ANNALS OF THE MISSOURI BOTANICAL GARDEN VOLUME 84 1997 Colophon The Annals is printed and distributed by Allen Press, Inc. of Lawrence, Kansas 66044, U.S.A. © Missouri Botanical Garden 1997 ISSN 0026-6493 1997 ALMEDA, FRANK. Chromosomal Observations on the Alzateaceae (Myrtales) — 305 BACIGALUPO, NELIDA M. (See Elsa L. Cabral y N6lida M. Bacigalupo) 857 BENITEZ DE ROJAS, CARMEN & WILLIAM G. D’ARCY. The Genus Ly- cianthes (Solanaceae) in Venezuela 167 BERRY, PAUL E. Book Review. Guide to the Vascular Plants of Central French Guiana. Part 1 by Scott Mori et al. 907 BLACKMORE, STEPHEN (See Sandra Knapp, Viveca Persson & Stephen Blackmore) __ ■ 67 BROWN, K. D. (See J. F. Smith, J. C. Wolfram, K. D. Brown, C. L. Carroll & D. S. Denton) 50 CABRAL, ELSA L. Y NELIDA M BACIGALUPO. Revision del Genera Gal- ianthe subg. Ebelia stat. nov. (Rubiaceae: Spermacoceae) 857 CARR, GERALD D. (See Harold Robinson, Gerald D. Carr, Robert M. King & A. Michael Powell) 893 CARROLL, C. L. (See J. F. Smith, J. C. Wolfram, K. D. Brown, C. L. Carroll & D. S. Denton) 50 CHASE, MARK W. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nick- rent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 CHAW SHU-MIAW (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nick- rent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) — 1 CROAT, THOMAS B. A Revision of Philodendron Subgenus Philodendron (Ar- aceae) for Mexico and Central America 311 DENTON, D. S. (See J. F. Smith, J. C. Wolfram, K. D. Brown, C. L. Carroll & D. S. Denton) - — __ — — — 50 D’ARCY, WILLIAM G. (See Carmen Benftez de Rojas & William G. D’Arcy) D’ARCY, WILLIAM G. A Review of the Genus Eccremocarpus (Bignoniaceae) 103 FRITSCH, PETER W. A Revision of Styrax (Styracaceae) for Western Texas, Mexico, and Mesoamerica — — 705 GILLESPIE, LYNN J. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nick- rent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 GOLDBLATT, PETER & MASAHIRO TAKEI. Chromosome Cytology of Iri- daceae — Patterns of Variation, Determination of Ancestral Base Num- bers, and Modes of Karyotype Change 285 GOLDBLATT, PETER & ANNICK LE THOMAS. Palynology, Phylogenetic Re- construction, and Classification of the Afro-Madagascan Genus Aristea (Iridaceae) ... — 263 GU HONG-YA & PETER C. HOCH. Systematics of Kalimeris (Asteraceae: Astereae) 762 HAHN, WILLIAM J. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nick- rent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 HAUK, WARREN D. A Review of the Genus Cydista (Bignoniaceae) 815 HILU, KHIDIR, W. & JOHN L. JOHNSON. Systematics of Eleusine Gaertn. (Poaceae: Chloridoideae): Chloroplast DNA and Total Evidence 841 HOCH, PETER C. (See Gu Hong-ya & Peter C. Hoch) 762 HOOT, SARA B. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 JOHNSON, JOHN L. (See Khidir W. Hilu, & John L. Johnson) 841 JOHNSON, LEIGH A. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 5 1 KING, ROBERT M. (See Harold Robinson, Gerald D. Carr, Robert M. King & A. Michael Powell) 893 KNAPP, SANDRA, VIVECA PERSSON & STEPHEN BLACKMORE. A Phy- logenetic Conspectus of the Tribe Juanulloeae (Solanaceae) 67 KRESS, W. JOHN (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 KRON, KATHLEEN A. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 KUZOFF, ROBERT K. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 LE THOMAS, ANNICK (See Peter Goldblatt & Annick Le Thomas) 263 LI JIE (See Li Xi-wen & Li Jie) 888 LI XI-WEN & LI JIE. The Tanaka-Kaiyong Line — An Important Floristic Line for the Study of the Flora of East Asia 888 MEDAIL, FREDERIC & PIERRE QUEZEL. Hot-Spots Analysis for Conser- vation of Plant Biodiversity in the Mediterranean Basin 112 NICKRENT, DANIEL L. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 PERSSON, VIYECA (See Sandra Knapp, Viveca Persson & Stephen Black- more) 67 PIESSCHAERT, FREDERIC, ELMAR ROBBRECHT & ERIK SMETS. Dialy- petalanthus fuscescens Kuhlm. (Dialypetalanthaceae): The Problematic Taxonomic Position of an Amazonian Endemic 201 PIRE, STELLA MARIS. G6nero Galianthe subg. Ebelia (Rubiaceae: Sperma- coceae): Estudio Palinoldgico 878 POWELL, A. MICHAEL (See Harold Robinson, Gerald D. Carr, Robert M. King & A. Michael Powell) 893 QUEZEL, PIERRE (See Fr6d6ric M6dail & Pierre Qu6zel) _ 112 ROBBRECHT, ELMAR (See Frederic Piesschaert, Elmar Robbrecht & Erik Smets) „ — 201 ROBINSON, HAROLD, GERALD D. CARR, ROBERT M. KING & A. MI- CHAEL POWELL. Chromosome Numbers in Compositae, XVII: Sene- cioneae III 893 ROHWER, JENS G. The Fruits of Jasminum mesnyi (Oleaceae), and the Dis- tinction Between Jasminum and Menodora _____ 848 SCHUTTE, ANNE LISE. A Revision of the Genus Xiphotheca (Fabaceae) _ 90 SMETS, ERIK (See Frederic Piesschaert, Elmar Robbrecht & Erik Smets) __ 201 SMITH, J. F., J. C. WOLFRAM, K. D. BROWN, C. L. CARROLL & D. S. DENTON. Tribal Relationships in the Gesneriaceae: Evidence from DNA Sequences of the Chloroplast Gene ndhF _ — 50 SOLTIS, IJOUGLAS E., PAMELA S. SOLTIS, DANIEL L. NICKRENT, LEIGH A. JOHNSON, WILLIAM J. HAHN, SARA B. HOOT, JENNIFER A. SWEERE, ROBERT K. KUZOFF, KATHLEEN A. KRON, MARK W. CHASE, SUSAN M. SWENSEN, ELIZABETH A. ZIMMER, SHU-MIAW CHAW, LYNN J. GILLESPIE, W. JOHN KRESS & KENNETH J. SYTSMA. Angiosperm Phylogeny Inferred from 18S Ribosomal DNA Sequences — 1 SOLTIS, PAMELA S. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nick- rent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 SWEERE, JENNIFER A. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 SWENSEN, SUSAN M. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 SYTSMA, KENNETH J. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 TAKEI, MASAHIRO (See Peter Goldblatt & Masahiro Takei) 285 TAYLOR, CHARLOTTE M. Conspectus of the Genus Palicourea (Rubiaceae: Psychotrieae) with the Description of Some New Species from Ecuador and Colombia - 224 THOMPSON, MAXINE M. Survey of Chromosome Numbers in Rubus (Rosa- ceae: Rosoideae) ; 128 WOLFRAM, J. C., (See J. F. Smith, J. C. Wolfram, K. D. Brown, C. L. Carroll & D. S. Denton) . 50 ZIMMER, ELIZABETH A. (See Douglas E. Soltis, Pamela S. Soltis, Daniel L. Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kuzoff, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma) 1 Annals • fiVS73 teF of the Missouri Botanical Garden 1997 S* Volume 84 Number 1 Volume 84, Number 1 Winter 1997 Annals of the Missouri Botanical Garden The Annals, published quarterly, contains papers, primarily in systematic botany, contributed from the Missouri Botanical Garden, St. Louis. Papers originating out- side the Garden will also be accepted. Authors should write the Managing Editor for information concerning arrangements for publishing in the Annals. Instructions to Authors are printed in the back of the last issue of each volume. Editorial Committee Henk van der Werff Editor, Missouri Botanical Garden Ihsan A. Al-Shehbaz Missouri Botanical Garden Amy Scheuler McPherson Managing Editor, Gerrit Davidse Missouri Botanical Garden Missouri Botanical Garden Diana Gunter Roy E. Gereau Missouri Botanical Garden Editorial Assistant, Missouri Botanical Garden Peter Goldblatt Missouri Botanical Garden Vicki Couture Secretary Gordon McPherson Missouri Botanical Garden Teresa Johnson Publications Order Processor P. Mick Richardson Missouri Botanical Garden For subscription information contact Depart- ment Eleven, P.0. Box 299, St. Louis, MO 63166-0299. Subscription price is $110 per volume U.S., $115 Canada and Mexico, $135 all other countries. Four issues per volume. The journal Novon is included in the subscription price of the Annals. amcpher@admin.mobot.org (editorial queries) deptll@mobot.org (orders) http://www.mobot.org © Missouri Botanical Garden 1997 The Annals of the Missouri Botanical Garden (ISSN 0026-6493) is published quar- terly by the Missouri Botanical Garden, 2345 Tower Grove Avenue, St. Louis, MO 63110. Pe- riodicals postage paid at St. Louis, MO and ad- ditional mailing offices. POSTMASTER: Send ad- dress changes to Annals of the Missouri Botanical Garden, Department Eleven, P.O. Box 299, St. Louis, MO 63166-0299. The mission of the Missouri Botanical Garden ii their environment, in order to preserve and enri s to discover and share knowledge about plants and ich life. @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Volume 84 Number 1 1997 Missouri Botanical Garden ANGIOSPERM PHYLOGENY INFERRED FROM 18S RIBOSOMAL DNA SEQUENCES 1 Kuzoff , 2 j i ji i . j a i iiijjii liiiii 3 I Jill I I 111 II i mini min 1 1 mm i I l I s My 8 41* s 5 b 55 5 1 js! i-8 liiliSS liiiii i i !iii i ! m i! iiliiil ill 1 1 in! 1 1 ill !> iiill Volume 84, Number 1 1997 Soltis et al. 18S Ribosomal DNA Phytogeny 6 Annals of the Missouri Botanical Garden ‘ DNA Phytogeny ► H j lliliillJiJ liu ii II I Iili I II Illllllilll 1111 II ill I IIII I i- llipi «!i Is * ill 5 J»ISJitiiH M i! !i! in ]•]]]]>] i t r DNA Phylogeny , I.. ! . JI . I J . HI iSIliluil iiinJ lilliiiiliJ 111 IHIIillll IIIIII 1IIIIIII1II i i i Ji |i ji ill!! ill ill ufuJiiiillli Iiiiil fjiuiiljli ih i« mi ‘i iiiiil III! li i si i ns Him ms t< . t t ii* i* 1» i* 1 J, 5 I, 3 1, ! I Pll'P ! II il-U j Il'ii i s 1 &= a ! 1 ?, IU S ;l iJJ I? 1 1 j 1 1 u i»i » a ir-s nun Illn 1 II II I] i !i 11 1 j i i & ll. i i f i L Illl li i iin 11 ■Kill 111 Jill Si >i Volume 84, Number 1997 Soltis et al. 18S Ribosomal DNA Phylogeny cn Z X ilj ! I I 1 ! I null !!! ill nn lilt I § i-i-i uli ■ -8 I il =3 c $ c/3 c$ c/3 c$ c$ c/) c$ 3 1 §§ II I J * & I I 1 II |2l2i- ^ *1 S V ^ Sill I n mill i 1 Ui'llfll j If 1 1 1 1 f 1 = | Garden . i | iLiii H'l'l I Hi Hilt t U it iSJ ! iiij iliiiSS! SiiilSSI S! ill ill! M5 i i 5 1 h] ti«i ! i? |i ; i A i i i i SiMlI I IS ii I Mi Hill lil 1 ii ii ill ill Mi §1 .1 n ii atiiii 14 Garden Annals of the Missouri Botanical Garden Table 3. Area initially thought to be prone to insertion ing difficulties. The underlined portion of the Hydrangea 215 Podophyllum AAAGGTTGACGCGGGCTTTGCCC AAAGGTCAACACAGGCTCTGCCT AAAGGCCAAC GCTTTGCCC AAAGGTCAACGCTTGCTTCGGCT AAAGGCCAAC GCTCTGCCC AAAGGTCGAC GCTTTGCCC AAAGGTCAACG GCTTTGCCC AAAGGCCGAT- -CGGCTCTGCCC AAAGGTCAAC ?????? 7CTGCCC AAAGGTCGA TGCC- ent in the gene itself and may lead to compressions and associated sequencing problems. More than one molecular systematist with substantial experi- ence in the sequencing of chloroplast genes such as rbcL has referred to the sequencing of 18S rDNA as “tricky.” We have found that preparation of sam- ples via cycle sequencing followed by automated sequencing yields reliable 18S rDNA sequences that appear more accurate than most manually gen- erated sequences. The critical procedural step is likely the cycle sequencing reactions, in which sec- ondary structure is reduced or eliminated by high temperature. Several specific regions of 18S rDNA are particularly difficult to sequence. These include base positions 215-230, 1355-1365, and 1705- 1715 (all positions mentioned in this paper corre- spond to those of Glycine max; Eckenrode et al., 1985), as well as several of those small regions not- ed earlier that are prone to insertion and deletion (positions 230-237; 496-501; 666-672; 1363- 1369) (see Appendix). We will use the first of these regions (base po- sitions 215-230) to illustrate the errors that can result in 18S rDNA sequencing. Based on manual sequences (generated by D. Soltis & R. Kuzoff), the base composition of this region in Saxifragaceae and several other rosid families initially appeared to involve a large deletion relative to some other available sequences (see Table 3). Similarly, the 18S rDNA sequences generated manually by other investigators, representing a diversity of angio- sperms, typically were lacking one or more base pairs in this region. Alternatively, researchers scored this region as uncertain, using either “?” or “N.” Thus, sequences available prior to this study suggested that this region was highly prone to in- sertion and deletion. As a result, alignment of this region was initially difficult. Alignment problems were exacerbated by the apparent occurrence of base substitution in the region. Further compound- ing the difficulty of alignment is the fact that the region just 3' of this area actually is prone to in- sertion-deletion, as well as to considerable varia- tion in primary sequence. Base positions 230-237 correspond to one of the variable helix termini not- ed above. Cycle and automated sequencing of over 100 taxa, however, revealed no indels in the area of positions 215-230. In fact, after resequencing this region in many taxa for which manual sequenc- es initially suggested the presence of numerous in- dels, we have concluded that indels in this region are either extremely rare or nonexistent. This region is G-C rich; as a result, sequencing “stops” often occurred, yielding only a portion of the base pairs actually present in the region. Alignment of these incomplete sequences suggested numerous indels in this region, leading to the misconception that indels were frequent. Similar sequencing problems were encountered in other portions of the 18S rRNA gene. Taken together, these regions had con- tributed to the view that insertion and deletion are common in 18S rDNA. Although the frequency of indels has been over- stated for the 18S gene, several regions of 18S rDNA are, in fact, prone to variation in primary sequence and length. However, these regions are small, easily located, and, as noted by Nickrent and Soltis (1995), typically confined to the termini of helices on the proposed secondary structure model for 18S rRNA (e.g., Nickrent & Soltis, 1995). Four such regions, represented by base positions 230- 237, 496-501, 666-672, and 1363-1369, corre- spond to the termini of helices E10-1, 17, E23-1, and 43, respectively (see Appendix). These regions are difficult to align over a broad taxonomic scale, such as all angiosperms, and were therefore not used in our phylogenetic analyses (see Materials and Methods above). On a lower taxonomic scale (e.g., closely related families), however, even these highly variable regions are easily aligned, permit- ting the use of these regions in more focused stud- ies (e.g., Polemoniaceae and related Asteridae S.I., Johnson et al., unpublished; portions of Saxifraga- ceae s.l., D. Soltis & Soltis, 1997; Orchidaceae, Cameron & Chase, unpublished). Because indels in 18S rDNA are neither as prev- alent nor as problematic as previously thought, alignment of clean 18S rDNA sequences is straight- forward. With the exclusion of the few small regions noted above, alignment of over 200 angiosperm se- quences was straightforward and easily accom- 17 '..IIS 84, Number 1 Volume 84, Number 1997 Soltis et al. 18S Ribosomal DNA Phylogeny related to the glucosinolate clade in any of the four searches. Phylogenetic analyses of 18S rDNA se- quences involving additional glucosinolate taxa fur- ther demonstrate the monophyly of the glucosino- late-producers, with the exception of Drypetes, and also clarify relationships among the members of this clade (Rodman et al., submitted). These results closely parallel findings based on the phylogenetic analysis of rbcL sequences (Rodman et al., 1993; Chase et al., 1993) and morphology (Rodman, 1991). Thus, both rbcL and 18S rDNA sequence data indicate that there were two independent ori- gins of the mustard oil syndrome (see Rodman et al., 1993, submitted). Nitrogen-fixing clade. Species of only 10 fam- ilies of angiosperms are known to form symbiotic associations with nitrogen-fixing bacteria in root nodules (Fabaceae, Betulaceae, Casuarinaceae, Coriariaceae, Datiscaceae, Elaeagnaceae, Myrica- ceae, Rhamnaceae, Rosaceae, and Ulmaceae). These families are distributed among four of Cron- quist's (1981) six subclasses of dicotyledons, im- plying that many of these families are only distantly related. Recent phylogenetic analyses of r6cL se- quences reveal, however, that representatives of all ten of these families occur together in a single clade (‘‘nitrogen-fixing clade"; Soltis et al., 1995). In addition to these ten families, this clade also contains several families not known to form asso- ciations with nitrogen-fixing bacteria, lacluding Moraceae, Cannabaceae, Urticaceae, Polygalaceae, Fagaceae, Begoniaceae, and Cucurbitaceae. Analyses of three of four 18S rDNA data sets (Figs. 1, 2, 4) suggest an alliance of taxa similar to that revealed by rbcL sequences. This clade in large part represents a subset of the taxa present in the r6cL-based nitrogen-fixing clade. The fami- lies in the 18S rDNA-based nitrogen-fixing clade include Betulaceae, Casuarinaceae, Datiscaceae, Elaeagnaceae, Rhamnaceae, and Ulmaceae, all families that form symbiotic associations with ni- trogen-fixing bacteria. Other families known to form such associations (i.e., Coriariaceae and Myrica- ceae) and that appeared in the rtcL-baaed nitrogen- fixing clade were not analyzed for 18S rDNA se- quence variation. Also part of the nitrogen-fixing clade retrieved here are Begoniaceae, Moraceae, Urticaceae, and Cucurbitaceae, families also found to be part of this alliance based on analyses of r&cL sequences. However, neither Rosaceae nor Faba- ceae, two families involved in nitrogen-fixing sym- bioses, are included within the 18S rDNA nitrogen- fixing clade, although both families are part of this alliance in the rhrL-based trees (Soltis et al., 1995). Searches involving the two larger data sets (1 and 2) also place three families of Malvales (Mal- vaceae, Bombacaceae, and Tiliaceae) within the ni- trogen-fixing clade; these taxa were not part of the nitrogen-fixing clade in the rfccL-based trees. In analyses of data set 4, however, these three families of Malvales are not part of the nitrogen-fixing clade (Fig. 4). No clear nitrogen-fixing clade emerged in analyses of data set 3; instead, these taxa are part of a grade that represents the first branches of a primarily rosid-dilleniid clade (Fig. 3). Asteridae sensu Into. Analyses of all four 18S rDNA data sets also reveal an expanded Asteridae clade (Asteridae s.l.) that agrees closely with that recovered by analyses of rhcl. sequences (Olmslead et al.. 1992, 1993; Chase et al.. 1993). In addition to the conventionally circumscribed Asteridae, this clade also includes a numlier of families placed in Dilleniidae, such as Ericaceae, Clethraceae, Pyro- laceae, Styracaceae, Ebenaceae, Actinidiaceae, Sarraceniaceae. Fouquieriaceae, Theaceae, ami Primulaccae. Also present in Asteridae s.l. are Nys- saceae, Pittosporaceae, Apiaceue, Araliaceae, and Hydrangeaceae, all members of Rosidae. In addi- tion. Eucommiaceae, a memlrer of llamamelidae. and Byblis, a genus of carnivorous plants usually placed in Rosidae, also appear within Asterirlae s.l. All analyses also place an expanded Caryophylli- dae (Caryophvllidae S.L) within the Asteridae s.l. clade, an unexpected result that is discussed in more detail below. Within Asteridae s.l., several subclades or grades can be identified that agree, in large part, with some of the groups identified in analyses of rbcL sequences (Chase et al., 1993; Olmslead et al., 1993). Perhaps most noteworthy of these is the ericalean grade (the asterid III clade of Chase et al., 1993) observed in all of the shortest 18S rDNA trees (Figs. 1-4). Other claries of Olmslead et al. are also observed to be monophyletic, including Dipsacales, Boraginales, Genlianales, Asterales s.l., and Lamiidae. Additional asterid taxa should be sequencer! for I8S rDNA to assess more rigor- ously the monophyly of these groups and their in- terrelationships. Caryophyllidae sensu lata. All analyses of 18S rDNA sequences reveal a clade composed of Nyc- taginaceae ( Mirabilis ), Chenopodiaeeae (Spinacia), Phvtolaccaceae ( Phytolacca ), Aizoaceae (Tetrago- nal), and Mollugmaceae (Molklgo). These five fam- ilies represent Caryophyllales (e.g., Cronquist, 1981), the monophyly of which is supported in this study by a jackknife value of 58%. as well as by numerous lines of morphological ami molecular 24 Annals of the Missouri Botanical Garden data (e.g., Rodman et al., 1984; Rettig et al., 1992). Sister to this clade of Caryophyllales is another strongly supported clade comprising Plumbagina- ceae and Polygonaceae (jackknife value of 77%): this group collectively represents Caryophyllidae (sensu Cronquist, 1981). The monophyly of Cary- ophyllidae is only weakly supported by cladistic analysis of morphological, chemical, anatomical, and palynological features (Rodman et al., 1984). Analyses of 18S rDNA sequences also suggest that two families of carnivorous plants, Droseraceae and Nepenthaceae, are sister to Caryophyllidae, and we refer to this entire assemblage as Caryophyllidae s.l. (Figs. 1-4). Phylogenetic analyses of rbcL sequences simi- larly recovered a Caryophyllidae s.l. clade com- posed of Caryophyllales, Polygonaceae, Plumbagi- naceae, Droseraceae, and Nepenthaceae (Chase et al., 1993). One of the broad analyses of rbc L se- quences (search A, Chase et al., 1993) placed Vi- taceae and Dilleniaceae with this expanded Cary- ophyllidae clade. In the analyses of 18S rDNA sequences, Vitaceae were not sampled, and Dille- niaceae are well removed from Caryophyllidae s.l. The anomalous placement of Dilleniaceae near the monocots (Figs. 1, 2) is discussed below. Santaloids. Analyses of all four data sets reveal a monophyletic santaloid clade or Santalales, which are represented here by only three families (Opi- liaceae, Santalaceae, and Viscaceae). However, in preliminary analyses in which Santalales are rep- resented by seven families (Opiliaceae, Santala- ceae, Viscaceae, Eremolepidaceae, Misodendra- ceae, Loranthaceae, and Olacaceae), santaloids again form a clade. These seven families are widely considered to form a natural group based on mor- phology (e.g., Cronquist, 1981) and have been shown to form a clade in previous, smaller analyses of 18S rDNA sequences (Nickrent & Franchina, 1990; Nickrent & Soltis, 1995). Although santaloids appear monophyletic, the position of this clade varies among the analyses. In analyses of data sets 1 and 2, santaloids are sister to Poly gala and closely related to the legumes. Analysis of data set 4 again places santaloids with Polygala and a legume (Pisum), as well as with Gunnera. Analysis of data set 3 results in an un- usual placement of santaloids with several paleo- herbs. These findings parallel those of Chase et al. (1993) based on rbcL sequences in which the po- sition of santaloids differed greatly between the 476- and 499-taxon searches. In the former, san- taloids and Gunnera form the asterid V clade; in the latter, santaloids are sister to the caryophyllids. but again appear near Gunnera. Thus, whereas both rbcL and 18S rDNA searches occasionally place santaloids near Gunnera, analyses of three of the four 18S rDNA data sets place santaloids close to Fabaceae and Polygalaceae. Celastroids. Another small clade revealed in all analyses consists of Lepuropetalon and Parnassia (Saxifragaceae s.l.), Brexia (Grossulariaceae), and Euonymus (Celastraceae). This clade, labeled ce- lastroids (Figs. 1-4), was also recovered in analyses of rbcL sequences (Chase et al., 1993; Morgan & Soltis, 1993). Although this initially appears to be an eclectic assemblage (Brexia is a genus of small trees; Lepuropetalon spathulatum is the smallest terrestrial angiosperm), embryological and morpho- logical data also unite these taxa (reviewed in Mor- gan & Soltis, 1993). The celastroid clade consists of two pairs of genera, each of which is strongly supported: Lepuropetalon-Parnassia (jackknife = 100%) and Brexia— Euonymus (jackknife = 67%). These same two pairs of genera also were revealed in analyses of rbcL sequences (Chase et al., 1993; Morgan & Soltis, 1993). Cunonioids. Bauera and Ceratopetalum (Cu- noniaceae) and Eucryphia (Eucryphiaceae) form a clade with a jackknife value of 53%. A close re- lationship among these genera also was revealed by a cladistic analysis of morphological features (Huf- ford & Dickison, 1992). Bauera, Ceratopetalum, and Eucryphia constitute the core of a very well supported clade (jackknife value of 89%) labeled cunonioids (Figs. 1—1) that also contains Cephalo- taceae, a family of carnivorous plants, and Sloanea (Elaeocarpaceae). A close relationship of Cephal- otaceae to these same representatives of Cunoni- aceae and Eucryphiaceae also is suggested by anal- yses of rbcL sequences (Chase et al., 1993; Morgan & Soltis, 1993). Sloanea was not represented in the broad analyses of r&cL sequences. Other taxa that appear closely allied with Cunoniaceae, Eucryphi- aceae, and Cephalotaceae in rbcL analyses include Tremandraceae and Oxalidaceae; these families were not included, however, in the 18S rDNA anal- Other noteworthy relationships. As recently re- viewed (Qiu et al., 1993), the placement of Lacto- ridaceae has been controversial, with relationships to Magnoliales, Laurales, and Piperales all pro- posed. Analyses of rbcL sequences suggested a close relationship of Lactoridaceae to Aristolochi- aceae (Chase et al., 1993), and analyses of 18S rDNA sequences similarly suggest that these two 84, Number 1 Garden Volume 84, Number 1 1997 Soltis et al. 18S Ribosomal DNA Phylogeny 27 1, 2, 4), as well as in the study of Hamby and Zimmer (1992), Ceratophyllum is allied with the monocots. Both studies also concur in suggesting that Nymphaeaceae appear near the base of the an- giosperm radiation. Nymphaeaceae are the sister group to all other angiosperms in Hamby and Zim- mer’s (1992) shortest trees; however, Amborella- ceae, Austrobaileyaceae, and Schisandraceae were not included in that study. In all of our shortest trees, Nymphaeaceae follow the latter three families and Illiciaceae as the sister group to all remaining flowering plants. Another similarity between the shortest trees in both studies is the placement of Drimys (Wintera- ceae). Drimys occupies an unusual phylogenetic position in trees presented by both Hamby and Zimmer (1992) and Nickrent and Soltis (1995), ap- pearing as sister to Glycine and Pisum (Fabaceae), rather than as an early-branching angiosperm. Dri- mys occupies an unusual position in trees derived from the current analyses, as well, appearing among the lower eudicots. In trees resulting from the anal- ysis of data set 3, Drimys again appears with Pisum. The 18S rDNA sequence of Drimys exhibits a num- ber of substitutions not found in other magnoliids. In an attempt to ascertain the relationships of Win- teraceae, we sequenced two species of Drimys, D. winteri and D. aromatica, and they have identical sequences. More recently, another member of Win- teraceae (PseudmvinUra) has been sequenced for 18S rDNA (Hoot, unpublished); this sequence is nearly identical to the sequences for Drimys. Add- ing Pseudowintera to the analysis does not alter the unusual position of Winteraceae (trees not shown). The unusual phylogenetic relationships that exist among the eudicots in the shortest trees of Hamby and Zimmer (1992) probably derive from insuffi- cient sampling in that study. The present analysis with its greater representation of eudicots reveals relationships much more in accord with recent clas- sifications (e.g., Cronquist, 1981; Takhtajan, 1987) and/or the rbcL topologies of Chase et al. (1993). Thus, the present study suggests that many of the highly unusual relationships seen in Hamby and Zimmer are likely to reflect low taxon density rather than an inherent inability of 18S rDNA sequences to resolve relationships. CAVEATS A number of limitations are inherent in any large phylogenetic study such as this. Several factors may contribute to the anomalous positions of certain taxa, including uncertainty regarding maximum parsimony, insufficient taxon sampling and/or den- sity, the presence of “older,” erroneous 18S rDNA sequences in the data matrix, and the overall lower rate of evolution of 18S rDNA compared to rbcL. We discuss these potential factors in more detail An analysis of this magnitude cannot be expect- ed to achieve maximum parsimony in a reasonable amount of time. It is likely that we did not find all classes of most-parsimonious trees, despite a search strategy (cf. Maddison et al., 1992) designed to identify multiple islands (Maddison, 1991) of shortest trees, and that even shorter trees exist that were not recovered. Furthermore, ulthough our search strategy involved well over two years of com- puter time, no search swapped to completion; there is no assurance, therefore, that these trees repre- sent even a local parsimony optimum. Although it is, of course, impossible to know how far from com- pletion any search is when it is truncated, the search design used here offers an insightful basis for comparison. Data sets 1 and 2, and 3 and 4 are identical except for the inclusion of two gap char- acters (indels) in data sets 2 and 4, each of which apparently accounts for only four steps on the shortest trees obtained. Thus, the fact that the shortest trees obtained in searches of data set 2 are seven steps longer than those obtained in searches of data set 1 indicates that the shortest trees ob- tained in our searches of data set 2 are three steps less parsimonious than trees derived from searches of data set 1. A similar comparison of the searches of data sets 3 and 4 reveals that the shortest trees from searches of data set 4 are two steps less par- simonious than those obtained from data set 3. We also sampled among the large set of equally parsimonious trees following Sanderson and Doyle (1993b). Using trees obtained in searches of data set 1, we examined the number of distinct compo- nents (clades) as a function of the size of the sample of trees (number of trees). We wanted to determine whether increasing the set of trees uncovers new components that bear on the relationships of par- ticular taxa or, in contrast, includes different sub- sets of the components that are essentially varia- tions on the same theme (Sanderson & Doyle, 1993b). We found that a plot of the number of dis- tinct clades versus the number of trees sampled reaches an asymptote for a small number of trees, suggesting that most of the clade diversity has been found, despite the fact that all most parsimonious trees have not been retrieved. The development of improved methods of phylogenetic analysis of large data sets will ultimately be one of the central issues of phylogeny reconstruction during the next several years (see discussions in Chase et al., 1993; Doyle Annals of the Missouri Botanical Garden et al., 1994; Mishler, 1994; P. Soltis & Soltis, 1997). Although the anomalous relationships described for some taxa may be unsettling, extremely short branches characterize most of the major clades in the 18S rDNA trees. The internal support for many branches is very low, as indicated by the parsimony jackknife analysis (Farris et al., 1997). Although the monophyly of the angiosperms is well supported (jackknife value of 100%), few major clades within the angiosperms have high jackknife values. For example, large clades such as eudicots and Rosidae do not have jackknife values above 50%; the sax- ifragoids represent the largest clade having a high jackknife value (jackknife value of 68%). The other monophyletic groups with high jackknife values are relatively small, such as cunonioids, Zingiberales, Malvales, Caryophyllales, Lactoridaceae-Aristolo- chiaceae, and Schisandraceae-Illiciaceae-Austro- baileyaceae. Significantly, a number of major clades seen in all shortest trees, as well as in trees many steps longer than the most parsimonious trees, do not have jackknife values above 50%, including monocots, glucosinolates, Caryophyllidae s.l., and Asteridae s.l. The majority of high jackknife values correspond to pairs of sister taxa representing ter- minal nodes (e.g., Calycanthus— Sassafras, Brexia— Euonymus, Lepuropetalon-Pamassia, Plumbago- Cacoloba, Helwingia-Phyllonoma, Tragopogon- Tagetes, Francoa-Greyia, Trochodendron-Telracen- tron, Menisftermum-Tinospora). Examination of trees obtained from searches that found trees one or a few steps longer than the short- est trees also suggests low internal support for some branches. The phylogenetic position of the mono- cots appears weakly supported. In some searches of data set 2, for example, trees only one step longer than the shortest trees place the monocots within the eudicots, as part of Rosidae, a position also observed in the shortest trees obtained from search- es of data set 3 (Fig. 3). Although all of the starting trees and shortest trees showed Amborellaceae, 11- liciaceae. Schisandraceae, and Austrobaileyaceae to be at the base of the angiosperms, one search of data set 2 resulted in trees two steps longer than the shortest trees and placed these four families near the monocots, with Acorns and Oncidium as the first-branching angiosperms. Trees two steps longer than the shortest trees show the Asteridae s.l. embedded within Rosidae, rather than sister to this large clade. In trees two steps longer than the shortest trees found for data set 3, Caryophyllidae s.l. are not part of Asteridae s.l. but instead are part of the large Rosidae clade. These few examples illustrate well the uncertain- ty that surrounds some angiosperm relationships in- ferred from analyses of 18S rDNA sequences. Fur- thermore, because relatively few character-state changes occur on many of the branches, a small amount of homoplasy or error in the data set may be sufficient to distort some relationships. Additionally, some of the anomalous placements could reflect insufficient and/or uneven taxon sam- pling. The somewhat uneven taxonomic distribution of the sequences presently available means that some groups, such as Asteridae, and much of Ros- idae and Hamamelidae,, are relatively well repre- sented here, whereas Magnoliidae, the monocots, Dilleniidae, Caryophyllidae, and several orders of Rosidae are under-represented. The importance of sufficient taxon density is re- vealed here by some of the differences in topology observed between trees resulting from analyses of the smaller and larger data sets. Many of the taxa not present in the two smaller data sets (3 and 4) represent monosulcates and lower eudicots. It is this portion of the overall topology that shows the most spurious relationships in trees derived from analyses of these two small data sets (the distinc- tion between the monosulcate grade and eudicots largely breaks down in Fig. 3, for example). In con- trast, the much more thoroughly represented Aster- idae s.l. and Rosidae clades are little affected by slightly decreased representation in data sets 3 and 4. These findings lend further support to the im- portance of sufficient and equal taxon density in attempts to infer angiosperm phylogeny (e.g., Syts- ma & Baum, 1996). One of the major lessons of this study is that the 18S rRNA gene is difficult to sequence, apparently due in large part to the secondary structure inher- ent in the rRNA. As a result, many published se- quences are erroneous, some highly so, and the ex- tent of insertion and deletion events has been greatly overestimated. We reiterate that whereas the total length of the aligned 18S rDNA data matrix of 64 taxa used by Nickrent and Soltis (1995) was 1853 bp, the length of our 228- taxon data matrix actually is shorter, 1850 bp. After resequencing over 20 dubious 18S rDNA sequences, we were able to remove numerous “false” indels and reduce the length of the aligned sequences. The great ma- jority (70%) of the 18S rDNA sequences used here were generated via cycle sequencing followed by automated sequencing, an approach that provides more reliable rDNA sequences. Additional “older” 18S rDNA sequences should be replaced with se- quences generated via this approach. The overall slower rate of evolution of 18S rDNA compared to rbcL (see Nickrent & Soltis, 1995) Volume 84, Number 1 1997 Soltis et al. 18S Ribosomal DNA Phytogeny 29 contributed, in part, to the widespread belief that 18S rDNA sequences would not contribute greatly to phylogenetic inference in angiosperms. Although this study and other recent papers employing entire 18S rDNA sequences (e.g., Nickrent & Soltis, 1995; Kron, 1996; D. Soltis & Soltis, 1997; Rod- man et al., submitted; Johnson et al., unpublished) have dispelled this notion, 18S rDNA sequences will, in most cases, not elucidate relationships to the degree possible with the more rapidly evolving rbcL. In some groups such as Orchidaceae, how- ever, 18S rDNA has been found to evolve faster than rbcL (Cameron and Chase, unpublished). FUTURE CONSIDERATIONS These exploratory analyses clearly illustrate the phylogenetic potential of 18S rDNA sequences for elucidating angiosperm relationships at higher tax- onomic levels. Future attempts to conduct broad phylogenetic analyses of 18S rDNA sequences should not only add more taxa, but should also in- volve the resequencing of the 18S rRNA gene for This study suggests that a broad, nuclear-based phylogenetic hypothesis for the angiosperms is achievable via sequence analysis of the 18S rRNA gene. One of the strengths of 18S sequence data appears to be the ability to recognize a suite of groups that appear in all shortest trees (e.g., glu- cosinolate clade, saxifragoids, Caryophyllidae s.l., Asteridae s.l., celastroids). This may reflect substi- tutions that occurred in highly conserved portions of the 18S rRNA gene during the early diversifi- cation of a lineage, resulting in a well-supported clade. Such substitutions are rare, however, and the result is limited resolution in some areas of the 18S rDNA topologies. Thus, our results also clearly re- veal that 18S rDNA topologies will, in most rases, not exhibit the degree of resolution and internal support possible with rbcL sequences. Increased sampling of angiosperms for 18S rDNA sequence analysis is desirable. However, to achieve a nucle- ar-based estimate of angiosperm phylogeny com- parable to that realized with rbcL, it probably will be necessary to include all, or portions of, the 26S rRNA gene as well. The utility of portions of the 26S gene for inferring family-level relationships has been demonstrated for angiosperms (Hamby & Zimmer, 1992). as well as for other groups of or- ganisms (e.g., Buchheim & Chapman, 1991; Chap- man & Buchheim. 1991; Chapela el al., 1994; Wa- ter* et al.. 1992). Conclusions This study provides general insights into the structure and evolution of the 18S rRNA gene in angiosperms and dispels certain “myths” about its evolution. Indels are neither as common nor us problematic for alignment as previously believed. Insteud, they are largely confined to a few, small, specific regions that correspond to the termini of certain helices present in the proposed secomlary structure for 18S rRNA. When these few. short ur- eas are eliminated from consideration, alignment of 18S rDNA sequences is straightforwunl and easily accomplished by eye across all ungiosperms. Con- versely, indels are rare throughout most of the 18S rRNA gene; when present, they typically involve a single base pair. Furthermore, indels present in highly conserved regions of the gene may, in fact, be phyiogenetically informative, such as the inser- tion that unites saxifragoids and the deletion that unites higher eudicots. Initial attempts to evaluate the impact of sec- ondary structure of the 18S rRNA transcript on phylogeny reconstruction in angiosperms suggest that both stem and loop regions appear to be sources of phylogenetic information, with a slightly greater proportion (58% vs. 42%) of informative sites found in stem rather than loop regions. Of the stem changes we analyzed, only 27% destroyed a base-pairing couplet; 73% restored or maintained stem base pairing and hence are considered com- pensatory. The most frequent type of stem change observed involved single base substitutions that changed one base-pairing couplet to another (e.g., U-G to C-G; U-A to U-G). The high frequency of compensatory change indicates that some down- weighting of stem characters relative to loop bases may be warranted in future broad analyses of 18S rDNA sequences. The phylogenetic trees obtained in these explor- atory, broad analyses of 18S rDNA sequences are largely concordant with those resulting from anal- yses of rbcL sequences. Areas of general concor- dance include the presence of a tricolpate or eu- dieot clade, which in tum includes two large clade* corresponding mostly to Rosidae ami Asteridae s.l., respectively. However, the latter clade also includes Caryophyllidae s.l. in 18S rDNA trees, but not in trees retrieved from analyses of rbcL sequences. In addition, the monocotyledons are monophyletir (with the possible exception of Acorns ) and gener- ally appear with other taxa having monosulcate pol- len. One of the most noteworthy differences be- tween this study and that of Chase et al. (1993) concerns the first-branching angiosperms. The 31 84, Number 1 etal. DNA Phylogeny Annals of the Missouri Botanical Garden 1 A Volume 84, Number 1997 Soltis et al. 18S Ribosomal DNA Phylogeny LOWER EUDICOTS Annals of the Missouri Botanical Garden ROSIDAE Volume 84, Number 1 Soltis et al. 37 Figure ID. Figure 2. One of 2508 shortest trees resulting from the exploratory phylogenetic analysis of 223 species of angio- sperms; two indels were included in the analyses. Each of the shortest trees has a length of 3930 steps. Cl = 0.235, and RI - 0.540. Arrows indicate nodes not present in the strict consensus of all shortest trees. The letters A and B indicate the occurrence of the indels described in Table 2. Because of its size, the tree has been broken into four parts (2A, 2B, 2C, and 2D). Volume 84, Number 1 Soltis et al. 39 1997 18S Ribosomal DNA Phylogeny 40 Annals of the Missouri Botanical Garden Figure 2C. LOWER EUDICOTS Volume 84, Number 1 1997 Soltis et al. 18S Ribosomal DNA Phylogeny Figure 3B. 3C Volume 84, Number 1 1997 Soltis et al. 18S Ribosomal DNA Phylogeny 45 EUDICOTS tE £ cE • Tacca Bowiea Chlorophytum Hippeastrum Gladiolus Xanthorrhoea Veitchia Cyanella Helmholtzia Maranta Zingiber Costus Canna Heliconia Musa Glomeropitcairnia Zea Oryza Colchicum Calla Oncidium Ceratophyllum Neiumbo Aristolochia ^ Asarum > Lactoris g Houttuynia x Peperomia g Saururus «/> Nymphaea Schisandra Austrobaileya lllicium Amborella Gnetum gnemon Gnetum nodiflorum Gnetum urens r— Ephedra sinica i— Ephedra torreyana 4A o z 8 I 1 Figure 4. Strict consensus of 2582 shortest trees resulting from the exploratory phylogenetic analysis of 194 species of angiosperms; two indels were included in the analyses. Each of the shortest trees has a length of 3507 steps, Cl = 0.249, and RI = 0.536. The letters A and B indicate the occurrence of the indels described in Table 2. Because of its size, the tree has been broken into three parts (4A, 4B, and 4C). MONOSULCATE GRADE Annals of the Missouri Botanical Garden 4B LOWER EUDICOTS/MONOSULCATES Volume 84, Number 1 Soltis et al. 47 1997 18S Ribosomal DNA Phylogeny 4C LOWER EUDICOTS / MONOSULCATES 48 Annals of the MONOSULCATES Figure 4D. TRIBAL RELATIONSHIPS IN j- F. Smith \ J. C. Wolfram™, K. D. THE GESNERIACEAE: dZo* C ' L Carr ° 11 ^ and S ' EVIDENCE FROM DNA SEQUENCES OF THE CHLOROPLAST GENE ndhY 1 drawn a great deal of attention (Annals of the Mis- * ’ * Ann. Missouri Bot. Gard. 84: 50-66. 1997. 84, Number 1 52 Annals of the Missouri Botanical Garden Table 1. Species sequenced in this study with Genbank submission numbers and voucher specimens. JFS - James F. Smith, WLW - Warren L. Wagner, DEB - Dennis E. Breedlove, SI - Smithsonian Institution, LG - Longwood Gardens. Letters in parentheses indicate herbarium where vouchers are deposited. Speci Achimenes skinneri Lindl. Aeschynanthus micranthus C. B. Clarke Agalmyla parasitica (Lam.) Kuntze Alloplectus meridensis Klotzsch Anna mollifolia (W. T. Wang) W. T. Wang & K. Y. Pan Asteranthera ovata (Cav.) Hanst. Besleria affinis Morton Boea hygroscopica F. Muell. Chirita sinensis Lindl. Codonanthe elegans Wiehler Columnea schiedeana Schlecht. Cyrtandra hawaiensis C. B. Clarke Cyrtandra umbellifera Merr. Diastema racemiferum Benth. Didissandra frutescens Clarke Didymocarpus albomarginata Hemsl. Drymonia stenophylla (J. D. Smith) H. E. Moore Fieldia australis Cunn. Gasteranthus corallirms (Fritsch) Wiehler Gesneria pedicellaris Alain Gesneria christii Urban Gloxinia sylvatica (HBK) Kunth Hemiboea henryi C. B. Clarke Kohleria spicata (Kunth) Oerst. Mitraria coccinea Cav. Monophyllaea hirticalyx Franch. Monopyle macrocarpa Benth. Napeanthus costaricensis Wiehler Napeanthus macrostoma Leeuwenberg Negria rhabdothamnoides F. Muell. Nematanthus hirsutus (Mart.) Wiehler Niphaea oblonga Lindl. Opithandra primuloides (Miq.) B. L. Burtt Omithoboea wildeana Craib. Paliavana prasinata (Ker-Gawl.) Fritsch Paraboea rufescens (Franch.) Burtt Pelrocosmea flaccida Craib Primulina tabacum Hance Ramonda myconi (L.) Rchb. Rhynchoglossum natonianum (Wall.) B. L. Burtt Rytidophyllum tomentosum (L.) Mart. Rytidophyllum auriculatum Hook. Samtpaulia ruptcola B. L. Burtt Sarmienta repens Ruiz & Pav6n Sirmingia (Uetda) brasiliensis (Regel & Schmidt) Wiehler Sinningia cooperi (Paxt.) Wiehler Solenophora obliqua D. L. Denham & D. N. Gibson Streptocarpus holstii Engl. ruanotrichum oldhamii (Hemsl.) Soler. SI 94-606 JFS 643 (WIS) SI 94-570 JFS 1182 (WIS) Skog 94-498 Stewart 12234 (SRP) LG870575 SI 89-041 SI 94-111 SI 82-45 JFS 288 (WIS) WLW 6753 (BISH) WLW 6701 (BISH) JFS 3539 (SRP) SI 85-98 SI 94-512 SI 94-509 JFS 2248 (WIS) Stewart s.n. (SRP) SI 94-243 SI 94-567 SI 94-507 Dunn 9012051 (SRP) SI 85-157 SI 94-552 SI 94-158 Stewart s.n. (SRP) no voucher Feuillet (US) Nordenstam 8608 (S) Olmstead & Reeves, 1995 SI 78-354 SI 93-073 SI 93-075 SI 78-368 Skog s.n. (US) SI 85-196 SI 93-040 Katzenstein s.n. (SRP) SI 94-378 SI 77-235 SI 94-524 SI 94-492 Stewart s.n. (SRP) Dunn 9104014 (SRP) SI 94-340 SI 94-554 DEB 71542 (CAS) Olmstead & Reeves, 1995 JFS s.n. (WIS) SI 86-106 U62177 U62169 U62171 U62158 U62188 U62204 U62162 U62205 U62189 U62178 U62164 U62172 U62165 U62173 U62156 U62190 U62207 U62159 U62196 U62163 U62192 U62191 U62157 U62180 U62181 U62182 U62193 U62168 U62197 U62198 U62161 U62195 L36404 U62160 U62183 U62166 U62174 U62206 U62184 U62167 U62185 U62179 U62200 U62199 U62176 U62194 U62175 U62201 U62186 U62202 L36415 U62170 U62187 84, Number 1 STc,, Digitalis grandiflora Mill. & K. il Garden 84, Number 1 ' Volume 84, Number 1997 Smith et al. Tribal Relationships in Gesneriaceae 58 Annals of the Missouri Botanical Garden Volume 84, Number 1 Smith et al. 59 Figure 3. Single most-parsimonious tree of 4613 steps (Cl = 0.29, RI = 0.38) from the analysis of the species in the Gesneriaceae with only Paulouinia (SC— Scrophulariaceae) designated as the outgroup. Displayed in this figure are the tribes of the Cyrtandroideae, K1 — Klugieae, Ti — Titanotricheae, Di — Didymocarpeae, Tr — Trichosporeae, and Cy Cyrtandreae. The Gesnerioideae are displayed in Figure 4. Numbers along branches are the synapomorphies that support those clades. Numbers in parentheses indicate those synapomorphies that are homoplastic in this tree. Numbers below branches are decay values. Branches with no value indicated have a decay value of 1. 60 Annals of the the Gesner Gesnerioideae, Co— Cot nerieae, and Gl— Gloxini >st-parsimonious tree of 4613 steps (Cl = 0.29, RI = 0.38) from the analysis of the species in only Paulownia designated as the outgroup. Displayed in this figure are the tribes of the le — Beslerieae, Na — Napeantheae, Si — Sinningieae, Ep — Episcieae, Ge — Ges- tandroideae are displayed in Figure 3. Numbers along branches are the syna- Numbers in parentheses indicate those synapomorphies that are homoplastic in 84, Number 1 84, Number 1 64 66 Annals of the Missouri Botanical Garden Wang, W. T., K. Pan & Z. Li. 1992. Keys to the Gesner- iaceae of China. Edinburgh J. Bot. 49: 5-74. Wiehler, H. 1983. A synopsis of the neotropical Gesner- iaceae. Selbyana 6: 1-249. & A. Chautems. 1995. A reduction of Lietzia to Sinningia. Gesnenana 1: 5-7. Wolfe, K. 1991. Protein-coding genes in chloroplast DNA: Compilation ol nucleotide sequences, database entries and rates of molecular evolution. Pp. 467-482 in L. Bogorad & I. K. Vasil (editors). The Photosynthetic Apparatus: Molec- ular Biology and Operation, vol. 7B, Cell Culture and So- ‘ ' ic Press, New York. malic Cell Genetics in Plants. Acaden Sandra Knapp 2 , Viveca Persson 2 , and Stephen Blackmore 2 A PHYLOGENETIC CONSPECTUS OF THE TRIBE JUANULLOEAE (SOLANACEAE ) 1 Abstract The tribe Juanulloeae has traditionally consisted of nine genera of rarely collected, epiphytic shrubs and small trees: podium. Here we present the results of a cladistic study of the relationships of the species of these genera and provide a conspectus of the genera as we define them. The number of genera in our treatment is reduced to six: Dyssochroma , Juanulloa, Marked, Merinthopodium , Schultesianthus, and Trianaea. Included in the key and conspectus is the genus Solandra, which at present is treated as a separate tribe, Solandreae. We also discuss the groupings on the tree and point out areas for future research in this group. A key to the genera is provided and for each genus a list of component taxa and their distributions is also given. Many of these genera have previously been illustrated, but illustrations are provided for some of the previously unillustrated taxa or for taxa where illustrations are difficult to find. The Solanaceae are an economically important, cosmopolitan family with over 2500 species that have traditionally been divided into two subfami- lies. The Cestroideae, including petunias, the oes- trums and their relatives, have non-compressed, of- ten prismatic seeds and tropane alkaloids. The Solanoideae, which contain the large majority of the species in the family, include Solarium and its rel- atives that have compressed seeds and steroidal al- kaloids. This traditional classification has recently been challenged by cladistic analyses using chlo- roplast and nuclear DNA data sets, and the family can now be divided into approximately seven mono- phyletic groups (see Olmstead et al., in press). The tribe Juanulloeae, first described by Hun- ziker (1977), is placed in the subfamily Solanoide- ae in both the traditional and phylogenetic systems: its members share flat, discoidal seeds and curved embryos with others in that subfamily. The tribe as defined by Hunziker (1977, 1979) is delimited by a combination of habit, anther, and seed characters and consists of nine genera: Dyssochroma Miers, Ectozoma Miers, Hawkesiophyton Hunz., Juanulloa Ruiz & Pav6n, Merinthopodium Donn. Sm., Markea Rich., Rahowardiana D’Arcy, Schultesianthus Hunz., and Trianaea Planch. & Linden. The Juan- ulloeae are thought to be closely related to the ge- nus Solandra Swartz, the only member of the So- landreae. Trianaea was previously considered a member of the Solandreae (Hunziker, 1979; Ber- nardello & Hunziker, 1987), but was transferred to the Juanulloeae (Hunziker & Bemardello, 1989) owing to its ex-endospermous seeds and almost straight embryos with oblique, accumbent cotyle- dons. Solandra differs from the members of the Juanulloeae in its incumbent rather than accum- bent cotyledons and its partially inferior ovary (D’Arcy, 1973 [1974]). Miers (1857) allied Solan- dra, Juanulloa, Markea, Sarcophysa Miers ( Juan- ulloa speciosa (Miers) Dunal), and Dyssochroma as the Solandreae, which he considered to be closely related to the shrubby neotropical genus Brunfelsia L. Brunfelsia is now considered to be related either to Salpiglossis L. and its relatives (Hunziker, 1979) or to Petunia L. (Olmstead et al., in press). In the recent molecular phylogeny of the family Juanul- loinae and Solandrinae are united in the tribe Juan- ulloeae (see Olmstead et al., in press). The general habit and morphological similarities of the Juan- ulloeae and Solandra have long been recognized and we have thus included Solandra in this anal- The genera of the Juanulloeae (here referred to in the broad sense, including Solandra ) are all neo- tropical, epiphytic trees and shrubs. Many of the species are myrmecophilous, especially in the ge- 1 We thank M. Nee, W. G. D’Arcy, R. G. Olmstead, and L. Freire de Carvalho for helpful discussions about the biology and taxonomy of Solanaceae; M. G. Bovini in Rio de Janeiro and T. Nunez in Quito for invaluable field assistance; D. M. Williams for help with successive approximation weighting; the curators of BM, F, K, MO, NY, QCNE, and US for loans and permission to sample pollen from specimens in their care; and the SEM and Photographic Units 2 Department of Botany, The Natural History Museum, Cromwell Road, Ixmdon SW7 5BD, United Kingdom. Ann. Missouri Bot. Gard. 84: 67-89. 1997. -SSSSKSS- RsSSwsr-- Volume 84, Number 1 1997 Knapp et al. Tribe Juanulloeae Figure 2. — A. Juanulloa mexicana (Schltdl.) Miers, cult. Royal Botanic Gardens, Kew, scale bar =1.5 cm. — sity of Texas, Austin, TX, originally collected at Las Tux- tlas, Veracruz, Mexico (photo J. Mallet), scale bar = 4 cm. Figure 3. — A. Merinthopodium neuranthum (Hemsl.) 2 cm. — B. Trianaea speciosa (Drake) Soler., Knapp et al. ers (for a complete listing of the host plant rela- tionships of these butterflies see Drummond & Brown, 1987). The young larvae make character- istic “C”-shaped damage in the leaves, but have rarely been reared and are difficult to collect from the forest canopy. In common with many other tropical epiphytes the leaves of species in the group tend to be thick and leathery, and the branches flexible with pliable bark. Flower shape in the Juanulloeae varies con- siderably from the long, red or bright orange, pre- sumably hummingbird-pollinated, flowers of Mar- kea coccinea Rich, and most of the species of Juanulloa (Fig. 2A) to the greenish, open campan- ulate flowers of the species of Trianaea (Fig. 3B), Merinthopodium (Fig. 3A), and Dyssochroma, which are bat-pollinated (Voss et al., 1980). In every case, apart from Markea panamensis, however, the flow- ers are sympetalous and have relatively long corolla tubes. Fruits of members of the Juanulloeae are generally fleshy to leathery berries, with some vari- ation in the thickness of the berry wall. Genera of the tribe have been traditionally delimited (Hun- ziker, 1977, 1979) using combinations of the fol- lowing characters: insertion of the anthers on the filaments, filament insertion on the corolla tube, and corolla aestivation. Characters and their states are discussed more fully in Materials and Methods. and this makes the assessment of characters diffi- cult since the extent of variability is not known. Woody tropical epiphytes are difficult to collect as they often grow high in the canopy and flower only rarely. Many of the species of the Juanulloeae are known only from flowering material, and thus de- cisions based on fruit characters tend to be rather ad hoc at best. With few specimens available it is nearly impossible to assess variability in charac- ters, and a tendency to overdivide at the generic level is apparent and understandable. Recent applications of molecular systematics have been extremely useful in providing broad frameworks for directing future study. However, at present, the limited range of taxa that has been 70 Volume 84, Number 1 1997 Knapp et al. Tribe Juanulloeae ris, 1989). Trees were also constructed using NONA (Goloboff, 1993) to confirm the actual ver- sus potential groupings on the tree, as NONA and Hennig86 treat zero length branches (potential groupings) in a different way. The following com- mands, as recommended by Goloboff (1993), were used: hold 100, hold/20, and mult*50. Successive approximation weighting (xs w option in Hennig86) was used to assess the reliability of fit of characters to the most parsimonious tree. The fitting function in the MS-DOS program Pee Wee (Goloboff, 1993) was also used to assess character reliability. Character weighting, when applied to the fit of char- acters, emphasizes those characters that best fit the initial tree topology. Successive weighting allows the characters to judge themselves in terms of their reli- ability: i.e., their best fit to the solution supported by all the data (Carpenter, 1994). Best fit is judged by the shortest tree (Farris et al., 1970), the shortest tree for the weighted data in terms of tree length (Farris, 1969), or the “heaviest” tree when calculated as a function of character weights (Goloboff, 1993). Char- acters that are more homoplasious are less reliable and are thus downweighted in these analyses. Suc- cessive weighting (Hennig86, xs w option) uses the rescaled consistency index (re) of Farris (1989) as the weighting function of each character: it is calculated as the product of ensemble RI (retention index) times ensemble Cl (consistency index) and scaled between 0 and 10. Goloboff (1993) calculates weights as the extra number of steps per character such that the weight = K/K+ESi, where ESi is the extra steps per character and K is the concavity constant (in our anal- yses set at K = 3). The characters were coded to be unordered, thus minimizing ad hoc weighting or polarity before anal- ysis. In this analysis we used 43 taxa and 38 char- acters (Table 2). Three taxa were selected as out- groups (see Tables 1 and 3), Nicandra physalodes , Atropa belladonna, and Lycium cestroides, represent- ing a range of putative sister taxa for the Juanulloeae. Choosing a range of outgroups (Watrous & Wheeler, 1981) has been thought to increase the likelihood of obtaining an accurately rooted tree. Recent work, however (Nixon & Carpenter, 1993), has shown that multiple outgroups perform no better at “polarizing” ingroup nodes, but that multiple outgroups might im- prove inference. One difference in this data set from that used in Persson et al. (1994) is the omission of Mandragora as one of the outgroups for the analysis. Olmstead and Palmer (1992) had originally identified Mandragora and Solandra as sister taxa using chlo- roplast DNA restriction site mapping. More recent work has revealed that Mandragora is an isolated tax- on of uncertain affinity, possessing many autoapomor- Table 2. Characters used in the cladistic analysis of the Juanulloeae. 0. Habit shrubs 0, herbs 1, epiphytes 2. present 1. 2. Inflorescence of solitary flowers 1, few flowers 0, many flowers 2. 3. Inflorescence axis condensed 0, elongate 1. 4. Calyx lobes shorter than the corolla tube 0, equal to 6. Calyx lobes acute 0, acuminate 1, long-acuminate 2, rounded 3. 8. Corolla aestivation overlapping 0, valvate 1. 9. Flower shape narrow tube 1, salverform 2, open 0, 10. Corolla color green/white 0, purple-purplish green 1, red, orange, or yellow 2. 11. Filaments straight 0, decimate 1. 12. Filament tube absent 0, present 1. 13. Filament base glabrous 0, pubescent 1. 14. Filament base pubescence dense 0, sparse 1. 15. Filaments inserted in anther basally 0, dorsally 1, ventrally 2. 16. Anthers included in the corolla tube 1, at mouth of corolla tube 0, exserted 2. 17. Anthers dehiscing separately 0, confluently 1. 18. Ovary superior 0, partially inferior 1. 19. Ovary ± conical 0, narrowly beaked 1. 20. Fruit pericarp membranous 0, coriaceous 1. 21. Seeds reniform 0, rectangular 1. 22. Cells in center of testa square 0, elongate 1. 