An internariopall Rad EG Beh aig plant systematic, phytogeographical - 9 Few! Tad oC AO, 3iGal publication Vol. 72 february 1992 No. 2 W YORK Mor CONTENTS ~ LITTLE, R.J., Nomenclatural changes in Viola (Miolaceae)e es Mou hes 77 - LITTLE, R.J., A nomenclatural change in Viola (Violaceae). ........... 79 - KARTESZ, J.T. & K.N. GANDHI, Nomenclatural notes for the North Amer- SEAM Aae MN MMUMOL MLN RG VASA AAR Lucie biota mad \ gu oy 80 ~ JONES, S.D. & G.D. JONES, Cynodon nlemfuénsis, (Poaceae: Chlorideae) prevonslyunreponted wm Mex! |.) 5') se) oh sas vives ak tle 93 ~ SMITH, L.B. & M.A. SPENCER, Reduction of Streptocalyz (Bromeliaceae: POM EVEMMER ES EV UNCER ni PMMV NIE Sadie Vlehu sy Vk RUDY NS yy 96 - RAYNER, T.G.J., Studies on the genus Bidens L. (Compositae) from the Eastern Hemisphere. 2. New combinations and a new subspecies for the fora of Tropical Bast Afriog area...) Oo) 0c// Las ove vel sco ke det 99 ~ TURNER, B.L., A new species of Coreopsis (Asteraceae) from southern Be a AN tad A a CV SE OPIN WOR OV A 103 - NESOM, G.L., A new species of Pedicularis (Scrophulariaceae) from western DURE AN CRS ete DLV UCM CLNNAL NAAN A Alek i 105 » TURNER, B.L., Two new species of Verbesina (Asteraceae) from southern EOD Mer aNee RAPPER AUREL YEE OE ABN MAL RUAN 50) 109 ~ TURNER, B.L., New species of Wedelia (Asteraceae, Heliantheae)from México and critical assessment of previously described taxa. .............. 115 —~-Contents continued on the inside cover. tele # MEM RNVOR Mag ane Ne ED DL EL DAN UY LIN CY TAN Rte CLAD UA Get a Published by Michael J. Warnock 185 Westridge Drive Huntsville, Texas 77340 U.S.A. PHYTOLOGIA is printed on acid free paper. (Contents continued) - TURNER, B.L., Two new species of Stevia (Asteraceae, Eupatorieae) from Mexico. haus Gee pe de ai sueie glove 4 jae d cele: Berber 127 - MACROBERTS, M.H. & B.R. MACROBERTS, Floristics of a sandstone glade in western Louisiana.) .......,.'../.'s.'.s's\sn' less iene ..130 GRANT, J.R., Two new introduced species in the Alaskan flora. ...... 139 _ ROBINSON, H. & J. CUATRECASAS, Thelechitonia Cuatrecasas, an older name for Complaya Strother (Ecliptinae-Heliantheae-Asteraceae). . 141 * ROBINSON, H., New combinations in Elaphandra Strother (Ecliptinae-Helian- theae-. Asteraceae), ) 0023.00 001 cdc ce dae ents er 144 ORZELL, S.L. & E.L. BRIDGES, Xyris isoetifolia Kral (Xyridaceae) new to Alabama and its range and habitats in Florida. ..................- 152 PHYTOLOGIA (ISSN 00319430) is published monthly with two volumes per year by Michael J. Warnock, 185 Westridge Drive, Huntsville, TX 77340. Second Class postage at Huntsville, TX. Copyright ©1991 by PHYTOLOGIA. Annual domestic individual subscription (12 issues); $36.00. Annual domestic institutional subscription (12 issues): $40.00. Foreign and/or airmail postage extra. Single copy sales: Current issue and back issues volume 67 to present, $3.50; Back issues (previous to volume 67), $3.00 (add $.50 per copy postage and handling US [$1.00 per copy foreign]). Back issue sales by volume: $17.00 per volume 42 (not all available as complete volumes); $21.00 per volume 67-present; add $2.00 per volume postage US ($4.00 per volume foreign). POSTMASTER: Send address changes to Phytologia, 185 Westridge Drive, Huntsville, TX 77340. vi NM ae Phytologia (February 1992) 72(2):77-78. NOMENCLATURAL CHANGES IN VIOLA (VIOLACEAE) R. John Little University Herbarium, University of California, Berkeley, California 94720 U.S.A. ABSTRACT Viola purpurea Kellogg var. grisea Jepson is recognized as a sub- species of V. pinetorum E. Greene. Viola quercetorum M. Baker & J. Clausen is recognized as a subspecies of V. purpurea. Viola lobata Benth. var. integrifolia §. Watson, and V. sororia Willd. var. affinis L.E. McKinney are changed to subspecific rank for the purpose of con- sistency. KEY WORDS: Viola, Violaceae, California A revision of the California Violaceae for the Jepson Manual Project has resulted in the need to make the following nomenclatural changes: Viola pinetorum E. Greene subsp. grisea (Jepson) R.J. Little, stat. et comb. nov. BASIONYM: Viola purpurea Kellogg var. grisea Jepson, Fl. Calif. Vol. 2:521. 1936. TYPE: U.S.A. California: Tulare Co., Templeton Meadows, near Kern Peak, 8500 ft, 5 July 1912, W.L. Jepson 4977 (HOLOTYPE: JEPS!). Viola purpurea Kellogg subsp. quercetorum (M. Baker & J. Clausen) R.J. Little, stat. et comb. nov. BASIONYM: Viola quercetorum M. Baker & J. Clausen, Leafl. W. Bot. 5:101-102. 1948. TYPE: U.S.A. California: Kern Co., 2.4 mi west of Glenville, on road to Woody, near summit (just W of Linn Valley), 2800 ft, 12 April 1935, D.D. Keck & J. Clausen 3186 (HOLOTYPE: CAS!; Isotypes: GH,MO,NY,UC!,US). In keeping with the use of subspecies in the Viola treatment for the Jepson Manual Project, the following changes in status are made: 77 78 PHY TOLO GIA volume 72(2):77-78 February 1992 Viola lobata Benth. subsp. integrifolia (S. Watson) R.J. Little, stat. et comb. nov. BASIONYM: Viola lobata Benth. var. integrifolia S. Watson, Bot. California 1:57. 1876. Viola sororia Willd. subsp. affinis (J.E. Le Conte) R.J. Little, stat. et comb. nov. BASIONYM: Viola sororia Willd. var. affinis (J.E. Le Conte) L.E. McKinney, Sida Botanical Miscellany, No. 7. 1992. ACKNOWLEDGMENTS I thank Dr. Barbara Ertter and Dr. Elizabeth Neese of the University Herbarium, University of California, Berkeley, for reviewing this paper. Phytologia (February 1992) 72(2):79. A NOMENCLATURAL CHANGE IN VIOLA (VIOLACEAE) R. John Little University Herbarium, University of California, Berkeley, California 94720 USA: & Landon E. McKinney Dept. of General Biology, Herbarium, Box 112, Station B, Vanderbilt, Tennessee 37235 U.S.A. ABSTRACT Viola primulifolia L. var. occidentalis A. Gray is changed to subspe- cific rank. KEY WORDS: Viola, Violaceae In keeping with the use of subspecies in the Viola treatment for the Jepson Manual Project, V. primulifolia var. occidentalis is changed to subspecific rank. Viola primulifolia L. ssp. occidentalis (A. Gray) L.E. McKinney & R.J. Little, comb. nov. BASIONYM: Viola primulifolia L. var. occidentalis A. Gray, Bot. Gaz. (Crawfordsville) 11:255. 1886. ACKNOWLEDGMENTS We thank Dr. Barbara Ertter and Dr. Elizabeth Neese of the University Herbarium, University of California, Berkeley, for reviewing this paper. 79 Phytologia (February 1992) 72(2):80-92. NOMENCLATURAL NOTES FOR THE NORTH AMERICAN FLORA. X. John T. Kartesz and Kancheepuram N. Gandhi North Carolina Botanical Garden, Dept. of Biology, Coker Hall, University of North Carolina, Chapel Hill, North Carolina 27599-3280 U.S.A. ABSTRACT The nomenclature of Ceratotheca triloba (Bernh.) E. Mey. ex Hook. f., Linaria vulgaris Miller, Pediomelum digitatum (Nutt. ex Torr. & Gray) Isely, P. hypogaeum (Nutt. ez Torr. & Gray) Rydb. var. scapo- sum (A. Gray) Mahler, P. hypogaeum var. subulatum (Bush) Grimes, P. pentaphyllum (L.) Rydb., Symplocarpus R.A. Salisbury ez Nutt., and S. foetidus Nutt. is discussed. Six new combinations are proposed: Draba arctica Vahl ssp. ostenfeldii (Ekm.) Kartesz & Gandhi and var. os- tenfeldii (Ekm.) Kartesz & Gandhi; Ipomopsis congesta (Hook.) V. Grant ssp. nevadensis (Tidestrom) Kartesz & Gandhi; Penstemon crandallii A. Nels. ssp. taosensis (Keck) Kartesz & Gandhi; Physalis subulata Rydb. var. neomexicana (Rydb.) Waterfall ex Kartesz & Gandhi; and Senna artemisioides (Gaudichaud ez A.DC.) Kartesz & Gandhi. Cardamine bulbosa (Schreb. ez Muhl.) B.S.P., Carissa macro- carpa (Eckl.) A. DC., Evolvulus arizonicus A. Gray, and Kalanchoe mar- morata Baker are recognized in place of Cardamine rhomboidea (Pers.) DC., Carissa grandiflora (E. Mey.) A. DC., E. laetus A. Gray, and K. grandiflora A. Rich., respectively. The quadrinomial Lupinus parv- iflorus Nutt. ec Hook. & Arn. ssp. myrianthus (E. Greene) Harmon var. myrianthus is treated as two trinomials: L. parviflorus ssp. myr- ianthus (E. Greene) Harmon and Lupinus parviflorus var. myrianthus (E. Greene) Harmon. KEY WORDS: Apocynaceae, Araceae, Brassicaceae, Convolvu- laceae, Crassulaceae, Fabaceae, Pedaliaceae, Polemoniaceae, Scro- phulariaceae, Solanaceae, Cardamine, Carissa, Ceratotheca, Draba, Evolvulus, Ipomopsis, Kalanchoe, Linaria, Lupinus, Pediomelum, Penstemon, Physalis, Senna, Symplocarpus 80 Kartesz & Gandhi: Nomenclatural notes - X. 81 INTRODUCTION Continuing with the “NOMENCLATURAL NOTES FOR THE NORTH AMERICAN FLORA” (Kartesz & Gandhi 1989, 1990a, b, c, 1991a, b, c, d, 1992), a tenth note in the series is presented here toward advancing our understanding of North American plant names. APOCYNACEAE Carissa macrocarpa The amatungulu, a coastal shrub of South Africa, belonging to the genus Carissa L., has become established in some areas of southern Florida. Palmer & Pitman (1972, p. 1901) and Retief (in Gibbs Russell et al. 1987, p. 152) as- signed the name C. macrocarpa (Eckl.) A. DC. (based on Arduina macrocarpa Eckl.; published in 1830) to amatungulu, whereas Rosatti (J. Arnold Arbor. 70:346. 1989.) used the name C. grandiflora (E. Mey.) A. DC. (based on A. grandiflora ©. Mey; published in 1836). Although Rosatti cited Palmer & Pit- man’s usage of the name C. macrocarpa, he provided no reason for accepting C. grandiflora. Based on priority, we assert that C. macrocarpa is the correct name to be used. Carissa macrocarpa (Eckl.) A. DC., Prodr. 8:336. 1844. BASIONYM: Ar- duina macrocarpa Eckl., S. African Quart. J. 1:372. 1830. Arduina grandiflora E. Mey., Comm. Pl. Afr. Austr. 191. 1836. Carissa grandiflora (EK. Mey.) A. DC., Prodr. 8:335. 1844. ARACEAE Symplocarpus foetidus Regarding the usage of the conserved generic name Symplocarpus R.A. Sal- isbury ex Nuttall, Gen. N. Amer. 1:105. Jul 1818. (fide Stafleu 1978, p. 305), Mabberley (Taxon 29:601. 1980.) amended the author and bibliographic cita- tion: “[Salisb. ez] Barton, Med. Bot. 1:124. 1817.” We agree with Mabberley regarding the correct bibliographic citation, but disagree with his author cita- tion. Our discussion follows on the authorship of the generic name and on the combination S. foetidus. On p. 123, W. Barton introduced the combination Symplocarpus foetidus, ascribed it to Salisbury, and cited numerous references (pp. 123-124), including “Lin. Sp. Pl. p. 1372” (which refers to the basionym Dracontium foetidum 82 PHY D2OL.O GIA volume 72(2):80-92 February 1992. L.) and “Nuttall, Gen. Am. plants, p. 105.” Since W. Barton mentioned the correct page number from Nuttall’s work seven months prior to its publication, we assume that W. Barton had access to page proof of Nuttall’s manuscript. Moreover, of the many references cited by W. Barton, Nuttall alone used the combination S. foetidus. On p. 124, W. Barton treated the genus Symplocarpus, ascribed the generic name to Salisbury, and provided Nuttall’s Latin diagnosis (copied verbatim from Nuttall’s manuscript), which was followed by a description of its repro- ductive parts, copied from his uncle B. Barton (Fil. Virginica). We consider that Nuttall’s generic diagnosis is fundamental in validating the generic name. Benjamin Barton’s description of it must be considered of secondary impor- tance. Since Nuttall ascribed the generic name Symplocarpus to Salisbury, the complete author citation is: R.A. Salisbury ez Nuttall. In personal commu- nication, both D. Nicolson (US) and S. Thompson (CM) concurred with this authorship. The authorship of Symplocarpus foetidus is more complicated. On p. 124, W. Barton provided an elaborate treatment of S. foetidus. He mentioned the six word description of Michaux (given under the name Pothos foetida [L.] Ait.) along with a detailed description quoted verbatim from Nuttall. However, a formal description was unnecessary for the validation of the name S. foetidus. For this reason, Nicolson and Thompson (pers. comm.) attributed the combination to: R.A. Salisbury ez W. Barton. However, we do not accept their conclusion for the following reasons. Although W. Barton ascribed both the generic and specific names to Salis- bury, W. Barton admitted (p. 126) that he had not seen Salisbury’s work and that he relied on Nuttall for the names. Had Barton not referenced Nuttall in the protologue of Symplocarpus foetidus, then W. Barton would be credited as the correct author for the species. However, that was not the case. First, W. Barton relied upon Nuttall for the generic and specific names in discussion (i.e., Symplocarpus and S. foetidus); second, he had access to galley proof of | Nuttall’s work; and third, he copied Nuttall’s description. Therefore, we as- | sert that W. Barton inadvertently published the combination for Nuttall; thus, | Nuttall must be credited for the authorship of the specific name. Since Nut- tall did not ascribe the combination to anyone else, he alone is the combining author. Symplocarpus R.A. Salisbury ez Nuttall in Barton, Med. Bot. 1:124. Dec 1817. Symplocarpus foetidus (L.) Nuttall in Barton, Med. Bot. 1:124. Dec 1817. Kartesz & Gandhi: Nomenclatural notes - X. 83 BRASSICACEAE Cardamine bulbosa and C. rhomboidea Presently, Cardamine bulbosa (Schreb. ex Muhl.) B.S.P. and C. rhomboidea (Pers.) DC. are considered to be conspecific. Al-Shehbaz (1988, p. 94) accepted the name C. rhomboidea for this complex and cited “Arabis rhomboidea Pers., A. bulbosa Schreber ex Muhl. (nomen nudum), C. bulbosa (Schreber ez Muhl.) BSP.” as synonyms. Regarding the nomenclature for these two epithets, our analysis follows. Merrill & Hu (1949, pp. 42-43) discussed the nomenclature of Arabis bul- bosa, but they were uncertain regarding its validity in Muhlenberg’s work (Trans. Amer. Philos. Soc. 3:174. 1793.). Although Muhlenberg provided no description for A. bulbosa, he cited the following reference: “Clayton 99, n. 45.” This citation refers to Clayton’s composite work: Gronovius’ Flora Virginica, ed. 2, which was based primarily on Clayton’s collections as indicated on the title page. On p. 99, Gronovius described six Arabis species, one of which was based on Clayton’s collection (no. 45). Since Gronovius used polynomials (e.g., “Arabis foliis ovatis denticulatis glabris” on p. 99), names in his work are considered as invalidly published (Greuter 1988, JCBN Art. 23.6c); albeit, his descriptions (e.g., “Hesperis flore specioso albo, ... siliqua longa tenui. Tota planta Cochleariae sapore praedita.” for Clayton’s plant) were effectively pub- lished. Since Muhlenberg cited a direct reference to an effectively published description, he indeed met the requirements for valid publication (JCBN Arts. 32.1c, 32.3, and 32.4). Therefore, the name A. bulbosa was validly published by Muhlenberg in 1793 and has priority over the name A. rhombozdea. Cardamine bulbosa (Schreb. ex Muhl.) B.S.P., Pre. Cat. 4. 1888. BA- SIONYM: Arabis bulbosa Schreb. ex Muhl., Trans. Amer. Philos. Soc. 3:174. 1793. HOLOTYPE: U.S.A. Virginia: Clayton 45 (BM). Arabis rhomboidea Pers., Syn. Pl. 2:204. Nov 1806. Cardamine rhom- boidea (Pers.) DC., Syst. Nat. 2:246. 1821. _ Draba arctica ssp. ostenfeldit and var. ostenfeldii Bocher (Biol. Skr. 14(7):19. 1966.) proposed the following combination: _ Draba arctica Vahl ssp. ostenfeldit (Ekm.) Bocher. Under ssp. ostenfeldii, he (p. 28) recognized two varieties: var. ostenfeldii (without an author) and var. oviboviana Ekm. However, for his new subspecific combination, Bocher failed to provide complete bibliographical particulars of the basionym as required by ICBN Art. 32.3. Hence, Bocher’s two combinations remain invalidly published. Since we follow Bocher’s treatment of this complex, we simultaneously validate Bocher’s usage of subspecific and varietal names below (as allowed by JCBN Art. 34.3). 84 Pay PO DOGTA volume 72(2):80-92 February 1992 Draba arctica Vahl ssp. ostenfeldii (Ekm.) Bocher ez Kartesz & Gandhi and var. ostenfeldii (Ekm.) Bocher ex Kartesz & Gandhi, ssp. et var. comb. nov. BASIONYM: Draba ostenfeldii Ekm., Sv. Bot. Tidskr. 23:491. 1929. Note: Dr. Ekman cited several specimens. To the best of our knowledge, no lectotype has been designated. CONVOLVULACEAE Evolvulus arizonicus Austin (Sida 14:278. 1990.) used the name Evolvulus laetus A. Gray (Proc. Amer. Acad. Arts 17:228. 1882.) and cited E. arizonicus A. Gray (“Syn. Fl. N. Amer. 2, 1:218. 1886.”) as a synonym. Prior to Austin’s publication, £. arizonicus was the accepted name for this taxon. Gray’s “Syn. Fl. N. Amer. vol. 2(1)” was first issued in 1878 and was reissued in 1886 as a second edition, under the title The Gamopetalae (Stafleu & Cowan 1976, p. 992). The name E. arizonicus was validly published in the first edition (7.e., in 1878), and thus has priority over the name EL. laetus. We communicated our finding to Mr. Barney Lipscomb (editor of Sida), who in turn communicated this information to Austin (for the purpose of providing accurate nomenclature within that journal). Based on our finding, Austin (Sida 14:447. 1991.) correctly cited the date of publication of Evolvulus arizonicus and reinstated it in place of EF. laetus. Evolvulus arizonicus A. Gray, Syn. Fl. N. Amer. 2(1):218. 1878. Evolvulus laetus A. Gray, Proc. Amer. Acad. Arts 17:228. 1882. CRASSULACEAE Kalanchoe marmorata For the plants commonly referred to as lifeplant, Austin & McJunkin (J. Arnold Arbor. 59:62. 1978.) used the name Kalanchoe grandiflora A. Rich. (published in 1847-48). Unfortunately, these authors failed to realize that Richard’s name is a later homonym of K. grandiflora Wall. er Wight & Arn. (published in 1839). Since K. grandiflora A. Rich. is an illegitimate name, it is replaced by K. marmorata Baker (published in 1892). Kalanchoe marmorata Baker, Gard. Chron. 2:300. 1892. Kalanchoe grandiflora A. Rich., Tent. Fl. Abyss. 1:310. 1847-48, non Wall. er Wight & Arn. 1839. Kartesz & Gandhi: Nomenclatural notes - X. 85 FABACEAE Lupinus parviflorus ssp. myrianthus and var. myrianthus Following Kartesz & Gandhi (1991e), the quadrinomial Lupinus parviflorus Nutt. ez Hook. & Arn. ssp. myrianthus (E. Greene) Harmon var. myrianthus is recognized with two trinomials as given below. Lupinus parviflorus Nutt. er Hook. & Arn. ssp. myrianthus (E. Greene) Har- mon and var. myrianthus (E. Greene) Harmon, Trans. Missouri Acad. Sci. 6:161. 1973. BASIONYM: Lupinus myrianthus E. Greene, Pittonia 4:134. 1900. Pediomelum Grimes (1990) made many new combinations in Pediomelum Rydb. Un- fortunately, three of those names were made prior to 1990 by other authors. Since Grimes based his new combinations on the same types used by previous authors, his new combinations “P. digitatum (Nutt. ex Torr. & Gray) Grimes,” “P, hypogaeum (Nutt. ex Torr. & Gray) Rydb. var. scaposum (Gray) Grimes,” and “P. pentaphyllum (L.) Grimes” must be treated as isonyms (neither as su- perfluous names nor as later homonyms; Nicolson 1975). Additional discussion is provided below, under the respective pediomelums. Pediomelum digitatum (Nutt. ez Torr. & Gray) Isely, Sida 11:430. 1986; Grimes, Mem. New York Bot. Gard. 61:71. 1990. BASIONYM: Pso- ralea digitata Nutt. ex Torr. & Gray, 1:300. 1841. Note: Although Isely made the above combination in 1986, he (1990, p. 129) apparently overlooked his own combination and attributed it to Grimes. Isely’s attribution to Grimes must be treated as a bibliographical error. For his “new combination,” Grimes cited Pediomelum digitatum var. parvifolium (Shinners) Gandhi & L.E. Brown as a synonym. The preceding trinomial should have alerted Grimes to the fact that prior to his 1990 work, the use of the binomial Pediomelum digitatum was indeed established. Pediomelum hypogaeum (Nutt. ez Torr. & Gray) Rydb. var. scaposum (A. Gray) Mahler, Sida 12:250-251. 