JOURNAL OF THE ARNOLD ARBORETUM HARVARD UNIVERSITY VOLUME 64 NUMBER 1 US ISSN 0004-2625 Journal of the Arnold Arboretum Published nc in January, April, July, and October by the Arnold Arboretum, Harvard Univers Subscription price $30.00 per year. Sul al I d be sent to Ms E. B. Schmidt, Arnold Arboretum, 22 Divinity Avenue, Cambridge, Massachusetts 02138, U. S. A. Claims will not be accepted after six months from the date of issue. POSTMASTER: send address changes to Ms E. B. Schmidt, Arnold Arboretum, 22 Divinity Avenue, Cambridge, Massachusetts 02138, U. S. Volumes I-51, reprinted, and some back numbers of volumes 52-56 are available from the Kraus Reprint Corporation, Route 100, Millwood, New York 10546, U.S.A EDITORIAL COMMITTEE S. A. Spongberg, Editor E. B. Schmidt, Managing Editor P. S. Ashton K. S. Bawa P. F. Stevens C. E. Wood, Jr. Printed at Allen Press, Inc., Lawrence, Kansas COVER: The stylized design appearing on the Journal and the offprints was drawn by Karen Stoutsenberger. Second-class postage paid at Boston, Massachusetts, and additional offices. JOURNAL OF THE ARNOLD ARBORETUM VOLUME 64 JANUARY 1983 NUMBER | THE 1980 SINO-AMERICAN BOTANICAL EXPEDITION TO WESTERN HUBEI PROVINCE, PEOPLE’S REPUBLIC OF CHINA B. BARTHOLOMEW, D. E. BOUFFoRD, A. L. CHANG, Z. CHENG, T. R. DuDLEy, S. A. HE, Y. X. Jin, Q. Y. Li, J. L. Luteyn, S. A. SPONGBERG, S. C. SUN, Y.C. TANG, J. X. WAN, AND T. S. YING THE BOTANICAL SOCIETY OF AMERICA’S Committee for Scientific Liaison with the People’s Republic of China, chaired by Peter H. Raven, director of the Missouri Botanical Garden, and a corresponding committee of botanists in the People’s Republic of China, chaired by Tang Pei-sun, director of the Institute of Botany, Academia Sinica, Beijing, successfully sponsored and arranged exchange visits of American botanists to the People’s Republic of China in 1978, and Chinese botanists to the United States in 1979. These delegations were limited to visiting botanical institutions, and the exchange resulted in a basic understanding by the participants of the current status of botanical research in both countries (see Thorhaug, 1978; Bartholomew, Howard, & Elias, 1979; Howard, Bartholomew, & Elias, 1979). At the conclusion of the visit by the Chinese delegation to the United States in early June of 1979, a meeting was convened at the University of California, Berkeley, to discuss the kinds of cooperative programs that might be developed in the future. The general con- sensus of the participants at that meeting was that steps should be taken to promote cooperative ventures beyond the delegation level, and that joint par- ticipation in botanical research was highly desirable. As a direct result of the Berkeley meetings, five American botanists were invited by the Academia Sinica to participate in a botanical expedition in China with Chinese colleagues. In return, five Chinese botanists were invited to the United States for a one-year period to study at botanical institutions and to take part in fieldwork. The purpose of this report is to summarize the activities and results of the 1980 Sino-American expedition, the first joint © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 1-103. January, 1983. 2 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 botanical expedition involving botanists from the United States and the Peo- ple’s Republic of China since the founding of the People’s Republic in 1949. The 1980 Sino-American Botanical Expedition to western Hubei! Province was conducted under the auspices of the Academia Sinica and the Botanical Society of America and consisted of a three-month (15 August-15 November 1980) visit to the People’s Republic by the American participants. Professor S. C. Sun, director of the Wuhan Institute of Botany and chairman of the Department of Biology, Wuhan University, was the expedition leader. Most of the field investigations were conducted in the Shennongjia Forest District in western Hubei Province, with additional fieldwork in the metasequoia region of Lichuan Xian in southwestern Hubei. The herbarium collections made in both areas are enumerated in the list presented below, but a report of the brief visit to the metasequoia region is presented in a separate paper (Bartholomew, Boufford, & Spongberg, 1983) that follows this report Once the fieldwork had been concluded, the American participants visited several botanical institutions and botanical gardens throughout China before returning to the United States. During that period, time was available for individual study in the herbaria and libraries and for observation and discussion of the research being conducted in those institutions. The following sections summarize the ecological, physical, and climatic con- ditions of the Shennongjia Forest District, previous botanical exploration in the area, and observations on current lumbering practices in the region. A list of determinations of the collections concludes this report. During the six weeks spent in the field, 2085 collections were made, of which 1715 numbers represent vascular plants. A separate report concerning the collections of nonvascular plants is planned. NATURAL CONDITIONS IN THE SHENNONGJIA FOREST DISTRICT GEOGRAPHIC POSITION. The Shennongjia Forest District (MAp |), with an area of about 3250 km? (slightly larger than the state of Rhode Island, and almost exactly the same size as Yosemite National Park, in the United States), is situated at 31°15’-31°57’ north latitude and 109°56’—1 10°58’ east longitude. It lies in the northwestern part of Hubei Province and is bounded on the west by Sichuan Province. The district was created in 1970 from what were parts of Badong Xian (county), Xingshan Xian, and Fang Xian and was set off as a separate administrative division at the level of xian. GEOLOGY AND GEOMORPHOLOGY. The Shennongjia Mountains are an extension of the southeastern Dabashan Mountain system, a subrange of the Wushan Mountains (Ying et a/., 1979). The region was gradually uplifted in the middle Devonian and was formed into a mountain body through the Yanshan, and then the Himalayan, mountain movements, with rising continuing after these movements. As a result of strong, constant erosion that kept pace with the IiThea Dp; - t fa ization i names of older collectors, authors, and places for which sane in spelling might cause confusion. In these instances the Wade-Giles spelling is given in parenthes 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 3 7 : ) Shaanxi ee “Anhui ; ea Henan Provi ‘N ; Province ~ ovince Province XN, noe (Shennongjia ee ~, ‘ N, j Forest District HANJIANG ( | . tees RIVER aay a7 Sichuan Hubei eat | Province oe Province —. pw YICHANG WUHAN CHANGJIANG \ RIVER = f a : 2° Jiangx ° : Pa 5 \ Hunan Province ad Province 1 iN Map 1. Outline of Hubei Province indicating location of Shennongjia Forest District (stippled). uplift, the range appears young geologically, as evidenced by high mountains, deep valleys, and steep slopes (FiGuRE 1). In some areas a karst topography has developed. The rock formation consists primarily of sedimentary rock (mainly Sinian siliceous limestone, Cambrian and Ordovician limestone and shale, Silurian calcareous shale, and Tertiary sandstone) interspersed with some metamorphic rock. Quaternary sediments are undeveloped. The mountains in the Shennongjia Forest District are oriented in a nearly west-southwest to east-northeast direction and have an average elevation of around 1800 meters. The main mountain peaks are located slightly south of the central part of the district and include Laojun-shan (2936 m), Xiaoshen- nongjia (3005 m), Dashennongjia (3052 m), and Wuming Peak (3105 m), the highest in Shennongjia. However, the areas along the river valleys, such as along the Yangriwan and Yinyu-he rivers, are at an elevation of only several hundred meters. The altitudinal differences between these areas and the main peaks are usually more than 2000 meters. Many deeply incised, V-shaped valleys occur in Shennongjia, forming a radiate drainage pattern (see Map 2), with the water flowing into the Changjiang (Yangtze) and Hanjiang (Han) rivers (the Songlo-he and Nicha-he rivers in the north and the Yinyu-he and Loyang- he rivers in the west flow into the Hanjiang; the Jiuchong-he and Dangyang- he in the southeast and the Yandu-he in the southwest flow into the Changjiang). CiimatTe. Shennongjia is situated in a transitional area between the higher mountains of southwestern China and the low, hilly regions of the southeastern part of the country, and its climate is characteristic of an east-west transitional zone. It also lies in the path of monsoons moving north. Moreover, due to the topographical features of high, steep mountains and deeply incised valleys, the microclimates in the areas that we explored vary greatly, providing habitats 4 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Figures 1,2. Shennongjia Forest District: I, view southwest from Xiaoshennongjia, alt. el thickets Of Sinarumnainaria nitida in right foreground; prominent conifer, Abies farges 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 5 TABLE 1. Meteorologic data for Songbaizhen and Dajiuhu, Shennongjia Forest District. No. oF ANNUAL RE- TEMPERATURE (°C) FROST FREE CIPITA- STATION & Mean Mean ean Maxi- DAYS TION ELEVATION annual January July Minimum mum _ Per YEAR (in mm) Songbaizhen, 935 m 12.2 1.0 23.5 —17.7 36.4 227 973.7 Dajiuhu, 1700 m 7.4 —4.9 18.5 =21.2 34.5 144 1528.4 that range from warm temperate-subtropical at the lowest elevations to essen- tially boreal at the summits of the highest peaks. Weather stations were only recently established in this district, but there are not enough of them to provide detailed climatic data for much of the area. The data in TABLE | are based on meteorologic observations recorded at Songbaizhen (935 m alt.) and Dajiuhu (1700 m alt.) and made available by the Meteorologic Service of the Shen- nongjia Forest District. Based on the available meteorologic data, the climate of Shennongjia appears to be warm temperate to temperate, with an abundance of moisture. There is far too little information from the areas along the river valleys at low elevations and from the highest peaks, but based on the distribution of the vegetation it is apparent that the lowest elevations support a few subtropical taxa while the highest elevations are occupied almost exclusively by boreal elements. Sorts. There are three main types of soil, which are vertically distributed in Shennongjia (Ying et al., 1979): a yellow-brown forest soil belt, a mountain brown forest soil belt, and a mountain gray-brown forest soil belt. The yellow-brown forest soil belt is found at elevations below 1500 meters. The parent materials forming this soil are mostly purple and red arenaceous shale (pH 6-7). Organic materials are thoroughly decomposed due to the mois- ture/heat conditions and the wet and dry seasons. The mountain brown forest soil belt occupies a position between 1500 and 2200 meters elevation. The parent materials are mainly limestone, micaceous sandstone, and quartz-containing sandstone (pH 4.5-6). This belt is densely covered with vegetation and has a rather good accumulation of humus. The mountain gray-brown forest soil belt is located at elevations above 2200 meters. The parent materials are limestone and quartz-containing sandstone (pH 4.5-6.5). The organic layer 1s well developed. FLORA AND VEGETATION. The floristic and vegetational characteristics of Shen- nongjia have been discussed to a greater or lesser extent by Peng (1957), Wu (1979, 1980), Ying et a/. (1979), and Ban (1980). However, we would like to comment on aspects of the flora and vegetation in the areas that we visited. Natural vegetation predominates over much of the area but 1s being altered 6 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 at an alarming rate. At lower elevations (generally below 1800 m) much of the natural vegetation has already been totally destroyed for timber or to provide crop land. Virgin forests are for all practical purposes nonexistent, although there are a few areas at middle elevations that support well-developed secondary forests containing a rich variety of tree species, and there are some relatively undisturbed coniferous forests at the highest elevations. Nevertheless, even these areas are being exploited at a rapid rate, and many steep slopes at middle and high elevations have been completely stripped of forest cover. This practice has led to serious erosional problems, which will definitely worsen as more areas are cut Coniferous forests are far less extensive than broad-leaved forests, due partly to the fact that coniferous forests that have been lumbered have mostly been replaced by broad-leaved ones. Except in certain areas at high elevations (e.g., on Mts. Laojunshan and Xiaoshennongjia; see FIGURE 2), pure, intact, conif- erous forests are scarce. Vertical zonation of the forests in Shennongyjia is not easily observable. The original vegetation does not cover large areas, and it is difficult to distinguish the upper and lower boundaries of the various vegetation types. Some idea of the altitudinal zonation can be achieved by piecing together data from various parts of the district, but this may not reflect the true distribution of the original forest components (however, see Ying ef al., 19 main vegetational patterns in Shennongjia can be classified into at least seven associations: evergreen broad-leaved forest, mixed evergreen and decid- uous broad-leaved forest, dwarf bamboo thickets of Sinarundinaria nitida (FiGuRE 3), deciduous broad-leaved forest (FIGURE 4), pine forest (either Pinus armandii or P. massoniana), boreal hardwood forest (consisting of Betula spp., Sorbus spp., Acer spp., and scattered gymnosperms, especially Abies), and fir forests (comprised almost exclusively of Abies fargesil). In addition, there are the different plant communities present in the subalpine-alpine meadows, the fens, and the marshes, as well as on the numerous vertical cliffs (FIGURE 5) located throughout the district. The complex topography, the wide differences in climate, the geographic position, and the floristic history of the Shennongjia region have resulted in the development of a rich and varied flora (which includes a number of relict families, genera, and species) in the area. Based on Anonymous (1980) and additions to the flora collected during the 1980 Sino-American Botanical Expe- dition, the vascular flora comprises 2090 species in 786 genera and 166 families. Of these, 64 families and 307 genera are primarily represented in the tropics and subtropics, while 61 families and 304 genera belong to groups that are predominantly temperate in distribution. However, only 21 families and 98 genera are mostly restricted to the tropics, leaving subtropical and temperate groups as the major components of the flora. This breakdown by larger taxo- nomic groupings does not give a clear picture of the actual situation. Of the subtropical elements, most of the families and genera are represented by only one or a few species, while in the primarily temperate groups the genera and families are often represented by several to many species. Examples of these include Dryopteris (15 spp.), Polystichum (19 spp.), Salix (16 spp.), Betula (6 | /O km ( ONY at j 48 x Baicaoping Songbaizhen Shennongjia Forest sees collecting localities of the 1980 Sino-American Botanical Expedition. Numbers correspond to collection Map 2. localities (see APPENDIXES |, 2), localities are in vicinity of en names. how exact localities, numbers not associated with dots indicate collecting areas. Unnumbered collection [£861 NOLLIGSd Xd NVOISAWNYV-ONIS “TV LA MANOTOHLYUVEA 8 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Ficures 3, 4. Shennongjia Forest District: 3, Sinarundinaria nitida forming dense, impenetrable thickets at - 2650 m alt. at Banbiyon (rock formation characteristic of any areas of aac on); 4, Loyang River gorge near Pingqian at 1300 m alt. (forest on i es above river composed largely of Tsuga chinensis, ele Cercidiphyllum, Acer aaa, Davidia, Pterocarya paliurus, Fagus spp., and Malu. 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 9 spp.), Carpinus (9 spp.), Ranunculaceae (63 spp.), Sorbus (8 spp.), Cotoneaster (13 spp.), Acer (18 spp.), Rhododendron (14 spp.), and Viburnum (17 spp.). The flora of Shennongjia is more similar to those of eastern and northeastern China and Japan than to that of southwestern or southern China. Many genera extend from eastern Sichuan—western Hubei into eastern China and to Japan (Ying et al., 1979). Examples of plants with this distribution pattern are Pleu- rosoriopsis (Pleurosoriopsidaceae); Akebia (Lardizabalaceae); Deinanthe, Rodgersia, and Schizophragma (Saxifragaceae),; Paulownia (Scrophulariaceae); Weigela (Caprifoliaceae); Peracarpa (Campanulaceae; extending to the Hima- layas); Euptelea (Eupteleaceae); Cercidiphyllum (Cercidiphyllaceae), Hosta, Reineckea, Tricyrtis, and Cardiocrinum (Liliaceae); Epimedium (Berberida- ceae); and Aucuba and Helwingia (Cornaceae; both extending to the Himala- yas). In addition, many species in more widely ranging genera occur only in the region extending from central China to Japan A smaller number of genera are related to groups more characteristic of the Himalayan region or of regions more to the south or west. These include Holboellia and Decaisnea (Lardizabalaceae), Brandisia (Scrophulariaceae), Streptolirion (Commelinaceae; also extending to Korea), Tupistra (Liliaceae), and Gastrochilus (Orchidaceae). Another group of plants in the Shennongjia Forest District includes those belonging to genera showing disjunct distribution between eastern Asia and eastern North America. Examples are Matteuccia (Onocleaceae), Maclura (Moraceae), Caulophyllum and Diphyilleia (Berberidaceae), Buckleya (Santa- laceae), Antenoron (Polygonaceae), Liriodendron (Magnoliaceae), I/lictum (Il- liciaceae), Schisandra (Schisandraceae), Sassafras (Lauraceae), Stylophorum (Papaveraceae), Decumaria and Penthorum (Saxifragaceae), Pachysandra (Buxaceae), Acer sect. NEGUNDO (Aceraceae), Cryptotaenia (Umbelliferae), Lyonia and Vaccinium sect. HuGERtA (Ericaceae), Trachelospermum (Apocy- naceae), Catalpa (Bignoniaceae), Phryma (Phrymaceae), Triosteum (Caprifol- iaceae), and A/etris (Liliaceae). There are also a number of genera endemic to central, or central and south- western, China (Li, 1953) that are represented in the flora of Shennongjia. Some of these are Pteroceltis (Ulmaceae), Saruma (Aristolochiaceae), Tetracentron (Tetracentraceae), Asteropyrum (Ranunculaceae), Sargentodoxa (Sargentodox- aceae), Sinofranchetia (Lardizabalaceae), Eucommia (Eucommiaceae), Sino- wilsonia, Sycopsis, and Fortunearia (Hamamelidaceae), Dipteronia (Acera- ceae), Clematoclethra (Actinidiaceae), Davidia (Nyssaceae, or Davidiaceae), Dipelta and Kolkwitzia (Caprifoliaceae), Emmenopterys (Rubiaceae), and Ischnogyne (Orchidaceae). Continental glaciers did not encroach upon all of the Shennongjia region during the Quaternary (Li, 1940), and it is thought that this may have allowed a number of relict genera and families to survive there. However, it is also possible that many of these plants could have migrated southward during the Pleistocene, returning only after the climate warmed at the end of the last glaciation. 10 JOURNAL OF THE ARNOLD ARBORETUM IVOL. 64 “= — FiGure 5. Shennongjia Forest District: steep, nearly vertical cliffs (ca. 400 m alt. at base) close to southern boundary of Forest District near confluence of Jiuchong and Dangyang rivers. Slopes support rich broadleaved-evergreen mixed deciduous forest association (but note corn field on slope at middle left). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 11 MAN’S INFLUENCE ON THE REGION Since the Shennongjia Forest District contains some of the roughest terrain in Hubei Province, the vegetation has until recently remained unexploited. However, the demand for lumber and other forest products in the People’s Republic has increased to the extent that even remote areas of the District are currently being made accessible. Areas at lower elevations and forests on mod- erate slopes had been largely lumbered prior to the 1980 expedition (FIGURE 6). In 1980 roads were in place, or under construction, into high mountain regions and areas of extremely rough topography; large tracts of forested land, including areas on the steepest slopes, were being clear-cut. The timber is removed from these sites by sliding the logs down the steep slopes to collecting and loading areas, resulting in serious erosion. Cables suspended across valleys facilitate removal of logs from more inaccessible locations. From collection points the timber is transported by truck to larger collection depots, or “forest farms” (for example, our primary base camp was Jiuhuping Forest Farm; see Map 2), where some of the logs are cut into boards for local use. Most of the logs, however, are taken to the Changjiang (Yangtze) River for transport as giant rafts downstream to eastern China, where the timber is processed for use. Although our party was unable to visit all areas of the Shennongjia Forest District, and although none of us professes a thorough knowledge of forestry, we were nonetheless struck by the alarming rate at which both lumbering and preparation for future cutting were progressing. If cutting continues at the present rate, it might be roughly estimated that the forests of the district will be depleted of useful timber within a ten-year period. Of equal concern was the apparent waste evident in the lumbering process, the general disregard for the vegetation as a whole (through clear-cutting and the log slides), and the apparent lack of reforestation. In 1980 many of the areas that had been lum- bered were under cultivation in food crops, and it was incredible to observe corn “fields” on slopes at angles greater than 45°. The yield in these fields is very low: in many areas the stalks produce no ears or only a single one. It is obviously very expensive to remove timber from the district. If this is to remain as a major timber-producing region, the expansion of the human pop- ulation into the lumbered areas—especially those more readily accessible at low elevations, where the yields could potentially be greatest—should be strongly discouraged. The middle and higher elevations should remain unpopulated since the land there appears unsuitable for agriculture. Reforestation of lum- bered areas, both to prevent extensive erosion and to provide a more continuous yield of forest products, should be the first priority in the district. Altered lumbering procedures would also allow for natural reforestation with native species. The members of the 1980 expedition hope that man’s presence in the district will be limited, and that steps will be taken to preserve representative vegetational areas within the area before they are completely destroyed. 12 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 FiGure 6. Shennongyjia Pore Distach: Songlo Commune along the Songlo River at 1000 m alt. Lower- and middle ' denuded of native vegetation and planted in corn; small groves of Pinus massoniana “also planted for timber. Slender trees in front of commune, Cunninghamia lanceolata, with lateral branches pruned for firewood 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION | 13 PREVIOUS BOTANICAL EXPLORATION IN THE SHENNONGJIA FOREST DISTRICT AND LICHUAN XIAN So far as can be determined, the first botanical collections from within the region that now constitutes the Shennongjia Forest District were made by Augustine Henry, an Irish physician and botanist, during the period 1882- 1889, when he was stationed as an officer of the Chinese Imperial Maritime Customs Service in Yichang (Ichang). Some of Henry’s collections from the district were probably made by Chinese collectors in his employ, since his own collecting activities were initially concentrated within a ten- to fifteen- mile radius of Yichang (Bretschneider, 1898). However, in July of 1888, Henry made a trip to Baokang Xian (Paokang Hsien), Fang Xian, and Wushan Xian. In a letter to the director of the Royal Botanic Gardens, Kew, Henry recounted that on that journey he “‘. . . travelled due north from Ichang till I reached the range separating the basins of the Yangtze and Han Rivers, and then I made my way along the ranges westwards as far as Szechuan, striking then the Yangtze on the Hupeh-Szechuan boundary line I returned two or three days ago by boat down the rapids. ... I reached at one point about 10,000 feet altitude, and found the mountains from 8,000 to 10,000 feet rich in plants not previously sent. ... No foreigner, not even the Roman Catholic missionaries, had ever been in these parts before .. .” (Henry in Thiselton-Dyer, 1889, p. 226). From these excerpts of the letter as well as from some of the plants he mentioned, it is apparent that Henry had traveled across what is now part of the Shennongjia Forest District. The second botanist to visit the region was E. H. Wilson, an English plant explorer, who began to collect in western Hubei and eastern Sichuan provinces early in 1900. Altogether, Wilson made four expeditions in China, the first two in the employ of the Veitch nursery firm of Chelsea, England, and the latter two under the auspices of the Arnold Arboretum (see Howard, 1980). The first two years of Wilson’s initial expedition were spent exploring the mountains of western Hubei, and to judge from the map Wilson published of his itinerary (Wilson, 1905; see also Clausen & Hu, 1980), there is evidence that he also penetrated the Shennongjia region. Moreover, in his A Naturalist in Western China Wilson (1913) described in some detail a trip undertaken from Yichang to Daning (Taning) in Sichuan during the fourth expedition. Leaving Yichang on 4 June 1910 and traveling via Xingshan Xian, Wilson journeyed to Qing- tianpao (Chin-tien-po), Xiaolongtan (Hsao-lung-tan), Dalongtan (Ta-lung-tan), and Dajiuhu (Ta-chu-hu), reaching Daning in 22 days. All four of these inter- mediate locations are today within the Shennongjia Forest District. Although both Henry and Wilson undoubtedly prepared herbarium speci- mens of many of the plants they encountered while in the Shennongjia Forest District, the label data on their specimens are usually inadequate for pinpointing exact locations. As a result, it is usually impossible (particularly with Henry’s specimens) to determine if a particular collection was made in the Shennongjia region. The first Chinese botanists to collect in the Forest District were W. Y. Chun, 14 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 S. S. Chien, and R. C. Ching. These botanists left Yichang in July, 1922, and traveled via Wan-chao-shan (Wan-tsao-shan) in western Xingshan Xian to Xiaolongtan and Jiucaiyazi. Their two-month journey yielded more than 1000 numbers, but the most complete set was destroyed by fire in 1924 (R. C. Ching, pers. comm.). Some of the duplicate collections were distributed, however, and are in the herbaria of Nanjing University (N), the Institute of Botany, Beijing (PE), the Arnold Arboretum (A), and the United States National Herbarium (us). Likewise, a collection of nearly 3000 numbers made by Y. Chen in 1926 may include specimens from the Shennongjia Forest District, but since his field notes are missing it is impossible to determine exact localities for these spec- imens, which are deposited in the herbaria of Nanjing University (N) and Nanjing Technical College of Forest Products (NF). In July and August of 1943, a forest survey was undertaken in the Shennongyjia region and additional herbarium specimens were collected. It is of historical interest that C. Wang (T. Wang or Wang Zhan) participated in this survey and then undertook a side trip to Wan Xian (Wan Hsien) and Modaoqi (Mo-tao- chi), where he was the first botanist to collect specimens of Metasequoia glyp- tostroboides. Subsequent to the early activities in the Shennongjia region in the 1920's and early 1940’s, extensive botanical collections were made in the 1950’s and particularly during the period 1976-1978 by the Shennongjia Plant Resources Expeditions. The participants of these expeditions analyzed the vegetation and plant resources of the region and collected more than 10,000 numbers (Anon- ymous, 1980). These collections, although distributed widely in herbaria throughout China, are deposited primarily in the herbaria of the Wuhan Insti- tute of Botany (His) and the Institute of Botany, Beijing (PE). The history of botanical exploration in Lichuan Xian, where Metasequoia glyptostroboides was discovered as an extant plant, has been summarized by Hu (1980) in her paper concerning the flora associated with Metasequoia and is outlined briefly by Bartholomew, Boufford, and Spongberg (1983). Among the more recent Chinese collectors to visit this area are C. Wang, C. Y. Hstieh, C. T. Hwa, K. L. Chu, W. C. Cheng, and Y. W. Djou. Prior to the American scientists on the present expedition, the last ones to visit the region were R. W. Chaney and J. L. Gressitt, whose activities in the area were prompted by the discovery of Metasequoia. As far as can be determined, the American participants in the 1980 Sino- American Botanical Expedition to western Hubei were the first Western bot- anists to collect in the Shennongjia Forest District since the days of E. H. Wilson, and in the metasequoia area of Lichuan Xian since Chaney and Gres- sitt. Although the activities of Chinese botanists have greatly increased our knowledge of the floras of both regions (Anonymous, 1976, 1979, 1980), it 1s significant that 15 taxa new to science (13 new species, | variety, and | forma) are proposed as a result of the 1980 expedition, and that numerous taxa from both Shennongjia and the metasequoia region are recorded from these areas for the first time. Although it can safely be said that the flora of western Hubei is relatively well known, our knowledge still remains incomplete. 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 15 ACKNOWLEDGMENTS We would like to express our gratitude and thanks to the many people, too numerous to mention, who gave freely of their time and advice in the planning and execution of the expedition, as well as to those who helped to prepare this report. Primary responsibility for this paper is shared by D. E. Boufford and S. A. Spongberg, who either wrote or edited sections of the manuscript prepared by Chinese and American participants. In China S. C. Sun and Y. C. Tan rendered similar service. Our thanks are also expressed to C. Z. Ji, who skillfully prepared the illustrations of the new taxa of pteridophytes. The help of M. Byrnes in the preparation of the manuscript and of E. B. Schmidt in its editing is also gratefully acknowledged. For their help in the planning and execution of the expedition, we express our particular thanks to W. S. Xie, X. Y. Wang, and J. Cheng, of the Wuhan Branch, Academia Sinica; H. Dai, Hubei Foreign Affairs Office; S. Y. Wang, Institute of Botany, Academia Sinica, Beijing; Y. L. Xu, Forestry Office of Hubei Province; and especially R. X. Ma, chief of the Shennongjia Forest District, and Z. Z. Ran, vice director of the Wuhan Institute of Botany, who were responsible for our itinerary and logistic support. For their ready assistance in the field, K. Y. Guan, H. Wang, and S. Y. Cai are to be thanked. We are also indebted to Professors P. S. Tang, director of the Institute of Botany, Academia Sinica, Beijing, T. T. Yii, vice director, Institute of Botany, Academia Sinica, Beijing, S. C. Sun, director of the Wuhan Institute of Botany, Academia Sinica, and expedition leader, and P. H. Raven, director of the Missouri Botanical Garden, for their unfailing efforts to bring the 1980 Sino- American Expedition to a reality and a successful conclusion. To the staffs of Institute of Botany, Beijing, and the Wuhan Institute of Botany, the Chinese host institutions, the American participants wish to express their great appre- ciation. We would also like to thank all the support personnel who greatly facilitated our stay and contributed to the success of the expedition in the Shennongjia Forest District. Moreover, officials in Lichuan Xian, as well as the directors and staffs of universities, botanical gardens, and other botanical, forestry, and agricultural institutions that we visited after the completion of the fieldwork, are thanked for their help and kind hospitality. Major funding for the expedition was provided by the Academia Sinica, while a grant (#2 133-80) from the National Geographic Society to the American participants provided for transportation to and from China and for field equip- ment. Additional support to the American participants provided by members of the American Association of Botanical Gardens and Arboreta allowed for essential follow-up activities once the expedition had been completed. This financial support is gratefully acknowledged. AN ENUMERATION OF THE VASCULAR PLANTS COLLECTED ON THE 1980 SINO-AMERICAN BOTANICAL EXPEDITION TO WESTERN HUBEI The following annotated list represents the determinations of the vascular plants collected both in the Shennongjia Forest District (numbers /—1924) and 16 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 in the metasequoia region of Lichuan Xian (numbers /925—2085). Numbers between / and 2085 not represented in this list were assigned to nonvascular plants that will be detailed elsewhere. Names followed by an asterisk indicate taxa previously unreported for the Shennongjia Forest District, based on Shen- nongjia Plants (Anonymous, 1980). Determinations of the collections were undertaken by both American and Chinese participants of the expedition, and lists of determinations were exchanged for comparison as the work progressed. Discrepancies were evalu- ated in herbaria in both the People’s Republic of China and the United States until agreement was reached. In the few instances where we were unable to make satisfactory determinations, specific epithets have been omitted. Working at the Harvard University Herbaria (A and GH), D. E. Boufford and S. A. Spongberg took primary responsibility for the identifications of the expedition materials in the United States. T. R. Dudley provided determinations for [/ex and Viburnum and assisted with additional genera, while B. Bartholomew identified the specimens of Camellia. The assistance of L. Constance (Umbel- liferae), K. Flinck (Cotoneaster), L. Garay (Orchidaceae), I. Hay (Gymnosper- mae), T. Koyama (Juncaceae, Cyperaceae), E. Landolt (Lemna), H. W. Li (certain Labiatae), E. McClintock (Hydrangea), P. O’Connor (Gramineae), P. H. Raven (Epilobium), and B. G. Schubert (Begonia, Dioscorea, and certain Leguminosae) 1s also gratefully acknowledged. Primary responsibility for identifications in the People’s Republic of China was assumed by Z. Cheng, at the Wuhan Institute of Botany, and Y. C. Tang and T.S. Ying, Institute of Botany, Beijing. R. C. Ching and K. H. Shing kindly studied the pteridophytes, while T. T. Yii, L. T. Lu, and T. C. Ku examined the specimens of Rosaceae, R. H. Shan examined the Umbelliferae, and A. L. Chang studied the Ericaceae. Prior to the departure of the American participants from China, the her- barium material, as well as the germ-plasm collections, was divided equally between the Chinese and American teams. A complete set of the 1980 Sino- American Botanical Expedition specimens of vascular plants (1715 numbers) is at HIB, and the second most complete set (1695 numbers) is at A; specimens beyond these two sets were divided more or less equally between the institutions of the participants (cm, 1413 numbers; KUN, 1431; NA, 1441; NAs, 1466; Ny, 1433; pe, 1605; sFpH, 1309; uc, 1489; and wu, 1199). Specimens in addition to the sets deposited in the institutions of the American participants have been sent as gifts to the Royal Botanic Garden, Edinburgh (£, 872 numbers), the Department of Botany, Kyoto University (kyo, 454 numbers), and the Missouri Botanical Garden (mo, 173 numbers). Unfortunately, a portion (approximately 200 sheets) of the herbarium specimens alloted to the American team inexplica- bly disappeared after their safe arrival in the United States and after the deter- minations had been completed, resulting in less complete sets than would have otherwise been possible. In an attempt to make the following enumeration of greater value than a mere floristic listing, we have appended brief notes where we found taxonomic or nomenclatural problems during the identification process. It is hoped that these notes will stimulate further investigations that will ultimately help clarify 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 17 problems in the systematics of these central Chinese plants. Acronyms of the herbaria (following Holmgren, Keuken, & Schofield, 1981, with the exception of the Shennongjia Forest District Herbarium, abbreviated sFpH) where the specimens are deposited are given for each collection number. Also, literature citations of original descriptions are provided for names not appearing in /mdex Kewensis or its Supplements, or in Index Filicum or its Supplements. APPENDIXES 1 and 2 provide a cross-reference to collection numbers and exact collecting localities, the elevation of each site, and the collection date. The collecting localities are indicated on Map 2. PTERIDOPHYTA? LYCOPODIACEAE Huperzia crispata (Ching) Ching, Acta Bot. Yunnanica 3: 293. 1981. (syn.: Lycopodium crispatum Ching, Acta Phytotax. Sinica 18: 236. 1980.) 1974 (A, HIB). Huperzia hupehensis Ching*, Acta Bot. Yunnanica 3: 301. 1981. 1386 (A, HIB, PE, UC). Lycopodium obscurum L. 768 (A, HIB, NY, PE, UC). SELAGINELLACEAE Selaginella involvens (Sw.) Spring oe (A, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1360 (A, HIB, KUN, AS, NY, PE, SFDH, UC, WH). ae labordei Hieron. 218 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 330 (A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /080a (A, NA, UC); 1593 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /69/ (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Selaginella moellendorffii Hieron. 1080 (A, HIB, KUN, NA, NAS, PE, SFDH, WH). Collection no. /080 represents a mixed collection, part of which (no. 1080a) is Selaginella labordei Hieron. Some of the specimens cited here as S. moellendorffii in the Chinese herbaria may belong to that species. Selaginella nipponica Makino 845 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Selaginella uncinata (Desv.) Spring* 1117 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). EQUISETACEAE Equisetum arvense L. 1874 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Equisetum hyemale L. 141 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). he syst ti gement of the families of Pteridophyta follows Ching (1978). 18 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Equisetum ramosissimum Desf. 435 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). BOTRYCHIACEAE Botrychium strictum Underw 503 (HIB); 540 (HIB); 1686 es HIB, PE, UC). HYMENOPHYLLACEAE Hymenophyllum barbatum v. d. Bosch 2 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). DENNSTAEDTIACEAE Dennstaedtia pilosella (Hooker) Ching* (distributed as Dennstaedtia hirsuta Nae ee 54 (A, , NY, PE, SFDH, UC). ie ia ee Chr ris 563 es ee HIB, KUN, ie NY, PE, SFDH, UC, WH); 1760 (A, CM, HIB, KUN, KYO, FDH rae ae ‘Coutt ) c Chr. 2052 (A, HIB, NAS, PE, UC). HYPOLEPIDACEAE Hypolepis punctata (Thunb.) Mett. 943 (A, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1944 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). LINDSAEACEAE Stenoloma chusana (L.) Ching (syn.: Sphenomeris chusana (L.) Maxon) 2036 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). PTERIDACEAE Pteridium revolutum (Blume) Nakai 674 (A, CM, E, HIB, KUN, KYO, NAS, NY, PE, SFDH, UC, WH); 1/673 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Pteris excelsa Gaudich.* 1432 (A, CM, HIB, KUN, NAS, NY, PE, SFDH, UC); 1898 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Pteris nervosa Thunb. 495 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1654 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, WH). Pteris vittata L. os CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1437 (A , NA, NAS, NY, PE, SFDH, UC, WH) re ee ee rdh Z (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 19 SINOPTERIDACEAE Aleuritopteris duclouxii (Christ) Ching* (syn.: Cheilanthes duclouxii (Christ) Ching) 1674 (A, HIB, PE). Aleuritopteris farinosa (Forsk.) Fée* (syn.: Cheilanthes farinosa (Forsk.) Kaulf.) 1754 (A, HIB). Cheilsoria chusana (Hooker) Ching, comb. nov. BasionyM: Cheilanthes chusana Hooker, Sp. Fil. 2: 95. ¢. J06B. 1852. 1600 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Leptolepidium kuhnii (Milde) Hsing & S. K. Wu*, Acta Bot. Yunnanica 1(1): 117. 1979. (syn.: Cheilanthes kuhnii Milde) 1769 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Onychium ipii Ching* 1599 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Onychium japonicum (Thunb.) Kunze 2051 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Onychium moupinense Ching 545 (A, HIB). ADIANTACEAE Adiantum capillus-veneris L. 1915 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Adiantum davidii Franchet* 1357 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Adiantum edentulum Christ* 327 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, wH); 590 (A, CM, HIB, KUN, NA, NY, PE, SFDH, UC); 1767 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Adiantum erythrochlamys Diels* 326 (A, HIB, PE, UC); 497 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, wH); 1712 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Adiantum myriosorum Baker 552 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Adiantum pedatum L. 1295 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1755 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); /8/9 (A, HIB). HEMIONITIDACEAE (GYMNOGRAMMACEAE) Coniogramme caudiformis Ching & Shing*, Acta Bot. Yunnanica 3: 233. 1981. (distributed as Coniogramme caudifrons Ching) 496 (A, HIB, NA, NAS, PE, UC); 1435 (A, CM, HIB, NA, NAS, PE, UC). Conlogramme intermedia Hieron. 504 (A, HIB, NAS, PE, UC); 1446 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). 20 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 Coniogramme robusta Christ (distributed as var. repandula Ching, apparently an unpublished name) 049 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). ATHYRIACEAE Allantodia okudairai (Makino) Ching*, Acta Phytotax. Sinica 9: 49. 1964. 1711 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Allantodia squamifera (Mett.) Ching*, Acta Phytotax. Sinica 9: 55. 1964. 321 (A, CM, HIB, NAS, PE, UC); 1237 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Athyrium amplissimum Ching, Boufford, & Shing, sp. nov. Species proxime affinis Athyrium multifido Rosenst., Japoniae, sed in statura multo majore, stipite virescente (nec purpurascente), pinnis pinnu- lisive multo majoribus, et pinnis breviter petiolatis differt. Plants ca. 80 cm tall. Rhizome short, erect. Stipe ca. 35-40 cm long, 4 mm in diameter, stramineous, glabrous above base, which is covered with light brown, lanceolate scales; lamina ample, 40-45 cm long, to 40 cm wide, ovate, acuminate at apex, broadly rounded at base, tripinnate; pinnae ca. 13 pairs, lower ones nearly opposite but becoming alternate upward, rather closely spaced, sessile, oblong-lanceolate, acuminate to truncate base, sub- sessile, the basal pair nearly as large as those next above, ca. 20 cm long, 8- 9 cm wide above narrowed base, bipinnate; pinnules ca. 25 pairs, spreading rachilets glabrous. Sori large, subrounded to oblong, 3 to 5 per segment, close; indusium small, fugaceous. Type. China, western Hubei Province, Shennongjia Forest District (31°30'N, 110°30'E), vicinity of Dalongtan and Xiaolongtan on the W side of the road, elevation 2300-2600 m, 9 September 1980, Sino-Amer. Bot. Exped. 914 (holotype, PE; isotypes, A, CM, E, HIB, KUN, KYO, NA, NAS, NY, SFDH, UC, WH). ADDITIONAL SPECIMENS EXAMINED. China. Huser: Shennongjia Forest District (31° 30’N, 110°30’E), mixed deciduous-coniferous-bamboo forest remnant at Chuifeng Pass, ca. 3000 m, fronds tufted erect abate 1980 Sino-Amer. Bot. Exped. 49 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); vic. of Xiaojiuhu, ca. 2000 m, semiopen, moist slope in ise forest, fronds tufted from erect rhizomes, stipes ae or brownish, 1980 Sino-Amer. Bot. Exped. 1348 (A, CM, E, HIB, KUN, KYO, MO, NAS, NY, PE, SFDH, UC, WH). Somewhat similar to Athyrium multifidum Rosenst., but differing in being much larger and in having green (not purplish) stipes, much larger pinnae and pinnules, and pinnae with shorter petioles. 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 21 Athyrium epirachis (Christ) Ching 201] (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Athyrium fallaciosum Milde* 1768 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). er filix-femina Ne ) Roth.* I (A, CM, HIB, KUN, NA, NAS, PE, UC). Fae mackinnoli (Hope) C Chr.* 320 (A, CM, HIB, KUN , NY, es eH, UC, WH). Athyrium vidalii (ane yy Say. )N 333 (A, HIB, KUN, PE, UC); 2053 i CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Athyrium ee ae ra 542 (A, , KUN, Athyrium ae amanaied ¢ as ae deflecum Ching, apparently an unpublished name) 1876 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Dryoathyrium dielsii (C. Chr.) Ching* cs oe CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /09/ (A, CM, E, B, , NY, PE, SFDH, UC, ee ean a eee) Ching S588 (A, CM, HIB, KUN, NA, NAS, NY, PE, ce oe (HIB); 62/ (HIB); 1000 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, Dryoathyrium okuboanum (Makino) Ching, ie Phytotax. Sinica 10: 303. 706 (A, HIB, KUN, NA, NY, PE, UC). Dryoathyrium unifurcatum (Baker) Ching* 1434 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Lunathyrium shennongense Ching, Boufford, & Shing, sp. nov. Species insignis, adspectu Lunathyrium vermiformi Ching, Boufford, & Shing proxime affinis, sed in stipite rachidique subtus glabris supra in sulco pilis minutis septatis sparsis praedito, costis pinnarum subtus similiter sparse villosis, segmentis angustioribus, sinubus latis, indusiis brevioribus latior- ibusque subconchatis cinereis membranaceis marginibus erosis (indusiis in pinnis supremis plerumque hippocrepiformibus), differt. Plants to 70 cm tall. Stipe 12 cm long, 2.2 mm in diameter, pale-stra- mineous, bisulcate on upper side, scaly at base, glabrous upward; lamina oblanceolate, to 60 cm long, 15 cm wide at middle, acuminate at apex, gradually narrowed toward base, which is ca. 5 cm wide; pinnae ca. 22 pairs, patent, the lower ones separated by broad intervals, the basal pair 3-4 cm long, somewhat deflexed, the middle ones 10 cm long, !.8 cm wide at base, sessile or subsessile, lanceolate, acuminate at apex, truncate at base, pin- natipartite; segments ca. 22 pairs, spreading at right angles to costa, ca. 7 m long, 3 mm wide at base, oblong, obtuse at apex, entire; veins 6 or 7 pairs in each segment, simple. Fronds drying green, glabrous except for costa on undersurface of pinnae, which has few fine, septate hairs, these more 22 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 at yn Le SS RSE — \ uae _ a SAS : RAS = ery is Sere S FiGure 7. a-c, Lunathyrium vermiforme. a, frond (bar = 2 cm); b, portion of pinna Vy NY AA = nee ve aes - with sori (bar = 5 mm); c, setae from rachis (bar = | mm). d-g, Pyrrosia caudifrons: d, fronds and rhizome (bar = 2 cm); e, : _ from rhizome (bar = | mm); g, stellate hairs from undersurface of lamina (bar = m). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 23 dense on upper surface of rachis. Sor oblong, contiguous, 4 or 5 pairs per segment; indusium subconchoidal (those on uppermost pinnae generally horseshoe shaped), gray, membranous, margin erose. Type. China, western Hubei Province, Shennongjia Forest District (31°30'N, 110°30'E), vicinity of Chuifeng Pass, elevation ca. 2700 m; openings in bamboo thickets, fronds with erect rhizomes, 26 August 1980, Simo-Amer. Bot. Exped. 353 (holotype, PE; isotypes, A, CM, HIB, KUN, KYO, NA, NAS, NY, SFDH, UC, WH Although Lunathyrium shennongense resembles L. vermiforme Ching, Boufford, & Shing morphologically, it differs in having a stipe and rachis that are glabrous beneath and sparsely hairy with fine, septate hairs in the groove on the upper side (the costa on the undersurface of the pinnae also has a few similar hairs), segments that are narrower with broader sinuses, and short, broad, subconchoidal, membranous, ash-gray indusia with erose margins. Lunathyrium vermiforme Ching, Boufford, & Shing, sp. nov. FiGuRE 7, a. Species e turma Lunathyrium giraldii (Christ) Ching, sed in pinnis infer- ioribus minus abbreviatus, rachidibus omnino setis crassis vermiformibus copice praeditis, pinnis majoribus, differt. Plants to 74 cm tall. Rhizome s‘.ort, erect; fronds fasciculate. Stipe ca. 20 cm long, 2 mm in diameter, de «-stramineous, glabrous above base, which is covered with dark brown. ' .aceolate scales; lamina elliptic-lanceolate, 55 cm long, 18 cm wide at middle, acuminate at apex, gradually narrowed toward base, bipinnatipartite; pinnae ca. 20 pairs, patent, separated by nar- row intervals, the lower widely separated, the lowest 4 pairs abbreviated, the basal pair auriculate, ca. | cm long, the middle ones to 10 cm long, 1.6 cm wide at truncate base, lanceolate, slightly falcate, acuminate at apex, aerial to narrow wing along costa; segments ca. 20 pairs, close, oblong, y 4mm, entire, apex obliquely truncate or subrounded. Fronds her- ete drying green, pinnae glabrous on both sides, rachis covered through- out with thick, pale setae. Sori linear, extending from midrib to near margin, indusia linear, entire, straight throughout, well spaced. Type. China, western Hubei Province, metasequoia region of Lichuan Xian (30°10’N, 108°45’E), vicinity of Zhuanjiaowan on the E side of the valley, elevation ca. 1500 m, 7 October 1980, Sino-Amer. Bot. Exped. 2025 (holo- type, PE; isotypes, A, CM, HIB, KUN, KYO, NA, NAS, NY, SFDH, UC, WH). Lunathyrium vermiforme is related to L. giraldii (Christ) Ching of the same region but differs in having larger pinnae, with the lower ones less abbreviated, and the rachis covered throughout with numerous thick setae. Lunathyrium wilsonii (Christ) Ching, Acta Phytotax. Sinica 9: 71. 1964. 2048 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). 24 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Lunathyrium sp. (mistakenly identified and distributed as Lunathyrium centro- chinense Ching 1239 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Pseudocystopteris atkinsonii (Bedd.) Ching, Acta Phytotax. Sinica 9: 78. 1964. 322 (A, HIB, UC); 909 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1238 (A, HIB, UC). eat longipes (Christ) Ching*, Acta Phytotax. Sinica 9: 78. 1964. 6 (A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 873 (A, CM, HIB, KUN, KYO, NA, ve NY, PE, SFDH, UC, WH TTHELYPTERIDACEAE Cyclosorus acuminatus (Houtt.) Nakai (syn.: Thelypteris acuminata (Houtt.) Morton) 1104 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Macrothelypteris oligophlebia (Baker) Ching, Acta Phytotax. Sinica 8: 309. 1963, var. elegans (Koidz.) Ching, ibid. 1436 (A, HIB, NA, PE, UC). Parathelypteris nipponica (Franchet & Sav.) Ching, Acta Phytotax. Sinica 8: 302. 1963. (syn.: Thelypteris nipponica (Franchet & Sav.) ae g) 306 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, Phegopteris decursive-pinnata (Van Hall) Fée (sue Thelypteris ie -pin- nata (Van Hall) Ching) 246 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1433 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 2047 (A, CM, E, HIB, KUN, , NY, Baa. He oe ee Feo ae Thelypteris phegopteris (L.) Slosson ex Rydb.) 913 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, Pronephrium penangiana (Hooker) Holttum (syn.: ie penangiana (Hooker) Ching) 492 (A, CM, HIB, KUN, NAS, PE, UC); ///6 (A, HIB, NAS, NY, PE, UC). ge a is pyrrorachis (Kunze) Ching*, Acta Phytotax. a a ae os a A CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Thelypteris subocthodes (Ching) Ching* (distributed as Pseudocyclosorus tsoi Ching) 1946 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). ASPLENIACEAE Asplenium incisum Thunb. 1692 (HIB Asplenium sarelii Hooker 328 (A, HIB); 430 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Asplenium trichomanes L 0 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/448 (A, HIB, UC). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 25 No. 760 was distributed as Asplenium trichomanes L. subsp. orientale Lovis. Asplenium tripteropus Nakai 2046 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Asplenium unilaterale Lam. 1920 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). PLEUROSORIOPSIDACEAE Pleurosoriopsis makinoii (Maxim.) Fomin* 296 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). ONOCLEACEAE ee intermedia C. Chr. , HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). yee orientalis (Hooker) Trev. 720a (A, CM, NA, UC); 1890 (A, HIB); 2050 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Collection no. 720 originally represented a mixed collection and was divided into two parts, 720 and 720a. Other than for the specimen serving as the holotype of Matteuccia orientalis f. monstra (distributed as Matteuccia cen- tro-chinense, an unpublished name), the identity and disposition of these two numbers in Chinese herbaria are not known. However, specimens were deposited in HIB, KUN, NAS, SFDH, and wH under no. 720 Matteuccia orientalis (Hooker) Trev. forma monstra Ching & Shing, forma FIGURE 8. A forma typica differt pinnis lateralibus angustioribus integrisque infra partem apicalem plus minusve incisas. Sterile fronds ca. 80 cm tall. Stipe to 30 cm long, 6 mm in diameter, stramineous, densely covered at base with lanceolate, light brown, entire scales ca. 1.6 cm long, subglabrous upward; lamina oblong, ca. 50 by 34 cm, imparipinnate; pinnae ca. 20 pairs, obliquely patent, opposite, the lower ones separated by intervals ca. 2-3 cm wide, becoming closer upward, the basal pinnae as long as those above, to 17 by 1.5 cm, narrower (ca. 7 mm) toward base, subsessile, linear, acuminate, margins crenate, the upper pinnae similar but with apical portion pinnatifid with segments ca. 6 mm long; veins in all segments simply forked. Fronds glabrous, drying brownish green. The fertile fronds ca. 60 cm tall; stipe 35 cm long, dark-stramineous, glabrous above base; lamina 28 cm long; fertile pinnae approximate, sessile, linear, dark brown, glossy, leathery, distichous, 6 by 4 mm, reflexed margin persistently incurved, indusium brownish, fimbriate. Type. China, western Hubei Province, Shennongjia Forest District (31°30'N, 110°30’W), vicinity of Muyuping Forest Brigade on the SE side of the watershed divide between the Changjiang (Yangtze) and Hanjiang rivers near km 73 from Xingshan Xian, elevation 1450-1600 m; cutover Fagus-Quercus-Bet- 26 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 SRI NK ~ li a Ficure 8. Matteuccia orientalis f. monstra: a, sterile frond (bar = 2 cm); b, portion of sterile pinna (bar = 5 mm); c, apical portion of fertile pinna (bar = 5 mm), 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 27 ula forest on steep, moist slopes, SE exposure; fronds tufted from erect rhizome, 7 September 1980, Sino-Amer. Bot. Exped. 