173 PHYTOLOGIA An international journal to expedite plant systematic, phytogeographical and ecological publication Vol. 73 August 1992 No. 2 CONTENTS ROBINSON, H., A new genus Vernonanthura (Veruonieae, Asteraceae). 65 RAYNER, T.G.J., Studies on the genus Bidens L. (Compositae) from the Eastern Hemisphere. 3. Typification of names of Bidens, Coreopsis L., Guizotia Cass., and Microlecane (Schultz-Bip.) Benth. & Hook. f. from ONL NSAT GC SORE ene eae tab areara MEL AACE A ALD IRL 1 Yr 77 ; | . 4 fos A. & S. CARLQUIST, Eight new taxa of Drosera from Australia. 98 he CLEMENTS, M.A. & D.L. JONES, New combinations in Appendicula Blume and Octarrhena Thwaites (Orchidaceae) in Australia. ............. 117 _—NESOM, G.L., Variation in Erigeron chiangii | Asteraceae: Astereae), with ‘ DEMIR SRL el tee er au tae bys hi bia ee eihate eee 119 Bees, J.T. & K.N. GANDHI, Nomenclatural notes for the North Amer- STS SS ay ee a Seah ad wee ae 28 ee Vetere he Conkle 124 __NESOM, G.L., A new species of Castilleja (Scrophulariaceae) from Durango, . MME Vo O LD Vi watt SAN Livin 's'd 8 bv vv 50RD tn Meg ea ee ee 137. - _“NESOM, G.L., A new species of Justicia (Acanthaceae) from northeastern MUMMERS RAs SOLO AYALA T. Nal ala b's vB pe bho yw ie (a/al baba ate idy aata e S 140 ~ TURNER, B.L. & J. PANERO, New species and combinations in Podachae- Gam (Asteraceae, Velianthese).. i... ieee tales da mesad eaee sae 143 : ~Contents continued on the inside cover f } r2 CCT 26 1992 WYEVY YORK +TANICAL GARDE? (Contents continued) _/ ROBINSON, H., Pour new species of Clibadiurn from northern South America (Asteraceae: Heliantheae).) i. 01 i aisle eye veins e wees tices tae 149 WINDLER, D., L. ADLER, & S. SKINNER, Crotalaria incana var. grandi- flora (Leguminosae), a new variety from South America. .......... 155 ‘HOLMES, W.C., Studies on Mikania (Compositae: Eupatorieae) - XVIII. New species from) Brazile sik) 5.2 sic. Bie fie ld dipkesalate ean hk 159 OCHOA, C.M., New synonyms in the tuber bearing Solanum. .........166 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 42-66 (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. Phytologia (August 1992) 73(2):65-76. A NEW GENUS VERNONANTHURA (VERNONIEAE, ASTERACEAE) Harold Robinson Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 U.S.A. ABSTRACT Vernonanthura is described for 62 Neotropical species previously placed in Vernonia with Baccharis brasiliana L. as type. The genus dif- fers from Vernonia in the woody habit, the erect sometimes xylopodial bases, the thyrsoid to pyramidal inflorescences, and the often prominent basal appendages on the anther thecae. KEY WORDS: Asteraceae, Vernonieae, Vernonanthura, Vernonia, new genus During the last decade, the author has produced a number of papers in which many genera have been segregated from Vernonia Schreb. Among these papers are many that treat the “Lepidaploa Complex” (Robinson 1987a, b, c; 1988a, b, c; 1990), the members of which are considered part of the subtribe Vernoniinae, but which were initially distinguished as a group from typical Vernonia only by the extremely scorpioid cymose or seriate cymose forms of their inflorescences or by their lophate pollen grains. In the papers on the Lepidaploa Complex, the limits of typical Vernonia in eastern North Amer- ica, México, and the Bahamas have been left imprecisely defined, and some Neotropical elements of the Vernoniinae that were less obviously distinct from Vernonia were left in the latter genus, in spite of the realization that they eventually needed to be removed. Improved knowledge of the limits of typ- ical Vernonia and the need for fully revised concepts for floristic treatments make the description of the present new genus necessary at this time. The new genus includes the largest group of Neotropical species that has remained in Vernonia. None of the existing generic names in the Vernonieae has been found to apply to this group. The new name is derived from the generic name Vernonia plus the complex ending derived from “anthera” (anther) and “oura” (tail), referring to the frequently tailed anther bases. 65 66 PHY TOTO G TA volume 73(2):65-76 August 1992 Vernonanthura H. Robinson, gen. nov. TYPE: Baccharis brasiliana L. Plantae erectae fruticosae vel arborescentes raro volubiles ( Ver- nonanthura cocleana |Keeley] H. Robinson); xylopodia saepe prae- sentia; caules teretes vel angulati (V. pinguzs [Griseb.] H. Robin- son) medulla solida, pilis simplices vel T-formibus. Folia alternata sessilia vel longe petiolata; laminae lineares vel lanceolatae vel late oblongae vel obovatae, base obtusae vel attenuatae raro cordatae vel auriculatae ( V. amplezicaulis [Fries] H. Robinson) margine inte- grae ad dense serratae apice obtusae vel breviter acuminatae supra glabrae vel pilosulae vel scabrae subtus glabrae vel tomentosae glandulo-punctatae, nervis secondariis patentiter pinnatis. Inflo- rescentiae thyrsoideae vel pyramidaliter paniculatae, ramis dense ad laxe vel leniter seriato-cymosis vel corymbiformis, foliis in nodis inferioribus inflorescentiae leniter vel valde minoribus, bracteolis superiores parvis, pedunculis nullis vel brevibus plerumque ad 8 mm longis raro longioribus (V. tweedieana [Baker] H. Robinson). Capitula homogama; involucra plerumque late campanulata; brac- teis 15-3plo numeriores quam floribus plerumque 16-30 (ca. 60 in V. crassa [Vell. Conc.] H. Robinson) in seriebus 4-10 valde imbricatis subcoriaceis ovatis vel oblongis apice obtusis raro bracteis inferi- oribus apice acutis et reflexis (V. angulata |H. Robinson] H. Robin- son) interioribus persistentibus; receptacula epaleacea. Flores 4-30 in capitulo; corollae lavandulae vel albae anguste infundibulares, faucibus distinctis interne glabris, lobis linearibus glanduliferous, pilis nullis, ductis interioribus longitudinalibus numerosis; thecae antherarum base breviter vel longe appendiculatae; appendices api- cales oblongo-ovatae, appendices et connectivae saepe abaxialiter glanduliferae, parietibus cellularum tenuibus; base stylorum dis- ciformiter nodulosi. Achaenia prismatica vel obovata glabra vel dense setulifera et glandulifera 8-10 costata, idioblastis non glan- dularibus plerumque in aggregatis 1-3 cellularibus, rhaphides sub- quadratis vel late oblongis; carpopodia superne constricta in part- ibus superioribus saepe setulifera et glandulifera; setae pappi biseri- atae, seriebus exterioribus brevibus et plerumque squamiformibus, seriebus interioribus capillaris apice plerumque distincte latiores. Grana pollinis in diametro plerumque 37-40 ym typus A tricolpo- rata, tectis perforatis inter colpis continuis, z = 17 (Jones 1979, 5 spp.; Keeley 1978, 1 sp.; Stutts 1988, 5 spp.). Like most American Vernonieae, Vernonanthura has a basic chromosome number of z = 17. The new genus is clearly a member of the subtribe Ver- noniinae in having a well developed basal stylar node and apical anther ap- pendages with thin walled cells and often with glands. Within the subtribe, Robinson: Vernonanthura, new genus in Vernonieae 67 the new genus is unlike members of the Lepidaploa Complex and similar to Vernonia in the Type A rather than lophate pollen, in the lack of truly scorpi- oid cymes or cymes with large foliose bracts in the inflorescence, in the lack of hairs or spicules on the corolla lobes, and in the subquadrate to short oblong rather than elongate raphids in the achene walls. In all these respects the new Neotropical genus seems to be a close relative of the North Temperate Vernonia. As presently recognized, Vernonia sensu stricto is mostly eastern North American, with a few species such as V. ervendbergu A. Gray, V. fausttana (Chapman & Jones) Turner, V. greggit A. Gray, and V. schaffneri A. Gray extending into México, and V. blodgettz: Small and V. tnsularis Gleason reach- ing the Bahamas. Unlike Vernonia, Vernonanthura ranges throughout the Neotropical Region from México to Argentina. In the West Indies it includes those species placed by Keeley (1978) in the subsects. Burifoliae Ekman and Polyanthes Ekman. Jones (1976) treated some of the Mexican species in his revision of subsect. Paniculatae ser. Umbelliformes Gleason. South American species were placed by Cabrera in subsects. Chamaedrys Cabrera and Nitidulae Cabrera, and many are placed in subsects. Nudzflorae Cabrera and Polyanthes Ekman. Species of the new genus were placed by Stutts (1988) in the ser. Brasilanae Stutts, Nitidulae (Cabrera) Stutts, and Puberulae Stutts, subser. Chamaedrys (Cabrera) Stutts, Larae Stutts, Nudiflorae Stutts, and Chague- nsis Stutts. In Baker (1873) members of the genus fell mostly into his infor- mal subsections Lepidaploae Scorpioideae, Scorpioidea Verae, and Lepidaploae Paniculatae. No previous treatment seems to have recognized the species of Vernonanthura as a single group apart from other elements of Vernonia. Gleason (1922) noted the most useful distinction between Vernonia sensu stricto and other species in tropical North America by his key character, “Inflo- rescence corymbiform, its branches irregular in length; heads few to many, on peduncles of irregular length or some nearly sessile; species of northern México, the United States, and the Bahama Islands,” All the typical Vernonia species examined show such inflorescences with shallowly rounded, flattened, or cy- mose form. In contrast, the inflorescence in Vernonanthura is pyramidal to thyrsoid in form with only individual branches showing corymbose or cymose shape. In cases where whole plants are available, a second difference between Vernonia and Vernonanthura can be seen. In Vernonia the bases of the stems arise from a creeping stem, and they are herbaceous. In Vernonanthura the stems are erect from a woody base without evident creeping stems. The bases often have xylopodia, as noted in the subsection Chamaedrys by Stutts (1988). The plants are shrubs or even trees. A third difference is significant in Ver- nonanthura, but is not found in all the species. Vernonia is recognized in the traditional concepts of the genus for its lack of tailed anther bases, and this is actually true of Vernonia sensu stricto. However, in Vernonanthura a number of species have tails. In some cases, such as V. diffusa (Less.) H. Robinson, 68 PHYTOL OG A volume 73(2):65-76 August 1992 V. discolor (Less.) H. Robinson, and V. petzolaris (A.DC.) H. Robinson, the tails are highly developed, as long as the tails of Pzptocarpha R. Br. of the Piptocarphinae, which is traditionally distinguished from Vernonza on the ba- sis of that character. The prominently tailed species of Vernonanthura, like many Old World species with tailed anthers (Robinson & Kahn 1986), should never have been placed in Vernonza as traditionally defined. The type species of Vernonanthura, V. brasiliana (L.) H. Robinson, has shorter, but neverthe- less, distinct tails. These sterile bases of the anthers are obvious in opened florets under the dissecting microscope by their texture. Most other species of Vernonanthura have some sterile cells at the anther base, often forming some teeth, but there are members of the genus that have anther bases essentially like those of Vernonia. Other possible distinctions of Vernonanthura remain to be critically exam- ined. The corolla lobes of many Vernonanthura have elongate internal struc- tures that appear to be thin walled, rather obvious, parallel ducts filling the lobe, and other species seem to share the character in a weaker form. These structures have not been studied in detail, but they have not been noticed in Vernonia or Eremosis (A.DC.) Gleason. The corolla lobes of Vernonia, have thicker and more roughened tips than those of Vernonanthura. In one case where a hair was seen on a corolla lobe in Vernonanthura, it was not T-shaped, but a few of the rare hairs seen microscopically in Vernonia were T-shaped. Most members of the new genus closest to the type have carpopodia with setulae or glands intruding onto the upper, constricted, sclerified part. The glands or setulae do not arise directly from sclerified cells, but their bases are often completely surrounded by such cells. The sclerified parts of carpopo- dia in Vernonia sensu stricto are not constricted above and bear no setulae or glands. j The genus contains the following 62 Neotropical species that have been examined by the author. The list excludes some species that have not been studied as carefully, and the synonymies exclude combinations made in Ca- cala Burm. by Kuntze (1891) except one name that is a basionym. Four comparatively familiar names, Vernonia missionis Gardner, V. nitidula Less., V. polyanthes Less., and V. ruficoma Schlecht. er Baker fall into synonymy respectively under the names Vernonanthura cymosa (Vell. Conc.) H. Robin- son, V. montevidensis (Spreng.) H. Robinson, V. phosphorica (Vell. Conc.) H. Robinson, and V. membranacea (Gardner) H. Robinson. Vernonanthura almedae (H. Robinson) H. Robinson, comb. nov. BA- SIONYM: Vernonia almedae H. Robinson, Phytologia 46:107. 1980. Vernonanthura amplexicaulis (Fries) H. Robinson, comb. nov. BASIO- NYM: Vernonia amplezicaulis Fries, Ark. Bot. 5(13):5. 1906. Robinson: Vernonanthura, new genus in Vernonieae 69 Vernonanthura angulata (H. Robinson) H. Robinson, comb. nov. BA- SIONYM: Vernonia angulata H. Robinson, Phytologia 45:170. 1980. Vernonanthura auriculata (Griseb.) H. Robinson, comb. nov. BASIONYM: Vernonia auriculata Griseb., Symb. Fl. Argent. 164. 1879. Vernonanthura beyrichii (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia beyrichi Less., Linnaea 4:275. 1829. Vernonia denticulata A.DC., Prodr. 5:36. 1836. Vernonanthura brasiliana (L.) H. Robinson, comb. nov. BASIONYM: Baccharis brasiliana L., Sp. Pl. ed. 2, 1205. 1763. Vernonia brasthana (L.) Druce, Rep. Bot. Exch. Cl. Brit. Isles 3:426. 1913 (1914). Vernonia scabra Pers., Syn. Pl. 2:404. 1807. Vernonia odoratissima H.B.K., Nov. Gen. Sp., ed. fol. 4:32. 1818. Vernonia assana Mart. in A.DC., Prodr. 5:38. 1836. Vernonanthura buxifolia (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia buzifolia Less., Linnaea 4:313. 1829; based on Lepidaploa buzi- folia Cass. Lepidaploa buzifolia Cass., Dict. Sci. Nat. 26:18. 1823; nom. inval., described in Lepidaploa prior to validation at generic level. Vernonia domingensis A.DC., Prodr. 5:30. 1836. Vernonia montana Gleason, Bull. New York Bot. Gard. 4:191. 1906. Vernonanthura canaminina (Gleason) H. Robinson, comb. nov. BA- SIONYM: Vernonia canaminina Gleason, Amer. J. Bot. 10:309. 1923. Vernonanthura catharinensis (Cabrera) H. Robinson, comb. nov. BA- SIONYM: Vernonia catharinensis Cabrera, Sellowia 13:180. 1961. Vernonanthura chamaedrys (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia chamaedrys Less., Linnaea 4:259. 1829. Vernonia iler Chodat, Bull. Herb. Boissier, ser. I]. 4:410. 1902. Vernonanthura cichoriifolia (Chodat) H. Robinson, comb. nov. BASIO- NYM: Vernonia cichoriifolia Chodat, Bull. Herb. Boissier, ser. I]. 2:300. 1902. Vernonanthura chaquensis (Cabrera) H. Robinson, comb. nov. BASIO- NYM: Vernonia chaquensis Cabrera, Darwiniana 6:358. 1944. Vernonanthura cocleana (Keeley) H. Robinson, comb. nov. BASIONYM: Vernonia cocleana Keeley, Brittonia 39:44. 1987. Vernonanthura condensata (Baker) H. Robinson, comb. nov. BASIONYM: Vernonia condensata Baker, J. Bot. 8:202. 1875. 70 PH YT. OL O1G ITA volume 73(2):65-76 August 1992 Vernonia sylvestris Glaz., Bull. Soc. Bot. France 56, Mém. 1(3):373. 1909; nom. nud. Vernonia bahtensts Toledo, Arq. Bot. Estado Sao Paulo, n.s. 1:52. 1939. Vernonanthura cordata (H.B.K.) H. Robinson, comb. nov. BASIONYM: Vernonia cordata H.B.K., Nov. Gen. Sp., ed. fol. 4:31. 1818. Vernonia morelana Gleason, Bull. Torrey Bot. Club 46:241. 1919. Vernonanthura crassa (Vell. Conc.) H. Robinson, comb. nov. BASIONYM: Chrysocoma crassa Vell. Conc., Fl. Flumin. 305. 1825 [1829]. Vernonia crassa (Vell. Conc.) Ekman er Malme, Kongl. Svenska Vetenskapsakad. Hand. III. 12(2):24. 1933. Vernonanthura cronquistii (S.B. Jones) H. Robinson, comb. nov. BA- SIONYM: Vernonia cronquistu S.B. Jones, Rhodora 78:194. 1976. Vernonanthura cuneifolia (Gardner) H. Robinson, comb. nov. BASIONYM: Vernonia cunetfolia Gardner, London J. Bot. 5:215. 1846. Vernonia itapensis Chodat, Bull. Herb. Boissier, ser. II. 2:301. 1902. Vernonanthura cupularis (Chodat) H. Robinson, comb. nov. BASIONYM: Vernonia cupularis Chodat, Bull. Herb. Boissier, ser. II. 2:299. 1902. Vernonanthura cymosa (Vell. Conc.) H. Robinson, comb. nov. BASIONYM: Chrysocoma cymosa Vell. Conc., Fl. Flumin. 327. 1825 [1829]; not Ver- nonia cymosa Blume, 1826. Vernonia missionis Gardner, London J. Bot. 4:115. 1845. Vernonanthura deppeana (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia deppeana Less., Linnaea 6:398. 1831. Vernonanthura diffusa (Less.) H. Robinson, comb. nov. BASIONYM: Vernonta diffusa Less., Linnaea 4:272. 1829. Vernonanthura discolor (Less.) H. Robinson comb. nov. BASIONYM: Vernonia discolor Less., Linnaea 4:274. 1829. Vernonanthura fagifolia (Gardner) H. Robinson, comb. nov. BASIONYM: Vernonia fagifolia Gardner, London J. Bot. 5:216. 1846. Vernonanthura ferruginea (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia ferruginea Less., Linnaea 4:271. 1829. Vernonia polycephala A.DC., Prodr. 5:39. 1836. Vernonia crenata Gardner, London J. Bot. 5:218. 1846. a Robinson: Vernonanthura, new genus in Vernonieae 71 Vernonanthura fuertesii (Urban) H. Robinson, comb. nov. BASIONYM: Eupatorium fuertesti Urban, Repert. Spec. Nov. Regni Veg. 17:9. 1921. Vernonia fuertesi (Urban) H. Robinson, Phytologia 38:149. 1977. Vernonia barkeri Ekman ez Urban, Ark. Bot. 23A(11):49. 1931. Vernonanthura havanensis (A.DC.) H. Robinson, comb. nov. BASIONYM: Vernonia havanensis A.DC., Prodr. 5:37. 1836. Vernonia cubensis Griseb., Cat. Pl. Cub. 144. 1866. Vernonia stictophylla Wright, Sauv. Anal. Cien. Havana 6:176. 1894. Vernonanthura hieracioides (Griseb.) H. Robinson, comb. nov. BASIO- NYM: Vernonia hieractoides Griseb., Mem. Amer. Acad. Arts 8:511. 1861. Vernonia ortentis Gleason, Bull. Torrey Bot. Club 40:330. 1913. Vernonanthura ignobilis (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia zgnobilis Less., Linnaea 6:658. 1831. Vernonanthura laxa (Gardner) H. Robinson, comb. nov. BASIONYM: Vernonia lara Gardner, London J. Bot. 5:214. 1846. Vernonanthura liatroides (A.DC.) H. Robinson, comb. nov. BASIONYM: Vernonia liatroides A.DC., Prodr. 5:34. 1836. Vernonia ehrenbergiana Schultz-Bip., Linnaea 20:513. 1847. Eupatorium tulanum Klatt, Abh. Naturf. Ges. Halle 15:324. 1882. Vernonia capreaefolia Gleason, Bull. New York Bot. Gard. 4:200. 1906. Vernonanthura loretensis (Hieron.) H. Robinson, comb. nov. BASIONYM: Vernonia loretensis Hieron., Bot. Jahrb. Syst. 22:676. 1897. Vernonanthura lucida (Less.) H. Robinson, comb. nov. BASIONYM: Ver- nonia lucida Less., Linnaea 4:260. 1829. Vernonanthura mariana (Mart. er Baker) H. Robinson, comb. nov. BA- SIONYM: Vernonia mariana Mart. ez Baker, Fl. Bras. 6(2):107. 1873. Vernonanthura membranacea (Gardner) H. Robinson, comb. nov. BA- SIONYM: Vernonia membranacea Gardner, London J. Bot. 5:217. 1846. Vernonia ruficoma Schlecht. ez Baker, Fl. Bras. 6(2):105. 1873. Vernonanthura menthaefolia (Poeppig ez Spreng.) H. Robinson, comb. nov. BASIONYM: Vernonia menthaefolia Poeppig ex Spreng., Syst. Veg., ed. 16, 3:412. 1826. Eupatorium perrinianum Spreng., Syst. Veg., ed. 16, 3:412. 1826. Vernonia ottonis Schultz-Bip., Linnaea 20:508. 1847. Vernonia grisebachii Schultz-Bip., J. Bot. 1:231. 1863. 72 PBX T O LOrG IA volume 73(2):65-76 August 1992 Vernonanthura montevidensis (Spreng.) H. Robinson, comb. nov. BA- SIONYM: Baccharis montevidensis Spreng., Syst. Veg., ed. 16. 3:460. 1826; not Vernonia montevidensis Nees ex Otto & A. Dietr. Vernonia nitidula Less., Linnaea 4:266. 1829. Vernonia gochnatioides Hook. & Arn. er A.DC., Prodr. 7:264. 1838. Vernonia florida Gardner, London J. Bot. 5:212. 1846. Vernonia arechavaletae André, Rev. Hort. 73:284. 1901. Vernonanthura mucronulata (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia mucronulata Less., Linnaea 4:266. 1829. Vernonia collina Gardner, London J. Bot. 5:213. 1846. Vernonanthura nudiflora (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia nudtflora Less., Linnaea 4:258. 1829. Vernonanthura oaxacana (Schultz-Bip. ez Klatt) H. Robinson, comb. nov. BASIONYM: Vernonia oaracana Schultz-Bip. ex Klatt, Leopoldina 20:74. 1894. Vernonanthura oligactoides (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia oligactoides Less., Linnaea 4:648. 1831. Vernonia sorocabae Schultz-Bip. ez Baker, Fl. Bras. 6(2):58. 1873. Vernonia conyzoides Chodat, Bull. Herb. Boissier, ser. I]. 2:303. 1902. Vernonanthura oligolepis (Schultz-Bip. ez Baker) H. Robinson, comb. nov. BASIONYM: Vernonia oligolepis Schultz-Bip. er Baker, Fl. Bras. 6(2):56. 1873. Vernonanthura paludosa (Gardner) H. Robinson, comb. nov. BASIONYM: Vernonia paludosa Gardner, London J. Bot. 4:113. 1845. Vernonanthura patens (H.B.K.) H. Robinson, comb. nov. BASIONYM: Vernonia patens H.B.K., Nov. Gen. Sp., ed. fol. 4:32. 1818. Vernonia baccharoides H.B.K., Nov. Gen. Sp., ed. fol. 4:32. 1818. Vernonia lanceolaris A.DC., Prodr. 5:37. 1836. Vernonia haenkeana A.DC., Prodr. 5:37. 1836. Vernonia micradenia A.DC., Prodr. 5:38. 1836. Vernonia pacchensis Benth., Pl. Hartw. 134. 1844. Vernonia aschenborniana Schauer, Linnaea 19:714. 1847. Vernonia stuebeli: Hieron., Bot. Jahrb. Syst. 21:337. 1895. Vernonia bangit Rusby, Mem. Torrey Bot. Club 6:52. 1896. Vernonia weberbaueri Hieron., Bot. Jahrb. Syst. 40:354. 1908. Vernonia monsonensis Hieron., Bot. Jahrb. Syst. 40:335. 1908. Vernonia salamana Gleason, Bull. Torrey Bot. Club 46:242. 1919. Vernonia vargasit Cuatr., Bot. Jahrb. Syst. 77:83. 1956. Robinson: Vernonanthura, new genus in Vernonieae 73 Vernonanthura petiolaris (A.DC.) H. Robinson, comb. nov. BASIONYM: Vernonia petiolaris A.DC., Prodr. 5:37. 1836. Vernonia hilariana Gardner, London J. Bot. 4:113. 1845. Vernonanthura phaeoneura (Toledo) H. Robinson, comb. nov. BASIO- NYM: Vernonia phaeoneura Toledo, Arq. Bot. Estado Sao Paulo 1(4):95. 1942. Vernonanthura phosphorica (Vell. Conc.) H. Robinson, comb. nov. BA- SIONYM: Chrysocoma phosphorica Vell. Conc., Fl. Flumin. 325. 1825 [1829]. . Chrysocoma arborea Vell. Conc., Fl. Flumin. 326. 1825 [1829]. Vernonia polyanthes Less., Linnaea 6:651. 1831. Vernonia psittacorum A.DC., Prodr. 5:36. 1836. Vernonia corcovadensis Gardner, London J. Bot. 5:218. 1846. Vernonanthura pinguis (Griseb.) H. Robinson, comb. nov. BASIONYM: Vernonia pinguts Griseb., Symb. Fl. Argent. 165. 1879. Vernonanthura piresii (H. Robinson) H. Robinson, comb. nov. BASIONYM: Vernonia ptresit H. Robinson, Phytologia 45:178. 1980. Vernonanthura prenanthoides (Gleason) H. Robinson, comb. nov. BA- SIONYM: Vernonia prenanthoides Gleason, Amer. J. Bot. 10:308. 