owe? fs Bf ee } bj Pd f x An international journal to expedite plant systematic, phytogeographical and ecological publication Vol. 73 September 1992 No. 3 CONTENTS DUDLEY, T.R. & F.S. SANTAMOUR, Jr., Cornus florida subsp. urbini- ana, (Rose) Rickett from México: The correct name for “C. florida var. NL, feta teh aria ah Fue oi eikentaeiy ein mies Be aia ws Lark aes eas oP bt 169 OCHOA, C.M., Determinations of chromosome number (27) and endosperm } balance number (EBN) in some little known tuber bearing Solanum. 180 OCHOA, C.M., Solanum lobbtanum Bitter, a little known Colombian tuber ELIMI R SS ties Wh Or wane ican sae wed Wiss dns dot aoa nai aid oak ate ye 183 ESOM, G.L., Taxonomic notes on Erigeron (Asteraceae: Astereae) of Cal- SPR SINR CUCL DANEEL ANTI ZOMG Gd Pp oie uc iaracs horse Sone A, vp arated oe cele Llewlas 186 NESOM, G.L., Erigeron and Trimorpha (Asteraceae: Astereae) of Nevada. -WIEHLER, H., New species of Gesneriaceae from the Neotropics. ..... 220 TURNER, B.L., A new species of Cymophora (Asteraceae, Heliantheae) from CITIES TAC Wis Foe fee ae a cate poe cieioe Paes bad kale do ee nae 242 ESOM, G.L., A new species of Castilleja sect. Euchroma (Scrophulariaceae) Pe MIR ACR, CVE SACO aye Ala si os he se hoe ears we islet petra asey 247 TURNER, B.L., A new species of Stemodia (Scrophulariaceae) from Costa Ra MORAN SEEN So tues incense anew asica boed era ps yan re ks 251 Contents continued on the inside cover (Contents continued) TURNER, B.L., A new species of Stenocarpha (Asteraceae, Heliantheae, Galinsoginae) from: México. i0 .ce sede see eve ee eee oben 255 _.. KEIL, D.J., A new combination in Hemizonia (Asteraceae: Madiinae. . 259 LIBRARY NOV 3 A 1990 NEW YOR BOTANICAL GARDEN PHYTOLOGIA (ISSN 00319430) is published monthly with two volumes per year by Michael J. Warnock, 185 Westridge Drive, Huntsville, TX 77340. Second Class postage at Huntsville, TX. Copyright ©1991 by PHYTOLOGIA. Annual domestic individual subscription (12 issues): $36.00. Annual domestic institutional subscription (12 issues): $40.00. Foreign and/or airmail postage extra. Single copy sales: Current issue and back issues volume 67 to present, $4.50; Back issues (previous to volume 67), $3.00 (add $.50 per copy postage and handling US [$1.00 per copy foreign]). Back issue sales by volume: $17.00 per volume 42-66 (not all available as complete volumes); $21.00 per volume 67-present; add $2.00 per volume postage US ($4.00 per volume foreign). POSTMASTER: Send address changes to Phytologia, 185 Westridge Drive, Huntsville, TX 77340. Phytologia (September 1992) 73(3):169-179. CORNUS FLORIDA SUBSP. URBINIANA (ROSE) RICKETT FROM MEXICO: THE CORRECT NAME FOR “C. FLORIDA VAR. PRINGLET’ Theodore R. Dudley & Frank S. Santamour, Jr. U.S. National Arboretum, Agricultural Research Service, U.S. Department of Agriculture, Washington, D.C. 20002-1958 U.S.A. ABSTRACT Plants from disjunct populations of Cornus florida L. in México are appearing in the United States nursery trade under the illegitimate name “C. florida var. pringlei.” The correct and legitimate name for this taxon is C. florida subsp. urbiniana (Rose) Rickett. New data are presented on morphological and flowering characteristics of the Mexican flowering dogwood to justify its subspecific status. KEY WORDS: Cornus, Cornaceae, México In a recent issue of the American Nurseryman, Lowrey (1990) described certain features of a very unusual flowering dogwood cultivated in Houston, Texas, that he named “C. florida var. pringlei.” The “Pringle dogwood,” a common name given by Lowrey (Lowrey 1990), now flowering in Texas and other locations as well, including the U.S. National Arboretum in Washington, D.C., originated from seed collected by Mr. Lowrey a number of years ago from a wild-occurring disjunct population of C. florida in the Sierra Madre Oriental mountains near Monterrey, Nuevo Leon State, Mexico. This plant is now, or soon will be, entering the commercial nursery trade and be available to consumers. The purposes of this paper are to: (1) show that the name or epithet “pringler” is not valid at the level of species or variety, or any other botani- cal rank; (2) provide a valid botanical name at an appropriate rank for this Mexican flowering dogwood; and (3) describe and elaborate on the unusual characteristics of this taxon. Application of correct botanical classification or identification (taxonomy) and the use of accurate and proper botanical names (nomenclature) are cor- nerstones of understanding essential to communication among and between 169 170 P..B YAP OL O GALA volume 73(3):169-179 September 1992 Figure 1. Cornus florida subsp. urbiniana (Rose) Rickett in México, Nuevo Leon State, Sierra Madre Oriental, near Monterrey. Photograph courtesy of Lynn Lowrey. Dudley & Santamour: Cornus florida subsp. urbiniana £71 plant scientists, horticulturists, and nurserymen. Such communication is es- pecially needed in the genus Cornus at the present time because of a renewed interest in the large bracted flowering dogwoods brought about by the dis- covery and rapid spread of the deadly dogwood anthracnose disease, which is now wreaking havoc among native United States populations of C. florida L. Santamour & Dudley (1992) recently presented a taxonomic and cytogenetic summary of Cornus to a group of plant pathologists and mycologists engaged in dogwood anthracnose research in order to increase their awareness and un- derstanding of the extraordinary complexity of this genus. As new flowering dogwood germplasm is being sought, worldwide, for evaluation, disease testing, hybridization, and introduction to the nursery industry, we must assiduously guard against the misapplication and use of incorrect plant names before they become firmly entrenched in the parlance of the nursery trade and “popular” horticultural publications. There never has been a properly described and validly published taxonomic plant entity called “C. pringlev” or “C. florida var. pringler.” All the standard references and appropriate floras have been consulted and checked. These names simply have never been published in a way that would grant them official taxonomic and nomenclatural siatus. Accordingly, these names should not be used for any living plants, cultivated or as they occur in natural wild habitats. The name “C. pringlei,” however, does exist in the literature, but only as a nomen nudum, having been published without any of the information required by the /nternatzonal Code of Botanical Nomenclature (Greuter, et al. 1988) to establish and recognize the name. Where, then, does the name “pringlez” come from? Cyrus Guernsey Pringle (1838-1911), a Vermonter, was an extraordinarily prolific and distinguished botanist and plant explorer. His explorations and collecting activity were con- centrated primarily in México between 1885 and 1909, when he amassed nearly 16,000 herbarium collection numbers in multicate. By 1936, his collections were the basis of approximately 29 new genera, 1,200 new species and over 100 new varieties (varietates in the botanical sense) of plants indigenous to México. There is, indeed, adequate evidence that C.G. Pringle collected spec- imens of the unique Mexican Cornus, allied to C. florida, in the Sierra Madre Orientale above Monterrey, México, Nuevo Leon State. It was from this lo- cation, so many years later, that Frederick G. Meyer (retired taxonomist of the U.S. National Arboretum), Lynn R. Lowrey (presently at Anderson Land- scape Services, Houston, Texas), and perhaps others, collected germplasm and introduced it into cultivation in the United States. In an edited compilation of Dr. Pringle’s field notes and diaries of his Mexican trips (Davis 1936), there occur several references to “C. pringlez Rose n. sp.” On page 61 of the edited diaries, Pringle reported that on June 15, 1889 “I walked to the canon of the Sierra Madre 10 miles to the southeast, collecting ...High in the canon | also found Cornus pringlet Rose n. sp., a 172 PB Yo OE. OG. TA volume 73(3):169-179 September 1992 b) tree eight to ten inches by 25 feet ....” Pringle did not provide in his diary an herbarium specimen collecting number for C. pringlez, as he did for other genera and species collected that particular day or as he did throughout his 24 - year diary. Later, from a site very close to the first collection on June 15 (or even from the same tree?), on August 23, 1889, Pringle wrote “Descending from the mountain we collect on the way Cornus pringle: Rose n. sp., a tree a foot in diameter and 25 feet in height....” Once again it is noted that the field collecting number is not cited. Then, in the “Classified List” (Davis 1936, p. 417) of Pringle’s collections “Cornus pringlez: Rose n. sp.” is designated as Pringle collection number 2409, deposited only (authors’ emphasis) in the Pringle Herbarium (University of Vermont, Burlington; VT). Cornus florida L. is also listed in the Davis (1936) “Classified List” as Pringle collection number 2409. In Davis’s “Numerical List” of Pringle’s collections Pringle No. 2409 is cited twice: first as Cornus florida L., without an herbarium of deposit indicated; and then secondly as “Cornus pringler Rose n. sp. P.,” the “P.” indicating the Pringle Herbarium as the only institution of deposit. Who, then, was this “Rose” indicated by Pringle in his 1889 Mexican diary: “Cornus pringle: Rose n. sp.?” Joseph Nelson Rose (1862-1928) was an emi- nent, and somewhat controversial, botanist at the Smithsonian Institution in Washington, D.C. Dr. Rose was, in fact, responsible, as the publishing author- ity, for a large number of Pringle’s collections being recognized and published as new genera and new species. It can only now be deduced that Pringle’s opinion about his collection No. 2409 vacillated between being C. florida and a new species, which Pringle himself designated as “C. pringlez,” assuming and anticipating that Dr. Rose would name it after him, the original collector. However, Dr. Rose did not ever describe, name or publish the Mexican flow- ering dogwood after Dr. Pringle. This may well be because Dr. Rose possibly never saw or studied Pringle’s collection No. 2409 from the Sierra Madre near Monterrey, México. Or, if Rose did see the Pringle 2409 collection, he may have decided not to describe it as a new species as no inflorescence bracts were present, and only very immature fruit were evident. This fact is well proven by the chapter “Pringle’s Plants in Type Herbarium at the National Herbarium” (the National Herbarium is in the Smithsonian Institution De- partment of Botany; US) in Davis (1936). In May, 1992 the first author of this paper talked with Dr. David Barrington, Curator of the Pringle Herbarium in Burlington, Vermont and discovered that, indeed, a specimen of Pringle No. 2409 is deposited there (VT). The typical printed label of this collection states: “Cornus pringle: Rose n. sp.,” and indicates the locality as México, Nuevo Leon State, Canon Sierras, near Monterrey. This Pringle 2409 speci- men is provided with an interesting annotation by the late L.A. Charette, a former curator of that herbarium in the 1960’s, that indicated that “Cornus pringle: Rose” is a synonym