720 (holotype, PE; isotypes, A, NY). A peculiar form differing from all other known members of the genus in having sterile fronds with narrowly linear, crenate pinnae (the upper ones pinnatifid apically) and simply forked veins. This may be an abnormal form of Matteuccia orientalis (Hooker) Trev.; further field observations are need- ed. Matteuccia struthiopteris (L.) Todaro* 999 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). WOODSIACEAE Woodsia polystichoides D. C. Eaton 853 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). BLECHNACEAE Struthiopteris eburnea coe Ching* (syn.: Blechnum eburneum Christ) 1896 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Woodwardia unigemmata (Makino) Nakai 486 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 2057 (A, HIB, UC). PERANEMACEAE Peranema cyathioides D. Don 2085 (A, HIB). DRYOPTERIDACEAE Cyrtomium fortunei J. Sm. 238 (A, HIB); 1500 os CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1569 (A, CM, E, HIB, KUN, MO, NA, NAS, NY, PE, SFDH, UC, WH). Cyrtomium Sane erie Tagawa 238a (A); 1092 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Dryopteris apicifixa Ching, Boufford, & Shing, sp. nov. Species proxime affinis Dryopteris juxtaposita Christ, sed in statura multo minore, rachi paleacea, pinnis pinnulisque multo brevioribus, soris ad pinnis supremas restrictis, differt. Plants ca. 50 cm tall. Rhizome short, erect; fronds tufted. Stipe ca. 23 cm ng, 1.2 mm in diameter, stramineous, densely covered at base with broadly lanceolate, thin, light brown scales; lamina oblong, 28 cm long, 15 cm wide at base, acuminate at apex, bipinnate; pinnae pinnate, ca. 6 pairs, opposite, short-petiolate, patent, the basal pair slightly larger, 9 by 4 cm, oblong, acuminate at apex, rounded at base, the upper ones similar in outline but gradually smaller than basal pair; pinnules ca. 10 pairs, closely spaced, bas- 28 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 iscopic ones longer than acroscopic, the basal one not abbreviated, as large as those upward, 2-3 cm long, | cm wide at base, sessile, lanceolate, acute at apex, pinnatifid with few rounded lobes, the acroscopic pinnules 1.5 cm long, 8 mm wide at base, oblong, blunt at apex, margins crenate. Fronds drying light green, glabrous except for rachis, which is covered with rather sparse, lanceolate, dark brown scales. Sori small, confined to apical part of lamina, 2 or 3 pairs per pinnule; indusium light brown, rather thin, curling, deciduous. Type. China, western Hubei Province, Shennongjia Forest District (3 1°30'N, 110°30’E), along the trail between Hongriwan construction camp and Qui jiaping, elevation 1200-1400 m; growing in bamboo thicket, 2 Sepicniber 1980, Sino-Amer. Bot. Exped. 543 (holotype, PE; isotypes, A, CM, HIB, UC). Dryopteris apicifixa is related to D. juxtaposita Christ but differs in its much smaller size, its scaly rachis, its much shorter pinnae, and its pinnules with the sori confined to the apical part of the lamina. Dryopteris bissetiana (Baker) C. Chr.* &49 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Dryopteris handelii C. Chr.* 1293 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Dryopteris infrapuberula Ching, Boufford, & Shing, sp. nov. FIGURE 9. Species insignis habitu Dryopteris yunnanensis Christ e provincia Yunnan, sed in stipite (e basi sursum) omnino glabro, pinnis ad basin conspicue dilatatis, venis in segmentis omnibus simplicibus, et pinnis subtus praecipue secus venas sparse puberulis, facile distinguitur. Plants ca. 80 cm tall; growing tip covered with dark brown, broadly lan- ceolate scales; fronds tufted. Stipe ca. 23 cm long, 3 mm in diameter, dark straw colored, nearly glabrous from base upward; lamina elliptic-oblong, 40 cm long, 23 cm wide at middle, acuminate at apex, slightly narrowed toward base, pinnate-pinnatilobed; pinnae ca. 23 pairs, patent, separated by narrow intervals, the basal pair somewhat abbreviated, to 7 cm long, horizontally patent, the middle ones to 13 cm long, 2 cm wide at base, sessile, lanceolate, acuminate at apex, dilated at base, lobate-pinnatifid, with lobes ca. 3 by 5 mm, roundish, entire; veins 5 pairs in each lobe, simple, anterior basal vein reaching sinus, posterior one reaching margin slightly above sinus. Fronds chartaceous, drying green, the rachis copiously covered with dark brown, linear, appressed scales, the undersurfaces of costae covered with small, lanceolate, brown scales with ciliate margins, the midribs and veins with few minute hairs. Sori rather small, 3 or 4 pairs per segment, medial; indusium dark brown, firm, ultimately deciduous. Type. China, western Hubei Province, Shennongjia Forest District (31°30'N, 110°30’E), vicinity of Quijiaping, elevation 1440-1650 m; moist, level areas along stream in mixed deciduous—broad-leaved evergreen—coniferous forest, 3 September 1980, Sino-Amer. Bot. Exped. 619 (holotype, PE; isotypes, A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 29 1983] Dryopteris infrapuberula: a, frond (bar = 2 cm); b, portion of pinna with FiGure 9. sori (bar = 5 mm); c, scale from undersurface of rachis (bar = | mm); d, scale from undersurface of costa (bar = 1 mm). 30 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 Dryopteris infrapuberula is an outstanding species of the group containing D. yunnanensis Christ, from Yunnan Province, but is easily distinguished by the dark straw colored stipe naked from the base upward, the pinnae with broadened bases, the veins in the segments all simple, and the undersurfaces of the pinnae (especially along the veins) sparsely puberulous. Dryopteris labordei (Christ) C. Chr. 2054 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Dryopteris nemagetae (Kurata) Kurata* 1292 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Dryopteris canals hued 759 : : , NAS, NY, PE, UC). vais pe (Diels) C. Chr. 31] (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Dryopteris submarginalis Ching, Boufford, & Shing, sp. nov. FiGure 10. Species ex affinitate Dryopteris pulcherrimae Ching, sed in statura multo majore (planta usque ad 78 cm alta), lamina latiore, pinnis longioribus, costis subtus distincte canaliculatis, et soris submarginalibus, recedit Plants to 78 cm tall; rhizome short, erect; fronds fasciculate. Stipe short, to 10 cm long, 5 mm in diameter, densely covered with dark brown, lan- ceolate scales; lamina to 68 cm long, 16 cm wide at middle, oblanceolate, acuminate at apex, gradually narrowed toward base, base ca. 8 cm wide, bipinnatipartite; pinnae numerous, ca. 38 pairs, patent, sessile, separated by narrow intervals, lower ones more widely separated, the basal pair 4 cm long, the middle ones to 8 cm long, 1.5 cm wide at base, lanceolate, acuminate at apex, truncate at base, pinnatipartite to narrow wing along each side of costa; segments ca. 22 pairs, patent, separated by narrow sinuses, oblong, ca. 6 by 2 mm, entire; veins 5 or 6 pairs in each segment, simple, oblique. Fronds herbaceous, drying green, costa canaliculate beneath, copiously cov- ered with brown, ovate, membranaceous scales, glabrous above except for some brown, deciduous fibrils along lower part of costa; rachis densely cov- ered beneath with ovate-acuminate, fimbriate scales, densely fibrillose above. Sori rather small, 3 to 5 pairs per segment, submarginal; indusium light brown, firm, ultimately curling, usually persistent. Type. China, western Hubei Province, Shennongjia Forest District (31°30'N, 110°30’E), S end of the Loyang River gorge near Pingqian, elevation ca. 1300 m; deciduous forest on steep slope along the river, 14 September 1980, Sino- Amer. Bot. Exped. 1356 (holotype, PE; isotypes, A, CM, HIB, KUN, KYO, NA, NAS, NY, SFDH, UC, WH). Dryopteris submarginalis is related to D. pulcherrima Ching of the same region, but differs in its much larger size, its broader lamina and longer pinnae, and its costae that are distinctly canaliculate beneath. Dryopteris supraimpressa Ching, Boufford, & Shing, sp. nov. Species affinis Dryopteris marginatae (Hope) Christ, sed in statura multo minore, lamina basi tripinnatifida nec tripinnata, pinnulis brevioribus api- 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 31 NON FH 7A { IRE s iN aN ft a Vik N \, Y 7] f ( p) . aA “Os od {a rt : wd 9/4 ’ \ Soe = v oe oS ai$ ey, fas Pi a yy, < <<8036- . a fF On I SS ». < SH RS ASAE ey} 7 of Noo . = RAFAL NTN ~ A ¢ f ga 58 fp / SS Rory SS ONGLER OOF: / w Je ). Ci, . y A > NGO SAF q.. ¥ 3 = c es, o Ny . VAS g BQ r = 5 Wy o Bog fotos) = . Bh IN RY 58 ~ > OF 7a SSF ‘ y , 4, Pyliy’ /4) o | Tt /, IOUT); Sy HS 19 0g SOS Whb op Legg, POD Cas 5 a ANY /_ SS a fil He < 33 Oe AMES EES BO ag eee Zoid yi NUL a OS” 2 cm); b, pinnule with sori Polystichum shennongense: a, frond (bar : = 5 mm); c, d, scales from stipe (bar = 5 mm); e, scales from rachis (bar = 5 mm). FiGure ILI. (bar 36 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Polystichum tsus-simense (Hooker) Moore 406 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /062 (A, CM, HIB, KUN, NAS, NY, PE, UC); 1653 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). POLY PODIACEAE Arthromeris cuneata Ching 1971 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 2029 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Drymotaenium miyoshianum @iakino) Makino 2058 (HI ae tuners drymoglossoides (Baker) Ching (syn.: Lemmaphyllum dry- Soe eae ie , CM, HIB UC). BAe anne ae Ching Fl. Tsinlingensis 2: 232. 1974. ‘M, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1363 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, Lepisorus asterolepis (Baker) Ching*, ie Zhi Wu Zhi (Fl. Jiangsuensis) 1: 74. fig. 112. 1977. 1163 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Lepisorus contortus (Christ) Ching A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 358 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Lepisorus distans (Tagawa) oe ees ities Sinica 10: 302. 1965. 1297 CM, HIB, KUN, KYO, NA, NA DH, ae ee Lepisorus ‘marginatus Ching, a soars - 333, 15 (A, HIB, NAS, NY, PE, UC); 6/ (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 5/5 (A, CM, HIB, KUN, NA, NAS, PE, UC Lepisorus paohuashanensis Ching*, Jiangsu Zhi Wu Zhi (Fl. Jiangsuensis) 1: 467. fig. 113. 1977. 62 (A, HIB, PE, UC). Lepisorus unre (Kaulf.) Ching 1817 Microsorium ae (Moore) Chin 26 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 556 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/430 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). vo subhastatum (Baker) Ching (distributed as Lepidomicrosorium subhastatum (Baker ee g) 2083 (A, CM, HIB, NY, UC). Neolepisorus ovatus oer ‘ ‘Ching, Acta Phytotax. Sinica 9: 99. 1964, forma deltoidea (Baker) Ching, ibid. 247 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/43] (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH) Polypodioides pseudo-amoenum (Ching) Ching, comb. nov. BasionyM: Polypodium pseudo-amoenum Ching, Fl. Tsinlingensis 2: 177. 974 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 37 332 (A, CM, HIB, KUN, NAS, PE, UC); 356 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH Pyrrosia assimilis (Baker) Ching* 805 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Pyrrosia caudifrons Ching, Boufford, & Shing, sp. nov. FIGURE 7, d-g. Species configuratione proxime affinis Pyrrosia linguae (Thunb.) Farw., sed in fronde ambitu elliptica ad apicem caudata, stipitis longitudine alli- quanto superante, textura tenuiore, differt, necnon P. heteractae (Mett.) Ching valde similis, sed statura minore, pilis stellatis solum e brachuis lanceolatis constitutis, differt. Plants 20-22 cm tall. Rhizome wide-creeping, ca. 2 mm in diameter, densely covered with lanceolate, brown scales with ciliate margins. Fronds distant, 4-5 cm apart. Stipe 10-12 cm long, firm, dark-stramineous, glabrous above base; lamina 13-15 cm long, 3-4 cm wide at middle, elliptic, caudate at apex, cuneate at base, subcoriaceous, buff colored, glabrous above, covered beneath with thin tomentum of |-layered, uniform, appressed stellate hairs consisting only of lanceolate arms, the lateral veins distinct to leaf margin, obliquely ascending. Sori subrounded or oblong, dense, in close rows of 4 or 5 between lateral veins, light colored when young, dark red at maturity. Type. China, western Hubei Province, Shennongjia Forest District (31°30'N, 110°30’E), vicinity of Duanjiangping, 11 September 1980, Sino-Amer. Bot. Exped. 1159 (holotype, PE; isotypes, A, CM, E, HIB, KUN, KYO, NA, NAS, NY, SFDH, UC, WH) pec naa SPECIMENS EXAMINED. China. Huser: Hefeng, H. J. Li 6840, 8393 (PE). SICHUAN: Emei Shan (Mt. Omei), K. H. Shing & K. Y. Long 519, 525, 1163 (PE); Ses K. J. Kuan et al. 2330 (re); Leis, Z. T. Kuan 8426 (pe); Daxiang Ling, S. S. Kung 3691 (PE). Pyrrosia caudifrons is similar to P. lingua (Thunb.) Farw. in habit, but differs in the elliptic outline, caudate apices, and thinness of the fronds, which are slightly longer than the stipe. Pyrrosia caudifrons is also very similar to P. heteracta (Mett.) Ching in habit, but differs mainly from this species in its smaller size and in the stellate hairs on the undersurface of the fronds consisting of only lanceolate arms. The stellate hairs in P. heteracta have both lanceolate and aciculate arms. Pyrrosia drakeana (Franchet) Ching 18 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 519 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 18/6 (A, HIB). Pyrrosia gralla (Giesenh.) Ching* 1714 (A, HIB, PE, UC). Pyrrosia lingua (Thunb.) Farw. 1929 (A, CM, HIB, NA, NAS, PE, UC). Pyrrosia mollis (Kunze) Ching 547 (A, HIB, UC). 38 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Pyrrosia petiolosa (Christ) Ching 38 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Pyrrosia pseudocalvata Ching, Boufford, & Shing, sp. nov. Species (critica ex affinitate) Pyrrosia calvatae (Baker) Ching valde simile (cul ambitu congruit), sed in pilis stellatis deciduis e branchiis aciculatis brunneis, aliter glabra, recedit. Plants 40-67 cm tall. Rhizome short, thick, procumbent, ca. 6 mm in diameter, densely covered with lanceolate, brown scales with ciliate margins. Fronds proximate. Stipe 8-13 cm long, greenish stramineous, glabrous above base; lamina 30-50 cm long, 2.5—3 cm wide at middle, lanceolate, acuminate at apex, gradually narrowed at base, decurrent, coriaceous, both sides yellow- green, glabrous above, sparsely covered beneath with single layer of decid- uous, stellate hairs with aciculate arms; veins indistinct. Sori subrounded, densely covering ne from middle to apex, yellow-green when young, brown at maturi Type. China, western Hubei Province, Xingshan Xian (31°05'N, 110°30’E), ca. | km W of the Leigutai Hydroelectric Power Station at the confluence of the Jiuchong and Dangyang rivers, elevation ca. 390 m; steep rocky slope, 5 September 1980, Sino-Amer. Bot. Exped. 1110 (holotype, PE; isotypes, A, CM, HIB, KUN, NA, NAS, NY, UC). ADDITIONAL SPECIMENS EXAMINED. China. GANsu: Wen Xian, 2 He mae (PE). GuizHou: Xingyi, An-Shun Exped. 640 (Pe); Qingzhen, Sichuan-Guizhou Ex, 66 (PE); Zunyi, Sichuan-Guizhou Exped. 1405 (pr). HUBEL Sheanongia ee Dee (31°30'N, 110°30’E), Honghua ravine on W side of Dangyang R. N of Honghua, elev. ca. 820 m, growing on rock in forest, /980 Sino-Amer. Bot. Exped. 1429 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); Shennongjia Forest District, vic. of Housh- anping on § side of Hou R., elev. 800-1050 m, 1980 Sino-Amer. Bot. Exped. 1663 (A, CM, HIB, NA, NAS, NY, PE); Badong, H. C. Chow 957 (pe); Enshi, H. C. Chow 1940 : i), . Shing 1083 (PE); Tivenkou. T. L. Dai 100121, 102488, 104590, 107167, 107796 (PE). Because of the similarity in general outline of the fronds, Pyrrosia pseu- docalvata has been considered identical to P. ca/lvata (Baker) Ching. It differs in having only a single layer of sparse, deciduous, stellate hairs with aciculate arms instead of a double layer, of which the ground layer consists of stellate hairs with usually persistent, wooly arms. The geographic distribution of the species also differs: P. ca/vata occurs in Guangdong, Guangxi, Yunnan, Fujian, and southern Zhejiang provinces, and the two meet only in Guizhou Prov- ince. Pyrrosia sheareri (Baker) Ching 1925 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 2055 (A, CM, HIB, KUN, NA, NAS, NY, SUG), Saxiglossum Rica oni (Giesenh.) Ching, Acta Phytotax. Sinica 10: 301. 428 (A, CM, HIB, KUN, NA, NAS, PE, UC); 697 (A, CM, HIB, KUN, NA, NAS, NY, 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 39 PE, SFDH, UC, WH); 852 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH LOxOGRAMMACEAE Loxogramme grammitoides (Baker) C. Chr.* 546 (A, HIB). Loxogramme saziran Tagawa* 427 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 5/4 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1294 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). AZOLLACEAE Azolla imbricata (Roxb.) Nakai 2 A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). GY MNOSPERMAE TAXACEAE Amentotaxus argotaenia (Hance) Pilger 533 (A, HIB, NA). Taxus chinensis (Pilger) Rehder 585 (A, HIB); 777 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC Hs ae CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1524 A, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). ae yo ee Pancha 584 (A, HIB); 625 (HIB). CEPHALOTAXACEAE Cephalotaxus fortunei Hooker f. 751 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1038 (A, CM, E, HIB, KUN, NAS, NY, PE, SFDH, UC, Cephalotaxus sinensis (Rehder z Wilson) H. hi 104 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1759 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1829 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). PINACEAE are ee Franchet 958 (A , E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). See eae (Bertr.) Beiss. 24 , E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Picea Dikons Mast. : 6 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1823 (A, CM, E, HIB, UN, KYO, , NAS, NY, PE, SFDH, UC, WH). Be pers ee 40 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 384 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC); 877 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Pinus massoniana male 72 , CM, E, HIB, NY, PE, SFDH, UC, WH). Pinna (amb var. ae ase (Mast.) Wu, Acta oie Sinica 5: 153. 1956. 1466 (A, CM, E, HIB YO, MO, NA, NAS, NY, PE, SFDH, UC, WH). oe ee Pee (Franchet) Priteel 5 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1827 (A CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). (CCUPRESSACEAE Cupressus funebris Endl. 1914 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). TAXODIACEAE Metasequoia glyptostroboides Hu & Chen 2082 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). ANGIOSPERMAE Dicotyledones SAURURACEAE Houttuynia cordata Thunb. § (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). CHLORANTHACEAE Chloranthus multistachys Péi 456 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1699 (A, CM, HIB KUN, NA, PE) SALICACEAE Populus davidiana Dode 1388 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Populus poner Oliver* 1343 (A, HIB). Populus wilsonii Schneider 742 (A, HIB). Salix fargesii Burkill 15 (A, HIB). Salix fargesti Burkill var. kansuensis (Hao) N. Chao, Feddes Repert. Spec. Nov. Regni Veg. Beih. 93: 61. 1936. 235 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 41 JUGLANDACEAE Juglans cathayensis Dode 450 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, WH). Platycarya strobilacea Sieb. & Zucc. 1 M, HIB, KUN, NA, NAS, PE). Pterocarya delavayi Franchet* Is M, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Pterocarya cas Bes, 940 (A, CM, E, HIB, PE, SFDH, WH). Pterocarya polilns Bs ce C ieee peas (Bavalin) inka (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, W Pterocarya stenoptera D 1573 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). BETULACEAE Betula albo- — Burkill 31 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Betula cf. wibo- sinensis Burkill 897 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). This collection may ee a a hybrid between oe albo-sinensis and B. utilis D. Don, another species credited to the Shennongjia region (Anon- ymous, 1980) but not collected by the 1980 Sino-American Botanical Expe- dition. Betula fargesii Franchet 4 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Betula insignis Franchet I CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Betula luminifera H. Winkler 734 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Carpinus chinensis (Franchet) Cheng (syn.: C. cordata Blume var. chinensis Franchet, J. Bot. 13: 202. 1899 1528 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Carpinus fargesti Franchet 1011 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Carpinus hupeana 1548 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Carpinus simplicidentata H. H. Hu 9 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1486 (A, CM, E, HIB, KUN, NA, , NY, PE, SFDH, UC, WH). Corylus ferox Wallich var. tibetica (Batalin) Franchet, J. Bot. 13: 200. 1899. 12 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 346 (A, HIB, NA, PE); 383 (A, CM, HIB, PE); 1048 (A, HIB, NAS, NY, PE, UC). Corylus heterophylla Fischer ex Trautv. var. sutchuenensis Franchet, J. Bot. 13: | 214 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 613 (A, CM, 42 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /255 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /882 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Corylus mandshurica Maxim. & Rupr.* (syn.: C. sieboldiana Blume var. mand- shurica (Maxim. & Rupr.) Schneider, Pl. Wilsonianae 2: 454. 1916.) 1535 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). An interesting species with a discontinuous ars from Sichuan (and now western Hubei) in central China, to northeastern China and Korea, and the northern Japanese island of Hokkaido. (See eee Pl. Wilsonianae 2: 454. 1916, for additional details of aa tion.) Corylus saat (Franchet) A. Cam 423 , CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). FAGACEAE Castanea henryi (Skan) Rehder & Wilson 1852 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). mer mollissima Blume 46 , CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 798 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, oer UC, WH); 1427 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1884 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 2034 (A, HIB). Fagus engleriana Seem. 1221 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1309 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Fagus longipetiolata Seem. 1479 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). Fagus lucida ees . Spee 1485 (A, CM, HIB, UC). Lithocarpus teistcarps Seem ) Rehder he Wilson 7 A, CM, E, , SFDH, UC, WH). oie polstachns ae ex aes ) Rehder* A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Pe acutidentata (ia ) Koidz. 378 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 743 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); /858 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Quercus engleriana Seem. S77 (A , UC) Quercus glandulifera Blume 1193 (A, HIB, ie hues he CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1931 (A, Cc NA, Y, PE, SFDH, Quercus glandulifera ie var. re (A. “DC. y Nakai, J. Arnold Arbor. 5: 76. Ve 8 (A, CM, E, HIB, KUN , PE, SFDH, UC, WH). Quercus glauca Thunb. var’ ek (Rehder & Wilson) A. ‘Conrus: Les Chénes 1: 285. 1938. 67 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 43 In Fl. Hupehensis 1: 116. 1976, this taxon is elevated to the rank of species, and a new combination— Quercus gracilis (Rehder & Wilson) Wuzhi—is made. This name is invalid as a later homonym of Quercus gracilis Korth. Verh. Nat. Gesch. Bot. 207. 1844 Quercus myrsinifolia Blume 729 (A, CM, E, HIB, _KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). cpl oxyodon M oe CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC); 1065 (A, CM, HIB NAS, PE, SFDH, UC, WH). nee SHU Bae ex Franchet 371 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); /3/0 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ULMACEAE Celtis biondii Pampan. 432 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 789 (A, CM, E, HIB, KUN, NA , NY, ie SFDH, UC, WH). Celtis vandervoetiana Se 1490 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Zelkova sinica Schneider 1706 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). MORACEAE seg ie papyrifera (L.) L>Hér. ex Vent. oe CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1585 (A, CM, HIB, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Ficus ee Wallich var. henryi King ex Oliver, Hooker’s Icon. Pl. 19: pl. 1824. 1889. (syn.: F. sarmentosa Buch.-Ham. ex J. E. Smith var. henryi (King) Corner (Gard. Bull. Singapore 18: 6. 1960.) in Fl. Hupehensis 1: 152. 1976.) 1603 (A, CM, HIB, PE, UC). Ficus heteromorpha Hemsley 550 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1074 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 145] (A, CM, E, HIB, KUN, , PE, SFDH, UC, WH). Ficus en on ae syn.: F. sarmentosa Buch.-Ham. ex J. E. Smith var. impressa (Champ.) Corner, Gard. Bull. Singapore 18: 6. 1960.) 1492 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). CANNABINACEAE Cannabis sativa L. 65la (pistillate) (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 65/b (staminate) (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). umulus scandens (Lour.) Merr. (syn.: H. japonicus Sieb. & Zucc.) 1639 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 44 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 URTICACEAE Boehmeria diffusa i 482 , HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Boehmeria peer vel WL Wr ae A, CM, E, HIB, KUN NAS, ae PE, SFDH, UC, WH). oe ichangense H. nee 1919 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Elatostema stewardii Merr. 524 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Gonostegia hirta (Blume) Miq. 1937 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lecanthus peduncularis (Royle) Wedd.* 49 3 , CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Oreocnide frutescens (Thunb.) Migq. (syn.: O. fruticosa (Gaudich.) Hand.-Mazz.) 1638 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Parietaria micrantha Ledeb. (syn.: P. debilis Forster var. micrantha (Ledeb.) Wedd. in DC. Prodr. 16(1): 235*. 1869.) 1833 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Pilea japonica (Maxim.) Hand.-Mazz. (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Pilea ies (Lévl.) Hand.-Mazz. , CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). is ee C. H. Wright 113 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1660 (A, HIB, KUN, NA, NAS, NY, PE, UC). Pilea sinofasciata C. J. Chen & B. Bartholomew, nom. nov., based on Pilea fasciata Franchet, Nouv. Arch. Mus. Hist. Nat. II. 10: 81. 1888, non Wedd. in DC. Prodr. 16(1): 120. 1869 62 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). LORANTHACEAE Loranthus levinei Merr. 1324 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ARISTOLOCHIACEAE Aristolochia heterophylla Hemsley 112] (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Saruma henryi Oliver 1592 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). BALANOPHORACEAE Balanophora involucrata Hooker f. 889 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1260 (HIB). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 45 Balanophora japonica Makino 458 (1B); 890 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). POLYGONACEAE Antenoron neofiliforme (Nakai) Hara 42? (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 764 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Polygonum amplexicaule D. Don var. sinense Forbes & Hemsley, J. Linn. Soc., Bot. 26: 333. 1891. ee ee CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 50/ (A, CM, E, HIB, KUN, , NY, PE, SFDH, UC, WH). pe a ee Blum 1347 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1956 (A, CM, E, HIB, N, NA, NAS, NY, PE, SFDH, UC, WH 57 (A, HIB, NA, PE). Polygonum niliinerve (Nakai) Ohwi Gane P. matiflorum Thunb. var. cillinerve (Nakai) A. N. Steward, Contr. Gray Herb. 88: 97. 1930.) 307 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Polygonum cuspidatum Sieb. & Zucc. 437 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 724 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1578 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1626 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, Polgonim hydropiper L. 954 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Peete multiflorum Thunb 44() (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1657 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, Polygonum muricatum Meisner nee P. strigosum R. Br. var. muricatum (Meisner) A. N. Steward, Contr. Gray Herb. 88: 89. 1930. 1955 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 20/3 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Polygonum persicaria L.* 1521 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 20/5 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ea pilosum (Maxim.) Forbes & Hemsley 41 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Palygonin thunbergii Sieb. & Zucc. 953 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). AMARANTHACEAE Achyranthes aspera L. 62 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /50/ (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Achyranthes bidentata Blume 49] (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 46 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Celosia argentea L. 1918 (A, HIB, NA, NAS, NY, PE). PHYTOLACCACEAE — aks ago eNOS: NY, PE, SFDH, UC, WH). ree a Cae Batalio 723 (A, HIB, KUN, NAS, NY, PE, SFDH, WH). Nowicke (Ann. Missouri Bot. Gard. a 294-364. 1968) did not see spec- imens of Phytolacca polyandra Batalin and therefore did not treat this species in her study of the Phytolaccaceae. Although our specimens have five (not eight) united carpels and sterile (not fertile) stamens, they agree most closely with the original description of that species. Nowicke admitted that her treatment of Phytolacca was not exhaustive and stated that the genus is in need of further, critical study CARYOPHYLLACEAE Cucubalus baccifera L. 409 (A, HIB, PE, UC). Dianthus superbus L. 144 (HIB); 945 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Silene fortunei Vis. 780 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1409 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 19/7 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Silene linearifolia Pampan. 973 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Silene tatarinowli Regel (syn.: Melandrium tatarinowli (Regel) Y. W. Tsui) 71 (A, HIB, KUN, NA, NAS, NY, PE, UC); 386 (A, CM, HIB, KUN, NA, NAS, PE, UC); 1859 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). TETRACENTRACEAE Tetracentron sinense Oliver 1183 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). EUPTELEACEAE Euptelea pleiosperma Hooker & Thomson 105 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); //5/] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/549 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, 216 (A, CM, E, UC, WH RANUNCULACEAE Aconitum cannabifolium Franchet 1290 (A, CM, HIB, KUN, NA, NAS, NY, PE). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 47 Aconitum hemsleyanum Pritzel 69 (A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 154 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 388 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, 1156 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1777 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Aconitum scaposum Franchet A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 19/ (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 357 (A, HIB); 42] (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC ee Actaea asiatica H 413 (HIB); ee es se a (A, HIB, NA, PE). Anemone bia sis Li 7 E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Anemone reer (Maxim.) Péi 53 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Aquilegia oxysepala Trautv. & Meyer* 258 (A, HIB, KUN, NA, NAS, NY, PE, UC); 869 (A, HIB, KUN, NA, PE, UC). Cimicifuga acerina (Sieb. & Zucc.) Tanaka a S CM, HIB, NA, NAS, PE); 622 (A, HIB); 136] (A, CM, E, HIB, KUN, NA, NAS, , UC, WH). GC imicificea, fot Le 253 (A HIB, NY, UC Gre ae Wornsk , CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Clematis gratopsis W. T. Wang 439 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1905 (A, CM, E, HIB, KUN, , NY, PE, SFDH, UC, WH). Clematis lasiandra Maxim. I (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); /8/ (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 2/7 (A, CM, E, HIB, KUN, MO, NA, NAS, NY, PE, SFDH, UC, WH); 3/0 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 700 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/92 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /328 (A, HIB, KUN, NA, NAS, NY, PE); 1648 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Clematis otophora Franchet 301 (A, CM, HIB, KUN, NAS, NY, PE, UC); 635 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1546 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). Clematis uncinata Champ. var. coriacea Pampan. Nuovo Giorn. Bot. Ital. 22: 288 580 (A, , HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). C lemons eine Franchet* 1940 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Belen naienini Huth* 4 A, HIB, PE). Paeonia obovata Maxim. var. willmottiae (Stapf) Stern, J. Roy. Hort. Soc. 68: 128. 1943 197 (A, HIB); 385 (A, HIB, KUN, NAS, PE, UC). 48 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 a przewalskii Maxim. _ CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). ne robustum = 572 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Thalictrum see oa het 148 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). LARDIZABALACEAE Akebia trifoliata (Thunb.) Koidz. 77 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1089 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Decaisnea fargesii Franchet 72 (A, CM, HIB, NY, PE); 6/8 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1055 (A, HIB, PE, UC). Holboellia fargesii Réaubourg 616 oi HIB); 1005 a _ HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1704 (A, HIB, NA, NA Che eres as (Eranchet) Hemsley 124] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1879 (A, HIB, NY, PE, SFDH, UC). BERBERIDACEAE Berberis circumserrata (Schneider) Schneider 255 (A, CM, HIB, KUN, NA, NY, PE, UC). geo dasystachya Maxim ae HIB, KUN, NY, PE, 08h 34] (HIB); 888 (A, CM, E, HIB, KUN, KYO, MO, ee ae aes A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Berberis henryana Schneider 373 (A, HIB, PE, UC); 1232 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, H). Berberis julianae Schneider 232 (A, CM, HIB, KUN, NAS, NY, PE, UC); 1804 (A, CM, HIB, KUN, NA, NY, PE). mes mitifolia Stapf* 04 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1553 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH);/86/ (A, HIB, KUN, NA, NAS, PE, SFDH, Berbers sargentiana Schneider 47 i , CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /938 (A, CM, E, HIB, KUN , NY, PE, SFDH, UC, rie ee ee ae Fedde 643 (A, CM, HIB, KUN, NA, NAS, PE, UC). Berberis virgetorum Schneider 2079 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Caulophyllum robustum Maxim. 91] (a1); 1202 (A, HIB, KUN, MO, NA, NAS, NY, PE, UC). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 49 Hee age eee |e Ge ea) 910 (A, ets ae Franchet 186 B Nandina domestica Thunb. 110] (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). MENISPERMACEAE Sinomenium acutum (Thunb.) Rehder & Wilson 696 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1067 (A, CM, HIB, KUN, NA, NAS, NY, PE). Stephania sinica ae 698 (A, CM, HIB , NY, PE, SFDH, UC, WH). ek sagittata (Olive nee 716 (A, HIB, PE). MAGNOLIACEAE Magnolia sprengeri Pampan.* 1181 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ILLICIACEAE Iilicium henryi Diels (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1028 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 149] (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). SCHISANDRACEAE Schisandra glaucescens Diels 6 (HIB); 10] (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 199 (HIB); 331 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 382 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC, WH); 645 (A, CM, HIB, KUN, NA, PE, UC); 711 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC); 1389 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH LAURACEAE Actinodaphne cf. reticulata Meisner var. forrestii Allen, Ann. Missouri Bot. Gard. 25: 412 1070 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lindera communis Hemsley 1423 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1650 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1679 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1906 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Lindera fragrans Oliver 1623 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lindera fruticosa (Hemsley) Gamble 50 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 765 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/481] (A, CM, HIB, KUN, , NAS, NY, PE, SFDH, UC, WH). Lindera glauca (Sieb. & Zucc.) Blume 447 (A, CM, E, oe KUN, NA, NAS, NY, PE, SFDH, UC, WH); 782 (A, CM, E, HIB KUN, KYO, NA, NAS, NY, PE, SFDH, UC). Lindera megan Hemsley 1900 (A , E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lindera crab Blum ) , E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lindera ae (Merr.) Merr.* 704 (A, HIB). Litsea ichangensis Gamble 303 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 5/0 (A, CM, HIB, KUN NA, NAS, NY, PE, SFDH, UC, WH); 63] (A, CM, E, HIB, KUN, KYO, NA, NAS, NY PE, SFDH, UC, WH); 18/2 (A, CM, HIB, PE). Litsea aff. sericea (Nees) Hooker f.* 3 A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 892 (A, CM, E, HIB, a NAS, NY, PE, SFDH, UC, WH); 1587 (A, CM, HIB, KUN, NA, NAS, NY, aE CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Neolise ‘onfoi (Hemsley) M , HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 583 (A, CM, HIB, KUN, NA, ae “ PE, SFDH, UC); 603 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC WH); 1027 (A, CM, HIB, KUN, NAS, NY, PE); 1085 (A, CM, HIB, KUN, NA, NAS NY, PE, a UC, WH); aie (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC WH); 1608 (A, CM, E, HIB N, NA, NAS, NY, PE, SFDH, UC, WH). Phoebe neurantha (Hemsley) Ganbie 2045 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). KUN PAPAVERACEAE Macleaya eae tgs ) Fedde 300 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Se lophorum lasipcarpum (Oliver) Fedde S4 (A, HIB); 1064 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC WH). FUMARIACEAE Corydalis davidii Franchet a (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 859 (A, CM, E, HIB, A, NAS, NY, PE, SFDH, UC, WH). o oda incisa (Chunb. )P A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). CRUCIFERAE Arabis pendula L. 1153 A, CM, HIB, KUN, NA, NAS, NY, PE, SEDH, UC, WH). Cardamine urbaniana O. E. Schulz I A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 51 Rorippa indica (L.) Hieron. 1020 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). CRASSULACEAE Sedum aizoon L. 107 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 12/2 (A, HIB). Sedum amplibracteatum K. T. Fu, Fl. Tsinlingensis 1(2): 425. 1974. (a nomen novum for S. bracteatum Diels, non S. bracteatum Viv.) 203 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 100] (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Sedum dielsii Hamet 1634 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Sedum erythrostictum Miq.* (syn.: S. telephium L. subsp. alboroseum (Baker) Fréd. Acta Horti Gothob. 5(Appendix): 61. 1930.) 1210 (A, HIB). ee filipes Hemsley 589 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 776 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 974 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH ae verticillatum L. 173 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). SAXIFRAGACEAE Astilbe rubra Hooker & Thomson 75 (A, CM, HIB, NAS, PE, UC). ep aes lanuginosum Hooker & Thomson* (incl. C. ciliatum Franchet) 12 , HIB). C. eens macrophyllum Oliver 1208 (A, HIB). Beenie pilosum Maxim. var. valdepilosum Ohwi, Repert. Spec. Nov. Regni Veg. 36: 52. 1934. 1207 (A, HIB). Decumaria sinensis Oliver 1454 (A, HIB, NA, NAS, NY, PE, UC); 1899 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH) Deutzia schneideriana Rehder* 1615 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Deutzia vilmorinae Lemoine 6 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/49 (A, CM, HIB, NAS, NY, PE); 349 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 539 (A, CM, HIB, NA, NAS, PE); 1032 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, aia febrifuga Lour. 2 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ene aspera D. Don subsp. robusta (Hooker & Thomson) McClint. Proc. Calif. Acad. Sci. 29: 194. 1957 oa) JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 370 (A, HIB, KUN, NA, NAS, NY, PE, UC); 606 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1197 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1353 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, WH). Hydrangea aspera D. Don subsp. strigose (Rehder) MeClint Proc. Calif. Acad. Sci. 29: 193. 1957. 467 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 693 (A, HIB, KUN, NA, NAS, NY, PE, UC); 1/06 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 19/3 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Hydrangea heteromalla D. Don 128 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /84 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 370a (A); 735 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 884 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH), 1341] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/529 (A, CM HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); [772 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Hydrangea scandens (L. f.) Sér. subsp. chinensis (Maxim.) McClint. Proc. Calif. Acad. Sci. 29: 206. 1957 2003 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Ttea ilicifolia Oliver 1107 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Parnassia delavayi Franchet 37 (A, CM, HIB, KUN, NA, PE, UC). Parnassia wightiana Wallich ex Arnott 186 (A, HIB, PE, UC). Philadelphus incanus Koehne 230 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 5/5 (A, HIB, NA, NAS, NY, PE); 748 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); [026 (A CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /6/4 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). ee sod sericanthus Koehne , CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 956 (A, CM, E, HIB, UN, KYO, , NY, PE, SFDH, UC). Rides ac natin a Wail ch 40) (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH): //2 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /58 . CM, HIB, KUN, NA, NAS, PE, SFDH, UC, ; 252 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). ae fascieulanim Sieb. * Zucc. var. ee Misc m. Mélanges Biol. Bull. Phys.-Math. Acad. Imp. Sci. Saint-Pétersbourg 9: 237. 1873; Bull. Acad. Imp. Sci. Saint- Besioue 19; 264. 1874. 1682 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Ribes moupinense Franchet 647 (A, HIB, PE); 969 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Ribes sp. 167 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rodgersia aesculifolia Batalin 79 (A, HIB); YOO (A, CM, HIB, KUN, NA, PE). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 53 Saxifraga flabellifolia Franchet 3 A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Saxifraga giraldiana Engler var. hupehensis Engler, Bot. Jahrb. Syst. 29: 366. 1901. 26 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /72 (A, CM, HIB, NA, NAS, NY, PE, UC) Schizophragma integrifolium Oliver a Me CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); /29/ (A, N, NA, NAS, NY, PE, SFDH, UC, WH). J6 Ae polyphylla D. Don 1864 (A, HIB, KUN, PE, UC). PITTOSPORACEAE Pittosporum glabratum Lindley var. neriifolium Rehder & Wilson, Pl. Wil- sonianae 3: 328 2030 (A, HIB). Pittosporum rehderianum Gowda 478 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/066 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1438 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/640 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, W Pittosporum i Pritzel 1616 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). HAMAMELIDACEAE Corylopsis platypelta Rehder & Wilson 13 M, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Corylopsis sinensis Hemsle 752 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Corylopsis veitchiana Bean 2033 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Liquidambar eee Chan. Bull. Sun Yat-sen tay. 2: 33. 1959: Fl. Reip. Pop. Sinicae 35(2): 5 979. 1950 (A, HIB, NAS, NY, PE, UC). Sinowilsonia henryi Hemsley 1483 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Sycopsis sinensis Oliver 248 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC) ROSACEAE Agrimonia pilosa Ledeb. 451 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Cotoneaster acutifolius baie 6 (A, CM, HIB, NAS, NY, PE, nee 344 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 365 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, 54 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 wu); 857 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); /55/ (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1731] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH Cotoneaster acutifolius uae var. aalacias Rehder & Wilson, Pl. Wilsonianae 1: 158 3 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC, WH). Cotoneaster bullatus Bios. 215 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC). Cotoneaster dielsianus Pritzel 1016 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1044 (A, CM, HIB, KUN, NA, NAS, PE, UC); en (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1346 CM, HIB, KUN , NAS, NY, PE, C otoneaster aff. Fae le el 1935 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cotoneaster divaricatus Rehder* 1318 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cotoneaster foveolatus Rehder & Wilson* 265 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 906 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC); 94/ (A, CM, E, HIB, eat NA, NAS, NY, PE, SFDH, UC, WH); 1332 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1344 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/750 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Cotoneaster fo foveolatus Rehder & Wilson 1854 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cotoneaster ee Dene. 100 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 375 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 779 (A, CM, E, HIB, KUN, KYO NA, NAS, NY, PE, SFDH, UC, WH); 1/609 (A, HIB, KUN, NA, NAS, NY, PE, UC); 1/830 oe E, ns, se nee NA, NAS, NY, PE, SFDH, UC, WH); 1933 (A, CM, E, HIB, xe) , KYO, Ci ooneater ns Bun nge 51 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1806 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1834 (A, CM, HIB, KUN, NA, NAS, Y, PE, SFDH, UC, WH). Cotoneaster seus Rehder & Wilson* 129 M, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 205 (A, CM, HIB, NA, PE, UC); 211 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 393 (A, CM HIB, KUN, NA, NAS, NY, PE, SFDH Cotoneaster salicifolius Franchet 676 (A, CM, HIB, KUN, NAS, PE, SFDH, UC, WH); 1086 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1907 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, WH). Cotoneaster salicifolius Franchet var. rugosus (Pritzel) Rehder & Wilson, PI. ilsonianae 1: . 1912. (syn.: C. rugosus Pritzel) A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /708 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 55 Cotoneaster aff. zabellii Schneider 575 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC); 1579 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Cotoneaster sp. 372 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /80/ (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). Crataegus cuneata Sieb. & Zucc.* 1736 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Crataegus wilsonii Sarg. A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 70 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, a 263 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 896 (A, CM, HIB, NAS, NY, PE); 1056 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, eee 1234 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Fragaria gracilis A. Los. 164 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Kerria japonica (L.) DC. 221 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 350 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 579 (A, CM, HIB, NA, NAS, NY, PE); 1279 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, Malus eae (Batalin) Schneider* 6 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 262 (A, HIB, PE, uc); 397 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 893 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Malus yunnanensis (Franchet) Schneider* 57] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1556 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 177] (A, CM, E, HIB, KUN, KYO, O, NA, NAS, NY, PE, SFDH H). Malus yunnanensis (Franchet) Schneider var. veitchii Hort. ex Rehder, J. Arnold Arbor. 4: 115. 1923 1301 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Malus sp. 106 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 484 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 767 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 12/8 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, UC); 1298 NAS UC, WH); 1885 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH The collections of Malus listed above have been quite difficult to identify, although all (with the possible exception of 1/298) are referable to either M. hupehensis or M. halliana Koehne. It has become apparent, however, that the relationships of these two species are in need of careful assessment and resolution before the above-listed collections can be named with any assur- 56 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ance. Unfortunately, time has not allowed more than a casual investigation of this problem, but it is hoped that further collaborative work can soon be undertaken in both China and the United States. Neillia sinensis Oliver 135 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 470 (A, HIB); 678 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1077 (A, HIB, NY, PE). Photinia beauverdiana Schneider 770 (A, CM, HIB, KUN, NAS, PE, UC); 1543 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, ae 1583 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC, WH); 1875 (A, , KUN, KYO, NA, NAS, NY, PE, SFDH, Photinn ee ae ere var. notabilis (Schneiden Rehder & Wilson, Pl. Wilsonianae 1: Dele 1022 (A, , PE, SFDH, UC, WH). Photinia ° arvifolia ‘Pritzed roe aa a CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 208/ (A, CM, E, HIB, , NA, NAS, NY, PE, SFDH, UC, Bits villosa (Thunb.) DC 1865 (A, HIB, NA, PE, SFDH, ud, Potentilla fruticosa L. 1 , CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); //84 (A, UN, NA, NAS, NY, PE, SFDH, UC, WH). i ee D. Don 937 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Prunus brachypoda Batalin 967 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Prunus salicina Lindley 139 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 336 (A, CM, HIB, NAS, NY, PE, SFDH, UC, WH). Pyracantha crenulata (D. Don) Roemer 1903 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /96/ (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Pyracantha fortuneana (Maxim.) H. L. Li 436 E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 466 (A, CM, cy Cc > HIB, KUN, N NY, PE, SFDH, UC, WH); 14/7 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, i WH); 1622 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, 1939 (A , E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). H); Pyrus pyrifolia ‘Bannan f. ) Nak S47 (A, CM, HIB, sae NA, NAS, NY, PE, UC). ee serrulata Rehder 1630 (A, CM, E, ee KUN, NAS, NY, PE, SFDH, UC, WH). Rosa banksiae Aito 442 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /4/0 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1632 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rosa banksiopsis Baker 1778 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 57 Rosa helenae Rehder & Wilson 747 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Rosa ie Boulenger dE , E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Rosa omeiensis Rolfe (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, wh); 125 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rosa saturata Baker sis (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 960 (A, , E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rosa aia Rolfe 1 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Ae sp. 10 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, wH); 147 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 187 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 732 (A, HIB); 876 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 957 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1053 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1406 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /534 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1633 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, wu); 1784 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1862 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1863 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1934 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, ). A situation similar to that mentioned under the Rac naned collections of Malus exists with the undetermined collections of Rosa. Although for the most part the collections fall into two or three groups, unclear or unresolved species limits and the possibility of hybridization have prevented assured determinations and indicate the need for further work. Rubus amphidasys Focke ex Diels 2076 (A, CM, HIB, NAS, NY, PE). Rubus flosculosus Focke 5 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, wu); 1545 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rubus ichangensis Eemsley & Kuntze* 551 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rubus innominatus S. Moore 449 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Rubus lambertianus Sér.* 499 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 502 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 799 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /4 A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UG, es | (A, CM, HIB, KUN, NA, NAS, NY, PE, UC); 1936 (A, CM, E, HIB, NA, , NY, PE, SFDH, UC, WH). roe eee Focke 22 , E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1058 (A, HIB, KUN, NA, NAS, NY, PE); 1312 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 58 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Rubus lasiostylis Focke var. hubeiensis Yii, Spongberg, & Lu, var. nov. A varietate lasiostylo in drupeolis indumento minore praeditis, aurantiaco- rubrae, et stylo parte distili glabro deciduo, differt Differing from var. /asiostylus in having less densely pubescent, orange- red drupelets and distally glabrous, deciduous styles Type. China, western Hubei Province, Shennongjia Forest District (31°30'N, 110°30’E), on a steep, NW-facing slope above the road to Chuifeng Pass, elevation ca. 2780 m; shrub 1.5—2 m tall, fruits red, 1-1.5 cm across, the young stems grayish, bark on older stems brownish red, 26 August 1980, Sino-Amer. Bot. Exped. 114 (holotype, PE; isotypes, A, HIB, KUN, NA, NAS, NY, SFDH, UC, WH). ADDITIONAL SPECIMENS EXAMINED. China. Huser: Shennongjia Forest District (31°30’N, 110°30'E), NW-facing slope below Chuifeng Pass in Sinarundinaria thickets and in disturbed meadow with Abies chensiensis, elev. ca. 2900 m; shrubs with orange-red, sweet, edible fruits, 1980 Sino-Amer. Bot. Exped. 