1987; Grimes, Mem. New York Bot. Gard. 61:76. 1990. BASIONYM: Psoralea hypogaea Nutt. ex Torr. & Gray var. scaposa A. Gray, Boston J. Nat. Hist. 6:173. 1850. 86 PHY £0 L.0:G LA volume 72(2):80-92 February 1992 Pediomelum hypogaeum var. subulatum Grimes (1990, p. 76) transferred Pediomelum subulatum (Bush) Rydb. (based on Psoralea subulata Bush) to Pediomelum hypogaeum (Nutt. ez Torr. & Gray) Rydb. at varietal rank and made the combination Pediomelum hy- pogaeum var. subulatum (Bush) Grimes. He cited Psoralea subulata var. minor Shinners as a synonym. Superficially, some workers may believe that Grimes erred in choosing the epithet subulata over the epithet mznor, since the latter existed as a varietal name, and since a name has priority only within its own rank (JCBN Art. 60.1). However, we assert that Grimes was correct in his treatment, and that any confusion of the nomenclature may be attributed to an absence of a discussion of the nomenclature of this taxon and to the in- complete citation of basionym details within Grime’s treatment. A discussion is provided here. When Shinners (Field & Lab. 19:23. 1951.) proposed Psoralea subulata var. minor, he automatically created P. subulata var. subulata (ICBN Att. 26.2). If these two varieties represent a single taxon, then the autonym (var. subulata) has priority over var. minor (ICBN Att. 57.3). Pediomelum hypogaeum (Nutt. er Torr. & Gray) Rydb. var. subulatum (Bush) Grimes, Mem. New York Bot. Gard. 61:76. 1990. BASIONYM: Pso- ralea subulata Bush (Annual Rep. Missouri Bot. Gard. 17:120. 1906.) var. subulata (automatically created by Psoralea subulata Bush var. mz- nor Shinners). Psoralea subulata Bush var. minor Shinners, Field & Lab. 19:23. 1951. Pediomelum pentaphyllum (L.) Rydb., N. Amer. Fl. 24:23. 1919, as to the type, but not as to the description; Grimes, Mem. New York Bot. Gard. 61:82. 1990. BASIONYM: Psoralea pentaphylla L., Sp. Pl. 764. 1753. Note: Since Rydberg’s description for Pediomelum pentaphyllum applied to a decumbent pediomelum (endemic to México), and since the type of its basionym (Psoralea pentaphylla) applied to the acaulescent species of Pe- diomelum found in sw United States (se Arizona, sw New Mexico, Trans-Pecos area of Texas) and adjacent México (Ockendon 1965), Grimes perhaps believed that Pediomelum pentaphyllum (L.) Rydb. was invalidly published. Grimes (pp. 82-83), therefore proposed “Pediomelum pentaphyllum (L.) Grimes.” It is unfortunate that Grimes failed to realize the following. On the trans- ference of Psoralea pentaphylla to Pediomelum by Rydberg, the resultant com- bination, Pediomelum pentaphyllum, must be retained for the species to which the type of Psoralea pentaphylla belongs, and that Pediomelum pentaphyllum Kartesz & Gandhi: Nomenclatural! notes - X. 87 must be attributed to Rydberg, even though Rydberg applied his combination erroneously to a Mexican species (JCBN Art. 55.2). Senna artemisioides The Australian legume, Cassia artemisioides Gaudichaud ez A. DC., has become established as a weed in Arizona and California (Isely 1975, p. 64). In their generic treatments of American Cassinae (Cassia L., Chamaecrista Moench, and Senna P. Mill.), Irwin & Barneby (1982) did not include this Australian legume. Based on its characteristics (e.g., fertile stamens 10 with terminal poricidal dehiscence; absence of bracteoles), we assign this legume to Senna and propose the following combination. Senna artemisioides (Gaudichaud ez A. DC.) Kartesz & Gandhi, comb. nov. BASIONYM: Cassia artemisioides Gaudichaud ez A. DC., Prodr. 2:495. 1825. TYPE: Gaudichaud (P). PEDALIACEAE Ceratotheca triloba Ceratotheca triloba, a south African native, has become naturalized in se U.S.A. This binomial has been attributed to E. Meyer (Soil Conservation Service 1982, p. 262); to E. Meyer ex Bernh. (Stapf in Dyer 1904); to Bernh. (Jackson 1895); and to (Bernh.) Hook. f. (Abels 1975, p. 199; Welman in Gibbs Russell et al. 1987, p. 187). Our analysis follows. Bernhardi (Linnaea 16:41. 1842.) proposed Sporledera as a segregate genus from Ceratotheca Endl. and proposed S. trilobaas a sp. nov. Although Sporled- era Bernh. is a later homonym of Sporledera Hampe 1837, and thus illegiti- mate, the binomial S. triloba Bernh. must be considered for purpose of priority (ICBN Art. 68.1). In its protologue, Bernhardi cited C. triloba E. Mey., a Manuscript name, as a synonym. At that stage, the correct authorship of C. triloba was: E. Mey. er Bernh., pro syn. (ICBN Rec. 50A.1, Ex. 1). In their treatment of the genus Ceratotheca, Bentham & Hooker (Gen. Pl. 2:1059. 1876.) mentioned the name S. triloba, but did not make the combination; hence, they did not validate the name C. triloba (ICBN Art. 33 Ex. 2). Ap- parently, Hooker (Bot. Mag. 114: t. 6974. 1888.) was the first to validate the combination C. triloba and he ascribed the name to E. Mey. Ceratotheca triloba (Bernh.) E. Mey. ex Hook. f., Bot. Mag. 114: t. 6974. 1888. BASIONYM: Sporledera triloba Bernh., Linnaea 16:42. 1842. 88 PHY TE ODAC BA volume 72(2):80-92 February 1992 POLEMONIACEAE Ipomopsis congesta ssp. nevadensis Cronquist (in Cronquist et al. 1984, p. 128), who recognized Gilia nevaden- sis Tidestrom at specific rank, remarked that Tidestrom’s plant appeared to be a compact expression of G. congesta Hook. ssp. palmifrons A. Brand, and deserved to be treated as an infraspecific taxon of G. congesta. Kartesz (1988, pp. 961-963) recognized the genus [pomopsis Michx. as distinct from the genus Gilia Ruiz & Pavon and treated Tidestrom’s plant at subspecific rank within Ipomopsis congesta (Hook.) V. Grant. He remarked that ssp. nevadensis repre- sented a unique expression within the J. congesta complex. Kartesz’s treatment is maintained here. Since the combination J. congesta ssp. nevadensis needs validation, it is formally proposed below. Ipomopsis congesta (Hook.) V. Grant ssp. nevadensis (Tidestrom) Kartesz & Gandhi, comb. et stat. nov. BASIONYM: Gilia nevadensis Tidestrom, Proc. Biol. Soc. Wash. 38:15. 1925. TYPE: U.S.A. Nevada: Toiyabe Natl. Forest, Bunker Hill, A. Hitchcock 865 (7). SCROPHULARIACEAE Linaria vulgaris Linaria vulgaris (toadflax), an Eurasian native, has become naturalized in temperate North America. Chater et al. (in Tutin et al. 1972, p. 232) and Dorn (1988, p. 282) attributed the name to Miller (Gard. Dict., ed. 8. Linaria no. 1. 1768), whereas Holmgren (in Cronquist et al. 1984, p. 460) attributed the name to Hill. Our analysis follows. Hill (Brit. Herb. 108. 1756) based Linaria vulgaris on Antirrhinum linaria L. However, he did not use the Linnaean system of binary nomenclature in this work. Whatever “binomials” may be found within that work are not true binomials, but mere descriptive phrases reduced to two words (see [CBN Art. 23.6c, Ex. 9). Therefore, the name L. vulgaris was not validly published in Hill’s work. Apparently, unaware of this fact, Holmgren (l.c.) erroneously used Hill as the author. To the best of our knowledge, Miller was the first to validate the name L. vulgaris. Linaria vulgaris (Hill, Brit. Herb. 108. 1756, nom. invalida]; Mill., Gard. Dict., ed. 8. Linaria no. 1. 1768. Antirrhinum linaria L., Sp. Pl. 616. 1753. Linaria hinaria (L.) Karsten, Deutsche Fl. 947. 1882 (Tautonym). Kartesz & Gandhi: Nomenclatural notes - X. 89 Penstemon crandallu ssp. taosensis We concur with Nisbet & Jackson (Univ. Kansas Sci. Bull. 41(5):726. 1960.) on their transfer of Penstemon linarioides A. Gray ssp. taosensis Keck to P. crandallu A. Nels. at infraspecific rank. However, we do not follow their varietal recognition of Keck’s taxon and reinstate its subspecific rank. Penstemon crandallii A. Nels. ssp. taosensis (Keck) Kartesz & Gandhi, comb. et stat. nov. BASIONYM: Penstemon linarioides A. Gray ssp. taosensis Keck, Bull. Torrey Bot. Club 64:373. 1937. Penstemon cran- dallit A. Nels. var. taosensis (Keck) Nisbet & R.C. Jackson, Univ. Kansas Sci. Bull. 41(5):726. 1960. TYPE: U.S.A. New Mexico: Taos Co., 30 Jul 1932, Nelson & Ruth 158 (UC). SOLANACEAE Physalis subulata var. neomezicana Waterfall (in Correll & Johnston 1970, p. 1392) proposed the combination Physalis subulata Rydb. var. neomezicana (Rydb.) Waterfall and cited the ba- sionym. Unfortunately, Waterfall failed to provide the complete bibliography of the basionym, and thus did not meet the requirements for validation of a new combination (ICBN Art. 33.2). Prior to proposing the preceding combi- nation, Waterfall (Rhodora 60:168-169. 1958.) treated Rydberg’s plant as a variety of P. foetens Poir. and lectotypified Rydberg’s binomial. We accept Waterfall’s 1970 treatment and provide the bibliography of the basionym to validate the combination. Physalis subulata Rydb. var. neomexicana (Rydb.) Waterfall ez Kartesz & Gandhi, comb. nov. BASIONYM: Physalis neomezicana Rydb., Mem. Torrey Bot. Club 4:325. 1896. Physalis foetens Poir. var. neomezicana (Rydb.) Waterfall, Rhodora 60:168. 1958. LECTOTYPE (vide Water- fall, Rhodora 60:168-169. 1958.): Fendler 678 (GH). ACKNOWLEDGMENTS The authors are grateful: to Dr. Paul A. Fryxell (TAES), Dr. Larry E. Brown (SBSC), and Dr. Dieter Wilken (CS) for helpful suggestions; and to Dr. Bryan Dutton (GH), to Ms. Ruth F. Schallert (US-Botany Library), and Ms. Tracy Delius (Biology-Forestry Library, DUKE) for literature used in this study. 90 Pe YO TWOrG, OF GareA volume 72(2):80-92 February 1992 REFERENCES Abels, J. 1975. Die Gattungen Ceratotheca Endl. und Dicerocaryum Boj. Mem. Soc. Broteriana 25:1-358. Al-Shehbaz, I.A. 1988. The genera of Arabideae (Cruciferae: Brassicaceae) in the southeastern United States. J. Arnold Arbor. 69:85-166. Correll, D.S. & M.C. Johnston. 1970. Manual of the Vascular Plants of Teras. Texas Research Foundation, Renner, Texas. Cronquist, A., et al. 1984. Intermountain Flora, Vol. 4: Subclass Asteridae (except Asteraceae). The New York Botanical Garden, Bronx, New York. Dorn, R.D. 1988. Vascular Plants of Wyoming. Mountain West Publishing, Box 1471, Cheyenne, Wyoming. Dyer, W.T. 1904. Flora Capensis, vol. 4(2). Lovell Reeve & Co., Ltd., Covent Garden, London, U.K. Gibbs Russell, G.E., et al. 1987. List of Species of Southern African Plants, ed. 2, part 2: Dicotyledons. Botanical Research Institute, Dept. of Agr. & Water Supply, South Africa. Greuter, W. (Ch. Ed. Comm.). 1988. International Code of Botanical Nomenclature. Adopted by the Fourteenth International Botanical Con- gress, Berlin, Jul-Aug 1987. Regnum Veg. 118. Grimes, J.W. 1990. A revision of the New World species of Psoraleeae (Leguminosae: Papilionoideae). Mem. New York Bot. Gard. 61:1-113. Irwin, H.S. & R.C. Barneby. 1982. The American Cassinae. Mem. New York Bot. Gard. vol. 35(1). Isely, D. 1975. Leguminosae of the United States: II. Subfamily Caesalpin- ioideae. Mem. New York Bot. Gard. vol. 25(2). . 1990. Vascular Flora of the Southeastern United States, vol. 3(2): Leguminosae (Fabaceae). The University of North Carolina Press, Chapel Hill, North Carolina. Jackson, B.D. 1895. Index Kewensis, vol. 1. Clarendon Press, Oxford, England, U.K. Kartesz & Gandhi: Nomenclatural notes - X. 91 Kartesz, J.T. 1988. A flora of Nevada. Ph.D. Dissert., University of Nevada, Reno, Nevada. . 1992. Synonymized Checklist of the Vascular Flora of the United States, Canada & Greenland, ed. 2 (in press). Timber Press, Portland, Oregon. & K.N. Gandhi. 1989. Nomenclatural notes for the North American flora - I. Phytologia 67:461-468. . 1990a. Nomenclatural notes for the North American flora - II. Phytologia 68:421-427. . 1990b. Nomenclatural notes for the North American flora - III. Phytologia 69:129-137. . 1990c. Nomenclatural notes for the North American flora IV. Phytologia 69:301-312. . 199la. Nomenclatural notes for the North American flora - V. Phytologia 70:194-208. . 1991b. Nomenclatural notes for the North American flora - VI. Phytologia 71:58-65. . 1991c. Nomenclatural notes for the North American flora VII. Phytologia 71:87-100. . 1991d. Nomenclatural notes for the North American flora VIII. Phytologia 71:269-280. . 199le. Varietal autonyms of nontypical subspecies - validly published combinations under today’s Code. Taxon 40:308-310. . 1992. Nomenclatural notes for the North American flora - IX. Phytologia 72:17-30. Mermill, E.D. & S-Y. Hu. 1949. Work and publication of Henry Muhlenberg, with special attention to unrecorded or incorrectly recorded binomials. Bartonia 25:1-66. Nicolson, D.H. 1975. Isonyms and pseudo-isonyms: identical combinations with the same type. Taxon 24:461-466. Ockendon, D.J. 1965. A taxonomic study of Psoralea subgenus Pediomelum (Leguminosae). Southw. Naturalist 10:81-124. 92 PHYTO LO GIA volume 72(2):80-92 February 1992 Palmer, E. & N. Pitman. 1972. Trees of Southern Africa, vol. 3. A.A. Balkema, Cape Town, South Africa. Soil Conservation Service. 1982. Natzonal List of Scientific Plant Names, vols. 1 & 2. U.S.D.A., SCS-TP 159. Government Printing Office, Wash- ington, DC. Stafleu, F.A. & R.S. Cowan. 1976. Tazonomic Literature, ed. 2. vol. I: A-G. Regnum Veg. vol. 94. Stafleu, F.A. (Ch. Ed. Comm.). 1978. International Code of Botanical Nomenclature. Adopted by the Twelfth International Congress, Leningrad, July 1975. Regnum Veg. vol. 97. Tutin, T.G., et al. 1972. Flora Europaea, vol. 3: Diapensiaceae to Myopo- raceae. University Press, Cambridge, U.K. Phytologia (February 1992) 72(2):93-95. CYNODON NLEMFUENSIS, (POACEAE: CHLORIDEAE) PREVIOUSLY UNREPORTED IN TEXAS ‘Stanley D. Jones & *Gretchen D. Jones 1S.M. Tracy Herbarium (TAES), Department of Rangeland Ecology and Management, Texas A&M University, College Station, Texas 77843-2126 U.S.A. *Department of Biology, Texas A&M University, College Station, Texas 77843-3258 U.S.A. ABSTRACT Cynodon nlemfuénsis var. nlemfuénsis, previously unreported in Texas has been found in Kenedy County. A key for the Texas species of Cyn- odon is included. KEY WORDS: Cynodon, Cynodon nlemfuénsis, Chlorideae, Poaceae, Texas Cynodon L. Rich., is a genus of nine species and ten varieties (Harlan et al. 1970). Clayton & Renvoize (1986) listed + eight species in the Old World, with one species being pantropical and extending into warm tem- perate regions. The species occupy inhabited, grazed, or other disturbed places. Cynodon nlemfuénsis Vanderyst var. nlemfuénsis is native to Africa from Ethiopia to Rhodesia (= Zimbabwe) and Angola (Harlan et al. 1970). Cynodon nlemfuénsis was not listed as occurring in Texas by Cory & Parks (1937), Correll & Johnston (1970), Gould (1975), Hatch et al. (1990), nor Lonard (in press). There are now three species of Cynodon reported for Texas: dactylon (L.) Pers., C. aethiopicus Clayton & Harlan, and C. nlemfuénsis var. nlemfuénsis. The following key to Texas Cynodon is modified from Clayton & Harlan (1970). KEY TO TEXAS CYNODON 1. Plants with rhizomes and stolons. ................ C. dactylon (L.) Pers. 93 94 Pap Vor Ol OG FA volume 72(2):93-95 February 1992 1. Plants without rhizomes, but with stolons. .............+.-0++--ee-- (2.) 2. (1.) Culms woody, very hard; racemes in 2-5 whorls (rarely 1), stiff, frequently red or purple. ....... C. aethiopicus Clayton & Harlan 2. Culms not woody (stolons may be woody), soft, racemes in 1 whorl (occasionally 2), flexuous, frequently green. ............+.-.-.-.. Dare ete ee eee C. nlemfuénsis Vanderyst var. nlemfuénsis Harlan et al. (1970) provide the following description of Cynodon nlemfuén- sis var. nlemfuénsis: stolons stout with long internodes; rhizomes absent; leaves linear-lanceolate, sometimes with scattered trichomes; whorls of inflo- rescence 1, rarely 2 or 3 verticillate; racemes (branches) slender, long, often somewhat flexuous, 4-7 cm long; spikelets rather widely spaced on raceme (branch), ca. 4 mm long; glumes subequal, longest ca. * the spikelet length; lemma pointed, pubescent on keel, sometimes lightly pubescent along margins. The growth habit of stout stolons and rather slender culms forms loose mats up to 60 cm high. Chromosome number 2n = 18, rarely 36. Specimen collected: UNITED STATES. Texas: Kenedy Co., 04 Novem- ber 1990, S. & G. Jones 6158 (PAUH, sj [Stanley Jones’ personal herbar- ium], TAES, jw [J.K. Wipff’s personal herbarium]). The collection site is 43.9 miles south on US 77 from its junction with Los Olmos Creek, north of Ray- mondville, Cynodon nlemfuénsis was frequently observed for ca. 5 miles south and north along US 77 from the collection site. The habitat is an open mesic coastal prairie (roadside) with the elevation being about 13 feet (4 m) above sea level. The soils are in the Nueces-Sarita series, are light gray fine sands, and are fairly well drained. They have moderately slowly and moderately rapidly permeable lower layers with a reasonable shrink swell potential. The soil is neutral to rather alkaline. The geology of the site is Wind Blown Deposits (Qs) (Recent). Associated species include Chloris gayana Kunth, Pennise- tum ciliare (L.) Link, Cynodon dactylon, Palaforia terana DC., Eriogonum multiflorum Benth., Sporobolus purpurascens (Swartz) Hamilt., S. cryptan- drus (Torr.) A. Gray, Panicum virgatum L., P. capillarioides Vasey, Eragrostis secundiflora Presl, EF. swallenii Hitchc., E. lugens Nees, Trichoneura elegans Swallen, Aristida purpurea Nutt., A. longispica Poir., Digitaria terana Hitchce., and Paspalum plicatulum Michx. Cynodon nlemfuénsis should not be expected to spread outside of region 6 or the southern part of region 1 as defined by Hatch et al. (1990) because it will not endure prolonged freezes. ACKNOWLEDGMENTS We thank Stephan L. Hatch (TAES), Robert I. Lonard (PAUH), and J.K. Wipff (TAES) for reviewing this manuscript. Jones & Jones: Cynodon nlemfuensis in Texas 95 LITERATURE CITED Clayton, W.D. & J.R. Harlan. 1970. The genus Cynodon L.C. Rich. in tropical Africa. Kew Bull. 24:185-189. Clayton, W.D. & S.A. Renvoize. 1986. Genera Graminum: Grasses of the World. Kew Bull. Additional Series XIII. Her Majesty’s Stationery Office, London, U.K. Correll, D.S. & M.C. Johnston. 1970. Manual of the Vascular Plants of Tezas. Texas Research Foundation, Renner, Texas. Cory, V.L. & H.B. Parks. 1937. Catalogue of the Flora of Tezas. ‘Texas Agric. Exper. Sta. Bull. 550. Harlan, J.R., J.M.J. deWet, W.W. Huffine, & J.R. Deakin. 1970. A Gurde to the Species of Cynodon (Gramineae). Oklahoma State Univ. Agric. Exper. Sta. Bull. B-673. Hatch, S.L., K.N. Gandhi, & L.E. Brown. 1990. Checklist of the Vascular Plants of Teras. Texas Agric. Exper. Sta. Bull. MP-1655. Lonard, R.I. (in press). A Guide to Grasses of the Lower Rio Grande Valley, Tezas. The University of Texas-Pan American Press, Edinburg, Texas. Phytologia (February 1992) 72(2):96-98. REDUCTION OF STREPTOCALYX (BROMELIACEAE: BROMELIOIDEAE) Lyman B. Smith & Michael A. Spencer Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 U.S.A. ABSTRACT Streptocalyz is reduced to Aechmea and new combinations are pro- vided where necessary. KEY WORDS: Aechmea, Streptocalyz, Bromeliaceae In Flora Neotropica Monograph No. 14 (Smith & Downs 1979), the genera Aechmea and Streptocalyr are distinguished from one another on the basis of the presence or absence, respectively, of petal appendages. More recently, petal appendages have proven unacceptable as a delimiting generic character in Bromeliaceae where groups of closely related species are segregated solely on the basis of this character. An example is Streptocalyz, which in all other respects is identical to Aechmea sensu Smith & Kress (1989). Baker (1889) was the first to comment on the artificiality of Aechmea versus Streptocalyz, and considered them “scarcely worth separating.” We agree completely with Baker and here reduce Streptocalyr to synonymy. The appropriate names in Aechmea for taxa previously treated as Streptocalyr follow. Aechmea Ruiz & Pavon, Fl. Peruv. Prodr. 