518, BM). Further, data through SEM studies of the epidermis of mature leaves of the two taxa also revealed significant differences (Fig. 3), supporting the above view. In case of Berberis replicata var. dispar (Fig. 3, A) the lower epidermal cells of leaves exhibited the convex nature of the periclinal wall cells, with more or less regular cell boundaries and smooth cuticle, whereas in case of 20 SIDAL6(1) 1994 He 4 FIG.3 Scanning electron micrographs of the abaxial surface of the leaf, (A) Berbers replicata var. dispar (Ludlow & Sherriff 1274, BM); (B) Berberis griffithiana var. pallida (Ludlow, Sherriff & Elfiot 12518, BM). Bar = 10 pm. B. griffithiana vat. pallida (Fig. 3, B) the periclinal wall cells are concave in nature, with irregular cell boundaries and the cuticle is more or less striate. The number of stomata also varied in the two species. The former with ten stomata in one field and the latter with six. HUSAIN, DATT AND RAO, Berberis 21 Berberis replicata var. replicata, B. replicata var. dispar and B. griffithiana vat. pallida are separated by the morphological characters in Table 1 Table 1. Berberis replicata var. iii B. replicata var. dispar and B. griffithiana var. pallida are separated by the morphological characte S.No. B. replicata var. replicata B. replicata var. dispar B. griffithiana var. pallida ibs Shoots slender, terete Shoots slender, terete Shoots angled 2, Leaves white below, Leaves white below, Leaves grey below, with enervate, margin strongly vate, margin strongly __ lateral veins clearly visible, revolute rev nel margin not revolute 2. Inflorescence usually 8-12 Inflorescence 2—6 flowered Inflorescence 4—8 flowered flowered 4, Outer sepals 4 mm long, Outer sepals 4-5 mm long, Outer sepals 5 mm long, apices subacute apices acute apices roundec Di. Glands on petals 0.5 mm Glands on petals 1.5-2 mm_ Glands on petals 1.2 mm long long long ACKNOWLEDGMENTS The authors are thankful to the Director, National Botanical Research Institute, Lucknow, for facilities and to the British Museum of Natural History (BM), and the Royal Botanic Garden, Edinburgh (E), for loan of specimens. REFERENCES AHRENDT, L.W.A. 1961. Berberts and Mahonia—A taxonomic revision. J. Linn. Soc., Bot. 57: CHAMBERLAIN, D.F. and C.M. Hu. 1985. A synopsis of Berberis section Wallichianae. Notes Roy. Bot. Gard. Edinburgh 42:529-557 SIDA16(1) 1994 N No BOOK NOTICES LONARD, Robert I. 1993. Guide to the Grasses of the Lower Rio Grande Valley, Texas. (ISBN 0-938738-08-9, pbk.) University of Texas-Pan American Press, Edinburg, TX 78539. Price not given. 240 pp. The introduction to this guide to 63 genera and 183 species of aaa considers nvironment, how to use the book, morphology of grasses, and grasses as elements in vegetation. Following the key to genera are descriptions of these genera and keys to their species. The keys “worked” well for ae 15 species I tried to key out; the descriptions are original and largely adequate. Closing the book are a glossary and illustrations of entire plants and/or diagnostic structures of about 100 grass species, four per page. These illustrations, appearing to be negative photos of line drawings, are of variable quality, some being useful aids, others less so. The index precedes the glossary and illustrations, a user- unfriendly fee hones W. Thieret. BarBour, MICHAEL, BRUCE PAVLIK, FRANK DRYSDALE AND SUSAN LINDSTROM. 1993. California’s Changing Landscapes: Diversity and Conser- vation of California Vegetation. (ISBN 0-943460-17-4, pbk). Cali- fornia Native Plant Society, Sacramento, CA 95814. $24.95. 224 pp. Gary Snyder, the wise California poet, in the preface notes the “elegance and economy” with which this important book makes California’s “vast and complex landscape available ous.” And that the book provides a “vision of a Cauione pwnose native plant populations do not become entirely degraded, and wl le.” It is very significant that the BEERS address ens landscape of vegetation and the natural dynamics that bott perturbed by human influences. Because it is from that perspective, ietiiks with HEdeistending, that we can develop the management biodiversity. Int this way the book is a model to lead Sane states, a role taken by the California Native Plant Society, publishers of this book, and several partner conservation organizations for several decades The book is readable and informative, for both lay and tec innieal audiences. introductory chapter describes the environmental forces that have shaped California’s landscapes as well as the changes that have occurred since European settlement. The following chapters describe general regions, welilscaed wie ners maps, photos, and diagrams. Two pis chapters raise new and important topics: the first on past influences on the vegetation by Native Californians, previously owen to have lived for many millennia in one slush eee without ee influence on its structure led places, a goal that or function; the s will become ever more neon ane to Ech ee the very practical and reasonable vision jas encouraged by Gary Snyder.—Larry Ford. HEMEROCALLIS HAKUUNENSIS (LILIACEAE) IN KOREA SOON SUK KANG and MYONG GI CHUNG Department of Biology Gyeongsang National University Chinju 600-701, THE REPUBLIC OF KOREA ABSTRACT Principal components ana shone pat a type of EG micrantha Nakai ineladed I poy fH ft hanes thern, central, and noethavestens Keres) H lis hak / is the correct name for this taxon. Also, the result shows that H. dnd fi (cencnal a northeastern Korea) is distinct from H. hakunnensis by having a capitate-type inflorescence, large inflorescence bracts, short perianth tube and a different flowering a ai Synonymy, descriptions, and distribution pattern of H. hakuunensis are included. RESUMEN El] andlisis d tes p — muestra ye el tipo de Hemerocallis aang Nakai esta incluido en si asa continuo de variacion i ED Gn LT ih Oa ” 7 , mk Nakai (sur, central y noroeste de Korea) nomb correcto para este taxon. Los resultados muestran también que H. es Cae y noreste de Korea) se diferencia de H. hakuunensis por tener inflorescencias de tipo capitado, a ee la enfigeescencia gence, tubo del PeHAante corto ae un peoge de floracién diferen j p VI H. hakuuensis. INTRODUCTION Hemerocallis L. is an economically important genus of approximately 30 species restricted to mainly eastern Asia (Matsuoka and Hotta 1966). Many species and cultivars are widely grown in gardens in Asia, Europe, and North America (Cohen 1986). Numerous nomenclatural and taxonomic problems exist within the genus (Matsuoka and Hotta 1966; J. Noguchi pers. comm.). The taxonomic difficulty has been attributed to the relative paucity of diagnostic characters, the fact that many species (e.g., H. awrantiaca Baker, H. flava L., H. fulva L., and H. thunbergii Baker) were described from cultivated plants of unknown origin (Kitamura et al. 1986; Matsuoka and Hotta 1966), the extreme difference in appearance between living plants and dried herbarium specimens, and possible widespread hybridization (Kitamura et al. 1986). In addition, many species of Hemerocallis are so variable ecologically and morphologically that a proper species concept requires morphological, ecological and biosystematic studies. Over 27,000 cultivars further confound the taxonomic status of several Hemerocallis SIDA 16(1): 23 — 31. 1994 24 SIDA16(1) 1994 species. Matsuoka and Hotta (1966) noted that Baily (1930), Nakai (1932), and Stout (1941) did not consider the variability of natural populations when developing their classification. Based on the taxonomic literature on the genus Hemerocallis (e.g., Nakai 1932; Matsuoka and Hotta 1966; Hotta 1986, Kitamura et al. 1986), the geographical and ecological distribution of the Japanese Hemerocallis is relatively well known. On the other hand, litle is known of the Korean Hemerocallis. After Nakai’s (1932) description of H. coreana Nakai as the endemic Hemerocallis species in Korea, he (Nakai 1943) described two other taxa; H. hakuunensis Nakai and H. micrantha Nakai. Matsuoka and Hotta (1966) recognized H. fu/va var. minor (Miller) M. Hotta, H. flava var. coreana (Nakai) M. Hotta, H. dumortieri Morren var. middendorffii (Tr. et Mey.) Kitamura, and H. hakuunensis as the indigenous species in Korea. Lee (1985) reported 6 species and two cultivated plants from Korea: H. fulva L. (cultivated plant), H. /ulva var. kawanso Regel (cultivated plant), H. flava, H. littorea Makino, H. dumortiert, H. middendorffit, H. minor and H. thunbergii. The recognition of Korean Hemerocallis species has varied depend- ing on authors. Consequently, there is an obvious need to study the group in detail to understand better the taxonomy of Hemerocallis. Principal components analysis of morphological data has been widely used to study species delimitation; e.g., Noguchi (1986) for the Hemerocallis pai: complex and Chung et al. (1991) for the Korean Hosta (Liliaceae). The goals of this study were to 1) analyze morphological variation between H. hakuunensis and H. middendorffii; 2) determine the overall geographic distribution patterns for H. hakuunensis and H. middendorffii; and 3) more closely conform the descriptions to the typical morphology of H. hakuunensis. More specifically, do the two species H. hakuunensis and H. micrantha merit taxonomic distinction at the rank of species? MATERIALS AND METHODS Several morphological characters were measured for five individuals from 16 populations in Korea from 1988 to 1993 (Fig. 1; Table 1). Observations were also made of the habitats, and notes were ralceh, on characteristics (€.g., soil types) of the populations. Voucher specimens of all collections are deposited at GNUC. In addition, five herbarium specimens of H. middendor- ffi and the two types of H. hakuunensis and H. micrantha (Table 1) were used to determine if 1) the types of the latter two names fall within one continuous field of variation and 2) H. hakuunensis and H. micrantha are morphologically distinct by comparison with H. middendorffii. The herbaria of Seoul National University (SNU) and Tokyo University (TI) were visited in order to search type specimens of H. hakuunensis and H. micrantha and determine the total range of geographic variation and the distribution of North Korean Hemero- callis (Fig. 1 KANG AND CHUNG, Hemerocallis 25 FIG. 1. Geographical distribution of Hemerocallis hakuunensis and H. middendorffii in Korea. Arrows with alphabetic codes (A-N, P and Q) indicate locations of populations from which samples were collected by M. G. Chung and &.S. Kane. Arrows ete codes (R-X) indicate locations from which herbarium specimens were collected. See Table 1 for the explanation of 23 alphabetic codes. 26 SIDA16(1) 1994 Taste 1. Locations of 16 populations and seven herbarium specimens examined for morphological oe analysis. All collections by M. G. Chung and S. S. Kang except others noted. Code — Locations Collection number 1) Hemerocallis hakuunensis, southern, central and peers Kore A ram-ri, Chi on-myeon, Uich’ang-gu ieee 1850 B Sorang-ri, Koje-myeon, Koje-gun, Koje Island, Gyeongsangnam-do 1851 Cc Mizo-myeon, Namhae-gun, Namhae Island, Gyeongsangnam- 1852 D — Masa-ri, Bonggang myeon, cape run, Chollanamdo 1853 E Pyeongmok-ri, Imwhe-myeon, Chindo-gun, Chin Island, Chollanam-do 1863 F Ch’osang-ri, Uishin-myeon, Chindo- -gun, Chin Island, Chollanam-do 1864 G ~~ Gurea-ri, Leeyang-myeon, Whasun-gun, Chollanam-do 1866 H — Yongdam-ri, Kumnam-myeon, Yeongi-gun, Ch’ungch’ongnam-do 1871 I Changgi-ri, Anmyeon-eup, Seosan-gun, Ch’ungch'’ongnam-do 1874 J Hwayangdong Provincial Park, Ch’ungch’ongbuk-do 1875 K — Danyang-gun, Sobaksan National Park, Ch’ungch’ongbuk-do 1878 L Yuljeon-ri, Dae-myeon, Hongch’on-gun, Kangwon-do 1881 M_ Angi-ri, Bookhu-myeon, Andong-gun, Gyeongangbuk-do 1884 N Sangbuk-myeon, Uljoo-gun, Gagisan Provincial Park, Gyrongsangnam-do 1887 P Komnam-myeon, Ch’olwon-gun, Kangwon-do te Kwangnung, Sohul-myeon, Poch’on-gun, Gyeonggi-do R Holotype, Mt. Baekun, Chollanam-do Nakai 5.n. eis 2) Hemerocallis micrantha, Hamyang-gun, Gyeongsangnam-do S Holotype, Hamyang-gun, Gyeongsangnam-do O. Syogo s.n.( TI) 3) Hemerocallis ae central and northeastern Korea, Manchuria, Amur T Ch'ungjin-shi, Hamgyeongbuk-do Nee sn. (TI) U- Mt. et Gyeongsung-gun, Hamgyeongbuk-do Toh 4507 (SN a Vo Mt. Myohwang, Hwich’un-gun, P’yeonganbuk-do Toh G Shim i 278 (SN W Kwangnung, Poch’un-gun, Gyeonggi-do Toh 900 VNU) X ~~ Mc. Myeon, Taebaek-shi, Kangwon-do ne 2. (SNU Previous studies (Matsuoka and Hotta 1966; Hotta et al. 1966; Hotta 1986; and Noguchi 1986) revealed that several qualitative (e.g., flowering time and period, shape of roots, odor, and type of an inflorescence) and quanti- tative characters (e.g., floral and leaf morphologies and wintering conditions of leaves) are important characters in recognizing Hemerocallis taxa. As most qualitative characters were not available from the herbarium specimens, only 14 morphological characters were selected in this study (Table 2). Phenetic analyses were utilized to assess the morphological variation present in 87 OTUs (operational taxonomic units) of H. hakuunensis, H. micrantha, and H. middendorffiz. Principal components analyses (PCAs) were performed to identify characters contributing to the separation of each taxon and assess inter-taxon variation using the Statistical Analysis System (SAS 1987). The data were standardized for each character, with the raw data matrix transformed such that each character had a mean of zero as a standard deviation (Sneath and Sokal 1973). The first three principal components were extracted by the PCAs. KANG AND CHUNG, Hemerocallis 27 TABLE 2. List of 14 characters used in the morphological analysis. Acronym Character derivation Unit or Category PSH Plant (scape) height > 50cm=0; < 5O0cm=1 I Length of inflorescence minus flowers > lem=0; 2-20cm=1 NES Number of flowers/scape # LLB Length of the lowest bracts cm WLB Width of the lowest bracts cm LPO Length of the perianth tube enclosing ovary cm LIP Length of the inner perianth cm WIP Width of the inner perianth cm LOP Length of the outer perianth cm WOP Width of the outer perianth cm WWL Width of the widest leaves cm ROL Ratio of inner to outer lobe length ratio ROW Ratio of inner to outer lobe width ratio SOI Shape of inflorescence head or cap ty Y-type or aie 1 RESULTS The cumulative variance of the first three principal components (64.8%) and the contributing value of each of 14 characters are shown in Table 3. Characters weighted heavily in the first principal component (27.4%) are width of the outer perianth (WOP), width of the inner perianth (WIP), shape of inflorescence (SOI), and plant height (PSH). The second component accounts for 25.0% of the total variation and is mostly concerned with length of the perianth tube enclosing the ovary (LPO), SOI, PSH, length of inflorescence minus flowers (LSJ), length of the outer perianth (LOP) and length of inner perianth (LIP). The third component accounts for 12.4% of the total variation and number of flower per scape (NFS), length of the lowest bracts (LLB), and width of the widest leaves (W/WL) are highly weighted. Results from two-dimensional plots of component 2 against components 1 and 3 (Figs. 2 and 3) show that the type specimen of H. micrantha is margin- ally included in the continuous range of variation described by the OTUs of H. hakuunensis (populations A-N, Pand Q) including the type (R). The range of variation for H. hakuunensis is distinct from that of H. middendorffi. TAXONOMIC DISCUSSION Hemerocallis middendorffiz, found on open areas of central and northeastern mountains in Korea(S.S. Kang pers. obs.;S.H. Oh pers. comm.), differs from H. hakuunensis by having a capitate-type inflorescence; large inflorescence bracts (2.0—-4.0 cm long, 1.0—3.0 cm wide); short perianth tube (1.0-1.8 long); short scape (30-47 cm long); and a different flowering period ae middle May to June. The results from principal components analysis revealed these two species are distinct from each other (Figs. 2—3). 28 SIDALG6(1) 1994 Table 3. First three principal components composed of 87 OTUs derived from 14 morphological characters. Acronym* PC | PC 2 PCS PSH -0.625 0.724 -0.034 LSI -0.503 0.694 0.207 NES -0.434 0.098 0.608 LLB 0.048 0.140 0.580 WLB 0.602 -0.194 0.502 LPO -0.075 0.722 -0.072 LIP 0.574 0.657 -0.317 WIP 0.790 0.271 -O.115 LOP 0.569 0.668 -0.308 WOP 0.828 0.31° 0.159 W 0.231 0.445 0.572 ROL -0.131 -0.301 0.082 ROW -0.453 -0.196 -0.406 SOI -0.625 0.724 -0.034 Eigenvalue 3.830 3.496 1.742 Proportion of variance 27.4% 25.0% 12.4% Cumulative variance 27.4% 52.4% 64.8% “See Table 2 for character explanation. Nakai (1943) originally described H. micrantha based on only one herbarium specimen collected by O. Syogo from Hamyang, Prov. Gyeongsang Nam of southern Korea. He noted that H. micrantha is distinct from other species by having small-sized orange-yellow flowers and a highly branched inflorescence. For example, he reported that the length and width of perianth lobes are 4 cm and 5—6 mm, respectively, and the length of inflorescence is 23 cm long. These values are included in the continuous range of variation observed during our field trips for H. bakuunensis. Individuals with 20-30 cm long inflorescences were encountered within the populations of D, K, and M. Individuals with small-sized (ca. 4—6 cm long) length of perianth lobes were also observed within the populations of A, B, D, E, 1, M, and N. In addition, the flowering date (17 July) and collection location (Prov. Gyeongsang Nam of southern Korea) for H. micrantha fall within the range of flowering period (June to July, August in the mountain- ous areas) and geographic distribution (southern, central, and northwestern Korea) of H. hakuunensis. It is highly probable that the type of H. micrantha is one of the variable individuals observed for H. hakuunensis as revealed by PCAs (Figs. 2—3). Although H. bakuunensis and H. micrantha were published in 1943 on the same volume of Journal of Japanese Botany (pp. 315-316), H. hakuunensis has priority over H. micrantha because the former was described ahead of the latter. So H. Aakuunensis is the correct name for the taxon. — KANG AND CHUNG, Hemerocallis a) QF ecco OL =a —2/ 1 it ] -4 -3 -2 at K kK ? P K @ D 1k Pp oD 4 LQ Q Q P HIQ P K M Je B F Lo Re F HHC JCA DCF th s OF D c ad 1 ee FA oM BBA tm A I a B ME H x -2/ x : Cosa -3 f \ ® = ‘ 4 4h | ! | i= —2 -1 1 2 3 PC3 . Diagrams showing oo on of PCAs. th and tm = holotype of Hemerocallis A (components | and 2) composed of 87 of 87 OTUs beanie end H. micrantha. FIG. 2 FIG. 3. PCA (components 2 and ie composec jon 30 SIDALG6(1) 1994 Matsuoka and Hotta (1966, p. 37) also considered H. micrantha as a synonymy of H. hakuunensis. The nomenclature 1s as follows: Hemerocallis hakuunensis Nakai, J. Ja Bot. 19:315. 1943. Type: KO- REA. Prov. CHOLLA Nam: Met. Baekun, 22 Aug 1934, Nakai s.n. (HOLOTYPE: TI!; ISOTYPE: TI!). ParatyPE: KOREA. Prov. CHoita Nam: Me. Chiri, 25 Jul 1937, 0 Syogo s.n. (TT). gins micrantha Nakai, J. Jap. Bot. 19:315-316. 1943. Type: KOREA. Prov. EONGSANG Nam: Hamyang, 17 Jul 1937, O. Syogo s.n. (HOLOTYPE: TT!). Herbaceous perennials. Roots usually tuberous (ca. 1.5—2.0 cm long, 0.8 cm wide), grayish yellow. Leaves 40-100 cm long, 1.2—2.5 cm wide, greenish yellow. Scapes ascendant or erect, 32-145 (80 + 15.62, mean + SD) cm long, nearly as long as or shorter than leaves, usually 1—2 lanceolate bracts below the inflorescence, 1.2—6.0 (9.5) cm long, 0.7-1.8 cm wide; inflorescence branched (rarely Y-type) with 4—27-flowers, 3-30 (10.82 + 4.43) cm long; bracts ovate, greenish, membranous on margin. Perianth (fresh) orange-yellow, 7-10 cm long, tube 1.5—3.0 cm long, green with orange tint; throat orange ae inner perianth lobes 4.5—9.0 cm long, 1.6—2.6 cm wide. Stamens 3—4 cm long, inserted; filaments orange-yellow, attached to the base of the on tube; anthers ca. 6 mm long, dark brown with purple tint. Style filiform, ca. 8.5 cm long, exerted beyond the stamens. Capsule usually oblong-oval, 2.0—2.5 cm long and 1.1—1.5 cm wide, cross- wrinkled when dried, surface usually covered with wart-like projections, apex emarginate; seeds shining black, angled, 6 mm long, 4 mm wide. Flowering from June to July (August in the mountainous areas) in Korea; fruit ripening late July to September. Korean name: Baekunsan-wonch'uri This species is commonly found on the humus or granitic soils and open areas or under pine-oak forests on hillsides of southern, central, and northwestern Korea, including Kojae, Namhae, Komun, Dolsan, Pogil, Chin, Anmyeon, and Kangwha islands (Korean endemic species). ACKNOWLEDGMENTS We thank Jungo Noguchi and Mitsuru Hotta for discussions on the classification of Hemerocallis. Special thanks go to Sun Gi Chung, Duk Seo Gu, and Ou Kun Chin for their company on field trips. Also, appreciation is extended to Barney Lipscomb and two anonymous reviewers for com- ments on the manuscript. We are also grateful to the directors and staffs of the National Parks and the Department of National Monuments of Culture Property preservation Bureau of Korea for permission to collect samples from Hallasan, Hanryohaesang, Sobaeksan, Sokrisan, Tadohaesang National Parks and Hong and Cheju islands. We especially thank the herbaria KANG AND CHUNG, Hemerocallis 31 directors and curators of GA, KYO, MO, SNU and TI. This research was supported in part by a National Science Foundation Dissertation Improve- ment Grant (BSR-8914430) to MGC, a Korea Research Foundation Non Directed Research Fund (1992) to MGC, and a Korea Science and Engineer- ing Foundation Grant (931-0500-031-2) to MGC. REFERENCES Battey, L. 1930. Hemerocallis: The day-lilies. Genetes Herb. 2:143—156. CHUNG, M.G., S.B. Jones, J.L. Hamrick and H.G. CuHunec. 1991. at ee and sozyme analysis of the genus Hosta (Liliaceae) in Korea. Plant Species Biol. 6:55—G69. ne S. 1986. The well- pisted ee Garden 16:19 (Sep/Oct). Horta, M. 1986. Hemerocal ntiaca group in northern Kyushu, Japan. Acta Phytotax. Geobot. 37:17-21. | Gin cane Krramura, S., G. Murata and T. Koyama. 1986. Colored illustrations of herbaceous plants ae {fonocaryledoneae. Hoikusha Publ. Co., Ltd., Osaka, Japan. (in Japanese) Lee, T.B. 1985. Illustrated flora of Korea. Hyangmoonsa, a . in Korean) Matsuoka, } ne and M. Horta. 1966. Classification of Hemerocallis in Japan and it’s vicinity. Acta Phytotax. Geobot. 22:25—43. (in Japanese) ne : 1932. Hemerocallis japonica. Bot. Mag. Tokyo 46:111—123. T. 1943. Notulae ad plantas Asiae Orientalis (XXVIII). J. Jap. Bot. 19:315-316. ou 1987. SAS/STAT Guide for personal computers, ver. Ged. SAS Inst. Inc., SNEATH, P.H.A. and R.P. SOKAL. 1973. penance taxonomy: The principles ae meee at numerical classification. Freeman Co., San Francisco, CA. Strout, A.B. 1941. Memorandum on a le oe of the genus Hemerocallis. Herbertia 8:67-71. SIDA16(1) 1994 eo ne) BOOK NOTICES INTERNATIONAL INSTITUTE FOR SUSTAINABLE DEVELOPMENT (IISD). 1992. Sour- cebook on Sustainable Development. (ISBN 1-895536-04-9, pbk). IISD, Winnipeg, Manitoba, Canada R3B OY4. $20.00. 133 pp. The sourcebook is an invaluable reference for individuals and organizations seeking information related to the goals and issues of * — nek ae which was raised to international atrention at the United Nations Conference on Environment and Develop- ment. The sourcebook also provides lists and dance for networking with worldwide organizations ee in this field. In addition to concise descriptions of such organizations and an extensive bibliography categorized by topics and regions, the sourcebook describes audio-visual organizations and materials available as well as a detailed ee to relevant computer networks, conferencing systems, and databases.—Larry Fore BANURI, TARIQ AND FREDERIQUE APFFEL MARGLIN (Eds). 1993. Who Will Save the Forests?: Knowledge, Power and Environmental De- struction. (ISBN 185649-159-5, hbk; 1985649-160-9, pbk). The United Nation University, World Institute for Development Econom- ics Research, Helsinki, Finland; Zed Books Ltd., London, England. $55.00 (hbk); $17.50 (pbk). 195 pp. The editors of this onan oriented volume compiled four case studies of forest communities to examine “the contrasting visions and attitudes toward resource conserva- jos tion implicit in Bre oo non-modern societies [and] the lessons that these visions anc practices provide for coping with the environmental crisis.” The case studies are: (1) a hill tribe practicing shifting ental on. in India; (2) a pedsartt community living in a ie modernized afforested region of Indi hold x in Finland; and (4) forest workers and (oteeeie: in Maine. hay Ford. AN UNDESCRIBED SACCHARUM (POACEAE: ANDROPOGONEAE) FROM JAMMU AND KASHMIR, NORTHWEST HIMALAYA, INDIA S. RAJESHWARI Botanical Survey of India Debra Dun-248 195, INDIA R.R. RAO and A. GARG National Botanical Research Institute Lucknow-226001, INDIA ABSTRACT A new species of Poaceae, Saccharum stewartii Rajeshwari, R.R. Rao & Garg from Jammu & Kashmir, northwest Himalaya, India, is described and illustrated. Palynology of the new species is analyzed and compared with its closely allied S. griffithit Munro ex Boiss RESUMEN Se describe e as una especie nueva de Poaceae, poaaatad sStewartit Rajeshwari, R.R. Rao & Garg, de Jammu y Cachemira, noroeste del Himalaya, India. Se analiza y compara la palinologfa de esta especie con la de su pariente cercana S. ee Munro ex Boiss. The genus Saccharum L. (sensu lato), with ca. 35—40 species, is distributed in the tropics but extends to warm temperate regions of the world. In India, the genus has ca. 16 species mostly in the tropical belt (Bor 1960). During a taxonomic study on grasses of Jammu & Kashmir, the authors came across an interesting specimen of the genus collected from the Kashmir Valley and carrying adeterminavit slip by N.L. Bor bearing the name Erianthus stewarti1 Bor. Critical examination of the specimen indicated that this is an unique species morphologically very near to Saccharum griffithii Munro ex Boiss. However, the binomial Erianthus stewartii was never published and further, as the species of Erianthus are now transferred under Saccharum, a new binomial, Saccharum stewartii is necessary and described here. Saccharum stewartii Rajeshwari, R.R. Rao & Garg, sp. nov. (Fig. 1). Evianthus stewartit Bor, nom. nud. 1 diff i l pec lunculis Proxime affinis S. griffithiz Munro ex Boiss pilosis, callo cum pilis usque ad 6.8 mm longis, aristis usque ad 3mm u loneis. antherisque usque ad 2 mm longis. Perennial; culms up to 3 m high. Leaf blades up to 50 cm long, tough, long attenuate at the tip. Panicle ca. 20 cm long; peduncle villous; racemes SIDA 16(1): 33 — 36. 1994 34 SIDAL6(1) 1994 . yes HES as eZ if by AY / A Paar oie {Ss vy e AE GEFEN Sie =KEs : EZ es Se Se y y a7 “ SSS: a4 HK \} Gs) AW “fh vA IN \ INN i, RWIS NY NA NX) we > wee Ay SSS Sy * — Se SN SSG SS) as SEES yee gee £7 iy oF, LEP Z Zy ZG ia ft A bY 4, ‘y me FIG. 1: Saccharum stewartii Rajeshwari, Rao & Garg. A. Portion of Inflorescence, B,, B> upper and lower glumes; C. lower lemma; D. upper lemma with awn; E. palea; E&G. lower glume and lower lemma of the pedicelled spikelet. 6-7.5 cm long; internodes hairy; hairs up to 6 mm long. Sessile spikelets up to 8 mm long; callus bearded with silky hairs; hairs up to 6.8 mm long, usually shorter than the spikelets and not concealing them; glumes equal, ovate-lanceolate, keeled, 1—3 nerved, scabrid on the keels, hairy on the back or glabrous; lower lemma 3 mm long, 3-nerved, ciliate at the margins, hyaline; lower palea membranous, as long as the lower lemma, ciliate at the RAJESHWARI, RAO AND GarG, Saccharum 45 margin; upper palea membranous, hyaline; upper lemma 2 mm long, entire, ciliate at the margins, awned; awn 3 mm long, straight; anthers 2 mm long. Pedicelled spikelets male only or bisexual; lodicules 2, cuneate or ovate- oblong, faintly ciliate; caryopsis not observed. Type: INDIA. Kasumir: Rupal to Gurikor, 2500 m, 23 Aug 1939, R.R. Stewart & LD. Stewart 18893 (HoLorype: DD). Saccharum stewartii closely resembles S. griffithii Munro ex Boiss in having awned spikelets in short, articulate racemes borne on long panicles, acute or acuminate, membranous glumes, and lower florets all neuter. The two species can be separated as follows: Panicle more than 20 cm; peduncle glabrous; racemes up to 2.5 cm long; callus hairs yellow or cream colored, 3-4 mm long; awn of upper lemma 4 mm long S. griffithii Panicle up to 20 cm; peduncle hairy; racemes 6-7 cm long; callus hairs white, mm long; awn of the upper lemma less than 3 mm long ................4- S. stewartit Palynology In addition to the above morphological differences the new species also reveals certain palynological differences with that of S. griffithiiz. The morphology of the pollen grains in the two species is discussed below. Saccharum stewartii Rajeshwari, R.R. Rao & Garg Grains 29.3 pm (29-29.6 pm) in diam, spheroidal with circular outline; typically monoporate, pores operculate, annulate. Annulus conspicuous. Exine surface finely granulate, granules of low relief, grouped in close proximity, sparse at annulus; exine 1 pm thick, thicker at the annulus region (2 225-qim) Chigs.2 é& 3) Saccharum griffithit Munro ex Boiss Grains 16 pm (15-18 pm) in diam, spheroidal with circular outline, typically monoporate, pores operculate, annulate. Annulus faint. Exine surface psilate with few punctae distantly scattered having a negative reticulum. Exine | pm thick, slightly varying in thickness near the aperture (Figs. 4 & 5) ACKNOWLEDGMENTS The authors are thankful to the Director of the National Botanical Research Institute, Lucknow for facilities and encouragement and to Dr. S.K. Jain, Emeritus Scientist, National Botanical Research Institute for going through the manuscript. One of us (SR) is thankful to the Director of the Botanical Survey of India, Calcutta for the award of a Senior Research Fellowship during the tenure of which this work was carried out. 36 SIDA1G6(1) 1994 wD FIGS. 2—5. Scanning Electron Micrographs of pollen of Saccharum stewartii Rajeshwari, R.R. Rao & Garg (Figs. 2 & 3) and Saccharum griffithii Munro ex Boiss (Figs. 4 & 5). Scale lines = 3 pm REFERENCE Bor, N.L. 1960. The grasses of Burma, Ceylon, India and Pakistan. Pergamon Press, London. POPULATION AND SITE CHARACTERISTICS OF A RECENTLY DISCOVERED DISJUNCT POPULATION OF CROTON ALABAMENSIS (EUPHORBIACEAE) GREGORY H. APLET The Wilderness Soctety 900 Seventeenth Street NW Washington, DC 20000, U.S.A. RICHARD D. LAVEN and MAURYA B. FALKNER Department of Forest Sciences Colorado State University Fort Collins, CO 80523, U.S.A. ROBERT B. SHAW! Center for Ecological Management of Military Lands orado State University Fort Collins, CO 80523, U.S.A. ABSTRACT A disjunct population of Croton alabamensis E. A. Smith ex Chapman, a Category 2 Candidate species being considered for addition to the List of Endangered and Threatened Plants, was discovered over 1000 km from any previously known population. This paper describes the extent and status of this disjunct population. The Croton population is situated along canyon bottoms and appears healthy and self-sustaining. Croton occurrence exhibits no association with overstory gaps, disturbance, or particular fluvial geomorphic features. The species appears to be restricted to canyon bottoms characterized only by mesic conditions provided by the presence of overstory cover and deep soils. Differences between the ecology of this newly discovered population and those of the heretofore considered endemic Alabama populations are discussed. RESUMEN Una poblacién alopatrica de Croton alabamensis E. A. Smith ex Chapman, Categoria 2 de Especie Candidata, que esta siendo considerada para ser agregada a la Lista de Plantas en Peligro de Extincién, fue descubier’a a una distancia de 1000 km de la poblacién previamente conocida. En este trab ajo se describe la extensién y estado de la poblacién alopatrica. La poblacién de Croton esta situada a lo largo del fondo de un cafion aparentando poder sobrevivir y crecer en buenas condiciones. La ocurrencia de Croton no parece tener asociacién con espacios desnudos, perturbaciones o caracterfsticas fluviales geomérficas 'Corresponding author SIDA 16(1): 37 — 55. 1994 38 SIDA16(1) 1994 particulares. La especie parece estar restringida al fondo del cafion caracterizado solamente por condiciones mesomérficas debidas a la presencia de cobertura y suelos profundos. Las diferencias entre las caracteristicas de esta poblacién recientemente descubierta y las de poblacién heterofftica en Alabama, antes considerada endémica, son discutidas. INTRODUCTION Croton alabamensis E. A. Smith ex Chapman (Euphorbiaceae) has been de- scribed as “one of the rarest shrubs in the United States” (Farmer and Thomas 1969). It is a short-lived (<20 y), multi-stemmed, monoecious shrub <3 m tall found primarily on limestone and shale outcrops along the Warrior and Cahaba Rivers, Tuscaloosa and Bibb counties, Alabama. It was collected once in 1899 in Tullahoma, Coffee County, Tennessee, but has not been reported from there subsequently. Croton alabamensis has long been thought to grow only in isolated populations within these two neighboring counties. Due to its restricted range, it is being considered for addition to the List of Endangered and Threatened Plants under the Endangered Species Act of 1973, as amended. It currently is designated as a Category 2 Candidate, which means that ‘there is some evidence for vulnerability, but ... there are not enough data to support listing proposals at this time” (USDI 1991). In early 1990, Croton alabamensis was discovered over 1000 km from any previously known population on the U.S. Army’s Fort Hood, Texas. The population was discovered by John Cornelius, a wildlife biologist with the Fort Hood Resource Management Department, during an excursion to view bird habitat in the Owl Creek Mountains. Cornelius showed the population several weeks later to Carol Beardmore of the U.S. Fish and Wildlife Service and Rex Wahl of Texas Parks and Wildlife. Ginzbarg (1992) mistakenly credited the discovery to Beardmore and Wahl. Not long afterwards, a second and a third population were discovered in Travis County, to the south. The species identity was determined by Steve Ginzbarg, a graduate student in botany at the University of Texas, Austin. Ginzbarg (1992) subsequently determined that the Texas plants are sufficiently distinct to merit varietal status. He consequently described them as Croton alabamensis E. A. Smith ex Chapman var. texensis Ginzbarg (Ginzbarg 1992). Key characteristics of the Texas variety are represented in Figures | and 2. In June 1991, we undertook to describe the extent and status of the population at Port Hood as part of the Department of Defense’s Legacy Resource Management Program. The Legacy Program was established to enhance conservation of biologically and culturally significant resources on military lands. Our study is aimed at gathering the requisite information to effect proper stewardship of this species at Ft. Hood. We therefore restricted our analysis to only one of the three Texas populations (i.e. the Fe. Hood population). We recommend that similar studies be undertaken in both APLET, LAVEN, FALKNER AND SHAW, Croton alabamensis 39 FIG. 1. Croton alabamensis var. texensis: (a) stem with flowers clustered at the apex of primary and secondary branches (flowers open before leaves are fully expanded), (b) stem showing fully developed leaves. 40) SIDA16(1) 1994 FIG. 2. Croton alabamensis vat. t s:(a)androg infl b) malefl bud, (c) male flower, (d) female flower with rudimentary stamens, (e) 3-celled capsule, (f) seed, (g) stellate trichome common on leaves, buds, and ovary, (h) trichome found on disk lobes of male and female flowers. APLET, LAVEN, FALKNER AND SHAW, Croton alabamensis 4] Texas and Alabama to assess the status of all populations of this rare and scientifically important species. STUDY AREA This study was conducted in several small watersheds of the Owl Creek Mountains, Bell and Coryell counties, in the Hill Country of central Texas (31°N, 97°W). The Owl Creek Mountains (elevation ~300 m) are com- posed of Mesozoic limestone overlain by clayey and loamy soils (McCaleb 1985). The formation rises to its summit plateau ~65 m above the west-to- east flowing Owl Creek in less than 2 km. Intermittent tributary streams run northward to the creeksuch that virtually all canyon walls have east- and west-facing aspects. The climate is hot in the summer, and the winters are generally mild, with an occasional cold surge. The average daily temperature during the summer is 28°C, with an average daily maximum of 36°C. Winter daily tempera- tures average 9°C, with an average low of 3°C. Rainfall is distributed uni- formly throughout the year, witha slight peak in the spring and an average annual total of 825 mm (McCaleb 1985). Owl Creek flows to some extent throughout the year, but the tributary streams are dry most of the year. The vegetation of the area (see Appendix 1 for complete plant species list) is primarily Ashe juniper (Janiperus ashei Buchh.) woodland of the appropriate character to meet the habitat needs of two rare birds, the black-capped vireo (Vireo atricapilla) and the golden-cheeked warbler (Dendroica chrysoparia). The vireo primarily inhabits the shrubby balds of the summit plateau, but the warbler requires the bark of relatively old junipers for nesting materials and mature hardwoods for feeding (J. Cornelius pers. comm.). Prior to our study, Croton alabamensis var. texensis had been identified from two of the tributary canyons of Owl Creek. We searched five additional canyons and found only three more plants in the canyon between the original two and several plants along the stretch of Owl Creek between the two tributaries; populations of Croton appear to be largely restricted to the two original canyons. Our study involved two phases: a description of the structure and habitat of the two Croton populations (canyons 1 and 3) and an attempt to discern why Croton is all but absent from the canyon between (canyon 2). Subsequent to the completion of this study, another population consisting of 35 individuals was located in a canyon three km east of the study area. This population is not considered in our analysis. METHODS The bed of each of the three canyons was mapped using a tape and a hand- held compass. The slope of the creek was measured using a hand-held clinometer. Three high density stands of C. a/abamensis var. texensis were 42 SIDALG6(1) 1994 identified for intensive sampling along the creekbeds in Canyons 1 and 3. Ateach sample location, two transects were established perpendicular to the creekbed to facilitate sampling of both east- and west-facing aspects. In Canyon 2, which lacked Croton, similar pairs of transects were established at distances up the canyon comparable to those identified in Canyons | and 3. Slope topography, overstory and understory cover, Croton seedling den- sity, Croton adult population structure, and soil depth were determined along each transect. At each site, a tape was stretched upslope beyond the extent of the Crofon population. In some cases, this was as far as 60 m from the creekbed, but was generally less. Transects in Canyon 2 were 30-40 m long. Slope breaks and important topographic features were noted along the slope, and a cross-section map was prepared for each transect. Soil depths were measured at 10 m intervals and at important topographic features by probing the soil with a | m rod. long the tape, cover of overstory (>2 m tall) and understory (<2 m but >10 cm tall) vegetation was determined using the line-intercept method (Mueller-Dombois and Ellenberg 1974). The line also served as the center of a 2 m wide Croton seedling belt in which individuals <30 cm tall were tallied. Croton >30 cm tall were sampled using a modified nearest-neighbor method adapted from those described by Mueller-Dombois and Ellenberg (1974). Starting at the beginning of the line (middle of the creekbed), a 180° arc was searched upslope for an individual of Croton >30 cm tall. The distance and azimuth to the nearest plant were recorded as were the height of the tallest shoot and the diameter of each live shoot (to the nearest 0.5 cm). From that plant, the process was repeated until no plants could be found within 10 m of the last plant (Fig 3). In only one case, the process was suspended and moved back to the tape when the search led to a plant >10 m from the tape. Thus, all plants were sampled within 10 m of the tape. Consistent with other nearest-neighbor methods, the distance between plants was assumed to be related to the share of the total area allotted to an indi- vidual plant. From these data, population density could be calculated as the total number of plants sampled divided by the sum of areas allotted to indi- vidual plants. Additionally, using a simple trigonometric conversion, the den- sity represented by individual plants could be plotted against distance from the creek to assess changes in density related to environmental variables. Sizes of individual, multi-stemmed Croton plants were described using the statistic of equivalent diameter (D,,). Equivalent diameter is the diameter of a single stem that possesses the same cross-sectional area as the sum of all the stems borne by the plant. For example, as the total cross-sectional area of four shoots, each 2 cm in diameter (4 X 3.14 cm”) is equal to that of one shoot, 4 cm in diameter (12.56 cm), the D., for the four shoots is 4 cm. APLET, LAVEN, FALKNER AND SHAW, Croton alabamensis 43 Transect_ a Pan Origin ~~ Nt . 20 Distance (m) FIG.3. Illustration of the modified nearest neighbor method for sampling Croton irae and diameter and calculating density. Solid squares indicate plant locations within a hypothetical transect. Solid squares connected by line ee sampled plants. RESULTS Autecology The population structures of Croton alabamensis var. texensis in the two canyons are presented in Figure 4. In both canyons, Croton was well represented 1n the smaller diameter classes, but also occurred as large mature plants. This decreasing monotonic population structure was found in all transects in which Croton was relatively dense. Therefore, both canyons appeared to support healthy populations of adults, juveniles, and new recruits. Croton density was highly variable within Canyons 1 and 3. Each canyon contained places in which scarcely an individual was found. In other places, Croton formed dense thickets in which it dominated to the near exclusion of other understory species. The densest stands occurred in Canyon 1 where Croton density (adults and seedlings) in two transects exceeded 100 plants/ 100 m2, but three transects in Canyon 3 exceeded 50 plants/100 m? (Table 1). Overall, we estimate the number of plants occurring in the 1.5 km of creekbed in Canyons 1 and 3 to be ~20,000 individuals. In general, Croton was found only in the canyon bottoms. No individuals were found >60 m from the creek, and of the 12 transects containing Croton, only four contained plants >40 m from the creek. Croton density fluctuated, but Croton did not decline proportionally with distance from the creek. Canopy gaps along the line intercept showed no correlation with Croton occurrence. Croton were found in openings and in the deepest shade. The can- yons of the Owl Creek Mountains have been subjected to harvesting of juni- per trees for fence posts, and some of the gaps may have been anthropogenic. dX peN SIDA16(1) 1994 35 Canyon 1 30+ | y P| Wlldnenee. Don oon 0 2 4 6 8 10 12 14 >15 Canyon 3 Plants per 100 m2 Oo no ooo oll 8 10 12 14 >15 Deq Class (mm) FIG. 4. Croton population size structures, indicating number of plants by equivalent diameter (D,,) class. See text for explanation. APLET, LAVEN, FALKNER AND SHAW, Croton alabamensis AS TABLE 1. Characteristics of the 18 transects. VEGETATION COVE CROTON Slope Watershed sity Canyon Tran >2m(%) <2m (%) (%) Aspect Size (ha) (#/100m?) 1 l 77 21 16.7 E 1400 16.2 232 ] 2 85 37 30.0 W 1400 46.8 61.2 l 3 69 18 16.7 E 925 5.1 28.2 l 4 50 39 10.0 W 925 13.1 12.6 l 5 re) 50 Reyes) W 800 98.6 91.0 | 6 94 38 33.3 E 800 13.0 18.2 2 1 74 16 53.3 Ww 425 0 0 2 2 25 16.7 E 425 0) 0 2 3 90 37 53.3 E 375 0 0 2 4 2 39 13:3 Ww 375 0 0 2 5 75 42 53:3 E B25 0) 0 2 6 66 37 30.0 Ww 325 0) 0 3 1] 89 23 25.0 E 600 2 13.4 5 2 84 49 Ww 600 2.0 4.4 3 3 96 54 60.0 E 350 9.7 55.8 3 4 99 30 21.7 Ww 350 254 19.9 3 5 99 47 30.0 Ww 325 23.4 28.6 3 6 77 39 60.0 E 325 0 21 In some cases, the loggers incidentally cut adult Croton along roads and skid trails, but the plants sprouted and were growing we Topographic features also showed no direct correla with Croton density. Croton occurred on bank slopes, terraces, and on toeslopes of each canyon. Soil depth, however, explained much of the variability in Croton distribution. A comparison of soil depth with the presence of Croton along the transect indicated a significant association of adult plants with deep soil (p<0.05; Kruskal-Wallis test). Our inability to measure depths in excess of 1 m means that the measured mean soil depth in the presence of Croton (0.80 m;n=64) was probably considerably less than the true mean. In contrast, soil measurements in the absence of Croton averaged only 0.42 m (n=33) in Canyons 1| and 3 Comparison a Croton density with overstory species composition sug- gests an association with mesic sites, as would be expected from the observed restriction of Croton to the canyon bottoms. Figure 5 shows the relationship of adult Croton density to the combined cover of Fraxinus texensis (Gray) Sarg. and Quercus mublenbergii Engelm. and to the combined cover of Juniperus ashei and Quercus texana Buckl. within 40 m of the creek. These four species dominated the overstory stratum in these canyons; only the woody vine Vitis ble cover over the study area. Of the mustangensis Buckl. contributed c AG SIDAILG6(1) 1994 2.0 02 0.4 0.6 0.8 1.0 Juniperus and Q. texana Cover Croton Density (plants/m? ) 0.0 Pa , @° | | | | ! | ! 0.0 0.1 0.2 0.3 0.4 05 Fraxinus and Q. muhlenbergii Cover FIG. 5. Relationship of adult Croton density to combined cover of Fraxinus texensis and Quercus mublenbergit and the combined cover of Juniperus ashet and Quercus texana. APLET, LAVEN, FALKNER AND SHAW, Croton alabamensis 47 three species of oak encountered during sampling, Q. muhlenbergii is the most mesic (Miller and Lamb 1985), and Fraxinus is found primarily in canyons (Correll and Johnston 1970). Juniperus is not so restricted, and Q. texana is described as occurring on “dry limestone hills and ridges” (Miller and Lamb 1985). In the study area, Croton was more abundant on mesic transects in which Fraxinus and Q. mublenbergii were common and less abundant on dry transects in which J. ashez and Q. texana were most common. Synecology The four tree species listed above dominated the vegetation of all three canyons. Figure 6 shows the contribution of these species to the overstory cover within 40 m of the creek (40 m was used as the limit of Croton habitat). The dominant species in all three canyons was clearly J. asher, which con- tributed about one-third of the cover. The other three species collectively con- tributed another third. These four were the only species to contribute > 10% relative cover in one of the three canyons, although some species, such as V. mustangensis, Ulmus crassifolia, Nutt. Juglans major (Torr.) Heller., and Ce/tis laevigata Willd. were abundant locally (Appendix 2). All overstory species contributing <10% relative cover are included in Figure 6 as “minor species.” In an attempt to discern differences in habitat characters between the canyons supporting Crofton populations and Canyon 2, we subjected transect cover data to principal components analysis (PCA) (SAS/STAT User’s Guide 1988). If the transects in Canyon 2 differed in overstory composition from the other two canyons, those sites would have segregated as a distinct habitat type. Canyon 2 showed no difference in habitat from the other two as expressed in overstory composition (Fig. 7). Likewise, understory cover displayed no pattern related to Croton occur- rence. With the exception of Croton itself, all three canyons supported similar understory communities (Appendix 2). In the two canyons in which it occurred, understory Croton cover averaged 10.4%. (Again, placement of the transects was biased by the presence of dense populations.) Other relatively abundant understory species included Fraxinus species (5.5%), V. mustangensis (3.9%), Rhamnus caroliniana Walt. (3.4%), J. ashei (3.2%), Rhus toxicodendron L. (2.6%), various grasses (2.4%), Q. texana (2.1%), and Ilex decidua Walt. (2.0%). he three canyons differed very little in both total overstory and under- story cover. Overstory cover ranged from 50% in one transect in Canyon | to 99% in two transects in Canyon 3, but no significant difference occurred among canyons (Table 1). Similarly, understory cover ranged from 16% in one Canyon 2 transect to 54% in a Canyon 3 transect, but canyon means were similar. 48 SIDA16(1) 1994 Forest Cover Canyon 1 Canyon 2 Canyon 3 Juniperus ashei MME Quercus texana MMB Fraxinus texensis MM Quercus muhlenbergii {| Minor species (Relative cover <10%) FIG. 6. Relative cover of the four dominant eee Species. All overstory species contributing <10% relative cover are included as “minor species. Steepness of canyon walls within 40 m of the creek varied widely from site to site (10% to 60%), but means from the three canyons were similar (Table 1). Canyons | and 3 were the largest watersheds, but watershed areas above Croton populations in Canyon 3 were no larger than those present in Canyon 2. Likewise, soils of a depth that favored Croton in Canyons | and 3 were abundant in Canyon 2. The only morphological feature distinguishing Canyon 2 from Canyons | and 3 was the steepness of the creekbed. Streambed gradient in Canyon 2 (3.81%) was significantly steeper than in Canyons 1 and 3 (1.92% and 2.11%) (p<0.05; Kruskall-Wallis test). DISCUSSION Croton alabamensis var. texensis in the Owl Creek Mountains grows in healthy, self-sustaining populations along the bottoms of tributary canyons and the connecting section of Owl Creek. Croton occurrence exhibits no association with overstory gaps, disturbance, or particular fluvial geomor- phic features. It appears to be restricted to canyon bottoms characterized only by mesic conditions provided by the presence of overstory cover and deep soils. There is some suggestion that high cover of F. texensis and Q. muhblenbergii indicates a good site for Croton. APLET, LAVEN, FALKNER AND SHAW, Croton alabamensis 49 Axis 2 G PO w Axis 1 FIG. 7. Results of principal components analysis of cover data for all three canyons. Symbols identify locations of plots by canyon. Lack of segregation indicates that Canyon 2 did not differ in overstory composition from Canyons | and 3. The distribution of Croton does not appear to be limited by the availability of sites. Canyon 2, which contained only 3 plants, supported a similar overstory and understory to Canyons | and 3. Canyon 2 also contained deep soils and a similar geomorphology to the other two canyons. Likewise, watershed size did not appear to explain Croton presence. The only feature that differed among canyons was stream gradient. Hupp (1988) discusses several ways in which stream channel geomor- phology affects vegetation, including fluvial landforms, flood frequency and duration, and stream gradient. It would be consistent with other observa- tions if the steeper gradient of Canyon 2 afforded a less mesic environment that tecouraeee © ae pale However, fluvial processes in these arid intermittent chan iffe from those of the eastern floodplain systems that Hupp discustee: and channel gradient is offered only specula- tively as a factor explaining Croton distribution. The larger question of how C. alabamensis vat. texensis came to occupy disjunct sites in Alabama and Texas is beyond the scope of this study. Although Ginzbarg (1992) discusses the possibility that the present distri- bution may be an ice age relict, he suggests that the Texas populations may be more easily explained as the result of relatively recent introduction by 50 SIDAI16(1) 1994 long distance seed dispersal by birds. Regardless, it is likely that the disjunct distribution resulted from a prolonged process of migration, colonization, and extinction of many sites. The absence of Croton from Canyon 2 may reflect not the lack of available habitat, but the vagaries of colonization and extinction. Perhaps a population once thrived there and has gone extinct, or seeds have, by chance, not recently reached this canyon. More likely, the canyons of the Owl Creek Mountains and similar features of the Edwards Plateau support metapopulations of Croton alabamensis var. texensis, each subpopulation establishing, thriving, and going extinct only to be replaced somewhere else. Only through repeated observations over long time periods will we understand the process. Regardless of the limitations on its range within the study area, the behavior of C. a/abamensis in the Owl Creek Mountains is dramatically different from its behavior in Alabama. Farmer (1962) describes Croton as occurring on shallow soils and rock outcrops at mid-slope positions in two counties in Alabama. He describes the habitat as “shallow soil ... on moderately- to steeply-sloping terrain; high temperature of soil and air during summer; intense drought; and freedom from fire.” Soil is described as “usually only a few cm thick and seldom more than 1 m thick.” In our study site, Crofon occurs on deep soils on toeslopes and fluvial deposits of canyon bottoms. The groves in Alabama are “marked by shrub dominance, few or no large trees, and a relative absence of herbs” (Farmer 1962). Occasional plants are found under the forest canopy surrounding the outcrops, but these are considered only “extensions of nearby thickets”. This contrasts sharply with Croton behavior in Texas. In our study sites, Croton occurs as an understory shrub in the company of many large trees and a healthy herb layer. Farmer considers adaptation to extreme drought to be an important factor in the ecology and distribution of C. alabamensis in Alabama (see above). In Texas, however, this species behaves as a drought avoider by remaining in mesic canyon bottoms. Perhaps these differences result only from perceptions relative to annual climate and surrounding vegetation. The 3—5 inches (7—12 cm) of monthly precipitation that Farmer (1962) reports for the “dry” season in Alabama is considerably greater that the 2—3 inches (5—7 cm) recorded for summer months in central Texas (McCaleb 1985). In Texas, where upland vegetation is necessarily drought adapted, a species requiring 7—12 cm of precipitation naturally would be restricted to the most mesic sites. nother interesting difference between Croton behavior in Alabama and Texas regards vegetative reproduction. Farmer (1962) discounts asexual reproduction in Croton by stating that “there are no rhizomes or adventitious rootings that result in plant reproduction.” In Texas, however, we observed APLET, LAVEN, FALKNER AND SHAW, Croton alabamensis 51 numerous plants that had produced “new” upright shoots through the nodal rooting (layering) of prostrate branches. We did not investigate the degree of connectivity between the layered offspring and the parent plants, but layering appears to be a potential mechanism of asexual reproduction in C. alabamensis var. texensis. CONCLUSIONS The disparate behaviors exhibited by this species over its disjunct range underscore the necessity for site-specific studies prior to making manage- ment recommendations. The ecology of Croton in Alabama would suggest a conservation strategy that might well be unsuccessful in Texas. The Ala- bama ecology implies preservation of, and population augmentation into, dry, open, limestone outcrops. This strategy, if adopted in Texas, likely would fail misera We nos very little of the disturbance ecology of C. alabamensis var. texensis in central Texas. The species apparently tolerates some degree of physical disturbanceas plants injured or cut off during the pole cutting operation appear to have recovered well. We know nothing, however, of the species’ response to fire or soil disturbance. In Alabama, fire is believed to be lethal to Croton (Farmer 1962). We saw nothing to indicate a dependence on disturbance for establishment. Conservation efforts, therefore, should focus on maintenance of the undisturbed nature of the mesic forests of the area. Construction of new roads for military training and logging should be discouraged, and old roads should be allowed to fall into disuse. The importance of this plant community has already been recognized for a rare bird; it now appears to be critical to the survival of a rare plant as well. ACKNOWLEDGMENTS e thank Dennis Herbert, Billie Ray Jones and John Cornelius, Division of Wildlife, Fort Hood, Texas for their assistance and encouragement during the study. A special thanks to Keith Schulz for field assistance and to Tracy Wager for the line drawings. Funding was provided by the USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Department of Defense Legacy Resource Management Program, and the Natural and Cultural Resources Division, Engineering and Housing Support Center. APPENDIX lL SIDAI16(1) 1994 Species encountered during line-intercept sampling. Taxonomy follows Correll, D.S. and M.C. Johnston. 1970. Manual of the vascular plants of Texas. Texas Research sae fae Texas. ACRONYM SPECIES FAMILY GROWTH FORM! ACLI Acalypha lindbetmert Euphorbiaceae H AMAR Ampelopsis arborea itaceae Vv AQHE Unidentified aquatic emergent H ARSI Argythamnia simulans Euphorbiaceae H BULA Bumelia lanuginosa Sapotaceae T CADR Calylophus drummondianus Onagraceae H CAREX Carex s Cyperaceae G CEHE Ceanothus husbonwe Rhamnaceae T CELA reltis laevigata Ulmaceae a5 CEOC es occidentalis Rubiaceae T CERE Celtis reticulata Ulmaceae T CLTE Clematis texensis Ranunculaceae H COD Cornus drummondit Cornaceae ay COER Commelina erecta Commelinaceae H CRAL Croton alabamen Euphorbiaceae T CYBA Cynanchum ee Asclepiadaceae H var. fexemsis CYU Cynanchum unifarinum Asclepiadaceae H DITE Diospyros texana Ebenaceae T FRTE FPraxinus texensis Oleaceae T GARE Galactia regularis Fabaceae H GRASS Various grasses” Poaceae G HESP Helianthus sp. Asteraceae H ILDE Llex decidua Ss T INMI Indigofera miniata Fabac H JUAS Juniperus asher Cupressaceae a JUMA Juglans mayo Jug aca T LEVI Lespedeza violacea Fabaceae H LOJA Lonicera Japonica Caprifoliaceae H MEAL Melilotus albus Fabaceae H IRE Mitchella repens Rubiaceae H MORU Morus ruby Moraceae T OPPH Opuntia oe Cactac H PALU pains area Vv PAPE Par es aes Urticacea H PAQU Parthenocissus winguefolia Vitacea Vv PLOC Platanus occidentalis Platanacea 1 PODO Polanisia dodecandra Capparidaceae H PRMU Prunus munsoniana Rosaceae T PRSE Prunus serotina Rosaceae T QUMU Quercus mublenbergri Fagaceae T QUSI Quercus Sinuata Pagaceae T var. breviloba QUTE Quercus texana Fagaceae T RHAR Rhus aromatica Anacardiaceae T RHCA Rhamnus cavoliniana Rhamnaceae T RHTO Rhus toxicodendron Anacardiaceae T/V RHVI Rhus virens Anacardiaceae T APLET, LAVEN, FALKNER AND SHAW, Croton alabamensis APPENDIX 1 continued ACRONYM SPECIES FAMILY GROWTH FORM! RUBI Rubus bifrons Rosaceae T/V SARO Salvia roemeriana Lamiaceae H SMBO Smilax bona-nox Smilacaceae Vv SOAF Sophora affinis abaces T SOSE Sophora secundiflora Fabaceae T STPL tyrax platanifolia Styracaceae T ULCR Ulmus crassifolia Ulmaceae T UNHE Various unidentified herbs H UNSP Ungnadia specto Sapindaceae T VEVI Verbesina virginica Asteraceae H VIMU Vitis mustangensis Vitaceae Vv YURU Yucca rupicola Liliaceae H 'G=Grass or ene planets; H=Herbaceous plants; T=Trees or shr ote easses encou ichanth 1M dA ubs; V= Vines red during sampling include coi te aaa Bothriochloa sp., Bromus pubescens cuminatum vat. implicatum, and Glycerta striata APPENDIX 2 Absolute percent cover and frequency of occurrence (no. of transects) of all species encountered during line-intercept sampling. Percent cover is reported for overstory (> 2m) and ee (< 2m) in each canyon. Species are listed in descending order of dominance over the study ar LCOV, whet ALLCOV was calculated as (overstory cover + understory cover)/3. Plot oe in canyons and 3 was biased by the presence of Croton (see text) OVERSTORY COVER UNDERSTORY COVER ACRONYM CYN1 CYN2 CYN3 FREQ CYN1 CYN2 CYN3_ FREQ ALLCOV JUAS 51.60 49.67 49.88 18 1.54 Neon 4.67 14 53.58 FRTE 16.50 13.18 26.21 L 7 3.88 4.93 ted 18 24.14 QUTE 15.59 29.09 20.11 17 3.69 2.23 0.42 11 23.71 VIMU 9.87 11.68 14.35 13 3.21 3.26 3.29. 15 15.89 QUMU 13.39 6.33 8.65 10 0.69 0.18 0.21 6 9.82 RHCA 4.58 2.92 4.87 10 1.21 LD: 7.00 13 Tow) ILDE 5.60 5.62 4.43 13 Ld 3.40 1.17 1 19 CRAL 0.22 1 11.24 9.46 1] 6.97 ULCR 2.02 4.14 10.67 7 0.75 0.63 3 6.07 JUMA 2.87 8.74 6 0.58 0.04 2 4.08 CELA 9.11 3 3.04 RHTO 2.28 2.90 7 15 2.58 GRASS 2.28 2.96 1.90 13 2.38 PRSE 1.95 2.61 6 122 1.10 5 2.29 QUSI Sided 0.33 3 0.83 1.58 0.04 8 1.99 SMBO 0.22 0.08 2.09 5 0.17 1.33 0.68 10 1.52 STPL 1.94 2 0.08 1.05 0.04 a, 1.04 CAREX 0.04 2.13: 0.90 1.02 BULA 0.14 2.19 B) 0.04 0.14 0.11 3 0.88 MORU 0.06 2:39 e) 0.81 RHVI 0.58 0.28 2 0.17 0.33 0.72 3 0.69 PLOC 2.03 2 0.68 54 SIDA16(1) 1994 APPENDIX 2 continued OVERSTORY COVER UNDERSTORY COVER ACRONYM CYN1 CYN2 CYN3 FREQ CYN1 CYN2 CYN3_ FREQ ALLCOV SOSE 1.21 l 0.04 0.21 0.47 4 0.64 PAQU 0.63 1 0.25 0.63 0.11 8 0.54 COER 1.36 3 0.45 HESP 1.15 3 0.38 EVI 0.45 0.39 4 0.28 YURU 0.33 0.50 2 0.28 CEOC 83 ] 0.28 AMAR 0.67 l 0.08 0.04 2 0.26 LEVI 0.43 0.25 2 0.23 PODO 0.29 0.21 0.13 6 0.21 NSP 0.57 l 0.19 DITE 0.33 1 0.23 2 0.19 RHAR 0.29 0.17 4 0.15 CERE 0.46 1 0.15 SARO 0.30 0.14 5) 0.15 INMI 0.42 l 0.14 AQHE 0.42 2 0.14 UNHE 0.08 0.12 0.17 5) 0.13 MEAL 0.29 l 0.10 SOAF 0.25 ] 0.08 PRMU 0.23 2 0.08 CODR 0.21 l 0.07 OPPH 0.08 0.10 2 0.06 PALU 0.17 l 0.06 CEHE O.11 1 0.04 GARE 0.08 l 0.03 LOJA 0.08 l 0.03 RUBI 0.08 l 0.03 MIRE 0.08 ] 0.03 CLTE 0.08 ] 0.03 CYBA 0.03 | 0.04 l 0.02 ACLI 0.05 l 0.02 ARSI 0.05 1 0.02 CYUN 0.05 l 0.02 PAPE 0.04 l 0.01 CADR 0.04 I 0.01 REFERENCES CorreLt, D.S. and M.C. JoHNsTon. 1970. Manual of the vascular plants of Texas. Texas Res ne oo Renner, TX. Farmer, J.A 2. An ecological life history of Croton alabamensis E.A. Smith ex Chapman. eee im D. Dissertation, University of Alabama. FARMER, J.A. an ae . Rca 1969. Disjunction and endemism in Croton alabamensis. lal 's(Euphorbi )fi Texas GINZBARG, S. 1992. i. new disjunct variety of Crot Sida 15(1):41-52. APLET, LAVEN, FALKNER AND SHAW, Croton alabamensis 55 oe C.R. 1988. Plant ecological aspects of flood geomorphology and paleoflood history. p.335—-356, in Flood See eas (Baker, V.R., R.C. Kochel and P.C. Patten, eds.). ae Wiley and Sons, Inc. New Miter, H.A. and - a Lams. 1985. te of North America. Naturegraph Publishers, Inc. Happy Camp, CA. McCates, N.L. on Soil survey of Coryell County, Texas. United States Department of Agriculture. Soil Conservation Service. U.S. Government Printing Office. MuELLER-Dompsois, D. and H. aati 1974. Aims and methods of vegetation ecology. ohn Wiley and Sons, Inc. New Sas/Stat User’s Guipe. Release 6.03 een 1988. SAS Institute, Inc. Cary, NC. 27512- 00. Uspr. 1991. Endangered and threatened wildlife and plants: review of plant taxa for listing as endangered or threatened species; notice of review. 50 CFR Part 17. Federal Register. 56 SIDA16(1) 1994 BOOK NOTICES SMITH, NiGEL J.H., J.T. Wittiams, DoNnaLp L. PLUCKNETT, and JENNIFER P. Tatsor. 1992. Tropical Forests and their Crops. (ISBN 0-8014- 2771-1, pbk). Comstock Publishing Associates, a Division of Cornell University Press, Ithaca, NY. $27.95 (pbk), $69.95 (hbk). 568 pp. “Many cultivated plants important for food and income in developing countries arose in tropical forests.” A number of these plants are the subjects of this book, grouped into “Beverage and confectionery crops” (e.g., coffee, cacao); “Major fruits” (e.g., mango, citrus, alae ‘Regional fruits” (e.g., durian, rambutan); “Rubber, oils, and resins” (e.g., oil palm, tropical pin nes); “Daily bread” (e.g., bananas, are “Fuelwood, fodder, ane oe grasses” (leucaena and bamboo); “Spices and natural food colorants” (e.g., e, cinnamon, annatto): and “Nuts” (SHEN, Brazil nut, macadamia). Among the oe eon a history, evolution spread, genetic resources, gene pools, gene banks, empl collection, breeding ghallenoes. conservation—not tall of oe for each ». Additional chapters discuss the search for new plant | : the need for a strategies of conservation. Most entries in the eee listof“References” a are post-1950 (largely from the 1980s). The book is illustrated with black-and-white photos and a few line hn W. Thi drawings and is well indexed.—John W. Thieret National ResEARCH CouNcIL. 1993. Vetiver Grass. A Thin Green Line against Erosion. (ISBN 0-309-04269-0, pbk). National Academy Press, Washington, DC. No price given. 171 pp. — Vetiver, Vetiveria zizanioides, has long been cultivated as the source of essential oil of vetiver used in perfumes and soaps. That it is also a pias: of much promise for controlling soil erosion in warm areas is eae increasingly realized. Hedges of this deep-rooted, robust, densely cespitose plant create impressive “botanical dams” strong enough to retard runoff. arma its use as a hedge plant ts not new, it has recently come much into the limelight. This book offers an excellent summary on vetiver: biology, past use, current research, and future prospects. Appendix A highlights “some environmental horrors and points to the role that vetiver might play in alleviating them”; Appendix B, “Other potential vetivers,” gives data on other grasses and some shrubs and trees that might be useful in erosion control; Appendix C has selected readings; and Appendix D lists worldwide research contacts. This book and the World Bank’s practical handbook Vetiver grass (noted in Sida 15: 558. 1993) are a most valuable pair for anyone wanting information on this grass.—John W. Teeves — a NOTES ON ERIOCHLOA WEBERBAUERI (POACEAE: PANICEAE) ROBERT D. WEBSTER United States Department of Agriculture Systematic Botany & Mycology Laboratory Bldg. 003, Rm. 235, BARC-West Beltsville, MD 20705, USA PAUL M. PETERSON Department of Botany National Museum of Natural History Smithsonian Institution Washington, DC 20560, USA ROBERT B. SHAW Department of Range Science Colorado State University Fort Collins, CO 80526, USA ABSTRACT Recent collections from Batiadar 4 have uncovered the EDEOIgHce of a species, Eriochloa weberbauert Mez, for which li isavailable. Previously reported only from Peru, this succics is characterized by features unique within the genus. A complete morphological description is presented and based on studies of all available specimens. Relationships with other species are discussed and based on the presence of shared taxonomically important characters. Diagnostic features of E. weberbaueri include a muticous and smooth upper floret, perennial habit, paired spikelets, spikelets 3.1—-3.9 mm long, and the absence of a first glu RESUMEN Recolecciones recientes en Ecuador han descubierto la presencia de una especie, Eriochloa weberbaueri Mez, de la que ae oe muy poca informacion taxonémica comparativa. Citada previamente solo de Pert, esta expecie presenta unas caracteristicas inicas dentro del sf género. Se presenta una cee onsen completa en base a estudios de todos los especimenes disponibles. Se discuten las relaciones con otras especies en base a la canes de caracteres compartidos a IE SONI Los caracteres diagn de E. weberbaueri incluyen una flor superior miitica y lisa, habito ales saa pareadas, espiguillas 3.1—-3.9 mm de largas y la ausencia de la primera glum INTRODUCTION The Paniceae is a tribe of grasses characterized by a paniculate inflores- cence, spikelets with two florets, the lower floret sterile, the upper floret SIDA 16(1): 57 — 61. 1994 58 SIDAI6(1) 1994 fertile and coarser in texture than the lower floret. Erioch/oa Kunth, a mem- ber of this tribe, consists of about 30 species and occurs throughout the tropical, subtropical, and warm temperate regions of the world. Presence of a cup-like callus at the spikelet base differentiates Eriochloa from the other 105 genera of the Paniceae. Additional significant characters used to differen- tiate Erioch/oa within the tribe include an indurate upper floret, disarticula- tion at the spikelet base, inflorescence with racemose primary branches, and the absence of a bristle (Webster & Valdez 1988, Webster et al. 1989). Significant characters used to differentiate the species of Erzochloa include spikelet shape, presence of an awn on the upper floret, spikelet length, presence of hairs in the inflorescence, spikelet arrangement, longevity, and the relative length of the florets (Webster 1987, Shaw & Webster 1987). Recent collections by Peterson and others in Ecuador have revealed the presence of specimens which do not match the morphological characteristics of taxa commonly associated with that area. The characteristics of this taxon were recorded in DELTA format (Dallwitz 1980, Dallwitz et al. 1993) and compared against all other recognized species of Eriochloa. This analysis indicated these specimens could not be differentiated from a poorly know Peruvian species, E. weberbauert Mez. Hitchcock (1927) states that this species was known only from the type specimen. Tovar (1993) list only two specimens, the type and Anderson 740, which was collected in the same area of Peru. A survey of collections at the U.S. National Herbarium revealed the presence of three specimens (A nderson 740, Sanchez Vega 1031, Camp E-23406) which possess the diagnostic characteristics of E. weberbaueri. In addition, G. Davidse has identified two MO collections (Sanchez et al. 3567 & Smith 6190) of the same species. Diagnostic features of Eriochloa weberbaueri include a muticous upper floret, a smooth upper floret, paired spikelets on the primary inflorescence branches, perennial habit, spikelet length (3.1—-3.9 mm long), first glume absent, and the absence of elongate rhizomes. Eriochloa distachya, E. michauxii, E. polystacyha, E. sericea, and E. weberbaueri are the only species with a muticous upper floret and perennial habit. Of this subgroup, E. distachya and E. weberbaneri are the only species with long hairs associated with the pedicel. Eviochloa michauxit, E. polystacyha, and E. weberbaueri share the presence of paired spikelets. Therefore, the presence of hairs at the pedicel apex and paired spikelets distinguishes E. weberbaueri within this subgroup of mor- phologically related species. Perhaps the most diagnostic characteristic of E. weberbaueri is the smooth and relatively shiny upper floret. This characteris- tic appears to be unique for E. weberbaueri within the genus. This species appears to be associated with drier environments at elevations between 1,500 and 3,000 meters. WEBSTER, PETERSON AND SHAW, Eriochloa weberbaueri 59 SPECIES DESCRIPTION oe ee Mez, Bot. Jahrb. Engler 56: Beibl. 125:11. 1921. t. Cajamarca: A. Weberbauer 4135 (HOLOTYPE: B; photographs and ies at for Plants perennial; stoloniferous; rhizomatous. Rhizomes with compacted nodes; with glabrous scales. Flowering culms 3—8 dm tall; not caespitose; erect from the base; not rooting at the lower nodes; with a hard knotty base; terminated by a solitary inflorescence; branched from the base; 2—5-noded. Nodes not swollen; pubescent. Internodes hollow; glabrous; smooth; neither viscid nor glaucous. Leaves cauline or basal; ; green; myihour auricles. Sheaths 6-13 cm long; overlapping or not overlapping; not i h; glabrous; rounded; not keeled; not ciliate; closed. Teale a fringe of hairs; 0.5-1 2mm long. Collar not differentiated; pubescent. Leaf blades 7-20 cm long; 1.5— 5 mm wide; linear; flat, involute, or convolute; flexuous; spreading; lax; smooth on the lower and upper surfaces; glabrous on the lower surface; glabrous or hairy (minutely pubescent) on the upper surface; with a truncate base; with flat margins; with margins not thickened; with scabrous margins; with glabrous margins; with the midrib not obviously differentiated; lacking obvious transverse veins; acuminate. Prophyllum with scabrous nerves; with pronounced keels. Peduncle 160-300 mm long; glabrous. Panicle 10-35 mm wide; fully exserted at maturity; oblong. Lowermost inflorescence node not differentiated; hairy. Main axis present; 60-150 mm long; straight; stout; hairy; pubescent; with hairy axils; with quaquaversal primary branches. Primary branches 10-30 mm long; 0.3—0.5 mm wide; appressed to the main axis or spreading; with appressed secondary branches; with secund spikelets; not whorled; 8-13 on the main axis; straight; not winged; smooth; hairy; setose; ciliate; with loosely arranged spikelets. Rachis terminating in a spikelet. Pedicels present; 0.2—0.5 mm long; straight; scabrous; hairy; setose; with hairy apices; discoid at the apex; perpendicular with the spikelet base; concave. Disarticulation at the spike- let base. Callus differentiated; not prolonged into a stipe; not flared to form a discoid receptacle; not oblique; rounded; smooth; glabrous. Cleistoga- mous inflorescence absent. Spikelets 3.1—3.9 mm long; 0.9-1.1 mm wide; paired; densely overlap- ping; evenly distributed on the rachis; not divergent from the axis; homo- morphic; adaxial; green; dorsiventrally compressed; lanceolate. First glume absent. Rachilla pronounced below the second glume; not pronounced between the florets. Second glume 3.0—3.7 mm long; 1 times spikelet length; 1 times the length of the lower floret; lanceolate; membranous; rounded on the back; smooth; 5-nerved; hairy; pubescent; acute to acumi- nate; muticous. Fertile florets 1. Sterile florets 1. Lower floret lacking 60 SIDA16(1) 1994 eS = ; —<— & Sa oe —— ww . °; . . NY SI = = ——— mt ‘i ——— 8 4 YW” p FIG. 1. Eriochloa weberbauert. A. Habit. B. Ligule. C. Primary inflorescence branch. D. Spikelet and pedicel apex. E. Spikelet base and pedicel apex. F Upper floret. G. Upper lemma enclosing stamens. H. Stamens and stigmas. WEBSTER, PETERSON AND SHAW, Eriochloa Weberbaueri 61 stamens. Lemma of lower floret 2.8-3.7 mm long; lanceolate; membranous; smooth; with equal internerve spacing; 3-nerved; the nerves pronounced but not swollen; the nerves smooth; not keeled; with glabrous margins; lacking acentral longitudinal groove; acute; muticous; hairy. Lower lemma hairs not forming a distinct horizontal line; shorter than the upper floret; pubescent; smooth and terete; without apical modifications; white. Palea of lower floret absent. Upper floret 0.9 times the length of the lower floret. Lemma of upper floret 2.3-2.8 mm long; lanceolate to ovate; cartilaginous; slightly convex; smooth; yellow; shiny; with involute margins; with margins of the same texture as the body; glabrous; 3-nerved; the nerves not pronounced; with smooth nerves; without basal modifications; not differentiated at the apex; entire; acute; Muticous. Germination flap conspicuous. Palea of upper floret well-developed; enclosed at the apex; cartilaginous; smooth; the base neither swollen nor protruding; the nerves pronounced but not winged. Lodicules about 0.15 mm long; the nerves not pronounced. Stamens 3. Anthers 1.5—1.7 mm long; yellow. Stigmas yellow to orange. South America: Ecuador & Peru. Specimens Examined: SOUTH BM ERICA ECUADOR. cae de Azuay: 10.2 km N of Ona on the Pan American Hwy at the g the Rio Leon, elev. 1880 m, 24 Apr 1990, PM. Peterson, ‘i = Annable and M.E. ene 8915 se 83 km S of Cuenca on road to Loja, crossing of Rio Leon, elev. 1800 m, 30 May 1990, P.M. Peterson and E.J. Judziewicz 9378 (US!); valley of the rio Paute, between Paute and Cuenca, 26 Apr 1945, Camp E-2346 (US!). Peru. Santa Cruz, I. Sanchez Vega 1031 (US!); 11 Apr 1948, Anderson 740 (US!). ACKNOWLEDGMENTS Gratitude is extended to Dr. Gerrit Davidse for making valuable sugges- tions concerning the taxonomy of this group. REFERENCES Datiwirz, M.J. 1980. A general system for coding taxon ;. Taxon 29:41-46. Datiwitz, M.J., Paine, T.A. and ZurcHER, E.J. 1993. ues a to the DELTA system: a general system for pr eon taxonomic descriptions. 4th ed. CSIRO Division of Entomology: Canberra, Austra Hircucock, A.S. 1927. The grasses cat Ecuador, Peru, and Bolivia. Contr. U.S. Natl. Herb. 24:42 SHAW, R.B. and R.D. WessTeR. sae ti genus Erioch/oa (Poaceae: Paniceae) in North and Central America. Sida 12:165— Tovar, O. 1993. Eriochloa in las gramineas (Poaceae) del Peru. Ruizia 13:343—-347. Wesster, R.D. 1987. The Australian Paniceae (Poaceae). J. Cramer, Berlin-Stuttgart. and J. VaLpes re 1988. Genera of Mesoamerican Paniceae (Poaceae: oe Sida 13:187—22 —___, J.H. Kirxsride - J. VALDES REYNA. 1989. New World genera of th Paniceae Geis Panicoideae). Sida 13:393-417. 62 SIDA16(1) 1994 BOOK NOTICES QueErROL, DANIEL. Undated. 1992. Genetic Resources: A Practical Guide to Their Conservation. (ISBN 1-85649-203-6, hbk; 1-85649-204- 4, pbk). Zed Books Ltd., London, England. $55.00 (hbk); $22.50 (pbk). 252 pp Daniel Querol originally compiled this book as a manual for his course on genetic resources and genebanks at the Chapingo Autonomous University in Mexico. He later adapted it to guide the development of the Nicaraguan Genetic Resources Program and a enetic Resources Program in Peru. Those experiences, over a decade, are reflected in the very practical orientation of this oe Translated from Spanish, it offers a perspective from the * Souch,” countries that have been exploited for genetic resources by the “North,” with unfair compensation. This book offers the basics on genetics, diversity, crop breeding, and wild species use, then describes the methods necessary for collecting material and establish- ing, maintaining, and using a genetic resource conservation program.—Larry Ford ACTA BOTANICA HUNGARICA. Volume 36, numbers 1990- a Teeny papers. El genero Moacroton Croiz ees genus Leucocroton (Euphorbiaceae) A. Boruipi.—New spontaneous taxa fie: genus euphoria. rowa ex E etveroes sem) in Cuba. ‘2%. KERESZTY. ae ae examina- f grasslands—Fumana procumbens and Euphorbia seer a. K. BABOs. —Xylotomical examinations of some Venezuelan species of tree — to the Caesalpiniaceae, Fabaceae and Mimosaceae families— Caesalpiniaceae IT. K. Basos, L.J.C. CUMANA.—Xylotomical examinations of some Venezu- elan species of tree bel sates to the — Caesalpiniaceae, Fabaceae and Mimosaceae— Caesalpiniaceae II. K. Banos, L.J.C. CUMANA —Comparison of the pollen of various angiospermous taxa and 6 sporae of a a maoline 1 quality. S. Gutyas, G. PAtri-DEIm.—C ea phate oo in the pollen spectranfCencral Europe L. STUCHLIK, M. KonzaLovA.—Lig ica L. and Cichorium L. pollen (Gonisesicae Lactuceae). M. Z. Haque, M.B. E, eee eee polinico de sedimentos marinos del Occidente de la Isla de la Juventud (Cuba). MitaGros MONCADA FERRERA, C.E. HERNANDEZ FUENTES, M. CABRERA CASTELLANOS.—Botanical identification of Ipomoea tricolor Cav. seed samples from Hungary and thin-layer chromatographic exami- nation of iotesieaioaeans as ergot alkaloids. L. Botz, E. Hahn, L. Gy. Szabo. me Volume 37, numbers 1-4, 1992. Nineteen papers. New names and new species in the Flora of Cuba and Antilles, IV. A. Boruipi.—A new Erythroxylum species in Cuba. R. Oviepo Prirro, A. BorHibI.—A review of Brazilian rece s of the chrysotricha species group in the genus Tabebuia Gomes ex DC. (Bignoniaceae). J. PACLT ay occa rhidi es yidi vum (Senecioneae: Asteraceae) in Hispaniola. A. oe. . GON oS oO A Iss, ZS. Ornosz-KovAcs.—Is there vegetation continum in mangrove enoee MOH UxkponG.—Xylotomical study of some Venezuelan tree species (Minosaceae I-IV). K Bapos, L. J.C. CUMANA TAXONOMY OF THE SIDA RHOMBIFOLIA (MALVACEAE) COMPLEX IN INDIA V.V. SIVARAJAN and A.K. PRADEEP Department of Botany University of Calicut Kerala 673 635, INDIA ABSTRACT The Sida rhombifolia complex in India is revised. Sida rhomboidea and S. scabrida are reinstated as distinct species; S. retusa, also reinstated as a distinct species, is treated under the older name S. al/nifolia. An artificial key for identification, amended synonymy, descriptions, and illustrations are provided. RESUMEN Se revisa el complejo Sida rombifolia en \a India. S. rhomboidea y § naples son restablecidas como especies diferentes; S. retusa, también restablecida e trata bajo el nombre mas antiguo de S. a/nifolia. Se ofrecen una clave anced para la DP eautesion, sinonimias corregidas, descripciones e ilustraciones The genus Szda L. is ically difficult. Among its species, S. rhombifolia continues to be the hardest nut of all. There is great diversity of opinion as to its circumscription. Even as a world-wide revision of the genus is still want- ing, we have tended to consider this species as a highly polymorphic taxon including widely different elements from different areas. The result is that classifications proposed for one region, say for Malesia (Borssum Waalkes 1966), Nigeria (Ugborogho 1980), Mexico (Fryxell 1988), China (Hu 1955), and Taiwan (Chang 1977), are not applicable to Indian elements. This makes the current taxonomy of the species a little suspect and it seems that the very concept of species needs change. From the treatments of the complex available, it seems that defining the species in this complex into well characterized, easily identifiable units on narrower lines would be of greater service than dumping everything intoa vague and hazy S. rhombifolia. This is the idea that emerged during a critical and exhaustive study of the Indian elements during a revisionary study of the Malvaceae of the area. Sida rhombifolia, as it is understood at present, is one of the most variable and widely distributed species of the genus and has been given some 30 binomials by different taxonomists (Ugborogho 1980). Linnaeus (1753, 1763) described three species now included in this complex, namely S. rhombifolia, 8. alnifolia, and S. retusa, all from India. Cavanilles (1785) added S. alba, 8. canescens, S. microphylla, and S. orientalis, and De Candolle (1824) recognized S. capensis, S. philippica, 8. planicanlis, and S. schrankit. That was SIDA 16(1): 63 — 78. 1994 64 SIDA16(1) 1994 the time when the species had narrower definitions, but they were soon rejected in favor of a broad concept; consequently most such species were sunk into synonymy of a broad S. rhombifolia or were recognized as subspe- cific taxa under it. It was Masters (1874) who broadened the definition of the species; he recognized five different varieties of it in India. Schumann (1892) recog- nized six varieties in Brazil “because he had no clear idea what S. rhombifolia sens. str. was” (Fryxell, pers. comm.). A similar course followed by Baker (1892) and, for the African elements, by Hutchinson and Dalziel (1958), who recognized three varieties of this species in Africa. Borssum Waalkes (1966), working on Malesian Malvaceae, elevated var. rhombifolia and vat. retusa into subspecies and reduced all other binomials into synonymy. This is the treatment widely followed at present. Subsequently, Hatusima (1971) recognized another new subspecies (subsp. z#sw/aris) in the Philippines; this was later reported also from Taiwan (Chang 1977). Ugborogho (1980) kept S. retusa and S. alnifolia as two different subspecies in Nigeria. The classification of this complex becomes all the more difficult because taxonomists differ among themselves in its circumscription. For example, S. orientalis Cav. has been included in S. rhombifolia by some (e.g., Rao 1985), but others treat it as a synonym of S. acuta (Borssum Waalkes 1966). The most confusing part of all is that very different elements from different parts of the world have been treated under the same name by different authors. Thus S. alba L. is actually S. spinosa L., but 8. alba Cav. is 8. rhombifolia. To add to the trouble, Paul and Nayar (1988) treated S. a/ba L. and S. spinosa L. as two different species. So also, 8. a/nifolia L., widely accepted as conspecific with. retusa L., is kept distinct by Ugborogho (1980), who recognized three subspecies in the S. rhombifolia complex in Nigeria, namely, subsp. rhombifolia, subsp. a/nifolia, and subsp. retusa. From his descriptions and plates, it is obvious that he treated at least the two latter taxa in a sense different from that of Linnaeus, because both of them seem to belong to subsp. rhombifolia (sensu Borssum Waalkes, 1966). It is pertinent to quote here the comments of Marais (1983) on this: “He uses some names in a sense other than that of van Borssum Waalkes and myself ... the photographs of the mericarps of S. rhombifolia sens. lat. (p. 83 Fig. 6) are so poorly reproduced that it is impossible to comment on his application of the three Linnaean epithets” (1.e., rhombifolia, alnifolia, and retusa). We undertake here a revision of this complex. We have not been able to study it from the entire range because of difficulties in procuring materials and literature. Instead, we have confined ourselves to the Indian elements. Linnaeus (1753) described S$. rhombifolia as “Sida foliis lanceolato- rhomboidibus serratis” and S$. a/nifolia as “Sida foliis orbicularis plicatis serratis.” Later, he described another species, S. retusa (Linnaeus 1763), as SIVARAJAN AND PRADEEP, Sida rhombifolia 65 “Sida foliis cuneiformibus retusis, serratis subtus tomentosis.” He described all these taxa based on Indian specimens; while describing S. retusa, he indi- cated that it 1s very close to S. alnifolia (“Varietarum S. a/nifoliae statuit Dillenius, mihi distincta visa est”). These three taxa have now been lecto- typified (S. rhombifolia: \ectotype 2. rhombifolia LINN 866. 3, BM—Fryxell 1988; S. alnifolia: lectotype Herb. Herm. III fol. 4. Linn. 260. BM—Borssum Waalkes 1966; S. retusa: lectotype Linn. 866.7, LINN—Borssum Waalkes 1966). We have now studied all these (as photographs) along with Indian specimens and relevant literature and are convinced that S. a/nifolia L. and S. retusa L. are conspecific. Sida alba, another name often reduced to the synonymy of S. rhombifolia, was originally described by Linnaeus (1763) as “Sida foliis cordatis sub- rotundis, stipulis setaceis, axillis trispinosis” based on Indian material (lectotype H.U. Herb. Linn. 866.2). He also indicated its close kinship with S. spinosa (“Similissima S. spinosae, sed folia rotunda flores albi, stigmatibus purpureis; petiolarum apices purpurascentis tota demum planta major’). Consequently, most authors have reduced it to synonymy of S. spinosa (Masters 1874; Borssum Waalkes 1966; Ugborogho 1980; Fryxell 1988). But Paul and Nayar (1988), in their revision of Indian Malvaceae, treated S. alba and S. spinosa as separate species based on presence or absence of stipular spines. We studied the types of S. spinosa (Lectotype: Herb. Linn. 866.1, LINN) and that of S. alba L. (1.c.). They differ mainly in shape of leaves, probably duet phological plasticity (as also acknowledged by Ugborogho 1980) and in stipular spines. However, “the subpetiolar spine (better: spur) is quite variable in expression (even from one node to the next of a given plant!) and its presence or absence is of little taxonomic value. Moreover, it sometimes occurs on several other unrelated species, at least in South America” (Fryxell, pers. comm., 6 Sep 1993). Our observations on Indian specimens endorse Fryxell’s viewpoint. These spines usually occur on the woody branches of older plants. The Linnaean specimen is, most likely, a young plant where spines have not yet developed. Otherwise, his S. spznosa and §. a/ba are identical and hence conspecific. Cavanilles, however, (1785) seems to have attributed the name S. a/ba to a different Indian plant (type: MA, Photo BM, not seen), which is now widely accepted as conspecific with S. rhombifolia. Sida obovata Wall. and S. microphylla Cav. from Bengal are two other names often associated with S. rhombifolia. In fact, Masters (1874) recognized them as two different varieties of the same species. We have not been able to study the type of the latter (Sonnerat s.n., P), but Paul and Nayar (1988) studied it, concluding that it is nothing but a smaller form of S. rhombifolia. We have, however, been able to study the type of S. obovata Wall.(lectotype, Wall. Cat. 1864 from Burma, K). It is a very distinctive plant with obovate-obtuse 66 SIDA16(1) 1994 leaves and long-peduncled, somewhat racemose, axillary inflorescences, which character combination is not known anywhere in the S. rhombifolia complex. In the absence of more material, we are not able to comment upon its exact identity, but are almost certain that its place is not in this complex. In his classification of this complex, Masters (1874) recognized five vari- eties of S. rhombifolia in the Indian subcontinent, i.e., var. scabrida, var. retusa, var. rhomboidea, var. obovata, and var. microphylla, we have already eliminated the last two from the present consideration. Subsequently Borssum Waalkes (1966) found that Malesian materials of this complex fall under two groups, S. rhombifolia and S. retusa, distinguishable by habit, leaf shape, flower size, and some other characters, which he has dealt with in detail (Borssum Waalkes 1953). He also observed that, though the taxa are sympatric, they do not interbreed and produce hybrids in nature (Borssum Waalkes 1966). Consequently, he reyected Masters’ treatment (“the differences between the groups are several, they cannot merely be regarded as varieties”) and gave them the rank of subspecies. But his subsp. rhombzfolia, still, is very highly polymorphic, including in it 8. scabrida and S. rhomboidea. After a critical study of living populations and herbarium specimens we are still at aloss to understand why we cannot consider S. retusa to bea species distinct from S. rhombifolia, in the strict sense of Linnaeus. The differences between the two, Backer (1943), Borssum Waalkes (1953, 1966), listed by Masters (1874), and Ridley (1922), are substantial and adequate for specific separation, notwithstanding the muddling by Ugborogho (1980: 65-75); their reproductive isolation renders them good “biological” species. We reinstate S. retusa as a species different from S. rhombifolia. But one of the problems that usually crops up in distinguishing S. retusa is that taxonomists often rely too much on leaf characters, especially the retuse leaf apex, but this character is not unique to this taxon; an unambiguous classification would not be possible unless we correlate it with mericarp fea- tures. Dr. Fryxell, while commenting on this manuscript (pers. comm.), wrote, I have six specimens in my herbarium, of which three (from India and Ceylon) have muticous mericarps (in agreement with your description), but the other three have long spines (to 3 mm) on the mericarps. These specimens are from Malaysia and two from the Philippines. All have pubescent fruits.” The awns of the mericarps are evident in the Malaysian specimen (Worthington 13027 UTEP, pf), a photocopy of which Dr. Fryxell kindly sent to us. However, after consulting the type of S. retusa and other Indian specimens, we are of the opinion that the Malaysian and the Philippine materials belong elsewhere than to S. retusa. But there is a nomenclatural problem involved. It was already mentioned that S. a/nifolia and S. retusa are conspecific, the epithet “alnifolia” antedating the latter. There is technically nothing wrong with Borssum Waalkes’ (1966) adoption of the later epithet “retusa” for the SIVARAJAN AND PRADEEP, Sida rhombifolia 67 subspecies, because a name does not have priority outside its own rank (ICBN Art. 60), but he has also rightly suggested that “in case this taxon is considered a distinct species, it should be named as S. a/nifolia L.” So, we treat this taxon under the earlier name. Since the publication of S. rhombifolia by Linnaeus (1753), Fleming (1810) described yet another, closely related Indian species under the name S. rhomboidea, based on a specimen named by Roxburgh (see Borssum Waalkes 1966: 197; Fryxell 1988: 403). Fleming (1810) and Roxburgh (1832) described it as differing from S. rhombifolia “in the arils having no horns.” But Wallich, apparently, attributed this name to a different plant (Wall. Cat. 1862 F) and observed that Roxburgh’s plant is S. rhombifolia (distinguished by the two long awns on the mericarps) (cf. Wight & Arnott 1834). It is probably from this time on that S. rhomboidea has been included in synonymy of the latter. However, Wight and Arnott (1834: 58) studied Roxburgh’s figures and material cultivated in the Calcutta Garden by Roxburgh and concluded that they are diff from Wallich’s S. rhombifolia. Borssum Waalkes (1966: 197) also made a critical study of these and concluded that “Wallich’s specimen is obviously misidentified.” He, in turn, designated the two specimens at Brussels (s. loc.,s. coll., no. 2228, photographs seen) as the lectotype of S. rhomboidea. Sida rhomboidea is widely distributed in India, and we had ample oppor- tunities to study it in the field and laboratory and to compare it with S. rhombifolia L. (s.s), which is also plentiful. They differ substantially in several respects, of which fruit characters are most striking. In the case of S. rhombifolia, the schizocarps are prominently beaked at the apex with the pro- jecting awns of the mericarps. The mericarps, each with two long awns, dehisce at the apex, exposing part of the seed before dispersal and they separate from each other even as the fruits remain on the plants. In the case of S. rhomboidea, however, the schizocarps are depressed, somewhat umbili- cated at top, as described by Roxburgh (1832), but not beaked. The mericarps are rather indehiscent, with a beak at apex (not 2-awned). They cohere together closely and are shed as a single unit by articulation of the pedicel, the mericarps separating much later. From our observations here, it is obvious that S. rhomboidea is a distinct taxon, morphologically and apparently reproductively isolated from S. rhombifolia and that it deserves to be treated as a distinct species on its ownas has been done by Fleming (1810), Roxburgh (1832), and Wight and Arnott (1834), instead of drowning it in a highly polymorphic S. rhombifolia. Sida rhomboidea seems to be more akin to S. unicornis Marais from Mauritius. It differs in its glabrous mericarps with obtuse or retuse beaks, whereas S. wnicornis is characterized by pubescent mericarps with sharp-pointed beak. 68 SIDA16(1) 1994 Sida scabrida was originally described as “whole plants sprinkled with rigid, simple or 2—3 (or more) partite hairs ... leaves rhomboid or oblong, lanceolate ... without tomentum ... pedicels jointed at the very base; carpels 9-11, bicuspidate” (Wight & Arnott 1834:57), based on the peninsular Indian specimen Wight 166 (K, CAL). We have collected and studied several specimens that perfectly match the type (photograph seen). Apparently, it is close to S. rhombifolia, so much so that Masters (1874) and Paul and Nayar (1988) treated it as a variety of the latter. But it differs from S$. rhombifolia (s.s.) in a number of characters, notably in the indumentum, coarse pubes- cence of leaves, calyces etc. Apart from that, the mericarps in S. scabrida are stellate-pubescent while those of §. rhombifolia are glabrous. Sida scabrida also closely resembles S. ywnnanensis Hu, with which it might easily be confused. But the former is characterized by adpressed simple hairs on the upper surface of leaves, pedicels articulated at base or not at all, and 7-10 mericarps in contrast to the stellate hairs on the upper surface of leaves, pedicels articulated above the middle, and G—7 mericarps, in S. ywnnanensis. Now the question arises whether these are sufficient reasons for segregat- ing S. scabrida as a distinct species or whether we should retain it as a variety of S. rhombifolia. The answer will have to focus on the manner of delimitation of species in the entire genus. With a circumtropical distribution and wide- ranging variability, species delimitation here has been based on a few character differences with, of course, support from apparent reproductive isolation. In this context, S. scabrida qualifies well for species status and so we are reinstating it as distinct. Our studies on vegetative features (from seedling stage to adult plants) and reproductive characters, especially mericarp morphology (Sivarajan et al. 1992),reveal that in India we have at least four distinct species, i.e., S. rhombifolia (s.s.), 8. alnifolia, 8. stabrida, and S. rhomboidea, which have been uitherto included in S. rhombifolia (s.1.). Amended descriptions and syn- onymy of these taxa are given below. KEY TO SPECIES la. Adaxial surface of leaves with appressed simple hairs, abaxial surface coarsely pubescent with simple and 2—3(5)-rayed stellate hairs; pedicels jointed at very base or not jointed 4.8. scabrida lb. Adaxial surface of leaves without simple hairs, abaxial surface densely pubescent with many- ee stellate hairs; pedicels jointed at about middle ............ 2 2a. Lower leaves aes obovate, retuse or truncate at nae ieee minutely s ibesc ent, mucronate at apex, mu retuse 1. S. alnifolia or emarginate at ape | ~) om Lower leaves not eee retuse or truncate at apex; mericarps glabrous, 2-awned or beaked 2 SIVARAJAN AND PRADEEP, Sida rhombifolia 69 3a. Staminal column stellate- oo mericarps beaked with a single, icous process, indehiscent 3. 8S. rhomboidea Sb: Spacrinal column oe mericarps prominently 2-awned, eee dehiscent S. rhombifolia 1. Sida alnifoliaL., Sp. Pl. 684. 1753. (Fig. 1). Tyre: Herb. Herm. Ill. fol. 4. Linn. 260 (Lecroryre: BM); Wight & Arn., Prodr. Fl. Pen. Ind. Or. 1:58. 1834. Sida baa Sp. Pl. 961. 1763. Type: Herb. Linn. 866.7, LINN; Wight & Arn., Prodr. Fl. Pen. Ind. Or. 1:58. 1834. S, ron var. retusa (L.) ee in Hook. f., Fl. Brit. India 1:324. 1874; Trimen, Ce 2b on 1:143. 1893. S, - a bsp. retusa (L.) Borss., Blumea 14:198. 1966; Mani. & Sivar., Fl. Calicut 44, 1982; Sald. maRaniech in Sald., Fl. Karnataka 1:260. 1984; aaa. in Nayar et al. (eds.), Fasc. Fl. India 19:216. 1988, non Ugborogho 1980. S. rhombifolia subsp. alnifolia Ugborogho, Bol. Soc. Brot. Ser. se 1980. Woody herbs or subshrubs to 50 cm, usually low and strongly branched; stems prostate or ascending, terete, green or purplish grey, stellate-tomen- tose to glabrescent. Leaf blades 0.5-5 X 0.5—4 cm, leaves towards stem base always obovate with retuse or emarginate apex, rarely truncate, upper leaves obovate to elliptic-lanceolate with rounded, subobtuse or acute apex, obtuse or rounded at base, margins irregularly serrate-dentate or crenate distally, entire proximally, upper surface sparsely pubescent with short many-rayed stellate hairs, lower surface densely greyish tomentose with short, many- rayed stellate hairs; petiole 3-5 mm long, stellate-pubescent, shortly pulvinulate below blade; stipules 4.5 mm long, equal, linear to subulate, glabrescent. Flowers axillary, solitary, sometimes in terminal clusters due to reduction of distal leaves; pedicels longer than petioles, 3-4 mm long in flower, to 30 mm in fruit, glabrous, jointed about middle; calyx 6-7 mm diameter, 6 mm long, pubescent with minute stellate hairs, glabrous within, 5-lobed, lobes 2 mm long, ovate-triangular; corolla 12 mm diameter, orange yellow, petals 7.8 x 6.5 cm, obliquely obovate, cuneate at apex, glabrous except for minutely hairy base; staminal column 3 mm long, glabrous or minutely stellate-hairy, antheriferous at apex; ovary 1.5 mm diameter, depressed globose, glabrous; styles 7-10, white; stigmas capitate, creamy- yellow. Mericarps 7-10, 2.5—1.5 mm, included in calyx, reticulate or rugose on sides and back, apex with a pair of short stellate-hairy mucros, mucros obtuse, retuse, or emarginate at apex. Seeds 2 mm long, black, glabrous except for puberulent hilum. Distribution and Ecology. Sida alnifolia is widely distributed in the plains and hills of Southern Peninsular India, and occurs along Ghat roadsides, forest clearings as secondary growth, lateritic hill slopes, and occasionally as a weed in upland cultivation. It flowers from August to April in Kerala. The flowers open at about 0900 am and wither by 1400. 70 SIDA16(1) 1994 t) yy LZ te Z w Sey / 3 wll N 3 3 3 3 woe Mu a Me 3 5 uy SS 7 ae wt bea dhe ood FIG. 1. Sida alnifolia. A. Flowering branch; B—E. Range of leaf variation; F. Indumentum; G. Schizocarp with accrescent calyx; H. Schizocarp with calyx removed; I-J. Mericarps, lateral and frontal view; K. Seed; L. Seedling. cimens examined: ANDHRA PRADESH: Cuddapah Dt.: without precise locality, a, 4634 (MH). Guntur Dt.: without precise locality, Barber 4632 (MH). East — Dt.: Bison Hill, Barber 5104 (MH). RNATAKA: Chikmangalur Dt.: Santaveri, Pradeep 47732 (CALI). Dakshin Kannad SIVARAJAN AND PRADEEP, Sida rhombifolia 71 ee Barber 2095 (MH); Uduppi, sels 5060 ee Nalgody-Kutajadri road, Pradeep 5041 (CALI). Kodagu Dt.: Sampaji, Barber 2209 (MH). Uttar Kannad Dt.: Karwar, Pradeep 5017, 5018; Shrunkeri, Pradeep 5025B; co ie s, cae 5041 (CALD. KERALA: Idukki Dt.: Munnar, etl oa 48577 (MH). Kannur Dt.: Kannoth R.E,, Ramachandran 57700 (MH). Kottayam Dt.: Kumaramperoor R.F., Subramanian 8246 (FRC). Kozhikode Dt.: Kadalundi, Pradeep 5013; Badagara, Pradeep 5025 (CALI). Palakkad : Dhony, Pradeep 5268 (CALI); Sebastian 21092 (MH); Venkatasubramanian & Sasidharan 10650(FRC). Thrissur Dt.: Peringalkuth, Pradeep 5228 (CALI); Peechi, Subramanian 9322 (FRC). Wynad Dt.: Tariodu R.F., Pradeep 51617 (CALI TAMIL NADU. Coimbatore Dt.: Maruthamalai, Sebastian 721 (MH); Fischer 2017 (FRC); ae alae CAL): Kottabomman Dt.: Daneel Be ee bho H). Nilgiri Dt.: Theppakad, V: 243051(MH).R ie 72378 (MH). Saleny: De Hunasur cattle farm, — amt 2925 (MH). 2. Sida rhombifolia L., Sp. Pl. 684. 1753. (Fig. 2). Type: “2 rhombifolia” (LECTOTYPE: LINN - 866.3: ISOLECTOTYPE: S$); Roxb., Fl. Ind. ed. Carey 3:176. 1832; Mast. in Hook. f., Fl. Brit. India 1:323. 1874, in part; Trimen, Handb. Fl. Ceylon 1:143. 1893, ace, var. 3 retusa; Dunn in Gamble, Fl. Pres. Madras 1:90. 1915; Ramam. in Sald. & Nicolson, Fl. Hassan Dt. 155. 1976; Britto & Matthew in Matthew, Fl. Tam. Carnatic 3:131. 1983, 1n part; Bevel, Syst t. Bot. Monogr. 25:403. 1988, in part, excl. syn. S. rhomboidea, Mani., Fl. Silent Valley 28. 1988, in part., Vajravelu, Fl. Palghat Dt. 84. 1990, in part, excl. syn. vars. obovata, retusa G rhomboidea. Sida rhombifolia subsp. rhombifolia: Borss., Blumea 14:193. 1966; Mani. & Sivar., Fl. Calicut 44. 1982; Sald. & Ramesh in Sald., Fl. Karnataka 1: 259. 1984 (all in part, excl. syn. S. rhomboidea). S. rhombifolia subsp. rhombifolia var. rhombifolia: Paul & Nayar in Nayar et al. (eds.), Fasc. ia 1 4. 1988. S. alba Cav., Diss. I. 22.t. 3.f.8. 1785, non Linn., 1763. Erect branched undershrubs to 1 m tall; stems terete, green or purplish, cinereous with many-armed short-stellate hairs. Leaf blades 5—6 x 2—2.5 cm, elliptic to rhomboid, rounded to truncate at base, 3-nerved, serrate distally, entire towards base, upper surface green, glabrescent, lower surface densely cinereous with short-stellate hairs, appearing farinaceous; petiole S—-6 mm long, densely pubescent, shortly pulvinulate at both ends; stipules up to 5 mm long, equal linear-lanceolate, 1-nerved, margins simple-hairy. Flowers axillary, solitary, sometimes in apparent racemes due to reduction of upper leaves; pedicels 0.5—1 cm long in flower, to 5 cm in fruit, glabrous, articu- lated above middle; calyx 5—6 mm diameter, campanulate, 10-ribbed at base, pubescent, 5-lobed, lobes 3 mm long, triangular, apex acute-acumi- nate, externally pubescent with short-stellate and simple hairs, glabrous within; corolla 1 cm diameter, pale yellow or creamy-white, veins some- times tinged with red at center; petals 9-10 x 6-7 mm, obliquely obovate, truncate or cuneate at apex, short stellate-hairy or glabrous at base; staminal column 3 mm long, glabrous, antheriferous at apex; ovary 1.5 mm diameter, conical, glabrous; styles 8-10; stigmas capitate, yellow or pale pink. Meri- 72 SIDA16(1) 1994 FIG. 2. Sida rhombifolia. A. Flowering branch; B—C. Range of leaf variation; D. Indumentum; E. Schizocarp with accrescent calyx; F. Schizocarp with calyx removed; G-H. Mericarps, dorsal and lateral view; I. Seed; J. Seedling. SIVARAJAN AND PRADEEP, Sida rhombifolia rie carps 8-10, 4 x 3 mm, included in calyx, trigonous with acute angles, pale when mature, distinctly and transversely rugose on sides and back; apex with a pair of short, glabrous, divergent awns 1-1.5 mm long (as long or slightly shorter than calyx). Seeds 2 mm long, black, minutely hairy at hilum. Distribution and Ecology. Sida rhombifolia is almost world-wide in distribution, occurring throughout the tropics of both the Old and New worlds and extending into temperate zones (Fryxell 1988). It is common throughout India and occurs along roadsides and wastelands at altitudes from 1000 to 2000 m. It flowers almost throughout the year, the flowers opening late in the morning, usually becween 1045 and 1100. Specimens examined. KERALA: Kottayam Dt.: Sabarigiri, Sivarajan 5211 (CALI. Thiruvananthapuram Dt.: Ponmudi, Pradeep & Sivarajan 50463 (CALI); Bonecaud, si a pee (MH). L NADU: Kanyakum «| Kothayar, Pradeep 44925, 44916 (CALI). Pi ease Dt.: Courtallum, ae 5105 (CALI). 3. Sida rhomboidea Roxb. ex Fleming, Asiat. Res. 11:178. 1810. (Fig. 3). Type: s. loc., 5. coll. 2228 (BR); Roxb., Hort. Beng. 50. 1814 & FI. Ind. ed. Carey 3:176. 1832; Wight & Arn., Prodr. Fl. Pen. Ind. Or. 57. 1834; Dunn in Gamble, Fl. Pres. Madras 1:90. 1915 Sida rhombifolia var. rhomboidea (Roxb. ex Fleming) Mast. in Hook. f., Fl. Brit. India S. rhombifolia subsp. rhombifolia sensu Borss., Blumea 14:193. 1966; Mani. & Sivar., Fl. Calicut 44. 1982; Sald. & Ramesh in Sald., Fl. Karnataka 1:259. 1984; Nair & Nayar, Fl. Courtallum 1:75. 1986; Fryxell, Syst. Bot. Monogr. 25:403. 1988; Mani., Fl. Silent Valley 28. 1988; Paul & Nayar in Nayar et al. (eds.), Fasc. Fl. India 19:214. 1988; Vajravelu, Fl. Palghat Dt. 84. 1990 (all in part). Erect much branched subshrubs to 2.5 m tall; stems terete, usually pur- plish, minutely pubescent with small stellate hairs. Leaf blades on younger shoots much larger (7-8 X 5—6 cm), obovate or suborbicular, truncate or rounded at base, subobtuse or acute at apex; those on flowering shoots smaller, 1-5 X 0.5—3 cm, rhomboid to lanceolate, 3-nerved from base, lateral nerves 3—5 pairs, nerves raised on lower surface, margins coarsely serrate to crenate, entire towards base, densely stellate-tomentose beneath, sparsely pubescent above; petiole 1-15 mm long, stellate-pubescent, shortly pulvinulate at both ends; stipules 9 mm long, equal, linear, slightly pur- plish, caducous. Flowers axillary, solitary; pedicels 6 mm in flower, to 30— 40 mm in fruit, filiform, glabrous, articulated at about middle; calyx 9 mm diameter, campanulate, 10-ribbed at base, 5-lobed, lobes 3 mm long, tri- angular, outer surface densely tomentose with minute stellate hairs, inside nearly glabrous, margins purplish; corolla 1.5 cm diameter, pale yellow; petals 7-8 X 5—6 mm, obliquely obovate, retuse or emarginate at apex, glabrous; staminal column 3 mm long, stellate-pubescent, antheriferous at apex; 44 SIDAI6(1) 1994 Wer Yi G SS SOY VE — FIG. 3. Sida rhomboidea. A. Flowering and fruiting branch; B—F. Range of leaf variation; G. Indumentum; H. Schizocarp with accrescent calyx; I. Schizocarp with calyx remvoed; J-K. Mericarps, lateral and dorsal view; L. Seed; M. Seedling. SIVARAJAN AND PRADEEP, Sida rhombifolia 75 ovary 1.5 mm diameter, depressed globose, glabrous; styles 8—10; stigmas capitate, yellow. Mericarps 8-10, completely included in calyx, closely co- herent, 3 x 2 mm, indehiscent, trigonous with acute angles, prominently reticulate on sides, reticulate or rugose on back, apex beaked with a single, glabrous muticous process. Seeds 2 mm long, brownish black, glabrous throughout. Distribution and Ecology. Sida rhomboidea is common in the plains of Peninsular India. It occurs along roadsides and waste lands generally at sea level. Because of its frequent confusion with S. rhombifolia, it is difficult to draw conclusion on its distribution elsewhere, based on literature. The plant flowers almost throughout the year. The flowers open between 1045 and 1100 and wither by 1415 Specimens examined. KERALA: Ernakulam Dt.: Wellington Island, Pradeep ren Cochin, hei 5179 (CALI). Kannur Dt.: Tellicherry, Punnol, Pradeep 5107 (CALI). Kottayam Dt.: Changanacherry, Antony 904 (MH). Kozhikode Dt.: West Hill, — D219; so ev Jayakumar 884; Cheruvannur, Sivarajan 1482 (CALI). Malappu Dt.: Parappanangadi, Pradeep G& Sivarajan 5002 (CALD. Thiruvananthapuram De: Benecud Mibaner 63246 (MH). TAMIL NADU: Ramanathapuram Dt.: Vajravelu 33853 (MH). 4. ee — Wight & Arn., Prodr. Fl. Pen. Ind. Or. 57. 1834. (Fig. 4). E: Peninsular India, Wight 166 (CAL, K,); Ugborogho, Bol. Soc. Brot., 54:100. at ss ee var. scabrida (Wight & Arn.) Mast. in Hook. f., Fl. Brit. India 1:324. oe combi o rhombifolia vat. scabr Sr ae ) Mast., Paul & Nayar in Nayar asc. Fl. India 19:21 Erect Seach subshrubs to 2 m ee stems terete, green or slightly tinged with purple, pubescent with minute stellate hairs intermingled with scattered, long, simple hairs. Leaf blades 6-8 x 3—4 cm, concolorous, rhom- boid or oblong-lanceolate, truncate at base, acuminate at apex, basally 3- nerved, lateral nerves 4—5 pairs, margins serrate-crenate distally, entire towards base, coarsely pubescent on upper surface by appressed simple hairs, on lower surface with few scattered 2—3 armed stellate and simple hairs especially on the nerves, never with tomentum; petiole 0.5—1 cm long, pubescent with minute stellate and long simple hairs; stipules 4-6 mm long, subulate or filiform, ciliate. Flowers axillary, solitary, sometimes in clusters of 3—5 due to reduction of distal leaves. Pedicel up to 1 cm in flower, to 3 cm in fruit, not articulate; calyx 6-7 cm diameter, 10-ribbed at base, campanulate, 5-lobed, lobes 4 x 4 mm, ovate, acuminate at apex; corolla 1.5 cm diameter, yellow; petals 1-12 x 7-8 mm, minutely stellate-hairy at base; staminal column up to 3 mm long, stellate-pubescent; ovary ovoid, 2 mm 76 SIDA1G6(1) 1994 eh K G. 4. Sida scabrida, A. Flowering and fruiting branch; B—E. Range of leaf variation; F. Indumentum, G. Schizocarp before dehiscence; H. Schizocarp showing dehiscence; I—J. Mericarps, dorsal and lateral view; K. Seed; L. Seedling. diameter, minutely simple-hairy towards apex; styles 7—10; stigmas sub- globose. Mericarps 7-10, 3 X 3 mm, enclosed in calyx, trigonous with acute angles, black when mature, minutely stellate-pubescent on back towards apex, prominently reticulate or transversely rugose on sides and back, apex with a pair of linear divergent simple and stellate-pubescent awns. Seeds 2 mm long, blackish or brown, pubescent at hilum. Distribution and Ecology. Sida scabrida is known from Peninsular India SIVARAJAN AND PRADEEP, Sida rhombifolia 77 and (disjunctly) from Nigeria in Africa. It occurs in disturbed sites in evergreen, semi-evergreen, and moist deciduous forests. It also grows along roadsides generally at an altitude of 250-2000 m. The plant produces flowers principally from August-April. The flowers open between 1045 and 1100. Specimens examined. KARNATAKA:. Chikmagalur Dt.: Bababudan Hills, Pradeep & Sivarajan 47726 (CALI). Kodagu Dt.: Mercara, Sampaji, Barber 2315 (MH KERALA: Kottayam Dt.: Pampa, Swbramanian 9969 (FRC). Kozhikode Dt.: Badagara, Pradeep 5101 (CALI, pf). Malappuram Dt.: Nedunkayam, Pradeep & Sivarajan 5097 (CALI). Palakkad Dt.: Silent Valley R.E, Nair 64374, Vajravelu 26103 (MH Pathanamthitta Dt.: Moozhiyar, Subramanian 9620; Mahadevan & party 9193 (FRC), dol Kumar 1513 (MH). Thiruvananthap Dt.: Ponmudi Hills, Mohanan 52542 (MH). r Dt.: Peringalkuthu, an $297 (CALI). Wynad Dt.: Lakidy, Pradeep 5157, eet near Pookkottu lake, Pradeep 50457 (CALI); Chandanathode, E//is 29438 (MH). TAMIL NADU: Coimbatore Dt.: Attakkati-Valparai road, Sivarajan es 47761 (CALI). Kanyakumari Dt.: Kothayar, Pradeep 44929 (CALI). Nilgiri Dt.: Devala R.E, Vajravelu 42831; Kattaicombai, oe 1080; without precise rea “Wight iat, (MH). Salem Dt.: Salem, Deb 31265 (M ACKNOWLEDGMENTS We are indebted to Dr. Paul A. Fryxell, Texas A&M University, for com- ments on an earlier version of our paper and for useful suggestions; to Dr. P. Bamps, National Botanical Garden of Belgium, for some of the type photo- graphs; to Dr. R.E. Ugborogho, University of Lagos, Nigeria, for help with literature; and to the Jt. Director, Botanical Survey of India, Coimbatore, for providing facilities to study materials at MH. AKP is grateful to the Council of Scientific and Industrial Research, New Delhi for a fellowship. REFERENCES Backer, C.A,. Bor Bexnopte flora veh Javea (em. - ) 4C, fam. 109. BAKER, E.G. 1 g | J. Bot. 30:136—-142, 235-240, 290-296, ae 332. BorssuM WAALKES, J. van. 1953. Enkele bloeiwaarnemingen bij twee Sida-soorten. Trop. Natuur 33: 21-2 . 1966. Malesian Malvaceae revised. Blumea 14:1—213. CaNDOLLE, A.P. de. 1824. Malvaceae. In: A.P. de Candolle (ed.), Prodromus 1:427—474. CavaNiLLes. A.J. 1785-1790. Monodelphiae classis dissertationes decem. Paris & Madrid. CHANG, CHING-EN. 1977. Malvaceae. In: Hui-Lin Li et al. (eds.) Flora of Taiwan 3:710—728. FLEMING, J. 1810. A catalogue of Indian medicinal plants and drugs, with their names in the Hindustani and Sanskrit languages. Asiatic Res. 11:153-196. FryXeLL, P.A. 1988. Malvaceae of Mexico. Syst. Bot. tess 25:1-522. Hatusma, S. 1971. Flora of the Ryukyus. Naha, Okina Fuso. 1. 1955, Mora . China, family 153: Malvaceae. a aica Plain: Arnold Arboretum of Harvard Univers HUurTcHINsON, J. ae DatzieL. 1958. Flora of West Tropical Africa. Ed. 2. London. Vol. I(2). Linnaegus, C. 1753. Species plantarum. Stockholm. 78 SIDA16(1) 1994 —__ 1763. Species plantarum, ed. 2. Stockholm. Marais, W. 1983. Notes on Mascarene Malvaceae. Kew Bull. 38:41—46. Masters, M.T. 1874. Malvaceae. In: J.D. Hooker (ed.), Flora of British India 1:317-353. PauL, T.K. and M.P. Nayar. 1988. Malvaceae. In: ae = Nayar, K. Thothachri, and M. Sunjapia (eds.), Fascicles of flora of India. 19:64—2 Rao, R.S. 1985. Malvaceae. In: Flora of Goa, Diu, oa Dade and Nagarhaveli. 1:31—4 RipLey, H.N. 1922. The flora of the Malay Peninsula 1:253—260. RoxpurGH, W. 1832. Flora Indica. Serampore SCHUMANN, K. 1892. Malvaceae I. In: Urban, I. (ed.) Flora Brasiliensis 12(3):253—456. SIVARAJAN, V.V., A.K. PRADEEP and E.J. GEORGE. 1992. Mericarp morphology of Indian species of Sida L. (Malvaceae) in relation to taxonomy. J. Taiwan Mus. 45:65—73. UcxBoroGuo, R.E. 1980. The taxonomy of Sida L. (Malvaceae) in Nigeria. I & I. Bol. Soc. , Sér. 2, 54:5—-40, 65-8 c.. R. and G.A.W. ARNOTT. 1834. Prodromus florae peninsulae Indiae orientalis. London. SOME OBSERVATIONS ON LEAF FORM IN ILEX VOMITORIA (AQUIFOLIACEAE) RICHARD STALTER Department of Biological Sciences Saint John’s University Jamaica, NY 11439, U.S.A. DWIGHT T. KINCAID Plant Sciences Doctoral Subprogram Lehman College, City University of New York Bronx, NY 10468, U.S.A. ABSTRACT We report a statistical analysis of leaf form for a sample of 481 collected from five ‘yaupon trees at Ponchatoula, Louisiana. Considerable variation exists within these “sun” leaves. Across the 5 trees, surface area varies 24-fold (22.1-533 mm/?), leaf specific mass 22-fold (7.67-167.83 g m”), leaf length 5-fold (6-31 mm), leaf width 4.5-fold (4—18 mm), and crenations 4.7-fold (7-33 per leaf). Leaf complexity (LC) varies from 1 to 6 Fourier frequencies, and the leaf dissection index (DI) varies from a nearly circular 1.036 toa high of 1.349. Trees i in this population a: are estausucally significantly different from each other in leaf eaf mass, leaf ,and in number of crenations per leaf. In logistic regression, Ene piopebility that: a leat wall develop more crenations, increases with leaf size. COMStdht of leaf di iond hologi e- Leat saree area is strongly related to nodal position on ae iss caidalle aodes ners produce the largest leaves. Key Worps: Ilex vomitoria, yaupon, leaf form RESUMEN Se realiza un andlisis estadistico de la forma de la hoja sobre una muestra de 481 hojas colectadas en cinco arboles en Ponchatoula, Louisiana. Existe una variaci6n Seton entre esta hojas de “sol.” En cinco arboles, el drea de la superficie varfa en 24 veces (de 22 mm_7), la masa especifica de la hoja en 22 veces (de 7.67-167.83 gm“), la ae ae en cinco veces ie 6-31 mm), la anchura de la hoja en 4.5 veces (de 4-18 mm), y las crenaciones en 4.7 veces (de 7—33 por hoja). La complejidad igs hoja eve varia de 1aG frecuencias de Tons y el indice de disecci6n ey ri 1.036 hasta 1.349. Los 4 ta poblacién muestran ienificati if & el tamafio foliar cn masa foliar, la masa foliar especffica, y en el nimero de crenaciones por hoja. En regresi6n logistica, la probabilidad de que una hoja desarrolle mas crenaciones aumenta con el tamafi de la hoja. Cuando varia el tamafio de la hoja, su porns permancee ta poblacién; la di 5n de la ho variable Se El] area ae la superficie foliar esta ae relacionada con la posicién nodal en la rama, los nudos medios producen generalmente las hojas mas grandes. SIDA 16(1): 79 — 92. 1994 80 SIDA16(1) 1994 INTRODUCTION Advances in computer-assisted image analysis expand the ability of bota- nists to use large sample sizes in leaf morphometric research (e.g., Kincaid and Schneider 1983, White et al. 1988). Our objective was to quantify leaf variability ina population of yaupon, I/ex vomitoria Aiton (Aquifoliaceae) at Ponchatoula, Louisiana. We collected leaves from five trees in order to answer these questions: (1) Are trees homogeneous in leaf size and shape? (2) Is leaf form related to nodal position along the twig? (3) Do predictive relationships exist among leaf specific mass (g dry weight / m? surface area), crenation number, mass, area, dissection index, and leaf shape complexity? (4) How do average leaf images per tree, reconstructed by Fourier transform, compare to conventional morphometric statistical analysis? Yaupon is a shrub and small tree common in forests along the Coastal Plain from southern Virginia to Florida, and west to Texas (Elias 1980). The leaves are small, flat, coriaceous, evergreen, elliptical, and have marginal mucronate crenations (Radford et al. 1968). METHODS A sample of 481 leaves was collected, on August 12, 1989, from 5 trees growing within 100 meters of each other along a sunlit edge of a pine forest at Ponchatoula, Louisiana. Leaves were individually numbered with a serial number and nodal position on the current year’s shoot, the petiole excised, and the blades placed into a plant press. After drying in a convection oven at 70°C, the leaves were weighed to the nearest 0.0001 gram. Crenations were counted using a stereo dissecting microscope. Maximum length and width were recorded for each blade. In our laboratory, leaf images are analyzed (Fig. 1) using the leaf boundary method of Kincaid and Schneider (1983) which ts based on Fourier transform. In an analysis of various computerized leaf morphometric methods, White, et al. (1988) found this method performed well in terms of discriminating power and in the reconstruction of synthetic, average leaf images. Leaves were photographically enlarged (Fig. 2), and the images bound- aries digitized into x, y coordinates using a graphics pad (Model CR1212, Summagraphics Corp., Fairfield, CT) attached toa Macintosh Ici computer (Fig. 1). Other details are in Kincaid and Schneider (1983), and in Figures —3. Image information les in the values of the Fourier coefficients at each frequency (Table 1). Leaves have the same size and shape, if and only if, their Fourier coefficients are identical (Kincaid and Schneider 1983). Using this method, leaf surface area, leaf complexity, and leaf dissection index were computed for 342 out of the 481 leaves. Leaf complexity (LC) is a dimension- less and discrete, ordinal variable providing a mathematical measure of the “complexity” of a leaf’s outline. For example, LC = 1 for pure ellipses, and STALTER AND KINCAID, Ilex vomitoria The Leaf Boundary Method Leaf of Tlex vomitoria Pressed Leaf or O 4 Photocopy Placed onto Digitizer Pad Digitize Vv Boundary Stylus Boundary is traced — counterclockwise in an x,y coordinate system. Points are obtained, smoothed, and redistributed to produce oo equally spaced points (e.g. 256, A Fourier transform performed on these points 512, 1024) along the boundary. generates 2 coefficients that completely specify the leaf image. These coefficients are stored on disk in a file for each leaf. FIG. 1. Leaf boundary method of Kincaid and Schneider (1983). 82 SIDA16(1) 1994 0 FIG. 2. Enlarged, photographic image of leaf #30 with ruler (small divisions are mm), as digitized. (Flat-bed scanners have now replaced photography as a preferred method of image capture and manipulation prior to digitization of boundary and Fourier Transform.) STALTER AND KINCAID, Ilex vomitoria 83 DOOD a ae el a, FIG. 3. Images of leaf #30 are reconstructed by reverse Fourier Transform, using various frequencies (M). Reconstruction with M = 1 (the first frequency) yields the “best fit” ellipse. = 3 reconstructs leaf #30 to within 95% of a full reconstruction (M = 128). The finer details of the leaf margin ar ded by the higher frequencies. Notice the image transition between M = 16 and M = 32, and that frequencies greater than M = 32 add no detectable marginal detail. For leaves of simple shape, and for studies such as this one where detailed 10 mm image analysis of mucronate crenations is not paramount, one need only deal with coefficients from the lower frequencies. LC = 1 for most elliptically shaped leaves with relatively smooth margins. The more complicated the leaf shape, relative to the best fit ellipse, the greater the integer value of LC (In terms of reverse Fourier transform, LC is the number of frequencies necessary to reconstruct the leaf image to within 95% of the actual image.). While LC captures a mathematical aspect of leaf shape, dissection index (DI) is the empirical relationship between leaf perimeter and leaf surface area expressed as a dimensionless, continuous measurement variable, perhaps amenable to biophysical interpretation of convective heat exchange. The minimum value of DI is that of a circle for which DJ always equals 1.0. The value of DI for a circularly shaped, entire leaf is slightly larger than 1.0, and the more deeply lobed, dissected or lanceolate a leaf shape, the larger the value [DI = Perimeter / {2*SRQ(Area * pi)}}. These variables, defined by Kincaid and Schneider (1983), are useful in making comparisons of leaf shape among leaves having different surface areas. 84 SIDA16(1) 1994 mae tb TABLE L. Fourier f| ] leaf (#4 30) of | li pl 1 for the first 14 frequencies (out of a total of M = 128 frequencies). M = 3 encodes 95% of this image and M = 8 is sufficient for data analysis for this species. Reconstruction of the image of a leaf using M = 1 (the +1 x : ] } i i a VG | Pe | : . tf bs ae) (ee tee a her and -1 freq.) y 1 r oO d / marginal details. If the Fourier coefficients are for a circle, the +1 real coefficient is the radius and all other coefficients are zero. Fourier Coefficients Frequency Real part Imaginary part l 12.34 -.26 -1 3.53 26 Z .O2 -2 19 02 3 eld -.1] -3 48 Ol 4 06 02 -4 0) -.01 5 2) -.09 -5 20 -.02 6 0) 03 -6 02 Ol 7 12 -.03 -7 09 @) 8 -.O1 Ol -8 O02 02 9 08 -.01 -9 .O7 0] 10 Ol 0 -10 -.02 O 11 06 -.02 -l1 O05 Ol 12 -.02 0 -12 O01 ( 13 06 -.02 -13 O03 QO] 14 0 0) -14 0) O01 We used Stat View (Abacus) and JMP (SAS 1989) on Macintosh comput- ers to perform the data analysis. The technique of logistic regression (Pagano and Gauvreau 1993; using JMP) was utilized to search for trends, and to visualize relationships between crenation interval (dependent variable) and leaf surface area, and between classification of leaves by tree (dependent variable) and number of leaf crenations. While logistic regression is a commonly used statistical tool in biomedical fields for analyzing discrete responses, it is rarely used in organismic biology. RESULTS AND DISCUSSION Based on our sample of 481 leaves, considerable leaf variation exists within these “sun” leaves. Across the 5 trees, surface area varies 24-fold (22.1—533 STALTER AND KINCAID, Ilex vomitoria 85 mm7), leaf specific mass 22-fold (7.67—167.83 gm“), leaf length 5-fold (6— 31 mm), leaf width 4.5-fold (4-18 mm), crenations 4.7-fold (7—33 per leaf), and leaf complexity (LC) from 1 to 6 Fourier frequencies. Leaf dissection index (DI) changes from a nearly circular 1.036 toa high of 1.349. Table 2 provides descriptive statistics and results of analysis of variance for these leaves grouped by tree. We found strong evidence for heterogeneity among the five trees for average leaf area, mass, leaf specific mass, crenation number, and dissection index (P < 0.00001 with R-square values for the main effect ranging from 0.153—0.542, Table 2). Interestingly, average LC was homogeneous among the 5 trees (F = 1.7; 4,338 df; P = 0.14) and homogeneous among shoots within each tree (P > 0.05). We predicted that leaf dissection, a variable providing linkage to convective cooling ability, would be positively corre- lated with leaf surface area. However leaf dissection did not correlate with any variable, indicating that as leaf size changes, shape remains relatively constant within this population. In these “sun” leaves, leaf weight (r = 0.91), leaf specific mass (r = 0.24), and crenation number (r = 0.48) increased with leaf size (P < 0.01 for each correlation coefficient). Figure 4 presents the relationship between leaf mass and leaf area. On an individual tree basis, dry weight of leaf is an excellent predictor of surface area (e.g., in Figure 5 for Tree 5, R-square = 0.976) but less so for all leaves (R-square = 0.826). Asa field technique, leaf width pro- vides the simplest predictor of leaf surface area (e.g., for Trees 1 & 4, Area = 25.30 * Width - 60.46, R-sq. = 0.91). Once it is determined that leaf shape changes little with leaf area, regression equations could be used to predict leaf area, as dry weight and/or blade width is easier to measure than leaf area. Increased leaf specific mass usually confers greater water use efficiency and photosynthetic capacity. Average leaf specific mass ranged from 82.6 in tree 4, to 141.3 g m™ in tree 5. We cannot explain why leaf specific mass varied so much in this study, among 5 trees growing within 100 meters of each other (Table 1). Indeed, trees 4 and 5, with essentially the same average leaf images (Fig. 8), had the most divergent values for leaf specific mass. Crenation number, grouped into 5 levels (7-10, 11-14, 15-18, 19-22, and 23—33 crenations per blade), was declared a “response” variable, and analyzed by logistic regression against leaf surface area as an explanatory variable (Figure 5). In logistic regression, crenation value is not predicted, rather, probabilities are estimated for each level of crenation “response, given leaf surface area. The resultant graph, partitions the outcome space into mutually exclusive regions. To use the logistic regression graph, one draws a vertical line at any desired surface area dividing the estimated probability into segments for each level of response. For example, as leaf size increases beyond 200 mm”, the probability that a leaf will have 7—10 1aOTn) ith pay a 2 | : sy H iC, ] tC | ariables f for five trees of y llected August os seer Ponchatoula, LA. Surface area is mm* ieee one side; mass is mg aie weight; leaf specific m mass is g dry weight / m* surface area; ‘dissection index is a dimensionless number for perimeter relative to surface area with circles oo a DI = 1 (Dissection Index = ne een / (2 * SQR(Area * pi)); leaf complexity is the number of Fourier frequencies necessary to reconstruct a leaf image to within 95% of a complete reconstruction. R- square is the proportion of the variability (total sum of squares) in each variable accounted for by the main effect ee oo by tree). ONE-WAY ANOVA VARIABLE Tree 1 Tree 2 Tree 3 Tree 4 Tree 5 F d P R-square Surface Area 138.2 254.6 130.1 98.5 92.6 71.8 4,337 <.00001 0.460 (53.09) (115.50) (39.34) (33.17) (22.33) Leaf Mass 9.3 31.6 12.1 6.9 13.4 140.9 4,476 <.00001 0.542 (.006) (.017) (.004) (.003) (.004) Leaf Specific Mass 92.8 Ly? 92.4 82.6 141.3 55.3 4, 338 <.00001 0.396 (27.76) (19.74) (11.84) (29.56) (15.74) Crenation Number 14.5 17.7 13.3 16.2 15.7 21.5 4, 476 <.00001 0.153 (2.52) (4.49) (2.52) (2.99) (5.48) Dissection Index 1.13 1.14 1.24 1.16 1.15 63.5 4, 338 <.00001 0.429 (.039) (500) (.060) (.041) (022) Leaf Complexity 2.95 2.97 3.16 3.07 3.09 ed 4, 338 0.14 ns 0.020 (.580) (.701) (541) (.464) (.384) 98 V66I (1)9TVAIS STALTER AND KINCAID, Ilex vomitoria 87 ann 1 y=6448.807x + 47.59, R-squared: .826 oO 5004 n=342 S 4 ® 400} @. Pg ° % fe) : fe) 3004 Bs Jo Oo 90g90 . eq E Trees 1-5 g — gs 08 8) = oO © 5 =) 7) ® 100. 4 fe) © O 80; O 60; q fe) fe) 40, Tree 5 20 T T T ¥ T ¥ T ¥ T T T T T T T T T T T T .002 .006 01 014 .018 022 Mass (grams dry weight) . 4. Relationship between leaf surface area and leaf mass. Upper graph: 5 trees (n = 342); outliers (less than 0.02 g) are immature leaves. Lower graph: Tree 5 (n = 33). Least squares linear regression linear regression (P< 0.0001 for both) with 95% confidence intervals for slope given for Tree 5. crenations becomes extremely low. Large leaves (> 400 mm7) are much more likely to develop 15—33 crenations than they are likely to develop 7-18 crenations. In Figure 6, we use logistic regression as a tool to visualize the estimated probability of tree “membership” for a leaf, given the number of crenations on a particular leaf. Trees 1 and 3 have the lowest average number of crenations per leaf; and Trees 2 and 4 have the largest average number (Table 2). The fitted logistic regression curves of Figure 6 provide a display of these trends across the entire range of number of crenations found in this study. Leaf surface area is strongly related to nodal position, but only when 88 SIDA16(1) 1994 1.00 5 4 > 23-33 _ 0.75 4 crenations ios To) ra § 3 © O — = sone £ au crenations c Oo 0.50 - O a - = 15-18 © : Cc crenations iS © +7) O Lu 0.25 - 11-14 2 crenations crenations 4 0 T q T q T q T T T 0 100 200 300 400 500 Leaf Surface Area (sq. mm) FIG. eee regression analysis of number of crenations versus leaf surface area as the explanatory variable. In an alysis of loglike *lthood, chi- -square = | 17.4, P < 0.00001, N = 339. analyzed ona per shoot basis (Fig. 7). For many shoots, 70-90% (R-square) of total variability in leaf area is accounted for by nodal position. This holds true even for long terminal shoots that have experienced, over the growing season, an episodic growth and/or a developmental switch from preformed to neo-formed (produced and released in current growing season) leaf buds (lower right graph in Fig. 7). Shoots within trees had the same average leaf size in ANOVA Average leaf images reconstructed for each tree, as computed from average Fourier coefficients for the first 8 frequencies, are displayed in Figure 8 along with principal component analysis of more conventional morphological variables. The Fourier transform captures only the two-dimensional leaf outline: We see from these average images that Trees 4 and 5 have very similar leaf sizes and shapes, and that Trees 1 and 3 have similar leaf sizes but different shapes. In principal component analysis, leaf area loaded heavily on the first component which accounted for 0.33 of total variance. The shape variables (DI and LC) loaded on the second component, accounting for 0.28 of the variance. Tree number, loaded on the third component, accounting for 0.19 of the variance. As a general rule for biological objects analyzed by STALTER AND KINCAID, Ilex vomitoria 89 1.00 pr OOo Pan = 0.75 - © Tree 4 a Oo OW > 0504 Tree 3 Oo O is 7) Tree 2 LU 9.25 4 Tree 1 0 | T if T qT 5 10 15 20 25 ~=30 35 Number of Crenations on Leaf FIG. 6. Visualization of nelavionship between ee on tree ae manent and are = number of | lo ISTIC = 84.2, P < 0.00001, N = 481). 2 fan) S juar principal components, “size” variables dominate the first component and “shape” variables the second component (Pimentel 1979). The scatterplot of the 342 leaves, graphed in PC space for the first two components, revealed clusters of points corresponding to tree. The 95% confidence ellipses of the bivariate means of each of these clusters are displayed in the lower part of Figure 8. Trees 3, 4 and 5 cluster rather closely together, while Trees 1 and 2 are distinct. Leaf surface area appears to be functionally related to nodal position, with the middle nodes producing leaves that often are 15 to 20 times larger in surface area than leaves at the early and late nodes (Fig. 7). This range in leaf surface area within single shoots, may transcend average differences from populations across the latitudinal and sun-shade extremes of the species. 90 SIDA16(1) 1994 Tree 4 shoot 3 aa Tree 2 shoot 2 190s . Tree 4 shoot 4 LEAF SURFACE AREA, (sq.mm) 4 R sq. = 0.85 ; ail R sq. = 0.79 . 8 T , T 1 a T T T T T T T T 17... 2 8 C 2 4 6 8 2 4 6 8 10 12141618 202224 Node Node FIG. 7. Leaf surface area is functionally related to nodal position. Seventy to ninety percent (R-square in second order polynomial regression) of the variability in area is accounted for by nodal position. This remains true even for long terminal shoots that have experienced, over the growing season, episodic growth and/or a developmental switch from preformed to neo-formed leaf buds (lower right figure). We have established nine study populations of yaupon. Three of these are inland stations: Aiken, South Carolina; Homosassa Springs, Florida; and, the subject of this paper, Ponchatula, Louisiana. We have also established six coastal stations from the species northern distributional limit, Virginia Beach, Virginia, to one of its southern limits, New Smyrna Beach, Florida. Exploratory analysis of the modestly sized data set of this paper has helped us plan collection strategies for the other stations. As botanical methods of computer-assisted image analysis advance (White et al. 1988), research designs should be able to accommodate larger sample sizes of leaves, involving more trees per population, and more shoots per tree. ACKNOWLEDGMENTS For a grant in support of field work, we thank the Research and Develop- ment Committee of the Holly Society of America, Inc., Barton Bauers, Sr., Chairman. We also thank Maria Bilelis, Mark Pokrzywa, Josephine Puppio and Robert Sica, students at St. John’s University, Jamaica, New York for morphometric assistance. STALTER AND KINCAID, Ilex vomitoria 91 O00)0 06 Tree 1 Tree2 Tree3 Tree4 _ Tree5 10mm Small area —_ ---------- > Large area 2 Nn o a QO. = w” Tree 1 & ie 8 5 A ; Tree 2 ! ® ' 2 Tree 4 2 oO Hl x 0-7 xo Y) 7) a Tree 5 i: oo ra @ D = 5 Tree 3 a -1 T T T T " qT -2 | 0 1 2 Leaf Area PC 1 IG. 8. Upper: Average leaf images per tree reconstructed from average Fourier coefficients. Lower: Principal component analysis with 95% confidence ellipses of the bivariate means for each Tree, N = 342 leaves. 92 SIDAILG6(1) 1994 REFERENCES Extas, T.S. 1980. The complete trees of North America. Van Nostrand Reinhold, New York. Kincaip, D.T. and R.B. SCHNEIDER. 1983. Quantification of leaf shape with a microcomputer and Fourier transform. Canad. J. Bot. 61:2333-23 PAGANO, M. and K. Gauvreau. 1993. Principles of biostatistics. Duxbury Press, Belmont, yh NO California. PIMENT EL, R.A. | 979. M | i multivariate analysis of biological data. Kendall/ Hunt Publishing Co. “Dubague Iowa. Raprorb, A.E., H.E. AHLés and C.R. BELL. 1968. Manual of the vascular flora of the Carolinas. Universit ty North Carolina Press, Chapel H SAS. 198 oe User’s Guide. SAS Institute Inc., Cary, North Carolina. Wuitrr, R.J., H.C. Prentice and T. Verwyst 1988. Automated image acquisition and oe description. Canad. J. Bot. 66:450—459 THE MYRICACEAE OF THE UNITED STATES AND CANADA: GENERA, SUBGENERA, AND SERIES ROBERT L. WILBUR Department of Botany Duke University Durham, NC 27708, U.S.A. ABSTRACT The compelling case for recognizing three genera of North American Myricaceae (Comptonia, Morella, and Myrica) is presented. Keys and descriptions are provided for the recognized supraspecific taxa. The nomenclatural basis of each of these genera, subgenera, and series is outlined and discussed if at all controversial. More//a Lour., the largest genus, is here treated as comprising two subgenera of which subg. eee (Tidestr.) Wilbur s found in the Americas and Africa and subg. More//a is restricted to eastern and ees Asia. There are at least four species of Moref//a in the United Aa and Canada and perhaps as many as six: M. cerifera (L.) Small, M. caroliniensis (Mill.) Small, M. inodora (W. Bartram) Small, ye M. cir Cha, & Schlitdl.) Wilbur. Those — which there is some ce do not have binomials in the genus More//a but are known as Myrica pel ania bel (= Cerothamnus plan ns (Mirbel) Moldenke) and Myrica pil Raf. Cniionais ae (Raf.) Small). Myrica L. is represented by the c b M. gale L. ade the Californian M. hartwegit S$. Watson. Comptonia L'Heér. ex Aiton with its only species, C. peregrina (L.) J.M. Coulter, is restricted largely to the northeastern United States and adjacent Canada. New cae nations and/or rankings are provided for the following taxa: Morell subg. Cerothamnus (Tidestr.) Wilbur and series Faya (Webb & Berthel.) Wilbur, ella californica (Cham. & Schltdl.) Wilbur, and for the Azorean M. faya (Aiton) Wilbur. RESUMEN Se presenta un estudio convincente para reconocer tres géneros de Myricaceae de Norte América pean Morella, y Myrica). Se ofrecen claves y descripciones de los taxa lacie reconocidos. Se bosqueja y discute la base nomenclatural de todos los éneros, subgéneros y series, aunque sea controvertida. More//a Lour., el género mas amplio, se trata aqui dividido en dos subgéneros de los cuales saa (Tidestr.) Wilbur se encuentra tanto en América del Norte como del Sur y en Africa, y Morella que esta restringido al este y sureste e de aie Hay al menos cua de Morell en los Estados Unidos y Canada, llegaraseis: M rs ca M. caroliniensis (Mill.) Small, M. znodora Cw. Baga) Small: y M. ealijorneca eee & Schledl.) Wilbur. Las me sobre las que hay dudas, no tienen ningtin binomen en el género More//a, pero se nocen como Myrica pensylvanica Mirbel (= Cerothamnus et (Mirbel) Moldenke) y 5 Mer abate Raf. (= Cerothamnus pusilla (Raf.) Small). Myrica L. esta representado por M. vale L., circumboreal, y M. Aartwegii S. Watson, de California. Comptonia L Hér. ex Aiton con su Unica especie C. peregrina (L.) J. M. Coulter esta restringida al noreste de los Estados Unidos y zonas adyacentes i Canada. Se dan nuevas combinaciones y/o rangos para los siguientes taxa: Morella . Certhamnus (Tidestr.) Wilbur y serie Faya (Webb & Berthel.) Wilbur, Morella — (Cham. & Schltdl.) Wilbur 5 y para M. faya (Aiton) Wilbur, de las Isla Azores. — SIDA 16(1): 93 — 107. 1994 94 SIDA16(1) 1994 For one whose formative years were spent ina section of the country where Small’s (1933) Manual of the southeastern Flora was the basic reference, numer- ous adjustments had to be made to relate to the prevailing, more cOnscivarIVe generic concept then dominating most of American botany. Small’s “micro- genera” were viewed as a provincial aberration. For one’s work to be under- stood by most of the botanical community, one had to convert the generic names employed by Small and other prolific practitioners of the so-called “New York School” suchas Britton, Rydberg, and Barnhart into the broader generic concepts employed by more conservative botanists. For example, how many of us know which common genera are intended when one encounters such generic names as Wallia, Cerothamnus, Tulipastrum, or Phenianthus? Brandegee (1901) argued vehemently against the generic splitting charac- teristic of the New York Botanical Garden under the leadership of Nathaniel Lord Britton. She suggested that genera should be so broadly delineated that all reasonably bright 10 year olds could be expected to know the principal genera of plants and animals in their neighborhood. Clearly Brandegee would be disappointed today for not only would most school boys and girls fail her test but so would most university biology professors. In fact, in the past 2 decades the pendulum has swung back strongly towards the generic standards of Britton, Rydberg, and Small. Who could have predicted 2 or 3 decades ago that the well-known genus Cassia would disappear from the flora of the Carolinas and be replaced by the segregates Chamaecrista and Senna, or that Psoralea would be confined to southern Africa and that those generic names that Rydberg and Small were castigated for using instead (Orbexilum, Pediomelum, and Rhytidomene) would now be very widely adopted for different elements formerly included in Psoralea s.1.? Turning to Myricaceae, we find that Small (1933) recognized three genera in the area of the southeastern United States: Myrica L. (1753), Comptonia LHeér. ex Aiton (1789) and Cerothamnus Tidestr. (1910). Radford et al. (1968) also recognized three myricaceous genera in the Carolinas, differing however in the generic names accepted and the groups to which the names applied: Gale Adans. (1763), Comptonia LHeér. ex Aiton (1789), and Myrica L. (1753). More conservative treatments of Myricaceae recognize, in addi- tion to the controversial and relatively little-known, monotypic, New Caledonian Canacomyrica Guillaumin, two genera sometimes combined as subgenera or sections within the broad concept of Myrica. Under such a conservative scheme, the genus Myrica would consist of three major taxa of very unequal size. The New Caledonian Canacomyrica will not be considered in this paper, but it should be noted that considerable doubt has been expressed as to its relationship with Myricaceae (Thorne 1973). The nomenclature of the three North American myricaceous genera is in part controversial as is discussed below. WILBUR, Myricaceae 95 The morphological evidence summarized in Table 1 argues strongly that there are at least three major groups within non New Caledonian Myricaceae. These three groupings are so fundamentally different that, in my opinion, generic rank should be accorded to each of them. With flowers as greatly reduced as those of Myricaceae, it is not surprising that the most striking differences are found elsewhere—especially in characteristics of fruit and vegetative features. These same groups were treated as three subgenera of Myrica by Engler (1894). Gleason (1952) and Gleason & Cronquist (1963) treated all species as Myrica with no indication given of infrageneric, supra- specific classification. Many authors including Rehder (1949), Fernald (1950), Hutchinson (1964-), Elias (1971), and Gleason & Cronquist (1991) in the past 5 decades have recognized two genera: Comptonia and Myrica (sometimes the latter with the two commonly accepted subgenera, Gale and Morella). Among previous authors recognizing three genera are Chevalier (1901), Rendle (1903), Small (1933), Radford et al. (1968), and Baird (1968). That there are three major groupings within Myricaceae seems generally agreed upon by most students of the family; the only question is the rank or ranks to be accorded to these seemingly natural, monophyletic taxa. Perhaps a comparable case is Quercus, since that genus is traditionally divided into three subgenera by most botanists although Schwarz (1936) treated the subgenera as genera and Oersted (1867) had originally treated those Asian species with connate cupular scales forming concentric rings as the genus Cyclobalanopsis. A table or chart comparing the differences between these three fagaceous taxa is both lengthy and impressive, but evaluation of the differences suggests that they are variations on the same theme as opposed to rather dramatic innovations and new themes expressed in the phyletic development of the three lines of Myricaceae. Although in Table 2 Baird has been outvoted 3 to 1, the preponderance of evidence suggests that he was nomenclaturally correct in the names he employed and the groups to which he applied the names. Small (1903) recognized both Comptonia and More//a in the southeastern United States but was unaware of the presence of Myrica gale there. It is unfortunate that Baird’s detailed investigation was never published. I here acknowledge my indebtedness to Baird (1968) for information summarized in his unpub- lished dissertation that strengthened the case made for the recognition of three genera. It must be admitted that in such matters one can not be dogmatic for the state of our science does not totally exclude individual taste. As long as the groups recognized are seemingly monophyletic (in the pre- cladistical sense) and perhaps reasonably equivalent in morphological differentiation, whether such groups are treated as three genera or three sections or subgenera of one genus is a matter of individual taste doubtlessly tempered by both tradition and the prevailing philosophy of the period. 96 SIDA16(1) 1994 Tasie |. Comparing the three genera of the Myricaceae. MORELLA MYRICA COMPTONIA Terminal buds Present Lacking Lacking Leaves Stipules Aments Stamen number Ovary Fruits Thick, usually persis- a tent, entire, toothed, or rarely incised Lacking Inserted on old wood mainly below the leaves 3-22 Covered with may or if present, non-adnate In very loose clusters; spherical to subspheri- cal nutlet covered by wax-secreting papillae Adnation of bracteoles None with fruit wa Ectocarp: Relative thickness ty Papll presence Tric es presence ae Cell type of mesocarp Wood Chromosome number Thick Absent to den Non-sclerified, waxy Sclerenchyma Diffuse porous n=8 Thin, deciduous, entire or weakly serrate distally Lacking Inserted at the summit of the branchlets of preceding year Usually 4 but ranging from 3-6 (-9) Smooth, flanked by 2 entire bracteoles which develop in fruit into wings, strongly adnate In dense, subcylindrical By. ar keeled nutlet made buoyant by two adnate, enlarged an inflated bracteoles Strongly adnate hin Parenchyma Absent sent Neither hard nor waxy Parenchyma Ring porous n=24, 48 > Thin, deciduous, roundedly pinnatifid Present Inserted in the axils on deciduous branchlets Usually 4 but ranging from 3-7 Smooth, flanked by 2 laciniate bracteoles emerging from the base and developing into a loose cupule In spherical ae conic to cylindri nutlet enveloped by the sas aes and persis- ent bracteoles and ae forming abur-like fruiting structur None Thin Sclerenchyma Absent Sclerified and not enveloped by wax Parenchyma Diffuse porous n=16 WILBUR, Myricaceae one TABLE 2. CHEVALIER (1901) RENDLE (1903) RADFORD ET AL. (1968) BairRD (1968) Myrica Myrica Myrica Morella Gale Gale Gale Myrica Comptonia Comptonia Comptonia Comptonia In recognizing genera, botanists would perhaps find that by placing greater emphasis on the currently minimized morphological adaptations between plants and their environment, the disparity between botanical and zoological practice would be less striking than now is the case. The impor- tance of the genus as an indicator of discrete evolutionary lines would be enhanced if the obvious correlation between muneulone morphology and broadly conceived environmental integration 1 (Inger 1958). In the case of Myricaceae, it should be pointed out that these Prcipholegical asapeaTions to DO enine means of fruit dispersal have not been ignored by nists even if they have not emphasized the functional role. These charac- teristics of a fruit have been recognized by even the most conservative botanists in their formal classification but usually at the subgeneric level. My account merely advocates recognition at the generic level. A principal reason for the widely divergent generic treatments of Myricaceae is not because of differences in interpretations of biological or morphological facts or even in the weight given to these facts; it is simply a disagreement as to the lectotypification of the generic name Myrica L. Chevalier (1901), Rendle (1903), and Radford et al. (1968) in effect all treated Myrica cerifera L. as the lectotype of the generic name Myrica; Baird (1968) accepted Myrica gale L. as the lectotype. This confusion exists whether we accept one genus with three equivalently ranked subgroups or recognize three independent genera, but the differences are obviously much more dramatic under the binomial system of nomenclature if three genera are recognized rather than one genus. Unfortunately the International Code of Botanical Nomenclature (ICBN 1988) still provides remarkably little guidance in the matter of choosing a lectotype. In the absence of clear directions from ICBN, botanists will of necessity flounder along with their divergent conclusions as to what the lec- totype of such Linnean genera as Myrica should be. A special Committee on Lectotypification was established by the Nomenclature Section of the Sydney Congress to resolve problems such as those pointed out by Stirton et al. (1981). This Committee recommended to the Berlin Congress that Art. 8.1 be amended to read “The author who, on or after | January 1935, first desig- nates a lectotype or a neotype must be followed. ...” It was hoped that this stipulation would eliminate the uncertainty connected with use of the term “type” by such early authors as Rafinesque whose concept of type surely was 98 SIDA16(1) 1994 different than that of a present-day investigator and also the uncertainty of the lectotypes designated under both the American Code and the Type Basis Code. Under this suggested requirement, the genus Myrica would have been typified by M. gale as that was the choice of Hitchcock and Green and pub- lished as an unofficial supplement to the ICBN (1935, p. 116). However, the International Botanical Congress meeting in Berlin in 1988 found itself unable to resolve the problems of lectotypification and, since one committee failed to solve the problem to everyone’s satisfaction, three committees were appointed to study and to report their recommendations to the 1993 Con- gress (Tokyo). As might be expected the complicated problems of lectotypi- fication have been largely passed on to the next Congress. Both Myrica gale and M. cerifera have been designated as the lectotype for the generic name Myrica. The species best known to Linnaeus as the only mem- ber of the genus occurring in Sweden, where it is abundant, is Myrica gale. Britton, operating under the American Code, designated this species as the type (= lectotype) of the genus. Hitchcock & Green (1929) made the same choice; their conclusions as to the “standard species” (= lectotype) were listed in the International Rules (1935) in a semi-official way. However, the desirability of Myrica gale as a lectotype was challenged by Hylander (1945) since that species with a very few close allies had been segregated as a small, independent genus. If the much larger clade, represented by Myrica cerifera, were to be segregated from Myrica this larger portion of the genus, a group of ca. 50 species, would belong to this segregate almost all requiring new combinations. Rehder (1949) also accepted Myrica cerifera as the lectotype of the genus Myrica. In spite of this consideration, it seems certain that Myrica gale, the historically best or at least the scientifically longest known species of Myrica, will be confirmed as the lectotype of the genus. Prelimi- nary reports of the subcommittee dealing with lectotypification of Linnaean generic names strongly suggest that Myrica gale will be recognized as the lectotype of Myrica L. KEY TO THE AMERICAN GENERA OF MYRICACEAE . Terminal buds present; distal axillary buds — mature fruit both papillose and wax-covered; anthers forming in the spring shortly prior to flowering; fruit a spherical or subspherical rae covered by wax- secreting g; y g papillae L. Morella . Terminal buds absent; distal axillary buds floral; mature fruit lacking papillae, somewhat resinous but never bearing a waxy coating; anthers formed in the fall preceding flowering; fruit either flattened and keeled or conical to cylindric, neither wax-coated nor shee Leaves stipulate; leaf margin entire or serrate; bracteoles of pistil moMerS 2 2. Bpreped, adnate to ae fruit; ie a Te ene aL ate wa ed nucletc enlarged, unlobed a racteole 2. Myrica WILBUR, Myricaceae 99 2. Leaves stipulate; leaf margins deeply pinnatifid with broadly rounded obes; bracteoles of pistillate flowers 2, deeply lobed into linear segment not adie to the conic to cylindric fruit; fruit a conic to cylindric ae enveloped at maturity in a bur-like involucre formed from the lobed 7 bracteoles and the accompanying bracteal scale 3. Comptonia 1. MORELLA Lour. Morella Lour., Fl. Cochinch. 548. 1790. Type: Morella rubra Lour. KEY TO THE SUBGENERA OF MORELLA } 1 Fs g several ovaries of which only — . Aments branched; pistillate bracts s one develops; fruit 6-8 mm in diameter, covered at maturity by imbricate, fleshy, succulent papillae subg. 1. Morella 1. Aments ae pistillate bracts usually solitary, simple, and 1-flowered; fruit 1-5 mm in diameter, covered at maturity usually by wax-secreting, neither cae nor succulent papillae subg. 2. Cerothamnus MORELLA subgenus MORELLA Morella out. Fl. Cochinch. 548. 1790. Tyre: Morella rubra Lou. Myrica sect. ee aie ) Benth. & ue f ee - 3:401. 1880. Myrica subg. Morella (Loutr.) r, Nat. fener Wi. 27: ee Morella atin contains fewer than 10 species of eastern Asia, the Philippines, and Malaysia. The differences between the species of subg. Morel/a and subg. Cerothamnus from both the Americas and central and southern Africa are so striking as to have elicited expressions of consterna- tion from Greene (1910) that Small (1903) would transfer our southeastern species to More/la; this dismay was shared by Nieuwland (1910). Greene could not believe that plants whose fruits were so palatable and wholesome as to be eaten both uncooked and cooked and so succulent and juicy as to be made intoa flavorful wine could belong to the same genus as the hard, waxy- fruited plants of the southeastern United States or the western coastal region from Vancouver Island throughout much of California. Perhaps Greene was correct in his assessment, but for the present, based largely upon the findings but not the conclusions of MacDonald (1978) and Abbe (1972), the Asiatic plants are here treated as merely subgenerically differentiated from their American and African congeners. Morella Subgenus Cerothamnus (Tidestr.) Wilbur, comb. & stat. nov. Cerothamnus Tidestr., Elys. Marian., Ferns. 41. 1910. Lecrorype: Cerothamnus arborescens astigl.) Tidestr. (= Myrica cer ee L.) ae designated}. Usually aromatic, dioecious or polygamo-monoecious shrubs to small trees with terminal buds. Leaves alternate, pinnately veined, deciduous or evergreen, entire or serrate, estipulate. Inflorescences borne proximally be- low or axillary to the lower leaves, the staminate erect or nearly so, at anthesis 100 SIDA16(1) 1994 thick-cylindric to broadly ellipsoid, bracts broadly to narrowly ovate and at anthesis shorter than the 1—22 stamens, the stamens yellow or becoming yellow, filaments simple or branching and arising from the staminal column at different levels, the secondary, tertiary and even quarternary bracts often present; pistillate inflorescences simple or basally branched, at anthesis ovoid to cylindrical, the rachis sometimes glandular and the bracts usually persistent, the pistillate flower subtended by secondary, tertiary, or even quarternary bracts forming a calyculus, the ovary either glabrous or pilose and almost completely covered by persistent, more or less globular, wax- secreting papillae. Fruit a nutlet, + spherical and mostly covered by a layer wax, the ovary wall glabrous to densely pilose, the papillae glabrous to puberulent, the associated secondary, tertiary, and even quarternary bracts, if all present, persisting until after fruit maturation and never enlarging. Cerothamnus, the largest subgenus in Myricaceae, contains the waxy- fruited species of Morel/a native to the Americas and Africa and their neighboring islands—i.e. the West Indies and the Atlantic islands lying off the northwestern coast of Africa. KEY TO THE SERIES OF SUBGENUS CEROTHAMNUS 4 . Staminate flowers with 3—7 stamens; staminal column branches each with only 1 anther; bracteoles 0-3; pistils solitary in axil of a bract; fruit wall glabrous or, if pubescent, then the papillae also pubescent .............. Ser. 1. Cerothamnus . Staminate ae with (6-)8—18(-22) stamens; staminal column branches often with 2 anthers; bracteoles 2—6; pistils 1—3 in axil of each bract; fruit wall but not the papillae densely pubescent Ser. 2. Faya — Morella series Cerothamnus (Tidestr.) Wilbur, comb. & stat. nov. Cerophora ae Cerocarpa Raf., Alsogr. Amer. 11. 1838. Lecroryre: Cerophora lanceolata Raf., herein designated [= Myrica cerifera L.} ee sect. Copia Ra) Chev., Mém. Soc. Sci. Nat. Cherbourg 32:223 (= Monogr. L959) 1901. ae Tidestr., Elys. Marian., Ferns. 41. 1910. Lecrorype: Cerothamnus arborescens (Castigl.) Tidestrom (= Myrica cerifera L.) Aments simple; pistillate usually solitary, simple, and 1-flowered. Fruit 1-5 mm in diameter. Papillae wax-producing, neither juicily succulent nor fleshy. In spite of the few species in More//a series Cerothamnus present in eastern North America, there currently exists a surprising amount of uncertainty as to just how few or how many species can be recognized in that well-collected area. The number of species are in any event few: perhaps no more than two and certainly no more than four. 1. sare cerifera a ) — - e U, S. 337 & 1329. 1903. Myrica cerifera Sp. Pl. 1024. 1 iferus (L.) Small, Fl. Miami 61 & 200. (26 Apr — ee WILBUR, Myricaceae 101 2. Asecond taxon has been proposed whose distinctness from Morella cerifera is denied by several of our more experienced students of the southeast- ern flora. The questioned taxon is usually a low bushy, rhizomatose plant with strikingly smaller leaves. It was treated by Michaux (1803) and Radford et al. (1968) asa variety (Myrica cerifera var. pumila Michx.) but as a species by Rafinesque (1838) and Small (1903 and 1933) (Myrica pusilla Raf., Morella pumila (Michx.) Small, and Cerothamnus pumilus (Michx.) Small). Many field observers and collectors of bayberries, are at least partly convinced that Myrica pusilla merits taxonomic recognition for these dwarf plants seem strikingly unlike the much more abundant and luxuriant M. cerifera that often grow in close proximity. In the herbarium on the other hand the claim to specific or even varietal status of M. pusilla seems much less certain perhaps because most specimens are not accompanied by notes as to either the height or habit of the plant; those two features together with the much smaller size of the leaves are what makes these plants in the field so strikingly unlike the ubiquitous M. cerifera. In spite of the apparent dis- tinctiveness noted in the field, it seems impossible to maintain as distinct cerifera and pusilla even in a group as plastic or as bereft of taxonomically useful characters as are the bayberries. Sufficient dissatisfaction with this conclusion exists, however, that I intend and would urge all other botanists with the opportunity to study and collect the plants in the field to do so and to record carefully notes on the height, habit and habitat of the plants— something that surprising has been largely neglected by most collectors even in recent decades. Thieret (1966), who has had extensive field experience with both species in Louisiana, noted that “from aclump that is otherwise typically M. pusilla, there will arise one main stem to ten or twelve feet tall and six inches in diameter—a M. cerifera stem. Examination reveals that all the stems of the clump comprise a clone. ... Plants intermediate in habitat between the two extremes are usually found in the vicinity.” Thieret concluded that “the habit differences ... are not reliable criteria but are simply responses to habitat differences” and “that che habit extremes pass insensibly into each other.” I too have often observed in North Carolina both pusz//a and cerifera growing in close proximity but have not yet noted the blurring of growth form and leaf size that Thieret has noted nor am I convinced that there is always a correlation of habitat and growth form. I have regrettably never tried to excavate the underground connection perhaps implied by Thieret and which, if proven, would provide convincing support for the treatment tentatively accepted here of non-recognition (unless root grafting occurred). Godfrey (1988), whose field experience with our southeastern plants is 102 SIDAILG(1) 1994 surely unrivalled, does not accept the dwarf plants as a taxon deserving a name either. Additional study 1s certainly needed, especially investigations centered in the field. 3. Morella caroliniensis (Mill.) Small, Fl. SE U.S. 337 & 1329. 1903 [as Carolinensis |. Cerothamnus caroliniensts (Mill.) Tidestr., Elys. Marian., Ferns. 41 Myrica heterophylla Raf., Alsogr. Amer. 9. 1838 {as “heterophyla”]: Type Loca.iry: ‘Carolina to Florida.” Cerothamnus heterophylla (Raf.) Moldenke, Phytologia 29:386. L975 4. Myrica pensylvanica Mirbel, Traité Arbr. Arbust. 2:190. 1804, [not Loisel. as poe! cited! }. on een pensylvanicus (Mirbel) etecake. Rev. Sudam. Bot. 4:16. 1937. Although my study is not completed, Iam doubtful that the above two species can be maintained as distinct. The alleged principal difference between the two is apparently to be found in the fruits: fruit of more northern species (Myrica pensylvanica) possess puberulence on both the papillae and ovary wall; fruit of the southern species (usually referred to as Myrica heterophylla Raf.) have glabrous papillae and glabrous walls. Due to the dense covering of wax, detection of the puberulence is often difficult and the amount varies from a dense covering to very few trichomes. The other alleged differences of fruit size, twig pubescence, leaf persistence, etc. seem even less consistently diagnostic. If only one species is to be recognized, the correct binomial is Morefla caroliniensis (Mill.) Small, a binomial that has been applied to the combined species in the past and to both of the species at different times when they were treated as distinct species. Miller’s name should be applied to the species with the more southern distribution if two species are represented as most authors have accepted for the past 75 years. 5. Morella tnodora SS Bartram) Small, Fl. SE U.S. 337 & 1329. 1903. Myrica inodora WW. Bartram, Travels Carolina, Ba 1791. Cerothamnus tnodorus (W. Bartram) Small, F an Trees 12 & 102. 191- In striking contrast to most other Myricaceae, but as indicated by its specific epithet, the crushed foliage of More/la inodora is not aromatic. The species ranges from southern Georgia westward into southeastern Louisiana. 6. Morella californica (Cham. & Schltdl.) Wilbur, comb. nov. Myrica californica Cham. & Schledl., Linnaea 6:535. 1835. Gale californica (Cham. & Schltdl.) eene, Man. Bot. San Francisco. 298. 1894. The natural range of this species apparently extends from central coastal Washington south into Los Angeles County, California at elevation of 150 m or less. WILBUR, Myricaceae 103 MORELLA series FAYA (P. Webb & Berthel.) Wilbur, comb. & stat. nov. Fayana Raf., Alsogr. Amer. 12. 1838. Type: Fayana azorica Raf. (= Myrica faya Aiton). Faya P. Webb & Berthel., Hist. Nat. Iles Canaries 3:272, t. 216. 1847. Type: Faya ae P. Webb & Berthel. (= ee jee Aiton). Myrica sect. Faya (P. Webb & el.) C. DC., Prodr. 16(2):151. Aments simple or branched; pistillate bracts subtending several ovaries of which only some usually develop. Fruits usually 4-6 mm in diameter, often forming a syncarpium. Papillae often producing wax but never fleshy. Series Faya is a small taxon of three geographically widely separated species. Besides the two North American representatives treated below, the section is composed of the type species of the section, Morella faya' of the Canaries, Madeira, and the Azores and possibly also of Portugal where it occurs but perhaps only as a naturalized introduction Burges (1964). MacDonald (1977, p. 2638), who has presented a series of papers on the morphology of the inflorescence of many myricaceous taxa, is of the opinion that “section Faya could easily be incorporated in section Cerophora” of the genus Myrica. This genus in his opinion would then consist of the fleshy- fruited section More//a of eastern Asia and Indonesia and section Cerophora, the largest taxon in the family, with both American and southern African representatives. The suspicion lingers that series Faya is not a proven monophyletic group not only because MacDonald questioned its morpho- logical distinctiveness but also because its distribution pattern is not one readily explained or matched by other examples. In contrast to the uncertainty existing in both the identities and names of the taxa comprising Morella series Cerothamnus, the taxa forming series Faya are morphologially most distinct and geographically widely separated. 2. MYRICA L. Myrica L., Sp. Pl. 1024. 1753; Gen. Pl. ed. 5.449. 1754. Lecrorype: Myrica gale L. seep Fl. Belg. 12. 1827. [Neither Ga/e Duhamel, (Traité Arbr. Arbust. 1:253. 5 lanson., (Fam. Pl. 1763.) Both Duhamel’s and Adanson’s reintroduc- ae of el were ‘lesieinn ate names as substitutions for Myrica L. and hence 1. pee superfluous (Art. 63 ICBN) Myrica “b” Gale {Tourn.] Endl., Gen. Pl. 272.1837. [Employed in the sense of the Myricaceae ie Comptonia.} Cerophora subg. Galestis Raf., Alsogr. Amer. 11. 1838. Lecrorype: Cerophora (Galestis) angustifolia Raf. (= Myrica wn * Angeia Tidestr., Elys. Marian., Ferns 37. 1910. Type: Angeia palustris (Lam.) Tidestr. (= Myrica palustris Lam., = My rica gale L.) Ba Wilbur, comb. nov. BASIONYM: Myrica faya Aiton, Hort. Kew. 3:397. 104 SIDA16(1) 1994 Aromatic, usually dioecious shrubs lacking terminal buds. Leaves alternate, pinnately veined, serrate to entire, deciduous, estipulate. Inflorescences borne distally, the staminate suberect to recurved, in bud ovoid to elliptical and at anthesis cylindrical, the bracts broadly ovate to triangular, at anthesis longer than the 3—G6 stamens, secondary bracts absent; pistillate inflores- cences simple, broadly cylindric at anthesis, the bracts persistent, the pis- tillate flowers each subtended by 2 secondary bracts partially adnate to the ovary wall, the ovary glabrous and lacking papillae. Nutlet flattened, keeled, not covered with wax but with few to many glandular trichomes, the brac- teoles persistent, becoming greatly enlarged and inflated, strongly adherent to the flattened fruit, inflated and forming a buoyant float. Myrica is represented in the Americas by two species that are morphologi- cally readily distinguished from one another and whose geographical ranges do not overlap. 1. Myrica gale L. This species has a broken circumboreal distribution pattern extending across Canada and Alaska and south to northern New Jersey, Pennsylvania, eastern Ohio, and about the Great Lakes from as far west as eastern Minnesota and in the west to the mountains of Oregon. Morphological variation apparently correlated with distribution occurs within the exten- sive range of this species but its taxonomic merit requires additional study. Hultén (1944 and 1968) presented a brief overview of the problem. 2. Myrica hartwegii S. Watson. This species is found along stream banks in yellow pine forests in the Californian Sierra between 300-1500 m and consequently not sharing any part of its range with any other member of the Myricaceae. 3, COMPTONIA LHer. ex Aiton Comptonia L'Hér. ex Aiton, Hort. Kew. 3:334. 1789. Tyee: Comptonia asplenifolia (L.) LHér. ex Aiton (= Comptonia peregrina (L. 7 M. Coulter). Myrica “c” Comptonia (L Her. ex Aiton) Endl., Gen. Pl. 2 837 Myrica sect. Comptonia (LHér. ex Aiton) Endl. ex C. oe ee 16(2): 151. 1864, Myrica subg. Comptonia (L Hér. ex Aiton) Engler, Nat. PAaareatim: III. 1:28. 1893. Aromatic, usually dioecious, colonial shrubs 1.5 m tall or less and lacking terminal buds. Leaves alternate, pinnately veined and lobed, deciduous, conspicuously semi-cordately stipulate. Inflorescences borne distally, the staminate suberect to strongly recurved, cylindrical, the primary bracts persistent, broadly ovate to quadrangular, at anthesis longer than the 3—7 stamens, adaxally bearing numerous glandular trichomes, secondary bracts WILBUR, Myricaceae 105 absent; pistillate inflorescences simple, broadly ovoid at anthesis, the rachis pubescent and glandular, the bracts persistent but in fruit obscured by the greatly enlarged lobes of the secondary bracts, the pistillate flower sub- tended by the 2 greatly enlarged and deeply lobed secondary bracts, the ovary glabrous and lacking papillae. Nutlet conical to cylindrical, non- ceriferous, 2.2-5.5 mm long, glabrous, surrounded by the enlarged lobes of the secondary bracts, which form a bur-like structure in fruit. The genus Comptonia is both monotypic and endemic to eastern North America. ki ee peregrina (L.) J.M. Coulter, Mem. Torrey Bot. Club 5:127. 1894. Liguidambar peregrina L., Sp. Pl. 999. 1753. Myrica peregrina (L.) Kuntze, : 638. 1891. Revis. Gen. Pl. Myrica asplenifolia L., Sp. Pl. 1024. 1753. Liquidambar asplenifolia (L.) C.F. Ludwig, Neuere wilde Baum 27.1783. Comptonia asplenifolia (L.) L Hér. ex Aiton, Hort. Kew 3:334. 1789. Comptonia peregrina vat. golesila (L. ) Fernald, Rhodora 40: 410. 1938. The variation in vegetative pubescence suggested by Fernald as a diagnos- tic feature distinguishing var. peregrina from var. asplenifolia (L.) Fernald does not delimit populations as sharply as Fernald’s account implied. There seems to be more continuous variation in pubescence than Fernald reported and the geographic range of the two pubescence types is less discrete than suggested. Well-collected areas of the mid-Atlantic States have many examples of both pubescence types within their borders. Examples of different pubescence types on different branches of the same shrub are not unusual. It surely can be con- cluded that varietal status within the monotypic genus Comptonia based on pubescence has not been convincingly proven. My rapid survey of hundreds of specimens does not suggest that a detailed, careful analysis of pubescence would likely demonstrate the presence of geographically based varieties. As shown by the above synonymy, Linnaeus treated this species twice in Species Plantarum. The species was these first included (p. 999) as Liguidam- bar peregrina and again (p. 1024) as Myrica asplenifolia. For nearly a century and a half the name most frequently adopted was Myrica asplenifolia or its derivative Comptonia asplenifolia. For the past century (except for Gleason (1952), Gleason & Cronquist (1963), and Wagner et al. (1990), the accepted name has been almost universally that based upon Liguidambar peregrina. The latter choice is mandated by the International Code not because of the American Code’s favoring of the name appearing first in a volume (“page priority”) but because Linnaeus, upon discovering the conspecificity of the two binomials, was the first to unite them (Syst. Nat. ed. 10. 2: 1273. 1759.) by placing Myrica asplenifolia in the synonym of Liquidambar peregrina. 106 SIDALG(1) 1994 REFERENCES Abbe, E. 1972. The inflorescence and flower in male Myrica esculenta var. farqubariana. Bot. az. (Crawfordsville). 133:206—213. eR 968. A taxonomic revision of the plant family Myricaceae of North America, north of Mexico. Unpublished thesis, Univ. of North Carolina, Chapel Hill. BRANDEGEE, K. 1901. Some sources of error in genera and species. Zoe 5:91—98. Burces, N.A. 1964. Myricaceae in T.G. Tutin et al. Flora Cee 1296, Cuevatier, A. 1901. Monographie des Myricacées; anatomie et histologie, organographie, classification, et description des espéces, distribution géographique. Mém. Soc. Sci. Nat Cl | 32:85-340. Enias, T.S. 1971. The — of the Myricaceae in the southeastern United States. J. Arnold ‘vee 52:305-31 ENGLER, A. 1894. mee Nat. Pflanzenfam. II]. 1:26—28. ee ee b, M.L. 1950. Gray’s manual of botany. 8th ed.: American Book Company, New k. [Myricaceae, 523-525.]} 3LEASON, H.A. 1952. The new Britton and Brown illustrated flora of the northeastern United States and adjacent Canada. 3 vol. New York: New York Botanical Garden, New York. [Myricaceae, 2:24—25 GLEASON, H.A. and A. CRonquist. 1963. Manual of Vascular plants of northeastern United States and adjacent Canada. Van Nostrand Co., Inc. [Myricaceae, 240-241 GLEASON, H.A. and A. Cronquist. 1991. Manual af vascular pints of roinenscens United oe 2nd edition. New York Bots Bronx. [Myricaceae, 81) o.. yo Rak, 1982; Trees, shrubs and woody vines of northern Florida and adjacent Georgia and Alabama. Univ. of Georgia Press. Athen ear: 483-489. ] GREENE, E.L. 1910. ee of the bayberries. Leafl. Bor. Observ. Crit. 2:37—40. Greene, E.L. 1910a. A new name for the bayberries. Leafl. < Observ. Crit. 2:101—104. Hircucock, A.S. & M.L. Green. 1929. Standard-species of Linnean genera of Phanerogamae (1753—54.) International Bot. Congress, Cz ae 1930 in Nomencl. Prop. Brit. Bot. 110-199. His Majesty’s Stationery London Hutreén, E. 1944. Flora of Alaska and the Yukon. foe Univ. Lund. Avd. 2.40: 572 {Myrica]. a 1968. Flore o> ae neighboring territories. Stanford University Press, Stanford, CA. Male p. ; Hien , J. 1964. The genera ‘er flowering plants. Dicotyledones. Oxford Univ. Press. London [Myricaceae, 2:120—121. 1967.] Hytanber, N. 1945. Nomenklatorische at ae ane Studien tiber Nordische Gefiasspflanzen. Uppsala Univ. Arsskrift. 337. [Myrica, 40.} INGgR, R.F. 1958. Comments on “he eee a ee Evolution 12:370-384. INTERNATIONAL RULES BOTANICAL NOMENCLATURE (ICBN). 1935. 3rd ed. xi & 151. Gustav Fischer, Jena. INTERNATIONAL Cope OF BoraNical NOMENCLATURE (ICBN). 1978. “Leningrad Code.” Regnum Veg. 97:xiv, 1-457. Bohn, Scheltema & Holkema. Utrecht ——*'*O9SB. “Berlin Code.” Regnum Veg. 118:xiv & 328. Koeltz Scientific Books. KG6nigstein, Germany. — DONALD, A.D. 1977. Myricaceae: floral hypothesis for Gale and Comptonia. Canad. J. ot. 535:2636-2651. a 1978: a a aa fl 1 fl f Myrica esculent. Canad. J. Bot. 56:2415- WILBUR, Myricaceae 107 Micuaux, A. 1803. Flora Boreali-americana ... Paris. [Myrica 2:227-228. NiguWLAnD, J.A. 1910. The name of our American wax bayberries. Amer. Midl. Naturalist 1:238-243. OerrsteD, A.S. 1867. Recherches sur la classification des chénes. Bianco Luno-F.S. Muhle, Copenhague. Raprorb, A.E., H.E. AHLEs and C.R. Bet. 1968. Manual of the vascular flora of the Carolinas. awe North Carolina Press, Chapel Hill. [Myricaceae, 360—361.] RAFINESQUE, C.S. 1838. Alsographia americana. Philadelphia. [Myricaceae, 8-12 Reuper, A. 1949. Bibliography of cultivated trees and shrubs. Arnold Arbor. eee Univ., Jamaica Plain, MA [Myricaceae, 88.] Renbie, A.B. 1903. Notes on ee 1 Bot. 41:82—-87. ScHWARZ, O. 1936. Entwurf zu einen Natiirlichen System der Cupuliferen und der Gattung Quercus L. Notizbl. Bot. Gart. eee 13:1-22. SMALL, ae 1903. Flora of the southeastern United States. New York. [Myricaceae, 336—338.] «1933. Manual of the southeastern flora. New York. [Myricaceae, 408-410. ] StTIRTON, C.H., D.V. Fretp, R.K. BRuMMitt and J. MCNEILL. 1981. Proposals on stability of names. Taxon 30:250-256. THIERET, J.W. 1966. Habit variation in Myrica pensylvanica and M. cerifera. Castanea 31:183-185 THORNE, R.F. 1973. The‘ semennlerar or Hamamelidae as an artificial group: a summary statement. ne 25:395-405. Wacner, W.L., D.R. Hersstr and S.H. Soumer. 1990. Manual of the flowering plants of Hawaii. 2 vol. Univ. Hawaii Press. [Myricaceae 1: 929-931.} 108 SIDA16(1) 1994 BOOK REVIEW RaMaAMoortnHy, T.P., R. Bye, A. Lorr and J. Fa (Eds). 1993. Biological Diversity of Mexico: Origins and Distribution. (ISBN 0-19-506674- X, hbk) Oxford University Press, New York. $79.95. 812 pp. As stated by the editors in this impressive volume’s introduction, the primary objective of this contribution is to assemble data pertaining to a wide variety of Mexican taxa and to make it available to Mexicans and the world community alike. In achieving this admirable and ambitious objective, the book is by and large successful. While many researchers are likely to find one to disagree with in the chapters relevant to their particular area of oF they are at the same time likely to rely on the volume as an invaluable source of and bibliographic aloe on the subject of ee ae dtl Although h h largest country in the wold, it is asually inked within the top three or four countries in the world in terms of the richness of its biological diversity. This volume does a very good job at explaining, in general and taxon by taxon, how biogeography, geology, and eras history have combined to bring about Mexico’s fantastic species richness. If the book does nothing more than draw well-deserved conservation attention to Mexico’s arid and montane ee ic will have performed a great service The book’s 26 chapters are organized into six sections. The first section’s three chapters focus on historical background, with chapters on geology and the origins and diversity of Mexico's flowering plant flora. The second section reviews six selected faunistic groups including: bees, butterflies, fishes, in seer and mammals. A chapter on ecological diversity in scrub jays, which is inclu seems out of place. The third section reviews 11 floristic groups including eee bryophytes, grasses, and legumes. The two chapters covering the genus Pinus and Quercus are particularly outstanding. The fourth section contains two chapters analyzing a a a Sa patterns in one oe tems: tropical rainforests and alpine habitats. T] with ethnobotanical themes, which eu unlikely to contain much infounacion chacds 1S new to those familiar with the field, provide a compilation of bibliographic references that is very useful. The sixth and final section contains an overview of Mexican biodiversity in terrestrial habitats. } 2 : kl ] ] 1 = 1. } WV hich for the present elas was held j in 1988. Other chapters have been recently dpa ec: and, in general, the lack of bibliographic references from the early ’90s is a minor drawbac one hae is probebly uneyoldanle ina volume : this Scope ae are translation ee: 1d turns of f phrase. One chapter in which eee translation seems particularly weak is ‘hae by Hernandez-X. on plant domestication in Mexico. of the chapters contain conservation recommendations at the end which although skeletal, is in some cases (e.g., the chapter on Pinus) make specific important recommen- dations. However, it is noc this volume’s place or role to produce comprehensive conserva- tion anunareas ee, but ho sear to pro bee the development of these by other Indeed, the bo nis to all the scholars of biodiversity, in hopes oe this ie will assist in the “endeavor to pr ae a world-wide conservation agenda.” I think that it will. Bs aie M. Syminton, Biodiversity Support Program, clo World Wildlife Fund, Washing- ton, D.C. 20 FLORA VASCULAR DE LA SIERRA DE LA PAILA, COAHUILA, MEXICO JOSE A. VILLARREAL Q. Departamento de Boténica Universidad Auténoma Agraria “Antonio Narro” Buenavista, Saltillo Coahuila 25315, MEXICO RESUMEN es seco. Se eleva desde ie 1200 m Bae los 2350 m. Las rocas le la eee son fundamentalmente calizas y los suelos litosoles. No presenta escurrimentos permanentes de agua. La vegetacion es clasificada en cuatro tipos: El Matorral peste Chihuahuense, Matorral Submontano, Bosque de Encino-pino y Zacatal. La flora esta compuesta por aproximadamente 87 familias con 387 géneros, 703 especies, con 712 taxa incluyendoa los taxa infraespecificos. Las familias mds diversas son Asteraceae, Poaceae y Fabaceae, los géneros Muhlenbergia, Dalea, Notholaena y Polygala presentan el mayor ntimero de especies. Es la localidad tipo de 21 especies y 9 de ellas son endémicas. ABSTRACT The Sierra de la Paila located in the southern portion of the mexican state of Coahuila, is an element of the discontinuous part of the Sierra Madre Oriental. The sierra ranges from the 1200 m to 2350 m in elevation, and the climate is dry. Limestones are the pencpa. substrate and lithosoles the principal soils. TI water f The vegetation is divided in oni types: Chihuahuan Decor Scrub, Submontane Scrub, Oak— pine Woodland and Grassland. The flora comprises approximately 87 families with 387 genera, 703 species, 712 taxa with the inclusion of the infraspecific taxa. The most diverse families are PSESEACEAS Poaceae and Fabaceae. The genera Muhlenbergia, Dalea, Notholaena and Polygal. her in number of species. It is the type locality of 21 species and about 9 of deena are ila. INTRODUCCION La sierra de la Paila es una formaci6n montafiosa aislada que forma parte del complejo de la Sierra Madre Oriental. Se localiza en el limite este del Desierto Chiuahuense y es un mosaico de asociaciones vegetales que cambian en cada uno de sus flancos. Por ser un area poco estudiada botanicamente y con el proposito de contribuir al conocimiento de la flora del noreste de México se presenta el siguente trabajo que aunque exhaustivo no pretende ser completo. SIDA 16(1): 109 — 138. 1994 110 SIDALG(1) 1994 DESCRIPCION GENERAL DEL AREA La Sierra de la Paila es una formaci6n montafiosa démica aislada, ubicada enelsureste del estado de Coahuila (Fig. 1). El area ocupada por la sierra tiene una forma rombica que se localiza entre los 101°25'y 101°49'W y los 25°45! y 26°16'N. Tiene aproximadamente unos 60 km de extensidén en su eje mayor (norte-sur) y unos 35 km en su eje menor, con una superficie de unos 1700 km2. Se eleva desde los 1200 m alcanzando las partes mas altas los 2350 m de altura. Esta surcada por una serie de cafiones que van del centro a la periferia. En la parte central se localizan valles altos y pequefios. Varias vias de comunicacién bordean la sierra totalmente. Por el sur pasa la carretera 40 y la linea de ferrocarril Saltillo-Torreén. Por el noreste pasa lacarretera 57 y el resto de la circunferencia los forman caminos de terraceria. El poblado de Hipolito, ubicado en el extremo sur, es el mas importante y proximo a la sierra. El ejido Las Coloradas se localiza en los valles cercanos al extremo norte. Los ejidos de El Cedral y Parrefios se encuentran dentro de la sierra (Fig. 2). De acuerdo a lo presentado por el INEGI (1983), los climas seco y semiseco (BS°,BS?) dominan en el area. Estos climas se caracterizan por presentar lluvias predominantes en el verano y temperaturas altas. La precipitacion promedio es de 400 mm anuales, y la temperatura media anual es de 20°C, Las rocas dominantes en el area son las calizas, en su mayoria del Cretacico. Los suelos predominantes son litosoles, arcillosos y ricos en materia organica, de color pardo oscuro 0 negro. Los valles intermontanos de la sierra presentan xerosoles, y las partes bajas y valles que rodean la sierra presentan suelos aluviales tipo castafiozem y ferozem (INEGI 1983). En la sierra no se presentan correientes permanentes ni depositos de agua, solo arroyos en los cafiones que drenan el agua durante la epoca de Iluvias. VEGETACION Aunque la colecta de muestras vegetales en la Sierra de la Paila se realizaron desde principios de siglo (Sousa 1969), la informacién sobre la vegetacion del area es escasa. En trabajos tradicionales como los de Muller (1947), pasa desapercibida. Solo recientemente se ha aportado informaci6n por Cano y Marroquin (1967) y Wehbe (1985). La vegetacion en el area de la Sierra de la Paila puede ser clasificada en cuatro tipos prnicipales, que de acuerdo con la revision sobre vegetacion de Coahuila hecha por Villarreal y Vladés (1992-1993) correspeonden a: Matorral Desértico Chihuahuense, Matorral Submontano, Bosque de Encino- pino y Zacatal. — VILLARREAL Q., Flora vascular de la Sierra de la Paila 11] 1032 57' 1032 29257' a 292 28° 272 262 252 24232! 1. Sierra del Carmen 7. Sierra San Marcos y San Antonio 2. Serranfa del Burro 8. Sierra La Paila Sierra Santa Rosa 9. Sierra de Arteaga 4. Sierra del Pino 10. Sierra de Parras 5. Sierra del Fuste ll. Sierra de Jimulco 6. Sierra de La Madera FIG. 1. Localizacién de la Sierra de la Paila y otros sierras en al Estado de Coahuila. Matorral Desertico Chihuahuense. Es el tipo de vegetaci6n mas comtn en toda el drea, circundando toda la sierra. Los valles que bordean la sierra y lomerios con suelos algo profundos y poco pedregosos son ocupados por matorrales microfilos de arbustos de 30-150 cm de alto. La especie mas frecuente es Larrea tridentata, asociada con Flourensia cernua, Fouquieria splendens, Partheninm incanum, Yucca filifera, Y. treculeana, Opuntia imbricata y Prosopis glandulosa. En algunos sitios de la porci6n oeste Sericodes greggii es frecuente y Grusonia bradtiana abunda en las porciones norte. Las laderas de las pociones mas externas de la sierra, asi como los lados mas expuestos de muchos cafiones con suelos pedregosos y someros se desarrollan 112 SIDA1L6(1) 1994 T T 101°40' 101225' A Monclova \ \ N - 26°15' ~ Cafién El Diente ae @Plan de Guadalupe Cafién E] Carmen Ejido Cedral Ejido Pa ‘ t 10 Km. H t te “a? - 25°45! / ———————— H A Saltillo A Torreén wd I A Saltillo FIG. 2. Mapa de la Sierra de la Paila, ubicacién y algunas localidades. matorrales bajos con comunidades de Agave lechuguilla, asociada frecuentemente con Parthenium argentatum, Euphorbia antisyphilitica y con presenciade A gave striata, Dasylirion palmer, Hechtia texensis, Opuntia microdasys, O. phaeacantha, Yucca carnerosana y Gochnatia hypolenca. Matorral Submontano. Este tipo de vegetacién tiene distribucién irregular, se localiza en los cafiones y usualmente se mezcla con porciones de bosque en el centro de la sierra. Esta formado por arbustos y pequefios arboles de 2-5 m que se desarrollan en comunidades densas en los lechos de los arroyos y areas protegidas con suelos profundos y algo de humedad. Se presentan frecuentemente Quercus invaginata, Q. intricata, Fraxinus greggit, Vauquelinia corymbosa, Cercocarpus mojadensis, Pistacia texana, Acacia berland- VILLARREAL Q., Flora Vascular de la Sierra de la Paila 113 ieri, Rhus virens, Flourensia retinophylla, Berberis trifoliolata, Ptelea trifoliata y Leucaena greggii. Brahea berlandieri crece en las laderas de los cafiones. En los lechos de los arroyos es frecuente Dodonaea viscosa. Bosque de Encino-Pino. Se localiza en la porcién central de la sierra, ocupa valles y cafones amplios con comunidades arboreas poco densas do- minadas por Quercus gravesi1, Q. lacey1, Juniperus flaccida, con elementos espar cidos de Pinus arizonica o P. cembroides. Otras especies frecuentes incluyen a Juniperus erythrocarpa, Arbutus xalapensis, Garrya ovata, Ungnadia speciosa, Fraxinus cuspidata y elementos de matorral submontano. Préximo al bosque es frecuente encontrar vegetaci6n arbustiva de tipo chaparral, que usualmente se mezcla con las especies arboreas del bosque. En estas comunidades esta presente Quercus intricata, Q. invaginata, Q. grisea y Rhus virens. En los arroyos es frecuente_ Juglans microcarpa. Las especies arboreas usualmente presentan especies de Ti//andsia y Phoradendron. Zacatal. En areas reducidas de los valles centrales elevados de la sierra con suelos profundos se Te el desarrollo de comunidades de gramineas. En ] dichos zac Sporobolus airoides, Nassela tenuissima, Stipa eminens y en otras areas Bouteloua gracilis y B. curtipendula, frecuentemente mezcaldas con elementos de bosque. FLORA La flora vascular documentada de la Sierra de la Paila comprende 703 especies (712 taxa incluyendo a los taxa infrespecfficos) distribuidas en unos 387 géneros y unas 87 familias. Es la localidad tipo de unas 21 especies, de las cuales 9 son endémicas (ver tablas 1 y 3). Las familias y géneros con mayor diversidad se muestran en la tabla 2. Las primeras colectas parecen haber sido hechas por Endlich en 1905 y C. A. Purpus en 1910 y 1911, cuyo material sirvi6 para la descripcion de nuevas especies. En los ultimos afios una docena de colectores ha trabajado en la sierra, como se presenta en la lista de abreviaciones. Los herbarios revisados para la elaboracién de la lista florfstica fueron el de la Universidad Atoméma Agraria Antonio Narro (ANSM) y el de la Universidad de Texas (TEX-LL). El arreglo que se sigue para la lista de plantas vasculares en los grupos principales es el de Cronquist (1981) para angiospermas y el de Crabbe et al. (1975) para helechos y grupos afines a helechos. Las categorias de familia, género y especie se presentan en secuencia alfabética. Cada nombre cientifico va seguido por el tipo de vegetacién donde frecuentemente se encuentra, luego los datos sobre el habitat y frecuencia. Al final la abreaviacién de los colectores es seguida de los nimeros de colecta. 114 SIDA16(1) 1994 Tasia |. Principales grupos componentes de la flora vascular. Taxa infraespecificos Division Familias Géneros Especies adicionales Lycopodiophyta l l 3 0 Polypodiophyta 2 4 14 0 Pinophyta 3 3 6 0 Magnoliophyta Clase Magnoliopsida 71 315 556 8 Clase Liliopsida 10 64 124 1 Torales 87 387 703 9 Tabla 2. Familias y géneros con mayor diversidad. Familia Géneros Especies Géneros Especies Asteraceae 75 125 Muhlenbergia 12 Poaceae 4l 91 Dalea 11 Fabaceae 26 54 Notholaena y Polygala 9 TABLA 3. Lista de especies tipo y endémicas de la Sierra de la Paila. ACANTHAC LAMIACEAE Carlow ae parvifolia Brandg. Hedeoma montanum Brandg. AGAVACEAE LILIACEA Yucca endlichiana Trel. +Muilla am ‘pusit Brandg. ASTERACEAL MALPIGHIA * Haplo Ein setosa Brandg. es ies Peulibita Turner AL, a S ea Flourensia retino, Stevia salicife CACTACEAE aie lani Bremer ee delaetii dun ta Giierke *Gymno S Ag wus Glass & Foster CONVOLVULACEAE *Ibomea zimmermanit McDonald FABACEAE Senna monozyx (Irwin & Barneby) Irwin FAGAC one invaginata Trel. *Especies endémicas ke C r integra (Blake) Rob. ~ Le ie (Blake) Nesom POACE Bittle johnstonii Swallen POLYGALACEAE Polygala nudata Brandg. RHAMNACEAE aes standleyana C. B. Wolf RUBIAC *Cont ene p ailensis Villarreal RUTACEAE *Thamnosma pailensis M. C. Johnst. CecrRopUTT ARTACEAF *Penstemon punctatus Brandg. VILLARREAL Q., Flora Vascular de la Sierra de la Paila 115 ABREVIACIONES EN LA LISTA FLORISTICA Tipos de vegetacion: MD, Matorral desértico chihuahuense. MS, Matorral submontano. B, Bosque de encino-pino. P, Zacatal. Principales colectores y herbarios donde estan depositados sus colecciones: C, M. A. Carranza P., (ANSM). Ca, G. Cano y J. S$. Marroquin (UNL). CQ, D. Castillo Q., (ANSM). Endlich, R. Endlich (MO). H, J. Henrickson (UC, TEX-LL). Hin, G. B. Hinton (TEX-LL). JM, J. S. Marroquin (ANSM). JW, J. Wehbe (ANSM). MCJ, M. C. Johnston, F. Chiang y T. Wendt (TEX-LL). Purp, F. A. Purpus (UC,EUS). R, A. Rodriguez G. (ANSM). VR, J. Valdés Reyna(ANSM)y V, J. A. Villarreal Q. (ANSM CATALOGO DE LAS PLANTAS VASCULARES DE LA SIERRA DE LA PAILA, COAHUILA, MEXICO LYCOPODIOPHYTA (Selaginelas) SELAGINELLACEAE Selaginella lepidophylla (Hook. & Grev.) Spring. MS. Paredes rocosas. Comtin. V3103, 3558. Selaginella pilifera A.Br. MS, MD. Paredes rocosas. Escasa. V3094, 3559, 5312. Selaginella wrightii Hieron. MD, MS. Laderas rocosas. Comtin. V3095, 3533, 4481, 4783. POLY PODIOPHYTA (Helechos) ADIANTACEAE Cheilanthes alabamensis (Buckl.) Kuntze. B. Cafiones y arroyos. Escasa. V3524, 4729. Cheilanthes notholaenoides (Desv.) Maxon. B, MS. Cafiones y arroyos. Comtin. V3101, 3978, 4479, 4730, 5288, VR2233. Cheilanthes pinkavae Reeves. MD. Laderas rocosas. Escasa. V3096, 48 Notholaena = KI. MS, MD, B. Laderas rocosas. Comin. ioe CQ875, V3543, 4728. Notholaena ee Maxon. MD, MS. Laderas y arroyos. Comtin. H16543, MCJ10099, 10109, 10511, R1486 Notholaena cochisensis Goodd. MD. Laderas rocosas. Comin. H16512, CQ783, V4727. Notholaena greggit (Mett.) Maxon. MD. Laderas rocosas. Rara. MCJ10520, W5562. Nicholatie eee eae ) Hevly. MD. Laderas rocosas. Escasa. CQ782, 787, R1481. on. MD. Laderas rocosas. Escasa. MCJ11711le, R794, V5201, 5259, eg ] 665 59. Notholaena parvifolia Tryon. MD, MS, B. Cafiones y arroyos. Comtin. MCJ10109b, CQ660, V3098, 3542, 4472 Notholaena sinuata an: ) Kaulf. MD, MS. Laderas rocosas. a V3097, 3591. Notholaena standleyi Maxon. MS. Laderas rocosas. Escasa. V448( 7 as (L.) Link. B, MS. Cafiones y arroyos. aan V 3923; 927 71,6060, 2197 ASPLENIACEAE Asplenium resiliens Kuntze. MS, B. Laderas rocosas. Rara. MCJ11697a. 116 SIDA16(1) 1994 PINOPHYTA (Gimnospermas) CUPRESSACEAE Juniperus erythrocarpa Cory. B. Valles y laderas. Comtin. V3937, 3956, 4176, 5830, VR2216. Juniperus flaccida Schlecht. var. flaccida. B, MS. Valles y laderas. Comin. V3561. EPHEDRACEAE Ephedra antisyphilitica Mey. MD. Laderas. Escasa. JM2327, MCJ10512, 10519, V2892, 2 Whe peace ee Ephedra aspera Wats. MD. Laderas. Escasa. JM320, C462, V3873, 5572 PINACEAE Pinus artzonica Engelm. B, MS. Valles y laderas. Comtin. V3620a, 3983. Pinus cembroides Zucc. B, P. Valles y laderas. Escasa. V3082 MAGNOLIOPHYTA (Angiospermas) MAGNOLIOPSIDA (Dicotiledoneas) ACANTHACEAE Anisacanthus linearis (Hagen) Henrick. & Lott. MS, MD. Arroyos y cafiones. Comin. JM2220, MCJ10095, V3312, 3540, 4686. Carlowrightia mexicana Henrick. & Daniel. MS. Areas protegidas. Escasa. V4642. .P 751 Carlowrightia parvifolia Brandg, MD. Arroyos. Comin. Purp4751, V5436. Carlowrightia aes (Torr.) Gray. ioe eee y laderas. Escasa. V4541, 5564 Dyschoriste linearis (T.& G.) Kuntze. MS, B, MD, P. Comtin. CQ710, V3904, 5252. Ruellia ae Gray. MD. hae y | ee Muy comtn. C0635, MCJ11690a, H16183, V4A498 ee pilosella (nes) Torr. MD, MS. Valles y laderas. Escasa. V150 ACERACEAE Acer grandidentatum Nutt. B, MS. Cafones, muy localizado. Escasa. R1328, V3550, 3886, 5262 AMARANTHACEAE Alternanthera repens (L.) Kuntze. B. Valles. Comun. V5829, VR2206. Amaranthus hybridus L. B. Valles y areas perurbadas. Dicraurus leptocaudus Hook. MD. Valles. Escasa. CQ806. Tresine heterophylla Stand. B, P. Valles. Escasa. VR 2243 Froelichia arizonica Thornb. MS, MD. Reaves Comun. HI 6187, V3902, 453 Froelichia interrupta (L.) Mog. MD. Arroyos. Rara. V5305. Tidestromia lanuginosa (Nutt.) Standl. MD, P. Areas perturbadas. peed ee aes ) Standl. var. — I.M.Johnsc, ay ‘lade ras. Muy comun. 17p4927, H16195, V2886, 4179, - 683. NACARDIACEAE Pistacia texana Swingle. MS, B. Arroyos y canadas. Muy comun. JM2222, MCJ11718, V3069, 6684. Rhus microphylla Engelm. MD, MS. Valles y lomerfos. Comtin. JM316, 997, V3562, 5218. Rhus muelleri Stand] & Barkl. MS, B. Arroyos. Escasa. MCJ11683, V3942, 4787, 5263, Rhus trilobata Nutt. vat. trilobata. MS, B. Araeas protegidas. Comtin. CQ849, MCJ11692a, V3261, 3949. VILLARREAL Q., Flora Vascular de la Sierra de la Paila 117 Rhus virens Lind. var. virens. B, MS. Arroyos y cafiones. Muy comtin. C466, JM757, 2235, MCJ11702b, V3545, 3970. APOCYNACEAE Amsonta grandiflora Alex. MS, B. Arroyos y orilla de caminos. Comin. R1338, V3520,4761, 5232, VR2212 Macrosiphonia macrosiphon (Torr.) Heller. MD. Valles y lomerios. Escasa. JM957. Mandevilla karwinskii (Muell. Arg.) Hemsl. MS. Laderas. Comin. MCJ11680, V3593, 4768, 5268 ASCLEPIADACEAE Asclepias brachystephana Torr. MD, MS. Maleza de dreas perturbadas. Escasa. V5607. Asclepias linaria Cav. MS, B. Arroyos. Comin. CQ665, HI6174, V4148 Asclepias texana Heller. MS, B. Comin. Mea 5254, VR2200 ee oP B, Laderas. Escasa. W5254¢ ‘(Dene.) Stand!. MD. ce agbastes bajos. Comin. MCJ11718a, V3538, OGL SE a (Gray) Henrick. MD. Sobre arbustos bajos. Comin. V3115, 4175, 4489, ae ae (Gray) Woods. MS. Valles y cafiones. Escasa. W6675. Sarcostema cynanchoides Dene. MD. Arroyos, rastrera. Escasa. W6678 ASTERACEAE (COMPOSITAE) Acourtia nana (Gray) Reveal & King. MR, P. Valles. Escasa. V4181. Acourtia parryi (Gray) Reveal & King. MD. Valles y lomerios. Escasa. JM2998, V4163. Acourtia runcinata (D.Don) Turner. MD. Laderas pedregosas. Escasa. JM1537, 3004, V2887. ourtia wrightii (Gray) Reveal & King. MS, B. Valles y laderas. Comtan. V5350 Ageratina callophylla (Blake) King & Rob. MS, MD. Laderas. Comtin. V4165, 4779. Ageratina ee - B.K.) King & Rob. MS, B. Laderas. Comtin. V3092, 3587, 4523, 4633, 4 A geratina ee > King & Rob. MS, B. Laderas. Comtin. V4794, 5440. Ageratum corymbosum Zucc. B. Valles y laderas. Comtin. V3622, 3948, 4795, 5370. Aphanostephus ramostssimus DC. var. ramosissimus. MD, B. Valles. Escasa. R803, V3916, 6663, 2241 Artemisia glauca Pall. B. Arroyos. Escasa. V3063a. Artemisia ludoviciana Nutt. MS, B. Valles. Comtin. V3109, 3931, 4160. Aster saenaes Benth. MD. Valles, en represas y arroyos. Escasa. JM303, 7 Baccharis havardii Gray. MS, B. Laderas y paredes rocosas. Comin. a 4140. Bahia obtinthifolia Benth. var. absinthifolia. MD, P. Valles. Comin. JM466, 771, 103 Baileya multiradiata Harv. & Gray. MD. Valles al SW. Comin. V6341. Barroetea subuligera (Schaver) Gray. MS, B. Valles y arroyos. Muy comtn. W3100, 3613, 4517, H16193. Bidens ae Gray. MS, B. Areas oe as. Escasa. V3624, 5289. Bidens ferulifolia Jacq.) DC. MS, B. Arroyos. Escasa. V4487. Brickellia chlorolepis (Woot. & Sti an Shinners. us P. Arroyos. Comtin. V3073, 4164. Brickellia cylindracea Gray. & Engelm. MS. Arroyos. Escasa. Purp4725, 4728, V4658. Brickellia laciniata Gray. MS, B. Arroyos y areas perturbadas. Escasa. V5465. eee lege Gray. var. conduplicata (Rob.) Turner. MS. Laderas pedregosas. Comun. V3521, 4167, 5372. Brickellia lemmonii Gray. var. nelsonit (Rob.) Turner. B. oes Cabs 118 SIDA16(1) 1994 Brickellia urolepis Blake. B. Comin. V4786, 5290. Brickellia veronicaefolia (H.B.K.) Gray. MS, MD. Escasa. V4166, 4781. Calyptocarpus vialis Less. MS, B. Areas perturbadas. Escasa. V5294. Centaurea americana Nutt. MS. Areas aaa Escasa. V3951, VR2198. nae bellioides (Gray) Shinners. MD, P. Valles. Comin. V3900, 4725, CQ683, 728. etopappa parryi Gray. MS, B. Laderas y panedée rocosas. Comin. V5253, 5302, 5304 eres hololeuca Greenm. MD. Valles y laderas. Rara. W3527 Chaptalia texana Greenm. MS, B. Valles. Muy comin. €Q1057, V3527a, 3977, 5375. Sean mexicana Gray. MD, MS. Valles y laderas. Muy comin. JM617, R1336, V5377. Chrysactinia pinnata Wats. MS, B. Laderas rocosas. Escasa. V4656, 5234 ce it eile (Nutt.) Spreng. MS. Arroyos y areas perturbadas. Escasa. MCJ11695a, Conoclininm greggit Gray. MM, MS, P. Areas protegidas. Comun. V3079. Conyza coulteri Gray. MS, P. Areas perturbadas. Escasa. W4142. Coreopsis tintoria Nuee var. tintoria. B, MS. Are Spa Escasa. V3544. Dichaetophora campestris Gray. MS. Areas sae rk Rara. CQ691. sets papposa (Vent.) Hitchc. MS, B, P. Maleza de areas perturbadas. Comin. V4767, 525 aces pinnatifida Nutt. MS. Areas perturbadas. Escasa. pa Erigeron bigelovii Gray. MS. Valles y laderas. Comin. V3588, Erigeron chiangit Nesom. MS, B. Arroyos, laderas rocosas. oe MCJ11700, V4662, 53 08. Evigeron modestus Gray. MS, B. Lugares hamedos. Comin CQ735, 1053, V4726, R1327, Erigeron pubescens H.B.K. MS, B, MD, P. Muy comtin. JM4503, CQ734, V4518, 5374. Fleischmannia pynocephala (Less) King & Rob.. a a y laderas. Comtin. V3577, 4780. Flourensia cernua MD. Valles. Escasa. JM Flourensia inaiiylle Blake. MS, MD. ce y oe as. Comtin. Purp4728, MCJ10110a, 10509a, 11702e, V3555, 4154, 5456. Gaillardia gypsophila Turner. MD. Lomerios. Escasa. V5590. Gnaphalinm roseam H.B.1K. B, MS. Laderas y valles. Escasa. WR2239, isa aioli dean DC. MS, B. Arroyos y dreas protegidas. Escasa. W4672, 4754. Gnaphalopsis micropotdes DC. MD, P, MS. Valles. Comin. JM1018, V2894, 4494, 4742. ah eres eae Hook. B, MS. Arroyos y dreas protegidas. Escasa. C475, V3932, 5291, 32: ee hypoleuca (DC.) Gray. MS, MD, B. Arroyos. Muy comin. JM767, 2253, ae )67. Gu ee — (DC.) Gray. MD, MS. Areas perturbadas. Comin. W3111, 3899, 414 a he (Pursh.) B. & Rusby. MD. Areas perturbadas. Escasa. C477, W4157. Gutierrezia sphaerocephala Gray. MS, MD, P. Areas perturbadas. Comin. JM1411, V3915, 4155 en glutinosum (Spreng.) Less. MS, MD, B, P. Areas perturbadas. Comin JM1070, V5361. 25, V5567. Helenium oe Labil. MD. Areas inundadas. Comin localmente. JM1022. Helianthella mexicana Gray. B. Areas protegidas. Comin. V3950, oe 5273; Helianthus annuus L. , MS, P. Maleza de areas perturbadas, C11 Heliopsis parvifolia Gray. MS, B. ce Comin. V3556, 3872, ao VR2223. ae nln ee Turner. MD. Partes bajas, N de la sierra. Comtin. Hin16555, Purp4708, MCJ10 VILLARREAL Q., Flora Vascular de la Sierra de la Paila 119 gars cae — ) Shinners. MS, B. Arroyos y paredes rocosas. Comin. V3963, Oe , VR jae ee as P, B. Areas perturbadas. Escasa. V415 — scaposa (DC. cial var. scaposa. B, MS, P. Valles. . €Q670, MCJ11688, V3602, 4762, R13 Iva paabraaepilie Gray. es perturbadas. Muy comin. V3911, ye brevifolia (Gray) Strother. MD. Laderas. Comun. ae ae 997, V4512, a guapbaleiaes (Gray) Strother. MD. Lado N. Rara. MCJ 11714a. Koanophyllum solidaginifolium (Gray) King & Rob. MS. Arroyos y laderas rocosas. Escasa. V4659 Leucelene ericoides (Torr.) Greene. MD. Valles y laderas. Escasa. V4133. Liatris punctata Hook. var. mexicana Gaiser. B. Rara. V3944. Machaeranthera gypsophila Turner. MD, P. Valles y lomerios. Comin. MCJ10123, V4158, 4544, 5574, 5588, 6363. Machaeranthera alas (Hook.) Shinners. var. pinnatifida. MD, P. Areas perturbadas. JM1936, V3927, Melampodium ee ie oe ex Rob.) Blake. MD, MS. Valles. Comin. Purp4730, R1479, Hin16563, V521( Nicolletia edwardsii Gray. MD. tien abundante, valles W de la sierra. JM754, V4137, 6344. Palafoxia texana DC. var. texana. MD. Areas perturbadas y laderas rocosas. Comin. JM774, 63, 1024, V3597, 3898, 4746, 5212 Parthenium argentatum Gray. MD. Laderas oa as. Comtin. JM770. V5649. Parthenium confertum Gray. var. lyratum (Gray) Roll. MS, B. Areas perturbadas. C788, V3941, 521 Parthenium incanum H.B.K. MD. Valles y laderas bajas. Comin. JM765, V5219. Pectis angustifolia Torr. var. tenella (DC.) Keil. MD. Valles. Comin. R1473, V4138. Pinaropappus roseus (Less.) Less. MD, MS. Laderas y valles. Comtin. V4631, 6354. Porophyllum amplexicaule Engelm. MD, P. Laderas rocosas. Escaso. JW083. Porophyllum Scoparium Gray. MD. Arroyos. Muy comin. JM795, W55 Psilactis tenuis Wats. P, MS. Areas perturbadas. Comin. V3972, 4139, VR2225. Psilostrophe washed DC. MD. Areas perturbadas. Muy coman. JM796. Sanvitalia angustifolia Engelm. B, MS. Areas perturbadas. Comin. VR2234. Sanvitalia ocymivides DC. MS. Areas perturbadas. Comin. V3947, VR. Sartwellia puberula Rydb. MD. Valles y lomerfos. Muy coman. V558¢ Sclerocarpus uniserialis Benth. var. frutescens (Brandg.) Feddma. B, MS eee ee B910; 5359, 6664 Senecio coahuilensis Greenm. B. Laderas y valles. Rara. V3946, 5371 Senecio ese DC. var. /ongilobus (Benth.) Benson. MD. Areas pertubadas, N y W de la sie mun. JM340, MCJ10515c, V2895. Sonsios neomexicanus Gray var. neomexicanus. B. Laderas y areas protegidas. Escasa. V479 1a. Solidago velutina DC. MD. Areas perturbadas. Escasa. W4790 Sonchus oleraceus L. MD, MS. Areas perturbadas, arroyos. Maleza comu Stevia micrantha Lag. B. Areas protegidas y arroyos. Rara. Purp 4721, V5325. Stevia ovata Willd. B, MS. Areas perturbadas. Escasa. V4782. Stevia pilosa Lag. MS, B. Valles. Escasa. V3934, 5285, VR2251. Stevia salicifolia Cav. var. integra (Blake) Rob. MS, B. Laderas rocosas. Escasa. Purp4722, 120 SIDAIG6(1) 1994 Stevia serrata Cay. MS. Areas perturbadas. Comtin. V3575. Stevia tomentosa H.B.K. B. Areas Seana Comin. JM1042, V4770. Tagetes lucida Cav. B. Areas perturbadas. Escasa. V3874, 5274. Tdapenalintin Ce iy. MS, B, P. Valles. ay C473, a L11c, V3606, 3893, 4162, Thelesperma megapotanucum (Spreng.) Kuntze. MD, MS, P. Laderas bajas y valles. Comin V5224, 5563. PBelperma simplicifolium Gray. MS, P, Areas perturbadas. Comin MCJ11683a, V3968, 2240. meee ma subaequale Blake. MS, P. Areas perturbadas. Comin. V3967, 5248. Thymophylla acerosa (DC.) Strother. MD. Valles y lomerfos. Comtn. V4141, 5587.6345. ee ees (DC.) Small. var. paberula (Rydb.) Strother. MD, MS. Areas p ry comun. JM1009, 1017, 1055, 2295, V2893, 3077, 3595, 5313, 6346. Thymophyl ee (DC.) Small. var. belenidium (DC.) Strother. MD. Escasa. CQ621. Thymophylta tae var. setifolia. MD. Valles y laderas. Escasa. C470. rixis californica Kell. var. californica. MD. Arroyos y laderas. Comin. V5597. Verbesina chihuahuensts Gray, MS, MD. men rocosas. Escasa. Purp4694, MCJ10108d, V3093, - Verbesina pe Turner. MS. Laderas. Escasa. V5287. Verbesina encelioides (Cav.) Benth. & Hook. MD, MS. Ma IDI Verbesina microptera DC. MS, B. Valles y arroyos. Comtin. V3557, 3966, 4756, 9363. Vernonia greggii Gray. var. ebrenbergii (Gray) Champan & Jones. B. Laderas y orilla de caminos. 