155 (A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rubus lasiostylus var. hubeiensis can be distinguished from var. /asiostylus by its less densely pubescent drupelets, by its distally glabrous (vs. pubescent) styles that tend to be deciduous (vs. persistent) prior to drupelet maturity, and by the orange-red color (pinkish when dry) of the drupelets (vs. whitish: tannish when dry). Based on the limited material available for comparison, other characters that may be of significance include the relative lack of prickles on the floricanes and an apparent difference in stipule and infruc- tescence-bract shape. In var. /asiostylus the floricanes are usually beset with numerous straight to slightly recurved, spreading prickles, and the stipules and bracts of the infructescence are narrowly lanceolate. By contrast, the stipules and bracts of var. hubeiensis tend to be ovate to suborbiculate As in Rubus lasiostylus var. dizygos Focke (in Pl. Wilsonianae 1: 53. 1911), the leaves on the flowering shoots of var. hubeiensis sometimes bear five (rather than three) leaflets. In other details of the vegetative parts, fruit aggregates, and seeds, var. hubeiensis agrees with var. /asiostylus. Rubus setchuenensis Bur. & Franchet 1952 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rubus simplex Focke 1130 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Sorbaria arborea Schneider 7 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 266 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 97] (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, , UC, WH); 1025 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Serbs eee (Sieb. & Zucc.) K. Koch A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1302 (A, cM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/400 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1779 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1855 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 59 ae caloneura (Stapf) Rehder _ CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1097 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 2039 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Sorbus folgneri (Schneider) Rehder 1008 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH), 1476 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1870 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Sorbus hupehensis Schneider 102 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 879 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 13/5 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Sorbus keissleri Rehder 1285 (A, HIB, KUN, NA, NAS, NY, PE, UC). Sorbus koehneana Schneider 880 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 964 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Sorbus xanthoneura Rehder 719 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, wu); 1317 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1342 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1739 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1776 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Sorbus zahlbruckneri Schneider 1555 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, wH), 1728 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Spiraea chinensis Maxim.* 16 A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Spiraea japonica L. f. var. acuminata Franchet, Nouv. Arch. Mus. Hist. Nat. Il. 8: 218. 1886; Pl. David. 2: 36. 1888 8 (A, CM, HIB, KUN, NAS, NY, PE, SFDH, UC, WH), 444 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 78/ (A, CM, HIB, NY, PE); S82 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Spiraea veitchii Hemsle A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 389 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC); 860 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, wH); 1155 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Stranvaesia amphidoxa Schneider (syn.: Photinia amphidoxa (Schneider) Reh- der & Wilson 01 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Stranvaesia davidiana Dcne.* Ds _ CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 213 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 5/3 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 6/4 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 772 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, wH); 1039 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1081] (A, CM, E, HIB, KUN, 60 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 NA, NAS, NY, PE, SFDH, UC, WH); 1496 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, Stranvaesia eee Dene. var. undulata (Dene.) Rehder & Wilson, Pl. Wil- sonianae I: 192. 191 2035 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). LEGUMINOSAE Amphicarpaea trisperma Baker / (A, HIB); 786 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1661 (A, HIB, NA, PE). Bauhinia hupehana Craib Caesalpinia sepiaria Roxb. 1114 (A, HIB). Campylotropis ichangensis Schindler O65 (A , HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cercis chinensis Bunge 1010 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1695 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cercis racemosa Oliver 1115 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Dalbergia dyeriana Prain & Harms 1428 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1624 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SEDH, UC, WH). Dalbergia a aaa a et ee E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /42/ (A, CM, HIB, E); 73 ( A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Ree bidhawli Oliver 1526 (A, HIB, KUN, NA, NAS, PE, UC). Desmodium nodocarpuin DC. subsp. podocarpum i , CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Dein podocarpum DC. subsp. fallax (Schindley Ohashi in Hara, FI. E. Himalaya 2: 65. 1971. (see also Ohashi, a 1068 (A, CM, HIB, KUN, NA, NAS, NY, PE, SF UC). Desmodium podocarpum DC. subsp. i Onan in Hara, FI. E. Himalaya 2: 65. 1971, var. oxyphyllum* 804 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1499 (A, cM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Indigofera amblyantha Craib (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1732 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, Indigofera pseudotinctoria Matsum. 407 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lespedeza buergeri Miq. 1607 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1860 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). i 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 61 Lespedeza cuneata eee -Cours.) G. Don 703 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ane ee (DC.) Nakai* 4 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 56 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1744 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH Vicia-cracca L. 1134 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Vicia pseudo-orobus Fischer & Meyer 1856 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). GERANIACEAE Geranium henryi Kunth 11 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, WH); 65 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH Geranium sibiricum L. 144] (A, CM, HIB, KUN, NA, NAS, PE, UC). RUTACEAE Euodia rutacarpa (Juss.) Bentham var. bodinieri (Dode) Huang, Acta Phytotax. Sinica 6: 113. 1957 1947 (A, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Euodia rutacarpa (Juss.) Bentham var. officinalis (Dode) Huang*, Acta Phy- totax. Sinica 6: 114. 1957. 1450 (A, E, HIB, KUN, NA, NAS WH). In Hartley’s recent revision car ‘Bull. Sing 34; 91-131. 1981) both of the above-mentioned taxa of Evodia are included in Tetradium ruticarpum (A. Juss.) Hartley. Hartley’s revision should be consulted for the basis of the division of the genera Tetradium Lour. and Euodia J. R. & G. Forster, as well as Melicope J. R. & G. Forster. Zanthoxylum armatum DC. (syn.: Z. planispinum Sieb. & Zucc.) 433 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 797 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1029 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 7408 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH Zanthoxylum dissitum Hemsley 1602 (A, CM, HIB, KUN, NAS, NY, PE, UC); 1/675 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Zanthoxylum undulatifolium Hemsley 633 (A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). POLYGALACEAE Polygala arillata Buch.-Ham. 189 (A, HIB, NY, PE). 62 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Polygala tatarinowil se 684 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC); 1688 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH) Polygala wattersii Hance 1698 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). EUPHORBIACEAE ao gales a 19 NAS, NY, PE, SFDH, UC). Hischotn: pay ae (Levi ) Airy. Shaw 3 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Euphorbia chrysocoma Lévl. & Vaniot 1200 (A, CM, HIB, KUN, NA, NAS, = PE, SFDH, UC, WH). Euphorbia hylonoma Hand.-Maz 364 (A, HIB, KUN, NA, NAS, NY, a 1IS& (A, CM, E, HIB, KUN, KYO, NA, NAS, Y, P E, H, Cc, id Leptopus ene (Bunge) Poyjark. (syn.: Andrachne chinensis Bunge) 1613 (A, HIB, NA, PE); 1678 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Mallotus contubernalis Hance 448 (A, CM, HIB, KUN, NAS, NY, PE, UC); 1644 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). DAPHNIPHYLLACEAE ee ina ee Chien , E, HIB, , KYO, ne NAS, NY, PE, SFDH, UC, WH). eae SENET iq.* 605 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). BUXACEAE Buxus microphylla Sieb. & Zucc. var. sinica Rehder & Wilson, Pl. Wilsonianae 223 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1397 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Pachysandra terminalis Sieb. & Zucc. 309 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 648 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /203 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1330 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, Wilson, Pl. Wilsonianae 2: 164. 305 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, UC); 403 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC , 6 , CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); ey (a, HB); 1350 (A, HIB, NY, PE); 7352 (A, CM, HIB, KUN, NA, PE); 1967 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). WH Sareoenten humilis Stapf (syn.: S. hookeriana Baillon var. humilis Rehder & 1914. 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 63 Sarcococca ruscifolia Stapf 1912 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ANACARDIACEAE Rhus chinensis Miller 1445 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rhus verniciflua Stokes 103 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). AQUIFOLIACEAE Tlex fargesii Franchet 717 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1304 (a, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1395 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1753 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1775 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH Tlex ne ae Oliver 19 CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Ilex Bia ey iq. 1227 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Ilex pedunculosa Miq. 741 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Ilex pernyi Franchet 239 (A, HIB); 673 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 800 (A, CM, HIB, KUN, NA, NAS, NY, PE); 1094 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, uc, WH); 1131 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1625 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Ilex shennongjiaensis T. R. Dudley & S. C. Sun, sp. nov. A Ilex sugerokii Maxim. ex Japonia et Taiwania persimilis et manifeste affinis sed ramulis primo non nisi puberulis; gemmis terminalis brevis stipita- tis; foliis laminis aliquantum maioribus et crassioribus; petiolis et stipulis brevioribus; fructibus persaepe maioribus; calycibus maioribus glabris, sinu- bus non profundus, louie eeloabus eciliatis; stigmatibus maioribus sessili- bus, lobis parus elevat re renis maioribus fere teretibus transversaliter circularis, dorsaliter bisulcatis et striatis, prae- cipue differt. Single-trunked tree to 10 m tall, 25 cm d.b.h. Mature bark glossy, whitish gray. First-year branchlets rigid, reddish brown, glistening, very sparsely puberulent becoming glabrous, 1-1.5 mm in diameter; third- and fourth- year growth terete, glabrous, grayish brown, glossy, 3-4 mm in diameter; leaves borne predominantly on first- and second-year branchlets, rarely per- sisting on older growth. Terminal buds shortly stipitate, ovoid or ellipsoid, 3-4 mm long, acute; stipes 0.5—1 mm long, sparingly puberulent; bud scales glabrous or remotely puberulent, not ciliate along margins; internodes 2-6 mm long; stipules deciduous, subulate, 0.3-0.5 mm long, glabrous. Leaves 64 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 with petiole 2.5-3(-4) mm long, deeply grooved on adaxial surface, only sparsely puberulent; blade elliptic-ovate, 2.5-4(-5) by (1-)1.5-2.5(-3) cm, acute to subobtuse and shortly mucronulate at apex, mucro 0.2-0.5 mm long and amber colored, short-attenuate or truncate to occasionally briefly cuneate at base and sparingly puberulent near petiole, thickly coriaceous, dark green, glossy and glabrous on upper surface, dull yellow-green on lower surface, the lateral veins obscure, 6 or 7 pairs, the midvein minutely puberulent, flattened to complanate or somewhat impressed on upper surface, conspicuously ele- vated and glabrous on lower surface, the margin serrulate-crenulate along ¥s—" of distal portion, with 3 to 7 serrulations per cm. Flowers unknown. Infructescences solitary and axillary; fruiting peduncles erect, |-1.5 mm long, rigid, glabrous, with 2 ovate, obtuse-cuspidate, glabrous prophylls at or above middle; pedicels 5.5—12 mm long, stout, rigid, more or less dilated toward base of fruit. Fruits spheroid and often oblate, 8-12 mm broad, dark cherry- red, glossy, glabrous, sweet tasting; calyces generally persistent, explanate, —5 mm across, glabrous, with shallow sinuses and 4 to 6 obtuse to subacute stigmas sessile, flattened, 1.5-2 mm in diameter, circular, with 4 (or 5) narrowly elliptic-fusiform, scarcely raised, separate lobes | by 0.2-0.3 mm. Pyrenes 4 or 5 per fruit, oblong, nearly circular in cross section, 4-5 by 2.5- 3 mm; dorsal surface moderately bisulcate, with 2 thin, prominent, raised marginal ribs and | raised medial rib; ventral and lateral surfaces smooth; endocarps cartilaginous. Type. China, western Hubei Province, Shennongjia Forest District (31°30'N, 110°30’E); infrequent between Yinpo and Qiaodonggou canyon, on bank along road between Jiuhuping Forest Farm and Bancang, elevation ca. 2100 m, 19 September 1980, Simo-Amer. Bot. Exped. 1554 (holotype, PE; isotypes, A, CM, E, HIB, KUN, KYO, NA, NAS, NY, SFDH, UC, WH). ADDITIONAL SPECIMENS EXAMINED. China, Huse: sae Forest District (31°30’N, 110°30’E); infrequent at Miaogou canyon, ca. 4 km N of Jiuhuping, elev. ca. 1768 m in thicket along edge of stream, 1980 Sino- ae. Bot. Exped. 236 (A, HIB, NA, PE). Ilex shennongjiaensis falls in ser. Cassinoides (Loes.) Hu of sect. PALTORIA (Ruiz & Pavon) Maxim. of subg. ILEx. The other — components of ser. Cassinoides are I. pedunculosa Miq., I. rockii S. u, 7. punnanensis Franchet, /. kirinsanensis Nakai, and I. sugerokii a Tlex shennongyjia- ensis has foliage and fruit that suggest close affinity with J. sugerokii, but it can be distinguished from that species by the larger pyrenes that are dorsally sulcate and striate; the larger, earlier-ripening, bright red, glossy, sweet fruit; the larger, sessile stigmas with flattened, conspicuously separate lobes; the larger calyces with shallow sinuses and glabrous lobes; the shorter pedicels that are dilated at both ends; the glabrous peduncles bearing longer, eciliate prophylls; the shorter, glabrous, deciduous stipules; and the deeply grooved, shorter petioles. The short, strongly grooved petioles of //ex shennongjiaensis are also rem- iniscent of J. kirinsanensis, endemic to Mt. Kirin-san in Japan. Most of the features that distinguish /. shennongjiaensis from I. sugerokii also apply to 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 65 and separate J. kirinsanensis. The lower surfaces of the leaves of J. kirin- sanensis are described as being glandular-punctate. After more material is studied, this taxon might realistically be relegated to infraspecific rank within I. sugerokii. Loesener (1901) tentatively assigned a Chinese collection, Farges 129, from Jiangkouzhen (““Tschen-keow-tin’’), Sichuan Province, to aoe forma brevipedunculata Maxim. (= var. brevipedunculata (Maxim.) S Hu), commenting that it was intermediate between that form and J. yun- nanensis. Hu (1949) regarded Farges 129 as I. yunnanensis var. gentilis (Franchet) Loes. ex Diels. aes wilsonii Loes. 031 (A, CM, HIB, NA, NAS, PE). is yunnanensis Franchet var. gentilis (Franchet) Loes. ex Diels*, Bot. Jahrb. Syst. 29: 435. 1900; Nova Acta Acad. Caes. Leop.-Carol. German. Nat. Cur. 646 (A, HIB, PE); 89/ (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, wu); 1303 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1763 (A, HIB, KUN, NA, NAS, NY, PE, UC); 1773 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). CELASTRACEAE Celastrus angulatus Maxim. 10 , CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1646 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Celastrus gemmatus Loes. 785 ue CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC); 1220 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Cela es Rehder & Wilson (syn.: C. rugosus Rehder & Wilson) 193 ies HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 3/3 (A, HIB, KUN, NA, PE, uc); 8 94 (A, HIB, KUN, NA, NAS, NY, PE, UC). coe orbiculatus Thunb. ~ , CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1006 (A, CM, E, HIB, UN, NY, PE, SFDH, UC, WH). Ce ea eae Loes. 1282 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). E enymus acanthocarpus Franchet 574 (A, CM, E, H , KYO, MO, NA, NAS, NY, PE, SFDH, UC). Euonymus alatus acon ) Sieb. (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 347 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 5/6 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, uc); 795 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/52 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1284 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Euonymus cornutus Hemsle 608 (A, HIB, NY, PE); 639 (A, CM, HIB, KUN, NA, NAS, PE); 865 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1289 (A, HIB). 66 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Euonymus Gaae peers ae E. elegantissimus Loes. & Rehder) IZ33(A , HIB NAS, NY, PE, SFDH, UC, WH). Euonymus eee ePurez, ) Hand. -Mazz. 1337 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Euonymus giraldii Loes.* 261 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Euonymus hamiltonianus Wallic 137 (A, CM, HIB, NAS, NY, PE); 1023 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY PE, SFDH, UC, WH); — (A, CM, HIB, KUN, NAS, NY, PE, SFDH, UC); 1233 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1552 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1749 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Euonymus kiautschovicus Loes. 2075 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Euonymus maackii Rupr.* 887 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Euonymus oxyphyllus Miq. 26 , CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Euonymus phellomanes Loes. 366 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 954 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH Euonymus porphyreus Loes. 864 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 965 (A, CM, HIB, KUN, PE, UC). Euonymus sanguineus Loes. 1004 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1799 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /87/ (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, , UC, WH). ye Tene vero Loes 1527 (A, CM, E, HIB YO, MO, NA, NA PE, SFDH, UC, WH). Euonymus nee ie Les var. Sige ee '& Rehder*, PL Wilsonianae 1: 493 398 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Microtropis triflora Merr. & Freeman 2018 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Perrottetia racemosa (Oliver) Loes.* 561 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1902 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). STAPHYLEACEAE Euscaphis japonica (Thunb.) Kanitz 1942 (A, HIB). Staphylea bumalda DC. 716 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Staphylea holocarpa Hemsle (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /084 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 67 ICACINACEAE Hosiea sinensis (Oliver) Hemsley & Wilson 1082 (A, HIB, KUN, NAS, NY, PE, SFDH, UC). ACERACEAE Acer amplum Rehder 738 (A, CM, HIB, KUN, PE). Acer caudatum site var. be naar (Maxim.) Rehder in Sarg. Trees Shrubs 1: 163. 1905. 110 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /6/ (A, CM, E, IB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, Keren to Bean (Trees Shrubs Hardy Brit. Tee ed. 8. 1: 220. 1970), the correct name of Acer caudatum Wallich is A. papilio King, the former name constituting a nomen ambiguum. Moreover, var. multiserratum 1s treated in the same work (1: 239) as a distinct species, A. multiserratum Maxim.; we have been unable to locate a proposal to treat this taxon at the varietal rank under A. papilio. Inasmuch as this problem is both taxonomic and nomenclatural, the varietal combination is not proposed here pending additional study of the taxonomy of the group, and the name is provisionally maintained as Acer caudatum var. multiserratum. Acer davidii Franchet 517 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 536 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 679 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, ae UC, WH); 1/25 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, Acer erianthum eeiuatin 1041 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /53/ (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/781] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Acer flabellatum Rehder 1127 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Acer franchetii Pax 83 (A, HIB); 1024 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1240 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /76/ (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /828 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, Acer griseum (Franchet) Pax 1482 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 157/ (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1747 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Acer henryi Pax 1740 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Acer maximowiczil Pax 368 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 896 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). 68 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Acer mono Maxim. 150 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Acer robustum Pax 1359 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Acer sinense Pax 1805 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Acer stachyophyllum Hiern 302 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 10/8 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /2// (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Acer J aedlatige Franchet* 2 , HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). ee sinensis Oliver 1063 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). SABIACEAE _ pene hae = pias 007 (A, S, NY, PE, SFDH, UC, WH). Melasma leno (Wellich ex ete & Arnott) Walp. subsp. cuneifolia (Franchet) Beus. Blumea 19: 442. 1971. a A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1198 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 7868 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Meliosma dilleniifolia (Wallich ex Wight & Arnott) Walp. subsp. flexuosa (Pampan.) Beus.* Blumea 19: 444 1100 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1949 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Meliosma veitchiorum Hemsley 1498 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). BALSAMINACEAE Impatiens blephorosepala Pritzel ex Diels* 1246 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Impatiens dicentra Franchet 55 (A, CM, HIB, KUN, NAS, NY, PE, UC); 560 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 60/ (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 558 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Impatiens cf. exiguiflora Hooker f. 1960 (A, CM, HIB, KUN, NAS, NY, PE, UC). Impatiens pterosepala Pritzel ex Diels 58 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 77 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 559 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 766 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); aga CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1037 (A, CM, KUN , NY, PE, SFDH, UC, WH). rae ee Hooker eg 1090 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 69 Impatiens stenosepala Pritzel ex Diels 558 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Impatiens sutchuanensis Franchet ex Hooker f.* 17 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 5/ (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). RHAMNACEAE Berchemia flavescens (Wallich) Brongn. 1 CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1798 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, U H). Berchemia floribunda (Roxb.) Brongn. 737 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rhamnus crenatus Sieb. & Zucc 771 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); / 1/94 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /488 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1702 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, wH); 1880 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rhamnus davuricus Pallas 2024 (A, HIB, NA, PE) Rhamnus dumetorum Schneider* (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rhamnus esquirolii Lévl.* 1683 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rhamnus Teinophyiius Schneider mee (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1047 (A, CM, E, HIB, , NA, , NY, PE, SFDH, UC, WH He tis eo li Schneider S508 (A, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 522 (A, CM, E, HIB, KUN, NA, NAS, NY, PE om UC, WH); 628 (A, CM, E, HIB, KUN, NA, NAS, NY, C, WH); 629 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /336 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). < jan} oi Ll 2S No PSG P oe. > e) s Rhamunus utilis Dene. 465 (A, HIB); 1327 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 14/9 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); /54/ (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). VITACEAE eae ecards Diels & Gilg nies KUN, NA, NAS, NY, PE, SFDH, UC, WH). ee ne (Lévl. & Vaniot) Gagnep. 1542 (A, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cayratia pseudotrifolia W. T. Wang 474 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1494 (A, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1697 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). A, 70 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 Parthenocissus henryana (Hemsley) Diels & Gilg 1694 (A, CM, HIB, KUN, S, NY, PE, SFDH, UC, WH). Parthenocissus himalayana (Royle) Planchon 1036 (A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Tetrastigma lineata Diels & Gilg 20 A, CM, HIB, NAS, UC). Tetrastigma eae Planchon 1621 (a, B, KUN, NA, PE, SFDH, UC). Tetrastigma ie (Wallich) eee var. pilosum Gagnep. Notul. Syst. Paris 1: 323. 1 1596 (A, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Vitis betulifolia Diels & Gilg* 192 (A, HIB, NAS, PE, UC); 23/ (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 334 (A, HIB, KUN, NA, NAS, NY, PE, UC); 13/3 (A, CM, HIB, KUN, NA, NAS NY, PE, SFDH, UC, WH). Vitis piasezkii Maxim. var. pagnuccii (Romanet ex Planchon) Rehder, J. Arnold rbor. 3: 223. 1922. 226 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 650 (A, CM, E, HIB KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1049 (A, CM, HIB, KUN, NA, NAS, NY PE, SFDH, UC, WH); 1345 (A, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /355 (A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1734 (A, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ‘TILIACEAE Grewia biloba D. Don 1647 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Tilia chinensis Maxim. 9S (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1307 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 13/9 (A, CM, E, HIB, KUN, KYO NA, NAS, NY, PE, SFDH, UC, WH); /533 (A, CM, E, HIB, KUN, NA, NAS, NY, PE SFDH, UC, WH Tilia oliveri Szysz. var. cinerascens Rehder & Wilson, Pl. Wilsonianae 2: 367. 1119 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/20 (A, CM E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC ACTINIDIACEAE Actinidia callosa Lindley* ie , CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /7/0 (A, CM, PE, SFDH, UC dean fee Tinley var. henryi Maxim. Trudy Imp. S.-Petersburgsk. Bot. : 1890. 249 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Actinidia chinensis Planchon A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 749 (A, CM, E, HIB KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 806 (A, HIB). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 71 Actinidia chinensis Planchon var. hispida C. F. Liang, Acta Phytotax. Sinica 13: 33. 1975 237 (HIB). Actinidia polygama Pe & pane — (A, CM, HIB, KUN, , NY FDH, UC). ee polygama Gicb. & Zuen) Movie var. lecomtei (Nakai) H. L. Li, J. Arnold Arbor. 33: 22. 1952. 339 a CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Actinidia tetramera Maxim. 8 , E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Ree e an Komarov , KYO, NA, NAS, NY, PE, SFDH, UC, WH). fe es Veen. Baillon 6) A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 944 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1532 (A, CM, E, HIB, KUN, NA, NAS, NY, P H, UC, WH). Clematoclethra lanosa Rehder* 1752 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). THEACEAE Camellia cuspidata (Kochs) Veitch 801 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 2037 (A, HIB, NA, NAS, PE, UC). Eurya alata Kobuski 1447 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1489 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Eurya brevistyla Kobuski 521 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Eurya eae Dunn , E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Eurya pane Chan 2004 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Stewartia sinensis Rehder he Wils 1504 (A, CM, HIB, NY, PE); 1568 ie CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). GUTTIFERAE Hypericum ascyron L. 363 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 79/ (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, Hypericum cf. Atienuahinn Choisy 108 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Hypericum patulum Thun 2021 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Hypericum perforatum L. 396 (A, HIB, KUN, NAS, NY, PE, UC); 1050 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 72 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 VIOLACEAE Viola grypoceras A. Gray var. pubescens Nakai*, Bot. Mag. Tokyo 36: 55, 89. 1502 (A, HIB). Viola prionantha Bunge 1645 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). Viola stewardiana W. Becker 1493 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). FLACOURTIACEAE Carrierea calycina Franchet 5 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). STACHYURACEAE Stachyurus chinensis Franchet 554 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1078 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1424 (A, CM, E, HIB, KUN, NA, NAS, NY, PE WH). Stachyurs pees Franchet var. /atus H. L. Li, Bull. Torrey Bot. Club 70: oe (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 194 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 600 (A, HIB, KUN, NAS, NY, PE, SFDH, uc, WH); 1128 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). BEGONIACEAE Begonia evansiana Andrews 493 (A, HIB). Begonia sinensis A. DC. 498 (A, HIB, KUN, NA, PE, UC); 1642 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Begonia sp. 2012 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ELAEAGNACEAE Elaeagnus henryi Warb. 2077 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Elaeagnus lanceolata War 95 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 145 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 459 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1030 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Elaeagnus umbellata Thunb. 1182 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/96 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 73 LYTHRACEAE Rotala indica (Willd.) Koehne 2084 (A, HIB). NYSSACEAE Davidia involucrata Baillon var. vilmoriniana (Dode) Wangerin in Engler, Pflanzenr. IV. 220a(Heft 41): 19. 1910 1473 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). ALANGIACEAE Alangium chinense (Lour.) Harms 452 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC), 160/ (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 163/ (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC Alangium an (Sieb. & Zucc.) Harms 604 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1738 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH) Alangium sinicum (Nakai) S. Y. Hu, Spongberg, & Z. Cheng, comb. nov.* BasionyM: Marlea sinica Nakai, Fl. Sylv. Koreana 17: 29. 1928. 1017 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1096 (A, CM, E, HIB, KUN, KYO, NA, NAS, a PE, SFDH, UC, WH); 1703 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, Two species sae A. chinense (Lour.) Harms and A. platanifolium (Sieb. & Zucc.) Harms, have been credited to the Shennongjia Forest District (Anonymous, 1980) and Hubei Province (Bloembergen, 1939, map on p. 163). Our field activities confirmed the presence of these species in the Shennongjia area but also revealed a third taxon that appeared distinct from both A. chinense and A. platanifolium. Subsequent studies and a comparison of the Shennongjia collections with specimens in the Harvard University Herbaria confirm this third taxon, which (based on specimens at A and GH) is not confined to western Hubei but is widely distributed in China in Shaanxi, Shandong, Shanxi, Henan, Jiangsu, Hubei, Guizhou, and Yunnan provinces. S. Y. Hu, in earlier unpublished studies utilizing specimens at A and GH also recognized this third species and stated that Nakai (1928) had provided it with a name under the genus Marlea. The appropriate new combination under Alangium is made here. The following key, based primarily on spec- imens collected from western Hubei but supplemented by other Chinese specimens, will aid in distinguishing the three species. 3 Key to the Species of Alangium in the Shennongjia Forest District . Peduncles 1.3—1.6 cm long; pedicels 1.3-1.6 mm long; corollas 0.8-1.9 cm long; styles pubescen nt, 2. Ovaries and fruits 1-celled; branchlets and petioles sparingly daa etn leaf surfaces glabrous except for tufts of hairs in vein axils.......... uni. 74 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 2. Ovaries and fruits 2-celled; branchlets, petioles, and leaf surfaces pubescent. inense. bee araeesecgs dat ien ts tals tasting a get. dndoeniees Daan a ieteote es eee e eee ota ee cee: .¢ 1. Peduncles 3.2-3.5 cm long; pedicels 2—3.8 cm long; corollas 2.8-3. , cm long: styles A. platanifolium. ONAGRACEAE Circaea alpina L. subsp. imaicola (Ascherson & Magnus) Kitamura, Fl. Afghan- istan, 279. 1960. 28 (A, HIB). Circaea erubescens Franchet & Sav. 850 (A, CM, E, HIB, KUN, MO, NA, NAS, NY, PE, SFDH, UC, WH). Circaea glabrescens (Pampan.) Hand.-Mazz.* 133 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 4/9 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Circaea mollis Sieb. & Zucc. 85] (A, CM, HIB, PE); 1636 (A, CM, HIB, KUN, NA, PE). Circaea repens Wallich ex Ascherson & Magnus* 1157 (A, CM, E, HIB, KUN, MO, NA, NAS, NY, PE, SFDH, UC, WH). Epilobium cephalostigma Hausskn. 5 (A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 119 (A, CM, HIB, KUN, NA, NAS, CM, HIB, KUN, NA, NAS, NY, PE). Epilobium parviflorum Schreber Il A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1832 (A, CM, E, HIB, KUN, MO, NA, NAS, NY, PE, SFDH, UC, WH). Epilobium sinense Lév1.* PE, UC). Epilobium aff. nepalense Hausskn.* 318 (A, 24 (A, HIB, KUN, NAS, PE); 3/9 (A, CM, E, HIB, oS KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 380 (A, HIB, NAS, NY, PE, UC); 599 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); (A, CM, E, HIB, KUN, KYO, MO, NA, 2 AS, NY, PE, SFDH, UC, WH); 10/3 (A, CM, HIB, KUN, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1831 (A, CM, E, HIB, KUN, MO, NA, NAS, NY, PE, SFDH, UC, WH). ARALIACEAE Acanthopanax giraldii Harms 115 (A, CM, E, HIB, NAS, PE, UC); 160 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Acanthopanax henryi (Oliver) Harms 1195 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1338 (A, CM, E, HIB, WH). Acanthopanax leucorrhizus oo Harms var. fulvescens Harms & Rehder, Pl. Wilsonianae 2: 558. 1 185 (A, CM, HIB, KUN, NY, PE, UC); 367 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 905 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1392 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 75 Acanthopanax leucorrhizus (Oliver) Harms var. scaberulus Harms & Rehder, Pl. Wilsonianae 2: 558. 1916. 975 (A, HIB, NY, PE, UC) Acanthopanax Ee is Harms ex Diels 109 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 219 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/29 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1757 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1797 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Acanthopanax trifoliatus (L.) Merr 1102 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Aralia chinensis L. 636 (A, HIB); 1123 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Aralia echinocaulis sane psi - 38] (A, CM, HIB, NA, UC). Hedera nepalensis K. nee var. sinensis (Tobler) Rehder, J. Arnold Arb. 4: 250. 1923. 702 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC, WH); 1083 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH) Nothopanax davidii (Franchet) Harms ex Diels 299 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 675 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /0/5 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1088 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC WH); 1627 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 2042 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, U Panax pseudo-ginseng Wallich Pal wean between vars. bipinnatifidus (Seem.) H. L. Li (Sargentia 2: 118. 1942) and japonicus (C. A. Meyer) Hoo & Tseng) 314 (A, HIB). Panax pseudo-ginseng Wallich var. japonicus (C. A. Meyer) Hoo & Tseng, Acta Phytotax. Sinica 11: 437. 1973 1803 (A, HIB, NAS, NY, PE). UMBELLIFERAE Angelica pubescens Maxim. 1487 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Angelica sp. 256 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1580 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Bupleurum chinense DC. 1641 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Bipleurum he Wallich ex DC. var. franchetii H. Boiss. Bull. Soc. Bot. France 53: 425. 1224 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Bupleurum longicaule Wallich ex DC. var. giraldii Wolff in Engler, Pflanzenr. IV. 228(Heft 43): 123. 1910. 111 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 254 (A, CM, E, 76 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 949 (A, CM, HIB, KUN, FDH, UC roan meant Taree var. porphyranthum Shan & Y. Li, Acta Phytotax. Sinica 12: 270. 404 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 903 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 12/3 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Bupleurum need aes 6 CM , NAS, PE, SFDH, UC, WH). Bay Japonica Hasek. Gas C pie eee (L.) DC. var. japonica (Hassk.) , Bot. Mag. Tokyo 22: 175. 1908.) py mn — 549 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Heracleum acuminatum Franchet 1774 CM, HIB, KUN, NA, NAS, NY, PE, UC); /S/0 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Heracleum moellendorffii Hance* 1253 (A, HIB, NA, NAS, PE, UC); 18/3 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, , E, H 69 Heracleum yungningense Hand.-Mazz. 1254 (A, HIB) Ligusticum daucoides Franchet* 948 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). a sinense Oliver dd , CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /38 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /S// (A, CM, E, HIB, KUN, NA, NAS, E, S WH). Melanosciadum pimpinelloideum H. Boiss.* 417 (A, HIB, NY, PE, UC); 686 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC); 1076 (A, HIB, NAS, NY, PE, UC); 18/4 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Nothosmyrnium japonicum Miq. var. sutchuensis H. Boiss.* Bull. Soc. Bot. France 61: 349. 1909 654 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /20/ (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1242 (A, CM, HIB, KUN, NAS, PE, UC). Oenanthe dielsii H. Boiss. 1458 (A, CM, HIB, KUN, NAS, NY, PE, SFDH, UC). Peucedanum praeruptorum Dunn 140 (us); 464 (A, HIB, PE, UC). Pimpinella arguta Diels 569 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /252 (A, HIB). Pimpinella diversifolia DC. 146 (A , HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Pimpinella diversifolia DC. var. stolonifera Hand.-Mazz.* Symb. Sinicae 7: 714. 1933 699 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 77 Pleurospermum giraldii Diels* A, HIB, PE, UC). Sanicula orthacantha S. Moore 506 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). CORNACEAE Aucuba chinensis Bentham* 615 (A, HIB, NA); 1676 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1756 (A, CM, ee KUN, NAS, NY, PE, UC); /90/ (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, Cornus ee Wangerin (syn.: Macrocarpium ee (Wangerin) Hutch.) 1696 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, W Cornus controversa Hemsley 19 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 644 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /0/4 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cornus kousa Hance (syn.: Dendrobenthamia japonica (DC.) Fang var. chi- nensis (Osborn) Fang, Acta Phytotax. Sinica 2: 105. 224 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 784 ee HIB); 1244 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 13/6 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cornus: kousa auc var. angustata Chun, Sunyatsenia 1: 285. 1934. (syn.: ngustata (Chun) Fang, Acta Phytotax. Sinica 2:95. 1953.) 203 et CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cornus macrophylla Wallich 369 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 883 (A, CM, E, HIB KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Cornus paucinervis Hance 162 CM KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Helwingia Pere ee i A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1897 (A, CM, HIB, KUN, Hevinga ae (Thunb) Dietr. var. japonica 195 (A, CM, HIB, KUN, NA, NAS, NY, PE); 355 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, Helwingia japonica (Thunb) Dietr. var. hypoleuca Hemsley ex Rehder, PI. Wilsonianae 2: 570. 191 107] (A, HIB, PE, UC). Helwingia japonica (Thunb.) Dietr. (intermediate between vars. japonica and hypoleuca) 862 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). CLETHRACEAE Clethra fargesii Franchet 485 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 773 (A, CM, E, HIB, KUN, KYO, ie: NA, NAS, NY, PE, SFDH, UC, WH); 1276 (A, CM, HIB, KUN, NA, 78 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 NAS, NY, PE, SFDH, UC, WH); 1469 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH PYROLACEAE Chimaphila japonica Miq. 912 (A, HIB). Monotropa hypopithys L. 5867 (A, CM, HIB, NAS, NY, PE, UC). Pyrola deigat Andre 1250 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ERICACEAE Enkianthus chinensis Franchet 626 (HIB); 718 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1132 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /3// (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Ee lee skin Sea ease 2014 (A WH). Lyonia oval Walch) Drage var. aye (Sieb. & Zucc.) Hand.-Mazz. Symb. Sinicae 7(4): 7 936. 1394 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Pieris formosa (Wallich) D. Don 769 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/474 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Rhododendron argyrophyllum Franchet subsp. hypoglaucum (Hemsley) Cham- NAS, NAS, NY, PE, SFDH, UC, WH); 1758 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, Giodedention augustinii Hemsle , CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 739 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC); 1748 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, uc, WH); 1780 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rhododendron concinnum Hemsley S75 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rhododendron fargesii Franchet 124 (A, HIB, KUN, NA, NAS, PE, UC); 942 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, H). soeare fortunei Lindley subsp. discolor (Franchet) Chamberlain, Notes ot. Gard. Edinburgh 37: 330. 1979. 2040 (A HIB). Rhododendron maculiferum Franchet A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 863 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 943 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 79 se eibalas mariesii Hemsley & Wilson 787 (A, HIB, PE). Rldodendron ee Turcz. 208 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). ee simsil Biacken 774 (A, HIB); 1425 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1932 (A, HIB, PE). Rhododendron sutchuenense Franchet 1231 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Vaccinium henryi Hemsley 1497 (A, CM, E, HIB, KUN, KYO, NAS, NY, PE, SFDH, WH). Vaccinium japonicum Mia. var. sinicum Dies Render J. Arnold Arbor. 5: 56. 1924. (syn.: Hugeria ane (Lévl.) H 730 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/475 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). MYRSINACEAE Ardisia crenata Sims 1455 (A, HIB); 1620 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1677 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). iy japonica (Thunb.) Blume 1926 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH). Mpyrsine africana L 479 (A, HIB, NA, NY, PE). PRIMULACEAE Androsace henryi Oliver 1456 (HIB). Lysimachia christinae Hance 359 (A, HIB). Lysimachia clethroides Duby 727 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1503 (A, HIB, NA, NAS, NY, PE, UC). Lysimachia congestiflora Hemsley 1093 (A, HIB, KUN, NA, NAS, NY, PE, UC). Lysimachia stenosepala Hemsley 200 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 570 (A, HIB, PE, UC). Primula ovalifolia Franchet 1472 (A, CM, HIB, NA, NAS, PE, UC). EBENACEAE Diospyros lotus L. 1478 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1605 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 80 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 SYMPLOCACEAE Symplocos anomala Brand* 1693 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Symplocos lancifolia Sieb. & Zucc. 2019 (A, CM, HIB, PE). Symplocos paniculata (Thunb.) Miq. 744 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 12/7 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /530 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /S5/ (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC, WH); 1886 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC). STYRACACEAE Styrax hemsleyana Diels* 348 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/33 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1390 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Styrax japonica Sieb. & Zucc. 763 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1484 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). OLEACEAE Forsythia suspensa (Thunb.) Vahl 694 (HIB); 1707 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Fraxinus paxiana Lingelsh.* 1 A, CM, HIB, PE). Jasminum floridum Bunge E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Jasminum lanceolarium Roxb. I A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 19/0 (A, CM, E, HIB, A, NAS, NY, PE, SFDH, UC, WH). Jasminum urophyllum Hemsley 2078 (A, HIB, KUN, NA, NY, PE, UC). Ligustrum acutissimum Koehne 242 (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 340 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /03/ (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 12/16 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Ligustrum henryi Hemsley 1911 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Ligustrum obtusifolium Sieb. & Zucc.* 1735 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Ligustrum quihoui Carriére 1444 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Osmanthus armatus Diels 9] (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 692 (A, HIB). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 81 Osmanthus fragrans Lour.* (collected from a plant in cultivation) 1575 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH) Syringa reflexa Schneider* 264 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 908 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 955 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH LOGANIACEAE Buddleja albiflora Hemsley 257 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). meen davidii Franchet 671 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 14/8 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). GENTIANACEAE Gentiana panthaica Burkill* 1 , HIB, KUN, NA, NAS, NY, PE, ie UC, WH). none poate eee (H. Sm.) M , HIB , KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Poe se D. ae var. elliptica oo A, CM, E, HIB, KUN, aan NAS, NY, PE, SFDH, UC, WH); 87/ (A, CM, E, HIB, ' MO, NA, NAS, NY, Halenia a. ae var. A eas Linn. Soc., Bot. 26: 141. 683 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 856 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 12/5 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lomatogonium bellum (Hemsley) H. Sm.* 174 (A, CM, HIB, KUN, NY, PE, SFDH, UC); 946 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Swertia bimaculata (Sieb. & Zucc.) Hooker & Thomson 180 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 228 (A, CM, HIB, KUN, re , NY, PE, SFDH, UC, WH); 3/7 (A, CM, HIB, KUN, NAS, NY, PE, UC); 1958 ae M, HIB, NY, PE). Swertia punicea Hemsley* 1595 (A, CM, HIB, KUN, Nes NAS, NY, PE, UC). Swertia tetragona Edge I A, CM, HIB, KUN, , NY, PE, , WH). Collection no. esos was Senne penne as Swertia punicea; the two plants are quite different. Tripterospermum affine (Wallich) H. Sm. 9 (A, CM, HIB, NAS, NY, PE); 143 (A, CM, HIB, KUN, PE, UC); 566 (A, HIB, NA, PE); 1927 (A, CM, HIB, NA, NAS, NY, PE). 82 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ASCLEPIADACEAE Cynanchum auriculatum Royle ex Wight 1225 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cynanchum inamoenum ieee ) Loes. 634 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Metaplexis hemsleyana Oliver (syn.: M. sinensis (Hemsley) Hu) 439 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1420 (A, CM, HIB, NAS, NY, PE); 1649 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). CONVOLVULACEAE Cuscuta japonica Choisy (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1570 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Porana racemosa Roxb. 1443 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). BORAGINACEAE Cynoglossum zeylanicum (Vahl) Thunb. 379 (A, HIB, NAS, NY, PE, UC); 790 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). VERBENACEAE Callicarpa bodinieri Lévl. var. bodinieri 408 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /68/ (A, CM, E, HIB, UN, NA, PE, SFDH, UC, WH); 2007 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, voi 206 0 (a CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 7086 (A, CM, E, , KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, CG ea hod nieri Lévl. var. giraldii (Hesse ex Rehder) eee J. Arnold Arbor. 15: 322. 1934 846 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Caryopteris incana (Thunb.) Miq. 1656 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Clerodendrum bungei Steudel 1928 (A, CM, HIB, NAS, PE, UC). Clerodendrum trichotomum Thunb. var. trichotomum 40] (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 602 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1878 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Clerodendrum trichotomum Thunb. var. fargesii (Dode) Rehder, Pl. Wilson- ianae 3: 37 916 &48 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 83 LABIATAE Clinopodium eo oa (Vaniot) C. Y. Wu & Hsuan*, Observ. Fl. Hwang- shanicum, 168. 1965 Sauer incorrect in Index Kewensis). 1968 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Elsholtzia aiaie (Thunb. : Pande 1135 (A, HIB, KUN, NA, NY, PE); 15/8 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1962 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Elsholtzia cypriani (Pampan.) S. Chow*, Observ. Fl. Hwangshanicum, 170. 1965 (citation incorrect in Index Kewensis). 1685 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Elsholtzia flava Bentham 1945 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Elsholtzia fruticosa (D. Don) Rehder 220 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1746 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, Kinostemon ornatum (Hemsley) cae (syn.: Teucrium ornatum Hemsley) 0 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC ee gta (Lour.) S. Y. Hu (incl. L. japonicus Hylander) 746 E, , KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Mosla ee han. )C.Y. Wu & H.W. Li 1965 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Origanum vulgare L. 568 (A, HIB, NA, PE); 866 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Phlomis umbrosa Turcz. 113 (A, HIB, KUN, NA, PE, UC); 352 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1866 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rabdosia excisoides (Sun ex C. H. Hu) C. Y. Wu & H. W. Li 142 (A, CM, HIB, KUN, NA, sean NY, PE, SFDH, UC, WH); 206 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, meres nervosa (Hemsley) C. Y. we & H.W. Li A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). oe maximowicziana Hemsley 182 (A, CM, HIB, KUN, NA, NAS, PE, UC); 376 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). SOLANACEAE Lycianthes lysimachioides (Wallich) Bitter 617 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Physalis alkekengi L. var. franchetii (Mast.) Makino, Bot. Mag. Tokyo 22: 34. 1908 475 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1550 (A, HIB, NY, PE). Scopolia sinensis Hemsley (syn.: Atropanthe sinensis (Hemsley) Pascher) 483 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). 84 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Solanum lyratum L. 1576 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Solanum eis L. 429 (A, Cc IB NA, NAS, NY, PE, SFDH, UC, WH). eee er ee Hemsley (A, CM, HIB, NA, NAS, PE). SCROPHULARIACEAE ae heterophyllum Wallich* 166 (A , E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH Melam won cee Maxim. var. obtusifolium (Bonati) Hong, Fl. Reip. Pop. Sinicae 67(2): 3 1979. 853 (A, CM, E, HIB, KUN, KYO, NA, NAS, PE, SFDH, UC, WH). Mimulus tenellus Bunge 1770 (A, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). se aaa holocalyx Hand.