23. Lateral cell walls straight 0, sinuate 1. 24. Dried seed color pale brown 0, dark brown 1, orange 2. 25. Undigested seed surface pitted 0, smooth 1. 27. Colpi extending nearly to the poles 0, relatively short 1. 28. Pollen without Ubisch bodies 0, covered with Ubisch 29. Colpi with continuous margins 0, lalongate apertures 1. 30. Exine around apertures thickened 0, not thickened 1. 31. Tectum continuous 0, not continuous 1. 32. Beak-like margo apertures absent 0, present 1. 33. Colpus border not thickened 0, thickened 1. 34. Outline in equatorial view spherical 0, oblate 1. 35. Outline in polar view obtuse, convex 0, acute, straight 1. 36. Nexine same thickness as the sexine 0, nexine thicker than the sexine 1. phies, both molecular (Olmstead & Sweere, 1994; Olmstead, pers. comm.) and morphological. The data matrix is presented in Table 3. Most of the characters are self-explanatory; those peculiar to the Juanulloeae are described in detail here and the palynological characters were discussed in Annals of the Missouri Botanical Garden Table 3. Data matrix used for the analysis of the Juanulloeae. Solandra brachycalyx Solandra boliviana Marked sessiliflora Marked formicarum Marked camponoti Marked lopezii Mar a ulei Merinthopodium neurar, Merinthopodium pendul Schultesianihus uniflon Schultesianthus megala Schultesianihus odorifei Schultesianthus dudleyi Jua, ll a h d ana Juanulloa globifera Juanulloa speciosa Juanulloa ochracea Juanulloa ferruginea Juanulloa membranacei Juanulloa parviflora Juanulloa mexicarm Dyssochroma longipes Dyssochroma viridijlora ’■evipes 1000100000000070200000001001001000100 1000000000110101000000000000000000100 0000000001100101200000000000001100100 2010010000210020001010001000001000000 20100100001100?000101?????01001000000 2000012000000100100001102001101100100 200001200000010010000?????01101100110 2000012002200100000001102001010100110 2000012000000100100001102001010100110 2000012000000100100001102001010100110 20110010002001001000??????11001000000 20100010001001001000????????????????? 2000QO00000OO0?1110011102001000010110 2000000002000101010011102001000010110 2001000013000111200010010100000001100 200101001300010?2000??????01010100110 2110010100000100 ?0001?????01101100100 211001010000010010001?????01101100100 2110010100100110000010001001101100100 2100010100010110000010010101100000000 2100010100010110000010010101100000000 210001010001011000001?????01100000000 2100010100010110000010010101100000000 210001010001011000001?????01100000000 2000110000001101100011000001001000001 20201100012000?1110101010001011000001 20201100011000?111010???????????????? 2010010001200101100010000000000001110 2000010001000101100000000000000001110 2001010001200101100000001001001000001 2001010001200101000000001001000000000 200100000120 01 011000??????01001000101 20010000010001011000?0001001001000101 20010100010001011000????????????????? 2000010001200101100000001001001000001 20101110130001001000??????00000100100 20101110130001002000??????00000100100 2001110013000112200010000000001100100 2010110000000112200010000001001000110 200011001300011220001???????????????? 2001110013000112200010000000001000110 20111100130001022000??????00000000100 201011000?000????000????????????????? Persson et al. (1994). Minute, peltate, glandular tri- chomes (character 1) on the leaves of Schultesian- thus were first described by Bemardello and Hun- ger (1991) and are either present or absent on both leaf surfaces. These trichomes are sunken into small pits in the mesophyll so that the ca. 24 cell head is at about the same level as the foliar surface. The trichomes appear as minute reddish dots to the naked eye and are usually, but not always, more abundant on the abaxial leaf surface. Corolla aes- Figure 4. One of the two equally parsimonious identical to that produced by NONA, and to the con states are shown in Table 2. For characters marked < and parallelisms (homoplasy) and solid marks non-1 cladograms from the Hennig86 analysis. ' rnsus tree. The characters are discussed in i the branches of the cladogram: stippled r tejrt, ^nXcharacter indicate reversals Volume 84, Number 1 1997 73 Knapp et al. Tribe Juanulloeae 74 Annals of the Missouri Botanical Garden ... . . . ■mn in 76 77 Garden 78 Volume 84, Number 1 1997 Knapp et al. Tribe Juanulloeae 79 Figure 7. Juanulloa pavonii (Miers) Benth. & Hook, as Ectozoma Pavonni (plate 48 from Miers, J. 1857. Illustrations dell, Portaea aurantiaca Tenore); J. ochracea Cuatr., Colombia to Peru; J. parasitica Ruiz & Pa- v6n, Ecuador, Peru, and Brazil (Ulloa parasitica (Ruiz & Pav6n) Pers.); J. parviflora (Ducke) Cuatr., Brazil, near Manaus (known only from the type) (Marhea parviflora Ducke); J. pavonii (Miers) Benth. & Hook., W Ecuador and NW Peru ( Ecto- zoma pavonii Miers, Markea pavonii (Miers) D’Ar- cy); J. speciosa (Miers) Dunal, Colombia and Ec- uador (Sarcophysa speciosa Miers); J. verrucosa (Rusby) Hunz., Bolivia (Markea verrucosa Rusby); J. wardiana (D’Arcy) S. Knapp, Panama (Raho- wardiana wardiana D’Arcy). The flowers of most of the species of Juanulloa conform to the classic hummingbird pollination syndrome: they are tubular, brightly colored and very thick and fleshy (Fig. 2A). However, J. pavonii (Fig. 7) and J.ferruginea (and also perhaps to some extent J. parviflora) have quite different flowers, which are greenish, with shorter tubes and some- what reflexed lobes. These perhaps represent a dif- ferent pollination syndrome. Annals of the Missouri Botanical Garden Figure 8. Markea coccinea Rich, (plate 45 from Miers, J. 1857. Illustrations of South American Plants). The very condensed globular inflorescence of Juanulloa wardiana and /. globifera is unique in the tribe, and in the family Solanaceae (illustrations can be seen in Knapp & D’Arcy, 1993). In these species the calyx as well as the corolla is brightly colored and quite showy. These flowers are appar- ently visited by hummingbirds, and produce copi- There are two types of pollen found in this genus (Persson et al., 1994). In Juanulloa speciosa and J. ochracea the pollen is 3-colporate with long, narrow colpi and scabrate/rugulate exine ornamentation. In the rest of the genus the pollen is 3-colporate with short, broad colpi and a scabrate/perforate exine 3. Markea Rich., Actes Soc. Nat. Hist. Paris 1: 107. 1792. TYPE: Markea coccinea Rich. Fig- Hawkesiophyton Hunz., Kurtziana 10: 39. 1979. TYPE: Hawkesiophyton panamense (Standi.) Hunz. (basio- nym Markea panamensis Standi.). Lamarkea Pers., Synopsis 1: 218. 1805. TYPE: Lamarkea Garden Annals of the Missouri Botanical Garden Figure 10. Schultesianthus leiwanthus (Donn. Sm.) Hunz. ( Hampshire & Whitefoord 469, BM). description of Solandra. The flowers of Schultesianthus are among the most showy in the Juanulloeae (see Fig. 10). They are sweetly fragrant and change color from white to a creamy yellow with age. Large bees have been observed visiting the flowers, but little is known about their biology. 6. Solandra Swartz, nom. cons., Kongl. Vetensk. Acad. Nya Handl. 8: 300-306. 1787. TYPE: Solandra grandiflora Swartz. Figures 2B, 11. Swansia Gmelin, Systems naturae 2: 296, 390. 1791. Solandera Kuntze, Rev. Gen. Plantarum. 452. 1891 (or- thographic variant). Woody lianas or high climbing epiphytic shrubs Volume 84, Number 1 1997 Knapp et al. Tribe Juanulloeae Trianaea neovisae Romero-Castafieda— Peru: D(az & Osores 2944 (MO). Trianaea nobilis Planch. & Linden— Colombia: Silver- stone-Sopkin 1188 (MO); Ecuador: Zak 1028 (K). Trianaea speciosa (Drake) Soler. — Ecuador: Boeke 547 (NY), Gentry, Bonifaz & Loot 30950 (MO), Dodson & Dodson 18592 (NY); Spruce 5527 (K); Peru: Barbour 4113 (MO). Trianaea sp. nov.— Peru: DCaz, Osores & Bustamente 3935 (MO). A REVISION OF THE GENUS XIPHOTHECA (FABACEAE ) 1 91 al Garden Volume 84, Number 1 1997 Schutte Revision of Xiphotheca * -V V V -V V -V- V -V- -V Taxonomic Treatment Xiphotheca Eckl. & Zeyh., Enum. PI. Afric. Aus- tral. 2: 166. 1836. TYPE: Xiphotheca rotun- difolia Eckl. & Zeyh. (lectotype, designated by Schutte & Van Wyk, 1993) [= Xiphotheca tecta (Thunb.) A. L. Schutte & B.-E. van Wyk], Priestleya DC. sect. Aneisothea DC., in Ann. Sci. Nat. 4: 91. 1825, Prodr. 2: 121. 1825. TYPE: Priestleya el- liptica DC. (lectotype, designated by Schutte & Van Wyk, 1993) [= Xiphotheca elliptica (DC.) A. L. Schutte & B.-E. van Wyk]. Woody shrubs or shrublets. Leaves alternate or rarely opposite or subopposite, simple, narrowly el- liptic to almost circular, mostly flat, sometimes with recurved margins, pinnately veined; petiole short, — 1 mm long; stipules inconspicuous, less than 0.5 mm long. Inflorescence axillary, 2-flowered, with the two flowers opposite, aggregated into synflores- cences of up to 20 flowers. Bracts linear to oblan- ceolate, fused at the base with pedicel for 0.5-1 .0 mm. Bracteoles minute, sometimes lacking. Corolla yellow, longer than the calyx, glabrous. Calyx nar- rowed to the base, rarely intrusive; upper two lobes fused higher up than the lower three lobes; carinal lobe sometimes longer than the upper four. Stan- dard petal suborbicular to elliptic; apex emarginate. Wing petals oblong, longer than the keel; the tips imbricate; pocket developed as a thickened lobe toward the inside. Keel petals widely obovate, with weakly developed pockets, apex obtuse. Stamens diadelphous, the vexillary filament free; anthers ± uniform in shape and size, alternately dorsifixed and subbasifixed. Pistil sessile; style slender, Iff Priestleya glaucaT. ^M. Salter, J. S. Mr. Bot. 8: 256. 1942. of^Hercules’ 1 ra^'jIstlXrg! A/^T^aec- designated by Schutte & Van Wyk (1993), 100 Volume 84, Number 1997 A REVIEW OF THE GENUS ECCREMOCARPUS (BIGNONIACEAE) 1 2 G. D'Atc? ■ work left by the late Alwyn H. Gentry and that of others 108 3. Eccremocarpus scaber Ruiz & Pav6n, Syst. Veg. 157. 1798. Calampelis scaber (Ruiz & Pa- v6n) Sweet, Brit. FI. Gard. ser. 2. 1. t. 30. 1831. [As u scabra. n ] TYPE: Chile. Provs. Col- chagua, Rancagua, and San Jacob, Ruiz & Pa- v6n 1798, (holotype, MA, photo, 029233, F). Eccremocarpus sepium Bert, in “Merc. Chil. 1829” cf. Bull. Ferussac, 20. 111. 1830. Eccremocarpus scaber var. [b] saepium (Bert.) A. DC., Prod., 9: 239. 1845. TYPE: Chile. Sepibus secus, vias prope Rancagua S. Yago et Quillota, Bertero 965 (BM G , F ! , FI S , MICH C , MO ! , P c ). [Accepted as a synonym by Munoz (1966).] Eccremocarpus scaber var. aurea [sic] Benary, Gartenflora 22: 608. 1903. TYPE: cult. Hort. Schonbrunn (?W, Eccremocarpus scaber var. carmineus [sic] Spigolatore, Bull. R. Soc. Toscana Orticultura 29(2): 339, fig. 22. 1904. TYPE: cultivated in ?France, not located. Eccremocarpus scaber carmineus Pynaert, Rev. Hort. Beige 31: 55. 1905. TYPE: cult. Belgium, not located. Eccremocarpus scaber var. roseus Huxley et al.. New Roy Hort. Soc. Diet. Gard. 2: 122. 1992. TYPE: Not lo- Vine, climbing and twining, 2-6 m long, basally woody, the stem drying with alternating ridges and canals, puberulent with short, erect, often gland- tipped hairs and occasional weak hairs to 2 mm i!lii!iWIK!!ni:BJ3!H Volume 84, Number 1997 D’Arcy 111 (MO G! ex TULV) 2 (longiflorus-G) COLOMBIA. VaUe. de Vries s.n. (AAU G ) 2 (longifloras-G) ECUADOR. Napo. Dombey s.n. (P ; ) 3 CHILE. Santiago. Ellenberg 4867 (MO G! ) 2 (vargasianus- ! ) PERU. Elliott 419 (BM C ) 3 CHILE. Elliott 247 (K ! ) 3 CHILE. Elwes s.n. (K ! ) 3 CHILE. Fournier s.n. (P 30 ) 2 (longiflorus-G,S) ECUADOR. Pi- chincha. Frbr. Bert. Jime. ? s.n. (MO G! ) 3 CHILE. Freire et al. 656 (QCA C ) 2 (longiflorus-G) ECUADOR. Chim- borazo. Frodin s.n. (NY G! ) 3 CHILE. Santiago. Frddin 635 (BM G ) 3 CHILE. Aconcagua. Gay s.n. (K s ) 2 (longiflorus-S) ECUADOR. Gays.ru (P°) 3 CHILE. Santiago. Gay s.n. (P°) 3 CHILE. Valdivia. Gay 20 (P°) 3 CHILE. Santiago. Goodspeed 16853 (MO ) 3 CHILE. Aconcagua. Goodspeed 23344 (K ! ) 3 CHILE. Aconcagua. Halpin s.n. (CLEMS 0 , K G ) 3 CHILE. Hartweg 148 (K ! ) 2 (longiflorus-G,S) ECUADOR. Loja. Harvey s.n. (K 3! ) 3 CHILE. Hastings 171 (NY G! ) 3 CHILE. Hirsch P1022 (K ! ) ye^llsT^MO 1 ) 3 EUROPE FrTnce. HumLhh's.n. (K\ MO ! , P 30 ) 2 (longiflorus-G,S) PERU. Loja. Jameson s.n. (K 30 ) 2 (longiflorus-S) ECUADOR. Pi- chincha. Jameson 56 (NY G! ) 2 (longiflorus-G) ECUA- DOR. Pichincha. Jameson 186 (K 3G! ) 2 (longiflorus-S) ECUADOR. Pichincha. Jameson 286 (BM 30 , F G , K, NY G , P 80 ) 2 (longiflorus-G, S) ECUADOR. Pichincha. Karsten s.n. (W 3 ) 2 (longiflorus-S) COLOMBIA. Pichin- cha. Karsten s.n. (W 3 ) 2 (longiflorus-S) ECUADOR. Cun- dinamarca. Kausel 1673 (F ! ) 3 CHILE. Santiago. King 589 (BM C ) 3 CHILE. King 712 (BM C ) 3 CHILE. Kuntze s.n. (NY C! ) 3 CHILE. Leeds s.n. (F ! ) 3 U.S.A. California. Lehmann 3149 (BM 30 , K S! ) 2 (longiflorus-G,S) COLOMBIA. Caldas. Lobb 391 (K S! ) 2 (lobbianus-S) PERU. Londono et al. 481 (MO G! ex-MEDEL) 2 (longiflorus-G) COLOMBIA. Antioquia. Luteyn & Tirira 1336A (MO ! ) 2 ECUADOR. Napo. Macbride 4371 (F, W 3 ) 2 PERU. MacLean s.n. (K-3 3! ) 2 (lobbianus-S) PERU. Macrare? s.n. (K ! ) 3 CHILE. Ma- thews 3176 (BM 30 , K') 2 (longiflorus-G, S) PERU. Ama- zonas. Mathews s.n. (BM 30 ) 2 (longiflorus-S) PERU. Amazonas. Mazzei s.n. (FI°) 3 CHILE. Santiago. Mc- Pherson 13159 (M0 C! ) 2 (longiflorus-G) COLOMBIA. An- tioquia. Menzies 91 (MO G! ) 3 CHILE. Mexia 7887 (BM°, F, MO G! , NY G! ) 3 CHILE. Cnrico. Meyen s.ru (P°) 3 CHILE. Meyer 9422 (K 3! ) 3 ARGENTINA. Chubut. Mid- dleton s.n. (BM°) 3 CHILE. Molau & Ohman 1635 (GB° = photocopy M0 ! ) 2 (vargasianus-G) PERU. Cuzco. Montero 67a (MO G! ) 3 CHILE. Colchagua. Montero 260 (K) 3 CHILE. Santiago. Montero 507 (M0°) 3 CHILE. Santiago. Morrison 16853 (M0 C! ) 3 CHILE. Aconcagua. Nunez & Galiano 13414 (M0 ! ) 1 PERU. Cuzco. Nunez & Luna 8841 (F, M0 G! ) 2 (vargasianus-G!) PERU. Cuz- co. Nunez et al. 14135 (M0° ! ) 1 PERU. Apurimac. 0llgaard et al. 38199 (AAU°) 2 (longiflorus-G) EC- UADOR. Chimborazo. 0llgaard et al. 98155 (AAU°) 2 (longiflorus-G) ECUADOR. Carchi. Ortiz s.n. (AAU°, QCA°) 2 (longiflorus-G) ECUADOR. Imbabura. Ortiz 30 (AAU°, NY!) 2 (longiflorus-G) ECUADOR. Pearce 533 (K 3! ) 2 PERU. Huanuco. Pearce 823 (K 3! ) 2 (vargasianus-S) PERU. Penland & Summers 1080 (F) 2 ECUADOR. Azuay. Pennell 12262 (F ! , GH, NY G! ) 3 CHILE. O’Higgins. Philippi s.ru (HB G ) 3 CHILE. San- tiago. Poeppig 2 (BM°) 3 CHILE. Prance 26595 (US°) 2 (longiflorus-G) ECUADOR. Napo. Prieto P-151 (NY G! = MO, photocopy) 2 (longiflorus-G, Wurdack) ECUADOR. Caiiar. Purdie s.n. (P 30 , K-2 S! ) 2 (longiflorus-G,S) CO- LOMBIA. Risaralda. Rauh & Hirsch P.1022 (K 3 ) 2 (vargasianus-S) PERU. Urubamba. Raddin s.n. (F ! ) 3 CHILE. Santiago. Reed s.n. (K ! ) 3 CHILE. Maule. Rossiter 398253 (MO ) 3 NEW ZEALAND. Ruiz & Pavdn 5/14 (MA-3 3 , BM) 2 PERU. Ruiz & Pavdn s.n. (BM 3 , F ex MA, FI 3 ) 3 CHILE. San- tiago. Sanderness 288 (K 1 ) 3 CHILE. Saunders 776 (K 3! ) 1 PERU. Apurimac. Schlatzers.n. (AAU°) 3 CHILE. San- tiago. Schmidt s.n. (HB°) 3 CHILE. Shepard 9 (NY C! ) 3 PERU. Puno. Simpson s.n. (P°) 3 CHILE. Sodiros.n. (P 30 ) 2 (longiflorus-G, S) COLOMBIA. Cundinamarca. THana s.n. (P 30 , COL°) 2 (longiflorus-G, S) COLOMBIA. Antioquia. Triana 223 (P 30 ) 2 (longiflorus-G,S) COLOM- BIA. Risaralda. Triana 4107 (?5107,?8107,?4107) (COL°, P 30 ) 2 (longiflorus-G,S) COLOMBIA. Cundina- Vargas 3034 (K 3 , US 30 ) 1 PERU. Cuzco. Vargas 5956 (K S! , M0 G! ) 2 (vargasianus-G,S) PERU. Cuzco. Vargas 20079 (MO C! ) 2 (vargasianus-G) PERU. Cuzco. Vargas 22644 (M0 G! ) 2 (vargasianus-G) PERU. Cuzco. Vieillard 84 (P°) 3 CHILE. Valparaiso. Weberbauer (F?) 2 (longiflorus-Macbride, S) PERU. Ca- jamarca. Weberbauer 4938 (?) 2 (longiflorus-Macbride,S) ECUADOR. Werdermann 482 (BM°, F ! , HB°, M0 C! , U°) 3 CHILE. Santiago. Wildenow s.n. (HB°) 3 CHILE. Zollner 6486 (L°) 3 CHILE. Aconcagua. Zollner 9350 (M0 C! ) 3 CHILE. Valparaiso. Zollner 11052 (M0 G! ) 3 CHILE. Santiago. nodiflora 107, 108 105, 107 108 105, 106 scaber 105. 107. 108 HOT-SPOTS ANALYSIS FOR Frtdtric MtdaiP and Pierre QuizeP CONSERVATION OF PLANT BIODIVERSITY IN THE MEDITERRANEAN BASIN 1 Gard. 84: 112-127. 1997. 114 Annals of the Volume 84, Number 1 Medail & Qu6zel 115 118 Missouri Botanical Garden Table 2. Plant biodiversity of the countries from the Mediterranean basin (islands excluded). The figures indicate the number of species, except for the values marked with an asterisk, which also include subspecies. Unless otherwise indicated, sources are: (a): Davis et al. (1986) and (b): Qu6zel (1985). Other sources: (c): original observations, (d): Greuter (1991), (e): Fennane & Ibn Tattou (1995), (f): Enriquez-Barroso & Gomez-Campo (1991), (g): Bartolo et al. (1977), (h): Boulos (1995), (i): Davis et al. (1994), (j): Davis (1988), (k): I.U.C.N. (1980), (1): Olivier et al. (1995), (m): Gomez-Campo et al. (1984). Countries Algeria Tunisia Syria Turkey 20,700 10,000 2,200 97,600 10,000 2,200 185.000 50,000 3,100 10,400 10,000 2,600 779.000 480,000 8,600 2.000 *165 (i) 7.5 1,800 (c) 145 (i) 6.6 2,600 (c) *395 (i) 12.7 2,600 311 (i) 12 5.000 2,651 0) 30.8 Continental Greece 107,000 Albania 28,700 Former Yugoslavia 255,000 5,700 *4,000 *742 (i) 3.000 2,200 46 (k) 5.000 2,500 *320 (i) Continental Italy Continental Spain Portugal 251,400 200,000 549,600 50,000 504,000 400,000 91,000 70,000 4,870 (c) 3,850 (c) 4,800 3,200 6,720 (d) 5,000 2,600 (d) 2,500 570 (c) 180 0) 1,286 (d) Table 3. Plant biodiversity of the Mediterranean and Macaronesian islands. The figures indicate th< species and subspecies, except for the values marked with an asterisk, which also include varieties. Rare ar taxa correspond to Ex, E, V, R, I, and K categories of I.U.C.N. Sources: (a): Davis et al. (1986), (b): Jeanmonod (1993), (c): Bocchieri (1995), (d): Raimondo et al. (1994), (e): Lanfranco (1995), (f): Turland (g): Alziar (1995), (h): Schmida & Werger (1992), (i): Dalgaard (1994), 0): I.U.C.N. (1980), (k): Original 0): MSdail & Verlaque (1997), (m): Polunin (1987), (n): Mus (1995, modified), (o): Verlaque pers. comm, al. (1985), (q): I.U.C.N. (1983). 3.7 19.1 4.4 Balearic Islands Sardinia Sicily Malta Crete (incl. Karpathos) Cyprus Madeira Archipelago endemics) J 5,014 1,450 (a) 8,748 2,354 (b) 24,090 2,054 (c) 25,708 2,700 (d) 316 700(e) 8,700 1,706 (f) 9,250 1,620 (g) 7,273 1,582 (h) 796 670 (i) 94 (j) 180 (o) 6.5 *130 (b) 270(b) 5.5 106 (k) 200 (k) 5.2 260(c) 310(c) 9.6 16(e) 32 (k) 2.3 171(f) 200 (m) 10 130(g) 170(g) 8 504(h) 600 (k) 31.8 113 (i) 175 (i) 16.9 90 (o) 39 (p) 654(d) 12 (p) 119 (q) 69 (p) 432 (k) 137 (q) 1 I f I 1 i ! i £ j ! I .5 1 5 i j fill t + + 1! * *: Ill xxx t.:: i\W + * + II! t H + + + I t |-i ♦ +'*» 1J ♦! : >*i 1 + t* + iit*l i 1 tilth i ttttt + i .i !l!jl! !i* liUllilfl SURVEY OF CHROMOSOME NUMRERS IN RUBUS (ROSACEAE: ROSOIDEAE ) 1 ra ? 1997 Chromosome Numbers in Rubus 139 Mitteleuropa. Sonderb. Naturwiss. Vereins Hamburg 4: 1-229. & W. Maurer. 1991. Kommentierte Checklisle der in Osterreich nachgeweisenen Arten der Gattung Rubus L. (Rosaceae). Phyton (Horn) 31: 67-79. Williams, C. F., B. W. Smith & G. M. Darrow. 1949. A pan-American blackberry hybrid. J. Heredity 40: 261- Yamell, S. H. 1932. Chromosome behavior as a factor in berry breeding. Proc. Amer. Soc. Hort. Sci. 28: 114-117. . 1936. Chromosome behaviour in blackberry- raspberry hybrids. J. Agric. Res. 52: 385-396. Yeager, A. F. & E. M. Meader. 1958. Breeding better fruits and nuts. New Hampshire Agric. Exp. Sta. Bull. 448. Ytl, T. T. (Editor). 1985. Flora Reipublicae Popularis Sin- icae. Tomus 37; viii, 10-218. Science Press, Beijing. Zhukova, P. G. 1980. Khromosomnye chisla nekotorykh vidov rasteniy yuzhniy Chukotskiy. Bot. Zum. SSSR 65: 51-59. . 1982. Chisla khromosom nekotorykh vidov ras- teniy severo-vostoka Azii. Bot Zum. SSSR 67: 360-365. , A. A. Korobkov & A. D. Tikhonova. 1977. Khro- mosomnye chisla nekotorykh vidov rasteniy vostoka arkticheskoy Yakutii. Bot. Zum. SSSR 62: 229-234. Zielinski, Q. B. & D. O. Galey. 1951. Chromosome num- bers of certain trailing blackberry clones. Proc. Amer. Soc. Hort. Sci. 57: 163-164. ?ii I m i , II! !!,! J, ill |ii M « Hi M ilBiMI ! I i» I , ill ill ii tiiiiiLii 2 228 S3 2 ill i ! ii I ! i ii * °** % * ****** ** * *** ** ** Ill l il Mi I i it i 11 !1 1111 ,iili ! I I I i i Sill S3 3 3333 38338&§3 333333383 J s I 8 Sill im -M, ill 111 ill 11 c« a; g ftjosa; a* as as as as a; as as as as a; as as as as as g 1 In JM Garden Volume 84, Number 1 I Garden = s sssssaas sag ggggg sags ggggs a Garden ffiff as os as ftjftjosa;a:osa:asa:osasasa;osa; Garden i 84, Number 1 Garden 1 £ Hit i S 3 3 8 8SS f=S 1 i. i I i 1 NOTICE of the CONTENTS Angiosperm Phylogeny Inferred from 18S Ribosomal DNA Sequences Douglas E. Soltis , Pamela S. Soltis, Daniel L Nickrent, Leigh A. Johnson, William J. Hahn, Sara B. Hoot, Jennifer A. Sweere, Robert K. Kyzojf, Kathleen A. Kron, Mark W. Chase, Susan M. Swensen, Elizabeth A. Zimmer, Shu-Miaw Chaw, Lynn J. Gillespie, W. John Kress & Kenneth J. Sytsma 1 Tribal Relationships in the Gesneriaceae: Evidence from DNA Sequences of the Chloroplast Gene ndhF - J. F. Smith, J. C. Wolfram, K. D. Brown, C. L Carroll & D. S. Denton 50 A Phylogenetic Conspectus of the Tribe Juanulloeae (Solanaceae) Sandra Knapp, Viveca Persson & Stephen Blackmore 67 A Revision of the Genus Xiphotheca (Fabaceae) Anne Lise Schutte 90 A Review of the Genus Eccremocarpus (Bignoniaceae) William G. D’Arcy 103 Hot-Spots Analysis for Conservation of Plant Biodiversity in the Mediterranean Basin - - Fridtric Medail & Pierre Quizel 112 Survey of Chromosome Numbers in Rubus (Rosaceae: Rosoideae) Maxine M. Thompson 128 Notice 165 Cover illustration. Eccremocarpus viridis Ruiz & Pav6n, by Phyllis Bick. Annals ip of the Missouri Botanical Garden 1997 m Volume 84 Number 2 Volume 84, Number 2 Spring 1997 Missouri Botanical Garden The Annals, published quarterly, contains papers, primarily in systematic botany, contributed from the Missouri Botanical Garden, St. Louis. Papers originating out- side the Garden will also be accepted. Authors should write the Managing Editor for information concerning arrangements for publishing in the Annals. Instructions to Authors are printed in the back of the last issue of each volume. Editorial Committee Henk van der Werff and Michael H. Grayum Co-Editors (this issue), Missouri Botanical Garden Amy Scheuler McPherson Managing Editor, Missouri Botanical Garden Ihsan A. Al-Shehbaz Missouri Botanical Garden Gerrit Davidse Missouri Botanical Garden Roy E. Gereau Missouri Botanical Garden Diana Gunter Editorial Assistant, Missouri Botanical Garden Peter Goldblatt Missouri Botanical Garden Vicki Couture Secretary Gordon McPherson Missouri Botanical Garden Teresa Johnson Publications Order Processor P. Mick Richardson Missouri Botanical Garden For subscription information contact Depart- ment Eleven, P.0. Box 299, St. Louis, MO 63166-0299. Subscription price is $110 per volume U.S., $115 Canada and Mexico, $135 all other countries. Four issues per volume. The journal Novon is included in the subscription price of the Annals. amcpher@admin.mobot.org (editorial queries) deptll@mobot.org (orders) http://www.mobot.org The Annals of the Missouri Botanical Garden (ISSN 0026-6493) is published quar- terly by the Missouri Botanical Garden, 2345 Tower Grove Avenue, St. Louis, MO 63110. Pe- riodicals postage paid at St. Louis, MO and ad- ditional mailing offices. POSTMASTER: Send ad- dress changes to Annals of the Missouri Botanical Garden, Department Eleven, P-0. Box 299, St. Louis, MO 63166-0299. © Missouri Botanical Garden 1997 The mission of the Missouri Botanical Garden is to discover and share knowledge about plants and their environment, in order to preserve and enrich life. © This paper i ' requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Volume 84 Annals * Number 2 of the 1997 Missouri MISSOURI BOTANICAL Botanical JUN2 7 1997 Garden THE GENUS LYCIANTHES Carmen Bender de Rojas* and (SOLANACEAE) IN william a D ' Arcf VENEZUELA 1 168 Annals of the Missouri Botanical Garden Figure 1. Selected Lycianthes species. A, B. Lycianthes paucijlora.—\. Fruits and calyx — B. Fruiting calyx.— C. Lycianthes asarifolia. Flowers and foliage. — D. Lycianthes pauciflora. Seeds.— E. Lycianthes lycioides. Pyrenes. A, B. wair mo? & ° ArCy 3271 (M0, MY) ' C ' After D Arcy 16261 (M0) ' D - After Bemtez et aL 5148 (MY) - E - After Nee supported the separation of Lycianthes from Sola- num and placed it near Capsicum. Stone Cells the form of stone cells or sclerocytes, which com- monly do not enclose the seed. These are aggre- gates of secondarily hardened parenchyma cells or sclereids. Bitter (1911, 1914) examined many So- lanaceae, and he found them in various berries of Lycianthes , , Solanum, and other genera of subfamily Solanoideae. He hypothesized that they might be relics from hard walls of ancestrally capsular fruits. Danert (1969) used electron microscopy to examine their genesis in Solanum. Stone cells have been used in the Solanaceae as taxonomic clues (Schil- ling, 1981), but there is still poor confirmation of their constancy in taxa of different levels. They may be diagnostic in some Lycianthes taxa, notably L lycioides, but are variable in others, such as L pau- ciflora. Their presence or absence was noted by Bitter, Barboza and Hunziker (1992), and Dean (1995) in their species descriptions. The Calyx In bud, the calyx in Lycianthes and Capsicum is fused to the top (complete prefloration), and the corolla, stamens, and other interior parts egress from it by a stretching of the calyx apex. This re- sults in a truncate margin subtended by a thin sleeve of tissue with reduced vasculature. Low in the calyx, the five principal calyx traces each di- verge into three traces, much as in a foliage leaf, and the adjacent lateral traces are fused together, resulting in a calyx with 10 nerves. Of these, 5 primary ones are produced by the continuation of the principal calyx traces, and 5 are produced sec- ondarily by fusion of the adjacent lateral traces. In i?rt 1997 Benitez & D’Arcy Lycianthes in Venezuela 169 170 Volume 84, Number 2 1997 Benitez & D’Arcy Lycianthes in Venezuela 171 examined 25 species of Lycianthes. Among these were six that occur in Venezuela: amatitlanensis, asarifolia, acutifolia (conicibaccata), lenta (vari- ifolia), radiata (goudotii), and stenoloba (stephan- ocalyx). She concluded that the elaboration of hair types in Lycianthes evolved independently from that in Solanum. She hypothesized that in Lycianthes stellate hairs arose from branched hairs, unlike in Solanum where they arose from gland-tipped hairs. Unfortunately, she did not distinguish between the simple, eglandular hairs ( Fingerhaare ) character- istic of many species of Solanum and the coarse, tawny strigose hairs of some Lycianthes, which seem quite different in nature. These simple hairs in Ly- cianthes may also be independent in origin from Uses Fruits of some species, Lycianthes asarifolia i Venezuela and elsewhere, are sometimes eaten, an some, L moziniana (Dunal) Bitter and others i Mexico (Williams, 1993; Dean, 1995), even reac fruit markets. Lycianthes rantonnei (Carrifcre) Bitt< is grown as an ornamental in upland tropical ga dens. No other uses are known. Chemistry Lycianthes has a suite of alkaloids comparable to other genera of Solanaceae (Roddick, 1986), but few species have been examined (Bradley et al., 1978; Evans & Somanabandhu, 1980; Lin et al., 1987; Ripperger & Porzel, 1992), none of these species occurrii Systematics of Lycianthes Lycianthes was divided into a hierarchy of sub- genera, sections, and series by Bitter (1920), only some of which are represented in Venezuela. The subgenera and sections in Venezuela are amply dis- tinct and not likely to be confused. Within the sec- tions, however, distinctions are often difficult to make, and species limits are sometimes poorly known. The type species, Lycianthes lycioides, has ire enclosed in Dtes a perhaps s and section. In the other four sections, seeds are numerous and not enclosed in sclerenchyma. Section Asaropsis includes one or perhaps three species of ground creepers. Section Simplicipila includes perhaps a dozen species of weak shrubs with strigose hairs. Section Polymeris includes many species of night-blooming climbers. Series Oligochondra and Virgatae are separated by presence or absence of stone cells in the fruit. Subgenus Lycianthes (L lycioides): seeds enclosed Section Lycianthes (L lycioides): erect, woody shrubs; flowers diurnal; stamens unequal (3+2). Subgenus Polymeris (Dunal) Bitter: seeds numer- ous, not enclosed in sclerenchyma; fruits sometimes with stone cells. Section Asaropsis Bitter (L, asarifolia ): plants procumbent; leaves cordiform; flowers solitary, diurnal; stamens equal; stone cells wanting. Section Polymeris: shrubs or climbers; flowers fasciculate, nocturnal; calyx teeth 10 in 2 un- equal series; stamens unequal (4+1); stone cells often present. Series Oligochondra Bitter ( L ferruginea, L pauciflora, L stenoloba): plants high climbing or erect, small shrubs; flowers fasciculate, noc- turnal; fruits dangling, stone cells often pres- Series Virgatae Bitter (L lenta, L sanctaemar- thae): plants low climbers; flowers fasciculate, nocturnal; fruits dangling, stone cells wanting. Section Simplicipila Bitter (L acutifolia, L ama- titlanensis, L inaequUatera, L radiata)’, sub- shrubs; flowers fasciculate, diurnal; stamens equal; fruits held erect, stone cells present or Geography of Lycianthes Lycianthes is confined to the Neotropics and to southeast Asia. Most of the species and most of the diversity is in the New World, with distinctive groups in Mexico and Central America. Section Ly- cianthes (L lycioides) is restricted to South Amer- ican uplands. Section Asaropsis includes one (or two) species that grow in lowlands in eastern South America, Lycianthes asarifolia, of this section, was recently reported as introduced into Texas (Darwin & Feibelman, 1991). This section may also em- brace L lysimachioides (Wall.) Bitter, a wide-rang- ing southeast Asian species that has similar growth form and anthers, but which has well-developed teeth on the calyx. Section Polymeris and its similar series Oligochondra and Virgatae are widespread at lower and middle elevations in South and Central America and the Antilles. Section Simplicipila has a similar range but is not known from the Antilles, and it may not range north of Nicaragua. Scope of the Present Study Both authors have studied plants of Lycianthes in various countries for more than two decades, but the present focus that led to this paper dates from Volume 84, Number 2 173 Volume 84, Benitez & D’Arcy Lycianthes in Venezuela 175 Bocon6, 2200-2400 m, Steyermark & Manara 125326 (VEN). Yaracuy: Sierra de Area. 9 km W de San Mi,**. 