Bx F ATF 4 7 Ey 72 1 ee PHYTOLOGIA An international journal to expedite plant systematic, phytogeographical and ecological publication Vol.70 January 1991 No.1 CONTENTS ineicn Ee f BRR ie A ey set ie? it Tek ENS We eS Sa RR EE a 1 J.D. BACON & J.A. MCDONALD, Nama quiexobranum (Hydro- phyllaceae): a new species from Oaxaca, MExicO osc eeee. 24 -~T. DUNCAN & C.S. KEENER, A classification of the Ranunculaceae with special reference to the Western Hemisphere ............ 24 _S. BOYD, T. ROSS, & L. ARNSETH, Ceanothus ophiochilus (Rhamnaceae): a distinctive, narrowly endemic species from FAVES SIC UIE 7 CCAMLOPTM ho Siw hereyeciecdat cbaredeeeescdetayenneess 28 -B.L. TURNER, A new species of Coreocarpus (Asteraceae - Coreopsideae) from Mexico State, MEXICO o....eecceeeeeeeeeee 42 | -B.L. TURNER, A new gypsophilic species of Mirabilis (Nyc- taginaceae) from Nuevo Léon, ME RIED. huss occa ola perisaty 44 ~W.C. HOLMES, Studies on Mikania (Compositae: Eupatorieae) -XVII: two new species from Minas Gerais, Brazil .............. 47 ; 1 ee SMITH, A new Pitcairnia (Bromeliaceae) from Peru oeceses. 52 mG; a NESOM, Solidago durangensis (Asteraceae: Astereae), a new species from MERICE ar watt cesT Rea S el Ny ies Sonera a a.s 58 Contents continued on the inside cover. ‘Published by Michael J. Warnock 185 Westridge Drive Huntsville, Texas 77340 U.S.A. PHYTOLOGIA is printed on acid free paper. M.J. WARNOCK, Book reviews 23.2.4...) ssn enseeeoe 60 Books received i )..bii.cisccebsesesesscneace dann dopheuditeymsleea nanan aan 65 Errors and additions, vOluME 69 oo... cee cecceccsecneseeccesesceneenectens 67 Publications dates, volume 69. ........:.¢.:.2:.2esecesacdeneye ee 68 PHYTOLOGIA (ISSN 00319430) is published monthly by Michael J. Warnock, 185 Westridge Drive, Huntsville, TX 77340-8916. Second Class postage at Huntsville, TX. Copyright ©1990 by PHYTOLOGIA. Domestic individual subscription (6 issues): $18.00: Domestic institutional subscription (6 issues): $20.00. 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 me 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 — 77340-8916. ee aaa LIBRARY APR -8 185 Phytologia (January 1991) 70(1):1-20. NEW YORK BOTANICAL GARDEN TAXONOMY OF GENTIANELLA (GENTIANACEAE) IN MEXICO Guy L. Nesom Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT The taxonomy is detailed for the twelve Mexican species of Gen- tianella (excluding Gentianopsis). Among them are four new species, Gentianella calycidon, Gentianella fimbrilinguis, Gentianella glossocarpa, and Gentianella tarahumarae, as well as Gentianella sandiana comb. et stat. nov. Distribution maps are included. KEY WORDS: Gentianella, Gentianaceae, México A detailed revision of the North American species of Gentianella was pro- vided by Gillett (1957), but both his generic and species concepts have been considerably restricted by later students of the Gentianaceae. Particularly, the segregation of Gentianella subgenus Eublephis, the “fringed gentians,” as Gentianopsis (Ma 1951) has been generally recognized as justified (Iltis 1965). Gentianella subgenus Comastoma also has been segregated as a separate genus (Tokokuni 1961; other transfers by Holub 1967), but these species have gen- erally been retained within Gentianella in recent floristic treatments (e.g., Holmgren 1984). Gentianella and Gentianopsis both differ from Gentiana in their epipetalous nectar glands, corollas without conspicuous folded plaits between the lobes, and calyx tubes usually without an inner membranous rim. Based on these features, both genera are perhaps more closely related to Frasera and Swertia than to Gentiana (Tokokuni 1963). Gentianella differs from Gentianopsis in its much smaller corollas and smooth (vs. papillose) seeds. Gentianella is a cosmopolitan genus comprising about 100-125 species of temperate regions. The greatest number of species are found in South Amer- ica, which were treated in large part by the studies of Fabris (1960) and Pringle (1981). Other centers of lower diversity are in North America, Europe, and Asia. Eleven species of Gentianella occur in North America north of Mexico (Kartesz & Kartesz 1980) and twelve in México; four of these cross the inter- national border in their distribution. A number of these represent taxa raised from subspecific rank (sensu Gillett 1957) to specific rank by Holub (1967). 1 2 PHYTOLOGIA volume 70(1):1-20 January 1991 The latter worker provided only a short paper that presented many new com- binations but no discussion or explanation of the taxonomic decisions. The present study evaluates and summarizes the taxonomy of the Mexican species. In my overview, all of the Mexican taxa recognized by Gillett are accepted, but all except one are recognized at the specific rank; however, some of the collections studied by Gillett are identified differently. Four new species are recognized, in addition to another recently described in a separate paper (Ne- som & Turner 1990). Several of the species appear to be somewhat variable in morphology, but all are distinct and there is no evidence of intergradation among any of the taxa. In Mexico, species of Gentianella that are most closely similar among themselves appear to be allopatric (Maps 1-3) and consequently have little or no opportunity to hybridize. Many of these species apparently are extremely narrow in their geographic distribution and at least three are probably in immediate danger of extinction. Plants of the Mexican species of Gentzanella are notably similar in their vegetative morphology, and as seen in the key and reflected in the descrip- tions, features that distinguish the taxa are almost completely restricted to the flowers and fruits. Among the species included here, the color of fresh corollas appears to be either primarily yellowish or variably purplish to bluish to nearly white. The pigments, however, have a strong tendency to become yellowish when dry, and use of flower color in the identification of dried spec- imens is not particularly reliable. Gentianella Moench, Meth. Pl. 482. 1794. Taprooted annuals, glabrous, smooth or minutely papillate-scabrous. Stems erect, simple from the base or rarely few branched, often ridged. Leaves op- posite, epetiolate, commonly subclasping, entire, 3-5 nerved, the basal usually absent by flowering [in the Mexican taxa]. Flowers (4-) 5-merous, convolute in bud, mostly in compact, axillary and terminal cymes, pedicels longest on the lower nodes; calyx usually with a short tube, lacking an inner membra- nous rim, with lobes equal or unequal in length; corollas yellow to purplish blue, tubular to funnelform or campanulate, 5-25 mm long, with epipetalous nectar glands at the base of the tube, the lobes spreading to erect, without plicae in the sinuses, usually with a row of fimbriae inserted at the base of the lobes. Stamens epipetalous, the filaments often narrowly winged basally, often adnate for about half the corolla tube length, the anther thecae 0.6-1.0 mm long, positioned near the apex of the throat, included within the tube or barely exserted. Ovaries unilocular, placentation parietal, the 2 stigmas persistent on the fruit apex; fruits sessile or short stipitate, usually slightly exserted from the corolla, capsular, 2 valved, septicidal, apically dehiscent. Seeds numerous, 0.6-1.0 mm long, brownish, more or less smooth, round to slightly flattened. Base chromosome number, z = 9 in subg. Gentianella (see comments by Pringle 1981), 2 = 5 in subg. Comastoma. Taxonomy of Gentianella in México Nesom: BUROTXoM Bynoe ‘dsqns e{ [ese UuOpTOATBO TUSZTTSTA Map 1 January 1991 volume 70(1):1-20 PHYTOLOGIA a - See OG. fimbrilinguis —|—- AG. microcalyx Xa. sandiana @G. canosoi 2... ~—— 2G. glossocarpa | OG. tarahumarae Map 2 Nesom: Taxonomy of Gentianella in México ited Map 3 6 PHY TOLOGIA volume 70(1):1-20 January 1991 Type species, Gentianella tetrandra Moench (= Gentianella campestris [L.] Borner), a European species. KEY TO THE MEXICAN SPECIES 1. Calyx lobes linear lanceolate, lanceolate, or ovate lanceolate, 2-12 mm long, the sinuses acute to sharply rounded. ..............00eeeeeeee (3) 1. Calyx lobes teethlike, 0.5-2.0 mm long, usually of a thicker texture than the tube, arising from a calyx tube with a relatively flat upper margin. 5 Ege! sate acheccigedh dos: tees: 5. $20. oe (2) 2. Calyx split to the base along one side to form a membranaceous sheath: Sioned he shee .2dh ties «oye Sek ere G. wislizent 2. Calyx a continuous tube, herbaceous in texture, not membranous. b nkPaadag outeeme ha eee coon ty snes cee ae. G. calycidon 3. Calyx tube (1-)2-3 mm long, with lobes 3-12 mm long; corollas 10-18 mm CL Sn ea a eer sen ee OM Me (5) 3. Calyx tube 0.5-1.0 mm long, with lobes 2-4 mm long; corollas 6-11 mm LON, ©... 03.5 apteplete ogieen axtegizn Pence Aten mice ieas ao (4) 4. Corollas without fimbriae, the tube 4.0-6.5 mm long; calyx lobes thin herbaceous, narrowly triangular; fruits sessile. .G. microcalyz 4. Corollas with numerous fimbriae, the tube 6-8 mm long; calyx lobes fleshy, linear, fruits stipitatesseees 2s: .<.2isccc see G. tarahumarae bo: Calyx tube 2:0-3:5 min longer ie8 2... 60-0004 oe < noe od (8) o. Calyx-tube-1*2 tam long: 3i2... sessed wa vebas cs des 9 ee (6) 6. Corollas 17-20 mm long; apical extensions of fruit 1.2-1.8 mm long, MareOWwly OBLONG ao cn/sasnie.civseisursennecm eee G. glossocarpa 6. Corollas 8-16 mm long; apical extensions of fruit 0.3-0.5 mm long, ae er errs (7) 7. Calyx lobes 6-12 mm long, with minutely papillate margins; corollas 13-16 mm long, the tube 9-11 mm long, the lobes relatively even in length, with fimbriae inserted at the base of the lobes. ........... G. amarella 7. Calyx lobes 3-7 mm long, with smooth margins; corollas 8-14 mm long, the tube 4-8 mm long, the lobes strongly uneven in length, with fimbriae inserted at about midlength of the lobes. ............. G. fimbrilinguis Nesom: Taxonomy of Gentianella in México 7 8. Calyx 5-11 mm long, the tube 2-3 mm long; corollas mostly bluish or purplish before drying, 10-18 mm long. ...............2++. (10) 8. Calyx (9-)12-15 mm long, the tube 3-4 mm long; corollas yellow belore drying, 17-25 mimilong. <5 ache = sients ae enters «oleh trae (9) 9. Corollas 20-25 mm long, the lobes 7-10 mm long, with attenuate apices, the throat densely finibriate.. <0. % »2 2200s doee ieee ened G. wrighti 9. Corollas 17-21 mm long, the lobes 5-7 mm long, with acute but not attenuate apices, the throat sparsely fimbriate. ........... G. hartwegit 10. Calyx lobes 4-8 mm long; corolla fimbriae inserted well above the base of the lobes:) ! cj: s sicx ated: dopa eae G. mezicana 10. Calyx lobes 3-4 mm long; corolla fimbriae inserted at the base of GheMaper: Ween see esa tne. oC es ee waa tine seme ee te sacs es (11) 11. Pedicels and the veins, lamina, and margins of calyx lobes and tube densely scabrous with erect, prominent papillae. ........... G. canosoi 11. Pedicels smooth, the calyx sometimes with minutely papillate-scabrous margins, otherwise smooth... 5... ccnens~ s2a~aeelemaewilaeels G. sandiana 1. Gentianella amarella (L.) Borner subsp. acuta (Michx.) Gillett, Ann. Mis- souri Bot. Gard. 44:253. 1957. BASIONYM: Gentiana acuta Michx., Fl. Bor. Amer. 1:177. 1803. TYPE: [CANADA. Quebec.] Near “Ta- doussack,” A. Michaur s.n.(P). Amarella acuta (Michx.) Rafin., Fl. Tel- lur. 3:21. 1837. Ericala acuta (Michx.) G. Don, Gen. Syst. Gard. Bot. 4:190. 1837. Gentiana amarella L. var. acuta (Michx.) Herder, Trudy Imp. S-Peterburgsk. Bot. Sada 1:428. 1872. Gentianella acuta (Michx.) Hiit., Mem. Soc. Faun. Fl. Fenn. 25:76. 1950. Gentiana amarella L. subsp. acuta (Michx.) Hultén, Ark. Bot., ser. 2, 7:107. 1968. Gen- tana amarella L. {. michauziana Fern. (homotypic with Gentiana acuta), Rhodora 19:151. 1917. Gentianella amarella (L.) Borner f. michauziana (Fern.) Scoggan, Fl. Canada 1:52. 1978. Michaux’s citation reads “in montibus altibus Carolinae et Ca- nada, prope Tadoussack.” The only species of Gentianella of the Appalachian Mountains in the “Carolina” region is G. guinquefolia (L.) Small. See Gillett (1957) for numerous other North American (north of México) synonyms of Gentiana acuta. Leaves spreading-ascending, ovate-lanceolate, 3-5 nerved, 15-25 mm long. Pedicels 3-12 mm long. Calyx 7-13 mm long, the tube 1-2 mm long, the lobes 6-12 mm long, often strongly unequal in width, less so in length, herbaceous, 8 PHY TOL0 GIA volume 70(1):1-20 January 1991 the margins distinctly papillate-thickened and commonly purplish. Corollas blue to purple, 13-16 mm long, the tube 9-11 mm long, the 5 lobes 4-5 mm long, lanceolate-ovate, erect, fimbriae numerous, inserted at the base of the corolla lobes. Staminal filaments slightly winged, adnate along the basal 1/4- 1/3 of the corolla tube, anthers purplish. Fruits sessile, 12-15 mm long, not exceeding the corolla at maturity, the apices erect. Disjunct populations in Durango and Nuevo Leon, widespread in the west- ern United States, eastern to western Canada and north into Alaska, the Aleutian Islands, northeastern U.S.S.R.; in Nuevo Leon, meadows, openings in conifer forests, just below to above timberline, ca. 3400-3800 m, July- September; 2000-2100 m in Durango. Additional collections examined: MEXICO. Durango: Mpio. Pueblo Nuevo, 5 km NE of El Palmito, steep barranca wall, 19 Oct 1983, Breedlove 58881 (MO); 21.9 mi NE of El Paraiso, Sinaloa, on road between Villa Union and El Salto, steep, moist embankment, 29 Sep 1953, Ownbey & Ownbey 1984 (GH, US). Nuevo Leon: Cerro Potosi, 27 Aug 1987, Bogler & Atkins 149 (TEX); Cerro Potosi, 21 Aug 1969, Hinton, et al. 17244 (TEX); Cerro Potosi, 23 Aug 1984, Lavin 4782 (TEX); Cerro Potosi, 26 Jul 1985, McDonald 1783 (MO, TEX); Cerro Potosi, 21 Jul 1935, Muller 2247(GH, MO); Cerro Potosi, 1938, Univ. Illinois Students 971 (MO, US). Gillett (1957) treated these plants as a subspecies of Gentianella amarella (L.) Borner, while noting that the subsp. amarella is restricted to central and western Europe, completely disjunct from its North American relatives. Oubsp. amarella and another more geographically restricted subspecies have been recognized in western Europe (Clapham, et al. 1987). In comparisons of the American and Eurasian populations of G. amarella, Fernald (1917) concluded that all were conspecific, but he did not identify the origin of the extra-American specimens he compared. Fernald (1950) and Gleason & Cron- quist (1963) recognized the North American plants as G. amarella, but similar plants that extend into northeastern Russia were identified as G. acuta Michx. (Grossgeim 1952). Hultén (1968) identified the same taxon from Alaska as Gentiana amarella L. subsp. acuta (Michx.) Hultén. Compared to the American plants of Gentianella amarella, the European ones tend to have larger corollas (12-22 mm long, Clapham, et al. 1987). In the northeastern U.S.S.R., they are 9-13 mm long (Grossgeim 1952) and 10-15 mm long in Alaska (Hultén 1968). Clearly, the taxonomy of this complex needs to be studied in detail from a worldwide perspective, where there perhaps is as much justification for treating subsp. acuta and subsp. amarella at the specific rank as combining them into a single species. Disjunct populations of subsp. acuta occur in Durango and Nuevo Leon, ap- parently as southern extensions of montane population systems in California- Arizona and in New Mexico, respectively. I can find no basis for separating these Mexican plants from others in the southwestern United States. Besides Nesom: Taxonomy of Gentianella in México 9 subsp. acuta, Gillett recognized four additional subspecies within Gentianella amarella in the New World. In his view, each of the other four American sub- species of G. amarella has a more restricted geographic range, each partially overlapping with the range of subsp. acuta. Marroquin & Rzedowski (1984) recognized two taxa in the Valley of México as subspecies of G. amarella, which are treated here as the separate species G. mezicana (Griseb.) Holub and G. hartwegit (Benth.) Holub. They noted that citations of subsp. acuta from the Valley of México refer to plants of G. mezicana; Gillett’s report of subsp. acuta from the state of México was based on two collections of somewhat immature plants of G. mezicana. Previous records of subsp. acuta from Baja California are identified here as G. fimbrilinguis Nesom. 