1923. Vernonanthura puberula (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia puberula Less., Linnaea 6:649. 1831. Vernonanthura rigiophylla (Kuntze) H. Robinson, comb. nov. BASIONYM: Cacalia rigiophylla Kuntze, Revis. Gen. Pl. 2:971. 1891; based on Ver- nonta rigiophylla Schultz-Bip. er Baker. Vernonza rigiophylla Schultz- Bip. ez Baker, Fil. Bras. 6(2):118. 1873; not Vernonia rigiophylla A.DC., 1836. Vernonia elsieae Stutts, Brittonia 35:351. 1983. Vernonanthura sambrayana (S.B. Jones) H. Robinson, comb. nov. BA- SIONYM: Vernonia sambrayana S.B. Jones, Fieldiana, Bot., n.s. 5:34. 1980. Vernonanthura serratuloides (H.B.K.) H. Robinson, comb. nov. BA- SIONYM: Vernonia serratuloides H.B.K., Nov. Gen. Sp., ed. fol. 4:33. 1818. Perezia paniculata A. Gray, Proc. Amer. Acad. Arts 21:393. 1886. Perezia vernonioides A. Gray, Proc. Amer. Acad. Arts 22:433. 1887. Vernonia jaliscana Gleason, Bull. New York Bot. Gard. 4:198. 1906. Vernonia umbellifera Gleason, Bull. New York Bot. Gard. 4:199. 1906. 74 PHY TOL, OGTA volume 73(2):65-76 August 1992 Vernonia vernonioides (A. Gray) Bacigalupi, Contr. Gray Herb. 97:77. 1931. Vernonia camporum M.E. Jones, Contr. West. Bot. 18:69. 1933. Perezta nervata M.E. Jones, Contr. West. Bot. 18:74. 1933. Vernonanthura sinclairii (Benth.) H. Robinson, comb. nov. BASIONYM: Vernonia sinclatru Benth., Bot. Voy. Sulphur 109. 1845. Vernonanthura squamulosa (Hook. & Arn.) H. Robinson, comb. nov. BA- SIONYM: Vernonia squamulosa Hook. & Arn., Companion Bot. Mag. 2:44. 1836. Cacalia praecor Kuntze, Revis. Gen. Pl. 3:139. 1898. Vernonia praecor (Kuntze) Schumann, Just’s Bot. Jahresber. 26(1):382. 1900. Vernonanthura stellata (Spreng.) H. Robinson, comb. nov. BASIONYM: Conyza stellata Spreng., Neue Entdeck. 2:142. 1820. Vernonanthura subverticillata (Schultz-Bip. er Baker) H. Robinson, comb. nov. BASIONYM: Vernonia subverticillata Schultz-Bip. ex Baker, Fl. Bras. 6(2):99. 1873. Vernonanthura tuerckheimii (Urban) H. Robinson, comb. nov. BASIO- NYM: Vernonia tuerckheimiu Urban, Symb. Antill. 7:421. 1912. Vernonia microphylla Alain, Mem. New York Bot. Gard. 21:156. 1971. Vernonia pusilliflora Alain, Mem. New York Bot. Gard. 25:279. 1973. Vernonanthura tweedieana (Baker) H. Robinson, comb. nov. BASIONYM: Vernonia tweedieana Baker, Fl. Bras. 6(2):99. 1873. Vernonanthura viscidula (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia viscidula Less., Linnaea 4:289. 1829. Vernonia corymbulosa Mart. ez Baker, Fl. Bras. 6(2):113. 1873. Vernonanthura westiana (Less.) H. Robinson, comb. nov. BASIONYM: Vernonia westiana Less., Linnaea 6:650. 1831. Vernonia hebeclada A.DC., Prodr. 5:36. 1836. Vernonanthura yurimaguasensis (Hieron.) H. Robinson, comb. nov. BA- SIONYM: Vernonia yurimaguasensis Hieron., Verh. Bot. Vereins Prov. Brandenburg 48:195. 1907. Vernonza albifila Gleason, Bull. Torrey Bot. Club 59:374. An additional species, Vernonia schulziana Cabrera, would belong to the genus according to the treatments by Cabrera (1944) and Stutts (1988), but material has not been seen. Some additional species treated by Jones (1976) in the ser. Umbelliformes of México, such as Vernonza karvinskiana A.DC., have Robinson: Vernonanthura, new genus in Vernonieae 75 narrow ducts in the interiors of their corolla lobes, and they may be related to Vernonanthura. Nevertheless, the tips of their involucral bracts are more expanded and their inflorescences are less pyramidal or thyrsoid, and so they are excluded for the present. LITERATURE CITED Baker, J.G. 1873. Compositae I. Vernoniaceae in C.F.P. Martius, Flora Brasiliensis 6(2):2-179. Cabrera, A.L. 1944. Vernonieas Argentinas (Compositae) Darwiniana 6:227- 231. Gleason, H.A. 1922. Carduales, Tribe 1. Vernonieae North American Flora 33(1):47-110. Jones, S.B. 1976. Revision of Vernonia (Compositae) subsection Paniculatae series Umbelliformes of the Mexican highlands. Rhodora 78:180-206. Keeley, S.C. 1978. A revision of the West Indian vernonias (Compositae). J. Arnold Arbor. 59:360-413. Kuntze, C.E.O. 1891. Revisio Generum Plantarum, vol. 2. Arthur Felix, Leipzig, Germany. Robinson, H. 1987a. Studies in the Lepidaploa complex (Vernonieae: Aster- aceae) I. The genus Stenocephalum Sch.-Bip. Proc. Biol. Soc. Wash. 100:578-583. 1987b. Studies in the Lepidaploa complex (Vernonieae: Aster- aceae). II. A new genus, Echinocoryne. Proc. Biol. Soc. Wash. 100:584- 589. 1987c. Studies in the Lepidaploa complex (Vernonieae: Aster- aceae). III. Two new genera, Cytocymura and Eirmocephala. Proc. Biol. Soc. Wash. 100:844-855. . 1988a. Studies in the Lepidaploa complex (Vernonieae: Asteraceae) IV. The new genus, Lessingianthus. Proc. Biol. Soc. Wash. 191:929-951. _______. 1988b. Studies in the Lepidaploa complex (Vernonieae: Asteraceae) V. The new genus, Chrysolaena. Proc. Biol. Soc. Wash. 101:952-958. . 1988c. Studies in the Lepidaploa complex (Vernonieae: Asteraceae) VI. A new genus, Aynia. Proc. Biol. Soc. Wash. 101:959-965. 76 Po Ys TOE. OLGA A volume 73(2):65-76 August 1992 . 1990. Studies in the Lepidaploa complex (Vernonieae: Asteraceae) VII. The genus Lepzdaploa. Proc. Biol. Soc. Wash. 103:464-498. . & B. Kahn. 1986. Trinervate leaves, yellow flowers, tailed anthers, and pollen variation in Distephanus Cassini (Vernonieae: Asteraceae). Proc. Biol. Soc. Wash. 99:493-501. Stutts, J.G. 1988. Taxonomic revision of Vernonia subsect. Chamaedrys (Compositae: Vernonieae). Rhodora 90:37-99. Phytologia (August 1992) 73(2):77-97. STUDIES ON THE GENUS BIDENS L. (COMPOSITAE) FROM THE EASTERN HEMISPHERE. 3. TYPIFICATION OF NAMES OF BIDENS, COREOPSIS L., GUIZOTIA CASS., AND MICROLECANE (SCHULTZ-BIP.) BENTH. & HOOK. F. FROM AFRICA T.G.J. Rayner Department of Botany, Plant Science Laboratories, University of Reading, Whiteknights, P.O. Box 221, Reading, RG6 2AS, U.K. ABSTRACT Typification is undertaken for names of Bidens, Coreopsis, Gutzotia and Microlecane considered synonymous with names of native species of African Bidens. Lectotypes are chosen for 31 names. Neotypes are selected for three names, viz.: B. diversa Sherfi, B. holsti (O. Hoffm.) Sherff (as C. holstii O. Hoffm.) and B. steppia (Steetz) Sherff (as C. steppia Steetz). KEY WORDS: Bidens, Coreopsis, Guizotra, Microlecane, Com- positae, typification, taxonomy, Africa INTRODUCTION During the course of revisional studies on the genus Bidens in Africa, a large number of problems have arisen concerning typification. Thus, it was decided to undertake a special study to help elucidate these difficulties, which have mainly arisen for the following four reasons. Firstly, numerous type specimens, formerly housed at B, are apparently no longer extant and presumably were destroyed during World War II. Because of this it has been necessary to undertake a comprehensive search of many herbaria (see acknowledgments for list) in order to try to locate duplicates of these specimens. vt 78 PHY TOL CiGTA volume 73(2):77-97 August 1992 Secondly, attempts to typify names published before the full adoption of the “type method” have led to difficulties when a collection, known to consist of two or more duplicates, has been cited in the protologue without reference to herbaria where specimens are housed. The usual practice, often employed by later authors, of adopting as the holotype the specimen of the collection located at the institution where the author of the name worked, or where his “own herbarium” is now housed, is considered unsatisfactory. In particular, this procedure is commonly adopted for names published by Otto Hoffmann in Engler’s Botanische Jahrbticher and Die Pflanzenwelt Ost-Afrikas where the sheets at B are designated as the holotypes by later authors, even though no indication is made in these works that the type material is housed at B, and sheets in other herbaria often bear annotations in Hoffmann’s hand. The flawed nature of these assumptions is well illustrated by a second example. Schultz-Bipontinus (1846) described Bidens schimperi citing only the collec- tion G.H.W. Schimper 1429, of which at least fourteen duplicates housed in nine herbaria are known. Mesfin (1984a) has subsequently designated the sheet of this number at P bearing Schultz-Bipontinus’s handwriting as the holotype. Clearly he has assumed that this specimen was the one used by Schultz-Bipontinus in the construction of his protologue. This assumption is probably based on the fact that this is the only sheet of Schimper 1429 which bears one of Cosson’s printed labels indicating that it was formerly in Schultz- Bipontinus’s own herbarium which is now housed at P. Mesfin’s designation, however, must be rejected on two grounds. First, Schultz-Bipontinus’s de- tailed description contains information that he could not have obtained from this sheet alone; second, other duplicates also possess annotations by Schultz- Bipontinus and were doubtless used by him in the construction of the proto- logue in conjunction with the specimen at P. For these reasons I consider that there is no holotype of B. schimperi. A more favourable approach to this kind of problem, and that adopted here, is to try to ascertain which of the duplicates, if any, were seen by the publishing author. Most frequently this information may be obtained from annotations, etc., in the publishing author’s hand. When it has been established which duplicates were examined then a lectotype should be selected from amongst these. If it is not possible to demonstrate which of the duplicates were seen by the author, then the specimen most closely matching the original description should be selected as the lectotype. In particular, one should avoid automat- ically choosing the specimen at the author’s own herbarium on the grounds that this may be considered a mechanical method of lectotype selection by future workers, and therefore liable to be rejected by invocation of Article 8.1 of the International Code of Botanical Nomenclature (ICBN) (Greuter et al. 1988). A similar procedure to the above has been adopted in those cases where more than one collection (syntype collections) has been cited by a publishing author. Rayner: Studies of Eastern Hemisphere Bidens 3 79 Thirdly, in his monographs of Bidens (Sherff 1937) and Coreopsis (Sherff 1936), Sherff usually adopted the procedure of choosing as the lectotype’ the first cited specimen, or a specimen of the first cited collection, from among the two or more syntypes or syntype collections of previously published names. On the fifteen occasions where he selected lectotypes for names of African Bidens and Coreopsis, in only one instance did he not follow this procedure. In this case, the selection of a lectotype for B. stuhlmanniz (O. Hoffm.) Sherff, his reasoning is as follows: “Hoffmann cited a specimen by Meyer first, but this was a mere scrap, utterly worthless for determination. The fairly large and much better specimen by Stuhlmann, though cited after Meyer’s, was clearly the type from which Hoffmann’s description was drawn.” Indeed, elsewhere Sherff states that the reason a particular specimen has been chosen as the lectotype is because it was the “first one cited” (1937:606). Thus I consider that he has selected lectotypes mechanically, and in accordance with Article 8.1 of the CBN these choices may be superseded. Although in many instances his choice of lectotype is satisfactory, in a number of cases better specimens are available and new lectotypes are here selected from among these. Fourthly, when describing a new species, Sherff occasionally cited more than one specimen of a collection in different herbaria as the type. In these cases I do not consider that he has designated a holotype as defined by Ar- ticle 7.3 of the JCBN, and so a lectotype has been chosen from among the original material. Further to this point, Sherff often adopted an extremely broad concept of a type specimen, usually including within it all sheets of a particular collection in any one herbarium. Clearly this is not the sense in which most contemporary practising taxonomists use this term. The [CBN does not expressly provide a definition of a specimen. Article 9.1, however, states that “The type...is a single specimen. ..except...for small herbaceous plants..., [when] the type may consist of more than one individual”. From this I interpret that the JCBN intends that, except for small herbaceous plants, etc., a type specimen should be one individual, or part thereof. It may, there- fore, be argued that in those cases where Sherff’s “type specimen” is obviously more than one individual, his act of typification is ineffective. I have, however, adopted the following procedures when dealing with this problem. If the “type material” cited by Sherff was formerly all at B, I consider that it may be taken to have been the holotype, as it is not possible in these circumstances to know if his concept of a specimen corresponds with that of the JCBN. The lectotype is then chosen from among the isotypes. If, on the other hand, the “type ma- terial” is still extant and is considered to be of more than one individual, the lectotype is chosen from among these specimens (Article 7.5 of the JCBN). Throughout this paper I have attempted to maintain a constant concept \Sherff did not use the term lectotype in either of these works. I consider, however, that his use of the phrase “type specimen” is, in most instances, “an equivalent” (Article 8.3 of the ICBN). 80 PHY TOL OG TA volume 73(2):77-97 August 1992 of a type specimen when designating lectotypes and neotypes. Usually this is equivalent to a single herbarium sheet, but occasionally, if a sheet bears specimens that are obviously from different plants, only part of this sheet has been selected as the type. The entries are arranged alphabetically within each genus. Except where indicated, it is not intended that these names should be considered as referring to accepted species. Indeed, all the names discussed under Coreopsis, Guizo- tia, and Microlecane are considered by me to be synonymous with species of Bidens, as are most of the names included under Bidens itself. Subsequent papers in this series will indicate the synonymic position of each of the names discussed. Bidens L. Bidens abyssinica Schultz-Bip. var. glabrata Vatke, Linnaea 39:500. 1875. TYPE: ETHIOPIA. Scholloda, 24 Sep. 1862, G.H.W. Schimper 285 (LECTOTYPE NOV. [here selected]: Z; Isolectotypes: BM[2 sheets], PRE). Vatke’s original description of Bidens abyssinica var. glabrata is followed by the citation, in parentheses, of the collection Schimper 285. In addition he in- cludes a discussion involving two other Schimper numbers (105 and 305), here reproduced in full: “Adest in coll. a. 1854 n. 105. e Gaha Meda prope Dschad- scha, a Schweinfurthio Beitr. 142 cum. var. altera (quadriaristata Hochst. fide ejusdem) n. 305 e Gageros confusa, a qua primo intuitu diversissima; nostra transitum praebere videtur ad B. bipinnatum L. a Kotschyo in Nubia reper- tum, cui forte stirps abyssinica reducenda.” Sherff (1937) selected the three sheets of Schimper 105 at B as the lectotype. As these are now apparently destroyed, Mesfin (1984a) has chosen the specimen at Z as the new lectotype. It is clear, however, that Vatke had intended that Schimper 285 should be taken as the type. Throughout this work, when describing new taxa, Vatke consistently cites in parentheses the collection described after the description and before any subsequent notes. Sherff’s failure to mention Schimper 285, except in a list of specimens examined (1937:402), leads me to believe that he simply did not notice Vatke’s citation of this collection. Vatke’s discussion shows that he was in fact comparing Schimper 285 with Schimper 105, but it is clear that he did not consider either Schimper 105 or 305 to be the same taxon as Schimper 285. Therefore, by invoking Article 8.1(b) of the JCBN, Sherff’s and Mesfin’s choice of specimens of the paratype collection Schimper 105 as lectotype is here rejected in favour of a duplicate of Schimper 285. Ac- cording to Sherff (1937:402) a specimen of Schimper 285 was formerly housed at B, but in the absence of any indication by Vatke that this, or any other, was the specimen on which his description is based, all the duplicates of this Rayner: Studies of Eastern Hemisphere Bidens 3 81 collection must have equal status and so are here considered cotypes*. None of the specimens of this number at BM, PRE, and Z bear Vatke’s handwriting, but all are undoubtedly from the same collection and each matches Vatke’s brief varietal description. There is no doubt, however, that the best specimen is that at Z. This sheet bears part of a plant with numerous more or less intact leaves, and capitula at various stages of development. Bidens bequaertii De Wild., Repert. Spec. Nov. Regni Veg. 13:204. 1914. TYPE: ZAIRE. Katanga, Elisabethville, 19 Mar. 1912, J. Bequaert 270 (LECTOTYPE NOV. [here selected]: BR; Isolectotype: BR). In his protologue of Bidens bequaertii, De Wildemann cited the two collec- tions Bequaert 270 and A. Corbister (H.A. Homblé) 605. He did not indicate where the specimens that he had studied were housed, but as the only sheets of these numbers to have been located are at BR, where De Wildemann worked, these specimens may be taken to be syntypes. Sherff (1937) selected as the type the two sheets of the first cited collection, Bequaert 270. As it has not been possible to ascertain whether these specimens are part of the same in- dividual, this designation must be considered contrary to the definition of a lectotype as employed by Article 7.5 of the JCBN, and consequently this choice is here rejected. The specimen chosen as the new lectotype of B. bequaertii is that attached to the sheet which has affixed to it De Wildemann’s manuscript copy of his original description. This specimen closely matches his descrip- tion, and was clearly the main element used by him in its construction. It possesses a branched stem bearing many lobed leaves and numerous capitula with mature fruit. Bidens crocea Welw. ex O. Hoffm., Bol. Soc. Brot. 10:177. 1892. TYPE: ANGOLA. Huilla District, Lake Ivantala, Feb. 1860, F.M.J. Welwitsch 3964 (LECTOTYPE NOV. [here selected]: BM; Isolectotypes: BM,BR, C,COI,G,K,LISU,M,P). The two collections Welwitsch 3964 and J.M. Antunes {s.n.|, both collected in the Huilla District of Angola, were cited by Hoffmann for Bidens crocea, but without indication of where specimens of these numbers were housed. Sherff (1937) selected the sheet of the former at B as the type, but as this is now apparently destroyed, a new lectotype is required. Interestingly, Sherff did not cite any collections by Antunes for this species, but he did include a specimen of E. Dekindt 861 from B, now apparently destroyed, also collected in Huilla. A sheet of B. crocea at LISC, also possessing the number 861, bears a hand written note indicating that it was collected by either Antunes or Dekindt. It 2A cotype is here defined as any duplicate of the type collection when no holotype was designated (see Frizzell 1933). 82 PHY TOTOGIA volume 73(2):77-97 August 1992 is possible, therefore, bearing in mind Sherff’s meticulous citation of the spec- imens he studied, that the Dekindt number at B and the specimen at LISC are part of the original material of B. crocea. Indeed, the well preserved spec- imen at LISC closely matches Hoffmann’s description. However, the existence of numerous duplicates of the other syntype collection avoids the necessity of choosing this specimen as the new lectotype. The two sheets of Welwitsch 3964 at BM are of especially high quality. One of these, that which has attached a copy of Welwitsch’s manuscript description of B. crocea, bears a specimen possessing all the important diagnostic characters of the species and thus is chosen as the new lectotype. Bidens diversa Sherff, Bot. Gaz. 76:159. 1923. TYPE: ANGOLA. Huilla, Lubango, Tundavala, at 12 km, source of the Inhames, 30 Apr. 1971, A. Borges 167 (NEOTYPE [here selected]: LISC; Isoneotypes: M,P, PRE,SRGH). Sherff cited for this species the two sheets of J.M. Antunes 315, collected in Mounyino, Angola, in March 1901, both formerly at B and now apparently destroyed. As no duplicates of this collection have been located, a neotype is here selected in accordance with Article 7.4 of the JCBN. The sheet of Borges 167 at LISC is chosen to serve as the neotype because of the close similarity of this specimen to Sherff’s original description of Bidens diversa. Indeed, all the sheets of this new type collection possess the important diagnostic characters of the fruit and florets which distinguish this species from the closely allied B. acuticaulis Sherff. Bidens flabellata O. Hoffm. in Warb., Kunene-Sambes: Ezped. 419, t. 11G. 1903. TYPE: ANGOLA. at Kuebe near Manonge, 22 Apr. 1900, H. Baum 847 (LECTOTYPE NOV. [here selected]: G; Isolectotypes: BM,BR,K,W). Hoffmann cited the single collection Baum 847 for this species, without designating a type specimen. As all the duplicates of this collection bear his handwriting, and clearly were seen by him, it is unlikely that he used only one specimen to describe this taxon. It follows, therefore, that he has not complied with the definition of a holotype as employed by Article 7.3 of the ICBN, and in accordance with Article 7.4 a lectotype may be designated. Sherff selected the sheet of this number at B as the type, but as this specimen is now appar- ently destroyed, a new lectotype is chosen here. The variation shown by Baum 847is illustrated by the shape and size of the leaves on the five duplicates seen. Hoffmann described this variation as “foliis... orbicularibus..., praeter basin integerrimam grosse crenatis vel flabellatim multilobatis, segmentis obtusis”. The original description, however, is accompanied by a figure which can be seen to possess only the former type of leaf shape. In my choice of lectotype, Rayner: Studies of Eastern Hemisphere Bidens 3 83 therefore, I have paid special attention to this diagram as most closely repre- senting Hoffmann’s concept of the species. In this regard I have chosen the specimen at G as the new lectotype due to the close resemblance between this specimen and Hoffmann’s figure. Bidens kivuensis Sherff, Bot. Gaz. 96:145. 1934. TYPE: ZAIRE. Mulungu, May 1932, J. Lebrun 5467 (LECTOTYPE [here selected]: BR; Isolecto- types: BR,F). In his original description of Bidens kivuensis, Sherff cited the type collec- tion, Lebrun 5467, as “lst and 2nd type sheets, Herb. Bruss.: 3rd type sheet, Herb. Field Mus.” As no holotype was indicated, a lectotype is here selected in accordance with Article 7.4 of the JCBN. These three sheets closely match Sherff’s original description and were clearly all used in its construction. I have decided to choose the “1st type sheet” as the lectotype, however, because this specimen is in a slightly better state of preservation than those attached to the other two sheets. Bidens kotschyi Schultz-Bip. in Walp., Repert. Bot. Syst. 6:168. 