351. Escasa. V3942, — eza de areas perturbadas. JM1007 Viguiera brevifolia Greenm. MD. Valles y laderas del W de la sierra. Escasa. Endlich916 (MO), Viquiera cordifolia Gray. MS. Valles. Escasa. V3079, Purp4693. Viquiera dentata (Cav.) Spreng. MD, MS, P. Areas perturbadas. Comtin. V3078, 3554, 5339. Viquiera gregeit (Gray) Blake. MD, MS. Laderas. Comtin. MCJ10104, Purp4698, V4152, 4538, 5439. Viqutera longifolia (Rob. & Greenm.) Blake. MS, P. Areas perturbadas. Comin. JM1068, 2240, V4146, 4776. Viquiera stenoloba Blake. MD. Laderas. Muy comtin. JM764, V5438, 5596. Wedelta acapulcensts Kunth. in H.B.K. var. hispida (Kunth.) Strother. MS. Valles y lomerios. Escasa. V396 Nanthium strumarium L. MD. Maleza de areas ene Escasa. V4692. Xylothamia pseudobaccharis (Blake) Nesom. MD. Valles. Escasa. Sfucateaa Zaluzania triloba Gray. MD. Valles y laderas. a. JM616, MCJ117 Zinnia acerosa (DC.) Gray. MD. Valles y laderas. Muy comin. _JM990, ee Hin16502, V5467, 5594. ae Zinnia grandiflora Nutt. MD. Valles y laderas. Comtin. V4516, 4639, 5451. BERBERIDACEAE Berberis eutriphylla (Fedde.) Muller. B. Cafiones y laderas. Escasa. eo 1709b. Berberis trifoliolata Moric. MD, MS. Valles y laderas. Comin. JM1( BIGNONIACEAE Chilopsts linearis (Cav.) Sweet. ssp. linearis var. linearis MD. Arroyos. Escasa. JM 2251, 5836, 6681 Tecoma stans (L.) Juss. MD, MS. Laderas y arroyos. CQ745, H16180, JM 1540, 2258. VILLARREAL Q., Flora Vascular de la Sierra de la Paila 121 BORAGINACEAE Antiphytum heliotropioides A.DC. MS, MD. Laderas. Comin. MCJ 10113, V3582, 4491, Cordia pine A.DC. MS. Arroyos al E. de la sierra. Rara. V5202. a iat mexicana (Brandg.) I.M.Johnst. MD. Arroyos y caminos. Comtn. CQ622, 2 confertifolium (Torr.) Gray. MS, MD. Escasa. V4540. Heliotropium sy Torr. MS. Arroyos. Escasa. V4126. Heliotropium torreyi 1.M.Johnst. MD, MS. Laderas y arroyos. Comtin. V4492, 6668. Lithospermum bie, I. M. Johnston. MS. Laderas. Comin. MCJ 11694a. Lithospermum viride Greene. MS, B. Areas protegidas. Comtin. V5256, 5344. Tiquilia canescens (DC.) Rich. MD. Valles y laderas. Muy comin. he V4542. Tiquilia greggit (T. & G.) Rich. MD. Laderas. Comtin. JM337, C Tiquilia mexicana (Wats) Rich. MD. Valles. Comin. CQ629, Ss 4677. BRASSICACEAE (CRUCIFERAE) Brassica kaber (DC) Wheeler. MD. Orilla de caminos y arroyos, maleza. Escasa. W5583. Cardamine macrocarpa Brandg. var. texana Roll. MS. Arroyos. Rara. V3940. Descurainia pinnata (Walt.) Britt. MD. Areas protegidas. Comtin.V6360. Eruca sativa Mill. MD, MS. ie de areas perturbadas. R 7 Lepidium virginicum L. MD. Maleza de dreas perturbadas. R80 ap te argyraea (Gray) Wats. ssp. diffusa (Roll.) Roll. & ae MD. Valles y laderas con an. V5582. eee fondleri (Gray) Wats. MD, MS. Comitin. R799, 1332, V2890. Lesquerella dies os Wats. ssp foliosa (Roll.) Roll. MS, B. Areas protegidas. scasa, Ds Nerisyrenia camporum (G os Greene. MD. Valles y laderas. Comtin. CQ790, V2891, 6342. Novirei lineal (Wats.) Greene. var. eats MD. Valles y laderas del W y N de la Sie n. MCJ10111, 10118a, 10510, R1467. Sisymbrium aur see Gray. MD. Arroyos y dreas aoe Escasa. V4129, 6359. Sisymbrium irio L. MS, B. Cafiones y valles con himedad. Escasa. V150 Sibara vierecki me ) Roll. MS. Arroyos y laderas con himedad. Escasa.R1332. Syathlipsis gregeii Gray var. greggii. MD. Areas perturbadas. Muy comuin. R798, V4679. Thelypodium see (Roll.) Roll. MS. Arroyos. Escasa. V3901, WR2204. BURSERACEAE Bursera fagaroides (H.B.K.) Engler. var. fagaroides. MD. Laderas. Escasa. MCJ10096, 11719, CACTACEAE inl ae mialaten ee ) Shummann. MD. Rara. Dietr.) B. & R. MD. Laderas. Escasa. Elizondo294 (ANSM), V5648. Cog bentia ee Purp. MD. Laderas y valles. Escasa. JW177, V5630 ou delaetiana (Quehl) Berger. MD, MS. Valles. Comin. JW153, ey pF, Gs 10a, Pies Dales (Quehl) Berger.MD, MS. Valles. Comtin. JW141, 1 Coryphantha echinus (Engelm.) B. & R. MD. Valles. Comin. Hiando295 nen Coryphantha laut Bremer. MD. Laderas. Escasa. Bremer476—3 (ASU). Echinocactus horizonthalonius Lem. MD. Valles y lomerios. Comin. pd Echinocereus blanckii (Posel.) Pal. MS. Valles. Escasa. JW146, 159, 183. 122 SIDA16(1) 1994 Echinocereus delaettii Gurke. B. Valles y laderas. Comtn. JW157 Echinocereus enneacanthus Engelm. var. dubius (Engelm.) Benson. MD. MS. Valles y lomerios. 8 Echtnocereus enneacanthus Engelm. var. enneacanthus. MD. Lomerios. Comin. ee Echinocereus pectinatus (Scheidw.) Engelm. MD. Lomerios y laderas. Comtn. JW123. Echinocereus stramineus (Engelm.) Engelm. MD. Lomerfos. Comin. JW 106, te Echinomastus mariposensis Hester. MD. Lader valles. Escasa. V5629. aed ite micromerts (Engelm.) Weber. var. greggii (Engelm.) Ber.MD. Valles y laderas. epee 162, V5231a. Escobar tha (Engelm.) B. & R. MD, MS. Laderas y paredes. Comtin. JW 158, 180, V5190a, 5.269. Escobaria tuberculosa (Engelm.) B. & R. MD, MS. Laderas rocosas. Comiun. V4798. Ferocactus ee (Muhl.) B. & R. MD, MS. Laderas y valles. Comun. JW109. Ferocactus stainesti (Hook.) B. & R. MD, MS. Laderas. Comin. JW125 Ferocactus uncinatus (Gal.) B. & R. MD. Valles y laderas. Comin. 1196 V5628. Grusonia bradtiana Coult. MD. Valles y lomerfos. Comin. V4 Gymnocactus aguirreanus Glass & Foster. MD. Laderas, W de la sierra. ae Glass & Foster 3044, 2206 (POMA) Leuchtenbergia principis Hook. MD. Valles y lomerios. Escasa. JW155. Lophophora williamsit (Lem.) Coult. MD. Valles. Rara. Mammillaria chionocephala Purp. MD, MS. Laderas. Escasa. Mammillaria melanocentra Posel. var. meiacantha Craig. MD. Valles y laderas. Comtn. Mammillaria pottsii Scheer. MD, MS. Laderas. Comin. JW108. Neolloydia conoidea (DC) B. & R. MD. Laderas y lomerfos. Comin. JW132, V5633. Neolloydia smithii (Muehl.) K. & F. MD. Laderas. Comin. JW131, 165, Elizondo478 M). Opuntia anteojoensis Pinkava. MD. Valles al N y W de la sierra. Escasa. Lopez sn (ANSM). Opuntia grahamii Engelm. MD. Lomerios y laderas baj as. Comuin. V5592. Opuntia imbricata (Haw.) DC. MD. Valles. Muy com 0. Opuntia kleiniae DC. MD.MS. Valles. Muy comin. 10120. Opuntia leptocaulis DC. MD. Valles y lomerfos. Muy comin. JW117 Opuntia lindheimeri Engelm. vat. aciculata (Griff.) Bravo. MS. Laderas. Escasa. Opuntia lindheimeri Engelm. var. lindhermeri. MD, MS. Valles. Comin. JW 100, 164, V5634. es lindheimeri Engelm. var. subarmata (Griff.) Eliz. & Wehbe. B, MS. Laderas y valles. a. JW 180. es lindheimeri sea var. tad (Griff.) Benson. MD, MS. Laderas. Muy comtn. Opuntia macrorhiza Engelm. Valles y laderas bajas. Escasa. Elizondo468, 469 (ANSM). Opuntia a (Lehm.) arte var. microdasys. MD. Laderas y valles. Muy comun. JW102, 148, V Opuntia ie Berger. MD. Valles al N de la sierra. Comtin. Elizondo342 (ANSM). Opuntia arte Engelm. var cates (Griff.) Benson & Walk. MD. Valles y laderas. n. Espinosa31, 50 (ANSM). Pe ae ere ree var. mayor Engelm. MD. Valles y laderas. Comin. Elizondo464 M). Opuntia rastrera Weber. MD. Valles y lomerios. Comun. Opuntia violacea Engelm. var. macrocentra (Engelm.) Benson. MD. Valles y lomerfos. Escasa. JW16 8 Thelocactus bicolor (Gal.) B.& R. var. bicolor. MD. Laderas. Comtin. JW105, 138, V5632. VILLARREAL Q., Flora Vascular de la Sierra de la Paila 140) Thelocactus rinconensis (Pos.) B.& R. var. nidulans (Quehl.) GI.&F. MD, MS. Laderas. Escasa. Gl. & F1689 (POMA). Wilcoxia tuberosa (Pos.) Berger. MD. Valles al e de la Sierra. Rara. CAPRIFOLIACEAE Abelia coriacea Hemsl|. MS. Cafiones y lugares protegidos. Rara. MCJ11678. Lonicera pilosa Willd. B. Sobre arbustos. Rara. V4769, VR2210. CARYOPHYLLACEAE Drymaria anomala Wats. B. Lugares hmedos. Escasa. V4528. Drymaria axillaris Brandg. MD. Arroyos y lomerios. Comin. W5579. Drymaria glandulosa Persl. var. ae MS, B. Lugares himedos y arroyos. Escasa. Drymaria polycarpoides Gray. MD. Lomerios. Escasa. S. Vasquez41 (ANSM). Drymaria subumbellata 1.M.Johnst. MD, MS. Laderas. Escasa. MCJ 10103, 10515b. Paronychia monticola Cory. MD. Laderas y valles. Rara. MCJ11681e. CELASTRACEAE ro Was latisepala 1.M.Johnst. MD. Laderas. Comtin. JM1054, R 1480, V4681. Imert Hemsl. MD.MS. Laderas. Escasa. CQ876, MCJ10115. ie cunetfolia Gray. MD. 7 Valles y laderas bajas. Escasa. V3104. CHENOPODIACEAE Atriplex canescens (Pursh.) Nutt. MD. Valles y laderas. Escasa. V5622. Chenopodium album L. MS. MD. Areas con disturbio. Escasa. V1509. Chenopodium graveolens Willd. B, MS. Areas hamedas y perturbadas.Comtin. V3590, 4168, 4775 CONVOLVULACEAE Convolvulus equitans Benth. MS. Maleza de dreas de cultivo. Escasa. as 6367. luscuta glabrior (Engelm.) Yuncker. MD. Parasita de Flourensia. Comtin. V6337 Dichondra argentea H. & B. MD, MS, B. Comtin. V3603, 4789, 5358, VR2230. Dichondra brachypoda Woot. & Standl. B, MS. Lugares protegidos. Comtin. V3604, 4683. Evolvulus alsinoides L. var. hirticulis Torr. MD, MS. Comin. R1468, V3917, 4533, 6666. Evolvulus sericeus 8.Wats. MD. Laderas y arroyos. Escasa. W4497, 5251, 6667. Ipomoea gs House. MS, B. Sobre arbustos bajos. Comtin. V3923, 4663, 5247, 5347, VR2242. Ipomoea Se Torr. MD. Sobre el suelo o arbustos bajos. Comtin. V4554, 4690, VR2222. Ipomoea muricata (L.) Jacq. MS. Sobre arbustos. Escasa. V. 4552, 4 Ipomoea purpurea (L.) Roth. MS. Sobre arbustos de areas perturbadas. ee V4551. Ipomoea sescossiana Baillon. MS. Arroyos. Escasa. MCJ11690, V4553. Ipomoea zimmermanit McDonald. MD. Laderas y arroyos. Escasa. Zimmernnan1948 (TEX- LL), V3982, 6672. CRASSULACEAE Echeveria strictiflora Gray. MD. Laderas rocosas. Escasa. H16197a, V6669. Len Laos oo Britt. MD, MS. Laderas rocosas. Escasa. MCJ10108b, Mee 4508. Sedum wrightii Gray. MS. Laderas y paredes rocosas de arroyos. Escasa. V418 9. Villadia ae Rose. MD. Laderas rocosas. Escasa. MCJ10109e, V4177. CROSSOSOMATACEAE Forsellesia spinescens (Gray) Greene. MD. Laderas y arroyos. Escasa. H16170. 124 SIDA16(1) 1994 CUCURBITACEAE Cucurbita foetidissima H.B.K. MD. Arroyos y areas perturbadas. Escasa. [hervillea tenutssecta (Gray) Standl. MD. Sobre arbustos bajos o el suelo. Escasa. V4670, 5617. Sicyos angulatus L. MS. Sobre arbustos. Rara. V4785. EBENACEAE Diospyros texana Scheele. MD, MS. Arroyos y cafiones. Escasa. V3070, 3564. ERICACEAE Arbutus xalapensis H.B.K. B. Comin. V3873, oe. Comarostaphylis polifolia (A.B.K.) Zucc. ss{ | is Henrick. B. Laderas y valles. Escasa. MCJ10101, 11682. EUPHORBIACEAE Acalypha lindbeimeri Muell.Arg. B. Arroyos 7 lugares aoe Escasa. V3592, VR2228. Acalypha monostachya Cav. MD, MS, B, P. Comtin. CQ783, V4547, 5243. aa Aumilis (Engelm. & Gray) Muell Arg. var. pupils MD, MS. Escasa. V4521, Ber . myrcifolia (Scheele) Wats. MD, MS. Cafiadas. Comtin. W3107, 3571, 4678. Chir eal oe (Muell.Arg) Pax. MS. Laderas de los arroyos, E de la sierra. Comtin. V3989, 4 Croton dioicus ce MD, MS. Laderas. Escasa. JM929, 1004. Croton fruticulosus Torr. Lugares hGmedos. Comin. JM985, V3572, 4665, 4741. Croton hypoleucus Schlecht. MS, B. Arroyos. Muy comtin. MCJ11710, V3108, 3924, 4478, 5 2905 VIR2199: Croton incanus H.B.K. MD, MS. Cafiones. Muy comtin. V3113, 3594. Croton pottsii (K1.) Muell.Arg. MD. Laderas bajas. Escasa. R1464, VR2220. Croton suaveolens Torr. MD, MS, B. Arroyos y laderas. Muy comin. MCJ11713b, 12753, V¥ 35334, 3926; 5275, VR2211. Euphorbia antisyphilitica Zucc. MD. Laderas bajas y valles. Muy comun. — 5627. ce hileg des Engelm. MS, B. Arroyos. Escasa. V5293, 5817, VR2235. y Engelm. MS, MD. Arroyos. Muy comun. CQ786, Mc 100, 11715, Vv. 4515, 4751, 5203, 5292. Euphorbia cyathophora Murr. B, MS. Lugares protegidos y himedos. V3894, VR2229. Ez uphor bia dentata Michx. MS. Arroyos y areas perturbadas. Comtin. V3885, 4527, 4750, R2231. Euphorbia scopulorum Brandg. MD. Laderas bajas del W de la sierra. Comtin. R1478. oo eis Scheele. MS, B. Arroyos. Muy comun. MCJ116811, V3623, 3897, Jat bois Cerv. MD. Valles y laderas bajas. Muy comtin. JM961 eee Bg Spreng. MD, MS. Arroyos y laderas bajas. MCJ11 697, V3922, 230: ee een Arg.) I.M.Johnst. MD. Cafiones. Escasa. JM13 Tragia amblyodonta (Muell.Arg.) Pax.&Hoffm. MS.MD. Areas cane is y arroyos. Comun. V5296. FABACEAE (LEGUMINOSAE). MIMOSOIDEAE Acacia berlandier? Benth. MS, MD. Arroyos y laderas. Muy comun. C467, V5624. Acacia constricta Gray. MD. Orilla de caminos y laderas bajas. Comtin. C467, JM2280. Acacia greggit Gray. MD, MS. Arroyos. ee V5598. Acacia hirta V.& G. MS, B. Escasa. V41 VILLARREAL Q., Flora Vascular de la Sierra de la Paila 125 Acacia neovernicosa Isley. MD. Laderas bajas y arroyos. Comtin. V5458. Acacia rigidula Benth. MS. Valles y arroyos del E de la sierra. Comtin. V3884. Acacia roemeriana Scheele. MD, MS. Arroyos y laderas bajas. Escasa. CQ929, 951, W5577. Calliandra conferta Benth. MD, MS. Arroyos y laderas. Comtin. CQ928, V3881, 4145, 4667. Desmanthus painteri (B.&R) Standl. MD. Rara. V3973. Desmanthus virgatus nae var. glandulosus Turner. MS, B. Comin. V4771, 5320. Leucaena greggii S.Wat oo y laderas de cafiones. Comtin. CQ909, H16172, MCJ10117, 73088, pe Mimosa biuncifera Benth. MS. ne y arroyos. Escasa. VW4536, 5434. Mimosa emoryana Benth. var. emoryana.MD. Valles y laderas bajas. Comtin. V3311, 4483, 5241, 5460, 5593. Minn texana en Small. MD. Comtin. CQ83 ray. MD. Valles y laderas bajas ot N y W de la sierra. Comtin. V5462, eas wpe A 5578. Mimosa zygphylla Gray. x M. emoryana Benth. MD. Laderas bajas. Escasa. W54 rosopts Hee pec var. glandulosa. MD. Arroyos y laderas bajas. Comun. aan 5650. Zapoteca media (Mart. & Gal.) Her. MS. Arroyos. Escasa. W5205. FABACEAE. CAESALPINOIDEAE Bauhinia ramosissima Benth. MD. Laderas. Comtin. JM334, 992, 1058,H16176, MCJ 10094, 7, V3610, 4146, 4486, 5461, 6686. Caesalpinia parryi (Fisher) Eifert. MD. Laderas. Escasa.V5589. Cercidium texanum Gray. MD. Arroyos, lado E y N de la sierra. Rara Chamaecrista greggti (Gray) Heller. MS. Laderas de cafiones. Comun. “V4. 490, 4695, 5206, 6688. Hoffmanseggia glauca (Ort.) Eifert. MD.MS. Areas perturbadas. Comun. a ated V3584. Hoffmanseggia oxycarpa Gray. MS. Laderas y arroyos. Escasa. V3975, CQ69 Senna bauhinivides (Gray) Irwin & Barneby. Valles. Escasa. CQ358, oS Senna sii alee (Scheele) Irwin & Barneby. MS. Arroyos. Escasa. V3616. Senna monozyx (Irwin & Barneby) Irwin & Barneby. MD. Valles NW de la sierra. Escasa. Hin16565. FABACEAE. LOTOIDEAE Astragalus sanguinens Rydb. B. Arroyos y areas con disturbio. Rara. W5355. Centrosema virginicum (L.) Benth. B, MS. Sobre arbustos. Escasa. V3990a Cologania angustifolia H.B.K. MS, B. Laderas de arroyos. Comtn. V4793, 5260, VR2188. Cunine (aca ) Benth. var. edwardsii (Gray) Hassl. MS, B. Lugares himedos. Comin. V3586, 3903, 4735, VR2193. Datlea aurea Nini, MD. Laderas bajas, entre arbustos. Escasa. V4772. Datea bicolor H.B.K. var. bicolor. MS, MD. Arroyos y cafiadas. Muy Comtin V3076, 3541, 4763, 5286. Datea frutescens Gray. MD. Laderas bajas. Escasa. CQ779, V3958. Datlea greggii (Gray) Heller. MD, MS, P. Comin. R801, V2886, 3908, 5573. oS cee (Rose) Bullock. MS. Arroyos. Muy Comtin. C468, V3957, 4170, 5216, be ae orn ) Willd. var. /utea. MS. Arroyos y laderas bajas. Comtn. W4549, 5240. Dalea melantha cae Rydb. MD. Arroyos. Comtin. V4164, 5463. Datlea nana Torr. var. carnescens (Rydb.) Kearn. & Peeb, MD. Val les. Escasa. V6689. Datlea ae Gray. MD. Valles y laderas bajas. Muy comin. CQ801, V3534, 3878, 126 SIDA16(1) 1994 Datlea radicans Wats. MS. Arroyos y valles. Comtin. MCJ11702c, V4800. Dalea wrightii Gray. MD. Valles y laderas bajas. Muy comin. R 1466, Purp4740, V4448, 5228, 5459 Desmodium grahamii Gray. MS. Arroyos. Escasa. V3962, VR2195. Desmodium neomexicanum Gray. B, MS. Arroyos y laderas protegidas. Escasa. W523 Eysenhardtia parvifolia Brandg. MD. Laderas N y W de la sierra. Comin. conde V4147, 4485. della ees as ) has MS. Arroyos y cafiadas. Comtin. JM2246, V3930. MD. Sobre arbustos bajos. Comtin. CQ680, 744, MCJ11720, V4482, 4638, in Indigofera acutifolia Schlecht. MS, B. Arroyos. Comin. CQ854, V3974, 5258. Nissolia platycalyx Wats. MS, B. Sobre arbustos. Comin. V3979, 5235, 6687. Phaseolus plagiocalyx Harms. B. Enredadera. Escasa. VR2194. Phaseolus xanthotrichus Piper. MS. Laderas, en cafiones. Escasa. V3961. Rhynchosia senna Hook. var. es (Gray) Grear. B, MS. Valles. Comun. VR2190. Sophora nuttalliana Turner. B. Valles. Comin. V5824, VR2253. Sophora secundiflora (Ort.) Lag. ee MS, B. Arroyos y laderas. Comuin. JM322, H16192, 16597, R1329, MCJ10517, V4724, 5557 FAGACEAE eee gravesii Sudw. B. Valles y laderas, cafiones. Muy comtin. JM1078, 2315, V3617, 38, 4759, 5357. nee grisea Liebm. MS, B. Cafiones y valles. Comin. W5825, VR 2209 Quercus intricata Trel. MS, MD. Arroyos y cafiones. Comin. JM1048, C464, V3118, VR2202 Quercus invaginata Trel. MS. Arroyos y laderas de cafiones. Muy comun. Purp5030, 5029, M2336, 3935, C465, V3117, 3563, 4178, 4666, 5238, 5334, 5570, MCJ11713a. Quercus ae Liebm. B. Valles y arroyos. Rara. V3938, 478 Quercus laceyi Small. B. Valles cafiones y laderas. Muy comin. 3937, 5939, 5350. Quercus mohriana Buckl. MS. Cafiones. Escasa. JM sn FOUQUIERIACEAE Fouquieria splendens Engelm. ssp splendens. MD. Laderas rocosas y valles. Muy comun. JM3263 GARRYACEAE Garrya ovata Benth. ssp lindhermeri (Torr.) Dahling. B, MS. Cafiones. Escasa. JM1076, MCJ11709a, V3074, 3952, 4792, 6682 Garrya wrightii Torr. B. Cafiones. Escasa. V4791. GENTIANACEAE Centaurium calycosum (Buckl.) Fern. var. calycosum. MS, P, MD. Valles. Comin. H16181, 74, V3918, 4159, 5367. HyYDROPHYLLACEAE Nama palmeri Gray. MS. Arroyos y laderas bajas. Comin. CQ651, 716, V4515, 4684, 6671. Nama undulatum H.B.K. MD, MS. Areas perturbadas. Comtin. CQ627, V6336. Phacelia congesta Hook. MS. Arroyos. Escasa. CQ701 JUGLANDACEAE Juglans microcarpa Berl. MS, B. Arroyos. Escasa. JM2248, V3567, 4736, 5839. VILLARREAL Q., Flora Vascular de la Sierra de la Paila 1277 KOEBERLINIACEAE Koeberlinia spinosa Zucc. MD. Arroyos y laderas bajas. Comin. JM319, 1005. KRAMERIACEAE Krameria grayi Rose & Painter. MD. Laderas bajas W . la sierra. Escasa. H16503, V5626. Krameria lanceolata Torr. MS, B. Valles. Escasa. V358 ; Krameria ramosissima (Gray) Wats. MD. Laderas ae Beas V4643. LAMIACEAE (LABIATAE) Hedeoma costatum Gray vat. costatum. B, MS. Paredes rocosas y laderas. Comtin. V3876. Hedeoma microphyllum Irving. MS, B. Valles y laderas. Comtin. MCJ11695, V3875, 5327. Hedeoma montanum Brandg. MD. Laderas bajas y valles N de la sierra. Comin. Purp4964, MCy! 0111, C461, 4539, 5435. Hedeoma nanun (Torr.) Brig. MD. Laderas y valles al W de la sierra. Comin. V6366. Hedeoma plicatum Tort. MS, B. Laderas pedregosas. Escasa. V5327. Marrubium vulgare L. MS, B. Areas perturbadas. Comin. W5575. Poliomintha glabrescens Gray. B, MS. Valles y areas cercanas a arroyos. Comtn. MCJ11684a, V3532, 3920. Salvia ballotaeflora Benth. MD, MS. Arroyos. Comtin. V4501, 4608. Salvia farinacea Benth. MS. Arroyos. Comin. Purp4754, V3075, 3529, 4747. Salvia greggii Gray. B, MS. Laderas y valles. Comtin. MCJ11691a, V3888, 5314. Salvia refleca Hornem. MS. Areas perturbadas y arroyos. Escasa. V3608. Salvia regla Cav. MS, B. Laderas. Escasa. V3611, 3887. Salvia roemeriana Scheele. B, MS. Arroyos. Comtin. V4777, 5334, 5834. Salvia tiliaefolia Vah\. B. Areas perturbadas. Escasa. V3547. Scutellaria hispidula Rob. MS, MD. ae y age Bscasa. MCJ116 Scutellaria wrightii Gray. MS, B. Valles ones. Comin. oe a0) 14732. 5332, VR2244. LINACEAE Linum lewisit Pursh. B. Arroyos. Escasa. JM2306, V3110, 4156. Linum nelsonit Rose. B. Areas protegidas y arroyos. Escasa. V4739 Linum rupestre (Gray) Gray. MS, MD. Muy comin. CQ636, MCJ10098, 10108f, V3151, 4696, 5217, 5249, 6349 Linum vernale Woot. MD. Laderas bajas y valles. Comtin. V5586. LOBELIACEAE Lobelia henricksoniti M.C.Johnst. MS. Arroyos. Lugares himedos. Escasa. MCJ11681g, V6665 LOGANIACEAE La sell Benth. MD. Arroyos y laderas. Comin. V3883. dd cordioides H.B.K. MS. Areas perturbadas. Escasa. JM1067. ee aa Torr. a Arroyos y laderas bajas. Cafiones. Comin. JM2226, V3066, 3619, 4537, 4646, 4748. Spigelia lindheimeri Gray. MS. Arroyos pedregosos. Rara. V4705, 5229. OASACEAE Cevallia sinuata Lag. MD. Laderas bajas. Escasa. CQ804. Eucnide bartonioides Zucc. MD, MS. Laderas de arroyos y paredes rocosas. Escasa. V4535. Eucnide lobata (Hook) Gray. MD, MS. Laderas de arroyos y paredes rocosas. Escasa. CQ843, 96 128 SIDAL6(1) 1994 Mentzelia asperula Woot. & Stand|. MS. Arroyos. Escasa. V5237. Mentzelia hispida Willd. MS, B. Valles y laderas bajas. Comtin. V4520. Mentzelia mexicana Thomps. & Zavortink. MD. Laderas y valles pedregosos. Escasa. Mentzelia saxicola Thomps. & Zavortink. MD. Laderas pedregosas. Escasa. MCJ10105. MALPIGHIACEAE Echinopterys setosa Brandg. MS. Laderas bajas de cafiones. Comtin. Purp4950, MCJ10121, V4519, 6367. (25 Mascagnia cana Small. MD. Laderas y arroyos. Comin. MCJ10120, V4150, 4524. Mascagnia lilacina (Wats.) Niedenzu in Endl. & Pranl. MS, B. Sobre arbustos, paredes rocosas. Comtin. V3308, 4530, 5199 MALVACEAE Abutilon hypoleucum Gray. B, MS. Escasa. V4654, 6679 Allowissadula holosericea (Scheele) Bates. MS. Areas protegidas. Rara. V3869. Anoda crenatifolia Ort. MS. Areas perturbadas. Escasa. V4660 Anoda pubescens Schlecht. MD. Arroyos. Rara. CQ1230 Anoda thurberi Gray. MD, MS. Areas perturbadas. Comtin. V4701, 4802. Batesimalva violacea (Rose) Fryx. MD. Laderas. Escasa. W3870. Hibiscus coulteri Gray. MD. Laderas rocosas. Comtin. JM760, V4130, 4653, 6361 Hibiscus martianus Zucc. MD, MS. Laderas y Valles. Comtn. JM761, V4652. Pavonia lasiopetala Scheele. MS. Areas protegidas. Comtin. V3612, 3913, 4651, 5328. Sida lindheimeri Engelm. & Gray. MS. Valles y dreas perturbadas. Comin: V3615,3959;..5376; Sida longipes _ MS, B. Valles. Comtin. V5215, 5319. Sida spinosa L. MS. Areas cena Escasa. V3615 Sibueraliin angustifolia (Cav.) D.Don. MD, B. Areas serenade, Comtn. CQ604, 782. Sphaeralcea endlichti Ulbride. en Laderas y lomerfos. Muy comtin. Endlich847, V3595, 4127. Sphaeralcea hastulata Gray. MD. Valles y laderas. Comtin. V4128, 6350. MARTYNIACEAE Proboscidea fragrans (Lindl.) Dene. MD. Orilla de caminos, areas perturbadas. Comuin. V4693, 5317. NYCTAGINACEAE Acleisanthes acutifolia Standl. MS. Orilla de arroyos. Escasa. W447: Hedet anes ely Gray. MD. Laderas rocosas. Comuin. a incarnata L. MD. Valles n. CQ79 Boer hate ae Gray. MD. Laderas rocosas. Comtin. V4473, 5224, 5452. C ee es (Mart. & Gal.) Standl. MS, MD. Arroyos. Conta. V3574, 3921, Mee i be ae (Ort.) I.M.Johnst. MD. Laderas. Escase, V5465a Mirabilis longiflora L. MS. Areas protegidas. Comtin. V3573, 4532, 5265. EACEAE Forestiera angustifolia Torr. MS. Valles y cafiadas. Comtin. V4495. Fraxinus cuspidata Torr. B, MS. Cafiadas y arroyos. Comtin. CQ776, 814, R1337, V3954, 4760, 5360. Fraxinus greegil Gray. var. gregei?. MS. Laderas y arroyos. Muy comin. JM1045, V3953, 4499, 5299. Menodora coulteri Gray. MD. Laderas. Escasa. Purp5020. Menodora longiflora Gray. MD. Laderas y valles. Muy comtin. JM775, 2274, 2231, H16173, V3880, 4738, VR2215 VILLARREAL Q., Flora Vascular de la Sierra de la Paila 129 ONAGRACEAE oe hartwegii (Benth.) Raven var. hartwegii. MS, MD. Valles. Comtin. MCJ10509b, 733, 5246a, 6339, VR2247. ee eye ae Raven. MS, MD. Valles y arroyos. Escasa. CQ1051, V3990. Gaura calcicola Raven & Gregory. MS, MD. Arroyos. Comtin. CQ1054, V3955, 5255, 5285, 6338 OROBANCHACEAE Orobanche cooperi (Gray) Heller. MD. Valles de Larrea y Fouquieria. Escasa. V4513, 6352. OX ALIDACEAE Oxalis albicaulis H.B.K. var. pilosa (Nutt.) Eiten. B, MS. Arroyos y areas protegidas. Escasa. V3598.VR2217 Oxalis latifolia H.B.K. MS, B. Arroyos. Comin. MCJ11681d, V2214. PASSIFLORACEAE Passiflora tenuiloba Engelm. MD. Valles y cafiones. Escasa. W6670. PAPAVERACEAE Argemone fruticosa Thurb. ex Gray. MD, Valles. Rara. Hin 16582. Argemone sanguinea Greene. MD, valles. Escasa. Hin 16600, V6368. PHYTOLACCACEAE Rivina humulis L. MS, B. Arroyos. Comin. CQ785, V4671, 5227. POLEMONIACEAE Gilia incisa Benth. B, MS. Areas himedas. Escasa. W4744. Gilia insignis (Brandg.) Cory & Parks. MS, MD. Cafiadas. Escasa. Gilia rigidula Benth. ssp rigidula. MD. oe bajas. Comtin. CQ662, V2896. Gilia stewartii 1.M.Johnst. MD, MS. Laderas rocosas. Comin. JM2237, oun. MCJ10515, 61, 6358, WR2245. Loeselia coerulea (Cav.) Cav. MD, P. Valles y laderas bajas. Comin. V4743, 5453. Loeselia greggii Wats. MD. Laderas pedregosas. Comtin. V3576, 4556, 5257. POLYGALACEAE Polygala alba Nutt. MD, MS. Cafiones. Comtin. Purp4760, JM2329, R795, H16186, V5271, MCJ11711. Polygala barbeyana Chodt. MD, B. Laderas y valles. Comtin. MCJ11703a, V3532, 4529, 4657, 5270, 5639, 6362.VR2246. Polygala dolichocarpa Blake. MD. Arroyos. MCJ10521b. Polygala pias a var. lindheimeri. MS. Arroyos y laderas bajas. Comuin. JM3985, 4749, CQ818, V58 673. Polygala eee Gr MD. Laderas rocosas. Muy comin. JM777, 783, CQ659, H16182, MCJ1052 ie) data ay MS, MD. Arroyos. Comtin. Purp4762, H16168c, V4698, 5239, 5560, 5615 5, MCJ10521d, 11720a. Polygala parrasana Brandg. MS. Laderas bajas. Escasa. MCJ116 Polygala scoparioides Chodat. MS. Cafiones y arroyos. Comin. en 685, 723 Wy58. Polygala watsonii Chodat. MD, MS. Laderas. Escasa. MCJ11698, V5581. POLYGONACEAE Eriogonum greggit T. & G. MD. Laderas Daas cae V3981. PELE yaa Benth. in DC. va lil h.) Stokes ex. Jones. B. Valles. Comin. Robert4488 (ANSM). 130 SIDAL6(1) 1994 PORTULACACEAE Portulaca munditla 1.M.Johnst. MD, MS. Valles y arroyos. Muy comtin. V3614, 3919, 4511, 52, VR2248. Portulaca retusa Engelm. MD. Arroyos. Rara. V4510. Talinopsis frutescens Gray. MD. Laderas bajas y paredes rocosas Escasa. V45 26. Talinum aurantiacum Engelm. MD, MS. Valles y laderas bajas. Escasa. W5220. RANUNCULACEAE Aquilegia longissima Gray. MS, B. Laderas y arroyos. Escasa. V3528. Clematis coahuilensis Keil. MS. Arroyos y paredes rocosas. Escasa. V4657. Clematis drummondit T. & G. MD, MS. Areas perturbadas. Escasa. CQ1055. Clematis pitcheri T. & G. Sobre arbustos bajos. Escasa. V3893, VR2226 Thalictrum grandifolinm Wats. B, MS. Areas protegidas y htimedas. Muy comtn. CQ905, V3611,9929, 5778: RESEDACEAE Oligomeris lintfolia (Vahl.) Macbr. MS. Laderas bajas pedregosas y lugares inundables. Comitin. JM468, V5600, 6343. RHAMNACEAE Ceanothus greggit Gray. MS, B. Valles y laderas. Comtin. V3609, 5842, VR Colubrina greggit Gray. MS, B. Cafiones y arroyos. Escasa. V36 Condalia ericoides (Gray) M.C — MS. Laderas bajas y ies Cmte. C460, V414 Condalia warnockii M.C ae . MD. Valles y laderas bajas. Comtin Condalia viridis 1.M.Johns "MS, Cafiones. Escasa.CQ1008, V3890, 5841. isan baer do Sc t.)Zucc. MS, MD. Valles wcatarlas Comin. JM993, V3064, 5 S| es betulifolia Greene. B. Cafiones. Escasa. W4757. Rhamnus standleyana C. B. Wolf. B. Valles y cafiones. Rara. Purp 4968, V5840, VR2207. Ziziphus obtusifolia (TV. & G.) Gray. MD.MS. Valles y cafiones. Comtin. JMsn, V5618. ROSACEAE nas denticulata (H.B.K.) Koch. MS, B. Laderas bajas, arroyos y cafiadas. Comin. V4149, 4771, 5342, 6685. Gear mojadensts Sdhneid. MS, B. Laderas en cafiones. Comin.JM2260, 2283, 3272, , MCJ10518, V3081. Pee ins montanits a var. ae (Wats.) F.L.Martin. MS, B. Cafiones. JM2296, V3080, 3914, 4764, Prunus serotina Ehth. var. virens (Woot & Standl.) Mc. Vaugh. B. Cafiones. Rara. V3943, 5821. Vauquelinia corymbosa H. & B. ssp. heterodon (1.M.Johnst.) Hess & Henrick. MS. Cafiones y — Muy comin. JM1518, H16171, 16501, MCJ10110b, 11684, V2888, 4496, 5298, 3091 RUBIACEAE Bouvardia ternifolia (Cav.) Schlecht. MS. Arroyos. Escasa. V3530, ae 5208. Coutaportla pailensis oe MS. Paredes rocosas. Rara. V3063, Galium uncinilatum Gray. B. Arroyos y areas Sens Po V: ne 635 Hedyotis acerosa Gray. MD. ae pedregosas y valles. Comin. JM988, Rl oa V3090, 3536, 3912, 4534, 4700, 4745. Hedyotis intricata Fosb. MS, MD. Laderas pedregosas. Comin. V3089. Hedyotis nigricans (Lam.) Posh: var. nigricans. MS, B. Arroyos. Comtin. V3625, 3904, 4734. Hedyotis palmeri (Gray) Lewis. MS. Arroyos. Escasa. V4546, VILLARREAL Q., Flora Vascular de la Sierra de la Paila 131 Randia pringlei Gray. MD, MS. Laderas rocosas y valles. Muy comtin.JM994, CQ650, H16178, V3106, 3307, 3569. Relbunium microphyllum (Gray) Hems!|. MS, B. Arroyos. Comin. MCJ11681h, V3579, 3909, 4645, 19 RUTACEAE Amyris marshii Standl. MS. Cafiones. Comtin.MCJ1 V3583, 4766, 6678. Choisya palmeri Gray. MS. Cafiones. Escasa. MCJ116 Ptelea trifoliata L. MS, B. Cafiones y arroyos. Muy comtin. MCJ11702d, R1331, V3531, Thamnosma pailensis M.C.Johnst. MD. Laderas bajas, N de la sierra. Rara. MCJ12751, L. Woodruf 369 (TEX-LL). ees texana (Gray) Torr. f. purpurea (Woot. & Stand.) Lundl. MS. Lechos de arroyos. asa. W4682, 5242. SAPINDACEAE Dodonaea viscosa Jacq. MS. Lechos de arroyos. Comtin. H16169, MCJ10119, V3309, 3546, DOD's Ungnadia speciosa Endl. MS, B. Arroyos y cafiones. Comin. V3580, 5316. SAPOTACEAE Bumelia lanuginosa (Michx.) Pers. var. rigida Gray. MS, MD. Cafiones. Escasa. CQ841, 953, V5831. SAXIFRAGACEAE Fendlera rigida 1.M.Johnst. MS. Laderas rocosas. Escasa. MCJ10108c, 11683b, VW5226. Fendlerella lasiopetala Standl. MS. Laderas rocosas. Escasa. JM1050, MCJ11696, V5818. SIMAROUBACEAE Castela erecta Turp. ssp texana (T. & G.) Crong. MD. Valles. Rara. JM333, 615, V5619. Holacantha stewartit Mull. MD. laderas rocosas. Escasa. JM3897, V6690. SCROPHULARIACEAE Castilleja lanata Gray. MD. Valles y laderas bajas. Escasa. JM321, V6343. Castilleja he iy H.B.K. MS, MD. Arroyos y laderas pedregosas. Comtin. JM3215, CQ687, V4755, 5845. Hemichaena spinulosa (Wats.) Thieret. MS, MD. Paredes rocosas. Comtin. MCJ11679, ¥3306, Vaan ntha (Cham. & Scht.) Ernst. MD. Lad 1 Escasa. MCJ10116c, V4509. ee Lencophyllum candidum 1.M.Johnst. MD. Laderas bajas y valles del W de la sierra. Escasa. Leucophyllum frutescens (Berl.) 1.M.Johnst. MS, MD. Valles y laderas bajas. Comtin. V3087, Leucophyllum minus Gray. MD. Laderas bajas. Comtin. MCJ10111b, V3086, 4131. Maurandya a H. & B. ssp antirrhiniflora. MD, MS. Sobre arbustos en arroyos. Escasa V4632 pes oni H. & B. SP hederifolia (Roth.) Elisens. MD. Sobre arbustos de s.Escasa. JM1081. ee ae (H.B.K.) — MD, MS, P. Aroyos y lugares himedos. Escasa. Penstemon barbatus (Cav.) Roth. MS, B, P. Valles y laderas bajas. Comtin. JM2332, V3925, 4804 132 SIDA16(1) 1994 Penstemon lanceolatus Benth. MS. Arroyos. Escasa. V4680. — punctatus — MS, B. Arroyos. Comtn. Purp4747, CQ1056, MCJ11685, , 3933, 4778, 52 aes ee ee Heller. ssp ania ifolius Penn. MS, B. Valles y arroyos. Escasa. R1339. Seymerta coahuilana (Penn.) Stand. MS, MD. Valles y laderas bajas. Comtin. MCJ10109c, V4132, C469. Somer ola Turner. var. falcata. MS. Valles y laderas bajas. Comin. V3607, 3965, 4685, 53 oe virgata (H.B.K.) Benth. MD, MS. Laderas rocosas. Escasa. MCJ11681b, VR2205. SOLANACEAE Ch 25 es (Moric. ex Dunal) Britt. MD. Valles al W de la sierra. Comtin. V5348. ee sonia (Dunal) Gray. MS. Valles y arroyos. Escasa. V4650. en ag Averett. MS, MD. Arroyos. Comtin. CQ1052, R1334, V4649, 5207 6662 Oeecals on) ea ae MS, MD. Arroyos. Comtin. V3556, 4648. Datura inoxia Mill. MD, MS. Arroyos, orilla de caminos. Comtin. JM964, V4694, 5571. uate wii H.B. K. MS, MD. Areas perturbadas y arroyos. Escasa. V4665. @(Torr.) Hunziker & Subils. MS. Arroyos. Comtin. R1340, JM754, V4647, 52 14 : Margaranthus solanaceus Schlecht. MS. Areas perturbadas y arroyos. Comuin. V4478, 4689, 5261. eae sree Grab. MD. Areas perturbadas y arroyos. Escasa. CQ796. var. oe ) Waterfall. MS, B. Lechos oe Comtn. V3552. shat microphysa Gray. B, MS. Cafiones y valles. Escasa. V5309. Quincula lobata (Torr.) ae MS. Valles. Comin. V3537. Solanum americanum Mill. MS, MD. Lechos de roe Comin. V3936, 4758. Solanum eleagnifolium Cav. MD. Areas perturbadas. Comin. Solanum triquetrum Cav. MS, MD. Entre arbustos y lugares protegidos. Comin.V3548, 4644, 5204 STERCULIACEAE Ayenia microphylla Gray. MS, MD. Valles y laderas bajas. Escasa. R1459, W5222. Ayenta pilosa Cristobal. MS, MD. Valles y laderas bajas. Comin. MCJ11681, V4477, 4688, 5349 VERBENACEAE Aloysia gratissima (Gill. & Hook) Troncoso. MS, MD. Cafiadas y valles. Comtin. CQ798, Aloysia wrightti (Gray) Hellor. MD. Laderas rocosas. Escasa. V4493, VWR2203. Bouchea spathulata Torr. MS, B, P. Valles. Escasa. MCJ10107, V3099, 6365. Citharexylum brachyanthum (Gray) Gray. MS, MD. Valles. Comin. V3560, 4674. Lantana horrida H.B.K. MS. Cafiones y arroyos. Escasa. V4636. Lantana macropoda Torr. MD, MS. Valles y laderas bajas. V4635. Lippia graveolens H.B.K. MD. Laderas. Muy comin. V3907, 5 nee Priva mexicana — ) Pers. B. Valles y arroyos. Escasa. VW3353, VWR22 Tetraclea coulteri Gray. MD, MS. Valles y deka Comin. V5437. Verbena salads Nutt. MD, MS. Valles. Com Verbena elegans H.B.K. MS. Valles. Escasa. W430 i Verbena neomexicana Te Small. MS, MD. Arroyos. Escasa. MCJ1168. VIOLACEAE Viola sororta Willd. B, MS. Arroyos y valles. Areas protegidas. Comtin. R1336, V3985, 4790. VILLARREAL Q., Flora Vascular de la Sierra de la Paila 133 VISCACEAE Phoradendron lanceolatum Engelm. MS, B. Sobre Quercus. Comin. MCJ10102, 11709, Vsil2, 3565, 5303 Phoradendron tomentosum (DC.) Gray ssp tomentosum. MS, B. Sobre Acacia. Escasa. V3068. Phoradendron villosum (Nutt.) Nutt. MD, B, MS. Sobre Quercus. Escaso. V3984. VITACEAE Cissus incisa (Nutt.) Desmoul. MD, MS. Sobre arbustos bajos. Escasa. V4673 Parthenocissus quinquefolia (L.) Planch. MS, B. Paredes rocosas de cafiones. oe V3578. Vitis berlandieri Planch. B, MS. Sobre arboles y arbustos. Comtin. V3525, 3892, 5823, 6661. ZYGOPHYLLACEAE Guaiacum angustifolium Engelm. MD, MS. Valles y cafiones. Comin. CQ621, V3105. Kallstroemia parviflora Morton. MD. Laderas bajas y valles. Comin V545 Larrea tridentata (DC.) Cav. MD. Valles y laderas bajas. Muy comin. ie Peganum mexicanum Gray. MD. Valles dreas perturbadas. Escasa. JM Sericodes greggii Gray. MD. Valles y laderas bajas del WS de la sierra. ve comin. V5602, 5609. LILIOPSIDA (Monocotideloneas) AGAVACEAE Agave lechuguilla Torr. MD, MS. Laderas rocosas. Muy comin. JMsn, H16196, V5610. Agave seein Todaro. Laderas. Escaso. H16185, os 20050. Agave parrasana Berger. MS. Laderas de cafiones. Escas Agave ee cae Dyck. ssp scabra. MD, MS. ne rocosas.Comtin. V5638 Agave striata Zucc. ssp. falcata (artis Gentry. MD, MS. Laderas y valles. Comun. Endlich879a, V5566, 5612. Dasylirion palmeri Trel. MD, MS. Laderas rocosas y arroyos. Muy comun. Endlich7 Hesperaloe funifera (Koch.) Trel. MD. Arroyos y valles del E y N de la sierra. Escasa. Nolina cespitifera Trel. MD, MS. Laderas bajas rocosas. Comin. JM2347, V5340. Yucca carnerosana (Trel.) McKelvey. MS. Laderas rocosas de cafiones. Comun. JMs.n., V2897, a} Yucca endlichiana Trel. MD. Valles al W de la sierra. a. Comin V2126, 6340. Yucca filifera Chabaud. MD. Valles y arroyos. Comin. V3986. Yucca torreyi Schafer. MD. Laderas pedregosas a ‘Comes, a 459, Yucca treculeana Carr. MD. Laderas y valles. Comin. V5611 AMARYLLIDACEAE Cooperia drummondii Herb. MD. Valles y arroyos. Escasa. V6676. Zephyranthes longiflora Hemsl. MD. Valles al N. Escasa. W667 ARECACEAE (PALMAE) Brahea berlandieri Bartlett. MS. Paredes rocosas y arroyos de cafiones. Comtin. V3620. BROMELIACEAE Hechtia texensis Wats. MD. Laderas y valles r s. Muy comtin. V5569, 5613, 5844. Mee recurvata L. MD, B. Sobre ee ee y Pinus. Comin. CQ817, V3114, 96 (ls DO2e ens usneoides (L.) L. B. Sobre Pinus y Quercus. Valles. Comin. V5837. COMMELINACEAE Commelina erecta L. var. angustifolia (Michx.) Fern. MD, MS, B. Laderas rocosas. Comun. V3971, 4488, 4640, 5366. 134 SIDAL6(1) 1994 Gibasis linearis (Benth.) Rohw. MD, MS. Valles y laderas pedregosas Comtin. V4641, 5338 Tradescantia brachyphyHa Greenm. MS, MD. Laderas rocosas. Escasa. W5266, VR2237. Tradescantia crassifolia Cav. MS. Paredes rocosas de canones. Escasa. V4702, 5337, VR2238. CYPERACEAE Carex schiedeana Kuntz. MS, B. Arroyos y cafiadas. Comin. MCJ10109a, 11691b, 11693, V4753, 6674. Cyperus acuminatus T. & H. MS. Lechos de arroyos. Rara. VW4500. IRIDACEAE Sisyrinchium scabra C & 8. MS, B. Arroyos y cafiones. Escasa. JM1071, MCJ1168 1a. LILIACEAE Allium kunthit Don. MD. Arroyos. Comtin. MCJ101 16a. — & Thomps) Cruden. B. Valles y laderas bajas. Comtin. V3960, R2208 Pei ee Benes. MD. Valles y laderas bajas. Escasa. Purp4959, V3980. Schoenocauton coultert Baker. MS, B. Valles, cafiones. Rara. V5245. Smilax bona—nox L.. MS, B. Sobre arbustos y arboles, cafiones. Escasa. V3976. ORCHIDACEAE a omanthus shag ‘TNUS sii & Lex.) Gray. MD. laderas pedregosas. Comtin. MCJ 100 ’ 4 - 4 os ye 27 o, oo pe G C. a inst. MS. Laderas rocosas. Rara. MCJ10108a. POACEAE (GRAMINEAE) Andropogon spadiceus Swallen. MS. Valles y arroyos. Escasa. V3084, 3630-6, VR2185. Aristida adscensionis L. MD, P. Valles y laderas. Comtin. Ca44, V3627, 4506, 5192. ristida arizonica Vasey. MD. Valles y lomerfos. Comin. Ca50, V5385. Avid ian Warnock. MD, MS. Valles. Escasa. W5469, » Nutt. var. nealleyi (Vasey) Allred. MD, MS. Valles. Comin. V5181, 5470 > ee 5. Aristida purpurea Nutt. var. wrightii (Nash) Allred. MD, MS, P. Valles y laderas. Muy comtin. Ca40, V4810. Aristida ternipes Cav. MD. Valles y laderas. Comtin. V4558, 5441 Bothriochloa barbinodis (Lag.) Herter var. barbinodis. MD.MS, B. Valles y arroyos. Comin. Ca56, V5188. Bothriochloa laguroides(DC.) Hertor. ssp forreyana (Steud.) Allred & Gould. MD, MS. Comin. I S 2194s Bothriochloa saccharoides (Wats) Rydb. ssp reevesti (Gould) Allred & Gould. MS. Valles. Escasa. V4808. Bouteloua barbata Lag. MD, MS. Laderas y lomefos. Muy comin. Ca9, V5193. Bouteloua curtipendula eae) Torr. MD, MS, P. Valles y laderas. Comtin. Cal 2, V5333, R2171 Bouteloua gracilis (A.B.K.) Lag. MD, P. Valles y lomerfos. Comin. Ca7, V5352, 5433. Bouteloua hirsuta Lag. MS. Valles y arroyos. Comtin. Ca63, V5336, 5450. Bouteloua johnstonii Sallen. MD. Laderas y valles. Escasa. IMJ8491, 8751, R1483, V4173. Bouteloua ramosa Scribn. MD. Laderas y lomerfos. Comtin. Ca8, V5189. Bouteloua triftida Thurb. MD, MS. Valles y arroyos. Comtin. Ca6, V3628, 4812, VR2180. Brachiaria meziana Hitche. B, P. Valles y areas perturbadas. Escasa. VR2162. Brachypodinm pringlet Scribn. MS, MD. Laderas. Escasa. MCJ11705, 11705a. Bromus anomalus Ropr. MS, B. Valles y laderas. Escasa. Ca51, WR2167. VILLARREAL Q., Flora Vascular de la Sierra de la Paila 13) Buchloe dactyloides (Nutt.) Engelm. MD, B, P. Valles y laderas. Comtin. VR2163. Cenchrus incertus M.A.Curtis. MD, MS. Valles. Escasa. Ca21, WR2154. Cenchrus ciliaris L. MD. Orilla de caminos. Escasa. V5445. Cottea pappophoroides Kunth. MS, B. Valles y arroyos. Comtn. V5187. Cynodon dactylon (L.) Pers. MS, B. Arroyos. Escasa 3 Chloris gayana Kunth. MS. Arroyos y areas eeneurhadas: Escasa. Ca34. Chloris virgata Swartz. MD, MS. Areas perturbadas. Escasa. Ca33, V4503. Chloris submutica H.B.K. B. Areas perturbadas. Rara. VR2166. Dasyochloa pulchella (A.B.K.) Willd. MD. Valles. Escasa. Cal0. Dichanthelium oligosanthes (Schult.) Gould var. seriberianum (Nash.) Gould. B.MS. Arroyos y ones. Escasa. V3630—7, 5306, VR2181. Digitaria californica Mend ‘ Heard, MS, B. Valles y arroyos. Comtin. Ca3 Digitaria hitchcockii aa MS, B. Valles y arroyos. Comin. Ca36, ou. VR2175. Echinocloa colonum (L.) Link. MS. Arroyos. Escasa. Ca36, V3629, VR2169. Enneapogon desvauxii Daveau. MS, MD. Valles y laderas. Comtin. Ca59, V5231. Eragrostis barrelieri Daveau. MD, MS. Valles y arroyos. Comin. V5194, 5447. Eragrostis ciltanensis Link. MD. Valles y areas perturbadas. Escasa. Ca32. Eragrostis curvila (Schard.) Nees. MS. Valles y arroyos. Comuin. Ca29 Eragrostis intermedia Nees. MS, MD. Valles. Comin. Ca61, 27, V3626, 5324, 5383, WR2167. Eragrostis mexicana (Hornem.) Link. MS. Arroyos. Escasa. V5470a. Eragrostis palmert Wats. MS. Valles y laderas bajas. Esacasa. V4806. Eragrostis pectinacea (Michx.) Nees. MD, MS. Valles y arroyos. Comin. Ca30, V3630-3. Erioneuron avenaceum (H.B.K.) Tateoka MD, MS. Valles y laderas. Comin. MCJ11708a, V5 4 Erioneuron nealleyi (Vasey) Tateoka. MD, MS. Valles. Comtin. Ca3, V5443. VR2155. Sails pilosum (Buckl.) Nash. MS. Valles. Escasa. Cal3, V 9. Heteropogon contortus (L.) Beauv. MS, MD. Orilla de caminos. Comtin. Ca24, V5195. Hilaria mutica (Buckl.) Benth. P. Valles. Escasa. Ca25, V3988 Koeleria pyramidata (Lam.) Beauv. B. Laderas. Rara. Cal5. ree dubia (H.B.K.) Nees. MS, B. Valles y areas perturbadas. Comin. Ca5, W4507, 5449. Eres i sal (Schult.) Chase. Arroyos. Comun. Ca28, V528( Lycurus phleoides H.B.K. MD, P. Valles y laderas bajas. Comun. me V5330: Melica montezumae Piper. MS, B. Laderas rocosas. Escasa. MCJ11704. Maal msn (Scribn.) Conert. MS. Valles y arroyos. Comtin. Ca58, MCJ10108, 10514, ee dubia Fourn. B. Valles laderas. Escasa. — VR2184. Muhlenbergia emersleyi Vasey. MS, B. Laderas. Escasa. Ca Muhlenbergia glauca (Nees) Mez. MS, Q tees Escasa. 08 84. Muhlenbergia lindheimeri Hitchc. MS, B. Laderas bajas. Escasa. W5436. Muhlenbergia microsperma (DC.) Kunth. MS. Laderas y valles. Comun. Ca Muhlenbergia monticola Buckl. MS. Laderas. Comtin. Ca45, MCJ11705b, ae 11708. Muhlenbergia porter Scribn. MD. Valles y laderas bajas. Comtin. Ca46, V5446. Muhlenbergia pubigluma Swallen. MS, B. Laderas. Escasa. V5464. Muhlenbergia rigens (Benth.) Hitche. wee B. Laderas. Escasa. Cal8 Muhlenbergia rigida (H.B.K.) Kunth. MS, B. Valles y laderas. Comtin. Ca47, V5379. Muhlenbergia ag Vasey. MS, MD. Laderas. Muy comin. Cal9, V4171, 4811, 5378, 5468, VR218 Muhlenbergia sie ae (H.B.K.) Kunth. MS, P. Valles y laderas. Muy comtin. V3084, 4504, 4557, 5186, 5448. 136 SIDA16(1) 1994 Nassella leuchotricha (Trin. & Rupr.) Pohl. B, Ms. Valles y laderas bajas. Comtin. W5264, VR2165 Nassella tenuissima (Trin.) Barkw. B, P. Valles. Escasa. Ca31, W5276. Panicum — Vasey. MS. Arroyos y areas protegidas. Comtin. Ca43, MCJ11716a, V3640— 7) 2164. Panicum ae H.B.K. MS, P. Valles y lugares inundados. Escaso. Ca57. Pappophorum bicolor Fourn. MD. Valles y laderas Escasa. Ca38, V3630-5, 5431. Pappophorum vaginatum Buckl. MD. Valles. Escasa. Ca60. Piptochaetium fimbriatum (H.B.K.) Hitche. B, ve Comin. Ca52, V5281. Poa annua L. , B. Lugares hGmedos. Rara. Cal Poa involuta Hitchc. MS, B. Laderas protegidas. Rara. MCJ11706. oo Scoparinm (Michx.) Nash. var. se (Nash.) Gould. MS. Arroyos. Comtn. Ca50, V3987, 4809, VR2158, 2160. Sion beuiilius Phil. MD, P. Valles y ne bajas. Rara. Ca54. Setaria grisebachii Fourn. MS. Valles y arroyos. Comtin. Ca22, V5184, 5383, VWR2159. oe leucopila (Scribn. & Merr.) Schumann. MS. Arroyos. Comin. Ca20, MCJ10122, 716b, V5182, 5382, VR2182. ie scheelei (Steud.) Hitchc. B. Valles. Escasa. VWR2186. Sorghasthrum brunneum Swallen. MS, B. Laderas. Escasa. V5311, VR2161. Sorghum halepense (L.) Pers. MD. Areas perturbadas. Rara. V3630-2 : Sporobolus airoides (Torr.) Torr. P, MS, MD. Valles y arroyos. Comtin. Ca49, V5183. Sporobolus cryptandrus (Torr.) Gray. MS, MD. Valles. Comin. Ca48, V3631—1, 5246. Sporobolus pyramidatus ag ) Hitchc. Valles y arroyos. Escasa. Ca42. Sporobolus wrightii Munro. Valles. E Stipa alta Swallen. MS. De ae MCT! 1704a, V5185, 5432, VR2174. Stipa eminens Cav. MS, B. Valles y laderas. Comtin. Ca55, V5442, VR2172. Stipa lobata Swallen. MS. eee Escasa. VR2168. Stipa multinodis Beal. MS, B. Laderas. Comtin. MCJ11707a, V4502. Stipa neomextcana (Thurb. ) . oo MS. Valles y laderas bajas. Rara. Ca62. Tragus berterontanus Schult. MS. Areas perturbadas. Escasa. Ca35. gee muticus (Torr.) Nash. MS. Valles y arroyos. Con Cad: V4505, 4807, 5310, 2153. eae texanus (Wats) Nash. MS, MD. Valles y arroyos. Escasa. Cal4, V3630. AGRADECIMIENTOS Agradezco la colaboraci6n de M. A. Carranza y A. Rodriguez G. por su ayuda en las colectas de campo. A los curadores de los herbarios TEX y MEXU por permitirme consultar sus colecciones. La contribucién de J. Valdés R. al revisar las gramineas, M. C. Johnston por permitirme revisar sus listados de colecta, a B. L. Turner y G. Nesom por sus determinaciones y a T. Wendt por la revision detallada del manuscrito. REFERENCIAS BRANDEGEE, T.S. 1911. Plantae Mexicanae Purpusianae II. Univ. Calif. Publ. Bot. 4:177-194. Bravo-Ho tis, H. y H. SANCHEZ-MEJORADA. 1991. Las Cactdceas de México. II. Universidad acional Autonoma de México. Bremer, L. 1979. oe laui Sp. Nov. A new species from Coahuila, México. Cact. Succ. J. 51:278-2 VILLARREAL Q., Flora Vascular de la Sierra de la Paila Ie Cano, G.J. y J.S. MARROQUIN. ar Las gramineas de la Sierra de la Paila, Coah. Bol. Soc. Nuev. Hist. Nat. “J. E. Gonzz.” 1: 59-106. CorreELL, D.S., M.C. ane 1970 The manual of the vascular plants of Texas. Texas Research Botind: Renner, Tex CraBBE, J.A., A.C. JERMY a. 1 Mickel. 1975. A new generic sequence for the pterydophyte herbarium. Fern. Gaz. 11:141-162. ne A. 1981. An integrated system of classification of flowering plants. Columbia Press. New York. ae M. 1976: Systematic study of the genus F/owrensia (Asteraceae-Heliantheae). Ph. D. disertation. University of Texas, Austin, T FRYXELL, P.A. 1988. Malvaceae of Mexico. Syst. Bot. Monogr. 25:1—522. GLASS C and R. Foster. 1972. sae aguirreanus a new species from southern Coahuila, México. Cact. Succ. J. 44:80-81. GENTRY, H.S. 1982. Agaves of pores North America. Univ. of Arizona Press. Tucson, Ariz HENRICKSON, J. 1981. A new subspecies of Comarostaphylis polifolia (Ericaceae) form Coahuila, Mexico. Madrofio 28:33-37 HEnRICcKSON, J.and D. Fryr. 1985.8 ics of Leucophyll. 1 Eremog Scrophul aceae) Sida 11:10—172. Hess, W.S. and J. HENRICKSON. 1987. A taxonomic revision of Vawquelinia (Rosaceae). Sida 12:101-163. INEGI (INSTITUTO NACIONAL DE ESTADISTICA, GEOGRAFIA E INFORMATICA). 1983. Sintesis Geografica de Coahuila. Secretaria de Programacién y Presupuesto. México D IRWIN, H.S. and R.C. Barnepy. 1975. Notes preliminary to an account of Cassa in the Chihuahuan Desert. Sida 6:7-18 ea I.M. 1944. Plants of Coahuila, eastern Chihuahua, and adjoning Zacatecas and Durango. III. J. Arnold Arbor. 25:43-83. ee L A. 1975. Revision of the Rhamnus serrata complex. Sida 6:67—79. JOHNSTON M.C. 1983. Thamnosma pailensis ee new species form the Sierra de la Paila, Coahuila, México. Phytologia 53:1 80. KiNG, R.M. and H. Rosinson. 1987. The genera Be eaiess (Asteraceae) Missouri Bot. Gard. Monogr. Syst. B McDonalp, J.A. 1987. "Three new species of Convolvulaceae from northeast México. Brittonia 39:106-111 Mutter, C.H. 1947. Vecetariva and climate of Coahuila, México. Madofio 9:33—57. Nesom, G.L., Y. SuH., D.R. MorGAn and B.B. Simpson. 1990. Xylothamnia eee ere a new genus related to Exthamia. Sida 14:101-116. Pinkava, D.J. 1979.1981. Vegetation and flora of the Bolson of Cuatro Cienegas Region, Coahuila, México. Bol. Soc. Bot. Mex. 38:35—73 (part I); 39:107—127 (part ID); 41:127- 151 (parc III). REVEAL, J.L. and R.M. King. 1973. Re-establishment of Acourtia D. Don (Asteraceae). Phytologia 27:228—232. Robinson, B.L. 1930. The stevias of North America. Contrib. Gray Herb. 90:90—160. Sousa-S.M. 1969. Las colecciones botanicas de C. A. Purpus en México. Periodo 1898— 1925. Univ. Calif. Press. Berkeley, Calif. TuRNER, B.L. 1975. Taxonomy of Haploestes (Asteraceae— Senecioneae). Wrightia 5:108—115. TuRNER, B.L. 1978. Taxonomic study of the scapiform species of Acourtia (Asteraceae- mutisieae). Phytologia 38:456—468. TuRNER, B.L. 1982. Revisional treatment of the Mexican species of Seymeria (Scrophulari- aceae). payelonia 51:403—-422 138 SIDA16(1) 1994 VILLARREAL -Q., J.A. 1987. Coutaportla pailensis (Rubiaceae), a new species from Coahuila, ». Sida 12:223-225. Vin a . 1992. Dos nuevos taxa de para el norte de México. Acta Bot. Mex. 20: VILLARREAL-Q.,J.A. y J. VALDES-R. 1992-1993. ee de Coahuila. Revista Manejo de Pastizales 6(1— 2): a 18. WenHBE-D., J.A. 1985. Estudio Floristico de las cactaceas s del municipio de Ramos Arizpe, Coahuila, México. Tesis, VANE, Saltillo, Coah. M ZANONI, T.A. and R.P. ADAMS. 1980. The genus es sup ressaecar) in ie and Guatemala: synonymy, key and distribution of the taxa. Bol. Soc. Bot. Mex 83- cere Mimosa (Leguminosae:Mimosoideae) VASCULAR PLANT TYPE SPECIMENS IN THE UNIVERSITY OF GEORGIA HERBARIUM, WITH A BRIEF HISTORY OF THE HERBARIUM MICHAEL O. MOORE and DAVID E. GIANNASI Department of Botany University of Georgia Athens, GA 30602, U.S.A ABSTRACT The University of Georgia Herbarium contains 162 vascular plant specimens that have been designated as types, and for which some form of documentation has been found. These types are catalogued here with label data (locality, date, collector), the kind of type, and a citation of the original publication. RESUMEN EI Herbario de la pave rsidad de Georgia contiene 162 especimencs de plantas vascular ares id tipo os y de los cuales se I que han : Estos tipos estan se ae aqui con datos de etiqueta (loci alidad: fecha, colector), especificaci6n concreta del tipo, y una cita de la publicacién original. The herbarium at the University of Georgia was founded in the 1920s by John N. Reade, and underwent its first reorganization by Joseph H. Pyron and Rogers McVaugh in the 1930s. It was expanded under the direction of Wilbur H. Duncan who served as curator until his retirement in 1978. Samuel B. Jones, Jr., was appointed curator in 1978 and Nancy C. Coile served as assistant curator. During 1978-1980, the herbarium was reorga- nized and its scope and activities broadened. In addition, the herbarium library was formally established. In 1981, Samuel B. Jones was named Direc- tor of the herbarium and Nancy C. Coile became the curator. Michael O. Moore became curator in 1989, and David E. Giannasi became Director in 1991. The University of Georgia herbarium serves as a repository of samples of southeastern U.S., with emphasis on the flora of the state and has as one of its major goals elucidating the flora of Georgia. During 1979, the her- barium staff prepared check-lists for the floras of Clarke County and the state of Georgia (Jones and Coile 1979a, 1979b). In 1985, the 2nd edition of the Georgia Plant List was issued (Coile and Jones 1985). In 1985-1986, an elec- tronic high density mobile storage system (compactor) was installed in the herbarium. This compactor system almost doubled the storage capacity of the herbarium and will allow for continued growth of the collection. In 1988, Jones and Coile published The Distribution of the Vascular Flora of Georgia (Jones and Coile 1988). This publication has distribution maps by SIDA 16(1): 139 — 151. 1994 140 SIDA16(1) 1994 counties for all native or naturalized plant species occurring in Georgia based upon voucher specimens deposited in institutional herbaria. Although 5,000 copies were printed, demand for this publication has been great, and it is now out of print. The publication es an important next step toward the development of a state Flor The herbarium now holds over 206 (000 mounted specimens, of which 162 have been documented as types. This latter number does not include topotypes, paratypes, or specimens suspected as being types, but for which no literature citation could be found. A number of these specimens are of historical significance, particularly collections of 22 taxa made by A.H. Curtiss, a prominent early collector in the southeastern United States from 1884-1899 Catalogued specimens were documented in at least one three ways: (1) through citation in the original publication; (2) through citation as a type in a secondary publication, and (3) through annotation as a type by a later researcher. Where the original or a secondary publication was consulted, the location of the holotype is also given. The label data (locality, date, collector, collection number), the reference citation, the kind of type, and source(s) of documentation are given for each specimen. The following catalogue is divided into four sections: Pteridophyta, Gymnospermae, monocotyledons and dicotyledons. Within each section, taxa are listed alphabetically by genus. CATALOGUE OF TYPE SPECIMENS PTERIDOPHYTA Isoetes tagetiformans P.M. Rury, Amer. Fern J. 68: 99-108. 1978. IsorypE. U.S.A, GeorGiIA. Columbia Co.: Heggies Rock, 3.8 mi E from Columbia Junior High School along GA 232 and County Rd 2122, 33° 32'30" N, 82° 15! 05" 7 21 Jan 1978, PM. Rury & M. Treiber 259; cited in original publication esse NCU). Thelypteris pilosa (M. Martens & Galeotti) Craw. var. alabamensis Craw., Amer. Fern J. 41:15-20. 1951. synrype. USA. howe Winston Co.: in fissures a Pottsville Sandstone on W fork of Sipsey River, 5 mi E of Double eee 26 Nov 1949, L.C. Crawford & C.B. Segars 1103; cited in original publication Woodsia scopulina D. ene ree laurentiana Windham, Contr. Univ. Michigan Herb. 19:59. 1993. isoryPE. CANADA. QueBec. Gaspe Co.: on sendscone sea-cliffs, 19-21 Aug 1905, Collins & Fernald ats annotated as isotype by M.D. Windham, 1993. GYMNOSPERMAE Podocarpus lucienti Laubenf., Brittonia 12:80. 1960. isorypE. NEW CALEDONIA: scattered and locally common on slopes above and N of Riviere Bleue River, 5 mi from confluence with Yate River, 10 Oct 1957, D.J. de Laubenfels P137; cited in original publication (HOLOTYPE: GH). ANGIOSPERMAE MONOCOTYLEDONS Aletris lutea Small forma albiflora E.T. Browne, Rhodora 63:305. 1961. HOLOTYPE. U.S.A. Georcia. Chatham Co.: low ground at edge of Cypress stand 15.3 mi W 4° S of MOORE AND GIANNASI, Type specimens, University of Georgia Herbarium 141 Savannah City Hall, NW of Old Ogeechee Canal on Bamboo Farm-Pembroke Road, 13 Jun 1958, WH. Duncan 20982; cited in original publication. Allium passeyi N.H. Holmgren & A.H. Holmgren, Brittonia 26:309. 1974. IsoTyPE. U.S.A. Uran. Box Elder Co.: SW part of Howell Valley, TL1N, RS W,S7, 13 Jun 1960, A.H. Holmgren, H.B. Passey & V.K. Hugie 13125, cited in original publication (HOLOTYPE: NY). ndropogon brachystachyus Chapm., Fl. S U.S. ed. 2:688. 1883. ISOLECTOTYPE. U.S.A. Froripa: dry pine barrens near Jacksonville, Oct no year, A.H. Curtiss 3632; annotated as isolectotype by C.S. Campbell, 1980 Andropogon virginicus L. var. eiaeus Hack. in DC., Monogr. Phan. 6:411. 1889. isoTyPE. U.S.A. FLroripa: dry pine barrens near Jacksonville, Nov no year, A.H. Curtiss 36386; annotated as isotype by C.S. Campbell, 1980. istida lanuginosa Scribner ex Hitchc., Contr. U.S. Natl. Herb. 17:278. 1913. ISOTYPE. MEXICO. JALISCO: hills near Guadalajara, 28 Oct 1889, C.G. Pringle 25 cited in original publication (HOLOTY S). alochortus rhodothecus Clokey, Bull. S. Calif. Acad. Sci. 37:1. 1938. isotyPe. U.S.A Nevapa. Clark Co.: Charleston Mountains, Charleston Park, 21 Jul 1937,1.W. Clokey 7479; cited in original publication. Carex X absconditiformis Fernald, Rhodora 44:387. 1942. isotype. U.S.A. VIRGINIA. Sussex Co.: bottomland woods along Nottoway River, E of Huske, 13 Jun 1941, M.L. Fernald & B. Cie 12969; cited in original publication, annotated as isotype by C.T. Bryson, 1980 (HOLOTYPE: GH). Carex amplisquama F.J. Herm., Rhodora 57:157—158. 1955. tsorype. U.S.A. GEORGIA. Gilmer Co.: 8 mi E of Chatsworth, 15 May 1938, J.H. Pyron & R. McVaugh 2951; cited in original publication (HOLOTYPE: Carex austrocaroliniana P.M. Bailey: Am. J. Sci. 45:173. 1843. tsorype. U.S.A. CAROLINA AND GeorGIA: in montibus, no date, S.B. Buckley s.n.; annotated as isotype by C.T. Bryson, 1980. Carex digitalis Willd. var. ees © Fernald, Rhodora 43:544. 1941. isotype. U.S.A VIRGINIA. Southampton Co.: steep wooded banks, ravines and clearings near Three Creek, NW of Applewhite Church, g May 1940, M.L. Fernald and B. Long 11791; cited in original publication, annotated as isotype by C.T. Bryson, 1980 (HOLOTYPE: GH). Carex interior FM. Bailey var. charlestonensis Clokey, Bull. S$. Calif. Acad. Sci. 38:1. 1939. 1soryPE. U.S.A. Nevapa. Clark Co.: Charleston Mountains, Charleston Peak, damp soil by ne me 1937,1.W. Clokey 7468; annotated as isotype by A.A. Reznicek, 1978. Car um Willd. ex Link var. austrolucorum J.H. Rettig, Sida 13:449-450. 1989. HOLoTYPE. U.S.A. NortH Carouina. Buncombe Co.: along NC 9, 5.0 mi N of jet. with US 74 in town of Bat Cave, 6 May 1986, J.H. Rettig 1554; cited in original publication, annotated as holotype by J.H. Rettig, 1988. ar rea aay Castanea 50:15-17. 1985. HOLOTYPE & IsoTYPE. U.S.A. Nor eee Macon Co.: Coweeta Hydrologic Lab, Stewart Trail, 17 May 1981, J).R. Manhart ae annotated as sole and isotype by C.T. Bryson, 1985. ar ohriana Mackenzie, N. Amer. Fl. 18:106. 1931. isorype. U.S.A. FLoripa: swamp near Wauchula, 15 Apt 1901, A.H. Curtiss 6761; annotated as isotype by A.A. Reznicek, 1978. Carex shortiana Dewey, Mss. Dewey, Car. XXX, p. 60. isotype. U.S.A. KENTUCKY: Lexington, no date, Short s.n.; annotated as isotype by 8. McDaniel, 1971 Cyperus granitophilus McVaugh, Castanea 2:103. 1937. HoLoTyPE. U.S.A. GEORGIA. Walton Co.: sandy depression in granite rocks, 1 mi E of Loganville, 3 Oct 1936, ].H. Pyron & R. McVaugh 971; cited in original publication. Eragrostis chiquitaniensis T. Killeen, Ann. Missouri Bot. Gard. 77:153. 1990. IsoTyPE. ~ ou 142 SIDA1G6(1) 1994 BOLIVIA. Santa Cruz, Chiquitos, Est. San Ignacio, 22 km N of San Jose, 17° 35'S, 60° 45! W, 1 Feb 1986, T. Ki/leen 1728; cited in original publication. Erythronium umbilicatum C. Parks & Hardin ssp. monostolum C. Parks & Hardin, Brittonia 15:254. 1963. isoryee. U.S.A. NorrH Carona. Swain Co.: rich woods near Indian ai 6 May 1961, J.W. Hardin 2360, cited in original publication eae NCSC). ythronium umbilicatum C. Parks & Hardin ssp. umbilicatum, Brittonia 15:252— o on IsoTyPE. U.S.A. NorTH CAROLINA. ide ll Co.: along zee iow stream, 9 m W of Statesville off 140, 13 May 1961, J.W. Hardin 2368, cited in original Sle ees CSC}. sey ales (Retz.) Vahl ssp. diffusa D.B. Ward, Castanea 33:127. 1968. isoTyPE. U.S.A. Froripa. Alachua Co.: FL 121, Williston Rd., ca. 4 mi SW of Gainesville near ioe corner ae Paynes Prairie, 13 Ang 1966, D.B. Ward 5867; cited in original ae eee PE: F enaria ble phar slots (Willd.) Hook. var. mea Ne Bot. Mus. Leafl. 9: a 156. 1941. isorype. U.S.A. Kentucky. McC Co.: in Sphagnum bog with Xyris 3 mi N of Whitley City, 27 Aug 1940, FE r McFarlend & HJ. Rogers 97; cited in original publication (HOLOTYPE: AMES). Hosta jonesit M.G. Chung, Ann. Missouri Bot. Gard. 76:920. 1989. HOLOTYPE. KOREA: KYEONGSAN NAM Do Province: Namhae Gun, Sangu Myeon, Mt. Kumsan, 28 Aug 1988, M.G. Chung 1613; cited in original publication. Hypoxis curtissii Rose in Small, Fl. SE U.S. 287, 1329. 1903. isoryps. U.S.A. FLorIpaA: swamps near - ksonville, 19 May 1894, A.H. Curtiss 4727; annotated as isotype by A. Herndon, 199¢ Najas major vn var. gracilis Morong, Bor. Gaz. 10:255. 1885. 1sorype. U.S.A. FLoripa: Palm Creek near Cape Romano, Jul no year, A.H. Curtiss 2705; annotated as isotype by R.R. Haynes, 1974. olina atopocarpa Bartlett, Rhodora 11:81. 1909. Isorype. U.S.A. FLoripa: Indian River, Eau Gallie, no date, A.H. Curtiss 5702; cited in original publication (HOLOTYPE: GH), Rhynchospora curtissii Britton ex Small. FlcSE U3:,195,.1 527: 19035 sOrrre: WS, FLORIDA: ie res near Milton, 8 Jul 1897, A.H. Curtiss 5929; annotated as isotype by S.M. Kooyman, Rhynchospora Sie A. Gray var. breviseta Gale, Rhodora 46:128—-130. 1944. isoTyPE. U.S.A. FLoripa: damp pine barrens near Jacksonville, 19 Jun 1896, A.H. Curtiss 5687; cited in original publication, annotated as isotype by $.M. Kooyman, 1980 (HOLO- TYPE: GH). pe secundifolia Kral, Brittonia 34:12—13. 1982. isorypr. U.S.A. ALABAMA, DeKalb Co.: cracks in sandy Aaviocle shallows, sandy shallow pools and depths up to 0.75 m even at a stage, Little River above AL 35 bridge, 27 Jul 1977, R. Kra/ 60639; cited in original publication (HOLOTYPE: US). oO Scirpus rubiginosus Beetle, Amer. J. Bot. 28:697. 1941. tsorype. U.S.A. CALIFORNIA. Marin Co.: tide flat N of Stinson Beach, 9 May 1941, A.A. Beetle 26806; cited in original publication. Sporobolus teretifolius R.M. Harper, Bull. Torrey Bot. Club 33:229-231. 1906. IsoTyPE. U.S.A. GeorGia. Colquitt Co.: moist pine barrens near Moultrie, 20 Sep 1902, R.M. Harper 1642, cited in original publication (HOLOTYPE: Trillium decipiens J.D. Freeman, Brittonia 27:17-18. 1975. ISOTYPE. U.S.A. GEORGIA. Early Co.: common in rich woods on W facing bluff of Chattahoochee River, just below Tenneco Oil Depot and S of GA 162, 28 Mar 1968, J.D. Freeman 615; cited in original publication, annotated as isotype by J.D. Freeman, 1968 (HOLOTYPE: GH). MOORE AND GIANNASI, Type specimens, University of Georgia herbarium 143 Trillium foetidissimum J.D. Freeman, Brittonia 27:31—32. 1975. isorypr. U.S.A. Mussissipp1. Adams Co.: in loess ravines under rich hardwoods S of US 61-84-98, 1.2 miE St. Catherine Creek in Natchez, 30 Mar 1967, J.D. Freeman 535; cited in original publication, adnotated as isotype by J.D. Freeman, 1968 (HoLoTyPE: GH). Raf., Autik. bot.:132. 1840. isoNEoTYPE. U.S.A. Floripa. Gadsden Co.: moist ines in woods, Chattahoochee, 14 Mar 1901, A.H. Curtiss 6747; cited as isoneotype by J.D. Freeman, 1975, Brittonia 27:10 (NEOTYPE: GH). rillium persistens W. Duncan, Rhodora 73:244—246. 1971. isoryPe. U.S.A. GEORGIA. Rabun Co.: open deciduous woods on S facing slope of Tallulah Gorge, 23 Mar 1971, WH. Duncan 23548; cited in original publication (HoLoTyPE: US). rillium reliquum J.D. Freeman, Brittonia 27:21. 1975. tsorype. U.S.A. GrorGIA. Columbia Co: (Richmond Co. on label and original publication, but Freeman says Columbiz oe rich woods on bluff slope at mouth of ravine, E facing bluff of Savannah River, ca. } mi ve Augusta, 30 Mar 1968, J.D. Freeman 625; cited in original publication, toe) as isotype by J.D. Freeman, 1967 (HOLOTYPE: ). Yucca smalliana Fernald, Rhodora 46:8. 1944. isorype. U.S.A. FLoribDA: sandy soil near Jacksonville, May no year, A.H. Curtiss 2950, cited in original publication (HOLOTYPE: GH). DICOTYLEDONS Agalinis acuta Pennell, Bull. Torrey Bot. Club 42:338. 1915. tsorype. U.S.A. Massa- CHUSETTS. Martha's Vineyard, Edgartown, dry sandy downs, 12 Sep 1901, M.L. Fernald 45; cited in original publication, annotated as isotype by J.M. Canne, 1985 (HoLorype: US) Agalinis edwardsiana Pennell, Proc. Acad. Nat. Sci. Philadelphia 73:522. 1922. ISOTYPE. U.S.A. Texas. Kendall Co.: dry adobe hills 5 mi NE of Boerne, 16 Sep 1913, EW. Pennell 5482; cited in original publication (HoLoTypE: PENN). Agalinis homolantha Pennell, Proc. Acad. Nat. Sci. Philadelphia 73:525. 1922. IsoTyPE. U.S.A. Texas. Colorado Co.: sandy oak woodland NW of Sheridan along SA & AP railroad, 21 Sep 1913, FW. Pennell 5522; cited in original publication (HOLOTYPE: PENN). Amorpha georgiana Wilbur, Rhodora 56:261—263. 1954. isorype. U.S.A. GEORGIA. Telfair Co.: 3 mi NW of Lumber City on US 23, dry savannah and river bank, 9 May 1953, R.L. Wilbur 3158; cited in original publication (HOLOTYPE: GH). Amorpha ouachitensis Wilbur, Rhodora 77:394—397. 1975. tsorype. U.S.A. OKLA- HOMA. LeFlore Co.: dry rocky (sandstone) hills, Page, 23 Sep 1921, E.J. Palmer 20572; cited in original publication, annotated as isotype by R.L. Wilbur, 1973 (HOLOTYPE: Amorpha virgata Small, Bull. Torrey Bot. Club 21:17. 1894. IsoLEcroryPE. U. S.A. GeorGia. DeKalb Co.: NW slope of Stone Mountain, 3 Jul 1893, ].K. Small s.n.; cited as isolectotype by R.L. Wilbur, 1977, Rhodora 77:398 (LECTOTYPE: ). — eae Tidestrom ssp. al Ss Clokey, Bull. S. Calif. Acad. Sci. . 1939. 1sorypPe. U.S.A. Nevaba. Clark Co.: Charleston Mountains, Charleston Peak, ene rocky slope ieee timberline, 16 Jul 1936, LW. Clokey 7094; cited in original publication. renaria | Mic hrb. var. longipedunculata W. Duncan, Phytologia 3:282.1950. 1sorypE. U.S.A. ce a ton Co.:S side of Chattahoochee River, E of Marsh Creek, 5 Jun 1949, WH. Duncan 9701; cited in original publication (HOLOTYPE: NY). Aster divaricatus L. var. deltoideus E.S. Burgess in Britton & Brown, Ill. Fl. N US. 3:357. 1898. tsoLectotyPe. U.S.A. Disrricr oF CoLuMBIA. Opposite Analostan, thickets, 27 Sep 1888, E.S. Burgess s.n.; cited as isolectotype by W.F. Lamboy & A.G. Jones, 1987, Brittonia 39:290 144 SIDA16(1) 1994 Aster erectus E.S. Burgess, Mem. Torrey Bot. Club 13:147. 1906. isotecrorype. U.S.A. New York: vicinity of New York City, Bryn Mawr Park, 26 Sep 1896, E.S. Burgess 5.n.; cited as oo by W.F. Lamboy & A.G. Jones, 1987, Brittonia 39:290. er Ie Lamboy, Syst. Bot. 13:192— ee 1988. HOLOTYPE & IsoTyPE. U.S.A pee Gta. Upson Co.: oak-hickory woods at E base of Pine Mountain, near Pasley Shoals, Ww of Thomaston, cf Sep 1947, A. Cronquist 4694; ce in original publication, annotated as holotype and isotype by W.F. Lamboy, 1988 Aster roscidus E.S. Burgess in Britton & Brown, Ill. Fl. N U.S. 3:360. 1898. IsoLectotyPe. U.S.A. New York: Bryn Mawr Park, 14 ie 1897 (18987), E. : Burgess $0.5 cited as isolectotype by W.F. Lamboy & A.G. Jones, 1987, Brittonia 39:290. Aster tenuifolius L. var. aphyllus R. Long, Rhodora 72:40—41. 1970. IsoryPs. U.S.A. FLoripa. Hillsborough Co.: NW of Tampa, S of FL580 and W of Rocky Creek, 24 Dec 1962, 0. Lakela 25610, cited in original publication, annotated as isotype by S. Sundberg, 1986 ). — (HOLOTYPE: Astragalus aequalis C 6:215. 1942. sorype. U.S.A. Nevapa. Clark Co. Soaaie Mountains, Seen belt, Harris Springs Road, 4 Jun 1937, 1.W. Clokey 572; cited in original publication. beeen hemigyrus Clokey, Madrofio 6:220. 1942. isotype. U.S.A. Nevapa. Clark Co.: Charleston Mountains, rock ledges, Larrea belt, S of Indian Springs, 18 Apr 1939, 1. W. Clokey 8409, cited in original publicatio Baptisia arachnifera W. Duncan, ieee 46:29—31. 1944. sorypr. U.S.A. GEORGIA. Wayne Co.: sandy soil in open pine woods 10 mi S of Jesup, 4 Jul 1943, WH. Duncan 5693; cited in original publication (HOLOTYPE: GH). aptisia pendula Larisey, A Missouri Bot. Gard. 27:171. 1940. tsorype. U.S.A. GeorGia. Decatur Co.: dry re woods bordering the Flint River near Bainbridge, 19 Jul 1901, A.H. Curtiss 6810; annotated as isotype by D. Isely, 1978. Bidens mariana S.F. Blake, Rhodora 31:88. 1929. tsorype. U.S.A. MARYLAND: sandy enters Point, Charleston, 17 Sep 1926, S.F Blake 9698; S). shore of Northeast River near Car —— In original publication (HOLO’ TYPE: gelowia nuttallii L.C. Anderson, Sida 3:460—461. 1970. isorype. U.S.A. GEorGIA. mine Co.: W bank of Ohoopee River 3.5 mi W of Reidsville, 12 Nov 1969, L.C. Anderson 3455; cited in original publication (HOLOTYPE: KSC). Bumelia thornei Cronquist, Castanea 14:103. 1949. HoLorype. U.S.A. Groraia. Early Co.: in dry live oak woods by cypress swamp, | mi E of Nantz Spring, 22 Oct 1947, R.E Thorne 7345; cited in original publication. Calycanthus brockiana Ferry & Ferry f., Sida 12:339. 1987. HoLoTYPE. U.S.A. GEORGIA. Lumpkin Co.: 12.8 mi SW of Dahlonega, S of E-W dirt farm road in ee deciduous forest, 10 May 1982, R.J. Ferry Sr. 637, annotated as holotype by R.J. Ferry & R.J. Ferry f., 1987 Castilleja christit N.H. Holmgren, Bull. Torrey Bot. Club 00:91. 1 973. 1soTyPe. U.S.A, IpaHo. Cassia Co.: Cache Peak Range, Harrison Mountain, near top, T13S, R24E, $9, 12 Jul 1966, N.H. Holmgren & J.L. Reveal 28606; cited in original publication (HOLOTYPE: NY). Castilleja clokeyi Pennell, Proc. Acad. Nat. Sci. Philadelphia 89:420. 1938. IsoTyPE. S.A. Nevaba. Clark Co.: Kyle Canyon, gravelly loam on N slope, 8 Jul 1936, 1. W. Clokey & C.B. Clokey 7322; cited in original publication Cirsium clokeyi S.F. Blake, Proc. Biol. Soc. Wash. 49:8. 1938. 1sorypE. U.S.A. NEVADA. Clark Co.: Charleston Mountains, ridge ne _ Lael open slopes, 6 Aug 1937, 1.W. a 7456, cited in original ae (voi US). Clematis morefieldii Kral, Ann. ne yt. a rd. 74:665. 1987. isoryee. U.S.A ALABAMA. Madison Co.: SE oe ae upslope side of eastward unpaved extension as MOORE AND GIANNASI, Type specimens, University of Georgia herbarium 145 of Deborah Ave., 0.75 mi SSW of Red Top Mountain, 17 Jun 1983, R. Kral 70176; cited in original publication. lematis socialis Kral, Rhodora 84:287—289. 1982. isotype. U.S.A. ALABAMA. St. Clair Co.: silt and ae of alluvial openings in bottoms of Dry Creek and contiguous hardwood forested edges, ca. 3 mi S of Ashville, 2 May 1980, R. Kral 64852; cited in original publication ole LS). Clitoria fragrans Small, Torreya 26:57. 1926. isorypr?. U.S.A. Floripa. Highlands Co.: sandhills near DeSoto City, 20 May 1925, J.K. Small & E.T. Wherry 12626; note on label states that Gleason could not find the type at NY in 1945, annotated as possible isotype by ie Bees 812 Cordylanthus glandulosus Se epee Proc. Acad. Nat. Sci. Philadelphia 90:11. 1938. isorype. U.S.A. Nevaba. Clark Co.: Charleston Mountains, brushy hills, Juniper Belc, Harris Springs Read. 16 Jul 1937, 2. W. Clokey 7715; cited in original publication. Coreopsis leavenworthit T. & G. var. curtissii Sherff, Bot. Gaz. 94:592. 1933. IsoTyPE. U.S.A. Froripa: dry pasture ground along the Suwannee River near Branford, 23 Oct 1900, A.H. Curtiss 6734; annotated as isotype by E.B. Smith, 1976. Coreopsis a a | var. godfreyi Sherff, Bot. Leafl. 6:3. 1952. HoLoTyPE. U.S.A. Georaia. Wheeler Co.: bogg easacieet arene in sandhills 7 mi S of McRae, 14 Oct 1950, .R.K, pee 50793; sanotareda ype” by E.E. Sherff, no date Coreopsis saxicola Alex. var. duncanii ‘ herff, Bot. Leafl. 6:2. 1952. HoLtorypE. U.S.A. GEORGIA. Pee Co.: shallow soil at edge of thickly vegetated drain at Echols Mill, SE of Point Peter, 4 Aug 1941, WH. Duncan 3832; annotated as “type” by E.E. Sherff, no date. Couepia dolichopoda Prance, Brittonia 26:302. 1974. 1sorype. PERU. Depr. oF Loretro: Varadera de Mazon from Rio Amazonas to Rio Napo, primary forest, 21 ‘Aagil O12, T.B. Croat 19382; cited in original publication (HoLoTYPE: NY Dedeckera cireleneis Rev. & J.T. Howell, Brittonia 28: 246-248. 1976. ISOTYPE. U.S.A. Cauirornia. Inyo Co.: Last Chance Range in a rock canyon, ca. 3 air mi SE of Eureka oe sand dunes and 3.5 air mi NW of Marble Valley VABM 7559 in T10S, R40E, 29 Jul , JL. Reveal, MAC. DeDecker G PW. DeDecker 3909; cited in original publication ee US). Dicerandra frutescens Shinners, Sida 1:89—90. 1962. IsorypE. U.S.A. FLoripa. High- lands Co.: 20 mi S$ of Sebring along Route 27, 28 Aug 1953, FH. Sargent 6600; cited in original publication, annotated as isotype by R.B. Huck, 1984 (HoLorypE: SMU). ipsacus sylvestris Hudson f. albidus Steyerm., Rhodora 60:175. 1958. IsoryPeE. U.S.A. ILLINots. Cook Co.: undeveloped section, Mt. Hope Cemetery, Chicago, 115th Street, 25 Aug 1957, K.E. Bartel s.n.; cited in original a. (HOLOTYPE: F). Enceliopsis nudicaulis (A. Gray) Nelson var. corrugata Cronquist, Bull. Torrey Bot. Club 99:246. 1973. 1soryPe. U.S.A. Nevapa. Nye Co ree hard limestone in Larrea desert about 15 (airline) mi SE of Lathrop Wells, T17S, R51E, about $20, 20 Apr 1966, A. Cronquist 10648; cited in original publication (HOLOTYPE: NY Eriogonum aliquantum Rev., Phyrologie 34:460-461. 1976. isorypr. U.S.A. New Mexico. Colfax Co.: 0.3 mi W of NM 21, 4.4 mi S of US 64 at Cimmaron on the Philmont Scout Ranch, 14 ul nae J L. Reveal 277; cited in original publication (HoLoTyPE: US). g 254. 1989. isotype. U.S.A. IDaHo. Custer Co Edict Pass, upper end of Gesues Basin ona ridge SW of the pass, 11 Jul 1975, /.L. ee 3876; cited in Be Ger publication Eriogonum lewisii Rev., Great Basin Nat. 45:277. 1985. isotype. U.S.A. Nevapa. Elko Co.: White Elephant a S of Elk Mountain, $4, T46N, R16E, 30 Jul 1976, J.L. Reveal 4596; cited in original publication. § Phelps a a 146 SIDA16(1) 1994 Eriogonum ochrocephalum S. Watson var. alexanderae Rev., Great Basin Naturalist 45:276. 1985. isorypr. U.S.A. Nevapa. Lyon Co.: along NV 3 in Wilson Canyon, between Smith and Mason, 12.8 mi NE of Smith, 21 Jun 1978, J.L. Reveal 4737; cited in original publication. Eriogonum ovalifolium Nutt. var. pansum Rev., Phytologia 66:259. 1989. IsoTyPE. U.S.A. Ipano. Boise Co.: along ID 21 ca. 12.5 mi S of Lowman, near West Ford Creek at ats 59.5, 12 Jul 1975, J.L. Reveal & BJ. Ertter 3883, cited in original publication. horbia gracilior Cronquist, Castanea 14:102—103. 1949. isorype. U.S.A. GEORGIA. ae Co.: sandhills 3 mi NW of Reidsville, 16 Jun 1948, A. Cronquist 5334; cited in CS aeere annotated. as es by M.J. Huft, 1979 (HoLorype: US). veria linearis Lag. var. latifoliaJ.R. aie Proc. Amer. Acad. Arts 39:289. 1903. syNntTyPE. U.S.A. Froripa. Palm oe Co.: shore of Lake Worth near Palm Beach, 31 Aug 1895, A.A. Curtiss 5524; cited as type in original publication, annotated as type by A.M. Powell, no date. Forestiera es ens Nutt. var. glabrifolia Shinners, Field & Lab. 18:99. 1950. ISOTYPE. U. . Texas. Bosque Co.: thin soil over limestone, 12.5 mi NNE of Walnut ek 26 an 1949, L.H. Shinners 11331; annotated as isotype by C.J. Brooks, 1975. Galactia minor W. Duncan, Phytologia 37, 59-60. 1977. HoLorypPeE. U.S.A. GEORGIA. Long a sandhills adjacent to Alramaha River aa SW of Ludowici, 2 Aug 1953, WH. Duncan 16993; cited in original publication, annotated as holotype by W.H. Duncan, 1977 alesia tetraptera J. Ellis, original publication unknown. ISONEOTYPE. SOUTH CaROUINA. Williamsburg Co.: foodplain on N bank of the Santee River along SC 17, 0.6 mi N of the Berkeley Co. line, about 2.7 mi N of Jamestown, 6 Apr 1975, J.L. Reveal & M.J. Seldin 3532: cited by J. Reveal and M. Seldin, 1975, Taxon 25:135. Helenium floridanum Fernald, Rhodora 45:494. 1943. isorypr. U.S.A. FLoripa,. Py Co.: low open woods near Fitzgerald, 14 Jun 1900, A.H. Curtiss 6663; cited in original publication, annotated as isotype by W.M. Bierner, 1971 (HOLOTYPE: GH). Hexastylis shuttleworthii (Britton & Bs Small var. harperi Gaddy, Sida 12:54. 1987. isotype. U.S.A. Georaia. Madison Co.: acid bog just N of GA 106, 14.2 mi NE of Athens, 9 May 1986, L.L. Gaddy s.n.; cited in original publication (HOLOTYPE: CLEMS). Hoita hirsuta Rusby, Mem. New York Bot. Gard. 7: 261. 1927 Isorype. BOLIVIA: tae 12 Jul 1921, O.E. White 154; annotated as isotype by i Grimes, 1990. noxys turneri K.F. Parker, Phytologia 20:192. 1970. tsorype. U.S.A. TEXas. Karnes . limestone soil along roadside 2 mi N of Karnes City, 21 Apr 1965, B.L. Turner 5154; cited in original publication (HOLOTYPE: US) Hypericum adpressum Barton var. spongiosum B. Robinson, Rhodora 4:136. 1902. isoryPe. U.S.A. MassacHusetts. Barnstable Co.: sandy margin of Flax Pond, Bourne, 15 Sep 1901, G.G. Kennedy, E EF. Williams & M.L. Fernald 234; cited in original publication (HOLOTYPE: ). Hypericum canadense L. var. galiiforme Fernald, Rhodora 49:154—155. 1947. Iso- Type. U.S.A. VirGINIA. Sussex Co.: sandy and peaty shore of Airfield Millpond, SW of Wakefield, 11 & 12 Sep 1945, M.L. Fernald & B, Long 14962; cited in original publication (HOLOTYPE: GH). Hypericum exile P. Adams, Contr. Gray Herb. 189:33. 1962. isorype. U.S.A. FLoripa. Gulf Co.: 2.5 mi SE of Port St. Joe, sandy dry soil of pine flatwoods, 20 May 1960, P. Adams 456; cited in original publication (HOLOTYPE: GH). Hypericum lissophloeus P. Adams, Contr. Gray Herb. 189:21. 1962. isorype. U.S.A. FLoripa. Bay Co.: shores of Merial Lake, 10 mi N of Panama City, 14 Jun 1960, R.K. Godfrey & JN. Triplett 59844; cited in original publication (HOLOTYPE: GH). Hypericum mutilum L. var. latisepalum Fernald, Rhodora 38:372. 1936. IsoryPE. pal MOORE AND GIANNASI, Type specimens, University of Georgia herbarium 147 S.A. Florina. Duvall Co.: low ‘grounds, Jun-Aug no year, A.H. Curtiss 264*, cited in original publication (HOLOTYPE: Ilex decidua Walter var. curtissii Fernald, Bot. Gaz. 33:155. 1902. isorype. U.S.A. Floripa. Suwannee Co.: woods along the Suwannee River near Branford, 24 Oct 1900, A.H. Curtiss 6736; annotated as isotype by R.C. Clark, 1983. Ilex glabra (L.) A. Gray f. leacocarpa F.W. Woods, Rhodora 58:25—26. 1956. 1soTyPE. U.S.A. Floripa. Jackson Co.: open woodland 4 mi S of Marianna, 20 Jan 1955, FW. Woods C.E.F.-2; cited in original publication, annotated as isotype by R.C. Clark, 1983 (HOLOTYPE: US). Justicia cooleyi E.C. Leonard & J. Monachino, Rhodora 61:184. 1959. isotype. U.S.A. Floripa. Hernando Co.: Indian Hill, 2 mi NW of Chinsegut Hill, 20 Nov 1957,G.R. Cooley 5481; cited in original publication (HoLoTYPE: NY). Lesquerella perforata Rollins. Rhodora 54:190. 1952. isorype. U.S.A. TENNESSEE. Wilson Co.: open field 5 mi N of Lebanon, 30 Mar 1952, R.C. Rollins & D. Rollins 5207; cited in original publication (HoLoryPE: GH) Lesquerella stonensis Rollins, Rhodora 57:255—256. 1955. isorypr. U.S.A. TENNESSEE. Rutherford Co.: field near E fork of Stones River, Walterhill, 26 Apr 1955, R.C. Rollins 55176; cited in original publication (HOLOTYPE: GH). Lesquerella thamnophila Rollins & Shaw, The genus Lesquerella:86. 1973. IsoTYPE. U.S.A. Texas. Zapata Co.: sandy soil 3 mi SE of Zapata, 19 Apr 1959, R.C. Rollins & D.S. Correll 5949; cited in original publication. Lobelia harrisii Urban, Symb. Ant. 5:520. 1908. synrypE. JAMAICA: near Troy, 28 Jun ee W. Harris 8695; cited in original publication. wigia raventi C.I. Peng, Syst. Bot. 9:129. 1984. isorype. U.S.A. SoutH CAROLINA. Root Co.: roadside sedgy, sandy ditch, 0.1—0.2 mi SW of the jct. of County Road 16 and County Road 6, on County Road 16 W of Moncks Corner, 9 Sep 1982, C.1. Peng 4402; cited in ee ae annotated as isotype by C.I. Peng, 1984 (HoLoryre: MO). upinus westianus Small, Torreya 26:91—93. 1926. isorype. U.S.A. FLoripa. Bay Co.: dunes oe St. Andrews Bay, 4 May 1926, ].K. Small, C.A. Mosier & PA. Matthews 12821; annotated as isotype by D.B. Dunn, 1969. ythrum curtissii Fernald, Bot. Gaz. 33:155. 1902. isorypr. U.S.A. Georcia. Calhoun Co.: miry place in swamp near Leary, 20 Aug 1901, A.H. Curtiss 6876; cited in original publication, anonymously annotated as isotype. Marshallia obovata (Walt.) Beadle & FE. Boynton var. scaposa Channell, Contr. Gray Herb. 181:90. 1957. tsorype. U.S.A. NortH Carouina. Brun Space Co.: coarse sand, plowed fire lane, 3 mi W of Leland along US 76, 21 May 1949, R.K. Godfrey 49184; cited in original publication, annotated as isotype by R.B. Channell, (HOLOTYPE: GH). Mirabilis macfarlanei Constance & Rollins, Proc. Biol. Soc. Wash. 49:148. 1936. ISOTYPE. U.S.A. OREGON. Wallowa Co.: lower Ae Landing, between mouth of Somer’s Creek and Pittsburgh Landing, T27N, R1-2W, 15 May 1936, L. Constance, R.C. Rollins, H.F. Clements & L.A. Dillon 1579, cited in original publication. ynarda X medioides W. Duncan, Rhodora 61:303—304. 1959. HOLOTYPE. U.S.A. GeorGia. Union Co.: highway right-of-way S of Blairsville, 25 Jun 1959, WH. Duncan 21628, cited in original publication. Myrica curtissit Chev., Mem. Soc. Sci. Nat. Cherbourg 32:269. 1901. 1sorype. U.S.A. FLoripA. Swampy places in pine barrens near Jacksonville, 10 Mar & 22 Oct 1894, A.H. Curtiss 4571; annotated as isotype by J.R. Baird, 1968. Nuphar X interfluitans Fernald, Rhodora 44:397-398. 1942. isorypr. U.S.A. Vir- GINIA. New Kent Co.: deep fresh tidal water of Chickahominy River, SE of Windsor Shades, 148 SIDALG6(1) 1994 9 Sep 1941, M.L. Fernald & B. Long 13607; cited in original publication, annotated as _ by E.O. Beal, 1954 and J.H. Mee 1982 (HOLOTYPE: GH). nothera linifolia Nutt., J. Acad. Nat. Sci. Philadelphia 2:120. 1821. isorype. U.S.A. a. DeKalb Co.: soil thinly ieee rocks, Little Scone Mountain, 11 May 1901, A.H. Curtiss 6778; annotated as isotype by G.B. a ey, 1976. pu witia Chailestoniens/sClokay, Madrofio 7:71. 1943. isorypr. U.S.A. Nevaba. Clark Co.: Charleston Mountains, rocky slope, oa Mixe: 16 Jul 1938, LW. Clokey 8029, cited in ae publication. multigeniculata C ee peer 7:69-70. 1943. isoryPE. U.S.A. NEvADA. Clark Co.: "C harleston Mountains, E of Wilson's ranch, ead from Blue oo Mine to Mill, rocky ridge, 13 Jul 1939, 1 W. C ee 8430, cited in original publication. Phacelia dubia (L.) Trel. var. georgiana McVaugh, Ecol. Monogr. 13:158. 1943. isorype. U.S.A. GeorGia. Oglethorpe Co.: shallow soil about granite outcrops, Echols Mill, 2 mi NE of Lexington, 4 Apr 1938, J.H. Pyron & R. Mc ied ‘2448; annotated as isotype i R. McVaugh, 1941. Physostegia serotina Shinners, Field & Lab. 24:17. 1956. tsorype. U.S.A. LoulmsiANa. Calcasieu Par.: clay ditch bank, E side of Lake Charles, 9 Oct 1955, L.H. Shinners 22108; annotated as isotype by P. Cantino, 1979. ortulaca cornata Small, Bull. Torrey Bot. Club 23:126-127. 1896. Isorypr. U.S.A GerorGia. DeKalb Co.: on Little Stone Mountain, 11 Sep 1894, J.K. Small s.n.; cited in sae publication een PE: ). ulaca smallii P. Wilson in Rydb., N Amer. FI. ee 1932. isorype. U.S.A. te 1A. DeKalb Co.: on Little Stone Mountain, 11 Sep 4, J.K. Small s.n.; cited in original publication (HOLOTYPE: NY). Potentilla beanii Clokey, Bull. S. Calif. Acad. Sci. 38:4. 1939. isoryprs. U.S.A. Nevaba. Clark Co.: Charleston Mountains, ridge to Charleston Bak 22 Jul 1938, 1.W. Clokey 7974; cited in a al publicatic Potentilla ee C lokey, Bull. S. Calif. Acad. Sci. 37:4. 1938. isorype. U.S.A. Nevapba. Clark Co.: Charleston Mountains, Charleston Peak, gravelly slope and flat, 18 Jul 1937, 1.W. Clokey 67: cited in original publication. Pouteria ucuqui Pires & R. Schultes, Bot. Mus. Leafl. 14:87. 1950. isorypr. BRAZIL. r Rio Negro, Igarape Uaba, opposite mouth of Rio Xie, Amazonas, 3-6 Jan 1942, R.E. Schultes & F. Lopez 9553, cited in original publication (HOLOTYPE: GH) Primula nevadensis N.H. Holmgren, Madrofio 19:27. 1967. tsorypr. U.S.A. NEVADA. White Pine Co.: Snake Range, S slopes of Mount Washington, T12N, RO8E, 20 Jul 1965, N.H. Holmgren 22106; cited 7 original mene ot Prunus injucunda Sm Bull. Torrey Bor. Club 25:149. 1898. Isotype. U.S Georcia. DeKi alls Co.: base 7 ee Stone Mountain, along Tom George Creek, 7 Jul on Small s.n.; cited in fe publication (HOLOTYPE: ). yen anchennins rvipes (Greene) E. Grant & Epling, Leafl. Bot. Observ. 2:140. 1911. ISOTYPE. USA oak DeKalb Co.: upper ae and summit of Scone Mountain, 27 Jul 1893, ].K. Small s.n.; cited as as by E. Grant & C. Epling, 1943, Univ. Calif. Publ. Bot. 20:209, annotated as isotype by J. Hamer, oo Gononver: US). Pyrrhopappus georgianus Shinners, Field & Lab 21:93. 1953. isorype. U.S.A. GEOR- Gta. Glynn Co.: sandy soil along roadside on St. hee Island, near Brunswick, 11 Apr 1947, A. Cronquist 4276; annotated as isotype by L.H. Shinners, 1953 ee arbutifolia (L.) L.f. var. glabra Cihauilse: ee 14: 101-102. 1949. HOLO- Type. U.S.A. Georaia. Wayne Co.: fine sandy soil in wet open pinelands SW of Jesup, 10 Apr oo W.H. Duncan 4659, cited in original publication. Quercus ajoensis C.H. Muller, Madrofio 12:140. 1954. isorype. U.S.A. ARIZONA. Pima MOORE AND GIANNASI, Type specimens, University of Georgia herbarium 149 Co.: wash in main fork of Alamo Canyon, Ajo Mountains, Organ Pipe Cactus National Monument, 15 Aug 1952, C.H. Muller 9519: cited in original publication Quercus oglethorpensis W. Duncan, Amer. Midl. Naturalist 24:755. 1940. ISOTYPE. U.S.A. Georaia. Oglethorpe Co.: well drained bottomland of Buffalo Creek 8 mi : Lexington, 11 Sep 1940, WH. Duncan 2870, cited in original publi YPE: : Sanicula deserticola C.R. Bell, Univ. Calif. Publ. Bot. 27:133—230. O54 ISOTYPE. MEXICO. Baja Cauirornia: Norte, rocky slopes 6 mi N of American Smelter Co. Mine, 4 Feb 1953, C.R. Bell 1355; cited in original publication. Senecio andersonii Clokey, Bull. $. Calif. Acad. Sci. 37:10. 1938. Isorype. U.S.A. Nevapa. Clark Co.: Charleston Mountains, Lee Canyon, brushy meadow, yellow pine belt, 5 Aug 1935, LW. Clokey 5638; cited in original publication Senecio ganderi T.M. Barkley & R.M. Beauch. Baten 26:106-108. 1974. IsoTyPE. U.S.A. CatirorniA: San Diego Co.: N slope of SE flbe. Lawson Peak, 23 May 1973, R. Moran & M. Douglas 21038, cited in original saan (HOLOTYPE: y ene andersonii Clokey, Bull. S. Calif. Acad. Sci. 38:2. 1939. isorypE. U.S.A. NEVADA. Clark Co.: Charleston Mountains, Levell Canyon, steep N pore slope, 9 Aug 1937, 1.W. Clokey 7514; cited in original publication. Solidago rigida L. var. laevicaulis Shinners, Field & Lab. 19:35. 1951. isorype. U.S.A Texas. Dallas Co.: 2 mi N of Cedar Hill, limestone soil, 11 Nov 1947, E. Whitehouse 19269: cited in original publication, annotated as isotype by S. Heard, 1986 (HoLotyPE: SMU). Solidago tarda Mackenzie ex Small, Man. SE Fl. 1355, 1509. 1933. Isorype. U.S.A Georaia. Clarke Co.: Winter ee Hulme Farm, 20 Oct 1926, J.H. Miller s.n.; cited in se publication (HOLOTYP 3 Tragia saxicola Small, Fl. SE U.S. 702, 1333. 1903. isotype. U.S.A. FLoriDA. Dade Co.: rocky pine woods between the Everglades and Biscayne Bay, SE Florida, Jun no year, A.H. Curtiss 2517; annotated as isotype by K.I. Miller, 1964. Triodanis holzinger McVaugh, Wrightia 1:45—46. 1945. isorype. U.S.A. OKLAHOMA. McClain Co.: open prairie near Washington, 12 Jun 1936, M. Hopkins & C.T. Eskew 088, cited in original publication, annotated as isotype by R.L. Wilbur, 1969 (HoLoTyPE: NA). Vaccinium ashei Reade, Torreya 31:71. 1931. isorype. U.S.A. FLoripaA. Okaloosa Co.: n small swamp near ees 26 Mar 1927 & 8 Jun 1929, WW. Ashe 5.n.; annotated as isotype by W.H. Camp, 19 Vaccinium elliottii Pees FI.S U.S. 260. 1860. sorype? U.S.A. Floripa: no location, no date, A.H. Chapman s.n.; annotated as isotype? By S. McDaniel, 1986. Vaccinium hirsutum Bu ckley, Amer. J. Sci. 45:175. 1843. isorype? U.S.A. CAROLINA AND GeEoraia: in montibus, no date, S.B. Buckley s.n.,; annotated as isotype? By 8. McDaniel, 1986 Vernonia albicaulis Pers. ssp. longistylis $.C. Keeley, J. Arnold Arbor. 59:374. 1978. HOLOTYPE. GUADELOUPE: Basse Terre, limestone cliffs along roadside, 2 km W of Pointe de la Saline along highway to St. Ann, 20 Dec 1975, J.E. Keeley & S.C. Keeley 1948a; cited in original publication. ernonia angustifolia Michaux var. mohrii S.B. Jones, Rhodora 66:397—398. 1964. HOLOTYPE. U.S.A. ALABAMA. Washington Co.: roadside longleaf pine woods 12.3 mi N of Citronelle on AL 17, 19 Jul 1963, S.B. Jones 1406; cited in original publication, annotated as Le by S.B. Jones, 1964. onia autumnalis McVaugh, Contr. Univ. Michigan Herb. 9:477-479. 1972. ISOTYPE. MEXIC O. JALISCO: mountains 3 miS of La ee road to Barra de Navidad, 3 Oct 1960, R. es oes annotated as isotype by S.B. Jones, 1972 a duncanii S.B. Jones, Brittonia 25:108-109. 1973. HOLOTYPE & ISOTYPE. i ac, etna limestone hillside 1 1—12 mi W of Cuidad Victoria, km 155, Mexico 150 SIDA16(1) 1994 Hwy 101, 6 Nov 1970, S.B. Jones 20559; cited in original publication, annotated as ae and isotype by S.B. Jones, 1972. a harleyi H. Robinson, Phytologia 44:287—288. 1979. isotype. BRAZIL. ie “19. 5 km SE of town of Morro do Chapeu on road BAOS2 to Mundo Novo, by the Rio Ferro Doido, 2 Mar 1977, R.M. Harley 19296; cited in original publication, annotated as isotype by J.G. Stutts, 1980 (HOLOTYPE: Vernonia liatrioides DC ssp. gentryi$.B joes: Rhodora 78:193-194. 1976. HOLOTYPE & IsoTyPE. MEXICO. DURANGO: aie mi W of Durango on Hwy 40, 7 Mar 1974, S.B. Jones 22527, cited in original publication, annotated as holotype and isotype by S.B. Jones, 1974. Vernonia libertadensis S.B. Jones, Fieldiana, Bot. 5:31. 1980. HoLoTyPE. PERU: Depro La Lipertab: Prov. Otuzco, Cerro Sango, 7 Jun 1953, A. Lopez 1947; cited in original Seen annotated as ore by S.B. Jones, 1978. nonia macphersonii S.B. Jones & J.G. Stutts, Britconia 33:546. 1981. IsoTyPE. ee ene steep mountainsides 3— 10 km generally E on the road to Mina del Cuale, from the jet. 5 km NW of El Tuito, Mpio. de Cabo Corrientes, 16-19 Feb 1975, R. McVaugh 26442, cited in original publication gas MICH). Vernonia macvaughit S.B. Jones, Brittonia 25:105. 1973. HoLoTyPE. MEXICO. Oaxaca: 4.1 mi S of Puebla-Oaxaca ae line and 6 mi N of Huajuapan de Leon, 23 Mar 1972,S.B. Jones 21064; cited in original publication, annotated as holotype by S.B. Jones, 1O72: Vernonia michoacana McVaugh, Contr. Univ. Michigan Herb. 9:482—484. 1972. IsoTyPE. MEXICO. MicHoacan: steep dry mountainsides ca. 8 km NW of Aguililla, road to Aserradero Dos Aguas, 2 Mar 1965, R. McVaugh 22668, annotated as isotype by S.B. Jones 1972; Vernonia nestor S. Moore, J. Linn. Soc., Bot. 35:317. 1902. Isorypr. MALAWI: Nyasaland, 1895, J. Buchanan 129; cited in original publication. ernonia obtusa (Gleason) S.F. Blake ssp. parkeri $.B. Jones, Brittonia 25:108. 1973. HOLOTYPE. MEXICO. Hipatco: dry limestone hillside 4-5 mi SW of Jalaca at La Placita, 7 Nov 1970, S.B. Jones 20564; cited in original publication, annotated as holotype by S.B. Jones, 1972 Vernonia oligactoides Less., Linnaea VI:648. 1831. NEoTyPE. BRAZIL: Sao Paulo, Mun. Itarare, PR151, Fazenda Esplanada, 17 Feb 1982, J.G. Stutts 1214, cited as noeotype by J.G. Stutts, 1988, Rhodora 90:86. nia paucartambensis Dillon, Brittonia 36:336. 1984. isorypE. PERU: Cuzco, Pasar, km 132 on Paucartambo-Pilcopata Road, 5 km below Pillahuata, 7 aac , A. Gentry, M. Dillon, P. Berry & J. Aronson 23553; cited in original publication Panna F). Vernonia petiolaris DC var. appendiculata Baker, Fl. Bras. 6:98. 1873. IsoryPE. BRAZIL. Minas Gerats: Caldas, no date, Regne// II] 655; annotated as isotype by S.B. Jones, 1980. Vernonia pugana S.B. Jones & J.G. Stutts, Brittonia 33:544-545. 1981. IsorypE. MEXICO. Jauisco: foothills of Sierra de Manantlan, 16-22 km S$ of El Chante, 3—6 Feb 1975, R. McVaugh 26172; cited in original publication (HOLOTYPE: MICH). Vernonia salicifolia (DC) Schultz-Bip. var. baadii oe Contr. Univ. Michigan Herb. 9:484. tsorype. MEXICO. Ja.isco: headwaters of Rio Mascota, 8-10 km § of El Rincon, on a a to Aserradero La Cumbre, 2—3 Apr 1965, R. ae 23473; annotated as isotype by S.B. Jones, 1972. ernonia tequiliana $.B. Jones & J.G. Stutts, Brittonia 33:544. 1981. IsoryPE. MEXICO. Jatisco: Barranca de los Tanques, desv. al-camino de San Martin de las Canas, MOORE AND GIANNASI, Type specimens, University of Georgia herbarium ja Mpio. Tequila, 3 Mar 1974, L.M. Villareal 6110; cited in original publication (HOLOTYPE: MICH). Vernonia woytkowskii $.B. Jones, Fieldiana, Bot. 5:29. 1980. IsorypeE. PERU: Lambayeque, Purculla ad Olmos, 29 Sep 1961, F Woytkowski 6770; cited in original publication, annotated as isotype by S.B. Jones, 1978 ane MO Viola charlestonensis Baker & R.T. Clausen, Madrofio 1945. ISOTYPE. U.S.A. Nevaba. Clark Co.: Charleston Mountains, Forest Camp No. i oe Dot Weer Creel 23 May 1937, 1.W. Clokey 7501; cited in original publication. Vitis acerifolia Raf., Amer. man. grape vines 14. 1830. Neorype. U.S.A. TEXas. Wilbarger Co.: along Beaver Creek on US 283 S of Vernon, in rest area 1.5 mi S of jet. with Farm Road 1763, 13 Jun 1986, M.O. Moore 700; cited as neotype by M.O. Moore, 1991, Sida 14:359. Vitis argentifolia T.V. Munson, Proc. Soc. Promot. Agric. Sci. 8:59. 1887. NEOTYPE. U.S.A. West VirGINniaA. Nicholas Co.: W side of US 19, 1.7 miS of the Braxton Co. line, | mi N of jet. with WV 55, 25 Aug 1987, M.O. Moore 886; cited as neotype by M.O. ae 1991, Sida 14:347-348,. Vitis bloodworthiana Comeaux, Sida 14:460—461. 1991. tsorypr. MEXICO. Du- RANGO: along Hwy 40 16.5 km W of Del Diablo and 35.5 km E of the tropic of cancer, 1 Jul 1985, B.L. Comeaux 4219; cited in original publication (HOLOTYPE: SMU). Zornia ey ie (Walter) J.R Gmel., Syst. Nat.:1096. 1791. 1sonsorypE. U.S.A. GeorGiA. McDuffie Co.: open pine Wie woods 5 mi E, 43° south of Thomson, 29 Jul 1950. We: ae 11557; cited as isoneotype by R.H. Mohlenbrock, 1961, Webbia 16:2 29- 30 (NEOTYPE: US). REFERENCES Come, N.C. and §.B. Jones. 1985. Georgia plant list. $ d edition. Department of Botany, University of Georgia. Athens. ae S.B. and N.C. Coe. 1979a. Plant list for Clarke County, Georgia. Department of Botany, University of Georgia. Athe JONES, ‘S ne nd N.C. Corte. 1979b. Cae plant list. Department of Botany, University Jones, $.B. and N.C. Come. 1988. The distribution of the vascular flora of Georgia. Department of Botany, University of Georgia. Athens. 152 SIDA16(1) 1994 BOOK REVIEW TAYLOR, WALTER KINGLSEY. 1992. The Guide to Florida Wildflowers. (ISBN 0-87833-747-4, hbk) Taylor Publishing Company, 1550 West Mockingbird Lane, Dallas, TX 75235. $24.95. 320 pp. Are you headed to Florida this year and would like to identify some of the roadside flowers you meet? If so, you should consider obtaining this handbook before embarking. This mee book is designed be a pictorial guide to many of the commonly occurring wildflowers, especially of central and northern Florida. Taylor’s prowess as a photographer and ee astic field botanist comes through clearly, as the user will find a usually high quality color photograph of each species discussed. The photos are further supported by thumbnail descriptions, flowering times and geographic occurrences within the state. Conspicuous fruits are shown as photo insets for a number of the species, as well. However, close- ups of some with small, diffusely arranged flowers would have been a welcomed addit One's first impression, based on the crisp layout and handsome photos, is es oe book is a user-friendly guide; frustration lies ahead. Species are arranged first by color (color tabs at the top of each page), and secondly by an outdated sequence of relationships. Thus, one is guided to the species by looking through all the photographs of a particular color group. This task is complicated by the blurred definition of color groups. That is, should a particular species be searched for in the pink or the violet group, the white or green group, or the white or yellow group? The author’s choice is not always obvious because the violet group contains both blue flowers and rose-pink ones. Furthermore, in the yellow group a number of species with light yellow petals are illustrated with washed-out pictures showing white petals. Florida always presents a problem to designers of plant identification manuals. In the frost-free tip of the state, there is a myriad of diverse tropical elements that occur only there r that have invaded from more tropical areas, with or without human assistance. National guides usually ignore all of the tropical species, while state guides struggle with which to include. For this reason, visitors traveling to South Florida will be frustrated further by this guide. One glaring omission is that of lantana, a prominently colorful flower of roadsides in not only South Florida but the citrus area of central Florida, as we ye author might be excused based on his statement, “...excluded for the most part the woods plants, ine for some trees, shrubs, and vines that are rather common and produce attractive flow However, lantanas are not amie woody to the casual observer; and furthermore, he did include the more north nd less showy scrub plum, red chokeberry, gopher apple, greenbriers, Itea and aes ne ate ts that are conspicuously absent: Abrus, Allamanda, Caesalpinia bonduc, Carica, Echites, Epidendrum, Ernodia, Guzmannina, Hamelia, Neptunia, Oncidinm, Ricinus, Thalia, T. cic Urechites, and Waltheria Although this book has several problems, it is certainly no worse than other similar guides for Florida, It is well worth . money, -y, if you do not aloud own one of the other guides and you do not expect to be guided through tropical Florida.—Roger W. Sanders, Botanical Research Institute of Texas. CAREX LUTEA (CYPERACEAE), A RARE NEW COASTAL PLAIN ENDEMIC FROM NORTH CAROLINA R.J. LEBLOND North Carolina Natural Heritage Program 132 Norris Road Swansboro, NC 28584, U.S.A. A.S. WEAKLEY North Carolina Natural Heritage Program Department of Environment, Health, and Natural Resources PO. Box 27087 Raleigh, NC 27611-7687, U.S.A A.A. REZNICEK University of Michigan Herbarium North University Building Ann Arbor, MI 48109, U.S.A. W.J. CRINS Ontario Ministry of Natural Resources P.O. Box 9000 Huntsville, Ontario, POA 1KO, CANADA ABSTRACT Carex lutea (section Ceratocystis) is described from the outer coastal plain of southern North Carolina. It is the only Species ot us section to occur south of New Jersey along the Atlantic Coast, and 1emic of wet savannas underlain by limestone deposits. From all other species of this section, C. /wtea is distinguished by its pale scales, tall ((40)65—110(125) ake slender culms, and elongate inflorescences (4.5)5.4—25(41) cm long with usually only one or two widely as pistillate spikes below the + long peduncled i. 3)1-6(10. - mm) staminate spi RESUMEN a } Fo beeeae a eee | ] | rat an | alcura a: ee 110¢- 125)cm), ites sdelgados, inflorescencias Blacoad as de (4.5- 5.4 —25(- 1 que usualmente II las + largamente pedunculadas (0. 3. 1 6(-10.2) mm) una o dos espiees mer ampliamente separadas. INTRODUCTION Field surveys in 1991 by R.J. LeBlond of a rare, wet savanna habitat underlain by coquina limestone deposits on the outer coastal plain in Pender SIDA 16(1): 153 — 161. 1994 154 SIDA16(1) 1994 County, North Carolina disclosed an unfamiliar Carex with clear affinity to the well-defined boreal section Ceratocystis, and more specifically with the C, flava L. complex as defined by Crins & Ball (1988, 1989a, 1989b). Sub- sequent searches for the plant by LeBlond and A.S. Weakley disclosed four additional populations in Onslow and Pender counties. However, the popu- lations were all clustered within a radius of about 4 km. Furthermore, only the original population was large; all the others consisted of fewer than 60 clumps each. Diagnostic characters of the Carex flava complex include globose to ovoid, more or less approximate, short-peduncled to sessile pistillate spikes with the perigynia spreading (the lowermost often reflexed) (Crins & Ball 1989b). Three species of the complex, C. eryptolepis Mack., C. flava, and C. viridula Michx. subsp. viridula, are known from the northern coastal plain, but none has been found south of northern New Jersey (Crins & Ball 1989b). Two additional subspecies of C. viridula, C. viridula subsp. oedocarpa (Andersson) B. Schmid and subsp. brachyrrhyncha (Celak.) B. Schmid, are known from boreal habitats farther north along the Atlantic coast in the Gulf of St. Lawrence region and in Eurasia. Several other, distinct species are found elsewhere in the old world and in the temperate parts of the southern hemisphere, but the relationships of the North Carolina plants are clearly with the North American members of the C. flava complex. Close examination of the North Carolina plant discloses a number of significant points of difference from the other species of the Carex flava complex in North America. The pistillate scales of the North Carolina plant are pale yellowish-green, immediately separating it from C. flava and most subspecies of C. viridula, the scales of which are coppery or tinged with red or brown. The larger (up to 5.2 (5.8) mm long), reflexed lower perigynia separate it from those subspecies of C. viridula that may have paler scales. The North Carolina plants can be separated from the pale-scaled C. cryptolepis by its usually sparsely serrulate perigynium beaks, as well as its taller stature and longer inflorescences Several features unique to the North Carolina plant separate it from all other species in the C. flava complex both in North Americaand world-wide. The North Carolina plant is taller than any other members of the section, with taller mature flowering individuals typically over 65 cm tall and reaching heights of up to 1.25 m. Even though the plants are tall, the number of pis- tillate spikes is normally only one or two, extremely rarely three. As well, the inflorescences of the North Carolina plants (particularly those of culms that produce two pistillate spikes) are much longer than any other species, the longest ranging from 18—41 cm. All other members of the C. flava complex are much shorter, rarely reaching heights of over 65 cm, and routinely pro- ducing 2—3 pistillate spikes in shorter inflorescences 1.5—12(20) cm long. LEBLOND, WEAKLEY, REZNICEK AND CRINS, Carex lutea 155 Based on its unique morphology, this rare and highly localized North Carolina plant is here described as a new species. Carex lutea LeBlond, sp. nov. (section Ceratocystis). (Fig. 1) Plantae cespitosae; culmi (40)65—110(125) cm alti; vaginae basales pallide brunneae, glabrae. Folia 3-7, plerumque basalia; laminae 5.5—28 cm longae, 1.8-3.8 mm latae; vaginae 2.5—11 cm longae; ligulae obtusae 1. a 3.6 mm lone Inflorescentia (4.5)5. 4 25(41)cm longae; spicae 2—3(4), ascendentes rae pistillatae; bracteae infimae laminis 5—25 cm longis, 1.2— 2, 5 mm latis et eae (0)0. 2-4 cm onps Squamae pistillatae pallide flavovirentes. Perigynia (3.5)3.9—-5.2(5.8) mm longa, 1.4—2.2 mm lata, patentia, extrorsus curvata, plus minusve trigona, lutea, glabra, in rostrum contracta; rostra 1.4—2.2 mm longa. Achenium 1.4—1.9 mm longum, 1.2—1.5 mm latum. Styli marcesentes; stigmata 3. Antherae 3, 2.1-3.6 mm longae. Plants cespitose in small to large (up to ca. 45 fertile culms) clumps, with short ascending rhizomes; roots pale brown, not densely felted with root hairs; fertile culms (40)65—110(125) cm tall, 0.6—0.9 mm wide at base of lowermost spike, central, more or less trigonous, smooth except at apex where often slightly scabrous-angled, with glabrous, stramineous to pale brown bladeless basal sheaths. Leaves 3—7, mostly basal; blades 5.5—28 cm long, much shorter than culms, 1.8—3.8 mm wide, plicate, yellowish-green, glabrous, the margins antrorsely scabrous, the widest leaves 2.4—3.8 mm wide; leaf sheaths 2.5—11 mm long, more or less tightly enveloping culms, glabrous, green; inner band of sheath glabrous, whitish-hyaline, the apex thin and friable, irregularly concave, more or less truncate, or occasionally slightly prolonged; ligules obtuse, 1.4—3.6 mm long, the free portion more or less entire, whitish-hyaline, to0.7 mm long. Vegetative shoots 40—65(90) cm tall; leaves S—11, similar to those of fertile culms but up to 65(90) cm long and 4.6 mm wide, more or less evergreen (at least the proximal portions of the blades); pseudoculms ca. S—13 cm tall. Inflorescences (4.5)5.4—25(41) cm long, with all spikes quite separate, the lowest 2 pistillate spikes (if present) (1.6)4.5-18(33) cm distant; spikes single at nodes, ascending; lowermost spikes with peduncles 0.4—4.5(16.5) cm long, the uppermost pistillate spike sessile; peduncles smooth and terete proximally, more or less trigonous and serrulate-angled distally; lowermost bracts usually reflexed (except when subtending long-peduncled spikes), with blades 5-25 cm long and 1.2—2.5 mm wide and sheaths (0)0.2—4 cm long, 0.5—1.3(1.9) times as long as the inflorescence, the uppermost bracts also reflexed, but much reduced. Spikes 2—3(4), the terminal staminate (very rarely with an addi- tional small, accessory staminate spike at base), the lateral pistillate (very rarely with a small staminate apex up to 8 mm long). Terminal spikes (9)17— 39 mm long, 1.4—2.5 mm wide, ca. 45—90-flowered, peduncles (0.3)1— 6(10.2) cm long, (0.1)0.7—2.5(5.1) times as long as the spikes. Lateral spikes 0.7—2.7 cm long, 8-11 mm wide, globose to ellipsoid or short-cylindric, 156 SIDA16(1) 1994 FIG. 1. Carex /ntea, A. Habit. B. Portion of inflorescence. C. Sheath and ligule. D Pistillate scale. E. Perigynium, side view. F. Perigynium, front view. G. Perigynium, top ie Perigynium and achene, transverse section. I. Achene, front view. J. Achene, top view. K. Staminate scale. L. Anther. Bar equals 5 cm in A, 5 mm in B, 2 mm in C, and 0.5 m in D-L. Drawn by Susan A. Reznicek from che ee LEBLOND, WEAKLEY, REZNICEK, CRINS, Carex lutea 157 densely 15—G60-flowered. Pistillate scales 2.1-3.3 mm long, 0.9-1.3 mm wide, lanceolate to narrowly ovate, obtuse to more or less acuminate, glabrous, pale yellowish-green with a green center and wide hyaline (sometimes stramineous-tinged) margins and apex, l-nerved, almost totally concealed by the crowded perigynia. Staminate scales 2.9-5.6 mm long, 1.2-1.9 mm wide, narrowly obovate to narrowly oblong, obtuse to acute, glabrous, stramineous to pale brown with a green center and wide hyaline margins and apex, 1(3)-nerved. Perigynia (3.5)3.9-5.2(5.8) mm long, 1.4— 2.2 mm wide, spreading, strongly outcurved and squarrose in the spikes, the lowermost strongly reflexed, somewhat inflated, irregularly and asymmetri- cally trigonous with convex, obovate sides, glabrous, bright yellow proxi- mally, pale yellowish-green distally, short-stipitate and tapered to the base, 7—13-nerved with two nerves much more prominent than the rest, con- tracted into a strongly deflexed beak; beaks 1.4—2.2 mm long, finely and sparsely serrulate or sometimes smooth, green, the apex bidentulate with teeth 0.2—0.5 mm long. Achenes 1.4-1.9 mm long, 1.2-1.5 mm wide, trigonous with flat to slightly concave, strongly obovate sides, essentially truncate apically, brown, sessile. Styles withering; stigmas 3. Anthers 3, Z Type: UNITED STATES. Nortu Caro.ina. Pender Co.: SW of N.C. Hwy 50, ca. 1.8 mi W of Onslow Co. line, ca. 11 mi NNW of Holly Ridge. S of secondary Rd 1532, 0.85 mi SW of N.C. Hwy 50, “Lanier Quarry Savanna,” 20 May 1993, A.A. Reznicek 8942 with S.A. Reznicek, R.J. LeBlond & B.A. Sorrie (HOLOTYPE: MICH; Isorypes: BRIT/SMU, FLAS, GA, NCU, TAES, US, VDB, herb. C.T. Bryson). Additional specimens examined: NORTH CAROLINA. Onslow Co.: NE side N.C. Hwy 50, ca. 10.4 mi NNW of Holly Ridge, “Powerline Savanna,” 20 May 1992, A.A. Reznicek Bese with S.A, Reznicek, RJ. LeBlond, & B.A. Sorrie (MICH, VPI). Pender Co.: Lanier Quarry Savanna, S of secondary road 1532 0.85 miSW of Hwy NC 50, 11 Apr 1990, RJ. LeBlond 1188 (NCU); 24 Apr 1991, R.J. LeBlond 1985 (NCU); 22 May 1991, R./J. LeBlond 2056 (MICH, NCU); 27 Jun 1991, RJ. LeBlond 2267 (NCU); 20 Apr 1992, R./J. LeBlond 2723 (NCU); Sandy Run Savanna along Hunt Club Rd 0.5 mi S of Hwy NC 50, 12 May 1992, A.S. Weakley & RJ. LeBlond s.n. (NCU); Watkins Savanna, 0.1 mi N of Hwy N.C. 50 ca. 0.5 mi W of the Onslow Co. line, 12 May 1992, A.S. Weakley G& R.J. LeBlond s.m. (NCU); N side N.C. Hwy 50, ca. | mi W of Onslow Co. line, ca. 11.5 mi NNW of Holly Ridge, “Watkins Savanna,” 20 May 1992, A.A. Reznicek 8949 with S.A. Reznicek, R.J. ee G B.A. Sorrie (MICH, USCH). DISCUSSION In the key to the Carex flava complex in Crins & Ball (1989b), C. lutea keys closest to C. cryptolepis because of its pale, yellowish-green scales, narrow leaves, and perigynia of similar size. The overall yellowish-green color of C. /utea is also similar to that of C. cryptolepzs. Insertion of the following couplet in place of the first lead of couplet 3, however, will complete the key, and 158 SIDA16(1) 1994 allow easy separation of C. /utea from C. cryptolepis. 1. Tallest culms 65-125 cm; lowermost pisuvate apie bracts 0.5—1.3(1.9) times as long as the inflo i peduncles mostl 0.7—2 g i a 2-1.5 mm wide; perigynium beaks often sparsely serrulate C. lutea ; times as il h 1. Tallest culms 25—50 cm; lowermost pistillate spike bracts about 1.5—4 times as long as the inflorescence; staminate spike peduncles 0.2—0.5 times as long as the staminate spikes; achenes 1—1.2 mm wide; perigynium beaks smooth C. eryptolepis Although Carex lutea keys with C. cryptolepis due to its pale scales, its elongate inflorescence is most similar in aspect to that of C. viridula subsp. brachyrrhyncha var. elatior (Schitdl.) Crins (C. /epidocarpa Tausch), a boreal and subarctic taxon of highly calcareous, open wetlands in the Gulf of St. Lawrence region and Europe. Carex /utea differs most obviously from C. viridula subsp. brachyrrhyncha var. elatior in plant size, inflorescence size, and paler scales. The two taxa can be separated by the following couplet. 1. Pistillate scales pale yellowish-green, nearly the same color as the perigynia and thus inconspicuous; tallest culms 65—110(125) cm; longest inflorescences 41 cm C. lutea — . Pistillate scales with brown coloration, conspicuous among the greenish to yellowish perigynia; tallest culms (10)20—60(85) cm; longest inflorescences 2—12(18) cm long C. viridula subsp. brachyrrhyncha var. elatior In addition to the scale color and size differences, Carex /utea differs from C. viridula subsp. brachyrrhyncha var. elatior in inflorescence proportions. In C. lutea, the bracts siete lowest a spikes are shorter than or equalling horter than their peduncles. In C. viridula subsp. ee var. elatior, the bracts of the lowest pistillate spikes often equal or exceed the inflorescence and most staminate spikes are longer than their peduncles. Because of the great variability in these proportions due to differences in the number of pistillate spikes in the inflorescence and the considerable plasticity of inflorescences depending on the vigor of the plants, there is modest ees in these proportions, rendering them unsuitable as key characters. C utea also has proportion- ately narrower leaves than C. viridula subsp. brachyrrhyncha var. elatior. Although a significantly shorter plant than C. /wtea, the widest leaves of C. viridula subsp. brachyrrhyncha var. elatior range up to 5.6 mm whereas those of C. /utea are at most 4.6 mm wide. Other differences include the larger staminate spikes of C. /wtez, which are up to 39 mm long, much larger than the inflorescence and most the maximum of 25 mm in C. viridula subsp. brachyrrhyncha var. elatior; and larger perigynia up to 5.2(5.8) mm long in C. /wtea whereas those of C. viridula subsp. brachyrrhyncha var. elatior are only up to 4.2 mm long. It is tempting to suggest that the character of pale scales, which uniquely defines the two North American endemics, C. cryptolepis and C. lutea, indi- LEBLOND, WEAKLEY, REZNICEK, CRINS, Carex lutea 159 cates a close relationship. However, until genetic, breeding system, and micromorphological evidence comparable to that available for the other taxa is known for C. /wtea, statements of putative relationship are speculative. Nevertheless, when a character compatibility analysis was run with C. /wtea added to the data matrix of Crins (1990), C. /utea consistently clustered with C. cryptolepis (anpublished data). Although genetic and micromorphological data are not yet available for C. /wtea, it can be recognized readily and warrants description. The ecology of C. /utea is quite distinctive within the section; not surprisingly, since its occurrence is so far removed from all the other species. ll the sites occur in sandy soils overlying coquina limestone deposits, and the species shows a preference for the ecotone between the longleaf pine savanna and nonriverine swamp forest communities. Soil from a Lanier Quarry microsite for C. /utea had a pH of 5.6. This is similar to pH levels recorded at Thalictrum cooleyi H.E. Ahles microsites at the Lanier Quarry Savanna. Thalictrum cooleyi is the most frequent associate of C. /wtea, and is similarly restricted in distribution. Soils not supporting T: coo/eyz at the Lanier Quarry site regularly test at lower pH levels. ost Carex /utea plants occur in the partially to moderately densely tree shaded savanna-swamp ecotone, with scattered shrubs and a moderate to dense herb layer. The savanna-swamp ecotone is subject to occasional to frequent fires which favor a herbaceous ground layer and suppress shrub dominance. Carex /utea isa subdominant to patch dominant at two of the five known sites. Occurrences are densest in areas of partial to moderate tree shading with an absence of a shrub understory. Frequent associates include Taxodium ascendens Brongn., Liriodendron tulipifera L., Acer rubrum L., Myrica cerifera L. var. cerifera, Thalictrum cooley, Aletris farinosa L., Carex lonchocarpa Willd., Osmunda regalis L. var. spectabilis (Willd.) A. Gray, Physostegia purpurea (Walter) S.F Blake, and Parnassia caroliniana Michx. Though the height of Carex /utea suggests that it might be easily found, the plants are slender, occur in areas with substantial graminoid cover, and, except for one site, are rare and localized. Thus, a colony may remain quite inconspicuous, even in fruit. This suggests the possibility that additional colonies may be discovered elsewhere in similar sites on the southeastern coastal plain, though the association with narrowly endemic species in an unusual habitat suggests that C. /wtea may be a highly localized species. he occurrence of a localized disjunct species of the boreal Carex flava complex on the outer coastal plain of North Carolina is quite striking. This constitutes a disjunction of about 750 km from the nearest known popula- tions of other members of sect. Ceratocystis. In Eurasia, species of this com- plex occur as far south as Spain, Morocco, Turkey, and Iran. The most sub- stantial range separation among close relatives in Europe is that between C. 160 SIDA16(1) 1994 viridula subsp. brachyrrhyncha vars. elatior and nevadensis (Boiss. & Reut.) Crins from southeastern France to southern Spain (ca. 800 km). The type locality of Carex /utea, Lanier Quarry Savanna, and the other locations where it has been found are ecologically highly unusual and phyto- geographically notable. The combination of fairly open conditions under- lain by a calcareous substrate is very rare on the Atlantic coastal plain. Over 26 species regarded as rare in North Carolina by the state Natural Heritage Program are found at the Lanier Quarry site. Nine of these are listed as endangered, threatened or candidate species by the U.S. Fish and Wildlife service. Many of these rare plants have very restricted distributions, either being endemic to small areas or with highly scattered occurrences. The affinities of these taxa are variable, but include connections to the calcareous savannas of the Gulf Coast states, alkaline marshes of the Atlantic tidewater, calcareous glades, barrens, and prairies of the Appalachian region, and pinelands of the Carolinas and southern New Jersey. Most notable are the two endemic species closely associated with Carex lutea, Thalictrum cooleyi and an undescribed species of Al/ium. Thalictrum cooleyi is endemic to about a dozen sites, all savanna margins in sites underlain by calcareous deposits, in North Carolina, Georgia, and Florida. It has the highest ploidy level known in the genus (30 at 2n = 210), suggesting its derivation from a more widespread northern and inland species suchas T. revolutum (Park 1992). The undescribed A//ium is related to the widespread inland species A. cernunum Rothand A. ste//atum Ker and is known from only 5 sites, four of them shared with C. lutea. Because of their unusual annie conditions, these savannas underlain by calcareous deposits have evidently served as a small archipelago of phytogeo- graphic ae for species poorly adapted to present conditions on the southeastern coastal plain. It seems likely that Carex /utea is a narrowly distributed, very rare endemic. It may reflect a southern extension of the C. flava complex during the Pleistocene glaciation, followed by isolation ina few suitable sites and speciation. Alternatively, C. /vtea could represent the relictual occurrence of a formerly more widespread and older species in the C. flava complex. ACKNOWLEDGMENTS Weare grateful to the North Carolina chapter of The Nature Conservancy for providing funding fora systematic survey of the flora of its Lanier Quarr 8 g , y ag preserve, which resulted in the discovery of this new species. REFERENCES CRINS, es L990. aa considerations below the sectional level in Carex. Can. J. Bot. 68:1433-14 LEBLOND, WEAKLEY, REZNICEK, CRINS, Carex lutea 161 Crins, W.J. and P.W. BALL. 1988. Sectional limits and phylogenetic considerations in Carex section Ceratocystis (Cyperaceae). Brittonia 40:38—47. Crins WJ. and P.W. BALL. 1989a. Taxonomy of the Carex flava complex (Cyperaceae) in North America and northern Eurasia. I. Numerical taxonomy and character analysis. Can. J. Bot. 67:1032-1047. Crins, W.J. and P.W. Batt. 1989b. Taxonomy of the Carex flava complex (Cyperaceae) in North America and northern Eurasia. II. Taxonomic treatment. Can. J. Bot. 67:1048— 1065. Park, M.M. 1992. A biosystematic study of Thalictrum section Lexcocoma (Ranunculaceae). Ph.D. Dissertation, Pennsylvania State University, State College, Pennsylvania. 162 SIDA16(1) 1994 ANNOUNCEMENT The Bulletin of Botanical Research (Zhiwu Yanjiu), formerly Bulletin of the Herbarium of Northeastern Forestry Institute (1959-78) and Bulletin of Botanical Laboratory of Northeastern Forestry Institute (1979-80), is an international journal of botany. It covers plant systematics, geography, paleobotany, morphology, physiology, and ecology, with emphases on new taxa and evolutionary and ecological patterns of species diversity. It is published quarterly by Northeast Forestry University (NEFU), Harbin, China, and has a circulation of over 3,000 in 65 countries. Subscription per year (4 issue): $35.00 outside China. Guidelines for contributors are available upon request. Manuscripts submitted to the Bulletin of Botanical Research must not have been published previously. All manuscripts are considered by an editorial committee and reviewed by at least two reviewers. Manuscripts may be in English or Chinese. Page costs of $20 per page are charged. Manuscripts can be up to 25 double-spaced manuscript pages, an original and two copies should be submitted to: Prof. Shaoquan Nie Editor-in-Chief Bulletin of Botanical Research Northeast Forestry University 8 Hexing Road, Harbin 150040, CHINA Authors outside China should contact: Dr. Shiyou Li The Tucker Center Stephen F. Austin State University Nacogdoches, TX 75962, U.S.A. Tel. 409-568-4600 Fax. 409-568-1195 E-Mail: f_lis@ccsvax.sfasu.edu SPOROBOLUS POTOSIENSIS (POACEAE: ERAGROSTEAE): A NEW RHIZOMATOUS SPECIES FROM SAN LUIS POTOSI, MEXICO, AND A NEW COMBINATION IN S. AIROIDES JOSEPH K. WIPFF and STANLEY D. JONES S.M. Tracy Herbarium Department of Rangeland Ecology and Management Texas AGM University College Station, TX 77843-2126, U.S.A. ABSTRACT £ Sporobolus potosz PENCE hi ies fi San Luis Potosi, Menicothandiies ; . . (ornt 39. airoides by | yh 2.) lacking cHichomies behind the ligule, and 3.) peneially smaller in stature. A new combination is proposed at the subspecific rank: S. aéroides subsp. regis. A key to the rhizomatous species of Mexican Sporobolus, as well as S. nealleyi and S. atroides, is provided. RESUMEN Sporolobus potosiensis es una nueva ae rizomatosa de San Luis Potosi, México, que diftere de suf I I airoides por 1.) tener rizomas conspicuos y mas delgados, 2 ) cia de tri detras de la ligula, y 3.) tamafio mas pequefio generalmente. Se propone una nueva combinaciédn de rango ee ces S. airoides subsp. regis. Se ofrece una clave para las especies rizomatosas mexicanas de Sorolobus, asi como para S. nealleyi y S. airoides Sporobolus isa genus of approximately 160 species, distributed throughout the tropics, subtropics and temperate areas (Clayton and Renoize 1986). The species of Sporobolus intergrade to such an extent that their limits are seldom sharply defined (Clayton and Renvoize 1986). There are approximately 60 species of Sporobolus known to occur in the New World (Judziewicz and Peterson 1989) and 28 species are reported from México (Beetle 1987), only two species of which have been reported as rhizomatous: S. virginicus (L.) Kunth and S. regis ILM. Johnston. Sporobolus virginicus, a strongly rhizoma- tous perennial, is found on sandy beaches and at the bases of sand dunes from Virginia to Texas, south to Brazil and Peru, the West Indies, and the tropics of the Old World (Reeder 1975; Pohl 1980). Sporobolus regis is only known from Coahuila, México (Beetle 1987). Johnston (1943) stated that his new species, S. regzs, was probably most closely related to S. azroides (Torr.) Torr. and 8. wrightii Munro ex Lamson- Scribner and differed from these species in having pubescent leaf sheaths, tufts of trichomes in the axils of the panicle branches, and very coarse SIDA 16(1): 163 — 169. 1994 164 SIDA16(1) 1994 rhizomes. Johnston (1943) also mentioned that “the bases of the culms and the younger nodes of the rhizomes bear shredded remnants of old leaves.” The presence of shredded remnants of old leaves on the rhizomes needed further investigation, since grass rhizomes usually produce modified leaves termed “scale leaves” (Gould and Shaw 1983). Examination of the holotype [Stewart 2653 (GH, acronyms according to Holmgren etal. 1990)}, revealed that S. regis is not rhizomatous and that the “rhizomes” are the portion of the culms covered by soil, with their sheaths decomposed and their nodes developing adventitious roots, thus resembling rhizomes. All of the tillers, or shoots, of the type specimen are erect and lack lateral (horizontal) shoots. Because of the obvious remnants of the culm sheaths, and the lack of “scale leaves” on the culms, there is little doubt that the “rhizomes” of S. regzs are in fact culms responding to being buried. Sporobolus regis is distinguished from S. airoides by its densely pubescent sheaths and tuft of trichomes in the axils of the panicle branches. In our opinion, these characters are not significant to warrant the recognition of S. regis at the specific rank, but are significant to warrant infraspecific recog- nition. The lack of rhizomes places this taxon in S. azrozdes. However, these unique and distinct characters coupled with its restricted distribution (SW Coahuila) justify the recognition of this taxon at the subspecific rank, thus necessitating the following new combination. a ea airoides (Torr.) Torr. subsp. regis (I.M. Johnst.) Wipff & $.D. , comb. et stat. nov. BASIONYM: Sporobolus regis 1.M Johnst., J. Arnold Arbor. 393- 394 (1943). Type: MEXICO. Coanulta: salt flat 4 km SE of Laguna del Rey, eae 18 Sep 1942, Stewart 2653 (HOLOTYPE: GH!). Sporobolus potosiensis Wipf & $.D. Jones is the second rhizomatous species known to occur in México. This new species was originally identified at TAES from specimens that were either unidentified or misidentified as S. nealleyi Vasey. Specimens were requested from MEXU, MICH, MO, TEX and US to determine whether additional collections could be found. Sporobolus nealleyi isa cespitose, non-rhizomatous, gypsophilous species of the southwestern United States and northern México. Chase (1951) and Correll and Johnston (1970) reported S. nealleyi as rhizomatous or subrhi- zomatous. None of the specimens examined, including the holotype (Ned//ey, US), had rhizomes. However, the densely tufted and persistent bases can be covered by soil and then appear to be shortly rhizomatous or subrhizoma- tous, as already discussed above. Sporobolus nealleyi, in México, is reported from Coahuila, Nuevo Leé6n, and San Luis Potosi (Reeder 1975; Beetle 1987). It appears that reports from San Luis Potosi are based on misidenti- fied collections of S. potosiensis. Based upon specimens examined, distribution of S. nealleyi in México is probably restricted to northern Coahuila and Nuevo Leén. WHIPFF AND JONES, Sporobolus 165 The closest putative relative of 8. potosiensis is S. atroides. Specimens of S. potosiensis superficially resemble depauperate individuals of S. azrozdes, but differ from S. azroides by 1.) having conspicuous and slender rhizomes, 2.) lacking trichomes behind the ligule, and 3.) being smaller in stature. Sporobolus airoides does not have rhizomes and usually has conspicuous trichomes behind the ligule, though some very depauperate specimens were examined that did not have trichomes. Sporobolus airoides is usually a robust plant to 150 cm tall, but depauperate specimens may resemble S. potoszenszs in general appearance, except for the conspicuous rhizomes. KEY TO THE RHIZOMATOUS SPECIES OF SPOROBOLUS IN MEXICO, AS WELL AS §, AIROIDES AND S. NEALLEY1 1. Plants without rhizomes 2 1. Plants with rhizomes 4 2. Inflorescences (5-) 15—25 cm wide; found on dry soils in open ground, prairies and along saline or alkaline : 2. Inflorescences 1—3.5 cm wide; restricted to gypsiferous soils........... S. nealleyi Vasey . Back of sheaths densely pul I finfl i branches yi ; I adaxially pubescent S. airoides subsp. regis (I.M. = Johnst ) wae S.D. Jones 3. Back of s! } glabrous; f infl primary | r aivoides subsp. airoides 4. Inflorescences 4.0-9 cm wide, 12-18 cm long, open; branches spreading; plants inland, not coastal ............::00 S. potosiensts eee Jones 4. Inflorescences 0.6—1.0 cm wide, 2—8 cm long, contracted; branc appressed; plants coastal S. virginicus (L.) Kunth Sporobolus potosiensis Wipff & S.D. Jones, sp. nov. (Figs. 1—4) ramen perenne, 18—51 cm altum; rhizomatibus ad 14.5 cm longis, 1.3—3.5 mm latis. Inflorescentia 12—18 cm longa, ad 9 cm lata, diffusa; spiculis 1.6-2.5 mm longis, glabris; antheris 3, 1.2—1.4 mm longis, luteolis. Plants (Fig. 1) perennial, erect, 18-51 cm tall with slender rhizomes (Fig. 2) to 14.5 cm long, 1.3—3.5 mm wide. Leaves 4—5, cauline, subcoriaceous; sheaths glabrous, margins ciliate; collar (Fig. 4) abaxially glabrous, yellowish, margins conspicuously long ciliate with trichomes to 5.5 mm long; blades 6.2—24.0 cm long, 0.6—2.5 mm wide, flat, folded or involute; abaxial surface glabrous; adaxial surface minutely papillose, antrorsely scaberulous along raised veins; margins antrorsely scaberulous; /zgu/es 0.1—0.3 mm long, ciliolate membrane, truncate, lacking long trichomes behind the ligule. Inflorescences an open panicle, 12-18 cm long, 4.0—9.0 cm wide, glabrous, lower branches 3.0—8.0 cm long, pO ee roned in the proximal 1/3—1/2; pedicels 0.2—2.2 mm long, glabrous; spil totertiary branches. Spikelets (Fig. 3) 1.6-2.5 mm long, 0.5—1.1 mm wide, glabrous, mottled purplish, with one floret; g/wmes unequal; first glumes 1.0-1.5 mm long, 166 SIDAL6(1) 1994 SEP 4 1968 ys HOLOTYPE > Det by J.K. Wipf & $.D. Jones 8 potosiensis Wiptt & SO. Jones Determined by JK Wipf A 8. D. Jonas March 1904 MENIAL ELA NACIONAL DE CIENCIAS BIOLOGICAS M KXICO, DF PLANTAS MEXICANAS Sporobolua nealleyi Vasey Granine. ay 9 km al Rfoverde, sobre la Sereeeh ee a a, Ray SAP INIS ror ‘IX/1967 bet piavinnad’ ct oe con vegetacién ¢e gacatal ne ve ilo con Proania Pied ki raminea dominante | J .Rzedowski FIG. | Photograph of Sporobolus potostensis [Rzedowskt 24790 (MICH)}. Bar = 5 cm WIPFF AND JONES, Sporobolus 167 FIG. 2. Photograph of the rhi f Sporobolus potosiensis {Bravo 14 MMEXU)}. Bar = 1 cm. veinless, lanceolate; second glumes 1.3—2.1 mm long; conspicuously 1-veined, vein green; lanceolate with white-hyaline margins in the distal half, slightly shorter than lemma; /emmas 1.5—2.3 mm long; conspicuously 1-veined, vein green, ovate-lanceolate, mottled purplish with white-hyaline margins in distal half; paleas 1.6—2.2 mm long, 2-veined, grooved or furrowed between veins, similar to lemma texture and color; anthers 3, 1.2-1.4 mm long, yellowish. Caryopses 0.9-1.0 mm long, 0.4—0.5 mm wide. Chromosome number unknown. The specific epithet refers to the State of San Luis Potosi, México; the only area in which this species is currently known to occur. Distribution: Known only from the Rfo Verde River Valley, San Luis Potosi, México. Typus: MEXICO. San Luts Porost: 9 km al E de Rio Verde, sobre la carretera a Rayon, alt. 1000 m, terrenos aluviales planos con vegetaci6n de zacatal hal6filo con Prosopis, graminea dominante, 16 Sept 1967, J. Rzedowski 24790 (HoLotyPE: MICH!; tsorypes: LL!, MICH!, TAES!, WIS). Additional ined (paratypes): MEXICO. San Luis Potosi: in the valley of the Rio Verde, between Rio Verde ial San Ciro, alt. 850-1000 m, 12 Sep 1954, Sohns 1228 (TAES); in the valley of the Rfo Verde and in the Sierra de Cuates along the route, Rio Verde- San Francisco-Patios-Cardenas-Rayén, 14—15 Sept 1954, Sohns 1254 (TAES); Mpio. Ciudad del Maiz, 0.7 mi N of Las Tablas (RR crossing in town), alkaline flats dominated by grass and Juncus, gypseous soil, scattered mesquite, endemic Hedyotis, Pinaroppapus, Chenopodium, Samolus, Viguiera, Flaveria, alt. 1010 m, 22°17'N, 99°52'W, 14 Sep 1988, Nesom 6680 & Wells (ARIZ, TEX); + 5 km al SW de Tablas, alt. 1000 m, 22 Jan 1959, Rzedowski 9613 168 SIDA16(1) 1994 oo aes electron micrograph of the spikelet of Sporobolus potosiensis [Rzedowski nn or ES)}. Bar = 0.5 mm. ie 4. Scanning electron ieee of the collar of Sporobolus potostensis {Rzedowski 4790 (T AES)}. Bar = 0.5 n Qu (MEXU); oe de Llanos de pin municipio de Rfo Verde, alt. 900 m, 29 Jul 1980, Bravo 14 (MEXU); 0.65 km (0.4 mi) N of the Escuela Primera in Las Tablas, on gravel road nning eee Hwys 70 and 80, alt. 1097 m (3600 ft), 27 Jul 1979, Lane 25386 GJ. E. Fryxell (TEX). ACKNOWLEDGMENTS We thank Tony A. Reznicek (MICH) for his assistance with the Latin description and review of the manuscript. We thank Charlotte G. and John R. Reeder (ARIZ), and an anonymous reviewer for their review and suggestions. We also would like to thank the Curators at the following herbaria for loaning specimens: GH, MEXU, MICH, MO, TEX/LL, and US. REFERENCES Beetle, A.A. 1987. Noteworthy eae from Mexico XIII. Phytologia 63:209—297. CHASE, ae ae A revision o tchcock’s Manual of the grasses of the United States, 2nd ed. U.S.D.A. Misc. Publ. on 200. United States Government Printing Office, nae D.C. Ciayton, W.D. and S.A. RENvoize. 1986. Genera graminum, — of the World. Kew Bull., Addit. XIII. Her Majesty's Stationery Office, Londor Corre, D.S. and M.C. JoHNston. 1970. Manual of the ae ‘plant of Texas. Texas Research rene Renner, TX. WIPEF AND JONES, Sporobolus 169 GouLp, FW. and R.B. SHaw. 1983. Grass systematics, 2nd ed. Texas A&M University Press, College Station, TX. HoimGren, P.K., N.H. HoLMGREN and L. C. Barnett. 1990. Index herbariorum. Part I: The herbaria of the world. Regnum Veg. 120. New York Botanical Garden, New York, NY. ane: I.M. 1943. Plants of Coahuila, eastern Chihuahua, and adjoining Zacatecas and Durango, II. J. seiee Arbor. 24:377-421 pcr eien E.J. and P.M. Peterson. 1989. ‘Svonobalie temomazremensis (Poaceae: Eragros- tideae): a new species ae northern South America t. Bot. 14:525-528. POHL, R.W. 1980. Gramineae. In: Flora Costaricensis, Family #15, ed. W. Burger. Fieldiana, Botany No. 4. Field Museum of Natural History, Chicago, IL. REEDER, C.G. 1975. Sporobolus. In: The grasses of Texas, Frank W. Gould. Texas A&M University Press, College Station, TX. 170 SIDA16(1) 1994 ANNOUNCEMENT The Department of Botany of the National Museum of Natural History, Smithsonian Institution, is pleased to announce the inauguration of a Gopher Server that will provide Internet access to databases and documents that our staff have created or now manage. The enclosed brochure describes these offerings. We wish co call your attention in particular to the Type Specimen Register of the U.S. National Herbarium, which with over 88,000 records is the world’s largest electronic database of plant types and will become an important reference for monographers and flora writers. Also please note that our Index to Historical Collections is the most compre- hensive account of the collectors represented in the U.S. National Her- barium. Neither of these databa available el tronically orin printed form before. Our Gopher Server provides electronic access with key word searching to several documents that previously were available only as printed matter. These include the Checklist of the Plants of the Guianas (Guyana, Surinam, French Guiana) (1992), the Biological Conservation Newsletter (198 |-present) and, on behalf of the American Society of Plant Taxonomists, the ASPT Newsletter (1987-present). We hope you will bring this announcement to the attention of your readers. We would also appreciate expressions of support for our efforts to make this information available. A strong and favorable response from the botanical community will enable us to make more databases and documents available in the future. NOTES ON CAREX, CYPERUS, AND KYLLINGA (CYPERACEAE) IN MISSISSIPPI WITH RECORDS OF EIGHT SPECIES PREVIOUSLY UNREPORTED TO THE STATE CHARLES T. BRYSON USDA. ARS. Southern Weed Science Laboratory Stoneville, MS 38776, U.S.A. RICHARD CARTER Herbarium, Department of Biology Valdosta State University Valdosta, GA 31698, U.S.A. ABSTRACT ield and herbaria explorations have Raa the following as new to Mississippi, U.S.A.: poe bicknellii var. opaca, C. fissa vat. fissa, Cyperus drummondit, C. elegans, C. entrerianus, C. tie G: ubibnaate and Kyllinga bea (iladades Two of these, < entrerianus and K. brevifolioides, have the potential to become tl land lawn weeds, respectively, in a southeastern United States. Addidional range extensions are presented or C. aggregatus, C. difformis, C. flavicomus, C. lancastrensis, C. ovatus, and C. pilosus in Mississippi. Locality and habitat data are presented for all species reported. RESUMEN nen de campo 7, de rae yan dado los siguientes taxa como nuevos para Mississippi, U.S.A.: Ca Miiva ca, C. fissa vat. fissa, Cyperus drummondit, C. elegans, C. entrerianus, C. eragrostis, C. bisinen Kyllinga brevifolioides. Dos so res C. entrerianus y K. brevifotoides, tienen potencial para llegaraser laagri- cultura y los cé éspede en el sudeste de los E | id t pliaciones de area para fe aggregatus, c ea mis, C. flavicomus, C. es C. ovatus, y C. pilosus en Mississippi. Se p localidad y habitat de todas las especies citadas. INTRODUCTION While continuing to prepare a synoptical treatment of Carex, Cyperus, and Kyllinga as a contribution to the Flora of Mississippi Project, we have examined herbaria specimens and conducted field surveys for species with potential to occur within Mississippi. We have also continued assessment of population size, distribution, and habitat requirements of recently reported species, especially those which have potential to become weedy. This article adds to the knowledge of Carex, Cyperus, and Kyllinga which has been reported in recent years (Bryson 1984a; Bryson & Jones 1990; Bryson et al. SIDA 16(1): 171 — 182. 1994 172 SIDA1G6(1) 1994 1991; Bryson & Carter 1992; Bryson et al. 1992; Carter et al. 1987; Morris & Bryson 1986; Naczi & Bryson 1990). As we previously discussed, the flora of Mississippi is still poorly known in comparison with several adjacent states. Lowe’s Plants of Mississippi (1921), although outdated, must continue to serve as a base line for the general floristic work in the state. Other references that we used as sources of distributional data are Mackenzie (1931-1935), Kiikenthal (1935-1936), McGivney (1938), Corcoran (1941), Horvat (1941), Radford et al. (1964), and Godfrey & Wooten (1979). The terminology of physiographic regions or resource areas in Mississipp1 follows Lowe (1921) as adapted by Morris (1989). Herbarium abbreviations follow Holmgren et al. (1990), except ctb, MMNS, and USMH (personal herbarium of Charles T. Bryson, Mississippi Museum of Natural Science, Jackson, and University of Southern Mississippi, Hattiesburg, respectively). NEW STATE RECORDS Carex bicknellii Britton var. opaca EJ. Herm. was described from three collections by Dr. Delzie Demaree from river terraces in Lonoke and Prairie counties, Arkansas (Hermann 1972). In his discussion of this puzzling variety of C. bicknellii, Hermann (1972) emphasized its large, nearly nerveless, and partially translucent perigynia and indicated they were similar to perigynia of C. brittoniana Bailey and C. merritt-fernaldii Mack. as wellasC., bicknelliz. However, the perigynia of C. bicknellii var. opaca are much larger (5.5—7 mm long, 4-4.75 mm wide) compared to those of C. merritt- fernaldii (4—5 mm long, 2.5—3.5 wide). Carex bicknellii var. opaca differs from the Oklahoma-Texas C. brittoniana in usually having S—7(9) (rarely only 4) spikes per culm; perigynia (2)4—7 nerved (sometimes faintly) over the achene ventrally; and staminate and pistillate scales obtuse to long-acumi- nate, but with the midrib not excurrent as a scabrous awn (rarely a short mucro present). Carex brittoniana has culms with normally (2)3—4(5) spikes; perigynia nerveless over the achene ventrally, or nearly so; and staminate scales (and sometimes the lowermost pistillate scales) with the midrib excurrent as a scabrous awn ().1—0.9(2.4) mm long. Carex bicknellii vat. bicknellii is a dry prairie species not known from Mississippi. Carex bicknelli var. opaca was found at the same site in the Black Prairie Region of Mississippi where C. o&/ahomensis Mack. was first collected in the state by the senior author in 1991 (Bryson et al. 1992). It was associated with C. bushii Mack., C. complanata Torr. & Hook., C. glaucodea Tuckerman, C. longi Mack., and C. vulpinoidea Michx. The following are data reporting C. bicknellii var. opaca from Mississippi for the first time. Voucher specimens: U.S.A. MISSISSIPPI. Lee Co.: E of Tupelo, NE of jcc. of hwy US 78 and Auburn Road, 26 May 1993, Bryson 12400 (ctb, IBE, MICH, SWSL). BRYSON AND CARTER, Cyperaceae in Mississippi 173 Carex fissa Mack. var. fissa is herewith reported new to Mississippi. Carex fissa was described by Mackenzie (1931) from specimens collected in eastern Oklahoma where it was believed to be endemic. Kolstad (1986) reported C. fissa from southeastern Kansas and eastern Oklahoma. Hermann (1965) described Carex fissa var. avistata from north central Florida. Jones et al. (1990) reported C. fissa new to Texas. The Mississippi collection of C. fissa var. fissa was found ina wet ditch between old hwy US 78 and railroad tracks in with C. annectens Bicknell, C. triangularis Boeck., and C. vulpinoidea Michx. in the North Central Plateau Region. It is possible that this species may have been introduced into Mississippi by highway or rail traffic. However, its habitat in Mississippi is similar to that in Arkansas, and it is likely that C. fissa is native to Mississippi. It is probable that C. fissa var. avistata will be found in southern Mississippi. The following data report C. fissa vat. fissa new to Mississippi. Voucher specimens: U.S.A. MISSISSIPPI. Benton Co.: 1.7 mi NW Hickory Flat, 18 May 1990, Bryson 9977A (ctb, IBE, MICH). Cyperus drummondii Torr. & Hook. [=C. virens Michx. var. drummondii (Torr. & Hook.) Kiikenthal} in North America is restricted to the coastal plain and is known from Texas and Louisiana (Denton 1978). Specimens have also been seen from Florida and Georgia (Carter, unpublished data). It is also known from Nicaragua, Jamaica, the Galapagos Islands, Surinam, and Brazil (Denton 1978). Although Denton (1978) considered C. drummondii to be a variety of C. virens Michx., we think the morphological differences are sufficient for recognition as a species. Cyperus drummondii is generally a taller plant with a more compact inflorescence and fewer primary inflores- cence bracts, shorter scales, and a greater achene length to scale length ratio. At the site reported here, C. drwmmondii was associated with C. haspan L., C. ovatus Baldwin, C. polystachyos Rotth., C. strigosus L., C. virens, and Fimbri- stylis spp. in the Coastal Pine Meadows Region. The following are data for the first collection of C. drummondii from Mississippi. Voucher specimens: U.S.A. MISSISSIPPI. Jackson Co.: Moss Point, ditch along W side of hwy MS 63 between Frederick Street and Dr. Martin Luther King Drive, 0.5 mi S Escatawpa River Bridge, 16 Sep 1993, — 11315 (eth, SWSL, VS@): Cyperus elegans L. is known from coastal Texas, Louisiana, and Florida; it has not previously been reported from Mississippi. It is related to C, oxylepis Nees ex Steud. and can be distinguished from that species by its black, obovoid achenes (0.7—1.0 mm broad) and grayish brown, semi- translucent scales versus the brown, oblong, ca. 0.5 mm broad achene and golden yellow to orangish scales of C. oxylepis (Bryson & Carter 1992). Like C. oxylepis, C. elegans has a distinctive and pleasantly aromatic fragrance 174 SIDAI6(1) 1994 somewhat like that of cured juniper wood, which can often be detected several meters away from live plants in the field. The authors have on several occasions smelled these species in the field prior to making visual contact. Both species are locally abundant in heavily disturbed fill areas along the coast in Jackson County, Mississippi, and the populations of C. e/egans and C. oxylepis consist of more than 2000 plants each and cover an area greater than 121 ha (300 acres). Cyperus elegans was found in association with C. odoratus L., C. oxylepis, C. entrerianus Boeckler, C. filicinus Vahl, C. flavescens L., C. surinamensis Rottb., and C. virens in the Coastal Pine Meadows Region. Collection data for C. elegans in Mississippi follow. Collection data. U.S.A. MISSISSIPPI. Harrison Co.: Orange Grove Community Center, 0.3 miN jet. hwys I-10 and US 49; W of US 49, 16 Oct 1993, Bryson 13160 and Carter (ctb, SWSL, VSC). Jackson Co.: Pascagoula, vicinity of Bayou Casotte, S jct. of Louise and Washington Streets, 9 Aug 1993, Bryson 12595 (ctb, IBE, MICH, VDB, VSC); Carter 11339 (VSC, others to be distributed). Cyperus entrerianus Boeck. is an apparent introduction from South America or Mexico (Carter 1990) and was reported from 21 counties in Georgia, Florida, Alabama, Louisiana, and Texas (Carter 1990; Carter & Jones 1991). Although C. entrerianus is not recognized in some floras, it is quite distinct from other taxa in the Cyperus section Luzuloidei in temperate North America, and a comprehensive discussion and key was provided for this section by Carter (1990). This perennial is a copious producer of seeds and possesses a hard, stout, deeply set rhizome. It seems to be a more aggressive weed than other members of Cyperus section Luzuloidei that occur in Mississippi (e.g., C. acuminatus Torr. & Hook., C. drummondit, C. eragrostis Lam., C. pseadovegetus Steud., C. surinamensis, and C. virens). Cyperus entrert- anus is often locally abundant in eastern Texas and southern Louisiana where it seems to be associated with rice production, and it has been found ina rice field in Paraguay (Carter 1990). Cyperus entrerianus was found growing in the Coastal Pine Meadows Region with C. e/egans and its associates listed in the preceding discussion. The following are data for the first collections of C. entrerianus 10 Mississippi. Voucher specimens: U.S.A. MISSISSIPPI. ae ee Co.: Pascagoula, vicinity of Bayou Casotte, S jct. of Louise and Was een streets,9 A nae Bryson 12598 (ctb); 31 Aug 1993, Bryson 12821 and Newton (BR U, ctb, DSC, AS, FSU, GA, IBE, KNK, MICH, MISS, MISSA, MMNS, MO, a aL TAES, ce VDB, VSC, others to be distributed). Cyperus eragrostis Lam. is native to California, Oregon, and Washing- ton in North America; Bolivia, Peru, and Surinam south to Argentina, Brazil, Chile, and Uruguay in South America; and Easter and Juan Fernan- dez Islands in the Pacific (Denton 1978). This species has not previously BRYSON AND CARTER, Cyperaceae in Mississippi 175 been reported from Mississippi although two specimens collected by Ken Rogers (cited below) were collected in Forrest and Hinds counties in 1971 and 1981, respectively, and determined by R. Kral (VDB) and R. Carter. In Rankin County, collections were made along shallow narrow ditches. These sites are across the Pearl River and within five miles of the earlier Hinds County site of Ken Rogers. The Mississippi plants seem to be intermediate in size between typical C. eragrostis var. eragrostis and the diminutive form C. eragvostis var. compactus (Desv.) Kiik. Denton (1978) treated C. eragrostis var. compactus as a synonym of C. eragrostis because the diminutive form occurs throughout the range of the species. When transplanted into pots and grown under controlled conditions (i.e., with regular watering and fertilizer) in the greenhouse at the USDA, ARS, Jamie Whitten Delta States Research Center, Stoneville, MS, individuals of C. eragrostis from the Rankin County collection produced new stems that were longer than those observed in the field and that were typical of Cyperus eragrostis var. eragrostis. Thus, it appears periodic mowing of ditches and roadsides apparently caused plants to be shorter than typical for the species in the Rankin County, Mississippi population observed by the senior author. In Rankin County, C. eragrostis was associated with Carex longii, Cyperus haspan, C. odoratus, C. pseudovegetus, C. strigosus, C. virens,and Kyllinga brevifolia Rottb. [=Cyperus brevifolius (Rottb.) Hassk.}. Each of the collections were made in the Longleaf Pine Belt Region. The following data are the first records C. eragrostis from Mississippi. oucher specimens: U.S.A. MISSISSIPPI. Forrest Co.: Hattiesburg, drainage ditch at Kemper Park, 16 Jul 1971, Rogers 6708 (VDB). Hinds Co.: Jackson, Riverside Park, 8 Jun 1981, Rogers 46975 (VDB). Rankin Co.: Flowood, 0.3 mi E of Pearl River; S of hwy MS 25, 7 Sep 1993, Bryson 12966 (ctb, VSC); Bryson 12975 (ctb, SWSL, VSC); Flowood, 0.7 mi E Pearl River; S of hwy MS 25, 14 Oct 1993, Bryson 13128 and Bryson (ctb, SWSL, VSC). Cyperus louisianensis Thieret was previously reported from only two sites, including the type locality, in Tangipahoa Parish, Louisiana (Thieret 1977). Its lenticular achene and two-branched style place it in subgenus Pycreus, and it appears to be closely related to the North American species, C. diandrus Torr. (Thieret 1977). The scales of C. louistanensis are similar to those of C. diandrus, but C. louisianensis is distinguished by three stamens and styles that are divided less than half way to the base (Thieret 1977). Cyperus louistanensis 1s also closely related to the widespread old world species, C. sanguinolentus Vahl (Thieret 1977). Our observations indicate that C. louistanensis is typically found in disturbed habitats, such as road ditches and margins of artificial ponds, where it often forms dense, nearly monotypic, stands. Also, it is found in association with introduced weeds, e.g. C. pilosus Vahl (Bryson & Carter 1992) and Sacciolepis indica (L.) Chase (Bryson & Lockley 1993), in the Coastal Pine Meadows Region. We have also observed 176 SIDA16(1) 1994 that C. /ouistanensis flowers and fruits from late August until frost and that it is much more easily detected late in the season when its distinctive reddish scale pigmentation has developed completely. Thus, this may explain why it has been overlooked previously in Mississippi. Cyperus louisianensis has been listed as a candidate for protection (U.S. Fish and Wildlife Service 1993). Additional field and herbarium studies by the authors are currently in progress to determine the status of this taxon. The following data are the first records of C. /owistanensis from Mississipp1. Collection data. U.S.A. MISSISSIPPI. Hancock Co.: W side of hwy MS 43, 0.44 mi N of jct. hwys US 90 and MS 43 in Waveland, 16 Sep 1993, Carter 11342 (VSC, SWSL, others to be distributed); 16 Oct 1993, Bryson 13166 and Carter 11545 (ctb, SWSL, VSC, others to be distributed); W of Mill Creek, between Mill Creek and Indian Ridge Road, S of MS 43, ca. 5 mi E jet. hwys MS 43 and I-59 in Picayune, 18 Oct 1993, Bryson 13265 and Carter 11567 (ctb, SWSL, VSC, others to be distributed); 9.0 mi NW of j jet. hwys 43 and 603 in Kiln, at Petroleum Pipeline crossing of hwy MS 43,S side of hwy MS 43, 18 Oct 1993, Carter 11568 and Bryson (VSC, others to be distributed); N of Kiln, 0.6 mi S of jct. hwys MS 43 and MS 603, by W side of hwy MS 43, 18 Oct 1993, Bryson 13267 and Carter 11569 (ctb, SWSL, VSC, others to be distributed); Kiln, beside Shifalo Baptist Church and across hwy MS 43 from Kiln Post Office, by W side ofhwy MS 43, 18 Oct 1993, Bryson 132608 and Carter 11570 (ctb, SWSL, VSC, others to be distributed); SE jct. hwys MS 43 and I-10, 18 Oct 1993, Bryson 13271 and Carter 11571 (ctb, SWSL, VSC, others to be distributed). Harrison Co.: Orange Grove Community Center, 0.3 mi N jct. hwys I-10 and US 49; W of US 49, 16 Oct 1993, Bryson 13164 and Carter 11544 (ctb, SWSL, VSC, others to be distributed), Orange Grove, 1.1 miS jet. hwy US 49 and O'Neal Road, 18 Oct 1993, Bryson 13276 (ctb, SWSL, VSC, others to be distributed); N Gulfport, 0.3 mi W of jct. of Harrison Drive and 34th Avenue, ditch along Harrison Drive, 18 Oct 1993, Carter 11574 (VSC, others to be distributed); 3.32 mi W of jct. of Popps Ferry Road and D’Iberville Boulevard (=hwy MS 67), along Popps Ferry Road, ca 50 m W of jct. with Camp Four Jacks Road, 18 Oct 1993, Carter 11577 (WSC, others to be distributed), NW jct. hwys I-10 and US 49, 18 Oct 1993, sae 13279 (ctb, SWSL, VSC, others to be distributed). Jackson Co.: Pascagoula, just SE ct. of Washington and Louise Streets, vic. Bayou Cassotte, 16 Sep 1991, Bryson 11032 and Neu ton a BRIT/SMU, ctb, DSC, GA, FLAS, FSU, IBE, KNK, MICH, MISS, MISSA, MMNS, MO, NLU, NY, SWSL, TAES, TENN, UARK, US, USMH, VDB, VSC, others to be eee 16 Sep 1993, Carter 11337 (VSC, others to be distributed); St. vec 0.2 mi N of jet. of Old Fort Bayou Road and Rosefarm Road, along Rosefarm Road S of creek, 18 Oct 1993, Carter 11579 (WSC, others to be distributed); vicinity of St. Martin, 1.13 mi W of jcc. of Old Fort Bayou Road and Yellow Jacket Boulevard, between Lancaster Road and Mayfair Road, 18 Oct 1993, Carter 11580 (VSC, others to be distributed); vicinity of St. Martin, 0.19 mi E of jct. of Fort Bayou Road and Yellow Jacket Drive, by Old Fort Bayou Road, 18 Oct 1993, Carter 11581 (VSC, others to be distributed). Pearl River Co.: Picayune, ca. 250 m N ject. hwys I-59 and MS 43 by frontage road along W side of I-59, 18 Oct 1993, Bryson 13222 and Carter 11565 (ctb, SWSL, VSC, others to be distributed); Picayune, 0.5 mi W jct. hwys I-59 and MS 43; N of MS 43, 18 Oct 1993, Bryson 13257 and Carter 11562 (ctb, SWSL, VSC, others to be distributed). Kyllinga brevifolioides (Thieret & Delahoussaye) G.C. Tucker {=C brevifolioides Thieret & Delahoussaye} (Tucker 1984) was cited from Con- necticut, North Carolina, Pennsylvania, and Virginia in North America by BRYSON AND CARTER, Cyperaceae in Mississipp1 Lye Delahoussaye and Thieret (1967) in the original description. Subsequently, it has been reported from Maryland (Sipple 1978; Naczi et al. 1986), Tennessee (Kral 1981; Webb et al. 1981), Alabama and Georgia (Webb et al. 1981), New Jersey (Snyder 1983, 1984), Delaware (Naczi 1984; Naczi et al. 1986), and Arkansas (Sundell & Thomas 1988). Like K. brevifolia, K. brevifolioides is a rhizomatous perennial; however, it can be separated from K. brevifolia by its smooth scale keel and 2—3 stamens versus the denticulate scale keel andasingle stamen of K. brevifolia. Like K. brevifolia and K. odorata, K. brevifolioides is weedy and was probably introduced from Asia (Ferren and Schuyler 1980; Webb & Dennis 1981). Ky//inga brevifolia and K. brevifolioides have been observed by the authors to be weeds in periodically wet or frequently irrigated areas of lawns, roadsides, ditches, cemeteries, golf courses, and flower beds, often associated with the following sedges: Carex longti, Cyperus polystachyos, C. psendovegetus, C. strigosus, Eleocharis obtusa (Willd.) Schult., and Ky//inga pumila Michx. The collections reported herein are from the Locs Bluffs and Tennessee River Hills Regions. The following are data for the first collections of K. brevifolioides from Mississippi. Voucher specimens: U.S.A. MISSISSIPPI. Alcorn Co.: Corinth, S of jct. hwy US 72 and Cass Street, 29 Sep 1993, Bryson 13082 (ctb, FSU, IBE, KNK, NYS, SWSL, VDB, VSC); Corinth, SE of jet. hwy US 72 and Harper Road, 29 Sep 1993, Bryson 13094 (ctb, IBE, NLU, SWSL, Mea VSC). De Soto Co.: Olive Branch, 0.3 mi W of jct. hwys US 78 (old) and MS 302 (=Gc an Road); N of Goodman Road, 7 Oct 1993, Bryson 13 102 aaa (BRIT/ SMU, ctb, DSC, fee GA, IBE, KNK, MICH, MISS, MMNS, NLU, SWSL, USMH, VDB, VSC); ca. 0.8 mi N jet of Goodman Road and Getwell Road; W of Getwell ae | 3 Oct 1993, Bryson 13124 (ctb, VSC). OTHER NOTEWORTHY COLLECTIONS Cyperus aggregatus (Willd.) Endl. {previously known as C. bwarmensis (H.B.K.) M.C. Johnst., C. cayennensis (Lam.) Britton, and C. flavus (Vahl) Nees; see Tucker (1985)} is reported from Florida (Kral 1966; Clewell 1985), Louisiana (Horvat 1941), and Texas (Correll & Johnston 1970). Since it was first discovered in Stone County, Mississippi, in 1991 (Bryson & Carter 1992), C. aggregatus has been observed to be an aggressive weed of open sandy hilltops, roadsides, poorly kept lawns, and disturbed vacant lots on coarse sandy soils in similar habitats and often growing in association with C. croceus Vahl and C. retrorsus Chapman. The number, sizes, and distribution of populations in Mississippi suggest that C. aggregatus, if not a native, was introduced many years ago into the Coastal Pine Meadows and Longleaf Pine Belt Regions of Mississippi. The following are additional records of this weedy species, including two new county records. Voucher specimens: U.S.A. MISSISSIPPI. Forrest Co.: Fruitland Park, E of hwy US 49, 9 Aug 1993, Bryson 12580 (ctb, IBE, SWSL, VSC). Harrison Co.: 0.2 mi N of Howison, E of hwy US 49, 9 Aug 1993, Bryson 12581 (ctb, IBE, NYS, VSC); Gulfport, vacant lot W — 178 SIDA16(1) 1994 of hwy US 49 between 34th and Madison Streets, 12 Aug 1993, Bryson 12664 (ctb, SWSL, VSC). Stone Co.: Wiggins, hilltop just NE of jct hwy US 49 and MS 26, 16 Oct 1993, Carter 11542 and Bryson (VSC, others to be distributed). Cyperus difformis L. ranks as one of the world’s worst weeds (Holm et al. 1991) and occurs from Europe, Asia, Central America, North America (Lipscomb 1980), Australia, the Pacific Islands (Kiikenthal. 1935), Mexico (McGivney 1938), and South America (S. McDaniel pers. comm.). It is a particularly pernicious annual weed because of its relatively short generation period (as little as 4 to 6 weeks from seed to seed) and high reproductive potential (Holm et al. 1991). This species seems to be spreading in the United States especially along major waterways, through introduction at ports-of-entry, and in rice production in California. In the United States C. difformis has been known from Virginia since 1934 and isa troublesome weed of rice in California (Bryson 1984b). It is also known from Alabama (Kral 1973), Florida (Burkhalter 1985), Louisiana (Thieret 1964), Mississippi (Bryson & Carter 1992), Nebraska (Lemaire 1970), North Carolina (Tyndall 1983), Pennsylvania (Smith 1986), Tennessee (Webb & Dennis 1981), and Texas (Carr 1988). During field work in 1993, new populations were discovered, and previously known populations (Bryson & Carter 1992) were observed. Population size, in area covered and number of individuals, had increased by 2- to 400-fold since 1991. The following are data for recently discovered populations of C. difformis in Mississippi from the Coastal Pine Meadows and Longleaf Pine Belt Regions. Voucher specimens: U.S.A. MISSISSIPPI. Forrest Co.: Hattiesburg, Alcorn Street between James and Tipton streets, beside Hattiesburg Water Works, 16 Oct 1993, Bryson 13148, Carter and Rosso ree Sac ctb, FSU, IBE, KNK, MISS, MMNS, MO, NLU, NYS, SWSL, TAES, USMH, VDB C); Hencoue Co.: Kiln, wet ditch W of hwy MS 43, 6 Aug 1992, Bivion 11941 (ctb, oa VSC). Harrison Co.: Gulfsore. wet ditch between Seaway Road and RR; S$ of hwy I-10, 6 Aug 1992, Bryson 11993 (ctb, IBE, VSC); Orange Grove, 0.3 mi N jet. hwys I-10 and US 49; W of US 49, 31 Aug 1993, Bryson 12805 and Newton (ctb). Jackson Co.: Moss Point, SE jet. hwys I-10 and MS 63, 31 Aug 1993, Bryson 12818 and Newton (ctb); Moss Point, NE of jet. hwys US 90 and MS 63, 1 Sep 1993, Bryson 12835 and Newton (ctb, IBE, NLU, VDB, VSC); Pascagoula, Bayou Casotte area, adjacent to City Animal Shelter just S of S end of Louise Street, 16 Sep 1993, Carter 11338 (VSC, others to be distributed). Cyperus flavicomus Michx. (=C. a/bomarginatus Martius & Schrad. ex Nees) was previously known from only Lafayette and Tishomingo counties, both in northeastern Mississippi in the North Central Plateau and Tennessee River Hills Regions. Cyperus flavicomus was found growing on wet soil often in shallow standing water in association with Carex longit, Cyperus difformis, C. esculentus L., C. tria L., C. haspan, C. odoratus, C. pilosus, C. polystachyos, C. strigosus, C. surinamensis, C. virens, Eleocharis obtusa, E. tuberculosa (Michx.) Roem. & Schult., Fimbristylis autumnatlis (L.) Roem. and Schult., E miliacea BRYSON AND CARTER, Cyperaceae in Mississippi 179 (L.) Vahl, Kyllinga brevifolia, and Rhynchospora corniculata (Lam.) Gray. Based on our examination of herbarium specimens and field surveys, C. flavicomus seems to occur sporadically throughout much of the southeastern United States. The following are additional county records from the Coastal Pine Meadows Region of southern Mississippi. Voucher specimens: U.S.A. MISSISSIPPI. Harrison Co.: Orange Grove, 0.3 mi N jct. hwys I-10 and US 49; W of US 49, 31 Aug 1993, Bryson 12815 and Newton (ctb, DSC, IBE, MISS, MMNS, MO, NLU, SWSL, USMH, VDB, VSC); Bryson 13 162 and Carter (ctb, VSC). Jackson Co.: Moss Point, SE of jct. hwys I-10 and MS 63, 31 Aug 1993, Bryson 12816 and Newton (ctb). Cyperus lancastrensis Porter in Gray was first reported from Mississippi by Morris (1988) in the North Central Plateau Region. Subsequently, it was reported from Lee and Tishomingo counties in the Tennessee River Hills Region (Bryson & Carter 1992). Following are data from an additional three counties all from northern Mississippi in the North Central Plateau and Tennessee River Hills Region. Cyperus lancastrensis was found growing in association with C. echinatus (L.) Wood and C. strigosus in small damp depressions on open hillsides. Voucher specimens: U.S.A. MISSISSIPPI. Itawamba Co.: 1.6 mi N Dorsey SW of jct. of hwy US 78 and Fawn Grove exit, 26 Aug 1993, Bryson 12784 (ctb, VSC). Marshall Co.: N of Galena, jct. of old and new hwys MS 7, 24 Aug 1993, Bryson 12731 (ctb, IBE, VDB, VSC). Tate Co.: Thyatira, S of hwy MS 4, 7 Oct 1993 Bryson 13114 and Bryson (ctb, VSC). Cyperus ovatus Baldwin {=C. pollardii Britton in Small} occurs through- out Florida and northward in the outer Atlantic coastal plain into North Carolina and westward along the Gulf Coast into Louisiana (Carter, in prep.). It is related to C. retrorsus and is discussed previously from Mississippi (Bryson & Carter 1992). At the time of our previous paper, we were unable to locate C. ovatus in Hancock County despite field surveys in suitable habitats (e.g., pond margins, edges of salt marshes, banks of black water streams, and bogs and ditches in pineland savannas). Following are data for additional recent collections of C. ovatus from Mississippi in the Coastal Pine Meadows Region. Voucher specimens: U.S.A. MISSISSIPPI. Hancock Co.: Port Bienville Waterfront Industrial Park, 17 Oct 1993, Bryson 13182 and Carter 11555 (ctb); Kiln, W of hwy MS 43; across hwy MS 43 from Kiln Post Office, beside Shifalo Baptist Church; ca. 2.0 miS jct. hwys MS 43 and 603, 18 Oct 1993, Bryson 13269 and Carter (ctb). Cyperus pilosus Vahl, a weed of tropical and subtropical regions of Asia, Australia, and West Africa (Kiikenthal 1935, McGivney 1938), was appar- ently introduced into North America from the old world and was first reported in the United States from Tangipahoa Parish, Louisiana, by O'Neill (1938). Subsequently, C. p/osus was reported from Florida by Burkhalter 180 SIDA16(1) 1994 (1985), relocated in Louisiana by Carter (Bryson & Carter 1992), and discovered in Mississippi by Bryson & Carter (1992). In Mississippi, C. pilosus was found most frequently growing in disturbed soil at the edge of standing water in ditches, along slow moving streams, and around ponds often in association with many other Cyperus species including: C. difformis, C, iria, C. lonisianensis, C. odoratus, C. strigosus, and C. virens. The following data report C. p//osus from additional counties in Mississippi in the Coastal Pine Meadows Region. Voucher specimens: U.S.A. MISSISSIPPI. Hancock Co.: SE of Napolean, in ditch E of jct. hwys I-10 and MS 607; N of I-10, 10 Aug 1993, ewe 12606 (ctb, IBE, MO, NLU, SWSL, VDB, VSC); N of Waveland, W of hwy MS 43, 0.44 mi N of jct. hwys US 90 and 43, 16 Sep 1993, Carter 11346 (VSC). Harrison Co.: a ditch between Seaway Road and RR; Sof I-10, 6 Aug 1992, Bryson 11996 (ctb, IBE, MICH, NLU, TENN, VDB, VSC); Gulfport, ditch along Industrial Road, 9 Aug 1993, Bryson 12582 and Lockley (ctb); Bryson 12584 (ctb, IBE, SWSL, VSC); Bryson 12625 (ctb, IBE, VDB, VSC); Orange Grove Cain Center, 0.3 mi N jct. hwys [I-10 and US 49; W of US 49, 31 Aug 1993, Bryson 12813 and Newton (ctb, VSC); Bryson 13158 and Carter (ctb, VSC). Pearl River Co.: Ca. | mi S Picayune; ca. 3 mi N Mississippi/Louisiana state line; E of hwy I-59 at Mississippi Welcome Center, 17 Oct 1993, Bryson 13220 and Carter (ctb, SWSL, VSC); Picayune, 0.5 mi W jct. hwys I-59 and MS 43; N of MS 43, 18 Oct 1993, Bryson 13254 and Carter (ctb, SWSL, VSC). ACKNOWLEDGMENTS Thanks are expressed to N.B. Bryson, T.C. Lockley, and T.E. Newton for help with field work and A.A. Reznicek (MICH) who confirmed the determination of Carex bicknellii var. opaca and C. fissa var. fissa, J. W. Thieret (KNK) who confirmed the determination of the 1991 Cyperus louisianensis collection, and $.D. Jones (TAES), R.E-C. Naczi (KNK), and A.A. Reznicek who reviewed the manuscript. Financial support of the junior author’s field work was provided bya U.S. Fish and Wildlife Service status survey contract on Cyperus louisianensis. REFERENCES Bryson, C.T. 1984a. Noteworthy additions to the Carex of Mississippi. Castanea 49:44. 1984b. Weed alert: smallflower umbrella sedge. Southern Weed Sci. Soc. ' 76, Bryson, C.T. and $.D. Jones. 1990. Carex comosa (Cyperaceae) new to Mississippi. Sida Bryson, C.T., S.W. Rosso and R.EC. Naczi. 1991. Carex baltzellii (Cyperaceae) new to Mississippi with notes on Carex picta and Carex impressinervia in Mississippi. Sida 14: 493-499 Bryson, C.T. and R. Carrer. 1992. Notes on Cyperus and Ky/linga (Cyperaceae) in Mississippi with records of six species new to the state. Sida 15:119-124. Bryson, C.T., R.E.C. Nacziand 8. McDanigL. 1992. Notes on noteworthy records of Carex oe from the southeastern United States. Sida 15:125—135 BRYSON AND CARTER, Cyperaceae in Mississippi 181 Bryson, C.T. and T.C. Lockey. 1993. Sacciolepis indica (Poaceae) new to Mississippi. Sida 15: 56 BURKHALTER ye - 1985. Aletris farinosa, Cyperus difformis, and Cyperus pilosus new for Florida. Sida 11:2 Carr, B. ioe oe difformis L. we one new to Texas. Sida 13:255-256. Carter, R. in prep. A taxonomic treatment of Cyperus section Umbellati in Nors America. Carter, R., C.T. Bryson and B.L ae OMB. ee Cyperus uniflorus (Cyy : f the Mississippi River. Sida 12:250. CARTER, R. oe Cyperus entrerianus (Cyperaceae) overlooked species in temperate North America. Sida 14:69-77. eae R. and $.D. JONEs. rae Additional records of Cyperus entrerianus (Cyperaceae) in the United States. Sida 14:615-616. CLewELL, A.F. 1985. Guide to the vascular plants of the Florida panhandle. University Bie es of Florida: Florida State University Press. Tallaha Corcoran, M.L. 1941. A revision of the subgenus Pyced t in » North and South America. Catholic Univ. Amer., Biol. Ser. 37:1—68. Correll, D.S. and M.C. Josie 1970. Manual of the vascular plants of Texas. Texas Research ee. Ren DeELAHOuSSAYE, A.J. and J.W. Tate RET. 1967. Cyperus subgenus Ky/linga (Cyperaceae) in the continental United States. Sida 3:128—136. Denton, M.F. 1978. A taxonomic treatment of the Luzulae group of Cyperus. Contr. Univ. Mich. Herb. 11:197-271 FERREN WR. and A.E. ScHuyLer. 1980. Intertidal vasc a plants of river systems near Pitladéishia: Proc. Acad. Sci. Philadelphia 132:86- Goperey, R.K. and J.W. Wooren. 1979. Aquatic ne ie plants of sout United States. wee eae The University of Georgia Press, Athens. p. 2 HERMANN, FJ. 1965. An eastern variety of Carex fissa (Multiflorae). Rhodor 67: 198. Hermann, EJ. 1972. A new variety of Carex bicknellii from Arkansas. Sida 5:49. Ho.m, L.G., D.L. Plucknett, J.V. PANCHO and J.P. HERBERGER. 1991. The anne s worst eds: distribution and biology. Krieger Publishing Company: Malabar, Florida. Cae N, P.K., N.H. HoLMGREN and L.C. Barnett (eds.). 1990 Index Herbariorum. Part 1: the fierbana of the world, 8th ed. New York Botanical Garden, Bronx. Horvat, M.L. 1941. A revision of the subgenus Maréscws found in the United States. Catholic nN Amer., Biol. Ser. 33:1-147. Jones, S.D., G.D. Jones and J.K. Wiprr. 1990. Carex fissa, section Mu/tiflorae (Cy peraceae), new to He Phytologia 68: 47-50. Kotstap, A.O. 1986. Cyperaceae. p. 1059-1113. In: Flora of the Great Plains. ed. R.L. McGregor, T.M. Barkley, R.E. Brooks, and E.K. Schofield, University Press of Kansas. Krat, R. 1966. Observations on the es of the southeastern United States with special reference to pnohinetn Louisiana. Sida 2:395—408. yeastern 39. Krat, R. 1973. Some notes on the flora of the southern states, particularly Alabama and middle Tennessee. Rhodora 75:366-4 Krat, R. 1981. Further additions to some notes on the flora of the Southern States, particularly Alabama and Middle Tennessee. Rhodora 83:301-315. - KENTHAL, G. 1935-36. oo -Scirpoideae-Cypereae. In: L. Diels, ed., Pflanzenreich V. 20 (Heft 101):1—67 i MAIRE, R.J. 1970. Re tee records for Nebraska. Rhodora 72:283-284. Lipscoms, B.L. 1980. Cyperus difformis L. (Cyperaceae) in North America. Sida 8:320—327. Lowe, E.N. 1921. Plants of Mississippi. Mississippi State Geological Survey, Bulletin 17, Jackson. 182 SIDAL6(1) 1994 MackeNzigE, K.K. 1931-1935. aoe aaa Cariceae. N. Amer. Flora 18. McGivney, M.V. 1938. A revision of the re Eucyperus found | in the United States. Catholic Univ. Amer., Biol. Ser. 26:1— 14 Morris, M.W. and C.T. BRYSON. 1986. Carex swanii in Mississippi. Castanea 51:226—227. Morris, M.W. 1988. Noteworthy vascular plants from Grenada County, Mississippi. Sida 7-186. «1989. Spiranthes (Orchidaceae) in Mississippi. os 11:39-48. Naczi, - EC. 1984, Rare sedges discovered and rediscovered in Delaware. Bartonia 50:31—35. Naczi, R.EC., R.J. Driskitt, E.L. PENNELL, N.E. Seyeriep, A.O. TUCKER and N.H. Dit. ee New records of some rare DelMarVa sedges. Bartonia 52:49—-57. Naczi, R.F.C. and C.T. Bryson. 1990. Noteworthy records of Carex (Cyperaceae) from the southeastern United States. Bartonia 56:49-58. O'NEILL, H.T. 1938. Cyperus pilosus Vahl in the United States. Rhodora 40:74. Raprorb, A.E., H.E. Antes and C.R. Bett. 1964. Manual of the es flora of the Carolinas, The University of North Carolina Press, Chapel Hill. p. 203. Sippce, W.S. 1978. An atlas of ge plant species es aS for tidewater er eere! Scene Publ. No. 1. Maryland Dept. Nat. Res., Annapoli SmitH, T.L. 1986. News and notes: eastern Pennsylvania a Lee inventory 1985 held highlights. Bartonia 52:83-84. SNYDER, D.B. 1983. Rare New pe y grasses and sedges. Bartonia 49:7 1-72. —_______. 1984. Botanical discoveries of Vincent Abraitys. Bartonia 50:54—56. SUNDELL, E. and R.D. THomas. 1988. Four new records of Cyperus (Cyperaceae) in Arkansas. Sida 13:259-261. THIERET, J.W. 1964. More additions to the Louisiana flora. Sida 1:294—295. . 1977. Cyperus aides eae aceae), a new species from southern Louisiana. Louisiana Acad. $ Tucker, G.C. 1984. A revision “of the ; genus olan Rottb. (Cyperaceae) in Mexico and Central America. Rhodora 86: 507-538. —______. 1985. The correct name for Cyperus cayennensis (C. flavus), Cyperaceae. Southw. Matuealas: 30:607 TYNDALL, R.W. 1983. Diba of Cyperus difformis L. (Cyperus) in the southeastern Wniced Seres Castanea 48:277—280. U.S. Fish AND WILDLIFE SERVICE. one lant taxa for listing as endangered or threatened species; notice of review. Federal Register 58(188):51144—51190. Wess, D.H. and W.M. DENNis. 1981. Additions to the flora of Tennessee. Sida 9:184—185. WeBB, D.H., W.M. DEeNNis and T.S. Parrick. 1981. Distribution and naturalization of Cyperus brevifolioides (Cyperaceae) in eastern United States. Sida 9:188—190. NEW TAXA AND NEW COMBINATIONS IN CHINESE PLANTS SHIYOU LI and KENT T. ADAIR The Tucker Center College of Forestry Stephen F. Austin State University Nacogdoches, TX 75962, U.S.A. ABSTRACT This paper describes two new conifers found in boreal China, namely, Larix gmelini (Rupr.) Rupr. genhensis S.Y. Li & K.T. Adair and Pinus sylvestris L. var. manguiensis S.Y. Li & K.T. Adair. Also, it presents two new combinations, Polemonium hingganicum (PH. Huang &S.Y.Li)S.Y. Li & K.T. Adair and aes ne L. subsp. chinense (Pursh) S.Y. Li & K.T. Adair. RESUMEN En este articulo se describen dos coniferas nuevas qu en el norte de China, Larix gmelini as ) Rupr. subsp./var. genhensis S.Y. Lig & K.T. Adair y Pinus sylvestris L. var. manguiensis S.Y. Li & K.T. Adair. También se presentan dos combinaciones nuevas Polem- onium lena (PH. Huang & S.Y. Li) S.-Y. Li & K.T. Adair y Penthorum sedoides L. subsp. chinense (Pursh) S.Y. Li & K.T. Adair This article describes two new taxa and two new combinations from China. The study is largely based on collections Shiyou Li made in a four-year forest resource investigation in Northeast China during 1984-1987. The speci- mens are mainly deposited in the herbarium of Northeast Forestry Univer- sity (NEFI) China. Larix gmelini (Rupr.) Rupr. var. genhensis S.Y. Li & K. T. Adair, var. nov. (Fig. 1) A var. genhensis brachlis longis et pendula differt. Larix gmelini(Rupr.) Rupr. var. genhensis (GENHEN LARCH) is distinguished from var. gmelini (DAHURIAN LARCH) by its long, slender and pendulous branches. The first-year branches of var. genhensis are about two times longer than those of var. gmelini. The tree reaches 20 meters tall and 50 cm in diameter in favorable conditions. Type: CHINA. NeIMONGGu: Genhe, 4 Jul 1986, 8. ¥. Li 861-643 (HOLOTYPE: NEFI!). Larix gmelini (Rupr.) Rupr. var. genhensis is sparsely distributed in Genhe, Da Xingan Ling, Neimonggu, China (Fig. 3). It usually grows in wetlands or occasionally in the Larix gmelini (Rupr.) Rupr. var. gme/ini forest on lower SIDA 16(1): 183 — 190. 1994 SIDAL6(1) 1994 FIG. 1. Larix gmelini (Rupr.) Rupr. var. gerhensis. slopes. It has withstood temperatures of -45°C. This larch should receive horticultural attention because of pendulous branching and cold tolerance. Pinus sylvestris L. var. manguiensis S.Y. Li & K. T. Adair, var. nov. (Fig. 2) Differt a Pinus sylvestris L. var. sylvestris, var. mongolica Litvin., var. sylvestrifomis (Takenou- chi) Cheng et C.D. Chu foliis 0.5—4.0 cm longis. Pinus sylvestris L. var. manguzensis is one of the shortest leaf pines in the world. This variety can be distinguished from other varieties of Pinus sylvestris by its extremely short and slightly or non-twisted needles (Table 1). LI AND ADAIR, Chinese plants 185 * 4 iy Tee 3 . ~ x te. : f z = is mJ i», 3 ° ll FIG 2. Pinus sylvestris L. var. manguiensis. Type: CHINA. NEIMONGGU: Mangui, 26 Jun 1986, S. ¥. Li 861-917 (HoLotyPe: NEFI}). Additional specimens examined: CHINA. NEIMONGGU: Jinhe, 20 Jun 1986, S. Y. Li 860- 300 (NEF). Pinus sylvestris L. var. manguiensis (SHORT-LEAF SCOTCH PINE) occurs in pure stands on dry rocky slopes (400-800 m at elevation) in Mangui and Jinhe, Da Xingan Ling, Neimonggu, China (Fig. 3). Pznus sylvestris L. var. manguiensis predominates the canopy and 1s occasionally associated with Larix gmelini. The regeneration seedlings under canopy are common, especially in gaps. The common shrubs in undergrowth are Rhododendron mucronulatum Tuarcz., Spiraea sericea Tarcz., Rosa acicularis Lindl., Vaccinium vitis-idaea L.,and Rubus sp. Main associated herbaceous plants include: Carex spp. Calamari sp., Iris ensata Vhunb., Pyrola incanata Fisch., Artemisia sp., Convallaria majalis L., Fragaria eee Lozinsk., and Sanguisorba officinalis L The new pine is a beautiful large tree with a short crown and a straight reddish-brown trunk. It is usually 20-25 m tall and 50 cm in diameter. It is adapted to an extremely continental climate. It has withstood tempera- tures of -47°C. The new pine should be managed for timber. Polemonium hingganicum (P.H. Huang & S.Y. Li) S.Y. Li & K.T. Adair, c omb. et stat. nov. Basionym: Polemonium boreale Adams subsp. hingganicum PH. 186 SIDAI6(1) 1994 Neimonggu er Heilongjiang © Pinus sylvestris L. var. manguiensis = Larix gmelini (Rupr.) Rupr. var. genhensis FIG. 3. Distribution of Pinus sylvestris L. var. manguiensis and Larix gmelini (Rupr.) Rupr. var. lak Heilongjiang FIG. 4. Distribution of Polemonium hingeanicum. Huang & S.Y. Li, Bull. Bot. Res. 5:151-154. 1985. Type: CHINA. HEILONGJIANG: Da Xingan Ling, S.¥. Lz 841-372 (HoLotyPE: NEFI!). Polemonium Ok et (XINGAN JACOBS-LADDER) was named by P.H. Huang and 8.Y. Li in 1985 and treated as a subspecies of the circumboreal P. boreale. ae we reexamined this taxon with its closely related taxa. The data (Table 2) indicate that the taxon should be treated as a separate LI AND ADAIR, Chinese plants 187 TABLE 1. Gradient Analysis of needle length of Pinus sylvestris population with geographic location in Chin Sample Specimen Needle Length (cm) Location Size Cited Range ean+s.d.* Taxon Name Mohe, Heilongjiang 1030 NEFI 3589 3.1-7.1 5.35+0.94 — var. sylvestris Gulian, Heilongjiang 630 Wang 109 3.5-8.0 5.5741.30 — var. sylvestris Mordaga, Neimonggu 760 Li 861-604 2.0-6.2 4.064#1.29 — var. sylvestris Nenjiang, Heilongjiang 160 Wang 105 3.4-8.4 6.08+0.93 — var. sylvestris Jixi, Heilongjiang 76 NEFI 0005 0-7.0 4.6241.78 — var. sylvestris Mangui, Neimonggu 1078 Li 861-917 0.8-3.7 2.43+0.73 var. manguiensis Jinhe, Neimonggu 950 Li 860-300 0.5-3.9 2.8140.58 — var. mangutensis Hailar, Neimonggu 108 = NEFI 630004 4.0-8.9 7.41+1.10 — var. mongolica Honghuarji, Neimonggu 240 ~=NEFI 630087 4.6-7.0 5.76+1.10 — var. mongolica Yirsi, Neimonggu 85 NEFI 630082 3.1-7.3. 5.09+0.83 — var. mongolica Guegang, Neimonggu 88 Dong 355, 357 2.1-7.9 5.244+1.21 var. mongolica Antu, Jilin 350 NEFI 4001 3.4-7.7 4.98+0.80 — var. sy/vestriformis Harbin, Heilongjiang 667 Li H-0086 7.3-11.0 9.48+0.73 — var. sylvestriformis * standard deviation species because it can be distinguished by densely glandular pubescence in parts of the plant, large plant size, mostly stem leaves, a long pedicel, short corolla, and exposed stamens and stigma. Polemonium hingganicum is similar to Polemonium pulchellum in Siberia, but differs from the latter with large size, mostly stem leaves, campanulate corolla, and exposed stamens and stigma. Penthorum sedoides L. Penthorum isa perennial herbaceous genus witha disjunct distribution be- tween eastern Asia and eastern North America. Although it isasmall genus, the taxonomic treatment of species is in dispute. Gray (1846) once doubted that the Asian plants (P. chinense and P. humile) would be the same as the eastern North American P. sedoides. Maximowicz (1883) treated P. chinense as a variety of P. sedoides, and considered P. humileas possibly based on abnormal specimens. However, in this century, most scholars tend to recognize two species P. sedoides and P. chinense in the genus. Li (1952) clearly stated that P. chinense differs from P. sedoides in that stems are usually simple or few- ranched, narrow leaves, has fewer flowered more corybose-like inflores- cences and white flowers. In P. sedoides, as Li mentioned, the stems are usually many-branched, the leaves broader, the inflorescences more paniculate and the flowers greenish. Li’s opinion is widely accepted by many scholars. The natural range of P. chinense is the northeastern Asia, including Far East Russia, Northeast and North China, Korea, and Japan. We examined all available Chinese specimens of Penthorum and related literature and collected TABLE 2. Morphological comparison of Polemonium hingganicum and its closely related species. P. acutiflorum P. boreale P. hingeanicum P. pulcherrimum P. pulchellum Distribution Habitat Stature Stem Leaves Leaflet Leaf surfaces Pedicel length Calyx Length Calyx lobe length Calyx surfaces Corolla length Stamens Stigma Specimens Cited References Europe, Siberia, & western North America moist & grassy slopes 10—80 cm glabrous below & glandular pubescent abov glabrous >calyx 6-11 mm > tube glabrous 11-22 mm not exposed not exposed Ohwi 1965, Polunin 1959 circumboreal dry rocky slopes 5—22 (40) cm glandular pubescent bove d mostly basal 13-23 viscid-pubescent tube pubescent 14-25 mm not exposed not exposed Polunin 1959, Welsh 1974 boreal region of China moist & fertile sites 30-90 cm densely short glandular pubescent mostly stem 15-25 glandular pubescent >calyx 5-7 mm =tube pubescent 6-10 mm exposed exposed Li 841-9008 Huang & Li 1985 western North America calcareous rocky slopes 10-40 (55) cm spreading-hair to glandular pubescent mostly basal up to 31 viscid-pubescent >calyx i—7 mm > tube glabrous 8-13 mm not exposed not exposed Raines 1313 Polunin 1959, Welsh Siberia rocky hills 8-25 cm short glandular pubescent mostly basal 11-25 glandular pubescent =calyx 4-8 mm >tube pubescent 7-12 mm not exposed not exposed Polunin 1959 1974 881 FOGI (LOT VAIS LI AND ADAIR, Chinese plants 189 hundreds of accounts of specimens during 1985-1987. In 1989, we exam- ined the American P. sedozdes in both herbarium and field. Finally, we found the Asian and American plants to be almost the same in all characteristics in nature, with the exception of leaf shape. However, descriptions are often individual in flower color. Asian plant flowers have been observed and recorded as white, yellow white, and light green; American ones have been described as green or yellow-green in much of the literature (e.g. Correll and Johnston 1979). Some field notes of specimens are also different: e.g. McCrary 1044 (Texas, 20 Jul 1984) recorded the flowers as yellow-green, and Nixon 15992 (Texas, 10 Sep 1987) recorded flowers as white. According to our field observations in the last few years, the corollas of Chinese and American plants are both white in nature, but they are often inconspicuous and fall early while other parts of flowers are all green in color. In most cases, the white sepals do not remain or at most are restricted to the upper flowers in the inflorescences so that many observers recorded the flowers as green. Thus, flower color is not a character on which to distinguish between Asian and American plants. Further, according to our observations, some Asian plants have simple or few-branched stems as Li (1952) described. Others have many-branched stems especially in dry and disturbed sites. Therefore, it seems there is no difference in stems between Asian and American populations. Also, Li’s descriptions on the fewer flowered more corybose-like inflorescences (Asian plants) and more paniculate (American plants), seem a result of partial observations. According to our specimen examinations, there is no distinct difference in inflorescences between Asian and American taxa. In addition, anatomical data (Haskins and Hayden 1987) failed to show any significant difference between these two taxa Actually, the only morphological difference between Asian and American taxa is the former has narrow and long willow-like leaves, while the latter has wide and shorter glabrous leaves. Geographically, they occupy similar habitats on separate continents. Thus, we treat the Asian taxon as a geographic subspecies of the American species. As a result, Penthorum becomes one of five monospecific disjunct genera in eastern Asia and North America. Penthorum sedoides L. subsp. sedoides Distribution. Eastern United States. eae Lean examined. UNITED STATES. Arkansas: D. Demaree 34302 (ASTC). lowa: H.B be iets Sto) Louisiana: C. Smith 1853 (ASTC). Massachu setts: H.E. Abhles 90326 (A STC). Texas: L.E. Bro 3732 (ASTC a M. McCrary rT (ASTC); E.S. Nixon 7649, 10717, 15992 (ASTC Pk es nand C. Burandt 7078 (ASTC); E.S. Nixon and M. McCrary 3591 (ASTC); E.S. Nixon and J.R. Ward ee 12390, 12422, 12434, 12547, 12922, 13043, 13136 (ASTC); C. Waters 174 (ASTC). 190 SIDA16(1) 1994 Penthorum sedoides L. subsp. chinense (Pursh) S.Y. Li & K.T. Adair, et stat. nov. BasionyM: Penthorum chinense Pursh, Fl. Amer. Sept. 323.1811; Koma-rovy, Fl. Manshur. 2:403. 1903; Kitagawa, Lin. Fl. Manshur. 246. 1939; Liou, Key Pl. NE China 122. 1959. Penthorum sedoides L. var. chinensis (Pursh) Maxim., Mélanges Biol. Bull. Phys.-Math. Acad. Imp. Sci. Saint-Pétersbourg. 11:774. 1883. Distribution. Korea, Japan, eastern China southward to Indochina. d. CHINA. Heilongjiang: S.L. Dong 11040 (NEFI);S.Y. Li 123, 84- 103 (NEFD, S.Q. Nie 964 (NEFD; Y.L. Zhang 1926 (IFP). Jilin: S.Y. Li 10043 (NEFI). Liaoning: 8S. Y. Li 853007 (NEFD. ACKNOWLEDGMENTS We are grateful to Li-Guo Fu (Chinese Academy of Sciences, Beijing, China), Barney L. Lipscomb (BRIT), Shao-Quan Nie (NEFD), and Elray S. Nixon and Darrel L. McDonald (ASTC) for reviewing the manuscript. Special thanks are due to Puhua Huang and Fucheng Shi (NEFI), Qingsheng eng (Heilongjiang Institute of Traditional Chinese Medicine, Harbin, China), Shulan Yin (Ministry of Forestry, Beijing, China), Jing Zhan (Mangui Forestry Bureau, Neimonggu, China), and Y. Wang (Genhe Forestry Bureau, Neimonggu, China) for aiding field investigations in Da Xingan Ling during 1984-1987. REFERENCES CorreLl, D.S. and M.C. JOHNSTON. 1979. Manual of the vascular plants of Texas. University of Texas, Dallas Gray, A. 1946. Analogy between the flora of Japan and that of the United States. Amer. J. fee Sci. II. 2:135-136. (Reprinted in Graham 1972, and Stucke 78). Haskins, M.L. and W.J. Haypen. 1987. Anatomy and affinities of Penthorum. Amer. J. Bot. 74:164-177. Hune, P.H. andS.Y. Lr. 1985. A new eee of Polemonium L. from the Greater Khingan Mountains, China. Bull. Bot. Res. 5:151-154. iraGawa, M. 1939. Lineamenta Florae one ae. Hsiuking. Komaroy, V.L. 1903. Flora of Manshuria II. Leningrad. Li, H.L. 1952. Floristic ee between eastern Asia and eastern North America. Trans. Amer. Philos. Soc. 42 20. Liu, S.E. (ed.) 1959. Key to plants in Northeast China. Science Press, Beijing. Onw!, J. 1965. Flora of Japan. Smithsonian Institution, Washington, D.C. PoLunin, N. 1959. Circumpolar Arctic flora. Clarenden Press, Oxford, Wetsu, S.L. 1974. Anderson’s flora of Alaska and adjacent parts of Canada. Brigham Young University, Utah A NEW SPECIES OF ANEMONE (RANUNCULACEAE) FROM CENTRAL TEXAS CARL 8. KEENER 208 Mueller Laboratory Pennsylvania State University University Park, PA 16802, U.S.A. BRYAN E. DUTTON Department of Botany University of Maryland College Park, MD 20742, U.S.A. ABSTRACT Five species of Anemone with tuberous rootstocks occur in Texas: Anemone caroliniana Walter, A. berlandieri Pritzel, A. tuberosa Rydberg, A. edwardsiana Tharp, and a new species, A. okennonii Keener & Dutton, endemic mainly to the Edwards Plateau in west-central Texas. RESUMEN Existen en Texas cinco especies de Anemone con rizomas tuberosos: Anemone berlandieri Pritzel, A. caroliniana Walter, A. edwardsiana Tharp, A. tuberosa Rydberg, y una nueva especie A. okennonit Keener & Dutton endémica principalmente de la Meseta Edwards en el centro-oeste de Texas INTRODUCTION The genus Anemone L. consists of ca. 150 species of perennial herbs featuring one or more radical leaves, involucrate peduncles, petaloid sepals, and achene fruits. Ina recent preliminary reclassification based on restriction site varia- tion and morphology of the subgeneric taxa of Anemone, Hoot et al. (1994) recognized two subgenera and seven sections. Section Anemone consists of four taxonomically unnamed groups, one of which (the Coronaria group) is characterized by distinctive tuberous rootstocks, typically heteromorphic 1—3 ternate radical leaves, simple or branched scapes with sessile involucral bracts positioned near the middle of the peduncle, numerous linear to obovate sepals, and numerous, usually densely woolly achenes borne in ellipsoidal to cylindroidal heads. According to Hoot etal. (1994), the ca. 18 species of the Coronaria group of section Anemone range from southern Europe to Mongolia, central and southern United States and adjacent Mexico, and South America below 10° south latitude (Hoot 1991; Joseph and Heimburger 1966). SIDA 16(1): 191 — 202. 1994 192 SIDA16(1) 1994 Texas has five species of anemones with tuberous rootstocks: A. caroliniana Walter, A. berlandieri Pritzel, A. tuberosa Rydberg, A. edwardsiana Tharp, and a new species, A. okennonii Keener & Dutton, herein described from the Edwards Plateau. The aim of our paper is to clarify the taxonomy of these species of Anemone in Texas, to provide keys to the taxa, and to describe the new species. Joseph and Heimburger (1966) reviewed the cytotaxonomic structure of these species (except A. ofennonii), and provided (p. 908) a helpful comparative chart and distribution maps; the reader is referred to their paper for additional details. The five species treated herein are distinguished chiefly on the basis of leaf dissection, scape pubescence, relative similarity of involucral bracts to the basal leaves, style length, and sepal number and size (see Table 1 for additional details). Until the species of Axemone with tuberous rootstocks are studied world- wide, it would be premature to speculate on phylogenetic relationships of the species treated in our paper (See Hoot et al. [1994] for further analysis of this problem). We believe, however, that Anemone okennonii is taxonomi- cally closest to A. edwardsiana, but whether these two species are closest phylogenetically remains in question. KEY TO TEXAS SPECIES OF ANEMONE WITH TUBEROUS ROOTSTOCKS ay .Involucre below middle of scape at anthesis; scape nearly glabrous below involucre; involucral bracts similar to at least some radical leaves; styles as long or longer than the ovoid achenes _ A. caroliniana —. .Involucre above middle of scape at anthesis; scape glabrous to pubescent below involucre; involucral bracts dissimilar or similar to radical leaves; se less than half as long as the orbicular, flat achenes. 2. Scape simple, bearing 1 flower; involucral bracts distinctly dissimilar to radical leaves; scape densely ae ee below involucte «.......:0s.0000 2. A. berlandieri 2. Scape usually branched, bearing 2 or e flowers per stem; involucral bracts similar or dissimilar to or ce scape nearly glabrous to a below involucre. . Involucral bracts similar to radical leaves; radical leaflets narrow, 1— cm wide; achenes 3—4 mm wide; rare in Texas (El Paso Co.) 3. A. tuberosa . Involucral bracts dissimilar to radical leaves; radical leaflets ore .5—2.5 cm wide; achenes 2—3(3.5) mm wide; uncommon, central to oe Tex ce eiaan iy ternate; leaflets usually crenate-dentate; ae HOWERINE 5 usually al the SalI LUITIC as | t 10— 20, oblanceolate, 2-3 mm wide 4. A. edwardsiana 4. Leaves (2)3 ternate; leaflets usually incised with sharp teeth; lateral flowering branches maturing much later than the central; sepals 7— 11, oblong, 3-4.5 mm wide 5. A. okennonti KENNER AND DUTTON, Anemone okennonii 193 1. Anemone caroliniana Walter, Fl. Carol. 157-158. 1788. Slender, apically villous herbs; stems simple, 0.5—3.5 dm tall at anthesis, from rhizomes bearing a series of small, +oblong to globose tuberous rootstocks 0.5—1 cm long and 0.5 cm thick. Radical leaves 1—2 ternate, to 3.5 cm wide; petioles 3—9 cm long. Leaflets 1-2 cm wide, glabrous, deeply 2—3-parted or divided, segments variously cleft, lobed or toothed, ultimate segments broad to narrow with crenate-dentate margins to variously linear with acute tips, sessile to long petiolulate. Involucral bracts usually similar to at least some radical leaves, reduced, opposite or whorled, sessile to subsessile, sparsely pilose, positioned below middle of scape at anthesis. Flowers solitary; scape villous above involucre, nearly glabrous below. Sepals 10-30, greenish white or pink to purplish blue, linear-oblong, 10-20 mm long and 1-6.5 mm wide; anthers yellow, 0.5—0.75 mm long. Fruiting heads ellipsoidal, to 2 cm long and 1.5 cm broad; achenes brownish, turgid, +ovoid, ca. 2 mm wide, densely woolly; styles greenish (lavender-tipped), erect, filiform, as long as or longer than achenes, projecting beyond the woolly tomentum; 27=16 Joseph and Heimburger 19 February to April. Prairies, moist sandy soils, calcareous dry grasslands, rocky hillsides, and alluvial flats, South Dakota to Texas, east to South Carolina and Georgia. Anemone caroliniana is distinguished by its rhizomatous rootstocks bear- ing a series of small tubers, its heteromorphic radical leaves some of which resemble the involucral bracts, its simple scape bearing involucral bracts below the middle, its scape pubescent above the involucre but glabrous below, and its densely woolly ovoid achenes bearing relatively long styles (Table 1). In Texas, A. caroliniana occurs chiefly in the eastern third of the state, with some populations in the west-central and northern regions (see Joseph and Heimburger 1966, Fig. 3, fora distribution map of this species). 2. Anemone berlandieri Pritzel, Linnaea 15:628—629. 1841. A. heterophylla N I] Jum insyn., Torrey & A. Gray, Fl. N. Amer. 1:12. 1838. A. caroliniana Walter var. heterophylla Torrey & A. Gray, Fl. N. Amer. 1:12. 1838 A. decapetala Arduino var. heterophylla (Torrey & A. Gray) Britton & Rusby, Trans. New York Acad, Sen-/17- 1697 Similar to, though often larger than A. caroliniana. Stems simple, (1)3—5 dm tall at anthesis, from thick clavate to oblong, tuberous rootstocks 2—4 cm long and ca. 1 cm thick. Radical leaves 1(2) ternate; petioles 3—20 cm long. Leaflets 2-4 cm wide, pubescent, often 2—3 lobed or parted, the cuneate segments usually broad or rounded apically with dentate or serrate to crenate margins, sessile to long petiolulate. Involucral bracts dissimilar to radical leaves, ultimately dissected into linear segments with acute tips, 194 TABLE 1. Comparison of Texas Anemone species wi SIDA16(1) 1994 th tuberous rootstocks. Anemone caroliniana berlandieri tuberosa edwardstana okennonit STEM height (dm) 0.5—3.5 (1)3-5 jae 3-5 2-3 branches one none 1 (rarely 2) 1-3 1-3 time of flowering central before — lateral with — central before latera central eral PUBESCENCE above involucre villous villous villous densely villous villous below involucre +glabrous villous subglabrous subglabrous glabrous to aa TUBEROUS ROOTSTOCKS shape +oblong to clavate to oblong- oblong- oblong- globose oblong obovate obovate obovate length (cm) 0.5-1.0 2-4 1.52.0 2-4 1-3 thickness (cm) 0. 1 1 1-2 l RADICAL LVS divisions 1—2 ternate 1(2) ternate 1—2 ternate 1(2) ternate (2)3 ternate leaflet width |- 2-4 1-2 2-2.5 0.5-1.5(2.5) cm) pubescence glabrous pubescent subglabrous glabrous subglabrous; marginally ciliate petiole length 3-9 3-20 5-7 8-15 5-10 BRACTS position below above above above mid-scape mid-scape mid-scape mid-scape mid-scape pubescence sparsely pilose pilose thinly pilose glabrous to appressed pilose sparsely pilose sessile/petiolate — +sessile sessile short-petiolate sessile short-petiolate similar to basal yes (some) no yes no leaves SEPALS number 10—30 7 8-10 10-2 al shape linear-oblong linear-oblong — linear-oblong — oblanceolate oblong length (mm) 10—20 7-15(20) 10-14 10-16 —12 ‘olor various various greenish white greenish white greenish to purplish to bluish white FRUITING HEAD shape ellipsoidal cylindroidal +cylindroidal ellipsoidal oblong- to oe tare cylindroidal ellipsoidal length/width = 1.5—2/0.5-1.5 2—3.5/0.5-1.3. 1.5—3/1- 1.5—3/0.6-0.8 1-3/0.5— cm ACHENES shape si + ovoid flat, orbicular — flac, orale: flat, orbicular flat, ovate width (mm) 2 2— 2—3(3.5) 2- pubescence ee woolly densely woo ly lence villous sparse densely white KENNER AND DUTTON, Anemone okennonti 195 TABLE 1. continued Anemone caroliniana berlandieri tuberosa edwardsiana okennonti STYLES shape filiform subulate filiform filiform filiform style/ach. length 51.0 ca <1/2 ca. 1/3 ca. 1/4 ANTHESIS Feb. to Apr. Feb. to Apr. Apr. to May Feb. to Apr. Mar. to Apr. HABITAT moist acid soils moist high dry canyons, high dry and calcareous alkaline soils rocky slopes in moist open ledges prairies shaded and slopes alkaline soils pilose, positioned above middle of scape at anthesis. Flowers solitary; scape villous throughout. Sepals 7-17, greenish white to pink or purplish blue, linear-oblong, to 7—15(20) mm long and 1.5—5 mm wide; anthers yellowish brown, 0.5—1 mm long. Fruiting heads cylindroidal, 2-3.5 cm long and 0.5—1.3 cm broad; achenes flat, orbicular, 2-3 mm wide, densely woolly; styles greenish (lavender-tipped), subulate with bent tips, ca. 1/3 length of achenes, not projecting beyond the dense woolly tomentum; 2”=16 (Joseph and Heimburger 1966). February to April. Limestone hills, grassy knolls, and stony ground, Texas and Oklahoma eastward to (rarely) North and South Carolina. Anemone berlandieri is distinguished by its ternate radical leaves with broad segments having typically crenately toothed margins, its simple scape bearing involucral bracts above the middle of the more or less uniformly pubescent scape, its bracts distinctly unlike the radical leaves, and its densely woolly orbicular, flat achenes bearing styles ca. 1/3 length of the achene (Table 1). In Texas, A. berlandieri occurs throughout the state except for the extreme western portions (i.e., High Plains and Trans-Pecos, Mountain and Basin regions; see Map 1, Correll and Johnston 1970). The taxonomy of this species is complicated, owing largely to whether the North American plants are conspecific with the South American species recognized as A. decapetala Arduino (Britton 1891), and the adoption of the name A. heterophylla, which is listed by Torrey and A. Gray (Fl. N. Amer. 1:12. 1838) merely as a synonym for A. caroliniana Walter var. heterophylla Torrey & A. Gray. Joseph and Heimburger (1966) clarified distinctions between the North and South American species, and Keener (1975) re- viewed the legitimacy of the name A. heterophylla. 3. Anemone tuberosa Rydberg, Bull. Torrey Bot. Club 29:151-152. L902 A. sphenophylla sensu Britton, Ann. New York Acad. Sci. 6:220. 1891, p.p., non Péppig, Frag. syn. 27. 1833 196 SIDA16(1) 1994 Robust, apically villous herbs; stems simple below, usually branched above involucre, 1-3 dm tall at anthesis, from brownish, oblong-obovate tuberous rootstocks 1.5—2 cm long and 1 cm thick. Radical leaves several, 1-2 ternate; petioles 5-7 cm long. Leaflets 1-2 cm wide, subglabrous, variously parted or cleft, with cuneate-obovate segments having acute tips, sessile to petiolulate. Involucral bracts 3, similar to radical leaves in dis- section, reduced, short-petiolate, thinly pilose, positioned above middle of scape at anthesis; secondary branches | or more, 2-bracteate. Flowers 1-several per scape; lateral flowering branches maturing later than central flower; scape villous above involucre, subglabrous below. Sepals 8—10, greenish white to purplish, linear-oblong, 10-14 mm long, 3—6.5 mm wide; anthers yellowish brown, ca. | mm long. Fruiting heads cylindroidal to ellipsoidal, 1.5—3 cm long and 1—1.5 cm broad; achenes flat, orbicular, 3-4 mm wide, densely villous; styles lavender, +erect, filiform, less than 1/2 length of achenes, not projecting beyond the dense woolly tomentum; 2”=16 Joseph and Heimburger 1966). April to May. High dry rocky slopes, southeastern California, southern Nevada, and southwestern Utah, southeastward to extreme western Texas (El Paso Co.), where it occurs on igneous rocky slopes (Joseph and Heimburger 1966). Anemone tuberosa is distinguished by its ternate to biternate radical leaves with acute-tipped oblong-ovate segments, its relatively large involucral bracts similar in dissection to the radical leaves, its typically branched scape withan involucre above the middle at anthesis, its scape pubescent above the involucre but more or less glabrous below, and its densely villous, flat, or- bicular, achenes having styles less than 1/2 the length of the achene (Table 1). Britcon (1891) regarded the North American populations conspecific (as A. sphenophylla) with similar populations in Chile, but Rydberg (1902) pointed out that in the Chilean plants “the whole scape is decidedly pubescent with ascending hairs and the segments of the involucral bases are narrower.” However, Lourteig (1951: 564) regarded A. sphenophylla as a taxonomic synonym of A. decapetala Arduino. 4, Anemone edwardsiana Tharp, Amer. Mid]. Naturalist 33:669. 1945. Slender, apically villous herbs; stems simple below, usually branched above involucre, 3—5 dm tall at anthesis, from brownish, oblong-obovate tuberous rootstocks 2—4 cm long and 1—2 cm thick. Radical leaves several to many, 1(2) ternate; petioles 8-15 cm long. Leaflets reniform in outline, 2—2.5 cm wide, glabrous, variously 2—4 cleft, lobes crenate-dentate to sharply toothed, sessile to petiolulate. Involucral bracts 3, sessile, dissimi- lar to radical leaves, 2-5 cm long, ultimately dissected into oblanceolate KENNER AND DUTTON, Anemone okennonii 197 segments with rounded to acute tips, glabrous to sparsely pilose, positioned above middle of scape at anthesis; secondary branches 1-3, 2-bracteate. Flowers l-several per scape; lateral flowering branches usually maturing with central flower; scapes villous above involucre, subglabrous below. Sepals 10-20, greenish white to bluish, oblanceolate, 10-16 mm long, 1.5—3(4) mm wide; anthers bright yellow, 0.5—1 mm long. Fruiting heads ellipsoidal tocylindroidal, 1.5—3 cm long and 0.6—0.8 cm broad; achenes flat, orbicular, 2—3(3.5) mm wide, typically sparsely pubescent to glabrous; styles erect to horizontal, filiform, ca. 1/3 length of achenes, not projecting beyond the woolly tomentum; 27=16 Joseph and Heimburger 1966). February to April. Alkaline soils in moist shaded canyons; shaded, moist, rocky limestone bluffs and ledges along eastern edge of Edwards Plateau (R.J. O’Kennon, pers. comm.). Two varieties: a. Achenes woolly, +dull var. edwardsiana a. Achenes glabrous, vernicose (polished) var. petraea 4a. Anemone edwardsiana var. edwardsiana Range and habitat of the species. 4b. Anemone edwardsiana var. petraea Correll, Madrofio 19:189. 1968. Tall, spindly plants with glabrous, vernicose achenes endemic to moist rocky crevices along Curry Creek near Kendalia, Kendall Co., Texas. Addi- tional population and cytotaxonomic studies are desirable to establish more clearly the range of this taxon as well as its phylogenetic relationships to other species in this complex. The most polymorphic of Texas anemones, A. edwardsiana combines fea- tures of both A. twberosa and A. berlandieri (cf. Table 1). It is like A. tuberosa in having a branched scape glabrous or slightly pubescent below the in- volucre, and like A. berlandier7 in having involucral bracts with long, narrow, sharply toothed segments quite dissimilar from the radical leaves. Achene shape, style length, and position of the involucral bracts are similar to those character states in both A. berlandieri and A. tuberosa. The combination of branched scapes, involucral bracts dissimilar to the radical leaves, and relatively few sparsely woolly to glabrous achenes demarks A. edwardsiana from the other anemones in Texas. Joseph and Heimburger (1966) pointed out that A. edwardsiana not only combines features of A. twberosa and A. berlandteri (= A. heterophylla in their treatment), but also “shows an intermediate geographic range and altitude preference.” They suggested that A. edwardsiana may bea hybrid derivative of A. tuberosa and A. berlandieri. However, based on restriction site analyses, Hoot et al. (1994) claimed that A. edwardsiana is a sister species to A. 198 SIDA16(1) 1994 berlandieri and possesses “no rDNA fragments in common with A. twberosa.” Additional biosystematic studies are desirable to clarify the phylogenetic relationships of this complex. Because of glabrous, vernicose achenes, Correll (1968) segregated several populations occurring near Kendalia in Kendall Co., Texas, as A. edwardsi- ana Vat. petraea Correll. These populations appear to be an ecotypic phase of the polymorphic A. edwardsiana. 5. Anemone okennonii Keener & Dutton, sp. nov. (Figs. 1,2) Herba beribus | I bl obovata, 1-3 cm longa, | cm crassa. Folia radicalia pluria, triternata, eee perl usque ad 10 cm longi. Foliola 0.5-1.5 (2.5) cm lata, profunde 2—4-fida; lobi cuneati-acuti; petioluli usque ad 2 cm longi. Scapi 2-3 dm nn basibus glabratis, apicibus pubescentibus; brac teae primariae involucri 3, 2-5 cm longae; ramuli axillares (0)1—2(3), unusquisque involucro secundario bi- ca ees ees 1.5—2.5 cm lati; ceteri parviores; se ne viridi-alba. Capitula liptica, 1-3 cm longa. Achenia lata, ovata, marginata, brevirostria; rostra ae minusve erecta, pilis basibus sub-aequantibus corpora pubescentia. Slender, hae a uous herbs; stems simple below, usually 1-3 branched above involucre, 2— tall at anthesis, from brownish oblong-obovate tuberous rootstocks i cm long and 1 cm thick. Radical leaves several to many, (2)3 ternate; petioles 5-10 cm long. Leaflets 0.5-1.5(2.5) cm wide, subglabrous but marginally ciliate, deeply 2-4 cleft, lobes cuneate-acute, reddish-glandular, petiolules to 2 cm long. Primary involucral bracts 3, short-petiolate, dissimilar to radical leaves, 2-5 cm long, variously cleft into linear, acute-tipped segments, appressed pilose, positioned above middle of scape at anthesis; secondary branches (0)1—2(3), 2-bracteate. Flowers 2- several per scape; /ateral flowering branches maturing later than central flower, scapes pubescent above involucre, glabrous or subglabrous below. Sepals 7— 11, greenish white, oblong, 6-12 mm long, (2)3—4.5 mm wide; anthers pale yellow, fading to creamy brown, 0.75—-1 mm long. Fruiting heads oblong-ellipsoidal, 1—3 cm long and 0.5—1 cm broad; achenes flat, ovate, 2-3 mm broad, densely white villous; styles lavender-tinted, +erect to horizontal, filiform, ca. 1/4 length of achenes, not projecting beyond the wooly tomentum. Chromosome num- ber unknown Type: U.S.A. Texas. Gillespie Co.: high on shallow roadcut in sandy loam, Fm 783, 2 mi S of Doss, 22 Apr 1993, Robert J. O’Kennon 11390 (HoLotyPE: BRIT; tsorypes, PAC, TEX; see Fig. 1). Named in honor of its discoverer, Robert J. O’Kennon, keen student of the flora of Texas. Representative specimens examined: U.S.A.: Texas: Brewster Co.: Glass Mts., 21 Mar 1941, Rose-Innes & Warnock 565 (SMU). Crockett Co.: 14.8 mi W of Ozona, 14 Mar 1949, Turner & Warnock 273 (SMU). Kimble Co.: Hwy 385, 2.5 mi S of Llano River, 9 Mar 1992, R. J. O’Kennon 8813 (PAC). Mitchell Co.: Lake Hollywood, 27 Mar 1945, Pohl 4744 (SMU). Pecos Co.: 20 mi NE of Ft. Stockton tc | McCamey, 27 Apr 1947, Warnock 5199 (SMU). KENNER AND DUTTON, Anemone okennonii 199 Tet S aD pe. : epee {2 CLE or ai ge a ae we Sa See Boe tov oo on =e Faeee base ONG —< Sans Aid, Sa Poste 3 Ses SE FIG 1. Anemone okennonii from the holotype. A) habit; B) primary involucral bracts; C) fruiting head. 200 SIDALG(1) 1994 FIG. 2. Habit of Anemone okennonit. KENNER AND DUTTON, Anemone okennonii 201 Terrell Co.: 31 mi S of Sheffield, 14 Mar 1949, Turner & Warnock 307 (SMU); 4.2 mi W of Sanderson, 14 Mar 1949, Turner G Warnock 319 (SMU). Val Verde Co.: 20 mi NNW of Del Rio, 31 Mar 1947, McVaugh 7729 (SMU). March to April. Dry sunny rocky limestone ledges, hills, and roadsides, in red sandy loam and igneous-based soils at an elevation of 490-530 m. Endemic from Gillespie and Mitchell cos., southwestward to Val Verde and Brewster cos., Texas, mainly in the Edwards Plateau vegetational area with outliers in the Rolling Plains and Trans Pecos, Mountain and Basin vegeta- tional areas (see Map 1, Correll and Johnston 1970). Anemone okennonit has been variously recognized as A. caroliniana Walter A. decapetala Arduino, A. edwardsiana Tharp, A. sphenophylla Poppig, A. tuberosa Rydberg, or possibly a hybrid between A. tuberosa and A. heterophylla Nuttall (= A. berlandieri Pritzel). Anemone okennonii differs from its nearest presumed congener, A. edward- siana, by its leaf dissection ({2}3 ternate vs. 1{2] ternate), its smaller leaflets (0.5—1.5 cm wide vs. 2—2.5 cm wide) with sharply cleft (rounded) lobes, its central flower maturing before (with) the lateral flowers, its sepal number (7-11 vs. 10-20), anther color (pale yellow vs. bright yellow), achene pu- bescence (densely white villous vs. sparse to glabrous), seedling leaves (similar to basal leaves vs. similar to bracts), relatively later blooming period (March to April vs. February to April), and habitat (drier, more exposed sites vs. more moist and shaded sites). The following chart summarizes differences be- tween A. okennonii and other Texas species of A vemone with tuberous rootstocks: 1. Radical leaves typically 3-cternate Distinctive 2. Leaves reddish-glandular on margins, tips istincti 3. Leaf divisions cuneate-acute Like A. tuberosa 4. Smaller flowers (cf. sepal length, below) Distinctive 5. Two or more flowers per stem Like A. edwardsiana and A. tuberosa 6. Central ower matures before lateral flowers Like A. tuberosa 7. Anthers pale yellow, fading to creamy brown =e 8. Styles lavender-tinted ve 9. Involucre with linear divisions Like a - ee 10. Sepals 7-11 Like A. tuberosa 11. Sepals oblong, 6-12 mm X (2)3—4.5 mm Distinctive Plant associates of A. okennonii include Acacia roemeriana Scheele, Aristolo- chia coryi 1.M. Johnston, Colubrina texensis (Torrey & A. Gray) A. Gray, Delphinium carolinianum Walter, Draba cuneifolia Torrey & A. Gray, Echinocereus enneacanthus Engelmann, Euphorbia acuta Engelmann, Ferocactus setispinus (Engelmann) L. Benson, Gilia rigidula Bentham, Haplopappus spinulosus (Pursh) DC., Melampodium lencanthum Torrey & A. Gray, Opuntia phaeacantha Engelmann, Prunus minutiflora Engelmann, Tetraclea coulteri A. Gray, and Thamnosma texana (A. Gray) Torrey (R. J. O’Kennon, pers. comm.). Anemone okennonii appears to have evolved from A. edwardsiana and may 202 SIDA1G6(1) 1994 actually be a xeric form of it (R. J. O’Kennon, pers. comm.). O’Kennon has suggested (pers. comm.) that millions of years ago when western Texas was more mesic than at present a precursor form of A. edwardsiana was most likely the predominant species of Anemone in this area. As western Texas became hotter and drier (and this trend continues at present), xeric forms began to evolve. Anemone he probably evolved first farther west, and then A. okennonii evolve t into the niche between the two. The three species seem not to oe anywhere in their ranges. The only species sympatric with A. okennonii is A. berlandieri, but there does not appear to be any intergradation between these two species. Critical biosystematic and population studies would be useful in establishing the most probable phylogenetic relationships among these five taxa of Texas anemones with tuberous rootstocks. ACKNOWLEDGMENTS Special thanks are due to Robert J. O’Kennon for pointing out the taxonomic problems in the Texas anemones and for providing specimens and data; to Dr. Sara Hoot and Dr. Ronald A. Pursell for reviewing an earlier manuscript; and to Robert FC. Naczi for providing the Latin diagnosis of Anemone okennonii. Any errors of fact or interpretation are strictly our responsibility. REFERENCES Britton, N.L. 1891. The American species of the genus Anemone and the genera which have ferred to it. Ann. New York Acad. Sci. 6:215—2 Correll, D.S. 1968. Some additions to the flora of Texas a Maco 19:187-192. [Type description mee edwardsiana Tharp var. petraea Correll, —________ and MC. Jonston. 1970. Manual of the ae oe of Texas. Texas Research ee Ret POOL. Fi Hoor, S.B. cee relationships in Anemone based on morphology and chloroplast DNA variation. Ph.D. dissertation, Univ. of Michiga . REZNICEK and J. PALMER. 1994. phylosenceie relationships in Anemone Ra snunculacene based on ieee and chloroplast DNA. Syst. Bot. Si 169-200. Josep, C. and M. Heimmpurcer. 1966. Cytotaxonomic studies on New World species of aye Cae Eviocephalus) with suse rootstocks. Canad. J. Bot. 44:899-928. . 1975. Studies in a Ranunculaceae of the southeastern United States. I. yee L. Castanea 40:36—4 Lourteic, A. 1951. nme teen de Sudamérica eee Darwiniana 9:397—G608. RybBerG, P.A. 1902. Studies on the Rocky Mountain flora—VII. Bull. Torrey Bot. Club 29:145-160. [Type description of Anemone vie Rydberg, 151-152.] DOCUMENTED CHROMOSOME NUMBERS 1994:1. KARYOTYPE OF JAIMEHINTONIA GYPSOPHILA (AMARYLLIDACEAE) ZAIMING ZHAO Department of Botany University of Texas Aisin, TX 7871S Oot ABSTRACT A somatic ehbomesonis count of the neeeney Sauces ernie gypsophila - L; ner(A along aryotype It a diploid count at of 2n=20 re is comprised af dine pairs sof latge submencentic a. somes, three pairs of medium metacentric chromosomes and four pairs of small chromo- somes, three of these metacentric and one acrocentric. RESUMEN Se presenta un OosOmic rico de | { | ita Jaimehintonia gypsophila B.L. Turner (Amaryllidaceae) junto con una fotograffa que muestra su Cariotipo. Presenta un recuento diploide de 2n=20 formado por tres pares de cromosomas grandes submetacéntricos, tres pares de cromosomas de tamafio medio metacéntricos y cuatro pares de cromosomas pequefios, tres de los cuales son metacéntricos y uno acrocéntrico. INTRODUCTION Jaimehintonia is a recently described monotypic genus of the Amarylli- daceae, tribe Allieae (Turner 1993). The single species, J. gypsophila, isa rela- tively rare edaphic endemic in to gypseous soils of northeastern Mexico. Its generic relationships are moot, but its author thought the species to be most closely related to Androstephium and allies, in the sense of Moore (1953). The present paper attempts to position the genus with more certainty using karyotypic data. MATERIALS AND METHODS Chromosome studies were made from root tips of seeds germinated on an agar nutrient medium ina petri dish. Seeds were collected at the type locality by George S. Hinton (Hinton et al. 20560, TEX). Root tips were placed in saturated 8-hydroxyquinoline solution for 4 hr, fixed in 3:1 (ethanol : acetic acid) about 4 hr, and transferred to 70% ethanol overnight. They were softened in 10% HCl at 60 C for 15 min, then squashed using standard procedures. SIDA 16(1): 203 — 205. 