2. Gentianella calycidon Nesom, sp. nov. TYPE: MEXICO. Chihuahua: 11 mi SW of El Vergel, 9200 ft, 7 Oct 1959, D.S. Correll & H.S. Gentry 22895 (HOLOTYPE: GH!). Gentianellae sandianae (Gillett) Nesom similis sed differt lo- bis calycis dentoidibus multo brevioribusque marginibus rasilibus, corollis brevioribus, et antheris luteis differt. Leaves ovate-lanceolate, 13-17 mm long, 5-9 mm wide, trinerved, spreading, with minutely papillate-scabrous margins. Pedicels 5-9 mm long, with flowers laterally oriented to slightly nodding. Calyx 4.0 mm long, the tube 2.5-3.0 mm long, the lobes 1.0-1.5 mm long, teethlike, linear-lanceolate and thickened relative to the tube, with smooth margins, the sinuses broadly rounded to nearly flat. Corollas pale lavender, drying yellowish, 15-18 mm long, the tube 10-12 mm long, the lobes ovate, not speckled, 4-5 mm long, fimbriae numerous, inserted at the very base of the corolla lobes. Staminal filaments broadly winged on the basal third, adnate to the basal 3-4 mm of the corolla tube, the anthers yellow. Fruits sessile, mature size not observed. Known only from the type collection. The calyx of Gentianella calycidon Nesom, with its flat margined tube and short, teethlike lobes with smooth margins, is very similar to that of G. wishizeni (Engelm.) Gillett, but the calyx is not as thin textured or cut to the base as in the latter. Gentianella microcalyr (Lemmon) Gillett has a calyx with short but scabrous margined lobes and much shorter tube, blue and shorter corollas, and no fimbriae in the corolla. Gentianella calycidon is different from G. sandiana (Gillett) Nesom in its fewer flowered inflorescence, shorter and differently textured calyx lobes without scabrous margins, shorter corollas, and yellow anthers. 3. Gentianella canosoi Nesom & Turner, Sida 14:227. 1990. TYPE: MEXICO. Durango: Mpio. Pueblo Nuevo, vicinity of El Salto, pine woods, 4 Oct 1981, S. Gonzdlez & S. Acevedo 2053 (HOLOTYPE: TEX!; Isotypes: GH!, MO!). 10 PAN DOL OG LA volume 70(1):1-20 January 1991 Stems often purple, the young portions densely papillate-scabrous, smooth below or remaining slightly scabrous along the ridges. Leaves subclasping, not basally connate, trinerved, lanceolate, 15-35 mm long, 3-6 mm wide, the mar- gins minutely papillate-scabrous. Pedicels 1-4 mm long. Calyx prominently scabrous on the veins and lamina with long, erect papillae, most densely so on the veins, 5-6 mm long, the tube 2.0-2.5 mm long, the lobes linear-lanceolate, 3-4 mm long, equal in length or nearly so, spreading at the apices. Corollas yellow, drying yellow to purplish, 13-16 mm long, the tube 8-10 mm long, the 5 lobes spreading-erect, 5-6 mm long, with attenuate apices, fimbriae numerous, inserted at the base of the corolla lobes. Staminal filaments narrowly winged basally, adnate to the corolla tube for about half its length, anthers yellow. Fruits sessile, 14-18 mm long, distinctly exceeding the corollas. South-central Durango; pine-oak woodlands, rich soil, ca. 2400-2650 m; September-November. Additional specimens examined: MEXICO. Durango: Sierra Madre W of Durango, Sep-Oct 1881, Forrer s.n. (US); Mpio. Pueblo Nuevo, 6 mi W of La Ciudad on Hwy 40, at Puerto de Buenos Aires, 7 Nov 1964, Flyr 276 (TEX); Mpio. Pueblo Nuevo, 5 km SW of El Salto, 4 Oct 1981, Gonzdlez & Acevedo 2033 (TEX); Mpio. Pueblo Nuevo, along Hwy 40 at the turnoff to La Campana, 3.2 mi W of Las Adjuntas and 14.7 mi W of El Salto, 26 Sep 1973, Reveal 3458 (TEX). 4. Gentianella fimbrilinguis Nesom, sp. nov. TYPE: MEXICO. Baja California Norte: Sierra San Pedro Martir, banks of La Sanca creek, 5 mi NW of La Grulla, 6700 ft, 17 Sep 1930, 1.L. Wiggins & D. Demaree 4845 (HOLOTYPE: GH!; Isotype: F). Gentianellae amarellae (L.) Borner subsp. acutae (Michx.) Gil- lett similis sed differt lobis corollarum longis lanceolatisque in lon- gitudine inaequalibus fimbriis ad ca. medio loborum insertis et fil- amentis antherarum ad basim tubi corollae insertis. Plants 20-45 cm tall. Cauline leaves sharply ascending, ovate-lanceolate, not basally connate, 3-5 nerved, 15-40 mm long, the margins minutely papillate- scabrous. Pedicels 4-14 (-20) mm long. Calyx 4-8 mm long, the lobes smooth margined, 3-7 mm long, with sinuses acute to sharply rounded, the tube 1.0- 1.5 mm long. Corollas purplish, 8-14 mm long, the tube 4-8 mm long, the 5 lobes lanceolate, erect, 4-6 mm long, often equaling or longer than the tube, usually very uneven in length, with one cut half the length of the corolla, each lobe bearing numerous fimbriae inserted in an arc at about midlength of the lobe. Staminal filaments not winged, inserted at the base of the corolla tube, anthers purple. Fruits short stipitate, 9-13 mm long, the apices erect to slightly divergent. Nesom: Taxonomy of Gentianella in México 11 Baja California Norte, endemic to the Sierra San Pedro Martir; 2050-2350 m; September-October. Additional collection examined: MEXICO. Baja California Norte: Sierra San Pedro Martir, llanitos ca. 6 mi from La Encantada on trail to Vallecitos, 22 Sep 1938, Wiggins 9075 (GH, US). Both collections of this species were cited by Gillett (1957) as Gentzanella amarella subsp. acuta, but they differ from the latter in the calyx tubes with smooth margins, anther filaments free to the base of the corolla tube, and shorter corollas with deeply cut, lanceolate corolla lobes of uneven length with fimbriae inserted at about the middle of the lobes. The fimbriae diverge from the corolla at about midpoint of the lobes and are adnate for a short distance below that. The corolla fimbriae of G. mexicana and G. glossocarpa Nesom are similarly positioned though not inserted so far distally. 5. Gentianella glossocarpa Nesom, sp. nov. TYPE: MEXICO. Durango: Mpio. Mezquital, 48 km WNW of Huejuquilla El Alto, Jalisco, on road to Canoas, Durango; crest of ridge, forest of Pinus, Arbutus, and Quer- cus, with steep slopes to north and south, 2530 m, 21 Oct 1983, D.E. Breedlove 59156 with F. Almeda (HOLOTYPE: MO!; Isotype: CAS). Gentianellae fimbrilingui Nesom similis tubis calycum brevibus, lobis calycum ad marginem laevibus, corollis caesiis, et lobis corol- larum fimbriis supra basim insertis, sed foliis ad marginem lae- vibus, corollis longioribus, et fructibus longioribus extensionibus apicalibus linguiformibus 1.2-1.8 mm longis differt. Plants 3-6 dm tall. Cauline leaves spreading-ascending, ovate-lanceolate, not basally connate, 3-5 nerved, 15-30 mm long, gradually reduced upwards, the margins smooth (not papillate). Pedicels 2-7 mm long. Calyx 5.5-7.0 mm long, the lobes smooth margined, 3.5-5.0 mm long, with acute to sharply rounded sinuses, the tube 2.0 mm long. Corollas “lavender,” drying purplish blue, 17-20 mm long, the tube 11-12 mm long, the 5 lobes lanceolate, erect, 7-8 mm long, relatively even in length, each with numerous fimbriae inserted in an arc ca. 1 mm above the base. Staminal filaments not winged, inserted ca. 3 mm above the base of the tube; anthers purplish, not exserted from the corolla tube. Fruits 18-21 mm long, on a basal stipe ca. 2 mm long, the apices somewhat divergent, bearing persistent, narrowly oblong, tonguelike, apical extensions 1.2-1.8 mm long, these usually at least slightly recurved. Known only from the type collection; the MO sheet bears three plants. Gentianella glossocarpa Nesom is similar to G. fimbrilinguis Nesom in its short calyx tubes, calyx lobes with smooth margins, blue corollas, and corolla lobes with fimbriae inserted above the base but different in its smooth leaf margins, longer corollas, and longer fruits with long, tonguelike apical exten- sions. These latter structures (the persistent stigmatic portions of the ovary) 12 PHY P@GLOGIA volume 70(1):1-20 January 1991 are found on Gentianella fruits of all species, but their shape and length on G. glossocarpa are distinct among the Mexican species. Further, the only other Mexican species with such long corollas and fruits are G. hartwegii and G. wrighti (A. Gray) Holub, which are closely related to each other but not especially close to G. glossocarpa. 6. Gentianella hartwegit (Benth.) Holub, Folia Geobot. Phytotax. 2:117. BASIONYM: Gentiana hartwegit Benth., Pl. Hartw. 47. 1840. TYPE: MEXICO. Michoacan: Angangueo, [Aug-Sep 1838], Hartweg 351 (prob- able HOLOTYPE: BM; Isotype: W, MO-photo!). Gentiana mezicana Griseb. subsp. hartwegiz (Benth.) Wettst., Oesterr. Bot. Zeitsch. 50:291. 1900. Amarella hartwegii (Benth.) Arthur, Torreya 12:33. 1912. Gen- tranella amarella(L.) Borner subsp. hartwegii (Benth.) Gillett, Ann. Mis- souri Bot. Gard. 44:260. 1957. Gentiana mezicana subsp. hartwegii f. pringlei Wettst., Oesterr. Bot. Zeitsch. 50:291. 1900. TYPE: MEXICO. México: Moist mead- ows, Nevado de Toluca, 11,000 ft, 6 Sep 1892, C.G. Pringle 4237 (HOLOTYPE: WU; Isotypes: F, GH!, MO!, US!). Gentiana citrina Pollard, Proc. Biol. Soc. Washington 13:130. 1900. TYPE: MEXICO. México: Wet meadows, valley of Toluca, 18 Aug 1892, C.G. Pringle 4196 (HOLOTYPE: US; Isotypes: F, GH!, MEXU, MO!, NY). Leaves ovate-lanceolate, spreading-ascending, 1.5-3.0 cm long, 7-12 mm wide, trinerved, basally rounded but not connate. Pedicels 5-25 mm long. Ca- lyx 10-15 mm long, thin herbaceous, often with noticeable reticulate venation, the tube 3.5-5.0 mm long, the (4-) 5 lobes 5-10 mm long, slightly uneven in width. Corollas yellowish or creamy white, drying yellow, 17-21 mm long, the tube 11-16 mm long, the (4-) 5 lobes lanceolate ovate to lanceolate, 5-7 mm long, the apices acute to rounded, fimbriae relatively few though prominent, inserted at the base of the corolla lobes or slightly above. Staminal filaments narrowly winged near the base, adnate to the lower 3-4 mm of the corolla tube. Fruit sessile, 20-23 mm long, the apices erect to slightly divergent. Michoacan, México, Distrito Federal; wet meadows, 2800-3400 m; July- October (-February). Additional collections examined: MEXICO. Distrito Federal: Near Di- namo de Contreras, 8 Aug 1965, Rzedowski 20397 (TEX). México: Dist. Temascaltepec, Crucero-Agua Blanca, 8 Oct 1935, Hinton 8320 (GH, US); Dist. Temascaltepec, Meson Viejo, 8 Feb 1932, Hinton 1317 (GH, US); Dist. Genes 9 mi E of Villa Victoria by Mex 15, 29 Jul 1965, Kral 25169 US): Gentianella hartwegit and G. wrightii both produce flowers on relatively short pedicels, large and somewhat foliaceous calyx lobes, and large yellow Nesom: Taxonomy of Gentianella in México 13 corollas, and they appear to be very closely related although distantly sepa- rated geographically. The latter differs chiefly in its slightly longer pedicels, smaller corollas with fewer fimbriae, and its smaller (on average) leaves. 7. Gentianella mezicana (Griseb.) Holub, Folia Geobot. Phytotax. 2:117. BASIONYM: Gentiana mezicana Griseb., Gen. Sp. Gentian. 243. 1839. TYPE: MEXICO. “Patria,” no other data, Schiede s.n. (GOET, not seen). Gentianella amarella subsp. mezicana (Griseb.) Gillett, Ann. Mis- souri Bot. Gard. 44:258. 1957. Amarella mezicana (Griseb.) Arthur, Torreya 12:34. 1912. Leaves spreading-ascending, ovate-lanceolate, trinerved, (6-) 10-25 mm long, 5-9 mm wide. Pedicels 3-14 mm long. Calyx 6-11 mm long, the tube 2.0-3.5 mm long, the lobes narrowly lanceolate, 4-8 mm long, often strongly unequal in width, less so in length, herbaceous, the margins minutely papillate- scabrous. Corollas bluish to purple, 10-18 mm long, the tube 6-9 (-14) mm long, the lobes 3.5-5.0 mm long, ovate-lanceolate, erect, fimbriae numerous, inserted in an arc at about midlength of the corolla lobes or slightly below. Staminal filaments very slightly winged, adnate to the basal 2-4 mm of the corolla tube, anthers greenish. Fruits sessile, 8-15 mm long, shorter than the corollas, the apices erect. Veracruz, Hidalgo, Distrito Federal, México, Oaxaca; open meadows, some- times near timberline, areas of pine to fir woods, 2350-3200 m; September- November (-December). Additional collections examined: MEXICO. Without locality, Ehrenberg 79 (GH). Distrito Federal: La Cima, 14 Oct 1908, Barnes & Land 383 (US); Desierto Vieja, Vallee de México, 17 Oct 1865-1866, Bourgeau 1125 (GH); Canada Contreras, Sep 1937, Lyonnet 1978 (US); Desierto de los Leones, Sep 1929, Lyonnet 517(GH, MO, US); S of Contreras, 17 Sep 1930, Russell & Sou- viron 197 (US). Hidalgo: Pachuca, 28 Oct 1900, Holway 5253 (GH); Mpio. Mineral del Chico, below national park, fir forest, 20 Oct 1946, Moore 1607 (GH). Mexico: Toluca, Oct 1827, Berlandier 1086 (MO-photo of G speci- men); Dist. Temascaltepec, Crucero, 21 Oct 1932, Hinton 2117 (GH, US); Dist. Temascaltepec, Crucero-Agua Blanca, 9 Oct 1935, Hinton 8330 (GH, MO, TEX, US); Dist. Temascaltepec, Sierrita, 29 Oct 1935, Hinton 8313(GH, MO, TEX, US); Dist. Temascaltepec, Oro, 28 Sep 1952, Matuda 27261 (MO); Dist. Temascaltepec, road to Nevado de Toluca, 4 Oct 1940, Moore 89a (GH); Dist. Temascaltepec, Sierra de las Cruces, 2 Oct 1892, Pringle 4227(GH, MO, US); Dist. Temascaltepec, Ixtaccihuatl, Oct 1905, Purpus 1757 (GH, MO, US); Dist. Temascaltepec, Cerro San Miguel, Nov 1912, Salazar s.n. (US). Oaxaca: Sierra de Clavellinas, 26 Oct 1894, Smith 667(MO, US). Veracruz: Cofre de Perote, La Simiento, 9 Dec 1930, Balls 5435 (US); Mpio. Las Vigas, Manzanares, 20 Oct 1972, Ventura A. 7222 (LL). 14 PHY T GROG TA volume 70(1):1-20 January 1991 These plants are different from Gentzanella amarella subsp. acuta in the distinctive epilobed insertion of the corolla fimbriae, longer calyx tube, and narrower leaves. ‘The only other species in south-central México, GC. hartwegii, has much more foliaceous calyx lobes without scabrous margins and longer, yellow corollas with fimbriae inserted at the bases of the lobes. 8. Gentianella microcalyz (Lemmon) Gillett, Ann. Missouri Bot. Gard. 44:246. 1957. BASIONYM: Gentiana microcalyz Lemmon, Pacific Rural Press 23:129. [Feb] 1882. TYPE: UNITED STATES. Arizona: [Cochise Co.], peaks of Chiricahua Mts., 30 Sep 1881, J.C. Lemmon 584 (HOLOTYPE: CAS?; Isotypes: GH!, MO, US). Gentiana microcalyr Engelm. ez A. Gray, Proc. Amer. Acad. Arts 17:222. 