1846. TYPE: SUDAN. Nubia, Mt. Arasch-Cool, 30 Sep. 1839, K.G.T. Kotschy 79 (LECTOTYPE [here selected]: P; Isolectotypes: G|4 sheets],K[2 sheets], L,M[2 sheets],MO,S,STU,UPS,W,WAG,WU). Schultz-Bipontinus’ original description of Bidens kotschyi omits a direct citation of the specimens used by him to describe this species. The protologue does, however, include the sentence “Crescit in Nubia ad stagna pluvialia in radice orientali montis Arasch-Cool et in paludosis Cordofanis”, which was clearly obtained by Schultz-Bipontinus from the sheets used by him in its construction. At P are specimens of two widely distributed Kotschy collec- tions, numbers 79 and 91, both formerly in Schultz-Bipontinus’ own herbar- ium and which have been determined by him as B. kotschyz. The printed label on Kotschy 79 reads “Kotschyi iter Nubicum. Ad stagna pluvialia in radice orientali montis Arasch-Cool”, while that of Kotschy 91 reads “Cord- ofan: in paludosis”. No other Kotschy collections, at P or elsewhere, possess labels with similar geographic and habitat information. Clearly, therefore, specimens of these two numbers alone were used by Schultz-Bipontinus in the construction of his original description. Indeed, specimens of both these col- lections closely match this description. Sherff (1937:369) stated that Kotschy 79 was the type collection of B. kotschyi, and cited eleven sheets from seven herbaria but without indicating which of these should be taken as the type. I consider this lectotypification to be ineffective because it is contrary to the definition of a lectotype employed by Article 7.5 of the JCBN. Mesfin (1984a) has subsequently designated the P specimen of this number as the holotype. 84 PHY TOW O1Gt A volume 73(2):77-97 August 1992 From this act, it is clear that he was unaware of the equal status of the collec- tions Kotschy 79 and 91, and on these grounds alone this designation must be considered ineffective. It has been strongly put to me, however, that by one interpretation of Article 8.3 of the JCBN, the designation of a specimen as a holotype should be considered an effective lectotypification. Article 8.3 says, “For purposes of priority under Art. 8.1, designation of a type is achieved only if the type is definitely accepted as such by the typifying author, and if the type element is clearly indicated by direct citation including the term ‘type’ or an equivalent.” As Article 8.1 is referring to the designation of lectotypes (and neotypes), the first two usages of the word type in this sentence can be replaced by lectotype (or neotype) without any alteration of meaning. Clearly Mesfin has not “definitely accepted” the P specimen of Kotschy 79 as the lectotype, and thus this act of typification must be considered ineffective. A new lectotype, therefore, is required for B. kotschyz. Both the sheets at P bear single specimens possessing mature capitula and many of the exceedingly distinctive leaves known to me only from these two collections. The specimen of Kotschy 79, however, has been selected to serve as the lectotype because of its decidedly superior state of preservation. Bidens leptolepis Sherff, Bot. Gaz., 76:85, t. 9, f. a-g. 1923. TYPE: ZAIRE. Mt. Kundelungu, 10 May 1908, T. Kassner 2725 (HOLOTYPE: Bf; LECTOTYPE [here selected]: Z; Isolectotypes: BM,K, P). In his protologue of Bidens leptolepis, Sherff cited as the type the sheet of Kassner 2725 at B, now apparently destroyed. Therefore, in accordance with Article 7.4 of the ICBN, a lectotype is here selected. There is little to choose between the isotypes at BM, K, P, and Z. The latter specimen has been selected to serve as the lectotype, however, because it possesses a number of well preserved capitula showing the very darkly coloured ray florets which Sherff (1937) considered the key difference in distinguishing between this species and B. urceolata De Wild. Bidens palustris Sherff, Bot. Gaz. 76:148. 1923. TYPE: ZAIRE. Kun- delungu, 13 Mar. 1908, T. Kassner 2599 (HOLOTYPE: Bf; LECTO- TYPE [here selected]: K; Isolectotypes: BM,BR,HBG,P,Z). Sherff cited the sheet of Kassner 2599 at B, now apparently destroyed, as the type of Bidens palustris. The isotypes, housed at BM, BR, HBG, K, P, and Z, are mostly of fairly poor quality. The specimen at BM, for example, has mostly insect damaged capitula, whilst that at Z has had all its heads removed. By contrast, the specimen on the sheet at K bears two well preserved capitula, one at anthesis and a second possessing many mature fruit, and so is here selected as the lectotype. Rayner: Studies of Eastern Hemisphere Bidens 3 85 Bidens paupercula Sherff, Bot. Gaz. 76:158, t. 12, f. a-g. 1923. TYPE: TANZANIA. Kyimbila, 22 Jul. 1912, A.F. Stolz 1442(HOLOTYPE: B[2 sheets|{; LECTOTYPE [here selected]: M; Isolectotypes: B,C,G,K,L,S, STU,W,WAG[2 sheets],Z{2 sheets]). In his protologue of Bidens paupercula, Sherff cited as the type two sheets of Stolz 1442 at B. A specimen of this number is extant at B, but an attached printed label shows that this sheet was only acquired from the herbarium of Rudolf Gross in 1946, long after Sherff had described B. paupercula, and s0 it — is not considered to be part of the holotype. As both of the type sheets are apparently destroyed, a lectotype is here selected in accordance with Article 7.4 of the ICBN. Numerous duplicates of Stolz 1442 are also housed elsewhere. Four of these (at C, G, M, and W) were cited by Sherff, in addition to the type, in his monograph of Bidens (1937). It is therefore considered that the lectotype should be chosen from among these specimens which clearly correspond with Sherff’s concept of this taxon. These specimens are of variable quality although all match Sherff’s original description. The best, however, is that at M which is here selected to serve as the lectotype. Bidens praecoz Sherff, Bot. Gaz. 92:450. 1931. TYPE: TANZANIA. Lindi District, Mayanga, 15 May 1903, W.C.O. Busse 2523 (HOLOTYPE: Bt; LECTOTYPE [here selected]: EA). Bidens rubicundula Sherff, Amer. J. Bot. 41:762. 1954. TYPE: ZAM- BIA. Chizera, 11 Jun. 1953, D.B. Fanshawe F64 (LECTOTYPE [here selected]: K; Isolectotypes: BR,F,K,SRGH). Sherff cited “two type sheets” at K and an “isotype” at F of Fanshawe F64 for this species. Clearly he considered the specimens at K to constitute the holotype. These two sheets bear a number of specimens, however, and in accordance with Article 7.3 of the JCBN this designation is contrary to the definition of a holotype and must be considered ineffective. A lectotype is thus required and, following Article 7.5, this must be chosen from among the specimens at K. The “ist type sheet” bears parts of perhaps three plants possessing immature and flowering capitula, but lacking mature fruit. The solitary plant on the “2nd type sheet” also bears flowering capitula, but in addition it possesses a number of fruiting capitula bearing extremely mature cypselas. As these and other characters of the mature capitulum are of crit- ical importance in distinguishing between this species and the closely related Bidens urceolata De Wild., this sheet is here selected as the lectotype. Bidens rupestris Sherff, Bot. Gaz. 76:144. 1923. TYPE: TANZANIA. Mt. Meru, 27 Nov. 1901, C. Uhlig 750 (HOLOTYPE: Bt; LECTOTYPE [here selected]: EA). 86 PHY TOLO GTA volume 73(2):77-97 August 1992 Bidens schlechteri Sherff, Bot. Gaz. 76:146. 1923. TYPE: SOUTH AFRICA. Transvaal, Houtboschberg, 30 Mar. 1894, F.R.R. Schlechter 4745 (LEC- TOTYPE NOV. [here selected]: G; Isolectotypes: C,K). Bidens schlechteri was described by Sherff from two sheets of Schlechter 4745 formerly at B. He cited these specimens as “type in Herb. Berl., two sheets”. Later, however, Sherff (1937) treated one of these sheets as the “type”, and the other as the “cotype”. Therefore, in the absence of any of this ma- terial it is probably best to consider that Sherff did not originally designate a holotype, but later chose a lectotype from between these two sheets. With the loss of these specimens it is necessary to select a new lectotype. Of the three duplicates, none of which were cited by Sherff (1937), those at C and K have well preserved leaves, but only a few badly damaged capitula. The specimen at G, however, although only possessing poor quality foliage, has a number of capitula at various stages of development with characters closely matching those of Sherff’s original description. Because the characters of the capitulum were considered by Sherff (1937) to be of great import in distinguishing be- tween this species and the closely related B. taylorii (S. Moore) Sherff and B. kivuensis Sherff, this specimen is here selected as the new lectotype. Bidens setigera (Schultz-Bip.) Sherff var. lobata Sherff, Bot. Gaz. 91:311. 1931. TYPE: Cultivated in garden of J. Veitch and Sons, 1908, from material collected in Kenya by Capt. Diespecker (LECTOTYPE [here selected]: K; Isolectotype: K). Sherff cited the two Diespecker sheets at K, labelled “1st” and “2nd” type sheets respectively, for Bidens setigera var. lobata. As these each bear. one specimen, not obviously both from the same individual, it is clear that Sherff has not designated a holotype as defined by Article 7.3 of the JCBN, and thus the selection of a lectotype is required. The accompanying description uses only foliar characters, obtained almost exclusively from the “lst” type sheet, to distinguish between this variety and var. setigera. In addition, Sherff noted that “The difference in general aspect due to the different amounts of foliar dissection is great, but such a difference...does not appear to justify specific segregation.” Therefore, I have decided to choose as the lectotype the “1st” type sheet because this specimen possesses numerous leaves encapsulating the entire range of the varietal description, and is clearly that on which Sherff’s concept of this variety is based. The specimen attached to the “2nd” type sheet, by contrast, bears far fewer leaves of decidedly poorer quality. Bidens somaliensis Sherff, Bot. Gaz. 90:395. 1930. TYPE: ETHIOPIA. from Biddum to Volghe, 15 Sep. 1893, D. Riva & E. Ruspoli 85( 1806) (LECTOTYPE [here selected]: FT; Isolectotype: FT). Rayner: Studies of Eastern Hemisphere Bidens 3 87 Sherff cited for Bidens somaliensis, the two sheets of Riva & Ruspol 85( 1306) at FT. As it is by no means clear that the specimens attached to these sheets are part of the same individual, and therefore do not correspond with the definition of a holotype as employed by Article 7.3 of the JCBN, it is thought desirable to select a lectotype from between them. The specimen on the “1st” sheet is here selected to serve as the lectotype because it bears many capitula at various stages of development, from buds to mature fruiting heads, as well as a number of well preserved leaves, and was clearly the main element used by Sherff in the construction of his description. The other sheet, by contrast, possesses only a few capitula, the most mature of which is just at anthesis. Bidens steppia (Steetz) Sherff var. leptocarpa Sherff, Bot. Gaz. 90:392. 1930. TYPE: TANZANIA. Kyimbila, Kaningwe, 26 May 1911, A.F. Stolz 729 (HOLOTYPE: B}; LECTOTYPE [here selected]: G; Isolectotypes: B,C,G,K,LE,LU,M,S,STU,W,WAG[2 sheets],Z[2 sheets]). Sherff cited as the type of this name a sheet of Stolz 729 at B, now appar- ently destroyed. The loss of this sheet requires the selection of a lectotype. A specimen of this number is extant at B but, as in the case of Bidens paupercula Sherff, this is not considered to be the holotype because it was only acquired from the herbarium of Rudolf Gross in 1946. In his protologue Sherff also cited two “cotypes” (1.e., isotypes) at G. As these sheets are part of the origi- nally designated type material on which the name B. steppza var. leptocarpa is based, the lectotype must be selected from between them in accordance with Article 7.5 of the ICBN. Sherff distinguished the eight varieties of B. steppia by employing characters of the cypselas, as well as in addition occasionally using other capitular characters. In the case of var. leptocarpa, Sherff (1937) separated this variety from vars. steppia and elskensz Sherff by the length of the cypselas and outer phyllaries, and from var. ambacensis (Hiern) Sherff by the nature of the cypselial aristae. Therefore, because of the importance given to these characters by Sherff in defining his varietal limits, | have decided to choose as the lectotype the specimen possessing the greater number of fruiting capitula and mature cypselas. This is the sheet bearing Stolz’s “Flora Africae Oriental.” herbarium label. The capitula of this specimen, unlike those on the other sheet, are not insect damaged, and also more closely match Sherff’s detailed original description. Bidens straminoides Sherff, Amer. J. Bot. 22:706. 1935. TYPE: RWANDA. Mt. Bohanga, s.a., H.F.A. Scaetta 2272 (LECTOTYPE [here selected): BR; Isolectotype: BR). Sherff cited the two sheets of Scaetta 2272 at BR for Bidens straminoides. As this collection clearly consists of two specimens, one on each sheet, I do 88 PHY TOLOGTA volume 73(2):77-97 August 1992 not consider that he has designated a holotype, as defined by Article 7.3 of the ICBN, and thus the selection of a lectotype is necessitated. These specimens are of exceedingly poor quality and should never have been made the type of anything. They possess a few damaged fruiting capitula and withered leaves, and were clearly at a very advanced stage of development at the time of their collection. The specimen on the first sheet, which both Mesfin and Lisowski have determined as B. grantz (Oliver) Sherff, possesses parts of three leaves and one capitulum completely lacking fruit. Fortunately, the specimen on the second sheet, apart from bearing a number of leaves, possesses two complete capitula with numerous mature cypselas. This sheet is thus reluctantly selected to serve as the lectotype. Coreopsis L. Coreopsis abyssinica Schultz-Bip. var. bipinnato-partita Chiov., Annuario Reale Ist. Bot. Roma 8:185. 1904. TYPE: ETHIOPIA. Eritrea, Ghinda, Donkollo, 14 May 1892, G.A. Schwetnfurth & D. Riva 2119 (LECTO- TYPE [here selected]: FT; Isolectotypes: G,K,P,Z). Chiovenda cited seventeen collections under his brief desctiption of this variety. With the exception of Schweinfurth & Riva 2119 and A. Pappi 4101, all are unicate specimens housed at FT. Duplicates of Schweinfurth & Riva 2119 are located at FT, G, K, P, and Z, while sheets of Pappi 4101 are at FT and RO. Mesfin (1984a) transferred this variety to the rank of subspecies un- der Bidens setigera (Schultz-Bip.) Sherff. Within this new taxon he included four of Chiovenda’s syntype collections, namely A. Terracciano & Pappi 1165, G. Scotti s.n., Pappi 4101, and Schweinfurth & Riva 2119, the others being included in either B. setigera ssp. setigera or B. camporum (Hutch.) Mesfin. He claimed that Schweinfurth & Riva 2119 was the “original collection”, and designated the specimen of this number at FT as the holotype. This claim, however, cannot be substantiated by an analysis of Chiovenda’s protologue which shows that all the collections are given equal status. Indeed, an exam- ination of the original material shows that Chiovenda saw all the specimens at FT as well as the sheet of Pappi 4101 at RO. Therefore Mesfin’s first claim must be rejected. His designation of the FT sheet of Schwetnfurth & Riva 2119 as the holotype is, of course, also incorrect. It is possible, however, that this designation can be considered an effective lectotypification by refer- ence to Article 8.3 of the ICBN. I refer the reader to the discussion under B. kotschyi Schultz-Bip. for an explanation of why I consider this unaccept- able. In an attempt to maintain current usage of the name C. abyssinica var. bipinnato-partita (Rec. 7B.5 of the JCBN), I have decided to choose as the lectotype a specimen of one of the four collections included by Mesfin in B. setigera ssp. bipinnato-partita (Chiov.) Mesfin. All the specimens in question Rayner: Studies of Eastern Hemisphere Bidens 3 89 match Chiovenda’s description, but because of the existence of the duplicates of Schweinfurth & Riva 2119, I have selected the specimen of this collection at FT as the lectotype. Coreopsis badia Sherff, Bot. Gaz. 76:90. 1923. TYPE: TOGO. 1908-09, G. de Gironcourt 256 (HOLOTYPE: Bt; LECTOTYPE [here selected): P); Coreopsis baumi: O. Hoffm. in Warb., Kunene-Sambes: Exped. 419. 1903. TYPE: ANGOLA. Mambunda region, between Kuma and Kutsi, H. Baum 883 (LECTOTYPE NOV. [here selected]: W; Isolectotypes: BM, COI,G,K,M). Hoffmann cited for Coreopsis baumiz, the collection Baum 883, without des- ignating a specimen as the type. As all the extant duplicates of this collection possess his handwriting and clearly were seen by him before their distribution from B, it is unlikely that he used only one specimen to describe this taxon. It follows, therefore, that he has not complied with the definition of a holotype as employed by Article 7.3 of the JCBN, and in accordance with Article 7.4 a lectotype may be selected. The sheet of this number at B was designated the “type” by Sherff (1937), but as this specimen is now apparently destroyed, a new lectotype is required. There is little to choose between those duplicates that I have seen. The sheet at W, however, is here selected as the new lecto- type because it most closely matches Hoffmann’s protologue. This specimen possesses a solitary branched stem with many leaves in good condition, four mature capitula, and the most mature fruits of any of the six cotypes. Coreopsis crataegifolia O. Hoffm., Bot. Jahrb. Syst. 30:431. 1901. TYPE: TANZANIA. Livingstone Mts., Yawulanda Mt., 18 Apr. 1899, W. Goetze 851 (LECTOTYPE NOV. [here selected]: BM; Isolectotype: BR). In his protologue of Coreopsis crataegifolia, Hoffmann cited the collection Goetze 851, without indicating which specimen was intended to be taken as the type. Both of the extant duplicates at BM and BR bear his handwriting, and so clearly were seen by him before their distribution from B. Therefore, as it is most likely that these sheets were used in the construction of the description, it is apparent that Hoffmann has not complied with the definition of a holotype as employed by Article 7.3 of the JCBN, and in accordance with Article 7.4 a lectotype may be designated. Sherff (1937) cited the sheet at B as the type. This specimen is now apparently destroyed and consequently a new lectotype is required. At least two duplicates are in existence, one each at BM and BR. Both are clearly Hoffmann’s taxon, but the former is chosen as the new lectotype because it possesses two branches, each with well preserved leaves and flowering capitula. The BR specimen, on the other hand, is a solitary broken branch with one damaged capitulum and many senescent leaves. 90 PRY TOroGcra volume 73(2):77-97 August 1992 Coreopsis ezaristata O. Hoffm. in Engl., Pflanzenw. Ost-Afrikas, C:414. 1895. TYPE: TANZANIA. Usambara, s.a., C.H.E.W. Holst 5002 (LECTO- TYPE NOV. [here selected]: WU; Isolectotype: B). Hoffmann cited the three Holst nos. 102, 207, and 5002 for Coreopsis exaris- tata, without reference to where specimens of these numbers were housed. Therefore, these three collections are here treated as forming a syntype col- lection consisting of all the duplicate specimens. Sherff (1937) mechanically selected a sheet of the first cited collection as the type, 1.e., no. 102 at B, and this choice is rejected in accordance with Article 8.1 of the JCBN. No duplicates of Holst 102 have been located, but a sheet of no. 207 is at BM, whilst specimens of no. 5002 are at B and WU. None of these sheets bears any indication that they were seen by Hoffmann. The specimen of Holst 207 is a mere scrap and wholly unworthy of lectotypification. The sheets of Holst 5002, on the other hand, bear large leafy specimens closely resembling Hoffmann’s original description. The specimen at WU is selected as the new lectotype, Coreopsis holstii O. Hoffm. in Engl., Pflanzenw. Ost-Afrikas, C:415. 1895. TYPE: TANZANIA. T3, Tanga Region, Lushoto Distr., West Usambara Mts., Shagayu F. R., summit 2.5 km ENE of Shagayu Sawmill, 14 Mar. 1984, A. Borhidi, Sebsebe Demissew, M. Hedrén, S.T. Iversen, W.R. Mziray, & T. Pécs 84873 (NEOTYPE [here selected]: MO; Isoneotypes: ETH n.v.,K,UPS). Hoffmann cited for Coreopsis holstii the collection C.H.E.W. Holst 76, col- lected in the Usambara region of Tanzania in October 1891, again without reference to where specimens were housed. According to Sherff (1923) a sheet of this collection bearing three flowering specimens was at B. Unfortunately, this sheet is now apparently destroyed and no duplicates of this collection have been located. As I consider this taxon to be a good species, as Bidens holstz (O. Hoffm.) Sherff, a neotype is here selected in accordance with Article 7.4 of the JCBN. A small number of specimens of this species have been collected from the type locality. One of these, here selected as the neotype, is the MO specimen of Borhidi et al. 84879. This specimen, the best of all the dupli- cates of this collection, closely matches Hoffmann’s original description of this species and Sherff’s figure (1937) drawn from the type specimen. It possesses numerous capitula at anthesis and in fruit, as well as many of the exceedingly characteristic leaves. Coreopsis kilimandscharica O. Hoffm., Bot. Jahrb. Syst. 20:234. 1894. TYPE: TANZANIA. Kilimanjaro, Uschiri, 14 Jun. 1893, G.L.A. Volkens 398 (LECTOTYPE NOV. [here selected]: BM; Isolectotype: G). Rayner: Studies of Eastern Hemisphere Bidens 3 91 Hoffmann cited for Coreopsis kilimandscharica, the three collections W.L. Abbott s.n. and Volkens 398 and 537, without reference to where specimens of these numbers were housed. Sherff (1937) selected the specimen of Abbott s.n. at B as the type because this “was the first one cited by Hoffmann.” Again this is a mechanical method of lectotype selection and, by invoking Article 8.1 of the JCBN, this choice is here rejected. Specimens of all three collections were formerly at B (Sherff, 1937), but these are now apparently destroyed. Fortunately, at least one duplicate of each collection is extant elsewhere. The only sheet of Abbott s.n. is at US. This is a particularly poor specimen, hence the rejection of Sherff’s choice of lectotype. The best specimens of the syntype collection are the sheets of Volkens 998 at BM and G. These closely match Hoffmann’s original description of this taxon. The BM specimen is chosen as the new lectotype, however, because of the superior quality of its flowering capitula. Coreopsis leptoglossa Sherff, Bot. Gaz. 76:88. 