1100-1500 rn. Liemer & GonzdUz 10038 (VEN). Zulias Swsare, 2700-3300 m. Wood & Berry 85 (VEN). 2. Lyrianthe* amatitlanenau (Coult. & J. D. Volume 84, Number 2 1997 Benitez & D’Arcy Lycianthes in Venezuela 1 77 3. Lycianthes asarifolia (Kunth & Bouch6) Bit- ter, Abh. Naturwiss. Vereine Bremen 24(1): 423. 1919 [1920]. Solatium asanfolium Kunth & Bouchl, Index Sem. Hort. Berol. p. 10. 1845. Solarium violifolium var. asanfolium (Kunth & BouchtS) Hassl., Repert. Spec. Nov. Regni Veg. 15: 221. 1918. TYPE: cultivated in Berlin from seed from Caracas (holotype, B destroyed, = F photo 2562). Solarium violaefolium var. majus Dunal, in DC., Prodr. 13(1): 164. 1852. TYPE: Bolivia. Santa Crux, d’Or- bigny 619 not seen. i 84, Number 2 Figure 7. Lycianthes asarifolia. Flowering and fruiting stems. After Licata et al. 32 (MY). Volume 84, Number 2 Benitez & D’Arcy 181 Figure 9. Lycianthes ferruginea. — A. Flowering and fruiting branch Benitez 5148 (MY). \t. — B. Opened flower. — C. Fruiting calyx. After 1997 Benitez & D’Arcy Lycianthes in Venezuela crescent, fixating pedicels 2 mm long; seeds ca. 70 per fruit, yellowish brown, 1.5-1 mm across, the con- spicuously thickened margin 0.5 mm wide. Figures 21 , 11 . Se caracteriza por ser pubescente, con hojas menores conspfcuas, los dientes del ciliz en ntimero de 10, liger- amente desiguales, en una serie y subulados, los cuales This species closely resembles L amatitlanensis, from which it differs in its lesser pubescence and shorter calyx teeth, which are usually less conspic- uous in fruit, and longer pedicels. Distribution. Venezuela and Bolivia. Cloud for- 1997 Benitez & D’Arcy Lycianthes in Venezuela 185 570: 1900. SYNTYPES: Mexico. Oaxaca: Palmer 533 (GH not seen); Tehuantepec, Seler 1625 (GH not Lycianthes lenta var. endopsila Bitter, Abh. Naturwiss. Ver- eine Bremen 24(1): 367. 1919 [1920]. SYNTYPES: Venezuela. Caracas, Humboldt 748 (B not seen); Var- gas s.ru (lectotype, designated here, US-601441). Solarium virgatum Lam. var. caracasanum O, E. Schulz, in Urhan, Symb. Antill. 6: 190. 1909. TYPE: Ven- ezuela. Near Cura, Humboldt 748 (holotype, P). Lycianthes lenta var. scotmophila Bitter, Abh. Naturwiss. Vereine Bremen 24(1): 367. 1919 [1920]. TYPE: Venezuela. Valle del Aragua bei San Mateo, Otto 788 (holotype, B destroyed). a (Vahl) Bitter subsp. tobagoensis Bit- i 24(1): 343. Garden Volume 84, Number 2 1997 Benitez & D’Arcy Lycianthes in Venezuela 189 Lycianthes pauciflora (Vahl) Bitter, Abh. Na- Solarium g turwiss. Vereine Bremen 24(1): 341. 1919 c J“*j [1920]. Not Sendtn. (1846). Solatium pauciflo- Cuuu rum Vahl, Eclog. Amer. 1: 20. 1796. TYPE: Martinique (holotype, C-hb Vahl). turn Vahl, Eclog. Amer. 1: 21. 17 ninata (Vahl) Bitter, Abh. Naturwi 1 24(1): 392. 1919 [1920]. TYPE: Rohr s.ru (C, = F photo 22887) Sol. 177. 1813. M3 31 Annals of the Missouri Botai 1,264, lo 192 closed during the day, are less conspicuous. The calyx has conspicuous, porrect, recurved or re- flexed, sometimes almost woody teeth. Distribution. Collections have been seen from Costa Rica, Panama, the Greater and Lesser Antil- les, and all tropical countries in South America ex- cept Chile and the Guianas, and it probably occurs in the latter region. Lycianthes pauciflora is the most wide-ranging species of the genus within Ven- ezuela, and it is variable as to its degree of pubes- cence and the size and form of its calyx teeth. Cloud forests and gallery forests 600-1500 m ele- vation. (Map, Fig. 18.) The species appears to flow- er and fruit throughout the year, but most flowering specimens were collected from April to July and in November and December. The concept employed here unites concepts of sev- eral regional treatments (D’Arcy, 1973; Barboza & Hunziker, 1992) under the oldest name for this wide- ranging species. The Panamanian plants for which D’Arcy used the name L guianensis Dunal have larg- er fruits and longer, thinner calyx lobes than those from the Antilles, Venezuela, and other eastern parts y represent a distinct t particulars, particularly in having fruits with 2 stone cells, the Venezuelan plants agree with the other plants discussed here. Plants of the species from the Antilles, particularly Dominica and Martinique, the type locality of L pauciflora, tend to have slightly smaller calyces and flowers and more rotund leaves than those of Venezuela and the Guian- as, but they are otherwise similar. Fruits examined from the Lesser Antilles had varying numbers of stone cells (Martinique, Duss 364 (US), no stone cells, Duss 4430 (US), 1 stone cell, Dominica, Ernst 1942 (US), 2 stone cells). Plants from lowland Paraguay, Bolivia, and Peru have dimensions like those of plants from Representative specimens examined. VENEZUELA. Distrito Federal: between el topo Macanillal & El Pico Izcaragua, 7—12 km E de los tanques de la Electricidad de Caracas, 700-800 m, Morillo et al 3272 (VEN). Ama- curo: Reserva forestal Sierra Imataca, 1988 m, Sanoja 2002 (MY). Amazonas: 5 to 7 km by river E of Cerro La Neblina, 140 m, Uesner & Funk 15838 (MO, MY, VEN). Apure: Reserva Forestal de San Camilo, Chiricoa, 200 m, Steyermark et al. 101704 (MO, NY). Aragua: Carre- tera Maracay-Choronf, 1200 m, Benitez et al. 4911 (MY). Soledad and Santo Domingo, 1300 m, van der Weiff & Ortega 6124 (MO, NY). Bolivar: El Dorado-La Gran Sa- bana, 1200 m. Bunting 2908 (MY). Carabobo: Colinas de Guaremales, en la Fortaleza, Pittier 8803 (VEN). Fal- con: Sierra de San Luis, S of La Tabla, 1450 m, Steyer- mark 98911 (MO). Lara: above Sanare toward Las Blan- quitas, 1500-1900 m, Badillo 6687 (MY). Merida: Carretera Estanquez-Las Nieves, 850 m, Benitez et al. 4826 (MY). Miranda: Parque Nacional Guatopo. La Ma- canilla, 600 m. Nee 17760 (MO, NY). Portuguesa: E of Chabasquen, 1450-1520 m, Steyermark et al. 126813 (VEN). Sucre: Cerro Patao, N of Puerto de Hierro, Stey- ermark & Agostini 91324 (VEN). Tachira: between Que- brada Grande and El Nula, border with Apure, 250 m. Gentry & Puig-Ross 14293 (MO). Trujillo: Vfa Escuque-El Socorro, Benitez 1952 (MY). Yaracuy: Ser- ranfa Santa Marfa-Certo La Chapa, 6 km N de Nirgua, 1200-1350 m, Meier et al. 3903 (MY, VEN). 9. Lycianthes radiata (Sendtn.) Bitter, Abh. Na- turwiss. Vereine Bremen 24(1): 433. 1919 [1920]. Solanum radiatum Sendtn., in Martius, FI. Brazil. 10: 53. 1846. TYPE: Hartweg 1293 [129 in publication] (BREM not seen, W not seen, B destroyed, = F photo 2586). Solanum goudoti Dunal, in DC., Prodr 13(1): 158. 1852. Lycianthes goudoti (Dunal) Bitter, Abh. Naturwiss. Vereine Bremen 24(1): 435. 1919 [1920]. TYPE: Co- lombia. Goudot 13 (holotype, G-DC not seen, - IDC microfiche, = F photo, 006772, W not seen). Lycianthes holocalyx Bitter, Abh. Naturwiss. Vereine Bre- men 24(1): 459. 1919 [1920]. TYPE: Ecuador. Prov- incia Santo Domingo, Sodiro 114/38 (B destroyed, - F photo 2577) & D’Arcy Volume 84, Number 2 1997 Benitez & D’Arcy Lycianthes in Venezuela 195 de Santa Marta in Colombia and in the Rfo Guasare watershed in Zulia State in Venezuela. Riverine for- ests from 500 to 600 m elevation. (Map, Fig. 22.) We have seen flowering specimens from May and June and fruiting specimens from August. Although Bitter reported an absence of stone cells in the fruits of this species, we found two in the fruit we examined ( Bunting & Kaujfman 10257). Representative specimens examined. VENEZUELA. Zulia: Disl. Mara, Cuenca de los rfos Socuy-Guasare, en la Paloma, 600 m. Bunting 10185 (MO), Bunting & Kauff- man 10257 (MO); E of rfo Guasare, 600 m, Steyermark et al. 122959 (NY, VEN); Cerro Los Manantiales, E of rfo ermark* aL 123281 (NY, VEN). 11. Lycianthes stenoloba (van Heurck & Muell.-Arg.) Bitter, Abh. Naturwiss. Vereine Bremen 24(1): 358. 1919 [1920]. Solanum stenolobum van Heurck & Muell.-Arg., Observ. 198 Annals of the Missouri Botanical Garden Aymard & Flores 216 (MY, PORT, VEN) (7). Aymard et al. 1643 (MY, PORT) (8); 2839 (MY, PORT) (3). Badillo 1816 (MY) (6); 1914 (MY) (8); 4426 (MY) (8); 5672 (MY) (4); 6584 (MY) (7); 6636 (MY) (8); 6687 (MY) (4). Badillo & Holmquist 6217 (MY) (8). Badillo et al. 7841 (MY) (8). Benftez 318 (MY) (6); 517 (MY) (3); 683 (MY) (7); 1181 (MY) (8); 1335 (MY) (8); 1353 (MY) (8); 1408 (MY) (4); 1418 (MY) (1); 1446 (MY) (4); 1551 (MY) (1); 1554 (MY) (6); 1562 (MY) (8); 1952 (MY) (8); 2068 (MY) (4); 2243 (MY) (8); 2614 (MY) (6); 3214 (MY) (6); 3621 (MY, NY) (3); 3868 (MY) (1); 3869 (MY) (4); 3898 (MY) (1). Benftez & Otero 4609 (MY) (1). Benftez & Pons 4651 (MY) (3); 4654 (MY) (11). Benftez & Rojas 3087 (MY) (8); 3993 (MY) (3); 5000 (MY) (8). Benftez et al. 3221 (MY) (3); 4184 (MY) (7); 4185 (F, MO, MY, NY, VEN) (9); 4235 (MY) (4); 4236 (MY) (1); 4259 (MY) (3); 4261 (MY) (8); 4611 (MY) (9); 4847 (MY) (9); 4884 (MY) (8); 4911 (MY) (8); 5104 (MY) (4); 5117 (MY) (4); 5129 (MY) (4); 5148 (MY) (4). Bemardi 433 (MER) (1); 1834 (VEN) (1); 2327 (NY, MER) (8); 5687 (NY) (8); 5792 (MER, NY) (3); 5930 (MER) (8). Berry 940 (VEN) (4); 944 (VEN) (1). Bianco 39 (CAR, VEN) (3); Bianca 302 (MER, MO, VEN) (8). Bonpland 71 (P) (6). Breteler 3628 (MER, VEN) (1). Bunting 2908 (MY) (8); 3020 (MY) (8); 4433 (MY) (6); 5808 (VZU, MO, VEN) (6). Bunting 10185 (MO) (10). Bunting & Kauffmann 10257 (MO) (10); Bur- kart 16324 (VEN) (6). Cardenas 4038 (MY) (8). Castillo 1930 (MY) (9). Car- nevali et al. 614 (MY, VEN) (8). Cesari (VEN-249120) (1). Colonello 930 (CAR) (8). Croat 54841 (VEN) (5). Cu- mana 1850 (IRBR, MY) (6). D’Arcy & Benftez 18257 (MO, MY) (9). Davidse 4035 (VEN) (4). Davidse & Steyermark 18168 (VEN) (7). Dav- idse & Gonz4lez 18899 (NY, VEN) (5); 21914 (NY) (8). Davidse & Miller 27476 (MO, MY) (8). Delascio 51 (CAR) (3) ; 122 (CAR) (1); 981 (CAR) (4); 9026 (CAR) (4). De- lascio & de Delascio 2692 (CAR, VEN) (3). Delascio & L6pez 12838 (MO, VEN) (3). Demey (CORO, MY-86695) (4) . Diederichs 173 (VEN) (4); 177 (VEN) (8); 270 (VEN) (8); 279 (VEN) (8). Dorr & Barnett 7183 (MER, MO) (8). Dorr et al. 4749 (MY, NY) (8); 5104 (NY, VEN) (9); 5263 (NY) (7). Duno et al. 184 (MY) (6). Edwards et al. 94 (MY) (8). Fendler 974 (MO, P) (1); 991 (G, MO, NY) (4); 1065 (NY) (3). Fernandez, A. 379 (MY) (3); 497 (MY) (6); 606 (MY) (8); 620 (MY) (8); 1112 (MY) (8); 3697 (MY) (1); 3762 (MY) (8); 3963 (MY) (8); Fernandez, F. 98 (VEN) (8). Ferrari 733 (MY) (8); 790 (MY) (3); 791 (MY) (11); 859 (MY) (7). Gentry & Puig-Ross 14293 (MO) (8). Gentry & Stein 47275 (MO, MY, VEN) (8). Humbert 26108 (MER, P) (1); 26156 (MER, P) (1). Humboldt 748 (P) (6). Ijjasz & Madriz 164 (MY, VEN) (7). Jahn 1211 (VEN) (6); 1248 (VEN) (6). Jeffrey & Trujillo 2510 (MY) (8). Knapp & Mallet 6766 (BH, MY) (2); 6807 (BH, MY, VEN) (1). Lasser 1068 (VEN) (4); 1203 (VEN) (4); 2044 (VEN) (8); 2212 (VEN) (8). Licata et al. 32 (MY, PORT) (3). Liesner 10038 (VEN) (1); 12833 (MY, VEN) (5). Liesner & Funk 15838 (MO, MY, VEN) (8). Liesner & Gonzalez 9732 (VEN) (8); 9773 (VEN) (2); 9969 (VEN) (8). Liesner & Guariglia 11631 (MO, NY, VEN) (8). Liesner et al 7803 (MY, MO, VEN) (3). Linden 437 (G) (4); 478 (P) (3). L6pez-Figueiras & Rodriguez 9080 (MERF) (7). L6pez- Palacios 1457 (MERF, MO, MY) (7); 1539 (MO) (4); 2206 (MERF, MY, VEN) (8). Ldpez-Palacios & Bautista 3485 (MER) (8). Magallanes (MY-9313) (6). Manara (MY-83285) (11); (VEN-71770) (3); (VEN-176544) (3). Marcano-Berti 1409 (MER) (8). Marcano-Berti & L6pez-Palacios 1758 (MER, MY) (4). Marcano-Berti & Carrillo 29-4-78 (MER) (8). Mocquerys 880 (P, VEN) (6); 978 (MY, P) (11); s.n. (P) (6). Montaldo 3756 (MY) (8). Morillo 2519 (VEN) (2); 3384 (VEN) (1); 11148 (MERF, MY) (7). Morillo & Garda 11472 (MERF, MY) (1). Morillo & Manara 1600 (MY, VEN) (1); 2022 (VEN) (3); 2070 (MY) (3). Morillo & Seres 8614 (VEN) (1). Morillo & Smith 6057 (MY, VEN) (4). Morillo et al. 2940 (VEN) (8); 3272 (VEN) (8). Moritz 1642 (G, P) (4). Nee 31048 (MO, MY, NY, VEN) (8). Nee & Whalen 16899 (MO, NY) (8); 17053 (VEN) (7); 17146 (NY, VEN) (3); 17760 (MO, NY) (8). Nilsson & Steyermark 221 (VEN) (8). Ortega 892 (VEN) (3). Pannier 199 (MERF, VEN) (5). Pefour 6 (MERC) (7). Penaloza 205 (CAR) (1). Pietrangeli 338 (MY) (9); 1331 (MY) (9). Pittier 5972 (NY) (3); 8803 (VEN) (8); 9129 (NY, US, VEN) (3); 9378 (NY, VEN) (1); 10036 (VEN) (1); 10057 (NY, VEN) (1); 11179 (VEN) (6); 11867 (M0, NY, VEN) (8); 13088 (MO, NY, VEN) (6); 13237 (M0, NY, VEN) (7); 13514 (MO, VEN) (4). Pittier & Nakichen- ovich 15543 (VEN) (8). Plowman 7766 (P, MO, NY) (1). Plowman et al. 13445 (NY) (8). Poelt & Oberwinkler 14991 (VEN) (7). Quintero 2173 (MER) (9). Quintero & Hemdndez 246 (MER) (9). Ramfrez 2071 (MY) (4). Ricardi 1093 (MERC) (1). Rodriguez, G. 770 (MY, VEN) (9); Rodriguez, H. 74 (MY) (3). Rodriguez et al. 1346 (MY) (1). Rodriguez & Cardozo 1712 (MY) (1). Romero 486 (MY) (6); 854 (MY) (8). Ro- sales 11 (MY, VEN) (4). Ruiz-Terdn 645 (MER) (8); 2128 (MERF) (1); 8802 (MERF) (7). Ruiz-Terdn & L6pez-Fi- gueiras 64 (MERF, MY) (1); 245 (MER, MERF, MY) (7). Ruiz-Ter4n & L6pez-Palacios 6708 (MERF) (7); 11470 (MERF, MY) (8); 12533 (MERF, MY) (7). Saer 118 (VEN) (6); 263 (NY, VEN) (3). Sanoja, E. 2002 (MY) (8). Schnee 830 (MY) (4). Schwarzkopf 12 (MERF, MY) (9); 48 (MY) (1). Smith V1341 (MY, UCOB) (1); V4464 (MY, UCOB) (8); V4465 (VEN) (4); V4466 (VEN) (4); V7557 (MY, VEN) (4). Smith V9384 (MY) (8). Sobel et al. 2062 (NY) (3). Stein & Gentry 1512 (MO, MY, VEN) (8). Stergios & Aymard 4451 (MY, PORT) (3). Stergios et al. 6360 (MY, PORT) (5). Steyermark 55405 (MY, VEN) (1); 55945 (MY, VEN) (4); 56015 (MY) (1); 56262 (VEN) (7); 56494 (NY, VEN) (4); 56628 (F, MY, VEN) (9); 56632 (VEN) (4); 56939 (MY, VEN) (4); 56957 (VEN) (4); 57433 (MY, VEN) (4); 60971 (MY, VEN) (3); 86204 (NY, VEN) (4); 87105 (VEN) (8); 89040 (VEN) (8); 89896 (NY, VEN) (8); 89951 (NY, VEN) (8); 90991 (VEN) (8); 91601 (NY) (1); 98911 (MO) (8); 99013 (VEN) (8 ; 102180 (MO, NY, VEN) (6); 106197 (MO, NY, VEN) (8); 106473 (MO, NY) (1); 115464 (MO, VEN) (4); 119551 (MO, VEN) (8); 120002 (VEN) (8); 121528 (VEN) (2 ; 121593 (VEN) (2); 121927 (VEN) (1); 126813 (VEN) (8 ; 126823 (VEN) (8). Steyermark & Agostini 91324 (VEN) (8). Steyermark & Davidse 116492 (VEN) (3); 116662 (VEN) (8). Steyermark & Espinoza 105842 (VEN) (&)■ Steyermark & Liesner 118330 (VEN) (9); 118386 (MO, VEN) (9); 119186 (VEN) (8); 120743 (VEN) (8); 120772 (MO, NY, VEN) (2). Steyermark & Manara 110995 (MO, NY, VEN) (6); 125326 (VEN) (1). Steyermark & Mass 123694 (MY, VEN) (8). Steyermark & Stoddart 118046 Volume 84, Number 2 1997 Benitez & D’Arcy Lycianthes in Venezuela (VEN) (8). Steyermark & Steyermark 95433 (VEN) (3). Steyermark & Rabe 96125 (P, VEN) (2); 96718 (VEN) (8). Steyermark & Wessels-Boer 100481 (MO, NY, VEN) (8). Steyermark et al. 92969 (VEN) (8); 98795 (MY, VEN) (4); 100742 (MY, VEN) (1); 101704 (MO, NY) (8); 102599 (VEN) (8); 104542 (VEN) (8); 105798 (MO) (8); 109926 (VEN) (3); 110242 (MY, VEN) (4); 122959 (NY, VEN, VZU) (10); 123500 (MO, VEN) (1); 124590 (MY, VEN) (3); 124685 (MY, VEN) (8); 124991 (VEN) (3); 126679 (VEN) (8); 127216 (MY, VEN) (3). Suarez & Gil 53 (MER) (7). Tamayo 283 (VEN) (6); 397 (VEN) (4); 1229 (VEN) (1). Trujillo 1260 (MY) (4); 2917 (MY) (1); 3325 (MY) (3); 5237 (MY) (8); 6358 (MY) (7); 13947 (MY) (8); 15829 (MY) (8); 15967 (MY) (3). Trujillo & Bunting 7598 (MY) (1). Trujillo & del Castillo 8265 (MY) (1). Trujillo & Fer- ndndez 362 (MY) (1); 368 (MY) (1); 372 (MY) (1); 10597 (MY) (3). Trujillo & Ponce 19715 (MY) (3). Trujillo et al. 16574 (MY) (6); 16662 (MY) (6). van der Werff 4767 (VEN) (8); 4811 (VEN) (8); 5453 (VEN) (8); 6124 (MO, NY) (8). van der Werff & Vera 864 (CORO, MY) (11). Velasquez 110 (CAR) (8); 310 (CAR) (1). Whetzel & Muller 485 (VEN) (1). Williams 10269 (VEN) (8); 10789 (VEN) (8); 11084 (MY, VEN) (3). Wing- field 12989 (CORO, MY) (11). Wood 448 (VEN) (3). Wood & Berry 85 (VEN) (1). Xena 626 (MY) (4). iiiissgsssssSsisSiisiiSs i s 5S ss s Missouri Botanical Garden DIALYPETALANTHUS FUSCESCENS KUHLM. (DIALYPETALANTHACEAE): THE PROBLEMATIC TAXONOMIC POSITION OF AN AMAZONIAN ENDEMIC 1 Volume 84, Number 2 1997 Piesschaert et al. 203 Dialypetalanthus fuscescens from Amazonia scope. In order to show the thickenings of the inner tangential walls of the exotestal cells, the outer tan- gential walls had to be removed. Therefore, boiled seeds were transferred to a 3:1 mixture of alcohol 96% and acetic acid 99% for 24 hours. Afterward they were put in a 3% aqueous mixture of sulphuric acid at 50°C for at least 10 hours (modified from Braune et al., 1967). Remaining specks of dirt were removed by ultrasonication. The methods used for acetolysis and breaking pollen grains are discussed in detail by Huysmans et al. (1994). Wood anatomical sections were pre- pared as described by Jansen et al. (in prep.). Results VEGETATIVE STRUCTURES Habit. Dialypetalanthus fuscescens is a large, usually smaller). It has a soft, fibrous (because of the formation of several concentric phellogen lay- ers), red to cinnamon-colored bark. The trunk is fluted toward the base. The wood is light but ex- tremely hard. It is used locally for house construc- tion (Rizzini & Occhioni, 1949), but as far as we know it has little economic value. Leaves. The simple leaves are decussately ar- ranged and have blades with entire margins. The petiole is well developed (0.6-3.5 cm) and shallow- ly sulcate above. The blades are broadly elliptic (L/W-ratio 1.5:1), elliptic (2:1), suborbiculate (1.2: 1), to narrowly (2:1) or widely obovate (1.2:1) (sometimes slightly asymmetric), with a shortly acu- neate to slighdy decurrent base (terminology follow- ing Hickey, 1988); they are (2-)6-17(-20) X (1-) 4— 11(— 14) cm. The leaf venation is pinnate and camptodromous. Hie veins, especially the midvein (Fig. 1A), are prominent on the abaxial side of the blade. The divergence angle between primary and secondary veins is about 35° in the center of the blade and gradually rises toward the base of the leaf. The veins are detectable up to the fifth order. la in size (upper ones much larger than the short, tri- angular lower ones; Fig. 2). Anatomically, the stip- ules have a characteristic adaxial epidermis, con- sisting of thickened cells. Scattered through the homogeneous chlorenchyma are numerous small vascular traces that are surrounded by fibers. Riz- zini and Occhioni (1949) recorded occasioned par- acytic tain silica-bodies (Fig. 4G, H). The perforation plates of the vessels are simple, bearing a single elliptical to almost circular opening (Fig. 5A). The vessel-ray pits are simple (Fig. 5B). The intervessel pits are alternately arranged and vestured (Fig. 5C). Internal phloem is absent. REPRODUCTIVE STRUCTURES Wood, anatomy. The wood of Dialypetalanthus shows indistinct growth rings (probably reflecting the precipitation cycle), marked by a transition from thin- to thick-walled fibers. The fibers are sep- tate (Fig. 4C) and have simple pits in vertical rows (Fig. 5D) (libriform fibers sensu Baas, 1986). The wood is diffuse-porous. The vessels are solitary or arranged in short radial rows (2-8 cells) (Fig. 4D, E). In young wood, close to the pith, the radial vessel rows are often longer. The outline of the sol- itary vessels is rounded (Fig. 4E), although the smaller vessels are often compressed between the larger ones. Axial parenchyma is present and oc- curs as scanty, paratracheal strands (Fig. 4B). The width of the rays varies from (l-)2 to 6 cells (Fig. 4A); their height may exceed 100 cells. In radial sections, the procumbent body-ray cells have a margin of one or often several layers of square cells (Fig. 4F). Sometimes a mixture of procumbent and more or less square cells occurs. The rays are vis- ible to the naked eye as clear, narrow, parallel lines. The pith and the rays contain very small cu- bic to prismatic crystals; navicular crystals occur as well. The septate fibers as well as the rays con- Inflorescence. The bloom of Dialypetalanthus is large inflorescences. The inflorescence of Dialype- talanthus (Figs. 6-8) was described by Kuhlmann (1925) as paniculate and racemose. Rizzini and Oc- chioni (1949) and Hutchinson (1959) also indicated it was a panicle. According to Weberling (1992), a panicle is characterized by terminal flowers on its main axis and side branches. Thus, a panicle is a determinate inflorescence. Dialypetalanthus, how- ever, does not have a terminal flower but a terminal bud that at first sight may be confused with a single terminal flower. Dissection of the terminal bud re- veals a floral meristem where acropetal inception of lateral flowers occurs (the flowers in the most terminal zone seem to be poorly developed) (Fig. 7). The inflorescence of Dialypetalanthus is thus indeterminate, contrary to a panicle. Dahlgren and Thome (1984) provided the first correct illustration of the inflorescence but did not describe it. Cron- quist (1981) was the first to correctly call the Di- alypetalanthus inflorescence a thyrse. More specif- ically, it is a frondobracteose, heterothetic, indeterminate thyrse with opposite branches (Fig. 206 Annals of the Missouri Botanical Garden !iH«»ii! !? i lilUM Garden Volume 84, Number 2 1997 Piesschaert et al. 209 Volume 84, Number 2 1997 Piesschaert et al. 213 Dialypetalanthus fuscescens from Amazonia Figure 13. Pollen morphology of Dialypetalanthus fuscescens (SEM). A-D. Entire pollen. E, F. Broken Polar view with well-developed apocolpium.— B. Equatorial view of ecto- and mesoaperture.— C. Detail ol pium.— D. Detail of the mesocolpium.— E. Endocolpus.— F. Endexine ornamentation; note the difference warty (white star) and weaker looking zone at the top of the endocolpus (black star). After de Albuquerque et al. 1295. .in 216 A B Figure 15. Fruit structure of Dialypetalanthus fuscescens. — A. Fruit in late stage of opening, showing deep septicidal and more shallow loculicidal slits. — B. Septal view of a half fruit in the same stage after artificial separation; cr = calyx remnants and scars, st = septum, se = seeds, pi = placenta. After Silva & Mono 3238 (A) and Silva 740 (B). per style) and pollen sacs closely associated with the filament instead of the versatile anthers that are typical of wind-pollinated species. Moreover, should bear in mind that Dialypetalanthus is a i forest tree. The dense vegetation of mostly < green species in rainforests, the large amount of precipitation, and the low turbulence due to the closed canopy, by which pollen cannot stay in the air long enough to cause successful pollination, make wind pollination unlikely. All of these mor- phological and ecological data strongly suggest an- imal pollination. A plausible type of pollination is cantharophily. Beetles are especially attracted to white, fragrant flowers with abundant and easily at- tainable pollen (Willemstein, 1987). All of this is offered by Dialypetalanthus . Grant (1950) stated that an inferior ovary may be an adaptation to de- structive pollinators (in this case beetles) as a pro- tection for the vulnerable ovules. Cantharophily would be a good explanation for the simultaneous ovary in Dialypetalanthus. The rough flower treat- ment that is so typical for beetles would also ex- plain the earlier-mentioned lignified zones in the stamens and the overall firm, fleshy structure of the On the other hand, pollination by bees seems to be an acceptable alternative, because the poricidal anthers may point to buzz-pollination. Moreover, D’Arcy et al. (1996) stated that the occurrence of oxalate packages is rather typical for bee-pollinated species. The pollen becomes mixed with the cal- by the insect, although it is still unclear what the crystals are used for. Field observations are needed to establish the pollination strategy of Dialypeta- lanthus. Distribution and Ecology In the past it was assumed that Dialypetalanthus occurs mainly in the eastern part of Brazil, around Betem (Robbrecht, 1994; Nicholas & Baijnath, 1994). This was probably due to the fact that the first collections ( Kuhlmann 1514; Ducke 17921 , 23660, and 21684) were from this region. The specimens presently available reveal, however, that the center of distribution is in the northern border Volume 84, Number 2 Piesschaert et al. 217 Dialypetalanthus fuscescens from Amazonia Botanical Garden In the CONSPECTUS OF THE GENUS PALICOUREA (RUBIACEAE: PSYCHOTRIEAE) WITH THE DESCRIPTION OF SOME NEW SPECIES FROM ECUADOR AND COLOMBIA 1 226 Annals of the Missouri Botanical Garden Figure 1. Approximate number of species of Palicourea by country, for the entire range of the genus. Numb followed by “+” represent conservative estimates for country floras not yet surveyed in detail. pears to be an important factor in the diversification of this genus. Adaptations for hummingbird polli- and even actively defend, Palicourea plants with green inflorescences and relatively short white co- rollas (e.g., P. calophlebia Standi., Colombia, pers. obs.). Phytogeography and Ecology of Palicourea Species of Palicourea are concentrated in tropi- cal South America; they are also well represented in Central America, particularly in montane Costa Rica and Panama (Fig. 1). Eight species are found in the Antilles, most of them in the Greater Antil- les, and more than 32 species are now known from Central America (Taylor, 1989, 1990). In South America, Palicourea is widespread in low-elevation wet tropical forests and also shows notable centers of species diversity in wet tropical montane areas. Palicourea is more species-rich in montane than lowland habitats. When montane species are de- fined as those predominantly distributed at or above 1000 m elevation (Gentry, 1988), 22 of the 26 spe- cies known from Costa Rica can be considered montane (Burger & Taylor, 1993), as can ca. 50 of the ca. 80 species known from Ecuador (Taylor, in prep.). The two subgenera recognized below ap- proximate this habitat distribution: of the 188 spe- cies of Palicourea classified below, 52 are placed uted at low elevations, while 136 are placed in sub- genus Montanae, which is primarily montane. Po*~ icourea is usually well represented locally in montane forests in both number of individuals and number of species (e.g., Kappelle & Zamora, 1995; Silverstone-Sopkin & Ramos-P6rez, 1995). The dis- Taylor Conspe i of Palicourea Taylor Conspectus of Palicourea Taylor . nov. TYPE: Pali- ^stur“ & Pav.) DC. Southwestern Taylor Taylor Missouri Taylor Consp€ scale; B, E, G to 5-mm scale. i 84, Number 2 Taylor faaSS licourea lugoana C. M. Taylor, sp. nov. TYPE: Taylor Taylor Conspt n,E,4,f Taylor Garden PALYNOLOGY, Peter Goldblatt 2 and Annick Le PHYLOGENETIC RECONSTRUCTION, AND CLASSIFICATION OF THE AFRO-MADAGASCAN GENUS ARISTEA (IRIDACEAE ) 1 264 Annals of the Missouri Botanical Garden Table 1. Species of Aristea examined, with voucher data and pollen grain dimensions. Voucher specimens are located at MO, with additional duplicates often at NBG, P, and PRE (herbarium acronyms after Holmgren et al. (1990)); collectors are abbreviated as follows G = Goldblatt, M = Manning. Species are arranged alphabetically within the sections recognized by Weimarck (1940 ). Aristea fimbriata, unknown to Weimarck, is assigned to section Racemosae, Section Eucapsulares Goldblatt (= sect. Euaristea Weim.) 58.5 (A. angustifolia Baker, / Thomas, 1992a) all h Section Trilobalae Weim 54.3 X 51.6 60.0 X 57.0 61.0 X 51.0 51.0 X 45.0 57.7 X 40.0 66.7 X 60.0 59.5 X 56.5 kitchingii Baker, S. Africa, W. Cape, Taylor 11009 (MO) S. Africa, Natal, G&M 8360; 9818; 9870 S. Africa, Natal, G&M 9857 S. Africa, Natal, G&M 9858 Malawi, Bidgood et al. 1310 Tanzania, la Croix 4267 Madagascar, Malcomber 1336 S. Africa, E. Cape, G&M 9588 S. Africa, E. Transvaal, G&M 9815; 9831 ind A. madagascariensis Baker (Goldblatt & L 1 diameter.) . Africa, E. Cape, van Wyk & Mathews 7727 Section Pseudaristea Pax A. biflora Weim. 71.2 X 70.5 A. cantharophila Goldblatt & J. Manning 69.2 X 66.4 A. ecklonii Baker 63.0 X 59.0 A. lugens (L.f.) Weim. 92.2 X 81.0 A. pauciflora W. Dod. 73.5 X 63.5 A. pusilla (Thunb.) Ker 59.6 X 58.5 A. simplex Weim. 71.1 X 67.5 A. spiralis (L.f.) Ker Gawl. 73.5 X 63.5 A. teretifolia Goldblatt & J. Manning 79.3 X 77.1 A. sp. ? off. pauciflora 81.2 X 76.2 also^ Goldblatt & Le Thoiias, 1992a) S. Afr S. Africa, S. Afr S. Africa, W. Cape, Goldblatt 8898 W. Cape, G&M 10284 E. Cape, ex hort W. Cape, Oliver 4739 W. Cape, G&M 10102 W. Cape, Bayliss 7635 W. Cape, G&M 9754 W. Cape, G s.n. (Cape Poin W. Cape, Bean 2785; Nann Cape, Drewe 466 A. confusa Goldblatt A. fimbriata Goldblatt ined. A. major Andrews Goldblatt Baker 54.0 X 51.1 73.5 X 70.5 54.0 X 49.5 52.5 X 48.0 S. Africa, S. Africa, S. Africa, S. Africa, S. Africa, S. Africa, W. Cape, G s.n. (Hout Bay) W. Cape, G&M 10167 W. Cape, Williams 891; Orchard 354 Cape, G s.n. (Hout Bay) W. Cape, G&M 9476A Cape, Oakes s.n. A. africana (L.) Hoffmsg. A. dichotoma (Thunb.) Ker Gawl. A. glauca Klatt A. oligocephala Baker 95.2 X 92.2 S. Africa, W. Cape, G&M 9352 ; 82.5 X 81.0 G&M 9505 ; 87.0 X 83.2 Bean 2789 ; 88.5 X 77.2 G&M 9750 97.5 X 93.0 S. Africa, W. Cape, G&M 9503; G&M 10154 82.5 X 75.0 S. Africa, W. Cape, G&M 9595A 97.5 X 85.5 S. Africa, W. Cape, Barker 412 Volume 84, Number 2 Goldblatt & Le Thomas 267 1997 Palynology of Aristea Table 2. Continued. 