of Cornus florida subsp. urbiniana (Rose) Rick- ett, Anales Inst. Biol. Univ. Nac. México 21:92, 1950 (Rickett 1950). Charette Dudley & Santamour: Cornus florida subsp. urbiniana 173 had apparently overlooked the 1945 validating publication of that combination, also by Rickett (1945). The first author of this paper discovered a duplicate specimen of Pringle 2409 collected on 23 August 1889, labeled C. florida, in the Gray Herbarium (GH) of the Harvard University Herbaria. This specimen had no inflorescence bracts, only developing fruit. Although Dr. Rose did not name the Mexican flowering dogwood in the late 19th or early 20th centuries as “Cornus pringlet,” he did finally describe, name, and publish the taxon as Cornus urbiniana Rose sp. nov. in 1903 (Rose 1903), naming the new species for the collector “my good friend Dr. (Manuel) Urbina, acting director of the National Museum of México, by whom it was collected, and to whom I am under many obligations.” He distinguished this new species from C. florida by its “much larger and comparatively narrow (involucral) bracts,” pinkish, 5 cm long, 2 cm or more wide, tapering at the base. The type specimen chosen by Rose was collected by Manuel Urbinain April, 1891 in Vera Cruz State on Cerro de San Cristébal near Orizaba, México, and the holotype was deposited in the type collection of the U.S. National Herbarium (US) at the Smithsonian Institution, Washington, D.C. A duplicate specimen (isotype) of this Urbina collection was deposited in the Herbario Nacional, México City; MEXU. It should be noted that this infraspecific taxon of C. florida apparently is represented in México by at least two separate populations: one in Nuevo Leon State, and the other in Vera Cruz State. Recent explorations indicate that it also occurs in Tamaulipas State (Fairey, pers. comm.). Cornus urbiniana Rose was reduced in rank from species status to C. florida var. urbintana by Wangerin (1910). Wangerin (1910) described his C. florida var. urbiniana, called “Corona de Montezuma” and “Corona de San Pedro” locally in México, as distinct from typical North American C. florida by hav- ing oblong or narrowly obovate, “subchartaceous” involucral bracts with very short acuminate apices that are not at all cordate incised; these bracts are not spreading after anthesis but remain suberect and arcuate, convergent at the apices. In the literature of this Mexican taxon Wangerin (1910) is, then, the only authority prior to Lowrey (1990) to mention the unusual and novel character of the involucral bracts remaining erect throughout flowering with their apices “convergent:” this botanical term also meaning connivent (com- ing together) and adherent. However, under no circumstances are these bract apices fused or connate in any way as stated by Lowrey (1990). Temperate México is given by Wangerin (1910) as the natural distribution of C. florida var. urbinzana, and he cited three herbarium specimens, two from Nuevo Leon State, and one from Vera Cruz State. In a more recent taxonomic treatment of Mexican Cornus, Rickett (1945) regarded C. urbiniana Rose as the basis and basionym for his C. florida subsp. urbiniana (Rose) Rickett, comb. nov., with C. florida var. urbiniana (Rose) Wangerin as a synonym. Five years later, Rickett (1950) again published the same new combination of subsp. urbiniana (Rose) Rickett under C. florida. 174 PH Y“eO iP O"Gr a volume 73(3):169-179 September 1992 However, the earlier Rickett article of 1945 has priority as the validating pub- lication. In publishing the new subspecies combination Rickett (1945) men- tioned the narrow, acute involucral bracts, the smaller number of flowers per inflorescence, and the reduced number of fruits per inflorescence cluster as minor, yet distinguishing, deviations from the “normal” C. florida. However, Rickett (1945) appeared to be unaware of the peculiar “fusion” or convergence, or adherence (to be explained below) of the involucral bract apices described and photographically illustrated by Lowrey (1990). It is noteworthy that the “fusion” (apical coherence) of the involucral bracts of subsp. urbiniana is characteristic of all of the native plants observed by Lowrey (pers. comm.) in Nuevo Leon State as well as in plants cultivated in widely separated localities in the United States, grown from seed originally collected in Nuevo Leon and Vera Cruz States. These cultivated plants in- clude all of those grown in Houston, Texas (Lowrey, pers. comm.), a plant (NA 57778) at the U.S. National Arboretum, Washington, D.C. (from Lowrey seed), a plant at the University of Washington Arboretum in Seattle from seed collected by F.G. Meyer on August 23, 1948 “along arroyo 6 mi. from Dulces Nombres in the Sierra Madre Oriental, Nuevo Leon State,” and another plant (access. no. 865-48) growing at the University of Washington Arboretum in Seattle reported as having been grown from seed collected in Vera Cruz State. It has been established that “Cornus pringlez” and “C. florida var. pringlev” are names without any official taxonomic and nomenclatural status, and that C. florida subsp. (or var.) urbiniana is the correct name for this plant. The question remains as to whether it should be named at the rank of variety (var.), or at the rank of subspecies (subsp.). Rickett (1945) based his catego- tization of urbiniana at the rank of subspecies on the fact of the geographic isolation of the Mexican population which is disjunct from the main body of the natural distribution range of C. florida var. florida, and the anomalous nature of the concepts and definitions of the term and rank of variety (vari- etas, pl. varvetates) in the botanical literature of the time. For many years American and European botanists argued extensively and excessively about the terms “variety” and “subspecies,” with the Americans, for the most part, resisting the term and botanical rank of “subspecies.” It was probably Davis & Heywood (1965) who established criteria for these ranks on an international basis. The rank of subspecies is used in that book for geographically (and/or ecologically) isolated entities or populations which demonstrate some (at least three) correlated, distinct, and definable morphological character discontinu- ities. The term or rank of “variety” should be reserved for botanical taxa that may occur sporadically and randomly throughout a natural distribution of a species, and possess one to three definable morphological character differences. In the case of this Mexican flowering dogwood, which could be called “Urbina’s dogwood,” with unusual and consistent differential morphological characters that are consistently correlated with the natural geographic disjunction, the Dudley & Santamour: Cornus florida subsp. urbiniana 175 use of the rank of subspecies (subsp.) is well justified. There is the possibility, as yet not investigated, that in addition to a definite geographical disjunction Cornus florida subsp. urbiniana may be ecologically isolated to some extent, in that its natural habitat is in fairly dry limestone mountains. Lowrey (1990) commented that, because of this ecological factor, the plant “may adapt better to San Antonio, Texas and the Edwards Plateau than C. florida.” However, based on the evidence of the thriving flowering tree of C. florida subsp. urbiniana at the U.S. National Arboretum, Washington, D.C. (USDA Hardiness Zone 7), this taxon successfully tolerates acidic and heavy clay soils. As mentioned above, the involucral bracts of Cornus florida subsp. urbint- ana from México are not “fused” as stated by Lowrey (1990). The tips of the bracts are, however, strongly adherent (or connivent, or coherent) from early development of the inflorescences and expansion of the involucral bracts (April in Washington, D.C.) through anthesis (the pollen shedding period) and present the “Chinese lantern” configuration described by Lowrey (1990). The following observations and measurements of C. florida subsp. urbinzana are based on a single, flowering tree growing at the U.S. National Arboretum (NA 57778), and on wild-collected and cultivated herbarium specimens de- posited in the Herbarium of the National Arboretum (NA), studied in 1992. Comparisons with “normal” C. florida subsp. florida, when mentioned, are based on the cultivars ‘Barton’ and ‘Cloud 9,’ two popular and widely grown nursery selections. Careful examination of the tips of the bracts of subsp. urbiniana, under magnification, revealed the reason for this apical coherence or adherence. Ventrally, the bract apices are covered with a very dense white indumentum composed of slender, sinuous, and undulating hairs. Throughout the development of the inflorescence and bracts in the previous year the hairs of all bract apices are very tightly intermeshed and interlocked. The pressure exerted by the expanding narrow bracts of C. florida subsp. urbznzana is not sufficient to pull the hairs, and subsequently the bracts, apart. Conversely, the pressure exerted by the expanding, and much larger and wider bracts of C. florida subsp. florida is sufficient, for the most part, to separate their bract apices. Occasionally, the involucral bracts of some inflorescences of subsp. urbiniana do come apart naturally, but they remain “twisted” and are more or less erect, never flattened, horizontal, deflexed, or sometimes “floppy” as they are for subsp. florida. Coherence or adherence of bract tips has also been noted in plants of C. florida subsp. florida from many United States provenances when forced into flower, earlier than normal, in a polyhouse or greenhouse. When these plants flowered outdoors, the adherent bract trait dis- appeared. Mr. Fred Galle, former Director of Horticulture, Callaway Gardens, Pine Mountain, Georgia, has related (pers. comm.) that over 40 years ago he observed a number of 4 inch caliper trees of Cornus florida subsp. florida in private gardens in Knoxville, Tennessee that consistently displayed apically 176 Po ¥ ThOeLsOnGriA volume 73(3):169-179 September 1992 “fused” involucral bracts. The Galle observation of coherent inflorescence bract apices on Cornus florida subsp. florida is documented by an herbarium specimen in the Gray Herbarium (GH) of the Harvard University Herbaria: Tennessee, Knoxville, lawn on Terrace Avenue, W.W. Wyatt 17289, 27 April 1953 (GH). Another bract anomaly on C. florida subsp. florida demonstrated by Arnold Arbore- tum Accession 22791 (no longer living) of C. florida is documented by two herbarium specimens collected by Alfred Rehder: one on 11 May 1944; the other on 18 May 1948 (both A). This accession which was growing near the Jamaica Plain gate of the Arnold Arboretum had four bracted inflorescences, of which only two opposing ones were upright and incurved at their apices, and only a few of these erect narrow bracts were coherent at their apices. Rehder noted that this accession did not show any reduction of the number of flowers in the inflorescence. It is obvious, and perhaps unfortunate, that the “Chinese lantern” feature (Lowrey 1990) caused by the coherence of the involucral bract tips is not a consistent diagnostic character displayed only by Cornus florida subsp. urbini- ana. It