1882. Homotypic with Gentiana microcalyr Lemmon but published separately and slightly later, based on Lemmon 584 (HOLOTYPE: GH!; Isotypes: MO, US). Leaves ovate, 1.5-3.5 cm long, 6-15 mm wide, not basally connate. Pedicels 10-25 mm long. Calyx 2.5-4.5 mm long, the tube 0.5-1.0 mm long, the lobes thin herbaceous, 2.0-3.5 mm long, usually noticeably unequal, the margins smooth edged. Corollas lavender-pink to bluish, often drying yellowish, some- times slightly speckled, completely lacking fimbriae, (6-) 7-10 mm long, the tube 4.0-6.5 mm long, the 5 lobes 2.0-3.5 mm long, ovate-lanceolate, erect. Staminal filaments slightly winged, adnate to the basal 2 mm of the corolla tube, anthers greenish to purplish, sometimes abortive. Fruits 8-9 mm long, on short (1 mm) stipes, the apices widely divergent. Sonora, Chihuahua, Coahuila, Durango, southern Arizona; commonly on seepy ledges or cliffs, in areas of oak-pine to pine and pine-fir woodlands, 1800-2650 m; August-October. Additional collections examined: MEXICO. Chihuahua: N of Basiguare, a few mi off Creel-Rio Urique road, 19 Oct 1977, Bye & Weber 8334 (GH); Mpio. Temosachi, Nabogame, 3 Dec 1987, Laferriere 1285 (TEX); Temosachic, Madera, Muller 3465 (GH, as cited by Gillett 1957). Coahuila: Sierra de la Madera, middle and upper Canon de la Hacienda, 21 Sep 1972, Chiang, et al. 9451 (LL); Sierra de la Madera, above Canon de la Hacienda, 5 Aug 1973, Henrickson 11945 (TEX); Sierra de la Madera, Corte Blanco fork of Charretera Canyon, 12-14 Sep 1941, Johnston 8990 (GH); Sierra de la Madera, high crest of main ridge ca. 2 km E of Picacho de Zozaya, 13 Sep 1941, Johnston 9030 (GH, LL); Sierra de la Madera, Cafion Desiderio, 29 Sep 1976, Wendt 1831 (TEX). Sonora: 4 mi E of El Bilito, Rio Bavispe region, 12 Oct 1941, White 4771 (US). Durango: Barranca, Sandia Station, Pringle 13660 (GH, as cited by Gillett 1957). Gentianella microcalyz is recognized by its very short calyx tube, consis- tent lack of corolla fimbriae, and fruits with widely divergent apices. It is most similar to and probably most closely related to G. tarahumarae Nesom, Nesom: Taxonomy of Gentianella in México 15 distinguished by features noted in the key. As cited above, two specimens of Gentianella microcalyz (Mueller 3465, Pringle 13660) that were examined by Gillett have not been relocated in the present study. 9. Gentianella sandiana (Gillett) Nesom, comb. et stat. nov. BASIONYM: Gentianella amarella subsp. sandiana Gillett, Ann. Missouri Bot. Gard. 44:259. 1957. TYPE: MEXICO. Durango: Mesa de Sandia, 9000 ft, 20 Oct 1905, C.G. Pringle 10111 (HOLOTYPE: MO!; Isotypes: F, GH!, MEXU, NY, US!). Leaves ovate-lanceolate, basally rounded, trinerved, 15-30 mm long, 6-10 mm wide, spreading-ascending. Pedicels 3-9 mm long. Calyx 5.5-7.0 mm long, the tube 2.5-3.0 mm long, with linear lobes 3.0-4.0 mm long, not fleshy, the margins and veins minutely but prominently papillate-scabrous. Corollas light bluish to lavender, drying yellowish, sometimes with red speckles, 11-16 mm long, the tube 7-10 mm long, the lobes 3.5-6.0 mm long, spreading, ovate lanceolate to obovate, fimbriae numerous, inserted at the very base of the corolla lobes. Staminal filaments very narrowly winged, adnate to the basal 3- 4 mm along the corolla tube, anthers greenish. Fruits 13-18 mm long, slightly longer than the corolla tube, sessile. Northern Durango and southern Chihuahua; 9000-9200 ft; October. Additional collection examined: MEXICO. Chihuahua: 11 mi SW of El Vergel, 9200 ft, 7 Oct 1959, Correll & Gentry 22895 (GH). Gentianella sandiana (Gillett) Nesom is similar to G. canosoi Nesom & Turner in the sizes and relative proportions of its calyces and corollas. The latter differs in its densely papillate-scabrous calyx and pedicels, and its yellow anthers. The tiny red “speckles” on the corolla lobes, noted by Gillett (1957) as distinctive of this species, are found on 5 of the 7 plants mounted on the three type sheets examined. They also occur on the corollas of some, but not all, plants of both Gentianella tarahumarae and G. microcalyz but not on corollas of G. canosoz. 10. Gentianella tarahumarae Nesom, sp. nov. TYPE: MEXICO. Chi- huahua: Mpio. Guachochic, between Cusarare and Bahichic, open slopes of mixed pine and oak forest, 6900 ft, 10 Oct 1974, R. Bye 7045 (HOLO- TYPE: TEX). Differt a Gentianella amarella(L.) Borner subsp. acuta (Michx.) Gillett foliis effusis crassis anguste lanceolatis minoribusque, ca- lycibus brevioribus lobis incrassatis linearibusque, corollis brev- ioribus, et ovariis stipitatis. 16 PHYTOLO GIA volume 70(1):1-20 January 1991 Leaves darkly pigmented, linear lanceolate to lanceolate, sometimes no- ticeably trinerved, not at all basally rounded, 1-3 cm long, 1.5-5.0 mm wide, strongly spreading to slightly deflexed. Pedicels 3-9 mm long. Calyx 2.2-5.0 mm long, 1/6-2/5 as long as the corolla, the tube 0.5-1.0 mm long, with fleshy, linear lobes 1.8-4.0 mm long, with sinuses acute to sharply rounded, the mar- gins minutely papillate-scabrous. Corollas purple to blue, drying whitish to blue or rose, 8.5-11. mm long, the tube 6-8 mm long, the 5 lobes ovate to ovate elliptic, spreading, 2.5-3.0 mm long, sometimes speckled, with numerous fimbriae inserted at the base of the lobes. Staminal filaments inserted the middle of the corolla tube, not winged, the anthers greenish, included within the corolla tube. Fruits 10-12 mm long, with erect to somewhat divergent apices, slightly exserted from the corolla tube, on a short (0.5-0.9 mm long) but distinct stipe. Southwestern to south-central Chihuahua; 2000-2200 m; September-Novem- ber. Additional collections examined: MEXICO. Chihuahua: Mpio. Guachochic, Cusarare, along arroyo just NW of Cusarare church, 2200 m, 14 Oct 1977, Bye & Weber 8095 (GH); Sierra Madre, 65 mi E of Batopilas, 7000 ft, 1-2 Oct 1898, Goldman 188 (US); SW Chihuahua, Aug-Nov 1885, Palmer 334 (GH-mounted on same sheet as isotype of G. wislizeni, US). The collections by Goldman and Palmer were identified by Gillett as Gen- tianella sandiana, but that taxon differs from G. tarahumarae in its ovate- lanceolate, ascending leaves, calyces with a much longer tube and broader lobes, longer corollas, and sessile ovaries. Gentianella tarahumarae is distinct from G. amarella subsp. acuta in its thick, short, narrowly lanceolate, spread- ing leaves, shorter calyces with thickened, linear lobes, shorter corollas, and stipitate ovaries. As noted by Bye on the label of the type collection of G. tarahumarae, this is “one of the few plants still in flower after frost; plants mashed in water and used to wash head and clothes.” 11. Gentianella wishzent (Engelm.) Gillett, Ann. Missouri Bot. Gard. 44:235. 1957. BASIONYM: Gentiana wishizeni Engelm., Trans. Acad. Sci. St. Louis 2:215. 1862. TYPE: MEXICO. Chihuahua: Llanos, mountains west of Chihuahua, 5 Oct 1846, Wislizenus 206 (HOLOTYPE: MO!; Isotype: GH!). Leaves ovate-lanceolate with rounded bases, not basally connate, trinerved, 2-4 cm long, 4-15 mm wide, spreading at right angles to somewhat ascending. Pedicels 4-18 mm long. Calyx tube thin textured, often hyaline, commonly purplish, 3.5-4.0 mm long, nearly flat across the top, with 5, greenish, linear toothlike lobes 0.5-1.5 mm long, the whole tube split to the base along one side to form a membranaceous sheath. Corollas whitish pink to violet purple, commonly drying distinctly yellowish, 8-14 mm long, the 5 lobes lanceolate Nesom: Taxonomy of Gentianella in México 17 ovate to broadly elliptic, spreading to erect, 3-4 mm long, of even length, fimbriae relatively dense to very sparse, inserted at the base of the lobes, sometimes completely lacking. Staminal filaments narrowly winged, adnate to the basal half of the corolla tube, the anthers wholly included to slightly exserted from the corolla tube. Fruit 10-12 mm long, sessile, with erect apices. Chihuahua, Durango, Arizona; rocky sites in pine-oak and pine woodlands, 2000-2550 m; September-October. Additional collections examined: MEXICO. Chihuahua: Mex. NW R{ail] R[oad], continental divide, ridge between Rio Chico and Rio Caballo, 30 Sep 1911, Barlow s.n. (US); Santo Domingo on Matachic-Ocampo truck road be- tween Concheno and Pinos Altos, 21 Oct 1945, Hewitt 78 (GH); Mesa, W of Hop Valley, 17 Sep 1903, Jones s.n. (US); 5.3 mi W of Madera on road to Rio Papigochic, 22 Sep 1984, Lavin 4947 (TEX); Colonia Garcia, 23 Sep 1934, Pennell 19097 (GH, US); Colonia Garcia, near First Meadow, 23 Sep 1934, Pennell 19136 (GH, US); Sierra Madre, cool slopes, Pringle 1328 (GH, US); Sierra Madre, cool slopes, 9 Oct 1888, Pringle 1662 (MO); near Colonia Garcia, 11 Sep 1899, Townsend & Barber 322 (GH, MO, US); 8 mi S of Cd. Guerrero, ca. 2 mi W of Rio Colorado, 23 Sep 1981, Warnock 2358 (TEX). Durango: Road between San Julian and Cerro Prieto, 9 Sep 1898, Nelson 4950 (US). The peculiar calyx of this species is unmistakable. The closest relative of Gentianella wishzeni probably is G. calycidon, as noted in the discussion following the latter. 12. Gentianella wrightu (A. Gray) Holub, Folia Geobot. Phytotax. 2:118. BASIONYM: Gentzana wrighttt A. Gray, Syn. Fl. N. Amer. 2(2):118. 1886. TYPE: MEXICO. Sonora: Valley near Santa Cruz, springy ground, 24 Sep 1851, C. Wright 1659 (HOLOTYPE: GH!; Isotype: GH!). Amarella wrightii (A. Gray) E. Greene, Leafl. Bot. Observ. Crit. 1:53. 1904. Gentianella amarella (L.) Borner subsp. wrightii (A. Gray) Gillett, Ann. Missouri Bot. Gard. 44:259. 1957. Amarella cobrensis E. Greene, Leafl. Bot. Observ. Crit. 1:56. 1904. LECTOTYPE (designated here): UNITED STATES. New Mexico: Santa Rita del Cobre, 11 Oct 1880, E.L. Greene s.n. (ND-G!). Apparently referring to the type, Gillett (1957) cited a col- lection by Greene (s.n.) from the “Pinos Altos Mountains” in Grant Co., New Mexico, but the specimens referred to in the original publication apparently are from a different gathering. The specimen chosen as lectotype is a sheet from Greene’s herbarium, filed by Greene in his Gentiana cobrensis folder. The specimen has label data exactly matching those in the 18 PHY TOLO GIA volume 70(1):1-20 January 1991 publication, although it has no original identification or anno- tation other than “Gentiana Amarella L. var.” Gentiana townsendu Briq., Candollea 4:329. 1931. TYPE: MEXICO. Chihuahua: Near Colonia Garcia, 1 Oct 1899, C.H.T. Townsend & C.M. Barber 358 (HOLOTYPE: G; Isotypes: F, GH!, MO!, NY, US!). Leaves ovate-lanceolate, spreading to ascending, 2-5 cm long, 8-17 mm wide, trinerved, basally rounded but not connate. Pedicels 3-5 (-15) mm long. Calyx 8-13 mm long, thin herbaceous, often with noticeable reticulate venation, the tube 3.5-5.0 mm long, the 5 lobes 6-8 mm long, slightly uneven in width. Corollas yellowish or whitish, rarely with blue streaks, drying yellow, 20-25 mm long, the tube 13-16 mm long, the lobes lanceolate, 7-10 mm long, with markedly attenuate apices, fimbriae numerous, inserted at the base of the corolla lobes or slightly above. Staminal filaments narrowly winged near the base, adnate to the lower 3-5 mm of the corolla tube. Fruit sessile, 22-26 mm long, the apices erect to slightly divergent. Sonora, Chihuahua, Sinaloa, Arizona, New Mexico; low, wet meadows or swales, area of pine-juniper to pine woodlands, 2100-2400 m; August- November. Additional collections examined: MEXICO. Chihuahua: SW of Creel, Mpio. Bocoyna, 17 Oct 1977, Bye & Weber 8262 (GH); 1-2 mi S of Creel, 25 Sep 1972, Henrickson 8031 (TEX); Chuichupa, 26 Sep 1903, Jones s.n. (MO, US); SW Chihuahua, Yerba Buena, Nov 1885, Palmer 306 (GH, MO, US); Colonia Garcia, 23 Sep 1934, Pennell 19118 (US). Sinaloa: Cerro del Viejo, San Ignacio, 19 Nov 1917, Montes & Salazar 83 (US). See comments following Gentzanella hartwegii, which is closely related. ACKNOWLEDGMENTS I thank Dr. B.L. Turner, Dr. D.R. Windler, and Dr. A. McDonald for their review and comments, Dr. Barney Lipscomb for help with obtaining pertinent literature, Dr. Barbara Hellenthal (Curator of ND-G) for help in interpretation of the specimen of Gentianella cobrensis, and GH, MO, ND-G, and US for loans of specimens. LITERATURE CITED Clapham, A.R., T.G. Tutin, & D.M. Moore. 1987. Flora of the British Isles (ed. 3). Cambridge Univ. Press, Cambridge, England. Nesom: Taxonomy of Gentianella in México 19 Fabris, J.A. 1960. El género Gentianella en Ecuador. Bol. Soc. Argent. Bot. 8:160-192. Fernald, M.L. 1917. Some forms of American gentians. Rhodora 19:149-152. _____. 1950. Gray’s Manual of Botany (ed. 8). American Book Co., New York. Gillett, J.M. 1957. A revision of the North American species of Gentzanella Moench. Ann. Missouri Bot. Gard. 44:195-269. Gleason, H.A. & A. Cronquist. 1963. Manual of Vascular Plants of North- eastern United States and Adjacent Canada. Van Nostrand Co., Prince- ton, New Jersey. Grossgeim, A.A. 1952. Gentiana. Fl. U.R.S.S. 18:538-620. Holmgren, N.H. 1984. Gentianella. Intermountain Fl. 4:12-15. Holub, J. 1967. Neue Namen innerhalb der Gattungen Gentzanella Moench, Gentianopsis Ma und Comastoma (Wettst.) Tokokuni. Folia Geobot. Phytotax. 2:115-120. Hultén, E. 1968. Flora of Alaska and Neighboring Territories. Stanford Univ. Press, Stanford, California. Iltis, HH. 1965. The genus Gentzanopsis (Gentianaceae): transfers and phytogeographic comments. Sida 2:129-154. Kartesz, J.T. & R. Kartesz. 1980. A Synonymized Checklist of the Vascu- lar Flora of the United States, Canada, and Greenland. Univ. North Carolina Press, Chapel Hill. Ma, Yu-Chuan. 1951. Gentianopsis: a new genus of Chinese Gentianaceae. Acta Phytotax. Sinica 1:5-19. Marroquin, J.S. & J. Rzedowski. 1984. Gentianella. Pp. 220-221. In Rze- dowski, J. & G.C. Rzedowski. Flora del Valle de Mézico. Vol. II, Di- cotyledoneae. Instituto de Ecologia, México, D.F. Nesom, G.L. & B.L. Turner. 1990. Gentianella canosoi (Gentianaceae), a new species from Durango, Mexico. Sida 14:227-229. Pringle, J.S. 1981. Nomenclatural transfers and taxonomic notes on some South American Gentianaceae. Phytologia 48:281-285. 20 PHY FOLOGIA volume 70(1):1-20 January 1991 Tokokuni, H. 1961. Separation de Comastoma, genre nouveau, d’avec Gen- tianella. Bot. Mag. Tokyo 74:198. ______. 1963. Conspectus gentianacearum Japonicarum, a general view of the Gentianaceae indigenous to Japan. J. Faculty Sci. Hokkaido Univ., Ser. 5 (Bot.) 7:137-259. Phytologia (January 1991) 70(1):21-23. NAMA QUIEXOBRANUM (HYDROPHYLLACEAE): A NEW SPECIES FROM OAXACA, MEXICO John D. Bacon & J. Andrew McDonald Department of Biology, The University of Texas at Arlington, Arlington, Texas 76019 U.S.A. & Department of Botany, The University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT Nama quiexobranum sp. nov. is known only from Cerro Quiexo- bra in the Sierra Madre del Sur, Oaxaca. In several aspects, N. qutezo- branum approaches N. sericeum Willd. ez Roem. & Schult. and N. ort- ganifolium H.B.K., but its seeds are most similar to those of the primi- tive species, N. hirsutum Mart. & Gal. and N. prostratum Brand. Nama qutezobranum appears to link the derived N. sericeum, N. origanifolium, and several annual species to these basal taxa. KEY WORDS: Nama, Hydrophyllaceae, México Nama quiexobranum Bacon & McDonald, sp. nov. N. sericeo Willd. er Roem. & Schult. caulibus fragilibus cymis laxis terminalibus ac lateralibus, corollis infundibuliformis ac ob- conicis usque ad 22 mm longis tangit, sed foliis angustioribus usque ad 7 mm latis sine trichomatibus densis sericeis in pagina inferiore, seminibus corrugatis costis minutis longitudinalibus 0.8-1.1 mm longis differt. TYPE: MEXICO. Oaxaca: 35 km ESE of Miahuatlan, 5 km NE of Santo Domingo Ozolotepec, Cerro Quiexobra, 16° 10’ N Lat., 96° 15’ W Long., 3650-3800 m. Timberline vegetation along ridges and in mountain “saddles,” dominated below by pine forest, occa- sional on upper margins of rock outcrops or less commonly in damp, shaded ravines, soft wooded perennials, flowers dark blue-purple, 4 Oct 1990, McDonald 3014 (HOLOTYPE: TEX!; Isotypes: GH!, MEXU!, NY!, US!). 