1 11111 11112 22222 22223 333 anceps 11121 10010 01000 00221 00001 1000? 100 cantharophila 20101 10000 32202 10200 00001 0000? 102 fimbriata 00010 ?1001 11??1 00100 00121 010?? 100 glauca 30000 11001 10121 00012 00011 0100? 100 lugens 20000 11000 32202 10200 10001 0000? 101 spiralis 21120 01000 32202 10201 01001 00001 101 S X ae M Hill Jim 000S0 Jo200 HZ HZ III goetzei 11120 10021 00000 00201 00001 0000? 100 cladocarpa 11121 10011 21000 00101 00001 0000? 100 madagascariensis 11120 10021 00000 01200 00001 0000? 100 kitchingii 11120 10021 00000 01100 00001 0010? 100 Xf HZ HZ ZS Zi HZ Z ranomafana 11120 10021 00000 01010 00001 0010? 100 racemosa 00010 00001 11011 00200 00101 001?? 100 pusilla 11121 10110 21000 00101 00001 0000? 100 angolensis 11111 10011 00000 00100 00011 00001 100 rr 11120 ?0011 ooooo 00200 ooooi oooo? 100 r “wdS” Jy’x^ analy'ds '(Swoflorf, became impractical to perform. 1997 Goldblatt & Le Thomas Palynology of Aristea 269 270 Missouri Botanical Garden Goldblatt, 1994; Le Thomas et al., 1996) in which there are two clearly defined zonasulculi (Figs. 25- 34), an extremely rare type of pollen grain in flow- ering plants. Two-zonasulculate apertures occur in all se species of section Pseudaristea (Table 1). Here, margins are usually clearly defined (Figs. 26, 31, 34), but in A. spiralis (Fig. 35) and A. paucifiora Volume 84, Number 2 1997 Goldblatt & Le Thomas Palynology of Aristea 271 (Le Thomas et al., 1996) they are somewhat diffuse of A. simplex, 2-zonasulculate grains occurred to- and the apertural membrane is covered with small, gether with some 1-zonasulculate grains. The scattered fragments of exine. This aperture type is 1-zonasulculate condition may represent the inter- usually constant within a species, but in our sample mediate phase in the evolution of the 2-zonasul- Goldblatt & Le Thomas Palynology of Aristea 273 Goldblatt & Goldblatt & Le Thomas CHROMOSOME CYTOLOGY OF IRIDACEAE— PATTERNS OF VARIATION, DETERMINATION OF ANCESTRAL BASE NUMBERS, AND MODES OF KARYOTYPE CHANGE 1 Gard. 84: 285-304. 1997. Muell.) Klatt 288 Annals of the Missouri Botanical Garden Diploid Species number 2 n 0. acorifolius (Kunth) Raven- 54 Collection data Venezuela, Merida, Grifo & Hahn 361 (BH) Cav. Venezuela, Dorr et al. 5044 (NY) Alophia drummondii (Gra- ham) R. C. Foster Calydorea azurea Klatt . xiphioides (Poepp.) Espino- 42 U.S.A., Texas, Bastrop Co., Lee sub Goldblatt s.n. Uruguay, Treinta y Tres to Tacuarembd, Castillo 1146 sub Gold- blatt s.n. Argentina, Entre Rios, Concepcidn del Uruguay, Goldblatt s.n. Argentina, Cordoba, Cerro Colorado, Goldblatt s.n. Chile, Coquimbo, Hoffmann s.n. (Standi.) Ravenna Cypella fucata Ravenna C. herbertii subsp. brevicrista- ta Ravenna subsp. woljheugelii (Hauman) Ennealophus e seb.) Raven E. foliosus (Kr Herbertia lahu nearifolius G. J. Lewis Watsonia dubia Eckl. ex Klatt 18 W. hysterantha Mathews & L. 18 -I Bolus W. minima Goldblatt 18 Tribe Ixieae Crocus longiflorus Raf. 28 Dierama inyangense Hilliard 20 Geissorhiza callista Goldblatt 26 G. foliosa Baker 26 G. roseoalba (G. J. Lewis) 26 Goldblatt Gladiolus aquamontanus 30 Goldblatt & Vlok Brazil, Maranhao, near Imperatriz, Plowman et al 9305 Honduras, Nelson s.n. (no voucher) Uruguay, Maldonado, Punta del Este, Castillo s.n. Argentina, Buenos Aires, Cerro Ventana, Lamberto & Mochel s.n. (BB 3881) Argentina, Misiones, Gamichos, Castillo s.n. (FAA) Bolivia, Tarija, Arce, Solomon 9972 Peru, Dillon 4514 (F) Chile, Omduff 9153 (UC); U.S.A., Louisiana, Shreveport, Hei- kamp, s.n. Brazil, Rio Grande do Sul, Rosario do Sul, Castillo s.n. Uruguay, Maldonado, Punta del Este, Castillo sub Goldblatt s.n. Brazil, Rio Grande do Sul, Uruguaiana, Goldblatt s.n. S. Africa, W. Cape, Cape Point Reserve, Goldblatt s.n. (no voucher) S. Africa, W. Cape, Cape Point, Goldblatt 5400 S. Africa, W. Cape, Malmesbuiy, Goldblatt 8708 S. Africa, W. Cape, Langebaan, Snijman 71 (BG) S. Africa, W. Cape, near Greyton, Goldblatt 8047 Italy, Sicily, Eloro, Goldblatt 5073 Zimbabwe, Nyanga, Clarke s.n. (no voucher) S. Africa, W. Cape, near Grayton, Goldblatt 8680 S. Africa, W. Cape, Strawberry Hill, Goldblatt 7948 S. Africa, E. Cape, Pootjeshoogte, Vlok 1663 S. Africa, W. Cape, Rust-en-Vrede, Vlok s.n. Goldblatt & Takei N.E B, 1 (L.) Garden Goldblatt & Takei Chromosome Cytology of Iridaceae Table 2. Contir Nemastylis (3/5) Onira (0/1) Sessilanthera (2/4] Tigridia (14/35) Goldblatt, 1982a Bamardiella (1/1) Belamcanda (1/1) Bobartia (8/14) Dietes (6/6) Ferraria (10/10) Galaxia (14/15) Gynandriris (7/9) Hermodactylis (1/1) Hexaglottis (6/6) Homeria (34/34) 9(8, 7, 6) 6(7,5) 20, 40, 60 20, 40, 60 18, 16, 14, 12 12, 24, 10, 9, 8 Moraea (95/130) 10(9, 8, 7, 6, 5) 20, 40, Pardanthopsis (1/1) 16 32 Roggeveldia (2/2) 6 12,24 Goldblatt, 1979c, 1984a Goldblatt, 1980b Simonet, 1932 Goldblatt, 1987 Goldblatt, 1980a, 1981b this large genus; both polyploidy and dysploidy are , 12, 24, 48, 10 Goldblatt, 1976, 1986a, 1986b Simonet, 1932 Goldblatt, 1992 •Goldblatt & Snow (1991) have shown that the plants associated with the counts of In = 12(-14) for Cipura paludosa nd 2/i = 14 for Eleutherine bulbosa (Sharma & Talukdar, 1959) were confused with one another. Counts of 2 n = 14 re for C. paludosa, while those of 2n = 12(-14) are for E. bulbosa. The count of 2 n = 14 for the latter reported by lao (1969) is probably also for C. paludosa. '.ft. V ;c © • •V L* V * r ■ . .* *5 f \ ^ * © © * *>V'\ ¥ S 0 V • # •» i * * © *' •*, > • & Figures 1-9. Mitotic metaphase in Iridaceae subfam.ly Niven, o.deae iAnstea and raursonu,) ana mao.aeae m, Sisyrinchieae (Orthrosanthus).-!. Aristea abyssinica (2n = 64).-2. A. anceps (2 n = 32). 3. A. angolensis (2n 32). — 4. A. juncifolia (2 n = 32).— 5. A. angustifolia (2n = 32J.-6. Patersonm senoea (2 n - 22).— 7. Orthrosanth polystachyus (2 n = 84).— 8. 0. aconfolius (2 n = 54).— 9. 0. chmboracensu (2n - 54). Vouchers as given in Table Scale bar, 10 gm. 1997 Goldblatt & Takei Chromosome Cytology of Iridaceae callista. One population examined has a karyotype African species are needed to elucidate the signif- consisting of matching pairs (Fig. 46A), but another icance of this situation. is karyotypically heterozygous (Fig. 46B) and has Unusual in Iridaceae, the only three species of one long metacentric chromosome and one very Homeria that exhibit dysploidy, H. pallida, n — 6 short metacentric. Further studies in this tropical and 4, and H. tenuis and H. ftavescens, 2 n = 10, G> 1 ( ® tVJ 1 -7r.< ‘*V have dipbad number* of 2a - 2b. 28, 40. 52. Si. 60. and 90 (Table 2i. m» another 1W •• aU a iwuarialde ton A Hevwnod. 1 Wit TW two aperies v hers baaed on 10. T mm rtumwm (2a * and f amrwnsau (2a • 60». alao have d* Oaiassa appears to be a dyspload genus with an nceatraJ haae of * « 9. The banc karyotype (Gold* latU 1979r» bn . The six a per i e s of whptiui £am* • dyspkod series Orth ddlevem -pe r - 9. 8. 7. and 6 Th» has beet. Hence, the base of g * 10 is prahshb also denied in has a - 26 and T mmstu has a - 1 4. and the 296 Annals of the Missouri Botanical Garden S or 12(14, 11, 10, 9) 26,2 £ 300 bifucata (= S. bicolor), 2n = 12. The latter count is for plants from Gauteng Province, South Africa, where the genus was not recorded until 1983 (de Vos, 1983). Basic number in Syringodea is most likely x = 6. The ancestral base number for Ixioideae is most likely * = 10, evidently the base number for tribes Pillansieae and Watsonieae (Table 3). The ancestral base number in Ixieae may also be x = 10, but genera less specialized as regards leaf anatomy and seed characters (Goldblatt & Manning, 1995) have higher base numbers, and it is equally likely that x = 10, present in a few genera of the tribe, is secondary. At least Geissorhiza, Gladiolus, Hesper- antha, Radinosiphon, Romulea, and Tritoniopsis, all with ancestral base numbers between x = 16 and 13 (Table 3), are paleopolyploid. We suspect that Syringodea (x = 6) is a dysploid derivative of an ancestor shared with Romulea. Both have inflores- cences reduced to solitary flowers and similar asymmetric corms with woody tunics. The single species of Syringodea with x = 11 (Goldblatt, 1971; de Vos, 1976) is almost certainly a secondary hypotetraploid. Notably low base numbers charac- terize Babiana and Zygotritonia (Table 3); although both have x - 1 , they are probably not closely related (Goldblatt, 1989b). Extensive dysploid series in Ixioideae are re- stricted to just 4 genera, Lapeirousia, Gladiolus, Ro- mulea, and Crocus, out of a total of 28. Limited intrageneric dysploidy occurs in Syringodea, as noted above, and is reported here for the first time in two species of Tritonia and one of Hesperantha. Thus, the pattern outlined by Goldblatt (1971) for Ixioideae of genera each having a single base num- ber must be modified. Although most genera do have a single base number and exhibit little or no polyploidy, significant dysploidy has been discov- ered in Gladiolus since that review (Goldblatt et al., 1993). Most species of the genus have x = 15, but among the small-flowered tropical African species numbers include n = 14, 13, 12, and 11. The pat- tern of dysploidy in Lapeirousia has also been found to be more extensive than previously thought (Goldblatt, 1990b). As hypothesized hy Goldblatt and Takei (1993) there is a striking example .of dys- ploid reduction from * - 10 to 4 followed by poly- ploid increase and'further reduction from ft = 6 to * — 3 in Lapeirousia subg. Panic-ulntn . Gladiolus and Fnpeiromiajnnsl therefore, be added to the - **•- extensive dys- . JploTd sequences. " • ” - : : 1 : : ; S5idd;^$^£:ecHi^entIy=t tf.be de- scending' Syr- ingoma unifoHa:^ ev*i&nti£ derived in their respective genera, the first specialized in its acaulescent habit and the second in having a solitary and terete leaf. In Gladiolus all the dys- ploid species appear more specialized than those with the presumably ancestral x = 15 (Goldblatt et al., 1993). Moreover, they appear to belong to at least three, and probably four, separate lineages. Likewise, in Romulea members of primitive sec- tions have x = 12 or 13, and those of derived sec- tions have * = 11, 10, or 9 (de Vos, 1972). Cytological patterns in Crocus, last of the genera of Ixioideae with extensive dysploidy, are complex (Mathew, 1982) and remain to be satisfactorily ex- plained. Both dysploidy and polyploidy have been significant in the cytological evolution of the genus. Provisionally we suggest an ancestral base of x = 6 for the genus, a hypothesis based on outgroup comparison (the immediately related Syringodea has x = 6) and on the pattern of counts in section Crocus (summarized by Mathew, 1982), which in- cludes the more primitive members of the genus. From this base we assume dysploid reduction to x = 4, perhaps in several lines, and repeated poly- ploidization on bases of 6, 5, 4, and 3. REVIEW OF GENOME SIZE Genome sizes have been established for a num- ber of Iridaceae and, as in many other families, have been found to vary considerably, even within genera. In Iridoideae, species of Sisyrinchium (Kenton et al., 1987) have basic genome sizes (ad- justed for polyploidy), i.e., 1C values, of 0.48-0.73 pg in subgenus Sisyrinchium (= sect. Bermudiana ) and 0.25-2.10 pg in subgenus Echthronema. The related Olsynium (= Phaiophleps plus Sisyrinchium^ sections Filifolium and Nuno) typically has larger genomes among the temperate southern South American species with 2.66-3.26 pg, but the toploid North American member of the genus, 0. douglasii, has a basic genome size of 0.49 pg. Among Tigridieae, which typically have larger chromosomes than Sisyrinchium, 1C genuine sizes range from 2.03-2.39 pg in Cypella and 1.24-1.34 pg in three species of Hesperoxiphion (Kenton et al., 1990), but H. huilense has a genome size of 4.38 pg, despite also being diploid. These genome sizes were determined cytophotometrically against a standard, Hordeum vulgare, genome size of which is 11.12 pg (2C) or 5.56 pg (1C), a value recently confirmed by Arumuganathan and Earle (1991)- Comparable genome sizes of 1.47-2.48 have also been reported by Martinez and De Azkue (1987) for five species of Eleutherine, Ennealophus, and Volume 84, Number 2 . 1990b. Cvtological ranahilily i„ the African ge- Bot. Card. 77: 375-382. X ' 01 *"* CHROMOSOMAL Frank OBSERVATIONS ON THE ALZATEACEAE (MYRTALES ) 1 306 Garden NOTICE Plant Evolution and Domestication 26-27 September 1997 Indiana University will hold a weekend symposium in honor of Dr. Charles Heiser’s prominent contribu- tions to Botany during his 50 years at IU. The sym- posium is entitled “Plant Evolution and Domestica- tion ” and will take place Friday evening, September 26 and all day Saturday, September 27. Speakers in- clude Greg Anderson, John Doebley, Jeff Doyle, Don Levin, Barbara Pickersgill, Charles Rick, Loren Rie- seberg, Doug Soltis, and Herb Wagner. Registration fees are $75.00 for regular participants and $25.00 for students. For further information contact Angi Bai- ley or Jennifer Jones, Department of Biology, Indiana University, Bloomington, IN 47405-6801. Phone: 812-855-6283, FAX: 812-855-6705, email: abail- ey@bio.indiana.edu. Editor’s Note CONTENTS The Genus Lycianthes (Solanaceae) in Venezuela Carmen Benitez de Rojas & William G. D'Arcy 167 Dialypetalanthus fuscescens Kuhlm. (Dialypetalanthaceae): The Problematic Taxonom- ic Position of an Amazonian Endemic - Frederic Piesschaert, Elmar Robbrecht & Erik Smets 201 Conspectus of the Genus Palicourea (Rubiaceae: Psychotrieae) with the Description of Some New Species from Ecuador and Colombia Charlotte M. Taylor 224 Palynology, Phylogenetic Reconstruction, and Classification of the Afro-Madagascan Genus Aristea (Iridaceae) Peter Goldblatt & Annick Le Thomas 263 Chromosome Cytology of Iridaceae — Patterns of Variation, Determination of Ancestral Base Numbers, and Modes of Karyotype Change — Peter Goldblatt & Masahiro Takei 285 Chromosomal Observations on the Alzateaceae (Myrtales) Frank Almeda 305 Notice 309 Editor’s Note and Editors of the Annals of the Missouri Botanical Garden 310 Cover illustration. Eccremocarpus viridis Ruiz & Pav6n, by Phyllis Bick. Annals of the Missouri Botanical Garden 1997 m Volume 84 Number 3 Volume 84, Number 3 Annals of the Missouri Botanical Garden The Annals, published quarterly, contains papers, primarily in systematic botany, contributed from the Missouri Botanical Garden, St. Louis. Papers originating out- side the Garden will also be accepted. All manuscripts are reviewed by qualified, independent reviewers. Authors should write the Managing Editor for information concerning arrangements for publishing in the Annals. Instructions to Authors are printed in the back of the last issue of each volume. Editorial Committee Michael H. Grayum Ihsan A. Al-Shehbaz Editor , Missouri Botanical Garden Missouri Botanical Garden Gerrit Davidse Amy Scheuler McPherson Missouri Botanical Garden Managing Editor , Roy E. Gereau Missouri Botanical Garden Missouri Botanical Garden Diana Gunter Peter Goldblatt Editorial Assistant, Missouri Botanical Garden Missouri Botanical Garden Gordon McPherson Missouri Botanical Garden Vicki Couture Secretary P. Mick Richardson Missouri Botanical Garden Teresa Johnson Henk van der Werff Publications Order Processor Missouri Botanical Garden For subscription information contact Depart- ment Eleven, P.0. Box 299, St. Louis, MO 63166-0299. Subscription price is $110 per volume U.S., $115 Canada and Mexico, $135 all other countries. Four issues per volume. The journal Novon is included in the subscription price of the Annals. amcpher@admin.mobot.org (editorial queries) deptll@mobot.org (orders) http://ww.mobot.org The Annals of the Missouri Botanical Garden (ISSN 0026-6493) is published quar- terly by the Missouri Botanical Garden, 2345 Tower Grove Avenue, St. Louis, MO 63110. Pe- riodicals postage paid at St. Louis, MO and ad- ditional mailing offices. POSTMASTER: Send ad- dress changes to Annals of the Missouri Botanical Garden, Department Eleven, P-0. Box 299, St. Louis, MO 63166-0299. © Missouri Botanical Garden 1997 The mission of the Missouri Botanical Garden is to discover and share knowledge about plants and their environment, in order to preserve and enrich life. @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Volume 84 Annals Number 3 of the 1997 Missouri Botanical Garden niMmmmmmm A. and C. DeCan to 120. Of these, 93 species were members of P. The final revisionary effortby EngkToii Philo- from the 1878 work. One section was raised to sub- “Gruppe” Achyropodium Schott: P. ■ST7 sect. Macrogynium Engl.: P. twffnwnnii < sensu Engl. (= P jacquinii Schott) 310 Am* of 182. P. 183. P tripartitum (Jacq.) Scholl ■ P sect. Schizophyllum (Scholl) Engl. luding 5 •preir* puktuhni by Grayum. 1 7 nrw •ubaprein*. and 1 nrw (am w nadr. Of thi* lolal only writ* reported for Ceft- r wrrr in P aubg. Pttr* hum. Thus* 22 aperies* of P. aubg. PhUoden- 7. P seel. Polytomium (Scholl) Engl.: 195. P radiatum Scholl 198. P. auguatinum K. Koch * 199. P polytomum Scholl - P. 200. P. uanzruirai K. Koch & 8. P. seel. MacroUmchium (Schott) Engl. as* follows*: Philodendron mi Schott). P ««' .HO! Williams*. P. hasu Ma- ! hemesu Standi.. P. daxidsomi Croat. P dreuteri G. S. Bunting. P. ertamonii 1. M. Johns*!. (- P. jacquuui), P gUuxduUferum Matuda. P. har- Umu I. M. Johns*!. = P hederaceum. P jamapanum G. S. Bunting i = P. sagiittfolium Licbm ). P- j°~ 1997 Philodendron Subgenus Philodendron 319 davisumum G. S. Bunting, P. lancigerum Standi. & were often poorly pn L 0. Williams (- P. sagittifolium ), P. latisagittium virtually all lacked ( Malm la ( = P mexuanum ), P. lundellii Bartlett ex observations or well-p Lundell ( = P. jacquinii), P. microstictum Standi. & aspects of the plants L 0. Williams. P. miduhoi Matuda = P. hedera - ceum, P. mirificum Standi. & LO. Williams ( = P. pttrutum K. Koch & Augustin), P. monticola Ma- tuda = P. adtena, P. platypetuAatum Madison, P. pleistoneurum Standi. &L0. Williams (= P. gran- dipes K. Krause), P. pseudoradiatum Matuda (= P. radiatum var. pseudoradiatum (Matuda) Croat), and P trisectum Standi. ( = P. are essential to the proper understanding of Philo- dendron and other Araceae. I believe that it is pos- sible that this confusion, coupled with the dearth of well-prepared specimens and the paucity of types of Ara nk, and only 1 1 taxa in P. subg. Philodendron remained. These were: P. brenesii , P. radiatum var. pseudoradiatum, P. auri- culatum, P. microstictum, P. ba.su . P. glanduliferum, P. dressleri, P. jodavisianum, P platypetiolatum, P. Most of the Central American Holistic work out- side of Mexico was carried out by Paul C. Standley, often working with his associate Louis 0. Williams. Standley worked initially at the Smithsonian and later at the Field Museum in Chicago, then at the herbarium of the Escuela Agricola Panamericana, where he died at Zamorano in Honduras. Standley described P. brenesii and P. trisectum (= P. aniso- tomum Schott) alone, and with L 0. Williams he also provided the following epithets: P. armigerum (= Syngonium armigerum (Standi. & L. 0. Wil- liams) Croat), P. auriculatum, P. brevinodum (= Monstera tuberculata Lundell (Standi. & L 0. Wi tiferum), F um), P. microstictum, P mirificum (= P pterotum), and P. pleistoneurum (= P. sagittifolium). It is unusual that despite the fact that there were many undescribed species of Philodendron subg. Philodendron in Costa Rica and Panama, these workers did not succeed in describing many of them since oS the nine described, three proved to belong to other genera, and three others proved to be synonyms of existing Philodendron names. This is particularly surprising since both Standley and L. 0. Williams were astute observers who were very familiar with the Central American flora in general. Their mistakes point out the complexity of the tax- onomy of Araceae and the bewildering array of ma- terial available to them at that time. Even when I began my own work with the Araceae in the late is (aside from types) t names. Specimens treatment of the Flora of Panama (Standley, 1944), small percentage of the total aroid flora that is cov- ered compared to what is now known to exist. In his treatment of some genera Standley (1944) seemed too willing to accept epithets of species de- scribed in Colombia, regardless of how well they “fit" Panamanian species. As a result many species were wrong. His treatment of Philodendron subg. Philodendron was much better from the standpoint of correct names, but he treated only 8 of the 104 taxa (7%) of Philodendron subg. Philodendron now known for Panama. Only one species, P. hqffmannii (now P. jacquinii ). had the wrong name. Philoden- dron jacquinii was also misapplied, being intended by Standley for P. hederaceum. The other species he included in the Flora of Panama were: P. bre- nesii, P. brevispathum, P. grandipes, P. paruimen.se. P. radiatum, P. tripartitum, and P. uendlandii. He did not do so well with members of P. subg. Pter- Schott was a mixture of two species and P. guttiferum was a mix- ture of three species. Perhaps the most curious thing about Standley’s Flora of Panama treatment is that by 1944, after Robert Woodson and his col- laborators had already made several expeditions to Panama, so few of the new species included in the present revision had been collected. Standley had collected some of the new species but failed to rec- ognize them as new. These included: P. crassispa- thum ( Standley & Valerio 51910), P. findens (Stan- dley & Torres 52355), P. purulhense (Standley 89902), P. strictum (Standley 51371), P. verapazense (Standley 91978), and P. wilburii var. wilburii (Standley 38300). See discussion of those species for additional details. See also section on “Collect- ing History of P. subg. Philodendron .” The Flora of Guatemala (Standley & Steyermark, 1958b) was much more accurate and complete in : of the total taxa of P. subg. Philo- while teaching fur FWhU ! former Cum) Zone. mule 16 caUrrtwms. S o i ilwm new to Kicocr P. annuUtum Croat A Comm, P doluhophyUum Croat. P. Uuom Croat. /» Oa*r*ur Croat, and P (tmwj Croat. Even the late Ah* to Gentry, who r scribed at tbe U Collecting in Honduras. Costa Rwa. and f between 1973 and 1993. Ron Lreaner collected 43 specie* of P. subg. Philodendron, 4 of them. P. al- licoia Croat & C i making (since 1984) 426 i m» (in both Central and South - tal. 153 v ies of P. s gustdobum Croat A Grayum. P. aromaUcum Croat & Grayum, P. bakeri, P. brunneicaule Croat & Gra- Croat & Grayum. P. dodumu Croat A Grayum. P grarumti Croat, P. rcalannene Croat & Grayum. P. straminicaule, P. thalassicum . and P wdburii. My own collecting activities with Philodendron began in 1967 with work - phonodorrae. and Peltandrear. Engler (1912) tidinae ( Burephalandra , Gamogine, Microcasia, and Ptptmpaiha) into subtribe Philodendnneae. 1997 Croat 323 Bogner and Nicolson (1991) left Engler's subfamily act, but Grayum (1984) made including an incorporation of cladistic analysis. While maintaining essentially e same alliances suggested by Grayum (1990), ayo et al. (1995) placed all araceous genera with lisexual flowers in subfamily Aroideae. Philoden- ' on is placed in tribe Philodendreae in the Phdo- i priority). The , Aglaonema , Peltandra, and Philodendron Alliances. In Gra- vums system. Philodendron, in its own tribe Phil- t shared the alliance with the . Both Dieffenbachia and 1 dump genera mostly occurring in southern South Am ccarum. Grayum (1990) later placed Philoden- an close to Homalomeninae and the African gen- > Culcasia and Cercestis (which had been placed the Pothoideae and Lasioideae, respectively, by igler). Thev all share similar stem and stamen Araceae (Grayum, 1990). Grayum believed that the Philodendroideae are a sister group to the Pothoi- deae (including Engler’s Monstemideae), which have in common the exclusive characteristics of ge- niculate petioles, cork formation in a with Furtadoa and Homalomena and tribe Anubi- adeae (Anubias only). Their cladistic analysis also shows Bognera to be a close ally of Dieffenbachia. Another cladistic analysis resulting from a study of chloroplast DNA restriction site variation in the Ariflorae by French et al. (1995) places Philoden- dron as a sister group to Homalomena, suggesting, according to Grayum (1996), that Homalomeninae is paraphyletic. According to French’s findings, An- of Homalomena, Furtadoa, chardia is a sister taxon to all four of these gem generic relatives of Philodendron, the classifies! by Mayo et al. (1995) has taken into account the evidence to date including the extensive i lecular studies by French. 1 distant placement of While there i sification of tril l of the major systems of classification at the suprageneric level was made by Croat (1990(19921 ). It included the systems of Hotta (1970), Grayum (1990), and Bog- ner and Nicolson (1991). Phenetic analysis on the Philodendroideae by Mayo (1986) shows Philodendron to be distinct but closest relatives were the African genera Culcasia and Cercestis. In his survey of sclerotic hypodermis in the roots of Araceae, French (1987a) provided evidence to link Philodendron to the West African genera Anubias, Culcasia, and Cercestis and the spathe, while those of Philodendron are closely c Dieffenbachia is immediate as long a visible. Dried sterile material without field notes de- noting terrestrial habit (consistently true of Dieffen- bachia, but rarely so of Philodendron ) or scent (usu- More recently, in an attempt to bridge differences in the systems of Bogner and Nicolson, Grayum, and Hay and Mabberley (1991), Mayo et al. (1995) conducted another sweeping survey and produced much more problematic. Dieffenbachia leaf blades are rarely ovate and never truly cordate, whereas this blade shape is common in Philodendron. Philoden- dron may, however, have blade shapes that closely match those of some Dieffenbachia. If the petiole is well preserved the presence of the petiole sheath is the best means to separate Dieffenbachia and P. subg. Philodendron ; the latter generally has a very short sheath, while it is rare that the sheath of Dief- middle of the petiole. Live material of neotropical Homalomena is not easily confused with Philodendron because the for- Garden fer^Nomencl. Bot 2: 674 1874.^[rankles s ] let. Bot. 19. 1832. TYPE: P. imbe Schott Pill Garden Garden 349 In hu review at leal morphology and function Ray (1987a) divided lra«m into ihw main type*: foliage Irate*, reduced Irate* (10-7041 liar »ur at aulralr al the base (Fig. 212). In either event they typically become more terete toward the middle and then obtusely somewhat flattened toward the apex (Fig. 126), P jocquimi, P. mseospalhum, and some- times in P. utgimfolium. Twenty-one Central American species of P. subg. ** termed a sy m p od ial leaf (Rav. 1967a). In P. subg. Philodendron all .dull leaves are of this type. Ju- venile leave* of P subg. Philodendron, on the other Hand, are all mnnnpodial leaves (Crayum. 1990). iously D-shaped in cross section. Examples of spe- cies with D-shaped petioles are: P copens*. P. fin- (Fig. 264). P Itgulaium var. heradioanum (Figs. 275. 276). P thalasucum. and P irrapasense. Pet- ioles of Philodendron adiena, P. ptenotum, P im- mixtum, and P. ligulatum var. ovatum are also broadly and sharply sulrate. e g.. P. bakeri, P. dav- idsonii, P helemae ’ P lentil (Fig. 264), P. Ugula- tum, P jodaiisianum, and P scalarinerve. Rarely and only 4 species, namely P bakeri, P. bmster- P. chirripoei petioles less than P radiatum var. t sually the petioles are erect-spreading from the with the blades either extending initially in the same plane or. more frequently, somewhat pen- denl from the end of the petiole. as in Anlhurium, but the variation is important tax- onomically. AU too often herbarium collections make no mention of this frequently critical diag- nostic feature. Typically petioles are obtusely some- what flattened or sometimes broadly and obtusely and with the petiole often much broader than thick, as in P. uendlandu (Fig. 453). In the Utter, the lateral margins may be very acute and directed out- ward. D-shaped petioles generally have the lateral margins weakly to prominently raised with the mar- ginal rib either acute or obtuse. Taxa with the pet- iole margins acute comprise: P. dot idsonii, P. lig- ulatum var. heraclioanum, and P. uendlandu i They are sometimes also acute on P. chiriquense, P. lin- dens, P. fortunense. and P. pterotum, the margins are prominently winged. The wing is erect-spread- ing and may be markedly undulate in the area of the geniculum as in P fortunense (Fig. 183) and P. fendens (Fig. 171). Sometimes the petioles of P. lig- winged (Figs. 275, 276). Even petioles not D-sha- ped often have a slightly thickened, slightly raised lateral margin on the adaxial surface. Examples in- clude P- annul alum. P. cretosum, P microsticlum . cr? 1 l5a -£SEEi25eaE£gg: lb ' ° r ’°°'^ l '-“'^- “ i * h ‘ h ' i "' < ™»‘ l ‘ r 'W l “^ Lg 0^”“ " k iH 32a ' SSSS3SSSS3SS ‘ ^H^lisia^f -a-sssssgss :zzr -^mmmir ^TegTon, r C or :M, K, M, MEXU, MO, W); 4.6 km beyond peak on road ^rNETfAho^deTa^ V. 700-750 m, 9°15'N, )); 5-10 km NE of Altos ssssss sendero de Interpretacidn, 1 km al este del Campamento B.g&i5£S= ssSSSIsSb SSStSSHSl sassus il ' km W of Bolfvar, 466 m, Croat 49298 (MO); 27 mi. W of 462 Annals of the Missouri Botanical Garden Pedregal de San Isidro, ca. 3 mi. S of Lake Coatepeque, ca. 850 m, Croat 42241 (MO). GUATEMALA. Los Amates, Kellemum s.n. (US); Watson 427 (GH). Alta Ver- apaz: road to El Estor (Lago Izabal), 7 mi. E of Highway CA-14 to Cob4n, 1000 m, Croat 41480 (MO). Escuintla: Santa Lucia, 1045 ft., Kellerman 4547 (US); 5285 (US). 