-Mazz 8 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). a. resupinata L. 567 (A, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /2/4 (A, HIB, KUN, NA, NAS, 9 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC, WH); 259 (A, HIB, KUN, NA, NAS, NY, PE, UC); 868 (A, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Scrophularia henryi Hemsley 177 (A, HIB, PE). Veronicastrum caulopterum (Hance) Yamazaki 2008 (A, HIB, KUN, NA, NY, PE, UC). Unidentified Scrophulariaceae 169a (A) OROBANCHACEAE Orobanche coerulescens Stephan 9OT (HIB GESNERIACEAE Hemiboea subcapitata C. B. Clarke CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Lystonotis naeonE Maxim 7S CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /705 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Paraboea sinensis (Olives) B. L. Burtt*, Notes Roy. Bot. Gard. Edinburgh 38: 471. 1980 1715 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 85 ACANTHACEAE Asystasiella chinensis (S. Moore) E. Hossain 16 A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). PHRYMACEAE Phryma leptostachya L. var. asiatica Hara, Enum. Spermatophyt. Jap. 1: 297. 1948. 762 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). RUBIACEAE Anotis hirsuta (L. f.) Boerl. 2006 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). E he pene henry Oliver 2 M, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Leptodermis aff. oblonga Bunge 1700 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH) Collection no. 1/700 is tentatively assigned to Leptodermis oblonga Bunge. There appears to be considerable confusion concerning the Chinese species, especially with regard to fruiting material, and the genus is in need of careful study both in the field and in the herbarium. Leptodermis wilsonii Hort. ex Diels* 581 (A, HIB, KUN, NA, PE, UC). Ophiorrhiza japonica Blume 509 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC, WH); 1/087 (A, CM, HIB, KUN, NAS, NY, PE, SFDH, UC, WH); 1449 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Paederia scandens (Lour.) Merr. 511 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1072 (A, HIB, KUN, NA, 6 (A, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). CAPRIFOLIACEAE ao engleriana (Graebner) Rehder A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 653 (A, CM, HIB, KUN, NA, Te NY, PE, SFDH, UC, WH). Abelia sp lee tee CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). as floribunda Max A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1684 (A, HIB). Lonicera gynochlamydea Hemsley (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 225 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 4/1 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 642 (A, HIB, NA, NAS, PE, UC); 680 (A, CM, E, HIB, KUN, 86 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 NA, NAS, NY, PE, SFDH, UC, WH); 1040 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1073 (A, CM, HIB, KUN, NA, NAS, NY, P DH, UC, WH); /281 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1467 (ate 1762 (A, CM, HIB, NA, NAS, PE); 1825 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lonicera henryi Hemsley* 240 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 342 (A, CM, E, HIB, YO, NA PE, SFDH, UC, WH); 736 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1046 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1095 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Lonicera japonica Thun 443 (A, HIB). Lonicera koehneana Rehder 315 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 345 (A, HIB, NA, PE); 394 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 907a (A, NA, UC); 1802 (A, HIB, PE, UC). Collection no. 907a was distributed as Lonicera koehneana Rehder var. longipes Rehder. This name was apparently never published; there is no type at A, and the single specimen bearing this name was not determined by Rehder. Moreover, the varietal name does not appear in Rehder’s published writings on Lonicera. Lonicera longa Rehder* 260 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Lonicera maackii Maxim. 44] (A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 47/ (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1442 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/452 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1584 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 16/2 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Lonicera nervosa Maxim. 157 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 966 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Lonicera cf. nervosa Maxim. 25] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lonicera pileata Oliver 705 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1099 (A, HIB, NA, NAS, PE, UC); 1908 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 194] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). any similis Hemsley 426 (A, HIB, NY, PE, UC). Lone papleiee Franchet , E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Lonicera ae Maxim. ote no. 907 represents a mixed collection, part of which (no. 907a) is Lonicera koehneana Rehder. The aol and the distribution of the specimens in Chinese herbaria are not know 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 87 Lonicera tragophylla Hemsley 611 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Lonicera trichopoda Franchet* 1743 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). ee adnata Wallich ex DC.* (syn.: S. schweriniana Rehder) 38 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH) eae chinensis pees 1108 (A, CM, E, HIB, , KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). h Triosteum Fee ce Wallic 0 (A, HIB, NA, NAS, NY, PE, UC); 175 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, FDH, UC, WH); A, CM, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Viburnum betulifolium Batalin 557 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1043 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /098 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 14/5 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Viburnum cylindricum Buch.-Ham. ex D. Don 4 A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 14/4 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1930 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, U Viburnum ee Wallich 32 (A, CM, HIB, KUN, NA, NAS, PE, UC); 947 (A, HIB, NA, PE). Viburnum flavescens W. W. Sm.* 1035 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). Viburnum foetidum Wallich var. rectangulatum (Graebner) Rehder* in Sargent, Trees Shrubs 2: 114. 1908 14] ] (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 2002 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Viburnum formosanum (Maxim.) Hayata* 434 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Viburnum hupehense Rehder subsp. hupehense* 14 (A, HIB, NA, PE); 73 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 190 (A, HIB, NA, PE); 24] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 390b (a; see V. lobophyllum for note on distribution of this collec- tion); 607 (A, HIB); 630 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1199 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1325 (A, HIB, NA, NY PE); 1577 (A, HIB, NA, NAS, PE, UC); 1/733 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1766b (A, NA). Collection no. /766 represents a mixed collection, part of which (/766a) is Viburnum ovalifolium Rehder. The identity and the distribution of the specimens in Chinese herbaria are unknown. Viburnum hupehense Rehder subsp. septentrionale Hsu*, Acta Phytotax. Sinica 11: 77. 1966 98 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC); 1/286 (A, HIB, NA, NAS, PE). Viburnum ichangense Rehder var. ichangense 1219 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 133/ (A, CM, E, HIB, 88 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 KUN, NA, NAS, NY, PE, SFDH, UC, WH); /889a (A, NA, NY); 2038 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Viburnum ichangense Rehder var. atratocarpum (Hsu) T. R. Dudley & S. C. un, comb. nov.* BasionyM. Viburnum erosum subsp. ichangense (Hemsley) P. S. Hsu var. atratocarpum P. S. Hsu, Acta ee Sinica 13(1): 127. 1975. 446 (A, CM, E, HIB, KUN, KYO, NA, NAS, PE, To UC, WH); 775 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, ne 889b—d (A, CM, NA, UC). Calleelion no. 1/889 represents a mixed collection, part of which (/88 9a) is Viburnum ichangense Rehder var. ichangense. The identity and the dis- tribution of the collections in China are not known i lobophyllum Graebner* 250 (A, CM, HIB, KUN, NA, NAS, NY, PE); 390a (A, CM, NA, NY, PE, UC); 874 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 14/6 (A, HIB, KUN, NA, NAS, NY, PE, UC). Collection no. 390 represents a mixed collection and was renumbered 390a and 390b (Viburnum hupehensis subsp. hupehensis). The identity and the distribution of the sheets in Chinese herbaria, other than the sheet at PE, are not known; however, HIB, KUN, NAS, SFDH, and wu all received specimens under no. 390. Viburnum ovatifolium Rehder 1124 (A, CM, HIB, KUN, NA, PE, SFDH, UC); 1766a (A, CM, NA, NY, WH); 1588 a CM, E, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Viburnum Tae Thunb. forma tomentosum ‘Thunberg) Rehder, J. Arnold rb ree 77. ade 1658 (A re pip naan Hemsle 462 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1069 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /4/3 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 170] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1904 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH Viburnum rhytidophyllum Hemsley 82 (A, HIB); 152 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 2/0 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 649 (A, HIB, NA, PE); 1002 UN, NA, NAS, NY, PE, SFDH, UC, WH). 247 (A, HIB); 1335 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1887 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1963 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 2043 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Viburnum sympouiale Graebner 1209 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1287 (A, CM, HIB, 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 89 KUN, NA, NAS, NY, PE, UC); 1334 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Viburnum utile Hemsley 1412 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1581] (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 16/8 (A, CM, HIB, NA, NAS, NY, PE). Viburnum veitchii C. H. Wright* 374 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); S61 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1052 (A, CM, HIB, KUN, NA, NAS, Y, PE, SFDH, UC, ; ee pa Thunb. var. sinica (Rehder) Bailey, Gentes Herb. 2: 49. 1929. 460 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 750 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /95/ (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). V ALERIANACEAE Patrinia angustifolia Hemsley 81 (A, CM, HIB, PE). Patrinia monandra C. B. Clarke 25 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 573 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC); 755a (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1222 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). DIPSACACEAE Dipsacus asper Wallich 756 (A, CM, HIB, NAS, NY, PE, UC). Dipsacus japonicus Miq. ‘ a CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 64 (A, CM, E, HIB, KUN, NAS, NY, PE, SFDH, UC, WH). a epee glandulifera Wallich 1349 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). (CUCURBITACEAE Gynostemma cardiospermum Cogn. ex Oliver (distributed as Gynostemma pentaphyllum (Thunb.) Makino, which has free or only slightly united fila- ments; our specimens have united filaments) 538 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Hemsleya eee Cogn. ex Forbes & Hemsley te CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, wh); 641 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). nee ee Cogn. eee CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); I251 (A, HIB, K NAS, NY, PE, SFDH, UC, WH). ene hens eee 62 (A, CM, E, HIB, KUN, NA, NAS NY, PE, SFDH, UC, WH). Thladiantha a, Cogn. 1323 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). 90 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 gi nudiflora Hemsley 637 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 794 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1477 (A, CM, E, HIB, KUN, NA, NAS, NY PE, SFDH, UC, WH 5 CAMPANULACEAE Adenophora axilliflora Borbas : A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 52 (A, CM, E, HIB, KUN, AS, NY, PE, SFDH, UC, WH); 335 (A, HIB, PE). Adenophor are Hemsley (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 20] (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1/22 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1/54 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Adenophora wilsonii Nannf. 1280 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Campanula punctata Lam. 99 (HIB), 179 (HIB); 324 (A, HIB, NY, PE); 360 (A, CM, HIB, KUN, NA, NAS, PE, UC). Campanumoea javanica Blume var. japonica (Maxim.) Makino, Bot. Mag. Tokyo 22: 155. 1908. (syn.: C. maximowiczii Honda) 455 (A, HIB, NY, PE, UC, WH); 2009 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, C, WH). Codonopsis pilosula (Franchet) Nannf. 1057 (A , HIB). Codonopsis tangshen Oliver 7 (A, HIB, KUN, NA, NAS, PE, UC); 186 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 234 (A, CM, HIB, KUN, NA, NY, PE, SFDH, UC); 188] (A, CM, HIB, , KYO, NA, NAS, NY, PE, SFDH, UC, WH). Praia eee er (Lam.) A. Br. & Ascherson 64 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). COMPOSITAE Adenocaulon himalaicum Edgew. 548 (A, HIB, NA, NAS, NY, PE); 1042 (A, HIB, NY, PE, UC). Ainsliaea gracilis Franchet 1970 (A, CM, HIB, NA, NAS, PE, ee a (A, HIB, NA, NY, PE, UC). Ainsliaea triflora ones aan am. a oe 1598 (A, CM, E, HIB AS, NY, PE, SFDH, UC, WH Anaphalis Reine L). Bente. fi Hooker subsp. japonica (Sch. -Bip.) Kitamura, Acta Phytotax. Geobot. 5: 148. 1936. (distributed as 4. margar- an var. lies (Sch.-Bip.) Makino) , CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Pe iar sinica Hance 44 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 963 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 91 Artemisia annua L. 757 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1567 (A, HIB, NAS, NY, PE, UC); 1643 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Artemisia argyi Lévl. & Vaniot 1520 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, Artemisia argyi Lévl. & Vaniot var. incana (Maxim.) Sees eres Giorn. Bot. Ital. 36: 451. 1930. 1597 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Artemisia alae pee ig , NA, NAS, NY, PE, SFDH, UC, WH). Artemisia eres Wallich 323 (A, CM, HIB, KUN, NA, NAS, PE, UC). Artemisia subdigitata Mattt. 59 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Aster ageratoides Turcz. var. ageratoides 122 (A, HIB, NAS, NY, PE, UC); 1277 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Collection no. 122 was distributed as 4. ageratoides var. scaberulus (Miq.) Ling, and collection no. 1277 as A. ageratoides var. micranthus Ling; both varietal names are unpublished. Aster ageratoides Turcz. var. laticorymbus (Vaniot) Hand.-Mazz. Acta Horti Gothob. 12: 214. 1 587 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 623 (A, CM, HIB, NA, NAS, NY, PE, SFDH, UC); 1857 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Aster ee Oe J Hand.-Mazz. 1061 (A, , KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Aster achalie Guanes 1619 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Bidens parviflora Wil 1662 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Bidens pilosa L. nel (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1405 (A, CM, E, HIB, , NA, NAS, NY, PE, SFDH, UC, WH Oe tripartita L. 1256 (A, CM, E, HIB, KUN , NY, PE, SFDH, UC, WH). Cacalia ainsliaeflora ee) "Hand. -Mazz. (syn.: C. leucanthema (Dunn) Ling) 1162 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Cacalia hastata L. var. ae Ledeb. FI. ‘Altaica 4; 52. 1833. 878 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Cacalia pie tes (Franchet) Hand.-Mazz. 3 , E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cacalia een (Dunn) Hand.-Mazz. 6 A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Cacalia sinica Lin 1783 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 22 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Cacalia tangutica (Franchet) Hand.-Mazz. 1 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 670 (A, CM, E, HIB, K KYO, MO, NA, NAS, , SFDH, el WH Cacalia vespertilio ae, a rH i 1687 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). Carpesium cernuum L 6 HIB, PE, UC). Carpesium divaricatum Sieb. & Zucc. 535 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC). Carpesium pease Winkler 487 (A, CM, E, HIB, KUN, KYO, MO , NAS, NY, PE, SFDH, WH). Paar eae boreale (Makino) Making® (syn.: Earn ce boreale (Makino) Ling) 895 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1329 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); /7 A, CM, E, HIB, KUN, NA, NAS, NY, eens enn indicum L. (syn.: Dendranthema indica (L.) Moulins) 132 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 959 (A, CM, E, HIB, A, NAS, NY, PE, SFDH, UC, WH). Cirsium fargesti ume i 54 (A, CM, E, HIB, KUN, , MO, NA, NAS, NY, PE, SFDH, UC, WH). Cirsium henryi (Franchet) Dies 938 (A, CM, E, HIB, KUN NAS, NY, PE, SFDH, UC, WH). Cirsium lineare (Thunb. Sch. ‘Bip var. intermedium (Pampan.) Petrak, Repert. Spec. Nov. Regni Veg. 43: 276. 1938 740 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Conyza canadensis (L.) Cronq. A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Erigeron elongatus Ledeb. 33 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC Eupatorium chinense L. ee E. japonicum ae ) eo CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 15/9 (A, CM, E, HIB, , NY, PE, SFDH, UC, WH). Galinoga ae Cav 431 CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Gynura pee Bentham 489 (A, CM, HIB, KUN, NAS, NY, PE, SFDH, UC, WH). Kalimeris een (L.) Sch.-Bip 1105 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Lactuca Ae (Wallich) De 201 , HIB, NY, PE). Leontopodium japonicum Miq. A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 46/ (A, CM, HIB, NAS, PE, UC); 70] (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Ligularia hodgsonii Hooker 362 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 93 Ligularia veitchiana (Hemsley) Greenman 1205 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Pertya sinensis Oliver 15 , CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). Picris APR AaTTEER IO subsp. japonica (Thunb.) Krylov, Fl. Altai, WH. 1904. I , HIB, KUN, NA, NAS, NY, PE, UC); 354 (A, CM, HIB, KUN, NA, NAS, NY, PE, uc): 951 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH Prenanthes tatarinowli Maxim. 681 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 968 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Rhynchospermum verticillatum Reinw.* 423 (A, HIB, PE, UC). Saussurea cordifolia Hemsley 153 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1765 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, H, UC, : Saussurea sees (DC) C. B. Clarke" A, CM, HIB, KUN, NAS, NY, PE, SFDH, UC, WH). Saussurea silvestrii Pampan. 400 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Saussurea veitchiana Drumm. & Hutch 162 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 953 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Saussurea sp. 1873 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Senecio scandens Buch.-Ham. ex re) 96 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 823 (A, CM, E, HIB, N, NA, NAS, NY, PE, SFDH, UC, WH). Siegesbeckia pubescens Makino 480 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Solidago decurrens Lour. a (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 20/6 (A, CM, E, HIB, N, NA, NAS, NY, PE, SFDH, UC, WH). sours deltoides (Aiton) Nakai 7 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Mee denticulata (Houtt. ) Kitamura (syn.: Ixeris denticulata (Houtt.) Steb- bins) 1617 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). MONOCOTYLEDONES GRAMINEAE Agrostis ae eee S. - 1321 (A, E, NAS, NY, PE, SFDH, UC, WH). ee haps (Thun - Makin 140 , NY, PE, SIDH, UC Artvason hops hanb, ) ies var. ee ee (Hackel) Honda’*, Bot. Mag. Tokyo 39: 277. 1925. 1655 ee E, HIB, KUN, NA, NAS, NY, PE, SEDH, UC, WH). 94 JOURNAL OF THE ARNOLD ARBORETUM Dactylis glomerata L. 972 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Deschampsia caespitosa (L.) Beauv. 47 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Deyeuxia sylvatica (Schrader) Kunth* 392 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Digitaria sanguinalis (L.) Scop. 1471 ( A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Erianthus fulvus Nees 1403 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Isachne nipponensis Ohwi 1959 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Melica onoei Franchet & Sav 687 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Microstegium nudum (Trin.) A. Camus* [VOL. 64 338 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1966 (A, CM, E, HIB , NA, NAS, NY, PE, SFDH, UC, WH). M saad igi iia HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). (A, Onlsmen canchutat folius (Ard.) Roemer & Schultes 42 , CM, HIB, KUN, NAS, NY, PE, UC); 688 (A, CM, E, HIB, KUN, NA, NAS, NY, E, SFDH, UC, WH). eee Wee (L.) Sprengel 1948 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Phleum alpinum L 43 (A, CM, HIB, KUN, NA, NAS, NY, PE). Poa nemoralis L., s.1.* 117 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC) Setaria excurrens (Trin.) Miq.* 111] (A, HIB, NAS, NY, PE). CYPERACEAE Bulbostylis densa (Wallich) Hand.-Mazz. 1969 (A, CM, HIB, KUN, NA, NAS, NY, PE, UC) Carex brunea Thunb. 1709 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Carex gentilis Franchet* 555 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Eriophorum comosum Nees 1916 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Rhynchospora chinensis Nees & Meyen 1188 (A, CM, HIB, NAS, NY, PE). Scirpus lushanensis Ohwi 1189 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 95 ARACEAE Arisaema consanguineum Schott 87 (HIB); 788 (A, HIB). Arisaema erubescens (Wallich) Schott* 1820 (A, HIB). Arisaema fargesii Buchet 1604 (A, HIB) Arisaema Cea eayune Blume* 1 A, HIB, NY, ae es lobatum Engler 243 (A, HIB, KUN, NAS, NY, PE). LEMNACEAE Lemna japonica Landolt*, Veroff. Geobot. Inst. ETH Stiftung Riibel Ziirich 70: 23. 1980. 1689 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH). ERIOCAULACEAE Eriocaulon buergerianum Koern. 2061 (A, HIB, NY Eriocaulon robustium Makino* 1187 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). COMMELINACEAE Streptolirion volubile Edgew. 1075 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1659 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC JUNCACEAE Juncus effusus L. 119] (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Juncus leschenaultii Gay 1190 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1259 (A, CM, HIB, KUN, NAS, NY, Cc). Juncus Nese Franchet* 42 (A, HIB, NY, PE). Juncus modicus N. E. Br. 118b (A, NA, PE, UC). Juncus potaninii Buch. 118a (A, PE). LILIACEAE Aletris stenoloba Franchet 15 A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). 96 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Allium cyaneum Regel 171 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Allium henryi C. H. Wright 870 (A, CM, HIB, KUN, NAS, NY, PE, SFDH, UC); 17/3 (A, CM, E, HIB, KUN, NA NAS, NY, PE, SFDH, UC, WH). Allium victorialis L. 1822 (A, CM, HIB, PE). Cardiocrinum giganteum (Wallich) Makino var. yunnanense (Leicht. ex Elwis) Stearn, Gard. Chron. 124: 4. 1948. 196 (A, CM, HIB, NA, NY, UC). Clintonia udensis pees i Meyer 88] (A, CM, HIB, PE, Disporum bodinieri (Lev. & Vaniot) Wang & Tang, Bull. Res. Peking Univ. 6: 20. 1949, fide Fl. Reip. Pop. Sin. 15: 44. 1978. 1112 (A, CM, HIB, KUN, NA, NAS, PE, UC). Disporum cantoniense (Lour.) Merr. 733 (A, CM, HIB, KUN, NAS, PE, UC); 1/60 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 1249 (A, HIB, NY, PE); 1393 (A, HIB, KUN, NA, NY, PE, UC); 1867 (A, HIB, NA, NY, PE). Hemerocallis minor Miller 3 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); /05/ (A, HIB, PE). Lilium lancifolium Thunb. 207 (A, CM, HIB, KUN, NAS, PE, UC); 36/ (A, HIB; distributed as Lilium sp.); 685 (HB); 855 (A, HIB, NA, NY, PE); 1059 (HIB). Lilium taliense Franchet 1480 (A, HIB). Lirlope Dee (L.) Baker 454 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Ophiopogon bodinieri Lévl. 1398 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH). Paris polyphylla Sm. 45 (up), 198 (A, CM, HIB, KUN, NA, NAS, PE, UC); 304 (A, CM, E, HIB, KUN, KYO, MO, NA, NAS, NY, PE, SFDH, UC, WH); 627 (A, HIB); SS6 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1396 (A, CM, E, HIB, KUN, NA, NAS NY, PE, SFDH, UC, WH). Polygonatum cyrtonema Hua 695 (HIB). Polygonatum odoratum (Miller) Druce 182] (A, HIB). Polygonatum sibiricum Delas ex Redouté 387 (A, CM, HIB, KUN, NA, NAS, NY, PE). Polygonatum verticillatum (L.) All. 343 (A, HIB, NA, NY); 1785 (A, CM, HIB, NAS, PE, UC). Polygonatum zanlanscianense Pampan. 48 . Reineckia carnea (Ander.) Kunth 507 (A, HIB); 1453 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 97 Smilacina henryi (Baker) Hara 30 (HIB). Smilax discotis Warb. 1582 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). en glauco-china Warb. 477 (A, HIB, PE); 1635 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Smilax megalantha C. H. Wright* 457 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 2005 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). Smilax menispermoidea A. DC 731 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC). Smilax polycolea Warb. 632 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1223 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1340 ss CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 1399 (A, CM, HIB, KUN, NA FDH, UC Smilax riparia A. DC. var. acuminata (C. tL Weight Wane rs Tang, FL ‘Rei. Pop. Sin. 15: 192. 1978. 1339 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 1786 (A, HIB, NY, PE). Smilax scobinicaulis C. H. Wright 792 (A, HIB, NAS, NY, PE, UC); 1586 (A, CM, HIB, KUN, NAS, PE, SFDH, UC); 1/690 A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Smilax stans Maxim 415 (A, HIB); 652 (s, HIB). Streptopus obtusatus Fassett (A, HIB, PE, UC). Tricyrtis maculata (D. Don) Macbr. 80 (A, HIB, NA); 15/1 (A, CM, HIB, KUN, NA, NAS, PE, SFDH, UC); 3/2 (A, CM, E, HIB, KUN, KYO, NA, NAS, NY, PE, SFDH, UC, WH); 505 (A, HIB, PE, UC). Tupistra chinensis Baker 033 (A, HIB, NAS). Veratrum oblongum Loes. f. 638 (HIB) DIOSCOREACEAE Dioscorea nipponica Makino var. rosthornii Prain & Burkill, J. Asiat. Soc. Bengal, Pt. 2, Nat. Hist. 73(Suppl. 7): 2. 1904. (syn.: D. giraldii R. Knuth) 233 (A, CM, HIB, KUN, NA, NAS, PE, UC); 39/7 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 537 CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH); 844 e CM, HIB, KUN, NA, NAS, NY, PE, Eas uc, WH); 1305 (A, CM, E, HIB, KUN, DH, UC, WH); 6 (A, HIB, KUN, NA, NAS, NY, PE, UC). eee envio L. o D. pane Thunb.) &02 (A, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC); 1470 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH WH Dioscorea zingiberensis @ a Wright 1652 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). 98 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 IRIDACEAE Belamcanda chinensis (L.) DC. 1651 (A, CM, E, HIB, KUN, NA, NAS, NY, PE, SFDH, UC, WH). Iris wilsonii C. H. Wright 952 (A, CM, HIB, KUN, NA, NAS, NY, PE, SFDH, UC). ZINGIBERACEAE Zingiber mioga (Thunb.) Roscoe 2022 (A, HIB, PE). ORCHIDACEAE Bletilla ochracea Schlechter 4 A, HIB, PE, UC). Coeloglossum bracteatum (Muhl.) Parl. (syn.: Coeloglossum viride (L.) C. J. Hartman var. bracteatum (Muhl.) Richter, Pl. Europ. 1: 278. 1890.) S85 (A, HIB, PE). Cypripedium fasciolatum Franchet* 1808 (A, HIB). Neottianthe cucullata (L.) Schlechter* 1796 (HIB); 1809 (A, HIB). Neottianthe de (Ames & Schlechter) Schlechter* 1185 (A, HIB Platanthera cf, seloeloiti Maxim.* 950 (HIB). Pleione bulbocodioides (Franchet) Rolfe 1391] (A, HIB). LITERATURE CITED Anonymous. 1976. Flora Hupehensis. Vol. 1. 508 pp. Wuhan Institute of Botany, . 1979. Flora Hupehensis. Vol. 2. 522 pp. Wuhan Institute of Botany, Wuhan. . 1980. Shennongjia plants. 467 pp. + 63 figs. Wuhan Institute of Botany, Wuhan. [In Chinese with Latin plant names.] Ban, J. D. 1980. Study on the vegetation of Shennongjia Mountains, western Hupeh. J. Centr. China Normal Col., Nat. Sci. Ser. 1980(Special Issue): 1-90. [In Chinese. ] BARTHOLOMEW, B., D. E. BoUFFORD, & S. A. SPONGBERG. 1983. Metasequoia glypto- stroboides—its present status in Central China. J. Arnold Arbor. 64: 105- , R. A. Howard, & S. — 1979. Phytotaxonomy in the People’s Republic of China. Brittonia 31: BLOEMBERGEN, S. 1939. eee of the genus A/angium. Bull. Jard. Bot. Buitenzorg BRETSCHNEIDER, E 1898. History of pane botanical discoveries in China. xv + pp. Sampson Low, Marston & Co., London CHING, 4 C. 1978. The Chinese fern cee a eenera: sare arrangement and historical origin. Acta Phytotax. Sin. 16(3): 1-19; 16(4): 1 CLAusen, K., & S. Y. Hu. 1980. Mapping the collecting aie of E. H. Wilson in China. Amoldia 40: 139-145. 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 99 Hotmaren, P. K., W. KEuKEN, & E. K. SCHOFIELD. 1981. Index herbariorum. Part 1. The herbaria of the world. ed. 7. Regnum Veg. 106: 1-452. Howarp, R. A. E. H. Wilson as a botanist. Arnoldia 40: 102-138, 154-193. _ BARTHOLOMEW, & T. Exias. 1979. The botanical gardens of the People’s Republic of China. Amer. Assoc. Bot. Gard. Arbor. Bull. 13: 33-4 Hu, S. Y. 1949. The genus //ex in China. J. Arnold Arbor. 30: 233- 344: 348-387. Ibid. 31: 39-80; 214-263. 1950. 1980. The Metasequoia flora and its phytogeographical significance. [bid. 61: Li, H. L. 1953. Endemism in the ligneous flora of eastern Asia. Proc. 7th Pacific Sci. Congr. Bot. 5: 212-216. [A Cs of this article into Chinese appeared in Biol. News (China) 1957(6): 5-8. Li, S. G. 1940. Outline of aa of Quaternary glaciation in western Hubei, eastern Szechuan, western Hunan and northern Kwangsi provinces. Geol. Rev. 5(3): 171-184. [In Chinese.] LogEsENER, T. 1901. Monographia Aquifoliacearum. Pars 1. Nova Actorum Acad. Caes. Leop.-Carol. German. Nat. Cur. 78: 1-598. Penc, Z. M. 1957. The vertical distribution of plants and plant communities in Shen- nongjia, Hubei. J. Centr. China Agric. Col. 2: 126-142. [In Chinese. THISELTON-DyerR, T. F. 1889. Maer productions, central China. Kew Bull. 1889: 225-227. [Includes letter from A. Henry to Thiselton-Dyer.] THORHAUG, A., ed. 1978. Botany in China: report of the Botanical Society of America Delesatiion to the People’s Republic, May 20—June 18, 1978. v + 154 pp. U.S.-China Relations Program, Stanford University, Stanford, Californ WILson, E. H. 1905. Leaves from my Chinese note-book. III. Kiating- -fu. Gard. Chron. 174. 913. A naturalist in western China. Vol. 1. frontisp. + xxxvii + 251 pp. + 56 pis. [Introduction, pp. umes by . a SARGENT.] Vol. 2. frontisp. + xi + 229 pp. + 43 pls. Doubleday, Page & Co., Wu, C. Y. 1979. The co aera of the ee flora. Acta Bot. Yunnan. 1(1): l- 22 , chief ed. 1980. The vegetation of China. vi + 1375 pp. + 3 fold-out charts + 96 unnumbered pp. of photographs + 2 maps in pocket. Science Press, Beijing. [In Chinese. ] Yina, T. S., C. G. Ma, & C. S. CHANG. 1979. Observations of the flora and vegetation of Mt. Shennongjia in western Hupeh, China. Acta Phytotax. Sin. 17(3): 41-60. [In Chinese, English summary.] 100 APPENDIX 1. JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Cross reference of collection numbers with aaah localities,* 1980 Sino-American Botanical Expediti COLLECTION COLLECTION COLLECTION NUMBERS LOCALITY NUMBERS LOCALITY NUMBERS LOCALITY 1-25 | 856-938 26 1505-1517 43 26-50 4 o39 27 1518 42 51-94 2 940, 941 28 1519, 1520 43 95-100 3 942, 943 pa 152] 42 101-109 2 94 28 1522-1526 43 110-125 4 945-966 27 1527-1541 44 126-128 5 967-976 28 1542-1553 45 129-150 20 977-998 27 1554, 1555 46 151-153 3 999-1050 29 1556 47 154-178 4 1051-1060 8 1557-1567 45 179-207 a) 1061, 1062 10 1568-1573 48 208-242 7 1063-1100 11 1574 50 243, 244 9 1101-1114 23 49 245-250 10 1115 17 1576-159] 48 251-269 4 1116-1118 23 1592-1596 5] 270-284 7 1119 22 1597-1674 52 285-298 1] 1120-1150 24 1675-1692 23 299, 300 12 1151-118] 29 1693-1727 54 301-329 6 1182 30 1728 D2 330-350 8 1183 31 1729 56 351-354 4 1184-1186 32 1730-1735 57 355-400 5 1187-1275 33 1736 56 401-427 11 1276-1300 34 1737 a7 428-443 13 1301-1323 35 1738-1770 58 444-485 12 1324 36 1771-1786 59 486, 487 13 192). 35 1787-1796 58 488-500 14 1326-1350 38 1797-1823 60 501-532 15 1351-1360 34 1824, 1825 61 533-538 16 1361-1363 37 1826-1830 62 539 15 1364-1378 34 1831-1835 61 540-550 17 1379-1381] a7 1836-1849 60 551-561 14 1382-1384 38 61 562-590 17 1385-1400 39 1851-1895 63 591-599 18 1401-1437 40 1896-1911 64 600-626 19 1438-1454 4] 1912-1920 65 627-650 20 1455 40 1921-1924 64 651 18 1456-1465 Bi, 1925-2000 66 652-669 20 1466 42 2001-2061 67 670, 671 18 1467-1472 43 ee 68 67 20 1473 42 69 673-716 21 — 1484 43 sre. 2081 70 717-750 24 1485 42 2082 71 751-854 25 1486-1503 43 2083-2085 67 855 —t 1504 42 *Collection localities are given in APPENDIX 2. +Vicinity of Jiuhuping Forest Farm, 8 September. 19 le 83] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 101 APPENDIX 2. Locality data, 1980 Sino-American Botanical Expedition. SHENNONGJIA ForesT DISTRICT . = of eee Forest Farm along W side of Jizigou canyon. Elev. ca. 1650 m. 25 2: Ca. 0. 5 ‘km S of Jiuhuping a Farm in side canyon on E side of Jizigou canyon. Elev. ca. 1700 m. 25 Augu 3. Side canyon on E side of, Tizizou canyon directly E of Jiuhuping Forest Farm. Elev. ca. 1700 m. 25 Augus 4. Vicinity of Chuifeng Piss Lae 2500-3000 m. 26 Augus 5. Along Qiaodonggou ca of road between Jiuhuping ee Farm and Bancang. Elev. ca. 195 6. Along Yuergou canyo on. “Elev. 2000-2200 m. 27 yates 7. Along Miaogou canyon. Elev. 1800-1900 m. 28 Augus 8. Side canyon on E side = Jizigou canyon directly E of Paes Forest Farm. Elev. ca. 185 . 28 Aug 9. Xiaodangyang, along paeaee River near km 53 from Xingshan Xian. Elev. ca. 1000-1050 m A 10. gees of Mucheng along Jiuchong River. Elev. ca. 1000 m. 29 Augus 11. W of Hongriwan yale camp along trail leading toward aa Bunline Mt. Elev. ca. 1300 m. 30 A 12. Laojunshan Yaowan canyon es W side of Jiuchong River, ca. 1 km S of Mucheng. Elev. 1000-1250 m. 31 Aug 13. Vicinity of Mucheng along eee River. Elev. ca. 940 m. 31 Augus 14. Side canyon of Mengjia Sa N of bridge crossing Mengjia River N of Mucheng. Elev. ca. 1000 m. 1 September. 15. E side of ee Ree in een of Hongriwan construction camp. Elev. ca. 1300 m. | Sept 16. Between Niiend and eas construction camp along Mengjia River. Elev. ca. 1000 m. 1 Septem 17. Along trail between Aoraiw an construction camp a Eee Elev. 1200-1400 m. 2 September (no. ///5 collected on 5 Septem 18. Vicinity of Qiujiaping. Elev. ca. 1435 m. 2 Septem 19. Vicinity of Qiujiaping. Elev. ca. 1440- 1650 m. 3 ae ber 20. Along trail from Quujiaping toward Laojun Mt. Elev. 1500- 2300 m. 3 September. 21. Vicinity of Qiujiaping in side canyons on E side of Mengjia River. Elev. 1500-1900 m. 4 September. 22: Disturbed slope above Jizigou. Elev. ca. 1800 m. 4 September 23. Ca. 1 km W of Leigutai Hydroelectric Power Station at confluence of Jiuchong and Dangyang rivers. Elev. ca. 390 m. 5 September 24. Vicinity of Muyuping Forest Brigade on SE side of watershed divide between Chang- jiang ea and Hanjiang rivers, near km 73 from Xingshan Xian. Elev. 1450- 600 m. 7 September. 25. NE of Guanmenshan along S side of Shicao River. Elev. ca. 1150 m. 8 September. 26. Vicinity a Dalongtan and Xiaolongtan on W side of road. Elev. 2300-2600 m. 9 Septem 27. Vicinity of pose rnen Elev. 2700-2900 m. 10 September. 28. Along road between Guanmenshan and Xiaoshennongjia. 2600-2650 m (no. 940 t 1500 m). 10 em Wh Ne} es) — 0 m). t nee . Vicinity ee ‘Duanjiangping Elev. 1300-1800 m. 11 September. 0. Gan ngou, is between Jiuhuping Forest Farm and Dajiuhu. Elev. ca. 1700 m. 12 cenes . Ma Mt., on oo between Jiuhuping Forest Farm and Dajiuhu. Elev. ca. 1700 m. 12 September *Locality number +Localities 23 and 65 are in Xingshan Xian, just south of the Shennongjia Forest District. JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 : eee on road between Jiuhuping Forest Farm and Dajiuhu. Elev. ca. 2650 m. 12 September : Znstansa Pass on W side of Dajiuhu basin. Elev. 1780 m. 13 September. . Send of Loyang River gorge near Pingqian. Elev. ca. 1300 m. 14 September. . Ma Mt., on on between Jiuhuping Forest Farm and Dajiuhu. Elev. 2150 m. 14 mber September. . E side of ridge between Pingqian and Dajiuhu. Elev. 2000 m. 14 September. . Along road between Pingqian and Dajiuhu. Elev. a m. 14 September. . Vicinity of Xiaojiuhu. Elev. 2000 m. 15 Septem . Hillside above Qianjiaping. Elev. 1400-1800 m. 17 September 0. Honghua ravine on W side of Dangyang River N of Honghua. Elev. ca. 820 m. 17 — - - HL HL Fa) & nN ~ September. . W side of Dangyang River at km 44 Xingshan Xian, side canyon on W-SW side of road. Elev. 725 m. 17 Septem . Along road aes Taizishang on SE ae of Yingyu River canyon. Elev. 2000-2050 m. 18 Septem . Vicinity of Taizishang along Yingyu River. Elev. 2000 m. 18 September. . S of Jiuhuping Forest Farm along Jizigou canyon bottom. Elev. 1900 m. 19 Sep- mber . Vicinity of Yinpo on road between Jiuhuping Forest Farm and Bancang. Elev. 1750 m. 19 September . Between Yinpo and Qiaodonggou canyon on road between Jiuhuping Forest Farm and Bancang. Elev. 2100 m. 19 September. . Canyon-side on slope of SE-draining tributary of Jizigou canyon on road between Jiuhuping Forest Farm and Bancang. Elev. 2000 m. 19 September. . Along road between Baicaoping and Bajiaomiao. Elevation ca. 1400 m. 20 Septem- er. . Headquarters of Songlo Commune. Elev. ca. 1000 m. 20 September. . Vicinity of Xinping. Elev. 1200 m. 20 September . Side canyon on E side of Songlo ma directly E of Songlo Commune headquarters. Elev. 1000-1200 m. 20 Septem . Vicinity of Houshanping on S side ‘of Hou River. Elev. 800-1050 m. 21 September. . Side canyon on E side of Songlo River directly E of Songlo Commune headquarters. Elev. 1000-1200 m. 22 September. : sae canyon on E side of Songlo River E of Jingyin. Elev. 1050-2000 m. 22 Sep- mber : Tust S of Tianmen Pass. aves 2300 m. 23 September 6. Along road between Gumiaoya and Panshul. Elev. 1700-1850 m. 23 September. ber . tem er. of Shibapan along Changping River. Elev. 1500 m. 25 September . Vicinity of Sanyuan. Elev. 1550-1900 pa erie no. 1874: road between Baishayuan and Sanyuan. Elev. ca. 2100 m). 26 Sept . Canyons and cliffs along road, ca. 12 km from town of Xingshan on road between Xingshan and Jiuhuping Forest Farm. Elev. 330 m. 29 September.t METASEQUOIA REGION OF LICHUAN XIAN . Metasequoia area in vicinity of Lojiaba on W side of valley. Elev. ca. 1500 m. 6 October . Metasequoia area in vicinity of Zhuanjiaowan on E side of valley. Elev. ca. 1500 m. 7 October 1983] BARTHOLOMEW ET AL., SINO-AMERICAN EXPEDITION 103 68. Metasequoia area in vicinity of Hongshaxi on E side of valley. Elev. ca. 1500 m. 7 69. Wanjiatang on road between city of Lichuan and Zhonglu. Elev. ca. 800 m. 8 cto 70. Meet area in vicinity of Yujiawan, E of Xiaohe. Elev. ca. 1500 m. 8 October. 71. Modaoai. 9 October B. B. Y.X. J. UNIVERSITY OF CALIFORNIA BOTANICAL WUHAN INSTITUTE OF BOTANY ARDEN ACADEMIA SINICA BERKELEY, CALIFORNIA 94720 WUHAN, HUuBEI Present address: CALIFORNIA ad OF SCIENCES pais GoLpEN GATE DEPARTMENT OF BIOLOGY SAN FRANCISCO, een 94118 WUHAN UNIVERSITY WUHAN, HuBEI D. E. B. CARNEGIE MUSEUM OF NATURAL J.LL 1 THE New YorkK BOTANICAL GARDEN 4400 Forses AVENUE Bronx, New York 10458 PITTSBURGH, PENNSYLVANIA 15213 SAS Present address HARVARD ee HERBARIA 22 Divinity AVEN CAMBRIDGE, Nur ue 02138 THE ARNOLD ARBORETUM OF HARVARD UNIVERSITY Vv UE CAMBRIDGE, MASSACHUSETTS 02138 KUNMING INSTITUTE OF BOTANY ACADEMIA SINICA KUNMING, YUNNAN WUHAN INSTITUTE OF BOTANY, AND DEPARTMENT OF BIOLOGY OF WUHAN UNIVERSITY Fo WUHAN, HUuBEI WUHAN INSTITUTE OF BOTANY ACADEMIA SINICA INSTITUTE OF BOTANY WuHan, HuBE! ACADEMIA SINICA T.R.D. BEIJING U. S. NATIONAL ARBORETUM WASHINGTON, D. C. 20002 : WUHAN INSTITUTE OF BOTANY : i ACADEMIA SINICA JIANGSU INSTITUTE OF BOTANY WuHan, HuBe Hortus BoTANICUS NANJINGENSIS T.S.Y EM. SUN YAT-SEN INSTITUTE OF BOTANY NANJING, JIANGSU ; ACADEMIA SINICA BEIJING 1983 BARTHOLOMEW ET AL., METASEQUOIA 105 METASEQUOIA GLY PTOSTROBOIDES—ITS PRESENT STATUS IN CENTRAL CHINA BRUCE BARTHOLOMEW, DAviD E. BOUFFORD, AND STEPHEN A. SPONGBERG DURING THE LATE SUMMER and early fall of 1980, we participated in the 1980 Sino-American Botanical Expedition to western Hubei Province in the People’s Republic of China.' Most of our fieldwork was conducted in the Shennongjia Forest District, but a brief visit (5-10 October) was also made to the metase- quoia region (lat. 30°10’N, long. 108°45’E) of Lichuan Xian (Hsien)? after most of the field activities were completed. To reach the metasequoia area we traveled west by boat up the Changjiang Verses) River from Yichang in Hubei Province to Wanxian in Sichuan Prov- ince (Map 1). From Wanxian we proceeded southward by jeep back into Hubei Province to the town of Modaoqi (Mo-tao-chi), where Metasequoia was first discovered as a living plant. After a brief stop at the type tree of Metasequoia glyptostroboides Hu & Cheng (see FiGureE 1), we continued on to the town of Lichuan (the county seat of Lichuan Xian), our base of operations for the next four days. We visited the metasequoia valley (about three hours away) in Xiaohe Commune on each of three consecutive days (6-8 October). Although this region has been visited by Chinese botanists and foresters over the past four decades, we had the very great privilege to be the first foreigners to visit the area since 1948 (see below). Part of our time in the valley was spent walking or traveling to noteworthy trees of M. g/yptostroboides and meeting with com- mune and local forestry officials to learn of their inventories, care, and pres- ervation of the naturally occurring trees and their program of seed collection and propagation. We also had an interesting and informative interview with e American participants were: Bruce Bartholomew, California Academy of Sciences (then at University of California Botanical Garden, Berkeley); David E. Boufford, Harvard University Her- of Harvard University. The principal Chinese participants included: A. L. Chang, Spaaen, Institute of Botany; Z. Cheng, Wuhan Institute of Botany; S. A. e, Jiangsu Institute of B y; Y. X. Jin, Wuhan Institute of Botany; Q. Y. Li, Wuhan University S. C. Sun (e oer ee ia Wuhan University and Wuhan Institute of Botany; Y. C. Tang, Institute of Botany, Beijing; J. X. Wan, Wuhan Institute of Botany; and T. S. Ying, Institute Botany, se The expedition was conducted under the auspices of Academia Sinica and the Botanical Society of America, with funding by Academia Sinica and additional support from the National Sane a (Grant #2133-80) and member rs of ihe Amiencan. Association of Boe Gardens and Arbo 2T he P Llanes and places for which aaa in spelling might cause eOnniSIOn. In fhieset instances the aie ae spelling is given in parenthes © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 105-128. aaa 1983. 106 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ij aif i gh eae ge ob, f " AAs . ab 1G UP” hs el a = (awe ‘ pat if ee av J Sf WNSS, LVI” n, AOS mA ¢B oF If i fp: tin, GE OP bas Bh; ei cigee lap an, fay #8 :. EN, ; >) 7 APE? Na Is af a £ i“ 4 =~ & ytd 1 bree irg th Ve, ee gh Sof ff: SF per on portion of NASA LANDSAT and Satellite Image (No. 8230502370500, taken November 1975 and reproduced by U. S. Geological Survey EROS Data Center) to show topography of region. Wanxian located on banks of Changjiang (Yangtze) River. 1983] BARTHOLOMEW ET AL., METASEQUOIA 107 two elderly members of the commune, who remembered the 1948 visit of Americans to the region. The remainder of the time was spent collecting spec- imens and determining the extent and nature of the vegetation associated with Metasequoia. No quantitative ecological data were gathered, but 161 collections of herbarium specimens were made. The identities of the vascular plants are included in the 1980 Sino-American Botanical Expedition report (Bartholomew et al., 1983) and are represented by numbers /925-2085. Those not previously recorded for the metasequoia area are listed below. Although we spent only a portion of three days in the valley, the limited number of firsthand reports in English concerning the natural occurrence of Metasequoia (Chaney, 1948; Chu & Cooper, 1950; Gressitt, 1953) makes our visit of interest. In addition, over 32 years have elapsed since the observations that served as the basis for these previous reports were made. Our impressions of current conditions in the valley are noted below. THE DISCOVERY OF METASEQUOIA GLYPTOSTROBOIDES ] L ae | The discovery of A “living fossil” has received a great deal of coverage in both the scientific and the popular literature (Fulling, 1976, 1977). However, to provide background information, we present a brief summary based on information given by Fulling (1976, 1977), Hu (1980), Bocher (1964), and Belder and Wijnands (1979) of the sequence of events surrounding the discovery and introduction of Metasequoia into cultivation outside of China. Metasequoia was discovered in 1941 at Modaogqi in Sichuan Province near the border with Hubei Province? by T. Kan (Gan Duo), of the Department of Forestry of National Central University. Kan, however, did not collect or make specimens, and it was not until 1943 that C. Wang (Wang Zhang) of the Central Bureau of Forestry made the first collections of herbarium material. The tree was initially thought to be a form of G/yptostrobus lineatus (Poiret) Drude (syn.: G. pensilis (Staunton) Koch), but W. C. Cheng (Cheng Wanjun), of National Central University in Nanjing, realized that it represented a new genus. As a result, Cheng sent one of his assistants, C. J. Hsueh (Xue Jiru), to collect more herbarium material in February and May of 1946. In the fall of 1946, H. H. Hu (Hu Xiansu), then director of the Fan Memorial Institute of Biology, Beijing, received material from W. C. Cheng. Hu recognized that the newly discovered tree belonged to the genus Metasequoia, described in 1941 from Pliocene fossils by the Japanese botanist Shigeru Miki (Miki, 1941). Miki determined that certain fossils, which for nearly 100 years had been variously assigned to either Sequoia or Taxodium, actually represented a new genus, which he named Metasequoia. At the time, he did not realize that a living species of Metasequoia was still extant in south-central China. In 1946 Cheng also sent herbarium specimens to E. D. Merrill, then director of the Arnold Arboretum of Harvard University. Merrill, realizing the signif- icance of the new discovery, immediately arranged through Hu to obtain seeds. 3After the founding of the People’s Republic of China, the boundary between Sichuan and Hubei provinces in this region was changed so that Modaogqi is now in Hubei 108 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 a * FiGuRE 1. American participants, 1980 Sino-American Botanical Expedition to west- bei Province, People’s Republic of China, at foot of type tree of pages a ag just outside hamlet of Modaogi: (from left) David E. Boufford, James ut e Bartholomew, Stephen A. Spongberg, and Theodore R. Dudley. eae characters on i trunk indicate ““Lichuan County Tree Number One” and illustrate record of this tree in Metasequoia count. 1983] BARTHOLOMEW ET AL., METASEQUOIA 109 Merrill provided $250 for Cheng to organize a collecting trip in the late summer and fall of 1947, when seeds would be mature, and Cheng assigned C. T. Hwa (Hua Jingan) to undertake the expedition. During the three-month collecting period, Hwa found trees of Metasequoia scattered over an area of about 800 km’, with the largest concentration (what he thought to be about 1000 trees) in the Shuishaba (Shui-sa-pa) valley of western Hubei.* About | kg of mature seeds was collected on this expedition, and Cheng sent the initial batch to the Arnold Arboretum in December, 1947. It has generally been thought that this was the first introduction of living Metasequoia into the West. However, based on evidence provided by Bécher (1964, in legend to fig. 7) and Belder and Wijnands (1979), Cheng apparently also sent seeds to Copenhagen and Amster- dam at the same time that he sent them to Merrill at the Arnold Arboretum; thus, the introduction of Metasequoia into the West was essentially simulta- neous in Europe and North America. (This information differs from that given by Fulling (1976)). Merrill immediately distributed seeds to institutions and interested individuals around the world, and one recipient was Ralph W. Cha- ney, a paleontologist at the University of California, Berkeley. Early in 1948 Chaney traveled to see Metasequoia in the wild. He was accompanied by Milton Silverman, who was then science writer for the San Francisco Chronicle. Chaney and Silverman were in the metasequoia area for five days in March, 1948, and returned with seeds and several seedlings of Metasequoia. Fulling (1976) believed that Chaney did not collect seeds of Metasequoia, but according to Silverman (pers. comm.), he and Chaney collected seeds from cones on the ground. Chaney and Silverman divided these seeds in China (Silverman returned to the United States first), and they each carried some back. The fact that Chaney brought seeds back is corroborated by Jean Spitzer, a stewardess on the Pan American flight that Chaney took from Wake Island to Honolulu. According to Mrs. Spitzer (pers. comm.), Chaney told her of the seeds, which he had in an inside pocket of his coat, and stated that he was not going to declare them at Honolulu—he feared that the seedlings of Metasequoia he had with him might be confiscated and did not want to lose the seeds as well. Later in 1948 two additional expeditions, one led by Cheng and the other by J. L. Gressitt, of Lingnan University, stayed for extended periods in the metasequoia area. Gressitt was mainly interested in collecting insects associated with Metasequoia, but he also made plant collections that, in addition to seeds and herbarium specimens from the Cheng expedition, were sent to the Arnold Arboretum. The herbarium material from these two expeditions served in large part as the basis for Hu’s floristic analysis (1980). Gressitt also collected seeds and seedlings of Metasequoia, and these were sent to Lingnan University and to the California Academy of Sciences in San Francisco (Gressitt, pers. comm.). *The valley containing the main Metasequoia population is usually referred to in the literature as Shuishaba (Shui-sa-pa); however, according to Xi Xingwen, head of the Xiaohe Commune, the name Shuishaba only refers to the region immediately around the hamlet of Shuishaba rather than to the entire valley 110 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 THE TYPE TREE OF METASEQUOIA GLYPTOSTROBOIDES The original tree from which the type specimens of Metasequoia glyptostro- boides were taken is still growing just outside of Modaogqi, about 35 km north of Xiaohe, where the main Metaseguoia population occurs. Gressitt (1953) reported that two smaller trees of Metasequoia occurred in a row with the type tree, but in 1980 only the one large tree remained at this location. Inquiries in Modaogqi about the other two drew a complete blank. There are now many Metasequoia trees along the road that passes through Modaoqi, but these have been planted during the past 30 years, along with Cryptomeria japonica (L. f.) D. Don, as the principal roadside trees in the area. In 1948 there was a small shrine on one side of the type tree and a shed on the other. Both of these structures have been removed. The tree looks essentially the same now as it does in photographs taken in 1948 (see Ficures 2, 3). It appears to be in excellent health and was bearing seed-filled cones in October, 1980. However, it has grown only very slightly over the past 30 years. We measured the buttress reported as 230 cm in diameter in the late 1940’s (Hu & Cheng, 1948): it was 241 cm across, and the diameter of the trunk above the buttress was 167 cm. Based on partial cores, Chaney estimated this tree to be 300 years old (Chu & Cooper, 1950). Extrapolations from partial cores made in 1977 yielded an estimated age of 450 years. PRESENT STATUS OF NATURALLY OCCURRING METASEQUOIA Almost all of the naturally occurring trees of Metasequoia grow in the central valley of Xiaohe Commune. Since 1974 the Bureau of Forestry of Lichuan Xian has maintained a staff of five people in the Commune, with one of their objectives being to measure each tree every four years. The Forest Bureau has counted and numbered 5420 trees with a d.b.h. of at least 20 cm. We were told that approximately 1700 to 1800 of these are mature, seed-producing trees. The tallest recorded trees in the valley are on the east side in the vicinity of Hongshaxi: several reach a height of about 50 m, but they have smaller trunk diameters than the one that grew at Wangjiaying (see below). The oldest tree in Xiaohe Commune (approximately 420 years, estimated from a partial coring made in 1977) is near the town of Xiaohe (FiGure 4); in 1980 it had a d.b.h. of 160 cm. There are several outlying populations and individuals (Ling, 1976; Liu et al., 1978: Zhang Fengyun, pers. comm.) in addition to the major Metasequoia population at Xiaohe. Near the town of Zhonglu (ca. 15 km southeast of Xiaohe), there are about ten trees, with a d.b.h. of 40-50 cm; in Shizhu Xian (about 40 km west of Xiaohe in Sichuan Province) there are two; in Longshan (in northwestern Hunan Province about 110 km southeast of Xiaohe), three (two growing together and one some distance away). Another outlying tree 1s the type tree at Modaoqi. There was previously an isolated tree at Wangjiaying, discovered by Hwa in 1947 (Hu & Cheng, 1948); with a diameter of 2.2 m and a height of about 50 m, it was the largest recorded Metasequoia (Chu & Cooper, 1950). This tree 1983] BARTHOLOMEW ET AL., METASEQUOIA 111 a. 4 Y om FiGureEs 2,3. Type tree of Metasequoia oe 2, photograph taken in 1948 by J. L. Gressitt. Note foot at peas of tree. 3, photograph take n on 5 October 1980. Shrine removed, and tree now su ee by rice ae and small, ditched stream. Buildings in eae ced pe 1948 was also seen by Gressitt in 1948 (Gressitt, 1953). Unfortunately, lightning struck it in 1951, splitting it into three parts and killing it. Another large tree, with a diameter of 2 m, was found between the main population in Xiaohe Commune and the three trees in Hunan Province, but it was cut down during the Cultural Revolution. In its natural habitat Metasequoia is now protected by the government, and not even small trees may be cut. The trees that we saw (including the two ancient ones at Modaoqi and Xiaohe) all appeared to be in good health. How- ever, we did not see any small seedlings. This differs from the situation in 1948, when Chu and Cooper (1950) found seedlings in thickets surrounding older Metasequoia trees. In 1980 vegetation was either absent around the trees of Metasequoia, or very closely cropped (see FiGuRE 5), presumably by the local people and not by animals. The lack of governmental protection of the habitat (and thus the lack of associated vegetation) probably accounts for the lack of seedling establishment. THE “METASEQUOIA FLORA” The occurrence of Metasequoia trees in the main valley is essentially the same as that reported by Chu and Cooper (1950) and Gressitt (1953): Metase- 112 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 Figure 4. Oldest Metasequoia tree in Xiaohe Commune (approximately 420 years old, based on partial corings made in 1977, and with d.b.h. of 160 cm in 19 quoia grows primarily along the sides of the valley, and most of the flat land in the valley is planted in rice. There have been a number of reports of plants associated with Metasequoia 1983] BARTHOLOMEW ET AL., METASEQUOIA 113 (Chaney, 1948; Chu & Cooper, 1950; Hu, 1980). It is of interest to compare these reports with the present situation of Metasequoia in Xiaohe Commune. Chaney (1948) enumerated a few species and genera that he recognized, but his report was fragmentary and not based on the collection of specimens. In contrast, the paper by Hu (1980) was based on all of the material in the Arnold Arboretum herbarium that was collected in the metasequoia area in the late 1940’s. It enumerates 550 species of vascular plants belonging to 301 genera and 127 families. Based on our firsthand observations, it is our impression that Hu’s paper gives a very deceptive picture of the plants associated with Metasequoia, and that it is not accurate to refer to these 550 species as the ““Metasequoia Flora.” Although most of the specimens seen by Hu were col- lected in the vicinity of the metasequoia valley, it is quite likely that many were collected at higher elevations than where Metasequoia grows, and some were, no doubt, from cultivated trees. Hu admits the cultivated status of Ginkgo biloba L., and it is probably also true for such others as Pinus armandii Franchet (Gressitt, 1953, p. 49, reported one large tree in Shuishaba (Gressitt 2536) but many more near Modaoqi), Juniperus formosana Hayata (Gressitt, 1953, p. 51, reported two trees at “Suen-wu” and another on the approach to ““Suen- wu’), and Keteleeria davidiana (Bertrand) Beissner (we saw two very large trees of this near a school a short distance from Xiaohe). It is also clear from the accounts of Gressitt’s treks in search of other populations of Metasequoia that many of the plants he collected were found at some distance from the known Metasequoia stands, and many were from high elevations or from habitats unsuitable for Metasequoia (Gressitt, 1953). Also, of the 18 species of gym- nosperms listed by Hu (1980) for the “‘Metasequoia Flora,” we saw only one, Cunninghamia lanceolata Hooker, outside of cultivation and in close proximity to Metasequoia. However, Cunninghamia is restricted to drier habitats on well-drained slopes above Metasequoia. Chu (Chu & Cooper, 1950) also found Taxus chinensis Rehder in his quadrats and reported Cephalotaxus fortune! Hooker as being nearby. Chu and Cooper (1950, p. 272) noted that of the 33 tree species growing either in the Metasequoia quadrats or nearby, “only four are gymnosperms.” Similar examples of cultivated angiosperms growing with Metasequoia can likely be found. Hu herself (1980, p. 49) considers that at least 13 listed angiosperms do not grow in the metasequoia community. Chu and Cooper’s study (1950), derived from Chu’s quadrat analysis, is the only report on plants associated with Metasequoia based on systematic studies. Chu’s quadrats were placed in areas that included Metasequoia and were made during the 1948 expedition to the metasequoia area led by W. C. Cheng. Specimens from this expedition labeled as collected by W. C. Cheng-H. T. Hwa were part of the material used by Hu (1980). However, although Chu and Cooper (1950, p. 273) stated that “nearly one hundred species of herbs, includ- ing the lower forms, have been collected from the area in which Metasequoia grows,” they did not indicate that vouchers for the quadrat studies had been made. There are plants reported by Chu (see below) that were not represented among the specimens available to Hu. If there were vouchers made for Chu’s quadrat studies, they were not among the material sent to the Arnold Arbore- tum. It is also unfortunate that not all of Chu and Cooper’s determinations are 114 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Ficure 5. Grove of Metasequoia preiren growing at bottom of small ravine above main valley floor near village of Xiaohe. Note absence of associated plants. complete. In light of the deterioration of the vegetation in the metasequoia valley, and especially that growing in close association with Metasequoia, if vouchers for this study are extant in some herbarium in China, a reworking 1983] BARTHOLOMEW ET AL., METASEQUOIA 115 of the determinations of the material would give the best picture of the plants originally associated with Metasequoia that will ever be possible. The habitat of Metasequoia is reminiscent of that of Taxodium distichum (L.) Rich. in the southeastern United States, a parallel previously drawn by Chaney (1948). Metasequoia is a riparian species, and before habitation the valley floor may well have been a Metasequoia forest (see below). Metasequoia trees that occur away from the valley floor are restricted to the moist bottoms of ravines and draws that drain into the main valley (see FiGurE 5). Taxodium commonly occurs in flat, poorly drained depressions behind natural levees along slow-moving rivers. Based on this similarity of habitats, on reports of the species associated with Metasequoia (Chaney, 1948; Chu & Cooper, 1950; Gressitt, 1953), on Hu’s (1980) enumeration of the ““Metasequoia Flora,” and on our own observations both in the southeastern United States and in the metasequoia valley in 1980, it is possible to hypothesize a past Metasequoia forest analogous to present-day Taxodium distichum forests. Among the dom- inant tree species usually found with Taxodium distichum are Nyssa aquatica L., N. sylvatica var. biflora (Walter) Sarg., Populus heterophylla L., Quercus spp., Liquidambar styraciflua L., Carpinus caroliniana Walter, Betula nigra L., Acer rubrum L., Ulmus americana L., Carya spp., Fraxinus spp., and Salix spp. The associated shrubs include //ex spp., Viburnum spp., [tea virginica L., Cornus spp., and Lindera benzoin (L.) Blume. The vines include Berchemia scandens (Hill) K. Koch, Bignonia capreolata L., Rhus radicans L., Decumaria barbara L., Parthenocissus quinquefolia (L.) Planchon, Vitis spp., Ampelopsis spp., and Smilax spp. While each of the species of this group has specific microhabitat requirements, all are usually found growing in close proximity to Taxodium. In their list of plants growing with Metasequoia, Chu and Cooper (1950) included species of many of the same genera. We noted several large trees of Liquidambar acalycina (L. formosana Hance in Chu & Cooper, 1950; and in Hu, 1980) and species of Salix, Acer, Pterocarya, and Quercus in habitats similar to those occupied by Metasequoia, but not on the adjacent slopes. Moreover, it seems likely that at one time the floor of the metasequoia valley was occupied by trees that were tolerant of periodic flooding, could grow in poorly drained soils, and occupied more or less specific microhabitats. Among the species listed as being associated with Metasequoia glyptostroboides by Chu and Cooper (1950), Gressitt (1953), and Hu (1980), the following grow in habitats similar to those of their American counterparts associated with 7ax- odium: Houttuynia cordata Thunb. (in place of Saururus cernuus L. in the southeastern United States); Populus adenopoda Maxim.; Salix spp.; Ptero- carya hupehensis Skan, P. paliurus Batalin, and P. stenoptera C. DC. (all in place of Carya spp.); Betula luminifera Winkler; Carpinus fargesii Franchet; Quercus spp.; Morus sp.; Cocculus orbiculatus (L.) A. P. DC.; Ulmus multinervis Cheng; Lindera glauca, Liquidambar acalycina Chang; Ilex spp.; Berchemia spp.; Nyssa sinensis Oliver; Cornus controversa Hemsley and C. macrophylla Wallich; Clethra fargesii Franchet; Styrax bodinieri Lévl. and S. suberifolius Hooker f. & Arnott (S. japonica Sieb. & Zucc. was reported by Chu and Cooper 116 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 (1950) as being in the quadrats but is not included in Hu’s (1980) enumeration); Viburnum spp.; and Smilax spp. These similarities, however, should not be taken as evidence that the asso- ciations we see now have existed over long periods of time. Hu (1980, p. 64) states that “the vegetation of the metasequoia area is a living sample of a comparatively well-preserved ancient flora.”’ It is highly doubtful that this is true. It would perhaps be better to say that the flora in the metasequoia region contains a number of genera that have fossil records dating back to the Tertiary and ecological adaptations enabling them to grow together at present. Davis (1976) has compiled data showing that the forests of eastern North America and Europe, which are often thought of as stable communities, are actually the result of differential migrations in the past of the individual components that have come together only fairly recently. She has shown that the current asso- ciations may be a feature of the present, and that the species growing together now often occurred in different associations and abundance even within the past few thousand years. The fact that we see members of genera with long geologic histories growing together today in the metasequoia region of China only indicates that their ecological tolerances overlap; it gives no indication that their migrational or evolutionary histories have coincided. Davis (1976) suggests that the glacial periods (perhaps as many as 16 in North America and Europe), each of which lasted 50,000-100,000 years, are the normal and most stable phases of the Pleistocene, while the interglacials, which lasted only 10,000—20,000 years each, were catastrophic periods. There is little reason to believe that the climate of eastern Asia remained unchanged during this time when the climates in North America and Europe were undergo- ing such wide fluctuations. It seems almost certain that there would have been widespread migrations in the Asian flora like there were in North America and Europe, and that what we see now are the assemblages of species (communities) resulting from the sorting out of the flora in response to the rather drastic changes in climate during the glacial and interglacial periods of the Pleistocene and Holocene. Chu and Cooper (1950) stated that Metasequoia appears to grow naturally extensive stands of Metasequoia. Additional evidence of a once more-wide- spread Metasequoia forest on the valley floor 1s provided by several large trunks of Metasequoia that we saw that had recently been unearthed in the center of paddy fields far from the nearest slopes and ravines where the trees now grow. Altogether, more than 200 of these trunks, many over two meters in diameter, have been found in the paddies along the level floodplain of the main river and side streams (T. S. Ying, pers. comm.). Also (according to Liu et a/., 1978), some of the houses in the valley were constructed of boards cut from Metase- quola. These houses are believed to be 200-300 years old and date roughly from the time of the original settlers. 1983] BARTHOLOMEW ET AL., METASEQUOIA HE hat) Ni Res pe us Te ee Fic s 6, 7. General views of metasequoia valley. 6, photograph taken in 1948 b ce er ait Note forest- sarees slopes in backgr ound. and fields only on level valley floor. 7, photograph taken 1980. Note extensive cultivation of rice on valley floor and on terraced fields on lower slopes. Columnar trees along edges of paddies an ames of Metasequoia ee and Cunninghamia lanceolata, with colum- ar habit due to pruning of lateral branches for firewood 118 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 CLIMATE There are no climatic data available for Xiaohe Commune. The nearest weather station is in Lichuan, 40 km northeast of Xiaohe, but data from this station’ give a much more accurate picture of the climate where Mefasequoia grows than do those reported by Chaney (1948), which were based on infor- mation obtained from Zhongjing (Chungking), or those used by Chu and Cooper (1950) based on records from Guiyang (Kweiyang). Lichuan 1s at an elevation of about 1070 m (essentially the same as the floor of the metasequoia valley) and is within the range of Metasequoia, if one includes the outlying populations and individuals. TABLE | presents temperature records from Lichuan for the years 1959- 1978. The lowest monthly mean minimum temperature is —6.1°C for January, and the absolute lowest temperature is —15.4°C recorded in February, 1972. These temperatures are nowhere near the physiological lower limit of Metase- quoia since trees survive the winters in Jamaica Plain, Massachusetts, where temperatures as low as —23°C have been recorded. The highest monthly mean maximum in Lichuan is 32.2°C for August, and the absolute highest temper- ature 1s 35.4°C recorded in August, 1959. This is also not the physiological upper limit for Metasequoia. In St. Louis, Missouri, where Metasequoia is used as an ornamental and street tree, temperatures higher than 38°C are not uncom- on Precipitation data for Lichuan are given in TABLE 2. Rainfall is very seasonal: two thirds of the total amount occurs in the five-month period from May through September, and less than one sixth from November through March. There were no months lacking rain between 1959 and 1978, although only a trace (1.1 mm) was recorded in January, 1963. During this 20-year period the wettest year was 1975, with 1529 mm of rain; the driest was 1966, with 863.1 m. Without data from Xiaohe Commune, it is impossible to compare the climate there with that of Lichuan, but the fact that the valley is completely enclosed would probably provide for some insulation and moderation of the climate. Of particular note 1s the ridge about 1500 m in elevation that bounds the valley to the north. One might expect that the absolute minimum temperature in the metasequoia valley would be somewhat higher than in Lichuan. THE HUMAN POPULATION IN THE METASEQUOIA VALLEY The Lu family, the first to settle in the metasequoia valley, moved into the area about 310 years ago.° They settled in the area now occupied by the hamlet of Jiantianba, about 7 km west of the present town of Xiaohe. Xiaohe itself ‘Weather records from the Lichuan Weather Station were supplied by Zhang Fengyun, director of the Lichuan Forest Research Institute. ‘Information on Xiaohe Commune was supplied by Xi Xingwen, head of Xiaohe Commune. TABLE |. Temperature records (°C) for Lichuan, Hubei Province, People’s Republic of China, 1959-1978. MonTH RECORD Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec Monthly mean 1.7 3.0 7.8 12.9 16.9 20.3 23.3 22.8 18.6 13.6 8.1 3.6 Monthly mean maximum 12.5 15.6 21.5 26.5 27.8 30.7 32.2 32:3 29.3 24.5 18.6 14.5 Monthly mean minimum —6.1 =5 —1.6 8.7 11.9 15.8 14.7 9.3 3.9 —1.2 —3.7 Absolute maximum 16.9 20.0 27.1 30.2 30.6 32.9 34.8 35.4 32.0 27.5 2271 17.7 Absolute minimum —13.8 —15.4 —3.5 —l 5.9 8.6 13.7 12.2 7.1 —0.6 —3.5 —8.2 TABLE 2. Precipitation records (mm) for Lichuan, Hubei Province, People’s Republic of China, 1959-1978. oe ANNUAL RECORD Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. Mean monthl precipitation 18.7 30.0 67.9 109.5 187.5 183.8 171.9 144.1 170.8 109.2 61.2 28.3 1282.9 Percentage of annual total 1.5 2.4 5.3 8.5 14.6 14.3 13.4 11.2 13.3 8.5 4.8 22 100 Lowest monthly precipitation 1.1 6.0 26.6 60.2 74.9 55.7 15.9 24.6 47.0 19.3 26.6 8.7 Highest monthly precipitation 40.7 71.1 117.3 179.7 321.1 383.2 293.9 307.6 378.9 204.0 121.9 55.6 [£861 VIONOASVLAW “TV LA MANOTOHLUVEA 120 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 was first settled by the Wu family about 200 years ago. Chu and Cooper (1953), however, reported that the Wu family was the first family to settle in the valley. The family names Lu and Wu are still common in the area. It seems probable that the flat valley floor with its Metasequoia stands was the first site to be cleared for agriculture when the original inhabitants settled the valley (Ficures 6, 7). The settlers most likely cleared vegetation from the richest and easiest areas for cultivation but left the trees in marginal areas such as immediately along the river or in the narrow ravines leading into the main valley. Perhaps some trees were also spared due to their large size or because of possible religious significance. The valley containing the main Metasequoia population is in Xiaohe Com- mune. This commune occupies a total area of 10,640 ha and has a population of 21,000 people, of whom 12,000 live in the valley itself, mostly in scattered hamlets of several houses situated along the edges of the valley. Within the valley there are 670 ha of rice paddies and 1400 ha of other fields; outside there are an additional 1270 ha of nonpaddy fields planted in corn and other crops. In addition to the fields, the commune contains 7300 ha of mountainous land. The high population density in Xiaohe Commune (if the figures we were given are correct, there are 6.287 persons per ha or only 0.159 ha per person!) has resulted in considerable damage to the local vegetation. Both Chu and Cooper (1950) and Gressitt (1953) reported that the forests had largely been destroyed by the time of their visits, and even the Metasequoia communities showed signs of alteration due to man’s activities. We found that conditions had deteriorated even more since these reports. Our observations indicate that there has been so much human and domestic animal disturbance that there are very few plants now associated with Metasequoia (FIGURE 5). However, there are areas in Xiaohe Commune, particularly in side ravines and on slopes on the east side of the main valley, where secondary forests are developing. Although these areas are close to the Metasequoia groves, they are separated from the riparian areas occupied by Metasequoia by cut-over slopes and cul- tivated fields. Comparison of the present condition of the forests with pictures taken in 1948 shows considerable destruction during the past 32 years (see Ficures 6-9). We were told that many large trees, particularly Castanea henryi Rehder & Wilson and C. mollissima Blume, were cut in the mid to late 1950's during the Great Leap Forward to make charcoal for smelting iron. However, no significant amount of iron was ever produced. The protected status currently given by the government to the remaining naturally occurring trees of Metasequoia will probably insure their survival for the immediate future, but the lack of protection for the surrounding habitat will likely result in little, if any, natural reproduction. The thickets that Chu and Cooper (1950) mentioned as being around many of the trees are no longer there, and it was in those habitats that they reported finding seedlings and small trees of Metasequoia. The efforts to monitor the natural populations of Metasequoia may have resulted in disturbance and clearing of other vegetation, thereby contributing to the destruction of suitable germination sites. 1983] BARTHOLOMEW ET AL., METASEQUOIA 121 Ficure 8. Small side valley connecting with eastern portion of metasequoia valley (photograph taken by J. L. Gressitt in 1948). Note relatively undisturbed forest sur- rounding terraced field. 122 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 FiGUREY. Highly oa es of side valley showing results oserae habitation (photograph taken in 1980). er slopes largely denuded of native vegetation and lower slopes given over to rice ae and corn fields. Tree behind building is Ginkeo biloba. 1983] BARTHOLOMEW ET AL., METASEQUOIA 123 ACKNOWLEDGMENTS We thank Zhang Fengyun, director of the Lichuan Forest Research Institute, and Xi Xingwen, head of Xiaohe Commune, for their invaluable assistance and the information they have provided. We welcome this opportunity to express our thanks and deep gratitude to our Chinese colleagues who facilitated our trip to the metasequoia region, and to our Chinese and American colleagues who comprised the expedition team. We would like to thank the local govern- mental officials of Lichuan County for their invaluable help and kind hospi- tality. We are indebted to Professors P. S. Tang, director of the Institute of Botany, Academia Sinica, Beijing, T. T. Yui, vice director of the Institute of Botany, Academia Sinica, Beijing, S. C. Sun, director of the Wuhan Institute of Botany, Academia Sinica, and expedition leader, and P. H. Raven, director of the Missouri Botanical Garden, for their unfailing efforts to bring the 1980 Sino-American Expedition (of which the trip to the metasequoia region was a part) to a reality and a successful conclusion. The editorial expertise of E. B. Schmidt is also gratefully acknowledged. We would like to express our particular thanks to J. Linsley Gressitt, who kindly reviewed the manuscript and made available for our use photographs he had taken in the metasequoia valley in 1948. It was with great sadness that we learned of Gressitt’s and his wife’s deaths in an airplane crash in China on April 26, 1982, when Gressitt was on his first return visit to the People’s Republic since the early 1950’s. As a tribute to his pioneering work in the metasequoia area, we dedicate this paper to his memory. ADDITIONS TO THE FLORA OF THE METASEQUOIA REGION Of the 160 numbers collected in the metasequoia valley on the 1980 Sino- American Botanical Expedition, 122 of them represent taxa of vascular plants. Of these, 71 have not been previously reported from the valley. The following enumeration presents these additions to the flora of the region. Collection numbers of the Sino-American Botanical Expedition are in parentheses. The 20 taxa reported as associated with Metasequoia by Chu and Cooper (1950) that are not in Hu’s (1980) enumeration are also included here; these names are followed by an asterisk. The arrangement of taxa follows the same scheme used in the report of the 1980 Sino-American Botanical Expedition to western Hubei (Bartholomew ef al., 1983), and reference to that paper is suggested for annotations and discussions of taxonomic problems and for descriptions of new taxa. PTERIDOPHYTA LYCOPODIACEAE Lycopodium crispatum Ching (1974) HY MENOPHYLLACEAE Hymenophyllum barbatum Bosch (2023) 124 JOURNAL OF THE ARNOLD ARBORETUM DENNSTAEDTIACEAE Microlepia marginata (Houtt.) C. Chr. (2052) HyPOLEPIDACEAE Hypolepis punctata (Thunb.) Mett. (1943, 1944) PTERIDACEAE Pteris wallichiana Agardh (2026) HEMIONITIDACEAE Coniogramme robusta Christ var. repandula Ching (2049) WoOoODSIACEAE Peranema cyathioides D. Don (2085) BLECHNACEAE Woodwardia unigemmata (Makino) Nakai (2057) ATHYRIACEAE Athyrium epirachis (Christ) Ching (20/1) Athyrium vidallii (Franchet & Sav.) Nakai (2053) Lunathyrium vermiforme Ching, Boufford, & Shing (2025) Lunathyrium wilsonii (Christ) Ching (2048) Pseudocyclosorus tsoi Ching (1946) Di TWP Lbomnitvrnt b/ATD Dryopteris labordei (Christ) C. Chr. (2054 Dryopteris supraimpressa Ching, Boufford, & Shing (2020) Polystichum lobatopinnulum Ching, Boufford, & Shing (2059) ASPLENIACEAE Asplenium tripteropus Nakai (2046) POLY PODIACEAE Arthromeris cuneata Ching (1971, 2029 Drymotaenium miyoshianum (Makino) Makino (2058) Lepidogrammitis drymoglossoides (Baker) Ching (2027) Lepidomicrosorium subhastatum (Baker) Ching (2083) AZOLLACEAE Azolla imbricata (Roxb.) Nakai (2056) ANGIOSPERMAE DicotTyLEDONES FAGACEAE Castanea seguinii Dode* Quercus glandulifera Blume (1937) [VoL. 64 1983] BARTHOLOMEW ET AL., METASEQUOIA 125 MORACEAE Ficus foveolata Wallich* Morus alba L.* URTICACEAE Gonostegia hirta (Blume) Mia. (1937) POLYGONACEAE Polygonum caespitosum Blume (1956) Polygonum hydropiper L. (1954) Polygonum muricatum Meisner (syn.: P. strigosum R. Br. var. muricatum (Meisner) A. N. Steward) (/955) Polygonum persicaria L. (2015) Polygonum thunbergii Sieb. & Zucc. (1953) RANUNCULACEAE Clematis urophylla Franchet (/940) LARDIZABALACEAE Akebia ee (Thunb.) Koidz.* Holboellia s BERBERIDACEAE Berberis sargentiana Schneider (/938) Berberis virgetorum Schneider (207 9) SCHISANDRACEAE Schisandra pubescens Hemsley & Wilson* LAURACEAE Phoebe neurantha (Hemsley) Gamble (2045) HAMAMELIDACEAE Corylopsis veitchiana Bean (2033) Liquidambar acalycina Chang (1950) ROSACEAE Cotoneaster aff. dielsiana ie (1935) Pyracantha crenulata (D. Don) ee 1) Peer (Oa rate )H.L.L Rosa henryi Boulenger* Rubus amphidasys Focke ex Diels (2076) Rubus lambertianus Sér. (1936 Rubus setchuenensis Bur. & Franchet (/952) LEGUMINOSAE Dalbergia stenophylla Prain* 126 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 SIMAROUBACEAE Picrasma quassioides Benn.* RUTACEAE Euodia rutacarpa (Juss.) Bentham var. bodinieri (Dode) Huang (1947) EUPHORBIACEAE Acalypha australis L. (1973) Mallotus tenuifolius Pax* (author as Muell.-Arg. in Chu & Cooper, 1950) BUXACEAE Sarcococca humilis Stapf (syn.: S. hookeriana Baillon var. humilis Rehder & Wilson) RHAMNACEAE Rhamnus davuricus Pallas (2024) CELASTRACEAE Euonymus kiatschovicus Loes. (2075) Microtropis triflora Merr. & Freeman (20/8) VITACEAE Parthenocissus tricuspidata (Sieb. & Zucc.) Planchon* Tetrastigma hemsleyanum Diels & Gilg (20/0) SABIACEAE Meliosma dilleniifolia (Wallich ex Wight & Arnott) subsp. flexuosa (Pampan.) Beus. 1949 Meliosma oldhamii Mig.* (syn.: M. pinnata (Roxb.) Maxim. subsp. barbulata Cufod. var. oldhamii (Maxim.) Beus.) BALSAMINACEAE Impatiens cf. exiguiflora Hooker f. (1960) THEACEAE Eurya loquaiana Dunn (2028) Eurya obtusifolia Chang (2004) LYTHRACEAE Rotala indica (Willd.) Koehne (2084) ARALIACEAE Nothopanax davidii (Franchet) Harms ex Diels* CORNACEAE Cornus kousa Hance var. angustata Chun (2032) 1983] BARTHOLOMEW ET AL., METASEQUOIA 127 ERICACEAE Rhododendron fortunei Lindley subsp. discolor (Franchet) Chamberlain (2040) MyRSINACEAE Ardisia crispa (Thunb.) DC.* Ardisia japonica (Thunb.) Blume (/926) OLEACEAE Jasminum urophyllum Hemsley (2078) GENTIANACEAE Tripterospermum affine (Wallich) H. Sm. (1927) LABIATAE Clinopodium polycephalum (Vaniot) C. Y. Wu & Hsuan (1968) Elsholtzia ciliata (Thunb.) Hylander (1962) Elsholtzia flava Bentham (1945) Mosla scabra (Thunb.) C. Y. Wu & H. W. Li (1965) SCROPHULARIACEAE Veronicastrum caulopterum (Hance) Yamazaki (2008) RUBIACEAE Anotis hirsuta (L. f.) Boerl. (2006) CAPRIFOLIACEAE Lonicera japonica Thunb.* Viburnum cylindricum Buch.-Ham. ex D. Don (/930) Viburnum foetidum Wallich (2002) COMPOSITAE Ainsliaea gracilis Franchet (1970, 2041) Lactuca graciliflora (Wallich) DC. (20/7) Solidago decurrens Lour. (2016) MONOCOTYLEDONES GRAMINEAE Isachne nipponensis ae (1959 Microstegium nudum (Trin.) A. Camus (1966) Pennisetum See (L..) Sprengel (1948) CYPERACEAE Bulbostylis densa (Wallich) Hand.-Mazz. (1969) ZINGIBERACEAE Zingiber mioga (Thunb.) Roscoe (2022) 128 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 LITERATURE CITED a gate B., D. E. Bourrorp, A. L. CHANG, Z. CHENG, T. R. DUDLEY, S. A. HE, n,Q. Y. Li, J. L. Luteyn, S. A. SPONGBERG, S. C. Sun, Y. C. TANG, J. S. Hi ie S. Yinc. 1983. The 1980 Sino-American Botanical Expedition to western Hubei Province, People’s Republic of China. J. Arnold Arbor. 64: 1-103. 1983. Bevper, J., & D. O. WuNANDS. 1979. ee glyptostroboides. Dendroflora 15, 16: 24, 25. [In Dutch, English summa BOcHER, T. W. 1964. Morphology of 7 vegetative body of Metasequoia glyptostro- boides. Dansk Bot. Ark. 24: 1-70. CHANEY, R. 948. The bearing of the living Metasequoia on problems of Tertiary paleobotany. Proc. Natl. Acad. U.S. A. 34: 503-515. Cuu, K. L., & W. C. Cooper. 1950. An ecological reconnaissance in the native home of Metasequoia glyptostroboides. Ecology 31: 260-278. Davis, M. B. 1976. Pleistocene biogeography of temperate deciduous forests. Geoscl. & Man 13: 13-26. er ee H. 1976. Metasequoia—fossil and living. Bot. Rev. 42: 215-315. Additions. . 43: 281-284. 1977. eee J. L. 1953. The California Academy-Lingnan dawn-redwood expedition. Proc. Calif. Acad. Sci. 28: 25-58 Hu, H. H., & W. C. CHENG. 1948. On the new family Metasequoiaceae and on Metase- quoia ‘glyptostroboides, a living species of the genus Metasequoia found in Szechuan and Hupch. Bull. Fan Mem. Inst. Biol. II. 1: 153-166. Hu, S. Y. 1980. The Metasequoia flora and its phytogeographic significance. J. Arnold Arbor. 61: 41-94. eae H. 1976. The living fossil—the past a present Metasequoia glyptostroboides u& — Acta Phytotax. Sin. 14(2): 51-54. Liv, = C. ees & P. L. Su. 1978. Ska 144 pp. Hubei People’s Press, Hubei. oa Chines Miki,S. 1941. On fe change of flora in eastern Asia since Tertiary Period. I. The clay or lignite beds flora in Japan with special reference to the Pinus trifolia beds in central Hondo. Japanese J. Bot. 11: 237-303 B. B. _E. B. UNIVERSITY OF CALIFORNIA CARNEGIE MUSEUM OF NATURAL BOTANICAL GARDE HISTORY BERKELEY, Eee 94720 4400 Fornes AVENUE Present address: PITTSBURGH, PENNSYLVANIA 15213 CALIFORNIA ACADEMY OF SCIENCES Present address: GOLDEN GATE P HARVARD Sy het HERBARIA SAN FRANCISCO, Cee 94118 22 Divinity AVE CAMBRIDGE, Reka eres 02138 S.A. S. THE ARNOLD ARBORETUM OF ARVAR I ITY 22 Diviniry AVENUE CAMBRIDGE, MASSACHUSETTS 02138 1983] DANIEL, HOLOGRAPHIS 129 SYSTEMATICS OF HOLOGRAPHIS (ACANTHACEAE) THOMAS F. DANIEL MEXICAN ACANTHACEAE have received little taxonomic attention since Stand- ley’s treatment included in his Trees and Shrubs of Mexico (1926). As a result of field work associated with several floristic projects and the efforts of numer- ous individual collectors, the number of collections of Mexican Acanthaceae has greatly increased during the five decades since Standley’s flora was pub- lished. However, there have been no comprehensive studies of Holographis or of the genera here considered congeneric with it. Work on these genera has been limited to regional floristic treatments and descriptions of new taxa. In the present treatment, the relationships of two genera recognized by Stand- ley and subsequent workers and a third, more recently described genus are discussed. Both Berginia Harvey ex Bentham & Hooker and Lundellia Leonard are considered to be congeneric with the older genus Holographis Nees. In addition to the new combinations in Holographis, three species from Mexico are described as new to science. This study also presents information on aspects of the ecogeography and phenology of the species. Holographis comprises ten species of shrubby perennials that are confined to arid and semiarid habitats in Mexico. Most species are restricted in their distribution and are known from relatively few collections. The genus is char- acterized by the following combination of characters: spicate inflorescences, a five-parted calyx, a relatively small, strongly zygomorphic corolla, four mono- thecous stamens of approximately equal length, pubescent thecae that lack any appendages, a short staminode, and an ellipsoid capsule bearing four seeds. In addition, the plants lack cystoliths, and all but three of the species have whorled leaves. TAXONOMIC HISTORY AND INTRAFAMILIAL RELATIONSHIPS Based on a Mexican collection of Ehrenberg, Nees described the genus Holo- graphis and the species H. ehrenbergiana Nees in the supplement to his treat- ment of the Acanthaceae in De Candolle’s Prodromus (1847). He placed the genus in tribe Gendarusseae subtribe Gendarusseae “section” Pseudo-Aphe- landreae and noted affinities with Hemigraphis Nees. Bentham and Hooker (1876) described the genus Berginia based on a Coulter collection from North America. The genus was placed in tribe Justicieae subtribe Asystasieae, and affinities with Stenandrium Nees were noted. Bentham and Hooker did not see any material of Holographis, but based on Nees’s description they included that genus in tribe Justicieae subtribe Eujusticieae, where it was grouped with © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 129-160. cm 1983. 130 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 Aphelandra R. Br. and Geissomeria Lindley. In 1885 Asa Gray described Pringleophytum A. Gray from a Pringle collection from Sonora, Mexico, noting an alliance between his genus and Holographis. Brandegee (1889) suggested that P. lanceolatum A. Gray and B. virgata Harvey ex Bentham & Hooker were conspecific. Rose (in Vasey & Rose, 1890) described a second species of Berginia, B. palmeri Rose, and confirmed Brandegee’s notion that P. /anceo- latum and B. virgata were the same species. Lindau (1895) provided accounts of both Holographis and Berginia in his influential treatment of the family in Die Natiirlichen Pflanzenfamilien. He placed both genera in subfam. Acan- thoideae tribe Aphelandreae, which included members of the family that have a corolla imbricate in bud and with an upper lip, an androecium consisting of four monothecous stamens, and typical Spaltenpollen. Holographis and Neri- acanthus Bentham were distinguished from other genera of the Aphelandreae (including Berginia) largely by the presence of a staminode. Berginia was also reported to have an unusual pollen type with three pores, each flanked by two furrows. Lindau’s conception of Berginia was probably based on a misidentified specimen, since the genus has a staminode and tricolpate pollen. In 1906 Brandegee described an unusual plant from the Chihuahuan Desert as Holographis ilicifolia Brandegee, but he was uncertain of its generic position. Subsequently, two additional species of Holographis (H. pallida Leonard & Gentry and H. parayana Miranda) and a species and a variety of Berginia (B. hintonii Leonard and 8B. virgata var. glandulifera Leonard & Morton) have been described. Leonard and Gentry (in Gentry, 1948) noted that H. pallida might eventually prove worthy of generic status due to the apparent uniqueness of its connivent anthers. They also noted the similarity of this species to species of Berginia and doubted whether the two genera could be kept separate. In 1959 Leonard described the genus Lundellia from southwestern Mexico. He claimed that its closest affinity was with the large American genus Aphelandra. Bremekamp (1965) suggested revisions of tribes and subtribes within the family. Although he did not list the genera that comprise his subfamilies and tribes, if one uses his scheme, the species under consideration in this study would be included in subfam. Acanthoideae tribe Aphelandreae, which contains members of the family having four monothecous stamens, two ovules per ovary cell, a distinctly bilabiate corolla, and prolate, tricolpate pollen (except in Sten- andrium subg. SPHAEROSTENANDRIUM), but lacking cystoliths. According to Bremekamp, this tribe is confined to the neotropics. The closest relatives of Holographis, and the only other members of Lindau’s Aphelandreae that occur in Mexico, are Stenandrium and Aphelandra. In Mex- ico these three genera can be distinguished by the following key: 1. Corolla subactinomorphic, lobes homomorphic; plants often acaulescent. ........ see ie ai a Bea ig ae co useage Ft a aca eens Gas Fae eg eee Stenandrium. 1. Corolla strongly bilabiate, lobes heteromorphic; plants rarely acaulescent. 2. Corolla orange, red, or purple (yellow in A. verticillata), 30-70 mm long. ...... BS tics Gyan Spann eons Gavecdet anges dokteanset tag cad an eneses, ace Paseo arna ten aoeaed Aphelandra. 2. Corolla yellow, pink, rose-purple, or white, 7-18 mm long. ...... Holographis. Holographis appears to be rather closely related to Aphelandra. There is no nown substantial morphological character that invariably separates them. 1983] DANIEL, HOLOGRAPHIS 131 Indeed, the two genera show some of the same trends (e.g., whorled leaves and spinose-toothed laminae), although they are usually readily distinguishable by a suite of characters. In Aphelandra the bracts are often brightly colored, the length of the corolla can reach 75 mm, and a staminode is rarely present (Wasshausen, 1975). In addition, Aphelandra is primarily a genus of damp habitats, with a center of diversity in South America. Only ten species are known to occur in Mexico (Wasshausen, 1975), and these are largely restricted to the wetter habitats in the southern part of the country. In contrast, Holo- graphis does not have colored bracts, the length of the corolla reaches only 18 mm, and a staminode is always present. The plants are restricted to Mexico, where they occur in arid and semiarid associations throughout the country. It is tempting to suggest that Holographis represents an evolutionarily spe- cialized line (adapted for the northern desert regions into which it radiated) of some part of the large and variable genus Aphelandra or its ancestor. Some of the differences between the genera may have arisen in response to (or may be maintained because of) the apparent differences in pollination mechanisms. The small, usually pale-colored flowers that contain little nectar (as observed in H. ilicifoliaand H. argyrea) suggest that Holographis is pollinated by a small bee or a fly. In contrast, the large, nectar-yielding, usually reddish flowers of Aphelandra are commonly hummingbird pollinated (Wasshausen, 1975). MORPHOLOGY Hasirt. Plants of Holographis are erect, often intricately branched subshrubs or shrubs to 2 m tall, arising from a stout, woody base. Variation in habit due to exposure is particularly observable in specimens of H. ilicifolia and H. ehrenbergiana. Exposed or drought-induced forms tend to be compact shrubs with small leaves and close nodes. Stems. The epidermis of the persistent, woody stems Stegner ny exfoliates 1 in S e surface is smooth, soon becoming conspicuously fluted, and is variably pubes- cent. Leaves. The leaves of Ho/ographis are either opposite (subopposite in some individuals) or whorled, depending on the species. Seven species (/1. il/icifolia, H. ehrenbergiana, H. pueblensis, H. anisophylla, H. pallida, H. hintonti, and H. parayana) have whorled leaves, with four at each node. Holographis tamau- lipica, H. argyrea, and H. virgata have opposite leaves. In H. pueblensis the lower leaves are sometimes opposite or subopposite, and the upper leaves are whorled. The leaves are sessile or petiolate, simple, and usually entire. The leaves of H. ilicifolia are unusual in their chartaceous texture and spinose- toothed margins. The lamina varies from lanceolate through ovate, elliptic, and orbicular to obovate and is 3-64 by 2-28 mm. The venation of all species of Holographis except H. ilicifolia is brochidod- romous; that of H. i/icifolia is craspedodromous. In his study of leaf architecture in the Acanthaceae, Sreemadhaven (1976) noted the occurrence of both of these 132 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 venation types in the Aphelandreae and considered pinnate craspedodromy to be derived from the pinnate-camptodromous condition. The leaves tend to have trichomes similar to those of the stem from which they arise, but dendroid trichomes are present on the unfolding leaves of Holographis argyrea. VestiTuRE. Holographis has both glandular and eglandular trichomes. The latter are more common, and when glandular trichomes occur, they are usually intermixed with eglandular ones. Trichomes of the vegetative axes are eglan- dular, 0.05-1 mm long, and erect to flexuose, retrorse, or ascendant. Depending on the species, the calyxes and the inflorescence axes, bracts, and bractlets can have glandular trichomes in addition to eglandular ones. The capitate glandular trichomes are 0.05—0.5 mm in length. Trichomes can also occur on most parts of the flowers, fruits, and seeds. Those of the seeds are unusual in being den- droid, often with circinate branches. Branched trichomes are not usually found on other parts of the plant. INFLORESCENCES. The inflorescences are axillary or terminal bracteate spikes (usually reduced to two flowers in Holographis ehrenbergiana) to 15 cm long. The spikes are congested, erect, and relatively short (1-6 cm long) in most species, but they are often loose, lax, and longer in others (especially taxa in the H. virgata complex). The flowers are sessile in the axil of two isomorphic bractlets and a bract and are opposite (alternate in H. i/icifolia) at the nodes of the inflorescence axis. FLowers. The calyx is gamosepalous, five parted, and radially symmetrical. The tube is shorter than the homomorphic lobes. The calyx is fairly constant in length within a species, and its outer surface is often glandular. The corolla is gamopetalous, pentamerous, and bilaterally symmetrical. It varies from yellow and white to pinkish and purplish, often with colorful markings. The orientation of the corolla varies from horizontal (e.g., in Holo- graphis pallida, H. parayana, and H. tamaulipica) to erect (e.g., in H. pueb- lensis, H. argyrea, and H. ehrenbergiana). The tube is usually shorter than the limb and is slightly ampliate toward the apex (except in H. argyrea). The limb is bilabiate, the upper lip with two petal lobes that are fused for most or all of their length, and the lower lip with three large, basally fused lobes. Two prom- inent ridges are often evident in the center of the lower lip. In bud the petal lobes are imbricate, with the upper lip innermost, concealed by the lateral and central lobes of the lower lip; the lower-central lobe is outermost. The corollas of most species have eglandular trichomes on the outer surface, but those of H. parayana, H. anisophylla, and H. hintonii have glands as well. The corolla of H. anisophylla is somewhat unusual in having glands on the inner as well as the outer surface; that of H. tamaulipica is glabrous on the outer surface, with the inner surface usually conspicuously pubescent along the central lobe of the lower lip and/or near the junction of the corolla and the filaments. The androecium consists of four epipetalous stamens and a staminode. The staminode is a short, filamentlike projection arising on the dorsal side of the corolla between the posterior pair of stamens; it lacks a theca, may be either glabrous or pubescent, and is 0.1-1.3 mm long. In most species the staminode 1983] DANIEL, HOLOGRAPHIS 133 emerges from the corolla at or about the same position as the filaments, although in Holographis ilicifolia it arises from the corolla up to 2 mm above the insertion of the filaments. The filaments of the fertile stamens emerge from the corolla near the apex of the tube and are either glabrous or pubescent. The anthers are monothecous and pubescent. The four thecae are frequently connivent at their apices and are basally rounded, lacking any appendages. Pollen of at least one individual of each species (except Holographis an- isophylla) was examined with a scanning electron microscope and/or a light microscope for shape, size, and sculpturing. Fresh pollen grains are prolate and generally uniform in size, with polar diameter (27.5-)35—40(—45) um and equa- torial diameter (17.5—)22.5-—27.5(-30) wm. The grains are 1.3 to 1.6 times longer than wide. FiGureE | illustrates some of the variation in pollen of species of Holographis. The micrographs show that the grains are tricolpate. The colpi are often bifurcate near the poles, and the bifurcations from each colpus can fuse with those of the other colpi below the poles. The gynoecium consists of a bicarpellate, superior ovary, a compound style, and usually two short stigma lobes. The ovary sits on a cuplike, fleshy disc. The style is filiform for most of its length but is conspicuously flared near the apex in some species. The presence or absence of trichomes on the style is usually constant for a species, but the styles of Holographis pallida and H. virgata can be either glabrous or pubescent. The stigma is either distinctly or inconspicuously bilobed, with the lobes often asymmetric. Fruits. The ovary develops into an indurate, two-valved capsule that dehisces explosively when mature. The capsule is ellipsoid in shape, splits to the base, and has no stalk; its outer surface is either glabrous or pubescent. The seed- bearing retinacula are hooklike and subequally inserted on the inner capsule wall. The seeds (up to four per capsule) are laterally flattened, suboval in shape, and either glabrous or pubescent. When present, the trichomes on the seeds are dendroid. ECOGEOGRAPHY AND PHENOLOGY Holographis is endemic to Mexico, from Baja California and northwestern Sonora, eastward to Tamaulipas, and southward to central Chiapas (Maps I, 3). Despite the extensive range of the genus in Mexico, most of the species have restricted distributions and are known from relatively few collections. Holographis hintonii and H. anisophylla are known only from the holotype, and H. argyrea, H. tamaulipica, and H. pueblensis are known from only two localities. Holographis ehrenbergiana and H. virgata have larger ranges, how- ever, and can be common in certain regions. There is no concentration of species in any one region of Mexico; in fact, the species are mostly allopatric, the only exceptions being Holographis ehren- bergiana, H. tamaulipica, and H. pueblensis, which overlap in range. There does appear to be a correlation of species with many of the arid and semiarid regions of Mexico, usually one species per region. Holographis ilicifolia 1S restricted to the Chihuahuan Desert, H. virgata to the Sonoran Desert region, H. pallida to the thorn forest south of the Sonoran Desert, H. tamaulipica to 134 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 El. Variation in pollen of Holographis species: a, H. virgata subsp. glandulifera Fic ~ var. slandulifera (Moran 3928), equatorial view, X 775; b, H. argyrea (T. Daniel 1242), equatorial view, < 900;c, H. ehrenber eana{ Gonsie Quintero 3581), equatorial view, x 900; d, H. pallida (H. Gentry 7022), equatorial view, X 775: e, H. ehrenbergiana (Gonzd- lez Quintero 3581), polar view, X 900: f, H. Pie giana (Gonzéilez Quintero 3581), equatorial view, X 900 oe the mesquite scrubland of central Tamaulipas, H. pueblensis to the arid Valley of Tehuacan, and H. argyrea and H. hintonii both to the arid tropical scrub of the Rio Balsas Basin (however, their ranges are not known to overlap); 1 ehrenbergiana is especially common in the Rio Panuco relict desert region of Hidalgo but also occurs in the semiarid regions throughout northeastern Mex- ico. Holographis parayana and H. anisophylla both occur in regions of tropical deciduous forest that are somewhat isolated from the ranges of other species in the genus. In his revision of the large genus Aphe/andra, Wasshausen (1975) noted the pronounced endemism of many of the species and the wide scattering of indi- viduals. Holographis appears to be closely related to Aphelandra (see discussion 1983] DANIEL, HOLOGRAPHIS 135 (ess Te Peal ee 100 200 3X0 400 Kilometers Map 1. Northern and central Mexico, showing distribution of Holographis aniso- phylla (plus sign), H. ehrenbergiana (solid squares), H. ilicifolia (solid dots), H. pallida (solid triangles), H. tamaulipica (open squares), and H. virgata (circles). under Taxonomic History and Intrafamilial Relationships) and shows the same phenomena. Additional, local species of Holographis will undoubtedly be dis- covered as botanical exploration of Mexico continues. The species occur in several habitats at elevations from near sea level to 2000 meters. The plants are generally found in arid and semiarid associations including desert scrub, thorn forest, and tropical deciduous forest. The most frequent habitats are on exposed rocky slopes and along gravelly washes and arroyos. Only Holographis ilicifolia, H. ehrenbergiana, and H. virgata are known from sufficient collections to determine their flowering periodicity adequately. Although both H. ilicifolia and H. ehrenbergiana have been collected during most months of the year, the majority of specimens of each species were taken during the period from July through September, corresponding to the summer rainy season in the Chihuahuan Desert region and the dry regions of north- eastern Mexico where these species occur. Members of the H. virgata complex have been collected in flower during every month except July, August, and September, although peak flowering occurs from March through May, corre- sponding to the end of the season of winter rains in the western section of the Sonoran Desert. The few collections of the other species of Ho/lographis suggest that H. pallida, H. hintonii, H. tamaulipica, and H. parayana flower predom- inantly January through April, H. pueblensis in June and July, H. argyrea in October and November, and H. anisophylla in November and December. Fruiting collections are not known for all of the species, and fruits and flowers are seldom found together on specimens of those species whose fruits are known. It appears that fruiting begins several weeks after flowering has started. 136 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 TAXONOMIC TREATMENT Holographis Nees in DC. Prodr. 11: 728. 1847. Type species: Holographis ehrenbergiana Nees. Berginia Harvey ex Bentham & a Gen. Pl. 2: 1096. 1876. Type species: B. virgata Harvey ex Bentham & Hoo Pringleophytum A. Gray, Proc. ere por Arts 20: 292. 1885. Type species: P. lanceolatum A. Gray Lundellia Leonard, Wrightia 2: 1. 1959. Type species: L. argyrea Leonard. Erect to spreading subshrubs to shrubs to 2 m tall, arising from stout, woody base. Older stems woody, pubescent, or glabrate; younger stems green, purplish, or silvery white, terete to quadrate to ellipsoid in cross section, the surface smooth to striate to fluted, pubescent, often densely so. Leaves opposite (rarely subopposite) or whorled (4 per node), sessile to petiolate; lamina linear-lan- ceolate to obovate, the margin entire (spinose toothed in H. ilicifolia), flat or revolute, the surfaces pubescent, often sparsely so, with abaxial surface punctate in some species. Inflorescences axillary or terminal spikes (usually reduced to 2 flowers in H. ehrenbergiana), flowers sessile, opposite (alternate in /. ilici- folia) along spike axis, subtended by 2 paired, isomorphic bractlets and a bract; bracts and bractlets usually similar in shape, often of different relative lengths. Calyx deeply 5-lobed, lobes homomorphic; corolla yellow or white to purplish or pinkish, bilabiate, pubescent (glabrous in H. famaulipica) on outer surface, the tube usually shorter than the lips, usually ampliate above, the upper lip bilobed, the lower lip trilobed, with lobes spatulate, central lobe usually con- spicuously larger and extending beyond lateral ones and usually with 2 prom- inent ridges down middle; stamens 4, included or slightly exserted and appressed to upper lip, the filaments glabrous or pubescent, the anthers monothecous, with thecae introrse, pubescent, often connivent; st de short, borne between posterior pair of stamens, glabrous or pubescent, theca lacking; style terminal, filiform, glabrous or pubescent, either flared near apex into distinct, equal or unequal stigmatic lobes or slightly flared and with stigmatic lobes inconspic- uous. Capsules brown, ellipsoid, bilocular, splitting to base, glabrous or pubes- cent on outer surface; retinacula 4 per capsule, hooklike. Seeds 4 per capsule (or fewer by abortion), laterally flattened, suboval, surface and margins glabrous or pubescent. DIsTRIBUTION. Mexico: Baja California Norte, Sonora, Coahuila, Nuevo Leon, and Tamaulipas, southward to Chiapas. KEY TO THE SPECIES OF HOLOGRAPHIS —_ . Leaves opposite (rarely subopposite). 2. Cauline trichomes ascendant-appressed; inflorescence spikes axillary; bracts 2-3 mm long, glabrous; calyx 4-5 mm long; corolla glabrous on outer surface. ..... fei aatidicnte Whe deter weed weg bodes ay ees bony ge ees ee aa ee eee ee eee aA. Seen aes Cauline trichomes straight, flexuose or retrorse; inflorescence spikes terminal o branches; bracts 3-8 mm long, pubescent; calyx 5-8 mm long; corolla sone on outer surface. N 1983] DANIEL, HOLOGRAPHIS 137 3. Cauline trichomes 0.05-1 mm long; corolla reddish pink, 9-14 mm long; stamens 2.8-5 mm long; Sonoran Desert region. ............ . H. virgata. 3. Cauline aa 0.05 mm ee or less; corolla yellow, 7—9.5 mm long; sta- Mion? GueneO; 4.02 ntraiccoetwst ede sags 3. AH. argyrea. 1. Leaves whorled, 4 per node ioe leaves of H. pueblensis sometimes opposite to subopposite). 4. Leaves orbicular, 0.8-1.3 times longer than wide, margin spinose toothed; bracts alternate; Chihuahuan Desert region. ..........----.++050-055 4. H. ilicifolia. Leaves lanceolate to obovate, 1.5—4.5 times one than wide, ae entire; bracts opposite; not of Chihuahuan Desert regio 5. Corolla 8-10 mm long, glandular on noe surface; stamens 2.5-3.5 mm long. 6. Cauline trichomes ascendant-appressed, 0.1-0.3 mm long; bracts triangular to lance-subulate, eglandular, emucronate; calyx 2-3 mm long; style 5-6 mm long; ovary glabrous. ............---+ sees eee eee 5. H. parayana. Cauline eae retrorse, 0.05-0.1 mm long; bracts lance-ovate to orbicu- lar, glandular, mucronate; calyx 4-5 mm long; style 3.5—-4 mm long; ovary pubescent. 7. Inflorescence axis glandular-pubescent with glandular trichomes 0.05 mm long and straight eglandular trichomes 0.05—-0.1 m m long; bracts lance-ovate to ovate, 2.5-3 mm long; bractlets lance-ovate to ovate, 2— 2.8 mm long; stamens 2.5-3 mm long; staminode pubescent. ....... diet cl Ria ete’ h daliaasa shied oe, ceo Bae ee ee 6. H. anisophylla. 7. Inflorescence axis eglandular with crooked, interwoven trichomes to 0.8 > ox mm long; bracts broadly ovate to orbicular, 1.5—2 mm long; bractlets peers 1.5-2 mm long; stamens 3-3.5 mm long; staminode gla- 7. H. hintonit. 1 1,,] 5. Corolla 10-1 8mm eae pubescent with eg trichomes on outer surface; stamens 3.5-11 m 8. Cauline ee retrorse, 0.05—-0.2 mm long; laminar margin flat; bracts mucronate; corolla white to lavender, 10-12 mm long, es at anthe- sis; style 5.3-6 mm Jong. «2... 2.1.00 ce ence nee tee aen es 8. H. pallida. . Cauline trichomes at to ascendant-appressed, 0.1-0.7 mm ae lam- inar margin revolute; bracts emucronate; corolla yellow, 12- 18 mm long, vertical at anthesis; style 12-16 mm long. 9. Inflorescence spikes many flowered, to 4.5 cm long; bracts 6.5—-11 by 1.5—2 mm; calyx 7-10.5 mm long. ............... 9. H. pueblensis. 9. peeere re spikes to 2 cm long, usually reduced to 2 flowers; bracts 1.5-5 by 0.5-1 mm; calyx 3.5-6 mm long. 10. H. ehrenbergiana. io) 1. Holographis tamaulipica T. F. Daniel, sp. nov. FIGURE 2. Frutex erectus. Caules juniores substriati pubescentes trichomatibus eglan- dulosis 0.2-0.6 mm longis. Folia opposita; laminae lanci-ovatae vel ellipticae, (15-)25-64 mm longae, (6-)9-28 mm latae, 2—4-plo longiores quam latiores, marginibus revolutis ciliatis. Inflorescentia spicata, bracteae triangulatae vel ovatae, 2-3 mm longae, |.5-2 mm latae; bracteolae lanceolatae, 1.5—2.2 mm longae. Calyx 4-5 mm longus; corolla 9-11 mm longa, extus glabra; stamina 3.5-4 mm longa filamentis 2—-2.5 mm longis, pubescentibus; staminodium 0 0.3 mm longum, pubescens; stylus 6-6.5 mm longus, glaber. Capsula 9 mm longa, glabra. Erect shrub. Older stems pubescent or glabrate; younger stems green, terete in cross section, the surface somewhat striate, evenly pubescent, the trichomes 138 JOURNAL OF THE ARNOLD ARBORETUM [vVoL. 64 Ficure 2. Holographis tamaulipica yes holotype): a, habit, x 0.6; b, spike, x 3.5; c, dissected flower, X 4; d, gynoecium, eglandular, ascendant-appressed, 0.2-0.6 mm long. Leaves opposite, ascendant to horizontal; petiole to 10 mm long, pubescent; lamina lance-ovate to elliptic, attenuate at base, acute at apex, (15—)25-64 by (6-)9-28 mm, 2 to 4 times 1983] DANIEL, HOLOGRAPHIS 139 longer than wide, the margin entire, revolute, ciliate with erect to flexuose trichomes 0.2-0.8 mm long, the surfaces subglabrous or sparsely pubescent. Inflorescences axillary or terminal, pedunculate spikes to 9 cm long; peduncles 1-9 mm long; axes pubescent with erect to ascendant trichomes 0.05—0.2 mm long; flowers opposite at spike nodes; lower bracts (rarely absent) sterile; upper bracts fertile, triangular to ovate, 2-3 by 1.5-2 mm, submucronate to mucro- nate at apex, outer surface glabrous, margin often ciliate; bractlets lanceolate, shorter than or approximately same length as subtending bract, |.5—2.2 by 0.7— 1 mm, pubescent like bracts. Calyx 4-5 mm long, the lobes lance-subulate, mucronate at apex, pubescent like bracts; corolla white (label data), horizontal at anthesis, 9-11 mm long, glabrous on outer surface, the tube 4-5 mm long, ampliate above, the upper lip 3.5-5 mm long, with lobes 2-3 mm long, the lower lip 5-6 mm long, with lobes spatulate, 3.5—4.5 by 1.5-2 mm, stamens 3.5-4 mm long, the filaments 2—2.5 mm long, pubescent, the thecae 1.5-2 mm long, pubescent; staminode 0.2-0.3 mm long, pubescent; style 6-6.5 mm long, glabrous, flared at apex, the stigma unequally bilobed, with longer lobe to 0.5 mm long. Capsules 9 mm long, 3.8 mm in diameter, glabrous. Seeds not seen. Type. Mexico, Tamaulipas, vicinity of Victoria, ca. 320 m, 1 Feb.