can be said, however, that apparently the coherent bract apices are the general rule, rather than the exception, for Cornus florida subsp. urbiniana; whereas coherent bract tips for C. florida subsp. florida are the exception, rather than the general rule. The involucral bracts of Cornus florida subsp. urbiniana are much narrower in relation to width when compared to the “normal” bracts of “typical” C. florida subsp. florida. For subsp. urbiniana, they average 5.2 cm long and 2.2 cm wide at anthesis; whereas those of subsp. florida are two to three times wider. While in the “Chinese lantern” configuration the involucral bracts of subsp. urbinzana are strongly keeled (folded) ventrally, and even if they separate naturally, often by wind action, they remain keeled. The average number of flowers per each inflorescence was 17 for subsp. urbiniana and 27 for the two cultivars of subsp. florida. The percentage of aborted pollen was higher (ca. 20%) for subsp. urbiniana than normal (ca. 1%). The diameter of stained pollen grains of subsp. urbznzana was 40 yp, slightly and insignificantly smaller than normal (43 «). Some meiotic abnormalities were suspected to occur in subsp. urbinzana, where 2-3% of the stained pollen ranged from 52.8 pu to 62.4 p. These large, apparently viable, pollen grains probably had more than the normal gametic complement (n = 11) of chromosomes, and could give rise to progeny with more than the diploid (2n = 22) number of chromosomes. A number of herbarium specimens deposited in the U.S. National Arbore- tum Herbarium (NA) were also studied, and are cited here for purposes of documentation. All of the “flowering” specimens of subsp. urbinzana cited be- low demonstrate adherent or coherent bract apices. However, many of these coherent bract apices separate with senescence, a fact also observed on the living material growing of the U.S. National Arboretum. Dudley & Santamour: Cornus florida subsp. urbiniana 177 MEXICO. Nuevo Leon State: Municipio de Villa Trinidad to Potrero Redondo, tree 1 ft x 40 ft., abundant along arroyo banks in dense pine-oak forest, C.H. Mueller 2945, August 23, 1939 (also examined at GH); ibid., 11] mi. w. from El Cercado and junction with México highway 85 and 7 mi. from Horsetail Falls in the high Sierra Orientale, elev. 5300 ft., small tree, 7-8 m tall, d.b.h. 2-3 dm‘, single trunks, pine-oak forest, common locally along forest edge and lightly within, n. facing slope, fruits greenish with slight pink tinge, W. Hess & G. Wilhelm 4354, June 12, 1978. CULTIVATED. UNITED STATES. Washington: University of Washing- ton Arboretum, Seattle, deciduous small tree, 25 ft. tall, with cream white bracts, grown from seeds collected in México, Nuevo Leon State, along arroyo 6 mi. east of Dulces Nombres, Sierra Madre Oriental, coll. F.G. Meyer, August 23, 1948, Joseph Witt, May 31, 1966 (also examined at A); ibid., Arboretum Accession No. 865-48, from Missouri Botanic Garden, St. Louis, seed collected near Vera Cruz, México, Herbarium Committee Arboretum Foundation Unit Council, June 6, 1975. District of Columbia: Washington, U.S. National Ar- boretum, NA Accession No. 57778, received as a 10 ft. plant from Dr. Allen G. Hirsh, American Red Cross, Bethesda, Maryland on June 3, 1986, “the C. florida collected by Lynn Lowrey, North Star Nursery, TX at Sierra Madre Oriental, Monterrey, México,” E.J. Garvey, February 15, 1992; ibid., E.J. Garvey, April 13, 1992; ibid., tree single trunked at base, ca. 15 ft. tall, base diam. ca. 4 inches, trunk branches into 2 main stems 2 ft. from ground level, each stem ca. 2 inches in diam.; bark smooth, light brownish grey; leaves mod- erate yellowish green and somewhat lustrous above, light bluish green, dull and pubescent below; petioles yellowish green; flowers mostly yellowish green with light bluish green sepals; involucral bracts white, mostly adherent at tips, with age some bract tips separating, A. Fournzer, May 12, 1992. Additional herbarium specimens of Cornus florida subsp. urbiniana were studied at the Harvard University Herbaria, and are important to cite here. MEXICO. Nuevo Leon State: Sierra Madre Oriental, northside of ridge above Puertos El Ceroado, 30 south of Monterrey, abundant in Arroyo, C.H. & M.T. Mueller, 13 May 1934 (A); ibid., Municipio de Villa Santiago, Potrero Redondo west to Puerto a Lacuna Sanchez and beyond, abundant in canon beyond west Puerto in dense oak woods, small tree up to 8” x 30’, C.H. Mueller 2115, 5 July 1935 (A - orig. det. as C. disciflora DC. var. floccosa [Wang.] Standley); ibid., mountains near Monterry, C.H. & M.T. Mueller, July 1933 (A - orig. det. C. disciflora DC.); ibid., below Alamar, about 15 m SW of Galeana, scattered in densely wooded waterway on canyon floor, C.H. & M.T. Mueller 1141, 21 July 1934 (A); ibid., Dulces Nombres and east to the border with Tamaulipas, tree to 40 ft. tall, bark corrugated, in dense oak- pine woods, near stream course, 1550 meters alt., F.G. Meyer & D.J. Rogers 2602, 20 June 1948 (GH); ibid., canyons of Sierra Madres near Monterey., C.G. Pringle 2409, 23 August, 1889 (A). Vera Cruz State: Cerro de San Cristobal 178 PHY TOL 0iG 7 A volume 73(3):169-179 September 1992 near Orizaba locus classicus, moist slope, 5000 ft, A.J. Sharp 4618, 8 March 1946 (GH); ibid., 2/3 way up Cerro de San Cristébal, 1/2 m south of Orizaba- locus classicus, elev. about 5000 ft., Wayne E. & Margaret S. Manning 53768, 4 August 1953 (GH); ibid., Huatusco, C.A. Purpus 8933, March 1921 (GH). Lynn Lowrey of Houston, Texas also arranged for us to examine a fruiting ‘specimen of Cornus florida subsp. urbiniana deposited in the herbarium of the Robert.A. Vines Environmental Science Center in Houston, Texas: MEXICO. Nuevo Leon, Monterrey, Chipinque, C.D. Peterson 1244, 24 July 1988 (SBSC). Professor John G. Fairey of the Department of Architecture, Texas A&M University, College Station, Texas recently reported (pers. comm.) that in the autumn of 1990 he discovered a large number of 60 foot tall trees of this Mexican flowering dogwood in a cloud forest area, 7000 feet elevation, at “E] Butano,” a very remote and seldom visited giant sinkhole 50 miles southwest of Monterrey, Nuevo Leon State, northeast México. These trees were grow- ing among Jlez rubra S. Watson and J. discolor Hemsley (also 60 feet tall), Tazus globosa Schlechter, Picea martinezu T.F. Patterson (a synonym of P. chthuahuana Martinez), and an evergreen Quercus. Another large population of Cornus florida subsp. urbiniana is reported by Professor Fairey (pers. comm.), and consists of thousands of trees in Nuevo Leon State near La Trinidad at 4500 feet elevation. These trees have extremely large inflorescence bracts, nearly twice as large as those of typical C. floridain the United States, yet all are coherent at their apices. The fruits of these trees are also reported as being exceptionally large. Professor Fairey has also dis- covered (pers. comm.) scattered trees with “soft pink” inflorescence bracts in a remote locality approximately 60 miles NW of Ciudad Victoria in Tamaulipas State. In summary, the correct botanical name for the Mexican flowering dogwood now being grown and promoted as “Cornus florida var. pringlez” is C. florida subsp. urbinzana (Rose) Rickett (1945). Whether it is logical to continue to call this plant the “Pringle dogwood” when it is really named in honor of Dr. Manuel Urbina is an issue that does not need to be resolved here. Cornus florida L. subsp. urbzniana (Rose) Rickett, Bull. Torrey Bot. Club 72:223. 1945. BASIONYM: Cornus urbiniana Rose, Contrib. U.S. Nat. Herb. 8:53. 1903. SYNONYMS: Cornus pringlez Pringle ez Davis, Life and Work of Cyrus Guernsey Pringle, pp. 61,65,417,505,721. 1936 - nomen nudum et illegitamum. Cornus florida L. var. pringlei (Pringle ez Davis) Lowrey, American Nurseryman 172 (6):142. 1990 - nomen nudum et combinationes illegitimum. Cornus florida L. var. urbiniana (Rose) Wangerin, Das Pflanzenreich (ed. A. Engler) Heft 41. IV. 229:87. 1910. Benthamidta florida (L.) Dudley & Santamour: Cornus florida subsp. urbiniana 179 Moldenke var. urbinzana (Rose) Moldenke, Revista Sudamer. Bot. 6:177. 1940. This identical combination was made by Hiroshi Hara eight vears after Moldenke (cf. Hara 1948). LITERATURE CITED Davis, H.B. 1936. Life and Work of Cyrus Guernsey Pringle. University of Vermont, Burlington, Vermont. Davis, P.H. & V.H. Heywood. 1965. Principles of Angiosperm Tazonomy. Oliver & Boyd, Edinburgh and London, U.K. Greuter, W. (Ch. Ed. Comm.) 1988. International Code of Botanical Nomen- clature. Adopted by the Fourteenth International Botanical Congress, July-August 1987. Regnum Veg. vol. 118. Hara, H. 1948. The nomenclature of the flowering dogwood and its allies. Jour. Arnold Arb. 29:111-115. Lowrey, L.R. 1990. Cornus florida var. pringlez. American Nurseryman 176(6):142. Moldenke, H.N. 1940. Some new names in the Apocynaceae and Cornaceae and in various American groups. Revista Sudamer. Bot. 6:176-178. Rickett, H.W. 1945. New combinations in Cornus. Bull. Torrey Bot. Club iae223: Rickett, H.W. 1950. Cornus in Mexico with notes on the evolution of the genus. Anales Inst. Biol. Univ. Nac. Mexico 21:83-94. Rose, J.N. 1903. Studies of Mexican and Central American plants - No.3. Cornaceae. The Mexican species of Cornus. Contrib. U.S. Nat. Herb. 8(1):53-55. Santamour, F.S., Jr., & T.R. Dudley. 1992. A taxonomic and cytogenetic summary of the genus Cornus. Proc. Sixth Regional Dogwood Workshop (April 14-16, 1992, Pipestem, WV), p. 8-13. Wangerin, W. 1910. Cornaceae. Das Pflanzenreich (Ed. A. Engler), Heft 41. IV. 229:1-110. Phytologia (September 1992) 73(3):180-182. DETERMINATIONS OF CHROMOSOME NUMBER (2N) AND ENDOSPERM BALANCE NUMBER (EBN) IN SOME LITTLE KNOWN TUBER BEARING SOLANUM C.M. Ochoa International Potato Center, Genetic Resources Department, P.O. Box 5969, Lima, PERU ABSTRACT Diploid chromosome numbers and endosperm balance numbers are reported for taxa of tuber bearing Solanum. KEY WORDS: Solanaceae, Solanum, chromosome number, cytology Solanum acroglosum Juz. (Bull. Acad. Sci. U.R.S.S. 3:313-314. 1937.): 2n=24: EBN = 2. Solanum albornozi Corr. (Wrightia 2:178. 1961.): 2n=24; EBN = 2. Solanum amayanum Ochoa (Amer. Pot. Jour. 66:1-4. 1989.): 2n=24; EBN =e Solanum anamatophilum Ochoa (Ann. Cient. Univ. Agr., Lima. 2[4]:391-395. 1964): 2n=24; EBN = 2. Solanum antacochense Ochoa (Amer. Pot. Jour. 58[3]:127-129. 1981.): 2n=24; EBN = 2. Solanum ayacuchense Ochoa (Agronomia, Lima. 26:312-313. 1959.): 2n=24; EBN = 2. Solanum blanco-galdosii Ochoa (Ann. Cient. Univ. Agr., Lima. 11{3-4]:157- 160° 19735); 2n—24- E BN = 2: Solanum buesii Vargas (Rev. Arg. Agron. Buenos Aires. 10:396-397. 1943.): 2n—24; EBN = 2. 180 Ochoa: Chromosome and endosperm balance numbers in Solanum 181 Solanum candolleanum Berth. (Ann. Sci. Agron., Paris, ser. 3. 2:185. 1911.): 2n—24-/EBN = 2; Solanum cantense Ochoa (Agronomia, Lima. 26:217-218. 1959.): 2n=24; EBN = 2. Solanum contumazaense Ochoa (Ann. Cient. Univ. Agr., Lima. 2{2]:148-151. 1964.): 2n=24; EBN = 2. Solanum chilliasense Bitt. (Lorentzia, Argentina) 4:9-11. 1981.: 2n=24; EBN es Solanum chiquidenum Ochoa (Biota, Lima. 1:5-7. 