6 mi. from Izabal, 65-600 m, Steyermark 38486 (F); ca. 7 mi. S of Puerto Barrios, 50 m, Croat 41811 (MO). Que- lombi a ai l^ mi. W of turnoff, 580 m, U 14 0 41'N, 91°48'W, Croat 63394 (B, BM, MO, US). San Marcos: near San Rafael, 600 m, Croat 40769 (MO). Suchitepequez: 1 mi. E of Mazatenango, <500 m, Croat 43757 (MO). HON- DURAS. Atlantida: Lancetilla Valley, Tonacatepeque, Pfeifer 2130 (BH, US); San Jos6 de Texfguat-El Chorizo, 100 m. Nelson 10565 (TEFH); 4 km S of Tela, 0-100 m, bias 190 (TEFH, UNAH); ca. 10 mi. SE of Tela, along Rfo Lancetilla, 10-150 m, Croat 42639 (MO); Quebrada Grande, ca. 10 km SW of La Ceiba, 80-140 m, 15°42'N, 86°51'W, Uesner 26335 (MO). Colon: Rfo Selen, 7 km E of Trujillo, Howler Site, Saunders 192 (MO). Comayagua: junction Rfo Yure-Rfo Humuya, 200 m, Nelson et al. 6182 (MO). Copan: 13 mi. E of Copdn, road to La Entrada, 750 m, Croat 42529 (MO); Sta. Rita village, 650 m, Mo- 42739 (MO). Gracias a Dios: Ahuas Bila, 200 km SW of Puerto Lempira, 100 m, Nelson & Cruz 9316 (UNAH); 9292 (TEFH, UNAH). Olancho: Mpio. San Esteban, near Santa Marfa del Carbdn, 21 mi. NE of San Esteban, along road to Bonito Oriental, 440 m, 15°25'25"N, 85°34'45"W, Davidse et al. 35571 (MO; Rfo Guyape, San Pedro de Catacamas-Poncaya, Blackmore & Heath 1984 (BM); Rfo Wampu, 8 km S of Pisijire, 500-700 m, 15°15'N, 85°25'W, Nelson & Clewell 594 (FSU, MO). MEXICO. Chiapas: El Triunfo, ca. 10 mi. NE of Escuintla, 300 m, Croat 43859 (MO); 2 mi. SW of Guatemalan border. High- way 200 to Tapachula, 300 m, Croat 43771 (MO); Bonam- pak, near ruins, 500 m, Matuda 38715 (MO); Mpio. Ocos- ingo, 5 km SW of Santo Domingo, 600 m, Davidse et al. 20425 (MO); Esperanza, Escuintla, 150 m, Matuda 17789 (MO); P alenque-Bonampak, 60 mi. SE of Pfdenque, ca. 400 m, Croat 40167 (MO); Palenque-Ocosingo, Highway 199, 27 mi. SW of Palenque, 210 m, Croat 40302 (MO); Cerro Vernal, NW side, 25-30 km SE of Tonal*, 400-000 m, Breedlove 25617 (DS). Guerrero: Tierra Colorada- Xalpatlahuac, Tierra Colorada, Rfo ComitlSn, 900-1000 m, Croat 45755 (MO); Pinotepa Nacional-Tlaxiaco, High- way 125, ca. 8.4 mi. S of Putla de Guerrero, ca. 1000 m, Croat 45807 (MO); Tierra Colorada-Acapulco, kms 366- 367, ca. 380 m, Moore & Bunting 8840 (BH). Jalisco: Puerto Vallarta, 100 m, Mexia 1314 (UC); 24.1 mi. from Authin, ca. 300 m, Moore & Bunting 8737 (BH); Quimix- to, Mexia 1201 (BM, CAS, DS, G, GH, MO, NY, UC, US). Nayarit: Miramar, ca. 10 km W of Jalcocotdn, Dressier & Wirth 2703 (US); Singaita, Lewis s.n. (BH); Philbrick 785 (BH); San Bias, Lewis s.n. (BH). Oaxaca: Pinotepa-Tlax- iaco. Highway 125, 4.4 km S of Putla de Guerrero, 850 m, Croat 45835 (MO); 12 km from Hwy. 200, road to Chayuco, 220 m. Miller & Tenorio L 524 (MO); Tuxtepec, Rinc6n del Tigre, Mpio. Acatl4n, 2 km from Acatl4n on road to Capilla, ca. 100 m, ca. 18°31'N, 96°36’W, Gereau et al. 2190 (CAS, MO, RSA); Tuxtepec-Oaxaca, 10 mi. S of Valle Nacional, 700 m, Croat 39802 (MO); 0.5 mi. S of Valle , Croat 39694 (MO); Esmeralda- lriO'N, 94°45'W, Croat , Hannon 63233 ( Puebla: Teziutliin-Nautla, Rancho Las Margaritas, Huey- tamalco, near border with Veracruz, 19°57'N, 97°16'W, Conradt 218 (MEXU). San Luis Potosi: 6 mi. NW of w5T& Lundell 7115 (CM); 6.5 mi. S of lE^IxtU- mel, 325 m, 21°18'N, 98°47'W, Thompson et al. 1320 (CM); 1321 (CM). Tabasco: Tacotalpa, Cowan 1999 (M0). Veracruz: 927 m, Birdsey 226 (UC); Huatusco-Puente 1200 m, Croat 44014 (MO); near Fortfn, Cervecerfa Moc- tezuma, 1000-1150 m, Croat 39408 (MO); Mpio. Nautla, ) Camin r Cerro Chic ■i 3600 (DS); Mpio. San Andres Tuxtla, Estacion de Biologfa Trop- ical Los Tuxtlas, LOTE 71, 400 m, Ibarra & Colin 3126 (MO), 400 m, 18°34-36'N, 95°04-09'W, Manriquez & Colin 3126 (MO). NICARAGUA. Boaco: along Hwy. 33 from Rfo Quilan, ca. 300-310 m, 12°35'N, 85°32'W, Ste- vens 9335 (MO); Cerro Mombachito, 500-900 m, ca. 12°24— 25'N, 85°32-33'W, Stevens & Grijalva 14749 (MO); Quebrada Rfo Grande, NE del Cerro Mombachito, 600-700 m, 12°25'N, 85°32'W, Moreno 354 (MO). Chin- andega: Rfo Chiquito, El Viejo, 0-100 m, Atwood 2635 (MO). Chontales: Rfo Bizcocho-Rfo El Jordan, 350-550 m, ca. 12°12-16'N, 85°15-17'W, Stevens & Montiel (MO); ca. 2.8 km N of Cuapa, 400-500 m, ca. 12°17'N, 85°23'W, Stevens 3696 (MO); 3 km N of Santo Tomas, 280-300 m, 12°05'N, 85°07'W, Moreno 16066 (MO); Juigalpa, La Libertad, Rfo El Bizcocho, ca. 17.4 km NE of Rfo Mayales, 350-400 m, ca. 12°12'N, 85°17'W, Stevens 4090 (BM, MO); 4 k nNW of Villa S km from Highway 12, near bridge of Rfo La Aduana, 80- 100 m, ca. 12°02'N, 86°31'W, Stevens 5394 (BM, LL, 2, km 28, ca. 700 m, 11°57'N, 86°20'W, Stevens 3990 (MO); 4 km from Highway 8 to Highway 2 intersection, 800-860 m, 11°58'N, 86°18-19'W, Stevens 4539 (M0); Escuela Nacional de Agricultura and Ganaderfa, Route 1, 12 km E of Managua, Atwood 2930 (MO). Matagalpa: Matagalpa— J inotega, km 140, ca. 900-1000 m, Guzmdn et aL 229 (MO); Cerro MusiSn, ca. 300 m, ca. 12°55'N, 85°16'W, Stevens 12032 (MO); Quebrada Malacal, Haci- enda La Bonanza, ca. 20 km from Matagalpa, 560 m, 13°01'N, 85°47'W, Castro 2391 (MO). Nueva Segovia: 7 km SE de Santa Clara, 600-700 m, 13°40'N, 86°14'W, Araquistain & Moreno 2191 (HNMN, LE, MO); El Terrero, 4 kms NE of El Jfcaro, 500-600 m, 13°45'N, 86°07 W, Stevens & Moreno 2215 (MO). Rio San Juan: Rfo Indio, San Juan del Norte, 2 m, Araquistain 3312 (K, M, MB . MEXU, MO). Rivas: Volcan Concepci6n, La Esperanza. 200-400 m, 11°31'N, 85°37'W, Robleto 1618 (ENCB, MO); Isla Ometepe, 140-350 m, 11°33-34'N, 85°36 W, Robleto 997 (MO); SE of “La Flor,” 300-800 m, H 32- 34'N, 85°37-38'W, Robleto 1915 (MO); 400460 m, 11°33'N, 85°37'W, Sandin " " ' main road, Stevens 12672 (MO); Ibo Tingni, __ , ;a . 14°9-11'N, 83°29- 31 'W, Stevens 10663 (CM, MO); Puerto Cabezas, 0-20 m, 14°01'N, 83°22-23'W, Stevens 10684 (MO); Rfo B Rfo Copalar, ca. 29 km E of Rfo Blar E of campo German Pomares, ca. 60-90 m, 11°36'N, 83°52'W, Moreno 15137B (MO); ca. 10 m, 11°36'N, 83°51'W, 15187 (MO); Cano Monte Cristo, 1 km before the camp German Pomares, ca. 10 m, 12°35'N, 83°51'W, , ca. 1.5 km N of mmimE » & Moreno 19278 (MO); Rto Matts, Want-Siuna, ca. 0-100 m, ca. 13°43'N, 84°49'W, Errjs5Tc.SK terson & Annoble 7287 (MO); Chiriqut Lagoon, von Wedel felfiisE” Vic. George W. Greco Part. Welch 19853 (MO. NY, RSA). Gualaca, 100 m, 8°29'N, 82°17'W, Churchill & de Never s reddish slightly paler than lyrSTdTCHe veins lacking; primary lateral veins <6-7)8-10(11- 13) per side, departing midrib at a .30-40" angle. 0 . 1 - 471 ^gspglg !«'Is§=e§ 15°46'3(TN, 85°41'W, Evans 1086 (MO). Copan: 10 mi! Riv^: Tola-^^linas^E^oyor ca°^3° k^Lyond W of Copan, road to La Entrada, 700 m, Croat 42517 entrance of Hda. Miramar, ca. 30-10 m, 11°23'N, ES SSSSSS S-SSSSSIS sSSStHS'S SSstSsssSHS ‘ ' “ of Tela, near border of Yoro and Atlintida, 50 to Cerro Pelado Radar Station, 0.5 km NW of Gamboa, N, 87°43'W, Croat & Hannon 64666 (MO). 75-150 m. Nee 7760 (MO, RSA); Curundu, Panque Me- I SSrSS SiSiS SSSS SaSSSS SSS5SSSS S55?<2SE2i?sS5 & Bunting 8908 (HB); 14 mi. N of Puerto Escondido on Williams 381 (NY). Darien: Cent, Pirre region, El Real, Rte. 131 to Oaxaca, 300 m, 16°13-47'N, 97°5-8'W, Croat & Porter 15460 (MO); Santa Fe region, Univ. of jsk: ^asaacraas :a Veracruz, E. 3277 (F, US); El Llano-Chepo, Gentry & Tyson 1727 ™nL‘, cZ^OO m! philodendron jeferun: Croat, ap. no». TYPE: Pan IT 28 "• 85 :f'«'. Moreno 10643 (MO); Rfo Las Cattas, Pa[iam4; Ce.ro Jefe, along road short c 477 Ka’i-KSEas 10.1 mi. NW of Los Planes de Homito, 1250 m, 8°45'N, 82°17'W, Croat 50041 (CM, MO); 10 mi. NW of Los Planes de Homito, 1260 m, Croat 50102 (MO, NY, PMA); Chiriquf Grande-Fortuna, 4.5-5 km N of dam over For- S^^ra ( nde M 8 mL^N Ha^de Homito, 3 L4 lifsliiifi iss£s-csr«*ss Rfo i cf S ci^5.7 I m!; 80°26'W, Croat 67538 (MO, PMA); El Copd region, Alto Jj* 4MO?5 ^ I AjSc^ 2^7S5^TO m,°8 0 39% ptswz £"S‘S«iS5SS fSBSE irre National Park, W side of Cerro Pirre, base [ Dlri^EsTcldn^RanclTo Fri^N bJSofS!^ HrfSSSiSs|i Trinidad Basin, 20-50 m, Pittier 4015 (US). Darien: Cer- ro Pirre National Park, W side of Cerro Pine, base camp, 50 in, 8°N, 77°48'W, Croat 68962 (M, MO, NY, US); near S=S.„ 9°19'N, 78°55'W, de Nevers et al 7342 (1 511 513 517 enus Philodendron KT£Kr:«5 «« ■ 5k: s r„ sruK zs asrsis^rsi r^'sr&s i 2 « 7 ,i M0); M ° nkey P ° int ’ 15 km NW ’ 1 ~ 5 m ’ U ° 36 ' N ’ Cerro Gaital, Churchill 3907 (MO). Colon: Sabanitas- J"* ™' S “™ ite >' «*« MSV 0 Mb C ^A!^Sm?»mrR!dteR^0.6to 19614 (F); Rfo Pis Pis, 0.5-1. 5 km from Plantel El Salto, from highway, ca. 380 m, Croat 34345 (MO); Mile 6.5, oo^T’io m, ‘ N ° f Con * Rf0 Cocalito ’ ca - 5 ™- from P- Cocalito, Whitefoord & □«M MO) 8 !? 7 W N C K°- & £Dry Dry Dry-> Wet BiModal Unknown albisuccus ' alticola 1 angustOobum 1 annulatum ! 1 aromaticum 1 baked ”1 basii breedlovei brenesi 1 1 bremt^ — i 1 chirripoense dewefi 1 cotonerne cretosum subsp. bocatofanum+A50 1 doiichophyllum 1 d^nKalense edenudatum ~ 1 1 SSL. gtanduifenjm subsp glanAifenjn granulate grayumi hammelii hebetatum hedetaceum var. hederaceum vat. kkkbridei hnlnninj jacquini — i 578 Missouri Botanical Garden Flowering I Pattern PHILODENDRON Alt Year Wet Wet->Dry Dry Dty-> Wet BrModal Unknown jodavisianum 1 cnappiae 1 lazorii 1 ligulatum var. Hgulaturn var. ovatum lanense madronense 1 1 pirrense pterotum U 1 radiatufn var radiatum var angustilaminatum 1 rothschuhianum sagittifolium sca^inerve smiths 1 squamipetiolatum strictum gftpy ^ tripartitum ^^ eTOe T verapazense verrucosum warszewiczfi wendlandu wilburii var wilburii — t zhuanum TOTAL 51 1 Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 579 Flowering 1 Pattern PHILODENDRON All Year | Wet | Wet->Dry| Dry | Dry-> Wet | BiModal IUnknown ... 28/16% 1 94^1 RBt»| 43151*1 3.88*1 4.85^ 582 w 1997 Croat 585 586 Annals of the mn 588 Annals of the Missouri Botanical Garden Figures 33-36. — 33 (top L). Philodendron hebetatum, showing spathe beginning to break up to expose mature berries (Croat & Monsalve 61396). —34 (top R). P. findens, showing scar left by fallen spathe and coarsely striate peduncle apex ( Croat 38218). —35 (bottom L). P. pseudauriculatum, showing mature infructescences with spathe and staminate portion of spadix fallen ( Croat & Zhu 76251). —36 (bottom R). P. brenesii, showing infructescence with exfoliating spathe and partially consumed berries (Croat 35519). 1997 Croat Philodendron Subgenus Philodendron 589 590 1997 Croat 591 Figures 45-48. P. alticola ( Croat 74906). — 45 (top L). Stem with post-anthesis inflorescence. — 46 (top R). Juvenile Figures 49-52. Philodendron angustilobum. —49 (top L). (Croat 61162) Habit, in cultivation. —50 (top R). Plants displaced from trees, with inflorescences (Croat & Hannon 64522). 51, 52. ( Croat 61162). —51 (bottom L). Inflores- cence at anthesis. —52 (bottom R). Habit in cultivation. Croat Philodendron Subgenus Philodendron 594 Missouri Botanical Garden 1997 Croat Philodendron Subgenus Philodendron Figures 61-64. Philodendron antonioanum (Croat & Zhu 76909). -61 (top L). Habit. -62 (top R). Leaf. -63 (bottom L). Inflorescences partially obscured by cataphylls. —64 (bottom R). Leaves, both lower and upper surfaces. 596 Missouri Botanical Garden 1997 Croat Philodendron Subgenus Philodendron 597 Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 599 1997 Croat Philodendron 601 Figures 85-88. — 85 (top L). Philodendron breedlovei, open inflorescence ( Breedlove 35181) (photo: D. Breedlove). 86 ~ 8 8- P brenesii, habit. —86 (top R). ( Croat 67578). —87 (bottom L). (Croat 68084). —88 (bottom R). Apex of stem w ith bases of petioles, cataphyll, and unopened inflorescence ( Croat 78806). Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron Annals of the Missouri Botanical Garden Figures 97-100. — 97 (top L). Philodendron brunneicaule, leaf blade adaxial surface ( Croat & Zhu 76581)- 98- 100. P. chiriquense ( Croat 69068). —98 (top R). Leaf blade adaxial surface. —99 (bottom L). Stem apex showing inflorescences emerging from cataphyll fibers. — 100 (bottom R). Inflorescence with tube portion cut open. 1997 Croat 606 Annals of the Missouri Botanical Garden Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 607 Annals of the Missouri Botanical Garden Figures 113-116. 113-115. Philodendron copense. —113 (top L). Leaf 115. (Croat 68765). —114 (top R). Uaf blade abaxial surface. -115 (botto. fibers. —116 (bottom R). P. correae, habit (Croat & Zhu 76395). Volume 84, Number 3 Croat 609 Croat Philodendron Subgenus Philodendron 611 612 Volume 84, Number 3 1997 Croat 613 Figures 133-136. 133-135. Philodendron davidsonii subsp. bocatoranui ultivalion. —134 (top R). Open inflorescence. —135 (bottom L). Leaf bla 'avidsonii subsp. davidsonii, open inflorescence (. Davidson 7097). 614 Missouri Botanical Garden Figures 137-140. Philodendron dodsonii. 137, 138. ( Croat 72982). —137 (top L). Habit. —138 (top R). Apex of stem wit h petioles, cataphylls, and unopened inflorescence. —139 (bottom L). Unopened inflorescence ( Croat & Hannon 79114). -140 (bottom R). Inflorescence on abaxial blade surface showing normal open nature. Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 615 Volume 84, Number 3 618 Missouri Botanical Garden Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 619 Annals of the Missouri Botanical Garden Figures 161-164. Philodendron edenudatum. 161, 162. ( Croat & Zhu 77087). —161 (top L). Leaf blade adaxml surface. —162 (top R). Stem apex. 163, 164. ( Croat 33988). —163 (bottom L). Blade abaxial surface, showing purple spots on petiole and lower midrib (photo: P. Malesevich). — 164 (bottom R). Inflorescence in cultivation in final stages 1997 Croat 621 Figures 165-168. Philodendron ferrugineum. —165 (top L). Habit (Croat 33732). —166 (top R). Habit (Croat 75155). —167 (bottom L). Cluster of inflorescences (Croat & Zhu 77029), juvenile foliage. —168 (bottom R). Juvenile foliage (Croat 75116). Figures 169-172. 169-171. Philodendron findens. — (Croat 67153). — 170 (top R). Showing divided blades wh L). Showing winged petiole (Croat & Zhu 76502). —172 i >p L). Showing young blade before shredding occu urally become pinnate (Croat 67919). —171 (bottom R). P. folsomii (McPherson 13619). 1997 Croat Philodendron Subgenus Philodendron 624 Missouri Botanical Garden Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 627 Figures 189-192. 185M91. Philodendron fragrantissimum. —189 (lop L). Habit ( Croat 11526). —190 (top R). Apex of stem with persistent cataphyll fibers ( Croat 53912). — 191 (bottom L). Juvenile leaves (Croat 9003). — 192 Missouri Botanical Garden Missouri Botanical Garden Figures 201-204. 201, 202 showing petiolar glands. — 202 203 (bottom L). Habit (displaced) ( Croat unopened inflorescences (Croat 61323) (Colombia. Valle: Bajo Call Philodendron glandulifemm var. glanduliferum. —201 (top L). ( Croat 43909) Leaves tnn m Showing weathered cataphyll fibers ( Croat 39753). 203, 204. P. grandipes. — ° Watt 70200). — 204 (bottom R). Apex of stem with intact cataphylls and 1997 Croat Philodendron Subgenus Philodendron 631 Figures 205-208. — 205 (top L). Philodendron gigas, blade adaxial surface with unopened inflorescence (Croat & Zhu 76988). — 206 (top R). P. grandipes, open inflorescence (Croat 33648) (photo: P. Malesevich). — 207, 208 (bottom L & R ) P. granulare (Croat & Porter 15543). Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 634 Annals of the Missouri Botanical Garden Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron Figures 221-224. Philodenron hederaceum. 221-223. P. hederaceum var. hederaceum. 221, 222. Habit. —221 (top L). ( Croat 70891). —222 (top R). (Croat 69834). —223 (bottom L). Cultivated at Kiev Botanical Garden (juvenile form with velvety leaves). — 224 (bottom R). P. hederaceum var. kirkbridei, close-up of stem showing minutely warty surface and anchor roots ( Croat 75108). 636 Annals of the Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 637 Figures 229-232. 229, 230. Philodendron helemae. —229 (top L). Cluster of inflorescences (one open) (Croat & Zhu 76738). —230 (top R). Open inflorescence with protniding spadix (Croat 61275) (Colombia. Valle: Bajo Calima). 231, 232. P jacquimi —231 (bottom L). Habit (Croat 69835). —232 (bottom R). Habit with inflorescences (Croat 12458). 638 Annals of the Missouri Botanical Garden 1997 Croat Philodendron Subgenus Philodendron 639 1997 Philodendron Subgenus Philodendron 641 67213). 642 Annals of the Missouri Botanical Garden / _2 i 9 v (t r ° P ):} Philode ndron hnappiae, habit (Croat 67982). 250-252. P. lazorii Croat. 250, 251. Croat 68953). -250 (top R). Leaf blade adaxial surface. -251 (bottom L). Stem apex with cataphyll fibers and closed inflorescences. -252 (bottom R). (Croat 69833) Stem with cluster of cataphylls. 644 Annals of the Missouri Botanical Garden Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 645 646 Annals of the Missouri Botanical Garden Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 647 Annals of the Missouri Botanical Garden Figures 273-276. —273 (top L). Philodendron ligulatum var. ovatum, habit ( Croat & Zhu 76888). 274-276. ?■ ligulatum var. heraclioanum ( Croat & Zhu 77098). —274 (top R). Habit showing blades spotted on lower surface. . 275 (bottom L). Leaf blade adaxial surface showing sharply D-shaped petiole with narrow wing and dark ring at petio e apex. — 276 (bottom R). Unopened inflorescence, stem, and D-shaped petiole. 1997 Croat Philodendron Subgenus Philodendron 649 Figures 277-280. 277, 278. Philodendron llanense. —277 278 (top R). Cluster of unopened inflorescences (Croat & Zhu (Hammel & McPherson 14526). (top L). Leaf blade adaxial surface (Croat 60505). — 76993A). —279, 280 (bottom L & R). P madronerue Missouri Botanical Garden Figures 281-284. Philodendron malesevichiae (Croat 74818). —281 (top L). Habit. —282 (top R). Stem with intact, persistent cataphylls. —283 (bottom L). Petiole with conspicuous, trichome-like glands. —284 (bottom R). Open 1997 Croat Philodendron Subgenus Philodendron Figures 285-288. Philodendron mexicanum. 285, 286. ( Croat & Bay 65778). —285 (top I tree. —286 (top R). Stem close up. 287, 288. Cultivated by Monroe Binlsey (photo: K. Up Inflorescence. — 288 (bottom R). Open spalhe. Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron m I Missouri Botanical Garden 1997 Croat Philodendron Subgenus Philodendron Figures 301-304. Philodendron pirrense. 301, 302. ( Croat 37944). — 301 (lop L). Adult leaf. 302 (top *ith detached fibrous cataphyll and inflorescence. 303, 304. (Croat 68952). — 303 (bottom L). Apical half with loose inflorescences. —304 (bottom R). Basal half of blade. I 1997 Philodendron Subgenus Philodendron 659 Missouri Botanical Garden 662 663 Missouri Botanical Garden Figures 337-340. Philodendron rothschuhianum. —337 (top L). Adult leaves. Tortuguero National Park (not col- lected; photo: M. H. Grayum). —338 (top R). ( Croat & Zhu 77224) Juvenile leaves. —339 (bottom L). Habit in cultivation at the Missouri Botanical Garden (l Croat 35657). —340 (bottom R). Stem with inflorescences (Croat 27704). Volume 84, Number 3 Croat Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron Figures 349-352. — 349 (top L). Philodendron roseospathum var. angustiUiminatum ( Hammel 3133). 350-352. P. sagittifolium. —350 (top R). Leaf blade adaxial surface ( Croat 67918). —351 (bottom L). Apex of stem with petiole bases, cataphylls, and inflorescences ( Croat & Grayum 60259). —352 (bottom R). Open inflorescence ( Croat 69731). Volume 84, Number 3 1997 Croat 669 670 Annals of the Missouri Botanical Garden 1997 Croat 671 672 Annals of the Missouri Botanical Garden if blade adaxial surface (Croat 66 lature inflorescences ( Croat 74991 » ( Croat 35620). —372 (bottom I Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 674 Annals of the Missouri Botanical Garden Figures 377-380. 377, 378. Philodendron smithiL — 377 (top L). „ 64588) —378 (top R). Open inflorescence ( Croat 40079). 379, 380. P. s 20450). —380 (bottom R). ( Breedlove 25142). . —379 (bottom L). ( Davidse et. > Volume 84, Number 3 1997 Croat 675 Figures 381-384. Philodendron squamicaule ( Croat & Zhu 76798). —381 (top L). Leaf b ade (82, 383 (top R & bottom L). Plant with clusters of inflorescences and persistent cataphyll fib •riginal site on steep road bank. —384 (bottom R). Cluster of inflorescences, one open. I 676 Annals of the Missouri Botanical Garden Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 677 Figures 389-392. Philodendron slraminicaule. —389 (top L). Habit ( Croat 67920). —390, 391 (top R & bottom L). ( Croat 69013) Open inflorescences. —392 (bottom R). Habit ( Croat 66584). I 678 Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 679 680 Missouri Botanical Garden Figures 401-404. 401, 402 (lop L & R). Philodendron subincisum ( Moore & Bunting sulcicaule, leaf blade adaxial surface (Croat 75167). -404 (bottom R). P. tenue, leaf blade 1997 Croat Philodendron Subgenus Philodendron 681 682 Annals of the Missouri Botanical Garden Volume 84, Number 3 1997 Volume 84, Number 3 Croat Philodendron Subgenus Philodendron Figures 421-424. Philode 422 (top R). Leaf blade abax 424 (bottom R). Habit (Croat endron tripartitum. -421 (top L). Cultivated at the Lyon Arboretum (#80-890), habit. — dal surface ( Croat 70136). -423 (bottom L). Stem with open spathe ( Croat 59160). — 56108). Annals of the Missouri Botanical Garden 1997 Croat Philodendron Subgenus Philodendron 687 Annals of the Missouri Botanical Garden ( !°P , L >- .»**• < C "»> 68638)- -126 (top R). Letf b , ith c l us L r inflZ!;' ~^ 2 (bottom L L) - blade adaxial surfaces ( Croat 66711) —428 (bottom R) iritn cluster of inflorescences, one spathe open (Croat & Zhu 76346). Volume 84, Number 3 1997 Croat Philodendron Subgenus Philodendron 15068). Annals of the Missouri Botanical Garden lmS L >- r* Annals of the Missouri Botanical Garden 1997 Philodendron Subgenus Philodendron Annals of the Missouri Botanical Garden Volume 84, Number 3 Croat Philodendron Subgenus Philodendron | 1997 694 Annals of the Missouri Botanical Garden Volume 84, Number 3 Croat Philodendron Subgenus Philodendron 696 Annals of the Missouri Botanical Garden I. — 465 (top L). Habit. —466, 467 (top R&h Volume 84, Number 3 1997 697 Croat Philodendron Subgenus Philodendron Type B Type C Type D 327 mm IlSift •SSS»88Sm3SS§3$S3SS88SS8g8SSS.$g|ll!SSigg88Bgg8S CONTENTS A Revision of Ph mnmnn§%m ss mmmnmnzsszs Annals || of the Missouri Botanical Garden 1997 m Volume 84 Number 4 Volume 84, Number 4 Fall 1997 Annals of the Missouri Botanical Garden The Annals, published quarterly, contains papers, primarily in systematic botany, contributed from the Missouri Botanical Garden, St. Louis. Papers originating out- side the Garden will also be accepted. All manuscripts are reviewed by qualified, independent reviewers. Authors should write the Managing Editor for information concerning arrangements for publishing in the Annals. Instructions to Authors are printed in the back of the last issue of each volume. Editorial Committee Michael H. Grayum Editor, Missouri Botanical Garden Amy Scheuler McPherson Managing Editor, Missouri Botanical Garden Diana Gunter Editorial Assistant, Missouri Botanical Garden Vicki Couture Secretary Ihsan A. Al-Shehbaz Missouri Botanical Garden For subscription information contact Annals of the Missouri Garden, % Allen Marketing & Management, P.O. Box 1897, Lawrence, KS 66044-8897. Subscription price is $120 per volume U.S., $130 Canada & Mexico, $155 all other countries. Four issues per volume. The journal Novon is included in the subscription price of the Annals. amcpher@admin.mobot.org (editorial queries) http://www.mobot.org © Missouri Botanical Garden 1997 The mission of the Missouri Botanical Garden is 1 their environment, in order to preserve and enricl Gerrit Davidse Missouri Botanical Garden Roy E. Gereau Missouri Botanical Garden Peter Goldblatt Missouri Botanical Garden Gordon McPherson Missouri Botanical Garden P. Mick Richardson Missouri Botanical Garden Henk van der Werff Missouri Botanical Garden The Annals of the Missouri Botanical Garden (ISSN 0026-6493) is published quar- terly by the Missouri Botanical Garden, 2345 Tower Grove Avenue, St. Louis, MO 63110. Pe- riodicals postage paid at St. Louis, MO and ad- ditional mailing offices. POSTMASTER: Send ad- dress changes to ANNALS OF THE MISSOURI Botanical Garden, % Allen Marketing & Management, P.O. Box 1897, Lawrence, KS 66044-8897. discover and share knowledge about plants and @ This | ' ANSI/NISO Z39.48-1992 (Permanence of Paper). Volume 84 Number 4 1997 / # * / ^ / # / Annals ^(k ofthe Missouri Botanical Garden A REVISION OF STYRAX (STYRACACEAE) FOR WESTERN TEXAS, MEXICO, AND MESOAMERICA 1 Peter W. Fritsch 2 Mi 1997 719 722 Annals of the Missouri Botanical Garden Volume 84, Number 1997 724 I Sf 728 !U{!. r iHwmiMwrrmnu!H Fritsch Revision of Styrax 735 Figures 58-63. 58-60. Styrax magnus. —58. 1 —60. Fruit, Breedlove 46223 (CAS). 61-63. Styrax adjacent stamens. —63. Fruit. 62, 63, Tenorio 1560 738 740 Missouri Botanical Garden nate up to the calyx margin distally; corolla lobes 5(6), imbricate in bud, 11-18 X 3-7 mm, overlapping, spreading, thin, elliptic, vesture on the outer distal third consisting of stellate hairs. Free portion of the stamen tube 1-6 mm long; distinct portion of filament 2-9 mm long, of equal width throughout, ventrally not auriculate and sparsely white-stellate-pubescent, be- coming glabrous toward the distal end of the filament, the hairs with arms to 0.15 mm long; anthers 3-5.5 mm long, the connectives not or only slightly pro- longed beyond the non-tapered anther sacs. Ovary densely short-stellate-pubescent; style pubescent at base only to nearly throughout; stigma 0.3-0.7 mm wide. Capsule 7-10 X 7-11 mm (broader when 2- 3-seeded), globose, 3-valve-dehiscent, the wall hard, stellate-canescent, smooth but usually transversely wrinkled post-dehiscence. Seeds 6-9.5 mm long; seed West-central Texas on the Edwards Plateau es- carpment and in the Davis Mountains, to Coahuila and Tamaulipas, Mexico, where it exists along the ! Garden 12 2.82 0.11 7 2289 40 Botanical Garden 758 Annals of the Missouri Botanical Garden 40 (1). Araquistain, M. 3635 (19). Arriaga, R. 176 (6). Arsene, B. G. 2841 (16); 5366 (16); 8472 (16). Atwood, J. A3 16 (11a). Avendaiio R., S. 238 (6); 258 (6); 331 (6); 353 (6); 793 (6). Avila, A. de 742 (2). Ayala, M. G. 678 (15) ; Azofeifa, A. 6 (6). Baez, C. G. 124 (6). Barkley, F. A. 7440 (16). Barthol- omew, B. 2914 (16). Bauml, J. 531 (1). Bello, E. 932 (11a); 1031 (6); 2158 (6); 2944 (6); 5310 (6). Berlandier, J. 429 (6). Boege, W. 1765 (16). Bossa, G. 8222 (1); 8335 (19). Botteri 1006 (16). Boutin, F. 2878 (4). Boyle, B. 3841 (16). Brave 2015 (19). Breedlove, D. E. 6366 (1); 6439 (1); 7619 (1); 8917 (6); 10188 (6); 14085 (1); 15588 (4); 16893 (4); 18269 (4); 18414 (4); 18584 (4); 20577 (1); 20635 (1); 21827 (19); 21943 (1); 22841 (19); 23352 (1); 24991 (6); 25152 (1); 25292 (19); 27646 (19); 28186 (6); 28487 (1); 29307 (1); 29696 (1); 30187 (6); 30255 (1) ; 30989 (19); 31045 (10); 31106 (10); 31368 (19); 32394 (10); 32773 (1); 33804 (1); 34377 (6); 39600 (10); 39966 (1); 40224 (19); 40415 (6); 41339 (19); 41446 (10); 41611 (19); 42421 (1); 45097 (8); 45657 (16); 46223 (10); 46979 (1); 48026 (1); 49750 (10); 50020 (19); 50206 (1); 50712 (1); 51000 (1); 52664 (6); 52994 (19); 55749 (19); 57053 (10); 58137 (6); 59921 (16); 61642 (8); 61786 (16); 64233 (16); 64437 (16); 64929 (2); 64955 (16); 65176 (2) ; 67582 (6); 68742 (10); 68828 (6); 68928 (3); 69019 (6); 69083 (19). Brenes, A. M. 180 (6); 298 (6); 345 [258] (1); 3683 (1); 3772 (19); 3971 (19); 3986 (6); 4462 (19); 5403 (19); 5506 (19); 6079 (6); 6380 (19); 6410 (1); 6464 (1); 6734 (6); 6823 (6); 11458 (19); 13599 (6); 17008 (1). Bullock, S. H. 2024 (15). Burger, W. 7390 (19); 7707 (19); 7713 (19); 12004 (6); 12064 (19); 12153 (6); 12400 (6); 12475 (6). Cabrera, E. 1404 (1); 1405 (1); 1988 (1); 3805 (10). Calvert, B. 53 (16). Calzada, J. I. 3076 (6); 4333 (6); 5249 (6); 5534 (6); 8791 (6); 9902 (1); 17816 (8); 17822 (8); 17840 (15); 17960 (6); 18805 (15). Campos, A. 3467 (15). Campos, P. 177 (6). Campos V., A. 3545 (1); 4740 (16). Cardenas, A. L. 79 (15). Carlson, M. C. 690 (19); 2710 (6). Carranza, E., 4033 (9); 4465 (9); 4519 (16). Castillo C., G. 1573 (6); 2203 (6). Cedillo T., R. 670 (6); 1986 (1); 2356 (6). Cevallos, J. 72 (16). Chacdn, A. 494 (19). Chacdn, I. A. 1562 (19); 1577 (1). Chavarria, M. M. 631 (19). Chavez, C. 521 (19); 540 (11a). Chazaro B., M. 1196 (6); 1511 (6); 4675 (6); 5485 (8). Chazaro, M. 2685 (6); 2691 (6); 3759 (6); 6993 (8). Chinchilla, M. 183 (6). Chrysler, M. A. 5590 (19). Cochrane, T. S. 10792 (16); 11942 (15); 12047 (15); 12546 (16). Contreras, E. 4915 (19); 4924 (19); 5102 (19); 11029 (19). Conzatti, C. 2338 (16) ; 2450 (16); 3465 (19); 3480 (6); 4558 (1). Cooper, J. J. 10321 (19). CiSrdoba, J. J. 754 (1). Correll, D. S. 29152 (14a); 29572 (14a); 37003 (14a); 37024 (14a). Cory, V. L. 34763 (14c); 34765 (14c); 34937 (14d); 34938 (14d); 34939 (14d); 34940 (14d); 37767 (14d); 37769 (14d); 38763 (14d); 38764 (14d); 38933 (14d); 38934 (14d); 42659 (14d); 42661 (14d); 42664 (14d); 42666 (14d); 42668 (14d); 42669 (14d); 42673 (14d); 42677 (14d); 42678 (14d); 42953 (14d); 42956 (14d); 49179 (14d); 49453 (14c); 49454 (14c). Cowan, C. P. 4612 (19). Croat, T. B. 34830 (19); 34981 (19); 37190 (13); 45262 (15); 48197 (16); 64211 (6). Cruz C., R. 534 (16). Cuevas G., R. 709 (15); 1251 (16); 1294 (16); 1366 (16); 1876 (15); 1927 (16); 2213 (16); 2391 (15); 2423 (16); 2864 (16); 2890 (16); 4032 (16). Cufodontis, G. 461 (19). D’Arcy, W. G. 9887 (19); 9992 (19). Dario, M. 59 (3); 72 (6); 119 (6); 142 (6). Davidse, G. 1511 (19); 1654 (19); 1656 (6); 10164 (19); 10216 (19); 24017 (19); 24409 (19); 24511 (19); 24632 (19); 26146 (19); 28381 (6); 28554 (19); 28606 (19); 28668 (19); 34298 (6). Davidson, M. E. 776 (19). Delgado S., A. 849 (19). de Nevers, G. 6018 (19); 6848 (13). De Niz L., D. 23 (15); 290 (16). Dfaz B„ H. 2106 (16); 2265 (16); 3656 (16); 4700 (16). Dfaz Luna, C. L. 3592 (16). Dieterle, J. V. A. 4023 (15). Diggs, G. 3804 (16). Dominguez C., R. 28 (4). Dorantes, J. 574 (6). Dorr, L. J. 1910 (14a); 1918 (14d). Dryer, V. J. 1261 (6); 1420 (11a); 1575 (11a). Dwyer, J. D. 454 (1); 7006 (19). Ebinger, J. E. 716 (19). Edwards, J. B. P-290 (1). Eh- renberg 498 (16). Emrick, G. M. 60 (16). Erazo, M. 63 (1). Escobedo, J. M. 1063 (16); 1783 (16). Espinoza, R. 128 (1); 169 (1); 234 (1); 682 (1); 1254 (19). Eugenio, J. 388 (1). Evans, R. 1726 (19). Fairey, J. T26M-79p-101490 (9); T39M-3.5h-031992 (6). Fernandez, A. 807 (6); 1444 (19). Ferrera B., A. 176 (1). Flores C., J. 315 (16). Flores F., G. 793 (16); 1907 (15); 2498 (15). Flores M„ A. 2680 (16). Flores-Franco, G. 3344 (15). Folsom, J. P. 6047 (19). Fonseca J., R. M. 163 (2). Frankie, G. W. 239 (1). Fuentes O., M. 56 (16); 90 (16). Fuentes, Z. 192 (11a). Galeotti, H. 1687 (16); 2851 (6); 2852 (6). Gamboa, B. 71 (19). Garcfa, A. R. 1902 (19). Garwood, N. 334 (19); 761 (1); 1404 (6). Gentry, A. 13797 (12?); 14073 (12?). Gentry, H. S. 4420 (4); 5164 (15); 6244 (4); 10485 (15); 10811 (15); 10863 (15). Ghiesbreght, A. B. 620 (1); 811 (19). Gilly, C. L. 64 (6). Glassman, S. F. 1956 (19). Gliess- man, S. R. CH-54 (19). Gold 129 (16). Goldman, E. A. 872 (1). Gdmez, L. D. 20777 (1). Gdmez-Laurito, J. 8012 (6); 9681 (19); 11345 (6); 11383 (19); 11445 (6). Gdmez- Pompa, A. 793 (6). Gonsoulin, G. 1287 (14a); 1288 (14a); 1289 (14a); 1305 (14a); 1306 (14a); 1307 (14a); 1308 (14a). Gonzalez, J. 108 (1); 340 (16); 627 (19). Gonzalez L„ M. 78 (16). Gonzalez Q., L. 3347 (16). Gonzalez-Es- pinosa, M. 1148 (19); 1279 (19); 1659 (19). Gonzalez- Villareal 4053 (1). Grayum, M. H. 6453 (19). Gregg, J. 955 (8). Guerrero 322 (8). Guerrero N., J. J. 649 (16). Gutierrez, G. 124 (6). Gutierrez, M. 239 (16). Guzman M., R. 2516 (16). Haber, W. 230 (lib); 357 (19); 363 (6); 404 (1); 2917 (19); 3173 (13); 3533 (1); 4330 (6); 5541 (19); 6116 (1); 7353 (6); 7526 (19); 9318 (13); 10631 (6); 11118 (19); 11381 (1). Haenke, T. s.n. and [148] (1). Hagen, C. von 1253 (1). Halbinger, C. 220 (16). Hammel, B. 5734 (19); 6729 (19); 13812 (6); 15271 (6); 19068 (19); 20076 (6). Hansen, B. 1479 (16). Hannon, W. E. 3056 (1); 3461 (1); 4160 (19). Hartshorn, G. S. 1075 (19); 1805 (1); 1894 (lib); 2119 (6). Hartweg 489 (6). Hatheway, W. H. 1258 (19). Hazlett, D. 18-22 Nov. 1974 (6); 667 (1). Heller, C. 234 (6). Helmrich, H. 1690 (6). Henrickson, J. 11499 (14e); 11561 (14e). Henry, M. G. 6557 (14a); 6560 (14c). Hernandez, G., L. 322 (16). Herrera, G. 297 (1); 472 (6); 854 (1); 2046 (6); 3532 (19); 4534 (5); 4911 (6); 5866 (17); 5957 (6). Heyde, E. T. 2915 (3); 6182 (1). Hinton, G. B. 3798 (16); 5282 (16); 5309 (15); 6145 (16); 6193 (15) ; 7410 (15); 7605 (16); 7606 (16); 7618 (16); 7685 (16) ; 10400 (16); 11831 (16); 14409 (16); 14742 (19); 14750 (2). Hitchcock, C. L. 7103 (16); 7105 (16). Hold- ridge, L. R. 2299 (17); 2338 (6). Huerta, M„ M. 233 (16). litis, H. H. 337 (16); 389 (15); 459 (15); 2516 (16). I INBio 16 (13). II InBio 164 (lib). Janzen, D. H. 12108 (1). Jardel, E. 214 (16). Jenny, G. 105 (14a). Jimenez M., A. 345 (1); 1285 (1); 1974 (19); 2305 (19); 3511 (1); 3653 (19); 3983 (19). Jimenez, 0. 