PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 122(2): 155-161. 2009. , a new of (Compositae or ) from the northern Andes

Vicki A. Funk* and Harold Robinson (VAF) U.S. National Herbarium, Department of Botany, MRC 166, National Museum of Natural History, P.O. Box 37012, Smithsonian Institution, Washington, DC. 20013-7012, U.S.A., e-mail: [email protected]; (HR) Department of Botany, MRC 166, National Museum of Natural History, P.O. Box 37012, Smithsonian Institution, Washington, DC. 20013-7012, U.S.A., e-mail: [email protected]

Abstract.—The genus (Kunth) Cass. is divided into two genera based on the two former subgenera and a new name was necessary for the non-typical subgenus. DNA sequence data have shown that these two groups are not sister taxa and that each is easily defined by morphological characters that include differences in the habit, position of the inflorescence, and the pubescence of the style branches. Sampera, with eight species, is named in honor of Dr. Cristian Samper, Director of the Natural History Museum, Smithsonian Institution.

The tribe Liabeae includes a distinctive ment of the tribe in the subfamily but strictly tropical group of American sensu lato near the Verno- Compositae not recognized at the tribal nieae and the Lactuceae, and this was level until Rydberg (1927) established a later confirmed by a morphological cla- separate tribe in his study of the Mexican distic study by Robinson & Funk (1987). and Central American species. In earlier The structural studies of Robinson studies of the family, the group was (1983a, 1983b) also seemed to have buried in either the (Cassini resolved the generic limits within the 1830) or the (Bentham 1873, group, recognizing 15 genera distributed Hoffmann 1894). The latter treatments from Mexico and Cuba in the north to were particularly unhelpful, with almost in the south. That apparent all the species placed in one genus, resolution was a major improvement over Adans., in the Senecioneae but previous classifications, but it was not the with related elements scattered in other final word. tribes: DC. in the He- New collections from northern Peru liantheae, Kunth in the Hele- revealed a previously undescribed genus, nieae, and Chionopappus Benth. in the Bishopanthus H. Rob. (Robinson 1983b). Mutisieae. The chaotic treatment was A morphological cladistic study of the partly because the tribe Liabeae possessed tribe (Funk et al. 1996) provided a a confusing combination of characteris- phylogeny that was similar to the intuitive tics, with styles like the Vernonieae, one described by Robinson (1983a, yellow rays like the Senecioneae, and 1983b) but with a few new sister groups often latex like the Lactuceae. proposed. Re-examination of some other The structural studies of Robinson specimens resulted in the reduction of (1983a) had essentially resolved the place- Austroliabum H. Rob. & Brettell to synonymy under Cabrera ! Corresponding author. (Robinson 1990). The application of 156 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