1923. TYPE: ZAIRE. Lofuku River, 25 May 1908, T. Kassner 2871 (HOLOTYPE: Bt; LECTOTYPE [here selected]: Z[p.p.min.]; excl. Isolectotypes: BM,K,P,Z|p.p.maj.|). Sherff cited for Coreopsis leptoglossa, the sheet of Kassner 2871 at B. As this is now apparently destroyed, a lectotype is here designated in accor- dance with Article 7.4 of the JCBN. According to Sherff’s original description this plant possessed “Folia... bipinnata, segmentis linearibus, plerumque circ. 1 mm., rarius 1.5-2 mm. latis...Involucri bracteae basi dense aliter leviter hispidae. .., interioribus lanceolatis”. An examination of the isotypes at BM, K, P, and Z clearly shows that this number is a mixed collection. Of all the duplicates seen only one of the two specimens on the sheet at Z fits this descrip- tion. The other specimens have entire to deeply lobed or pinnatisect leaves’ with segments rarely less than 5 mm wide, outer phyllaries mostly densely hispid throughout, and inner phyllaries more or less oblong. Therefore, the right hand specimen on the sheet at Z, although a rather poor specimen with only one immature head, is here selected as the lectotype. Coreopsis lineata Klatt, Ann. K. K. Naturhist. Hofmus. 7:103. 1892. TYPE: ANGOLA. Pungo-Andongo, Jan.-Apr. 1879, A. von Mechow 131 (LEC- TOTYPE [here selected]: Z; Isolectotype: GH). Klatt cited the collection Mechow 131 for Coreopsis lineata, without desig- nating a specimen as the type. It is by no means clear which of the duplicates were used by Klatt in the construction of his protologue. As both the du- plicates at GH and Z bear annotations in his hand, however, it is apparent that they were seen by him, and were probably used, at least in part, to de- scribe this taxon. Therefore, Klatt cannot be considered to have complied with the definition of a holotype as provided by Article 7.3 of the JCBN, and 92 PHYTOL OG TA volume 73(2):77-97 August 1992 in accordance with Article 7.4, a lectotype is here selected. Sherff (1937:600) cited sheets at B and GH for this name, but did not designate either of these as the type. As the sheet at B is now apparently destroyed, the lectotype is here chosen from between the specimens at GH and Z. The latter specimen is here selected as the lectotype because it possesses two flowering capitula with many fairly mature fruit, and a number of leaves which closely match Klatt’s original description. The specimen at GH, on the other hand, has only one capitulum at anthesis, and a few damaged leaves. Coreopsis lupulina O. Hoffm., Bot. Jahrb. Syst. 30:432. 1901. TYPE: TAN- ZANIA. Usafua, Beya Mts., 27 Jun. 1899, W. Goetze 1069 (LECTO- TYPE NOV. [here selected]: P; Isolectotypes: BM,BR,K,L). Hoffmann cited for Coreopsis lupulina, the collection Goetze 1069, with- out designating a specimen as the type. As all the extant duplicates of this collection possess his handwriting and clearly were seen by him before their distribution from B, it is unlikely that he used only one specimen to describe this taxon. It follows, therefore, that he has not complied with the definition of a holotype as employed by Article 7.3 of the CBN, and in accordance with Article 7.4 a lectotype may be selected. Sherff (1936) cited as the “type spec- imen” the two sheets of this number at B. As these are now both apparently destroyed a new lectotype is required. Duplicates of Goetze 1069 are extant at BM, BR, K, L, and P. Of these, the sheet at P is here selected to serve as the new lectotype. The specimen on this sheet is the apex of a branch possess- ing many mature leaves and nine capitula at various stages of development, and shows all the important diagnostic characters attributed to this taxon by Hoffmann. Coreopsis ochracea O. Hoffm., Bot. Jahrb. Syst. 30:431. 1901. TYPE: TAN- ZANIA. Uhehe, Bweni, 11 Mar. 1899, W. Goetze 731 (LECTOTYPE NOV. [here selected]: BM). Hoffmann based this name on the collection Goetze 791, but failed to des- ignate a specimen as the type. The only extant duplicate, at BM, bears Hoff- mann’s handwriting, and thus clearly was seen by him before its distribution from B. Sherff (1925) designated a sheet of this number at B as the type. It is likely that both this specimen and the one at BM were used by Hoffmann in the construction of his description. Therefore it follows that Hoffmann has not complied with the definition of a holotype as employed by Article 7.3 of the ICBN, and in accordance with “Article 7.4 a lectotype may be designated. The apparent loss of the sheet at B requires the selection of the specimen at BM as the new lectotype. Rayner: Studies of Eastern Hemisphere Bidens 3 93 Coreopsis pinnatipartita O. Hoffm., Bot. Jahrb. Syst. 30:432. 1901. TYPE: TANZANIA. Usafua, Poroto Mt., 17 Jun. 1899, W. Goetze 1041 (LEC- TOTYPE NOV. [here selected]: P; Isolectotypes: BM,K,L,Z). Hoffmann cited the two collections, Goetze 1041 and J. Buchanan 380, in his protologue of Coreopsis pinnatipartita. The former was cited in large print immediately following the Latin and German descriptions. The latter, however, was cited in a smaller type face in some notes at the end of the protologue. The first sentence of this reads: “Ein von Buchanan im Jahre 1891 im Nyassaland gesammeltes Exemplar (n. 380) gehort offenbar zu derselben Art.” This is followed by a perfunctory description in German. This evidence alone would suggest that Hoffmann based his concept of C. pinnatzpartita on the collection Goetze 1041, and that Buchanan 980 was considered by him to be just another example of this species. In addition, examination of the duplicates of both collections clearly shows that the species description was based solely on the specimens of Goetze 1041. For this reason | consider that Goetze 1041 should be taken as the type collection, with Buchanan 380 treated as a paratype collection. No indication was made by Hoffmann as to where specimens of Goetze 1041 were housed. As all the duplicates at BM, K, L, P, and Z bear his handwriting, and were clearly, therefore, seen by him before their distribution from B, it is unlikely that only one specimen was used by him to describe this taxon. It follows, therefore, that he has not complied with the definition of a holotype as employed by Article 7.3 of the JCBN, and in accordance with Article 7.4 a lectotype may be selected. Sherff (1936) cited the two sheets of Goetze 1041 at B as the type, but as these are both now apparently destroyed a new lectotype is required. Wild (1967) selected the K sheet of the paratype collection Buchanan 380 as the neotype. This choice is rejected, however, because the existence of duplicates of Goetze 1041 deem this selection ineffective (Article 8.1{a] of the JCBN). The sheet at P is here selected to serve as the new lectotype because of the high quality of its leaves and capitula, some of which bear fairly mature fruit. Coreopsis scabrifolia Sherff, Bot. Gaz. 76:86. 1923. TYPE: ZAIRE. Kun- delungu, 15 May 1908, T. Kassner 2776 (HOLOTYPE: Bt; LECTO- TYPE [here selected]: BM; Isolectotypes: BM,K,P,Z). When Sherff described ‘Coreopsis scabrifolia from the collection Kassner 2776, in addition to the holotype at B, he cited “two co-type sheets” at BM. As the B sheet of this number is now apparently destroyed, I have selected as the lectotype one of these two sheets in accordance with Article 7.5 of the JCBN. The sheet chosen is that bearing Kassner’s collecting label. This specimen possesses many capitula at various stages of development, and shows more clearly than any of the other duplicates the dimorphic leaf shape exhibited by this collection and described by Sherff. 94 PHY TOLO GTA volume 73(2):77-97 August 1992 Coreopsis steppia Steetz in Peters, Naturw. Reise Mossambique 6:496. 1864. TYPE: MOZAMBIQUE. Mocuba, Lugela District, Namagra, 1946- 7, H.G. Faulkner Kew 10 (NEOTYPE [here selected]: K; Isoneotypes: BM,BR,COI,EA,FT,G(2 sheets),K(2 sheets),P,PRE,S,SRGH). Steetz cited for Coreopsis steppia the solitary specimen, mounted on two sheets, of W.C.H. Peters 57 at B, collected from Rios de Sena, Mozambique, between 1842 and 1848. Because I consider C. steppia to be a good species, as Bidens steppia (Steetz) Sherff, the apparent loss of the Peters specimen necessitates the selection of a neotype in accordance with Article 7.4 of the ICBN. In choosing a neotype I have attempted to find a collection from near the type locality, which closely matches Steetz’s long and detailed diagnosis and description, and which is also well distributed throughout many of the world’s major herbaria. A collection which amply fulfills these requirements is Faulkner Kew 10. Many of the specimens of this number would make perfectly serviceable neotypes. One of the best, however, is that at K labelled “Sheet 1”, which is here selected. This plant possesses most of the characters used by Steetz in his original description, including the diagnostically important characters of the mature fruit. Guizotia Cass. Guizotia bidentoides Oliver & Hiern in Oliver, Fl. Trop. Afr. 3:386. 1877. TYPE: MALAWI. 3 miles from river Shire, Aug.-Sep. 1861, C.J. Meller s.n. (LECTOTYPE [here selected]: K). Guizotia bidentoides was described by Oliver & Hiern from a sheet at K cited in their protologue as “Manganja Hills, 500-3,000 ft. alt., Kirk!”. The sheet in question, however, bears three separate gatherings. The first, collected by Meller, bears a printed “Livingstone’s Zambesi Expedition” label saying “About Lat. 16 South, Long. 35E. Aug. & Sept. 1861”, with the additional hand written note “From a hill 500ft - 3 miles from river Shire”. The second, also collected by Meller, bears the printed label “Manganja Hills. Sept. to Nov. 1861”. The third has attached a hand written note by J. Kirk saying “Karizakwwo, Entr. to Bangue Pass 3000ft.” Clearly, therefore, it can be seen that all three collections were used to construct the protologue. The Kirk specimen cannot be considered to be the holotype because, by the definition employed in Article 7.3 of the JCBN, the holotype is “the one specimen. . . used by the author or designated by him as the nomenclatural type.” It is clear from the above that Oliver & Hiern attributed the whole sheet to Kirk, instead of just the small part actually collected by him. Thus, Oliver & Hiern cannot be considered to have designated one specimen as the holotype, and hence the need for the selection of a lectotype. It is also clear that all three plants were Rayner: Studies of Eastern Hemisphere Bidens 3 95 used in the construction of the description. Oliver’s notes on the sheet refer mostly to the Kirk specimen and Meller specimen from the Manganja Hills. Unfortunately, both these collections are badly damaged. The measurement of leaf size in the published description, however, is clearly taken from the other Meller collection which is here selected as the lectotype. This specimen, which corresponds closely with Oliver & Hiern’s original description, is the apical portion of a branch bearing a number of well preserved leaves and capitula. Microlecane (Schultz-Bip.) Benth. & Hook. f. Microlecane carinata Hutch., Bull. Misc. Inform. Kew 1916:41. 1916. TYPE: ETHIOPIA. Gondar and vicinity, s.a., R.E. Massey 74 (LECTOTYPE [here selected]: K). Hutchinson published his description of Microlecane carinata in 1916 citing the three syntypes Massey 74, G.H.W. Schimper 1386, and J.A. Grant s.n. There are sheets of all these numbers at K bearing Hutchinson’s handwriting. Previously, Hoffmann (1906) had cited the collections Schimper 1386 and F.G. Rohlfs & A. Stecker s.n. for his Coreopsis schimperi. Mesfin (1984b) has sub- sequently claimed that because of the inclusion of “type material of another ‘name’, M. carinata Hutch. is an illegitimate name according to article 63 of the International Code of Botanical Nomenclature.” However, Article 63.1 clearly states that in order for a name to be illegitimate it must include “all syntypes” of another name, instead of, as in this case, only one specimen of the syntype collection. Thus it can be seen that M. carinata is a legitimate name. The three syntypes of this name are well preserved specimens, pos- sessing capitula with mature fruit, and were apparently used fairly equally by Hutchinson in the construction of his diagnosis and description. The selection of the lectotype, however, has been determined by factors governed by articles of the ICBN. The specimen of Schimper 1386 at K has been designated the new lectotype of C. schimperi O. Hoffm. by Mesfin (1984a), and so is excluded from selection by Article 63.1. The specimen of Grant s.n. has been included in B. negriana (Sherff) Cuf. by Mesfin (1984a), and so in order to preserve the current usage of the name M. carinata is not here considered a candidate for lectotypification (Rec. 7B.5). Thus the only specimen available for selection as the lectotype is Massey 74, which is here chosen. ACKNOWLEDGMENTS I would like to thank the directors and curators of the following herbaria for placing material at my disposal, and for undertaking or organizing searches of 96 PHY TOLOGIA volume 73(2):77-97 August 1992 their herbaria for type specimens: B,BM,BP,BR,C,CAIM,CGE,COI,DSM,E,EA, F,FHO,FR,FT,G,GB,GE,GH,GOET,GRA,H,HBG,ILCA,JE,K,KIEL,L,LD,LE, LISC,LISU,LMU,LY,M,MHU,MO,MPU,MRSC,NH,NY,OXF,P,PAD,PH,POM, PR,PRE,RO,S,SRGH,STU,TO,TOM,TUB,UC,UPS,US,UZL,W,WAG,WRSL, WU,YBI,Z. In addition, my gratitude is extended to C. Piggott (K), M. Tardelli (FT), and G. Wagenitz (GOET) for providing examples of autographs. Lastly, my sincerest thanks go to R.K. Brummitt (K) for finding time within his exceedingly busy schedule to criticize the manuscript and to endure detailed and extensive discussions of nomenclatural problems. BIBLIOGRAPHY Frizzell, D.L. 1933. Terminology of types. Amer. Midl. Naturalist 14:637- 668. 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. Hoffmann, O. 1906. Compositae africanae. IV. Bot. Jahrb. Syst. 38:196-211. Mesfin Tadesse. 1984a. The genus Bidens (Compositae) in NE tropical Africa. Symb. Bot. Upsal. 24(1):138pp. Mesfin Tadesse. 1984b. Microlecane (Sch. Bip.) Benth.: a congener of Bidens (Compositae - Heliantheae). Nordic J. Bot. 4:737-746. Schultz-Bipontinus, C.H. 1846. Compositae. In W.G. Walpers, Repertoritum Botanices Systematicae 6:88-367. Leipzig, Germany. Sherff, E.E. 1915. Studies in the genus Bidens. II. Bot. Gaz. 59:301-316. Sherff, E.E. 1923. New or otherwise noteworthy Compositae. Bot. Gaz. 76:78-94. Sherff, E.E. 1925. New or otherwise noteworthy Compositae. II. Bot. Gaz. 80:367-389. Sherff, E.E. 1931. New or otherwise noteworthy Compositae. VI. Bot. Gaz. 91:308-319. ’ Sherff, E.E. 1936. Revision of the genus Coreopsis. Field Mus. Nat. Hist., Bot. Ser. 11:277-475. Rayner: | Studies of Eastern Hemisphere Bidens 3 97 Sherff, E.E. 1937. The genus Bidens. Field Mus. Nat. Hist., Bot. Ser. 16:1- 709. Wild, H. 1967. The Compositae of the Flora Zambesiaca area, 1. Kirkia 6:1-62. Phytologia (August 1992) 73(2):98-116. EIGHT NEW TAXA OF DROSERA FROM AUSTRALIA Allen Lowrie! & Sherwin Carlquist? 16 Glenn Place, Duncraig, Western Australia 6023, AUSTRALIA ?Rancho Santa Ana Botanic Garden & Department of Biology, Pomona College, Claremont, California 91711 U.S.A. ABSTRACT Six new species and two new subspecies of Drosera (Droseraceae) are described. Drosera citrina, D. lasiantha, D. nivea, D. silvicola, and D. stelliflora are pygmy droseras from southwestern Australia; D. bicolor and D. macrophylla subsp. monantha are tuberous droseras from southwestern Australia. Drosera whittakeri subsp. aberrans is a tuberous species from South Australia and Victoria. Features that distinguish these taxa from nearest relatives are presented, along with other data. KEY WORDS: Australia, Drosera, Droseraceae, South Australia, Victoria, Western Australia INTRODUCTION Since the publication of the first two volumes of Carnivorous Plants of Australia (Lowrie 1987, 1989), many new Drosera taxa have been discovered. Most of these have been reported by Lowrie & Carlquist (1990), Lowrie & Marchant (1992), and Marchant & Lowrie (1992). In the present paper, an additional eight taxa are described. The southwest of Western Australia is remarkable for the large number of droseras found there. With the current additions, 104 Drosera taxa are known from this area; of these, 73 are recognized at the species level. The reasons for this high level of speciation in the region may relate primarily to edaphic factors, and the insular way in which soil types suitable for Drosera are distributed. Some of the species are known to be restricted to special soil conditions. For example, D. parvula Planchon is found only in deep white sand, 98 Lowrie & Carlquist: New Drosera from Australia 99 whereas D. citrina Lowrie & Carlquist is known only from yellow sand. Drosera barbigera Planchon grows in a mix of laterite and silica sand on hilltops; when this species grows on flat ground, investigation generally shows that the area is a remnant of an eroded hilltop, a barely perceptible rise. Drosera localities for the world at large can be characterized as acidic. The silica sands in which Drosera occurs in Australia are likely acidic to various degrees, and the soil texture and topography (valley vs. hill) may dictate dif- ferent degrees of water availability that also govern the range of Drosera taxa. Rainfall in southwestern Australia decreases markedly from the southwestern tip of the continent toward the interior, the north, and the east. Western Australia’s first resident botanist, James Drummond (1784-1863), explored and collected extensively for most of his life in the southwest of Western Australia. He observed that plants in this region seemed restricted to limited areas, as he stated in an 1844 letter to W.J. Hooker in Kew: “When we consider the great numbers of species known to grow in only one spot, and these spots exhibiting no very remarkable conditions of land or aspect, etc., it is impossible to calculate the amount of novelty which might reward the researches of a naturalist.” The genetic and cytological patterns underlying the species of Drosera need elucidation. NEW SPECIES AND SUBSPECIES The new species are presented first. For convenience of comparisons, the first five species presented are pygmy droseras, followed by the new tuberous species. New tuberous subspecies conclude the paper. Drosera citrina A. Lowrie & S. Carlquist, spec. nov. HOLOTYPE: AUS- TRALIA. In deep yellow sand between low shrubs on Brand Highway, 24 km S of Regan’s Ford, Western Australia, 9 October 1983, Allen Lowrie 83/011 (PERTH); Isotype: RSA. Herba perennis 1.5 cm in diametro, folia rosulata. Caulis 1 cm longi, stipularum foliolorum dense vestitus. Lamina late elliptica, 1.5 cm longa, 1.2 mm lata, cupulata. Petiola anguste lanceolata, 5 mm longa, ad basi 1 mm lata, ad apice 0.3 mm. Gemma stip- ularum ovoidea-globosa, 4.5 mm longa, 5.5 mm in diametro, mu- cronasta. Stipula trilobata, 4.5 mm longa, 3 mm lata, lobus cen- tralis in 3 segmenti divisus, segmenti setosi; lobi laterales serrato- marginati, apex bisetosum. Inflorescentia racemosa. Scapi 1-2, 4 cm longi, minute glandulosi, ad apicem dense glandulosi. Flores ca. 12 per scapi. Pedicelli 2.5-3.0 mmlongi, glandulosi, in fructibus semi-erecti, ebracteolati vel ad apicem inflorescentiae bracteolati, bracteolae 2.5 mm longae. Sepala ovata, 2 mm longa, 1.2 mm lata, 100 PBY PTODG:GTA volume 73(2):98-116 August 1992 sparse glandulosa, margine integra sed ad apicem serrata. Petala lutea, e basi alba, obovata, 5 mm longa, 3.5 mm lata. Ovarium turbinatum, 0.6 mm longa, 1 mm in diametro. Styli 3(-4), alba, filiformes, patentia, 3.0-4.5 mm longi, e basi 0.15 mm in diametro, acuminati. Fructus ignotus. Fibrous rooted perennial herb, forming a compact rosette 1.5 cm in diam- eter. Stem to 1 cm long, covered with the remains of the previous season’s leaves and stipules. Lamina broadly elliptic, 1.5 mm long, 1.2 mm wide, deeply cupped. Petiole narrowly lanceolate in outline, 5 mm long, 1 mm wide near the base, tapering to 0.3 mm wide at the apex. Stipular bud broadly ovoid, globelike with a small pointed projection at the apex, 4.5 mm long, 5.5 mm in diameter. Stipules trilobed, 4.5 mm long, 3 mm wide, the central lobe divided into three segments, each segment narrowing to setae; upper outer margins of lateral lobes serrate, apex with two short setae, one seta equaling in length the central lobe present on the inner margin. Inflorescence a raceme. Scapes one or two, to 4 cm long, covered with minute glandular hairs, densely glandular near apex. Flowers 12 or more per inflorescence. Pedicels 2.5-3.0 mm long, glandular, semierect in fruit, bracteoles mostly absent; when present, at the apex of the inflorescence and 2.5 mm long. Sepals ovate, 2 mm long, 1.2 mm wide, surface sparsely covered with glandular hairs, margins entire except for the serrate apex. Petals lemon yellow, white at base, obovate, 5 mm long, 3.5 mm wide. Ovary at anthesis turbinate, 0.6 mm long, 1 mm in diameter. Styles three, sometimes four, style (including stigma) white, filiform, spreading horizontally, 3.0-4.5 mm long. 0.15 mm in diameter at the base but tapering toward the apex. Fruit and seeds unknown. _ Drosera citrina (Fig. 1) belongs to subgenus Rorella (DC.) Diels section Lamprolepis Planchon. Drosera nivea Lowrie & Carlquist is considered to be the closest relative (see discussion below under that species). Drosera citrina may be mistaken for fruiting or nonflowering specimens of D. pycnoblasta Diels, a species with a smooth globelike stipular bud. Drosera pycnoblasta differs from D. citrina by having no pointed projection at the apex of its stipular bud and by having trilobed stipules with entire margins. Drosera citrina is illustrated in the first printing of Carnivorous Plants of Australia, Vol. 2 (Lowrie 1989, pp. 154-157) as D. rechingerz Strid; this will be corrected in subsequent printings. For “Drosera coolamon,” a nomen nudum in Lowrie (1989, pp. 38-41), the earlier name D. rechingeri should have been used. Drosera lasiantha A. Lowrie & S. Carlquist, spec. nov. HOLOTYPE: AUS- TRALIA. In loamy laterite soils on open areas with dwarf jarrah (£u- calyptus marginata) woodland on the higher scree slopes, top of crest, west end of the Porongorup Range, Porongorups, Western Australia, 25 November 1991, Allen Lowrie 523 (PERTH); Isotype: RSA. Lowrie & Carlquist: New Drosera from Australia 101 A. Locorie /285. Figure 1. Drosera citrina. A. Habit of plant in flower. B. Sepal. C. Petal. D. Ovary with styles. E. Stigmatic portion of style, enlarged. F. Leaf. G. Section of petiole. H. Stipule. Scales = 1 mm. 102 PHY? OL .OGirA volume 73(2):98-116 August 1992 Herba perennis, folia rosulata, infima deflexa. Caulis ad 4 cm longi, stipularum petiolorumque rudimentis dense vestitus. Lam- ina elliptica, 3 mm longa, 2 mm lata, petioli 7.5 mm longi, e basi 1 mm lati. Gemma 4 mm longa, 2.5 mm lata. Stipulae scariosae, trilobatae, subdivisi, ad apicem trisetosae, margine serrata. Inflo- rescentia racemiformis, 2 cm longa. Scapi 1-4, dense glandulosi et lanati. Flores ca. 12. Pedicelli 1.3 mm longi, semi-patenti, ebracte- olati. Sepala elliptico-ovata, 4 mm longa, 1.3-2.0 mm lata, margine serrata, albolanata. Petala atrorosea, obovata, 6.5 mm longa, 5.5 mm lata. Ovarium turbinatum, 0.8 mm longum, | mm in diametro ad anthesim. Styli 3, albi, semi-erecti, 0.7 mm longi. Stigma rosea, clavata, e basi falcata, 1 mm longa. Capsula ignota. Rosette perennial forming a compact convex rosette 1.5 cm in diameter, leaves reflexed. Stem 4 cm long, covered with the withered remains of the previous season’s growth and supported below on prop roots. Active insect trapping leaves about 10 in number. Lamina elliptic, 3 mm long, 2 mm wide. Petiole 7.5 mm long, 1 mm wide at the base, narrowed to 0.5 mm at the apex. Stipule bud ovoid, 4 mm long, 2.5 mm wide, setose. Stipules three lobed, 7 mm long, 4 mm wide, the central lobe divided into three segments, each segment in turn divided into more segments narrowing to setae, lateral lobes serrate on the outer margin with the apex divided into three short setae; on the outer margin two setae are present. Inflorescence a raceme to 2 cm tall. Scapes 1-4, densely covered with short glandular hairs and white nonglandular woolly hairs. Flowers about 12 in number. Pedicels 1.3 mm long, semierect in fruit, bracteoles absent. Sepals narrowly ovate, 4 mm long, 1.3-2.0 mm wide, margins irregularly serrate, surface covered with white, nonglandular long woolly hairs. Petals dark pink, obovate, 6.5 mm long, 5.5 mm wide. Ovary turbinate, 0.8 mm long, 1 mm in diameter at anthesis. Styles three, white, semierect, 0.7 mm long. Stigmas pale to dark pink, clavate, falcate near the base, 1 mm long. Drosera lasiantha (Fig. 2) belongs in subgenus Rorella section Lamprolepis. Drosera scorpioides Planchon and D. dichrosepala Turcz. are considered the closest relatives of D. lasiantha, but differ from it in the features cited below. It is locally abundant, and the population extends into Porongorups National Park, where it is safe from agriculture. Scapes glandular, petals elliptic: .... s E. Enlarged portion of style and stigmas. F. Leaf. G. Section of petiole. Scales = 1 mm. 116 PHYTO LO Gra volume 73(2):98-116 August 1992 LITERATURE CITED Lowrie, A. 1987. Carnivorous Plants of Australia, Vol. 1. University of Western Australia Press, Nedlands, Western Australia. Lowrie, A. 1989. Carnivorous Plants of Australia, Vol. 2. University of Western Australia Press, Nedlands, Western Australia. Lowrie, A. & S. Carlquist. 