47. 1793, nomen conservandum. Streptocalyr Beer, Flora 37:348. 1854. syn. nov. Aechmea arenaria (Ule) Smith & Spencer, comb. nov. BASIONYM: Sitrep- tocalyr arenarius Ule, Verh. Bot. Brandenb. 48:134. 1907. TYPE: PERU. Ule 6335 (HOLOTYPE: B; Isotype: MG). Aechmea biflora (L.B. Smith) Smith & Spencer, comb. nov. BASIONYM: Streptocalyz biflorus L.B. Smith, Phytologia 24:448, pl. 5, figs. 15, 16. 1972. TYPE: ECUADOR. Dodson & Thien 2070 (HOLOTYPE: WIS). 96 Smith & Spencer: Reduction of Streptocalyz 97 Aechmea beeriana Smith & Spencer, nom. nov. Based on: Streptocalyz poeppigit Beer, Bromel. 141. 1856; non Aechmea poeppigu Baker, Handb. Bromel. 37. 1889. TYPE: BRAZIL. Poeppig s.n. (LECTOTYPE: B). Aechmea brachystachya (Harms) Smith & Spencer, comb. nov. BA- SIONYM: Streptocalyz brachystachys Harms, Notizbl. Bot. Gart. Berlin 9:1152. 1927. TYPE: PERU. Tessmann 4645 (HOLOTYPE: B). Aechmea colombiana (L.B. Smith) Smith & Spencer, comb. nov. BA- SIONYM: Streptocalyz colombianus L.B. Smith, Caldasia [1], No. 5:9, fig. 5. 1942. TYPE: COLOMBIA. Cuatrecasas 6863 (HOLOTYPE: US; Isotype: COL). Aechmea colombiana (L.B. Smith) Smith & Spencer var. laxa (E. Gross) Smith & Spencer, comb. nov. BASIONYM: Streptocalyr colombianus L.B. Smith var. larus E. Gross, Trop. Subtrop. Pflanzenwelt 65:72, figs. 44,45. 1989. TYPE: COLOMBIA. Rauh 37422 (HOLOTYPE: HEID). Aechmea curranii (L.B. Smith) Smith & Spencer, comb. nov. BASIONYM: Streptocalyz currani L.B. Smith, Contr. Gray Herb. 95:44, pl. 11, figs. 7-9. 1931. TYPE: BRAZIL.. Curran 143 (HOLOTYPE: US). Aechmea floribunda Martius ez Schultes f. in Roemer & Schultes, Syst. 7(2): 1271. 1830. TYPE: BRAZIL. Martius s.r. (HOLOTYPE: M). Strep- tocalyz floribundus (Martius ex Schultes f.) Mez in Martius, Fl. Bras. 3(3):284. 1892. Aechmea fuerstenbergi E. Morren & Wittmack, Belg. Hortic. 29:42, pl. 2. 1879. TYPE: BRAZIL. Porte s.n. (HOLOTYPE: LG). Streptocalyz fuerstenbergiz (E. Morren & Wittmack) E. Morren, Belg. Hortic. 33:16. 1883. Aechmea geminiflora (Harms) Smith & Spencer, comb. nov. BASIONYM: Streptocalyr geminiflorus Harms, Notizbl. Bot. Gart. Berlin 12:527. 1935. TYPE: ECUADOR. Diels 1109 HOLOTYPE: B). Aechmea kentii (Luther) Smith & Spencer, comb. nov. BASIONYM: Streptocalyz kentii Luther, Selbyana 12:83, 86, fig. 14. 1992. TYPE: ECUADOR. Kent s.n. (HOLOTYPE: SEL; Isotype: QCA). Aechmea lanata (L.B. Smith) Smith & Spencer, comb. nov. BASIONYM: Streptocalyz lanatus L.B. Smith, Contr. Gray Herb. 95:45, pl. 11, figs. 4, 5. 1931. TYPE: BRAZIL. Curran 138 (HOLOTYPE: US). Aechmea longifolia (Rudge) Smith & Spencer, comb. nov. BASIONYM: Bromeha longifolia Rudge, Guyan. 1:31. pl. 49. 1805. TYPE: FRENCH 98 PELY TO 50-614 volume 72(2):96-98 February 1992 GUIANA. Martin s.n. (HOLOTYPE: BM). Streptocalyz longifolius (Rudge) Baker, Handb. Bromel. 31. 1889. Aechmea lugoi (Gilmartin & Luther) Smith & Spencer, comb. nov. BA- SIONYM: Streptocalyz lugoi Gilmartin & Luther, J. Brom. Soc. 40(5):208, fig. 8. 1990. TYPE: ECUADOR. Lugo 308 (HOLOTYPE: §; Isotype: MO). Aechmea murcae (L.B. Smith) Smith & Spencer, comb. nov. BASIONYM: Streptocalyz murcae L.B. Smith, Phytologia 7:110, pl. 1, figs. 20-24. 1960. TYPE: BRAZIL. Murca-Pires 951 (HOLOTYPE: IAN). Aechmea napoensis Smith & Spencer, nom. nov. Based on: Streptocalyz pallidus Luther, J. Brom. Soc. 30(6):256-258, fig. 1980; non Aechmea pallida L.B. Smith, Mem. New York Bot. Gard. 10:40, fig. 1964. TYPE: ECUADOR. Luther s.n. (HOLOTYPE: SEL). Aechmea poitaei (Baker) Smith & Spencer, comb. nov. BASIONYM: Streptocalyz poitaei Baker, Handb. Bromel. 31. 1889. TYPE: FRENCH GUIANA. Poiteau s.n. (HOLOTYPE: K). Aechmea seidelii (Leme) Smith & Spencer, comb. nov. BASIONYM: Streptocalyz seidelii Leme, Bradea 4(39):313. 1987. TYPE: BRAZIL. Seidel 1056 (HOLOTYPE: HB). Aechmea williamsii (L.B. Smith) Smith & Spencer, comb. nov. BASIO- NYM: Streptocalyz williamsii L.B. Smith, Contr. Gray Herb. 98:14, pl. 4, fig. 7. 1932. TYPE: PERU. Williams 2722 (HOLOTYPE: F). ACKNOWLEDGMENTS We would like to thank Harold Robinson and Stephen Smith for reviewing the manuscript. LITERATURE CITED Baker, J.G. 1889. The Handbook of the Bromeliaceae. George Bell and Sons, London, U.K. Smith, Lyman B. & R.J. Downs. 1979. Bromelioideae (Bromeliaceae). Flora Neotropica Monograph No. 14(3), New York Botanical Garden, New York. Smith, L.B. & W.J. Kress. 1989. New or restored genera of Bromeliaceae. Phytologia 66(1):70-79. Phytologia (February 1992) 72(2):99-102. STUDIES ON THE GENUS BI/DENS L. (COMPOSITAE) FROM THE EASTERN HEMISPHERE. 2. NEW COMBINATIONS AND A NEW SUBSPECIES FOR THE FLORA OF TROPICAL EAST AFRICA AREA T.G.J. Rayner Department of Botany, Plant Science Laboratories, University of Reading, Whiteknights, P.O. Box 221, Reading, RG6 2AS, UNITED KINGDOM ABSTRACT Four new combinations and a new subspecies are provided for the forthcoming account of the genus Bidens for the Flora of Tropical East Africa: Bidens scopulorum comb. nov. (= Coreopsis scopulorum), Bidens elgonensis ssp. morotonensis comb, et stat. nov. (= Coreop- sis morotonensis), Bidens elgonensis ssp. cheranganiensis ssp. nov., Bidens odora comb. nov. (= Coreopsis odora) and Bidens acuti- caulis var. filirostris comb. nov. (= Bidens paupercula var. filirostris). A lectotype is selected for the latter name. KEY WORDS: Bidens, Coreopsis, Compositae, taxonomy, Africa Bidens scopulorum (Sherff) T.G.J. Rayner, comb. nov. BASIONYM: Coreopsis scopulorum Sherff, Bot. Gaz. (Crawfordsville) 88:302. 1929. TYPE: KENYA. K1, Northern Frontier Province, summit of Mt. Gar- guez, 26 Aug. 1911, E. Heller s.n. (U.S. Nat. Herb. No. 634308] (HOLO- TY Piss US): Bidens scopulorum, known only from the type locality in central Kenya, is a member of a group of closely related species each with a restricted distribution occurring at altitude and endemic to eastern Africa. It is apparently most closely related to B. chippii(M.B. Moss) Mesfin Tadesse and B. isostigmatoides Sherff, both found only in the Imatong Mts. of southern Sudan. From these two species, B. scopulorum may be distinguished by its capitula ca. 1 cm wide at anthesis (2-4 cm in B. chippii and B. isostigmatoides), outer phyllaries 8 or fewer (not to 20+), inner phyllaries ca. 3 mm long (not to 7 mm), paleae apices obtuse or rounded (not acute), and cypselas setose and biaristate (not nude and exaristate). 99 100 Pay TOL O GLA volume 72(2):99-102 February 1992 Bidens elgonensis (Sherff) Agnew ssp. morotonensis (Sherff) T.G.J. Ray- ner, comb. et stat. nov. BASIONYM: Coreopsis morotonensis Sherff, Amer. J. Bot. 34:157, f. 5. 1947. TYPE: UGANDA. UY. Northern Province, Karamoja, Mt. Moroto, Jun. 1942, [.R. Dale U-261 (HOLO- TYPE: EA; Isotype: K). Bidens morotonensis (Sherff) Agnew, Upland Kenya Wild Flowers. 466. 1974. Another member of this group of species is Bidens elgonensis from east- ern Uganda and central-western Kenya. This species, originally published by Sherff (1925) as Coreopsis elgonensis, was based on the single specimen of R.A. Dummer 3304 at K collected on Mt. Elgon. Later, (Sherff 1947) he described a closely related species, C. morotonensis, from the nearby Mt. Moroto. This he distinguished from C. elgonensis by its not exserted capitula (exserted in C. elgonensis), peduncles 1-2 cm (not 2-9 cm), leaves 2.0-3.5 cm long (not 1.0-1.8 cm), etc. Subsequent collections have maintained these vegetative dis- tinctions and have added a further character, namely specimens collected on Mt. Elgon have at least some internodes 2 cm or more long towards the apices of the flowering branches, whereas specimens collected on Mt. Moroto usu- ally have internodes 1.5 cm or less. Although specimens from these two areas show constant vegetative differences, characters of the inflorescence are uni- form throughout the range of the species. Therefore, because differences in characters of the capitula are mostly employed to distinguish between species in African Bidens, the distinctions between these two variants are not here considered worthy of specific delimitation. For this reason I have decided to reduce C. morotonensis to the level of subspecies within B. elgonensis. Bidens elgonensis (Sheriff) Agnew ssp. cheranganiensis T.G.J. Rayner, ssp. nov. TYPE: KENYA. K3, Rift Valley Province, Cherangani, east slope of Kaisungor, Dec. 1971, D.R. Tweedie 4192 (HOLOTYPE: k). Subspecies haec a Bidente elgonensi (Sherff) Agnew subspecie elgonensi et Bidente elgonensi subspecie morotonensi differt ca- pitulis exsertis, quidem foliis aliquibus duobus centimetris longis plerumque multo longioribus, necnon quidem internodiis aliquibus versus apices caulium florentium plus quam duobus centimetris longis. Bidens elgonensis also occurs in the Cherangani Hills, a large dissected plateau to the east of Mt. Elgon. In this area its distribution is much more widespread and vegetative characters show a greater degree of variation within and between specimens. Populations in this area, however, are distinct from those growing on Mt. Elgon and Mt. Moroto. The internodes toward the apices of the flowering branches are mostly over 2 cm long and the capitula Rayner: Studies on Bidens from Eastern Hemisphere 101 are exserted on long peduncles, thus resembling B. elgonensis ssp. elgonensis. The leaves, however, resemble those of B. elgonensis ssp. morotonensis, but may be as much as 5 cm long. The leaf segments also tend to be much wider. Bidens odora (Sherff) T.G.J. Rayner, comb. nov. BASIONYM: Coreopsis odora Sherff, Kew Bull. 11:445. 1957. TYPE: TANZANIA. T7, South- ern Highlands Province, Mdapo, Mar. 1954, S.R. Semsez 1655 (HOLO- TYPE: EA; Isotype: K). Bidens odorais restricted to the Njombe district of southern Tanzania, oc- curring in montane grassland between about 2150 and 2300 m. It is perhaps most closely allied to B. elgonensis but may be distinguished from this species by its bipinnatisect leaves with segments to 1 mm wide (tripartite with seg- ments 2-13 mm wide in B. elgonensis), capitula several to many in terminal corymbs (not solitary or few at branch apices) and cypselas biaristate and narrowly winged (not exaristate and unwinged). Bidens acuticaulis Sherff var. filirostris (P. Taylor) T.G.J. Rayner, comb. nov. BASIONYM: Bidens paupercula Sherff var. filirostris P. Taylor, Hooker’s Icon. Pl. 36:t. 3580. 1962. TYPE: TANZANIA. T8, South- ern Province, Songea District, ca. 140 km east of Songea, 4 Jun. 1956, E.W.B.H. Milne-Redhead & P.G. Taylor 10547(LECTOTYPE (here se- lected]: K; Isolectotypes: B,BR,EA,K(2 sheets], LISC,P,S,SRGH). A comparison of the type collections of Bidens acuticaulis (1915) and B. paupercula (1923) has shown them to be virtually identical. The key distinction used by Sherff (1937) to separate these two species, namely “rami acutissime angulati et fere subalati” for B. acuticaulis as opposed to “rami tantum mod- erate angulati vel subtetragoni” for B. paupercula, is wholly unreliable even as a character to distinguish between specimens determined by Sherff himself. Clearly, therefore, B. paupercula should be considered a synonym of B. acuti- caulis. With the uniting of these two names it becomes necessary to rename B. paupercula var. filirostris. This variety is mainly distinguished from the type variety by the enormously elongated cypselas which may be up to 26 cm long. Taylor (1962) indicated that the holotype of var. filirostris was at K, but here there are three sheets of the type collection Milne-Redhead & Taylor 10547 bearing parts of at least five plants. As none of these bear any indication by Taylor that they were intended to be the holotype, I do not feel that he has complied with the definition of a holotype as provided by Article 7.3 of the International Code of Botanical Nomenclature (Greuter et al. 1988). Therefore I consider it is necessary to select a lectotype from among the sheets of the 102 PHWe TO: E30cG A volume 72(2):99-102 February 1992 type collection at K. Of these “Sheet 2” is here chosen as the lectotype. In addition to having mature fruit this sheet, unlike the others, also possesses well preserved leaves. BIBLIOGRAPHY Greuter, W., H.M. Burdet, W.G. Chaloner, V. Demoulin, R. Grolle, D.L. Hawksworth, D.H. Nicolson, P.C. Silva, F.A. Stafleu, E.G. Voss, & J. McNeill. 1988. International Code of Botanical Nomenclature, adopted by the Fourteenth International Botanical Congress, Berlin, July-August 1987. Regnum Veg. 118:328 pp. Sherff, E.E. 1915. Studies in the genus Bidens. II. Bot. Gaz. (Crawfordsville) 59:301-316. Sherff, E.E. 1923. Studies in the genus Bidens. VI. Bot. Gaz. (Craw- fordsville) 76:144-166. Sherff, E.E. 1925. New or otherwise noteworthy Compositae. II. Bot. Gaz. (Crawfordsville) 80:367-389. Sherff, E.E. 1937. The genus Bidens. Field Mus. Nat. Hist., Bot. Ser. 16:1- 709. Sherff, E.E. 1947. New or otherwise noteworthy Compositae. X. Amer. J. Bot. 34:138-158. Taylor, P. 1962. Bidens paupercula Sherff. Hooker’s Icon. Pl. 36:t. 3580. Phytologia (February 1992) 72(2):103-104. A NEW SPECIES OF COREOPSIS (ASTERACEAE) FROM SOUTHERN MEXICO B.L. Turner Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT A new species, Coreopsis oaxacensis B. Turner, is described from southeasternmost Puebla and closely adjacent Oaxaca. It belongs to the sect. Pseudo-Agarista where it relates to C. rhyacophylla, from which it differs in being a smaller plant, with smaller, more pubescent leaves and smaller heads with shorter rays. KEY WORDS: Coreopsis, Asteraceae, México Routine identification of Mexican Asteraceae has revealed the following novelty. Coreopsis oaxacensis B. Turner, sp. nov. TYPE: MEXICO. Oaxaca: Mpio. de Tamazulapan, Cerro Pericon, 3 km N of San Pedro Nopala on road to Yosocuno, 2600 m, 4 Nov 1991, J.L. Panero, P. Davila, and P. Tenorio 2606 (HOLOTYPE: MEXU; Isotype: TEX). Coreopsi rhyacophilae Greenm. similis sed foliis multo minoribus (2-5 cm longis vs. 6-12 cm) pariter puberuli-hispidulis (vs. glabris vel fere glabris) et capitulis minoribus corollis radii brevioribus (8- 10 mm longis vs. 12-20 mm). Suffruticose perennial herbs or shrublets to 60 cm high. Leaves semileath- ery, opposite, sheathing at the very base, mostly 2-6 cm long, 1-3 cm wide, mostly pinnately dissected with 3-5 primary divisions, the ultimate divisions usually shortly trilobed; petioles mostly 3-10 mm long; blades ovate to ovate deltoid in outline, nearly glabrous to conspicuously and evenly hispidulous- puberulous at first but glabrate with age. Heads 2-6, terminal, the ultimate peduncles 1-3 cm long, sparsely puberulous. Involucres turbocampanulate, 6-8 103 104 PHY TO LOvGHA volume 72(2):103-104 February 1992 mm high, 10-12 mm wide (pressed), the outer bracts (calyculus) 8-11, linear- oblanceolate, 3-5 mm long, the inner bracts elliptic to elliptic oblanceolate, 6-7 mm long, 1.5-2.5 mm wide, sparsely puberulent. Receptacle plane to some- what convex, glabrous or nearly so, the bracts linear-lanceolate, striate, hirsute below. Ray florets 8-11, neuter, the ligules yellow, striate, mostly 8-12 mm long, 2-6 mm wide. Disk florets 30-60 per head, the corollas glabrous, yellow, 4-5 mm long, the tube ca. 2 mm long, the throat ca. 2 mm long. Achenes ca. 4 mm high, 1.5 mm wide, markedly ciliate along the margins and along the midline on the adaxial side, otherwise glabrous; pappus of two persistent hispidulous-ciliate awns 4-5 mm long. ADDITIONAL SPECIMEN EXAMINED: MEXICO. Oaxaca: Mpio. Zapoquila, Portezuelo Majada Grande, Cerro Yolotepec al S de Membrillos (18° 03’ N, 97° 32’ W), ca. 2400 m, 9 Nov 1986, Pedro Tenorio L. 12391 (TEX,MEXU). Puebla: Mpio. Caltepec, Cerro Chicamole, al E de Membril- los (18° 08’ N, 97° 34’ W), ca. 2400 m, 8 Nov 1986, Tenorio L. 12371 (TEX). Coreopsis oaracensis belongs to the sect. Pseudo-Agarista (sensu Crawford 1976) where it relates to C. rhyacophila Greenm. It differs from the latter in being a smaller plant (30-60 cm high vs. ca. 100 cm), having markedly smaller leaves and smaller heads with smaller rays, as noted in the Latin diagnosis. Ac- cording to its collectors, Coreopsis oazacensis is apparently abundant locally. Type material differs from the remaining collections in having somewhat larger heads and nearly glabrous foliage, otherwise the plants differ but little. ACKNOWLEDGMENTS 1 am grateful to Guy Nesom for the Latin diagnosis and to him and Carol Todzia for reviewing the manuscript. José Panero kindly provided his excellent collections to serve as type material. LITERATURE CITED Crawford, D.J. 1976. Taxonomy of Coreopsis sect. Pseudo-Agarista (Com- positae) in Mexico with additional comments on sectional relationships in Mexican Coreopsis. Brittonia 28:329-336. Phytologia (February 1992) 72(2):105-108. A NEW SPECIES OF PEDICULARIS (SCROPHULARIACEAE) FROM WESTERN MEXICO Guy L. Nesom Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT Pedicularis chihuahuensis sp. nov. is described from southern Chihuahua, México. Based on similarities in floral morphology, it is most closely related to the Mexican species P. mezicana and P. orizabae but differs from both in having tuberous roots, villous vestiture, foliar teeth with aristate apices, and merely toothed (unlobed) floral bracts. The new species is allopatric with both of its putatively close relatives. KEY WORDS: Pedicularis, Scrophulariaceae, México A taxonomic overview of the Mexican species of Pedicularis with pinnately parted leaves was presented by McVaugh & Mellichamp (1975), and another such species was described by McVaugh & Koptur (1978). Yet another previ- ously unknown one, has been collected in southern Chihuahua and is described here. Pedicularis chihuahuensis Nesom, sp. nov. TYPE: MEXICO. Chi- huahua: Mpio. Guadalupe y Calvo, 55.2 km SW of El Vergel on road between Parral and Guadalupe y Calvo, N side of town of Catedral in valley with moist meadows surrounded by dry hills with pines, fenced area protected from grazing, ca. 2400 m, 23 Aug 1983, G. Nesom 4756 (HOLOTYPE: TEX!; Isotype: MEXU!). Pediculari mezicanae Zucc. ez Benth. ac P. orizabae Cham. & Schlecht. similis morphologia florali sed a speciebus ambabus differt radicibus tuberantibus, vestimento villoso, et bracteis floralibus marginibus nonlobatis. 