1954.): 2n=24; EBN = 2. Solanum dolichocremastrum Bitt. (Fedde Repert. Sp. Nov. 12:3-4. 1913.): 2n—24~ BN = T° Solanum guzmanguense Whalen & Sagast. (in Whalen, M.D., Sagastegui A., & Knapp, S. Brittonia 38/1]:9-12. 1986.): 2n=24; EBN = 1. Solanum humectophilum Ochoa (Darwiniana 15({3-4]:550-553. 1969.): 2n=24; PBN — Solanum hypacrarthrum Bitt. (Fedde Repert. Sp. Nov. 11:367-368. 1912.): 2n—24; EBN = 1. Solanum immite Dun. (in A.DC., Prodr. 13(1):32. 1852.): 2n=24; EBN = 1. Solanum ingaefolium Ochoa (Agronomia, Lima. 26:319. 1959.): 2n=24; EBN = 1. Solanum irosinum Ochoa (Amer. Pot. Jour. 58(3]:31-33. 1981.): 2n=24; EBN =e Solanum jalcae Ochoa (Agronomia, Lima. 19:167. 1954.): 2n=24; EBN = 2. Solanum litusinum Ochoa (Phytologia 48(3]:229-232. 1981.): 2n=24; EBN = Ze Solanum lobbianum Bitt. (Fedde Repert. Sp. Nov. 12:446-447. 1913.): 2n=48; EBN = 2. Solanum minuttfoliolum Corr. (Wrightia 2:191. 1961.): 2n=24; EBN = 1. Solanum nemorosum Ochoa (Amer. Pot. Jour. 60/6]:389-392. 1983.): 2n=72; EBN = 4. Solanum neovavilovii Ochoa (Amer. Pot. Jour. 60[11]:919-923. 1983.): 2n=24; EBN = 2. 182 PY IPO re OrGr kA volume 73(3):180-182 September 1992 Solanum nubicola Ochoa (Ann. Cient. Univ. Agr., Lima. 8/3-4}:143-146. 19702) 2n—=48: BBN =: Solanum olmosense Ochoa (Ann. Cient. Univ. Agr.. Lima. 3/1|:33-37. 1965.): on 24 Bele Solanum orophilum Corr. (Wrightia 2:192-193. 1961.): 2n=24; EBN = 2. Solanum pampasense Hawkes (Bull. Imp. Bur. Pl. Breed. & Genet. Cam- bridge. 125. 1944.): 2n=24; EBN = 2. Solanum paramoense Bitt. (ex Pittier, Man. Pl. Usual. Venez. 329. 1926.): 2n=48; EBN = 4. Solanum peloquinzanum Ochoa (Amer. Pot. Jour. 57{1|:33-35. 1980.): 2n=24; BE Nee. Solanum pillahuatense Vargas (Las Papas Sudperuanas, Univ. Ncl. del Cusco. 2:53-54. 1956.): 2n=24; EBN = 2. Solanum piurae Bitt. (Beibl. Bot. Jahrb., No. 119. 54:5-6. 1916.): 2n=24; EBN = 2. Solanum quillonanum Ochoa (Phytologia 67(3]:235. 1989.): 2n=24; EBN = 75 Solanum raquzalatum Ochoa (Agronomia, Lima. 19:172-174. 1954.): 2n=24; PBN Solanum suffrutescens Corr. (Wrightia 2:183-184. 1961.): 2n=24; EBN = 2. Solanum tacnaense Ochoa (Agronomia, Lima. 18[74]:133,135-136. 1953.): 2n=24- EE BN — 2. Solanum taulisense Ochoa (Ann. Cient. Univ. Agr., Lima. 1{3]:216-220. [9G3..):)2n—24-. BIN i= 2. Solanum trinitense Ochoa (Ann. Cient. Univ. Agr., Lima. 2[3]:245-247. 1964.): 2n=24; EBN = 1. Solanum urubambae Juz. (Bull. Acad. Sci. U.R.S.S. 2:312-313. 1937.): 2n=24; EBN— 2. Solanum villuspetalum Vargas (Las Papas Sudperuanas. Univ. Ncl. del Cusco. 2:04. 1956.): 2n=24; EBN = 2. Solanum wittmackiw Bitt. (Fedde Repert. Sp. Nov. 12:546. 1913.): 2n=24; EBN = Phytologia (September 1992) 73(3):183-185. SOLANUM LOBBIANUM BITTER, A LITTLE KNOWN COLOMBIAN TUBER BEARING SPECIES C.M. Ochoa International Potato Center, Genetic Resources Department, P.O. Box 5969, Lima, PERU ABSTRACT New information and an expanded description are provided for the seldom collected Solanum lobbianum. KEY WORDS: Solanaceae, Solanum, taxonomy Solanum lobbianum was described by George Bitter in 1913 (Fedde Repert. Sp. Nov. 12:446.). The type of this wild tuber bearing potato species was collected circa 1856 by William Lobb, a British plant collector. However, Lobb failed to record his collection number, date of collection, and the exact location where collected. Its type specimen was housed in the Hookerianum Herbarium in 1867 with the simple label notation “Colombia, Lobb.” This specimen, which represents the holotype of the species, is currently housed in Kew Gardens Herbarium in London. In 1918, another collection of Solanum lobbianum was made by M.T. Dawe, under his number 742. This collection has three specimens which were dis- tributed to herbaria at Kew, The New York Botanical Garden, and the Na- tional Herbarium of the United States of North America (Smithsonian). Dawe’s specimens bear only a brief note indicating that they had been collected in “Ruiz, Colombia, 3000 m alt.” During my explorations in Colombia in July 1980-nearly 124 years after Lobb’s first collection-I found Solanum lobbianum while ascending the volcano of El Ruiz, between Las Juntas and Termales, 2800-3200 m alt., Department of Manizales. Although the original diagnosis of Solanum lobbianum is extensive and detailed, it was based on only one specimen. Here, I will give some comple- mentary data that could be useful for a better understanding of this species. Plant (15-)25-30(-40) cm tall, slightly rossulate at the base when growing in the open field under exposed conditions, densely pubescent, stoloniferous 183 184 Pol YOO CiOsG ar A volume 73(3):183-185 September 1992 1. El Ruiz Peak, near 5000 m alt.,Colombia. Frontground in the low 3200 m alt., habitat of Solanum iobbianum. 2. Flower dissection of Solanum lobbianum Ochoa: Solanum lobbianum, a little known species 185 and tuber bearing. Stem usually branched, 3-5 mm in diameter, sparsely pi- lose, narrowly winged, wings straight and barely distinguishable, internodes (1.0-)2.5-4.5(-5.5) cm long. Leaves imparipinnate, 9.5-20.0 cm long by 6.5-16.0 cm wide, 3-5 pairs of leaflets, 3-8 pairs of sessile interjected leaflets, from two to three sizes, small suborbicular to orbicular; lateral leaflets widely elliptic to elliptic lanceolate, first and second upper pair 2.5-4.5 cm long by 1.7-2.5 cm wide, 2-7 mm petiolulate, obtuse or subacuminate apex, petiolules some- times with 1-2 small interjected leaflets; the lower pair of leaflets much smaller and sessile. Pseudostipular leaves widely elliptic-lanceolate with the apex very obtuse, small, 7.0 x 5.0 mm. Inflorescence cymose, 7-9 flowered. Peduncle 3.5- 7.0 cm long, 1.5-2.0 mm in diameter toward base, pilose as the pedicels, calyx and flower buds; hairs multicellular of unequal length sparsely alternated with short glandular tetralobed hairs; pedicels 18-22 mm long, articulated slightly above the middle. Calyx 7-11 mm in diameter, lobes narrowly lanceolate, acute. Corolla very pale violet or light violet-lilac with whitish acumens, ro- tate pentagonal or pentagonal, 1.6-2.0 cm in diameter, acumens densely pilose. Filaments 1.0-1.5 mm long, glabrous; anthers widely elliptic-lanceolate, small, 4 mm long. Style 7.0 mm long, covered with little prominent papillae to- ward the one-third of the basis; stigma shortly conical-capitate, small. Ovary conical. Fruit long conical. Solanum lobbianum is a typical species of the series Conzcibaccata, with a chromosome number of 2n=48. Although S. lodbianum has some similarity with S. bukasovii Juz., and S. brevicaule, as Correll states (The Potatoe and its Wild Relatives, Texas Research Foundation. 1962.), it differs from these and from all known tuber bearing species because of its conspicuous pubescence, silvery-white hairs noticeably longer on the stem, calyx, and flower buds, pe- duncles shorter and thinner than in S. bukasovii, pedicels slender, corolla small (2.0-2.5 cm in diam.), very pale violet, fruits long conical. Contrary to S. bukasovn or S. lobbianum f. multidissectum (Hawkes) Correll, which in my opinion is only a variety of S. bukasoviz, the corolla of S. bukasovti is pentago- nal, larger (up to 4 cm or more in diameter), dark blue violet or dark purple, flower buds are not too densely pilose, nor are the hairs very long; fruit globose to ovate and its chromosome number is 2n=24. Specimens Examined: COLOMBIA. Dept. Manizales: Heights of Las Jun- tas, 2800 m alt., July 22, 1980, C. Ochoa 14109 (CIP,COL,GH). Above Ter- males, ascending towards volcano Ruiz, 3200 m alt., July 22, 1980, C. Ochoa 14110 (CIP,COL,GH). Indefinite: Lobb s.n. (HOLOTYPE: K). “Ruiz, 3000 m alt.”, 1918, M.T. Dawe 742 (K,NY,US). Phytologia (September 1992) 73(3):186-202. TAXONOMIC NOTES ON ERJGERON (ASTERACEAE: ASTEREAE) OF CALIFORNIA, NEVADA, AND ARIZONA Guy L. Nesom Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT In connection with taxonomic treatments of Erigeron from Cali- fornia, Nevada, and Arizona (submitted or in press), details are pro- vided regarding some of the taxonomic decisions, most significantly those involving E. lobatus and E. divergens, E. peregrinus, the E. eatontt complex, E. aphanactts, and the taxa centered around £. chrysopsidis. Three new combinations are proposed (E. eatonii var. nevadincola; E. eatonii var. sonnei; E. chrysopsidis var. austiniae) and com- ments are provided regarding possible sectional realignments of species centered around E. linearis. New distribution records are reported for E. lobatus (first records for Nevada and California) and E. pumilus (first records for New Mexico). KEY WORDS: Erigeron, Asteraceae, Astereae, California, Nevada, Arizona, New Mexico Before the publication of relatively condensed taxonomic treatments of Erigeron L. for the states of California (51 species, 71 taxa in total, submitted for the developing Jepson’s Manual), Arizona (submitted for the developing Flora of Arizona, 36 species, 38 taxa in total), and Nevada (33 species, 38 taxa in total, Phytologia-this issue), several associated taxonomic and nomenclat- ural problems need to be addressed in greater detail. In addition, species described since Cronquist’s revision of the North Amer- ican species (1947) are now known from Arizona (Nesom & Roth 1981; Nesom 1989b, 1990c, 1990d) and California (Nesom 1992), concepts of Arizona taxa have been modified (Nesom 1989a), and a species previously known only from Mexico has been recorded for the state (Nesom & Baker 1991). I am treating the North American species of Erigeron sect. Trimorpha (Cass.) DC. as the genus 7rimorpha Cass. (Nesom 1989c). Some species variably treated in the 186 Nesom: Erigeron in California, Nevada, and Arizona 187 past as Erigeron or Conyza] recognize as Conyza (Nesom 1990a), while others I] treat as the genus Laennecia Cass. (Nesom 1990b). Erigeron ortegae S.F. Blake (Aster sprnosus Benth.) has been segregated as the genus Chloracan- tha, with a single species (Nesom et al. 1991; Sundberg 1991). A synopsis of the infrageneric taxonomy of Erigeron in North America is available (Nesom 1989d). 1. The Erigeron foliosus Nutt. group. In his monograph of the North American species of Erigeron, Cronquist (1947) recognized six species and a total of eighteen taxa, including vari- eties, among the representatives of the E. foliosus group in the western United States. A recent treatment of this same group (Nesom 1992) recognizes four- teen species and a total of 28 taxa, including varieties. Of these, E. ozyphyllus E. Greene occurs only in Arizona and adjacent México; most of the rest of the taxa are centered in California. 2. Erigeron algidus Jeps. The name Erigeron petiolaris E. Greene (Leafl. Bot. Observ. Crit. 2:205. 1912.) generally has been used for the monocephalous, high elevation species of California and Nevada, which is closely related to E. sempler E. Greene of the cordillera of the Southern Rocky Mountains. Greene’s E. petiolaris, however, is a later homonym of E. petiolaris Vierh. (1906), a species of Siberia. The earliest available name for the North American species is E. algidus Jepson (Man. Fl. Calif. 1052. 