1994 204 SIDA16(1) 1994 RESULTS The chromosome number of this species is 2n=20 (Fig. 1). Its karyotype can be described as follows (Fig. 2): three pairs of large submetacentric chromosomes, three pairs of medium metacentric chromosomes, three pairs of small metacentric chromosomes, and one pair of small acrocentric chromosome. DISCUSSION Jaimehintonia appears to have a distinctive karyotype that, along with its morphological features, supports its recognition as a distinct genus. Among the tribe Allieae, chromosome numbers of 2n=20 are found only in two species of the genus Muilla, but chromosomes of the latter differ in shape from those of Ja:mehintonta. (Both M. maritima {[Torr.} S$. Wats. and M. transmontana Greene have five medium to long submetacentric chromo- somes, four short submetacentric, and one short nearly metacentric chromo- somes; Lenz 1975.) An anomalous count of n=20 pairs has been reported for Brodiaea lutea (Lind|.) Morton var. anilina (Greene) Munz (Niehaus 1965), but the chromosome numbers of most taxa of Brodiaea are based on x=6, 7 8, and 9 (Federov 1969). This, along with differences in floral morphology, appears to rule out a close relationship between Jasmehintonia and Brodiaea. In contrast, the similarity between Jasmehintonia and taxa of the An- drostephinm group (within the Allieae), as noted by Turner (1993), indicate FIG.1. Microphotograph of root tip cell of Jatmehintonia eypsophila (X 1000). ZHAO, Karotype of Jaimehintonia gypsophila 205 FIG.2. Karyotype of Jaimehintonia gypsophila, obtained from Fig.1. that a close evolutionary relationship is probable. Unfortunately, there are no chromosome number reports for Androstephinm. ACKNOWLEDGMENTS Iam grateful to Dr. Guy Nesom and Dr. B.L. Turner for helpful sug ges- tions in preparing the manuscript, and to Dr. Richard Starr for providing the nutrient medium. REFERENCES Feporov, A.A. (ed.). 1969. Chromosome numbers of flowering plants. Acad. Sci. U.S.S.R., Moscow. Lenz, L.W. 1975. Th I {Bi 7 | Mazila (Lili | g for Muzlla coronata and M. transmontana. Aliso 8:259—262. Moore, H.E., Jr. 1953. The genus Muwis//a and its allies. Gentes Herb. 8:263-294. Nienaus, T.F. 1965. In: Documented chromosome numbers of plants. Madrofio 18: 122-126. Turner, B.L. 19934 _Jaimebintonia (A llid Allieae), a new genus from northeastern Mexico. Novon 3:86-88. Z 206 SIDA1G6(1) 1994 BOOK REVIEW BeLousova, L.S. and L.V. Denisova. 1992. Rare Plants of the World. (ISBN 90-6191-482-5, hbk). A.A. Balkema Publishers, Rotterdam, Netherlands. $75 (hbk). 348 pp. Translated from Russian, this book describes about 2000 rare and endangered plants of the world. The authors intended to provide a larger list and more information about each plant than is available from the Red Data Book of Blents spay ished a ae ete i jon Conservation Union (IUCN). Selections are based g Data Book of Plants, and it uses the IUCN rarity categories. Iti is easy to peruse, with brief botanical descriptions of each species and common names printed in bold type. It is organized by continent with plant families in habetical order. Notes on distributions, ‘onservation status, endangerment listings, and causes of rarity are also included. There are sixty-four plates each with four taxa illustrated; ca of the plates are in color. Treated in the book are about 118 taxa out of 818 listed as extinct or most probably extinct and endangered in North America (1342 additional taxa are in the vulnerable category for North America). When examining endangered, threatened, or protected native plants of Texas one finds a number of taxa, 17 to be exact, not included. Out of the 21 species listed for Texas, four are treated in the book. Two out of nine cactus 2 Soe are listed for Texas, REICHENBACH’S ECHINOCEREUS (Echinocerens reichenbachti) and TOBUSCH’S ANCISTROCACTUS (Ancistrocactus tobuschi1). HINCKLEY’S OAK (Quercus hinckley’) and a Witb-RIcE (Zizania a are the other two taxa listed. e book would be eee bya disc ussion ag and endangerment in pants; for with at a summation conservation it pea Ford. NOTES VALIDATION OF KARINIA (CYPERACEAE)—It has been called to our attention that the name Karinia Reznicek & McVaugh, Flora Novo- Galiciana 13:386. 8 Dec 1993, may not be validly published because, even though monotypic, the sole species was not stated in the protologue also to be the holotype of the generic name. It is difficult to understand how the generic name could have any other type, but for the peace of mind of those who would consider the name not validly published, the name is here explicitly typified to fulfil this require- ment for valid publication, and the combination Karinia mexicana is remade now that the generic name is valid. Karinia Reznicek & McVaugh HototyPe: Karinia mexicana (Britton) Reznicek & McVaugh, comb. nov. Scirpus mexicanus C.B. Clarke ex Britton, Trans. New York Acad. Sci. 11:77. 1892. —A.A. Reznicek & R. McVaugh, University of Michigan Herbarium, North University Building, Ann Arbor, MI 48109, U.S.A. PENSTEMON THURBERI (SCROPHULARIACEAE) NEW TO TEX AS—Penstemon thurberi Torr. (P. scoparius A. Nels., Leiostemon thurberi Greene, P. ambiguus var. thurberi Gray) is a woody perennial distributed from Southern California through Arizona to Southcentral New Mexico and Northern Mexico. In New Mexico, it is found in the southwestern counties east to Dona Ana and Lincoln counties (Martin and Hutchins 1981). This is the first report of P. thurberi in Texas, a range extension of approximately 320 km from the nearest known site in Dona Ana county, New Mexico, to the two populations on the slopes of Little Pinto Canyon and Cinco de Mayo Canyon in the Chinati Mountains of Presidio county, Texas. The two populations, approximately 2 km apart, one with about 150 specimens and the other with about 300, are in loose sandy shale, surrounded by plants typical of the Chihuahuan Desert grassland. Collection data are: TEXAS. Presidio Co.: Little Pinto Canyon, Warnock 121, 9 Aug 1988 (Fig. 1). —Barton H. Warnock, Professor Emeritus, Department of Biology, Sul Ross State University, P.O. Box 158, Alpine, TX 79831, U.S.A.; John Mac Carpenter, Native Plant Society of Texas, P.O. Drawer 430, Ft. Stockton, TX 79735-0430, U.S.A SIDA 16(1): 207. 1994 208 SIDA16(1) 1994 FIG. 1. Penstemon thurbert. REFERENCES Kearney, T.H. and R.H. Persies. 1951. Arizona Flora. University of California Press, Berkel MartIN, WC. and C.R. Hurcnins. 1982. A flora of New Mexico. J. Cramer, in der A.R. Gantner Verlag i oiinen aed lochae. Wooton, E.O. and P.C. STaNDLEY. 1915. Flora of New Mexico. Contr. U.S. Natl. Herb. 19:1-794. Government Printing Office, Washington. EUPHORBIA GRAMINEA (EUPHORBIACEAE) NEW TO FLORIDA—Enphorbia graminea Jacq. apparently came into southern Florida during the late 1980s as a weed in horticultural stock. John Popenoe found plants without flowers in nurseries for several years (personal communica- tion), but it was not until Jan-Feb 1993 that Popenoe and I independently found flowering populations of E. graminea separated by several miles. Plants of this species were found again, and in new areas, during the fall of 1993 so it seems fair to regard the species as established in southern Florida. Euphorbia graminea isan erect annual reaching 35 cm in height with leaves well separated along the stem and branches. Stem leaves are 2-4 cm long, alternate and, on all Florida plants seen, are ovate with undulate margins. Leaves on the branches of the inflorescence are slightly shorter, opposite or nearly so and narrowly elliptic. Cyathia are terminal or solitary in the axes SIDA 16(1): 208. 1994 NOTES 209 of dichasia and about | mm long. A variable number (1-4) of glands is pro- duced along the upper edges of the cyathium, with many specimens having two glands on the cyathium. Glands have white appendages that are about equal in size to the glands. The capsule is held beneath the gland appendages until maturity and may not be apparent without close inspection. Mature capsules are about 3 mm long and wider than long. Euphorbia graminea is widespread from southern Mexico to northern South America, and has weedy tendencies throughout this range. It shows consid- erable variation in leaf shape and pubescence, leading to the proposal of several subspecific groups. There is no consensus on the validity of these subspecific taxa (Webster and Burch 1968), so no attempt was made to identify the Florida plants below the specific level. Within the genus Euphorbia (sens. lat.), E. graminea is the lectotype species of the section Cyttarospermum (Wheeler 1943). It has also been treated under the segregate genus Exmecanthus. This group is characterized by petiolate leaves and ecarunculate seeds with an areolate pattern of raised bumps covering the surface (Subils 1977). All collections from Florida so far have been made in cultivated sites. Within these cultivated areas, the Exphorbia evidently prefers bare patches. Seeds are dispersed locally by the explosively dehiscent capsules, but the pri- mary means of spread in southern Florida seems to be through human trans- port of ornamental plants, especially field grown trees. Voucher specimens: FLORIDA. Dade Co.: weed in residential lawn H 1,28 Jan 1993, Herndon 3338 (FTG); weed at the Dave Romney farm, Homestead, 5 Feb 1993, Popenoe 2476 (FTG); growing in planted bed along the right-or-way of Old Cutler Road, ca. mi S of Fairchild Tropical Garden, 20 Feb 1993, Popenoe 2480 (FTG); weed in potted plant by residence, Perrine, 20 Aug 1993, Herndon 3346 (F I am grateful to John Popenoe for sharing his knowledge of the early history of this species in Florida.—A/an Herndon, Department of Biological Sciences, Florida International University, Miami, FL 33199, U.S.A REFERENCES Susi1s, R. 1977. Las especies de Euphorbia de la Republica Argentina. Kurtziana 10:8—248. Wester, G.L. and D. Burch 1968. Family 97. Euphorbiaceae. In: Woodson et al., Flora of Panama, Part VI. Ann. Missouri Bot. Gard. 54:211—350. WHEELER, L.C. 1943. The genera of the living Bupho-biacéae: Amer. Midl. Naturalist 30:456- 5.05: SIDA 16(1): 209. 1994 210 SIDA16(1) 1994 MELICA SUBULATA (POACEAE: MELICEAE): THE FIRST REPORT FOR COLORADO—Melica L., a genus of + 80 species, is found in tem- perate regions throughout the world, except Australia (Clayton and Ren- voize 1986). Species of Melica are found in a variety of habitats, from woodland shade to dry stony slopes (Clayton and Renvoize 1986). There are 17 species of Me/ica in the United States. Boyle (1945) reported three taxa as occurring in Colorado, M. bulbosa Geyer ex Porter & J. M. Coulter, M. porteri Lamson-Scribner var. porteri, and M. spectabilis Lamson-Scribner. In the summer of 1992, Stanley and Gretchen Jones discovered a fourth taxon in the state, M. swbulata (Grisebach) Lamson-Scribner. This represents a range extension of approximately 400 km (250 miles) from the closest previously known locality in Fremont County, Wyoming, as reported by Hallsten et al. (1987). Melica subulata has not previously been reported as occurring in Colorado (Boyle 1945; Chase 1951; Harrington 1964; Weber 1987, 1990; Weber and Wittmann 1992). This species is found in mesic forests from Alaska south to Canada, Washington, Oregon, California, Idaho, Montana, Wyoming, and Colorado. The following is a key to identify the species of Me/ica found in Colorado. KEY TO MELICA OF COLORADO 1. Pedicels sharply bent; spikelets disarticulating below the glumes M., portert var. portert 1. Pedicels not sharply bent; spikelets disarticulating above the glumes ................0008 2 2(1). Lemmas strigose over the veins; palea about 1/2 as long as the lemma; lemma apex acuminate to subattenuate M. subulata 2. Lemmas glabrous or scabrous over the veins; palea about 2/3—3/4 as long as the lemma; lemma apex rounded to acute : 3(2). Second glumes less than 1/2 the length of the spikelet ................ M. spectabilis 3. Second glumes 2/3—3/4 the length of the spikelet M. bulbosa Voucher specimen. COLORADO. einstein Co.: Gunnison Nationa! Forest, 0.6 mi S on USFR 898 from McClure Summit, hill spruce forest with gray silty clay loam soil, elev. 2660 m (8727 ft), 24 Jul 1992, S. se GC. Jones 9509 (jkw, sdj, TEX, UTC). Associates: Veratrum californicum Durand, Zig Pursh, Carex stenoptila F. Herm., C. microptera Mack.., Senecio bigelovii A. Gray, Rudbeckia eal: is Nutt., Mertensia sp., Piston: tilla spp., Rosa sp., Delphinium sp., Geranium sp., Populus tremuloides Michy., and Picea s We would like to thank Mary E. Barkworth (UTC) for her review of this manuscript and Gretchen D. Jones (USDA) for her assistance with the field work.—Joseph K. Wipff and Stanley D. Jones, S.M. Tracy Herbarium, Department of Rangeland Ecology and Management, Texas AGM University, College Station, TX 77843-2126, U.S.A SIDA 16(1): 210. 1994 NOTES 211 REFERENCES Bovis, W.S. 1945. A cytotaxonomic study of the North American species of Melica. rofio 8:1—32. Cuase, A. 1951. A revision of A. S. Hitchcock’s manual of the grasses of the United States, 2nd ed., U.S.D.A. Misc. Publ. No. 200. U.S. Government Printing Office, Washing- ton D.C. CLayTon, W.D. and S.A. RENvoize. 1986. Genera graminum, Grasses of the World. Kew Bulletin Additional Series XIII. Her Majesty’s Stationary Office, London. HALLsTEN, G.P., Q.D. SKINNER and A.A. BEETLE. 1987. Grasses of Ny OuunS: 3rd Edicion. University of Wyoming Research Journal 202. Agricultural E Laramie Wyoming. Harrincton, H.D. 1964. Manual of the plants of Colorado. 2nd ed. The Swallow Press, Inc. Chicago, Hlinois WEBER, W.A. 1987. Colorado flora: Western Slope. Colorado Assoc. Univ. Press, Niwot, Colorado. Weper, W.A. 1990. Colorado flora: Eastern Slope. Univ. Press of Colorado, Niwot, Colorado Weer, W.A. and R.C. WITTMANN. 1992. Pan of the Colorado flora: a biodiversity baseline. Univ. Press of Colorado, Niwot, Colora EUSTACHYS GLAUCA AND E. CARIBAEA (POACEAE: CHLO- RIDEAE): THE FIRST REPORTS FOR MISSISSIPPI—Eustachys Desv. is a small tropical and subtropical genus of ten species, that is principally found in the New World (Clayton and Renvoize 1986). Three species are now known to occur in Mississippi: Eustachys caribaea (Sprengel) Herter, E. glauca Chapm. and E. petraea (Sw.) Desv. Eustachys caribaea, native to South America, has been reported from Louisiana (McKenzie et al. 1987) and Texas (Wipffand Hatch 1992). Eustachys glauca, native to North America, has been reported from Florida, Georgia and North Carolina and is found in brackish marshes, wet prairies and swamps (Chase 1951). Eustachys petraea is found in the SE U.S.A., eastern Mexico, Belize, Honduras, Costa Rica, and Panama (Pohl 1980). In the United States, E. petraea occurs in the Coastal Plain from North Carolina west to Texas on coastal sands, sandy fields and open pine woods (Chase 1951). KEY TO EUSTACHYS OF MISSISSIPPI ry . Lower (fertile) floret glabrous, rarely sparsely pubescent on the midvein and margins; second glume apex is rounded or minutely notched; lower (fertile) lemma dark chocolate brown to black E. glauca . Lower (fertile) floret conspicuously pubescent on the midvein and margins; second glume apex is conspicuously bifid, the lobes are rounded to acute; lower (fertile) lemma dark chocolate brown to black or pale to golden brown............ 2 — SIDA 16(1): 211. 1994 212 SIDA16(1) 1994 2(1). Lower (fertile) lemma dark chocolate brown to black; lower lemma margins glabrous on the lower 1/2—2/3; the upper portion of the lemma margins with short usually appressed trichomes 0.1—0.4 mm long; upper most (sterile) lemma 0.8—1.0 (1.2) mm long E. petraea . Lower (fertile) lemma pale to golden brown; lower lemma margins pubescent the entire length; trichomes on the lower lemma margins ho arg appressed to spreading and more than 0.5 mm long; uppermost (sterile) lemma 1.3—1.5 mm long E, caribaea Eustachys caribaea (Sprengel) Herter Voucher specimens. MISSISSIPPI: Harrison Co.: Gulfport, W jct. of Hwy 49 and 19th Street, T8S, R11W, S4; along edge of road and in a vacant lot, 11 Aug 1993, C.T. Bryson 12661 (SWSL); Gulfport, W of Hwy 49, between 34th and Madison Street; vacant lot, open area, on sandy soil, 12 Aug 1993, C.T. Bryson 12662 (SWSL). Eustachys glauca Chapm. Voucher specimens. MISSISSIPPI. Hancock Co.: 1 mi S jct. Hwys US 90 and MS 607; 90; open roadside on humic sandy soil, 17 Oct 1993, C.T: Bryson 13181 & R. Carter (IBE, SWSL, TAES). Harrison Co.: Gulfport, Mill Rd.; T7S, R11 W, Sec. 25; old field adjacent to Bayou Bertrand, 11 Oct 1990; T:C. Lockley s.n. (SWSL). —J.K. Wipff, Stanley D. Jones, 8.M. Tracy Herbarium, Department of Rangeland Ecology and Management, Texas AGM University, College Station, TX 77843- 2126, U.S.A. and Charles T. Bryson, USDA-ARS, Southern Weed Science Laboratory, Stoneville, MS 38776, U.S.A. REFERENCES Cuase, A. 1951. Hitchcock’s manual of the grasses of the United States, 2nd ed., U.S.D.A. Misc. Publ. No. 200. U. S. Government pou: Office, Washington Crayton, W.D. and S.A. RENvoIzE. 1986. Genera ea a of the World. Kew Bull. Addit. ae re Her Majesty’s ane Office, L McKenzie, P.M., L.E. Urparscu and C. AULBACH-SMITH. ae -eareaeintetliaes (Poaceae), a species new to ee United States. an 12:227-232 POHL, R.W. 1980 oe In: Flora Costaricensis, family No. 15, W. Burger, ed. Fieldiana, Bot. 4:1—-608 Wiper, J.K. and S. L. ae 1992. Eustachys caribaea (Poaceae: Chlorideae) in Texas. Sida 15:160- 162 ; SIDA 16(1): 212. 1994 NOTES 213 SCHISANDRA GLABRA (SCHISANDRACEAE) NEW TO KEN- TUCK Y—Field searches for rare native plants on the Daniel Boone Na- tional Forest in east central Kentucky are conducted yearly. During such a search in fall 1991, an unusual vine was found in McCreary County. The vine, growing with Parthenocissus quinquefolia and at first mistaken for an aberrant form of the latter, was eventually identified as Schisandra glabra (Brickell) Rehd., starvine. The plant was growing ona shale/sandstone talus slope at a break in a sandstone cliff. The species is typically a high climbing vine up to 3 cm in stem diameter, but the plants in this population were all creeping along the ground, some rooting at nodes. The site was partially shaded but did receive direct east sun. accuse in the area consisted of Acer rubrum, Liriodendron tulipifera, Tsug , and scattered Quercus species. Other species immediately adj acent to the site included C letbra acuminata, Kalmia latifolia, Mitchella repens, and Solidago caesia. Schisandra glabra was not reported for McCreary County by Rogers (1941) or for Kentucky by Braun (1943). More recently, the taxon was not reported for Kentucky by Johnson and Nicely (1990) or Browne and Athey (1992). Medley (1993) referenced my collection in his dissertation. This new location is about 250 km disjunct from the nearest population of S. glabra, in Stephens County, Georgia (Jones and Coile 1988; Ettman 1980) and represents the first record for the northern Cumberland Plateau. Other sites are from the southern Cumberland Plateau (Alabama), the Piedmont Plateau, the Coastal Plain, and the Mississippi Embayment. The leaves of starvine are alternate, typically pale green, somewhat fleshy, and coarsely and remotely serrate. Shorter branch stems have leaves crowded at the branch tip, almost appearing as palmately arranged leaflets. Leaves on trailing stems are widely spaced. Flowers are difficult to see as they are usually high above the ground. Fruits are scarlet and are arranged in loose racemes on pendulous peduncles. Illustrations may be found in Duncan (1967) and Stone (1968). I thank Michael A. Vincent (MU) for providing helpful comments and suggestions. Voucher specimen: ee aeons McCreary Co.: single population on shale/sandstone talus slope at break 1 ff on W side of Wolfpen Branch, ca. 1 mi S of White Oak Junction, ca. | miE pe Knob Church, Barthell Quadrangle, elev. ca. 1220 ft, 8 Oct 1991, D.D. &. L.A. Taylor 16351 (BEREA, KNK, MU, US). —David D. Taylor, USDA Forest Service, Dantel Boone National Forest, 1835 Bighill Road, Berea, KY 40403, U.S.A. SIDA 16(1): 213. 1994 214 SIDAL6(1) 1994 REFERENCES E.L. 1943. An annotated Ns of spermatophytes of Kentucky. Privately pert ee by ie author, Cine Browne, E.T. and R. AtHey. 1992. Sacculat plants of Kentucky: an annotated checklist. The Universiey Press of Kentucky, Lexington Duncan, W.H. 1967. Woody vines of the southeasiod United States. Sida 3:1—76. Errman, D. 1980. A study of Schisandra glabra Brickell) Rehder, a rare species endemic to the southeastern United States. Unpublished master’s thesis, Emory University, Atlanta JOHNSON, G.P. and . A. au. 1990. The Magnoliales of Kentucky. Trans. Kentucky Acad. Sci. 51:14— Jones, S.B. and N. . Cong: 1988. The distribution of the vascular flora of Georgia. Department of Botany, University of Georgia, Athens jan) Meptey, M.E. 1993. An annotated catalog of known or reported vascular flora of Kentucky. Unpublished dissertation, University of Louisville, Louisville. Rocers, J.H. 1941. The flora oieara County, Kentucky. Unpublished master’s thesis, University of Kentucky, Lexingto Stone, D.E. 1968. Cae a ee cuondioweal notes on the southeastern endemic Suhiatian glabra (Schisandraceae). J. Elisha Mitchell Sci. Soc, 84:35 1—3 THLASPI MONTANUM (BRASSICACEAE) AND OENOTHERA MIS- SOURIENSIS (ONAGRACEAE) NEW TO COAHUILA—Recent collec- tions in northern Cahuila, México, have included two species not previously reported for the state. Nesom (1992) found a similar situation for three species of Aster collected in the Serranfas del Burro in Coahuila. Thlaspi montanum L. var. montanum grows on moist or dry, open, rocky scree or talus slopes, alluvial fans or flats, Limestone cliffs, and forest clearings. It has a wide distribution in the western United States, where it occurs from Washington, Oregon and southward along the Rocky Mountain Cordillera into New Mexico, Arizona, western-most Texas and in one locality in the La Bufa Mountains of northern Chihuahua, México. This new record extends its distribution to the northwestern Coahuila. In addition to its wide geographical distribution, this variety of Th/aspi montanum has wide morphological variation (Holmgren 1971). The plants growing in Coahuila are 20—30 cm tall with racemes 8—10 cm long, peduncles 1 cm long and fruit 1 cm long and S—6 mm wide. The habitat is semi-mesic submontane scrub. 1: MEXICO: Coahuila: Mpio. Muzquiz: Sierra del Carmen, Ranchos is tetas y San Isidro, Aprox. 178 km nw Muzquiz (Carr. 53) 28°47'N, 102°30'W, matorral rosetofilo de Agave lechuguilla, Leucophyllum frutescens, Prosopis glandulosa y Larrea tridentata, 1300 m, 27 Mat 1992, M. A.C Carranza 1385, J]. Nortega, y L. Garcta (ANSM, MEXU). ee Muzquiz: Sierra la Encantada, cuesta Malena, aporx. 170 km nw Muzquiz (Carr. 53) 28°54'N, 102°30'W, matorral de Cercocarpus montanus, Garrya ovata, Juniperus flaccida, > SIDA 16(1): 214. 1994 NOTES 215 , Ungnadia speciosa, Quercus gravesil : . invaginata, 1600 m, 29 Mar 1992, M. A. Ce 1523, J. Noriega y L. Garcia (ANSM Oenothera missouriensis Sims. cimens examined: MEXICO: Coahuila: Mpio. Muzquiz: Carr. Muzquiz-Boquillas ©) a Rancho la Babia, area cercana al arroyo la Babia, 28°33'N, 102°05'W, matorral de ae Futescens, Acacia rigidula con Rhus virens, R. me a ne 950 m, 17 Sep 2, J. A. Villarreal 6971, M. A. Carranza y R. Vasquez A. (AN U). res missouriensis is widely distributed in limestone knobs, prairies and dry hills in the High Plains through Oklahoma, west Arkansas, Missouri to Kansas and Nebraska and into northcentral Texas (the Edwards Plateau). The coahuilan population belongs to the var. zwcana gray and 1s located in an intermountain valley in central Coahuila disjunct from the closest Texan location. Numerous other species show a similar pattern of distribution as noted by Nesom (1992), considering that the two species here recorded havea primarily North American distribution extending into northern México. Ithank G. Nesom for comments on the manuscript.—JoséA. Villarreal Q., Universidad Autinoma Agraria Antonio Narro, Departamento de Boténica, Buenavista, Saltillo, Coahuila 25315, México. REFERENCES Corre.L, D.S. and M.C. JoHNsToN. 1970. Manual of the vascular plants of Texas. Texas Research eee Renner, Texas. HOLMGREN, Be 71. Abiosystematic study ee North American Th/aspi montanum and allie some m. ce York Bot. Gard. 21:1-1( NEsom, G. L. 1993. Three species of Aster pee Astereae) disjunct in northern Coahuila, México. Phytologia 74:296—304. A STATE RECORD FOR CYPERUS GRAYOIDES (CYPERACEAE) IN ARKANSAS—C yperis erayoides Mohlenbrock is an obscure member of Cyperus section Laxiglumi. It has previously been reported from Illinois, Missouri, Texas, and Louisiana. In Illinois its habitat is dry sand prairie located on the terraces of outwash plains of the Mississippi and the Illinois rivers (White & Madany 1978). In Missouri it is locally abun-dant on the Scotco sandridges of the southeastern lowlands (Carter & Bryson 1991). The Texas and Louisiana occurrences are described as located on pine barrens, xeric riparian sandhills, sandhill woodland, and deep sand savanna (Bridges & Orzell 1989). An Arkansas station was discovered as follows: SIDA 16(1): 215. 1994 216 SIDA16(1) 1994 Voucher specimens: U.S.A. ARKANSAS. Ouachita Co.: ca. 2.5 km E of Chidester, NW 1/4 of NE1/4 Section 6, T12N, R19W, on a ceils i om 1993, oe 1993-95 (ISU, UARK, VSC). Approximately 100 plants were observed in a sunny opening of a pine sandhill community. Associate species include Polygonella americana, Stylisma pickeringit, Quercus incana, Q. stellata var. margaretta, Polansia erosa, Haplopa- ppus divaricatus,and Monarda punctata. The site is undergoing active erosion. The presence of active disturbance to the site corresponds to descriptions given for sites in other states (Bridges & Orzell 1989; Carter & Bryson 1991; Herkert 1991). The pine sandhill community where the species was found corresponds more closely to the descriptions of C. grayoides locations in Texas and Louisiana (Bridges & Orzell 1989) than to those in Missouri or Ilinois (Carter & Bryson 1991). Discovery of C. grayoides in Arkansas indicates that this species should be looked for in other sandhill communities of southwestern Arkansas. Addi- tionally, the presence in northeastern Arkansas of soils derived from sand de- posits related to the sandy soils of southeastern Missouri (Saucier 1978) in- dicates that the species should also be searched for in that portion of the state. Iam grateful to the following individuals for their assistance: Dr. Richard Carter, Valdosta State College; Dr. Paul McKenzie, U.S. Fish and Wildlife Service; and Sherry Holmes, Missouri Department of Conservation.—John M. Logan, Arkansas Natural Heritage Commission, 1500 Tower Building, 323 Center St., Little Rock, AR 72201, U.S.A. REFERENCES Bripces, E.L. and §.L. ORzELL. 1989. Additions and noteworthy vascular plant collections from Texas and Louisiana, with historical ecological and geographical notes. Phytologia 66:12-69. Carter, R. and C.T. Bryson. 1991. A report of Cyperus grayoides and Cyperus retroflexus yaad new to Missouri and notes on other selected Missouri Cyperus. Sida 14: 481. re J.R. (Ed.) 1991. Endangered and threatened species of Illinois: status and distribution Volume 1: plants. Illinois Endangered Species Protection Board. SAucIER, R.T. 1978. Sand dunes and related eolian features of the lower Mississippi River Alluvial Valley. Geosci. and Man 19:23—40 Wuite, J. and M. Mapany. 1978. Classification of natural communities in Illinois. In: White, J. ed. Ilinois Natival Areas Inventory Technical Report. Vol. I: Survey Methods and Results, p. 311-405. Illinois Department of Conservation. SIDA 16(1): 216. 1994 BOOK REVIEWS Horn, B., R. Kay and D. Ase. A Guide to Kansas Mushrooms. (ISBN 0-7006-0570-3, hbk; ISBN 0-7006-0571-1, pbk.). University Press of Kansas, 2501 West 15th Street, Lawrence, KS 66049-3904. (913)864- 4154. $19.95 (pbk); $29.95 (hbk) 298 pp., 6 figures, 160 color photographs. nsas is famous for many things; wheat, cattle, oil, gas, aircraft manufacture, politi- cians, basketball, mile runners and now you can add mushrooms to the list. Most people not native to Kansas who plan a trip using Interstate 70 conjure up im of a monotonous, flat landscape dominated by endless fields of wheat. Nothing could be “Farther from the truth for eastern and central Kansas. The westernmost extension of the broadleaf deciduous forest reaches into eastern Kansas, providing diverse habitats for mushrooms. The Flint Hills in the central part of the state also provide rolling terrain with glens and canyons that are heavily wooded with a mixture of deciduous trees and junipers. These diverse habitats provide for a much greater biodiversity of flowering plants, mushrooms, Myxomycetes, mosses, and liverworts than would normally be expected. Bruce Horn, Richard Kay, and Dean Abel have given us a book that documents over 750 species of mushrooms with 150 species described and illustrated with color photographs. In the words of the authors “This book is a guide to the fungi of Kansas” and later, “Because this guide | has grows out of our forays 1 in Kansas, it is a personal document, written by | heir enthusiasm no less than their experience. WIlO Wal lt p It ie that and much m ne anEEOTNCHOR contains topical sections on the naming of mushrooms, mushroom parts substrates, sex and mushrooms, mycology in Kansas, first fungi Riess Haliats: aK -alendar of fungal fruitings, collecting mushrooms, photogtaph- ing mushrooms, mushrooms in the kitchen, Kans oe the edibility issue, and identifying mushrooms. A bibliography follows of section. Some of the general non Kansas information is available in other mushroom books bi it some is not. The section on Kansas habitats destroys the myth that Kansas is a flat desert. First fungi informs the reader about the easy to recognize groups of fungi. A Kansas calendar of fruitings oS the frequency of occurrence and the months ae for 150 fungal species. There is a very informative section on techniques, equipment and tips on collecting mushrooms. There i is a spoof on page 36 named Gigantomyces kansensis, depicted in color and located in Mushroom State Park near Kanopolis Lake. A Kansas historical marker along Interstate 70 reveals the true origin of these “mushroom rocks. “Formed through millions of years of erosion, and known geographically as the say Hills, these Dakota sandstone mounds cover a wide area. Notice the natural rock outcroppings and toadstools” in this park. Here also are stone fence posts still used in this portion of Kansas There is an interesting chapter on the ag of delienl in Kansas that highlights the activities of W.H. Kellerman at Kansas State University, Manhattan, who founded the Journal ee the forerunner ee ie that today represents ae official publication of the Mycological Society of America. Elam Bartholomew who was a farmer near Stockton in Rooks County and a self taught Ee inser “became the greatest collector of American fungi in his generation.” The numbers of ae he collected are staggering: “...292, specimens of fungi, which he divided and distributed in 427,700 labeled packets”, and the remainder represented the largest private herbarium in the United States. In the 1950s SIDA 16(1); 217. 1994 218 SIDA16(1) 1994 Clark T. Rogerson and Robert L. Shaffer, currently distinguished mycologists, collected thousands of fungal specimens for the Kansas State University Herbarium and began an annotated checklist of Kansas nnet Bue here 1s cach more E Backed: into these six pages. An Anthology of Kansas M } Accomvycorina with keys to orders, to families, and to species. The dichotomous keys are based on macroscopic features and are easy to use. The authors are to be commended for giving numbers along with species names in the keys that guide the user to the species descriptions and photographs. Species are presented in systematic order and with the non technical species description and color habit photograph on the same page. The color habit photographs are Me and of adequate size to demonstrate the appropriate field characters. Als | ery helpful notes that will aid in locating and identifying each species. There are 66 gilled Redes. 10 Boletaceae, 27 Aphyllophorales, 5 Tremellales, 13 Gasteromycetes, 2 Nidulariales, 3 Phallales, and 21 Ascomycetes. The nomenclature is current with scientific names given without author citations, some synonyms and common names are provided, and there are frequent references to related species. Each species is designated as being either edible, ec libilicy unknown, not edible, caution, or poisonous. There are 3 appendixes and a four-page glossary. The first appendix is a summary of fungal classification of the fleshy Basidiomycotina and Ascomycotina anda listing of genera known from Kansas. The second appendix is entitled A Lifelist for che Kansas Mycophile and lists 548 Kansas species alphabetically by genus. The last appendix is an introduction to mycological Latin pronunciation. The Annotated Bibliography contains 26 references, mostly current and published lies 1975, that will serve as field guides to identification. The handy size of the book will facilitate its use in the field. The backside and outer edge of the book is divided into a 20 centimeter ruler that is an expedient way to measure specimens. i This book is well organized and easy to use for the beginner but has sufficient detail to satisfy the advanced student. There is a wealth of information noc found in many regional mushroom books that warrants buying this book. It is the best of the regional books on mushrooms. It is mo ou estly priced and worth every penny when one considers the number of color photographs. This book will help you enjoy and get the most out of hunting mushrooms in your neck of the woods. It should be on the bookshelf of every mushroom enthusiast wherever they live.-—Harold W. Keller. SIDA 16(1): 218. 1994 BOOK REVIEWS 219 Dick-Peppiz, WILLIAM A. 1993. New Mexico Vegetation, Past, Present, Future. (ISBN 0-8263-1361-2, hbk) University of New Mexico Press, Albuquerque. $ 244 pp. + map and color photographs. New Mexico has a richly diverse vegetation that is ina continual state of flux due to both nces and long-term climatic changes. This theme pervades New Mexico Vegetation in which Dr. Dick-Peddie summarizes and synthesizes data from his long career at the University of New Mexico. Not only does this volume provide a detailed account of the present vegetation of New Mexico, but also gives an overview of the physical factors, vegetational history, and future concerns about rare species and fragmented natural areas. The book is divided into twelve chapters which fall into three natural sets. Chapters One to Four, plus the Introduction, provide the reader with a general background on the vegetation. In particular, immediately preceding Chapter One there are a county map and 29 handsome, colored plates that depict the vegetation types and phenomena discussed. Chapter One, covering the physical environment, includes maps of the physiographic regions, land/wacer areas in geologic time, average frost-free days, average annual precipi- tation, and soils. Chapter Two provides a reconstruction of past vegetation since the Tertiary. The ein and documentation of desertification over the last 100 years is important. Chapter Three presents general concepts of vegetation patterns in relation to the physical environment and to interspecies dynamics. Chapter Four discusses methods wee vegetation, conventions used in this book, and differences with other, earlier The detailed descriptions of the vegetation types and their constituent ance types (or vegetation units) in Chapters Five through Ten, make up the second section, the bulk of the book. Each chapter contains tables summarizing the hierarchy of these vegetation and habitat types and the major plant species in each vegetation type. Dick-Peddie recognizes the following vegetation types in New Mexico: Alpine Tundra, Subalpine Coniferous Forest, Upper Montane Coniferous (mixed conifer) Forest, Lower Montane Coniferous Forest, Aspen Distiepance SoM cS sOleTEFOUS Woodiane, Mixed Woodland, Savanna extensive woodland-g Montane Grassland, Plains-Mesa Grassland, Desert Grassland, Montane — Plaing-Mesa Scrub, Great Basin Desert Scrub, Chihuahuan Desert Scrub, Successional Montane Scrub, and several riparian vegetation types. Chapter Ten discusses special types including aquatic, lava-flow, and gypsophilic vegetation. W.H Moir contributed the chapter on alpine t tundra and montane forests. The final chapters shite the focus from the description of present vegetation types to concerns of and fact ting the future of New Mexico NeECu Richard Spellenberg contributed a lengthy discussion (Chapter 11) on species of special concern, i.e., endemics of limited distribution, and species in decline. Besides reviewing the problems, needs, and possible solutions, he gives detailed accounts of each of the types of threats and the plant species involved. In Chapter 1 2, Dick-Peddie advances a view of the future for the vegetation as a whole. In particular, his advocates a series of natural areas and a means to secure them. n I first saw this book, I knew that it would be invaluable as a lecture and student reference for a field botany course in New Mexico that I am developing. Even so, I was somewhat disappointed when the book was evaluated against the rigorous demands that I placed upon it. I was happily composing charts of vegetation on a grid of mean annual rainfall vs. mean annual temperature, which can be gleaned from the descriptions in Chapter Five, when I discovered that there are no parallel presentations of that information in the remaining descriptions. Furthermore, an introductory map showing physical features such SIDA 16(1): 219. 1994 220 SIDA16(1) 1994 as Mountain ranges, rivers, etc. would help one to quickly pinpoint examples given in the Xt. Even with these short ings, the volume is sure to be of use to anyone interested in the biology and/or geography of the Southwest.—Roger W. Sanders, Botanical Research Institute of Texas. SIDA 16(1): 220. 1994 BOOK REVIEWS 221 Neusert, H., W. Nowotny and K. BAUMANN. 1993. Die Myxomyceten Deutschlands und des angtenzenden Alpenraumes unter besonderer Beruchschtigung Osterreichs. Band 1, Ceratiomyxales, Echinosteliales, Liceales, Trichiales. (SBN 3-929822-00-8, hbk) Karlheinz Baumann Verlag, Gomaringen, Germany. DM190 (ca. $110.00). 344 pp., 158 line drawings, 96 SEM, 192 colored photo- graphs. English translation of the title, The Myxomyceres of Germany and its bordering Alpine Areas, with special attention to Austria. Volume 1 Ceratiomyxales, Echinosteliales, Liceales, Trichiales. This is volume one of a three volume series. Volume two on the Physarales and volume three on the Stemonitales will follow in intervals of about two years. The volumes ma purchased separately. The acquisition of a volume does not obligate one to purchase later volumes. Volume I is 344 pages, bound in cloth with a color jacket, 190 German marks and Poi $1 10 at current exchange r Introductory pages have ek of colored plates taken from “Bulliard’s Histoire des Champignons de la France” that illustrate Stemonitis axifera, Comatricha typhoides, and Arcyria cinerea and Ditmar’s “Die Pilze Deutschlands” that illustrate Enteridium Lycoperdon, Trichia iia Trichia decipiens and Arcyria incarnata. The first chapters treat in a general ay the cl ation of the Myxomycetes, give their occurrence and distribution, describe and illustrate their life cycle, and discuss the general morphology of the fruiting bodies. The color photographs of the fruiting body habits are the highest quality I have ever seen. The ook is worth the cost just for the excellent color photographs and scanning electron micrographs. Some of the color a are of fruiting bodies immature and freshly matured state. This is true of Ceratiomyxa, Dictydiaethalium plumbeum, Hemitrichia clavata, Lycogala oe Lrichia decipiens var. olivacea, and Tubifera ferruginosa. The illustrations of color photographs and line drawings are conveniently included in the text on the same page as the species description. The SEM’s are included as a group at the back of the book. The photographs and line drawing illustrations have a voucher specimen number cited as the source of the illustration. The authors are to be commended for following this practice. A special feature of this book is the effective use of boldface headi I ide tl d hrough the g , speci 1 species descripti The high quality paper, -double column fe a a] a bs C:]] betes d wero) book very user friendly. All three volumes are in German and represent the most comprehensive treatment of the Myxomycetes in German since the work of Schinz in 1920. According to the authors, their work will help fill the gap since that time. In a special section dichotomous keys lead the reader from the four orders to families, genera, and BBecles: wien ae species and varieties described in detail and 157 indexed. The | There are also synoptic keys to some series for example, the very difficult and ae genus Cribraria, with line drawings of naps on 1 side by side pages next to the key. This is d YCIry uscl friend] ke | are three species described as new: Horna C. macrospora, and C, tellilova: The authors prepared this work for professional botanists who may wish to identify Myxomycetes and also serve as an introduction for the lay botanist. The unique beauty of SIDA 16(1): 221. 1994 222 SIDA16(1) 1994 the Myxomycetes and largely unknown distribution in Germany are given as the additional reasons for publishing these volumes. There are sections on collecting and herbaria in Germany, on microscopic slide preparation, a glossary of 45 defined terms and the etymology is given for generic and species names. Commemorative peoples names listed for genera and — also have their birth and death dates, country of origin, and professional disciplin The aoe classification system considers these organisms aligned with the fungi retaining the proper cen ha eninge ees Py the Inteinations ed of Botanical Nomenclature. The subordinal losely y Nannenga- pee since many of ea taxa, arene genera and feat lies, are recognized. Trichiales toa new family, the Minkatellaceae, in the order Liceales. The order Liceales was a ee unnatural assemblage of unrelated taxa before the transfer of Minakatella and it is even more so now. There is no new convincing evidence to recognize two new families, the Minakatellaceae and Dictydiaethaliaceae, in the Liceales. In the case of Minakatella, the over emphasis of the property of birefringence of spores and capillitium, supposedly lacking in ae results in the tyranny of a single unproven character; a character that is variable w species but is used to move taxa from different orders and establish new families. Much of the name juggling that has gone on in the Myxomycetes should await state of the art DNA sequencing techniques to analyze genetic relationships within the Myxomycetes and to — other taxa such as the protostelids and dictyostelids. The literature cited section is cae valuable since it contains much of the published licerature on the Myxomycetes sin eG. W. end and nee PRERODOUNOS published their world monograph, “The M Lee "in 1969. T p with author citations from chesughout the world. This book oe pet on bane bookshelf every student of the Myxomycetes. The authors the Myxomycetes on both sides of the Rivne Faas W. Keller. SIDA 16(1): 222. 1994 Carex lutea (Cyperaceae), a rare new coastal plain endemic from North Carolina RJ. LeBlond, A.S. Weakley, A.A. Reznicek and W.J. Crins 153 Sporobolus potgsiensis (Poaceae: Eragrosteae): A new rhizomatous species from San Luis Potosi, México, and a new combination in $. airoides Joseph K. Wipffand Stanley D. Jones 163 Notes on Carex, Cyperus, and Kyllinga (Cyperaceae) in Mississippi with records of eight species previously unreported to the state Charles T. Bryson and Richard Carter 171 New taxa and new combinations in Chinese plants Shiyou Li and Kent T. Adair 183 A new species of Anemone (Ranunculaceae) from central Texas Carl S. Keener and Bryan E. Dutton 191 Documented chromosome numbers 1994:1. Karyotype of Jaimehintonia gypsophila (Amaryllidaceae) Zaiming Zhao 03 NOTES Validation of Karinia (Cyperaceae) 207 Penstemon thurberi (Scrophulariaceae) new to Texas 207 Euphorbia graminea (Euphorbiaceae) new to Florida 208 Melica subulata (Poaceae: Meliceae): The first report for Colorado 210 Eustachys glauca and E. caribaea (Poaceae: Chlorideae): The first reports for Mississippi 211 Schisandra glabra (Schisandraceae) new to Kentucky 213 Thlaspi montanum (Brassicaceae) and Oenothera missouriensis (Onagraceae new to Coahuila 214 A state record for Cyperus grayoides (Cyperaceae) in Arkansas 215 Announcements 162, 170 Book reviews and notices 16, 22, 32, 56, 62, 108, 152, 206, 217 } ( ISSN 0036-1488 CONTENTS Miconia skeaniana (Melastomataceae: Miconieae), a new species from eastern Cuba Walter S. Judd 225 Cytological studies on North American species of Saccharum (Poaceae: Andropogoneae) D. M. Burner and Robert D. Webster 233 Systematic study of Texas populations of Phacelia patuliflora rophyllaceae) Jennifer Ann Moyer and Billie L. Turner 245 Ashicaulis, a new genus for some species of Millerocaulis (Osmundaceae) William D. Tidwell 255 oe : New and noteworthy Malesian Myrsinaceae, VII. _ Benjamin C. Stone — oo on A new species of Carex (Cyperac and Texas; typification of seet Bracteosae and Phaestoglock Stanley D. Jones 341 CONTRIBUTIONS TO BOTANY ¢ VOLUME 16 NUMBER 2 DECEMBER, 1994 (continued on back cover) CONTRIBUTIONS TO BOTANY FOUNDED BY LLOYD H. SHINNERS 1962 Wim. F. Mahler Publisher 1971-1992 Director Emeritus S.H. Sohmer Director Barney L. Lipscomb Editor Botanical Research Institute of Texas, Inc. 509 Pecan Street Fort Worth, Texas 70102-4060 817 332-4441 / 817 332-4112 FAX John W. Thieret Prof. Dr. Felix Llamas Associate Editor daca Spanish Editor Biological Sciences Dept. Dpto. de Botanica, Facultad de Biologia Northern Kentucky University Universidad de Leon ‘-24071 Leon, Spain — Highland Heights, Kentucky 41076 Guidelines for contributors are available upon request nd on the inside back cover of the last issue of each volume. teen per year: $20. Individual, $30. U.S. Institutions, $40. Outside U.s.; numbers issued twice a year Sf oO © SIDA, CONTRIBUTIONS TO BOTANY, ea 16, Number 2, pages 225-388, copyright 19 Botanical Research Institute of Texas, Inc, ISSN 0036-148 MICONIA SKEANIANA (MELASTOMATACEAE: MICONIEAE), A NEW SPECIES FROM EASTERN CUBA WALTER S. JUDD Department of Botany, 220 Bartram Hall University of Florida, Gainesville, FL 32611-8526; and Herbarium, 209 Rolfs Hall, Florida Museum of Natural History Gainesville, FL 32611-0530, U.S.A ABSTRACT ra Miconia skeaniana, which occurs in moist montane forests, cloud forests, and thicl 750-1974 m in the Sierra Maestra, is described and illustrated. It is compared with the ie related M. alternifolia, of the mountains of northern “Oriente,” a species with which it has been consistently confused. RESUMEN Sed ibe e il Miconia skeni eenlosb 1 htimedos bosques de nubes y matorrales entre los 750 y 1974 men an Maestra. Se compara con la cer emparentada M. alternifolia de las montafias del norte de “Oriente,” una especie con la que ha sido confundida frecuentemente. In the course of taxonomic study of herbarium material in connection witha monograph of the Antillean members of Miconia section Chaenopleura Benth. & Hook.f., the following undescribed species of Miconia was recognized. Material representing this new species previously has been identified as M. alternifolia (Griseb.) Alain (= PF lesiuee grisebachii Triana). The new species clearly is referable to Miconia sect. Ch (see Cogniaux 1891), a distinctive, diverse, and presumably monophyletic group (at least within the Antilles) possessing an actinomorphic androecium (i.e., stamens forming a radially symmetrical pat- tern around style) of glabrous, white stamens of which the obovate anthers open by longitudinal slit-like pores Judd & Beaman 1988; Judd & Skean 1991). The group is also characterized by an indumentum of usually ferrugineous, + ir- regularly stellate-branched hairs, globose fruits that turn from red to pale blue at maturity, and angular-obovoid seeds with a + smooth testa (see Judd & Skean 1991, fig. LOB). The section is especially well developed in the Greater Antilles, where many narrow endemics occur. The new species is described and compared with M. alternifolia, a species with which it has been consistently confused. Muconia alternifolia occurs in the mountains of northern “Oriente”! from the Sierra de Cristal eastward, while M. skeaniana is restricted to the Sierra Maestra. 'For convenience the four easternmost provinces of Cuba, i.e., Granma, Santiago de Cuba, y as “Oriente,” the pre-1959 name of the political unit comprising eastern Cuba. Sipa 16(2): 225 — 231. 1994 226 Stipa 16(2) 1994 Miconia aacuneuia aie sp. nov. (Fig. 1) Mi | ifoliae (Griseb.) Alain affinis, sed foli ionil gi valde abaxiale elevatis, i.e., venis tertiariis leviter vel manifeste elevatis, et aliquot vel omnibus venis Be caaeacies leviter devs ats (vs. v venis tertiariis leviter elevatis vel plus minusve planis, et planis), foli apice late obtuso, acuto, vel acuminato), pilis sreepulanicee stellatis ca 0.09— 0.2 mm latis in foliorum pagina abaxiali, interdum cum pilis minutis et globulosis (vs. pilis globulo-stellatis ca 0.04—0.09 mm lactis), et antherae parte fertili 1.21.9 mm longa, i.e., 66-77% antherae, (vs. 0.9— 1.4 mm longa, i.e., 39-58% antherae). Shrub to 4m tall. Indumentum of multicellular, ferrugineous, minute-globu- lar, globular-stellate, or irregularly stellate-branched, to elongate-branched hairs. Young twigs not ridged, + rectangular in cross-section, 1.7—4.5 mm wide, be- coming + terete with age, the indumentum of moderate to dense, irregularly stellate-branched to elongate-branched hairs, these + persistent; internodes 1— 4.5 cm long. Leaves opposite, with petiole 0.7—2.9(—3.2) cm long, the indumen- tum similar to that of the twigs; blade ovate to elliptic, 5—11.2(-16) X 1.7- 4.1(—4.4) cm, flat, coriaceous, the apex acuminate, the base acute to rounded or very slightly cordate, the margin plane to revolute, obscurely serrulate to undu- late-serrate, especially distally, proximal 0-40% of margin entire, but frequently appearing entire throughout (when revolute), the largest teeth to 0.2-0.4 mm (occasionally to 0.8 mm) long; venation acrodromous, slightly suprabasal to + basal, with prominent midvein (primary vein) and 4 secondary veins, with 2 conspicuous secondary veins placed 2—6.5 mm in from margin, 2 inconspicuous secondary veins placed closer to margin, and numerous percurrent tertiary veins oriented subperpendicular to midvein, the tertiary veins sometimes partially separated by composite inter-tertiary veins but usually connected by 1 to nu- merous quaternary veins, higher order veins + orthogonal-reticulate; adaxial surface very soon glabrescent, the midvein and major secondary veins impressed, minor secondary veins, tertiary veins, and some quaternary veins slightly im- pressed, and higher order veins flat, the surface appearing minutely wrinkled- papillose after drying, with sparsely to densely Stas druse crystals; abaxial surface with moderate to dense, persistent, irregularly stellate-branched or globu- lar-stellate hairs, 0.09—O.2 mm across, on lamina and smaller veins, sometimes also with minute globular hairs, intergrading with slightly larger stellate- branched to elongate-branched hairs on midvein, the midvein and major sec- ondary veins prominently raised, tertiary veins prominently to slightly raised, minor secondary veins, some to all quaternary veins, and sometimes even a few higher order veins slightly raised. Inflorescences terminal, many-flowered, broadly- rounded cymes of 3 or 4 branch-pairs, 3-8 cm long, 3-8 mm across; proximal segment of lowermost inflorescence branches 1—2.9 cm long, distal internodes shorter, ultimate branches 1—6.5 mm long, with + moderate, irregularly stel- late-branched to globular-stellate or elongate-branched hairs; peduncle 1.5—5 cm long, with similar indumentum; each branch associated with early caducous, Jubb, Miconia skeaniana 221 pitts, LP ie panne Lil. ne : W 0 ee Z Mey stale Fic. 1, Hlustration of Miconia skeaniana, along with selected features of M. alternifolia. Miconia skeaniana: A. flowering branch (Ekman 8855); B. leaf, abaxial surface (Ekman 6928); C. detail of abaxial leaf surface (Ekman 6928); D. stellate hair from abaxial leaf surface (Seifriz 1072); E. flower (E&man 8855); F. st 8855); G. young berry (Sefriz 1072). Miconia alternifi lia: H. leaf, abaxial surface (Ekman 6833); 1. leaf, abaxial surface (Wright 179); J. detail of abaxial leaf surface (Shafer 8198); K. stellate hair from abaxial leaf surface (Ekman 6833); L. e portion of the anther (E&man 6833). stamen, with elongate basal and steril 228 Sipa 16(2) 1994 slightly ovate to obovate bract, ca 1.74 x 1-2 mm, the apex acute to rounded, the lowermost pair sometimes leaf-like; flowers in dichasia, usually distinctly separated from each other, each subtended by 2 caducous, + ovate bracteoles, 1.5—2.2 x 0.8—1 mm, the apex acute to obtuse; pedicel O—-0.5 mm long. Hy- panthium cylindrical, free portion 1.6-1.9 mm long, the outer surface with sparse to moderate, minute, stellate-branched hairs, the inner surface glabrous and strongly ridged, i.e., with 10 prominent ridges alternating with 10 weaker ridges, the apices of the stronger ridges projecting, 0.05—0.1 mm. External calyx lobes (=teeth) 5, 0.3—0.8 X 1.4+2.5 mm, broadly triangular, with acute to acumi- nate apex, indumentum as on hypanthium; internal calyx lobes 5, 0.5-1.5 x 1.4— 2.5 mm, + triangular to ovate-triangular, green, glabrous or with sparse stellate hairs, apex rounded, margin entire, sometimes minutely fringed; calyx tube (.2-0.5 mm long. Petals 5, broadly ovate to obovate, 2.8-4.2 x 2.6—2.9 mm, glabrous, white to pink-tinged, imbricate and apically interlocking in bud, with apex emarginate, with an asymmetrically located notch; margin entire. Stamens 10, geniculate, glabrous; proximal segment (filament) 1.9—2.9 mm long; distal segment (anther and connective) 2.6—3.2 mm long, with minute dorsal projection (pointing + toward anther apex), the anther 1.8—2.5 mm long, with fertile portion (anther sacs) 1.2-1.9 mm long, opening by 2 longitudinal slits, the connective extending 0.6—0.9 mm beyond the base of the anther. Ovary (2- or) 3-loculate (N=1, 10), ca 1/2-inferior (immature) to ca 2/3-inferior (mature), short-ovoid to subglobose, 2.1—3 x 2.4-3.5 mm, glabrous and strongly ridged, with fluted apical projection to ca 0.5 mm encircling the base of style; style 3.8— 5 mm long, terete, glabrous; stigma truncate. Berries globose to subglobose, ca 4.5-7 x 5.5-7 mm, pale blue. Seeds angular-obovoid, ca 0.5—0.9 mm long; testa smooth. Type: CUBA. Prov. SANTIAGO DE Cusa [=Oriente, p.p.}: Sierra Maestra, Cordillera de la Gran Piedra, La Gran Piedra, cloud forest, ca 1200 malt., 10 Nov 1917, E.L. Ekman 8855 (HOLOTYPE: S!; IsoTYPE: NY'!). Etymology: It is a pleasure to name this distinctive species after Dr. James Dan Skean, Jr. (b. 1958), plant systematist at the Department of Biology, Albion College, Michigan, who has assisted the author during fieldwork conducted in connection with a taxonomic revision of the Antillean species of Mzconza sect. Chaenopleura. Additional Specimens Examined: CUBA. Prov. GranMa [=Oriente, p.p.}: Sierra Maestra, La Bayamesa, on the ridge between Rio Oro and Rio Yao, 1100-1400 m, Ekman 7215 (F); peak of Punta de Palma Mocha, south of Yara, 1400 m, Ekman 14317 (NY). Prov. SANTIAGO DE Cusa [=Oriente, p.p.}: Sierra Maestra, summit of Pico Turquino, 1960 m, Acuia 6760 (NY); Pico Turquino, oe slopes, ca 1750 m, Ekman 5275 (S); ibid., Ekman 5389 (S), between Finca Reunion and peak of Loma del Gato, 750 m, Ekman 6928 (NY), near summit of Pico Turquino, Leén 10744 (GH, NY); between the arroyos Peladero and Indio, 3000—4500 ft, Lépez-Figueiras 406 (US), Gran Piedra, 1250 m, Lopez-Figueiras 2661 (IJ, US); Pico Turquino, Jupp, Miconia skeaniana 229 SS 9) ae, = - ene one Pa saad 77 76 75 FIG. 2. Distribution of Miconig theaniana in astern Cuba. south slopes, Seé/riz 1072 (US); Gran Piedra, ca 1500 m, Shafer 9028 (NY); eastern Cuba, ithout definite locality, but probably collected at Loma del Gato, 1856-1857, Wright 179 (GH, GOET, MO). Miconia skeaniana is endemic to Cuba and known from several localities in the Sierra Maestra, both the Cordillera de Turquino and the Cordillera de la Gran Piedra, of southern “Oriente” [prov. Granma and Santiago de Cuba] (Fig. 2), where it occurs in moist montane forests, cloud forests and thickets from 750-1974 m elev. The vegetation of the higher elevations of the Sierra Maestra is summarized in Leén (1924, 1946), Seifriz (1943), and Borhidi (1991). Miconia skeaniana is probably most closely related to M. alternifolia, with which it consistently has been confused. It can be readily distinguished from this species by its more strongly abaxially raised-reticulate leaf venation, i.e., midvein and major secondary veins prominently aie tertiary veins slightly a prominently raised, minor secondary veins, some to all q “y veins, and sometimes even a few higher order veins slightly raised (vs. midvein and major secondary veins prominently raised, minor secondary veins slightly raised, tertiary veins slightly raised to + flat, and higher order veins flat). The leaves of M. skeaniana have consis- tently acuminate apices while those of M. a/ternifolia vary from broadly obtuse to acute or acuminate. Miconia skeaniana possesses a ferrugineous indumentum of irregularly stellate-branched to occasionally globular-stellate hairs, ca 0.09— 0.2 mm across, on minor veins and lamina of the abaxial leaf surface. These hairs are Someries intermixed with minute-globular hairs. In contrast, the hairs of are more darkly ferrugineous, consistently globular-stellate, and smaller a e., 0.04—0.09 mm across). The basal and sterile portion of the anthers 230 Stipa 16(2) 1994 of M. skeaniana is shorter than that in M. a/ternifolia, i.e., anther with fertile portion 1.2—1.9 mm long and occupying 66-77% of anther length in M, skeaniana (vs. 0.9-1.4 mm long and occupying 39-58% of anther length in M. alternifolia). Additionally, the tertiary veins of M. skeaniana are frequently more closely spaced than those of M. a/ternifolia, i.e., 5-12 (avg. 9) veins vs. 4-11 (avg. 7) per 2 cm in middle portion of leaf. The two species are completely allopatric and are, thus, geographically isolated. The type of Miconia alternifolia, collected by Charles Wright (no. 179), ts actually a mixed collection composed of material collected at Monte Verde (Prov. Guantdnamo) on March 2nd, 1859 (see original label on isotype at GH and label on holotype at GOET) and presumably at Loma del Gato in the Sierra Maestra (Prov. Santiago de Cuba) in 1856-1857. The exact locality of the 1856— 1857 collections is not known because no original label is present; however, plants matching these specimens occur only in the Sierra Maestra and it is known that Wright did collect in the Loma del Gato area of the Sierra Maestra in 1856-1857 (Underwood 1905; Howard 1988). These collections of Charles Wright were all assigned the same number by Asa Gray (Howard, 1988). The holotype of Méconia alternifolia (at GOET) represents the “Monte Verde ele- ment” (collected on March 2, 1859), which is considered to be conspecific with phenetically similar collections from other localities in northern “Oriente,” 1.e., prov. Holguin and Guantanamo; isotypes representing this collection are found at BM, GH, NY, and S (all with printed labels giving the incorrect date of 1860-1864). Collections of Wright 179 at GH, GOET, and M (collected in 1856-1857, probably at Loma del Gato) actually represent the taxon here con- sidered to be M. skeaniana. As discussed above, this species is restricted to the Sierra Maestra. The name M. a/ternifolia is, of course, linked to the holotype specimen of Wright 179 (which represents the northern “Oriente” taxon). Thus, the plants of the Sierra Maestra, if considered specifically distinct from M. a/- ternifolia, are left without a name, and herein are described as M. skeanzana. The specimens of Wright 179 collected in 1856-1857 have sometimes been considered as a distinct variety, i.e., M. grisebachit var. reticulata Cogn., but the type of this name actually is a Puerto Rican plant of the Sierra de Luquillo (and this name is asynonym of M. pychoneura Urb.). The recognition of Miconia skeaniana brings the number of Antillean species of Miconia sect. Chaenopleura to 43 (Judd, unpublished data). Other members of this section occurring in Cuba include: M. alternifolia, M. cubensis (Griseb.) C. Wright, M. rwfa (Griseb.) Triana (probably conspecific with M. plumeriaefolia Britton & P. Wilson), and M. subcorymbosa Britton (probably conspecific with M. calycina Cogn.). Jupp, Miconia skeaniana 231 ACKNOWLEDGMENTS I thank Wendy B. Zomlefer for her excellent illustrations and Kent D. Per- kins for his assistance in processing specimen loans. This research was supported by NSF Grant BSR-9016793 REFERENCES Boru, A. 1991. gy of Cuba. Akadémiai Kiad6, Budapest. COGNIAUX, A. 1981. vee aceae. cre A. C. de Candolle, editors. Monographiae phanerogamarum. Vol. 7. Masson, Paris. Howarp, R.A. 1988. Charles Wright in Cuba, 1856-1867. Chadwick-Healey, Alexandria, irginia. Jupp, W.S. and R.S. BeaMaN. 1988. Taxonomic studies in the Miconieae hfe astomaraccss) II. aaa of the Miconia subcompressa complex of Hispaniola, including th 3 ia 40:368-391. two new species. Britton WS. and J.D. SKEAN OR. 1991. Taxonomic studies in the Miconieae (Melastomataceae). eneric realignments among terminal-flowered taxa. Bull. Florida Mus. Nat. Hist., Biol. 7 36:25-84 Leon, HNno. 1924. Una excursion al Pico Turquino. Mem. Soc. Cub. Hist. Nat. “Felipe Poey” 6:127-140. Hno. 1946. Flora de Cuba. Vol. 1. Contr. Ocas. Mus. Hist. Nat. Colegio “De La Salle” A411. oF rrIZ, W. 1943. The plant life of Cuba. Ecol. ce 13 an miaes Unpberwoop, L.M. 1905. A summary of Charles W ration in Cuba. Bull. Torrey Bot. Club 32:291—300. 232 Sipa 16(2) 1994 BOOK NOTICE Karresz, JOHN. T. 1994. A Synonymized Checklist of the Vascular Flora of the United States,Canada and Greenland. Second Edition. ISBN 0-88192-204-8, hbk, 2 volumes); Timber Press Inc., 9999 S. W. Wilshire, Suite 124, Portland, OR 97225. (800) 327-5680. $159.00. 1, 622, vol 2, 816 pp. The long awaited second edition of John Kartesz’ A Synonymized Checklist of the Vascular Flora of the United States, Canada and Greenland is finally available, and it was well w This is officially a second edition, but it is more than just a new edition, it is rather a much expanded major rework which includes a new format and a second volume. This is a significant oprovernent over his 1980 work, which has been the standard io: fourteen years. monographic research the | process is in constant flux, and will be for many years to come. Also, much of the classification of our flora is a tangle of individual taxonomic opinion. This work does not actempre to se all cise problems, but it is a quantum leap forward, and is a solid new souadiatlo non oe e. With Kartesz’ new Checklist we now have an incredibly nd prel ive effort that sets a new standa rd. The new edition has two volumes. The first is the familiar checklist that is alphabetical by families, genus, species, etc., and the synonyms of each of these taxa. The second volume is a new thesaurus that is a nie reference source with all names and authors used following ex- ~ those in the body of the checklist with each name listed alphabetically 2 eae Accepted ames appear in Roman print and stand alone, aes eas s appear in italics and a followed by their accepted names. The th {i from the checklist “ offering a rapid location system for all names, save those names above the genus level. The thesaurus and the checklist are together intended to provide a comprehensive summary of accepted names and their synonyms of the North American vascular flora north of Mexico: the continental United States, Hawaii, Puerto Rico, and the U.S. Virgin Islands; Canada; the French islands of St. Pierre and Miquelon; and Greenland. This ts a major work which is used many times a day in my work of compiling floras, manuals, articles, books, and plant checklists of Texas and its many regions. This definitive Checklist on plant nomenclature and its ayAcoyIRy is an a tool of immense reference value, and isa must on the desk and bookshelf of ystematists, botanists and horticulturists who must kno | four flora.—Robert J. O ‘Kenn CYTOLOGICAL STUDIES ON NORTH AMERICAN SPECIES OF SACCHARUM (POACEAE: ANDROPOGONEAE) D.M. BURNER United States Department of Agriculture Agricultural Research Service Sugarcane Research Unit Houma, LA 70361, U.S.A. ROBERT D. WEBSTER United States Department of Agriculture Agricultural Research Service Systematic Botany & Mycology Laboratory Ide. 003, Rm. 235, BARC-West Beltsville, MD 20705, U.S.A. ABSTRACT Cytology of the Saccharum L. Gaciuding Erianthus MEER ) Spsciss native to Nezth America has not been well characterized. Our objecti ves were f 60 clones representative of the eight native species and varieties and attempt crosses with elite sugarcane (Saccharum spp. hybrids). Counts are reported for the first time for S. brevibarbe var. brevibarbe (2n = 60), S. coarctatum (2n = 60), and S. giganteum (2n = 30, 60, and 90). The latter represents the first report of a ae enn series within S. giganteum and the first count of 2n = 90 or the Erianthus group. Counts also were made for S. alopecuroideum (2n = 30), 8. baldwinii (2n = 30), and S. brevibarbe var. contortum ne = 60). Five putative crosses were made between sugarcane hybrids and native Saccharum, yielding 4 to 448 seeds per cross. RESUMEN Lac itologia de | F ies de Saccharum L. (inc luyendo Erianthus Michx. ) nativas de Norte América no ie sco bien caracterizada. oe objecivos oo determinar el numero Cruces le élite (hibtidosde Sacch Se ometen por primera vez recuentos de S. brevibarbe var. ise . ©), y>. atten n= = oo), y S. giganteum (2n = 30, 60, y 90). Este alti una serie poli S. gigantenm y el primer recuento de 2n = 90 para el grupo aie Se hicieron cache bién recuentos de S. alopecuroidenm Meee = I Slurp 30), 8. baldwinii (2n = 30), y S. brevibarbe var. sii (2n = GO). Se hicieron cinco cruces putativos entre hibridos de cafia de azuicar y aru stres, produciendo de 4 a 448 semillas por cruce. INTRODUCTION There has been disagreement among taxonomists concerning the treatment and placement of Saccharum L. and Erianthus Michx. A brief history of the vari- Sipa 16(2): 233 — 244. 1994 234 Sipa 16(2) 1994 ous concepts of these genera is presented in Webster & Shaw (1995). North American taxonomists have followed the concepts of Hitchcock (1951) and have recognized both genera. European agrostologists have placed Erianthus within Saccharum. Characteristics used to distinguish the genera are difficult to apply when the full range of variation is considered. Therefore, we are following the concept presented in Clayton & Renvoize (1986) that recognizes Erianthus as asynonym of Saccharum. Sugarcane breeders recognize separate genera and generally include five to six species within Saccharum: S. edule Hassk. (2n = 60 to 80), S. officinarum L. (2n = 80), 8. robustum Brandes & Jesw. ex Grassl (2n = 60, 80, to about 200), S. szmense Roxb. (2n = 111 to 120), and S. spontanenm L. (2n = 40 to 128). A sixth species, S. barberi Jesw. (2n = 81 to 124), is sometimes included in S. sznense. Chromo- some counts (Daniels et al. 1975) for closely related taxa called the “Saccharum complex” by Mukerjee (1957) include Old World Erianthus Michx. sect. Rip- idium Henrard (2n = 20, 30, 40, 60); Miscanthus Anderss. sect. Diandra Keng (2n = 40); Narenga Bor (2n = 30); and Sclerostachya (Hack.) A. Camus (2n = 30). These and a few other genera have at times been placed in Saccharum (Daniels and Roach 1987; Whalen 1991). The basic genomes within Saccharum (s. str.) appear to be x = 8, 10, and 12 (Sreenivasan et al. 1987), and that of Erianthus may be x = 5, typical of the Andropogoneae (Celarier 1956a). Harlan and de Wet (1975) noted that Saccharum had an “oversplit taxonomy,” implying that many taxonomic divisions may be artificial. As evidence of this, interspecific and intergeneric crosses within the Saccharum complex are usually successful (Gill and Grass! 1986; Grass] 1980; Daniels and Roach 1987). The taxonomic relationships among the taxa of the Saecharum complex have been neither carefully studied nor well-defined (Webster and Shaw 1995) and they conclude that Erianthus is best treated as a synonym of Saccharum. Five species and one variety of Saccharum L. were recognized by Webster and Shaw (1995) as being native to North America. They are S. a/opecuroidenm (L.) Nutt. {= Erian- thus alopecuroides (L.) El], S. baldwinii Spreng. (= E. strictus Baldw.), 8. brevibarbe (Michx.) Pers. var. brevibarbe, 8. brevibarbe (Michx.) Pers. var. contortum (Nutt.) R. Webster (= E. contortus Ell.), S. coarctatum Fern. (= E. coarctatus Fern.), and S. gigantenm (Walt.) Pers. {= E. gigantens (Walt.) C.E. Hubb.}. Old World species previously treated in Erianthus (sect. Ripidium) are used in sugarcane breeding (Heinz 1991), particularly for its disease resistance (Burner etal. 1993; Grisham et al. 1992) and freeze tolerance (Moore 1987). Chromo- some numbers of many clones have been reported (Babu and Srinivasan 1960; Mohan and Sreenivasan 1983). Hybrids between Saccharum spp. and North American species placed in Erianthus have not been reported, although Gill and Grassl (1986) reported hybrids between Sclerostachya fusca (Roxb.) A. Camus and E. brevibarbis Michx. (= S. brevibarbe), E. tracyi Nash. (= 8. alopecuroidenm), E. Webster, Cytological studies on Saccharum 235 contortus Baldw., and between Miscanthus sp. and E. brevibarbis Michx. (= S$ brevibarbe). Except for the few clones (five clones in four species) reported by Celarier (1956b) and Gill and Grass! (1986), there have been no cytological studies of the North American species traditionally placed in Erianthus. The objectives of this research were to determine chromosome numbers and characteristics ina complete collection of North American species of these taxa and attempt hy- bridization with elite sugarcane hybrids (interspecific and intergeneric Saccharum spp. hybrids). MATERIALS AND METHODS Clones were collected as rhizomes from natural populations in the mid-At- lantic and southeastern U.S.A. (Fall 1992) and southcentral U.S.A. (Fall 1993) and were grown at Houma, Louisiana (29° 35'N 90° 44'W) in 18.9 L cans filled with a soil mix of 3:2:2 (soil:sand:peat moss). Taxonomy was verified according to the concepts presented in Webster and Shaw (1995). Voucher specimens are deposited at TAES. Inflorescences were collected at early boot stage and fixed in Carnoy’s B: etha- nol, acetic acid, and chloroform (6:3:1 mixture by volume) (Smith 1947) anda few drops of saturated ferric chloride. Anthers were squashed in 5 g L"' propi- onocarmine. Chromosome number was determined from pairing configurations at diakinesis or metaphase I (MI) in 5 to 25 microsporocytes per plant. Chromo- some number for some plants was determined or verified in squashes of root tip cells using standard procedures. Pollen stainability, an estimate of pollen viabil- ity, was measured by staining mature anthers in 10 g L The cross-sectional area of individual bivalent chromosomes was measured at MI for each of nine clones (range 1 to 15 cells clone™!, mean 6 cells clone™') using a Cue-2! image analyzer (Galai 1990). Area (jim*) was determined from the number of pixels in the bivalent chromosome. About 100 completely filled pollen grains from mature anthers of 25 plants were collected in Fall 1993, stained in 10 g L"! I5-KI, and imaged at 400x. Cross-sectional area was calculated as described above. Average cross-sectional radius (im) of each pollen grain was the mean of eight Martin’s radii measured at 0, 45, 90, 135, 180, 225, 270, and 315° (Galai 1990). Volume (jim?) was calcu-lated from average radius assuming that pollen grains were perfect spheres. Analysis of variance of pollen area and volume was done by the general linear models procedure (SAS Institute 1990). Sources of variation were species (7 df), 'Mention of a trademark or proprietary product does suarantee or warranty of the product by the U.S. Department of enue and does not imply its approval to the exclusion of other products that may also be suitable 236 Stipa 16(2) 1994 clone within species (16 df), and residual (2,773 df). The effect of species was tested using clone within species as error; clone within species was tested using residual as error. Means were compared using least significant difference at the 5% level of significance (Steel and Torrie 1980). The cluster procedure (SAS Institute 1990) was used to assign clones to non-overlapping hierarchical groups ased on multivariate analysis of chromosome number, pollen area, and pollen volume. Prior to conducting the cluster analysis, data were standardized toa mean of 0 and standard deviation of 1 using procedure “standard” (SAS Insti- tute 1990). Crosses between the native North American taxa of Saccharum and elite sug- arcane clones, cytoplasmically male-sterile (cms) Sorghum bicolor (L.) Moench, and Vetiveria zizantoides (L.) Nash (used as females) were attempted in Fall 1993. Flowering of sugarcane clones was induced using standard photoperiod treat- ment (Dunckelman and Legendre 1982). Flowering of Sorghum, North Ameri- can Saccharum, and V. zizanioides occurred under natural photoperiod. Conven- tional methods of crossing, seed maturation, seed germination, and seedling establishment were used (Dunckelman and Legendre 1982). RESULTS AND DISCUSSION Chromosome number varied among and within the North America Saccharum species (Table 1). Saccharum alopecuroidenm and S. baldwinii were 2n = 30. Counts of 2n = 30 in E. strictus from ‘Texas (Celarier 1956b) and E. tracyi (Gill and Grass 1986) are consistent with our data. A count of 2n = 60 in S. alopecuroidenm (Celarier 1956b) is inconsistent with our data. However, two clones of S. gigan- team (2602 and 2603) with 2n = 60 and characteristics similar to 8. a/opecurot- deum were collected in Tennessee and Arkansas. It seems probable that Celarier’s count of 2n = 60 may be based on S. g/ganteum according to present taxonomic concepts. Clones with 2n = 30 and 2n = 60 are probably 6x and 12x, respectively. Saccharum giganteum consisted of clones with 2n = 30, 60, and 90 chromo- somes. These are the first counts for the species, the first indication that the species is a polyploid series, and the first report of 2n = 90 (18x) in the taxa traditionally placed in Erianthus. The 2n = 30 types were collected from Mary- land south to Alabama and in southern Louisiana; the 2n = 60 types were col- lected in Alabama, Arkansas, Georgia, Louisiana, and Tennessee; and the 2n = 90 types were collected in southern Georgia and Florida. This indicates a geo- graphic effect on distribution of cytotypes. Future studies are planned to define the relationship between chromosome number and morphology within this species. Saccharum bengalense Retz. {= E. bengalense (Retz.) Bharadw.} has also been shown to be a polyploid series with 2n = 20, 40, and 60 chromosomes (Mehra et al. 1968). Wesster, Cytological studies on Saccharum 257 TABLE oe oh be ] 1] P ; £ : North A ‘ : Ee L S. alopecuroideum 2n = 30: ALABAMA. Blount Co.: Hwy 278, 32 km W of Gadsden, 6 Nov 1992, Webster 2551 (TAES). Lamar Co.: Hwy. 18, 24 km E of Fayettsville, 6 Nov 1992, Webster 2548 (TAES). GEORGIA Banks Co.: Interstate 85, 3 km E of Commerce, 7 Nov 1992, Webster 2553 (TAES). Brooks Co.: Hwy 84, 8 km E of county road 219, 27 Oct 1992, Webster 2533 (TAES). Forsyth Co.: Hwy 9,6kmS of Cummings, 7 Nov 1992, Webster 2552 (TAES). MISSISSIPPI. Choctaw Co.: Natchez Trace Pkwy., 13 km S of Hwy 9, 28 Oct 1992 Webster 2544 (TAES). TENNESSEE Putnam Co.: Interstate 40, just E of mile marker 293, 8 Nov 1993, Webster 2600 (TAES). S, baldwinti 2n = 30: ALABAMA. Tuscaloosa Co.: Hwy 82, 8 km N of Tuscaloosa, 28 Oct 1992, Webster 2541 are GEORGIA. eee Co.: Hwy 23 & 121, 8 km S of Folson, 26 Oct 1992, Webster 2525 (TAES). Lowndes Co.: Hwy 84, 8 km W of Valdosta, 27 Oct 1992, Webster 2532 (TAES) S. brevibarbe var. brevibarbe 2n = 60: ARKANSAS. Perry Co.: Hy 10; 2 km E of Hwy 324 jcc, 10 Nov 1993, Webster 2607 (TAES). White Co.: Hwy 367, y limits, 9 Nov 1993, Webster 2605 (TAES). S. brevibarbe var. contortum 2n = 60: ALABAMA. Autauga Co.: Hwy 82, 2 km E of county aA 29, 28 Oct 1992, Webster 2538 (TAES). Chilton Co.: Hwy 82,7kmS pw road 65, 28 Oct rane es 2539 (TAES). Houston Co.: Hwy 84, at AL state line, 27 Oct 1992, oe 2536 (TAE mar Co.: Hwy 96, 8 km from MS srate line, 6 Nov 1992, rene 2547 (TAES). Pickens cc. Hwy 82,13km from MS state line, 28 Oct 1992, Webster 2542 (TAES). ARKANSAS. White Co.: Hwy 64, 6 km W of Beebe, 10 Nov 1993, eee ene Yell Co.: ase 10,2 km W of Birta, 10 Nov 1993, Webster 2608 (TAES). GEORGIA. Decatur Co.: Hwy 84, 1 km from jct 285, 14 km E of Donaldsonville, 27 a 1992, ae 2535 (TAES). ae. Somerset Co.: octaw Hwy 9, 28 Oct 1992, Webster ak (TAES). NORTH CAROLINA. Greene me Hwy 13 Hwy 97, 24 Oct 1992, ee 2508 (TAES). SOUTH CAROLIN A. eee oe Has 301, Jacksonville, 13 Nae 93, Webster 2609 (TAES). VIRGINIA. Accomack Co.: ae 13,91 m off of Parkway, 2 km N of Keller, 23 Oct 1992, a ra oe (TAE S. coarctatum 2n = 60: ALABAMA. Henry Co.: Hwy 95, 29 km N of state line, 27 Oct 1992, Webster 2537 (TAES). FLORIDA. Clay Co.: Hwy 301, 27 km N of Starke, 26 Oct 1992, Webster 2527 (TAES). GEORGIA. Beane Co.: Hwy 301, 14km N of Nahunta, 26 Oct 1992, Webster 2523 (TAES). Evans Co.: Hwy 301, 8 km N of Canoochee River, 25 Oct 1992, Webster 2520 (TAES). Grady Co.: Hwy 84 & oes Road, 27 Oct 1992, Webster 2534 (TAES). Screven Co.: Hwy 301, 8 km S of state line, 25 Oct 1992, Webster 2519 (TAES). Wayne Co.: Hwy 301, 3 km N of Jones Creek, Bethel Church, 25 Oct 1992, Webster 2522 (TAES). SOUTH CAROLINA. Dillon Co:: Hwy 301, Little Pee Dee River, 8 km S of the state line, 24 Oct 1992, Webster 2513 (TAES). Clarendon Co.: Hwy 301, 1 km N of Fox Tindal Road, 25 Oct 1992, Webster 2516 (TAES). Orangeburg Co.: Hwy 301, 11 km E of Orangeburg, 25 Oct 1992, Webster 2517 (TAES). S. giganteum 2n = 30: ALABAMA. Bibb Co.: Hwy 82, 2 km from ject Hwy 91, 28 Oct 1992, Webster 2540 (TAES). FLORIDA. Gilchrist Co.: Hwy 26, 13 km E of Trenton, 26 Oct 1992, Webster 2528 (TAES). 238 Stipa 16(2) 1994 Table 1. Continued LOUISIANA. Terrebonne Parish: Spanish Trail Farm, Chacahoula, Barner 2554 (TAES). NORTH CAROLINA. Robeson Co.: Hwy 301, 5 km N of Little Marsh Swamp, 24 Oct 1992, Webster 2512 (TAES). Wayne Co.: Hwy 13, 2 km N of Rachel Jackson Hwy, 24 ies — Webster 2510 (TAES). MARYLAND. Wicomico Co.: Hwy 50, just W of exit 331, 1992, Webster 2501 (TAES). SOUTH CAROLINA. Bamberg Co.: Hwy 301, 11 = S of Bamberg, 25 Oct 1992, Webster 2518 (TAES). Clarendon Co.: Hwy 301, 2 km N of leans 25 Oct 1992, Webster 2515 (TAES). VIRGINIA. Accomack Co.: Hwy 13, 3 km N of Hwy 704, 23 Oct 1992, Webster 2505 (TAES). Southampton Co.: Hwy 258, 3 km N a ~ state line, 24 Oct 1992, Webster 2506 (TAES). 2n = 60: ALABAMA. Cullman Co.: Hwy 91, at 4 mile marker, 6 Nov 1992, Webster 2550 (TAES). Lamar Co.: Hwy 50, at AL state line, 6 Nov 1992, Webster 2546 (TAES). Walker Co.: Hwy 69, 3 km S of Jasper, 6 Nov 1992, Webster 2549 (TAES). ARKANSAS. White Co.: Hwy 367, 91 m E of Bradford city limits, 9 Nov 1993, Webster 2603 (TAES). GEORGIA. Evans Co.: Hwy 301,8km N of Canoochee River, 25 Oct 1992, Webster 2521 (TAES). LOUISIANA. St. Landry Parish: Hwy 49, 53 km N of Opelousas, 30 Oct 1992, Webster 2545 (TAES). TEN- NESSEE. Carroll Co.: Hwy 70A bypass, N of Huntington, 8 Nov 1993, Webster 2602 (TAES). 2n = 90: F neve Dixie Co.: Hwy 19 & 98, 2kmS of county road $358, 26 Oct eek, Webster 2529 TAES). Taylor Co.: Hwy 51, 10 km from the - If of Mexico, 26 Oct 1992, Webster 2530 CAS ae 221,6km N of Perey, 27 Oct 1992, Webster 2531 (TAES). = i Charlton Co.: H 3&121,2kmN of the FL line, ee 1992, Webster 2526 (TAES Saccharum brevibarbe vars. brevibarbe and contortum and S. coarctatum were 2n = 60. Gill and Grass] (1986) reported that a clone of E. brevibarbis was 2n = 60, which confirmed our finding for the species. However, they reported that a clone of E. aan was 2n = 30, which may be the result of misidentification Saccharum brevi var. contortum is characterized as having callus hairs equal to or shorter than the spikelet, while callus hairs are either absent (S. ba/dwini1) or longer (S. eigantenm and S. alopecuroideum) than the spikelet in 2n = 30 types (Webster and Shaw 1995 Meiosis was normal in all clones and univalents and quadrivalents were ex- tremely rare (Fig. 1). Celarier (1956b) noted bivalent size polymorphism in North American Saccharum. Mean bivalent area of 4.4 43 uum- (range 1.58 to 8.94 jum*), 3.25 um? (range 1.10 to 8.77 rm”), and 4.43 pam? (range 1.86 to 7.61 1m”) in 2n = 30, 60, and 90 biotypes, respectively, differed little among cytotypes. Bivalent area was not normally distributed; the distribution was generally shifted toward the smaller size classes. Bivalents of an Old World clone (E. rafipilus, 2n = 20) averaged 3.89 ppm? (range 3.25 to 5.28 am’), and those of an elite sugar- cane clone LCP 81-30 (2n = 105, 108) averaged 3.42 ppm? (range 1.55 to 6.99 pm?) (Burner and Legendre 1994; Burner unpublished data). Thus, North American Saccharum tends to have wider ranges of bivalent area than Old World Saccharum, but there is little difference in area or range between New World Saccharum and sugarcane. The data support the observation by Gould (1956) that there is no obvious correlation between chromosome number and chromosome size in the Andropogoneae. WesSTER, Cytological studies on Saccharum 239 EE Fic, 1. Meiotic metaphase I chromosomes of North American Saccharum. Bar in each figure represents 10 pm. (A) S. gigantenm(n = 15 bivalents) [x 1600}. (B) S. brevibarhe var. contortum (n = 30 bivalents) {1600}. (C) S. g/gantenm(n = 45 bivalents) [x 1000}. 240 SIDA 16(2) 1994 Pollen area and volume differed significantly among species and clones within species (Table 2). As expected, species with 2n = 30 had smaller pollen than those with 2n = 60 or 2n = 90 (Gould 1957). Pollen size has been used as a taxo-nomic character to separate closely related taxa differing in chromosome number (Gould 1953 & 1957). However, variation in pol- len size among clones within species indicates that pollen size would be an unreliable predictor of species. Clones were assigned to clusters based on chromosome number, pollen area, and pollen volume. The dendrogram beginning with seven clusters (R“ = and ending with one cluster (R* = 0.00) is shown in Fig. 2. There was a Group ID 2516 S. coarctatum 2545 S. giganteum 2546 S. giganteum 2550 S. giganteum 2529 S. giganteum 2531 S. giganteum 2504 S. brevibarbe var. contortum 2509 S. brevibarbe var. contortum 2607 S. brevibarbe var. ea aoe 6 var. contortu 2513, S% aan ee 2523 S. coarctat 2547 S. arc ae var. contortum eum 2541 S. Balai eo a et 0.97 0.96 0.94 0.87 0.76 0.50 0.00 -squared Fic. 2. Dendrogram of relationships among 25 North American Saccharum clones based on pollen area, pollen volume, and chromosome number. WEBSTER, Cytological studies on Saccharum TABLE 2. Mean poll 1 poll | f North American Saccharum clones. Pollen Species/clone Area (ym?) Volume (jim?) S. alopecuroitdeum 1052 25761 2544 1028 24840 2551 1044 25421 2552 1084 27022 S. baldwinti 1218 32146 292 1282 34478 2532 1243 33248 2541 1129 28712 S. brevibarbe var. brevibarbe 1926 63635 2607 1926 63635 S. brevibarbe var. contortum 2020 68814 2504 1854 59960 2509 1914 63254 2536 2476 92818 2547 1752 55298 2606 2106 7274) S. coarctatum 1751 55459 2513 1711 33577 2516 1603 48565 2517 1924 63544 2523 1765 56152 S. giganteum 2n = 30 1184 30987 2506 1271 34365 2510 1277 34685 2512 1045 25494 2515 1144 29405 S. gigantenm 2n = 60 1447 41065 2545 1529 45172 2546 1417 40494 2550 1394 39148 S. gigantenm 2n = 90 1478 43090 2529 1447 41883 2531 1510 44297 Ce 1511 45366 CV (%) 12.3 18. LSD (0.05) - Species 207 10324 52 2397 LSD (0.05) - lone within species 242 Sipa 16(2) 1994 TasLe 3. Seed yield of crosses with North American Saccharum. Parents Viable Cross no. Female Male seeds (no.) 3003 A4692! 2508 ) 3004 A3TX7000! 2535 0 3006 AoTX632! 2504 2 3007 A83E! 2551 l 3008 ATX 623! 2525 0 3011 POJ 2222° 2543 26 3031 IND 81-53" 2532 0 3144 CP 65-357? 2531 148 3145 CP 91-534? 2531 92 3206 CP 92-6703 2530 12 3209 CP 65-357? 2533 { 3210 CP 88-755? 2533 0 ! i. ides 2n = 20 (D.M. Burner, blished data). 3 E likes sugarcane clone. 10% loss in explained variance when seven clusters were combined to four clusters. Four-cluster analysis (R2 = 0.87) seemed to be most informative. Clones of S. giganteum (2n = 60) and Webster 2516 (S. coarctatum, 2n = 60) were assigned to cluster group 1. Morphology of Webster 2516 is otherwise typical of S. coarctatum. Clones of S. giganteum with 2n = 90 chromosomes were assigned to cluster group 2. Clones of 8. brevibarbe vars. brevibarbe and contortum and S. coarctatum, except for Webster 2516 were aebigned to cluster group 3. Clones with 2n = 30 were assigned to cluster group 4. Thus, the four-cluster analysis was generally consistent with present taxonomic con- cepts (Webster and Shaw 1995) and provided evidence of diversity be- tween the cytotypes of S. giganteum. There was a further loss of 11% of explained variance when the 2n = 60 and 2n = 90 cytotypes of S. gzgantenm were joined to form three clusters. The three-cluster analysis explained 76% of variance. Only 50% of total variance was explained by two clusters. Crosses were successful between elite sugarcane and North American Saccharum (Table 3). A cross with Webster 2531 (2n = 90) yielded 448 seeds, and other crosses yielded 0 to 92 seeds. The potential agronomic value of these F, hybrids will be evaluated in subsequent tests. Two crosses of cms Sorghum * North American Saccharum (five crosses attempted) yielded some seed, but grow-out evaluation showed that the F; progeny were not hybrid. Crosses of cms Sorghum X elite sugarcane also failed to yield hybrid ey (Burner unpublished data). Vetiveria zizanioides (2n = 20) X Webster 2532 (2n = 30) was also unsuccessful. Pistil- late sterility is frequently observed in V. z7zanzoides (Ramanujam and Kumar WessTER, Cytological studies on Saccharum 243 1963), and despite several attempts we have never obtained viable seeds from this species. Love (1951) noted that differing chromosome numbers within a species, as we found in S. gzganteum, indicates the species may include more than one distinguishable taxon and needs closer taxonomic inspection. Léve (1951) and Nannfeldt (1938) further argue that intraspecific difference in ploidy level has fundamental systematic value sufficient to justify the rec- ognition of their respec-tive members as species. Cytomorphological study of S. giganteum should be conducted and crosses between 6x and 18x forms, to attempt synthesis of the intermediate 12x form, would further our un- derstanding of the evolution of the genus. ACKNOWLEDGMENTS We thank B. Duet and S. Michalisko, USDA-ARS, Houma for technical assistance, and Drs. P. Bramel-Cox (Manhattan, Kansas) and F. Miller (Col- lege Station, Texas) for providing sorghum seed. REFERENCES Basu, C.N. and K. Srinivasan. 1960. Chromosome numbers in the genus Erianthus. Sci. Cult. 26:230-231 Burner, D.M., M.P. Cusaian and B.L. LEGENDRE. oe Resistance of sugarcane relatives injected ae Ustilago scitaminae. Plant Dis. 77:1221-1223. Burner, D.M. and B.L. LeGenpre. 1994. Cytogenetic i fertility characteristics of elite sugarcane clones. Sugar Cane 1994 (1):6-10 CELARIER, R.P. 195G6a. Additional evidence for five as a basic number of the Andropogoneae. Rhodora 58:135-143. _.. 1956b. Cytotaxonomy of the Andropogoneae I. Subtribes Dimeriinae and Saccharinae. Cytologia 21:272-291. Ciayton, W.D. and $.A. Renvoize. 1986. Genera graminum: Grasses of the World. Kew, London: Royal Botanical Gardens DANIELS, J., P. SMITH, N. PATON, a : A. WiLuiaMs. 1975. The origin of the genus Saccharum. Sugarcane Breed. Newsl. 36:2 DanizLs, J. and B.T. Roacu. | | eee and evolution. pp. 7-84. In: D.J. Heinz (ed.) Sugarcane improvement through breeding. Elsevier, New York. DUNCKELMAN, ean and B.L. LEGeNprRE. 1982. Guide to sugarcane breeding in the temper- ate zone DA-ARS ARM-S-22. Agric. Res. Serv. Southern Region. USDA, New Or- leans, is ouisiana. GaLal PRODUCTION Lrp. 1990. Cue-2 cae analyzer operation manual, version 3.2. Olympus Corp., Lake Success, New York. Git, B.S. and C.O. Grassi. 1986. ae of genetic transfer in intergeneric hybrids of sugar cane. Sugar Cane 1986(2):2-7 GouLb, F.W. 1953. A cytotaxonomic study in the genus Andropogon. Amer. J. Bot. 40:297— 306. 1956. Chromosome counts and cytotaxonomic notes on grasses of the tribe Aacianovancae Am. J. Bot. 43:395—404. 244 Stipa 16(2) 1994 __. 1957. Pollen size as related to polyploidy and speciation in the Andropogon wari A. barbinodis ee as Brittonia 9:71-75. Grass, C.O. 1980 Breeding A pog the generic level for biomass. Sugarcane Breed. Newsl. 43:41-57. GrisHaM, M.P., D.M. Burner, and B.L. LeGenpre. 1992. Resistance to the H strain of sugarcane mosaic virus among wild forms of sugarcane and relatives. Plant Dis. 76:360— 362. — HARLAN co R. er M - de Wer. 1975. On O Winge and a prayer: The origins of polyploidy. Bot HEINZ, ah el ; eee cytogenetics. pp. 279-293. In: T. Tsuchiya and P.K. Gupta (eds.) Chromosome engineering in plants: Genetics, Breeding, Evolution, Part B. Elsevier, msterdam. Hircucock, A.S. 1951. Manual of the grasses of the United States, 2nd ed. (revised by A Chase). U.S. Dept. Agric., Misc. Publ. No. 200, Washington, D.C. Love, A. 1951. Taxonomical evaluation ae pee 263-284 Maan. P.N., PK. KHosta, B.L. Koni, and J.S. K 1968. Cytological studies in the North Indian grasses wee 1). Res. Bull. Punjab. Univ. a7 230. Monan, N. and T.V. SREENIVASAN. 1983. Distribution and chromosome number in the genus Erianthus. Sugarcane Breed. News|. 45:50-63. Moor, P.H. 1987. Breeding for stress tolerance. pp. 503-542. In: D.J. Heinz (ed.) Sugarcane mprovement through breeding. Elsevier, New York. cos S.K. 1957. Origin and distribution of Saccharum. Bot. Gaz. 119:55-61. NAnneELpt, J.A. 1938. Poa maroccana Nannf.n. sp. and P. rivalorum Maire & Trabut, two more tetraploids of sect. ee A. & Gr., and some additional notes on Och/opoa. Svensk. Bot. Tidskr. 32:295-32 RAMANUJAM, S. and S ae 1963. Irregular meiosi tated with pollen sterility in Vetsverta zizantoides (Linn.) Nash. Cytologia 28:242—24 Sas INstITUTE. 1990. SAS/STAT user’s guide, version 6, fourth ed. SAS Institute, Cary, North Carolina. SmiTH, L. 1947. The acetocarmine smear technic. Stain Technol. 22:17—31. pa, T.V., B.S. AHLOOWALIA, and D.J. Heinz. 1987. Cytogenetics. we 211-253.In einz (ed.) Sugarcane improvement through breeding. Elsevier, New Ye Suet, R. o D. and J.H. Torrie. 1980. Principles and procedures of statistics. McGraw- Hill, New York. Wesster, R.D. and R.B. SHaw. 1995. Roan of the native North American species of Saccharum (Poaceae: Andropogoneae). Ann. Missouri. Bot. Gard. (submitted) WHALEN, M.D. 1991. Taxonomy of Saccharum (Poaceae) Baileya 23:109-125. SYSTEMATIC STUDY OF TEXAS POPULATIONS OF PHACELIA PATULIFLORA (HYDROPHYLLACEAE) JENNIFER ANN MOYER ' and BILLIE L. TURNER Department of Botany The University of Texas Austin, TX 78713, U.S.A. ABSTRACT Populations of Phacelia patuliflora, a — of the southeastern United States and closely adjacent Mexico, were investigated in Texas. The species was formerly treated as having two varieties, P. patulifora vat. patuliflora and P. ie var. teucriifolia. The present investigation ae that an additional regional variety from south Texas be recognized for which the name P. patuliflora var. austrotexana J. oyer, var. nov. is proposed. A key to the varieties of this complex along with a map showing hee distribution is included. RESUMEN Se investigaron en exes arora niles patuliflora, una especie Gel sureste de los Esta- dos Unidos yt Méx iedades, P. patu- lifora var. patuliflora y P. patuliflora var veucrifolia La i igacion sugiere que puede ser reconocida una variedad regional adicional del sur a Texas, para ina que se propone el nombre P. patuliflora var. austrotexana J.A. Moyer, var. nov. Se incluye una clave para las variedades de este complejo junto con un mapa que muestra su distribuci6n. Key Worps: Hydrophyllaceae, Phacelia, Texas Phacelia patuliflora belongs to section Cosmanthus Brand of the Hydrophy]l- laceae. This section is characterized by the presence of a gland or nectary on the principal vein of the corolla (Constance 1949). Cosmanthus consists of 14 species witha south to north distribution starting in the Guatemalan and the Mexican highlands and extending northwards into Texas and Oklahoma and eastward through the Ozarks to the Appalachians. Other species of Phacelia do not over- lap the distribution of the subgenus comes, ee a ane of P. ae Hook. and its few relatives, which are n cytologically similar to Cosmanthus (Constance 1949). Giles el 968), in che oe work on Cos- manthus, also recognized 14 species in the section. Turner (1991) has recently added three additional new species, P. a/totonga, P. carmenensis, and P. neffiz, to the Cosmanthus complex. The Phacelia patuliflovra complex is most commonly aligned with P. strictiflora and P. laxa (Constance 1949, 1950; Gillett 1968). As indicated in Fig. 1 and in the taxonomic treatment that follows, we recognize three infraspecific elements 'Present address: U.S. Army Corps of Engineers, Planning Division, P.O. Box 1715, Baltimore, MD 21203-1715. Swwa 16(2): 245 — 252.1994 246 Stipa 16(2) 1994 in the group, all of which are distinctly allopatric and show little or no tendency to intergrade when observed in the field. The var. austrotexana, a segregate from var. patuliflora as treated by Constance, is probably closer to P. /axa Small than 1t is to P. patuliflora var. patuliflora {which appears equally close to P. strictiflora (Engelm. & Gray) A. Gray, according to Constance (1949)}. Indeed, Constance in his discussion of P. /axa, states that the species is: . exceedingly difficult to separate from P. patuliflora in the herbarium ... since all the chinoaeies of P. laxa, except perhaps the low number of ovules, have been difubeds into P. eile lal its distinctness rests solely upon a combination of characters, none of which is clearly distinguishing in itself. He goes on to note, —— that in the field it is easily recognized by its small, pale flowers, petiolate cauline leaves, sparse pubescence and preference for shaded alluvial situations, all of which the senior author also observed in her field excur- sions. Nevertheless, Constance opined that P. /axa has been very nearly “swamped out” by the c ion from its own recombinational products with P. patuliflora var. fencriifolia. Our field observations and examination of distributional ranges, make it difficult to accept that P. /axa has, to any considerable extent, hybrid- ized with P. patuliflora var. teucriifolia. Rather, the distribution pattern and mor- phological data strongly suggest that P. patuliflora vat. austrotexana is a region- ally differentiated taxon little affected by hybridization from P. /axa or other taxa. Whatever its origin, it would appear on both morphological and geo- graphical grounds that P. /axa is much closer to P. patuliflora var. austrotexana than it is to P. patuliflora var. teucriifolia. BIOGEOGRAPHY The three varieties of Phacelia patuliflora in Texas are distinctly separated geo- graphically (Fig. 1). Interestingly, the collective distribution of the subgenus Cosmanthus indicates a migration northward from Mexico (Constance 1950; Turner 1991). Within Texas this migration presumably allowed for geographi- cal diversification and habitat specialization for each variety. Phacelia patuliflova is characterized by hard-to-locate but often relatively dense populations which appear to be thriving. The species is not present in many apparently suitable areas near the known populations. Frequent human distur- bance or overgrazing at the sites where P. patuliflora does occur might explain the difficulty in locating sizeable colonies. TAXONOMIC TREATMENT Phacelia Sgr (Engelm. & A. Gray) A. Gray, Proc. Amer. Acad. Arts 10:321. 1875. Excota patuliflora Engelm. & A. Gray, Boston J. Nat. Hist. 5:45. 1845. TYPE: s. A: Texas: Austin Co.: in a woods near San Felipe, Mar 1843, Lindheimer II- 280 (HOLOTYPE: GH; Isorypes: NY, MO). Moyer AND TURNER, Systematics of Phacelia patuliflora 247 aie ele Buckl., Proc. Acad. Nat. Sci. Philadelphia 13:463. 1862. non A. Gray 8. Typ S.A.: Texas: Travis Co.: Austin, 1860, Buckley 5.2. (HOLOTYPE: PA; ISOTYPES: a MO), KEY TO THE VARIETIES OF PHACELIA PATULIFLORA 1. Peduncles with small, well-defined, glandular hairs, these ca 0.1 mm in length and oe among a rather even vestiture of mostly short, non-glandular hairs 0.5 mm in length or les var. patuliflora 1. Peduncles Sen glandular lane the vestiture various but not as described in the abo (2) ve 2. Calyx lobes markedly acute, elongating in fruit to twice the length of the invested capsules; calcareous soils of west-central Texas and adjacent Mexico teucritfolia Z. Calyx lobes rounded or rather abruptly obtuse; mostly sandy soils of southern Texas var. austrotexana Phacelia patuliflora (Engelm. & A. Gray) A. Gray var. patuliflora Annual herbs, 8—60 cm high, branching at the base with branches stiffly as- cending to decumbent. Leaves rarely rosulate or persistent, petiolate, truncate at base, oblong to oval, 2-10 cm long, 14 cm broad, pinnately lobed to pinnati- fid with 1—7 pairs of smaller lobes or remote leaflets at base and a larger terminal lobe, all coarsely dentate or lobed with obtuse or acute teeth, strigulose to hir- sute on both surfaces, the cauline leaves reduced, mainly sessile, oblong and coarsely dentate, distinctly lobed. Inflorescence a scorpoid cyme of 5—30 flowers, loosely hirsute to strigulose or canescent; the peduncles with small, well-de- fined glandular hairs, the latter about 0.1 mm in length; the mature pedicels weakly ascending to widely spreading or decumbent, 3—15 mm long. Corolla, purplish-violet to lavender, commonly with white centers, broadly campanu- late to rotate-campanulate, 8-16 mm broad, the lobes finely crenulate, moder- ately to sparsely pilose on back. Stamens 5-8 mm long, included, the anthers 1.0-1.5 mm long. Style included, cleft to midpoint, hirsute below. Capsule globose, 4-6 mm across, the seeds 10-15, about 2 mm in length, ovoid-angled, brown, areolate and finely alveolate. Chromosome number, n = 9 pairs. Distribution and Ecology: Central and eastern Texas (Fig. 1) mainly in sandy soils of roadsides, railroad embankments and alluvial soils of stream and river beds. The plants occur in disturbed areas in full sun or in partial shade of de- ciduous roadside thickets. Often grazed, plants begin to flower in mid-March, peak in early April and have set seed by the middle of May. Population Site Descriptions: (Fig. 1) Three populations, 5—15 individuals each, of var. patuliflora were found in Burnet County on April 13, 1991. Populations 1, 2 and 3 were located on sandy, well-drained roadside banks in the Buchanan Dam area. These sites were situated on large sandstone outcrops with smaller flags present on the surface. Some of the plants in all of the sites had been grazed. 248 Stipa 16(2) 1994 | ) med dy yA A eaten amano | aie ° var. teucriifolia A yar. patuliflora | * var. gustrotexana Fig. 1. Distribution of Phacelia patuliflora in Texas based upon specimens at LL, TEX. Popula- y. ee tions examined in the field and discussed in the text are labeled according Interestingly, plants were confined to the sandy areas, although tracts with abundant humus were in close proximity. Grasses and occasional plants of Opuntia macrorhiza characterized the area. Population 4 of var. patuliflora was observed at or near the type locality in San Felipe, Texas on April 3, 1992. This population was rather dense, with 1000-1500 plants covering approximately half an acre. The plants were growing in very sandy loam with a few sandstone outcrops and were located in full sun. Other plants growing in this area were Lupinus texensis, Pyrrhopappus pauciflorus, Gaillardia pulchella, Cirsium texanum and Castilleja indivisa. MOYER AND TURNER, Systematics of Phacelia patuliflora 249 vee Heat var. teucriifolia (1.M. Johnst.) Constance. Tver: MEXICO. CoanuiLa: Muzquiz, 12 Apr 1936, Marsh 2120 (HoLoryPe: GH; Isotype: a Annual herbs, 10-60 cm high, branching at the base with branches stiffly ascending. Leaves rarely rosulate or persistent, petiolate, truncate at base, ob- long to oval, 2-10 cm long, 14 cm broad, pinnately lobed to pinnatifid with 1-7 pairs of smaller lobes or remote leaflets at base and a large terminal lobe, all coarsely dentate or lobed with obtuse or acute teeth, strigulose to hirsute on both surfaces, the cauline leaves reduced, mainly sessile, oblong and coarsely dentate, distinctly lobed. Inflorescence a scorpoid cyme of 5—30 flowers, loosely hirsute to strigulose or canescent, the mature pedicels weakly to stiffly ascend- ing, 3-15 mm long. Corolla, purplish-violet to deep bluish-purple, commonly with white centers, broadly campanulate to rotate-campanulate, 12-20 mm broad, the lobes finely crenulate, moderately to sparsely pilose on back. Stamens 5—8 mm long, included, the anthers 1.0-1.5 mm long. Style included, cleft to midpoint, hirsute below. Capsule globose, 4-6 mm across, the seeds 10-15, about 2 mm in length, ovoid-angled, brown, areolate and finely alveolate. Chro- mosome number, n=9 pairs. Distribution and Ecology: Edwards Plateau region of Texas southward to Mexico (Figure 1). Plants commonly occur in calcareous soils in draws and river beds as well as along roadsides. Populations are frequently located in the partial shade of deciduous trees and in the full sun of disturbed roadside ditches. Flowering begins in late March with its peak in mid-April. Seeds are set and plants die back by late May. Population Site Descriptions: (Fig. 1) Populations 5 and 6 occurred in Montell, Uvalde County and were sampled in mid-April of 1991. Both were located on highway embankments. Populations were comprised of 20-50 plants which were widely scattered, located in full sun and on the naturally calcareous soil of the western Edwards Plateau region. Associated species included, De/phininm cavolinianum, Callirboe involucrata, Verbena bipinnatifida and Phlox drummondii, as well as assorted grasses. Populations 7, 8 and 9 were all located along the San Saba River in Menard County. These sites were characterized by calcareous soils of a grainy nature that seemed well drained. Populations 7, located at a picnic area on highway 83, two miles south of Menard, and population 8, located in the town of Menard, were situated in the partial shade of pecan bottoms. Both populations were dense, comprised of 500-1500 individuals. Other species in these areas included Ocnothera speciosa, Gaura coccinea and Penstemon triflorus. Population 9 was located on the banks of a road cut along FM 1311 where it crosses the San Saba River and was comprised of 24 individuals. The soil at this site was sand that appeared to have been recently deposited. It was also in 250 Sipa 16(2) 1994 partial shade. Associated species were grasses and abundant individuals of Clematis pitcheri. Phacelia patuliflora var. austrotexana J.A. Moyer, var. nov. Differt a Phacelia patuliflora (Engelm. & A. Gray) A. Gray var. patuliflora trichomatibus i] Teecrecaaa aaa aia Annual herbs, 8—60 cm high, branching at the base with branches stiffly ascending to decumbent. Leaves rarely rosulate or persistent, petiolate, truncate at base, oblong to oval, 2—10 cm long, 1-4 cm broad, pinnately lobed to pin- natifid with 1—7 pairs of smaller lobes or remote leaflets at base and a large terminal lobe, all coarsely dentate or lobed with obtuse or acute teeth, strigulose to hirsute on both surfaces, the cauline leaves reduced, mainly sessile, oblong and coarsely dentate, distinctly lobed. Inflorescence a scorpoid cyme of 5—30 owers, loosely hirsute to strigulose or canescent; the mature pedicels weakly ascending to widely spreading or decumbent, 3—15 mm long. Corolla, pur- plish-violet to lavender, commonly with white centers, broadly campanulate to rotate-campanulate, 7-16 mm broad, the lobes finely crenulate, moderately to sparsely pilose on back. Stamens S—8 mm long, included, the anthers oblong 1.0-1.5 mm long. Style included, cleft to midpoint, hirsute below. Capsule globose, 4—6 mm across, the seeds 10-15, about 2 mm in length, ovoid-angled, brown, areolate and finely alveolate. Chromosome number, n=9 pairs. Type: U.S.A.: TEXAS: San Patricio Co.: along railroad ca. 2.5 mi. SW of Sinton, 22 Mar 1969, D.S. Correll 36834 (HoLoryPE: LL!; soryre: TEX! Representative specimens: TEXAS. Bee Co.: just south of Pettus, 2 Mar 1968, Corre// 35511. Brooks Co.: King Ranch, Encino Division, 26 Apr 1949, Lundell 14975, Cameron Co.: Ar- royo Colorado, Harlingen, 13 aes 1964, Correll 28973. Dimmit Co.: Carrizo ae 12 Mar 1964, Cuellar 43. Duval Co.: mi W of Hebbronville, 5 Mar 1962, Trivino 4. Goliad Co.: open places, Mar 1927, C.B. ff illiams 45. Hidalgo Co.: S of Encino, 19 Mar 1942, Lundell 10829. Jim Hogg Co.: Hebbronville, 14 Mar 1931, Tharp s.n. Jim Wells Co.: field at Alfred, 30 Mar 1970, Corre// 38310. Kenedy Co.: Sarita along Hwy 96, 14 Mar 1941, Runyon 2469. Kleberg Co.: 10 mi S of Kingsville, 6 Mar 1959, Turner 4467. Live Oak Co.: along US Hwy. 59, 20 SW of Georgewest, 14 Mar 1964 Fowler & Vergara 99. Nueces Co.: Petronila Creek, between Bishop and Chapman Ranch, 12 Mar 1964, Corre// 28925. San Patricio Co.: near Mathis, 29 Mar 1941, Lundell 10081. Webb Co.: 8.3 mi S of Mirando City, 16 Mar 1966, Correll 32259. Distribution and Ecology: South and southeastern Texas in the sandy, well- drained soils of roadsides and alluvial stream beds (Fig. 1). Found in the follow- ing counties: Bee, Brooks, Cameron, Dimmit, Duval, Goliad, Hidalgo, Jim Hogg, Jim Wells, Kenedy, Kleberg, Live Oak, Nueces, San Patricio, Webb. Populations were found most frequently in full sun and often comprised the only vegetation in highly disturbed areas. Flowering begins in March with maximum anthesis in early April. Population Site Descriptions: (Fig. 1) Population 10, located in Bee County, MOYER AND TURNER, Systematics of Phacelia patuliflora 251 was sampled on 12 March 1991. It was comprised of approximately 500— 700 plants which were found growing on the banks of a roadside drainage area in downtown Beeville. The drainage ditch walls and associated bank were comprised of sand with gravel and was located in the full sun. The population extended to a railroad embankment located approximately 20 meters from the roadside area. This embankment, also comprised of sand and gravel, was located in full sun and was well-drained. Individuals were not found in the 20 meter swath of loamy sand that lay between the drain- age area and the embankment. Associated species found growing in the sand and gravel included Corydalis micrantha and isolated individuals of Opuntia macrorhiza. Population 11 was collected in Cameron County on 14 March 1991. The site was in an abandoned commercial area off of the access road to Highway 77. Approximately 100-150 individuals were growing in what appeared to have been a sand and gravel parking lot in direct sun. No plants were found ina nearby (20 meters distance) overgrown garden area of rich loam. Kenedy County was visited on 14 March 1991 and three populations were located. Population 12 was located along Highway 77 on the sandy berm of a drainage area. The berm, located in full sun, appeared to have been deposited some years prior and was comprised entirely of sand. Approximately 50 indi- viduals were growing in association with Sisyrinchinm ensigerum and varied grasses. Population 13 was the only partially cultivated population found. Several thou- sand individuals were growing in the sandy loam of Sarita Square in the center of Sarita. These plants were being watered daily. All individuals were located in full sun in association with Gaillardia pulchella, Castilleja indivisa, Sisyrinchium ensigerum, Gaura coccinea, Argemone albiflora and varied herbs and grasses. Population 14 was located along a roadside and the railroad embankment running parallel to and approximately 15 meters from Highway 77 near Ytturia. This site was characterized by extremely dry, well-drained sand and gravel. Approximately 100-200 individuals were present with only sparse grasses. Three populations were found in Kleberg County on 14 March 1991. Popu- lation 15 was comprised of 15 plants and was located on the alluvial sand of Escondido Creek just south of Kingsville. Individuals were found only in the sand of the banks, and not in the surrounding grassy field. All individuals were located in full sun. Associated species included Buchnera floridana and Phlox drummondit. Population 16 occurred on a railroad embankment along Highway 77 at its junction with FM 1118 inalluvial sand and gravel. The 22 plants observed were in full sun and were widely mee Associated vegetation was comprised of Corydalis micrantha, Cirsium texanum, Opunti, hiza, and Sz Population 17 was also growing ona railroad embankment running ald to Highway 77 just south of its junction with FM 628, in gravelly sand. 292 Sipa 16(2) 1994 This population was widely scattered and was comprised of approximately 50-60 individuals. However, this group was in the partial shade of a large Acacia grove. Corydalis micrantha and Sisyrinchium ensigerum were the only other species found on the embankment. Population 18 was located in Nueces County and was sampled on 13 March 1991. It was comprised of 25-50 individuals growing in sand at the foundation of a roadside warehouse off of the Padre Island Expressway in Flour Bluff. The individuals were restricted to the sandy soils that appeared to have been placed at the foundation for the purpose of creating a drainage bank. The site was located in full sun; grasses were the only other species present. ACKNOWLEDGMENTS The present study represents part of a Master's thesis by the senior author. It is based upon field work, as indicated, and examination of approximately 300 specimens of the taxa concerned on file in the Plant Resources Center at the University of Texas at Austin (LL, TEX). We are grateful to Guy Nesom for the Latin diagnosis and to the McLean Family, Red Stoner and Marky Miles for their generous donation of time and effort in locating populations. REFERENCES Constance, L. 1949. A revision of Phacelia subgenus Cosmanthus (Hydrophyllaceae). Contr. Gray ane 168:1-48 950. Some fecerseciie relationships in Phacelia subgenus Cosmanthus. Proc. Amer. hen 78:135-147. GILLETT, G.W. 1968. gee, relationships in the Cosmanthus and Phaceltas (Hydrophyl- laceae). Brittonia 20:3 4, JOHNSTON, I.M. 1943, oe sppanemieanis from Mexico. J. Arnold Arbor. 24:90-98. Turner, B.L. 1991. New species of Mexican Phacelia, subgenus Cosmanthus (Aydrophyllaceae). Phytologia 71:445-452. ASHICAULIS, A NEW GENUS FOR SOME SPECIES OF MILLEROCAULIS (OSMUNDACEAE) WILLIAM D. TIDWELL Department of Botany and Range Science Brigham Young University Provo, UT 84602, U.S.A. ABSTRACT Millerocaulis Erasmus ex Tidwell was established as a new genus for Miller's “Osmundacaulis herbstii group.” Australosmunda — Forsyth & Green was proposed subsequently for osmunda- ceous stems anatomically similar to Mi/lerocaulis, but without leaf gaps. However, a few species of Millerocaulis, including its PRS, M. eee: lack or have occasional leaf gaps. Therefore, Australosmunda is ced in the synony of Millerocaulis. Milleroca ulis is emended to include those osmundaceous species wit abnes or evn only an occasional leaf gap. inte the new arrangement, Millerocaulit includes four species M. dunlopii, M. indica, M. indentata, an chubutensis. A new genus, Ashicaulis, is proposed for species formerly in Mi/ler iets bavine merous leaf gaps. It contains 22 species, all of which are recognized nomenclaturally by new combinations. RESUMEN Millerocaulis Erasmus ex Tidwell se establecié nuevo género para el grupo de Miller i nace her bstii”. Se propuso por ello Australosmunda Hill, Forsyth & Green para los tallos Millerocaulis ae sin ene salience Sin emee EO; le Millerocaulis, incluyendo su tipo 1. dunlopii r mente elueuna foliares. Por ello, Australosmunda se pasa a i sinonimia de Mé/lerocaulis. Se enmienda Millerocautis para incluir llas especies osmundaceas que no tienen ae foliar o que la sdlo ocasi saalmanee. En la iene M elaian M. ndiea M. as y: be chubutenst 1S. Be propone un género nuevo, Mobicale para erocaulis. Contiene 22 eee: que se reconocen n nonenclariralmente con nuevas combinaciones. s n nueva, Mi/lerocau/is i INTRODUCTION The genus Mi/lerocaulis was established as a new taxon based upon the infor- mal “Osmundacaulis herbstii group” that was originally discussed by Miller (1967). The “Osmundacaulis herbstii group” was ineftectively published (Art. 29) as a new genus Mzllerocaulis by Erasmus in his unpublished doctoral dissertation (1978). Herbst (1981:37) accepted Erasmus’ M7//erocau/is but failed to validate it witha description (Art. 32.1)” (Tidwell 1986:402). Mil/lerocaulis was validated by Tidwell (1986) Australosmunda was proposed for osmundaceous stems similar to Mz/erocaulzs, but without leaf gaps (Hill et al. 1989; Fig. 1A). However, there are a few species in Millerocanlis, including its type, M. dun/opit (Kidston and Gwynne-Vaughan Sipa 16(2): 253 — 261. 1994 254 Sipa 16(2) 1994 C@mD Fic. 1. Xylem cylinders and part of xylem cylinders of four species of Mi//erocaulis and two species of Ashicaulis gen. nov. A. Millerocaulis indentata (Australosmunda indentata), illustrating ack of leaf i B,C, D. Millerocaulis euaetaee Pi urt es continuous xylem cylinder, C stages of leaf trace fc nd departure. E rbstitand F. Ashicaulis wader Tow ing many leaf gaps in their le Sas G Mille cals indica with continuous xylem cylinder and forming leaf trace. H, I, J, K 6, H. xylem without leaf gaps, I, J, K, leaf trace ee and is without leagas gaps. (Bars = 1mm). (A. Drawn from holotype of Austalosmunda (=M.) indentata, Hill etal. 1989; B, C, D, Drawn from holotype of M. a Kidston & pean Vaughan 1907; E. Redrawn from Archangelsky & de la Sora 1963; F Drawn irom holotype of Ashicaulis u ede. Tidwell & Rushforth 1968; G. Redrawn from Sharma 197 . Drawn from holotype of M. chubutensis, Herbst 1977, LJ,K. Redrawn from Herbst io TIDWELL, Ashicaulis, a new genus in the Osmundaceae 255 1907; Figs. 1B, C, D), that also lack or have an occasional leaf gap. Since the type of Mi//erocaulis essentially lacks leaf gaps, Australosmunda is placed in the synonymy of Mi/lerocaulis, and Millerocaulis is restricted to fossil osmundaceous axes lacking or having very rare or only an occasional ill-deflned leaf gap. Ashi- caulis is proposed for those species formerly in Mz//erocaulis that have numerous definite leaf gaps (Figs. 1E, F). TAXONOMY Millerocaulis Erasmus ex Tidwell emend. Tidwell Fossil osmundaceous rhizomes or arborescent axes containing stem or branch- ing stems surrounded by a mantle of leaf bases and roots. Stele an ectophloic siphonostele, with xylem cylinder usually continuous, generally 3-15, some- times up to 20, tracheids thick radially, with leaf gaps either totally lacking, incomplete, or very rare. Incomplete leaf gaps extending up to three-quarters through the metaxylem cylinder. Leaf trace separating from the xylem cylinder with only one protoxylem cluster and often, but not always, lacking axillary sclerenchyma. Scelerotic outer cortex of stem usually much wider than paren- chymatous inner cortex. Petiole bases stipulate, and adventitious roots arising either singly or in pairs. Type: M, oa (Kidston & Gwynne-Vaughan) Tidwell (Basionym: Osmandites dunlopii Kidston & Gwynne-Vaughan “dunlopz’). Millerocaulis chubutensis (Herbst) Tidwell, comb. nov. Basionym: Osmundacaulis chubutensis Herbst, Facena 1:25. 1977 Millerocaulis dunlopii (Kidston & Gwynne-Vaughan) Tidwell, comb. nov. ASIONYM: Osmundites dunlopii Kidston & Gwynne-Vaughan, Trans. Roy. Soc. Edinburgh 45:759. 1907 ( wees ). Osmundacaulis duntopii (Kidston & Gwynne-Vaughan) Miller, Contr. Mus. Paleo. Univ. Michigan [21:146. 1967 (“dunlopi” )t nom. Hae under Art. 33.2 —no a nee to basionym} 23: fas 1971. Osmund: ‘s Marshall, Tra c. New Zealand Inst. 56:210. Milieroc oie iodentars (Hill, ee) Tidwell, comb. nov. Basionym: ee eee Hill, Forsyth & Green, Palaeontology 32: baka 1989 Millerocaulis indica (Sharma) eee co mb. nov. BasionyM: lacaulis indica Sharma, che 140B:157 Ashicaulis Tidwell, gen. nov. Fossil osmundaceous rhizomes, rarely arborescent axes, containing stem or branching stems surrounded by a mantle of leaf bases and roots. Stele an ec- tophloic-dictyoxylic-siphonostele (Miller 1971), with a xylem cylinder up to 20 tracheids thick, with many deflnite leaf gaps. Leaf trace separating from the xylem cylinder with only one protoxylem cluster and often, but not always, lacking axillary sclerenchyma. Commonly, the outer sclerenchymatous cortex is wider than the parenchymatous inner, rarely the same width. Petiole bases stipular and adventitious roots arising either singly or in pairs. 256 Sipa 16(2) 1994 TYPE: Ashicaulis herbstii (Archangelsky & de la Sota) Tidwell (Osmundites herbstit Archangel- sky & de la Sota) The generic name honors my good friend and colleague Professor Sidney R. Ash of Weber State University of Ogden, Utah, in recognition of his numerous significant contributions to paleobotany. The list of species assigned to A shicaults and their synonyms are as follows: Ashicaulis amajolensis (Sharma) Tidwell, comb. nov. BasionyM: Osmundacaulis amayjolensis Sharma, sisal a aoe 140B:156. 1973. Millerocaulis amayjolensis (Sharma Tidwell, Sida 11:402. 1 Ashicaulis ae Go Tidwell, comb. nov. Basionym: Osmundacaulis fle Ghaige as Can. J. Bot. 56:3034. 1978. Millerocaulis beardmorensis (Schopf) Tidwell, Sida 11:402. Ashicaulis brogani Cee Munzing & Banks) Tidwell, comb. nov. Basionya: well, Munzing & Banks, Palacnatogsachicd 223B:98. 1991. hic seine (Sharima Done Sine) NOweNe comb. nov. Basionyn: ee & Singh, Phytomorphology 8:61. 1979 (“estipu- lare”), Milly ears Sama Bohra & Singh) Tidwell, Sida 11:403. 1987. Ashicaulis iana (Kidston & Gwynne- Vaughan) Ti well, comb. nov. BASIONYM: ee gibbiana Kidston & Gwynne- Vaughan, Trans. Roy. Soc. Edinburgh 45:763. 1907. Osmundacaulis gibbiana (Kidston & Gwynne- ea hams: Contr. Mus. Paleontol. Univ. Michigan {21:146. 1967, nom. invalid. under Art. 33.2 — no page reference to basionym} 23 : - 1971. Millerocaults gibbiana ee & Gwynne- Vaughan) Tidwell, Sida 11:403. 1 Ashicaulis guptal Sica Tidwell, comb. nov. Basionym: Osmundacaulis guptal harma, Palaeontographica 140B:154. 1973. Millerocaulis guptai (Sharma) Tidwell, Sida tile Tan 1987. Ashicaulis hebeiensis (Wang) Tidwell, comb. nov. Basionym: Osmundacaulis hebeiensis Wang, Rev. Palaeobot. Palynol. 39:93. 1983. Millerocaulis hebeiensis (Wang) Tidwell, Sida 11:403. 1987. Ashicaulis herbstii (Archangelsky & de la Sota) Tidwell, comb. nov. Basion Osmundites ji Archangelsky & de la Sota, Ameghiniana 3:135. 1963. ee herbstti (Arc gon & de la Sota) Miller, Contr. Mus. Paleontol. Univ. Michigan (21:14 1967, nom. invalid. under Art. 33.2 — no page reference to basionym) 23:134. 1971. Millerocaulis herbstii (Archange disley & de la Sota) Tidwell, Sida 11:403. 1987. Ashicaulis johnstonii (Tidwell, Munzing & Banks) Tidwell, comb. nov. BasionyM: Millerocaulis johnstonii Tidwell, Munzing & Banks, Palaeontogra ne 223B:94. LOD 1., es kidstonti Seay) Tidwell, comb. nov. BasionyM: Osmundites kidstonii Stopes, Ann. Bot. 35:55 21 (kidstoni’). Osmundacaulis kidstonii (Stopes) Miller, Contr. Mus. Paleontol. Univ. Mic ones :146. 1967 (“kidston?”), nom. invalid. under Art. 33.2 — no page reference to basionya) 23:136. 1971. Millerocaulis kidstonii (Stopes) Tid well, Sida 11:403. 1987 Ashicaulis kolbei (Seward) Tidwell, comb. nov. Basionym: Osmundites kolbei Seward, Geol. Mag., NS. 4:482. 1907. Osmumdacaliskolbe (Seward) Miller, Contr. Mus. Paleontol. Univ. Michigan (21:146. 1967, nom invalid. under Art. 33.2 — no page reference to basionym) 23:136. 1971. ee Sida ie 403. 1987. psmicauls ee (Wu & Shao-lin) Tidwell, comb. nov. Basiony: s Wu & Shao-lin, Acta Palaeontol. Sinca 30:717. 1991. TIDWELL, Ashicaulis, a new genus in the Osmundaceae 257 eine patagonica (Archangelsky & de la Sota) Tidwell, comb. nov. BASIONYM: Osmundites patagonica Archangelsky & de la Sota, Aimee biiaiana 25a F962. Oma tag (Archangelsky & de la Sota) Miller, Contr. Mus. Paleotol. Univ. Michig 1:146. 1967, nom. invalid. under sionyml 25 136.1 11:403. ca rajmahalensis (Gupta) Tidwell, comb. nov. Basionym: Osmundites aie et Proc. Ind. Sci. Congr. Varanasi 55: a Le OTA Tap alate 2s Gupta, Palaeontographica 130B:174. 1970. Os Pa laconcgraphic 140B:152. 1973. Millerocaulis vapilabalensis (Gupta) Tidwell, er 11:403. 1 reer Ean (Tidwell) Tidwell, comb. a Basionym: Millerocanlis richmondti Tidwell, Pap. & Proc. Roy. Soc. tama) 26:12. 1992. A shica is sahnii Dele Tidwell, comb. Aov. Baoan Osmundites sabnii Mittre, Paleobotanist 4:113. Art. 33.2 — no page reference to ba- 1971. Millerocaulis patagonica (Archangelsky & de la Sota) Tidwell, Sida Miller, Contr. Mus. Palaeontol. Univ. Michigan [21: i 1967, nom. invalid. under Art. 33.2 — no page oo to basionym] 23:135. 1971. Millerocaulis sabnii (Mittre) Tidwell, Sida 11:403, Ashicaulis santaecrusis (Herbst) Tidwell, comb. nov. Basiony: ee Santaecrusis Herbst, Facena 1:21. 1977. Ashicaulis spinksii (Tidwell, Munzing & Banks) Tidwell, comb. nov. Ba- SIONYM: Millerocaulis spinksii Tidwell, Munzing & Banks, Palaeontographica 223B:96, L991 Ashicaulis swanensis (Tidwell, Munzing & Banks) Tidwell, comb. nov. BasionyM: M7/lerocaulis swanensis Tidwell, Munzing & Banks, Palaeontographica 223B:99, 1991. Ashicaulis wadei (Tidwell & Rushforth) Tidwell, comb. nov. Basiony: Os- mundacaulis wadet Tidwell & Rushforth, Bull. Torrey ae - ub 97:137. 1970. Millerocaulis wadei (Tidwell & Rushforth), Tidwell, Sida 11:403. 1 Ashicaulis websterii (Tidwell, Munzing & ee Tidwell, comb. nov. Ba- SIONYM: Millerocaulis websterii Tidwell, Munzing & Banks, P: alae oer aue flit 223: 97. 1991. shicaulis wrightii (Tidwell, Munzing & Banks) Tidwell, comb. nov. Ba- SIONYM: Millerocaulis wrightii Tidwell, Munzing & Banks, Palaeontographica 223B:93. 1991 DISCUSSION Hill et al. (1989) considered Australosmunda indentata the first fossil osmun- daceous rhizome described with a simple siphonostele and a parenchymatous pith without realizing that the holotype of Mi/lerocaulis, M. dunlopii, also has a simple siphonostele. The reason is the considerable confusion surrounding the presence or absence of leaf gaps in M. dun/opii. As pointed out by Kidston and Gwynne- Vaughan (1907) in describing the holotype, “The most important anatomical character of this species is the almost complete absence of leaf gaps in the xylem ring of the stem. The fact that most of the leaf traces, if not all depart without in any way interrupting the continuity of the xylem ring, so that the “medullary rays” characteristic of the osmundaceous stele in general are almost or completely absent. Most of these breaks in the xylem ring are clearly 258 Sipa 16(2) 1994 due to accident, but it is just possible that some of them may have been occu- pied by thin-walled cells which decayed before fossilization. Even if such med- ullary rays actually were present in the living plant, they must have been ex- tremely narrow and very rare. and reference to the longitudinal sections will make it clear that no medullary ray is caused by the departure of the xylem of the leaf-trace and that the continuity of the deeper portion of the xylem ring is undisturbed.” (p. 760-761). Thus, this species lacks leaf gaps. Sinnott (1910), however, attributed the apparent lack of leaf gaps in the holotype of M. dunlopii to its having very short and narrow gaps that were oblit- erated by the process of fossilization, an intriguing explanation but without any real evidence. It is interesting to note that Sinnott also illustrated a diverging leaf trace for Osmunda regalis in an “apparently gapless fashion” and one for O. cinnamomea and others for Todea superba and T. hymenophylloides that had not formed complete gaps. The lack of leaf gaps in extant Osmundaceae is rare. Miller (pers. comm. 1993) mentioned that stems of modern osmundaceous taxa may be with- out complete leaf gaps near branches. Several osmundaceous specimens with leaf gaps have been reported as Osmun- dites (= Millerocanlis) dunlopii from the Jurassic of New Zealand (Sinnott 1914) and the Jurassic near Purga, Queensland (Sahni 1920). However, none appear to 7 «6 be that species. One of Sinnott’s specimens, for example, is most likely Ashicau- lis wrightii described from Tasmania (Tidwell et al. 1991). His other specimens also belong to Ashicaulis but are too poorly preserved to be identifiable to spe- cies. Of the five leaf gaps Sahni (1920) noted in the transverse section of the stem of one of the four incomplete specimens he attributed to 0.(=M.) dunlopi, two very narrow gaps may be valid, but even these, could be due to fracturing prior to preservation. None of his alleged gaps have traces opposite them. Sahni further stated, “In the only longitudinal section that was made, three leaf-traces are seen arising, but while it is impossible to say whether all of these have been cut medially, in no case is the cauline strand interrupted opposite an off-coming leaf trace” (p.13). Furthermore, it is doubtful if the specimen (#140) Sahni (1920) illustrated as text — fig. 2 is M. dunlopii, but more likely represents a species of Ashicaulis. He founded much of his identification on the curved leaf traces, scle- rotic rings, and “the two large sclerotic strands inside the curved ends of the leaf-trace, which is characteristic of O. dvn/opit.” He used these criteria based on the limited knowledge of the Osmundaceae in Australia at the time. It is now known that the first two characters occur in essentially all the extant and extinct species of the Osmundaceae (Hewitson 1962; Miller 1967, 1971) and the last feature in nearly all described species of Mi/erocaulis and Ashicanlis trom that continent (Tidwell et al. 1991). The distribution of sclerenchyma in the stipular wings, which has proven to be useful in species determination (Hewitson 1962; Miller 1967), was largely ignored in the specimens reported by Sahni (1920) and Sinnott (1910, 1914) as TIDWELL, Ashicaulis, a new genus in the Osmundaceae LS), Osmundites(=M.) dunlopi. The distribution in the wings of one petiole figured by Sahni (pl. I, fig.4) was noted to be like that described for M. dunlopii, but that particular petiole was not attached to any stem. The stem Sahni (1920) figured (pl. I, fig.3) had attached petiole bases, but none was figured or described. Con- sequently, it is doubtful whether any were really M. dunlopii. Edwards (1933) pointed out in discussing his fossil osmundaceous specimen from central Australia that it had a continuous or almost continuous xylem cylinder that was not broken into separate strands similar to the Jurassic species of Osmundites (= Millerocaulis) dunlopii. Medwell (1954), Sharma (1973), and Andrews (1961) all stated or implied that 0. (=M.) dun/opii possessed a continu- ous xylem ring, meaning no leaf gaps. Miller (1967, 1971) in his papers on the Osmundaceae accepted the fossils of Sinnott (1910, 1914) and Sahni (1920) as specimens of 0. (=M.) dunlopii and concluded that the species had very narrow, some immediate, some delayed and some incomplete leaf gaps. In addition, he mentioned that the near absence of gaps in the type specimen of this species is not typical for the species. More recently, however, Miller (pers. comm. 1993) commented that in reviewing his notes he had not observed any evidence of leaf gaps when examining slides (#1242—1247) of the holotype of M. dunlopii from Gore (Otago), New Zealand at the British Museum of Natural History. He further stated that the holotype of M. dunlopii does not have leaf gaps, and that there was never any indication of the short narrow leaf gaps that might have been obliterated in the process of fossilization as per Sinnott’s (1910) explanation. Furthermore, Miller noted that Kidston and Gwynne-Vaughan’s (1907) photos (pl. I, fig. 3; pl. II, figs. 4-5) do not show any ‘obliteration’ of the xylem tracheids either. In conclusion, the only definite specimen of M. dunlopii known, is its holotype, and it lacks leaf gaps. Consequently, Australosmunda indentata is reassigned to Millerocaulis. Millerocaulis dunlopii, the type, and M. indica, which also lacks leaf gaps (Sharma 1973; Fig 1G), are retained in this genus. Based on photographs of Osmundacaulis ae sis from Argentina (Herbst 1977), this species also appears to be without leaf gaps (Figs. 1H, I, J, K) and is placed in M/erocaulzs as well. Millerocaulis, as now defined, with a Triassic species (M. indentata) and three Jurassic species illustrates her phylogenetic line in the Osmundaceae that is presently known only from the Southern Hemisphere. Osmundaceous taxa of this phylogenetic line in lacking gaps and having parenchymatous piths lend support to the hypothesis that evolution of the parenchymatous pith and evolu- tion of the leaf gap in the xylem cylinder were independent (Hill et al. 1989). The other species previously assigned to Mz//erocaulis have numerous definite leaf gaps are hereby transferred to the new genus Ashicaulis. Ashicaulis, presently composed of 22 species, is known from both the Northern and Southern Hemi- spheres and is currently known from the Triassic to Early Cretaceous. 260 Sipa 16(2) 1994 ACKNOWLEDGMENTS Tam grateful to Dr. Dan H. Nicolson of the Smithsonian Institution, Dr Charles N. Miller, Jr. of the University of Montana, Dr. Alfred Traverse of Penn State University, and Dr. Sidney R. Ash of Weber State University for their assistance with nomenclatural problems as well as reviewing earlier versions of this paper. REFERENCES ANbrews, H.N. Jr. 1961. Studies in peleono ey ee and Sons, New York. 1962. ARCHANGELSKY, S. and E.R. De La Soro. a [Pena Ui estipite petrificado au “Osmundites,” de edad Jurasica, procedente del Gran Bajo de San Juli ian, Provincia de Santa ruz. Ameghiniana 2:153—164. a 1963. Osmundites erbstii, nueve petrificacion Triasica de el Tranquilo, Provincia de Santa Cruz. Ameghiniana 3:135—140 ios W N,.1933: Osmundites from central Australia. Ann. Mag. Nat. His. 11:661—663. Erasmus, T. 1978. The anatomy and evolution of Osmundacaulis Miller emend. with notes on the geometry of the xylem framework of the osmundaceous stele. University of Pretoria unpublished Ph.D. dissertation. Gupta, K.M. 1970. Investigations on the Jurassic flora of Rajmahal Hills 8. On a new species of Osmundites, O. rajmahalense sp. nov. from the Rajmahal Hills, Bihar. Palaeontographica 130B: 173-180 Hersst, R. 1977. om nuevas especies de Osmundacaulis (Osmundaceae, Filices) y ostros restos de Osmundales de Argentina. Facena 1:19-44. Hersst, R. 1981. Guairea milleri nov. gen. et sp. y Guaireaceae, nueva Familia de las Osmundales (sensulato) del Permico superior de Paraguay. Ameghiniana 18:35—5O Hewrrson, W.H. 1962. Comparative morphology of the Osmundaceae. Ann. Missouri Bot. Gard. 49:59-93. Hut, R.S.,S.M. Forsyt and F. Green. 1989. A new genus of osmundaceous stem from the eee ee of Tasmania. Palaeontology 32:287-296. a R., and D.T. GwyNNE Bh IGHAN. 1907. On the fossil Osmundaceae, Part I. Trans. . Soc. Fdlinbpreh 45:759-78 = iALL, P. 1924. A new species mie from Kawhai, New Zealand. Trans. & Proc. New Zealand Inst. 56:210—213 we DWELL, L.M. 1954. A review aad revision of the flora of the Victorian Lower Jurassic. Proc. Roy. Soc. Victoria 65: Mier, C.N. 1967. ee of the fern genus Osmunda. Contr. Mus. Paleontol., Univ. Michi- gan 21:139-203. Mutter, C.N. 1971. Evolution of the fern family Osmundaceae based on anatomical studies. Contri. Mus. Paleontol., Univ. Michigan 23:105—169 ee, V. 1955. Osmuniites sabnit sp. nov., anew species of petrified osmundaceous rhizomes m India. Palaeobotanist 4:113-118. es B. 1920. Petrified plant remains os the Queensland Mesozoic and Tertiary forma- tions. Queensland Geol Sur. Publ. 267:3-4 ScHopr, J.M. 1978. An unusual at ie specimen from Antarctica. Canad. J. Bot. ) lect SHARMA, B.D. 1973. Anatomy hi 1 from the Middle Jurassic of Amajola in the Rajmahal Hills, India. Palaeontographica 140B:150-160. TIDWELL, Ashicaulis, a new genus in the Osmundaceae 261 ,D. Bonra, and R. SINGH. 1979. Osmundacaulis estipulare ve nov. a rhi- micron Jukessie of Rajmahal Hills, India. eae 29:4 ene E.W. 1910. Foliar gaps in the Osmundaceae. Ann. Bot seen a 107-118. StTopes, M.C. 1921. The missing link in ait fon Bot. (Oxford) 35:55-61. TIDWELL, W.D. 1986. Millerocaulis, a new genus with species formerly in 0. lis Miller fossils: Osmundaceae). Sida 11:401—405. TipweLl, W.D. 1992. Millerocaulis richmondit sp. nov., an osmundaceous fern from Mesozoic strata near Little Swanport, Tasmania. Pap. & Proc. Roy. Soc. Tasmania 126:1-7. and $.R. RUSHFORTH. 1970. Osmundacaulis wade, a new osmundaceous species form the Morrison Formation (Jurassic) of Utah. Bull. Torrey Bot. Club 97:137-1 144. ________, G.E. Munzinc and M.R. Banks. 1991. Mi/lerocaulis species (O 1 from Tiana Australia. Pal hi 2238 91-105. Wana, Z. 1983. Osmundacaulis ee is, a new species of fossil rhizomes from the Middle as of China. Rev. Palaeobot. and Palynol. 39:87—107. and Z. SHAO-LIN. 1991. A new species of 7 eal rhizome from Middle Jurassic see China. Acta Palaeontol. Sin. 3:7 1¢ 262 Stipa 16(2) 1994 BOOK NOTICES Everitt, JAMES H. and D. Lynn Drawe. 1993. Trees, Shrubs & Cacti of South Texas. Texas Tech University Press, Lubbock, TX 79409-1037 (1-800-832-4042). ISBN 0-89672-252X; ISBN 0-89672-253-8 (pbk). $18.95 pbk. Text of colored photographs. ntents include the preface, introduction with colored Meee map, text with 204 a photograpt ace seas we glossary, list of vertebrates, and in he text “contains a color photograph, the family name, sc ane name, common name, general description, Ree range, and cee characteristics for each Species: of the majority of trees, shrubs, and cacti occurring in the 14 southernmost counties of Texas The photographs are good and the photos are lance ak mostly two per page to provide a nice aspect. Only one word was found to be mispelled in browsing through the book (“Baccharts,” p. 47) This is an exellent pictorial book for the south Texas area and should be on your bookshelf or with you daily—Wm. F. Mahler. Burt, Pace. 1991. Barrenland Beauties. (ISBN 0-919315-25-9, pbk). Outcrop Ltd., The Northern Publishers, Box 1350, Yellowknife, N.W.T., Canada X1A 2N9. $25.00 (Canadian; includes postage). 246 pp. This is a guide to some 100 species of “showy plants” of the Canadian arctic coast. The color photos—about 200 of them, a few full-page eee most eh grouped two to 9 four per page excellent. Notes on habitat, ecology, and n Inuinaktun, the written form of the dialect commonly spoken in the central arctic). Described, too, are the setting, history, and botany of seven major coastal communities. The book has a glossary, a bibliography, and an index to common and scientific names. Page Burt is a naturalist at Bathurst Inlet Lodge, to which every summer come guests to learn about arctic natural history (I recom- mend the lodge highly). Although her fine book was written to fill the need for a guide to the marvelous, low-arctic flora of the inlet, its usefulness extends from there to Ellesmere Island. Enjoy this book in your living room (bonus: no mosquitoes); it is the next-best thing to going north.—John W. Thieret. NEW AND NOTEWORTHY MALESIAN MYRSINACEAE, VIII' BENJAMIN C. STONE’ Botanical Research Institute of Texas 509 Pecan Street Fort Worth, TX 76102-40060, U.S.A. ABSTRACT New taxa of paleotropical Myrsinaceae are described from the Philippines and New Guinea in the genera Ardisia, Fittingia, and Tapeinosperma. Ardisia anceps var. borneensis is shown to be a species of Theaceae, probably Adinandra, which is not a myrsinaceous plant. The subge- neric position of Ardisia mystica is established. Ardisia brassii Stone (non Sleumer) is renamed A. brassiella. RESUMEN Se describen taxa de Myrsz | icales, que peteanene a los = BenCTOS Ardisia, ph y Tapeinosperma, procedentes de i PAlisinas yk Ardisia var. borneensis es en realidad, una ayeG de la familia Theaceae, ee un ene ns género Ardinandra. Se ubic subgénero Pimelandra. Ardisia brassi Stone Sl ) se renombra como eae ba Pane INTRODUCTION This paper is a continuation of a series of studies of paleotropical Myrsinaceae from the Malesian region. Earlier papers in the series are included in the refer- ences. This paper is also Contribution no. 2 in the Precursors series for the Flora of the Philippines project. In the course of revising the Myrsinaceae in the Malesian region and in col- lecting intensively in the Philippines under the Philippine Plant Inventory project (PPI), new taxa, new biological and distributional data necessitate taxonomic and nomenclatural changes. New taxa are described in Ardisia, Fittingia, and Tapeinosperma. Also, one purported variety of Ardisia is in fact theaceous and some further notes on species of Ardisza are included. A new species of Ardisza Swartz Ardisia reynosoi B.C. Stone, sp. nov. (Fig. 1). Frutex glaber humilis usque ad 150 cm altus, ramulis 2.5—3 mm diametro adscendentibus, ‘Flora of the Philippines Project, Precursors, Contribution no. 2. Preceding paper: see Pacific Sci. 47(3): 276-294, 1993. This study was partly supported by NSF Grant DEB-9007742. +Benjamin C. Stone died 19 March 1994 in Manila, Philippines. Siwa 16(2): 263 — 272. 1994 264 Sipa 16(2) 1994 £9168 OSONAITY & LNIDY Fic. 1. A Stone (drawn from holotype). A. Leafy flowering branchlet. B. Single flower. C. Decne calyx (2 lobes cut off), adaxial view. D. Corolla lobe, interior view; glands ament. E. Stamen, abaxial view, with glandu- — toward apex, papillae above small staminal attrac lar connective. F. Ovary and style. G. Placenta, the lower figure ue enlarged to show rows of ovules. J. Trichomes from calyx margin. UO a apicem Vv iose foliatis. Folia congestae, parvae, petiolis 3—6 (raro ad 15) mm longis, laminis eee cbouatie subrigide coriaceis, ad 45(— ais ee 17.5(-25) mm latis, apice subacuti vel obtusi vel rocundati, marginibus integris rev ‘osta infra elevato; venis secundariis 6—9-paribus infra parce elevatis, tertiariis sabinaniectia vel subobscuris; glandulis minime pustulatis a sis. Inflorescentia axillaris, brevis, 2—3-floriferis, arin gece 11 (-20) mm longis, glabris, alabastris ideis 2—3 mm longis; calycis lobis 1 —1.2(—-1.8) mm longis subsotundatis le punctatis, marginibus minime ciltatis, basi perbreviter imbricatis; coroll lobivovatls bolas ad 4 4mm mes glabris, Braneblosts, intus basin versus > sparse papillosi filamentatis, antheris deus: acniis, connectivo dorso bilinedliter glandulosis, pleats indivisis; ovario hemiovoideo, 1 mm alto, in stylo 1.5-2(-4) mm longo producto, stigmate acuto punctiforme; placenta hemiovoideo, ovulis ca. 35 tetraseriatis. Fructus (PPI 6925) oblatus 5 X 7 mm ultime atropurpureus. en PHILIPPINES. Sipuyan (Romblon Province): Magdiwang, western extremity of Mt. Giting-giting, ridge to Mayos Peak, steep slope in open grassy area at 1400 m alt., 29 Aug 1989, G. Argent & E. Reynoso 89-163 (HoLoTyPE: PNH; isotype: E). Paratypes: SIBUYAN: ridge to Mayos Peak, 1300 m alt., submontane for- est, 24 Aug 1989, Argent & Reynoso 89-87 (E, PNH); same locality, 1400 m alt., dwarf shrub, 5 Jun 1992, B.C. Stone & E. Reynoso PPI 6925 (L, PNH, US). STONE, Malesian Myrsinaceae 265 Etymology: Named for the collector Ernesto Reynoso wh tensi ence in the Philippines qualified him as a team leader in the ice Plant Inventory Project. Many botanists and other collectors have been guided by “Estong” over the years, so this dedication is well deserved. Notes: This small, shrubby Ardisza occurs in elfin forest and partly open areas on the steep upper ridges of Mt. Giting-giting; so far it is only known from there and is apparently endemic. The side-branches (usually 15—20 cm long) are ascending and bear small crowded ascending leaves. These may reach 8 cm but are usually shorter; the midrib beneath is tawny brown. The inflorescences are axillary and terminal, umbelliform, with umbels usually few (2-3) flowered; in fruit the calyx lobes attain a length of almost 2 mm and are slightly imbricate at base. Ripe fruits are dark purple, conspicuously obloid, and are about 5 mm long and 7 mm wide. In general appearance this species is very similar to A. gitingensis Elmer. The infrageneric position of Ardista mystica Ardisia mystica Stone, Malayan Forester 45:114, fig. 5. 1982. This species was described from imperfect material and there was doubt as to which of the subgenera it pertains to. After examining type material of Ardisia chrysophyllifolia King & Gamble (K) it was evident that the two species were closely related and that A.mystica had to belong in the same subgenus. Both in fact belong to Subg. Pimelandra. The ovary of Ardisia mystica is minutely tomentellous and the flowering pedicels are about 10 mm long; while in A. chrysophyllifolia the ovary is glabrous and the flowering pedicels are 5 mm long. The identity of Ardista anceps var. borneensts Scheff. Ardisia anceps var. borneenis Scheff., Myrs. Archip. Ind. 79. 1865. “Folia sessilia denticulata longiuscule abrupte acuminata, in sicco rubescentia.” (Fig. 2). The type specimen bearing this name (Korthals s.n. from Gunong Rantau, South Borneo, in L!) is not myrsinaceous, but appears to be theaceous, propa a species of Adinandra. It has acicular pellucid unicellular trichomes; the crenate-denticulate leaf margins with the distinct teeth are extensions of the veinlets. The habit of the leafy branchlets somewhat resembles an Ardisia with slight ridges (alulae) as prolongations of the petiole bases, and a pattern of glands on the leaf lamina. In several herbria, specimens of Adinandra and other similar genera of Theaceae have been found lodged among undetermined material of Myrsinaceae, showing that there is a possibility of confusion among specimens of these families. The trichomes of Myrsinaceae are often ferrugineous or brown, seldom pale tawny, and the species with denticulate leaves have Sipa 16(2) 1994 LAL SSE SS Z LLC sah MLL LE LE LD SSNS LE k 0.4 mm 09-S&Z'E06 PIH (O3NdOd) STVHLYOM Fic. 2. “Ardisia hac aa var. borneensts Scheffer” = Theaceae sp. Details of the type specimen. A. w (showing pair of ridges on stem). B—D. Views of leaf ith strigose midrib, laminar glands, -like tri- oe Stem with lea n undersurface co with Lee part tof middle, a nd apex) w and margin margin with spinulose crenations, shown enlarged. F. Spinule chomes from an STONE, Malesian Myrsinaceae 267 more delicate teeth with a different relationship to the veins. Scheffer him- self was doubtful as to the affinity of this specimen, stating “An Ardisia? Nec fructiferam nec floriferam vidi.—Rami ancipites sunt.—Myrsinea ex Korthals.” A new name for Ardista brassti Stone — eens C. Stone, nom. nov. A. brassii Stone, Proc. Natl. Acad. Sci. U.S.A. 1990 (non Sleumer, Blumea 33:130. 1988). Type: PAPUA NEW GUINEA: ee ce Mt. Riu, W slopes, 300 m, 31 Aug 1956 (Al, fr), Brass 27954 (HOLOTYPE: US). The name A. brassiz Stone was inadvertently used in my 1990 study for the species represented by Brass 27954 and 27825, an error for A. brassiella which was intended, as Sleumer’s earlier use of A. brassiz (for a New Guinea species) preempts that epithet. Fittingia Mez This genus is endemic to New Guinea. The following new species is the sixth known; Sleumer (1988) recognized five species. Fittingia mariae B.C. Stone, sp. nov. (Figs. 3, 4). Fruticulus tenuis, stipite erecto ad 4 mm diam., atrobrunneo, glabro, apicem versus foliatis, ramulis lateralibus tenuioribus 1-2 mm diam., 10-15 cm longis, tereribus, apice excepto hau foliatis, novellis anes vel sparsiter lepidotis. Folia anguste elliptico- lanceolatae, usque ad i cm aon 4cm latae, integerrimae, chartacea, sae ue sensim acuminato-acutatae, infra ins 0 leviter reticulatae; petiolis 10-15 mm By longis, subteretibus, supra canalicularis: ae marginatis; costa media supra anguste 60 impresso, infra bene elevato basin verus c nato; venis ane 10—14-paribus, ad ca. emittis, len Balas distaliter antrorse curvtis et ete neies rete nervarum tenue submanifesto. Inflorescentiae in axillas foliorum ramulorum, breviter angusteque paucipaniculatae ad she cm longae, axibus omnibus Somer et sparsicer lepidoculis, pedunculis brevibus singularibus vel paucifasciculatis, ramis ad 10-12 mm longis, suboppositis, bracteis subulatis glandulosis 1-2 mm longis, pedicellis ca. 2mm longis, cymis 1—3-floriferis, floribus ca. 2 mm ong! is, 4-meris, Geena omnino unisexualibus (masculis solis hic descriptis), calyce gl eee x0, lobis ovatis integerrimis multi landulosis oo g (glandulis i j ] it ey g 7 mm lon s, basi ite (0. 5 mm) eoalitle, t i landulosis, 1 ino d pa illo Osis; ae subieteroisc ca. 0.5 mm longi, e basi lobi corollae are antheris oe oblongis fere 1 ngis api 2 natis, pallidis, eglandulosis; gynoecio rudimentale stylo 1.1 ngo, stigmate capitato oe Ovario vedic 0.3 mm longo quam stylo Vix latiore, mm lon ovulis nullis. =: NEW GUINEA. Morose Province: Boana, 1,219 m in mossy montane scrub, a ne shrublet, flowers dull pinkish, 4 Jul 1940, M.S. Clemens 4155 1-A (HoLoTYPE: MICH). P pe: NEW GUINEA. Monrose Province: Boana, 1,219 m, 4 Jul 1940, M.S. Clemens 41550 (MICH). Etymology: Fittingia mariae is named for the indefatigable collector Mary 268 Siva 16(2) 1994 Pe OF U0 Fa SSlt SNIW312 SW Fic. 3. Fittingia mariae Stone (drawn from holotype). A—D. Floral details. A. flower in profile with pedicel and part of partial inflorescence with bracts. B. Interior view of part of corolla, showing densely papillate inner faces of corolla-lobes, - short glabrous stamens with emar- ginate anthers. C. Stamen, abaxial view; note the non-glandular connective. D. Pistillode with discoid stigma. E. Diagram of inflorescence. F. Leaf. STONE, Malesian Myrsinaceae ‘ 269 Fic. 4. Fittingia mariae Stone. P I jotograph of holotype. 270 Sipa 16(2) 1994 249508 i Collected by L. J, Himsa on the ath {March-Nowember Fic. 5. Tapetnosperma filipes Stone. Photograph of holotype. STONE, Malesian Myrsinaceae 271 Strong Clemens who with her husband Joseph, and later alone, obtained numerous plant collections in the Philippines, Borneo, and New Guinea. Notes: The dioecious habit, 4-merous flowers, and axillary inflorescences indicate that this species belongs to the genus Fettimgia. Among other gen- era of the family that occur in New Guinea only Loheria could be consid- ered as a possible alternative, but that has short lateral shoots with flowering panicles subtended by a subinvolucral group of bracts. In the absence of pistillate flowers, some doubt may remain, but the general habit and slen- der panicles, sharply acuminate leaves and low habit suggest Fttingia much more than Loheria. In Fittingia also the corolla seems to lack internal papillosity, while it is common in Loheria. Among the species of Fittingia, F. mariae may be readily distinguished by the slender, sharply acuminate leaves and very slender panicles. Collec- tion of pistillate materials including fruits remains a desideratum. Tapeinosperma Hook.f. Sleumer (1988) lists four species of Tapeznosperma in New Guinea. Among the L.J. Brass collections the following new species has been identified. Tapeinosperma filipes B.C. Stone, sp. nov. (Fig. 5). Arbuscula plerumque monocaulosa usque ad eo be m alta; caulis apicem versus dee ad 6 mm diametro; foliis spiraliter confertis; pergracilil illaribus 1—3-foliatis elongatis ceretibus a aes cm tones linosenones fe ineo-tomen i piso. 0. 05— ; ms ] 0.06 mm] 34-plo eee Rolie near: innceolees 208 32 cm longae, 2—2.2 cm latae utrinque acutae, basi sensim decurrente attenuato, chartaceae, integerrimae, ineras sat epaieior, supra glabra, infra glabrata (costa media excepto); costa supra basi ang , infra prominente; venis lateralibus prominulentibus ca ae paribus venis a ntersecencatils subacqualibus intercalatis, oa in pag manifestis; dispersis 1 in areolis somnib pees convexus, gracilis, 10- 12mm Donieas, primo minute rovagntellag demum glabratus. Ramuli fertili tereti laevi 1-2.5 mm diametro, += 26 cm ng (usque ad reliant pean, foliis reductis 7-11 cm longis subulatis; axibus, comentellis. Inflorescentia terminalis. Flores minute remoti si balobesi. ssedicdll is 2.5—3 mm longis; calyce cupulato 5-lobato, basi | sparsiter hirtello, lobis ovato-deltatis 0. 61 mm ee late ovato-subrotundatis, ec glabris 5-venatis, glan- | 00. duloso-punctatis, integris. Stamina ca. 1.3 mm longa, filam eu longo, antheris oblongo- ovatis, subacutis, ie sublaeve vix glanduloso vel glanduli ris singulis in connectivo positis. Ovarium ca. 0.5 mm longum, glabrum, sensim in tee 0.7 mm awae re stigmate minute truncato obscuriter discoideo et minime papilloso. Placenta ca. 0.4 mm alta, ovulis 2 vel 3, terminalibus. Type: PAPUA NEW GUINEA. MILNE Bay District: Maneau Range, Mt. Dayman, 700 m ., rainforest, shrub with pinkish flowers and fruit, 14-24 Jul 1953, Leonard J. Brass 23403 (HOLOTYPE: US) ia o Known only from this collection from lower montane rainforest. Etymology: The epithet, filipes, refers to the very slender, almost thread- 272 Sipa 16(2) 1994 like inflorescence axes (fila = thread; pes = foot, stalk). Notes: This distinctive species was cited as an indeterminate Ardisia by Sleumer (1988: 103), but the discoid-truncate stigma and strongly reduced ovular number exclude it from that genus. The flowers are very small but seem to be hermaphroditic, and in this respect conform to the condition in Tapeimosperma. The very long slender leaves and extremely slender open panicles borne on long, slender axillary branchlets bearing 1 or 2 reduced leaves are further characters of this species. REFERENCES SLEUMER, H. 1988. The genera Discocalyx Mez, Fittingia Mez, Loberia Merr. and Tapeinosperma Hook.f. (Myrsinaceae) in New Guinea. Blumea 33:81— . STONE, es C. 1989. Myrsinaceae. In: ae FS.P. (ed). Tree flora of Malaya. Vol. 4:64-84. STONE, B.C. 1990. Studies in Malesian Myrsinaceae V. Additional new species of Ardisia Sw. Proc. Acad. Nat. Sci. Philadelphia L- 42:21-58. A NEW SPECIES, SY MPHORICARPOS GUATEMALENSIS (CAPRIFOLIACEAE), FROM GUATEMALA WITH A KEY TO THE MEXICAN AND GUATEMALAN SPECIES JUSTIN KIRK WILLIAMS Department of Botany University of Texas Austin, TX 78713, U.S.A. ABSTRACT Recent studies of Mexican and Central American Symphoricarpos have revealed the existence of a new species, described and illustrated below, from the Cuchumatanes Highlands of Guate- mala. In addition, keys and distribution maps for the species of Symphoricarpos from Mexico and Guatemala are provided. RESUMEN 10 eae Ee Pe DP a ener Ans eee : . 1 13-1 1 pos Mexicanos y de una nueva especie, q lescribe e ilustra a continuacién, procedente de los Cuchumatanes de Guatemala. Ademas, se ofrecen cl y mapas de distribucién de | ee a eeiry eee de México y Guatemala. Key Worps: Symphoricarpos, Caprifoliaceae, Guatemala, Mexico. Symphoricarpos guatemalensis J.K. Williams, sp. nov. (Fig. 1) Symphoricarpt microphyllo ABK. similis sed differt f -oroll glabra et floribus ac foliis dimidia parte brevioribus. Erect shrub, 2 m tall, much-branched, young twigs strigose. Leaves oval with rounded apices, tapering toward base, 3-6 mm long, 2—3 mm wide, dark green above, margins entire, glabrous to slightly villous, coriaceous; petioles up to 1 mm long. Flowers solitary or in pairs; pedicels up to 0.5 mm long, with two apical bracteoles 2-3 mm wide; calyx glabrous, 1-2 mm long, irregularly 5-toothed; corollas white, narrowly campanulate, actinomorphic, papillate to- wards base within, 4-6 mm long, tubes 3—5 mm, lobes 1—2 mm, erect; stamens 5, epipetalous, barely exserted, anthers 1.0—1.5 mm long, ca. half the length of filaments; style glabrous, 3-4 mm long, stigma capitate. Fruit a globose berry, 6-7 (10) mm long, white, and tipped with persistent calyx. Distribution and Ecology: Limestone ridges of the Sierra de los Cuchuma- tanes in Guatemala (Fig. 2), 3000-3500 m; flowering July to September. Type: GUATEMALA. HUuEHUETENANGO: 3 mi SW of San Mateo, Ixtatan, 5 Aug 1965, D.E. Breedlove 11544 (HoLoryPe: TEX!; isorype: F!, MO). Sipa 16(2): 273 — 280. 1994 No ~J AN SIDA 16(2) 1994 G. 1. A-B. ae A. Habit. B. Flower. (Breedlove 11544). C. Symphoricarpos mich ih HLus flower Additional specimens examined: GUATEMALA. Huehuetenango: Sierra de los Cu- chumatanes, on and below road from Huehuetenango, & 4 W of San Juan Ixcoy, E of Caprtzin, 15° 35'N 91° 28'W, 11 Jan 1976, I//tis G-271 (F); between Paquix and San Juan fae 8 Jan 1974, Molina 30022 (F, MO); between Paquix and Llanos San Miguel, road to San Juan Ixcoy, 17 Nov 1967, Molina 21248 (F); dry slopes of El Mirador, 12-23 Jan 1966, Molina 16505 (F); along road to San Pedro Soloma, 3 mi SW of San Mateo mabe 5 Aug 1965, Breedlove 11544 (F); limestone ridge near Chemal, Sierra de los Cuchumata 29 Nov 1962, L.O. Williams 21979 (F); just — Calveras, 29 Nov 1962, L.O. os ee 22389 (F); Sierra de) os Cuchumatanes, at KM 311 on Ruta Nacional 9N (between Paquix and ae small limestone ridge with ry Standley, 2 Aug 1959, Beaman 2989 (T ear Tunima, Sierra de los Cuchumatanes, 6-7 Jul 1942, Steyermark 48304 (F); along a in region of Chemal, 31 Dec 1940, Standley 81624 (F); rocky ridges beneath pines, Sierra Cuchumatanes, 15 Sep 1934, Skutch 1253 (F). WILLIAMS, Symphoricarpos guatemalensis 21D 100 mi 91 90 Fic. 2. Distribution of S. gvatemalensis (hatched area) and S. microphyllus (squares) in Guatemala. This narrow endemic of the Sierra de los Cuchumatanes is a member of subgenus Anisanthus G.N. Jones (Jones 1940), as evidenced by the corolla obes shorter than the corolla tube and its included style. Symphoricarpos guatematensis is most similar to S. microphyllus but differs in its glabrous (vs. villous) corolla throat, dark brownish-green leaves (vs. light green), and leaves and flowers of half the dimension. The leaves of S. gwatemalensis are 3—6 mm long and 2—3 mm wide, compared to those of 8. microphyllus which are 10—25 mm long and 7-15 mm wide (Fig. 3). In addition, the corolla length of S. gwatemalensis is 4—6 mm, compared to 9-10 mm in S. microphyllus (Fig. 1). In his monograph, Jones (1940) reported that species of Symphoricarpos, other than S. mzcrophyllus, are rarely found in Mexico and further south. Accumulating collections from Mexico, however, have shown that the ge- nus is more prevalent there than once suspected. In order to place the new species in a broader perspective and to provide a preliminary taxonomic evaluation of the Mexican and Central America species, I have constructed a key to the species of Symphoricarpos in this area based on specimens housed 276 Sipa 16(2) 1994 at BRIT, E, LL, and TEX. Characters used to delineate taxa have been es- sentially adopted from Jones (1940) and Gray (1873). Symphoricarpos is divided into two subgenera: subg. Symphoricarpos, char- acterized by an exserted style and corolla lobes longer than the tube, and subg. Anisanthus, characterized by an included style and corolla lobes shorter than the tube. There is only one member of subg. Symphoricarpos orbiculatus, native to Mexico. All other species in Mexico and Central Ammecicn belong to subg. Anisanthus. KEY TO THE MEXICAN AND GUATEMALAN SPECIES OF SYMPHORICARPOS 1. Corolla lobes equal to or longer than tube; fruits red Leis ee 1. Corolla lobes shorter than tube; fruits w . Anthers sessile, style pilose or oe 1 basal nectary 2.8. inns es with distinct filaments, style always glabrous, 5 basal nectaries .............:00 3. Leaves 4-6 mm long, 2-3 mm wide; corolla 4—6 mm long, glabrous within .S. guatemalensis 3. Leaves 6—25 mm long, 3-18 mm wide; corolla 6-10 mm long, villous within 4 4. Shrubs erect, 2—3 m tall; mature leaves of older branches entire ... 4. 8. microphyllus 4, Shrubs trailing, branches arching, 1—1.5 m tall; mature leaves of older branches lobed to crenate 5 5. Leaves of flowering and non-flowering branches ly crenate with 5.8. palmeri yi 5. Leaves of flowering branches mostly entire with rounded to acute apices, leaves of non-flowering branches lobed with rounded to mu- cronate apices 6.8. parishii AN OVERVIEW OF THE MEXICAN AND CENTRAL AMERICAN SPECIES Symphoricarpos orbiculatus Moench This is the only member of the subg. Symphoricarpos native to Mexico; other species of the subgenus occur throughout North America and Asia. This par- ticular species is unique to the genus in having red fruits (Fig. 5). Representative specimens: MEXICO. Coahuila: Muzquiz, along Sabine River, 11 Jul 1936, Marsh 475 (TEX). Nuevo Leon: Monterey, Diente Canyon, 20 Jul 1930, Mueller 103 (FE, TEX). Symphoricarpos longiflorus A. Gray This species is distinguished by its salverform corollas 11—13 mm long, the largest of any corolla among the Symphoricarpos of Mexico. Other features that characterize this species are its sessile anthers and solitary basal nectary (all other Mexican taxa have five nectaries). Another feature which helps separate this species from other Mexican Symphoricarpos is a pubescent style; this is a plastic character, however, and not always reliable. The foliage and style of populations in Mexico range from completely glabrous (in Baja California) to pubescent (in northern Chihuahua) (Fig. 4). WILLIAMS, Symphoricarpos guatemalensis 277 Representative specimens. MEXICO. Baja California Norte: Sierra San Pedro Martir, east slope of Cerro de la Culpa, 31° 02'N, 115° 27'W, 27 Jul 1970, Moran 17890 (TEX). Chihuahua 23 mi ENE of Villa Ahumada, 30° 40'N, 106° 07'W, 12 Sep 1973, Henrickson ee (TEX). Symphoricarpos microphyllus HBK. Symphoricarpos microphyllus is the largest of the Symphoricarpos species, some- times attaining a height of over 2 meters. It is readily distinguished from S. palmeri and S. parishiz in its erect habit and by having all its leaves entire (Fig. 3). This is the most widespread Symphoricarpos species in both Mexico (Fig. 4) and Guatemala (Fig. 2). Throughout most of its range S. microphyllus is pubes- cent, but there is a rather large isolated population of glabrous individuals in the Sierra Madre Oriental along the border of southern Coahuila and the Midwestern Nuevo Leon border ae 4), Representative specimens. GUATE A. Chimaltenango: slopes of Volcan de Acatenango, above Las Calderas, 3 Jan 1939, ae pee (F); Volcan de Agua, 22 Jul 1937, J.R. Johnston 910 (BE). El Progresso: between Calera and summit of Volcan Siglo, 21 Jan 1942, Steyermark 43042 (F). Huehuetenango: across river from San Juan Atitan, 8 Sep 1942, Steyermark 52018 (F); Cerro sees above S San Ildefonso Ixtahuacan, 15 Aug 1942, Steyermark 50566 (F). Quezalte o: Summit of the Sierra Madre Mts, 15 km N of Ostuncalco, 9 Dec 1963, L.0. ee oe ao (F); a Madre Mts, 5 km N of Ostuncalco, 8 Dec 1963, L.0. Williams 25472 (F); Cuesta de El Caracol, Sierra Madre Mountains, 5-8 km N of San Juan Ostuncalco, department of Quezaltenango, 11 Dec 1962, Williams Do 768 (F, TEX); mts SE of Palestina, along new road to San Juan Ostuncalco, 21 Jan 1941, Standley 84354 (F), Volcan Santo Tom 22 Jan 1940, Steyermark 34804 (F); Volcan Santo Tomas, 22 Jan 1940, Steyermark 34804 o Volcan Santa Maria, bet Santa Maria de Jesus, Los Mojadas, and summit of volcano, 12 Jan 1940, ela 34050 (F); slopes of Volcan Santa M above Palojunoj, 6 Mar 1939, Stand- ley 67668 (F); mts above Olintepeque, 20 Feb 1939, nes 65956 (F); Volcan Santa Maria 27 Jul ie a 870 (F). Sacatepequez: Slopes of Volcan de Agua, above Santa Maria de Jesus, 11 Feb 1939, Standley 65108 (F). San Marcos: between Sibinal and summit of Volcan Tacana, 19 Feb 1940, Steyermark 36080 (F). Totonicapan: ravines in Sierra Madre mts near Villa Las Cruces, 8 km SW of Totonicapan, 13 Dec 1963, L.0. Williams 22969 (F); Nebuloso de Maria Tecun, 21 Nov 1965, Molina 15886 (F). MEXICO. Chiapas: Zinacantan, 6 Oct 1986, Ton 9533 (TEX); San Juan Chamula, 2 Jun 1988, Ruiz 883 (TEX); Valley floor of Zinacantan Center, 13 Apr 1966, mee 668 (F); slope with Oak and Pine, along road to Zontehuitz near Piedrachitas, 25 Aug 1964, Breedlove 7127 8); near Ik'al Nab lake, on the boundary between Chamula and oe antan, ep road t antan center, 21 Sep 1965, Breedlove 12394 (F); along small stream 7 km NE of Huistan, along road to Och and Ocosingo, 21 Sep 1972, Breedlove 27750 (F). Coahuila: Sierra E] Para Arteaga, 17 Jun 1991, Hznton a 2 (TEX); Sierra Pilote de Fierro, 25° 10'N 25'W, 24 Oct 1991, Carranza 1221 (TEX); Las Vigas rie Sierra de Arteaga, 25° ae 100° 39'W, 5 Jun 1987, eames (TEX): Sierra la Marta E of Cerro Moro, 22 Jul 1985, Ginzburg 141 (TEX); Muzquiz, Sabinas river, 11 Jul 1936, oe 475 (BRIT-SMU). Distrito Federal: volcanic soil, SE side of D.E, 12 Jul 1947, Barkley 2215 (TEX); Desierto de los Leones, 5 Nov 1942, Steyermark 52250 (F). D go: canyon head, 51 mi SW of Durango, 13 Aug 1957, Waterfall 13727 (BRIT-SMU, F, TEX). Guanajuato: San Miguel de las Casitas, 8 Sep 1990, Ventura 8692 (F). Hildago: on Hwy 104 NE of Pachuca, 49.7 mi E of Ajacuba, 10 Jul 1985, Cowan 5493 (TEX); San Vicente, 16 Aug 1937, Fisher 37191 (F); upper cliffs 278 Stipa 16(2) 1994 . 3. Leaf shape of selected Symphoricarpos species. A. Symphoricarpos guatemalensis. B. S. micro- ae C.S. palmeri. d. S. parishii, non-flowering branch. E. 8. parishit, flowering branch. tepec Station, 27 Jun 1904, Pringle 13010 (F); Sierra de Pachuca, 20 Jul 1901, sae 9480 (Ey: aie Nerds de Colian: i, brecha entre E] Izote y la Micronda las Viboras, 5 Feb 1992, Chazaro 6808 (TEX); NE slopes of Nevado de Colima, below Canoa de Leoncito, 10 Sep 1952, McVaugh 12875 (BRIT-SMU). Mexico: 11 mi E of Amecameca, 31 Jul 1975, LeDoux 2037 (TEX); apes Toluca and Mexico, 24 Jul ae ? eave 9 (TEX): pine forest of Las Crucea, 12 Jul 1932, Hinton 1013 (F); 2 km SE of Totoloapan, 18 km E of Tezcoco, 10 Jul 1979, Koch ae (F); valley in mountains, 16 mi E of Toluca, 18 Aug 1957, Waterfall 14067 (F). Michoacan: on road to Uruapan, 4 mi S of Cheran, 3 Nov 1958, Jones 23326 (TEX). hills of Patzcuaro, 11 Oct 1892, Pringle 4261 (F); Cerro Huashan, 2 km S of Nahuatzen, 28 Jun 1990, Garcia he (F); Zitacuaro, 19 Aug 1938, Hinton 13120 (F); 1 km W of Rancho Seco on Hwy 37, 2 km W oe Morelos, 13 Sep 1986, Swagel 167 (F); 6 km SE of Villa Madero, 30 Sep Nunez 4732 (F). Morelos: Tepoztlon, Jul 1938, pias A340 (BF). ieee neat 6 km SE of ce ‘Tnidad: summit of one of the aes of Sierra Cebolla, 25° 1 100° 07'W, 5 Aug 1988, Patterson 6174 (TEX); ra La Marta, Galan 6 Sep a ad 18343 (TEX); Cerro, El Potosi, 21 Jun 1969, ae 17115 (TEX); 1.5 mi SW of Pabillo, 21 Jul 1958, Corre// 19953 (TEX); Municipio Galeana, open pine forest near Cerro Potosi, 23 Jul 1938, Schneider 953 (F); peak of Cerro Potosi, Municipio de Galeana, 21 Jul 1935, Muweller 2256 (F); Sierra Potosi, by the N hogback, 20 mi NE of Galeana, 26 Jul 1934, Mueller 124 — Lerios Mts., Coahuila, 45 mi E of Saltillo, 10-13 Jul 1880, Balinese sae (F). Oaxaca: Cerro Yeas. It il 1908, bales 3518 (F). Puebla: El Cimarron, 28 Jul 197 ee 3954 (LE X); el Chamizal, 20 Jul 1970, Ventura 1722 (F); in moist sandy loam, a and Juniper woodland, 10 mi W of San Salvador de Seco, 2 Aug 1947, Barkley 2469 (F). Queretaro: 0.5 km of Cuatro Palos, along road to La Canada, 20 Sep 1989, Gia 2110 (F); Campo Alegre, 3 km S of San = 24 May 1986, Fernandez 3298 (TEX). San Luis Potosi: 7 km SW of Pozuelos and 22 km SW of San nae Potosi, on aie to ern 22° 03'N, 101° 10'W, 21 Aug 1973, Johnston 12271. (TEX). a WA Symphoricarpos palmeri G. N. Jones Symphoricarpos palmeri is a relatively restricted endemic, occurring in the Trans-Pecos region of Texas and the Sierra del Carmens of northern Coahuila, WILLIAMS, Symphoricarpos guatemalensis 279. Fic. 4. Psi ipaneD of S. ie (sq mae a CODE (circles) in Mexico. Open char- acters represent g rs represent pubescent populations. Mexico. It is recognized by its trailing habit and crenate-mucronate leaves of both flowering and non-flowering branches (Fig. 3). The species has mostly pubescent stems and leaves, but individuals with glabrous stems occur throughout its range (Fig. 5). Represent: ative specimens: MEXICO. Coahuila: 3.5 km N of Cerros El Centinela Creek and 4.8 km S of Canyon El Jardin, a 06'N, 102° 35'W, 19 May 1992, Mayfield 1449 (TEX); Sierra Santa Fe del Pino, WNW of Hacebuches, 28° 13'N, 103° 02'W, 26 May 1973, Johnston 11236 (F, TEX); 4 mi W and 10 mi S of Ocampo, 16 Jun 1956, Graber 163 (TEX); N igneous slope of Picacho del Centinela, 24 Aug 1953, Warnock 11603 (TEX). Symphoricarpos parishii Rydb. Symphoricarpos parishii is distinguished from the other Mexican taxa by its trailing habit, entire leaves of the flowering branches, and lobed leaves of the older non-flowering branches (Fig. 3). Specimens studied from Mexico have glabrous stems, although pubescent individuals are found scattered throughout its range in the United States. This species has been regarded as a variety of both S. oreophilus A. Gray by Cronquist et al. (1984) and S. rotundifolius A. Gray by Dempster (1992), but this problem is beyond the scope of the present paper (Fig. 5). 280 Sipa 16(2) 1994 a 30 ; a r Zor 20 1S + | l n \ | 1 15 110 105 100 Q5 90 Fic. 5. Distribution of S$. orbiculatus (triangles), S. palmeri (circle), 8. parishii (square) in Mexico. Representative specimens. MEXICO. Baja California Norte: Parque Nacional Sierra San Pedro Martir, Vallecitos, 31° 02'N, 115° 27.5'W, 1 Sep 1985, Thorne 61437 (TEX); Sierra de Juarez, Parque Nacional Constitucion de 1857, Laguna Hanson & adjacent rocky flats and slopes, 26 May 1987, Thorne 62362 (F); upper east end of Vallecitos meadow, 31° OO'N, 115° 27'W, 3 Jul 1973, Moran 21082 (TEX) ACKNOWLEDGMENTS I would like to give special thanks to the very warm and generous Dr. Guy Nesom for suggesting this project to me, as well as for the Latin diagnosis and many fine readings of the manuscript. Nancy Webber was beneficial in provid- ing beautiful and expedient illustrations. Alan Prather and Dr. B. L. Turner provided helpful comments. Cooperation from BRIT-SMU, FE, and TEX-LL was also much appreciated. REFERENCES Cronguist, A., A. HOLMGREN, N. HOLMGREN, J. REVEAL and P. HoLMGREN. 1984. Capritoliaceae. In: Intermountain flora; vascular plants of the Intermountain West, U.S.A. Hafner, New York. 4:539. Dempster, L.T. 1992. A nomenclatural change in sea Madrofio 39:77—78. Gray, A. 1873. Revision of the genus Symphoricarpos. J. Linn. Soc., Bot. 14:9—12 jours. GN 1940. A monograph of the genus Symphoricarpos. ae Arbor. 21: 201-252. SPECIES POOLS IN EASTERN ASIA AND NORTH AMERICA SHIYOU LI and KENT T. ADAIR The Tucker Center College of Forestry Stephen F. Austin State University P.O. Box 6109 Nacogdoches, TX 75962-6109, USA. ABSTRACT A species pool refers to the total species in a defined region at a given time. The species pool in eastern Asia is the product of a long history with low speciation rates, while the species pool n northwestern North America is largely due to recent rapid speciation. Data fail to support a theoretical dominancy of ecological determinism in the development of regional species aes A specs pool 1 is the product of evolution Sind pene and extinction and thus rally d t species richness to be an historical artifact rather tl eter- ae reality. RESUMEN La gran riqueza de especies en el este de Asia es el efecto de una historia larga con unas tasas de extincién y especiacion bajas, mientras que la gran riqueza en el noroeste de Norte América es debida a especiaci6n reciente rdpida. Los datos no explican la dominancia teérica del determinismo ecolégico en favor de la historia geofisica peptone sSpecuice como base a explicar la diversidad regional de especies. Se propone un “pool” p la evolucién, inmigracion y extincion, y por ello se postula que la ri n ies vivientes es un artefacto hist6rico en vez de una realidad ecol6gicamente detemainads, INTRODUCTION Species diversity is a central theme of environmental studies. Scientists have accumulated much data on species diversity over the past 250 years. However, there are still no answers to some basic questions. One critical question that 1s frequently asked by scientists is why there are just so many species in a region, and not more or less than that number. In other words, what factors determine regional species richness? In fact, a comprehensive, rigorous, and general theory of species diversity is lacking (Solbrig 1991). Ricklefs (1987) thus claimed that species diversity is “the major, unexplained pattern in natural history,” one that “mocks our ignorance.” One dominant hypothesis is that available energy determines regional spe- cies richness (Hairston et al. 1960, MacArthur and Wilson 1967, Currie and Paquin 1987, Moore 1987, Adams and Woodward 1989, Andersson 1990, Currie 1991). Currie and Paquin (1987) and Currie (1991) clearly claimed that Sipa 16(2): 281 — 299. 1994 282 SipA 16(2) 1994 regional tree species richness can be explained in terms of actual evapotranspira- tion (AET). Alternatively, White (1983) claims that the tree species diversity pattern of the Northern Hemisphere is the product of extinction. Goldsmith (1985) believed that periodic, or episodic, catastrophic destruction of life was caused by comet or asteroid impacts. However, there are three critical problems with these positions: First, almost all studies of species richness concentrate, very naturally, on just one taxonomic group (e.g. pines, oaks) (Moore 1987). But each group may have some peculiar requirement of its own that obscures the general causes of species diversity (Moore 1987). Second, studies were often made using small biogeographic regions. Micro-environmental variations greatly influence the result when comparing two small regions. Third, more impor- tantly, the conceptual definition of a species has not been consistent from tax- onomist to taxonomist nor from region to region. To minimize the observation biases caused by the factors mentioned above, this study examined all seed plants in the larger biogeographic regions in the Northern Hemisphere. Eastern Asia in this study refers to the eastern forest region of China, Far East and eastern Siberia of Russia, and Korea (Fig. 1). North America includes the continental regions of Canada and United States. The forests of North America are naturally divided into those west and those east of the Great Plains (Fig. 2). The boundary of the boreal zone in the North- ern Hemisphere follows Larsen (1980) with modification in eastern Asia. The boreal region in Europe includes the area westward from the Ural Mountains (Fig. 3: I). Boreal eastern Asia includes eastern Siberia in Russia and the Da Xingan Ling Mountains of China (Fig. 3: I). The east-west division of the boreal zone in North America follows Daubenmire (1978). The eastern section includes the area east of the Rocky Mountains (100°W) (Fig. 3: 1V) and the western or Cordilleran Section extends westward (Fig. 3: IID). In eastern Asia, the temperate forest region includes the Xiao Xingan Ling, Wanda, and Changbai mountains in Northeast China, southern portion of Far East of Russia, and northern Korea (Fig. 3: V). This region is known as the Korean pine conifer and hardwood forest region (Chou and Li 1990). In western North America, the temperate forest region covers portions or all of the Pacific Northwest states of Oregon, Montana, Idaho, Washington, and southern Brit- ish Columbia (Fig. 3: VI). This region includes the Pacific Northwest forests (Franklin 1988) and the northern portion of the forests of the Rocky Mountains (Peet 1988). In eastern North America, the temperate forest region extends from northern Minnesota upper Great Lakes region and eastward to southern Canada and New England, including, toward the southeast, much of the Appa- lachian Plateau in New York and northern Pennsylvania (Braun 1967, Fig. 3: VII). The region is known as the Hemlock-white pine-Northern Hardwoods region (Braun 1967, Vankat 1979) as well as the mixed conifer and deciduous forest region (Bailey 1976). LI AND ADAIR Species pools | MM Eastern Asia cegecae es | Fic. 1. Map showing the definition of eastern Asia in this study. The main data sources of this study include field investigations, specimen observations, personal communications, and literature (including flora and fos- sil records). The primary data for eastern Asia were collected from a comprehen- sive forest survey conducted by aut — qors and their colleagues at Northeast For- estry University and numerous forestry agencies during 1984-1988. Data for Stipa 16(2) 1994 in) [ee) ys 2on{i | oe \ 130° a eres ae Nes - <7 a I Western North America | II Eastern North America | 10°N i: ee Scale of Miles 110° 100° Fic. 2. Map showing the definition of western and eastern North America in this study. North America is based on literature, seas where possible, by field collec- tions, herbarium 1 investigations, and personal interviews during 1989-1992. By using established methods and the same el chen in the herbarium and field, the observational biases that often plague broad-scale comparisons were minimized. Ll AND ADAIR, Species pools 285 130° 420° 110° 100° 90°E 80° 70° 60° Fic. 3. Map showing the boreal and temperate forest regions involved in this study (I-IV— boreal forest zone: I-Europe, [-eastern Asia, []I]-western North America, [V-eastern North America; V—-VJI—temperate forest zone: V-eastern Asia, VI-western North America, VU— v Five sheets of voucher specimens of each taxon of seed plants found in the boreal and temperate forests of eastern Asia were collected during the field inves- tigations. The location, life history, growth forms, function of woody plants, vegetative organs, habitats, and other characteristics such as flower color of each species were recorded for each specimen. The voucher specimens were retained in the Institute of Plant Sciences at Northeast Forestry University in Harbin, China. Manuals involving the floras of East Siberia, Far East, Korea, and Japan were also 286 Stipa 16(2) 1994 used to identify voucher specimens. Herbarium collections, located in Northeast Forestry University, the Botanical Institute in Beijing, and the Applied Ecologi- cal Institute in Shenyang were also used to confirm identification of voucher specimens. Seed plant checklists of the boreal and temperate forests of eastern Asia were created separately from these comparisons. Additional floristic check- lists were created separately for boreal Europe, boreal western North America, boreal eastern North America, temperate Japan, temperate western North America, and temperate eastern North America using scientific references: Ohwi (1965), Polunin (1959), Gleason and Cronquist (1991), Hitchcock and Cronquist (1974), and Scoggan (1978-1979). In both boreal and temperate forest comparisons of Europe, eastern Asia, and western and eastern North America, floristic analyses were made at the family, genus, and species levels. Evolutionary analyses of flora were based on the diversi- fication ratios (family and genus size), diversification rate (Eriksson and Bremer 1992), and Hower characters. Phylogenetic analyses were made in terms of three categories: |) primitive—subclasses Magnoliidae and Alismatidae; 2) intermedi- ate—subclasses Caryophyllidae, Hamamelidae, Dilleniidae, Rosidae, Arecidae, Commelinidae, and Zingiberidae; and 3) advanced—subclasses Asteridae and Liltidae (following Cronquist’ system in 1988). EASTERN ASIA: A LONG EVOLUTIONARY HISTORY COUPLED WITH LOW SPECIATION RATES The great plant species diversity of eastern Asia is well-known and has been widely emphasized by scientists (Sargent 1913, Hu 1935, Takhtajan 1969, Li 1952, Wuand Wang 1983, Boufford and Spongberg 1983, Latham and Ricklefs 1992). To date, 25,480 species of native vascular plants have been identified in eastern Asia, representing 300 families and 2,875 genera (Li 1993). This amounts to 10.7% of the total known species in the world and about 1.6 times the figure for North America (Table 1). The great plant diversity in eastern Asia is the product of long evolutionary history. The geological and paleobotanical records failed to support the existence of continental Quaternary glaciers in most of eastern Asia and thus extinction is not a main process in the development of the extant eastern Asian flora. Also, eastern Asian flora consists of a host of taxa that are presumed to be phylogenetically primitive, with many occurring as mono- typic taxa in the subtropical or tropical regions. These factors have led to the concept that eastern Asia was the evolutionary source of modern flora in the Northern Hemisphere (Takhtajan 1969, Latham and Ricklefs 1992). Ferns are phylogenetically primitive vascular plants. 52 families (80.0% of the total flora of the world), 204 genera (46.0%), and 2,300 species (19.5%) of ferns are known tn eastern Asia. The family, genus, and species numbers are 3.5, 3.4, and 6.7 times those in North America, respectively. Gymnosperms, another phylogenetically primitive group of plants, are represented by ten families LI AND ADAIR, Species pools 287 Taste 1. Comparison of taxon richness of native vascular plants of eastern Asia and North America. Eastern Asia * North America ** World *** Family Ferns 52 ib) 65 Gymnosperms 10 5 11 Angiosperms 238 202 542 Total 300 222 620 Genus Ferns 204 60 443 Gymnosperms 34 19 Du Angiosperms 2637 2261 12500 Total 2875 2340 13000 Species Ferns 2300 341 11820 Gymnosperms 180 118 70 Angiosperms 23000 15827 225000 Toral 25480 16285 237490 Notes: * from Li (unpublished, 1988); ** based on Checklist of D. E. Moerman (Nancy Morin, pers. comm.); *** from Wu and Wang (1983). in eastern Asia, but only five in North America. Eastern Asia contains almost all primitive monotypic families, including Ginkgoaceae, Cycadaceae, and Gneta- ceae. Ginkgo, for example, is the oldest of all trees, ic dates back to the Triassic some 200 million years ago (Flora of North America Editorial Committee 1993). It is now naturally restricted to China although it was quite widespread in the world during the Cretaceous (Hsu 1983). Conifers apparently represent the greatest taxonomic diversity in eastern Asia. They are distributed among Eu- rope, eastern Asia, and western and eastern North America, respectively, ap- proximately in the ratio 1.0: 3.4: 2.1: 1.6 by genus and 1.0: 5.9: 2.4: 1.2 by species (Table 2). There are 2,637 genera and 23,000 species of angiosperms belonging to 238 families native to eastern Asia (Li 1993). The family, genus, and species num- bers in eastern Asia are 1.4, 1.2, and 1.6 times those in North America, respec- tively. Almost all ancestral families of angiosperms based on fossil records have greater diversities in eastern Asia than in North America. Chloranthaceae, as the earliest angiosperm (appeared in the early Cretaceous, Muller 1981) has three genera and 16 species in eastern Asia and only one genus and one species in North America. Aquifoliaceae, one of the earliest woody angiosperm families (Muller 1981) represents 168 species in eastern Asia and about 20 species in North America. Moreover, Magnoliaceae, the most phylogenetically primitive woody family of angiosperms (Hutchinson 1973), contains 12 genera and 250 species ranging from Asia to North America (Wu and Wang 1983). Ten genera and 100 species of this family are known in eastern Asia, but only two genera 288 Sipa 16(2) 1994 TaBLe 2. Distribution of conifer species among Europe, eastern Asia, and western and eastern North America. Eastern Western North Eastern North Family/genus Europe sia merica America Cephalotaxaceae Cephalotaxus 0) fi 0 0 axacea Amentotaxus 0 4 0 0 Pseudotaxus 0 1 0 0 Taxus 1 6 1 2 Torreya 0 3 i 1 Taxodiaceae Cryptomerta 0 1 0 0 Cunninghamia 0 2 0 0 G/yptostrobus 0 1 0 0 Metasequoia 0 l 0 0) Sete Lys 0 0) 0 0 Sequota 0 0 l 0) Sequotadendron 0) 0 l 0 Taiwania 0) 2 0 0 Taxodium 0 0) 0 2 Cupressaceae Chamaecyparts 0 1 2 2 Calocedrus 0 2 ] 0 Cupressus 0 7 12 0 Juniperus 6 19 5 5 huja 0 | 1 Thupopsis 0 0 0 0 Pinaceae Abies 5 22 7 2 Cathaya 0 ] 0 0 Cedrus 1 0 0 0 Keteleerta 0 9 0 0 Larix l 10 2 it Picea 3 16 6 ] Pinus 9 2 19 13 Pseudolarix 0 l 0 0 Psendotsuga 0 6 2 0 Tinea 0 6 3 3 Total Genera 7 24 15 1] Total Species 26 153 63 32 and 9 species are known in North America. Hamamelidaceae, another primi- tive family, has 25 genera and 90 species widely distributed in eastern and southeastern Asia and North America with some reaching south to Austra- lia and Africa. 17 or 18 genera and 70 species of this family are known in eastern Asia. Disanthus, Exbucklandria, and Rhodoleia, ancestral genera of LI AND ADAIR Species pools 289 the family (Takhtajan 1969), are all found in southern China. Only three genera and five species are present in North America. The data above imply that eastern Asian flora has a long evolutionary history. However, the analysis of diversification rates (Eriksson and Bremer 1992) of families based on fossil records shows that species diversification rates. of most families in eastern Asia are much lower than those in North America. 20.7% of the total species in eastern Asia belong to families with low diversification rates (R<0.15 my~'), while only 11.2% of the total spe- cies in North America belong to the same families. Thus, we concluded that the great taxonomic richness of eastern Asia is largely the result of a lengthy evolutive history coupled with low extinction and low diversification rates. NORTHWESTERN NORTH AMERICA? A SHORT EVOLUTIONARY HISTORY COUPLED WITH HIGH SPECIATION RATES Geologically, northwestern North America is young. But the northwestern North American flora has the highest species diversity among all northerly re- gions in the Northern Hemisphere (Table 3 and 4). This great species diversity results from high speciation ina short evolutionary history. e taxonomic and evolutionary structure of extant flora is consistent with the concept of a geologically young western North America. In the temper- ate zone, for example, western North America has a total 3,161 species, which is about 50% of the total species of all three temperate floras and about 1.5 times that of either eastern Asia or eastern North America. However, this flora has 19 families fewer than in eastern Asia and 27 families fewer than in eastern North America (Li 1993). Therefore, families of temperate floras, on average, have more species in western North America (roughly 30 species per family) than in eastern Asia (roughly 17) and eastern North America (15, see Table 5). Theoretically, the larger the family (containing more species), the more evolved (Stebbins 1981), because family is parental to species. From this point of view, the temperate flora in western North America is young. The evolutionary analysis of flowers strengthens this statement. The solitary flower, both terminal and axillary, is the initial form of arrangement for flowers, while inflorescence is more advanced and of a greater biological advantage (Takhtajan 1991). In the temperate zone, western North America contains the highest proportion of inflorescence (87.8% of the entire flora, compared with 80.4% in eastern Asia and 79.1% in eastern North America) (Li 1993). Flower color is an important factor affecting pollination of plants. Green flowers are *Diversification rate (R) refers to the speciation rate minus the extinction rate and is mea- sured as the number of extant species over the time since the first appearance of a family. R is measued by the species number per million years (my”'). 290 Sipa 16(2) 1994 Taste 3. Family, genus, and species richness of the seed plants in the three temperate regions. Eastern Western Eastern Total Asiz North America North America Family Total 125 106 133 144 Genus “Trees 30 23 43 Shrubs 74 54 77 119 Lianas 10 2 5 13 Herbs 537 574 495 909 Toral 652 650 615 1090 Species Tees 5 50 95 200 Shrubs 228 235 263 652 Lianas 13 ¢ 10 31 erbs 1786 2867 1632 5505 Total 2092 3161 2000 6388 TABLE 4. Family, genus, and species richness of the seed plants in the four boreal regions. Western Eastern Europe Eastern North North Total Asia America merica Family Total 84 95 76 80 104 Genus Tr 18 20 14 20 25 Shrubs 35 44 37 59 Lianas @) 5 0 1 5 Herbs 215 311 278 206 385 Total 266 379 326 268 470 Species Trees 25 36 28 45 117 Shrubs 84 113 127 100 279 Lianas 0 6 0) 7 Herbs 597 1006 1012 612 1970 Toral 706 1161 1167 758 2373 largely pollinated by wind, and thus the anemophilous plants are claimed to be primitive in evolution (Takhtajan 1991). Alternatively, bright colored flowers are usually pollinated by insects, and entomophilous plants are considered more advanced in evolution. In western North America, only 21.8% of the total species have green flowers, a lower percentage than in either eastern Asia (27.8%) or eastern North America (33.6%). However, colorful flowers (white, yellow, pink, rose, purple, blue, and red) are more frequent in western North America LI AND ADAIR, Species pools 291 TABLE 5. Comparisons of cae ncaa levels of the temperate and boreal floras based on the ratios of genera/family spec ies/fam Eastern Western Eastern orth Europe Asia America America Genus No./Family Temperate Diz 13 4.6 3.2 4.0 4.3 3 Species No. maa Tempera 16.7 29.8 15.0 Boreal 8.4 12.2 15.4 9.5 Species No./Genus Temperate Zone Trees 2.2 2.2 2.2 Shrubs 3.1 4.3 3.4 Lianas 1.3 4.5 2.0 Herbs 3.3 5.0 3.3 Total 3.2 4.9 3.3 Boreal Zone rees 1.4 1.8 2.0 2.3 Shrubs 2.4 2.6 3.4 22 Lianas 0.0 1.2 0.0 1.0 erbs 2.8 3.2 3.6 3.0 Total 2.7 31 3.6 2.8 and comprise 78.2% of the total flora. The same colors account for 72.2% in eastern Asia and 66.4% in eastern North America. Life history, as the result of natural selection, reflects the long-term adapta- tion of a plant to its environment. Annuals are unknown in primitive ferns and gymnosperms, but their sexual reproduction processes enable them to persist in habitats that restrict the establishment of perennials. Therefore, annuals are believed to arise relatively late in the evolution of terrestrial plants (Bazzas and Morse 1991). Annuals are distributed among eastern Asia, western and eastern North America at the ratio of 1.0 : 2.5: 1.0. Alternatively, as the early evolved group, trees are distributed among eastern Asia, western and eastern North America at ratios based on total numbers of both genus and species as follows: 1.3:1.0: 1.9. Obviously, the temperate flora in western North America is relatively young in evolution relative to the other two temperate floras. Further, phylogenetic data strengthen the argument. Magnoliidae and Alismatidae are the most primitive taxa of flowering plants in Cronquist’s system (1988). They, together with gymnosperms represent the lowest diversity in tem- perate western North America (16 families and 56 genera) and the highest in temperate eastern Asia (23 families and 72 genera). But Asteridae and Liliidae, the most advanced taxa in Cronquist’s system have the highest diversities in tem- 292 Stipa 16(2) 1994 perate western North America (243 genera, compared with 233 genera in tem- perate eastern Asia and 199 in temperate eastern North America respectively). In brief, all these evidences from taxonomic structure of flora, flower diver- sity, life history and phylogeny show that northwestern North American flora is relatively young in evolution. However, data on diversification rates (R) for families show that there is high species diversification in northwestern North America. A total of 25 families with low diversification rates (R<0.10 my!) are recorded in the temperate regions, and they are distributed among eastern Asia, western and eastern North America, respectively, at a species ratio of 2: 1 : 3 (Table 6). Western North America clearly has the smallest taxon diversity among families with low diversification rates. In contrast, the nine families with higher diversification rates (R>0.60 my!) are distributed at a ratio of total number of species of | : 3: 1 in eastern Asia, western and eastern North America. Further, three genera with the highest diversification rates, Astraga- lus, Senecio, and Carex display maximum species richness in western North America and exceed that of any other region (279 species relative to 51-145 species in the temperate regions and 144 species relative to 56-90 species in the boreal regions). Boreal floras show a similar pattern (Li 1993). Thus, there is little doubt that the data indicate that the northwestern North American flora is the product of high species diversification and rela- tively recent evolution. Considering that the species-rich northerly flora of western North America has over 50% of the total species restricted to this region, we conclude that this great species richness in northwestern North America is largely the effect of rapid speciation rather than immigration. SPECIES POOL HYPOTHESIS Based on our studies, we argue that species exist in a pool ina given region at a given time. This regional species pool 1s a relatively static spot ina dynamic river of species diversity development over time. It is the result of two additive processes: speciation and immigration; and, one subtractive process: extinction (both locally and globally) (Fig. 4, Table 7)! Thus, a species pool has largely an historical rather than an ecological basis for its existence (Li 1993). Speciation is the formation of new species from pre-existing ones usually by a process of improved adaptation of survivors to the environment. New species may be better adapted to the environment and replace the less well adapted ones. Speciation began long ago and is still going on. It is best characterized as a regional subset of the global genetic pool. The rate of speciation depends not only on environmental diversity within the pool, especially under conditions of geographic or ecological isolation, but also on chance. Thus, it is impossible to predict the speciation rate for a region. Immigration, however, does not involve global genetic variability, but does Li AND AbalrR, Species pools 293 TABLE 6. Evolutionary levels of temperate and boreal floras based on diversification rates (R). Western Eastern Diversificate Rate(R) Europe Eastern North Asia America America R< 0.10 my" Temperate Family Number 21 11 22 Species Number 77 42 111 % of Total Species 3.7 1.3 5.6 Boreal Family Number 10 13 7 11 ecies Number 21 35 7 11 % of Total Species 3.0 3.0 1.7 4.6 R > 0.60 my" Temperate Family Number 8 9 9 Species Number 95 258 91 % of Total Species 4.5 8.2 4.6 Boreal Family Number 6 8 7 6 Species Number 4S 83 108 42 % of Total Species 6.4 7.1 9.3 5 Notes: *species per million years. TABLE 7. Basic patterns of development of regional species pool. Environmental Pattern Historical mcts + + + + : : a 2 Heterogeneity in Pool + + - - + + - - Barriers ae Pool - ; - + - + - + Process Extinction - + + + - - - Speciation + ‘ 7 - + + e z migration + - + - + - + - Species Diversity M M“ L L H H¢ L L Example Cl NW NE EE x SEA NEA SP Notes: “+” represents great and “—” suena weak; L—low, M—median, H—high; ©@ represents ore endemic species. CI lis slands, NW—-northerly western North America; NE— ~ —unidentified region; SEA—south- northerly eastern North America; EE erly eastern Asia; NEA—northerly eastern Asia; SP—specialized habitat. Sipa 16(2) 1994 No \o dy Adaptation to environments SPECIATION IMMIGRATION V SPECIES POOL Global Local EXTINCTION Failed adaptation to environments Fic. 4. Species pool ina given region at a given time (After Li 1993). contribute to regional changes. During migration, species may evolve. If genetic change makes it possible to distinguish new species from the parent species, this process is evolutionary (speciation), if not, it is called immigration. Both natural selection and chance play roles in immigration of species to a region. This also means that the absence of a species in a region may result purely from chance rather than deterministic natural processes. Extinction of a species can occur locally (exclusion or emigration), in which one or more populations vanish but others survive elsewhere, or globally, in which all members of the species population everywhere in the world vanish. Extinction of a species is a natural process. The fossil record indicates that most, if not all, species have a finite life span, averaging between 1 and 10 million LI AND Apair, Species pools 295 years (Solbrig 1991). Drastic environmental change is a major cause of species extinction especially on a global scale. Because the environment is in a constant state of transformation, some species are always being lost while others are added. Some changes in the physical environment are cyclic and periodical, while others are less predictable. Thus, it is impossible to establish a normal rate of species change and based on the data presented in this study, it may be impossible to do so at a A species pool has the following main characteristics. Liquidity and Irreversibility A species pool is a dynamic flow of species through a storage area. Adapted to new environments, species that have become locally extinct (exclusion) may re- migrate back into a region. However, the natural process from speciation to extinction is irreversible. Species that have become globally extinct cannot be recovered naturally. Thus, conservation of biodiversity is needed on a global level. But, it is impossible for humans to interrupt the natural extinction pro- cess and store all species on a global scale. Man can temporarily store as many species as possible in regional species pools, but even these are present only on a dynamic basis. A species pool is not the accumulation of species only over long evolutionary time spans, and species-rich communities may not always be the oldest as Whittaker (1977) has stated. Western North America has a young geological history. The rapid rise of the Rocky Mountains in the Tertiary and extensive glaciation in the Pleistocene caused widespread extinction of life in western North America. However, western North America has the greatest species pool in the northerly regions of the Northern Hemisphere. The northerly flora of western North America, which is both young and species-rich, is a striking example that fails to support Whittaker’s hypothesis. This fact is also not con- sistent with White’s hypothesis that emphasizes that the differences in taxon richness among eastern Asia, North America, and Europe results from differing Pleistocene extinction rates in the three areas (White 1983). In fact, it is not realistic to focus on preservation of static ecosystems for preservation of maximum species richness or biodiversity. Stochasticity and Non-equilibrium Nature is not in balance. Disturbances and irregularities of all sorts are not aberrations, but integral parts of nature (Solbrig 1991). Both chance and natural selection produce the steady coming and going of species through a pool. Speciation, immigration, and extinction function continuously. Thus, the size of a species pool (species richness) is indeterminate and unpredict- able. The theory of species equilibrium, namely, that the steady-state num- ber of species, found on an island or isolated patch of habitat due to a 296 Sipa 16(2) 1994 balance between the immigration of new species and the extinction of old residents (MacArthur and Wilson 1967, Wilson 1992) is inadequate be- cause it denies the reality of chance. Heterogeneity and Non-saturation The basic arguments of the currently dominant energy hypothesis are: (1) species richness is a measure of available energy, and (2) the product of a balance between immigration and extinction which causes richness to approach its theo- retical maximum over time (MacArthur and Wilson 1967). Therefore, this hy- pothesis 1s that species richness in similar environments is the result of inevi- table convergence. In general, tree species richness supports this and increases in direct relation to precipitation or AET, suggesting that a positive relationship exists between diversity and productivity of the habitat (Latham and Ricklefs 1992). But this is not always true. There are many examples of species richness decline at high habitat productivity levels (Ricklefs 1987, Latham and Ricklefs 1992). The total number and species of plants within a community may reflect total productivity, but species richness does not. Different species have dif- ferent individual size, population and distribution patterns and thus have different energy needs and different ecological roles in an ecosystem. Some species, known as keystone species (Wilson 1992), affect the survival and abundance of many other species in the community in which they live. In contrast, the presence of some species may be caused by, and be largely dependent on, the existence of other species. Clearly, species 1s not an energy or ecological unit. Theoretically, every spe- cies has a unique niche because every species has at least one physical or behav- ioral characteristic that defines it from other species (Solbrig 1991) even though the current measures of environmental parameters are too crude to distinguish all differences among species. The ecological inequality of species indicates that the number of species present is not a function of the physical environment. On the other hand, all regions exhibit heterogeneity and patchness and no environ- ment 1s completely homogeneous. This variability and patchness in the envi- ronment provides a foundation for the coexistence of species. Consequently, competition seems less important as a determinant of species richness throughout a large scale biogeographic region than on a local site. How- ever, species richness does not always tend toward its possible maximum largely due to the effect of chance. There is reason to assume that nowhere in the world have resources been fully utilized by plants. A measure of energy availability cannot even allow prediction of the maximum number of species in a region because plants have the ability to expand local envi- ronmental constraints during colonization. If species saturated biological communities exist within limits set by local conditions, new species could not join the community without the compensating disappearance of others Li AND ADAIR, Species pools 297 (Ricklefs 1987). But many successful introductions of exotic species can- not be shown to have caused an apparent loss of native species. In subtropi- cal eastern Asia, for example, where tree species richness is about four times that of subtropical eastern North America, about 50 species of trees and woody vines introduced from eastern North America have successfully colo- nized without the compensating disappearance of the native species. Ameri- can Robinia psendoacacia L., Campsis radicans Seem., Sabina virginiana (L.) Antoine, Pinus elliottii Engelm., P. taeda L., P. rigida Mill., Magnolia gran- diflora L., Carya tllinoensis (Wangenh.) Koch, and Liriodendron tulipifera L., all now grow well in China and do even better locally than their eastern Asian species pairs. Of them, Robina pseudoacacia and Campsis radicans have escaped from plantations and gardens and became naturalized species in China (He and Gu 1990). In these cases, competition between exotic and native species has lead to increased diversity rather than extinction and loss of species in the area. Thus, a climate or energy hypothesis alone cannot explain this fact of additional species richness because both neglect histori- cal plant evolution and migration. Previous studies have usually concentrated on a small taxonomic group of organisms and thus failed to provide a general theory for the existence of species pools. In short, what is present on a given site at a given time, or interval of time, is a product of chance. The natural processes flowing in a river of information are relatively deterministic, but the residents on site at a given time are only those who then happen to reside there. ACKNOWLEDGMENTS We are grateful to Barney L. Lipscomb of Botanical Research Institute of Texas, Hebe R. Mace, Darrel L. McDonald, Elray S. Nixon, Hershel C. Reeves, and Kenneth G. Watterston of Stephen E. Austin State University, and an anony- mous reviewer for reading the draft and providing valuable comments and suggestions, REFERENCES Abas, J.M. and EI. Woopwarp. 1989. Patterns . tree species richness as a test of the rlacial extinction hypothesis. Nature 339: Sc ANDERSON, L. 1990. The diving : S| pts and ecology. Taxon 39: 375-382 BaiLey, R.G. 1976. Ecoregions eae Unted States. ‘USDA Forestry Service, Intermountain Regionm Ogden, Utah, Ma Bazzaz, F.A.and$.R. Morse. 1991. Annual plants: potential responses to multiple stresses. In: demic Press, San Diego, New York, Boston, London, Sydney, Tokyo, Toronto. pp 283-306. BourrorbD, D.E. 1992. Affinities in floras of Taiwan and North America. In: C.I Peng (ed.), Phytogeography and botanical inventory of Taiwan eae. of ROC-USA Symposium 298 Sipa 16(2) 1994 on phytogeography and botanical inventory of Taiwan, March 11-12, 1992, Taipei). Institute of Botany, Academia Sinica, Taipei. pp 9. Bourrorb, D.E. and S. A. SPONG BERG. 1983. ies Asian- eastern North American Pay togeographical rele tury. Ann. Missouri Bot. Gard. 70:423—439. Braun, E.L. 1967. Deciduous forests of eastern North America. Hafner Publication Com- Cuou, Y.L. and S.Y. Li. 1990. Forests of China. Science Press, Beijing. (Chinese). Cronauist, A. 1988. — evolution and classification of flowering plants. 2nd ed. New York Botanical Garden, Bronx, New Yor Cutorta, E. 1994. ee gather co mix of policy, science in Nashville. Science 265: 1178-1179. Curriz, DJ. 1991. ag and large-scale patterns of animal-and plant-species richness. Amer. Naturalist 137:27-49. Currir, D.J. and V. Paquin. 1987. Large-scale biogeographical patterns of species richness of trees. Nature 329:326-327 ae ee R. 1978. Plant geography, with special reference to North America. Academic s, New York, San Francisco, Londor ee O. and B. Bremer, 1992. Pollination systems, ee modes, life form, and diver- sification rates in angiosperm families. Evolution 46:258— FLORA OF Now AMERICA EpIroriAL COMMITTEE. 1993, Flora of ‘North America North of Mexico. 2+ vols. Oxford Univ. Press, New York. ae 7 F. 1988. Pacific Northwest forests. In: M.G. Barbour and W.D. a eds. North American terrestrial vegetation. Cambridge Univ. Press. pp. 103—130. GLEASON, H.A. and A, Cronguist. 1991. Manual of vascular plants of northeastern United States and a Canada (2nd edition). New York Botanical Garden, Bronx, New York. GoLpsMITH, E. 1985. eels succession rehabilitated. Ecologist 15:104—112. AIRSTON, N.G., FE. SMITH L.B. SLopopkin. 1960. Community structures, population pope id competition oe Naturalist 94:421—424. He, S.A. wad Y. Gu. 1990. A study on mutual introduction of main woody species between China and the United States. Bull. Nanjing Bot. Gard. 1990:1—21. Hrrcncock, C.L. and A. Cronquisr. 1974. Flora of the Pacific Noga an illustrated manual University of Washington Press, Seattle and London Hsu, J. 1983. Late Cretaceous and Cenozoic vegetation in Ching, emphasizing their connec- tions with North America. Ann. Missouri Bot. Gard. 70:490—508. i, H.H. 1935. A comparison of the ligneous flora of China and eastern North America. Bull. inese Bot. Soc. 1:79—97 Ht Hurtcuinson, J. 1973. The families of flowering plants arranged according to a new system based on their area phylogeny. 3rd ed. Clarenden Press, Oxford. LaruaM, R.E. ’. RIcKLEFS. 1992, Continental comparisons of temperate-zone tree species diversity. Pages 294-314 In: R.E. Ricklefs and D. Schluter, eds. Species diversity: his- torical and geographical perspectives. Univ. of Chicago Press, Chicago. Li. H.H. 1952. Floristic oo between eastern Asia and eastern North America. Trans. Amer. Philos. Soc. 42: —429. (Reprinted with forward and additional references, as a Morris Arbor. Monogr. Li, 8.1993, Flotisnic me ecological comparisons of the continental forests in eastern he and North America se aasag es in 1 the boreal and temperate zones. Doctoral Disser- tation of Stephen F. Austin State Univ., Nacogdoches, Texas Li, S.Y. and K.T. Abair. 1995 (in i Gite s00ls of seed ssleats in eastern Asia and LI AND ADAIR Species pools 299 North America. Henry M. Rockwell Monograph, Stephen F. Austin State Univ., Nacogdoches, Texas. MacArr THUR, R.H. and E.O. Wison. 1967. The theory of island biogeography. Princeton Univ. Press, ee Vion: P.D. 1987 ae saul a forest rich? Nature 329:292. Mutter, J. 1981. Fossil pollen records of extant angiosperms. Bot. Rev. 47:1-42. Ouwi, J. 1965. Flora of Japan. Smithsonian Institution, saan ee D.C. Pret, R.K. 1988. Forests of the Rocky Mountains. In: M.G. Barbour and W.D. Billings, eds. North American terrestrial vegetation. Cambr ridge Univ. a pp. 103-130. Potunin, N. 1959. Circumpolar Arctic flora. Clarenden Press, Oxfor Rick ers, R.E. nee Community diversity: relative roles of local a tensor processes. Sci- ence 235:167-17 SARGENT, C. , LOL: Lesa In: E.H. Wilson, A naturalist in western China. Methuen & Co. Ltd., London. pp. x SCOGGAN, HJ. 1978- 1979. oie of Cans ada. Parts 1-4. Natl. Mus. Nat. Sci. Canada, Publ. Bot. No. 7, Ottawa. STEBBINS, 1981. Why are there so many species of flowering plants? BioScience 31: 573-57 sai oO e 1991. The origin and function Sose aan Environment 33:15—38. TAKHTAJAN, A. 1969. Flowering plants, origi 1. Edinburgh ea A. 1991. Evolutionary aoa in dewenes plants. Columbia Univ. Press, New ork. Vankat, J.L. 1979. The natural vegetation of North America. John Wiley & Sons, New York, Chichester, Brisbane, Toronto. Wuire, P.S. 1983. Eastern Asian-eastern North American floristic relations: the plant commu- nity level. Ann. Missouri Bot. Gard. 70: oe eae R.H. 1977. Evolution of versity in land communities. Evol. Biol. 10: 6. ee E.O. 1988. The current state of biological diversity. In: E.O. Wilson and EM. Peter, eds. Biodiversity. National Academy Press, Washington, D.C. pp. 3-18. WiLson, E.O. 1992. The diversity of life. The Belknap Press of Harvard Univ. Press, Cam- bridge, Massachusetts. Wu, Z.Y. and H.S. WaANnc. 