21 22 PHYTOLOGIA volume 70(1):21-23 January 1991 Erect, mealy-glandular perennials to 40 cm tall, the brittle stems branch- ing above their base, moderately to densely puberulous tomentose, or bearing scattered, hirsute-hispid trichomes to 1 mm long. Leaves 10-28 mm long, 1.5- 7.0 mm broad, oblong, elliptic, oblanceolate, or obovate, plane, or the younger weakly to strongly revolute, acute to rounded at apex, attenuate and sessile or with a distinct petiole 1.0-3.5 mm long, invested with moderate to dense, appressed, short, strigillose trichomes. Flowers in loose, terminal or lateral, racemoid cymes; peduncles and pedicels slender, (1.7-) 4.5-18.0 (-34.0) mm long; sepals 6.0-11.5 mm long, broadly linear to spatulate, acute to rounded apically, and moderately hispid-hirsute; corollas 18-22 mm long, broadly tubular-obconic, dark blue-purple; filaments 11.0-12.8 mm long, the free portions distally terete but soon becoming somewhat flattened and ex- panded some distance above their insertion 5.0-6.3 mm above corolla base, about twice as long as the prominently winged adnate portions; mature styles 6.4-8.2 mm long. Capsules 6.8-7.4 mm long, 2.8-4.4 mm broad, ovoid to ellip- soid; seeds 0.8-1.1 mm long, multifaceted, rarely somewhat ovoid, brown, with weak transverse corrugations and longitudinal ridges. Chromosome number, n = 7 pairs. Nama quierobranum Bacon & McDonald is an interesting taxon in that it combines features found in a number of species of Nama. Its obconic-broadly tubular corollas, often exceeding 20 mm in length, and loose, terminal inflo- rescences are similar to those of N. sericeum Willd. ex Roem. & Schult.; it also has the unusual brittle stems found in both N. sericeum and N. origani- foltum H.B.K. The latter two species are Mexican perennials distributed to the north of the state of Oaxaca. All three species are readily differentiated, how- ever, by leaf shape, size, and pubescence or flower size. Nama origanifolium differs from N. quiezobranum by corollas 4.5-7.5 mm long, its usually velvety- villous indument, and smaller, black seeds 0.3-0.5 mm long. Nama sericeum is distinguished by broader leaves (7.0-22.0 mm wide), densely sericeous leaf undersurfaces, and weakly reticulate, smaller seeds (0.6-0.8 mm long). Seeds of Nama quierobranum and N. hirsutum Mart. & Gal. (a perennial found in southern México and northern Guatemala), are strikingly similar in their size (0.8-1.1 mm long), shape, and texture, exhibiting weak transverse corrugations and longitudinal ridges. In addition, wall thickenings in cells of the outermost testa of both species are essentially alike; columnar, often basally branched, generally extending from the radial wall into and across the lower transverse wall that forms the floor of the cell. In wall thickening pat- terns, seeds of N. quierobranum and N. hirsutum are much like those of N. pro- stratum Brand, another perennial distributed in central and southern México. Based on seed features, both N. hirsutum and N. prostratum were suggested by Chance & Bacon (1984) to be primitive elements in Nama. Thus, similarities of seed features among these taxa suggest that N. quierobranum represents yet another basal member in the genus. At the same time, morphological Bacon & McDonald: New species of Nama from México 23 similarities among N. quierobranum, N. sericeum, and N. origanifolium argue that the latter two species are to be positioned near the former. As Chance & Bacon (1984) allied N. sericeum and N. origanifolium with several annual species found in México and southwestern United States (see Bacon 1987), N. quierobranum appears to link these derived species with basal elements in the genus. ACKNOWLEDGMENTS We thank Dr. Guy Nesom and Dr. B.L. Turner for critically reviewing the manuscript. LITERATURE CITED Bacon, J.D. 1987. Systematics of Nama (Hydrophyllaceae) seed coat mor- phology of Lemmonia californica and Nama species allied with Nama demissum. Aliso 11:441-450. Chance, G.D. & J.D. Bacon. 1984. Systematic implications of seed coat morphology in Nama (Hydrophyllaceae). Amer. J. Bot. 71:829-842. Phytologia (January 1991) 70(1):24-27. A CLASSIFICATION OF THE RANUNCULACEAE WITH SPECIAL REFERENCE TO THE WESTERN HEMISPHERE Thomas Duncan! & Carl S. Keener? ' University Herbarium, University of California at Berkeley, Berkeley, California 94720 U.S.A. ? Department of Biology, 208 Mueller Laboratory, The Pennsylvania State University, University Park, Pennsylvania 16802 U.S.A. ABSTRACT A classification of the Ranunculaceae of the Western Hemisphere is presented as consisting of three subfamilies, 16 tribes, two subtribes, 30 genera, and an estimated 437 species. Two tribes (Eranthideae, Xanthorhizeae) are described as new, and six taxa are given a new tribal status (Adonideae, Aquilegieae, Coptideae, Thalictreae, Trautvetterieae, Trollieae). KEY WORDS: Ranunculaceae, classification, nomenclature, tax- onomy, Western Hemisphere The following classification of Ranunculaceae portrays a subfamilial, tribal, and subtribal classification of the 30 genera of Ranunculaceae native or nat- uralized within the Western Hemisphere. Other genera will be noted and in- cluded in a subsequent and more detailed paper. In the classification proposed here, Ranunculaceae consists of three subfamilies, which are classified further into tribes, and in one case, subtribes. Data and cladistic analyses that sup- port the classification system adopted here will be published elsewhere. Every attempt has been made to secure the oldest correct name for each taxon rec- ognized. The authors will be grateful if any errors of omission or citation are called to their attention. The numbers of species, in some cases estimated, are given within parentheses following the reference to the protologue of each genus. We estimate there are 437 species of Ranunculaceae within the Western Hemisphere. 24 Duncan & Keener: Classification of Western Hemisphere Ranunculaceae 25 RANUNCULACEAE AL. de Jussieu, Gen. pl. 321. 1789. Crowfoot Family. Subfam. I. Hydrastidoideae Engler & Gilg, Syllabus, ed. 7. 188. 1912. 1. Hydrastis L., Syst. nat., ed. 10, 2:1069, 1088, 1374. 1759. (1). Golden seal, Orangeroot. Subfam. II. Ranunculoideae Tribe 1. Ranunculeae 2. Ranunculus L., Sp. pl. 1:548. 1753. Gen. pl., ed. 5. 243. 1754. (141). Buttercup, Crowfoot. 3. Krapfia D.C., Syst. nat. 1:228. 1818. (8). 4. Myosurus L., Sp. pl. 1:284. 1753. Gen. pl., ed. 5. 137. 1754. (5). Mouse-tail. 5. Hamadryas Commerson ez de Jussieu, Gen. pl. 232. 1789. (5). 6. Laccopetalum Ulbrich, Bot. Jahrb. Syst. 37:404-408. 1906. (1). Tribe 2. Trautvetterieae (Tamura) Duncan & Keener, stat. nov. BA- SIONYM: ‘Trautvetteriinae Tamura, Sci. Rep. Osaka Univ. 16:42. 1967. 7. Trautvetteria Fischer & Meyer, Index sem. hort. petrop. 1:22. 1835. (1). False-bugbane. 8. Kumlienia E. Greene, Bull. California Acad. Sci. 1:337. 1886. (2). Tribe 3. Anemoneae DC., Syst. nat. 1:129. 1818. Subtribe 1. Anemoninae Spach, Hist. nat. vég. 7:190. 1839. 9. Anemone L., Sp. pl. 1:538. 1753. Gen. pl., ed. 5. 241. 1754. (36). Anemone. 10. Hepatica P. Miller, Gard. Dict., abr. ed. 4. 1754. (1). Hepatica, Liverleaf. 11. Pulsatilla P. Miller, Gard. Dict., abr. ed. 4. 1754. (2). Pasque- flower, Prairie-smoke. 12. Barneoudia Gay, Fil. chil. 1:19. 1845. (3). 13. Oreithales Schlechtendal, Linnaea 27:559-560. 1954. (1). Subtribe 2. Clematidinae Lotsy, Vortr. bot. Stammesgesch. 3:584. 1911. 14. Clematis L., Sp. pl. 1:543. 1753. Gen. pl., ed. 5. 242. 1754. (47). Clematis. Tribe 4. Helleboreae DC., Syst. nat. 1:130, 306. 1818. 15. Helleborus L., Sp. pl. 1:557. 1753. Gen. pl., ed. 5. 244. 1754. (2). Hellebore. Tribe 5. Cimicifugeae Torrey & A. Gray, Fl. N. Amer. 1:34. 1838. 16. Cimicifuga Wernischeck, Gen. pl. 298, 321. 1763. (6). Bugbane. 26 PHYTOLOGIA volume 70(1):24-27 January 1991 17. Actaea L., Sp. pl. 1:504. 1753. Gen. pl., ed. 5. 222. 1754. (3). Baneberry. Tribe 6. Eranthideae Duncan & Keener, tribus novum. Diagnosis: Herbae pumilae rhizomatibus tuberosis; fo- lia palmata; folia rosulae petiolata; folia caulis sessilia ver- ticillata; flores solitarii sessiles; sepala lutea grandia per- sistentia; petala peltata bilabiata nectarifera; stamina nu- merosa (18-44); folliculi stipitati. Typus genus: Eranthis Salisbury. Dwarf herbs with tuberous rhizomes; leaves palmately divided; basal leaves petiolate; cauline leaves sessile, whorled; flowers solitary, sessile; sepals yellow, large, persistent; petals peltate, 2 lipped, nectariferous; stamens numerous (18-44); follicles stipitate. One genus, Franthis Salisbury, native to the Mediterranean region. 18. Eranthis Salisbury, Trans. Linn. Soc. 8:303. 1807. (1). Winter- aconite. Tribe 7. Nigelleae Schrodinger, Abh. K. K. Zool.-Bot. Ges. Wien 4, Heft 5, p. 58. 1909. 19. Nigella L., Sp. pl. 1:534. 1753. Gen. pl., ed. 5. 238. 1754. (1). Love-in-a-mist. Tribe 8. Adonideae (Spach) Duncan & Keener, stat. nov. BASIONYM: Adonidinae Spach, Hist. nat. vég. 7:222. 1839. 20. Adonis L., Sp. pl. 1:547. 1753. Gen. pl., ed. 5. 242. 1754. (3). Pheasants-eye. Tribe 9. Caltheae J.S. Presl in K.B. Presl, Fl. sicul. 1:20. 1826. 21. Caltha L., Sp. pl. 1:558. 1753. Gen. pl., ed. 5. 244. 1754. (7). Marsh-marigold. Tribe 10. Trollieae (Heintze) Duncan & Keener, stat. nov. BASIONYM: Trolliinae Heintze, Cormofyternas fylogeni 103. 1927. 22. Trollius L., Sp. pl. 1:556. 1753. Gen. pl., ed. 5. 243. 1754. (2). Globe-flower. Tribe 11. Delphinieae Schrodinger, Abh. K. K. Zool.-Bot. Ges. Wien 4, Heft 5, p. 58. 1909. 23. Aconitum L., Sp. pl. 1:432. 1753. Gen. pl., ed. 5. 236. 1754. (13). Monkshood, Wolfsbane. 24. Delphinium L., Sp. pl. 1:530. 1753. Gen. pl., ed. 5. 236. 1754. (73). Delphinium, Larkspur. 25. Consolida(DC., Syst. nat. 1:341. 1818) S.F. Gray, Nat. arr. Brit. pl. 2:711. 1821. (3). Garden Larkspur. Duncan & Keener: Classification of Western Hemisphere Ranunculaceae 27 Subfam. III. Isopyroideae Schrodinger, Abh. K. K. Zool.-Bot. Ges. Wien 4, Heft 5, p. 59. 1909. Tribe 12. Coptideae (Tamura) Duncan & Keener, stat. nov. BA- SIONYM: Coptidoideae Tamura, Sci. Rep. Osaka Univ. 17:52. 1968. 26. Coptis Salisbury, Trans. Linn. Soc. 8:305. 1807. (4). Goldthread. Tribe 13. Xanthorhizeae Duncan & Keener, tribus novum. Diagnosis: Fruticuli ligno noninduratis; radices lutae; folia pinnata nitida; foliola acute fissa vel incisa vel serrata; inflorescentiae paniculatae cernuae; flores parvi; sepala 5; petala 5; integumenta 2 integumento interiore quam in- tegumento exteriore longiore; folliculi monospermi (ovulo unico obortivo). Typus genus: Xanthorhiza Marshall. Shrublets with wood not becoming hard; roots yel- low; leaves pinnate, shiny; leaflets sharply cleft, incised, or serrate; inflorescences paniculate, drooping; flowers small; sepals 5; petals 5; integuments 2, the inner integument longer than the outer; follicles one seeded (one ovule aborts). One genus, Xanthorhiza Marshall, native to southeastern United States. 27. Xanthorhiza Marshall, Arbust. amer. 167. 1785. (1). Yellow Root. Tribe 14. Isopyreae Schrodinger, Abh. K. K. Zool.-Bot. Ges. Wien 4, Heft 5, p. 59. 1909. 28. Enemion Rafinesque, Jour. Phys. 91:70. 1820. (5). False Rue Anemone. Tribe 15. Aquilegieae (Tamura) Duncan & Keener, stat. nov. BA- SIONYM: Aquilegiinae Tamura, Sci. Rep. Osaka Univ. 17:48. 1968. 29. Aquilegia L., Sp. pl. 1:533. 1753. Gen. pl., ed. 5. 237. 1754. (24). Columbine. Tribe 16. Thalictreae (Heintze) Duncan & Keener, stat. nov. BA- SIONYM: Thalictroideae Heintze, Cormofyternas fylogenz 103. 1927. 30. Thalictrum L., Sp. pl. 1:545. 1753. Gen. pl., ed. 5. 242. 1754. (35). Rue. Phytologia (January 1991) 70(1):28-41. CEANOTHUS OPHIOCHILUS (RHAMNACEAE): A DISTINCTIVE, NARROWLY ENDEMIC SPECIES FROM RIVERSIDE COUNTY, CALIFORNIA Steve Boyd, Timothy Ross, & Laurel Arnseth Rancho Santa Ana Botanic Garden, Claremont, California 91711, U.S.A. ABSTRACT A newly discovered species, Ceanothus ophiochilus, is described from Riverside County, California. Unusually small, narrow, semi-terete leaves distinguish this species from other members of section Cerastes. Its blue to pinkish lavender flowers, rather than white, further separate the taxon from other members of the section in southern California. The solitary population is restricted to an unusual pyroxenite rich outcrop on privately held land being considered for development. KEY WORDS: Ceanothus, Cerastes, Rhamnaceae, edaphic en- demic, pyroxenite endemic, endangered species. INTRODUCTION The species of Ceanothus section Cerastes are characterized by persistent, coriaceous leaves with stomata in sunken pits; thick, darkly colored, corky stipules; and flowers in axillary umbels. All but two species have opposite leaves and most have capsules bearing three horns (McMinn 1942; Munz 1974). Raven (1977) reports that of the 21 species in the section, 20 occur in Cal- ifornia, with 17 endemic to the California Floristic Province. The section is especially well represented in the central and northern part of the state, where several of the taxa are edaphic endemics. At least two species are confined to serpentine soils (Nobs 1963; Kruckeberg 1984); one is limited to Franciscan marine sandstones of Jurassic age; and five are restricted to Pliocene Sonoma volcanics (Nobs 1963). Depending on which treatment is followed, seven to eight taxa, representing five or six species and four varieties, have tradition- ally been recognized for Southern California (McMinn 1942; Munz 1959, 1974). In this paper, we describe a narrowly endemic species of Ceanothus, section 28 Boyd, et al.: Ceanothus ophiochilus new species in California 29 Cerastes, which we encountered in March, 1989, while conducting floristic sur- veys of a large, privately held parcel surrounding Vail Lake in southwestern Riverside County, California. This plant is an edaphic endemic and is known only from the type locality where it is restricted to an unusual, pyroxenite rich outcrop. The continued existence in the wild of this attractive species is threatened by potential urbanization of the site. TAXONOMY Ceanothus ophiochilus Boyd, Ross, & Arnseth, sp. nov. Figure 1 Frutez rotundatus, ramosissimus, (3.0-) 12-15 (-20.) dm altus. Caules basilares 1-aliquot, trunco principale ad 7.2 cm diametro basi, interdum aspectu funis textilem. Cortez laevis comparate, cinereus ad schistaceum vel castaneum in veteribus truncis decor- ticantibus. Ramzi divaricati, cinerei, internodiis (2.0-) 2.5-7.0 (-13) mm longis; ramis hornotinis ferrugineis, pilos parcos breves geren- tibus, mox glabrescentibus; stipulis hornotinis ferrugineis, sed ma- turis ferro-griseis, suberosis. Folia opposita, glabra, coriacea, fas- ciculata in ramis maturis, flavovirentia ad viridia, ambitu anguste- oblanceolata ad obovata, (2.0-) 3.5-7.0 mm longa, (1.0-) 1.5-2.5 (-3.0) mm lata, supra concava subtus gibbosa, 0.8-1.0 mm crassa; apicibus denticulatibus vel rotundatibus vel emarginatibus; margin- ibus integris vel 1-2 (-3) paribus denticulatibus; costis distinguibili- bus sed nervatura secundaria obscura generaliter; petiolis 0.55-1.0 mm longis, 0.4 mm latis. Flores 6-8 in umbellis lateralibus, sub- lazulini vel subrosei vel lactei. Bracteae inflorescentiae 2, rotun- datae, ciliatae, 1.7-2.0 mm diametro, evanescentes. Pedunculus 1-3 mm longus, pedicellis 2.2-5.0 mm longis. Calyz rotatus, glabrus, 3- 5 mm diametro, sepalis ovato-deltoideis, 1.0-1.5 mm longis. Discus glandularis violaceus, 1.5 mm diametro. Stylus 1.5-1.8 mm longus, stigmate trilobato, lobis circa 0.1 mm longis. Filamenta staminum 1.4-1.8 mm longa, antheris 0.8 mm longis. Petala unguiculata, 2 mm longa, leniter deflexa, laminis cyathiformis 1 mm longis. Capsula globosa, circa 3.8 mm diametro, rubella ubi immatura, brunnea hebetata ubi matura, cornibus lateralibus absentibus vel raro vestigialibus, cristis intermediis absentibus. Semina 3, poli- tae, brunneae ad fusco-nigras, 2.2 mm longae. Florescentia in circa medio Februario ad Martio, fructibus maturescentibus post 2.5-3.0 menses. TYPE: U.S.A. California: Riverside Co., Vail Lake area, 1 mile W of lake, T8S, R1W SEt SEt section 8, swt swt section 30 BBY ST OVE O. GLA volume 70(1):28-41 January 1991 9; 2000-2099 feet. Restricted to series of N-facing slopes. 