105 106 PRIN OT ZO SE OIG TA. voluine 72(2):105-108 ‘ebruary 1992 Perennial herbs arising from thick, fleshy roots, sometimes apparently tap- rooted or often the main axis of the root with 1-few thick branches; stems single from the base, unbranched, erect, 60-75 cm tall, loosely villous with spreading to deflexed, jointed, white vitreous hairs. Leaves cauline (the basal mostly absent by flowering), 4-10 along the stem, lanceolate to oblanceolate in outline, slightly villous along the veins, once pinnately parted, with 23-30 toothed to lobed segments, the teeth aristate, the petiole and rachis barely if at all winged, lower leaves 15-25 cm long, 1.5-5.0 mm wide (at the widest point), the uppermost greatly reduced in size. Racemes (flowering) 12-15 cm long, (fruiting) 20-35 cm long, 30-40 flowered, with sessile or nearly sessile flowers, the floral bracts ascending, 10-14 mm long, with ciliate, nearly en- tire to shallowly toothed margins, spatulate with an abruptly narrowed and linear-attenuate apex 3-7 mm long. Calyces campanulate, the tube 4.5-5.0 mim long, with 5 nearly equal lobes ca. 4 mm long. Corollas 20-22 mm long, “pink and yellow,” bilabiate, the lower (abaxial) lip 16-18 mm long, the 3 lobes suborbicular, the terminal lobe 5 mm wide, the lateral lobes ca. 6 mm wide, the upper (adaxial) lip (the galea) 12-14 mm long, smoothly incurved from the adaxial side, without a beak, not prolonged beyond the curve. Stamens 4, didynamous. Style arcuate, the stigma slightly exserted at anthesis. Capsules 9-15 mm long, broadly ovoid, slightly raised on a broad stipe 1-2 mm long; seeds blackish, ca. 4 mm long, with reticulate surfaces. Additional collections examined: MEXICO. Chihuahua: Mpio. Guadalupe y Calvo: along stream in meadow, San Juan, Sierra Chinatu, 8900 ft. (2700 m), flrs. pink and yellow, 8 Oct 1959, D.S. Correll & H.S. Gentry 22927 (LL); along stream (tributary of Rio Soldado) in gorge of conifer forest, near La Rocha, NE slope of Sierra Mohinora, 7500 ft (2270 m), 14 Oct 1959, D.S. Correll & H.S. Gentry 23060 (LL). Pedicularis chihuahuensis apparently is endemic to the vicinity of Cerro Mohinora and Mt. Chinatu in southern Chihuahua, a relatively small area of high elevation known to harbor numerous other local endemics in a number of different families. The three collections were made within a radius of about 25 kilometers. The only other species of Pedicularis that occurs in the same area is P. angustifolia Benth., which has linear, undivided leaves and loose, few flowered inflorescences. McVaugh & Mellichamp (1975) observed that the Mexican species of Pedic- ularis with pinnately parted leaves are not particularly closely related among themselves but divisable instead into three groups, based on floral morphol- ogy: (1) the galea prolonged into a linear, recurved-ascending beak (P. glabra McVaugh & Mell.; P. jonesii Brandeg. [type MO!] also belongs here); (2) the galea downwardly curved, short beaked or merely enlarged and pointed on the lower side (P. mexicana Zucc. ex Benth. and P. orizabae Cham. & Schlecht., and P. gordon McVaugh & Koptur also apparently belongs here); and (3) the galea clavate, rounded at apex, neither beaked, curved, nor enlarged to Nesom: New Pedicularis from western México 107 a point on one side (P. tripinnata Mart. & Gal. and P. hintonit McVaugh & Mell.). Among these, P. chthuahuensis clearly is most similar to P. orizabae and P. mezicana in its short curved galea and to P. orizabae with its unbeaked galea and the lobes of its lower corolla lip shorter than the galea. Pedicularis orizabae is known from high peaks in the states of México, Veracruz, and Oax- aca; P. mexicana is more widespread, occurring from Puebla to Michoacan and central Durango. The new species differs from both in its vestiture, root morphology, foliar teeth apices, and floral bract morphology, but it is com- pared in the following couplet to P. mezicana, because the latter more closely approaches the geographic range of P. chihuahuensis than does P. orizabae. 1. Roots numerous, thin fibrous; stems, leaves, and floral axes glabrous; fo- liar teeth with callose-dentate apices; floral bracts trilobed; lower corolla lip as long as the galea, the galea abruptly incurved nearly at a right angle, narrowed beyond the curve into a short, truncate beak; Tlaxcala, Hidalgo, Puebla, Morelos, México, Distrito Federal, Michoacan, appar- ently disjunct to central Durango. 22.0. .2.0. omen ek rene P. mexicana 1. Roots mostly with 1-3 tuberous thickened primary branches; stems, leaves, and floral axes villous; foliar teeth with aristate apices; floral bracts without lateral lobes; lower corolla lip shorter than the galea, the galea smoothly incurved from the adaxial side, without a beak; southern Chi- PPberUR Mes ee Peacsneeoe Acre cin helo BT cI nisih |= Oe ee Ie P. chihuahuensis Apart from Mexican species that are putatively closely related to it, Pedic- ularis chihuahuensis also is similar and probably closely related to P. procera A. Gray, which occurs in the United States from Colorado to northern New Mexico and adjacent Arizona. Pedicularis proceraalso produces a villous vesti- ture, aristate foliar teeth, and usually unlobed floral bracts, but differs in its thin fibrous roots, much taller plants (8-18 dm tall), longer and denser inflo- rescences, longer flowers (27-34 mm long), and lanceolate, longer floral bracts (25-35 mm long) that are coarsely villous with stipitate glandular hairs on the surfaces and margins. ACKNOWLEDGMENTS I thank Dr. B.L. Turner and Dr. T.P. Ramamoorthy for their review and comments on the manuscript. 108 PHY £0.05 OyGiA volume 72(2):105-108 February 1992 LITERATURE CITED McVaugh, R. & T.L. Mellichamp. 1975. Mexican species of Pedicularis (Scrophulariaceae) hitherto confused with P. tripinnata Mart. & Gal. Contr. Univ. Michigan Herb. 11:57-63. McVaugh, R. & S. Koptur. 1978. A new species of Pedicularis from Jalisco, Mexico. Contr. Univ. Michigan Herb. 11:298-300. Phytologia (February 1992) 72(2):109-114. TWO NEW SPECIES OF VERBESINA (ASTERACEAE) FROM SOUTHERN MEXICO B.L. Turner Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT Two new species of Verbesina are described from México: V. fayii B. Turner from Guerrero and V. kimii B. Turner from Veracruz. The former relates to the V. virgata complex of central México. The rela- tionship of V. kimi is moot since it is a small tree with white rayed flowers and has a columnar receptacle, characters which do not readily position it among the known sections of the genus. KEY WORDS: Asteraceae, Verbesina, México Routine identification of Mexican Asteraceae has revealed the following novelties. Verbesina fayii B. Turner, sp. nov. TYPE: MEXICO. Oaxaca: “Steep, moist, brushy, granitic slopes in deciduous forest region, ca. 23 km N of Putla, and 65 km SW of Tlaxiaco”, ca. 1300 m, 30 Oct 1970, A. Cronquist & J. Fay 10856 (HOLOTYPE: TEX!; Isotypes: NY,US!). Verbesinae virgatae Cav. similis sed foliis elliptici-lanceolatis ad extrema duo pariter gradatim descrescentibus, ac paginis inferis aequaliter molliter pubescentibus valde venosisque differt. Slender shrubs 1-3 m high, stems without wings, densely short pilose. Leaves alternate, markedly venose, mostly 8-12 cm long, 2.5-4.5 cm wide; petioles 1-10 mm long; blades elliptic to elliptic lanceolate, equally taper- ing at both ends, pinnately nervate, pilose above and below, more so below with somewhat appressed soft hairs, especially along the veins, the margins minutely serrulate to entire. Heads numerous and terminal in broad leafy cymose panicles. Involucres campanulate to somewhat hemispheric, 3-5 mm 109 110 PHY TOLOGYTA volume 72(2):109-114 February 1992 high, 6-9 mm wide (pressed), the bracts 2-3 seriate, subgraduate to subequal. Receptacle conical, ca. as high as wide, the bracts linear oblanceolate with acute recurved apices. Ray florets pistillate, fertile, mostly 8-13, the ligules yellow to yellowish orange, 5-10 mm long, 2-4 mm wide. Disk florets numer- ous (50-100), the corollas yellow, 3-4 mm long, the tube pubescent, ca. 1 mm long, the throat mostly glabrous, ca. 2 mm long, the lobes ca. 0.3 mm long. Anthers yellow, the appendages ovate. Achenes oblanceolate, ca. 2 mm long, 0.8 mm wide, scarcely winged, if at all, appressed pubescent, the pappus of two, readily deciduous awns 1.0-1.5 mm long. ADDITIONAL SPECIMENS EXAMINED: MEXICO. Guerrero: 11 mi W of Chilpancingo, ca. 6000 ft, 21 Oct 1963, Cronquist 9711 (TEX,US). Oaxaca: 47 mi S of Sola de Vega, ca. 5700 ft, 7 Nov 1965, Cronquist & Sousa 10510 (GH,TEX,US); 16 km Oeste de Sola de Vega, 2080 m, 22 Nov 1977, Delgado S. et al. 630 (TEX); 56 km S of Tlaxiaco, growing in crevices and on vertical cliffs, 2000 m, 9 Nov 1990, Panero 2105 (TEX); “Cerro Tres Cruces” al S de el Limén el cual esta a 11.1 km al SW del entronque carr. Tehuantepec - Oaxaca - Buenos Aires, Distr. de Tehuantepec, 9 Dec 1983, Torres C. 4290 (TEX). Verbesina fay is closely related to the widespread, highly variable, V. virgata complex (Fig. 2) but is readily distinguished from among those by its leaves that are broader, elliptic-lanceolate, equally tapering at both ends, and more markedly venose and densely soft pubescent beneath. The species is named for Dr. John Fay, co-collector of the type material and esteemed monographir of the mostly Mexican genus Perymenium. Verbesina kiinii B. Turner, sp. nov. Fig. 1. TYPE: MEXICO. Veracruz: 3 km S of Cd. Mendoza on road to Tehuacan, mountain summit near the Puebla- Veracruz border, steep limestone slopes, disturbed oak forest (ca. 18° 45’ N, 97° 18’ W), ca. 3000 m, 13 Jan 1987, Ki-Joong Kim 10062 (HOLOTYPE: TEX!; Isotype: MEXU). Verbesinae chilapanae B. Turner similis flosculis discii numero- sis flavisque et receptaculis conicis sed differt habitu (arbor parva usque ad 5 m alta vs. frutex demissus 0.5-1.0 m altus), capitulis multo minoribus numerosioribusque (30+ vs. 5-7), et floribus radii ligulis albis (vs. flavis). Trees to 5 m high. Stems thick, pithy, winged, velutinous-tomentose. Leaves alternate, 15-20 cm long, 6-10 cm wide; petioles 3-6 cm long, winged throughout; blades ovate, pinnately nervate, scabrous above, densely veluti- nous beneath, markedly venose, the margins irregularly serrate with ragged shallow lobes. Heads radiate, numerous in congested terminal cymose panicles, Pe Two new Verbesina from southern México Turner: from holotype. Verbesina kimi, Eig. 1. 112 Pra YY T'O'LO'G-RA volume 72(2):109-114 February 1992 VERBESINA cinerascens culminicola cymbipalea liebmannii macvaughii oreopola platanara Ppterocarpha virgata o Of ee Op gap Fig. 2. Distribution of Verbesina virgata and closely related taxa. —s \rea occupied by V.favi is shown by diagonal lines. Turner: Two new Verbesina from southern México Tals’ the ultimate peduncles densely pilose, mostly 3-10 mm long. Involucres turbo- campanulate, 4-6 mm high, the bracts 2-3 seriate, somewhat graduate to sube- qual, the outer series pilose, oblanceolate. Receptacles, at maturity, columnar, 4-5 mm high, ca. 0.8 mm wide, the bracts numerous, folded, oblanceolate, pubescent, persistent, the apices erect, acute. Ray florets 13-32, pistillate, fertile; ligules white, 2-4 mm long, ca. 1 mm wide, 3-5 nervate. Disk florets numerous; corollas white (?), ca. 3 mm long, pubescent throughout, the tube ca. 0.8 mm long, the throat ca. 2 mm long, the lobes 0.2-0.4 mm long. An- thers dark (blackish), the appendages narrowly lanceolate, 2-3 times as long as wide. Style branches glabrous or nearly so, the appendages gradually attenu- ate. Achenes broadly winged, the peripheral achenes 3 sided and 3 winged, the more central achenes flattened and with 2 wings, 4-5 mm long, ca. 3 mm wide, the wings corky, ca. 0.8 mm wide, the pappus of two or rarely 3 persistent awns 1-2 mm long. Because of its white rays, large alternate leaves and treelike habit, Verbesina kimi apparently appears to belong to the section Ochractinia, where it has no close relatives. In Blake’s (1926) treatment of the trees and shrubs of México, the species will key to the common and widespread V. turbacensis H.B.K., which it superficially resembles. In Olsen’s (1985) synopsis of section Ochrac- tinza, the material concerned will key to the widespread, largely herbaceous, V. microptera DC., a common species of the Gulf coastal montane regions of México. Verbesina kimzi differs from both of these taxa in having a markedly columnar receptacle with numerous florets. Indeed, it is possible that V. kimi is more closely related to species of the section Verbesinaria (sensu Robinson & Greenman 1899) than it is to section Ochractinia, largely because it has numerous disc florets on a pronounced, columnar or conical receptacle, which is found in certain species of the section Verbesinaria (e.g., V. oazacana DC. and V. chilapana). It is a pleasure to name this extraordinary species for its only collector, Dr. Ki-Joong Kim, an extraordinary doctoral student of mine, a superb synan- therologist and Korean national who monographed Krigia for his doctoral dis- sertation and is currently deeply involved in DNA analyses of the Asteraceae with Dr. Robert Jansen at the University of Texas, Austin. ACKNOWLEDGMENTS I am grateful to Guy Nesom for the Latin diagnoses and to him and T.P. Ramamoorthy for reviewing the manuscript. Nancy Webber provided the illustration. I am grateful to J. Panero for providing information relating to the relationships of Verbesina kimi. 114 PHY POLO GPA volume 72(2):109-114 February 1992 LITERATURE CITED Blake, S.F. 1926. Verbesinain Trees and Shrubs of Mezico. Contr. U.S. Natl. Herb. 23:1572-1586. Olsen, J. 1985. Synopsis of Verbesina sect. Ochractinia (Asteraceae). PI. Syst. Evol. 149:47-63. Robinson, B. & J. Greenman. 1899. Synopsis of the genus Verbesina. Proc. Amer. Acad. Arts 34:534-566. Phytologia (February 1992) 72(2):115-126. NEW SPECIES OF WEDELIA (ASTERACEAE, HELIANTHEAE) FROM MEXICO AND CRITICAL ASSESSMENT OF PREVIOUSLY DESCRIBED TAXA B.L. Turner Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT Two new species, Wedelia pimana B. Turner and W. talpana B. Turner, occurring in Chihuahua and Jalisco, respectively, are de- scribed as new. In addition, justification is given for the acceptance of W. gentry: B. Turner, and W. cronquisti B. Turner, both of which were treated as synonyms of W. greenmanii B. Turner, in Strother’s (1991) treatment of the genus for North America; evidence is also pre- sented for the submergence of W. tegetts Strother in the earlier W. mezicana (Schultz-Bip.) McVaugh. When appropriate, distributional maps of these several taxa are provided. KEY WORDS: Wedelia, Asteraceae, México Appearance of the carefully constructed revisional study of the North American species of Wedelia and related groups by Strother (1991) has oc- casioned the present study, largely because the nomenclature proposed by him has impinged upon my treatment of the Mexican species (Turner & Nesom, in prep.). Strother recognized seventeen species as occurring in México (exclud- ing W. trilobata L., which he included in the newly proposed genus Complaya Strother). In addition, he recognized seven infraspecific taxa in his concept of the widespread highly variable W. acapulcensis Kunth complex, all of these oc- curring in México. Strother, in his basically conservative treatment, also called to the fore one or more set of specimens which he thought might ultimately prove deserving of specific status. In my own appraisal of Wedelia in North America (largely based upon the study of material at ARIZ,ASU,CAS,F,GH,LL,MEXU,MICH,TEX, and WIS (including most of the types for the names concerned) I would recognize 23 specific taxa as occurring in México (excluding the probably introduced W. trilobata), two of these described in the present contribution. Additionally, 115 116 PHY PT OLbsOrGH A volume 72(2):115-126 February 1992 while I agree with many of Strother’s taxonomic conclusions, I cannot accept his broadly inclusive treatment of Wedelia acapulcensis, and yet other taxa. Because of this, I place on record here my reasons for rejecting some of the taxonomy proposed by him, and the nomenclature that might accompany such rejection. Wedelia pimana B. Turner, sp. nov. TYPE: MEXICO. Chihuahua: “small grassy area above a small waterfall not more than 100 m from the center of town [Nabogame] ... a somewhat anomalous area in the vicinity, sunlit with a stream flowing right through the middle towards the water- fall. The parent rock material in that area was a purplish-red shale which generally made for richer soil than the siliceous rhyolite on the east side of the valley.”, 1800 m, 3 Aug 1987, J.E. Laferriére 646 (HOLOTYPE: TEX; Isotype: ARIZ). The quoted information is taken from a letter to the present author from the collector, dated 5 Dec 1991. Wedeliae scabrae (Cav.) B. Turner similis sed floribus radii neu- teris et foliis valde laceratis differt. Suffruticose erect herbs to 70 cm high. Stems reddish, pubescent through out with coarse spreading hairs to 1 mm long, these interspersed among a finer hispidulum of much shorter hairs, some of them uncinate or apically recurved. Leaves opposite, scarcely reduced upwards, mostly 5-7 cm long, 1.5-2.0 cm wide; petioles 2-4 mm long, pubescent like the stems; blades ovate to ovate lanceolate in outline, principal nerves 3, pubescent beneath like the stems, but the surface also subglandular, or glandular, the margins markedly lacerate or sublobate. Heads arranged 3-5 in terminal cymes, the ultimate peduncles mostly 2-6 cm long. Involucres campanulate, ca. 1 cm high, 0.8-1.0 cm wide, the bracts linear-lanceolate, subequal, spreading hirsute with stiff white hairs. Receptacle convex, the pales linear-lanceolate, 8-10 mm long, readily detached, the single dorsal nerve purple, the apices acute. Ray florets ca. 8, neuter; corol- las yellow, the tubes glabrous, ca. 3 mm long, the ligules 5-8 mm long, ca. 2 mm wide. Disk florets ca. 35, the corollas yellow, 6-7 mm long, glabrous, the tubes ca. 2.5 mm long, the lobes narrowly acute, ca. 1.5 mm long. Achenes ca. 5 mm long, 2.2-2.5 mm wide, wingless, moderately appressed pubescent, the pappus of 2 principal awns 1-3 mm long, between these a shorter, lacerate, crown of scales, the base with a distinct elaiosome. Strother (1991) called attention to the present taxon, noting, “I thought that this odd collection [646] might represent a hybrid between the two { Wedelia chthuahuana and W. greenmani]. I found pollen stainability ... to be 98% for a single floret from Laferriére. Perhaps the specimen will prove referable to a distinct taxon.” Turner: New and noteworthy Wedelia from México 1 LG Additional study, and correspondence with Laferriére (cited, in part, above and below) has convinced the present worker that the plant concerned repre- sents a good species, presumably a localized endemic with distant relationships to the widespread Wedelia scabra (sensu Turner 1988), to judge from the vesti- ture, but perhaps equally close to W. chihuahuana. I assigned the collection concerned to the latter taxon initially, but W. pimana differs in habit, vestiture and details of the disk florets (having glabrous lanceolate lobes vs. hispidulous and deltoid). I had intended to name this taxon for its only known collector, Dr. Joseph E. Laferriére, who has made a remarkable series of collections of the region concerned in connection with an ethnobotanical study of the Mountain Pima of Chihuahua. However, the collector deigned have his name eponymized, suggesting instead the name pimana, which is adopted here. Wedelia talpana B. Turner, sp. nov. Fig. 1. TYPE: MEXICO. Jalisco: Mpio. Talpa, km 9 of the road to Minas del Cuale, 1050 m, 7 Mar 1992, J.L. Panero, Lidia Cabrera, & A. Campos 2858 (HOLOTYPE: MEXU; Isotype: TEX). Wedeliae