1983. Natural geogre Press, Beijing. (in Chinese). Plant geography (1).Science 300 Stipa 16(2) 1994 BOOK NOTICE SWINK, Floyb and GERoOULD WILHELM. 1993. Plants of the Chicago Re- gion. Fourth Edition. (ISSN 1-883362-01-6, hbk). Indiana Academy of Science, 1102 North Butler Avenue, Indianapolis, IN 46219-2918. $28.00 (Academy members); $35.00 (non-members) (+ postage: $2.50 domestic, $5.00 foreign). 921 pp. e “Chicago Area,” lying within a 75-mile radius of the Windy City’s center, includes parts of Illinois, Indiana, Michigan, and Wisconsin. The first (1969) and second (1974) editions of this work were by Swink alone; the third (1979) added Wilhelm and well-done keys. This fourth edition, with 226 more taxa than its predecessor, is the most splendid of them all. Even though Chicagoland is urbanized and depressingly industrialized, it still has many “natural” areas and a flora of 2530 taxa, about 35% of them introduced. The arrangements of families, genera, and species ts strictly alphabetical, as in an ordinary dictionary. Look, then, for the key to genera of Cyperaceae under the C’s; the key to, and species entries for, Sc#rpus under the S's. TY are AE ae aE SIGE - ; ' I ) I rod data, but also listings, often lengthy, of associated species. The glossary is well illustrated; the bibliography has about 1750 entries. Swink & Wilhelm bring to this impressive and authori- tative work a total of some 75 years of experience in passionate pursuit of plants in their region. Could a flora be written by authors more qualified for the task? I doubt it.—John W. Thierer. SYNOPSIS OF THE GENUS LYCORIS (AMARYLLIDACEAE) HSU PING-SHENG' Laboratory of Systematics and Evolutionary Botany Institute of Botany, Chinese Academy of Sciences Beijing 100093, PEOPLE'S REPUBLIC OF CHINA SIRO KURITA Laboratory of Phylogenetic Botany Faculty of Science, Chiba University 1-33 Yayoi-cho, Chiba 260, JAPAN YU ZHI-ZHOU and LIN JIN-ZHEN Hangzhou Botanical Garden, Taoyuanling Hangzhou, Zhejiang 310013 PEOPLE'S REPUBLIC-OP CHINA ABSTRACT Species of the East Asian genus Lycoris asily hybridized; di hological features occur frequently among them in nature and in culdivarion: This has cauiked: ues difficulty in identifications. Based on taxonomic and karyological investigations and crossing tests, 20 spe- cies are recognized of which four are regarded as hybrids: Lycors hice ardit, L. ees, i, x Hunt, W.L. 1963. Lycoris sanguinea and L. “cinnabarina.” P1. Life 19:50—-53 INARIYAMA, S. 1931. Cytological studies in the genus Lycoris (Prel. Notes). as ie: Tokyo INARIYAMA, S. 1932. Cytological studies in the genus Lycoris I. Conjugation of chromosomes in meiosis of L. albiflora Koidz. Bot. Mag. Tokyo 46:426-434. eae S. 1933. Natsuzusen ni okeru senshekutai-gun no Anarisisu. Rep. Jap. Sci. Congr. 8:3 1. dn Japanese) a. S. 1937. Karyotype studies in Amaryllidaceae I. Sci. Rep. T. B. D. Sect. B. 3:95— 113. INARIYAMA, S. 1944. Origin of Lycoris radiata and L. albiflora. Jap. J. Genet.20:87-88. InarryaMa, S. 1948. Origin of Japanese Lycoris. J. Jap. Genet. 23:15-1 INARTYAMA, S. 1951. Cytological studies in the genus Lycoris (1). Sci. Res. T. B. D. Sect. B. 6:74:100. INARIYAMA, S. 1952. Higanbana Zoku no Keito. Iden 6(10):12-15. anne INARIYAMA, S. 1953. Cytological studies in Lycoris. Rep. Kihara Inst. Biol. R ) , H. and M. Koyama. 1954. Offspring abiaued by self-pollination “of Lycoris radiata Herb.,a criploid species. Jap. J. Genet. 29:160-161. Kim, M. and S. Ler. see A taxonomical study on the Korean Lycoris (Amary J. Pl. Taxon. 21:123-139. Koyama, M. 1954. ae studies in the genus Lycoris (1). ee yer een on the hybrid of L. radiata x L. sanguinea. Ann. Rep. Doshisha Women’s Coll. 4 4l. een M. 1959. Offsprings rele vodiate obtained by artificial ee Ann. Rep. Doshisha Women’s Coll. 10:388-394. Koyama, M. 1962. Somatic ee in the genus Lycorzs. Ann. Rep. Doshisha Women’s .12:1-8. rani C.S. 1850. Enumeratio Plantarum 5:545—546. Kurita, S. 1980. Lycoris aurea versus L. traubii. J. Jap. Bot. 35:287-288 Kurita, S. 1985. Geoclinal change in the pollen ornamentation of Lycoris sanguinea Maxim. var. ee i? be Bot. Sone Kuri 8 iati tion on the karyotype of Lycorss, San eeee Il. Karyo- —_— lidaceae). Korean typea ree of te n taxa among which seven are native to China. Cytologia 52 Kurita, S. 1987b. Variation and evolution in the karyotype of Lycorzs, Ainsealidacene Il. Intraspecific variation in the karyotype of L. trawbii Hayward. Cytologia 52:117—128 Kurita, S. 1987c. Variation and evolution in the karyotype of Lycoris. Amaryllidaceae IV. In- traspecific variation in the karyotype of L. radiata (LV Hér.) Herb. and the origin of this trip- 2. loid species. ee 52:137-14 Kurita, S. 1988a. Variation and evolution in the karyotype of Lycoris, Amaryllidaceae VI. In al and/or intraspecific variation in the karyotype of L. sdanes — var. Riusiana and L. sanguinea Maxim. var. koreana (Nakai) Koyama. Cytologia 53:3 21. Kurita, S. 1988b. ae VII. Modes of karyotype peo within species < a a e trend of | in wk ae eee a 53:3 Kurita, S, 1989.1 iat t nthe eae of ari Amarlidaceae) Chro- mosomal variation in L. sanguinea ee Pl. Spec. Biol. 4 330 SIDA 16(2) 1994 Lamarck, J.B.A.P.M. de. 1783. Encyclopedie methodique. Botanique. 1:1 Ler, S. and M. Kim. eae ae study of some Lycoris species ne Korean J. Pl. Taxon. 17:147- Lit). 25 2.2.70 ee = (P.S. Hsu). 1990. ey aa and breeding of Lycoris. In: He, S.A. et al.(ed.), Proc. Int. Symp. Bot. Gard. 557-568. Liu, Y. and P.S. Hsu. 1989. A study on aes of the genus Lycoris. Acta Phytotax. Sin. 27:257-264 Liu, Y. and B.S. Xu (P.S.Hsu). ie Mechanism of sterility of diploid hybrid in genus Lycorss. Renner Shanghai 6:27— Margkawa, F. 1943. Serene ecculees plants to Japan proper. Acta Phytotax. Geobot. 13:274-279. Makino, T. 1943. On Lycoris albiflora. Acta Phytotax. Geobot. 13:17—19. Nakamura, T. 1978. Gyolopical studies on medical evan C hromosome number and karyo- types in some species of ee aires La Kramosomo II—10:271—281. (In Japanese) NisHikawa, K., Y. Furuta and H. E 1979 CG a of the chromosome evolution on the basis of muteleae DNA content ane total chromosome length in Lycorss. Jap. J. Genet. 54:387-396. NISHIYAMA, I. 1928. Reduction division in Lycor7s. Bot. Mag. Tokyo 42:509-5 13. Onwit, J. 1978. Flora of Japan, new ed. Sh — Co., ad Publishers Tokyo. (In Japanese) Repouté, P.J. 1822. Amary/lis aurea. Bot. Reg. 8:t.6 SATO, D. 1942. Karyotype alteration and nee in ies and allied families. Jap. J. Bot. 161. SPRENGER, C. 1888. ae as Lycoris Herb. (Della famiglia delie Amaryllidaceae). Bull. Soc. Tosc. Ortic. 13:323—32 Tak, K.H., $.C. Ko and 2 a ie A cytotaxonomic study on genus Lycorzs in Korea. ana Pl. Taxon. 17@Q):1 oa E. 1961. sae ee and cytological studies on artificial hybrids in the genus Lycoris 1. On based F, hybrid between L. sprengeri Comes and L. straminea Lind]. Bot. Mag. Tokyo 74:524-531 TAKEMURA, E. ee Monies and cytological studies on artificial hybrids in the genus Lycoris 11. Artificial hybrids among the different species having only rod-shaped chromo- somes. Bot. Mag. Tokyo 75:72-79. TAKEMURA, E. 1962b. Morphological and clan oe on — hybrids in the genus Lycoris I. An artificial hybrid having four V- ot. Mag. Tokyo 75:324— 0 TAKEMURA, E. 1965. Studies on Lycoris sanguinea complex. Proc. 30th Ann. Mat. Bot. Soc. Jap. (Oral report). Traus, H.P. 1957. Lycorts haywardit, L. houdyshelti er ae Pl. Life 1 48. Traus, H.P. 1958. Two new Lycoris species. Pl. Life 1 Traus, H.P. 1963. Lycoris xlajolla hybr. nov. Pl. Life ‘e a Traus, H.P. 1964. Lycoris x jacksoniana Traub, hybr. nov. Pl. Life 20:52 Traus, H.P. and H.P. MoLpenxke. 1949. Amaryllidac.: Tribe Amaryll. hse Pl. Life Soc. Cali- fornia WILLIAMS, M. 1983. Lycoris “cinnabarinum’ —A hybrid between Lycoris sanguinea and L. traubir? Pl. Life 39:95-99. Wors.ey, A. 1928. ees A garden review. Gard. Chron. ser. 3, 84:169, f. 7 Xu, B.S., J.Z. Lin, Z.Z. Yu and S.E Huan. 1986. An evaluation of the oe relation- ships of Lycoris based on eae viability and rate of seed set after crossing. Acta Genet. Sin. 13:36 6 Hsu, Synopsis of Lycoris real | Xu, Y., Z.B. Hu, X.L. Huanc and G_J. = 1982. New taxa of the genus Lycoris from China. Acta. Phytotax. Sin. 20:196-198. f.1- YOneEzZAWA, N. 1989. A new oo a a new form of Lycoris radiata (L Hérit.) Herb. J. Phytogeogr. Tax. 37:74. f.1-2. Yosuipa, M. 1972. Karyological ae on the genus Lycoris L. Sand-Dune Res. 18:20—36. NOTES ON THE GENUS CYBIANTHUS SUBGENUS CYBIANTHUS (MYRSINACEAE) IN COLOMBIAN AMAZONIA JOHN J. PIPOLY II Botanical Research Institute of Texas 9 Pecan Street Fort Worth, TX 76102-4060, U.S.A. ABSTRACT Studies of Myrsinaceae in Colombian Amazonia have revealed two new species in the genus Cybianthus subgenus Cybianthus: C. barbosae Pipoly and C. i Monae ee a are de- ] scribed and illustrated herein, and their phylogenetic RESUMEN Al estudiar las Myrsenaceae de la amazonia colombiana | bri los especies nuevas, pertenecientes al genero Gia ae bani Se ies Cybianthus barbosae Pi- poly y C. ruforamulus Pipoly, asf mismo se ilustran y se discute su parentesco. The genus Cybranthus Martius contains ten subgenera and perhaps 150 spe- cies (Pipoly 1987, 1991, 1992). Subgenus Cybianthus contains approximately 52 species, including 7 Peruvian ones as yet undescribed. The subgenus is dis- tributed in the Andes from Colombia to Bolivia, thence eastward through the Guayana Floristic Province (sensu Maguire 1979) to French Guiana and south- eastward through Amazonia to southeastern Brazil. The subgenus is defined by a vestigial pistillode in staminate flowers, basifixed anthers as wide or wider than long, and a staminal tube which is developmentally fused to the corolla tube; the stamens thus appearing epipetalous. During phytodiversity studies in the Amazon Basin of Colombia, concomi- tant with floristic and monographic studies of the genus Cybianthus for Flora Neotropica and Flora de Colombia, [have encountered the following new species, described herewith. Cybianthus barbosae Pipoly, sp. nov. (Fig. 1) Propter ramulos angulatos rufo-stellato- POR RERTOSOS folia gaara deren lobos calycinos nthe late ovatos vel suborbiculares ad apicem rotundatos ores quam longiores ad apicem obtusas vel truncatas e venezuelanum arcte affinis, sed ae eor ae dense rufo- stellato-tomentosis et glanduloso-lepidotis (non solum ae stellato-tomentosis), laminis embranaceis (non oe pete olis marginatis aliquantum Saale es (non canaliculatis (nec O a ~S ao 5—1 (nec 1.5—2.5) cm longis, calyce chartaceo (non carnoso) | mm longo, corolla eee (non carnosa) 1.8—2.5 (nec 1.6—1.8) mm ae corollae tubo Sioa 16(2): 333 — 339. 1994 334 Sipa 16(2) 1994 Fic. 1. Cybianthus barbosae Pipoly. A. Habit. B. Adaxial and abaxial leaf bases, showing margin- ate, pulvinate petiole, punctations and decurrent leaf base. C. Branchlet apex, showing tomen- tum. D. Branchlet close-up, showing stellate tomentum and translucent glandular-lepidote scales. E. Portion of inflorescence, showing floral habit, linear floral bracts, cylindrical pedicels, hyaline calyx lobe margins, and notched apex of corolla lobes. F. Staminate flower, showing glandular granules on corolla tube and lobes, quadrate corolla tube, and prominent ventra punctations on anther connectives. A-F. drawn from holotype. Prpoty, Notes on the genus Cybianthus 335 quadrato (nec tereti), lobis corollinis depresso-ovatis (nec late triangularibus), denique antheris : lial A SN : fs Shrub or small tree to 2 m tall; growth following Rauh’s architectural model. Branchlets angulate, 3-4 mm diam., with prominent, narrow longitudinal ridges, ensely rufous-stellate and translucently glandular-lepidote, the tomentum per- sistent. Leaves pseudoverticillate; leaf blades membranaceous, linear-lanceolate, very narrowly oblanceolate or rarely oblong, (10—)15—22(—27) cm long, 2—3.5 cm wide, apex long-attenuate or acuminate, the acumen 1.5—2.5 cm long, base long attenuate and cuneate, decurrent on the petiole, the blade bullate when fresh, subbullate when dry, nitid, minutely and sparsely orange-punctate above, below densely pellucid-punctate, sparsely black-punctate-lineate and with scat- tered stellate trichomes below, venation brochidodromous, costa slightly raised above, prominently raised below; secondary veins 18—24 pairs, deeply impressed above, prominently raised below, the margin irregular, undulating, flat; petioles marginate, 0.5—1 cm long, sparsely stellate above, densely stellate below, slightly pulvinate at base, glabrescent. Staminate inflorescence a simple, erect raceme 8— 11 cm long at maturity, the rachis and pedicels densely rufous-glandular-granu- lose and rufous-stellate-puberulent; inflorescence bract unknown; floral bracts membranaceous, linear, (1.5—)2—2.4 mm long, 0.2—0.3 mm wide, equal to or longer than the pedicel, apex subulate, glabrous above, densely stellate below, carinate, the margin entire, with scattered stellate hairs; pedicel cylindrical, thin, (1.5—)2—2.3 mm long, densely glandular-granulose and with a few scat- tered stellate hairs, glabrescent. Flowers chartaceous, 4-merous, green, pendent at maturity; calyx 1.2-1.4 mm long, deeply divided, tube 0.3—-0.4 mm long, lobes very widely obovate to suborbicular, symmetric, 0.9—1.1 mm long, 0.8— 1.1 mm wide, translucent, apex broadly rounded to truncate, thickened medi- ally, densely and prominently orange-punctate, abruptly constricted basally, the margin white to hyaline, irregularly erose, sparsely glandular-ciliate; corolla subrotate, 1.8—2.5 mm long, translucent, the tube quadrate in cross-section, 1— 1.1 mm long, densely rufous-glandular-granulose and rugose without, espe- cially on the surface alternate with the calyx lobes, densely translucent-glandu- lar-granulose within on surface alternate with anthers, the lobes depressed-ovate, (.8—1.4 mm long, 1.8—2.2 mm wide, apex very broadly rounded to truncate, with at least one apical notch, densely and prominently orange-punctate and rufous-glandular-granulose without, densely translucent glandular-granulose within except near base; anthers widely triangular, apparently sessile and epipe- talous (the staminal tube not easily discernible from corolla tube), 0.4—-0.6 mm long, 0.8—-1 mm wide, apex obtuse, base truncate, appearing truncate at anthe- sis, birimose, dehiscent by large terminal pores opening into wide longitudinal slits for ca. 3/4 anther length, connective adnate to corolla tube adaxially, densely red-punctate abaxially; pistillode conic, 0.3—-0.6 mm long, 0.2—0.4 mm diam., 336 Sipa 16(2) 1994 terete, hollow, apex translucent-glandular-granulose. Pistillate inflorescence unknown. Fruit unknown. Type: COLOMBIA. Amazonas. Municipo de Leticia: Parque Nacional Natural Amacayacu, Quebrada Agua Pudre, ca. 1.5 km NE of Quebrada mouth at rfo Amacayacu, permanent 25- hectare plot, 03°47'S, 70°15'W, 200-220 m, 18 Nov 1991 (stam. fl), J. Pipoly et al. 16450 (HOLOTYPE: COL: IsOTYPES: BRIT, MO, K, TEX, US). Paratypes: COLOMBIA. Amazonas. Municipio de Leticia: Parque Nacional Natural Amacayacu, eee Rea Pudre, ca. 1.5 km NE of Quebrada mouth at rfo Amacayacu, permanent 25-hectare ae 03°47'S, 70°15'W, 200-220 m, 18 Nov 1991 (ster.), J. Pipoly et al. 15790 (COL, FMB, MO); (ster.), J. Prpoly et al. 16509 (BRIT, COL, FMB, MO, TEX); (stam. fl), J. Pipoly et al. 16513 (BRIT, COL, F, FMB, MO, TEX); (stam. fl), J. Pipoly 16528 (BRIT, COL, FMB ) Distribution, ecology and conservation status: Cybianthus barbosae grows 1n tall wet forest on lateritic soils and is presumably endemic to the Leticia area. The canopy of the forest in which it occurs is approximately 35 m in height, and the domi- nants include Virola, Osteophloem and Iryanthera (Myristicaceae), Eschweilera (Lecythidaceae), Licania (Chrysobalanaceae), and Parkia (Mimosaceae). Cybianthus barbosae grows specifically along small streambanks on rotten logs, at 200-220 m elevation. In the area of a quantitative inventory in the park, I observed six individuals of C. barbosae per hectare, a relatively high frequency for a species of Myrsinaceae. Etymology: This species is dedicated to César Eduardo Barbosa Castillo, Direc- tor of the Herbarium, Unidad Investigativa Federico Menem (FMB), Mininstry of Environment, Colombia. César Barbosa is a specialist in Pithecellobinm and Rhizophoraceae of Colombia, and a keen student of the Amazonian flora. He first spotted Cybianthus barbosae growing next toa 40 meter tree of Sterigmapetalum (Rhizophoraceae) sp. nov Widely ovate or suborbicular calyx lobes, rounded or truncate apically, an- thers as wide as long and obtuse or truncate at apex, and rufous-stellate-tomen- tose, angulate branchlets indicate that Cybianthus barbosae is most closely related to C. venezuelanus, a taxon dispersed from Guyana west through Amazonian Brazil and Venezuela to Colombia, thence south through the Andes to Peru. The widely trangular anthers, quadrate corolla tube, glandular granules inter- spersed with rufous stellate hairs on the branches, marginate and somewhat pulvinate petioles, and the chartaceous and longer perianth, immediately sepa- rate C. barbose from C. venezuelanus. Cybianthus ruforamulus pen P. nov. oe 2) Ob folia late oblanceolata, fl i | bdel jon aneall lobos latiores quam longiores C. nit vm simulans, sed ab eo laminis chartaceis tang mem ipprepacels) rufo- hirtello- ae stellato- Be ie as _ 1.7-2 (non? Foss songs) rae Sbcordatis) plabris (nec citreo Daal lulc Prpoty, Notes on the genus Cybianthus STNG Us pany < eat Oe Ki Fic. 2. Cybianthus ruforamulus Pipoly. A. Habit, showing terete branchlets. B. Adaxial and ases, showing hirtellous tomentum. C. Ab abaxial leaf b axial leaf surface, showing biramose hirtellous trichomes. D. Branchlet apex, showing hirtellous tomentum. E. Portion of staminate inflorescence rachis, showing subsessile flowers, tomentum, prominent punctations that appear verrucose and hyaline, and tomentose margins. 338 Stipa 16(2) 1994 Tree to 8 m tall; growth following Rauh’s architectural model; all vegetative portions densely rufous hirtellous-tomentose, the trichomes biramose, branched just above base (Y-shaped), stiff, persistent. Branchlets terete, 8-10 mm diam. Leaves alternate; leaf blades chartaceous, oblanceolate, (34—)38-46(5 2) cm long, (9—)11—13(—16) cm wide, apex acute to short-acuminate, the acumen ().3—0.5 cm, base long-attenuate and cuneate, decurrent on the petiole, subbullate, sor- did above, pallid below, densely prominently and minutely black punctate above and below, more densely tomentose along the costa and veins below, venation brochidodromous, the costa impressed and hirtellous-tomentose above, prom1- nently raised and densely hirtellous-tomentose below, secondary veins (15—)20— 24 pairs, deeply impressed above, prominently raised below, the margin entire, flat; petiole deeply canaliculate, pulvinate, 2—3 cm long, densely tomentose, persistent. Staminate inflorescence a simple, erect raceme 8—10 cm long, the rachis, pedicels, and calyx densely tomentose; inflorescence bract unknown; floral bracts linear, 1.3—1.5 mm long, 0.2-0.3 mm wide, apex subulate, hyaline, densely long-hirsute abaxially, glabrous adaxially, margin entire; pedicel cylindrical, 0.4— 0.6 mm long, densely tomentose, persistent. Flowers chartaceous, 4-merous, greenish-maroon, translucent, pendent at maturity; calyx cotyliform, deeply divided, 1.1—1.4 mm long, tube 0.2—0.3 mm long, lobes widely ovate to subor- bicular, 0.9-1.1 mm long, 0.7—0.9 mm wide, apex obtuse, densely and promi- nently red punctate, the punctations crowded an aring verrucose throughout except at scarious margin, tomentellous alemar as with the calyx lobes, mar- gin irregular, entire, bifid-hirtellous; corolla subrotate, 1.3—1.7 mm long, the tube 0.3—0.5 mm long, the lobes depressed-ovate 0.9—1.2 mm long, 1.3—1.5 mm wide, apex broadly rounded, not notched, densely and prominently red- punctate and appearing verrucose and tomentose on areas between calyx lobes abaxially, densely glandular-granulose throughout within, the margin entire, hyaline, tomentose; anthers subsessile, subdeltate, 0.3-0.4 mm long, 0.4—0.5 mm wide, apex acute, base truncate, apically dehiscent by pores, the pores not birimose, opening ca. 3/4 anther length, the oo ee and promi- nently punctate dorsally; pistillode lageniform to obturbina 0.5—0.7 mm long, 0.1—0.2 mm, glabrous, style short, stigma selina Pistillate inflorescence: like the staminate but 6—9 cm long. Fruit sessile, calyx 1.3—1.5 mm long, the tube ca. 0.1 mm long, the lobes widely ovate, 1.2—1.3 mm long, 0.7—0.9 mm wide, apex obtuse, margin as in staminate. Fruit globose, 8-10 mm diam., orange, then purple, then black at maturity, rugose, inconspicuously black- punctate, exocarp thick, juicy. TYPE: COLOMBIA. Amazonas: along rio Yari, near mouth of Quebrada El Mochilero, ca. 00°30'N, 72°53'W, 120-200 m, ae 1986 (stam. fl), G. Galeano, J. H. Torres, J. Huitoto, G se ae 1105 aed pe: COL). F ARATYPES: COLOMBIA. ee Quebrada Aduche, 200 m, 5 Aug 1977 (fr), J. M. ee (COL). vend Acre: Estrada Alemanha, Cruzeiro do Sul, 14 Apr 1971 (fr), G. Pipoty, Notes on the genus Cybianthus 939 Prance et al. 11908 (IAN, INPA, K, MG, NY); 6 May 1971 (fr), PB Maas et al. P12737 (F, IAN, INPA, K, N Distribution, ecology and conservation status: Cybianthus ruforamulus is known from terra firme forests, near 200 m, in Amazonas, Colombia and Acre, Brazil. According to one collector (Galeano, pers. comm.), it grows along small brooks and other minor watercourses in primary forests, and may be considered an indicator of environmental quality. Etymology: The specific epithet describes the rufous of the branchlets, composed of forked trichomes unique within the genus. Cybianthus ruforamulus appears to be closely related to C. minutiflorus, a poorly known taxon from the rim of the Amazon Basin in Peru, but is easily recognized by the biramose-hirtellous tomentum of the branchlets, leaves and inflorescence, the shorter petioles, the chartaceous, greenish-maroon perianth, and the sub- deltate, glabrous anthers. The tomentum of Y-shaped hairs (Fig. 2C) is the first of its kind reported for the genus and should not be confused with the mal- pighiaceous trichomes unique in Cybzanthus to subgenus Triadophora. ACKNOWLEDGMENTS Research for this paper was completed at the Missouri Botanical Garden, where my studies in Amazonian biological diversity were supported by grants from the John D. and Catherine T. MacArthur Foundation. Cybianthus barbosae was found during my course in forest inventory and tropical tree architecture offered to graduate students of the Universidad Nacional de Colombia, Insti- tuto de Ciencias Naturales, in Parque Nacional Natural Amacayacu. I thank my Colombian counterpart, Agustin Rudas LI. for logistical support. I also owe special thanks to Oscar Pinto, former regional director of INDERENA, César Barbosa, Jaime Aguirre, Clara Inés Orozco, Pilar Franco, Adriana Prieto, Pablo Palacios, Santiago Duque, José Murillo, Roscio Cortés, Marta Gonzalez, Roberto Sanchez, and Hector Esquivél, for attending the course or facilitating arrange- ments. Linda Ellis executed the fine line drawings of the taxa. REFERENCES Macuire, B. 1979. Guayana, region of the Roraima Sandstone Formation. ao Larsen and Holm-Nielsen, eds. payee Botany. Academic Press. London. Pp. 2 Mez, Ce. 1902. Myrsinaceae. In: A. Engler, ed. Das PRanacnrcich 9 AV, 236): 187 Leipzig. PrpoLy, J. 1987. nee es revision eee genus C (Bentham) Pipoly (Myrsinaceae). M m. Ne w York Bot, Gard. eS. l- 76. 1991. Nuevas especies del género Cy g pha (My ) hae csle: Calombia: Clie. 16(78):257- 284, . The genus Cy hy, Z Cc ‘ph (My ‘ Vie } Ann Missouri Bot. Gard. 79:908—957. A NEW SPECIES OF CAREX (CYPERACEAE: PHAESTOGLOCHIN) FROM OKLAHOMA AND TEXAS; TYPIFICATION OF SECTION PHAESTOGLOCHIN, AND NOTES ON SECTIONS BRACTEOSAE AND PHAESTOGLOCHIN STANLEY D. JONES Botanical Research and Consulting Box 0717 Bryan, TX 77805-6717, U.S.A. ABSTRACT Carex perdentata, is described from Palo Pinto a: Texas. A key to the species of section Phaestoglochin occurring in Arkansas, Louisiana, Oklahoma, and Texas is included. Carex, sec- tion BOGEN TRLOIEERAS 1S ae Mis lecto-typified with a muricata. The South American section Bracteosae, nonymous with section Phaestoglochin, is considere RESUMEN Se describe Carex perdent. ita, seccion Phaestoelochin del condado de Palo Alto, Texas. Se incluye una clave para las especies de la seccién Phaest toglochin n que viven en Arkansas, Louisiana, Okla- homa y Tees, Se lec a la secci6n én Phat del género Cae con 1G. muvicata. La Bracteosae, considerada a 5n Phaestoglochin nejdera distinra Carex section Phaestoglochin {Sy = section Bracteosae sensu Mackenizie (1931), non Pax}, subgenus Vignea (P. de Beauvois ex Lestiboudois f.) W. Petermann was described from Europe by Dumortier in 1827. This section now includes 27 species. However, most of the taxa are native to the New World. Today, all of the European species have been introduced into North America. Their distribution in the New World is primarily in temperate North America. However one species, C. xalapensis C. Kunth, is found in México and northern Guatemala. Section Phaestoglochin is characterized by 1) an inflorescence with 3—25 sessile spicate branches, frequently with ten or less branches, the lower branches occa- sionally being compound in certain taxa (especially C. muehlenbergii C. Schkuhr ex C. Willdenow var. enervis W. Boott); 2) spikes androgynous, rarely pistillate, or having short staminate spikelets arising laterally from an androgynous or pistillate spike; 3) perigynia plano-convex or unequally biconvex, with the bod- ies of the perigynia more or less abruptly contracted into a beak; 4) achenes two- sided, either lenticular or slightly biconvexed; 5) style jointed with the achene; 6) stigmas two. Since section Phaestoglochin has never been typified, C. muricata Sipa 16(2): 341 — 353. 1994 342 Sipa 16(2) 1994 L. is here designated as the type for section Phaestoglochin. The typification of C. muvicata has been reviewed by Reznicek and Ball (1980 Section Bracteosae F. Pax differs from section Phaestoglochin by 1) a denser inflorescence with more spikes; 2) perigynia with a spongy area at the basal end of the ventral surface, occasionally on the dorsal surface as well; 3) and usually with wart-like bumps over the lower part of the ventral surface and sometimes across the dorsal surface of the perigynia. Kunth (1837) was the first to use the name Bracteosae as an infrageneric category, but without rank or description. Pax (1889) published Bracteosae at the sectional rank and provided a scant and am- biguous, but valid description. Pax did not cite Kunth, giving authorship of section Bracteosae solely to himself. Pax cited Carex cephalophora Muhlenberg ex Willdenow of North America as belonging to this section. This is undoubtedly where Mackenzie (1931) got his concept that the North American taxa that he treated belonged to section Bracteosae. Kiikenthal (1909) recognized the North American taxa that Mackenzie (1931) recognized as section Bracteosae as section Muthlenbergianae, a superfluous name for section Phaestoglochin, and recognized the South American taxa as belonging ina section distinct from those in North America. While researching the following complex of species in section Phaestoglochin; C. austrina (J. Small) K. Mackenzie, C. cephalophora H. Muhlenberg ex C Willdenow, C. /eavenworthii C. Dewey, C. mesochorea K. Mackenzie, C. muehlenbereti Schkuhr ex Willdenow var. enervis W. Boott, and C. muehlenbergii C. Schkuhr ex . Willdenow var. muehlenbergii, a new species of Carex (C. perdentata S.D. Jones) was discovered. Means, variances, standard deviations, and ranges were measured for nine populations consisting of 10 plants per population. Additional specimens were examined but no measurements were found to lie outside of the range recorded for the nine populations. The mean, plus and minus two standard deviations, for all parametric data is presented in the species description. If the ranges ex- tend beyond the mean, plus or minus two standard deviations, then the exten- sion 1s listed as parenthetical. Fruiting dates are based on a statistical mean, plus and minus two standard deviations. Ranges greater than and/or lesser than the two standard deviations are listed as parenthetical. Carex perdentata S.D. Jones, sp. nov. (Fig. 1) Plantae cespitosae: culmis fertilis (1 7—O—)26.8—74.2(—90.0) cm altis. Foliis (4—)5—9, laminis (18.0—)20.6-46.4(-55.5) cm ee 2.2-4.5(-4.6) mm latis; paginis adaxialibus papillosis; vaginis arctis. Inflorescentiis (S—)6—12(—14) androgyniis spicis, 13.5— 28.0) mm longis, (9.0-)9.5—14.5 mm latis. Perigynio (3.3—)3.4—5.2(-5.6) mm lense: 1, 9- 2. 7 2.8) mm lato, lenticularis, sine venae vel 1 —5(—8) tenuis venae ventraliter; sine venae vel 1—10(—11) renuis venae dorsaliter. Achaeniis 1.8—2.6(—2.8) mm longis, |.5—2.1(—2.2) mm latis, lenticularis, ovatis. Stigmatibus 2, rubellus- a brevis; antheris (1.3—)1.4-1.8(-1.9) mm longis. Jones, New species of Carex from Oklahoma and Texas 343 [oe Be ET ae RS ts Fic. 1. Carex perdentata (S.GG, Sine oS wee isotypes). A. Habit. B. Perigynium, E. Spke dorsal view. C. Perigynium, ventral view. D. Ache ther H. Sheath and ligule. Bar equals : cminA, - mm in E, 1 mminB,C,D, F/G, and H bract. FE. Pistillate scale. G. An- 344 Stipa 16(2) 1994 Habit perennial, cespitose; rhizomes short to long, dark brown to black, fibrillose; fertile culms (17 .O—)26.8—74.2(—90.0) cm tall, erect, stiff, 2.0—3.2 (-3.8) mm wide ca. 2 cm above rootstock, sharply triangular, scaberulous above, leafy on lower third, conspicuously exceeding leaves; /eaves with well-developed blades (4—)5—9 per fertile culm; hypostomous; blades (18.0—-)20.6-46.4(-55.5) cm long, 2.2-4.5(4.6) mm wide, erect-ascending to slightly arching, thick, light green, Hat, long-acuminate, margins antrorsely serrulate; abaxial surface without papil- lae or papillose distally; midvein raised, becoming increasingly more antrorsely scabrous distally; epidermal cells rectangular and raised above the braided pat- terned cell walls; adaxial surface papillose, papillae arising perpendicular from distal end of embedded clavate shaped epidermal cells (Fig. 2); stomata narrowly elliptic-oblong, paracytic with subsidiary cells mostly triangular with some semi- circular in shape; sunken below adjacent epidermal cells, restricted to intercos- tal zones on abaxial surface; sheaths tight around culm; dorsal sheath surface not septate-nodulose or rarely inconspicuously septate, papillose, pale green; ventral sheath surface whitish-hyaline, striate, papillose (at least proximally), deeply con- cave and more or less callused at apex of distal end, yellowish-brown tinged; ligule 0.4—1.9(—2.5) mm long, membranous, more or less linguiform, or infre- quently acute at apex; /nflorescences of (S—)6—-12(—14) androgynous spikes, 13.5— 25.3(—28.0) mm long, (9.0—)9.5—-14.5 mm wide, usually 0.5—2.5(—3) times as long as wide, but occasionally as wide as long; /ateral spike second from bottom (4.5—)4.7-7.9(-8.0) mm long; staminate flowers few, with ovate-lanceolate cus- pidate scales; /owest inflorescence bract 7.0-56.6(-105.0) mm long including awn, pistillace scale-like but with larger and conspicuously longer awns; mpper bracts pistillate scale-like but with longer awns, mid-point of awns ().1—0.5(—0.9) mm) wide; lateral branch, second from bottom (spike) (4.5—)4.7—7.9(-8.0) mm long; gap (internode) between the lowest two spikes (0.5—)0.7—3.5(—4.2) mm; pistil- late scales (1.6—)1.8—3.1 mm long, 1.4—2.0(—2.2) mm wide, 1-veined, rarely 3- veined, ovate, brownish or greenish-hyaline, narrower than and (excluding awn) about length of bodies of perigynia, apex usually conspicuously awned (rarely acuminate), awn (O—)0.1—3.6 mm long , green mid-stripe, 0.2—0.4(—0.5) mm wide; perigynia (3.3—)3.4—5.2(-5.6) mm long, 1.9—2.7(—2.8) mm wide, (2—)4— 1 9(—24) per spike, ascending or at maturity spreading, flattened-plano-convex, ovate, round-tapering and often slightly spongy at base, ventral perigynia sur- face veinless, or 1—5(—8) fine veins, dorsal perigynia surface veinless or 1—10(— 11) fine veins, submembranous, sharp-edged to base, serrulate above middle, tapering or abruptly contracted into a serrulate beak, beak bidentate, sutures conspicuous; /eeth (1.0—)1.4—1.7(-1.8) mm long, narrowly-triangular; achenes 1 .8—2.6(—2.8) mm long, 1.5—2.1(—2.2) mm wide, lenticular, strongly flattened, ovate (Fig. 3) to suborbicular, substipitate 0.1—0.3 mm long, minutely apicu- late, occasionally retuse at apical end; epidermal cells are nonisodymetric; single Jones, New species of Carex from Oklahoma and Texas 345 Ee: 2s Carex pendent SEM puichogtaph (S.& G. Jones 8349, isotype: TAES). Adaxial leaf & pet pendic ular from the distal end ofem bedded clavate shaped epidermal éelis. Bar equals 50 m. large central silica body and no satellites perched on edge of silica platform (Fig. 4); central body occasionally composed of two fused bodies; sty/e straight, short, slender, enlarged at base, jointed with achene; st/gmas two, reddish-brown, short; anthers (1.3—)1.4—1.8(-1.9) mm long; chromosome number unknown; fruiting (13 Mar-) 21 Mar - 1 Jun (-12 Jun); eco/ogy a facultative sciophyte, primarily in alfisols or inceptisols with sandy or sandy loam soil, sandstone outcrops, granitic outcrops, or thin soil over limestone, open mesic to submesic hardwood forests, or open hardwood-juniper forests, or woodlands in savannas in granite outcrops; elevation 175-525 m; distribution (Fig. 5), central Texas north to Oklahoma; economic importance forage value for livestock is low in palatability but is of use for wildlife, especially for rabbits, rodents, deer, and birds; it is also important in soil erosion, Typus: U.S.A. Texas. Palo ee o.: 2.3 miSon FR 4 from its jct. with FR 3137, S of Palo Pinto, mesic to submesic live oak 1 W/-facing slope with reddish sandy loam soil and sandstone, 23 Apr 1992,8. GG. ee 8349 (HOLOTYPE: MICH; tsorypes: BRIT, MO, OKL, SAT, TAES, TEX, US, VDB, WARM). ditional specimens: OKLAHOMA. Comanche Co: 1.3 mi N of Meers along W side of Re. 115, lightly shaded, moist soil along trail through deciduous forest, frequent, 15 May 1990, Naczi sy B (MICH). Cleveland Co.: low, damp, ground, 28 May 1939, Bebb i085 346 Stipa 16(2) 1994 7 Fic. 3. Carex perdentata, achene, SEM micrograph (S.GG. Jones 8349, isotype: TAES). A. Apical end. Bar Sonate 0.5 mm. (OKL). ees ia . o.: below Dwight Mission, 6 mi NE of Sallisaw; creek woods, 7 May 1955, Sar ees TEX a re 2.1 mi N on TX 95 from its jet. with FR 228, then 2.8 mi NE on a ie from its jct. with TX 95, mesic open to wooded roadside with graveley soil, 14 Apr 1990, S. & G. Jones 4340 (BRIT/SMU, ctb, MICH, MO, SAT, TAES, TEX, US DE, WAR Bexar Co.: San Antonio, 22 Apr 1911, Clemens and Clemens 386 (CAS). cent Co. 3 mi NW of Clifton, limestone gravel and silt, Meridian Creek bank, under trees, 16 Ape 1953 3, Shinners 14241 Cuma Burnet Cos: NE corner of US 281 and RR 2147 in Marble Falls, mesic roadside roade with granitic soil, 30 Mar 1992,S. GG. Jones 8213 ene ctb, MIC ff MO, SAT, TAES, TEX, US, VDB, WARM). Colorado W leaves 2.54.5 mm wide; most culms forming less JONES, New species of Carex from Oklahoma and Texas aot than a 50° angle with the ground, usually much g rreatly less; Thaly bract and/or ¢ exceeding the perigynia aes aks . Some dorsal leaf sheaths white or pale green with darker green veins and darker green septate-nod- ules, but some ‘Sheaths green mottled with white; y; widest leaves (3.0-)3.5-5.0(6. Om mm wide; most culms forming an angle greater than 70° with the ground; most bract and/or aaa scale awns not excoeing the 1] cs C. austvina ica) apex of the perigynia s y afew bract awns surpass the ee. C. aggregata 16. Perigynia 2.0-3.5 mm long, 1.3—2.3(—2.4) mm wide ...........0..4.. 19 19. Perigynia bodies ovate-deltoid; perigynia beaks 0.3— 0.7(-0.8) mm long with a single row of serrations, abruptly « ne from the apex of the perigynium; wid- est leaf blade 3.0(—4.0) mm wide; leaves per fertile culm 2—6(— aes width, ca. 2 cm above pee. 1.0-2.4(—3.5) mm wide; pistillate ae. 4—)1. (—2.5) mm long; pistilate scale awn 0-0.8(-1. mm long; dorsal leaf sheath frequently green mottled with e dot C. leavenworthti whi c bod (ee ee Sept Tile 19. Perigynia perig a s o as In mm long iD gdouble noe 8 ions, gradu widest leaf blade (1. 9)2.5—4.4 mm wide; ievesp per fertile culm oe )5—8; culm width, ca. 2 cm above rootstock, 1.7 3.2(—3.3) mm wide; pistillate scale 1.1—-1.7(-1.9) mm long: pistillate scale awn 0O—3.2 mm long; dorsal leaf sheath mostly green, infrequently green mottled with white dots . cephalophora 15. Abaxial, adaxial, or both leaf surfaces minutely pay tllose (sand paper-like), at least near distal end 20. Inflorescence capitate, 12-19 mm long, 9-14 mm wide; leaves conspicuously shorter than culm, (6.5—)8.4—21.0 (—23.0) mm long; ventral surface of perigynia veinless, dorsal surface veinless or rarely with 1—4 incomplete narrow veins (ca. 0.1—0.2 mm wide) C. mesochorea . Inflorescence short-oblong, oblong, or linear, (1 2.0—)13.5— 47 mm long, 6.0-18.0(—-28.0) mm wide, the central axis visible, at least between some spikes, usually the lowest two; leaves short or long, 11.3-46.4(—55.5) mm long; ventral surface of perigynia 0—15-veined, dorsal surface O—1 2-veined ........ 21 Ventral surface of perigynia with (S—)6—15 oe. broad veins (ca. 0.5 mm wide); dorsal surface v 0 2 broad Weinsnis oases C. mue iene ‘eit var. mueblenbergit 20 a 21. Ventral surface of perigynia with 0—6(—8) narrow veins (ca. O mm wide); dorsal surface with 0-1 1(—14 narrow veins ope Sipa 16(2) 1994 22. Pistillate scales 3.0-4.2(4.3) mm long, (1.0—)1.6— 2.6(—3.0) mm wide; segs stripe 3-veined, rarely 1- veined: culms oe gan angle of 50° or with the ground C. austrina Pistillace scales (1 5—)1.8—3.1 mm long, (1.0—)1.2— 1.8(—2.2) mm wide; mid- stripe 1-veined, occasion ally 3-veined; culms usually forming an angle of 70° or more with the ground 23 23. Beaks of perigynia 0.2—0.6(-1.0) mm long, abruptly arising from apex of perigynium; peri- gynia broadly ovate, (1.5—)2.5—3.8 mm long; dorsal leaf sheaths frequently green with mottledwhite dots; plants wide spread, from Texas east to Georgia and north to Can bho No anada adie ‘gui Var. enervis 23. Beaks of perigynia (1.0—)1.4—1.7(-1.8) mm long, tapering from shoulders or occasionally abruptly ene from apex of perigynia; peri- bo Oo )mm ions most dorsal leaf sheaths infrequently mottled with white dots; plants restricted to central Texas north to Oklahoma ............0..4.. C. perdentata This treatment does not recognize varieties of Carex gravida. However, fut- ther research may prove that infraspecific taxa are warranted [1.e., C. gravida var. /wnelliana (Mackenzie) EJ. Hermann], but until that time, I have opted for a more conservative approach. In floras where C. gravida, C. aggregata, and C. sparganioides occur together, they have classically been separated from other members of section Phaestoglochin by having baggy sheaths. However, I have encountered specimens of C. gravida and C. aggregata, both in the field and on herbarium sheets, that have tight leaf sheaths. It is possible that a single reces- sive gene is responsibile for tight leaf sheaths, or developmental through ontog- eny. Another plausible explanation provided by A.A. Reznicek [((MICH) pers. comm.|} is that the individuals with tighter leaf sheaths are growing in less than optimum habitat. Regardless, C. gravida and C. aggregata usually have baggy leaf sheaths but can have tight leaf sheaths so they key out under baggy and tight leaf sheaths. The classical spelling of C. muhlenbergii should be corrected to C. muehlenbergii. This was pointed out by Peter Ball {(TRTE) pers. comm.}. Willdenow’s original spelling was C. méhlenbergii with a diacritical sign (an umlaut over the “u”). Greuter et al. (1988), Article 73.6 state that diacritical signs are not used in Latin plant names. In names which are drawn from words in which such signs appear, the signs are to be suppressed with the necessary transcription of the letters so modified (i.e., ii becomes ue). Carex perdentata’s closest putative relative is C. mesochorea. Carex perdentata differs in having longer perigynia with their bases slightly spongy, slightly nar- Jones, New species of Carex from Oklahoma and Texas 5) rower and conspicuously longer leaves, usually longer inflorescences, more spikes per inflorescence but with fewer perigynia per spike, and for the most part, longer anthers. The teeth of the beaks of C. perdentata are noticeably longer and spreading at maturity in contrast to the short and usually straight teeth of C. mesochorea which rarely reach 1 mm long. The species epithet “perdentata” refers to the conspicuous teeth on the beaks of the perigynia. ACKNOWLEDGMENTS Tam grateful to Gretchen D. Jones (USDA, AWPMRU), my wife, for re- viewing this manuscript, for her tireless enthusiastic help in the field, but most important for her encouragement, understanding, and patience throughout this research. I thank A.A. Reznicek (MICH) and J.K. Wipff(TAES) for their manu- script review. Lam grateful to Paul A. Fryxell and A.A. Reznicek for correcting my Latin description, to Keith Westover for providing the illustration, and to Peter Ball (TRTE) for pointing out the orthographic correction of Carex muehlenber git. om REFERENCES Dumortigr, B.C. 1827. Florula Belgica, Operis Majoris, Prodromus. Staminacia, Tournay (J. Casterman). P. 146. GREUTER, W., ET AL. 1988. Chairman. International Code of Botanical Nomenclature, ga by the Fourteenth International Botanical Congress. Koeltz Scientific Books, a G. 1909. eee Caricoideae. Das Pflanzenreich, IV, 20 (Heft 38) 396-353, ngelmann, Leipzig, Germ me CS 1837 ee ne synoptica dive Enumeratio Cyperacearum. Vol. 2. Enumeratio Plantarum. Stutgardiae et Tiibingae. Pp. 378-381. Mackeniziz, K.K. 1931. North American Flora. 18. Cyperaceae, tribe 2, Caricae. New York Botanical Garden, Bronx, N Pax, F. 1889. Cyperaceae In: Engler und = — Die natiirlichen Pflanzenfamilien. W. Englemann, pee Germany. Pp. 98-13 Reznicek, A.A. and P.W. BALL. 1980. ee taxonomy of Carex section alee in North American and ‘North of exe. Contr. Univ. Michigan Herb. 14:153—2 NOTES ON CAREX (CYPERACEAE), WITH C, GODFREYI NEW TO ALABAMA AND C. COMMUNIS AND C. SCOPARIA NEW TO MISSISSIPPI CHARLES T. BRYSON USDA, ARS Southern Weed Scrence Laboratory Stoneville, MS 387706, U.S.A. JOHN R. MACDONALD Department of Biological Sciences Massisstppt State University Massassippi State, MS 39762, U.S.A. RANDY WARREN Department of Biological Sciences Massesstppi State University Massissippi State, MS 39762, U.S.A. ABSTRACT Field explorations have yielded C ifreyi new to Alabama and C. communis and C. scoparia new to Mississippi, U.S.A. Additional range extensions are presented - G: oo var. Dae and C. oklahomensis in Mississippi. Locality and h RESUMEN I ploraciones de campo han dad resultado Carex godfreyi nuevo para Alabama, y C. communis y C. scoparia nuevos para Mississippi, U.S.A. Se presentan extensiones de areal para C. bicknellit var. opaca y C. oklabomenis en Mississippi. Se indican la localidad y el habitat de todas las especies citadas. INTRODUCTION In preparing a synoptic treatment of Carex as a contribution to the Flora of Mississippi Project, the senior author has continued to examine herbarium speci- mens and conduct field surveys for species with potential to occur within Mis- sissippi and other southeastern states in the U.S.A. The authors have also con- tinued assessment of population size, distribution, and habitat requirements of recently reported species C. bicknellii Britton var. opaca FJ. Herm. and C. oklaho- mensis Mack., especially because these may have potential to become weedy. This mauscript adds to the knowledge of Carex that has been reported in recent years (Bryson 1984a; Bryson & Jones 1990; Bryson et al. 1991; Bryson et al. Spa 16(2): 355 — 361. 1994 356 Sipa 16(2) 1994 1992; Bryson & Carter 1994; Carter et al. 1990; Morris & Bryson 1986; Naczi & Bryson 1990). As previously discussed, the flora of Mississippt is still poorly known in comparison with several adjacent states (Bryson & Carter 1994). Lowe's Plants of Mississippi (1921), although outdated, must continue to serve as a base- line for the general floristic work in the state. The terminology of physiographic regions or resource areas in Mississippi follows Lowe (1921) as adapted by Morris (1989). Herbarium abbreviations follow Holmgren et al. (1990), except ctb, MMNS, and USMH (pers. herb. of Charles T. Bryson; Mississippi Museum of Natural Science, Jackson; and Uni- versity of Southern Mississippi, Hattiesburg, respectively). Carex communis Bailey var. communis belongs to the section Acrocystis (Rettig 1988) and is known from dry to mesic rich woods in mountains, steep ravines, rocky ledges, and calcareous soils from Quebec to Ontario and Minne- sota, south to Georgia, Tennessee, Alabama, and Arkansas (Mackenzie 1931; Naczi 1993; Radford et al. 1964; Rettig 1988; Steyermark 1963). The follow- ing are the first collections of C. communis from Mississippi. Voucher specimens: U.S.A. MISSISSIPPI. Monroe Co.: ca. | mi N Lake Monroe, T13S R7E S10 SW/4, 12 May 1992, MacDonald 4561 (ctb, IBE, MICH, others to be distributed); 28 Apr 1993, MacDonald 5928 & Warren (ctb, IBE, MICH, SWSL, others to Bs distributed); 19 May 1994, Bryson 13051 & MacDonald (ctb, SWSL, others to be di At this site in the Tennessee Hills Region, C. communis var. communis was found at an elevation of about 75 to 90 m ina mesic beech-maple-oak forest near the crest of a N- to NE-facing slope on Ruston and Cuthbert soils. It was associated with Acer barbatum Michx., Actaea pachypoda EMiott, Adiantum pedatum L., Asimina triloba (L.) Dunal, Arabis canadensis L., Carex abscondita Mack., C. vaphalinbnie Willd., C. corrugata Fernald, C. blanda Dewey, C. gracilescens Steudel, C. rosea Willd., C. laxiflora Lamarck var. serrulata FJ. Herm., C. wi//denowi: Willd., C oe vireimanum L., Decumaria barbara L., Fagus grandifolia Ehrhart, Geranium macilatum L., Hepatica americana oC ) Ker, on lera benzoin (L.) Blume, Morus rubra L., Obolaria virginica L., ee longistylis (Torrey) DC., Panax quinquefolium L., Quercus alba L., Rhamnus caroliniana Walter, Thelypteris hexagonoptera (Michx.) Weath. Carex godfreyi Naczi occurs in wet hammocks, swamps, and floodplains in the coastal plain from southern North Carolina southward to the central penin- sula of Florida and west and to southwestern Georgia and nearby portions of the Florida panhandle. It inhabits shaded, mesic to wet areas in calcareous muck or sandy loam soils (Naczi 1993). Carex godfreyi belongs to a complex of species that include C. amphibola Steudel, C. corrugata Fernald, and C. grisea Wahlenb. Bryson, MACDONALD AND WARREN, Notes on Carex 357 Carex godfreyz is distinguished from the preceding species by its more loosely cespitose habit, its leaf sheaths and cataphylls with more extensive purple-red pigmentation, and its narrower leaf bases. The following citations are the first report of C. godfreyi from Alabama. Voucher specimens: U.S.A. ALABAMA. Houston Co.: ca. 9 mi S of Dothan, vicinity of Big Creek; on road to Madrid at 0.2 mi SE of Hwy US 231, 18 May 1992, MacDonald 4594a (ctb, IBE, KNK, others to be distributed); Chattahoochee State Park, area where Irwin Mill Creek meets Hwy AL 95, 5 May 1993, MacDonald 5975 ( ctb, IBE, KNK, SWLS, others to be distributed). At the site about nine miles south of Dothan, C. godfreyi grew in a Fagus grandifolia-Magnolia grandiflora L.-Magnolia virginiana L.-Pinus glabra Walter- Quercus lauvifolia Michx.-Quercus michauxii Nutt. forest above Big Creek. Addi- tional associates included Carex abscondita, C. complanata Torr. & Hook., C. debilzs Michx., C. leptalea Wahlenb., C. styloflexa Buckley, Gordonia lasianthus (L.) El- liott, Uvularia floridana Chapm. At the Chattahoochee State Park site, C. god- freyi grew in an Acer barbatum-Magnolia grandiflora~-Magnolia virginiana-Persea palustris Raf.) Sarg.-Quercus hemisphaerica Bartram forest above Irwin Mill Creek ona clay soil. Associates present at this site also included Aristolochia serpentaria L., Bumelia lanuginosa (Michx.) Persoon, Conopholis americana (L.) Wahl., Cornus stricta L., Dryopteris ludoviciana (Kunze) Small, [ex cassine L., I. montana Torr. & Gray, Lyonia lucida (Lamarck) K. Koch, Osmanthus americana (L.) Benth. & Hook., Pedicularis canadensis L., Ponthieva racemosa (Walter) C. Mohr, Rhododendron canescens (Michx.) Sweet, Sanicula canadensis L., 8. marilandica L., Thelypteris palustris Schott, and Viburnum obovatum Walter. Carex atlantica Bailey subsp. capi/lacea (Bailey) Reznicek, C. leptalea, Cicuta mexicana Coult. & Rose, Decodon verticillatus (L.) Elliott, Nasturtium microphylum (Boenn.) Rchb., Rosa palustris Marshall, and Zizaniopsis miliacea (Michx.) Doell & Asch. grew in wetter areas nearby. Both collections are from the Doughtery Plain District of the East Gulf Coastal Plain of Alabama (Sapp & Emplaincourt 1975). Carex scoparia Willd. var. scoparia is known from open areas in swampy or wet river bottoms, valleys, prairie swales, upland prairies, margins of sink-hole ponds, and roadside ditches from Newfoundland to British Columbia, south to South Carolina, Tennessee, Arkansas, Oklahoma, New Mexico, and Oregon (Kolstad 1986; Mackenzie 1931; Radford et al. 1964; Steyermark 1963). This collection is first of C. scoparia from Mississipp1. Voucher specimens: U.S.A. MISSISSIPPI. Lafayette Co.: Presbyterian Camp Hopewell, ca. 6 mi NE Oxford, T8S R2W S9, 28 May 1994, Bryson 13910 (ctb, IBE, MICH, SWSL, others 1 ted) Jiot At this site in the Central Hills Region, C. scoparia var. scoparia grew inan open area on coarse sandy soil above a small spring-fed lake in association with 358 Stipa 16(2) 1994 Carex albolutescens Schwein., C. festucacea Willd., C. laevivaginata (Kiik.) Mack., C. longii Mack., C. lurida Wahlenb., C. triangularis Boeck., C. vulpinoidea Michx., Fuirena squarrosa Michx., Rhynchospora glomerata (L.) Vahl, Xyris torta J.E. Smith. OTHER NOTEWORTHY COLLECTIONS Carex bicknellii Britton var. opaca EJ. Herm. was described from three collections of Delzie Demaree from river terraces in Lonoke and Prairie coun- ties, Arkansas (Hermann 1972). Carex bicknellii var. opaca was first found in Mississippi at a site in the Black Prairie Region of Mississippi where C. ok/aho- mensis Mack. was first collected in the state (Bryson & Carter 1994). The follow- ing citation is the second report for Mississippi. Voucher specimens: U.S.A. MISSISSIPPI. Itawamba Co.: N of Dorsey, SW of junction of Hwy US 78 and Dorsey-Fawn Grove Road Exit, T9S R7E $25, 25 May 1994, Bryson 13802 (ctb, MICH). Three plants of C. bicknellii var. opaca were observed at this site in close asso- ciation with C. oklahomensis Mack. Each plant was depauperate compared to those from a Lee County site observed in 1993 (Bryson & Carter 1994), but plant size was also reduced at the Lee County site in 1994, possibly due to dry conditions. In Itawamba County, C. bicknellii var. opaca was also associated with C. bushti Mack., C. cherokeensis Schweinitz, C. complanata Torr. & Hook., C. glau- codea Tuck., C. longii Mack., C. vilpinoidea Michx., Cyperus echinatus (L.) Woods, C. lancastriensis Porter in Gray, C. odoratus L., C. strigosus L., Festuca pratensis Huds., Eleocharis obtusa (Willd.) Schult., Fimbristylis autumnalis (L.) Roem. & Schult., and RAynchospora capitellata (Michx.) Vahl Carex oklahomensis Mack. is known in Mississippi only from Lee County, just east of Tupelo (Bryson et al. 1992). The following data are for an additional site in Lee County and for the first stations in Itawamba and Lowndes counties. Voucher specimens: U.S.A. MISSISSIPPI. Itawamba Co.: N of Dorsey, SW of jct. of Hwy US 78 and Dorsey-Fawn Grove Road Exit, T9S R7E 825, 24 May 1994, Bryson 13803 (ctb, MICH). Lee Co.: Tupelo, SW of jct. Hwy US 78 and Veterans Blvd. (=old Saltillo Road), TOS RGE 821, 24 May 1994, Bryson 13806 (ctb, MICH). Lowndes Co.: ca. 5 mi N of Columbus, Columbus Air Force Base, T16S R18W S30 SE/4 of E/2, oS 1993, Warren 2469 (ctb, IBE, MICH, MISS, MMNS, SWSL, USMH, VDB, VSC, additional sf to be distributed); 19 May 1994, Warren 2595, MacDonald & Bryson nea MICH, skin spunea to ae a uted); MacDonald 7197, Warren & Bryson (ctb, IBE, MICH, additio uted); Bryson 13680, Warren €-MaDinald BRITISM oth ai = MICH, MISS, MMNS, USMH, VDB, VSC, additional specim The additional collection from Lee einen and the one from Itawamba County are both along highway US 78. As previously speculated (Bryson et al. 1992), it is likely C. o&/ahomensis was introduced with blown hay for erosion control BrysON, MACDONALD AND WARREN, Notes on Carex 359 along the construction corridor of Highway US 78 in Itawamba and Lee coun- ties. Associates at both sites are the same as those previously discussed for C. bicknellii var. opaca. It is likely that C. ok/ahomensis was introduced at the Lowndes County site in much the same manner as the Itawamba and Lee county sites or as a contaminate of grass seeds planted for erosion control. At the Lowndes County site, C. oklahomensis grew west of Independence Boulevard along and south of a small drainage ditch through a pine-hardwood forest. In 1993, C. oklahomensis plants were 1.4 to 1.7 m tall. At the time the specimens were collected in June, plants were drooping from the weight of the infructescences. Plants were only 1.0 to 1.2 m tall in 1994; the reduced plant height may have been caused by increased competition from grasses and drier than normal con- ditions. Associates at the Lowndes County site are Carex frankii Kunth, C. longit Mack., C. triangularis Boeck., Eleocharis obtusa (Willd.) Schult., Festuca pratensis, ae Glyceria septentrionalis el In Mississippi, C. oklahomensis in- habits open sites with mucky clay soil, which are transitional between the Black Prairie and Tennessee Hills regions. ACKNOWLEDGMENTS Thanks are expressed to A.A. Reznicek (MICH) who confirmed the determi- nation of Carex bicknellii var. opaca, C. oklahomensis and C. scoparia, R.F.C. Naczi (KNK) who confirmed the determinations of Carex godfreyz; and Richard Carter (VSC), S.D. Jones (TAES), R-EC. Naczi, and A.A. Reznicek who reviewed the manuscript. REFERENCES BRYSON, C.T. 1984a. Notewot thy additions to the Carex f Mississipy 1. Castanea 49: 44, Bryson, C.T. and S.D. Jones. 1990. Caves comosa (Cyperaceae) new to MESisipat Sida 14:311- Bryson, C.T., S.W. Rosso, and R.EC. Naczi. 1991. Carex baltzelli ey new to Missis- sippi with notes on Carex picta and Carex impressinervia in Mississippi. Sida 14:493—499. Bryson, C.T., R.EC. Naczi, and 8. McDanirL. 1992. Notes on eee ot of Carex (Cyperaceae) from the southeastern United States. Sida 15:125—135 Bryson, C.T. and R. Carter. 1994. Notes on Carex, Cyperus, and Kyllinga Cype ae in Mis- sissippi oie records nae species are uorepontes 6) eb state. Sida 16:175—186. Carter, R.M.W. Morris, and C.T. Bryson. 199 interesting vascular lants from the Delta Region of Mississippl ae tanea 55:40-55. HERMANN, F.J. 1972. A new variety of Carex bicknellii from Arkansas. Sida 5:49. HoiMGren, P.K., N.H. Hoimcren, and L.C. Barnett (eds.). 1990 Index herbariorum. Part 1: the herbaria of the world, 8th ed. New York Botanical Garden, Bronx ee A.O. 1986. Cyperaceae. In: Flora of the Great Plains. R.L. Mc Gregor M. Barkley, Brooks, and E.K. Schofield, eds. University Press of Kansas. pp. 1059-1113. oe EN. 1921. Plants of Mississippi. Mississippi State Geol. Surv., Bull. > l= Q SS % \ Fic. 1. Hygrophila polysperma, A. Distal portion of submersed shoot with abundant adventitious roots. B. Terrestrial form, drawn to same scale. C. Flower. A and B drawn by Amy L. Mahloch from live material; C redrawn after an illustration provided by the Center for Aquatic Plants, University of Florida, Gainesville. 368 Stipa 16(2) 1994 Hygrophila polysperma (Roxb.) T. Anderson, J. Linn. Soc., Bot. 9:426. 1876. Justicia polysperma Roxb., Fl. Ind. 1:120. 1820. Hemidelphis polysperma (Roxb.) Nees in Wall., Pl. Asiat. Rar. 3:30. 1832. Perennial rhizomatous terrestrial or aquatic herbs to 1.5 m tall. Stems ascen- dant or rarely erect, more or less 4-angled, puberulent to glabrate, with abun- dant elongate or rarely rounded cystoliths in the epidermis. Leaves opposite, broadly elliptic to oblanceolate, acute at apex, attenuate to a subpetiolar base, minutely denticulate to entire, 7—65 mm long, 2-10 mm wide, mostly gla- brous but those subtending the flowers hispid, especially on the margins, the abaxial and adaxial surfaces with abundant elongate cystoliths. Flowers solitary in the axils of uppermost leaves, sessile. Bracts narrowly lanceolate, 4-5 mm long, herbaceous. Calyx equally 5-lobed, the lobes scarious-margined, 4—5 mm long, united basally, hispid. Corolla bluish-white (yellowish in dried specimens), 5—6 mm long, the upper lip 2-lobed, the lower lip 3-lobed, puberulent. Fertile stamens 2, included, filaments glabrous, anthers 2-celled, ca. 1 mm long. Ovary hispid distally, the style ca. 3 mm long, sparingly hispid, the stigma flattened, ca. 0.3 mm long. Capsule 6—7 mm long, mostly glabrous but with a few distal hairs. Seeds flattened, round, ca. 0.8 mm diameter. Blooming mostly Sep—Oct. In the San Marcos and Comal rivers, H. polysperma is most often found grow- ing completely submersed, although we have frequently encountered both emer- gent and terrestrial individuals along the upper San Marcos River. Submersed individuals have longer stems (0.3—1.5 m tall) with elongate internodes (20-54 mm long), relatively large oblanceolate leaf blades (17-65 mm long), and pro- duce abundant adventitious roots at the upper nodes. Terrestrial individuals are of shorter stature (10-20 cm tall) with shorter internodes (2-16 mm long), smaller elliptic leaves (7—25 mm long), and bear roots almost exclusively on the rhizome. We have observed flowers and fruits primarily on terrestrial individu- als and, occasionally, on emergent shoots of partially submersed plants; in both cases the distal leaves subtending the flowers are marginally hispid. These ob- servations accord well with those of Sculthorpe (1967) who cited numerous examples of normally terrestrial plants (e.g., species of A/isma L., Bacopa Aublet, Campanula ., Gratiola L., Nomaphila Bl. Renn ulus L., and Rotala L.) that can grow sory submerged even in deep water, leading to their common use as ornamental plants in aquaria; submersed individuals of such species were usu- ally found to be characterized by elongation of the leaf blade, loss of pubescence, and sterility. Two previous studies of the aquatic macrophytes of the San Marcos River failed to document the occurrence there of H. polysperma. Lemke (1989) incor- rectly identified collections of this species as H. /acustris, while Staton (1992) misidentified her collections as Ludwigia repens. Hygrophila lacustris, a native of the southeastern U.S., occurs in only a few counties in southeast Texas (Brazoria, ANGERSTEIN AND LEMKE, Hygrophila polysperma from Texas 369 Chambers, Fort Bend, Hardin, Harris, Jackson, Montgomery, Orange, and Walker counties), where it grows as a terrestrial or emergent plant along muddy stream and pond margins (Wasshausen 1966, Correll & Correll 1975). The plants are typically erect herbs to 80 cm tall with leaves mostly 5—12 cm long and flowers borne in distinct axillary clusters along the length of the stem. Hygrophila polysperma, in contrast, is usually found submersed, the leaves of the aerial shoots are typically 7-25 mm long, and the flowers are solitary in the axils of the uppermost leaves. When first introduced to the aquarium plant market, H. polysperma was though to bea species of Ludwigia L. and was given the common name “oriental ludwigia” (Innes 1947). Vegetatively, the terrestrial shoots of H. polysperma are very similar to those of L. repens, a native species also known from the San Marcos and Comal rivers; both species have small, opposite, elliptic leaves and short internodes. Even in the absence of reproductive structures, however, the two species are readily separable by nodal morphology. The petioles of L. repens are subtended by a pair of minute, glandular stipules less than 1 mm long, while the connate leaf bases of H. polysperma are exstipulate but bear a number of setiform hairs to 1.5 mm long (Fig. 2). Hyerophila poly sperma is the second adventive aquatic member of the Acan- thaceae to be reported from Texas in recent years. Ramamoorthy and Turner (1992) documented the occurrence of Nomaphila stricta (Vahl) Nees, another Malaysian species, from San Felipe Springs in Val Verde County and surmised that this species was also introduced by aquarists. 1G, 2. Comparison of nodal morphology in Hygrophila and Ludwi — A. Hygrophila polysperma. B. Ludwigia repens. Drawn by Amy L. Mahloch from live material. 370 Stpa 16(2) 1994 aa Specimens examined. TEXAS. Comal Co.: Comal River, Landa Park, City of New Braunfels ca. 100 m upstream of Pecan Island, mid-channel, in |—2 ft of water, 21 Feb 1994, Angerstein 94-1 (BRIT, SWT, tees Comal River, Landa Park, City of New Braunfels, in spring run underneath Landa Dr., 23 Feb 1994, Angerstein 94-2 (SWT). Hays Co.: San Marcos River below Southwest Texas Spare University campus, 30 Nov 1969, Tabler s.n. (SWT); shoreline sandbar formed at mouth of Sessoms Creek, 22 Nov 1975, Litchfield s.n. (SWT); San Marcos River downstream from Clear Springs Apts. een 5 - ee Litchfield s.n. (ST); South- west Texas State University raceway in gravel sub 8 Aug 1976, Litchfield s.n. (SWT); San Marcos River at Thompson's Island below county fe ei é os ar 1991, Staton s.n. (SWT), San Marcos River upstream from Purgatory Island, 26 Apr 1991, Staton s.n. (SW'T); San Marcos River at University Blvd., 26 Jun 1991, Brerner 91-71 (TEX); Rio Vista Park, Purgatory Is- land, City of San Marcos, 30 Mar 1994, Angerstern 94-5 (SW’T, TAES); San Marcos River, Lions’ ark area, San Marcos, 30 Mar 1994, Angerstein 94-6 (SW/T); San Marcos River, Sewall Park, Sn Marcos, along edge of concrete channel, 30 Mar 1994, Angerstein 94-7 (SWT, TEX); South- west ee State University campus pond beside Freeman Bldg., 30 Mar 1994, Angerstein 94-8 (SWT. : AES); Southwest Texas State University campus pond directly 1 in front of J. C. Kellam Bldg., - Mar 1994, Angerstern 94-9 (BRIT, SW’T). — ACKNOWLEDGMENTS We would like to thank the Parks and Recreation Department of the City of New Braunfels, Texas, for permission to collect in Landa Lake, the Center for Aquatic Plants at the University of Florida (Gainesville) for assistance in locat- ing certain literature items, and the curators of TEX-LL and USF for loans of specimens. REFERENCES Borts, PS. rd M. LAWRENCE, B.W. Wirz and C.W. Kovacn. 1990. Plasticity in morphology, proxi and energy content of Hygrophila polysperma (Roxb.) Anders. Aquatic Bot. 361 207- 214 CorreELt, D.S. and H. B, CorreLi. 1975. Aquatic and wetland a of the southwestern United States, 2 vols. Stanford University Press, Stanford, Californ Goprrey, R.K. and J.W. Wooren. 1981. Aquatic and wetland ae of the southeastern United States. eras University of Georgia Press, Athens. Hannan, H.H. 1969. The introduction and esa blishmnent of Ceratopteris in Texas. Amer. Fern Jeo; Innes, W.T. 1947. Hygrophila, a new aquarium plant. Aquarium (Brooklyn) 16:30—31. Kovacnu, C.W., J.P. IRDZIEL, R. Bow /MAN, J. WAGNER and J.M. es 1992. The effect of stress and di on proximate compo , allocation of production, te sis, respiration, and chlorophyll levels in apa sperma (Roxb.) Anders. (Acanthaceae). eae Exp. Bot. 32:479-486. Lantz, P.S. 1993. Florida’s most invasive species. Palmetto 13:6—7. LEMKE, b. E. 1989. Aquatic macrophytes of the upper San Marcos River, Hays Co., Texas. Souda. Naturalist 32:289-291. Les, D.H. and R.P. WUNDERLIN. 1981. Hygrophila polysperma (Acanthaceae) in Florida. Florida Sci. 44:189-192 Lona, R.W. 1970. The genera of Acanthaceae in the southeastern United States. J. Arnold Arbor. 51:257—309 ANGERSTEIN AND LEMKE, Hygrophila polysperma from Texas oil RamMamoortny, T.P. and B.L. TurNER. 1992. Nomaphila stricta (Acanthaceae), a Bee discerned aquatic weed in Texas, and the first report for North America. Sida 15:115—11 SCULTHORPE, C.D. 1967. The biology of aquatic vascular plants. St. Martin's ae Ney York, NY. SPENCER, W. and G. Bowes. 1985. Linnophila and = “ophila: a review and aes aSSeSS- ent of their weed potential in Florida. J. Aquatic Pl. Mana pea 23 oe rATON, L.L. 1992. Assessment of changes in the aquatic ni in the upper Marcos River. M.S. thesis, Southwest Texas State Caney San Mar Unrrep Stares DEPARTMENT OF AGRICULTURE. 1983. Noxious weeds. Fed. Reg. 48: 30037- 20047. UNITED ue Fist AND WILDLIFE SERMeys I 994, Draft San Marcos and Comal Springs and y plan. Albuquerque, N Van Duk, G. M. ,D.D. Teaver and W.T. Hatter. 1986. Growth ote orabhale and Hydrilla in flowing wa a Aquatic Pl. Managem. 24:85-87. W assHAUSEN, D.C. 1966. Acanthaceae. In: C.L. Lundell, ed. Flora of Texas 1:223-282. THOMAS, TOWNSEND, OR TOW NSHEND— WHAT WAS T'S. BRANDEGEE’S NAME? PAUL A. FRY XELL Department of Botany University of Texas at Austin Austin, TX 78713-7640, U.S.A. HORACE R. BURKE Department of Entomology Texas AGM University College Station, TX 77843-2475, U.S.A. JOHN EF REED Director of the Library New York Botanical Garden Bronx, NY 10458-5126, U.S.A. T.S. Brandegee and his wife Katharine Brandegee were important California botanists (Setchell 1926) who contributed greatly to our knowledge of the flora of that floristically rich region during the early decades of the 20th century. The nature and extent of their contributions are amply dealt with elsewhere (see references) and need not detain us here. We became aware of the fact, however, that T.S. Brandegee’s full name is referred to in the literature with several vari- ants, which may be a source of confusion. We wished to ascertain which alterna- tive is correct and to share our findings. In his own publications Brandegee usually cited his name with initials only, as “T.S. Brandegee.” Others often referred to him in this same way, but some- times also as “Townsend Stith Brandegee” (Lanjouw & Stafleu 1954, Lindsay 1955, Thomas 1969, Stafleu & Cowan 1976), sometimes as “Townshend Stith Brandegee” (Setchell 1926, Jones 1929, Ewan 1950, Lindsay 1955, McVaugh 1956, Langman 1964, Thomas 1969, Hunt Institute 1972, Lenz 1986, pp. 26, 84, Bonta 1991, p. 87, Reveal & Pringle 1993), and sometimes as “Thomas Stith Brandegee.” (Jones 1929, Rodgers 1968, Lenz 1986, p. 200). Indeed, four of the references cited above (Jones 1929, Lindsay 1955, Thomas 1969, Lenz 1986) are internally inconsistent in citing two diffe variants of Brandegee’s name. We therefore conclude that no single reference can be considered authori- tative in solving this problem, nor do we believe that this sort of question can be resolved by majority rule. All references agree on his second name “Stith,” taken from his mother’s Stipa 16(2): 373 — 374. 1994 374 Sipa 16(2) 1994 maiden name, so that is not a source of controversy. The first name (whether Thomas, Townsend, or Townshend), however, is clearly a source of uncertainty. But two references give an unequivocal answer and thus resolve the uncertainty, and we wish to share this information in the present note. One paper published by Brandegee (Brandegee 1913) departs from his usual practice of using only initials in the byline, and clearly states “Townshend Stith Brandegee.” The second and more persuasive bit of evidence is the portrait of Brandegee at the age of 80 published by Setchell (1926: plate 14), which in- cludes the full signature of Brandegee, reading “Townshend Stith Brandegee.” Thus, since it is attested in his own hand, there can be no doubt about the correct spelling of Brandegee’s given name, which was not “Townsend” and certainly was not “Thomas.” Even more compelling but less readily accessible are documents in the Brande- gee archives kept at the | ee of the University of California (UC), includ- ing his “doctoral diploma from Yale, pension papers, and discharge from the U.S. Army,” all of which specify his given name as “Townshend” (Barbara Ertter, pers. comm.). REFERENCES Bonta, M.M. 1991. Kate Brandegee, accidental botanist. In: Women in the Field. Texas A&M University Press. 1991. pp. 85-92. BRANDEGEE, T.S. 1913. Plantae Mexicanae Purpusianae, V. Univ. Calif. Publ. Bot. 4:375—388. Ewan, J. 1950. Rocky Mountain naturalists. University of Denver Press. Hunt Boranicat Liprary. 1972. Biographical dictionary of one represented in the Hunt Institute portrait collections. G.K. H fall & Co.: Boston. Jones, M.E. 1929. The Brandegees. Contr. W. Bot. 15:15-18. LANGMAN, I.K. 1964. A selected guide to the literacure on the flowering plants of Mexico. University of Pennsylvania Press. 1013 pp. Lanjouw, J. and F.A. Srarteu. 1954. Index herbariorum, Part II, Collectors (A—D). Lenz, L.W. 1986. nase us E, gon Rancho Santa Ana Botanic eae LInpsay, G. 1955.2 : the botanical explor xploration of Lower Califor- nia, Mexico. Belvedere Scientific cael McVauGu, R. 1956. Edward Palmer, explorer of the American West. University of Oklahoma Press REVEAL, J.L. and J.S. PRINGLE. 1993. eure botany and flristics. In: Flora of North America Edi torial Committee, eds. 1993+. Flora of North America North of Mexico. 2+ vols. Oxford rsity Press, New York. Vo 1 1, pp. on 192, renee A.D. II. 1968. American botany 1873-1892, decades of transition. Hafner: New York. SETCHELL, W.A. 1926. Townshend Stith — and Mary Katharine (Layne) (Curran) Brande- gee. Univ. Calif. Publ. Bot. 13:155-178 + 2 plates SrarLeu, EA. and R.S. Cowan. 1976. Taxonomic literature-II. Vol. 1 (A—G):303. Tuomas, J H. 1969. Botanical explorations in Washington, Oregon, California, and adjacent regions. Huntia 3:5—62. NOTES NOTEWORTHY NEW RECORDS FOR NEOTROPICAL GENTIANACEAE—New distribution records, based on specimens I have re- cently identified, are reported here for six Mexican and South American species of Gentianaceae. Gentiana mirandae Para MEXICO. Veracruz: Mun. Huayacocotla, Viborillas, Huayacocotla, 2200 m, Herndndez M. & Trigos 1188 (E, GH This is the first record of this showy but rarely collected species for Veracruz, although the locality is near a previously reported locality in Hidalgo (Pringle LED) Gentianella achalensis (Hieron. ex Gilg) T.N. Ho & S.W. Liu BOLIVIA. Tarya. Prov. Arce: 12.5 km N of Emboroza-Sidras road, on road to Padcaya, 22° 12'S, 64° 37'W, 1150 m, Solomon 10160 (MO). Prov. Cercado: 54.9 km E of Tarija- Padcaya road, on road to Entre Rfos, 21° 29'S, 64° 20'W, 2050 m, Solomon 10324 (MO) These records extend the known range of this species hitherto recorded only form Argentina, into southernmost Bolivia. Field work in Bolivia by J.C. Solomon was supported by the National Science Foundation. Gentianella armerioides (Griseb. ex Gilg) J.S. Pringle BOLIVIA. La Paz. Prov. Loayza: along the S fork of the Rio Jiskha Choque Kkota which down from Cerro Huatia Kkota, ca. 4 km SE from Viloco, 16° 29'W, 4300-4650 m, Lewis 871320 (MO). This record extends the known range of this species, hitherto recorded only form Depto. Puno, Pert, into adjacent Bolivia. Gentianella primuloides ee Pringle BOLIVIA. La Paz. Prov. B vedra: cerca de la carretera nueva de Cotacampa a Ulla- Ulla, Cerro Pifiita, 4500 m, perc: 1782 (HAM, LPB). The isotype at G, from Depto. Puno, Pert, which is of particularly good quality, is closely matched by this recent collection from just across the border in Bolivia, indication that this species is correctly attributed to the floras of both countries. The corollas were described by Manhofer (in sched.) as red (rojas), and show contrasting whitish bases, corresponding to Gilg’s (1916) description. Sina 16(2): 375. 1994 376 Sipa 16(2) 1994 Gentianella vargasii Fabris BOLIVIA. La Paz. Prov. Murillo: La Paz-Calacoto 64 km hacia el nevado Ilimani, sobre el pueblo de Pinaya, pie de Illimani, 4200 m, Beck 9073 (HAM, LPB) This record extends the known range ca. 570 km to the southeast and is the first for Bolivia, the species hitherto being known only form Depto. Cuzco, Pera (Fabris 1958). This indicates that the range of G. vargasii may be more ext than that of most Andean Gentianella species, even though the species may be lee tively uncommon. A parallel is approached, however, in the relatively extensive range of G. punicea (Wedd.) Holub, which extends from Depto. La Paz, Bolivia, along the mountain ranges northeast of Lake Titicaca, well into Pera. These ranges are essentially continuous from the international border to the southeast- ern part of Depto. Cuzco from which G. vargasii was previously reported. T Steyerm. Pp PERU. Amazonas. Luya Prov.: Camporredondo-Tullanya, trocha hacia el Cerro Huicsocunga, 2350 m, Diaz & Campos 3733 (HAM, MO). Cajamarca. Cutervo Prov.: San Andrés de Cutervo, Parque Nacional de Cutervo, entrando por “Chorro Blanco” sobre invernas del Sr. Nicolas Navarro, 2300—2400 m, Diaz et al. 3951 (HAM, MQ). These records, which are the basis of my inclusion of this species in Brako & Zarucchi’s (1993) Catalogue of the Flowering Plants and Gymnosperms of Peru, are the first for the genus and the species for Pert, T’. zamoranum hitherto having been reported only from southern Ecuador (Steyermark 1951). — —James §. Pringle, Royal Botanical Gardens, Box 399, Hamilton, Ontario, Canada L8N 3H8. This is Contribution No. 85 from the Royal Botanical Gardens, Hamilton, Ontario. REFERENCES Brako, L., and J.L. Zaruccut. 1993. Catalogue of the flowering plants and Gymnosperms of Peru/Catalogo de e oo eal y Gimnospermas de Pert. Monogr. Syst. Bot. Missouri Bot. Gard. 45 xl pp. aa. ee tas sobre Gentianella del Peri. Bol. Soc. Argent. pe oe Gite, E. 1916. Gentianaceae Andinae. Bot. Jahrb. Syst. 54 (Beibl. 118):4 sae Se J.B. 1959. Gentianaceae. Gentian Family. In: Flora of Peru. aa ve Nat. Hist., er. ee 270-363. Bay , J.S. 1977. Taxonom 1 distribution of Gentiana (Genti ) in Mexico and Cen- Cees . Sect. enn Sida 7:174—217. ee ned 1951. The genus Tapeinostemon (Gentianaceae). Lloydia 14:58-64 Sipa 16(2): 376. 1994 NOotESs 377 SAGINA PROCUMBENS (CARYOPHYLLACEAE), NEW TO ILLINOIS—Sagina procumbens L., procumbent pearlwort, is a Eurasian native that has been found at various sites in North America, well established prima- rily along both the Atlantic and Pacific coasts (Crow 1978). Within the Great Lakes Region, it has been reported from scattered sites in Michigan, Minnesota, Ohio, and Wisconsin; all but four sites, three in the Lower Peninsula of Michi- gan (Crow 1978, Voss 1985) and one in central Ohio (Crow 1978), are along the shore of Lake Superior. Sagina procumbens can now be reported from four counties in Illinois from the following collections: ILLINOIS. Cook Co.: in patio blocks, 2601 Lake Ave., Wilmette, last week Jun 1976, E. McArdle & R. H. Fargo s.n. (MOR, SIU); sidewalk cracks, 5300 Block of N Spaulding Ave., Chicago, 27 Jun 1985, Hattaway 5326 (MOR, {SIU]). Kane Co.: sidewalk cracks, 40 W 693 McDonald Rd., N of Wasco, 13 Jul 1991, K. Johnson s.n. (MOR). Lake Co.: between stones or brick, Farwell & Lydia Lee’s residence, Lake Forest, 6{?} Jul 1965, Rulison 49-X (SIU). Peoria Co.: in brick sidewalk cracks, 200 block N Garfield St., Peoria, 15 Jun 1955, Chase 14252 (ILL{2}). The reports of Sagina decumbens (Elliott) Torrey & A. Gray from Cook and Lake counties by Mohlenbrock and Ladd (1978) and Swink and Wilhelm (1979) are based on specimens of S. procumbens cited here. Sagina decumbens is native to sandy areas in the southeastern United States and, except for a single collection from Peoria Co., is “occasional in the s. 1/2 of the state (Mohlenbrock 1986).” When the dots for Cook and Lake counties are removed, the map in Mohlen- brock and Ladd (1978) for the distribution of S. decumbens in Illinois is once again correct. Hattaway 5326 is the basis of his report of Arenari, as new to Illinois and the Midwest (Hattaway, 1987); it was erroneously identi- fied as that species. Sagina procumbens can be distinguished from S. decumbens via the following ey: 1. Plant perennial, stems procumbent, sepals diverging from the dehiscent landica (Retz.) Sprengel es oO sule S. procumbens 1. Plant annual, stems filiform and mostly erect, sepals remaining appressed to the dehiscent capsule S. decumbens Two other characters are commonly cited as separating these two species, features that are diagnostic in many, but not all cases. The flowers of S. procumbens are predominantly 4-merous with petals much shorter than the sepals or absent while those of S. decumbens are predominantly 5-merous with petals about equal- ling the sepals. Some populations of S. decumbens include plants that are smaller and more branched and that often have 4-merous apetalous flowers (specimens Spa 16(2): 377. 1994 378 Stipa 16(2) 1994 seen from Coles and Wabash cos., ILLS); Crow (1978) included these within the range of variation of S. decumbens. ACKNOWLEDGMENTS I thank the curators of ILL, ILLS, MOR, and SIU for the use of their collec- tions and Garrett E. Crow and an anonymous reviewer for comments on this manuscript. —Richard K. Rabeler, University of Michigan Herbarium, North University Build- ing, Ann Arbor, MI 48109-1057, U.S.A. REFERENCES Crow, G.E. 1978. A taxonomic revision of Sagina (Caryophyllaceae) in North America. Rhodora 80:1-91. Harraway, R.A. 1987. Arenaria groenlandica (Retz.) Spreng. var. groenlandica in the Midwest. Trans. Illinois Acad. Sci. 80:343-344 MOHLENBROCK, R.H. 1986. Guide to the vascular flora of Illinois, rev. ed. Southern Ilinois Univ. Press, Carbondale & Edwardsville. MOHLENBROCK, R.H. and D.M. Lapp. 1978. Distribution of Illinois vascular plants. Southern Illinois Univ. Press, Carbondale & Edwardsville. SWINK, F. and G. WILHELM. 1979. Plants of the Chicago region, rev. ed. The Morton Arbore- tum, Lisle, IL. Voss, E.G. 1985. Michigan flora. Part II. Dicots. (Saururaceae-Cornaceae). Bull. Cranbrook Inst. Sci. 59 and Univ. Michigan Herbarium. EMILIA FOSBERGI (ASTERACEAE: SENECIONEAE), A NEW INTRO- DUCTION TO TEXAS—Emilia is an Old World genus with approximately 45 species, three of which have become neotropical weeds (Nicolson 1975). Until recently Emilia fosbergit Nicolson has been reported in the United States only as a casual weed growing in southern Florida (Cronquist 1980; Barkley & Cronquist 1978). In the fall of 1993, a population of approximately 40 individuals of Emilia fosbergii was found growing in and around a gravelly path in a south Austin nursery. Upon inquiry the manager informed me that the nursery has suppliers in Florida, which might explain the presence of this weed. Voucher specimen: TEXAS. Travis Co.: City of Austin, A-1 Grass Nursery, two blocks S of Barton Skyway along S Lamar St., growing in gravel of parking lot and planting area, 20 Nov 1993, Williams s.n. (TEX). Emilia fosbergii is distinguished from other Texas genera in the Senecioneae by the red disk flowers and absent ray flowers. The receptacle is flat to slightly convex, the leaves are alternate. Stipa 16(2): 378. 1994 NOTES 379 —Justin K. Williams, Department of Botany, University of Texas, Austin, TX 78713, REFERENCES BARKLEY, T.M. and A. Cronouist. 1978. Emilia. In: N. Amer. Fl., Ser. I, 10:147-150. Cronautst, A. 1980. Vascular flora of the southeastern United States. Vol. I. Asteraceae. Univ. North Carolina Press, Chapel Hill Nicoison, D.H. 1975. Emilia fosbergii, a new species. Phytologia 32:33. NEW COLLECTION RECORDS FOR THE AQUATIC MACROPHYTES CERATOPTERIS THALICTROIDES (PARKERIACEAE) AND LIMNOPHILA SESSILIFLORA (SCROPHULARIACEAE) IN TEXAS—The spring systems that arise along the Balcones fault zone of central Texas support a diversity of aquatic macrophytes, including a number of adventive species that have not been reported from elsewhere in the state (Lemke 1989, Ramamoorthy & Turner 1992, Angerstein & Lemke 1994). The upper San Marcos River in Hays County, Texas, supports a macrophyte community comprising thirty-one species (Lemke 1989), two of which, Ceratopteris thalictroides (L.) Brongn. and Limnophila sessiliflora B\., have not previously been reported elsewhere in Texas (Correll & Johnston 1970). Recent collections of aquatic macrophytes from Landa Lake, a small reservoir formed by the damming of the headwaters of the Comal River in New Braunfels, Comal County, Texas, have documented the presence of these two species in this river system as well. Ceratopteris thalictroides is a tropical, free-floating, homosporous aquatic fern that has been introduced into Florida, Louisiana, Texas, and California (Lloyd 1993). The species was first reported from Texas by Morton (1967) and its introduction into the San Marcos River by a local aquarium plant supply com- pany was documented by Hannan (1969). The following collection represents only the second county record for the species in Texas: Voucher specimen: TEXAS. Comal Co.: free-floating along north shore of Landa Lake, Landa Park, City of New Braunfels, 16 Apr 1994, Lemke 4103 (SWT). Limnophila sessiliflora is a sabmersed or emergent macrophyte one ees to India and Southeast Asia. It resembles the native species C Bu Gray in gross morphology, but can be easily distinguished 1n the vegetative state by its bright green coloration, more compact growth habit, and verticillate leaves. Limnophila sessiliflora is reported to be sporadically naturalized in Florida and Georgia; the following collection represents only the second county record for the species in Texas: jee Voucher specimen: Texas. Comal Co.: submerged along north shore of Landa Lake, Landa Park, City of New Braunfels, 25 May 1994, Lemke 4171 (SWT). Siwa 16(2): 379. 1994 380 Stipa 16(2) 1994 It seems likely that these species were introduced into Landa Lake either intentionally, as has been documented for C. thalictroides in Texas by Hannan (1969) and for L. sessi/iflora in Florida by Mahler (1980), or unintentionally through careless dumping by aquarists, as has been surmised for the aquatic weeds Nomaphila stricta (Vahl) Nees (Acanthaceae) by Ramamoorthy & Turner (1992) and Hygrophila polysperma (Roxb.) T. Anders. (Acanthaceae) by Angerstein & Lemke (1994 Although C. Pereren has not been reported to occur so abundantly as to cause problems in aquatic ecosystems, L. sessiliflora is of concern as a potential aquatic weed. Several species of Limnophila, including L. sessiliflora, are respon- sible for major weed infestations throughout much of Southeast Asia and virtu- ally all herbicides registered for use in aquatic systems have proven ineffective in controlling these species (Misra & Tripathy 1975, Takematsu et al. 1976, Mahler 1980). Spencer & Bowes (1985) reported that L. sessé/iflora has several character- istics that could provide it with a competitive advantage over native aquatic plants in Florida, such as a substantial reproductive capacity, the potential fora low photorespiration rate, and the ability to photosynthesize effectively under low light regimes. Limnophila sessiliflora was reported to be uncommon in the San Marcos River by Lemke (1989) but has been found to be abundant in parts of Landa Lake, where several individuals were observed to be flowering in the summer of 1994. Although the species has a limited capacity for asexual reproduction, this may be offset by its profuse sexual reproduction. Spencer & Bowes (1985) reported that each flower of L. sess7/iflora may set between 200 and 300 seeds with a germination rate as high as 96 percent. Presently there is no evidence that L. sessiliflora is having a deleterious effect on the growth of native aquatic plant species in either the San Marcos or Comal rivers; however, the spread of this species in the spring systems of central Texas should be closely monitored. —David E. Lemke, Department of Biology, Southwest Texas State University, San Marcos, TX 78666 U.S.A. REFERENCES ANGERSTEIN, M.B. and D.E. LeMKE. 1994. apes records of the aquatic weed Hygrophila polysperma ones from Texas. Sida 16:365— CorreLl, D.S. and M.C. i 1970. cee of the vascular plants of Texas. Texas Re- aay Bonnet of Renner, T HAnnaNn, H.H. 1969. The are and establishment of Ceratopreris in Texas. Amer. Fern 59:122 Lemke, D.E. 1989. Aquatic macrophytes of the upper San Marcos River, Hays County, Texas. SouthW. Naturalist 32:289-291 Stipa 16(2): 380. 1994 Notes 381 Lioyp, R.M. 1993. eee In: Editorial oe a eds. Flora of North America North of Mexico, vol. 2, pp. 119-121. Oxford Univ. Press, New Mauter, M.J. 1980. Cpl a NeW eXOtic pest. oe a te ‘2 Misra, G. and G. TripatHy. 1975. Studies on the control of aquatic weeds of Orissa India. 2. Effect of chemical herbicides on some aquatic weeds. J. Indian Bot. Soc. 54:65—71. RAMAMOORTHY, T.P. and B.L. Turner. 1992. eae stricta (Acanthaceae), a newly discerned weed in Texas, and the first report for North America. Sida 15:115—117. SPENCER, W. and G. Bowes. 1985. Limnophila and Hyprophila: a review and physiological assess- ent of their weed digo: in Florida. J. Aquatic Pl. Managem. 23:7-1 Seats , 1., M. Konnal, Y. TAKEUCHI and N. Icuizen. 1976. Weeds ee neiee fields and herbicides in China. Bull Coll. Agric. Utsunomiya Univ. 9:91-107. QUERCUS LACEYI (FAGACEAE) NEW TO THE LLANO UPLIFT AREA OF CENTRAL TEXAS—The Lacey oak, Quercus laceyi Small, is a small to moderate-sized tree that is commonly found on mesic north-facing slopes and in canyons on the Edwards Plateau of central Texas. The range of the species in Texas extends from Bexar and Medina counties in the east to Terrell and Brewster counties in the west; it is also found on the eastern slopes of the Sierra Madre Oriental in Coahuila and Nuevo Leon, Mexico (Muller 1951, Nixon & Muller 1992). These plants were included in Q. g/aucoides Mart. & Gal. by Tre- lease (1924) and Correll & Johnston (1970), but have recently been shown to represent a distinct allopatric species, with true Q. glaucoides being confined to central and southern Mexico (Nixon & Muller 1992), Quercus laceyi was described by Small (1901) from material collected “on lime- stone hills” in Kerr County, Texas, and subsequent descriptions of the habitat of the species (e.g., Muller 1951, Correll & Johnston 1970, Nixon & Muller 1992) have always emphasized the calcareous substrate. In October, 1993, I discovered a stand of several dozen individuals of Q. /acey7 in a mesic canyon at Enchanted Rock State Natural Area in Gillespie and Llano counties, Texas. The individuals were mostly mature trees 20—30 cm in diameter growing in association with Celtis reticulata Torr. and Diospyros texana Scheele. The following collection cita- tion is the first documentation of the occurrence of Q. /aceyi on the igneous- derived sandy soils of the Llano Uplift rather than the limestone-derived soils of the surrounding Edwards Plateau: Voucher specimen: TEXAS. Llano Co.: Enchanted Rock State Natural Area, in canyon between Enchanted Rock and Little Rock, 13 Oct 1993, Lemke 4138 (SWT, TEX). Given that several vegetation studies have been conducted in the area now included within Enchanted Rock State Natural Area (Whitehouse 1933, Butterwick 1979, Walters 1980) and that the collection site is situated along a frequently used hiking trail, it is surprising that the occurrence of Q. /acey7 has Spa 16(2): 381. 1994 382 Sipa 16(2) 1994 not been previously noted. This may be due to superficial similarities between Lacey oak and post oak (Q. stellata Wang.), which 1s common throughout the park. I myself have walked past this stand of trees numerous times over the last 18 years and never noticed them until last fall. Upon close examination, how- ever, the characteristic bluish-green upper leaf surface and waxy deposits on the pulvinus readily distinguish Q. /acey/ from Q. stellata —David E. Lemke, Depart- ment of Biology, Southwest Texas State University, San Marcos, TX 786006, U.S.A. REFERENCES 1 Gill Burrerwick, M. 1979. A survey of the flora of Enchanted Rock and vicinity, Llat counties, Texas. In Enchanted Rock, a natural area survey (no. 14), pp. 41— 102. iyadouk. Johnson — eee Affairs, Univ. of Texas, Austin Corre, D.S. an C. Jonnsron. 1970. Manual of the vascular plants of Texas. Texas Re- oh sean Renner, Texas Mutter, C.H. 1951. The oaks of Texas. Contr. Texas Res. Found. 1:21—311. Nixon, K.C. and C.H. Mutter. 1992. The taxonomic resurrection of Quercus laceyi Small (Fa- gaceae). ee a 15:57-69. SMALL, J.K. 1901. Trees and shrubs of the sout —— nern states-IV. Bull. Torrey Bot. Club 28: 356-361. Trecease, W. 1924. The American oaks. Mem. Natl. Acad. Sci. 20:1-255. Watters, T.W. 1980. Vascular flora and eevee of granite outcrops in the central mineral region of Texas. Masters thesis, Texas A&M Uniyv., College Station. WHITEHOUSE, E. 1933. Plant succession on on Texas granite. Ecology 14:391—405. SOLANUM VIARUM (SOLANACEAE), NEW TO MISSISSIPPI—Solanum viarum Dunal, TROPICAL SODA APPLE, is a perennial shrub that belongs to section Acanthophora of subgenus Leptostemonum (Nee 1991). It is native to Brazil and Argentina but has become a weed in other areas of South America and in Africa, India, Nepal, West Indies, Honduras, and Mexico (Nee 1991) and Florida, U.S.A. (Coile 1993, Mullahey et al. 1993a, 1993b, 1993c). Mature plants of 8S. viarum are 1 to 2 m tall and are armed on the leaves, stems, pedicles, petioles, and calyxes with broad based white to yellowish prick- les up to 12 mm long (Nee 1991, Mullahey et al. 1993c). Leaves and stems are pubescent; corollas are white with five recurved petals and white to cream col- ored stamens that surround the single pistil (Coile 1993). Immature fruits are mottled whitish to light green and dark green (.e., like a watermelon) (Mullahey etal. 1993c). The mature fruits are smooth, globular, yellow, and 2 to 3.2 cm in diameter with a leathery skin surrounding a thin-layered, pale-green, mucilagi- nous, scented pulp and moderately flattened, reddish-brown seeds (Coile 1993; Mullahey et al. 1993a). Each S. vzarum plant has the capability to produce over 50,000 seeds (Mullahey et al. 1993c). The first known U.S.A. collection of S. viarum was from Glades County, Florida Siva 16(2): 382. 1994 NOTES 383 in 1988, but it may have been present in Florida as early as 1981 or 1982 (Coile 1993). Since its introduction into Florida, $. viarum has spread rapidly. By 1993 it was reported to infest over 61,000 ha of improved pastures, citrus groves, sugar cane fields, ditches, natural areas (oak hammocks and cypress heads), etc. (Mullahey et al. 1993b). Drs. Robert Eplee and Randy Westbrooks (pers. comm., U.S. Dept. Agric., APHIS, PPQ, Whiteville, NC) estimated that by early 1994 S. viarum infested as many as 285,000 ha in Florida. Because of its rapid popu- lation explosion 1n Florida, S. vzarum was placed on the Florida noxious weed list on February 28, 1994. The primary means of dispersal of S. vzarum seems to be livestock and wild- life, such as raccoons, deer, feral hogs, and birds feeding on the fruits (Coile 1993, Mullahey et al. 1993a). Mullahey et al. (1993a) reported that S. viarum foliage is unpalatable to livestock although cattle will eat the mature fruit. Scari- fication of seeds by digestive systems of livestock and wildlife seems to promote seed germination. Intra- and inter- county and state movement of livestock that have recently fed on S. viarum fruit may be the primary vectors for its spread. However, contaminated hay may also serve as a means of dispersal (Wunderlin et al. 1993). Once S. viarum is established, livestock and wildlife continue to disperse seeds resulting in population levels that may increase exponentially. Collection data for S. viarum in Mississippi follow with herbarium abbrevia- tions after Holmgren et al. (1990), except MMNS and USMH (Mississippi Museum of Natural Science, Jackson and University of Southern Mississippi, Hattiesburg, respectively). Voucher specimens: U.S.A. MISSISSIPPI. Forrest Co.: NW of Hattiesburg, 5 mi NE jet. of Hwys I-59 & US 98, S of US 98, Southeast Mississippi Livestock Auction, 27 Jul 1994, Bryson 13995 (BRIT/SMU, IBE, MISS, MMNS, NLU, PIHG, SWSL, USMH, VSC, VDB, additional Spectncns ] yee cistuibuted); 15 ie ee t, Bryson 14188 & Rosso (SWSL, additional speci- ). Lamar Co.: Oloh, 2.8 mi S of Hwy US 98 and to - Ore anyiss ea Rd, 27 Sep on Bryson 14454, on > Rouse (SWSL, additional spec oe 14456, Byrd, & Rouse IBE, NY, SWSL, additional seccinens to be See: Oak ve Community, SE of jet. Oak Grove Road (= old Hwy US 11) & Griffith Rd, 27 Sep 1994, Bry son 14457, rae < Rouse (SWSL, additional specimens to be distributed). Pearl River Co.: SE of Poplarville, 4.3 mi SE jct. of Hwys I-59 and MS 53, Eof MS 53, 14 Jul 1994, Byrd s.n. CBE, SWSL); 2 = 1054, Bryson 13996 (BRIT/SMU, DSC, IBE, MISS, MMNS, NLU, PIHG, SWSL, USMH, VSC, VDB, additional specimens to be distributed); 16 Aug 1994, Bryson 142066 & Lockley (SWSL, additional specimens to be distributed); 8 Sep 1994, Bryson 14367 (NLU, SWSL); SE of Poplarville, 2.1 mi SE jct. Hwys I-59 & MS 53, N of Restertown Rd, 17 Aug 1994, Bryson 14274 & Byrd (SWSL); SE a 2.1 mi SE of jct. Hwys I-59 & MS 53, Eon Restertown, W of Bert Gentry Road, 17 Aug 1994, Bryson 14276 & Byrd (SWSL). Stone Co.: ca. 11 mi WSW of Perkinston, 1.5 miS of jct. Ridge Rd & Red ee Rd on Ree Gap Rd, 17 Aug 1994, Bryson 14279 & Byrd (SWSL, esas specimens The senior author first observed several plants of S. viarum growing in the Siwa 16(2): 383. 1994 384 Sipa 16(2) 1994 median of Hwy I-10 between mile markers 17 and 18 in Hancock County, Mississippi on 17 October 1993 but did not realize the significance of the obser- vation until the annual meeting of the Weed Science Society of America in February 1994. Subsequent trips to this site and surrounding areas in May, June, and early July 1994 were unproductive in relocating S. viarum because the median had been mowed. An additional sighting was reported from Hinds County, Mississippi in late 1993. At the Forrest County location six S. viarum plants were found along a live- stock holding-pen fence. Plants were about | m tall and were possibly estab- lished the previous year, each bearing up to 50 mature fruits per plant. About 200 S. viarum plants ina 2 to 4 ha area were present at the inital Pearl River County location, ranging in size from seedlings to 1.5 m tall with mature fruit. Additional seedling plants were observed with each successive visit to this loca- tion. At this site, plants were observed growing in the shade of pecan and oak trees but most were growing in full sun. Dead plants with mature fruit from the previous season were observed along a fence row that had been treated witha herbicide. The land owner stated that the plants had been at this Pearl River County location since 1991 and he pointed out several plants that had produced fruit in 1993, Evidently S. vzarum is at least a short-lived perennial in southern Mississippi. At che Stone County location, 8. viarum infests several 100 ha of pastures and holding-pens around a farm headquarters barn at the end of Red Gap Road. In Lamar County, S. vzarum infests a total of about 50 ha at two locations. At each of the locations in Lamar, Pearl River, and Stone counties, Solanum viarum was found in pastures where livestock had been introduced from Florida within the past few years and population size seems to be directly pro- portional to the total number of livestock imported from infested areas in Florida. At each location in Mississippi, plants were growing ina sandy loam soil. Mature plants were dug up at each location. In mature plants, 8S. vzarum root systems were up to 3 dm deep with lateral roots up to 1 m long and 1.5 to 2.5 cm in diameter. In laboratory and greenhouse experiments at Stoneville, MS, it was discov- ered that S. vzarum fruit will ripen in sunlight if the green fruit were greater than 2 cm in diameter. Seed germination from these fruit was greater than 70% in greenhouse conditions. In addition to dispersal by livestock and wildlife, it was discovered that S. vzarum fruit may be dispersed by water. Solanum viarum fruit from less than 1 cm in diameter to mature yellow or dry fruit are extremely buoyant. Thus, control of S. viarum requires elimination of immature and ma- ture fruit as well as the whole plant including the root system. In order to prevent further dispersal of this pernicious weed, the U.S. Dept. Agric., APHIS, PPQ and the Mississippi Cooperative Extension Service are on Stipa 16(2): 384. 1994 NOortEs 385 initiating an education and notification campaign on the potential weed prob- lem of S. viarum. Additional surveys and an eradication program in Mississippi may be required to prevent S. viarum spread not only within Mississippi but into other states. Botanists, livestock producers, weed scientists, and regulatory officials should be alerted that S. viarum may already be in their region. Early detection is paramount to eliminate this weed, which has the potential to infest millions of ha of pasture lands, row crops, truck crops, sod farms, lawns, forests, and natural areas, especially in tropical and subtropical regions of the U.S.A. and other countries. Additional research is needed to determine S. viarum fitness to more northern areas of the U.S. ACKNOWLEDGMENTS We thank Michael Nee (NY) for verification of the initial collections of S. viarum and Richard Carter (VSC), Nancy Coile (PIHG), Jeff Mullahey (SW Florida Research and Education Center, Immokalee, FL), and Robert Eplee and Randy Westbrooks (U.S. Dept. Agric., APHIS, PPQ, Whiteville, NC) for pro- viding information on the biology and spread of S. vzarum in Florida. —Charles T. Bryson, USDA, ARS, Southern Weed Science Laboratory (SWSL), Stoneville, MS 38776, U.S.A. and John D. Byrd, Department of Plant and Soil Sci- ences, Mississippi State Univ., Mississippi State, MS 39762, U.S.A. REFERENCES Come, N.C. 1993. Tropical soda apple, Solanum viarum Dunal: the plant from hell. Fla. Dept. Agric. & Consumer Services, Div. Plant Industry. Bot. Circ. No. 27, Hoimcren, P.K., N.H. Hotmaren, and L.C. Barnett (eds.). 1990. Index Femationant Part 1: the herbaria of the world, 8th ed. New York Botanical Garden, Bronx. Mutauey, J.J. and D.L. Cotvin. 1993a. Tropical soda apple: a new noxious Si in Florida. Univ. Florida, Florids Cooperative Extension Service, Fact Sheet WRS-7 MUuLLAHEY, J.J., JA. CORNELL, and D.L. Cotvin. 1994b. Tropical soda seule ae viarum) control. Weed echo: T2344 ai. Mutaney, J.J., M. Nez, R.P. WUNDERLIN, and K.R. DELANEY. 1993c. Tropical apple (Solanum viarum): a weed threat in subtropical regions. Weed Technol. 7:783-786 Nee, M. 1991. Synopsis of So/anum section Acanthophora: A revision of interest for glyco alka- loids. p. 258-266 in J.G. Hawkes, R.N. Lester, M. Nee, and N. Estrada, eds. Solanaceae III: Taxonomy, Chemistry, Evolution. Royal Romine Sans Kew, Richmond, Surr K. WUNDERLIN, R.P., B.E. HANSEN, K.R. DELANEY, M. NEk, and J.J. MULLAHEY. oes Solanum viarum and S. tampicense (Solanaceae): two mae meee new to Florida and the United States Sida 15:605-611. Siva 16(2): 385. 1994 BOOK REVIEW Bremer, K. (with assistance of A. A. Anderberg, P.O. Karis, B. Nordenstam, J. Lundberg, and O. Ryding) 1994. Asteraceae: Cladistics and Clas- sification. (ISBN 0-88192-275-7, hbk). Timber Press, Portland, Or- egon. $79.95 (hbk). 752 pp., 89 figures, 24 tables, 6" x 9". Often viewed as a nontraditionalist for his early espousal of cladistic methodology, Kare Bremer has produced a monograph of the Asteraceae in the finest Benthamian tradition. This is no easy task in today’s world of multiple sources of comparative data and conflicting taxonomic opinions among the disproportionately populous community of synantherologists. However Bremer, who heads a Swedish research team in studying the systematics of the Wo Hd composite tribes, has become a recognized authority on the Asteraceae as a whole. Bremer has created a handbook of genera and suprageneric taxa set in the context of insights from recent ee research and of unresolved problems. However, unlike recent symposium vol- umes which provide a forum for disparate views and often uneven treatments, this volume is the expression of relationships and classification seen froma single perspective. The book is composed of four introductory chapters and 19 systemati hapters that treat the three subfamilies and 17 tribes in detail. The introductory ee cover 1) a brief over- view of cladistic methods, 2) the history of classification of the Asteraceae, 3) morphological criteria, and ~ the origin of the family and major unresolved problems relevant to the family as a whole. Each of the agstemate chapters ia De an overview can variation in and past treatments of the subfamily or tribe, 2) a new hed istic analysis, 3) discus- sion of the taxon’s evolution, and 4) formal seenocens éithe eabes and genera. The treat- ments of the Inuleae, Plucheeae, Gnaphalieae, Calenduleae, Helenieae, Heliantheae, and Eupatotiese were contributed 1 in whole or part by Bremetr’s associates. zoal is “...not to provide a ince cso of the Asteraceae, but rather to review the strengths and weaknesses 0 o facilitate further revisions by future taxonomists” and “to oe a fairly even and see treatment of the tribes”. In my opin- ion, he has succeeded in both. Even though the treatments of the Old World tribes are given somewhat more attention because the Stockholm group is most familiar with them, the dispar- ity does not create a noticeable unevenness, and the added insights are a plus for the North American synantherologist who ie not ventured beyond the Isthmus of Tehuantepec. In asense, the value of the b nds on the audience. I ee that it will be of greatest use to students, to herbarium curators, Sand to amateur and general taxonomists. These people can use it as a reference to the genera of composites, as well as to understand the evolution of the cae as Snes However, professional synantherologists, who already have references with ;, will find the value to be in the cladistic analyses and discussions. Bremer uld have eaaheaned the force in his views, as well as greatly reduced the size and price, by forgoing the generic descriptions. In fact, the information present is summarized in coded form in the data matrices provided. Along this same line, one could criticize the cladistic analyses because the monophylesis of the generic OTUs is not explicitly substantiated in the text or tables. One must wonder whether variable characters of certain la arge genera are coded Hos the mnest pisiomespate state of a syna omorphic transition or whether the most pl I transitior (relative to segregates) was used. This may not be a problem for the supsapeneric classification if the segregate and the large potentially paraphyletic genus are treated in the same subtribe. Sipa 16(2): 387. 1994 388 Stipa 16(2) 1994 However, the suprageneric classification is problematic in cases such as Dendroseris, which is treated in its own subtribe. Sanders et al. (1987) reviewed the suggestions for the origin of this oceanic island endemic. Included are Hieracium (probably sect. Stenotheca), Sonchus, pale za, and a few other small genera. Bremer cites a paper that suggests the transfer t. Stenotheca to Crepis. Each of ee four possible ancestors is in a separate subtribe. In Bremer’ S ena. Dendroseris is several branches and 10 steps away from Stephanomeria, —_— 6 steps away from Crepis, and one step away each from Hieracium and Sonchus. Undoubt- edly, sect. ae should have oe treated as a separate OTU and/or the large genera should have been represented by potentially monophyletic subgroups to more accurately assess the subtribal alignment of Dendroseris and circumscription of the pertinent subtribes. Bremer and his associates have thoroughly referenced the pertinent literature. However, I did find at least one minor error. T. F. Stuessy is omitted as a co-author of Sanders et al. (1987). This is especially unfortunate because Prof. Stuessy designed the study, obtained funding, led the field work, and kept a reign on me while I carried out ae ees and drafted the initial text. verall, I recommend this book based on its scholarly } and provocative content. Depending on one’s needs, this may be a “must have” reference. However, many specialists may wince at having to buy another large, expensive review of the Asteraceae. Literature Cited: SANDERS, R. W., T. F. Stugssy, C. Marricorena, and M. Sitva O. 1987. Phytogeography and evolution of Dendroseris and Robinsonia, tree-Compositae of the Juan Fernandez Islands. Opera Bot. 92: 195-215.—Roger W. Sanders. Stipa 16(2): 388. 1994 Notes on Carex (Cyperaceae), with C. godfreyi new to Alabama and C. communis and C, scoparia new to Mississippi Charles T. Bryson, John R. MacDonald and Randy Warren 455 Further notes on the genus Ardisia (Myrsinaceae) in Madagascar Jobn J. Pipoly HI 40] ‘irst records of the aquatic weed Hygrophila polysperma (Acanthaceae) from Texas Michelle B. Angerstein and David E. Lemke 405 Thomas, Townsend, or Townshend—What was T.S. Brandegee’s name? Paul A, Fryxell, Horace R. Burke and Jobn F. Reed 373 Notes Noteworthy new records for neotropical Gentianaceae 375 Sagina procumbens (Caryophyllaceae), new to [linois 377 Emilia fosbergii (Asteraceae: Senecioneae), a new introduction to Texas 378 New collection records for the aquatic macrophytes a thalictroides (Parkeriaceae) and Limnophila sessiliflora (Scrophulariaceae) in Texas Quercus laceyi (Fagaceae) new to the Llano Uplift area of central Texas 381 Solanum viarum (Solanaceae), new to Mississippi 382 pd Book reviews and notices 232, 262, 300, 387 Vv N ISSN 0036-1488 CONTENTS A new species of Zigadenus (Liliaceae) from New Mexico, with additional comments on the section Anticlea William J. Hess and Robert C. Sivinski 389 Una nueva especie del género Pteropepon (Cucurbitaceae) de Colombia Alvaro Cogollo P. and fobn J. Pipoly HI 4Q A new Tachigali (Fabaceae: Caesalipinioideae) from western Amazonia Jobn J. Pipoly Hl 407 Anatomical study of Erioneuron and Dasyochloa (Poaceae: Chloridoideae: Eragrostideae) in North America Jestis Valdés-Reyna and Stephan |. Hatch 2115) On the hybrid nature of Quercus basaseachicensis (Fagaccae, sect. Quercus) Richard Spellenberg 427 Nomenclatural changes in Setaria and Paspalidium (Poaceae: Paniceae) Robert D. Webster A taxonomic inyestigation of Cuscuta attenuata (Cuscutaceae) and related taxa L. Alan Prather, Ronald ]. Tyrl and William D. Warde 447 A new species of Dirca (Thymelacaceae) from the Sierra of northeastern Mexico Guy L. Nesom and Mark H. Mayfield 459 Miscellaneous notes on Haplophyton (Apocynace:c: Plumericae: Haplophytinae) Justin Kirk Williams 469 New taxa of Rhamnaceae from China Fan Guo-Sheng and Deng Li-Lan ion Amur honeysuckle (Lonicera maackii; Ca prifoliace::c); Its ascent, decline, and tal ee ee James O. Luken and John W “thteret CONTRIBUTIONS 479 TO BOTANY pad VOLUME 16 NUMBER 3 AUGUST, 1995 CONTRIBUTIONS TO BOTANY FOUNDED BY LLOYD H. SHINNERS 1962 Wm. F. Mahler Publisher 1971-1992 Director Emeritus x @ S.H. Sohmer Director Barney L. Lipscomb Editor Botanical Research Institute of Texas, Inc. 509 Pecan Street Fort Worth, Texas 76102-4060, USA 817 332-4441 / 817 332-4112 FAX John W. Thieret Prof. Dr. Félix Llamas Associate Editor Contributing Spanish Editor Biological Sciences Dept. Dpto. de Botanica, Facu Northern Kentucky University ltad de Biologia Universidad de Leon lighland Heights, Kentucky 41076, USA E-24071 Leon, Spain Guidelines for contributors are available upon request and on the inside back cover of the last issue of each volume. Subscription per year: $20. Individual numbers issued twice a year , $30. USA Institutions, $40. Outside USA: © SIDA, CONTRIBUTIONS TO BOTANY, Volume 16, Number 3, pages 389-610 Copyright 1995 Botanical Research Institute of Texas, Inc. Printed in the United States of America ISSN 0046-1488 A NEW SPECIES OF ZIGADENUS (LILIACEAE) FROM NEW MEXICO, WITH ADDITIONAL COMMENTS ON THE SECTION ANTICLEA WILLIAM J. HESS Morton Arboretum Lite di 60532, 0.5.A: ROBERT C. SIVINSKI New Mexico Forestry and Resources Conservation Division P.O. Box 1948 Santa Fe, NM 87504, U.S.A. and Herbarinm University of New Mexico Albuquerque, NM 87131-1091, U.S.A. ABSTRACT A new species, Zigadenus mogollonensis, is described and illustrated. Its large lowers are campanulate, cernuous, and have a distinct purplish tinge. It is endemic, but quite locally common, to the Mogollon Mountains of southwestern New Mexico. A discussion on re- lated species characteristics and variation within the section Anticlea of the genus Zigade- nus is also presented. RESUMEN e describe e ilustra una nueva especie, Z/gadenus head Sus flores grandes son campanuladas, colgantes y tienen un tinte pirpura manifiesto. Es endémica, aunque muy comiun localmente, de las montafas Mogollon del ae de Nuevo México. Se ofrece también una discusi6n de las caracterfsticas y variacién en las especies emparentadas dentro Qo. de la seccién Anticlea del género Zigadenus. In 1968, Hess collected a Zigadenus from the interior of the Mogollon Mountains in southwestern New Mexico and identified it as Zigadenus virescens (Kunth) J.F. Macbr. In the summer of 1993, he returned to the same general area and recollected this plant. Upon becoming familiar with the true Z. virescens, it was evident that the original collection was misidentified and that this Zigadenus was undescribed. Through inquiry, Hess discovered that Sivinski was also studying this same plant in the Mogollon Mountains, and had independently arrived at the same conclu- sion that it was a new and undescribed species of Zigadenus. Sipa 16(3): 389 — 400. 1995 390 Sipa 16(3) 1995 Zigadenus mogollonensis Hess & Sivinski, sp. nov. (Fig. 1) Perennius robustus, (4.5—)5.5—8.5(—9.5) dm altus. Bulb tenues ovoidei, 2—3 x 1.3— cm, tegmento papyraceo tenue vel tegmento crasso fibrarum e basibus éslioenm veterum dm ad nfra terram. Folia basalia ad 35 cm longa, 10-18 mm lata, pace leviter scabri; ela caulina Pee sursum deminuta. Inflorescentiae racemosae (8—)10—2 1(—28) nodis, |O—25(—35) x 3.5—8 cm, interdum paniculati ramis ex infernis 2—3 ie Badicell 4 crassi, 1.3-2 cm longi, recurvati sub anthesi, fructificantes erecti; bracteae infimae 2—4 x ).5—1.2 cm, acutae vel acuminatae, pallide virides, saepe tinctae purpurae per longitudinem marginum, longiores pedicellis, interdum extensi trans or Flores campanulati et cernut, 6 x f tepala elliptica, margines saepe recurvi, |2—1¢ —9.5 mm, rucilanctia vel margines rutilantes, virescentes (pallide eee vel meee prope venam mediam abaxialem, adaxialem vulgo pallidiores; apices rotundati, aliquando leviter emarginati; gradatim angustati ad basem vel late unguiculati; venae viridulae, arcuatae versus apicem,; glandes —5(-6) mm, leviter expansae ad extremitates, apices emarginati vel undulati, tiguancum bilobati, flavovirentes; stamina cum name ntis 7-8 mm longa, dilata basi, mm lata, decrescentia ad 0.5 mm lata articul antherae 2.4—2.6 x 1.7— 2 mm, hippocrepiformia, dehiscentia per totam a aan pollen striatum, monosulcatum 3042p * 18-24; pistilla tricarpellata, ovaria partim inferiora, trilocularia, 3 styli et stigmata; ovula numerosa, plana, breviter lanceolata. Capsulae trilobatae, 12-15 mm longae, lobi tepalae orientes 4-5 mm super pedicellos; nulla semina observata. Robust perennials, (4.5—)5.5—8.5(—9.5) dm tall. Bulbs slender, ovoid, 2-3 x 1.3-2 cm, with chin, papery covering to thick, fibrous covering from old leaf bases, to 1 dm below ground. Basal leaves to 35 cm long, 10— 18 mm wide, margins slightly scabrous; cauline leaves 1—3(—5), reduced upwards. Inflorescences racemose, with (8—)10—21(—28) nodes, 10—25(— 35) X 3.5-8 cm, occasionally paniculate with branches from lower 2 or 3 nodes. Pedicels stout, 1.3—2 cm long, recurved at anthesis, erect in fruit; lowest bracts 2-4 x 0.5-1.2 cm, acute to acuminate, pale green, often purple-tinged along the margins, longer than pedicels, occasionally ex- tending beyond the flowers. Flowers campanulate and cernuous; tepals el- liptic to broadly elliptic, the margins often arching inward, 12—16 x 7— 9.5 mm, purplish-red or margins purplish-red blending to pale yellow-green or green near midvein abaxial side, mostly lighter colored adaxial side; apices rounded, occasionally slightly emarginate; gradually tapering to base or broadly clawed; veins greenish, arching towards apex; glands 5—G6.5 X 3.2-5(-6) mm, slightly expanded terminally, apices emarginate or undu- late, somewhat bilobed, greenish-yellow; stamens with filaments 7-8 mm long, dilated at base, 1.5—2 mm see tapering to 0.5 mm wide at anther connection, anthers 2.4—2.6 X 1.7—2 mm, horseshoe-shaped, dehiscing entire length, sone striate, ee 30-42p X 18-24p; pistil tricarpellate, ovary partially inferior, trilocular, 3 styles and stigmata; ovules many, flac, aus lanceolate. Capsules 3-lobed, 12-15 mm long, tepal lobes originating 4—5 mm above pedicels; no seeds seen. Flowering late August to early September. Fruiting September. HEss AND SIVINSKI, A new species of Zigadenus from New Mexico 391 tt , | CEOS N aad e i 15cm y Zl ( Fic. 1. Zigadenus mogollonensis. A. Habit. B. Flower with outline of glands on tepals. C. Flower and bract. D. Pendulous flowers 392 Sipa 16(3) 1995 Type: U.S.A. NEW MEXICO. Catron Co.: Mogollon Mts., of the Gila Wilderness, Gila National Forest, vicinity of Mogollon Baldy, Whitewater Baldy, Black and Sacaton on Little Dry Creek Trail from Apache Springs, elev. 2860 m, 20 Aug 1968, W Hess 3213 2 (HOLOTYPE: MOR; Isorypes: ARIZ; NCU; NMC; NY; OKLA; SMU; US). Additional collections examined: U.S.A. NEW MEXICO. Catron Co.: Mogollon Mts., S of Whitewater Baldy, center of sect. 30, T11S R1I8W, elev. 3140 m, understory of ae mature forest, 9 Sep 1980, R. Fletcher 4856 (UNM); 15 mi NE of Mogollon, | Aug 1938, C.L. Hitchcock, R.V. Rethke, R. van Raadshooven 44603 (WTU); along Bursum Road E of Mogollon, TIL1S R18W sect. 3, elev. 2770 m, 19 Jul 1994, C.A. Huff 1660 aes between Apache Cabin & Sacaton, Baldy, Mogollon Mts., 9000 ft, 20 Jul 1959, Kriuckebere 4671 (WTU,; UC); Gila Forest Trail 182, ca. 2.5 a S of NM 78, T118 sections 2 and 11, elev. 3050 m, 2 Aug 1987, G.A. Levin 1909 (NMC); Gila Wilderness, Whitewater Balay spruce-fir ores with Vaccininm Ke and Rubus parviflorus, infre- quent, elev. 3020 m, 2 Aug 1974, WH. Moir 315 (NMC; RM); Mogollon Mts., Forest Trail . 182, Sandy Point, first 2 mi, ponderosa pine- ic forest, 15 Aug 19 : eae E. Lehto & T. Reeves P12558 (ARIZ; W TU); Gila National Wilderness, Mogollon ,ca 10 air mi ESE of Mogollon and | mi from Hwy 78 on Trail 206 to Redstone Park, on W-slope, soil rocky clay with humus, Populus tremuloides, Pteridinm, and Rubus, elev. ae m, 5 Sep 1976, J. Reitzel, D. Hill & R. Spellenberg 33 (NMC); Mogollon Mts., Forest ervice Road 159 between Silver Creek Divide and Sandy Point, T11S R18W sect. 3 ay '/4, on rhyolitic soil, N-facing slope in understory of mixed conifer forest of ela menziesii, Abtes concolor, Acer glabrum, Geranium abe Viola canadensis, elev. 2800 n 12 Aug 1993, R. Siinski & K. Lightfoot 2517 (MOR, UNM); on NM Hwy 78 ae Mogollon and Beaverhead, T eat IS R18W sec ere /s, with spruce, fir, limberpine, aspen thimbleberry, Geranium, Cornus, Oxalis, Sener cardamine, elev. 2770 m, 16 Aug 1985, R Spellenbere & N. Zucker 8237 (NMC); Mogollon Mts., Mogollon Road, 8 Aug 1900, E.0. Wooton sn. (US Zigadenus mogollonensis is apparently endemic to the mixed conifer and spruce-fir forests of the Mogollon Mountains between 2650 and 3200 meters in elevation. It is a common understory component of these high elevation forests and usually occurs on highly organic soils with a thick humus layer. Variation within this species is represented by its completely racemose or branched lower inflorescence, its range of tepal lengths between 12 and 16 mm, and corolla pigmentation of pale yellowish-green with pale purple margins to green with dark brownish-purple margins. The exceptional speci- men of Reztze/ et al. 33 (NMC) has green tepals with only little anthocyanic pigmentation, and is here placed within Z. mogollonensis because of its long (13 mm) tepals and campanulate corolla. Previous collectors have usually identified this species as Z. virescens, but also as Z. e/egans Pursh and porrifolius Greene. The nearest related species to Zigadenus mogollonensis are Z. elegans and Z. virescens. Both are widespread species, with Z. e/egans ranging from Alaska to northern Mexico and Z. virescens occurring in the Sierra Madres of Mexico north to the southern mountains of Arizona and New Mexico. Zigadenus mogollonensis and Z. elegans have a similar inflorescence, which is usually racemose, but can have one or few lower panicled inflorescence branches. HEss AND SIVINSKI, A new species of Zigadenus from New Mexico 393 Rotate to rotate-campanulate corollas, erect pedicels at anthesis, and shorter (S—10 mm) tepals are characteristics of Z. elegans, whereas Z. mogollonensis is distinguished by its campanulate corollas, cernuous flowers at anthesis, and longer (12-16 mm) tepals. Zigadenus virescens is a more delicate plant with thinner pedicels and panicle branches, shorter bracts (usually less than the length of the pedicel), small (4-7 mm) tepals, campanulate corollas, and recurved pedicels at anthesis. Zigadenus mogollonensis has more robust pedicels and branches, larger bracts (as long or longer than the pedicels) and much larger tepals. The anterior margins of its tepals are also suffused with anthocyanic red over yellowish-green, which gives the flower a brown- ish-purple appearance. Both Z. elegans and Z. virescens frequently have anthocyanic bracts; however, this pigmentation infrequently extends to the flowers, and then is confined to the base of the tepals on the abaxial surface. Their open flowers are white, ochroleucous, or pale green. The references to purplish-flowered Z. e/egans in western American floras (Correll & Johnston 1970; Harrington 1964) are overstated and apparently derived from the description of purplish tepals for Z. coloradensis Rydberg (1900), which is a synonym of Z. elegans. The type specimen of Z. coloradensis is white-flowered, with small amounts of anthocyanic pigment on the abaxial surface of the tepals. The purplish pigmentation of Z. mogollonensis is much more intense and colors the anterior margins of the tepals. Zigadenus elegans is very rare in the Mogollon Mountains of New Mexico and is known from that range by a single 1881 collection (Rusby 406 US, only the flowering plant on sheet). However, Z. e/egans is more common in the adjacent White Mountains of eastern Arizona and northern mountain ranges of Arizona and New Mexico. Zigadenus virescens is common in the Mogollon Mountains and occurs with Z. mogollonensis on the same forested slopes near Silver Creek Divide. No evidence of hybridization was observed at this point of sympatry. Pollen samples from four Z. mogollonensis and two djacent Z. virescens were tested with Alexander’s stain and found to be 99- 100% viable. Late summer flowering times here are somewhat overlapped, but most Z. virescens flowered a little earlier and produced fruit at the time Z. mogollonensis was in full flower. A few unusual specimens of Z. virescens from the northern Mogollon Mountains (Daniel & Nelson 3598 ASU; Hubbard s.m. UNM) have large, leafy bracts that could possibly indicate a past cross- ing event with Z. mogollonensis. Zigadenus mogollonensis appears most similar to the large-flowered forms of Z. volcanicus Benth. in the Sierra de los Cuchumatanes in the Huehueten- ango district of western Guatemala. These two populations have long tepals, large bracts, cernuous/campanulate flowers and thick pedicels. They are differentiated by the somewhat larger, purple tepals of Z. mogollonensis com- pared to the white to ochroleucous tepals of the plants in Guatemala. 394 Sipa 16(3) 1995 ADDITIONAL COMMENTS ON SECTION ANTICLEA The above species belong to the section Anticlea of the genus Zigadenus, which is characterized by a single bilobate or obcordate gland at the base of each tepal and a partially inferior ovary. Preece (1956) recognized five American taxa (Z. elegans var. elegans, Z. elegans var. glaucus (Nutt.) Preece, Z. virescens, Z. vaginatus (Rydb.) J.E. Macbr., Z. volcanicus) in this section with one, or possibly two, additional species in Asia. Turner (1992) re- cently added another Mexican species (Z. Aintoniorum B.L. Turner) to the section Anticlea. The species of this section are usually distinct, but among the taxa intergradations and inconsistent variations are common. Species diagnosis must often rely on a combination of descriptive, qualitative fea- tures rather than definitive, measurable characteristics. Species descriptions that do not account for the range of variation have made this genus misun- derstood, and is evidenced by a long list of synonymy and the questionable rank of a few taxa. After studying the types and several hundred specimen sheets, the following discussion attempts to clarify some of the morpho- logical tendencies of the American taxa in the section Anticlea of Zigadenus. The two most important characteristics in separating taxa in the section Anticlea are the position of the flower (erect or nodding) and flower shape (rotate or campanulate). The flowers at anthesis are either on erect pedicels with perianth widely spreading or rotate-campanulate, or cernuous pedicels with perianth consistently campanulate. Unfortunately, these characteris- tics are often difficult to assess on pressed and dried specimens, and collec- tors are well advised to note the floral aspects of fresh plants. It is also important to assess these characteristics on flowers in anthesis, since taxa with erect flowers can have cernuous buds, while those with nodding flowers will usually become erect in fruit. Plants with erect/rotate flowers are best represented by the Z/gadenus elegans complex. This is the common North American species that extends from Alaska and eastern Canada to the southern Rocky Mountains and is sporadic in northern Mexico. It is very polymorphic, but retains some con- sistency by its erect pedicels and rotate corollas. Its flowers can often be attached to the erect pedicels at an angle that turns the faces of the flowers toward the outside of the inflorescence and perpendicular to its axis. This may give the false appearance of nodding flowers in some pressed speci- mens. Such minor and inconsistent variation is especially common in the mountains of central New Mexico, but occasionally occurs on specimens rom Colorado, Utah, northern Arizona, and Chihuahua. This variant of- ten, but not always, occurs in combination with very large bracts. The eastern variety, Zigadenus elegans var. glaucus, and a low elevation species from the canyonlands of western Utah, Z. vaginatus, are the only HEss AND SIVINSKI, A new species of Zigadenus from New Mexico 395 other erect pedicel-rotate corolla taxa presently recognized. The inflores- cence of Z. vaginatus is similar to Z. elegans, particularly with those variants from central New Mexico that have a perpendicular flower attachment be- low the calyx. Most flowers of Z. vaginatus are often smaller, but within the range of typical Z. elegans. Cronquist et a/. (1977) considered Z. vaginatus a synonym of Z. elegans. Preece (1956) and Welsh (1989) maintained these taxa as distinct species because the former occurs at lower elevations in hanging garden seeps on canyon walls and has a later flowering period of July through September. Z7gadenus elegans in Utah and Colorado grows at higher elevations and flowers earlier during June and July. Preece (1956) and Welsh (1989) stated that Z. vaginatus is closely related to Z. volcanicus in Guatemala. However, we believe it is more closely related to Z. elegans because of its erect pedicels and rotate corollas rather than to Z. volcanicus, which has cernuous/campanulate flowers. Zigadenus elegans also occupies similar habitats and tends to flower later south to the Guadalupe Moun- tains on the New Mexico/Texas border. A feature unique to Z. vaginatus is its numerous hard, persistent leaf bases, which allows it to be separated from Z. elegans. Zigadenus elegans is rare in northern Mexico, with a few known locations in Chihuahua and Coahuila. The Mexican populations differ from typical Z. elegans and have been treated variously. For instance, there is a large- Hlowered (tepals 10 X 5 mm long), large-braccted variant with rotate corol- las and perpendicular flower attachment on erect pedicels called Z. mohinorensis Greenm. Greenman (1903) does not mention flower angle, but it is evident on the type specimen (Ne/son 4875 US). Preece (1956) placed Z. mohinorensis, known only from a few collections from Mt. Mohinora in Chihuahua, into synonymy with Z. elegans, but with some reservation. Zigadenus gracilentus Greene (1901) was described, based on plants col- lected in the Sierra Madre of Chihuahua (Pringle 1383 F, NY, US) and origi- nally identified as Z. elegans. They had campanulate corollas, tepals 6—7 mm long, short stamens, and widely spreading pedicels. Greene described this new species as strictly dioecious and with different tepal characteris- tics for the staminate and pistillate plants. There is some variation in tepal morphology; however, the types studied for this discussion had obvious stamens and maturing ovaries. Preece (1956) placed Z. gracilentus into syn- onymy with Z. e/egans. In addition to the campanulate corollas and long, spreading pedicels, the flowers might have been cernuous at anthesis. The inflorescence bracts are narrow, acuminate, and 2/3 the length of the pedicels, and are identical to the long, linear-bracted form of Z. virescens that 1s com- mon in the northern Sierra Madres of Chihuahua and Sonora, and the Huachuca Mountains of southern Arizona. Zigadenus gracilentus is interme- 396 SipA 16(3) 1995 diate between Z. virescens and Z. elegans and cannot be comfortably placed into synonymy with either species. The precise type locality of Z. gracilentus is unknown, but if ic were to be relocated and the population had plants with consistently erect pedicels and campanulate corollas, ic might be wor- thy of taxonomic recognition. South of the Rocky Mountains to Guatemala, the dominant form of Z7- gadenus has cernuous/campanulate corollas and is best illustrated by the variable Z. virescens. This species occurs predominantly in Mexico and usu- ally characterized by small tepals (4-7 x 1.5—3.5 mm), exserted stamens, short bracts, and slender, recurved digs Zigadenus volcanicus, compared to Z. virescens, ranges further south into Guatemala and has slightly to greatly larger flowers (tepals 6-15 mm long), included stamens, and broader tepals (3-6 mm). Zigadenus mogollonensis is a New Mexico endemic with large purple flowers. There are many examples of herbarium specimens with inconsistent char- acter combinations for particular species. For instance, intermediate forms between Z. elegans and Z. virescens, within a fairly broad area of sympatry in eastern Arizona, southwestern New Mexico and northern Mexico, frequently have the tepal measurements, thicker pedicels and somewhat longer bracts of Z. elegans, and cernuous/campanulate corollas of Z. virescens. They are the dominant form in some localized areas (1.e., Mc. Baldy in the White Moun- tains of eastern Arizona). They also are in mixed populations since they share the same herbarium sheet with typical Z. verescens (.e., White Mts., AZ, Peebles 12522 ARIZ) or with typical Z. elegans (i.e., Mogollon Mts., NM, Rusby 406 US). Several attempts have been made to recognize taxonomically these scat- tered, larger-flowered forms of Zigadenus virescens. The southwestern New Mexico plants were named Z. porrifolius Greene (1881), a name later ap- plied to several collections from eastern Arizona and an aberrant form from southern Coahuila (Hemsley 1885). In 1940, O. S. Walsh annotated many of these specimens as Z. virescens var. porrifolivs, a combination that was never published. Kearney and Peebles (1951) placed Z. porrifolius into syn- onymy with Z. virescens, but stated that it might be worthy of varietal status. Preece (1956) placed Z. porrifolius in synonymy with Z. virescens and attributed its larger flower size to the influence of Z. elegans. Zigadenus porrifolius is probably not worthy of taxonomic distinction because of its sporadic distribution and mixed populations. However, it is frequent enough to cause difficulties in species determination. In eastern and southern Mexico, Z. virescens may also intergrade with Z. volcanicus and possibly Z. elegans. There are several collections with long bracts, thick pedicels, and broader tepals of Z. volcanicus and Z. elegans, but with the distinctly decurved pedicels, campanulate corollas, and longer HEss AND SiviNski, A new species of Zigadenus from New Mexico 397 stamens of Z. virescens. A few larger-flowered collections from the border area between Coahuila and Nuevo Leon strongly resemble Z. volcanicus, and the collection (McDonald 1522 TEX) from Coahuila could easily be placed within that species. This, and most of the tall forms of Z. virescens in eastern Mexico, have recently been named Z. Aintoniorum (Turner 1992). The holotype of Z. intoniorum is somewhat aberrant, but it and the paratypes are clearly the highly variable species Z. virescens. In fact, the broad limits given to Z. Aintontorum would include Z. porrifolius and almost all of Z. volcanicus. Some Zigadenus from the Sierra Madre Orientale display odd character combinations that may be taxonomically divisible on a smaller scale; however, Z. bintoniorum is not correctly circumscribed and represents another attempt to segregate several intergrading and variable populations that cannot be comfortably assigned to Z. virescens, Z. elegans, or Z. volcanicus. Zigadenus volcanicus is a poorly understood species that has been incor- rectly represented by Preece (1956) and Turner (1992). Preece (1956) de- scribed Z. volcanicus as a short plant (2—3.5 dm) with erect or spreading pedicels and a corolla diameter of 1—1.4 cm, which could only be accurate if the corolla were rotate. However, among the specimens Preece annotated as Z. volcanicus, was one 7.5 dm tall (Sefer 23235 US), and the type specimen (Hartweg 626 NY) was cut into several sections to fit on the sheet! These, and all other specimens of Z. volcanicus seen in this study, were greater than 3 dm tall and had cernuous/campanulate flowers. Turner (1992) inadequately characterized this species in his key to the section Anticlea and distinguishec Z. volcanicus on the basis of tepals 14-15 mm long at anthesis. His descrip- tion was obtained from a single Guatemalan specimen (Beaman 3097 TEX), which represented either an extreme variation or undescribed species with very large flowers that are not at all typical of Z. volcanicus. Duplicates of this unusual collection (Beaman 3097 US and MSC) have somewhat smaller tepals (12-13 mm long), less paniculate inflorescences, and a strong resem- om blance to Z. mogollonensis. However, Z. mogollonensis is readily distinguished from it by its purple flowers and occurs 3,000 km north of the Guatemalan plant. The type specimen and original description of Zigadenus volcanicus (Bentham 1842) represented a species with tepals 6-8 mm long. Standley and Steyermark (1952), from an independent collection (Steyermark 50153 US) for their Flora of Guatemala, agreed with Bentham (1842). Baker (1879) recognized Z. volcanicus as a cernuous-flowered species and distinguished it from Anticlea mexicana Kunth (= Z. virescens) by its somewhat broader tepals. Many of the specimens of Z. volcanicus that we studied could be further distinguished from Z. virescens by their longer bracts (often as long or longer than the pedicels) and stamens that are shorter than the tepals. The com- bined features of broader tepals and longer bracts also describe Z. porrifolius 398 Stipa 16(3) 1995 and Z. Aintoniorum (both = Z. virescens). Like Z. elegans, bract length was not consistent in Z. volcanicus and varied from one-half to two times the length of the pedicel. Bract length in Z. virescens is also variable, but is usually less than the length of the pedicel. Other distinguishing, but inconsistent, fea- tures of Z. volcanicus were noted in this study. Several specimens displayed a wavy, and often twisted, central axis of the inflorescence. The flowers were often clustered at the ends of long panicle branches. The significance of these characteristics must wait for the collection and study of additional specimens. The variation in Zigadenus volcanicus suggests that it may be further di- visible, at least at an infraspecific level, and ict is tempting to name the large-flowered form on the Sierra de los Cuchumatanes in northern Guate- mala. However, in several collections from that location, the tepals range in length from an extreme 15 mm to the 7 mm of the more typical form. One collection (Steyermark 50317 F) from Sierra de los Cuchumatanes is a mixed sheet containing a plant with tepals 10 mm long and another 7 mm long. Further collections and field studies are necessary before a large- flowered taxon can be separated from Z. volcanicus. The following key to the taxa in Zigadenus section Anticlea accepts a broad concept of species variability and, therefore, somewhat artificial lim- its of taxon circumscription. The most ambiguous separation occurs be- tween Z. virescens and Z. volcanicus. Zigadenus hintoniorum will mostly key out to Z. virescens, but because there is no consistent and distinguishing criterion for Z. hintoniorum, it may also key to Z. volcanicus. We prefer to extend the limits of variation for Z. virescens and the geographic distribu- tion of Z. volcanicus rather than continue the use of a new and variable taxon that cannot be separated from existing species. Also, Z. porrifolius (= Z. virescens) in southeastern Arizona will occasionally key to Z. volcanicus. We suspect it is the result of intergradation with Z. e/egans, and not be- cause of a relationship to Z. volcanicus. KEY TO THE AMERICAN TAXA IN ZIGADENUS SECTION ANTICLEA . Pedicels erect at anthesis; corolla rotate to rotate-campanulate; plants of United States and Canada with sporadic populations in northern Mexico. 2. Lower stem sheathed with numerous hard, persistent leaf bases; he 1- 8 mm long; plants of hanging gardens on cliffs in southeastern Z.. Vaginalus an J. F. Macbr. 2. Old leaf bases not persistent; tepals 5-10 mm long; plants widespread. 3. Inflorescence usually racemose, occasionally with one or a few erect a ee e branches in the lower inflorescence; plants of western N Am elegans Putsh var. efegans 3, ia usually paniculate with a few to several lax branches, rare racemose, plants of eastern N. America ....... Z. elegans var. glaucus (Nutt.) Preece Hess AND SrvinskI, A new species of Zigadenus from New Mexico 399 1. Pedicels decurved or widely spreading, attached to cernuous flowers at an- thesis; corolla campanulate; plants of Arizona, New Mexico, Mexico, and Guatemala. 4. Tepals 4-10 mm long at anthesis. 5. Stamens usually as long or longer than the tepals; oe 1. a as ae mm wide and < 8 mm long; floral bracts usually shorte subtended ee Z. virescens nae F. Macbr. 5. Stamens shorter than the tepals; tepals 3-6 mm wide and > 6 mm long; floral bracts '/2 to longer than the length of the subtended pedicels Z. volcanicus Benth. 4. Tepals 12-16 mm long at anthesis. 6. Tepals eee with pale to dark purple la plants of conifer forests in the Mogollon Mts. of SW New Mexic Z. mogollonensis 6. Tepals ae to ochroleucous; plants of high sloine meadows on the Sierra de los Cuchumatanes of N. Guatemala cf. Z. volcanicus ACKNOWLEDGMENTS The senior author thanks Ian MacPhail for translating the description into Latin. Kim Altvatter and Nick Stoynoff accompanied him into the Gila Wilderness of the Mogollon Mountains for the field work, and he is eternally grateful. Floyd Swink did his superb editing for grammatical er- rors and Fayla Schwartz kindly reviewed the manuscript. The fine illustra- tion of this new species is by the hand of Nancy Bartels. The junior author thanks the curatorial staffs ac ARIZ, ASC, ASU, CAS, DS, EF, LL, MOR, C, NY, TEX, UNM, RM, US, and WTU for their prompt atten- tion to loans of specimens and type materials. He also appreciates the ef- forts of Guy Nesom at TEX, Steve Reed at UNM, and Susan Richardson at US in locating and providing several obscure pieces of literature. REFERENCES Baker, J.G. 1879. The aberrant tribes of Liliaceae. J. Linnean Soc., Bot. 17:480—483. BENTHAM, G. 1842. Plantae Hartwegianae. pg. 96. CorreLt, D.S. and M.C. JoHNsToNn. 1970. Manual of the vascular plants of Texas. Texas Reseeich Foundation, Renner, Texas Cronouist, A., A.H. HotmGren, N.H. Hoimoren, J.L. Revear and P.K. HoLMGREN. 1977. MUSA ices ee Volume 6. The New York Botanical Garden, Bron oe . I . New species of plants, chiefly New Mexican. Bull. Torrey Bot. Club 8:12 _ pee : L. 1901. New or noteworthy species XXVIII. Pittonia 4:238—241. GREENMAN, J.M. 1903. New and otherwise ea ee, angiosperms from Mexico and entral America. Proc. Amer. Acad. Arts 39:69-1 Harrinctron, H.D. 1964. Manual of the ane ona Swallow Press, Inc., Chicago. Hemstey, W.B. 1885. Biologia Centrali-Americanae, Botany 3:382. , T.H. and R.H. Peesres. 1951. Arizona flora. University of California Press, Berkeley. 400 Sipa 16(3) 1995 Preece, S. 1956. A cytotaxonomic study of the genus oo (Liliaceae). Unpubl. Doc- heat Eanes Washington State University, Pu RypperG, P.A. 1900. Studies on the Rocky Mountain flora_—Il Bull. Torrey Bot. Club 27:528-538. STANDLEY, P.C. and J.A. STEYERMARK. 1952. Flora of Guatemala. Field Mus. Nat. Hist., Bot. ser. 24(3) TURNER, B.L. 1992. A new species of Zigadenus (Liliaceae) from eastern Mexico. Phytologia 78-382. Wris LSH, i 1989. On the distribution of Utah’s hanging gardens. Great Basin Naturalist 49:1- — BOOK NOTICE Dr Waat, Louise C., Lois E. Cuitp, P. Max Wape, and Joun H. Brock (Eds.). 1994. Ecology and Management of Invasive Riverside Plants. (ISBN 0-471-94257-X, hbk.) John Wiley and Sons Led., Chichester, West Sussex, PO19 1UD, England. $95.00. 217 pp. This book, one in the “Landscape Ecology” series, seeks to bring together researchers and practitioners to solve plant invasion problems. Considering the fact chat many differ- ent habitats are invaded by non-indigenous plants, | was somewhat intrigued why the riverside habitat was chosen. The preface provides scant information on this problem; the only clue offered is that the river environment is “especially prone to invasion.” Among the 20 chapters, all concerned with plants that grow in or near rivers, 11 deal with control methods; 6, with basic biology of weeds; and 3, with rates of spread. Data from the Czech Republic, Denmark, England, Ireland, Scotland, Sweden, western United States, and Wales are presented. The book provides much information for the practitioner wanting to elimi- nace weeds. However, it not much use for those wanting to understand how and why riparian environments are invaded. Furthermore, there is scant information about the po- tential ecological functions of non-indigenous plants in riverside habitats.—James O. Luken. UNA NUEVA ESPECIE DEL GENERO PTEROPEPON (CUCURBITACEAE) DE COLOMBIA ALVARO COGOLLO P. Fundacion Jardin Botdnico, “Joaquin Antonio Uribe” Apartado Aéreo 51407 Medellin (Antioquia) COLOMBIA JOHN J. PIPOLY III Botanical Research Institute of Texas 509 Pecan Street Fort Worth, TX 76102-4060, U.S. A. RESUMEN Se describe y se ilustra Preropepon oleiferum y se discute su parentesco. Es una especie unica dentro del género y de la familia debido al hecho de que tiene solamente un estambre con una antera bilocular insertada en un filamento conico, y ad es dioica. Sin embargo, esta mejor colocada en Preropepon que en cualquier otro géner ido hasta la fecha. Esta especie es a tercera especie nueva descrita de la regidn de Rio Gli (Antioquia) en los tones tres afios, como resultado de las 1 investigaciones de los autores. La zona donde crece n alto porcentaje de la flora aparentemente endémica, presentando también muchos ae disyuntos y otros aspectos biogeogrdficos interesantes. ABSTRACT s discussed. ec ee is described, illustrated, and its phylogenetic relationship It isa un ies within the genus and family, based on its one stamen and one bilocular anther on a conic filament, concomitant with its dioecious habit. Nevertheless, we feel that the new entity is best placed in Preropepon than in any other cucurbitaceous genus known, because of its laterally compressed, fibrous and winged fruit, and staminate flowers in panicles. As a result of the authors’ collaborative research, this is the third new species to be described from the Rio Claro area of Antioquia since 1991. The area is calcareous, apparently with a large percentage of the flora endemic, disjunct, or otherwise biogeographically interesting. INTRODUCTION El género Preropepon (Cogniaux) Cogniaux (1916) fue descrito para incluir las especies de Sicydium Schlechtendal que son monoicas, con flores estaminadas en paniculas, las pistiladas en pares, anteras uniloculares y sésiles o subsésiles; frutos lateralmente comprimidos, fibrosos y alados. Las otras especies que quedaron en Sicydium son dioicas, con flores estaminadas y pistiladas en paniculas, el androceo de 2 anteras con 2 tecas y una con una teca, filamentos pequefios, y bayas globosas. En Das Pflanzenreich, Cogniaux reconociO dos especies, ambas ae sureste de Brasil. Veinte afios después, Stipa 16(5): 401 — 406. 1995 402 Sipa 16(3) 1995 Macbride (1937) describi6é una variedad de Preropepon deltoideus var. peruvianus, segtin las determinaciones de Harms, para entonces llegar a un total de 3 taxa en el género. Desde entonces, Jeffrey (1978) y Kearns (1993) no han reconocido var. peruvianus como entidad meritoria, aunque no han explicado el porqué. Adicionalmente, Crovetto (1950, 1952) describid Preropepon parodii and P. argentiniense, respectivamente, y cred la secci6n Micropteropepon, para incluir las especies que tienen frutos pequefios y membranaceos, incrementando el total de los taxa hasta cinco. Como resultado de las investigaciones en la regidn de Rio Claro en el Valle del Rio Magdalena en Colombia, encontramos una especie que se ubique mejor en este género que en cualquier otro. La especie es otro ejemplo del alto nivel de endemismo en esa zona, como fue el caso con otras especies pertenicientes a los géneros Erythroxylum (Cogollo & Pipoly 1993) y Cybianthus (Pipoly 1991). La nueva especie aqui descrita se separa de las especies previamente conocidas por la siguiente clave. CLAVE A LAS ESPECIES DEL GENERO PTEROPEPON 1. Plantas didicas; frutos pequefios, membranaceos, bases y apices emarginados sect. Micropteropepon Crovetto 2. Inflorescencias euance 8-12 cm de largo 0, mayores las hojas; 1. S; flores pistiladas brevemente racimosas; laminas tenuement membranaceas racimosas; | Tdlliitlds COMMU P. parodii Crovetto 2. Inflorescencias estaminadas 1.5—5 cm de largo, menores o subiguales a las hojas; flores pistiladas solitarias; laminas subcoridceas 0 coriaceas P. argentiniense Crovetto 1. Plantas monoicas; frutos grandes, fibrosos, bases estipitados, apices apiculados sect. Preropepon 3. Lamina eee triangular o subdelroidea, apice acuminado, base etpenee, estambres 3, anteras uniloculares. 4.Sépalos lanceolado-triangulares; pétalos lanceolado-lineares, papilosos; frutos costados P. monospermus (Veloso) Cogniaux 4. Sépalos subulados; pétalos anchamente ovados, glabros; frutos no costados leltoidens (Cogniaux) Cogniaux 3. Lamina coridcea, ovada 0 oblongo-elfptica, dpice agudo, base redondeada; estambre ];anteras biloculares .......... P. oleiferum Cogollo and Pipoly Pteropepon oleiferum Cogollo and Pipoly, sp. nov. (Fig. 1) Quoad flores staminatos paniculatos denique fructum complanatum certe generi Preropepon pertinet, sed a specibus caeteris foliis coriaceis (non membranaceis) ovatis vel oblongo-ellipticis (nec triangularibus vel subdeltoideis) ad apicem acutis (nec acuminatis) ad basim rotundatis (nec truncatis), scamini | (non 3) necnon antheris bilocularibus (nec unilocularibius) praeclare distinguitur. Trepadora dioica; tallos graciles, surcados, glabrescentes, ramificados. Pecfolos graciles, 2.5-4.2 cm de largo, glabrescentes; laminas rfgidas, COGOLLO AND Pipoty, A new species of Pteropepon from Colombia 403 ey. SS, otis oo — we) Se 6 o-66~ f&D / fiesta C1994 _ 1. Preropepon oleiferum Cogollo a Pipoly A. Habito, mostrando el fruto lateralmente comprimido, y caracteres foliares. B. Inflorescencia estaminada. C. Sépalos y pétalos estaminados. D. Detalle sepalar, ea las papilas. E. Estambre ents los dos léculos de la antera. F. Semilla. A y F: de A. Cogollo & R. Borja 764; B-E: del holotipo coridceas, ovadas a oblongo-elfpticas, (6.5—)7.5—16.0(—17.0) cm de largo (4.5—)8.5—10.0(-13.2) cm de ancho, el apice agudo, la base redondeada, glabras por ambas caras, las nervaduras laterales 3 a cada lado, impresas a ligeramente prominentes en la cara superior, prominentes en la cara infe- rior, el margen entero; zarcillos graciles, redondos a ligeramente surcados 404 Sipa 16(3) 1995 bifidos en el apice. Inflorescencias estaminadas paniculadas; pedinculo prin- cipal plurifloro, gracil, surcado, glabro a ligeramente papiloso, 3.5—30.0 cm de largo, las ramas divergentes, las basales de 2.5—12.0 cm de largo, mis cortas hacia el apice, subtendidas por unas bracteas de 2.0—7.0 mm de largo. Flores estaminadas pequefias, verdosas, 2.0—2.5 mm de didmetro; receptaculo glabro, rotado; sépalos 5, oblongos, atenuados en el apice, uninervios, abaxialmente papilosos, adaxialmente glabros; corola dividida hasta el receptaculo; pétalos 5, ovados, glabros, alternandose con los sépalos, un poco mas grandes que los sépalos; estambre |, pequefio, el filamento cénico, de 0.3—0.5 mm de largo, la antera 1, amarilla, bilocular, l6culos ca. 0.5 mm de largo, 0.3 mm de diametro. Flores pistiladas desconocidas. Fruto fibroso, coridceo, eliptico a obovado, glabro, 9.0—-10.5 cm de largo, 6.0-8.0 cm de ancho y 2.0—2.5 cm de espesor, alado, con 2 costillas prominentes largo, 4.0-5.0 cm de on longitudinales. Semilla 1, elfptica, de 5.0-6.0 cm de ancho, verrugosa, con testa lenosa. Tipo. COLOMBIA. aang er Municipio de Puerto Triunfo, (ek renee Santafe de Bogota, sector Rio Claro-Rio Magdalena, camino de la autopista hacia la gruta ae Condor,” ca. 400 m de la autopista; 05° 56'N, 74° 5O'W, 350-400 m, 27 Feb 1994 (fl. estam.), A. Cogollo & G. ia 8020 (HoLotTIPO: JAUM; isoTipos: BRIT, COL, a HUC, JBGP, MEDEL, Paratipos. COLOM re Anco Munici S uis, Canon de Rio Claro, margen derecha, sector sur, 05° 53 el 1° 39 W, 52541 Om, 3 a 1983 (fr), A. Cogollo & R. Bory 764 (COL, JAUM, MO), ° 1983 (fl. estam.), A. Cogollo 1040 (COL, JAUM, MO): corregimiento “El Prodigio,” vereda ° . Confusas,” 06° 03'N, 74° 47'W, 350-500m, 6 Mar 1990 (fl. estam.), D. Cardenas & J. Ramirez 2545 (COL, JAUM, MO); corregimiento | Prodigio,” Cafio El Tigre, 06° 06'N, 74° 48'W, 350-700 m, 21 Nov 1990 (fr), D. Cardenas & J. Ramirez 3034 (COL, JAUM, MO); municipio de Puerto Triunfo, autopista Medellin- Bogota, sector Rio Claro- Rio Magdalena, camino de la autopista ba la gruta de “El Condor,” ca. 400 m de la autopista, 05° 56'N, 74° 50'W, 350-400 m, 27 Feb 1994 (fl), A. Cpille & G. Jaramillo 8021 (BRIT, COL, HUA, HUC, JAUM, JBGP, MEDEL, MO) Distribucién.—Pteropepon oleiferum es conocida hasta ahora unicamente de la regién de rfo Claro, en las faldas orientales de la Cordillera Central de los Andes, hacia el valle medio del Rio Magdalena, en los municipios de San Luis y Puerto Triunfo, Departamento de Antioquia, Colombia, entre 300 y 700 m de altitud Etimologia.—E] epiteto especifico “oleiferum” es del latin, “oleum” (aceite) con el sufijo adjectival “-fer” (de portar, o contener); asf que se refiere al hecho de que la semilla contiene aceite. Ecologra.—P. oleiferum crece en orillas de quebradas, en bordes de bosques y en rastrojos; principalmente en suelos calcdreos; sus frutos se dispersan por agua. Usos y Condiciones Actuales en Cuanto a Conservacion: Las semillas de COGOLLO AND Pipoty, A new species of Pteropepon from Colombia 405 P. oleiferum contienen una gran cantidad de aceite, el cual es extraido en forma rudimentaria por algunos habitantes de la region, y es usado para lubricacién de herramientas. La gente local aprecia el valor de la especie y por lo tanto, en algunos lugares hay planes de domesticarla. Pteropepon oleiferum es una especie tinica dentro del género, por su flor estaminada con un solo estambre, antera bilocular, insertado en un filamento cénico, y por su condicién didica. Sin embargo, debido a sus frutos lateralmente comprimidos, fibrosos y alados; flores estaminadas y paniculadas, P. o/e¢ferum no podria ser acomodada en otro género conocido. El bosque de la zona del Rio Claro sobre rocas calcdreas contiene no solamente las tres nuevas especies susodichas, sino otra medio docena de entitades no identificables hasta la fecha y posiblemente nuevas también. Estudios actualmente progresando indica que la zona tiene una mezcla rara de elementos chocoanos, andinos, amaz6énicos y centroamericanos, verificando su importancia como fuente de biodiversidad Unica en el departamento de Antioquia. AGRADECIMIENTOS La colaborac6én de nuestros compafieros del herbario respectivos se ha facilitado el trabajo para ambos autores, y les agradecemos. En particular, | trabajo de Dayron Cardenas (COA), Luz Marina Vélez, y Juan Guillermo Ramirez nos facilit6 mucho en compilar los datos sobre la especie. Lindsay Woodruff revisé el manuscrito y hizo comentarios valiosos. Apreciamos el apoyo financiero para el primer autor del Fondo Colombiano de Investigaciones Cientificas y Proyectos Especiales, “Francisco José de Caldas”-COLCIENCIAS lo que ha permitido realizar exploractions en la region de Rio Claro; también el apoyo de la National Geographic Society, cuya subvencién para estudios en el Parque Nacional Natural “Las Orquideas” nos ha permitido reunirnos bianualmente tanto para estudios de la flora del parque como estudios sobre regiones adyacentes, como el presente. También, comentarios sobre el manuscrito, hechos por Michael Nee (NY), Charles Jeffrey (K), y Dennis Kearns (MO) enriquecieron y mejoraron la presentaci6n. Apreciamos el trabajo de la dibujante Consuelo Garcia (HUA) por la ilustraci6n aqui presentada. REFERENCIAS CoGniaux, A. 1916. Cucurbitaceae- niece et Melothrieae. Pp. 1-265. In: A. Engler, (ed.), Das Pfanzenreich 66(1V, 275—1):1— Coco.tio, A. and J. Pirory. 1993. ae nueva ee del género Erythroxylum (Erythroxylaceae) de Antioquia, Colombia. Novon 3:126—128. Crovetto, R. 1950. Nueva especie de “Preropepon” Cees de la fora Argentina. Bol. Soc. Argent. Bot. 3:174-176 406 Sipa 16(3) 1995 —________.. 1952. El género Preropepon (Cucurbitaceae) en la Republica Argentina. Bol. Soc. Argent. Bot. 4:177-182. JerrReY, C. 1978. Furcher notes on Cucurbitaceae. 1V. Some New World taxa. Kew Bull. 33:347-380 Kearns, D. 1993. Cucurbitaceae. Pp. 377-384. In: L. Brako and J. Zarucchi (eds.), Cata- logue of the eae as plants and Cpe: of Peru. Monogr. Syst. Bot. Missouri Bot. Gard. / AS. Macsripe, J.F. “Grbac Pp. 321 eo n: J.B. Macbride (ed.), Flora of Peru. Field Mus. Nat eee Bot. Ser. cn 322-3 ee, Pipoty, J. 1991 1eS eee ybianthus suabgénero Conomorpha (Myrsinaceae) de Cisnbe Claas ere eae 64 BOOK NOTICE LEAKE, DoroTHY VANDyKE, JOHN BENJAMIN LEAKE, and MARCELOTTE LEAKE Roeper. 1993. Desert and Mountain Plants of the Southwest. (SBN 0-8061-2489-x, pbk). University of Oklahoma Press, P.O. B. 787, Norman, OK 73070-0787. $18.95. 239 pp A problem with plant identification guides such as this one lies not in what is svcluded but what is excluded. By implication, the book includes desert and mountain plants from western Texas to southern California and north through the southern half of Nevada, Utah, and western Colorado (map on page 2), a large order. The “over 250 species” repre- sented in the book are thus but a fraction of the total desert/mountain flora of this region. True, the geographical coverage shrinks markedly on page vii, where it is said to be espe- cially “the pare of the Sonoran Desert located in Arizona.” The book’s title is obviously misleading. The illustrations are line drawings, some of which show well the features of the plants pictured. Placement of the illustrations is erratic—some on pages facing the text, some on ae before or after the text (text page-numbers are not given at the illus- trations). A “Guide to plant families,” which attempts to lead the user to the ay ope ee “group ae the plant in hand, is not highly successful. Following an appendix— “Plant characteristics illustrated,” which novices should find useful—are a glossary, bibli- oe. and index. The book could have used the services of a botanically astute and ritical editor.—John W. Thieret A NEW TACHIGALI (FABACEAE: CAESALIPINIOIDEAE) FROM WESTERN AMAZONIA JOHN J. PIPOLY HI Botanical Research Institute of Texas S509 Pecan Street Fort Worth, TX 76102-4060, U.S.A. RESUMEN describe una nueva especie del género Tachigalt, T. vasquezi1, procedente de la cuenca amazénica colombiana y peruana. Se ilustra la especie y se discuten sus relaciones filogenéticas. Ademas, otras especies anteriormente clasificadas en el género Sclerolobium, que se encuentran en varias flérulas amaz6nicas actualmente en proceso de redaccién, 0 con las que tienen parentesco, se cransfieren al género Tachigali, resultando aan nuevas sade ciones Tachigali rugosa (Martius) Zarucchi & Pipoly, T) micropetala (Ducke) Zarucchi Pipoly, Tachigali brlateaia (ams Zarucchi & Pipoly. Tachigali micropetala y T. bracteosa son nuevas Citas para Peru. ABSTRACT A new species of Tachigalt, T. vasquezii, is described from the Amazon Basin of Peru and Colombia. The new species is illustrated and its phylogenetic relations are discussed. In addition, other species aah classified in the genus Sc/erofobinm now being treated in amazonian florulas in final preparation, or closely related to them, are transferred to Tachigali, resulting in the new combinations: Tachigali rugosa (Martius) Zarucchi and Pipoly, T.: micropetala (Ducke) Zarucchi and Pipoly, and 7: bracteosa (Harms) Zarucchi and Pipoly. The occurrence of T. micropetala, and T. bracteosa are new reports for Peru. INTRODUCTION Tachigali Aublet is a neotropical genus now considered to contain ap- proximately 60 species, including the 35 species formerly included in Sclerolobium Vogel (Zarucchi & Herendeen 1993). Historically, the genera were distinguished by position of the piscil stipe with reference to the re- ceptacle cup (Sclerolobinm, centrally; Tachigali, eccentric) more than any other character (Dwyer 1954, 1957). However, Zarucchi and Herendeen indi- cate that there is a continuum in the stipe character and no other either morphologcial or anatomical character may successfully be used to separate the two groups. Both are frequently myrmecophilous, a trait otherwise rare in the Caesalpinioideae. While carrying out a series of florulas in the western Amazon Basin of Peru and Colombia, a new taxon was discovered which is assigned to Tachigali and described herewith. Sipa 16(3): 407 — 411. 1995 408 Stipa 16(3) 1995 Tachigali vasquezii Pipoly, sp. nov. (Fig. 1) Ob ramulos angulosos, laminas coriaceas ad apices acuminatos ad bases truncatos, stipulas bifoliolatas ovatas coriaceasque ad T. rvgosa valde arcte affinis sed ab ea ramulis foliolisque subter chocolati (non aureo-) tomentosis, foliolis oblongis (non lanceolatis) desuper planis (nec bullatis), denique petiolulis teretibus (non subteretibus), 7-13 (nec 3-5) mm longis praeclare distat Emergent canopy tree to 30(-45) m tall, 57 cm DBH; branchlets subterete to 3—5-angled, bark brown and gray in alternating longitudinal bands, 0.8—1.2 cm diam., minutely chocolati-tomentellous, early glabrescent; pith large, soft; stipules foliaceous, bifoliolate, coriaceous, ovate, |.8—3.5 cm long, 1.3—2.8 cm wide, apex acute, base asymmetric, truncate, midrib im- pressed aboue prominently raised below, secondary veins 5—9 pairs, bullate and densely chocolati-tomentose and glandular along the midrib and sec- ondary veins above, densely chocolati-tomentose below, the margin entire, revolute, glabrous. Leaves paripinnate, alternate, petiolate; petioles (S—)9— 14(-27) cm long, 0.7-1.5 cm diam., pulvinate basally; rachis solid, subcerete, with an adaxial furrow, without myrmecodomatia, densely and minutely chocolati-tomentellous, (14—)24—36(—47) cm long, apica — ly mu- cronate, the mucron 1—2 mm long, caducous, eglandular; petiolules terete, (0.7—)0.9-1(—1.3) cm long, 2—3 mm diam., densely chocolati-tomentellous, the tomentum persistent; leaflets coriaceous, oblong, 12—18(—31) cm long, 5—7(-13) cm wide, apex abruptly acuminate, the acumen (0.5—1 cm long, base truncate, midrib impressed above, prominently raised below, densely tomentulose, secondary veins 14—20(—28) pairs, densely tomentellous and somewhat impressed above, prominently raised and tomentose below, sub- marginally loop-connected to the next distal secondary vein, tertiary veins somewhat impressed above, prominently raised below, nitid above, densely chocolati-tomentellous-velutinous below, the margin revolute, entire. In- florescence an axillary panicle, 26-29 cm long, peduncle 6—9 cm long; pedicels 3—7 mm long. Flowers unknown. Fruit flat, oblong or rarely ellip- tic, 12.5—15 cm long, (3—)5—G6.5 cm wide, apex rounded, base acute, testa reddish-coffeate, exfoliate at maturity, one-seeded. peas PERU. Amazonas: Valle del Rio Santiago, 65 km N of Pinglo, Quebrada Caterpiza, 3 km behind Caterpiza, 200 m, 1 Feb 1980 (fr), Vi Huashikat 1910 (HOLOTYPE: MO; IsOoTYPES: AMAZ, UC). ParatyPEs. COLOMBIA. Amazonas: Municpio Leticia, Parque Nacional Natural Amacayacu, Quebrada Agua Pudre, 1.5 km NE of Rio pegs ae MO Strategy Inventory site, tree No. 158, 03°47'S, 70°15'W, 200-220 LOOT (Sten) J Pipoly et al. 15851 (BRIT, COL, FMB, MO). PERU. aes. = ann Codo de Pozuzo; alluvial floodplain of Rio Pozuzo after it emerges from mountains, trail N of ears » Rfo Mashoca, 75° 25'W, 9° 37'S, 500 m, 19 Oct 1982 (ster.), R. Foster 9329 (F, USM). Loreto: Prov. Maynas, Distrito Iquitos, Allpahuayo, Estaci6n Experimental del rae de Investigaciones de la Amazonfa Peruana, 04°10'S, 73°30'W, 150-180 m, Dec Popoty, Tachigali from western Amazonia 409 Fic. 1. Tachigali vasquezii Pipoly. A. Branchlet, showing the paired stipules and oblong leaflets with trunctae bases. B. Infructescence, showing fruits. Drawn from holotype. 1990 (ster.), R. Vasquez GN. gee ee (AMAZ, MO, USM). Madre de Dios: Prov. Manu, Parque Nacional Manu, Rio Manu, Cocha Cashu Station, 350 m, 23 Nov 1980 (ster.), R. Foster 5830 (F, USM), sre Station, Rio Manu, North Trail, 350 m, 20 Nov 1980 (fr), R. Foster 5780 (E, USM), Zone 1, one km N of camp, 11° 56'S, 71° 16'W/, 350 m, 22 Dec 1988 (ster.), R. Foster GS. Baldeén 12671 USM). Prov. Tambopata, Tambopata, along Rio Tambopata, 12°49'S, 89°18'W, 280 m, 19 Feb 1984 (ster.), A. Gentry et al. 45635 (AMAZ, CUZ, MO, USM); Tambopata Tourist Camp, at jct. of Rios ee and La Torre, 12°49'S, 69°43'W, 280 m, 22 July 1985 (ster.), A. Gentry et al. 51088 (CUZ, MO, USM), 26 May 1987 (ster.) A. Gentry et al. 57661, 30 May 1987 (ster.), A. Gentry et al. 57948; Las Piedras, Cusco Amazonico, Permanent Inventory Plot, 12°29'S, 69°03'W, 200 m, 25 Nov 1991 (ster.), M. Timand & N. Jaramillo 3404 (CUZ, MO, MOL, USM), 31 Oct 1991 (ster.), M. Timand & N. Jaramillo 2934 (CUZ, MO, MOL, USM); 18 Jun 1989 (ster.), O. Phillips et al. 412 (CUZ, MO), 21 Jun 1989 (ster.), O. Phillips et al. 553 (CUZ, MO, USM). Pasco: Prov. Oxapampa, Valle del Palcazu, Iscozacin, camino a Villa America, 400 m, 8 Aug 1981 (ster.), R. Foster 4688 (F, USM), PEPP Arboretum, 09° 50'- 10°45'S, 68° 00'-68° 30'W, 300-600 m, (ster.), G. Hartshorn, J. Quijano G E. Meza 2871 (CR, E, US, USM); Palcazu, Rio Alto Iscozacin, Ozuz to Rio Lobo, 10° 19'S, 75°16'W, 10 May 1985 (fr), R. Foster & B. d'Achille 10070 (F, USM). 410 Sipa 16(3) 1995 Common name.—‘chaira pacae amarilla” (Peru). Distribution: Throughout the western Amazon Basin, from near Leticia, Colombia, south and westward through the departments of Loreto, Amazonas and Pasco, to Madre de Dios, Peru, at 150-600 m elevation. In addition, reported to be common in the departments of Pando, Northern La Paz, and Beni, Bolivia (R. Foster- pers. comm.). Ecology.—Common in the canopies of humid forests, on mostly old allu- vial clays, especially sandy clays, or rarely, sands. T. vasquezii is frequently encountered on non-inundated terraces of old floodplains and low hilltops and slopes of the Upper Amazon and Andean foothills. Quantitative forest inventories containing permanent plots, established near Leticia, Iquitos and Cusco Amazonico have shown that 3—5 individuals of T. vasquezii may be found per hectare, sometimes clumped. Other data from those plots have revealed that T) vasquezii regularly grows in the same habitat as T- ptychophysca Spruce ex Bentham, but the two do not grow in close proxim- icy. According to Robin Foster (pers. comm.), this species is monocarpic— individuals flowering once and then dying slowly as the seeds mature. The behavior is similar to that described for T. versicolor (Foster 1977), except there seems to be greater frequency of individuals in which only a part of the tree flowers and dies within a given year. Etymology.—This species is dedicated to Rodolfo Vasquez Martinez, friend, colleague and director of the Peruvian exploration project for the Missouri Botanical Garden. He is principal author of the Florala of the Bio- logical Reserves of Iquitos, containing nearly 3,000 species. He is also an au- thority on the systematics of the Hypericaceae, Ebenaceae and Myrsiticaceae of Peru, and the genus Caraipa (Clusiaceae) throughout its range. The angulate branchlets, coriaceous leaves with acuminate apices and truncate bases, concomitant with the bifoliolate, ovate and coriaceous stipules, indicate that Tachigali vasquezii is most closely related to the vicariant 7. rwgosa (Martius) Pipoly and Zarucchi of the cerrado formations from eastern Brazil. However, T. vasquezii is easily recognized by the choco- late tomentum of the branchlets and abaxial leaflet surface, smooth oblong leaflets, and terete, longer petioles. Whether it is coincidence that the re- cently described Affonsea Pipoly and Vasquez, is also most closely related to an eastern Brazilian taxon remains uncertain pending further studies. NEW COMBINATIONS IN TACHIGALI As was previously mentioned in the introduction, Tachigali and Sclerolobinm cannot be separated (Zarucchi & Herendeen 1993) on any char- acter other than relative position of pistil stipe within the receptacle cup. The plasticity of this character, and lack of any other known character to separate the two groups necessitates the transfer of three species of Popoty, Tachigali from western Amazonia 411 Sclerolobinm related to, or occurring in florulas now in final preparation, to Tachigali. Tachigali rugosa (Martius) Zarucchi and Pipoly, comb. nov. Basiony™: Sclerolobinm pois Martius ex Bentham in Hooker, J. Bot. 11: 237. 1850. Type. AZIL. Maro Grosso: Cuiabé, da Silva Manso s.n., Martius herbarium 1155, (Ho- LOTYPE: M ae seen); photo at F Neg. 6264). Tachigali micropetala (Ducke) Zarucchi and Pipoly, comb. nov. BasionyM: Sclerolobium micropetalum Ducke, Bol. Tech., Inst. Agron. Norte Belém. Type. BRA- ZIL. AMAzONAS: Manaus, without date, Dacke 1219 (syntypes: K, MO, NY) Tachigali bracteosa (Harms) Zarucchi a a comb. nov. BasionyM: Sclerolobinm bracteosum Harms, Verhandl. Bot. Ver. Brandenb. 48:168. 1907. Type. BRAZIL. AMazonas: Rio Marmelos and Rio oo peas F- fragment, HBG, photos at K, M ). “~~ ACKNOWLEDGMENTS This paper is the result of research on Amazonian plant diversity con- ducted during my tenure at the Missouri Botanical Garden, under the aus- pices of the John D. and Catherine T. MacArthur Foundation. Supplemen- tal research on the Flora of Peru, which also contributed to fieldwork, was carried out through a grant from the Andrew W. Mellon Foundation. Rodolfo Vasquez, who provided the excellent illustration, is the director of the Peru program at the Missouri Botanical Garden, and is also supported by grants from the John D. and Catherine T. MacArthur Foundation for permanent inventory plot studies and the Andrew W. Mellon Foundation for the Peruvian exploration program. I thank Jim Zarucchi (MO), Alex Lasseigne and Robin Foster (F) for their reviews and helpful suggestions concerning the manuscript. — REFERENCES Dwyer, J. 1954. The poe American genus Tachigalia Aublet (Caesalpiniaceae). Ann. Missouri Bot. Gard. 41:223-260. 1954. eae entre stipe et coupe receptaculare dans la classification aes Acahernciene (Caesalpiniaceae). VIII. Cong. Int. Bot. Rapp. and Comm. Sect. 4:52 1954 «1957. The tropical American genus Scferolobinm Vogel (Caesalpiniaceae). Lloydia 20:67-118. 195 oeere eee eres 7. Supplement to The tropical American genus Sclerolobium Voge (Caesalpiniaceae). Lloydia 20:266—267. Foster, R. 1977. Tachigalia versicolor is a suicidal neotropical tree. Nature 268(5621):624— 52 Piece HI, J. and P. HERENDEEN. 1993. Tachigali (Fabaceae). Pp. 1254-1255. In: L. Brako, L. and J. Zarucchi (eds.), Catalogue of - eg: plants and Gymnosperms of Peru. a Syst. Bot. Missouri Bot. Gard. Stipa 16(3) 1995 aN — No BOOK NOTICES PoweLL, A. MicuacL. 1994. Grasses of the Trans-Pecos and Adjacent Areas. (ISBN 0-292-76553-3, hbk; ISBN 0-292-76556-8, pbk). University of Texas Press, P.O. pee 7819, Austin, TX 78713-7819. $75.00 hbk; $29.95 pbk. 377 pp. Texas west of the Pecos River is a wonderful land of deserts and mountains, with many peaks a mile or more high to 8751 ft on Guadalupe Peak). Powell’s book accounts for 83 genera and 268 species of Trans-Pecos Poaceae (half of the grasses known from Texas). Endemic to that part of the state are 53 species. Most of the illustrations are the by-now- so-familiar ones from the Hitchcock Manual (1951), Artist Manning contributed a few original drawings; other illustrations are 22 black-and-white photos of Trans-Pecos scen- Descriptions of species are given only in genera with but one species in the area; otherwise the genera alone are desc ribed, a disadvantage. Introductory chapters tell of the region, discuss grasses and grasslands, list che tribes and genera of Trans-Pecos Poaceae, and give a key to the 16 tribes included. Keys to genera and species are at ioe places in the “descriptive grass flora” (75% of the book). The keys worked well for the species of grasses I used to test them. Final chapters are a glossary, literature cited (ca. 1 entries), and an index. Frequently mentioned are taxonomic/nomenclatural aspects of vari- ous taxa. Nomenclature is up to date.—John W. Thieret Jones, Davib L. 1993. Cycads of the World. Ancient Plants in Today’s Landscape. (ISBN 1-56098-220-9, hbk). Smithsonian Institution Press, 470 LEnfant Plaza, Washington, DC 20560. $45.00 (plus $2.25 vostage). 312 pp. — This is a one-volume encyclopedia devoted to the 185 species of cycads. The illustra- tions—of plants, leaves, and cones—are the book's principal glory. Many (ca. 250) are fine color photos, some are original line drawings, and a few are reproductions of old engrav- ings, color or black-and-white, from 19th century and later botanical works. Part 1, about 30% of the volume, contain a most impressive wealth of information: discussions of the plants’ history and prehistory, conservation, structure, economic impor- tance, biology, cultivation, pests and diseases, and propagation. (Data on more recondite topics—e.g., gametophyte development, anatomy, chromosomes—must be sought else- 1ere.) On the world map of distribution of cycads their occurrence in the United States— Florida and Georgia—was somehow overlooked; the error is corrected in the map of Zamia. Part 2, the taxonomic section, sia a key to the 11 cycad genera and then an account ccounts include a key to species (exception: no such of each genus and species. key is given for Cyeas, es “and Zamia, because of the “confused state of the tax- onomy” or the “size and complexity” of these genera). Then follow paragraphs on deriva- tion of name, generic description, notable generic features, recent studies, habitat, cultiva- tion, and propagation. Species accounts include description, notes on distribution and habitat, other notes, cultivation, and ae The bibliography has about 200 refer- es; the index contains mostly scientific names. I recommend this excellently made book for cycadophiles and non-cycadophiles alike. i relatively low price makes it one of the book bargains of the year.—John W. Thieret. ANATOMICAL STUDY OF ERIONEURON AND DASYOCHLOA (POACEAE: CHLORIDOIDEAE: ERAGROSTIDEAE) IN NORTH AMERICA JESUS VALDES-REYNA Departamento de Botdnica Universidad Autonoma Agraria “Antonio Narro” Buenavista, Saltillo, Coahuila 25315, MEXICO STEPHAN 1. Aare S.M. Tracy Herbarium Department of Rangeland Ecology and Management Texas AGM University College Station, TX 77843-2126, U.S.A. ABSTRACT Leaf blades of Ertonenron and Dasyochloa species were analyzed for anatomical details. Analysis of transverse sections and epidermis show that both genera can be clearly delim- ited. Differences between E. avenacenm and E. ea aay were not found, supporting the interpretation that these are conspecific. Differences between E. avenaceum, E. pilosum, and E. nealleyi were found, suggesting that these be ee as distinct entities. Differences between Dasyochloa and Erioneuron support the recognition of Dasyochloa as a monotypic North American genus. RESUMEN Fueron analizadas laminas de las especies de Erioneuwron y Dasyochloa, eas un estudio anatémico detallado. El andlisis de la lamina en secci6n eanswense! ok a epidermis, muestra que los dos géneros pueden ser claramente definidos. No se encontraron diferencias tre E. avenaceum : Es; ae um, 1o que apoya la interpretacién de que no son especies oe Se encontraron algunas diferencias entre E. avenaceum, E. eee E. nealleyi, que if 2 sugieren que éstas se pueden mantener como especies distintas. Las diferencias entre Dasyochloa y Erionenron, apoyan el reconocimiento de Dasy ochloa como un género monotipico para Norteamérica. INTRODUCTION Caceres (1950) studied the leaf blade anatomy of Munroa mendocina and Blephavidachne benthamiana Hitche. and compared these species with spe- cies of Eragrostis and Tridens pilosa (Buckl.) Hitche. var. argentina (O. Kuntze) L.R. Parodi {syn. Erionenron pilosum (Buckl.) Nash var. longiaristatum (Kurtz) Anton}. Tateoka (1961) examined the leaf epidermis of Tridens, including Sipa 16(3): 413 — 426. 1995 414 Sipa 16(3) 1995 Erioneunron and he concluded that the epidermal features were of the Chloridoid subtype as described by Prat (1936). Sanchez (1979a) studied the leaf blade anatomy of three species of Tridens and two species, including their varieties, of Erzonewron from Argentina. Observations of the unique anatomical features of Dasyochloa pulchella (H.B.K.) Willd. ex Rydberg [syn. Evioneuron pulchellum (H.B.K.) Tateoka] were made. Sanchez (1979b, 1981) reported the e coven ment of the “Kranz” structure in the stems of Mwanroa, Blepharidachne, and Erioneuron from Ar- gentina. Sanchez (1983) analyzed ane anatomical differences between Dasyochloa and Ertonenron and proposed the segregation of Dasyochloa as an independent genus following, in part, criteria as outlined by Parodi (1934). The grass genus Erionevron was described by Nash in Small’s (1903) Flora of the Southeastern United States, based on Erioneuron pilosum. Tateoka (L961), in a biosystematic study of the genus Tridens, presented cytologi- cal, morphological, and anatomical differences to support recognition of two distinct genera. One prominent difference was in the basic chromo- some numbers, x = 8 for Erionenron and x = 10 for Tridens. The results of the examination of the leaf anatomy show that all species of Tridens and Erionenron have a Chloridoid subtype of the panicoid type of leaf epidermis, characterized by the presence of globose or club-shaped bicellular microhairs and saddle-shaped silica bodies. Particular kinds of bicellular microhairs were observed in Tridens and Erionenron. Tridens had bullet-shaped hairs, and Erronenron had ellipsoidal hairs. Differences in the leaf cross-section were found between genera in leaf margin, midrib of the leaf blade, and in the number of vascular bundles within the leaf. Based on these results Tateoka recognized the genus Erionewron and suggested affinities with the genus Muanroa, rather than with Tridens. Sanchez (1979b) studied the leaf anatomy of the species and varieties of Tridens and Erioneuron from Argen- tina. That study included a table of characters used to separate the genera, with keys to the species and varieties based on anatomical characters. This classification has been well accepted by agrostologists and used in several grass floras (Correll & Johnston 1970; Nicora 1973; Gould 1968, 1975, 1979; Anton 1977; and McVaugh 1983). The generic name Dasyochloa Willd. first appeared in Steudel (1840) as a synonym of Uralepis. Two species of Dasyochloa were listed, D. avenacea Willd. and D. pulchella Willd., both as “nomina nuda.” Although Rydberg (1906) validated Dasyochloa with an English description as part of a key, it has been regarded as a synonym of Er/onenron. Recently Caro (1981) trans- ferred the species of the genus Erionenron (excluding E. pilosum) to Dasyochloa, based on the lemma morphology. Sénchez (1983) reported that the ana- tomical features of Dasyochloa formed the most important difference within Vatbés-REYNA AND Hatcu, Anatomy of Eroneuron and Dasyochloa 415 Erioneuron s.\. She recognized Dasyochloa as a monotypic North American genus, consisting of D. pulchella. Leaf blade anatomy of North American species of Erzonexron has not been fully studied. This study analyzes the leaf anatomy of Erionenron and com- pares the results with the studies from Argentina. The results of this analysis will add to the anatomical characteristics of the leaf epidermis not reported by Sanchez (1983). MATERIAL AND METHODS Leaf blades were selected from population samples from across the geo- graphic range of each species. Table 1 lists the specimens analyzed. At least three leaf blades were selected from each specimen so that leaf cross-sec- tions and adaxial and abaxial epidermes could be observed and compared. Leaf blade sections one cm long were taken from the specimens. Leaf blades collected in the field were fixed in FAA for 24 hours and then transferred to 70% ethanol. Leaf blades from dried specimens were placed in a high mo- lecular weight solution of 1:3, photo-flo 200 and water until thoroughly imbibed, and then transferred to 70% ethanol. Leaf blades for cross-section analysis were embedded in paraffin using the standard techniques (Berlyn & Miksche 1976). Fifteen sections per plant specimen were sectioned with a rotary microtome at 10—15 pm thickness, stained with safranin and fast green, and made permanent with Permount. Adaxial and abaxial leaf epidermes were studied following rehydration in a high molecular weight solution. Leaf segments were then placed in a clorox solution for a few minutes to bleach the chlorophyll, placed flat on a microscope slide and scraped with a razor blade until the epidermis, meso- phyll, and vascular bundles were removed. The remaining epidermis was stained with Azo-Black, washed with drops of 90% ethanol, and made permanent using Euparal and ethanol. In some instances only small frag- ments of the epidermis could be obtained because of the furrowed leaf blades and sclerenchyma on the margins. In such cases the remaining epidermis was not stained to avoid the possibility of losing the tissue in the destain- ing process. The silica cells, silica bodies, microhairs, macrohairs, and prickle hair variations were recorded for each taxon. The descriptions of leaf transections and epidermal structures follow the terminology used by Metcalfe (1960) and Ellis (1976, 1979). All drawings are original and were made with the aid of a drawing tube. RESULTS AND DISCUSSION Examination of the leaf blade of Erzonewron confirms that this genus has the Cloridoid leaf anatomy as described by Brown (1958), Metcalfe (1960), 416 Sipa 16(3) 1995 TABLE 1. Specimens of Er‘onewron and Dasyochloa used in analysis of leaf transverse sections and epi- dermal scrapes, listed by species Set aaa country, state, collector and collection number. Speci- mens are deposited at ANSM and TAE Erioneuron avenaceum (H.B.K.) Tateoka MEXICO. San Luis Porosi: Guadalcazar, Valdés-R. 1641, 1636, 20 mi N of San Luis Potosi, Valdés- R. 1644, Enrique Estrada, Vald&-R. 1694. Coatniuita: Saltillo, Valdé-R. 1701. E. grandiflorum (Vasey) Tateoka MEXICO. Nuevo Leon: El Salero, Va/dés-R. 1623, 1627. E. nealleyi (Vasey) Tateoka MEXICO: Coanuita: Arteaga, Hatch et al. 5033, Saltillo, Hatch et al. 6045, 5050, Valdé&-R. 1455, Parras, Valdé&-R. 1576. UNITED STATES. Texas. Presidio Co.: Valdés-R. 1689 E. pilosum (Buckl.) Nash area CoanulLa: Saltillo, Valdés-R. 1502, 1537. Nuevo Leén: Cerralvo, Valdés-R. 1632 UNITED STATES. New Mexico. Grant Co. Velde: 1081. Texas. Travis Co.: Valdé-R. 1653. Dasyochloa pulchella (H.B.K.) Willd. ex MEXICO. Coanuia: Saleillo, Valdé-R. 1532, Hatch et al. SO55a. SAN Luis Potosi: Guadalcazar, Valdés-R. 1642, 20 mi N of San Luts ee Valdé-R. 1702. UNITED STATES. New Mexico. Grant Co.: \ Valdés-R. 1683, Otero Co.: Morden 672. Texas. Presidio Co.: Valdé-R. 1686, 1691. Tateoka (1961), Sanchez (1979a), and Renvoize (1983). Each bundle sheath, with its associated radial chlorenchyma, constitutes a discrete structural unit separated from similar adjacent units by large, clear bulliform cells. The radial chlorenchyma cells form one layer and are long, narrow, radially arranged, and contain few chloroplasts. This anatomical structure indicates that these grasses belong to the “C4 plants” or “Kranz PS” as described by Brown (1977). The leaf epidermis of these grasses corresponds to Renvoize’s (1983) general description of the Eragrostideae. Complete description of each species is given in Valdés-Reyna (1985). — ERIONEURON NASH Laminas are V-shaped, i.e. conduplicate, narrow to standard, the adaxial side slightly sinuous with ribs and furrows present. The abaxial side of the lamina is sinuous with well-marked ribs and furrows present, containing a prominent central vein. Sclerenchyma are associated with the vascular bun- dles (VB’s), and a small strand sometimes forms a girder between the bundle and the abaxial and adaxial epidermis. Bulliform cells are inflated and fan- shaped. The leaf epidermis contains intercostal long cells on a h epidermes that are moderate to strongly sinuous. Stomata occur in 1—2 rows per inter- costal zone with triangular subsidiary cells. Intercostal short cells occur as silico-suberose pairs between long cells. One papillae per cell occurs on the adaxial surface. Prickle hairs are distributed on intercostal and costal zones. Vatp£s-REYNA AND Hatcu, Anatomy of Eroneuron and Dasyochloa 417 Microhairs have a spherical distal cell. The macrohairs are unicellular. The costal zone has a row of short cells, each alternating with silica cells. Silica cells are mostly dumbbell- to saddle-shaped. Erioneuron pilosum (Buckl.) Nash Transverse section (Figs. 1 & 5).—Lamina is V-shaped, with a narrow to standard angle, 2.4 mm wide, 0.12 mm thick, 13—15 vascular bundles wide. The adaxial side is slightly sinuous with furrows and the abaxial side with slight to shallow furrows, the ribs well-marked. The prominent midvein has a well-developed keel, and the colorless cells, bulliform cells, and sclerenchyma cells are associated with the median vein. The bulliform cells penetrate through the abaxial epidermis. Sclerenchyma are absent adaxially and very abundant abaxially. One primary VB comprises the keel. The position of the VB is at the same level for all bundles of different orders. These bundles are situated in between the center and the adaxial surface. Three primary, (including the midvein), two secondary, and 10 tertiary VB’s are present, and round in outline. Mestome sheath cells are small, thick-walled, and surround the primary and secondary VB’s, inter- rupting the tertiary VB’s. Parenchyma sheath cells are large and thin-walled in the primary VB’s and continuous or sometimes interrupted by scleren- chyma. The sclerenchyma that is associated with primary VB’s forms a girder between the bundles and abaxial epidermis, and sometimes the adaxial epidermis. Secondary and tertiary VB’s have adaxial sclerenchyma present as a minute strand consisting of a few subepidermal fibers. Sclerenchyma cell walls are very thick; with a well-developed sclerenchyma cap present on the margins of the leaf. This sclerenchyma cap is not in contact with the lateral bundle, and is wider than tertiary VB’s. The chlorenchyma are radi- ate, one cell layer thick, in cabular arrangement, and interrupted by the sclerenchyma where associated with the bundle. The chlorenchyma of suc- cessive VB’s are separated by bulliform cells and colorless cells. The color- less cells form a girder-like extension to the opposite epidermis and are associated with the bulliform cells. The colorless cells are smaller than bulliform cells, not inflated, and have one extension from each group. Across most of the blade, 3—6 bulliform cells occur in the furrows, and are exten- sive over midvein and in the adjacent furrow. The epidermal cell walls are thick and covered by a distinct cuticle that is continuous over the epider- mal cells. Prickle hairs are present. Macrohairs are absent. Papillae are present on adaxial surface and restricted to the intercostal zone. There is one pa- pilla per cell that is relatively broad, but not much more than half the width of the epidermal cells. Abaxial epidermis (Fig. 9b).—The intercostal long cells have margins that are deeply undulating, strongly corrugated, 50-170 ppm long, and about 418 Sipa L6(3) 1995 100 um pel ho Fics. 1-4. Whole leaf blade transverse sections of Erionenron and betas Blackened 1 Fi areas indicate sclerenchyma tissue. Fig E. pilosum, Valdé 1502. Fig. 2. E. avenaceum, Valdés 1644. Fig. 3. E. nealleyi, Hatch a 5050. Fig. 4. D. ee Valdés 1680, Va.tprs-REYNA AND Hatcu, Anatomy of Eroneuron and Dasyochloa 419 10 ym wide. Stomata are in 1—2 rows per intercostal zone with triangular subsidiary cells. Intercostal short cells are solitary or paired between long cells; when paired the cells are silico-suberose, with tall and narrow smooth outlines. Papillae are absent. Bicellular microhairs occur as a single row attached to short cells, 20—25 pm long. The proximal cell is about 10 pm long, and the inflated and rounded distal cell is about 15 pm long. Prickle hairs are absent. Macrohairs are present on the intercostal zone. Costal zones have a row of paired short cells that are silico-suberosed and separated by normal costal short cells. These short cells have sinuous walls. Silica bodies, 8—10 pm long, are mostly dumbbell-and saddle-shaped, and relatively short. Adaxial epidermis (Fig. 9a).—Epidermis is similar to the abaxial surtace except for the following: intercostal zone long cells are slightly to moder- ately undulating; dome-shaped, inflated papillae are present, one per cell, with unthickened walls. Prickle hairs are infrequent on the intercostal zone. Erioneuron avenaceum (H.B.K.) Tateoka Transverse section (Figs. 2 & 6).—Lamina is V-shaped with a narrow to standard angle, 2.2 mm wide, 0.14 mm thick, 12—16 VB’s wide. The adaxial side has ribs with slight co medium furrows, less than a quarter of the leaf thickness. The abaxial ribs are the same size co taller than the adaxial, appearing moniliform. The prominent midvein has a keel with one VB comprising the keel. The colorless cells, bulliform cells, and sclerenchyma cells are associated with the midvein. The bulliform cells in the adaxial epidermis are located above the median bundle. Sclerenchyma are absent adaxially and abundant abaxially. The position of the VB’s is the same for all bundles of different orders. All bundles are positioned midway between the adaxial and abaxial surface. Five primary bundles (including the midvein) and 10 secondary bundles are present. Tertiary VB’s were absent. Primary bundles are circular to elliptical in outline and vertically elongated; while secondary bundles are circular in outline. Mestome sheaths are composed of 10-12 small, thick-walled cells surrounding the primary and secondary VB’s. The parenchyma sheath, of large thin-walled cells, 1s continuous or sometimes sclerenchyma interrupts the primary VB. Sclerenchyma are as- sociated with the VB’s on both the abaxial and adaxial surface and the strands consist of a few subepidermal fibers, sometimes forming a girder- like structure between the bundle and abaxial epidermis. The sclerenchyma cell walls are thick, with a sclerenchyma cap on the leaf margins. The chlo- renchyma are radiate, in tabular arrangement, and separated from succes- sive VB’s by the bulliform cells and colorless cells. Colorless cells form a girder-like extension to the opposite epidermis, and are associated with bulliform cells but smaller. These cells are not inflated, and form one ex- tension from each group. The furrows have 2—4 bulliform cells and are _— Stipa 16(3) 1995 Nine Boy 020% (+) Y Ge: ors “i WOS TC" eae SO. Valdés 16 Fics. 5-8. Detail of primary and secondary vascular bundles of Erionewron and Dasyochloa. hella, Fig. 5. £. pilosum, Valdés 1502. Fig. 6. E. avenaceum, Valdés 1644. Fig. 7. E. nealleyi Hatch and Valdés 5050. Fig. 8. D. pul Vatpfs-REYNA AND Hatcu, Anatomy of Eroneuron and Dasyochloa 42] extensive over the midvein and in the adjacent furrows. The bulliform cells are inflated and fan-shaped. The cuticle is continuous over the epidermal cells. Prickle hairs are present on the adaxial surface. Macrohairs are ab- sent. Papillae are present on the adaxial surface but restricted to the inter- costal zone. Abaxial and adaxial epidermis (Figs. 10a & 10b).—The anatomy of both surfaces is not different from that of FE. pi/osum epidermal surfaces; except that stomata, when seen, form one row per intercostal zone. Erioneuron grandiflorum (Vasey) Tateoka Significant differences were not observed between this species and E. avenaceum. Metcalfe (1960) reported the leaf anatomical features of E. grandiflorum {syn. Tridens grandiflorus (Vasey) Woot. & Standl.} and they correspond to our description of E. avenaceum. Erioneuron nealleyi (Vasey) Tateoka Transverse section (Figs. 3 & 7).—This species is similar to E. avenaceum and E. pilosum except that the lamina is 20-25 VB’s wide. The adaxial surface is slightly sinuous, and the abaxial surface is moderately to shal- lowly ribbed and furrowed. Abaxial snes (Fig. 11b).—Epidermis of FE. nea/eyi is similar to E. avenaceum and E. pilosum except that the intercostal long cell margins are moderately undulating. Adaxial epidermis (Fig. 11a).—Similar to the species mentioned, except that the prickle hairs are more abundant. DasyOCHLOA Dasyochloa pulchella (H.B.K.) Willd. ex Rydb. Transverse section (Figs. 4 & 8).—Lamina is U-shaped, without a definite angle formed with the midrib, 0.6-0.9 mm wide, 0.1 mm thick, 7 VB's wide. The abaxial and adaxial longitudinal ribs and furrows are monili- form. The bulliform cells are fan-shaped, the central one relatively small, and not much larger than bundle sheath parenchyma cells. Sclerenchyma are associated with the VB’s on the adaxial epidermis with a well-devel- oped strand that follows the shape of the adaxial rib. The well-developed sclerenchyma strand on the abaxial epidermis is wider than deep. The po- sition of the VB’s is at the same level for all orders. The three primary bundles and four secondary bundles are situated midway between the abaxial and adaxial surface. These types of bundles have a round outline. The thin- walled cells of the mestome sheath surround the VB’s. The parenchyma sheath, of large thin-walled cells, is continuous or interrupted by scleren- Sipa 16(3) 1995 aca ecco aa aa rio wie - SS apy ” = ; Cod co Se Leu INNO! LDN” S0cistagus toes OU RVOME Fics. 9-12. Detail of adaxial and abaxial surtace of leaf epidermis of Erionenron and Dasyochloa. Fig. 9. E. prlosum, Valdés 1653. Fig. 10. E. avenaceum, Valdés 1623. Fig. V1. E. nealleyt, Hatch and Valdé 5050. Fig. 12. D. pulchella, Valdés 1532, Morden 672. Vatpés-REYNA AND Hatcu, Anatomy of Eroneuron and Dasyochloa 423 chyma on the primary VB’s. The sclerenchyma are associated with the pri- mary VB’s forming a girder between the bundle and abaxial epidermis, and sometimes the adaxial epidermis. In the secondary VB’s of the adaxial and abaxial epidermis the sclerenchyma are present as a minute strand consist- ing of few subepidermal fibers. The sclerenchyma cell walls are thick, and the lumen almost excluded. The well-developed sclerenchyma cap is present on the margin of the leaf, but not in contact with the lateral bundle. Chlo- renchyma cells are radially arranged around the VB’s in one cell layer, and interrupted by sclerenchyma when associated with the bundle. The chlo- renchyma of successive VB’s is separated by colorless cells and bulliform cells. Colorless cells form a girder-like extension to the opposite epidermis and are associated with and smaller than the bulliform cells. These cells are not inflated. Bulliform cells form a girder with associated colorless cells. The cuticle of epidermal cell walls is thickened and the associated cell wall occupies less than half of the depth of the cells. Prickle hairs present on both epidermes. Macrohairs are present. Papillae are absent on both epidermes. Abaxial epidermis (Fig. 12b).—Intercostal long cells are 60-150 im long, is about 10 am wide, with deeply undulating margins. Stomata occur in | row per intercostal zone, and are dome-shaped, with rounded subsidiary cells. Intercostal short cells are solitary or paired and situated between long cells. Papillae are absent. Bicellular microhairs are 20-24 ppm long and occur over intercostal long cells in | or 2 rows. The proximal cell is about 10 pm long, and the distal cell is about 12 pm long, and appears inflated and rounded. Prickle hairs, attached to short cells, are medium sized with the base as long as the stomata. They occur over the intercostal and costal zones. Macrohairs are present. Costal zones have a row of short cells with sinuous walls. Silica bodies are mostly dumbbell-shaped. Adaxial epidermis (Fig. 12a).—This epidermis is similar to abaxial sur- face except for the following: The intercostal zone cells are slightly undu- lating; prickle and macrohairs are abundant. The descriptions of E. pilosum, E. avenaceum, and D. pulchella correspond with the anatomical features presented by Sanchez (1979a, 1983) and Caceres (1950). DISCUSSION Transverse section—In general Erioneuron is characterized by having a V- shaped lamina with ribs and furrows present. The abaxial ribs in Erzoneuron are sometimes the same size or larger on the adaxial surface depicting a moniliform structure. However, Dasyochloa pulchella has U-shaped lamina with abaxial/adaxial ribs and furrows that form the moniliform structure. The distribution of sclerenchyma in the leaf appears to be a reliable char- acter relative to taxonomic relationships. The sclerenchyma associated with 424 Sipa 16(3) 1995 the VB’s sometimes forms a girder between the bundle and the abaxial and/or adaxial epidermis, but generally the girder is represented by small strands. Dasyochloa pulchella difters from Erionenron by having well-devel- oped sclerenchyma girders on the abaxial/adaxial epidermis. The scleren- chyma associated with the primary vascular bundle comprising the keel have been a key taxonomic character for the separation of E. avenaceum and E. pilosum (Sanchez 1979). Erionexron pilosum has a prominent midvein with sclerenchyma occupying more than half of the keel, whereas E. avenaceum has one third or less sclerenchyma occupying the eeu. The bulliform cells of Erionenron species are fan-shaped as in D. pulchella, except the central one is smaller. Macrohairs are usually common on the adaxial epidermis of both genera. However, as mentioned by Sanchez (1983), D. pilchella macrohairs are more abundant on both epidermes. Leaf surfaces.—Features of the abaxial and adaxial epidermis are similar in most characters between the two genera. In Erionenron, papillae were present only on adaxial surface, whereas in D. pa/chella, they were not ob- served, Macrohairs are present in all species. However, they are more abundant in E. nealleyi on the adaxial surface, and in D. pulchella they occur on both surfaces. A summary of the differences of the anatomical characters are presented in Table 2. The blades of E. avenaceum and E. grandiflorum had no differ- ences in the anatomy. Dasyochloa pulchella has characteristics that differ from all Erionenron. Therefore, these data support the recognition of Dasyochloa and Erioneuron as anatomically distinct genera. TABLE 2. Diagnostic characters of the leaf anatomy in the genera Erronewron and Dasyochloa. Species are chen’ toas: A = D. pulchella, B = FE. prlosum, C - E. avenacenn, D = E. grandiflorum, and E = E. nealleyi. CHARACTER SPECIES B Cc D Transverse Section Lamina |) V-shaped, 2) U-shaped > ] l | Adaxial and abaxial furrows 1) slight-co-deep, 2) moniliform 2 l l | 1 Sclerenchyma associated with VB 1) few fibers, 2) a strand, 3) a girder 3 | 2 2 2 Bulliform cells 1) fan-shaped, regular, 2) fan-shaped, irregular 2 l | l I Epidermis ree eee with one papillae on long-cells 1) present, 2a ? | | Prickle hairs on adaxial surface 1) absent, 2) present l 2 ) 2 l X bo iw) VaLp£s-REYNA AND Hatcu, Anatomy of Eroneuron and Dasyochloa ACKNOWLEDGMENTS We thank Stanley D. Jones and J.K. Wipff (TAES) for their review and suggestions on this manuscript. This is Technical Bulletin TA 31740, Texas Agricultural Experiment Station. REFERENCES ANTON, A.M. 1977. Notas criticas sobre gramineas de Argentina. II El género Erionenron en Argentina. Kurtziana 10:57—67 BERLYN, a P. and J.P. MikscuHe. 1976. Beenie! microtechnique and cytochemistry. Ames: The Iowa State Univ. Press. Brown, W.V. 1958. Leaf anatomy in grass systematics. Bot. Gaz. 119:170-178. . 1977. The Kranz Syndrome and its subtypes in grass systematics. Mem. Torrey Bot. Club. net Caceres, M.R. 1950 5micos foliares de Munro locina y Blepharidachne gine Rex Argent. Agron. 17:233-240. . Rehabilitacién del género Dasyochloa (Gramineae). Dominguezia 2:1— 17, i oe Correll, D.S. and M.C. Jounston. 1970. Manual of the vascular plants of Texas. Renner, Texas: Texas Research Foundation. a R.P. 1976. A procedure for standardizing com - the leaf as viewed in transverse section. Bothalia 12:65-1(¢ . 1979. A procedure for standardizing easy aoe in the Poaceae parative ce anatomy in the Poaceae — II.: the epidermis as seen in surface view. aa en 12:64 GouLp, F.W. 1968. Grass systematics. New York: McGraw- Pil 1975. Grasses of Texas. oie Station, Texas: Texas A&M Univ. Press. . 1979. A key to the genera of Mexican grasses. Tex. Agr. Exp. Sta. MP- 1422. oe Station, TX. McVauecu, R. 1983. Hee Novo-Galiciana: A descriptive account the vascular plants of Western Mexico. Ann Arbor: The Univ. of Michigan Press. Vol. 14:436 MEtTcALrE, C.R. 1960. Anatomy of the monocotyledons, I. ae Oxford: Clarendon Press. Nicora, E.G. 1973. Novedades agrostolégicas patag6nicas. Darwiniana 18:80—106, f. 1-6. Paropt, L.R. 1934. Las gramineas del género Munroa. Rev. Mus. La Plata 34:171-193. Prat, H. 1936. La systematique des Graminees. Ann. Sci. Nat. Bot. ser. X, 18:165-258. ReNnvoize, S.A. 1983. A survey of leaf-blade anatomy in grasses IV. Eragrostideae. Kew Bull. pages RypserG, P.A. 1906. Flora of Colorado. Colorado as Exp. Sta. Bull. 100. SANCHEZ, E. 1979a. Estructura Kranz en tallos de Gramineae (Eragrosteae). Kurtziana | 2— 13:113-118. 1979b. Anatomia foliar de las especies y waniedades argentinas de los géneros Tridens Reeds: et Schult. y oo Nash (Gramineae Darwiniana 22:159-175, f 1981. Sink eine ore en tallos de Gramineae. Lilloa 35:37—40. 1983. Dasyochloa Willdenow ex Rydberg (Poaceae), genero monotipico de Norteamérica. Lilloa 36:131-138. SMALL, J.K. 1903. Flora of the southeastern United States. New York: Published by the author. Era orosteae). last lay 426 Stipa 16(3) 1995 StTeupeL, E.T. 1840. Nomenclator botanicus I. Typis et sumptibus J.G. Cortae, Sturtgartiae et Tubingae. pp. 731, 732 Tareowa, T. L961. A. biosystematic study of Tridens (Gramineae). Amer. J. Bot. 48:565— kDa —_______, S. Inoug, and S. Kawano. 1959. Notes on some grasses. IX. Systematic significance of bicellular microhairs of leaf epidermis. Bot. . 121:80-91. Vatprs-ReyNa, J. 1985. A biosystematic study of the genus eee Nash (Poaceae: Eragrostideae). Ph.D. Dissertation, Texas A&M University, College Station, Texas. BOOK NOTICE Connor, S. 1994. New England Natives. (ISBN 0-674-61350-3, hbk.). ers a Press, 79 Garden Street, Cambridge, MA 02138- 9983. $39.95. 274 pp. This book was written to reveal “the evolving interaction between the people anc . by describing the wide range of uses, during different eras, to le various species of trees and shrubs ... in this region ” Chapter 1, 1 che plants of New England . which men and women have put t and describing the qualities of these plants and noting their distributions ... "is devoted largely to relationships between plants and Native ’ focusses on European settlers. “A place in the forest, Americans of yesteryear. Chapter 2, “This wooden world,’ Chapter 3, “Trees in the Marketplace and in the Garden,” considers uses and decorative value of trees. And chapter 4, “The New Yankee Forest,” discusses ecology and more uses. A prominent member of the saline s dramatis personae is the Arnold Arboretum. The book is abundantly illustrated with 24 color plates of plants (mostly trees) and 194 black-and white illustrations (mostly photographs but also reproductions of old illustrations related to trees and tree products). Well written, the book certainly shows well the “overwhelm- ing significance of wood in the history of the Republic” and “the use of the forest as more than mere scenery.” It would be a fine addition to the list of “suggested readings” for any general botany or biology course or course on economic botany. And it is good reading for just about anybody interested in plants and people.—Johu W. Thieret. ON THE HYBRID NATURE OF QUERCUS BASASEACHICENSIS (FAGACEAE, SECT. QUERCUS) RICHARD SPELLENBERG Department of Biology New Mexico State University Las Cruces, NM 88003-0001, U.S.A. ABSTRACT ee hasaseachicenis C. H. Muller, collected by LeSueur in 1936 and named by Muller 1938, was not recollected until 1985. Since that time this white oak (sect. Qvercws) has ee discovered at several sites in the Sierra Madre Occidental in Chihuahua and extreme northern Durango, Mexico. For the first time fruiting material is described. Quercus basaseachicensis occurs with two other white oaks, the low, shrubby, rhizomatous Q. depressipes Trel. and the robust shrub or tree Q. rugosa Née, and is intermediate between them. Evi- including pubescence) is ae to “— dence based on habitat, habit, leaf morphology indicate the hybrid nature of Q. basaseachicensis. The name Q. Xbasaseachicensis C. H. Muller, pro sp., is proposed. RESUMEN Quercus basaseachicensis C. H. Muller, que fue colectado por LeSueur en el afio 1936 y descrito por Muller en 1938, no ha sido vuelto a recolectado hasta 1985. Desde entonces este roble blanco (secc. Quercus) ha sido encontrado en algunos lugares en la Sierra Madre Occidental de Chihuahua y en el norte extremo de Durango, México. Se describen por el primer vez plantas con flores y frutos. Quercus basaseachicensis convive con dos otros robles blancos muy diferentes, uno el pequefio, arbusto, rizomatoso Q. depressrpes Trel., el otro el arbéreo o arbusto grande, Q. rugosa Née, siendo intermedio entre los dos. Se presentan evidencias basadas en el habitat, habito, morfologia y indumento i las hojas para indicar la naturaleza hibrida de Q. basaseachicensis. Se propone el nombre Q. Xbasaseachicensis C. H. Muller, pro sp. INTRODUCTION Harde LeSueur, from the University of Texas, was the first botanist to collect at what is presently the Parque Nacional “Cascada de Basaseachic” on the west slope of the Sierra Madre Occidental in southwestern Chihua- hua, approximately 300 air km. W of Cd. Chihuahua. Among the exten- sive collections he made in 1936 was a peculiar white oak, which C. H. Muller named Quercus basaseachicensis in 1938. LeSeuer's specimens were vegetative and Muller’s new species, therefore, necessarily lacked descrip- tions of flowers and fruit. Nevertheless, on leaf morphology alone Muller placed the new species in the subgenus Lewcobalanus Engelm. (= subg. Quercus) and, with excellent insight, in the series Reticu/atae Trelease. Muller Stpa 16(3); 427 — 437. 1995 428 SIDA 16(3) 1995 noted that a study of the fruit would probably remove the species from this section, to which he felt it was only distantly related. He also noted that “at present there exists no other series to which the species could be re- ferred, and its characters are not sufficiently plain to serve as the basis of a new series.” With no additional information, Camus (1938-39) and Martinez (1956) followed Muller's classification of Q. basaseachicensis exactly. Nearly 50 years passed before this oak was collected again. In 1985, during general floristic collecting at Basaseachic, and in subsequent le of botanical exploration in the northern Sierra Madre Occidental, I and associated collectors discovered about 20 plants referable to Q. basaseac een (Fig. | and Appendix). A few of these were in flower or in fruit. This paper, then, reports upon these collections, provides a description of habit of the plant and its flowers and fruits, and provides morphological and ecological evidence that Q. basaseachicensis is a rarely formed hybrid, which at Basaseachic at least shows some segregration or backcrossing to the paren- tal types. A plant closely resembling the isotype was determined to have 2n=24 chromosomes (Rodriguez & Spellenberg 1992), the usual number in Quercus. The parental species are believed to be the extensively rhizomatous Q. depressipes Trel., a patch-forming low shrub, and the robust shrub or tree Q. rugosa Neée, two distantly related white oaks placed in different series (Trelease 1924; Camus 1938-39; Martinez 1956). This paper follows Nixon’s (1993) ine asenen: classification. He also notes that those oaks such as Q. depressipes, which are from Mexico and the southwestern United States and have connate cotyledons, form the G/awcoideae and probably are distinct at the subsectional level from those with free cotyledons, but a classification has yet to be devised. The hybrid between these very different white oaks that is under consideration here has not been observed to form self-per- petuating populations, and for this reason the hybrid name Q. x basaseachi- censis C. H. Muller, pro sp., is proposed (basionym: Q. basaseachicensis C. H. Muller. 