420 (1); Jim6nez, Q. 308 (1); 699 (6); 749 (1). Jim6nez, W. 99 (19). Johnson, J. C. 599 (14c). Johnston, M. C. 7304 (14d); 7408 (9); 7428 (14d); 11888 (14e); 11925 (14e); 12845 (9). Juarez, L. G. 62 (6). Judziewicz, E. J. 4792 Volume 84, Number 1997 Fritsch Revision of Styrax 759 (16); 4899 (16); 4973 (16); 5038 (16); 5098 (16); 5212 Kappelle, m’. MK23 (6); MK74 (19); MK118 (6); MK119 (19); MK407 (19); MK417 (19); MK483 (19); MK505 (6); MK861 (6); MK892 (19); MK1088 (19); MK 1176 (19); MK1535 (19); 2800 (19); 2875 (19). Keeney, T. 1456 (14d); 1457 (14d). Kellerman, W. A. 5664 (1); 6415 (1); 7642 (1). Koch, S. D. 7643 (16); 7929 (16). Kowal, R. R. 2805 (16). Kuntze, O. 2286 (19). LaFrankie, F. V. 1306 (18). La Gaza s.n. (6). Lagos- Witte, S. 86 (19). Laguna, A. 279 (1); 351 (1). Lamb, F. B. 115 (19). Lankester, C. H. 139 (6). Lao, E. A. 349 (19); 359 (19). Laughlin, R. M. 408 (16). Lavin, M. 4591 (1). Lems, K. 5340 (19). Lent, R. W. 1681 (19); 1932 (6); 3920 (19). Le6n, J. 269 (19); 295 (19); 3675 (19). Liebmann, M. F. 424 (6); 589 (6); 596B (16). Liesner, R. 14480 (6); 14600 (6); 15533 (6); 26612 (6). Linden, J. J. suppl. 7 (6); 76 (6); 358 (19). Little, E. L. 6043 (19); 9932 (10); 20061 (1). Lopez, E. 7555 (1). Ldpez R„ R. M. 503 (6). Lorea H., F. G. 43 (16). Lott, E. J. 2540 (15). Lundell, C. L. 12631 (6); 19214 (17); 19450 (7); 19618 (7); 20452 (19); 20933 (7); 21126 (6); 21172 (6); 21219 (17). Lyon- net, E. 2106 (16); 1341 (16). MacDaniels, L. H. 942 (6). MacDougall, T. H155 (15); H156 (15). Machuca N., J. A. 6636 (16). Madriz V., A. 9 (19); 67 (1). Magallanes, A. S. 1534 (15); 3432 (15). Ma- gana, R. H. 5607 (6); 5785 (6); 6109 (6). Makrinus, E. 551 (15). Mancera O., A. MO-315 (16). Mdrquez, W. 141 l; 847 (6); 882 (6). . 501 ( 1 j.. S. , M. 55 (16). Martfnez, J. L. 1300 (6). Martfnez S., E. 327 (16); 429 (16); 2857 (16); 3908 (16); 4906 (2). Martfnez-Ico, M. 160 (19). Matuda, E. 467 (19); 568 (19); 2668 (19); 2982 (10); 3930 (19); 4135 (19); 4179 (6); 4335 (6); 5102 (6); 5215 (1); 15410 [5410] (6); 15507 (3); 15922 (6); 15995 [5995] (6); 16265 (19); 17816 (6); 28364 (16); 30405 (16); 30566 (16); 30833 (16); 30858 (16); 31979 (16). Mayfield, M. H. 1632 (15). McCaffrey, D. 77 (1). McPherson, G. 9255 (19). McVaugh, R. 541 (15); 10304 (16); 11764 (16); 11766 (16); 12061 (8); 12100 (15); 13369 (8); 13914 (16); 14279 (16); 15025 (16); 21521 (16); 22813 (16); 23250 (15); 23415 (15); 23515 (15); 23593 (16); 25492 (15); 26020 (16); 26376 (15). Meave, J. 1162 (1). Medina G., 0. 1750-a (16). Med- rano, F. G. 2488 (6); 5032 (16); 5044 (16); 5176 (16); 5191 (16); 6201 (19); 6706 (2); 6785 (16); 6821 (16); 6833 (16); 17054 (16); 17405 (6); 17436 (6); 17455 (6). Mendoza, A. G. 3802 (16); 3862 (16). Mexfa, Y. 1351 (15); 1504 (16); 1861 (15); 9059 (16). Meyer, E. 267 (16). Miller, J. S. 1402 (19); 2608 (16); 2752 (19); 3092 (16); 3114 (16); 3140 (15); 5145 (14c). Miranda, F. 161 (16); 1214 (16); 1580 (16); 3066 (16); 3460 (6); 5295 (1); 5927 (1); 7020 (6); 7055 (19); 7232 (19); 9170 (10); 9173 (6); 9199 (6). Molina R., A. 627 (19); 6460 (19); 6522 (19); 7233 (6); 7524 (1); 12236 (19); 17086 (19); 20321 (1); 21341 (1); 21505 (19); 24404 (6); 24561 (1); 25328 (1); 27644 (19); 34417 (1). Montes, M. 714 (15). Moore, H. E. 3231 (6); 6225 (6). Mora, C. E. 3, p.p. (19). Mora, G. 229 (19). Moraga, M. 263 (6). Morales, J. F. Ill (19); 4266 (6). Moran, R. 10161 (15). Moreno G., S. 38 (16); 176 (16). Moreno, P. P. 480 (19); 1883 (1); 10854 (1); 16366 (11a); 17875 (1); 21064 (1); 21279 (1); 22001 (1); 22569 (1); 23512 (1). Mori, S. 7359 (19); 7416 (19). Mo- rones G., Aj 209 (16). Motte, M. E. 37 (16). Mufioz, E. 4 (19); 36 (16). Museo Nacional 17924 (1). Narave, H. 42 (6); 55 (6); 468 (6). Nee, M. 24719 (6); 26290 (6); 27636 (19). Nees s.n. [1885?] (1). Negrete A., M. 45 (16). Neill, D. 2313 (19). Nelson, E. E. 3750 (6); 6886 (16). Nevling, L. I. 2441 (6). Nieves H., G. 50 (8); 411 (15); 546 (15). Norris, D. H. 14959 (15). Nufiez, J. C. S. 4739 (16); 4785 (16); 5243 (6); 6291 (16); 6328 (16). Opler, P. A. 523 (1). Orozco, J. M. 329 (1). Orated, A. S. 591 (1); 11194 (19). Ortega, J. G. 60 (15); 4118 (15); 5082 (15). Palacios E., E. 34 (1); 1442 (1); 161 (10). Palafox R., M. 1 (6). Palmer, C. W. 97 (6). Palmer, E. J. 9843 (14c); 10237 (14c); 10907 (14a); 11474 (14c); 11528 (14c). Pa- nero, J. 3475 (2); 5254 (2). Paray, L. 681 (6). Parks, H. B. 1009 (14c); 1942 (14c); 40988 (14c). Ptrez J., L. A. 814 (15). P6rez, S. 17 Sep. 1987 (8). Peterson, C. D. 1243 (14b). Pichardo A., V. 14 (16); 28 (16); 111 (16); 212 (16). Pittier, H. 606 (1); 1583, p.p. (1); 1583, p.p. (19); 2140 or 2265 (6); 4242 (12); 5076 (1); 6583 (6). Plowman, T. 14547 (15). Poole, J. M. 1205 (14c). Poveda, L. 223 (1); 1728 (19); 1730 (6); 3843 (1). Pringle, C. G. 2978 (8); 3486 (8); 4416 (8); 6848 (16); 8023 (16); 8129 (6); 9000 (16); 11012 (8); 13104 (6). Proctor, G. R. 25210 (6); 25299 (19); 25514 (19); 31886 (19). Purpus, C. A. s.n. May 1925 “with 10057” (1); 46 (1); 82 (19); 157 (19); 1925 (6); 5317 (19); 7422 (3); 9280 (6); 10082 (19); 10088 (6); 10521 (6); 10531 (1); 10545 (6); 10555 (19); 10612 (6); 13001 (1); 14301 (1). QuinSs, M. 180 (6). Ramfrez D., R. 259 (8); 477 (15). Ramirez, M. 77 (19). Ramfrez R., R. 894 (15). Ramfrez-Marcial, N. 504 (6); 553 (19). Ramos, P. X. 86 (16). Reko, B. P. 3642 (15); 4136 (6). Reverchon, J. 1551 (14a). Reyes Garcfa, A. 356 (1); 463 (1). Reyes, J. R., 1640 (6). Reyna B„ O. 309 (16). Riskind, D. H. 1682 (14c); 1691 (14c); 1915 (14«); 2079 (14e). Rivera, G. 198 (19); 749 (1); 1851 (6). Robles, L. 813 (15). Robleto, W. 87 (6). Rodriguez, J. V. 2004 (1); 2100 (1); 3669 (1); 3760 (1); 10194 (1). Rose, J. N. 7236 (16). Rubio, H. 1583 (6); 1711 (6). Rzedowski, J. 14553 (16); 17211 (6); 18362 (16); 18371 (16); 18529(2); 19771 (16); 22118 (16); 30304 (16); 39745 (16). Salazar L., S. 5 (8). Sanders, A. C. 4405 (16). Sandoval 14 (1). Santana M., F. J. 8 (16); 2695 (16); 3303 (15); 4536 (16); 4776 (15). Sartorius 92 (6). Saynes V., A. 1437 (15) . Schmalzel, R. J. 1608 (19); 1756 (19). Schmitz 505 (16) . Schwabe, W. 224 (1); 77566 (16). Servfn, S. 15 (6); 258 (6); 1056 (6). Sess6, M. 4967 (16). Sharp, A. J. 3460 (6); 4673 (6); 4676 (6); 4677 (19); 4684 (6); 45107 [451047] (3); 45262 (19); 45382 (6); 45499 (16); 45559 (6); 45834 (6); 46154 (6); 46173 (6); 46209 (6). Skutch, A. F. 1935 (1). Smith, A. A113 (19); P1971 (19); P2623 (19); 65 (19); 176 (19); 2700 (19); 4165 (19); 4232 (19); 10079 (19). Smith, C. E. 4473 (19); 4512 (6). Smith, J. 517 (14c); 678 (14c). Smith, J. D. 2266 (1). Smith, J. F. 393 (19). Smith, L. C. 347 (16). Smith, R. F. M149 (14b). Smith, S. G. 6012 (6). Snyder, B. 274 (14c). Soejarto, D. D. 5306 (16). Solheim, S. 1397 (16). Soto Arenas, [M. A.] 1179 (16). Sousa, M. 11635 (1). Standley, P. C. 1674 (1); 1739 (1); 5007 (1); 5686 (1); 6344 (1); 8019 (19); 12351 (1); 14121 (19); 15170 (1); 15875 (1); 15947 (1); 20147 (1); 20405 (1); 32549 (19); 33500 (6); 34672 (19); 41410 (19); 42216 (19); 42254 (19); 42345 (19); 42656 (19); 51137 (6); 51175 (6); 51309 (6); 57958 (1); 58420 (19); 62879 (1); 63714 (19); 63786 (19); 64767 (1); 67865 (6); 68618 (10); 70316 (6); 77656 (1); 80536 (19); 80605 (19); 80659 (19); 80684 (19); 81378 (1); 82830 (1); 82938 (1); 83018 (1); 83697 (6); 85559 (10); 91569 (19); 91577 (19); 92302 (19); 92400 (19); 92574 (19); 92631 (19). Stern, W. L. 1993 (19); 2045 (19). Stevens, W. D. 86 (19); 6687 (6); 13694 (19); 15494 (1); 20401 (11a); 22180 (19); 760 Annals of the Missouri Botanical Garden 22709 (1). Steyermark, J. A. 30624 (1); 30672 (1); 30728 (1); 31150 (1); 32340 (19); 32827 (19); 33138 (1); 33315 (19); 33926 (19); 35418 (1); 36234 (10); 36712 (19); 36801 (10); 36807 (6); 37378 (6); 37989 (19); 42075 (1); 42742 (1); 43199 (19); 43279 (3); 43410 (19); 43593 (19); 43937 (19); 46750 (6); 46759 (19); 47397 (19); 47924 (1); 48441 (19); 48666 (17); 48848 (17); 49738 (6); 49982 (3); 50429 (1); 50699 (1); 50776 (1); 50801 (1); 51100 (1). Stone, D. E. 3286 (6). Sullivan, J. R. 543 (6). Tapia, L. 80 (6). Taylor, C. M. 3210 (19). Taylor, J. 2956 (6); 11848 (19); 17390 (1). Tellez V., O. 10139 (16); 10282 (15); 10412 (15). Templeton, B. C. 7000 (15). Ten- orio L., P. 2642 (2); 8364 (15); 8441 (15); 14351 (16); 15605 (15); 16929 (15); 16984 (15). Tenorio, P. 5707 (1). Terry, M. E. 1338 (19); 1348 (19). Thomas, H. 26 (7); 474 (19); 496 (6). Ton, A. S. 404 (1); 3431 (1); 6494 (1); 7814 (6); 7963 (10); 9329 (16); 9734 (10). Tonduz, A. 459 (19); 7079 (1); 7447 (1); 7667 (6); 9816 (1); 11744 (19); 17924 (1). Toriz A., G. 464 (6); 475 (6). Torres C., R. 2686 (6); 4982 (6). Torres, R. 13321 (16). TnSchez, L. 17 (1). Tyson, E. L. 826 (19); 827 (19). Ugent, D. 1635 (16). Umafla, G. 71 (19); 144 (19). Utley, J. 4557 (19). Valdez, G. L. 616 (2). Valerio, Man. 1378 (1); 1396 (19). Valerio, Mar. 145 (19); 174 (19). VSsquez S., J. 2174 (16). Vazquez, A. 1195 (15); 4164 (15); 4625 (6). Vazquez M., A. 4529 (16). Vela G., L. 2050 (16). Velazquez H., A. 13 (8). Velzen, H. van 2765 (19). Ventura A., F. 784 (6); 866 (6); 1195 (6); 3408 (6); 3511 (6); 5211 (6); 8071 (6); 8116 (6); 12945 (6); 20283 (6); 20303 (6). Vergas, C. 285 (6). Villarreal, L. M. 4585 (8). Villasefior, J. L. 183 (16). Vincelli, P. C. 424 (19). Voorhies, B. 3-11 (1); 5-3 (1); 16-9 (1). Walker, S. 72.021 (16); 73H07 (15). Warscewicz, J. 40 (1); 203 (19). Wawra 1011 (6). Weaver, R. E. 1745 (6); 1754 (6). Webster, G. L. 12230 (19); 12496 (11a). Wendt, T. 148b (14e); 582 (14e); 3576 (12?). Werff, H. van der 6301 (19). Wetter, M. A. 2007 (16); 2057 (16). White, G. 70 (19). White, S. S. 5260 (1). Wiemann, M. C. 32 (19). Wilbur, R. L. 1766 (16); 8821 (19); 16166 (19); 16223 (19); 18026 (1); 21135 (19); 22584 (19); 23494 (19); 27228 (19); 29475 (19). Williams, L. 0. 10491 (1); 11279 (1); 12136 (1); 12314 (1); 12648 (19); 12709 (1); 13270 (1); 13318 (19); 13581 (19); 13709 (19); 14028 (1); 16067 (1); 16131 (19); 16260 (19); 17455 (19); 20001 (19); 23955 (11a); 24844 (19); 25087 (1); 28194 (19). Wood- son, R. E. 484 (19); 919 (19). Woronow, G. 99b (16). Wynd, F. L. 340 (14b). Yuncker, T. G. 5589 (1); 6175 (6). Zamora V., N. 1213 (6). Zavala C., F. 295 (6). Zola B„ M. G. 526 (6); 612 (6). Zuniga, R. 169 (6); 575 (6); 607 (1). Appendix 1. Comparison of Gonsoulin’s (1974) treatment of Styrax from western Texas, Mexico, and Mesoamerica available at the time of Gonsoulin’s revision. (D) = deciduous. S. glabrescens var. glabrescei S. platanifolius var. platanifolius (D) S. platanifolius var. stellatus (D) S. texanus (D) S. youngiae (D) S. S. S. platanifoHu S. platanifoliii Total: 19 species, 24 taxa Volume 84, Number 1997 761 Benzoin Hayne Cyrta Lour. Darlingtonia Tor grandiflorus E. Carranza (2) hintonii (Bullock) Gonsoulin (16) , micranthus (Perkins) D’Arcy (16) var. pilosus Perkins (6) guatemalensis Donn. Sm. (6) hintonii Bullock (16) incamatus P. W. Fritsch (7) .. jaliscanus S. Watson (8) ' - - Fritsch (9) . 8 (5) • leiophyllus Mie limoncillo (Humb. & Bonpl.) Mie magnus Lundell (10) micranthus Perkins (16) myristicifolius Perkins (1) nicaraguensis P. W. Fritsch (11) subsp. ellipsoidalis P. W. Fritsch (lib) .... °% nal “. • • Jal “ Ca . nm .. . (S ' • . Wat '°. n) • Pe ‘ orizabensis Perkins (16) panamensis Standi. (12) penivianus Zahlbr. (13) platanifolius Engelm. ex Torr. (14) subsp. mollis P. W. Fritsch (14b) subsp. stellatus (Cory) P. W. Fritsch (14c) subsp. texanus (Cory) P. W. Fritsch (14d) . subsp. youngiae (Cory) P. W. Fritsch (14e) var. stellatus Cory (14c) ....... polyanthus Perkins (1) polyneurus Perkins (19) psilophyllus A. DC. (5) radians P. W. Fritsch (15) ramirezii Greenm. (16) .... var. micranthus (Perkins) Perkins (16) var. orizabensis (Perkins) Perkins (16) squamulosus M. F. Silva (5) steyermarkii P. W. Fritsch (17) texanus Cory (14d) tuxtlensis P. W. Fritsch (18) vestitus Lundell (6) vulcanicola Standi. & Steyerm. (10) ......... warscewiczii Perkins (19) StHgMa™ w. 0ry . argentea (C. Presl) Miers (1) glabrata (Schott) Miers (5) leiophylla (Miers) Miers (5) psilophylla (A. DC.) Miers (5) SYSTEMATICS OF KAL1MERIS Hong-ya Gu 1 and Peter C. Hoch> (ASTERACEAE: ASTEREAE)' 766 Annals of the Missouri Botanical Garden Figures 1-5. Morphology of achenes. — 1. Boltonia diffusa-, A. Front view. B. Side view. — 2. Kalimeris indica ; Ray achene. B, C. Disc achene of front view and side view, respectively. — 3. / hispidus ; A. Ray achene. B, C ~ ' * * * ... - view. B. Side view. Volume 84, Number - Gu & Hoch Systematics of Kalimeris 771 772 Missouri Botanical Garden Gu & 774 5.8) mm long, the throat campanula!*-, with 5 broadly lanceolate or triangular, equal, subequal, or unequal lobes each (0.4-)0.7-l .8(-2.2) X 0.4- 0.7(-0.8) mm, abruptly narrowed below into a tube 0.6-2 mm long, V4-V4 as wide as the throat, covered with biseriate glandular hairs; androecium synge- nesious, with five anthers 0.7-1. 8(-2.2) mm long, each with a triangular or broadly lanceolate ap- pendage 0.2-0.5 mm long; style 1.5-5 mm long, the two branches linear, (0.4— )0.6-1.2(-1.6) mm long, the lower part with marginal stigmatic lines and the upper part topped by two triangular or broadly lanceolate appendages 0.2-1 .6 mm long, these covered with collecting hairs on the outer sur- face and glabrous on the inner surface. Ray and disc pappus bristles uniseriate, whitish to reddish brown, equal or unequal; ray bristles (0-)0.1-0.8(- 1) mm long; disc bristles (0-)0.1-0.8(-1.5) mm long. Achenes straw-colored to dark purple, or greenish, obovoid to broadly obovoid, compressed laterally, loosely covered with biseriate, 4-cell ed. glandular hairs apically, or with only glandular hairs at the top; the margins light-colored, ciliate, thickened, 0. 1-0.4 mm wide; ray achenes 3(-4)- ribbed, 1 .5-3.2 X 0.6-1 .5(-2J2) mm; disc achenes 2(-3)-ribbed, 1.5-^.2(-3.5) X (0.7-)l.l-2.1(-2.9) Phenology. Flowering period from late May to November. Typification of this generic name is problematic. When Cassini published Aster subg. Kalimeris in 1822, he ascribed to it only one species using the illegitimate combination Kalimeris platycephala. In 1825, he elevated this subgenus to a genus, Kali- meris (Cass, in F. Cuvier) Cass, in F. Cuvier, but no species was transferred (see King & Dawson, 1975). Unfortunately, no specimens that were pos- sibly used by Cassini have been located. Kalimeris platycephala Cass, in F. Cuvier ex Nees (1832) was validly published but with synonymy including ear- lier species names, e.g., Grindelia incisa (Fisch.) Spreng., Aster tataricus L. f. (as A. tartaricus), A. incisus Fisch., and A. sibiricus Hort. plurr. It is clear that K platycephala was an illegitimate name then because all other specific names included in the originally described as K platycephala and thus are in conflict with it, and they both are treated as spe- cies of Aster by contemporary botanists (Kitamura, 1937b; Tamamshjan, 1959; Ling et al., 1985). Be- cause K platycephala is neotypified in this treat- ment (see the discussion under K incisa), Cassini’s designation of the type of the genus Kalimeris is followed here. synonymy were earlier, validly published names. The correct name Nees should have used for that species was K. tatarica, because A. tataricus L. f. (1782) is the oldest name (Article 11.4, Interna- tional Code of Botanical Nomenclature , Greuter et al., 1994). But A. tataricus and A. sibiricus have different morphological characters than the plant Leaves entire to deeply lobed, glabrous or sparsely covered with thick, ascending, non- glandular hairs .. 7. K. lautwran* 1997 Gu & Hoch 775 1. Kalimeris incisa (Fisch.) DC., Prodr. 5: 258. 1836 ["Calimeris"]. Aster incisus Fisch., M£m. Soc. Imp. Natural istes Moscou 3: 76. 1812. Gnndelm incisa (Fisch.) Sprang., Syst. veg. 3: 575. 1826. Heteropappus incisus (Fisch.) Sie- bold & Zucc., FI. jap. fam. nat. 1: 182. 1843. Boltonia incisa (Fisch.) Miq., Ann. Mus. Bot. Lugduno-Batavum 2: 170. 1866. Asteromoea incisa (Fisch.) Koidz., Bot. Mag. (Tokyo) 37: 56. 1923. TYPE: Russia.“Aster incisa, Sibi- ria,” “ Fisch. = Calimeris i. DC." (lectotype, here designated, LE). Kali mens plaiycephala Cass, ex Nees, Gen. sp. Aster. 226. 1832 I '‘Calimeris']. TYPE: Russia. Amur Blagoves- chensk. Karo 182 (neotype, here designated; isoneo- types, BM, E, G (three sheets), K, P). Root fibrous; rhizomes condensed, forming root- stocks up to 3 cm diam. Stem 30-120 cm tall, cov- ered with ascending, uniseriate, nonglandular hairs. Cauline leaves narrowly oblong to obovate, serrate, crenate, or deeply pinnate-lobed, rarely en- tire, (2.6-)3.6-8(-10) X 0.7-2.4(-3.4) cm, glabrous or with scattered ascending nonglandular hairs adaxially, sparsely to evenly pubescent abaxially (mostly along the veins). Capitula 10-55; pedun- cles (1.8— )4— 14(-18) cm long; bracts 1-9 along the peduncle, the lower ones narrowly to linearly ob- long, 8-26 X l-3(-6) mm, the upper ones linear, (2.1-)3.4-8 X 0.3-1. 4(-2.5) mm; involucre hemi- spherical, 7-11 mm broad, 5-7.5 mm high. Phyl- laries in 3—4 whorls, imbricate, covered with short hairs or glabrous, the upper Vt-Vs part of the phyl- ceous and green, the lower part chartaceous, the margins ciliate and membranous, sometimes pur- plish; those of the outermost whorl narrowly oblong or lanceolate, (2.5-)3.5-5(-5.2) X (0.7-)0.9-1.4 mm, apex acute, those of the middle whorl(s) ob- lanceolate, (4-)4.2-5.5(-6.4) X 1.6(-2) mm, the apex acute, those of the inner whorl oblanceolate to oblong, (4.3-)4. 7-6(-6.5) X l.l-1.8(-2.2) mm, the apex acute. Receptacles convex or subconical, (l.l-)1.5-2.4 mm wide, (0.7-)1.2-2.1 mm high. Ray florets (13-)16-27(-29); corollas pale lilac to pale purple, ligules narrowly oblong or oblanceo- late, (8.5-) 1 1 .5-20(-22) X (1.8-)2-2.8(-3.2) mm, the tube (0.9-)l.l-1.6 mm long; style (1.6-)1.9- 2.8 mm long, the branches 0.7-1 .2 mm long. Disc florets (4G-)58-115; corollas 3-4(-4.2) mm long, the lobes unequal, the longer one 0.9-1.6(-2) mm long, the shorter ones 0.6-1.2(-1.5) mm long, both 0.3-0.6 mm wide; tube (0.6-)0.9-1.3(-1.5) mm long and usually covered with biseriate, nonglan- dular or glandular hairs; anthers 0.9-1 .4(— 1.6) mm long; style 2.5-4.3 mm long, the branches 0.8-1 .3 mm long, each with a triangular appendage 0.3-0.4 mm long. Ray and disc pappus bristles whitish to light brown, sometimes unequal; ray bristles (0.2-) 0.5-0.8(-l) mm long; disc bristles (0.5-)0.6-l.l(- 1.5) mm long. Achenes greenish to straw-colored, obovate to broadly obovate, covered with 4-celled, biseriate, nonglandular hairs or these sometimes in- terspersed with biseriate, glandular hairs at the top, the margins 0.2-0.4 mm broad; ray achenes 3- ribbed, 2.6-3.1 X 1.1— 1.5 mm; disc achenes 2- ribbed, 2.6-sl.l X 1.4-2 mm. Chromosome num- ber n = 9. Figure 9A-D. Phenology. Flowering period: late June to Octo- From the two specimens Tamamshjan (1959) se- lected as the type of Kalimeris incisa, the one with the annotated label “Aster incisa, Sibiria” and an added note “Fisch. = Calimeris i. DC” is chosen as the lectotype for K. incisa here. The type spec- imen of K. platycephala was based on a cultivated strain labeled as Aster sibiricus from the “Jardin du Roi,” now the Museum National d’Histoire Natu- relle in Paris. No authentic specimens have been located. In order to follow the typification of the genus Kalimeris made by the first author, Cassini in this case, a well-duplicated specimen collected by F. Karo is selected as the neotype of K platy- cephala here (Greuter et al., 1994: Article 9.6 and 9.11). This species has the most northeastern distri- butional range in the genu subspecies: subsp. incisa la. Kalimeris incisa subsp. incisa Aster incisus var. australis Kitag., Rep. Inst. Sci. Res. Manchoukuo 1: 323. 1936. Kalimeris incisa var. aus- tralis .(Kitag.) Kitag., Neo-lin. fl. Manshur. 652. 1979. TYPE: China. Jilin: Wui-hu-lin Railway Sta- tion, 2 Sep. 1936, M. Kitagawa s.n. (holotype, TI). Stems 40-120 cm tall; branches ascending at an angle from the stem less than 45°. Cauline leaves oblanceolate or narrowly oblong, serrate, less often crenate, or rarely entire, 4.7-7(-10) X (0.7-)0.8(- 776 Annals of the . — D. Achene of K. i 780 Volume 84, Number 4 1997 Gu & Hoch Systematic^ of Kalimeris 783 Volume 84, Number 4 1997 Gu & Hoch Systematics of Kalimeris 785 (IFP), Kitagawa s.n. (TI); Xiuyan Xian, Tangchi-xiang, Toudao-dadui, Xiaozhezigou, W. Wang et al. 454a (IFP); Zhangwu Xian, Namuqiantu-cun, T. Wang et al. 2721 (IFP). Nei Mongol: Hulun Buir Meng, Ergun Youqi, Shangkuli, Kitagawa s.n. (TI); Ju Ud Meng, Hexigten Qi, Guangxingyuan-gongshe, 1200 m, Mengning Team 1150 (PE). Hebei: Badaling, Sato 8870 (IFP, KYO, PE); Bei- daihe, Cowdry 119 (K), Schnack 135d (W); Changli Xian, W. Y. Hsia 1945 (E); Changshanyu, Nakai et al. s.n. (TI); Dongling-shan, Hongsongken, H. F. Chow 40593 (PE); Jizhou Xian, “Pan-shan,” Y. Yabe s.n. (PE); “Paita,” 1015 m, Licent 9 (W); Shanhaiguan, Jiao-shan, 1579 (BM, P, W); Xinglong Xian, Wuling-shan, Hongmeisi, 650 m, P. Y. Fu et al. 4568 (IFP, PE), Nakai et al. s.n. (TI); from Chaihekou to Xinglongtang, Nakai et al. s.n. (TI); from Xinglongtang to Beiyingfang, Nakai et al. s.n. (TNS), Clemens 4153 (E). Beijing: Changping Xian, Xiakou, Xigou, PE Team 1585 (PE); Gubeikou, Wuohu-shan, Sato 10287 (IFP, PE); Qinglongqiao, Kanasiro 4130 (KYO). 4. Kalimeris indica (L.) Sch.-Bip., in Zoll., Syst. Verz. 125. 1854 [ u Calimeris”]. Aster indicus L., Sp. pi. 876. 1753. Asteromoea indica (L.) Blume, Bijdr. 901. 1826. Boltonia indica (L.) Benth., FI. Hongk. 174. 1861. TYPE: China: without locality (holotype, LINN-997/42 not seen; pho- tos, GH, NY; microfiche, MO). Matricaria cantoniensis Lour., FI. cochinch, ed. 1: 498. 1790. Hisutua cantoniensis (Lour.) DC., Prodr. 6: 44. 1838. Asteromoea cantoniensis (Lour.) Matsum., Nip- pon shokubut-sumei ed. 2: 41. 1895. TYPE: China. Guangdong: Guangzhou; type not located. Martinia polymorpha Vaniot, Bull. Acad. Int. G6ogr. Bot. 12: 32. 1903. Kalimeris indica f. polymorpha (Vaniot) Kitam., J. Jap. Bot. 19: 340. 1943. Kalimeris indica var. polymorpha (Vaniot) Kitam. ex Y. Ling, in Y. Ling et al., FI. reipubl. popularis sin. 74: 102. 1985. TYPE: China. Guizhou: Guiyang, 31 July 1897, E. Bodinier 1739 (lectotype, designated by Lauener (1976), E; isolectotypes, E, P (two sheets)). Hisutua serrata Hook. & Am., Bot. Beechey Voy. 265. 1838. TYPE: Japan. Ryukyu Islands: type not locat- ed. Kalimeris indica f. epapposa J. Q. Fu, Bull. Bot. Res., Harbin 3: 111. 1983. TYPE: China. Shaanxi: Kang Xian, Qing-he Forestry Station, 1340 m, 18 Oct. 1963, Z. Y. Zhang 17324 (holotype, WUK). Root fibrous; rhizomes stoloniferous, 0.5-18 cm long. Stem green or purplish, (12-)20-150 cm tall. Cauline leaves linearly oblong, oblong, oblanceo- late, or obovate, entire to deeply pinnate- or lan- ceolate-lobed, (0.7-)2-7(-9) X (0.3-)0.5-1.8(-2.5) cm, glabrous, or densely covered with ascending thin or thick nonglandular hairs, sometimes with scattered biseriate glandular hairs. Capitula 8—120; peduncles (0.3-)l-9(-13) cm long; bracts l-8(-12) along the peduncle, the lower ones 6.2-19(-27) X l-4(-6) mm, the upper ones 1.2— 6.3 X 0.2— 0.9(— 1.3) mm; involucre hemispherical, rarely campan- ulate, 4.5-10 mm broad, 3-6 mm high. Phyllaries in 3-4 whorls, imbricate, usually obtuse at apex. sometimes purplish; those of outermost whorl lan- ceolate or obovate, 1.6-3.7(-4) X (0.5-)0.8-l.l(- 1.4) mm, those of middle whorl(s) oblanceolate, ob- ovate, or spathulate, 2.3— 5.4 X 0.7—2 mm, those of inner whorl oblong to obovate or spathulate, 2.5- 5.7 X 0.8-2 mm. Receptacles convex, 1,1— 1.7(- 2.5) X 0.5-0.9(-1.2) mm. Ray florets 10-26; co- rollas pale lilac to pale purple; ligules narrowly ob- long or oblanceolate to elliptical, glabrous or rarely with scattered long biseriate glandular hairs near the corolla tube, 5.2-16 X 1-3.2 mm, the tube 0.4- 2 mm long; style 1. 3-2.6 mm long, the branches 0.4-1 .3 mm long. Disc florets 40-105; corollas 2.2^4.2 mm long, the lobes subequal or unequal, the longer one (0.8-)l-1.4(-1.7) mm long, the shorter ones 0.6-1 .2(-l .4) mm long, both 0.4-0.6 mm wide; the tube 0.6-1. 2(-l .5) mm long; anther 0.8-1 .6 mm long; style 1. 7-3.8 mm long, the branches (0.4-)0.5-l.l(-1.4) mm long, each with a triangular appendage 0.2-0.4 mm long. Ray and disc pappus bristles equal or unequal; ray bristles (0-)0. 1-0.3 mm long; disc bristles (0-)0.1-0.3(- 0.5) mm long. Achenes brownish to dark purple, obovate to broadly obovate, covered with 4-celled nonglandular hairs and with biseriate glandular hairs at the top, the margins 0. 1-0.3 mm broad; ray achenes 3(-4)-ribbed, 1.5-2.6 X 0.5-1 mm; disc achenes 2(-3)-ribbed, 1.5-2.9 X 0.7-1.4 mm. Chromosome number: n = 18, 27. Figure 14. Phenology. Flowering period: May to November. This is a very variable and widely distributed species. It ranges from central China to South Ko- rea and Japan, and to northern Indochina. It is here divided into three subspecies, as follows: Key to the subspecies of Kaumeris indica la. Leaves glabrous or loosely covered with thick nonglandular hairs 4a. subsp. indica lb. Leaves densely to evenly covered with thin or thick nonglandular hairs and with scattered glan- 2a. Cauline leaves obovate to oblong, covered with thick nonglandular hairs; phyllaries of middle whorl(s) as wide as long; rhi- zomes 5-15 cm long 4b. subsp. collina 2b. Cauline leaves lanceolate to oblong, covered with thin nonglandular hairs; phyllaries of middle whorl(s) Vg-A as wide as long; rhi- zomes 0.5— 1(— 3.5) cm long 4c. subsp. stenolepis Annals of the Missouri Botanical Garden Gu & Hoch 790 1472 (PE), Beixiangqu Forestry Station, H. Gu & Z. Li m, P. X. 300 8 m, 59944 '*SBfl2scaj«;a " isr™* S.tLLan 664 792 Annals of the Missouri Botanical Garden the ray bristles 0. 