19 March 1989, Steve Boyd, Tim Ross, Laurel Arnseth 3020 (HOLOTYPE: RSA; Isotypes: CAS, MO, SD, SJSU, UC, UCR, US, and 14 others to be distributed). Paratypes. - U.S.A. California: Riverside Co., same loca- tion as above, 12 March 1989 (fl.), Steve Boyd, Tim Ross, Laurel Arnseth 2960 (CAS, MO, SJSU, RSA, UCR); same location as above, 31 March 1989 (fr.), Steve Boyd, Tim Ross, Laurel Arnseth 3097 (CAS, MO, RSA, SD, SJSU, UC, UCR, US, and 14 others to be distributed). Rounded, divaricately branched shrub, (3.0-) 12-15 (-20.) dm tall. Basal stems one to several, with the main trunk to 7.2 cm in diameter and eventually developing a braided appearance. Bark relatively smooth, ash gray to slate gray or occasionally ranging to reddish brown on exfoliating trunks. Young twigs reddish brown maturing to ashy gray with internodes (2.0-) 2.5-7.0 (-13) mm in length. Stipules on new growth are reddish brown, ultimately becom- ing iron gray with a corky texture. Leaves glabrous, coriaceous, yellow green to medium green, opposite on stems of recent growth, becoming fascicled on short axillary spur branches in older wood; narrowly oblanceolate to obovate, (2.0-) 3.5-7.0 mm long, (1.0-) 1.5-2.5 (-3.0) mm wide, 0.8-1.0 mm thick; the adaxial surface shallowly concave folded; abaxial surface strongly convex, + gibbous; the margins entire, or occasionally with 1-2 (-3) pairs of minute teeth; the apex bearing a tooth, or rounded, or emarginate. Midrib visible on abaxial surface, but lateral veins generally obscured. Petioles 0.55-1.0 mm long, 0.4 mm wide, sparsely appressed pubescent in youngest leaves but soon glabrous. Inflorescences axillary on peduncles 1-3 mm long, mostly 6-8 flowered. Floral bud scales 2, pinkish, broadly ovate to orbicular, 1.7-2.0 mm long, sparsely puberulent abaxially towards the apex, with ciliate margins; evanescent. Pe- duncles 1-3 mm long. Pedicels 2.2-5.0 mm long, usually deeply pigmented at anthesis. Calyx at anthesis 3-5 mm broad, glabrous, the lobes ovate-deltoid, 1.0-1.5 mm long. Flowers pale blue or rarely pinkish lavender, fading to white after anthesis due to loss of pigmentation in pedicel and glandular disk. Petals ladle shaped, 2 mm long, equally divided between the filiform claw and the deeply saccate blade. Style 1.5-1.8 mm long, trilobate, each lobe circa 0.1 mm long. Staminal filaments 1.4-1.8 mm long; anthers 0.8 mm long. Capsule glo- bose, 3.3 mm in diameter, hornless or rarely with 3 rudimentary lateral horns; shiny and + reddish when young, becoming dull light brown when mature. Seeds medium to dark brown with polished surfaces, convex abaxially, angled on inner face; 2.2 mm long. Flowering occurs in approximately mid-February to March, with maturation of the capsules in about late May to mid-June. Close observation of the rounded leaf margins reveals a distinctive pat- tern resembling the labial scales about the mouths of some snake species. Boyd, et al.: Ceanothus ophiochilus new species in California 31 This characteristic patterning served as the inspiration for the specific epi- thet, ophiochilus (Greek, ophis (snake, serpent]; chezlos [lip]). Because the sole population overlooks Vail Lake, we would like to suggest the name “Vail Lake Ceanothus” as the vernacular for this species. DISCUSSION Ceanothus ophiochilus is known only from the type locality where it grows on a series of contiguous north-facing slopes and ridge tops between 1980’ and 2090’ on the eastern slopes of Oak Mountain', 1 mile west of Vail Lake. The entire population is restricted to about 20 acres of a pyroxenite rich outcrop which totals about 40 acres in area. It is surprising that a plant as distinc- tive as C. ophiochilus could remain undetected in a region as relatively well explored botanically as Southern California. Historical factors have certainly played a role in this regard. The Vail Lake area has long been inaccessible to botanists, having been a part of the sprawling Vail Ranch and later held in private ownership by recreational vehicle parks and now developers. The area did receive some attention from such notable botanists as Philip Munz, F.W. Peirson, Marcus E. Jones, and Edmund Jaeger during the early part of this century. However, at that time, the C. ophiochilus population, which is sur- rounded by dense chaparral on all sides, would still have been some distance from any contemporary access roads. It is likely that past botanical explo- ration in this region has been confined to those areas most readily accessible from the main highways. An extensive area from Mount Palomar northward to Bautista Canyon apparently remains completely unexplored botanically. It appears unlikely that other populations of Ceanothus ophiochilus exist, however. The Oak Mountain pyroxenite rich outcrop is unusually large, yet it has not been reported by geologists working in the area (Morton 1989, pers. comm.; Seay 1964; Mann 1955; Rogers 1965). Morton (1989, pers. comm.) indicated that other large pyroxenite rich outcrops are not known from the surrounding area and that the presence of similar large outcrops is doubtful. He suggested a possible link between the outcrop and past thermal activity associated with a small, extinct volcano of late Pleistocene or sub-Recent age (Mann 1955) at the mouth of Temecula Creek Canyon, approximately 1 mile to the northwest. No distinction of soil type is indicated by U.S. Soil Conservation Service maps (Knecht 1971). However, Tullock, et. al. (1989, pers. comm.) found that soil samples taken within the areas supporting C. ophiochilus were deficient in phosphorus, with ridgetop soils having no detectable amounts. '{Note: The name ”Oak Mountain” has been applied by the USGS to two important physical features in the Vail Lake area; in this case we refer to the 2130-foot “Oak Mountain” (USGS 7.5’ Vail Lake quadrangle) which corresponds to Mann’s (1955) “Vail Mountain” vs. the 2705-foot “Oak Mountain” north of Vail Lake (USGS 7.5’ Sage quadrangle).] 32 PAHY-T OLO GIA volume 70(1):28-41 January 1991 Ceanothus ophiochilus is unique among the Cerastes taxa from Southern California in having blue to pinkish lavender flowers rather than the usual white to cream white. However, the leaves are the most striking morphological feature of C. ophiochilus and unequivocally distinguish it from all other taxa in section Cerastes. No other species in the section is reported to have leaves as small and narrow, and no other species in either section of Ceanothus has been reported to possess leaves with a strongly gibbous lower surface (Figure 2, Figure 3). Another morphological feature apparently unique to C. ophiochilus relative to other members of section Cerastes is the general lack of pubescence, especially on the abaxial leaf surface. Except for the very youngest leaves at growth tips, leaf pubescence is limited to the trichomes guarding the entrance to each stomatal crypt (Figure 3, Figure 4). Stomatal crypt structure appears to be Type II as defined by Nobs (1963), with the trichomes confined to the border of the crypt aperture. The tiny, fascicled leaves of Ceanothus ophiochilus give vegetative plants a strikingly similar appearance to Adenostoma fasciculatum Hook. & Arnott, the codominant shrub at the type locality. This condition may. also help to explain the relatively late discovery of this species. Except during the brief flowering period, C. ophiochilus is difficult to differentiate from the surrounding Adenostoma without relatively close examination. Using McMinn’s (1942) treatment of the genus, Ceanothus ophiochilus keys to C. ramulosus (E. Greene) McMinn var. fascicularis McMinn. This is based on the shared characters of hornless capsules, relatively entire leaf margins, leaves borne in axillary fascicles in older wood, and light blue to pale lavender flowers. That taxon, however, is endemic to coastal terraces in Santa Bar- bara and San Luis Obispo counties, California, and differs in having slender, spreading branches; a larger capsule (4.7 mm); and larger (6-20 mm), pla- nar leaves which are dark green on the upper surface and minutely canescent on the lower surface (McMinn 1942; Munz 1959). Ceanothus ramulosus var. fascicularis also differs in possessing Type I stomatal crypts (Nobs 1963). Using Munz’s (1974) floristic treatment for southern California, Ceanothus ophiochilus keys (albeit not smoothly) to C. greggii A. Gray. This is based primarily on the concave upper leaf surface and relatively small, lateral horns on the capsules (at most, vestigial in C. ophiochilus, but also vestigial in some forms of C. greggii). Ceanothus greggii is a highly variable, primarily desert, species which ranges from Oaxaca, México, in the south, northward through the Chihuahuan, Sonoran, Mojave, and Great Basin Deserts, as well as into the arid interior cismontane regions of southern California (McMinn 1942). The species is represented by three more-or-less geographically defined varieties: the typical var. greggzi; var. vestitus (E. Greene) McMinn; and var. perplezans (Trelease) Jeps. (McMinn 1942). Variety greggii ranges northward from Oax- aca, México, into Texas, New Mexico, and Arizona. Variety vestitus ranges northward from Arizona through Nevada and Utah, and westward into Cali- 33 ophiochilus new species in California Boyd,:.et.al: Ceanothus CEANOTHUS OPHIOCHILUS (RHAMNACEAE): A DISTINCTIVE, NARROWLY ENDEMIC SPECIES FROM RIVERSIDE COUNTY, CALIFORNIA Steve Bovd, Timothy Ross. Laurel Arnseth Figure 1. Ceanothus ophiochilus Bovd, Ross, & Arnseth, sp. nov. A) Flowering stem il- lustrating divaricate branching pattern and fascicled leaves on older wood. B) Flower detail. Glandular disk surrounding ovary provides much of the floral color. C) Leat fascicle and four leaves removed from one plant to show variation in size. shape. and margin. Note vestigial marginal teeth on severai leaves. D) Inflores- cence (lower flowers removed). E) Infructescence, one capsule having dehisced. F) Individual seed, abaxial and adaxial views. Scale bar increments = 1mm. Pay ot Ger A volume 70(1):28-41 January 1991 CEANOTHUS OPHIOCHILUS (RHAMNACEAE): A DISTINCTIVE, NARROWLY ENDEMIC SPECIES FROM RIVERSIDE COUNTY, CALIFORNIA Steve Boyd, Timothy Ross, Laurel Arnseth Figure 2. /n situ photograph of Ceanothus ophiochilus in late anthesis showing early fruit development. Photo taken April 1, 1989. Boyd, et al: Ceanothus ophiochilus new species in California 35 —_ a - CEANOTHUS OPHIOCHILUS (RHAMNACEAE): A DISTINCTIVE, NARROWLY ENDEMIC SPECIES FROM RIVERSIDE COUNTY, CALIFORNIA Steve Boyd, Timothy Ross, Laurel Arnseth ‘igure 3. SEM photograph showing a near-apical view of Ceanothus ophiochilus leat with three vestigial teeth, one apical and two marginal. Note the glabrous surfaces. pubescence being limited to the entrances of the stomatal crypts. Scale bar = 200 microns. 36 PHYTOL OGI-A volume 70(1):28-41 January 1991 2 CEA NOTHUS OPHIOCHILUS (RHAMNACEAE): A DISTINCTIVE, NARROWLY ENDEMIC SPECIES FROM RIVERSIDE COUNTY, CALIFORNIA Steve Bovd, Timothy Ross, Laurel Arnseth Figure +. SEM photograph of cross-section through a Ceanothus ophiochilus leat illustrat- ing concave upper surface. rounded margins, strongly convex underside. and rela- lively large, Type II stomatal crypts with trichomes limited to the crypt aperture. Photosynthetic mesophyll is apparently limited to the area immediately surrounding the stomatal crypts with most of the remaining area occupied by bundle sheath ex- tension cells. Scale bar = 200 microns. Boyd, et al.: Ceanothus ophiochilus new species in California 37 fornia through the Mojave Desert ranges as far as southern San Luis Obispo County. Variety perplezans occurs in California from the southern slopes of the San Bernardino and Little San Bernardino Mountains southward through the peninsular ranges into northern Baja California. This third taxon occurs as a common element in desert transition chaparral vegetation on the Anza Bench, less than 10 miles east of Vail Lake. Ceanothus greggit differs from C. ophiochilus in possessing larger leaves (9- 19 mm long, 6-9 mm wide) which are not fascicled in older wood, and which are grayish canescent on both surfaces in vars. greggiz and vestitus and on the lower surface in var. perplerans. These taxa also differ from C. ophiochilus in possessing Type III stomatal crypts (Nobs 1963). Likewise, the capsules are generally larger (3-5 mm in diameter), and the flowers are white. In overall morphological aspect, Ceanothus ophiochilus more closely ap- proaches C. greggit than C. ramulosus var. fascicularis. The shallowly concave folded upper leaf surface of C. ophiochilus suggests a xeromorphic reduction of C. greggii (s.l.) leaves and the yellowish green leaf color is similar to that in C. greggit var. perplerans. Likewise, the branching patterns of C. ophiochilus and C. greggii are similar. On the basis of shared morphological features, it would appear that the closest affinities of C. ophiochilus may lie with the C. greggi complex. We are hesitant, however, to speculate further on the relationships or possible origins of C. ophiochilus solely on the basis of morphology. The com- plexities inherent in edaphic endemism (Raven 1964; Mason 1946a,b) may, in this instance, be further complicated by hybridization and introgression with contemporary sympatric Ceanothus species over geologic time. The relation- ships of C. ophiochilus to other taxa in the section would best be elucidated through