1938. Amer. Mid]. Naturalist 19:582; Type: Mexico: Chihuahua, Cascada de Basaseachic, 6 Jul 1936, LeSueur 549; holotype: Muller’s pers. herb., transferred to BH; isotype: TEX! {photo at NMC}). — ADDITIONS TO DESCRIPTION OF QUERCUS X BASASEACHICENSIS Shrubs 1—1.8 m call, few to many stemmed, rarely rather extensively rhizomatous and forming patches up to ca. 5 m across. Staminate aments 15—42 mm long, with 4—20 flowers in the distal 80%, sparingly stellate tomentose; perianth sparingly to rather densely tomentose, 1.5 mm wide, about as long; anthers 3-7, glabrous, tan or reddish brown, 1.0—1.5 mm long. Pistillate flowers 1—4, in distal 1/3 of sparingly stellate pubescent to glabrate peduncle 15-41 mm long, usually only 1(—3) maturing. Cups SPELLENBERG, The hybrid nature of Quercus basaseachicensis 429 ARIZONA NEW MEXICO 110° SONORA CHIHUAHUA DURANGO Fic. 1. Locations in the northern Sierra Madre Occidental, northwestern Mexico, oe x basaseachicensis. Numbered circles correspond to sites described in Appendix and follow ing. Site 1, Chihuahua, Municipio Ocampo, Parque Nacional “Cascada de Basaseachic,” the type locality. Site 2,-Chih. SS : asas Grandes, 14.4 road km E of Altamirano. Site 3, Chih., Mcpio. I. Zaragoza, 1.1 km E of pass over Sierra Catarina. Site 4 nae Mcpio. ae 13 km W of crossing of Rio - Juan. Site 5, Chih., Mcpio. Guerrero, 14.4 road sm W of Tomochic. The small shaded circle is the location ar Cd. Chibctiun given for reference. hemispheric, 10-12 mm wide, 8-10 mm deep, the mature scales with reddish or brown densely appressed-pubescent bases, narrowed to thin, glabrate, reddish or tan, round or more or less acute tips that are puberu- lent on margins. Acorns ovoid, 11-12 mm long, 8-10 mm wide, light brown, about 1/2 included. Cotyledons pale pink, fused by their edges, separate in the center in the basal 2/3. HYBRID NATURE OF QUERCUS XBASASEACHICENSIS Oaks are notorious for the frequency of hybridization within subgenera (see, for example, Stebbins 1950, pp. 61-66). Evidence for the hybrid na- ture of Quercus Xbasaseachicensis is provided by the habitat and by macro- and micromorphological intermediacy. The hybrid is always uncommon. Presently QO. xbasaseachicensis is known from five sites (Fig. 1). In Sites 14 both putative parents are immediately sympatric; at Site 5 Q. x basaseachicensis Stipa 16(3) 1995 grew immediately beneath Q. rvgosa, but the nearest Q. depressipes (along the highway) was noted at ca. 1.5 km distant. Habitat. —Quercus depressipes and Q. rugosa have about the same elevational range in Chihuahua and northern Durango, that is, ca. 1950-2600 m, the latter presently known to extend somewhat higher than the former. Quercus x basaseachicensis occurs between 1980 and 2225 m. Quercus rugosa com- monly occupies sites seeming to be more mesic, commonly in canyons, on north- or northeast-facing slopes, and often on deeper soils. oe -s depressipes commonly occurs on open sites, often on thin rocky soi As 1s virtually common knowledge, and as reviewed by Grant (1 08 l,p L99 ff.), hybrid plants are particularly common in disturbed areas. At Basaseachic (Site 1 in Fig. 1) 12 putative hybrid plants have been located, all along the trail leading from the parking lot to the top of the falls. In this area, Q. rugosa is common, Q. a is rare. At Basaseachic Q. depressipes 1s Common on open rocky slopes with little tree cover, but in the forested area where Q. basaseachicensis plants occur Q. depressipes is not fre- quent. It may have been more common here in the past, and is now suc- cumbing to succession after fire (pines in the area are ca.125 years old or younger, and bases of many large Cupressus are fire-scarred). Where it oc- curs in the Sierra Madre, Q. depressipes may increase very rapidly in open areas generated by fire (R. Corral D. , pers. comm). One intermediate at Basaseachic, a few-stemmed shrub aie 1.2 m tall near the top of the stairs where the trail crosses the Rio Basaseachic, very closely resembles the isotype (TEX) (Fig. 2) in macromorphology and in the characteristics of the pubescence on the abaxial surface of the leaf (Figs. 4, 6) (in the various collections made during this study, this plant has on designated plant “#2”). No plants referable to Q. < fay 5 a qute ee der LENGTH, VERMIFORM Hairs DIAMETER VERMIFORM Hairs ca. 10 microns FRUITING peduncle 0.7—2.5 cm, INFLORESCENCE fruits 1-2 near tip, often only | maturing COTYLEDONS connate (except in center near base): flesh white Shrub, occasionally fairl rhizomatous, usually not l—-3 m pale bluish green to deep green 4—7.5 cm 1.9-S.1 cm Entire to toothed throughout Slightly rugulose che larger veins somewhat impressed; smallest veinlets somewhat paler than areolae ‘rom rather smooth and with cel convex and cells aire convex Minutely pubescent, with small golden vermitorm hairs, some o) ants also with sparse stellate hairs 0.09—0.21 mm ca. 20 microns eduncle 2—4 em, fruies 1—4 in upper 30%, often only 1 maturing connate at margins, center free; flesh pale pink Tree, occasionally non-rhizomatous - ub (2)3—20 m deep, dark green 5-13 cm —10.0 cm Usually irregularly toothed throughout, rarely 4 ntire Rugulose, all buc the smallest veins impressed; smallest veinlets notably paler than areolae Strongly bul late and y strongly convex Noticeably pubescent, with large golden vermiform hairs and with stellate hairs ca. 30 microns (2)6—13 cm, in upper 60% ped uncle ruics |— most mat uring free; flesh deep pink best to eliminate another common white oak, Q. arizonica Sarg., which occurs at Basaseachic sympatrically with Q. rugosa, as a parent instead of Q. rugosa. Characteristics of the piscillate inflorescence (Fig. 7) and the coty- ledons provide insight. Quercus ~ ww Distribution.—Australia: arid grasslands and shrublands of Western Aus- tralia, Northern Territory, South Australia, Queensland, and New South Wales. Setaria criniformis (S.T. Blake), R.D. Webster, comb. nov. Paspalidinm as Blake, Proc. Roy. Soc. Queensland 62:98. 1952. Type: 8. T. Blake 5282 (B Paspalidium gracile (R. Br.) Hughes var. debile Vickery, Contr. New South Wales Natl. Herb. 1:331. 1951. Type: J.H. Camfield 5, 1901 (NSW no. 9168). Distribution. —Australia: woodlands and shrublands of Queensland and New South Wales. sae oe (Trin.) Veldkamp, eos 39:376. 1994. Paspalidium distans n.) Hughes, Bull. Misc. Inform. 317. 1923. Panicum distans Trin., Spec. Gram. oe . 1829, non Salez. ex Steud. be Tyee: R. Brown 6098 (HOLOTYPE: K; ISOTYPE: BM!). es commixtum Steud., Syn. Pl. Glum. 1:59. 1853. Type: @’ Hie SM. ade Paspalidinm a Vickery, Contr. New South Wales Natl. Herb. 1:332. 1950. ne E. Cheel 3, 1907 (HoLtotyre: NSW, - Paspalidinm pea mai ee Proc. Roy. Oueendancs 84:65. 1973. Type: 8.7. Blake 19899 (HoLotypE: BRI). Distribution.—Australia: woodlands of Queensland and New South Wales. Malesia: New Guinea. Setaria hee ae (A. Rich.) Pilg. in Engl. & Prantl, Nat. Pllanzenfam. Ss 14e:72. 1940. Paspatidium wails wm (A, Rk: ) Davidse & R. a Now 106. 1992. Panicum distantiflorum A. Rich. in Sagra, Hist. Fis. one 850. Type: Sagra s.n. Gone P). gS. Distribution.—West Indies: Cuba and Bahamas. Setaria flavida (Retz.) Veldkamp, Blumea 39:376. 1994, Paspalidium flavidum (Retz.) A. Camus in Lecomte, Fl. Gen. de |’ fade: Chine 7:419. 1922. Panicum Haden Retz., Obs. Bot. 4:15. 1786. Type: Koenig s.n. (HOLOTYPE: LD?). Distribution.—India, southeast Asia, Pacific islands, and Australia (Queensland). Wesster, Nomenclatural changes in Setaria and Paspalidium 443 Setaria gausa (S.T. Blake) R.D. Webster, comb. nov. Paspalidium gausum S.T. Blake, Roy. Soc. Queensland 84:68. 1973. Type: S.T. Blake 19914 (HotoryPe: BRI). Distribution. —Australia: woodlands of Queensland and New South Wales. Setaria geminata (Forssk.) Veldkamp, Blumea 39:377. 1994. eae geminatum (Forssk.) Stapf, Fl. Trop. Afr. 9:585. 1920. Panicum geminatum Forssk., Fl. Aegypt.-Arab. 18. 1775. Type: Forsskal s.n. (HOLOTYPE: C). Distribution. —Native to Africa and Asia, now widely introduced in the tropics and subtropics of the world. Setaria geminata (Forssk.) Veldkamp var. paludivaga tae & Chase). R. iD ebster, comb. nov. ee ae geminatum (Forssk.) Stapf var. pala & Chase) Gould, Southw. Naturalist 15:391. 1971. Paspalidinm paludi ne & Chase) Parodi, Gram. Bonaer. ed. 3 3.85, 89. 1939. Paice valaduiawuan Hitche. & Chase, Contr. U.S. Natl. Herb. 15:132. 1910. Type: Nash 746 (HoLoryre: US!). Distribution. —Wet areas of United States a Mexico, Guatemala, and West Indies. Setaria niet (Domin) R.D. Webster, comb. nov. Paspalidinm oo (D ert. in) Hughes, Bull. Misc. Inform. 317. 1923. Panicum globoideum Domin, R nine Nov. Regni Veg. 10:119. 1911. Type: Wath 5.2, (LECTOTYPE: Distribution. —Australia: woodlands of Queensland and New South Wales. Setaria grandispiculata (B. K. Simon) R.D. Webster, comb. nov. Paspalidiam ial ee B. K. Simon, secre ce 465. 1982. Type: Peart 1990 (HOLO- : BRI!; tsorypes: CANBI, NSW'!). Distribution. —Australia: woodlands of Queensland. Setaria Penis (FE. Muell.) R.D. Webster, comb. nov. Paspalidinm ae (F. Mue iell.) Hughes, Bull. Misc. Inform. 317. 1923. Panicum inaequale F. aes 8:18 89. 1874. Type: Mueller s.n. (HOLOTYPE: MEL; tsoryPe: K) Distribution. —Australia: tropical woodlands of Queensland. Setaria ue (Trin.) R.D. Webster, comb. see Paspalidium ead eda n.) ng ee Bull. Misc. Inform. 317. 1923. Panicum jubiflorum Trin., es . 2:130. 1826. Panicum a var. lalate (Trin. Y Domin, scene "Bor. 30(85): 300. 1915. Type: Lindley s.n. (HOLOTYPE: E?). Distribution.—Australia: arid grassland to sub-humid woodlands of Western Australia, Northern Territory, South Australia, Queensland, and New South Wales. Setaria leonis (Ekman ex Hitchc.) Léon, Contr. Ocas. Mus. Hist. Nat. Cole egi io “De La Salle” 8: 163. 1946. Paspalidium leonts (Ekman ex Hitchc.) Davidse & R. W. Pohl, Novon 2:106. 1992. Panicum leonts Ekman ex Hitche., Man. Grasses W. Ind. 295. 1936. Type: Ekman 13155 (HoLotyee: US!). Distribution.—W/est Indies: Cuba. 444 Sipa 16(3) 1995 Setaria ophiticola (Hitche. & Ekman) pons pone Ocas. Mus. Hist. Nat. Colegio “De La Salle” 8:163. 1946. P la (Hitche. & Ekman) Davidse & R. W. Pohl, Novon 2:106. 1992. Dean ies Hitche. & Ekman in Hitche., Man. Grasses W. Ind. 293. 1936. Type: Ekman 12712 (Hotoryre: US!). Distribution.—West Indies: Cuba. Setaria pradana (Léon ex Hitche.) Léon, Contr. Ocas. Mus. Hist. Nat. Colegio ° ‘De La Sara 8:164. 1] 946. Pebalidinm pradanum (Léon ex Hitchc.) Davidse & R. W. , Novon 2:106. 1992. Panicum pradanum Léon ex Hitchc., Man. Grasses W. a er 1936. Type: Léon 11710 (HoLotypeE: US!). Distribution.—West Indies: Cuba. Setaria punctata (Burm. f.) oe Blumea 39:381. 1994. Paspalum punctatum (Burm. f.) Stapf ex Ridl. Malay Penins. 5:218. 1925. se aca punctatum (Burm. f.) A. = amus in nee Fl. Gen. Indo-Chine 7:419. 1922. Pani- cum punctatum Burm. f., Fl. Ind. 26. 1768. Type: Plikenet (LECTOTYPE, ene by Clayton & Renvoize ree M). — Paspalidium mucronatum (Roem. & Schult.) Ohwi, Acta Phytotax. Geobor. 11:33. 1942. Panicum mucronatum Roem. & Schult., Syst. Veg. 2:425. 1817. Type: Heyne (B, prob- ably destroyed). Distribution.—Africa, Asia, and Pacific Islands. Setaria rara(R. Br.) R.D. Webster, comb. nov. aaa ees 3ull. Misc. Inform. 318. 1923. Panicum rarum R. Br., Pre 189. LSLO. Type: R. Brown 6100 (HOLOTYPE: K). Distribution.—Australia: arid grasslands and sub-humind woodlands of Western Australia, Northern Territory, Queensland, and New South Wales. Setaria reflexa (R.D. Webster) R.D. Webster, comb. nov. aaa R.D. Webster, Australian Paniceae 166. 1987. Type: Latz 4847 (HOLOTYPE: Distribution. —Australia: arid grasslands of Western Australia, Northern Territory, and South Australia. Setaria retiglumis (Domin) R.D. Webster, comb. nov. Paspalidum retiglume (Domin) Hughes, Bull. Misc. Inform. 317. 1923. Panicum retigdume Domin, Repert. Spec. Nov. Regni Veg. 10:119. L911. Type: Mueller s.n. (HOLOTYPE: MEL; isotype: K). Distribution. —Australia: woodlands of Western Australia, Northern Ter- ritory, and Queensland. Setaria scabrifolia (S.T. Blake) R.D. Webster, comb. nov. Paspalidium scabrifolium 8.T. Blake, Proc. Roy. Soc. Queensland 84:69. 1973. Type: Blake 19983 (HOLOTYPE: BRI). Distribution.—Australia: woodlands of Queensland. Wester, Nomenclatural changes in Setaria and Paspalidium 445 ae i gras (S.T. Blake) R.D. Webster, comb. nov. Paspalidinm spartellum lake, Proc. Roy. Soc. Queensland 62:97. 1952. Type: Blake 9939 (HOLOTYPE: aca se se gracile (R. Br.) Hughes, Bull. Misc. Inform. 318. 1923 Panicum gracile R rodr. 190. 1810, non Setaria gracilis Kunth (1824), nec Spreng. ex Trin. (1835). - PE: a Brown 6096 (HOLOTYPE: K). Distribution.—Australia: woodlands, shrublands, and grasslands of West- ern Australia, Northern Territory, Queensland, and New South Wales. Setaria subtransiens Hitchc. & Ekman in Hitchc., Man. Grasses W. Ind. a5 1. 1 Be el subtranstens (Hitche. & Pleman) Davidse & R. W. Pohl, Novon 2 106. 1992. Type: Ekman 16828 (HOLOTYPE: US). Distribution.—West Indies: Cuba. — goatee (Hack.) R.D. Webster, comb. nov. Paspalidiu vlatum (Hack.) Hubb., Bull. Misc. Inform. 448. 1934. ee eon Hack, Bot. Jahrb. eee 1885. Type: Navmann s.n. (HOLOTYPE: K). Distribution.—Australia: shrublands and grasslands of Western Australia. Setaria uda(S.T. Blake) R.D. Webster, comb. nov. Paspalidinm udum S.T. Blake, Proc. Roy. Soc. Queensland 62:98. 1952. Type: S.T. Blake 16659 (HoLotyPe: BRI; NB) Distribution.—Australia: tropical woodlands of Northern Australia and Queensland. Setaria ese (Scribn.) Pilg. in Engl. & Prantl, Nat. Pflanzenfam. (ed 14e:72. 1940. Paspalidinm utowanaeum (Scribn.) Davidse & R. W. Pohl, eae. 7106: 1992. pees utowanaenm Scribn. in Millsp., Publ. Field Columbian Mus., Bot. Ser. 2:25. 1900. Type: Millspaugh Senet Ae F). Distribution. —West Indies: Greater and Lesser Antilles. ACKNOWLEDGMENTS Appreciation and acknowledgements are extended to Dr. John Wiersema for his critical review and significant recommendations on the nomencla- ture in this manuscript. REFERENCES Ciayton, W.D. and S.A. Renvoize. 1986. Genera graminum: Grasses of the World. Kew, London: Royal Botanical Gardens. Davipse, G. and R.W. Pont. 1992. New taxa and nomenclatural combinations of Mesoamerican grasses (Poaceae). Novon 2:81—110. Hitcucock, A.S. and A Ct HASE. 1910. The North American species of Panicum. Contr. S. Natl. Herb. 15:1 Hircucock, A.S. 1951. ee of the grasses of the United States, 2nd ed. Revised by A. Chase. Washington, D. C.: U.S. Dept. Agric. Misc. Publ. 200. 446 Sipa 16(3) 1995 Pircer, R. 1940. Gramineae.3. (Unterfamilie Panicoideae). In: Engl. & Prantl, Nat. Pflanzenfam, ed. 2, 14e:72, ee Romincer, J.M. 1962. Taxonomy of Setaria (Gramineae) in North America. Illinois Biol. Mon. 29:1-132. Srapr, O. 1920. Gramineae. In: D. Prain (ed.), Flora of Tropical Africa, vol. 9. London: L. Reeve & Co. VELDKAMP, J.F. 1980. Setaria clivalis (Ridl.) Veldk. comb. nov. (Gramineae). Liber gratulatorius in honorem H.C.D. de Wit. Misc. Pap. Landbouwhogeschoo!l Wageningen 19:315—320. serine a Paper 19. ane eee mises Vencous: notes on Southeast Asian Gramineae. LX. Setaria and Paspalidinm. eae 39:3 Wessrer, R.D. 1987. The eee Paniceae (Poaceae). J. Cramer, Berlin. . 1988. Genera of the North American Paniceae (Poaceae: Panicoideae). Syst. 13:576-609. WeBSTER. R.D. ¢ na. VaLprs-REYNA. 1988. Genera of Mesoamerican Paniceae (Poaceae: Panicoideae). Sida 13:187—2 Wesster, R.D., J. KirksrIDE, oe J. Vatpeés-ReEYNA. 1989. New World genera of the Paniceae (Panicoideae; Poaceae). Sida 13:393—4 Wessrer, R.D. 1992. Old World genera of the Paniceae (Poaceae: Panicoideae). Sida 15: 9-40. 1993. Nomenclature of Setaria (Poaceae: Panicoideae). Sida 15:447—489. A TAXONOMIC INVESTIGATION OF CUSCUTA ATTENUATA (CUSCUTACEAE) AND RELATED TAXA L. ALAN PRATHER' and RONALD J. TYRL Herbarium, Department of Botany Oklahoma State University Stillwater, OK 74078-0289, U.S.A. WILLIAM D. WARDE Department of Statistics Oklahoma State University Stillwater, OK 74078, U.S.A. ABSTRACT Examination of the taxonomic relationships of the rare parasitic vine Cuscuta attenuata Waterf. to other species of the genus reveals that it should be positioned in subsect. lwdecorae Yunck. rather than Lepidanche cay ae as previously placed. Analyses of morphological variation, via univariate analysis aN ses principal component analysis, and iscriminant analysis, indicate a vat it is distinct from C. compacta Juss. and C. cuspidata relm. but morphologically similar to C. indecora Choisy, especially var. — Yunck. as shee in a program of interspecific hybridizations reveal that C. attenuata 1s reproductively isolated from the other taxa. These data suggest that C. attentata Merits continued recognition as a distinct species. RESUMEN El examen de las relaciones taxonémicas de la rara parasita Cuscuta attenuata Con otras especies del género revela que debe ser colocada en la subsect. Idecorae en vez de la subsect. Lepidanche como se hab{a hecho previamente. Los andlisis de la variacién morfoldgica, mediante andlisis univariante, agrupamiento UPGMA, anadlisis de componentes principales y andlisis discriminante, indican que es distinta de C. compacta Juss. y de C. cuspidata Engelm aunque maadtolo cicamente semejante a C. /ndecora Choisy, especialmente a la var. loarepile unck. Los resultados obtenidos en un programa de hibridaciones interespecificas revela que C. attenuata esta aislada reproductivamente de los otros taxa. Estos datos sugieren que C. attenuata debe continuar reconociéndose como una especie distinta. INTRODUCTION Because of its rarity and uncertain taxonomic status, Cuscuta attenuata Waterf. has merited much attention (Tyrl et al. 1978; Taylor & Taylor 1980; Prather 1990; Prather & Tyrl 1993). The species was described from plants 'Present address: Plant Resources Center, Department of Botany, University of Texas, Austin, TX 78713-7640 Sipa 16(3): 447 — 458. 1995 448 Sipa 16(3) 1995 collected from two neighboring populations in the Red River floodplain of extreme southeastern Oklahoma (Waterfall 1971). Prior to 1989 the spe- cies was known from only four populations within a few km of one another (Tyrl et al. 1978; Taylor & Taylor 1980). Because of its limited geographi- cal distribution, it was considered to warrant possible designation as an endangered species and was declared a Category | species by the U. S. Fish & Wildlife Service's Office of Endangered Species (1980 FR 45:82500; 1985 FR 50:39526). Prather (1990) and Prather and Tyrl (1993) reported four extant and six historical populations in Kansas, Oklahoma, and Texas. Asa result, an status of the species was modified to Category 2 (1993 FR 26) bl ne ll (1971) tentatively placed C. attenuata in subsect. Lepidanche Engelm. on the basis of his interpretation that the calyx was polysepalous. He suggested, however, that formation of a new subsection to accommo- date the species might be appropriate because its capsule shape and the distribution of its floral bracts were not consistent with Yuncker’s circum- scription of subsect. Lepidanche (Yuncker 1965). Tyrl et al. (1978) echoed Waterfall’s uncertainty about the species’ relationships and called for taxo- nomic investigations of the species and its putative relatives. Among the species of subsect. Lepidanche, Waterfall stated that C. attenuata most closely resembled C. compacta Juss. but distinguished them on the basis that: (1) C. attenuata has a pedicel while C. compacta does not, (2) C. attenuata has only one floral bract which is at the base of the pedicel while C. compacta has 1-10 bracts which are situated along the length of the pedicel, and (3) C. attenuata has lanceolate, attenuate sepals while C. compacta has ovate, obtuse sepals (Table 1). Waterfall also stated that C. attenuata resembled C. cuspidata Engelm. in the presence of pedicels, which all other species in the subsect. Lepidanche lack. They are different in that C. cuspidata has a much more open inflorescence; ovate, cuspidate sepals; and usually one or two bracts along the pedicel (Table 1). During a preliminary examination of herbarium specimens, including the holotype, it was discovered that the calyx of C. attenuata is gamosepalous and not polysepalous as Waterfall stated in his diagnosis. Because of this fusion, C. attenuata clearly seems better placed within subsect. [ndecorae Yunck. rather than in subsect. Lepidanche. The two subsections are not thought to be closely related though both are positioned within sect. Cleistogrammica Engelm. (Yuncker 1932, 1965). With the exception of its somewhat dense inflorescence, all of the characters of C. atrenuvata are better accomodated in subsect. Indecorae (Table 1). Within the subsection, C. attenuata is morphologically similar to C. indecora vat. longisepala Yunck., and specimens of C. attenuata key to var. PRATHER ET AL., Taxonomic investigations of Cuscuta attenuata 449 Tas_e |. Comparison of morphological characters of Cuscwta species. Calyx Bracts Inflorescences Subsection Lepidanche C. compacta polysepalous, 1-10 along compact ovate, obtus pedicel C. cuspidata polysepalous, 1 at base and open ovate, cuspidate 1-2 along pedicel Subsection Indecorae C. attenuata gamosepalous, 1 at base of pedicel |= somewhat compact lanceolate, attenuate C. indecora var. indecora gamosep: alous, 1 at base of pedicel — open var. longisepala ile gee 1 at base of pedicel — open lanceolate, 2 longisepala in Yuncker’s 1965 key. Waterfall’s description of C. attenuata (Waterfall 1971) and Yuncker’s description of C. indecora var. longisepala (Yuncker 1965) are similar for every character state. In 1965, Yuncker ex- amined one of the specimens identified in this study as C. attenuata (C.J. Eskew 1395, OKL). Recognizing the distinctiveness of the specimen, he made the following annotation: “C. sndecora Choisy. The long narrow calyx lobes would make it var. longisepala Yunck. However, the specimen looks teratological and may not be the variety but only an abnormal form.” Cuscuta indecora is widespread in North and South America and is highly variable throughout its range (Hunziker 1950; Yuncker 1965, Beliz 1986). The infraspecific classification is somewhat controversial and Yuncker (1920, 1932, 1965) accepted different varieties in each of his treatments. Cwscuta indecora var. longisepala occurs throughout much of the range of C. indecora, including South America where the variation is similar to that found in North America (Hunziker 1950). Beliz (1986) does not mention C. indecora var. longisepala or list it in synonymy. Variability over the range of C. zndecora is not discussed nor were representative specimens west of Arizona cited. The few chromosome numbers known for taxa of the two subsections are invariant. Cuscuta indecora and C. glomerata Choisy (subsect. Lepidanche) are both reported as n=15 (Freeman & Brooks 1988; Pinkava et al. 1974) while C. attenuata is reported as 2n=30 (Prather & Tyrl 1993). An investigation of the relationship of C. attenuata to other taxa of Cuscuta by means of analyses of morphological variation and a program of interspe- cific hybridization was undertaken in an attempt to clarify the taxonomic position and rank of the species. 450 Sipa 16(3) 1995 MATERIALS AND METHODS Analyses of morphological variation —To examine morphological variation within and between taxa, 186 herbarium specimens of Cuscuta from DUR, ECSC, LL, NLU, NOSU, NWOSU, OCLA, OKL, OKLA, SMU, TAES, TEX, TULS, UARK, and the herbarium at Cameron University in Lawton, Oklahoma were examined—10 of C. attenuata; 50 each of C. compacta, C. cuspidata, and C. indecora var. indecora; and 26 of C. indecora var. longisepala (Prather 1990). In a preliminary analysis, five flowers each were examined on several specimens of each taxon. It was determined that variation among flowers of the same plant was negligible, therefore one flower was used from each specimen, which was treated as an OTU. One individual per population was examined. The data recorded for each specimen are presented in Prather (1990). Because Cuscuta lacks roots and well-developed leaves, and because stem features could not be accurately scored from herbarium specimens, only floral characters were used. One flower, with its pedicel and bract(s), was removed from each speci- men. To minimize variation resulting from flower age, all flowers selected had dehiscing anthers, a stage which lasts only a short time (Prather & Tyrl 1993). Observations were made after softening each flower by boiling in water to facilitate examination of the inner floral parts (Yuncker 1920). Samples were examined with a dissecting microscope at a magnification of 30x to score characters. Measurements were recorded to the nearest 0.1 mm using an ocular micrometer. Many of the measurements were incorpo- rated in the analyses as ratios to minimize the effect of size which may be influenced by environmental factors. Forty-four characters were scored for each sample. The first 27 characters (Table 2) were qualitative and there- fore suitable for use in all analyses. Characters 28—44 (Table 3) were quali- tative characters and thus could only be used in the multivariate analyses. For characters 1 and 5—27 an unprotected LSD test was performed. Char- acters 2—4 were excluded because the values were invariant among indi- viduals of at least one taxon. The multivariate analyses comprised UPGMA clustering, a principal component analysis, and a discriminant analysis. Statistical Analysis System (SAS Institute, Inc. 1985) was employed to perform these analyses except the unprotected LSD tests which were per- formed using Statview (Abacus Concepts, Inc. 1992). The UNIVARIATE procedure of SAS was used to confirm that the assumption normality of the test was not violated. The UPGMA clustering analysis was performed using the AVERAGE option of the CLUSTER procedure on data that had been standardized by the STD option which changes the mean to zero and the standard deviation to one. The varimax rotation method was used in the principal component analysis. Because the other statistical methods PRATHER ET AL., Taxonomic investigations of Cuscuta attenuata 451 Tasie 2. Means, standard deviations, minima, maxima, and results of the unprotected LSD test of characters 1-27 ae (S.D.) Min. - Max. Significance* . Pedicel Length (mm) C. attenuata (+0.49) 0.6-2.2 LL C. 1. longisepala 2.6 (+1.3) 0.5-5.8 A,M,S C. i. indecora 2.4 (40.98) 0.8-4.9 A,M,S C. compacta 1.0 (+0.42) 0.3-1.8 LL,s C. cuspidata 1.6 (+1.0) 0.2—-4.0 I,L,M 2. Number of Bracts at Pedicel Base** C. attentata 1.0 (40.00) 1-] C. t. longisepala 1.0 (+0.00) 1-1 C. t. indecor 0.9 (+0.24) 0-1 C. compacta 1.1 (40.35) |—2 C. cuspidata 1.0 (+0.14) 0-1 3: a of Bracts Along Pedicel** C. attenuata 0.0 (+0.0 0-0 C. 7. longisepala 0.0 ole 0 C. t. indecora 0.0 (40.14) i C. compacta 2.8 (+1.3) 1-7 C. cuspidata 0.7 (+0.88) 0-3 4. Number of bracts at Pedicel Apex** C. attenuata a (+0.00) 0-0 C. 1. longisepala 0.0 (+0.00) 0 C. 1. indecora 0.0 (+0.00) 0-0 C. compacta 1.6 (40.50) ]|—2 C. cuspidata 1.2 (+0.82) Q-3 5. Bract Length/Calyx Length C. attentata 0.82 (40.22) 0.56—1.27 = C. 1. longisepala 0.78 (40.19) 0.45-1.11 1,M,S C. 1. indecora 0.91 (40.24) 0.50-1.63 L C. compacta 0.89 (+0.13) 0.59-1.15 L C. cuspidata 0.85 (40.14) 0.44-1.13 L 6. Brace Length/Bract ss C. attenuata 2.2 (40.53 —3.3 1,L',M,S C. 1. longisepala 1.8 (40.38) ‘ 3-2.6 All 1,M,S C. i. indecora 1.4 (+0.30) 1.0- = A,L,M,S C. compacta 0.72 (40.12) ).39-0.99 A,I,L,S C. cuspidata 1.2 (+0.23) cee 8 A,1,L,M 7. Calyx Length (mm) C. attenuata 2.6 (+0.52) 1.5-3.2 1,L,M,S C. 1. longisepala 2.1 (40.44) 1.6-2.9 55 C. i. indecora 1.4 (+0.27) 0.8—2.0 A,L,M,S C. compacta 2.1 (40.25) 1.7-2.6 18 C. cuspidata 1.7 (+0.17) 1.3-2.1 A,1,L,M 8. Calyx Tube Length/ — - oe ie C. attenuata 0.22 (+0.06) 2 ILLS C. 1. longisepala 0.33 (40.12) 0. ie 3- 153 A,I,M,S C. t. indecora 0.50 (+0.11) 0.31 ae Teo) A,L,M,S C. compacta 0.20 (40.09) 0.05—0.53 LL.S C. cuspidata 0.13 (+£0.04) 0.07—-0.20 A,I,L,M 452 Sipa 16) 1995 TABLE 2. (Continued) Mean (S.D.) - Max. Significance* o. oe rae alyx wie C. attenuata 2.7 (+0.45) 2-3.8 1,L'.M,S C. 7. longisepala 2.2 ae L1-4.0 A'I,M,S C. i. indecora 1.4 (+0.30) 0.96—2.3 A,L,M,S! C. compacta 1.1 (40.19) 0.20—1.6 ATS C. cuspidata 1.3 (+0.16) 0.95-1.7 A,I',L,M 10. Corolla Length/Calyx Length 2 (+0.16) C. attentata 1,L,M,S C. 7. longisepala 1.6 (+£0.33) 1.0-2.3 A,I,M,S C. 1. indecora 2.2 (+0.37) 1.5—3.3 A,L,M C. compacta 1.8 (40.23) 1.4-2.4 A,L,I,S C. cuspidata 2.2 (+0.33) 1.4-3.2 A,L,M LC oyolla L secre Corolla Length C. attenuata 0. er (+0.0 ).53-0.64 1M C. 4. longisepala 0.54 ee ees Al 'M,S C. i. indecora 0.57 (+0.04) 0.46—0.67 Lis: C. compacta 0.72 (+0.06) 0.59—0.86 A,I,L,S C. cuspidata 0.61 (+0.12) 0.46-1.4 I',L.M 12. Corolla Lobe Length/Corolla Lobe Width C. attenuata 3.3 (40.42) 2.8-4.3 1,M,S! C. 1. longisepala 3.1 (+£0.36) 2.4-3.8 1'.M,S 1. indecore 3.0 (40.29) 2.4-3.6 A,L'\M,S C. compacta 3.7 (+0.45) 2.9-5.0 ALJ C. cuspidata 3.8 (40.57) 2.0-4.8 ALLL 13. Beane Length (mm) C. attenuata 3.4 (+0.4 2. 84.3 IM! Ss! C. 1. longisepala 3.1 en 2. 4-3.8 M,S C. 1. indecora 3.0 (40.30) 2.43.6 A,M\S C. compacta 3.7 (40.45) 2.9-5.0 ALE C. cuspidata 3.8 (£0.57) 2.0-4.8 ALLL 14. Number of Fringes Per Corolla Appendage C. attenuata 26.8 (+3.3) 24-35 M . i, longisepala 24.7 (+6.6) 6-36 M ©. 2. indecora 26.6 (+5.8) 17-46 - compacta 13.2 (+2.1) 7-18 A,LL,S C. cuspidata 26.4 (+4,.8) 16-36 Ls eae of Corolla pppencas (mm) C. attenuata 1.8 (+0.43) 1.0—2.6 I'M C. 1. longrsepala 1.9 (+0 63) 1.2-4.6 I'M C. i. indecora 1.6 (+0.27) Ll=29 ALLIM,S C. compacta 2.3 (40.30) 1.5—3.4 1,L,8 C. cuspidata 2.1 (40.44) 1:.042:8 IM 16. os a alera - Corolla Tube C. attenuata 0.92 - C. 1. longisepala 1.1 are er I,M,S C. 1. indecore 0.94 (+0.12) 0.68—-1.3 L,M C. compacta 0.86 (+0.09) 0.74-1.1 LS! C. cuspidata 0.91 (+0.13) 0.37-1.5 LM! PRATHER ET AL., Taxonomic investigations of Cuscuta attenuata TABLE 2. (Continued) one Cu oa ANA Sacna one as) io Anan Shea a cuspidata attenuata i. longisepala 1. indecora compacta cuspidata athenuata i. longisepala , sndecora compacta cuspidata attenuata 1. longisepala 1 des ‘ord cus, ee attenuala 7. longisepala 1. indecora compacta cuspidata attenuata i. longisepala 1. indecora compacta cuspidata attenuata cuspidata 7. Proportion of Appendage Fused to Corolla 0.45 (+0.07) 0.42 (+0.09) 0.27-0.71 0.44 (+0.07) 0.31-0.65 0.46 (+0.09) 0.29-0.78 0.61 (+0.06) 0.49-0.77 18. nas a Width 1.6 (+0.28) l. 1.8 oo l i 1.8 (+0.26) 1.2-2.4 4.2 (+0.65) 2.3-5.6 3.8 (40.61) 2.7-5.6 19. Filament Length (mm) 0.8 (+0.2) 0.5-1.0 0.7 (+0.1) 0.4-0.9 0.6 (+0.2) 0.3-0.9 0.2 (+0.1) 0.1-0.4 0.7 (+0.2) 0.4-1.0 20. Anther Length (mm) ) 0.8 (+0.1) 0.7-1.0 0.7 (+0.2) 0.4-1.1 0.7 (+0.1) 0.5-1.0 0.4 (+0.1) 0.1-0.6 0.7 (+0.1) 0.4-0.9 . Filament ce aaa a 0. ne (+0. 0.7 1- ere 0.50-1, 4 0.80 (+0.27) 0.38-1.8 0.54 (+0.28) 0.20—2.0 0.96 (+0.24) 0.50-1.5 22 pes Length/ aan Aves h 1.3 (£0.18) 1.6 1.3 ee os 6 1.4 (+0.21) 0.92-1.8 1.2 (+0.26) 0.74-1.7 1.9 (+0.34) 1.1—2.7 23. Longer Style Length (mm) 1.4 (+0.45) 0.8 1.1 (+0.38) 0.5-1.8 1.1 (+0.30) 0.6-2.1 1.3 (+0.30) 0.6—2.0 2.3 (+0.63) 1.3-3.9 24. Longer Style rca Style Length 1.1 (+0.06) O- 1.1 (+0.07) ee 1.1 (+0.10) 1.0-1.4 1.3 (£0.19) 1.0-2.1 1.3 (+0.16) 1.0-1.8 454 Sipa 16(3) 1995 TABLE 2. (Continued) 25; ee ength/Stigma Width C. attenuata 0.73 (40.6 0.61—0.85 _ C. 7. longisepala 0.75 (40 0.57-0.91 S| C. 7. indecora 0.75 (40.08) 0.55—0.86 S C. compacta 0.75 (40.08) 0.56—0.94 S C. cuspidata 0.70 (+0.09) 0.46—0.89 I,L'.M 26. Ovary Length (mm) C. attenuata 1.4 (+£0.25 1.0-1 LL's C. 1. longisepala 1.2 (20.21) 0.8-1.6 A'\M,S C. 1. indecora .2 (+0.20) O.8—1.6 A,M,S C. compacta 1.4 ).17) L.O-1.8 ILS C. cuspidata 0.8 ( er 0.5-1.2 A,I,L,M . Ovary cuaadi! hae C. attenuata 1.0 (+0.] 2 1,M!\S C. 1. longisepala 0.97 (+0. 5) eee | I,M,S C. t. indecora 0.54 (+0.12) 0.54-1.1 A,L,M C. compacta 1.1 (40.10) 0.80-1.2 A'L,LS C. cuspidata O.81 (40.13) 0.60—1.3 A,L,M *No — ant difference between taxa except those listed under significance and the taxa labelling the row; p<0.01 unless varie indicated, / .. attenuata, | = C. indecora var. indecora, G: 's = C. cuspidata. *Not tested because he values for some taxa were invariant. I S = a S x Q = sith 7 Var. es hoe only at p<0.05 had established that C. compacta and C. cuspidata were easily distinguished from C. attenuata, only C. attenuata and the two varieties of C. indecora were examined in the discriminant analysis. Prior probability of the discrimi- nant analysis was set proportional to the number of specimens of each taxon used in the analysis. Interspecific hybridizations. —Parasitized host plants of all taxa were trans- ported co the plant growth facility ac OSU and maintained as described previously (Prather & Tyrl 1993). Attempts to maintain C. compacta in the laboratory were unsuccessful because transplanting its woody hosts was not possible and cuttings did not survive under laboratory conditions. Pol- len of C. compacta, therefore, was collected from five individuals in the field and used immediately in crosses in the laboratory. Vouchers of all popula- tions used were deposited in OKLA. Individual flowers of C. attenuata were emasculated before anther de cence and mature pollen from individuals of another taxon was manually transferred to the stigmas (Radford et al. 1974). Reciprocal crosses with all taxa, except C. compacta, were performed in the same manner. Twenty crosses were performed between C. attenuata and C. compacta, 9 between C. attenuata and C. cuspidata, 16 between C. attenuata and C. indecora var. indecora, anc 25 between C. attenuata and C. indecora var. longisepala. — nis- en PRATHER ET AL., Taxonomic investigations of Cuscuta attenuata 455 TabLe 3. Qualitative morphological characters (28—44) of Cuscuta which were used in the multivari- ate analyses No. Character No, Character 28. Bract orientation 37. Presence of calyx papillations 29. Calyx orientation 38. Presence of calyx laticifers 30. Corolla orientation 39. Shape of corolla ‘ges 31. Shape of bract margin 40. Shape of corolla apex 32 Shape of bract apex 41. Presence of corolla aoe 33. Presence of bract laticifers 42. Inflexing of corolla lobe t 34. Overlap of calyx lobes 43. Orientatin of styles 35. Shape of calyx margin 44, Presence of stylopodium 36. Shape of calyx apex RESULTS AND DISCUSSION On the basis of univariate and multivariate analyses, C. attenuata 1s a morphologically distinctive taxon albeit similar to C. zndecora, particularly var. longisepala. Means, standard deviations, and minimum and maximum values for all characters are given in Table 2, as well as LSD test results between all taxa for those characters tested. As revealed by the univariate analysis, the means of four characters are significantly different (p<0.01) between C. attenuata and each of the other taxa (Table 2, Fig. 1). The mean of C. attenuata differs significantly (p<0.05) from that of C. compacta, C. cuspidata, and C. indecora var. indecora, for over half of the characters (15, 16, and 15, respectively); and from C. éndecora var. longisepala for one-third of the characters. Multivariate analyses revealed that C. attenuata is distinct from C. compacta and C. cuspidata but morphologically similar to C. indecora. In the UPGMA analysis, C. compacta and C. cuspidata each formed distinct groupings. How- ever, C. attenuata, C. indecora var. indecora and C. indecora var. longisepala did not form distinct groupings, but rather one large cluster. In the principal component analysis, the first three components explained 56.4% of the variation (Fig. 2). The remaining variation was accounted for by the other factors in 1-4% increments. The first principal component, which accounted for 31.1% of the variation, was weighted for characters 28, 29, 34, and 41. The second principal component, which accounted for 18.2% of th je variation, was weighted for characters 14, 21, 31, 32, 39, and 40. The third principal component, which accounted for 7.1% ofl he varia- tion, was weighted for characters 5, 7, 9, 10, and 37. As in the UPGMA analysis, C. compacta and C. cuspidata formed distinct clusters and the three remaining taxa did not. The third principal component was weighted for characters which dealt with variation in the calyces. For those characters, — C. indecora var. longisepala was always intermediate between C. indecora var. indecora and C. attenuata. 456 Sipa 16(3) 1995 —_~seeerofeacs—__ attenuata ae —— longisepala ee a —_—_ indecora | fe __ oh cuspidata ap of compacta a mew i T : T bs T y 225). «05.1325 ‘1. 15. 25 ‘2. 75°3.25 -75 1.25 1.75 2.25 2.75 3.25 Bract Length/Bract Width Calyx Length ornare: attenuata oe _ longisepala ee “anaes ——§ indecora a a cuspidata — a —aj— compacta fe ———T | a cae F a iia 5, 2 3.5 5S 1 15 2 25 3 Calyx Length/Calyx Width Corolla Length/Calyx Length Fic. 1. Univariate analysis of morphological characters in Cuscuta taxa. Means (vertical lines), standard deviations (broad horizontal lines), and minima and maxima (narrow hori- zontal lines) of the four characters for which the means of C. attenuata differ significantly (p<0.01) from each of the other taxa. Names in the center are epithets except indecora (=C. imdecora var. indecora) and longisepala (=C. indecora var. longisepala), and apply to the figures on each side. The discriminant analysis yielded probabilities of >.93 that each speci- men was appropriately designated, and thus demonstrated that C. attenuata and the two ae of C. indecora, can be distinguishec Interspecific hybridizations. —None of the crosses Deegan C. attenuata and any of the other taxa, erent both varieties of C. sndecora, produced fruits or seeds. In a related study, populations of C. attenuata were shown to be interfertile using these same methods (Prather & Tyrl 1993), and in some cases even the same individuals. These intraspecific crosses produced 81— 92% fruit set and 38—45% seed set. The lack of fruit and seed set from all interspecific crosses, but high rate of success of intraspecific crosses, sug- gests that C. attenuata is reproductively isolated from the other taxa. Taxonomic implications—Although once thought to be related to C. compacta and C. cuspidata, C. attenuata is definitely distinct from these spe- cies and should be positioned in subsect. Indecorae on the basis of its fused sepals rather than in subsect. Lepidanche as proposed by Waterfall (1971). On the basis of the systematic data generated in this study, it is concluded that C. attenuata is a distinct species albeit morphologically similar to C. indecora. In the absence of reproductive isolation we might treat C. attenuata as a variety of C. indecora. PRATHER ET AL., Taxonomic investigations of Cuscuta attenuata 457 - ° 4A ° 4 Aa ° ° = cae Pa Ds a, Ba oo. 6° 3 o ° A aakhe® 4 ° & 4A A 1 22... Ss °o A A 4, 9 00 ° A A a 3 - - R ane ° A = ° co) 8 ef ° & 0 eS oO a A=] Vv By -1 2 re 0 1 2 Second Component Fic. 2. Principal components analysis of morphological characters in Cuscwta taxa. First component plotted against the second component. The addition of C. attenuata to subsect. Indecorae brings its total number of species to five: C. attenuata parasitizing Iva annua in Kansas, Oklahoma, and Texas; C. coryli Engelm. parasitizing a wide variety of hosts in the central and eastern U.S.; C. stenolepis Engelm. parasitizing unknown hosts in Ecuador; C. warneri Yunck. parasitizing Phyla cuneifolia in Utah and Arizona; and C. indecora with two recognized varieties: var. indecora occur- ring on a wide variety of herbaceous hosts and widespread in North and South America and the West Indies, and var. dongisepala parasitizing a wide variety of herbaceous hosts in the southwestern U.S., Mexico, and South America (Yuncker 1965). Interestingly, C. warneri is also listed as a Cat- egory 2 species and is thought to be extinct (1993 FR 58:51159). ACKNOWLEDGMENTS The authors thank P. Buck, the late J.K. McPherson, and two anony- mous reviewers for their helpful comments. Appreciation is also extened to the curators who kindly loaned specimens of Cascuta for study. This re- search was supported by the U.S. Fish & Wildlife Service (Project Number 201811-89-00420). 458 Sipa 16(3) 1995 REPERENCES ABACUS CONC epTS. 1992. Statview. Abacus Concepts, Inc., Berkeley, Beuiz, T.D.C. 1988. A revision of Cuscuta sect. Cleistogrammica using ae and cladistic analyses with a comparison of reproductive mechanisms and host preferences in species from California, Mexico, and Central America. University Microfilms International, Freeman, C.C. and R.E. Brooks. 1988. Documented plant chromosome numbers 1988: Chromosome counts for North American plants—I. Sida 13:241—250. HUNZIKER, . L950. Las especies de Cuscuta (C renee de Argentina y Uruguay. gb yior ve Bot. Univ. Nac. Cordoba (Argentina) 1:1 PinKaAva, D.J., R.K. Brown, : H. Linpsay, and L.A. Mc ae 1974. IOPB chromosome number report XLIV. Taxon 23:373—380. _ HER, L.A. 1990. The ee of Cuscuta attenuata Waterfall. Unpubl. Master’s Thesis, i. St. Univ., Stillwater, OK ee L.A. and R.J. Tyre. 1993. oT he biology of Cuscuta attenuata Waterfall (Cuscutaceae). Okla. Acad. Sci. 73:7—13 Rabrorb, A.E., W.C. Dickson, J. R. Massry, and C.R. Bei. 1974. Vascular plant system- tics. Harper & Row, NY. oe Institute INc. 1985. SAS user's guide: Statistics, Sth ed. SAS Institute Inc. Cary, NC. Taytor, R.J. and C.E. Taytor. 1980. Cuscata attenuata: Status report, Submitted to U.S. Deparment : Interior, Fish & Wildlife Service, Albuquerque, NM. sig Seas on Sam , P.G. Risser, and J.J. Crockerr. 1978. Field surveys and status evalwation on ee and candidate Sue Let species in Oklahoma, project report to a epartment of Interior, Fish & Wildlife Service, Albuquerque, NM. WATERFALL, 1971 . New species of oe and from Oklahoma. Rhodora 73:575— YUNCKER, T.G. 1920, Revision ye the North American and West Indian species of Cuscuta. Illinois Biol. fee 6:1-14 YuNCKER, T.G. 1932. The genus mn Bull. Torrey Bot. ee 18:113-331. oa ER, LG. i Cuscuta. North American Flora. Ser. [f.4:1—5] A NEW SPECIES OF DIRCA (THYMELAEACEAE) FROM THE SIERRA OF NORTHEASTERN MEXICO GUY L. NESOM and MARK H. MAYFIELD Department of Botany University of Texas Austin, TX 78713, U.S.A. ABSTRACT A newly discovered population of Dirca from the Sierra Madre Oriental of Tamaulipas, Mexico, is described and illustrated here as D. mexicana sp. nov. It expands the genus to three species and it is first of the genus known to occur in Mexico. The new species is similar to D. palustris of the eastern United States in its long calyx tube with abruptly ing limb, but its greatest overall similarity is with the rare and restricted D. occidentalis the San Francisco Bay region in west-central California. Dirca mexicana resembles the ee in its bud scales with whitish pubescence, sessile flowers and fruits, distinctly lobed calyces, and persistent pubescence on the young twigs and abaxial leaf surfaces. The new species differs from D. occidentalis in its Howers with a longer, narrower tube, shorter limb anc i lobes, and shorter style and filaments; the two population systems are separated by more than 2500 kilometers. Dirca mexicana is separated from the closest population of D. palustris by more than 1100 kilometers. Key Worps: Dirca, Thymelaeaceae, Mexico RESUMEN Una poblacién de Dirca recientemente descubierta en la Sierra Madre Oriental de Tamaulipas, México, se describe e iconografia aqui como D, mexicana sp. nov. Con e lo se eleva a tres el ntimero de especies en el género, siendo ésta la primera que se conoce en éxico. La nueva especie se parece a D. palustris, del este de los Estados Unidos, por su largo tuba del caliz con limbo abruptamente acampanado, pero su mayor parecido general es con la rara y restringida D. occidentalis de la regién de la Bahia de San Francisco en la region centro-oeste de California. Dirca mexicana se parece a esta ultima en las escamas de las yemas con pubescencia blanquecina, flores y frutos sésiles, calices netamente lobulados y la pubescencia persistente en las ramas jévenes y en el envés de las hojas. La nueva especie difiere de D. occidentalis por sus flores con un tubo més largo y mas estrecho, con limbo y lébulos mas cortos, y estilo y filamentos mds cortos. Los dos sistemas de poblaciones estan separados por mas de 2500 kil6metros. ta alejada mas de 1100 kil6metros e la poblacion de D. palustris mas proxima. The genus Dirca L. has previously included only two species, D. palustris L., scattered but widespread in the eastern United States and adjacent Canada, and D. occidentalis A. Gray, endemic to six counties of che San Francisco Bay region in west-central California (McMinn & Forderhase 1935; Vogelman 1953; Nevling 1962; various floristic manuals; Fig. 1). On a Sina 16(3): 459 — 467. 1995 460 Stipa 16(3) 1995 occidentaii ENA LK a O tas POS o\ aN Nao mexicana . a ° | pegs = | £ So ipa Sui | Fic. 1. Geographic distribution of the three species of Dirca. botanical reconnaissance in September 1994 in the Sierra Madre Oriental of Tamaulipas, Mexico, we observed a population of low shrubs that we recognized as Direa, but the plants were sterile and not identifiable to spe- cies. In early March 1995 we were able to study the population at peak flowering and beginning of fruit. The Mexican plants most closely resemble D. occidentalis, but there are several consistent morphological differences between them and they are geographically disjunct by more than 2500 kilometers (California/Tamaulipas, see Fig. 1). The closest known popula- tions of Dirca palustris, at the southwestern corner of its range in eastern Louisiana, Arkansas, and southeastern Oklahoma, are more than 1100 ki- lometers disjunct from the Mexican plants. We believe that it is justifiable and desirable to recognize this Mexican population as a species distinct from both of its more northern relatives. Dirca mexicana Nesom & Mayfield, sp. nov. (Fig. 2) Differt a Dircae oecidentali A. Gray calycibus tubo longiore angustiore limbo breviore, filamentis supra medium calycis insertis, et stylis ac filamentis brevioribus. Woody shrubs 6—20 dm tall, averaging ca. 16 dm, from a single trunk branched near the base, with spreading, flexuous branches, the stems 2— 3(—5) cm wide near the base of the plant, the ultimate branches 2—3 mm wide, bark smooth and grayish to reddish-brown, the current year’s growth NeESOM AND MayeleLp, A new species of Dirca from Mexico 461 SN “\ _ Branches and details of Dirca mexicana. A and B. Sterile branch and leaf era 2085). C-E. Flowering branch, flower cluster, and single flower, opened (Nesom 7863). 462 Sipa 16(3) 1995 persistently loosely and sparsely pubescent, glabrous below. Leaves deciduous, alternate, entire, at maturity broadly elliptic to slightly ovate, basally rounded, 4-8 cm long, 2.5-6.0 mm wide, 1.3—1.7 times longer than wide, the largest distally situated on the branchlets, glabrous above, the lower surface persistently sparsely strigose-sericeous on the lamina and along the veins, petioles 1-2 mm long. Buds covered by the enlarged peti- ole base, mixed (flowers and leaves), the apical apparently falsely terminal, the 4 bud scales whitish-sericeous, forming a foliaceous, deciduous involu- cre to the flowers. Flowers appearing before or concurrently with the leaves, sessile in axillary and apical fascicles, deflexed to somewhat more nodding at full anthesis, the axillary fascicles invariably producing 3 flowers, the apical fascicles sometimes ap panenely twinned” with (S—)6 flowers; corolla absent, not represented by infracalycular structures; calyx (7—)8—L10 mm long, petaloid, the tube narrow but slightly widening distally, 5-7 mm yi long, abruptly expanded into a flaring limb 2—4 mm long, the tube/limb ratio (of length) (1.5—)1.7—3.0, the 4 { calyx lobes 1.5—3.0 mm long, the lobes and limb yellow, the nubél lighter yellow and drying creamy; stamens 8, filaments inserted within the calyx tube essentially at a single level (at the throat, 2.0—3.5 mm below the lobe apex), the longest exserted 2-3 mm above the calyx lobes, those inserted below the lobes |—2 mm longer than those inserted below the sinuses, the thecae 0.4—0.8 mm long, basifixed; “hypogynous disc” or “disc” (sensu Heinig L951) a ring of connate, irregu- lar, slightly fleshy scales ca. 0.2-0.4 mm high and wide, basally adnate to the inner surface of the calyx base; stigma minutely capitate, above the level of the pre-dehiscent anthers on a fully elongated style, equalled or slightly surpassed by the dehiscing anthers. Fruits drupaceous, |-seeded, pyriform to ovoid, sessile, green when young, mature fruits not observed. Type: MEXICO. Tamautipas. Municipio. Hidalgo: along mountainous road from Sta. Engracia (Tamaulipas) to Dulces Nombres (Nuevo Leon), Arroyo Obscuro, 2.0 road mi NE of Los Caballos toward Canada E] Mimbre, 15.0 road mi W of a crossing of arroyo El Mimbre, 23°59'09"N, 99°28'37"W/, ca. L800 m, 3 Mar 199 5, Guy Nesom 7863 with Mark Mayfield and Greg Anderson (HOLOTYPE: ve isotypes: AAU, ANSM, ARIZ, ASU, BH, BRIT, CAS, CHAP, COLO, CONN, DAV, DUKE, ENCB, FE, FSU, FTG, GH, GUADA, IEB, K, KANU, LSU, M, MICH, MO, MSC, NCU, N I NLU, NY, OBI, OKL. OS, OSC, P, RM, RSA, S, TENN, TEX, UARK, UAT, UC, UCR, UNL, US, VDB, WIS, WTU, XAL Additional collection examined: MEXICO. Tamautipas: type locality, [sterile, leaves only}, 23 Sep 1994, Mayfield 2085 with Nesom (TEX). ee Description of the locality.—The site at which Dirca mexicana was observed and collected (Arroyo Obscuro) is a steeply sloping, north-facing, mesic cove surrounding a rocky watercourse. The area is limestone with karstic tendencies. The elevation near the road is ca. 1800 meters, but plants of Nesom AND MayrigLp, A new species of Dirca from Mexico 463 Dirca extend to at least ca. 20 meters (elevation) above the road and to at least 80 meters below it. They may well occur over a broader area than we investigated, particularly downslope, but this is the only locality where Dirca was encountered along the 35 kilometer road from Sta. Engracia to Dulces Nombres. Indeed, we found no other site along this road with a similar physical and floristic definition. The dominant canopy trees at the site are Carya ovata (P. Miller) K. Koch, Pinus patula Schlecht. & Cham., Pseadotsuga menziesii (Mirb.) Franco, and Quercus laurina H. & B. Large trees of these reach about 20-25 meters in height, with call Pseadotsuga perhaps reaching 30 meters. Judging from stumps in the area, particularly large individuals (presumably of Psewdotsuga and Pinus) have been removed by logging. Large trees of Carpinus caroliniana Walt. and Liguidambar styraciflua L. are scattered at the locality but do not reach the full height of the canopy. The woody understory includes the following: Cornus urbiniana Rose, Croton virletianus Muell.-Arg., Garrya macrophylla Benth., Ilex aff. rubra S. Wats., Litsea pringlet Bartlett, Persea pododaenia Blake, Philadelphus calcicolus Hu, and Taxus globosa Schlecht. Most of the canopy species are deciduous and were just beginning to break bud at this site; the understory includes a predominance of evergreen species. Arboreal bromeliads, including three species of Ti//andsia, are conspicu- ous. Ferns form a major part of the herbaceous flora at this site; among the most common are species of Adiantum, Woodwardia, Botrychium, Phanerophlebia, Polypodium, and Polystichum. Other common herbs include species a Chimaphila, Chiropetalum, Goodyera, Sisyrinchium, and Stachys. Characteristics of the population and biological observations, —We observed 800-1000 plants (by estimate) at the Dirca site, where they occur for about 300 meters along the road and adjacent slopes on the east side of the water- course. They tend to be densely clustered in more open-canopy microsites, occurring most abundantly toward bottom of cove, but the plants occur — singly and more scattered over a larger area. We did not unearth any plants, but there was no indication that they reproduce clonally. A report of rhi- zome production in D. occidentalis (McMinn & Forderhase 1935) has not been corroborated by more recent observations (Spongberg pers. comm.). The entire population of Dirca mexicana is strongly synchronous in flowering. Had we arrived 3 or 4 days later, it is likely that we would have been unable to observe floral features, as the flowers appear to wither quickly with the onset of fruit maturation. The flowers appear to be weakly protogynous and there is evidence that they also may be self-compatible Initially, before full development of the calyx, the style (with apical stigma) usually is elongated past the level of the anthers. At this time, the stigma appears slightly moist and presumably receptive; we also observed that 464 Sipa 16(3) 1995 many at this stage appear to have pollen atrached. At full anthesis, the filaments have raised the open anthers to the level of the stigma or slightly beyond it. Further, there is little time separating intial receptivity of the stigma and dehiscence of the anthers, and little distance between the stigma and open anthers. Visiting the flowers were one species of bee (individuals ca. 7 mm in length) and at least four species of butterflies. Pollination is effective, judg- ing from the apparently ubiquitous deposition of pollen on the stigmata of pre-staminate flowers. Bees and butterflies were probing inside the flowers, and although it seems highly likely that nectar was available to them (prob- ably produced by the hypogynous disc), its production was not evident in numerous flowers that we examined. Nor could we detect any fragrance (D. occidentalis was noted by Howell {1970} to be fragrant). Fruit matura- tion was beginning only on a few plants and we were unable to make an estimate of the success of fruit and seed production. We did not determine what feature or features account for the remark- able rarity of Dirca mexicana. All of the other species yet identified from Arroyo Obscuro are found in other sites in the same general area—none except the Dirca could be considered rare (but see comments below regard- ing Viburnum). Rarity, however, also is a feature of both other species of the genus, particularly D. occidentalis (Stebbins 1942; Johnson 1994). Dirca palustris is widespread but of uncommon occurrence. Morphological comparisons. —Dirca mexicana resembles D. occidentalis in most of the features that have been used to distinguish the latcer from D. palustris (Vogelmann 1953; see key below): vestiture, presence or absence of a pe- duncle and pedicel, and the distal morphology of the calyx. The flowers of D. mexicana, however, are more similar in general configuration to those of D. palustris: both have a relatively long and narrow tube with a shorter, abruptly widening limb, the staminal filaments are inserted above the middle of the calyx, and the style and anthers are exserted for a relatively short length. In D. occidentalis, the tube is shorter than the broadly funnel- form limb, the staminal filaments are inserted below the middle of the calyx, and the style and stamens are long-exserted. Vogelman (1953, p. 80) emphasized the taxonomic usefulness of the level of filament insertion, which in turn is indicative of the throat position and flower shape: “In most in- stances this character alone is sufficient to distinguish the two species.” — The comparative illustrations furnished by Vogelmann, however, do not accurately represent this difference, nor does the detailed illustration of D. palustris in Cronquist (1981, p. 635); those by Holm (1921) are more simi- lar to our own observations. Differences between the new species and D. occidentalis are summarized in the following key. NeEsoM AND MAYFIELD, A new species of Dirca from Mexico 465 1. Bud scales with brown or reddish-brown pubescence; young twigs and both leaf surfaces completely glabrous; flowers and fruits pedicellate, che whole cluster often pedunculate; calyx margin merely crenulate-undulate, with- out distinct lobes; eastern United States D. palustris . Bud scales with whitish pubescence; young twigs and abaxial leaf surfaces Sea pubescent; flowers and fruits sessile; calyx distinctly and dee ; San Francisco Bay region of California or northeastern MEXICO ou. ee ceeeee eee ee eee (2) 2 Cala tube 2-4 mm long, broadened into a broadly funnelform limb 4— 6 mm long; staminal filaments inserted below the middle of the calyx, ede exserted 3—4(—5) mm above the flower at maturity; style and stigma (1-)2—4 mm above the anthers at maturity; San Francisco Bay — region D. occidentalis . Calyx tube 5-7 mm long, abruptly broadened into a flaring limb 2-4 long; staminal filaments inserted above the middle of the calyx, an- NO thers exserted 2—3 mm above the flower at maturity; style and stigma about level with the anthers at maturity; sierra of northeastern Mexico D. mexicana Biogeographic pattern. —It does not seem possible at this point to provide a morphologically based hypothesis of relationship among the three spe- cies of Dirca, because the genus appears to be relatively isolated, its closest relatives (and thus the evolutionary polarity of character states) difficult to specify (see Domke 1934; Nevling 1959). Still, it is surprising to find a greater overall similarity between D. mexicana and D. occidentalis, in view of the well-known pattern of close relationship and disjunction between species of the eastern and southeastern United States and the sierra of north- eastern Mexico (Miranda & Sharp 1950; Graham 1973). In the immediate area of Dirca mexicana are numerous species that are disjunct from their primary range in the eastern United States (e.g.): Carpinus caroliniana, Carya ovata, Chimaphila umbellata (L.) W. Barton, Desmodinm glutinosum (Muhl. ex .) Wood, Liguidambar styraciflua, Pedicularis canadensis L., anc Polystichum acrostichoides (Michx.) Schott. In addition, there is a rare and yet undescribed species of Viburnum (Nesom in prep.) in the close vicinity that apparently is most closely related to V. obovatum Walt. of the southeastern United States. Another recently recognized, disjunct species from the same area of Tamaulipas, Scrophularia sp. nov. (Mayfield and Nesom submitted), is closely similar co S. marilandica L. of the eastern U.S. species and S. californica Cham. & Schlect. of the Pacific region. Taxus globosa, one of the understory species at the Dirca site, also has close relatives widely separated in the eastern and western United States. Biotic disjunctions between the eastern and western United States, similar to that in Dirca, are well known though not particularly common (e.g., Sharp 1951; Wood 1970), and other examples can easily be added (e.g., Sericocarpus, lonactis). This pattern is emphasized by the recent discovery of 1995 SS 466 Sipa 16(3 Nea: a second species of Neviusia in California (Shevock et al. 1992), both species extremely rare. In contrast, clearly established acne tions of extant plants from the western United States to the Sierra Madre Oriental of Mexico are unknown to us. Numerous disjunctions exist between the Sierra Madre Oriencal and $.M. Occidental, but most or all of these appear to be deriva- tives of the general pattern described by McVaugh (1952), where the evo- lutionary antecedents occupy a more southern position. A few Mexican species have a broken but still somewhat continuous distribution from the eastern sierra through northern Coahuila into the mountains of southwest- ern Texas and southeastern New Mexico (e.g., Populus tremuloides Michx., Pseudotsuga menziesti {Mirb.} Franco, and see Nesom 1993). In the broadest view of the origin of Dirca mexicana, the simplest hy- pothesis is that it belongs to the floristic element with its closest evolu- tionary ties to the fora remaining in the southeastern United States, the geographic continuity between them probably established during the middle to late Miocene (Graham 1973). The disjunction between D. palustris and D. occidentalis may be considerably older, as a vegetation probably includ- ing these species was spread across North America through much of the Tertiary, beginning as early as the Eocene (Graham 1972, 1993). In a con- trasting view, Axelrod (1975) hypothesized that the closely related dis- juncts found in eastern Mexico, the Appalachians, and the West coast of the United States represent remnants of a continuous forest earlier spread into Mexico and more simultaneously fragmented as a result of a spreading dry climate in the mid-Oligocene. Evaluation of the latter theory is diffi- cult at present because of the lack of evidence for northern temperate ele- ments in Latin American prior to the Late Tertiary. These elements have not been recovered from the Oligo-Miocene Simojovel Group of Chiapas, Mexico; Quercus first appears in Panama in the Mio-Pliocene, and in South America A/nus arrives by about one million years ago and Quercus not until about 340,000 years ago. Many eastern North America disjuncts were present in eastern Mexico by the middle Pliocene, and global paleotemperacure history suggests that cooling in the middle Miocene may have been an appropriate time for their principal introduction. ACKNOWLEDGMENTS We thank Greg Anderson for help with collecting, observations on floral biology of the new species and his comments on the manuscript. We are graceful co Billie Turner, Steve Spongberg, Alan Graham, and Beryl Simpson for comments on the manuscript, to John Taylor, Gene Wofford and Bob Kral, and Sidney McDaniel for information on Dirca in Oklahoma, Ten- nessee, and Mississippi, respectively, to Linda Vorobik for the illustration, and to Jaime Hinton an his encouragement and company on fall and spring Nesom AND MayrieLp, A new species of Dirca from Mexico 467 collecting trips and for the example of his remarkable intelligence and con- tinuing vigor. REFERENCES AxeELrob, D.J. 1975. Evolution and eae . the Madrean-Tethyan sclerophyll vegetation. Ann. Missouri Bot. Gard. 62:280— Cronauistr, A. 1981. An integrated system of eee of flowering plants. Columbia University Press, New York. Domke, W. 1934. Untersuchungen uber die systematische und geographische Ghlederung der Thymelaeaceen. Bibliot. Bot. 27(111):1-151 GranuaM, A. 1972. Outline of the origin and a recognition of floristic affinities etween Asia and eastern North America. In: A. Graham (ed.), Floristics and paleo- floristics of Asia and eastern North America. eee Sci. Publ. Co., Amsterdam. Pp. 1-18 _,s«1:973. History of the arborescent temperate element in the northern Latin American biota. In: A. Graham ed., Vegetation and Vegetation History of Northern Latin America. Elsevier Sci. Publ. Co., Amsterdam. Pp. 301-314 _ «993. History of the vegetation: Cretaceous (Maastrichtian) - Tertiary. In: Flora oe Neath America Editorial Committee, eds, 2+ vols. Oxford University Press, New York. Vol. 1, pp. 365-366 Heinic, K.H. 1951 “Studies in the foral morphology of the Thymelaeaceae. Amer. J. Bot. 38:113-132. Hou, T. 1921. Dirca palustris L. A morphological study. Amer. J. Sci. 2:177-182. How ane 1970. Marin Flora (ed. 2). University of California Press, aoe JOHNSON, B.G. 1994. Plight of the Western Leatherwood. Four Seasons 9:38—41. Mayeletp, M.H. and G.L. Nesom. Submitted. A new species of Scrophularia ies from northeastern Mexico. Britton McMinn, H.E. and B. ForbERHASE. oe Noces on Western Leatherwood, Dirca occidentalis —_ Gray. Madrofo 3:117—120. McVaucu, R. 1952. Suggested phylogeny of Prius serotina and other wide-ranging phylads in North America. Brittonia 7:317—346. Miranpa, F. and A.J. SHarp. 1950. Characteristics of the vegetation in certain temperate regions of eastern Mexico. Ecology 31:313—333 Nesom, G.L. 1993. Three species of Aster (Asteraceae: Astereae) disjunct in northern 4. Coahuila, Mexico. Phytologia 74:296—30 Nevuine, L.I. 1959. A revision of the genus Daphnopsis. Ann. Missour Bot. Garden 46:257—358. celles aay Oe. he Thymelaeaceae in the southeastern United States. J. Arnold Arbor. 43:428—434. saul A.J. 1951. Relationships a the floras of California and southeastern United es. Contr. Dudley Herb. 4:59-61. ae a B. Ertrer, and Be Tayior. 1992. Nevinsia cliftonit (Rosaceae: Kerrieae), an intriguing new relict species from California. Novon 2:285-289. Sreppins, G.L. 1942. The genetic approach to problems of rare and endemic species. Madrono 6:241-258. VoGELMAN, H. hae 3. - ae ison of Dirca palustris and Dirca occidentalis (Thymeleaceae). Asa Gray Bull. Woop, Cx. L970. a floristic relationships between the southern Appalachians and western North America. In: P.C. Holt (ed.), The distributional history of the biota of the Southern Appalachians. Part II. Flora. Virginia Polytechnic Inst. State Univ. Res. Div. Monogr. 2:331-404. 468 Sipa 16(3) 1995 BOOK NOTICE FautIn, DAPHNE GAIL, DouGLas J. FuruyMa, and FRANCES C, JAMES. 1994 Annual Review of Ecology and Systematics, Volume 25. (ISSN 006-4162, hbk.). Annual Reviews, Inc., 4139 El Camino Way, Palo Alto, CA 94303-0139. $47.00 US, $52.00 elsewhere. 686 pp. “Chapters in this volume directly relevant to application of ecological and systematic knowledge are those by Schulze et tL , who apply new technology to some pressing global issues, Chambers and sears who compare natural and managed systems, and Leigh Brown and Holmes on HI t and Lawton and Kuris and Lafferty, among others, ex- plicitly relate principles _ ee to human concerns. Linkages between and among chapters in this volume abound. Concern with evolution, the interface between ecology and systematics, is implicit in many chapters; it is explicit in those by Moran on complex life cycles, Moreno on genetic architecture, Raff et al. on metazoan phylogeny, and Vrijenhoek on unisexual fishes, ve in thac by Vermeij entitled “The Evolutionary Interac- tion among Species: Selection, Escalation, and Coevolution.” “Three chapters concern sys- tematics of particular groups, two of them plants—legumes by Doyle, and Mesembryanthemaceae by Ihlenfeldt.” There are 25 chapters: 1) Algal eee Limita- tion and the Nutrition of Aquatic Herbivores; 2) Genetic Architecture, Genetic Behavior, and Character Evolution; 3) Molecular Approaches to Population Biology; 4) Unisexual Fish: Model Systems for Studying Ecology and Evolution; 5) Cooperation and Conflict in the Evolution of Signal Interactions, 6) Evolutionary Biology of Human Immunodefi- ciency Virus; 7) ii ane Dynamics and Genetics; 8) Community Structure: Larval ae in Snail Hosts; 9) The Evolutionary Interaction Among Species: Selectior Escalation, and C Coevolution: 10) The Systematics of Coral Genus Acropora: Implications of w Biological padinne for Species Concepts; 11) A Day j in the Life of Seed: Movements a Fates of Seeds and Their Implications For Natural and Managed Systems; 12) The Evolution of Vocalization in Frogs and Toads; 13) Phylogeny of the Legume Family: An Approach to Understanding the Origins of Nodulation; 14) Using DNA Sequences to Unravel the Cambrian Radiation of the Animal Phyla; 15) ee and Ecology of Whales and Dolphins; 16) Fisheries Ecology in the Context of Ecological and Evolutionary Theory; 17) Ecology and Evolution of Reproduction in Milkweeds; 18) The Nature and Conse- quences of Indirect Effects in Ecological Communities; 19) Integrative Approaches to opr Ecology: Anolis Lizards as Model Systems; 20) The Ecological Consequences of Shared Natural Enemies; 2 iversification in an Arid World: The tee nee 22) Genetic Divergence, Reproductive Isolation, and Marine Speciation, 23) Adaptation and Constraint in Complex Life Cycles of Animals; 24) Male Parental Behavior in Birds; 25) Relationships Among Maximum Stomatal Conductance, Ecosystem Surface Conductance, Carbon Assimilation Rate, and Plant Nitrogen Nutri- tion: A Global Ecology Scaling Exercise. A subject index is included MISCELLANEOUS NOTES ON HAPLOPHYTON APOCYNACEAE: PLUMERIEAE: HAPLOPHYTINAE) JUSTIN KIRK WILLIAMS Department of Botany University of Texas Agtiy, LX F371 3, US as ABSTRACT The genus oa ton has typically been regarded as monotypic by various authors. A close examination of herbarium specimens coupled with field observations in Chiapas, Mexico, has rane new evidence that strengthens the recognition of two distinct spe- cies, H. cimicidum and H. crooksii. Further evidence suggests that Haplophyton should be removed from the subcribe Alstoniinae and placed back into its own subtribe, the Haplophytinae. Distribution maps and a key to the species are provided. Key Worps: Haplophyton, Alstonia, Haplophytinae, Alstoniinae, Apocynaceae RESUMEN El género Haplophyton ha sido tratado como monotipico por varios autores. Un cuidadoso examen de especimenes de herbario junto con observaciones de campo en Chiapas, México, ha puesto de manifiesto nuevas evidencias que etueieey el reconocimiento de dos especies distintas: H. cimicidum y H. crooksii. Otras | ue Haplophyton deberfa ser sacado de la subtribu Alstoniinae para valveils a eoloean: en su propia subtribu Haplophycinae. Se ofrecen mapas de distribucién y una clave de especies. De Candolle (1844) described Haplophyton as a monotypic genus based upon H. cimicidum. Benson (1942) added a variety to the species, var. cro0ks77, later elevating it to the rank of species (Benson 1943) when material of mature seeds was examined. Benson (1954) regarded H. crooksi7 as a dis- tinct taxon because of its smaller leaves and smaller discontinuously ridged seeds. Various authors Johnston 1990; Kartesz 1994; Leeuwenberg 1994; Pichon 1950), however, disregarded these differences and recognized only H. cimicidum. My examination of the specimens housed at LL, TEX coupled with field observations in Chiapas, Mexico, however, has uncovered new evidence that strengthens the recognition of two distinct species. The genus Hap/ophyton is placed in the subfamily Plumerioideae because of its undifferentiated free anthers and indole alkaloides. These characters, in particular the indole alkaloides, are regarded as indicative of a mono- phyletic lineage (Leeuwenberg 1994). In addition, sinistrorse aestivation of the corolla (overlapping of the petals in bud to the left) distinguishes the on Stpa 16(3); 469 — 475. 1995 470 Sipa 16(3) 1995 A A A 30 A A 30 A A A A P Aa Aa A 110 100 0 300 mi a eer Fic. 1. Documented distribution of Haplophyton crooksii. Plumerioideae from the Apocynoideae, which has dextrorse aestivation (over- lapping of the petals in bud to the right) (Leeuwenberg 1994). The aestivation of Haplophyton, however, has been under considerable question. De Candolle (1844) described Hapl/ophyton as sinistrorsely contorted, while Bentham and Hooker (1873) listed the taxon as having a dextrorse aestivation. Pichon (1950) also pointed out that Haplophyton had a “curi- ous” dextrorse aestivation. Woodson (1938) on the other hand reported a sinistrorse aestivation for Haplophyton, as did Standley and Williams (1969), and Correll and Johnston (1970). Benson (1942, 1943, 1954) did not de- scribe the aestivation. An examination of herbarium specimens, however, reveals that the above reports are all correct. The southwestern United States and northern Mexican (Fig. 1) members of Haplophyton have an aestivation to the left, while the Guatemalan and southern Mexican (Fig. 2) members have an aestivation to the right. The correlation of this character with the allopatric distribution of the two populations provides strong evidence that Haplophyton is comprised of two distince species. My field studies in El Chorreadero, Chiapas (W7//iams & Plum, 95-34) confirm that southern popu- lations have a dextrorse aestivation. When de Candolle (1844) described Haplophyton cimicidum he reported a sinistrorse aestivation, and listed as the type a specimen from Tehuantepec, Oaxaca. Topotypes from this region, however, have a dextrorse aestivation. on Witiiams, Notes on Haplophyton 471 Fic. 2. Documented distribution of Haplophyton cimicidum. Because I have not had an opportunity to examine the type, its aestivation remains at present undetermined. I suspect, however, that it 1s to che right and that the reports of a sinistrorse aestivation are either from specimens examined from the north or are the result of the presupposition that all Plumerioideae flowers have aestivation to the left. Differences in the corolla tube and calyx dimensions have not until now been discussed as segregating factors between the two species, but my ex- amination of herbarium specimens shows that H. c/micidum possesses a longer corolla tube than H. crooksi/, as well as smaller calyx lobes. An examination of the seeds, in particular their size and surface structures, also supports Benson's (1943) assertions that the two species are distinct. A character that seems to tentatively hold true is the color of the coma which is typi- cally golden yellow in H. cémicidum and charcoal grey in H. crooksi. The populations in Guaymas, southwestern Sonora, superficially resemble H. cimicidum in leaf size; accordingly Benson (1943, 1954) recognized them as such. A close inspection of these individuals, however, shows that their aestivation is to the left and that the seeds are 6-8 mm long, placing them with H. crooksiz. With its sinistrorse aestivation, larger calyx lobes, smaller corolla tube, smaller seeds, and isolated geographical range, it seems evident that 472 Sipa 16(3) 1995 Haplophyton crooksi constitutes a distinct element from H. cimicidum and, consequently, necessitates the recognition of two species for Haplophyton. KEY TO SPECIES OF HAPLOPHYTON 1. Petals of the bud overlapping to the right; leaves 22-65 mm long, 11-26 mm wide; corolla tube 8-10 mm long; sepals 3-5 mm long, less than half the length of corolla tube; seeds 8-11 mm long, el ridges of seeds continuous; coma yellowish; Guatemala and southern Mexico ............. L. H. cimicidum . Petals of the bud overlapping co the left; leaves 11—35 (50) mm long, 3-12 mm wide; corolla tube 6-8 mm long; sepals 4-11 mm long, longer than or aay ~~) equal to half the length of corolla tube; seeds 6-8 mm long, vertical ridges of seeds broken by transverse grooves; coma white to charcoal grey; south western United States and northern Mexico 2. H. crooksii |. Haplophyton cimicidum A. DC., Prodr. 8:412. 1844. Typr: MEXICO. Oaxaca: Tehuantepec, 1833, Andrienx 250 (HOLOTYPE: G-DC). eae co specimens. GUATEMALA. HUEHUETENANGO: between the villages of Nenton and Llano Grande (15°48'N; 91°45°W), growing in dry subtropical forest, 16 Nov | sie Castillo 1638 (B). MEXICO. Cuiapas: steep wooded slopes 9 km N of Tuxtla Gutiérrez along road to E oo. 27 Oct 1965, Breedlove 13850 (TEX); steep canyon, ie ses deciduous forest, 15 km SW of Suchiapa, eae road to Villa Flores, 750 m, 26 Sep 1972, Breedlove 28008 (TEX), ELC acdc ro, 200 yds. da entrance to waterfall park, along road, growing in the open, below rocky ee with Capraria frutescens, 7 Jan 1995, Williams & Plum 95- 34 (TEX). Morexos: in limestone soil on side of the ruins of Xochicalco, 16 Aug 1947, Rodrigo 710B (TEX). MicHoacan: in El Carrizo, 30 km SW of Ti as epec, 970 m, 25 Aug 1980, Nijez 2496 (TEX). Oaxaca: tropical deciduous forest on SE slopes of mera Madre del Sur, 22 km W of Tequisisclan on Pan-Am hwy to oo L000 m, (16°2 95°45'W), 4 Jul 1969, Marcks 993 (TEX); 55 mi SE of oe ons road to Tehuz saben in mountains 9 mi NW of La Junta, 13 Sep 1971, Clarke 20463-1 (TEX); 52 km § of pears along the road to Oaxaca, 31 Jul 1985, Salinas F-2671 (TEX); open areas 9— 10 km E of La Ventosa along Pan-Am hwy (rt 190), 50 m, 16 ne L959, Kine 1717 (TEX); ruinas et Cerro Guiengola (16°21°N; 95 ae ae 450 m, 26 Nov 1986, Torres 533 (TEX); La Huerta, 20 km NE Weare of Morelos, 9 Oct 1970, Cisneros 2560 (TEX); 15 km N of La Ventosa, 120 m, 25 Nov 1986, Mand za 2708 (TEX). Pursea: 4 km S of oo 1100 m, 29 Jul 1983, Chiang 2370 (TE about 6.5 km ‘SW of Axusco (18°1 97°12’ W), 27 Jun 1987, Salinas 4079 eee — 2. Haplophyton crooksit (L. D. Benson) L. D. Benson, Amer. J. Bot. 30:630. 1943. Basiony: tila var. crooksii L. D. Benson, Torreya 42:9. 1942. Type: U.S.A. Arizona: Pima Co.: “prison road” soldier trail hwy, Santa Catalina Mts, 27 Dec 1939, Crooks & Darrow s.n. (HOLOTYPE: ARTZ; isorypre: B). Representative specimens. U.S.A. ARIZONA: Pima Co.: end of Roger road, Rincon foot- hills, E of — 3200 ft, 25 Asai 1985, Van Devender 85-179 (TEX). New Mexico: Dona Ana Co. na ie Mts, ca 15 mi N of Las Cruces, S ee of Summertord Mt, 26 Aug 1973, ae ym. (TEX). Texas: El Paso Co.: Frankli 1.1 mi W jet Trans- Mountain road (loop and Gateway S$ (31 °53'N; oe x southern exposure of granite boulders, 4600 ft, 26 Aug 1978, Worthington 3207 (TEX). Hudspeth Co.: lime- Witiiams, Notes on Haplophyton 473 stone crevices throughout the cae at the head of the als Arroyo, an intermontane basin in the Quitman Mts, 17 Apr 1976, Butterwick 2377 (TEX). Presidio Co.: near ith of narrow canyon draining s slope of the W es among igneous boulders, i800 ft, oe 1977, Butterwick 38576 (TEX). XIC HIHUAHUA: ca 20 km ENE of Ciudad Jiménez, in limestone arroyo in can- yon aa of summit of eee de Chupaderos (27°12’N; 104°43’W), 5100 ft, 2 Oct 1973, Henrickson 13765 (TEX); ca 31 mi NW of Julimes in SW facing canyon aoe Rancho El Recuerdo in Sierra de Carrasco (28°47'N; 105°09'W), 4400 ft, 15 Sep 1973, Henrickson 12941 (TEX); 46 km W of Ojinaga on hwy to Chihuahua City and S 2 km to the deep canyon of Rio Conchos in the Sierra de Pegiiis (29°32'N, 104°48’°W), 1000 m, 20 Oct 1972, Chiang et al. 9757 (TEX); S slope and top of Sierra del Roque, NNE of Julimes approached from Mina Las Playas via Rancho El Saucito (28°39’N; 105°18'W), 19 Jun 1973, M.C. Johnston et al. 11386B (TEX). Coanutta: § side of Cerro San José de las Piedras (28°42' N; 102°51’W), 1000-1400 m, 8 Jun 1972, — et Aas 7555 (TEX); 0.5 km E of Matrimonio Viejo, E end of limestone ridge (27°08'N; 103°07'W), 1125 m, 2 Sep 1972 Chiang 9122 (TEX). Sonora: Sierra Bojihuacame SE oa, 800-2000 ft, 17- 25 Oct 1954, Gentry 14482 (TEX), talus slope of low basaltic hill, 15 mi S of La Palma, between La Palma and Guaymas, 2 Sep 1941, Wiggins 226 (TEX); Rio Mayo area, hil ; vo microwave tower, 9 mi NW of Alamos and 4 mi SSW of Piedras Verde (27°C : 109°02’W), eat m, 6 Sep 1989, Sanders 9435 (TEX); Cafion de Nacapules, 6 i NE of Bahia San Carlos, deep riparian canyon, 19 Oct 1984, Felger 84-122 (TEX). In his classification of Haplophyton, Pichon (1950) erected the monogeneric subtribe, Haplophytinae. He discussed its affiliation with the Catharanthinae, but ended by questioning the relationship, remarking that “the lignification of the branches {in Hap/ophyton]| seems to be rather rapid and the plant has nothing in common in this point of view with the Catharanthinae” (Pichon 1950; p. 161). Leeuwenberg (1994) subsequently positioned Haplophyton with Alstonia R. Br. in the Alstoniinae, near the Catharanthinae, on the basis that both genera are the only two taxa in the Plumerioideae to possess hairy seeds. Seed pubescence in the Alstoniinae, however, is heceromorphic. Hap/ophyton has seeds with both apical and basal coma, while A/stonia has seeds with either membranous ciliations (sect. Tonduzia) or with both apical and basal coma (sect. Monuraspermum) (Monachino 1949). he dextrorse aestivation found in Haplophyton, suggests a close rela- tionship with Adstonia, which also displays dextrorse aestivation. Indeed, Haplophyton cimicidum may be related to the Oceanic species of Adstonia (sect. Monuraspermum), which have flowers with dextrorse aestivation and seeds with both apical and basal coma (Monachino 1949). Presently, how- ever, it is not known whether or not dextrorse aestivation in Haplophyton is ancestral or derived. Alstonia is the only other genus in the tribe Plumerieae, besides the gen- era of the Catharanthinae (excluding Vinca (Lawrence 1959)), to retain a relexed membranous appendage basal to the stigma throughout its ontog- eny (Pichon 1950). Woodson (1928) noted, however, that a swollen region Aq4 Sipa 16(3) 1995 occurs below the stigmas of Haplophyton. In his view, this character prob- ably represents a primitive stage of the more complex appendage found in the Catharanthinae. Preliminary studies of Haplophyton herbarium speci- mens, by the author, reveal that the style heads of immature buds have a membranous skirt below them. This feature, however, was not present in mature (opened) flowers in either the field or herbarium specimens. Pichon (1950) also suggested an affiliation between Haplophyton and Anechites, stating that the two genera are unique in the family by having hairs with multicellular bases (Fallen 1983). Pichon, however, placed each of these genera in monotypic subtribes, suggesting that their relationships were unresolved. Anechites was subsequently positioned in the subtribe Condylocarpinae of the subfamily Plumerioideae (Leeuwenberg 1994). Haplophyton differs from Alstonta sect. Monuraspermum in being a suffruticose herb (vs. tree), having alternate (vs. whorled) leaves, hairs with multicellular bases (vs. simple), no nectary (vs. an annular nectary), soli- tary flowers (vs. cymous), and its restriction to the New World (vs. Oceania). These differences suggest that Haplophyton is not related to Adstonia. Until further evidence is presented, the most suitable and conservative course favors placing Haplophyton in the monogeneric subtribe Haplophytinae (Pichon 1950) of the tribe Plumerieae, and not with A/stonia in the Alstontinae (Leeuwenberg 1994) whose resemblance to Haplophyton is at present questionable. ACKNOWLEDGMENTS Iam grateful to Carol Todzia, James Henrickson, and the two anony- mous reviewers for their attentive editing of the manuscript, and to my companion Jennifer Forrest for translating Pichon’s work from the original French. B.L. Turner provided helpful comments. Cooperation from F, and TEX-LL was also much appreciated. Finally, lacknowledge my friend Jon Plum, for having the tenacity to accompany me to Chiapas, despite the volatile political situation. REFERENCES Benson, L.D. 1942. Notes on the flora of Arizona. Torreya 42:9-1 1. —______. 1943. Revisions of status of southwestern desert trees and shrubs. Amer. J. Bot. 30:630-31. and R.A. Darrow. 1954. The trees and shrubs of the southwestern Deserts. The ie of Arizona, Tucson, Arizona. Pp. 2§ 32. BENTHAM, G. and J.D. Hooker. 1873. Apocynaceae. In: Genera plantarum. Lovell Reeve & Co., London. 2:722—723. ue , A. Dr 1844. Apocynaceae. In: A.P. de Candolle, Prodromus systematis naturalis regni vereeilis Fortin, Masson & cie, Paris. 8:412—413. Wituiams, Notes on Haplophyton 475 Corrett, D.S. and M.C. Jounston. 1970. Manual of the vascular plants of Texas. Texas Research Foundation, Renner, Texas. P. 1211. FALLEN, M.E. 1983. A systematic revision of Anechites (Apocynaceae). Brittonia 35:222— 243 > Jounston, M.C. 1990. The vascular plants of Texas; A list up-dating the manual of th vascular plans : Texas, 2nd. ed. Published by author. Karresz, J.T. _ A synonymized checklist of the vascular flora of the United States, Cans | 2nd ed. Timber Press, Portland, Oreg LAWRENCE, G.H.M. 1959. Vinca and Catharanthus. Baileya. 7: 113. LEEUWENBERG, A.J. M. 1994. Taxa of the Apocynaceae above the genus com Agric. Univ. ageningen Papers 94(3):45-60. Picuon, M. 1950. Classification des age aee VUE, Supplément aux Plumériotdées. Mém. Mus. Natl. Hist. Nat., Sér. B., Bot. -173 Rosatti, T. 1989. The genera aE eiiban lee Apocyinee (Apacynaceae and oe aeuiae in the southeastern United States, ee J. Arnold Arbor. 70:307-4 SCHUMANN, K. 1895. Apocynaceae. In: gler & K.A. ae Die a ies Pflanzenfamilien. Wilhelm Engelmann, Teipzie. 4(2):109-18 STANDLEY, P.C. and L.O. Winuiams. 1969 . Apocynaceae, in Flora eae Fieldiana, Bot. 24(8):335—407. Woopson, R.E., JR. 1928:. Studies in the i III. A monograph of the genus Amsonia. Ann. Missouri Bot. Gard. 15:3 _ «1938. Apocynaceae. In: N. , Britton et al., North American Flora. New York Botanical Garden, New York. 29:103-192. rule AT6 SIDA 16(3) 1995 BOOK NOTICES Rick, ELroy L. 1995. Biological Control of Weeds and Plant Diseases. Advances in Applied Allelopathy. (ISBN 0-8061-2698-1, hbk.) University of Oklahoma Press, 1005 Asp Avenue, Norman, OK 7301 9, (405) 325-S111. $55.00. 448 pp, 14 figures. “Biological Control of Weeds and Plant Diseases. Advances in Applied Allelopathy, is a timely study, by a world-renowned authority, of the production, by plants and microorganisms, of compounds that affece the growth, health, and oie biology of other plants and microorganisms. Elroy L. Rice focuses on new devel elopments in allelopathy in agriculture and forestry, where the deleterious side effects of synthetic pesticides necessitate more wides pread use of biological control techniques. Summarizing the explosion of knowledge during the last decade, Rice adds his own insights to the various topics reviewed. He tells how, in many cases, biol logical control can be substituted for chemical control with no decrease in crop yields, and how, in other cases, only a minimum of research remains to be done before the data are conclusive.” The following chapters are included: 1) Introduction and Al lelopathic Effects of Crop Plants on Crop Plants; 2) Allelopathic Effects of Weeds on Crop Plants; 3) Other Roles o Allelopathy in Agriculture; 4) All lelopathy in the Biological Control of Weeds; 5) Allel- opathy in Bacterial and Fungal Diseases of Plants: 6) Allelopathy in the Biological Control of Plant Diseases: Host Plants A—M: 7) Allel opathy in the Biological Control of Plant Diseases: Host Plants N-Z; 8) Allelopathy in Forestry. Frey, Kenneru J., Eb. 1995. Historical Perspectives in Plant Science. (SBN 0-8138-2284-X, hbk.). Iowa State University Press, 2121 S, State Avenue, Ames, [A 50014-8300, 1-800-862-6657. $44.95. 216 pp4 6 x 9, illus, Historical Perspectives in Plant Science is a compilation of leccures presented at the 1991 Plant Science Lecture Series sponsored by the lowa State Unive rsity Departments of Agronomy, San Forestry, Horticulture, and Plant Pathology. The lecture series brought together eight scholars who have been called the “makers of plant science history” over the ~_ os half century. The subject matter covered is restricted to higher plants with some agi icultural impor- tance. Included is a general overview of plant science, devel opment of the history of plant physiology, plant pathology, quantitative genetics, cytogenetics, molecular biology, and the me of plane breeding methodology and eh cee The eight chapters in- clude: 1) Biological Revolutions of Thought during the Twentieth C entury, by G. Ledyard Stebbins; 2) Historical Developments in Biological Nitrogen Fixation, by Robert H. Burris; 3) Historical Perspectives on Contributions of Quantitative Genetics to Plant Bree ding, by Bruce Griffing; 4) Contributions of Plant Pathology to the Biological Sciences, by Arthur Kelman; 5) Perspectives on Germplasm Manipulation, by Ralph Riley; 6) Current Perspectives: The Impact of Biotechnology on Plant Improvement, by Charles S. Levings IH, Kenneth L. Korth, and Gerty Cori Ward; 7) Plant Bre eding—A Vital Part of Improve- ment in Crop Yields, Quality, and Production Efficiency, by John W. ie and 8) His- torical Perspectives on Plant Breeding Methodology, by Neal F. Jense NEW TAXA OF RHAMNACEAE FROM CHINA FAN GUO-SHENG and DENG LI-LAN Forest Plant Laboratory Southwest Forestry College Kunming, Yunnan 650224, CHINA ABSTRACT Sageretia yunlongensis G.S. Fan & L.L. Deng sp. nov. from Yunlong Xian, Yunnan, China and Rhamnus hemsleyanus Schneid. var pancinervatus G.S. Fan & L.L. Deng var. nov. from Wuyi Shan Mountain, Fujian, China, are described and illustrated. Kry Worps: New taxa; Rhamnaceae; China RESUMEN Se describen e ilustran Sageretia yunlongensts G.S. Fan et L.L. Deng del condado de Yunlong en vaunan (China) y Roane hemsleyanus Schneid. var. paucinervatus G.S. Fan et L.L. Deng var. nova de la montafia Wuyi Shan en Fujian (China). Sageretia yunlongensis G.S. Fan & L.L. Deng, sp. nov. (Fig. 1) Species haec praesertim habitu S. hewry; Dumm. et Sprag. valde similis, a qua differt foliis oblongo-ellipticis, 10-15 cm longis, 4.5—6.5 cm latis; axibus inflorescentiarum et foliis subtus et costis super folia et petiolis pilosis. Shrub 4 m tall, unarmed; branchlet cylindric. Leaf thick and papery, oblong-elliptic, 10-15 cm long, 4.5—6.5 cm broad, obtuse-rounded to short- acute at apex, rounded at base, pubescent beneath; the midrib pubescent above; the lateral veins 7—8 per side, impressed above, raised beneath; peti- ole thick, 14-20 mm long, densely pubescent. Inflorescence paniculate, 9— 10 cm long in fruit; axis of inflorescence densely pubescent, persistent in fruit; aon unknown. Fruit globose, 3-4 mm diam.; pyrenes 2—3; stem of fruit 2-3 mm long Type: CHINA. YUNNAN: Yunlong Xian, 1300 m alt., Yin Wa-Yuan et al. 161, 26 Oct 1987 (HoLoryPe: SW/FC). Pararypes: CHINA. Tibet, Moers Nian, Tibet Team 74 1702, 17 Aug 1974 (KUN This new species is similar to Sageretia henryi Dumm. & Sprag., but it is easily recognized by its oblong-elliptic, 10-15 cm long, 4.5—6.5 cm broad leaf, and axis of the inflorescence, the lower leaf surface, the midrib above, and the petiole with pubescence. Rhamnus hemsleyanus Schneid var paucinervatus G.S. Fan & L.L. Deng, var. nov. A typo recedit nervis lateralibus paucioribus, 4—5 jugis; petiolis glabris. Sipa 16(3): 477— 478. 1995 478 Sipa 16(3) 1995 Fic... Sageretia yunlongensis G.S. Fan & L.L. Deng, sp. nov. 1. Sheet with fruits, 2. Fruic, 3. Seed, 4. amplified lower surface of leaf. The new variety is similar to var. hems/eyanus, but it can be recognized by its fewer lateral nerves (4—5 pairs) and glabrous petioles. Type: CHINA. Fuyian: Wuyi Shan Mountain, Wang Min-Jing, et al. 1823. 9 Nov 1956 (HOLOTYPE: NAS). Paratypes: 8 Nov 1956, Wang Min-Jing 1782, 19 Apr 1955, 3326 (NAS). AMUR HONEYSUCKLE (LONICERA MAACKH; CAPRIFOLIACEAE): ITS ASCENT, DECLINE, AND FALL JAMES O. LUKEN and JOHN W. THIERET Department of Biological Sciences Northern Kentucky University Highland Heights, KY 41099-0400, U.S.A. ABSTRACT 0-year chronology of interaction between Lonicera maackii and people reveals the diverse roles of arboreta, boranical gardens, and government agencies in plant introduc- tion and eventual naturalization. The species’ biology, horticultural value, and conserva- tion use are descri RESUMEN na cronologia de 150 afios de interaccién entre Lonicera maackii y la gente revela los roles diversos que han tenido los arboreta, jardines botdnicos, y agencias del gobierno en la incroducci6n y eventual naturalizacion de esta planta. Se describe la biologia de esta especie, su valor en horticultura, y su uso para conservacion. In an effort to understand how non-indigenous plants occupy new geo- graphic areas, two processes have received the most attention: population spread as documented from herbarium records (Forcella 1985) and popula- tion growth as mediated by plant traits and community-level interactions (Mack 1985). Although most plant invasions result from accidental or in- tentional introduction by people, relatively little attention has been given to human cultures and the historical and extant exchange systems contrib- uting to plant invasion. A description of these systems, when they operate with plants that eventually escape and become naturalized, may be useful for understanding rates and areal extent of the invasion process. We present here a 150-year chronology of events that eventually led to introduction and naturalization of the eastern Asiatic shrub Lonicera maacki1 (Rupr.) Herder (Caprifoliaceae), Amur honeysuckle, in North America. Throughout most of this time L. maackii was highly valued in gardens and conservation plantings. However, the tendency of the species to naturalize and spread beyond points of original introduction established it as a woody “weed” of concern in eastern U.S. The documented history of interaction between Amur honeysuckle and people is both extensive and varied. The chronology given here may be of value as regulatory decisions are made about future plant introductions. Sipa 16(3): 479 — 503. 1995 480 SIDA 16(3) 1995 The addition of L. maackii to the alien flora of North America can be traced to three historical interactions between the plant and Homo sapiens: (1) Discover the plant, classify it, and describe it as a member of the flora of eastern Asia; (2) Introduce the plant to western horticulture for its attrac- tive foliage, abundant flowers, and showy fruits; and (3) Use the plant to achieve conservation goals, e.g., soil stabilization and/or wildlife-habitat improvement. These three interactions contributed to widespread intro- duction, thus necessitating a fourth and final one: control or remove the plant from the many biotic communities that have been invaded. AMUR HONEYSUCKLE? ITS PEREGRINATIONS From the Pacific to Western oe The Russian Role In the mid 19th century, Russian possessions in coastal eastern Asia were all north of the 55th parallel, a somewhat less than hospitable region. To extend their holdings into more favorable areas, the Russians initiated a series of southward explorations into the relatively greener pastures of Manchuria (Bretschneider 1898). This area, chen “but loosely held in the feeble grasp of the Chinese government,” had excellent harbors and abun- dant resources, including timber. Among the targets was the territory north of the Amur River and that bounded by the Amur and Ussuri rivers, the Sea of Japan, and the Korean frontier. Russia eventually annexed these lands by treaty from China in 1858 and 1860, thus extending its domain south to the latitude of present-day Vladivostok. One of the scientific expeditions sent to explore the valley of the Amur began at Irkutsk in April 1855, returning to Irkutsk 9 months later (Bretschneider 1898). Accompanying the expedition was a naturalist, Ri- chard Maack (1825-1886), professor in the Gymnasium of Irkutsk. Maack is remembered today primarily in the name of a genus of Fabaceae, Maackia, and in the specific epithets of several species, including one in Lonicera. Among the species that Maack found along the Amur in June was the yet-to-be-described Amur honeysuckle; he made but a single collection of the planc—in the Bureja Range north of the Amur about midway between Khabarovsk and Blagoveshchensk (Maximowwicz 1878; Ruprecht 1857). His specimens of woody plants from the trip were sent to St. Petersburg, where they provided the basis for part of the first publication devoted to plants of “Amurland” (Ruprecht 1857), that area on both sides of the Amur between ca. 42° and 55° north and 131° and 141° east (Maximowicz 1859). In chat work the honeysuckle was described as a new species, Ny/osteum maackit, by Ruprecht (Ruprecht 1857). Maack (1859) gave an account of his journey along the Amur; in this work is the first published illustration (at least in western literature) of Amur honeysuckle (Fig. 1). The species was soon included in Maximowicz’s Primitiae florae Amurensis (Maximowicz on) LUKEN AND THIERET, Lonicera maackit 481 Fic. 1. The earliest drawing of Amur honeysuckle (Lonicera maacki1) in western literature. Reproduced from Richard Maack’s Journey on the Amur (1859). Sipa 16(3) 1995 123 45 67 8 9 10¢ ILFORD CIBACOPY To No Unxey, fon NAN-TO ayo From Dn. Fortune, 1845 { dic awcahter Cig : LAL ¢ INDEX FLORA SINENSTS i oA Fic. 2. One of two earliest herbarium specimens (center of sheet) of Amur honeysuckle (Lonicera maackii). The collection, Fortune A34, was made by Robert Fortune in China in 1845, probably at Amoy, and deposited in the Herbarium of the Royal Botanic Gardens, Kew. LUKEN AND THIERET, Lonicera maackii 483 1859), the first flora of Amurland, written after Maximowicz’s first trip (1853-1857) to eastern Asia. Five years after publication of the flora, the species was transferred to Lonicera as L. maackii (Ruprecht) Herder (Herder 1864). Maximowicz did not see the plant in nature before he wrote Primitiae florae Amurensis. However, in a later expedition (1859-1864) he obtained specimens of it for St. Petersburg from five localities (Herder 1878): near the mouth of the Amur, Don . 1947. Plants for permanance: the late honeysuckle. Fl. Grower 34:18. sei D. 1938. The world was my garden. Travels of a plant explorer. Charles Scribner’s ons, New York. FarRINGTON, E.]. 1931. Ernest H. Wilson ed hunter. a a list of his most important introductions and where to get them. The Stratford Co., FERNALD, M.L. 1950. ile s manual of on 8th ed. D. Nocuane Blew: York. FLorAL Committee. 1908. Floral Committee, June 11, 1907. J. Roy. Hort. Soc. 33:Ivii-lix. ee ae ee ce C ommittee, June 8 1915. J. Roy. Hort. Soc. 41:¢xx—cxxiil. FORCELLA, . 1985. Final distribution is related to rate of spread in alien weeds. Weed Res. 25:18 9] eee a 1915. The flora of north-western Yunnan. J. Roy. Hort. Soc. 41:200—208, fig. 70-77 1916. Notes on the flora of north-western Yunnan. J. Roy. Hort. Soc. 42:39 46, Froune, D. and H.J. PrANnper. 1984. A colour atlas of poisonous plants. A Wolfe Science ook, London. GLEASON, H.A. 1952. The new Britton ne Brown illustrated flora of the northeastern United States . adjacent Canada. Vol. 3. New York Botanical Garden, Bre eer, A. Cronquist. 1963. Man ee of vascular seg of Se ea United States and ae Canada. D. Van Nostrand Co., New Y anc 1991. Manual of vascular planes of northeastern United States and adjacent Canada. 2nd ed. New York Botanical Garden, oe Happen, N.G. 1925. A selection of honeysuckles. Garden 89:638— Hansen, N.E. 1909. T he wild alfalfas and clovers of Siberia, with a perspective view of the alfalfas of the world. U.S.D.A. Bur. Pl. Industr. Bull. 150. LUKEN AND THIERET, Lonicera maackii 499 Hao, K.-s. 1934. Caprifoliaceae. In T.-n. Liou, Flore illustrée du nord de la Chine 3:1—94. 37 pls. Hara, H. 1983. A revision of Caprifoliaceae of Japon with reference to allied plants in other cee and the Adoxaceae. Ginkgoana 5:1—336. Harty, EM. 1993. How IIlinois kicked the exotic fabio Paves 195-209 mm B.N. McKnight ed. Biological pollution: The control and impact of invasive exotic species. Indiana Academy of Science, Indianapolis. Harvit_, A.M. Jr., T.R. BRADLEY, and C.E. Stevens. 1981. ne of the Virginia flora. Part II. Dicoerledoas Virginia Botanical Associates, Farmville, et al. 1992. Atlas of the Virginia flora. III. ie not indicated, Burkeville, VA. HENry, 1932. A few unusual loniceras. Natl. Hort. Mag. 11:165-166. HERDER, . v. ae Plantae Raddeanae erase aes Die Monopetalen Ostsibiriens. Bull. Soc. Imp. cree Moscou 37(1):190—235. _—s«*d'8 78. Adlerndiaa ett emendands ad plantas Raddeanas monopetalas. Bull. Soc. Imp. nee Moscou 53:1— Horto UNIVERSITATIS ee 1924, ie seminum in Horto Universitatis A. 1924 collectorum. Botanic Garden, University of eee Copenhagen, Denmark. Hortus BOTANICUS ae PETROPOLITANUS. 1887. Delectus seminum quae Hortus Botanicus Imperialis Petropolitanus promutua commutatione offert. Imperial Botanic Garden, St. pee «899. Delectus seminum quae Hortus Botanicus Imperialis Petropolitanus pro mutua commutatione offert. Imperial Botanic Garden, St.Petersburg Hsu, P.-s and H.-j Wana. 1984. Notes on the genus Lonicera Linn. of China. Acta Phytotax. Sin. 22:22-41. Un Chinese; English summary. and LO .«d'‘O988. Leonicera. Pages 143-259. In: Flora Reipublicae Popularis ere 72. (In Chinese ) INGOLD, J.L. and M.J. Craycrarr. 1983. Avian a on honeysuckle (Lonicera) in south- western sai .A. Ohio J. Sci. 83:256-2 IwaTsuKI, K. et al., eds. 1993. Flora of Japan. a 3a. Kodansha, Tokyo. JANAKI AMMAL, EK. and B. SauNpERs. 1953. Chromosome numbers in species of Lonicera. Kew Bull. 1952:539-541. JARDIN BOTANIQUE DE L'UNIVERSITE D’AMSTERDAM. 1929. Catalogue des graines récoltées dans le Jardin os a de Université d’ Amsterdam. Botanic Garden, University of neeat ae Amsterdam, Nethe S KoeEHNE, E. 1893. Deuescie Dendrologie. Ferdinand Enke, Stuttgart. KOKHNO, 1986. Derev 'iai kustarniki [Trees and shrubs}. Naukova Dumka, Kiev. (In Russian.) KOLBEK, — and M. Kucera. 1989. A brief survey of selected woody species of North Kor .P.R.K.). Botanical Institute, Czechoslovak Academy of Sciences, Pruhonice, -choslovakia. ae. ss 1907. Flora Manshuriae. Vol. 3. Imperial Botanic Garden, St. Petersburg. (In Russian.) Kurata, S. 1971. Illustrated important trees of Japan. Vol. 3. Chikyu Shuppan Co., To- kyo. (In Japanese and English.) Lee, T.B. 1989. Illustratec flora of Korea. ee Seoul. (In Korean.) Leveau, A.M., M. DuraANb, and R.R. Paris. 1977. Sur toxicité des fruits de divers Lonicera (Caprifoliacées). Pl. Med. Picicas 11:94-1 LEVvEILLE, H. 191 5S—1916. Catalogue des plantes du Yun- te Chez l’Auteur, Les Mans. Ya jon 500 Stpa 16(3) 1995 Lorenz, D.G., W.C. SHarp, and J.D. Rurener. 1989. Conservation plants for the North- east. U.S.D.A., Soil Conservation Service, Program Aid 11 Luken, J.O. 198s. Population structure and biomass cea of the naturalized shrub Lonicera maackit (Rupr.) Maxim. in forest and open habitats. Amer. Midl. Naturalist 119:258—267 we ee 1990. Forest and iat communities respond differently to cutting of exotic Amur honeysuckle (Kentucky). Restor. Manage. Notes 8:122-123. ete and NY, ee ee 993: Seedli ing distribution and potential persistence of the exotic shrub Lonicera maackii in fragmented forests. Amer. Mid]. Naturalise 133: 124-130. and D.T. Mattimiro. 1991. Habitat-specific resilience of the invasive shrub Amur See e (Lonicera maackii) during repeated clipping. Ecol. Applic. 1:104— 109. and J.W. Thieret. 1987. Sumac-directed patch succession on Northern Ken- tucky aor embankments. Trans. Kentucky Acad. Sci. 48:51—54. Maack, R. 1859. Puteshestvie na Amur {Journey on the Amur]. Tip. K. Vulfa., St.Petersburg. (In Russian.) Mack, R.N. 1985. Invading plants: their potential contribution to population biology. Pages 127-142 im J. White, ed. Studies on plant demography: John L. Harper Festschrift. ademic Press, London. Maximowicz, C.J. 1859. Primitiae florae Amurensis. Mém. Acad. Imp. Sci. St. Pétersbourg Divers Savants 9:1-504. 10 p/s. (Reprint 1859 by Kaiserliche Akademie der Wissenschaften, Saint Petersburg 1878. eee eee Diagnoses plantarum novarum Asiaticarum. II. Bull. Acad. Imp. Sci. Saint-Pétersbourg, 3rd ser. 24:26—-89. (Published also in Mélanges Biol. Bull. Phys.- Math. Acad. Imp. Sct. Saint-Pétersbourg 10:43—134. 1880.) McCain, W.E. and E.A. ANDERSON. 1990. Loss of hill prairie through woody plant inva- sion at Pere Marquette State Park, Jersey County, Illinois. Nat. Areas J. 10:69-75. MITCHELL, R.S. 1986. A checklist of New York state plants. New York State Mus. Bull. ior rer, 8S. 1907. Chévrefeuilles nouveaux. Rev. Hort. 79:298—300. eine Db Hisrotrre Natrureiie. 1931. Index seminum Muse: Parisiensis anno 193 collectorum. Muséum d’ Histoire Naturelle, Paris. Nakal, T. 1909. Flora Koreana. Pars prima. J. Coll. Sci. Imp. Univ. Tokyo 26. ere eee 11. Flora Koreana. Pars secunda. J. Coll. Sci. Imp. Univ. Tokyo 31 —___.. 192 la. Flora sylvatica Koreana. Pars XI. Caprifoliaceae. The Government of Chosen, an a Japanese and Latin.) 192 rear eee ey . Tentamen systematis Caprifoliacearum Japonicarum. J. Coll. Imp. Univ. hs en —_____.. 1938. A new classification of the genus Lonicera in the Japenese ou ae together with the diagnoses of new species and new varieties. J. Jap. Bot. 14:359—376. NicHoson, G. 1888. Illustrated peo! of gardening. Vol. 4. L. Upcote Gil London. Nopa, M. 1971. Flora of the n.-e. province (Manchuria) of Cl no place indicated. (In ee ae English.) Nysorr, R. 1992. Vegetation management guideline: bush honeysuckles—Tatarian, Morrow's, Belle, and Amur honeysuckle (Lonicera tatarica L., L. morrowti Gray, L. xbella Zabel, and L. maackii [Rupr.}] Maxim.). Nat. Areas J. 12:218—219. Onw1, J. 1965. Flora of Japan. Smithsonian enn Washington, DC. PALiBIN, J. 1898. Conspectus florae Koreae. Pars prima (Ranunculaceae-Campanulaceae). Acta Horti Petrop. 17(1): 1-128 hina. No publisher indicated, LUKEN AND THIERET, Lonicera maackii 501 Penkovsky, V.M. 1901. Derev’ia i kustarniki [Trees and shrubs]. Vol. 3. O.D. Khodshinoy, Kherson. (In Russian.) Perripes, G.A. 1986. A field guide to trees and shrubs. 2nd ed. Houghton Mifflin, Boston. Poyarkova, A.I. 1958. Caprifoliaceae Vent. Pages 419-584 in Flora URSS 23:419-584. (In Russian.) PRINGLE, J.S. 1973. Lonicera maackii (Caprifoliaceae) adventive in Ontario. Canad. Field- Naturalist 87:54—55 Purpus, A. 1900. Lonicera maacki {sic} Maxim. Gartenwelt 4:235. ReGeL, E. 1884. Lonicera maackii Maxim. Gartenflora 33:225—226, Table 1162. REHDER, i — Synopsis of the genus Lonicera. Annual Rep. Missouri Bot. Gard. 14:27— 23252 1913. Neue oder kritische Gehélze. Mitt. Deutsch. Dendrol. Ges. 1913:254— 265: 1949a. Bibliography of cultivated trees and shrubs hardy in the cooler tem- perate regions of the northern hemisphere. Arnold Arboretum of Harvard University, Jamaica Plain, MA 949b. Manual of cultivated trees and shrubs. 2nd ed. Macmillan, New York. Roya Botanic GARDEN, EDINBURGH. 1924. List of seeds collected in the Royal Botanic Garden, Edinburgh during the year 1924. Royal Botanic Garden, Edinburgh. Royat Botanic GARDENS, DusLin. 1919. List of seeds for exchange. Department of Agri- culture and Technical Instruction for Ireland, Dubli Roya. Botanic GARDENS, Kew. 1934. Hand-list of trees — shrubs (excluding Coniferae) cultivated in the Royal Botanic Gardens{,}Kew. His Majesty's Stationery Office, Lon- don Roya. Garbens. 1896. Hand-list of trees and shrubs grown in arboretum. Part I. Gamopetalae to monocotyledons. Her Majesty’s Stationery Office, London. 18 = Ruprecut, FJ. 1857. Die ersten botanischen Nachrichten iiber das Amurland. Zweite Abtheilung: Baume und Straucher, Conn von Richard Maack, bestimmt von F.J. Ruprecht. Bull. Cl.Phys.-Math. Acad . Sci. Saint-Pétersbourg 15:356—-383. (Re- published in Mélanges Biol. Bull. Phys.- on Acad. Imp. Sci. Saint-Pétersbourg 2:5 13— 568. 1858.) SarGENT, C.S., ed. 1913. Plantae Wilsonianae. Vol. 1. Publ. Arnold Arbor. 4. 1922. The first fifty years of the Arnold Arboretum. J. Arnold oe :128- Scrvti C.K. 1911. Illustriertes Handbuch der Laubholzkunde. Elfte Lieferung. Gustav Fischer, Jena an W.C. 1970, "Rem-Red’honeysuckle. Amer. Nurseryman 132(9):7, 32. R. Belcher. 1981. 'Rem-Red’ honeysuckle—A multipurpose land- scape shrub. Amer. Nurseryman 153(12):94—95 SOUTHMEADOW Fruit Garbens. 1994. Choice and unusual fruit varieties for the connoiseur nd home gardener. Southmeadow Fruit Gardens, Lakeside, MI. Swink, FE. and G. WitHeLM. 1994. Plants of the Chicago region. 4th ed. Indiana Academy of Science, ea Taytor, H.J. 1941. To ee ae prairies and plains. The life and work of Niels Ebbesen Hansen. Bios, - Ver TuHatTcHER, A.E. 1922 eae ak at Aldenham. Gard. Chron., 3rd ser. 71:114—115, 123,137,179; 1995215: Tozer, E. 1994. On the trail of E. H. Wilson. Horticulture 72(9):50—-59. 502 Sipa 16(3) 1995 RISEL, D.E, and D.L. Gorcuov. 1994. Regional distribution, ecological impact, and leaf "eae of the invasive shrub, Lonicera maackii. Bull. Ecol. Soc. Amer. (Suppl.) 75(2):23 1-232. (Abstract.) TRUELOVE, W. L915. Lonicera maackti. Gard. ia 3rd ser. 57:348. L917. Lonicera maackit. Gard. Chron., 3rd ser. 62:252. UNIVERSITET Borne Have. 1917. ee seminum, fructum, sporarum quae Hortus Botanicus Christianiensis mutua offert. University Botanical Garden, Oslo. U.s.p.A. 1899a. Foreign seeds and plants imported by the section of seed and plant intro- duction. Numbers 1—1000. U.S.D.A. Div. Bot. Invent. —______.. 189 9b. Foreign seeds and plants ea by the section of seed and plant introduction. Numbers 1001-1900. U.S.D.A. Div. Bot. In . 1909. oe and plants Imporecd during the eal rom April 1 to June 30, 1908. U. S.D.A. r. Pl. Industr. Bull. 142. eee LIS, se and plants imported ss the period from January | to March S1., 1912. US.D.A. r. Pl. Induscr. Bull. 28 Peer eee nee of seeds and eee imported by the Office of Foreign Seed and Plant eine des une the period from April | to June 30, 1912. U.S.D.A.B Pl. Induser. Seeds ee . Inventory oF seeds and plants imported by the Office of Foreign Seed and Plane pene during the period from October | to December 31, 1916. U.S,D.A, Bust Pe Industr. Invent. Seeds 49. —_—_____.. 1923. Inventory of seeds and plants imported by the Office of Foreign Seed and Plane frereduction during the period from April 1 to June 30, 1921. U.S.D.A. Bur. Pl. Induscr. es Seeds 67. —_________.. 1926. Seeds and plants imported by the Office of Foreign Seed and Plant ncn, Bureau of Plant Industry, during the period from April 1 to June 30, 1923. U.S.D.A. Invent. 7 ey ——_______.. 1928. Plant material introduced by the Office of Foreign Plane Introduc- tion, Bureau ie Plant Industry, January | to March 31, 1926. U.S.D.A. Invent. 86. ee rare 29. Plant material introduced by the Office of Foreign Plant Introduc- tion, Bien a he Industry, January | to March 31, 1927. U.S.D.A. Invent. 90. ——_______.. 198]. Plant material introduced January | to December 31, 1978. U.S.D.A. PI. Invent. a —__________.. 1982. Plant material introduced January | to December 31, 1979. U.S.D.A. PI. Invenc. 187. —________. 1983. Plant material introduced January | to June 30, 1980. U.S.D.A. PI. Invent. 188(1). —____.. 1984. Plant material introduced January | to December 31, 1983. U.S.D.A. Pl. Invenr. 191. —________.. 1985. Plant material introduced January | to December 31, 1984. U.S.D.A. Pl. ee LOZ: Voir’, E. 1899. Materialy dlya izucheniya russkikh” vidov” zhimolostey [Materials for the study of the Russian species of honeysuckle]. Izv. $.-Peterburgsk. Lesn. Inst. 3:1— 47, pls. 1-15. (In Russian.) Wana, C.-w. 1961. The forests of China with a survey of grassland and desert vegetation. Harvard University, Cambridge, MA. Wittiams, C.E., J.J. Rattey, and D.H. Taytor. 1992. Consumption of seeds of the invasive Amur honeysuckle, Loniera maackti (Rupr.) Maxim., by small mammals. Nat. Areas J. 12:86-89 LUKEN AND THIERET, Lonicera maackit 503 Witson, E.H. 1913. A naturalist in western China. Vol. 1. Doubleday, Page and Co., New York. 1917. Aristocrats of the garden. Doubleday, Page & Co., Garden City, NY. . 1925. Honeysuckles in bush and vine. House & Gard. 47:90-91, 114, 116. «1927. Lonicera korolkowii floribunda. Bull. Popular Inform. Arnold Arbor., 3rd ser. 1:44. 1928. More aristocrats of the garden. Stratford, Bos 1929. Lonicera maackit podocarpa. Bull. Popular oo Aina Arbor., 3rd ser. 3:34—3 6. Woerkorr, A.D. 1941. What can the Manchurian flora as well as the flora of neighbouring countries give to gardens of Manchuria itself and other countries with cold climates. Publisher not indicated, Harbin Wyman, D. 1962. The honeysuckles. Arnoldia 22:57-67. 1969. Shrubs and vines for American gardens. The Macmillan Company, London. Yost, S.E., S. ANTENEN, and G. HartwiGsen. 1991. The vegetation of the Wave Hill natural area, Bronx, New York. Bull. Torrey Bot. Club 118:312—325 504 Sipa 16(3) 1995 BOOK NOTICES COMMITTEE ON MANAGING GLOBAL GENETIC Resources. 1993. Managing Global Genetic Resources. Agricultural Crop Issues and Poli- cies. (ISBN 0-309-04430-8, hbk.). National Academy of Sciences Press, 2101 Constitution Avenue, N.W., Washington, D.C. 20418, (202) 334-3180. $49.95. 449 pp., 6 x 9. “Managing Global Genetic Resources. A oo Crop Issues and Policies examines the struc- ture that underlies efforts to preserve genetic material, including the worldwide network of genetic collections, the role of aie and a host of scientific, institutional, legal, economic, and political issues that surround management and use. The Executive immary, with the committee's major recommendations; an Overview, which introduces the subject for those readers without a background in genetic conservation; and two parts. Part one addresses basic science issues and entails Chapters 1 to 10. Part two addresses policy issues and entails Chapters 11 to 15.” The 15 chapters are: 1) Genetic Vulnerability and Crop Diversity; 2) Crop Diversity: Institutional sal aise 3) In Situ Conservation of Genetic Resources; 4) The Science of Collecting Genetic Resources; 5) The Science of Managing Genetic Resources; 6) Using Genetic Resources; 7) Biotachinoloay and Germplasm Conservation; 8) Documentation of Genetic Resources; 9) The Conservation of Genetic Stock Collections; 10) The Genetic Resources of Microorganisms; | 1) Exchange of Genetic Resources: Quarantine; 12) Exchange of Genetic Resources: Proprietary Rights: 13) Genetic Resources: Assessing Economic Value, 140 Conflicts Over Ownership, Man- agement, and Use; and 15) National and International Programs. PRIMACK, RICHARD B., 1993. Essentials of Conservation Biology. (ISBN 0-87893-722-6, hbk.). Sinauer Associates, Inc., Sunderland, Massa- chusetts 01375, U.S.A. 564 pp. This elegance text provides a well-organized and comprehensive summary of the primary issues as well as the biological and economic principles involved in the new discipline of conservation biology. The text opens with concise discussions of the origins of the disci- pline and definitions of key terms. Concluding chapters review the applications of conser- vation biology principles to problems of the effective management of habitat and remnant reserves as well as human dominated lands and degraded lands to be restored Essentials of Conservation Biology will be useful in pone eres introductory courses. Abundant use is made of interesting worldwide case examples in the narratives as well as in side boxes. The pane figures are clear and well-se mae to illustrate the narrative. Lists of art readings are provided at the end of each chapter. Talented teachers of conservation biology, who usually represent a wide interdisciplinary background anc deep personal commitment, will find this text an important foundation upon which to custom- ize their courses. —Lawrence D, Ford. Qu A SYNOPSIS OF THE GENUS CLUSIA SECTIONS CRIUVOPSIS AND BRACHYSTEMON (CLUSIACEAE) IN NORTHERN SOUTH AMERICA JOHN J. PIPOLY II Botanical Research Institute of Texas 509 Pecan Street Fort Worth, TX 76102-4060, U.S.A. ALLISON GRAFF PO, Box Gt San Gregorio, CA 94074, U.S.A. ABSTRACT The genus Clusia sections Crinvopsis and Brachystemon are revised, in preparation for treatment of the Clusiaceae for Flora of the Venezuelan Guayana and Flora de Colombia. i plified descriptions, keys to the species and lectotypifications are provided for each section. In Clusia section Crinvopsis, 3 species are recognized, of which one, Clusia araracuarae, is described as new, illustrated, and its phylogenetic relationships discussed. A mplified de- scriptions, illustrations and discussions of phylogenetic relationships are oe for Clusia amazonica and C. martiana. Clusia oedematopoidea is reduced to synonymy under C. amazonica, and C. uleana under C.martiana. In Clusia section Brachystemon, 3 taxa are recognized, of which one, Clusia aymardii, is described as new, illustrated, and its phylogenetic relation- ships are discussed. Amplified descriptions and discussions 0 phylogenetic relationships are provided for Clusia spathulaefolia and C. penduliflora. RESUMEN Como resultado de investigaciones para preparar un tratamiento taxonomico del género Clusia para Flora of the Venezuelan Guayana y Flora de Colombia, se presenta una revision de las secciones Criavopsis y Brachystemon. Se presentan descripciones actualizadas para las dos seccibnes, claves para separar sus especies, y se lectotipifiquen.. Se reconoce tres especies en la seccién Criuvopsis, entre las cuales, C. araracuare, se describe como especie nueva, se ilustra, y se discute su parentesco. Se provee descripciones ampliadas, ilustraciones y comentarios sobre ee ieee para Clusia amazonica,y C. martiana. Se relega C. ai ei a sinonimia bajo C. amazonica y, a C. uleana ajo C. martiana. In seccion chystemon, se teconoce tres especies, entre las cuales, C. oat se ors como especie nueva, se ilustra, y se discute su parentesco. También, se provee ione pliadas, y comentarios sobre relaciones filogenéticas para C. pubuediles y C. 5 ne INTRODUCTION The genus Clusia Linnaeus contains over 250 species (with 336 names), distributed throughout the Neotropics, with two, C. rosea Jacquin, and C. minor Linnaeus, widely cultivated throughout the tropics. Despite taxo- Sipa 16(3): 505 — 528. 1995 506 Sipa 16(3) 1995 nomic overdescription, it is estimated that there are at least an additional 40 as yet undescribed species in the Andes of Colombia, Ecuador, Peru and Bolivia, particularly in the premontane, montane (including ceja de selva), subpdramo, paramo and jalca habitats. The genus is easily recognized by its capsular, septicidally dehiscent fruits, with few to numerous stamens, and copious cream, yellow or white latex. Characters traditionally used to define the sections of the genus include the structure of the androecium in staminate flowers, and corresponding structure of the stigmas and staminodes of the pistillate. The last comprehensive revision was that of Vesque (1893), in which Engler’s (1888) classification was simplified. The taxonomy of the group is difficult owing to its sexual lability, with stami- nate, pistillate, bisexual (Hammel 1986) and apomictic (Hammel 1986; Maguire 1976) individuals known. In addition, I have seen staminate flowers on reiterative (sensu Hallé et al. 1978) shoots of otherwise functionally piscillate plants in Clusza pusilla Steyermark and in C. grandiflora Splitgerber, in the Pakaraima Mountains of Guyana and on the Chimanta Massif com- plex of tepuis in the eastern portion of the state of Bolivar, Venezuela. While preparing a diagnostic treatment of the genus for the E/ora of the Venezuelan Guayana and a more comprehensive one for Flora de Colombia, sections Crinvopsis and Brachystemon were revised. These two sections form a natural group, closely related to section Crivva Bentham. Engler (1893) considered both sections Criwvopsis and Brachystemon to be subsections within subgenus Crivva sensu Bentham (1862). In fact, this study grew out of the difficulty in distinguishing young plants of Clusia (Brachystemon) penduliflora from C. (Crinvopsis) martiana. Because the formats of the floristic treat- ments in preparation do not allow for complete descriptions and synonymy, the present treatment is intended to provide data for the groups over their entire range, until a more comprehensive revision is prepared. TAXONOMIC TREATMENT Clusia L. section Criuvopsis Planchon and Triana, Ann. Sci. Nat., Ser. 4 13:322. 1860. Lecrorype Species (here designated): Clusia amazonica Planchon and Triana, Ann. Sci. Nat., Ser. 4 13:358. 1860. Clusia section Cri#va Bentham and Hooker, subsection Crinvopsis (Planchon and Triana) Engler, Fl. Bras. 12(1):401. 1888. J. Vesque in A. DC and DC, Monogr. Phan. 8:77. 1893. Clasia subgenus Crivva (Bentham and Hooker) Engler, section Criavopsis (Planchon and Triana) Engler, Nat. PAlanzenfam. 3(6):225. 1895. Glabrous lianas or hemiepiphytes; latex white to translucent. Leaves petio- late. Inflorescence a terminal panicle, the fowers cymose; secondary inflo- rescence bracts 2, coriaceous; bracteoles 4, decussate, coriaceous. Perianth spreading; sepals 4—5, chartaceous, opposite the petals; petals 5, contorted, Pipoty AND Grarr, Synopsis of Clusia 507 very thickly carnose; androphore pentagonal, stamens numerous, the fila- ments short, flat, apically free, connate at the base, fleshy, often thickened at the base; anthers basifixed, muticous, equalling or slightly longer than the filaments, truncate to emarginate apically, dehiscent by longitudinal slits; pistillode absent, the central receptacle resiniferous or not. Pistillate inflorescence and flowers as in staminate, but staminodia 5, free, linear, with vestigial anthers; carpels 5, stigmas chin, elliptic to oblong, peltate; ovules horizontally ascending;. Fruit a capsule, ovoid to oblongoid, stigma and styles persistent. Distribution.—Three species, distributed from Panama to Colombia, Venezuela, Ecuador, Peru, Bolivia and Amazonian Brazil. Clusia section Crinvopsis appears to be most closely related to C. section Brachystemon by virtue of its short, thick filaments of the fertile stamens, sessile stigmas, and linear anthers deshiscent by longitudinal slits. How- ever, section Criavopsis may be easily distinguished from section B rachystemon by the flat filaments connate basally, and very thin inflorescence rachises. —_— KEY TO SPECIES OF CLUSIA SECTION CRIUVOPSIS 1. Leaf blades oblong, bullate; sepals 4; petals 5; anthers rounded to slightly ots C. advardcnardae emarginate apically and basally .. 1. Leaf blades obovate to elliptic, smooth; sepals 5, petals 5; anthers muticous, 2. Leaf base obtuse to broadly rounded, leaves coriaceous, adaxial latex canals inconspicuous, adaxial venation and submarginal collecting vein prominent; liana. .... 2. C. amazonica 2. Leaf base acute to cuneate, leaves subcoriaceous, adaxial latex canals highly conspicuous, adaxial venation inconspicuous, submarginal collecting vein barely discernible; scandent, epiphytic shrub, at times appearing 5 lianous 3. C. martiana 1. Clusia araracuarae Pipoly, sp. nov. (Fig. 1) Propter lamina coriacea ad bases obtusa vel late rotundata, nervos secundarios conspicuous nervo submarginale conjunctos, C. amazonicae valde arcte affinis, sed ab ea laminis oblongis (non obovatis vel ellipticis), necnon bullatis (nec laevibus), sepalis 4 (non 5), tan- dem antheris ad apices basesque emarginatis vel rocundatis (non muticis), praeclare distat. Scandent lianous shrub, to 23 m long, 4 cm diam. glabrous; branchlets tetragonal in cross section, squarrose, 3-5 mm diam., bark of apical por- tions rufous, appearing scaly, checking transversely, gray at maturity, rug- ose, latex white, copious. Leaves decussate; blades coriaceous, oblong, (8)13- 18 cm long, (2.5-)3—3.5(—4) cm wide, apex abruptly acuminate to rostrate, base widely rounded to truncate, deeply bullate, nitid above at first, then pallid above and below at maturity, midrib deeply impressed above, promi- nently raised below, secondary veins 17—32 pairs, perpendicular to midrib, deeply impressed above, prominently raised below, united by a prominent 508 Sipa 16(3) 1995 y \ , a SSS ma THUR Try WF t t can 2 “AS fh ites Ne PIG. 1. Clasta araracuarae Pipoly. A. Habit, showing the oblong, bullate leaves. B. Closeup of cyme, showing callose keels of bracteoles. C. Quadrate pedicel. D. Outer sepals. E. Inner sepals. F. Petals. G. Androecium, showing pentagonal morphology and apically rounded anthers. AG, drawn from holotype. PIPOLY AND GrafF, Synopsis of Clusia 509 submarginal collecting vein, latex canals linear, conspicuous, margin revo- lute, entire; petioles deeply canaliculate, 2.5—2.8 mm long, glabrous. Stami- nate inflorescence terminal, erect, pyramidally paniculate, i? cm long and 5—7 cm wide in bud, the branches cymose, flowers in 3’s; peduncle 3.5—5 cm long, subtended by two reduced leaves similar to vegetative leaves but 3—5 cm long, 1.4—1.8 cm wide; secondary inflorescence bracts carnose, depressed-ovate, 1.3—1.6 mm long, 1.6—-1.8 mm wide, apex rounded, me- dially carinate, the margin opaque, not scarious, entire; pedicel squarrose, 2.8-4.2 mm long; bracteoles four, 2 decussate pairs, carnose, depressed- ovate to suborbicular, apex obtuse, carinate medially, the keel callose, hya- line, the margin entire, glabrous, the outer 2—2.5 mm long, 2.3—2.6 mm wide, the inner 2.5—3 mm long and wide. Staminate flower buds cream; outer sepals 2, opposite, coriaceous, orbicular, 2.3-6 mm long and wide, apex very widely rounded, cucullate, hyaline except apically, the margin entire; inner sepals 2, decussate to the outer ones, coriaceous, as outer se- pals but reddish, hyaline apically; petals 5, contorted, carnose, oblong, 0.8— 1.1 mm long, 0.4—0.6 mm wide in bud; androphore convex, pentagonal as seen from above; stamens numerous, 0.9—1 mm high, the filaments flat, 0.6-0.8 mm long, the anthers linear, 0.1—0.2 mm long, emarginate or rounded apically and basally; pistillode absent, not resiniferous. Pistillate inflorescence unknown. Fruit unknown. Type: COLOMBIA. Amazonas: Araracuara, along Rfo Caquetd, right bank, 3 km above Sumaeta, 0°36'S, 72°10'W, riparian floodplain, 200-300 m, 30 Sep 1990 (stam. fl bud), E. Alvarez, A.C. Londotio, A. Rodriguez, & F. Moreno 116 (HoLoTYPE: COAH!; tsorype: NY!). Distribution.—Known only from the type. Ecology and conservation status. —Clusia avavacuarae occurs in Igapé veg- etation (Prance 1979), forming mats in canopies of trees along black water river margins. The area in which this species occurs is subject to extreme disturbance owing to forest destruction, largely for firewood. Araracuara is a significant population center and busy port on the Rio Caqueta, and there- fore, this species should be considered threatened. The long inflorescences, lianous habit, leaves with prominent secondary venation and submarginal collecting vein and obtuse to broadly rounded bases, indicate that Clusia ; is most Closely related to Clusia amazonica However, the bullate leaf blades, fewer sepals, and anthers that are rounded to emarginate apically and basally clearly set C. araracuarae apatt. 2. Clusia amazonica Planchon and Triana, Ann. Sci. Nat., Ser. 4 13:358. 1860. (Fig. 2). Typr. BRAZIL. Amazonas: near Panure along the Rio Uaupes, Oct 1852 ae el fl), R. Spruce 2878 (HOLOTYPE: P!; NY Neg. 5033; IsoTyPE: P!; NY Neg. 503 510 Stipa 16(3) 1995 Il. Amazonica. Lge = Clusia amazonica Planchon et Triana. Habit (center); stcaminate bud, pistillate flower sad ee ite flower, stamens in abaxial and lateral views, androecium, and longisection through pistillate flower (counterclockwise, from upper left). Adapted from Engler in Martius, Flora Brasiliensis, 12: Plate 83. C Aust vedematopoided setlo Phytologia 39:73, 1978. Type. PANAMA. Panama: J. ke 13557 (HoLoryee: NY: isoryPr: MO!). Liana; branchlets seteeontl in Cross section, squarrose, (2.5—)3—5.5(—7) mm diam., bark of apical portions rufous, appearing scaly, checking trans- versely and exfoliate, beige at maturity, rugose; latex white, copious. Leaves decussate; blades coriaceous, oblong, elliptic or rarely obovate, (8.0—)10.5— 20.0(—26.0) cm long, (3.5—)4.0-4.6-10.5(-12.2) cm wide, apex broadly rounded, with an abruptly short-acuminate tip, the acumen 0.5—1 cm, base obtuse to broadly rounded, not decurrent on the petiole, smooth, sordid above, pallid below, latex canals varying in conspicuousness above, but never readily visible without maginification below, midrib prominently raised above and below, secondary veins numerous, prominently raised above, slightly raised below, united by a fine submarginal collecting vein, the mar- PIPOLY AND GRafFF, Synopsis of Clusia 511 — gin subrevolute, entire, glabrous; petioles canaliculate, (2.1—)2.7—4.0 (5.8) cm long. Inflorescence terminal, erect, pyramidally paniculate, (6—) 12—24 cm long, 6-10 cm wide at maturity, the flowers cymose, 3 per cyme; pe- duncle (2—)4—8(-11.5) cm long, subtended by normal leaves; secondary bracts coriaceous, very widely ovate, 1.5-2.5 mm long, 1.5—2 mm wide, apex obtuse to rounded, carinate, the ae yalnG scarious, entire. Stami- nate inflorescence with tetragonal pedicels, 3-5 mm long; bracteoles 4, decus- sate, the outer 2 coriaceous, very widely ovate, 1.5—2.5 mm ohne 1.5-2 mm wide, apex obtuse to rounded, carinate, the margin hyaline, scarious, entire, the inner 2 suborbicular to orbicular, 2.5—3 cm long and wide, apex rounded, flat, the margins entire, hyaline, scarious. Staminate flowers white; sepals 5, contorted, membranaceous, brittle, suborbicular to orbicular, 5S—6 mm long and wide, apex broadly rounded, translucent, densely lineate, the margin hyaline, scarious; petals 5, contorted, carnose, somewhat cucullate, oblong, 3.5—4 mm long, 2.5—3 mm wide, apex obtuse to rounded, some- what cucullate, densely lineate, the margin hyaline; aaa flat or slightly concave, pentagonal, the stamens numerous, free, 2.5—3.5(—3.7) mm long at maturity, the filaments short, flattened, ca. 1-1.2 mm long, the anthers linear, 1.5—2.5 mm long, 0.7—1 mm wide, apex truncate, base truncate, dehiscent by wide longitudinal slits, abruptly expanding at base to filament, the connective flat, darkened; pistillode obsolete, resin scanty. Pistillate inflorescence as in staminate but (1.5—)2—3(—5) mm long; outer bracteoles deltate, 1.8—2 mm long and wide; inner bracteoles oblate, 2.7— 3.5 mm long, 3—3.7 mm wide. Pistillate flowers white, as in staminate but sepals orbicular, 6-7 mm long and wide; petals strongly cucullate, 6—6.5 mm long, 3.5—4 mm wide, staminodes 5, flat, 2.5-3.5 mm long, the fila- ment ca. | mm long, abruptly tapering at apex, the anther linear, ca. 1.5— 2.5 mm long, apex broadly rounded, not well-differentiated from apex of filament, the thecae not fully differentiated and devoid of pollen, the con- nective darkened; pistil 3-5 mm long, 4-G mm wide; carpels 5; styles absent, stigmas thinly coriaceous, suborbicular, peltate, 1.5—-2 mm long and wide, thickening with age, flat in early ontogeny, convex at maturity; ovules numerous. Fruit orale cream, suffused with purple along the carpel sutures, oblongoid, 2—3.5 cm long, 1.5—2.5 cm diam. Common name.—Jubaga bakoko” (Mui language, Colombia) Distribution.—From lowlands of central Panama southward along the Pacific coast of Colombia (Chocé Floristic Province), along and below the western slopes of the Andean Cordillera Occidental of Ecuador, eastward through the Amazon Basin of Colombia and Brazil, and south-southwest- ward through the Amazon Basin of Ecuador, Peru and Bolivia, 100-400 (—1,690) m elevation. 512 Sipa 16(3) 1995 Ecology and conservation status. —Clusia amazonica is a riparian species, tol- erant of moderate amounts of disturbance but not capable of surviving long periods of indundation. It usually occurs high enough up in the canopy to permit some of the foliage to remain above periodic floodwaters. It oc- curs in varzea, igap6, tahuampa and other periodically inundated habitats. Unlike Clusia martiana, C. amazonica spends its entire life as a liana, and does not form large mats over the canopies of the trees. It is only very locally common, but is not considered threatened. Specimens examined. COLOMBIA. Amazonas: Rio Caqueta River a 00°50'S, 71°50'W, 20 Nov 1991 (ster.), J. Duivenvoorden et al. 1259 (BRIT, COAH, COL, U); Mpio. Leticia, 03°47'S, 70°15'W, 100 m, 14 Apr 1991 (fr), J. Pipoly 15398 - COL, FMB, MO); Antioquia: sae San Francisco, a iaae Aquitania, Finca La [usi6n, 1,200— 1,500 m, 2 Apr 1992 (stam. fl; pist. fr- mixed coll.), R. Fonnegra et al. 4129 (BRIT, COL, HUA, JAUM, Vo VENEZUELA. Amazonas: Dae. Atabapo, along Rio Cunucunuma, near Cul — 03°44'N, 65°44'W, 210 m, 15—16 Feb 1985 (stam. fl), R. Liesner 17516 J, VEN), (fr), R. Lieswer 17535 (BRIT, MO, VEN); Salto Yureba, Cafio Yureba, Bajo — 04°03'N, 66°01'W, 120-150 m, 4 Nov 1981 (stam. fl bud), F Delascio & F. Gudnchez 10841 (MO, NY, US, VEN), 350 m, 15-16 Mar 1985 (pist. fl, fr), R. Leesner 18755 (MO, VEN); Dea. Atures, near Remo a 04°34'N, 67°18'W, 180 m, ae 1989 (fr), E. Foldats & J. Velazco 9580 (MO, PORT, VEN); Depto. Casiquiare, 12 km SE of San Fernando de Atabapo, Sector “El Pozo,” CVG Experiment Station, 03°50'N, ae 110 m, 10-16 Feb 1988 (pist. A), G. Aymard et al. 6391 (MO, PORT, VEN); G. Aymard et al, 6404 (MO, PORT, VEN); Depto. Rio Negro, Cerro de la Neblina, vicinity of base camp along Rfo Mawarinuma, afluent of Rio Baria, 00°50'N, 66°10'W, 110-130 m, 8-10 Jan 1984 (stam. fl bud), J. Steyermark & J. Luteyn 129759 (B, MO, NY, US, VEN); Margins of Rfo Gavilan, between Raudal Gavilancito and Cerro Pel6n, 05°37'N, 67°22'W, 80— 120 m, 10-11 Feb 1992 (pist. fl), G. Romero et al. 2373 (AMES, MO, VEN). ECUADOR. Esmeraldas: Ecological Reserve Cotocachi-Cayapas, Parroquia Luis Vargas Torres, Rfo Santiago, Pote Salt ae 00°49'N, 78°45'W, 250 m, 23-27 Oct 1993 (fr), M. Tirado et al. 536 IT, 3, MO). Napo: Cantén Orellana, Parque Nacional Yasunf, “Maxus” road and oil ee = 20, 00°33'S, 76°30'W, 250 m, 11-15 Aug 1993 (pisc. fl bud), M. Autestia cane (BRIT, MO, QCNE); ae well “Amo II,” 00°52'S, 76°05'W, 230 m, 29-30 Ma (fc), F. Covello 152 (BRIT, MO, QCNE); Cantén Tena, ee Sacha Biological ee , Rio Napo, 8 km E of Mista, 01°04'S, 77°36'W, 450 m, 20 Jan 1990 (fr), C. Cerén 8386 (BRIT, MO, QCN 22-24 Feb 1988 (fr), W Palacio 2489 (BRIT, MO, QCNE), 400 m, 17-28 May oe a W. Palacios 4224 (BRIT, MO, QCNE), 17-28 May 1989 (fr), W Pais 4307 (BRIT, MO, QCNE), 6 May 1990 (fr), W. Palacios et al. 4954 (BRIT, MO, QCNE), 13 May 1990 (fr), W. Palacios & E. Fretre 5105 (BRIT, CNE). 14 May 1990 (fr), W. Palacios G& E. Freire 5145 (BRIT, MO, QCNE), Perma- ne Acne along Rio Chiguipino, 450 m, 22 May 1992 (fr), z Guidino & J. Zuleta 1660 (BRIT, MO, QCNE). Pastaza: Cantén Pastaza, “Masaramu” oil well, of UNOCAL, 40 km NNW of Montalvo, 00°44'S, 76°52'W, 400 m, 1-16 May 1990 (fr), E. Guidiiio 362 (BRIT, MO, QCNE), PetroCanada road, Via Auca, 115 km $ of Coca, 5 km S of Rio ea 01°15'S, 76°55'W, 320 m, 26-31 Jan 1989 (fr), D. Neil G EF Hurtado 8749 T, MO, QCNE), 1-6 Mar 1989 (ster.), V Zak 4119 (BRIT, MO, QCNE). Sucumbios: sae Gonzalo Pizarro, Campo Bermejo No. 6 Norte, 30 km NE of Lago Agrio, 00°14'N, 77°13'W, 1,050 m, 23 Mar 1990 (fr), C. Cerén 9227 (BRIT, MO, QCNE). PERU. Loreto: Prov. Maynas; Allpahuayo, Institute of Amazonian Investigations (IIAP) Experimental PipoLy AND GrarF, Synopsis of Clusia 513 Station, 04°10'S, 73°30'W, 150-180 m, Permanent Inventory, Plot A, Subplot 10, Nov 1991 (stam. fl), R. Vasquez & N. Jaramillo 14830 (AMAZ, BRIT, MO, USM), 27 May 1991 (ster.), R. Vasquez G N. Jaramillo 10618 (AMAZ, BRIT. MO, USM); Dtto. Fernando Lores, Caserfo Serafin, Quebrada Tamshiyacu, from Caserfo Serafin to Caserfo Constancia, 125-130 m, 10 May 1991 (fr), C. Grdndez et al. 2606 (AMAZ, BRIT, MO, USM); Dtto. Iquitos, 1977 (stam. fl), J. Revilla 3312 (AMAZ, BRIT, F, MO, USM), Caseria Rio Nanay, right bank, 20 Dec 1976 (stam. fl), J. Revilla 89 (AMAZ, F, MO, USM); Rio Yarapa, 04°20'S, 73°30'W, 122 m, 20 Nov 1989 (stam. fl bud), C. Grandez — Sea 1466 (AMAZ, MO, USM). BOLIVIA. Beni: Prov. Ballivian, area San Borja Hwy, E side of Serranfa del Pilén Lajas, 15°13'S, 67°03'W, 850 m, 2 Nov 1989 (fr), . Smith & V. Garcia 13865 (LPB, MO). La Paz: Prov. Nor Yungas, near . at entrance to road to Chairo, 2 km from Chairo, 2,430 m, 5 Sep 1987 (pist. fl), §. Beck 12969 (LPB, MO), Prov. Morillo, Valle de Zongo, along trail from end of road at the Cahua Power Plant, 1 ,660— 1,690 m, 23 Nov 1980 (pist. fl), 7: Croat 51413 (LPB, MO), 16°05'S, 68°03'W, 1,400— 1,600 m, 22 Apr 1982 (fr), J. Solomon 7491 (LPB, MO). BRAZIL. Amazonas: Alto Rio Solimées, Mpio. de Sdo Paulo de Olivenga, plateau S of city, road to Bom Fim, 25 Nov 1986 (stam. fl), C. Cid et al. 8552-A (BRIT, IAN, INPA, MG, NY). Pard:Ilha do Marajo, Rio Mocées, 2 km upriver from Anajas, 00°57'S 49°56) W, 13 Nov 1987 (fr), 8. Beck et al. 484 (BRIT, INPA, MG, NY), center of Iha ene 27 Oct 1984 (stam. fl), G. Sobel et al. 4829 (BRIT, INPA, MG, NY). Rondénia : 28 km from Vilhena on road to Colorado, 13°00'S, 60°00'W, 29 Oct 1979 (fr), B. Ne/son et al. 330 (BRIT, IAN, INPA, MG, NY). Clusia amazonica is most closely related to C. araracuarae, but is readily distinguished by its smooth, obovate to elliptic leaf blades, muticous an- thers, and inconspicuous latex canals. This species is apparently sympatric with C. martiana, but occurs in riparian forests above and away from the inundation zone. Forest inventory data indicate that C. amazonica is much more infrequent than C. martiana, and does not form large mats as the latter species does. The type of C. vedematopoidea represents populations with smaller, less coriaceous leaves and smaller, more compact inflorescences than the Ama- zonian populations of C. amazonica. A collection from Antioquia, Colom- bia (R. Fonnegra et al. 4129) is identical to the type of C. vedematopordea. Collections from Bolivia (S. Beck 12969, T. Croat 51413, J. Solomon 7491) resemble C. martiana in quantitative features of the leaves, but the promi- nent secondary veins and submarginal collecting vein, and, absence of sig- nificant numbers of latex canals visible on the adaxial leaf surface, clearly indicate they belong to C. amazonica. 3. Clusia martiana Engler, Fl. Bras. 12(1):411. 1888. (Fig. 3). Type: BRA- ZIL. AMazonas: Alto Amazonas, in forests of the Rio Japura, Dec 1852 (stam. fl), C. Martius 3033 (HOLOTYPE: M, n.v.). es uleana Engler, Bot. Jahrb. 58:Beibl. 130:1. 1923. Syn. Nov. Type: BRAZIL. AZONAS: near Fortaleza, along Rio Jurua, 6 Nov 1901 te fl bud), E. Ule 6006 rie pe: B-destroyed, F Neg. 9205 !). A search for remaining duplicates of the type collection is underway and neo- or lectotypification 1s ee until the where- abouts of duplicate collections is known, or if any are extant. Sipa 16(3) 1995 Fic. 3. Clusia martiana Engler. Habit (center); androecium, anthers in abaxial and lateral views, staminate Hower bud, staminate Hower bud with sepals and bracteoles removed, top view of staminate flower in anthesis (from left to right). Adapted from Engler in Martius, Flora Brasiliensis, 12:Plate 95 Scandent epiphytic shrub, at times appearing lianous to 4 m tall; branchlets terete to somewhat tetragonal or pentagonal in cross section, at times squarrose, 2.5—3.5 mm diam., bark of apical portions rufous, scaly, checking transversely and exfoliate, rufous and glabrous at maturity, latex clear, moderate. Leaves decussate; blades subcoriaceous, obovate to oblan- ceolate, (6.5—)10-14.5 (-19)cm long, (2.3—)3.5—5.0(—6.5) cm wide, apex narrowly acute to abruptly acuminate, the acumen 0.3—1.5 cm long, base acute to cuneate, not decurrent on the petiole, smooth, glabrous above and below, sordid above, pallid below, latex canals dense and conspicuous above and below, midrib slightly raised above, prominently raised below, second- ary veins numerous, smooth, inconspicuous above, slightly raised below, submarginal collecting vein barely discernible, the margin flac, entire, gla- brous; petioles canaliculate, 1—1.5(—2) cm long, glabrous. Inflorescence ter- minal, pendent, pyramidally paniculate, 2-4 cm long, 3 | oe 3-5 cm wide at PIPOLy AND Graff, Synopsis of Clusia S15 maturity, the flowers 3 per cyme; peduncle 0.8—1.5 cm long, subtended by normal leaves; sceondary bracts coriaceous, oblate, 2.3—2.6 mm long, 3.5— 4 mm wide, apex widely rounded, carinate, the margin somewhat hyaline, very narrowly scarious, entire, glabrous. Staminate inflorescence with tet- ragonal pedicels, the lateral 1.8—2.2, the terminal 2.5—3.5 mm long, gla- brous; bracteoles 4, the outer 2 coriaceous, oblate, 2.3—2.6 mm long, 3.5— 4 mm wide, apex on dey rounded, carinate, the margin somewhat hyaline, very narrowly scarious, entire, glabrous, the inner 2 stiffly coriaceous, ob- late, 2.5—3 mm long, 3.3—-4.5 mm wide, apex broadly rounded, somewhat carinate, the margins entire, somewhat translucent, not scarious, glabrous. Staminate flowers white to yellow; sepals 5, contorted, membranaceous, very widely ovate, 5-7 mm long, 4-6 mm wide, apex broadly rounded, somewhat cucullate, densely lineate, me ce hyaline, scarious, glabrous; petals 5, contorted, carnose, oblong, 5.3-5.8 mm long, 1.8—2.2 mm wide, apex rounded, flat, obscurely lineate, be margin entire, opaque, not scari- ous, glabrous; androphore flat, pentagonal, the stamens numerous, free, 2.1—3.3 mm long at maturity, the filaments short, flat, anther 1-1.3 mm long, apex muticous, base truncate, dehiscent by longitudinal slits, the connective flat, truncate, darkened; pistillode absent. Pistillate flowers white, as in staminate but sepals orbicular, 3.5-4 mm long and wide; petals 4.5— 5 mm long, 2.8—3.3 mm wide, apex obtuse to rounded; staminodes 5, 3— 3.5 mm long, the filaments wide, | mm long, 2 mm wide, abruptly con- ee apically to anthers, the anthers linear, ca. 1 mm long, 0.5 mm wide, apex muticous, base not distinguishable, with iene eclinal slits, de- void of pollen, the connective darkened; pistil 4 mm long, 5 mm diam.; carpels 5; styles absent, stigmas thinly coriaceous, oblongoid, peltate, 1.8— 2.2 mm long, 1—-1.2 mm wide, flat in early ontogeny, convex at maturity, ovules numerous. Fruit greenish white with red hue in intercarpelary areas, ovoid, 2—3.5 cm long, 1.5—2.5 cm wide, Distribution. —Amiazon Basin of Colombia, Venezuela, Peru, Bolivia and Brazil, 1O0O—450(—1,850) m elevation. Ecology and conservation status. —Clusia martiana is a hemiepiphyte, grow- ing along riverbanks, where it forms mats in and over the canopies of low trees, eventually killing them. This species can withstand extended peri- ods submerged. I have seen the fruits floating on rivers, but it is doubtful that dispersion by fish is significant, because of the large numbers of small birds I have seen feeding from open capsules, on the red arillate seeds. It is commonly found on the margins of varzea, and “tahuampa” forests, which are on lateritic soils, but is also rarely found in sandy loams (‘varillal”) and very rarely found in riparian forests with some pockets of white sand. Clusia —_— martiana is one of the few weedy species of Clusia. 516 Sipa 16(3) 1995 Specimens examined: COLOMBIA. Amazonas: Mpio. Leticia, Parque Nacional Natu- ral Amacayacu, Matamata trail, 03°47'S, 70°15'W, LLO—120 m, 28 Oct 1991 (ster.), J. Pipoly 15610 (BRIT, COL, FMB, MO), near Quebrada Matamata, 100 m, 11 Mar 1991 (fr), A. Rudas et al. 1542 (BRIT. COL, FMB, MO), along Quebrada Bacaba, 100 m, 16 Apr 1992 (fr), A. Rudas & A. Prieto 4286 (BRIT, COL, FMB, MO), along ‘ae of Rio Amacayacu, 100 m, 9 Apr 1991 (ster.), J. Pipoly et af. 15133 (BRIT, COL, FMB, MO); Vereda Puerto Narino, 03°45'S, 70°15'W, 100 m, 3 Aug 1989 (ster.), R. Viigues et al. 12510 (AMAZ, BRIT, COL, FMB, MO, USM). Antioquia: Mpio. Amalff, 8-27 km NE of Almafi, en route from Vetilla to Fraguas, near Salazar and Marengo, 06°00'N, 75°04'W, 1,150—1,450 7 Dec 1989 (pist. fl), R. Callejas et al. 9114 (BRIT, COL, HUA); Mpio. Campamento, Vereda Llanadas, 6—12 km W of Campamento en route to Mina Las Brisas, 07°05'N 75°20'W, 1,650-1,810 m, 7 Sep 1989 (pist. fl), R. Callejas et al. 8266 (BRIT, COL, HUA). VENEZUELA. Amazonas: Isla Sebastian, Rio Casiquiare above Chapezo6n, between Boca and Solano, 01°58'N, 67°03'W, 120 m, 31 Jan 1980 (pist. fl), R. Liesner & H. Clark 8941 (MO, VEN); Depto. fg: 50, oe Iguapo, Alto Orinoco, 15 km SE of La Esmeralda, 03°00'N, 65°28'W, | ‘eb 1990 (stam. fl), G. aide Delgado 8227 (BRIT, PORT, VEN); Danie, eee me Jénita floodplain, 7 km before a amo river mouth, 02°46'N, 64°54'W, 170 m, Feb 1990 (stam. fl), A. Pee 7318 (BRIT, PORT, VEN) Rio Mayaca, 02°01'N, 65°07'W, 228 m, 6 Feb 1989 (stam. fl), A. ae et al. 987 (BRIT, NY, VEN); Alto Orinoco, along riverside 2 ioe above La Esmeralda, . Mar 1953 (stam. fl), B. Maguire & J. Wardack 34708 (MO, NY, VEN); Rio Padamo, 10 km above mouth, 27 Mar 1953 (stam. fl), B. Maguire GJ. W a 34718 (MO, NY, ae Depto. Casiquiare, near Bea along Rio Casiquiare, 5 Feb 1991 (piste. fl, fr), ML. Collela et al. 1829 (BRIT, NY, VEN); Isla Sebastian, along Rio Casiquiare between Boca and Solano, 01°58'N, 67°03'W, o m, 31 Jan 1980 (pist. fl), R. Liesner & H. Clark 8941 (MO, VEN). PERU. Cuzco: Dtto. cine 8 km W of Quincemil, right margin of Rio Maniri, toward the mouth of Rfo Araza, 13°17'S, 70°48'W, 720 m, 17 Jul 1990 (seam. ff), M. Timand & H. Astete 646 (BRIT, CUZ, MO, USM). Junin: Mazamari, 1,000 m, 13 Sep 1960 (pist. Al, fr), F Woytkowski 6019 (MO, US, USM). Loreto: Prov. Maynas, Buena Vista, Rio Tahuayo, 04°15'S, 73°10'W, 140 m, 24 Jan 1981 (stam. fl), R. Vdsquez & N. Jaramillo 1229 (AMAZ, F, MO, USM); Explornapo Camp, Quebrada Sucusari, Rio Napo, 03°15'S, 72°55'W, 130 m, 30 May 1991 (ster.), A. Gentry et al. 74277 (AMAZ, BRIT, MO, USM), Explornapo Camp, 03°20'S, 72°55'W, 140 m, 18 Apr 1991 (fr), R. Vasquez GN. Dh a 16141 (AMAZ, BRIT, MO, USM); Vicinity of Iquitos, 1977 (stam. fl), J. Revilla 3254 (AMAZ, BRIT, F, MO, USM); Rio Mamén, 6 Sep 1972 (stam. fl), T: Croat 20028 (AMAZ, MO, USM); Santa Maria de Nanay, 10 a W of Caserfa Mishana, Reserva Cocha Yaramé, along ee Sie 03°55'S, 73°35'W, 130 m, 15 Mar 1991 (ster.), J. Pipoly et al. 15031 (AMAZ, BRIT, MO, USM), (fr), /. os et 15036 (AMAZ, BRIT, CUZ, MO, USM), Rio i ane San Juan de Muniches, ca. 40 min above Iquitos), 120 m, 19 Mar 1977 (fr), A. Gentry et al. 18422 (AMAZ, F, MO, USM); Rio Nanay, 6 turns ssses Iquitos, 28 Dec 1976 (pist. fl, fr), J. Revilla 2098 (AMAZ, F, MO, TEX, USM); Rio Nanay between a and Puerto Almendras, 13 Jul 1976 (stam. fl), A. Gentry & J. Revilla 16710 (AMAZ, F, MO, USM); see Iquitos, Rio Nanay, near Santa Clara, 19 Aug 1976 (stam. fl), J. Reale 1144 (AMAZ, F, MO, USM), 13 Nov 1976 (stam. fl), J. Revilla ane ences, CUZ, F, HUT, MO, cn 14 Nov 1976 (pist. fl bud), J. Revilla 1817 (A Z, F, MO, USM); Rio Nanay, near Santa Clara, ca. 10 km above Morona Cocha, 150 m, o Dec 1976 (stam. fl), C. Davidson & J. Revilla 5415 (AMAZ, F, MO, USM); Nauta, Quebrada Saragosa, 04°29'S, 73°35'W, 200 m, 29 Mar 1987 (fr), R. Vasquez & N. Arévalo 9048 (AMAZ, MO — > wa PipoLty AND GRAFF, Synopsis of Clusia 517 USM); Rio Yavarf, across river from Brazilian village of Paomari, 23 Nov 1977 (stam. fl), A. Gentry & J. Revilla 20814 (AMAZ, BRIT, CUZ, F, HUT, MO, TEX, USM); Along Rio Blanco, above Tamshiyacu, 140 m, 17 Mar 1978 (fr), C. Diaz et al, 215 (AMAZ, CUZ, F, , USM); Caserfo Nina Rumi, along Rfo Nanay, 23 Feb 1976 (fr), J. Revilla 199 (AMAZ, MO, USM); San Antonio, Rio Pintuyacu, 03°40'S, 73°54'W, 160 m, 21 Apr 1986 (fr), RK. Vasquez et al. 7487 (AMAZ, CUZ, F, MO, USM); Quebrada Yanayacu, tributary of Rio Manati, 03°45'S, 72°55'W, 110 m, 25 Jan 1989 (pist. fl, fr), R. Vasquez G N. Jaramillo 11573 (AMAZ, BRIT. MO, US). BOLIVIA. La Paz: Rio Saturiapo, 14 km SW of Ixiamas, second ridgetop, 13°53'S, 68°15'W, 620-650 m, 2 Jun 1990 (bud), A. Gentry & R. Foster 70864 (F, LPB, MO); Prov. Nor Yungas, 4.5 km below Yolosa, then 14 km W on road along Rio Huarinilla, on abandonded road to Hacienda Sandillan, S side of river, 16°12'S, 67° 50 W, 1,200-1,300 m, 24 Jan 1983 (fr), J. Solomon 9391 (LPB, MO); Prov. Murillo, 1/ 2-3/4 by trail Sa es the Cahua Hydropower Plant, 16°05'S, 68°03'W, 1,400— 1,600 m, 22 Apr 1982 (fr), J. Solomon 7491 (LPB, MO); 44 km below Lago a o Dam vicinity of Cahua ee plant, 16°03'S, 68°O L'W, 1,200 m, 12-15 Sep 1983 ia, fl), J. Solomon 10829 (LPB, MO). BRAZIL. Roraima: SEMA Ecological oe Ilha de Maracé, Parimiu, 03°20'N, 62°58'W, 24 May 1987 (fr), W Milliken & S. Bowles 276 (MO). Clusia martiana is isolated within the section, but may be more closely related to C. amazonica than to C. araracuarae. It is readily separated from both of those species by the subcoriaceous leaves with acute to cuneate bases, the highly conspicuous, numerous linear latex canals, and barely discernible submarginal collecting vein. I have observed that C. martiana 1s an epiphytic shrub at first, then becomes lianous with age. Clusia martiana is most frequently confused with C. (section Brachystemon) penduliflora be- cause they share the nearly the same habit, have relatively short inflores- cences (compared to C. amazonica), and are partially sympatric. However, the oblongoid stigmas, flattened, longer filaments and subcoriaceous leaves of C, martiana easily distinguish it from C. penduliflora and C. amazonica. The type of Clusia uleana represents those populations of C. martiana with short branches and inflorescences, and slightly larger lowers than those of the average population. However, it is otherwise identical to the major- ity of C. martiana populations. The type specimen was destroyed in Berlin, and no duplicate has been found in HBG or other herbaria thus far. How- ever, the Field Museum of Natural History (F) photograph clearly shows the characteristic squarrose branchlet tips with horizontally checking and exfoliating bark, numerous, very highly conspicuous foliar linear latex ca- nals, and obscure secondary venation, thus confirming its identity as Clusia martiana. Clusia section Brachystemon (Engler) Pipoly, comb. nov. Sie section Crinva pubsecdon oe Engler in Martius, Fl. Bras. 12(1):412. 1888. LecroryPE ): Clusia puialaaile Engler in Martius, FL. . 12(1):412. 1888. 518 Sipa 16(3) 1995 Clusia section Criuva subsection Crinvopsis, pro parte, Vesque, Monogr. Phan. 8:34. 1893. Clusia subgenus Crivva Bentham section Brachystemon (Engler) Engler, Pflanzenfam. 3(6):225. 1893. Free-standing trees or hemiepiphytes; latex white or cream. Leaves peti- olate. Inflorescence a terminal panicle; branches cymose; secondary inflo- rescence bracts 2, carnose; bracteoles 4, decussate, carnose or coriaceous. Perianth erect, sepals 5—9, membranaccou or coriaceous, the outer 2 op- posite, the others contorted; petals 4—5, contorted, chartaceous, coriaceous or thickly carnose; androphore te stamens numerous, subsessile on free, short, terete or prismatic filaments; anthers linear, basifixed, muticous, much longer than the filaments, apically truncate or emargin- ate, dehiscent by longitudinal slits; pistillode absent or very reduced, the central receptacle resiniferous. Pistillate inflorescence and flowers as in stami- nate, but staminodia 5, free, che filament broad, abruptly tapering apically, anther linear, apex muticous; carpels 5; ovules horizontal: stigmas sessile, pelcate, elliptic or narrowly triangular. Fruit a capsule, oblongoid or de- pressed-globose, stigmas persistent. Distribution.—Three species in the Amazon Basin and adjacent Andes of Colombia, Venezuela, Ecuador, Peru, Bolivia and Brazil. Clusia section Brachystemon is most closely related to section Crinvopsts, but is separated by the short, subterete or prismatic filaments of the an- thers, and compact inflorescences with relatively thicker rachises. It is re- stricted to the Amazon Basin and adjacent Andes of South America, often along riverbanks. The three species of the section can be separated by the following key. KEY TO SPECIES OF CLUSIA SECTION BRACHYSTEMON 1. Hemiepiphyces; leaf blade coriaceous, dull above and below, the apex broadly rounded to an abruptly acuminate tip; flowers on pedicels 3—10 mm long; stigmas elliptic; fruits smooth, not ribbed 1. C. penduliflora 1. Free-standing trees; leaf blade stiffly coriaceous to cartilaginous, nitid above, pallid below, the apex broadly rounded to truncate, without acuminate tips Howers sessile to subsessile, the pedicels less than | mm | form, fruits strongly ribbed. reer &sugn 2. Leaf blades oblanceolate to obovate, the base attenuate, decurrent on the petiole, che margins revolute; petioles broadly marginate, 1-1 ace cm long; branchlets 7-10 mm diam C. spathulaefolia 2. Leaf blades oblong to elliptic, the base obtuse to broadly nounee not Taeeantent on the petiole, the margins flat; petioles subterete, 3-4.5(— long; branchlets 10-22 mm diam. 3. C. aymardti 1. Clusia penduliflora Engler, Fl. Bras. 12(1):412. 1888. (Fig. 4). Tyee. BRAZIL. Amazonas: Near Panure along Rio Uaupes, Dec 1852 (stam. fl), R. Spruce 2792 (HOLOTYPE: G, n.v.; ISOTYPE: P!), Hemiepiphyte; branchlets quadrate to subterete, 3.5—5 mm diam., gla- PIpOLy AND Graff, Synopsis of Clusia bab!) : le nN Fe ae ours a - a led ee hs : | _ CLUSIA 4. |, penduliflora. 7 a | Fic. 4. Clusia penal Engler. Habit (center); staminate flower bud, staminate flower bud with sepals and bracteoles removed (lower left, from left to right); top view of see (upper right); anthers in adaxial, abaxial and lateral views (right side, left to right). Adapted from Engler in Martius, Flora Brasiliensis 12:Plate 84 brous, bark gray to grayish-brown, not transversely checked or exfoliating; glabrous; latex white to cream, scanty. Leaves decussate; leaf blades coria- ceous, obovate to rarely elliptic, (8.8—)14—22.5 cm long, (3—)6.5-11.0 cm wide, apex broadly rounded to an abruptly acuminate tip, or rarely acute, base obtuse to acute, not decurrent on the petiole, dull above and below, at times with numerous, conspicuous latex canals, midrib raised above, promi- nently raised below, secondary veins inconspicuous, numerous, extending to the margin, without submarginal collecting vein, the margin flat, en- tire, glabrous; petioles canaliculate, 1.3—1.7(—2.3) cm long, glabrous. In- florescence: terminal, pendent, pyramidal paniculate, 3—7(—8.5)cm long, 3-8 cm wide, the branches cymose, 3-flowered at inflorescence apex, 2—3 flowers per cyme below; peduncle (0.7—)1—2 cm long, subtended by nor- mal leaves; secondary bracts carnose, depressed ovate, 2—3 cm long, 2.5— 3.5 cm wide, apex rounded, carinate, the margin opaque, not scarious, en- tire, glabrous. Staminate inflorescence with tetragonal pedicels, 3-10 mm long, glabrous; bracteoles 4, decussate, carnose, the outer 2 suborbicular, 3-4 mm long, 3—3.5 mm wide, apex broadly rounded, medially carinate, the margins entire, opaque, not scarious, the inner 2 coriaceous, orbicular, 5—6 mm long and wide, apex broadly rounded, carinate, the margin entire, opaque, scarious. Staminate flowers greenish pink; sepals 7—9, the outer se- 520 Sipa 16(3) 1995 pals 2, opposite, coriaceous, suborbicular to oblate, 5—6 mm long, 6-7 mm wide, apex very widely rounded, subcucullate, the margin entire, scarious, the inner sepals 5—7, contorted, membranaceous, brittle, very widely ovate, 6-7 mm long, 5—7 mm wide, translucent, apex rounded, venation promi- nent, the margin hyaline, scarious; petals 5, contorted, coriaceous, widely obovate, 5—7 mm long, 3-4 mm wide, apex truncate to very broad| ly rounded, linear latex canals numerous, prominent, the margin hyaline; androphore flat or slightly convex, the stamens numerous, free, 3.5—3.8 mm long, the filaments short, terete, ca. | mm long, the anthers linear, 2.5—3.5 mm long, ca. 0.5 mm wide, apex truncate, base truncate to broadly obtuse, dehiscenc by longitudinal slits, the connective terete, darkened: pistillode subobsolete to obsolete, resiniferous. Pistillate inflorescence as in staminate but with outer bracteoles 2.5—3 3.5 mm long, 2.5—3.5 mm wide, inner 2 bracteoles suborbicular to oblate, a 3.5 mm long, 2.5—4 mm wide pedicels subobsolete to 3 mm long. Pistillate ae green, as in the stami- nate but sepals suborbicular, 3.5—5 mm long, 5—7 mm wide, the inner sepals suborbicular to very widely ovate, 5-8 mm long, 5—7 mm wide; petals prominently eee staminodes 5, flat, 3.5-4 mm long, the fila- ments 2—2.5 mm long, 2—2.2 mm wide, abruptly tapering apically; an- thers rectangular, 1-1.2 mm long, 0.5—0.6 mm wide, muticous apically, the connective darkened, the thecae not fully differentiated and devoid of pollen; piscil ss chose. 4—6 mm long and in diam.; carpels 5; styles ab- sent; stigmas thinly coriaceous, elliptic, peltate, 1.5-2 mm long, 1-1.5 mm wide, persistent, flat in early a convex at oe ovules numerous. Fruit green, oblongoid, 2—3.5 cm long, 1.3—2.5 cm diam. Common names.—‘ Matapalo” (Napo-Ecuador). Distribution.—Central and western Amazonia of Brazil, Colombia, Ec- uador and Peru, 5O—500(-1,800) m elevation. Clusia penduliflora is expected to occur in Amazonian Venezuela in forests on terra firme. Its apparent ab- sence there is probably a collection artifact. Ecology and conservation status. —Clusia penduliflora is a hemiepiphyte that grows in primary lowland moist, wet and rain forests, on alluvial terraces just above floodplains. It may be found in tree canopies in areas of high incident oe near the transition zones with varzea forests. Sp |. COLOMBIA. Chocé: Along Rio San Juan, near Docord6, 0-100m, 04°15! N, 77°22'W, 29 ee fe (stam. fl), E. Forero et as 4328 (COL, MO); road to Caby, SW 7 Tutunendo, 100 m, 19 Jan 1979 (pist. fl), A. Gentry & E. Renteria 24444 (COL, JAU MO). ECUADOR. Esmeraldas: Lita-San Lorenzo Road, ca. 30 km NW of Lita, Ae NL, 78°40'W, 300-500 m, 12 May 1991 (fr), A. Gentry et al. 70025 (BRIT, cay QCNE). Morona-Santiago: Campamento La Playa, road construction camp 23 km SE o San Juan Bosco, 1,050 m, 28 Jan 1981 (stam. fl), A. G entry et al. 30932 (MO, SEL). ee Cantén Aguarico, Reserva Faunistica Cuyabeno, Laguna Zancudo Cocha (Iripari), 00°33'S, Prpoty AND Grafr, Synopsis of Clusia a2 75°32'W, 230 m, 28 Sep 1991 (pist. A), W. Palacios et al.. 7770 (MO, QCNE); Cantén Tena; Jatun Sacha Biological Station, Rio Napo, 8 km below Misahualli, 01°04'S, 77°36'W, 450 m, 17 Jan—6 Feb 1987 (pist. fl), C. Cerén 614 (MO, QCNE), 17-28 May 1989 (fr), W. Palacios 4274 (BRIT, MO, QCNE), 14 May 1990 (fr), W. Palacios & E. Freire 5151 (MO, QC 17-24 Feb 1988 (stam. fl), C. Cerén 3622 (MO, QCNE), 24 Nov 1987 (pist. fl), D, Neil &S. eed 8013 (MO, QCNE); Parque Nacional Yasunf, “Daimi” Oil Well 2, 00°55'S, 76°11'W, 200 m, 26 May-8 Jun 1988 (fr), C. Cerén & F. Hurtado 4084 (MO, QCNE); an ee NE of runway, Feb 1970 (stam. fl), R. Mowbray 7026 (MO, SEL). Pastaza: Via Auca, 110 km S of Coca, 10 km from Rfo Tigiiino, sector Cristal, 01°15'S, 76°55'W, 320 m, 7 Jan 1989 (pist. fl), W Palacios et al. 3433 (BRIT, MO, QCNE). Canton astaza, “Ramirez” Oil Well, 20 km S of Curaray, 01°32'S, 76°51'W, 300 m, 21-28 Feb 1990 (fr), V. Zak et al. 5292 (BRIT, MO, QCNE). Pichincha: Cooperativa Santa Marta No. 2, along Rio Verde, 2 km SW of Sto. Domingo de Los Colorados, 530 m, 5 Feb 1979 (pist. fl), C. Dodson 7418 (MO, QCA, SEL). Sucumbios: Canton Gonzalo Pizarro, Campo Bermejo 6N, 30 km NW of Lago Agrio, 00°14'N, 77°13'W, 1,050 m, 23 Mar 1990 (fr), C. Cerin et al. 9120 (BRIT, MO, QCNE); Parroquia Reventador, PreCooperativa Garcia Moreno 3rd line N of road, near Rio Due, 00°03'N, 77°35'W, 1,800 m, 23 May 1990 (fr), C. Cerdn et al. 9841 (BRIT, MO, QCNE). PERU. Cuzco: Prov. Quispicanchis, Camanti, Maniri along Rio Maniri, trail to Quebrada Garrote, 13°71'S, 70°45'W, 720 m, 8 Sep 1990 (stam. M. Timand 924 (AMAZ, BRIT, CUZ, F, MO, USM), hills around Rio Araza between n de Azucar and Quince Mil Airport, 292 km from Cusco, 13°13'S, 70°45'W, 643 m, ir Aug 1 a (stam. bud), P. Nuwfez 14081 (CUZ, HUT, MO, USM). Loreto: Prov. Maynas; forest eos Caseria Mishana, 150 m, 13 Nov 1977 (pist. fl), A. Gentry et al. 20673 (AMAZ, MO, USM); Rio Gueppi, tributary of Rfo Putumayo, Peru/Ecuador border, 200 m, 14 a 1978 (fr), A. Gentry et al. 21842 (AMAZ, MO, USM), Drto. Fernando Lores, Caserio Constancia (Quebrada Tamshiyacu), 04°08'S, 72°55'W, 120-130 m, 7 May 1991 (ft), C. Grdndez et al. 2472 (AMAZ, BRIT, MO, USM); Drto. Pevas; Caserio Colonia (Quebrada Sumé6n), tributary of Rio Yahuasyacu, trail to Rio Putumayo, 03°20'S, 71°50'W, 120-130 m, 30 Mar 1991 (fr), Grdndez et al. 2334 (AMAZ, BRIT, MO, USM); San Anto- nio, Rio Itaya, 04°10'S, 73°20'W, 13 Dec 1982 (stam. fl), R. Vasquez & N. Jaramillo 3558 (AMAZ, MO, USM); Prov. Requena, Jenaro Herrera, Rio Ucayal, ie ae 22 Feb 1987 (fr), A. Gentry et al. 56361A (AMAZ, MO, USM); Sapuena, Bagazan-Rio Ucayali, 04°45'S, 73°38'W, 130 m, 14 Jan 1987 (pist. A), R. Vasquez & N. Jaramillo 3789 (AMAZ, MO, USM). BRAZIL. Amazonas: Deto. Agropecuario, Reserva 1051 ( ‘km 41”) of the WWE/INPA MCS Project, 02°25'31" 4 59°45'5S0" W, 50-125 m, 20 Nov 1988 (stam. fl), B. Boom et al. 8568 (BRIT, INPA, NY), 02°24'26"- 02°25'31"S, 59°43'40"- 59°45'50" W, 6 Dec 1988 (piste. fl), B. Boom et al. 8764 (BRIT, INPA, NY), (stam. fl), S. Mori et al. 20194 (BRIT, INPA, NY); Municipality of Sao Paulo de Olivenga, near Palmares, 11 Sep— 26 Oct 1936 (stam. fl), B. A. Krukoff 8324 (MO, NY, US); Municipality Oriximina, BR 163 to 7 km N of Cachoeira Porteira, near ES-7, 01°02'S, 57°02'W, 20 Aug 1986 (stam. fl), C. A. Cid et al. 7929 (BRIT, MG, INPA, NY). Clusia penduliflora is most closely related to C. spathulaefolia, but 1s readily separated by its hemiepiphytic habit, leaves with long petioles, cuneate bases, obscure secondary venation and conspicuous latex canals, and the relatively small, pendent inflorescences. This species is often confused with Clusia martiana, but may be readily separated in flower by its subsessile anthers. Vegetatively, Clusia penduliflora generally has larger leaves and an 5322 Sipa 16(3) 1995 erect habit, but this character overlaps between the two species consider- ably. In fruit, C. penduliflora has an oblongoid fruit with elliptic stigmas, while C. martiana has ovoid fruits with oblongoid stigmas. 3. Clusia spathulaefolia Engler, Fl. Bras. 12(1):412. 1888. Type. BRAZIL. Amazonas: Alto Amazonas, along streams near Panuré, along Rio Uaupes, Dec 1852 (stam. fl), R. Spruce 2782 (HoLotypr: B-destroyed, F Neg. 32279!; tecrorype (here designated): P!). Free-standing trees to 20 m tall, 35 cm DBH; branchlets 4-angled with narrow ridges approximately 2 mm tall, 7-10 mm diam., bark gray to grayish-brown, smooth, not transversely checked or exfoliating, glabrous; latex white, copious. Leaves decussate; leaf blades stiffly coriaceous to carti- laginous, oblanceolate to obovate, rarely elliptic, (12—)19—28 cm long, (3.5) 10(-16.8) cm wide, apex obtuse to broadly rounded, base attenuate and decurrent on the petiole, nitid above, dull below, latex canals inconspicu- ous, midrib raised but canaliculate above, prominently raised below, sec- ondary veins numerous, raised above, slightly raised below, extending toa fine submarginal collecting vein, the margin revolute, entire, glabrous; petioles broadly marginate, 1-1.5(—2.2) cm long, glabrous. Inflorescence terminal, erect, pyramidal paniculate, 3—8(—10) cm long, S—7(—10) cm wide, the branches cymose, the flowers solitary or 3 per cyme; peduncle 1—2.5(— 5) cm long, subtended by normal leaves; secondary bracts carnose, widely ovate, 3-7 mm long, 4—6 mm wide, apex obtuse to rounded, carinate, the margin scarious, entire, glabrous. Staminate inflorescence with tetragonal pedicels, within each cyme, the pedicels of the lateral flowers 0.6—-1 mm long, those of the terminal flower 3—5 mm long, glabrous; bracteoles 4, the outer 2 carnose, widely ovate, 3-7 mm long, 4-6 mm wide, apex obtuse to rounded, carinate to crested medially, the margin scarious, hyaline, entire, glabrous, the inner 2 coriaceous, orbicular to oblate, 4—4.5 mm long, 3— lat carinate, the margins scari- — 3.5 mm wide, apex broadly rounded, somew ous, hyaline, entire, glabrous. Staminate flowers white; sepals 7(—8), the outer sepals 2, opposite, chartaceous, brittle, orbicular, 5-7 mm long and wide, apex broadly rounded, cucullate, the margin entire; the inner ones 5, contorted, membranaceous, brittle, very widely ovate, 4-6 mm long, 3-5 mm wide, largely opaque, apex very broadly rounded to subtruncate, trans- lucent latex canals dense, cucullate, the margin hyaline, entire, scarious; petals 5, thickly carnose, obovate-spathulate to oblong, S—7 mm long, 2- 2.5 mm wide, apex truncate, venation and latex canals prominent, medi- ally thickened, the margin undulating, entire, translucent, glabrous; androphore flat or slightly concave, quadrangular, the stamens numerous, free, 2.5—3.3 mm long, the filaments prismatic, 0.5—0.8 mm long, the PIPOLY AND GRafF, Synopsis of Clusia 523 anthers linear, 2—2.5 mm long, 0.5 mm wide, apex truncate, base truncate, dehiscent by longitudinal slits, the connective prismatic, darkened; pistillode obsolete. Pistillate inflorescence as in staminate but with pedicels 2—3 mm long; outer bracteoles suborbicular, 3—3.5 mm long, 3-4 mm wide, the inner 2 4—5.5 mm long, 5-6 mm wide. Pistillate lowers white, as in the staminate but sepals suborbicular, me mm long, 3.5—5 mm wide; petals obovate-spathulate, 5-7 mm long, 2.5—-3 mm wide, apex rounded, abruptly constricted toward base; staminodes 20, to 2 mm long, the fila- ment flat, 0.3-0.5 mm long, anther linear, 1.3-1.5 mm long, 0.3—0.5 mm wide, not bearing pollen; pistil subglobose, 5-8 mm long, and in diam., narrowing toward the apex; carpels 5, = ovules numerous, = styles ab- sent; stigmas sessile, cuneiform, convex, 2—3.5 mm long, 1.5—2 mm wide. Fruit pale green, tinged with pink in ae eee grooves, depressed- globose, 1.5—1.8(—2) cm long, 2.2—2.5(—2.8) cm diam. Common names.—COLOMBIA: “Matapalo” (Spanish); “Detziva” (Mirafia language); “Atbo-caja-jubaga-bacoco-moho” (Mui language), “Jubagagaimio” (Mui language- elders); “Jipeo” (Hui language). VENEZU- ELA: “Cupf banero hoja gruesa,” “Cupi”. Distribution. —Endemic to the Guayana Crystalline Shield of Venezuela, Brazil, and Colombia, with a disjunct population in Peru, at 50-200 (—1,650) m elevation. Ecology and conservation status. —Clusia spathulaefolia is a medium-sized, free-standing tree growing in gallery scrub forests (“Bana”) and savannas (“campinarana” or “amazonian caatinga”) along black water rivers, on deep white sands. These formations are near rivers but do not food (Macedo & Prance 1978; Prance 1979; Prance & Schubart 1978). Specimens examined. COLOMBIA. Caquetd: Araracuara, N side of the central portion of the airstrip; 27 Apr 1988 (ster.), M. Sanchez et al. 131 (COAH, MO), Near the airport, 00°37'S, 72°24'W, 17 Oct 1990 (A bud), J. Duivenvoorden & A. Cleef 258 (BRIT, COAH, COL, U), 15 Nov 1991 (ster.), J. Duivenvoorden 980 (BRIT, COAH, COL, U). Amazonas: Rio Caquetd, right bank, 1.8 km above the mouth, Quinche, 2.7 km; 11 May 1988 (fr), M. Sanchez et al. 341 (COAH, MO); Rio Caqueta, left bank, 1 km above the W end of Marifame Island, 3.5 km; 13 Jun 1988 (fr), M. Sanchez et al. 744 (COAH, MO), Jun ean (ster.), L. Urrego 578 (COAH, MEDEL, MO), Jul 1989 (ster.), L. Urrego 721 (CO MEDEL, MO); Cuenca Rio eee 00°50'S, 71°50'W, 22 Nov 1991 (stam. fl a. 7 Sef etal. 13061 (BRIT. AH, COL, U); Quebrada el peas 8 km N of mouth, 0 Sep 1988 (immature M. Sanchez et al. 1347 (COAH, MO). Vaupés: Ama- zon Basin Rio Negro, San Felipe de Vaupés, 200 m, 13-25 Nov 1952 (ster.), H. Humbert 27511 (COL, P). VENEZUELA. Amaconess Depto. a ae SE bank of middle portion Cafio Yagua at Cucurital Yagua, 03°36'N, 66°34'W, 120 m, 8 May 1979 (pist. fl, fr), G. Davidse et al. 17404 (MO, US, VEN), 2 ow Salto ee on tributary of headwa- ters of Rio [guapo, 03°35! ~ 65°23'W, 1500-1650 m, 12 Mar 1985 (fr), R. Liesner 18601 (MO, VEN); third slope, on Rfo Atacavi, 03°12'N, 67°24'W, 100 m, Nov 1989 (fr), /. 524 Sipa 16(3) 1995 Velazco 1126 (BRIT, PORT, VEN): Depto. Atabapo/Casiquiare, El Almid6én, Rio Atacavi, slope 2, 03°04'N, 67°06'W/, 80 m, Nov 1989 (stam. fl bud), J. Velazco 870 (BRIT, PORT, VEN). Depto. Atures, Second camp 110 km above Rio Guayapo, 04°18'N, 67°28'W, 120 m, May 1989 (fr), E. Foldats & J. Velazco 9318 (BRIT, NY, MO, PORT): Depto. Casiquiare, Cano San Miguel, sector “Las Tinajas,” 02°39'N, 66°45'W, 160 m, 25 Apr 1991 (fr), G. Aymard 9232 (BRIT, PORT, VEN); Depto. Rio Negro, Bona, 10 km NE of San Carlos de Rio Negro, ca. 20 km S of confluence of Rio Negro & Brazo Casiquiare, 01°56'N, 67°03'W, 119 m, 21 Dec 1978 (pist. fl), H. Clark & P. Maquirino 6907 (BRIT, MO, VEN), 6 May 1979 (pisc. fl, fr), H. Clark 7172 (MO, VEN), 2 Mar 1979 (pist. fl, fr), H. Clark 7053 (MO, VEN), 7 Apr 1979 (pist. fl, fr), R. Liesner 6302 (MO, VEN), 16 Apr 1979 (pisce. Al, fr), R. Liesner 6692 (MO, VEN), 25 Jan 1985 (stam. fl), B. Boow et al. 5361 (GH, NY, US, VEN); Alto Rio Negro, Yapacana Savanna, 16 Mar 1953 (pisc. fl), B. Maguire & J. Wardack 34547 (MO, NY, US, VEN), (stam. fl), B. Maguire & J. Wurdack 34574 (F, MO, NY, US, VEN), B. Maguire & J. Wurdack 34575 (F, MO, NY, US, VEN); Sierra de Unturdn, 01°33'N, 65°12'W, 1,150 m, 3 Feb 1989 (stam. fl), A. Henderson 941 (BRIT, NY, VEN); Cerro de la Neblina base camp, Rio Mawarinuma, 00°50'N, 66°10'W, 140 m, 8 Feb 1995 (pist. fl), B. Boom & A. Weitzman 5698 (BRIT, F, GH, MO, NY, US, VEN). PERU. Loreto: Campamento Petrolero San Jacinto, Rio Tigre, 02°15'S, 75°50'W, 200 m, 16 Sep 1979 (pist. fl, fr), C. Diaz & N. Jaramillo 1463 (AMAZ, F, MO, USM). BRAZIL. Amazonas: Municipio Pres. Figueredo, Rioi Uatuma, Igarapé Caticu, right margin, 01°- 02°S, 59° 60°W, 29 Mar 1986 (fr), C. Cid et al. 7014 (BRIT, INPA, MG, NY); Km 130, Manaus- Caracarai Rd., | Dec 1974 (stam. fl), A. Gentry 12960 (INPA, MO); Municipio Sao Gabriel do Cachoeira, Ilha Tamandua, at mouth of Rio Uaupés (Rio Catari), 2 km above Cumunidade Jowavira, near Ilha de Flores, .30 Dec 1989 (fr), B. Nelson et al. 1719 (BRIT, MG, INPA, Y); Rio Negro, near mouth of Rio Xié, opposite S40 Marcelino, DOP 9S NO 7"15 We 2 1 Oct 1987 (fr), D. Stevenson et al.779 (BRIT, INPA, MG, NY). Para: Municipio de Itaituba, Serra do Cachimbo, Base Aérea, 5 km behind the Cachimbo Airport, along bank of Rio Formiga, 27 Apr 1983 (fr), M. N. Silva et al. 136 (BRIT, INPA, MG, NY). — Clusia spathulaefolia is most closely related to C. aymardii, the two hav- ing free-standing, terrestrial habic, stiffly coriaceous to cartilaginous and nitid leaf blades with rounded to truncate apices, the subsessile to sessile flowers, cuneiform stigmas and strongly ribbed fruits in common. Clusia spathulaefolia, along with Clusia pusilla Steyermark, and several species of Myrsinaceae, most notably Cybianthus spicatus (H.B.K.) Agostini, and C. Julvopulverutentus (Mez) Agostini subsp. magnoliifolius (Mez) Pipoly, all grown on Serra do Cachimbo, suggesting that it may be a tepui satellite not pre- viously listed in Maguire’s (1979) treatise on the phytogeography of the Guayana Highland. 3. Clusia aymardii Pipoly, sp. nov. (Fig. 5) Ob habitu arboreum, folia rigido-coriacea vel cartilaginosa, supra nitida subter pallidaque, ad apicem late rocundata vel truncata, flores sessiles vel subsessiles, pedicelos usque | mm longos necnon stigmatibus cuneiformibus denique fructus praeclare costatus C spathulaefoliae valde arcte affinis, sed ab ea laminis oblongis vel ellipticis (nec oblanceolatis vel obovatis), ad basem obtusis vel late rotundatis (nec attenuatis), secus marginis planis (nec revolutis), petiolis subteretibus (non marginatis) 3—5 (nec 1.5—2.2) cm longis, et ramulis 10-22 (non 7—10) mm in diametris facile cognoscitur, — Pipoty AND GrarFr, Synopsis of Clusia A ho WA mie & —~St Efi = Fic. 5. Clusia aymardii Pipoly. A. Habit, showing cartilaginous leaves with broadly rounded bases and subterete petioles, erect inflorescence and sessile flowers. B. Fruit, showing cos- tae. C. Staminate flower at anthesis, showing numerous stamens and somewhat clawed petals with hyaline, scarious margins. D. Staminate flower bud, showing sepals (on left), and petals (above and on right). E. Androecium, showing emarginace anthers. A, C-E, drawn from type; B, drawn from G. Aymard 5918. All unmarked scale ars represent 5 mm. 526 Stipa 16(3) 1995 et Glabrous, free-standing tree to 3 —15(-22) mm diam., longitudinally ridged when dry. Leaves cartilagi- nous, the blades oblong elliptic, (10—)12—18 cm long, (S—)7-11 cm wide, apex very widely rounded to truncate, base obtuse to widely rounded and slightly inequilateral, nitid above, pallid below, costa raised above and be- low, secondary veins numerous, inconspicuous, latex canals, narrow, vein- like, margin entire, opaque; petiole subterete, (3—)3.5—4(—5) cm long, lon- gitudinally ridged. Inflorescence erect, glomerate, (4—)8—12-flowered, peduncle essentially terete, 1-2 cm long; floral bracts 4, decussate, carti- laginous, oblate to depressed-ovate, (5—)8—10 mm long, 10-13 mm wide, apex obtuse, base subauriculate, carinate medially, margin scarious; pedicels m tall; latex white, branchlets terete, obsolete. Staminate flower with latex canals not obvious in any perianth ae outer ee 2, Opposite, coriaceous, depressed-ovate, 7.7—8.4 mm long, 9.8-11.4 mm, wide, apex truncate, margin scarious, entire, inner sepals 5, as coriaceous, obovate-spathulate, 9-12 mm long, 7.7- 9.8 mm wide, apex broadly rounded, margin scarious, entire; petals 5, spi- rally arranged, chartaceous, 13.5—17.2 mm long, white, somewhat clawed: limb widely obovate, 7-8.4 mm long, 9.1—10.8 mm wide, apex truncate, base cuneate, with numerous longitudinal linear resin canals, the margin entire, undulate, hyaline, scarious, the claw 6.5—9.8 mm long, 3.6-4.2 mm wide; stamens numerous, 2.5—2.9 mm long, free; filaments prismatic, 1.5-1.8 mm long; anthers muticous, developmentally adnate to abaxial stamen surface, 0.8—1.1 mm long, distally highly recurved, emarginate at apex and base. Pistillode absent. Pistillate flower (in bud) like staminate but .sepals 7, the outer 2, decussate, oblate, 5 mm long, 8-10 mm wide, the inner 5, spiral, same shape, size and texture as in staminate flower; stami- nodes chartaceous, resiniferous, numerous, in one whorl, at times several con- nate in phalanges, anantherous, 6-8 mm long, 2-3 mm wide, apex Muticous; ae oblongoid, 5 mm long, 4 mm diam., styles absent, stigmas cuneiform, a. 1.5 mm long, | mm wide. Fruit a 7-carpelled capsule, oblongoid, crimson, 3 cm long, 1.5 cm diam., strongly ribbed, the stigmas sessile, cuneiform, flat, persistent. _— Type. VENEZUELA. Botivar: summit of Cerro Guaiquinima, falls at Rfo Szczerbanari (Rio Carap6), 1-2 km upriver from Szczerbanari Falls, 05°44'04" N, 63°41'08" W, central part of mountain, 750 m, 20-25 Jan 1977 (stam. fl), - Steyermark GG. C.K. GE. Dunsterville 113157 (HoLotyPE: VEN!; tsorypes: F!, MO!, NY, PARATYPES. VENEZUELA. Botivar: Det mee a Guaiquinima, Camp 2, 05° N 63°00'W, 1,200 m, 7 Feb 1990 (stam. fl), : Boom 9430 (BRIT, NY, VEN); central base of Guaiquinima-Tepuf, scrub forests along Quebrada Martinez, 85 km S of La Paragua, 06°04'N, 63°22'W, 500 m, 11 May 1987 (fr), G. Aymard ave (PORT); Mpio. Raul Leon, Cerro Guaiquinima, ca. 15 km NE of Rio Paragua, ( 3'N, 63°35'W, 800 m, Nov 1988 (pist. fl bud), Y. Ferndndez 275 (BRIT, PORT, a PIpOLY AND GrarF, Synopsis of Clusia S27 Distribution —Endemic to Cerro Guaiquinima, state of Bolivar, Venezu- ela, at 750—1,200 m elevation. Ecology and conservation status.—Clusia aymardii occurs only in scrub for- ests on the slopes and summit of Cerro Guaiquinima, and as an endemic, may be considered threatened. Clusia aymardii is most closely related to C. spathulaefolia, but is easily recognized by the oblong to elliptic leaf blades with obtuse to broadly rounded bases and flat margins, the subterete, longer petioles, and thicker branchlets. While Cerro Guaiquinima is known to be poor in terms of num- bers of endemic species (Pipoly 1992), C. aymardii is a remarkably distinct species. ACKNOWLEDGMENTS Research for this treatment was greatly facilitated by the Missouri Bo- tanical Garden, which provided loans and numerous gifts for determina- tion. Fieldwork conducted during the senior author's tenure at MO, was supported by generous grants for documentation of Amazonian Phytodiversity from the John D. and Catherine T. MacArthur Foundation, and for botanical exploration in Peru from the Andrew W. Mellon Founda- tion. Rodolfo Vasquez, Agustin Rudas, César Grandez, and Nestor Jaramillo accompanied the senior author in the field, provided logistical support, and cordial hospitality. The hospitality of personnel at AMAZ, COAH, COL, HUA, JAUM and USM greatly facilitated his visits to those institu- tions. We thank Jon Ricketson, Catherine Mayo and Roy Cummings (MO), and Lindsay Woodruff (BRIT) for technical assistance. Linda Ellis provided the excellent line illustrations of new taxa. P. Mick Richardson (MO) fa- cilitated the participation of the junior author in the project by arranging a one term study rotation in the senior author's laboratory. We also thank Peter Stevens (GH) and Carol Todzia (TEX) for reviewing the manuscript and making helpful suggestions. REFERENCES BENTHAM, G. 1862. Guttiferae. In: G. ae and ay: D. Hooker, eds. Genera plantarum. Lo Reeve & Co. London. Vol. 1, pp 112 nee A. 1888. nT In: C. Martius, a aoe Brasiliensis Vol., 12(1), pp. 382- 474 893. Clusia. In: Die Natiirlichen Pflanzenfamilien 3(6). Verlag von Wilhelm Pasian, ee r, Hammet, B. 1986. New species of Clusiaceae from ae America with notes on Clusia and synonymy in the tribe C/uszeae. Selbyana 9:112—120. Macepo, M. and G. K. 1978. Notes on the ve pecan’ of Amazonia II. The dispersal of plants in oo white sand campinas: the camp functional islands. Brittonia 30:203-215. 528 Sipa 16(3) 1995 Macuirr, B. 1976. Apomixis in the genus Clusia (Clusiaceae). - A preliminary report. Taxon 25:241-244. 1979. Guayana, region of the Roraima Sandstone Formation. In: K. Larsen and L, Poi Nielsen, eds. Tropical botany. Academic Press. New York. Pp. 223-238. Pipoty, 1992. The genus Cybranthus ee us ihe (Myrsinaceae) in Guayana. Ann. Missouri os Gard. 79:908-957 PRANCE, G. 1979. Notes on the vegetation of Amazonia III. The terminology of Amazo- nian oe types subject to inundation. Brittonia 31:26—38. dH. Scuusarr. 1978. Notes on the vegetation of Amazonia I. A prelimi- nary note on the origin of the white sand campinas of the lower Rio Negro. Brittonia 30:60-63. VesQue, J. 1893. Gutciferae. In: A. De Candolle, ed. Mongraphie Phanerogarum Vol. 8 pp. 1-669 BOOK REVIEW GRAHAM, Linpa E. 1993. Origin of Land Plants. (ISBN 0-471-61527-7, hbk.) John Wiley and Sons, New York. $89.00. 287 pp. 189 figures. Linda Graham’s entire career had been devoted to research on the origin of embryo- phyces from algal progenitors. In the Origin of Land Plants, she summarizes and synthesizes not only her own suo renee research, but the history of the debate, the pertinent re- search by other workers, and further unresolved issues. Although anyone would be tempted to bias such a synthesis in favor of one's own favorite hypothesis, she had provided a bal- anced, almost detached, review of the questions of embryophyte origin The first three chapters provide background to the problem of embryophyte origin, including early Paleozoic environments, current research methods being applied, and the consensus that the Charophyceae are the closest algal relatives of land plants. The next examine the Charop A= vyceae in detail and — them to land plants. Three more chapters follow with in- denth discussions on the evolution of significant features such as the cytoskeleton, cytokinesis, origin of ae. sporophytes, nutrient transfer regions, control or meiosis, phytochrome systems, and phenolic biosynthetic pathways. The final chapter recapitulates the ideas presented in the previous chapters. The book is clearly aimed at advanced undergrads, graduate students and professional bocanists. In the final chapter Dr. Graham reveals her hopes that che book (or least chat chapter) will be used by authors of introductory texts to update the often mistaken view students have of plant evolution. Certainly I found the book, especially chapters 3, 7, an 10, to be of great help in preparing lectures for an introductory botanical survey course at Southern Methodist University. pace, The book ts a welcomed addition to the modern botanical literature, and I recommend it for every botany professor and college library.—Roger W. Sanders. SUBTRIBAL CLASSIFICATION OF THE NEW WORLD ERAGROSTIDEAE (POACEAE: CHLORIDOIDEAE) PAUL M. PETERSON Department of Botany National Museum of Natural History Smithsonian Institution Washington, D.C. 20560, U.S.A. ROBERT D. WEBSTER Systematic Botany and Mycology Laboratory Agriculture Research Service Bldg. 003, Rm 235, BARC-West Beltsville, MD 20705, U.S.A. JESUS VALDES-REYNA Departamento de Botdnica Universidad Autonoma Agraria “Antonio Narro” Buenavista, Saltillo Coahuila 25315, MEXICO ABSTRACT A subtribal classification system for 38 genera of Eragrostideae that occur in the New World is presented. Seven recognized subtribes include: Eleusininae, car pien A Reel a Muhlenbergiinae, Munroinae, Sporobolinae, and Uniolinae. A com parative discussion outlining the recent classification history 1s included. Diagnostic fully comparative descriptions are given for each subtribe and a new subtribe, Munroinae, is validated. RESUMEN OQ. Se — un sistema de clasificaci6n tribal de 38 géneros de Eragrostideae que viven e Mundo. Las siete subtribus reconocidas son: Eleusininae, Eragrostidinae, Medion Muhlenbergiinae, Munroinae, Sporobolinae, y Uniolinae. Se incluye una discusié6n comparativa que esboza la historia de la clasificaci6n reciente. Se ofrece una descripcién diagnéstica comparativa completa de cada subtribu y se valida una nueva subtribu, Munroinae. INTRODUCTION The cribe Eragrostideae in the subfamily Chloridoideae includes approxi- mately 80 genera and 1000 species or about one-tenth of all grasses. These Sipa 16(3); 529 — 544. 1995 530 Stipa 16(3) 1995 taxa are primarily distributed in the subtropical to tropical regions in arid climates with centers of distribution in the savannahs of southern Africa and north central Mexico. In the New World or Western Hemisphere, we recognize 38 genera and about 427 species with the highest concentration in north central Mexico and southwestern United States. Within the Eragrostideae there is considerable variation in morphology, anatomy, and cytology (Peterson 1988, 1989: Peterson & Annable 1990, 1991, 1992; Peterson et al. 1989, 1993, 1996; Peterson & Herrera 1995; Valdes-Reyna & Hatch 1991). It is difficult, if not impossible to select diagnostic characteristics that exclusively delimit the Eragrostideae from other tribes in the Chloridoideae. However, the tribe commonly has pan- iculate inflorescences (occasionally racemose), laterally compressed or ter- ete (rarely dorsiventral) spikelets, lemmas 3-nerved (occasionally 1 or more than 3-nerved), and disarticulation typically above the glumes (Peterson et al. 1996). In Flora Capensis, Stapf (1898) first used the tribal name Eragrosteae (= Eragrostideae) where he recognized four genera: Desmostachya (Hook. f) Stapf, Diplachne P. Beauv. (= Leptochloa), Eragrostis, and Pogonarthria Stapf. Pilger (195G) increased the number of taxa in the Eragrostideae to include 53 genera in six subtribes. Five out of the six Eragrostideae subcribes (Eragrostinae, Lycurinae, Muhlenbergiinae, Scleropogoninae, and Sporobolinae) used by Pilger contain indigenous New World genera. Relationships among these genera have been problematic and agrostologists have attempted co sort out a classification that reflects com- mon ancestry. The primary objective of the present study is to present a comprehensive, fully comparative classification system for the 38 native and introduced New World genera of the Eragrostideae. In addition to presenting our new classification, we include a review of the major classifi- cations used in the lasc half century. A proposed classification recognizing seven subtribes is given to elucidate probable phylogentic relationships among the New World genera. METHODS A discussion is presented for those classification treatments that were considered important for the New World genera of the Eragrostideae and these include: Hitchcock and Chase (1951), Pilger (1954, 1956), Gould and Shaw (1983), Clayton and Renvoize (1986), and Watson and Dallwitz (1992). An outline of these classifications is presented in Table 1. Associ- ated with Table | is a discussion on the characters used by the authors to delimit the categories. This information was extracted from their keys and descriptions. Within each classification outline we primarily included only those categories pertinent to the New World genera. Recognition of — PETERSON ET AL., Eragrostideae of the New World 33:1 subtribes was based on a synthesis of data from relevant systematic disci- plines and includes morphological, anatomical, cytological, ecological, and molecular characteristics. In addition, this classification reflects our knowl- edge gained from experience studying these taxa in the laboratory and field. The DELTA system (Dallwitz 1980; Dallwitz et al. 1993) was used in the selection of characters and production of diagnostic comparative de- scriptions of the seven recognized subtribes. Initially, a list of 90 characters was constructed to account for variation at the generic level for the Eragrostideae. A full range and adequate number of herbarium specimens were analyzed and data were recorded for the 38 recognized New World genera. A comprehensive taxonomic account for these genera is forthcom- ing (Peterson et al. 1996). INTKEY was used to combine data for the genera of a subtribe. For example, the DELTA format data of the Monanthochloinae was produced through INTKEY by combining the data for the members, specifically AJ//olepis, Distichlis, Jouvea, Monanthochloé, Reederochloa, and Swallenia. This combined DELTA format data was con- verted to the natural language description presented here. Selection of the 21 characters presented in these descriptions was based on their diagnostic value and on the value of the character in distinguishing among the subtribes. Additional methods concerning the application of the DELTA programs can be found in Webster (1988, 1992a, 1992b). HISTORICAL CLASSIFICATION It is important to present a discussion of the relevant classifications of this group. Included here in chronological order are discussions of the treat- ments by Hitchcock and Chase (1951), Pilger (1954, 1965), Gould and Shaw (1983), Clayton and Renvoize (1986), and Watson and Dallwitz (1992). Important characteristics used to distinguish among their catagories are outlined together with the placement of our 38 New World genera. The Eragrostideae are treated by Hitchcock and Chase (1951) in three tribes: Agrostideae, Chlorideae, and Festuceae (Table 1). Their Agrostideae (Blepharonenron, Calamovilfa, Crypsis, Lycurus, Mublenbergia, Sporobolus) is based on the presence of perfect, 1-flowered spikelets, and paniculate inflo- rescences. Although the number of florets per spikelet is a very diagnostic generic characteristic, its use to determine tribal and sufamily relation- ships has proven fallacious. Therefore, members of their Agrostideae have been subsequently re-aligned to other tribes in the Poaceae. Their Chlorideae included six genera (Dactyloctenium, Eleusine, Leptochloa, Munroa, Trichoneura, Tripogon) are characterized as having 1-to several-flowered spikelets in two rows along one side of a continuous rachis. The inflorescence consists of spikes or spike-like racemes, either solitary, digitate, or racemosely arranged along the main axis. The Festuceae as interpreted by Hitchcock and Chase 532 Sipa 16(3) 1995 (1951) includes the following genera: Blepharidachne, Distichlis, Eragrostis, Eragrostis (in part, Neeragrostis), Monanthochloé, Redfieldia, Scleropogon, Ectosperma (=Swatlenia), Neyraudia, Tridens (in part, Dasyochloa, Erioneuron), Triplasis, Uniola, Vaseyochloa. Nine genera (A/lolepis, Bealia, Chaboissaea, Gouinia, Jouvea, Neesiochloa, Peretlema, Sohnsia, Steirachne, Reederochloa, Tetrachne) were not treated by Hitchcock and Chase (1951) since they do not occur in the United States. A progressive classification of the grass family was given by Pilger (1954, 1956) who placed our 38 genera into five different tribes in two different subfamilies (Table 1). It is worthy to note that he states exceptions to most if not all characters used to define his categories. The following characteris- tics were used to differentiate the Eragrosteae from the other five tribes in subfamily Eragrostoideae: spikelets in loose or contracted panicles (often compounded in racemes or spikes); branches spirally inserted; glumes usu- ally shorter than the lemma; upper floret in many-flowered spikelets often sterile; lemma usually | or 3-nerved, membranous to leathery; lemma apex entire, crenate, toothed or lobed, if aristate, the awn arising from the mid- nerve; lemma nerves often hairy; stamens 2—3; fruits usually loosely sur- rounded by the palea (often becoming free); pericarp thin (often loosening and leaving the seeds naked); hilum small and basal; base chromosome number 8 or 10. Pilger’s Eragrosteae included six subtribes (Eragrostinae, Scleropogoninae, Lycurinae, Garnotiinae, Sporobolinae, and Muhlenbergiinae). The Eragrostinae (including Blepharidachne, Dactyloctenium, Eleusine, Eragrostis, Leptochloa, Munroa, Neesiochloa, Neyraudia, Redfieldia, Stetrachne, Trichoneura, Tridens, Triplasis, Tripogon) was separated by multi-Alowered spikelets with perfect forets. The Scleropogoninae with only Scleropogon was characterized by unisexual spikelets with the female spikelets long aristate. Pilger’s Lycurinae (including Lycwrs) was characterized by glumes finely aristate and single-flowered spikelets. The Sporobolinae which includes Blepharoneuron, Crypsis, and Sporobolus, was differentiated by single-flowered spikelets, thin glumes, aristate lemmas, and rounded fruits. The Muhlenbergiinae (including Mhlenbergia) was separated by glumes shorter than the lemma, single-flowered spikelets, and narrow cylindrical cary- opses. Pilger’s final tribe chat includes New World genera, the Jouveeae (Jouvea), was characterized by dimorphic, unisexual spikelets, with or with- out a rachilla extension (in female spikelets), 1-Aowered female spikelets, rudimentary glumes, inflorescence a spike with few spikelets, and a thick, grooved rachis. Pilger (1954) placed eight of our genera in the subfamily Festucoideae in three tribes (Festuceae, Aveneae, Arundineae, see Table 1). The Festuceae was separated by paniculate inflorescences, spikelets 1 to many-flowered, PETERSON ET AL., Eragrostideae of the New World 533 glumes shorter than the florets, and lemmas 3 to 5-nerved. His treatment included six of our 38 genera in three subtribes: Monanthochloinae, Festucinae, and Melicinae. The Monanthochloinae (including Monanthochloé) was characterized by spikelets without “outer” glumes, unisexual spikelets (dioecious or monecious), and short stiff leaves. The Festucinae (including Distichlis, Uniola, Tetrachne) was defined by spikelets with two outer glumes, lemmas 5 to several-nerved (rarely 3-nerved), and the lemma apex entire or bidentate. The Melicinae (including Ectosperma and Vaseyochloa) was distin- guished by a several to many-nerved lemma, a paniculate inflorescence and lemmas blunt and rounded on the back. Calamovilfa was placed in the Aveneae, a tribe characterized by disarticulation above the glumes, rela- tively long glumes, and mostly S-nerved lemmas. The Arundineae (in- cluding Gowinia) was described as mostly tall reed-like in habit, rachilla or lemma hairy, outer glume shorter or equalling the spikelet length, 3 or 5- nerved lemmas, and lemmas with a straight or twisted awn. New World genera not treated by Pilger included AJ//olepis, Reederochloa, and Sohnsia. The classification presented in Gould and Shaw (1983) represents a sig- nificant advance over that given by Hitchcock and Chase (1951). This ad- vancement was the result of Gould’s (1968, 1983) synthesis of the advance- ments from 1950 to 1983 (eg. Stebbins & Crampton 1961) in agrostology, specifically in the areas of anatomy, cytology, ecology, and morphology. Gould and Shaw (1983) divided the Chloridoideae into eight tribes, whereas Hitchcock and Chase only include four tribes with chloridoid genera (Agrostideae, Chlorideae, Festuceae, Zoysieae). Five tribes recognized by Gould and Shaw (1983) do not include genera that we place in che Eragrostideae. Each of these five possess diagnostic characters that separate them from the genera of the Eragrostideae. A key to the tribes is not given by Gould and Shaw; however, diagnostic charac- ters can be extracted from the brief descriptions associated with each tribe. The Pappophoreae is best distinguished by the presence of lemmas with 9 or mores nerves and awns. Secondary characters include a paniculate inflo- rescence with 3 to several florets per spikelet, ligule a line of hairs, and point of disarticulation close to the glumes. The Chlorideae is distinguished by an inflorescence of “a unilateral spike or of few to several unilateral spicate primary branches.” Secondary diagnostic characters are spikelets with a single fertile floret and one or more reduced florets, 3-nerved lem- mas, and ligule a line of hairs. The Orcuttieae is distinguished by the ab- sence of a ligule and the undifferentiated leaf blade and sheath. Additional characters include persistent spikelets, 5-nerved lemmas, and paniculate inflorescences with several-flowered spikelets. Gould and Shaw define the Aristideae by 1-flowered spikelets, disarticulation above the glumes, and lemmas with an awn column usually bearing 3 awns. In addition they give 534 Sipa 16(3) 1995 many anatomical characters, such as: dumbbell-shaped silica cells, two pa- renchyma sheaths, and elongate chlorenchyma cells. The last tribe lacking genera we place in the Eragrostideae, Zoysieae, is differentiated by a ligule as a line of hairs, inflorescence a contracted raceme of 1-flowered spikelets, and disarticulation at the base of the spikelet. As Table 1 indicates, Gould and Shaw (1983) place our New World Eragrostideae genera in three tribes (Aeluropodeae, Eragrosteae, Unioleae), with some not treated. The Unioleae, including only Uniola in the U. S., was distinguished by disarticulation below the glumes, ligule a line of hairs, large paniculate inflorescences, and several to many-flowered spike- lets chat have the lowermost and uppermost florets sterile. Additional ana- tomical characters include a leaf epidermis with club-shaped bicellular microhairs and square or saddle-shaped silica cells. Their concept of the Aeluropodeae included four of our New World genera (A//olepis, Distichlis, Monanthochloé, Swaltenia). Their tribe was differentiated by the presence of a stoloniferous or rhizomatous habit, usually occurring in saline habitats, short pungent leaf blades, and inflorescences a contracted panicle or raceme, usually unisexual. Excluding those genera not treated by Gould and Shaw, the remaining 22 New World genera were lumped into the Eragrosteae. This cri — de is poorly defined with numerous exceptions to all characters used to recognize the group. However, important characters include a 3- nerved lemma, |-several flowered spikelets, a paniculate inflorescence with some rebranching of the primary branches, and disarticulation mostly above the glumes. New World genera not treated by Gould and Shaw (1983) include Bealia, Chaboissaea, Gouinia, Jouvea, Neesiochloa, Ni eyraudia, Peretlema, Reederochloa, Sohnsia, Steirachne, and Tetrachne. All of which occur in regions south of the United States. Clayton and Renvoize (1986) separate the Chloridoideae into five tribes (Pappophoreae, Orcuttieae, Eragrostideae, Leptureae, and Cynodonteae) and nine subcribes; with five subtribes in the Eragrostideae and four subtribes in the Cynodonteae. Our New World genera of the Eragrostideae are all placed in their concept of the Eragrostideae. The Pappophoreae includes five genera and is distinguished by a many-nerved, lobed, and awned lemma, a narrow paniculate inflorescence, spikelets with several florets (commonly with the lower 2 or more florets bisexual), and disarticulation above the glumes. The Orcuttieae is delimited on leaves that are scarcely differenti- ated into a blade and sheath, absence of a ligule, viscid and aromatic leaf blades, lemmas with 1 3—15 nerves, and absence of lodicules. The Leptureae, consisting of Lepturus, is separated from other members of the subfamily by having a single cylindrical bilateral raceme with the spikelets alike, borne edgeways on and embedded in hollows in the fragile rachis. There do not appear to be any New World chloridoids that are morphologically similar on PETERSON ET AL., Eragrostideae of the New World 535 Tasie 1. Tribal classifications of the Chloridoideae. Hitchcock & Chase (1951) Subfamily Festucoideae 1. Agrostideae = Blepharonenron, Calamovilfa, Crypsis, Lycurus, Muhlenbergia, Sporobolus eae 3, Bambusea 4. Chlorideae = coe sae Lepimaina, Mania, cae uialis i eDiee 5. Festuceae = Blepharidachne Monanthochloé, Redfieldia Scleropogon, ao eee ones iain part Dictiocilid. Beioneuan yaa Uniola, Vaseyochlo 6. Hordeae 10. ats ted = Allolepis, Bealia, Chaboissaea, Gouinia, Jouvea, Neestochloa, Pereilema, Sobnsia, Steirachne, es ies Tetrachne Pilger (1954, 1956) Subfamily Eragrostoideae —_ . Eragrosceae a. Eragrostinae = Blepharidachne, Dactyloctenium, Eleusine, Eragrostis (including Neeragrostis), Leptochloa, Munroa, Neesiochloa, es idia, Redfieldia, Steirachne, Trichoneura, Tridens (includes Dasyochloa and Erionenron), Triplasis, Trrpogon Scleropogoninae = Scleropogon . Lycurinae = Lycurus, Peretlema Garnotiinae _ Sporobolinae = Blepharonenron, Crypsis, Eptcampes (= Muhlenbergia in part), Sporobolus f. Muhlenbergiinae = Mwhlenbergia (including Bealia and Chabotssaea) oon 2. Phaenospermeae 3. Chlorideae . Lappagineae 6. Jouveeae = Jouve oT Festucoideae estuceae a. Monanthochloinae = Monanthochloé b. Festucinae = Distichlis, Uniola, Tetrachne c. Melicinae = Ectosperma (= Swallenia), Vaseyochloa . Aveneae = Calamovilfa Ww N . Arundineae = Gowinia Not Treated = A//olepis, Reederochloa, Sohnsia Gould & Shaw (1983) 1. eae ae Seas a Co Crypsis, Dactyloctenium, Eleusine, isyochloa), Le loa, Lycurus, Mublenbergia, Munroa, Neeragrostss, ae Na sede Spools, ath neue, ee a " Thiplasis, Tripogon, Vaseyochloa 2. Aeluropodeae = A//olepis, Distichlis, Monanthochloé, Swallenia . Unioleae = i 536 Stipa 16(3) 1995 TABLE 1. (Continued) 4. Pappophoreae 5. Chlorideae 6. Orcuttieae Aristideae 8. Zoysieae Not Treated = Bealia, Chaboissaea, Goninia, Jouvea, Neesiochloa, Neyrandia, Peretlema, Reederochloa, Sobnsia, Steirachne, Tetrachne Clayton & Renvoize (1986) 1 Aca 2. Orcuttieae 3. eet dea b. Sc oe = Scleropogon a. Triodiinae b. Uniolinae = Tetrachne, Uniola c. Cee = Allolepis, Distichlis, Jouvea, Monanthochloé, Reederochloa, Swallenia d. Eleusininae = Blepharidachne, Blepharoneuron, Dactylocteninm, Elensine, Eragrostis (including Neeragrostis), Erioneuron (including Dasyochloa), Gouinia, Leptochloa, Munroa, Neesiochloa, Neyraudia, Redfieldia, Scleropogon, Sohnsia, Stetrachne, Trichonenra, Tridens, Triplasis, Tripogon, aseyochloa e. Sporobolinae = Calamovilfa, Crypsis, Lycurus, Muhlenbergia (including Bealia and Chaboissaea), ereilema, Sporobolus 4. Leptureae d. Zoysiinae Watson & Dallwitz (1992) 1. Triodie 2 Papronors 3. Orcuttie 4. Chloridea alee Bealia, is alse Sia pie ilfa, Chaboissaea, Crypsis, eet aye va, Distichlis, Elewsine, Eragrostis, Eri uron, Gouinia, Airs Leptochloa, Lycurus, Monentol ea. Munroa, Neeragro: ory tochloa, Neyraudia, Peretlema, Redfieldia Scleropogon, Sobnsia, Sporobolus es oo Swallenia, Tetrachne, cs sa neaes Tr aa, oti se Cuil: Vaseyoc a to Lepturus. Clayton and Renvoize (1986) characterize the Cynodonteae, which includes the subtribes Chloridinae and Boutelouinae, as having tough unilateral racemes (these single digitate or scattered along the axis), spike- lets with one fertile floret (with or without additional staminate or sterile florets), 1—3(—5)-nerved glumes, nerveless or 3-nerved lemmas, and fruit occasionally with a free pericarp. Some genera of the Chloridinae are morphologically similar to members of our Eragrostideae. Specific examples include: Eustachys, Chloris, and Cynodon with 2-many digitate racemes: PETERSON ET AL., Eragrostideae of the New World 37 Microchloa with a single raceme bearing spikelets on a semiterete rachis; and Schedonnardus with several racemes on a single axis bearing distant ap- pressed spikelets on a three-angled rachis. As indicated in Table 1, Clayton and Renvoize (1986) separate the Eragrostideae into five subtribes (Triodiinae, Uniolinae, Monanthochloinae, Eleusininae, and Sporobolinae). The Triodiinae includes only Australian genera and is best distinguished by highly xeromorphic pungent and needle- like leaf blades and disarticulation below each floret. The leaves have greatly reduced chlorenchyma tissue that occurs as narrow tracts along the sto- matal grooves. The Uniolinae included the New World genera Tetrachne and Uniola and is diagnosed by the presence of several to many fertile florets, disarticulation below the glumes, and glumes shorter than the strongly keeled lemmas. Clayton and Renvoize’s third subtribe the Monanthochloinae includes the New World genera A//olepis, Distichlis, Jouvea, Monanthochloé, Reederochloa, and Swallenia. Significant diagnostic characters of this group include are distichous often pungent leaf blades, spikelets with several to many perfect florets, disarticulation usually below each floret, and lemmas 5—13-nerved and thick textured. The Sporobolinae included six New World genera (Calamovilfa, Crypsis, Lycurus, Mublenbergia, Pereilema, Sporobolus). Their concept of Muhlenbergia included Bealia and Chabotssaea. Important characters for this subtribe included inflorescence a panicle, spikelets 1- flowered, the rachilla not extending above the floret, and lemma 1—3-nerved, usually membranous. Clayton and Renvoize’s (1986) Eleusininae includes 20 New World genera: Blepharidachne, Blepharoneuron, Dactyloctenium, Eleusine, Eragrostis (including Neeragrostis), Erionenron (including Dasyochloa), Gouinia, Leptochloa, Munroa, Neesiochloa, Neyraudia, Redfieldia, Scleropogon, Sobnsia, Steirachne, Trichoneura, Tridens, Triplasis, Tripogon, Vaseyochloa. Im- portant characters in this diverse subtribe include: spikelets with 2-many fertile florets, lemmas 3-nerved, and an inflorescence more often a raceme but sometimes a panicle. Watson and Dallwitz (1992) recognized four tribes within subfamily Chloridoideae. These include the Triodieae, Pappophoreae, Orcuttieae, and Chlorideae. The distinguishing characters among these tribes are not clearly defined in their database nor hardcopy publication. However, in the inter- active database, genera of the Orcuttieae can be separated from the Chlorideae by not having leaves that are clearly differentiated into a sheath and blade. Genera of the Pappophoreae can be characterized by lemmas with 7-many nerves and 7-many lobes, disarticulation above the glumes, paniculate inflorescences, and relatively long glumes. New World Troidieae (including only Triodia) can be differentiated from other chloridoids by indurate lemmas and hard, needle-like leaf blades. All of our New World 538 Sipa 16(3) 1995 Eragrostideae are grouped by Watson and Dallwitz (1992) in the Chlorideae without recognition of subtribal categories. PROPOSED CLASSIFICATION Based primarily on morphological characteristics, we have attempted to place the genera of Eragrostideae into natural groups. We have included the naturalized genera along with the autochthonous genera, even though many of the remaining genera in each of the subtribes may be centered in a continents. By using results from our phenetic analysis (Peterson et 1996) and review of the literature, aca on data from chloroplast ae restriction site es (Duvall et al. 1994), the New World Eragrostideae is appropriately divided into seven subtribes: Eleusininae, Eragrostidinae, Monanthochloinae, Muhlenbergiinae, Munroinae, Sporobolinae, and Uniolinae (see Table 2). Dactyloctenium and Eleusine are the New World representatives of the Eleusininae. An inflorescence with digitate primary branches is the diagnos- tic characteristic for this subcribe. Additional significant characters that de- fine the Eleusininae include the presence of a free pericarp and a membra- nous ligule. The Eleusininae obviously has arisen in the Old World tropics, quite possibly in East Africa where species diversity of Dactyloctenium and Eleusine is greatest. The few species represented in the New World of these two genera suggest migration from the Old World has occurred recently. The Eragrostidinae is, at best, an unnatural grouping of convenience, since the relationships among these 15 genera is poorly understood (Table 2). However, based on the full range of characters used here, these genera are presently excluded from our other subtribes. More than one floret per spikelet, fruit with an adnate pericarp, and base chromosome number of 10, 20 or 30 distinguishes the Eragrostidinae from the Eleusininae, Munroinae, Sporobolinae, and the Uniolinae. Important characters that can be used to separate the Eragrostidinae from most of the Monanthochloinae and Muhlenbergiinae include leaf arrangement (distichous in Monanthochloinae), sexuality (dioecious in Monanthochloinae, except Swallenia), lemma texture (relatively thick in Monanthochloinae), and num- ber of lemmatal nerves (3-nerved in Muhlenbergiinae). No characters ex- clusively distinguish the Eragrostidinae from the Muhlenbergiinae. Even- tually, as new data becomes available we suspect that many of these 15 taxa will be aligned within other current subtribes or placed within new, smaller monophyletic assemblages. The Uniolinae is a curious small subcribe that includes only four genera worldwide (Clayton & Renvoize 1986), three distributed in Africa and Uniola centered along the subtropical and tropical coastal regions of North PETERSON ET AL., Eragrostideae of the New World Be Taste 2. The subtribal classification of the New World Eragrostideae. Eleusininae Dumort., Anal. Fam. 63. 1829. Type: Elewsine Genera: Dactyloctenium Eleusine Eragrostidinae Presl, Rel. Haenk., 1:273. 1830. Type: Eragrostis Genera: Eragrostis Sobnsia OMNI Steirachne Leptochloa Trichonenra Neeragrostis Tridens Neesiochloa Triplasis Neyraudia Tripogon edfieldia Vaseyochloa Scleropogon Monanthochloinae Potztal, Willdenowia 5:472. 1969. Type: Monanthochloé Genera: Allolepis Monanthochloé Distichlis Reederochloa Jouvea Swallenia Muhlenbergiinae Pilger, Nat. Pfl. Fam. ed.2, 14d:168. 1956. Type: Muhblenbergia Genera: Bealia Lycuris Blepharoneuron Mubhlenbergia Chaboissaea Pereilema Munroinae Parodi ex P. M. Peterson, Gram. Bonar. ed. 4:28. 1946, nom nud. Type: Manroa Genera: Ble se idachne Evioneuron Dasyochloa Muanroa Sporobolinae Bench., J. Linn. Soc., Bot. 19:30. 1881. Type: Sporobolus Genera: Calamovilfa Sporobolis PV Ypsis Uniolinae Clayton, Kew Bull. 37:417. 1982. Type: Uniola Genera: Tetrachne Uniola and Central America. Presence of S—20 florets per spikelet, fruit with a free pericarp, and ligule a line of hairs are characters, when used in combina- tion, that serve to distinguish the Uniolinae from the other subtribes. Char- acters of secondary significance include: a 3—9-nerved lemma, primary in- florescence branches racemose or paniculate, and lemma entire at the apex. The Uniolinae shows some affinities with the Eleusininae, since both subtribes contain species with free pericarps, laterally flattened spikelets, and spikelets that are two-ranked along primary inflorescence branches. 540 Sipa 16(3) 1995 Calamovilfa, Crypsis, and Sporobolus are the New World members of the Sporobolinae, which is exclusively defined by the presence of a 1-nerved lemma. Secondary diagnostic characters include one floret per spikelet and a line of hairs for a ligule. The Sporobolinae are closely linked to the Eleusininae and Uniolinae, as all three subtribes possess fruits with free pericarps. Our analysis does not indicate a close affinity between the Muhlenbergiinae and the Sporobolinae even though both have a one-flowered spikelet. Origin of the Sporobolinae probably lies in Africa or the Eastern Mediterranean where species diversity is greatest Allolepts, Distichlis, Jonvea, Monanthochloé, Reederochloa, and Swallenia are the New World members of the Monathochloinae (Table 2). For most taxa, the presence of distichous leaf arrangement, thick textured lemmas, and the dioecious habit distinguishes this subtribe. Secondary diagnostic char- acters include florets 2-25 per spikelet, fruic wich an adnate pericarp, and emmas entire at the apex. Numerous morphological and anatomical adop- tions, 1.e., distinctly distichous leaf arrangement and bicellular microhairs with enlarged bases, are found in some members of this subtribe. These adaptations are in direct response to the environment, since most of the species occur in saline habitats. Based on our phenetic analyses and those by Soderstrom and Decker (1963, 1964, 1965), the Monanthochloinae seems to be a good monophyletic unit. The Monanthochloinae is primarily New World in distribution, only Ae/wropus Trin. is restricted to the Mediterra- nean, northern China, Ethiopia, and Sri Lanka. The distribution of the Muhlenbergiinae is almost entirely New World, only approximately eight of the 160 species of Muhlenbergia are known to occur in southern Asia. The six genera of the Muhlenbergiinae (Bea/ia, Blepharoneuron, Chabotssaea, Lycurus, Muhlenbergia, Peretlema) are character- ized by 1—3 florets per spikelet, 3-nerved lemmas, true caryopses, and a base chromosome number of 8 or 10. These characters do not adequately distinguish the Muhlenbergiinae from the Munroinae. However, the Munroinae always has 2—12 florets per spikelet and the lemma apex is usu- ally emarginate to cleft, whereas, the Muhlenbergiinae usually has a single floret per spikelet and the lemma apex is mostly entire. Evidence from molecular data supports this subcribe as being monophyletic (Duvall et al. 1994) The Munroinae, originally named by Parodi, is formally described be- low. At present, this subtribe consists pore of the following New World genera: Blepharidachne, Dasyochloa, Erionenron, and Munroa. Diag- — — nostic characteristics of the subtribe include lemmas with emarginate to cleft apices and a base chromosome number of x = 7 or 8. Molecular studies support the conclusion that the Munroinae evolved from a common ances- tor (Duvall et al. 1994). PETERSON ET AL., Eragrostideae of the New World 541 Munroinae Parodi ex P. M. Peterson, subtribus nov. Parodi, Gram. Bonar. ed. 4:28. 