1925.). Erigeron sumplez was reported by Cronquist (1947) to occur in Nevada, but he cited no specimens from that state; Spong- berg (1971) located no collections of E. stmplez from Nevada. Erzgeron algidus and E. simpler might be treated as varieties of a single species, but because of their morphological distinction and wide disjunction, they may be justifiably kept apart. 3. Erigeron compositus Pursh. Several varieties have commonly been recognized within this species, based on leaf morphology (most commonly var. compositus, var. discoideus A. Gray, and var. glabratus Macoun; for others, see Cronquist 1945). Recent studies of cytogeography and molecular variation within Erigeron compositus (Beaman unpublished; Noyes et al. 1987; Noyes 1988) provide evidence that these tra- ditional infraspecific categories are artificial and replace them with a detailed and presumably more natural system of classification. Until the publication of Beaman’s formal system of nomenclature for these variants, | am treating the species without any formally recognized infraspecific elements. 188 PAY TOO GIA volume 73(3):186-202 September 1992 4. Erigeron lobatus A. Nelson and E. divergens Torr. & Gray. Erigeron lobatus is a distinctive species closely related to the Arizona en- demic E. piscaticus Nesom and the widespread, primarily Mexican species E. velutipes Hook. & Arn. (Nesom 1989b). Some plants of E. lobatus, appear to be influenced by genes of E. divergens, but the two species for the most part maintain their own identity when occurring in close proximity. They are distinguished by the following contrasts. a. Stems and phyllaries minutely but densely and prominently stipitate glan- dular and very sparsely hairy with eglandular, straight, spreading hairs 0.5-2.0 mm long; basal and lower cauline leaves with deep lobes with rounded apices; phyllaries ovate-lanceolate, the margins thin hyaline; flowering December-May(-June, -September), in habitats at 1200-3800 filimeAmzona)) 5 "yAcayay. Jeisiccs aah aes Joos eae. os Reema E. lobatus a. Stems and phyllaries densely and evenly pubescent with eglandular, spread- ing, often slightly crinkly hairs 0.1-1.0 mm long, sometimes also minutely granular glandular but the glandularity not prominent; basal and lower cauline leaves entire, or if lobed the lobes shallow to deep with acute apices; phyllaries lanceolate, the margins relatively thick; flowering (Feb- ruary-)April-October, in habitats at 3500-8050 ft (in Arizona). ....... BE eae tern OPI Oe 5 Set NS W S MISS SERS Ce E. dwergens Erigeron lobatus is restricted to eastern Arizona and adjacent Sonora, Mexico, except for one collection from Nevada and one from California, each the first report of the species from its respective state, and both very near the Arizona border. (1) Nevada, Clark Co., Valley of Fire, 11 Apr 1937, Maguire 17874 (UT). This plant is slightly different from typical E. lobatus in its vestiture, mostly lacking conspicuous stipitate glands and producing a denser portion of eglandular hairs. In overall morphology, it is similar to E. lobatus, but it is perhaps genetically influenced by E. divergens and approaches the “accedens form” of E. divergens (see below). (2) California, San Bernardino Co., vicinity of Stepladder Mountains and Ward Valley, 20 mi E of Essex, 1800 ft, 14 May 1978, Faulkner 608 (TEX,UCR). This collection is of typical E. lobatus. A form of Erigeron divergens with persistent,.deeply pinnatifid basal leaves with long petioles, elongated lower internodes, and a strong tendency toward perenniality has been described both as FE. accedens E. Greene and E. cali- fornicus Jepson (see citations below). The form is particularly distinctive in California, where E. divergens otherwise produces basal leaves that are mostly entire. ] have annotated a number of specimens of the atypical form as the “accedens form” of E. divergens, although Jepson’s name predates Greene’s. Nesom: Erigeron in California, Nevada, and Arizona 189 Such plants are commonly identified as F£. lobatus, and they may be of hy- brid origin between EF. divergens and LE. lobatus, but they are more similar in vestiture to E. divergens. They are found not only in the area where both pu- tative parental species occur (primarily Arizona, where the “accedens form” is more abundant than elsewhere) but outside of it as well (central and southern California to Arizona, and much less commonly in New Mexico, trans-Pecos Texas, northern Sonora, and Chihuahua), and it seems likely that such plants have had a single origin. The plants of the “accedens form” for which chromo- some counts are available all are triploid (and probably agamospermic) (from Arizona: Keil & Pinkava 1976; Pinkava & Keil 1977; Solbrig et al. 1969; from Texas and New Mexico, Turner & Zhao in prep.), although other more typical forms of E. divergens have also been reported as triploid. The “accedens form” is broadly sympatric with more typical E. divergens and is one of several mor- photypes in the E. dzvergens complex that probably deserve formal taxonomic recognition at some rank. Erigeron accedens E. Greene, Pittonia 4:155. 1900. TYPE: UNITED STATES. Arizona: Greenlee Co., Clifton, Apr 1899, Dr. A. Davidson s.n. (HOLO- TYPE: ND-G!). Erigeron californicus Jepson, Bull. Torr. Bot. Club 18:324. 1891. TYPE: UNITED STATES. California: Sutter Cc., Marysville Buttes, [edge of the] summit of South Peak, 20 Apr 1891, W.L. Jepson s.n. (HOLO- TYPE: JEPS!; Isotype: JEPS!). 5. Erigeron multiceps E. Greene. The plants of the type collection (see below) are distinctively perennial, with thick, branching caudices, and the stem pubescence is upwardly ascending- appressed. Otherwise, in both floral and vegetative features, as noted by Cron- quist (1947), the plants of Erigeron multiceps are very similar to the most common form of E. divergens in California. A number of collections of rela- tively typical E. divergens have been made in the immediate area of the type locality of E. multiceps, and if the latter is indeed distinct from E. divergens, it seems that it must be a rare species. Plants of Erigeron divergens with appressed stem pubescence, though un- common, have been collected from the southern part of its range (e.g., Esmer- alda and Clark cos., Nevada; Cochise Co., Arizona). Similar variation occurs in other species of Erigeron, and a rarely occurring shift in the orientation of stem vestiture provides weak justification for segregating taxa, particularly as E. divergens is variable in other features as well. Field work is needed to clarify the identity of E. multiceps and its relationship to E. divergens, but until such is available, I have maintained the former as a separate species. 190 PB Y.T O4.OG TA volume 73(3):186-202 September 1992 Erigeron multiceps E. Greene, Pittonia 2:167. 1891. TYPE: UNITED STATES. California: [Kern Co.], gravelly spots near river banks, N. Fork of the Kern River, 7-15 Jun 1888, E. Palmer |and Wright] 121 (LECTOTYPE (designated here]: ND-G!; Isolectotypes: NY!,US!,US photo-CAS!). 6. Erigeron peregrinus (Banks ez Pursh) E. Greene. Cronquist (1947) regarded Erigeron peregrinus subsp. callianthemus (E. Greene) Cronq. as the southern segment of a widely distributed and variable species of western North America, with the typical subspecies occurring from the Aleutian Islands through Alaska and western Canada into Washington of the United States. Within subsp. callianthemus, var. hirsutus Cronq. of west central California and immediately adjacent Nevada appears to be weakly but justifiably maintained as a variety, but var. scaposus (Torr. & Gray) Cronq. and var. angustifolius (A. Gray) Crong. can be recognized apart from the typical variety (var. callianthemus [E. Greene] Cronq.) only as arbitrarily dis- tinguished and intergrading populations. These may be ecotypically differen- tiated in some areas, but they do not appear to represent coherent taxa with geographic integrity. Cronquist (1947) noted that var. scaposus occurs “ap- proximately over the range of the subspecies,” usually in alpine habitats. He placed the range of var. angustzfolius from California and adjacent Nevada to British Columbia, also completely sympatric with the typical plants, though sometimes occurring to lower altitudes. While it does appear that plants with narrowly lanceolate to oblanceolate leaves, usually with acute apices, are most common in the Pacific states, they do not form a well differentiated group of populations that can justifiably be given formal taxonomic status. A study of variation within the two subspecies (sensu Cronquist) of E. peregrinus is underway (Nesom in prep.). 7. The Erigeron eatonu A. Gray complex. In their study of Erigeron eatonu and allied taxa, Strother & Ferlatte (1988, p. 77) noted that all of the taxa treated by them “variously intergrade morphologically and may constitute a single, polymorphic species.” Such a radical treatment would not be necessary, although I believe that two of the species recognized by them are better subsumed as varieties of FE. eatonz: E. sonnez E. Greene and E. nevadincola $.F. Blake. Two new combinations are required: Erigeron eatonii A. Gray var. sonnei (E. Greene) Nesom, comb. nov. BA- SIONYM: Erigeron sonnez E. Greene, Pittonia 1:218. 1888. Erigeron nevadensis (sic) A. Gray var. sonnez (E. Greene) Smiley, Univ. Calif. Publ. Bot. 9:373. 1921. TYPE: UNITED STATES. California: Nevada Nesom: Erigeron in California, Nevada, and Arizona 191 and Placer counties, W slope Washoe Mts., 22 Jul 1888, C.F. Sonne 2 (HOLOTYPE: ND-G!). See Strother & Ferlatte (1988) for comments on the typification. Erigeron eatonii A. Gray var. nevadincola (S.F. Blake) Nesom, comb. et stat. nov. BASIONYM: Erzgeron nevadincola S.F. Blake, nom. nov., Proc. Biol. Soc. Wash. 35:78. 1922. Based on: Erigeron nevadensis A. Gray, Proc. Amer. Acad. Arts 8:649. 