rosei (Greenm.) McVaugh similis sed foliis minoribus (plerumque 2-3 cm longis vs. 3-8 cm) marginibus involutis ac pagi- nis infernis brevi-hispidulis et capitulis plerumque 2-3 subsessilibus (in pedunculis 2-10 mm longis) in aggregatis differt. Brittle stemmed low shrubs or shrublets 20-30 cm high. Stems densely hispid-pilose. Leaves opposite, mostly 20-30 mm long, 3-8 mm wide; petioles 1-2 mm long; blades narrowly lanceolate to ovate elliptic, harshly pubescent above and below with erect or ascending hairs, the surface itself somewhat glandular-furfuraceous, the margins mostly enrolled and seemingly entire. Heads campanulate, mostly arranged in aggregations of 2-3 nearly sessile heads, when single, the peduncles 2-10 mm long. Involucres mostly 7-11 mm high, the bracts subgraduate to subequal in 2-3 series, the outer series herbaceous and pilose, the inner series scarious, ciliate, and with purple striations. Receptacle convex, the bracts lanceolate with purple midribs. Ray florets 5-8, neuter, sterile; tubes glabrous, ca. 2 mm long; ligules yellow, 5-7 mm long. Disk flo- rets ca. 30; corollas ca. 5 mm long, yellow, glabrous or nearly so, the tubes ca. 1.5 mm long, the limbs 3.0-3.5 mm long, the lobes ca. 0.7 mm long, sparsely hispidulous. Anther sacs black, the appendages yellow. Achenes obovoid, very plump, 3.5-4.0 mm long, ca. 2 mm wide, the surface black and finely striate, rather evenly, but moderately pilose throughout, the base with a well devel- oped elaiasome; pappus a crown of short fimbriate scales arising from a boss or neck ca. 0.4 mm high. February 1992 volume 72(2):115-126 PHYTO NO Gla 118 Wedelia talpana (from Diaz L. 20971, TEX). roe le Turner: New and noteworthy Wedelia from México 119 ADDITIONAL SPECIMENS EXAMINED: MEXICO. Jalisco: Mpio. de Cabo Corrientes, 3-10 km generally east on the road to Mina del Cuale, from the junction 5 km NW of El Tuito, 850-1150 m, 16-19 Feb 1975, Mc Vaugh 26367 (MICH); Mpio. Talpa, km 18 along the road from El] Tuito to La Mina de Zimapan, 1850 m, 14 Oct 1989, Diaz L. & Lomeli S. 20971 (TEX); between Cumbre del Tejamanil and Cuale, 1790 m, 6 Mar 1971, R. Gonzales T. 125 (ENCB,MICH). Strother (1991) called attention to two of the above cited specimens, not- ing that “On further study, supplemented by additional collections, these may prove to represent a distinct species.” With the collection of additional spec- imens from the area concerned (including the type), which possess all of the characters called to the fore in the above description, I have no hesitation in making formal its recognition. According to J. Panero (pers. comm.), who collected the type, the species was relatively common in the area concerned. Wedelia aggregata (Greenm.) B. Turner In my transfer of this taxon from the genus Aspilia (Turner 1988), not having examined its type, I accepted McVaugh’s (1984) viewpoint that Wedelia aggregata was but an aberrant form of the well known, widely distributed, W. roset. So treated, the correct name for the taxon would be that of the earlier name, W. aggregata. Having now examined type material (GH!) of the latter, I concur with Strother’s interpretation: W. aggregata appears to be a good morphological species known only from type material. Wedelha gentry: B. Turner Strother placed this name in synonymy within his concept of Wedelza green- mani B. Turner. I have equivocated in my acceptance of these two taxa (pers. comm. to Strother), largely because at the time of my study Strother had most of the types, and material relating to these, under purview. | have now ex- amined the entire complex and conclude that two good taxa are involved: W. gentry: and W. greenmanit. The latter is readily distinguished from the for- mer by its very minute, closely appressed, strigose hairs, these occurring in a rather uniform fashion on the foliage, stems and involucral bracts. Wedelia gentryz possesses uniformly larger and longer, erect or ascending, hairs on all of its vegetative parts and usually has broader midstem leaves with blades rather clearly trinervate below (vs. mostly uninervate or pinnately nervate, not clearly trinervate below). So far as known plants referable to Wedelia greenmani and W. gentry: have not been found to occur together, nor have I seen clear intermediates between these. That they are sympatric or nearly so, cannot be doubted, as shown in 120 PHY TOL O GIA volume 72(2):115-126 February 1992 SON CHI COA ~ Cd. Durango a fy ZAC WEDEIELIA CS A aperegata O ceronquistii a gonvulecziarum Vi ieeaypi 3 @ meaicunum ce) O rusci Wb otalpana WO Gov Aly Fig. 2. Distribution of Wedelia spp with respect to Seemann's 1849-50 trip to the “Sierra Madre". Turner: New and noteworthy Wedelia from México 121 Fig. 3. Indeed; since both W. greenmanu and W. gentryz occur in relatively close proximity in the area of Nabogame, Chihuahua (ca. 28° 30’ N, 108° 30! W), both collected by Dr. J.E. Laferriere while engaged in an ethnobotanical study of the region concerned (W. gentryi 1101 (ARIZ,TEX]; W. greenmani 1186, 1950 |ARIZ,TEX]), I requested that the collector assess their niche re- lationships, if any, within this region (including that of W. pzmana, quoted in the above account). He replied as follows: Collection #646 [W. pimana| was from a small grassy area above a small waterfall not more than 100 m from the center of town. It was a somewhat anomalous area in the vicinity, sunlit with a stream flowing right thru the middle toward the waterfall. The parent rock material in that area was a purplish-red shale which generally made for richer soil than the siliceous rhyolite on the east side of the valley. #1101|W. gentryz| was from a very different area about a mile east, in a riparian pine/oak/ Cupressus arizonica forest in the rhyolite area. #1186 and #1187|W. greenmanii| were from a dry granitic site about half a mile west of town. That area was composed of many small but steep hills, large boulders, and small cliffs; “the maze” I used to call it because it was often difficult to weave my way thru the broken terrain. I made two separate collections of the Wedelia there because the first had reddish stems, the second grayish-brown stems, a slight difference, but in the field it can be hard to tell what is significant ...| Hence to answer your question about habitats, the three areas were quite different habitats, both in soil and in dominant vege- tation. Another difference is that the 646 site was grazed only in the winter, the others all year round. Given the limited number of collections, however, it can be difficult to say whether these ob- servations would hold up if someone went down to make a more detailed survey. Since the original description of Wedelia gentry: (Turner 1988), known then only by the type (collected in Sierra Surotato, Sinaloa), | have examined the above mentioned collection of Laferriére, that of Breedlove & Thorne 18403 (MICH), and Pennington 92 (TEX), all of these examined by Strother and cited as W. greenmanii. In short, my examination of most of the material examined by Strother that relates to this complex has reaffirmed my original conviction that two taxa are involved, which I opt to treat as species, there being little suggestion, if any, that the characters which mark them intergrade. Finally, it should be noted that Strother also placed Wedelia cronquistu B. Turner in synonymy under his concept of W. greenmanit. ‘This is discussed in more detail, as follows. 122 PHY TOLO.GiA volume 72(2):115-126 February 1992 os 100, 0 ' nn6 WEDELIA 20 O chihuahuana | @® gentryi O greenmanii | A pimana Fig. 3. Distribution of Wedelia spp. _ Turner: New and noteworthy Wedelia from México 123 Wedelia cronquistu B. Turner This taxon is known only by the type, collected just east of Guadalajara, Jalisco (Fig. 4). Strother contends (p. 77) that the type differs “from other specimens of Wedelia greenmanii primarily in the greater density, but not in the quality, of the foliar indument of the plants from the disjunct population in Jalisco.” Actually, the leaves, and vestiture on undersurfaces of these, dif- fer markedly from those of both W. greenmanzi (which, as noted above, has a very minute appressed pubescence) and W. gentry: (which has scattered, longer, nonappressed, hispid hairs, these occurring on relatively broad, triner- vate, midstem leaves), having narrow, highly reticulate venose, undersurfaces and a vestiture of closely strigose or closely packed short or suberect coni- cal hairs. In fact, one might more certainly position W. cronquistu within an expanded W. rose than within W. greenmanii, for its habital aspects are more like the former and, of course, such a position would make better bio- geographical sense; indeed, I suspect that more intensive study will show that the population referable to W. cronquistw is closer to both W. rose and W. gonzaleziarum B. Turner (from southernmost Durango) sensu Strother, than it is to the morphologically and geographically more remote W. greenmanii. In short, I intend to retain W. cronguisti, believing this to be a localized en- demic in the area concerned, much as appears to be the case for W. aggregata, W. gonzaleziarum, W. pimana, and W. talpana, all localized and seemingly distinct species. Wedelia mezicana (Schultz-Bip.) McVaugh Strother recognized this taxon, but believed it to be known only by the type, collected somewhere between Mazatlan, Sinaloa and Cd. Durango, Du- tango, or possibly between the latter and northern Nayarit, areas traversed by Seemann, who collected in this region from 23 Nov 1849 to 22 Feb 1850 (cf. stippled lines, Fig. 2). Strother distinguished Wedelia mezicana from his newly described W. tegetis Strother by internode length (mostly 2-5 cm long in W’. mezicana vs. 0.5-1.5 cm) and shape of leaf blades (mostly deltate to ovate, rounded to subtruncate at base in W. mezicana vs. elliptic to lanceo- late, cuneate at the base). I do not find these compelling differences; indeed, a photograph of the type collection housed at K (GH!) shows that a range of leaf shapes varying from deltoid to elliptic lanceolate occurs among the several shoots attached to that sheet; in addition, there is much variation in internode length. Besides, nearly all of the species of Wedelza are notoriously variable as regards leaf shape and internode length, as any cursory examination of speci- mens of this or that species will show. In short, I suspect that Strother’s W. tegetis is the same as W. mezicana. Indeed, it is likely that Seemann collected the type of W. mexicana in the same general area from which the type of 124 PHY T.O-L0:GI4 WEDELIA @ aggregata A cronquistii O rosei A talpana | volume 72(2):115-126 February 1992 Fig. 4. Distribution of Wedelia spp. Turner: New and noteworthy Wedelia from México 125 W. tegetis was obtained, for his journey to Cd. Durango took him directly through the route (along present day Mexican Highway 40) along which the type locality of W. teyetis occurs. Numerous recent workers have collected along this route and forms referable to both W. tegetzs and W. mezicana are invariably collected along the Sinaloa-Durango border (Concordia, Sinaloa to El Espinazo del Diablo, Durango). Following his arrival in Cd. Durango, See- mann left the city on 2 Jan 1850, taking a southerly route through Mezquital, Durango, then hence to Santa Maria (northernmost Nayarit) from whence he returned to Cd. Durango via a more westerly route (through La Guajolota, Durango). He returned hastily to Mazatlan over the same route (along present day Highway 40) he had traversed earlier, arriving in that port city on 22 Feb 1850. The important part of the above, which I have taken from Seemann’s own account of his travels as given in the Voyage of the Herald, is that the only place W. tegetis (or W. mexicana) is known to occur in the region traversed by Seemann, is along the Sinaloa-Durango border (Fig. 2). The route south of Cd. Durango to Santa Maria, Nayarit, has now been fairly well collected and only W. rosei, W. gonzaleziarum, or W. grayi McVaugh has been obtained near or along this route. Strother, however, cites two white rayed plants (MICH!) of W. tegetis from the area near Jesus Maria, Nayarit (Fig. 3) but I would include both of these, somewhat aberrant collections, in my concept of W. grayi, which is distinguished from W. mezicana by its white rays and longer, more oblanceolate, leaf blades. The above discourse relating to Wedelia tegetis may be more academic than need be, for Strother (p. 83) wisely notes that “As more specimens become available ... the type of one or more of the names W. gray1, W. hintoniorum B. Turner, and W. tegetis may prove to be conspecific with that of W. mez- icana.” My only quibble here is, that based on the material available at the present time, it is almost certain that W. tegetis is the same as W. mezicana. ACKNOWLEDGMENTS I am grateful to Guy Nesom for the Latin diagnoses and to him and T.P. Ramamoorthy for reviewing the manuscript. Special thanks are due Dr. J.E. Laferriere for providing me with information relating to the occurrence of Wedelia greenmaniu, W. gentryz, and W. pimana in the area of Nabogame, Chihuahua. Dr. José Panero kindly provided material from his collections (after my preliminary study) for the typification of W. talpana. 126 PHY ..O.L0 GLA volume 72(2):115-126 February 1992 LITERATURE CITED McVaugh, R. 1984. Compositae. In Flora Novo-Galiciana, ed. W.R. Ander- son, 12:1-1157. University of Michigan Press, Ann Arbor, Michigan. Strother, J.L. 1991. Taxonomy of Complaya, Elaphandra, Iogeton, Jefea, Wa- malchitamia, Wedelia, Zermenia, and Zyzyria (Compositae-Heliantheae- Ecliptinae). Syst. Bot. Monogr. 33:1-111. Turner, B.L. 1988. New species names and combinations in Wedelza (Aster- aceae — Heliantheae). Phytologia 65:348-358. Phytologia (February 1992) 72(2):127-129. TWO NEW SPECIES OF STEVIA (ASTERACEAE, EUPATORIEAE) FROM MEXICO B.L. Turner Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT Stevia hintoniorum B. Turner, a species from near Iturbide, Nuevo Leon, México and S. pratheri B. Turner, from Chiapas, are described as new. The former is related to the locally sympatric S. incognita Grashoff, the latter to S. tomentosa H.B.K. KEY WORDS: Asteraceae, Eupatorieae, Stevia, México Routine identification of Mexican Asteraceae has revealed the following novelties. Stevia hintoniorum B. Turner, sp. nov. TYPE: MEXICO. Nuevo Leon: Mpio. Galeana, above Agua Blanca, “oak and pine woods on limestone,” 2390 m, 25 Oct 1991, G.B. Hinton et al. 21611 (HOLOTYPE: TEX). Steviae incognitae Grashoff similis sed foliis majoribus ellipticis vel oblanceolatis fere glabrisque et bracteis involucro sin glandulis stipitatis differt. Erect or sprawling herbs 40-80 cm high. Stems tawny to purplish, puberu- lent. Leaves sparsely puberulent to subglabrous, opposite throughout, at mid- stem mostly shorter than the internodes, elliptic ovate to oblanceolate, mostly 4-9 cm long, 2.0-4.5 cm wide; petioles mostly 2-10 mm long; blades trinervate from above the base, glandular punctate on lower surfaces, the margins ser- rate. Heads sessile and numerous in congested terminal corymbs, the branches below the heads decidedly glandular pubescent. Involucres 7-9 mm high, sub- glabrous or sparsely puberulent, sometimes atomiferous glandular (7.e., with sessile amber colored resinous glands) but not at all glandular pubescent, the bracts linear-lanceolate with acute apices. Corollas purple, 7-8 mm long, the 127 128 PHY TOLOGIA volume 72(2):127-129 February 1992 tube and throat both sparsely pilose and atomiferous glandular, the lobes more so. Achenes ca. 5 mm long, sparsely hispidulous, the pappus a fimbriate crown 0.3-0.8 mm high. ADDITIONAL SPECIMENS EXAMINED: MEXICO. Nuevo Leon: Mpio. Galeana: above Agua Blanca, 2050 m, 25 Oct 1991, Hinton et al. 21626(TEX); 10.6 km W of Ejido La Purisima, ca. 2100 m, 27 Oct 1982, Sundberg 1922 (TEX). Mpio. Iturbide: Iturbide to La Purisima, 2070 m, 25 Oct 1991, Hin- ton et al. 21664 (TEX); 3.3 km W of Ejido La Purisima, ca. 1600 m, 27 Oct 1982, Sundberg 1897 (TEX). In Grashoff’s (1972) account of the North American stevias, the above cited specimens key with ease to Stevia clinipodioides Greenm. The latter, however, is a taxon of southcentral and southern México (type from Serrania de Ajusco, Distrito Federal, ca. 3000 m) and is a slender rhizomatous perennial to 35 cm high with oblanceolate, conduplicate, leaves 1.5-3.5 cm long and 0.5-1.2 cm wide. As indicated in the description, 5. hintoniorum is a robust herb of lower elevations, being 40-80 cm high and having much larger nonconduplicate leaves. Overall, 5. hintoniorum appears most closely related to the poorly defined, but widespread, S. incognita Grashoff. It differs from the latter in having eglandular involucral bracts and mostly elliptic, nearly glabrous leaves. Stevia incognita occurs in Nuevo Leoén at higher elevations (e.g., on Cerro Potosi and Pena Nevada) than does S. hintoniorum but occasional hybrids between S. incognita and S. hintoniorum presumably occur, to judge from seemingly intermediate plants from the lower slopes of Pena Nevada (Ayers 504 [TEX]). Stevia pratheri B. Turner, sp. nov. TYPE: MEXICO. Chiapas: Mpio. Ixtapa, 4.2 km S of Ixtapa along route 195, steep dry slope with Junzpe- rus, Acacia, and Pinus, ca. 1150 m, “Rare in full sun on dry rocky banks of creek,” 8 Jan 1992, Alan Prather 1144 (with J. Soule) (HOLOTYPE: TEX; Isotype: MEXU). Steviae tomentosae Kunth similis sed differt foliis omnino op- positis et capitulis in pedunculis primariis nudiusculia (usque ad 20 cm longis foliis deminutia paribus tantum 2-3). Perennial herb 30-40 cm high. Stems white tomentose. Leaves opposite throughout, bicolored, mostly 2.5-3.5 cm long, 6-12 mm wide; petioles 8-12 mm long; blades ovate, trinervate, sparsely tomentose above, the lower surfaces densely white felty tomentose, the margins crenulodentate. Heads arranged in an obpyramidal corymbose terminal cluster, these borne on a rather naked primary peduncle ca. 20 cm long, the latter having only 3-4 pairs of reduced leaves, the ultimate peduncles tomentose, 1-3 mm long. Involucres mostly 5-7 mm long, the bracts linear-lanceolate, tomentulose, the apices acute. Florets Turner: Two new Stevia from México 129 5 per head, the corollas white, ca. 5 mm long, the lobes ca. 0.8 mm long, sparsely pubescent and atomiferous glandular on the outer surfaces. Achenes black, somewhat tangentially compressed, sparsely hispidulous, the pappus of 4 of the achenes a minute fimbriate crown ca. 0.3 mm high, the remaining achene with a pair of weak bristles 1-2 mm long, in addition to the fimbriate crown. Because of its markedly bicolored leaves and head clusters borne upon rather naked primary peduncles, this is a very distinctive species. The some- what flattened achenes with weakly developed pappus bristles also distinguish the taxon. It will not key in Grashoff’s treatment of the genus, but it might be compared with Stevia tomentosa, the latter readily distinguished by having stems about equally leafy throughout, the upper leaves mostly alternate, and the pappus bristles more numerous and stouter, up to 6 mm long. According to Grashoff (1972), Stevia tomentosa is distributed from south- ern Puebla to southern Coahuila, mostly on the Central Plateau. Stevia pratherz is apparently restricted to Chiapas, there being nothing like it from Oaxaca or Guatemala (Grashoff 1976). It is a pleasure to name this species for its primary collector, Mr. Alan Prather, doctoral student at the University of Texas working on a monograph of the genus Cobaea (Polemoniaceae). ACKNOWLEDGMENTS I am grateful to Guy Nesom for the Latin diagnoses and to him and Carol Todzia for reviewing the manuscript. LITERATURE CITED Grashoff, J.L. 1972. A systematic study of the North and Central American species of Stevia. Doctoral Dissertation. The Univ. of Texas, Austin. 