DNA sequencing showed that a new Sampera substituted for the subgeneric genus, Dillandia V. A. Funk & H. Rob. names Oligactis and Andromachiopsis. (Funk & Robinson 2001), was needed for three species, in a group placed previously Inflorescence axillary or terminal, with a in Ruiz & Pav. Now, new subglomerate, spiciform, or racemose molecular data from nuclear and chloro- arrangement of heads, heads usually plast DNA sequencing has not only with 6-10 florets; anther appendages supported the establishment of Dillandia, with papillose surfaces Oligactis Inflorescence terminal, with a corymbose but also demonstrated the need for arrangement of heads; heads with another new genus in the Liabeae (Dillon 18-50 florets; anther appendages et al. 2009, Funk & Chan pers. comm.). smooth Sampera Within the Liabeae, the two genera, Liabum and Oligactis (Kunth) Cass., are The characteristics cited remain the clearly closely related (Robinson 1983a). most valid structural distinctions avail- These two genera can be distinguished able for the two genera. In addition, in from other members of the tribe by their habit, typical Oligactis is a more com- pale anther thecae with teeth at the base, plete vine with slender stems, whereas the absence of stomata on the disk corolla Sampera is shrubbier in appearance, with lobes, the lack of latex, and the presence of thicker stems. The style branches of the subquadrate raphids in the achene walls. disk florets are longer than the hispidu- Of the two genera, Liabum has leaves that lous upper part of the style shaft in are trinervate or triplinervate with ascend- Sampera, rather than shorter as in Oli- ing lower secondary veins, whereas Oli- gactis. gactis has leaves with pinnate venation The new genus is named for Cristian (the apomorphic character). However, it is Samper, Director of the National Muse- Oligactis that has been shown to be non- um of Natural History (U.S.A.), former monophyletic. Robinson & Brettell (1974) Director of the Institute Alexander von recognized two subgenera in Oligactis. Humboldt, Santafe de Bogota, Colombia, Because molecular data indicate that these former Acting Secretary of the Smithso- two subgenera are not sister taxa, the nian Institution, 2007-2008, and Honor- atypical subgenus must be elevated to the ary Curator of Tropical Ecology in the genus level. Oligactis subg. Andromachiop- Department of Botany, National Muse- sis H. Rob. & Brettell is therefore removed um of Natural History. Dr. Samper has from Oligactis and placed in the new made a lifelong commitment to the genus Sampera. understanding of life on earth and its conservation, and it is fitting that a genus Systematic Account with some of its taxa in Colombia is named after him. It is interesting to note The DNA sequence data (ITS, ?r«L-F, that one of the eight species in this new ndh¥) separated the two groups in Oli- genus carries the name Sampera cuatre- gactis that were already known to have at casasii (M. O. Dillon & Sagast.) V. A. least subgeneric differences. What was Funk & H. Rob., combining the names of not evident from the morphology was Dr. Samper and the famous Smithsonian that one of the two subgenera was the botanist Dr. Jose Cuatrecasas, who sister taxon of Liabum but the other was worked for many years in Colombia. not (Funk & Chan pers. comm.). Robin- son (1983a) furnished a key to the Sampera V. A. Funk & H. Rob., subgenera that is given here in revised gen. nov. form with the genera Oligactis and Type: Liabum coriaceum Hieron. VOLUME 122, NUMBER 2 157

Oligactis subgen. Andromachiopsis H. of achene wall with subquadrate raphids; Rob. & Brettell, Phytologia 28(1):58 carpopodium short, annuliform, with (1974). Type: Liabum pichinchense small subquadrate cells in 3-5 series, Hieron. walls of cells moderately thickened; pap- Plantae frutescentes subscandentes non pus with 20-35 rather persistent inner laticiferae; folia opposita; inflorescentiae capillary setae sometimes with broadened corymbiformes; thecae antherarum palli- tips, with teeth of setae appearing mucro- dae; appendices apicales antherarum nate-tipped in Hoyer's solution, with laeves; raphidis acheniarum subquadratis. outer series of 7-20 short setae or Scrambling , moderately branch- squamellae. Pollen spherical, 30-37 um ing, stems terete to strongly hexagonal, in diameter, with spines rather unevenly mostly tomentose, without latex; nodes dispersed, with distinct group of inner with or without disks; leaves opposite; columellae under each spine. petioles with or without wings, sometimes The genus Sampera, with eight species, included in perfoliate leaf pairs; blades is distributed from Colombia southward sharply delimited and rounded to slightly to northern Peru, with the majority of the cordate at base or confluent with petiole, species in . The genus Oligactis, ovate to oblong-ovate, margins subentire as now delimited, has species found in to serrate, never angulate, upper surface Costa Rica, Panama, Colombia, and flat to slightly bullate, densely tomentose Venezuela. below, pinnately veined. Inflorescence terminal on branches, corymbiform, pe- Key to the species of Sampera duncles less than 5 cm long, thinly to la. Median involucral bracts ovate, densely tomentose. Heads with involucres shortly acute; longer pappus bristles broadly campanulate, involucral bracts thickened, especially at tips; upper 30-55 in 4-5 series, narrowly ovate to surfaces of leaf blades smooth or lanceolate, tips obtuse to narrowly acute, with exsulcate venation 2 outer surface puberulous to arachnoid lb.All involucral bracts lanceolate, nar- tomentose or hirsute. Ray florets 6-18, rowly acute; longer pappus bristles female; corollas yellow, with elliptical slender with tips only slightly broad- limbs; style branches not spiraled. Disk ened; upper surfaces of leaf blades florets 10-34, bisexual; corollas yellow, often rugose or bullate 4 2a. Nodes with stipuliform disks to 3 or narrowly funnelform, lobes 5, linear 4 times as wide as the stem; heads lanceolate, slightly longer than throat, usually with 12 ray florets and 12 without evident stomata; anther filaments disk florets S. ochracea smooth, with weak annular thickenings in 2b. Nodes without stipuliform disks or walls; thecae pale, digitate at base; with disks not more than twice as endothecial cells oblong, with oval or wide as the stem; heads with 2-10 strap-shaped sclerified bands, with single ray florets and ca. 10-15 disc florets thickening on each transverse wall; apical 3 appendages oblong-ovate, smooth with 3a. Heads with 2-4 ray florets and ca. 15 short-oblong to subquadrate cells. Nec- disk florets; lower surfaces of leaves with thin tomentum allowing dark tary short, not or scarcely lobed. Style color of secondary veins to show. . . . base scarcely broadened; style branches S. cuatrecasasii up to 20 times as long as wide, ca. 1.5 3b. Heads with 8-10 ray florets and ca. times as long as hispidulous part of upper 10 disk florets; lower surfaces of style shaft, tips narrowly obtuse. Achenes leaves with thick tomentum obscur- prismatic, with 5-8 ribs, bearing contort- ing color of secondary veins ed setulae and small glandular dots, cells S. coriacea 158 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