1990. A new species of tuberous Drosera from Western Australia. Phytologia 69:160-162. Lowrie, A. & N. Marchant. 1992. Four new Drosera taxa from Western Australia. Nuytsia (in press). Marchant, N. & A. Lowrie. 1992. New names and combinations in 34 taxa of Western Australian tuberous and pygmy Drosera. Kew Bulletin (in press). Phytologia (August 1992) 73(2):117-118. NEW COMBINATIONS IN APPENDICULA BLUME AND OCTARRHENA THWAITES (ORCHIDACEAE) IN AUSTRALIA Mark A. Clements & David L. Jones Australian National Botanic Gardens, G.P.O. Box 1777, Canberra, A.C.T., AUSTRALIA 2601 ABSTRACT Embryological and morphological studies indicate that two species, Podochilus australtensis and Phreatia baileyana, have been misplaced in their respective genera. The new combinations Appendicula aus- traliensis (Bailey) M. Clements et D. Jones, and Octarrhena pusilla (Bailey) M. Clements et D. Jones are provided to reflect the relation- ships of these species. KEY WORDS: Appendicula, Octarrhena, Podochilus, Phreatza, Australia, Orchidaceae INTRODUCTION This is the first of a series of papers required to adjust the nomencla- ture of Australian orchids following detailed studies by the authors. Prelimi- nary examination of the developmental embryology of species in the subtribes Podochilinae and Thelasiinae has revealed errors in the generic placement of some of the Australian taxa. Associated examination of fresh flowers of Podochilus australiensis (Bailey) Schltr. shows the pollinarium to consist of six pollinia. This is a characteristic of the genus Appendicula Blume, not of Podochilus Blume which has four pollinia. Parallel studies on material of Phreatia baileyana Schitr. shows that plants are sympodial with indeterminate growths and the flowers lack a mentum, both of which are characters of the genus Octarrhena Thwaites not of Phreatia Lindley. In light of these studies new combinations are made: 117 118 BBY POLO GLA volume 73(2):117-118 August 1992 Appendicula australiensis (Bailey) M. Clements et D. Jones, comb. nov. BASIONYM: Eria australiensis Bailey, Occas. Pap. Queensl. Fl. 1:8. 1886.; Podochilus australiensis (Bailey) Schltr., Repert. Spec. Nov. Regni Veg. Beih. 3:316. 1907., syn. nov. TYPE: AUSTRALIA. Queensland: Johnstone River, T.L. Bancroft s.n. (HOLOTYPE: BRI!). Illus. pl. 428 (Nicholls 1969). Octarrhena pusilla (Bailey) M. Clements et D. Jones, comb. nov. BA- SIONYM: Oberonia pusilla Bailey, Votes & Proc. Legis. Ass. Qld. 4:1227. 1889. and Rep. Bellenden Ker Exped. 23. 1889. Phreatia pusilla (Bailey) Rolfe, Orchid Rev. 11:344 (1903), non (Blume) Lindley 1830. Phreatia batleyana Schlitr., Repert. Spec. Nov. Regni Veg. Beih. 9:433. 1911., syn. nov. TYPE: AUSTRALIA. Queensland: Bellenden Ker at about 4000 ft, F.M. Bailey s.n. (HOLOTYPE: K!). Illus. p. 651 (Dockrill 1969). ACKNOWLEDGMENTS We wish to thank Surrey Jacobs, Royal Botanic Gardens, Sydney and Ben Wallace, Australian National Botanic Gardens for refereeing this paper. REFERENCES Dockrill, A.W. 1969. Australian Indigenous Orchids, vol. 1. Society For Growing Australian Plants, Sydney, Australia. Nicholls, W.H. 1969. Orchids of Australia. Nelson, Melbourne, Australia. Phytologia (August 1992) 73(2):119-123. VARIATION IN ERIGERON CHIANGII (ASTERACEAE: ASTEREAE), WITH A NEW VARIETY Guy L. Nesom Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT Erigeron chiangii var. lamprocaulis, var. nov., comprises a group of populations with linear leaves and the stems completely glabrous below the peduncular portions. Erigeron chiangit (including the new and typical variety) is restricted to Coahuila and adjacent Nuevo Leon, México; a distribution map of both taxa is provided. Erigeron scopart- otdes, the sister species of E. chiangti, is endemic to Tamaulipas. KEY WORDS: Erigeron, Asteraceae, Astereae, México In a paper formally describing Erigeron chiangii Nesom (Nesom 1979), I noted that several collections of atypically linear leaved plants primarily from the northern portion of its range could be distinguished from the rest of the species. After the accumulation of additional collections (TEX), reexamination of most of the original set of specimens of FE. chiangii, and observations of populations in the field, it is apparent that two taxa can be distinguished. The difference between them is primarily in vestiture and leaf shape, and although it is not clear that they intergrade, the relationship between them is so close that they are considered here as varieties of a single species. Erigeron chiangii Nesom var. lamprocaulis Nesom, var. nov. TYPE: MEXICO. Coahuila: Mpio. Melchior Muquiz, ca. 130 road km NW of Muquiz along Coahuila Hwy 2-A (Muquiz-Boquillas), then into Sierra La Encantada; bottom of deep, N-S trending canyon on SW side of La Encantada basin, Sierra Buenavista of the Sierra La Encantada, on SE side of road going up into mining area of the sierra; 102° 30’, 28° 34’; ca. 1650 m, 3 Jun 1992, G. Nesom 7408 with M. Mayfield (HOLO- TYPE: TEX; Isotypes: ANSM,ARIZ,ASU,CAS,CHAPA,COLO,ENCB, F,GH,MEXU,MO,NY,OBI,PATZ,RM,S,SRSC,UAT,UC,UNM,US). 119 120 PHY TOLO GIA volume 73(2):119-123 August 1992 Differt a Erigeronte chiangit Nesom typicus caulibus omnino glabris et foliis lineari-filiformibus. Additional collections examined: MEXICO. COAHUILA. Sierra del Car- men: higher elevations in the Sierra Jardin, 1 Sep 1966, Flyr 1210 (SMU); Mina El Popo, ca. 2 km S of Canon El Diablo, 29 Jul 1973, Johnston et al. 11922 (LL); canyon descending E from high pass N of Sierra Jardin, 6 mi E of Rancho El Jardin, 23 May 1968, Powell et al. 1602 - voucher for chromosome count of n=18 (TEX). Serranias del Burro: ca. 10 km SW of R. San Miguel at Ejido Santa Eulalia in Canon de los Burros, NE side of Serranias del Burro, 2 Jun 1972, Chiang et al. 7519C (LL); ca. 1 mi N of “La Laguna,” Canon del Mulato, 8 Sep 1963, Gould 10596 (TAES). Sierra Santa Rosa: Rancho Agua Dulce, E slope of the Sierra de San Manuel, 30 Jun 1936, Wynd & Mueller 387 (MO). Sierra de la Encantada: ca. 140 road km NW of Muquiz, Cuesta del Plomo area, ca. 6 road km S along well-maintained road originating just SE of Coahuila Hwy 2-A (Muquiz-Boquillas) crest at Cuesta del Plomo, 102° 29’ 30”, 28° 40’ 30”, area of scattered pine-oak woodlands with Yucca and Dasylirion; ca. 1820 m, 4 Jun 1992, G. Nesom 7461 with M. Mayfield (ANSM,MEXU,TEX). Sierra de la Gavia: Canon de la Gavia, S of Rancho de la Gavia, 2-3 Aug 1973, Johnston et al. 12085B (LL). NUEVO LEON. Mpio. Villaldama, Sierra Gomas in Canyon E] Alamo: northern exposed riparian community of Quercus-Ostrya-Acer with Tilia, Cornus, Carya, and Abzes on higher slopes, limestone derived soil, 1100-1400 m, 15 Aug 1988, Patterson 6694 (TEX) and 1850-2150 m, 15 Aug 1988, Patterson 6749 (TEX). The two varieties of Erigeron chiangii are distinguished by the following contrasts: a. Stems minutely glandular near the heads, otherwise glabrous; basal leaves deciduous by flowering, cauline leaves all linear, 2-6 cm long, 0.5(-1.0) mm wide, at least the upper cauline leaves with blades glabrous except for the sparsely spreading ciliate margins; 750-1820 m; May-Aug. ..... F cra ate aia a ovaoietahety oe arsine Wtietie osu wlan aseeree weaves seeseib le Ss var. lamprocaulis a. Stems glandular from top to base, at least the basal half also sparsely to moderately hirsute-pilose with stiff to loose nonglandular hairs, rarely with only nonglandular hairs from base to tip; basal leaves often persis- tent, the cauline oblanceolate on lower half of stem, (1-)2-4 cm long, 1-4 mm wide, linear-lanceolate above, the blades moderately hirsute-pilose and usually markedly glandular as well; 1300-2500 m; (Mar-)Apr-Sep (Oct) a 250 x ee ed thie ai) stators Sic ie MASA ate wipe Rial cee var. chiangit The two varieties differ in several, apparently genetically unrelated, mor- phological features, but the most consistent difference is in vestiture. While plants of var. lamprocaulis can always be distinguished by its lack of stem hairs, Nesom: Variation in Erigeron chiangi Q. 4a 2 TEX CHI O y COAHUILA ) ez O O @ UR 2 NUE i O O var. chiangii Q var. lamprocaulis Map 1. Distribution of the varieties of Eriveron chiaugii. Map 1. Distribution of the varieties of Erigeron chiangit. 121 122 PHY TOLO GTA volume 73(2):119-123 August 1992 considerable variation in vestiture exists within var. chiangit, particularly on the stems, where the degree of glandularity and the density of nonglandular hairs are variable. The stem surfaces of both species are shiny, but those of var. lamprocaulis often appear more so because of their lack of vestiture. Where the geographic ranges of the two taxa are known to overlap closely (in the Sierra de la Encantada; Map 1), var. chiangw occurs in drier, more exposed habitats (7.e., cracks, crevices, and ledges of rock outcrops and in shallow, stony soil); var. lamprocaulis in the same area grows in deep to shallow soil in considerably more protected sites, commonly at least partially shaded ones. B.L. Turner in 1979 first recognized the distinctiveness of Erigeron chi- angit, but he relinquished its description to me. The type specimen he had chosen represented the element described and named here (var. lamprocaults); my selection (a reselection) of the type established the concept of typical E. chiangit. That the two taxa were not originally distinguished taxonomically is an indication of their close similarity. Erigeron chiangii is most closely related to E. scoparioides Nesom (Ne- som 1989, 1992). Both species are especially unusual in the genus in their production of a rhizomatous root system without a central axis. They are further distinguished by their stems and leaves with a distinctively shiny sur- face texture and their narrow, entire, cauline leaves, erect buds, and strongly graduated phyllaries. Erigeron scoparioides is set apart in its completely eg- landular vestiture of short, upwardly appressed hairs (rare or absent on the phyllaries) and its apparent complete lack of basal leaves. Although Nesom (1979, 1992) has speculated that this species pair may be related to the E. fo- hosus Nutt. group (centered primarily in California), to a strong degree they appear to be geographically and morphologically isolated from any putative relatives. ACKNOWLEDGMENTS I thank Dr. B.L. Turner and Mark Mayfield for their review of the manuscript. LITERATURE CITED Nesom, G.L. 1979. A new species of Erigeron (Compositae) from Coahuila. Sida 8:65-70. . 1989. Four new species of Erigeron (Compositae: Astereae) from northern México. Phytologia 66:488-495. Nesom: Variation in Erigeron chiangit 123 . 1992. Revision of Erigeron sect. Linearifolii (Asteraceae: Aster- eae). Phytologia 72:157-208. Phytologia (August 1992) 73(2):124-136. NOMENCLATURAL NOTES FOR THE NORTH AMERICAN FLORA. XI John T. Kartesz & 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 Epilobium brachycarpum Pres] is accepted to be the correct name in place of E. paniculatum Nutt. ez Torr. & Gray. The disposition of Websteria confervoides (Poir.) Hooper is discussed. The quadrino- mial Sidalcea malviflora (DC.) A. Gray ez Benth. subsp. lacintata C.L. Hitchc. var. lacintata C.L. Hitchc. is recognized as two trinomials: Stdal- cea malviflora subsp. laciniata C.L. Hitchc. and Sidalcea malviflora var. laciniata C.L. Hitchc. The authorship of Amelanchier alnifolia (Nutt.) Nutt. er M. Roemer, A. pumila (Torr. & Gray) Nutt. er M. Roemer, Aronia arbuttfolia (L.) Pers., Choenomeles japonica (Thunb.) Lindl. ez Spach, Malus pumila C. Bauhin ez P. Mill., M. sieboldit (Regel) Re- hder, M. sylvestris P. Mill., and the new Vitaceae names proposed in Gray’s 1897 work is discussed. “Senna artemisioides (DC.) Kartesz & Gandhi” is treated as an isonym of S. artemisioides (DC.) Randell. One new combination is proposed: Spiranthes confusa (Garay) Kartesz & Gandhi. ._ KEY WORDS: Cyperaceae, Fabaceae, Malvaceae, Onagraceae, Orchidaceae, Rosaceae, Vitaceae, Amelanchier, Aronia, Choenome- les, Epilobium, Malus, Senna, Sidalcea, Spiranthes, Websteria, Bailey, Gray INTRODUCTION Continuing with the “NOMENCLATURAL NOTES FOR THE NORTH AMERICAN FLORA” (Kartesz &-Gandhi 1989, 1990a, 1990b, 1990c, 1991a, 1991b, 1991c, 1991d, 1992a, 1992b), an eleventh note in the series is presented here toward advancing our understanding of North American plant names. 124 Kartesz & Gandhi: Nomenclatural notes XI 125 CYPERACEAE Websteria confervoides Scirpus confervoides Poir. is a leafless, submerged aquatic found through- out the tropics and subtropics. It has a diffuse growth habit, slender terete stems, capillary branches in pseudo-whorls, and pedicelled ascending spikelets. Each spikelet has two distichous scales, with the upper scale subtending 6-11 retrorsely barbed bristles, 3 stamens, and an ovary with two style branches. Because of its unique morphology, its taxonomic disposition has been in dis- pute, as discussed below. Hooper (Kew Bull. 26:582. 1972.) transferred Scirpus confervoides to the genus Websteria S.H. Wright [W. confervoides (Poir.) Hooper]. Without ref- erencing Hooper, Kern (1974, p. 505) retained Poiret’s plant within Scirpus L. Tucker (J. Arnold Arbor. 68:388. 1987) transferred it to the genus Eleocharis R. Br. and proposed the combination “FE. confervoides (Poir.) Tucker.” Prior to Tucker’s publication, Miquel (1856, p. 303) made the combination E. con- fervoides. Since both Miquel and Tucker independently made the same combi- nation based on the same type, Tucker’s later combination must be considered an isonym (cf. Nicolson, Taxon 24:461-466. 1975). Ironically, both Miquel’s and Tucker’s combinations are later homonyms of FE. confervoides Steud. (Syn. Pl. Glumac. 2:82. 1854-1855.). For the inclusion of Poiret’s taxon in Eleocharis, Tucker derived support from a lack of leaf blades and from some embryological characters. {The em- bryo in Eleocharis is a variant of the Fimbristylis-type: top shaped, with basal coleoptyle, lateral rootcap, and first leaf more or less protruding from the germ pore (fide Kern 1974, pp. 446, 523).] Regarding the embryological character, Kern (p. 447) remarked that more than one type of embryo may occur within a genus or a single embryo type may occur in more than one genus. He stated that S. confervoides “has its own embryo type, similar to that of Eleocharis, but somewhat more differentiated.” He further suggested the retention of Poiret’s taxon within Scirpus because of its floral characters. We disagree with both Kern’s and Tucker’s disposition of Poiret’s taxon. We believe that inclusion of Poiret’s taxon either within Scirpus or within Eleocharis is anomalous. We concur with Wright’s (Bull. Torrey Bot. Club 14:135. 1887.) and Hooper’s (I.c.) analyses that the unique morphology of Poiret’s taxon deserve its disposition in the monotypic genus Websteria S.H. Wright. Websteria confervoides (Poir.) Hooper, Kew Bull. 26:582. 1972. BASIONYM: Scirpus confervoides Poir. in Lam., Encycl. 6:755. 1804. Eleocharts con- | fervoides (Poir.) Miq., Fl. Ind. Bat. 3:303. 1856.; Tucker, J. Arnold Arbor. 68:388. 1987., non Steud. 1854-1855. 126 Po OYAT. OFnOVGa A volume 73(2):124-136 August 1992 FABACEAE Senna artemisiordes Unaware of the publication of Senna artemisioides (Gaud. er DC.) Randell (J. Adelaide Bot. Gard. 12(2):220. 1989.), we (Phytologia 72:87. 1992.) made the identical combination. Barneby (NY) brought Randell’s combination to our attention. Prior to making our combination, we searched Index Kewensis, Kew Index, and Gray Herbarium Card Index to insure that this combination did not exist. At the time of our search, the NCU Library had holdings of Kew Index for the years ending 1989; unfortunately, Randell’s combination is not listed in the Kew Index for 1989 and is most likely listed in the Kew Index for 1990 (for which we do not have access to date). Nevertheless, we regret our oversight and consider our combination as an isonym of Randell’s combination. Senna artemisiotdes (Gaud. ez DC.) Randell, J. Adelaide Bot. Gard. 12(2):220. 1989.; Kartesz & Gandhi, Phytologia 72:87. 1992. MALVACEAE Sidalcea Hitchcock (1957, pp. 29-30) proposed Sidalcea malviflora subsp. laciniata, under which he recognized two varieties: S. malviflora subsp. laciniata var. lactntata and S. malviflora subsp. laciniata var. sancta C.L. Hitchce. Hitch- cock provided a Latin description for his subspecies and a Latin diagnosis for his var. sancta. He provided a diagnosis in English for var. laciniata. Since he intended var. laciniata to be the autonym of subsp. lactniata (he cited the same type for both ranks) and since the International Code of Botanical Nomenclature (ICBN) (Lanjouw 1956, Art. 26) mandated at that time that a) autonyms were to be cited without an author and b) that autonyms did not have taxonomic standing, Hitchcock neither provided a Latin diagnosis nor authorship details for S. malviflora subsp. laciniata var. laciniata. Under the present JCBN (Greuter 1988), autonyms exist only for species, but not for infraspecific taxa, such as subspecies and varieties. Furthermore, all names (including infraspecific ranks, but excluding autonyms of species) must have an author citation (ICBN Art. 46.1). Since var. lacinzata is not an autonym of S. malviflora, the former name requires authorship. According to Art. 34 Ex. 11, an author can simultaneously validate a sin- gle combination at different infraspecific ranks within a species. Example 11 illustrates that the description of “Malvastrum bicuspidatum subsp. tumidum S.R. Hill var. tumidum, subsp. et var. nov.” (Hill, Brittonia 32:474. 1980.) simultaneously validated both M. bzcuspidatum subsp. tumidum S.R. Hill and Kartesz & Gandhi: Nomenclatural notes XI 127 M. bicuspidatum var. tumidum S.R. Hill. In his work, Hill also proposed M. bicuspidatum subsp. tumidum var. glabrum S.R. Hill and provided the follow- ing diagnosis: “A varietate typica foliorum lamina angustata ovato-lanceolata petiolo 8-plo usque longiori, bracteolis calycem aequantibus, schizocarpis ma- turis glabris aut scabrellis et in vivo apice rubris, mericarpiis minoribus ca 5.0 mm diametro differt.” An analysis of the Latin description of M. bicuspidatum subsp. tumidum var. tumidum indicates that Hill did not compare the length of the blade with that of the petiole, nor did he compare the bracteoles with that of the calyx, or the size of the mericarps. However, he did describe the mericarps as “often tinted rose-red drying to red-brown or brown” with the vestiture restricted to the apical and cusp surfaces. Furthermore, Hill pro- vided a single description to validate both ranks (subsp. tumzdum and var. tumidum); hence, this description is not the sum of var. tumzdum and var. glabrum. According to Art. 25 Ex. 1, an infraspecific taxon, for nomenclatural purposes, is regarded as the sum of its subordinate taxa. In Hill’s treatment, it is evident that var. glabrum is taxonomically differentiated from var. tumidum, but for nomenclatural purposes, it is questionable whether subsp. tumzdum can be regarded as the sum of its vars. tumidum and glabrum. Hill’s and Hitchcock’s treatments are not identical. Unlike Hill’s treat- ment, Hitchcock’s Latin description of subsp. laciniata is explicitly a sum of its two varieties (var. laciniata and var. sancta). Evidently, Hitchcock met the requirements of Art. 25.1, but whether var. laciniata was validly published is questionable. In personal communications, Nicolson (US) and McVaugh (NCU) asserted that var. laciniata was validly published in Hitchcock’s treat- ment. Their assertion is in accordance with Rec. 26A.1. Stdalcea malviflora (DC.) A. Gray ex Benth. var. lacinzata C.L. Hitchc., Univ. Washington Publ. Biol. 18:30. 