609 pp. —_——. 1976. Stevia, in Flora of Guatemala. Fieldiana 24 (pt. 12):115-128. Phytologia (February 1992) 72(2):130-138. FLORISTICS OF A SANDSTONE GLADE IN WESTERN LOUISIANA M.H. MacRoberts & B.R. MacRoberts Bog Research, 740 Columbia, Shreveport, Louisiana 71104 U.S.A. ABSTRACT The floristics, species richness, geology, and edaphic conditions of a western Louisiana sandstone glade are described. KEY WORDS: Sandstone glade, Kisatchie National Forest, floris- tics, Louisiana INTRODUCTION The eastern United States is basically a forested region. Within this region, there are occasional permanent forest openings — bogs, savannahs, glades, barrens, balds, prairies— often of small size (e.g., Greller 1988; DeSelm 1986, 1990). These openings are almost always the result of unusual edaphic, geo- logic, and hydrologic factors. Sandstone glades are one such open community in the longleaf pine region of western Louisiana. As Martin & Smith (1991) have described them, these are usually ridge tops underlain by Catahoula sandstone with a thin layer of acidic silty clay loam. Glades are open with exposed sandstone boulders and ledges. There is no true overstory or midstory but scattered patches of trees and shrubs with a grass dominated, low diversity herbaceous layer. Ever since the Louisiana Natural Heritage Program began rating communities, this one has received rankings from critically imperiled to rare or uncommon (Craig et al. 1987; Smith 1988; Martin & Smith 1991). Within Louisiana, the community is restricted to a small part of the western section of the state and little is known about it; in fact, the entire previous literature on this and related communities in the western gulf coastal plain can be found in Smith 1988; Bridges & Orzell 1989; Orzell 1990; Martin et al. 1990, and Martin & Smith 1991. Sandstone glades of the type we are describing here are not to be confused with calcareous prairies or other prairielike communities, often 130 MacRoberts & MacRoberts: Western Louisiana sandstone glade floristics 131 referred to as barrens, glades, or prairies, occurring in this area (Bridges & Orzell 1989; Marietta & Nixon 1984; Smith et al. 1989; George & Nixon 1990). The purpose of this paper is to describe the floristics and species richness (number of species) of one glade. Additionally, we will make some observations on biotic, edaphic, and climatic conditions of glades. STUDY SITE AND METHODS Sandstone glades are locally common in the Kisatchie Ranger District of the Kisatchie National Forest, Natchitoches Parish, Louisiana. “Flat Glade,” the one we studied, is located about 11 km WSW of Cloutierville, Louisiana, on the eastern edge of T6N R7W S12. Its exact location is given by Martin & Smith (1991:288-291). It is approximately 100 meters above sea level. While the specific fire history of Flat Glade is uncertain, it is embedded in the pyrogenic longleaf pine community and thus probably has burned with regularity in the past (Smith 1991). It did not burn the year prior to this study. We visited Flat Glade every two weeks between March and December 1991 to sample vascular plants. We follow MacRoberts (1984, 1989) for botani- cal nomenclature. We measured species richness in three randomly selected permanent one meter square plots, pine seedling success (number of seedlings surviving to November) in two permanent four meter square plots, and pine seedling numbers in ten randomly selected nonpermanent one meter square plots. We ran a transect through Flat Glade in which we counted number of trees and measured diameter at breast height (dbh) of each tree. The transect was 213 meters long and three meters wide, and ran through the longest part of the glade. Soil samples were taken from the upper 15 cm of the glade and were analyzed by A. & L. Agricultural Laboratories, Memphis, Tennessee. We made increment borings of “relic” (predate the massive cutover of the early 1900’s) longleaf pine trees to assess age and counted rings in a number of smaller trees. We looked at many other glades in the area to familiarize ourselves with variations in the community type. RESULTS A profile of Flat Glade is shown in Figure 1. The glade measures about 1.9 hectares and is larger than most glades in the area. It is a mesalike outcrop Tising in two tiers of about seven meters each above the surrounding upland longleaf pine forest. The transition from glade to longleaf forest is not usually abrupt. Pines are scattered over Flat Glade and are stunted and dwarfed. The largest are not over twelve meters tall, are flat topped, and have sparse crowns. Canopy cover is about ten percent overall. A few oaks occur in Flat Glade. 132 POY TOLOGTA volume 72(2):130-138 February 1992 0) 50 100 | a meters Figure 1. Profile of Flat Glade MacRoberts & MacRoberts: Western Louisiana sandstone glade floristics 133 These are also stunted, gnarled, and scarcely rise above the few shrubs, Ilex vomitoria Ait. and Vaccinium spp. There were 25 trees 1.5 meters tall or taller in the 639 square meter transect: 15 longleaf pine, 9 loblolly pine, and one oak. This translates into one tree per 26 square meters. These trees had a mean dbh of 11.32 cm (range 2.4-38.0 cm, s.d. 9.1 cm). In these measurements of trees, Flat Glade lies between hillside seepage bogs and longleaf pine forest in the area. Hillside seepage bogs have a more open habitat while longleaf pine forest is much less open (MacRoberts & MacRoberts 1990). Why are glades open? Possible answers are poor seed production, poor seed germination, high seed predation, or high seedling mortality. While we have been studying this and will report on it more fully elsewhere, some data collected at Flat Glade are germane to this paper. To begin with, all pine species occurring on Flat Glade produce abundant cones. This is evident from the cones seen on trees and those on the ground from previous years. Also seeds readily sprout. On July 6, 1991 in ten one meter square plots randomly selected on the top of Flat Glade, we found two longleaf pine and nineteen loblolly pine seedlings. From these observations and from our general survey of the glade, there were undoubtedly thousands of seedlings on the 1.9 ha area. In the two four meter square plots described above, we counted 26 and 27 loblolly pine seedlings in May. By November there were ten alive in one plot and none in the other. All of the dead seedlings appeared to have succumbed to dessication and to scorching as they were without shade. And this occurred even though it was one of the wettest years in recorded history but by no means the hottest (pers. comm., National Weather Service). The seedlings that were still alive showed some signs of stress (brown lower needles). In many of the glades we visited, we noticed that trees had fallen, exposing shallow roots. The causes for this appeared to be shallow depth of the impermeable sandstone layer resulting in shallow rooting and susceptibility to windthrow. In order to examine growth rate, we cut four small longleaf pines (diameter 4.5-7.0 cm at ground level) and made increment borings in some of the largest longleaf pines present (dbh 29-38 cm) and compared these with trees from the surrounding longleaf pine habitat. We found that longleaf pines grow slowly in Flat Glade. In the four small trees, average ring width was slightly less than 1 mm (11 rings to 1 cm), and in the larger trees it was comparable but an average was harder to determine because the rings were sometimes so packed that counting was difficult. Nonetheless, the growth rate of trees in Flat Glade was about one-third that of trees in local longleaf pine forest. All of the large trees at Flat Glade appear to be “old growth” being easily in excess of one hundred years and probably much older. Soil information on Flat Glade is given in Table 1. Thre= samples were taken: two from the top and one from the middle mesa. The soil is highly acidic and essentially infertile. The surface geology is resistant beds of Catahoula 134 Pat Onl Ove A volume 72(2):130-138 February 1992 TABLE 1. Soil Characteristics Exchangeable Ions (ppm) Sample | pH | P| K | Ca K-1 4.9/4] 65 | 900 K-2 4.8 | 4] 44 | 53 K-3 4:5 |.0.| 127.) 730 sandstone overlain with Kisatchie soil — dark to light grey silty loam (fine montmorillonitic, thermic Typic Hapludalfs) (Martin et al. 1990:104). Part of Flat Glade is severely eroded, forming irregular sandstone studded gullies and minigorges of exposed soil. Although the soil is always thin and shallow, on the stable areas prairielike vegetation dominates, while on the more sparse and rocky areas where the soil is either very thin or absent, lichens and mosses are commonly the major ground cover. Plant roots are limited by the rock below. The soil moisture conditions of glades range from saturated and sticky after rains especially during winter and spring, to dry cracked and hard during summer and fall following droughts. Water runoff is fairly rapid. The year of this study, 1991, was one of the wettest in recorded history with about 170 cm of precipitation (43 cm above normal) (pers. comm., Natchitoches Station, National Weather Service), and only occasionally did Flat Glade dry out. In most years it would have been dry for periods of at least two to three weeks and possibly longer during the hottest months, July, August, and September. In July and August soil surface temperatures become very high, there being little or no shade. Additional climatic data for the area are given in Martin et al. 1990. The vascular plants found in Flat Glade are listed in Table 2. We recorded 63 taxa representing 46 genera and 26 families. The dominant species were ray- less goldenrod (Bigelowia nuttallii Anderson) and little bluestem (Schizachyrt- um scoparium [Michx.] Nash). The dominant families were grasses and com- posites, accounting for a third of the taxa. The three one meter plots had 9, 9, and 11 species, which is about half the richness of comparable plots in local hillside seepage bogs (MacRoberts & MacRoberts 1991). Herbaceous plant ground cover was not always 100 percent, with many bare areas of rock or soil often with numerous lichens and mosses. Flat Glade, like other glades we have observed, clearly has plants associ- ated with other communities, notably upland longleaf pine forest and sand woodlands (Martin & Smith 1991). These communities tend to be dry. Quite spectacularly, about ten percent of the species characterize hillside seepage bogs: Aletris, Drosera, Pingutcula, Platanthera, Sabatia gentianoides, and As- clepias longifolia Michx. (MacRoberts & MacRoberts 1991). In the glades, these plants take advantage of the open, seasonally saturated soils caused by MacRoberts & MacRoberts: Western Louisiana sandstone glade floristics 135 underlying sandstone “catchments” and ephemeral seepage areas. Pingutcula is notorious in bogs for blooming for about nine months a year. In Flat Glade, Pinguicula bloomed in late spring and early summer and again in December. As mentioned in the introduction, glades are quite variable from one to an- other. During the course of the year we visited many glades which ranged from prairielike with many grasses to bare sandstone outcrops almost totally devoid of vegetation. Interesting species not found at Flat Glade but common at other glades include: Eryngium yuccifolium Michx., Fimbristylis puberula (Michx.) Vahl., Hedyotis nigricans (Lam.) Fosberg, Lechea tenutfolia Michx., Selaginella arenicola Underw. ssp. riddellii (Van Eselt.) Tryon, Silphium laciniatum L., and Talinum parviflorum Nutt. ex Torrey & Gray. DISCUSSION/SUMMARY A major factor controlling plant community structure and composition of glades is stress. The thin, nutrient poor acidic loam soils underlain by impermeable sandstone characterized by high erosion, quick alternation of wet and very dry periods, occasional burning, and high summer temperatures, combine to produce a unique community — dwarfed and sparse, slow growing woody vegetation and hydric, mesic, and xeric herbaceous plants in relatively low numbers with large expanses of bare, or lichen or moss covered ground. While we have carefully investigated only one example of a glade community, we have looked at many of them and recognize that these communities differ in many ways undoubtedly depending on variations in soil, hydrology, and geology. Some have no hydric plants; others have many. Some have no trees because of a lack of soil, while others have many. Clearly, therefore, the habitat needs further investigation. ACKNOWLEDGMENTS We acknowledge the staff of the Kisatchie National Forest for their help and cooperation in this, as in all of our work on the National Forest. We especially thank Karen Belanger, Wildlife Biologist, and Tom Fair, District Ranger, Kisatchie Ranger District. Latimore Smith and Nelwyn McInnis, Louisiana Natural Heritage Program, made many helpful comments on earlier drafts of this paper. 136 Pali ft ©: LOG 1A volume 72(2):130-138 February 1992 Table 2. Taxa present in Flat Glade. AMARYLLIDACEAE — Aypozis rigida Chapm. CYPERACEAE — Rhynchospora globularis (Chapm.) Small, R. plumosa Ell., Sclerta ciliata Michx. JUNCACEAE — Juncus marginatus Rostk. LILIACEAE — Aletris aurea Walt., A. farinosa L., Alum canadense L. ORCHIDACEAE — Platanthera nivea (Nutt.) Luer, Spiranthes tuberosa Raf., S. vernalis Engelm. & Gray. PINACEAE — Pinus echinata P. Mill., P. palustris P. Mill., P. taeda L. POACEAE — Andropogon elliottii Chapm., A. ternarius Michx., Anthaenantia villosa (Michx.) Beauv., Aristida virgata Trin., Dichanthelium aciculare (Desv. ex Poir.) Gould & Clark, D. sphaerocarpon (Ell.) Gould, Schizachyrium scoparium (Michx.) Nash, Schizachyrium tenerum Nees, Sporobolus junceus (Michx.) Kunth. ASCLEPIADACEAE — Asclepias longifolia Michx., A. obovata Ell. ASTERACEAE — Aster dumosus L., A. linaritfolius L., Bigelowia nuttallu Anderson, Eupatorium leucolepis (DC.) Torrey & Gray, Helianthus angustifolius L., Heleastrum hemisphericum ( Alex.) Shinners, Heterotheca graminifolia (Michx.) Shinners, Liatris pycnostachya Michx., ZL. squarrosa (L.) Michx., Senecio tomentosus Michx., Solidago nitida Torrey & Gray, Vernonia terana (A. Gray) Small. AQUIFOLIACEAE — Ilex vomitoria Ait. CAMPANULACEAE — Lobelia sp. CLUSIACEAE — Hypericum gentianoides (L.) B.S.P., H. hypericoides (L.) Crantz. DROSERACEAE — Drosera brevifolia Pursh. ERICACEAE — Vaccinium arboreum Marsh, V. corymbosum L., V. stamineum L. EUPHORBIACEAE — Euphorbia corollata L. FABACEAE — Stylosanthes biflora (L.) B.S.P., Tephrosia virginiana (L.) Pers. FAGACEAE — Quercus falcata Michx., Q. marilandica Muenchh., Q. stellata Wang. GENTIANACEAE — Sabatia gentianoides Ell. LENTIBULARIACEAE — Pinguicula pumila Michx. LINACEAE — Linum medium (Planch.) Britt. LOGANIACEAE — Gelsemium sempervirens (L.) St. Hill. MYRICACEAE — Myrica cerifera L. POLYGALACEAE — Polygala mariana P. Mill., P. nana (Michx.) DC. RUBIACEAE — Diodia teres Walt. SCROPHULARIACEAE — Agalinus obtusifolia Raf., Aureolaria pectinata (Nutt.) Penn., Penstemon tubaeflorus Nutt. VIOLACEAE — Viola pedata L. MacRoberts & MacRoberts: Western Louisiana sandstone glade floristics 137 LITERATURE CITED Bridges, E.L. & S.L. Orzell. 1989. Longleaf pine communities of the west gulf coastal plain. Natural Areas Journal 9:246-262. Craig, N., L.M. Smith, N. Gilmore, G.D. Lester, & A.M. Williams. 1987. The natural communities of coastal Louisiana: classification and description. Louisiana Natural Heritage Program, Department of Wildlife and Fish- eries, Baton Rouge, Louisiana. DeSelm, H.R. 1986. Natural forest openings on uplands in the eastern United States. Pp. 366-375 In D.L. Kulhavy & D.N. Conner, eds., Wilderness and Natural Areas in the Eastern United States. A Management Chal- lenge. Stephen F. Austin State Univ., Nacogdoches, Texas. DeSelm, H.R. 1990. Flora and vegetation of some barrens of the eastern highland rim of Tennessee. Castanea 55:187-206. George, R.J. & E.S. Nixon. 1990. The herbaceous flora of three Weches formation outcrops in eastern Texas. Sida 14:117-127. Greller, A.M. 1988. Deciduous Forests. Jn M.G. Barbour & W.D. Billings, eds., North American Terrestrial Vegetation. Pp. 288-316. Cambridge University Press, New York. MacRoberts, B.R. & M.H. MacRoberts. 1991. Floristics of three bogs in western Louisiana. Phytologia 70:135-141. MacRoberts, D.T. 1984. The vascular plants of Louisiana. Bull. Museum of Life Sciences, No. 6. Louisiana State University, Shreveport, Louisiana. MacRoberts, D.T. 1989. A Documented Checklist and Atlas of the Vascular Flora of Louisiana. Bull. Museum of Life Sciences, Nos. 7-9. Louisiana State University, Shreveport, Louisiana. MacRoberts, M.H. & B.R. MacRoberts. 1990. Size distribution and density of trees in bogs and pine woodlands of west central Louisiana. Phytologia 68:428-434. Marietta, K.L. & E.L. Nixon. 1984. Vegetation of an open, prairie-like community in eastern Texas. Texas Journal of Science 36:25-32. Martin, D.L. & L.M. Smith. 1991. A survey and description of the natural plant communities of the Kisatchie National Forest, Winn and Kisatchie Districts. Unpublished report, Louisiana Natural Heritage Program, De- partment of Wildlife and Fisheries, Baton Rouge, Louisiana. 138 PO NOT OMLIONG TeA volume 72(2):130-138 February 1992 Martin, P.G., C.L. Butler, E. Scott, J.£. Lyles, M. Mariano, J. Ragus, P. Mason, & L. Schoelerman. 1990. Soil survey of Natchitoches Parish, Louisiana. United States Department of Agriculture, Soil Conservation Service. Orzell, S.L. 1990. Texas Natural Heritage Program inventory of National Forests and National Grasslands in Texas. Texas Natural Heritage Pro- gram. Parks & Wildlife Department. Austin, Texas. Smith, L.M. 1988. The natural communities of Louisiana. Unpublished report, Louisiana Natural Heritage Program, Department of Wildlife and Fisheries, Baton Rouge, Louisiana. Smith, L.M. 1991. Louisiana longleaf an endangered legacy. Louisiana Con- servationist, May/June, 24-27. Smith, L.M., N.M. Gilmore, R.P. Martin, & G.D. Lester. 