1-0.3 mm long; the disc bristles 0.1-0.4(-0.5) mm long. Achenes brownish, the margins 0. 1-0.2 mm broad; ray achenes 3(-4)- ribbed, 2-2.5 X 0.7-1 mm; disc achenes 2(-3)- ribbed, 2-2.5 X 0.9-1 .3 mm. Chromosome num- ber: n = 18. Figure 14I-K. Phenology. Flowering period: June to late Octo- ber. Distribution (Fig. 16). Kalimeris indica subsp. stenolepis is endemic to China, where it is especial- ly abundant in Hubei and surrounding central Chi- na. It occurs in Shanxi, Henan, Shaanxi, Gansu, Anhui, Zhejiang, northern Jiangxi, northern Fujian, Hubei, Sichuan, and Guizhou. Its primary eleva- tional range is 200-2000 m, but it occasionally reaches 3000 m in Sichuan. It usually grows in relatively shady and protected habitats. The type collection is from Chaping, Fujian, at the southern limit of the range, and as such does not represent the taxon well, since it includes spec- dividuals from near the geographic center. Handel- Mazzetti’s concept of this taxon was broader than that in the present treatment. He included in it some narrow-leaved plants that we treat as Kali- meris indica subsp. indica, which has some char- brous, thicker leaves and broader phyllaries. Some specimens cited by him (e.g., Tang and Hsia 196 and Bois-Reymond 805a, W) doubtlessly belong to subspecies indica. Some plants from Jiujiang, Jiangxi (e.g., Allison 4, GH; Migo 46, KYO; Sherrers.n. in 1873, K) have less hairy leaves and relatively elongated rhizomes, which are not typical for this subspecies, but other characters indicate that they should be included in subspecies stenolepis. The pappus bristles of this taxon tend to be conspicuous and are often united at the bases. Two collections from Hubei, H. C. Chow 378 (NY) and Henry 2099 (GH, TI), have pappi up to 1 mm long. Additional specimens examined. JAPAN. Nagasaki: Nagasaki Shi, H. Ando 27 (MAK). Kagoshima: Ohshima Gun, Amamiohshima, Ito 510 (TNS); Tarumizu Shi, Shin- midoo, from Kami-shinmidoo to Horikiri, 50-500 m, Mur- ata 11496 (TI). Okinawa: Ishigaki Shi, Ishigakijima, Iwasaki 2 (KYO), Yamasaki s.n. (TI); Itoman Shi, Furuse 1589 (K); Kikai Island, Miyagi 7277 (MAK), between Gu- suku and Hyakunodai, Yoshinaga 523 (MAK); Nago Shi Harano, Kanasiro 18 (KYO), Yaedake, 400 m, Yamazaki s.n. (TI); Naha Shi, Hantagawa, Amano 6806 (KYO, TI, TNS), Yogi Agricultural Inst., Walker et al. s.n. (KYO); Okinawa Shi, Nishihara, Tawada 3 (KYO); Okiyerabu-shi- ma, Ohba 48 (KYO); Shimajiri Gun, Itoman-shi, Koidzumi s.n. (KYO); Shuri, Hatusima 17309 (TI); Yoron Island, Chabana, Tagawa & Iwatsuki 2442 (KYO), Hatusima & Sako 30887 (MAK). KOREA. Junnam: Damyang, Baku s.n. (KYO); Namwon, Mori 335 (TI). Cheju Island: Mal- laisan, Taquet 226 (BM, E, G, W), Taquet 9679 (TI), Nakai 6520 (TI). CHINA. Shanxi: Hongtong Xian, Tong-pu rail- way, T. Kanasiro 4757 (KYO, PE). Shandong: Qingdao Shi, Zhongshan Park, H.Gu&Y. Gao 292 (MO, PE), 293 (MO, PE), 293a (MO, PE), 295 (MO, PE). Henan: Lushi Xian, Laochun-shan, K. M. Uu 4842 (PE), 380 m, J. Q. Fu 365 (KUN, WUK); Luanchuan Xian, Shifangyuan, Henan Team 2080 (PE); Luoyang Xian, K. S. Hao 37699 (PE). Shaanxi: Ankang Xian, Huohe-gongshe, 1100 m, P.Y.U 7928 (KUN, WUK), Langao-gongshe, Shangyi- xiang, 8631 (KUN, WUK); Baocheng Xian, Huangguan- ling, Y. L Qiao 353 (WUK); Feng Xian, Huangniupu, Huji-shan, Z. Y. Zhang 189 (HIB, KUN, PE, WUK); F.op- ing Xian, Gaojiaba, 1150 m, K. T. Fu 4945 (PE, WUK); Langao Xian, Hengbin River, 1800 m, P. Y. Li 8153 (KUN, WUK), Tiefo-gongshe, 1800 m, 8512, (KUN, WUK); Liuba Xian, Miaotaizi, 1600 m,K.T.Fu 6326 (PE, WUK), Zibai-shan, 850 m, 6185 (PE, WUK); Ningqiang Xian, Kuanchuan, 800 m, R Y. Li 94 (WUK), Liejinba, 14 (KUN, WUK), Yangpingguan, Y. L Qiao 108 (WUK), 750 m, T. N. Liu & C. Wang 50 (PE); Lueyang Xian, from Luotuo-xiang to Lianghekou, 870 m, C. L Tang 976 (WUK), from Wangjiatou to Lueyang, 800 m, H. Gu & Q. Han 206 (MO), 208 (MO); Yanglin-zheng, Weihe River, H. Gu & X. Hao 6 (MO), 7 (MO), 8 (MO); Pingli Xian, from Pingli to Yaofu-shan, 1200 m, P. Y. Li 9101 (KUN, WUK), Yaofu-shan, 1200 m, 9178 (KUN, WUK), Weiru- gongshe, 1800 m, 109010 (KUN, WUK), Zafu-shan, 1500 m, 8723 (KUN, WUK), Zhangbaodian, 800 m, 9635 (KUN, WUK); Qinling Mts., Qinfeng-shan, 1500-1700 m, Fenzel 136 (W); Shang Xian, 750 m , T. P. Wang 16022 (PE, WUK); Shiquan Xian, Gangtie-gongshe, Huangguan River, 1020 m, J. Q. Xing 7992 (WUK); Puhe-gongshe, 840 m, 9342 (WUK), 1010 m, 10308 (WUK), 10063 (WUK); Xixiang Xian, Zha-zheng, 600 m, P. Y. Li 7574 (WUK); Xianyang Xian, Fenzel 2471 (WUK); Yang Xian, Huayang, 1350 m, J. X. Yang 1441 (WUK), T. N. Liu & P. C. Tsoong 3500 (WUK); Zhashui Xian, Caiyuyao-qu, 1100 m, X. X. Zhan et al 1067 (WUK); Zhouzhi Xian, Pai 1340 (PE); Ziyang Xian, Chengguan-gongshe, Moshi- gou, P.Y.U 7330 (KUN, WUK); Liangmaya, 700 m, 7004 (KUN, WUK); Maoba, 700 m, 7056 (KUN, WUK); Shu- anghe-gongshe, 1750 m, 7211 (KUN, WUK), 4791 (KUN, WUK), 4933 (KUN, WUK), 7194 (KUN, WUK). Ningxia: Jinyuan Xian, Xisha, Ningxia Team 1-0228 (WUK). Gan- su: Hucheng Xian, from Taihe-gongshe to Zu-zhuang, 1400 m, Z. Y. Zhang 662 (WUK); Guyuan Xian, Wating, 1400 m, T. P. Wang 17106 (PE, WUK); Kang Xian, Changba, Shijiagou, Z. Y. Zhang 16908 (WUK); from Da- caochuan to Xiaocaochuan, W. Y. Hsia 6338 (PE); Tian- shui Xian, Baiyanglin, Tangshangoumen, 1600 m, J. M. Uu 10477 (PE, WUK); Wen Xian, Bikou-gongshe, 900 m, Z. E Zhang 14791 (WUK), 650 m, Y. Q. He 1102 (PE); Dongjiaheba, 14897 (WUK). Anhui: Anqin Xian, Q. Z. Hu 3 (PE), Maekawa 741 (TI); Chu Xian, Langyia-shan, Huadong Field Station Team 3170 (KUN); Guangde Xian, 280 m, Anhui Expedition Team 3300 (NAS); Huaining Xian, Ganlusi, Maekawa 9 834 (TI); Heyuezhou, 7 854 (TI); Linhu-xiang, 745 (TI); Maohuling, 8 B22 (TI); Huang-shan, Chow 33 (A, K, KYO, MO, PE, US), from Chuishilin to Tangkou, T. N. Uu & P. C. Tsoong 2170 (PE), around Tangkou, 2365 (PE); Jiuhua-shan, 520 m, C. S. Fan &Y.Y. U 133 (E), 700-800 m, S. C. Sun 1239 (NY), 1259 (NY); Dongyiesi, Huadong Field Station Team 4795 (PE). Jiangsu: Baoying Xian, Kenzhichang, Nanzha, LJxLfa^Tsioo™^ P. sswwass&tts , ?mZJ f“uu & r. SSSSSSiS SSitSssSSlSc m&5S5Si IFangr & Y. Uu 82264 (KUN, PE), 1300 m, T. T Tsai (-5-1) 1-2.1 : both 0.4r-0.6(-0.7) mm wide, the tube 0.8-1. 4 ray bristles (0-)0.2-0.5 mm long; disc bristles 0.2- mm, the upper ones (0.8-)1.8-5.4 X 0.4-1 mm; involucre 5.5-10 mm broad, 4.2-6.5(-7.5) mm ray ac^n^ 3H)-ribbedL 2.1-3 X 0.9-1. 6 mm; 2.3-3 X 1.4- 1.9 mm. Chro- 61, 62, 63, 66, 70, 71, 72. Figure 17A-C. 796 Annals of the J 1975) r Th?taxo" ^wrfrom “* Gun, “YukSm^&uto “• (K ™*' Cl, ' l, ^ u '“^ i mm Gu & Volume 84, Number 4 1997 Gu & Hoch Systematics of Kalimeris — G. Achene. 1997 Gu & Hoch Systematics of Kalimeris 810 814 Annals of the Missouri Botanical Garden 18770a (5b); Tsai, H. T. 53509 (4a), 53573 (4a), 60729 (4a), 61066 (4a), 62482 (4a), 62827 (4a); Tsang, W. T. 776 (4b), 20844 (4a), 21400 (4b), 21594 (4b), 22759 (4b), 22994 (4b), 23247 (4b), 28036 (4b); Togashi 2641 (8); Tsiang, Y. 1189 (4b), 3282 (4b), 10223 (4a), 10770 (4a); Tsiang, Y. & Wang, H. 16414 (4a); Tsoong, P. C. 26 (4b), 39422 (4a); Tsuchiya, K. 2121 (5a); Tsui, T. M. 696 (4b); Tsutsui, S. 25505 (5a), 27260 (5a), 27532 (5a); Tung, Y. Y. 123 (4a) Umemura 6 (5b); Uno 94 (5b), 22976a (la) Wang, C. 40255 (4b); Wang, C. & Liu, Y. X. 982 (7a); Wang, C. S. 128 (la), 1441 (8), 172 (la), 508 (3), 3371 (8); Wang, C. S. et al. 33 (8), 72 (3), 300 (8), 329 (8), 1846 (3), 1897 (3), 2138 (3), 2367 (3), 2371 (3), 2373 (3), 2381 (3), 2386 (3), 2390 (3), 2424 (8), 2644 (3), 3155 (8), 3180 (3), 3451 (3), 3452 (3), 3453 (3), 3524 (3), 3634 (8), 3772 (3), 3820 (3), 3830 (3), 3855 (3), 3868 (3), 4231 (3); Wang, C. W. 80971 (4a); Wang, F. T. 453 (4a), 2247 (4a); Wang, F Z. 328 (7a); Wang, G. 498 (la); Wang, G. & Li, Q. 26 (la); Wang, G. L. 630160 (8); Wang, G. Z. 6 (8), 18 (8), 177 (8), 321 (3), 834 (8), 2820 (8); Wang, G. Z. & Wang, W. 725 (8); Wang, G. Z. et al. 1015 (8), 1018 (3); Wang, H. 41433 (4a); Wang, H. C. 912 (4a), 4093 (4a); Wang, J. W. & Wu, F. T. 120 (4c); Wang, Q. S. 78 (4a), 302 (4c), 306 (4c), 370 (4c); Wang, T. 1671 (la), 2465 (la), 2572a (la); Wang, T. et al. 2182 (la), 2590 (8), 2721 (3); Wang, T. P. 887 (4a), 1436 (6), 2957 (6), 3626 (6), 5045 (6), 5062 (6), 5975 (6), 8282 (4a), 8910 (4a), 10791 (4c), 11322 (4c), 11865 (4a), 12098 (4a), 13620 (7a), 14891 (7a), 15931 (4c), 16022 (4a), 16313 (6), 17106 (4a), 18561 (7a); Wang, W. T. 2441 (7a), 67644 (4a), 67686 (4a); Wang, T. W. & Liu, Y. 82264 (4a), 84131 (4a); Wang, W. 3526 (3); Wang, W. et al. 178 (8), 454a (3), 548 (8), 1001 (7a), 2647 (8), 2656 (3); Wang, X. W. et al. 192 (4b); Wang, Y. C. 513 (7a), 639 (7a); Wang, Z. X. 168 (7a); Wawra 1214 (7a), 1269 (8); Weboter 196 (8); Wei, Z. P. 1047 (4c), 1966 (4c), 2044 (4c), 3855 (4a), 12199 (4b); Wenyon 1887 (4a); Wilson 1693 (4c), 1703 (4a), 1727 (6); Wright 142 (4a); Wu, K. M. 60165 (4b), 60435 (4b) Xiamen University Team 333 (4b); Xing, J. Q. 1075 (4a), 9342 (4a), 9660 (4c), 10063 (4a), 10308 (4a), 10555 (4c), 10652 (4c), 10901 (4c), 10975 (4c), 11073 (4c), 11273 (4c), 11679 (4c), 12096 (4c), 12619 (4c), 12738 (4c), 12820 (4c), 13056 (4c), 13089 (4c), 17789 (4c), 17990 (4c); Xinjin Bot. Exp. Team 9 (7a), 56 (7a); Xiong, J. H. & Chou, Z. Y. 91840 (4a), 92444 (4a), 94009 (4a); Xiong, S. 7109 (4a); Xiong, Y. G. 10070 (6); Xu, X. H. & Lin, H. F. 153866 (4b); Xu, Y. B. 10051 (4b) Yamamoto 384 (5a), 386 (5a), 3446 (5a); Yamanaka 26157 (5a), 54349 (5a); Yamasaki 8986 (2); Yamatsuta 13 (2) , 80 (3), 84 (3), 85 (3), 87 (la), 93 (la), 94 (3), 98 (8), 99 (la), 104 (3), 105 (3), 111 (7a), 114 (3), 115 (3), 277 (3) , 290 (3), 4050 (2), 4299 (5a); Yamazaki et al. 425 (5a); Yang, G. H. 56939 (4a), 65084 (4a); Yang, J. X. 1441 (4a), 2085 (4c), 5793 (7a); Yang, X. J. 3625 (6); Yang, X. X. 10836 (4a); Yang, Z. B. & Yao, J. 1164 (4c); Yao, K. 8530 (8); Yao, Z. W. 4042 (4a); Ye, H. H. 462 (4b); Yi, W. Q. 60-1238 (4a); Yingkou Team 294 (8); Yogo 3557 (5a); Yoshioka 12 (5a); Yoshinaga 523 (4a); Yu, P. H. 489 (4a), 992 (4a); Yue, J. S. et al. 1706 (4b), 2181 (4a), 2826 (4b), 3865 (4a), 3865 (4b), 4894 (4a) Zhan, X. X. et al. 1067 (4a); Zhang et al. 1864 (3); Zhang, G. C. et al. 81 (4b); Zhang, H. D. 1929 (4b); Zhang, R. M. 679 (4c); Zhang, R. Z. 25652 (4c), Zhang, S. Y. 1052 (4a), 4194 (4a), 6828 (4a), 7107 (4a); Zhang, X. M. 162 (8), 332 (6), 332 (8); Zhang, X. Y. 1131 (4a); Zhang, Y. & Wang, S. 657 (la); Zhang, Y. et al. 1226 (la), 2117 (la); Zhang, Y. L. & Liu, T. S. 8 (8); Zhang, Y. L. et al. 945 (3), 1812 (8), 1833 (8), 2291 (8), 2292 (3); Zhang, Z. R. 25884 (4c); Zhang, Z. W. 2855 (7a); Zhang, Z. Y. 189 (4a), 662 (6), 1237 (4c), 3541 (4c), 3617 (4c), 3789 (4c), 6366 (4c), 6573 (4c), 6839 (4c), 7065 (4c), 7094 (4c), 7884 (4c), 8827 (4c), 9240 (4c), 10815 (4c), 11802 (4c), 13029 (4c), 13732 (4c), 14791 (4a), 16908 (4a); Zhao, D. C. & Wang, T. 1457 (8); Zhao, S. D. & Cui, S. C. 661 (7a); Zhao, Q. S. & Liu, S. S. 5704 (4a); Zhao, Z. X. 550 (4a), 890 (6); Zhejiang Team 27655 (4a); Zhe- jiang Expedition Team 28893 (4c), 29121 (4c), 29159 (4c); Zheng, J. H. 96 (4c); Zheng, Z. 101 (4c); Zhong, S. Q. A60092 (4b), A60821 (4b), A60881 (4b), A62622 (4b); Zhou, H. B. 3021 (6); Zhou, T. P. & Liu, Z. F. 2421 (4b); Zollinger 895 (4a), 899 (4a) (4c), 7992 A REVIEW OF THE GENUS CYDISTA (BIGNONIACEAE ) 1 - 2 array of somewhat overlapping characters (Table 1; Gentry 1977a, 1997; Gentry & Tomb, 1979; Tomb & Gentry, unpublished). The genus most similar to Cydista is Roentgenia, which Gentry (1978) consid- ered “barely separable from Cydista .” Roentgenia differs from Cydista in its trifid tendrils, racemose inflorescence, and colpate pollen according to the generic concepts established by Miers (1863) and Urban (1916), and later maintained by Macbride ^ (1961) and Gentry (1977a, 1978, 1982). Gentry (1978) suggested that Cydista lilacina “links” the A genera Cydista, Roentgenia, and Clytostoma by its elongate racemose-paniculate inflorescence and 4 overall leaf appearance. Clytostoma differs from Cydista in its distinctive echinate fruit and glan- 4 dular-warty ovary, although Clytostoma has a u Cy- Volume 84, Number 4 Hauk 817 822 Annals of the Missouri Botanical Garden Month Flower Fruit -x- Managua - 4 - Caracas Month Flower Fruit -x- Managua 824 Missouri Botanical Garden 9 10 11 12 10 11 12 j Flower ♦ Fruit Managua - + Maracaibo | 1 2 3 4 5 6 7 8 9 10 11 12 3(-4)-colpate 3-colpate Soc. 3: I Garden derms de \ Guevara 1532 3^2156 U^la, 150 la, 164 la, 1669 la, 2376 la, 4054' 3, 428 la, 9 zr&rz ytru 6 ?^ Tones 1089' la; Chacdn & Chaedn 2031 lb; Chaffanjon 153 la; Chaing 234 6; Chan 1145 3, 1393 3, 1823 3, 191 6; Cfom & Burgos 1417 6, 1452 6, 556 4; CAon & 4028' la; Co/e/Za la foZbs 532^ sskhsss la; Coroolio PS 3; Coma 4 Bcoiar 178S la; Gmw, 302^6; Corf 189 la, 78 la; C remCTJ 5025 la, 5227 JfefS jrs& 42276' 3, 42506 3, 43&wl, 43749% 47641^5119 la, 5162' la, 5208' la, 5426' 4, 5439' la, 55333' 3, 5589' 7176^ iVl^ 8552' 6 1^ 8755> 1^9566' 32 ^^^491^C^V GaS ^ C ^ zo ^- la; G ^ m & Dorantes 16 4 ’ <*»**» 51' 3, 6466 la; clrda-Baniga '& JaramOlo 20561 la; Gar- Z JT?Umie?%32 A, 3; LuZ IMS, 238183, 23888 6, 24024 6, 24267 3, 24298 3, 24345 , S^^isSXii&ASL 2359 4 248(>*U 2 2S62'\ ^OIS 'Z 2629 l ' 4^84 4, 3859 la, 3957 la, 4020 4, 4059 4, 4088 4, 45)01 5, 4539 4, 4564' la. 4580 la, 46257 5, 46475 rs-SSS'a 70713; 2, 70828' la, 70878' 5, 710' la, 71485A la. lSh,™ BmI W8' fffjpit^l Kant V 3; Reefer 7 77 3; Kelly 133 6; Kenoyer 533 la; la, 594 la; Knight 69-87 la; Koch 4908 3; Km/ 25028 jisssts.'u-ssir-is 2; Kukle & BoomW l^vist et al3M\*^ ^ ^ SSHSSHFS MM6 la; W 54«J 3»1* SKSS-SJK^gg 840 Annals of the Missouri Botanical Garden Phryganocydia 817, Pleonotoma 815, gssigsigi SYSTEMATICS OF ELEUSINE GAERTN. (POACEAE: CHLORIDOIDEAE): CHLOROPLAST DNA AND TOTAL EVIDENCE 1 Volume 84, 1997 THE FRUITS OF JASMINUM MESNYI (OLEACEAE), AND THE DISTINCTION BETWEEN JASMINUM AND MENODORA 1 f (Oleaceae) Volume 84, Number 1997 Jasminum mesnyi (Oleaceae) 851 852 Annals of the Missouri Botanical Garden 4' «'^^). chaUzal end ' Volume 84, Number 1997 ' (Oleaceae) ii intibiiH jL iiiil.ii! 1 „*i i r iHil.it!] j. y ititiiLnib. },. itllllLilnj, La I liiiifi!,,?!, If* liiitiilfii,,!, J. iijJJil 1 ’ 1 ii Jil m iffll i (Oleaceae) Missouri Botanical Garden REVISION DEL GENERO EUa L. Cabral* y rnida M. GALIANTHE SURG. EBEL1A STAT. NOV. (RUBIACEAE: SPERMACOCEAE) 1 Bar. Gard. 84: 857-877. 1997. Tabla 1. Caracteres del Ebelia subg. Ebelia y del subg. Galianthe. Subg. Galianthe Subg. Ebelia Fruto de mericarpos dehiscentes Fruto de mericarpos indehiseentes Flores siempre heterostilas Flores casi siempre heterostilas, excepto en G. dichotoma Seraillas a veces complanadas, con bordes aliformes Semillas nunca complanadas H&bito erecto, con frecuencia xilopodio muy desarrollado HAbito variado, erecto, postrado, trepador Tallo nunca alado Tallo casi siempre alado Cerca de 40 sp., America del Sur 9 sp M Centro y Sudamdrica 10° -35°S 20°N-35°S ger y Taylor (1993) reconocieron Diodia y Sper- macoce y subordinaron en este gdnero las especies de Borreria. De ahf que un estudio global de los gdneros de Spermacoceae analizando el mayor numero de car- acteres de las especies que los representan, es ne- cesario para fundamentar los lfmites y las rela- ciones de los mismos. Con este fin se ha continuado con el estudio de los gdneros Diodia y Borreria. De estos se han separado algunas pocas especies que ianthe (Cabral, 1991; Pire & Cabral, 1992), pero del que se diferencian por los frutos de mericarpos indehiseentes. Por esto se propone ampliar los lfm- especM . Ebelia Materia 3 y MEtodos Este estudio se ha realizado con material de los herbarios nacionales y extranjeros cuyas siglas se registran de acuerdo con Holmgren et al. (1990) (BA, BAB, BR, CTES, F, HAS, ICN, G, K, UL, LP, MBM, MNES, MO, NY, P, PACA, R, RB, SI, SP, SPF, US). Tratamiento TaxonOmico En el estudio emprendido de las especies amer- icanas de los gdneros Spermacoce, Borreria y Diod- ia se han separado unas especies que no reunen los caracteres de estos gdneros: Borreria dichasia Sucre & C. G. Costa; Diodia brasiliensis Spreng., D. brasiliensis var. angulata (Benth.) K. Schurn., D. cymosa Cham., D. dichotoma (Willd. ex Roem. & Schult.) K. Schum. y D. hispidula A. Rich, ex DC.; y Spermacoce bogotensis Kunth. De esta lista, Diodia dichotoma £ue descrita er- r6neamente en el gdnero Knoxia L., ya que este gdnero es considerado actualmente en una tribu in- dependiente, Knoxieae, que se caracteriza por pre- sentar flores con 6vulos de placentaci6n apical. pdndulos, de mierdpila supera. Este error fue en- mendado y la especie fue reubicada sucesivamente en los gdneros Spermacoce, Borreria, y Triodon DC. por distintos autores y finalmente como Diodia di- chotoma (Schumann, 1889). Borreria dichasia fue descripta como Borreria a pesar de no haber contado con material con frutos. Diodia cymosa y D. hispidula fueron descritas den- tro del gdnero Diodia por sus frutos de mericarpos indehiseentes. Spermacoce bogotensis fue reubicada en el gdnero Diodia o Borreria por distintos autores. Diodia brasiliensis Spreng. fue separada por De Candolle (1830) junto con D. anthospermoides Cham. & Schltdl. y D. polymorpha Cham. & Schltdl., en un nuevo genero, Triodon. Este autor caracterizo a dicho genero por el hdbito sufiuticoso, las inflorescencias en fascfculos espigados y los frutos de mericarpos in- dehiscentes y lo nomino haciendo alusion a los ties diminutos dientecitos, correspondientes a restos de hacecillos vasculares, persistentes en el dpice de los pedicelos al caer los frutos. Posteriormente Bentham agrego dos nuevas especies, Triodon angulatum de Mexico y T. laxum de Ecuador. Este gdnero no fue aceptado por algunos autores (Schumann, 1888; Stan- dley, 1930; Steyermark, 1974) y posteriormente sus i dentro del gdnero Diodia, de mericarpos indehiseentes. De Candolle al describir Triodon no registry el dimoxfismo floral, a pesar de que incluy6 como basonirno de T. polymorphus a Diodia polymorpha Cham. & Schltdl. En esta especie cripcidn detallada de las flores brevistilas y longistilas que tienen, si bien no us an estos terminos. Schumann (1889) considers a las tres especies citadas por De Candolle, sinonimos de Diodia polymorpha, a pesar de que registrb a una especie anterior D. brasiliensis Spreng. (1825) entre sus sindnimos. El grupo de especies aquf estudiado se correspon- de con los caracteres del genero Triodon, pero este nombre gendrico estd invalidado por un anterior (Tabla 1) y en su lugar debe use Cabral y Bacigalupo 859 Galianthe Subg. Ebelia alianthe S Junfn, on the Rfo Blanco, small affluent of the rfo Piedras, 19 Feb. 1944, Fosberg 21484 (NY); Rfo San Francisco, above Bogota, 13 set. 1917, Pennell 1934 (NY); Sabana Cabral y Bacigalupo Galianthe Subg. Ebelia Cabral y Be Galianthe S 866 Annals of the Volume 84, Number 4 Cabral y Bacigalupo 867 1 .6-2.8 Galianthe Subg. si:. Garden 871 872 Annals of the Missouri Botanical Garden Volume 84, Number 4 1997 Annals of the Missouri Botanical Garden Cabral y Bacigalupo 877 Galianthe Subg. Ebelia gEnero galianthe subg. EBELIA (RUBIACEAE: SPERMACOCEAE): ESTUDIO PALINOLOGICO 1 Palinologi* ST (A° Rich, ex I Volume 84, Number 4 TIPO :UrMr * *• •> o* • i •• Volume 84, Number - Estudio Palinologico 885 Cabral, E. L. 1985. THE TAN AKA-KAIY ONG Li Xi-wen 2 and Li Jie 2 LINE— AN IMPORTANT FLORISTIC LINE FOR THE STUDY OF THE FLORA OF EAST ASIA 1 Li & Li Botanical Garden CHROMOSOME NUMBERS IN Harold Robinson \ Gerald D. Carr » COMPOSITAE, XVII: Robert M ' Kmg ‘' and A ‘ Michael Po SENECIONEAE IIP a Robinson et al. Chromosome Numbers in Compositae Table 1. Continued. *S. hieronymi Grise S. jacobaea L. S. lauius G. Forst. ■ S. lautus S. linearifolius . S. linearifolius S. linearifolius S. linearifolius S. lueidus (Sw.) S. lueidus A. Rich. DC. S. nuidagas cariens is Pi S. minimus Poir. S. nivalis (HBK) Cuatr S. odoratum Homem. S. pterophorus DC. S. pterophorus *S. subulatus G. Don e *5. subumbellatus Phil *5. tephrosioides Turc S. vagans F. Muell. S. veilleiodes A. Cunn S. veilleiodes S. viravira Hieron. S. warszewiczii A. Br. & Bouch6 2 n 2 n 2n = 10 pairs 2 n = 20 pairs 2n = ca. 20 pairs 2 n = 30 pairs 2n - 30 pairs 2n = ca. 30 pairs 2n = ca. 30 pairs n = ca. 50 2n = 50-52 pairs 2 n = 30 pairs 2 n = 20 pairs 2 n = 10 pairs 2n = 10 pairs 2 n = 10 pairs 2 n = 10 pairs 2n = 10 pairs n = 17 or 20 2n = 20 pairs 2n = ca. 20 pairs Argentina. Buenos Aires: K10289 Australia. Victoria: K9785 Australia. Western Australia: K9531 Australia. Western Australia: K9597 Australia. Victoria: K9776 New Zealand. North Island: K10266 Australia. Victoria: K9739 Australia. Victoria: K9752 Australia. Victoria: K9786 Australia. New South Wales: K9975 Dominica. St. George: K6379 France. Martinique: K10647 Australia. New South Wales: K9972 Australia. Victoria: K9741 Ecuador. Pichincha: K10045 Australia. Victoria: K9754 South Africa. Natal: K10205 South Africa. Cape: K10215 South Africa. Cape: K10212 South Africa. Natal: K10207 South Africa. Cape: K10220 Mexico. Veracruz: K6497 Bolivia. Cochabamba: K9639 Argentina. Rfo Negro: K9349 Argentina. Mendoza: K9423 Ecuador. Sucumbfos: K10126 Australia. Victoria: K9750 Australia. Victoria: K9748 Australia. Victoria: K9788 Argentina. Rfo Negro: K9360 Argentina. La Rioja: K9457 Guatemala. Chimaltenango: K7200 Guatemala. Quezaltenango: K7029 and the genus Ligularia and its relatives with n = 30, and a discussion of the Senecw aureus group with n = 22-24. The paper ends with, “The oc- currence of species of Senecio with n = 10 in por- tions of Europe and Africa, and the concentration of Senecioninae and Othonninae with n = 10 in Africa strongly suggest an Old World origin for the tribe, with subsequent wide migration and diversi- fication nearly throughout the world.” These points are enlarged upon below. The present paper adds and discusses a few chromosome records of note from the Andean genera that were previously re- ported cytologically only by Turner et al. (1967) and Powell and Cuatrecasas (1970). Within the Senecioneae, at the subtribal and ge- neric levels, two particular microcharacters have come into taxonomic use: (1) the anther collar being jylindrical versus balusterform (with a zone of en- arged cells below, Robinson & Brettell, 1973b; Sfordenstam, 1978; Vincent, 1996), and (2) the stig- natic surface covering the whole inside of the style branch versus forming two separated lines on each branch. These characters were observed and illus- rated by Cassini (1818), but they were cited only sparingly afterwards until they were used to delimit he generic concepts of Senecw sensu lato and the Iussilaginae (as cacalioids) by Robinson and Bret- tell (1973b) and Nordenstam (1978). This narrower concept of Senecio has been resisted by some au- thors, ana uic , ... been questioned due to some variability (Wetter, 1983); however, the value of the microcharacters is now generally recognized (Vincent, 1996) and the narrower generic concepts are now generally ac- ?mri ' i of n = 10, ca. * ^^yTcapbfd r^TIhow^'T 108 ’ ^ studies are still to be done. Recognition of Packera as a distinct genus seems premature at this time. (3) Mulgedifolii group, n = 20, 40 The Mulgedifolii group of Mexico lacks ray flo- rets and has red or white, but never yellow, corollas. The combination of characters led some early au- thors to place some of the species in the broad concept of Cacalia, and they appear among the ex- cluded taxa listed by Pippen (1968). The Mulge- difolii have the characteristic balusterform anther collars and separate stigmatic lines of the Sene- cioninae. The group was retained in Senecio by Barkley et al. (1996). There is a report of n = 22 for S. runcinatus Less. (Keil & Stuessy, 1975), and the same species is reported here as n — 17 or 20. Kowal (pers. comm.) has indicated that the n = 22 was probably wrong. The unpublished summaries of the Mulgedifolii chromosome counts by Villasefl- or list 10 counts for 6 species (1986) and 33 counts for 15 of the 17 species (1991). All listed counts, except the dubious n = 22, are n — 20 with one rio (Baldwin, 1946). The related Af- rican species of Senecio that were reported with n = 5 by Turner and Lewis (1965) and Lawrence (1985a, b) have all been transferred to Emilia by Jeffrey (1986). The lower numbers correlate with the short-lived, weedy habit of the genus. This sit- uation parallels that of Fleischmannia microstemon (Cass.) K. M. King & H. Rob. (Baker, 1967) and many other short-lived species with lower DNA content (Bennett, 1972). The low number is con- sidered here a derived condition in Emilia, a re- organization of chromatin on fewer centromeres as in Fleischmannia microstemon (Baker, 1967) and Crepis (Tobgy, 1943). Such a reorganization, or “ge- nome congealing” (Wagner et al., 1993: 422-423) has been supported by a measurement of the nu- clear DNA content of E. discifolia (Oliv.) C. Jeffrey (as Senecio discifolius Oliv.) (Lawrence, 1985b). In structural features, Emilia is a member of the Senecioninae, with paired stigmatic lines and bal- eral subgroup with Othonna and Dendrosenecw in the cpDNA study of Kadereit and Jeffrey (1996). (5) Senecioninae with n = 10, 20 10 in the neotropical Pseudogynoxys, questioned by Jeffrey (1992), must be incorrect. All other reports for the genus are n = 45-48 including P. cheno- podioides (HBK) Cabrera with n = 45 + 5 (Turner et al., 1962, as S. confusus Britton) and our report of P. sonchoides (HBK) Cuatrec., representing two morphological extremes of the genus. Jeffrey’s (1992) summary also gives both n = 10 and n = 20 for S. flocculus Less, of Mexico and S. bras- iliensis (Spreng.) Less, of South America. These iso- lated occurrences of n = 10 in basically n = 20 number, but are more likely reductions resulting from polyhaploidy or aneuploid reduction as in Em- ilia and some Tussilagininae such as Doronicum, Ligularia, and Petasites. An interesting close relationship between an Af- rican species with n = 10 and an American species with n = 20, both autogamous, annual, desert spe- cies, is suggested by isozyme analysis (Liston et al., 1989) and cpDNA evidence (Liston & Kadereit, 1995). In these papers, the North American Senecw mohavensis A. Gray was considered a closely re- lated derivative, through long-distance dispersal, of S.flavus (Decne.) Sch. Bip. subsp. breviflorus Kad- ereit of the Saharo-Arabian deserts. Radford et al. (1995) indicated that the aggres- sively weedy taxon commonly identified as Senecio madagascariensis Poir. was consistently 2 n = 20 based on material from Madagascar, South Africa, Australia, and Argentina specimens, and our count is the same. In contrast, Radford et al. (1995) re- ported the common Australian species S. lautus G. Forst ex Willd. as consistently 2n = 40 (see also the present study; Turner, 1970; Lawrence, 1980; Webb, 1988). However, see the report here of 2n = 10 pairs for material determined by Lander, Short, and Wilson as S. lautus from Western Aus- tralia. Sindel (1996) summarized the literature on the four subspecies of S. lautus. (6) Andean genera Lasiocephalus and Pentacalia, n = 20, 30, 45-52 Our counts confirm the few previous counts for two Andean genera. Further work on chromosome numbers of Andean Senecioneae is needed. Lasi- ocephalus (including Aetheolaena) has previously been known cytologicaUy from one species, L - mm iHieron.) Cuatrec. (Turner et al., 1967). 84, Number Robir Chror J. R. Foret. & G. I i H. Rob. & Brettel V & E Asia 5 & SE Asia, W Robinson et al. (4) Emilia, n = 5. og Garden BOOK REVIEW 908 Annals of the Missouri Botanical Garden Still, botanists from as far away as Manaus have found a great deal of overlap with this area and their own local flora. One helpful addition would be to include some information about each species’ overall distribution, for instance, if it is a narrow endemic or a widespread American weed. Altogether, this volume ranks right at the top of its class. It follows the tradition of extremely infor- mative local floras like Tom Croat’s Flora of Barro Colorado Island and then provides the kind of vi- sual aids that will entice even casual aficionados to explore the flora of lowland South America, be it vicariously, browsing through this book, or by getting their feet dirty and visiting a now well-doc- umented site such as the region surrounding the French Guiana village of Saiil. — Paul E. Berry, Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166, U.S.A. NOTICE Ann. Missouri Bot. Gard. 84 : 909 . 1997 . ANNALS OF THE of Plant CONTENTS A Revision of Styrax (Styracaceae) for Western Texas, Mexico, and Mesoamerica ..... Peter W. Fritsch 705 Systematics of Kalimeris (Asteraceae: Astereae) Hong-ya Gu & Peter C. Hoch 762 A Review of the Genus Cydista (Bignoniaceae) Warren D. Hauk 815 Systematics of Eleusihe Gaertn. (Poaceae: Chloridoideae): Chloroplast DNA and Total Evidence Khidir W. Hilu & John L Johnson 841 The Fruits of Jasminum mesnyi (Oleaceae), and the Distinction Between Jasminum and Menodora Jens G. Rohwer 848 Revision del G^nero Galianthe subg. Ebelia stat. nov. (Rubiaceae: Spermacoceae) ... Elsa L Cabral y N6lida M. Bacigalupo 857 G6nero Galianthe subg. Ebelia (Rubiaceae: Spermacoceae): Estudio Palinoldgico Stella Maris Pire 878 The Tanaka-Kaiyong Line — An Important Floristic Line for the Study of the Flora of East Asia Li Xi-wen & Li Jie 888 Chromosome Numbers in Compositae, XVII: Senecioneae III Harold Robinson, Gerald D. Carr, Robert M. King & A. Michael Powell 893 Book Review. Guide to the Vascular Plants of Central French Guiana. Part 1 by S. A. Mori et al., reviewed by Paul E. Berry 907 Notice 909 Checklist for Authors . 910 Cover lUv