-9 April 1907, Palmer 135 (holotype, GH!; isotype, F!). This species is known only from the type specimens and one additional collection, both of which were collected near Victoria, Tamaulipas (Map 1) in the early spring. The paratype was collected on a shaded limestone ledge in oak woods at 1200 meters. Holographis tamaulipica is unique among the known species of Holographis in having a glabrous corolla. It does not appear to be very closely related to the other two species with opposite leaves. In cauline pubescence, flower size, and corolla orientation it resembles H. parayana. ADDITIONAL SPECIMEN EXAMINED. Mexico. TAMAULIPAS: 12 mi from river at Victoria on road to Jamauve, Moore & Valiente M. 6159A (us). 2. Holographis virgata (Harvey ex Bentham & Hooker) T. F. Daniel, comb. nov. Berginia virgata Harvey ex Bentham & Hooker, Gen. PI. 2: 1097. 1876. Type: “Cal- ifornia incola,”’ without date, Coulter 603 (holotype, k!; isotypes, GH!, K (2 sheets)!). Pringleophytum lanceolatum A. Gray, Proc. Amer. Acad. Arts 20: 293. 1885. Type: Mexico, Sonora, rocky hills 50 mi below Altar and 30 mi from Gulf of California, 1884, Pringle s.n. (holotype, GH; isotypes, F!, Ny!, Us!). Erect shrub to 2 m tall. Older stems often whitish, pubescent or glabrate; younger stems pale green, soon becoming either whitish or purplish, subquad- rate to quadrate in cross section, the surface striate, soon becoming conspic- uously fluted, evenly pubescent, with trichomes eglandular and erect, flexuose, or retrorse, 0.05-0.7(-1) mm long. Leaves opposite (or subopposite), ascendant, sessile to petiolate; petiole (if present) to 5 mm long, pubescent; lamina linear- lanceolate to obovate, acute to attenuate at base, acute to rounded (rarely emarginate) at apex, (4-)6-46 by 2-26 mm, 1.1 to 9.2 (to 17) times longer than 140 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 wide, the margin entire, flat to revolute, ciliate, the surfaces pubescent. Inflo- rescences terminal, loosely or densely bracteate spikes to 15 cm long; axes pubescent with eglandular trichomes or with mixture of eglandular and glan- dular trichomes; flowers opposite (to subopposite) at spike nodes; lower bracts (up to 6 series) often sterile, triangular to orbicular; upper bracts fertile, some- what foliaceous, lance-ovate to ovate to elliptic to orbicular, 3-8 by 1.5-5 mm, usually pubescent like inflorescence axis; bractlets lance-subulate to linear to lanceolate, approximately equal to or shorter than subtending bract, 2-6 by 0.6-1.5 mm, pubescent with eglandular or mixture of eglandular and glandular trichomes. Calyx 5-8 mm long, the lobes lanceolate or oblanceolate, pubescent like bractlets; corolla reddish pink with white markings, 9-14 mm long, pubes- cent on outer surface with eglandular trichomes (rarely with glandular trichomes as well in subsp. glandulifera var. glandulifera), the tube 5-8 mm long, ampliate above, the upper lip (2.5-)3-5 mm long, with lobes 1.5—4 mm long, the lower lip 4-8 mm long, with lobes spatulate, 2.5-6 by 1.5-3.5 mm; stamens 2.8-—5 mm long, the filaments |.5-4 mm long, pubescent or nearly glabrous, the thecae 1.6—2.5 mm long, pubescent (anterior pair usually more densely so); staminode 0.5—0.8 mm long; style 5-7 mm long, glabrous or pubescent, flared at apex, the sg subequally bilobed, 0.3-0.5 mm long. Capsules 7-12 mm long, 3- 5 mm in diameter, pubescent or essentially glabrous, with trichomes often aa to apex, eglandular, 0.05—0.3 mm long, the retinacula 1.5—2.5 mm long. Seeds obliquely cordate to widely elliptic, 2—-3.5 by 2-3 mm, pubescent with dendroid trichomes to 0.3 mm long. As treated in this study, Holographis virgata comprises three taxa formerly recognized in Berginia. The most perplexing problem remaining in Holographis is the status of these taxa from Baja California and Sonora. The geographic discontinuity between the eglandular northern populations and the glandular southern populations justifies their recognition at the rank of subspecies. Due to minor differences in pubescence and an apparent difference in habitat ele- vation among the southern populations, two varieties are recognized in the southern subspecies. I feel confident that these plants are all part of the same species. The infraspecific treatment presented here, however, may presuppose a greater degree of understanding of this complex than can be obtained from herbarium material alone. Although the differences in pubescence among the three taxa are clear and most specimens can be easily identified, the nature of these distinctions is subtle. Pubescence can be influenced by environmental factors. Field work in Baja California and cultivation of these plants in an experimental garden will undoubtedly increase our understanding of their rela- tionships. Key to the Infraspecific Taxa of Holographis virgata 1. Inflorescence axes, bracts, bractlets, and calyx eglandular (rarely with inconspicuous glands on bractlets and calyx); bractlets 0.6—1 mm wide; Sonora and Baja California HOTH-Or 2S. suey ana dny keen yh teen Gores 2a. H. virgata subsp. virgata. . Inflorescence axes, : briots. bractlets, and calyx conspicuously glandular; bractlets 1- mm wide; Baja California, mostly south of lat. 28°N. — 1983] DANIEL, HOLOGRAPHIS 141 2. Leaves and young oat scaberulous, with trichomes retrorse, 0.05-0.1 mm long; usually below 600 m 2b. H. virgata subsp. glandulifera var. glandulifera. 2. Leaves and young ies hirtellous, with trichomes straight to flexuose, 0.1-1 mm long; usually above 600 m alt. ... 2c. H. virgata subsp. glandulifera var. palmeri. 2a. Holographis virgata (Harvey ex Bentham & Hooker) T. F. Daniel subsp. virgata Trichomes of younger stems erect to retrorse, 0.05-0.1 mm long. Leaves sessile or subsessile; lamina linear-lanceolate to lanceolate to elliptic (to obovate), (4-)7-43 by 2-9(-14) mm, (1.6 to) 3 to 17 times longer than wide. Inflorescence axes pubescent with eglandular, erect to retrorse trichomes 0.05-0.2(-0.5) mm long; bracts 1.5-3 mm wide, pubescent like inflorescence axis; bractlets 0.6-1 mm wide, pubescent like bracts (rarely also with scattered glandular trichomes to 0.2 mm long). Lower lip of corolla 5-8 mm long, lobes 4-6 by 2.5-3.5 mm. Capsule essentially glabrous to pubescent, trichomes often restricted to apex. DISTRIBUTION AND HABITATS. Sonoran Desert of western and central Sonora and southern Baja California Norte (Map 2); on gravelly slopes and along arroyos and washes, 15-1300 m alt.; with species of Jatropha L., Pachycereus (A. Berger) Britton & Rose, Pachycormus Cov. ex Standley, /dria Kellogg, Larrea Cav., Opuntia Miller, Acacia Miller, Cercidium Tulasne, Simmondsia Nutt., Croton L., and Carnegiea Britton & Rose. FLOWERING AND FRUITING. Flowering October through June; ee simul- taneously, more commonly during latter portion of flowering perio REPRESENTATIVE SPECIMENS. Mexico. BAJA CALIFORNIA Norte: Mina Desengafia, ca. 16 mi N of Punta Prieta, H. Gentry & Cech 8887 (MEXU, MICH, US); ca. 22 mi from Bahia Los Angeles on road to San Borja, H. Gentry & McGill 23313 (ASU, DES, MICH), 7 mi W of San Francisquito Bay, Harbison 41665 (sp); Las Animas Bay, J. Johnston 3509 (cas, GH, NY, UC); Paredones, Montufar 42 (eNcB); Agua de Higuera, Moran 7956 (Ariz, DS, MICH, SD, UC); Sierra San Borja, San Juan Mine, Moran 8060 (ps, sp, uc); Turners Is. Moran 13024 (sp, uc); Calmalli, Purpus 56 (ps, F, NY, UC, US); 8 mi N of Mission San mosillo, Abrams 13305 (ps, F); canyon back of Palma, 40 mi S of Hermosillo, Abrams 13348 (ps); ca. 31 mi E of Punta Cirio between Puerto Libertad and Caborca, Bowers & McLaughlin s.n., 13 April 1979 (Ariz); 10 mi NW of Hermosillo, C. Carter s.n., 13 April 1932 (micn); Beer Guaymas, Dawson 1075 (F, MICH); San Miguel de Horcasitas, Eisen s.n., May 1892 (us); Picu Mts., Altar Distr., H. Gentry 4479 (DES, MICH, MO); Sierra Cajon del Suen H. Gentry 11622 (MEXU, MICH); 22.2 mi N of Bahia Kino Nuevo, Hastings & Turner 64-40 (Ariz, DS, SD); 0.9 mi E of Pitiquito, Hastings & Turner 64- 7553 (ARIZ, F, MICH, MO; us): ca. 30 mi NE of Obregon, Soaaidae 75313 (ARiz): 8 mi 20.7 mi NE of Desemboque de San Ignacio, Van Devender & Kearns s.n., 17 Feb. 1977 (ARIZ). 142 JOURNAL OF THE ARNOLD ARBORETUM 1loow 200 Km [VOL. 64 2. Northwestern Mexico, showing distribution of infraspecific taxa of Holo- graphis virgata: subsp. virgata (solid squares), subsp. glandulifera var. glandulifera (solid triangles), subsp. glandulifera var. palmeri (open squares). 1983] DANIEL, HOLOGRAPHIS 143 This is the most widely distributed and most variable subspecies of Holo- graphis virgata. It occurs primarily to the north and east of the range of H. virgata subsp. glandulifera, but the two subspecies do come together in southern Baja California Norte near Calmalli (lat. 28°08’N, long. 113°24’W). At this locality Purpus collected specimens of H. virgata subsp. virgata, H. virgata subsp. glandulifera var. glandulifera, and intermediates. The intermediate plants on Purpus 56 (uc) have sparsely glandular infl , while other plants on the sheet have densely glandular (H. virgata subsp. glandulifera var. glan- dulifera) and eglandular (H. virgata subsp. virgata) axes. Duplicates of Purpus’s Brandegee 443 from the Sierra de la Laguna in the Cape Region of Baja California Sur contains sprigs of H. virgata subsp. virgata and of H. virgata subsp. glandulifera var. palmeri. Although there is some variation among the sprigs of H. virgata subsp. virgata in the length of the pubescence on the inflorescences, the variation falls within the range normally encountered in that subspecies. Annetta Carter, of the University of California (Berkeley) Herbarium, notes (in litt.) that during Brandegee’s trip from Magdalena Bay to San Quintin in 1889, he was in the vicinity of Calmalli in late April. This was the only time he passed through the range of Holographis virgata subsp. virgata. She notes that at that time, Brandegee could have collected the material of H. virgata subsp. virgata as represented on Brandegee 443. While working on his collec- tions of 1889 and 1890, he might have mixed material of the two subspecies. Holographis virgata subsp. virgata shows considerable variation in the length of the eglandular pubescence on the inflorescence axes and in the amount of pubescence on the capsules. It can be distinguished from H. virgata subsp. glandulifera by the general absence of glandular trichomes on the inflorescence; however, in some individuals the bractlets and calyx can contain inconspicuous glands 2b. Holographis virgata subsp. glandulifera (Leonard & Morton) T. F. Daniel, comb. et stat. nov., var. glandulifera Berginia Pe var. Regal Leonard & Morton, Contr. Dudley Herb. 4: 24. 1950. Type: Mex Baja California Sur, 32 miS of Mulegé, 15 March 1935, Shreve 710] (holotype, oe isotypes, ARIz!, ps!, F!, MICH!). Trichomes of the younger stems retrorse, 0.05—0.1 mm long. Leaves sub- sessile; petiole to 3 mm long; lamina lance-ovate to oblanceolate, 6-43 by 3- 22 mm, 1.3 to 3.8 times longer than wide. Inflorescence axes pubescent with mixture of glandular and eglandular trichomes, the former 0.1—0.3 mm long, the latter erect to flexuose, 0.05—0.3 mm long; bracts (2—)2.5-5 mm wide, lower ones pubescent like leaves, upper ones pubescent like inflorescence axis; bract- lets I1-1.5 mm wide, pubescent like inflorescence axis but usually with more numerous glands. Lower lip of corolla 4-5 mm long, lobes 3-4 by 1.5-—3 mm. Capsules pubescent. DISTRIBUTION AND HABITAT. Known from southern Baja California Norte, but more common in eastern and central Baja California Sur (MAP 2); rocky slopes, 144 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 arroyos, and gravelly washes to about 600 m alt.; with Lysi/oma Bentham, Mimosa L., Jatropha, Bursera Jacq. ex L., Pachycereus, Lemaireocereus Britton & Rose, Ruellia L., Fouquieria Kunth, and Cyrtocarpa Kunth FLOWERING AND FRUITING. Flowering November through May and into June; fruiting simultaneously, more commonly during latter portion of flowering season. REPRESENTATIVE SPECIMENS. Mexico. BAJA CALIFORNIA Norte: 20 mi W of Bahia San Francisquito, Humphrey 6830a (Ariz); diggings at Calmalli, Purpus 56, pro parte (uc). BAJA CALIFORNIA SuR: Purisima to Comondi, Brandegee s.n., 15 Feb. 1889 (uc); Comon- di, Brandegee s.n., 15 Feb. 1890 (F, ee 12.4 km SE of La Paz on road to Los Planes, A. Carter 2625 (DS, MEXU, MICH, SD, us); vic. of Rancho Los Burros, E of Llanos de San Pedro, NE of Comonda, A, Carter & Ferris 3430 (ps, MEXU, SD, UC); Arroyo Carrizal, E of Rancho El Horno (NE of San Xavier), 4. Carter & Ferris 3829a (cas, uc); S of Tinaja de Naucajoa, Cerros de Naucajoa (W of Llanos de San Juan), 4. Carter 4502 (uc); Puerto Escondido, Dawson 1098 (F, MicH); Las Cuevitas below Comondt, H. Gentry 4235 (DES, DS, MICH, MO); between San Ignacio and Los Martiles, H. Gentry 7872 (ARIZ, DS, MICH, UC); San Nicholas Bay, /. Johnston 3729 (cas), Carmen Is., Puerto Ballandra, uc, us); San Marcos Is., Arroyo de los Chivos, Moran 8999 (sp); Danzante Is., Moran 9249 (sp); island in Conception Bay, Rempel 197 (Ariz), 208 (Ariz); 14 mi S of Mulege, Shreve 7086 (ARIZ, F, GH, MICH, MO This variety resembles Holographis virgata subsp. virgata in the pubescence of the vegetative axis, and where the ranges of these taxa overlap, intermediates occur. In its distribution and in characters of inflorescence pubescence and bract width, this variety appears closer to H. virgata subsp. glandulifera var. palmeri. These latter taxa are distinguishable from each other by the relative lengths of the foliar and cauline trichomes and by the differences in elevation of their habitats. 2c. sarriaa eae subsp. glandulifera var. palmeri (Rose) T. F. Daniel, mb. et stat Berginia palmeri Rose, Contr. U. S. Natl. Herb. 1: 86. 1890. Type: Mexico, Baja California Sur, ie Rosalia, 15 March 1890, ie 272 (holotype, us!; isotypes, A!, ps!, F!, GH!, NY!). Trichomes of the younger stems erect to flexuose, 0.1-0.7(-1) mm long. Leaves sessile to petiolate; petiole to 5 mm long; lamina lance-ovate to obovate, (6-)11-46 by (4-)7-26 mm, 1.1 to 3.3 times longer than wide. Inflorescence axes pubescent with mixture of eglandular and glandular trichomes, the former 0.2-1 mm long, the latter 0.2-0.5 mm long; bracts 2.5—4.5 mm wide, pubescent like inflorescence axis; bractlets 1-1.5 mm wide, pubescent like inflorescence axis although glandular trichomes usually more numerous. Lower lip of corolla 5-6.5 mm long, lobes 2.5—4 by 2-3 mm. Capsules pubescent. DISTRIBUTION AND HABITAT. Eastern Baja California Sur (especially common in and around Sierra de la Giganta (Map 2)); among rocks on slopes and in arroyos above 600 m alt.; with Jatropha, Hechtia Klotzsch, Pachycormus, Franseria Cav., Lysiloma, Lemaireocereus, Colubrina Rich. ex Brongn., Bur- 1983] DANIEL, HOLOGRAPHIS 145 sera, Ficus L., Erythrina L., Agave L., Bernardia Miller, Viguiera Kunth, and Croton. FLOWERING AND FRUITING. Flowering October through June; fruiting predom- inantly March through June. REPRESENTATIVE SPECIMENS. Mexico. BAJA CALIFORNIA Sur: Sierra de Laguna, Brandegee 443 (uc); Arroyo Carrizal, E of Rancho El Horno (NE of San Xavier), A. Carter & Ferris 3819, pro parte (CAS, MEXU, UC, US); N ee of Pilon de las Parras, W of UN ai Carter & Sharsmith 4215 (CAs, F, MO, NY, UC, US); Sierra de la Giganta, S of La P W end of Valle de los Encinos (S side of on Giganta), A. Carter & Reese 4574 | ae MEXU, UC, US); Sierra de la Giganta, Mesa de San Geronimo, N of Rancho Viejo, A. Carter 5131 (CAS, GH, MICH, SD, UC, US); 5 mi S of San Bartolo, Gallagher 139 (asu); Sierra Giganta above Pto. Escondido, H. Gentry 3762 (A, ARIZ, DES, MICH, MO, UC); Volcan las Tres Virgenes, Moran 20459 (sp); La Ciénega Arroyo, Distr. of Buena Vista, Peters 206 (uc); 8 mi E of pass of Tres Virgenes Peak, Shreve 705]a (Ariz); El Purgatorio grade W of Santa Rosalia, Wiggins & Wiggins 18206 (CAs, DS, MEXU). Most (if not all) of the specimens from the vicinity of the Sierra de la Giganta were collected above 600 meters in elevation; however, Peters 206 (from the Cape Region) was collected at about 33 m alt. Ms Carter notes (in /itt.), “In the Cape Region, some plants occur both at low and higher altitudes whereas to the north (Sierra de la Giganta) the same taxon is restricted to higher altitudes. A case in point is Jatropha vernicosa which occurs above 2000 feet in the Sierra de la Giganta, but in the Cape Region it grows on coastal plain as well as at higher elevations in the mountains.” Peters’s specimen is unusual in having relatively few glands in the inflorescence, but it has the long, eglandular tri- chomes characteristic of this variety. All three collections of this variety from the Cape Region are unusual in their pubescence. The trichomes of the younger stems are scaberulous and 0.05-0.2 mm long, resembling those of Holographis virgata subsp. glandulifera var. glandulifera. The surfaces of the leaves are also scaberulous, but the margin and the veins have the long, flexuose trichomes characteristic of H. virgata subsp. glandulifera var. palmeri. Although some duplicates of A. Carter & Ferris 3819 from the Sierra de la Giganta contain both Holographis virgata subsp. glandulifera var. glandulifera and H. virgata subsp. glandulifera var. palmeri, Ms Carter notes (pers. comm.) that the material of the latter variety was collected at about 688 m. Material of the former variety (A. Carter & Ferris 3829a) was collected on the same day at about 584 m. It is not known if some of these collections were inadvertently mixed or if the two varieties indeed grow together. The difference in elevation between these two collections is not great. 3. Holographis argyrea (Leonard) T. F. Daniel, comb. nov. Lundellia argyrea Leonard, Wrightia 2: 1. 1959. Type: Mexico, Guerrero, cliff along stream near Chilpancingo, km 276, 21 Oct. 1943, Lundell 12603 (holotype, us!; isotypes, LL, MICH!). Erect, rounded shrub to | m tall. Older stems splitting in several lines, giving rise to numerous pronounced channels, glabrate; younger stems purplish, epi- ON JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 1lOo2°W =) amet Sega: nears ae ie) 50 100 150 200 Kilometers Southern Mexico, showing distribution of Holographis argyrea (solid dots), H. hintonii (solid triangle), H. parayana (solid squares), and H. pueblensis (open squares). dermis soon becoming silvery white, quadrate to ellipsoid in cross section, the surface fluted, evenly pubescent, with trichomes often very sparse, eglandular, retrorse, 0.05 mm long or less. Leaves opposite, ascendant to horizontal; petiole to 10 mm long, pubescent like stems; lamina ovate to orbicular (to obovate), rounded to acute (to subattenuate) at base, rounded to acute at apex, | 2-64 by 10-35 mm, 0.9 to 2.2 times longer than wide, the margin entire, flat to subrevolute, ciliate with erect to flexuose trichomes 0.05-0.6 mm long, espe- cially near base, the surfaces sparsely pubescent or subglabrous (surfaces of unfolding leaves densely tomentose with dendroid trichomes), abaxial surface punctate. Inflorescences densely bracteate, terminal spikes to 4.5 cm long; axes pubescent with mixture of glandular and eglandular trichomes, the former 0.2- 0.5 mm long, the latter erect, 0.1-0.3 mm long; flowers opposite at spike nodes; bracts lanceolate, 5.5-—7 by 2—2.5 mm, pubescent like inflorescence axis; bract- lets lance-subulate, approximately same length as subtending bract, 5-6.5 by 0.7-1.2 mm, pubescent like bracts. Calyx 6.5-7.5 mm long, the lobes lance- subulate, pubescent like bracts; corolla yellow with maroon markings, vertical at anthesis, 7-9.5 mm long, pubescent on outer surface with eglandular tri- chomes, the tube 5—5.5 mm long, slightly ampliate, the upper lip 1-2.5 mm long, with lobes 0.5-1.5 mm long, the lower lip 3.5-4 mm long, with lobes spatulate, 2.5-3.5 by 1.5-3.5 mm; stamens |.5 mm long, the filaments 0.5 mm long, pubescent, the thecae 1-1.3 mm long, pubescent; staminode 0.2-0.3 mm long, glabrous; ovary glabrous, the style 4 mm long, glabrous, flared at apex, 1983] DANIEL, HOLOGRAPHIS 147 the stigma unequally bilobed, with longer lobe to 0.5 mm long. Capsules not seen. DISTRIBUTION AND HABITAT. Known only from semiarid Rio Balsas basin in central Guerrero (MAP 3); rocky cliffs in low, deciduous thorn forest ca. 700 m alt.; with columnar cacti. FLOWERING AND FRUITING. Collected only during late October, but undoubtedly flowering well into November; fruiting apparently several weeks after flowering. ADDITIONAL SPECIMENS EXAMINED. Mexico. GUERRERO: Cafién de Zopilote, along road to Filo de Caballo, 0.6 mi W of jet. Hwy. 95, T. Daniel 1193 (aAsu, ENCB, GH, MICH), 1242 (CAS, DUKE, MEXU, MICH). Leonard (1959, p. 1) noted a relationship of Holographis argyrea with Aphe- landra. He believed the genus Lundellia could be distinguished from Aphe- landra by its corky, fissured stems, its oval, silvery laminas, its dendroid hairs, and its “compact solitary terminal glandular spikes of small yellow flowers with their deeply included, almost sessile stamens. ...”” Although H. argyrea would be an anomaly in Aphelandra, none of these features distinguish it from Holographis. Its closest relative appears to be H. virgata; both have opposite leaves, silvery stems, and terminal spikes. 4. Holographis ilicifolia Brandegee, Zoe 5: 236. 1906. Type: Mexico, Coahuila, Pefia (25°28'N, 102°34'W), Feb. 1905, Purpus 1327 (holotype, Uc!; iso- types, GH!, Mo!, NY!). FIGURE Twiggy, erect to spreading, often rounded subshrub to 3(—5) dm tall, arising from stout or tortuous woody base to 10 mm in diameter. Older stems woody, pubescent or glabrate; younger stems green, terete in cross section, the surface smooth to striate, evenly and densely pubescent, often making the stems appear pallid, the trichomes eglandular, retrorse to erect, 0.05—0.2 mm long. Leaves whorled, ascendant or decurved, sessile to subsessile; petiole (if present) 0.1— 0.5 mm long, pubescent; lamina coriaceous, orbicular, usually partially folded lengthwise along midvein, cuneate to truncate to subcordate at base, acute- aristate at apex with spine to 3 mm long, (1.5—)3—10 by 2.8-10 mm, 0.8 to 1.3 times longer than wide, the margin spinose toothed, with spines remote, 3 to 6 per side, 0.1-1.5 mm long, the surfaces sparsely pubescent, with trichomes erect to retrorse, 0.01—0.2 mm long. Inflorescences axillary spikes to 4 cm long; axes pubescent like younger stems; flowers alternate along spike axis; bracts ovate to triangular to subulate, (1—)1.5-3 by 0.6—-1 mm, pubescent like inflo- rescence axis; bractlets subulate, longer than subtending bract, (2-)2.5—4 by 0.6—1 mm. Calyx 5-8 mm long, lobes linear-subulate, outer surface pubescent like inflorescence axis, margin densely ciliate with flexuose, eglandular tri- chomes to 0.3 mm long and glandular trichomes to 0.05 mm long; corolla yellow, usually with dull maroon markings on outer surface, 10-14 mm long, pubescent on outer surface with eglandular trichomes, the tube 5-7 mm long, ampliate above, the upper lip 4-5.5 mm long, with lobes 1-3 mm long; the lower lip 5—7.5 mm long, with lobes spatulate, 4-5.5 by 2.5-4 mm; stamens 4.5—6 mm long, the filaments 3.5-4.5 mm long, more or less densely pubescent 148 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Holographis ilicifolia: a, habit (T. Daniel 659), X 0.6; b, leaf (7. Daniel Ficure 3. 674), X 7; c, flower (T. Daniel 674), X 3.5; d, aa (Correll & I. Johnston 21435), X 3.8; e, seed (Correll & I. Johnston 21435), X | with trichomes to 0.3 mm long, the thecae 1.2—1.5 mm long, pubescent along dorsal side; staminode 0.5—0.9 mm long, pubescent at base if at all; style 6- m long, glabrous, flared at apex, the stigma lobes indistinct. Capsules brown, ellipsoid, 8-10 mm long, 2—3.5 mm in diameter, glabrous or pubescent, 1983] DANIEL, HOLOGRAPHIS 149 trichomes to 0.1 mm long (rarely restricted 7 apex); retinacula 1-1.5 mm long. Seeds oval, asymmetric at base, 2.5-3.5 by 1.5—3 mm, densely pubescent with dendroid trichomes to 0.5 mm long. DISTRIBUTION AND HABITAT. Central and southwestern Coahuila and adjacent regions of northern Durango (Map 1); on rocky slopes and ridges and in gravelly washes of desert scrub zone, 850-1600 m alt.; with species of Larrea Cav., Agave, Viguiera, Yucca L., Fouquieria, Acacia, Hechtia, Mortonia A. Gray, Colubrina, Randia L. ep eheniim oe petra Zucc., Leucophyllum Humb. & Bonpl., Jatropha, Miinesa. and Opun tia FLOWERING AND FRUITING. Flowering February through September; fruiting during same perio ADDITIONAL SPECIMENS EXAMINED. Mexico. COAHUILA: Puerto de Ventanillas along Hwy. 30 N of San Pedro, 7.1 mi SW of turnoff to Delicias, 7. Daniel 659 (Asu, GH, MEXU, MICH, TEX, US); Valle de San Lucas in Sierra del Venado, N of El Mezquite, 7.5 mi W of Hwy. 30, T. Daniel 674 (DUKE, ENCB, MICH); ca. 23 mi NW of Las Delicias in valley N of Sierra de las Delicias, Henrickson 6110 (asu); ca. 32 mi NE of Tlahualilo in NW portion of Sierra de las Delicias, Henrickson 12217 (micnH); ca. 72 mi SW of Cuatro Ciénegas on E side of Sierra de las Delicias, 1.5 mi SW of Las Delicias, Henrickson 12257 (micu); ca. 67 mi SW of Cuatro Ciénegas, 1.5 mi SW of Las Delicias, Henrickson 12466 (ASU); vic. of Aguaje del Pajarito, 2-3 km N of Puerto Colorado, J. Johnston 8675 (F, GH, LL); NE edge of Cerro Bolas, 10.7 km of El Solon on Rte. 40, M. Johnston et al. 8282 (LL); 12.6 km NNE of Las Margaritas in canyon in E face of Sierra de las Margaritas, M. Johnston et al. 9504b (LL), ee (LL), 10362A (LL); Cafion de Fora, ca. 35 km W of Cuatro Ciénegas, M@. Johnston 10910 (LL); S part of Sierra de los Organos, 9.5 km E of Puerto del Gallo, M. Johnston et al. 12137 (LL); ca. 7 mi W of Cuatro Ciénegas, M. Powell et al. 2286 (TEx, wis); Cafidén de la Fora, ca. 1.1 mi by road E of Est. Socorro, Wendt & Lott 1199 (LL); 22 mi W of Cuatro Ciénegas, White 1949 (Ariz, GH, MEXU, MICH). DURANGO: 7 mi SE of Conejos, Correll & I. Johnston 21435 (LL, us). Brandegee (1906) was somewhat doubtful of the generic position of this species when he described it based on Purpus’s collection from southwestern Coahuila. He stated (p. 237), “This plant is not strictly congeneric with the little known genus Ho/ographis, but it seems best to place it there.” Standley did not see any specimens of H/. i/icifolia and did not include a description of it in his treatment of the Acanthaceae of Mexico (1926). Now known rather extensively from the Chihuahuan Desert region, this species is the most unusual in the genus. It is unique among the known species of Holographis in its spinose-toothed, conduplicate leaves and its alternate arrangement of flowers along the spikes. In addition, the orientation of the corolla is somewhat intermediate between vertical and horizontal, and the staminode arises from the corolla up to 2 mm beyond the filaments of the fertile stamens. Although the generic position of this apparently highly spe- cialized species is no longer in doubt, its closest relatives are not obvious. 5. Holographis parayana Miranda, Anales Inst. Biol. Univ. Nac. México 24: Type: Mexico, Chiapas, arriba La Chacona, cerca de la carre- tera a San Fernando, unos 10 km NO de Tuxtla Gutiérrez, 7 Jan. 1951, Miranda 6812 (holotype, MEXu!; isotype, F'!). 150 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Erect shrub to 2.5 m tall. Older stems pubescent or glabrate; younger stems greenish to purplish, terete to subquadrate in cross section, the surface striate, evenly pubescent, with trichomes ascendant-appressed, 0. 1-0.3 mm long. Leaves whorled, ascendant to horizontal; petiole to 4 mm long, pubescent; lamina lance-ovate to ovate, attenuate to acute to truncate at base, acute at apex, 7- 50 by 3.5-15 mm, 1.5 to 3.3 times longer than wide, the margin entire, flat, ciliate, the surfaces pubescent, abaxial surface often more densely so. Inflores- cences axillary spikes to | cm long; axes pubescent with eglandular, ascendant to flexuose trichomes 0.05—-0.3 mm long, occasionally also with glandular tri- chomes to 0.1 mm long; flowers opposite at spike nodes; bracts triangular to lance-subulate, 22.8 by 0.9-1.2 mm, pubescent like inflorescence axis; bract- lets triangular to lance-subulate, shorter than subtending bract, 1.5-2 by 0.8- 1 mm, pubescent like bracts. Calyx 2-3 mm long, the lobes lance-subulate, pubescent like bracts; corolla pinkish (label data), horizontal at anthesis, 9-10 mm long, pubescent on outer surface with mixture of glandular and eglandular trichomes, the tube 3.5-4 mm long, ampliate above, the upper lip 3-3.5 mm long, with lobes 1-1.5 mm long, the lower lip 4.8-6 mm long, with lobes spatulate, 3-3.5 by 1-2 mm; stamens 3 mm long, the filaments 2 mm long, sparsely pubescent, anterior pair often more densely so, the thecae 1.2-1.5 mm long, very sparsely pubescent; staminode | mm long, pubescent at apex; ovary glabrous, the style 5-6 mm long, glabrous, not conspicuously flared at apex, the stigma unequally bilobed, longer lobe to 0.5 mm long. Capsules 12-13.5 mm long, 5 mm in diameter, glabrous; retinacula 3 mm long. Seeds nearly circular in outline, 2.5 mm in diameter, pubescent with dendroid trichomes 0.1-0.3 mm long. DISTRIBUTION AND HABITAT. Known only from vicinity of Tuxtla Gutiérrez, west-central Chiapas (Map 3); low deciduous forests at ca. 900 m alt FLOWERING AND FRUITING. Flowering in January; fruiting in June. ADDITIONAL SPECIMENS EXAMINED. Mexico. CHIAPAS: cerca y arriba Pifiitas, NE Tuxtla , Enriquez 6851 (mexu); arriba Encafiada, NO Tuxtla G., Miranda 5874 (MExv); pera Chacona-San Fernando, NO Tuxtla G., Miranda 6827 (us); Chacona, Miranda 7209 (MEXU). Miranda (1953) noted the relationship between this species and Holographis ehrenbergiana. He claimed that the two could be distinguished by the flat and larger leaves, the smaller flowers, and the strongly curved corolla tube of H. parayana. Comparison of numerous characters in all of the species of Holo- graphis indicates that the affinities of H. parayana are probably with //. hintonii and H. anisophylla, with which it shares numerous features including a rela- tively short, glandular corolla. 6. Holographis anisophylla T. F. Daniel, sp. nov. FIGURE 4. 1 +31 1 A 1 | fay Frutex erectus. Caules juniores pul test O5- 0.1 mm longis. Folia verticillata, in quoque nodo quatuor; laminae lanceolatae vel lanci-ovatae, 9-27 mm longae, 2-8 mm latae, 2.2-4.5-plo longiores quam latiores, marginibus planis ciliatis. Inflorescentia spicata; bracteae lanci-ovatae 1983] DANIEL, HOLOGRAPHIS [Si leaves, < 1.4; c, spike with mature flower, < 3.8; d, dissected flower, x 5. vel ovatae, 2.5—3 mm longae, 1-2 mm latae; bracteolae lanci-ovatae vel ovatae, 2-2.8 mm longae; calyx 4.5-5 mm longus; corolla 8-9 mm longa, utrinque glandulosa; stamina 2.5-3 mm longa filamentis |.3—1.6 mm longis; stamino- dium 0.3 mm longum, apice pubescens; stylus 3.5 mm longus glaber, ovarium basi glabrum, apice pubescens. Capsula ignota. 152 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Erect shrub, Older stems glabrate; younger stems purplish turning white, terete in cross section, the surface striate, pubescent with eglandular, retrorse trichomes 0.05-0.1 mm long, trichomes more or less evenly distributed or in 2 vertical lines. Leaves whorled, horizontal to ascendant, sessile to subsessile; petiole, if present, to | mm long; lamina lanceolate to lance-ovate, long-atten- uate at base, acute to acuminate (to falcate) at apex, 9-27 by 2-8 mm, 2.2 to 4.5 times longer than wide, the 4 leaves at each node unequal in size, with leaves subtending inflorescences considerably reduced, the margin entire, flat, ciliate with ascendant trichomes to 0.5 mm long, the surfaces pubescent. Inflo- rescences axillary spikes to 17 mm long; axes pubescent with mixture of glan- dular and eglandular trichomes, the former 0.05 mm long, the latter erect, 0.05-0.1 mm long; flowers opposite at spike nodes; lower bracts (up to 7 series) sterile, triangular, 1-2 mm long; upper bracts fertile, lance-ovate to ovate, 2.5- 3 by 1-2 mm, mucronate at apex, pubescent like inflorescence axis, margin lanate with trichomes to 0.5 mm long; bractlets lance-ovate to ovate, approx- imately same length as subtending bract or slightly shorter, 2-2.8 by | mm, mucronate at apex. Calyx 4.5—5 mm long, the lobes lanceolate, mucronate at apex, pubescent like bracts; corolla horizontal at anthesis, color unknown, 8- 9 mm long, pubescent on outer surface with mixture of glandular and eglandular trichomes, the tube 4-5 mm long, ampliate above, the upper lip 1.5-2 mm long, with lobes | mm long, the lower lip 4 mm long, with lobes obovate, 2.5 by 2.5-2.8 mm; stamens 2.5-3 mm long, the filaments 1.3-1.6 mm long, pubescent with flexuose trichomes to 0.8 mm long, the thecae 1.8-2 mm long, connivent at pubescent apex; staminode 0.3 mm long, pubescent at apex; ovary glabrous below, pubescent with eglandular trichomes at apex, the style 3.5 mm long, glabrous, the stigma lobes equal, 0.4 mm long. Capsules not seen. pe. Mexico, Colima, Desviacién cerca Rio Salado, 4 Dec. 1959, Miranda oe (holotype, MExv!). Holographis anisophylla is known only from the type, a flowering collection from east-central Colima (Map 1), which was collected in tropical deciduous forest (McVaugh, pers. comm.). It is unique among the known species of /olo- graphis in having glands on the inner, as well as the outer, surface of the corolla. It appears to be most closely related to H. pallida, a species that occurs approx- imately 800 km to the northwest of the type locality. Miranda’s collection was, in fact, originally identified as the latter species. These two species resemble each other in the form of the cauline trichomes and in the shape of the leaves and bracts; in addition, the whorled leaves of H. pallida are frequently ani- sophyllous. The species can be distinguished from each other by the characters in the following couplet: 1. Bracts 3-5 mm long; bractlets lance-subulate, 3-4 iat ae calyx 5-7 ee long; corolla 10-12 mm long, eglandular; stamens 3.5-4 mm long. ........ H. pallida. 1. Bracts 2.5-3 mm long; bractlets lance-ovate to ovate, os 2.8 mm long; calyx 4.5-5 mm long; corolla 8-9 mm long, glandular; stamens 2.5-3 mm long. ............. 1983] DANIEL, HOLOGRAPHIS 153 7. Holographis hintonii (Leonard) T. F. Daniel, comb. nov. Berginia hintonii Leonard, Kew Bull. 1938: 64. 1938. Type: Mexico, Guerrero, District of Coyuca, between Coyuca de Catlan and Chamerito, 26 April 1934, Hinton et al. 5956 (holotype, K!). Erect shrub. Older stems covered with conspicuous lenticels, glabrous; youn- ger stems subquadrate in cross section, the surface striate-fluted, sparsely pubes- cent in 2 vertical lines, with trichomes inconspicuous, eglandular, retrorse, 0.05-0.1 mm long. Leaves whorled; petiole up to 8 mm (or more?) long, pubescent with erect to flexuose trichomes to 0.2 mm long; lamina apparently ovate-elliptic (see discussion), attenuate at base, to about 45 (or more?) by 20 (or more?) mm, the margin entire, flat, ciliate, the surfaces sparsely pubescent with flexuose trichomes to 0.8 mm long, the veins on the lower surface densely pubescent. Inflorescences congested axillary spikes to 10 mm (or more?) long; axes obscured by dense mats of crooked, interwoven trichomes to 0.8 mm long (villous); flowers opposite at spike nodes; lower bracts (up to 4 series) sterile, triangular, smaller than upper bracts; upper bracts fertile, broadly ovate to orbicular, 1.5-2 by 1.2-1.5 mm, mucronate at apex, pubescent with mixture of erect glandular and eglandular trichomes to 0.2 mm long, margin villous- ciliate; bractlets lanceolate, approximately same length as subtending bract, 1.5-2 by 0.7-0.9 mm, mucronate at apex, pubescent like bracts. Calyx 4—-4.5 mm long, the lobes lanceolate, pubescent like bracts (but glands often more numerous and conspicuous); corolla color unknown, orientation unknown, 8- 9.5 mm long, pubescent on outer surface with mixture of eglandular and glan- dular trichomes, the tube 2-2.5 mm long, the upper lip 3-4 mm long, with lobes 2-3 mm long, the lower lip 6-7 mm long, with lobes orbicular, 3-3.5 mm in diameter; stamens 3—3.5 mm long, the filaments 1-1.5 mm long, densely pubescent with flexuose trichomes to 0.8 mm long, the thecae 2 mm long, sparsely pubescent; staminode 0.4 mm long, glabrous; style 4 mm long, pubes- cent, the stigma lobes 0.5 mm long. Capsules 10-11 mm long, 3-4 mm in diameter, pubescent, with trichomes eglandular, 0.1-0.2 mm long. Seeds 4 by .5 mm, pubescent with dendroid trichomes to 0.3 mm long. This species is presently known only from the type, which was collected in flower and fruit in late April in the semiarid region of northwestern Guerrero (Map 3). Although this specimen of Holographis hintonii is leafless, the size of the leaf scars and a fragment of the basal portion ofa leaf in a packet attached to the specimen indicate relatively large, petiolate, ovate-elliptic leaves, prob- ably similar in form to those found in H. tamaulipica. When he described this species (as Berginia hintonii), Leonard (1938) noted that it was distinctive in Berginia in having “‘short bract-covered peduncles and short tomentose spikes.” In Holographis its closest relatives appear to be H. parayanaand H. anisophylla; all three have entire, flat leaves and relatively small (8-10 mm long), glandular corollas. 154 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 8. Holographis pallida Leonard & Gentry in Gentry, Brittonia 6: 324. fig. 5. 1948. Type: Mexico, Sinaloa, Cerro Llano Redondo, W of Caymanero, 25 April 1944, H. Gentry 7022 (holotype, MICH!; isotypes, DEs!, F!, GH!, Ny!, us!). Erect shrub to 1.3 m tall. Older stems pubescent or glabrate; younger stems gray-green, terete in cross section, the surface smooth to striate, evenly and densely pubescent making the stems appear pallid, the trichomes eglandular, retrorse, 0.05—0.2 mm long. Leaves whorled, ascendant or decurved, subsessile to petiolate; petiole to 5 mm long, pubescent like younger stems; lamina lan- ceolate to ovate to subelliptic, long-attenuate at base, acute to acuminate (to falcate) at apex, 8-50 by (2-)4-21 mm, 2.5 to 4 times longer than wide, the 4 leaves at a node often unequal in size, the margin entire, flat, ciliate, the surfaces sparsely pubescent, with trichomes often restricted to major veins, abaxial surface somewhat punctate. Inflorescences densely bracted axillary spikes to 3 cm long; axes hirsute, with trichomes erect to recurved, 0.2—0.8 mm long (axes often with inconspicuous, capitate glands to 0.3 mm long as well); flowers opposite at spike nodes; lower bracts (up to 4 series) mostly sterile, triangular, 1-2 mm long; upper bracts fertile, lanceolate to ovate, 3-5 by I-1.8 mm, mucronate at apex, pubescent like inflorescence axis, with trichomes particu- larly evident along margin; bractlets lance-subulate, approximately same length as subtending bract, 3-4 by 0.8—1 mm, mucronate at apex. Calyx 5—7 mm long, the lobes lance-subulate, mucronate at apex, pubescent like bracts; corolla “white to pale lavender” (label data), horizontal at anthesis, 10-12 mm long, pubescent on outer surface with eglandular trichomes, the tube 5-6 mm long, ampliate above, the upper lip 2-3 mm long, with lobes 0.8-1.5 mm long, the lower lip 5-6 mm long, with lobes spatulate, 3-5 by 2-4 mm; stamens 3.5-4 mm long, the filaments 1.5—2 mm long, pubescent with flexuose trichomes to 0.4 mm long, the thecae |.5—2 mm long, pubescent along dorsal side, connivent around style; staminode reduced to minute, glabrous projection 0.1 mm long; style 5.3-6 mm long, glabrous or sparsely pubescent, flared at apex, the stigma equally bilobed, with lobes to 0.5 mm long. Capsules 7-9 mm long, 3 mm in diameter, pubescent on outer surface with eglandular trichomes 0.1-0.2 mm long; retinacula 1.5 mm long. Seeds oval, 2.2 by 1.9 mm, densely pubescent with dendroid trichomes to 0.2 mm long. DISTRIBUTION AND HABITAT. Southern Sonora and central Sinaloa (Map 1); in arroyos and on rocky slopes in thorn forest, ca. 50-500 m alt FLOWERING AND FRUITING. Flowering in February and April, when leaves appear to begin flushing; fruiting in April. ecuees SPECIMENS EXAMINED, Mexico. SoNoRA: San Bernardo, H. Gentry 1350 (a, , F, MO, UC), 3675 (Ariz, F); Sierra Bojihuacame, SE of Cd. Obregon, H. Gentry 14492 ne Leonard and Gentry (in Gentry, 1948) noted the apparent uniqueness among American Acanthaceae of the monothecous stamens with connivent anthers in this species and suggested that the plant may be worthy of generic status. 1983] DANIEL, HOLOGRAPHIS 155 In the present study this character was found to be almost universal among species of Holographis. Leonard and Gentry further noted the similarity of this species to members of Berginia, indeed, Gentry’s label on the type bears a preliminary name in this genus. Finally, they concluded that Holographis and Berginia were probably congeneric. The closest relative of Holographis pallida appears to be H. anisophylla. The similarities between them are discussed under the latter species. 9. Holographis pueblensis T. F. Daniel, sp. nov. FIGURE 5. Suffrutex vel frutex erectus usque ad 5 dm altus. Caules juniores pubescentes trichomatibus eglandulosis, 0.2-0.8 mm longis. Folia supera verticillata, in quoque nodo quatuor; laminae ovatae vel ellipticae, 5-24 mm longae, 3-15 mm latae, 1.5-2-plo longiores quam latiores, marginibus revolutis ciliatis. Inflorescentia spicata; bracteae anguste 1_.1ceolatae vel lanci-subulatae, 6.5-11 mm longae, 1.5-2 mm latae; bracteolae lanci-subulatae, 5.5-9 mm longae. Calyx 7-10.5 mm longus; corolla aurea, 15-18 mm longa, extus pubescens; stamina 9-11 mm longa filamentis 8—8.5 mm longis apicibus glabris basibus pubescentibus; staminodium 0.4-0.5 mm longum pubescens; stylus 12-16 mm longus glaber, ovarium glabrum. Erect subshrub to shrub to 5 dm tall arising from stout, woody base to 5 mm in diameter. Older stems pubescent or glabrate; younger stems terete in cross section, the surface smooth to striate, evenly (often densely) pubescent, with trichomes eglandular, ascendant-appressed, 0.2-0.8 mm long. Leaves whorled (lower leaves sometimes opposite or subopposite), ascendant to hor- izontal, subsessile to petiolate; petiole to 8 mm long, pubescent with flexuose to ascendant trichomes to 1 mm long; lamina ovate to elliptic, attenuate to acute to truncate at base, acute to rounded at apex, 5-24 by 3-15 mm, 1.5 to 2 times longer than wide, the margin entire, revolute, ciliate, the surfaces pubescent, the abaxial surface often punctate. Inflorescences axillary, densely bracted spikes to 4.5 cm long; axes pubescent with erect to ascendant trichomes 0.2-0.5 mm long; flowers opposite at spike nodes; bracts narrowly lanceolate to lance-subulate, 6.5-11 by 1.5-2 mm, tapering to point but not mucronate at apex, pubescent like inflorescence axis; bractlets lance-subulate, approxi- mately same length as .ubtending bract, 5.5-9 by 1-1.8 mm, pubescent like bracts. Calyx 7-10.5 rm long, the lobes lance-subulate, outer surface sparsely pubescent (often glabrous near base), margin lanate; corolla yellow, vertical at anthesis, 15-18 mm long, pubescent on outer surface with eglandular tri- chomes, the tube 7.5-9 mm long, ampliate above, the upper lip 7-8 mm long, with lobes 0.3-0.5 mm long, the lower lip 7.5-9 mm long, with lobes spatulate, 4.5-6.5 by 2.2-4 mm; stamens 9-11 mm long, the filaments 8—8.5 mm long, glabrous above, pubescent at base, the thecae 2-3 mm long, pubescent; stam- inode 0.4-0.5 mm long, pubescent; ovary glabrous, the style 12-16 mm long, glabrous, not flared at apex, the stigma lobes inconspicuous. Capsules not seen. Type. Mexico, Puebla, vicinity of San Luis Tultitlanapa, near Oaxaca, July 1908, Purpus 3346 (holotype, uc!; isotypes, F!, GH!, Mo!, Ny!, Us!). 156 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Figure 5. Holographis pueblensis: a, habit (Purpus s.n., June 1907), X 0.6; b, node with leaves (Purpus 3346), X 2.5; c, young spike with flowers (Purpus 3346), X 2.2: d, gynoecium (Purpus 3346), x 3. 1983] DANIEL, HOLOGRAPHIS ibe DISTRIBUTION AND HABITAT. Known only from two Purpus collections from semiarid regions of southeastern Puebla (Map 3) FLOWERING AND FRUITING. Flowering June and July; time of fruiting unknown. ADDITIONAL SPECIMEN EXAMINED. Mexico. PUEBLA: Tehuacan, Purpus s.n., June 1907 (uc) This species appears to be most closely related to Holographis ehrenbergiana, which occurs primarily north of the range of H. pueblensis (see discussion under H. ehrenbergiana). 10. Holographis ehrenbergiana Nees in DC. Prodr. 11: 728. 1847. Type: Mex- ico, “supra la Haciende de Guadela [?],”’ Ehrenberg 1224 (holotype, B, destroyed). FIGURE 6 Erect, often rounded subshrub to shrub to 1.3 m tall, arising from stout, often tortuous woody base to 12 mm in diameter. Older stems pubescent or glabrate; younger stems pale green, terete to subquadrate 1n cross section, the surface smooth to striate, evenly (often densely) pubescent, with trichomes eglandular, erect to ascendant, 0.1-0.7 mm long. Leaves whorled, ascendant to horizontal, subsessile to petiolate; petiole to 8 mm long, pubescent; lamina lanceolate to obovate, attenuate to acute at base, acute to rounded at apex, 5— 35 by 2-16 mm, 1.9 to 3.5 times longer than wide, the margin entire, revolute, ciliate, the surfaces pubescent, abaxial surface often more densely so (sometimes becoming lanate). Inflorescences axillary spikes to 2 cm long, usually reduced to 2 flowers; axes pubescent like younger stems; bracts triangular to subulate, 1.5-5 by 0.5-1 mm, emucronate, sparsely to densely pubescent with erect to flexuose trichomes to | mm long; bractlets subulate, approximately same length as subtending bract or slightly longer, (1.5-)2.5—5 by 0.5 mm, pubescent like bracts. Calyx 3.5-—6 mm long, the lobes lance-subulate, pubescent like bracts; corolla yellow, vertical at anthesis, 12-18 mm long, pubescent on outer surface with eglandular trichomes, the tube 6-10 mm long, ampliate above, the upper lip (6-)7.5—9 mm long, with lobes 0.3-1 mm long, the lower lip (5.5—)8-10 mm long, with lobes spatulate, 4.5-6 by 2—4.5 mm; stamens 9-11 mm long, the filaments 8-10 mm long, glabrous or pubescent at base, the thecae 1.5-2.3 mm long, densely pubescent; staminode 0.5—1.3 mm long, pubescent at base; style 12-14 mm long, sparsely pubescent (some individual styles glabrous, but at least some styles sparsely pubescent on each plant), not flared at apex, the stigma lobes inconspicuous. Capsules 10-15 mm long, 4—4.5 mm in diameter, glabrous; retinacula 2-3 mm long. Seeds suboval, 3.5-4.8 by 2.5-3.5 mm, glabrous DISTRIBUTION AND HABITAT. Central Nuevo Leon and Tamaulipas south through San Luis Potosi and Hidalgo, apparently to southeastern Puebla (MAP 1); rocky limestone slopes at 150-2000 m alt.; in arid and semiarid associations with species of Mimosa, Flourensia DC., Acacia, Colubrina, Agave, Neopringlea S. Watson, Cordia L., Bonetiella Rzedowski, Fouquieria, and Yucca. 158 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Figure 6. Holographis ehrenbergiana: a, habit (Purpus 4921), X 0.6; b, node with leaves and flower (Moore & Wood 4406), X 2; c, capsule (Purpus 4921a), X 3.8; d, seed (Purpus 4921a), X 6.8. 1983] DANIEL, HOLOGRAPHIS iso FLOWERING AND FRUITING. Flowering May through October; fruiting a month or two after flowering. REPRESENTATIVE SPECIMENS. Mexico. HipaLco: km 147, SE of Ixmiquilpan, Lundell & Lundell 12342 (LL, Micu); Barranca de Toliman, above mines on road from Zimapan Ixmiquilpan, Purpus 1409 (F, GH, Mo, Ny, Uc); 10 km NW de Zimapan, Gonzdélez Quintero 3238 (ENCB, MICH); 5 km NW de Actopan, Soe Quintero 3581 (Ds, ENCB, MICH, MSC). Nuevo Leon: Cafién de Huasteca, near Monterrey, near upper spring of Rancho de Ed. Aguirre-Pequefio, Kruckeberg 4857 (MIcH). Be Tlacuilotepec, Purpus 3936 (uc). SAN Luis Potosi: 20 km E of San Francisco on San Luis Potosi-Rio Verde Hwy., 22°00'N, 100°37'W, M. Johnston et al. 8182 (LL); 10 mi W of Rio Verde on road to S.L.P. at microondas station, M. Johnston et al. 11189 (11); Rio sie Palmer 4 (F, GH, MO, NY, UC, US); Minas de mare Rafael, he 4921 (F, MO, UC, US), 49214 (GH, MO, co 5225 (DES, F, GH, MEXU, MO, NY, UC); E de Niifiez, km 48 carretera . L.P.-Antiguo Morelo, Rzedowski 5605 (ENCB); ca. 15 km NE de Guadalcazar, Rzedowski 6031 (ENCB, MEXU, MICH); 40 km S de Matehuala, 6 km E del km 575 de la carretera México—Piedras Negras, Rzedowski 8254 (ENcB). TAMAULIPAS: Sierra de San Carlos, vic. of San Miguel, Bartlett 10623 (F, Micu); Cerro El Platero above Rancho EI Platero, 0.5 mi S of Ejido El Higuerén (Mpio. Aldama) in Sierra de Tamaulipas, 7. Daniel 268 (mic); 1 km S of Carabanchel, Gilbert 74 (TEx); 3 mi E of the San Fernando-Santander Jiménez Hwy. on road to Loreto, M. Johnston & Crutchfield 5593 (micH, TEX); by Mex. 85, 10.6 mi N of San José and 16 mi N of Ejido, Kral 27326 (mo); Buena Vista Hda., Wooton s.n., 14 June 1919 (us). Locality unknown: Coulter 121] (GH) Purpus’s specimen from Tlacuilotepec, Puebla, was apparently collected near where Holographis pueblensis was growing. Sousa Sanchez (1969) noted that Tlacuilotepec 1s probably in the region of San Luis Tultitlanapa. It is from this latter locality that the type of H. pueblensis was collected. If this locality infor- mation is correct, then the range of H. ehrenbergiana overlaps those of both H. tamaulipica and H. pueblensis. Holographis ehrenbergiana shows considerable variation in the amount of pubescence on the abaxial leaf surface. At one extreme (e.g., M. Johnston & Crutchfield 5593) the lower leaf surface is evenly but relatively sparsely pubes- cent, while in some specimens (e.g., Moore & Wood 4406) the trichomes are so dense that the surface is not visible. Holographis ehrenbergiana appears to be most closely related to H. pueb- lensis, which it superficially resembles. Characters of the inflorescence, bracts, bractlets, and calyx readily distinguish these species, however. ACKNOWLEDGMENTS Most of this study was completed while I was a lecturer in botany at the University of Michigan. I am grateful for the facilities provided by the Uni- versity of Michigan Herbarium and for loans of specimens received from the following herbaria: A, ARIZ, ASU, CAS, DES, DS, ENCB, F, GH, K, LL, MEXU, MICH, MO, NY, SD, TEX, UC, US, and wis. I wish to thank Dr. W. R. Anderson for assistance with the Latin descriptions, Dr. James Henrickson for sending me collections of Holographis, Annetta Carter for information on H. virgata, Skye Baker for preparing the illustrations, and Drs. D. C. Wasshausen and Rogers 160 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 McVaugh for useful comments on the manuscript. Field support was provided by NSF grant DEB 78-09321. LITERATURE CITED BENTHAM, G., & J. D. Hooker. 1876. Genera plantarum. Vol. 2. viii + 747 pp. Reeve and Co., London. BRANDEGEE, T. S. 1889. A collection of plants from Baja California, 1889. Proc. Calif. Acad. Sci. If. 2: 117-216. _ 1906. New species of Mexican plants collected by Dr. C. A. Purpus. Zoe 5: 231-241. BREMEKAMp, C. E. B. 1965. Delimitation and subdivision of the Acanthaceae. Bull. Bot. Surv. India 7: 21-30. Gentry, H. S. 1948. Additions to the flora of Sinaloa and Nuevo Leon. Brittonia 6: 09-331. Gray, A. 1885. New genera of Arizona, California, and their Mexican borders, and two additional species of Asclepiadaceae. Proc. Amer. Acad. Arts 20: 290-296. LEONARD, E. C. 1938. Contributions to the flora of tropical America: XXXIV. Plantae Hintonianae: VI. Kew Bull. 1938: 59-73. : A new genus of Acanthaceae from Mexico. Wrightia 2: 1-3. LINDAU, G. 1895. Acanthaceae. In: A. ENGLER & K. PRANTL, eds., Nat. Pflanzenfam. IV. 3b: 274-354. Miranpa, F, 1953. Plantas nuevas o notables de la flora de Chiapas. Anales Inst. Biol. Univ. Nac. México 24: 69- NEES AB ESENBECK, C. G. 1847. Acanthaceae. In: A. P. pE CANDOLLE, ed., Prodr. 11: 46-519; Suppl., pp. 720-7 Sousa SANCHEZ, M. 1969. Las colecciones Sy a de C. A. Purpus en México periodo 1898- eee Univ. Calif. Publ. Bot. 51: 1-36. saan ,C. P. 1976. Leaf ae and systematics of Acanthaceae and ated familics 234 pp. Unpubl. Ph.D. dissertation, University of South Florida, Sees STANDLEY, P. C. 1926. Trees ee ae of Mexico (Bignoniaceae—Asteraceae). Contr. .S. Natl. Herb. 23: 1313-1 Vasey, G., & J. N. Rose. 1890. of plants collected by Edward Palmer in lower California and western Mexico in 1890. Contr. U. S. Natl. Herb. 1: 63-90. WASSHAUSEN, D. C. 1975. The genus Aphelandra eae Smithsonian Contr. Bot. 18: 1-157. DEPARTMENT OF BOTANY AND MICROBIOLOGY ARIZONA STATE UNIVERSITY TEMPE, ARIZONA 85287 1983] CARLQUIST, BELLIOLUM 161 WOOD ANATOMY OF BELLIOLUM (WINTERACEAE) AND A NOTE ON FLOWERING SHERWIN CARLQUIST Tue GENUS Belliolum Van Tieghem occurs in the Solomon Islands and in New Caledonia (Smith, 1943). Of the Solomon Islands species, three were described by Smith (1942); a fourth, B. haplopus (Burtt) A. C. Sm., was trans- ferred from Bubbia Van Tieghem. Although four New Caledonian species of Belliolum were recognized by Van Tieghem (1900), all except B. pancheri (Baillon) Van Tieghem are known from inadequate material or are otherwise of uncertain status. This situation will doubtless be remedied by the studies of W. Vink, who has kindly determined my New Caledonian material as B. pancheri. The present study gives comparative data on the wood of Belliolum, which has previously been known only from an illustration in Bailey (1944); his article contained a description of the wood of Winteraceae as a whole but no char- acterization of genera and species. The paucity of material available at that time rendered comparative studies within the family impossible. However, more abundant material now at hand renders comparative studies of winter- aceous woods feasible. Patel (1974) produced an admirable monograph of the woods of the genus Pseudowintera Dandy, and a study of woods of Zygogynum Baillon (Carlquist, 1981) has been published. Comparative study of woods yields not merely taxonomic dividends, but a better understanding of wood in relation to ecology as well. A few illustrations of flowers of Be/liolum pancheri and brief descriptions of details of anthesis are appended to supplement the account of pollination of the Winteraceae given by Thien (1980). MATERIALS AND METHODS The wood samples of Belliolum pancheri analyzed were collected in mid- elevation forests of the Plateau de Dogny, New Caledonia, where the species grows as an understory tree on humid, shady slopes. Wood samples were prepared by removing the bark to facilitate drying. Samples could not be completely dried prior to shipping, so they were enclosed in plastic bags in which paraformaldehyde sufficient to retard fungal growth had been added. Upon receipt in Claremont, these wood samples were extracted, washed, and air dried. This method prevented the mold that would otherwise have grown. The wood sample of B. pancheri was 10 cm in diameter. Wood samples of the Solomon Islands species were provided through the © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 161-169. January, 1983. 