1957. ONAGRACEAE Epilobium brachycarpum For 140 years (1840-1981), the Tall Annual Willowherb was known by the name Epilobium paniculatum Nutt. ez Torr. & Gray (published in 1840). In 1981, Hoch & Raven (Taxon 30:666) stated that E. brachycarpum Presl (pub- lished in 1831) and E. paniculatum were conspecific. These authors remarked that the name E. brachycarpum “has been incorrectly applied to plants of the E. ciliatum complex in the only four treatments in which the name has been used in the last hundred years.” Furthermore, these authors believed E. paniculatum to be an universally accepted name for the willowherb in ques- tion. Based on JCBN Art. 69, Hoch & Raven proposed to reject the name E. brachycarpum. 128.5 JP YT On, 0 GTA volume 73(2):124-136 August 1992 In anticipation of the acceptance of Hoch & Raven’s proposal by the nomen- clatural committee of the JCBN, Solomon (1982, p. 330) recognized the name Epilobium paniculatum and cited E. brachycarpum as a synonym. However, the nomenclatural committee (Taxon 33:300. 1984.) remarked that the name E. brachycarpum was neither widely nor persistently misused and that JCBN Art. 69 was not designated to cover a case such as Hoch & Raven proposed. Therefore, the committee unanimously rejected Hoch & Raven’s proposal and stated that the name EF. paniculatum must be replaced by the earlier name E. brachycarpum. Without referencing the committee’s rejection, Hoch (1986, p. 508) used the name Epilobium paniculatum, whereas Welsh et al. (1987, p. 441), Dorn (1988, p. 204), and Gleason & Cronquist (1991, p. 317) correctly used the name £. brachycarpum. . Epilobium brachycarpum Presl, Rel. Haenk. 2:30. 1831. Epilobium paniculatum Nutt. ex Torr. & Gray, Fl. N. Amer. 1:490. 1840. ORCHIDACEAE Sptranthes confusa Garay (1980) segregated members of Detregyne Schlechter from Spiranthes L.C. Rich. sens. lat. According to Garay, D. durangensis (Ames & C. Schwe- inf.) Garay (= S. durangensis Ames & C. Schweinf.) does not occur in Texas. He assigned the Texas orchids, known by the preceding name, to D. confusa Garay and stated that D. confusa differs from D. durangensis “in having glan- dular pubescent sepals, a differently proportioned lip with a different callus at its base and shape of the rostellum.” We concur with Garay’s treatment of D. durangensis, but for the North American flora, we include it within Spiran- thes sens. lat. (including Deiregyne), requiring the new combination proposed below. Spiranthes confusa (Garay) Kartesz & Gandhi, comb. nov. BASIONYM: Deiregyne confusa Garay, Bot. Mus. Leafl. 28:283. Apr 1982 (Sep 1980). TYPE: MEXICO. Hidalgo: Lagoon of Metztitlan, Gonzdles s.n., sub Nagel 2194 (AMES). Spiranthes durangensis auct. non Ames & C. Schweinf. Kartesz & Gandhi: Nomenclatural notes XI 129 ROSACEAE Amelanchier alnifolia Jackson (1895), Jones (1946, p. 67) and Phipps et al. (1990, p. 2231, no. 1) attributed the combining authorship of Amelanchier alnifolia to (Nutt.) Nutt. (J. Acad. Nat. Sci. Philadelphia 7:22. 1834.). In Nuttall’s work, the combination Amelanchier alnifolia was cited without authorship. Nuttall nei- ther provided a description nor cited a direct or an indirect reference to the basionym Aronia alnifolia Nutt. Since Nuttall did not meet the requirements of ICBN Arts. 32.1c, 32.3, and 32.4, the combination Amelanchier alnifolia was not validly made in Nuttall’s 1834 work. To our knowledge, M. Roemer was the first to validate the preceding combination. Amelanchier alnifolia (Nutt.) Nutt. ec M. Roemer, Fam. Nat. 3:147. 1847. BASIONYM: Aronia alnifolia Nutt., Gen. N. Amer. 1:306. 1818. Amelanchier pumila Jackson (1895) attributed Amelanchier pumila to Nutt. ex Torr. & Gray, whereas Phipps et al. (1990, p. 2232, no. 27) to Nuttall. Nuttall’s manuscript name A. pumila was validly published in Torrey & Gray’s work at varietal rank. Hence, none of these authors must be credited with authorship of the preceding binomial. To our knowledge, M. Roemer was the first to elevate Nuttall’s taxon to specific rank, which he attributed to Nuttall. The full author citation is given below. Amelanchier pumila‘ (Torr. & Gray) Nutt. exc M. Roemer, Fam. Nat. 3:145. 1847. BASIONYM: Amelanchier canadensis (L.) Medik. var. pumila Torr. & Gray, Fl. N. Amer. 1:474. 1840. Aronia arbutifolia The name Aronza arbutifolia has been attributed to either Medicus (Jack- son 1895) or to (L.) Ell. (Soil Conservation Service 1982, p. 140; Phipps et al. 1990, p. 2232, no. 1). In our study, we found that neither is correct. Medicus (1789, pp. 140, 155) cited the generic name Aronia and the binomial Mespilus arbutifolta. This does not constitute valid publication of the combination A. arbutifolia, since Medicus did not definitely associate the epithet arbutifolia with the generic name Aronia; hence, Medicus must not be credited with au- thorship of the combination (JCBN Art. 33, Ex. 2). Although Elliott (1821, p. 556) validly used the name A. arbutifolia, Persoon made this combination fourteen years prior to Elliott’s work. Accordingly, Persoon is the combining author of this binomial. 130 PHY TOO) Gala volume 73(2):124-136 August 1992 Aronia arbutifolia (L.) Pers., Syn. Pl. 2:39. 1807. BASIONYM: Mespilus arbutifolius L., Sp. Pl. 478. 1753. Choenomeles japonica The combining authorship of Choenomeles japonica has often been at- tributed to Lindley (Jackson 1895; Ohwi 1965, p. 547; Phipps et al. 1990, p. 2233, no. 2). Although Lindley (Trans. Linn. Soc. London 13:97. 1822.) proposed the genus Choenomeles and cited the name Pyrus japonica Thunb. (as the type), this does not constitute valid publication of the combination C. japonica, since Lindley did not definitely associate the epithet japonica with the generic name Choenomeles; hence, Lindley must not be credited as the au- thor of this combination (JCBN Art. 33, Ex. 2). To our knowledge, Spach was the first to associate the epithet japonica with the generic name Choenomeles and he attributed the combination to Lindley. The full author citation is given below. Choenomeles japonica (Thunb.) Lindl. ex Spach, Hist. Nat. Veg. Phan. 2:159. 1834. BASIONYM: Pyrus japonica Thunb., Fl. Jap. 207. 1784. Malus pumila The Soil Conservation Service (1982, p. 144) attributed the name Malus pumila to “(L.) Mill.” and assigned an * to the authorship indicating that the nomenclature was verified. However, Terrell et al. (1986, p. 92) and Phipps et al. (1990, p. 2234, no. 11) did not cite a parenthetical authorship for M. pumila. Our study follows. In his treatment of Malus, Miller (1768) referenced Linnaeus. For M. pumila, Miller cited a reference to Bauhin’s M. pumila (a pre-1753 publica- tion). Pyrus malus L. var. paradisiaca L. (1753, p. 479) was also based on Bauhin’s M. pumila. Although both Miller and Linnaeus referenced Bauhin, they recognized different taxonomic ranks and used different epithets. Even in the second edition of Species Plantarum, Linnaeus (1762, p. 686) maintained his 1753 treatment of P. malus var. paradisiaca. Since Miller did not base his M. pumila on the Linnaean work, the authorship of M. pumila must not include a parenthetical authorship, as indicated by Terrell et al. Malus pumila P. Mill., Gard. Dict., ed. 8. no. 3. 1768. Pyrus malus L. var. paradisiaca L., Sp. Pl. 479. 1753. Kartesz & Gandhi: Nomenclatural notes XI 131 Malus sieboldu Jackson (1895) attributed the name Malus sieboldi to Dippel (1893; p. 406), but Phipps et al. (1990, p. 2235, no. 29) attributed it to Rehder. Since Dippel made a reference to “Mal. Sieboldii Rgl. in Gartenflora 1859. S. 82,” some authors may believe that the reference to Regel was an indirect reference to the basionym Pyrus sieboldii Regel and that Dippel made the combination M. sieboldiz. However, Dippel cited the preceding name as a synonym of M. toringo Sieb. Since Dippel did not accept the name M. szeboldi, he must not be credited with its authorship (JCBN Art. 34.1a). To our knowledge, Rehder was the first to validly make the combination. Malus sieboldii (Regel) Rehder, Rev. Hort. IV. 451. 1870. BASIONYM: Pyrus sieboldi: Regel, Ind. Sem. Hort. Petrop. 51. 1858. Malus sylvestris Some authors, such as Rehder (1940, p. 391), believe that the name Malus sylvestris P. Mill. was based on “Pyrus malus var. sylvestris L.” In his treat- ment of the genus Malus, Miller (1768) indeed referenced Linnaeus. Under P. malus L., Linnaeus (1753, pp. 479-480) cited six epithets, of which sylvestris was the first. Superficially it might appear as though Linnaeus used the epithet sylvestris at varietal rank. Our study follows. Of the six epithets, the last five are associated with Greek letters, whereas sylvestris is not. According to Stearn (1957, pp. 90, 93), if Linnaeus considered his varieties to be well marked from the species, then such varieties were given epithets and Greek letters, whereas the species proper were given no Greek letter. In several cases, species proper were given an additional epithet (but no Greek letter) to contrast them with his other varieties. Based on Stearn’s analysis, we conclude that Linnaeus used the epithet sylvestris to distinguish the common expression of Pyrus malus from its other five varieties. Hence, “P. malus sylvestris L.” was never validly published and Malus sylvestris must not have a parenthetical authorship. Malus sylvestris P. Mill., Gard. Dict., ed. 8. Malus, no. 1. 1768. 132 PHY TOLTOGIA volume 73(2):124-136 August 1992 VITACEAE Authorship of the New Vitaceae Names Proposed in Gray’s Syn. Fl. N. Amer. In his treatment of Vitis, Moore (Sida 14:351, 357, 359. 1991.) attributed V. cinerea (Engelm.) Millard. var. canescens to (Engelm.) Bailey ez A. Gray; V. candicans Engelm. ez A. Gray var. coriacea to (Shuttlew. er Planchon) Bailey ez A. Gray; and V. longii Prince var. microsperma to (Munson) Bailey ez A. Gray. However, for V. cordifolia Michx. var. helleri and for V. rupestris Scheele var. dissecta, Moore (pp. 352, 361) credited Bailey with authorship. We speculate that Moore most likely distinguished between comb. nov. (e.g., vars. canescens, coriacea, and microsperma) and var. nov. (e.g., vars. dissecta and helleri) and that he considered Gray to be responsible for the comb. nov., while Bailey to be responsible for the var. nov. Regarding the authorship of the new names proposed within Vitaceae (treated in Gray’s work) and their bibliographies, our analysis follows. Between 1878 and 1897, Gray’s Synoptical Flora of North America was issued in two volumes. The second volume appeared first, followed by part 2 of the first volume, and finally number 2 of part 1 of the first volume. The preceding no. 2 was published nine years after Gray’s death. Although its text was chiefly written by Gray, it included several contributions from other workers. On p. 419 of no. 2, the treatment of Vitaceae was credited to Bailey, with a footnote stating that “ordinal and technical generic characters by A. Gray.” Based on this information, perhaps, the treatment of Vitaceae may be credited to both Bailey and Gray, with Bailey as the first author. Regarding the authorship of ten new varietal names (seven comb. nov., two var. nov., and one nom. nov.) proposed within Bailey and Gray’s work, eight were ascribed to Bailey, one to Eggert, and one to Gray. Excluding the name credited to Gray, the remainder (nine names), should be credited to Bailey, since he (not Gray) was responsible for them. Bailey based Vitis candicans var. coriaceaon V. coriacea Shuttlew. ez Plan- chon (in DC., Monogr. Phan. 5:345. 1887., non Miq. 1863.). Since the pre- ceding basionym is a later homonym and thus, illegitimate, it must not be taken into consideration for purpose of priority (ICBN Art. 45.3). Therefore, no parenthetical authorship should be cited for var. coriacea and it must be treated as a nom. nov., with its priority from 1897 (JCBN Art. 72, Note 1). Moore (l.c., p. 357) noted the illegitimacy of V. cortacea, but failed to omit parenthetical authorship for var. cortacea. The correct authorship of the ten names and their bibliographies are given below. Ampelopsis quinquefolia Michx. var. heptaphylla (Buckley) A. Gray, Syn. Fl. N. Amer. 1[1(2)]:432. 1897. Kartesz & Gandhi: Nomenclatural notes XI 133 Ampelopsis quinquefolia Michx. var. pubescens (Schlect.) Bailey 7n A. Gray, Syn. Fl. N. Amer. 1{1(2)]:432. 1897. Vitis aestivalis Michx. var. glauca (Munson) Bailey zn A. Gray, Syn. FI. N. Amer. 1[1(2)]:427. 1897. Vitis aestivalis Michx. var. bourquiniana (Munson) Bailey in A. Gray, Syn. Fl. N. Amer. 1[1(2)]:428. 1897. Vitts candicans Engelm. ez A. Gray var. cortacea Bailey in A. Gray, Syn. FI. N. Amer. 1[1(2)]:429. 1897. Vitis cinerea (Engelm.) Millard. var. canescens (Engelm.) Bailey in A. Gray, Syn. Fl. N. Amer. 1[1(2)]:425. 1897. Vitis cordifolia Michx. var. heller: Bailey 1n A. Gray, Syn. Fl. N. Amer. 1[1(2)]:424. 1897. Vitts longw Prince var. microsperma (Munson) Bailey in A. Gray, Syn. FI. N. Amer. 1[1(2)]:423. 1897. Vitis rupestris Scheele var. dissecta Eggert ez Bailey in A. Gray, Syn. Fl. N. Amer. 1[1(2)]:422. 1897. Vitis vulpina L. var. praecor (Engelm. ez Bailey) Bailey in A. Gray, Syn. Fl. N. Amer. 1[1(2)]:422. 1897. ACKNOWLEDGMENTS The authors are thankful to: Larry E. Brown (SBSC), Bryan Dutton (GH), Jim Reveal (MARY), and Rogers McVaugh (NCU) for helpful suggestions, and to Ms. Tracy Delius (Biology-Forestry Library, DUKE) for some of the literature used in this study. REFERENCES Dippel, L. 1893. Handb. Laubholzk., vol. 3. Berlin, Germany. Dorn, R.D. 1988. Vascular Plants of Wyoming. Mountain West Publishing, Box 1471. Cheyenne, Wyoming. . Elliott, S. 1821. Sketch Bot. S. Carolina, vol. 1. Charleston, South Carolina. 134 PRN O1,0)G ICA volume 73(2):124-136 August 1992 Garay, L.A. 1982. A generic revision of the Spiranthinae. Bot. Mus. Leafl. 28:277-426. Gleason, H.A. & A. Cronquist. 1991. Man. Vasc. Pl. N.E. U.S., ed. 2. The New York Botanical Garden, Bronx, New York. 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. Hitchcock, C.L. 1957. A study of the perennial species of Sidalcea, part I: Taxonomy. Univ. Wash. Publ. Biol., vol. 18. Hoch, P.C. 1986. Epzlobium. Pp. 504-509. Jn: Great Plains Flora Associa- tion, Flora of the Great Plains. University Press of Kansas, Lawrence, Kansas. Jackson, B.D. 1895. Index Kewensis, 2 vols. University Press, Oxford, England. Jones, G.N. 1946. American species of Amelanchier. Illinois Biol. Monogr. 20(2). Kartesz, J.T. 1988. A flora of Nevada. Ph.D. Dissertation, University of Nevada, Reno, Nevada. . 1992. A Synonymized Checklist of the Vascular Flora of the United States, Canada, and 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. . 1991a. 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. Kartesz & Gandhi: Nomenclatural notes XI 135 . 1991c. Nomenclatural notes for the North American flora - VII. Phytologia 71:87-100. . 1991d. Nomenclateral 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. . 1992a. Nomenclatural notes for the North American flora - IX. Phytologia 72:17-30. . 1992b. Nomenclatural notes for the North American flora - X. Phytologia 72:80-92. Kern, J.H. 1974. Flora of Malestana. I. 7(3). Cyperaceae. Noordhoff International Publishing, Leyden, The Netherlands. Lanjouw J. (Ch. Ed. Comm.). 1956. International Code of Botanical Nomen- clature. Adopted by the Eighth International Congress, Paris, July 1954. Regnum Veg., vol. 8. Linnaeus, C. 1753. Sp. Pl., ed. 1, 2 vols. Stockholm, Sweden. Facsimile edition issued in 1957-1959 by Ray Society, London, Great Britain. —_______.. 1762-63. Sp. Pl., ed. 2, 2 vols. Stockholm, Sweden. Medicus, F.K. 1789: Phil. Bot., vol. 1. Mannheim, Germany. Miller, P. 1768. Gard. Dict., ed. 8. London, Great Britain. Miquel, F.A.W. 1856. Fl. Ned. Ind., vol. 3. Leipzig, Germany. Ohwi, J. 1965. Flora of Japan. Smithsonian Institution, Washington, DC. Phipps, J.B., K.B. Robertson, P.G. Smith, & J.R. Rohrer. 1990. A checklist of the subfamily Maloideae (Rosaceae). Canad. J. Bot. 68:2209-2269. Rehder, A. 1940. Manual of Cultivated Trees and Shrubs, ed. 2. The Macmil- lan Company, New York, New York. Soil Conservation Service. 1982. National List Scientific Plant Names, vol. 1. U.S.D.A., SCS-TP 159. Government Printing Office, Washington, DC. Solomon, J.C. 1982. The systematics and evolution of Epilobium (Ona- graceae) in South America. Ann. Missouri Bot. Gard. 69:239-335. 136 PHY T.OLOGIA volume 73(2):124-136 August 1992 Stearn, W.T. 1957. An Introduction to the Species Plantarum and Cognate Botanical Works of Carl Linnaeus. Facsimile edition of Species Plan- tarum, ed. 1, issued by Ray Society, London, Great Britain. Terrell, E.E., S.R. Hill, J.H. Wiersema, & W.E. Rice. 1986. Checklist of Names of 3,000 Vascular Plants of Economic Importance. Agric. Handb. No. 505, U.S.D.A., Washington, DC. Welsh, S.L., N.D. Atwood, S. Goodrich, & L.C. Higgins. 1987. Utah Flora. Great Basin Naturalist Mem., no. 9. Phytologia (August 1992) 73(2):137-139. A NEW SPECIES OF CASTILLEJA (SCROPHULARIACEAE) FROM DURANGO, MEXICO Guy L. Nesom Department of Botany, The University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT. Castilleja chlorosceptron, sp. nov., is described from the Sierra Madre of western Durango. In its rhizomatous habit, scabrous stems, and evenly divided calyces with acute lobes, C. chlorosceptron is most similar and probably most closely related to C. durangensis. The new species differs from the latter particularly in its glabrous leaves with nonclasping bases and its much smaller flowers. KEY WORDS: Castilleja, Scrophulariaceae, México Two recent collections from western Durango, México, have brought to attention a previously undescribed species of Castilleja. Castilleja chlorosceptron Nesom, sp. nov. TYPE: MEXICO. Durango: Mpio. Canelas, on the road to Topia and Canelas, 6 km E of Canelas, 30 km E of jct of this road with road to Topia (at Cuevacillas); 1770 m, W slope, eroded rounded ridges, reddish igneous soil, more or less barren beneath, with yellow pines and oak (Q. eduardi, Q. viminea, Q. scytophylla). Rare, in small patches particularly on knoll tops; 29 Jun 1992, R.W. Spellenberg 11059 with J. Bacon (HOLOTYPE: TEX!; Isotypes: CIIDIR,NMC). Castillejae durangensi Nesom similis sed foliis ac bracteis glabris, foliis angustioribus basibus non amplectentibus, et floribus multo minoribus differt. Herbaceous perennials arising from a system of slender (almost wiry), scale leaved rhizomes, apparently without a central axis. Stems erect, 5-15 cm tall, slender, eglandular, minutely but densely scabrous with deflexed hairs 0.05- 0.10 mm long (to 0.5 mm in the upper inflorescence). Leaves densely crowded 137 138 PaY TT OL0' GIA volume 73(2):137-139 August 1992 and overlapping from bottom to top of the stem, on nodes ca. 1 mm apart, ascending, glabrous, 3 veined, entire, linear or becoming lanceolate upwards, 2-3 cm long, 1.5-2.5 mm wide, not at all basally clasping. Floral bracts mostly green, red at the very apex, glabrous, lanceolate, entire, 11-20 mm long, about equaling the associated calyces. Calyces greenish, slightly pink tinged, red only at the very apex, the upper 1/4 densely granular glandular, minutely hispidulous on the proximal portion, 12-13 mm long, the primary lobes 2-3 mm long, the secondary lobes 0.5-0.8 mm deep with acute tips. Corollas 15- 16 mm long, the teeth of the lower lip narrowly oblong, ca. 1 mm long, the galea 2-3 mm long, “blushed reddish, flanges red,” with a densely granular glandular dorsal surface without other hairs, exserted 3-5 mm from the calyx; stigma bilobed, 1.4-1.6 mm wide, yellow-green, prominently exserted. Mature fruit not seen. Additional collection examined: MEXICO. Durango: Mpio. El Salto, 32 km W of El Salto, woods of oak, pine, and some Abies, 2300 m, 10 Jul 1982, Herndndez M. 7844 (WIS). The plants of Castilleja chlorosceptron (named for their primarily green, wandlike aspect) are herbaceous and diminutive in stature, with stems aris- ing from slender rhizomes without a central axis, and they produce glabrous, crowded ascending, linear, entire, nonclasping leaves, minutely scabrous stems, lanceolate floral bracts, shallowly and evenly divided calyces, and small corol- las. In its rhizomatous habit, minutely scabrous stems, crowded, narrow, entire leaves, and evenly divided calyces with acute lobes, C. chlorosceptron is most similar to and probably most closely related to C. durangensis Nesom, secondarily to C. aspera Eastw. (Nesom 1992), both members of the group centered around C. scorzoneraefolia Kunth and treated by Eastwood (1909) within Castilleja sect. Euchroma (Nutt.) Benth. The new species differs from C. durangensis in its much thinner rhizomes, glabrous leaves and bracts, leaves with nonclasping bases, and much smaller flowers (calyces 12-13 mm long vs. 18-22 mm long; corollas 15-16 mm long vs. 21-25 mm long), the galeas with only glandular hairs (vs. glandular as well as pilose, eglandular hairs). Castilleja durangensis is represented by many collections from the area of El Salto in southwestern Durango (Nesom 1992), where it is endemic. Castilleja chlorosceptron is partially sympatric with C. durangensis, and although it has been collected only twice, it appears to have a significantly wider geographic distribution. ACKNOWLEDGMENTS I thank Billie Turner and Carol Todzia for their review of the manuscript. Nesom: New Castilleja from Durango, México 139 LITERATURE CITED Eastwood, A. 1909. Synopsis of the Mexican and Central American species of Castilleja. Proc. Amer. Acad. Arts 44:563-591. Nesom, G.L. 1992. New species and taxonomic evaluations of Mexican Castilleja (Scrophulariaceae). Phytologia 72:231-252. Phytologia (August 1992) 73(2):140-142. A NEW SPECIES OF JUSTICIA (ACANTHACEAE) FROM NORTHEASTERN MEXICO Guy L. Nesom Department of Botany, The University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT A new species of Justicia sect. Pentaloba is described from Nuevo Leon, México: J. hintoniorum. It is most closely related to J. pilosella and J. turneri, differing from both in its hairy capsules. KEY WORDS: Justicia, Siphonoglossa, Acanthaceae, México Hilsenbeck (in Henrickson and Hilsenbeck 1979) segregated four species of the genus Siphonoglossa Oerst. as Siphonoglossa sect. Pentaloba Hils. At the conclusion of his studies, Hilsenbeck (1990) transferred sect. Pentaloba to Justicia, but even within the latter genus, the species of sect. Pentaloba still form a distinctive group, particularly on the basis of their terete and greatly elongated floral tubes. The species of sect. Pentaloba differ from typical Stphonoglossa in the following features: floral bracteoles foliaceous and oblong to lanceolate (vs. subulate-bracteate), flowers axillary (vs. mostly spicate), calyces 5 lobed (vs. 4 lobed), stigmas 2 lobed (vs. 1 lobed), capsules ovoid (vs. fiddle shaped), seeds strongly flattened (vs. thickened), and a base chromosome number of z=14 (vs. z=11). All four species of sect. Pentaloba have their primary geographic range in northeastern México and adjacent Texas, west into Arizona. In the identification of recently collected specimens and accompanying cu- ration of other LL,TEX accessions, a previously undescribed species of Justicia sect. Pentaloba (Hils.) Hils. has been discovered. Justicia hintoniorum Nesom, sp. nov. TYPE: MEXICO. Nuevo Leon: Mpio. Aramberri, N of Aramberri, IRF Lampacitos, 995 m, 16 Jun 1990, Hinton et al. 20854 (HOLOTYPE: TEX!). Justiciae pilosellae (Nees) Hils. et J. turneri Hils. similis sed ab ambobus fructibus strigosi-hirsutulis differt; a J. turneri corol- lis purpureis lobis majoribus differt; a J. pilosella foliis majoribus tenuioribus discoloribusque differt. 