1946, nom nud. Vaginae ciliatae; spiculae 2—20 flosculis; lemmata 3-nervia, villosa, aristata, emarginata vel lobata; stamina 1—3 flava; chromosomata: x = 7 vel 8 Typus: Muvnroa Torr. DESCRIPTIONS FOR RECOGNIZED SUBTRIBES Eleusininae Plants hermaphroditic. Ligule a membrane or a ciliate membrane. Leaf blades linear. Primary branches of the inflorescence digitate. Spikelets 4—6 mm long; laterally compressed. First glume 1-nerved. Second glume shorter than lower lemma; 1—3(—5)-nerved. Rachilla pronounced between the florets. Florets 3-15 per spikelet. Sterile florets present. Lemma entire, awned, mucronate, or unawned; glabrous; 3—5-nerved; membranous. Palea mem- branous. Lodicules truncate or acuminate. Fruit with a free pericarp. Base chromosome number, x = 9, 10, or 12. Genera Included: Dactyloctentum Willd., and Eleusine Gaertn. Eragrostidinae Plants hermaphroditic, dioecious, or monoecious. Ligule a membrane, a ciliate membrane, or a line of hairs. Leaf blades filiform, linear, or triangu- lar. Primary branches of the inflorescence not digitate. Spikelets 1-40 mm long; laterally compressed or terete. First glume 1—5-nerved. Second glume shorter than or about the same length as the lower lemma; 1—9-nerved. Rachilla pronounced between the florets. Florets 2-60 per spikelet. Sterile florets present (occ. absent in Eragrostis). Lemma entire, emarginate, lobed, or cleft; awned, mucronate, or unawned; glabrous or hairy; 3—9-nerved; hyaline, membranous, chartaceous, coriaceous, or indurate. Palea hyaline, membranous, or chartaceous. Lodicules truncate, rounded, or acuminate. Fruit with an adnate pericarp. Base chromosome number, x = 10, 20, or 30. Genera Included: Eragrostis Wolf, Gouinia Benth., Leptochloa P. Beauv., Neeragrostis Bush, Neestochloa Pilger, Neyraudia Hook. f., Redfieldia Vasey, Scleropogon Phil., Sohnsia Airy-Shaw, Steirachne Ekman, Trichoneura Anderss., Tridens Roem. & Schult., Triplasis P. Beauv., Tripogon Roem. & Schult., and Vaseyochloa Hitche. Monanthochloinae Plants hermaphroditic or dioecious. Ligule a membrane, a ciliate mem- brane, or a line of hairs. Leaf blades linear. Primary branches of the infloresence not digitate. Spikelets 5-30 mm long; laterally compressed or terete. First glume 0—8-nerved. Second glume shorter than or about the same length as the lower lemma; 1—11-nerved. Rachilla pronounced or not pronounced between the florets. Florets 2-25 per spikelet. Sterile florets 542 Sipa 16(3) 1995 present. Lemma entire; unawned; glabrous or hairy; 3—13-nerved; chartaceous, or coriaceous, or indurate. Palea membranous, chartaceous, coriaceous, or indurate. Lodicules truncate or cuneate. Fruit with an adnate pericarp. Base chromosome number, x = 10, 19, or 20. Genera Included: A //olepis Soderstr. & Deck., Distichlis Raf., Jouvea Fourn., Monanthochloé Engelm., Reederochloa, and Swallenia Soderstr. & Deck. Muhlenbergiinae Plants hermaphroditic or andromonecious. Ligu — ea membrane or a cili- ate membrane. Leaf blades filiform or linear. Primary branches of the inflo- rescence not digitate. Spikelets 0.5—8 mm long; laterally compressed, ter- ete, or dorsiventrally compressed. First glume 1—3-nerved. Second glume shorter than, about the same length as, or longer than lower lemma: 1—4- nerved. Rachilla pronounced or not pronounced between the florets. Flo- rets 1-3 per spikelet. Sterile florets present or absent. Lemma entire, emar- ginate, or lobed; awned, mucronate, or unawned: glabrous or hairy; 3-nerved; hyaline, membranous, or chartaceous. Palea hyaline, membranous, or chartaceous. Lodicules truncate. Fruit with an adnate pericarp. Base chro- mosome number, x = 8, or 10. Genera Included: Bealia Scribn., Blepharonenron Nash, Chaboissaea Fourn., Lyenras Kunth, Muhlenbergia Schribn., and Pereslema Pres. Munroinae Plants hermaphroditic, monoecious, or gynomonoecious. Ligule a cili- ate membrane, a line of hairs, or absent. Leaf blades linear, or triangular. Primary branches of the inflorescence not digitate. Spikelets 5—12 mm long; aterally compressed. First glume |-nerved. Second glume shorter than lower lemma, about the same length as the lower lemma, or longer chan lower lemma; l-nerved. Rachilla pronounced between the florets. Florets 2-12 per spikelec. Sterile florets present. Lemma emarginate, lobed, or cleft: awned; hairy; 3-nerved; membranous, or coriaceous. Palea membranous. Lodicules truncate, or cuneate. Fruit with an adnate pericarp. Base chro- mosome number, x = 7, or 8. Genera Included: Blepharidachne Hack., Dasyochloa Rydb., Erioneuron Nash, and Munroa Torr. — Sporobolinae Plants hermaphroditic. Ligule a line of hairs. Leaf blades filiform, or linear. Primary branches of the inflorescence not digitate. Spikelets 1-10 mm long; laterally compressed or terete. First glume nerveless or 1-nerved. Second glume about the same length as the lower lemma; | -nerved. Rachilla hot pronounced between the florets. Florets | per spikelet. Sterile florets absent. Lemma entire; unawned; glabrous or hairy; 1-nerved; hyaline, mem- PETERSON ET AL., Eragrostideae of the New World 543 branous, or chartaceous. Palea hyaline, membranous, or chartaceous. Lodi- cules truncate. Fruit with a free pericarp. Base chromosome number, x = 8, Ovor 10; Genera Included: Calamovilfa (A. Gray) Scribn., Crypsis Ait., and Sporobolus Br. Uniolinae Plants hermaphroditic. Ligule a line of hairs. Leaf blades linear. Primary branches of the inflorescence not digitate. Spikelets 3-30 mm long laterally compressed. First glume 1—5-nerved. Second glume shorter than lower lemma; 1—5S-nerved. Rachilla pronounced between the florets. Florets 5-20 per spikelet. Sterile florets present. Lemma entire; unawned; glabrous; 3—9- nerved; membranous or coriaceous. Palea membranous or chartaceous. Lodi- cules truncate. Fruit with a free pericarp. Base chromosome number, x = 10. Genera Included: Tetrachne Nees and Uniola L. ACKNOWLEDGMENTS Appreciation is extended to Joseph H. Kirkbride for providing an En- glish translation of Pilger’s (1956) treatment of the Eragrostoideae. Stephan L. Hatch, David W. Hall, and Harold E. Robinson are thanked for review- ing the manuscript on short notice. REFERENCES CLAYTON, W.D. and S.A. Renvoize. 1986. a graminum, grasses of the World. Royal Botanic Gardens, Kew Bull. Add. Ser me M.J. 1980. A general system a coding taxonomic descriptions. Taxon 29: 41-46. ane M.J., T.A. Paine, and E.J. ZuRcHER. 1993. User’s guide to the DELTA System: A general system for ees taxonomic descriptions. 4th ed. CSIRO Division of Entomology: Canberra, Austra DUVALL, a P.M. PETERSON, ae A H. Curisrenrsen. 1994. Alliances of Muhlenbergia (Poaceae) within New World Eragrostideae are identified by phylogenetic analysis of mapped restriction sites a plastid DNAS. Amer. J. Bot. (in press) GouLp, EW. 1968. Grass sytematics. New York, McGraw-Hill Book Co. GouLp, F.W. and R.B. Suaw. 1983. Grass systematics, 2nd ed. Texas A&M University Press, College Station, Texas. Hircucock, A.S. and A. CHASE. 1. Manual of grasses of the United States, 2nd Ed. U.S. ub D Peterson, P.M. 1988. ae numbers in the annual Mablenbergia (Poaceae). Madrofio 35:320-324. 1989. A re-evaluation of Bealia mexicana (Poaceae: Eragrostideae). Madrono 36:260-265. PreTeRSON, P.M. and C.R. ANNABLE. 1990. A revision of Blepharoneuron (Poaceae: Eragrostideae). Syst. Bot. 15:515-525. _ «1991. Systematics of the annual species of Muhlenbergia (Poaceae- Biveroetidess). Syst Bot. Monogr. 31: 9 544 SIDA 16(3) 1995 ——s*d' 92. A revision of Chahoissaea (Poaceae:Eragrostideae). Madrofio 39:8—30. Pererson, P.M., C.R. ANNABLE, and V.R. FRANcESCHI. 1989. oo leaf anatomy of the annual Mshlabenpy (Poaceae). eee Bot. 8:575-58 PETERSON, P.M., M.R. Duvatt, and A.H. CurisrENsEN 1993. ieee differentiation among Bealia mexicana, iva nee argentea, and M. lucida (Poaceae: Eragrostideae). Saou ee 148— PETERSON, and Y. Herrera A. 1995. Allozyme variation in the amphitropical dis- y I junct, Buea see Eragrostideae). Madrono. (in rev PeTerson, P.M., R.D. Wepsrer, and J. VaLpis-REYNA. | ioe. a of New World Eragroscideae (Poaceae: Chloridoideae). Smithsonian Contr. Bot. (in press) Pitcer, R. 1954 Ssysecin der Gramineae unter Auschluss der Bambusoideae. Bot. Jahrb. Syst. 76:281— ee ey i Cannacie Il. Unterfamilien: Micraioideae, Eragrostideae, Oryzoideae, Olyroideae. In: H. Melchoir and E. Werdermann, eds. Die Nat. Pflanzenfam. 2nd ed. 4:1-168. Soperstrom, T.R. and H.F. Decker. 1963. Swallenia, a new name for the California genus Ectosperma (Gramineae). Madrono 17:88. 1964. Reederochloa, A new genus of dioecious grasses from Mexico. Brittonia 16:334-339. 1965. Adlolepis: A new segregate of Distichlis (Gramineae). Madrofio 18: ce STAPF, 1898. again In: W. T. Thiselton-Dyer, ed. Flora Capensis. L. Reeve & Co., ane Vol. 7:578— STEBBINS, G.L. and B. c RAMPTON. 1961. A suggested revision of tt perate North America. Recent Adv. Bot. 1:133—145 Watson, L. and M.J. Datiwirz. 1992. The grass genera of the World. C.A.B. Interna- tional, Wallingford, U.K Wesster, R.D. 1988. Genera of the North American Paniceae (Poaceae: Panicoideae). Syst. Bor., 13:576—-609 1992a. Old World genera of the Paniceae (Poaceae: Panicoideae). Sida 15: le grass genera of tem- 9-40, ——_____—.. 199 2b. Character significance and generic similarities in the Paniceae (Poaceae:Panicoideae). Sida 15:185—213 Vatpes-Reyna, J. and S.L. Harcu. 1991. iene micromorphology in the Eragrostideae (Poaceae). Sida 14:531—549. MARSILEA MINUTA (MARSILEACEAE): NEW TO FLORIDA AND NORTH AMERICA JAMES R. BURKHALTER Michael 1. Cousens Herbarium Building 58, Room 120 University of West Florida Pensacola, FL 32514, U.S.A. ABSTRACT Marsilea minuta L. (Marsileaceae), discovered in a streetside ditch in Escambia County, Florida, is here reported as newly introduced for Florida and North America. A probable explanation of its introduction via waterfowl and its potenti: ie as a weed are considered. ABSTRACT Se cita Marsilea minuta L. Sees descubierta en una cuneta del borde de la carretera en el Condado de Escambia, Florida, como reciente ne6fito para Florida y Norte América. Se discute una explicacién ree de su introducci6n mediante las corrientes de agua y su potencial como mala hierba. An obviously spontaneous population of the primarily tropical water- clover fern Marsilea minuta L. has been found growing ina streetside ditch at the northwest corner of Lownde Avenue and Grundy Street in a residen- tial subdivision on the southwest side of the Pensacola area in Escambia County, Florida. At the time of discovery it was realized that the nearest documented sites for Marsi/ea in Florida were many kilometers to the east and south in the peninsula (Ward and Hall 1976; Johnson 1986) and in Apalachicola (Anderson 1986) for the species M. vestita Hook. & Grev. and westward in Mobile, Alabama, for M. macropoda Engelm. ex A. Braun in Kunze (Burkhalter 1989). It was also realized that Marsi/ea in the Pensacola area constituted a new county record and a new species of vascular plant for the entire western Florida panhandle (Clewell 1980; Wilhelm 1984). Only the genus of the water-clover fern, namely Marsilea—a taxonomically dif- ficult group—was recognized at the time of discovery, so a sporocarp-bear- ing specimen (Burkhalter 13220) collected 3 May 1992 was sent to David M. Johnson of Ohio Wesleyan University for determination. Dr. Johnson identified it as Marsilea minuta L. Johnson 1986). Duplicates of that speci- men were deposited at UWFP (Fig. 1). Additional fertile specimens (Burkhalter 13304) collected at the discovery site 19 June 1992 were sub- sequently deposited at UW FP, FSU, FLAS, USE and NY. According to the information provided by Dr. Johnson (pers. comm. 27 August 1992) Sipa 16(3): 545 — 549. 1995 546 Sipa 16(3) 1995 Marsilea minuta is hereby reported as new to Florida and North America (also see Flora N. Amer. Ed. Comm. 1993, p. 333). At the discovery site M. minuta is the dominant vascular species in the described 60-cm-deep ditch for a distance of approximately 30 m north- ward from the stated intersection. Associated species in this section of the ditch include Hypericum mutilum, Ludwigia decurrens, Hydrocotyle umbellata, Mtkania scandens, Lipocarpha maculata, Juncus elliottii, Paspalum urvillei, Pani- cum repens, Sacctolepis striata, Thelypteris palustris, and Osmunda regalis. The substrate is composed of black muddy silt mixed with tan sand, and stand- ing or running water is almost constantly present. To the south of the in- tersection the ditch is very shallow and grades into adjacent residential lawns. There only a few scattered individuals of M. minuta occur over a distance of approximately 20 m in association with Lilgeopsis attenuata, Cardamine pensylvanica, and lawn grasses. This section of the ditch is sub- jected to periodic mowing. Less than 250 m southward the ditch drains into Bayou Grande, a large lagoonlike extension off the western side of Pensacola Bay. A detailed search of all ditches and other low moist habitats in the general geographic area has revealed that M. minuta occurs only in the ditch where it was discovered. An interview with Andrew L. Lucas, who resides across the street from the Marsilea-containing ditch, has provided informa- tion which leads me to conclude that the M. minita colony is not very old and perhaps became established less than five years prior to its discovery. Marsilea minuta is primarily tropical in its eastern hemisphere distribu- tion (Johnson 1986). There it occurs as a common and widespread weed in Africa and India. In the western hemisphere, where it is introduced, it has heretofore been known only from the islands of Trinidad and Tobago and from the state of Pernambuco in eastern Brazil (Mickel 1985; Johnson 1986). The New World colonies occur at low elevations near the coast in freshwa- ter or occasionally brackish habitats (Trinidad and Tobago) or in seasonal ponds (Brazil). The question of how M. minuta became introduced into the Pensacola area deserves consideration. Johnson (1986) has reviewed the literature con- cerning long-distance dispersal of Marsé/ea sporocarps in the digestive tracts of migratory aquatic birds; and this mode of transport for certain species of Marsilea, particularly via various ducks (genus Anas, and possibly also Azx), is an established fact and a not uncommon occurrence. Dennis and Webb (1981) have commented on similar long-distance dispersal of Pilwlaria spo- rocarps by waterfowl. It thus seems apparent that the most probable mode of introduction of M. muta into the Pensacola area is long-distance sporo- Carp transport via waterfowl from South America. This is particularly likely in view of the proximity of the discovery site to a waterfront area (Bayou Grande) visited by migratory waterfowl and the fact that a number of BurKHALTER, Marsilea minuta in North America 547 Marsiles mimite Lb. fet, OG. M. Jehneon GWU (Gapl.) Pants of FLORIDA bin Go., tar Fonsaeol, KW corner of lownde &ve and wide ditch. dominant in d@ re pers. | HERBARIUM OF THE UNIVERSITY OF WEST FLORIDA | are 5 MY lose CoH, James &, Burkhalter Fic. 1. Herbarium specimen of Marsilea minuta L. (Burkhalter 13220, UWFP). 548 Stipa 16(3) 1995 migratory waterfowl species with suitable dietary habits (including species of Anas and Porphyrula) use flyways which pass over Pensacola and the known sites for M. minuta in Brazil, Trinidad, and Tobago (cf. Pough 1951, Weston 1965; Kale & Maehr 1990; Ffrench 1991). Of course, one bird transport- ing one sporocarp ts all that would have been required to initiate the sub- ject colony of M. minuta near Pensacola. More than two years of observations have revealed that Marsilea minuta is evergreen in the Pensacola area and that it easily withstood two days of near- and below-freezing temperatures on 12—13 March 1993 and another unusually cold period on 17-18 January 1994. These facts, together with the documented invasive and weedy tendencies of this fern in other areas VJohnson 1986), lead to the conclusion that M. minuta indeed has the po- tential to become a weed of some importance throughout the southeastern United States. Suitable habitats for possible invasion by M. minuta include roadside ditches, sunny edges of freshwater and brackish marshes and swamp forests, and other similar low moist sites. Based on the luxuriant growth of M. minuta in the ditch at the discovery site, there is some con- cern that ic would grow as vigorously if it became introduced into other similar sites. Fortunately, as of this writing, this interesting invader from South America is apparently restricted to the single streetside ditch where it was discovered. ACKNOWLEDGMENTS Tam grateful to Dr. David M. Johnson for specific identification of the Marsilea minuta specimens cited in this paper. I also thank local ornitholo- gists C. W. Milmore and Robert and Lucy Duncan for sharing their exper- tise and information concerning possible avian vectors. REFERENCES ANbeRSON, L.C. 1986. Noteworthy plants from north Florida. I. Sida 11:379—385. BURKHALTER, J.R. 1989. Marsilea macropoda—stll a weed {in Mobile, Alabama]. Fiddle- head Forum 16:20 CLEWELL, ALF. 1985. Guide to the vascular plants of the Florida panhandle. Florida State Uniy. Pr., Tallahassee DeNNis, W.M. and D. H. Wome 1981. The distribution . — americana A. Br. ee eS in North America north of Mexico. Sida 9:1 FrrencH, R. 1991. A guide to the birds of Trinidad & ate on edition. Comstock Publ. Assoc., Ithaca, FLora oF NortH America Eprrortat Committee. 1993. Flora of North America, Vol. 2 Preridophytes and sien Oxford Univ. Press, JOHNSON, D.M. 1986. Systematics of the New World species - Marsilea (Marsileaceae). Bot. Monogr. 1 ee Kate, H.W. and D.S. Marner. 1990. Florida's birds. Pineapple Press, Sarasota, FL. BuRKHALTER, Marsilea minuta in North America 549 Mickel, J.T. 1985. Trinidad pteridophytes. New York Bot. Garden, Bronx, NY. PouGu, R.H. 1951. cai water bird guide. Boubleday & Co., Garden Cy NY. Warp, D.B. and D.W. Hatt. 1976. Re-introduction of Marsilea vestita into Florida. Amer. Fern Jour. 66:113-115. WESTON, EM. 1965. A survey of the birdlife of northwestern Florida. Tall Timbers Res. Sta. No. WILHELM, a S. 1984. ee a of the Pensacola region. Ph.D. dissertation, Southern Illinois Univ., Carbondale, 550 Sipa 16(3) 1995 BOOK REVIEW Stugssy, Top F. 1994. Case Studies in Plant Taxonomy: Exercises in Applied Pattern Recognition. (ISBN 0-231-07611-8, pbk.) Colum- bia University Press, New York. $25.00. 171 pp., 59 figures, 38 tables. As an experienced instructor of plant systematics at both the introductory and advanced levels, Tod Scuessy has both perceived and sought to fill the gaps in instructional materi- als. He has been using case studies in his own graduate course for several years. Now he has edited the class exercises in a handy volume to be available for other instructors to use. It will serve as a es teaching resource for any one training young systematists or for self-instructior It is See to be used in conjunction with a comprehensive text, such as Stuessy’s, Bien Taxonomy: The Systematic Evaluation of Comparative Data, or with an instructor's free- standing lectures. However, his inclusion in the first three chapters of some basic concepts of systematics, taxonomy, classification, taxonomic hierarchy, genera, species, infraspecific categories, and the range of taxonomic data, allows the book to be used somewhat inde- pendently. Although I found these chapters to be redundant with introductory texts, they do not detract from the overall usefulness of the book. The first two chi ipters, in particular, give us insight into Stuessy’s perspective on these concepts. The remaining introductory chapter is a guided example of discerning taxonomic pat- —- terns. Insightfully, Stuessy directs students to pay attention to “(1) pone es of charac- Z ter states among taxa; and (2) discontinuities in character states between taxa.’ as dis- appointed that he did not stress statistical analysis of characters to discern disconeiawies in these examples. However, that problem is corrected in the actual case studies, which are rich in bar diagrams, character polygons, and scatter diagram The majority of the book consists of 10 case studies representing original journal ar- ticles. The taxonomic problem, method, data sets, journal reference and a brief discussion of the possible solutions are provided, but the original authors conclusions are not. The student 1s expected to reach his or her own conclusion from the pattern ta in the data. Comparisons among students and to the original paper are encouraged. The case studies progress from simpler to more complex and from data sets of morphology/geogra- phy to those also incorporating macromolecular data. rall, I think that Stuessy is successful in providing a set of exercises that give the sane d scudent experience with interpreting taxonomic patterns. Especially in light of e short time left for botanical exploration, increasing the rate at which students can ee the skills to become practicing systematists makes this book is a welcomed addi- tion to our training resources.—Rover W. Sanders, Sipa 16(3): 550. 1995 TAXONOMY OF THE NATIVE NORTH AMERICAN SPECIES OF SACCHARUM (POACEAE: ANDROPOGONEAE) ROBERT D. WEBSTER Systematic Botany and Mycology Laboratory Agriculture Research Service Bldg. 003, Rm. 235, BARC-West Beltsville, MD 20705, U.S.A. ROBERT B. SHAW Department of Rangeland Ecosystem Science Center for Ecological Management of Military Lands Colorado State University Fort Collins, CO 80523, U.S.A, ABSTRACT Saccharum L. consists of about 40 species concentrated in the tropics and subtropics of the world. Five species and one variety are recognized in this revision of the native North American taxa. Recognized taxa include S. alopecuroideum (L.) Nutt., 8. baldwinii Spreng., S. brevibarbe (Michx.) Pers. var. brevibarbe, 8. brevibarbe (Michx.) Pers. var. contortum (Nutt.) ter, S. coarctatum (Fern.) R. Webster, S. giganteum (Walt.) Pers. A review of the relevant taxonomic history of the taxa is presented. A comprehensive set of morphological, geographical, and nomenclatural data was collected. A discussion is given on the a and distribution of the 234 characters applied in the taxonomic analysis. The LTA computer system was used for the analysis and production of a me to the species and comparative species descriptions. Given for each taxon is a representative set of specimen citations and a discussion of diagnostic characters and biological ona among the recognized specie RESUMEN Saccharum L. esta formado por cerca de 40 especies concentradas en las zonas tropicales y subtropicales del mundo. ee esta revision de los taxa nativos de Norteamérica se reconocen cinco especies y una variedad. Los taxa reconocidos son: S. a So (L.) Nutt., 8 baldwinii Spreng., S. brev an Miche Pers. var. contortum (Nutt.) R. We 1, S. coarctatum ebster y S. gigantenm (Walt.) Pers. Se presenta una revisi6n ae la historia taxon6mica de los taxa. Se colecié un conjunto extenso de datos morfolégicos, geograficos y nomenclaturales. Se hace una disucsién del rango y diseribucién de los 234 caracteres utilizados en el andlisis taxonédmico. Se empled el sistema DELTA para el analisis, la elaboraci6n de la clave de especies y para las descripciones comparativas de las especies. De cada taxon se da un conjunto representativo de citas y se hace una discusion de los caracteres diagnoésticos y de las relaciones biolégicas entre las especies que se reconocen. Sipa 16(3): 551-580. 1995 552 Stpa 16(3) 1995 INTRODUCTION North American floristic authors traditionally treated the native species of Saccharum L. under Erianthus Michx., which was established in Flora Boreali Americana (Michaux 1803). Hitchcock (1951) recognized both genera and separated them on the traditional character of presence or ab- sence of the upper lemma awn. That is, Saccharum species possess an awned upper lemma, whereas Erzanthus species lack the awn. For the grasses of South Africa, Chippendall (1955) treated only the introduced species, S. officinarum L., but described the genus as awnless or rarely awned. Bor (1960) used the same character in his treatment of Asian grasses to differentiate the genera; however, he listed Erianthus species under both genera. South American writers (Burkart et al. 1969; Rosengurtt et al. 1970; Smith et al. 1982) separated the genera on the same character. Bor’s (1970) “Gramineae” in Flora Iranica recognized only Saccharum and submerged Erianthus as a section differentiated on the absence of an awn. Taxonomists associated with the sugarcane breeding programs (Dutt & Roa 1950) consistently placed the species in separate genera. In Grasses of the Soviet Union, Tsvelev’s (1976) separated the genera on presence of an awn and length of the callus hairs. Hsu’s (1978) treatment of Taiwanian grasses separated Erianthus and Saccharum on whether culms are solid, reduction of the upper lemma, and presence of an awn. Finally, Clayton and Renvoize (1986) concluded that placing the awned and unawned species into separate genera was artificial and recognized only Saccharum. In the absence of conclusive taxonomic evi- dence to the contrary, it seems appropriate at the present time to follow the concepts of Clayton and Renvoize (1986), Bor (1970), and Renvoize (1984) and treat Erianthus as a synonym of Saccharum. Saccharum consists of about 40 species which mostly occur in the tropics and subtropics of the world. The center of diversity is tropical Asia, with approximately 25 native species occur. There is apparently only one native species in Africa and no species are known to be native to Australia. Four to 10 species occur in North America north of Mexico, excluding the Carib- bean and 3 to 7 species are native to South America and Mesoamerica. The taxonomic relationships among the native taxa of Saccharum have not been carefully studied nor well-defined. The objective of the present study is to define this relationship for the North American taxa. The introduced spe- cies {S. spontaneum L., 8. officinarum L., and S. ravennae (L.) P. Beauvy.} are not treated here. This study was based on field studies and loaned specimens from BM, BR, BRI, BRIT, C, CANB, F, GH, ISC, K, L, MICH, MO, NY, PH, and US. The DELTA system (Dallwitz 1980; Dallwitz et al. 1993) was used to gather, analyze, and present the taxonomic data. These procedures are dis- WEBSTER AND SHAW, North American species of Saccharum Ss) cussed in previous studies by Webster (1988, 1992) but require a brief statement here. Initially, a list of characters and associated states was con- structed to account for taxonomically significant variation within the com- plex. Classical methods of specimen analysis were used for the recognition of the taxa. Specimens of a recognized taxon were used to record the full range of data for each of the 285 characters. This data together with the authors concepts of species and character reliability were incorporated into the DELTA programs (CONFOR, KEY, & INTKEY) for the production of the identification key and descriptions. TAXONOMIC HISTORY The objective of this section is to provide a chronological explanation of the taxomonic development of the group, which begins with Linnaeus’s Species Plantarum (1753) where Saccharum L. and two species, Andropogon alopecuroides L. and Andropogon divaricatus L., were named. Thirty-five years later Walter (1788) named the taxon Anthoxanthum giganteum and created a taxonommic problem which persists today. Andre Michaux (1803) in Flora Boreali-Americana defined Erianthus Michx. and two species, E. brevibarbis Michx. and E. saccharoides Michx. Persoon (1805) transferred E. brevibarbis and A. giganteum, to Saccharum, conversely, Muhlenberg (1813) transferred A. giganteum to Erianthus. Even at this relatively early date there were dif- ferences in opinion concerning the relationship between Erzanthus and Saccharum. Elliott’s Sketch of the Botany of South Carolina and Georgia (1816) contains two new species, E. contortus and E. strictus, and transfers the epi- thet Andropogon alopecuroides to Erianthus. His work was soon followed by Nuttall (1818), who placed E. strictus, E. contortus, and E. alopecuroides in Saccharum. Sprengel (1815) made the new combination, S. strictum (Host.) Spreng., foran Old World species, making Nuttall’s combination [S. strictum (Baldw.) Nutt.} illegitimate. Sprengel (1825) provided the name, 5S. baldwinii, for this taxon. Chapman (1860) treated Andropogon brevibarbis and E. contortus as varieties of E. alopecuroides. Between 1895 and 1900 Nash named four new species and one variety of Erianthus (E. smallit, E. compactus, E. laxus, E. tracyi, and E. alopecuroides var. hirsutis). Fernald (1943) con- cluded that the name brevibarbis had been misapplied and named a new species and two varieties (E. coarctatus , E. coarctatus var. elliottianus, and E. saccharoides vat. compactus). As with many species complexes of flowering plants the most useful taxonomic information is contained in the regional floristic treatments. The earliest significant treatment is Mohr (1901), who recognized five spe- cies (E. alopecuroides, E. saccharoides, E. brevibarbis, E. strictus, and E. smallit), but does not give a key. In that work, E. giganteum and E. saccharoides are 554 Sipa 16(3) 1995 nomenclaturally confused, the concept of E. brevibarbis is incorrect, as is the relationship between E. smallii and E. contortus, Nash's (1903, 1913) treatments in Small’s manual follows these concepts and recognizes his four new species. Albert Hitchcock contributed the Gramineae for the 1933 3rd edition of Small’s manual, where five species (E. brevibarbis, E. contortus, E. divavricatus, E. saccharoides, and E. strictus) were recognized. Nash’s taxa were placed in synonomy. Most of the morphological and nomenclatural confusion occurring in Mohr (1901) was duplicated in Hitchcock’s (1933) treatment. Hitchcock (1935) presented a similar morphological treatment but uses the name E. a/opecuroides in place of E. divaricatus, and E. gigantens in place of E. saccharoides. Fernald (1950) treatment incorporated his clari- fication (Fernald 1943) of E. brevibarbis and recognized six species and one variety (E. giganteus var. compactus). Hitchcock (1951) followed Fernald’s concepts and recognized the same taxa. Mukherjee (1958) presented a world- wide revision of Erianthus and recognized eight North American species; however, his conclusions were not based on a detailed study of New World material or collections. Radford et al. (1964) recognized five species and apparently discounted or did not appreciate Fernald’s contributions. Their treatment is similar to Hitchcock (1935). Strausb Augh and Core (1970) in their flora of West Virginia included one species (E. alopercuroides). Correll and Johnston (1970) follow the concepts presented in Mukherjee (1958), and provide a key to four species, which does not account for the known morphological variation within the complex. Long and Lakela (1971) rec- ognizes only one species, E. giganteus, as occurring in tropical Florida. A recent explanation of the nomenclature of E. giganteus is given in Gandi and Dutton (1993). Gould (1975) followed Hitchcock’s (1951) morpho- logical concepts, and reported five species from Texas, and included a key to the species. Allen (1975) uses Mukherjee (1958) as a reference and re- ports five species for Louisiana. The most complete morphological data for these species is given in Godfrey and Wooten (1979), where five species are listed; however, their species concepts appear identical to Hitchcock (5D) — — CHARACTER VARIATION The purpose of this section is to define characters used in this study and to discuss the range in variation found among the taxa as compared to all grasses. Of the 234 characters considered in this study, 63 apply to vegeta- tive structures, 49 describe the inflorescence, 103 describe spikelet parts, and 19 apply to topics other than morphology (eg. geography and ecol- ogy). Characters common to all taxa are applicable at the generic level. All taxa are hermaphroditic, in that, no sexual differences exist among the spikelets of a plant. This is an exclusively perennial complex, lacking stolons, but with distinctly compacted rhizomes which produces a knotty WEBSTER AND SHAW, North American species of Saccharum 555 crown. Rhizomes are relatively short, except in S, alopecuroideum, and soon become erect to produce a new flowering culm. Leaves of the rhizomes are glabrous and striate. Flowering culms are not lignified, not caespitose, erect, never root at the lower nodes, and typically vary from 0.8 to 2.5 meters in height. The shortest species is S. baldwinii and the tallest S. gégantenm, however, height overlaps exist for these taxa. Each flowering culm is un- branched, terminates in a solitary inflorescence, and consists of 4—8 nodes. Culm nodes are not swollen and may be glabrous as in S. baldwini or long pilose as in S. a/opecuroideum. Internodes are glabrous, smooth, and hollow at maturity; however, immature internodes may be solid to spongy. Glau- cous internodes occasionally occur in S. alopecuroidenm, otherwise viscid or glaucous internodes do not occur in this complex. A basal cluster of leaves is not present, as all leaves originate from flowering culm nodes. Pronounced leaf auricles are not present; however, minute sheath au- ricles, measuring 0.3—1.0 mm long, are present in S$. baldwinii and S. coarctatum. Leaf sheaths are smooth, closed, rounded on the back, do not overlap, and the length has no diagnostic value. Sheaths are typically gla- brous but are hairy in some immature specimens of S. giganteum. The apex of the sheath is ciliate in S. alopecuroideum and S. giganteum, but otherwise undifferentiated. The ligule is a rounded, ciliate membrane, measuring |— 6 (usually 2—3) mm long, and cannot be relied on to differentiate among the taxa. However, Saccharum giganteum is the only taxon with a differenti- ated collar with long pilose hairs. Leaf blades are linear, flat, lax, flexuous, and spreading in all taxa. Shorter and narrower blades are found in S. haldwinii and S. giganteum; however, range overlaps exist for all taxa and these characters have little or no diagnostic value. Surface of leaf blades are adaxially and abaxially smooth in all taxa and the margins are smooth or minutely scabrous. At maturity, S. g/gantewm is the only species which may possess hairy (pilose) leaf blades. Margins of the blades are flat and not thickened; however, the midvein of the blade is pronounced or swollen in all taxa. Significant variation was not found in shape of the acuminate apex and truncate base. Morphology of the prophyllum was not included. The peduncle (stalk of the inflorescence) is typically elongate, but rela- tively short in S. baldwinii and S. coarctatum. It is hairy, especially at the apex, in S, alopecuroideum and S. giganteum, but otherwise glabrous and un- differentiated. The inflorescence is a terminal, fully exserted panicle vary- ing from linear, oblong, to lanceolate. Inflorescence shape is determined by the amount and length of hairs associated with the callus. In S. baldwiniz, the callus hairs are essentially absent and the inflorescence is narrow and linear; whereas, in S. alopecuroideum and S. giganteum the hairs are long, spread at maturity, and the inflorescence is oblong to lanceolate. Swccharum brevibarbe and S. coarctatum are intermediate between these forms. The low- 556 Sipa 16(3) 1995 ermost inflorescence node is differentiated only in the sense that it is hairy in some taxa. The main axis of the inflorescence is present and relatively stout. There are no taxonomically significant differences among the taxa based on length of the main axis. The amount of hairs on the main axis varies from densely hairy co essentially glabrous, and is positively corre- laced with the amount of hairs associated with the callus. Primary branches of the inflorescence potentially originate at all points on the main axis (that is, quaquaversal and not secund or distichous), have appressed secondary branches with distichous spikelets, and are not whorled. The primary branches are best described as appressed; however, presence of dense inflo- rescence hairs produces some spreading and at anthesis the branches spread and produce a slightly open inflorescence. Number of primary branches is difficult to determine, but varies from 8—30, and was not given high taxo- nomic significance in this study. Primary branches are straight, smooth, and glabrous or hairy. Hairiness of the branches tends to occur in lines and is correlated with the amount of hairs on the main axis. Length of the branches varies between 2-18 cm with the rachis internode slightly longer than the pedicel, but shorter than the spikelets. Pedicels are straight, not distinctly grooved, smooth, truncate, and gla- brous or hairy. Hairiness of the pedicels is positively correlated with the amount of hairs associated with the callus and main axis. Disarticulation is identical for all taxa with distinct points at branch internodes and at the base of the pedicelled spikelet. Morphological differences associated with the callus hairs are taxonomically significant in this complex. Other than the presence of hairs, the callus is undifferentiated. The hairs are com- pletely absent or sparse in 8. baldwinii and dense and long in S. alopecuroidenm and $. gsganteum. Other taxa are intermediate between these extremes. Color of the hairs is diagnostically important. Silvery hairs are characteristic of S, alopecuroideum, whereas the hairs of the remainin g taxa vary from white to rown. The single most reliable character for distinguishing among the taxa is the length of the callus hairs relative to spikelet length. They are relatively short in S. baldwinit, long in S. alopecuroidenm and S. giganteum (but of difference color), and intermediate in §. breviharhe and S§. coarctatum. Cleistogamous spikelets were not observed in the taxa; however, certain South American taxa are characterized by this feature. Spikelets are paired, abaxial, slightly overlapping, not embedded in the rachis, evenly distrib- uted on the rachis, and either heteromorphic or homomorphic. Here, het- eromorphism is a result of differences in the amount of hairs. Hacept (ors, baldwinii, it is typical for the glumes of the sessile spikelet to be glabrous or not as densely hairy as the pedicellate spikelet. The sessile and pedicel- late spikelets are otherwise identical. Spikelet color has minor diagnostic importance and cannot be used with confidence to distinguish among the on WEBSTER AND SHAW, North American species of Saccharum 557 taxa. Saccharum baldwinii and S. coarctatum are dark brown in color, whereas the other taxa are best described as straw-colored. Spikelets are dorsiven- trally compressed, lanceolate, and attenuate at the base. Frequently the second glume is slightly keeled. Spikelet length and width are of minor diagnostic value. The smallest spikelets are associated with species charac- terized by long callus hairs. These species may rely on wind dispersal of seeds, whereas species with larger spikelets and shorter hairs may depend more on dispersal by water. For example, 8. ba/dwinii has relatively short hairs or completely lacks hairs and grows in shaded stream bottoms closely associated with moving water; whereas, S. alopecuroideum with long hairs and relatively smaller spikelets inhabits open areas not necessarily associ- ated with water. The first glume is 5 nerved, encircles the spikelet base, and equals spike- let length, spikelet shape, and length of the second glume. It is cartilagi- nous in texture and slightly denser than the second glume. The surface is smooth, except in §. baldwinti where it is scabrous. The degree of hairiness of the first glume is correlated with the amount of hairs associated with the callus. In S. alopecuroidenm and S. gigantenm the first glume is hairy, whereas in S. baldwinii it is glabrous. Apex of the first glume is acuminate and muticous for all taxa; however, commonly the apex is minutely knotched or emarginate. The rachilla is not pronounced in the complex. The second glume is 3 or 5 nerved, essentially smooth, not ciliate, equals spikelet length and shape, acuminate, and muticous. The lower floret consists of a well- developed lemma, with the palea and other structures missing. The lower lemma is glabrous, hyaline, lanceolate, smooth, not keeled, acuminate, and muticous. It usually lacks distinct nerves in S. a/opecuroideum, 8. brevibarbe, and S. giganteum, whereas, in S. baldwinii and S. coarctatum the lower lemma is 2—3-nerved. Rarely, in these latter species the central nerve extends into a pronounced awn. Length of the lower lemma ranges from 3—8 mm and is of minor diagnostic value, as is the relative length to the upper lemma. The upper lemma is lanceolate, hyaline, smooth, glabrous, acuminate, awned, and commonly purple at maturity. Differences exist among the taxa on length of the upper lemma, however, these differences are corre- lated with overall spikelet length. Number of nerves on the upper lemma ranges from 1—3. Saccharum alopecuroideum and S. giganteum are |-nerved, whereas the others are distinctly 3-nerved. In those taxa with a straight awn the upper lemma apex is entire, whereas taxa with a coiled awn have a bifid apex. The bifid apex produces lateral lobes which measure approxi- mately 2 mm long. Morphology of the callus and upper lemma awn are taxonomically the most significant and reliable characters for the complex. This awn is mi- nutely scabrous and glabrous. Length is variable and is of limited taxo- — 558 Stipa 16(3) 1995 nomic value. The awn may be basally flat or terete, spiraled or not, and straight or geniculate. These three characters are all interrelated. That is, a flattened awn results in spiraling and spiraling frequently produces geniculation. A germination Hap is not pronounced in the complex. The palea of the upper floret is hyaline, well-developed, and about 1/2 the length of the upper lemma. It is ovate, smooth, and acute or cleft at che apex. Lodicules are pronounced and measure about 1 mm in length. The taxonomic value of this structure lies in whether the nerves extend into hair-like projections, which is best illustrated in mature specimens of S. coarctatum. Taxa of the complex have 2 stamens, whereas the introduced species, S. ravennae, a na- tive of southern Europe, is characterized by 3. Anthers are red to purple, about 1.7 mm long, and lack taxonomic significance. The caryopsis has a punctiform hilum and the length is correlated with spikelet length. The embryo measures about 1/2 the length of the caryopsis. Base chromosome number for the complex is 10. Cytology of the complex is the topic of a separate study (Burner & Webster 1994). Plants ower mostly from June to November, grow in helophytic to mesophytic conditions, and are locally common throughout the southeastern United States. KEY TO THE SPECIES 1. Awn of upper lemma basally spiraled 2 2. Callus hairs equal to or shorter than the spikelet; callus hairs white to brown; callus hairs less than 7 mm long; main axis sparsely hairy .......... S. brevibarbe 2. Callus hairs longer than the spikelet; callus hairs silvery; callus hairs more than 7 mm long; main axis densely hairy S. alopecurotdenm |. Awn of upper lemma not basally spiraled 3 3. Callus hairs longer than che spikelet; lemma of upper floret 1-nerved; lowermost inflorescence node densely hairy S. gigantenm Callus hairs equal to or shorter than the spikelet; lemma of upper floret 3-nerved; oe rmost inflorescence node not densely hair 4 4. Callus hairs absent or up to 2 mm long; primary ee a inflorescence 10-25 mm wide S. baldwinii — 4. Callus hairs more than 2 mm long; primary branches hairy; c@otes: cence more than 25 mm wide 5 5. Awn of upper lemma basally flattened; lemma of lower floret not distinctly nerved; upper lemma 0.9-1.0 times the length of the lower lemma; leaves sheaths without distinct auric 5. pwn of upper lemma basally terete; lemma of ‘lower floret typically 3-nerved; upper lemma 0.7—0.8 times the length of the lower lemma; leaves with minute sheath auricles S. coavctatum — S. brevibarbe TAXONOMIC TREATMENT Saccharum eo eevee (L.) Nutt., Gen. Pl. 1:60. 1818. Andropogon alopecuroides L., Sp. Pl. 1045. 1753. oe alopecuroides (L.) Elliott, Sketch Bot. S. arolina 1:38. cae ane VIRGINIA, Clayton 601 (HOLOTYPE: LINN 1211.9, pho- tograph seen). I WEBSTER AND SHAW, North American species of Saccharum 0? so eee . , Sp. Pl. 1045. 1753. Erianthus divaricatus (L.) Hitchc., Contr. Herb. 12:125. 1908. Type: VIRGINIA, Clayton 70 (HOLOTYPE: LINN, not en Erianthus tracyt Nash, Bull. as Bot. Club 24:37. 1897. Type: MISSISSIPPI, Starkville, Tracy 5.n. (HOLOTYPE: NY!?; tsorype: US!). Ertanthus sie (L.) Elliott var. birsutus ets in Small, Fl. SE U.S. 55. 1903. TYPE DA, Chapman s.n. (HOLOTYPE: N Rhizomes with elongate internodes. Flowering culms 10—25 dm tall. Nodes hairy (occ. glabrous at maturity; the hairs usually 7-12 mm long). Internodes solid; occasionally glaucous. Leaves without auricles. Sheaths apically ciliate. Ligule 1-3 mm long. Collar not differentiated; glabrous. Leaf blades 30-60 cm long; 14-28 mm wide; glabrous at maturity. Pe- duncle 40—60 cm long; hairy (pilose below the inflorescence). Inflorescence oblong to lanceolate; 3-10 cm wide. Lowermost inflorescence node hairy. Main axis 15-34 cm long; densely pilose. Primary branches appressed to the main axis; 3-12 cm long; hairy; ciliate. Rachis internode 3— long; 0.3 mm wide; hairy. Pedicels 2.5-4 mm long; hairy. Callus hairy; with silvery hairs (frequently tinged in purple). Callus hairs 9-14 mm long; longer than the spikelet. Spikelets heteromorphic; straw-colored; 6— 7 mm long; 1.1—-1.4 mm wide. First glume 5-nerved; smooth. Second glume 3 or 5-nerved. Lemma of lower floret 4.8-5.6 mm long nerveless or 1-nerved. Upper floret 0.6—0.8 times the length of the lower floret. Lemma or of upper floret 44.6 mm long; |-nerved; bifid. Lateral lobes of upper lemma 1.8—2.2 mm long (ciliate). Awn of upper lemma 14-20 mm long; basally flattened; basally spiraled. Lodicules with nerves not extending into hair-like projections (occ. with a few short hairs). Saccharum alopecuroidenm occurs occasionally over a wide area of the south- eastern United States, but is evidently rare or non-existent on the sandy coastal plain. Its western geographic limit is eastern Texas and Oklahoma, whereas the northern limit is southern Missouri, Illinois, Indiana to New Jersey. In addition, there is a paucity of specimens from southern Florida and the higher elevations of the Appalachians. It is common to find S. alopecuroidenm associated with S. gigantenm; however, S. giganteum occupies low moist areas and S. a/opecuroidenm exists on the dry clay upper slopes. Vegetatively, S. alopecuroideum is differentiated from the other species by the presence of pronounced elongate rhizomes up to three inches long. The rhizomes of other members of this complex are short with few nodes and immediately produce an erect flowering culm. Additional diagnostic veg- etative characters include the presence of long, white and pilose hairs on the culm nodes and base of the leaf blades. This hair type also occurs at the 5 mm peduncle apex, main axis, and callus. Presence of silvery callus hairs ex- ceeding spikelet length and a twisted geniculate upper lemma awn differ- Sipa 16(3) 1995 560 Sars <<< a eee LS Ze Le Lg YI G7. {ff YA, Yi Vy Ye4Y r i, hie ALY y, Y 7 V ; / Tf / i. 1. Saccharum alopecurotdeum. A. Habit, B. Ligule, C. Spikelet, pair, D. First Glume, E. Upper Lemma Awn. Oo WEBSTER AND SHAW, North American species of Saccharum 561 entiates this species from others of the group. Saccharum alopecuroidenm is most similar to S. gigantewm, but easily distinguished on the presence of the spiraled upper lemma awn. Specimens with intermediate characteristics are rarely encountered. Representative specimens examined. ALABAMA: Cherokee Co.: 2 mi N Leesburg, 14 Sep 1968, Kral 33376 (BRIT, C). Lawrence Co.: 8.8 mi S Meulten, 23 Sep 1970, Kral s.n. (MO). Lee Co.: 14 Oct 1897, Earle & Baker s.n. (MO). Auburn, 25 Sep 1897, Earle & Baker sn. (MICH). Marengo Co.: 3.5 mi N Dixon’s Mill on US 43, 7 Oct 1967, Kral 29585 (BRIT). Shelby Co.: | mi E of Harpersville by WS.251, Oce Los 1. Kral 44615 (NY). Winston Co.: border of fields, 13 Sep 1897, Eggert s.n. (GH, MO). ARKANSAS: Carroll Co.: Eureka Springs, 24 Sep 1913, Palmer 4470 (MO). Hot Springs Co.: P.O. Malvern, Demaree 34477 (GH). Izard Co.: 3 mi S of Brandenburg, 24 Sep 1963, Robinson DIA] (NY). Jefferson Co.: P.O. Pine Bluff, | Oct 1942, Demaree 24038 (NY). Lawrence Co.: US. 62 between Imboden & Ravenden, 10 Apr 1963, Robinson 2324 (NY). Miller Co.: Texarkana, 17 Oct 1894, Letterman s.n. (BM). Yell Co.: around Mt. Nebo, 30 Aug 1939, Demaree 20545a (MO, NY). FLORIDA: Leon Co.: near Tallahassee, Berr s.n. (NY). GEOR- GIA: Fulton Co.: College Park, Ga., 10 Oct 1963, Schallert 573 (C). Walker Co.: Lula Falls at Lookout Mt., 28 Aug 1883 (NY). ILLINOIS: Jackson Co.: 14 Sep 1878, French sn. (E). 2.6 mi § of Gorham, 26 Sep 1956, Thieret 2760 (F). Saline Co.: 2 mi NW of Herod, 7 Oct 1949, Fuller 14927 (F). INDIANA: Crawford Co.: 4 1/2 m1 SE of Taswell, 1 Sep 1938, Tryon, Jr. 4194 (BRIT). KENTUCKY: Rowan Co.: Re. 1274 S of Clearfield, 16 Sep 1987, Cusick & Hammer 27056 (NY). Warren Co.: Green River, Young’s Ferry, 10 Sep 1892, Price s.n. (MO). MISSISSIPPI: Jackson Co.: Ocean Springs, Miss., 17 Oct 1898, Tracy 4761 (F, MICH, NY). MISSOURI: Barry Co.: T22, R27W, NW 1/4 Sec. 35, 8 Sep 1979, Hornberger 842 (MO). Butler Co.: near Rombauer, 2 Sep 1938, Steyermark 6414 (F, MO). Carter Co.: Clubhouse, | 1 Sep 1897, Trelease s.n. (MO). Christian Co.: 21 Sep 1905, Bush 3337 (GH, MO, NY). Douglas Co.: T25N, R13W, Sect. 9, 27 Oct 1982, DeLozier 781 (MO). Dunklin Co.: Campbell Mo., 14 Sep 1893, Bush s.n. (MO), throughout SE Mo., 24 Nov 1892, Bush s.n. (MO). Ozark Co.: near Tecumsen, 8 Oct 1927, Palmer 32942 (GH). Ripley Co.: Little Black River between Greenville ford and Pennington ford, | Sep 1946, Steyermark 63964 (F). Shannon Co.: 3 mi SW of Midridge, T3ON, R3W, Sect. 10, 27 Sep 1936, Steyermark 20155 (MO). NEW JERSEY: Hudson Co.: near the Hudson river at the base of the Palisades, 8 Nov 1936, Beads s.n. (GH, PH); Palisades at Coytesville, 3 Sep 1916, Wiegmann 1709 (GH). NORTH CAROLINA: Catawba Co.: 5 mi ESE of US 64-70, on Startown Rd., 14 Sep 1978, Solomon 3956 (MO). Forsyth Co.: 14 Sep 1941, Schallert s.n. (NY). Macon Co.: 5 mi § of Franklin, 22 Aug 1936, Correll 6678 (GH). Moore Co.: 7.8 mi E of Carthage, 24 Oct 1936, Correll 7001 (MICH). Orange Co.: 3 mi E of Hillsboro, 29 Sep 1939, Blomquist 10945 (NY, PH). Robeson Co.: 4.5 mi SSW of St. Pauls along Co. 1765, 10 Oct 1964, Britt 3064 (C, GH, MICH, NY). Rowan Co.: S bank of Yadkin River, E of U.S. 29 bridge, 11 Oct 1956, Horton 522 (NY). Stokes Co.: 3 mi SW of King, Radford 41268 (MICH); Cascades, 22 Aug 1938, Blomquist 10,423 (F); 2.2 mi N of Moores Springs, 2 Oct 1958, Radford 41382 (BRIT). Yadkin Co.: 6 mi E of Yadkinville, highway 421, 6 Sep 1937, Blomquist & Anderson 9870 (GH). OKLAHOMA: Pushmataha Co.: 1 mi N and 1/2 W of Honobia, 9 Aug 1948, Waterfall 8532 (GH). SOUTH CARO- LINA: Anderson Co.: Anderson, 24 Sep 1919, Davis s.n. (MO). Greenwood Co.: Green- wood, 22 Oct 1913, E.B.B. 3310 (PH). Lexington Co.: US. 378, 4 mi W of Lexington, 5 Oct 1957, Radford 29876 (GH). Orangeburg Co.: Eutawville, 9 Sep 1939, Godfrey 8198 (F, GH, MO, NY, PH). TENNESSEE: Anderson Co.: 30 Sep 1934, Jennison 3336 (PH). 562 Sipa 16(3) 1995 Blount Co.: 4 mi N of Tallahassee ¢ alongside U.S. 129, 17 Sep 1964, Thomas 33216 (BRIT). Campbell Co.: near Norris Lake basin, 30 Sep 1934, Underwood 1378 (NY). Carroll Co.: 2 mi SE of Hollow Rock Jc., 27 Aug 1922, Svenson 452 (GH). Cocke Co.: between Paint Rock and Del Rio, 12 Sep 1897, — 939 (MO). Grainger Co.: Thorn Hill, 31 Aug 1847, Moldenke 19347 (NY, PH). Knox Co.: Cherokee Bluffs, Knoxville, 18 Sep 1928, Anderson 1098 (GH), top of hill at Mr. Beane place out beyond U.T. farm, 7 Oct 1937, (MO). Macon Co.: few mi E of Beech Bottom, 14 Oct 1968, Rogers 42914 (NY). Mont- ea Co.: Adams, Sep 1927, Piaa s.n. (GH). TEXAS: Gregg Co.: summer of 1941, York s.n. (GH). Rusk Co.: 6.5 mi NW of Tatum, 14 Oct 1962, Corre// 26303. VIRGINIA: Albemarle Co.: 7 mi SE of the Monticello mansion, Wrebofdt M—675 (PH). Brunswick Co.: Rattlesnake Creek, at old Clippers Mill, $.W. of Triplett, 13 Sep 1944, Fernald & Lewts 14690 (GH, PH). Dickenson Co.: 28 Aug 1942, Carr 1070 (GH). i oo baldwinii eae Syst. Veg. 1:282. 1825. Saccharum strictum Baldw.) Nuct., Gen. Pl. 1:60. 1818, non Sprengel (1815). Erianthus strictus Baldw., in Elliott, Sketeh Bot. S. Carolina 39.1816. Type: GEORGIA. Savannah, Boldin 5.m. (HOLOTYPE: PH!). Fig Pollinia dura Trin., Acad. St. See Mem. VI. Sci. Nat. 2:91. 1836. Andropogon durus (Trin.) Steud., bom Bor. ed. 2. 1:91. 1840. Type: CAROLINA (not located) Rhizomes with aes nodes. Flowering culms 9-18 dm tall. Nodes glabrous or munutely pubescent (white hairs ca. 0.5 mm long). Internodes spongy or hollow; neither viscid nor glaucous. Leaves with sheath auricles. Auricles 0.5—1 mm long. Sheaths not ciliate. Ligule 1-3 mm long. Collar not differentiated; glabrous. Leaf blades 18—60 cm long; 5-12 mm wide; glabrous on the upper surface. Peduncle 30-40 cm long; glabrous. Inflo- rescence linear; 10-25 mm wide. Lowermost inflorescence node smooth. Main axis 10-35 cm long; g are or sparsely hairy. Primary branches appressed to the main axis; 6-18 cm rong glabrous; the margins glabrous. Rachis internode 3-5 mm = 0.3-0.4 mm wide; glabrous. Pedicels 3—5 mm long; glabrous. Callus glabrous or hairy. Callus hairs straw-colored; 0— 2 mm long; shorter than the spikelet. Spikelets brown: 7-10 mm long; 1.I-1.5 mm wide. Firse glume 5-nerved; scabrous. Second glume 3-nerved. Lemma of lower floret 6-8 mm long; 2-nerved. Upper floret 0.9—1 times the length of the lower floret. Lemma of upper floret 5.5—8 mm long; 3- nerved; entire. Awn of upper lemma 17—24 mm long; basally terete; not basally spiraled. Lodicules with nerves extending into hair-like projections. Nuttall (1818) was evidently unaware of Sprengel’s (1815) earlier name in Saccharum. Since North American authors consistently placed this taxon in Ertanthus, a nomenclatural conflict has not existed. Acceptance of Saccharum, in the present study, necessitates the use of Sprengel’s (1825) combination. Saccharum baldwinii occurs throughout the southeastern United States, but is rare or completely absent from higher elevations of the Appalachians. Its western geographic limit is eastern Texas and Arkansas, and the north- 563 WEBSTER AND SHAW, North American species of Saccharum == 2 =——-= —_=——. ———= ———= ==> —_ = = a= ——_ ———, ———— = = — ————— — Ss —_—S—S-= SS rs SS ae — —— ee Ss == Fic. 2. Saccharum baldwinii. A. Habit, B. Ligule, C. Spikelet pair, D. First Glume, E. Upper Lemma and awn. 564 Sipa 16(3) 1995 ern limit is southern Missouri, Tennessee, to northern Virginia. It com- monly occurs in sandy shaded river and stream bottoms. Other members of the complex inhabit open wet or seasonally wet areas as well as forest mar- gins where moisture is not a limiting factor. Because of these limiting eco- logical factors, 8. baldwinii is not as common as other members of this complex. Vegetative characters that are important in distinguishing S. baldwinii include culm height, absence of culm hairs, and auricle shape. Typically, specimens of S. ba/dwinii are much shorter than other members of the complex. In addition, the culm internodes are not thick and the leaf blades are narrow. The most interesting vegetative characteristic is the pres- ence of minute sheath auricles which are ovate and measure about | mm in length. Significant characteristics of the inflorescence include the complete lack of pilose hairs, a narrow inflorescence, and homomorphic spikelets. The narrow inflorescence is correlated with the lack of hairs. That i is, the long inflorescence hairs of S. alopecuroidenm and S. gigantenm result in a wide inflorescence. Significant spikelet characters include glabrous spikelets and a straight upper lemma awn. The most morphologically similar species to S. baldwinii is §. coarctatum. Representative specimens examined. ALABAMA: Autauga Co.: | mi NE of Autaugaville, 24 Sep 1934, Harper 3208 (GH, MO, NY, PH). Tuscalosa Co.: Hurricane aa 24 Oct 1908, Harper 136 (FR, GH, MO, NY). ARKANSAS: Clay Co.: P.O. Corn- ing, 8-25-1939, Demaree 20330 (MO, NY); N Knobel, 21 Aug 1896, Eggert s.n. (SC, MO). ees id Co.: near McNab, 4 Oct 1923, Greenman : 385 (MO). Lonoke Co.: Carlisle, 3 Oct 1931, Demaree 8401 (GH, MO, NY, BM). Miller : Texarkana, Pine Woods, 20 Oct 1894, Litterman 9 (MQ). Pulaski Co.: Sep 1886, ie Ja. (NY); Aug 1885, Hasse sn. (F). FLORIDA: 1879, Curtiss 5.2. (F). Baker Co.: old field and pasture, Sep 1920, King & McRainey 2695 (PH). Columbia Co.: Osceola National Forest, 2 Nov 1981, Corre// 53099 (NY). Gadsden Co.: N of US Hwy 90 bridge, Ochlockenee River, Kra/ 1641 (GH). Mana- tee Co.: Palmetto, Orange Bend, 23 Sep 1907, Chase s.n. (US). Monroe Co.: Ponce de Leon, 7 Oct 1901, Curtiss 6936 (GH, MO, NY, US). Walton Co.: 6.7 mi S of Ponce de Leon, 27 Sep 1957, Kral & Godfrey 5991 (GH). GEORGIA: Berrien Co.: SW of Tifton, 29 Sep 1902, Harper 1091 (EF, GH, MO, NY, US). McIntosh Co.: Cox, 10 Aug | ssi ee SSM. (PH). Richmond Co.: Augusta, 6 Feb 1973, Baldwin s.n. (PH). Upson Co.: spurs at Pasley Shoals of Flint R., 9 Sep 1947, Piedmont 4674 (GH, MICH, a ae ANA: Allen Parish: 10 mi W ot Kinder 20 Oct 1940, Brown & Nyland 8710 ae mi E of junct. with La. 26, ca. 3 mi E of Elton, 12 Nov 1960, Reese 3986 (F). Cameron Parish: Oberlin, 9 Sep 1898, Ba// 207 (GH, NY). East Baton Rouge Parish: 5 mi E of Harelson store on Hanel’s Ferry road, 9 Oct 1927, Brown 1927 (NY, US); Elton, 12 Nov 1960 Reese 3986 (F, GH). Evangeline Parish: 4 mi E Barber on La. Bias: 10, Bayou Ne cee LO Jun 1960, Ewab 19985 (US). Franklin Parish: 4.7 mi NNE of Winnsboro, 19 § re Shinners 24,653 (GH). Livingston Parish: 4 1/4 mi S of Denham Springs. tee 7 Sep 1966, Thieret 24813 (US). Ouachita Parish: Sec. 31 TITN R 4E, 3 Oct 1988, ae 107,586 (MO). Rapides Parish: Alexandria, Hale s.n. (PH, US). St. Teaiaay Par- ish: Convington, 1832, Drammon s.n. (BM); Slidell, 5 Oct 1891, Langlois sin. (NY); Covington, Sulphur Springs, Oct 1919, Arsene 11566 (US): 3 mi from Covington, 24 Sep WEBSTER AND SHAW, North American species of Saccharum 565 1913, 5.2. (US). gate Parish: E of Robert, 23 Oct 1946, Brown & Bell 8800 (GH). Union Parish: Sec. 20 T2ON, R3E, 28 Sep 1987, Thomas 102,051 (NY). Vernon Parish: Oberlin, 9 au 1898, Ball 207 (US); along Devils Creek, section 4. ca. 5 mi NNW of Temple, 30 Oct 1966, Thieret 25199 (US). Winn Parish: along La. 501, 1.4 mi S of Mill, 20 Sep oa 1893 (L). MISSISSIPPI: Harrison Co.: Nicholson, along Pearl River, 11 Oct 1896, Kearney Jr. 365 (NY); Biloxi Kashtaw, 13 Oct 1898, Tracy 4672 (GH, NY). ee Co.: Vancleave, Ocean Springs, 24 Sep 1953, Demaree 34378 (US). Pearl River : 2 mi SW of Picayune, 27 Sep 1966, Sargent 8929 (MO); 2 mi W of Picayune, 19 Sep es Sargent 9467 foes seaitiiatenr ee Co.: Campbell, 7 Oct 1910, Bush 6384 (GH, MO, NY, US). NORTH CAROLINA: Anson Co.: 3 mi S of Ansonville, 18 Sep 1950, Boyce 1501 (NY). Columbus | a River Bridge on Co. Rt. 1928, 23 Oct 1968, Leonard & Radford 2203 (BM, BR, ISC, MICH). Cumberland Co.: along stream and low woods, 16 Aug 1932, Blomquist 334 (US), 3 mi E of falcon, Rt. 102, 14 Oct 1951, Boyce & Fox 1682 (ISC, US). Harnett Co.: near Dunn, 6 Oct 1933, Blomquist 728 (F). Pender Co.: Burgaw, | Sep 1938, Godfrey 6489 (GH, US); edge of Angola Bay Highway 53, 16 Sep 1937, Blomquist 10,088 (GH, NY, PH). SOUTH CAROLINA: Georgetow Co.: 5 mi S of Georgetown, 9 Sep 1939, 7 8127 (F, GH, MO, NY, US). Dine ous Co.: 8-19-1905, Hitchcock 236 (F, GH MO, NY, US). Williamsburg Co.: 1 mi E of Greeleyville, 22 Aug 1957, Radford ee (NY): 1.8 mi NE of Greeleyville, 19 Oct 1957 Radford 31158 (GH). TENNESSEE: Carroll Co.: Hollow Rock Jc., 27 Aug 1922, oo 463 (GH). Coffee Co.: 15 Oct 1880, Gettinger s.n. (ISC, MO); Jul 1886 (NY). Davidson Co.: Nashville, 162 (NY). Madison Co.: Jackson bottoms, Sep 1892, Bain s.n. (US); Tullahoma, Sep 1882, Curtiss s.n. (GH, US); Tullahoma, Sep 1879, Gattinger 3629 (BM, F, ISC, MO, NY, US); SE side of Manchester near jct US 41, 20 Aug 1992, Kral 43,672 (BM); Oak barrens N of Manchester, 6 Aug 1938, Svenson 8934 (US); Sep 1892, S.M.B. 198 (NY). TEXAS: Anderson Co.: along Catfish Creek at Palestine, 19 Sep 1971, Hatch 1059 (MO); 6 mi NW of Tennessee Colony, 22 Oct 1983, Hatch 5080 (MICH). Bowie Co.: Texarkana, 20 Oct 1894, Letterman s.n. (NY, BM, PH). Harris Co.: Houston, 1842, Engel- mann s.n. (GH); Houston, Jun 1841, Lindheimer s.n. (MO); Houston Co.: Grapeland, 22 1917, Palmer 12823 (NY). aia Co.: McFadden Beach, 5 Oct 1934, Cory 11021 (en. Forest Lawn Cemetery, Beaumont, 3 Oct 1945 Cory 50010 (GH, MICH, NY, US); S of Beaumont, 30 Sep 1940, Si/veus 6444 (US). Liberty Co.: 2.5 mi E of Rye on Hwy 105, 24 Sep 1966, Gould 121005 (US). Upshur Co.: between S.L. & S.W. railroad and the Gilmer to Big Sandy Hwy, 28 Nov 1941, Moon 110 (US); 3.3 mi S of New Diana, 15 Sep 1953, Shinners 15987 USC). VIRGINIA: Dinwiddie Co.: W of Winfield’s Mill, 13 Oct 1941, Fernald & Long 13885 (GH, PH). Greenville Co.: N of Emporia, Three Creek 19 Sep 1938, Fernald & Long 9241 (GH, PH). Prince George Co.: SE of Disputanta, 7-20-1938, Fernald & Long 8581 (F, GH, MO, PH, US). Southampton Co.: E of Drewryville, Terrapin Ridge, 21 Aug 1938, Fernald & Long 8917 (GH, PH); sandy alluvia boctomlands of Three Creek, 14 Sep 1941 Fernald & Long 13550 (GH). Sussex Co.: SE of Waverly, 10 Sep 1937, Fernald & Long 7299 (GH, NY, PH, US); 2 mi E of Stony Creek, 24 Aug 1938, Fernald G Long 8918 (GH, PH). York Co.: NW of Grafton, 18 Sep 1937, Fernald & Long 7300 (GH, PH, US). Saccharum brevibarbe (Michx.) Pers., Syn. Pl. 1:103. 1805. Erianthus brevibarbis Michx., Fl. Bor.-Amer. 1:55. 1803. ane alopecir cides (L.) Elliott var. brevibarbis (Michx.) Chapm., Fl. South. U.S. 583. 1860. Erianthus saccharoides Michx. subsp. brevibarbis (Michx.) Hack., Monogr. ae : 131. 1889. Type: TENNESSEE and CAROLINA, Michawx 5.n. (HOLOTYPE: P, photograph seen). Fig. 3. 566 Sipa 16(3) 1995 Rhizomes with compacted nodes. Flowering culms 8—25 dm tall. Nodes glabrous or hairy. Internodes hollow. Leaves without auricles. Sheaths not ciliate. Ligule 1-2 mm long. Collar not differentiated; glabrous. Leaf blades mostly 40—60 cm long; 7-25 mm wide; glabrous. Peduncle mostly 45-75 cm long; glabrous (occ. pubescent or minutely pilose). Inflorescence linear or oblong; mostly 4-10 cm wide. Lowermost inflorescence node smooth. Main axis (10—)30—50 cm long; glabrous or sparsely hairy; pilose. Primary branches aes to the main axis; typically 7-14 cm long; hairy; ciliate. Rachis internode 4—6 mm long; 0.4—0.5 mm wide; hairy. Eales 3-4 mm long; hairy. ie hairy. Callus hairs white to straw-colored; 3—G.5 mm long; shorter than the spikelet. Spikelets heteromorphic; ok or straw-colored; 6.5—10.5 mm long; 1.2—1.5 mm wide. First glume 5-nerved; smooth (scabrous at the apex). Second glume indistinctly 5-nerved. Lemma of lower floret 5.5—-8 mm long; nerveless. Upper floret 0.9-1 times the length of the lower floret. Lemma of upper floret 5.5—7.5 mm long; 3- nerved; entire or bifid. Lateral lobes of upper lemma 2—2.5 mm long. Awn of upper lemma 10—22 mm long; basally flattened; basally spiraled or not basally spiraled. Lodicules with nerves extending into hair-like projections or not extending into hair-like projections. Saccharum brevibarbe occurs throughout the southeastern United States. Its western geographic limit is the pineywoods of east Texas and Okla- homa, with the northern limit from Tennessee to Delaware. Diagnostic features include the presence and relative length of the callus hairs, length of the upper lemma awn, and whether the awn is basally flattened. It is most similar to S. coarctatum, but can be distinguished by the above-stated characters. Ertanthus smallii Nash was described to account for specimens with long (8-10 mm) spikelets and dense hairs associated with the apex of the peduncle and main axis. Examination of type material and numerous similar specimens show continuous variation of these characters; therefore, E. smatti 1s placed as a synonym. The typical variety is common in central and southern Arkansas, eastern Oklahoma, the pineywoods vegetation region of eastern Texas, and north- ern Louisiana. Relatively few collections were found in Mississippi, Ala- bama, and Tennessee. One collection was found in a coastal county of North Carolina. Saccharum brevibarbe var. contortum (El liott) R. Webster occurs throughout the range previously given for the species. Both varieties occur in similar habitats. They commonly occur in clay or loamy open or mat- ginal sites chat are prone to seasonal flooding; however, seasonal flooding is not a critical factor since this species frequently inhabits well-drained clay slopes. The varieties differ only in morphology of the lemma awn and corre- WEBSTER AND SHAW, North American species of Saccharum 567 ae aN i iis nh ae . Hist i) nN , / i Fic. 3. Saccharum brevibarbe var. contortum. A. Habit, B. Ligule, C. Spikelet Pair, D. First Glume (upper spikelet), E. First glume (lower spikelet), F Upper lemma and awn. 568 Sipa 16(3) 1995 lated characters. In variety contortum the awn is relatively long, basally flat, geniculate, and tightly coiled or spiraled at the base. Typically 2 to 4 (usu- ally 3) tight spirals occur at the base and 2—4 loose spirals occur above the base. Early in inflorescence development, prior to release from the upper sheath, the awn is not coiled, the apex of the upper lemma is not bifid, and lateral lobes are not present. Once the inflorescence is released from the sheath and as the spikelets mature, the awns quickly develop the spiraling morphology described above. Development of the basal spirals results in tearing the lemma apex from the central awn and producing the lateral lobes characteristic of variety contortum. In S. brevibarbe var. brevibarbe the awn is usually shorter, basally fat, but not tightly spiraled at the base. As a result the lateral lobes are not present. Absence of tight spirals and lateral lobes may be the result of a relatively shorter awn. The status and interpretation of E. brevibarbis Michx. (= §. brevibarbe var. brevibarbe) has been the source of taxonomic confusion. This problem was addressed in detail by Fernald (1943), who failed to compare this taxon to E. contortus Baldw. (= S. brevibarbe var. contortum). Fernald recognized that Michaux’s type of E. brevibarbis was morphologically different from the specimens to which the name was being applied. For these specimens he provided the new name, E. coarctatus Fernald. The name E. brevibarbis was restricted to the type and only one additional specimen (Demaree 8228). That concept was followed by Hitchcock (1951). Michaux’s type was clearly illustrated in Fernald (1943) and an origi- nal photograph was examined for the present study. Spikelets from the type exhibit callus hairs shorter than the spikelets and the upper lemma awn slightly coiled at the base. Numerous specimens were found showing the same features. These specimens are otherwise identical to specimens previously treated as E. contortus Elliott. Intermediate specimens between the taxa are not uncommon. Combining these taxa as varieties of S. brevibarbe was considered the most appropriate way to reflect the biological relationship. KEY TO THE VARIETIES Awn of upper lemma basally spiraled; lemma of upper floret bifid with lateral lobes about 2.0 mm long S. brevibarbe var. contortum Awn of upper lemma not basally spiraled; lemma of upper floret entire S. brevibarbe var. brevibarbe Saccharum brevibarbe (Michx.) Pers. var. brevibarbe Lemma of upper floret entire. Awn of upper lemma 10-18 mm long; not basally spiraled. Representative specimens examined. ALABAMA: Russell Co.: 4.5 mi SE Hurtsboro, WEBSTER AND SHAW, North American species of Saccharum 569 12 Sep 1968, Kra/ 33291 (BRIT). ARKANSAS: Ashley Co.: P.O. Hamburg, 27 Sep 1937, Demaree 16386 (BRIT, MO). Clark Co.: P.O. Okolona, 11 Oct 1939, Demaree 20713 (BRIT). Drew Co.: low waste meadows, P.O. Monticello, 8 Oct 1938, Demaree 18287 (BRIT, MO). Garland Co.: Hot Springs National Park, 6 Sep 1934, Demaree 11017 (BRIT). Howard Co.: mineral springs, 14 Oct 1932, Demaree 9699 (BRIT, GH); mineral springs, 14 Oct 1932, Demaree 9744 (BRIT, MO, NY, US). Lafayette Co.: P.O. Lewisville, 19 Oct 1959, Demaree 41930 (BRIT). Pike Co.: Murfreesboro, 28 Sep 1932, Demaree 9362 (BRIT, GH, MO, NY); near Tokio, 22 Oct 1932, Demaree 9939 (BRIT, GH, MO, US). Polk Co.: 20 mi N onde Queen, 30 Nov 1958, Van Schaack 3625 (MO). Pulaski Co.: Pulaski Hts., Little Rock, 21 Sep 1031, Demaree 8228 (BRIT, GH, MO, NY, US). celine Co.: P.O. Benton, 6 Sep 1942, Demaree 23959 (BRIT, MO). Union Co.: El Dorado, 4 Oct 1953, Hotberg 371 (BRIT); 10 mi NE El Dorado, 7 Oct 1953, Hoiberg 378 (BRIT). LOUISIANA: Caddo Parish: about 3 mi S of Longwood, 3 Oct 1965, Thiret 21194 (US). Lincoln Parish: 2 mi W ee 14 Sep 1970, Hil! 28 (MO). Ouachita Parish: La. 557 just N of Cypiess turnoff W of Luna, 7 Oct 1985, Thomas 93,843 (MO). MISSISSIPPI: Copiah Co.: P.O. Barlow, 18 Sep 1954, Demaree 36190 (BRIT, US). Jackson Co.: Ocean Springs, 23 Sep 1898, Tracy 4540 (MICH, MO, US). O Olubbehaco: Starkville, 1893, Tracy 2228 (US); near Starkville, 27 Sep 1896, Kearney Jr, 29 (MO), near Starkville, 29 Sep 1 896, Kearney Jr. 56 (US). NORTH CAROLINA: Beaufort Co.: 1.4 mi NE of Washington, 12 Oct 1958, ele 42167 (NY). OKLAHOMA: Lefore Co.: near Page, 8 Sep 1913, Stevens 2662 (GH, NY, US); Sec. 9-10, T. 3S., R. 25 E, State Game Preserve, 2 Jul 1930, gale 501 (US); 1 near Bokhom 11 Oct 1937, nae 44090 (MO). TEXAS: AN Co.: Texarkana, Oct 1894, nae SM. hae NY, PH, US); near Texarkana, Aug 1883, ce man s.n. (MQ); S of Dalby Cass Co.: Hwy 11 E, = ypress Creek, 25 Sep 1948, ae 20291a (BRIT). Harrison Co.: 3.2 mi of Marshall, 31 Oct 1953, Shinners 16766 (BRIT, ISC). Walker Co.: 9 1/2 mi N of Huntsville, 29 Sep 1934, Cory 10356 (GH). Saccharum oo (Michx.) Pers. var. contortum (Elliott) R. Webster, comb. nov. Erianthus contortus Elliott, Sketch Bot. S. Carolina 1:40. 1816. nae contortum fells Nutt., Gen. Pl. 1:60. 1818. Erianthus alopecuroides (L.) Elliott var. contortus (Elliote) Chapm., Fl. South. U.S. 582. 1860. Erianthus saccharoides Michx. su ae contortus (Elliott) Hack., Monogr. Phan. 6:131. 1860. Type: GEORGIA, Baldwi Calamagrostis rubra Bosc ex Kunth, Enum. PI. 1:478. 1833 (fide Hitchcock 1950). Erianthus eee Nash, N.Y. Bot. Gard. Bull. oo. 1900. Type: GEORGIA, Small s.n. (HOLOTYP Y!). Lemma of upper floret bifid. Lateral lobes of upper lemma 2—2.5 mm long. Awn of upper lemma 15-22 mm long. Awn of upper lemma basally spiraled (typically with 2-4 complete spirals). Representative specimens examined. ALABAMA: Barbour Co.: 20 Oct 1943, Koepper, pees and ae sn. (NY). Cherokee Co.: 2 mi E Leesburg on US 431, 14 Sep 1968, Kral 78 (BRIT, C). Clay Co.: jct Ala. 77 and 49 near Mellow Valley, 13 Sep 1968, Kral 7 ae Cleburne Co.: 1 mi E Piedmont on cty 70, 16 Sep 1971, Kral 44175 (MO). Cullman Co.: 12 Sep 1897, Eggert s.2. (MO, US); N Johnson, 24 Sep 1898, Eggert 5. (BM, MO). Greene Co.: NE Eutaw on Ala 14 toward Clinton, 8 Oct 1968, Kral 33894 (MO); Ala. 69, 4.6 mi N Greensboro, 20 Sep 1971, Kral 44451 (NY). Lee Co.: Auburn, 25 Aug 1897, Baker 1090 (NY); Auburn, 10 Sep 1897, Earle & Baker s.n. (NY), Auburn, 14 Oct 1900, Earle s.n. (NY). Macon Co.: 4 mi E of Tuskegee, 14 Aug 1927, Wiegand G Manning 570 Sipa 16(3) 1995 124 (GH). Marion Co.: 8 mi N of Haleyville on Ala. 5, 6 Oct 1967, Kral 29478 eae Mobile Co.: W of open Pine see Sep 1876, Modr s.n. (US). Mentone Co.: Mont- gomery, 19 Oct 1943, Isely 2979 (ISC, NY); Montgomery, Feb 1890, McCarthy 1888 G CH, PH). Shelby Co.: - mi Nj jet Ala 25 and Calera, 4 Oct 1968, Kra/l 33585 (BRIT). Tuscaloosa Co.: 12 Oct 1966, Deramus, Jobnson & N ee 902 (GH); in Hurricane Creek, 24 Oct 1908, Harper 137 (F, GH, MO, NY); 15 mi SW of Tuscaloosa along Hwy 59, 20 Aug 1981, Hatch 4607 (MO). ARKANSAS: Calhoun Co.: small bottoms, 20 Oct 1941, Demaree 22687 (BRIT, US). Drew Co.: P.O. Monticello, 12 Sep 1936, Demaree 13688 (BRIT); meandering stream bottoms, P.O. Plantersville, 30 Sep 1937, Demaree 16410 (BRIT, MO, NY); low waste meadows, P.O. Monticello, 8 Oct 1938, Demaree 18512 (BRIT, ISC, MO, US). Grant Co.: small creek bottoms P.O. Sheridan, Demaree 16548 (BRIT, MO). Logan Co.: Magazine Mt. , Sep 1947, be ae Nevada Co.: P.O. Prescott, 6 Oct 1940, Demaree 21744 (BRIT, GH, ISC O). Pike Co.: prairie creek Murfreesboro, 13 Oct 1932, Demaree 9689 (BRIT, MO). ay Co.: Pulaski Hes. Little Rock, 15 Sep 1931, Demaree 8165 (BRIT, MO, NY); W of Little Rock, Sep 1835, Engelmann s.n. (MO). Yell Co.: P.O. Aly, 19 Sep 1970, Demaree 62853 (BRIT). DELAWARE: | mi NW of Georgetown, 13 Sep 1936, Fogg, Jr. 11469 (GH, PH). FLORIDA: Jefferson Co.: 5 mi W of Monticello, 10 Oct 1957, Kral 6164 (GH). Leon Co.: near Tall: ahassee, Berg s.n. (NY); frequent in open pine-grassland, 14 Oct 1974, Godfrey 74002 (BM); 7 mi E of Concord, 7 Oct 1956, Kral sie (GH). GEORGIA: Bartow Co.: 4.8 mi E35 degree’s of eve ee 18 Sep 1951, Duncan 13140 (GH). Columbia Co.: above Little Kickee Creek, 24 Apr 1936, Leeds G Harper 2770 (PH, US). De Kalb Co.: Stone Mountain, 16 Oct 1907, Chat ee 20 CISC, US); Stone Mountain, Aug 1905, Hitchcock s.n. (US). Elbert Co.: ca 1.5 mi S of Coldwater Creek, 10 Oct 1979, Credle 2649 (NY); Atlanta, 21 Sep 1895, Lippincott 129 (PH). Gwinnett Co.: near Mcquire’s Mill, 20 Jul 1893, Hee n. (NY, US). Jackson Co.: 11 mi N of Athens, 21 Sep 1947, oe GH, PH, US). Jasper Co.: Monticello, Jul 1846, Porter som. (MO), | Co.: near Gainesville, - Aue 1 1936, Correll 6613 (GH). Muscogee Co.: moist, sandy an Columbus 7 Sep 1899, 38586 (US). Oglethorpe Co.: 9.3 mi N 45 degree E of Lexington, 23 Sep 1967, Blake 944 (PH). Putnam Co.: near Rock Eagle Lake, 30 Sep 1947, Cronquist 4763 (GH, NY). Rockdale Co.: 6 mi SW of Logansville, 18 Oct 1936, Pyron & MeV Angh 1118 (US). Wilkes Co.: 9.3 mi 8 60 degree E of Washington, 16 Oct 1949, Duncan 10634 (BM). LOUISIANA: Bienville Parish: roadside near Driskill Mtn., 27 Jul 1955, Moore 6252 (US); highway 507 near Driskill | Mountain, 17 Oct 1987, Poms 102,917 (NY). Bossier Parish: 3 mi NE of Alden Bridge, 7 Aug 1938, Correll 10129 (F, NY). Caddo Parish: 4.7 mi SW of oo. 14 Jul 1955, Shinners 20740 (BRIT). Caldwell Parish: 3 mi SW of Grayson, 7 Oct 1955, Shinners 21884 oar GH). e Soto Parish: 2 mi W of Hunter, 10 Aug 1938, Corre// 10183 > (GH, US): 4 mi SW of Mansfield, 2 Sep 1967, Thieret 27341 (US). Jackson Parish: 2 mi S of Ansley, , Sep 1955, Shinners 21203 (BRIT, GH). St. Helena Parish: 3 mi NW of Greensburg, 25 Sep 1966, Thieret 24873 (US). Ouachita Parish: 10 mi SW of Monroe, | Oct 1941, Smith s.n. (GH): Sec. 6, TION, RIE, 18 Sep 1987, Thomas 101,981 (NY). MARYLAND: Dorchester Co.: near Cambridge, Fork Neck Rd., 7 Oct 1989, Cohen | (US). Somerset Co.: Costen Station, 16 Oct 1935, Fernald, Long & ae 9575 (GH, PH); Princes Anne, Sep 1866, ¢ Canby s.n. (NY, PH). Wicomico Co.: Re. 313, 2.9 mi S of Rt. 348, Shar ptown, 12 Oct 1981, H7 mad Riefner 10851 (GH, NY); 0.2 mi W of Walsto n, 2 Oct 1983, Hill 13280 (GH ). Worcester Co.: Snow Hill, 30 Sep 1931, Moldenke 6591 (NY, US). MISSISSIPPI: ae Co.: near US. 13, 3.7 mi NW of Jason, Radford 40346 (C). Harrison Co.: Biloxi, 27 Oct 1893, Tracy 2297 (US): Biloxi 9 Oct 1898, Tracy 4670 (NY). Jasper Co.: 6 mi via rd. from Enterprise along Souinlovey Creek, 8 Sep 1967 Jones 15150 (US). Leake Co.: Natchez WEBSTER AND SHAW, North American species of Saccharum as. Trace Pkwy, 14 Sep 1947, McDougall 1494 (US). Lowndes Co.: Starkville, 12 Oct 1907, Chase 4462 (US); Starkville, 22 Oct 1893, Tracy s.n. (US); Starkville, 6 Oct 1895, Tracy s.n. (NY); Starkville, 4 Oct 1890, Tracy 5.2. (NY); pine lands Columbus, 28 Sep 1900, 5858d (US). Tishomingo Co.: 20 mi NW of Iuka, 12 Oct 1956, Ray, Jr. 7559 (GH). Wayne Co.: Waynesboro, 8 Aug 1896, Pollard 1247 (GH, MO, US). NORTH CAROLINA: Anson Co.: open dry hillside near Polkton, 25 Oct 1936, Correll 7102 (MICH, US). Beaufort Co.: 5 mi E of Washington, Hwy 264, 10 Oct 1936, Blomquist 8000 (NY). Bertie Co.: near Windsor, 15 Oct 1938, Godfrey 7002 (GH). Craven Co.: 6.8 mi SE. of New Bern, 10 Sep 1958 Radford 40150 (NY). Halifax Co.: Scotland Neck, 15 Oct 1938, Godfrey 7019 (GH). New Hanover Co.: Carolina Beach, 30 Aug 1938, Godfrey 6376 (US). Harnett Co.: near Angier, 15 Aug 1932, Blomquist s.n. (F), 2 mi N of Pine View Station, 29 Feb 1940, Walker 1412 (PH). Hyde Co.: near Leechville, 13 Oct 1938, Godfrey & White 6862 (GH). Lee Co.: pine woodland near Sanford, 14 Oct 1938, Godfrey 6921 (GH), 3.8 mi S of Harnett—Lee county line on NC 87, 29 Sep 1956, Laing 351 (US). Stokes Co.: 2.5 mi NE of Walnut Cove, 2 Oct 1958, Radford 41284 (BRIT). Pamlico Co.: near Grantsboro, 13 Oct 1938, Godfrey & White 6803 (GH). Pitt Co.: 6 mi E of Greenville, Hwy 264, 11 Oct 1936, Blomquist 8113 (PH). Polk Co.: 1 mi E of Columbus, 16 Oct 1953, Freeman 53589 (US). Tyrrell Co.: along re. 64, 3.8 mi E of Columbia, 22 Oct 1970, Terrell 4364 (US); 1.4 mi E of Columbia, along rt. 64, 22 Oct 1970, Terrell 4365 (US). Wake Co.: Raleigh, 7—25- 1930, Blomquist 44 (US); Raleigh, 10 Sep 1938, Godfrey 6600 (GH). OKLAHOMA: Lefore Co.: 20 mi N of Broken Bow, 16 Oct 1937, Hopkins & Cross 2493 (US); 2.5 mi E of Page, 13 Oct 1948, Robbins 3201 (BRIT, NY); 7 mi E of Broken Bow, 28 Aug 1938, Smith 502 (MO); 12 mi N of Bethel, 14 Oct 1951, Waterfall 10505 (US). SOUTH CAROLINA: Aiken Co.: Oct 1889, Ravenel s.n. (US). Anderson Co.: Anderson, 2 Sep 1919, Davis 1280 (F); Anderson, 26 Aug 1921, Davis 2063 (BM, MO); Anderson, 14 Sep 1919, Davis 9109 (BM); Anderson, 5 Oct 1920, Davis s.n. (US); McKinney Springs, 8 Sep 1917, Davis 5.1. (MO): between Pendleton Place apts. and landfill entrance gate, 13 Sep 1987, Hill 18791 (GH, NY). Florence Co.: jet of S.C. 237 and US 52, 17 Oct 1968, Leonard & Radford 2149 (MO), Greenwood Co.: Bradley, Sep 1920, Davis 5.n. (BM). McCormick Co.: 6 mi SW of McCormick, 18 Sep 1949, Duncan 10395 (US): 6 mi SE of Clarks Hill, 13 Oct 1957, Radford 30638 (GH). Oconee Co.: 1 Oct 1897, A nderson 1426 (GH, US). Orangeburg Co.: 16 Aug 1905, Hitchcock 234 (BM, C, F, GH, NY, PH, US). Union Co.: Carlisle, 11 Jul 1906, House 2478 (US). TENNESSEE: Hardin Co.: ca. 15 mi SE Maddox, 13 Sep 1971, Kral 43905 (MQ). Polk Co.: Hiwassee Valley, 30 Sep 1893, Ruth s.n. (CUS); Hiwassee River, Aug 1895, Ruth s.n. (MO), Hiwassee River, Aug 1894, Ruth s.n. (NY). McNairy Co.: Pine—oak forest W of Ramer., 16 Oct 1949, Woods 14654 (US). TEXAS: Bowie Co.: Springs, 20 Oct 1961, Correll & Correll 24771 (GH). Camp Co.: Pittsburg, 13 Sep 1923, Tharp 1970 (US). Gregg Co.: 27 Sep 1941, York s.n. (GH, MO). Henderson Co.: 3 mi E of Athens, 9 Nov 1962, Corre// 26699 (MO). Hopkins Co.: 7.8 mi of Sulphur Springs, 11 Sep 1949, Turner 1419 (BRIT). Houston Co.: 3 mi SE of rt. 21, | Dec 1962, Correll 26906 (NY). Leon Co.: 1/4 mi E of Keechi, 5 Oct 1937, Cory 25229 (US). Morris Co.: 2 mi W of Omaha, 17 Oct 1962, Correll 26317 (NY). Rusk Co.: Texas, 1884, Nealley s.n. (US). Smith Co.: western Tyler, N of Pine Burr Road, 18 Aug 1949, Cory 56858 (BRIT, US); 8 mi NE of Tyler, 10 Oct 1943, Moore, Jr. 545 (GH, US); glades in oak—hickory woods Amigo, 14 Oct 1945, Moore Jr. 1053 (BM, C, F, GH, ISC, MICH, NY, PH, US); near Gumwood, 27 Sep 1926, Palmer 31751 (US). Titus Co.: 6 mi SW of Mt. Pleasant, 23 Sep 1971, Amerson 733 (BRIT). Upshur Co.: 3.3 mi $ of New Diana, 15 Sep 1953, Shinners 16020 (BRIT, ISC). VIRGINIA: Norfolk Co.: near Norfolk, 10 Nov 1890, Blanchard s.n. (F); 19 Jul 1898, Kearney 1741 (US), Virginia Beach, 22 Sep 1900, Williams 3112 (GH, US). Accomack 572 Sipa 16(3) 1995 Co.: Oak Hall, 16 Oct 1935, Fernald, Longe & Fogg, Jr. 5576 (GH, PH); 1882, Mears s.n. (US). Charles City Co.: near Malvin Hill, 26 Sep 19306, Erlanson 134 (US); near Malvin Hill, 26 Sep 1936, Erlanson 163 (US). Henrico Co.: W of Elko Station, 21 Sep 1938, Fernald & Long 9243 (F, GH, PH); Richmond, 1876, Wood et al 2509 (US). Greensville Co.: E of Slagle’s Pond, N of Emporia, 20 Sep 1938, Fernald & Long 9245 (GH, PH). Isle of Wight Co.: N of Windsor, 13 Jul 1938, Fernald & Long 8580 (GH, PH). James City Co.: 20 mi W of Williamsburg, Highway 60, 31 Aug 1956, Meyer s.n. (MO). Nansemond Co.: Factory Hill, 26 Aug 1936, Fernald & Long 6457 (GH, PH); 2 mi SE of Cleopus, 15 Oct 1938, Fernald & Long 9507 (GH, MICH, PH, US); Portsmouth, 9 Sep 1895, Noys 48 (US). Norfolk Co.: near Ocean View, 11 Nov 1898, Kearney, Jr. 2398 (US). Northampton Co.: Meadows, 10 Oct 1891, Canby s.n. (NY); Belle Haven, 8 Sep 1935, Fogg, Jr., 9726 (GH, PH). Southampton Co.: Swale, Courtland, 11 Sep 1937, Fernald & Long 7302 (GH, PH), Saccharum coarctatum (Fern.) R. Webster, comb. nov. Erianthus coarctatus Fernald, Rhodora 45:246. pl. 758. 1943. Type: VIRGINIA, Fernald & Long 7301 (HOLOTYPE: US!). Fig. 4. Ertanthus coarctatus var. elliottianus Fernald, Rhodora 45:246. 1943. Type: FLORIDA, Curtiss 6940 (HOLOTYPE: US!), Rhizomes with compacted nodes. Flowering culms 10—25 dm tall. Nodes hairy (che hairs ca. 1-3 mm long). Internodes spongy or hollow. Leaves with sheath auricles. Auricles 0.3—1 mm long. Sheaths not ciliate. Ligule |-2 mm long. Collar differentiated or not; glabrous or hairy. Leaf blades L5—40 cm long; 7-12 mm wide; glabrous. Peduncle 35—45 cm long; gla- brous. Inflorescence linear or oblong; 30-70 mm wide. Lowermost inflo- rescence node smooth. Main axis 13-35 cm long; glabrous or sparsely hairy; pilose. Primary branches appressed to the main axis: 5—12 cm long; hairy; ciliate. Rachis internode 3-6 mm long; 0.3—0.4 mm wide: hairy. Pedicels 3-5 mm long; hairy. Callus hairy. Callus hairs white or straw-colored hairs; 3—5 mm long; shorter than the spikelet. Spikelets homomorphic or hetero- morphic; brown; 6—8 mm long; 0.9-1.2 mm wide. First glume 5-nerved; smooth or scabrous. Second glume 3-nerved. Lemma of lower floret 5.8— 7.5 mm long; 3-nerved. Upper floret 0.7—0.8 times the length of the lower Horet. Lemma of upper floret 4—5.5 mm long; 3-nerved; entire. Awn of upper lemma mostly 16-26 mm long; basally terete; noc basally spiraled. Lodicules with nerves extending into hair-like projections. Saccharum coarctatum is common on the coastal plain of the southeastern United States from Louisiana to Florida and north to Delaware. Species such as S. alopecuroideum, S. gigantenm, and S. brevibarbe occur commonly in the central part of the southeast. Saccharum coarctatum is restricted to the lower elevations of the sandy coastal plain. Its northern geographic limit is southern Louisiana, Mississippi, Alabama, Florida, and Georgia. On the east Coast it is Common in eastern Georgia, South Carolina, North Caro- lina, Virginia, Maryland, and Delaware. Saccharum coarctatum prefers open, moist, and loamy habitats, which are seasonally flooded. Diagnostic veg- WEBSTER AND SHAW, North American species of Saccharum 373 i fh N ry" | H, ae hm sid | ld N AN iN I (s ( hy’ \ Fic. 4. Saccharum coarctatum. A. Habit, B. Ligule, C. Spikelet Pair, D. First Glume, E. Second Glume, F. Upper Lemma and Awn. S74 Sipa 16(3) 1995 etative characters for 8S. coarctatum include relatively narrow leaf blades narrow culm internodes, and essentially glabrous culms. These a are similar to those of 8S. baldwinii, which is best Sep arated from S. coarctatum on the presence of hairy nodes; however, the nodes of S. coarctatum fre- quently become glabrous at maturity. The most reliable distinguishing characters for S. coarctatum are the straight non-spiraled awn of the upper lemma and callus hairs that are about half the length of the spikelet. An interesting characteristic of this species is that the nerves of the lodicules extend into hair-like projections that measure about 0.6 mm long. Fernald (1943) described the taxon, Erianthus coarctatus var. Sir Fern. and distinguished it from the typical variety on the basis of its rela- tively wider culm internodes, higher culm height, wider leaf blades, longer inflorescence main axis, and a wider inflorescence. When the variation among all specimens is considered, these characters intergrade with the typical variety. Therefore, no formal taxonomic rank was given to these somewhat — larger specimens. Representative Lee eas ALABAMA: Covington Co.: pine savanna 9.5 mi S Opp., 5 Oct 1968, ie 9 (MO). Escambia Co.: 0.5 mi E Canoe, 7 Oct ne Kral 33888 (MO). Mobile Co. . miS Theodore on Ala. 59, 8 Oct 1967, Kra/ 29731 (US), DELAWARE: Sussex oy es sandy soil, Sep 1894, Canby s.n. (PH); 5 mi W of ae on road to Laurel, 12 Oct 1898, Canby, : ke G& Muir sn. ne Ellendale, 25 Sep 1873, Commons s.n. (NY, PH, US); fencerow, 1/4 mi E of Ellendale, 12 Oct 1940, Tatnall 4745 (GH, PH); few mi W of Ellendale, 06 Nov 1961, Terre// & Thornton 3625 (US). FLORIDA: Alachua Co.: 3 mi W of Gainsville, 5 Nov 1938, Swa/len 5565 (US). Suwannee Co.: Live Oak, 9-17 —1900, Curtiss 3858 (US): near Live Oak, LO Oct 1901, Cvrtiss 6940 (NY, US). Union Co.: W of Lake Butler, 26 Oct 1940, Sifvens 6746 (US). GEORGIA: Berrien Co.: base of sand—hills of Little River, SW of Tifton, 29 Sep 1902, oe 1693 (BM, GH, MO, NY, US). McIntosh Co.: 2.3 mi W of S tip of Blackbeard Island, 17 Oct 1956, Duncan 0662 (ISC, US). Tift Co.: Brookfield, 9-1978, Quvarim 3605 (C, MO). LOUISIANA: Calcasieu Parish: Oberlin, 8 Sep 1898, Ball 194 (NY, US). Rapides Parish: pears. Hales.n. (PH). MARYLAND: Dorchester Co.: Robinson’s Neck, ce or’s Island, Oct 197¢ Reese s.n. (US). Somerset Co.: 5 mi N of Princess Anne, 2 Oct 1937, Tatnall 3574 es H). Wicomico Co.: 3 mi SW of Salisbury, 4 Sep 1937 ie 5364 (PH). NORTH CAROLINA: Brunswick Co.: S of Wilmington, 8 Oa 1933, Blomguist 721 (NY); ere 8—1905, Hitchcock s.n. (US). Columbus Co.: old Dock, 8-29-1938, Godfrey 6339 (GH, US). Duplin Co.: between Faison and Warsaw, 5 Oct 1935, Correll & Blomquist 4823 (GH, US). Pea Co.: between Clayton and Smithfield, 2 Sep 1932, Blomaguist 333 (F US). Robes o.: 5.1L mi SW of Sc. Pauls, 18 Oct 1958, Britt 2857 (MICH). SOUTH on Georgetown Co.: 5 mi S of Andrews, 11 Sep 1939, Godfrey 8191 (F, GH, MO, NY, PH, US); 5 mi N of Georgetown, 15 Sep 1939, Godfrey & Tryon 8229 (GH, US). Sumter Co.: 5 mi E of Sumter, 4 Oct 1957, Site 29591 (G a NY). VI GINIA: Sussex Co.: 4 mi NW of Homerville, 20 Se ep 1937, Fernald & Long 7301 (F, PH); NW of Owen's Store, 14 Oct 1941, Fernald & Long 1 3884 (GH, PH). Saccharum giganteum (Walt.) Pers., Syn. Pl. 1:103. 1805. Anthoxanthum eigantenm Walt., Fl. Carol. 65. 1788. Erianthus saccharoides Michx., Fl. Bor.-Amer. WEBSTER AND SHAW, North American species of Saccharum 575 1:55. 1803, nom. illeg. Erianthus giganteus (Walt.) C. E. Hubb., Rhodora 14:166. 1912, non Muhlenb. 1813. Saccharum erianthordes Raspail, Ann. Sci. Nat., Bot. 5:308. 1825, nom. illeg. Andropogon erianthus Link, Hort. Berol. 1:243. 1827, nom. ie Tyree: SOUTH CAROLINA. Georgetown Co.: Godfrey 8192 (NEOTYPE: MO, he designated). Fig. 5 Erianthus saccharoides Michx. var. michauxii Hack., in Mart., Fl. Bras. 2:57. 1883 (fide Hitchcoc Erianthus panies Nash, Bull. Torrey Bot. Club 24:344. 1897. Type: FLORIDA. Paola, Swingle 1732a (HOLOTYPE: NY!). Erianthus ee Nash, Bull. Torrey Bot. Club 22:419. rea) Erianthus saccharoides Michx. var. compactus Sak, Fernald, Rhodora 45:252. 1943. Type: Washington, D.C., Nash s.n. (HOLOTYPE: Rhizomes with uses nodes. Flowering culms mostly 10-25 dm tall. Nodes hairy (the hairs ca. 5 mm long). Internodes hollow. Leaves with- out auricles. Sheaths ciliate at the apex. Ligule 2-6 mm long. Collar differ- entiated; hairy. Leaf blades mostly 35—70 cm long; 8-30 mm wide; hairy on the upper surface. Peduncle mostly 40-80 cm long; hairy (pilose). Inflo- rescence oblong or lanceolate; 6-15 cm wide. Lowermost inflorescence node differentiated; densely hairy. Main axis mostly 15—30 cm long; hairy; pi- lose; densely or sparsely hairy. Primary branches appressed to the main axis (spreading at anthesis); 2-13 cm long; hairy; ciliate. Rachis internode 2— 5.5 mm long; about 0.3 mm wide; hairy. Pedicels 2.5—5.0 mm long; hairy. Callus hairy. Callus hairs straw-colored or brown; 7-25 mm long (mostly 15-20); longer than the spikelet. Spikelets heteromorphic (the lower spikelet commonly glabrous and the upper spikelet pilose); straw-colored; 4.2—6 mm long; 0.8—1.1 mm wide. First glume indistinctly 5-nerved; smooth. Second glume indistinctly 3 or S-nerved. Lemma of lower floret 3-5 mm long; nerveless. Upper floret subequal to the lower floret. Lemma of upper floret 2.5-3.5 mm long; 1-nerved; entire. Awn of upper lemma usually 12—26 mm long; basally terete; not basally spiraled. Lodicules with nerves extending or not extending into hair-like projections. Saccharum gigantenm is a polymorphic species that occurs throughout the southeastern United States and extends into Central America where its morphological relationship with S. tr/nii (Hack.) Renv. is unclear. In North America its western limit is the savannas and prairies of east Texas, and its northern limit is Arkansas through Tennessee to southern Pennsylvania and New Jersey. Within this complex, 5. gégantewm is distinguished by the presence of a straight upper lemma awn and callus hairs longer than the spikelet. It is most similar to 8. a/opecuroideum, and intermediate specimens are occasionally encountered. Saccharum giganteum commonly exists in a variety of ecological conditions ranging from standing water and loamy soils to dry red clay on open hillsides. It prefers open marshy conditions 576 Sipa 16(3) 1995 as va ff 4mm apa iie - \\ PAIN REE). il Tb \ | ae inp aaa Ue md Si = Sa er oe oe Z FOE 7 a me Sse. a — —— = NN bk — = ee mW — ~~ == = — a > ) SS SS re S 2S ee SE c= re ~ x — <~> ert an, oe > = Se. = Fic. 5. Saccharum gigantenm. A. Habit, B. Ligule, C. Spikelet Pair, D. First Glume, E. Upper Lemma and Awn. WEBSTER AND SHAW, North American species of Saccharum 577 where it frequently grows to 2.5 m. in height with large, bushy inflores- cences. It is also common in the clay pineland savannas of northern Florida and southern Mississippi, where it may be only a meter in height with a relatively small compacted inflorescence. Nash (1913) recognized three taxa that Hitchcock (1951) placed in syn- onymy. Nash Sep arated E. /axus from E. saccharoides (= S. giganteum) on the basis of elongate primary branches, longer rachis internodes, and longer basal hairs. These characters and others exhibited by the type specimen cannot be used to differentiate E. /axus from the typical form; therefore, the name E. /axus was placed in synonymy. Nash delimited E. compactus on the basis of a compact panicle with short branches, short rachis internodes, and crowded small spikelets. These characters cannot be used in a consis- tent manner. Many of the specimens with these characteristics also exhibit evidence of seasonal burning, mowing, or growth in less than favorable habitats. It was suspected that these smaller plants with compacted inflo- rescences, which are common in Delaware and New Jersey, are the result of the above negative growth factors and E. compactus was placed as a syn- onym. Nash's E. tracyii poses a different problem. The type specimen has a densely pilose inflorescence with long white callus hairs and the awn of the upper lemma is spiraled at the base. Based on the presence of the spiraled awn, E. tracyii was placed as a synonym of S. alopecuoideum. Other speci- mens [Pollard 1341 (NY), Kearney 6 (PH, NY), Tracy 5332 (NY)], col- lected near Starkville, Mississippi in the fall of 1896 are similar to the type of E. tracyiz but possess a short straight awn. These specimens appear simi- lar to the South American species, S. ¢rimiz. It seems possible that S. ¢rizz was introduced at the agricultural station at Starkville, escaped, collected, and since disappeared. Future studies of the variation in the South Ameri- can species may show that S. ¢rinii may rarely occur in southern Florida. Representative specimens examined. ALABAMA: Aut Auga Co.: 2 mi NE of Aut Augaville, 24 Sep 1934, Harper 3272 (GH, PH). Cullman Co.: 17 Sep 1897, Eggert s.n. (GH, NY). Escambia Co.: 2 mi W of Brewton, 6 Oct 1968, Kral 33730 (BRIT). Geneva Co.: 4 mi W of Geneva on county 4, 6 Oct 1958, Kral 22771 (BRIT). Mobile Co.: 1 mi N Mt. Vernon, 7 Oct 1967, Kra/ 29520B (BRIT). ARKANSAS: Ashley Co.: Wilson Lake, 11 Oct 1936, Demaree 13892 (BRIT, GH, NY). Bradley Co.: Drew and Bradley Co. line, 5 Oct 1940, Demaree 21822 (BRIT, GH, MO, NY). Calhoun Co.: P.O. Tinsman, 20 Oct 1941, Demaree 22688 (BRIT, NY). Columbia Co.: P.O. Magnolia 6 Oct 1940, Demaree 21785 (BRIT, NY). Crittenden Co.: RO. W of Memphis, 3 Oct 1969, Demaree 61271 (BRIT). Phillips Co.: P.O. W of Helena, 15 Oct 1950, Demaree 30246 (BRIT). DELA- WARE: New Castle Co.: 1 1/2 mi SW toward Vandyke, 20 Sep 1907, Van Pelt s.n. (PH). Sussex Co.: 1.3 mi N of Bethany Beach, 12 Oct 1940, Chase 12626 (MICH). DISTRICT OF COLUMBIA: Brightwood, 7 Oct 1905, Hitchcock 235 (GH). FLORIDA: Brevard Co.: Okeechobee region, 22 Sep 1903, Fredholm 6017 (GH). Calhoun Co.: near Jackson- ville, Curtiss 3627 (MICH). Dade Co.: SW of Homestead, 12 Oct 1962, Craighead 9228 578 Sipa 16(3) 1995 (GH). Dixie Co.: 4 mi N of Suwannee, 12 Oct 1957, ae 56180 (GH, NY). Duval Co.: Fe. Caroline and Merrill Roads, F.C. & Greager 282 (GH). nilton Co.: 1.2 mi WNW of Jennings—Blue Springs exit Hwy I-75, 13 Nov 1970, a 325 (NY). Lee Co.: Central Sanibel, 6 Oct 1972, Brumbach 8046 (GH). Leon Co.: 5 mi N ee eeags 11 Oct 1957, Godfrey 56136 (GH). Madison Co.: 7 mi W of Greenville, 2 Oct 1955, Kral & a 53999 (NY). Seminole Co.: Altamonte Ee 5 Oct 1959, ae 486 (MIC H).G GIA: Colquit Co.: in small pine—barren stre 23 Sep 1902 sie a 1662 (GH). oe tosh Co.: 5 mi NW of ea 10 Mar 194( 0, Ww ker 1486 (PH). LOUISIANA: Avoyelles Parish: 8 mi S of Bunkie, 17 Sep 1968, Harvey 8098 (MICH). eye Parish: 5.8 mi N of Waverly, 8 Sep 1979, ae et al 1499 (NY). Natchitoches Parish: Sec. 1 T12N R6W, 17 Oct 1986, Thomas & Gilmore 98848 (MO, NY). St. Landry Parish: 3.3 mi W of Opelousas, 5 Oct 1956, Shinners 25,008 (BRIT). Tensas Parish: 3.7 mi S of La. 4 and Newellton, | Oct 1987, Thomas 102,200 (NY). MARYLAND: Caroline Co.: | mi N of ee cross- roads, 14 Sep 1986, Hill 17378 (NY). Dorchester Co.: 11 mi SW of Cambric lge, 5 Sep 1937, Earle 1636 (GH, PH). Prince Georges Co.: Kenilworth swamp, 24 ei 1897, Preters sn. (MICH). Talbot Co.: 2 1/2 mi NE by N of Easton, Earle 4774 (GH); 1 1/2 mi WNW. of Easton, 17 Sep 1967, Earle 5401 (P Hy). Wicomico Co.: Rt. 50 ae at Nanticoke River, 12 Oct 1981, Hi// 10859 (GH). MISSISSIPI: Franklin Co.: 4 mi W of jct Miss. 33 near Leesdale, 25 Sep 1970, Westmoreland 707 (B). Pearl River Co.: Bene 6 Oct 1964, Sargent 8476 (MICH). NEW JERSEY: Burlington Co.: Indian Mills, 9 Sep 1922, Drisbork 1144 (EF, GH, PH). Camden Co.: | Sep 1871, Parker s.n. (F). Cape May Co.: Ocean City, 14 Sep 1907, Van Pelt 1907 (GH, PH). Cumberland Co.: W of Cedarville, 17 Mar 1935, Long 45538 (PH). Gloucester Co.: 18 Sep 1868, Parker s.n. (GH). Ocean Co.: Forked river, 3 Nov 1937, Koster 05-85-2 (MICH). NORTH CAROLINA: Anson Co.: near Polkton, 25 Oct 1936, Corre// 7101 (MICH). Beaufort Co. Aeon near Edward, 13 Oct 1938, Godfrey 6890 (GH). Brunswick Co.: The Oaks, 4 Oct 1941, Radford 10110 (NY). Buncombe Co.: near Biltmore, 2 Sep nee 5847 (F, GH). Cartaret Co.: 1 Sep —1938, Godfrey 6499 (GH). Craven Co.: New Bern, 11 Nov 1903, £.B.H. 4177 (GH). Durham Co.: Duke Forest, 20 Oct 1932 » Blomquist 708 a ). Edgecombe Co.: 3.2 mi S of Battleboro, 23 Sep 1958, Radford 40595 (BR). Gates Co.: near Gatesville, 15 Oct 1938, Godfrey 7043 (GH). New Hanover Co.: Carolina Beach, 30 Aug 1938, Godfrey 6375 (GH). Hoke Co.: 6 mi S of Southern Pines, 25 Oct 1935, Blomquist 7176 ies 1). Hyde Co.: near Leechville, 13 Oce 1938, — & White 6863 (GH). Macon Co.: near Franklin, 10 Sep 1933, Alexander et al s.n. (NY). Nash Co.: Pineland at Middlesex, 9 Oct 1938, Godrey & Kerr 6616 (GH). Pamlico Co.: oe near Grantaboro, | 3 Oct 1938, Godfrey G& White 6820 (GH). Pite Co.: 1.5 mi sec. of Stokes, 9 Oct 1958, ee (NY). Swain Co.: Smoke Mountains, 20 Aug 1891, pei & Kofoid s.n. (GH). Wake Co.: sphagnum bog at Method, 10 Oct 1938, Godrey 67 GH). Wilson - o.: near US. 264, 3 mi W of Sims, 25 Sep 1958, Radford 40709 nie at PENNSYLVANIA: Bucks Ce bank of Delaware river, Martindale s.n. (GH). SOUTH CAROLINA: Berkeley Co.: Swale, St. Stephens, 15 Sep 1939, Godfrey 8199 (GH). Georgetown Co.: 5 mi S of Andrews, Godfrey 8192 (GH, MO). Jasper Co.: 4 mi NNW of Hardeeville, 12 1974, mie 15848 (GH). Lee Co.: 4 mi WSW of Ashland, 3 Oct 1957, Radford 29301 (GH ). Williamsburg Co.: 5.4 mi ene of Kingstr 19 Oct 1957, Radford 31233 (BRIT). TENNESSEE: Sumner Co.: pond near Mitchel ville Aug 1886, Gattinger s.n. (GH). Tipton Co.: 9.7 mi SW of nae 23 Oct 1958, SAinners 27,674 (BRIT). TEXAS: Scone Co.: 3 mi N of EM. , 28 Oct 1974, Waller & Bauml 3293 (GH). Cass Co mi SW of Linden, 18 a ee Correll 26424 (MO). Galveston Co.: 3/4 mi E I. ‘ 45, N of Dickinson, 18 Nov 1974, Waller & Baum! 3301 (GH). Gonzales Co.: Palmetto State Park, 3 Oct 1943, Barkley 13697 (BRIT, GH, NY, WEBSTER AND SHAW, North American species of Saccharum 579 PH). Henderson Co.: 2 mi SE of Athens, 12 Oct 1962, Correl/ 26187 (NY). a Co.: 2 mi W of McGee Bend Dam., 13 Nov 1963, Correll 28644 (NY). Milam Co.: 4 mi NW of Milano, 30 Oct 1933, Wolff 4873 (BRIT). Orange Co.: 6 1/2 mi W of Orange, 16 Nov 1945, Cory 50885 (BRIT, MICH). San Augustine Co.: Boykin Spring Camp, E of Zavalla, 19 Sep 1952, Gould & Leinweber 6544 (BRIT). Wood Co.: near Little Hope, E. of Quitman, 19 Jan 1979, Gritz 102 (BRIT). VIRGINIA: Sepa Co.: 3/4 mi W of Waterlick road, 28 Sep 1947, Freer 1913 (GH). Dinwiddie Co.: 4 mi S of Petersburg, 11 Oct 1938, Fer ba & Long 9505 (GH). Fairfax Co.: near Fairfax, 21 Sep 1936, Allard ae (GH, NY); USS. paul 50, E of Hs rai 22 Sep 1939, Hermann & Martin s.n. (MICH, NY). ee a NW of Dahlia, 18 Sep 1938, Fernald & Long 9244 ae Nuaseaiond Co.: awe “Hil 1, 26 ee 936, Fernald & Long 6456 (GH). Prince George Co.: N of Gray urch, 25 Aug 1936, Fernald & Long 6455 (GH, PH). Stafford Co.: 1/4 mi E, 3 Sep 1939, Hermann 10395 (GH). Southampton Co.: Swale, Courtland, 11 Sep 1937, Fernald & Long 7304 (GH, PH). Sussex Co.: NW of Wakefield, 11 Sep 1937, Fernald & Long 7303 (GH) REFERENCES ALLEN, C. M. 1975. Grasses of Louisiana. University of Southwestern Louisiana, Lafayette, Louis Bor, Si oe. The grasses of Burma, Ceylon, India and Pakistan. New York: Pergamon Press. _ «1970. Gramineae in Rechinger’s Flora Iranica. Akademische Druck Verlagsanstalt, Austria. Burkuart, A., J.-A. Caro, K.A. OKaba, R. . Paracios, and Z. E. R. AGRASAR. cane rises ilustrada de Entre Rios (Argentina): Part Il, Gramineas. Buenos Aires: I. Burner, D. and R.D. Wesster. 1994. sae of the native North aeun species of Saccharum. Sida 16:233-24 CuapMaNn, A.W. 1860. Flora of elie southeastern United States. Ivison, Phinney & Co. New Os ei ae 1955. The grasses and pastures of South Africa. Hafner Publishing Co.: New Y CLAYTON, W D. and $. A. RENVOIZE. faa Genera graminum: Grasses of the World. Kew, London: al Botanical Gardet Corre, D.S. and M.C. JOHNSTON. 197 70. Manual of the vascular plants of Texas. Texas Research Fndation Renner, Texas. pe M.J. 1 . A general system for coding taxonomic descriptions. Taxon 29: A Datiwirz, aiae T.A. Paine, and E.J. Zurcner. 1993. User's guide to the DELTA system: A general system for processing taxonomic descriptions. 4th ed. CSIRO Division of Entomology: Canberra, Australia. Dutt, N.L. and J.T. Rao. 1950. The present taxonomic position of Saccharum and its congeners. Proc. Int. Soc. Sug. Tech., VI] Congr., Brisbane, 288-293. Euuott, S. 1816. A sketch of the botany of South Carolina and Georgia. J. R. Schenck, Charleston, South Carolina. FERNALD, M.L. 1943. Erianthus brevibarbis Michx. and other species. Rhodora 45:246-255. 1950. Gray’s manual of botany. D. Van Nostrand Co.: New York, New Ganp!, K.N. and B.E. Durron. 1993. Palisot de Beauvois, the correct combining author of Erianthus gigantens (Poaceae). Taxon 42:855-856 580 SIDA 16(3) 1995 Goprrey, R.K. and J.W. Wooren. 1979. Aquatic and wetland plants of southeastern United States. University of Georgia Press, Athens, Georgia GouLp, FW. 1975. The grasses of Texas. Texas A&M Daneiy Press, College Station, Texas. Hircucock, A.S. 1933. Gramineae, In: J.K. Small, Flora southeastearn United States. New York: By the author. . 1935. Manual : the grasses of the United States. Washington, D.C.: U.S. ept. ee Misc. Publ. 2 eee eee Se . Manual 7 the grasses of fe United States. 2nd ed. Revised by A ae ase. Washineean, D.C., U.S. Dept. Agric. Misc. Publ. 200. be C1078, Camus In: Flora of Taiwan. Epoch Publishing Co.: Taipei, Taiwan. LINNAEUS, was 1753. Species le Facsimile ed., Royal Society London, England. LONG, R.W. and O. Laketa. 1971. A flora of tropical Florida. University of Miami Press, coral Gables, Florida. Mies: ix, A. 1803. Flora boreali-Americana. Caroli ral Monr, C. 1901. Plane life of Alabama. Contr. U.S. Natl. Herb. 6. MUHLENBERG, G.H.E. 1813. Cat. Pl. er. Sep c. Wi ae Hamilton, Lancaster. MUKHERJEE, S.K. 1958. Revision of on genus Erranthus Michx. (Gramineae). Lloydia, —188. Nasu, G. 1903, 1913. Gramineae, In: J. K. Small’s flora southeastern United States. New York: By the author. Nurtait, T. 1818. The genera of North American plants. D. Heart, Philadelphia, Penn- sylvania Prrsoon, C.H. 1805. Synopsis plantarum. Parisiis lutetiorum, Par Raprorp, A.E., H.E. AHLes and R. Bett. 1964. Manual of the es flora of che Caroli- nas. The Speics of North Carolina Press, Chapel Hill, North Carolina. RENVOIZE, 1984. The grasses of Bahia. Royal Botanic Gardens, Kew, England. oe - ,B. ARRILLAGA DE Marre, and P. [IzAGuirre bE ArTuUCIO. 1970. Gramineas Uruguayas. Waive: ad de la Republica, oe de Publicaciones, Montevideo, Uruguay. SmitH, L.B., D.C. WassHausen, and R.M. 1982. Gramineas in flora ilustrada Chibdmnetise: Smithsonian Institute, une D.C. SPRENGEL, K. 1815. Pugillus 2:16 SPRENGEL, K. 1825. Systema Vegetabilium 1:282. Gottingae. SrrausB AuGH, P.D. and E.L. Core. 1970. Flora of West Virginia. Seneca Books, Inc., Grantsville, West Virginia. Tsvetev, N.N. 1976. Grasses of the Soviet Union. Academiya Nauk SSSR, Botanicheskiy Institut im. V.L. Komarova. {English translation by Amerind Publishing Co.: New Delhi 1983.} Watrer, T. 1788. Flora Caroliniana. J. Fraser, London, England. Wester, R.D. 1988. Genera of North America Paniceae (Poaceae:Panicoideae). Sida 13:576-609. 1992. Old World genera of the Paniceae (Poaceae:Panicoideae). Sida 15: NOTEWORTHY PLANTS FROM NORTH FLORIDA. VI LORAN C. ANDERSON Department of Biological Sciences Florida State University Tallabassee, FL 32306-2043 U.S.A. ABSTRACT e following appear to be first reports for the state of Florida: Anredera baselloides, aes maculata, Scirpus subterminalis, Silene caroliniana, and Zigadenus leimanthoides. Some additions to the Florida panhandle are documented also, and several significant range extensions, particularly for rare or endangered taxa within our area, are given. Qu RESUMEN Las plantas siguientes parecen ser primeras citas para el estado de Florida: Anredera baselloides, Chimaphila maculata, Scirpus subterminalis, Silene caroliniana y Zygadenus eee Se documentan también algunas adiciones para el extremo de Florida y se d an s de drea significativas, particularmente para plantas raras o en peligro dentro de este area. Recent botanical discoveries in the western part of the state prompted this sixth installment of a series (Anderson 1984, 1986, 1988a, 1989, 1991) to update our knowledge of the flora of the Florida panhandle and Clewell’s (1985) guide to the flora. The area of coverage is from the Suwannee River west to the Alabama state line. New discoveries—1.e., taxa not listed by Clewell—and range extensions of selected rare or otherwise noteworthy taxa are given here. Only general locations are given for some of the rarer taxa; most Eglin Air Force Base localities are closed to the public, and the area is patrolled and protected by military and civilian personnel. Exotics that appear to be adventive or natu- ralized are also listed. Herbarium specimens are at FSU unless noted otherwise. TAXA NEW TO THE AREA Amaranthus palmeri Watson. Gadsden Co.: fallow field, Quincy, 3 Nov 1965, R. K. Godfrey 67534 (FLAS, FSU), originally identified as A. hybridus, Jefferson Co.: frequent bordering pastures along Hwy 221 just N of Ashville, 19 Sep 1988, L.C. Anderson 11833; Leon Co.: roadside weed in Tallahassee, 12 Jul 1991, L.C. Anderson 13469 (staminate plant), 15470 (piscillace plant); naturalized, new to Florida panhandle. Anredera baselloides (H.B.K.) Baill. Leon Co.: locally common along Sipa 16(3): 581-587. 1995 582 Sipa 16(3) 1995 railroad embankment, W of Cadiz Street in Tallahassee, 4 Aug 1993, L.C. Anderson 14506, 27 Oct 1993, L.C. Anderson 14614; naturalized, new to Florida. A tendency toward imperfect or functionally imperfect fowers exists in the genus (Bogle 1969). Our plants have bisexual flowers; the pollen appears fertile (as determined by microscopic examination of stained grains), but no fruits have been found. Another species, A. /eptostachya (Mogq.) Steenis occurs sporadically in peninsular Florida. Its flowers have three bifid styles and are subtended by separate, caducous bracteoles, whereas flowers of A. baselloides have single styles with three capitate stigmas and are subtended by connate, persistent bracteoles. elamcanda chinensis (L.) DC. Leon Co.: along fence at San Luis Ridge park, | Aug 1994, L.C. Anderson 15168; naturalized, new to Florida pan- handle. Blechnum serrulatum L.C. Richard. Bay Co.: large, local population in wet a flatwoods in St. Andrews State Recreation Area near Grand Lagoon, 22 Jan 1995, L. O’Kane s.n.; naturalized (or native), new to Florida panhandle. Chimaphila maculata (L.) Pursh. Leon Co.: 15—20 stems seen in shade of magnolia-beech-oak woodland near Chaires, 27 Apr 1993, L.C. Ander- son 14185; native, new to Florida. Ficus pumila L. Leon Co.: Tallahassee, R.K. Godfrey ; Florida State Uni- versity campus, Tallahassee, 12 Oct 1994, L.C. Anderson 15306; natural- ized in Tallahassee for many years (apparently overlooked by Clewell). Galactia ellotti Nutt. Taylor Co.: locally common in coarse sandy soil along Rte 361, ca. 10 air mi NW of Steinhatchee, 13 Jul 1992, L.C. Ander- son 13649, native, new to Florida panhandle. Lindera subcoriacea Wofford. Okaloosa Co.: shaded floodplain of Metts Creek, Eglin Air Force Base, 31 Jul 1992, L.C. Anderson 13851, 15 Jul 1993, L.C. Anderson 14451; native, new to Florida panhandle (possibly new to the state). Some collections from Sanna Florida have been tenta- tively identified as this species (McCartney et al. 1989), but B. Sorrie (pers. comm.) claims those plants represent L. a (L.) Blume var. pubescens Palmer & Steyermark. Melampodium divaricatum (Rich. in Pers.) DC. Leon Co.: Leon Sinks Geological Area in Apalachicola National Forest, 25 Oct 1993, L.C. Anderson 15285. A single plant of this Central American species was found (it is available through local nurseries for horticultural use and may become natu- ralized). Monotropa hypopithys L. Okaloosa Co.: slopes of Silver Creek, E of Rte 85 on Eglin Air Force Base, 27 Jun 1994, A. ne 1100; Walton Co.: dried flowering stems infrequent along crest of steep head, Eglin Air Force Base, generally N of Choctaw Beach, 30 May 1994, A. Schotz 1157; new to ANbERSON, Noteworthy plants from north Florida 583 Florida panhandle. This endangered species (Coile 1993b) is known to oc- cur in central peninsular Florida (Ward 1966). Scirpus subterminalis Torr. Santa Rosa Co.: collected via canoe along edge of Big Coldwater Creek, T3N, R28W, Sec 13, 30 May 1994, L. Chafin 2051; frequent in coarse sand of shallow, shaded shoal, submersed in 10— 15 cm water with S. etwberculatus (Steud.) Kuntze along bank of Big Coldwater Creek between Bass Brinks Creek and Wolfe Creek, ca. 7.5 air mi NNE of Milton, | Jul 1994, L.C. Anderson 15053; native, new to Florida. Koyama (1962) concluded that S$. sa#bterminalis and S. etuberculatus were conspecific and made the combination S. swbterminalis var. cylindricus (Yorr.) Koyama for the latter. He reasoned that S. torreyi Olney should be merged with S. subterminalis and that S. etuberculatus was just a southern form of S. torreyi (hence all three belonged to one species), but Gleason and Cronquist (1991) treat all three as distinct species. Apparently, S. s#bterminalis and S. etuberculatus do hybridize in South Carolina (A. Pittman, pers. comm.). He suggests the two morphologies (i.e., S. subterminalis with flaccid, terete submerged stems with inflores- cences as single spikes and S. etuberculatus with rigid, triangular emergent stems with umbellate inflorescences) may be due to the respective plants growing at different levels in the water column. I found the two growing side by side on the same shallowly submerged shoal with no sign of in- tergradation; further, populations of typical S. etwberculatus are scattered across much of Florida, yet S. swbterminalis has just now been found in the state—it should be more wide spread if it is indeed just an ecotypic vari- ant. I consider the two as distinct species that may exhibit some hybridiza- tion in limited portions of their ranges. Silene caroliniana Walt. Okaloosa Co.: infrequent at base of shaded N- facing bluff bordering Rocky Bayou, E of Niceville, 21 Jun 1994, L.C. Anderson 15021; native, new to Florida. This may be a relictual population persisting in a suitable niche in an otherwise inhospitable environment; a similar circumstance has been suggested for S. virginica L. (Anderson 1991; Kral 1966). Solanum viarum Dunal. Leon Co.: wooded slope above Dog Lake, Apalachicola National Forest, 20 Oct 1994, R.K. Godfrey 84074, single plant seen and destroyed (naturalized?), first report for Florida panhandle. This noxious weed was recently found in southern Florida (Coile 1993a) and now infests considerable acreage there. Sphagneticola trilobata (L.) Pruski. Leon Co.: roadside ditch, Tallahas- see, 20 Jul 1994, L.C. Anderson 15107; naturalized, new to Florida pan- handle. The species is widely cultivated and naturalized in peninsular Florida and around New Orleans; this new report represents the northern-most population known for the species. 584 Sipa 16(3) 1995 This species was placed in Wedelia for many years. Strother (1991) said it should be in a separate genus, which he named Complaya, but Pruski (1995) noted the earliest available generic name is Sphagneticola. Zigadenus leimanthoides A. Gray. Escambia Co.: along creek § of Nine Mile Rd, W of Pensacola, 27 May 1984, ].R. Burkhalter 9390 (FLAS, FSU); Okaloosa Co.: shaded floodplain along Metts Creek, Eglin Air Force Base, 31 Jul 1992, L.C. Anderson 13847; shaded floodplain along Middle Creek, Eglin Air Force Base, 2 Jun 1994, L.C. Anderson 14913; sunny seepage slope under power line along Turkey Creek, Eglin Air Force Base, 9 Jun 1994, L.C. Anderson 14944; Walton Co.: northwest {part of county], 17 May 1938, E.G. Hume s.n, (FLAS), shaded bay-gall bordering seepage slope, Blount Mill Creek drainage, 29 Jul 1994, L.C. Anderson 15140; native, new to Florida. The Burkhalter specimens are labeled Z. densus (Desr.) Fern., but that species is characterized as having racemose inflorescences as opposed to paniculate ones in Z. lesmanthoides. Preece (1956) states the two species are closely related but distinct (he did not see the Hume specimen), whereas W. McDearman (pers. comm.) considers the two conspecific. The two spe- cles appear ecologically and morphologically distinct in the Florida pan- handle. The taxonomic relationships are complicated further because the type specimens of Z. leimanthoides are relatively small plants from the moun- tains of North Carolina, whereas plants from the Florida coastal plain are very robust (they may represent a separate taxon). ADDITIONAL RANGE EXTENSIONS Asclepias viridula Chapm. Walton Co.: scattered in wet flatwoods at Alaqua Point, S of Rte 20, 1 Jul 1994, A. Schorz 1104; new to western panhandle (Wilhelm 1984, pers. comm.). This taxon is considered threat- ened in Florida (Coile 1993b). Calycanthus floridus L. Walton Co.: in semi-shade of oak-sourwood- pine woodland along Rocky Creek, 6 mi NE of Niceville, 26 Apr 1994, A. Schotz 1019; new to western panhandle (Wilhelm 1984, pers. comm.). This species is considered endangered in Florida (Coile 1993b). oelorachis tuberculosa (Nash) Nash. Santa Rosa Co.: infrequent in depression marsh at Goose Ponds, Blackwater River State Forest, 11 Sep 1994, J. Jensen s.n.; new to western panhandle (Wilhelm 1984, pers. comm.). Crataegus phaenopyrum (L. f.) Medic. Liberty Co.: swamp, Ochlockonee River, 4 May 1987, H.G. Grant s.n. (FLAS); Wakulla Co.: frequent on Ochlockonee River floodplain with Nyssa ogechee, N. biflora, Liquidambar styraciflua, and Taxodium ascendens, Apalachicola National For- est, T35, RAW, Sec 18, 1 Jul 1993, G. Anglin s.n., 12 Jul 1994, L.C. Ander- son 15067, Walton Co.: Nyssa ogeche swamp, N side Sister River in ANDERSON, Noteworthy plants from north Florida 585 Choctawhatchee River delta, 17 Oct 1988, A.F. Clewell s.n. (FLAS). The Wakulla County trees are 7-12 m tall, and the erect, leafless trunks (8-10 cm DBH) are armed with patches of thorns. This species is rare in Florida (Anderson 1986) and is listed as endangered (Coile 1993b). Coker and Totten (1934) reported this species (as C. youngii Sarg.) from Wakulla County; it has now been rediscovered for the county. Cyperus hystricinus Fern. Okaloosa Co.: infrequent in dry sand of open hickory-pine woods at Rocky Bayou State Recreation Area, Niceville, 21 Jul 1994, L.C. Anderson 15014; this constitutes the second record for this species in the state. Carter (1988) reported the species from Walton County. Eleocharis confervoides (Poir.) Tucker. Okaloosa Co.: common (veg- etative) in Blue Spring, Eglin Air Force Base, 12 Mar 1994, L.C. Anderson 14670; Santa Rosa Co.: abundant (fruiting) in shallow waters of Atwell Pond on Eglin Air Force Base, 31 Jul 1992, L.C. Anderson 13833. This infrequently collected species was long assigned to the genus Websteria. Liatris earlei (Greene) K. Schumann. Okaloosa Co.: open pine-oak wood- land, Rocky Bayou State Recreation Area, 28 Jul 1994, L.C. Anderson 15127, infrequent in northern Florida, new to western panhandle (Wilhelm 1984, pers. comm.). Linum westii Rogers. Okaloosa Co.: frequent in peaty sand along mar- gin of small Taxodinm ascendens-llex myrtifolia swamp on Eglin Air Force Base, N of Wynnehaven Beach, 14 Aug 1992, L.C. Anderson 13911. This ts a significant extension westward for this endangered species (Coile 1993b). Matelea alabamensis (Vail) Woodson. Walton Co.: locally common on wooded slopes above Piney Creek, N of Choctaw Beach, 17 Jul 1994, A. Schotz 1125. This species is endangered in Florida (Coile 1993b) and new to the western panhandle (Drapalik 1970; Wilhelm 1984, pers. comm). Panicum nudicaule Vasey. Walton Co.: locally common on seepage slopes, Blount Mill Creek drainage, 28 Jul 1994, L.C. Anderson 15120, 29 Jul 1994, L.C. Anderson 15142. Clewell (1985) included this species in Dichanthelinm dichotomum (L.) Gould {i.e., P. dichotomum}, but the two are very distinct in morphology and habitat preference. This rare species (Hitchcock 1950, Lelong 1988) is considered a species of special concern by the Florida Natural Areas Inventory. Pinguicula ionantha Godfrey. Wakulla Co.: frequent with P. pumila along upper edge of wet, roadside depression, N edge of St. Marks, 11 Mar 1986, L.C. Anderson 9117. This is a new county of record for this endan- gered species (Coile 1993b). Rhexia parviflora Chapm. Santa Rosa Co.: outer edge of small gum pond in Blackwater River State Forest, ca. 7 air mi E of Munson, 21 Sep 1989, S.L. Orzell & E.L. Bridges 12482; locally abundant along edge of I/ex myrtifolia marsh at Goose Ponds, Blackwater River State Forest, 9 Jul | 994, 586 Sipa 16(3) 1995 J. Jensen s.n.; Walton Co.: edge of titi-gallberry depression at Alaqua Point, I Jul 1994, A. Schotz 1103; drainage ditch in pine flatwoods at Alaqua Point, 16 Jul 1994, L.C. Anderson 15095. These represent two new coun- ties of record for this “rarest of the Rhexia species” (Bounds 1987) that is considered endangered in Florida (Coile 1993b). Rhynchospora crinipes Gale. Okaloosa Co.: forming mats with Sphag- uum along Malone Creek, 30 Jul 1992, L.C. Anderson 13807: infrequent (semi-floating) on shaded, wet bank of Metts Creek, Eglin Air Force Base, ca. 10 air mi W of Rte 85, 15 Jul 1993, L.C. Anderson 14450. These consti- tute an additional county of record in the Florida panhandle for this rare species (Anderson 1988b), Rhynchospora leptocarpa (Chapm.) Small. Okaloosa Co.: frequent in wet peat near headwaters of Point Lookout Creek, Eglin Air Force Base, 23 Jul 1992, L.C. Anderson 13746; upper edge of tidal marsh on Rocky Bayou, 0.5 mt E of Niceville, 21 Jun 1994, L.C. Anderson 14980. This Species is rarely collected in Florida. Kral (pers. comm.) considers this species conspecific with the widespread R. capitellata (Michx.) Vahl. The latter is a more delicate plant with darker brown floral scales from higher elevations (piedmont and mountains), whereas the coastal plants of R. /eptocarpa are more robust and frequently form tussocks with 30-40 culms which are gracefully arching to 12 dm long. The two geographical populations appear to be distinct in achene morphology (Blake 1918); this complex needs further study. Ruellia pedunculata Torr. ex Gray ssp. pinetorum (Fern.) R.W. Long. Liberty Co.: frequent in seasonally wet depressions in pine woods, Apalachicola National Forest, 11 air mi SSW of Bristol, 26 Jun 1992, L.C. Anderson 13660, 13675. These represent the second county of record in the Florida panhandle for this infrequently encountered species. Sideroxylon lycioides L. Santa Rosa Co.: floodplain swamp near E end of Grimes Lake, S of Yellow River, Eglin Air Force Base (ca. 1.95 mi E of Parkerville), | Aug 1994, A. Schotz 1142. This species, formerly known as Bumelia (see Pennington 1991 for evidence supporting transfer to Sideroxylon), is endangered in Florida (Coile 1993b); it is new to the west- ern panhandle (Wilhelm 1984, pers. comm.). ACKNOWLEDGMENTS My field work was sponsored in part by the Florida Natural Areas Inven- tory (through a grant from the U. $. Department of Defense Legacy Pro- gram) and the Florida Division of Parks and Recreation. Al Schotz kindly provided specimens from his extensive, ongoing plant survey on Eglin Air Force Base, and Kent Perkins (FLAS) supplied pertinent herbarium data. ANDERSON, Noteworthy plants from north Florida 587 REFERENCES ANDERSON, L.C. 1984. Noteworthy plants from north Florida. Sida 10:295—297. «1986. Noteworthy plants from north pee II. Sida 11:379-384. 1988a. Noteworthy plants from north Florida. IH. Sida 13:93-100. 1988b. Status of endangered ae crinipes (Cyperaceae). Syst. Bot. — 1S: 407-A10. _ 1989. Noteworthy plants from north Florida. IV. Sida 13:497— 504 1991. Noteworthy plants from north Florida. V. Sida 14:467—47 4, BLAKE, S.F. 1918. Notes on the Clayton Herbarium. Rhodora 20:21-28. Bocte, A.L. 1969. The genera of Portulacaceae and d Basellaceae in the southeastern United States. J. Arnold Arbor. 50:566-5 Bounps, R.R. 1987. Rare species of ee L. Castanea 52:304—308. Carter, R. 1988. Cyperus hystricinus (Cyperaceae) new to Florida. Sida 13:118-119. Ciewe, A-F. 1985. Guide to the vascular plants of the Florida pant handle. Florida State University Press/University Presses of Florida, Tallahassee. Come, N.C. 1993a. Tropical soda apple, So/anum viarum Dunal: the plant from hell (Solanaceae). Florida Dept. Agric. & Consumer Services, Division of Plant Industry Botany Circular 27 CoiLe, N.C. 1993b. Florida’s endangered and threatened plants. Fla. Dept. Agriculture & onsumer Serv., Division of Plant Industry - Botany Section Contrib. 29, Gainesville. Cock, W.C. and H.R. Totten. 1934. Trees of the southeastern states. Univ. North Caro- lina Press, Coupe Hill. Drapatik, D.J. | _ A biosystematic study of the genus Matelea in the te aaa United States. a doctoral dissertation, Univ. North Carolina, Chapel H Gurason, H.A. and A. Cronquist. 1991. Manual of the vascular plants of sae United States and adjacent Canada. 2nd ed. New York Botanical Garden, Bronx. Hrrcucock, A.S. 1950. Manual of the grasses of the United States, 2nd ed., A. Chase. Publ. 200, U.S. Dept. Agriculture, Washington, anes T. 1962. The genus Scirpus Linn. Some North cites aphylloid species. Canad, J. Bot. 40:913-937. Krat, R. 1966. cere on the flora of the southeastern United States with special reference to HORUIEED Louisiana. Sida 2:395—408. Letonc, M.G. 1988. Noteworthy monocots of Mobile and Baldwin counties, Alabama. Sida 13:101-113. McCartney, R. - K. Wurpack, and J. Moore. 1989. The genus Lindera in Florida. Pal- metto 9(2):3 PENNINGTON, ab. 1991. The genera of Sapotaceae. Royal Botanic Gardens, Kew, Rich- mond, = UK. Preece, S. J. 1956. A cytotaxonomic study of the genus Zigadenus (Liliaceae). Unpubl. doctoral sera Washington State Univ., Pullman. Pruski, J.P. 1995. Compositae of the Guayana eats Tuberculocarpus and other aye in “a Heliantheae: Ecliptinae. Novon (in press). SrroTHER, J.L. 1991. Taxonomy of Complaya, os logeton, Jefea, Wamalchitamia, Wedelia, Zener and Zyzyxia (Compositae—Heliantheae—Ecliptinae). Syst Monogr. 33:1-1 WARD, B. 1966. oe of Monotropa hypopithys in Florida. Quart. J. Florida Acad. Sci 20, Peas G S. 1984. Vascular flora of the Pensacola region. Unpubl. doctoral dissertation, Southern Illinois University, Carbondale. 588 Sipa 16(3) 1995 BOOK REVIEW ZOMLEFER, WENDY B. 1994. Guide to Flowering Plants Families. (ISBN 0-8078-2160-8, hbk.; ISBN 0-8078-4470-5, pbk.). The University of North Carolina Press, Chapel Hill and London. $55.00 (hbk); $27.50 (pbk). 430 pp., 165 text figures, 302 glossary figures, 22 tables, 2 appendices. Rarely are the talents of a superb botanical illustrator and a qualified botanist combined in one person. Wendy Zomlefer is such a person, and her Guide to Flowering Plant Families is proof. Even a quick perusal of the illustrations shows them to be accurate, botanically significant, life-like, and crisp. The book is intended to be adopted as a laboratory manual or atlas of flowering plant families for a college/universities plant systematics or field botany course; although, it also makes a very nice reference for professionals and enthusiasts, as QV well. Approximately 325 pages are devoted to 130 family treatments consisting of 1) full page plates illustrating whole plants, details, dissections, and even diagnostic anatomical features;. 2) a written diagnosis and important characters; 3) lists of important genera; 4) distribution; 5) economic members, 6) and commentaries on reproductive biology and phylogenetic problems and/or current studies. 1e choice of taxa illustrated focuses on warm temperate and tropical U.S. representa- tives (a few cultivated ones also) to aid students in mentally generalizing from typical members of the family. Often, several taxa showing the range of variation are illustrated. Of special interes is the inclusion of tropical families usually left out of an introductory course but which are of major world wide importance, notably the Lauraceae, Sapotaceae, Sterculiaceae, Bombacaceae, Rhizophoraceae, Malpighiaceae, Rutaceae, Meliaceae, Sapindaceae, Combretaceae, Myrtaceae, Bignoniaceae, Palmae, Bromeliaceae, Zingiberaceae, and Marantaceae. The book has several added bonuses for the student. First and foremost is an illustrated glossary showing structures of real species (cross referenced to the family plate), not some diagrammatic idealization. Secondly, in Appendix B is a complete chart comparing each family in term of size, distribution, vegetative habit, floral formula, fruit types, and any special diagnostic field characters. Also, there are introductory chapters on 1) illustration techniques and floral diagrams, and 2) cladistic concepts that are important in under- standing the commentary. Because she has adopted Thorne’s system (which she defends well), Cronquists’ system is outlined in Appendix A as a cross reference. In summary, this comprehensive atlas is up-to-date, well drawn, well planned, and user- friendly. It is highly recommended as a supplement to a text on systematic principles or as a general reference in every botanist’s library.—Roger W. Sanders. Stipa 16(3): 588. 1995 NOTES ARISTIDA DESMANTHA (POACEAE), NEW TO MISSOURI.— Aristida desmantha Trin. & Rupr. is an annual grass occurring on dry, sandy soil in Texas, Louisiana, Illinois, and Nebraska (Hitchcock 1951; Mohlenbrock 1973; Gould 1975; Allen 1992). During recent surveys for Cyperus grayoides Mol k in southeastern Missouri, a population of Aristida desmantha was discovered in Scott County. Voucher specimens: U.S.A. MISSOURI. Scott Co.: ca. 1 mi E of the intersection of Rt. H and Co. Rd. 504; last field on north side of Co. Rd. 504 before intersection with Co. Rd. 505; T28N, RI4E, SW1/4 of SEL/4 of SEI/4 of Sec. 28, 15 Oct 1992, McKenzie 1165 (LSU, MO, NMC, UMO); same location, 7 Sep 1993, McKenzie 1286 with Sherry Holmes and John Logan (MO). Habitat at the discovery site is a dry, remnant sand prairie with recent history of disturbance. The deep, sandy soil is eolian (1.e., wind deposited) in origin (Saucier 1978; Bridges & Orzell 1989), and has been classified as belonging to the Scotco Association (Festervand 1981; Carter & Bryson 1991). Such habitats in southeastern Missouri are remnants of what were formerly more extensive sand prairies as classified by Nelson (1985). These sand prairies were apparently derived from sand dune formations associ- ated with deposition of Wisconsinan Stage glacial outwash by the Missis- sippi-Missouri rivers east and west of Crowley’s Ridge in southeastern Mis- souri and northeastern Arkansas (Saucier 1978). Most remaining tracts of QuMan — remnant sand prairies in the region have been greatly altered by activities (Carter & Bryson 1991). At the discovery site, Aristida desmantha was associated with Andropogon ternarins Michx., Digitaria cognata (Schultes) Pilger, Diodia teres Walt. var. teres, Froelichia gracilis (Hook.) Mogq., Cyperus grayoides, Opuntia humifusa Raf. var. humifusa, and Triplasis purpurea (Walt.) apman. During the 7 Sep 1993 visit, tens of thousands of plants were estimated to occur at the site. Searches in similar, adjacent habitat have failed to produce any additional populations. The Scott County site is ca. 320 km SSW of the nearest locality in west central Illinois. While Aristida desmantha is currently known only from the discovery site, further surveys in appropriate sand prairie habitat in southeastern Missouri are likely to yield additional records. The species should be considered as a candidate for state-endangered status following guidelines established by the Mis- souri Department of Conservation (1992). I thank the following for assistance: George Yatskievych, Missouri De- partment of Conservation- Flora of Missouri project; and Kelly Allred, New Mexico Sate University. Doug Ladd, The Nature Conservancy; and Tim Stipa 16(3): 589. 1995 590 Sipa 16(3) 1995 Smith and Doug Newman, Missouri Department of Conservation reviewed an earlier draft of the manuscript.—Pau/ M. McKenzie, U.S. Fish and Wild- life Service, 008 E. Cherry St., Rm. 200, Columbia, MO 65201, U.S.A. REFERENCES ALLEN, C.M. 1992. Grasses of Louisiana. 2nd ed. Cajun Prairie Habitat Preservation Soci- ety, Eunice, LA. Bribces, E.L. and $.L. OrzeLL. 1989. Additions and noteworthy vascular plant collections ical notes. Phytologia from Texas and Louisiana, with historical ecological and geograp 96:1 2-69, Carrer, R. and C.T. Bryson. 1991. A report of Cyperus grayordes and Cyperus retroflexus (Cyperaceae) new to Missouri and notes on other selected Missouri Cyperms. Sida 14:475— 481. FESTERVAND, D.E. 1981. Soil survey of Cape Girardeau, Mississippi, and Scott counties, Missouri. U.S.D.A./Soil Conserv. Serv., in cooperation with the Missouri Agric. Exp. Sta. GouLbp, EW. L975. a grasses of Texas. Texas A&M Press, College Station. Hircncock, A.S., 1951. Manual of ae ae of the United States. 2nd ed., revised by Agnes Chase. U. S. a A. Misc. Publ. 2 Missourt DEPARTMENT OF Seer os Rare and oo species of Missouri checklist. Missouri Department of Conservation, Jefferson City. MoH ENBROCK, R.H. 1973. The illustrated flora of ilinois, G rasses. Panicum to Danthonia. Illinois Univ. Press, Carbondal NELSON, PW. 1985. The terrestrial saeutel communities of Missour:. Missouri Natural i, Areas Committee, Jefferson C Saucier, R.T. 1978. Sand dunes He related eolian features of the lower Mississippi River alluvial valley. Geosci. & Man 19:23—40. Stipa 16(3): 590. 1995 Notes 591 MALAXIS WENDTII (ORCHIDACEAE) IN THE UNITED STATES.—Recently Gerardo Salazar (Orquidea 13:281—284. 1993) sepa- rated Malaxis wendtii Salazar from Malaxis ebrenbergii (Reichb.f.) Kuntze based on its papillose flora segments (vs. glabrous) and its narrowly sagit- tate lip (vs. broadly triangular hastate). The type specimen and other speci- mens cited by Salazar are all from Coahuila, Mexico, two of the latter being from Sierra del Carmen and Serranias del Burro directly east across the Rio Grande from the Chisos Mountains of Big Bend National Park. It is there- fore not surprising that the two specimens at SRSC collected by Barton Warnock (Warnock 23382) in the Chisos Mountains and labeled M. ehrenbergii are typical M. wendti1. All specimens of M. ebrenbergii examined at ASU, TUC, NMC, and UNM that are from Arizona or New Mexico show the papillae characteristic of M. wendtii as do two specimens from Durango and Sonora. Specimens there from further south in Mexico are typical M. ebrenbergiz. Specimens are known from Apache (Worthington 8680, OCLA), Cochise, Pima and Santa Cruz counties in Arizona; from Catron (Worthington 7587, OCLA), Grant (Worthington 7497, 7501, 7505, OCLA), Lincoln (Worthington 7463, 12267, OCLA), Los Alamos and Otero (Worthington 8680, 12330, OCLA) counties in New Mexico; and from Brewster County in Texas. It is interesting that the flowers in the 10x photograph of M. ehrenbergii shown of Plate 81 of Luer’s The Native Orchids of the United States and Canada (1975), which he photographed in Cochise County, Arizona, exhibit the papillae of M. wendtiz. I chank Lawrence K. Magrath (OCLA) for providing additional collection data from Arizona and New Mexico.—Thomas K. Todsen, Department of Bi- ology, New Mexico State University, Las Cruces, NM 88003, U.S.A. ry Sipa 16(3): 591. 1995 592 SipA 16(3) 1995 CAREX AMPLIFOLIA AND CAREX ROSSI (CYPERACEAE), NEW TO NEW MEXICO AND A KEY TO SECTION MONTANAE IN NEW MEXICO.—Carex amplifolia F. Boott, a western wetland caric-sedge, and C. rossi? F. Boott, a predominantly Rocky Mountain caric-sedge, previously unreported for New Mexico, have been found in the northern Black Range of the Gila National Forest. Carex amplifolia section Anomalae Carey, was reported from Idaho to Brit- ish Columbia, and southward to San Mateo and Tulare counties, California by Mackenzie (1935). Mackenzie cited specimens from Idaho, British Co- lumbia, Washington, Oregon, and California. Other authors (Hermann 1970; Cronquist et al. 1977; Hickman 1993) have mirrored Mackenzie’s distribution. This species was not included in the flora of New Mexico by Correll and Correll (1972) and Martin and Hutchins (1980). With this new location, the range of this species has been extended ca. 1000 km (625 mi) east from Tulare County, California. Carex amplifolia is the only representative of section Anomalae in New Mexico. The most closely related taxon in New Mexico is C. /imosa L. of section Limosae Tuckerman. Carex limosa differs from C. amplifolia by hav- ing beakless or very slightly beaked perigynia and pistillate spikes 2.5 cm or less long. However, the existence of C. /imosa in New Mexico 1s dubious. We have not seen specimens from New Mexico and the only reference to its existence in the state is Martin and Hutchins (1980). Excluding C. /imosa, C. amplifolia differs from all other Carex in New Mexico that share a trigonous achene, a style jointed with the achene, and glabrous perigynia by having its lower bracts being sheathless or very short sheathing. Carex amplifolia is characterized by having long, stout rhizomes with erect culms up co ca. 7.5 dm tall; leaves usually greater than 1 cm wide; the terminal one or two spikes being staminate with the lower four to five spikes being pistillate; each pistillate spike bearing over 100 perigynia; and perigynia are ovoid, obtusely trigonous, smooth or nearly so, and abruptly contracted into a conic, more-or-less excurved beak. Carex amplifolia was located along the edge of Diamond Creek in asso- ciation with C, bella L.H. Bailey, C. disperma C. Dewey, Pteridinm aquilinum (L.) Kuhn var. pubescens Underwood, and Alnus tenuifolia Nuttall ina mixe conifer forest of Picea engelmannit Parry ex Engelmann var. engelmannii, Psendotsuga menziesti (Mirbel) Franco var. glauca (Beissner) Mayr, and Pinus ponderosa P. Lawson vat. scopulorum Engelmann. The elevation of the collec- tion location is ca. 2438 m (8000 ft) and is situated along the border of Sections | and 2 of Range 10W, Township 12S. This population was lo- cated near the NW corner of the Aldo Leopold Wilderness, which can be hu Stpa 16(3): 592. 1995 NOTES 593 accessed from NM 51 along U.S. Forest Routes 226 and 500. The popula- tion is ca. 2 miles S by foot from the terminus of USFR 500. Voucher specimen. U.S.A. NEW MEXICO. Sierra Co.: along edge of Diamond Creek, ca. 3.5 air mi N of Diamond Peak, W-SW of Winston, 16 Jun 1994, Roalson 879 (NMCR, y= pers. herb. of Stanley D. Jones). Carex rossit, section Montanae Fries, is known from the mountains of Colorado to middle California, and northward to Yukon, eastward to the Black Hills of South Dakota, and locally to northern Michigan (Mackenzie 1935). Hermann (1970) and Cronquist et al. (1977) reported a similar range with the addition of the Kaibab Plateau in northern Arizona. This location extends its known range by ca. 490 km (306 mi) from northern Arizona. This species was not included in the flora of New Mexico by Mar- tin and Hutchins (1980). While looking through unidentified specimens of Carex at UNM, a specimen of C. rossi# collected in 1963 was discovered. This specimen was collected in Taos County, New Mexico, N of the small town of La Lama. Other representatives of section Montanae in New Mexico are C. heliophila Mackenzie, and C. geophila Mackenzie. KEY TO CAREX SECTION MONTANAE IN NEW MEXICO la. Fertile culms all alike, bearing both staminate and pistillate spikes; bas spikes absent. C.h ee Mackenzie lb. Fertile culms of two types, some short, 1—5 cm. long, partly t e den among the densely tufted leaf-bases and bearing only pistillate spikes, other culms elongated, 5-30 cm. long, and bearing both staminate and pistillate spikes 2 2a. Bract of the lowest nonbasal pistillate spike leaf-like, normally ex- ceeding the culm. C. rossii F. Boot 2b. Bract of the lowest nonbasal pistillate spike squamiform, shorter than che culm. C. geophila Mackenzie Carex vossii is characterized by having medium-sized, more-or-less densely cespitose clumps, without long horizontal stolons; culms 5—30 cm high, slender but erect and strict; leaf blades usually less than 6 cm long, 1—2.5 mm wide, thin but firm; staminate spike sessile or short-peduncled, erect, 3-15 mm long, exceeding the contiguous pistillate spike; pistillate spikes —5, sessile or short-peduncled, perigynia 3-15, lower bract leaf-like, nor- mally exceeding the culm; perigynia 3—4.5 mm long, abruptly contracted into a conic, ciliate-serrulate, deeply bidentate beak. Carex rossii was col- lected on N-facing slopes that had been burned ina forest fire in July 1990. The area is now dominated by Populus tremuloides A. Michaux var. aurea (Tidestrom) Daniels, Physocarpus monogynus (Torrey) Coulter, Rubus strigosus A. Michaux var. arizonicus (Greene) Kearney and Peebles, and Robinia Sipa 16(3): 593. 1995 594 Sipa 16(3) 1995 neomexicana A. Gray var. neomexicana. Tufted plants of Carex deweyana Schweinitz were commonly associated with the tufts of C. rossi. Zigadenus virescens (Sunth) Macbride and Symphoricarpos spp. were common on the slopes. The elevation at the collection site is ca. 2865 m (9400 ft) and ts located in Range 10W, Township 12S, Section 24. The Continental Divide Trail runs along this slope in several switchbacks from the base to summit of the ridge. Voucher specimen. U.S.A. NEW MEXICO. Sierra Co.: silty slopes ee Continental Divide Trail, N of Diamond Peak, N-facing slope, SW of Winston, ae 994, Roalson 975 (NMCR, MICH). Taos Co.: 2 mi N of La Lama, $13 T28N Rl ort ft, 30 Jul 1963, Goodrow 375 (UNM). We thank A.A. Reznicek (MICH) for verifying our specimens. —Erw H. Roalson, Department of Animal and Range Sciences, Box 3-1, New Mexico State University, Las Cruces, NM, 88003 U.S.A.; Stanley D. Jones, S.M. Tracy Herbarium, Department of Rangeland Ecology and Management, Texas AGM University, College Station, TX, 77843-2126 U.S.A.; and Kelly W. Allred, Department of Animal and Range Sciences, Box 3-1, New Mexico State Univer- sity, Las Cruces, NM, 88003 U.S.A REFERENCES CorreLt, D.S. and H.B. Corre. 1972. Aquatic and wetland plants of southwestern United States. Vol. 1. U.S. Printing Office, Washington, D.C. Gioncuee A., A.H. HoLtMGREN, N.H. HOLMGREN , J.-L. Reveat and P.K. HOLMGREN. Pai intecmennian Alora: Vascular plants of the Intermountain West, U.S.A., Vol. 6. The New York ae Garden, HERMANN, EJ. | etre of the cree of the Rocky Mountains and Colorado Basin. Agri. eee . Forest Service, U.S. Department of Agric, Washington, DC. HICKMAN, J.C. (ed.). oe “he Jepson manual: Higher plants of California. University of California Press, Se Mackenzir, K.K. 1935. Cyperaceae-Cariceae. North Amer. Flora 18:201 Martin, W.C. and C.R. Hurcuins. 1980. A flora of New Mexico. Vol. Le -ramer, pub- lisher, Vaduz, West Germany. — Sipa 16(3): 594, 1995 Notes 595 SALVINIA MINIMA (SALVINACEAE), NEW TO TEXAS.—Recent collections from the J.D. Murphree Wildlife Management Area near Port Arthur, Texas, have yielded specimens of Sa/vinia minima Baker. This 1s a new record for Texas according to Hatch et al. (1990) and Correll and Johnston (1970). Previous collections have been reported from Louisiana, Georgia, Florida, Alabama; Mexico; the West Indies; and Central America (Nauman 1993) (Lellinger 1985). Salvinia minima, water spangles or floating fern, is locally abundant in the water of J.D. Murphree Wildlife Management Area SW of Port Arthur, Texas in association with Echinochloa, Zizaniopsis, and Leptochloa. This spe- cies is so abundant that populations are shading out some submerged plants species. This floating species has long thin rhizomes and floating fronds. The frond blades are elliptic to round with a round to cordate base, a round to notched apex, and with rows of whitish, multicellular hairs on the upper surfaces. The apex of each hair is divided into four separate branches. Voucher specimen: TEXAS. Jefferson Co.: J.D. Murphree Wildlife Management Area, SW of Port Arthur, 30 Oct 1993, Hatch 6403 (TAES). —Stephan L. Hatch, §. M. Tracy Herbarium, (TAES), Department of Range- land Ecology and Management, Texas AGM University, College Station, Texas, 77843-2126, U.S.A REFERENCES Corre, D.S. and M.C. a 1970. Manual of the vascular plants of Texas. Texas Research Foundation, Renr Hare H,S.L., K.N. GANDHI, and L. oe N. 1990. Checklist of the vascular plants of Texas. xas Agric. Exp. Sta. Bull. M Pande ER, D.B. 1985. A field an of he ferns and fern- allies of the United States and Canada. ae Institution Press. Washington, D. Nauman, C.E., 1993. Salvinaceae. In: Flora of North ee Editorial Committee, eds. 1993+. i of a Americ a North of Mexico. 2+ vols. Oxford University Press, New York. Vol. 2, pp. —33 RH YNCHOSPORA GAPIT ELLATA (CYPERACEAE), NEW TO KAN- SAS.—Five species of RAynchospora have been verified in the Great Plains (Great Plains Flora Association 1986), with R. g/obularis (Chapman) Small var. elobularis, R. harveyi W. Boott, and R. macrostachya Torr. ex A. Gray occurring in Kansas (Magrath and Johnson 1971). Field surveys in central Kansas during the fall of 1994 yielded a beakrush previously unknown 1n the state. RAynchospora capitellata (Michx.) Vahl was discovered at a single site in extreme southeastern Rice County. Plants were growing in several Sipa 16(3): 595. 1995 596 Stpa 16(3) 1995 low, mesic to wet-mesic depressions among stabilized sand dunes in grazed, sand prairie. Soils in the general area are coarse-textured and formed in eolian sands of Pleistocene age (Horsch 1974). The dominant vegetation in the prairie is Andropogon hallii Hack., Sorghastrum nutans (L.) Nash, Schizachyrium scoparium (Michx.) Nash, and Panicum virgatum L. Common associates in the swales include Fimbristylis autumnalis (L.) Roemer & Schultes, Bulbostylis capillaris (L.) Clarke, Lipocarpha drummondit (Nees) G.C. Tucker, A ge/inis tenuifolia (M. Vahl) Raf., Juncus scirpoides Lam., J. diffusissimus Buckley, and Eleocharis obtusa (Willd.) Schultes var. detonsa (A. Gray) Drapalik & Mohl. Gale (1944) mapped the range of RAynchospora capitellata from Nova Scotia to southern Ontario and Wisconsin, south to northern Florida and the Gulf Coast, and west to eastern Texas, eastern Oklahoma, and western Missouri. She also reported isolated occurrences in northern California and southern Oregon. The species was included in the Flora of the Great Plains (Great Plains Flora Association 1986) based on Missouri populations. Steyermark (1963) reported R. capitellata from the southern one-third of Missouri, with stations in the southwest part of the state in Newton, Jasper, Barton, Dade, Lawrence, and Greene counties. The Kansas population—in os roll- ing, sandhill prairie with scattered, interdunal ponds and wet depr is some 200 miles west of these nearest eastern stations. Additional field surveys are needed to determine if other populations of this species exist in similar habitat elsewhere in the Arkansas River Lowlands of central Kansas. Voucher specimen: KANSAS. Rice Co.: locally abundant in wet depressions, 6 mi mi N of Medora just N of Highland Estates and Country Club, elev 1,660—1,690 . = li sec 35, T21S8 Ce 13 Sep 1994, Freeman 6710 (KANU). —Craig C. Freeman, R.L. MeGregor Herbarium, The University of Kansas, 2045 Constant Ave., Lawrence, KS 66047, U.S.A. REFERENCES GALE, S. 1944. SC section Paes in Canada, the United States, and the West Indies. Rhodora 46:88—134, 159-197, 207-245, 255-278. GREAT PLAINS FLORA ea 19806. a of the Great Plains. University Press of Kansas. Lawrence, KS. Horscu, M.L. 1974. Soil survey of Rice county, Kansas. U.S. Dept. of Agriculture, Soil Conservation Service. Washington, D.C. Macratn, L.K. and K.L. Jonsso 1971. The genus Rhynchospora (Cyperaceae) in Kansas. Southw. Naturalist 15:3 STEYERMARK, J.A. 1963. ae of Missouri. Iowa State University Press. Ames, IA. Stipa 16(3): 596. 1995 Nores 597 POSSIBLE EPONYMY OF THE GENERIC NAME BRASENIA SCHREB. (CABOMBACEAE).—The genus Brasenia Schreb. comprises only B. schreberi J.F. Gmel., water-shield, a floating-leaved aquatic species native from scattered tropical American localities north to the Alaska pan- handle and the Gaspé Peninsula of Québec, and in eastern Asia at compa- rable latitudes, southern Africa, and eastern Australia. Standard floras and other references generally indicate that the generic name is “of uncertain origin” (Fernald 1950; similarly, “unexplained,” Britton 1901; “of unknown meaning,” Gleason 1952; “origin unknown,” Stone 1993, among others). The only suggestions of its derivation that I have encountered were Gray's (1895) speculative “unexplained, perhaps named for some obscure bota- nist,” and Rafinesque’s (1828) more definite “from a German botanist, Brasen.” Rafinesque, however, provided no given name or initials, nor any other identifying data such as association with an institution, city, or any of the kingdoms or duchies into which Germany was divided at that time (admittedly not always feasible when boundaries were unstable and politi- cal refugees were numerous), that would indicate that he was speculating any less than Gray as to the existence of such an individual. Schreber’s (1789) publication of the name Brasenia, in keeping with the format of the successive editions of Genera Plantarum, included no state- ment as to the derivation of the name. Unless evidence is discovered in extant correspondence written by or to Schreber, therefore, it will remain impossible to be certain as to his basis for coining this name. Circumstan- tial evidence, however, suggests that the genus was named for Christoph Brasen, a Moravian missionary who collected botanical specimens in Greenland and Labrador. Christoph Brasen was born 3 January 1738 at Ripen, Jutland (now Ribe, Denmark; the “new style” calendar was already in use there). As a mission- ary of the Moravian Church (Church of the United Brethren; Unitas Fratrum; Briider-Gemeine), he served first at Godthab (now Nak), Greenland. In 1767 and 1768, he collected plant specimens in the vicinity of Godthab Fjord (Porsild 1935). In 1771, by which time he had established a reputa- tion as being “knowledgeable in botany,” he was selected to be superinten- dent of the mission to be founded at Nain, Labrador, to which he went that year with his wife, née Maria Catharina Federhahn, also from Ripen. Brasen brought the valuable expertise of a “skilled surgeon” to the new mission, and was already fluent in Inuktitut from his years in Greenland. On 15 September 1774, en route back to Nain from a reconnaissance voyage to the site where the Okak mission was subsequently established, Brasen’s Sipa 16(3): 597. 1995 598 Sipa 16(3) 1995 vessel was wrecked and he drowned in the attempt to reach shore (Anony- mous 1835; Roemer et al. 1871; Davey 1905; Peacock 1976). Brasen is known to have collected botanical specimens in Labrador be- cause of those cited by Georg Heinrich Weber (1784), professor at Kiel, in his Plantarum Minus Cognitarum Decuria. (This work is sometimes attrib- uted to Sebastian Grauer. According to Sprague [1922}, Weber was the author of this work and of the new botanical names therein. Grauer, the “respondent,” discussed or defended it as a requirement for his doctorate.) Ribes glandulosum (Grossulariaceae), described as a new species, was based on a specimen collected by Brasen in Labrador (Lysaght 1971). Johann Christian Daniel von Schreber (1739-1810), who named the genus Brasenta, was professor of natural history and director of the botani- cal garden at Erlangen, Bavaria (Stafleu & Cowan 1985). He was one of many botanists of his time and later who acquired specimens collected by Moravian missionaries. Most such specimens were initially sent from the missions to the Moravian Church headquarters in Herrnhut, Saxony, where some of the officials were themselves active in botany and aware of the on interest among botanists in specimens from distant parts of the world. From Herrnhut, some specimens were given to botanical scholars among the clergy and lay members of the Moravian Church, and some were sold to other botanists to raise funds for the missions. These recipients sometimes di- vided the collections further for exchange purposes. Schreber acquired speci- mens collected in Labrador by Benjamin Gottlieb Kohlmeister (1756— 1844), one of the most important early contributors of botanical specimens from Labrador. Schreber did not publish upon these plants, but after his death the specimens were studied by Franz von Paula von Schrank, profes- sor and director of the botanical garden at Munich, who published a paper on the flora of Labrador (Schrank 1818), which included an account of the history of these specimens (Pringle 1992). he specimens from Schreber’s herbarium do not indicate a direct link with Brasen, buc they do indicate that during the latter part of the eigh- teenth century, Schreber communicated with Moravian churchmen who were interested in botany, from whom he acquired specimens collected by at least one of the Moravian missionaries in Labrador. It seems quite likely that he learned from his Moravian correspondents that one of the mission- aries in Labrador, whose own knowledge in botany had been respected and who had contributed specimens for botanical research, had died. Such a message might have included a suggestion that, if and when Schreber dis- covered a plant new to science, it might be named as a memorial to this early botanical collector. None of the Labrador specimens acquired by Sipa 16(3): 398. 1995 Notes 599 Schreber represented an unnamed genus, so Schreber could not name a genus native to Labrador for Brasen, but he was able to give such a name to a genus native to North America, in (according to Stafleu & Cowan 1985) his first major botanical publication following Brasen’s death. No speci- mens were cited by Schreber (1789), but the specimens from his herbartum now at Munich (M) might be searched for any that would indicate the basis of his knowledge of Brasenia and the source of any specimens of that genus that Schreber might have acquired.—James S. Pringle, Royal Botantwal Gar- dens, Box 399, Hamilton, Ontarto, Canada LSN 3H8. This is Contribution No. 89 from the Royal Botanical Gardens. REFERENCES ANonymous. 1835. The Moravians in Labrador, ed. 2. Edinburgh: William Oliphant and Britton, N.L. 1901. Manual of the flora of the Northern States and Canada. New York: Henry Holt and Compan Davey, J.W. 1905. The fall of Torngat, or, the Moravian mission on the coast of Labrador. ndon: S.W. Partridge. ee M.L. 1950. Gray’s manual of botany, ed. 8. New York: American Book Com- pany. Gieason, H.A. 1952. The new Britton and Brown illustrated flora of the northeastern United States and adjacent Canada. Bronx: The New York Botanical Garden Gray, A. 1895. Nymphecee [except Nupbarh In: ue A., 8. Watson, & B.L. Robinson: Synoptical flora of North America. New York, ie and Chicago: American Book Company; ae publishers elsewhere. 1(1):72—77. LysaGut, A.M. 1971. Joseph Banks in rhe ie and Labrador, 1766: His diary, manu- scripts and ces London: Faber and Faber; Berkeley and Los Angeles: University of California Press. Peacock, E{A.]W. 1976. The Moravian Church in Labrador. Them Days 1(3):3-17. PorsILD, M.P. 1935. Stray contributions to the flora of Greenland. X. A — excur- n to the fiords of the Godthaab district. Meddel. Grgnland 93(3):52-74. oe Jd 9071 1 Contributions by ats an missionaries to - knowledge of the flora of Labrador. Canad. Hort. Hist. 2:187- — C.S. 1828-1830. Medical flora; or, sae lorehe medical — . United of America. Philadelphia: Atkinson & Alexander. (Brasenza, vol. 8.) ee ET AL. 1871. Retrospect of the history of the mission of the see s Church in Labrador - the past hundred years. Periodical accounts relating to the missions of the Church of the United Brethren, eater among the Heathen, ser. I, 28:1—19, 53-72. Also published anonymously, | as: History of the mission of the Church of the United Brethren in Labrador for the Hundred Years. London: W. Mallalieu & Co. SCHRANK, F. von P. von. 1818. Aufzahlung einiger eee aus os mit Anmerkung. Denkschr. Konigl.-Baier. Bot. Ges. Regensburg 1(2):1— ScureBer, J.C.D. von. 1789-1791. Caroli a Linne.. ae era, ,ed. 8. Frankfurt Main: Varrentrapp et Wenner. 2 vols. (Brasenia in vol. 1. 1789.) SPRAGUE, T.A. 1922. Grauer’s Decuria. J. Bot. 60:267-272. Sipa 16(3): 599. 1995 600 Sipa 16(3) 1995 STAFLEU, F.A., and R.S. Cowan. 1985. Taxonomic literature: A selective guide to botanical publications and collections wit Regnum Veg. 112. STONE, W. J. 1993. Cabom manual — 1 dates, commentaries and types, ed. 2. Vol. V:Sal-Ste. baceae: Watershield family. In: Hickman, J.C., ed. The Jepson : Higher plants of California. Berkeley, Los Angeles, and London: University of California Press. P. 450 [Weper, G.H.] 1784. Plantarum minus cognitarum decuria. Kiel: University dissertation (Sebastian Grauer). Sipa 16(3): 600. 1995 BOOK REVIEWS STEPHENSON, STEVEN L. and Henry StemPen. 1994. Myxomycetes: A Hand- book of Slime Molds. (ISBN 0-88192-277-3, hbk.). Timber Press, Inc., 133 8. W. Second Avenue, Suite 450, Portland, OR 97204-3527. $34. 95. 200 pp, 16 color plates, 15 b&w photos, 69 line drawings, 6” x 9” The authors state in the preface that the intent of this book is to serve as a field guide to 175 common and cosmopolitan species of Myxomycetes for the general naturalist, ama- teur mycologist or interested lay person and as a useful introductory text for che more serious student of the Myxomycetes. The descriptions and keys included in the book are based mostly upon specimens collected in eastern North America The book is divided into seven chapters: an introduction that discusses general topics such as the life cycle and a definition of the Myxomycetes and related groups. A chapter on structural features of Myxomycetes that describes, defines and illustrates with black and white line drawings and photomicrographs, the fruiting body types and their structural parts. Chapter three is devoted to the collection, culture and study of Myxomycetes both in the field and in the laboratory. There is a topical section here on moist chamber cultures using bark from living trees. The distribution of Myxomycetes around the world is a chap- ter that discusses geographical distribution, snowbank SS high-latitude species, desert species, species on living trees and litter, dung and soil as a microhabitat. Some of this information is general knowledge but some is new, such as the high-latitude species from Alaska, with many of the references cited after 1970. The chapter on the ecology of Myxo- mycetes represents the special forte of Stephenson and much of the information under ae stages, substrate relationships, fruiting season, insect associates, myxomyceticolous fungi and bryophyte-myxomycete associations is new and not available in other books on the Myxomycetes. The chapter entitled classification and identification includes a discus- sion of taxonomic names and reversible dichotomous keys that lead to the orders of the Myxomycetes: Ceratiomyxales, Liceales, Echinosteliales, Trichiales, Physarales, Stemonitales This is followed by keys to selected spe in the orders and, finally, descriptions of the ecies. Families are not included in any of the keys. Species descriptions are given in sufficient detail to identify the species and compare the species with the specimen at hand. There is also a substrate and comments section, the latter contains the distinguishing characters of the species and notes on similar and look-alike species. There are 175 species mentioned in the book but only 54 species are illustrated and given species descriptions. Detailed watercolor portraits (16), pen-and-ink drawings (69), and black and white light elie an at ek (15) illustrate the form and features of the various life stages of myxomycetes. The line drawings are of high quality, accurate and illustrate the habit of the fruiting a the various structural parts such as peridial dehiscence, See nd spores. The light photomicrographs illustrate che capillitium and s s. Unfortu- nately, some of these illustrations are at too low a magnification to show ee features in geen such as the pseudocap1 —_— licium of Lycogala epidendrum in Figure 2-8 and the crys- tals of Mucilago crustacea in Figure 6-4. The water color plates in many cases appear flat and lack the ee ee colors that characterize many species of Myxomycetes. Examples of this are Elaeomyxa miyazakiensis (a rare species in the United States), Lamproderma sautert (restricted to alpine areas) and Diachea leucopodia, anc jon in the latter two species the Sioa 16(3): 601. 1995 602 Sipa 16(3) 1995 cence or metallic sheen is not apparent. The beautiful and accurate rendition of the white Diderma effusum plasmodiocarp is an exception and contrasts nicely against the brownish surface of a leaf. References, a comprehensive glossary and an index complete the work. The glossary has 159 terms that are defined. There is a well balanced selection of 150 references on the biology, ecology and floristics of Myxomycetes that represent publications mostly since 1960. Many of these references do not appear in other books on the Myxomycetes. The text is carefully edited and free from typographical errors. The book is apparently only avail- able in a hardback edition and is rather expensive for a book of 183 pages. Persons vencur- ing into the wonderful world of Myxomycetes will find this book a welcome addition and helpful in picture keying many of the more common species. —Harold W. Keller, Research Associate, BRIT. SMITH, Epwin B. 1994. Keys to the Flora of Arkansas. (ISBN 1-55728- 312-5, pbk.). The University of Arkansas Press, Fayetteville, AR 72701. $30.00. 363 pp, 7" x 10". A brief introduction notes that the author has devoted more than 25 years to the study of the Arkansas flora. This publication follows the author's 1988 “An Atlas and Annotated List of the Vascular Plants of Arkansas” that provided synonymy, distribution by county dot maps, and chromosome numbers. The dichotomous keys are designed specifically for the Arkansas flora. There are general keys to the ferns, gymnosperms and flowering plants; the latter divided into monocots and dicots. Dichotomous keys lead to the families of ferns and fern allies, families of gymnosperms, the dicot families with a series of separate keys leading to the aquatic dicots, woody dicots with leaves absent at flowering time or par- tially expanded, woody plants with leaves opposites or whorled, woody plants with leaves alternate and plants herbaceous and terrestrial. The keys ultimately identify genera and species represented by 2,518 taxa that are presently known for Arkansas with an addi- tional 300 species most likely to occur. The author requests voucher specimens for any of the possible additional species since these would represent new state records. The families are arranged according to the old Engler and Prantl system traditionally used in floras. The genera are arranged alphabetically within families. Common names are provided for families and genera but no synonymy is given. ma A glossary defines many of the terms used in the keys with examples of representative taxa given in a few cases. More examples of taxa that illustrate a given defined term would make the glossary even more user friendly. The four corners of the outside edge of the book are rounded to prevent the cover and pages from becoming dog-eared from use over time. This book represents an important contribution toward a comprehensive flora of Arkansas that should eventually include keys, species descriptions, and illustrations to all of the known species in the state—Harold W. Keller, Research Associate. BRIT. Sipa 16(3): 602. 1995 Book REVIEWS 603 NATIONAL RESEARCH COUNCIL, COMMITTEE ON THE FORMATION OF THE Na- TIONAL BIOLOGICAL SuRVEY. 1994. A Biological Survey for the Na- tion. (ISBN 0-309-04984-0, pbk. 2nd printing). Academy Press, 2101 Constitution Avenue, N.W., Washington, D.C. 20418. $26.00. pp. 205,06" Oo": he mission of the National Biological Survey is “to gather, analyze, and disseminate the information necessary for the wise stewardship of our Nation's natural resources, and to foster an understanding of our biological systems and the benefits they provide to soci- ety. The NBS will act as an independent science bureau without advocating positions on resource management issues and without regulatory or land and water development au- thorities.” The National Research Council, in an advisory role to the Secretary of the Inte- rior, conducted a study on the formation of the NBS. This book represents the findings of the Committee on the Formation of the NBS. This commictee consisted of 19 people from diverse backgrounds chaired by Peter H. Raven from the Missouri Botanical Garden. The central theme of the book provides the answers to a series of questions: 1. What should a biological survey for the nation entail? 2. What should the NBS in the Depart- ment of Interior (DOI be to serve the needs of the department and others: 3. How should information relevant to the survey be managed? 4. How can existing and new survey re- lated activities and information be made most useful for policy management, and scien- tific purposes within and outside the DOI? 5. How can federal and other entities best collaborate for these purposes? The NBS will be a new administrative entity in the DOI but formed from preexisting programs. Ic will have responsibilities for inventorying, map- ping, and monitoring biotic resources; performing basic and applied research on species, groups of species, populations, and ecosystems; and providing the scientific support and technical assistance needed for managerial and policy decisions by DOI. There is a preface and executive summary that outlines the parameters of the NBS, followed by five units: 1. The Value of a Biological Survey for the Nation; 2. Meeting the Information Needs of NBS clients; 4. Coordination of the National Partnership for the Biological Survey; 5. Implementation. What are the needs for the NBS? The following represent the issues outlined: finding ways to preserve the nation’s biological heritage; managing biological resources in a sus- tainable manner; maintaining essential ecological services, such as water supply, flood and erosion control, and climate amelioration; understanding the impact of human settlement patcerns on biological resources, maintaining contributions of our nation’s biota to the aesthetic quality of life, understanding che effects of climate change; deriving new eco- nomic wealth from biological resources; restoring degraded environments. To create the NBS, the Secretary of the Interior has proposed combining a substantial portion of the biological research and survey activities in three DOI bureaus —the Fish and Wildlife Service, the National Park Service, and the Bureau of Land Management — with smaller portions of five other departments — the Minerals Management Service, the Office of Surface Mining Reclamation and Enforcement, the Bureau of Reclamation, the U.S. Geological Survey, and the Bureau of Mines. National Partnership for Biological Survey is proposed that will be able to conduct creditable science; be able to stimulate and coordinate appropriate research; be organized for program continuity; be user friendly and adapted to a variety of users. The key scien- tific objectives should include: determining what biotic specimens and data exist in our Sipa 16(3): 603. 1995 604 SIDA 16(3) 1995 nation’s institutional collections; discover, describe, classify and map species of selected taxa; establish taxonomic specialists, collections, and databases for large and important taxa; study the biology of selected species of importance; develop classification systems for ecological units and a set of core ecosystem attributes and protocols; develop predictive models to facilitate sustainable management; perform research on the restoration of de- graded environments; perform research to develop biological protocols for pollution; es- tablish collaborative pilot projects for interdisciplinary research on biological resources in selected regions. The book uses bold face type to highlight paragraphs and a series of 38 recommenda- tions. The reader can quickly scan the recommendations and grasp the overriding themes of this book. After certain topical headings, bullets are used to highlight specific points. One of the specific recommendations of the committee was to develop a National Biotic Resource Information System. This effort would develop a system of shared databases de- signed to make existing information more accessible and to establish mechanisms for effi- cient, coordinated collection and dissemination of new informacion. It is interesting to note that none of the distinguished panel members assembled to conduct this scudy had a extensive research background in fungal biology. Yet it is clear in a number of sections of the book that our understanding of fungi and the specific expertise his point is addressed for many groups to identify fungal taxonomic groups is lacking in many cases. T under che section entitled Prese rving the Nation's Biota, for example, “... of organisms (e.g., fungi, mites, nematodes, and marine invertebrates), most species have named, even within the borders of the United States.” not yet been described and ormation on committee members and profes- Appendix A: gives the biographical inf sional staff; Appendix B: gives the national biological survey Fiscal Year 1994 budget justification; and the Index is helpful in locating , topical subject matter. This book should . required reading for anyone who wishes to be well informed about the future of the biota of the United States of America.—Harold W. Keller, Research Associate, BRIT. Vines. (ISBN 0 521 39250 0, hbk.). Press Syndicate of the Univer- sity of Cambridge, 40 West 20th Street, New York, NY 10011-4211. Price unknown. 526 pp, 6 3/4" x 10", In che Foreword Paul Richards notes that vines have attracted relatively little scientific attention save for Charles Darwin in his Movement and Habits of Climbing Plants (1875) and ealing mainly with stem anatomy (1892-93). Purz, Francis E. and Harotp A. Mooney, eds. 1991. The Biology of Schenck in Germany with his two mem« This book should fill a gap in the general I biol logy of vines. It is divided into five parts: an introduction devoted to the distribution and evolution of climbing plants; climbing me- chanics and stem form; vine physiology and devel lopment; community ecology of vines; economic importance of vines. Each of the contributed chapters have a summary or con- cluding remarks that enable the reader co assimilate the salient poin The numbers of climbers are impressive. A total of at least 133 ee include a few climbers. At least 97 seed plant families have climbers that are found in the New World. The three largest families of climbers in the New World are the Asclepiadaceae with 1,000 species, the Convolvulaceae with 750 and Leguminosae with 7: he contribution by Oliver Phillips entitled The Bipnoboran and Economic Botany of Troperal Vines has a treasure trouve of information about vines that will be of interest to the Sipa 16(3): 604. 1995 Book Reviews 605 general reader. Factoids are cited such as 80% of the population of developing countries rely on traditional medicine for primary health care. About a billion people that live in industrialized nations take about 40% of their prescribed medicines that contain chemi- cals originally isolated from plants. Many tropical vines are harvested for their fiber con- tent and used in cordage, baskets, hammocks, and fishing nets. The best known fibrous lianas are the rattans (climbing palms from Southeast Asia). Starchy tubers from Dioscorea and Ipomoea make up a major contribution to the diet of many people in the tropics. There are 16 families listed and each with genera that have useful climbing plants in the tropics. Several examples will suffice. Philodendron provides the widest range of uses among aroid lianas. In Colombia P. craspedodromum Schultes is used by the Desana Indians for a fish oison. Other species of Philodendron are are used for treating snake-bites, removing warts, for skin rashes and used as a female contraceptive. The Arecaceae (the palms) provide spices, waxes, gums, poisons, food, shelter, fuel, fiber and medicines. The Bignoniaceae is the most speciose neotropical liana family and has an impressive array of ethnobotanical uses. This family has been scarcely sampled by chemists for biologically active medicinal com- pounds. The Jivaro in northern Peru use an Arrabidaea as a cure for thrush, a fungal disease of the throat and mouth. The juice extracted from the root of Martinella is used by many different Indian eribes in different places to treat eye inflammation or irritation. The Convolvulaceae is mostly a tropical climbing family that is best known for its edible tu- bers (sweet potato). Ipomoea tricolor (cultivated as ‘Morning Glory”) from tropical America has egoline alkaloids in its seeds. This was a sacred plant of the Aztecs and became popular in the 1960's for its LSD-type effects. The Cucurbitaceae produce fruits that have signifi- economic importance as food plants, for example, cucumbers, squashes, pumpkins, — and watermelons. There are many other examples to numerous to mention that could be cited. This chapter is valuable as a reference source for information presented in tabular form (genus name, use as food or medicinal properties and author - literature reference) for the families and genera of climbers There is a general subject ie and also a taxonomic index that facilitates locating specific taxa in the book. This book should be on the bookshelf of every botanist that has an interest in vines or the collection of plants in tropical areas of the world.—Harold W. Keller, Research eons BRIT: CaLttaway, M. Bretr and CHartes A. FRANcIs, eds. 1993. Crop Improve- ment for Sustainable Agriculture. (ISBN 0-8032-1462-, hbk.). University of Nebraska Press, 901 N. 17th Street, Lincoln, NE 68588- 0520. $35.00. 261 pp., 6" x 9 1/4". The primary focus of this book is the adaptation of crop plants to agricultural systems and prevailing conditions rather than the expensive modification of the field environment to fit current crops. The crop improvement theme is presented in a series of 11 chapters: Crop Improvement for Future Farming Systems; History of Crop Improvement in Sus- tainable Agriculture; Choosing Germplasm for Breeding Program Success; The Role of Seed Companies in Crop Improvement; Crop Breeding Objectives and Methods; Breeding for Resistance to Insects and Plane Pathogens; Crop Tolerance to Weeds; Tree Improve- ment for Agroforestry Systems; Contributions of Biotechnology to Crop Improvement; Genotype by Environment Interaction in Crop Improvement; Statistical Design and Analysis of Intercropping Experiments. Sipa 16(3): 605. 1995 606 Sipa 16(3) 1995 Early in the book there is a review of definitions used for sustainable agriculture that range from Wendell Berry's, the most desirable systems are those that degrade “neither land nor people”, to Francis and Youngberg’s “Sustainable agriculture is a philosophy based on human goals and on understanding the long-term impact of our activities on the envi- ronment and on other species. Use of this philosophy guides our application of prior expe- rience and the latest scientific advances to create integrated, resource-conserving, equi- table farming systems. These systems reduce environmental degradation, maintain agricultural ee promote economic viability in both the short and long term, and maintain stable rural communities and quality of lif Crop improvement in the future will place new saaece on tolerance to biotic and abiotic stress. There is a discussion of stress tolerance to cold, prolonged heat and drought. Special, potential new crops such as perennial cereal grains and legumes, as proposed by Wes Jackson at The Land Institute, holds promise for a perennial polyculture that would be represented by multiple species of plants with different life cycles, nutrient, and water needs, and contributions to food and feed supply. This cropping system would have the added advantage of providing soil protection by a permanent ground cover, low cost main- tenance and less tillage, and restructuring of the soil profile found in a native prairie eco- systern Each chapter ends with a set of references that for the most part are based on publica- tions that have appeared in the last ten years. Readers will find a wealth of information and Qu — ideas about breeding programs that will lead to a more sustainable agriculture as an alter- native to the high use of chemical pesticides, monocultures, and high tillage practices chat ultimately degrade the environment.—Harold W. Keller, Research Associate, BRIT. PaHLow, MANNFRIED. 1978. Healing Plants. (ISBN 0-8120-1498-7, pbk.). Barron’s Educational Series, Inc., 250 Wireless Boulevard, Hauppauge, NY 11788. $16.95. 224 pp, 86 full-color photographs, 6 1/2" x 9 1/4". This book focuses on healing plants used for the self treatment of the most common ailments, complaints, and disorders such as the common cold, sleep dysfunctions, and stomach complaints to mention a few. The organization and narrative are written for a general audience. The introductory pages serve as a guide and warning about the proper treatment procedures. There are four rules to follow given as topical headings and along with an explanation: |. Evaluate your complaint properly; 2. Select the application care- fully; 3. Prepare and administer the application as directed; 4. Observe the limits of self treatment. The table of contents consists of self-help with healing plants; nervousness and sleep disturbances; colds; bladder and kidney complaints; stomach and intestinal com- plaints; rheumatism and gout; gallbladder and liver complaints; childhood diseases; gyne- cological complaints and menopause; geriatric complaints; cardiac and circulatory disor- ders; minor injuries, skin irritations; compendium of medicinal plants; appendix There are precautions and warnings that che reader assumes full scones for ac- tions ve and the admonition to always consult with a physician before treating yourself with healing planes. But these warnings may go unnoticed because in some cases they are not highlighted with oversized and boldfaced letters that stand out against the rest of the text narrative. The self-treatment preparations derived from medicinal plants include teas, tinctures, baths, iahdlntions compresses, poultices, and washes. The examples selected by Sipa 16(3): 606. 1995 Book REVIEWS 607 the author are primarily those medicinal plants that have been subjected to extensive sci- entific research and are recognized as possessing therapeutic effects, and, in most cases, approved as natural medicines by the German Office of Healt There is an informative section on what you need to know ous medicinal plants and includes a topical section on - how do medicinal plants act? and - medicinal plants and the substances they contain. Information is given about how to store medicinal plants cor- rectly, weighing, mixing, and measuring plant substances, collecting medicinal plants yourself, and how to grow plants under the section compendium of medicinal plants. About half of the book deals with self-help treatments organized as units that focus on complaints or ailments. Rheumatism and gout will serve as an example of these units. There is a section that defines these metabolic diseases and the help given from healing plants. The healing plants used to provide relief and efficacy and the preparation proce- dures before application or use are described in — detail. The plant preparations are based on specific recipes and dosages. The direc 1 instructions for using plant prepa- rations appear detailed and specific. he last half of the book is a Compendium of Medicinal Plants arranged perpnieccn by common name. Each plant or distinctive plant part is illustrated in color as a quarter 0 half page photograph. The color reproduction is of outstanding quality. The ee name is given for each plant, requirements for cultivation, uses of the plant in general and in folk medicine and in some cases interesting facts pertaining to historical accounts of the plant The appendix contains a section on using healing plants properly and instructions for preparing and applying teas. There is an index that has common and scientific names and subject words that facilitates finding information. A brief glossary of 74 terms, mostly botanical, helps the lay reader to understand the boranical descriptions. There is a section on sources of herbs, sources of herb seeds, and sources of information that gives names and addresses. There is a section for further reading that has a list of 11 books and four ad- dresses for journal publications. This book will appeal to readers with a general interest in natural medicines, to garden- ers who may wish to grow these plants, and botanists who wish to collect medicinal plants for purely botanical reasons.—Harold W. Keller, Research Associate, BRIT. Arvico, Rosirra and MicHaeL Batick. 1993. Rainforest Remedies, One dred Healing Herbs of Belize. ISBN 0-914955-13-6, pbk.). Lotus Press, Box 325, Twin Lakes, Wisconsin 53181 $9.95. 219 pp, 100 black and white line drawings (text illustrations by Laura Evans), size 5 1/4" x 8 1/2" This book chronicles the work of The Belize Ethnobotany Project initiated in 1987 that has recorded the author’s findings about plants used as medicines, foods, fiber, in construc- tion and agriculture, during religious ceremonies, and as part of spiritual beliefs. This project has resulted in the creation of The Ix Chel Tropical Research Foundation, dedi- cated to traditional medicine, ethnobotanical studies, Belizean culture, and rainforest con- servatio The ee pages describe the country of Belize under the topical heading back- ground. The methodology and purpose, scope of book, and organization of text and illus- trations explain the formatting of the descriptions and arene of plants that follow. Sipa 16(3): 607. 1995 608 Sipa 16(3) 1995 The English common name is given in bold and capital typeface at che top of each page and used as the primary means of identification . The plants are arranged alphabetically ac- cording to these names. Fortunately the scventific name and author are also given for each plant to eliminate confusion. The plant family name is given according to the classification system of the late Arthur Cronquist. Under the heading description Se a brief feld description of the plant giving the habit, type of leaves, inflorescence and fruits. The habi- tat represents the most common areas where the Bia is found either in os wild or in cultivation. Traditional uses represents a compilation of information from healers and per- sonal observations of plant use in Belize. Research resw/ts are the results of laboratory re- search with appropriate references cited. [//ustrations are full page black and white line drawings which illustrate one or more key features useful in the field identification of the plane. There is a glossary of medical and botanical terms that is helpful to better under- stand the plane descriptions and us jon There is a 12-page list of ee mostly from the last 15 years. A number of the journals cited are published in South and Latin America and this adds to the utility of the book, especially for many in the United States. There is an index of scientific and common names. The authors have made a valuable contribution to preserve the traditions and customs of the indigenous peoples of Belize. The authors have donated the proceeds from the sale of this book to support the native traditional healers and ecosystem preservation in Central American rainforests. The sensitivity, concern, and goodwill shown by Rosita Arvigo and Michael Balick for native peoples represents a model for others to emulate.—Harold W. Keller, Research Associate, BRIT. Sipa 16(3): 608. 1995 609 ANNOUNCEMENT Volume | is on the way! Volume 1| of the 12 volume Useful Wild Plants of Texas, the Southeastern and Southwestern United States, the Southern Plains, and Northern Mexico will be available in the fall of 1995. Co-authored by Scooter Cheatham and Marshall Johnston, this series is a mammoth quest to sys- tematically chronicle the economically useful plants of the region centered in Texas and radiating through the southern half of the United States and the northern part of Mexico. Information on these plants has been gathered not only from the area itself, but draws on the extensive use of these and closely related species around the world. This work underscores Texas’ unique position in North America and the need to “save the rainforest in our own backyard.” Volume 1 includes 86 genera from Abronia through Arundo. Volume 2 will begin with the genus Asc/epias. The complete series will include de- scriptions, photographs, distribution maps, and use information on over 3,000 species of native or naturalized Texas plants. This will be a hands-on resource for botanists, field researchers, forest- ers, ranchers, landowners and leaseholders, archeologists, teachers, biochem- ists, chefs and many other professions and interests, with its extensive cov- —_— erage of thousands of uses of native plants for food, medicine, pharmaceuticals, oils, rubbers, fuels, and fibers, and many other domestic, industrial, and commercial uses. It underscores the need to develop new crops and future resource bases from native plants. In addition to the ecological, recreational, aesthetic, and ethical argu- ments for conservation, this work provides an economic basis for arguing for the preservation of land and the plant life that is inextricably linked to the land. This work will also support environmental impact statements, the development of wise land-use policies, and the development of educa- tional curricula. This series is being published by Useful Wild Plants, Inc., For informa- tion on ordering direct correspondence to Useful Wild Plants, Inc., Dept. B95, 2612 Sweeney Lane, Austin, TX 78723, USA, or call 512-928-4441. Specs: 592 pages, 304 color photos, 268 color distribution maps. Stipa 16(3): 609. 1995 610 Sipa 16(3) 1995 ANUNCIO jE] Volumen 1 esta en camino! EI primero de los 12 voltimenes de Useful Wild Plants of Texas, the Southeastern and Southwestern United States, the South- ern Plains, and Northern Mexico estaré disponible en el otofio de 1995. Los autores son Scooter Cheatham y Marshall Johnston. Esta serie es una busqueda gigantesca para hacer la crénica sistematica de las plantas con utilidad econdédmica de la regi6n que teniendo su centro en Texas se extiende por la mitad sur de los Estados Unidos y la parte norte de México. La informacion sobre estas plantas ha sido recogida no sélo en este Area sino que se recurre al uso extensivo de estas especies y otras muy relacionadas de todo el mundo. Este trabajo subraya la posicién Unica de Texas en Norteamérica y la necesidad de “salvar la pluvisilva en nuestro propio pa- tio”. El volumen | incluye 86 géneros, desde Abronia hasta Arundo. El volumen 2 comenzara con el género Asclepias. La serie completa incluira descripciones, fotografias, mapas de distribucién, e informacién sobre el uso de mas de 3.000 especies nativas o naturalizadas en Texas. Sera una fuente de transmisi6n para botanicos, investigadores de campo, forestales, rancheros, propietarios y arrendatarios de tierras, arquedlogos, profesores, bioquimicos, cocineros, y otras muchas profesiones e€ intereses, con su extensa cobertura de miles de usos de plantas nativas para alimento, medicina, farmacia, aceites, gomas, combustibles, fibras y muchos otros usos domésticos, industriales y comerciales. Subraya la necesidad de desarrollar nuevas plantas de cultivo y futuras bases de recursos a partir de plantas nativas. Ademias de los argumentos ecolégicos, recreativos, estéticos y éticos para la conservacion, este trabajo ofrece una base econémica para apoyar que la conservacin del territorio y la vida vegetal, que esta indisolublemente ligada a dicho territorio. Este trabajo también tendra utilidad en la evaluacién de impactos ambientales, el desarrollo de politicas de uso del territorio prudentes y sabias, y el desarrollo de curricula educacionales. Esta serie se publica por Useful Wild Plants, Inc. Para informacién sobre los pedidos, dirija su correspondencia a: Useful Wild Plants, Inc., Dept, B95, 2612 Sweeney Lane, Austin, TX 78723, USA, o lame al 512.928.4441 Especificaciones: 592 paginas, 304 fotos en color, 268 mapas de distribucién en color. Sipa 16(3): 610. 1995 A synopsis of the genus Clusia sections Criuvopsis and Brachystemon (Clusiaceae) in northern South America John J. Pipoly HL and Allison Graff 5 5( Subtribal classification of the New World Eragrostideae (Poaceae: Chloridoideae ) Paul M. Peterson, Robert D. Webster and Jesus Valdes-RKeyna 529 Marsilea minuta (Marsileaceae): New to Florida and North America James R. Burkbalter Taxonomy of the native North American species of Saccharum (Poaceae: Andropogoneae ) Robert D. Webster and Robert B. Shaw Noteworthy plants from north Florida. VI Loran C. Anderson 581 Notes Aristida desmantha (Poaceae), new to Missouri 389 Malaxis wendtii (Orchidaceae) in the United States 591 Carex amplifolia and Carex rossti (Cyperaceac), new to New Mexico and a key to section Montanae in New Mexico 592 Saloinia minima (Salvinaceae), new to Texas 595 Rbynchospora capitellata (Cyperaceae), new to Kansas 595 Possible eponymy of the generic name Brasenia Schreb. (Cabombaceae) 597 JO, 406, 412, 420, 438, 408, 476, 504, 528, 550, 588, 601-608 Book reviews and notices 4 Announcement, Anuncio 609, 610 ISSN 0030-1488 CONTENTS A revision of Mirabilis section Mirabilis (Nyctaginaceae) Alice Le Duc 613 The genus Clusia section Criuva (Clusiaceae) in Guayana John J. Pipoly Hl and Allison Graff 049 based on multivariate analyses of Taxonomy of ile (Nyctaginaceae geographic variatio Matthew Mabrt and Richard Spellenberg A synopsis of the genus Packera (Asteraceae: Senecioneae) in Mexico Craig C. Freeman and Theodore M. Barkley 699 Historical evidence of the native presence of Sabal mexicana (Palmae) north of the Lower Rio Grande Valley Landon Lockett 7i1 Taxonomic notes on new varieties of species of Clitoria (Leguminosae - Phaseoleae - Clitoriinae) Paul R. kantz 721 Two new species of Ichthyothere (Heliantheae: Asteraceae) from Ecuador and Peru Harold Robinson 731 Clusia fabiolae, a new species, with: sv nopsis of Clusia section Anandrogyne (Clusiaceae) in Guayana Jobn J. Pipoly II 737 The status of era perfoliatus (otamoy«\onaceae ) in Lake Pontchartrain, Lou Jolin W. Burrs fr, us Poirier and Kris P. Preston Documented chromosome“aumbers 1995: 1) Chromosome number of oe 7 _ Cornus sessilis (Cornaceae); Phylogenetic affinity and evolution of chromosome CONTRIBUTIONS TO BOTANY numbers in Cornus On u-Yun Xiang and Richard H. Hyde 705 VOLUME 16 NUMBER 4 DECEMBER, 1995 CONTRIBUTIONS TO BOTANY FOUNDED BY LLOYD H. SHINNERS 1962 Wm. E. Mahler Publisher 1971-1992 Director Emeritus S.H. Sohmer Director Barney L. Lipscomb Editor Botanical Research Institute of Texas, Inc. 509 Pecan Street Fort Worth, Texas 76102-4060, USA 817 332-4441 / 817 332-4112 FAX Electronic mail: sida@brit.org Home page at the URL: http://www.brit.org/sida/ John W. Thieret Prof. Dr. Felix Llamas Associate Editor Contributing Spanish Editor Biological Sciences Dept. Dpto. de Botanica, Facultad de Biologia Northern Kentucky University Universidad de Le6n Highland Heights, Kentucky 41076, USA E-24071 Leon, Spain Guidelines for contributors are available upon request and on the inside back cover of the last issue of each volume. Subscription per year: $25. Individual, $35. USA Institutions, $45, Outside USA; numbers issued twice ay aS) x G © SIDA, CONTRIBUTIONS TO BOTANY, Volume 16, Number 4, pages 613-809. Copyright 1995 Botanical Research Institute of Texas, Inc. Printed in the United States of America ISSN OO36-1488 A REVISION OF MIRABILIS SECTION MIRABILIS (NYCTAGINACEAE) ALICE LE DUC! Department of Botany University of Texas BUN, ES FEPLS Oaks ABSTRACT The genus Miradilis includes the formerly recognized genera A//ionia in part, Hesperonia, Oxybaphus and Quamoclidion. It is comprised of about 60 species of tropical and temperate herbs distributed primarily in the Americas. Currently the — is arranged in six sections, generally along the Roane of the former genera. This study considers the section Mirabilis. Based on comparative morphology, including extensive field studies and obser- mpa vations in the greenhouse, scanning electron microsopy studies of ee and fruit (anthocarp) characters, two species, M. ae M. poloni and one variety M. sanguinea var. breviflora are described as new. In addition, one new combination is proposed, M. gracilis. RESUMEN El género Mirabilis incluye los géneros anteriormente reconocidos A//ionia en parte y Hesperonia, Oxybaphus y Quamoclidion. Esté compuesto por suave 5 60 especies de plantas tropicales y clima templado, distribuido principalmente en las Américas. Actualmente el género esta organizado en seis secciones noe dentro de los limites del anterior género. Este estudio considera la secci6n Mirabilis. Esta basado en n mortologia comarativs, incluyendo estudios extensivos de campo, observaciones en invernadero, caracteristicas del polen y fruto (antocarpo) estudiadas con microscopia tee se describen dos nuevas especies M. ee M. polonti y una variedad M. sanguinea var. breviflora. Ademas se propone una nueva combinacion: ML. gracilis. The genus Mierabilis, in the family Nyctaginaceae, comprises approxi- mately 60 species distributed primarily in tropical and temperate regions of the Americas. A large number of species are centered in the warm tem- perate regions, especially the deserts, of North America. Several species occur in Mesoamerica, some of these extend into northern South America, while other species are exclusively South American. One species, M. himalaica (Edgew.) Heimerl is reported from the Himalayas, the only spe- cies native outside of the western hemisphere. Several of the American spe- cies are common and widespread; others appear as localized endemics. His- torically, generic and specific delimitations of Mirabilis have varied. Mirabilis was formally proposed by Linnaeus in 1753. The first synopsis of the genus was provided by Choisy (1849). He recognized in the genus Mirabilis only 'Present address: Department of Horticulture, Forestry and Recreation Resources, Kansas State University, 2021 Throckmorton Plant Science Center, Manhattan, KS 66506. Stipa 16(4): 613-648. 614 Stipa 16(4) species which today are included in Mirabilis section Mirabilis. His treat- ment regarded as distinct the genera Quamoclidion Choisy and Oxybaphus LHer. The treatment by Asa Gray (1859) recognized only Mérabilis and Oxybaphus as distinct genera. Gray separated Merabilis (incl. Quamoclidion ) and Oxybaphus on characters of the involucre, stamen number, and fruit. Beginning in 1889, Dr. Anton Heimerl of Vienna, probably the foremost authority on this group, contributed several excellent discussions and treat- ments, in over four decades of study (1889, 1897, 1934). Standley (1909) considered Heimerl’s treatment of Mirabilis to be exceedingly conserva- tive, largely because the latter considered the long recognized genus A//ionia Loefl. (Oxybaphus) to be only a section of Mirabilis. Standley’s treatment (1909, 1911, 1918) of the North American species used characters of fruit, involucre, and flowers to recognize A//ionia, Allioniella Rydb., Quamoclidion Hesperonia Standley, and Mirabilis as genera. However, by the 1930s both Standley (1931la, 1931b) and Heimerl (1934), recognized but a single ge- nus, Mirabilis. Standley, reflecting on his new treatment, said, “If only the species of North America are considered, such genera as Oxybaphus, Quamoclidion, and Hesperonia seem to be differentiated by good and con- stant characters, but as so often happens, when extralimital species are taken into account, the characters supposed to separate the groups break down. It seems necessary, therefore, to follow Heimer! in considering all the plants of the group as representing a single genus.” Most subsequent workers in North America, and several workers before Standley’s time (Gray 1859; Jepson 1914; Macbride 1918), have recognized the segregate genera, al- beit as subgenera within an expanded M/érabilis. Shinners (1951) consid- ered the species of Oxybaphus to be quite distinct, but satisfactory as a sub- genus of Mirabilis. Pilz (1978) maintained Quamoclidion as a subgenus, but reduced the heretofore regarded monotypic genus Hermidium S. Wats., to subgeneric rank within Merabi/is. The results of this study and additional studies by the author (Le Duc 1993) support the aforementioned expanded genus Mirabilis. However, after spending three summers in the field in Mexico (from Neuevo Leon to Chiapas) and day to day observations of collected plants grown in the green- house for four years, I support a sectional treatment. No set of characters is distinctive enough to elevate any section to subgeneric level when such species as M. triflora, M. exserta, M. urbanii, and M. sanguinea are taken into account. Mirabilis L., Sp. Pl. 1:177. 1753. Nyctago Juss., Gen. 90. 1789. Tyre: Mirabilis jalapa \.. Herbaceous perennials, stems erect, semidecumbent or decumbent, simple or branched from the base, with a pseudodichotomous branching Le Duc, Revision of Mirabilis section Mirabilis 615 pattern; roots often tuberous. Leaves opposite, petiolate or sessile, the blades variously shaped, linear, lanceolate, ovate, obovate, cordate or round, gla- brous or pubescent, some glandular, green or glaucous. Flowers axillary or in terminal inflorescences, or both. Involucres 1-many flowered, 5-lobed, sometimes enlarged and membranous in fruit. Perianth constricted above the ovary, the tube campanulate, funnelform or salverform. Stamens 3-6, unequal in length, filiform, incurved, united at the base into a fleshy cup around the ovary. Anthocarps usually 5-angled or 5-ribbed, glabrous or pubescent. Perisperm mealy. Base chromosome numbers, reportedly x = 20; 29, 33. Adapting Hooker's (1880) treatment, Heimerl (1934) defined six sec- tions of the genus Mirabilis as follows: 1. Section Mirabilis. Involucres + narrowly campanulate, 1-flowered, slightly accrescent in age. Perianth conspicuous, funnelform or nearly (tu- bular) salverform, the limb expanded. Stamens 5. Anthocarps ellipsoid, + pubescent, + angular or ribbed, surface smooth or warty, not mucilaginous when wet (Mirabilis sensu stricto). 2. Section Watsoniella. Involucres + narrowly campanulate, 1-flowered, lobes unequal in length, slightly accrescent in age. Perianth slender, the tube becoming wider above, the margin plain, scarcely lobed. Stamens 3. Anthocarps ellipsoid, short hairs, 5 obtuse ribs, tuberculate, constricted at the base (monotypic M. watsontana Heimer). 3. Section Quamoclidion (Paramirabilis). Involucres broadly campanulate, —12 flowered, slightly accrescent in age. Perianth broadly funnelform to funnelform-campanulate, the tube consistently longer than broad, deeply constricted just above the ovary, the limb expanded. Stamens 5. Anthocarps obovoid, ellipsoid to almost spheroid, consistently glabrous. (Hiemerl ex- cluded M. triflora.) Following the Int. Code of Botanical Nomenclature (1989), Article 22, Section Paramirabilis of Heimerl becomes Section Ouamoclidion, the first valid publication at the sectional level (Hooker 1880). 4. Section Mirabilopsis. Involucres broadly campanulate, 2—3 flowered, slightly accrescent in age. Perianth campanulate-funnelform, consistently longer than broad, deeply 5-lobed. Stamens 4-5. Anthocarps obovoid, ob- tuse apex, obtuse ribs and narrow furrows, fine pubescence, mucilaginous when wet (monotypic M. coccinea (Torr.) Hook.). 5. Section Oxybaphus. Involucres campanulate 2—3(-1) flowered, very accrescent, membranous, flattened, lobes equal in age. Perianth campanu- late, funnelform or almost rotate (deeply constricted just above ovary), the tube lacking or very short. Stamens 3—5. Anthocarps ellipsoid, obovoid or clavoid, 5 + strong ribs, mostly pubescent, base truncate, mucilaginous when wet (incl. Ad//ionia in part). 6. Section Oxybaphoides. Involucres campanulate, 1-flowered, lobes equal, 616 Stipa 16(4) only slightly accrescent in age. Perianth campanulate, funnelform or al- most rotate, (deeply constricted just above ovary), the tube lacking or very short, seldom narrowly campanulate. Anthocarps ellipsoid or obovoid, sur- face nearly always glabrous, rough or somewhat angled, mucilaginous when wet (incl. Oxybaphus in part, Hesperonia, & Alfioniella). KEY TO SECTIONS OF THE GENUS MIRABILIS ou 1. Involucre 1-flowered, only slightly enlarged in frui nthocarp mucilaginous when wet; ee perennials; roots not tuberous. Section Oxybaphoides sear ii not mucilaginous when wet; perennials; roots tuberous. 3. Stamens 3; perianth limb not noticeably lobed. ...........000.... Section Watsontella 3, ce 5. perianch limb distinctly 5-lobed, (AL. Aintoniorum only shal- lowly lobed.) ection Mirabilis . Involucre 2—3-flowered or more, sometimes much enlarged in fruit. 4, Anthocarp ellipsoid, obovoid or clavoid, glandular or nong — andular pubescent, mucilaginous when wet 5. Involucre only slightly enlarged and not membranous in \ fru stamens ection i trabilopsis 5. sree very enlarged and membranous in fruit; stamens 3 or 5. Section Oxybaphus 4. Anthocarp obovoid, ellipsoid to spheroid, glabrous, mucilaginous when wet, Or not. Section Oxvamoclidion The purpose of this treatment is to provide a difinitive means of identi- fication and circumcription of the species of section Mzrabilis. These spe- cies are characterized by single-flowered involucres only slightly accrescent after anthesis, the perianth conspicuous with limb expanded, and five sta- mens. This study recognizes ten species within the section, two species, a variety newly discribed herein and a variety elevated to specific level. The native distribution is prodominently in Mexico the exception Miradbilrs longiflora var. wrightiana is also found in the Mountain regions of southern Arizona, southern New Mexico and extreme western Texas. Habitat is mainly disturbed or open areas in subtropical deciduous and scrub vegetation. Again, M. /ongiflora var. wrightiana differs, it has a desert to juniper wood- land habitat. Mirabilis jalapa, the common Four O' clock, widely used as a garden plant by the Pre-Columbian people of Mexico and Europeans, has secome a weed in many areas of the world. MORPHOLOGY The morphological characters of taxonomic significance within the sec- tion Mérabilis include: stem size, internodal length; leaf blade shape and attachment; pubescence; inflorescence structure; involucre shape during anthesis and maturation of the anthocarp; periant —w 1 shape and color; stamen length and color, and anthocarp shape, topography and indumentum. Le Duc, Revision of Mirabilis section Mirabilis 617 Section Mzrabilis is comprised of perennials which grade from herba- ceous to suffruticose. Most species have erect or ascending branches. Occa- sionally, branches of M. longiflora, M. sanguinea, M. urbanii are slender and only weakly ascending to semidecumbent. In well established plants, lower stems may be very stout, 4-6 cm in diameter particularly with M. gracilis, M. jalapa, and M. polonii. All species have a pseudodichotomous branching pattern with swollen nodes bearing a transverse line of puberulence. Inter- node length varies from S—7 cm long in M. wrbanii to 13-23 cm in M. sanguinea, with most species having internodes ranging from 7—12 cm in length. Plant height usually varies from ca. 2 dm for M. sanguinea and M. urban to ca. 1.5 m for M. gracilis and M. wie All species have swollen, fleshy, tuberous roots which range from 3.5—4.0 cm in diameter and ca. 12 cm long in seedlings, to 3 dm diameter and 6-8 dm length in older estab- lished plants of M. gracilis, M. jalapa, M. longiflora, and M. pringlei. A cau- dex, 1-8 cm long, may develop above the tuberous root of very mature plants. Leaves are opposite and quite variable in size, large leaves (9-15 cm long) of lower stems to small leaves (2-6 cm long) subtending the inflores- cences. Leaf size is greatly effected by environmental conditions. Leaves that subtend the inflorescences often are quite reduced and lanceolate in M. longiflora, M. sanguinea. Petiole length also decreases toward the shoot tips, with the uppermost leaves sessile or subsessile in M. exserta, M. hintontorum, M. longiflora, M. urbanit. Blade outline of most species is ovate to deltoid, bases vary from cordate in M. exserta, M. longiflora, M. pringlei to truncate or subtruncate and asymmetrical in M. donahooiana, M. gracilis, M. polonii to asymmetrical with the blade grading down the petiole in M. wrbanii. Leaf apex may be short-acute M. exserta, M. /ongiflora, M. pringlei or long-at- tenuate M. donahooiana, M. gracilis, M. jalapa, M. polonii. Pubescence is almost always found on the veins of the upper surface, and may be present on the lower surface as well in M. /ongiflora var. longiflora, M. pringlet, M. sanguinea var. sanguinea, M. urbanii. The absence of pubescence on the undersurfaces of M. /ongiflora var. wrightiana and M. sanguinea var. breviflora is a distinguishing feature. Terminal multiple cymose inflorescences vary from open in M. exserta, M. hintoniorum, M. pringlei to aggregate or glomerate M. jalapa, M. longiflora var. longiflora, M. sanguinea. Often, however, the first flowers are solitary and axillary. The peduncles are pubescent, often densely so in M. donahooiana, M. exserta, M. longiflora, M. poloni, M. pringlet, M. sanguinea, M. urbanii. This pubescence is predominantly glandular in M. exserta, M. longiflora, M. pringlei. Flowers are perfect, involucrate, with one flower per involucre. The five-lobed involucre appears as a false calyx under a petaloid perianth. This gives the flower every appearance of having a symsepalous calyx and a sympetalous corolla. All species have similar campanulate or narrowly campanulate involucres with 5 lobes as long as, 618 Stipa 16(4) or slightly longer than, the fused portion. Involucres, of all species, are only slightly accrescent in age but display variation in shape at anthocarp maturity. Some involucres are rotate M. gracilis, M. jalapa and some are campanulate M. polonii, M. pringlei, M. sanguinea, M. urbanii. In M. longiflora, the involucral lobes are extremely attenuate and valvate until well after the anthocarp has matured. The perianth is composed of a showy petaloid ca- lyx Joshi & Rao 1934) at least twice as long as the involucre. It consists of three sections: the base which is constricted above the ovary, the tube, and the limb (the basal portion persists and encloses the ovary to become the fruit or anthocarp, the tube and limb abscise at the constriction point and fall off after pollination). The tube in most species is funnelform, though in some it is narrowly so, M. donahooiana, M.exserta, M. jalapa, M. sanguinea var. sanguinea; others are distinctly salverform M. gracilis, M. longiflora, M. polonii. Perianth color ranges from white in M. gracilis, M. longiflora, M. polonit to pink in M. exserta, M. pringlei, M. urbanii to lavender, purple and red in M. donahootana, M. jalapa, M. sanguinea. Orange appears only in M. hintontorum and among occasional populations of M. jalapa. The perianth limb terminates in five nearly equal usually broadly obtuse lobes with emar- ginate apices and five nerves which extend along the tube and limb to terminate in tuffs of pubescence. These lobes are induplicate and plicate in bud. Mirabilis pringler has distinctive acute triangular lobes and M. hintoniorum has very obscure lobes. In M. exserta, M. gracilis the shallow emarginate lobes give the perianth limb a ruffled appearance. All species have circinate stamens that are united at the base, forming a collar around the single ovary. This collar may completely contain the ovary in M. /ongiflora and M. pringlei or expose as much as the upper 2/3 of the ovary as in M. urbanii. Above the collar, stamens are free, though most are appressed to the perianth in the region of constriction, and some remain appressed part way up the perianth tube, M. gracilis, M. polonii. Stamens are usually un- equal in length with presentation to one side of the perianth creating a weakly zygomorphic flower. Filaments of most species are lavender to lav- ender-pink, except M. Aintoniornm and the yellow and white flowering forms of M. jalapa, which have filaments the same color as the perianth. In most species, the stamens are well-exserted beyond the throat of the perianth tube, the exceptions: M.donahooiana, M. urbanii, many populations of M. jalapa, and some populations of M./ongiflora var. longiflora. Pollen grains are spheroidal, pantoporate, and the sexine sparsely tubuliferous and spinu- lose. They range in size from 100 mm to 190 mm, (to 210 mm, according to Nowicke 1970). The ovary is superior with a single ovule. The capitate stigma and style, which are longer than the stamens, often remain extended in senescent flowers. The fruit or anthocarp formed from the persistent basal portion of the perianth and the enclosed ovary, may be spheroid, Le Duc, Revision of Mirabilis section Mirabilis 619 elliptic, or oblong; 5-angled or ribbed; glabrous or pubescent; smooth or warty; black, light brown, brown, or orangish brown. In a survey study of Mirabilis anthocarps ( forty of the sixty taxa) (Le Duc 1993) no other sec- tion displayed as much variablility of anthocarp characters between species as did section Mirabilis. The glabrous, elliptic anthocarp with five to ten furrows of M. exserta, is similar to several species of section Quamoclidion (Plate I-4). The anthocarps of M. Aintoniorum, M. sanguinea, and M. urbanii (Plate H-1, 2, 3 & 4) are oblong-ellipsoid, pubescent, 5-angled, ridges tuberculate, with a truncate base and an acute apex. These features are commonly associated with species in section Oxybaphus. The other entities of section Mirabilis ( Plate 1-3, 5 & 6; Plate H-5, 6, 7) display various combinations of characters intermediate between the afore described spe- cies. One significant character distiguishes all species of section Mirabilis from the other sections, a lack of any mucilage production when the anthocarps are wet. (Section Quamoclidion includes the only other species that reportedly do not produce mucilage when wet, but it also includes several species that become mucilaginous.) For most individual species in section Mirabilis the anthocarpal features remain quite constant. However, considerable variation exists in the two species that are known to have been cultivated as garden plants, first by the Pre-Columbia people, and then by the Europeans, M. jalapa and M. Jongiflora. Plate I-1 & 2 and 3 & 5 show two common forms for each of these species. TAXONOMIC TREATMENT Section Mirabilis Hook., in Benth. & Hook., Gen. Pl. 3:1—11. 