4a. Leaf blades ovate-lanceolate; pedun- Sampera cuatrecasasii (M. O. Dillon & cles with appressed arachnoid tomen- Sagast.) V. A. Funk & H. Rob., tum; from near or below comb. nov. 2500 m S. asplundii 4b. Leaf blades mostly oblong-ovate; Oligactis cuatrecasasii M. O. Dillon & peduncles with floccose tomentum; plants mostly above 2500 m 5 Sagast., Arnaldoa 2(2):25 (1994)[1995]. 5a. Base of leaf blade abruptly acute to Distribution.—Peru. subtruncate 6 5b. Base of leaf blade acuminate to Sampera cusalaguensis (Hieron.) V. A. decurrent 7 Funk & H. Rob., comb. nov. 6a. Heads with (5—) 6-8 ray florets; inner pappus bristles 3.5—4.0 mm long . . . Liabum cusalaguense Hieron., Bot. Jahrb. S. ecuadoriensis Syst. 29:55 (1900). 6b. Heads with 14-18 ray florets; inner Oligactis cusalaguensis (Hieron.) H. Rob. pappus bristles 4.5-5.5 mm long .... & Brettell, Phytologia 28(1):58 (1974). S. pichinchensis Distribution.—Ecuador. 7a. Petiole winged, continuous between decurrent base of leaf blade and Sampera ecuadoriensis (Hieron.) V. A. nodal disk S. cusalaguensis 7b. Petiole not winged S. pastoensis Funk & H. Rob., comb. nov.

Sampera asplundii (H. Rob.) V. A. Funk Liabum ecuadoriense Hieron., Bot. Jahrb. & H. Rob., comb. nov. Syst. 19:60 (1894). Oligactis ecuadoriensis (Hieron.) H. Rob. Oligactis asplundii H. Rob., Phytologia & Brettell, Phytologia 28(1):58 (1974). 35(3):200 (1977). Distribution.—Ecuador. Distribution.—Ecuador. Sampera ochracea (Cuatrec.) V. A. Funk Sampera coriacea (Hieron.) V. A. Funk & & H. Rob., comb. nov. H. Rob., comb. nov. Figs. 1, 2 Liabum ochraceum Cuatrec, Collect. Bot. Barcilone 3:302 (1953). Liabum coriaceum Hieron., Bot. Jahrb. Oligactis ochracea (Cuatrec.) H. Rob. & Syst. 29:58 (1900). Brettell, Phytologia 28(1):58 (1974). Oligactis coriacea (Hieron.) H. Rob. & Distribution.—Peru. Brettell, Phytologia 28(1):58 (1974). Liabum scandens Domke in Diels, Bib- Sampera pastoensis (Cuatrec.) V. A. Funk loth. Bot. 116:167(1937). & H. Rob., comb. nov. Oligactis scandens (Domke) H. Rob. & Brettell, Phytologia 28(1):59 (1974). Liabum pastoense Cuatrec, Not. Fl. Liabum granatense Cuatrec, Feddes Re- Colomb. 6:36 (1944). pert. Spec. Nov. Regni Veg. 55:128 Oligactis pastoensis (Cuatrec.) H. Rob. & (1953). Brettell, Phytologia 28(1):58 (1974). Oligactis granatensis (Cuatrec.) H. Rob. Distribution.—Colombia. & Brettell, Phytologia 28(1):58 (1974). Oligactis coriacea var. granatensis (Cua- Sampera pichinchensis (Hieron.) V. A. trec.) H. Rob., Phytologia 35(3):201 Funk & H. Rob., comb. nov. (1977). Distribution.—Colombia, Ecuador, Liabum pichinchense Hieron., Bot. Jahrb. Peru. Syst. 29:56 (1900). VOLUME 122, NUMBER 2 159