140 Nesom: New Justicia from México 141 Perennials with ascending-erect stems 8-18 cm tall, slightly woody at the base, arising from slender rhizomes, densely pilose-hispid with a mixture of stiffly spreading to slightly deflexed hairs 1.0-1.5 mm long and shorter (0.3-0.5 mm long), strongly deflexed hairs, the hairs with a slight tendency to oc- cur in lines on the stem. Leaves opposite, relatively thin, discolorous (lighter beneath), ovate to elliptic, mostly (15-)30-80 mm long with petioles 3-15 mm long, the blades 10-30 mm wide, eglandular, strigose-puberulent beneath, mod- erately to sparsely strigose-hispid above, the petioles with spreading cilia up to 1.5 mm long. Flowers axillary, sessile, solitary in upper half of the plant; paired bracteoles obovate to slightly spatulate, 10-14 mm long, 3-4 mm wide; calyx 7-10 mm long, the lobes linear-triangular, 0.3 mm wide at base, 6-9 mm long, equal, united for ca. 1 mm at the base; corollas distinctly purplish, moderately hispid, the tube 16-22 mm long, ca. 2 mm wide, the upper lobe erect, oblong, apically 2 toothed, 8-9 mm long, 2.5-3.0 mm wide, the lower 3 lobes spreading, oblong-oblanceolate, apically rounded, 6-10 mm long, 3.5-6.0 mm wide; stamens 2, exserted 3-6 mm from the tube, the thecae subparallel, 1.0-1.5 mm long, lower with a spurred base; styles ca. equal the tube length, slightly hispid-strigose near the base, the stigmatic lobes 0.1 mm long. Cap- sules ca. 10 mm long, basal stipe strongly flattened, 4-5 mm long, the head ovoid, 5-6 mm long, 3-4 mm wide, brown, prominently pubescent with stiff, retrorsely appressed hairs on the basal 2/3 and shorter, erect hairs near the apex. Seeds 4, bright orange, 2.5-3.0 mm long and wide, strongly flattened and disciform, without a thickened margin, the faces with bullate tuberculate incrustations 0.1-0.2 mm broad. Chromosome number unknown. Additional collections examined: MEXICO. Nuevo Leon: Mpio. Iturbide: Iturbide to Camarones, oak and pine woods, 1305 m, 6 Sep 1991, Hinton et al. 21416 (TEX); Iturbide to Agua Blanca, mixed forest of oak and walnut, 1385 m, 21 Aug 1991, Hinton et al. 21216 (TEX); 16 mi W of Linares, rocky mt. side from Hwy 60, on shale, mixed oak, Mimosa, with Zamia and Agave, 8 Sep 1962, Turner & Powell 1051 (TEX). The distinctiveness of this species was focused by the three recent col- lections (including the type) made by the Hinton family. The first known collection (Turner & Powell 1051), however, was made almost 30 years ear- lier and has been identified by Hilsenbeck as Justicia pilosella, although the salient features of 5S. hintoniorum were not noted in his description (1990) of J. pilosella. Fruiting specimens of J. hintoniorum can be immediately distin- guished from the other four species of sect. Pentaloba by their prominently strigose-hirsute capsules; the new species differs from J. pilosella in its larger, thinner, discolorous leaves and from J. turneriin its purple corollas with much larger lobes. These three species are most closely related among themselves; the other two species of sect. Pentaloba occur further west in México and the southwestern United States and are characterized by linear leaves. All four collections of Justicia hintoniorum have been made in a relatively 142 PHY TOLO GIA volume 73(2):140-142 August 1992 small area of southeast-central Nuevo Leon. The geographic range of J. hinto- niorum is essentially contiguous with that of J. turneri (which occurs mostly to the east and northeast) and apparently completely overlapping with that of J. pilosella. Both of the latter species have been collected very near or within the range of J. hintontorum. ACKNOWLEDGMENTS I thank B.L. Turner and T.P. Ramamoorthy for their review of the manuscript. LITERATURE CITED Henrickson, J. & R.A. Hilsenbeck. 1979. New taxa and combinations in Stphonoglossa (Acanthaceae). Brittonia 31:373-378. Hilsenbeck, R.A. 1990. Systematics of Justicia sect. Pentaloba (Acanthaceae). Pl. Syst. Evol. 169:219-235. Phytologia (August 1992) 73(2):143-148. NEW SPECIES AND COMBINATIONS IN PODACHAENIUM (ASTERACEAE, HELIANTHEAE) B. Turner & José Panero Department of Botany, The University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT. A new species from Chiapas (Reserva E] Triunfo), Podachaenium chiapanum B. Turner & Panero, is described and illustrated, and the new combination, P. standleyi (Steyermark) B. Turner & J. Panero, is made. The latter occurs in Guatemala and has been considered to be synonymous with the Oaxacan species P. pachyphyllum (Klatt) S.F. Blake by recent workers. Characters which distinguish these several taxa are listed. Along with the widespread P. eminens, the genus now contains four species. KEY WORDS: Asteraceae, Heliantheae, Podachaentum, México Routine identification of Mexican Asteraceae has occasioned the present paper. Prior to the transfer of Calea skutchii S.F. Blake into Podachaentum by H. Robinson (1978), the genus was thought to be monotypic. Jansen et al. (1982), however, transferred the latter to the genus Squamopappus but, at the same time, transferred into Podachaenium another species, P. pachyphyllum (including C. standleyz), that had long resided in Calea. In the present paper we accept Podachaenium as enlarged by Jansen et al. (1982), add an additional species P. chiapanum to the genus, and resur- rect the name Calea standleyi, giving it specific status within an expanded Podachaentum. As we currently view the genus, Podachaenium contains four species, as keyed, and discussed in the account that follows. KEY TO SPECIES OF PODACHAENIUM 1. Leaves palmately veined, broader than long, or nearly so; widespread. .. Miata sere IAS OIE ees eee sears ae Nea cece te leecees P. eminens 144 PY TOLOGLEA volume 73(2):143-148 August 1992 1. Leaves trinervate from well above the base, or pinnately nervate, 2-4 times as long as broad; Oaxaca, Chiapas, Guatemala. ..............2..-. (2) 2. Stems sparsely pilose to glabrous; involucral bracts ca. biseriate, mostly 3-OnmMiE AON. APPLessed. 5 -. 520 5. once mec ov oe see (3) 2. Stems densely and permanently rusty pilose; involucral bracts 3-4 seriate, mostly 6-12 mm long, loose and spreading; Chiapas. ..... RPM ec POCTRons pate ee ela wies Go are ew oka SR NE ee P. chiapanum 3. Leaves pinnately nervate; petioles mostly 0.5-1.0 cm long; blades elliptic, widest at or near the middle; ray florets ca. 13, the ligules 15-20 mm Ong tig hice acai rei ON oe ae Bel be oiste me os P. pachyphyllum 3. Leaves with 3(5) principal nerves; petioles mostly 2-7 cm long; blades ovate to trullate ovate, wider well below the middle; ray florets 8-11, i221, im LONE AGUAtemialas a.) fie cians oo oie «dda ake eee P. standleyi Podachaenium chiapanum B. Turner & J. Panero, sp. nov. TYPE: MEX- ICO. Chiapas: Mpio. Jaltenango, Reserva E] Triunfo, 15° 39’ N, 92° 48’ W, evergreen cloud forests, 2200 m, 12 Jun 1990, M. Heath & A. Long 956 (HOLOTYPE: TEX). Podachaenium pachyphyllo (Klatt) S.F. Blake similis sed differt caulibus dense persistente ferruginei-pilosis, foliis pinnatinervibus in sicco ferruginei-nigris, capitulescentiis capitulorum 8-10 com- positis, pedunculis ultimis 1-5 cm longis dense persistente brevipi- losis, capitulis hemisphaericis ca. 4 cm latis trans radios expansos (vs. 1.5-2.0 cm), flosculis radii ca. 21 ligulis plerumque 15-20 mm longis (vs. flosculis radii 13 ligulis 8-10 mm longis). Shrub or small tree 1.7-3.0 m high. Stems angulate, brownish black, densely and persistently short pilose but glabrescent with age. Leaves drying brown- ish black, mostly 10-20 cm long, 3-8 cm wide; petioles 1.5-4.0 cm long; blades elliptic-ovate, pinnately nervate, the nerves usually persistently pubescent with appressed rusty hairs, the margins serrate. Heads 10-20, arranged in irregu- lar cymose panicles, the ultimate peduncles mostly 2-5 cm long, densely pilose with persistent rusty red short hairs. Involucres hemispheric, 15-20 mm across, 8-12 mm high, the bracts linear-lanceolate, 3-4 seriate, subequal, loosely im- bricate and usually reflexed for about 1/2 their length, rusty red pubescent throughout. Receptacle hemispheric, the bracts persistent, shorter than the florets, the apices mostly obtuse. Ray florets pistillate, fertile, the ligules white, mostly 15-20 mm long, 2-4 mm wide. Disk florets numerous, the corollas yel- low, ca. 3 mm long, the tubes ca. 0.8 mm long, sparsely pubescent. Anthers Turner & Panero: New species and combinations in Podachaenium 145 brownish black. Achenes, those of the ray, 3 sided, 3 awned, those of the disk radially flattened, the body about 2 mm long, pubescent above, the pappus of 2 coarsely hispidulous awns ca. 1.5 mm long, between these 1-4 minute deciduous scales. ADDITIONAL SPECIMENS EXAMINED: MEXICO. Chiapas: al NW de la Reserva del Triunfo, en el Cerro del Filo, ca. 1980 m, 11 May 1982, Calzada et al. 8776 (TEX); Reserva El] Triunfo, path to Cerro El Triunfo, 2000 m, Apr 1989, Heath & Long MA14 (TEX). Podachaenium chiapanum is distinguished from both P. pachyphyllum and P. standleyi by a number of characters, as indicated in Table 1. Podachaenium eminens (Lag.) Sch.-Bip. This widespread species with its distinctive broad, palmately veined, leaves offers no problem in identification and is not considered further here. Excellent illustrations of the taxon have been provided by Nash (1976) and McVaugh (1984). The long ignored Podachaenium andinum E. André (Rev. Hortic. 64:414, 1892.), from the illustrations accompanying its description, appears to belong to P. eminens. Podachaentum pachyphyllum (Sch.-Bip. ez Klatt) Jansen, Harriman, & Ur- batsch, Syst. Bot. 7:482. 1982. Altamirania pachyphylla Greenm. Aspilta pachyphylla Sch.-Bip. ez Klatt Aspihopsis pachyphylla (Greenm.) Greenm. Calea pachyphylla (Sch.-Bip. ez Klatt) S.F. Blake Jansen, Harriman, & Urbatsch (1982) have discussed adequately the nomen- clatural history of this taxon, noting most of its salient features. The species ; is known only from northcentral Oaxaca. Jansen et al. (1982) included within their concept of this species material from Guatemala which we consider to be sufficiently distinct from P. pachyphyllum so as to be recognized as a species, as noted in Table 1, and by the account that follows. Podachaenium standleyi (Steyermark) B. Turner & J. Panero, comb. nov. BASIONYM: Calea standleyi Steyermark, in Standley & Steyermark, Publ. Field Mus. Nat. Hist., Bot. Ser. 22:299. 1940. Verbesina standleyi (Steyermark) D. Nash. 146 PHT OL OG 1A volume 73(2):143-148 August 1992 Table 1. Comparison of characters. P. pachyphyllum 'P. chiapanum | P.standleyi | 1. Stems: glabrescent 2. Leaf blades: trullate ovate widest below middle mostly with 3 principal nerves glabrous drying green 3. Petioles: ca. 1 cm long 4. Capitulescence: heads 15-30 | 15-30 cm across 5. Peduncles (ultimate): glabrescent 6. Heads: globose ray florets ca. 13 8-10 mm long 7. Involucres: 3-4 mm high bracts biseriate appressed 8. Disk Achenes: glabrous or nearly so densely pilose ovate trullate widest near middle pinnately nervate persistently pubescent along veins beneath drying rusty black 1-2 cm long heads 10-20 8-10 cm across persistently rusty pilose hemispheric ray florets ca. 21 15-20 mm long 6-8 mm high bracts 3-4 seriate loose and spreading like the rays markedly coarsely hispid above glabrous elliptic widest near middle pinnately nervate glabrous drying green 2-7 cm long heads 15-30 15-40 cm across sparsely white pilose to glabrate + globose ray florets 8-11 12-15 mm long 3-6 mm high bracts biseriate appressed glabrous 147 New species and combinations in Podachaenium Turner & Panero: from holotype Podachenium Biget 148 PHY? OE OxGiA volume 73(2):143-148 August 1992 The type of this taxon is from Guatemala: Dept. San Marcos, between Todos Santos and Finca E] Porvenir, Volcan Tajumulco, 1300-3000 m, 1 Mar 1940, Steyermark 37004 (HOLOTYPE: F!). Jansen et al. (1982) also exam- ined the holotype but thought it not sufficiently distinct from P. pachyphyllum so as to be distinguished. In addition to the holotype we have examined five other collections of Podachaenium standleyi, all from the Department of San Marcos, as follows (Steyermark 36277, 36859, 36341; Williams et al. 26269, 27189 — all these at F). The most obvious differences that distinguish P. standleyi from P. pachyphyllum are the leaves, the former having trullate to ovate leaves with three principal nerves, being widest well below the middle, the petioles mostly 2-7 cm long (vs. elliptical and pinnately nervate with petioles 1-2 cm long). Additional differences are listed in Table 1. On total characters, however, P. standley: appears to stand somewhat closer to P. pachyphyllum than it does to P. chiapanum. ACKNOWLEDGMENTS We are grateful to Guy Nesom for the Latin diagnosis and to him and R. Jansen for reviewing the manuscript. James Grimes kindly provided critical literature relating to the present study. LITERATURE CITED Jansen, R.K., N.A. Harriman, & L.E. Urbatsch. 1982. Squamopappus gen. nov. and redefinition of Podachaenium (Compositae: Heliantheae). Syst. Bot. 7:476-483. McVaugh, R. 1984. Podachaenium in Flora Novo-Galiciana, 12:752-755. Nash, D. 1965. Podachaenium, in Flora of Guatemala. Fieldiana 24, pt. 12:289-290. Robinson, H. 1978. Studies in the Heliantheae (Asteraceae). X. The rela- tionship of Calea skutchi. Phytologia 38:413-414. Phytologia (August 1992) 73(2):149-154. FOUR NEW SPECIES OF CLIBADIUM FROM NORTHERN SOUTH AMERICA (ASTERACEAE: HELIANTHEAE) Harold Robinson Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 U.S.A. ABSTRACT Clibadium funkiae is described from Colombia and C. alatum, C. napoense, and C. zakii are described from Ecuador. KEY WORDS: Asteraceae, Heliantheae, Clibadtum, Ecuador, Co- lombia The descriptions provided here are the result of a review of Clibadtum for the Flora of Ecuador undertaken prior to sending the material on loan. Dur- ing the study, three undescribed species have been discovered from Ecuador. Resolution of material from adjacent Colombia has shown an additional new species from Antioquia. While the new species are described below, it should be noted that two previously described species of the present author from Pert are now considered synonyms, C. rimachii H. Robinson (1988) is the same as C. divaricatum S.F. Blake, and C. vargasianum H. Robinson (1979) is only an unusually hirsute variant of the widespread C. surinamense L. Clibadium funkiae H. Robinson, sp. nov. HOLOTYPE: COLOMBIA. Antioquia: 3 km SE of Santa Elena on road from Medellin to Santa Elena and Rio Negro, 15 km NW of Rio Negro (centro), 2450 m, 30 Jul. 1979, Stuessy & Funk 5709 (US). PARATYPES: COLOMBIA. An- tioquia: Municipio Frontino, Corregimiento Nutibara, region Murri car- retera hacia La Blanquita, Finca Palmera, 1700 m, 15 Jul. 1986, Acevedo, Martinez, Orrego, Restrepo, Sanchez, & Silva 1344 (NY,US). Municipio de Yarumal, 2 km antes del alto de Ventanas, camino a Vereda Alegre, 2030 m, 20 Aug. 1986, Callejas, Churchill, Acevedo, & Saldarriaga 2488 (HUA,NY,US). Municipio de Caldas, Vereda La Corrala, 2440 m alt., 14 Sept. 1987, Escobar, Veldsquez, & Marulanda 7911 (US). Municipio 149 150 Po YT OL. O'GIA volume 73(2):149-154 August 1992 Frontino, road to Murri, 15 km W of Nutibara (Altos de Cueva), 1850 m, 17 Oct. 1987, Brant & Martinez 1378 (MO,US); km 17 of road Nutibara - La Blanquita, region of Murri, 1860 m, 3 Nov. 1988, Zaruccht, McPher- son, Roldan, & Escobar 7097 (MO,US) region of Murri, c. 13 road-km from Nutibara, 2000 m, 9 Dec. 1988, McPherson 18396 (MO,US). Plantae fruticosae scandentes ad 2 m altae et ad 6 m longae; caules atrobrunnescentes teretes vel subhexagonales hispiduli et leniter retrorse vel antrorse appresse strigillosi, ramis ad angulum ca. 90° patentibus. Folia opposita, petiolis 1-3 cm longis; lam- inae ovatae plerumque 7-14 cm longae et 2.5-6.0 cm latae base acutae vel leniter acuminatae margine antrorse serrulatae apice anguste breviter acuminatae supra et subtus appresse strigillosae quinquenervatae, nervis secundariis majoribus e fere ad basem et 0.8-2.0 cm supra basem ad marginem inferiorem subparallelis. In- florescentiae in ramis foliosis et in ramulis recte patentibus termi- nales subdense corymbosae, ramulis dense puberulis, in capitulis sessilibus vel subsessilibus 1-3 terminatis. Capitula 4-6 mm alta et 2-4 mm lata; bracteae involucri steriles 2-4 et bracteae feminei 3 vel 4 subcoriaceae late ovatae vel late oblongae 3-5 mm longae et 1.5- 3.0 mm latae apice obtusae extus strigillosae, bracteae masculini 3 vel 4 scariosae obovatae 2-4 mm longae et 1.0-1.5 mm latae apice leniter erosae minute fimbriatae extus glabrae. Flores radii 3 vel 4; corollae albae tubulares 2.5-3.0 mm longae, lobis 0.3-0.5 mm longis extus subglabris vel pauce glanduliferis et dense argute breviter se- tuliferis. Flores disci 4-6; corollae albae 3.0-4.5 mm longae, tubis basilaribus 0.5-1.0 mm longis, faucibus 2-3 mm longis, lobis 0.5-0.8 mm longis et 0.3-0.7 mm latis extus dense longe argute setuliferis, ductis validis corollarum 10 ad nervos et mediolobatis; thecae an- therarum nigrae 1.3-2.0 mm longae, appendicibus apicalibus nigris ovatis 0.2-0.3 mm longis. Achenia radii biconvexa obovata ca. 2.5 mm longa 1.5 mm lata distaliter dense argute pilosula, rostris api- calibus minutis deciduis; achenia disci 2.0-2.5 mm longa pilosula superne densiores. Grana pollinis in diametro ca. 25 um. The Antioquian species is distinct from the widely distributed Chbadium surinamense L. by the appressed strigillose pubescence of the leaves, the bear- ing of the heads often in sessile or subsessile groups of 2 or 3, and the 10 strong ducts in the disk corollas. The specimens with branches have those branches spreading at right angles in a manner not seen in C. surinamense. Material of the new species shows some variation. The type and Escobar et al. 7911 from 2450-2500 m alt. have spreading to slightly retrorse hairs on the stems and the heads are mostly 5-6 mm long with 4 sterile bracts. Most of the spec- imens from 1700-2000 m have antrorsely strigillose stems and smaller heads Robinson: Four new Clibadium from South America 151 with about 2 sterile bracts. One specimen, Callejas et al. 2488 from 2030 m is like the type specimen in its heads and like the lower elevation material in its stem pubescence. The specimens were originally identified as Clbadium pentaneuron S.F. Blake because of the two pairs of well developed secondary veins near the base of the leaf blade, but the Blake species has a broader, divaricately branched in- florescence and is restricted to southwestern Colombia. Clibadium pentaneuron is closer to C. larum S.F. Blake of Ecuador and southwestern Colombia. The specimens are cited from wet montane forests, edge of primary forests and in a transition zone to subparamo. Clibadium alatum H. Robinson, sp. nov. HOLOTYPE: ECUADOR. Carchi: environs of Maldonado, wet montane forest, alt. 1450-1650 m, 3 June 1978, Madison, Plowman, Kennedy, & Besse 4940 (US). Plantae suffruticosae ad 6 m altae; caules atrobrunnescentes teretes dense appresse strigillosi, ramis ad angulum 30°-50° as- cendentibus. Folia opposita, petiolis 3.5-14.0 cm longis ad basem distincte anguste alatis; laminae late ovatae vel suborbiculares 17- 29 cm longae et 8-23 cm latae base late subcordatae vel sub- truncatae vel rotundatae abrupte in alis petiolarum acuminatae margine crenato-serratae vel dentatae apice breviter acutae bre- viter acuminatae supra minute strigillosae subtus appresse strig- illosae e 4-7 cm supra basem valde trinervate, nervis secundariis majoribus ascendentibus, nervis secundariis inferioribus 2-4 mi- noribus congestis recte patentibus. Inflorescentiae in ramis foliosis terminales late corymboso-cymosae ad 20 cm latae, ramulis dense puberulis, ramulis distalibus seriate cymosis subdense sessiliter vel subsessiliter capituliferis. Capitula ca. 5 mm alta et 3.5-4.0 mm lata; bracteae involucri steriles 3 vel 4 et bracteae feminei 7-9 sub- coriaceae late ovatae vel late oblongae 4-5 mm longae et 1.8-3.5 mm latae apice obtusae vel breviter acutae distaliter breviter ciliatae extus distaliter pauce appresse strigillosae; bracteae masculini 6 vel 7 scariosae ca. 3 mm longae 1-2 mm latae distaliter erosae breviter ciliatae. Flores radii 7 vel 8; corollae albae tubulares ca. 2 mm longae, lobis 0.3-0.5 mm longis glabris. Flores disci ca. 8; corollae albae 3.5 mm longae, tubis basilaribus ca. 0.8 mm longis, faucibus anguste campanulatae 2.0-2.2 mm longae, lobis 0.5-0.7 mm longis et 0.4-0.5 mm latis extus dense scabridulis, ductis 5 ad nervos sub sinubus terminatis; thecae antherarum nigrae ca. 1.5 mm longae, appendicibus apicalibus nigris ovatis ca. 0.3 mm longis. Achenia radii biconvexa submatura ca. 1.8 mm longa 1.2 mm lata distaliter 152 PY OL O.G TA volume 73(2):149-154 August 1992 dense argute hispidula, rostris apicalibus subpersistentibus glabris 0.6-0.8 mm longis; achenia disci 2.0-2.7 mm longa inferne glabra distaliter dense argute pilosula. Grana pollinis in diametro ca. 23 pm. The species has leaves with large blades and very long petioles, but seems to be the only member of the genus with distinct wings on the petioles. The wings reach the bases of the petioles, and the wings of adjacent leaves nearly meet at the node. The rostrum of the ray achenes is longer than those in most species of Clibadium. Clibadium napoense H. Robinson, sp. nov. HOLOTYPE: ECUADOR. Napo: Canton Napo, Zatzayacu, alt. 500 m, 28 Mar. 1935, Mezza 7110A (US). PARATYPES: ECUADOR. Napo: Rio Napo, Chiroisla, alt. 250 m, 24 Aug. 1979, Holm-Nielsen, Jaramillo, & Coello 19785 (AAU,QCA, US). Left margin of Rio San Miguel, near outlet to Rio Putumayo, ca. 200 m, 1 Aug. 1980, Andrade 33147 (AAU,QCA,QNA,US). Canton Archidona, N bank Rio Suno, 15 km NW of Loreto, 8 km W of El Pro- greso, 600 m, 12 Dec. 1989, Neill, Hurtado, & Alvarado 9156 (MO,US). Plantae fruticosae 1.5-2.5 m altae; caules brunescentes teretes vel subhexagonales dense strigillosi, ramis ad angulum 30° vel 50° patentibus. Folia opposita, petiolis 