a comprehensive analysis of introgressive hybridization and reticulate evolution within the section as a whole. The xeromorphic features exhibited by Ceanothus ophiochilus, especially reduction of leaf size, plant stature, and pubescence, correspond with mor- phological responses to serpentine substrates as exhibited by numerous other taxa (Kruckeberg 1984). While the Oak Mountain substrate is not serpentine, it is similar to many serpentine outcrops in its rocky, poorly developed soil strata, low levels of calcium, and extremely low amounts of available phospho- rus (Kruckeberg 1984). Because of the combination of harsh physical features and the stresses related to the chemical composition of the soil, it is not un- expected to observe similar morphological responses on the Oak Mountain outcrop. Today, selective survivability on this substrate may play an important role in maintaining the integrity of Ceanothus ophiochilus as a distinct species. Extensive interspecific hybridization is well known in Ceanothus, especially in section Cerastes (McMinn 1944; Nobs 1963). It is not surprising, therefore, that several individuals of suspected hybrid origin were located within the C. ophiochilus population (Boyd, Ross, & Arnseth N°s 3017, 3018, 3098, 3099 [all 38 PHY TO LO:.G'TA volume 70(1):28-41 January 1991 RSA]). Large populations of C. crassifolius Torr. are present on sedimentary substrates less than one-half mile to the south and east of the C. ophiochilus population. In 1989, anthesis in the C. crassifolus populations in the Vail Lake area overlapped and extended beyond that of C. ophiochilus; consequently, it appears that ample opportunity exists for gene flow between the two taxa. Ceanothus crassifolius differs considerably from C. ophiochilus in characters and general appearance. This white flowered species is a stout, erect, openly branched shrub 2.0-3.5 m in height with tomentose twigs. Leaves are broadly elliptic to + elliptic obovate, 15-30 mm long, strongly revolute to nearly planar, olive green and glabrous above, and white tomentose beneath. The viscid fruits are globose with short (but prominent), subdorsal horns and are 7-8 mm broad (Munz 1959). Ceanothus crassifolius does, however, share with C. ophiochilus the Type II stomatal crypt (Nobs 1963). The hybrid plants that we observed displayed various stages of intermediacy between these two taxa. Interestingly, all suspected hybrids were found growing at the margins of the C. ophiochilus population, near the contact zone with adjacent metasedimentary substrates. The species of Ceanothus section Cerastes lack the ability to crown-sprout and only reproduce from seeds stored in the soil, generally in response to fire or physical disturbance (Hadley 1961; Raven 1977). We suspect that C. ophiochilus, while occasionally hybridizing with C. crassifolius, is maintained in relatively pure stands due to differential establishment of seedlings following fire. Ceanothus ophiochilus is probably better able to tolerate the unusual edaphic conditions present on the outcrop than is the regionally more abundant C. crassifolius. Apparently, hybrid individuals become established and flourish only at the margins of the outcrop where unfavorable edaphic conditions have been ameliorated. The long term prospects for the continued existence of Ceanothus ophiochi- lus in the wild are uncertain at this time due to the threat of human encroach- ment. Southwestern Riverside County is currently undergoing some of the most rapid urbanization anywhere in the state and, regrettably, C. ophiochilus occupies a locality now considered suitable for development. It does not stand alone. The Vail Lake region also harbors the largest known populations of both the state listed Mahonia nevinii (A. Gray) Fedde [= Berberis nevinii A. Gray] and the federally listed Dodecahema leptoceras (A. Gray) Reveal & Hardham [= Centrostegia leptoceras A. Gray; Chorizanthe leptoceras (A. Gray) S. Wat- son], as well as significant populations of several other sensitive taxa (Boyd, Arnseth, & Ross 1989). While these other taxa occur elsewhere in limited numbers, the development of the site on which the sole population of Cean- othus ophiochilus occurs would result in the extirpation of the species in its wild state. Even partial development of the land surrounding the site presents a serious threat to the species. Changes in fire regime, either through suppres- sion or through the introduction of unnaturally more frequent burns, could prove disastrous. Likewise, given the unique dynamics involved between the Boyd, et al.: Ceanothus ophiochilus new species in California 39 species and its substrate, we completely reject ez-situ preservation alone as an option, especially given the degree to which Ceanothus are known to hybridize in the artificial confines of cultivation. We recommend that any development plans implemented in the region provide reasonable and effective protection for these plants. Ceanothus ophiochilus should be immediately added to list 1B of the Cal- ifornia Native Plant Society’s Inventory of Rare and Endangered Vascular Plants with a R-E-D code of 3-3-3, its most sensitive rating. The species is limited to a solitary, highly restricted population; is endemic to the state of California; and is currently endangered by proposed development of the area. In addition, we strongly urge the immediate listing of Ceanothus ophiochilus as an endangered species by the California Department of Fish and Game, and by the United States Fish and Wildlife Service to provide this species the greatest legal protection possible from encroachment by human activities. ACKNOWLEDGMENTS We express our thanks to Dr. Clifford Schmidt, Dr. Sherwin Carlquist, and Andrew C. Sanders for reviewing this manuscript and offering various com- ments and suggestions. Special thanks to Kendall for providing the excellent line drawing in spite of our pitiably low budget. Thanks to Dr. Robert Tullock, Leta Barber, Vince Gallegos, and Matt Riha for having devoted so much time to analyzing soil samples from the Vail Lake site. Thanks also to Dr. Douglas Morton for personal observations relative to the geology of western Riverside County with specific regard to the substrate at the site, to Dr. Carlquist for providing the SEM photos, and to Dr. Scott Zona, Michael Hanson, Hugo Cota, and several others for assistance in various ways. REFERENCES Boyd, Steven D., Laurel E. Arnseth, & Timothy S. Ross. 1989. Botanical assessment of Bedford Properties holdings surrounding Vail Lake, Pauba Valley, Western Riverside County, California (Technical Appendix A). 51 pp. In Nelson, S. Gregory, S.D. Boyd, L.E. Arnseth, & T.S. Ross. 1989. Biological assessment of Bedford Properties holdings surrounding Vail Lake, Pauba Valley, Western Riverside County, California. Riverside County Planning Department, Riverside, California. 124 pp. Hadley, Elmer B. 1961. Influence of temperature and other factors on Cean- othus megacarpus seed germination. Madrono 16:132-138. 40 PHYEOLOGTIA volume 70(1):28-41 January 1991 Knecht, Arnold A. 1971. Soil survey of western Riverside area, California. Soil Conservation Service, U. S. Government Printing Office, Washing- ton, D.C. 216 pp. Kruckeberg, Arthur R. 1984. California Serpentines: Flora, Vegetation, Ge- ology, Sotls, and Management Problems. University of California Press, Berkeley, California. 180 pp. Mann, John F., Jr. 1955. Geology of a portion of the Elsinore Fault Zone, California. California Division of Mines and Geology Special Report 43. 22 pp. Mason, Herbert L. 1946a. The edaphic factor in narrow endemism. I. The nature of environmental influences. Madrono 8:209-226. 1946b. The edaphic factor in narrow endemism. II. The geo- graphic occurrence of plants of highly restricted patterns of distribution. Madrono 8:241-257. McMinn, Howard E. 1942. A systematic survey of the genus Ceanothus. pp. 131-304 in Van Rensselaer, Maunsell & H. E. McMinn, 1942. Ceanothus. Santa Barbara Botanic Garden, Santa Barbara, California. 308 pp. Morton, Douglas M. 1989. Personal Communication. Department of Geol- ogy, University of California, Riverside, California. Munz, Philip A. 1974. A Flora of Southern California. University of Cali- fornia Press, Berkeley, California. 1086 pp. —___— & David D. Keck. 1959. A California Flora. University of California Press, Berkeley, California. 1681 pp. Nobs, Malcolm A. 1963. Experimental studies on species relationships in Ceanothus. Carnegie Institution of Washington Publication 623. Wash- ington, D.C. 94 pp. Raven, Peter H. 1964. Catastrophic selection and edaphic endemism. Evo- lution 18(2):336-338. ————. 1977. The California flora. 30 pp. in Barbour, Michael G. & Jack Major. 1977. Terrestrial Vegetation of California. John Wiley & Sons, New York. 1002 pp. Rogers, Thomas H. 1965. Geologic map of California, Santa Ana sheet. California Division of Mines and Geology. Scale 1:250,000. Boyd, et al.: Ceanothus ophiochilus new species in California 41 Seay, Mervyn W. 1964. Geology of the Aguanga Basin in the vicinity of Vail Lake, southern Riverside County, California [Bachelor’s Thesis]. Univer- sity of California, Riverside, California. 57 pp. Tullock, Robert, Leta Barber, Vince Gallegos, & Matt Riha. 1989. Personal Communication. Department of Soil Science, California State Polytech- nic University, Pomona, California. Phytologia (January 1991) 70(1):42-43. A NEW SPECIES OF COREOCARPUS (ASTERACEAE - COREOPSIDEAE) FROM MEXICO STATE, MEXICO B.L. Turner Department of Botany, The University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT A new species, Coreocarpus ixtapanus, from the state of Mexico, is described. It is closely related to C. congregatus (S.F. Blake) E.B. Smith of Sinaloa and Durango. Both of these taxa appear to be anoma- lous in Coreocarpus. The taxonomic implications of this anomaly are briefly discussed. KEY WORDS: Asteraceae, Coreopsideae, Coreocarpus, México. Routine identification of Mexican Asteraceae has revealed the following novelty. Coreocarpus ixtapanus B. Turner, sp. nov. Coreocarpo congregato (S.F. Blake) E.B. Smith similis sed plan- tis minoribus foliis ac capitulis minoribus, et antheris luteis (vs. purpureis) differt. TYPE: MEXICO. Estado México: Mpio. Tonatico, along Mex- ican hiway 55, a few mi S of Ixtapan de la Sal, ca. 1800 m, 14 Oct 1962, D.L. & M.L. Denham 172 (HOLOTYPE: COLO!). Delicate slender annual 10-26 cm high. Stems terete, sparsely hirsute, more so at the nodes. Leaves opposite, 3-6 pairs to a stem, 8-15 mm long, 3-10 mm wide; petioles 1-2 mm long; blades mostly tripinnatisect, sometimes merely trilobate, sparsely hirsute with clear multiseptate hairs 0.5-1.5 mm long. Heads 1 or 2 to a stem, the peduncles mostly 2-3 cm long at maturity. Involucres ca. 6 cm high; outer herbaceous bracts 5-8 in a single series, 3-4 mm long, 0.5-0.8 mm wide; inner petaloid bracts yellow, 5-6 mm long, ca. 2 mm wide, the apices acute. Receptacular bracts linear-lanceolate, yellow, longer than the florets, the apices narrowly acute. Ray florets 5-8, yellow, 2-5 mm long, neuter, presumably sterile. Disk florets 7-10 per head, the corollas ca. 42 Turner: New Coreocarpus from México 43 2.5 mm long, the throats ca. 0.8 mm long, sparsely glandular pubescent, the limb ca. 1.7 mm long. Anther sacs yellow with narrowly acute appendages. Style branches abruptly apiculate apically. Achenes (the outer series) broadly obovate, black, incurved, ca. 2.5 mm high, ca. 2 mm wide, the margins with a narrow somewhat corky wing, epappose. The species is known only by the type sheet upon which are mounted 10 specimens; 9 of these are very delicate with rather minute heads, the remainder is ca. 26 cm high. Coreocarpus iztapanus Turner is clearly related to C. congregatus (S.F. Blake) E.B. Smith, of Sinaloa and closely adjacent Durango, but the latter is a much more robust plant with larger leaves, larger heads with more numerous ray and disk florets, the anthers decidedly purple (vs. yellow). Blake originally described Coreocarpus congregatus as belonging to the genus Coreopsis, but Smith (1989) transferred the species to Coreocarpus where it appears to be anomalous. Melchert (pers. comm.), who has excep- tional familiarity with the genus Bidens, excludes it from the latter genus. As already noted, Coreocarpus congregatus and Coreocarpus iztapanus differ markedly from other species of Coreocarpus in possessing dimorphic involucral bracts, much as in Bidens; additionally, the ray florets of these two species are neuter, whereas those of the remaining taxa are pistillate. Smith (1989) positioned Coreopsis congestus in Coreocarpus, largely because of the incurved peripheral achenes, which resemble those of most taxa of Coreocarpus, but he also called attention to its resemblance to Coreocarpus hintoni Sherff and Coreocarpus cronquisti Sherff, both of which are positioned within Bidens by Melchert (cf. Melchert & Turner 1990). In short, Coreocarpus congregatus and Coreocarpus iztapanus appear to stand somewhere between Coreocarpus and Bidens, and both species might ultimately reside in their own genus, if not within an expanded Bidens. ACKNOWLEDGMENTS I am grateful to Dr. Bill Weber for the loan of material from COLO; Dr. Guy Nesom provided the Latin diagnosis and reviewed the manuscript, as did Dr. Andrew McDonald. LITERATURE CITED Melchert, T. & B. Turner. 