rf,/?.-fffi6ref?/A)/ zocy

Fig. 1. Sampera coriacea (Hieron.) V. A. Funk & H. Rob. A. Habit, note pinnate venation. B. Head, note pubescence on involucral bracts. C. Ray corolla and style. D. Disk corolla, anthers and style. Drawing by Alice Tangerini (US). Mostly from Ollgaard & Balslev 8277. 160 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON

>fTZ{xfT

Fig. 2. Sampera coriacea (Hieron.) V. A. Funk & H. Rob. A. Longitudinal section of disc corolla showing bases of anther thecae. B. Style showing pubescence on the upper part of the style. C. Achene and pappus showing glands and hairs on achene and two types of pappus. Drawing by Alice Tangerini (US).

Oligactis pichinchensis (Hieron.) H. Rob. Cassini, H. 1830. Zyegee 60. Pp. 506-519 in G. & Brettell, Phytologia 28(1):59 (1974). Cuvier, ed., Dictionnaire des sciences natur- elles, Paris, [reprinted in R. M. King & H. W. Liabum hallii Hieron., Bot. Jahrb. Syst. Dawson, eds., 1975, Cassini on Compositae, 29:57 (1900). New York, Oriole Editions]. Oligactis hallii (Hieron.) H. Rob. & Dillon, M., V. Funk, H. Robinson, & R. Chan. Brettell, Phytologia 28(1):58 (1974). 2009. Liabeae. In V. A. Funk, A. Susanna, T. Distribution.—Ecuador. Stuessy and R. Bayer, eds., Systematics, Evolution, and Biogeography of the Compos- itae (in press). Acknowledgments Funk, V. A., & H. Robinson. 2001. A bully new genus from the Andes (Compositae: Lia- Thanks go to Alice Tangerini for her beae).—Systematic Botany 26(2):216-225. , , & M. O. Dillon. 1996. The Liabeae: lovely illustration and to two anonymous , Phylogeny, and Biogeography. reviewers. Pp. 545-567 in D. J. N. Hind, H. J. Beentje and P. D. S. Caligari, eds., Proceedings of the International Compositae Conference, Kew, Literature Cited 1994, Vol. 1. Compositae: Systematics, Kew Publishing, Kew. Bentham, G. 1873. Ordo 88: Compositae. 2(1) Hoffmann, O. 1893-1894. Compositae, 4(5). 163-533, 536, 537 in G. Bentham, & J. D. Pp. 87-387 in A. Engler and K. Prantl, eds., Hooker. Genera Plantarum. London, Lovell Die Natiirlichen Pflanzenfamilien. 4 Teil, Reeve. Nacht. 1^1. Leipzig. VOLUME 122, NUMBER 2 161

Robinson, H. 1983a. A generic review of the tribe , & V. A. Funk. 1987. A phylogenetic Liabeae (Asteraceae).—Smithsonian Contri- analysis of Leiboldia, Lepidonia, and a new butions to Botany 54:i-iv, 1-69. genus Stramentopappus (Vernonieae: Astera- . 1983b. Studies in the Liabeae (Asteraceae). ceae).—Botanische Jahrbiicher fur System- XVI. New taxa from Peru.—Phytologia atic, Pflanzengeschichte und Pflanzengeogra- 54(l):62-65. phie 108:213-228. . 1990. A redelimitation of Microliabum Rydberg, P. A. 1927. (Carduales) Carduaceae, Cabrera (Asteraceae: Liabeae).—Systematic Liabeae, Neurolaeneae, Senecioneae (pars). Botany 15(4):736-744. —North American Flora 34(4):289-360. , & R. D. Brettell. 1974. Studies in the Liabeae (Asteraceae), II: Preliminary survey of the genera.—Phytologia 28(l):43-63. Associate Editor: Carol Hotton.