1873, non Wedd. 1857. TYPE: UNITED STATES. Nevada: Storey Co., Cedar Hill and Mt. Davidson, near Virginia City, 1863-64, H.G. Bloomer s.n. (LECTOTYPE [Cron- quist 1947]: GH!). Cronquist also (pers. comm.) observed a close similarity between Erigeron sonnei and £. eatonz var. plantagineus (E. Greene) Cronq. In his 1947 study, he treated E. sonnet only as a synonym of var. plantagineus. In the key and comments of Strother & Ferlatte (1988), E. sonnei and var. plantagineus are separated essentially only by a greater, though overlapping, number of pappus bristles per achene in the former, and intermediates between the two occur at least in Sierra Co., California. With a number of specimens of both taxa at hand, var. sonnei can be seen as nearly intermediate in head size between typical var. nevadincola and var. plantagineus. My treatment of EL. nevadincola, in turn, is based on its intergradation with E. sonnei. Var. nevadincola is distinctive among the varieties of Erigeron eatonu in its large heads and corollas. Erigeron sonnei also is relatively large headed, compared to the rest of E. eatonw. An unpublished study of mine of the E. nevadincola-E. sonnei complex in Nevada (on file in the E. eatonz reprint folder at TEX) indicates that E. sonnez and E. nevadincola are morphologically intergrading and can be separated only arbitrarily, at least in the central part of Nevada. The characters studied were leaf length and width, head width, phyllary length, disc corolla length, style appendage length and shape, and pappus bristle number (my counts of pappus bristles include numerous plants of E. sonnei and FE. nevadincola with up to 35 bristles per achene, in contrast to Strother & Ferlatte’s maximum of “24+” bristles). Strother & Ferlatte (1988, p. 91) themselves observed that ”These intergrading taxa [E. sonnei and £. nevadincola| are probably conspecific; we maintain them at specific rank for the present in order to preserve established nomenclature.” In fact, E. sonnei has already been treated as a variety of E. nevadincola, but the combination is invalid, as it was based on an illegitimate later homonym, EL. nevadensis A. Gray; further, if the two are considered conspecific and distinct from E. eatonii, the earliest correct name for the species is E. sonnei (1888) vs. E. nevadincola (1922). On the east and north sides of the range of Erigeron eatonit var. nevad- incola, closely related taxa are completely separated from it geographically 192 PeY TOL OLG TA volume 73(3):186-202 September 1992 (Strother & Ferlatte 1988, Figs. 1 and 2), and there is no morphological in- tergradation between var. nevadincola and any of them. The connection of var. nevadincola with the rest of E. eatoni is only through E. sonnet. Erigeron sonnei and var. nevadincola are more similar to each other than either is to var. plantagineus. Var. nevadincola and var. plantagzneus are rela- tively clearly distinguished where they approach each other along the Nevada- California border, but Strother & Ferlatte (1988) noted the occurrence of “occasional plants” (unspecified) intermediate between them in Washoe Co., Nevada. They did not record the presence of var. plantagineus in Nevada, and my own observations are accordant with their Fig. 1, which shows var. nevadincola in California only in a small area of Sierra Co. and the adjacent southernmost portion of Lassen Co. Cronquist (1947, 1955) and Munz (1959) both noted the occurrence of Erigeron eatonii var. plantagineus in Nevada, but the plants upon which these reports were based probably are those identified by Strother & Ferlatte as E. sonnet. Such observations, however, indicate the close similarity between var. plantagineus and E. sonnei. As treated here, the differences among the vari- eties of E. eatonii that occur in California and Nevada are strictly quantitative. Distinctions between other of the varieties involve size as well as variation in vestiture. ; In treating other taxa of the Erigeron eatonit group, I have followed Strother & Ferlatte (1988) in considering EF. fleruosus Cronq. merely a synonym of E. lassenianus E. Greene but note that the populations identified as the former apparently are completely disjunct from the range of typical E. lassenzanus. As noted by Strother & Ferlatte, the disjuncts tend to produce a slightly dif- ferent vestiture, and a more detailed study would be valuable in assessing whether they might deserve formal recognition at some rank. 8. Erigeron concinnus (Hook. & Arn.) Torr. & Gray. This species was regarded by Cronquist (1947) as a subspecies of Erigeron pumilus Nutt. (as subsp. concinnoides Cronq., with two varieties). A recent study (Nesom 1983) concluded that E. concinnus does not intergrade with E. pumilus, and the two are better recognized as distinct species. Study of ad- ditional collections has corroborated this, and the published distribution map (Nesom 1983, Fig. 1) remains essentially correct. The primary modification is the extension of the range of FE. pumilus (var. pumilus) into north central New Mexico (several collections from Rio Arriba Co. - NMC,TEX, and UNM), a state from which that species has not been previously reported (Martin & Hutchins 1980). Several collections of Erigeron pumilus var. intermedius Crongq. from the vicinity of Carson City, Nevada, are of thick stemmed, monocephalous, large leaved plants with highly abortive pollen. Similar plants also occur in Washoe Nesom: Erigeron in California, Nevada, and Arizona 193 Co., Nevada, and adjacent Lassen Co., California. A triploid chromosome count (2n=27 univalents, Raven 18545-TEX, Solbrig et al. 1969) is known for such a plant from adjacent Storey Co., Nevada. Other chromosome counts from FE. pumilus var. intermedius have been diploid (n=9 bivalents; Taylor & Brockman 1966; Semple 1985; Chinnappa & Chmielewski 1987; Sundberg 1468-TEX, apparently unpublished). 9. Erigeron aphanactis (A. Gray) E. Greene. This species almost certainly is most closely related to those centered around Erigeron pumilus (Nesom 1989d, 1990d). This view is based in large part on the distinctive, reflexing ligules found in all of the taxa involved. The characteristically hairy disc corollas and the prominent, scaly, outer pappus of E. aphanactts are nearly identical to those found in the radiate E. concinnus. The geographic range and relative position of E. concinnus suggest, in con- trast, that it is more closely related to E. pumilus (consisting of two varieties), all three taxa being derived from a widespread, evolutionarily fragmented, im- mediate ancestor (Nesom 1983, Fig. 1). Erigeron concinnus and EL. aphanactis have broadly sympatric, nearly congruent ranges, and no unequivocal interme- diates are known between FE. aphanactis and any other species. The following nomenclatural paragraphs summarize my view, in agreement with Cronquist’s, of the taxonomy of E. aphanactis: Erigeron aphanactis (A. Gray) E. Greene, Fl. Franc. 4:389. 1897. Erigeron concinnus (Hook. & Arn.) Torr. & Gray var. aphanactis A. Gray, Proc. Amer. Acad. Arts 6:540. 1865. TYPE: UNITED STATES. Nevada: Carson City, 1864, C.L. Anderson 205 (HOLOTYPE: GH!; Isotypes: MO! .NY!,US!). Erigeron aphanactis (A. Gray) E. Greene var. aphanactis Erigeron aphanactis (A. Gray) E. Greene var. congestus (E. Greene) Cronq., Brittonia 6:177. 1947. BASIONYM: Erigeron congestus E. Greene, Leafl. Bot. Observ. Crit. 2:218. 1912. TYPE: UNITED STATES. California: San Bernardino Co., Gold Hill, Bear Valley, 7000 ft, 2 Jun 1901, S.B. Parish 4886 (HOLOTYPE: US!; Isotype: NY!). The scapose, monocephalous plants of Erigeron aphanactts (var. congestus) occur primarily along the southeastern margin of the species (Map 1), although scattered, similar variants within populations may be found elsewhere. Similar variation in habit occurs within the closely related E. concinnus. Erigeron pumilus usually produces several heads per stem, and if it is assumed to be ancestral or phyletically coordinate with E. aphanactis and E. concinnus, the scapose habit is evolutionarily derived. 194 PAY IO iO GavA volume 73(3):186-202 September 1992 E. aphanactis @® var. aphanactis @® var. congestus a ee hae ~ Resin: c : De a 1 Map 1. Distribution of Erigeron aphanactis. Open symbols in southeastern Utah (Grand Co. and San Juan Co. ) are added from Albee et al. (1988). Nesom: Erigeron in California, Nevada, and Arizona 195 Erigeron concinnus is superficially very similar to E. chrysopsidis A. Gray var. austiniae (E. Greene) Nesom (see topic 10, below), particularly in the production of disciform heads, the pistillate flowers present but with extremely reduced ligules. This condition is not strictly homologous in these two species because E£. austznzae E. Greene is apparently intimately related to a different group of species. Similar reduction of ligules also occurs in other, yet more distantly related, species of the genus. 10. The Erigeron chrysopsidis A. Gray complex. My study of this complex is essentially in agreement with Cronquist’s treat- ment of the same taxa, except for the provision of an additional taxonomic cat- egory for the southern segment of Erigeron chrysopsidis, which produces dis- ciform heads. Erigeron chr ysopsidis subsp. chrysopsidts (radiate) and subsp. austinzae (E. Greene) Crongq. (disciform) are allopatric and apparently closely related vicariads (Map 2). The latter might be maintained as a separate species, but in northwest Owyhee Co., Idaho, and in northern Malheur Co., Oregon, near the range of typical E. chrysopsidis, there are many plants, rep- resented by numerous collections from different populations, with yellow, short ligules 1-3 (rarely 4) mm long. Elsewhere in the range of subsp. austinzae (E. Greene) Cronq., even the production of tiny ligules occurs only rarely. While these ligulate variants clearly belong with subsp. austinzae rather than with typical E. chrysopsidis, | cannot find any other character besides ligule length that would consistently distinguish them. Cronquist (1947) noted within one radiate population of E. chrysopsidis from Umatilla Co., Oregon, a rare ten- dency for the reduction of ligules parallel to that of subsp. austiniae. Erigeron austiniae is justifiably regarded as conspecific with E. chrysopsidis; | treat it as a variety in order to provide a category of coordinate rank with the other varieties of E. chrysopsidis. In the rest of the genus, “variety” is used as the primary taxonomic category to describe infraspecific variants. Cronquist (1947) initially treated the plants of Erigeron chrysopsidts var. brevifolius Piper as a synonym of the typical element of E. chrysopsidis, but he later acknowledged their distinctive morphology and accepted their formal taxonomic recognition (1960; Hitchcock & Cronquist 1973). Although the differences of var. brevifolius from the typical variety are primarily quantitative, the former is easily distinguished and is restricted to the Wallowa Mountains of Wallowa Co., Oregon (Map 2), at high elevations (6100-10300 ft), whereas var. chrysopsidis occurs at 2800-4300(-6000) ft over its range. Intergrades exist but apparently are few, and more detailed study, including field observations, are needed for a more critical assessment of the nature of the relationship between the two taxa. Keys that distinguish Erigeron chrysopsidisand E. piperianus (1947; Hitch- cock & Cronquist 1973) have relied primarily on differences in stem vestiture 196 PHYTOLOGIA volume 73(3):186-202 September 1992 A E. piperianus 2 ns a ccaase ie E. chrysopsidis fF @ var. chrysopsidis 4 @ @ var. brevifolius @ var. austiniae > chrysopsidis Ca ; Map 2. Distribution of Erigeron chrysopsidis and E. piperianus. Nesom: Erigeron in California, Nevada, and Arizona 197 and in size of heads and plants. With a greater range of specimens available for comparison, it can now be seen that these putative morphological dif- ferences between E. chrysopsidis var. brevifolius Piper and FE. piperianus are considerably overlapping. Both are separated from typical EF. chrysopsidis by a tendency for the stem hairs to be appressed and by a reduction in head and flower size. Compared to var. brevifolius, however, E. ptperranus has leafier stems and the stem indument lacks the minute Type B trichomes and stipitate glands (Type C trichomes) commonly produced by E. chrysopsidis. There are four yellow rayed species of North American Erigeron: E. chrysop- stdis, E. piperranus, E. linearis (Hook.) Piper, and E. aureus E. Greene. If it is acknowledged that FE. chrysopsidis and E. piperianus are sister species, there are three yellow rayed lineages, each apparently independently derived from white or blue rayed ancestors. Erigeron linearis may be closely related to E. chrysopsidis and E. piperianus (see notes below, topic 11), but they are not derived from an immediate common ancestor. Following is a formal nomenclatural summary and key to Erigeron piperi- anus and the varieties of E. chrysopsidis. Erigeron pipertanus Crongq., Brittonia 6:197. 