1990. New species, names and combinations in Mexican Bidens (Asteraceae: Coreopsideae). Phytologia 69:20-31. Smith, E.B. 1989. A biosystematic study and revision of the genus Coreo- carpus (Compositae). Syst. Bot. 14:448-472. Phytologia (January 1991) 70(1):44-46. A NEW GYPSOPHILIC SPECIES OF MIRABILIS (NYCTAGINACEAE) FROM NUEVO LEON, MEXICO B.L. Turner Department of Botany, The University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT A new species, Mirabilis nesomii, is described and illustrated. It is known from only three collections, all of which were obtained from exposed gypseous outcrops at the western base of Cerro Pena Nevada in southern Nuevo Leon, México. KEY WORDS: Mirabilis, Nyctaginaceae, México Routine identification of plants from northern México has revealed the following novelty. Mirabilis nesomii B. Turner, sp. nov., Figure 1. Mirabih glabrifoliae (Ort.) I.M. Johnston similis sed laminis foliorum lineari-lanceolatis plerumque 1-3 mm latisque (vs. ovatis vel cordatis plerumque 10-40 mm latisque) differt. TYPE: MEXICO. Nuevo Leén: ca. 30 km ENE of Dr. Arroyo, 8 km ENE of San Antonio de Pena Nevada, W base of Cerro Pena Nevada, large area of gypsum hills and outcrops, 1950-2050 m, 30 Jul 1983, Guy Nesom 4704 (HOLOTYPE: TEX!; Isotype: MEXU). Perennial glaucescent herbs 25-35 cm high. Stems erect, glaucescent, glan- dular punctate, pubescent at nodes, with vestiture extending between nodes in narrow bands mostly ca. 1 mm wide, otherwise glabrous. Leaves linear- lanceolate, mostly 3-8 cm long, 0.1-0.3 cm wide, glabrous or nearly so, the margins enrolled, most of the foliage confined to the lower 1/3 of the stem, the upper leaves much reduced and scalelike. Flowers on 2-3 terminal, bracteate, branches, the latter 3-10 cm long, the ultimate involucres borne on pilose pe- duncles 5-6 mm long. Involucres ca. 8 mm high at maturity, 5 parted, the bracts united for about 5/8 their length, pubescent with multiseptate tri- chomes. Petaloid structures apparently pink, ca. 10 mm high and twice as 44 New Mirabilis from México 46 PHYTOLOGIA volume 70(1):44-46 January 1991 wide, the stamens ca. 12 mm long, exserted. Fruiting bodies (anthocarp) broadly clavate, ca. 3.5 mm long, the stipe ca. 0.6 mm long, the body with 5 tuberculate ribs, between these a very fine vestiture of minute glandular hairs. ADDITIONAL SPECIMENS EXAMINED (Paratypes): MEXICO. Nuevo Leon: 2.5 km ENE of San Antonio de Pena Nevada, W base of Cerro Pena Nevada, ca. 2000 m, 3-5 Aug 1981, Nesom 4275 (TEX, MEXU); ca. 7 km NE of San Antonio Pena Nevada, Jul 1977, Wells & Nesom 506 (TEX, unicate). Mirabilis nesomiu Turner, with its markedly linear-lanceolate, glaucescent leaves, superficially resembles M. linearis (Pursh) Heimerl, but is clearly more closely related to the widespread M. glabrifolia (Ort.) I1.M. Johnston in char- acters of the involucre and fruit, the latter appearing almost indistinguishable from that of M. nesomiz. Mirabilis nesomii differs from M. glabrifolia in being a much smaller plant, with fewer flowering branches and, as noted, by having linear or linear-lanceolate leaves, mostly 1-3 mm wide (vs. ovate to cordate and 10-40 mm wide). ACKNOWLEDGMENTS I am grateful to Guy Nesom for the Latin diagnosis and to him and Linda Escobar for reviewing the manuscript. Nancy Webber provided the illustra- tion. Phytologia (January 1991) 70(1):47-51. STUDIES ON MIKANIA (COMPOSITAE: EUPATORIEAE) — XVII: TWO NEW SPECIES FROM MINAS GERAIS, BRAZIL Walter C. Holmes Department of Biology, Baylor University, Waco, Texas 76798-7388 U.S.A. ABSTRACT Mikania citriodora and Mikania hartbergii, two new species from the Serra do Espinhacgo, Minas Gerais, Brazil, are described and illustrated. KEY WORDS: Compositae, Eupatorieae, Mikania, Minas Gerais, Brazil Continued study of the genus Mikania has resulted in the recognition of the following new species from the Serra do Espinhago, Grao Mogol, Minas Gerais, Brazil, an area characterized by a high rate of endemism. Mikania citriodora W. Holmes, sp. nov. (Figure 1). TYPE: BRAZIL. Minas Gerais: Serra do Espinhaco, Grao Mogol, ca. 2 km from center of town via Vila Nova, 950 m; sandy soil over sandstone; common, 12 Jun 1990, W.C. Holmes 5064 (HOLOTYPE: MBM; Isotypes: BAYLU, IBE, NLU, TEX). Species ad Mikaniam rufescem Schultz-Bip. similis sed differt planta tomentosa (non glabra) et foliis crenato-dentatis (non inte- gris). Herbaceous to semiwoody sprawling to twining vines growing from elon- gated knotty caudices; stems terete, glabrate (at bases) to tomentose (upper parts); internodes to 20 cm long. Leaf blades ovate to ovate deltate, 2.2-4.0 x 1.3-4.0 cm, apices obtuse to rounded, margins crenate-dentate, bases ob- tuse to truncate to an acute insertion at the petioles, trinervate from near the base, surfaces tomentose, spotted with glandular resinous dots; petioles 1.0-1.3 cm long, tomentose. Capitulescences corymbose, 3-6 x 5-8 cm; branch- lets terete, tomentose; bracts ovate, 0.8-1.7 x 1-2 cm, tomentose; ultimate 47 48 BPAY PTOLOGIA volume 70(1):47-51 January 1991 Figure 1. Mikania citriodora W. Holmes. A and B. leaves; C. head with phyllaries and subinvolucral bract; D. flower and achene; and E. branched trichome from upper part of achene. Holmes: Studies on Mikania — XVII 49 branchlets 1.5-5.0 mm, tomentose. Heads 8-10 mm long; subinvolucral bracts oblanceolate to elliptic, 4-6 mm long, apices acute, surfaces pilose to tomen- tose. Phyllaries elliptic-oblong, 5.5-7.0 mm long, apices acute, surfaces pilose to tomentose. Corollas 3.8-5.0 mm long, creamy white, tubes 1.7-2.5 mm long, throats funnelform to semicampanulate, ca. 1 mm long, teeth ovate, triangular to triangular ovate, 1.1-1.5 mm long. Achenes 2.6-3.5 mm long, 7 ribbed, the ribs upwardly scabrid, surfaces olivaceous, pilose (to tomentose at the summit) with branched trichomes. Pappus bristles 5.0-5.5 mm long, white, 40-50, the margins scabrid. PARATYPE: BRAZIL. Minas Gerais: Serra do Espinhacgo, Grao Mogol, mountains to the west of town; 1170 m, sand over sandstone; common, 14 Jun 1990, W.C. Holmes 5070 (BAYLU, IBE, MBM, NLU, TEX). Mikania citriodora W. Holmes has several very unusual characteristics for the genus. The injured fresh stem has a faint smell of lemon, hence the specific name. While most species of Mikanza have five ribbed achenes, the new species has achenes with seven ribs. Several species of erect Mzkania, formerly included in the segregate genus Kanimia, are reported to have ten angled achenes. Certainly the most interesting trait is the presence of multicellular branched trichomes on the achenes, a trait not known in other Mikania. Typically, Mikania have multicellular, prominently jointed, but unbranched trichomes. This is one of the few Mikania species reported to be aromatic. Others include M. anisodora Hassler, of Paraguay and Parana, Brazil, the fresh foliage reportedly having an anise odor (Hassler 1915) and M. aromatica Oersted (Scharling & Oersted 1863), a Brazilian plant described as having the odor of cumin. The latter name is a synonym of M. smilicina DC. Mikania hartbergii W. Holmes, sp. nov. (Figure 2). TYPE: BRAZIL. Minas Gerais: Serra do Espinhaco, Grao Mogol, mountains to the west of town, 1250 m, 14 Jun 1990, W.C. Holmes 5071 (HOLOTYPE: MBM; Isotypes: BAYLU, IBE, NLU, TEX). Species ad Mikaniam neurocaulum DC. similis sed differt cauli- bus teretibus (non profunde sulcatis) et foliis brevissime petiolatis (non longe petiolatis). Erect to ascending suffrutescent herbs, 0.5-1.7 m tall, single to multi- stemmed from knotty rootstocks; stems terete, velutinous, ca. 1 cm in diameter at the base; internodes 2.0-3.5 cm long. Leaf blades ovate, 2.2-4.6 x 1.7-3.2 cm, semicoriaceous, apices acute to a mucronate point, margins entire to den- ticulate, often revolute, bases truncate to subcordate, venation subpinnate with 2 pairs of secondary nerves separating from the midvein within the lower 8 mm of the blade; upper surfaces hirsute to pilose, prominently reticulate, 50 PHYTOLOGIA volume 70(1):47-51 January 1991 V Fa vA fy pf | wh Y { i st 1 \ Me f in) \! PUM ch a ry 4 : weylal! Figure 2. Mikania hartbergii W. Holmes. A capitulescence; B. leaf; C. flower and achene; D. head with phyllaries and subinvolucral bract. Holmes: Studies on Mikania — X VII 51 lower surfaces velutinous to pilose, reticulate; petioles 1.0-2.5 mm long, veluti- nous. Capitulescences thyrsoid corymbs, 9-15 x 11-15 cm, the heads ultimately disposed in ternately branching and congested corymbs 1.0-1.5 x 1.5-3.0 cm; branchlets terete, velutinous; bracts similar to leaves but reduced in size; ul- timate branchlets 0.5-3.5 mm long. Heads 6-8 mm long; subinvolucral bracts linear, 2.0-3.5 mm long, pilose especially on the apices and margins. Phyllaries ovate-oblong, ca. 3.8 mm long, apices rounded, ciliate-pilose, margins ciliate, surface glabrate to remotely puberulent; bases slightly calcarate. Corollas white, 4.8-5.2 mm long, tubes 1.6-1.7 mm long, throats funnelform to semi- campanulate, 1.5-2.0 mm long, teeth lance-ovate, 1.2-1.5 mm long, sparingly pilose at the apices. Achenes (immature) ca. 1.8 mm long. Pappus bristles ca. 6 mm long, white, 35-40, margins scabrid, apices slightly thickened. PARATYPE: BRAZIL. Minas Gerais: Serra do Espinhago, Grao Mogol, mountains to the west of town, 1030 m, 14 Jun 1990, W.C. Holmes 5068 (BAYLU, IBE, MBM, NLU, TEX). The new erect to ascending species of Mikania is known only from the Serra do Espinhago near Grao Mogol. Several colonies of about 10-12 plants were observed from 1030 to 1250 m altitude, but specimens were collected only from the two colonies that possessed mature flowers. Plants were usually rooted in dry, sandy crevices in sandstone. The species appears closely related to Mikania neurocaula DC., but can be distinguished by its terete stems, truncate to subcordate leaf bases, and subsessile to very shortly pedicellate leaves. Mikania neurocaula is described as having profoundly sulcate stems, acute leaf bases, and leaves with petioles of about 1.5 cm long. It is a pleasure to name this species for W. Keith Hartberg, Professor and Chairman of the Biology Department of Baylor University. ACKNOWLEDGMENTS I am grateful to Baylor University for providing financial support that made this study possible. I also wish to thank Sidney McDaniel and R. Dale Thomas for review of the manuscript and Gert Hatschbach for other assistance. LITERATURE CITED Hassler, E. 1915. Ex Herbario Hassleriana: Novitates Paraguariensis. XX. Feddes Repert. 14:172. Scharling, E.A. & A. Oersted. 1863. [Modet den 9 Januar.]. Overs. Kongel. Danske Vidensk. Selsk. Forh. Medlemmers Arbeider 10:1-12. Phytologia (January 1991) 70(1):52-57. A NEW PITCAIRNIA (BROMELIACEAE) FROM PERU Lyman B. Smith Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 U.S.A. ABSTRACT Pitcairnia cerrateana sp. nov. (Bromeliaceae) from Peru is de- scribed. KEY WORDS: Pitcairnia, Bromeliaceae, Peru Pitcairnia cerrateana L.B. Smith, sp. nov. Figures A-E. TYPE: PERU. Dep. Ancash: ascending to Huaylas, cerro subxerofilo, Prov. Huaylas, 1880-1920 m alt., 3 July 1988, E. Cerrate 8967 (HOLOTYPE: USM!). Ab omnibus speciebus adhuc cognitis foliorum laminis anguste triangularibus omnino serratis, inflorescentia simplici sepalis max- imis differt. PLANT flowering over 6 dm high. LEAVES 27 cm long; sheaths ovate, 5 cm long, laxly and minutely pale lepidote; blades narrowly triangular, 13 mm wide, laxly serrate with antrorse, slender, 3 mm long red spines. INFLORES- CENCE simple, 27 cm long (immature), lax. FLORAL BRACTS suberect, ovate, acute, 11 cm long, exceeding the pedicels, subcoriaceous, farinose, flow- ers spreading, glabrescent. SEPALS oblong-lanceolate, 6 cm long, yellow (! Cerrate). 52 Smith: New Pitcairnia from Peru A new Pitcairmie from Peru Figure A. Pitczirnia cerrateana ].. b. Smith, bese of floral «xis. 53 54 PHYTOL OCTA volume 70(1):52-57 January 1991 A new Pitceirnia from Peru Figure B. Pitcairnia cerrateana L. b. Smith, bese of floral axis. Smith: New Pitcairnia from Peru A new Pitcairnia from Peru “, Figure C. Pitcairnia cerrateana L. B. Smith, inflorescence. 55 56 PHY TOLOGSFA volume 70(1):52-57 January 1991 A new Pitcaimia from Peru Figure D. Pitcairnia cerrateana L. B. Smith, young flower artificislly opened. Smith: New Pitcairnia from Peru A new Pitcairnia from Peru ' @ Figure E. Pitcairnia cerrateana L. b. Smith, infrutescence. 57 Phytologia (January 1991) 70(1):58-59. SOLIDAGO DURANGENSIS (ASTERACEAE: ASTEREAE) A NEW SPECIES FROM MEXICO Guy L. Nesom Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT Solidago durangensis sp. nov. is known by only a few collections from the vicinity of the city of Durango, and it may now be extinct. The closest relative of the new species is hypothesized to be S. paniculata DC. KEY WORDS: Solidago, Astereae, Asteraceae, México Solidago durangensis Nesom, sp. nov. TYPE: MEXICO. Durango (Edo.): city of Durango and vicinity, Apr-Nov 1896, E. Palmer 363 (HOLO- TYPE: US!; Isotype: US!). Solidagini paniculatae DC. similis statura elata et capitules- centia paniculati-corymboideia ampla sed vestimento valde evoluto hispidulo, phyllariis brevioribus, et acheniis pubescentibus differt. Herbs, probably perennial (base not seen), evenly hispidulous on the stems and leaves with white, erect or slightly crisped hairs less than 0.1 mm long, the stems apparently 1 m or more tall, yellowish green to slightly purplish. Leaves narrowly oblanceolate to elliptic-lanceolate, 1 or 3 nerved, 6-10 cm long at midstem, strongly reduced in size in the capitulescence, 6-9 mm wide, the margins slightly revolute, entire or the lower leaves with a few, minute teeth, axillary fascicles not produced. Heads secund in dense corymboid panicles, on bracteate pedicels; phyllaries in 3-4 subequal series, very thin, with a narrow orange midvein, the inner 3.0-3.5 mm long, triangular-lanceolate, the outer minutely fringed-ciliate, otherwise glabrous. Ray flowers 12-18, 2.5 mm long, the ligule ca. 1.5 mm long. Disc flowers 9-14, 3.5-3.8 mm long, the lobes 1.2-1.3 mm long. Achenes sparsely strigose with very thin hairs, 1.4-1.6 mm long, obtriangular-fusiform, with 6 orange resinous nerves; pappus of numerous barbellate bristles. 58 Nesom: New Solidago from México 59 Additional collection examined: MEXICO. Durango: city of Durango and vicinity, Apr-Nov 1896, FE. Palmer 217(F, MO). Although these plants are known only from a few relatively old collec- tions, their morphology is extremely distinctive. Among the Mexican species of Solidago, the tall stature, relatively long leaves, large, paniculate-corymboid capitulescences, and the large number of ray flowers of S. durangensis Nesom are most similar to those of S. paniculata DC. In contrast, the newly de- scribed species differs from S. paniculata in its evenly and strongly developed, hispidulous (vs. glabrous) vestiture, much shorter (vs. 5-6 mm long) phyllaries, and pubescent (vs. glabrous) achenes. Solidago paniculata is closely related to other species occurring in México, S. sempervirens L., S. stricta Ait., and S. confinis A. Gray, all of which are glabrous and characteristically occur in wet habitats. Although the habitat of Solidago durangensis is not known, it is likely that the collections were made from the wet, saline prairies on the east and northeast side of the city (1900-1950 m in elevation), where other narrowly endemic species of vascular plants are known to occur. Since recent collectors of Compositae from this area of Durango apparently have not encountered additional plants of this singular species, the original population may now be extinct, and a careful search for it needs to be made. In the first phases of my study of Mexican Solidago, I annotated some specimens of S. durangensis as S. missouriensis Nutt., and although they have similarities at least in habit, the Mexican representatives of the latter differ from 5. durangensis in their shorter stature, axillary fascicles of leaves, nearly glabrous stems and leaves, shallowly serrate blades, and smaller number of disc flowers. Solidago missouriensis in México is known only from central Coahuila. ACKNOWLEDGMENTS I thank Dr. B.L. Turner and Dr. A. McDonald for their review of the manuscript, Genevieve Lauer for her comments, the staffs of F, GH, and MO for loans of specimens, and the staff of US for their assistance during a visit there. Phytologia (January 1991) 70(1):60-64. BOOK REVIEWS Michael J. Warnock Department of Biological Sciences, Sam Houston State University, Huntsville, Texas 77341 U.S.A. The Encyclopedia of Evolution, Humanity’s Search for its Origins. Richard Milner, foreword by Stephen Jay Gould. Facts on File, Inc., 460 Park Avenue South, New York, NY 10016. 