1989. Keiffer cal- careous prairies/forest complex: a research report and preliminary man- agement plan. Unpublished report, Louisiana Natural Heritage Program, Department of Wildlife and Fisheries, Baton Rouge, Louisiana. Phytologia (February 1992) 72(2):139-140. TWO NEW INTRODUCED SPECIES IN THE ALASKAN FLORA Jason R. Grant Departinent of Botany, University of Maryland, College Park, Maryland 29742-5815 U.S.A. ABSTRACT Two taxa previously unreported from Alaska have been found in the Fairbanks area. These are Conyza canadensis (L.) Cronquist and Sonchus uliginosus Bieb. KEY WORDS: Asteraceae, Conyza, Sonchus, Alaska Weedy species are often overlooked in the field. In Alaska, collectors tend to focus on the native arctic flora giving little attention to disturbed areas with introductions from more southerly regions of North America. With the development in Alaska during the last 50 years, invasive plants common in the U.S. Pacific Northwest and the north central plains areas of Canada and the contiguous United States, have been introduced into the most intensely developed areas. The pioneer seeds are probably carried by way of vehicles traveling the Alaska Highway from the Yukon Territories, British Columbia, and Alberta in Canada, and Washington and Montana in the United States into the interior of Alaska. From there they may be spread further south along the Parks, Richardson, and Glenallen Highways toward the Anchorage metropolitan area. The other possible means by which these introductions may be appearing is the increase in agricultural establishments in the Fairbanks, Big Delta, and Matanuska Valley regions. Within the grain and vegetable crops raised in these areas, there are most certainly seeds of weedy taxa which themselves are capable of producing seeds able to survive the long winter. These may then spread to roadsides and other disturbed areas. Recent studies of disturbed areas in the Fairbanks area have led to the discovery of two species previously unrecorded in Alaska. Two members of the Asteraceae, Conyza canadensis (L.) Cronquist and Sonchus uliginosus Bieb. are new introductions to the state. 139 140 PHY TOLOGTIA volume 72(2):139-140 February 1992 Conyza canadensis (L.) Cronquist SPECIMEN EXAMINED: UNITED STATES. Alaska: Fairbanks Quad., Fairbanks, Eton Blvd., 64° 54’ N, 147° 42’ W, 214 m, 22 August 1990, Grant 90-01273 (ALA). Conyza canadensis, a native of North America has become a semicos- mopolitan weed. For this reason it is not too surprising to see it arriving in Alaska. Sonchus uliginosus Bieb. SPECIMENS EXAMINED: UNITED STATES. Alaska: Fairbanks Quad., Fairbanks. Corner of Taku Rd. and Farmers Loop Rd. near the University of Alaska campus, 64° 51’ 30” N, 147° 49’ 20” W, 24 August 1988. Grant 88- 200 (ALA); Corner of Taku Rd. and Farmers Loop Rd. near the University of Alaska campus, 64° 51’ 30” N, 147° 49’ 20” W, 22 August 1990, Grant 90-01281 (ALA,US); corner of Lathrop Rd. and Mitchell Expressway, 64° 50’ N, 147° 43' W, 22 August 1990, Grant 90-01291 (ALA,US); Corner of S. Cushman and Van Horn Rd., 64° 50’ N, 147° 43’ W, 22 August 1990, Grant 90-01292 (ALA,US). Sonchus uliginosus has been naturalized throughout the northeastern Unit- ed States from Europe and is moving westward and northward with continued development. ACKNOWLEDGMENTS I would like to thank Tass Kelso, James L. Reveal, and James R. Rundell for review of the manuscript; and Al Batten for providing correct latitude and longitude coordinates. Phytologia (February 1992) 72(2):141-143. THELECHITONIA CUATRECASAS, AN OLDER NAME FOR COMPLAYA STROTHER (ECLIPTINAE-HELIANTHEAE-ASTERACEAE) Harold Robinson & Jose Cuatrecasas Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 U.S.A. ABSTRACT New combinations are made for the four species belonging to the genus Thelechitonia Cuatr., based on T. muricata Cuatr. (= Wedelia brachycarpa Baker). The genus includes the three species recently placed in Complaya Strother, based on Silphium trilobatum L. (= Wedelia trilo- bata [L.] Hitchc.). KEY WORDS: Asteraceae, Heliantheae, Thelechitonia, Complaya, Wedelia trilobata Recent considerations of the Wedelia group of the Heliantheae subtribe Ecliptinae have brought out two points of significance regarding the genus Th- elechitonia Cuatrecasas (1954). Cuatrecasas has noted, but not published, the fact that the type species of his genus, T. muricata Cuatr. of Colombia, has proven to be the same as the mostly central South American Wedelza brachy- carpa Baker based on examining a fragment of the type collection of the latter, Balansa 8550 (US) and other specimens at US. Cuatrecasas and Dr. Cabr- era of Argentina had corresponded and agreed on the topic in 1965 (Cabrera, Cuatrecasas, personal communications). Also, it is noted that Wedelia brachy- carpa, according to personal observations, was a fourth species of the genus Complaya Strother that has been omitted by Strother (1991) in the delimita- tion of his genus. The Baker species, which occurs mostly in Paraguay and adjacent regions, had all the characteristics cited by Strother for his segregate of Wedelia Jacq., including the prostrate habit with roots at the nodes, the solitary terminal heads that become laterally displaced, the black anther ap- pendages, the thickened and often roughened achene walls, and the rostra and pappi that are obscured in mature achenes by development of a corky collar. As illustrated in the publication of Thelechitonia (Cuatrecasas 1954), the bases 141 142 Pa Y 7 OO GTA volume 72(2):141-143 February 1992 of the achenes in Wedelia brachycarpa are often more narrowly pointed than illustrated by Strother (1991), but the bases vary in the species. The leaves of Wedelia brachycarpa are not so succulent as those of some material cited by Strother, but many noncoastal specimens of the common Wedelia trilobata (L.) Hitchc. have leaves that are equally herbaceous. The result of the observations cited above is that Thelechitonia is an older name for Complaya Strother. Since it has been the intention to follow most of the Strother alignments in the treatments of the Heliantheae in the Flo- ras of Ecuador and Peru now in preparation by Robinson, new combinations are needed. The three species included in Complaya by Strother have been checked, and all three are transferred here. The fourth species also needs a combination since the name for the type species described by Cuatrecasas (1954) is not the oldest available name for the species. Thelechitonia Cuatrecasas, Bull. Soc. Bot. France 101:242. 1954. TYPE: Thelechitonia muricata Cuatrecasas (= T. brachycarpa (Baker) Robinson & Cuatrecasas. Complaya Strother, Syst. Bot. Monogr. 33:10. 1991. TYPE: Complaya trilobata (L.) Strother. The genus contains the following four species. Thelechitonia brachycarpa (Baker) H. Robinson & Cuatrecasas, comb. nov. BASIONYM: Wedelia brachycarpa Baker in Martius, Flora Brasiliensis 6(4):181. 1884 (1885). Bolivia, Brasil, Colombia, Paraguay, Argentina, Venezuela. Specimens seen from Venezuela were determined as Wedelia paludosa var. villosa. Thelechitonia chinensis (Osbeck) H. Robinson & Cuatrecasas, comb. nov. BASIONYM: Solidago chinensis Osbeck, Dagb. Ostind. Resa 241. 1757. Eastern Asia. Thelechitonia gracilis (Richard in Persoon) H. Robinson & Cuatre- casas, comb. nov. BASIONYM: Wedelza gracilis Richard in Per- soon, Syn. Pl. 2:490. 1807. Antilles. Thelechitonia trilobata (L.) H. Robinson & Cuatrecasas, comb. nov. BASIONYM: Silphium trilobatum L., Syst. Nat. ed. 10, 2:1233. 1759. Pantropical. Synonyms include Wedelia brasiliensis (Spren- gel) S.F. Blake and Wedelia paludosa DC. Robinson & Cuatrecasas: Thelechitonia supersedes Complaya 143 LITERATURE CITED Cuatrecasas, J. 1954. Nouvelles Composées de |’Amerique du Sud. Bull. Soc. Bot. France 101:242-246. Strother, J.L. 1991. Taxonomy of Complaya, Elaphandra, logeton, Je- fea, Wamalchitamia, Wedelia, Zermenia, and Zyzyria. (Compositae- Heliantheae-Ecliptinae). Syst. Bot. Monogr. 33:1-111. Phytologia (February 1992) 72(2):144-151. NEW COMBINATIONS IN KLAPHANDRA STROTHER (ECLIPTINAE-HELIANTHEAE-ASTERACEAE) Harold Robinson Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 U.S.A. ABSTRACT Aspilia Thouars is formally reduced to synonymy, and the previous placement of Gymnolomia H.B.K. in synonymy under Aspilia is cor- rected. The type of Aspilia, A. thouarsu DC., is transferred to Wedelia, and the lectotype of Gymnolomia, G. tenella H.B.K., is transferred to Eleutheranthera Poit. ez Bosc. Nine species from northern South Amer- ica are transferred to Elaphandra Strother from Aspilia, and Elaphan- dra paucipunctata is described as new from Ecuador. KEY WORDS: Asteraceae, Heliantheae, Elaphandra, Aspilia, Wedelia, Eleutheranthera, Gymnolomia The recent study of various members of the Wedelia relationship of the Heliantheae (Strother 1991) is the culmination of a series of studies of the | limits of Wedelia Jacq. and Zermenia La Llave (Becker 1975a, 1975b, 1979; Rindos 1980; McVaugh 1972, 1984; Robinson 1978, 1984a, 1984b; Strother 1987, 1989a, b; Villasenor & Strother 1989). One aspect of the Strother study, as well as the previous McVaugh (1972, 1984) studies, is the dismissal of the traditional concept of Aspilia Thouars, which has contained Wedelza-like species with neutral rays. Most of the neutral rayed species within the Strother | study area have been transferred by McVaugh (1972) and Strother (1991) directly into Wedelia. However, one newly described Panamanian species is | placed by Strother in his new genus Elaphandra, and some additional South | American species were mentioned and annotated as possible members of the | new genus. Unfortunately, Strother, like Rindos (1980), chose not to publish | a number of the combinations that were the inevitable result of his study. Elaphandra Strother was rather well defined within the limited geogra- | phy of the Strother (1991) paper by its erect to scrambling habit, lateral leaf, 144 Robinson: New combinations in Elaphandra 145 veins reaching near the leaf tip, lack of resinous glandular punctations on the leaves, herbaceous outer involucral bracts, neutral rays, black anther ap- pendages, and narrow rather stipitate based achene bases with no elaiosomes and small carpopodia. The base of the achene lacking an elaiosome and lack- ing a large carpopodium is a primary distinction from the Strother concept of Wedelia. There are also three tendencies found in some but not all species of Wedelia that are not known in Elaphandra: fertile rays such as those of typical Wedelia yellow anther appendages, and resinous glandular punctations on the leaves. Some emphasis is given by Strother to the unique nonrostrate or scarcely rostrate, epappose or shortly bicornute apex of the achene in the type of the genus Elaphandra, Elaphandra bicornis Strother, but Strother sug- gests probable close relationship to the Colombian species named by Blake as Aspilia quinquenervis in which the rostrum and corona are more highly devel- oped. The lack of tuberculae on the achene is also used by Strother as a key character distinguishing Elaphandra from Eleutheranthera Poit. ex Bosc. and Thelechitonia Cuatr. (= Complaya Strother). The Strother separation of Elaphandra from Wedelia is accepted here. The separation from Aspilia is also accepted on the basis of the original description of that genus (Petit-Thouars 1806) and the description and illustration of the type A. thouarsiz DC. by Humbert (1963), which indicate that Aspilia is a synonym of Wedelia. The following combination formalizes the reduction of Aspilia to synonymy under Wedelia. The eventual dispositions of many species presently placed in Aspilia still need to be resolved. Wedelia thouarsii (A.DC.) H. Robinson, comb. nov. BASIONYM: Aspzlia thouarsi A.DC., Prodr. 5:561. 1836. The species of Elaphandra belong to neither Wedelia nor Aspilia among the pre-existing genera, but the problem of distinguishing the Strother concept of Elaphandra from Eleutheranthera Poit. ex Bosc. and Gymnolomia H.B.K. is not as easily solved. One of the key differences from Eleutheranthera used by Strother (1991) is the lack of tuberculae on the achenes of the former, but a number of the potential members of Elaphandra from South America have tuberculae. A second difference used by Strother, the lack of rays in Eleutheranthera, fails if the Colombian Aspilia tenella (H.B.K.) S.F. Blake is transferred to that genus, as suggested by Strother in his annotations of spec- imens. Nevertheless, four characteristics have been noted in this study that distinguish the expanded concept of Elaphandra from the expanded concept of Eleutheranthera. First, as noted by Strother, Elaphandra lacks resinous glandular punctations on the leaves, but they are present in Eleutheranthera. Second, the disk corollas of Elaphandra always have distinct fiber sheaths along the veins of the throat, a feature lacking in Eleutheranthera. Third, the anther 146 Poll YT OFsOrG TA volume 72(2):144-151 February 1992 appendages of Elaphandra are of ordinary oblong-ovate shape with no glands. The appendages of Eleutheranthera are very blunt, wider than long, and have glands abaxially. Fourth, the style branches of Elaphandra are blunter and more densely papillose distally. The style tips of Eleutheranthera are attenu- | ate with sparse spreading papillae. A secondary effect of the expanded concept of Eleutheranthera is the reso- lution of the genus Gymnolomia. The latter genus was originally credited with four species. Three of the species were transferred to Aspilia by Blake (1924), and two proved to be Aspilia tenella, which Blake selected as the lectotype of Gymnolomia. The fourth species, G. rudbeckioudes H.B.K., was transferred by Blake to Hymenostephium Benth. and was transferred later by Robinson (1977) to Viguiera H.B.K. The lectotypification of Gymnolomia by G. rud- beckioides, as was belatedly suggested by D’Arcy (1975, p. 1156-1157), was contrary to D’Arcy’s own suggestion that Gymnolomia might be the correct | name for much of the New World material placed in the genus Aspilia. Thus, Gymnolomia, which has been previously placed in the synonymy of Aspilia, proves to be a synonym of Eleutheranthera. The following summary of the resolution of Eleutheranthera and Gymnolo- | mia is possible. Eleutheranthera Poit. ez Bosc., Nouv. Dict. Hist. Nat., ed. 1. 7:498. 1803. LECTOTYPE: Eleutheranthera ovata Poit. ex Steud., nom. nud. (= | Eleutheranthera ruderalis [Swartz] Schultz-Bip.). Ogiera Cass., Bull. Soc. Philom. 1818:32. 1818. TYPE: Ogiera triplin- ervis Cass. (= Eleutheranthera ruderalis [Swartz] Schultz-Bip.). Gymnolomia H.B.K., Nov. Gen. Sp., ed. fol. 4:170. 1818. LECTO- TYPE: Gymnolomia tenella H.B.K. (Blake 1924). Fingalia Schrank, Syll. Ratisb. 1:87. 1824. TYPE: Fingalia heragona | Schrank. Gymnopsis A. DC., Prodr. 5:561. 1836. nom. superfl. for Gymnolomia. Kegeha Schultz-Bip., Linnaea 21:245. 1848. TYPE: Kegelia ruderalis (Swartz) Schultz-Bip. Eleutheranthera tenella (H.B.K.) H. Robinson, comb. nov. BA- SIONYM: Gymnolomia tenella H.B.K., Nov. Gen. Sp., ed. fol. | A171. 1818: Gymnolomia hondensis H.B.K., Nov. Gen. Sp., ed. fol. 4:171. 1818. Robinson: New combinations in Elaphandra 147 Aspilia tenella (H.B.K.) S.F. Blake, Contr. U.S. Natl. Herb. 22(8): 620. 1924. The expanded concept of Elaphandra consists at this time, of species lack- ing resinous glandular punctations on their leaves. having lateral leaf veins reaching the distal fourth of the leaf, bearing neutral rays or no rays, having fiber sheaths on the veins of the disk corolla throat, having black, ovate anther appendages without glands, having style branches densely papillose to the tip, and having achenes narrowed at the base without obvious elaiosomes or car- popodia. The bases of the achenes are usually not so elongated as in the type of the genus. As such, the genus is enlarged from the single species of Strother (1991) to include four additional groups of species previously placed in the genus Aspilia. One group is of apparent immediate relatives of the type, such as A. quinquenervis S.F. Blake that was mentioned by Strother (1991). A sec- ond group includes species notable for black spots on their leaves or black lines in their involucral bracts. A member of the group, A. verbesinoides (A.DC.) §.F. Blake, was annotated by Strother as a possible Elaphandra. The third group consists of an Ecuadorian species that is notable for a lack of rays. A fourth group consists of a recently described Venezuelan species with compara- tively small heads and pointed pales. Also, in an effort to provide for all names known to be needed for the Flora of Ecuador treatment, the opportunity is taken to describe an additional species from that country which belongs to the second group. Elaphandra Strother, Syst. Bot. Monogr. 33:17. 1991. TYPE: Elaphandra bicornis Strother, eastern Panama. Group I. Elaphandra macrolepis (S.F. Blake) H. Robinson, comb. nov. BA- SIONYM: Aspilia macrolepis S.F. Blake, Contr. U.S. Natl. Herb. 22:617. 1924. This seems closest to the type of the species being transferred, but the outer involucral bracts are longer and narrower, and the corona of the achene is more developed. Colombia. Elaphandra quinquenervis (S.F. Blake) H. Robinson, comb. nov. BASIONYM: Aspilia quinquenervis S.F. Blake, J. Wash. Acad. Sci. 18:26. 1928. The species is variable in the density and inclination of hairs on the lower leaf surface, and the Ecuadorian specimens have generally broader and shorter outer involucral bracts. Colombia and Ecuador. 148 PHY TO O'e FA volume 72(2):144-151 February 1992 Group II. Elaphandra archeri (H. Robinson & Brettell) H. Robinson, comb. nov. BASIONYM: Aspilia archeri H. Robinson & Brettell, Phy- tologia 32:419. 1975. Colombia. Elaphandra eggersii (Hieron.) H. Robinson, comb. nov. BASIONYM: Aspilia eggersu Hieron., Bot. Jahrb. Syst. 28:606. 1901. A related species is described below based on material once identified as this species. Ecuador. Elaphandra lucidula (S.F. Blake) H. Robinson, comb. nov. BA- SIONYM: Aspilia lucidula Proc. Biol. Soc. Wash. 36:52. 1923. Aspilia steinbachii H. Robinson & Brettell, Phytologia 32:420. 1975. Bolivia. Elaphandra ulei (Hieron.) H. Robinson, comb. nov. BASIONYM: As- pila ule Hieron., Verh. Bot. Ver. Brandenb. 48:205. 1906. (1907). Western Brazil. Elaphandra verbesinoides (A. DC.) H. Robinson, comb. nov. BA- SIONYM: Gymnopsis verbesinoides A. DC., Prodr. 5:561. 1836. Aspilia nigropunctata S.F. Blake, Proc. Biol. Soc. Wash. 24:119. 1911. Aspilia verbesinoides (A. DC.) S.F. Blake, Proc. Biol. Soc. Wash. 34:120. 1921. Trinadad, Tobago, Venezuela. Group III. Elaphandra pastazensis H. Robinson, comb. nov. BASIONYM: As- | pilia pastazensis H. Robinson, Phytologia 55:417. 1984. Ecuador. Group IV. | Elaphandra falconiensis (Badillo) H. Robinson, comb. nov. BA- | SIONYM: Aspilia falconiensis Badillo, Ernstia 9:13. 