1880. Tyee: Mirabilis jalapa Herbaceous or suffruticose perennials, erect, ascending or semidecumbent, the root fleshy, the stems slender or stout, puberulent or glabrous. Mid-stem leaves opposite, petiolate; blade thin or slightly succulent, ovate to broadly so; base cordate, truncate or grading into the petiole, veins prominent. In- volucres 1-flowered, + narrowly campanulate, 5-lobed, slightly accrescent in age. Perianth showy, funnelform or nearly salverform, limb with 5 emar- ginate lobes. Stamens 5, circinate before anthesis, the filaments unequal, capillary, connate at base into a sheath about the ovary. Anthocarp ellipsoid, + pubescent or glabrous, + angular or ribbed, surfaces smooth or warty. Not mucilaginous when wet. Base chromosome number possibly x = 29. KEY TO SPECIES OF SECTION MIRABILIS I, Stamens exserted, filaments at least twice as long as perianth. at anth tube slightly swollen above the ovary, the lobes acute. .......... .M. pringlet a8 Pent tube not swollen above the ovary, the lobes obtuse. . Anthocarp broadly dies or ovoid, glabrous; perianth white, ae tinged to pale lavender. 2. M. exserta 620 Sipa_ 16(4) 6 Piare [. 1. M. jalapa—Mexico, Veracruz. Le Duc & Sydor 158 (TEX). 2. M. jalapa—Texas, Travis oo Le Duc s.n. (TEX). 3. ML. ee var sek ae oe Le Duc 222 (TEX). 4. M. exserta—Mexico, Baja Cali eeteraed 43339 (MO). 5. M. longiflora var a Sn Msc. ie Le a 180 @ “EX). 6. Duce & Sydor 71 (TEX). Bar Aci ae Jalisco. Le Le Duc, Revision of Mirabilis section Mirabilis 621 Pate II. 1. TM. 3 SANQHINEA Var. SANG 2357 (TEX). 2. M. sanguinea var. orate =~ MENG, alisco. Le Duc 254 TE ae me ene Michoacan Hinton 13909 Sn _M. press ee oes Le Duc 245 (TEX). 5. M. polonti— Mexico, Sinaloa. Le Duc et al 178 (TEX). 6. M. donahootana—Mexico, Michoacan. Le Duc 247 (LER). 7; i: penal Nees. Jalisco. Le Duc 63 (TEX). Bar = 1.0 mm 622 Stipa 16(4) . Ant gee agh _ pubescent; perianth orange. .......... 3. M. bintoniorum iv ee if exserted, less than twice as long as perianth., 4. Perianth > "6 cm long. . Perianth curved downward (especially before anthesis); leaf base trun- cate; anthocarp obovoid, 6-7 mm long, constricted and truncate at the base and apex. 4. M. polonii 5. Perianth straight, erect (especially before anthesis); leaf base cor- date; anthocarp ellipsoid to obovoid, 7-8 mm long, constricted anc truncate at base only. . M. longiflora . Perianth < 6 cm long 6. Perianth ih aa < orange hair 6. Periant 15 mm long; anthocarp with prominent stiff 6. M. urbanit 1S mm long or longer; anthocarp without stiff 1 aaah es hair . nee 13-23 cm long 7. M. sanguinea 7. Internodes < 13 cm long. j 8. Inflorescence open, few-flowered; anthocarp oblong to ob ae obovate. _ M. donahooiana 8. Inflorescence compact, many-flowered; anthocarp a to ellip 9, Pe nee tube at least twice as long as limb-width; flowers white with lavender staminal filaments; anthocarp ell long 9. Periancth cube | ipsoid 9-11 mm 9. M. gracilis ess than twice as long as limb-width; if flowers white then staminal filaments white; anthocarps ellipsoid to ovoid, 6-9 mm long. 0. M. jalapa 1. Mirabilis pringleti oie Proc. Amer. Acad. Arts 45:424. 1910. (PI. 3-A). Type: MEXICO. Guerrero: under limestone cliffs, Iguala es 23 Jul 1907, Pringle 10384 poise GH); tsorypes: F!, LL!, MICH!, RSA!, UC!). Herbaceous or suffruticose perennials, erect or ascending, | m an much- branched, roots tuberous. Stems slender, finely viscid-pubescent, lower in- ternodes 10-15 cm long, nodes villous. Mid-stem leaves: petiole slender, —15 cm long; blade 3.5-13.0 cm long, 2-9 cm wide, thin, bright green, sparsely puberulent; base cordate or deltoid-ovate, cordate to unequally subcordate, apex acute to short acuminate, margin ciliolate (pink on young leaves). Inflorescences open, terminal, composed of many-flowered cymes these subtended by small, short-petioled leaves. Involucre campanul 5-8 mm long, glandular-puberulent, slightly revolute in age, lobes ca. 2 mm long, obtuse or acute, margin ciliolate; ultimate peduncles 1-6 mm long, densely glandular-pubescent. Perianth 2—3 cm long, sparsely glan- dular-pubescent, white to pink, cubular, slightly swollen above ovary, lobes 9-10 mm long, ca. 5 mm wide, triangular, reflexed after anthesis, apices acute. Stamens exserted, nearly twice as long as the perianth, pink to lav- ender. Style and stigma white. Anthocarp dark brown to grayish brown broadly obovoid to oblong, 5-7 mm long, 4-5 mm wide, 5-angled, sparsely ate, —_ Le Duc, Revision of Mirabilis section Mirabilis ON No Wo I. A. Mirabilis as el ae Due 63, TEX). B. M. polonit (Le Duc 178, TEX). C. M. een (Le Duc 248, M. gracilis (Le Duc 176, TE 624 Stipa 16(4) warty between ribs, slightly puberulent to glabrate, constricted at both ends, base truncate. Phenology.—Flowers from late July to September. Flowers open in the evening. Distribution (Fig. 1).—The Pacific slopes of the Sierra Madre Occidental and western Central Plateau; in full sun to partial shade, crumbly or rocky igneous soil. Altitude 300-2000 m. Representative specimens: MEXICO. Guanajuato: Empalmede, Rusby 118 (NY). Guerrero: Casa Verde, Xochipla, amas de Rio, Rz ese ees (UC). Jalisco: El Corcovado, Bridge over Rio San Pedro, L c & Sydor 6. <); Mezquitan, between Autlan & es eee Le Due & Sydor 70 ne “ 25 mi : of a Grullo & W of Ciudad Guzman, Le Duc 174 (TEX). Mexico: Palmar, Temascaltepec, Hinton 6418 (GH, MICH); Los Cuervos, 8.7 mi NE of Mexico state line along Hwy 130, Le Due et al 238 (T oe Michoacan: WSW of Apatzingdn, road to Dos Aquas & Aguililla, Dieterle 4325 (CA MICH, MEXU),; Tuzantla-Tiquicheo, Zitacuaro, ee ie 0 (F, GH, MEXU, a NY, RSA, UC, ne Puerte San Salvador, 54.3 mi N of La Mira junct. HWY 200 & 37, Le Due et os 244 (T Mirabilis a closely resembles M. exserta and M. hintoniorum in veg- etative characters. It differs markedly from them in perianth structure, having the tube salverform and lobes reflexed. The anthocarps also differ in M. pringle: having few contours or warty areas and they are sparsely covered with short hairs. 2. Mirabilis exserta Brandegee, Proc. Calif. Acad. Sci. H. 3:165 1891. Type: MEXICO. Baja CattroRNiA Sur: “Summits of the spurs of Sierra de San Francisquito,” 20 Oct 1890, T'S. Brandegee 480 (HOLOTYPE: UC!; tsorypes: GH!, US!). Herbaceous or suffruticose perennials, erect, 4-6 dm high, much branched. Stems slender, upper densely viscid puberulent, meee below, internodes 7-13 cm long. Mid-stem leaves: petiole slender 1.0—2.5 cm long; blade 6-11 cm long, 4.0-10.5 cm wide, thin, bright green, eens: lent when young, soon glabrate, base ovate-orbicular, broadly ovate-del- toid or cordate-ovate, subcordate or truncate, apex acute, acuminate to broadly rounded, margin minutely ciliolate. Inflorescences open, terminal, many-Hlowered cymes, subtended by small sessile or subsessile orbicular or ovate-lanceolate, often puberulent leaves. Involucre broadly campanulate, 6-11 mm long, densely viscid-villous, lobes broadly triangular, obtuse or occasionally acute, shorter than tube, margin ciliolate; ultimate peduncles 1-5 mm long, densely viscid-villous. Perianth 4-5 cm long, sparsely glan- dular-villous, white tinged with pink to pale lavender, narrowly funnel- form, limb 1.5—2.5 cm broad. Stamens ca. twice as long as perianth; style exceeding stamens in length. Anthocarp dark brown, broadly obovoid or oval, 6-8 mm long, obscurely angled, smooth, tapered at both ends. Phenology.—Flowering from late September to December. Le Duc, Revision of Mirabilis section Mirabilis 625 if if I T T T | wee loe° lore ra Pa 30° . n ae \ = a3 E : ) : 2 i Sy c (a - wi x 2 (e) Y \2 b \ ' i \. e) : a 3 f \ V4 C - a Y — 10° @ M. donahooiana é as a | | ~ 3 mon \ #18 uA 0 100 200 300 ; L ag oe" : Fic. 1. Distribution of Mirabilis exserta, M. donahooiana and M. pringlet. Distribution (Fig. 1).—Upper elevations of the mountains of the Cape Region, Baja California. Altitude 1700-2000 m. Reon specimens: MEXICO. Baja California Sur: La. Chuparosa, Brandegee . (F, GH); Valley (La cla : of Pico . Aguja on the Sierra La Laguna, Breedlove 3339 (MEXU); Laguna, Lagun , Jones 27304 (NY, RSA, UC); Los Limpios, Sierra la oe E of Todos Santos, aati et 10586 (RSA). This species closely resembles M. pringle: in vegetative characters. The flowers resemble those of M. gracilis but its stamens are more exserted. The anthocarp is unique within the section Mirabilis; it resembles those of sec- tion Quamoclidion in shape (elliptic) and having a smooth, glabrous surface with only slight indication of furrows. 1992. Tyee: MEXICO 3. Mirabilis hintoniorum Le Duc, Sida 15:53. 11 Jul 1939, Hinton MicHoacaANn: District Coalcoman, Villa Victoria, dense woods, 13909 (HoLotyPE: TEX!; tsorypes: G!, MO!, UC!). Herbaceous or suffruticose perennials(?), erect, 7-8 dm high. Stems slen- der, striate, nodes puberulent, otherwise glabrous. Mid-stem leaves: peti- ole slender 1-2 cm long; blade glabrous, 5—9 cm long, 2.0-6.5 cm wide, base broadly to narrowly ovate, asymmetrically cordate or slightly trun- cate, apex acuminate, margin sparsely ciliolate. Inflorescences terminal, arranged in few-flowered cymes, subtended by sessile or subsessile, ovate to ovate-lanceolate, pubescent leaves. Involucre narrowly campanulate, 2—3 Sipa 16(4) ; ! T T T T l T T T we joe ‘nes si Me 30° = = \y + 2 a ; " f & . | s os No OE a ws } NY a ; ie ie te = 24° 3 ‘ MY , on 5 . op, : Fr as \ é e Q M. hintoniorum am SAN, ae ; m™ M. gracilis o — 10° 4 M. polonii Cl ei rx oe @ M. urbanii ; RL ae Oo 100 200 300 . i‘ fe 100" lors . , Fic. 2. Distribution of Mirabilis hintoniorum, M. gracilis, M. polonit and M. urbanit. mm long, glabrous or slightly puberulent, lobes narrowly triangular, less than 1/2 the length of tube, margin ciliolate; ultimate peduncles 3-5 mm long, pubescent. Perianth 2.0-2.5 cm long, glabrous or nearly so, orange, tube dilated upwards, limb 5-8 mm broad, lobes obscure, ciliolate. Sta- mens ca. twice as long as the perianth. Anthocarp dark brown, oblong- ellipsoid, 6-7 mm long, 3 mm wide, 5-angled, the ridges tuberculate, moderately puberulent, constricted at both base and apex. Phenology.—¥ lowering in July. — Distribution (Fig. 2).—Known from only type locality, in tropical de- ciduous forest. Altitude 700 m. The open terminal inflorescences and ovate leaves with cordate bases of M. hintoniorum are most similar to those of M. exserta and M. pringle:. How- ever, it differs in the extreme reduction of the perianth lobes, the lack of viscid-villous pubescence, and its few-flowered inflorescences. The anthocarp of M. hintoniorum most closely resemble those of M. donahooiana and M. polonii but also shows a resemblance to many anthocarps of Mirabilis sec- tion Oxybaphus differing from the latter in being nonmucilaginous. The flower color of M. Aintoniorum, as noted by label data, is unusual for the genus, and might be questioned. However, I have collected M. jalapa with orange flowers, from a small population in the state of Mexico (Le Duc 94 TEX), thus giving credibility to Hinton’s notation. Le Duc, Revision of Mirabilis section Mirabilis 627 Fic. 3. Mirabilis polonit habit and details from holotype showing: upper leaves and flowers with an enlarged detail of the anthocarp. 4, Mirabilis polonii Le Duc, sp. nov. (Pl. 3-B; Fig. 3) Mirabilis gracilis (Standl.) Le Duc foltis similis, M. Aintontorum Le Duc morphologia anthocarpiorum similis; M. /ongiflorae L. similis perianthiis in longitudine similibus sed differt perianthiis arcuatis (deorsum curvatis), follorum basibus truncatis, et anthocarpiorum apicibus truncatis. Herbaceous or suffruticose perennials, erect, 7.5—10 dm high, much- branched, roots tuberous. Stems slender or stout, glabrous or with pubes- cence restricted to 2 longitudinal lines, internodes 7-9 cm long, nodes puberulence. Mid-stem leaves: petiole slender, 0.5—3.0 cm long; blade 7— 11 cm long, 4—5 cm wide, thin, green (often reddish beneath), puberulent, broadly ovate, base asymmetrical, subtruncate or occasionally subcordate, apex attenuate, margin ciliolate. Inflorescences terminal, somewhat open, many-flowered cymes, subtended by small ovate-lanceolate leaves. Involu- cre slightly pinkish in age, narrowly campanulate, 8-10 mm long, slightly puberulent, lobes 4-5 mm long, lanceolate-oblong, acute, margin ciliolate; 628 Sipa 16(A) ultimate peduncles 2-5 mm long, puberulent. Perianth 9-11 cm long, glabrate, white, fragrant, tube very elongate, curved downward, ca. 2 mm wide, limb 2.0—2.5 cm broad. Stamens exserted, slightly less than half length of tube, 11-13 cm long, lavender. Style ca. | mm longer than sta- mens. Anthocarp dark brown to brownish black, obovoid, 6—7 mm long, 4-5 mm wide, 5 angled, ridges tuberculate with warty areas between, pu- bescent, constricted and truncate at both base and apex. Tyree: MEXICO. Sinatoa: along Hwy 40, 3 mi SW of La Guayanera & 2 mi N of the Copolita spur, between Matzatlin and Durango, N 23° 23', W 105° 55', altitude ca. 700 m, 29 Jul 1991, Le Duc 178 (HOLOTYPE: TEX!; tsoryprs: MEXU!, others to be distributed). Representative specimens: MEXICO. Nayarit: Mpio Nayar Cerro Cangrejo, Cafiada al NE poblado Villa de Sadie Tenorio & Flores 16206 (RSA). ao Leon: Cafién 3.6 mi SW of Los Ajuntas & 7.4 mi of La Trinadad, Le Dac 259 (TEX); Cafién del Pasaje de . Osos, al Pte. de Ybanis, Santiago, Maman 1330 (TEX); Trail i. Potrero Redondo Las Ajuntas, Mueller 2990 (GH, UC). Oaxaca: Rio Coyula a 7 km al SE de le Limon, ae Cedillo 1688 (LL, MEXU). Gan Luis Borel ean Mpio San Antonio, Al/corn 1838 (TEX). Sin ores sr San Ignacio. La Cebolla + 40 km N de San Ignacio, Vega, R. y S. Palazuelos 781 (MEXU); 33 mi SW of Rivalcaderos, Wé ne 12733 (TEX). Tamaulipas: 10 km NW . El ie which is ca. 18 km NW of Ocampo, Standford et al. 1040 (GH, MO, NY, UC). — an —Flowers from late July to September. Flowers open in the evening. Distribution (Fig. 2).—Moist Pacific slopes of the Sierra Madre Occiden- tal, Sierra Madre del Sur, and the Eastern protected canyons of the Sierra Madre Oriental; in tropical deciduous forest, semi to full shade, soil crum- bly, igneous or limestone. Altitude 900-1500 m. This species resembles M. gracilis in foliage, having thin dark green with truncate bases. The anthocarp, however, is like that of M. Aintoniornm and M. donahooiana. The perianth is at least 6 cm long, resembling that of M. longiflora, but the latter is erect while that of M. poloni is arching (curved downward). The species is named in honor of David Polon, an anthropologist who worked in Mexico. Without his encouragement I would not have focused my studies on a genus from Mexico. He was very positive in his beliefs that —_ Caves ‘on more research was needed on the plants of Mexico. D: a longiflora L., Kongl. Svenska Vetensk. Acad. Handl. 176. t 9. 1755. Jalapa lnsifina (L.) Moench, Metl oe 50 : 1794. Nyctago longiflora (L.) ree Prodr. Stripium Chap. Allerton 57. 1796. Type: MEXICO. without spe- cific locality or date. Hlustration t. 6 ac ee typi this name. There is one specimen 240.3 LINN in the Linnean collection. No ae ition is indicated on the ner ele- ae sheet as to the origin of the material or if Linnaeus studied it. The only ot ment cited by Linnaeus is the illustration in Rerum Medicarum Novae Histpaniae Thesaurus f. 2, p. 170 by Francisco Hernandez, 1651. Herbaceous or suffruticose perennials, erect, 0.5—1.5 m high, much Le Duc, Revision of Mirabilis section Mirabilis 629 branched, roots tuberous. Stems slender or stout, densely viscid-puberu- lent or short villous, lower internodes usually longer than leaves. Mid-stem leaves: petiole slender, 2-6 cm long; blade 6-12 cm long, 3—7 cm wide, thin, bright green, densely viscid-puberulent to sparsely so, cordate-ovate to narrowly deltoid-ovate or lance-ovate, base cordate, apex acute to at- tenuate, margin ciliolate. Inflorescences dense, axillary or terminal, many- flowered cymes, subtended by sessile or subsessile reduced leaves. Involu- cre campanulate, 1.0—1.5 cm long, densely glandular-pubescent, lobes about as long as tube, triangular to narrowly triangular-lanceolate, very acute to long-attenuate, usually exceeding anthocarp in fruit, margins ciliolate; ul- timate peduncles ca. 3 mm long or less, densely glandular-pubescent. Peri- anth 7-17 cm long, densely viscid-villous outside, white, throat tinged with pink or purplish-red, fragrant, salverform, tube very slender, ca. 2 mm in diameter, limb 2—3 cm broad, lobes broad shallow-rounded. Sta- mens exserted, ca. 2.5 cm beyond throat, purplish-lavender. Anthocarp dark brown, puberulent, constricted and truncate at base. Chromosome number n = 29 (Showalter 1935; Kruszewska 1961) Inflorescence congested, glomerate, subtending leaves sessile; niece ob- hee 5-angled, slightly puberulent, prominent white-warty areas, the apex truncate. Sa. M. iif var. longiflora ees open, subtending leaves short-petiolate to subsessile; anchocarp 5-angled, puberulent, warty, the apex acute. ............. Sb. M. ne var. wrightiana Sa. Mirabilis longiflora L. var. longiflora Phenology.—Flowering from July to September. Flowers open in the evening. Distribution (Fig. 4)—Mostly the Trans-Mexican Volcanic Belt and Cen- tral Valleys of Mexico, growing under Juniperus, Magey or other similar plants, at the margins of cultivated fields. Altitude 1800-2800 m. Representative specimens: MEXICO. agra ae: Xichu road, ee 2298 (G HIDALGO: 6 km N of Pachuca, Herndndez 4346 (GH); Cerro Jazmin, 2 km NE of Apan, ee D-10 (UC). Nuevo Leon: eee Orcutt 1228 (US). Oaxaca: Escuela rie Snes Conzatti 973 (GH). Puebla: Santa Ana, Nicolas 5317 (NY); ca. 2 km N of Saltillo ee Hwy 140 from Jalapa to pace Poole 1555 (TEX), San ae pak Pew 374 (F, GH, MO, NY, UC, US); Mc. eae Esperanza, Seaton 493 ae WSW Tlaxco on road to Apan, Hwy 119, Le Due 170 (TEX); E of ae on Hwy 136, Le Duc et al. 224 (TEX). Verses “4 srote, Balls B5518 (UC, US); Tenextepec, Mpio Perote, Chazaro & Acosta 3739 (MIC ; Cerros arriba de Santiago, Nevling & Gomez-Pompa ee (CAS, GH, MEXU, RSA); near town of Alchichica, Ramos 284 (GH, MEXU), near ncho El Camino 7 & La Gloria, Ramos 226 (GH); ca. 20 air km a of Perote, ner 15209 (TEX); 3 km S of Totalco, Vazquez 4843 (MO). = 5b. Mirabilis longiflora var. wrightiana ( A. Gray) Kearney & Peebles, J. Wash. Acad. Sci. 29:475. 1939. Mirabilis wrightiana A. Gray ex. Britton & Sipa 16(4) ! _ ~ Te ! | | 1 1 y | : lors 96° 90° ‘.3 # ° e r— 30° @e fay 30° } e % nd > a 43QN ; \ \ rs aa ™~ a of aaa : !“ “a 24° e / \ AS yO ; : tie i v5 f(% Cc eww ‘, n a ae longiflora. var. longiflora Pane EN 1/ pie @M. longiflora var. wrightiana ‘ y. Re ‘< 7 Oo 100 200 300 at es Km 10e° loz* * a ! | i t i Fic. mee of Mirabilis longiflora var. longiflora and M. longiflova var. wrightiana in eae and U.S.A. Kearney, in Trans. New York Acad. Sci. 14:28. 1895. TYPE: es acs New oer Grant Che Santa Rita de Cobre, ak of eh Coppermines creek, 4 Aug 1851, Wright 150 (HoLoTyPE: GH!; isorype: NY!). A uted by A. Gray under the exsiccata ae l these are all isotypes. umber of specimens were ete on It has not been ascertained if Mirabilis suaveolens H.B.K., Nov. Gen. Sp. 2:213.1817. Type: MEXICO. Guanajuato: nity of Guanajuato, Auigese 803, Hanblr G Bonpland SM. Fi eeakes P). Mirabili tubiflora Fries ex. Heimerl, Beitr. Syst. Nyc 0. var. tubiflora (Heimerl) Heimerl, Notizbl. Bot. a. Berlin- ai lls “ype: Cultivated plane, Leipzig Botanical Garden grown from seed provided Sy Th. M. Fries. (HOLOTYPE: W, destroyed; isorypr: F!). Material was destroyed in 1945, dur- ing the war, pers. comm., Harald Riedl, Director W. oS 1897. Mirabilis ah eee 450. 1932. Differs from M. /ongiflora var. longiflora in stems more slender, upper most leaves short-petiolate. Inflorescences open, involucre slightly viscid- puberulent. Stamens slightly more exserted, ca. 3 cm. Anthocarp dark brown, oblong to ellipsoid, ca. 8 mm long, ca. 5 mm wide, S-angled, pu- berulent, except on swollen areas, warty, base constricted and truncate apex acute. Flowers from July to September. Flowers opening in the evening. — Distribution (Fig. 4).—Central Mexico northwards to the mountain ranges Le Duc, Revision of Mirabilis section Mirabilis 631 of SE Arizona, SW New Mexico, and Big Bend area of Texas, U.S.A.; usu- ally in the shade of trees and shrubs, occasionally in open grasslands, in rocky soil. Altitude 1500-3700 m. Representative specimens: MEXICO. Aquascalientes: 6 km E of Tepezala, — ski & McVaugh 1208 (MICH). Chihuahua: Soldier Canyon, Sierra Madre Mts., Jones 5.n.(RSA), NW end of Sierra del Diablo, Stewart 960 ( GH); 24 km NW of Balleza, ee a 9917 (RSA, TEX); Santa Eulalia, Rosalia, Wi/kerson s.n. (UC); Gallejo Springs, Wislizenus 122 (MO). Coahuila: Cafidn above Palomas, Saltillo, Gregg 331 (GH); Mpio Villa Acuna, Ha- cienda Piedra, Canyon of Sentenela, Wynd & Mueller 585 (MO, NY). Distrito Federal: Sierra de Guadelupe, cerro Grande 5 km al NNW of Cuautepec, Moreno 275 (MICH); Teutli, Ventura 1876 (NY, MEXU, RSA). Durango: Hwy 30 between La Zarca & La Cadera, 22 km E of Hwy 45, Cruden 2035A (UC), Hwy 40 at the crossing of arroyo de Los Mimbres, 5 mi W of Guadalupe Victoria, Le Duc 180 (TEX). Guanajuato: Guanajuato, Dugis s.n (GH); 30 mi E of San Luis de la Paz toward Xichi, Straw & Forman 1466 (MICH). Hidalgo: Puerto de la Zorra NE of Jacala, Moore 3524 (GH, UC); upper slopes of El Monte on trail from Zimapan to mines of El Monte, Moore, Jr. 4474 (MEXU, MICH). Jalisco: Mpio de Z an, Rio Caliente La Primavera, Diaz-Luna 333 (MICH), Mpic Tlajomulco, San Lucas Evangelista, eae sere 2736 (MICH). Mexico: Dist. Temascaltepec, Panon, Hinton sae (GH, MICH, MO, US); Dist. Temascaltepec, Salitree, Habe 4313 (GH, MICH, RSA, US); a ee 1560 (C AS, MEXU, UC); E of Tenango del Aire, Rio es Pineda 519 (CAS, MICH, UC); M pe eadeibe tlan, Presa de la Concepcion, Rzedowski 22891 (CAS, MICH, TEX). pee ate Egeler 124 (MO). MORELOS: Huajojuctla, Alexander & Herndndez 2017 (CAS, GH, MICH, NY, UC, US), Tepoztlan near ee eee Le Duc, et al. 234 X); Barranca near Cuernavaca, Pringle 6377 (CAS, GH, MO, NY, UC). Nuevo Leon: entrance de ie Cave, Grutas, 32 mi NW of Monterrey, Ward 5666 (MICH). San Luis Potosi: 7 km SW of Pozuelos and 22 km SW of San Luis Potosi on the hwy to Guadalajara, Johnston et a 12267 (CAS). Sonora: Imuris, Abrams 12771 (RSA); Fronteras, Hartman 976 (MO). Zacatecas: 95 mi W of Sombrerete, Taylor 6247 (NY); 3 mi W of Villanueva, Wa/ker 76H48 (MO, NY) U.S.A. ARIZONA: Cochise Co.: Dragoon Mts., Sorin ee Daniel 3079 (MICH, NY); Huachuca Mts. Carr Canyon, Guid et al, 2428 (UC); 1.5 mi W of Turkey Creek atu Station, Holler et al. 1024 (N Portal, aie et al. 2671 (NY). Gila Co.: Workman Creek, at ce Wagener 327 (UC). Pima Co.: Fresnel Canyon, Baboquivari Mts., Gilman 49 (G O, NY). Santa Cruz Co.: Atascosa Mts. near Yanks Canyon, Franklin 5390 (N ee MEXICO: Grant Co. Forest Nursery, Fort Bayard water shed, Blumer 231 (GH, NY); Santa Rita, on dirt road called Miner's Legend, Le Duc 185 (TEX): naw eas of Fierro, Le Due 190 (TEX). Socorro Co.: San Mateo Springs, 10 mi W of Hwy 85, Socorro/ Sierra Co. ee Baad 1349 (MICH); Mogollon Mts. mid fork of the Gila river, ” Metcalf 432 (GH, UC). TEXAS: Brewster Co: Alpine, ae 240 (NY). Jeff Davis Co.: Limpia Creek W of Fort aoe Correll 33672 (GH, UC); summit of cles in Davis Mts., Correll 34971 (NY); Limpia Canyon, Hwy 118 near Ft. Davis, 195 (TEX); Davis Mik: Sawtooth, Palmer 31895 (MO); old Kent road W of Mc. soe ‘Sei 1123 (NY); Fern Canyon, Steiger 1256 (NY); Fern Canyon, W. arnock TO34 (NY); 2 m Fort Davis, Warnock 8034 (MICH). Presidio Co.: Cibolo Creek, 5 mi E of Russ. Nae ranch headquarters, Warnock 3671 (NY); Cibolo Creek, Cieniguita, 10 mi N of Shafter, Warnock & Hinckley 4500 (UC). Mirabilis longiflora was cultivated by the Aztecs as an ornamental plant and, perhaps, as a medicinal herb much as it is today in parts of Mexico 632 Stipa 16(A4) (Alcorn 1984). Many of the populations in the Central Plateau valleys of Mexico show evidence of hybridization between M. /ongiflora and M. jalapa. Putative hybrids have also been recorded in the literature of Europe (Lepeletier 1806). The plant described by Linnaeus was probably from a European garden and its description is consistent with the numerous M. longiflora populations I have observed cultivated or commensal from the ntral Mexican Plateau valleys to Oaxaca. Several characters seem to be shared by these various populations. The terminal glomerate many-flowered cymes characteristic of M. jalapa are also characteristic of M. ongiflora var. longiflora. In many populations, plants with flowers resembling those of M. jalapa have vegetative characters resembling M. /ongiflora. Mirabilis longiflora var. wrightiana of the mountain areas of southern Arizona, New Mexico and Texas, and the northern desert regions of Mexico, differs from M. longiflora var. longiflora in inflorescence and anthocarp characters, appears to be the wild progenitor. It is possible that M. /ongiflora was not originally native to the Central plateau valleys of Mexico but was introduced from the more northern mountains in pre-Columbian times. 6. Mirabilis meee Hei merl, Oesterr. Bot. Z. 56:250. 1906. Tyee: heey O. MIcHOACAN: § of San Saleen 11-12 Jul 1898, Langlassé 240 (HOLOTYPE: W, de- stroyed; ISOTYPES: - K!, P!). The holotype was destroyed during the war in 1945, pers. comm., Harald Riedl, Director W. Herbaceous perennials, ascending or semidecumbent, 10-30 cm high, much branched, roots swollen or tuberous. Stems slender, young puberu- lent, mature glabrous or with pubescence restricted to 2 longitudinal lines, internodes 5—7 cm long. Mid-stem leaves: petiole slender, 1.0—1.5 cm long; blade 3.54.5 cm long, 2.5-3.0 cm wide, thin, bright green, puberulent, ovate-deltoid, base asymmetrical, grading into the petiole, apex attenuate or acute, veins few, weak or little branched. Inflorescences solitary in the leaf axils, or terminal and aggregate in small 2—3 (—4) flowered cymes, subtended by few small subsessile ovate-lanceolate leaves. Involucre nar- rowly campanulate, ca. 7 mm long, glabrate, the lobes ca. the same length as tube, lanceolate-oblong, subobtuse, the margin ciliolate; the ultimate peduncles 3-4 mm long, short-villous. Perianth 1.5—2.5 cm long, short- villous below, purplish-red to lavender-pink, cube funnelform, the limb 12-14 mm broad. Stamens slightly exserted, pink. Flowers may be cleisto- gamous late in season. Anthocarp brown to dark brown, oblong-ellipsoid, 6-7 mm long, 5-angled, ridges slightly tuberculate; distinct pubescence of orange-brown scalarified trichomes containing cystoliths; constricted at both base and apex. Phenology. —Flowers from late July to September or October. Flowers open in the morning. Le Duc, Revision of Mirabilis section Mirabilis 633 Distribution (Fig. 2).—Pacific slopes of the Sierra Madre del Sur in Michoacaén; crumbly igneous soil. Altitude 600-900 m oe specimens: MEXICO. Michoacan: 4.9 mi S$ of Puerte San Salvador, along , Le Duc et al. 245 (TEX); 20 km N of Infiernillo, Nw#ez 1687 (CAS). ea urbaniit possesses several distinctive characters. Its leaves which are truncate with the blade base grading into the petiole, and the anthocarp with its distinctive bristle-like, scalarified trichomes. Because of this, the position of M. wrbanii in section Mirabilis is somewhat questionable. It is similar to M. sanguinea, differing from the latter in the above mentioned leaf and anthocarp characters. 7. Mirabilis sea ten aes Notizbl. Bot. Gart. Berlin-Dahlem 11:451 1932. Type: MEXICO. Guerrero: Campo Morado, 14 Jun 1899, Langlassé 1058 (HOLOTYPE: W, pena isotype: F!, GH!). The holotype was destroyed during the war in 1945, pers. comm., Harald Riedl, Director W. Isotype at GH has the collec- tion no. 1058 written in, I could not discern if this was actually Langasse’s number but make the assumption that it is. Herbaceous perennials, erect or semidecumbent, 30-40 cm high, multi- stemmed, roots tuberous. Stems slender, glabrous to very puberulent, in- ternodes strongly elongate, 13-23 cm long, the nodes with lateral puberulence. Mid-stem leaves: petiole slender, usually half as long as or longer than blade; blade 60-70 mm long, 35-55 mm wide, thin, green, glabrous, rhombic-orbicular, cordate-deltoid, base cordate to subequal trun- cate, apex acute, margin minutely ciliolate. Inflorescences dense, terminal many-flowered cymes, subtended by small ovate-lanceolate to lanceolate leaves. Involucre narrowly campanulate, ca. 4 mm long, 2-5 mm wide, lobes slightly acute, margin ciliolate, exceeding anthocarp in fruit; ulti- mate peduncles 1.5—3.0 mm long, somewhat puberulent to densely pu- berulent. Perianth 15-35 mm long, glabrous or upper half long villous, blood red to lavender-pink, funnelform to salverform, tube narrow, ca. 0.5 mm, limb gently expanding to 13 mm broad, lobe apices obtuse. Stamens exserted, 17-37 mm long, lavender. Anthocarp brown to brownish-black, obovate-elliptic, 3.5-4.5 mm long, 2.0—2.5 mm wide, 5-angled, ridges tuberculate, hirsute, constricted near base, with nipple-shaped apex. Perianth 20-35 mm long; flowers red Ja. M. sanguinea var. sanguinea Perianth 15—20 mm long; flowers cerise to lavender-red ... 7b. M. sanguinea var. breviflora 7a. Mirabilis sanguinea Heimerl var. sanguinea Phenology.—Flowers from July to September. Distribution (Fig. 5).—Western slopes of the Sierra Madre del Sur, in- land to mountains along the Guerrero-Mexico and Morelos state lines and south to Chiapas. Altitude 800—2500 m. 634 Sipa 16(4) + 24° L_ige “4M. sanguinea var.sanguinea OM. sanguinea var.breviflora 0 100 200 300 Leche i on loe" l l l 1 Fic. 5. Distribution of Mirabilis sanguinea var. sanguinea and M. sanguinea var. breviflora. Representative specimens: MEXICO. Chiapas: Suchiapa, road to Villa Flores, Breedlove 28076 (NY); El ee . 6 mi E of Chiapa de Corozo, Breedlove 10679 (F, LL, MICH); Rio Grijalva Canyon, Hwy 211 N of eee de la Frontera, pee 359 (TEX). Guerrero: Aquazarca, Mina, Hinton Pe (F, GH, MICH, MO, NY, RSA, TEX, UC); Campo Marado, Mina, Hinton 14320 (F, GH, MICH, NY. RSA, ie US). anions La Florida, Zitacuaro, Hinton 11976 (F, GH, MICH, MO, NY, RSA, TEX, UC, US). Mexico: Nanchititla, Temascaltepec, Hinton 4521 ce UC, — 7b. Mirabilis sanguinea var. breviflora Le Duc, var. nov. (Fig. 6) Varietati typicae similis sed floribus perianthio breviore et limbo latiore differt. Distribution (Fig. 5)—Western slopes of the Sierra Madre del Sur, in- land to mountains along the Guerrero-Mexico and Morelos state lines north through the Sierra Madre Oriental and Sierra Madre Occidental. Altitude 800-2500 m Type: MEXICO. Jatisco: Ejido Santa Catarina Balneario, 0.9 mi N of Hwy 104 & 401 jet, 26 Jul 1991, Le Due et al. 251 (HOLOTYPE TEX!; 1sorypes: MEXU!, others to be distrib- uted). Representative specimens: MEXICO. Jalisco: Road to Tapalpa, 10.6 mi from jct. of ‘ oy 54, 0.8 mi from before microwave tower, Le Duc et al. 254 (TEX). Nuevo Leon: 13 m al e de San Antonio Pena Nevada, Mpio Zaragoza, Herndndez S. et al. 2716 (TEX). The entities from Jalisco and Nuevo Leon differ from M. sanguinea var. Le Duc, Revision of Mirabilis section Mirabilis 635 Fic. 6. Mirabilis sanguinea var. breviflora habit and details from holotype showing: upper leaves and flowers with an enlarged detail of the anthocarp. 636 Stipa 16(4) sanguinea. They have a shorter perianth tube,15—17 mm long, which flairs more and is lavender-purple in color. The Jalisco plants, which I have ob- served, all bloom in the morning. Mirabilis sanguinea is similar to speices of the section Oxybaphus in peri- anth, and in the shape and pubescence character of the anthocarp. This species differs from section Oxybaphus in having 1-flowered involucres that are not enlarged in fruit, and nonmucilaginous anthocarps. Within the section Mirabilis, M. sanguinea appears to be most similar to M. wrbanii. 8. Mirabilis donahooiana Le Duc, sp. nov. (Pl. 3-C; Fig. 7) Mirabili hintoniorum Le Duc ac M. polonit Le Duc morphologia anthocarpiorum similis; M. jalapae L. proprietatibus foliorum ac florum similis. Differt a M. salapa morphologia anthocarpiorum, floribus matutino aperientibus, et corollarum limbo magis profunde lobo. Herbaceous or suffruticose perennials, erect, 75-120 dm high, much branched, roots tuberous. Stems slender or stout, younger with pubescence restricted to 2 longitudinal lines, mature glabrous, internodes 7-11 cm long, nodes with lateral puberulence. Mid-stem leaves: petiole slender, 0.5— 2.5 cm long; blade 2-8 cm long, 1.0-4.5 cm wide, thin, green, slightly puberulent, broadly ovate, base asymmetrical, subtruncate, apex attenu- ate, margin ciliolate. Inflorescences open, terminal, many-flowered, cymes, subtended by small ovate-lanceolate leaves. Involucre narrowly campanu- late, LO-12 mm long, lobes 5—6 mm long, puberulent. Perianth 3.0—3.5 cm long, ca. 2 mm wide, limb 1.5—2.5 cm broad, pubescent, lavender- pink. Stamens only slightly longer than tube, lavender-pink. Style slightly longer than stamens. Anthocarp brown, oblong-obovate, 5-6 mm long, ca. 3mm wide, puberulent, ridges tuberculate, slightly warty between ridges; constricted and truncate at both base and apex. Phenology.—Flowering from July to September. Flowers open in the morning. Distribution (Fig. 1)—Area around Aquila, Michoacén, Mexico, in tropi- cal deciduous forest, in partial shade at the foot of rocky cliffs or road cuts, in crumbly igneous soil. Altitude ca. 25 m. Type: MEXICO. MicHoacan: road to Aquila, 4.8 mi NW from jet of Hwy 200, 3.5 mi before La Joya ac N 18° 37' 30" W 103° 30', 8 Aug 1992, Le Duc et al, 248 (HoLo- "EX!; tsorypes: MEXU!, others to be distributed). Seg pose MEXICO. Michoacan: Aquila, Dist. of Coalcoman, G. Hinton 16017 (LL, MICH, NY, RSA, UC); side road to Aquila, 8.3 mi NW from jct of Hwy 200, just before La Joya ek : Duc et al. 247 (TEX). — Anthocarps of this species are similar to those of M. Aintoniorum and M. poloni7, in other respects the species resembles M. jalapa. Mirabilis donahooiana differs from the latter in having a narrower perianth with deeply Le Duc, Revision of Mirabilis section Mirabilis 637 Fic. 7. Mirabilis donahootana habit and details from holotype showing: upper leaves and flowers with an enlarged detail of the anthocarp. lobed limb. It blooms in the morning, and is closed before four o'clock in the afternoon, the time when M. ja/apa begins to bloom. This species is named in memory of Absalom Donahoo, the author's great grandfather. Mr. Donahoo was a pioneer plantsman in Nebraska in the 1860s and 70s. It was his legacy of plant knowledge that led the author to her career in plant systematics. 638 Sipa 16(4) 9. Mirabilis gracilis (Standl.) Le Duc, comb. & stat. nov. (Pl. 3-D; Fig. 8). Mirabilis jalapa var. gracilis Standl., Contr. U.S. Natl. Herb. 12:367. 1909 Type: MEXICO. Srnatoa: vicinity of Culiacan, 17 Sep 1904, T’S. Brandegee s.n. (HOLOTYPE: UC!) Herbaceous or suffruticose perennials, erect, 75-100 dm high, much branched, roots tuberous. Stems slender or stout, glabrous or with pubes- cence restricted to 2 longitudinal lines, lower internodes 10—12 cm long, (occasionally 28-30 cm), the nodes + villous. Mid-stem leaves: petiole slen- der, 1.5—3.0 cm long; blade 6-9 cm long, 3.0-4.5 cm wide, thin, dark green, glabrous, deltoid-ovate, ovate, base asymmetrical, truncate, apex acuminate, margin minutely ciliolate. Inflorescences terminal few-flowered cymes, subtended by short- perce’: small leaves. Involucre campanulate, 8-10 mm long, glabrous, lobes 4-5 mm long, triangular, apices acute, margin ciliolate; ultimate peduncles ca. 3 mm long, villous. Perianth 6.0— 6.5 cm long, glabrous, fragrant, white, salverform, tube ca. 3 mm in diam- eter, limb ca. 2.5 cm broad. Stamens exserted, 8—9 cm long, lavender. Anthocarp dark brown, broadly obovoid to oval, 9-10 mm long, 4-5 mm wide, 5 angled, warty, pubescent, base constricted and truncate, apex acute. Phenology.—Flowering from late July to October. Flowers open in the evening. Distribution (Fig. 2).—Mostly Pacific slopes of the Sierra Madre Occi- dental and Sierra del Sur; Chihuahua to Michoacan, Mexico, occasionally in mts. of the Central Plateau; in tropical short tree or deciduous forest, partial to full shade, most often at the foot of a cliff. Altitude LOO—2000 m. on Representative slog ns: MEXICO. Chiapas: along Hwy 190 in Zinacantdn paaje of Multajoc, Mpio Ixtapa, Breedlove 13995 (F). Chihuahua: Guasaremos, Rio Mayo, Gentry 1549 (CAS, F, GH, MO, UC). Guerrero: Campo M nies oo Hinton 14489 (LL, MICH, NY, UC). Jalisco: 4.5 mi N of El] Rincon, Hwy 80, Le Duc & Sydor 71 (TEX); Chamela, Seen de Biologia Chamela, UNAM. Magan Es 22 (F, MEXU). Miesica: 5 mi SW of Santo Tomas de los Platanos (19.09N, 2W), G. Webster 21189 (MEXU),. Michoacan: Huizontla, Dist. Coalcoman, oe ere (LL, MICH, NY, RSA, UC). axaca: Presa Temazcal, Vertederos de le presa, Distr. Tuxtepec, Cortés, Ly R. Torres 49 (MEXU). Queretaro: Mpio of Landa de Matamoros, Tangajo, ca. 15 km W of Santa Ines, Ferndndez 3153 (NY This species closely resembles M. ja/apa, but differs in the few-flowered inflorescences, the perianth slightly longer, the stamens well exserted, and lavender rather than the same color as the perianth. The large anthocarp with its unique sulptured topography is unlike any other within the sec- tion Mirabilis. 10. pce jalapa L., Sp. Pl. 177. 1753. a icine Moench, Methodus 508. 1794. Nyctago ee Salisbury, Prodr. Stirp. C n 57. 1796. Nyctago jalapae De Candolle, Fl. Frang. 426. 1805. Type: in eur collector unknown, probably from cultivated material obtained originally in Mexico or the West Indies. Le Duc, Revision of Mirabilis section Mirabilis 639 Fic. 8. Mirabilis gracilis habit and details from holotype showing: upper leaves and flowers with an enlarged detail of the anthocarp. 640 Sioa 16(A) Herb. Clifford 53, Mirabilis 1[eta} (Lecrorype: BM! designated by Le Duc, in Regnum 1276) Mirabilis odorata L., Cent. a 1755. Mirabilis dichotoma L., Sp. Pl. 252. 1762. (The reference Arsen, Acad. : oo in the original description, is a citation of the reprint of the original description of M. odorata). Jalapa dichotoma Crantz, Inst. Rei Herb. lus 2 dae ye Crantz references the illustration on p. 90 of Clusius, Rar PL Hist., pe 160 e makes no refer pecimens examined. mee is possibly meant to : a NOV. pete of ML dione. ote undulata Moench, Methodus ce 196. LW a =} om ani ps. 1802. (Moench eee to M. dichotoma L.). Mirabilis see odorata (L. Bot. Jahrb. Syst. 21:616. 1896. Type: MEXICO. 240.1. (Lectotype here ie LINN). Gain material includes the specimen 240.1 which was annotated M. odorata 1 this annotation and added dichotoma, the su- a oy Linnaeus. He later crossed throug perfluous 1762 name Nyctago mirabilis J.St.-Hil., Expo. Fam. Nat. 1:212, t. 37. 1805. Type: not located, if any. Illustration t. 37 published with the original description here designated as lectotype. Mirabilis pedunculata Stokes, Bot. Mat. Med. 1:311. 1812. In his protologue Stokes mentions a garden grown plant and makes reference to M. jalapa L., Sp. Pl. 252. 762. Type: cultivated garden plant, collector and conc Ca Mirabilis divaricata Lowe, Trans. Cambridge Phil. Soc. 17. . TYPE: eleva and escaped garden plant on the islands of Madeira and ce cen 183 o collec- tions were nee If Lowe actually collected a voucher, then the specimen is probably at BM ¢ Mirabilis ne Bertol., Novi Comment. Acad. Sci. Inst. Bononiensis 3:15. t. 1. 1839, Mirabilis jalapa var. bea (Bertol.) Choisy in DC., Prodr. 13:428.1849. Type: culti- vated plant (HOLOTYPE: BOLO?; Lecrorype here desiginated: illustration t. 1 in lieu of a specimen) Mirabilis planiflora Trauty., Bull. Acad. Imp. Sci. Saint Pecersbourg 6:216. 1840. Mirabilis iene var, oe we nee Choisy in DC., Prodr. 13:428. 1849. Type: cultivated cal Garden. oe OTYPE: let “Raf, Autik. Bot. 1:12. 1840. Type: not located, if any. Rafinesque not a specific ealliceian: - did mention plants from Central America and ‘ aaa one plant with characteristics that blend in with a jalapa.” ee nally published as T. levigata, detencarars a acd Miah . Trauty., Linnaea 15: Lit. Ber. 97. 1841. Mirabilis jalapa var. ambigna v.) Choisy, in DC “Prodn, 13:428. 1849. Type: aes plant, Kiev Botanical ae 1840 ae Mirabilis jalapa subsp. ciliata Send. ok U.S. Natl. Herb. 12:368. 1909. Type: XICO. O ‘A: valley of Oaxaca, 1 Oct 1894, Smith 791 onde MO!). Mirabilis yalapa subsp. lindheimert Standl., Contr. U.S. Natl. Herb. 12:368. 1909. Mirabilis jalapa var. lindbeimeri (Standl.). Stand]., Rhodora - 405. 1936 Mirabilis | (Standl.) Shinners, Field & Lab. 19:17 3. Type: U.S.A. Texas. Comal C Nev unfels, Jun 1846, Lindhermer 158 (Lecrorypr here designated: M OF). $ avecal ie sheets exist at MO, some of these annotated by Standley; but these are eee dated and ey reflect a ont of collections. I have selected one of sheets annotated by Standley as a lectot Mab jalapa subsp. volcanica Standl., Cont atl. Herb. 12:367. 1909. Type: 2O. Disrriro FEDERAL: Pedre zal oe ba, aie of Mexico, 19 Aug 1896, ance 6433 (HOLOTYPE: MO!; Isorypes: GH!, Le Duc, Revision of Mirabilis section Mirabilis 641 Herbaceous or suffruticose perennials, erect, 0.5-1.0 m high, much branched, roots tuberous. Stems slender or stout, glabrous, puberulent or rarely short-villous, sometimes viscid. Mid-stem leaves: petiole slender, 0.3-5.0 cm long; blade 4-14 cm long, 2.0-8.5 cm wide, thin, bright- green, glabrous or rarely puberulent, ovate-deltoid, broadly ovate, ovate- oblong, or rarely lance-ovate, base subcordate to truncate and asymmetri- cal, apex acute to attenuate, margin usually ciliolate. Inflorescences terminal glomerate many-flowered cymes subtended by numerous reduced leaves. Involucre campanulate, 7-15 mm long, glabrous, puberulent, or short- villous, lobes longer than tube, linear-lanceolate to lance-ovate, acute to attenuate, margin usually ciliolate; ultimate peduncles 1-2 mm long or wanting. Perianth 3.0—5.5 cm long, purplish-red, white, yellow, orange, or variegated, glabrous or sparsely villous, tube 2-5 mm in diameter, gradu- ally dilated upward, the limb 2.0-3.5 cm broad, lobes shallow and broadly rounded. Stamens same length as perianth or occasionally exserted 1-2 cm beyond the perianth, white or the same color as perianth. Anthocarp dark- brown or black, elliptical, obovoid, oval to round, 7-9 mm long, 5 angled, warty or rugose, glabrous or puberulent, base truncate. Chromosome num- ber n = 29 (Kruszewska 1961) Phenology.—Flowers from May to November, or year round in cultiva- tion in areas that receive no freezing temperatures. Flowers open in the evening around 4 p.m. and close the following morning Distribution (Fig. 9)—In Mexico found cultivated in most villages and towns, though often seen as escaped, persisting long after abandonment. Selected strains of Mirabilis jalapa have become ubiquitous weeds through out the tropical and subtropical areas of the world. Altitudes mainly LOO— 3000 m. Representative specimens: MEXICO. Chiapas: Mpio Ocosingo, ruins of Yaxchilan, on banks of Rio Usumacinta, Breedlove 33906 (TEX); San Judn Chamula, Santiz-Ruiz 970, 829 (TEX). gees Federal: Pedregal, Pringle 6433 (F, ane NY, UC); Pedregal, Lyonnet 129 (G, XU, MO, NY); Xochimil co, Ventura 1579 (MEXU, NY, RSA). Chi- huahua: Moj anes Kuobiecd 5289 (F); Mpio Tuxtla erie Monte Grande, Ventura & Lopez 1656 (MEXU, NY), Amatenango del Tie Breedlove 14444 (LL, MICH, NY): be- tween San Richardo & Ocezucuantla, Ne/son 2987 (G). Colima: Rancho El Jabali 22 km NNW of oe at Jalisco line, Hacienda San Antonio NW of Lago EI — Garcia et . 8208 (RSA, TEX). Durango: vicinity of City of ee Palmer 030 (E, G : UC); vee de Santiago Papasquiaro, 3.5 km W a Soladad, 11 km on SE Papasquiaro, Diaz 824 (NY); Mina la Amparo 6 km pee of Las Higueras, Mpio de Rodeo, Torres et al 4225 (RSA). anal aese: near SuanAeA Kenoyer 1755 (G). HIDALGO: Rio Tula, Ra Moore 3309 (G); Rd N of Mineral (Real) del Monte, Straw & Gregory 1126 (MICH A). Tee: Valle de Guadelupe, Hwy 80, Le Duce & Sydor 42 (MEXU, TEX); 3 mi N - Tapalpa, Walker 78H40 (NY); Mpio Tuxpan, near Colima, Fuentes 551 (MICH). Mexico: Texcoco, Runyon 1362 (TEX); Temple of Quetzalcoatl, Barkley et al. — 642 Sipa_16(4) T qT | I ! | ] | ] \ | 4° los" tor: 96° 90° aaa ° 30° ee ? Lar 7 a wl Le \ a 24° » ‘ AY ws — eee | . Pe — 10° ) 4 18° — ae ®M. jalapa e e 0 100 200 300 ee - ——| Km 108° tore . . n ! l ! n ui n Ma Fic. 9. Distribution of Mirabilis jalapa in Mexico and Texas. 7259 (TEX), Hwy 54, 6 mi N of Ixtapan del Sal, Le Duc & Sydor 94, 96 (MEXU, TEX); Molino de la Flor, Texcoco, Matuda 18936 f - Mpio Temascaltepec, Tejupilco, Hinton 4385 (G, MICH, NY). Michoacan: ca. 4.5 n of Maravato, Soule 2472 (TEX), Patzcuaro, near ae ee Vasco, Le noe 4 (TEX); Hwy ‘ Patzcuaro to Morelia, Le Duc & Sydor 90, 91 (MEXU, TEX); 5 mi W of Cd Hidalgo, Sawer 1110 UO, Morelos: Along trail from town of sae to ruins, Ayers G Scott 111 (TEX); Cuernavaca, pyramids of Teopanzolco, Leon MXOOI (RSA). Nayarit: near Tepic, Rose 2137 (US). Nuevo Leon: Hacienda Pabillo, Galeana, Taylor 116-B, 116-a (F, TEX). Oaxaca: Valley of Oaxaca, Smith 791 7 NY); 5.5 km adelante de San Pedro Jocotipac, Dist. Cuicatlan, Salinas 4335 (TEX); 10 mi W of Mitla, Rowell et al 17M485 (TEX); Hwy 175, village of Guelatao, Le Duc & dor 127 (TEX); Yegul, Le Duc & Sydor 109 (TEX), Monte Alban, in the ruins, Le Duc & Sydor 107 (TEX); Yanhuitlan at Dominican Monastery, Le Duc & Sydor 106 (TEX). Puebla: Tehuacan 4 km al E de Azumbilla, Sanchez-Ken 252 (TEX); Meseta de San Lorenzo, Salinas F-4010 (TEX); Puebla, Aréne s.n. (NY); San Luis emi Purpus 3375, 3522 (UC), Tehuacan, (F, MO). Chaing et al. 2304 (RSA, TEX). Queretaro: Querétaro, Arsene . Agatel 10526 San Luis Potosi: Mpio San Antonio, “ani Alcorn 3332 (TEX); 2 mi S of rte. 70 on road to ees, Moran 7650 (MICH, RSA). Sinaloa: near eee Rose et al. 13767 (US). Sonora: Las Tierritas de E] Temblor, Sierra El Tigre, WAzte 3400 (G, MICH). Tabasco: Mpio Huimanguillo, E] Arenal, Ventura 20043 (GH). Tamaulipas: - mi S of Cd. Victoria, Hwy 85, Wilson 12278 (TEX). Tlaxcala: Tlaxco, Azedérraga s.n. (TEX). Veracruz: Hwy 125 to Jalapa, junct. with rd to Jalcomuco, Le Duc & Sydor 159 oa Mpio Alcotonga, Dodds 48 (MICH, NY); Mpio Perote, Tolalco, Ventura 9163 (LL, MICH); Bafios del Carrizal, Purpus B5494 (UC); Biological Stat. Los Tuxtlas, Gomez-Pompa 4613 (G); Vista Hermosa, Jilotepec, Ortega 579 (F); Coatepec, Pedraza & Ortega 297 (F). Yucatan: Izamal, Gawmer 548 (F); Chichen Itza, Steere 1479 (F, MICH); Xocén, Mpio Vallodalid, Acosta 238 (RSA). Le Duc, Revision of Mirabilis section Mirabilis 643 Zacatecas: 10 mi W of Fresnillo on rd to Valparaiso, E of Santa Cruz 15 km, Anderson & aie 3587 (G, ae NY,US). US TEXAS. mal Co.: New Braunfels, Lindheimer 1103 (TEX). El Paso Co.: El Paso, ae 13 a Travis Co.: Austin, Ferguson s.n. (TEX). Mirabilis jalapa is an exceedingly variable species, typical of a horticul- turally important plant. Emmart (1940) shows that M. jalapa was culti- vated by the Aztecs for its medicinal properties and for its showy fragrant flowers, long before the Spanish conquest of Mexico. Collections of M. jalapa were introduced into England within 75 years of the Conquest. By the time of Linnaeus (1753, Species Plantarum), the plant had been in cultiva- tion in Europe for about 200 years. The specimens from which the species was described were those of cultivated plants. The numerous early syn- onyms are a result of attempts to seggregate the various cuultivated strains. Considerable propagation of the species had been done and many plants were well established in Aztec gardens before the Conquest (1.e., prior to 1521). In all my field work in Mexico, I have never seen any population that was not in cultivated or formerly cultivated areas (herbarium collec- tions from remote areas ascribed to M. jalapa have on closer examination proved to be misidentified.) It is questionable whether there is any extant population that represents a true wild progenitor. Today, distribution of the species, in Mexico, encompasses all areas which were part of the Aztec empire and sphere of influence, particularly around ancient ruins. It is also dominant in the towns and cities established by the Spanish during the colonial period. Names applied to putative hybrids between M. jalapa x M. longiflora Mirabilis hybrida Lepeletier, Ann. Mus. Natl. Hist. Nat. 8:481. 1806. Type: Cultivated in the garden of M. Lepeletier, 1806. Apparently no voucher collection was made. Lepeletier’s plant was grown from seed received from M. Fabus d’Attichy of Champagne, France, who found a single natural hybrid in his garden in 1802. Mirabilis jalapa vat. oaxacana Heimerl, Notizbl. Bot. Gart. Berlin-Dahlem 11:450. 1932. Type: MEXICO. Oaxaca: ca. the city of Oaxaca, 1842, Franco s.n. (HOLOTYPE: W — PHOTOHOLOTYPE: F!; isoryPE: F!). The. holotype was destroyed during ve war in 1945, pe comm., et Riedl, Director W. There is a photograph of the holotype and as mi aietial at F. The name ne oaxacae Heimerl is a nomen nudum which appeared in Beitr. Syst. Nyctag. 20. 7, and is therefore not valid. Considering the site from which the above referenced entity was ob- tained, and its apparent intermediate morphology, it is presumed to be of hybrid origin. The foliage resembles M. jalapa, but the floral characters resemble those of M. /ongiflora var. longiflora. It differs from both, in having an obovoid anthocarp with prominent ridges and warty areas. The anthocarps of M. longiflora var. longiflora are warty throughout, with indistinct ridges, while those of M. jalapa are ovoid or elliptical. Stipa 16(4) NN ise ay ACKNOWLEDGMENTS A special thanks goes to Ruben Mitchell and Dennis Brown for allow- ing me access to the SEM in the Cell Research Institute, University of Texas at Austin, and Stephen Hall for his assistance and access to the pollen laboratory. The following people also gave me invaluable help in a variety of ways and I am most grateful: for their advice and support Jerry Brand, Linda Escobar, Virginia Hainesworth, Richard Spellenberg, Heinz Schulze, and Billie Turner; for being pleasant field companions Alice Hempel, Mark Mayfield, Lindsay Woodruff, Tom Patterson and Jacqui Soule, who made traveling in Mexico an experience to be remembered; for their suggestions and advice in the writing and editing of the manuscript Ted Barkley, Carol Todzia, Guy Nesom, and Elezar Reyes, and especially Guy for the Latin and Elezar Reyes for the Spanish translations. In addition, I must give special credit to Marcia Sydor, my sister, without her keen eyes many col- lections would not have been made, and Jackie Shanahan for her illustra- tions. Iam most grateful to the curators of CAS, MEXU, MO, RSA, UC, UCLA for their hospitality during visits (especially Mario Sousa at MEXU) and the curators of CAS, EF G, GH, K, MICH, MO, NY, PB, RSA, UC, US who kindly lent specimens for this study. Financial support for my research was made possible in part by a grant from the AAUW Educational Foundation and an honorarium from the Bernice Moore Scholarship Fund, University of Texas at Austin. APPENDIX Additional Specimens Examined Abrigo, R. s.n.(10) Breedlove, D. 10439(10), 14643(10), Alcorn, J. 1401, 297310 47T494(10) Arguelles, E. 1083(5b), 1145(5b), 2648(10) Burger, W. & T. Antinio 10939(10), Arsene, G. s.n., 1675(10) 10855(10) Ascencio, M.A. 74(10) Bush, B.F 1209(10) Avila, S. 135(10) Bustillo, S. 219(10) Balls, E. B 5494(10) Cabrera, E. & O. Tellez 2398(10) Barkley, FA. 14/a521(5b), 16024(Sb) Caec & Seler 53(10) Barneby, R.C. 5112(Sb) Caldron, S. 1800(10) Bartlett, H.H. 12381(10) Calzada, | 75(10) Beals, J.M. s.n. (Sb) Calzada, J.P. et al. 6315010) Benito, S. 21¢10) Campos, A. & G. Toriz 3209(7a) Bingham s.n. (Sb) Carlson, M.C. 109(10 ikley, E.R. B-S67(Sb) Castrejon, J. G10) 23(10) Blumer, J.C. 2190(Sb), 2205(5b) thiang, F et. al. 12267(5b), 9551 Fb) Bossé, G. 7860(10) Conzatti 136(10) Bonilla, R. & E. Monsalvo 30(10 Correll 33672(5b) Bourgeau, M. 61(10) Correll, D.S. & I.M. Johnston Brandegee, TS. s.n.(2) 20192(5b) 992 10), Le Duc, Revision of Mirabilis section Mirabilis Cory, V.L. ee ere il E. & E. Palm 68(5b) ss =e A. fone) 855(5b) Diaz, I. 39501 Di Hoe N.P. 68 ne , M. et al. ne Ze Edwards, & a s.n.(10) Eapeien eer Epling, C. & Stewart s.n.(5b) Ferris, R.S. & C 2579(5b) Fisher, G.L. 2113(10) Flores et al. 1713(10 M. & J. Dorantes 146(10) Garcia-Saucedo 2589(10), 2635( Gaumer, G.F. 1843(10), fete ae 548(10) Genelle, P. & G. ee 886(10) Gentle, P. 814, sey 32(10) ntry, mae 1911, 672M(10), en Gonzalez, g 10) Greene, E.L. s.n.(5b); 12511(5b) Greenman, J.M. & M.T. 5777(10) Gregg, J. s.n., 344, 231(10) Guzman, M. & D. Castro 1406(10) 3 1) Hanson, C.A. 507(5b), 503(Sb) Harde Le Sueur 613(5b) Harker & Mellowes 168(10) & T.H dean, 5818(Sb) ernandez A. C. eon 5(10) eee 22, 7 et al. 171(10) Hern: andez, H. 28(10) inkley & Te 4682 1(5b) Hinkls s.n., 3168, 129(5b) ton, G., et al. 20491(5b), 4356(1), 10207(78) D. Duncan 2519(5b), 588M(10), 645 Hinton, G.B. 4577(10), a 1333(10), 4644(10), 6554(10), 6 10) Jimenez a Pal 00038(10) Jones, M.E. s.n. ae WW. eds: 386(5b); 178(5b); 56(5b) ea . 984(10) Kenoyer, . S.n., King, R.M. Kunze ae Kusche, J. s.n.(Sb) Lavin, M., et al. 4850(9) Laughlin, R. 734, 2828(10) Le Duc, A. & M. Sydor 75(10), 45(10), 46(10), 74(10), 158(10), 127010), 90(10), 171(10), 42(10), 159(10), 47(10) Le Duc, A. et al. 255(7b) Leavenworth, W.C. 911(5b) eee E. & T. Reeves P12426(5b) mmon s.n. 2867(5b) ne 567(10) Loew, D. s.n. (Sb) Loomis, HE 7267(5b); 1267(5b) Lopez, R a Villarreal 902(5b) Li A600(10) Soderstrom 4616(10) Luckow, B. & C. Maguire 35500 (NY, VEN). Bolivar: Chimantdé-Massif, Chimanta-tepuf, C section, above summit cam 1,940 m, 4 Feb 1955 (stam. fl), J. Steyermark & J. Wurdack 412 (NY, VEN), poke yee on above summit camp, between Middle and Upper Falls of Rio Tirica, 1,925—1,940 m, 4 Feb 1955 (pist. fl), J. Steyermark & J. Wurdack 592, 593 (NY, VEN), summit, along 8 Mojado, 1,985-1,910 m, 23 Feb 1955 (fr), J. Steyermark & J. Wurdack 1107 (NY, VEN); NW slopes of Churi-tepuf (Murt-tepui), 25 Jan 1953 (piste. fl, fr), J. Waurdack 34198 (NY, VEN); Uaipan-tepuf, 1,900 m, 1-15 Feb 1948 (fr), K. Phelps & C. B. Hitchcock 394 (NY, VEN) Clusia hexacarpa is closely related to C. radiata, but easily distinguished by the larger bracteoles, fewer sepals and petals, and most notably, the perianth persistent in fruit. The type of subsp. praritepuiana Steyermark is notable only for its rose markings on the petals and purple stigmas. In other species of C/usia I have seen in the field, stigmas are often more purple when exposed to the sun. Maguire had intended to describe another subspecies of C. hexacarpa, bearing the subspecific epithet “octocarpa”, and several specimens were dis- tributed bearing that epithet. However, Maguire never published the name, and study of those specimens indicate that population differs from others only in having eight instead of six carpels. Many more specimens were available to me than were to Maguire, showing that floral merosity can change significantly even within populations. 668 Stipa 16(4) 8. ee ueoeta a & Phelps, Bot. Soc. Venez. Cienc. Nat. 14. 1952. "PE. EZU . AMAZONAS: Cerro Guanay, SE escarpment, 1,800 m, 4 Feb oe ee A, fr), B. Maguire, D. Phelps, C. B. Hitchcock & G. Budowski 31758 (Ho10- tyPE: NY!; isorypes: F!, VEN). Free-standing shrub to small tree to 5 m tall. Branchlets tetragonal, 6— 8 mm diam., the angles obtuse, not alate. Leaves petiolate; blades cartilagi- nous, widely obovate, widely elliptic to suborbicuar, 6-10 cm long, S—7 cm wide, apex and base widely rounded, base not decurrent on the petiole, symmetric, nitid above, pallid and without magenta glands below, midrib slightly elevated above, prominently elevated below, the secondary nerves barely distinguishable above, numerous, united by a submarginal collect- ing vein, linear latex canals inconspicuous, the margin entire, opaque, some- what scarious, glabrous; petioles broadly marginate, 6-10 mm long, gla- brous. Staminate inflorescence unknown. Pistillate inflorescence a single (rarely 3) terminal, pendent flower(s); peduncle angulate, 1.5—-3 cm long, subtended by 2 small leaflike braces 1.5—5 cm long, 1.2—3.5 cm wide; bracteoles 2, carnose, oblate, 12-14 mm long, 7-9 mm wide, apex broadly rounded, carinate, the margins entire, opaque, scarious. Pistillace flowers jhe white; sepals 4, decussate, carnose, very widely ovate, 12-14 mm long, 7— 92 mm wide, apex broadly rounded, linear latex canals inconspicuous, promi- nently carinate, the margin opaque, scarious, entire, glabrous; petals 8, cartilaginous, the outer and inner identical in shape, progressively smaller acropetally, the outer 4 decussate, the inner 4 imbricate, very widely obo- vate, 20-25 mm long, 16-20 mm wide, apex sol rounded, linear latex canals few, moderately conspicuous, the margin entire, opaque, not scari- ous, staminodes numerous, 3.8-4 mm long, the filaments flat, 3—3.5 mm long, the anther linear, 0.3-0.5 mm long, apex truncate, base not distin- guishable from filament, the connective darkened; pistil globose; carpels 7-10; styles minute, 0.8-1.0 mm long; stigmas carnose, cuneiform, subpeltate, 34 mm long and wide in fruit. Fruit ovoid, 3.54 cm long, 3.04.0 cm diam. Distribution.—Endemic to Cerro Guanay, Amazonas, Venezuela, at 1,800— 2,000 m elevation. Ecology and conservation status.—Clusia radiata occurs on rocky outcrops along steep slopes. Cerro Guanay is a remote mountain and although noth- ing is known of the population biology of the species, it is likely that it is not threatened. Representative specimen examined: VENEZUELA. Amazonas: Cerro Guanay, sum- mut, 2,000 m, 4 Feb 1953 (fr), B. Maguire et al. 31749 ( Clusia radiata is most closely related to C. hexacarpa, but is easily distin- guished from it by the wider leaves, longer bracteoles, uniform and more numerous sepals, and more numerous petals. Pipoty AND Grarr, Clusia section Criuva in Guayana 669 9. Clusia maguireana Pipoly, sp. nov. (Fig. 4) Quoad petiolos late ee flores solitarios vel 3, fructum ovoideum, laminas cartilagineas C. orang valde affinis, sed ab ea petiolis obsoletis vel 6 (non 6-15) mm longis, floribus 3—6 (nec 1-3), i (nec terminalibus), petalis suborbicularibus (nec orbicularibus vel seine denique staminodiis 4 (non 8-9) nee distinguicur. Shrub to small tree to 4(-10) m tall; branchlets tetragonal, strongly ridged but not winged, (4.5—)5—9 mm diam., glabrous. Leaves sessile; blades cartilaginous; very widely obovate, or rarely oblong, (3.5—)5—7(-10.2) cm long, (2.3-)3—S(-8) cm wide, apex rounded to truncate, base obtuse, not decurrent on the petiole, symmetric, nitid above, pallid and without ma- genta glands below, midrib raised above and below, secondary veins 20—30 pairs, not or barely visible, glabrous, w/o visible latex canals, the margin thick, opaque, revolute, entire, glabrous; petioles obsolete to 7 mm long, deeply canaliculate, ligulate, glabrous. Staminate inflorescence axillary, pendent, a twice-branched cyme, 2.5—3 cm long, the peduncle 5—7 mm long, the pedicels tetragonal, 5-7 mm long, glabrous; bracteoles 2, carti- laginous, suborbicular to oblate, 44.2 mm long, 4.5-4.7 mm wide, apex rounded, basally rugose, carinate, slightly cucullate, the margin entire, thick, opaque, glabrous; sepals 2, decussate to bracteoles, thinly coriaceous, ob- late, 6.3-6.5 mm long, 7—7.2 mm wide, apex rounded, cucullate, the latex canals obscure, uacoage the margin chartaceous, ca. 0.7 mm wide, opaque, entire, glabrous; petals 4, decussate in pairs and to the sepals, thickly carnose, dimorphic, latex canals conspicuous the outer pair sublinguiculate, 5.2— 5.9 mm long, the claw 0.7—1 mm long, 2.1 mm wide, the limb oblate, 4.5-4.9 mm long, 5.7-5.9 mm wide, apex broadly rounded, the margin thick, opaque, entire, glabrous, the inner pair unguiculate, 44.3 mm long, the claw 1-1.1 mm long, 1.3-1.4 mm wide, the limb oblate, 3—3.2 mm long, 3.5—3.7 mm wide, apex broadly rounded, the margin undifferenti- ated, opaque, entire, glabrous; receptacle convex, cubic, 1.6-1.8 mm high, 1.1-1.2 cm long and wide, without resin; androphore cubic, stamens 26, equal in size, free, 1-1.1 mm long, the filaments fleshy, flattened, free, 0.2-0.3 mm long, the anthers oblong, 0.9—-1 mm long, ca. 0.5 mm wide, apex emarginate, base deeply cordate, on connective darkened, glabrous, longitudinally dehiscent over entire length; pistillode absent. Pis- tillate inflorescence pendent, a reduced cyme, 1.5—2(—2.5) cm long, the peduncle 2—3(—5) mm long, the pedicels tetragonal, 2—3 mm long, gla- brous; bracteoles 2, cartilaginous, suborbicular to oblate, 2.1—2.2 mm long, mm wide, apex obtuse, basally rugose, carinate, slightly cucullate, the margin entire, thick, opaque, glabrous; sepals 2, decussate to bracteoles, thinly coriaceous, orbicular, 6.4—-6.6 mm long, and wide, apex rounded, cucullate, the latex canals obscure, marginate, the margin chartaceous, Ca. 0.7 mm wide, opaque, entire or sparsely incised, glabrous; petals 4, decus- 670 Sipa 16(4) Fic. 4. Clisia maguireana Pipoly. Habit. B.Fruit. C. Pistillatce cyme. D. Staminate pedicel, and separated sepal, showing scarious margin. E. Petals and androecium. A-C, drawn from type; D-E, drawn from Steyermark 93542. sate in pairs and to the sepals, thickly carnose, dimorphic, latex canals con- spicuous, the outer pair suborbicular, 4.5-4.9 mm long and wide, apex broadly rounded, the margin thick, opaque, entire, glabrous, the inner pair suborbicular, 4.24.3 mm long and wide, apex broadly rounded, the mar- Pipoty AND GrafFF, Clusia section Criuva in Guayana 671 gin undifferentiated, opaque, entire, glabrous; staminodes 4, resembling stamens, 2—2.4 mm long, the filaments flattened, 1.4—-1.6 mm long, broadly expanded at base, the sterile anthers orbicular, ca. 0.7 mm long and wide; pistil 10—carpellate, strongly ribbed, subglobose, 6-6.3 mm long, 5.6-5.8 mm diameter, styles obsolete, stigmas attached apically and basally, cunei- form, 2—2.1 mm long, 1.4—1.6 mm wide, persistent. Fruit subglobose 1.7-2 cm long, 1.4-1.6 cm wide at maturity, strongly ribbec Typr:. VENEZUELA. Botivar: Deto. Piar, Macizo del Chimanta, sector N-central of Chimantd-tepu/, eastern headwaterscabeceras of Cafio Chimanta, 5°18'N, 62°09'W, 2, 000 m, 26-29 7. 1983 (pisc. fl, fr), J. Steyermark 127980 (HOLOTYPE: VEN; isorypes: BRIT, I MO-2 shts, US). vase VENEZUELA. Botivar: Saddle between Terekeyuren and Murisipan-teput, 1,650 m, 22 Mar. 1987 (fr), B. Holst 3469 (BRIT, MO, US, VEN); Deto. Piar, central & W part of saddle between Camarcaibarai-tepui and Tereké-Yurén-tepuf, 1,800-1,900 m O>°o2 NN; 62°0 DW, 23 May 1986 (fr) R. Liesner et al. 21006 (BRIT, MO, US, VEN), puyalie tepuf, sum of central portion of NE arm (W range), between “Drizzly Camp” and “Rio Lomita ae 1,800-1,850 m, 5 May 1964 (pist. fl, fr), J. A. Steyermark 93442 (E, NY, US, VEN), same general area, woods beside small creek among savannas S of Jimmy Angel crash site, 1,800 m, 7 May 1964 (stam. fl), J. A. Steyermark 93542 (NY, US, VEN); central-NW section of E arm, Auydn-tepuf, 05°57'N, 62°25'W, 1,950 m, 27 Aug 1983 (stam. fl), O. Huber et al. 8096 (MYF, NY, VEN); Uei-tepui, between SE slope and summit, between Luepa & Cerro Venamo, vic. of km. 125, S of El Dorado, 1,100-1,300 m, 7 Mar. 1962 (fr), J. A. Steyermark & L. Aristeguieta 20 (F, US, VEN); Macizo del Chimanta, Deto. Piar, central-southern section, wide valley between NE border of Feces tepuf and central section of Chimanta-tepui, $ sper 5°16’N, 62°09'W, 2,1 , 11-15 Feb. 1985 (stam. oy, J. Pipoly et al. 7261 (MYF, NY, MO, VEN), (pist. f . i 0. Huber, J. Pipoly et al. 101 MYE, NY, VEN), (scam. i O. Huber, J. Pipoly e al. 10219 (MYF, NY, VEN); eee section of Chimantd-tepuif, E headwaters of Cafio Chimanta, 05°18'N, 62°09, 2,000 m, 26-29 Jan 1983 (pist. fl, fr), O. Haber & J. Steyermark 6934 (MYF, NY, VEN), (pist. fl, fr), J. Steyermark et al. 127980 (NY, VEN); Abécapa-tepui, Bonnetia foresc, NW part of summit, 2,125-2,300 m, 13 Apr. 1953 (fr), J. A. Steyermark 74861 (F, US VEN); Abdcapa-tepuf, above Ist line sandstone blufts, 2,000—2125 m, i Apr. 1953 (fr), J. A. Steyermark 75002 (BRIT, F, MO, US, VEN), Apdcara-teput, elfin forest foumanon on plateau of SE-facing pe shoulder of Apdcara-tepur, 2,000 m, 19 ie 1953 (fr > Steyermark 75717 (F, S: oo tepuf, SE-facing forestee ale ‘helow escarpment, 1,880-1 30 m, a Feb. 1955, J. A. Steyermark & J. J. Wurdack 1166 (pist. fl, fr), (FE. NY, US, VEN). G NA. Potaro- Soarani Region: Kaieteur National Park, N of Menzie’s Landing, 05° . oe 400 m, 26 Jan 1987 (fr), J. Pipoly & G. Gharbarran 10029 (BRIT, BBS, CAY, FDG, K, NY, P, U, US); Pakaraima Mts., Mt. Wok nS summit ridge of Kamiewah Pinnac ae NEt 6S Pinnacle, “Little Avanganoa 05°O 59°52'W, 1,550-1,650 m, 17 Nov 1993 (pist. fl), T. Henkel et al. 4495 (BRG, BRIT, a Distribution —Endemic to the eastern tepuis of the state of Bolivar, Ven- ezuela, and adjacent Guyana, at (400—)1,100—2,300 m elevation. Ecology and conservation status.—Clusia maguirveana is restricted to rocky outcrops, usually on overhanging ledges, and is subject to high winds and driving rains. I observed small patches of this species on the Kaieteur Pla- teau, but most of them were not fertile. It appears that the species has a 672 Sipa 16(4) considerable range, and because it occurs in a very hostile habitat, it is most likely not threatened. Etymology.—It is with great pleasure that I dedicate this species to the late Bassett Maguire, prodigious fieldworker, student of neotropical Clusiaceae and the flora of the Guayana Highland during his long career at the New York Botanical Garden. He conducted many expeditions to the most remote localities of the Guayana Highland, where he collected excel- lent specimens despite the harsh field conditions. Clusia maguireana is most closely related to Clusia hexacarpa, but may be distinguished from it by the axillary inflorescence with greater number of owers, the subsessile leaves, suborbicular petals and fewer staminodes. 10. Clusia opaca Maguire, Bot. Mus. Leafl. 15:62. 1951. Type: BRAZIL. mazonas: Path between headwaters of Ira-Igarapé and headwaters of Igarapé Abid, affluent of Rio Taraira, 4-6 Jul 1948 (stam. fl), R. E. Schiltes & F Lopez 10192 (HOLOTYPE: NY!; IlsoryPpe: GH) Clusia reducta Steyerm., Fieldiana, Bot. 28:391. 1952. syn. nov. Type. VENEZUELA. Amazonas: Cafo San Miguel, above mouth OE Ichana, Guainia, 125 m, 26 Mar 1942 (pist. fl, fr), LZ Williams 14898 (HoLoryer: Fl; isorype: US!), Free-standing shrub to tree 6 m tall; eas subterete, 3—4 mm diam., sparsely longitudinally ribbed, glabrous. Leaves petiolate; thickly coria- ceous, oblong or elliptic, (4.5—)7—9 cm long, 2.5—3.5(—5) cm wide, apex and base broadly rounded, the base not decurrent on the petiole, symmet- ric, nitid above, pallid and without magenta glands below, midrib im- pressed above, prominently raised below, the secondary veins numerous, inconspicuous, united by a submarginal nerve, linear latex canals incon- spicuous, the margin revolute, entire, glabrous; petiole canaliculate, (0.6-) I—1.5 cm long, glabrous. Staminate inflorescence terminal, pendent, a com- pact compound cyme, 6—18-flowered, 1.5—2 cm long, and wide; subtended by a pair of leaflike bracts, 1.5—4 mm long, 0.7—1.2 cm wide, apex broadly rounded, base obtuse, the margins revolute, entire; peduncle ca. 3 mm long, subterete, glabrous; bracts carnose, oblate, 1.8-2.2 mm long, 2—2.4 mm wide, apex rounded, carinate, the margin entire, opaque, not scarious, glabrous; bracteoles 4, decussate, oblate, carnose, 1.8—2.2 mm long, 2—2.5 mm wide, apex obtuse, carinate, the margin entire, opaque, scarious, gla- brous; pedicels angulate, 3-7 mm long, glabrous. Staminate flowers white; sepals 4—S(—6), the outer opposite, decussate to the bracts, membranaceous, orbicular, 3-5 mm long, and wide, apex broadly rounded, linear latex ca- nals few, conspicuous, the margin entire, hyaline, scarious, glabrous, the inner ones imbricate, 3.8-4.2 mm long and wide, apex broadly rounded, the margin entire, hyaline, scarious; petals 4—5, coriaceous, oblong, the outer ones opposite, the inner imbricate, all similar in shape, acropetally decreasing in size, 4-6 mm long, 3.8-4.5 mm wide, apex broadly rounded, Pipoty AND GraFF, Clusia section Criuva in Guayana 673 cucullate, the linear latex canals obscure, margins entire, opaque, not scari- ous, glabrous; androphore pentagonoid, concave; stamens numerous, 0.9— 1.5 mm long, the filaments flattened, connate basally, the distal ones 0.2— 0.3 mm long, the interior 0.5—0.7 mm long, the anthers linear, oblong, 0.7-1 mm long, 0.2-0.3 mm wide, apex rounded, base obtuse; pistillode obsolete, not resiniferous. Pistillate inflorescence as in staminate, but bracteoles 2—2.5 mm long, 2.3-2.5 mm wide. Pistillate flowers as in stami- nate, but sepals 4-6, 4—4.5 mm long, 4.5-5 mm wide, petals 4-5 mm long, 3.3-4 mm wide; staminodia numerous, 0.8-1.2 mm long; filaments flat, linear, anthers barely differentiated, ca. 0.2 mm long, apex rounded, base not differentiated; pistil subglobose, 3-5 mm long and wide; carpels 5; styles subobsolete; stigmas sessile, orbicular, ca. | mm diam., ovules numerous. Fruit globose, 1.2—1.5 cm long and in diam. Distribution.—Western Amazon Basin of Brazil, Venezuela and Colom- bia, at 80-160 m elevation. Ecology and conservation status.—Clusia opaca is endemic to “Amazonian caatinga,” campinas, and “Bana” formations, all of which are lowland subxeric areas of deep white sands, often near black water rivers, but not subject to inundation (Macedo & Prance 1978; Prance 1979; Prance & Schubart 1978). These environments all share essentially the same nutri- ent cycling regime, and vary only in terms of local species composition. The white sands are coarse in texture, extremely well-drained, and derived from eroded tepuis. Clusia opaca is locally common, and because it occurs near the major black water rivers of the Amazon Basin, which support relatively heavy river traffic, it should be considered threatened. Common names,—‘Pai-nan-ge” (Brazil, Maka language); “copei,” “upihi,” “banitha,” “cupi” (Venezuela). Specimens examined: COLOMBIA. Caqueta: Araracuara, sandstone plateau ee military campy 00°37'S, 72°24'W, 18 Oct 1990 (fr), J. Duivenvoorden & A. Cleef 314 (B , U). VENEZUELA. Depto. Atabapo: SE bank of the a part of Cafio oa at ere de Yagua, 03°36'N, 66°34'W, 120 m, 8 May ee ), G. Davidse et al. 17561 (MO, MYF, NY, VEN); El Almidén, limit of Depts. of Atabapo ne Rio Atacavi, slope 2, 03°04'N, 67°06'W, 80 m, Nov 1989 (fr), J. Velazco 869 (BRIT, PORT, VEN), Near San Antonio, Alto Orinoco, along Rfo Orinoco, 120 m, 15 Aug 1982 (pist. fl), 7: Rutz et al. 3964 (MY, VEN); Rio Guainia, 14 ay 1953 (pist. fl), B. Maguire & J. W ee 35642 (F, MO, NY, US, VEN); savanna 5 km E of Maroa, 130 m, 6 Oct 1957 (stam. fl), B. al et al. 41705 (FE, NY, US, VEN), (pisc. fl bud), B. Magure et al. ana NY, US, EN) Maroa- Naina rox acl, p PennieeD Rio Guainfa and Cano Pimichin, ca. 2 km beyond ee airport, 02°43'N, 67°38'W, 8 Oct 1978 (pist. fl, fr), H. Clark 6863 NY, US VEN); Maroa, i pane 127 m, 1942 (fr), LA Wiliams 14254 (% US); Pimichin, 128 m, 2 Jul 1942 (fr), LA Williams 14183 (E, US); savanna 0.5-1.5 km N of Puerto Colombia, opposite a. a Abia, 130 m, 12 Oct 1957 (fr), B. Maguire et al. 41843 (BF, NY, VEN); Depto. Casiquiare, Rio Casiquiare, 40 km beyond the mouth and 5 km NE of N): Ri camp, 28 Jan 1991 (fr), M. Colella et al. 1610 (BRIT, NY, VEN); Rio Casiquiare, 162 kms 674 Stipa 16(A) from the mouth, 3 Feb 1991 (fr), AL. Cole/la et al. 1752 (BRIT, NY, VEN){mixed collec- tion with C. gaudichaudii Choisy ex Pl. & Tr}; Cano San Miguel, sector “Las Tinajas,’ 02°39'N, 66°45'W, 160 m, 25 Apr 1991 (fr), G. Aymard 9237 (BRIT, PORT, VEN). Clusia opaca 1s most closely related to C. gwvayanae, burt easily distin- guished by the obtuse to broadly rounded leaf bases, the shorter peduncle, suborbicular sepals, coriaceous petsls, numerous staminodes and sessile, orbicular stigmas. 1. Clusia guayanae Pipoly, sp. nov. (Fig. 1) Species haec quoad petiolos canaliculacos, flores 6-18 in = insidens, fructus globosum, aevem, non costatoque, C opacae valde arcre affinis, sed ab ea laminis ad basem obtusis ve unculis 5-12 (non 3—5) mm longis, sepalis 2 — late rotundatis (non acutis vel obtusis), pec (non 4—6), oblatis (nec Seen petalis cartilaginosis (non coriaceis), staminodiis 4 (non numerosis), denique stigmatibus pentagonis (non orbicularibus) perspicue recedit. Shrub to 2 m; branchlets tetragonal, 3.5-3 mm diam., glabrous; latex white. Leaves petiolate; blades coriaceous, oblanceolate to obovate or rarely suborbicular, (3—)4—6(—7) cm long, (1.5—)2—3.5(-4) cm wide, obtuse to truncate, base obtuse to rounded, nitid above, pallid and without magenta glands below, glabrous, midrib slightly raised above, prominently raised below, secondary veins numerous, at a steep angle from midrib to a large submarginal collecting vein, ca. 0.3 mm from margin; petiole canaliculate, 3-5 mm long, with a ei channel at base. Staminate inflorescence termi- nal, a pendent panicle, 2—3 cm long, 2—3 cm wide, the branches cymose, in sets of 3; peduncle squarrose S—8(—12) mm ae secondary branch bracts 2, carnose, depressed ovate, 1.8-2 mm long, 2.1—3 mm wide, apex broadly rounded, glabrous, margins entire, opaque; bracteoles 4, decussate, carnose, depressed ovate, 2.2—2.4 mm long, 2.6—2.8 mm wide, apex broadly rounded, glabrous, margins entire, opaque; sepals 6, coriaceous, decussate, the outer 2, depressed-ovate cucullate, carinate, 4-4.5 mm long, 5—5.5 mm wide, apex broadly rounded, margins entire, opaque, glabrous, the inner ones oblong, cucullate, 5—5.3 mm long, 44.5 mm wide, apex truncate to widely rounded, the margin scarious, entire; petals 4, cartilaginous, oblong to widely oblong, cucullate, apex obtuse to truncate, 3.5—5 mm long, 2-3 mm wide, margin irregular, opaque, thick; androphore concave, cuadrate, 0.8—l1 mm long, 1.4-1.6 mm diam.; stamens 8, oblong, muticous, fila- ments and anthers undifferentiated, 0.8—1 mm long, 0.3—0.4 mm wide, apex rounded; pistillode absent. Pistillate inflorescence as in staminate, but 1.5—2 cm long, 1.5 cm wide, secondary branch bracts 2, as in stami- nate; bracteoles 4, decussate, as in staminate flowers; sepals 6, decussate, depressed ovate, the outer ones sepals 3.8-4 mm long, 4.2-4.5 mm wide, inner ones 4.8-5 mm long, 4—4.2 mm wide, coriaceous, cucullate, cari- nate, apex broadly rounded, the margin scarious; petals deep pink, 4—5, Pipoty AND Grarr, Clusia section Criuva in Guayana 675 a en a D A {il @ Fic. 5. Clusia genie Pipoly. A. Habit. B. Immature inflorescence. C. a shore and ie um. D. Bracts and sepals. E. Corolla. F. Staminodia and ovary. G. Fruit. A-C, drawn from Steyermark 6g Holst 20984; D-G, drawn from Haber 9330. cartilaginous, oblong, 5-7 mm long, 2-3 mm wide, cucullate, apex broadly rounded; staminodes 4, 1.8—2.2 mm long, thin, strap-like, bases united by a flat tube 0.2 mm long, the filaments 1-1.2 mm long, broadly ovate, the 676 Stipa 16(4) anthers widely oblong, 0.8—1 mm long, 0.6-0.8 mm wide, apex muticous, with narrow longitudinal slits; pistil 5-carpellate, subglobose, 2.7—3 mm long and diam., the stigmas peltate, subsessile, pentagonal. Fruit depressed- globose, 0.8-1 cm long, 1—1.2 cm diam., pinkish yellow. Tyree. GUYANA. Cuyunt- nates ReGIon: Mt. Ayanganna, E side on steep slopes, 5°27'N, 59°57'W, 1,250-1,300 m, 12 Mar 1987 (pisc. fl), J. Pipoly, G. Gharbarran, G. Samuels, J. Chin i (HOLOTYPE: ie. isoTyPes: BRIT, NY, US,). Paratryprs. VE UELA. Botivar: Detto. Heres: Macizo del ne central- NE section, slope ae §; 5°54’ - 63°42°W; 1,350 1 Apr. 1984 (pisc. fl, fr), O. Huber 9330 (MYFE, NY, VEN); Deto. Piar, Central aa western part of ai between Camarcaibarai-tepuf and Tereké- Yurén. teput, 05°52’N, 62°01’ W, 1,800-1,900 m, 23 May ao (stam. fl), R. Liesner, et al, 20984 (MO, US, VE N): Camarcaibarai-teput, shoulder of / slope, easternmost ay of Aparaman-tepuf range, 05°52'N, 62°01'W, 1,800 m, 24 ae 1986 (pise. fl, fr), B. Holst et al) 2887 (MO, VEN); Auydn-tepui, summit, south- central region, headwaters of Rio Churtin, 05°5 1'N, 62°32'W, 1,700 m, 29 Mar 1987 (fr), B. Holst 3738 (MO, VEN), plain in westernmost section of W arm of Auydn-cepuf, 25 km SE of Canaima, a JO'N, 62°43'W, 1,650 m, 13 Nov 1984 (stam. fl), O. Haber 9728 (MYF, NY, VE Qu. Distribution. —Endemic to the easternmost tepuis of Bolivar state, Ven- ezuela, and their satellites, at 1,250—1,900 m. Ecology and conservation status. —Clusia guayanae occurs in low, scrub cloudforest formations along edges of sandstone blufts, dominated by Clusia melchiorit, Clusta crassifolia Planch. & Triana, Bonnetia spp., numerous Myrtaceae and Rubiaceae. Its habitat has a hostile climate, with strong winds and rains. Because this species appears to inhabit the most inhospi- table of climates, it is not considered threatened. Etymology.—The specific epithet, “guayanae” is in the Latin locative, for this member of the autochthonous flora of the Guayana Highland. Wit globose fruits without ribs, C gwayanae is most closely related to C. opaca. However, C. gudyande 1s a separated from C. opaca by the broadly acute to obstuse leaf bases, longer peduncle, fewer, oblate sepals, cartilaginous petals, 4 nt h canalicuate ae flowers in 6-18-Howered cymes, and smooth, staminodes and pentagonal stigmas on styles. ACKNOWLEDGMENTS Research for this treatment was greatly facilitated by the Missouri Bo- tanical Garden, which provided loans and numerous gifts for determina- tion. Fieldwork was supported by the New York Botanical Garden and Smithsonian Institutions during the senior author’s tenure at those insticu- tions. We thank Jon Ricketson, Catherine Mayo and Roy Cummings (MO); June Cunningham and Lindsay Woodruff (BRIT) for technical assistance. Peter Stevens (A), Scott Mori (NY) and Barney Lipscomb (BRIT) provided many useful comments on the manuscript. Linda Ellis provided the excel- Pipoty AND GrarF, Clusia section Criuva in Guayana 677 lent line illustrations in aa 1, 4 and 5, while Juan Carlos Pinzén pro- vided Figures 2 and 3. P. Mick Richaedcon (MO) facilitated the participa- tion of the junior iho in the project by arranging a one term study rotation in the senior author's laboratory. REFERENCES ENGLER, A. wee Guttiferae. In: C. Martius, ed. Flora Brasiliensis 12(1):382—474. «1893. Clusia. In: Die Natiirlichen Pflanzenfamilien 3(6). Verlag von Wilhelm Bieta ane Leipzig. Macepo, M. and G. Prance. 1978. Notes on the vegetation of Amazonia II. The dispersal of plants in Amazonian white sand campinas: the campinas as functional islands. Brittonia 30:203-215. Macuire, B. 1979. Guayana, region of the Roraima Sandstone Formation. In: K. Larsen and L. Holm-Nielsen, eds. Tropical Botany. Academic Press. New York. Pp. 223-238. Pipoty, J. and A. Grarr. 1995. The genus a sections Crimvopsis and Brachystemon (Clusiaceae) in northern South America. Sida 16:505—528 Prance, G. 1979. Notes on the vegetation of Amazonia III. ‘The terminology of Amazo- nian forest ie ea subject to inundation. Brittonia 31:26—38 Prance, G. and H. ScHuparr. 1978. Notes on the vegetation of Amazonia I. A prelimi- nary note on the origin of the white sand campinas of the lower Rio Negro. Brittonia 30:60-63. Vesque, J. 1893. Guttiferae. In: A. De Candolle, ed. Mongraphie Phanerogarum 8:1—669. NUMERICAL LIST OF TAXA 1. Clusia duidae Gleason 7. Clusia hexacarpa Gleason 2. Clusia grammadenioides Pipoly 8. Clusia radiata Maguire & Phelps 3. Clusia multilineata Pipoly 9. Clusia ieee om y 4, Clusia asymmetrica Pipoly 10, Clusia opaca Maguire 5. Clusia melchiorii Gleason 11. Clusia guayanae Pipoly 6. Clusia cardonae Maguire LIST OF EXSICCATAE The figures in parentheses refer to the numbers from the numerical list of taxa. Agostini, G & T. Koyama 7458 (6). Amaral, I. do; J. Pipoly et al. 1593 (5). Aymard, G. 9237 (10). Beck, H. et al. 938 (5). Bernardi, L. 931 (6). Berry, P. et al. 4886 (5); 4926 (5). Boom, B.10863 (5). Boom, B. & G. Samuels we (5); 9201 (5). Boom, B. & D. Gopaul 7705 (6). Cardona, F. 2064 (5); 1036 (5). Clark, H. 6863 (10). Collela, M. et al. 1752 (10); 1610 (10). Cowan, R. & J. Wurdack 31257A (5); 31258 (5); 31244 (5); 31198 (7). Davidse, G. et al. 17361 (10). Duivenvoorden et al., J. 2673 (2); 314 (10). Florschiitz, P. & P Maas 3040 (5) Gentry, A. & B. Stein 46596 (7). Ceacie J. J. De 5401 (5); 8762 (5); 8761 (). Henkel, T. et al. 4495 (9); 1535 (5). Hoffman, B. et al. 3104 (6); 3567 (5); 2086 (6); 1184 (5). Holst, B. 2887 (11); 3469 (9); 3738 (11); 3784 (5); 3387 (5). Huber, O. et al. 9728 (11); 9330 (11); 11135 (6); 8096 (9); 12437 (5); 12098 (6). Huber, O., J. Pipoly et 678 Sipa 16(4) al. 10171 (9); 10219 (9). Huber, O. & L. Hernandez 11721 (6). Huber, O. & J. Steyermark 34 (9). Irwin et al., H. 54864 (5). Jansen-Jacobs, M. 875 (5). Koyama, T. & G. Agostini 7473 (6); 7482 (5). iesner, R. et al. 8576 (2); 16649 (3); 17603 (1); 18533 (1); 20984 (11); 21006 (9); 24972 (4); 25135 (1). Liesner, R. & G. Carnevali 22489 (7); 22575 (7); 22797 (4). Liesner, R. & R. Hall 3447 (2). Londofio, C. et al. 1150 (2). Maguire, B. et al. 24585 (5); 24585A (5); 24641 (5); 29124 (5); 30028 (7); 30033 (7); 30179 (7); 30642 (5); 30740 (5); 29535 (5); 30017 (1); 30029 (7), 30927 (5); 31578 (8); 31665 (5); 31746 (5); 31749 (8); 32990 (5); 33283 (7); 33318 (6); 33319 (6); 33391 (5); 36905 (5); sie (7); 37003 (7); 37022 (7); 37144 (7); 37247 (7); 37248 (5); 41705 (10); 41706 (10); 41843 (10); 42132 (7); 42169 (7); 42377 (5); 42731 (5); 42758 (5). Maguire, . & C. Maguire 35174 (5); 35419 (7); 35500 (7); 35393 (7). Maguire, B. & L. Politi 27781 (5); 27894 (5); 28014 (5); 28302 (5); 28797 (5); 32891 (5); 40308 (5); pee (); 46038A (5); 46081A (5). Maguire, B. & J. Wurdack 33905 (5); 35642 (10). Mori, S. et al. 8770 (5). Phelps, K. & C. B. Hitchcock 78 (5); 394 (7); 412 (6). Pinkus, A. 161 (5). ea al. 6614 (5); 7261 (9); 10293 (5); 10330 (5); 10330 (5): 11102 (6). Pipoly, J. & K. Alfred 7711 (5); 7821 (5); 7789 (S). Pipoly, J. & G. Gharbarran 10029 (9). Pipoly, J. & G. Samuels 6864 (5). Prance, G. - Pipoly et al. 29209 (5). Rodriguez, H. 2875 (4). Rufz. T. et al. 3964 (10). Schultes, R. E. & : Lépez 10192 (10). Schulz, J. 10303 (5); 10296 (5). Stein, B. & A. Gentry 1535 (5); 1651 (7). Steyermark, J. et al. 58945 (5); 59866 (7); 59903 (5); 74861 (9); 75002 (9); 75005 (5); 75717 (9); vn (5); 93442 (9); 93542 (9); 92502 (6); 93538 (5) 93881 (5); 98144 (5); 103992 (5); 103159 (5); 103847 (5); 103973 (7); 103975 (7); 105958 (5); 107282 (5); 109363 (5); 109654 (5); 117320 (5); 127980 (9); 129648 (1) 131998 (6); 132175 (6). Steyermark, J. & L. Aristeguieta 20 (9); 22 (6); 90 (6). Steyermark, & S. Nilsson 112 (6); 204 (6). Steyermark, J. & J. Wurdack 412 (7), 592 (7); 593 (7), 942 (5); 1017(5); 1107 (7); 1166 (9). Tate, G. 429 (1); 705 (5). Tavares, A. S. et al. 139 (5). Velazco, J. 869 (10). Williams, Ll. 14183 (10); 14254 (10); 14898 (10). Wurdack, J. 34198 (7); 34092 (5). TAXONOMY OF CYPHOMERIS (NYCTAGINACEAE) BASED ON MULTIVARIATE ANALYSES OF GEOGRAPHIC VARIATION MATTHEW MAHRT and RICHARD SPELLENBERG Department of Biology New Mexico State University Las Cruces, NM 88003-0001, U.S.A. ABSTRACT The two species of Cyphomeris (Nyctaginaceae) occur in arid regions of central southern orth America. ee = eee (Mart. & Gal.) Standl., the more widespread, consisted of two subspec . crassifolia (Standl.) Standl., is restricted to the eastern margin of the range 7 ; enus. iE survey of herbarium specimens reveals consid- erable variation particularly in fruit characteristics, leaf shape, and foliar aan and some intergradation in these features among Ls aes ons. We used 114 specimens and 29 population samples to examine relationships of 14 vegetative and 10 fruit ee Cluster analysis, principal pienaee analysis, aero analysis, and multivariate analysis of variance indicate the presence of two morphological groups that correspond roughly to the species-level taxa of previous authors, groups that can be separated by morphology, geog- raphy, and ecology. These two species are maintained. The sole infraspecific taxon ae vat. oe I. ohnst., is noted to be an extreme and eoeapicdly re- ted form intergradient wen nearby ee and is not eae taxonomically. RESUMEN La variacién geogrdfica de macromorfologia de Cyphomeris, un género pequefio de Nyctaginaceae de zonas dridas de la parte sur-centro de Norte América, fue examinada mediante el uso de métodos estadfsticos multivariantes. Autores anteriores reconocieron O ecias, una [C. a ae (Mart. & Gal.) Standl.} con dos cee y lao crassifolia Standl.} restringida a la parte este del drea del género. Usamos 114 muestras y ejemplares de 29 poblaciones para revisar las afinidades de 14 caracteres vegetativos y 10 caracteres de los frutos. Los datos fueron analizados por andlisis de agrupacién, andlisis de componentes oo. andlisis discriminante, y andlisis de multivariante. Los resultados indican la cia de dos grupos m ae que se corresponden el eee con los taxa anne el de especie de los autores anteriores; que pueden ser separadas por su morfologia, geografia, y ecologia. Unicamente no se reconoce el taxon scl csadiacd. C: Tea var stewartii I. M. Johnston, pero es considerado como una forma extrema, termedia con poblaciones mas 0 menos aisladas en las montafias proximas. INTRODUCTION Cyphomeris Standl. (Nyctaginaceae) is a small genus of perennial herbs distributed in arid and semi-arid regions of northern Mexico and the south- western United States. It has been considered a subgenus of Boerhavia L. Stipa 16(4): 679-697. 1995 680 Sipa 16(4) (Fosberg 1978). Standley (1911) and Reed (1970) recognized two species, C. gypsophiloides (Mart. & Gal.) Standl. and C. crassifolia (Standl.) Standl., distinguished by leaf shape, leaf pubescence, and features of the fruit, each having a geographic range more or less exclusive of the other. I. M. Johnston (1944) described a third taxon, C. gypsophiloides var. stewartii, from La Sierra el Diablo in southeastern Chihuahua, noting its robust habit and glandular hairs. He refrained from describing it at specific rank because he thought collections of C. gypsophiloides var. gypsophiloides from extreme southwestern Coahuila were transitional to it. The two species of Cyphomeris are not completely distinguishable mor- phologically. There is a wide range of variation (Fig. 1) within and between species, some apparently more or less continuous, other clearly discontinu- ous. The geographic limitation of the genus to an ecologically definable region and fragmentation of populations according to available habitat al- lows an assessment of relationships of morphological variation to geogra- phy and topography throughout the range of the genus. These results were used to examine the significance of the characters used by previous authors in the delimitation of species of Cyphomeris, and to aid in our own tax- onomy of the genus. CHARACTERISTICS OF POPULATIONS AND POLLINATION Throughout much of its range, and particularly west of the Sierra Madre Oriental, Cyphomeris often exists in small groups of 20-30 individuals con- fined to rocky soils of road cuts, washes, and other rough terrain, generally separated from other such groups by large distances. Thus, the overall popu- lation is highly subdivided and the majority of outcrossing probably oc- curs between genetically related individuals. In addition, moderate to high levels of self fertilization also occur. Monitoring of a natural population of Cyphomeris in southern New Mexico indicates that plants produce fruits from all or mostly cleistogamous flowers during the dry months (typically May-July) of the growing season (Fig. 2). After the onset of summer rains (typically July-September) plants produce mostly or all chasmogamous flowers. In each of two growing seasons mean percentage of cleistogamous flowers began at 100%, their relative frequency decreasing as the season progressed, corresponding to an increase in precipitation. Pooled data (across years) reveals a significant correlation (r = -0.74, P < 0.05) between the mean percentage of cleistogamous flowers and cumulative rainfall. Chasmogamous flowers, about a centimeter across and varying from deep to pale pink or rarely nearly white, are visited by small bees, flies, and butterflies during early morning hours. Insects tend to move from flower to flower on the same plant, afterward moving to a nearby plant where the pattern of intraplant visitation often is repeated. Whether or not a flower is MAHRT AND SPELLENBERG, Taxonomy of Cyphomeris 681 Qem M 2 cm op) Fic. 1. Leaves and anthocarps of Cyphomeris. cae in n circles show abaxial leaf surface at 40. Sample numbers are given in parenthesis and as te drawings with vouchers listed in ee A and localities on map (Fig. 4); (a-c) C. crassifolia (17), (d-f) intergrade from wes- ern Texas (8), (g-i) linear-leaved example of C. gypsophiloides (24), (j-l) C. gypsophiloides var. stewartii (25), (m-n) intergrade from central San Luis Potosi (20), (p-r) C. Lanes (29). 682 Sipa 16(4) wy 100 + ES eee ee a gece Sl tia 5 7% A e) = | s&s S20 z : 80 4 d | tO) H- 2545 = 2 ad < § B 2 . ry see - i 6075 =< si we 6) a | F200 72) w 40 & 2 OY 4A aS BO a Hise 8 2 nt 20 Se x 4 y z ) a |_| eras 4 WwW = ! q T T T T T T oO O Oo Oo a a a re Kb =) D5 = a) w Ww Ww 9 (3) < < < Lt 2) 7) 2) e) to) - = = = ° ie) 9 ° a - “ o — ~N o a a DATE i 2. Plots ae the relationship between the mean percentage of flowers that are s (squares) and cumulative rainfall (triangles) in a population of ie opvphinds nie along the W slope of the Organ Mountains, approximately 16 km E Las Cruces, New Mexico. visited , the stamens of chasmogamous flowers curl over the style and stigma by late morning. This process has been described for similar flowers in the nyctage genera Acleisanthes and Ammocodon (Spellenberg and Delson 1977), Boerhavia (Chaturvedi 1989), and Mirabilis (Cruden 1973; Hernandez 1990). It results in self-pollination and seed set in all species examined in these genera that have cleistogamous flowers. We have no reason to suspect oth- erwise for Cyphomeris. Such observations suggest that selfing often constitutes a major mode of reproduction in Cyphomeris, particularly in dry years or areas of low annual rainfall. As a result, the level of inbreeding is likely higher than that im- plied by population substructure, a situation conducive to rapid and local differentiation by the combined action of genetic drift and selection (Wright 1938, 1940, 1943). MATERIALS AND METHODS FOR MULTIVARIATE ANALYSES Data for numerical analyses were obtained from two sources. The first consists of 297 collections from 15 herbaria, including all holotypes. Of these specimens, 114 represented the genus well geographically, possessed complete location data, and exhibited all characters selected for analyses. Only these specimens (OTU’s) were measured (Mahrt 1993). In addition to the herbarium specimens, 29 population samples were collected from most of the range of the genus (Appendix A). Each collec- tion represents a distinct locality and consists of 6-10 branches, each bro- ken at the base of a different plant. Vouchers are at NMC, with duplicates MAHRT AND SPELLENBERG, Taxonomy of Cyphomeris 683 variously distributed to ARIZ, ASU, BRIT, CIIDIR, ENCB, F, IBUG, IEB, MEXU, MO, MT, NMC, NY, SLPM, TEX, UC, US (Holmgren et al. 1990). Data were taken on 14 vegetative and 10 fruit characters (Appendix B). Characters included those considered diagnostic by previous authors as well as those known to exhibit geographic variation. Although statistical prob- lems are known to occur with their use (Phillips 1983), two ratios (MLR and ULR) were included since they provided the best measure of impor- tant aspects of leaf shape. Each character was measured an average of three times at different locations on each specimen and a set of character means was computed for each specimen. These were used for all numerical analy- ses. Character measurements were confined to mature organs that did not represent infrequent extremes for a specimen. Characters were measured with an ocular rule of 100 units in a dissecting microscope at 10-40. Hair densities (HD and GHD) were the number of hairs occurring in 1.5 x 2 mm rectangle. Data were not transformed since a preliminary analysis of residuals revealed reasonably normal distributions for all characters (in- cluding ratios and meristic characters). OTUs were first assigned to one of three a priori categories based on Reed’s (1970) taxonomy: 1) C. gypsophiloides; 2) C. crassifolia, and 3) inter- grade. Three OTUs of C. gypsophiloides var. stewartii, a taxon outside the range of Reed’s treatment, were assigned to category one. We use these categories throughout the following analysis. Data from the 114 herbarium specimens were analyzed in three steps. First we examined relationships among OTUs by centroid hierarchical clus- ter analysis (CL) of all 24 characters. Next we employed a principal compo- nents analysis (PCA) of the character correlation matrix. Finally, we per- formed an a@ posteriori multivariate analysis of variance (MANOVA) on selected phenetic assemblages (exposed by CL) to identify their distinguish- ing characters. Data from the 29 populations were analyzed initially with a procedure outlined by Zimmerman and Ludwig (1975). This procedure employs dis- criminant analysis (DA) to detect clusters of related sample populations by identifying pairs for which the Mahalanobis distance (D’) is non-signifi- cant. For each pairwise comparison between populations, DA constructs a discriminant function, calculates D*, and computes an F statistic that tests if D? is significantly greater than zero. Population localities then are plot- ted on a map and pairs for which D° is non-significant are connected with lines. Aggregates of associated populations can then be visualized. Complimentary to DA, a MANOVA was conducted, followed by a mul- tivariate “means separation” technique developed by Smith et al. (1993). In this technique, the population multivariate sample means first are stan- 684 Sipa 16(A) dardized and subjected to a CL. Then a “group” variable is created, each group now treated as a single sample unit. Next, a MANOVA is run to determine if there are significant differences between groups. If differences exist, the CL is consulted and the closest groups (those joined first) are combined into a super group. A nested MANOVA with super groups, and groups within super groups, is then conducted and the significance levels of the resulting tests compared. The test for super groups should indicate significant differences whereas the test for groups within super groups may be either significant or non-significant. If the test is significant, then that is an indication the joining of the most recently combined groups was inap- propriate. If the test is not significant, then the super groupings to this point are reasonable and one may proceed to the next joining as indicated by the CL. The end result of this process is the formation of groups of sample populations whose members are homogeneous, but for which there are significant differences between groups. Analyses were performed using the Statistical Analysis System (SAS) on the mainframe computer of New Mexico State University. Cluster analysis, PCA, DA, and MANOVA were conducted using the CLUSTER, PRINCOMP, DISCRIM, and GLM procedures respectively (SAS Institute 1990). Standardization of sample population means was accomplished us- ing IML (SAS Institute 1989) RESULTS The CL of herbarium Specinens; using 24 characters, resulted in three major groups of OTUs (Fig. 3, branches A, B, and C). Clusters were arbi- trarily defined at approximately 0. 90 on the distance scale. One minor cluster (branch D) and four ungrouped specimens are also present at this level. Branch B constitutes a nearly homogenous collection of OTUs of C. crassifolia. Branch A contains the majority of C. gypsophiloides specimens, but with a mixture of @ prior? classified OTUs from C. crassifolia and inter- grades. Branch A-1 is mostly OTUs of C. gypsophiloides from northern and western areas (Fig. 4, black stars). Branch A-2 is mostly OTUs of inter- grades and C. crassifolia that are largely southern and eastern in distribu- tion (Fig. 4, open stars). Branch C consists of seven OTUs of linear-leaved forms of C. gypsophiloides. Of the three OTUs assigned to C. gypsophiloides var. stewarti?, two from La Sierra el Diablo in southeastern Chihuahua) re- main distinct, whereas the third, from the nearby La Sierra el Rosario in northeastern Durango, is grouped with the majority of specimens repre- senting C. gypsophiloides (branch A-1) A dendrogram (Fig. 5) from the CL of standardized population means reveals a pattern similar to that of the CL of herbarium specimens (Fig. 3). Two major clusters and six distinct populations are defined at 0.60 on the SN ice wi MaHRT AND SPELLENBERG, Taxonomy of Cyphomeris 1(1) | 1(1)* Cc 1(1) 1(1) 1(3) 1(1) 1) 12), 3(21), 1(5) 1 1), 1(3), 1(4)" 5 4), 1(12) 1 (SNSWI0SdS 4O YASWNN) dNOYD IIHdVWHDOAD EUCLIDIAN DISTANCE Fic. 3. Partial dendrogram from the cluster analysis of herbarium specimens. Values jee the horizontal axis signify the normalized euclidian distance between cluster centroi AS ee 1989). N Numbers at the tip of each branch indicate the composition of te specimens contained therein. The first digit refers to @ prior? group (1 = C. ele 2 = C. crassifolia, 3 = eee The value in parentheses gives the number of specimen belonging to that group on that branch. Clusters discussed in the text are identified d by capital letters and arabic numerals near the base of each cluster. An asterisk at the end of a branch signifies an OTU identified as C. gypsophiloides var. stewartii. — distance scale. Branch A has two subgroups: A-1 has four C. gypsophiloides samples, all northern; A-2) has 14 intergrade populations and a single sample representing C. gypsophiloides var. stewartii. Four samples of C. crassifolia comprise branch B. Three unclustered populations of linear-leaved C. eypsophiloides form branch C at 0.90 on the distance scale, implying a loose but explicit morphological relationship. The remaining samples, one each of C. eypsophiloides, C. crassifolia, and C. gypsophiloides var. stewartit, join after the major branches A, B, and C have combined. Thus, they have little mor- phological affinity among themselves or with any of the other populations. Results from PCA are summarized in Table 1. The first three principal components account for 58.6% of the total sample variance (Component 1| 686 Sipa 16(4) 1G. 4. Map showing the geographic range of Cyphomerzs and the results of DA. Small symbols represent the locations of selected herbarium specimens and the a priori taxo- nomic group to which they were assigned (see methods). BLACK STARS designate C. gypsophilordes. BLACK SQUARES denote C. crassifolia. OPEN STARS indicate plants that are more or less intermediate, particularly in leaf width and leaf pubescence. In the north these may be intergrades; in the south they may represent ancestral types. OPEN SQUARES indicate linear-leaved forms of C. gypsophiloides. Large BLACK CIR tions of sample populations (Appendix A). Pairs of localities connected by a solid line have values of D° that are nonsignificant at P = 0.05. Those connected by a dotted line have values that are significant at P = 0.05 but not at P = 0.01 (see methods). MaurT AND SPELLENBERG, Taxonomy of Cyphomeris 687 nN - oOo —_ . © a Oo _- >a —_ a i Oo | oO o ae) o =) A Lo ine) = © 99(2) — 98(2) 97(2) 2d 92(3) 104(3) 105(3) 103(3) A 101(3) 102(3) 96(3) 95(3) 93(3) 87(3) T T T T I T 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 EUCLIDIAN DISTANCE Fic. 5. Dendrogram from the cluster analysis of population samples. Values along the horizontal axis signify the normalized euclidian distance between cluster centroids (SAS Institute 1989). Population localities are identified by the number at the tip of each branch (Appendix A). The values in parentheses indicate the a priori group (see caption, Fig. 3) to which each sample was assigned. Clusters discussed in the text are referred to by capital letters and arabic numerals near the base of each cluster. An asterisk at the end of a branch signifies an OTU identified as C. gypsophiloides var. stewartit. (dNOYD DIHdVYDOAD) AIdWVS NOILVINdOd [o) accounting for 31.5% of the total). The remaining 21 each contribute 7% or less. Eleven characters (ULW, UND, UDP, MLW, HTW, MLR, ULR, PLB, PLW, and HD [Appendix B}), summarizing leaf shape, leaf pubes- cence, and wartiness of the fruit, load more or less equally on Component 1. Six characters (SLL, PLL, FL, PLW, ULP, and MLP) have high loadings on Component 2. The first four describe overall fruit length; the last two measure petiole length. Aspects of leaf length and width (ULL, MLL, MLPR, ULPR, and MLW ) have high loadings on Component 3. 688 Sipa 16(4) Tape 1. Character loadings for the first three principal ca The proportion of the coral variance ae for by each component is given in parenthe Loadings Character Component | (31.5%) Component 2 (15.1%) Component 3 (12.1%) M -0.2 0.09 0.42 *MLW 0.27 0.14 0.24 0.21 0.26 0.16 MLPR -0.19 0.12 0.37 *UND 0.29 -0.04 0.13 *UDP 0.28 0.01 0.16 U -0.17 0.14 0.43 *ULW 0.29 0.12 0.17 0.22 0.28 0.07 ULPR -0.14 0.24 0.36 *H 0.22 -0.16 0.01 GHD 0.1 -O.11 0.08 -0.14 0.35 -O.15 FDA 0.13 0.15 0.1 F -0.09 -O.17 0.07 *PLL -0.09 0.35 -0.17 *PLW -0.23 0.28 -0.12 *PLB 0.23 0.18 0.06 SLL 0.06 0.37 -0.21 SLW -0.08 -0.12 SLB 0.18 0.09 -0.1 FHT W 0.27 -0.04 0.12 *MLR -0.25 -0.21 0.13 *ULR -0.26 -0.2 0.16 * Denotes character discussed in the text (see Materials and Methods). A plot of OTUs of components | and 2 reveals three fairly well defined groups (Fig. 6) primarily separated by Component 1; characters UND, UDP, MLW, ULW, HTW, HD, and PLB contribute most. OTUs of C. crassifolia form a small cluster to the right of a larger cluster of C. gypsopbiloides and intergrade OTUS. In comparison to C. gypsophiloides OTUs of C. crassifolia have leaves that are shorter, wider, more undulate, and more pubescent, and fruits that are shorter and distinctly warty, characters used by Standley (1911) to distinguish the species. OTUs of intergrade and C. gypsophiloides var. stewartit overlap extensively with OTUs of C. gypsophiloides but mostly are distinct from those of C. crassifolia. The third cluster (lower left corner of Fig. 6) exhibits a high degree of separation on both components. This group consists of C. gypsophiloides OTUs that separate as a result of high values for MLR and ULR combined with low values for MLP and ULP. Little else distinguishes these plants, which occur in widely scattered loca- tions throughout the central Chihuahuan Desert (Fig. 4, open squares). A single outlier, an OTU from the Guadalupe Mountains of western Texas, MAHRT AND SPELLENBERG, Taxonomy of Cyphomeris 689 6 = Oo N 4 -— LJ a LJ O a ° i eR 2 = 2 Eh G AA a. i O s |_| CO O 0 ;— Bo 5 & AM Oo CS cu an mn a oH & A A a -2 ;_ g 4o De - ms oO oF q : RA = Oo E+ oe 6 ca oe | Lrerrtirpiry Ly -10 5 0 5 PRINCIPAL COMPONENT 1 Fic. 6. Plot of specimen OTUs on principal components one and two. Symbols indic the a ae taxonomic group to which each OTU was assigned (squares = C. eer triangle '. crassifolia, citcles = intergrades). Specimens representing C. gypsophiloides var. sfewartii are indicated by large squares. Type specimens (or, in the case of C. gypsophiloides, a specimen collected from near the type locality) are indicated by black figures. (far left, Fig. 6), has high values of MLR, ULR, FL, PLL, and PLW. It may represent an aberrant individual. It had little affect on the overall results of CA an and was not removed from the analysis. A similar pattern is produced by a plot of OTUs of components | and 3 (Mahrt 1993). Cyphomeris crassifolia and C. gypsophiloides OTUs form more or less discrete clusters whereas those of intergrades and C. gypsophiloides var. stewartii overlap extensively with OTUs of C. gypsophiloides. A plot of Us on components 2 and 3, which primarily measure aspects related to overall size (fruit size for Component 2, leaf size for Component 3), shows little to no separation of @ prior? taxonomic groups (Mahrt 1993) except for two OTUs of C. gypsophiloides var. stewartii. These show strong separation on Component 3, a reflection of their robust habit and large leaves, attri- butes noted by Johnston (1944) in the original description of the variety. Values of D° (most are highly significant, P < 0.01) for each pair of sam- ple populations are given in Mahrt (1993). Samples joined in Fig. 4 with a solid line have values of D° of P = 0.05 or greater. Those joined with a dotted line are significant at P = 0.05 but not at P = 0.01 (Zimmerman and Ludwig 1975). Five groups of interconnected populations are apparent. 690 Sipa 16(4) Tas_e 2. Results of multivariate means separation (see text for explanation of the testing procedure). Siraples joined at each step are listed in Appendix A. Significance levels are for the test of popula- tions within groups (Wilks’ Lambda). Group Samples Combined Prob. > F (Wilks’ Lambda) I 20 with 22 0.64 2 15 with 17 0.61 3 1 with 29 0.7 I 12 with 13 0.36 > 16 with group 2 0.045 1* * Indicates step at which the algorichm was stopped. The first (samples 1, 2, and 29) consists of three C. gypsophiloides popula- tions from New Mexico and Texas. The second (15, 16, and 17) constitutes three populations of C. crassifolia from the states of Tamaulipas and Nuevo Leon. The remaining groups are more or less central in distribution and consist entirely of intergrade populations (5, 7, 8, 10, 12, 13, 18, 19, 20, 21, and 22). Populations of C. gypsophiloides var. stewartii (24 and 28) are distinct from all others in the analysis, including each other. Ten addi- tional population samples are also individually distinct. Five of these (3, 23, 25, 26, and 27) are assigned to C. gypsophiloides. Three (4, 6, and 9) are intergrades. The remainder (11 and 14) represent C. crassifolia. Results from the MANOVA indicated significant differences between the centroids of the 29 sample populations (Wilks’ Lambda, P = .001). The means separation technique resulted in four groups (Table 2). These repre- sent a subset of the results from DA. Groups one and four each consist of two intergrade sample populations. Group two is composed of two C. crassifolia sample populations. Group three results from the combination of two C. gypsophiloides samples. DISCUSSION Results from the CL of population means and the PCA of herbarium specimens indicate that morphological variation in Cyphomeris is mostly discontinuous. Two geographically and morphologically defined groups exist, corresponding roughly to the species recognized by Reed (1970) and Standley (1911). The first group comprises C. crassifolia from east of the Sierra Madre Oriental, plants characterized by short, warty fruits and short, broad, undulate, moderately to densely pubescent leaves (Fig. 1, a-c). Plants of C. gypsophiloides and those designated as intergrades between the two species comprise the second group, an assemblage ranging from southern New Mexico in the United States southward (west of the Sierra Madre Oriental) to the Mexican state of Puebla. They have long, striate fruits and long, narrow, plane to slightly undulate, glabrous to sparsely pubescent leaves. This group can (more or less) be divided into two subgroups. Plants MauHRT AND SPELLENBERG, Taxonomy of Cyphomeris 691 from the northern portion of the range (Fig.1, p—r) usually have glabrous, plane, lanceolate leaves and striate fruits. Intergrade plants (Fig. 1, d-f and j-n) are southern and eastern in distribution, and have sparsely pubescent, lanceolate, plane to slightly undulate leaves and striate to weakly warty fruits. The two forms intermix across western Texas in a more or less clinal fashion. A third group, consisting of Cyphomeris gypsophiloides plants with sessile leaves that have high length to width ratios (Fig. 1, g—1), has no strong geographical unity (Fig. 3, branch C, Fig. 5, branch C). Such plants occur in the widely scattered mountain ranges of western Coahuila, eastern Chi- huahua, northeast Durango, and the Big Bend region of Texas. The CL of herbarium specimens illustrates the problem of the presence of conspicuous variation and the construction of a useful classification in the genus. Twelve individual Cyphomeris crassifolia cluster with C. eypsophiloides and intergrades (Fig. 3, branch A-2). These plants have large leaves with less pubescence than those of the discrete C. crassifolia cluster. Observations of C. crassifolia populations in the field suggest certain char- acters are affected by immediate environmental conditions. Shaded plants tend to have larger, longer, less pubescent leaves with larger length to width ratios than those growing in open sunlight, perhaps resulting in the dis- placement of C. crassifolia specimens by CL. Supporting evidence comes from two sources. First, a CL using only those characters found to be most diagnostic by PCA fully separates the C. crassifolia specimens from the C. gypsophiloides-intergrade complex, although each group of C. crassifolia speci- mens forms a separate cluster. Second, no displacement occurs in the CL of sample population means, which measures the central tendencies of the respective Popurenons. When these central tendencies are clustered, as opposed to the characteristics of individual specimens, C. crassifolia forms a homogeneous group discrete from both C. gypsophiloides and intergrades. If the displacement described above is discounted, the CL of herbarium speci- mens reveals a pattern similar to that from the CL of population samples. The results from DA and the means separation algorithm show that individual populations of Cyphomeris are highly differentiated, most likely due to the effects of selection superimposed on genetic drift resulting from inbreeding and population subdivision, but degree of differentiation varies within the range of the genus (Fig. 4). Populations in the south and east have a higher number of interpopulation connections than those to the north and west. Field observations indicate that populations of Cyphomeris in the south and east are less subdivided. In general, they are also character- ized by more frequent and dependable rainfall. Occasionally very different morphs will be found in close proximity. Samples 24 and 25 (Fig. 4) from the Sierra el Diablo, Chihuahua (Appen- 692 Sipa 16(4) dix A), occur within | km of one another and occupy similar habitats (sample 24 comes from slopes with a more southern aspect). Sample 24 has narrow, linear, glabrous leaves (Fig. 1, g—-i). In contrast, sample 25 is classic C. oo var. stewartit, a broad, lanceolate, moderately pubescent leaves (Fig. 1, j-1). An extensive search revealed no intergradient plants. Despite numerous instances of local differentiation, populations of Cyphomeris form broadly similar pone groups with at least some geographic unity, as evidenced by the results of both CL and PCA. Species circumscriptions based on such assemblages are subjective and in general do not represent “real” units in terms of evolution (Ehrlich and Raven 1969). However, they often represent “real” units in terms of geography and ecol- ogy, and as such provide the basis for a useful taxonomy. In the case of Cyphomeris the most utilitarian approach seems to be division of the genus into two species. Cyphomerts crassifolia comprises an entity distinct from the complex formed by C. gypsophiloides and the intergrades. The two groups can be separated morphologically by a combination of characters, specifi- cally leaf shape, leaf pubescence, and length and wartiness of the fruit. In addition, the two groups can also be distinguished ecologically. Cyphomeris crassifolia is more or less confined to relatively mesic areas east of the Sierra Madre Oriental, and occurs in a variety of settings, from farm fields to subtropical scrub. Densities within a given population are relatively high and the apparent degree of population subdivision relatively low. In con- trast, members of the C. gypsophiloides-intergrade complex inhabit dry, rocky on sites, have low population densities, and a high degree of population sub- division, Within C. gypsophiloides influence of C. crassifolia decreases northward and westward across southern Texas in a more or less clinal fashion where suitable habitat for Cyphomeris is comparatively undissected but becomes progressively drier. This pattern is particularly evident along southward draining river systems such as the Devil’s River. Specimens from the drier limestone rangelands to the north are very much like typical C. gypsophiloides, whereas those from the brushy areas at the confluence with the Rio Grande resemble C. crassifolia. \n general, however, these plants show more affinity to C. gypsophiloides. Corresponding with the abrupt habitat changes associ- ated with the Sierra Madre Oriental to the south the two species are rela- tively sharply delimited. 3 and Fig. 5, branches A-1 and A-2) suggests that infraspecific taxa might be recog- The existence of subgroups within C. gypsophiloides (Fig. nized. It is not clear, however, whether intergradient plants result from of primary intergradation or secondary contact between northern representa- tives of C. gypsophiloides and C. crassifolia, or both. Widely accepted hy- potheses of the northward expansion of the Chihuahuan Desert and the Maur AND SPELLENBERG, Taxonomy of Cyphomeris 693 evolution of the Mexican flora suggest that the southern intermediate types (from San Luis Potosi south to Puebla) represent an ancestral form that gave rise to the northern, more or less glabrous, component of C. gypsophiloides and the eastern, very pubescent C. crassifolia. The latter two groups may have then met secondarily to produce the northern intergrades found from Coahuila north to western Texas. Some evidence exists for this. A single population (sample 96) that appears to represent a hybrid swarm has been documented from central Nuevo Leon. Plants in this population exhibit almost the entire range of leaf variation, from plane, glabrous, lanceolate leaves to examples that are undulate, pubescent, and rhombic. The collec- tion Spellenberg, Zimmerman, and Zucker 8374, MEXICO. Coahuila: Cuesta de la Muralla, N of Saltillo, 11 Oct 1985, (NMC) is even more suggestive, combining the leaf characters of C. gypsophiloides (plane, glabrous, lanceolate leaves) with the short, warty fruits of C. crassifolia. Until a more thorough understanding is gained of both the nature and extent of apparent hybrid- ization, and also of local differentiation that results in such extreme leaf types as the linear glabrous leaves in many scattered populations or the large pubescent leaves of robust plants in the local phased named as the var. stewartii, we do not recommend formal recognition of infraspecific taxa. TAXONOMY Cyphomeris Standl., Contr. U.S. Nat. Herb. 13:428. 1911. Lindenia Mart. & Gal., Bull. Acad. Sci. Brux. 10:358. 1843. es as & Gal., Bull. Acad. Sci. Baise X1:240. 1844. Senkenbergia Schauer, Linnea 19:7 1847 Perennial herbs with woody taproots, glabrous to a pubescent with pale, curved, bristle-like to capitate glandular hairs. Stems divaricately branched, erect, ascending, trailing or clambering through shrubs, to 2m, often brittle, upper internodes much longer than the lower ones, each in- ternode with a glutinous band near the midpoint. Leaves opposite, those at a node unequal, exstipulate, petiolate or rarely sessile, weakly succulent; blades linear to lanceolate, ovate, broadly oblong, or rhombic, entire or undulate. Inflorescences axillary or terminal racemes; pedicels subtended by a lanceolate, sparsely ciliate, caducous bract 2.5—6 mm long; pedicels 0.5—1 mm. Flowers chasmogamous or cleistogamous. Perianth of chasmogamous flowers pale pink to red-violet (rarely greenish-white), broadly funnelform from a short, weakly curved tube constricted above the ovary; limb slightly oblique, 5 lobed, lobes emarginate. Stamens 5, exserted. Pistil unicarpe — late; style filiform, exserted beyond the anthers; stigma capitate. Anthocarp stipi- tate, clavate, gibbous on the dorsal side, generally concave ventrally, pen- dant or refracted, longitudinally striate with 10 fine striations and/or series of elongate interrupted ridges, exuding mucilage when wetted from con- spicuous wart-like protuberances 694 Sipa 16(4) In its racemose inflorescence and clavate fruits Cyphomeris resembles some species of Boerhavia and has been retained in that genus by FE. R. Fosberg (1978) Type SPECIES: Cyphomeris gypsophiloides (Mart. & Gal.) Stand. KEY TO THE SPECIES OF CYPHOMERIS . Leaves glabrous or sparsely pubescent, blades lanceolate to linear, less than one half as wide as din enure to shallowly undulate; anthocarps 8-11 1 striate to weakly warty . Leaves pubescent, “blad one half as oF ti s ophiloides es ovate, broadly oblong, or rhombic, greater a wide as long, undulate; anthocarps 6—8 mm, prominently warty, especially on dorsal side C. crassifolia 1, Cyphomeris a (Mart. & Gal.) Standl., Contr. U.S. Natl. Herb. 13:428. 1911. oa >: MEXICO. Oaxaca: Plains near Tehuacan, Galeotti 577 (HOLOTYPE: BR)). Lindenia ea Mart. & Gal., Bull. ear Sci. 10:358. 1843. Tinantina ae: oides (Mart. & Gal.) Mart. & Gal., Bull. Acad. Sci. Brux. 11:240. 1844. Senkenbergia mips (Mart. & Gal.) Ben me & Hook., Gen. Pl. 3:6. 1880. Boerhavia gypsophiloides (Mart. & Gal.) Coulter, Contr. U.S. Natl. Herb. 2:354, 1894. Senkenbergia annulata as Linnea 19:711. 1847. Type: Mexico: location unknown, Aschenborn exs. #253 (not en). Boerhavia gibbosa Pavon ex C hoisy in DC. Prodr. 13:457. 1849, nom. non leg. Plants glabrous to sparsely pubescent with pale curved, occasionally gland-tipped hairs primarily along the stems and the veins of leaves. Sven erect to ascending, 0.5—1.5 m. Leaves broadly oblong-lanceolate to nar- rowly lanceolate or rarely, linear (and then sub-sessile), less than one half as wide as long, the distal blades much broader than the proximal (except in linear leaved forms), proximal blades 1-9 cm « 1-30 mm, entire to shal- lowly undulate. Perianth light pink to red violet, 7-10 mm above the ovary. Anthaarp 8-1 1(-14) mm, striate to weakly warty, usually gibbous. Distribution.—Southern New Mexico and western Texas in United States south to the Mexican states of Puebla and Oaxaca; rocky soils of washes slopes, and roadsides, common on limestone, desertic scrub to open and dry pine-oak woodland; 500-2500 m. Blooming from mid-April to early November, but mostly mid-summer to early fall. Highly variable (Fig. d-f, g-i, m-o, jen l, The holotype of C. gypsophiloides is fragmentary, consisting only of a leafless branch and several semi-mature fruits, making most of its original charac- ter states impossible to determine. The collection Chiang et. al. 2265 (MEXICO. Puebla: Plateau of San Lorenzo, near Tehuacan, 7 Aug 1981 [TEX}]) comes from very near the type locality. It has the elliptic-lanceol slightly pubescent leaves, and moderately warty fruits that are cl tic of southern intermediate plants (Fig. 1, m-n). ate, laracteris- MAHRT AND SPELLENBERG, Taxonomy of Cyphomeris 695 2 ela ack crassifolia (Standl.) acten Contr. U.S. Natl. Herb.13:428. LOL, MEXICO. Sopouiay Saltillo and vicinity, Palmer 172 (HOLOTYPE: UCI; ISOTYPES: F c MO!, NY!, US!). oe crassifolia Standl., Contr. U.S. Natl. erb. 12:373. 19 Plants finely pubescent with pale, curved hairs (rarely capitate glandu- lar). Stems ascending to trailing, 0.5—2.0 m. Leaves rhombic to ovate or broadly oblong, one half as wide as long or wider, the distal blades nearly as broad as the proximal ones, proximal blades 1-5 cm x 840 mm, undulate and coarsely sinuate lobed or rarely more or less entire. Perianth pale pink (rarely greenish-white) to deep pink or red-violet, 6-8 mm above the ovary. Anthocarp 6-8(-11) mm, usually strongly gibbous, prominently warty, es- pecially on dorsal side. Distribution. —Extreme southern Texas in the United States and parts of Nuevo Leon and Tamaulipas in Mexico, primarily east of the Sierra Madre Oriental; fine or rocky soils on flats, washes, slopes, and roadsides, desertic scrub, more common in fallow fields and in semiarid or subtropical scrub; 1200 to 2500 m. Blooming mid-April to mid-December, but mostly late summer to late fall. Compared to C. gypsophiloides, C. crassifolia comprises a more cohesive, well defined morphological unit (Fig. 1, a-c) occurring in rela- tively mesic habitats and often having high population densities. It is likely that this species hybridizes with C. gypsophiloides (see discussion, preced- ing). As a result, populations exhibiting various degrees of intermediacy can be expected, particularly along the northern edge of the Sierra Madre Oriental where the geographic barriers to gene flow are less pronounced. In southern Texas, at the northern limit of the species’ range, C. crassifolia is known from only two collections; Starr Co., NW of Roma, 17 Dec 1963, Wood 723 (TEX) and between Sullivan City and Rio Grande, 13 Apr 1941, Runyon 2547 (TEX US). APPENDIX A t of population see Bold numbers at the beginning of each citation indicate the sample “indies (Fig. 5). Italicized numbers following each citation and adjacent to parentheses refer to the 4 priori group to which = sample was assigned (1 = C. gypsopbiloides, 2 = C. eee 3 = intergrade). Samples representing C. i al var. stewartit are marked with an asterisk. Numbers in paren- theses indicate the sample size. Unless otherwise noted all collections are those Mahrt and various associates. Vouchers are Sere at NMC . Duplicates are widely distributed. “ATES, New Mexico. Dofia Ana Co.: W slope of Organ Mtns, 17.6 air km E of Las Cruces, Slee 5953, 1(10). 1—Socorro Co.: White Sands Missile Range, San Andres Mtns., 60 km NE of Truth or Consequences, 29 Jul 1990, 84 1(9). 6 Texas. Crockett Co.: 6.4 km SE of eee a ie 1990, 89 3(10). 2—El Paso Co.: Franklin Mtns., 0.4 see E of Trans Mountain Road 1990, 85 1(10). 3—Hudspeth Co.: Guadalupe Mtns., 8 km SW of Pine Springs, 12 Oct 1990, ae Hao) 27—Jeff Davis Co.: 16 km SE of Fort Davis, 21 Oct ee 110 1(10). 4—Pecos Co.: 22.4 km E soe ed 12 Oct 1990, 87 3 (10); 5—89.6 km E of Fort oe 12 Oct mee 88 3(9). seen Co: 4.8 km § of Shafter, 21 — 1990, 109 1(6). 7—Val Verde Co.: 88 km NW of Del Rio, 13 Oct 1990, 90 3(9); 8—28.8 km of Comstock, 13 Oct 1990, 9/ can se km NW of Del Rio, 13 Oct 1990, 92 1110). 696 Sipa 16(4) 24—MEXICO. Chihuahua: Sierra el Diablo, 80 km E of Jimenez, 19 Oct 1990, 107 1(10); 25— Sierra el Diablo, 80 km E of Jimenez, 20 se nee 108 1*(10). 10—Coahuila: Sierra la Gavia, 33 km S of Monclova, 14 Oct on 93 3(10); 11—27.2 N of Saltillo, 14 Oct 1990, 94 2(10). 23— Durango: 17.6 km S of Cuencame, 18 ae oo 106 1(10); 28—N end of Sierra el Rosario, 20 km SW of Mapimi, oe ane 1*(10). 12—Nuevo Leon: ae Madre Oriental, 16.6 km W of Irurbide, 15 Oct 1990, 95 3(10); 13—Sierra Madre Oriental, 14.4 km W of ead 15 Oct 1990, 96 3(10); are km S of oo 15 Oct 1990, ee he 15—14.4 km §S of Linares, 15 Oct 1990, 98 2(10). 20—San Luis Porosi: 110 km NE of San ss Potosi, : Oct 1990, 103 3(9); 21—38 km NE of San Luis Potosi, 16 Oct 1990, 104 3(10); 22—33.6 km W of San Luis Potosi, 17 Oct 1990, 105 3(10). 16 Beton 9.6 km N of El Barretal, 150 1990, 99 2110); 17—E slope of Sierra Madre Oriental, 9.6 km SW of Ciudad Victoria, 16 Oct ce 100 2(10); 18—W slope of Sierra Grande, Sierra Madre One 27.2 km NE of Jaumave, 16 Oct 1990, 107 3(10); 19—Sierra Madre Oriental, 16 km SW of Jaumave, 16 Oct 1990, 102 3(9). APPENDIX B Characters measured for multivariate analysis of Cyphomeris. The letter in parentheses refers to scoring method (m = mensural, c = meristic). Units of measurement are indicated for mensural characters. Unless otherwise near fruit measurements are in ocular units at 10x. Primary line mea- surements were taken on the first lateral line of fruit. Secondary line measurements were one for a secondary line directly beneath the primary line. VEGETATIVE CHARACTERS: GHD (c)—total number of gland tipped hairs in a 1.5 x 2 mm 5 x 2 M square at 25x; HD (c)—total number of hairs in a I. 2mm se at 25x; MLL (m)—mid leaf length, in mm; MLP (m)—length of mid leaf petiole, in mn m)—distance from tip of mid- =a to widest point, in mm; MLR (m)—ratio of MLL oe a (m)—mid-leaf width, in mm; m)—upper lea length, in mm; ULP (m)— ne of ge : eaf petiole, in mm; U (m)— ae from the tip of upper leaf to widest point, in mm; ULR (m)—ratio of Ul to ULW; ULW (m)—upper leaf width, in mm; UDP (c)—depth of deepest eer UND (c)—number of undu- lations along right margin of mid leaf. FRUIT CHARAC S: (m)—fruit angle, in degrees; FD ( depth of fruit, distance from the dorsal to ventral surface; . EL (m)—length of fruit; HTW (m)—height of tallest wart on ventral surface, in ocular units at 25x; PLB (c)—number of breaks in primary line; PLL (m)—length of primary line; PLW (m)—length of longest wart of primary line; SLB (c)—number of breaks in secondary line; SLL (m)—length of secondary line; SLW (m)—length of longest wart of oe line REFERENCES CuHaturveb!, S.K. 1989. A new device of pollination in Boerhavia diffusa L. EN jetapinacean) Beitr. Biol. Pflanzen 64:55— Cruben, R.W. 1973 coal biology of ies “a cultivated Mirabilis (Nyctaginaceae). Amer. J. Bot. 60:802—809. Enric, P. and P. Rav 1969. Differentiation of populations. Science oe 1228-1232. FosBerG, ER. 1978. sai in the genus Beerhavia L. (Nyctaginaceae), | — 5. Smithsonian Contr. Bor. 39:1—2 HERNANDEZ, H.M. eet Autopolinizacién en Mirabilis longiflora L. (Nyctaginaceae). Acta ot. Mex. 12:25—30. Hoimcren, P.K., N.H. Ho_mGren, and L.C. Barnetr. 1990. Index herbariorum, ae Ee The herbaria of De world, 8th ed. New York Botanical Garden, Bronx, New Yor JOHNSTON, LM. 1944. Plants of northern Mexico, V. J. Arnold Arbor. 25: 172-173. Maurer, M.G. 1993. Multivariate analysis of geographic variation in Cyphomeris (Nyctaginaceae). Master's Thesis, New Mexico State University, Las Cruces, New Mexico. PHitiips, R.B. ne Shape characters in numerical taxonomy and problems with ratios. Taxon 32:535-544. MAHRT AND SPELLENBERG, Taxonomy of Cyphomeris 697 Reep, C.F 1970. Nyctaginaceae. In: D.S. Correll and M.C. Johnston, Manual of the vascular plants of Texas. Texas Research Foundation, Renn SAS Instirute INC 1989. SAS/IML Software: ann aad Reference, Version 6, lst ed Cary, North Carolina. «1: 990. SAS/STAT User's Guide, Version 6, 4th ed., Vols. 1, 2. Cary, North Carolina SMITH, D.W., P.W. BosLanp, N. Lownps, J. ANDERSON, and M.M. GONZALEZ-SALAN. 1993. An integrated analysis of means of multivariate responses in horticultural studies. J Hort. Sci. (submitted). SPELLENBERG, R. and R.K. DeLson. 1977. Aspects of reproduction in Chihuahuan Desert Nyctaginaceae, in R. H. Wauer and D. H. Riskind (eds.), Transactions of the symposium on the biological resources of the Chihuahuan Desert region, United States and Mexico U. S. Dept. ee Nationa Li. th ceedings series #3. STANDLEY P.C. e Allioniaceae of Mexico and Central America. Contr. U.S. Natl Herb. eee Park Service transactions and p WRIGHT, S. 1938. Size of population and breeding structure in relation to evolution. Sci- ence 87:430-431. wet ts LAD, oe structure of populations in relation to speciation. Amer. Naturalist 14: 232- . 1943. distance. Genetics 28:114—138 ZIMMERMAN, 7 R. and J.A. Lupwic. 1975. Multiple-discriminant analysis of geographical variation in the aquatic beetle ee gutticollis (Say) (Dytiscidae). Syst. Zool. 24:63— 698 Sioa 16(4) BOOK NOTICES SHAW, RicHarD J. 1995. Utah Wildflowers, A Field Guide to Northern and Central Mountains and Valleys. (ISBN 0-87421-170-0, pbk). Utah State University Press, Utah State University, Logan, UT 84322- 7800. $12.95. 218 pp, 102 color photographs, 47/s" x 81/2". This is a convenient sized book that is a portable pictorial guide to the wildflowers of Utah. Its geographic range is limited to the 21 northern and central counties of Utah. Brief introductory — include aes ee wildflowers, eae of flowers by color, among other topics. There are species of flowering see illustrated by 1; excellent color adorn Ea rch ae none eon He 1s /2 of each al and the nontechnical species description, common and scientific name occur on the > 2 1 cent preceding page so that a reader views the species porerer al and color eres side by side. There are no keys to the identification of the species other than the pictures. There is a single page of selected references, common botanical terms, and the index finalizes the book. Outdoor enthusiasts such as hikers, bikers, and campers that visit Uta h and the Rocky Mountains in neighboring states will find this book a welcome addition to field guides to our western states, Harold W. Keller, Research Associate, BRIT. Primack, RicHarpd B. 1995. A Primer of Conservation Biology. (ISBN 0-87893-730-7, pbk). Sinauer Associates, Inc., Publishers, 23 Plumtree Road, Sunderland, MA 01375. $18.95. 230 pp, 110 b/w illustrations, 7'/4" x 9!/s". This book, according to the author, follows two ie textbooks used in eee ory conservation biology courses. This primer represents a “quick” guide to the for those who want a basic familiarity with conservation biology that covers the major concepts and oe It is designed for use in short courses in conservation biology and as a guide for professionals who require a back none: in the subject but who have no need for the in n depth case ne and discussions presented in the other two book Chapter h sue s are as follows: Conservation Biology and Biological Diversity; Threats to Biological Div ; Conservation at the Population and Species Level; Conservation at the Community Level: Conservation and Sustainable Development. Each chapter is fol- lowed by a summary of major points that facilitates the quick recovery of the most impor- tant information. Suggested readings, mostly published in the 1990's, are located at the nd of each chapter and a more extensive bibliography of over 450 references concludes the book. The appendix has a list of selected € environmental organizations and sources of infor- mation that will be helpful in obtaining assistance from national and international organi- zations.—Harold W. Keller. Retearch pee BRIT. Sipa 16(4): 698. 1995 A SYNOPSIS OF THE GENUS PACKERA (ASTERACEAE: SENECIONEAE) IN MEXICO CRAIG C. FREEMAN McGregor Herbarium University of Kansas 2045 Constani Avenue Lawrence, KS 66047-3279, U.S.A. THEODORE M. BARKLEY Herbarium, Division of Biology Kansas State University Manhattan, KS 66506-4901, U.S.A. ABSTRACT The species of Packera (Asteraceae: Senecioneae) have been treated in the past as the “Aureoid” Jae within the genus Senecio. Fifteen species and three varieties of Packera are recognized in Mexico; a key, an account of the nomenclature, and statements of ranges and habitats are el Three new combinations are aes Packera scalaris var. parvasiana (Greenm.) C.C. Freeman & T. arkley, comb. nov.; Packera scalaris var. carmenensts (CC. Freeman) C.C. Freeman & TM. Barkley, comb. nov.; and Packera zimapanica (Hemsley) C.C. Freeman & T.M. Barkley, comb. nov. RESUMEN Las especies de Packera (Asteraceae: Senecioneae) han sido tratadas en el pasado como rupo “Aureoide” dentro del género Senecio. En México se reconocen quince especies y tres ee de Packera. Se presentan una clave, un informe nomenclatural y otro relativo a las drcas y los habitats. Se proponen tres combinaciones nuevas: Packera scalaris var. parrasina (Greenm.) C.C. Freeman & T.M. Barkely, comb. nov., Packera scalaris var. carmenensis (C. C Freeman) C.C. Freeman & T.M. Barkely, comb. nov., and Packera zimapanica (Hemsley) C. Freeman & T.M. Barkeley, comb. nov. Two items prompt this paper: one is the now widely accepted realign- ment of the generic concepts in the Asteraceae, and the other is the revival of interest in floristics. This paper provides a catalog for the Mexican spe- cies of the segregate genus Packera and an account of their nomenclature, plus a key for their determination The species referable to Packera A. Live & D. Love have been treated traditionally as an informal assemblage called the “Aureoid” group within the super-genus Senecio; the biology of this group was discussed in detail by Barkley (1988). Packera was erected in 1976, and its circumscription has grown to accomodate the whole Aureoid complex. The species of Packera Stipa 16(4); 699-709. 1995 700 Sipa 16(4) are distinguished from other members of Senecio, s.l., by sharing all or most of the following characters: perennial herbs arising from creeping rootstocks or a caudex; basal leaves well developed, cauline leaves progressively re- duced upward; leaf margins without callose denticles; roots fibrous, thin, and branching; haploid chromosome numbers 22 or 23, or numbers de- rived therefrom. A justification for the segregation of Packera is presented by Barkley, Clark, and Funston (in press). Packera includes some 60 species, ranging from Mexico to the Arctic and into eastern Siberia, but most of them are in temperate North America. Many occur in sites of continued mild disturbance, but few are truly weedy. The species are often imprecisely defined, and this imprecision is reflected in a complicated taxonomic history (Barkley 1988). The generic name com- memorates Dr. John Packer of the University of Alberta, Canada, an in- trepid student of the northern flora. The comparative cytology of the members of Packera is complicated by polyploidy and some apparent miscounts that have entered the literature. The comparative cytology is the subject of a study in preparation by Dr. R.R. Kowal of the University of Wisconsin, in collaboration with Freeman and Barkley. The information in this paper rests upon a study of the Aureoid senecios in Mexico, which was presented in a dissertation by Freeman in 1985 (q.v. for distribution maps, descriptions, lists of exsiccatae, and phyletic consid- erations). Adjustments have been made to account for knowledge accumu- lated since that time. The Mexican members of Packera are not so well known as the species farther north, and we expect that future field work will add considerably to what is presented here. Doubtless, the known ranges of several of the spe- cies will be expanded. KEY TO PACKERA IN MEXICO 1. Herbage glabrous or essentially so at maturity, or with light pubescence in the axils of the leaves and among the heads; basal leaves entire to dentate. The Avwrez species group. 2. Heads 1—2(—5); involucres hemispheric; basal and lower cauline leaves often deeply anthocyanic on the abaxial surface. Southern Sierra Madre ccidental. 9. P. roset i) Heads more than 5; involucres campanulate; basal and lower cauline leaves infrequently anthocyanic on the abaxial surface. 3. Blades of lowermost leaves cordate or distinctly truncate at the ss Southern Sierra Madre Occidental. . P. quebradensis 3. rae es of lowermost leaves tapering to the petiole. 4. Blades of lowermost leaves lanceolate to oblanceolate; plants taprooted or with spreading caudex, but without long stolons. — eastern Coahuila, southeast to Pefia Neva . P. hintoniorum FREEMAN AND BarKLey, Packera in Mexico . Blades of lowermost leaves obovate or oblong-ovate to orbicular plants usually stoloniferous. Coahuila. Herbage noe Uigeseue at eile and/or basal and lower cauline leaves pinnately lobed or pin 5. Basal leaves entire to ne or if lobed, then distinctly loose-tomen- e. The Tomentosi es 6. "Planes 4.5—8(-10 2 Ay i call: basa and lower cauline leaves typically onger a than wide. Baja Californ 6. Plants 0.5—5(—10) dm tall; longer overall than wide. > basal and lower a leaves (3—)4—12™ . Margins of basal and lower cauline leaves entire or nearly so, at most undulate to obscurely and unevenly dentate 8. Adaxial surface of basal leaves floccose at flowering time, abaxial surface permanently appressed-tomentose olcanica Transveral and Oaxaca uplands. 8. Adaxial and abaxial surfaces of basal leaves permanently densely tomentose to lanate. Northern Sierra Madre Occidental. ... 6. . Margins of basal and lower cauline leaves variously den to glabrate. Sierra ~~ ntate, some- times irregularly or obscurely so. Sierra Madre Occidental and cen- ral Coahuila. 7. 5. Basal leaves pinnatifid or pinnate, glabrous to sparsely pubescent, rarely densely pubescent. The Lobati species group 9. Annuals; stems arising from a taproot. Gulf Coast Plain, Rio mes Valley. 9. Perennials; stems arising from a caudex or spreading rootstock | young plants may have a taproot. 10. Terminal lobe of basal and lower cauline leaves prominent, dis- tinctly larger than the lateral lobes 701 . P. obovata 4. Po moranit 5. P. bellidifolia P. candidissima P. neomexicana . P. tampicana 11. Heads aka, Southern Sierra Madre Occidental. .......00..000.... 9. P. roset ll. ae mostl re. Abaxial surface of the basal and lower cauline leaves gla- brous to floccose-tomentose, but never with short, yellow- ish-brown, multicellular hairs. Northern Sierra Madre Occidental, Coahuila. ] 12. Abaxial surface of the basal and lower cauline leaves with few to many short, yellowish-brown, multicellular hairs, especially on the veins 13. Ray florets ca. "5 or 8; phyllaries ca. (8—)13; involucres cylindrical. Sierra Volcdnica Transversal and eee lands. 13. Ray florets ca. 8 or 13; phyllaries ca. (13-)21; — 0. P. scalars . P. sanguisorbae cres campanulate. Sierra Madre Oriental. .......... 12. P. coahuilensis 10. Terminal lobe of basal and lower cauline leaves no larger than the ateral lobes. 14. Basal and lower cauline leaves variously pinnatifid; the sinuses between the lobes aes 1/5 to 4/5 the distance to the midrib, thus the midrib is ged. 15 e with coarse, multicellular hairs; basal leaves often runcinate. Northern Sierra Madre Oriental. 15. Herbage glabrous to floccose-tomentose; hairs not obvi- wigplnedss 13. P. zimapanica 702 SIDA 16(4) ously multicellular; basal leaves crenate or sublyrate to pin- natifid. Northern Sierra Madre Occidental, Coahuila. ... 10. P. scalaris Basal and lower cauline leaves pinnate; the sinuses Berean the lobes extending to an essentially unwinged midri Aa 16. Cauline leaves absent, or rarely 1-2, the plant ati scapose, heads 2-4(-5); herbage glabrate or rarely densel tomentose. Nuevo Leon 14. P. montereyana 16. Cauline leaves (1—)3—11(—14); heads (1—)3—2 1(—32):; herb- age glabrous or sometimes tomentulose when young. Coahuila, Chihuahua. LS. PB. muillelobata 1. Packera 5 oa (Greenm.) — Weber & A. Live, ae 9:48. 1981. pee ides Gree Ann. Missouri Bot. Gard. 3:117. 1916. Type: MEXICO. »: Quebrada ‘Honda, 20-21 May 1906 ee TOTYPE, here designated: GH!; mae MO: IsocectotyPeEs: F!, MO!, NY!, UC!, Mesic sites in oak or pine-oak woodlands, in the southern half of the Sierra Madre Occidental; southwestern Chihuahua and northeastern Sinaloa south to Cerro de Tequila in Jalisco; 1000-2800 m. Flowering chiefly in March but year-round in the southern end of the range. This species is similar to P. aurea of the eastern U.S. and adjacent Canada but is shorter, leafier, and with more consistently dentate leaves. Collec- tions from southwestern Durango have leaves with notably obtuse teeth and darkened denticles, plus a distinctive aspect, and may deserve taxo- nomic recognition. 2. Packera hintoniorum (B.L. Turner) C. Jeffrey, Kew Bull. 47:101. 1992. Senecio hintoniorum B.L. Turner, Phytologia 67:238. 1989. Typr: MEXICO. Nuevo Leon: Cerro Potosi, rocky treeless summit, 3850 m, 25 Jun 1969, G.B. Hinton et al. 17048 (HoLoTYPE: TEX!) Open slopes in oak and pine-oak woodlands, upward to gravelly alpine meadow on Cerro Potosi; northern portion of the High Sierras of Sierra Madre Oriental from southeastern Coahuila, south-southeast to Pefia Ne- vada in southeastern Nuevo Leén and adjacent southern Tamaulipas; 2000— 3700 m. Flowering May through October; lower elevation plants generally begin flowering in May and are in fruit in July, whereas alpine plants flower mostly July through October. At high elevations, P. Aintoniorum has a low, subcaespitose aspect, with numerous basal leaves; a stout, weakly spreading caudex; a compact inflo- rescence with relatively large heads; and persistent tomentum. At lower elevations, the plants tend to be taller and have an erect or suberect caudex that may surmount a taproot; an inflorescence of numerous, smaller heads; and glabrous or glabrate herbage. This species resembles the more northern P. tridenticulata, and differ- ences between the two are summarized by Freeman (1985) and by Barkley FREEMAN AND BarkKLEY, Packera in Mexico 703 (1988). Packera hintoniorum was treated as a portion of P. scalaris by Muller (1939) in a discussion of the vegetation of Nuevo Leén. 3. Packera obovata (Miihlenb. ex Willd.) W.A. Weber & A. Lov Phytologia 49:47. 1981. Senecio obovatus Miihlenb. ex Willd. Sp. Pl. 3 Hees 1804. Type: “America boreali” (HoLoryPE: Willd. Herb. no. 15788, B!). This species has a rich synonymy, but the types are from temperate North America; see Barkley (1962, 1978) for summary. Widespread across the eastern half of temperate North America, but known in Mexico from only a few locations around seeps and along streams in oak and oak-pine woodlands, Sierra de la Madera and Sierra del Carmen, Coahuila; 1450-2100 m. Flowering March through May. 4, Packera moranii oo Barkley) C Paatea Kew Bull. 47: 101. 1992. Senecio moranii T.M. Barkley, Brittonia 30:69. . Type. MEXICO. Baja CaLiror- NIA Norre: common among eeu ee ne on i slope of Cerro Venado Blanc San Pedro Martir, near 31° OS'N, 29'W, ca 2725 m, 15 Sep 1968, R. es 15669 (HOLOTYPE: NY!; ISOTYPES: at CAS!, Gut. KSC!, MEXU!, MO!, SD!, US!). Seasonally damp sites in rocky igneous and metamorphic soils in open coniferous woodlands or sarcophyllous desert, San Pedro Martir of central Baja California Norte and Sierra San Francisco of northern Baja California Sur; 1400-2725 m. Flowering August through October. This is the only Packera that is known in Baja California. > ce bellidifolia (Kunth) W.A. Weber & A. Live, Phytologia 49:45. cee Re dee Kunth, Nov. Gen. & Sp. 4{folio]: 137. 1818; 4{quarto}: lee ). Typ CO. ene Mexico: “Crescit in monte ignivomo Jorullo, alt. “580 on span oe ieee anon, & Bonpland, s.n. (HOLO- : P; photo: Field Mus. Neg. 37883 MICH!, MO!, US!; tracing and frag- ments aN ioe B pe at oN ee Rape oy Kunth, Nov. Gen. & Sp. 4{folio}:13. 1818; ea 176. 1820. EXI noe THERN Mexico: “Crescit locis temperatis juxta Moran et Regla as anorum, alt. 1300 hex., Floret Majo,” Humboldt & neta 5.m. (HOLOTYPE: P; photo: Field Mus. Neg. 37881, F!, US!; tracing of holotype by J. M. Greenman on 29 900 in GH!). Senecio pauciflorus Kunth, Nov. Gen. & Sp. 4[folio}:138. pl. 365. 1818, ee 176. tab. sce 1820; non S. pauciflorus Pursh, 1814. S. vilneraria DC., Prodr. . 1838. Type: MEXICO. Veracruz: “Crescit in montibus altissimus Rexel — xicani (Cofre de Peraté) juxta eee alt. 1500 hex., floret Februario,” Humboldt & Bonpland s.n. (HOLOTYPE: P; a o: Field Mus. Neg. 37891 F!, MO!, US!). Senecio lethal Sessé & Mosia , Fl. Mex. ed. 2. 186. 1894. Type: MEXICO. Distrito i Habitat in een ae et frigidissimus S. Heremi montibus. Floret Julio,” er 1803, Sessé & avd 3139 (Lectotype: MA (McVaugh 1977); synrype: F!) Senecio vulnerarinus Sess Mociifio, Fl. Mex. ed. 2. 186. 1894, non cee vulneraria (SiC) DC. Tyee: co Distrito FEDERAL: “Habitat in Here .P. Carmelitarum,” 1787-1803 Sessé & Mociito 3147 (Lecrorype: MA [McVaugh eee SYNTYPE: F!). — 704 Stipa 16(4) Frequent in pine and pine-fir forests and in open, high meadows in the Sierra Volcanica Transversal; less frequent in the Oaxaca uplands, the Sierra Madre del Sur, and the Sierra Madre Oriental in Coahuila and Nuevo Le6én; 2740-4420 m. Flowering almost year-round, but most of the flowering specimens were collected March through August. This species is similar in aspect to P. cana of western temperate North America (Barkley 1988 for key) and to P. candidisstma of Mexico (q.v.). It is evident in the synonymy that the epithet “vulneraria” was used twice; once as Senecio vilneraria (sic) by DeCandolle to replace the pre-occu- pied S. pauciflorus of Kunth, and again (as Senecio vulnerarius) by Sessé & Mocino. Despite being homonyms, the two names ride on different type collections. 6. Packera candidissima (Greene) W.A. Weber & A. Live, Phytologia 49:46. 1981. Senecio candidissimus Greene, Pittonia 4:110. 19( o eee svaeoee CHIHUAHUA: near Colonia Garcia in the Sierra Madres, alt. 2285 m, 24 1899, oe T. Townsend & C.M. Barber 1 (Lecroryee, ae designated: N ND.C!: ISOLECTOTY PES: , GH!, MEXU!, MO!, MSC!, NMC!, ~ush; oo sites, often on thin soils of ene origin, in pine oak or pine-fir forests; northern half of the Sierra Madre Occidental from northwestern Chihuahua (Colonia Judrez) south to southwestern Chihuahua on Cerro Mohinora; 2070-3200 m. Flowering April through June. The type locality, Colonia Garcia, was a small Mormon settlement in the Sierra Madre Occidental, about 55 miles southeast of Casas Grande in north- western Chihuahua (Goldman 1951). The type collection for P. scalaris (q.v.) was also from this area. This species is somewhat weedy and forms spreading colonies from branching caudices. Ethnobotanical studies show that it has been used medicinally by the Indians of Chihuahua. Herbarium labels and correspon- dence with Dr. Robert A. Bye indicate that he is actively studying these medicinal uses. 7. Packera neomexicana (A. Gray) W.A. Weber & A. Love, ees 49:47. 1981. — neonteXicanus ce Gray, 9:55. 1883, nomen nudum. Syn Amer. ae) ee . Type: UNITED STATES. Arizona. Pima Co: Santa ae Mts, Apr 1880 . a an 49 ae TOTYPE: GH!). Typification of this name follows Turner i and replaces an earlier notion by Barkley (1978 Senecio hee Greenm., Monographie der nord- und ag oe Arten der Gattung Senecio. I. Teil. 24. 1901, nomen nudum. Engl. Bot. Jahrb. A902, nomen a. Ann. Missouri Bot. Gard. 5:44. 1918 Type: MEXICO. ees uerta de St. Diego, alt. 1980 m, 12 Apr 1891, C.V. Hartman 623 (Lecrorype, here designated: GH!; tsorecrorype: US!; photo: KSC!). This species has a complicated taxonomy and synonymy; see Barkle (1978, 1980) for summary. FREEMAN AND BarkKLey, Packera in Mexico 705 Widespread and frequent in southwestern U.S. but represented in Mexico by rather few collections; rocky soils, mostly in oak scrub and open oak- pine woodlands, scattered in central Coahuila, the northern Sierra Madre Occidental, and northeastern Sonora; 1650-4210 m. Flowering March through August. Three varieties are more or less discernable in the U.S. (Barkley 1978, 1980), and a case can be made for recognizing more than one species within the complex (Turner 1993). The Mexican plants are regarded here as be- longing to the widespread and variable var. neomexicana. 8. Packera ee uae (DC.) C. Tess ae 47:101. 1992. Senecio tampicanus DC., Prodr. 6:427. 1838. . TAMAULIPAS: Tampico de Tamaulipas, 1827 . sagas 186 (LECTOTYPE, oo pete G-DC, Prod. Herb. eh 800. 1141: I, 1; isotecroryre: F!; photo: Field Mus. Neg. 33823: F!, !, sae and fragments aie taken by J. M. Greenman at B are a at GH!). Senecio imparipinnatus Klatt, Abh. Naturf. Ges. Halle 15:333. 1882. Type: UNITED ere ean ‘Be on a la villa de Austin,” Apr 1828, J. Berlandier 1741 (LecTo- type: GH! [Barkley 1978}). ee eregeti Rydb. Bull. Torrey Bot. Club 27:170. 1900. Type: MEXICO. CutHuanua. valley of Rio Parral, near Santa Rosalia (= Ciudad Camargo, according to Goldman 1951), 21 Apr 1847, Dr Gregg 17 (HOLOTYPE: NY!; Isorypes: GH!, MO!). Senecio eae Greenm., Monographie der nord-und centralamerikanischen Arten der ctung Senecio, 1. Teil. 24. 1901, nomen nudum. Engl. Bot. Jahrb. 32:19. 1902, nomen nudum. Publ. Field Columbian Mus., Bot. Ser. 2:275. 1907. Type: MEXICO. VERACRUZ: near Tantoyuca, province of Huasteca, Oct 1858, C.L. Ervendberg 90 (Ho- LOTYPE: GH!). Monocarpic weed in open, disturbed sites, especially along the Coastal Plain in Tamaulipas, Nuevo Leon, and northern Veracruz, and less frequently in the Basin and Ranges of Chihuahua; 0-1400 m. Flowering mostly Feb- ruary through April. It also occurs northward into the central U.S., as far as Kansas. 9. Packera rosei (Greenm.) W.A. Weber & A. Love, Phytologia 49:48. a ene. as ae der nord- und centralamerikanischen Arten der Gattung Sene- , 1. Teil. . L901, nomen nudum. Engl. Bot. Jah 32:20. 1902, nomen cee Publ. Field on Mus., Bot. Ser. 2:276., Pl. 3, fig. 1. 1907. Type: MEXIC ARIT: Territory of Tepic, in the Sierra Madre, near Santa Teresa, 10 Aug vo TN. Rose 2157 (eo GH! {McVaugh 1984}; isotecrorype: US!). Marshy ground or mesic sites in pine forest regions, southern Sierra Madre Occidental from southwestern Durango (La Ciudad) south to Santa Teresa in northwestern Nayarit; 2130-2740 m. Flowering mostly in August and September. Packera rose is a facultative aquatic and can grow in saturated ground or shallow puddles, at least for a short time. This species belongs to the Lobati species group but it is also keyed among the Avrer for convenience. 706 Stipa 16(4 NS 10. Packera scalaris (Greene) C. Jeffrey Three varieties are recognized here and are distinguished as follows: |. Basal and lowermost cauline leaves abaxially glabrous or essentially so. 2. Inflorescence a compact, subumbellate or corymbiform cyme; cypselas glabrous; lower stem and basal leaves seldom anthocyanic. Northern Sierra Madre Occidental. 10a. P. scalaris var. scalarts 2. Inflorescence a loose corymbiform cyme; cypselas hirtellous on the ee lower stem and basal leaves frequently anthocyanic. Southwester Coahuila. 10 Ob. P oe var. parrasiana . Basal and lowermost cauline leaves abaxially lightly tomentose or floccose- tomentose. Northern Coahuila lOc. P. scalaris var. carmenensis 10a. Packera scalaris (Greene) C. Jeffrey var. sealants Kew Bull. iene 1992. Senecio scalaris Greene, Pittonia 4:108. 1900. ees MEXICO. Cuinuanus near Golonia Garcfa in the Sierra Madres, alt. 7600 ft, 3 Jul 1899, CLT. unvend & C.M. Barber 131 (Lecroryre, here designated: US!; isotecrorypes: F!, GH!, KSC!, MO!, MSC!, NMC!, NY!, TEX!, US!). Gravelly to sandy soils in open pine forests and in cleared pastures, in the northern Sierra Madre Occidental from northwestern Chihuahua south to northwestern Durango; 1525-3140 m. Flowering mostly June through August but infrequently into October. 10b. eae aaa var. parrasiana (Greenm.) C.C. Freeman & 7 Barkl ley mb. nov. Senecio parrasianus Greenm., Ann. Missouri Bot. Gard. af 1 a S. eee var. ‘ssieaaia (Greenm. ) C.C. Freeman, Phytologia 67:23 38. 1989. Type: MEXICO. Coanutta: Sierra de Parras, Jul 1910, C.A. Purpus 457 (LecrorypE, here designated: US!: 1 ISOLECTOTYPES: Fl, MO!). Gravelly, calcareous sites in open pine-oak-juniper woodlands and scrub in the Sierra de Parras of southwestern Coahuila; 1500-2880 m. Flowering mostly May through July. 10c. Packera — var. carmenensis (C.C. Freeman) C.C. Freeman & T.M. Ba Barkley, mb. nov. Senecio scalaris var. carmenensis C.C. Freeman, Phytologia 67:249. 1989. ee MEXICO. Coanuita: Mpto. Villa Acuna, Sierra del Carmen Canon de Senrenela (= Cafion del Centinel) on Hacienda Piedra Blanca; moist stream side, 6 Jul 1936, EL. Wynd & C.H. Mueller 546 (HoLoTyPE: NY!; 1sorypes: GH!, MICH! MO!, TEX!). Gravelly loam that apparently is derived from rhyolite, in open pine-oak woodlands in the Sierra del Carmen of northwestern Coahuila; 1500—2590 m. Flowering May through September. 11. Packera sanguisorbae (DC.) C. cee ae Bull. 47:101. 1992. Sene- cio sanguisorbae DC. we 6: ce 1838. Type: XICO. México: “circa Toluccanum {Toluca}, Floret Aprili,” 1834 Raita 292 i LecTtoTyPE: G-DC [McVaugh, 1984], Prod. Herb. IDC (on 141: I, 3; photo: Field Mus. Neg. 38811, F!, MICH!, MO!). FREEMAN AND BarKLEY, Packera in Mexico 707 Cineraria pinnata La Llave in La Llave & Lexarza, Nov. Veg. Descr. 1:29. 1824, non Senecio Pee Poiret. Senecio pinnatisectus DC., Prodr. 6:427. 1838. Type: MEXICO. ERACRUZ: “crescit ab umbram in petrosis, amque martio florentem inve declivitate quae ducit ad ultimum cataractam fluminis Blanco,’ P. La Llave os O- TYPE: G) Along streams and on mesic sites in pine, pine-oak, and pine-fir forests throughout the Sierra Volcanica Transversal, the southern end of the Sierra Madre Oriental, and the Oaxaca Uplands of the Sierra Madre del Sur; 1465— 3355 m. Flowering mostly May through November but also throughout the year. 12. Packera coahuilensis (Greenm.) C. Jeffrey, Kew Bull. 47:101. 1992. Senecio nnn Greenm., Monographie der nord- und centralamerikanischen Arten er tung Senecio, : Teil 23. 1901, nomen nudum. ae Bot. ee 32:19. 1902, nomen ae Publ. Field Columbian Mus., Bot. Ser. 2:275. 19, fig. 2. 1907. XICO. a Lerios (= Lirios?), Feb Oct eH E Pal 755 (LECTO- TYPE, here designated: GH!, tsotecrorypes: K, NY!, US!; photo '. The NY speci- men the following: “mt section 15 leagues . : Saltillo, i 10000 fe, Jul 10- 13, e Senecio leonensis Greenm., nfonogteoie der nord- und centralamerikanischen Arten der Gattung Senecio. 1. Teil 23. 1901, nomen nudum. Engl. Bot. Jahrb. 32:19. 1902 nomen nudum. Publ. Field Columbian Mus., Bot. Ser. 2:276. pl. 19, fig. 1. 1907. Type: . Nuevo Leon: Sierra Madre, near Monterrey, | Jun 1899, C.G. Pringle 2894 Gomme GH!; photo: MO!). Senecio a eke Greenm., Publ. Field Columbian Mus., Bot. Ser. 2 1907. TYPE: MEXIC SN: Sierra Madre above Monterrey, 2500— 3000 ar ee Mar 1906, C.G. are es anes GH1I; tsorypes: F!, GH!, KANU!, LL!, MICH!, MSC!, NMC!, SD!, US!, WIS!; photo: F!, MO!). Senecio hypotrichus Greenm., Ann. Nissen Bot. Gard. 2:612. 1915. Tyre: MEXICO. San Luis Porosi: chiefly in the se of San Luis Potosi, 6000- 8000 ft, 1878, C.C. Parry & E. Palmer 533 (Ho.oryre: US!; photo with fragments, MO!). Greenman (1915) noted that Parry & ee 533 is a mixed collection, me that the species was de- scribed from the specimen at US. The specimens of that number at K GH, and NY are included here within P. tampicana. Along streams and in open mesic to mildly xeric sites in pine-oak, pine, and pine-fir woodlands and occasionally in alpine meadows, throughout the Sierra Madre Oriental but most frequent in the High Sierras of west- central Nuevo Leén and southeastern Coahuila and the northern sierras of Coahuila; 1345-3720 m. Flowering mostly April through August, but also at other times. The name Senecio cyclophyllus has been used for plants on which the ter- minal lobe of the basal and lower cauline leaves is much enlarged and the lateral lobes are relatively reduced. Examination of a series of specimens suggests that this morphotype is merely a part of a large cline and it is accomodated comfortably within P. coahuilensis. 708 Stipa 16(4) 13. Packera zimapanica (Hemsley) C.C. Freeman & T.M. Barkley, comb. NOV. Senecio zimapanicus Hemsley, Biol. Centr. Amer. Bot. 2:248. 1881. Type: MEXICO. Hipatco: Zimapan, J.M. Colter 423 (Putative HOLOTYPE: K [Greenman L915). Gravelly soils and stream banks in pine woodlands in the Sierra Madre Oriental from Cerro Potosf in southeastern Nuevo Leén, southward to north- ern Hidalgo; 2740-3430 m. Flowering July and August. A highly distinc- tive and poorly known species. 14. Packera montereyana (S. Watson) C. Jeffrey, Kew Bull. 47:101. 1992. Senecio Haka Watson, Proc. Amer. Acad. Arts 25:255. 1890. Type: MEXICO. Nurvo Leon: dry shaded ledges on the Sierra Madre, near Monterrey, 27 Jun “ie C.G. Pringle 1922 (LecrorypE, here designated: GH!; 1soLecrorypes: COLO!, F MICH!, MO!, NY!, UC!, US!). Known only from limestone ledges in pine forests in the Sierra Madre above Monterrey, Nuevo Leon; 1370-1525 m. Flowering April through September and rarely at other times. 15. Packera millelobata (Rydberg) W.A. Weber & A. Léve, ce cc 49: 47. 198 en eat Rydb. Bull. Torrey Bot. Club 27: : pl. 5 1900. Type: UNITED STATES: Texas: hills on the Limpia, 1851-1852 Wright 1287 (HOLOTYPI en ISOTYPES: a MO!, NY!, PH!, US!; photo: a Mesic open sites in oak woodlands of northern Coahuila and central Chihuahua and north of the Rio Grande in adjacent Trans-Pecos Texas; 1066-2130 m. Flowering mostly March through September, but also at other times. ACKNOWLEDGMENTS The auchors gratefully acknowledge the assistance of the staff of the Herbario Nacional (MEXU), Instituto de Biologia, Universidad Nacional Auto’noma de México for making the facilities of that institution available and for collegial kindness. Particular thanks are due to Dr. José Luis Villasehior (MEXU) for his interest in the work and for sharing his consid- erable knowledge. This paper is Contribution no. 95-549-] from the Kan- sas Agricultural Experiment Station, Manhattan. REFERENCES BARKLEY, T.M. 1962. A revision of Senecio aureus Linn. and allied species. Trans. Kansas Acad. Sci. 65:318—408. . Senesio linn N. Amer. FL. TL 10:50-39, —C«d' OBO. Vaxonomic notes on Senecto tomentosus and its allies (Asteraceae). Brittonia 32:291—308. 1985a. Generic boundaries in the Senecioneae. Taxon 34:17—21. —__________. 1985b. Infrageneric groups in Senecio, s.l., and Cacalia, s.l., (Asteraceae:Senecioneae) in Mexico and Central America. Bical 37:211-218 FREEMAN AND BarKLey, Packera in Mexico 709 1988. Variation amore the aureoid senecios of North America: a geohistorical cena Bot. Rev. 54:82—10 Bark ey, T.M., B.L. CLarK, a A.M. FUNSTON (In press). The paar genera of Senecio, s.l., and oe s.l., (Asteraceae: Senecioneae) in Mexico and Central America. In: D. J.N. Hind, et al., eds. a lar of the International Gs Conference, Kew, 1994. Royal Beanie Gardens, Kew. Bremer, K. 1994. Asteraceae; ae tics & classification. Timber Press, Portland. FREEMAN, C.C. 1985. A revision of the aureoid species of Senecio (Asteraceae:Senecioneae) in Mexico, with a peri ne and Ser aek ppsiatenie of the aureoid com- ersity, Manhatta plex. Doctoral dissertation, Kansas State Univ F Go.tpMaN, E.A. 1951. Biological investigations in Mexico. eee Misc. Coll. 115: 1-476. GreENMAN, J.M. 1915. Monograph of the North and Central American species of the genus Senecio - Part I. Ann. Missouri Bot. Gard. 2:573—626. Jerrrey, C. 1992. The tribe Senecioneae (Compositae) in the Mascarene Islands with an annotated world check-list of the genera of the tribe: notes on Compositae VI. Kew Bull. 47:49-109. McVauGu, R.J. 1977. Botanical results of the Sessé & Mocifio ate (1787-1803) I. Summary of excursions and travels. Contr. Univ. Michigan Herb. 11:97-195. «984. Flora Novo-Galiciana. vol. 12, Compositae. 7 Michigan Press, Ann Achor Mutter, C.H. 1939. Relations of the vegetation and climatic types of Nuevo Leén, Mexico. mer. Midl. Naturalist 21:687—72¢ Turner, B.L. 1993. Lectotypification of Senecio neomexicanus A. Gray. Phytologia 75: 5 2 710 Siwa 16(4) BOOK NOTICES TERANISHI, Roy, RoN G. Burrery, and Hirosui Sucisawa (Eds.). 1993. Bioactive Volatile Compounds from Plants. ACS Symposium Se- ries. ISBN 0-841 2-2639-3, hbk). American Chemical Society, P.O. Box 57136, Washington, DC 20037-0136. $79.95. 309 pp, 6" x 9". This book was developed from a symposium sponsored by the Division of Agricultural and Food Seacoes at the 203rd National Meeting of the American Chemical Society. Chapter one is an overview of the book contents that discusses some of the latest methods and gs in aa preparation, analyses and identification of volatiles from various vor chemists from industry, government and ~— owers, leaves, and fruits. Perfumery and flav academia i identified and catalogued tens of thousand of compounds and this book resents the latest findings of volatile bioactive compounds from plants. There are 20 — divided into thematic sections: Biogenesis and Biochemistry; Essential Oils; Flow- . Some see a of the chapter headings include more general subject matter and in most cases very specific topics: Formation of Some Volatile Components from Tea; Anti- microbia fee of Green Tea Flavor Components: Effectiveness against Streptococcus mutans, Therapeutic Properties of Essential Oils and Fragrances; On the Scent of Orchids; Volatile Components of Apricot Flowers; Flower Scent of Some Traditional Medicinal Plants. The invited and peer reviewed papers are highly technical and will be of special interest to biochemists and chemists working in the this special field.—Harold W. Keller, Research Associate, BRI Manaster, JANE. 1994. The Pecan Tree. (ISBN 0-292-75153-2, hbk). The University of Texas Press, P.O. Box 7819, Austin, TX 78719. The Corrie Herring Hooks Series, No. 27. $17.95. 112 pp, Eight color and b/w photographs, with six of the color by Paul Montgomery, This is another volume in the natural history series for the general readership. It is of interest to the personnel of the pecan industry as well as to the public and especially Texans, since the Pecan Tree (Carya #/inoinensis) has been designated as the state tree of Texas. The Chapters cover: Introduction and Range, Botanical Niche, History, Cultivation and Improvement, Texas: A Case Study, Orchards, Animal eee The Pecan Industry, Nutrition, Recipes, Conclusion, and Resources. 70 to the present time, the history of the pecan and its uses by humans From 7000 B.C. t The pecan tree itself has evidently been found as fossilized unfolds in a narrative style. remnants in the lower Cretaceous period (135 million years ago). In the 1700s during the westward immigration, the settlers encountered the pecan in the country of the Ilinois Indians. The “Illinois Nuc” became popular back east and in Europe where it was first epee in a — article from a plant in cultivation. Wangenheim designated the specific epithet (species name) as “é//inoinensis” that means belonging to the Illinois even though ie geographical area is today in the northeastern part of its native range. The pecan is not native to Georgia in spite of the fact that today Georgia i is the top producer of pecans in the nation. This is an excellent book relating to the many historical up to currence aspects of the pecan and its importance in today’s life Won 1 Mahler. Sipa 16(4): 710. 1995 HISTORICAL EVIDENCE OF THE NATIVE PRESENCE OF SABAL MEXICANA (PALMAE) NORTH OF THE LOWER RIO GRANDE VALLEY LANDON LOCKETT 3210 Stevenson Avenue Austin, TX 78703, U.S.A. ABSTRACT References by Lindheimer (1845a and b, 1879) and Engelmann (Gray 1850) to Texas central coast palm trees, and a description of the distribution and uses of Sabal mexicana Mart. by Sargent (1905, 1922), add to the evidence that a wild population of the species discovered in Jackson and Victoria counties in 1989 is indigenous. RESUMEN Referencias de Lindheimer (1845a y b, 1879) y Engelmann (en Gray 1850) a palmeras en la litoral central de Texas, y una descripcién de la distribucién y los usos de Sabal mexicana Mart. hecha por Sargent (1905, 1922), contribuyen a la evidenciar que una poblacién natural de la especie descubierta en los condados de Jackson y Victoria en 1989 es autéctona. Lockett and Read (1990) report the discovery of a wild population of Sabal mexicana (syn. 8. texana) along Garcitas Creek (Jackson and Victoria counties), 200 miles north of the Lower Rio Grande Valley (LRGV), the region previously considered the northern limit of the species range (Correll & Johnston 1970, p. 341). Since S. mexicana has been widely cultivated in Texas, Lockett and Read (1990) presented historical evidence, going back to the colony established on Garcitas Creek by French explorer René-Rob- ert Cavelier, Sieur de La Salle, in 1685, in order to show that the Garcitas Creek population is not a case of escape from cultivation. Further, Lockett (1991, p. 66) presents historical evidence that the species may have also been native to San Antonio Springs, the headwaters of the San Antonio River. The purpose of the present paper is to present historical evidence discovered since Lockett and Read (1990) and Lockett (1991). While Ferdinand Lindheimer was collecting specimens in Texas for Asa Gray he corresponded with George Engelmann, and his letters to Engel- mann are now in the possession of the Missouri Botanical Garden. The late Minetta Goyne performed the arduous task of translating these letters, and the translation appears in Goyne (1991). As Goyne explains in her preface, the letters were hand written in German under field conditions and in an archaic script. Accordingly she assures us (p. xii) that “Translating the let- ters was not nearly so great a challenge as deciphering them.” Sipa 16(4): 711-719. 1995 712 Sipa 16(4) In reading Goyne’s entire book, two passages in her translation of Lindheimer’s letters caught my attention: Passage 1.—In a letter dated “22 January 1845” from “Camp on the Agua Dulce,” which Lindheimer describes as “7 miles from Port Lavaca on the West Matagorda Bay,” and in a paragraph titled “Cactus.” (Goyne 1991, pe LiL) I collected seeds of the fanshaped agave [“Facherpulke’} with stem that is often 20! to 40! high; Passage 2.—In a letter from “New Braunfels, Bexar County” dated “be- ginning of August 1845” and describing the vegetation as one moves in- land, up the rivers, from the shore of Matagorda Bay. (Goyne 1991, p. Loi) Farther upstream occasional single elms, the aforementioned Yacca and Opuntias, but less vigorous ones, often ... with trunk about 15! high, a palmetto (Chamaerops palmetto). Since the description of the plant referred to in Passage 1 seemed to fit a palm tree better than any known agave, and Passage 2 clearly referred to a palm (even though unclear as to whether the palm or another plant had the 15’ trunk), I contacted Goyne, told her of the discovery of Sabal mexicana on Garcitas Creek (15 miles north of Port Lavaca), and asked her to re- evaluate her interpretation of the two passages in question. She replied that when she translated the letters she had no idea palm trees could be native near Port Lavaca, but since receiving my communication had decided that the correct interpretation of the handwritten word she had transcribed as Facherpulke (literally “fan pulque”) was Facherpalme (‘fan palm”). The origi- nal translation (“fanshaped agave”) was due, she explained, to her having interpreted Lindheimer’s scrawled @ as ~, and his m, which had an overly large first loop, as &e. As this interpretation yielded a German spelling of the Spanish word pa/que, which refers to a Mexican alcoholic beverage made from the juice of the agave, she reasoned that, since she knew that Lindheimer had lived in Mexico and liked to play with words, he had used -pu/ke to mean “agave.” Her placing the word Facherpulke in brackets in her first translation indicates, of course, her original doubt about the -pw/ke inter- pretation. Her emended translation of Passage 1, which appears in a signed statement (see Appendix) she sent both to me and to the Texas A&M Press, reads as follows: I collected seeds of the fan palm [Facherpalme} with trunk that is often 20! to 40! high; Figure | is a photocopy of the numbered paragraph of Lindheimer’s let- ter (1845a) that contains Passage 1. A transliteration of Passage l, in mod- ern orthography, reads as follows: Lockett, Sabal mexicana north of the Lower Rio Grande Valley pale) Uae en oe re igs ae fiver (hin 34 LE xen ones Zo. ee) LAGI gal Fic. 1. Excerpt from Lindheimer letter dated “22 January 1845.” Passage 1 begins with second word of fourth line from top and ends with period just right of center of fifth line. The transliteration reads as follows: Samen, von der grossen Facherpalme mit Stamm oft 20 bis 40 Fuss hoch, habe {ich} gesammelt. Samen, von der grossen Facherpalme mit Stamm oft 20 bis 40 Fuss hoch, habe {ich} gesammelt Although Passage 1 clearly includes the adjective grossen “large,” Goyne omitted it from both her original and revised translations. But because she included the translation of this word in an intermediate version of her trans- lation of the sentence (pers. comm.), and because its absence does not sig- nificantly change the meaning, I assume its omission was an oversight. Reexamining Passage 2, Goyne determined that the ellipsed word is nicht “not,” and that she had accidentally placed the ellipsis after Adufig when it should have preceded it. Her emended translation (see Appendix), which makes it clear that the 15 foot trunk pertains to the palm, is as follows: Farther upstream occasional single elms; the aforementioned Yucca and Opun but less vigorous ones; not often, with trunk about 15' high, a fan palm sei a palmetto). Figure 2 is a photocopy of the portion of Lindheimer’s letter (1845b) containing Passage 2. A transliteration, in modern orthography, of the por- tion of Passage 2 pertaining to palms (the segment beginning with “not often” and ending with “pa/metto”) reads as follows: nicht haufig aber mit ohngefaéhr 15 Fuss hohem Stamm eine Facherpalme (Chamaerops palmeto). The correctness of Goyne’s revisions is confirmed by Lindheimer himself “Overview of the Flora of Texas,” an essay published decades later in Cas (Lindheimer 1879, p. 37). In a description of the coast, going inland from Indianola, there is a passage paralleling Passage 2 above. As translated by Goyne it reads: A bit farther from the water Sabal adansonii. Farther upstream the aforementioned large Yucca and Opuntia but not as vigorous as [those] in the vicinity of the sea Gulf}. Occasionally the palm Sabal palmetto. 714 Spa 16(A4) GOLD EG fe oO PS LE Gp 8 . GPS GpccetS dois goa i leg, HEE Sh RE Ge BOE IS oo Peni cee are tetE: x ee, A - oe =e Cead . Wat ol 7 SPAS PGi Ld. Dae Bop age bo poo P? Ki pag SG TRE. Sarai ag ar ‘ 254 —~ pa ; . 7 < eg - : < a Za oe en th c Thee Ease Z ects abhi he + = ho . co- \-= a 7 m 4 bed _ = Syne t 2 > é s c 4 rae ee . Mp CoS a! SEED EMIS i Ing ia = ae ll Ses ie ES Zope Fic. 2, Fecoat from Lindheimer letter dated “beginning of August 1845.” The portion of Passage 2 pertaining to palms begins with the word nicht near the end of the next to last line of the excerpt, and continues till the end of the excerpt. vicht appears after the comma r the first 2 of Chamaerops. The transliteration reads as follows: 7 haufig aber mit ee 15 Fuss hohem Stamm eine Facherpalme (Chamaerops palmeto). In 1845 Lindheimer had, like others in his time, incorrectly applied the genus name of a European palm (Chamaerops) to an American palm. In the essay he corrects this to Sabal. The epithet palmetto actually refers to a tall palm native to the southeastern United States. This error is to be expected, however, since in his time the Texas palms had not yet been identified as a species distinct from S. palmetto, which they closely resemble. His separate reference to S. adansonii (syn. S.minor) makes it clear his S. palmetto was a tall palm, and not dwarf palmetto. Engelmann, who apparently never visited the area but was the recipient of Lindheimer’s letters, also confirms Goyne’s emended translations of Pas- sages 1 and 2. In an addendum to Gray’s (1850, p. 235) description of plants collected by Lindheimer during 1845-1848, Engelmann identifies the area where the specimens were collected as that drained by the Colo- rado, Navidad, Lavaca, Guadalupe and San Antonio rivers. In an account of the vegetation of that area he first describes the plants of the bay shore, then writes: Some miles higher up the rivers, on the clayey soil, solitary Elms and Palm trees are seen; the prairies have a stiff black soil thickly matted with grass. The prevalent tree now becomes the Live Oak along the rivers, as well as in small groves on the prairi es. Garcitas Creek lies between the Lavaca and the Guadalupe. The Engel- mann quote (“Some miles higher up the rivers”), and both Lindheimer quotes (“Farther upstream”) fit the Garcitas Creek population today. Be- cause Sabal mexicana is not salt tolerant, as is S. minor, the Garcitas Creek palm trees do not begin until about four miles upstream from the bay. Lockett, Sabal mexicana north of the Lower Rio Grande Valley yal: Since Engelmann writes of “rivers”, the implication of his statement is that S. mexicana occurred along all the rivers from the Colorado to the San An- tonio. As reported in Lockett and Read (1990, pp. 82-83), the central coast population of S. mexicana was almost entirely removed, due to a demand for use of the trunks as wharf pilings, and for landscaping. There is, however, evidence that the range of S. mexicana may have reached as far northeast as the San Bernard River. Sargent (1905, p. 109) gives the distribution of the species as “Rich soil of the bottom-lands near the mouth of the Rio Grande, in Texas, and southward in Mexico in the neighborhood . the coast.” In Sargent (1922, p. 104), however, the distribution is given : “Rich soil of the bottom-lands on the Bernado {sic} River, Cameron Cane and near the mouth of the Rio Grande, Texas, and southward in Mexico in the neighborhood of the coast.’ The Sargent (1922) statement is unclear, and partly redundant. First, there is no “Bernado River” in Cameron County, or anywhere in Texas. Second, since Cameron County is the southernmost county in Texas, and is bordered by the Rio Grande on the south and the Gulf of Mexico on the east, any place “near the mouth of the Rio Grande, Texas” is in Cameron County. But, while there is no “Bernado River” in Texas, there is the San Bernard (from Spanish: San Bernardo) River, which flows into the Gulf 240 miles northeast of the Rio Grande, and is the next river up the coast from the Colorado. Further, six miles southwest of the San Bernard River, in the extensive bottomland forest that covers western Brazoria County, there is a small population of tree-sized palms that appear to be hybrids of Sabal mexicana and S. minor (Lockett 1991). In spite of the ambiguity of Sargent’s (1922) statement, we could hy- pothesize that sometime between 1905 and 1922 Sargent was informed, probably orally, that S. mexicana grew in the bottomlands of the “Bernardo River.” Since Sargent, like almost all other botanists since him (for an ex- ception see Cook 1908, p. 5n.a; and 1913 p. 11), had theretofore believed that the Texas range of the species was restricted to the , he then assumed that the river in question must be in the LRGV. The discrepancy in spelling between the informant’s “Bernardo” (although we don’t know how this person actually pronounced it) and Sargent’s “Bernado” could be attributed to Sargent’s Bostonian speech, characterized by loss of postvocalic r. Since the San Bernard is little known outside of Texas, we could assume Sargent had never seen the name “Bernardo” written. Adding to the likeli- hood of Sargent’s misinterpretation is the fact that his informant could have also spoken with an r-dropping dialect, since such speech is common in Texas, especially in Brazoria County, where the hybrids occur, and through which the San Bernard flows. 716 Spa 16(A4) In both the 1905 and 1922 editions of Sargent the description of the uses of S. mexicana reads as follows: “On the Gulf coast the trunks are used for wharf-piles, and on the lower Rio Grande the leaves for the thatch of houses.” Aside from confirming historical evidence reported in Lockett and Read (1990, p. 82), this statement suggests the existence of two separate Although in his letter dated 22 January 1845 Lindheimer told of col- lecting seeds from the tall palms he refers to, searches by curators at GH, A, MO, BH, TEX and BRIT failed to reveal any Lindheimer palm speci- mens other than a single specimen of Saba! minor. Likewise searches by curators at GH, A, BH, TEX and BRIT (including a search of Sargent’s correspondence at A), failed to reveal any specimen or correspondence that would form the basis for Sargent’s “Bernado River” reference. So far as I know the only well established wild population of S. mexicana that, on the basis of historical evidence, does not appear to be an escape from cultivation (although it is presumably second growth) is the Garcitas Creek population. This population is scattered through the bottomland forest for several miles along the east side of Garcitas Creek, and contains many mature (reproducing) specimens. The other wild, reproducing popu- lations known to me are apparent escapes from cultivation, and have only one mature palm each. One is along the Aransas River on and adjacent to the Welder Wildlife Refuge; another is near the Guadalupe River, south of Victoria; and the third is along a normally dry tributary of the Medina River, 12 miles northwest of Bandera. The fact that S.mexicana is establish- ing itself in the wild in the apple-growing region of the Hill Country, almost 300 miles northwest of Brownsville, should discourage hasty as- sumptions about its hardiness, or just how far north it might have once ranged. A factor that has probably contributed to the long-standing assumption that S. mexicana is not native north of the LRGV is the widespread belief that Spanish explorers named the Rio Grande River the Rio de las Palmas, presumably because of the S. mexicana palms growing along its banks. Ac- cording to Weddle (1985, pp. 104-105) and Chipman (1995), however, this belief, often expressed in articles and even textbooks, stems from a misinterpretation of Spanish explorers’ journals. Although there is a river that was called the Rio de las Palmas, it is the Rio Soto la Marina, in Mexico, and not the Rio Grande. There are probably more wild populations of S. mexicana in the thick bottomland forests along South Texas rivers, either as remnants of the origi- nal population, as on Garcitas Creek, or escapes. To the extent conserva- Lockett, Sabal mexicana north of the Lower Rio Grande Valley 717 tionists are able to locate such populations and persuade land owners to protect them, and reintroduce the species in appropriate habitat, perhaps Sabal mexicana can be restored to its former range. Palms with trunks of 20 to 40 feet will overtop the live oaks, creating a scene that may seem exotic to us, but apparently would have been familiar to La Salle and the Karankawas. ACKNOWLEDGMENTS I am grateful to Martha Riley (MO) for furnishing me photocopies of Lindheimer’s letters; to Helga von Schweinitz and Barbara Bresslau for help in translation; to Chris Durden, of the Texas Memorial Museum, for calling my attention to the statements in Sargent; and to Del Weniger, Anders Saustrup, and the staff of the University of Texas Center for Ameri- can History for help in searching for early references to palms. For their help in my search for herbarium specimens and early botanical records I am grateful to Judith Warnement of The Botany Libraries, Harvard Univer- sity; to Emily W. Wood (GH, A); to Steve Spongberg, Jennifer Brown and Richard Howard (A); to James C. Solomon and Chris Dietrich (MO); to Philippe Morat (P); to Natalie Uhl (B); and to Barney Lipscomb and Yonie Hudson (BRIT). I am especially grateful to Robert Read (US) and Billie Turner (TEX) for their guidance, encouragement and interest in my re- search. I very much appreciate the generosity of David Marlowe, Emily Dial, Patricia Martin and Charles W. Weber.Jr. for letting me look for palms on their ranches; and of James Teer and Lynn Drawe of the Rob and Bessie Welder Wildlife Foundation for letting me stay at the foundation’s wildlife refuge while searching for palms there. For helping me in this fieldwork Iam grateful to my wife Carol, Mike Rayburn, Mikel Borg, George Stevenson, and H. Dixon Hoese. I also want to thank Benny Simpson, Dorothy Mattiza, William FE. Mahler and the other members of the Native Plant Society of Texas for their continued help, support and recognition of my research. Finally, I want to dedicate this article to the late Minetta Altgelt Goyne, without whose research and cooperation it would not have been possible. 718 Stipa 16(4) APPENDIX Photocopy of Goyne's signed statement presenting her emended translations of Passages 1 and 2. Below are my emended translations of passages in my book A Life among the Texas Flora: Ferdinand Lindheimer's Letters to George Engelmann (TASMUP, 1991), which emendations I undertook at the suggestion of Landon Lockett, who is conducting research into native varieties of Texas palms. On closer examination of the photocopies of Lindheimer's letters in my possession, I have become convinced that these translations should be as follows, largely because Mr. Lockett has made me aware that large palms are not unknown in the region Lindheimer was des- cribing, information that I did not have while preparing the book for publication. The error on p. 111 resulted from illegible letters in the word that evidently was Fucherpalme (German for fan palm), The errors on p. 123 resulted from very faint punctuation marks (causing an ambiguity) and an accidentally misplaced ellipsis, which proved on closer examination to indicate the word preceding hdufig (German for often/common) was nicht (German for not). p- 111, 11, 10-13: I collected seeds of the fan palm /“Fcherpalme"/ with trunk that is often 20' to 40' high; a large Melocactus? (perhaps the same as the one sent) grows on marshy ground. p- 123, ll. 4-6: Farther upstream occasional single elms; the aforementioned Yucca and Opuntias, but less vigorous ones; not often, with trunk about 15' high, a fan palm (Chamaerops palmetto). oe Le eee copies sent to Landon Lockett and Texas A&M University Press Lockett, Sabal mexicana north of the Lower Rio Grande Valley REFERENCES CHIPMAN, D.E. 1995. Alonso Alvarez de Pineda and the Rio de las Palmas: Scholars and mislocation of a river. Southw. Hist. Quart. 98:369—38 Gon. O.F 1908. erie of vegetation on the South Texas prairies. U.S.D.A. Bur. PI. Industr. Circ. 14. 191 . A new ornamental palmetto in southern Texas. U.S.D.A. Bur. Pl. Industr. Cie 7 113:11- ORRELL, M. Johnston. 1970. Manual of the vascular plants of Texas. Texas Re- search Foundation, Renner. oo 1991. A life states the Texas flora: Ferdinand Lindheimer’s letters to George gelmann. Texas A&M Univ. Press, College Station. es A. 1850. Plantae Lindheimerianae, Part II. Boston J. Nat. Hist. 6:141-195, 209— 240 iaeuenee, F. 1845a. Letter to George Engelmann, dated Jan. 22,1845. Archives, Mis- Louis ouri Botanical Garden, St. Letter to Geore Engelmann dated “beginning of August 1845.” is. Archives, Missouri Botanical Garden, St. Lou site Vor Dutsare ind Aphendiungen von Ferdinand Lindheimer in Texas. Theodore Wane. Frankfur Lockett, L. and R. W. Reap. 1990, Extension of the native range of Sabal mexicana (Palmae) in Texas to include Central Coast. Sida 14:79-8 . 199 LOCKETT, Native Texas palms north of the capes Rio Grande Valley: recent dis- eries. ee 35: een C.S. 1905. Manual the trees of one America (exclusive of Mexico). Houghton, Mifflin and ete Boston and Nev _ «1: 922. Manual of the trees of America (exclusive of Mexico). Houghton, Mifflin and Company, Boston and New York. Wepb.E, R.S. 1985. Spanish sea: the Gulf of Mexico in North American discovery, 1500— 1685. Texas A&M Univ. Press, College Station. 720 Sipa 16(4) BOOK NOTICE Buat1a, SAtTIsH C. 1994. Moss Protonema Differentiation. (ISBN 0-471- 94438-6, hbk). John Wiley & Sons Inc., 605 Third Avenue, New York, NY 10158. $79.95. 296 pp. This is also Series No. 13 of Research Studies in Botany and Related so Fields re Studies Press Ltd., Taunton, Somerset, England: ISBN 0 86380 The See ie titles describe ane coverage: 1) Structure and nee of Spores, 2) The Fi orme otonema, 3) Structure and Development of Caulonema, 4) Orientation of oe by Light and Gravity, 5) Regulation of Caulon ma Formation by Auxins, 6) Auxin-cyclic Adenosine pn cee 7 P) Interaction in Protonema Differenciation, 7) Cytokinin Action on Caulon s, 8) Calcium Mediation Cyto- kinin-induced Bud Formation, 9) Nucleic Acid “and poe Synthesis in Relation to Pro- tonema Differentiation, and 10) Concluding Remarks From the Preface: eed for the present volume was felt because there has been no effort so far to criti- cally ae and compile the available information, in a single comprehensive volume, or a some of the topics have been discussed occasionally in various symposia and conferences. This work provides an up-to-date, in- pa analysis of our present under- ice of cell differentiation in moss protonema. It is intended for advanced students and professionals interested in cell differentiation | : plants in general and mosses in par- ticular. ten in ten chapters, the book provides an insight into the ultrastructural, physi- ological, biochemical and molecular events associated with moss protonema differentia- tion The Veni cited in the References extend from page 255 to 291 and is indicative of the depth of the subject matter that is discussed and evaluated throughout the text. —Wm. FE Mahler. Sipa 16(4); 720. 1995 TAXONOMIC NOTES ON NEW VARIETIES OF SPECIES OF CLITORIA (LEGUMINOSAE - PHASEOLEAE - CLITORIINAE) PAUL R. FANTZ Department of Horticultural Science Box 7609, North Carolina State i ae Raleigh, NC 27695-7609, ABSTRACT New varieties of 10 species of Clitoria (Leguminosae - Phaseoleae - Clitoriinae) are de- scribed with commentary. Characteristics of segregation from other varieties within the species are presented. Newly described varieties include: C. arborea var. pseudoamazonica Ww Brazil), C. hues var. angustissima (Brazil: Parana), C. guianensis vat. ache aicains (Brazil: Minas Gerais), C. /eptostachya var. fruticosa (Brazil: Para), C. mariana United States: oo. Cc, pe Aya var. congesta (México: México), C. i stachya var. let (México: Morelos), C. pozuzoensis var. ee (Pert: Loreto). ew combinations ce C. falcata var. aurantiaca (Southern Brazil), C. falcata var. latifolia ane Para), C. heterophylla var. — (Madagascar, es tius), C. gwianensis var. chapadensis (Brazil: Mato Grosso), and C. stipularis var. latifolia (Brazil: Ceara, Maranhao). RESUMEN Se describen variedades nuevas de diez especies de C/itoria (Leguminosae - Phaseoleae - Clitoriinae). Se entan las caracteristicas de segregacién de otras variedades dentro de las especies. a ae nuevamente descritas incluyen las soe C. arborea vat. pseudoamazonica (Brasil o ie C. epetiolata var. angustissima (Brasil: na), C. guianensis var. macrocleistogama (Brasil: Mineas Gerai - C. leptostachya var. ae “(Brasil Para), C. mariana vat. pubescentia (Estados Unidos: Florida), C. polystachya var. congesta (México: México), C. polystachya var. pringler (México: Mor nd : pozuzoensis var. schunker (Pert: Loreto). Se incluyen las siguientes combinaciones nu C. falcata var. aurantiaca (Brasil del: sur). G; wee var. latifolia (Brasil: Paré), C. eae var. ee (Madagascar, Mauritius), C. gaianensis var. chapadensis (Brasil: Mato Grosso), y C. stipularis var. latifolia (Brasil: Ceara, Maranao). INTRODUCTION Clitoria (Leguminosae, Phaseoleae, Clitoriinae) comprises 60 species within the pantropical-subtropical belt. Several species are divisible into populations that are sufficiently distinct, and have been recognized at the level of varietas. All types cited have been examined, unless noted. The objective of this paper is to validate these names. TAXONOMY 1. Clitoria arborea Hoftm. ex Benth. var. pseudoamazonica Fantz, var. nov. Varietas nova psendoamazonwa distinguibili a Clitoria arborea var. arborea foliolis acumi- natos, infra pubescentibus, calyce glabrato et vexillo subglabrato, et inflorescentia elongato. Stipa 16(4): 721-730. 1995 722 Sipa 16(4) Leaflets ovate-elliptic, acuminate with acumen 7-15 mm long, conspicu- ously pilose on nerves below. Inflorescence 10-30 cm long, primary branches 8-15 mm long that bear the pedicels. Calyx glabrate, appressed trichomes concentrated on ventral margin. Vexillum subglabrate with few appressed hairs along nerves and concentrated near the folded margin. Western Brazil. Type: BRAZIL. AMazonas: Mun. Sao Paulo de Olivencia near Palamares, Basin Rio Solimées, 11 Sep-26 Oct 1936, Krukoff 8208 (noLotyre: US 2169816; isoryprs: A, F 927714,G 295 & 296, MICH, MO 1175978 & 1250450, S, U 38199A. Paratypes: BRAZIL. AckE: near mouth Rio Macauhan, Basin Rio Purus, 27 Aug 1933, Krukoff 5719 (A, BM-2 sheets, F, G-2 sheets, LA, M, MICH, MO, NY-2 sheets, PR-2 sheets, 8, U, UC). These specimens have been identified by botanists frequently as Clitoria amazonumn Mart. ex Benth. Closer examination indicates that these specimens are distinquished easily from C. amazonum by (1) the smaller flowers, brac- teoles, calyx, gynoecium and androecium, (2) a different pubescence on the leaves, calyx and vexillum, and (3) the larger stipules, stipels and inflores- cence. Other botanists have reached the same conclusion, as evidenced by their annotations, but disagree upon the placement of this taxon. Specimens commonly have been assigned to Clitorta arborescens R. Br., Clitoria glaberrima Pittier, C/itoria javitensis (Kunth) Benth. or Clitoria racemosa Benth. {= Clitorta fairchildiana R.A. Howard}, but rarely co C. arborea. Occasionally, an annotation note has raised the question of a new species. These specimens agree with C. arborea in (1) habit, (2) inflorescence type, (3) flowers and its associated floral parts, and (4) legumes. There are too many characteristics in agreement with C. arborea to warrant its separation from this species. Variety arborea is distinguished from var. psewdoamazonica by the obtuse to abruptly short-acuminate leaflets (acumen 0.3—5 mm long) that are glabrate below, short inflorescences (2-7 cm long), calyces with appressed trichomes over its surface and a vexillum with moderately dense, appressed trichomes abaxially. 2. Clitoria epetiolata var. angustissima Fantz, var. nov. [Clitoria gayanensis (Aubl.) Benth. f. angustissima Hoehne, nom. in sched.} Varietas nova angustissima distinguibili a Clitoria epetiolata var. epetiolata Burkart foliolis angustatl. Leaflets 3-5 mm wide, lower surface bearing a few, scattered, appressed trichomes ca 0.5 mm long. Flowers 5.5—-6.5 cm long. Calyx tube 16-19 mm long, 6-9 mm wide at the throat, lobes 7—9 mm long. Bracteoles S—7 mm long. Cerrado and campos, Parana, Brazil. Type: BRAZIL. Parana: Jaguariahyva [= Jaguariaiva}, 5 Nov 1928, Hoehne 23404 (Ho- LOTYPE: GH). Hoehne was the first to recognized this taxon as distinct, named it angustissima, and assigned it as a form of C. gayanensis [= C. guianensis} in FANTZ, Taxonomic notes on Clitoria 723 annotations of herbarium specimens. Currently, no record has been found of the publication of this name. I used Hoehne’s name, changed the status to the rank of variety, and assigned it to C. epetiolata. Hoehne was unaware of C. epetiolata as Burkart (1949) did not circumscribed this closely related species until twenty years later. This variety is represented poorly in herbarium collections, but is dis- tinguished quickly by the narrower leaflets. Variety epetiolata (Paraguay and Corrientes, Argentina) has leaflets commonly 6—12(17) mm wide and bracteoles 7-12 mm long, with other characters similar to var. angustissima. Variety /atiuscula Burk. (Paraguay and Parana, Brazil) has leaflets 15-30 mm wide, sericeous below with trichomes ca | mm long, flowers 6—8.5 cm long with calyx tube (17)22—25 mm long, 13-15 mm wide at throat, lobes 11-13 mm long, and bracteoles 9-11 mm long. 3. Clitoria falcata Lam. This species is characterized as an herbaceous vine with white flowers (pale to deep yellowish in dried state) on elongate axillary peduncles from filiform, voluble, densely rufo-pilose stems. The petioles, rachis, and calyx surface commonly are rufo-pilose with leaves densely sericeous (thinning with age) beneath. Legumes are costate to ecostate. Table 1 contrasts the four varieties. Variety falcata has medium-sized flowers (3.5—5.5 cm) with larger leaflets 3-5.5 cm wide. The calyx tube is 10-16 mm long, moderately to densely pilose with lobes ovate-lanceolate, 2.54 mm wide. Bracteoles are ovate- lanceolate, acute, broadly cuneate below, 7-11 mm long, 3-4 mm wide. Legumes commonly are costate with a prominately raised medial costa ex- tending nearly the entire length of the valve, infrequently ecostate or weakly costate with a slightly elevated costa extending a 1/4—3/4 the length of the valve in f. heteromorpha (Griseb.) Fantz. This variety is widespread in the neotropics, from southern Mexico to Peru, east to Paraguay and southern Brazil, and introduced infrequently into Africa and recently documented (Guanes et al. 1990) from Taiwan. Variety glabrescens (Verdc.) Fantz is distinguished easily by the glabrescent appearance of the stem, petiole and calyx, lower leaf surface, and only ecostate fruits. Other characters agree with var. falcata. This variety is found in West Africa with limited collections cited by Fantz (1990) from Guadeloupe and Martinique (introductions?). Two new varieties are added as follows: 3a. Clitoria — var. aurantiaca (Benth.) Fantz, comb. nov. Clitoria g/yci- Ae DC. var. aurantiaca Benth., Mart. Fl. Bras. 15(1):119. 1862. Type: BRAZIL. ande: ne range flushed red, Fox 325 ee pe: K-hb. Hooker; photo of K: 8). oe sty DC. var. megapotamica Malm 2 Bot, 25a1o) 2) 151. Tyee: L. Rio Granbe Do Sur: Cruz Alto, 16 oe on Mare H:1102 (LECTOTYPE: S, 724 Sipa_ 16(4) mounted on two sheets). SYNryPES: BRAZIL. Minas Gerais: Caldas, 18 Jan 1860, Regnell I11:437 (S-2 sh.); 18xx, sa IIL: yee (S); 26 Dec 1862, Regnell III:437 (S). ees Sala Benth., Ann. Wiener Mus. Mee. 116. 1837; caged (Benth.) Benth., J. Proc. Linn. Soc. Bot. 2:39. 1858. Tyre: BRAZIL NAS GERAIS: in campis editis ad Tejuco et i do Principe, Martius s.n. lecisere: 12446). Vine, stem with upper poon conspicuously pubescent, trichomes dense, erect, rufus. Leaves conspicuously pubescent below, leaflets oblong, 3—-5.5 cm wide. Flowers (5.5) 6-7.5 cm long, vexillum white becoming dark yellowish-orange in the dried state. Calyx tube 16-20 mm long, laxly pi- lose with trichomes primarily along ventral and dorsal surfaces; lobes oblong, abruptly acuminate, broad, 4-7 mm wide. Bracteoles oblong, ob- tuse, widening above the middle, 7-11 mm long, 3-4 mm wide. Legume costate. Southern Brazil. Bentham (1862) published his varietal name with a question mark and a short diagnosis that included the orangish Howers with broad calyx lobes, two of the diagnostic characteristics. Bentham cited the Fox collection in Hooker's herbarium, but neglected to cite the collection number. Only one specimen was cited with a location of ee thus there is no doubt that Fox 325 is the holotype. Malme (1931) did not designate a type, but cited two collections (syntypes) without locality of deposit. Syntype Ma/me I[:1102 was selected as the lectotype because Malme cited it immediately follow- ing the diagnosis, and preceding a more detailed description. Syntype Regnell I11:437 was cited after a note that interpreted Bentham’s name as pub- lished with an incomplete description. Clitoria rufescens is rarely mentioned in the literature. Rizzini (1963) cited it as a Brazilian species, but did not examine any specimens of it. The type agrees with Fox 325 and the circumscription of this variety. 4b. Clitoria falcata var. boas (Rizzini) Fantz, stat. et comb. nov. ae hal gna ee ex Pers. f. /atifolia Rizzini, Arch. Jard. Bot. Rio de Janeiro 17:1 RAZ TL. Pe Mayan Tapajés, Cachoeira do Mangabal, 8 Feb 1 07 Dacbe 16740 (HOLOTYPE: RB 11861, mounted on two sheets; Isorypes: BM! 17171, non vidi). Vine, stem with upper portion conspicuously pubescent, trichomes dense, erect, rufous. Leaves conspicuously pubescent on lower surface, leaflets elliptic- oblong to oval, 5-8 cm wide. Flowers 3.5-4 cm long, white, drying pale yellow. Calyx tube 12-15 mm long, pilose; lobes ovate-lanceolate, acumi- nate, 24 mm wide. Bracteoles oblong-lanceolate, abruptly acuminate, 10— 15 mm long, 4-6 mm wide. Brazil: Para. Rizzini’s form is oe as a variety distinguished easily by the broader leaflets and longer bracteoles. Rizzini’s taxon Clitoria rubiginosa Juss. ex Pers. f. longifolia Rizz. (Arq. Jard. Bot. Rio de Janeiro 17:180. 1963) is part of the natural variation of the species, and is regarded as a synonym of var. falcata. Fantz, Taxonomic notes on Clitoria Tas_e 1. A comparison of the varieties of Clitorta falcata Lam. CHARACTER falcata glabrescens auvantiaca latifolia Pubescence Rufo-pilose Glabrate Rufo-pilose LEAFLETS Larger-width 3-5.5 cm 3-5.5 cm 3-5.5 cm Pubescense below Sericeous Glabrate Sericeous FLOWERS a: 3 n Color dried Pale to deep yellow Pale to deep yellow Dark yellowish- Rufo-pilose 5-8 cm Sericeous Pale to deep orange to orange yellow CALYX Tube length J-16 mm —16 mm 16-20 mm 2-15 mm Lobe width 2.54 mm 2.5—4 mm 4-7 mm 2.54 mm Pubescence Pilose Glabrate Lax Pilose Pilose BRACTEOLES Ovate-lanceolate Lanceolate to Oblong Ovate- ovate-lanceolate lanceolate Length 7-11 mm 7-11 mm 7-11 mm 10-15 mm Width 3-4 mm 3-4 mm 3-4 n 4-6 mm LEGUMES Costate Costate or ecostate Ecostate Costate Costate 4, Clitoria guianensis (Aubl.) Benth. This species is characterized as a subshrub with subsessile, 3-foliolate leaves with narrow elongated leaflets, axillary peduncles bearing a pair of large (5.5—7.5 cm) blue to lavender flowers, and costate legumes. Cleisto- gamy with reduced, apetalous flowers is common. Members are found from southern Mexico to Brazil, and western Cuba (Pinar del Rio, Isla de Pinos) in savannas, rocky cerrado, pine or pine-oak forests usually in dry sandy soil. Although well documented in collections, specimens with fruits borne from chasmogamous flowers are rare (2%, 8 of 347 collections), a pattern not observed in other species of C/itoria with cleistogamy. Two new variet- ies are noted below: 4a. Clitoria guianensis var. ee (Malme) Fantz, stat. nov. Clitoria chapadensis Malme 3:82. 1931; Clitoria guyanensis (Aubl.) Benth. f chapadensts (Male) Rizzini, gia Jard. os Rio de Janeiro 17:184. 1963. TyPE BRAZIL. Maro Grosso: Santa Anna de Chapada, 5 Aug 1902, Malme I1:2067 ; (LECTOTYPE: S; ISOLECTOTYPES: 8, 24 Jul 1902, Malme 2067; 2 Aug 1902, Malme 2067, 10 Aug 1902, Malme 2067d; 26 Sep 1902, Malme 2067e. Leaves 3-foliolate, occasionally 1-foliolate; leaflets shorter and broader (L/W ratio is 2—2.5:1), oblong-elliptic to oblong-obovate, 5.5—9 cm long, (2) 2.54 cm wae Petioles 3-6 mm long; rachis 2-5 mm long; stipules 7-12 mm long, 4—5 mm wide. Calyx tube of cleistogamous flowers 5—8 726 Sipa_ 16(4) mm long. Legumes costate; stipes 9-14 mm long. Brazil (Mato Grosso: Chapada dos Guimaraes). Malme (1931) cited the type as Ma/me 2067a-e. This group of specimens represents a series of plants collected from one generalized locality over a period of time. Thus, they are treated as syntypes. Malme 2067, 20672, and 2067e represent immature plants with small leaves and early flowering stages. Ma/me 2067c includes one plant with cleistogamous flowers and fruits. Ma/lme 2067d has more material, but lacks 1-foliolate leaves. Ma/me 20676 has the best material, represented well in flowering, with both |- and 3-foliolate leaves; thus it is selected as the lectotype. Vanlety gutanensis can be distinguished by the linear to lanceolate leaflets (7-20 cm long, 1—2.5 cm wide, with L/W ratio 3—6:1), smaller stipules (S—9 mm long, 2-4 mm wide), and elongate leaf axes (petiole 3-10 mm long; rachis 2-10 mm long). 4b. Clitoria guianensis var. macrocleistogama Fantz, var. nov. Varietas nova i cis la a ee cu Clitoria guianensis var. guianensis petiolo et rhachidi et stipice elongato et floribus cleistogamis cum macrocalycibus. Leaves 3-foliolate; —e linear to lanceolate, 7-16 cm long, 1-1.8 cm wide, with L/W ratio 3-8:1. Petioles 10-20 mm long; rachis 8-18 mm long; stipules 5-8 mm long, 2-3 mm wide. Calyx tube of cleistogamous flowers 9-12 mm long. Legumes weakly costate, the medial vein extend- ing nearly the entire length of the valve; stipes 14-20 mm long. Brazil: Minas Gerais. Type: BRAZIL. Minas Gerais: Serra da Caracol prope prodium Brata in campo alto audo preta, 20 Dec 1875, Mosen 4082 (HoLoryPE: S, mounted on two sheets). Variety gaianensis 1s distinguished by the smaller calyces of cleistoga- mous flowers (4-8 mm long), shorter leaf axes, and shorter fruit stipes (6— 14 mm long). ae os — ar. pedunculata me ex Benth.) Fantz, stat. NOV. Clitoria pedunculata Bojer ex Benth., Ann. Wiener Mus. Naturgesch. se va 1837 (non vee 1875); Ternatea a ‘Bojer ex Benth.) Kuntze, Revis. Gen. Pl. 1:210. 1891. Type: MADAGASCAR: inter frutices ad margines fluviorum, Mar em — v.77 (LECTOTYPE: W). Vine. Leaflets (3)5—9, 3-8 mm wide, upper surface micro-uncinate with scattered, short (O0.2—0.5 mm) subappressed, macrotrichomes. Peduncle elongate, arcuate, (15) 25-70 mm long; pedicel commonly solitary, borne laterally from peduncle apex, occasionally paired. Legume 3.5—6 cm long, pubescence erect to subappressed. Madagascar, with one collection from Mauritius. Bentham (1837) did not cite any collection with the original publica- tion of Bojer’s species; thus no holotype. However, a later publication pro- Fantz, Taxonomic notes on Clitoria T2i vided clues as to a probable type. Bentham (1858) cited one collection for the species as “Boyer (herbaria Musaei Vindobensis).” Descriptions published by Bentham consistently noted 3-foliolate leaves, an uncommon leaflet number in the species (typically 5-11-foliolate). Only Bojer v.77 (W) has some 3-foliolate leaves and bears the identification “C/itoria pedunculata mihi” (handwriting unknown; Bojer or Bentham?). Bojer v.77 (W) was a specimen utilized by Bentham in describing the plant, thus selected as the lectotype Variety heterophylla (common in Mauritius, infrequent in Madagascar) is distinguished easily by narrower leaflets (1-4 mm wide) glabrate above, shorter peduncles (3-10 mm long) with pedicels solitary, borne medially from the peduncle apex, and shorter legumes (2—3.7 cm long). 6. Clitoria leptostachya var. fruticosa Fantz, var. nov. Varietas nova fruticosa distinguibili a C/itoria cannes var. /eptostachya fruticosa cum apice interdum scandens, bracteola longis, et legumine angustato Shrub with apices occasionally climbing. Tele an midrib raised on upper surface. Bracteoles lanceolate, 3-4 mm long. Flowers rose to lilac. Legume 12-16 mm wide. Brazil (Amazonas, Para). Type: BRAZIL. ParA: prope medium fl. Tapejoz [Rio Tapajés], loco Quataguara, 15 Aug 1923, Ducke 17244 as = isorypes: G-hb. Barbey-Boissier (209 & 210), RB 17244 (two sheets), U 63828, US 1442540. Variety /eptostachya a is distinguished as a liana bearing pinkish- white to mauve-white flowers with a smaller bracteole (2 mm long), broader fruits (15-20 mm wide), and leaflets with the midrib weakly raised on the upper surface. 7. Clitoria mariana var. pubescentia Fantz, var. nov. BasionyM: Clitoria mariana var. mariana f. pubescentia Fantz, nom. in sched Varietas nova pubescentia distinguibili a Clitoria mariana var. mariana foliolis infra piloso- sericeis et in Florida meridionala endemica. Leaflets moderately to densely pilose-sericeous below, the trichomes con- spicuous, suberect to erect, thinning with age, but not widely scattered nor confined to veins; upper surface bearing inconspicuous micro-uncinate tri- chomes (vidi 20—30x) beneath conspicuous whitish macrotrichomes (0.3—1 mm long), trichomes falcate to subappressed, deciduous with age; stipules 2— 3 mm wide at the base. Inflorescence short, stout, |—4 cm long, 1-2-flowered. Calyx pubescence of micro-uncinate trichomes with moderate to scattered, subappressed to slightly spreading, 0.3-1 mm long macrotrichomes that are deciduous with age. Legume stipes from chasmogamous flowers 12—17 mm ek Endemic to central and southern Florida. ype: UNITED STATES. Fioripa Eee County}: near edge of Lake at Leesburg, 7 Jun eee Baltzell 120 (HOLOTYPE: FLA 34), 728 Stipa 16(4) Specimens of this variety had been annotated as a form of var. mariana because the distinction was based upon one character, an increase in pubes- cence. Populations from the coastal plains of other states (North Carolina to Texas) maintain their sparse degree of pubescence, similar to upland to mountainous members. When representatives of both varieties were grown together in the greenhouse, they retained their characteristic pubescence. Other botanists have encouraged me to treat this taxon as a variety because of its geographical isolation and the constancy in characters when trans- planted. Members of var. mariana are rarely found south (Alachua, Lake, Marion Counties) of an imaginary line across north-central Florida (Cedar Key to Ponte Verda Beach), but maintain their glabrate characteristic. Variety mariana is distinguished by its leaves being glabrate below, the trichomes subappressed and scattered when young, and confined to major veins in the mature state. The upper leaf surface lacks macrotrichomes. The macrotrichomes on the calyx are sparse to lacking. Florida populations are found in the panhandle and counties north of the line noted. Variety orientalis Fantz is endemic to Southeast Asia (Fantz and Predeep 1992), and distin- guished by elongate, more lax inflorescences (S—15 cm) with 2—6 flowers, shorter stipes (S—9 mm long), and broader stipules (3—S mm wide at base). 8. Clitoria polystachya Benth. This species is characterized as a —_ with axillary, multiflowered, pan- iculate inflorescences bearing small (2.5—3 cm long) white flowers or ecostate, strigose fruits that are weakly depressed between the seeds. Members range from southern Mexico to western Panama. Two new varieties are noted below: 8a. Clitoria polystachya var. congesta Fantz, var. nov. Varietas nova congesta distinguibili a Clitoria polystachya var. polystachya inflorescentia cum floribus congestis. Leaflet base rotund to subcordate. Stipules 6-12 mm long. Inflorescence contracted, 1-3 cm long; peduncle 1-2 cm long; flowers congested near the peduncle apex, pedicels appearing to be nearly fasciculate. Calyx tube —) mm long, lobes 2-3 mm long. Bracteoles 6-9 mm long. Endemic to District Temascaltepec, state of México, México. Type: MEXICO. Mexico: Dist. Temascaltepec, Rincén, 1960 m, 21 Aug 1933, Hinton 4447 (HOLOTYPE oa ISOTYPE: GH), PARATYPES: ME CO. Mexico: Dist. Temascaltepec, San Lucas, 24 Oct 1935, ple 8597 (F 1497 ey G-excluding vine around specimen, MO 1800932, NY, US 1979941,W 13453). This variety would be very ornamental if cultivated. It is recognized easily by the globular cluster of congested flowers on contracted inflores- FaNntz, Taxonomic notes on Clitoria 729 cences typically shorter than the petiole. The paratype collection has fruits borne from both chasmogamous and cleistogamous flowers. Variety polystachya has elongated inflorescences (3-16 cm) with flowers loosely ar- ranged and pedicels paired at the nodes, an elongated calyx tube (8-12 mm) and lobes (3-4 mm), shorter stipules (4-7 mm) and bracteoles (4—6 mm), and leaflets rotund, non subcordate. 