1990. xii. 481 pp. $45.00 (hard- cover). ISBN 0-8160-1472-8. This book is not an encyclopedia in the usual sense of the word, where entries are included with brief definitions or descriptions. Instead, most of the entries in this work are given extensive treatment (the discussions of many entries are more than one page in length). In addition to the expected terms, personages, places, events, and themes of evolution, a wide variety of topics not usually associated with the idea of evolution or the teaching of evolution are also included. Some of the unexpected entries include “Fantasia,” Lonesome George, Napi, King Kong, “Quest for Fire,” Thomas Jefferson, the Sinclair Dinosaur, and many others. These additional entries tremendously expand the usefulness of the work by placing such entries in an evolutionary context, particularly as they pertain to the development of societies’ perceptions of evolution. The extensive use of cross referencing of entries and inclusion for many entries of related reading on the topics is an added feature not common in such works. Both techniques greatly enhance the value of the book and provide access to a much larger stock of knowledge than could otherwise be practically included in such a book. Milner’s book is one which should be included in any educational library, from elementary school to universities. In light of the fact that it provides a context for many of the public’s (mis)conceptions about evolution (a context that scientists are unfortunately often unaware of), Milner’s book will also be useful for, and should be included in the personal library of, biologists who are in positions to communicate the concepts of evolution to nonscientists. 60 Warnock: Book reviews 61 Indicator Plants of Coastal British Columbia. K. Klinka, V.J. Krajina, A. Ceska, & A.M. Scagel. University of British Columbia Press, #303- 6344 Memorial Road, Vancouver, British Columbia V6T 1W5 CANADA. 1989. ix. 288 pp. $36.95 (flexible cover). ISBN 0-7748-0321-5. Klinka, et al. summarize the usefulness as ecological indicators, of a number of plants from coastal British Columbia. Plants are described as indicators of climate, soil moisture, soil nitrogen content, and ground surface material. The content of the book is based largely on previous studies by the authors (partic- ularly Klinka and Krajina). A summary of historical use of plants as ecological indicators in British Columbia is included. The concept of indicator species groups is described and reviewed. As one might expect, a number of plant species parallel one another in their responses to one or more environmental influences. The main body of the book consists of a species listing in which over 400 species of plants are included. Each species treatment includes a color photo- graph of the plant, and a summary of ecological and geographic distribution of the species. Included plants are arranged in the listing alphabetically by genus (and by species within genera). A large diversity of species are included, from lichens, mosses and liverworts, to flowering plants. Although the book contains a wealth of ecological information, it may end up being used more by amateur plant collectors who will use the high quality photographs to identify their plants by comparison with the pictures, rather than being heavily used by professional ecologists. The construction of the book appears to be solid. It is ostensibly built for field use, with the flexible water-resistant cover. The interior pages, although coated, might not withstand heavy field use, especially in a wet environment such as coastal British Columbia. One minor personal complaint about the book is the map on page 3, which is difficult to read, the symbols “CH” and “SH” appear on the map but not the key, and the lower panhandle of Alaska (which is almost completely surrounded by British Columbia and the coast) appears as a large open area in the map-not marked by the climatic zone boundaries found on the British Columbia portion of the map. All in all, the book should be useful for individuals undertaking ecological studies in the coastal regions of British Columbia. 62 PAR POLOGIEA volume 70(1):60-64 January 1991 Packrat Middens, The Last 40,000 Years of Biotic Change. Julio L. Be- tancourt, Thomas R. Van Devender, & Paul S. Martin (eds.). The Uni- versity of Arizona Press, 1230 North Park, Suite 102, Tucson, AZ 85719. 1990. vii. 469 pp. $55.00 (clothbound). ISBN 0-8165-1115-2. The present volume is a compendium of 21 chapters by 26 authors (some authors contributing to multiple chapters and most chapters written by mul- tiple authors), summarizing the use of data from middens to infer the past vegetation of a region. Following the introduction (Chapter 1), the next five chapters (Part I) provide a basis for use of packrat middens as a sampling tool for vegetation analysis. These chapters describe what constitutes a packrat ( Neotoma) midden, the behavior and ecology of packrats, accuracy of sampling from midden data as compared to contemporary field methods, and a sum- mary of sources of material (by radiocarbon dates and geographic location) for data used in the analyses described in later chapters. Part II includes chapters 7-12, each of which describes a summary of Late Quaternary vegetation and climate within a specific part of the study area (Chihuahuan, Sonoran, and Mohave deserts, as well as the Great Basin, Grand Canyon, and Colorado Plateau. Part III (Chapters 13-17) includes reports of use of packrat midden data for purposes other than describing complete regional vegetation patterns. These chapters compare midden data to pollen data at a site in Oregon, grass ecology in certain Sonoran Desert sites, mammal and arthropod distributions from material found in middens, and deuterium variations in cellulose. The final section of the book includes three chapters describing the use for sampling paleovegetation, of middens produced by other animals in other parts of the world. The suggested possibilities include Hyrax and Dassie Rat middens in Africa, Hyrax from the Middle East, and Stick-Nest Rat middens from Australia. The last chapter in the book is a summary of the use of middens and prospectus for future use in vegetation studies. The book provides considerable insight on vegetation in southwestern North America over the past 40,000 years. Not all of the data from packrat middens agrees with previous attempts to describe the vegetation of the region over the same time period. The book will be useful to many individuals, but particu- larly to systematists attempting to interpret variation patterns found in many plant and animal groups of the region. Many of the chapters are illustrated with drawings and photographs which enhance the text. Warnock: Book reviews 63 The Healing Forest, Medicinal and Toxic Plants of the Northwest Amazo- nia. Richard Evans Schultes & Robert F. Raffauf. Volume 2 of the His- torical, Ethno- & Economic Botany Series edited by Theodore R. Dudley. Dioscorides Press, subsidiary of Timber Press, 9999 S.W. Wilshire, Port- land, OR 97225. 1990. 484 pp. $59.95 + $3.00 shipping (hardcover). ISBN 0-931146-14-3. The Healing Forest is a compendium of the use of medicinal and toxic plants from the Amazonian regions of Colombia, Ecuador, Venezuela, and far northwestern Brazil. Much of the field work leading to this text was completed in Colombia by the authors and their students. The primary sources of infor- mation about the plants were conversations with the native medicine men in the region of study and herbarium specimens on which such information was reported by other collectors. The local experts provided vernacular names for and uses of the plants. Specimens were collected and Latin names applied to the plants. In some cases, the plants have been subjected to chemical extrac- tion (either by the authors, or more commonly, by other workers) to determine the molecules responsible for the active properties of the plants. The book be- gins with an extensive (27 pages) preface in which the northwest Amazonian region is briefly described, and the societies and tribal interactions of some of the native peoples are summarized. The preface is generously illustrated with black and white photographs, many of which show the tribal medicine men who served as sources of information. The main body of the text consists of a listing of many of the medicinal and toxic plants of the region. Entries are ordered alphabetically by family, genus, and species. Each family included is briefly described, with a summary of number of genera and species in the family, geographic distribution of the family, economic importance of the family, and known chemical properties of the family. Similar summaries are found for each genus included in the book. Species are listed with full authorities and publication citations (no synonyms), a positive feature of the book that is not often found in similar books. Species listings include vernacular names and the name of the tribe(s) that uses the name. Native uses and methods of preparation of the plant are included in each species listing. Names and collection numbers for specimens from which data on plant use were collected are included where applicable. Ease of use of the book is somewhat impaired by the fact that although genus and species epithets are highlighted by boldface type where they are found in the alphabetized listing, the same names are not even italicized when found in the text of discussion of a particular taxon. Thus (for instance), where genera are mentioned in the general description of a family, the names for those genera are in the same type face as, and blend into the rest of the text. Many of the plants are illustrated by line drawings or black and white photographs. Each family listing ends with a literature summary of studies (mostly chemical) of 64 PHYTOLOGIA volume 70(1):60-64 January 1991 members of the family. This book should prove to be a valuable reference for workers attempting to discover new sources of treatments for various ailments. It will provide at least a starting point in searches for future medicinal treatments. Unfortunately, many of the species listed in this book may soon disappear as the Amazonian forest is destroyed. Also unfortunate is the fact that the information about use of other species not listed in this book will probably disappear as the tribes who use them are brought into the “modern” world and encouraged not to use their traditional methods of medicine. Phytologia (January 1991) 70(1):65-66. BOOKS RECEIVED Annual Review of Entomology, volume 36. Thomas E. Mittler, Frank J. Radovsky, & Vincent H. Resh (eds.). Annual Reviews Inc., Palo Alto, CA. 1991. ix. 703 pp. $40.00 (cloth) ISBN 0-8243-0136-6. As with other volumes in this series, current literature and re- search in Entomology are summarized. Twentyeight papers are included, treating topics ranging from Abamectins to Zophobas atratus. Ecology of Plant Communities, A phytosociological account of the British vegetation. Jack Rieley & Susan Page. Longman Scientific & Technical, John Wiley & Sons, Inc., New York, NY. 1990. ix. 178 pp. $98.00 (hardcover) ISBN 0-582-44639-2. More than the title implies, this book is not only a phytosoci- ological account of the British vegetation, but includes brief sum- maries of plant ecology in general as well as phytosociology (meth- ods and terms) in particular. A brief account of worldwide vege- tation is also included. Orchid Biology, Reviews and Perspectives, V. Joseph Arditti (ed.). Timber Press, Inc., 9999 S.W. Wilshire, Portland, OR 97225. 1991. 451 pp. $58.00 + $3.00 shipping (hardcover) ISBN 0-88192-170-X. This book is an interesting and unusual compilation of reports of scientific study, combined with nonscientific articles. As one might expect from the title, most of the information presented would be of use primarily to individuals particularly interested in orchids. Pieces included in the book include such diverse items as a chapter entitled “Orchids in my life” and a report of a rather detailed study of Rhizanthella gardneri, a parasitic orchid in which only the flower protrudes above the ground. Other articles include 65 66 PHYTOLOGIA volume 70(1):65-66 January 1991 summaries of self pollination in Orchidaceae, water relations in or- chids, a literature review of Dactylorhiza, use of orchids in space re- search, and citations of orchids in literary works. A rather lengthy (100+ pages) appendix summarizes the known months of flowering for orchids found in cultivation. Most articles are illustrated (black and white). The Encyclopedia of Evolution, Humanity’s Search for its Origins. Richard Milner, foreword by Stephen Jay Gould. Facts on File, Inc., 460 Park Avenue South, New York, NY 10016 1990. xii. 481 pp. $45.00 (hard- cover) ISBN 0-8160-1472-8. See review on page 60. Phytologia (January 1991) 70(1):67. ERRORS AND ADDITIONS, VOLUME 69 Volume 69, Issue 6: Title Page, line four reads: Vol. 69 December 1990 No. 5 Replace with: Vol. 69 December 1990 No. 6 67 Phytologia (January 1991) 70(1):68. PUBLICATION DATES FOR VOLUME 69 Volume 69(1): Date on title page-July 1990; Date of mailing-27 July 1990. Volume 69(2): Date on title page-August 1990; Date of mailing-14 Septem- ber 1990. Volume 69(3): Date on title page-September 1990; Date of mailing-17 October 1990. Volume 69(4): Date on title page-October 1990; Date of mailing-14 Novem- ber 1990. Volume 69(5): Date on title page-November 1990; Date of mailing-22 December 1991. Volume 69(6): Date on title page-December 1990; Date of mailing—6 Febru- ary 1991. 68 otani: sical Garden Libra i nM ill 3 5185 0 information for Authors 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 but ~ must be written in DOS format as flat ASCII files. 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. Magocrintoe submitted in dot matrix print cannot be accepted. Use underscore 5 (not italics) for scientific names. Corrections made on typescript — manuscripts must’be complete and neat as the scanner 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 4 nomenclatural summaries should not appear in Literature Cited. 4 Nomenclatural work should include one paragraph per basionym ~ and must provide proper (as defined by the current International ~ Code of Botanical Nomenclature) citation of sources of epithets: on and combinations. 3 Articles from botanical systematics and ecology, including ’ a a Authors should arrange for two workers in the appropriate fi field to 4 review the manuscript before submission. Copies of reviews should ~ be forwarded to the editor with the manuscript. Manuscripts a not be published without review. 5 Cost of publication is currently $12.00 US per page for canoes a tion without reprints. Publication with 100 reprints is provided for — $16.50 US per page, 200 reprints for $20.00 US per page. Page a 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 chafges 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 (horizonal) by © 219 mm(vertical), not including running head and page numbe Title page should include title, author(s) name(s) and address(e Two blank lines should appear above and below section heading (Abstract, Discussion, Literature Cited, etc.) in the manuscript. extra Charge is made for line drawings provided they conform limitations of size and proportion for normal text. Halftones it an extra charge of $5.00 US per page.