1983. Venezu- | ela. Robinson: New combinations in Elaphandra 149 New species of Group II. Elaphandra paucipunctata H. Robinson, sp. nov. HOLOTYPE: ECUA- DOR. El Oro Prov.: 5 km W of Pinas on the road to Saracay, elev. 3600 ft., low spreading shrub 0.5 m tall, ray florets yellow, disk florets black but the lobes yellow, 4 Feb. 1979, King & Almeda 7969 (US). PARATYPE: ECUADOR. El Oro Prov.: Road from Pinas to Sta. Rosa, km 16, elev. 620 m, creeping vine in shrubs, flowers yellow, 7 Oct. 1979, Dodson, Gentry, & Shupp 8943 (US). Plantae volubiles vel patentiter frutescentes ad 0.5 m altae; caules erecte vel leniter retrorse hirsutuli. Folia opposita, petio- lis 6-12 mm longis; laminae ovatae plerumque 5-10 cm longae et 2.5-5.5 cm latae base breviter acutae margine serrulatae apice bre- viter acuminatae supra scabridulae subtus antrorse pilosae et pauce nigropunctatae fere ad basem distincte trinervatae. Inflorescentiae unicapitatae longe pedunculatae, pedunculis saepe 5-12 cm longis leniter retrorse hirsutulis. Capitula ca. 1 cm alta sine radii 1.0-1.5 cm lata; squamae involucri exteriores herbaceae oblongo-ovatae 9- 10 mm longae et ca. 5 mm latae apice breviter acutae extus dense pilosulae, bracteae interiores sensim membranaceae obovatae ca. 10 mm longae ad 6.5 mm latae pauce nigropunctatae apice late rotundatae extus dense scabridulae; paleae scariosae stramineae apicae breviter acutae suberosae extus glabrae. Flores radii ca. 8; corolla flavae glabrae, tubis ca. 2 mm longis, limbis oblongis ca. 17 mm longis et 6 mm latis apice distincte bilobatis. Flores disci ca. 35; corollae nigrescentes inferne et in lobis flavescentiores extus glabrae, tubis 1.5-2.0 mm longis faucibis ca. 3 mm longis, lobis ca. 1 mm longis in marginis interioribus dense longe papillosis; thecae et appendices antherarum nigrescentes, thecae ca. 2 mm longae. Achaenia immatura. The new species is related to the more widespread Ecuadorian Elaphandra eggersu with the same darkened corollas of the disk florets. The species differs by the much shorter tips of the leaves, the sparse occurrence of black spots on the undersurface of the leaves, and the much shorter outer involucral bracts. The black spotting of the leaves is obvious compared to the unspotted condi- tion seen in £. eggersii, but is still much less obvious than that seen in species like E. archeri, E. verbesinoides, or E. lucidula. 150 PHY TOs OETA volume 72(2):144-151 February 1992 LITERATURE CITED Becker, K.M. 1975a. New combinations in Wedelia Jacq. (Asteraceae). Phy- tologia 31:25. _______.. 1975b. New combination in Lasianthaea DC. Phytologia 31:297. . 1979. A monograph of the genus Lasianthaea (Asteraceae). Mem. New York Bot. Gard. 31(2):1-64. Blake, S.F. 1924. New American Asteraceae. Contr. U.S. Natl. Herb. 22(8):587-661. i-xi, pl. 54-63. D’Arcy, W.G. 1975. 73. Viguiera, in Flora of Panama. Annals Missouri Bot. Gard. 62:1156-1161. Humbert, H. 1963. Aspilia. In Flore de Madagascar et des Comores (Plantes Vasculaires) 189° Famille Composées 3:648-654. McVaugh, R. 1972. Compositarum mexicanarum pugillus. Contr. Univ. Michigan Herb. 9:359-484. . 1984. Compositae. In Flora Novo-Galiciana, ed. W.R. Anderson, 12:1-1157. University of Michigan Press, Ann Arbor, Michigan. Petit-Thouars, L.M.A. 1806. Genera Nova Madagascar, secund. pp. 1-30. Paris, France. Rindos, D. 1980. Generic delimitation in the ves|r]besinoid Heliantheae (Compositae). Amer. J. Bot. 68:206-215. Robinson, H. 1977. Studies in the Heliantheae (Asteraceae), VIII. Notes on genus and species limits in the genus Viguzera. Phytologia 36:201-215. . 1978. Lundelhanthus, a new genus from Guatemala (Heliantheae- Asteraceae). Wrightia 6:43-45, pl. 82. . 1984a. Studies in the Heliantheae (Asteraceae), XXXI. Additions to the genus Dimerostemma. Proc. Biol. Soc. Wash. 97:618-526. —____—. 1984b. Studies in the Heliantheae (Asteraceae), XXXIV. Redelim- itation of the genus Angelphyum. Proc. Biol. Soc. Wash. 97:961-969. Strother, J.L. 1987. Damnzanthodium (Compositae-Heliantheae) a new genus from Mexico. Syst. Bot. 12:41-43. Robinson: New combinations in Elaphandra 151 . 1989a. Oblivia, a new genus for Zexmenia mikanioides (Compositae- Heliantheae). Syst. Bot. 14:541-543. 1989b. Expansion of Lundellianthus (Compositae-Heliantheae). Syst. Bot. 14:544-548. 1991. Taxonomy of Complaya, Elaphandra, Iogeton, Jefea, Wa- malchitamia, Wedelia, Zermenia, and Zyzyzia (Compositae-Heliantheae- Ecliptinae). Syst. Bot. Monogr. 33:1-111. Villasenor, J.L. & J.L. Strother. 1989. Tuztla, a new genus for Zexmenia pittiert (Compositae-Heliantheae). Syst. Bot. 14:529-540. Phytologia (February 1992) 72(2):152-156. XYRIS ISOETIFOLIA KRAL (XYRIDACEAE) NEW TO ALABAMA AND ITS RANGE AND HABITATS IN FLORIDA Steve L. Orzell Florida Natural Areas Inventory, 1018 Thomasville Rd., Suite 200-C, Tallahassee, Florida 32303 U.S.A. & Edwin L. Bridges | Florida Department of Environmental Regulation, 2600 Blair Stone Rd., Tallahassee, Florida 32399 U.S.A. ABSTRACT Xyris isoetifolia Kral was previously thought to be a narrow endemic | of the Florida panhandle and was represented by collections from six stations. It is here reported as new to Alabama, from the sandy margin of a sinkhole pond in Covington County. The habitats of five newly discovered Florida panhandle stations are discussed and related to the Alabama station. KEY WORDS: Xyris isoetifolia, Xyridaceae, Alabama, limesink ponds Xyris isoetifolia Kral was described in 1966 from northwest Florida as oc-. curring in Bay and Gulf counties (Kral 1966) and has been reported as endemic to Florida (Muller et al. 1989; Anderson 1989). Kral (1966; 1983) described the habitat for X. isoetifolia as moist sands or sandy peats of savanna bogs, flatwoods pond margins, and shores of limesink ponds and lakes. Anderson (1989) reported that the Gulf County record (Kral 1966; Clewell 1985) was in error, and that the only other record for X. tsoetzfolia was actually from Washington County. There is a Washington County specimen at VDB, “Sandy shore of Chain Lakes, W of Greenwood,” Godfrey & Henderson 62954, 6 July 1963; which is also cited in Kral’s unpublished list of exsiccatae for his 1966 152 Orzell & Bridges: Xyris tsoetifolia 153 Xyris monograph. Anderson made additional collections from either side of an unimproved timber access road that straddles the Gulf and Bay county line. Anderson (1989, p. 501) erroneously reported his collections #11,713 and #11,776 as a county record for Gulf County, since the labels indicate that these numbers are actually from Bay County. However, Anderson did collect X. isoetifolia in Gulf County, but the collections are numbered 11,712 and 11,775, based upon our examination of specimens at the FSU herbarium. In summary, X. isoetifolia had previously been known from six collections at four Bay County stations (Kral 15651 [holotype], Kral 23193, Godfrey & Houk 61551, Godfrey & Kral 62425, Anderson 11713, Anderson 11776), one Washington County station, and two collections from a single station in Gulf County (Anderson 11712 and 11775). Our collections have added five new sta- tions for Florida; four in Bay County (Orzell & Bridges 14218, 15240, 17107, and 18278) and one in Washington County (Orzell & Bridges 9784). Our Florida collections are from three distinct community types: seep- age herb bogs or seepage savannas (Bay and Gulf counties), coastal scrubby flatwoods (Bay County), and margins of sandy limesink ponds (Washing- ton County). The Bay and Gulf county seepage habitats are dominated by Pleea tenuifolia Michx. In these areas Xyris isoetifolia grows in associa- tion with Bigelowia nudata (Michx.) DC., Burmannia capitata (Walt.) Mart., Cliftonia monophylla (Lam.) Britt. ez Sarg., Dichanthelium dichotomum (L.) Gould, Drosera capillaris Poir., Drosera tracy: MacFarlane, Erigeron vernus (L.) Torrey & Gray, Eriocaulon compressum Lam., Eriocaulon decangulare L., Fuirena breviseta (Coville) Coville in Harper, Gaylussacia mosiert Small, Hypericum brachyphyllum (Spach) Steud., Lachnanthes caroliniana (Lam.) Dandy, Lachnocaulon anceps (Walt.) Morong, Lachnocaulon digynum Korn., Lachnocaulon minus (Chapman) Small, Linum medium (Planch.) Britt. var. teranum (Planch.) Fern., Lophiola aurea Ker-Gawl., Lycopodium alopecuroides L., Lycopodium prostratum Harper, Panicum rigidulum Bosc ex Nees, Physoste- gia godfreyi Cantino, Pinguicula lutea Walt., Polygala lutea L., Rhexia ahifanus Walt., Rhezia lutea Walt., Rhynchospora baldwinti A. Gray, Rhynchospora chapmanu M.A. Curtis, Rhynchospora curtissw Britt. ec Small, Rhynchospora fikfolia A. Gray, Rhynchospora oligantha A. Gray, Rhynchospora plumosa Ell., Sarracenia flava L., Sarracenia psittacina Michx., Scleria reticularis Michx., Smilaz laurifolia L., Syngonanthus flavidulus (Michx.) Ruhl., Tofieldia race- mosa (Walt.) B.S.P., and Xyris ambigua Beyr. ez Kunth. Within less than one mile of the Gulf of Mexico and its embayments in coastal Bay County, Xyris isoetifolia can be locally abundant along disturbed sandy road embankments through scrubby flatwoods. Scrubby flatwoods occur where a relatively deep sandy surface overlies an impervious spodic layer, re- sulting in alternating conditions of subsurface saturation and drought. These flatwoods typically have a pine canopy with a dense tall shrub layer primarily of evergreen scrub oaks and ericaceous shrubs. Closely associated plants in these 154 PRY TOL OGCTA volume 72(2):152-156 February vs habitats include Xyris brevifolia Michx., X. elliottis Chapm., and Lachnocaulon anceps. Other associates recorded at scrubby flatwoods in coastal Bay County include: Asclepias cinerea Walt., Carphephorus odoratissimus (J.F. Gmel.) Hebert, Conradina canescens (Torrey & Gray) A. Gray, Euphorbia telephioides Chapman, Hypericum tetrapetalum Lam., [ler glabra (L.) A. Gray, Kalmia hir- suta Walt., Liatris tenuifolia Nutt., Lyonia ferruginea (Walt.) Nutt., Lyonza lucida (Lam.) K. Koch, Pinus palustris P. Mill., Polygonella gracilis (Nutt.)| Meisn., Quercus minima (Sarg.) Small, Rhynchospora plumosa, Smilax au- riculata Walt., Cyrilla racemiflora L., and Gratiola subulata Baldw. Where) seepage emerged on a road edibanbinent through scrubby flatwoods in coastal | Bay County, the following associates were recorded: Anthaenantia rufa (Ell.)) Schultes, Bigelowia nudata, Burmannia capitata, Carphephorus pseudoliatris| Cass., Drosera capillaris, D. tracyi, Erigeron vernus, Hypericum brachyphyl-| lum, Juncus scirpoides Lam., Lachnocaulon anceps, Lobelia puberula Michx., | Lycopodium carolinianum L., Pleea tenuifolia, Polygala cruciata L., Polygala lutea, Rhezia alifanus, Rhezia lutea, Rheria mariana L., Rhynchospora chap-. mani, Rhynchospora oligantha, Rhynchospora plumosa, Sarracenia flava, Scle- ria reticularis, Syngonanthus flavidulus, Utricularia juncea Vahl, and Xyris elliott. At our Washington County site Xyris isoetifolia grows on the upper margin of a relatively steep sided sinkhole lake, where seepage emerging from a nearby steephead stream saturates the exposed sandy lakeshore. Both the seasonal and long term fluctuating water levels of this sinkhole lake and others charac- teristic of the Greenhead Slope physiographic region, in Bay and Washington counties, keep the upper shoreline from being invaded by most woody plants, thereby producing an intermittently exposed sandy margin which varies in width according to the lake water level. During extended periods of low wa- ter conditions at this sinkhole lake, X. isoetifolia is abundant and grows in close association with Amphicarpum muhlenbergianum (Schultes) A.S. cng Centella asiatica (L.) Urban, Drosera filiformis Raf. (at one of its few sta- tions south of North Carolina), Fuirena scirpoidea Michx., Hypericum reduc: tum P. Adams, Lachnocaulon minus, Rhynchospora pleiantha (Kukenth.) Gale, Sagittaria isoetiformis J.G. Sm., Syngonanthus flavidulus, Utricularia cornutc Michx., U. subulata L., Waedvandia areolata(L.) T. Moore, Xyris baldwinianc Schulte »& ran idee Malme, and X. longisepala Kral. Clethra alnifolia L.| Gliffonia monophylla, and Ilex naka form a scrubby zone at the mean seasona. high water level of the lake, just upslope from the X. isoetifolia. Hypericun| lissophloeus P. Adams can occur as scattered shrubs in association with X isoetifolia. During our visits to the site (Sept. 1989, Aug. 1991) when higl lake water levels inundated the bare sandy margin, we did not observe any X isoetifolia. While conducting floristic studies on the Gulf Coastal Plain we collectec Xyris isoetifolia in southern Alabama at a site about 120 km northwest of th: Orzell & Bridges: Xyris tsoetifolia 155 nearest known locality for it in Florida. The collection data for our Alabama specimens are as follows: Xyris isoetifolia Kral (Xyridaceae). UNITED STATES. Alabama: Covington Co.: Sandy fluctuating shoreline of limesink pond on N shore of Blue Pond, at picnic area at end of FS Rd. to E of AL Hwy. 1387 at a point 1.1 mi N of jct. with Covington Co. Rd. 24; Conecuh Wildlife Management Area in Conecuh National Forest; NWQ, NEQ, Sec. 28, T2N R15E, Wing 7.5’ Quad.; 31° 06’ 51” N, 86° 33’ 11” W; Elev. 214-220 ft.; 12 May 1989, Orzell & Bridges 9820 (VDB,TEX,USF). Same locality, 11 Sept 1989, Orzell & Bridges 11711 (FSU,GA,MO,NCU,NY,VDB,TEX,WIS). Xyris isoetifolia occurs on the upper sandy margin of a limesink pond just below the perennial graminoid dominated zone at the Alabama site. It grows in small tufts on moist, exposed sand in association with X. longisepala, Lachnocaulon minus and occasionally X. jupicai L.C. Rich. Eupatorium lepto- phyllum DC. and Hypericum fasciculatum Lam. occur in this zone as scattered individuals and partially shade Xyris isoetzfolia. The Blue Pond location has been the source of several new and interesting plant records for Alabama (Kral 1973; Kral 1976; McGinty 1983). Kral 1973; 1976) reported several plant taxa new to Alabama from Blue Pond, including Eupatorium leptophyllum (Kral 1976) and Xyris longisepala (Kral 1973) which are associates of X. isoetifolia. Kral (1973) also mentioned several other rare taxa for Alabama which occur at Blue Pond (Rhezia salicifolia Kral & Bostick, Eriocaulon lineare Small, Lachnocaulon minus, and Sagittaria isoetiformis). A search by the authors of nearby limesink ponds in the vicinity of Blue Pond has thus far failed to locate any additional populations of X. zsoetifolia. Based upon our field observations in Florida and Alabama, and the relative abundance of Xyris isoetifolia at these stations, it appears to be most frequent on the outer coastal terraces of Bay County, Florida. Within the Gulf Coastal Lowlands region of the Florida panhandle, it occurs in seepage herb bogs and seepage savannas in both Bay and Gulf counties and in scrubby flatwoods of coastal Bay County. It is also locally abundant further inland on seepage sat- urated exposed sandy margins of steep sided, seasonally fluctuating limesink ponds in Bay and Washington counties on the sandy upland of the Greenhead Slope. Here, X. isoetifolia is often found in association with a number of en- demic vascular plants of these habitats. In southern Alabama it is apparently rare and presently known only from the sandy margin of a similar limesink pond in Covington County. Xyris isoetifolia occurs abundantly within micro- habitats where there is some periodic seepage or capillary saturation of sand or sandy peat. It appears to be most abundant where either natural (7.e., fire, soil slumping) or artificial disturbance of the habitat (z.e., occasional soil dis- turbance due to roadside maintenance, clearcutting, etc.) has occurred in the 156 PEY)T OLTO GTA volume 72(2):152-156 February 1992 recent past. Xyris zsoetifolia thrives where such disturbance reduces competi- tion from the surrounding herbaceous vegetation. Within these microhabitats it can form small tufts, which under favorable conditions can form small areas of turf. Here it may be locally dominant, but it is unlikely to persist as a long term component in severely disturbed habitats once taller perennial vegetation becomes established. Additional localities for this infrequently collected Xyris should be sought in similar habitats elsewhere in the southeastern Coastal Plain. ACKNOWLEDGMENTS We thank Guy Nesom and Carol Todzia at TEX for their helpful review comments. LITERATURE CITED Anderson, L.C. 1989. Noteworthy plants from north Florida. Sida 13:497- 504. Clewell, A.F. 1985. Guide to the Vascular Plants of the Florida Panhandle. Florida State University Press, Tallahassee, Florida, 605 pp. Kral, R. 1966. Xyris (Xyridaceae) of the continental United States and Canada. Sida 2:177-260. . 1973. Some notes on the flora of the southeastern states, particu- larly Alabama and middle Tennessee. Rhodora 75:366-410. . 1976. Additions to some notes on the flora of the southern states, particularly Alabama and middle Tennessee. Rhodora 78:438-456. 1983. A report on some rare, threatened, or endangered forest- related vascular plants of the South. U.S.D.A., Forest Service, Tech. Publ. R8-TP2. 1305 pp. McGinty, D.T. 1983. Inventory of proposed threatened or endangered plant species of the Conecuh Division, Conecuh National Forest, Alabama. Contract No. 53-43ZP-1-00860. 108 pp. Muller, J.W., E.D. Hardin, D.R. Jackson, $.E. Gatewood, & N. Caire. 1989. Summary report on the vascular plants, animals, and plant communities endemic to Florida. Florida Game and Freshwater Fish Comm. Nongame Wildl. Program Tech. Rept. No. 7. 113 pp. Wi 5 00288 4813 Information for Authors Articles from botanical systematics and ecology, including biographical sketches, critical reviews, and summaries of literature will be considered for publication in PHYTOLOGIA. Manuscripts may be submitted either on computer diskette, or as typescript. Diskettes will be returned to authors after action has been taken on the manuscript. Diskettes may be 5.25 inches or 3.5 inches and may be written in any IBM or MacIntosh compatible format. Typescript manuscripts should be single spaced and will be read into the computer using a page scanner. The scanner will read standard typewriter fonts but will not read dot matrix print. Manuscripts submitted in dot matrix print cannot be ac- cepted. Use underscore (not italics) for scientific names. Corrections made on typescript manuscripts must be complete and neat as the scan- ner will not read them otherwise. Language of manuscripts may be either English or Spanish. Figures will be reduced to fit within limits of text pages and therefore, should be submitted with an internal scale and have dimensions proportional to those for text pages. Legends for figures should be included in figures whenever possible. Each manuscript should have an abstract and key word list. Specimen citations should be consistent throughout the manuscript. Serial titles should be cited with abbreviations used in Botanico Periodicum Huntianum. References cited only as part of nomenclatural summaries should not appear in Literature Cited. Nomenclatural work should include one paragraph per basionym and must provide proper (as defined by the current /nferna- tional Code of Botanical Nomenclature) citation of sources of epithets and combinations. Authors should arrange for two workers in the appropriate field to review the manuscript before submission. Copies of reviews should be forwarded to the editor with the manuscript. Manuscripts will not be published without review. . Cost of publication is currently $13.00 US per page for publication without reprints. Publication with 100 reprints is provided for $18.00 US per page, 200 reprints for $21.50 US per page. Page charges are due with manuscript and no paper will be published before payment is received in full. Reprints must be ordered and paid for in advance. Page charges will be determined on the basis of a typescript page (single spaced, 10 points, blank line between paragraphs) with all type inside a rectangle 143 mm (horizontal) by 219 mm (vertical), not including running head and page number. Title page should include title, author(s) name(s), and address(es). Two blank lines should appear above and below section headings (Abstract, Discussion, Literature Cited, etc.) in the manuscript. No extra charge is made for line drawings provided they conform to limitations of size and proportion for normal text. Halftones require an _ extra charge of $10.00 US per page.