8b. Clitoria polystachya var. pringlei Fantz, var. nov. Varietas nova pringlei distinguibili a C/itoria polystachya var. polystachya stipulis et bracteolis elongatis. Leaflet base rotund. Stipules 7-10 mm long. Inflorescence elongate, 3— 16 cm long; peduncle 1.5—7 cm long; flowers separated. Calyx tube 8-12 mm long, lobes 3—4 mm long. Bracteoles 6-8 mm long. Endemic to Morelos, México. Type: MEXICO. Moretos: Barranca near Cuernavaca, 4 Aug 1896, Pringle 7253 (HOLO- TYPE: GH; tsorype: MICH) Variety polystachya is distinguished by the smaller stipules (4—7 mm long) and bracteoles (4-6 mm long). Collections of C/storia from the state of Morelos are rare. Variety polystachya is well represented in herbaria from southern México, but has not been documented from Morelos. 9. Clitoria pozuzoensis Macbride var. schunkei Fantz, var. nov. Varietas nova schunkei distinguibili a Clitoria pozuzoensis var. pozuzoensis fruticosa cum apice scandentis, inflorescentia et pedicello elongat« Shrub with apices scandent. Inflorescence dieneste 8-19 cm long, weakly flexuous towards bite pedicels 10-12 mm. Peru (Loreto). Type: PERU. Loreto: en bosque bajo, a 1 km abajo de Iparia al noroeste 7 Rio Ucayali, 250m, ; Sep ee: Sinks 2773 (HoLoryPe: F 1688611; Isorypes: G-2 sheets, NY) This species is poorly represented in herbarium ieee. Variety pozuzoensis is distinguished as a liana with a short, weakly flexuous inflores- cence (1.5-6 cm long), pedicels that are 5-7 mm long, and has a broader distributional range (Peri: Huanuco, Loreto; Ecuador: Manabfé, Orientale. 10. Clitoria stipularis Benth. var. latifolia aes Pant stat. ne — stipularis Benth. f. latifolia Rizz., Arch. Jard. Bot. Rio de Janeiro 17:18 63. Type: BRAZIL. MaraNnuao: Rio Jurupy [Gurupi], Seti de Pirocana, 25 e on Col Ignotus 10387 (LectoryPE: RB 5697; IsoLecroryPes: BM, G-hb. Delessert, MG 10387 -non vidi, US 1044109). ie elliptic, obtuse becoming emarginate, more or less mucronate, 3.5-8 cm long; length/width ratio is 1.2-1.7:1. Stipules broadly ovate, 10-13 mm wide. Petiolules 3-4 mm long. Bracteoles 16-19 mm long. Brazil: Ceara, Maranhdao. Rizzini (1963) described a new form based upon broader obtuse leaflets. He cited one collection consisting of two specimens deposited at different 730 Sipa_ 16(4) institutions (MG,RB). These are regarded as syntypes since a holotype was not designated (Fantz 1993). I have not seen the MG collection (MG un- able to locate it, possibly distributed to another institution?), thus desig- nated the RB specimen as the lectotype. No collector was listed on the specimens examined (BM,G,RB,US) that were distributed from the Herbario do Museu Parense Goeldi. It may be a collection of Ducke. Rizzini cited Ducke 10387 (R 2435-non vidi) under f. stipularis, not the new form. Variety stipularis is distinguished by the oblong to oblong-elliptic leaflets, being longer (7-13 cm) and appearing to be narrower (length/width ratio = 2—-3:1), longer petiolules (4-8 mm) and bracteoles (17-24 mm), and narrower stipules (7—9 mm wide). This variety is found in Bahia, Brazil with one collection from Amazonas. Leaflets in var. /atifolia give the false appearance of being broader as widths are similar, but leaflet length is shorter than the typical variety, providing a smaller ratio of length to width. ACKNOWLEDGMENTS Thanks and appreciation are extended to George Wilson for providing the Spanish translation of the abstract, and to the curators of the following herbaria who provided the specimens for study: A, APCR, ARIZ, ASC, ASU, BM, BR, BRIT/SMU, CAL, CGE, CM, DUKE, E, E, FLAS, FSU, FTU, G, GH, HAL, K, LA, M, MG, MIN, MO, MPU, NCSC, NCU, NY, ORE, OS, OSC, P, PENN, PH, PR, RB, S, SI, SIU, TENN, TTC, U, UC, UMO, UNA, US, USCH, USEF, VEN, VPI, VSC, W, WIS. REFERENCES BentHam, G. 1837. Commentationes de Leguminosarum Generibus. Ann. Wiener Mus. Naturgesch. 2:111—120 BENTHAM, G. 1862. Chivea In: Martius & Eichler’s, Flora Brasiliensis 15(1):117—124. FE. Heo Miinchen. Sue. 1949. Nuevas especies de Leguminosas Subamericanus. Darwiniana 5:61—66. FANtTz, a rie Clitoria (Leguminosae) Antillarum. Moscosoa 6:152—166. Fanrz, PR. and S.V. Prepreep. 1992. Comments on four legumes (Clitoria, Centrosema) reported as occurring in India. Sida 15: Fantz, P.R. 1993. Notes on C/itoria (Leguminosae) in southeast Asia. Novon 3:352-255. Huana, T.-C., K.-C. Yana, and S.-F. Huanc. 1990. Notes on the flora of Taiwan (7) - three new eel Taiwania 35:1—6. Mame, G. 1931. Die Leguminoseen der weiten Regnellschen Reise; Plants collected 1901- 03 Rio ae do = and Matto Grosso, Brazil, and Mendoza, Argentina. Arch. Bot. 32A(4, no. 13):32- Rizzini, C. 1963. Clitoriae Brasiliensis (Leguminosae). Arch. Jard. Bot. Rio de Janeiro. 17:171-198. TWO NEW SPECIES OF ICHTHYOTHERE (HELIANTHEAE: ASTERACEAE) FROM ECUADOR AND PERU HAROLD ROBINSON Department of Botany National Museum of Natural History Smithsonian Institution Washington, DC 20560, U.S.A. ABSTRACT The tuberculate-fruited Ichthyothere pastazensis from Ecuador and the pinnately veined I. macdanielit from Peru are described as new. RESUMEN escriben como nuevas especies [chthyothere aerate con frutos tuberculados, de Ecuador y 1. macdamielii, de hojas pinnatinervias, de P Andean material of the genus Ichthyothere has been reviewed for the treatment of the Heliantheae in the Flora of Ecuador, and two undescribed species have been detected. Both new species are related to I. peruviana Poepp. but occur to the north of the range of that species. Ichthyothere pastazensis of eastern Ecuador has tuberculate achenes while 1, macdanielzi has pinnately veined leaves. Ichthyothere macdanielii H.Rob., sp. nov. (Figs. 1, 3, 6) Ad I. peruvianae affinis sed in foliis pinnate nervatis differt. Shrubs to 1.2 m high, moderately branched; stems blackish, hirsute with pale hairs, internodes 4-12 cm long, fistulose. Leaves opposite, petioles 0.5-1.0 cm long; leaf blades elliptical to obovate, thinly herbaceous, 9-21 cm long, 2.5—7.0 cm wide, base acute to cuneate, margins minutely den- ticulate to serrulate, apex shortly and broadly acuminate, upper surface with weak, long pilosity, lower surface pilosulous, especially on veins, with sparse, minute glandular dots; secondary veins pinnate, 5 or 6 on each side, arching but mostly ending below distal third of blade. Inflorescences ter- minal and sessile on leafy stems and branches, glomerate or with few short, dense seriate-cymes, branches covered with subulate bracteoles ca. 5 mm long. Heads sessile, subspherical to somewhat obovate, 4-6 mm high; outer large involucral bracts 2 or 3, subtending female florets, moderately con- cave, broadly oblong to suborbicular, 4-6 mm long and wide, broadly api- culate, with many stout, multiseriate hairs outside; inner bracts and pales ca. 30, subtending male florets, obovate, 3.5—-4.5 mm long, 2.5—3.5 mm Sipa 16(4): 731-736. 1995 732 Stipa 16(4) 1mm ] | lem 2 Fics. 1-5. [chthyothere, leaves, 1 cm scale, corollas, 1 mm scale. 1, 3. : macdantelti, 1, leaf showing pinnate venation, 2, corolla with short a and anthers. 2, 4. 1. pastazensis, leaf showing trinervate venation, 2, corolla showing short throat and anthers. 5. 1. peruviana, corolla showing long throat and anthers. 1. McDaniel et al. 27587 (US). 2, 4. Brandbyge & Asanza 30856 (US). 3. McDaniel & Rimachi 18978 (US). 5. Var ‘eas 11742 (US). wide, subangulate with scarious wings on upper margin, apex obtuse and apiculate. Female florets 2 or 3; corollas tubular, strongly bent to one side at base, ca. | mm long, with gland-tipped hairs and uniseriate and biseriate non-glandular hairs; style branches scarcely broadened. Male florets ca. 30; corollas whitish, 3.0—3.5 mm long, mostly glabrous, basal tube 1.2—1.5 mm long, throat, narrowly campanulate to nearly cylindrical, 1.3-1.5 mm long, with slender veins, lobes, ca. 0.7 mm long, with numerous glands outside; anther thecae 0.8-1.0 mm long. Achenes of female florets oblong to obovate, 4-5 mm long, 2.0—-2.5 mm wide, with 3 costae on inner and outer surfaces when mature, not tuberculate in material seen, glabrous. Pollen grains ca. 25 jam in diam. Type: PERU. Lorero: Dtto. Fernando Lores, Quebrada de Tamishiyacu, trail from Alianza toward Algodon, 6 Jul 1974, McDaniel & Rimachi 18978 (HoLotyrE: US: isorypEs: AMAZ, IBE, US, USM). Paratype: PERU. Lorero: Maynas. Dtto. Fernando Lores, Quebrada de Tamishacu, trail Rosinson, New species of Ichthyothere 733 PLANTS OF PERU sehthyeck ere sp. nov.? de S. McDaniel 1982 LORETO. . Fernando Lores. Quebrada ae Tantshiyacu, trail from sane Loverd a ore » Mature forest y uplands mn Ex. UNITED STATES 3032975 lehthyothe eis Jislebae sities MeDaniel 16978 NATIONAL HERBARIUM y & Manuel Rimach Fic. 6. Ichthyothere macdanielii H. Rob., holotype, United States National Herbarium. 734 Stipa 16(4) from Nueva Chachapoya to Rfo Manatf, 100-120 m, 7 Jan 1984, McDaniel, Rimachi & McMannes 87 (AMAZ, IBE, US, USM). The species was collected in mature “upland” forest over clay or chacra. The species distinctions are discussed below. Ichthyothere pastazensis H.Rob., sp. nov. (Figs. 2, 4, 7) Ad I. peruvianae affinis sed in floris masculinis minoribus et acheniis radiis tuberculatis differt Small shrubs ca. 0.7 m high, moderately branched; stems brownish, stri- ated, antrorsely pilose with pale hairs, narrowly fisculose. Leaves opposite, petioles 0.2—0.4 cm long; blades membranceous to thinly herbaceous when dry, ovate to broadly elliptical, 5-11 cm long, 2-5 cm wide, acute to shortly acuminate at base and apex, margins remotely, minutely denticulate, up- per surface pilose, lower surface with antrorsely appressed pilosity denser on veins, with many minute glandular dots; two pairs of ascending second- ary veins near base, pairs separated by ca. 0.5 cm, upper of the pairs reach- ing distal fifth of blade. Inflorescences terminal and sessile on leafy stems and branches, with few, short branches bearing heads sessile in dense seriate- cymes, bracteoles to 8 mm long, bracteoles at base of heads 2-3 mm long. Heads 4—6 mm high, subspherical; outer large involucral bracts 2 or 3, subtending female florets, broadly obovate, 4-5 mm long, ca.3 mm wide, apex shortly apiculate, outer surface with numerous stout, multiseriate hairs; inner bracts and pales ca. 20, subtending male florets, obovate, ca. 4 mm long, 3 mm wide, with scarious, denticulate angles on upper margins, apex shortly obtuse and apiculate, with few or no hairs outside. Female florets 2 or 3; corolla tubular, without evident basal bend, ca. 1 mm long, with dense brush of long trichomes, biseriate-stalked glands mixed with uniseriate and biseriate non-glandular hairs; style branches not or scarcely broadened. Male florets ca. 20; corollas yellowish green, ca. 3 mm long, tubes ca. 1.2 mm long, throats ca. 1.5 mm long, very narrowly funnelform to nearly cylindrical, with narrow veins, lobes ca. 0.7 mm long and wide, with nu- merous glandular dots outside; anther thecae ca. 1 mm long. Achenes of female florets obovate, ca. 5 mm long, 3 mm wide, surface strongly tuber- culate with age on margins and costae, glabrous, without striations or fur- rows. Pollen san ca. 30 pm in diam. Type: ECUADOR TAZA: Lorocachi, on path to Lagartococha, 01°38'S, 75°58'W, 200 m, 25 May 1980, pes & Asanza 30856 (HoLoryre: US; isoryre: AAU). The species is known only from the type collected in a wet primary rain forest. The related Ichthyothere peruviana is distributed from San Martin south- ward to Cuzco, Madre de Dios and Puno in Peru. The more typical north- RosiNnson, New species of Ichthyothere 735 ais . rs = v. oa vl—e — — gf 5 sle=: — £16 _3 bal ao FLORA OF ECUADOR sai pie oe 30856 Compositae & Endt.) Baker | of + y { lehthy There pes fozensis Heh det. H. Robinson Province PASTAZA i : ; I tole ype P UNITED STATES Hype (75° 58° W 01° 38°S) Al. 200 m. 25 May 1980 3206230 i NATIONAL HERBARIUM Botankal Insitute, Aarbus University, Denmark (AAU) Fic. 7. Ichthyothere pastazensis H. Rob., holotype, United States National Herbarium. 736 Sipa_ 16(4) ern material has its lower leaves with longer petioles and longer acuminate leaf blade bases. The specimens of the species vary in the hirsute pubescence of the stems and inflorescence. N heless, all elements recognized here in L. permviana have distinctly broadened ike branches in the ray florets, long and slender throats on the male corollas, 2.5—3.0 mm long, and slender anther thecae, nearly 2 mm long (Fig. 5). The leaf tips are usually narrowly acuminate. The details of the northern element of 1. perwviana are represented well in the plate by Poeppig (1843, pl. 252), the long petioles of the lower leaves, long lower secondary veins of the leaf, elongate anther thecae, even distribution of hairs on the ray corolla, and broad branches of the ray style. The two new species. what disjunct to the north of lchthyothere peruviana but have all the leaves with shorter petioles as in the southern material of peruviana. Both new species have acute to slightly acuminate leaf tips, longer hairs on the ray corollas concentrated on the distal half, style branches of the rays scarcely broadened, shorter disk corolla throats, ca. 1.5 mm, and shorter anther thecae, 0.7-1.0 mm (Figs. 3, 4). The species from Loreto, Peru, is particularly unusual in the genus by the pinnate leaf venation (Fig. 1). Other species of Ichthyothere, including 1. pernviana and 1. pastazensis (Fig. 2), have leaves that are obviously tri- to quinque-nervate or have lower veins that spread at the base but arch upward to near the leaf apex. The pilosity on the upper leaf surface of 1. macdanielit 1s longer than that of related species, and the female corolla is strongly bent at the base. The type of 1. macdanielii has unusually long cuneate bases on some leaf blades, but such bases occur in only some leaves of the type and are lacking in the paratype. From near the Peruvian border in Ecuador, lchthyothere pastazensis has smaller leaves (Fig. 2), more branched plants, and the mature achenes are strongly tu- berculate. The pressed specimens of the Ecuadorian plant show a number of branches, while the longer segments of 1. macdanielit and I. peruviana on the herbarium sheets rarely show any branching. The tuberculate condition of the achenes in I. pastazensis develops only at maturity, but such tuberculae have not been seen in any apparently mature achenes of either 1. macdanielii or I. peruviana. ACKNOWLEDGMENTS Sidney McDaniel, of Mississippi State University, has furnished infor- mation on the distribution of his collections. Photographs by John Steiner, Staff Photographer, National Museum of Natural History. REFERENCE PorppiG, E. 1843. Nova genera ac species plantarum 3(5—6):33—52, pl. 241-260. CLUSIA FABIOLAE, A NEW SPECIES, WITH A SYNOPSIS OF CLUSIA SECTION ANANDROGYNE (CLUSIACEAE) IN GUAYANA JOHN J. PIPOLY II Botanical Research Institute of Texas 509 Pecan Street Fort Worth, TX 76102-4060, U.S.A. ABSTRACT In preparation of the treatment of Clusia section Anandrogyne tor Flora of the Venezuelan Guayana, Flora of the Guianas, and Flora of Central French Guiana, the section is lectotypified and its description is emended. A key to the species, along with updated descriptions with citation of types and representative specimens examined, are also provided for each spe- cies, concomitant with discussions of distribution, ecology and conservation status. Dis- cussions of postulated phylogenetic relationships are presented for ac taxon. Clusia fabiolae is described and illustrated. Clusia scariosa is relegated to >a under Clusia pachyphylla. Clusia duartei is placed in the section for the first time, based on new data from flowering material, and C. savannarum is newly recorded for Venezuela. Clusia sessilis is a homonym and a nomen novum, Clusia wurdackiana, is provided for it. RESUMEN Al preparar un tratamiento taxondmico del género Clusia secc. Anandrogyne para las floras de la Guayana Venezolana, de las Guayanas, y de la Guayana Francesa Central, se lectotipifica la seccién y se amplia su descripcién. Ademias, se ofrece una clave para distinguir — as especies, junto con descripciones actualizadas, tipos y especimenes citados. Se discute el parentesco, la distribucién geografica, ecologfa y estatus en cuanto a conservaciOn para cada especie. Finalmente, se discute lo que sabe de la filogenia de dichas especies. Se de- scribe, se ilustra y se discute el parentesco de Clusia fabiolae. Se relega C. scariosa a la sinonimia bajo C. pachyphylla. Se ubica Clusia duartei en la seccién por la primera vez, en base a nuevo material en flor, y se cita C. savannarum como nuevo para Venezuela. Clusia sessilis es un ee y por lo tanto, se ofrece para ella un nombre nuevo, Clusia wurdackiana. INTRODUCTION In preparation for a forthcoming treatment of the genua Clusia for the Flora of the Venezuelan Guayana, Flora of the Guianas, and Flora of Cen- tral French Guiana, it became clear that a revision of section Avandrogyne was needed. Overall, the section consists of 69 species, of which 19 are as yet undescribed, and is defined by the largely anantherous staminodes of the pistillate flowers and the pluriseriate, acropetally longer stamens of the staminate flowers with anthers dehiscent by wide longitudinal slits. This section, the “Clusia multiflora Group” of Hammel (1986), is the largest and Sipa 16(4): 737-755. 1995 738 Stipa 16(4) most taxonomically complex of all infrageneric groups within the genus. While apomixis has not been proven for the group, several individuals of a Clusia elliptica H.B.K. population in northern Colombia have been seen with young fruit forming before the petals have completely opened, imply- ing that apomixis may occur. As a result of the present study, seven species are now known from the Guayana Floristic Province (sensu Maguire 1979). TAXONOMIC TREATMENT Clusia s oo a Planchon & Triana, Ann. Sci. Nat. 2 sér 14 1860. A. Engler, Nat. Pflanzenfam. BG): EN 1895, LECTOTYPE am i diccinoteal) Clusia multiflora H. B. K., Nov. Gen. Sp. 5:200. 1822. Clusia section Crinva Bentham & Hooker eae Anandrogyne (Planchon & Triana) Engler in Martius, Fl. Bras. (1):402. Cua ae Thysanoclusia Vesque section ae ne (Planchon & Triana) Vesque, in A. DC. & DC. Monogr. ee 8:29. 1893. ie shrubs or trees; latex white. Leaves subsessile or petiolate; petioles canaliculate and/or marginate, usually with prominent adaxial margined pits basally. Staminate inflorescence a terminal cyme or panicle of cymes, subtended by foliaceous bracts, the bracts early caducous or rarely, persistent (C. wardackiana), peduncle tetragonal; secondary inflorescence racts 2, cartilaginous or coriaceous; floral bracts 2, cartilaginous; pedicels tetragonal or angulate; bracteoles 2—6, cartilaginous. Staminate flowers pink, yellow or white; sepals (4-)6—8(-10), cartilaginous, the outer opposite, the inner decussate or contorted; petals (4—)6—8(—10), carnose, coriaceous or membranaceous, the outer decussate, the inner decussate and/or contorted, sometimes progressively larger acropetally; stamens numerous, the nee poorly developed, cubic or hemispherical, the filaments apically free, basally connate, linear, flattened, coriaceous, the anthers muticous, extrorsely or latrorsely dehiscent by wide longitudinal slits, the thecae lin- ear or ovate; pistillode absent. Pistillate Howers as in staminate but staminodes 4—numerous, linear, anantherous or rarely with ovate antherodes (sterile anthers), producing mostly abortive pollen or sometimes ier sterile, sometimes early caducous; pistil ovoid or subglobose; carpels (4— )5-10(-1 2); styles obsolete or equalling carpels in number; stigmas cunei- form or corniculoid. Fruit ovoid or subglobose, rarely urceolate, the stig- mas and styles persistent. Distribution.—Sixty-nine species, throughout the Neotropics, but with centers of diversity in the Andes of Colombia and Peru, principally in premontane and montane moist, wet or pluvial forests. Clusia section Anandrogyne is one of the most difficult groups within the genus, not only because of the lack of adequate collections for many of the Andean taxa, but also because of the tendency for the staminodes to be Pipoty, Clusia section Anandrogyne ia? early caducous. Staminodia frequently begin to fall out from beneath the sepals once they become reflexed, even if the sutures of the capsule have not begun to open. When sterile or without staminodia, some members of section Anandrogyne are virtually indistinguishable from other species in sections Polythecandra or Retinostemon. In addition, many taxa are known only from mature fruit, and have been placed in the section because no staminodes are known. Likewise, for those taxa, we do not know if staminodes were “lost” or if they ever occurred, thus increasing the prob- ability chat the section is paraphyletic as currently defined Members of the group are usually locally common. My field experience has shown that they often occur in numbers exceeding 30 individuals/ha. Therefore, given their ecological importance in the C/wsia scrub forests of many tepuis and the thickets of the Andean subparamo life zone, a com- prehensive systematic revision of the group throughout its entire geographic range is needed. The present treatment is designed to provide descriptions, specimen citations, and explain concepts thus far established in detail not possible for the aforementioned floristic treatments, until a more compre- hensive monograph can be undertaken. Morphological terminology fol- lows Pipoly and Graff (1995). Clusia multiflora H. B. K. was chosen as the lectotype species instead of Clusia ducu Bentham, the other taxon mentioned in the protologue, be- cause C. multiflora more adequately fits the most narrow interpretation of the section Anandrogyne concept, with its large flowers, 4—S-merous calyx and 5—6-merous corolla, the prominent androphore, and 5 anantherous staminodia. KEY TO GUAYANA SPECIES OF CLUSIA SECTION ANANDROGYNE 1. Branchlets 2—7(—8) mm diam., pedicels thin, 2.5—6.5 mm long; stigmas cuneiform. 2. Branchlets verrucose, decussately tetragonal, 2-3 mm diam., not semisucculent; leaf blades (2—)3.5 cm long, (1.5— : 53.3 cm wide, ie bases cordate; sepals 5 _C. fabiolae Branchlets smooth, simply tetragonal, or eenaes cage eee (-8) mm diam., semisucculent; leaf blades 5—9 an long, G.4 oe 8 cm ae the bases truncate or broadly acute; sepals - 3. Petioles without conspicuous adaxial eae pits; leaf bases subacute; peduncle obsolete to 1 cm long; petals 5; staminodia numerous, a caducous. _ C. phelpsiae Petioles with large adaxial margined pits; leaf bases rounded to trun- — No Qo cate; peduncle (1.2—)2.5-5 cm long; petals 4; staminodia 4—10, per- sistent in fruit. 4. Inflorescence 3—4 times compound, 9—12-flowered; antherodes producing small amounts of pollen; styles obsolete; fruit ovoid; at margins of scrub forests with small savannas, in open, rocky areas, 1,250—2,000 m, southern Amazonas, Venezuela ..........0000..... 3. C. rotundifolia 740 Stipa 16(4) Inflorescence ee 1—3-flowered; antherodes without pollen; styles un, I-1.5 mm long; fruit subglobose; shrub islands in savannas, 460-1,000 m, eastern Bolt ivar, Venezuela and adjacent Guyana. 4. C. savannarum 1. Branchlets (8—)8.5—15(—22) mm diam.; pedicels thick, 7-15 mm long; deen corniculoic 5.1 Leaf blades aa iceous to coriaceous, obovate to oblanceolate, the bases acute, “ margins flat; sepals stiffly coriaceous, the margins entire; styles thick, radiate, 3-5 mm long at maturity; ovary and fruit smooth, not sulcate 5. C. wurdackiana 5. Leaf blades cartilaginous, elliptic to suborbicular, the bases obtuse to trun- cate, sepals chartaceous, the margin variously incised; styles obsolete; ovary and fruit deeply sulcate. ». Branchlets 8-11 mm diam.; petioles 1—-1.5(2) cm long; leaf blades with secondary veins ae raised above and below; floral bracts 4, 0.8-1.2 cm long. 6. C. pachyphylla 6. Braarh ees 15-22 mm diam.; petioles 2—3 cm long; leaf blades with secondary veins inconspicuous above and below; floral bracts 2, 1.5—3 cm long. 7. C. duartei I. Clusia fabiolae Pipoly, sp. nov. (Fig. 1) Ob ramulos angulatos, pedicelos usque ad 6.5 mm longos, laminas cartilagineas, inflorescentiam terminalem necnon stigmates cuneiformes C. rotundifoliae valde arcte affinis, sed ea ramulis tetragonis (non sulcatis) 2—4 (nec 5—8) mm diametris, lamina ad basem cordata (non rotundata vel truncata), secus mz as plana vel subrevoluta (nec manifeste revoluta), pedunculo obsoleto vel usque ad 1 (non 2.5—5) cm longo, denique fructu globoso (non ovoideo) statim separablis Terrestrial, ee shrub or small tree to 4(—7) m tall. Branchlets decussately tetragonal, with a few subacute angles, 2—3 mm diam., verru- cose, ultimately terete with age, glabrous. Leaves subsessile; ilades carti- laginous, ovate to oblong, (2—)3.5-4.5 mm long, (1.5—)2.5—3.3 mm wide, symmetric, apex obtuse, base cordate, adaxially nitid, eee dull, with conspicuous, black linear latex canals, the midrib prominently raised and the secondary veins not visible above and below, the margin translucent, thick, flat or rarely subrevolute when dried; petioles very small, 3—5 mm long, deeply canaliculate, verrucose when dried, the adaxial pits broadly margined.. Staminate inflorescence a terminal panicle of 3-flowered cymes, 2—3 cm long and wide; peduncle tetragonal, obsolete to 1 cm long; second- ary inflorescence bracts 2, opposite, coriaceous, oblate, 3—3.2 mm long, 4.24.3 mm wide, apex obtuse, carinate, medially crassate and somewhat verrucose, the margin scarious, ca. 0.3 mm wide, entire; pedicels tetragonal, like branchlets, 2.5—5.5(—6.5) mm long, glabrous; bracteoles 2, opposite, coriaceous, suborbicular, 5.4—5.6 mm long, 5.6-5.8 mm wide, apex rounded, medially crassate, carinate, verrucose, without latex canals, che margin undulate, incised, scarious, the margin ca. 0.2 mm wide. Flowers pink; sepals 5, the outer whorl 2, decussate to the bracteoles, subcoriaceous, oblate, Pipoty, Clusia section Anandrogyne 741 Fic. 1. Clusia fabiolae Pipoly. A. Habit. B. Cyme. C. Bracteoles and flower bud. D. Petals of staminate flower. E. Sepals. F. Androphore and androecium. G. Calyx and fruit. A-B, drawn from Steyermark 93190 (staminate); C-E, drawn from Liesner 21021 (staminate); F, drawn from Steyermark 128298 (pistillate). 742 Stipa 16(4) 5.96.1 mm long, 7—7.2 mm wide, apex rounded, carinate, verrucose, with brown latex canals conspicuous subapically, the margin subentire, hyaline, scarious, with a few incisions, glabrous, the inner whorl 3, contorted, chartaceous, suborbicular, 5.5-8 mm long, 6-8.5 mm wide, apex rounded, medially crassate, hyaline, wich brown latex canals conspicuous everyhwere but the margin, the margin subentire, hyaline, scarious, with a few inci- sions, glabrous; petals 4, contorted, membranaceous, a abruptly obovate-spathulate, the proximal portion 2.1—2.2 mm long, 2.9-3 mm wide, the distal portion suborbicular, 6.6-6.7 mm long and wide, apex rounded, flat, the margin undulate, entire, glabrous; receptacle concave, without resin; stamens numerous, more than 60, the androphore hemi- spherical, 0.4—0.5 mm tall, 2.3-2.4 mm diam., the distal ca. 2—3 mm long, progressively longer acropetally to 5.5(—6) mm long, the filaments linear, flat, as wide as the anthers, 1-3 mm long, 0.3-0.5 mm wide, the anthers muticous, 0.7-0.8 mm long, 0.5—-0.6 mm wide, apex emarginate, base cordate, dehiscent by broad longitudinal slits running entire length, the connective medially darkened on both sides; pistillode absent. Pistil- late inflorescence as in staminate but secondary inflorescence bracts very widely ovate, 3.8-3.9 mm long, 4-5 mm wide, apex short-acuminate, medially crassate, carinate, rugose, the margin hyaline, entire, glabrous. Flowers; sepals 4, the outer pair decussate to the bracts, chartaceous, subor- bicular, 6—6.2 mm long, 5.5-5.7 mm wide, asymmetric, hyaline, apex rounded, medially crassate, rugose, carinate, cucullate, latex canals incon- spicuous, the margin hyaline, scarious, ca. 0.2 mm wide, with 1—2 inci- sions, glabrous, the inner pair decussate, chartaceous, widely ovate, 5.9— 6.1 mm long, 5—5.2 mm wide, symmetric, hyaline, apex obtuse, medially Crassate, rugose, carinate, cucullate, latex canals inconspicuous, the margin scarious, hyaline, ca. 0.1 mm wide, with 1 incision, glabrous; petals 4, contorted, thinly carnose, widely obovate, 6.4—6.6 mm long, 4.8-5 mm wide, apex rounded, — rugose, flat, the margin somewhat translu- cent, flat, entire; staminodes 4, flat, linear, anantherous, 2.3—2.5 mm long, 0.8-1.1 mm wide, persistent in early fruit; pistil ovoid, 5556 mm long, 4—4.5 mm diam.; carpels 5; styles erect, short, ca. 0.3—-0.5 mm long, the stigmas sessile, cuneiform, 1.2-1.5 mm long, 0.9-1 mm wide, apex acute, base obtuse, attached toward base and apex. Fruit globose, 1-1.2 cm long, 0.9—1 cm diam. — Type: VENEZUELA. Bolivar: Deco. Piar, ce ee tepui, summit; fourth of four tepuis from W to E in Aparamdn range; 5°53'N, 61°59'W; highly eroded tepui summit with deep canyon S to 100 m; 26 Mar 1987 ole ms B. Holst 3626 (HOLOTYPE: VEN; isoryPes: BRIT, F, see , US). — VENEZ LA, Bolivar: Abacapa-tepui, Macizo del Chimanta, 2,125—2,300 m, 13 Apr 1953 (F . aN MO, NY, US, VEN); Amuri-tepur, Macizo del Chimanta, Pipoty, Clusia section Anandrogyne 743 section W of Acopdn-tepui, 5°10'N, 62°07'W, 1,850 m, 2-5 Feb 1991 (pist. fl, preco- cious), J. Steyermark et al. 128555 (MO, VEN), Apacara- Eile E-C section, 21-22 June 1953 (pist. fl, fr), J. Steyermark 75935 (F-2 shts, NY, US, VEN); Apacara-tepuf, Macizo del Chimanta, : base, 5°20'N, 62°12'W, 2,200 m, 30 Jan—1 ae 1983 (fr), J. Steyermark 128298 VEN); Auyan-tepuf, above Salto La ne C section, E of Rio Churim, 1,950 m, 27-28 Dec 1977 (fr), C. Brewer-Carias s.n. (MO, VEN); Auyan-tepui, summit, 1800 m, 5 Feb 1988 (fr), FE Delascio & R. Lopez 13573 G VEN); Auyaén-tepui, summit, SE, between “Boggy Camp” & “Oso Woods Camp”, 2,200 m, 1 May 1964 (stam. fl), J. Steyermark ke (F, NY, US, VEN); summit, C portion of NW branch, near Jimmy Angel airplane, 800 m, 7 May 1964 (stam. fl), J. Steyermark 93479 (EK, NY, US, VEN); summit Auyan- {, SE section, 5°42'N, 62°26'W, 2,140 m, 26 Feb 1978 (stam. fl), J. Steyermark s al, ae (EF, MO, VEN). Dtto. Piar, Camarcaibarai-tepuf, E of Auydn-tepui, 2,400—2,50( 25-26 Mar 1987 (pist. fl, fr), EF Delascto 13111 (MO, US, VEN); C & W part of Sih = tween Camarcaibarai-tepuf and Tereké-Yurén-tepuf, 1,800-1,900 m, 23 May 1986 (stam. fl), R. Liesner & B. Holst 21021 (COL, F, MO, US, VEN); summit of Tereké-Yurén, W edge, 5°52'N, 62°02'W, 2,135 m, 26 May 1986 (pist. fl), R. Liesner et al. 21066 (MO, US, VEN), (stam. fl), Liesner et al, 21070 (MO, US, VEN); Central section, Macizo del Chimanta, 1940 m, 4 Feb 1955 (fr), J. Steyermark & J. Wurdack 414 (F, NY, VEN); Murey-(Euroda)- tepuf, NE sector of Macizo del Chimantd, 5°23'N, 62°03'W, 2 — m, 1-3 Apr 1989 (scam. fl), J. Pruski & O. Huber 3588 (NY, VEN). Ptari-tepuf, summit, 5°45'N, 61°45'W, N of Sta. Teresita de Kavanayén, 2,360—2,420 m, J. Steyermark et o 115697 (MO, US, VEN) Distribution —Endemic to the tepuis of the southeastern portion of Bolivar state, Venezuela, at 1,600—2,500 m elevation. Ecology and conservation status.—Clusia fabiolae grows in rocky savannas below large rock walls. It is locally common, and is a conspicuous element of the vegetation. While its range is extremely restricted, most of it is protected and thus, C. fabiolae ee not to > be threatened at this time. Fabiola Mone -Pipoly (née Monje Silva), who, with the late Mario Monee, were instrumental in funding Venezuelan botanical projects while employed at Seguros Anauco, S. A., in Caracas. She has accompanied me in the field on numerous occasions and provides critical moral support, despite my frequent absences to pursue fieldwork. The angulate branchlets, compound inflorescences, thin pedicels, cunei- form stigmas and cartilaginous leaves of Clusia fabiolae indicate a close re- lationship with C. rotundifolia. However, the thinner, decussately tetrago- nal, non-succulent branchlets, leaf blades with cordate bases and flat or subrevolute margins, shorter or obsolete peduncle and globose fruit imme- diately separate Clusia fabiolae froom C. rotundifolia. The Tereké-Yurén popu- lations of Clusia fabiolae are diminutive compared to those of Auyan-tepui and the Chimantd-Massif, but are qualitatively identical. 2. Clusia phelpsiae Lasser & Maguire, Brittonia 7:81. 1950. Type: VEN- EZUELA. Amazonas: Cerro Yavi, 1,850—2,200 m, 1-3 Mar 1947 (fr), W. Phelps & C. Hitchcock 69 ——— NY!; tsorype: VEN). 744 Sipa_ 16(4) > Small tree, co 3 m tall. Branchlets subtetragonal, semisucculent, deeply sulcate when dried, 5-7 mm diam., glabrous. Leaves subsessile; blades co- riaceous, obovate, 5-9 cm long, - 56 cm wide, apex broadly rounded, base subacute, midrib prominently raised above and below, secondary veins eon numerous, prominulous above and below, the submarginal vein inconspicu- ous, the margin thick, translucent, entire; petioles obsolete to 5 mm long, without adaxial margined pits. Inflorescence terminal, cymose, 3-flowered; peduncle obsolete to 1 cm long; floral bracts early caducous; pedicels obso- lete co 5 mm long. Flowers yellow; sepals 4, decussate, coriaceous, widely ovate, 6-8 mm long, 4-6 mm wide, apex broadly rounded, medially cari- nate, the margin entire, opaque; petals 5, carnose, obovate-spathulate, 12— 15 mm long, 6-8 mm wide, apex broadly rounded, the margin subentire; staminodia numerous, early caducous; pistil ovoid, the carpels 5; styles thick, 5 mm long; stigmas sessile, cuneiform, 2.5 mm long. Fruit urceolate, 2.5 cm long, 1.5 cm wide. Distribution.—Known only from Cerro Yavi, state of Amazonas, Ven- ezuela, at 1,850—2,200 m elevation. Ecology and conservation status.—Clusia phelpsiae is a rare species, occuring only near the summit of Cerro Yavi. It is from one of the more inhabited areas of the Guayana in Amazonas, and therefore should be considered threat- ened. PARATYPE examined: VENEZUELA. poe ee Cerro Yavi, 1,850—2,200 m, 1—3 Mar 1947 (pist. fl), Wo Phelps & C. Hitchcock 69 (NY, Clusia phelpsiae is similar to, and may be oe related to C. rotundtfolia, but the broadly acute leaf bases, obsolete peduncle, higher floral merosity, early caducous staminodia, obsolete styles and larger, urceolate fruit, all allow easy recognition of the species. Clusia phelpsiae should not be con- fused with C. phelpsiana Maguire, of section Clusiastrum. 3. Clusia rotundifolia Gleason, Bull. mee Bot. Club 28:406. 1931. Tyee: VENEZUELA. Amazonas: Cerro Duida, Savanna Hills, 1,341 m; Aug 1928—Apr 1929 (stam. fl), G. Tate 798 (HOLOTYPE: NY!; ISOTYPE: FI), Tree to 3 m tall. Branchlets tetragonal, the angles acute but not winged, 5-8 mm diam., semisucculent, deeply sulcate. Leaves subsessile; blades cartilaginous, suborbicular, to very broadly obovate, rarely oblong, 6—8(— 9) cm long, (4—)6-8 cm wide, apex broadly rounded, truncate or rarely retuse above, base truncate, midrib slightly elevated above and below, the secondary veins numerous, the collecting vein 1-2 mm from margin, the margin revolute. Staminate inflorescence a compound cyme, each branch bearing 3 flowers; peduncle tetragonal, 3-4 cm long, deeply sulcate; sec- ondary inflorescence bracts cartilaginous, oblate, 2-3 mm long, 4-6 mm wide, apex broadly rounded, prominently carinate, the margin opaque, not Pipoty, Clusia section Anandrogyne 745 scarious, entire; secondary inflorescence bracts cartilaginous, oblate, 1—2 mm long, 3-4 mm wide, apex broadly rounded, prominently carinate me- dially; pedicels 34.5 mm long; bracteoles 6, decussate, cartilaginous, the outer pair widely triangular, 1.5-2 mm long, 3-4.5 mm wide, apex sub- acute, the following pair stiffly coriaceous, suborbicular, 2.5—-3.5 mm long and wide, apex broadly rounded, the margin thin but not scarious, entire; the innermost pair stiffly coriaceous, suborbicular to oblate, 34 mm long, 4—5 mm wide, apex broadly rounded, the margin scarious, entire. Stam1i- nate flowers whitish cream; sepals 4, cartilaginous, the outer pair opposite, suborbicular, 5-7 mm long and wide, apex broadly rounded, the margin hyaline, scarious, entire, the inner pair decussate, obovate, 6-8 mm long, —6 mm wide, apex broadly rounded, the margin scarious, irregularly notched; petals 4, decussate, carnose, obovate, 9-1.1 mm long, 5—G6 mm wide, apex truncate, irregularly notched, somewhat cucullate, gradually narrowed to 3 mm wide basally; stamens numerous, 34.5 mm long, the androphore cubic, 1-1.2 mm diam., the filaments linear, fleshy, 1-1.5 mm long, the anthers linear, 1-1.5 mm long, the apex muticous, latrorsely dehiscent by wide longitudinal slits; piscillode absent. Pistillate inflores- cence; as in staminate but peduncle (5S—)25-45 mm long; primary bracts 10-12 mm long, 12-14 mm wide; secondary bracts 11-12 mm long, 12— 13 mm wide; pedicels subobsolete to 8 mm long; floral bracts 6, as in staminate but the outer 1.3—1.5 mm long, 3.5-4 mm wide; the following pair 3—3.2 mm long and wide; the innermost pair 3—3.2 mm long, 4—4.2 mm wide. Pistillate flowers; sepals 4, as in the staminate but 7—7.3 mm long and wide, the inner ones 7.5—7.7 mm long, 6-6.2 mm wide; petals 4, obovate, 10-12 mm long, 6-6.5 mm wide, gradually tapering to 3 mm wide at base; staminodes 16, 3—3.5 mm long, the filaments basally fused to 0.3 mm, apically free 1.3-1.5 mm, the antherodes producing small amounts of pollen, ovate, 1.5—1.7 mm long, the apex obtuse, latrorsely dehiscent by wide longitudinal slits; pistil ovoid; carpels 4; stigmas cunieform, sessile, the angles rounded, 1.3-1.5 mm long and wide. Fruit ovoid, 1.2—-1.5 cm long, 2.3-2.5 cm wide. Distribution.—Endemic to Cerro Duida, Cerro Parti, and Sierra Parima, the area between tributaries of the Rio Ventuari and Rio Orinoco, between the Rio Paru to the north and the Rio Padamo to the south, with a disjunct population on Sierra Parima, Amazonas, Venezuela, at 1,250—2,000 m. Ecology and conservation status.—Clusia rotundifolia occurs at the margins of scrub forests with small savannas, on rather open, rocky areas. The popu- lation density and pressure to cultivate are extremely low in the area, so the a is not considered threatened. Specimens examined: VENEZUELA. Amazonas: Depto. Atabapo, Plateau of Cerro ene pe Culebra, 03°36'N, 65°42'W, 1,250 m, 2 Mar 1985 (stam. fl), R. Liesner 746 Siwa 16(4) 18168 (MO, US, VEN); Cerro Duida, Rio Cunucunuma, ridge W of Cafio Culebra, 1,800 m, 22 Nov 1950 (fr), B. Maguire et al. 29658 (F, MO, NY, US, VEN); Cerro Duida, N Sela 1,600 m, 23 Nov 1950 (stam. fl), B. Maguire et a 29659-A (MO, NY, VEN); a, headwaters of Rio Matacuni, along the Venezuelan-Brazilian border, No. 7, 04°05" " 64°24 W, 1,500 m, 19 May 1973 (fr), J. Steyermark et al. 107525 (BRIT, MO, NY, US, VEN); Serrania Pari, Rio Pari, Cafio Asisa, Rio Ventuari, 12 km N along W rim, 2,000 m, 4 Feb 1951 (fr), R. Cowan & J. Wurdack M31259 (MO, NY, US, VEN); Depto. Atures, Serrania Pard, Central Plain, SW sector, 04°25'N, 65°32'W, 1,200-1,250 m, 5-7 Mar 1991 (fr), P. Berry et al. 4995 (MYF); Serrania del Parti (Aroko), 3rd central mesa, central-northern sector of the Serranfa, 04°31'N, 65°35'W, 1,100 m, 3-4 Oct 1979 (stam. fl), O. Huber 4298 (MYF, NY, US, VEN); eee Pari, Rio Pari, Cafio Asisa, Rio Ventuari, along W rim from Camp Cajfio, 2,000 m, 4 Feb 1951 (stam. fl), R. Cowan & J. Waurdack M31256 (E, NY, US, VEN). Clusia rotundifolia is most closely related to C. savannarum, but is easily distinguished by the compound inflorescence, antherodes bearing pollen, albeit frequently abortive, sessile stigmas, ovoid fruit and montane tepui scrub forest habitat. 4. Clusia savannarum Maguire, Bull. Torrey Bot. Club 75:422. 1948. Type: GUYANA. Poraro-SiparuN! REGION: Kaieteur Savanna, 6 May 1944 (pist. fl, fr), B. Maguire & D. Fanshawe 2367 (Hovoryre: NY!; isorypes: F!, FDG). Tree to 3 m tall. Branchlets rounded quickly below, obtusely tetragonal near apex, 3.5—5 mm diam., glabrous. Leaves petiolate; blades cartilagi- nous, widely elliptic to suborbicular, (3—)4—8 (9.5) cm long, G—-)5—7 cm wide, apex and base broadly rounded, nitid above, pallid below, midrib raised at bottom of adaxial groove, prominently raised below, secondary veins numerous, the collecting vein on the revolute margin; petioles subobsolete to 5 mm long, with broadly margined adaxial pits. Staminate inflorescence unknown. Pistillate inflorescence 1—3(—7)-flowered; peduncle tetragonal, not flattened, 8-12 mm long; primary inflorescence bracts op- posite, stiffly coriaceous, suborbicular to elliptic, 3-10 mm long, 24 mm wide, apex subacute to obtuse, base obtuse to broadly rounded, the margin revolute; secondary bracts 2, opposite, carnose, suborbicular,2.5— 3 mm long, 1.5-2 mm wide, apex obtuse to subacute, sessile, medially promi- nently carinate, the margin scarious, flat, entire; pedicels 2-3 mm long. Flowers; sepals 4, decussate, stiffly chartaceous, widely ovate to suborbicu- lar, 5-7 mm long and wide, apex broadly rounded, the margin scarious, with several notches; petals 4, rarely 5, the fifth one vestigial, contorted, carnose, obovate-spathulate, 7-10 mm long, apex truncate, irregular, ta- pering to 2.5—3.5 mm wide at base, white and often pink toward base; staminodia 10, 1.8-2.7 mm long, the filaments flat, 1-1.5 mm long, con- nate 0.2—0.3 mm into a tube, the apical free portion 0.8—1.2 mm long , the antherodes ovate, 1—-1.5 mm long, apex subacute to obtuse, latrorsely de- hiscent, containing small amounts of pollen; pistil ovoid, 3-5 mm long, Pipoty, Clusia section Anandrogyne 747 7-8 mm wide, carpels (4—)5, stigmas cuneiform, peltate, on short styles to 1.5 mm long. Fruit subglobose, yellowish green at maturity, 1-1.2 cm long, 1.3-1.5 cm diam. Distribution.—Clusia savannarum is known only from the eastern portion of the state of Bolivar, Venezuela, and the Kaieteur Plateau of Guyana, at 460—1,000 m elevation. Ecology and conservation status.—Clusia savannarum grows in the shrub islands in savannoid formations on white sands from sandstone below the tepuis. The shrub islands are characterized by the presence of Cybianthus fulvopulverulentus (Mez) Agostini (Myrsinaceae), Clusia pusilla Steyermark (Clusiaceae), Emmotum spp. (Icacinaceae), Humiria balsamifera St.-Hilaire (Humiriaceae), and Licania spp. (Chrysobalanaceae). The Kaieteur Plateau of Guyana lies in Kaieteur National Park, and is therefore protected, so this species cannot be considered threatened. Specimens examined: VENEZUELA. Bolivar: Dtto. Roscio, slope N eee 05°25'N, 61°25'W, 1,000 m, 4 Mar 1983 (fr), O. Huber et al. 7303 (MYF, NY, VEN). GUYANA. Cuyuni-Mazaruni Region: Pakaraima Mts., Base camp on tributary an Partang River, 8.6 km NE of Imbaimadai, 05°46'N, 60°16'W, 650 m, 19 May 1992 (fr), B. Hoffman et al.. 1681 (BRG, FDG, US). Potaro-Siparuni Region: Savanna near Kaieteur Falls, 460 m, 4 Mar 1962 (fr), R. Cowan & T. Soderstrom 2037 (FDG, NY, US). Clusia savannarum is most closely related to C. rotundifolia, but may be recognized by its simple, 1-3-flowered inflorescence, the completely sterile antherodes, thin styles 1—-1.5 mm long, and subglobose fruit. The shru island habitat in the white sand savannas below the tepuis is quite different from the scrub forests on the tops which houses C. rotundifolia. 5. Clusia wurdackiana Pipoly, nom. nov. Clusia lee ears ex Engler in teat Flora Brasiliensis 12(1):105. 1888, syn nov. A [GUYANA]: Roraima, Nov 1842 (stam. ff), oe Sele 1037 ee B- destroyed; F oe 9200!), non Clusia sessilis G. For Austr. 74. No. 391. 1786, = Garcinia sessilis (G. Foster) Seeman Clusia ¢ cerroana Steyermark, Fieldiana, Bot. 28:386. 1952. Typ . VENEZUELA. BOLivar: {, S-facing slopes between plateau portion and ‘ Gus Camp,” 1,700—-1,800 m, ia 1944 (fr), J. Steyermark 59702 (HoLoryPe: F!; isorype: VEN). Shrub to small tree to 5(—15) m tall. Branchlets somewhat angulate, at times tetragonal but with obtuse angles, 8-10 mm diam., the bark trans- versely checked and exfoliating in the uppermost nodes. Leaves petiolate; blades chartaceous to coriaceous, obovate to widely oblanceolate, 9-24 cm long, 6-11 cm wide, apex broadly rounded to truncate or rarely, emargin- ate, base acute, midrib prominent above and below, the secondary veins numerous, prominulous above and below, the margins entire, flat; petioles canaliculate and marginate, 3-5 mm long, the adaxial margined pit as wide as the petiole. Staminate inflorescence a 6—9-flowered panicle of 3-flowered 748 Sipa 16(4) cymes; peduncle 3—5 cm long; secondary inflorescence bracts coriaceous, suborbicular, 0.8—1.3 mm long and wide, apex broadly rounded, cucullate but not carinate, the margin scarious, entire; pedicels 1-1.5 mm long; bracteoles 4, decussate, the outer 2 stiffly coriaceous, oblate, S—7 mm long, 8 mm wide, apex broadly rounded, medially carinate, the margin scari- ous, entire, the inner 2 chartaceous, obovate, 7-9 mm long, 4-6 mm wide, apex broadly rounded, the margin scarious, entire; Staminate flowers white; sepals 6, the outer 2 opposite, stiffly coriaceous, suborbicular, 5—7 mm long and wide, apex broadly rounded, medially carinate, the margin scari- ous, entire, the inner 4 contorted, increasingly larger, to 10 mm long and wide, cucullate but not medially carinate, the margin scarious, irregular, entire; petals 6, contorted, membranaceous, oblanceolate, 12—15 mm long, 5-7 mm wide, apex broadly rounded, cucullate, the margin opaque, not scarious, entire; stamens numerous, 15-20 mm long, the androphore cubic 6-9 mm diam., the filaments 9-11 mm long, the anthers linear, 5-7 mm long, apex muticous, the base truncate, latrorsely dehiscent by wide longi- tudinal slits; pistillode absent. Pistillate inflorescence a 3-flowered cyme; peduncle obsolete to 2(—5) cm long; secondary inflorescence bracts resem- bling leaves but narrowly oblanceolate to obovate, 4-8 cm long, 2-4 cm wide; bracteoles 2, cartilaginous, oblate, 4-6 mm long, 10-12 mm wide, apex broadly rounded, medially carinate, the margin scarious; pedicels ob- solete to 1 cm long. Pistillate flowers like the saminate but sepals 4, decus- sate, suborbicular, 3-7 mm long and wide, apex broadly rounded, cucul- late but not medially carinate, the margin scarious; petals unknown; scaminodia obsolete; ovary subglobose, 5-7 mm long and wide; 5-carpel- late; styles thick, radiate, 3—S mm long at maturity; stigmas corniculoid, 5-7 mm long and wide, convex. Fruit subglobose, 1.5—2 cm long and wide, smooth, not sulcate. Dystribution.—Known from the Guayana Highland of Venezuela, Guyana, Surinam and adjacent Brazil, 330—2,000 m elevation. Ecology and conservation status. —Clusia wurdackiana is a common element on rocky outcrops throughout its range. It frequently forms large colonies and reproduces rapidly from seeds as well as lammas shoots. It appears to be fire resistant, judging from populations observed in Brazil, which had distorted trunks from fire exposure. This is the most resilient species of the genus because it tolerates the greatest amount of ecological disturbance. Therefore, it is not considered threatened. Etymology.—This species is dedicated to John Julius Wurdack, Curator of Botany Emeritus at the U. S. National Museum of Natural History, Smithsonian Institution. John is best known for his outstanding plant col- lections from the Guayana Highland of Venezuela and the Andean/Amazo- nian transitional region of Peru and for his pre-eminence as an authority on on _— Pipoty, Clusia section Anandrogyne 749 the systematics and ecology of Neotropical Melastomataceae. His willing- ness to share his wealth of knowledge and to provide encouragement to younger systematists throughout the botanical community has had a pro- found effect on all of us who work in neotropical plant systematics. Representative specimens examined: VENEZUELA. Amazonas: Cerro Aracamuni, OO Proa Camp, 01°26'N, 65°47'W, 1,550 m, 16 Oct 1987 (fr), R. Liesner & F. Delascto 22021 (MO, US, VEN); o1°321N 65°49'W, 1,400 m, 26 Oct 1987 (fr), R. Liesner & G. Carnevali 22485 (MO, US, VEN); Cerro Aracamuni, N part, 01°32'N, 65°49'W, 1,415 m, 16-18 Oct 1987 (fr), FE Delascio & R. Liesner 13518 (MO, NY, US, VEN); N slopes of Cerro Duida, 2 km S of Culebra, 03°43'N, 65°45'W, 700 m, Apr 1990 (fr), A. Ferndndez 7757 (BRIT, PORT, VEN); on Plateau of Cerro Duida above Culebra, 03°36'N, 65°42'W, 1,250 m, 2 Mar 1985 (fr), R. Liesnmer 18151 eh US, VEN); Base of cliff on slope of Cerro Huachamacari, 03°39 'N, 65°43'W, 1,000—1,300 m, 5 Mar 1985 (fr), R. Liesmer 18312 (MO, US, VEN); Cerro Huachamacari, base main wall and slope below it, E side, 03°49'N, 65°43'W, 800-1,300 m, 5 Nov 1988 (stam. fl), R. Liesmer 25870 (BRIT, MO, US, VEN), (fr), R. Liesner 25876 (BRIT, MO, US, — Cerro Huachamacari, Rio Cunucunuma, 5 Dec 1950 (stam. fl), B. Maguire et al, 29854 (NY, US, VEN); vicinity of Summit Camp, Cerro een 1,800 m, 6 Dec 1950 (stam. fl), B. Magaire et al. 30016 (MO, NY, US, VEN), 1,600 m, 15 Dec 1951 ) (stam. fl bud), B. Maguire et al. 30254 (NY, US, VEN), 1,620 m, 15 Dec "1950 (fr), B. Maguire et al. 30255 (NY, US, VEN); Slope of Cerro Marahuaca, above Rio Yameduaka, 03K 38'N, 65K 28'W, 1,225 m, 19 Feb 1985 (fr), R. Liesner 17692 (MO, US, VEN); Cerro de la Neblina,S face of Pico Phelps Massif, 00°48'N 66°00'W, 1,550-1,650 m, 13 Apr 1984 (fr), A. Gentry & B. Sten 46600 (MO, NY, US VEN), Cerro de la Neblina, 6.5 km SSW of base camp, $ cca of range, 00°47'N, 66°11'W, 1,600 m, 18 Apr 1984 (fr), B. Stern et al. 1652 (BI MO, NY, US, VEN); Vicinity of Camp VI, Cerro Neblina, on ridge on Venezuelan- — ne border, 3, of Pico Zuloaga, 00°53'N, 65°56'W, 2,000 m, 13-15 Apr 1984 (fr), W. aan: & T. oo man 3059 (BRIT, EF MO, NY, US, VEN); Above one IIL Cerro de la Neblina, Rio Yatua, 1,600—-1,700 m, 17 Nov 1957 (stam. fl), : Maguire et al. 42044 (NY, US, VEN), Above ae IV, Cerro de la Neblina, 1,650-1,700 m, 12 Nov 1957 (stam. fl), B. fanan et al. 42056 (NY, US, VEN); Cafio Grande, Cerro os a Neblina, E of Cum af 1,100-1,300 m, 25 Nov 1957 (fr), B. Maguire et al. 42228 (NY, US, VEN); Certo la Neblina: « ulong W escarpment, 1,900-2,00 m, 6 Jan 1954 (fr), B. Maguire et al. 70 (NY, US, VEN); Cerro Sipapo (Pardque), below Camp Grande Terrace, 1,400 m, io _ 1948 (fr), B. Maguire & L. Politi 27686 (NY, US, VEN), Cait Sipapo, S terraces of Peak IV, 1,800 m, 3 Jan 1949 (fr), B. Maguire & L. Politi 28133 (NY, US, VEN); In Cafio Profundo, Cerro ae 10 Jan 1949 (fr), B. Maguire & L. Polit 286 28 (NY, US, VEN); Serranfa de Tapir Campo Tamacuari, stream trail from camp, 01°14'N, 64°40'W, 1,300 m, 10 Feb ios (fr), H. Beck et al. 939 (BRIT, NY, VEN), (fr), A. Henderson 1025 (BRIT, NY, VEN); On plateau N of unamed 1,760 m peak, 9 km NW of settlement of Yutajé, 4 km W of Rio Coro-Coro, W of Serranfa de Yutajé, O5°41'N, 66° 10'W, 1,050— : 300 m, 7 Mar 1987 (fr), R. Liesner & B. Holst 21732 (MO, US, VEN), 5-8 km NW of é, S slope below Serrania de Yutajé, 3 km W of Rio Coro-Coro, W of Serrania, N, ae 700—1,000 m, 10 Mar 1987 (fr), R. Liesner & B. Holst 21841 (MO, NY, US, VEN); Cumbre of Cerro Yutajé, E sector of Serranfa de Yutajé, headwaters of Cahlo Yutajé, 05°45'N, 66°03'W, 1,800 m, 21 Mar 1988 (fr), O. Huber 12618 (MYF, NY, US, VEN), SE summit of ieee Yutajé, Rio Manipiare, 2,100 m, 17-19 Feb 1953 (fr), B & C. Maguire 35343 (NY, US, VEN). Bolivar: ealns tepuf, summit, in S-central region, headwaters . Rio Churtin, 05° S1'N, 62°32'W, )m, 29 Mee ong (fr), B. Holst 3730 (MO, U 750 Sioa 16(4) VEN), SE of second ie a Rio Churtin, near S camp, 1,690 m, S section, W division of Auyan-tepuf, 3 May 4 (fr), J. Seek 93319 (F, MO, NY, US, VEN); Rio Lomita Camp, summit of portion of NW arm, W division of Auyan-tepui, near “Rio Lomita Camp”, 1,800 m, 10 May 1964 (fr), J. Steyermark 93569 (MO, NY, US, VEN); Chimanta Massif, Central Section, NNW of Summit Camp, 1,970 m, 19 Feb 1955 ae Steyermark & J. Wurdack 941 (F, MO, NY, US, VEN); Alto Rio Cuyuni, above escarpment of La Escalera, 850 m, 20-21 Aug 1962 (fr), B. Meat et al. 46834 (NY, US, VEN); Cerro Guaiquinima, Rio Paragua, below rim of W Escarpment, 1,600-1,700 m, 31 Dec 1951 (fr), B. Maguire 32893 (NY, US, VEN); Hu- tepuf, Gran Sabana, 1,500 m, 8 Mar 1952 (fr), B. Maguire 33315 (NY, US, VEN); Sierra de Lema, headw of Rio Chicandn, at base of NE-facing bluffs, 80 km SW of El Dorado, 06°05'N, pane 500 m, 28 Aug 1961 (fr), J. Steyermark 89550 (F, MO, NY, US, VEN); Mcpio. Cedefio, Serrania de Maigualida, sand- stone mountains 20 km E of San José de Kayamd, 45 km N of Cerro ee 06°19'N, 65°12'W, 1,250 m, Apr 1969 (fr), A. Ferndndez 5397 (BRIT, PORT, VEN); Cerro Impacto, 06°00'N, 65° 10'W, 1,250 m, Jun 1988 (fr), 8. Elcoro 348 (BRIT, PORT. ee Trirepéon- tepui, SW side of Rio Asaporpo, 1,300 m, 7 Jan 1953 (fr), J. Wardack M34040 (F, NY, US, VEN); Cerro Venamo, along right bank of W slope near spade ae 950-1,150 m, 29-30 Dec 1963 (fr), J. Steyermark et al. 923 63 (EF, , NY, US, VEN). BRAZIL. Amazonas: Mpio. de Barcelos, Platé da Serra Araca, Serra ace rte, 6 i W of well camp, 00°5 L'N, 63°22'W, 1,150 m, 20 Feb 1984 (fo, I o Amaral, J, Pipoly et al. 1635 (INPA, K, MG, NY); Pico Rondén, 0-3 km N of km 211 of Perimetral Norte Hwy, Pico Rondén, 01°32'N, 62°48'W, 3 Feb 1984 (fr), J Pipoly et al. 6617 (INPA, K, MG, NY, US). GUYANA. Cuyuni-Mazaruni Region: Mt. Ayanganna, E-most peak, 05°25'N, 59°57 'W, 1,350-1,380 m, 11 Mar 1987 (fr), J. Pinaly et al. 11113 (B, CAY, FDG, NY, P, U, US). Potaro-Siparuni Region: Kaieteur Plateau, Kaieteur Falls, along W rim of Potaro Gorge, 330 m, 18 Feb 1962 (fr), R. Cowan & T. Soderstrom 1896 (BRG, K, US); Kateteur Falls 05°12'N, 59°29'W, 500 m, 2 Apr ine (fr), W. Hahn et al. 4113 (BRG, NY, US); Ma dle River Valley, cributary of Potaro River, E-facing escarpment of — Ebini, 700 m, 15 Oct LOSI (fr), B. Maguire 32123 (FDG, NY, US); Pakaraima Mts., Mt. Kukuinang, adjacent W edge of prin ge meee 05°04'N, 59°57'W, 900-1 i. m, 26 Feb 1993 (fr), 7. Henkel et al. 3, T, US); Mt. Wokomung, summit ridge of Kamiewah Pin- nacle NE to : Nie re Aianeaone® 05°04'N, 59°52'W, 1,550-1,650 m, 17 Nov 1993 (fr), 7) Henkel et al. 4459 (BRG, US). Surinam: Wihelmina Gebergte, upper slopes and summit of Juliana Top, 15 km N of Lucie River, 1,000—1,230 m, 18 Aug 1963 (stam. fl), H. S. Irwin et al. 54862 (BBS, NY, U, US). The thick branchlets and pedicels and corniculoid stigmas of Clusia wurdackiana indicate that it is closely related to C. pachyphylla. However, the chartaceous to coriaceous, obovate to oblanceolate leaf blades, with acute bases and flat margins, stiffly coriaceous sepals with entire margins, thick, radiate styles and smooth ovary and fruit clearly distinguish Clusia wurdackiana from C. pachyphylla. — wurdackiana is a polymorphic ochlospecies (sensu Prance 1982; White 1962; Pipoly 1983), with great quantitative variation across a broad ee range. Steyermark (1981) corrected the common misconception that Schomburgk’s Roraima collections took place on the Guyanese side of the mountain, when in fact, he approached the upper slopes from the Venezu- elan side, hence the correction in the type citation. Pipory, Clusia section Anandrogyne 751 6. se oe Gleason, Bull. Torrey Bot. Club 58:405. 1931. Type: ELA. Amazonas: Cerro Duida, summit of Peak 7, 2,230 m, Aug 1928— i nee (pist. fl), G. Tate 609 (HoLoTYPE: NY!). Clusia scariosa Lasser & Maguire, ae’ 7:81. 1950. syn. nov. Type: VENEZUELA. MAZONAS: Cerro Yavi, 2,200 m 3 Mar 1947 (fr), Phelps & C. Hitchcock 51 (HOLO- TYPE: — ISOTYPE: WEN). Free-standing tree to 6 m tall; branchlets tetragonal, 8-11 mm diam., strongly sulcate, glabrous. Leaves petiolate; blades thickly cartilaginous, elliptic to obovate, (S—)8—10(—18) cm long, (2.5-)6.5—-10 cm wide, apex roadly rounded to broadly obtuse, midrib slightly raised above, promi- nently raised below, decurrent to petiole base, secondary veins numerous, prominently raised above and below, the collecting vein on the revolute margin; petioles trigonal, sub- to paar marginate, 1-1.5(—2) cm long, the adaxial pit deep, strongly margined, less than petiole diameter. Stami- nate inflorescence 1—3-flowered; peduncle 1—1.5 cm long; tetragonal, ae sulcate; flora bracts 4, decussate, suborbicular, cartilaginous, 0.8-1. long and wide, apex obtuse, prominently carinate, the margin scarious, opaque, irregularly notched; pedicels obsolete to 3 mm long. Staminate flowers; sepals 8—10, the outer four decussate, cartilaginous, suborbicular, 8-15 mm long, 10-16 mm wide, apex broadly rounded, the margin whit- ish scarious, variously incised, the inner ones as in the outer but contorted, acropetally larger, obovate to reniform, 15-20 mm long, 18-23 mm wide, petals G-8, the outer two decussate, the inner ones contorted, all similar in shape and size, coriaceous to carnose, obovate, 2.5—3 cm long, 1.3-2 cm wide, apex deeply erose, stamens numerous, 5—7 mm long, the androphore cubic, 1.5—2 mm long, the filaments linear, thick, 2-3 mm long, the an- thers linear, 3-4 mm long, extrorsely dehiscent by wide longitudinal slits, the connective muticous; pistillode absent. Pistillate inflorescence as in staminate but peduncle 1.5—2 cm long; bracts as in staminate; pedicels obsolete to 5 mm long. Pistillate flowers as in staminate but sepals 8-15 mm long 11-18 mm wide; petals but 2.0-2.5 cm long, 1.2-1.6 cm wide, persistent in fruit; staminodia numerous, as in staminate but not on androphore, 3-6 mm long, the filaments 3—3.5 mm long, irregularly grouped in phalanges, the antherodes 1-1.5 mm long, some of them pro- ducing pollen; pistil ovoid, 5-8 mm long; carpels 6-8, the stigmas sessile, convex, corniculoid, the corners acute, 4—S mm long, 3-5 mm wide. Fruit subglobose, 2.3-2.8 cm long, 3.3-4 cm diam. Distribution —Endemic to the NW portion of the state of Bolivar, west- ward through southern Amazonas, Venezuela, at 1,700—2,200 m Ecology and conservation status.—Clusia pachyphylla is infrequent on rocks in savannas on the summits of the tepuis it inhabits. The areas in which it oc- curs are quite remote and therefore, this species is not considered threatened. 752 Stipa 16(4) Specimens examined: VENEZUELA. Amazonas: Summit of Cerro Coro-Coro, NW headwaters of Rio Manipiare, (NW sector of Serrania Yutajé), 05°46'N, 66°12'W, 2,200 m, 12 Nov 1987 (fr), O. Huber 12310 (MYF, NY, VEN); Summit of Cerro Duida, 1,800— 2,075 m, 4 Sep 1944 (fr), J. Steyermark 58344 (F, VEN), Depto. Acures, Sierra Maigualida, NW sector, small valley poe an oa tributary of Cafio Iguana, 05°30'N, 65°15'W, 2,000 m, 28 Feb—3 Mar 1991 (fr), P. Berry se 4862 (BRIT, MO, VEN),(fr), 4883 (BRIT, MO, VEN), Sas fl), 4885 pe MO, N), 2 Mar 1991 (fr), O. Huber et al, 13103 (BRIT, MO, N); Serrania Uasadi, cee ctor, summits at E headwaters of Rio Asita, right ie i. Rio Ventuari, 05°21'N, oa 2'W, 1,850 m, 22 Nov 1988 (fr), O. Huber 12850 (MYF, US, VEN); Cerro Yavi, 2,200 m, 1-3 Mar 1947 (fr), K. Phelps & C. Hitchcock 53 (NY, VEN); Summit of Cerro Yavi, nea walels of Rio Parucito, W tributary of Rio Manip, in the NE sector of summit, 05°43'N, 65°52'W, 2,100 29 Oct | —— fl), O. Huber 11880, (MYF, NY, VEN), (piste. { fl, fr), O. Huber 118 Be (MYF, NY, VEN); Sanaa mit of Cerro Yutajé, E sector of Serrania Yutajé, headwaters of Cafio poe 16, 05°45! N, 66°03'W, 1,800 m, 21 Mar 1988 (fr), O. Haber 12622 (MYF, NY, US, VEN). Bolivar: Auyan-teput, Sep 1937 (fr), FE Cardona 232 (F, NY, US, VEN); gy Guaiquinima, Rio Paragua, N Valley, 1,700 m, 2 Jan 1952 (fr), B. Maguire 32926 (E JS, VE Summit, Cerro Guaiquinima, | km NW of Cumbre Camp, 26 Dec ee (fr), B. Maguire 32788 (F, NY, US, VEN); Meseta de Jaua, Central-Southern sector, southern ee head- waters of Rio Maran; headwaters of Rio Cacaro, 04°48'N, 64°32'W, 1,750-1,800 20 Nov 1989 (stam. fl), O. Huber 13020 (BRIT, MYF, VEN): Deto. Cedefio, Sierra oF Maigualida, NE sector, a on the headwaters of the Rio Chajura, W tributary of the Rio Erebato, ca. 100 km SW of Campamento oe 05°33'N, 65°13'W, 2,100 m, 28 Mar 1988 (stam. fl), O. _. 12728 (MO, M VEN); Sierra de Maigualida, NE sector, headwaters of Rio Chajura, W cribut tary - abate 1 00 km SW of Campamento Entrerios, 05°33'N, 65°13'W, 2,100 m, 18 Nov 1988 (fr), O. Haber & L. Izquierdo 12789 (MYF, NY, US, VEN). a ’ > > The thick branchlets and pedicels, and corniculoid stigmas of Clusia pachyphyla indicate a close relationship to C. wardackiana. However, the cartilaginous, elliptic to suborbicular leaf blades, with obtuse to truncate bases, chartaceous sepals with incised margins, the obsolete styles, and deeply sulcate fruit clearly distinguish C. pachyphylla. This species is one of a very few with prominent secondary venation despite the cartilaginous texture of the leaf blades. The type of Clusia scariosa Maguire differs only in the smaller, more closely veined leaves, larger petals and more numerous staminodia, all quan- titative characters which have now been found to vary widely based on analysis of many more gatherings than were available to Maguire. One collection from Sierra Magualida (P. Berry et al. 4883) has fruits with much narrower stigmas whose margins are fimbriate, and may represent an undescribed taxon. More collections from the Sierra and environs will be necessary to fully understand variation in Clusia pachyphylla. 7. Clusia duartei Maguire, Moscosoa 4:215. 1986. Typr: VENEZUELA. eee Cerro Sipapo, Campo Grande, 1,500 m, 10 Dec 1948 (stam. fl), B. Maguire . Politi 27575 (HoLoryee: NY!; isotypes: F!, G, JBSD, K!; MO!; US!, VEN). Pipoty, Clusia section Anandrogyne i3 Tree to 10 m tall; branchlets tetragonal, the angles acute but not alate, 15—22 mm diam., semisucculent, not sulcate. Leaves petiolate; blades car- tilaginous, sabosbicul ar to very widely elliptic, 20-27 cm long, 19—21.5 cm wide, apex very broadly rounded to subtrunctate, base very broadly rounded, midrib slightly raised above ca. 3/4 its length, decurrent to peti- ole base, prominently raised below, secondary veins numerous, inconspicous above, barely visible below, the submarginal collecting vein ca. 2 mm from margin, the margin flat; petioles trigonal, somewhat marginate, 2-3 cm long, the adaxial pits strongly marginate. Staminate inflorescence 1—3-flow- ered; peduncle tetragonal, the angles acute, 2 cm long; primary inflores- cence bracts foliaceous, opposite, cartilaginous, suborbicular, petiolate, the blades 4.5-7.5 cm long, 3.5—-6.5 cm wide, apex truncate, base obtuse, the midrib and secondary veins as in leaves, the petioles ca. 1 cm long, the adaxial pit strongly marginate; bracteoles 2, carnose, suborbicular, 1—1.5 cm long and wide, apex broadly rounded, prominently carinate, the mar- gin narrowly scarious, subentire, with occasional incisions; pedicel 3-10 mm long. Staminate flowers; sepals 8-10, decussate, the outer stiffly chartaceous, oblate, 1.2—1.5 cm long, 2—2.5 cm wide, apex broadly rounded, the margin very narrowly scarious, entire but undulate when dried, the inner acropetally thinner, more orbicular and larger, 2—2.5 cm long and wide, the scarious margin prcgrcssive’y wider; eee 8-10, contorted, co- riaceous, oblong, contorted, 4.5-5 cm long, 1.5—2 cm wide, not gradually tapering toward base, apex very broadly rounded to truncate, the margin scarious, Opaque, entire; stamens numerous, 1.5—2 cm long, the androphore cubic, 8-10 mm long, the filaments united basally, linear but fleshy, 3—5 mm long, the anthers linear, 3-5 mm long, apex muticous, dehiscent by introrse , wide longitudinal slits; pistillode absent. Pistillate inflorescence unknown, except floral bracts 2, cartilaginous, suborbicular, 1.2—1.5 cm long and wide, the apex broadly rounded, prominently keeled, the margin opaque, somewhat scarious; pedicels obsolete to 4 mm long. Pistillate flowers (from mature fruit); sepals 8, persistent in fruit, as in staminate but 1.5— 1.8 cm long, 2—2.5 cm wide, apex broadly rounded, the margin widely scarious, irregularly incised, acropetally larger and more suborbicular to 2.5 cm long, 2.3 cm wide; petals 6, very widely oblong to suborbicular, 2.5—3 cm long and wide; staminodes 12-16, 6-10 mm long, barely con- nate by filaments basally, the filaments 3—5 mm long, the anthers ovate, 5— 7 mm long, apex obtuse, latrorsely dehiscent by a long longitudinal slit, devoid of pollen or producing small amounts of deformed pollen; pistil unknown; carpels 10-12, deeply sulcate along the sutures, the stigmas sessile, subapical, corniculoid, the angles rounded, 4-5 mm long, 3—4 mm wide. Fruit subglobose, 5—G cm long, 6-8 cm wide. 754 Stipa 16(A4) Distribution.—Endemic to the western portion of the state of Amazonas of Venezuela, known from Cerros Sipapo and Cuao, at 1,500—-1,580 m. Ecology and conservation status.—Clusia duarter occurs in low scrub forest along river margins. It is known from only four collections, and should be considered rare, but not enough is known of its population biology to de- termine if it is threatened. Specimens examined: WENEZUELA. Am : Depto. Atures, Cerro Cuao, Cafio Cabeza de Manteco, 73 km SE of Puerto ae. "05" 06'N, 67K 24'W, 1,580 m, Sep 1989 (fr), A. Ferndndez et al. 6271 (MO, PORT, VEN); Cerro Sipapo, Campo Gane 1,500 m, 10 Dec 1948 (stam. fl), B. Maguire & L. Politi 27543, (fr), 28683 (NY, VEN) Clusia duartet appears to be most closely related to C. pachyphylla, but can immediately be distinguished by its thicker branchlets, longer peti- oles, massive leaf blades and fewer, longer floral bracts. Cursory compari- son of vegetative parts may result in intial confusion with some individuals from populations of C. (sect. Clusiastrum) crassifolia Planchon & Triana, but the yellow latex, minute stigmas on thin, finger-like styles, and antheriferous staminodes with dehiscence via cha pores (characteristics of all members of sect. Clusiastrum) clearly differentiate C. crassifolia. ACKNOWLEDGMENTS This study was greatly facilitated by the Missouri Botanical Garden and The New York Botanical Garden, which supplied numerous gifts for de- termination and loans. Fieldwork in Brazil was conducted under Programa Flora, Projeto Flora Amaz6nica, NSF Grant BSR-8106632X3 to Ghillean Prance during his tenure at NY. The Fund for Neotropical Research at the New York Botanical Garden allowed me to visit and consult the collec- tions from the Guianas. I also thank Vicki Funk, Carol Kelloff and John Boggan for their hospitality while I visited US under the auspices of the Smithsonian Biodiversity of the Guianas Program. John Wurdack (US) and Paul Berry (MO) reviewed the manuscript and provided useful sugges- tions. I thank Catherine Mayo and Jon Ricketson (MO) along with June Cunningham and Lindsay Woodruff (BRIT) for help in specimen process- ing. Linda Ellis skillfully prepared yet another excellent illustration. REFERENCES Hammel, B. 1986. New species of Clusiaceae from Central America with notes on Clusia and Ss my in the tribe Clusieae. Selbyana 9:112-120. Macuire, B. 1979, Guayana, region of the Roraima Sandstone Formation. In: K. Larsen & L. ae Nielsen, eds. Tropical Botany. Academic Press, London. Pp. 223-238 Pipoty, J. 1983. Contributions toward a monograph of Cybianthus (Myrsinaceae): II. A Revision of subgenus a Brittonia 35:61-80. Prpoty, J. and A. Grarr. 1995 ynopsis of the genus C/asia sections a and Brachystemon (Clusiaceae) in i South America. Sida 16:505— Pipoty, Clusia section Anandrogyne ie) Prance, G. 1982. Forest refuges: evidence from woody angiosperms. In: G. Prance, ed. Biological Diversification in the Tropics. Columbia University Press, New York. Pp. STEYERMARK, J. 1981. Erroneous citation of Venezuelan localities. Taxon 30:816-817. Wuire, FE. 1962. SCR variation and speciation in Africa with particular reference to oO He D. Nic ed. Taxonomy and Geography. The Systematics Association . No. 4, London, vu. K. Pp. 71-103 756 Sipa 16(4) BOOK NOTICE Porter, C.S., J.1. Conen, and D. JaANczewski (Eds.). 1993. Perspectives on Biodiversity: Case Studies of Genetic Resource Conservation and Development. (ISBN 0-87168-5 12-4, pbk) American Associa- tion for the Advancement of Science, Washington, DC 245pp. From the prologue, by Ehrlich and Wilson to the many case studies covering such sub- jects as agroecosystems, fisheries and wildlife, managed fore t ecosystems and conservation and regional d evelopment, this volume is full of literature reviews, republication of previ- ously published key works, and original, site-specific research. The book is divided into two parts: Themes in species and genetic resource conserva- tion, and case studies of biodiversity conservation in natural habitats. The rst. section consists of four papers, which include: t tactics and conflicts in preserving genetic resources (Soule), theoretical concerns regarding detection of congruent phylogenetic patterns among radiating lineages, and thus, detection of centers of evolutionary radiation (Erwin), bal- ance of species preservation vs. economic considerations (Morowitz), and finally, biological and socioeconomic factors in conservation of crops and flora (Williams). The secon tion consists of purely case studies, which deal with every level from the local farmer (Altierta & Montecinos), to conservation of specific crops (e.g., Wilkes, on maize), habitats (e.g. Twilley et al., on mangroves), mani ee ecosystems (Hartshorn & Pariona on Peru, Padoch & Peters, on forest gardens in West Kalimantan, and Wilcox & Olson) on Pacific forests of the Pacific Northwest I was somewhat disappointed that conservation of such pivotal groups as birds, bats, reptiles, and marine mammals, was totally ignored in this volume. Ic is, nonetheless, use- ful, the binding is well-done and the reproduction clear. T here are several critical papers which will be fodder for citations in our grant proposals, and I anticipate more soon. In summary, I think every organismic biologist should have access to this book as a model for preservation of case studies, and a source of background reading material.—Jobn Pipoly IIT. Sipa 16(4): 756. 1995 THE STATUS OF POTAMOGETON PERFOLIATUS (POTAMOGETONACEAE) IN LAKE PONTCHARTRAIN, LOUISIANA JOHN W. BURNS JR., MICHAEL A. POIRRIER and *KRIS P. PRESTON Department of Biology, *Department of Geography University of New Orleans New Orleans LA 70148, U.S.A. ABSTRACT Potamogeton perfoliatus L. is listed as a sensitive plant in the Louisiana Coastal Zone. Four erratic records from Lake Pontchartrain constitute the southwestern extent of its range in North America. To obtain a better understanding of its current status, aquatic habitats near New Orleans were surveyed. It was found in eastern Lake Pontchartrain at three new localities (Point Platte, Big Point, and Irish Bayou) and at two previously reported locali- s. The Point Platte bed is large (1.8 ha) and in a remote area. The erratic historic distri- ee of P. perfoliatus in Lake Pontchartrain may be due to the establishment of small, scattered, temporary beds in more accessible areas by dispersal from the large Point Platte population. Its restricted Louisiana distribution appears to be due to the suitable sediment and water quality conditions occurring in eastern Lake Pontchartrai — RESUMEN Potamogeton perfoliatus L. se incluye entre las plantas amenazadas de la zona litoral de Louisiana. Cuatro citas errdticas a orillas del lago Pontchartrain constituyen la parte suroeste de su Area norteamericana. Con objeto de obtener una mejor comprensién de su estado actual se observaron ciertos habitats acudticos cercanos a New Orleans. La especie fue hallada en tres nuevas localidades al este del lago Pontchartrain: Pointe Platte, Big Pointe, e Irish Bayou, asf como en otras dos logalidadss previamente cicadas. El lecho correspondiente a en una zona remota. La distribucién Point Platte es extenso (18 m*’ y se encuentra da al hist6ricamente erratica de P. perfoliatus en el lago Pontchartrain puede ser debi establecimiento de lechos pequefios, dispersos : temporales de zonas mas accesibles, debidos a dispersiones que tuvieron lugar a partir de la poblacién de Pointe Platte. Su distribucién restringida en Louisiana parece deberse a las condiciones de sedimento apropiado y calidad del agua que se dan en la zona oriental del lago Pontchartrain. INTRODUCTION Potamogeton perfoliatus L. (CLASPING PONDWEED) is a vascular plant that occurs in fresh and low salinity estuarine waters. It is currently listed as “extremely rare” in the Louisiana Coastal Zone by the Louisiana National Heritage Program (Lester 1988) with Louisiana records limited to Lake Pontchartrain. It is considered to be a widely distributed north temperate species that in North America occurs primarily on the northeastern coastal Stipa 16(4); 757-763. 1995 758 Sipa 16(4) plain (Sculthorpe 1967). Ogden (1943) described its distribution to be Newfoundland to Florida, Ontario, Ohio, and Louisiana, buc common only in the northeastern area of its range. Godfrey and Wooten (1979) reported that it occurs in calcareous or brackish ponds and streams with a distribu- tion including Mississippi but not Louisiana. Louisiana records represent the southwestern extent of its range in the United States. Lake Pontchartrain is a shallow, estuarine embayment in the Mississippi River Deltaic Plain in southeastern Louisiana. It encompasses 1,630 km? and has an average depth of 3.7 m and an average salinity of about 4 ppt (Sikora & Kjerfve 1985). Sebamed aquatic vegetation (SAV) occurs as discontinuous bands along the northeastern and southeastern shorelines and has been documented to be in decline (Burns et al. 1993; Mayer 1986; Montz 1978; Turner et al. 1980). These SAV bands, which are dominated by Vallisneria americana, extend as far as 200 m from shore in some areas, but rarely extend beyond a water depth of 1.5 m (Burns et al. 1993). Past Louisiana reports did not give a varietal form for P. perfoliatus. These records are: Riddell, Tchefuncta River lighthouse, 1883 (Ogden 1943); Brown, Mandeville Beach, 1945 (Haynes 1968); Montz, Pointe aux Herbes, 1973 (Montz 1978); and Brantley and Platt, beeween Goose Point and Bayou Lacombe, 1990 (Brantley & Platt 1991). Biological surveys of Lake Pontchartrain that included submersed aquatic vegetation were conducted by Chabreck oa 2), Lester (1988), Mayer (1986), Perret et al. (1971), Suttkus et al. (1954), and Thompson and Verret (1980). Alctt rough P. perfoliatus was not reported in these general surveys, it may have been present in the estuary and simply overlooked due to its limited cena At- tempts to find it prior to its being listed were unsuccessful (Lester 1988). Based on past records, its occurrence in Lake Pontchartrain is erratic and factors affecting its temporal and spatial distribution are unknown. There is a need for more information regarding the environmental factors affect- ing the distribution of submersed aquatics in estuaries to better under- stand why populations have declined worldwide (Dennison et al. 1993). Our study of P. perfoliatus was conducted to determine the following: (1) its current distribution in Lake Pontchartrain, (2) whether it occurs in other wetland habitats near Lake Pontchartrain, and (3) its habitat requirements. eon METHODS Our study area encompassed the entire shoreline of Lake Pontchartrain, including tidal areas of streams and passes, and aquatic habitats surround- ing Lake Pontchartrain (Fig. 1). The study area was bounded by La. Hwy. 36 to the north, Lafitte, La. to the south, La. Hwy 55 to the west and the Pearl River to the east. Field work began in November 1991 and ended in August 1993. BurNs ET AL., Potamogeton perfoliatus in Lake Pontchartrain 759 NC avn an ‘ ag s a s a Ly J yo? Study Area 4 ak A f Cay aa iS Wg ht {7h eseA ) 27 . y : if ‘gt Sry if ea y ‘Dy: oo Pee SAS J PY ne AS ARE PK ye PAA AES ee OY cient 4 CSA PINE HT Sah 9 14 iS ea 5 4 PVA Cee o/ BARA REST 5 PERS MSR Statute Miles I 3 5 Fic. 1. The historical distribution of Potamogeton perfoliatus in the Lake Pontchartrain estu- ary, LA: Riddell 1838 (A); Brown 1945 (B); Montz 1973 (C); Brantley and Platt 1990 (D); 1991-1993 distribution ( Goose Point (E), Bayou Lacombe (F), Pointe Platte (G), Big Point Beach (H), Pointe aux Herbes (I), Irish Bayou (J). This map is modified from NOAA navigational map no. 11369, 33% Ed. Species identification followed Godfrey and Wooten (1979); voucher specimens were deposited in the University of New Orleans Herbarium. Areal cover of monotypic stands of P. perfoliatus was determined by direct measure. Percent foliar cover in stands containing other species was deter- mined by the line intercept method (Westman 1985). Secchi disc (20 cm) transparency, water temperature, and water and sedi- ment samples were taken at each site. Salinity was calculated from chloride values determined by the mercuric nitrate method (Harvey 1957; Standard Methods 1989 ). Sediment textural classes were assigned by grain size analy- sis (Folk 1980). RESULTS Potamogeton perfoliatus was found at five sites in eastern Lake Pontchartrain (Fig. 1). It occurred in protected shoreline embayments and near the mouths of streams. This plant exhibits a wide range of variation in both morphol- ogy and color due to light intensity, sediment type, and water quality dif- ferences (Ogden 1943). Specimens from all sites fit the description of P. 760 Stipa 16(4) perfoliatus L. var. bupleuroides (Fern.) Farw; a smaller plane chat fruits freely and is distinguished from P. perfoliatus var. perfoliatus by a slender stem (diameter = 0.4-1.5 mm), delicate leaf (width = 0.5—2 cm) and few promi- nent nerves (7-17 nerves, 1-5 prominent) (Godfrey and Wooten 1979; Ogden 1943). Although our specimens fit the description of P. perfoliatus L. var. buplenroides (Fern.) Farw, several authors have not recognized this variety (Kartesz 1994; Thomas and Allen 1993). Southshore beds were present at Pointe aux Herbes (between the La. Hwy 11 bridge and I-10 bridge behind a cement erosion control structure) and the east side of the Pointe aux Herbes peninsula adjacent to the west mouth of Irish Bayou. Three beds occurred on the north shore between Goose Point and the mouth of Bayou Lacombe (500 m east of Goose Point, 200 m west of Bayou Lacombe and 1500 m west of Bayou Lacombe). Other north- shore beds were found near Point Platte (4.0 km east of Bayou Lacombe) and the swimming beach at Big Point. Physicochemical and habi- tat data from these sites are presented in Table 1. Potamogeton perfoliatus bed size ranged from four plants at Big Point to a 1.8 ha bed at Point Platte. Percent foliar cover of P. perfoliatus at Point Platte was 28%; and at Pointe aux Herbes, 72%. Total areal cover of P. perfoliatus was 2.2 ha and represents ca. 2.5 % of the total areal cover of all submersed aquatic vegetation in Lake Pontchartrain (Burns et al. 1993). No seasonal differences in bed size were noted, although individual shoot lengths were considerably shorter between December and February. A 20 x 30 m section of the Pointe aux Herbes population was found stripped of its leaves during September 1991. Waterfowl are known to feed upon the leaves, seeds, roots, and rhizomes of P. perfoliatus and are most likely responsible for the removal of leaves at this site. All populations occurred at water on) on depths between 31 and 122 cm and were rooted in sand and loamy sand substrates. Secchi disc transparency ranged from 30 to 217 cm, tempera- ture from 17 to 32 °C, and salinity from 1.0 to 8.2 ppt in P. perfoliatus beds (Table 1). DISCUSSION Although the areal cover of submersed aquatic vegetation in Lake Pontchartrain has declined by more than 50% since 1973 (Burns et al. 1993; Mayer 1986; Turner et al. 1980 ), there may have been an increase in the area occupied by P. perfoliatus. Irish Bayou, Big Point, and Pointe Platte are new locality records for P. perfoliatus, and it was found at Pointe aux Herbes where it had not been reported since 1973. The largest bed ever reported from Lake Pontchartrain was found at Pointe Platte during our study. The plants found at Big Point on 20 June 1991 disappeared by March 1992 and have not returned. The direct cause for the disappearance BurRNS ET AL., Potamogeton perfoliatus in Lake Pontchartrain 761 Tasie 1. Habitat data for Potamogeton perfoliatus in Lake Pontchartrain, Louisiana. Point Big Irish Bayou Pointe Platte Point Bayou Lacombe aux Herbes Foliar Stand Size (m2) 18000 4 plants 1920,180,300 300,119,150 1014 Flower & Fruit Present Absent Present Absent Present Water Depth (cm) 30.5-91.4 30.5-61.0 30.5-91.4 30.5-61.0 61.0-122.0 Sediment Type Loamy Sand — Sand Loamy Sand Sand Sand Species Present P,V,N,M,E P,V,R P,V,R,M P,V,N,M,E P,V,R Shoreline Stability Er St Er Er St Historical Recorc a a a b c Secchi Disk (cm) 210-217 196-206 31-62 30-72 46-81 Temperature (°C) 28-30 26-28 27-31 17-32 26-32 Salinity (ppt) 4.5 7.0-8.2 1.5-3.7 1.0-4.1 1.7-3.8 P=Potamogeton perfoliatus, V=Vallisneria americana, R= ae maritima, M= ihe ae sas Spicatum, ajas guadalupensis; E=Eleocharis parvula; St=stable shoreline r=eroding shoreline ecor for this area; b=first record 1990 (Brantly & Platt 1991), eae ee 1973 tents 1978). of P. perfoliatus in this area is not known. However, increased wave energy, associated with cement bulkheads along che shoreline, and “eatouts” by waterfowl may be responsible. The relative frequent occurrence and disap- pearance of scattered P. perfoliatus populations suggests that small popula- tions may persist by occasional recolonization rather than regulation of population size. Beds of P. perfoliatus greater than 300 m? in size had plants that flowered and produced fruit during March and April. The 1.8 ha bed found at Point Platte might be the source of small temporary beds located elsewhere in the estuary. These beds, which may become extirpated during adverse con- ditions, could become established through the dispersal of seeds and veg- etative propagules from the large Pointe Platte bed by waterfowl and cur- rents. Because P. perfoliatus occurs in sand and loamy sand sediments, it 1s restricted to the sites in northern and southeastern Lake Pontchartrain where these sediments occur. The large Point Platte bed may be important in maintaining the disjunct Lake Pontchartrain populations. Potamogeton perfoliatus is an inland species found in alkaline or brackish ponds and streams (Godfrey and Wooten 1979); it can also tolerate low- salinity estuarine conditions (Den Hartog 1981). It is present in many Atlantic Coast estuaries including the upper Chesapeake Bay (Orth et al. 1992) and can tolerate salinities up to 12 ppt (Twilley and Barko 1990). The stable, low salinity of Lake Pontchartrain (Sikora and Kjerfve 1985) is within the range of water quality conditions reported by other investiga- tors and is suitable for P. perfoliatus growth. The absence of P. perfoliatus from other sites in the study area is probably due to water quality and bottom sediment type. Eastern Lake Pontchartain 762 Stipa 16(4) is the only site in the study area that has low-salinity water and sand sedi- ment combined. Aquatic habitats north of Lake Pontchartrain have sand sediment but unsuitable acidic water low in dissolved solids. Other habi- tats south of Lake Pontchartrain with alkaline or brackish water have un- suitable silt and clay sediments. Herbivory and competition from other aquatic plants probably also affect its abundance and distribution. Potamogeton perfoliatus is an excellent food source for waterfowl; redheads, canvasback, mallard, ring-necked duck, black duck, Canada geese and tun- dra swans are known to feed on the seeds, leaves, stems and rhizomes (Hurley 1990). Waterfowl are probably responsible for the removal of leaves from P. perfoliatus at Pointe aux Herbes and the disappearance of all P. perfoliatus from Big Point during our study. Competition with other established aquatic plants, particularly in nutrient-rich floodplain habitats, may also limit its distribution. ACKNOWLEDGMENTS This research was partially funded by the UNO Urban Waste Manage- ment and Research Center and a kind gentleman who would like to remain anonymous. REFERENCES BRANTLEY, C.G. and S. if Prarr. 1991. Decne ence of Potamogeton perfoliatus L. (Potaaio pecan aiese) in Louisiana. Bae 14:617-618 Burns J.W., M.A. Poirrier, and K.P. Preston. 1993. iets of urban runoff on the envi- oneal quality of Lake Donechuretain: Louisiana. Sub-Project: Effects of New Or- leans urban runoff on the distribution and structure of submerged aquatic vegetation communities in Lake Pontchartrain, LA. University of New Orleans Urban Waste Man- agement & Research Center, Research Reporte No. 92-05. Cuasreck, R. 1972. Vegetation, water and soil characteristics of the Louisiana coastal region. Louisiana Share Univ. Agric. Exp. Sta., Bull. No. 664 Den Hartroc, C. 1981. Aquatic a communities of poikilosaline waters. Hydrobiologia 81:15—22. DENNISON, W.C., R.J. OrrH, K.A. Moore, J.C. Stevenson, V. Carrer, S. Koitar, PW. BERGSTROM, and R.A. Batiuk. 1993, eae water quality with pees: aquatic aia Habitat requirements as barometers of Chesapeake Bay health. BioScience 86-94. FoLk, RL. pee 7 of sedimentary rocks. Hemphill Publishing Co., Austin. Goprrey, R.K. < .W. Wooren. 1979. Aquatic and wetland plants of southeastern United States ee ee The University of Georgia Press, pee Harvey, H.W. 1957. The chemistry and fertilicy of sea waters. Cambridge University Press, London. Haynes, 1968. Potamogeton in Louisiana. Proc. Louisiana Acad. Sci. 31:82—90. Hur.ey, 1990. Field guide to the submerged aquatic vegetation of Chesapeake Bay. U.S: a aa Wildl. Serv., Chesapeake Bay Estuary Program, Annapolis. Burns ET AL., Potamogeton perfoliatus in Lake Pontchartrain 763 Kartesz, J.T. 1994. A — checklist of the vascular flora of the United States, Canada, and Greenland. Timber Press, Portland. Lester, G. 1988. Plants and animals of ig concern in the Louisiana coastal zone. Loui- siana Wildl. and Fish. Comm., Louisiana Nat. Heritage Progr., Special Pub. No.2 Mayer, M.S. 1986. The eee aquatic vegetation of the Lake Pontchartrain Estuarine System, Louisiana. M. S$. Thesis. Univ. New Orleans, New Orleans, Louisiana Montz, G.N. 1978. The submerged vegetation of Lake nee ees Tee Cine 43:115-128. Open, E.C. 1943. The broad-leafed ae of ii ipa al of North America north of Mexico. Rhodora 45:57—105, 119-163, 171- Ortu, R.J., J.B Nowak, G.E oe. = ia and J.R. Wuirinc. 1992. Distribu- tion of submerged aquatic vegetation in the Chesapeake Bay and tributaries and Chincoteague Bay - 1991. Virginia Institute of Marine Science, School of Marine Sci- ence, College of William and ee Gloucester Point, Perret, W.S., B.B. Barrerr, W.R. Latapie, J.F POLiarp, WR. Mock, G.B. ADKINS, W.J. Gatpry, and C.J. Wuire. 1971. oe Gulf of Mexico estuarine inventory and study, Louisiana. Phase I, area description. Louisiana Wildl. and Fish. Comm., New leans. ea ScutrHorPe, C.D. 1967. The biology of aquatic vascular plants. Edward Arnold Ltd., Lon- SIKORA, W.B. and B. Kyerrve. 1985. Factors influencing the salinity regime of Lake rain, ne ae a shallow coastal lagoon: Analysis of a long-term data set. Pon ee 8: 80. STANDARD METHODS. 1989, Chloride. In: L.S. Clesceri, A.E. Greenberg, R.R. Trussell, eds. Standard methods for the examination of water and wastewater, 17th ed., American Public Health Assoc., Aes Water Works Assoc., Water Pollution Control Federa- tion, Washington, DC. Pp. 67— SutrKus, R., R. DARNELL, and J peau + Biological study of Lake Pontchartrain, Louisiana. Tulane ean New Orlean Tuomas, R.D. and C.M. ALLEN. 1993. Atlas of the vascular flora of Louisiana. Vol. I: Ferns soa fen allies conifers, ee monocotyledons. Louisiana Wildl. and Fish., Baton Rouge. THompson, B.A. and J.S. VERRET. Nekton of Lake Pontchartrain, Fon siane and its surrounding pre ds. Pp. 711-864. In J. H. Stone (Ed.), Environmental analysis of Lake Pontchartrain, Louisiana, its Suntoundine wetlands and ene land uses. Coastal Ecology ae Center for Wetland Resources, Louisiana State University, Baton R = Turner, R., R. Darnew, and J. Bonp. 1980. Changes in the supe macrophytes of Lake ne ee (Louisiana): 1954— ae North. Gulf Sci. 4:44-49. Twitty, R.R. and J.W. Barko. 1990. rrowth of eer macrophytes under ex- perimental salinity and light paneer Estuaries. 13:311—321. WesTMan, W.E. 1985. Soe impact assessment, and ed planning. John Wiley and Sons, New Yor 764 Stipa 16(4) BOOK NOTICES GRETCHEN D. Jones, VAUGHN M. Bryant, Jr., MerepirH HoaG Lirux, STAN- LEY D. JONEs, and Pere D. LINGREN. 1995. Pollen of the Southeast- ern United States: With emphasis on Melissopalynology and Entomopalynology. (ISSN 0160-8843, AASP Contribution Series 30, hbk, “Hidden wire’O” metal spiral binding). AASP Foundation, c/o Vaughn M. Bryant, Palynology Laboratory, Texas A&M Univer- sity, College Station, TX 77843-4352. $27.00. 76 pp, 616 b/w pho- tographs on 104 photographic plates. Palynologists are always in need of additional reference materials. As the authors of this book point our, it would be impossible for any one researcher to know the pollen ey ON ogy of all of the more than 250,000 flowering plant species. This compilation o scanning Fe micrographs of pollen species from the southeastern United States in- cluding Texas will serve as a useful aid to anyone working on pollen from this region. Although ree book emphasizes pollen of are age is broad enough to be of significant value aa Ous plant taxa, the taxonomic cover- Positive features of this book include a eompreneaave listing of other published pollen floras, detailed descriptions of techniques used in scanning electron microscopy, voucher specimen preparation, and collection of pollen from voucher specimens, readily accessible and clear listings by family, genus and aperture type of pollen taxa figured in this book, and lastly, high quality prints of the 616 pollen image The authors made the photographs primarily for i who study pollen associated with insects by using scanning electron microscopy. Thus, there are no light micrographs for direct comparison with views of pollen under a light microscope. Nevertheless, the images remain valuable to all palynologists working with southeastern assemblages. There is now more in- formation on which to base pollen identifications, and that’s good for us all.—Bonnie Jacobs. 4 Reip, WaLrer V., SARAH A. Latrpb, Carrie A. MEYER, RODRIGO GAMEX, NA SITTENFELD, DANIEL H. JANZEN, MICHAEL A. GOLLIN, AND CALESTOUS Juma. 1993. Biodiversity Prospecting: Using Genetic Resources for Sustainable Development. (ISBN 0-915825-89-9, pbk). World Resources Institute, PO. Box 4852, Hampden Station, Baltimore, MD 21211 (1-800-822-0504). No Price Given. 341 pp. The aia chapters are included in the book: Foreword and Acknowledgments. 1) Nev Life; 2) Costa Rica’s Conservation Program and National Biodiversity eae (INBio); 3) Biodiversity Prospecting by INBio; 4) Contracts for Biodiversity Prospecting; 5) Research Management Policies: Permits for Collecting and Research in the Tropics; 6) An Intellectual Property Rights Framework for Biodiversity Prospecting; 7) Policy Options for Scientific and Technological i oases Annexes: a) The Role of the Parataxonomists, Inventory Managers, and Taxonomists in Costa Rica's Na- tional Biodiversity Inventory, b) Biodiversity Prospecting Contract c) The Convention on Biological Diversity and Intellectual Property Rights, d) The United Nations Convention on Biological Diversity. =) Stipa 16(4): 764. 1995 DOCUMENTED CHROMOSOME NUMBERS 1995:1. CHROMOSOME NUMBER OF CORNUS SESSILIS (CORNACEAE): PHYLOGENETIC AFFINITY AND EVOLUTION OF CHROMOSOME NUMBERS IN CORNUS QIU-YUN XIANG and RICHARD H. EYDE! Department of Plant Biology Ohio State University 1735 Neil Avenue Columbus, OH 43210-1293, U.S.A. ABSTRACT The chromosome number of the Californian ana cherry, Cornus sessilis Torr., is determined to be 27=20, differing from the number (27=18) reported for other cornelian cherries. This is the first report of a chromosome ee for C. sessilis. The finding of 2n=20 in C. sessilis along with evidence from morphology and molecular data helps clarify the evolutionary trend of chromosome numbers in Corzus and hoa an early divergence between the species and its close relatives, C. mas - C. officinalis - C. chinensis. RESUMEN El determina el nimero cromosémico del corncjo de California Cornus sessilis Torr. Como 2n=20, que difiere del nimero (2=18) citado de otros cornejos. Este es el primer recuento cromosémico de C. sessilis. El hallazgo de 27=20 en C. sessilis junto con la evidencia de datos morfolégicos y moleculares ayuda a clarifica las tendencias evolutivas de los numeros cromosémicos en Cornus y sugiere una divergencia temprana entre esta especie y sus parientes mas proximos, C. mas - C. officinalis - C. chinensis. The cornelian cherries, Cornus subgen. Cornus L. consist of five geographi- cally isolated, red-fruited species (C. mas L. in Europe; C. officinalis Seib. et Zucc. in eastern China and Japan; C. chinensis Wangerin in southwestern China; C. sessilis Torr. in California; and C. volkensii Harms in Africa). Chro- mosome numbers of 27=18 have been reported for C. mas and C. officinalis (Dermen 1932). This paper is the first report of a chromosome number for the Californian species C. sessilis, and discusses the phylogenetic affinity of the species and the evolutionary trend of chromosome numbers in Cornus. METHODS AND RESULTS Root-tips from germinating seeds of Cornus sessilis were pretreated with saturated p-dichlorobenzene for three hours before fixation in a solution of 'Deseased May, 1990. Sipa 16(4): 765-768. 1995 766 Sipa 16(4) 3:1 ethanol:glacial acetic acid. Seeds were collected in the field (voucher: Eyde 134 (US), northern California, May 1989) and germinated in the green- house. Chromosome counts were made from 15 mitotic cells of root-tip material using standard squash procedures. All cells were determined to be 2n=20 (Fig. DISCUSSION The distribution pattern of the cornelian cherries suggests an old age for the group. The geographic isolation of the various cornelian species paral- lels the morphological diversity found in the group. For example, the Chi- nese cornelian cherry, Cornus chinensis, is distinct in the genus in having monopodial axes (i.e., terminal leaf buds and axillary flower buds; Xiang 1987) whereas all other species of Cornus have sympodial axes (terminal flower buds and axillary leaf buds). The African cornelian cherry, C. volkensiz, is unique in the genus in being dioecious, and the Californian cornelian cherry, C. sessi/zs, is distinct from other cornelian cherries in its winter bud morphology. A winter bud of the cornelian cherries (except C. chinensis) consists of three buds, two lateral leaf buds each with a scale at its outer side, and one terminal inflorescence bud covered by four bracts. The leaf- bud scales in C. sessilis are modified and expand to cover completely the inflorescence bud to form the outer-most layer of protective sheaths for the inflorescence bud. In contrast, the leaf-bud scales in other cornelian cher- ries are small and not modified. Also, the peduncules of the preformed inflorescence of C. sess#/is are not precocious as they are in other cornelian cherries (Murrell 1993). As a result of the morphological diversity, the taxonomy of the cornelian cherries has been controversial. Cornus chinensis and C. volkensii have at times been separated from other cornelian cherries (C. mas, C. officinalis, and C. sessilis) and have been recognized as distinct subgroups within Cornus or as a distinct genus, such as, Cornus subgen. Sinocornus tor C. chinensis (Xiang 1987), and Afrocrania or Cornus subgen. Afrocrania for C. volkensii (Ferguson 1966a; Hutchinson 1942: Murrell 1993: Xiang 1987 Phylogenetic relationships within the cornelian cherries have been pro- posed by Eyde (1988). Based on morphological and fossil analyses, Eyde (1988) placed Cornus sessilis as the sister of a clade that consists of C. officinalis, C. mas, and C. chinensis with the first two as sisters. Eyde considered C. sessilis a line that diverged second in the cornelian cherry group, following the divergence between C. volkensii and the remaining cornelian cherries. This hypothesis is supported by molecular data. The result of a recent chlo- roplast DNA (cpDNA) restriction site analysis of Cornus by Xiang et al. (in press) revealed that C. sessz/zs is sister to C. mas - C. officinalis (C. chinensis and C. volkensii were not included in the cpDNA restriction site study). NOTrES 567 & . Mitotic chromosomes of root-tip cells from germinating seeds of Cornus sessilis Torr. ee ae Metaphase Furthermore, a phylogenetic analysis of a combined molecular data set of the cpDNA restriction sites and the rbcL - matK sequences (Xiang, unpubl.) also recognizes C. sessilis as a distinct lineage sister to a clade containing C. mas, C. officinalis, and C. chinensis with C. mas and C. officinalis as sister species. In addition, a cladistic analysis of Cornus using morphological data by Murrell (1993) suggested relationships within the cornelian cherries identical to those proposed by Eyde (1988). A high number of mutations were detected between C. sessi/is and other cornelian cherries. For example, 11 restriction site mutations between C. sessilis and C. mas - C. officinalis, and a total of 22 mutations (including restriction site mutations and all base substitutions in récL and matK) between C. sessilis and C. mas - C. officinalis - C. chinensis were found (Xiang et al. in press; Xiang unpubl.), suggesting an early divergence of the species in the cornelian cherry group. A different chromosome number in Cornus sessilis (2n=20 rather than 27=18) and the morphological divergence of the species also support a long history of isolation of C. sessé/zs from other cornelian cherries. The cornelian cherries have long been known to have a chromosome number of 2”=18, which was reported for Cornus mas and Cornus officinalis (Dermen 1932; also see Ferguson 1966b). The chromosome numbers of the other three species in this group have remained unknown. In Cornus, a 768 Stipa 16(4) chromosome number of 27=20 has also been documented in the two alter- nate-leaved, blue-fruited dogwoods, C. controversa Hemsley and C. alternifolia L. f.. In addition, 2”=22 has been reported for the big-bracted dogwoods, the dwarf dogwoods, and the opposite-leaved, blue-fruited dogwoods (Bain & Denford 1979; Dermen 1932). The close relatives of Cornus, such as Alangium, Mastixia, and Nyssa have also been reported to have a basic chro- mosome number of x=11 (Goldblatt 1978). Therefore, 2”=22 might be plestomorphic in Cornus, and 27=20 and 18 both represent derived states as proposed by Eyde (1988). The molecular phylogeny of Cornus (Xiang et al. in press; Xiang unpubl.) is consistent with this hypothesis. The finding of 2n=20 in C. sessilis provides new insight regarding the evolutionary trend of chromosome numbers in Cornus. Given the cpDNA phylogeny, the chro- mosome number of 27=20 could have evolved twice in the genus, once in the alternate-leaved blue-fruited species and once in the cars cherries be- fore the divergence between C. sessilis and C. chinensis - C. officinalis - C. mas. The chromosome number of 27=18 found in C. officinalis and C. mas was derived from ancestors having 27=20 (see discussion in Xiang et al. in press). ACKNOWLEDGMENTS Iam grateful to Daniel J. Crawford for reading the manuscript and pro- viding valuable comments. This work was conducted in 1989 in the late Dr. Richard H. Eyde’s lab at Smithsonian Institution and supported by a fellowship provided by the Smithsonian Institution for short-term visitors. REFERENCES Bain, _ and K.E. DENForb. se The herbaceous members of the genus Cornus in NW 1 America. Bot . 132:121-129 Beene 1932. Cyl ee studies of Cornus. J. Arnold Arbor. 13:401—417. Eypr, RH. 1988. suas een Cornus: Puzzles and progress in the systematics of the dogwoods. Bot. Rev. 54 —351. FerGuson, LK. 1966a. aie on the nomenclature of Cornus. J. Arnold Arbor. 47:100—105. —______.. 1966b. The Cornaceae in the southeastern United States. J. Arnold Arbor. 47:106-116. GotpsLatr, P. 1978. A contribution of cytology in Cornales. Ann. Missouri Bot. Gard. 65:650-655 Hurcuinson, J. 1942. Neglected generic characteristics in the family Cornaceae. Ann. t. (London), ser. 2, 6:83—93 Murrell, Z.E. ee Phylogenetic mn in Cornus (Cornaceae). Syst. Bot. 18:469-495. XIANG, 0.-Y. - . A neglected character of Cornus L. s. 1. with special reference to a new ben Q. Y. Xiang. Acta pene Sin. 25:125-131 eT S| BRU ELD, DE Sottis, and P.S. Souris (In press). Phy tionships in Cornus based on chloroplast DNA restriction sites: Implications for bioge- ography and character Evolution. Syst. Bot. — ogenetic rela- NOTES NEW REPORTS OF ERAGROSTIS (POACEAE: CHLORIDOIDEAE) FROM BRAZIL—Eragrostis Wolf is the largest genus in the subfamily Chloridoideae (tribe Eragrostideae) with +350 species occurring in the trop- ics and subtropics throughout the world (+120 in the New World and +225 from southern Africa). There is no modern account of the entire ge- nus and although the species appear very similar to one another, relation- ships within the genus are only speculative at this time. During prepara- tion of the taxonomic revision of Eragrostis in Brazil, we became aware of nine species that were not reported in the literature (Allem & Valls 1987; Braga 1976, Pilger 1902; Usteri 1911). In addition, Eragrostis paniciformis and E. tremula are new records for the New World We consulted the literature for an indication of how common each of the nine species was in other countries. Eragrostis macrothyrsa and E. orthoclada are reported from Argentina, Bolivia, and Paraguay (Killeen 1990; Zuloaga et al.1994). Eragrostis barrelieri is mentioned as being introduced in United States, Mexico, Jamaica, Haiti, Puerto Rico, Dominican Republic, Virgin Islands, Uruguay, Argentina (Adams 1972; Beetle et al. 1991; Hitchcock 1936; Liogier & Martorell 1982), Africa (Gibbs Russell et al. 1991), Burma, India, Somalia (Cope 1995), Pakistan (Bor 196O), Canada, and Panama (Peterson & Harvey in on Peterson in press). Eragrostis pastoensis occurs in Argentina (Zuloaga et al. 1994), Venezuela (Graterol et al. 1989), C lombia, Peru (Brako & Zarucchi 1993), Ecuador (Hitchcock 1927), Bolivia (Hitchcock 1927), Paraguay, and Uruguay (Peterson & Harvey in press; Peterson in press). Eragrostis gangetica, E. lebmanniana, E. mokensts, E. paniciformis, and E. tremula are primarily African species (Clayton 1972 Clayton et al. 1974; Gibbs Russell et al. 1991; Ibrahim & Kabuye 1987; Koechlin 1962; Zon 1992). In the New World, E. ees E.mokensis, and E. lebmanniana are reported in Venezuela (Graterol et al. 1989). Eragrostis gangetica and E. lebmanniana are also found in the United States (Correll & Johnston 1970; Davidse 1994; Peterson & Harvey in press), with the latter species extending into Mexico (Lebgue & Valerio 1986) and the former found in Belize (Davidse 1994). The following section lists the nine new records and includes informa- tion taken from the specimens. Species Citations 1. Eragrostis ire Daveau, J. Bot. pie 8:289. 1894. BRAZIL. No specific location, 1892 M.Glaziou 20119 (US 55412, 1162998). 2. es gangetica (Roxb.) Steud., Syn. PL. Glumac. 1:266. 1854. BRAZIL. Paraiba: Stipa 16(4): 769. 1995 770 Sipa_ 16(4) Campina Grande, 24 Jun 1935, B.Pickel 3832 (NY,US). Pernambuco (3 km to Jurema- Fazenda Boca da ie 780 m, 26 Jul 1966, E. Tenorio 66-138 (NY,US). . Eragrostis lehmanniana Nees, Fl. Afr. Austral. Ill. 402. 1841. BRAZIL. Ceara: Poredew nes vivo do Instituto de Zootecnia), collector unknown (SP 165964). 4. Eragrostis macrothyrsa Hack., Repert. Spec. Nov. Regni.Veg. 8:47. 1910. BRA- ZIL. Mato ce o: Miranda (a 6 km Je sede da Fazenda a no Rumo ESE da oe - Jun 1973, TS.Silva 100,107 (SP.US), TS. Silva 101 (SP 5. Eragrostis mokensis Pilg., Bot. Jahrb. Syst. 51:419-420. a BRAZIL Minas a Ouro Preto (Vila Rica), 100 m, 7 is 1925, A.Chase 9358 (NY,SP.US). 6. Eragrostis orthoclada Hack., Bull. Herb. Boissier ser.2, 4(3):281. o 4. BRAZIL. Mato Grosso do Sul: Anastacio, 22 Apt Valls et al. 8657 (CEN); Aquidauana, 5 Apr 1986, |. EM. Valls et al. 9891 (CEN), Bela Vista, 18 Jun 1946, J. Swallen 9469 (US), 21 Apr 1984; J.EM. Valls et al. 7654 (CEN); Corumba (on Rio paras), 200-250m.,31 Apr 1930, A.Chase 11124 (US). Miranda, 11 Jun 1973, Silva 85B (SP). 7. Eragrostis paniciformis (A.Br.) Steud., Syn. Pl. Glum. 1:268. 1854. BRAZIL. Sao Paulo: Caraguatatuba (3 km ao sul do Rio Mococa), 50 m, 27 May 1983, /. E M. Valls 7385, 7386 (CEN, ICN); 28 May 1986, |.EA ee 10270 : EN, ICN). Sao Luiz do Paratinga, 940 m, 29 May 1986, J.EM. Valls et al. 10280 (CEN, ICN). . Eragrostis pastoensis (Kunth) Trin., ees Imp. Acad. Sci. Saint-Petersbourg, Ser. 6, Sci. Math., Seconde Prt. Sci. Nat. 2 (1):71. 1836. BRAZIL. Maranhao: Grajau to Porto Franco, 8-13 Mar 1934, J.Swallen 3870 (US). Mato Grosso: Caceres, 25 Oct 1985, JEM. Valls 9381 (CEN), Campo Grande, E.F Nienstedt 149 (US); Sao Paulo: Perus, 25 Mar 1907, A. Ustert s.n. (SP 1012 9. Eragrostis tremula Steuc d. , Syn. Pl. Glum. 1:269. 1854. BRAZIL. No specific loca- tion, 21 Nov 1952, Camargo 6 (US 2205753). Serra ACKNOWLEDGMENTS The first author would like to thank the directors of Brazilian herbaria for the loan of specimens and to CAPES for financial support. Appreciation is extended to numerous people in the Department of Botany, Smithsonian Institution and especially to Marjorie Knowles for technical assistance. Sonja De Castro Boechat, Departamento de Botanica/UFRGS, Av. Paulo Gama, 40, 90049.060, Porto Alegre, RS, Brasil, and Paul M. Peterson, Department of Botany, National Museum of Natural History, Smithsonian Institution, Wash- mgton, DC 20560, U.S.A REFERENCES Apams C.D, 1972. The flowering plants of Jamaica. University of the West Indies, Mona Jamaica. AtteM, A.C. and J.R.M. Vauts. 1987. Recursos forrageiros nativos do Pantanal Mato- Grossense. Empresa Brasileira de Pesquisa Agropecudria Centro Nacional de Recursos Ecos Brasilia, BEETLI Be sin Manniour Forceck, }.A. MIRANDA SANCHEZ, V. JARMILLO LUQUE, A. CHIMAL ieee and A.M. RopreGcuez Ropricurz. 1991. Gramineas de Mexico. Tomo III. Secretaria de Agricultura y Recursos Hidréulicos, Mexico. Pp. 50-96 Sipa_ 16(4): 770. 1995 NOTES rai Bor, ee ne The grasses of Burma, Ceylon, India, and Pakistan. Pergamon Press, New Yor 495-516. Bree - 1976. oe do nordeste especialmente do Ceara. 3 ed. Mossoré, Escola Supe- r de Agricultur Bano, L. and J.L. 7 aRuC HI. 1993. Catalogo de las aN ial y gymnospermas del eru. Monogr. Syst. Bot. Missouri Bot. Gard. 45:945— a AYTON, W.D. 1972. Gramineae. In: J. Hutchinson, J.M. oe and EM. Hepper, Eds., ra of West Tropical Africa. Crown Agents for Oversea Governments and Admin trations. Pp. 383-392. ape W.D., M. Puytups, and S.A. Renvoize. 1974. Gramineae. In: R.M. Polhill, ed. ora of Tropical East Africa. London, Crown Agents. pt 2. Pp. ae 244. pn T.A. 1995. Some new Somali grasses I. Kew Bull. 50:10 CorRELL, DS. and M.C. JoHNsTon. 1970. Manual of the as . of Texas. Texas Research Foundation, Renner, Texas. Pp. 204—2 Davinse, G. 1994. Eragrostis Wolf. In: Davidse, Sousa, and A.O. Chater, eds. Flora oamericana. Universidad Nacional Dascnone de México, Missouri sie Gar- — and The Natural History Museum-London, Mexico. vol. 6. Pp. 263-2 Gipss RUSSELL, Watson, M. Korkemoer, L. SMook, N.P. BARKER, a ANDER- SON, and MJ. Sa 1991. Grasses of Southern Africa. In: O.A. Leistner, ed. Mem. Bot. Surv. S. Africa 58:139-163. Graterot, A., P. Torreciia, and B.K. Trujitto. 1989. Comentarios criticossobre el ane Eragrostis Wolf en Venezuela y clave para la identificacion de sus especies. Erns 51:16-28. Hircucock, A.S. 1927. The grasses of Ecuador, Peru, and Bolivia. Contr. U.S. Natl. Herb. 24:335-344. Hircucock, A.S. 1936. Manual of the grasses of the West Indies. U.S.D.A. Misc. Publ 243:1-439 IBRAHIM, K.M and C. Kasuye. 1987. i illustrated manual of the grasses of Kenya. EA.O., United Nations, Rome. Pp. 146-16 Kueen, T. 1990. The grasses of ak Santa Cruz, Bolivia. Ann. Missouri Bot Gard. 77:125-201. Korcuun, J. 1962. Famille des Graminees. In: A. Aubreville, Flore du Gabon. Museum National d’Histoire Naturelle, Paris. Pp. 221-231. Lescue, T. and A. VALERIO. Chihuahua, Mexico. Pp. 95— Liocier, A.H. and L.F ee Manual para identificar las Gramineas de Chihuahua. eos LL. 1982. Flora of Puerto Rico and ace islands: systematic synopsis. Editorial de la Universidad de Puerto Rico, Puerto Peterson, P.M. and L.H. Harvey (In Press a). Eragrostis Wolf. In: Ric M.E. ace ed. Manual of North American grasses. 83 manuscript p p. Prrerson, PM. (In Press.) Eragrostis Wolf. In: S. Laegaard, ed. Grasses of Ecuador. 78 manuscript pp. Pucer, R. 1902. Beitrag zur flora von ee am Bot. Jahrb. Syst. 30:128-141. Usrert, A. 1911. Flora der Umgebung der Stadt Sao Paulo. Weimar, ZON, VAN DER A.P.M. 1992. —— du Cameroun v.II. Wageningen on Uni- versity, Cameroon. Pp. | 37. ZULOAGA, E.O., E.G. es Z.E. RUGOLO DE AGRASAR, O. Morrone, J. PENsIERO, and A.M. CIaALpeLLa. 1994. Catalogo de la familia Poaceae en la Republica Argentina. Monogr Syst. Bot. Missouri Bot. Gard. 47:62—66 Stipa 16(4): 771. 1995 772 Sipa 16(4) CAREX CONJUNCTA (CYPERACEAE) VERIFIED FOR ARKAN- SAS, AND NOTES ON THE RANGE OF CAREX OKLAHOMENSIS— Species of Carex subgenus Vignea, section Vulpinae, are plants of swampy sites, buffalo wallows, open-hydric roadside ditches, and wet meadows. Section Vilpinae is composed of about 17 species, eleven of which are widely distributed in North America (Mackenzie 1931). Smith (1994) marked Carex conjuncta WW. Boott as a possible addition to Arkansas but had no verifiable specimen. Its distribution has been given as New York and New Jersey to the District of Columbia, and westward to South Dakota and eastern Kansas (Mackenzie 1931). Gleason and Cronquist (L991) gave a similar distribution, New York to Minnesota and South Da- kota, south to Virginia and eastern Kansas. Our collections verify C. conjaucta as occurring in Arkansas Until recently, the range of Carex oklahomensis was considered to be from SW Missouri (Steyermark 1963) and adjacent Kansas (Great Plains Flora Association 1991) south through western Arkansas and eastern Oklahoma to northeastern Texas (Mackenzie 1931; Jones et al. 1991). Carex oklahomensis was considered a hybrid by Smith (1988, 1994), and not mapped for Ar- kansas, but many Arkansas records for C. stipata belong here. This species is now known from Mississippi (Bryson et al. 1992, 1994). Reported here are recent collections extending its range to southeastern Missouri and into the “Boot Heel,” and one isolated collection from western North Carolina. Almost all of the recent eastern records, including all the Mississippi records, are from along roadsides, suggesting that the species has recently spread eastward, rather than having been merely overlooked. Carex oklahomensts should be watched for in other southeastern states. Both of these species are members of subgenus Vignea section Vulpinae and are characterized by distigmatic flowers, lenticular achenes, compound sessile androgynous spikes, and bidentate perigynia with beaks no more than half the length of the perigynial body. Carex conjuncta grows to over one m tall with cespitose, erect but soft and easily crushed, somewhat spongy, winged, scabriously margined culms. It grows in damp woods and shady creek banks. The dorsal leaf sheaths are green with septate nodules; ventral leaf sheaths fragile, w tered red dots, and cross rugulose. — on —_— ute with scat- Carex oklahomensis is a loosely cespitose, stout, erect sedge growing to over one meter tall. Its habitat includes open wet sites, usually in calcare- ous or basic soils, and it is a facultative to obligate heliophyte. Its blue- green dorsal leaf sheaths with conspicuous white dots without septate nod- ules are characteristic. The perigynia are relatively short beaked in contrast Sipa 16(4): 772. 1995 Nores a75 to most species of section Vi/pinae. In western Missouri, where some native prairies have been preserved, this taxon is frequently found in what appears to be historic buffalo wallows, however, it can also be found in open-hydric roadside ditches and along streams through prairies. In southeast Kansas, we found this plant along open-hydric roadsides. As used in the following key, incomplete veins are veins that do not extend from the base of the perigynia to the apex. A heliophyte is a plant that is normally found in full sun; a sciophyte is a plant normally found in the shade. Although, all taxa are not in all states the following artificial dichotomous key is for section Vi/pinae in the southeastern United States. KEY TO CAREX SECTION VULPINAE IN THE SOUTHEASTERN UNITED STATES ike — of perigynia shorter than the body pe es somewhat abruptly contracted into a beak ca. 1/2 the length of the perigynia body, ventral surface of perigynia with several incomplete veins basely; culms sharply triangular and narrowly winged, somewhat spongy and easily crushed; dorsal leaf sheath green; ventral leaf sheath with scattered red dots, transversely rugose; sciophytes C 2. Perigynia tapering into a beak, much shorter than the perigynia body, ventral surface of . conpuncta perigynia with several inconspicuous complete veins; culms inconspicuously triangular to roundish without wings, not spongy and not easily crushed; dorsal leaf sheath dark blue-green with conspicu- ous white dots; ventral leaf sheath without scattered red dots, not trans- versely rugose; heliophytes C. oklahomensis — S . Beaks of perigynia as long as or longer than the body 3. Ventral leaf sheath margins with orange-red dots; achenes ovate-lanceolate; serigynial wall adhering to achene C. cris-corvi — 3. Ventral leaf sheath margins without orange-red dots; achenes broadly ovate to ovate-orbicular; perigynial wall little to not at all adhering to achene 4. Ventral leaf sheath transversely rugose, + convex at apex and prolonged upward past the base of the blade, friable; eae includes C. wherior (C. Mohr) K. Mackenzie = C. stipata var. maxi > Chapman asec wired C. stipata . Ventral leaf ee he not transversely rugose, + cave at apex and not oe upward past the base of the Shae pet not friable C. laevivaginata Specimens collected: (Carex conjuncta) ARKANSAS. Benton Co.: NE side of Osage Creek and US 412/AR 68, E of eae Springs, mesic woodland edge along berm of creek, ‘a. 335 m, 26 May 1994, S. 7. Jones 11139 and A, A. G&S. A. Reznicek (BRCH Mic _ UARK). - side I-412, on ae of Osage Creek, ca. 9 mi E of Siloam Springs, 33, TISN, R32W, 4.1 mi “ oF the Washington Co. line, wet creek bottom pasture oe elev. ca. 335 m, 26 Ma 4A. Reznicek 9792 and S. A. Reznicek, S. D. Jones, and G. D. Jones (BRCH, MICH, MO. UARK). Associated species included other Carex spp., Sina spp., Juncus, Scirpus, Scleria, Plantago, Laportea, Festuca, and Verbesina. Specimens collected: (Carex ies MISSOURI. Dunklin Co.: E side Co. rd 203, > Stipa 16(4): 773. 1995 174 Sipa 16(4) se sect. 22, T22N, R82, ca. 7 mi NW of Campbell, 22 May 1993, A. A. Reznicek et al. 9432 (BRCH, ctb, KNK, MICH, MO). Stoddard Co.: cei Crowley Ridge Roadside Park, along N side rte. 50, 22 May 1988, — 21 1962 (MICH); S side of US 60, 1.4 mi W of jet with Co. rd TT at Dudley, 22 May 1993, Reznicek et al. 9420 (I 3RCH, ctb, KNK, FTG, GENT, MICH, MO, VDB, VPI). NORTH CAROLINA. Graham Co.: ee: road to Joyce Kilmer Memorial Forest, 7.6 mi from jet with Hwy 129 at Robbinsville, 15 May 1984, Reznicek & Reznicek 7339 (MICH, NCU, NY). aley D. Jones, Botanical Research Center. BRCH Herbarium, P. O. Box 6717, Bryan, TX 77805-6717, U.S.A. and A. A. Reznicek, MICH Herbarium, University of Michigan, North University Building, Ann Arbor. MI 48109- TOD 7. REFERENCES Bryson, C. T., R. E C. Naczi, and S. McDanieL. 1992. Notes on noteworthy records of Carex (Cy] eee an the southeastern United States. Sida 15:125—135 Bryson, C. T., J. R. MacDonatp, and R. Warren. 1994, Notes on Carex (Cyperaceae), with C. Cilia new to Alabama and ©. communis and C. scoparia New to Mississippi. Sida 16:355—361. Gieason, H. A. and A. Cronquisr. 1991, Manua States and adjacent Canada. (2nd ed.). New York ene Garden, Bronx, GREAT PLAINS FLORA PO GIATION: 1991. Flora of the Great Plains. University Dress of Kansas, goers Jones, 8. D., G. D. ee and $. L. Harcu. 1991. The deletion of Carex stipata (Cyperaceae) from the a flora. Phytologia 71: Mackenzig, K. K. 1931. North Nenean flora. 18. Cyperaceae, tribe 2, Caricae. New York Botanical Garden, Bronx, NY. SmitH, E. B. 1988. An atlas a as list of the vascular plants of Arkansas. (2nd ed.). Published by author, Fayetteville, SmitH, E. B. i Keys to the flora of hiassas: The University of Arkansas Press, Fay- of vascular plants of northeastern United oa ae etteville, Sieeciniane: a _ 1963. Flora of Missouri. The Iowa State University Press, Ames, IA. Stipa 16(4): 774. 1995 775 NOTEWORTHY PLANT SPECIES FROM THE OKEFENOKEE SWAMP, GEORGIA—The Okefenokee Swamp is located on the Atlantic Coastal Plain in southeast Georgia and northeast Florida. The swamp sur- face area of 3826 km? includes parts of Ware, Clinch, Charlton, and Echols counties, Georgia, and Baker County, Florida. There are a number of dis- tinct habitat types within the swamp, including swamp forests, tree and shrub islands, grass-sedge marshes, emergent macrophyte marshes, and open lakes (Greening & Gerritsen 1987). Sand and peat are the two soil types in the swamp. Marine or lacustrine sandy soils occur in large islands, uplands, and immediately east of the Suwannee River Sill, i.e. a low earthen dam impounding the Suwannee River (Parrish & Rykiel 1979). The majority of the swamp substrate is peat (96% organic matter) up to 5m deep (Cohen LOTS): The flora of Okefenokee has been of interest to botanists since the early 1900s when Roland M. Harper made the first extensive investigation of the swamp vegetation (Trowell 1988). Over the following years further description of the swamp flora has been provided by Wright and Wright (1932), Cypert (1961; 1972; 1973), and Schlesinger (1978). The most com- prehensive checklist of the vascular plants of Okefenokee was compiled by Loew and Jones (1984) from plant surveys conducted during the growing seasons of August 1978—September 1980. Their checklist documents 101 species of vascular plants representing 53 families. The large surface area and inaccessibility of remote regions of the Swamp have hindered comprehensive botanical surveys. As a result a few species, particularly sedges, have been overlooked. This manuscript documents spe- cies which should be added to the checklist of vascular plants occurring in the Okefenokee Swamp, Georgia. The observations reported herein result from more than two years (July 1992—December 1994) of collecting and observations made in conjunc- tion with a project entitled “The Effects of Hydrologic Alterations on the Ecology of the Okefenokee Swamp” (Loftin et al., unpublished data). Plants were collected by various methods. Many specimens were collected during species composition, biomass, and cover estimates along 80 permanent transects established in four emergent macrophyte prairies (Chesser, Durdin, Floyd’s, and Sapling Prairies), and the Suwannee River floodplain near the Sill. Some specimens were obtained as seedlings germinating from soil cores collected along the permanent transects. These soil cores were used in seed bank experiments, and seedlings were germinated from the cores under greenhouse conditions near the Suwannee Canal Recreation Area (Charlton Co., Georgia). Before harvesting the sample, the seedlings were grown un- Sioa 16(4): 775. 1995 776 Stipa 16(4) til an inflorescence was produced. Many specimens were also collected dur- ing routine forays in and around the swamp. NOTEWORTHY SPECIES The following is a list of nine noteworthy species collected from the Okefenokee Swamp during the course of our investigations. Although most specimens (except Rhynchospora alba) are fairly common species reported previously in Georgia, there is no documentation for these species from the Okefenokee Swamp. These species should be added to the checklist of vas- cular plants of Okefenokee. Voucher specimens for those species described herein have been donated to the herbaria of the University of Georgia, Athens (GA), and the University of Florida, Gainesville (FLAS). As noted previously, some specimens were obtained from soil cores during seed bank experiments. For these specimens, the locality stated is the site were the soil core was obtained. Although all species described herein were observed in the field, the soil core specimens were of herbarium-quality. Nomencla- ture follows that of Godfrey and Wooten ( 1979a, b). Poaceae Erianthus gigantens (Walt.) Muhl. Georgia. Charlton Co.: 1.6 km N of Dinner Pond, 500 mi E of Sapling Prairie boat trail, 15 Dec 1994, Loftin & O'Neill 138 (GA); Floyd's Prairie, Okefenokee NWR, 28 Oct 1992, Wé//ives & Loftin 20a (FLAS); Williges & Loftin 20b (GA). Erianthus giganteus is a cool season grass flowering September—October (Radford et al. 1968). It is fairly common and found growing throughout most prairies of the swamp. Although Loew and Jones (1984) reported collecting E. brevibarbus Michx. during their surveys, we have yet to observe this species in the field. Be- cause Loew and Jones (1984) collected during the growing season, they may have overlooked E. gigantes since it is most conspicuous in the fall. It is questionable whether E. brevibarbus actually occurs in the swamp. E. gigantens has been previously reported from Charlton County jones & Coile 1988) te Cyperaceae Carex verrucosa Muhl. Georgia. Ware Co.: 650 m E of Craven’s Ham- mock Island, 300 rn S of Craven’s Lake, 14 Jan 1993, Wéilliges G Loftin 46 (FLAS); Suwannee Creek Bridge on Swamp Perimeter Road, ca 200 m S of Piney Woods Lake, 200 m N of Suwannee Lake, 31 Aug 1993, W7//iges G O'Neil 2.09 (GA). Because this species preters sandy sites in the refuge, is not found in the “swamp proper,” i.e. where peat soils predominate. For this reason, it has been excluded from previous swamp studies. Since this Stipa 16(4): 776. 1995 Nores 577 species is common in relatively shallow water east of the Suwannee River Sill, it should be included in the flora of Okefenokee. These specimens represent new records for Ware County (Jones & Coile 1988). Cyperus erythrorhizos Muhl. Georgia. Charlton Co.: ca 900 m E of the Suwannee River Sill, 900 m S of Mack’s Island, 16 Sep 1993, W2//iges, Loftin & O'Neill 111 (GA), Williges, Loftin & O'Neill 1,12 (FLAS). Although these voucher specimens were grown from soil cores, this species has been observed growing in disturbed swamp sites, such as floating peat recently dredged from the canoe trails of Chesser Prairie. These specimens are new records for Charlton County Jones & Coile 1988). Eleocharis vivipara Link. Georgia. Charlton Co.: ca 20 m E of the Sapling Prairie Boat Trail, 2.4 km N of Dinner Pond, 25 May 1993, Wél/liges, Loftin, & O'Neill 68 (FLAS), Georgia. Ware Co.: ca 3.5 km NE of Suwannee River Sill, on Craven’s Hammock Canoe Trail, 19 Apr 1994, Williges & O'Neill 127 (GA). These specimens represent ae few E. vivipara we observed with mature spikelets. Eleocharis vivipara is one of four spikerush species that reproduce vegetatively by shoots produced from sterile spikelets at the culm apex (Ward & Leigh 1975). Submersed mats of vegetatively reproducing Eleocharis are common directly east of the Suwannee River Sill. Most of our collections in this area are either E. viipara or undetermined species. Loew and Jones (1984) reported collecting E. baldwinii (Torr.) Chapm., also a submersed, vegetatively reproducing species. We suspect that both E vivipara and E. baldwinii are found in the swamp, and possibly a third spe- cies, E. microcarpa Torr., with species dominance determined by hydrologic conditions. Rhynchospora alba (L.) Vahl. Georgia. Charlton Co.: Durdin Prairie ca 1.5 m S of Flag Lake, 100 rn W of boat trail, 16 Sep 1993, Wéilliges, Loftin & O'Neill 110 (FLAS); Durdin Prairie ca 300 rn SE of Durdin Lake, 100 m NE of boat trail, 24 Sep 1993, Wélliges, Loftin & O'Neill 114 (GA). These specimens represent the first records of this species from Southeast Georgia and are the southernmost records in the U.S. Prior records of R. alba exist for Rabun County in the Blue Ridge province of northeast Georgia (Duncan & Kartesz 1981). Of interest is that the habitat of R. a/ba in Okefenokee is not what is commonly associated with this species in other regions. Ac- cording to Gale (1944), R. alba is considered rare in Virginia, and found southward only in scattered mountain bogs of West Virginia, North Caro- lina, and Puerto Rico. Godfrey and Wooten (1979a) state this species 1s found in open, sphagnous bogs from Newfoundland to Maryland, and south in the mountains to North Carolina. Radford et al. (1968) report this spe- cies is very rare in North Carolina. Our voucher specimens were germ1- — Sipa 16(4): 777. 1995 778 Sipa_ 16(4) nated from soil cores collected from Durdin Prairie. Although it is not common, we have observed R. alba growing in the southern edge of Durdin Prairie, where the white scales of this species make it fairly conspicuous. Durdin Prairie is one of the least studied regions of the swamp. The un- stable floating vegetation mats characteristic of the prairie make travel in this area extremely difficult without the aid of air boats. For this reason few plant investigations have been done in this region, which may explain why this species has been overlooked. We have also found this species reproduc- ing in Southeast Chesser Prairie. Although this prairie in general has a more solid, submersed peat substrate, the prairie fringe consists of floating peat mats similar to those found throughout Durdin Prairie. The Chesser Prairie samples were found in this floating peat mat fringe. Rhynchospora cephalantha Gray. Georgia. Charlton Co.: Floyd's Island Prai- rie, 700 m E of Suwannee River, 100 m N of Floyd's Prairie Canoe Trail, 20 Jul 1993, Wélliges & Loftin 88 (FLAS), Williges G Loftin 89 (GA). Loew and Jones (1984) reported collecting four species of Rhynchospora: R. fascicularis (Michx.) Vahl, R. inundata (Oakes) Fern., R. microcephala Britt. x Small, and R. wrightiana Boeck. Rhynchospora cephalantha and R. microcephala are similar in general features and are not always distinguish- able (Godfrey & Wooten 1979a). Records of R. phalantha have been re- ported from Charlton and Echols counties (Jones & Coile 1988). Although we have not observed R. microcephala in the swamp, it is possible both spe- cies are present. Therefore, it is likely that six species of Rhynchospora occur in the Okefenokee Swamp. Scleria reticularis Michx. Georgia. Charlton Co.: Floyd’s Island Prairie, ca 700 m E of Suwannee River, 100 rn N of Floyd’s Prairie Canoe Trail, 18 Aug 1993, Williges, Loftin & O'Neill 102 (FLAS). Ware Co.: Chesser Prai- rie, ca 700 m W of Seagrove Lake, 700 rn E of boat trail, 4 Aug 1993, Williges, Loftin & O'Neill 100 (GA). These specimens were germinated from soil cores and represent new records for Ware and Charlton counties (Jones & Coile 1988). Scleria reticularis has also been found on exposed, floating peat mats in south Chesser Prairie. Its occurrence in the field corresponds to periods of summer drawdown, when the peat surface becomes exposed. — Juncaceae Juncus repens Michx. Georgia. Charlton Co.: Suwannee River Sill area, ca 300 m E of Middle Island, 24 May 1993, Wélliges, Loftin & O'Neill 62 (FLAS); Suwannee River Sill area, ca 1100 m SE of Pine Island, 24 May 1993, Willies, Loftin & O'Neill 63 (GA). These voucher specimens were ger- minated from soil cores in which they produced a diagnostic inflorescence. Sipa 16(4): 778. 1995 NOTES 779 This species is commonly found in relatively shallow water and sandy soils near the Suwannee River Sill and the Craven’s Hammock canoe trail. We have observed only the submersed form in the field, spreading by rooting at the nodes, without flowering stems. It usually does not flower unless exposed during a drawdown (Godfrey & Wooten 1979a). This species is not found in the swamp proper, but should be included in the Okefenokee flora due to its commonness within the refuge. Jwnucs repens has been previ- ously collected from Charlton and Ware counties ( Jones & Coile 1988). Onagraceae Ludwigia alata Ell. Georgia. Charlton Co.: Suwannee River Sill area, ca 300 m E of Middle Island, 19 Jul 1993, W7//iges, Loftin G O'Neill 83 (FLAS); Williges, Loftin & O'Neill 84 (GA). This species is frequent in sandy soils near the Suwannee River Sill. Lidwieia alata is usually found along the edge of canoe trails in relatively shallow water, frequently established on floating logs, and produces numerous, submersed, basal stolons. Previous field collections consisted of the stoloniferous form only, and it was not until our voucher specimens germinated from soil cores that we obtained specimens with mature flowers. It is a frequently occurring plant in the western portions of the refuge on sandy sites. These specimens are new records for Charlton County Jones & Coile 1988). — ACKNOWLEDGMENTS The authors wish to thank D. O'Neill for his assistance with sample collection, D. Hall for specimen identification verification, E. Bridges for verification of R. alba samples, K. Perkins for assistance with archival sample preparation, and the Okefenokee National Wildlife Refuge for access to field sites and permission to collect samples. This research was funded by the U.S. Fish and Wildlife Service and the National Biological Service un- der Cooperative Agreement #14-16- — 1544, Research Work Order #87. —Kent A. Williges and Cynthia S. Loftin', National Biological Service, Coop- erative Fish & Wildlife Research Unit, P.O. Box 11 0450, University of Florida, Gainesville, FL 32611-0450, U.S.A. REFERENCES ConEN, A.D. 1973. Petrology of some ea peat sediments from the oes Swamp- marsh sere of southern Georgia. Geol. Soc. Am. Bull. 84:3867—387 Cypert, E. 1961. The effects of fire in the aoe Swamp in 1954 a 1955. Amer. Midl. Moca 66:485—503 ‘Author to whom correspondence should be sent. Stipa 16(4): 779. 1995 780 Stipa 16(4) 1972. The origin of houses in the Okefenokee prairies. Amer. Midl. Natu- ralist 87: sae oan 973, Planet succession on burned areas in eee Swamp following fires of 1954 and 1955. Tall Timbers Fire Ecology Conf. 12:199-217. Duncan, W.H. and J.T. Karresz. 1981. Vascular fora of sag an ainataiel checklist. University a ‘Gecmia Press, Athens. Gate, S. 1944. Rhynchospora, section se le in Canada, the United States, and the West Indies. Rhodora 46:89—2 Goprrey, R.K. and J.W. a 1979a. Aquatic and wetland plants of southeastern United ae ie ui University of Georgia Press, Athens. 97 9b. Aquatic and wetland se of southeastern United States, Dicoty- ledons. en of Georgia Athe GREENING, H.S. and J. Gerritsen. 1987. Oe in macrophyte community structure - aes ae in the Okefenokee Swamp, Georgia, U.S.A. Aquatic Bor. 28:113 oe 5.B. Jr. and N.C. Cole. a The distribution of the vascular flora of Georgia. University of Georgia Press, Ath Loew, J.A. and S.B. Jones, Jr. 1984 "C hecklise of a vascular plants of the Okefenokee Swamp. I : A.D. Cohen, D.J. Casagrande, M.J. Andrejko, and G.R. Best, eds. The Okef fenokee eee its natural oy geology, oe recat Wetland Surveys, _os Alamos, New Mexico. =e 702-709 PARRISH, RK. and E.J. Rykie. 1979 ‘Glernde Swamp origin: review and reconsidera- tion. J.Elisha Mitchell Sct. _ 95:17-31 Raprorp, A.E., H.E. AHLrEs and CLR. Bat. 1968. Manual of the vascular plants of the Carolinas. University of North Carolina Press, Chapel Hill. SCHLESINGER, W.H. 1978. Community structure, dynamics anc Okefenokee cypress swamp-forest. Ecol. Monog. 48:4 Trowel, C.T. 1988. Exploring ie Okefenokee: Rol and 7 Harper in the Okefenokee Swamp. Research paper no. 2, South Georgia College, Douglas. Warp, D. B. and E.M. (Hopcson) LeiGu. 1975. Contributions to the flora of Florida --8, Eleocharis (C en Castanea 40:16— — nutrient cycling in the Wricut, A.H. and A.A. Wricnr. 1932. The | yabitats and composition of the vegetation of Okefenokee Swamp. Ecol. Monog. 11:1 10-232 Sipa 16(4): 780. 1995 781 DESMODIUM LINDHEIMERI (LEGUMINOSAE) IN MEXICO AND TEXAS—When first described, Desmodium lindheimeri Vail (1891) was known from only four collections: three from Mexico and one from Texas. Since then, many more collections have been made, making it possible to draw conclusions about the range and habitat of this species. Label data on the sheets in the University of Texas collection (LL, TEX), combined with the localities cited by Vail, indicate that the primary range of D. lindheimeri \ies in Mexico in the states of Coahuila, Nuevo Leon, San Luis Potosi, and Tamaulipas (Fig. 1). Desmodium lindbeimeri is usually found at an altitude of 1,000—2,000 m in Mexico. Of 27 Mexican collections in the University of Texas herbarium, only four were found below 1,000 meters. These were in Nuevo Leon at 985, 800, 600, and 450 meters. A majority of the collections were made in pine or oak forests and sometimes in a mixed pine/oak association. Plant associates include Quercus mublenbergi Engelm., Quercus gravesti Sudw., Quercus glaucoides Mart. & Gal., Colubrina greggi Wats., Ungnadia speciosa Endl., Acacia berlandieri Benth., Zanthoxylum fagara (L.) Sarg., Fraxinus cuspidata Vorr., Ostrya virginiana (Mill) K. Koch, Acer grandidentatum Nutt., Carya ovata (Mill) K. Koch, Juglans magor (Torr.) Heller, and Populus tremuloides Michx. The bloom period is apparently from early August through October, possibly into November. Flower color 1s usually whitish to pale pink, though occasionally rose-pink to purple. The stems are up to 2 m long. In contrast to the situation in Mexico, the passage of time has not pro- duced numerous collections of D. /indheimeri from a similarly broad area in Texas. Until recently, this taxon had not been collected at all in Texas since Lindheimer’s Comal County collection in November 1850. On March 12, 1992, the author found a population of D. /indbeimeri in Comal County, Texas, distributed along the rocky bed of a dry ravine, here designated Locality 1 (precise locations of all Texas populations withheld). On Octo- ber 15, 1995, populations were found at two other localities. Locality 1 lies on the midpoint of a line drawn between Locality 2 and Locality 3, which are 10 km apart. Distribution along this 10 km line, which crosses several separate ravine systems, is probably more or less continuous. The popula- tions are at an altitude of about 270 m. At Locality 1, the majority of plants are distributed along the rocky bed of a ravine with a few found among brush on the banks 0.3—1.3 m above the dry ravine bed. The author has walked a little over one kilometer of this ravine and found the plants scattered throughout, as well as along one- third kilometer of a side ravine. It is apparent that D. /indbeimeri is a favor- ite of the local deer population, with about 80% of the plants suffering Stipa 16(4): 781. 1995 79? Sipa 16(4) COAH. 100 KM. ———_—— Fic. 1. Documented distribution of Desmodium lindheimeri Vail. Sioa 16(4): 782. 1995 Notes 783 browsing severe enough to prevent successful flowering and fruiting. At Locality 2, the plants occur on the very steep, tree-covered, 6-12 m high bank of a ravine but not in the bed of the ravine. At this site, the steep bank and uncertain footing combine to protect the plants from deer. Here, for 0.4 kilometer, about 90-95% of the plants are flowering and fruiting un- molested. At Locality 3, the plants are distributed in three clusters along the roadway for 200 m. They grow on flat ground in dry caliche, protected from the afternoon sun by the shade of the treeline. This right of way was probably colonized by plants from adjacent ravines. Whether due to nearby homes or a high volume of road traffic, the plants at this locality are almost untouched by deer and are producing an abundance of fruit. At all three sites, the plants range from 0.5 to 1.3 m tall, often with partially decum- bent stems to 2 m long. There does not appear to be anything remarkable about the habitat in which these Texas plants are found. It is a live oak/juniper association found throughout much of the Edwards Plateau consisting of Quercus fusiformis Small and Juniperus ashei Bucch. with Quercus texana Buckl., Ungnadia speciosa Endl., Ulmus crassifolia Nutt., Croton fruticulosus Torr., Rhus toxicodendron L., Aloysia gratissima (Gill. & Hook.) Troncoso, Pavonia lasiopetala Scheele, Forestiera pubescens Nutt., and Bernardia myricifolia (Scheele) Wats. In Texas, D. lindhetmeri is abundant where found and seems to be a weedy, adaptable species which does well in a variety of habitats. It is surprising, then, to find it apparently restricted to a small area of Comal County. Members of this genus are known by the common name “tickseed” be- cause of the tenacity with which the hooked hairs of the loments attach them- selves to clothing and animal hair, and, are well-known for ensuring seed dis- persal through animal transport. Although it is possible that the Comal County material represents a disjunct relictual population (Nesom 1993), it is also possible that D. /indheimeri was one of the first plant invaders of Texas. North- eastern Mexico was the natural trading partner of the San Antonio region of Texas through the Eighteenth Century into the early Nineteenth Century. There was no shortage of candidates, animal or human, for seed transport. In 1689, Alonzo de Leon, the governor of Coahuila, reached the general area of present day San Antonio (Williams 1979). Only two years later, in 1691, Don Domingo Teran de Los Rios left the main body of his expedi- tion behind, and with four companions, visited the Comal Springs in what is present day Landa Park in New Braunfels. The Espinosa-Olivares-Aguirre Expedition, in 1709, discovered the San Pedro Springs and the San Anto- nio River and then passed through present day New Braunfels. The Domingo Sioa 16(4): 783. 1995 784 Stipa 16(4) Ramon Expedition, in May of 1716, camped at the site of present day New Braunfels. While on a trading trip to Saltillo, Coahuila, a resident of San Antonio de Bexar wrote in a letter dated January 6, 1827, “To me there is nothing more agreeable than to wake in the morning, when among the mountains, and listen to the bleating of the calves and lambs, the lowing of the cows, the braying of the mules and donkies, and to behold the bakharas herding the cattle, the horses and mules, while on every cliff and rock, sheep and goats may be seen sporting from rock to rock, and leaping over every dan- gerous cliff. .... The exports of this country are wool, cochineal, and fruits; besides this a vast number of mules are driven into the States for sale.” (Dewees 1852). Certainly, there were opportunities for seed dispersal from Mexico to Texas and colonization by D. lindheimeri prior to Lindheimer’s collection of 1850. The University of Texas collection of this taxon was greatly augmented several years ago when Dr. Guy Nesom reviewed the Mexican Desmodinm and found numerous misidentified sheets of D. /indheimeri. These collec- tions, combined with the recent Texas collections, make it possible to up- date and modify the description of the species. Desmodium lindheimeri Vail, Bull. Torrey Bot. Club 18: 120. 1891. Merbomta lindbeimert (Vail) Vail, Bull. Torrey Bot. Club 19:111. 1892. Erect branching perennial herb 4-15(-18) dm tall; stem angulate, grooved, uncinulate-puberulent and -pubescent and sparsely scattered-pi- lose with slender white trichomes; stipules ovate, long-attenuate, densely pilose on the outer surface with long white trichomes, reflexed at maturity, not long persistent, 6.5-8 mm long, 1.5—2 mm wide; stipels slenderly lance-attenuate, 1.5—2 mm long; petioles densely uncinulate-puberulent and -pubescent and somewhat long spreading-pilose, 14—35(-45) mm long; leaf rachis similar, 6O-15(-19) mm long; leaflets acute at apex, cuneate to obtuse at base, uncinulate-puberulent and more or less soft white-pilose above, densely long and soft white-pilose or tomentose below with promi- nent venation; terminal leaflet ovate to mostly rhombic in outline, 5—9(— 10.5) cm long, 3—5.8(—7) cm wide; lateral leaflets more nearly ovate or elliptic, somewhat asymmetrical, 4.3—-6(—7) cm long, 2-4 cm wide; inflo- — rescence paniculate, the rachis ridged and grooved, uncinulate-puberulent and -pubescent; primary bracts ovate-attenuate, striate, long appressed pi- lose on outer surface, ciliate, glabrous within, noc long persistent, 4.5—9 mm. long, 2-3 mm wide; secondary bracts essentially glabrous but ciliate, 0.8—3 mm long, 0.8—1 mm wide; pedicels rather finely pilose with multi- Stipa 16(4): 784. 1995 Notes 785 cellular trichomes which are glandular at base; flowers pinkish-white to pale pink, occasionally rose-pink to purple; calyx finely puberulent, some- what ciliate, the long white trichomes along central tooth of lower lobe reaching 3 mm in length; corolla to 7 mm long; loment stipitate, to 7- articulate; stipe 3-8 mm long; articles subrhombic to semiovate in out- line, the isthmi slightly excentric, the articles appearing somewhat con- torted because of the infolding of their margins, surfaces glabrous, reticulate at maturity, the suture densely uncinulate-puberulent, 7—11(-13) mm long, 5-8 mm wide; seeds 4—5(—6) mm long. The relationship of this species is with D. canescens and its relatives, espe- cially D. ochroleucum M.A. Curtis of southeastern United States, which it resembles particularly in the characters of the loments (Description adapted from Correll & Johnston 1970). cimens examined: MEXICO. Coahuila: Musquiz swamp, 15 Sep 1936, E. G. Marsh, - ne (TEX); Musquiz Palm Canyon, 19 Sep 1936, E.G. Marsh, Jr. 987 (TEX), Sierra la Gloria, Canon El Cono, a side canyon of C. Chilipitin draining in from N n Chilipitin, at lowest pouroff in steep-walled limestone canyon, 6 Sep 1976, T. Wendt ; - Riskind 1610 & 1618 (TEX); Mpio. de Musquiz, In sheltered moist drainage of next spring WNW of “slump” spring, 24 Aug 1975, T. Wendt, E. Lott, & D. Riskind 1313 (TEX). Nuevo Leon: Ejido Santa Rosa, Mpio. Iturbide, 29 Aug 1989, A. Eduardo Estrada C. 1648 (TEX); Mountains near Monterrey, JUL 1933, C.H. & M.T. Mueller 491 (TEX), Rio Ramos en el municipio de Allende, 17 Sep 1983, A. Rodriguez, . Carranza, & Grupo ICCAC 958 (TEX); Mpio. Villaldama, Sierra Gomas, in Canyon El] Alamo on limestone talus, 15 Aug 1988, TE Patterson 6757 (TEX); Galeana, Haciendo eae 26 Aug 1936, M. Taylor 243 & 219 (TEX); Areas cercanas a Cola de Caballo, cercana a corrientes de agua, no date, J.A. Villarreal, M.A. Carranza & M. Vasquez R. 2861 (TEX), Mpio. Linares, near Ejido Los Alamos, 7.2 mi S of Mex. 60, 28 Oct 1982, J. Grimes with K. Nixon, L. Dorr, S. Sundberg 2370 (TEX); Mpio. Iturbide, in a gully NW of ge es Rosa, 4.1 mi S of rbide, Loma la Banderra, 25 Oct 1982, J. Grimes with K. Ni ae & S. Sundberg 2334 (TEX); Iturbide to Camarones, 6 Sep 1991, Hinton et ys oe 2 (TEX); Mpio. de Montemorelos, La Trinidad, 19 Aug 1939, C.H. Maller 2833 (TEX); ae to Camarones, 17 Sep 1991, Hinton et al, 21545 (TEX), N of Aramberri, 9 Sep 1990, ie et al. 20589 & 20556 (TEX); North of Mpio. Villa de Santiago, Canon la Boca, camino a Cola de Caballo-Laguna de Sanchez, 10 Sep 1983, J.A. Villareal, M.A. Carranza, G M. Moreno 2360 (TEX); Mpio. Montemorelos, 5 km SE of La Trinidad, on eastern side of Sierra ane near Ojo de Agua, 7 Aug 1988, T:F. Patterson 6278 (TEX); Monterrey, in canyon abov Diento, Oct 1961, R. FE Smith M581 (TEX). Tamaulipas: Mpio. Hidalgo, road from = Engracia toward Dulces Nombres, N.L., 0.3 a mi W of Paraje de Los Caballos, 11.4 road mi E of Dulces Nombres, 4.6 road mi from ey of deep canyon of Arroyo Ramirez Luna, N-facing slope, 21 Sep 1994, G. Nesom with M. Mayfield & J. Hinton 7462 (TEX); Hidalgo, Puerto Purificacion, 23 Sep 1994, Hinton et al. 24837 (TEX); Hidalgo, Los Caballos, 21 Sep 1994, Hinton et al. 24812 (TEX); 11 mi ou road W of Victoria toward a =A Sep. 1950) ALG, ou & J. Graham se ne X). S.A. TEXAS: Comal Co.: New Braunfels, NOV 1850, FE Lindheimer 7654 (= L. 499) a ae 1, 12 Mar 1992, - oe 2089 ae - Mar 1992, M. Enquist 2116 Sipa 16(4): 785. 1995 786 Sipa 16(4) (ANSM, BRIT, GH, MEXU, MO, SRSC, TAES, TEX, UVST), 28 Oct 1994, M. Enquist 2593 (ANSM, BRIT, GH, MEXU, MO, SRSC, TAES, TEX, UVST), cultivated in Texas Tech Greenhouse, material from ne 1 in Comal County, 31 Oct 1994, M. Enquist 2596 ean GH, MEXU, MO, TAES, TEX); ie 2, 15 Oct 1995, M. Enquist (ANSM, BRIT, GH, MEXU, MO, SRSC, TAES, eas i duea Locality 3, 15 Oct M. Eas 2892 (BRIT, GH, MEXU, MO, TAES, TEX) ACKNOWLEDGMENTS I would like to thank Janet Atyia, Cynthia McKenney, and Kevin Mitchell of the Plant & Soil Science Department of Texas Tech for their expert cul- tivation of D. lindheimeri from Comal County. The Comal County material cultivated in the Texas Tech greenhouse was, in December 1994, placed in the care of Paul Cox of the San Antonio Botanical Gardens. My thanks to Paul and the staff of the San Antonio Botanical Gardens for accepting these plants and continuing their cultivation. —Marshall Enquist, 11511 Metric Blvd. #1033, Austin, TX 78758, U.S.A. REFERENCES Correll, D.S. and M.C. JoHNston. 1970. Manual of the vascular plants of Texas. Texas Research Foundation, Renner Dewees, W.B. 1852. Letters fom an early settler of Texas. Louisville, Kentucky: Morton & Griswold. Fac. ed. 1968, Waco, Texas: Texian Press. Nesom, G.L. 1993. Three species of Aster (Asteraceae: Astereae) disjunct in northern Coahuila, ae Phytologia 74:296-3 VatL, A.M.. 1891. An undescribed tie from Texas and Mexico. Bull. Torrey Bot. Club 18:120. —______.. 1892. A preliminary list of the species of the genus Methomia Heist., occur- ring in the United States and British America. Bull. Torrey Bot. Club 19:107—118. WituiaMs, J.W. 1979. Old Texas trails. Burnet, Texas. Eakin Press. Spa 16(4): 786. 1995 NOTES 787 VALIDATION OF TACHIGALI MICROPETALA (DUCKE) ZARUCCHI & PIPOLY (FABACEAE: CAESALPINIOIDEAE)—An in- advertent error caused the recently published new combination of Tachigali micropetala (Ducke) Zarucchi & Pipoly to be invalid (Pipoly 1995). The International Code of Botanical Nomenclature (Greuter et al. 1994) Ar- ticle 33.2 clearly states, “A new combination or avowed substitute (nomen novum), published on or after 1 January 1953, for a previously and validly published name is not validly published unless its basionym or the re- placed synonym is clearly indicated and a full and direct reference given to its author and place of valid publication with page or plate reference and date.” The article mistakenly left off the volume, page number and date of the article in which the basionym was published. The correct combination is validated below: Tachigali micropetala (Ducke) Zarucchi & Pipoly, comb. nov. Basiony: Sclerolobium micropetalum Ducke, Bol. Tech. Inst. Agron. Norte Belém 2:20. 1944. Type. BRAZIL. AMAzonas: Manaus, without date, A. Ducke 1219 (SYNTYPES: K, MO, NY). I thank Dr. K. N. Gandhi of the Gray Herbarium of Harvard University for kindly pointing out this unfortunate error. —John J. Pipoly I, Botanical Research Institute of Texas, 509 Pecan Street, Fort Worth, Texas 76102-4060 U.S.A. REFERENCES Greuter, W., ER. Barriz, H.M. Burner, W.G. CHALONER, V. DEMOULIN, D.L. HAWKsworTH, P.M. JORGENSEN, D.H. NicHoison, P.C. Sitva, P. TREHANE, and J. McNem (eds.). 1994. International code of botanical nomenclature (Tokyo Code). Regnum Veg. 131:1—389. Piroty, J.J. 1995. A new Tachigali (Fabaceae: Caesalpinioideae) from western Amazonia. Sida 16:407-411. Sma 16(4): 787. 1995 Sipa_ 16(4) ANNOUNCEMENT THE BARBARA J. HARVILL BOTANICAL RESEARCH FUND FOR FLORISTIC RESEARCH IN VIRGINIA Small research grants for floristic field work in Virginia and/or travel to herbaria are available to botanists without an institutional base of support for such work. This fund was endowed by friends and family of the late Barbara J. Harvill to encourage floristic and revisional work in Virginia. Most awards requested to date have been for mileage costs, but other ex- penses, such as lodging and certain kinds of field equipment (plant presses, for example) can also be covered. Please send your letter of application for 1996 awards by May 15 to Donna M.E. Ware, Sec., Virginia Botanical Associates, Department of Bi- ology, College of William and Mary, Williamsburg, VA 23187. Awards will be made by June 15, 1996. Siva 16(4): 788. 1995 BOOK NOTICES Catalogue of the Library of the Massachusetts Horticultural Society. 1994? Facsimile of the 1918-1920 first editon. Maurizio Martino Publisher, 746 Mansfield City Road, Mansfield, CT 06268. $150.00. 587 pp, hbk. The catalogue is divided into two parts. Part I consists of the Preface, Explanation of Abbreviations and Signs, Author Catalogue, Additions, and Further Additions. Part II is Table of Subjects, Subject Catalogue, Corrections, and Index. Included in the Table of Subjects are the following subjects: 1) Works of Reference, 2) Horticulture, 3) Trees and Shrubs, 4) Agriculeure, 5) Economic Plants, 6) Soils, 7) Entomology, 8) Plant Pathology, 9) Botany, 10) Natural History, 11) Rural Life and Plant and Garden Lore, 12) Voyages and Travels, 13) Evolution, 14) Addresses, 15) Color and Design, and 16) Miscellaneous Publications. Garrett, J. Howarp. 1995. The Dirt Doctor’s Guide to Organic Gardening: Essays on the Natural Way. (ISBN 0-292-72780-1, hbk; 0-292-72781-x, pbk). University of Texas Press, P.O. Box 7819, Austin, TX 78713-7819. $14.95 (pbk), $35.00 (hbk). 232 pp, Sth" x B82", From the Introduction: “A truly organic program, on the other hand, is built around a ‘healthy’ seco and has the primary goal of establishing a natural balance of soil, water, air and biology. Looked at in the simplest terms, an organic program increases the air and the organic matter in the soil, uses nz aturally balanced fertilizers, keeps all bare soil mulched, and increases soil life and insect life. If you want to go natural with your gardening, then this is the book for you. “Plant growers of all ee are turning to organic products and procedures for vigorous, healthy plants. Since 1988, J. Howard Garrett has been in the forefront of the organic gardening movement, dispensing well-tested advice through ‘The Natural Way,’ his radio program and column in the Dallas cae News. In this expert guide, he offers a complete program for organic gardening. e book is a compliation of revised versions of some of the author's past seniere stories, oe speeches, and thoughts on what he calls “The Natural Way.’ However, th book reflects new concepts, new science, and new techniques. These ‘new essays’ have ie rewritten to be clearer, more accurate, and more up-to-dat Here is what you will find in those 232 pages of essays: 1) Basics of Organics, 2) Soil eee 3) Fertilization, 4) Pest Control, 5) Organic Landscaping, 6) Food Crops, 7) Wildlife, and an Index. Stipa 16(4): 789. 1995 790 Sipa 16(4) BOOK REVIEWS BENJAMIN, Denis R. 1995. Mushrooms: Poisons and Panaceas. (ISBN 0-7167-2600-9, pbk). W. H. Freeman and Company Publishers, 41 Madison Avenue, New York, NY 10010. $34.95. 422 pp, 32 color photographs, 27 b/w illustrations, 6" x 9", This is a must buy book for libraries, poison control centers, mushroom clubs, mycolo- gists, toxicologists, hospital emergency room physicians and staff, and amateur and pro- semen naturalists. The book is divided into three major parts: 1) mushrooms and health; 2) mushroom poisoning, and 3) mushroom poisoning syndromes. The reader will also find four Paes 1) a discussion of the chemistry of mushroom toxins and methods of their analysis; 2) a listing of national and regional field guides for mushroom identification; 3) a geographical cre of mycological associations and consultants, and 4) a reference list of mushroom cookbooks. e author covers the history of mushroom consumption and differing cultural attitudes mee mushrooms in a delightful, engaging, and humorous way. He has not only selected eeroae anecdotal examples of cultural attitudes but is a good storyteller. He then goes on o discuss the signs, symptoms, and treatment for mushroom poisoning, and suggests the oe questions to ask in order to arrive at a proper — The book contains color ee of high quality to aid in picture keying toxic mushroom species, as well as clear tables an schematic diagrams to aid in understanding the on sms of intoxication. However, this not a field guide with keys and the p rare are limited to deadly poisonous and toxic fangi. The author's style and the layout of the book makes it easy and enjoyable to read and et it is most educational at the same time. The work is well referenced after each chapter and it is clearly evident on the author has carried out an extensive search of the current international literature, in order to include the most up-to-date knowledge on the subject of cetism Some of the chapters deserve special mention since the information is up-to-date and especially seve ealth Benefits and Medicinal Properties of Edible Mushrooms; ushroom Poisoning Not Caused ee Mushroom Toxins; Spectrum, Incidence, Sociology, Diagnosis an . Managen ment of Mushroom Poisoning; Amatoxin Syndrom e; Gryomitrin Poisoning; ‘Antabuse’ Syndrome; Hallucino ogenic oe ndrome; Muscarine Poison ning; Gas- trointestinal Syndrome; Miscellany of Toxins. These are just examples but there is far more in this book that will interest the ae eG ae as the chapter on the Introduction to the mre of Mushrooms and the valuable resource information j in the species If Ic afford to buy only one book about mushroom poisoning, ne would be When one measures the amount of im ortant information availa lable in thi ae ok against price, one can only conclude: What a beetle W. Keller, Re: eee BRIT. Qu Stipa 16(4): 790. 1995 Book REVIEWS 791 BEssETTE, ALAN E., ORSON K. MILLER, JR., ARLEEN R. Bessette, and Hope H. Mier. 1995. Mushrooms of North America in Color, A Field Guide Companion to Seldom-Illustrated Fungi. (ISBN 0-8156- 0323-1, pbk). Syracuse University Press, 1600 Jamesville Avenue, Syracuse, NY. 13244-5160. $17.95. 188 pp, 74 color photographs, oy ue eeu A book like this one is long overdue. Congratulations go to Syracuse University Press for publishing a book that fills a mycological and scientific need by adding to the iconog- raphy of fungi but avoids duplicating the common edible species that adds pagination and higher costs. The authors estimate that of the more than 5,000 species of mushrooms that occur in North America less than half have been illustrated with color photographs. Most monographic publications use either line drawings or black and white photographs or rarely color paintings or photographs. In any case the species descriptions are highly tech- nical and have limited value for a more general readership. The objective of this book is to provide an accurate but nontechnical, species description, a color illustration of high qual- nd a discussion of its distinguishing characteristics. Each species is given a morpho- logical description that highlights the fruiting body (cap, gills, stalk, flesh, technical fea- tures and spore print color). Under the topical heading fruiting is noted the habit, habitat and seasonality of occurrence. And lastly, the edibility, observations, and notes give a bet- ter understanding of the edibility, distinguishing characters and geographic distribution of each species. Each color photograph is about 2'/2" x 37/4" and occupies about one-third of a page. The photographs are of high quality and show the necessary gross morphological characters for identification There is a glossary, works cited and index that concludes the ook. There are no keys. The author's state that “The species selected are those which are uncommon, rare or not well illustrated in the current literature.” This book is intended to a companion to other popular field guides. I highly recommend this book for any seri- ous student of the fungi— Harold W. Keller, Research Associate, BRIT. Jones, Micuart. 1994. Flowering Plants of the Gambia. (ISBN 90 5410 197 0, pbk). A.A. Balkema Publishers, Old Post Road, Brookfield, VT 05036 (fax 802-276-3837). $40.00. 132 pp, 160 color photo- graphs, 61/2" x 91/2". Gambia is located in West-Central Africa and is one of the most densely populated countries of Africa. Much of the true forest has been destroyed by human encroachment and deforestation. This book will help document the need for preservation and greater appreciation for what is left of the original flora. There are introductory topical sections that briefly describe the geography, climate, ely and topography, soils, vegetation and phytogeography, botanical exploration, exsiccata, and botanical literature. These sections give a good general survey of the flora of The Gambia. ecies descriptions, mostly four to a page, are nontechnical and understandable by the general public. Flowering dates, habitats and local names and uses are given. The taxa are arranged into two classes: the Dicotyledoneae and Monocotyledoneae, and then into families, followed by introduced and well-known taxa. The color ae are arranged so that five photographs of varying size fill most of the page, with the scientific binomial of the plant and cross referenced page numbers where the species rere can be found, Stipa 16(4): 791. 1995 792 Sipa 16(4) as a caption at the bottom. The photographs are of good quality and show the general habit of the plane with flowers. The flowers are not shown in enough detail to discern the floral parts and familial nares The book is concluded by a short glossary, a list of about 35 references of mostly Gambian origin, and an index. It should appeal especially to foreign tourists who visit Gambia in ever increasing numbers.—Harold W. Keller, Research Associate, BRIT GUARINO, LuiGi, V. RAMANATHA RAO and Rosert Reip (Eds.). 1995. Collecting Plant Genetic Diversity. (SBN 0-85198-964-0, hbk). The University of Arizona Press, 1230 N Park Avenue, Suite 102, Tucson, AZ 85719. $120.00. 748 pp, 6'/2" x 9'/2", This is a compendium of generic, i ae theoretical and practical information that is aimed at both new and experienced collectors of plant germplasm. It is a comprehensive treatment published on behalf of the ow. al Plant Genetic Resources Institute in association with the Food and Agriculture Se onerane of the United Nations, The World ation Union and the United Nations Environment Prog There are 39 chapters, each prepared by different authors and each with a set of exten- sive references. The contents are organized around several themes: the introduction covers a brief os of plant germplasm collection, legal issues, and methods and procedures of collectin Before eu out includes: assessing the threat of genetic erosion, sampling strategy theory and practice, collecting wild species, classification of intraspecific variation in cr plants, published sources of information (on existing germplasm collections, on natural and human environments, on wild plant species), aids to taxonomic identification, biblio- graphic databases, Sa see surveys, mapping distribution of biodiversity, geographic information systems and remote sensing for plant germplasm. There is a wealth of infor- mation in this section that ‘would be useful as a reference source for anyone working on and a — U interested in the preservation of plant germplasn In the field includes: collecting plant genetic resources; gathering and recording data in the field; collecting seeds; collecting vegetativel pai ae crops, collecting grasses, legumes, woody perennials, pollen, RAizobiam, Frankia , mycorrhizal fungi, and herbarium vouchers. These sections contain detailed information on ee to collect various plant sources of germplasm but also provide more general information on how to handle plant struc- tural parts. Plant co ee interested in field procedures and collections in a broader ap- plication than just germplasm would do well to read this section k at base includes: processing and reporting germplasm data and collecting mis- sions. A series of case studies, for example, collecting the rice gene pool, collecting rare species in Florida, and collecting Andean root and tuber crops, among others. There is far more information than can be covered briefly in a single review but anyone seriously inter- ested in plant field collecting will benefit from reading this book.—Harold W. Keller, Research Associate, BRIT. Sipa 16(4): 792. 1995 Book REVIEWS 793 McDape, Lucinpa A. Kamayir S. Bawa, HENRY A. HESPENHEIDE, AND GARY S. HarrsHorN (Eds.). 1994. ISBN 0-226-03952-8, pbk). LA SELVA, Ecology and Natural History of a Neotropical Rain Forest. The University of Chicago Press, 5801 S Ellis Avenue, Chicago, IL 60637. $28.95. 486 pp, 8'/2" x 11". La Selva is a nature reserve and field station in Costa Rica that represents one of the most intensively studied tropical field sites in the world. This area has been a major focus of research on rain forest ecology, flora, and fauna for the last 30 years. This book covers La Selva's climate, soils, and physical oe its plane and animal life, and agricultural de- velopment and land use in nearby areas Part 1 summarizes research on ~ physical setting and environment of the rain forest, as well as the history of the research station. Chapters in this part focus on climate, geo- rphology, aquatic systems, soils, nutrient ae and ues of energy. ee 2 ne asizes the plant community with sections vegetat types, plane diversity aa demography, spatial patterns of trees, the ae of wee gaps on forest structure and dynamics, physiological ecology and plant reproductive systems. Part 3 covers the animal community suminaiing information on butterflies, fishes, amphibians, reptiles, birds, and mammals. Part 4 addresses interactions between plants and animals, focusing on her- bivory and frugivory. Part 5 considers the impact of land use and agricultural development 1 Selva and other areas of Costa Rica. One chapter examines land co ee and conservation in Sarapiqui, the county in which La Selva is located; another covers subsis- e and commercial agricultural development in Atlantic lowland regions; and a third treats 2 research ie in Costa Rica. A final chapter highlights the prospects for a comparative tropical ecology. The appendixes address research productivity at La Selva, the ah and governance of the field station and checklists of vascular plants and vertebrate pak The bibliography is extensive covering 57 pages in length and located at the end of the book instead of following each chapter. In the hands of conservation- minded se an book will be a valuable resource of information for anyone interested in tropical biology— Harold W. Keller, Research Associate, BRIT. LINDEN, Perer J. VAN DER and Donatp R. Farrar. 1993. Forest and Shade Trees of Iowa. Second Ed. UGSBN 0-8138-0734-4, hbk). Iowa State University Press, 2121 South State Avenue, Ames, IA 50014-8300. $22.95. 150 pp, 325 plus b/w photographs, line drawing illustra- tions, 8'/2" x 11". This second edition contains a information on the characteristics, origins, loca- tion, and uses of trees found in Iowa. It is a guide to the identification, distribution, cl history, and uses of Iowa trees. It is pro ofusely illustrated with aerial black and white photographs of landscapes, groundshots of habitats, habits of individual tree species and structural parts used in identification: leaves, twigs, bark patterns, and fruits. The photography is Aen ais and illustrates the essential key characters used in the identi- fication of spe he > lar arge-page format includes the photographs, which occupy about half the page, my species descri an on the same page. Keys to trees in both summer and winter conditions are ain me of the chapter headings include: Iowa's forests; se- lecting and transplanting trees; Se trees (an explanation of key characters used to Stipa 16(4): 793. 1995 794 Sipa 16(4) identify trees). The ae chapter is Trees of Iowa that covers 95 pages and over 140 species of mostly trees and a few shrubs. For each species there is a nontechnical descrip- tion of tsingusbing characteristics, a list of similar trees, "distri bution, and several para- graphs that give interesting information about the tree, for example, their use in landscap- ing and voi advantages and disadvantages of planting a particular tree species, growth patterns, fall coloration, and problems with diseases. Chapter 5, Key to Trees of Towa, has a general, dichotomous, one-way key to selected species of trees with key charac- ters illustrated with line drawing illustrations. The key in most cases is simple to use 2 user friendly. The general key is followed by keys to pines, spruces, willows, poplars, hicko- ries, oaks (the greatest number of species at 14), elms, les and dogwoods as examples. ere 1s s separate winter key to common deciduous trees. This section is followed by a bibliography of about 100 a. There is an index organized alphabetically by ge- neric and common names. This book should appeal to the lay public who venture into our midwestern forests to enjoy springtime walks or marvel at the brilliant display of fall colors, who may have use of forest trees in landscape design, and to answer the question, What tree is that?, in the course of their daily lives—Harold W. Keller, Research Associate, BRIT. Ortiz DE MONTELLANO, BERNARD R. 1990. Aztec Medicine, Health, and Nutrition. (ISBN 0-8135-1563-7, pbk). Rutgers University Press, 109 Church Street, New Brunswick, NJ 08901. $15.00. 308 pp, 40 b/w illustrations, 6" x 9", The first chapter briefly summarizes pre and post-Columbian Aztec history, culture, and religion as well as the Spanish Conquest and colonial organization. Aztec medicine was holistic. Ilness and disease were seen as the interactions of the supernatural, magical, and natural causes. Evaluation of Aztec medicine demonstrates that the Aztecs correctly observed the physiological effects of plants . The Aztecs were clearly very accurate observ- ers of nature and used empirically derived remedies. The chapter on Aztec religion, world view, and medicine discusses Aztec religious beliefs, concepts, practices, and structure wich oe placed on human physiology, health, and medicine; the ae of the cos- mos, creation, destruction, and human sacrifice; man as microcosm; complexes of deitie the preservation of cosmic order as a justification for social controls; ae ae nee and religious stratification; and shamanism. Moderation in diet, exercise and behavior was an essential component of a balanced body. Good health was based on equilibrium, mod- eration, and performance of duct s a small force of Spaniaias led by Hernan Cortés able to defeat and overthrow the jee Empire i in such a short Le of time from 1519 to 1521? There are chapters to support the case that at the time of the conquest by Cortés the Aztecs were a thriving, well-nourished, and healthy oe They had a highly sophisticated and productive agri- cultural system, a coherent set of medical beliefs and effective health measures and no need to rely on cannibalism for protein. To support the author's ideas evidence is provided from anthropology, folklore, pharmacology, iene geography, demography, linguistics, history of medicine, relic ous studies and psychoneuroimmunology. Some of the chapters that support these ideas are: Population ae Carrying Capacity of the Basin of Mexico; The Aztec Diet: Food Sources and Their Nutritional Value; eevee Diagnosing and Explaining Illness; Curing Illness; Syncretism in Mexican Folk Medicin Sipa 16(4): 794, 1995 Book REVIEWS 795 The book concludes with two appendixes: Nutritional Values and Amino Acid Compo- sition of Aztec Foods; Empirical Evaluation of Aztec Medicinal Herbs and a section of notes and an extensive bibliography of over 450 references and an index. This book will be fascinating reading for anyone interested in the fall of the Aztec Empire and also the use of erbal medicines by peoples of Mesoamerica.—Harold W. Keller, Research Associate, BRIT. Isety, DuANg. 1994. One Hundred and One Botanists. (ISBN 0-8138- 2498-2, hbk). Iowa State University Press, 2121 S State Avenue, Ames, IA 50014. $32.95. xiv + 351 pp, 6" x 9", With a talent for edutainment, an academic vernacular speaking and writing style, and often poking into deep crevasses of ne lives, Isely has produced a readable book on a variety of botanists through the development of that science. Starting with Aristotle (384 B.C.—322 B.C.) and ending with Winona Hazel Welch (1896-1991), one would expect that the development of this science would be well — and reflected by the discussions on discoveries and developments made by each individual. In a large part, this is true, the development of taxonomy of vascular plants, mosses and oe i fairly well reflected; ecology, plant physiology and even horticulture also receiving coverage by choice of pertinent participants. However, as will always be true of such books where the selec- tion of characters discussed is subject to pa bias, there are a gaps in botanical science that are hardly mentioned or not covered at all. Algae, the basic plant ple and often the recipient of such neglect by large segments - the botanical community, are hard- ly mentioned in this book. The only person discussed who could be classified as a phycolo- gist is N. Pringsheim, whose contributions to this discipline are relatively insignificant. Having established the a a of Schleiden’s discoveries on plant cell structure, Isely missed a grand opportunity for continuing the fascinating plant cell story by failing to follow through with later eee by individuals show a progress in modern plant cytology. The beginnings of the realization that symbiosis was involved in development of eukaryotic plant cells would have been especially interesting oak discussions of the work and hypotheses of Famintzin and/or Mereschkovsky in the early part of this century. The cha eal are often peppered with amusing comments such as the statement that many botanists ‘odd ducks” (p. 124) which, I hope, will not offend most of us profes- sional botanists ae will accept it in the jocular mood in which it was probably intended. Describing Charles Deam as “correspondence-incontinent,” however, is a bit unkind, par- es pyel not knowing whether or not the recipients of his letters were appreciative. Also, a less amusing note, the revelation is fascinating that Mary Agnes Chase was jailed twice as a suffragette. There are a few errors that should be noted, namely the family name ‘Proteraceae’, instead of ee on page 111 for the family in which Banksia is placed. Also, on page 227 the ‘Bradypods’ are mentioned in reference to an animal group which I feel sure should be the brachiopods eran sats In general, however, the book is commendable by providing concise surveys of a se- lected 101 botanists, 3 it is hoped that Isely will follow through with a second book overing a similar number of other botanists, those forgotten in this first volume c —Richard E. Norr. Sipa 16(4): 795. 1995 796 Sipa_ 16(4) PILBEAM, JOHN 1995. Gymnocalycium, A Collector’s Guide. (ISBN 90- 5410-192 X, hbk). A.A. Balkema, P.O. Box 1675, 3000 BR Rotterdam, Netherlands (fax +31-10-4135947),; A.A. Balkema Publishers, Old Post Road, Brookfield, VT 05036 (fax 802-276-3837). $70.00. ix + 191 pp, 124 color photographs, 98 figures, 10 maps, So <1". Gymnocalycium is a favorite genus for amateur as well as professional collectors of cacti. It well meets the qualifications of an almost ‘perfect’ cactus in having many spines, often quite long, that are beautifully arranged in areolae, flowers that are often large, fragrant and very attractive and, most importantly, plant sizes and shapes that are easily accommo- dated in most collectors’ facilities. This volume will be heartily welcomed by English speaking enthusiasts who, to this time, have had similar treatises available mostly in for- eign languag ncluded in the book is a fine section on cultivation of these cacti, a review of the classi- egies of Gymnocalycinm species, mainly including a review of Schiitz (1986) and Krainz (1968), but also providing, in Chapter 7, an annotated checklist of species and synonyms I &, yy y according to Hunt (1992). The discussion in the section on seed, fruit, flowers, and spines is very brief. | would have liked to have had a more detailed discussion on the fruit and seeds of Gymnocalycium, particularly the latter because of the importance placed on seed characters in formulating subgenera and sections of the genus. Scanning electron micro- scope photographs of seeds of a few species are provided but without much ment. Perhaps light ree a photographs of seeds of species showing differences in : different sub oaiets would have been more useful. The chapter on Geography and Distribution of Gymnocalycium is written by Metzing and translated by C. Walton. It is erroneously stated in this section that “The area inhab- ited by o genus Gymnocalycium | is entirely in South America, and extends roughly from 18° N to 45° Su. . It is made clear, however, that the genus occurs no farther north chan central va in South America, and does not occur as far north as the latitude 18° N which passes through Mexico, Jamaica and Puerto Rico. Locations of different species in each of the subgenera are illustrated on full-page maps. A very short chapter is provided on “Discovery and collection of species in the wild” in which the maintenance of collectors’ numbers by growers of Gymnocalycium is emphasized so that their origin ts clearly identifiable. Chapter 8 lists 23 pages of numbers of collectors, giving ie for most in 1 variable detail. The glossary is aimed primarily at etymology of the species names, but the roots gyno and calycium are not exp ae -d. Books important to the taxonomy of Gymmnocalycium are itemized in the Bibliography, but publishers names are not provided, making it difficult for the uninitiated to find them. Chapter 6, containing a “Commentary on species,” provides a description of each ac- cepted species as well as those species that are ‘provisional’. Two types of ‘provisional’ species are designated, those that may have been accepted in the past but now, for various reasons, “await resolution,” i.e. G. bayrianum, and those that are previously undescribed or illegally described, 1.e. G. altagraciense. Species that are considered in this volume to be synonyms are also listed and discussed. All itemized species in this section are well refer- enced, although some of the references are not listed in the Bibliography section. Eac 1 accepted and provisional species is illustrated on a color photograph of a flowering plant Sipa_ 16(4): 796. 1995 Book REVIEWS 797 as well as a close-up black and white ihc of areolae. Each accepted species is as- signed to one of Schiitz’s subgenera and also to the series in which Buxbaum placed it. The descriptions include details of plant form oh growth habits as well as flowers, fruits, and seeds. Comments on related species or past confusion of plants in cultivation are often made. Location of wild collections are listed and collectors numbers provided for each species. My curiosity stimulated, and becoming a very small scale collector, I purchased seven unidentified plants, without flowers, appearing to be the genus Gymzocalycium from a local nursery, and have attempted to identify them using this book. The identification process ais mae through the book and comparing the descriptions and photographs with my p _ In about an hour and a half I have probably accurately identified two species, possibly ae but I must await flowering of the remaining plants to be able to try placing a species. A key to the subgenera and species would have made my task much = em easier ae possibly more accurate Despite these limitations, this beiaeiailly produced | book i is one that collectors of succu- lene plants, especially cacti, will require—Rzichard E. Norr HENDERSON, ANDREW, GLORIA GALEANO, and Roprico BERNAL. 1995. Field Guide to the Palms of the Americas. ISBN 0-691-08537-4, hbk). Princeton University Press, 41 William Street, Princeton, NJ 08540. $75.00. 352pp, 256 color illustrations on 64 plates, 42 line draw- ings, 554 maps, 6"x 9". the guest author of the forward to Field Guide to the Palms us the America, Robin Foster ae two bold statements: “Amazing. I oe it would be another twenty years before s bo something as useful as this came out ... ” and “* ok is a oe in the arm of the study of palms.” I could not ee more. This field ie will impact the taxonomy, ecology and enor of palms well into the twenty-first century. field guide?” you say, “After all, che authors state at the outset, “The Guide is not a taxonomic treatment but a field guide for nonspecialists’”. However, the term “nonspe- cialists” is not restricted to school children and casual tourists; it includes anyone who ts not involved in systematic rese a particular group, even the Bactris specialist who has never delved into ee or oe genera. It is ate to i used by anyone who needs to provide rapid, reasonably accurate Pent eestone of palms seen, studied, or col- lected in the field: the ecotourist wanting to appreciate the ecosystem, the palm enthusiast collecting seeds for distribution through . International Palm Society, the tropical ec gist inventorying biodiversity and plant-animal interactions, the natural resource man- — =) ager evaluating habitats, and the conservationist cataloguing endangered species for a world- fi wide database. What makes this Guide so significant is that, until now, the taxonomy of American palms had never been nee and simplified to bring order to all species in all genera and provide standard names among countries. In fact, only a handful of genera have been treated in modern comprehensive monographs and, ae are not readily acces- lacing the species feval taxonomy of American ee in historical perspective, the arbos authors state in the Introduction: “It had been left by a Rodrigues, Burret, Bailey, and others in a chaotic state with literally hundreds of names ie no way to apply them to real species in nature. ... Over the last decade a new generation of botanists, many of them Sipa 16(4): 797. 1995 798 Sipa_ 16(4) natives of tropical countries, has become interested in palms, and our knowledge of the taxonomy of American palms is now based on extensive field work, well-collected speci- mens, ian of variation in nature, and realistic species concepts. Our Guide is based on the work of these botanists.” Indeed, the three authors are themselves part of the new generation of palm biologists, representing research centers in both North and South America and a cotal of over 30 years of extended periods of field experience throughout the neotrop! ane making sense and order out of the American palm species for nonspecialists comes at a cost in terms of some “cherished” names and ne conceptualizations of nee in horticulcural use. The authors ex} ee it this way: “Our species concept in this ide is necessarily a broad one. ... We have tended to combine closely related and doubefally distinct species and also groups = species that we consider to be part of species complexes. ... Many species are thus quite ‘messy’ and not easy to understand. Herbarium taxonomy does not have the eens (and certainly not the number of specimens) to fully understand this kind of variation ... [which] falls more into other fields. ... Our particular concept means that in several instances we have not accepted species of previous botanists. We do not imply that we are ‘right’ and the others are ‘wrong’ but stress we have different concept ... often the difference in opinions are merely a question of ranki For example, the authors’ concept of Coccothrinax miraguama is expanded to encompass all the variation denoted by all the species with loosely, coarsely woven but non- -spiny leaf sheaths. Likewise Coccothrinax jamaicensis, C. sonia C. proctori, and C. readii are all placed in synonymy under the more familiar C. argent. a) aa y accounting for all known variation of all ‘ee palms within a simplified hier- archy of conceptually clear-cut, easily distinguished species, this guide accomplishes some- thing unique to the American palm literature: Any palm specimen can be c juickly identi- fied correctly to species (i.e. species complex), quickly plugged into the research network and become a vital resource for eventually understanding the biological basis of species variation in palms. This is the crux of the book's impact A successful field guide must also be easy to use and ee organized. By any standards, the Field Guide to the Palms meets these criteria, as well. Twenty-two pages of introduc- tory material (taxonomic concepts, palm geography, morphological structures, and con- ventions used in the book) precede an illustrated key to the genera. The genera and species accounts occupy the next 200 pages. Each species is consistently treated to list common names, field characters, distribution, uses, and notes on taxonomic problems or recent rereeuch, The next section includes a distribution map for each species. Appendices in- clude country by country checklists and lists of complete synonymy, which aaa an easy way to Compare species concepts without the species accounts becoming cluttered. The book ends in a spectacular display of 256 color ee ae en: the more com- mon, evolutionarily significant or structurally beautiful spe Because this is intended to be used in the field or with ‘field collections, the authors make extensive use of distributions and habitats to distinguish species, as well as high- lighting diagnostic field characters in the species accounts. Within the generic accounts, keys are provided for genera exceeding ten or so species. For smaller genera, the combina- tion of distribution, habitars/elevation, and diagnostic features are sufficient to readily distinguish species. For example, of the eleven species of Ceroxylon, only five occur in the ral region of the Eastern Cordillera of the Colombian Andes. Ceroxylon alpinum and C. quindoense both have horizontally held leaves and regularly arranged, horizontal or pendu- ous leaflets. Ceroxylon parvifrons has arched leaves and regularly arranged leaflets held in a Sipa 16(4): 798. 1995 Book REVIEWS 799 “V” formation. Ceroxylon vogelianum and C. sasatmae both have arched leaves and clustered, multi-ranked leaflets. Ceroxylon alpinum occurs below 1800 m and has pebbled fruits, whereas C. quindoense occurs above 2000 m and has smooth fruits. Ceroxlyon sasatmae occurs below 1800 meters and has smooth fruits, while C. vogelianum occurs above 2000 m and has grooved fruits. Although, some palm specialists and knowledgeable enthusiasts may grumble at the radically expanded species limits, the Field Guide is still the book to use by anyone want- ing to understand neotropical palm species. It is not only a field guide; it is a unified summary of our current understanding of American palm species and exposes the taxo- nomic problems that will provide research subjects or years to come. My congratulations to Henderson, Galeano and Bernal.—Roger W. Sanders, Research Associate, BRIT. Sipa 16(4): 799. 1995 800 Stpa 16(4) REVIEWERS FOR VOLUME 16, 1994-1995 The following individuals have kindly supported Spa through their time and expertise in reviewing manuscripts submitted and/or published in vol- ume 16. Without your interest and support, SIDA would not be the journal you have come to expect. The demand on an individual's time to review a manuscript is under- stood and taken into consideration. Your support is vital and appreciated. Reviewer support is appreciated and with it SipA can remain a top quality systematic botany journal. Sipa’s subscription base continues to grow with are 401 domestic and 443 foreign subscriptions. Outside of the United States, Brazil has the most subscribers. Sipa, CONTRIBUTIONS TO BOTANY is now distributed in 90 countries. Volume 16 is the largest published to date with some 809 pages and 82 manuscripts. We thank all authors, reviewers, subscribers, and readers for your con- tinued interest and support.—Barney Lipscomb (BRIT), Editor; John W. Thieret (NKU), Associate Editor; Félix Llamas (LEB), Contributing Spanish Editor. Allred, Kelly W. Gandhi, K.N. Nee, Michael Tai, William Austin, Daniel F nas Steve Nesom, Guy Taylor, W. Carl Barkley, Theodore M. Hall, David W. Nicolson, Dan H. Taylor, Connie Barneby, Rupert C. Benz, Bruce F. Berry, Paul David E. rown, Larry E Bryson, Charles. T. G hamberlain, D.F Boufford, Evans, Dan K Foster, Robin B. Freeman, Craig C. Fryxell, Paul A. Stipa 16(4): 800. 1995 Hansen, Bruce EF. Johnson, Dale E Johnson, David M. oo Stanley D. Kearns, Denis Keeney, Toney McKenzie, Paul M. Magrath, Lawrence K. Manhart, James Meerow, Alan W. Montz, Glen Moore, Michael O. ae Scott A. Naczi, Robert EC Niezgoda, Christine ee nnon, ae Kent Perkins, Peterson, Paul - ae HI, John J. Pringle, James S. Read, Monique D. Reznicek, ae ann N.K.B. Sanders, ahs Ww. Schwartz, Fayla po B. ellenberg, Richard - pson, Bery sti Peter FE. Strother, John L. Stuckey, Ronald L. Thieret, eae W. Tho Yale i. pe E. Tucker, J.M eae B.L. aldés-Reyna, Jestis aaa Michael J. Weber, William A. Webster, Robert Wendt, Thomas L. Wiersema, John W. Wilson, Hugh D. Woodruff, Lindsay Wurdack, J.J. Zanoni, Thomas Zarucchi, James L. INDEX TO VOLUME 16, 1994-1995 TITLes OF ARTICLES WITH AUTHORS A new species of Anemone (Ranunculaceae) from central Texas 2 Cari 5S. KEENER and BryaN E. Dutron 191 new species of Carex (ee Phaesto- glochin) from Oklahoma and Texas; aoe of section Phaestoglochin, and otes on sections Bracteosae and Phaesto- cae by ee D. Jones 341 A new species of Dirca eee from the Sierra of northeastern Mexico by Guy L. Nesom and Mark H. Mayrie_p 459 new species of Zigadenus See from New Mexico, with addit > nal comments A new species, Sy ad oan Caprifoliaceae), from witl Wesiean a Guatemalan species by JUSTIN KirK WILLIAMS 27 A new ae (Fabaceae: Casali pinioideae) from western Amazonia by OHN : Pipoty III 407 A revision of Mirabilis section Miérabilts (Nyctaginaceae) by Atice Le Duc 613 synopsis of the genus C/wsza sections Crinvopsis and Brachystemon (Clusiaceae) in northern South America b Preoty IH and ALLISON Grarr 505 A synopsis of the genus Packera (Asteraceae: Senecioneae) in Mexico by Craic C > FREEMAN and THEODORE M. BaRKLEY A taxonomic investigation of Cuscuta attenuata (Cuscutaceae) and related taxa by L ALAN PRATHER, RONALD Wittam D. Warpe 447 Amur honeysuckle (Lonicera maackii; Caprifoliaceae): its ascent, decline and fall by JAMes O. LUKEN and JOHN W THIERET 479 An undescribed Saccharum (Poaceae: Andro- goneae) and Kashmir, India by S. J. Tyrx, and northwest Himalaya, Sipa 16(4): 801. 1995 RaAJESHWARI, R.R. Rao, and A. GARG Anatomical study of Erioneuron and Dasyochloa ees ri ener Brag tostidets) in 4 America b ; VALDES- cy STEPHAN L. Ashicaulis, a new genus for some species of Millerocaulis (Osmundaceae) by WILLIAM D. TipweELt 253 Bulbiferous Acontivm (Ranunculaceae) of the NALD E. Carex conjuncta (Cyperaceae) verified for Arkansas, and notes on the range of Carex oklahomensis by STANLEY D. JONES and A.A. REZNICEK 77 Carex lutea Seca a rare costal cos endemic from Nor by LEBLOND, 3ZNICEK, anc h Carolina Clusia fabiolae, a new species with ¢ a synop- sis of Clusia section Anandrogyne (Clusiaceae) in Guayana by JOHN Preoty II 737 c an studies on North American spe- of Saccharum (Poaceae: Andropo- kone) a D.M. Burner and Roserr D ER 233 Dena pe (Leguminosae) in as by MARSHALL 1 Documented chromosome numbers 1994:1. Karyotype of hits gypsophila (Amaryllidaceae) by ZAIMING ZHAO Documented chromosome numbers 1995:1. Chromosome number of Cornus sessilis (Cornaceae): Phylogenetic affinity a evolution of chromosome ie: in Cornus by QIu-YUN XIANG and RicHarbD H. Eypbe 76 First records of the aquatic weed Hygrophila polysperma (Acanthaceae) from Texas by 802 MICHELLE B. ANGERSTEIN and Davin E. LEMKI Flora vasc cular de la Sierra de la Paila, Coahuila, by José A. VILLARREAL 109 Further notes on the genus Ardisia (Myrsina- ceae) in Madagascar by JOHN J. PIroty 361 Hemerocallis hakuunensis (Liliaceae) in Korea SOON SUK KANG and Myonc G1 CHUNG 23 Historical evidence of the native presence Sabal mexicana (Palmae) north of the Lower Rio Grande Valley by LaNnpon CKETT 711 Marsilea minuta (Marsileaceae): new to Florida th America by James R. BuRKHALTER 545 Muconta skeaniana (Melastomataceae: Mico- i New species eastern Cuba ALTER 8. JUDD Miscellaneous notes on ae (Apocyna- ceae: Plumerieae: Haplophytinae) by IRK WILLIAMS 469 New and noteworthy Malesian Myrsinaceae, VII by Benjamin C. STONE 263 New reports of Eragrostiy (Poaceae: Chlori- doideae) from Brazil by Sonja De Castro BorcHat M. PETERSON 769 by Wa and PAauL New taxa and new combinations in Chinese plants by Sutvou Li and Kent T. Apair 83 New taxa of Rhamnaceae from China by FAN Guo-SHENG and DeNG Li-LAN 477 Nomenclatural changes in Setaria and Paspalidium (Poaceae: Panice Roserr D. Wester 439 Notes on Carex (Cyperaceae), with C. godfreyi new to Alabama and C. communis and C. scoparia new to Mississippi by CHARLES T. Bryson, JOHN R. oe D, and Ranby WARREN 355 Notes on Carex, Cyperus, and Kyllinga hs ta- ae) in Mississippi with recor ee species previously ae to the state by CHARLES T. Bryson and RICHARD CARTER 171 jon ae) by ie} hy Stipa 16(4): 802. 1995 Stipa 16(4) Notes on Erjochloa weberbaneri (Poaceae: Paniceae) by Roperrt D. Wesster, PAUL M. Prrerson, and Rosert B. SHAw 57 Notes on the genus Cybianthus subgenus Cybranthus (Myrsinaceae) in Colombian Amazonia by JOHN J. Pipoty III Noteworthy plant 9a bP] 3355 species from the kefenokee Swamp, Georgia by KENT Witiices and CyntHIA S. Lorrin Noteworthy plants from north Florida. VI by Loran C. ANDE 581 On the hybrid nature er Quercus basa- seachicensis (Fagaceae, sect. Quercus) by RICHARD SPELLENBERG 427 On the identity of two taxa of Berberis (Berberidaceae) from Tibet by Tariq - BuHaskar Darr, and R.R. Rao Patan and site characteristics of a re- ’ discovered disjunct population of a alabamensis (Euphorbiaceae GREGORY PLET, RICHARD D. Laven, Maurya B. FALKNER, Rosert B. SHAW 37 Some observations of leaf form in I/ex vomitorta (Aquifoliaceae) by RicHaRD STALTER and DwiGHt T. Kincaip 79 Species Le in eastern Asia and North a by Suivou Li and Kenr T. os. 281 —— potostensis (Poaceae: Eragrosteae): a rhizomatous species from San Luis ae i, México, and a new combina- tion in §. asroides by — K. WIpFE and StaNntey D. JONE Submersion of Dugaldia at Plummera in Hymenoxys (Asteraceae: Helia Gaillardiinae) by Mark W. BiERNER | Subtrial classification of the New Worlc oo (Poaceae: ae ee by PauL M. Peterson, Roser WEBSTER, eT JESUS ae 529 Synopsis of the genus Lycoris (Amaryllida- y Hsu PING-SHENG, Siro Kurita, Yu ZuHi-ZHou, and Lin JIN-ZHEN 301 SS anc poo INDEX Systematic study of Texas populations of Phacelia patuliflora (Hydrophyllaceae) by JENNIFER ANN Moyer and BILLIE TurRNER 245 Taxonomy of Cyphomeris (Nyctaginaceae) ed on multivariate analyses of geogra- phic variation by MarrHew Manarr and RICHARD SPELLENBERG 679 Taxonomic notes on new varieties of species of Clitoria aed - Phaseoleae - Clitoriinae) by PauL R. Fanrz 72 Taxonomy of the native “Nowh American species of Saccharum (Poaceae: Andropo- goneae) by Roserr D. Wesster and Roperr B, SHAW 551 Taxonomy of the Sida rhombifolia (Malva- ceae) complex in India by V.V SIVARAJAN and A.K. PRADEEP 63 The genus Clusia section Crimva (Clusiaceae) in Guayana by JOHN J. Pipory III and ALLISON GRAFF 649 The Myricaceae of the United States and Canada: genera, subgenera, and series by Roperr L. WiLBur 93 Stipa 16(4): 803. 1995 803 The status of Potamogeton perfoliatus (Potamogetonaceae) in ake Pontchartrain, Louisiana by JOH Burns Jr., MICHAEL A. Sees AND Kris P. se 757 Thomas, Townsend, or Toamend what was ndegee’s name? A ee R. Burke, and tons 242 ee Two new species of Ichthyothere (Heliantheae: eas from Ecuador and Peru by Haroitp RoBINsON 731 Una nueva especie del género Preropepon (Cucurbitaceae) de Colombia by ALvARO CoGoLLo P. and JOHN J. PIpoLy HI 401 Validation of Tachigali micropetala (Ducke) rucchi & Pipoly (Fabaceae: Caesa pinioideae) by JOHN J. Prroty III 787 Vascular plant type specimens in the Uni- versity of Georgia Herbarium, with a brief _ of the herbarium by MicuHakt O. Moore and Davip G. cee 139 804 Sipa 16(4) BOTANICAL NAMES New names are in bold face. Aconitum columbianum 12, ssp. colum- 1anum 13, ssp. viviparum 13 Amaranthus palmeri 581 Anemone berlandieri 193, caroliniana 193, pees 196, var. edwardsiana 197, var. petraea 197, tuberosa 195 Anredera baselloides 581 Ardisia anceps 265, okennonii 198, ur. borneensis 265, prassiella 267, brasii 267, capuronii 1, mystica 265, reynosoi 263 ‘Aeisids Eee rae 589 Asclepias ee 584 Ashicaulis 255, amajolensis 256, beard- morensis 256, broganii 256, estipu- laris 256, irate 256, guptai 256, hebeiensis 256, herbstii 256, john- stonti 256, easend 256, kolbei 256, liaoningensis 256, patagonica 257, rajmahalensis 257, richmondii 257, sahnii, 257, santaecrusis 257, spink- sit 257, swanensis 257, wadei 257, 257, wrightii 257 Belamcanda chinensis 582 Berberis 17 websterii — Blechnum serrulatum 582 Brasenia 597 Calycanthus floridus 584 Carex aggregatus 177, amplifolia 592, bick- nellii var. opaca 172, communis var, communis 356, conjuncta 772, crus- corvi 773, fissa var. fissa 173, godfreyi 356, laevivaginata 773, lutea 155, montanae 592, oklahomensis 358, 772, perdentata 342, rossii 592, coparia var scoparia 357, stipata 773, verrucosa 776 eratopteris thalictroides 379 Chimaphila maculata 582 fat Clitoria 721, arborea var. pseudoama- zonica 721, epetiolata var. angustis- sima 722, falcata 723, var. aurantiaca 723, var. latifolia 724, guianensis 725, var. chapadensis 72 cleistogama 726, , Var. macro- var. pedunculata Sipa 16(4): 804. 1995 726, leptostachya var. fruticosa 727, mariana var. pubescentia 727, poly- stachya 728, var. congesta 728, var. pringlei 729, ee 729, stipularis var, latifolia 7) pozuzoensis var Clusia 505, 649, 737, tion Anandrogyne 738, araracuarae 507, asymmetrica 657, aymardii 5 amazonica 509, sec- section ee ee 517, cardonae 663, section Criuva 650, section Criuvopsis 506, ee 752, duidae 652, fabiolae 740, grammadenioides 653, guayanae 674, hexacarpa 665, maguireana 669, martiana 513, mel- chiorit 660, multilineata 655, opaca 672, pachyphylla 751, penduliflora 518, phelpsiae 743, radiata 668, rotun- difolia 744, savannarum 746, laefolia 522, wurdackiana 4 Coelorachis tuberculosa 584 Comptonia peregrina 105 Cornus 765, chinensis 765, mas 765, officinalis 765, sessilis 765 Crataegus iio 584 37 spathu- Croton alabame Cuscuta attenuata 447 Cybianthus barbosae 333, subgenus Spies 333, ruforamulus 336 Cyphomeris 679, 693, crassifolia 695, gypsophiloides 694 Cyperus aggregatus 177, difformis 178, drummondii 173, elegans 173, entreria- nus 174, eragrostis 174, flavicomus 178, grayoides 215, hystricinus 585, lancastrensis 179, louisianensis 175, ovatus 179, pilosus 179 Dasyochloa 413, pulchella 421 Desmodium lindheimeri 781 irca mexicana 460 Dugaldia Eleocharis ee 585 Emilia fosbergii 378 Eragrostis 769, barrelieri 769, gangetica 769, lehmanniana 770, macrothyrsa INDEX 770, mokensis 770, orthoclada 770, paniciformis 770, pastoensis 770, tremula 770 sa hus giganteus 776 riochloa weberbaueri 59 sien 413, avenaceum 419, grandi- florum 421, nealleyi 11, pilosum 417 Eyphomts graminea 208 ustachys caribaea 212, glauca 212 Ficus pumila 582 Fictingia mariae 267 Galactia elliott 582 Gentianella achalensis 375, armerioides 375, mirandae 375, primuloides 375, vargasil 37 Elaplopnyson 469, cimicidum 472, crooksii 472 Hemerocallis hakuunensis 23 Hygrophila polysperma 368 Hymenoxys 1, ambigens 6 subgenus galdia 5, hoopesii 6, integrifolia 5, inetorum ‘ subgenus plummera 6, microcephala 6 Ilex vomitoria i. Ichthyothere 731, pastazensis 734 Jaimehintonia gypsohpila 203 a pas s 778 Kari Kylings (cai 176 arix gmelini var. genhensis 183 8 du- macdanielit 731, arix can earlei Limnophila sessiliflora 379 Lindera subcoriacea 582 inum westii Lonicera maackii 479 ae alata 779 Lycoris a xalbiflora nee anhuiensis 312, argentea 308, aurea 3 urea var. an- oe ala 318, aurea var. aurea 315, aurea var. surgens 318, caldw ellii 513% chinensis 314, elsiae 322, guangxiensis 314, xhaywardii 308, xhoudyshelii 320, incarnata (310, aes 312, var. flava ata 323, var. kazukoana 326, vat. amit 335, xrosea 326, sanguinea 310, var. kiusiana 311, var. koreana 311, shaanxiensis 314, spren- Stipa 16(4): 805. 1995 805 geri 307, oe 309, straminea 21, trau 318 Malan abies ae Marsilea minuta 545 a alabamensis 585 odium divaricatum 582 210 atelea Melam Melica cilia Miconia skeaniana 226 Millerocaulis 255, chubutensis 255, dunlopii 255, indentata 255, indica 255 Mirabilis 613, donahooiana 635, exserta 624, gracilis 639, hintoniorum 625, jalapa 640, longiflora 628, var. wrigh- tiana 630, section Mirabilis 619, polo- nii 627, pringlei 622, sanguinea 633, var. breviflora 635, — nii 632 Monotropa hypopithys ‘ rella 99, californica i: )2, caroliniensis 102, subgenus cerothamnus 99, se- ries cerothamnus 100, series By 103, inodora 102, pensylvanica 102 Myrica 103, gale 104, hartwegii 104 Oenothera Sere 215 jae 699, bel idifolia 703, candidissima 707, hintoniorum nontereyana 1 703, seen 704, 1 7103, quebradensis 702, rosei 703, 5 peer 706, var. carmenensis 706, var. parrasiana 706, eee 706, tampicana 705, zimapanica Panicum nudicaule 585 a ues 439 Penstemon thurberi 207 sn eee sedoides ssp. chinense 190, sedoides ssp. sedoides 189 Phacelia patuliflora 246,var. austrotexana 2 o, patuliflora 247, teucriifolia 249 eon ionantha 585 Pinus a var. manguiensis 184 Plummera 1 eae hingganicum 185 Potamogeton perfoliatus 757 Pteropepon o 4 r Quercus basaseachicensis 427, laceyi 381 806 Rhamnus hemsleanus var. paucinervatus 477 yarviflora 585 Rhynchospora capitellata 595, crinipes 586 leptocarpa 586 Rhexia — > Ruellia pedunculata ssp. pinetorum 586 Sabal mexicana 7 Saccharum 233, 551, 237, brevibarbe 237, brevibarbe var. contor- tum 237, coarctatum 237, 572, gigan- eriffithi 35, stewart 33 Sageretia yanlongensis 477 Sagina procumbens 377 Salvinia minima 595 alopecuroideum 237, baldwinti revibarbe var. teum 237, Schisandra glabra 213 Scirpus subterminalis 583 Setaria eye a 441, aversa 441, basic- ie 44 ae 441, chapmanii , Seca tii 442, constricta 442, ee 442, distans 442, dis- tantiflora 442, flavida 442, gausa 443, permis 443, geminata var. paludi- 443, globoidea 443, grandis- piculata 443, inaequalis 443, jubi- Vaga - Stipa 16(4):; 806. 1995 Stipa 16(4) flora 443, leonis 443, ophiticola 444, pradana 444, punctata 444, rara 444, ‘Aa retiglumis 444, scabri- pis 444, spartellum 445, subran- siens 445, tabulata 445, uda 445, utowanaea 445 Sida alnifolia 69, rhombifolia 71, boidea 73, scabrida 7‘ Sideroxylon lycioides S86 Silene caroliniana 583 Solanum viarum 382, 583 Sphagneticola trilobata 583 Sporobolus airoides ssp. regis 16 siensis 165 rhom- 4, poto- Symphoricarpos guatemalensis 273, longiflorus 276, microphyllus 277, orbiculatus 276, palmeri 278, parishii 279 Tachigali bracteosa 411, micropetala 411, 787, rugosa 411, vasquezii 408 Tapeinosperma 271, filipes 271 Tapeinostemon zamoranum 376 Thlaspi montanum var. montanum 214 Zigadenus 389, leimanthoides 584, ollonensis 390 mog- Adair, Ken Anderson, Lon C. a Aplet, Gre ee) Barkley, ae M 699 Bierner, Mark W. Boechat, Sonja De ee 769 Brink, Donald E. 9 Bryson, Charles T. 171 Burkhalter, James R. 545 Burns Jr., John W. 757 Carter, Richard 171 Chung, Myong Gi 23 Cogollo P., Alvaro 401 Crins, WJ. me Datt, Bhaskar 1 Dutton, Bryan : 9 Enquist, Marshall 781 Eyde, Richard H. 765 Falkner, eee 37 Fantz, Paul R. cae Craig e a Garg, ofS) Giannasi, David E Graff, ie 505, a Guo-Sheng, Fan 477 Hatch, Stephan L. 413 Hess, William J. 3 Husain, Tariq Kincaid, Dwight T. 7 7 Laven, Richard D. 3 tBlond, R.J. 153 Loftin, Cynthia S. 775 Stipa 16(4); 807. 1995 AUTHORS INDEX Luken, James O. 479 Mahrt, Matthew 679 Mayfield, Mark H. 459 Moore, Michael O. 139 Nesom, Guy L. 4 5 Peterson, Paul M. 57, 529, 76S 807 Pipoly, John J. 401, 407, 505, 649, 737, 787 Poirrier, Michael A. 757 2a) Vimo}, Reznicek, A.A. 153, 772 Robinson, Harold 731 Shaw, Robert B. 37, 57, 551 a V.V. 63 nski, Robert C. 389 a ee ee oa 427, 679 Stalter, Richa Rajeshwari, S. 33 ao, Stern, Kingsle i 9 Thieret, John W. 479 Tyrl, Ronald J. 447 Valdés-Reyna, ae 413, 529 es Q., José A. 109 "Willam D. 447 anne S. 153 Webster, ae D. 439, 529, 551 Williams, Justin Kirk 469 Williges, Kent A. 775 Wipff, Joseph K. 163 Woods, Jennifer A. 9 Arang, Qiu-Yun 765 Zhao, Zaiming 203 808 Sipa 16(4) SIDA GUIDELINES FOR CONTRIBUTORS Stpa, CONTRIBUTIONS TO BOTANY is an international journal of systematic botany con- taining primary research papers sensu lato, eee anatomy, biogeography, chemotax- onomy, Cladistics, ecology, floristics, genetics & evolution, numerical taxonomy, paleo- botany, and palynology. Sipa is open to all ne anywhere; coverage is not restricted to any geographical area All manuscripts submitted to SIDA are considered by at least two reviewers. Manu- scripts may be in English or Spanish. Page costs are $35 per page but may be waivered or reduced under certain circumstances. Please contact the editor and make arrangements before publication. Manuscripts are not rejected due to lack of financial aes Submit manuscripts and editorial questions to Barney Lipscomb, Botanical Research Institute of Texas, 509 Pecan Street, Forte Worth, TEXAS 76102-4060, U.S.A. Our joie aes mail address is sida@brit.org. Access our SIDA home page at the URL Aftp://www. brit.org/sida/ Manuscripts must be double spaced throughout and three copies sent to the editor. Upon acceptance for publication please return two copies along with a copy in . word- processing program format (with format commands) and a copy in an ASCII t file (without format commands) on a either 3!/2 or 5'/4 data disk. Consult the ane issue of Sipa for format of articles and notes regarding title, author, and address. Ranges of num- bers are separated by double oe (6--8). References, footnotes, and legends should be typed on separate pages and inserted at end of manuscript. jon Abstract Every paper should include both an English and Spanish, or another major language abstract. Spa is has a subscription base of 850 individuals and institutions in 90 countries with over half outside the United States. Specimen Citations Use the collector's last name and number or (s.n.) and underline. Abbreviate the months of the year by the first three letters of the month without a period. Example: Texas. Jeff Davis Co.: | mi W of Fort Davis, 3 Jan 1972, Smith 118 (BRIT). Author Citations Use the author abbreviations as given in the Author Abbreviations compiled at the Her- hare Rove Botanic Gardens, Kew (1992). Use et or the ampersand (&) between two ; for more than two authors, restrict to the first one followed by et al. Do not italicie or underline in che manuscript the terms et, et al., or in Latin wie ayes or Diagnoses scription or diagnosis of a new taxon should mention the taxonomic characters that repeats the taxon from its allies. These diagnostic characters should be underlined in the description. Since a diagnosis of a new taxon refers only to the taxonomic characters distinguishing it from other taxa, no See: in che manuscript is necessary. The (Rec. 32B.1) recommends that the main points in which the taxon differs from its allies should be mentioned; italicizing (ms) these in the description is in accord with that recommendation. Sipa 16(4): 808. 1995 GUIDELINES 809 Figures A scale should be in or on the illustration or eae sea reduction will a reduce the scale and subject proportionately. W this is not feasible, the proofs wil returned with the galleys for the author to ae the magnification of the final es tion. In half-tones, sharp glossy photographs with good contrast are necessary for good reproduction. Please do not mount illustrative material (half-tones, line drawings, etc.) on stiff boards. This prevents scanning with high quality equipment. References This includes all of the literature cited in the text and may include other article cita- tions the author fe deam desirable. Normal text references should be cited as follows: ‘Smith (1976) stated ...’ ‘the latest revision (Smith 1980) when reference is used as authority for a statement. Whe n there are three or more authors use only the name of the first author followed by et al.: ‘Smith et al. (1933) stated’. References at the end of the article are arranged alphabetically and Pama ei making use of a,b, etc. if an author had more than one publication in a given year. Author’s names are typed in lower case xcept for initial capitals. Only the first letter of the initial word, proper nouns, and proper adjectives of titles should be capitalized. When the auchor of the relevant part of a book is not an author of the book, the sequence, punctuation, and spacing for references to that part are as follows. Author(s) of the part. Year. Title of the part. In: Author(s) or are Title of the book. Place of publication: Publisher. Pages of the part. For journal abbrevia- tions, use Botanico-Periodicum-Huntianum. Abbreviations When the pan: abbreviations are used the period is omitted. Distance: mm, cm, m, km, ft, mi; directions (in caps): $ , E, W, months: first 3 letters only, Jan, Beh. etc. Example: iia Co.: 2 km W of Dot; 5 Jun 1971, Smith 118 (SMU). Notes When references are few in number, cite them in the text and omit reference section. Text citations of literature are in parentheses and should state the journal abbreviation, volume, pages, and year. The citation of a book includes the last name of the author, if no mentioned previously, the full unabbreviated title of the book, and the year published in parentheses. Documented Plant Chromosome Numbers The year and number of each is assigned by the editor. Refer to Sipa 16:765—768 1995 Sipa 16(4): 809. 1995 Notes New reports of Eragrostis (Poaceae: Chloridoideae) from Brazil 769 Carex conjuncta (Cyperaceae) verified for Arkansas, and notes on the range of Carex a5 ~~ oklabomensis Noteworthy plant species from the Okefenokee Swamp, Georgia 775 Desmodium lindbetmeri (Leguminosae) in Mexico and Texas 781 Validation of Tachigali micropetala (Ducke) Zarucchi & Pipoly (Fabaceae: Caesalpinioideae) 7% OD ( Reviewers for volume 16, 1994-1995 &¢ G0 Index to volume 16, 1994-1995 Sipa guidelines for contributors 808 Book notices and reviews 698, 710, 720, 756, 764, 789-799 Announcement 788 VY N ISSN OOA6- 1488