162 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 aliudu lily dolly ll way yyy | 1 } mi ais A | Nl aT . hy . | pe! ie Me Yi i \ oA \ \ \ ‘“ ve, ks i \ i — peat \ ‘ eit, = : ) he | | f ees | ; A) i ety | | Hees f 2 “a | - =| ji a == 0.9) 4 , : i : : “ ! " a my 3 Woh 4 =) Pi RB i, Pe i Ficures 1-5. Belliolum haplopus (MADw-22694), wood sections: 1, transverse sec- tion, illustrating long and short tangential parenchyma bands; 2, tangential section, showing thin-walled rays; 3, transverse section, tracheids wide and thin walled; 4, radial section, scalariform and transitional pitting on overlap areas of tracheid; 5, radial section, borders of pits on ray-cell walls. Scale: 1, 2, finest divisions = 10 wm; 3-5, divisions = LO um. 1983] CARLQUIST, BELLIOLUM 163 TABLE 1. Features of the wood of Belliolum. TRACHEID TRACHEID WALL TRACHEID DIAMETER THICKNESS LENGTH Ray SPECIES AND COLLECTION (um)* (um) (um) WIDTHT B. gracile, MADw-29483 60 1.4 6290 6.2 B. haplopus, MADw-22694 56 4.8 4902 6.7 B. haplopus, MADw-29285 54 5.4 4570 6.3 B. kajewskii, MADw-29294 50 5.0 3958 6.2 B. pancheri, Carlquist 15564 42 Wak 4260 4.4 *Mean diameter at widest point. +Number of cells of multiseriate rays at widest point. kindness of the Forest Products Laboratory of the U.S. Forest Service (MADw). The provenance of these samples is Bougainville Island, as follows: Belliolum gracile A. C. Sm., MADw-29483, R. Schodde & L. Craven 3718 (CANB), pri- mary montane rain forest, 2700 ft (tree 15 m, DBH ca. 20 cm), S foothill slopes of Lake Loloru crater, ca. 17 mi N of Buin; B. haplopus (Burtt) A. C. Sm., MADw-22694, no further locality, and MADw-29285, R. Schodde 261 (CANB), primary rain forest, ca. 2300 ft (small bushy tree ca. 7 m tall, DBH 6 cm), lower S slopes of Lake Loloru crater, ca. 14 mi N of Buin; B. kajewskii A. C. Sm., MADw-29294, L. A. Craven & R. Schodde 307 (CANB), primary montane rain forest with a number of small watercourses, ca. 2200 ft (tree 9 m tall, DBH 12 cm), lower S slopes of Lake Loloru crater, ca. 15 mi N of Buin. The identity of the specimen supplied as B. gracile is questionable because that species is not reported by Smith (1942) to occur on Bougainville Island, the provenance of the sample. The wood sample of Bel/liolum pancheri was boiled and softened with eth- ylene diamine before being sectioned on a sliding microtome, according to the technique of Kukachka (1977). The wood of B. pancheri is moderately hard for a winteraceous wood. Woods of the Solomon Islands species, however, proved too soft for sectioning on a sliding microtome after being boiled (but without any treatment with ethylene diamine). Consequently, an alternative method was used for these species. They were treated with ethylene diamine according to Kukachka’s method—but for one month. The woods were then washed, put through a tertiary butyl alcohol series (Johansen, 1940), embedded in paraffin, and sectioned at 15 mm. This novel use of ethylene diamine as a softening agent prior to paraffin sectioning yielded excellent results (see FIGURES 1-5) and is strongly recommended for woods that are extremely soft. All wood sections of Belliolum were stained with safranin, while macerations were pre- pared with Jeffrey’s fluid and also stained with safranin. ANATOMICAL DESCRIPTIONS As shown in TABLE |, quantitative data for the Bellio/um collections were obtained on a limited number of wood features. The figures show that the 164 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 l | | jet lene al i | | fh oe | | wu I yy yu yyy a Basests sesaes eas iesieas 7 IMT owe. . , a2, Ss , le ee ae ’ — 3 \ : 74) 9 4 FIGURES 6— 9. Belliolum pancheri (Car/quist 15564, RSA), wood sections: 6, transverse of tracheid walls and bands of axial parenchyma; 9, radial section, a biseriate pits of overlap areas of tracheids. Scale: 6, 7, finest divisions= = 10 um; 8, 9, divisions = 10 um 1983] CARLQUIST, BELLIOLUM 165 tracheids of B. gracile are thinner walled than those of other species. Compared to B. pancheri (FiGure 8), the Solomon Islands species as a whole (e.g., B. haplopus, FiGuRE 3) have much wider tracheids with thinner walls. Overlap areas (end walls) of tracheids in some Winteraceae show patterns markedly different from those on lateral walls. Pitting is generally denser on end walls than on lateral walls, and in Belliolum can frequently be biseriate (as shown for B. pancheri in FiGure 9) or triseriate. In addition, overlap areas of tracheids in all Belliolum species except B. kajewskii (which has only the biseriate and triseriate end-wall pitting mentioned above) also bear scalariform, scalariform transitional to opposite, and opposite pitting. The scalariform and scalariform transitional to opposite end-wall pitting types are shown here for B. haplopus (FiGure 4). Pits on lateral walls of tracheids of Belliolum are circular, about 7 um in diameter, and with elliptic apertures like those of the pits shown in FiGuRE 9. Patel (1974) has reported similar pits on Pseudowintera tracheids. The helical thickenings and trabeculae reported by Patel (1974) for tracheids of Pseudowintera are not present in Be/liolum. In groups with characteristics that reflect mesic environmental conditions, helical thickenings may relate to adaptation to areas where frost occurs. This tendency can be traced in such families as Illiciaceae, where such thickenings occur in vessel elements. Axial parenchyma expressions vary within Belliolum. In B. pancheri short bands, usually a single cell thick, can be seen in addition to isolated diffuse cells (FiGuRE 8). In B. haplopus bands are one to three cells thick, but diffuse cells are also present (FiGurRE 3). In B. gracile and B. kajewskii, however, only diffuse cells, occasionally grouped two or three together, were observed. Although the range of axial parenchyma types is somewhat greater in Zygogynum (axial parenchyma Is scarce in some collections of Zygogynum), the two genera are very similar. Pseudowintera also has the same range of axial parenchyma types as is found in Belliolum (Patel, 1974). Axial parenchyma strands are long in Belliolum, commonly ranging from 17 to 24 cells. Rays vary relatively little in width in Be//iolum (TABLE 1) but are appreciably narrower in B. pancheri (FIGURE 7) than in the Solomon Islands species such as B. haplopus (FiGureE 2). This cannot be related to stem size, for the stems of B. pancheri are comparable in size to those of B. gracile and B. haplopus. As judged from Patel’s (1974) data, the rays are much narrower in Belliolum than in Pseudowintera, but are about the same as those in Zygogynum. Ray cells have thicker walls in Belliolum pancheri than in the Solomon Islands species. In no species of Belliolum did rays contain the ethereal oil idioblasts or sclereids found in some species of Zygogvnum (Carlquist, 1981). Ray cells, asin Zygogynumand Pseudowintera (Patel, 1974), often have borders (FicurE 5). In none of the Belliolum species, however, are borders present in all ray cells. Resinlike deposits (FIGURE 5) could be observed in ray cells in all Belliolum species but are much more conspicuous in B. pancheri (FIGURES 6— 8). As in other species of Winteraceae, ray histology corresponds to Kribs’s (1935) Heterogeneous Type I. Growth rings are absent in Be/liolum but present in Pseudowintera (Patel, 1974) and other species of Winteraceae from markedly temperate climates. 166 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 CONCLUSIONS The species differ in tracheid length, probably in relation to size of plant. spe and Faull (1934) showed that tracheid length increases steadily with age n Sequoia, a pattern confirmed for other gymnosperms and for vesselless ieee: by Carlquist (1975). Tree (or shrub) size and trunk diameter appear to correlate with tracheid length. Tracheid diameter also appears to correlate with tracheid length in vesselless woods (Bannan, 1965). Compared to those of the other species studied here, the tracheids of Belliolum gracile are the longest (and widest); they were obtained from what is probably the largest tree (DBH = 20 cm) of the group. The longest tracheids (5580 ym) hitherto reported for vesselless angiosperms were from a sample of Bubbia semecarpoides that was 12 cm in diameter (Carlquist, 1975). Although trunk diameter and tracheid length are generally correlated (as they are in gymno- sperms), the relation is not at all precise, as proved true in Zygogynum (Carl- quist, 1981). Tracheids of Agathis are slightly longer than trunk diameter and plant height might suggest, but not greatly so (Carlquist, 1975). Siiss (1980), who attempts to use tracheid length as a phylogenetic indicator in plants with vesselless wood, seems unaware of correlations between plant height and size and tracheid length (and diameter). The differentiation of overlap areas in the tracheids of Winteraceae was noted by Bailey (1944) and has been illustrated for Bubbia semecarpoides and Zygo- gynum bicolor (as “Z. cf. pomiferum’’) (Carlquist, 1975), as well as for three other species of Zygogynum (Carlquist, 1981). As in Belliolum, scalariform end-wall pitting is clearly present in some collections of Zygogynum but absent in others. In Pseudowintera it is present only on enlarged traumatic tracheids (Patel, 1974). The presence of these scalariform pits on end walls in wintera- ceous tracheids has been claimed to be a kind of prevessel specialization of the tracheid for maximizing conductive ability (Carlquist, 1975, 1981). The pres- ence on the end wall of scalariform pits rather than merely crowded circular pits may also be seen as a remnant of the capability—present in primitive vesselless woods—to produce scalariform pitting. The range of axial parenchyma of Belliolum ( d diffuse-in-aggregates, plus bands up to three cells wide in some collections) resembles that in Zygo- ynum, as does the typical ray width. The ray cells, however, do not show the division of labor (ethereal oil cells, sclereids) in Belliolum that they do in Zygogynum. The wood of Belliolum is less similar to that of Pseudowintera, as described by Patel, and most closely resembles those of Bubbia and Zygo- gynum, two genera phytogeographically closer to Belliolum than Pseudowinte- ra. Some distinctive features of Pseudowintera (growth rings, helices in tra- cheids) are doubtless related to the strongly temperate climate of New Zealand, and thus the woods of Bubbia, Belliolum, and Zygogynum (which lack these features) reflect a mild and relatively seasonless climate. On the basis of a limited number of samples, the Solomon Islands species of Belliolum differ from B. pancheri from New Caledonia. Belliolum pancheri has thicker-walled, narrower tracheids and narrower rays with thicker-walled cells than do the Solomon Islands species. Smith (1943) mentioned possible 1983] CARLQUIST, BELLIOLUM 167 section, X 7.5, ovules in carpel at right displaced by sectioning: 14, dissection of flower at late anthesis, X 6, showing anthers open. sectional differences between the Solomon Islands and New Caledonian groups of species on the basis of carpel morphology; xylarly characters might be taken into account in this connection. Bongers (1973) mentioned that stomata are much more deeply sunken in leaves of B. crassifolium (Baillon) Van Tieghem, 168 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 a New Caledonian species, than they are in leaves of B. haplopus, from the Solomon Islands. A NOTE ON FLOWERING Thien (1980) has described the stages of anthesis in various Winteraceae, including Belliolum. Figures 10-14, eee of B. pancheri taken in New Caledonia, illustrate additional aspect The opening mechanism of the ae of Belliolum is hygrochastic, or depen- dent upon swelling of the flower rather than drying for opening to occur. Dissection of a bud (FiGurE 12) reveals petals and stamens to be relatively thin and smooth before anthesis. By the time when anthesis occurs, however, the petals and stamens have enlarged appreciably and have developed ridges and warts (Ficures 10, 11, 14). The deltoid stamen tips may thus play a role in the opening of the flower. During early anthesis the deltoid tips of the innermost stamens often appear to fit between the stigmas, so that the stigmas are exposed with certainty. The stamens soon reflex away from the carpels, however. The anther sacs are more nearly lateral than adaxial, but may be described as extrorse-latrorse (FiGuREs 12, 14). Damage to anther tips by a chewing insect was noted (FiGure 14). Sectioning the carpels (FiGuRE 13) revealed that ovules are clearly positioned on the adaxial side of the locule, although the stigma extends down very little from the apex onto the adaxial face of the carpel. The stigmas would not be exposed at anthesis, because of carpel number and carpel crowding, if they were on adaxial faces; thus, one should expect the stigmas to be mostly apical in orientation in this genus. In the Solomon Islands species of Be/liolum (Smith, 1942, 1943), the stigmas are even more clearly apical in nature than in B. pancheri, although the ovule position may be the same (Bailey & Nast, 1943). LITERATURE CITED Baitey, I. W. 1944. The comparative morphology of the Winteraceae. HI. Wood. Jour. Arnold Arb. 25: 97-103. & A. F. FAuLL. 1934. The cambium and its derivative tissues. IX. Structural variability in the redwood, Sequoia sempervirens, and its significance in the iden- tification of fossil woods. Jour. Arnold Arb. 15: 233-254. & . Nast. 1943. The comparative morphology of the Winteraceae. II. Carpels. Jour. Arnold Arb. 24: 472-481. BANNAN, M. W. 1965. The length, tangential diameter and length/width ratio of conifer tracheids. Canad. Jour. Bot. 43: 967-984 Boncers, J. M. 1973. Epidermal leaf characters of the Winteraceae. Blumea 21: 381- 411. CarLQuist, §. 1975. Ecological strategies of xylem evolution. Univ. California Press, Berkeley and London. 981. Wood anatomy of Zygogynum eared field observations. Bull. Mus. Hist. Nat. Paris. Sect. B. Adansonia IV. 3: 281-292 JOHANSEN, D. A. 1940. Plant microtechnique. McGraw-Hill Book Co., Kriss, D. A. 1935. Salient lines of specialization in the wood rays of ee Bot. Gaz. 96: 547-557. 1983] CARLQUIST, BELLIOLUM Kobe, ] studies. U.S. Forest KuKACHKA, B. F. 1977. § fractory woods for Service Res. Note FPL-0235: | -9. PATEL, R. N. 1974. Wood sa of the dicotyledons indigenous to New Zealand. 4. Winteraceae. New Zealand Jour. Bot. 12: 19-32. Smit, A. C. 1942. Studies of Papuasian plants. V. Jour. Arnold Arb. 23: 417-443. 1943. Taxonomic notes on the Old World species of Winteraceae. Ibid. 24: 119-1 64. Suss, H. 1980. Uber einiger holzanatomische Entwicklungsrichten bei Laubholzen. Pp. —27 in W. VENT, ed., Beitraége zu Prinzipien und Problemen der Systematik und nee eters enn aus dem Museum fiir Naturkunde. Humboldt University, Ber- lin. aan L. B. 1980. Patterns of pollination in the primitive angiosperms. Biotropica 12: -13. aed P. vAN. 1900. Sur les dicotylédones du groupe des Homoxylées. Jour. Bot. Morot 14: 259-297, 330-361. DEPARTMENT OF BOTANY CLAREMONT GRADUATE SCHOOL AND POMONA COLLEGE AND RANCHO SANTA ANA BOTANIC GARDEN CLAREMONT, CALIFORNIA 91711 JOURNAL OF THE ARNOLD ARBORETUM INSTRUCTIONS FOR AUTHORS General policy The Journal of the Arnold Arboretum is primarily a staff journal, and staff papers have priority. Other papers are accepted, as space permits, from former staff or former students, and from other botanists who have worked on our collections or who have done research on a plant group or in a geographic area of interest to the Arboretum. Submission of manuscripts Manuscripts should be submitted in duplicate to Ms E. B. Schmidt, Managing Editor, Journal of the Arnold Arboretum, 22 Divinity Avenue, Cambridge, Massachusetts 02138. A copy of the manuscript should be retained so that when reviews and/or editorial suggestions are received, any necessary correc- tions can be made and the appropriate portions of the paper resubmitted. For ease of editing, an outline of the paper (not to be published) showing the basic structure of the manuscript should be included. Preparation of manuscripts Papers should be triple spaced throughout (including title, text, citation of specimens, footnotes, acknowledgments, bibliography, and figure legends), on bond (not erasable) paper, with wide margins on all four sides. Nothing should be underlined except generic and infrageneric scientific names, italics when present in a quotation, and the collector and collection number ofall specimens cited. FORM AND STYLE. This can be determined from a recent issue of the Journal. The title should be as short as possible; it should usually contain the name of the family concerned but not authorities of scientific names. Abstracts are generally not used. Abbreviations should be employed only when two or more letters will be saved, and with the exception of units of measure, compass directions, and herbarium designations, they should always be followed by a period. Metric measurements should be used when possible, but information on specimen labels should not be changed. Authority names should be given for all generic and infrageneric taxa the first time they are oe in the text unless they are included in the formal taxonomic treatmen Acknowledgments should be placed at the end of the paper me the bib- liography; the author’s current address should follow the bibliography. Foot- notes should be kept to a minimum and should be numbered consecutively throughout the paper, with the exception of those appearing in tables. Here either standard symbols or lower-case letters should be used. In case of question, the latest edition of Words into Type, by Marjorie E. Skillin et a/. (Prentice-Hall), should be consulted. CITATION OF SPECIMENS. Currently accepted geographic names should be used, with spelling according to a standard source. Names of countries should be in English and should be typed in regular capital and lower-case letters. Below the country level, names may be in the language of the country involved. If this option is taken, careful attention should be paid to consistency, spelling, and accent marks. Arrangement of areas should be consistent (geographic or alphabetic, preferably the former) within a paper. Label information should not be changed unless it is obviously wrong or lacking critical information; in this case, additions or corrections should be bracketed. When a collector has a common surname, his initials should be given. With specimens from the Indo-Malesian area, care should be taken to deter- mine whether collections are institutional or not. In the case of an institutional collection, the institutional series and number (e.g., LAE 20257) should be given. This, in addition to the location, is all the information needed for a brief, unambiguous citation. If it is desirable to include the collector, this information should be placed after the institutional series and number (e.g., LAE 20257, Foreman or LAE 20257 (Foreman)). The list of exsiccatae should be arranged by institutional numbers, where applicable, for ease of use. ILLUSTRATIONS AND LEGENDS. Reference must be made in the text to all maps, figures, and plates. Insofar as possible, their sequence should be determined by the order in which they are mentioned. Plates (illustrations grouped together at the end of an article) and figures (illustrations scattered through the text of an article) should be prepared with Journal page proportions in mind. The maximum size after reduction is 4.25 by 6.5 inches (10.8 by 16.5 cm) for figures, and 4.25 by 6.75 inches (10.8 by 17.1 cm) for plates. A figure may occupy any portion of the length of a page; plates should be more or less full- page size. To facilitate mailing and handling, mounted illustrations must be of a manageable size. Line drawings showing habit and plant parts should be lettered in the same order (e.g., habit, leaf, inflorescence, whole flower, calyx, corolla) on each illus- tration throughout a paper. Photographs should be trimmed, grouped appropriately, and mounted with no space between them on stiff white cardboard with a margin of at least | inch left on all four sides. The author’s name and the figure number(s) should be noted on the back of each illustration. To prevent bending or other damage, the art work should be wrapped carefully and shipped flat. A clear copy of each illustration should be included for review purposes. Legends should be written in telegraphic style (see back issues for examples). They should be grouped in numerical sequence on a separate page, rather than placed below each figure or plate. Illustrations of each type (i.e., figures, plates, or maps) should be numbered consecutively and separately, figures and maps with Arabic numerals and plates with Roman numerals. For example, a paper could include figures 1-3, maps 1-3, and plates I-VII. If the illustrations are to be figures, each figure should be numbered separately, with the numbers running in order through the text. Subdivisions of the figures should be indi- cated with letters. Thus, figures 1-5 may appear on one page, figures 6-8 on the following page, and figure 9, A-C, on the next. In the case of plates, each page of photographs 1s a separate plate; the individual photographs comprising the plates are numbered or lettered consecutively. Illustrations with dark back- grounds should have white letters or numbers and vice versa. Whenever pos- sible, scales should be included in the illustrations; any magnifications necessary in the legends should be calculated to include reduction of illustrations to our page size Illustrations are not returned to the author after publication unless this is requested. TABLES. Titles for tables should be short, with all explanations placed in footnotes. Tables should be as simple as possible and must be neatly typed. Long and/or complicated tables can be photographed directly if they are in good order and the copy is clear enough (in this case the copy should not be triple spaced, but should have spacing appropriate to contents and headings); this eliminates the chance for error and the need for proofreading. Tables should be numbered consecutively using Arabic numerals. Each table should be cited in the text. BIBLIOGRAPHY. The Guide to Citation of Botanical Literature in the Interna- tional code (all editions through 1972) should be followed. When possible, reference should be made to past issues of the Journal for form. Titles should be abbreviated according to Journal (Schwarten & Rickett, Bull. Torrey Bot. Club 85: 277-300. 1958) or Botanico-Periodicum-Huntianum style; the two styles should not be mixed within a paper. Runovers should be indented. Except in cases where confusion would result, only authors’ initials are used in place of first and middle names. All typing should be done in regular capital and lower-case letters, and nothing should be underlined except generic and infra- generic scientific names. Titles of articles and books should not be capitalized except for the first word, scientific names, and proper nouns and adjectives. Page charges Authors are requested to help defray printing costs. Although actual printing costs are much higher, $20.00 is the customary charge, and authors are expected to make every effort to pay. Under special circumstances the fee may be reduced or waived altogether, if this is agreed upon in advance. Ability or inability to pay will in no way affect acceptance or handling of a manuscript. . POSTAL SI T. ERVIC! STATEMENT OF OWNERSHIP, | MANAGEMENT AND CIRCULATION 1. TITLE OF PUBLICATION = A. PUBLICATION NO. ]2 DATE OF FILING Journal of the Arnold Arboretum | 0 (oh ie) | ed | 2| 5| 9/14 ; A B. ANNUAL SUBSCRIPTION ANNUALLY PRICE Quarterly 4 $30.00 22 Divinity Avenue, Cambridge (Middlesex County), Massachusetts 02138 22 Divinity Avenue, Cambridge, Massachusetts 02138 FULL NAMES AND COMPLETE MAILING ADDRESS OF PUBLISHER, EDITO PUBLISHER (Name and Complete Malling Address) Arnold Arboretum of Harvard University, 22 Divinity Avenue, Cambridge, Massachusetts 02138 Stephen A. Spongberg, 22 Divinity Avenue, Cambridge, Massachusetts 02138 Elizabeth B. Schmidt, 22 Divinity Avenue, Cambridge, Massachusetts 02138 be given d fii i d address h individual b ji /f the publica- Saeiags f , ) L NAME | Arnold Arboretum of Harvard University Cambridge, Massachusetts 02138 | | | 8. KNOWN BONDHOLDERS. MORTGAGEES. AND OTHER SECURITY HOLDERS OWNING OR HOLDING 1 PERCENT OR MORE OF TOTAL AMOUNT OF BONDS, MORTGAGES OR OTHER eeeeiee (I there are none, so state) FULL NAME COMPLETE MAILING ADDRESS President and Fellows of Harvard College Cambridge, Massachusetts 02138 ee 9. FOR COMPLETION BY NONPROFIT ORGANIZATIONS AUTHORIZED TO MAIL AT SPECIAL RATES (Section 411.3, ou only) The purpose, function, and nonprofit status of this organization and the exempt status for Federal income tax purposes (Check one) (1) (2) HAS NOT CHANGED DURING HAS CHANGED DURING (If changed, publisher must submit explanation of PRECEDING 12 MONTHS PRECEDING 12 MONTHS change with this statement.) ———— ee COPIES EACH ACTUAL NO. COPIES OF SINGLE 10. EXTENT AND NATURE OF CIRCULATION ‘ISSUE GUAING PRECEDING ISSUE PUBLISHED NEAREST TO 12 MONTHS FILING DATE 800 800 B. PAID CIRCULATION 1. SALES THROUGH DEALERS AND CARRIERS, STREET VENDORS AND COUNTER SALES none none 2 MAIL SUBSCRIPTION TO4 TOT Tou 107 D. FREE DISTRIBUTION BY MAIL, CARRIER OR OTHER MEANS SAMPLES, COMPLIMENTARY, AND OTHER FREE COPIES 3 3 707 710 6. “Vor nor DISTRIBUTED fli 1 USE, L x PRINTING 93 90 none none arora Samat Elet ands A 800 800 u. 1 certify that the statements made by MANAGER, OR OWNER G me above are correct and complete (Ligpbee ee 7. ewe tl Cy COPY PS Form (Page 1) July 1981 3526 Journal of the Arnold Arboretum January, 1983 CONTENTS OF VOLUME 64, NUMBER | The 1980 Sino-American Botanical E ali to Western Hu- bei Province, People’s Republic of Chi B. BARTHOLOMEW, D. E. BOUFFORD, i i: CHANG, Z. CHENG, T. R. DubL_ey, S. A. HE, Y. X. Jin, Q. Y. Li, J. L. LuTeyn, S. A. SPONGBERG, S.C. sun; Y, C PANG. Av WAN. AND T. 5S. YING wid eee eewnds Metasequoia ge nei Present Status in Central Chin BRUCE Cee am Davip E. BOUFFORD, AND STEPHEN-A. Sa ose ak pe Ga ee ee ee es Systematics of Holographis (Acanthaceae). RESIN die A AAU ish args end aoa oa se OE Ee BA LO Wood Anatomy of Belliolum (Winteraceae) and a Note on Flow- ering. SHERWIN (2ARUOUISE o4.4.0b ind 2ct-ende GaSe dw se ee hese08t ibax Volume 63, Number 4, including pages 337-530, was issued December 30, 1982. JOURNAL oF tre ARNOLD ARBORETUM HARVARD UNIVERSITY VOLUME 64 NUMBER 2 US ISSN 0004-2625 Journal of the Arnold Arboretum Published quarterly in January, April, July, and October by the Arnold Arboretum, Harvard University Subscription price $30.00 per year. Subscript d t I i be sent to Ms E. B. Schmidt, Arnold Arboretum, 22 Divinity Avenue, Cambridge, Massachusetts 02138, U.S. A. Claims will not be accepted after six months from the date of issue. POSTMASTER: send address changes to Ms E. B. Schmidt, Arnold Arboretum, 22 Divinity Avenue, Cambridge, Massachusetts 02138, U.S. A. Volumes I-51, reprinted, and some back numbers of volumes 52-56 are available from the Kraus Reprint Corporation, Route 100, Millwood, New York 10546, U.S.A EDITORIAL COMMITTEE S. A. Spongberg, Editor E. B. Schmidt, Managing Editor P. S. Ashton K. S. Bawa P. F. Stevens C. E. Wood, Jr. Printed at Allen Press, Inc., Lawrence, Kansas COVER: The stylized design appearing on the Journal and the offprints was drawn by Karen Stoutsenberger. Second-class postage paid at Boston, Massachusetts, and additional offices. JOURNAL OF THE ARNOLD ARBORETUM VOLUME 64 APRIL 1983 NuMBER 2 SYSTEMATICS OF THE ANDROPOGON VIRGINICUS COMPLEX (GRAMINEAE) CHRISTOPHER S. CAMPBELL Tue ANDROPOGON VIRGINICUS COMPLEX i$ a Closely interrelated group of nine species that range over much of the northern half of the New World (Map 1). These grasses are weedy, diploid, sexual, cespitose perennials. The taxonomy of the broomsedges, as they are commonly called, has been confounded by the paucity and subtlety of taxonomically useful morphological characters and by the strong similarity of many of the taxa. Moreover, chromosome number and flavonoid chemistry have not provided additional usable variation. In this study, taxonomic rank is therefore based on morphological distance, supported by ecological and geographic differences. On the basis of extensive field ob- servations and laboratory and herbarium study of the plants, twenty taxa are recognized. Four of the nine species include two or more varieties, and four of these varieties contain nine taxa, here called variants, that are not sufficiently distinct morphologically to warrant formal nomenclatural status. These vari- ants have been given English names describing some aspect of their mor- phology, ecology, geography, or general nature. Six of the variants correspond to previously described taxa, and three are discussed here for the first time. The center of diversity of the group, both for number of taxa and abundance of individual plants, is the Coastal Plain of the southeastern United States. The combination of effective wind dispersal of the fruit, frequent inbreeding, and relative competitive superiority makes the plants of the virginicus complex successful colonizers under conditions of density-independent mortality. Many of the taxa rapidly form large, dense populations wherever there is full sun and in all but the poorest and driest of soils. Andropogon virginicus L. var. virginicus dominates all other vegetation in the early stages of old-field succession in much of the eastern United States (Keever, 1950; Golley, 1965; Bazzaz, 1975). Like many other inbreeding, colonizing groups, the virginicus complex contains numerous taxa that are very similar morphol lly (Stebbins, 1957; Lelong, 1965: Kannenberg & Allard, 1967). These very similar taxa frequently grow together but rarely produce apparent © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 171-254. April, 1983. 172 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 th 16-18 taxa Map |. Distribution of the Andropogon virginicus complex. hybrids (Campbell, 1980). They are effectively reproductively isolated from one another without being separated by large morphological gaps. These sibling species have been a taxonomic problem because, although they represent dis- crete variation, they are very difficult to distinguish. All of them belong to the three most widespread and weedy species, Andropogon gyrans, A. virginicus, and 4. glomeratus. It is not surprising that authors of previous treatments of these species do not agree on the number of taxa that should be recognized (Tae |). Nash (1903) and Hitchcock (1951) overdescribed these species, and Roberty (1960), Radford et a/. (1964), and Long and Lakela (1971) did not describe the variation sufficiently. Hackel (1889) and Fernald and Griscom (1935) misinterpreted some of the relationships within the virginicus complex, but the treatments by these authors provide the most insightful contributions to an understanding of the taxonomy of the group. TAXONOMIC POSITION AND DEFINITION OF THE VIRGINICUS COMPLEX The Linnaean concept of the genus Andropogon corresponds roughly to the current circumscription of the tribe Andropogoneae. The 12 species recognized by Linnaeus in the first edition of Species Plantarum are now distributed among nine genera. The tribe is a natural assemblage, with a center of distribution in southeastern 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX ee Tas_e 1. Number of species (total taxa) recognized in previous taxonomic treatments of Andropogon gyrans, A. virginicus, and A. glomeratus, as delimited in this work. SPECIES (total taxa) RECOGNIZED AUTHOR gyrans virginicus _glomeratus Total Hackel, 1889 1 (3) 1 (5) 1 (5) 3 (13) Nash, 1903 4 3 (5) ig (12) Fernald & Griscom, 1935 1 (3) 1 (8) 6) 2 (11) Hitchcock, 1951 3 3 (4) ] 7 (8) Roberty, 1960 0 1 (7) 0) 1 (7) Radford er al., 1964 2 l 0 3 Long & Lakela, 1971 2 2 (3) ] 5 (6) Campbell (present paper) 1 (3) 1 (5) 1 (5) 3 (13) Asia (Hartley, 1950, 1958), and its cohesiveness rests on similarity in many aspects of morphology (Clayton, 1972), cytology (Celarier, 1956, 1957, 1958, 1959), anatomy (Metcalfe, 1960; Chaudra & Saxena, 1964; F. W. Gould, 1967), and biochemistry (Guttierrez ef al., 1974). It has been the focus of several major taxonomic studies (Hackel, 1889; Keng, 1939; Roberty, 1960) and has been considered the most advanced tribe of the Gramineae. If one agrees with Stebbins (1974), it is the most advanced taxon of all angiosperms. Although the tribe is easily circumscribed, the delimitation of its genera has been problematic. Clayton (1972) used a numerical approach based on 41 morphological characters to cluster the genera. He admitted that his findings were tentative and that much remained to be done. For Andropogon he accepted Stapf’s (1919) classification (1967, and pers. comm.). His concept of the genus is narrow and does not include Bothriochloa Kuntze, Dichanthium Willimet, or Schizachyrium Nees. However, the genus remains large and heterogeneous (F. W. Gould, 1967). The largest of Stapf’s four sections, LepropoGon Stapf, contains approximately 55 species (Clayton, pers. comm.), with about 20 in Africa and the balance in the New World. The complex of species centering around A. virginicus includes the majority of species of the section that occur in the northern areas of the New World. Only three species of sect. LEPToPOGON not included in the virginicus com- plex have been collected in the United States. Andropogon ternarius Michaux (including A. ternarius var. cabanisii (Hackel) Fernald & Griscom) is common in much of the eastern United States and rare in Mexico. Andropogon bicornis L., widespread in South and Central America and the West Indies, has been collected only once in the United States (Florida, Monroe County, Craighead s.n., 1962 (FrG)). Andropogon gracilis Sprengel is rather frequent on the Miami oolite of Dade and Monroe counties, Florida. Andropogon gerardii Vitman and 4. hallii Hackel of sect. ANDROPOGON are the only other species of the genus in the United States. South of the United States, 12 other species of sect. LEPTOPOGON occur within the range of the virginicus complex. Andropogon virgatus Desv. (probably better treated as Hypogynium virgatum (Desv.) Dan- dy) is morphologically the most distant of these species and will not be con- 174 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 sidered further. The nine other species outside the virginicus complex in the West Indies are 4. urbanianus Hitche., A. gracilis, A. reedii Hitchc. & Ekman, A. reinoldii Leon, A. nashianus Hitche., A. lateralis Nees, A. leucostachyus HBK., A. selloanus (Hackel) Hackel, and A. bicornis. The last four also grow in Central America, and in Mexico along with three Mexican endemics, 4. bourgaei Hackel, A. pringlei Scribner, and A. spadiceus Swallen. The four species that occur both east and west of the Gulf of Mexico also extend into South America. The section is well represented there, but only one species, J. arenarius Hackel, is very similar to species of the virginicus complex. Finally, A. leucostachyus is the only species found in both the New and Old Worlds (possibly introduced into Africa (Stapf, 1919)). Three African species of sect. Leptopocon, A. eucomus Nees, A. huillensis Rendle, and 4. laxatus Stapf, resemble the species of the virginicus complex in general aspect but are excluded from the complex on morphological grounds (see below). The essential features characterizing all the taxa of the virginicus complex and combining to distinguish them from the other species of sect. LEPTOPOGON are presented in Taste 2. The reduction in stamen number from three to one, the single most distinctive feature of members of the virginicus complex, is found in many other genera of grasses: Cinna L., Uniola L. (Booth, 1964), Stipa L. (Brown, 1949), Bothriochloa (Heslop-Harrison, 1961), Briza L. (Mur- ray, 1974), Imperata Cyr. (Clifford, 1961), Deschampsia Beauv. (Parodi, 1949), Chikusichloa Koidz. (Connor, 1979), Poa L. (Weatherwax, 1929), and Vulpia K. C. Gmelin (Cotton & Stace, 1977). In these genera, as in the virginicus complex, the reduction in stamen number is associated with the capacity for cleistogamy (Campbell, 1982b). The only other species with one stamen in sect. LEPTOPOGON are Andropogon gracilis and the two closely related Mexican endemics, A. pringlei and A. spadiceus. Significantly, all three produce cleis- togamous flowers, although cleistogamy in these taxa differs from that in the virginicus complex (Campbell, 1982b). Because 4. gracilis has one raceme per inflorescence unit, several authors have placed it in Schizachyrium (S. gracile (Sprengel) Nash). However, it does not have the cupuliform rachis internodes and rounded glumes of the sessile spikelets that Clayton (1964) considered to be important features of Schizachyrium. In addition to having solitary racemes, A. gracilis differs from the virginicus complex in being tetraploid (n = 20: Davidse & Pohl, 1972). Andropogon pringlei and A. spadiceus differ from the virginicus complex in three morphological characters (TABLE 2). f the 16 species listed in TABLE 2, only Andropogon huillensis is separated from the virginicus complex in less than two ways. There may be a second difference for this species, depending upon what its chromosome number really is (two different ploidy levels have been reported—see TaBLe 2). Six species differ in only two ways. The wide-ranging 4. /ateralis (and the numerous species closely related to it) is quite different from the virginicus complex in having large, functional, pediceled spikelets. For some of the other five species listed in TABLE 2, there are additional differences that are not given in the table. The racemes of A. selloanus and A. eucomus are more densely pubescent. The lower glumes of 4. nashianus are ovate in shape and distinctly different from the lanceolate to oblong lower glumes of plants of the virginicus complex. Mor- TABLE 2. Comparison of the Andropogon virginicus complex with its closest relatives. SPIKELET CHROMO- STAMEN LENGTH AWN BASE PEDICELED SOME SPECIES NUMBER (m AWN TWISTING INTERNODE NUMBER A. virginicus complex 3-5 Present Moderate 0 to vestigial Thin n= 10* A. arenarius 3 4-6 Usually _ 0 to functional Thin Unknown absent A. bicornis 3 3-4 Absent — 0 to functional Thin n = 30 (Pohl & Davidse, 1971) A. bourgaei 3 ca. 4 Absent —- 0 to functional Thin Unknown A. eucomus ) 2.5-3 Present Moderate 0 to vestigial Thin n= 10 (de Wet, 1954) A. huillensis 3 ca. 4 Present Moderate O to vestigial Thin n = 30 (de Wet, 1960) n= 10 oe 1978) A. lateralis 3 3.5-5 Present Moderate O to functional Thin Unkno A. laxatus 3 4-5 Present Strong O to vestigial Thin aiiew A. leucostachyus 3 2.5-3.5 Abse — 0 to vestigial Thin n=10 E W. Gould, 1956) A. nashianus 3 2.5-3.5 Present Moderate 0 to vestigial Thin Unknown A. pringlei/spadiceus 6-7 Present Strong 0 to vestigial Thick Unknown A. reedii 3 4-6 Present Stron 0 to functional Thin Unknown A. reinoldii 3 2.5-3 Present Moderate 0 to vestigial Thin Unknown A. selloanus 3 ca. 4 Absent _ 0 to vestigial Thin n= 10(F. W. Gould, 1956) A. ternarius 3 5-6.5 Present Moderate 0 to vestigial Thick = 20 (F. W. Gould, 1956) A. urbanianus 3 4-6 Present Strong 0 to functional Thin n= 40 (Davidse & Pohl, 1972) *See Table 3. XATMNOO SNOINIDUIA NODOdOUGNY “TISddNVO — [E861 I GL 176 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Figures 1-7. 1, 2, dispersal units, scale bar = 2 mm: 1, old-field variant of Andro- pogon virginicus var. virginicus (Campbell 3770), common eseaey oe ee var. gyrans (Campbell 3782). 3, bases of raceme sheaths showing sp moderate, and e pubescence, bar = 2 mm. 4, ligules (arrow) of A. g/omeratus var. ee Camp- 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX LIT phologically, A. Auillensis, A. reinoldii, and A. laxatus remain as the species closest to the virginicus complex. The uniformity of the virginicus complex in the features shown in TABLE 2 argues for its naturalness. Other evidence for this supposition comes from the chemical similarity of the taxa. Eighteen of the taxa were surveyed with standard methods of paper chromatography (Mabry et a/., 1970): sixteen had nearly identical spot patterns. The two that vary somewhat, Andropogon arctatus and the robust variant of A. glomeratus var. pumilus, differ from the remainder of the virginicus complex in other ways as well. Andropogon arctatus has rhi- zomes, the most primitive flowering mode of the virginicus complex, and a different sort of life history. The robust variant has the widest tolerance for soil salinity and pH of all taxa of the virginicus complex. GENERAL DESCRIPTION OF THE VIRGINICUS COMPLEX The yearly growth of plants of the virginicus complex is divided into two phases. In the first part of the season, growth is vegetative in the form of a proaxis of basal leaves with very shortened internodes. In July, August, or September (presumably in response to short day-length, as in other species of the Andropogoneae (Tompsett, 1976)), the proaxis extends rapidly to produce a flowering stem. The stems are terete and 0.2-3.1 m in height. The leaves may be glabrous or pubescent with unicellular macrohairs, sca- brous on the adaxial blade surface and, in some taxa, on the abaxial sheath surface; and glaucous or green. The leaf blades are flattened to occasionally folded, and laxly recurving to infrequently erect. The ligules' are membrana- ceous and 0.2—2.2 mm long, with margins glabrous to long-ciliate. The inflorescence is a complex branch system. Stapf (1919) described the branching of Andropogon as cymose. At each node of plants of the virginicus complex, one to eleven branches may develop on a very short common axis. The largest branch is next to the parent axis, and successively smaller ones arise distally on the common axis. The branches alternate in two ranks that are close to one another on the adaxial side of the common axis. The first structure on the common axis is a two-nerved, hyaline prophyll separating the first branch from the parent axis. Between each of the distal branches there is also a prophyll (Ficure 10). The tips of the ultimate branches of the inflorescence bear structures here called inflorescence units (see APPENDIX A and FiGureE 11). The raceme sheath iCertain terms, of a technical nature or specialized in their meaning, are frequently used in this paper. They are defined in AppeNpIx A, and some are illustrated in Ficures 1-11. bell 3805) and deceptive variant of A. virginicus var. virginicus (right, Campbell 3747), bar = 1 mm. 5, fruits of commen variant of A. gyrans var. gyrans (Campbell 3872) (note marcescent anther at apex of left and center fruits), bar = 1 mm. 6, 7, scanning electron micrographs of stem sheaths, scale bar = | um: 6, deceptive variant of A. virginicus var. virginicus (Campbell 3870) (note bicellular microhairs); 7, A. glomeratus var. glomeratus (Campbell 3973) (note apically directed prickle hairs). 178 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 a ® - common axis Kw - stem sheath > - prophyll © - inflorescence unit Ficures 8-10. 8, 9, dispersal units: 8, old-field variant of eel ee virginicus var. VIrginicus (pubescence not included), showing one-keeled upper glume and base of awn; below middle of keels of lower glume). 10, diagram of inflorescence node of plant of virginicus oo c = callus, p = pedicel, ps = pediceled spikelet, r = rachis internode, s = spikele subtends the peduncle, which bears at its apex two or more racemes. The raceme is made up of a straight axis, the rachis, which has four to fourteen nodes. At each node there is a pair of spikelets (a major character for the tribe Andropogoneae). In the virginicus complex, one of the spikelets is sessile and 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 179 the other is vestigial or entirely absent and represented only by its pedicel. The sessile spikelet has two subcoriaceous glumes. The lower glume has two keels, and the upper glume has one. The glumes enclose two florets, the lower of which is sterile (as in most members of subfam. Panicoideae) and is represented only by a hyaline lemma. The upper floret consists of a long-awned, hyaline lemma, a vestigial (0.4-1.5 mm long) palea, and a flower with two lodicules, one stamen, and a two-styled ovary. The ovary matures into a brownish or purplish caryopsis. FLOWERING The 15 taxa observed flower consistently around dawn in the late summer and early fall. Populations in the northeastern United States tend to flower in August and September, and those in the south from September to November. In any region there is great overlap in flowering time between most of the taxa. Exceptions are noted under individual taxa. Vernal flowering occurs in some taxa (e.g., Andropogon brachystachyus and especially A. /ongiberbis), apparently in response to burning. In Mexico, Central America, and the West Indies late summer to fall flowering predominates. Variation in flowering in the virginicus complex is considerable (Campbell, 1982b). Andropogon arctatus, A. tracyi, A. floridanus, A. brachystachyus, and some taxa of A. virginicus and A. glomeratus produce mostly chasmoga- mous flowers (see character 33 of TABLE 4). The synchrony of peduncle elon- gation and floral maturation exposes the spikelets above the raceme sheath at anthesis. The lodicules can then open the spikelet. The filament elongates in about an hour, and the anther and styles protrude for potential cross-polli- nation. If the peduncle does not elongate sufficiently to free the spikelet above the raceme sheath, the lodicules cannot open the spikelet, and cleistogamy re- sults. When the peduncle does not elongate appreciably, the combined action of many lodicules forces the racemes out between the margins of the raceme sheath. The lodicules then open the spikelets. If, however, the flowers mature before the inflorescence unit emerges from the stem sheath in which it developed, the lodicules cannot open the spikelets. Cleistogamy is far more common in short- peduncled than in long-peduncled racemes (see character 33 of TABLE 4). CHROMOSOME NUMBERS Counts of a haploid chromosome number of 10 have been made for all species and for 16 of the 21 taxa (TABLE 3). Because of the possibility of misidentification of these morphologically close grasses, published reports are less reliable than those for which a voucher has been examined. The two polyploid counts reported for Andropogon virginicus are strongly questioned _ because of the uniform diploidy of the complex. The count of 2m = 40 may have been based on a mitotic division in the anther similar to the one shown in FicurE 16. The chromosomes appear to be meiotic, and the nucleolus marks 180 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Figure 11. Inflorescence units: A, B, common variant of Aen gVvrans v a (Campbell — C, D, ye ea gvrans var. stenophyllus (Campbell 3822): , tenuous variant of A. gyrans var. gyrans (Campbell 3873), G, A. brachystachyus (Campo 3884), H, A. floridanus (Campbell 3754), 1, A. longiberbis (Campbell 3729). . iebmannii var. pungensis (Campbell 3948). K, A. arctatus (Campbell 3944); L, M, 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 181 the stage as prophase, but another cell from the same anther, a pollen mother cell in prophase I, showed ten bivalents. For the 16 counts reported in this work, young spikelets were fixed in Farmer’s solution (3 parts 100% ethyl alcohol, | part glacial acetic acid), transferred to 70% ethyl alcohol after 24 hours, and stored at about 4°C. Fixation early in the morning yielded far more meiotic figures than that done later in the day. The ten new counts reported here are shown in Ficures 12-21. In all taxa studied the basipetal sequence of floral maturation facilitated finding prophase I to anaphase I. In racemes of wholly chasmogamous flowers, the developmental gradient is steeper than in those with cleistogamous flowers. In the former group, one raceme may contain nearly mature pollen in the apical spikelets and pollen mother cells in the basal spikelets. In contrast, meiosis may occur concurrently in all flowers of a raceme in the frequently or predominantly cleistogamous taxa. GEOGRAPHIC DISTRIBUTION Eighteen of the taxa grow on the Coastal Plain of the southeastern United States (see Map 1). The distribution of 12 is entirely limited to the region extending from southernmost New Jersey to eastern Texas. Andropogon arc- tatus, A. floridanus, A. longiberbis, and A. brachystachyus grow principally in Florida and infrequently in nearby states or the Bahamas. Only one taxon of eastern United States broomsedges, 4. liebmannii var. pungensis, does not occur extensively in Florida. The common variant of Andropogon gyrans var. gyrans, the old-field variant of A. virginicus var. virginicus, the robust variant of A. glomeratus var. pumilus, and the two varieties of 4. /iebmannii show a pattern of eastern United States— Mexican distributions common to many other vascular plants (Graham, 1973). Because of the uncertainties about the phytogeographic relationships of these two regions and the paucity of andropogonoid fossils, assertions concerning the direction of migration are conjectural. However, one line of reasoning suggests migration from the center of distribution to Mexico. This rests on the hypothesis (developed more fully in the discussion under A. brachystachyus) that A. virginicus is derived from A. brachystachyus ancestry. The range of the latter taxon and most of the variants of 4. virginicus is restricted to the south- eastern United States. Only the old-field variant of var. virginicus grows in Mexico, possibly after migration there. Closely related taxa in Andropogon virginicus and A. glomeratus have the old-field variant of A. virginicus var. virginicus (Campbell 3849), N, deceptive variant of A. virginicus var. virginicus (Campbell 4081), O, drylands variant of A. virginicus var. glaucus (Campbell 3898), P, robust variant of A. glomeratus var. pumilus (Campbell 3850); Q, A. glomeratus var. hirsutior (Campbell 3804), R, A. glomeratus var. glaucopsis (Campbell 3806); S, A. glomeratus var. glomeratus (Campbell 3915), T, A. tracyl (Camp- bell 4100). Bar = 5 cm. TABLE 3. Chromosome numbers in the Andropogon virginicus complex. TAXON COUNT LOCALITY VOUCHER SPECIMEN OR REFERENCE A, arctatus n= 10 Florida, Liberty Co. Campbell 4060 (Gu) A. brachystachyus n= 10 Florida, Putnam Co. Campbell 4264 (Gu) A. floridanus 2n = 20 No locality given Carnahan & Hill, 1961 n= 10 Florida, Marion Co. Campbell 4194 (Gu) A. glomeratus n= 10 No locality given Church, 1936 var. glaucopsis 2n = 20 Florida, Duval and St. Johns cos. Church, 1940 var. glomeratus n= 10 Massachusetts, Plymouth Co. Campbell 3973 (GH) 2n = 20 Rhode Island, Washington Co. Church, 1940 var. hirsutior n= 10 Georgia, Berrien Co. Celarier (Mo) Alabama, Baldwin Co. Campbell 3804 (GH) var. pumilus robust variant n= 10 Texas, Fort Bend Co. . W. Gould, 1956 Georgia, Berrien Celarier A- 2600-I (mo) Alabama, Montgomery Co eh ae (GH) No locality aes Gage Mexico, V Campbell 3648 ue Dominican eects Espaillat vidse & Poh 2n = 20 Texas, Robertson Co eee 1940 Texas, no county given Brown, 195 Mexico, Chiapas F. W. Gould & Soderstrom, 1970 A. gyrans var. gyrans common variant n= 10 Gould RF 506 (TAES) Louisiana, Rapides Parish Georgia, Tift Co. Florida, Jackson Co. Celarier A-2608-I (Mo) Campbell 3815 (GH) WOLAYOUAUV GIONYV AHL JO TVNANOL c81 p9 104] tenuous variant var. stenophyllus A. longiberbis A. liebmannii var. pungensis A. tracyi A. virginicus var. glaucus wetlands variant var. virginicus deceptive variant old-field variant 2n = 20 n= 10 n= 10 n= 10 n= 10 n= 10 n= 10 n= 10 2n = 20 2n = 40* n= 40* n= 10 n= 10 Georgia, McIntosh Co. Mexico, Chiapas Florida, Highlands Co. Florida, Liberty Co. Florida, Dade Co. Alabama, Washington Co. Florida, Putnam Co. Florida, Jackson Co. Massachusetts, Barnstable Co. Texas, Robertson and San Augustine cos. Virginia, Princess Anne Co. Tennessee, no county given No locality given Florida, Jackson Co. Rhode Island, Washington Co. Georgia, Berrien Co. Florida, Jackson Co. Florida, Putnam Co. Church, 1940 Tateoka, 1962 Campbell 3746 (GH) Campbell 4259 (GH) Campbell 3734 (GH) Campbell 3695 (GH) Campbell 4100 (Gu) Campbell 3818 (GH) Alava 3246, 3252 (Mo) Mohlenbrock, 1973 Campbell 3778 (Gu) Celarier A-2620-I (Mo) Campbell 4265 (Gu) *Probably misinterpretations. XITdNOO SNOINIDUIA NODOdOUGNY “TIAddNVO [E861 I €8 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 184 a Oe ds * ad ‘% f & te By’ . p ; oY ? % * * t ee ee 15 SMe '6) ‘ : r e fee of ¥ ay »#,\¥j "> — % S » , , 18 Chromosomes (see TABLE 3 for voucher sey unless otherwise noted, 720: 13, deceptive variant 15, 4. Ficures 12-21. all meiotic figures): 12, Andropogon floridanus, ey renee Az Hany, diakinesis, X 720; of A. virginicus var. virginicus, anap : liebmannil var. pungensis, diakinesis, X 8 ans var. stenophyllus, mitotic 18, 4. lon prophase in an anther, X 965; 17, A. a oe. X 825; - 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 185 most similar ranges. The wetlands, drylands, deceptive, and smooth variants of the former and Andropogon glomeratus vars. glaucopsis and hirsutior are found on the Coastal Plain of the southeastern United States, especially from southern Virginia and Mississippi (MAps 13-16, 18, 19). ECOLOGY The requirements of high levels of sunlight and little competition usually confine populations of the virginicus complex to areas undergoing early succes- sion or to persistently disturbed sites; they characteristically grow along roadsides or under powerlines, or they invade abandoned agricultural land or clear-cut timberlands (Ficures 22-27). The old-field variant of Andropogon virginicus var. virginicus is so successful as a colonizer (Keever, 1950; Golley, 1956; Bazzaz, 1975) that the three- to five-year period after an environmental opening has occurred is called the broomsedge stage. The success of members of the virginicus complex as weeds is certainly due in part to their dispersibility, both locally and to newly available sites. The long, spreading hairs of the dispersal unit provide aerodynamic drag for wind dispersal; the terminal velocity of the dispersal units is roughly equivalent to that of the common dandelion (Campbell, 1983). Because they are considerably taller than dandelions, the individual broomsedge plants have a greater poten- tial dispersal range. The probability of their long-distance dispersal increases through their capacity for autogamy. The increase in homozygosity that ac- companies inbreeding apparently promotes close adaptation to a particular environment (Allard, 1975). The most successful colonizers of the virginicus complex—the common variant of Andropogon gyrans var. gyrans, the old- field and deceptive variants of A. virginicus var. virginicus, and A. glomeratus vars. hirsutior and glaucopsis—bear mostly cleistogamous flowers. The one no- table exception to this generality is the robust variant of 4. glomeratus var. pumilus. These andropogons will flourish in partial shade if competition is suppressed through fire. In two pine plantations in the southeastern United States, there are extensive populations of the virginicus complex where annual burns prevent hardwood growth. Over large areas in both localities, the fall reproductive growth of the broomsedges dominates the undergrowth (FIGURE 26). At the plantation in Thomas County, Georgia, there are large populations of eight taxa of the complex, and at the Bladen County, North Carolina, site six taxa grow together. Clayton (1969) pointed to a possible adaptation for fire ecology of Hyparrhenia Andersson of the Andropogoneae in the demonstrated ability of their dispersal units to bury themselves. The hygroscopic awn and the callus berbis, diakinesis, X 715; 19, wetlands variant of 4. virginicus var. glaucus, diakinesis, x 790; 20, tenuous variant of 4. gyrans var. gyrans, diakinesis, x 940; 21, A. arctatus, diakinesis, X 715. (Arrows in 12 and 18 indicate two bivalents not resolvable in this focal plane.) 186 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 PE yada Wek, FIGURES 22-27. 22, 23, old-fleld variant of Andropogon virginicus var. virginicus: 22, under powerline, Kent Co., Maryland; 23, in four-year- old Pinus elliottii timberland, Decatur Co., Georgia. 24, A. ae in clearing in flatwoods, Highlands Co.., Florida. 