1947. TYPE: UNITED STATES. Washington: Grant Co., sagebrush slopes N of Soap Lake in Grand Coulee, 18 May 1935, J.W. Thompson 11490 (HOLOTYPE: NY!; Iso- types: GH!,LL!,MO!,RM!,TEX!,US!). Erigeron chrysopsidis A. Gray Chrysopsis hirtella DC., Prodr. 5:327. 1836. TYPE: UNITED STATES. [probably Oregon]: Columbia River, [ca. 1826,] D. Douglas s.n. (de- Candolle cited only the Douglas collection, “v.s. comm. ab hon. soc. hortic. Lond.;” photo DS!, photo and fragment UC, fide Cronquist 1947). Erigeron ochroleucus Nutt. var. hirtellus (DC.) A. Gray, Proc. Amer. Acad. Arts 16:90. 1880. Erigeron chrysopsidis A. Gray {nom. nov.], Syn. Fl. N. Amer. 1(2):210. 1884 (not E. hertel- lus DC. 1836.). Erigeron curvifolius Piper [nom. nov.]|, Bull. Tor- rey Bot. Club 27:396. 1900. The proposal by Piper of a second nomen novum for deCandolle’s species apparently reflected both a misunderstanding of the nomenclatural code and a misidentifica- tion of the species. The plants Piper associated with his concept of E. chrysopsidis were those later named by Cronquist (1947) as E. piperianus. The type of E. chrysopsidis is not mapped (Map 2) or precisely known, as Douglas apparently collected at many sites along the Columbia River, including areas where the species is known to occur. 198 PHY 2026 O-G TA volume 73(3):186-202 September 1992 Erigeron chrysopsidis A. Gray subsp. chrysopsidis Erigeron chrysopsidis A. Gray subsp. chrysopsidis var. chrysop- sidis Erigeron chrysopsidis A. Gray subsp. chrysopsidis var. brevifolius Piper, Bull. Torrey Bot. Club 27:395. 1900. TYPE: UNITED STATES. Oregon: [Wallowa Co.,] subalpine ridge of the Wal- lowa Mts. near the [Wallowa] Lake, 29 Jul 1899, W.C. Cusick 2270 (HOLOTYPE: WS; Isotypes: GH!,MO!,NY!). Erigeron chrysopsidis A. Gray subsp. austiniae (E. Greene) Crongq., Brittonia 6:196. 1947. BASIONYM: Erigeron austiniae E. Greene (below) Erigeron chrysopsidis A. Gray subsp. austiniae var. austiniae (E. Greene) Nesom, comb. et stat. nov. BASIONYM: Erigeron aus- tintae E. Greene, Erythea 3:100. 1895. TYPE: UNITED STATES. California: Modoc County, Davis Creek, May 1894, Mrs. R.M. Austin s.n. (LECTOTYPE [designated here]: ND-G!; Isolectotypes: NY!,PH,UC). Key to the taxa of the Erigeron chrysopsidis group a. Ligules of ray flowers completely absent or inconspicuous and not or barely exceeding the involucre; longest phyllaries 4.5-6.0 mm long; disc corollas 3.5-4.5 mm long; stem pubescence spreading. ........... var. austiniae a. Ligules of ray flowers well developed and conspicuously exceeding the involucre; phyllary and corolla length various; stem pubescence variable. ee ee eee (b) b. Longest phyllaries 5.0-6.5(-7.5) mm long; disc corollas 3.0-4.5 mm es ligules (6-)8-11 mm long; stem pubescence spreading. ...... ROR ec tbh UA i sto le tad ee var. chrysopsidts b. Longest phyllaries 3.5-5.5 mm long; disc corollas 2.5-4.5 mm long; ligules 4-7 mm long; stem pubescence appressed to ascending ap- pressed or spreading, the hairs of even length or markedly uneven. sla tays atin ais tole alebeymaishe sie epeiant. ago sistiaraual = GWiuryees Ro a pay ores eee C. tolucensis Kunth 3. Calyx without red coloration, sparsely hispid-pilose, eglandular; leaves 2-4 lobed, lobes nearly filiform with acute apices; corollas exserted 8-10 mm from the calyx; Oaxaca (Cerro Zempoaltepetl). ...5....085.-ccecerne- cago eh: SRN Bev Ne En a aS a C. zempoaltepetlensis Nesom 4. Plants usually single stemmed from the base, always with an erect SE BTA t eta Gh rare UP. re. 2 es, aid ba ia ots Solana ete ee ee ete nia (6) 5. Calyx green, the veins and lobe margins sparsely pilose with loosely spread- ing, vitreous hairs; leaves mostly 4-6 mm wide at midpoint; Nuevo Leon amoecoahttila.. Sirs SNe. stediasteas 2s Sots Meee onan icle C. bella Standley 5. Calyx cinereous, the veins and lobe margins densely ciliate with stiffly spreading, white hairs; leaves mostly 1-3 mm wide at midpoint (below the divergence of the lobes); Hidalgo, México, Puebla. C. moranensis Kunth 6. Plants annual from a very slender, short taproot; Edo. Mexico to | a Ca aac Nr en a Prete ee MAES SRE HR (8) 6. Plants perennial with a woody, thickened root; Veracruz to Oaxaca. Batista arecis toulrors tts. num pokes Mee. SOR ae OO (7) 7. Calyx with a red apex; Veracruz and Puebla (Orizaba and Perote). “iS Ne ROR EI ee he eA GE tape er Paced eap sree Sete, ie C. falcata Eastw. 7. Calyx completely green except for a yellow, narrow but prominent apical ARTE WOH CONE ts jah 552 318, 0, Bia. od Sis SEY 2 tes ee C. conzattw Fern. 8. Calyx green with a prominently red apex; stems sparsely invested with spreading hairs, stipitate glandular hairs usually present. (10) 8. Calyx greenish to yellow, commonly without red coloration (some- times barely red at the apex in C. sphaerostigma); stems with de- flexed or retrorsely appressed hairs, eglandular. .............. (9) 9. Stems moderately invested with thin, antrorsely appressed hairs; basal leaves pectinately lobed, the cauline entire; floral bracts lanceolate, en- tire; stigmas with lobes 2 mm long; Edo. México and eastern Michoacan. ee are ee acd vin auc Se ysieix ele alee SES C. macrostigma B. Rob. 250 PHY FO LO Gra volume 73(3):247-250 September 1992 9. Stems densely invested (nearly floccose in the inflorescence) with relatively thick, deflexed hairs; lower and cauline leaves usually with 1-3 pairs of linear lobes; floral bracts with apex somewhat dilated and shallowly lobed or distinctly scalloped to coarsely crenate; stigmas nearly capitate (thickened and slightly notched); west central Zacatecas and adjacent Durango, ors come Os ciiicte ato etele Wace ols Nee ls C. sphaerostigma Eastw. 10. Leaves with filiform lobes; floral bracts entire; northwestern Mi- CHoacanr yes carpe omer ethic ca sietcgs oS C. j1lquilpana Nesom 10. Leaves with oblanceolate to spatulate lobes; floral bracts lobed; Durango paces Hoesen ee oN ei ee aee C. saltensis Eastw. ACKNOWLEDGMENTS I thank Dr. B.L. Turner and Mark Mayfield for their review of the manuscript and the staff of GH for a loan of specimens. LITERATURE CITED Eastwood, A. 1909. Synopsis of the Mexican and Central American species of Castilleja. Proc. Amer. Acad. Arts 44:563-591. Nesom, G.L. 1992. New species and critical evaluations of Mexican Castilleja (Scrophulariaceae). Phytologia 72:231-252. Phytologia (September 1992) 73(3):251-254 A NEW SPECIES OF STEMODIA (SCROPHULARIACEAE) FROM COSTA RICA Billie L. Turner Department of Botany, University of Texas, Austin, Texas 78713 U:S-A. ABSTRACT A new species, Stemodia costaricensis B. Turner, is described from Cartago Province, where it is apparently locally abundant. It is known by twenty or more collections, all from the vicinity of Tapanti. The species is closely related to S. reliquiarum D’Arcy of Chiriqui Province, Panama, but is readily distinguished by several consistent characters including leaf shape, lack of vestiture, and corolla size and shape. A key to the two taxa is provided along with a map showing their distribution. KEY WORDS: Stemodia, Scrophulariaceae, Costa Rica, Panama D’Arcy (1979) described a new species of Stemodia, S. reliquiarum D’Arcy from Panama. It is remarkable within Stemodia because of its pubescent, seemingly sessile anthers, and he commented that it “might warrant recog- nition at the generic level.” Discovery of the present species, which contains the essential elements of S. reliquiarum, including pubescent anthers, adds an additional taxon to the complex concerned. The following couplet will distin- guish between the two taxa: Leaves mostly broadly ovate to subdeltoid; pedicels and sepals glabrous; corollas strongly bilabiate (Lobelza-like), the central lower lobe 4-6 mm long; Costa Rica (Cartago Province). ............-.--- S. costaricensis Leaves mostly ovate-lanceolate; pedicels and sepals glandular hirsute; corol- las not strongly bilabiate, the central lower lobe 2-3 mm long; Panama Mie een COVATICE i oie Sige pete eee eter = 0 oases lata ee S. reliquiarum 251 September 1992 volume 73(3):251-254 PHY FOTO GIA 252 Fig.1. Distribution of Stemodia costaricensis (open circle) and S. reliquiarum (closed circle) in Costa Rica and Panama respectively. Turner: New Stemodia from Costa Rica 253 Stemodia costaricensis B. Turner, sp. nov. TYPE: COSTA RICA. Cartago Province: “1-4 km beyond first bridge within Hydroelectric Plant Prop- erty (Instituto Costaricensis Electricidad) enroute to the reservoir at the road terminus,“ 4800-4900 ft, common but very local, 4 Mar 1981, F. Almeda & K. Nakai 4734 (HOLOTYPE: TEX!; Isotype: CAS). Stemodiae reliquiaro D’Arcy sed foliis late ovatis (vs. lanceolati- ovatis), pedicellis ac calycibus glabris (vs. pilosis), et corollis ma- joribus lobis longissimis 4-6 mm longis (vs. 2-3 mm) differt. Sprawling or trailing suffruticose glabrous perennial herbs 10-100 cm long, 1.0-1.8 cm wide; petioles mostly 3-8 mm long; blades broadly ovate to triangu- lar ovate, trinervate, minutely punctate beneath, the margins serrulate. Flow- ers arranged in terminal bracteate racemes 3-8 cm long, the pedicels glabrous, mostly 8-14 mm long. Calyx glabrous, ebracteolate, mostly 3-4 mm long, the lobes essentially alike and free to the base. Corollas reportedly deep violet blue and “Lobelia-like,” the tube ca. 3 mm long, the upper 2 lobes 2.5-3.0 mm long, the lower 3 lobes mostly 3-6 mm long, the central lobe 4-6 mm long. Capsule ovate, ca. 4 mm long. Seeds numerous, brown, ovoid, finely reticulate, ca. 0.5 mm long. ADDITIONAL SPECIMENS EXAMINED: COSTA RICA. Cartago: In- stituto Costaricensis de Electricidad, ca. 13 km beyond the bridge at Tapanti, 1400-1600 m, 10 Jul 1977, Almeda 3022 (CAS,F); Refugio Nacional de Fauna Silvestre Tapanti, along river just beyond Puente dos Amigos, 1500 m, 23 Mar 1986, Almeda 5761 (TEX); ca. 10 km S of Tapanti, ca. 1600 m, 14-17 Jul 1971, Burger 7537 (DUKE); Tapanti Hydroelectric Reserve, along Rio Dos Amigos, 1600-1700 m, 23 Jun 1976, Croat 36214 (F,MO); Tapanti Watershed Preserve, ca 20 mi SW of Paraiso, 5 Feb 1979, Croat 46990 (MO); 12 km S of Tapanti, 1520 m, 5 Nov 1967, Lent 1454, 1571 (F); Tapanti, mountain above town, ca. 1500 m, 23 Jun 1972, Primack 180 (DUKE); ca. 12 km from the bridge at Tapanti, 1400-1500 m, 11 Aug 1981, Taylor 1888 (DUKE); 7-12 km SE of Orosi, 1400-1600 m, 12 May 1975, Utley 2500 (F); wooded slopes above Rio Grande de Orosi, 1400-1700 m, 18 Feb 1976, Utley 4157 (DUKE); and all of the following collections by R.L. Wilbur of DUKE which were collected in the same general region about Tapanti: 18068, 18495, 22385, 22484, 30795, 30875, 32939, 38083. The twenty or more collections cited above were all obtained from the vicinity of Tapanti and all are exceedingly uniform. D’Arcy (1979) cited one of these (Croat 36214) as belonging to his concept of Stemodia reliquiarum. The two taxa, while closely related, are clearly distinct and by “specific standards” established in the genus Stemodia by previous workers appear to be worthy of specific rank. 254 PRY TO E.O'G:iA volume 73(3):251-254 September 1992 ACKNOWLEDGMENTS I am grateful to Guy Nesom for the Latin diagnosis and to him and Carol Todzia for reviewing the manuscript. LITERATURE CITED D’Arcy, W.G. 1979. Stemodia, in Flora of Panama, Ann. Missouri Bot. Gard. 66:252-262. Phytologia (September 1992) 73(3):255-258. A NEW SPECIES OF STENOCARPHA (ASTERACEAE, HELIANTHEAE, GALINSOGINAE) FROM MEXICO B.L. Turner Department of Botany, University of Texas, Austin, Texas 78713 U.S.A. ABSTRACT A new species Stenocarpha ritovegana B. Turner, is described from the vicinity of Surutato, Sinaloa. It differs from the only previ- ously known species of the genus, S. filiformis (Hemsl.) S.F. Blake, in having slender rhizomes and 1-3 much larger heads borne upon elongate peduncles. KEY WORDS: Stenocarpha, Galinsoga, Asteraceae, Heliantheae, Galinsoginae, Mexico, Sinaloa Routine identification of Mexican Asteraceae has revealed the following novelty. Stenocarpha ritovegana B. Turner, sp. nov. Fig. 1. TYPE: MEXICO. Sinaloa: Mpio. Badiraguato, alrededores de Surutato, bosque de pino, 1600-1800 m, 11 Dec 1987, Rito Vega 2593, with F. Hernandez y A. Hernandez (HOLOTYPE: TEX; Isotype: EACS). Stenocarphae filiform: (Hemsl.) S.F. Blake similis sed differt plantis perennis rhizomatibus tenuibus exorientibus (vs. annuis radice palari exorientibus), capitulis majoribus (12-15 mm latis trans radios expansos vs. 5-8 mm), et capitulescentia capitulis pau- cioribus instructa (1-3 per caulem vs. 5-numerosis). Erect scapose perennial herbs 20-30 cm high. Stems mostly less than 1 mm in diameter, arising from slender rhizomes. Leaves all basal, occasionally 1 or 2 pairs of much reduced stem leaves, 2-4 cm long, 1.0-1.5 cm wide; petioles 5-10 mm long; blades lanceolate to oblanceolate, tapering upon the petioles, 255 256 PHYT @LOGLA volume 73(3):255-258 September 1992 ) Ae ATHY if “ir itl ia \ \\ cs 3 ij Me FZ ao jl Wt fay Pe / E ‘ "} BETAS aa Fig.1. Stenocarpha ritovegana, from holotype. Turner: New Stenocarpha from Mexico 207 glabrous or nearly so, trinervate to subpinnately nervate, the margins crenulo- denticulate. Heads 1 to 3 to a stem, the common peduncle up to 15 cm long, the ultimate peduncles mostly 4-10 cm long. Involucres campanulate, 3-4 mm high, 6-7 mm across, the bracts 2-3 seriate, subequal, linear-elliptic, glabrous, striate, the apices obtuse or rounded. Receptacle conical, ca. 2 mm across at the base, ca. 3 mm high, the bracts linear, persistent, 2.5-3.0 mm long, ca. 0.15 mm wide. Ray florets ca. 11, pistillate, fertile, not forming a complex with the pales as in Galinsoga, the ligules white, 3-6 mm long, ca. 2 mm wide, the tube ca. 1.5 mm long. Disk florets numerous, the corollas yellow, ca. 2mm long, the tube hispidulous, ca. 0.6 mm long, the lobes minutely hispidulous, ca. 0.25 mm long. Achenes, those of the ray similar to those of the disk but epappose, those of the disk black, striate, narrowly obpyramidal, somewhat 4 sided, ca. 1 mm long, 0.3 mm wide, moderately strigose, the pappus of ca. 8 persistent white lanceolate scales, 1.0-1.5 mm long, erose at the apices, 4 of these usually possessing setaceous awns. Canne (1977) would have presumably positioned this taxon in the section Elata of her concept of the genus Galinsoga, where it would nestle next to its previously only known species, G. filiformis Hemsl. Blake elevated G. fil- uformis in taxonomic rank to the monotypic genus Stenocarpha, and Turner (1965) maintained the genus, but it must be admitted that the generic lines among phyletic groupings in the subtribe Galinsoginae are tenuous at best, as painfully noted by Turner (1990). Regardless, in my treatment for México, I intend to restrict Galinsoga to those species having an achene/phyllary com- plex that falls as a unit at maturity. Stenocarpha, which lacks such a complex, might ultimately be found to be related more closely to Sabazza than to Galin- soga. Whatever the generic lines, there can be little doubt that the present species belongs within Stenocarpha since it contains all of the previously de- scribed characters of that group, except for the few headed capitulescence and rhizomatous habit. ACKNOWLEDGMENTS I am grateful to Guy Nesom for the Latin diagnosis and to him and Dr. T.P. Ramamoorthy for reviewing the manuscript. Nancy Webber provided the illustration. LITERATURE CITED ‘Canne, J. 1977. A revision of the genus Galinsoga (Compositae: Heliantheae). Rhodora 79:319-389. 258 PHY TOLOGQIA volume 73(3):255-258 September 1992 Turner, B. 1965. Taxonomy of Stenocarpha (Compositae-Heliantheae-Galin- soginae). Southwest. Naturalist 10:238-240. . 1990. A re-evaluation of the genus Alepzdocline (Asteraceae, He- hantheae, Galinsoginae) and description of a new species from Oaxaca, Mexico. Phytologia 69:387-392. Phytologia (September 1992) 73(3):259-260. A NEW COMBINATION IN HEMIZONIA (ASTERACEAE: MADIINAE) David J. Keil Biological Sciences Department, California Polytechnic State University, San Luis Obispo, California 93407 U.S.A. ABSTRACT Hemizonia leucocephala Tanowitz is reduced to subspecific rank as H. congesta DC. ssp. leucocephala (Tanowitz) Keil. KEY WORDS: Hemizonia, Asteraceae, Madiinae Based on examination of types and a review of the application of names, Tanowitz (1983) recognized that the taxon that had been treated in recent floras as Hemizonia multicaulis Hook. & Arn. should properly be called H. congesta DC., and the the taxon that had been treated as H. congesta (sensu stricto) has actually never been named. Tanowitz therefore described the latter taxon as H. leucocephala Tanowitz. Members of the Hemzzonza congesta complex have been treated by Babcock & Hall (1924) as one variable species with six subspecies and by Clausen (1951) and Keck (1958, 1959, 1960) as seven species. Babcock & Hall presented strong evidence that these taxa are all more of less interfertile and that intergradation in nature is common where ranges overlap. Clausen (1951) presented a crossing polygon that confirmed that these taxa are weakly to moderately interfertile in various combinations. Nevertheless, Clausen & Keck treated these plants as distinct species. I have been unable to document their rationale for recognition of these taxa at the species level. In the Jepson Manual, I am following Babcock & Hall in treating these taxa as one polymorphic species. I recognize the same taxa that Babcock & Hall (1924) recognized. However, because of the nomenclatural changes that resulted from Tanowitz’s (1983) study, application of names has changed for some taxa. I am treating as Hemizonia congesta ssp. congesta, the taxon that Babcock & Hall (1924) called H. congesta var. lutescens (E. Greene) Babcock & Hall, and that Keck (1959, 1960) split into H. lutescens (E. Greene) Keck, H. multicaults Hook. & Arn., and H. multicaulis ssp. vernalis Keck. The taxon that Tanowitz treated as H. leucocephala has no name at the subspecific level. I therefore propose the following new combination: 259 260 PHY TOV Oras volume 73(3):259-260 September 1992 Hemizonia congesta DC. ssp. leucocephala (Tanowitz) Keil, comb. nov. BASIONYM: Hemizonia leucocephala Tanowitz, Bull. Torrey Bot. Club 110:15. 1983. ACKNOWLEDGMENTS I thank Dr. Bruce G. Baldwin and Dr. Rhonda L. Riggins for reviewing this paper. LITERATURE CITED Babcock, E.B. & H.M. Hall. 1924. Hemizonza congesta, a genetic, ecologic, and taxonomic study of the hayfield tarweeds. Univ. California Publ. Bot. 13:15-88. Clausen, J. 1951. Stages in the Evolution of Plant Species. Cornell University Press, Ithaca, New York. Keck, D.D. 1958. Taxonomic notes on the California flora. Aliso 4:101-114. . Hemizonia Pp. 1117-1125 in P.A. Munz in collaboration with D.D. Keck, A California Flora. University of California Press, Berkeley, Cal- ifornia. . 1960. Hemizonia Pp. 172-184 in L. Abrams & R.S. Ferris. JIlus- trated Flora of the Pacific States, Washington, Oregon, and California. IV. Bignoniaceae to Compositae, bignonias to sunflowers. Stanford Uni- versity Press, Stanford, California. Tanowitz, B.D. 1983. Taxonomic status of Hemizonia congesta DC. and Hemizonia corymbosa (DC.) T. & G. (Asteraceae: Madiinae). Bull. Torrey Bot. Club 110:12-17. — New ini Information for Authors Articles from botanical systematics and ecology, including biographical sketches, critical reviews, and summaries of literature will be considered for publication in PHYTOLOGIA. Manuscripts may be submitted either on computer diskette, or as typescript. Diskettes will be returned to authors after action has been taken on the manuscript. Diskettes may be 5.25 inches or 3.5 inches and may be written in any IBM or MacIntosh compatible format. Typescript manuscripts should be single spaced and will be read into the computer using a page scanner. The scanner will read standard typewriter fonts but will not read dot matrix print. Manuscripts submitted in dot matrix print cannot be ac- cepted. Use underscore (not italics) for scientific names. Corrections made On typescript manuscripts must be complete and neat as the scan- ner will not read them otherwise. Language of manuscripts may be either English or Spanish. Figures will be reduced to fit within limits of text pages and therefore, should be submitted with an internal scale and have dimensions proportional to those for text pages. Legends for figures should be included in figures whenever’ possible. Each manuscript should have an abstract and key word list. Specimen citations should be consistent throughout the manuscript. Serial titles should be cited with abbreviations used in Botanico Periodicum Huntianum. References cited only as part of nomenclatural summaries should not appear in Literature Cited. Nomenclatural work should include one paragraph per basionym and must provide proper (as defined by the current /nterna- tional Code of Botanical Nomenclature) citation of sources of epithets and combinations. Authors should arrange for two workers in the appropriate field to review the manuscript before submission. Copies of reviews should be forwarded to the editor with the manuscript. Manuscripts will not be published without review. Cost of publication is currently $13.00 US per page for publication without reprints. Publication with 100 reprints is provided for $18.00 US per page, 200 reprints for $21.50 US per page. Page charges are due with manuscript and no paper will be published before payment is received in full. Reprints must be ordered and paid for in advance. Page charges will be determined on the basis of a typescript page (single spaced, 10 points, blank line between paragraphs) with all type inside a rectangle 143 mm (horizontal) by 219 mm (vertical), not including running head and page number. Title page should include title, author(s) name(s), and address(es). Two blank lines should appear above and below section headings (Abstract, Discussion, Literature Cited, etc.) in the manuscript. No extra charge is made for line drawings provided they conform to limitations of size and proportion for normal text. Halftones require an extra charge of $10.00 US per page.