1-5 cm longis; laminae ovatae plerumque 10-18 cm longae et 3.5-8.5 cm latae base rotundatae vel obtusae interdum breviter acuminatae margine appresse antrorse serrulatae apice anguste breviter acuminatae supra minute scabridu- lae planae subtus appresse strigillosae inferne quinquenervatae, nervis secundariis fere ad basem et e 1-2 cm supra basem insertis ad marginem basilarem plerumque subparallis. Inflorescentiae in ramis foliosis terminales laxe corymbosae, ramulis seriate cymosis dense rigide puberulis. Capitula sessile contigua vel non contigua 3.0-3.5 mm alta et ca. 3 mm lata; bracteae involucri steriles 3 vel 4 et bracteae feminei 5 vel 6 subcoriaceae interiores aliquatum scar- iosae majores late ovatae vel late oblongae 3.0-3.5 mm longae et 2.5-3.5 mm latae apice obtusae vel breviter acutae margine bre- viter ciliatae extus distaliter scabridulae; bracteae masculini tenu- iter scariosae ca. 2.5 mm longae. Flores radii 5 vel 6; corollae albae tubulares 1.8-2.2 mm longae, lobis 3 ad 0.5 mm longis ex- tus subglabris vel breviter puberulis. Flores disci 8-11; corollae 2.8-3.0 mm longae, tubis basilaribus 0.6-0.8 mm longis, faucibus 1.5-1.7 mm longis, lobis 0.6-0.7 mm et longis 0.4-0.6 mm latis ex- tus dense breviter argute puberulis, ductis 5 ad nervos dispositis Robinson: Four new Clibadium from South America 153 sub sinubus terminatis; thecae antherarum nigrae 1|.0-1.2 mm lon- gae, appendicibus apicalibus nigris ovatis ca. 0.25 mm longis et 0.2 mm latae. Achenia radii biconvexa ca. 2.5 mm longa 2 mm lata supra mediam dense argute pilosula, rostris minutis deciduis breviter lobatis; achenia disci 2.0-2.2 mm longa argute pilosula su- perne densius pilosula. Grana pollinis in diametro 23-25 ym. The new species is related to Clibadium glabrescens S.F. Blake, which also occurs on the eastern slopes of Andes in an area reaching farther north into southernmost Colombia and farther south to the Ecuadorian Department of Morona-Santiago. The Blake species is distinct in its glabrous stems. The new species seems to occur primarily on wooded or sandy streambanks. Clibadium zakii H. Robinson, sp. nov. HOLOTYPE: ECUADOR. Bolivar: Carretera Chillanes - Bucay, en la hacienda “Tiquibuso” del Sr. Gonzalo Gomez, alt. 2100 m, 10 Sept. 1987, Zak & Jaramillo 2881 (MO,US). Plantae fruticosae scandentes 2-3? m altae; caules atrobrunnes- centes teretes vel subhexagonales dense hispiduli, ramulis ad angu- lum 45°-60° patentibus. Folia opposita, petiolis 1.0-1.3 cm longis; laminae ovatae plerumque 7-8 cm longae et 4.0-4.5 cm latae base obtusae vel breviter acutae non acuminatae margine minute ser- tulatae apice anguste breviter acuminatae supra sparse scabridae et dense minute scabridulae subtus plerumque in nervis et nervulis hispidulae e 0.5-1.0 cm supra basem trinervatae, nervis secundariis basilaribus pertenuibus. Inflorescentiae in terminis ramis foliosis et in axiles superioribus dispositae, ramulis rotundate dense corym- bosis distaliter non seriate cymosis, bracteolis lanceolatis 2-3 mm longis et ca. 0.5 mm latis hispidulis, pedunculis 0-2 mm longis dense hispidulis. Capitula subcontigua 4-5 mm alta et 3 mm lata; bracteae involucri steriles 2 vel 3 coriaceae ovatae ca. 4 mm longae et 2.5 mm latae acutae margine ciliatae extus dense strigillosae; bracteae feminei ca. 5 subcoriaceae oblongae ca. 4 mm longae et 2.2-2.5 mm latae apice obtusae margine superne in pilis brevibus argutis fimbriatae extus subglabrae; bracteae masculini 1 vel 2 obo- vatae scariosae ca. 2.5 mm longae et 1 mm latae. Flores radii ca. 5; corollae albae tubulares ca. 2.5 mm longae breviter puberulae et dense breviter glanduliferae supra mediam et in lobis densiores, lo- bis 3, 0.3-0.4 mm longis. Flores disci ca. 5; corollae albae supra me- diam in pilis argutis et glandulis brevibus obsitae in lobis densiores, tubis basilaribus ca. 0.4 mm longis, faucibus ca. 1.5 mm longis, lo- bis ca. 0.5 mm longis et latis, ductis 5 ad nervos et 3-5 mediolobate 154 Peay 2.00 O'G TA volume 73(2):149-154 August 1992 dispositis; thecae antherarum nigrae ca. 1.2 mm longae, appendi- cibus apicalibus ovatis 0.50-0.55 mm longis et 0.25-0.30 mm latis. Achenia radii biconvexa immatura ca. 1.5 mm longa, 1.2 mm lata supra mediam pauce argute pilosae et dense breviter glanduliferae, rostris deciduis ca. 0.3 mm longis late infundibularibus; achenia disci immatura ca. 1 mm longa supra mediam flexuose argute pilo- sula in partibus terminalibus densiores. Grana pollinis in diametro 25-27 wm. The species was first recognized as distinct by the rounded corymbose form of its inflorescence branches. The inflorescence does not have distal branches bearing any cymose series of sessile heads of the type found in more common members of the genus,’ such as Clibadium surinamense. The new species is also distinguished from C. surinamense by the presence of reddish ducts in the middle of the lobes of the disk corollas. The anther appendages are longer than those seen on any other species in the study, twice as long as the appendages of some other species. The numerous, short, glandular hairs of the distal halves of the corollas and disk achenes are also rather distinctive, but similar glands have been seen on the disk achenes of one specimen of C. surtnamense (Ecuador: Los Rios, Dodson, Dodson, Embree, & Perry 7037 [SEL,MO,US}) and may occur more widely in that species. LITERATURE CITED Robinson, H. 1979. New species of Clibadium, Heterosperma and Mullerza from Peru (Heliantheae: Asteraceae). Wrightia 6:46-48. pl. 82-85. 1988. Studies in the Heliantheae (Asteraceae). XXXXII. New species of Clibadium, Monactis and Verbesina from the northern Andes. Phytologia 65:50-61. Phytologia (August 1992) 73(2):155-158 CROTALARIA INCANA VAR. GRANDIFLORA (LEGUMINOSAE), A NEW VARIETY FROM SOUTH AMERICA D. Windler, L. Adler, & S. Skinner Department of Biological Sciences, Towson State University, Towson, Maryland 21204 U.S.A. ABSTRACT Crotalaria incana var. grandiflora, var. nov., a large flowered variety from Colombia, Ecuador, Peru, Bolivia, and Paraguay is de- scribed. KEY WORDS: Crotalaria, Leguminosae, Colombia, Ecuador, Peru, Bolivia, Paraguay, South America During preparation of a treatment of the genus Crotalaria for Flora Neotrop- ica, an analysis of the variation in the wide ranging and weedy species C. incana L. was conducted (Adler & Windler 1983). The variation in morpho- logical characters was examined in the light of the differences in geography and topography. Varieties incana and australis have the broadest range in the New World and are distinguished primarily by their indument types. Va- riety australis is characterized by its dense spreading trichomes and variety mcana is characterized by a more sparse, loosely appressed indument. Senn (1939) described var. nicaraguensis to recognize Central American plants with glabrous or nearly glabrous stems and fruits. Windler & Adler’s study found a fourth variety, characterized by flowers that are substantially larger than those of the other three varieties. Bentham had affixed the name C. grandiflora to a herbarium sheet but never submitted it for publication. N.L. Britton (1889) included the epithet in a paper, enumerating the collections of H.H. Rusby in 1885 and 1886, but did not validate it. In order to make the varietal name available for our use, we submit the following diagnosis: Crotalaria incana L. var. grandiflora Windler, Adler, & Skinner, var. nov. TYPE: BOLIVIA. Depto. La Paz, Prov. Sud Yungas, Yanacachi, abajo del “Hogar Fe y Alegria,” 1950 m, exp. SSE, 20-30 degree slope sendero por abajo, hacia chacras y el rio, Matorral, May 4, 1987, R. Seidel & E. Richter 874 (HOLOTYPE: US!). 155 156 PHY TOL OGIA volume 73(2):155-158 August 1992 Crotalaria incana L. forma microphylla Chodat & Hassler, Bull. Herb. Boissier 4(2nd series):835. 1904. TYPE: PARAGUAY. Cordillera, San Bernardino, E. Hassler 3526 (HOLOTYPE: G!). Diagnosis: Flores grandes (14-15 mm longi), calycibus 11-13 mm longis, carinis 11-14 mm longis (Fig. 1). This large flowered variety of Crotalaria incana is native to upper eleva- tions in tropical South America and appears at lower elevations in the more temperate climates of southern South America. Representative collections are as follows: COLOMBIA. Atlantico, Nr. Barranquilla, Dugand 127 (F). Bolivar, Nr. Cartagena, roadsides nr. sea-level, Killip & Smith 14160 (F). Magdalena, Isla de Salamanca, Romero-Castaneda 10523 (MO). ECUADOR. Azway, Valley of Rio Paute, between Paute and Cuenca, 7200-8000 ft., Camp E-2318 (F,MO). Chimborazo, Nr. Ituigna, Rio Chanchan, 4000-4500 ft., Camp E-2975 (F). Loja, Nr. Loja, Steyermark 54848 (F,MO). PERU. Amazonas, Prov. Bongara, on road to Pomacocha, 13 km N of Puente Ingenio, Hutchison & Wright 6834 (F). Ancash, Huaraz, Banos de Chancos, Sandeman 4603 (F). Apurimac, Prov. Abancay, Quebrada, 1 km E of Abancay, Stork et al. 10554 (F,MO). Cuzco, Pauwcartamboa, Balls B6677 (F). Junin, Tarma, Carretera Central, 1-3 km E of Palca, Edwin & Schunke 3934 (F,IAN). Piura, 40-43 km E of Olmos on road to Pucara, A. Gentry et al. 22660 (BALT). Puno, Region of Puno, Soukup 516 (F). BOLIVIA. La Paz, Coripati Yungas, Bang 2067 (F,MO). Tarija, Ruta Tarija- Ville Montes, Serere, Krapovickas et al. 19149 (CTES). PARAGUAY. Amambay, Sierra de Amambay, M, Hassler 12089 (RB). Caaguazu, in region of Igatimi, in field, E. Hassler 4723 (P). Cordillera, San Bernardino, Costa de Lago Ipacaray, Quarin et al. 1475 (CTES). San Pedro, Rio Tapiracuay, 8 km de San Estanislao, Krapovickas et al. 18911 (CTES). ARGENTINA. Entre Rios, Depto. Concordia, alrededores del obrador Salto Grande, N.S. Troncoso et al. 1714 (BALT). ACKNOWLEDGMENTS The authors thank Dr. Dan H. Nicolson of the Smithsonian Botany Depart- ment and Dr. Guy L. Nesom for reviewing and commenting on the manuscript, and Dr. Nesom for preparing the Latin diagnosis. Flower drawings were pre- pared by Ms. Carolyn Westbrook. 157 New variety of Crotalaria incana Windler, et al.: eIO/oueib “JEAN “q 718 Jayyoiy A japles BUCDU! “JCA ‘S eo 8ZI1 uelseg 158 PILY T OW O'G TA volume 73(2):155-158 August 1992 REFERENCES Adler, L. & D.R. Windler. 1983. An analysis of the subspecific classification in the New World Crotalaria incana L. Unpublished. Britton, N.L. 1889. An enumeration of the plants collected by Dr. H.H. Rusby in South America. 1885-1886. VII. Bull. Torrey Bot. Club 16:192. Senn, H.A. 1939. The North American species of Crotalaria. Rhodora 41:317-366. Phytologia (August 1992) 73(2):159-165. STUDIES ON MIKANIA (COMPOSITAE: EUPATORIEAE) - XVIII. NEW SPECIES FROM BRAZIL Walter C. Holmes Department of Biology, Baylor University, Waco, Texas 76798-7388 U.S.A. ABSTRACT Mikania nana, M. pacei, and M. reynoldsii, three new species from Brazil, are described and illustrated. KEY WORDS: Compositae, Eupatorieae, Mzkanza, Minas Gerais, Parana, Brazil Continued study of the genus Mikania has resulted in the recognition of the following new species from Brazil. . Mikania nana W. Holmes, sp. nov. (Fig. 1). TYPE: BRAZIL. Parana: Rio Pequena (mun. Sao José dos Pinhais), 900 m, 5 November 1969, Gert Hatschbach 22847 (HOLOTYPE: MBM). Species ad Mikaniam vimineam DC. similis sed differt foliis elliptico-ovatis (non linearibus) et petiolatis (non sessilibus). Semidecumbent to erect; stem 21 cm long, striate, villous to semitomentose. Leaves opposite, lanceolate to elliptic ovate, 1.5-3.6 x 1.0-1.7 cm, bases cuneate to attenuate, with 3 conspicuous nerves, usually with another pair of obscure nerves below these, apices acute, margins irregularly serrate to doubly serrate, apices of teeth denticulate, upper surfaces pilose, muricate, lower surfaces pilose particularly on the nerves; petioles 0.3-0.5 cm long, sparingly pilose; internodes 1-5 cm long, gradually becoming longer above. Capitulescence racemo-corymbose, dense, 2.0 x 3.5 cm; branchlets villous. Heads borne more commonly in clusters of three, the central one nearly sessile, the outer short pedunculate, or heads borne singly; ultimate branchlets (peduncles) 0.5-2 mm long, villous. Heads 9-10 mm long; subinvolucral bracts linear to oblanceolate, 5-7 mm long, obscurely nerved, apices pilose. Phyllaries oblong to oblong 159 160 PHYTOLOGIA ,\\ volume 73(2):159-165 August 1992 Wun Figure 1. Habit of Mikania nana W. Holmes. Holmes: Studies on Mikania X VIII 161 Figure 2. Mikania pacei W. Holmes. D. papain E. leaf; and F. head with subinvolucral bract. 162 Peay Om OrG LA volume 73(2):159-165 August 1992 Figure 3. Mikania reynoldstt W. Holmes. A. leaf; B. stem and leaves; and C. capitulescence and stipulelike appendage. almies:. Giudienouulibania, xviii 163 ovate, 5.0-5.5 mm long, mostly glabrous, bases slightly gibbous, apices acute to obtuse, slightly pilose. Corollas white, 5.0-5.3 mm long, tubes 2.5-3.0 mm long, throats semicampanulate to funnelform, 1.00-1.75 mm long, teeth ovate, 1.00-1.25 mm, densely villous; stigmatic surfaces densely hirsute. Achenes 2.0-2.7 mm long, lightly villous. Pappus bristles 50-60, 5-6 mm long, margins scabrid. The new species is known only from the type material. It is characterized by its diminutive size, racemo-corymbose capitulescence, 9-10 mm long heads, densely villous corolla teeth, and hirsute stigmatic surfaces. Mikania pacei W. Holmes, sp. nov. (Fig. 2). TYPE: BRAZIL. Minas Gerais: Senador Mourao (mun. Diamantina), 24 January 1978, Gert Hatschbach 40884 (HOLOTYPE: MBM; Isotype: BAYLU). Species ad Mikaniam obtusum DC. similis sed differt foliis ovatis et orbicularibus (non oblongis) cum basi auriculata (non obtusa). Erect, 1.0-1.3 m tall; stems terete, glabrate, internodes 4-5 cm long. Leaves opposite, broadly ovate to nearly orbicular, 2.5-5.0 x 1.8-3.8 cm, bases auricu- late, apices rounded and apiculate, margins entire, venation semipalmate from just above the base of the blade; upper and lower surfaces glabrate, reticu- late; petioles ca. 1 mm long, 2 mm wide (the leaves appearing nearly sessile), glabrous. Capitulescence paniculate, 15-23 x 15-20 cm, heads in corymbiform clusters of ca. 8 x 7 cm; lower bracts leaflike, upper bracts becoming smaller, bases cuneate; branchlets glabrous, terete; ultimate branchlets (peduncles) 4- 9 mm long, thin, glabrous. Heads 8-9 mm long. Subinvolucral bracts linear, glabrate, borne at base of the peduncle. Phyllaries ovate to elliptic oblong, 4.0-4.8 mm long, apices rounded, puberulent, margins ciliolate, especially in the upper half. Corollas cream, ca. 4.5 mm long, tube ca. 2 mm long, throat funnelform to semicampanulate; teeth ovate, ca. 1 mm long. Pappus bris- tles white, 38-40, 5.0-5.5 mm long, margins scabrid. Achenes 2.2 mm long, gradually tapering from summit to base, brown. PARATYPE: BRAZIL. Minas Gerais: 15 km ao norte de Cunha Mag- alhaes'(mun. Cunha Magalhaes), 20 January 1972, Gert Hatschbach 29013, L.B. Smith, & E. Ayensu (MBM,BAYLU). Mikanta pacer is similar to M. obtusata DC., a species with oblong leaves with acute to obtusely narrowed bases (Barroso 1958). Mikanza pacez has broadly ovate to nearly orbicular leaves with auriculate bases. The species is named in honor of Lula Pace: botanist, geologist, and first woman on the Baylor University faculty to hold the doctor of philosophy degree (Trantham 1925). Her plant collections made between 1903-25 are the basis of the university collection. Dr. Pace, one of Baylor’s most revered professors, 164 PHYTOLOGTA volume 73(2):159-165 August 1992 made significant original contributions to the botanical literature, primarily in cytology and embryology. Mikania reynoldsii W. Holmes, sp. nov. (Fig. 3). TYPE: BRAZIL. Minas Gerais: Trilha dos Garimpeiros (mun. Grao Mogol), 1100 m, 12 Febru- ary 1991, Gert & Marta Hatschbach 55084 & O.S. Ribas (HOLOTYPE: MBM; Isotype: BAYLU). Species ad Mikaniam micranthum HBK. similis sed differt foliis sagittatis ad subhastatis (non triangulo-ovatis) et pseudostipulis manifestis (non obscuris). Twining vines to 5-6 m long; stems terete, striate, glabrate; internodes 7-13 cm long. Leaves opposite, triangular, blades 5-8 x 3.5-5.0 cm, bases sagittate to subhastate, acute at the insertion of the petiole, trinerved from the bases; apices acute to acuminate, margins entire to remotely denticulate, glabrous above, with resinous glands below; petioles 2.0-4.5 cm long, thin, glabrate; op- posite petiole bases connected by stipulelike appendages ca. 5 mm wide and 1.5 mm long, margins serrate to fimbriate to lobed. Capitulescence a compound corymb, ca. 10 x 10 cm; heads disposed in corymbs, 4 x 6 cm; branchlets glabrate, angular; bracts similar to leaves but reduced in size. Heads 6.5-7.0 mm long; ultimate branchlets (peduncles) 0.5-3.5 mm, glabrous; subinvolu- cral bracts elliptic to obovate, ca. 3 mm long, acute to acuminate, glabrous. Phyllaries elliptic oblong to ovate, ca. 4 mm long, acute, glabrous except for the puberulent to slightly ciliolate apices. Corollas white to cream, ca. 3 mm long, tube ca. 1.2 mm long, throat funnelform to campanulate, ca. 1.3 mm long, teeth deltate-ovate, ca. 0.6 mm long, acute-acuminate. Achenes ca. 2.4 mm long, tan, tapering from summit to base, with globular resinous glands. Pappus bristles 31-36, 2.5-3.0 mm long, white, scabrid. PARATYPE: BRAZIL. Minas Gerais: Grao Mogol (mun. Grao Mogol), mountains to the west of town, 1130 m, 14 Jun 1990, W.C. Holmes 5069 (BAYLU,MBM). The new species appears similar to the well known and widely distributed Mikania micrantha HBK., a species with triangular-ovate leaves with cordate bases, heads 5-6 mm, and a flat topped corymbose capitulescence (Robin- son 1934). Mikania reynoldsii is distinguished by its triangular leaves with sagittate to subhastate bases and more branched capitulescence. Both species possess stipulelike enations, but those of M. reynoldsii are considerably more prominent than those of M. micrantha. The new species commemorates Dr. Herbert -H. Reynolds, President of Baylor University, whose dynamic stewardship of the University has promoted creative activities and research. His foresightedness in recognizing the impor- tance of international studies has greatly advanced botanical studies at Baylor. Holmes: Studies on Mikanza X VIII 165 This paper and a forthcoming treatment of the Afikanza of Parana, in collab- oration with Dr. Gert Hatschbach of Museu Botanico Municipal of Curitiba, are a direct product of his patronage. ACKNOWLEDGMENTS I am grateful to Baylor University for providing financial support that made this study possible. I also wish to thank Sidney McDaniel (IBE) and Richard Carter (VSC) for review of the manuscript. Gert Hatschbach of MBM provided the specimens that made this study possible. My sincere appreciation is also extended to Lance Hampton, a Baylor undergraduate, for providing the illustrations. LITERATURE CITED Barroso, G.M. 1958. Mikaniae do Brazil. Arq. Jard. Botanico Rio de Janeiro. 16:239-333. Robinson, B.L. 1934. Mikania scandens and its near relatives. Contr. Gray Herb. 104:55-71. Trantham, H. 1925. Dr. Lula Pace: An appreciation of her life and services. Baylor Monthly 5:1-2. Phytologia (August 1992) 73(2):166-168. NEW SYNONYMS IN THE TUBER BEARING SOLANUM C.M. Ochoa International Potato Center, Genetic Resources Department, P.O. Box 5969, Lima, PERU ABSTRACT Names recently determined as synonyms of recognized taxa of tuber bearing Solanum are listed. KEY WORDS: Solanaceae, Solanum, nomenclature During preparation of my monographic work on the South American tuber bearing Solanum (sect. Petota), I found so far, some species which to me are only synonyms of those previously quoted by other authors. Among these new synonyms are the following: Solanum bukasovii Juz., Bull. Acad. Sci. U.R.S.S., Ser. Biol. 2:303. 1937. Solanum amabile Vargas, Las Papas Sudperuanas. Univ. Ncl. del Cusco, 2:60. 1956. Solanum canasense Hawkes, Bull. Imp. Bur. Pl. Breed. & Genet. Cam- bridge. 123. 1944. Solanum canasense Hawkes var. alba Vargas, Las Papas Sudperuanas. Univ. Ncl. del Cusco. 2:57. 1956. Solanum canasense Hawkes var. calcense Vargas, Las Papas Sudperua- nas. Univ. Ncl. del Cusco. 2:58. 1956. Solanum canasense Hawkes var. intthuatanense Vargas, Las Papas Sud- peruanas. Univ. Ncl. del Cusco. 2:58. 1956. Solanum canasense Hawkes var. latifolia (Vargas) Ochoa, Los Solanum Tuberiferos Silvestres del Peru. Lima, Peru. 170-175. 1962. Solanum canasense Hawkes var. lechnoviczii (Hawkes) Ochoa, Los Sola- num Tuberiferos Stlvestres del Peru. Lima, Peru. 168. 1962. 166 Ochoa: New synonyms in Solanum 167 Solanum espinarense Vargas, Las Papas Sudperuanas. Univ. Ncl. del Cusco. 2:60. 1956. Solanum hapalosum Ochoa, Bol. Soc. Arg. Bot. 22(1-4):297-299. 1983. Solanum lechnoviczii Hawkes, Bull. Imp. Bur. Pl. Breed. & Genet. Cam- bridge. 123. 1944. Solanum lechnoviczit Hawkes var. latifolia Vargas, Las Papas Sudperu- anas. Univ. Ncl. del Cusco. 2:61. 1956. Solanum ochoae Vargas, Las Papas Sudperuanas. Univ. Ncl. del Cusco. 2:59. 1956. Solanum pampasense Hawkes, Bull. Imp. Bur. Pl. Breed. & Genet. Cam- bridge. 125. 1944. Solanum pumilum Hawkes, Bull. Imp. Bur. Pl. Breed. & Genet. Cam- bridge. 124. 1944. Solanum punoense Hawkes, Bull. Imp. Bur. Pl. Breed. & Genet. Cam- bridge. 123. 1944. Solanum sicuanum Hawkes, nom. nov., The Potato, Evolution, Biodi- versity & Genetic Resources. London, Great Britain. 148. 1990. Solanum soukupi: Hawkes, Bull. Imp. Bur. Pl. Breed. & Genet. Cam- bridge. 122-123. 1944. Solanum soukupii Hawkes var. espinarense (Vargas) Ochoa, Los Solanum Tuberiferos Silvestres del Peru. Lima, Peru. 278. 1962. Solanum bukasovii Juz. var. multidissectum (Hawkes) Ochoa, comb. nov. BASIONYM: Solanum multidissectum Hawkes, Bull. Imp. Bur. PI. Breed. & Genet. Cambridge. 124. 1944. Solanum colombianum Dun. in A.DC., Prodr. 13, 1, 33. 1852. Solanum cacetanum Ochoa, Phytologia 46(7):495. 1980. Solanum marinasense Vargas, Las Papas Sudperuanas. Univ. Ncl. del Cusco. 2:53. 1956. Solanum canasense Hawkes var. rerophylla (Vargas) Ochoa, Los Solanum Tuberiferos Silvestres del Perti. Lima, Peru. 170. 1962. Solanum canasense Hawkes var. zerophilum (Vargas) Hawkes, The Potato, Evolution, Biodiversity & Genetic Resources. London, Great Britain. 138-139. 1990. 5 Solanum lechnoviczti Hawkes var. zerophylla Vargas, Las Papas Sudpe- ruanas. Univ. Ncl. del Cusco. 2:61-62. 1956. 168 PRY T OT O:GLA volume 73(2):166-168 August 1992 Solanum raphanifolium Card. et Hawkes, Jour. Linn. Soc. Bot. 53:94-95. 1946. Solanum hawkesii Card., Jour. Linn. Soc. Bot. 53:95-96. 1946. Solanum tuberosum L. subsp. andigena (Juz. et Buk.) Hawkes, Proc. Linn. Soc. Bot. 166:130-137. 1956. Solanum paramoense Bitt. ez Pittier, Man. Pl. Usual. Venez. 329. 1926. ~OMATMENI il 3 ot oS 00232 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 Interna- 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. nt te it ee a a ee a.