25, clear-cut P. clausa timberlands colonized by A. ae 26, several taxa 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 187 hairs work the spikelet into the soil, where it is more protected from fire. The broomsedge dispersal unit may function in the same way, although the awn is often barely twisted (and therefore weakly hygroscopic). Most of the taxa of the virginicus complex have rather narrow tolerance for soil moisture. For example, in the panhandle of Florida, one characteristically finds the wetlands variant of Andropogon virginicus var. glaucus at the bottom of roadside ditches. Andropogon gyrans var. stenophyllus often borders the same ditches, and above the ditches one is likely to see the deceptive variant of A. virginicus var. virginicus, the common variant of A. gyrans var. gyrans, and the drylands variant of A. virginicus var. glaucus. The old-field variant of var. virginicus, tolerant of a wide variety of soil moisture conditions, may occur anywhere on the slope. While human activity has been decidedly beneficial to these broomsedges, the plants are commercially undesirable since they supplant other vegetation of greater economic value as fodder. Cattle generally eschew most of these andropogons. The only direct economic value of the plants is in their leaves and stems, which have long been used for dyeing fabric and for brooms. EVOLUTION AND SPECIATION A dominant trend in the evolution of the virginicus complex has been the shift from chasmogamy to cleistogamy. Stebbins (1957) presented convincing evidence for the primitive nature of outcrossing and the derived condition of inbreeding. The derivation of cleistogamy in the virginicus complex involved shortening the peduncle and developmental shifts in the maturation of the flowers (Campbell, 1982b). Smaller spikelets and anthers and increased - weediness are often associated with a shift toward cleistogamy. The reduction in anther length and pollen-producing capacity may result from relaxed selec- tion for the copious pollen needed for wind pollination. Evidence suggests that the change from chasmogamy to cleistogamy has repeatedly occurred in the virginicus complex (Campbell, 1982b). Both the tenuous variant of Andropogon gyrans var. gyrans and var. stenophyllus of the same species consist of two kinds of plants, one with some long peduncles and chasmogamous flowers and one with all short peduncles and mostly cleistog- amous flowers. The plants are otherwise similar, but the cleistogamous form of both taxa is more widespread and common. Another derivation of cleistogamy involves the precocious sexual maturation of ancestors of Andropogon brachystachyus to produce the deceptive variant of A. virginicus var. virginicus and, by other character changes, the rest of A. virginicus. Finally, A. glomeratus var. glomeratus appears to be the ancestral stock for both vars. Airsutior and glaucopsis. nnn =. of A. gyrans, A. virginicus, and A. glomeratus in annually burned P. palustris plan- tation, Thomas Co., Georgia. 27, robust variant of A. glomeratus var. pumilus in roadside ditch, Putnam Co., Florida. 188 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Cleistogamy presumably brings to the virginicus complex both greater col- onizing ability (see ecology section) and a barrier to gene flow between the taxa (Campbell, 1982b). This barrier contributes to the reproductive isolation of the taxa that grow together. Mixed populations of broomsedges are so com- mon now that one might assume that they have always grown together and that speciation was sympatric. Cleistogamy might have provided the interrup- tion in gene flow essential to divergence. However, the phenomenon of mixed populations may have developed largely in the presence of man’s tremendous disturbance of the habitat. The taxa that are so common today may once have been restricted to infrequent openings caused by fire and other natural distur- bance. Their divergence from one another may have taken place in isolation in pre-Columbian times. Many of the taxa that frequently grow near one another in mixed populations are very closely related. For example, two varieties of Andropogon glomeratus, vars. hirsutior and glaucopsis, form mixed populations in 67 percent of the 28 populations I have studied (Campbell, 1980). I have not detected plants of intermediate morphology that would suggest gene flow between these or most of the other closely related taxa. HYBRIDIZATION Hybridization between taxa of the virginicus complex is rare. There are ample opportunities for gene flow between taxa because they frequently grow together and flower at the same time of day and (mostly) at the same time of year. I have observed two taxa growing within one to three meters of one another over four hundred times. In only five of these opportunities for hybridization were there plants whose intermediate morphology suggested that they were hybrids. In the rare instances when hybridization does take place, there are few mature hybrid individuals. I have found only twelve putative hybrid individuals in the five localities where hybridization is suspected. The parents outnumber these hybrids by between five and one hundred or more to one. Four of the suspected hybridizations involve the old-field variant of Andro- pogon virginicus var. virginicus. Andropogon longiberbis and the robust variant of A. glomeratus var. pumilus are the other parents in two cases each. In the fifth instance the robust variant and A. longiberbis produced six putative F, plants. In all these instances the putative hybrids are morphologically intermediate between the parents. The three putative F, hybrids of the old-field variant and Andropogon longiberbis fall between the parents in the orientation of sheath pubescence, the number of racemes per inflorescence unit, and the lengths of the racemes, spikelets, and callus hairs. The three putative hybrids of the old- field and robust variants are intermediate in leaf-blade and raceme-sheath width and in maximum number of inflorescence branches. In leaf-blade length and number of inflorescence units per stem, however, the hybrids from both pop- ulations exceed the mean of these characters for both parents. For the robust variant and A. /ongiberbis there are ten characters distinguishing the parents for which the putative hybrids are more or less intermediate: stem height, leaf- 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 189 5T e ° e °o 4+ ~-_~ + ° E S ao "5 = 3 ao 6G Oo z= WW —_ to a? 22 Lu x oO i?) 1+ 1 2 3 4 RACEME SHEATH WIDTH (mm) Ficure 28. Graph of spikelet length vs. raceme-sheath length for seven populations each of robust variant of Andropogon glomeratus var. pumilus (squares) and A. longiberbis (circles; darkened circles indicate two populations with identical values), and for six putative hybrid individuals represented by Campbell 4266, Lake County, Florida (cross). Each symbol shows mean of four or five parent plants or six hybrids. Walue for each individual is mean of ten measurements of spikelets and raceme sheaths. pubescence orientation, leaf-blade length and width, number of inflorescence units per stem, raceme-sheath width, peduncle length, spikelet length, callus- hair length, and spikelet-keel scabrousness. Two of the best characters for distinguishing the two parent taxa are raceme-sheath width and spikelet length (Figure 28). Variation in these characters for the two parents and the six putative hybrids has been analyzed using discriminant functions (Campbell, 190 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 1982a). The six putative hybrids are relatively much less fertile than either parent I have observed only two barriers to gene flow in the virginicus complex: cleistogamy and, less importantly, nonsynchronous flowering. Interspecific pol- len-style incompatibility, in need of further study, may also contribute to re- productive isolation. MORPHOLOGY IN RELATION TO TAXONOMY TABLE 4, which contains information concerning the taxonomically useful variation in the virginicus complex, replaces separate descriptions for the taxa. This format facilitates comparisons between the character states of different taxa and may be used as a multiple-entry key. An explanation of the mea- surements and abbreviations used is presented in the “Character States” section on pages 194-199, The first 11 of the 33 characters listed in TABLE 4 pertain to vegetative parts of the plants. Stem height and leaf-blade dimensions vary together and divide the virginicus complex into three rough groups: Andropogon glomeratus and A, floridanus have the largest dimensions for these characters; 4A. gyrans, A. tracyl, and A. liebmannii var. liebmannii have the smallest; the rest are mostly intermediate. Because of the overlap between many of the taxa for these char- acters, they are secondary in importance. Much of the overlap is apparently due to phenotypic response to varying levels of soil moisture and nutrients. The glaucousness of the leaves and stem internodes of certain taxa is due to a wax that can be rubbed off. It is most conspicuous on young, fresh material. Glaucousness of the leaf is a more reliable character than that of the stem, and it distinguishes both Andropogon glomeratus var. glaucopsis and A. virginicus var. glaucus from all other taxa. Infrequently, the leaves of the smooth variant of A. virginicus var. virginicus are somewhat glaucous. Variability in the amount and distribution of two types of trichomes is taxonomically useful. The first type is the macroscopic, unicellular hair found on the leaves, below the raceme sheaths, and in the racemes. In this work “pubescence” refers to these hairs. Glabrous leaves lack such hairs. Generally the best place to find pubescence on leaves that are sparsely hairy is near the collar. Glaucous leaves are generally glabrous. Leaf pubescence may disappear as the foliage ages and may also become more appressed. It is therefore best to examine young leaves when trying to determine whether pubescence is spreading or appressed. The second type of trichome is the microscopic prickle hair; these hairs are usually directed apically on the leaves and keels of the lower glume of the spikelets. When sufficiently abundant, as in plants of Andropogon glomeratus (Figure 7) and A. floridanus, they make the stem sheath scabrous. Pressing the stem sheath firmly with the tip of the finger and moving the finger down the sheath will indicate the absence or the sandpapery effect of the presence of these prickles. They are generally best developed near the collar. The length of the ligule and its marginal ciliations (FiGURE 4) and the color of the ligule are useful characters; they separate most of Andropogon glomeratus 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 191 from the remainder of the virginicus complex. Previous authors have noted the characteristically long ligules of A. gyrans var. stenophyllus. Inflorescence shape is the most conspicuous character of the broomsedges. Since there is considerable variation in this character, authors of earlier works have focused on it and have been confused by the taxa with similar shapes. For example, the range of inflorescence shapes of the robust variant of Andro- pogon glomeratus var. pumilus (FiGuReE 51) includes those of the rest of A. glomeratus, A. floridanus (FiGuRE 40), and parts of A. virginicus (FIGURES 42, 45) and A. longiberbis (Ficure 41). The similarity of inflorescence shape in A, glomeratus and A. virginicus is one important factor in the union of the two species by recent American authors (see TABLE 1). Andropogon brachystachyus and sometimes the drylands variant of A. vir- ginicus var. glaucus produce long inflorescence branches that spread away from the main stem in smooth arches (Ficures 24, 43, 44), allowing ready distinction of these plants in the field. Herbarium specimens of these two taxa tend to have rather open inflorescences. Characters 14 and 15 are attempts to quantify the denseness of the inflores- cence. Order of branching refers to the number of rebranchings. For example, Ficure 10 depicts third-order branching (i.e., three branches at the stem node). These two characters and the number of inflorescence units per stem divide the virginicus complex into three groups that correspond roughly with the groups based on vegetative size. Like stem height and leaf-blade dimensions, they often reflect the suitability of a particular individual’s growing conditions. Most of the remaining characters vary independently of stem and leaf dimen- sions, inflorescence density, and environmental conditions. Categories of pubescence density below the raceme sheath are shown in Figure 3. Seven of the 20 taxa express more than half of the states for this character. There is no overlap in character states in only 25 of the 190 different pairwise combinations of taxa. The location of the raceme sheath is unique in the common variant of Andropogon gyrans var. gyrans. Toward the apex of the stem, the internodes are shortened and the stem sheaths become inflated and strongly overlapping. The inflorescence units remain mostly concealed within these inflated stem sheaths. Even though some of the peduncles may elongate and expose the racemes, their subtending raceme sheaths are rarely visible until senescence (FiGures 29, 30). Raceme-sheath dimensions, particularly the width, are very useful in distin- guishing many of the taxa because of nonoverlapping ranges of mean low and high measurement values. Peduncle length, spikelet length and width, anther length, and mean percent chasmogamy reflect flowering mode. Mostly chasmogamous plants have long peduncles, large spikelets, and large anthers that are not marcescent at the apex of the fruit within the spikelet (Campbell, 1982b). Plants with predominantly cleistogamous flowers have short peduncles and smaller spikelets and stamens, and the anthers are usually marcescent within the spikelets (see below). Raceme number distinguishes Andropogon liebmannii and is often helpful 192 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 TABLE 4. Character states. Character number: 2 3 4 Stem Leal Sample glaucous-— glaucous- pubescence Taxon size height (m.) ness ness amount ARCT 17 (8) (0.9): 423° 127) O (1) 0 O to D GYRA 68 (28) (0.3) 0.8 (1.4) (O) 1 (2) 0 O to D GYRA 37 (17) (0.5) 0.8 (1.4) (O) 1 (2) 0) O to D comm 23 (10) (0.5) 0.8 (1.4) (0) 1 0 O to D tenu 14 (7) (0.5) 0.8 (1.0) (0) 1 (2) 0 O to § STEN 31 (11) (0.3) 0.7 (1.0) 0,1 (2) 0 0 (S) TRAC 13 (7) (0.5) 0.8 (1.2) O51 0 0 to 8 LIEB 12°C) (0.2) 0.9 (1.7) 0 0 S to D LIEB 6 (5) (0.2) 0.6 (0.9) 0 0 S to D PUNG 6 (6) (0.8) 1.2 (1.7) 0 0 M to D FLOR 18 (8) CO) 1a C2 ob) O (1) 0 0 (S) LONG 38 (13) (0.5) 0.9 (1.5) O51 0 S to D VIRG 147 (75) (0.4) 1.2 (2.1) 0,1,2 0,1 0 to D VIRG 98 (43) COA): 1221) Os ay 0, O to D oldf£ 61 (26) (0.4) 1.2 (2.1) 0,1 0 S to D dece 19 (10) O27) V2: CLD) 0 0 S to M smoo 18 (7) (0.9) 1.4 (1.8) yi O51 ce) GLAU 49 (32) (0.6) 1.2 (1.8) 2 Hi 0) dryl 20 (15) (0.7) 1.2 (1.8) 2 1 (@) wetl 29 (17) (0.6) 1.2 (1.7) 2 iL O BRAC 11. 2C7) (1.1) 1.9 (3.1) 0 0 S GLOM 111 (63) (0.6) 1.3 (2.5) O (2) 0,1 O to D GLOM 25 (10) (0.6) 1.0 (1.6) 0 0 (0) S to D HIRS 25 (16) (1.0). divG- (230) 0 0 (O) S to D GLAP 22 (13) (1.2) 1.6 (2.2) 2 1 O (S) PUMI 39 (24) (0.2) 1.4 (2.5) 0 0 O to D robu 24 (14) (0.2) 1.4 (2.5) 0 0 (0) S to D sout 15 (10) (0.8) 1.2 (1.5) 0 0 0 to § in separating the old-field and smooth variants of A. virginicus var. virginicus from the rest of the virginicus complex. The racemes of the drylands variant of A. virginicus var. glaucus, the deceptive variant of A. virginicus var. virgin- icus, A. brachystachyus, and A. glomeratus var. glaucopsis are shorter than those of other taxa. The remaining characters pertain to structures of the dispersal unit. The rachis internodes of Andropogon floridanus are uniformly pubescent for their full length and do not become glabrous toward the base, as in other taxa. The 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 193 TaBLe 4. Character states (continued). 5 6 7 8 Sheath Sree Blade pubescence scab- Taxon orientation rousness length (cm.) width (mm.) ARCT A,S 0 (15) 21 (35) (3.0) 4.5 (8.0) GYRA S (A) 0 (6) 18 (48) (0.8) 2.2 (5.0) GYRA S (A) 0 (8) 20 (48) Cis4)) 2.3 +@:-0) comm S (A) 0 (8) 20 (48) (G2. 6: (C5:.0) tenu S 0 (11) 21 (30) le35)942". 0. (330) STEN S 0 (6) 11 (28) (0.8) 1.8 (3.0) TRAC S 0 (10) 16 (22) (1.2) 1.9 (2.6) LIEB Ss 0 (1) (3) 16 (35) CRN ast “CF .3) LIEB 5 0 (1) (3) 8 (15) 6255) 422° (6.5) PUNG S 0 (15) 24 (35) (2.5) 4.6 (7.5) FLOR iS) 0,1 (32) 42 (61) (2.9) 3.8 (5.0) LONG A (S) 0 (11) 26 (50) (20 82 Cae ay VIRG S 0 (1) (11) 25 (52) (1.7) 3.6 (6.5) VIRG S 0 (1) (11) 28 (52) (17)° 3.3 (535) oldf S 0 (1) (11) 30 (52) (1.7) 3.0 (5.0) dece Ss 0 (15) 24 (35) (245), 3.6: (5.5) smoo - 0) (22) 27 (38) (25-3023 .5-(5..0:) GLAU Fa 0) (12) 19 (38) (220): 43° C65) dryl - 0 (12) 17 (27) (220) 3.5:.C5.0) wetl - 0 (13) 20 (38) (2.6) 4.9 (6.5) BRAC S 0 (21) 33 (54) (2.3) 3.7 (6.0) GLOM S (A) 1 (0) (13) 41 (109) (2.9) 4.8 (9.5) GLOM S 1 (16) 37 (55) (2.8) 4.3 (7.5) HIRS S 1 (0) (24) 41 (60) (2.0) 4.2 (6.0) GLAP S 0 (1) (33) 40 (75) (3.0) 5.0 (7.0) PUMIL S (A) 0,1 (13) 44 (109) (37.0) 5-4 209.5) robu S (A) 0,1 (13) 46 (109) (3.0) 6.0 (9.5) sout Ss 1 (30) 41 (66) (3.5) 4.4 (6.0) length of the callus hairs is positively correlated with the general density of hairs on all parts of the dispersal unit. Andropogon gyrans and A. longiberbis, with callus hairs often to 5 mm long, have more densely pubescent rachis internodes and pedicels than all other taxa except 4. floridanus. The keels of the lower glumes of the spikelets bear prickles that often extend below the middle of the glume in 4. arctatus and A. glomeratus var. pumilus. The awn of the fertile lemma is more or less twisted where it joins the lemma and 1s relatively short in A. arctatus, A. floridanus, and A. brachystachyus. Variation 194 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 TasLe 4, Character states (continued). 9 10 1l Ligule ciliation Taxon length (mm.) length (mn.) color ARCT (0.3) 0.6 (0.9) 0-0.5 Brown GYRA (0.3) 0.9 (1.5) 0-0.7 Light brown (brown) GYRA (0.3) 0.6 (1.1) 0-0.3 Light brown to brown comm (0.3) 0.6 (1.1) 0-0.7 Light brown tenu (0.4) 0.6 (0.8) 0-0.3 Light brown to brown STEN (0.8) 1.1 (1.5) 0-0.1 Light brown TRAC (0.2) 0.4 (0.5) 0.2-0.8 Brown LIEB (0.5) 0.8 (1.2) 0-0.4 Whitish to light brown LIEB (0.5) 0.7 (1.9) 0-0.2 Whitish to light brown PUNG (O59) 0:29. (C2) 0-0.4 Whitish to light brown FLOR (0.4) 0.7 (1.2) 0.2-1.3 Brown LONG (0.2) 0.4 (0.6) 0.3-0.6 Light brown to brcwn VIRG (0.2) 0.5 (1.0) 022-13 Brown (light brown) VIRG (0.2) 0.5 (1.0) 0.2-1.3 Brown (light brown) oldf (0.2) 0.5 (0.8) O.2-1.3 Brown dece (0.3) 0.5 (0.7) 0.5-1.1 Brown smoo (9.4) 0.6 (1.0) 0.3-0.9 Brown (light brown) GLAU (0.2) 0.4 (0.5) 0.3-1.2 Brown dryl (0.2) 0.3 (0.5) 0.4-1.2 Brown wetl (0.2) 0.4 (0.5) 0.3-0.8 Brown BRAC (0.2) 0.4 (0.5) 0.6-1.5 Brown GLOM (0.6) 1.2 (2.2) 0-0.9 Whitish (brown) GLOM (1.0) 1.2 (2.0) 0-0.3 Whitish to light brown HIRS (0.7) 1.2 (2.0) 0-0.3 Whitish to light brown GLAP (0.9) 1.5 (2.0) 0-0.2 Whitish to light brown PUMI (0.6) 1.1 (2.2) 0.2-0.9 Light brown to brown robu (0.6) 0.8 (1.3) 0.2-0.9 Light brown to brown sout (1.0) 1.5 (2.2) 0.2-0.5 Light brown in anther length and the marcescence of the anther are discussed in the following section. The fruits are oblong to linear and are usually 2-3 mm long. Their size varies sufficiently with environmental conditions so as to be of little taxonomic value. CHARACTER STATES The measurements in TABLE 4 are based on herbarium specimens (primarily those of the author) and field observations covering the taxonomically useful 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 195 TasLe 4. Character states (continued). ale: 13 14 LS 16 ; Inflorescence orientation maximum maximum Taxon shape of branches branching branches units/stem ARCT Oblong to ovate Erect 1-3 1-5 (5) 16 (45) GYRA Linear to ovate Erect 1-2 1-5 (2) 9 (31) GYRA Linear to ovate Erect 1-2 1-5 (3) 9 (31) comm Linear to ovate Erect 1-2 2-5 (3) 12 (31) tenu Linear Erect 1-2 1-3 (3) 8 (13) STEN Linear Erect 1-2 1-3 (2) 11 (26) TRAC Linear Erect 1-2 1-2 (3) 8 (11) LIEB Linear to oblong Erect 1-3 2-3 (1) 16 (50) LIEB Linear Erect 1 (2) 1-2 (1) 4 (7) PUNG Linear to oblong Erect 2-3 2-3 (7) 27 (50) FLOR Obla to obovate Erect 2-3 2-5 (9) 51 (210) LONG Linear to oblong Erect 2-3 2-4 (7) 45 (97) VIRG Linear to obpy' Erect or arching 2-4 2-6 (6) 62 (195) VIRG Linear to oblong Erect 2-4 2-5 (6) 55 (175) oldf Linear to oblong Erect 2-3 2-5 (6) 51 (150) dece Linear to oblong Erect (arching) 2-4 2-4 (20) 82 (175) smoo Linear to obla Erect 2-3 2-3 (12) 28 (60) GLAU Linear to obpy Erect or arching 2-3 2-6 (19) 78 (190) dryl Linear to obpy Erect or arching 2-3 2-3 (31) 84 (290) wetl Linear to oblong Erect 2-3 2-6 (19) 72 (113) BRAC Ovate to obpy Arching 2-3 3-5 (12) 75 (190) GLOM (Linear to) obpy Erect 2-5 3-11 (10) 123 (600) GLOM Oblo to obpy Erect 2-4 3-4 (25) 107 (200) HIRS (Linear to) oblong Erect 2-4 4-5 (10) 96 (205) GLAP (Linear to) oblong Erect 3-4 4-5 (30) 159 (400) PUMI Obla to obpy Erect 2-5 3-11 (15) 132 (600) robu Obla to obpy Erect 3-5 3-11 (28) 169 (600) sout Obla to obpy Erect 2-3 3-7 (15) 95 (200) morphological variation and the geographic range of the taxa. The sample size reflects the complexity of the taxa or (with Andropogon liebmannii var. lieb- mannii and the tenuous variant of A. gyrans var. gyrans) a low frequency of occurrence in nature. The first number for sample size in TABLE 4 is the total number of plants measured; the number of populations included in the mea- surements is in parentheses. For some taxa many plants have been measured for certain critical characters and fewer plants for other characters less impor- tant in distinguishing the taxa. Also, numerous herbarium specimens have been spot checked for many of the characters. For all quantitative measurements, 196 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 TABLE 4. Character states (continued). 17 18 19 20 Pube nce below raceme —— Raceme Sheath = Taxon sheath location length (cm.) width (mm.) ARCT O to D 0 (3.3) 3.8-6.8 (9.0) (2.5) 3.2-4.0 (5.0) GYRA 0 to D 0,1 (3.5) 4.6-6.9 (13.5) (1.5) 2.7-4.7 (8.0) GYRA O to D 0,1 (3.5) 4.6-7.0 (13.5) (1.5) 2.8-4.7 (8.0) comm (O to) D 1 (0) (3.5) 4.4-5.5 (13.5) (2.0) 3.1-5.1 (8.0) tenu S to D 0 (4.0) 4.9-7.9 (9.5) (1.5) 2.5-4.3 (5.0) STEN 0 to D 0 (2.6) 4.2-6.6 (8.5) (2.0) 2.8-4.8 (6.3) TRAC O to D 0 (2.8) 4.1-5.8 (7.2) (3.0) 4.0-4.7 (5.8) LIEB O to D 0 (4.0) 4.9-7.4 (10.0) (3.0) 4.2-6.1 (10.1) LIEB O to S 0 (4.2) 4.7-7.0 (7.5) (3.0) 3.2-4.2 (4.5) PUNG M to D 0 (4.0) 5.1-7.8 (10.0) (4.1) 5.3-8.0 (10.1) FLOR O to D 0 (3.0) 4.0-5.9 (7.0) (1.5) 2.0-2.7 (3.6) LONG S to D 0 (2.5) 3.0-4.5 (6.0) (2.5) 3.2-4.1 (5.5) VIRG O to D 0 (2.1) 3.1-4.6 (6.7) (1.7) 3.0-3.8 (5.6) VIRG O to M 0 (2.2) 3.2-4.8 (6.7) (1.7) 3.0-3.8 (5.6) oldf O to S (M) 0) (2.3) 3.4-5.2 (6.7) (2.2) 3.3-4.4 (5.6) dece (0) S to M 0 (2.2) 2.5-3.8 (4.5) (1.7) 24-31, (4.0) smoo 0 (S) 0 (2.8) 3.3-4.7 (6.7) (3.0) 3.2-3.8 (5.2) GLAU oO to D 0 (2.1) 2.9-4.3 (6.0) (2.7) 3.1-3.8 (5.5) dryl 0) 0 (2.1) 2.6-3.8 (4.9) (2.7) 3.0-3.5 (4.5) wet] S to D 0 (2.4) 3.2-4.8 (6.0) (2.7) 3.2-4.2 (5.5) BRAC (O to) M ) (2.1) 2.4-3.5 (4.1) (2.3) 2.6-3.0 (3.8) GLOM S to D 0 (2.0) 2.9-4.4 (6.5) (1.3) 2.3-3.1 (4.7) GLOM S to D 0 (2.5) 3.4-5.2 (6.5) (2.0) 2.5-3.4 (4.7) HIRS (S) M to D 0 (2.5) 2.9-4.6 (5.7) (2.0) 2,4-3.1 (4.0) GLAP (S) M to D 0 (2.0) 2.4-3.6 (4.4) (1.3) 2.0-2.5 (3.0) PUMI S to D 0 (2.0) 2.9-4.4 (6.3) (1.5) 2.1-2.9 (4.4) robu (S) M to D 0 (2.0) 2.9-4.3 (5.2) (1.5) 2.0-2.5 (3.0) sout S to M ¢) (2.3) 2.9-4.5 (6.3) (125). -2.5353..3: (4.4) if two numbers are provided (characters 10, 14, 15, 22, 26, and 27) they represent the range of values recorded. Where three numbers are given (char- acters 1, 7-9, 16, 29, and 31), the first and last, which are enclosed in paren- theses, indicate the range around the mean value. For four numbers (characters 19-21, 23, and 25) the first and last again indicate the range, while the two middle values represent the mean of the smallest and largest values recorded for all the plants. For all other characters, states enclosed in parentheses rarely occur in the taxon. If two states are given, they are both of intermediate 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX Lo7 Tape 4. Character states (continued). ; 22 og 7 eduncle = Raceme Taxon length (mm. ) number length (cm.) ARCT (9) 26-66 (115) 2 (3, 4) (2592) 226-45 345.3) GYRA (1) 5-31 (195) 2-5 (1.5) 2.8-4.2 (5.5) GYRA (1) 5-42 (195) 255 Cle5) 25653729 652.5) comm (4) 6-60 (195) 2-5 (145)27-5 9777-0.) tenu (1) 5-25 (50) 2-4 (243942. 7-482 (525) STEN (3) 4-16 (45) 2-363.) (2.5) 4.0-4.5 (6.0) TRAC (9) 14-31 (65) 2 (1.5) 2.4-3.6 (4.2) LIEB (10) 24-68 (130) 2-13 (2.0) 2.4-4.0 (5.0) LIEB (20) 32-88 (130) 2-9 (2.0) 2.5-4.1 (5.0) PUNG (10) 15-47 (70) 2-13 (2.0) 2.3-3.9 (4.7) FLOR (10) 19-48 (93) 2 (-4) (2.0) 2.5-3.7 (4.5) LONG (1) 3-4 (13) 2 (3) (1.3) 1.8-2.6 (4.0) VIRG (1) 4-6 (30) 2-7 (0.5) 1.7-2.8 (4.4) VIRG (1) 4-6 (30) 2-7 (0.5) 1.7-3.0 (4.4) oldf (2) 4-6 (12) 2-5 (-7) (0.5) 1.9-3.3 (4.4) dece (1) 4-9 (30) 2 (3) (1.3) 1.5-2.3 (3.0) smoo (2) 3-4 (6) 2-4 (-7) (1.1) 1.5-2.9 (3.6) GLAU (2) 3-4 (10) 2 (3) (1.4) 1.8-2.7 (4.0) dryl (2) 3-4 (5) 2 (1.4) 1.7-2.4 (3.2) wetl (2) 3-5 (10) 2 (3) (1.5) 2.0-3.0 (4.0) BRAC (13) 20-31 (43) 2 (3) (1.2) 1.5-2.1 (2.6) GLOM (1) 6-14 (60) 2 (-4) CLe0)E 15722535) GLOM (4) 11-35 (60) 2 (-4) (1.5) 2.1-2.9 (3.5) HIRS (2) 3-5 (8) 2 (1.3) 1.7-2.8 (3.3) GLAP (1) 2-4 (6) 2 (1.0) 1.3-2.0 (2.3) PUMI (2) 7-13 (40) 2 (-4) (1.3) 1.8-2.4 (3.0) robu (2) 8-15 (40) 2 (23): 7245-20) sout (2) 5-10 (16) 2 (-4) (1.7) 1.9-2.3 (2.8) frequency (or both rare if in parentheses). For glaucousness (characters 2 and 3) “O’? means absence and “1” presence. For stem glaucousness “1” further means that the glaucousness does not occur in the internode below about 2 cm under the node, and “2” indicates general occurrence of glaucousness in the internode. Absence or presence of sheath scabrousness (character 6) is shown with a “0” or a “1.” In the pubescence-density characters (4 and 17) “0” is absence, ‘‘S” sparse, ““M’”’ moderate, and “D” dense. Pubescence orientation (character 5) is either appressed (“A”), spreading (“S”), or a combination of the two: when the leaves are always glabrous, this character is left blank. 198 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 TABLE 4. Character states (continued). 24 25 26 Rachis internode SS plkelet Taxon pubescence ieee (mm. ) width (mm.) ARCT 0,1 (4.3) 4 5.4 (6.1) 0.7-1.0 GYRA 0 (3.0) 3.9-4.7 (5.7) 0.5-0.7 GYRA 0 (3.5) 4.1-4.5 (5.7) 0.5-0.7 comm 0 (3.5) 4.2-4.5 (5.7) 0.5-0.7 tenu 0) (3.5) 4.0-4.5 (5.2) 0.5-0.6 STEN 0 (3.0) 3.4-5.1 (5.5) 0.5-0.6 TRAC 0 (4.0) 4.8-5.0 (5.5) 0.6-0.8 LIEB 0 (3.0) 4.0-4.5 (6.0) 0.5-1.0 LIEB ) (3.0) 3.4-3.9 (4.2) 0.5-0.9 PUNG 0 (4.3) 4.7-5.1 (6.9) 0.7-1.0 FLOR ue (3.8) 4.4-4.8 (5.5) 0.5-0.7 LONG 0 (3.5) 4.1-4.5 (5.0) 0.5-0.8 VIRG 0 (2.6) 3.5-3.8 (4.7) 0.4-0.6 VIRG 0 (2.9) 3.5-3.8 (4.7) 0.4-0.6 oldf 0 (2.9) 3.7-3.9 (4.7) 0.4-0.6 dece 0 (3.0) 3.3-3.6 (4.0) 0.4-0.6 smoo 0 (3.0) 3.5-3.7 (4.2) 0.4-0.6 GLAU 0 (2.6) 3.4-3.7 (4.4) 0.4-0.5 dryl 0 (2.6) 3.2-3.5 (3.9) 0.4-0.5 wetl 0 (3.0) 3.5-3.9 (4.9) 0.4-0.5 BRAC 0 (4.1) 4.4-4.6 (5.0) 0.5-0.7 GLOM 0 (3.0) 3.7-4.0 (5.0) 0.4-0.6 GLOM 0 (3.8) 4.1-4.4 (5.0) 0.5-0.6 HIRS 0 (3.4) 3.6-3.8 (4.6) 0.4-0.6 GLAP 0 (3.0) 3.2-3.5 (3.8) 0.4-0.6 PUMI 0 (3.0) 3.8-4.1 (5.0) 0.4-0.6 robu 0 (3.0) 3.4-3.8 (4.5) 0.4-0.6 sout ¢) (3.5) 4.2-4.5 (5.0) 0.4-0.6 serena shape (character 12) is linear, oblanceolate (abbreviated “‘obla’’), oblong, ovate, obovate, or obpyramidal (abbreviated “‘obpy’’). Raceme sheaths are mee (character 18) so as to be either largely exposed (‘‘0’’) or mostly hidden within enlarged stem sheaths (‘‘1’’) after anthesis. Pubescence on the rachis internode (character 24) is either relatively sparse toward the base of the internode (“‘0”’) or uniformly distributed along the internode (“‘1”’). Lower glume scabrousness (character 28) either occurs only above the middle of the glume (*‘O”’) or extends below the middle (“‘1’’). Percent chasmogamy (character 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 199 TabLe 4. Character states (continued). a 27 28 29 30 ximum callus hair glume keel Sw Taxon length (mm.) scabrousness length (mm.) base ARCT 1.5-2.5 1 (0.5) 0.8 (1.6) (0) 1 GYRA 15, 0 (0.8) 1.8 (2.4) 0,1 GYRA 2-5 0 (223) 52 6222) 0,1 comm 2.5-4 0 (1.53) 2128: -@53) (0) 1 tenu 2-5 0 (123) 225506 (2:24) (0) 1 STEN 1-2.5 0 (0.8) 1.5 (2.2) 0,1 TRAC 19-35 0 C1 1D) 56223) 0,1 LIEB p>? 0,1 (1.2) 1.8 (2.4) 1 LIEB Teo=2 0 (1.2) 1.4 (1.6) 1 PUNG L.5=2 0,1 (1.7) 2.1 (2.4) 1 FLOR 1-3 0 (0.5) 0.9 (1.5) 0 LONG P.5=5 0 (1.0) 1.6 (2.1) 0,1 VIRG 1-3 0 (1) (0.6) 1.4 (2.1) 0 (1) VIRG 1-2.5 0 (1) (0.8) 1.4 (2.0) 0 oldf 12\.5 O (1) (1.0) 1.5 (2.0) 0 dece 1-2.5 0 (0.8) 1.2 (1.7) 0 smoo 1-1.5 0 (1) (1.2) 1.5 (1.8) 0 GLAU 1-3 0) (0.6) 1.2 (2.1) 0 (1) dryl 1-3 0 (0.6) 1.1 (1.5) 0 (1) wetl 1-2.5 0 (0.9) 1.4 (2.1) On) BRAC 1-1.5 0 (0.2) 0.7 (1.1) 0 GLOM L=2:25 0,1 (0.6) 1.3 (1.9) O (1) GLOM Lome 0 (1) (0.8) 1.2 (1.8) 1 HIRS 1-2 0 (1) (1.0) 1.4 (1.7) (0) 1 GLAP das O (1) (0.9) 1.2 (1.6) (0) 1 PUMI T3245 ae (0.6) 1.2 (1.9) (0) 1 robu 1.5-2.5 al (0.9) 1.4 (1.9) (0) 1 sout 1-2 i (0.6) 1.1 (1.7) 1 33), given as the mean (one standard error), is determined by looking for the presence (cleistogamy) or absence (chasmogamy) of a marcescent anther and stigmas in at least 30 postanthesis spikelets per plant. TAXONOMIC TREATMENT Taxonomic rank within the complex is based primarily on morphological distance between the taxa. Distances between the taxa have been determined 200 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 TABLE 4. Character states (continued). 31 32 33 Anther Mean percent Taxon length (mm.) color chasmogamy + S.E ARCT 20) 255-0.) Red TOO! <0 GYRA (0.6) 1.1 (1.4) Yellow or purple 38 t 3.3 GYRA (0.7) 1.1 11.4) Yellow or purple 36% 4.7 comm (0.9) 1.2 (1.4) Yellow (purple) 38 + 6.1 tenu (0.7) 0.9 (1.2) Yellow or purple 34 5.2 STEN (0.6) 1.2 (1.7) Yellow 40 t 6.5 TRAC (12) 5. 12) Yellow 98 0,2 LIEB (0.7) 1.2 (1.4) Yellow 95 $1.2 LIEB (1.1) 1.2 (1.4) Yellow 98 t 0.7 PUNG (0.7) 1.1 (1.4) Yellow 91 t 2,2 FLOR (1.3) 1.5 (2.0) Yellow (purple) 98 + 0.7 LONG (0.9) 1.2 (1.6) Yellow 84 + 3.4 VIRG (0.6) 0.9 (1.5) Yellow or purple 64 + 2.8 VIRG (0.6) 0.8 (1.5) Yellow or purple 60 + 2.9 oldf (0.6) 0.9 (1.2) Yellow or purple 4745.9 dece (0.6) 1.0 (1.5) Yellow 88 + 3.6 smoo (0.6) 0.7 (1.0) Purple 44 + 4.8 GLAU (0.7) 1.0 (1.2) Yellow 7o t1.5 dryl (0.9) 1.0 (1.5) Yellow 97 + 0.9 wetl (0.7) 0.9 (1.1) Yellow 42 t 2.3 BRAC (1.7) 2.1 (2.4) Red 99 + 0.01 GLOM (0.5) 0.9 (1.5) Yellow, red, or purple 66 + 1.6 GLOM (0.9) 1.2 (1.5) Yellow 79 t 3.4 HIRS (0.7) 0.8 (1.1) Yellow or purple 1a Uae Hee GLAP (0.5) 0.7 (0.9) Yellow or purple 25 + 2.9 PUML (0.7) 0.9 (1.1) Yellow 93 + 0.9 robu (0.7) 0.9 (1.1) Yellow 96 t 0.6 sout (0.8) 0.9 (1.0) == 88 $1.9 through comparisons of the 33 characters (see TABLE 4) used in this study, ssigning a value of * a “1” for O” for no difference, ‘*2”’ for intermediate overlap, and little or no overlap between the character states. (The rules for deter- mining these values for each of the characters are presented in APPENDIX B.) The sum of these values for all the characters for any single pair of taxa is the distance between them (TABLES 5, 6). This distance represents the probability of misidentifying a specimen of one taxon as another taxon. The ma distance between any pair of taxa 1s 21.5 (Andropogon tracyi and A. glomeratus var. glaucopsis), and the minimum between any of the lowest-ranking taxa 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 201 Tas_e 5. Morphological distance between the species.* - Pubescence below raceme sheaths absent; raceme sheaths (2.1-)2.6-3.8 (-4.9) cm long; racemes (1.4-)1.7-2.4(-3.2) cm long; spikelets (2.6-)3.4-3.7 (-4.4) mm long; plants of sae drained soils on Coastal Plain from southern Virginia to southern Alabama. ........... 006. e beeen eee eens b. |. Drylands variant of A. virginicus var. glaucus. Pubescence below raceme sheath sparse to dense; raceme sheaths (2.4—)3.2- 4.8(-6) cm long; racemes (1.5—)2-3(-4) cm long; spikelets (3-)3.5-3.9(-4.9) mm long; plants of a drained soils on Coastal Plain from southern New Jersey to eastern Texas. 2.0.0... sc cece ee enter enn ees 7b. 2. Wetlands variant of A. virginicus var. glaucus 2: ei green (sometimes siehiily glaucous in 4. gyrans var. stenophyllus and mooth variant of 4. virginicus var. virginicus). 5. Stem sheaths scabrous with upwardly directed prickle hairs; leaf blades usually more than 35 cm long. 6. Ligules usually less than | mm long; raceme sheaths lr less than 2.5 mm wide: keels of lower glume often scabrous below middle. .......... 9d. 1. Robust variant of A. ee var. pumilus. 204 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 6. pee usually more than | mm long; raceme sheaths mostly more than wide; keels of lower glume glabrous or scabrous below middle only in eh ae variant of A. glomeratus var. pumilus. Keels of eg glume rarely scabrous below middle; plants of eastern United State pene (linear to) oblong; spikelets usually less than 4 mm long; anthers usually marcescent within spikelets; peduncles less than 10 mm long; plants of Coastal Plain ay southern Virginia to south- ern MissisSippl. ................ b. 4. glomeratus var. hirsutior. inflorescences oblong to obpyramidal: spikelets usually more than — oo [oe] — Massachusetts to southern oe on Coastal Plain and wel MlaNdy ci osceecs aoe way bes 4. glomeratus var. glomeratus. 7. Keels of lower glume usually Here Belew middle; plants of south- western United States. 2.00.00. 0000 eee ee iat axe 9d. 2. Southwestern variant of 4. glomeratus var. pumilus. 5. Stem sheaths smooth, very rarely scabrous; leaf blades usually less than 35 cm long (except in robust variant of A. glomeratus var. pumilus). 9. Inflorescence units sale hidden within broadened, strongly overlapping uppermost stem sheaths. .......0.0.0000 000000 cece eeey noni 1. Common ae gvrans var. gyrans. Inflorescence units mostly exposed at maturi 10. Leaves glabrous. 11. Ligules (0.8-)1.1(-1.5) mm long, with ciliations is than ae mm 2 a basal leaves often filiform, less than 1.5 mm wide, s dyed Ga auteteee todas 20. Ube stenpht Ll. Liles less than 0.7 mm long (rarely to m, but usually longer n southwestern variant of J. eas var. peeing with cil- on usually greater than 0.1 mm long; basal leaves usually more wide, soon arching. 12. Stem int infl peciall at stem and branch apices) with 3 or more racemes. 13. Stems less than | m long; inflorescence units less than 14 per stem; spikelets usually more than 4 mm long 2a. enuous variant of 4. gyrans var. gyrans. 13. Stems more > than | m long; inflorescence units usually more than 14 per stem; spikelets usually less than 4 mm ] . Smooth variant of A. virginicus var. virginicus. 2. Stem internodes green, glaucous just below node only: inflorescence units aD racemes (infrequently more). 14. Ligules more than | mm long; keels of lower glume often scabrous to below middle; spikelets usually greater than 4 mm long; plants of southwestern United States and northwestern Mexico. .....00..00..0 0000 c eee ee 9d. 2. Southwestern variant of A. glomeratus var. pumilus. 14. Ligules less than | mm long; keels of lower glume sca- brous only above middle; tae less than 4 mm long; plants of ee United Sta eo eee ae 7a. 2. Deceptive a of A. virginicus var. Virginicus. ~) g 10. i pubescent, at least on margin near collar. 15. Keels of lower ue often scabrous below middle; leaves usually more than 44 cm lo 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 205 16. Spikelets usually less than 4 mm long; ligules (0.6-)0.8(-1.3) ong less than 2.5 mm wide; wi ide- ranging plants of most of temperate North and Central Amer- ica except southwestern United States. .................. week 9d. 1. Robust variant of A. glomeratus var. De es 16. Spikelets usually longer than 4 mm; ligules (1-)1.5(-2.2) mm long; raceme sheaths usually more than 2.5 mm wide; oo of southwestern United States and northwestern Mex 9d. 2. Southwestern variant of . a umilus. 15. Keels of lower ee scabrous only above middle; ae usually less than 31 cm lo We. ae of more than 4 mm long: hairs on rachis internode and pedicel ense, long. 6 longiberbis. 17. Pubescence of young stem sheaths spreading; spikelets mostly less than 4 mm ane hairs on rachis internode and pedicel rather sparse, shor 18. Raceme pane (2.7-)3.3-4.4(-5.5) mm wide; racemes often more than 2 per inflorescence unit; widespread through most of eastern United States, absent from all but southernmost Florida. ...................00.05: Ja. 1. Old-field variant of A. virginicus vIrginicus. . Raceme sheaths (1.7-)2.4-3.1(-4) mm wide; racemes usually 2 per inflorescence unit; Coastal Plain from Flor- ida to southern Virginia, common only in Florida. .... 7a. 2. Deceptive variant of A. virginicus var. virginicus. ve } +t} > | iy | lat 1] — co 1. One or more postflowering peduncles more than 15 mm long 19. All raceme sheaths largely hidden before senescence ane broadened, strongly overlapping, uppermost stem ear stash ies Beal rep teense am aS a 8 Axa eB aise BE iiceaee bs wie gag gan eee ok: Common variant of 4. gyrans var. gyrans. Mostor all ths cl xposed after anthesis; uppermost stem sheaths reduced, not eae overla Spine, 20. Inflorescence units with (2 to) 4 to 6 (to 13) racemes. _ Stems less than 0.9 m tall: leaves less than 15 cm long; stem sheaths more or less glabrous (to densely areas spikelets less than 4.2 mm long; plants of Mexico. .......... 4a. A. liebmannii var. liebmannit. . Stems more than 0.8 m tall; leaves more fan 15 cm long; stem sheaths usually densely woolly-pubescent, at least near collar; spikelets more than 4.2 mim lone; plants of United States. jut fee oe A. liebmannii var. pungensis. AG NO 20. Se units with 2 (or more) race 2 m sheaths scabrous with upwardly Aiea prickle hair a Rachis internodes becoming sparsely pubescent pe ibase: stem a usually more or less pubescent; plants of moist or disturbed an ae (1-)1.2(-2) mm long, with ciliations less than 0.3 mm long; keels es oe glume scabrous only above middle. Heaney deg ena aay seen, sek rs a. A. glomeratus var. ease 24. Ligules (0.6-)0.8(-1.3) mm long, with ciliations 0.2-0.9 m long; keels aes glume often scabrous below middle. .... caunasiaee 9d. 1. Robust variant of A. pats var. pumilus. 23. Rachis ae dances! and uniformly pubescent, not becoming JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 sparsely esaee toward base; stem sheaths ve pubescent plants epee ands ee ee Te ee ene rem re ee 22. Stem sheaths sm 2 5. Many or all eee longer than raceme sheaths at maturity, racemes then fully exserted above apex of raceme sheath 26. Racemes usually less than 2.1(-2.6) cm long; ca foneecence branches long and arching, aun open , lo - Re ee ee ee eee ne ee een ee eet 4. br ach rstac Ayus. 26. Racemes exceeding 2 cm in length, Aone more than 2.5 ¢c Hee inflorescence branches erect, inflorescences more or less dens 27. ane less than 0. 9 m tall; leaves less than 2 mm wide: fully included within raceme sheaths at maturity on peduncles less than 10 mm long; anthers often mar event iy GENE ee AA SE ease es eta 2a. 2. Tenuous variant of A. gyrans var. gyrans. ; Stems asually more en l m tall, leaves mostly more ae No — wide; ed at maturity, eirets aeely iesuhan 15 mm long; anthers rarely marce 28. Leaves ee blade mostly more than 35 cm long; stamen | per flower, anther less ae 2 mm lon avant te pe ives one det otiea ease meetin ftonee _ A. Won as 28. Leaves often pubescent, blade less . 35 cm long; stamens 3 per flower (if only 1, anther more than 2 mm long). 29. Stamen(s) mostly |; lower glumes more or less folded; awns less than | cm long. ............ Peeee ng Ae te eetuaes een ates ee A 1. A. arctatus. 29. Stamens 3; lower glumes flat; awns more than | cm lon 30. Lower glume nerved between keels (most eee seen on adaxial surface); endemic of Florida. ........ A. ternarius var. cabanisit. oa ume not nerved between keels; widespread through most of eastern United States. ......... A. ternarius var. ternarius. 25. Peduncles all shorter than raceme sheaths, at a bases of racemes not fully exserted above raceme sheath a : Stems less than 1.2 m tall; leaf blades a than 30 cm by 3 we 2S Ls) — 32. Raceme sheaths (2.2-)2.5-3.8( usually less than 4 mm long. ........................ see dae eae ee 7a. 2. Deceptive variant of A. virginicus var. vIrginicus. 32. Raceme sheaths usually more than 4 cm long; spikelets generally more than 4 mm long. 33. roe more than 0.8 mm long, with ciliations less n 0.1 mm long; ete of bogs and ditches. ..... shen tices og eden eer ee 4. gyrans var. stenophyllus. . Ligules less than 0.8 mm long, with ciliations often more than 0.2 mm long; plants of well-drained soils. 34. Spikelets (4-)4.8-5(-5.5) by 0.6-0.8 mm: ra- cemes paired; anthers more than |.2 mm long. 4. 5). cm long: coeieis WfiaiR Es Min PAE RSS Uae BAe Saceipehiae Dede LLACVI. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 207 34. Spikelets (3.5-)4-4.5(-5.2) by 0.5-0.6 mm; ra- to 4 per inflorescence unit; anthers less than: 1.2: mnamilong gage renee er Sais ee ee ae 2a. 2. Tenuous variant of A. gyrans var. gyrans. 31. nae i more than 1.2 m tall; leaf blades often more han ee oa 3 mm; inflorescence units usually more than 5 per s 35. ae usually less than 2.1(—2.6) cm long; inflorescence branches long and arching, inflorescences open, loose Lisette tated ae a8 SPE 8. A. brachystach) pus. . Racemes often longer than 2 cm; inflorescence branches erect (infrequently somewhat spreading in deceptive vari- ant of A. virginicus var. virginicus), inflorescences more or less dense. 36. Rachis internodes densely and uniformly pubescent, not becoming sparsely hairy toward base; many pe- duncles regularly more than 30 mm long. ......... etd Havas ieee Ee Wiech ea em 5. A. floridanus. . Rachis internodes sparsely pubescent toward base; peduncles rarely more than 20 mm long. 37. Raceme sheaths (1.5—)2—2.5(-3) mm wide; keels Ww nN Ww io ee 9d. 1. Robust variant of 4. glomeratus var. pumilus. . Raceme sheaths (1.7—)2.4-3.1(-4) mm wide; keels ower glume scabrous only above middle; stems less than 1.7 m tall; leaves less than 35 cm by 5.5mm, ......... 7a. 1. Deceptive variant of A. virginicus var. VIrginicus. ios) ~ 1. Andropogon arctatus Chapman, Bot. Gaz. 3: 20. 1878. Based on Andropogon tetrastachyus Ell. var. distachyus Chapman, Fl. So. U.S. 581. 1860. Not Andropogon distachyus L., 1753. Type? Florida, scat 5.n. (lecto- type, Ny!). GURES 11, K; 34. Sorghum arctatum (Chapman) Kuntze, Rev. Gen. Pl. 2: 791. 1891. Leptopogon eee Ange) G. Roberty subvar. arctatus (Chapman) G. Roberty, Boissiera 9: 197. Diacnosis. Inflorescence units averaging 16 per stem; peduncles all long; spike- lets long and wide, the lower glume more or less concave at maturity, with keels scabrous to below middle; anthers long, red; flowers rarely, if ever, chas- mogamous DISTRIBUTION. Flatwoods, bogs, and scrublands. Florida (except extreme south and Suwannee River drainage) and southern Alabama (Map 5). REPRESENTATIVE SPECIMENS. United States. FLoripA: Bay, Godfrey 76774 (Fsu); Brevard, Fredholm 6110 (us); Calhoun, Woods s.n., 1954 (TENN); Charlotte, Lewis 107 (us); Clay, Swallen 5584 (us); Collier, Lakela 31145 (puKE); Escambia, Campbell 3944, Franklin, Hausman s.n., 1867 (Ny); Gulf, Silveus 6735-A (rAes); Highlands, Campbell 4118 (Gu); Hillsborough, Shuey 1507 (usr); Jackson, Godfrey et al. 76736 (Fsu); Liberty, Godfrey 2A Chapman specimen from ny, annotated in his script, is here designated the lectotype. 208 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 3 2 § 4 2 “a | 4 a FH 4 4 3 5 | 4 = “ 2 z u 3 Z e 4 BI 2 3 | | magasey FIGURES 29-35, 29-31, Andropogon gyrans var. gyrans, inflorescences: 29, common variant (Campbell 4143), from one population, showing variation from mostly cleistog- amous (left) to mostly chasmogamous (right) flowering; 30, common variant (Campbell 4095), senescent; 31, tenuous variant, showing plants with predominantly cleistogamous (left, Campbell 3746) and mostly chasmogamous (Campbell 3873) flowers. 32, 33, A. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 209 74577 ou NCU, vDB); Manatee, Simpson s.n., 1889 ie Osceola, Ray et al. 10490 (GH, Us, USF); Santa Rosa, Combs 486 (us); Walton, A. H. Curtiss 6924 (GH, MO, US); neon, ‘Combs 661 (us). ALABAMA: Covington, oe 44769 (GH, VDB). Andropogon arctatus is unique in the virginicus complex in four respects. First, it does not grow in the Suwannee straits, a corridor along the Suwannee River from the Okefenokee Swamp in southern Georgia to the Gulf of Mexico. The distribution of other members of the virginicus complex that are common in Florida includes the Suwannee straits. Second, it appears to have a life history different from that of the successful colonizing taxa of the virginicus complex. The six densest populations of this taxon observed in my field studies grew in recently burned areas. Three of these were clear-cut slash pine (Pinus e/liottii Engelmann) timberlands in north- western Florida, where fire is routinely used by the forestry industry in site preparation. One of these localities, in Liberty County, was observed by Dr. Robert K. Godfrey in 1976; the following year there were only a few individuals where there had previously been hundreds. The same marked reduction in the number of flowering plants occurred in the other two “pine fields” in the second year after site preparation. In one of these localities, numerous basal clumps of leaves of this species were found in the third year. It appears that Andropogon arctatus remains in this vegetative state until it is burned, then flowers abun- dantly before returning to a vegetative state. Restriction of flowering to a year or two following fire probably accounts in large part for the supposed rarity of this taxon. (Removal of the forest canopy by heavy winds or lumbering op- erations may, in certain circumstances, also stimulate flowering.) Although other andropogons are stimulated by fire (FiGuURE 26), the number of their flowering stems tends to increase greatly two and three years later. Also, the other andropogons infect a burned site with dispersal units from nearby pop- ulations rather than remaining in situ until fire occurs. Third, in contrast to the mostly cespitose habit of the rest of the virginicus complex, the base of the plant of Andropogon arctatus is a short, rather thick rhizom Fourth, the anthers of Andropogon arctatus are longer than those of any other taxon in the virginicus complex: the lower limit of the range of A. arctatus overlaps the upper limit of that of only A. brachystachyus. Although the flowers of A. arctatus usually have one stamen, out of 138 flowers from six individual plants in one population in Liberty County, Florida (Campbell 4060), 68 per- cent had one stamen, 29 percent two, and 3 percent three. In the virginicus complex as a whole, the frequency of two- and three-stamened flowers 1s less than 0.1 percent. In addition to stamen number and anther length, other features associated with chasmogamy—long peduncles and large spikelets— point to the relatively primitive position of 4. arctatus in the virginicus complex. In overall aspect Andropogon arctatus closely resembles A. ternarius, which is the only other common United States species in sect. LEPTOPOGON outside the virginicus complex. Andropogon ternarius, however, is an unlikely ancestor rans var. stenophyllus (Campbell 38 13): 32, two stems; 33, basal leaves. 34, 4. arctatus (Campbell 3944), inflorescence. 35, A. tracyi (Campbell 4100), stem, Scale = 15 ¢ 210 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 of A. arctatus because it 1s tetraploid. Andropogon ternarius var. cabanisii (Hackel) Fern. & Griscom (chromosome number unknown) differs from A. arctatus only in stamen number and in the size and intercarinal nerves of the lower glumes. Knowledge of the chromosome number of 4. fernarius var. cabanisii is necessary to clarify its relationship to A. arctatus. The closest relatives of Andropogon arctatus in the virginicus complex are probably 4. floridanus and the common variant of A. gyrans var. gyrans. All three taxa may produce very long peduncled racemes that are more or less densely pubescent with whitish hairs. 2 aga gyrans Ashe, Jour. Elisha Mitchell Sci. Soc. 15: 113. 1898. Neortype: North Carolina, Wake or Orange County, Ashe 2034, 1896 (woul). DiaGnosis. Stems short; leaves narrow; inflorescences little branched and bear- ing few inflorescence units; raceme sheaths long and wide; peduncles all short or short and long on same plant; racemes long; callus hairs long (except taxon 2b). DISTRIBUTION. Dry or wet soils. New Jersey to Missouri, south to Florida and Texas; rare in Central America and West Indies. This common and widespread taxon has long been known as Andropogon elliottii Chapman. Chapman published this name (FI. So. U. S. 581. 1860) as new for A. argenteus Ell. (Bot. S. Carolina Georgia 1: 148. 1816), a later homonym of A. argenteus DC. (1813). Elliott’s description and type (Scribner, 1901), however, clearly place this name 1n synonymy with A. ternarius Michaux (1803). The synonymy of A. ternarius is presented in APPENDIX D The next oldest name at the species level is Andropogon clandestinus Wood, a later homonym of A. clandestinus - (see Synonymy under taxon 2a below). The next available name is 4. gyra The specimen chosen as the coe matches Ashe’s protologue in most respects. The stems on this sheet fit his description in stem height, leaf-sheath glabrousness, and raceme number. The specimen label bears, in Ashe’s script, the following information: “Andropogon. On sandstone between [?] and Chapel Hill, N.C. Wake or Orange Co. No. 2034 Date Oct 1896 Coll. W.W. Ashe.” At the end of his protologue, Ashe stated “collected by the writer in pine woods in Durham county, N.C., Oct. 1896.” It should be noted that Wake and Orange counties lie on either side of Durham County. It may well be that this specimen was the basis for Ashe’s Andropogon gyrans and that the discrepancies of label and protologue are due to his carelessness. This specimen cannot, however, be definitely tied to Ashe’s name and so 1s designated as neotype The equivalence of Ashe’s specimen to the widespread taxon long known as Andropogon elliottii is questionable. The upper stem sheaths of this species are usually conspicuously overlapping and inflated, a distinctive and unique fea- ture; those of the neotype are remote and not inflated. Less striking characters (short stems, dense hairs at the base of the raceme sheaths, number of racemes per inflorescence unit, long callus hairs, and twisting of the bases of the awns), however, clearly equate the neotype with the widespread taxon. This specimen Maps 2-8. Distributions: 2, Andropogon doe brachystachyus, 4, A. Tames 5, A. arctatus, 6, A. liebmannii var. pungensis; 7, A. liebmannii var. liebmannii (central Mexico); 8, A. XT1Id WOO SNOINIDUIA NODOdOUGNY “TIAddNVO — [E861 IC I Pale. JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 and a few others from throughout the Coastal Plain of the southeastern United States are intermediate between the common and tenuous variants of 4. gyrans var. gyrans, they appear to be plants of the common variant in which over- lapping and inflation of the upper stem sheaths are not strongly expressed. There are two clusters of characters uniting the three taxa of this species. First, they are all small plants with small inflorescences, but the inflorescence unit and many of its parts are larger than those of most other taxa. This contrast of sizes provides a quick means of distinguishing these three taxa from de- pauperate individuals of Andropogon virginicus. Second, the other broomsedges have either all short peduncles, all long peduncles, or more or less continuous variation from short to long within a single plant. Variation is bimodal on an individual in the common variant of A. gyrans var. gyrans, or between indi- viduals in the tenuous variant and 4. gyrans var. stenophyllus, since some have only short peduncles and some mostly long peduncles. Occasionally, short- and long-peduncled plants of these two taxa occur in the same population, but more often they are separated in different populations. This bimodality reflects a variability that tends to correspond to ecological and geographic differences. Hackel’s subvar. stenophyllus is here made a variety of Andropogon gyrans for the first time. In addition to the characters uniting this taxon to the common and tenuous variants of var. gyrans discussed above, var. stenophyllus is sep- arated from these two variants by a distance of 5 (TABLE 6) Key to the Subspecific Taxa of Andropogon gyrans — : Tene pad less than 0.8 mm long; plants of well-drained soils, rarely of poorly drai 2. pee sheaths usually hidden ae inflated, strongly overlapping upper stem sheaths. cc . Common variant of A. gyrans var. gyrans. ee ise eee ena tee te Beene a. 2. Tenuous variant of A. gyrans var. gyrans. Ligules (0.8—-)1.1(-1.5) mm long; oe of bogs and ditches. .................... 2b a" 2a. Andropogon gyrans Ashe var. gyrans Andropogon clandestinus Wood, Class-book Bot. 809. 1861, ex char. (as clandestina). Not Andropogon clandestinus Nees, 1854. Type: unknown. Andropogon elliotti Chapman var. age Hackel in DC. Monogr. Phanerog. 6: 415. mae Typ ae Florida, Duval Cou A. H. Curtiss 3636a moles, not seen; 1s0- ! Fsu!, GA!, GH!, ce ver MO (two sheets)!, Ncu!, us!). Andropogon hae eg ne, Nash in Small, FI. SE. U.S. 63. 1903. Andropogon elliottii Chap- man f. gracilior (Hackel) Blomq. Grasses N. Carolina, 203. 1948. Andropogon campyloracheus Nash, ies New York Bot. a. 1: 431. 1900. Based on Andropogon elliottii Chapman var. laxiflorus Scribner, Bull. Torrey Bot. Club 23: 146. 1896. Noi ied 1854. Type: Florida, Lake County, Nash 1738, 1894 (holotype, us!; isotypes, GH!, MicH!). Anatherum virginicum (L.) subvar. /axiflorum (Scribner) G. Roberty, Boissiera 9: ae 1960. i subtenuis Nash in Small, Fl. SE. U. S. 63, 64. 1903. Type: Mississippi, Harr ounty, Tracy 2243, 1896 (holotype, NyY!). Feri pene virginicus L. var. graciliformis Leon, Bull. Torrey Bot. Club 53: 457. 1926. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX og ey Maps 9-11. Distribution of Andropogon gyrans: 9, var. gyrans, common variant (also outside United States; see representative specimens); 10, var. gyrans, tenuous vari- ant: 11, var. stenophyllus. (In 10 and 11 dots represent populations of plants with short peduncles; stars, populations of plants with long peduncles.) 214 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Type: Cuba, Santa Clara, Sabana de Montembo, Ledn & Loustalot 11343 (holotype, Colegio de la Salle, Vedado, Havana, not seen; rye GH Andropogon elliottii Chapman var. projectus Fern. & Griscom, Rhodora 37: 138, 139. 1935. Type: North Carolina, Buncombe County, ee 1421°, 1898 (holotype, GH!; on ncu!, Ny!, Us (two sheets)!). DiaGnosis. Ligules short. DISTRIBUTION. Same as the species. 2a. |. Common variant Ficures 11, A, B; 29; 30. DiaGnosis. Raceme sheaths hidden within overlapping and more or less in- flated upper stem sheaths DisTRIBUTION. Generally dry, sandy soil of roadsides, embankments, fields, pine or oak woods; occasionally in moister soil. New Jersey to Missouri, south to Florida and Texas; Mexico (Beetle, 1977); Belize; Honduras; Cuba (Map 9). REPRESENTATIVE SPECIMENS. United States. New Jersey: Atlantic, Long 14026 (Gu); Camden, Long 15430 (GH); Cape May, Mackenzie 6695 (DUKE, MICH, MO, NY); Cum- erat, Long 29696 (GH); Gloucester, pene n., 1894 (GH); Salem, Long s.n., 1909 Ny). PENNSYLVANIA: Bucks, Long 65639 (Gu); Chester, Pennell 8889 (ny); Delaware, ee 1383 (Gu); Seige ee py (GH); Sa ae. Crosby s.n., 1942 (GH). DELAWARE: New Castle, Long 57766 (Gu); Sussex, Churchill s.n., s.d. (MO). MARYLAND: Baltimore, Taylor s.n., 1890 (micu); Caroline, Wilkens 5057-a fe u); Cecil, Tatnall 4411 (GH); Montgomery, Hitchcock 251 (GH, Mo, NY, US); Prince George’s, Blake 9733 (GH); Queen Anne’s, Commons s.n., 1870 (Ny). District oF CoLuMBIA: Steele s.n., 1896 (DUKE, GH, NY). West VIRGINIA: Cabell, Millender s.n., 1939 (Ny); Jackson, Richardson s.n., 1955 (Ncu). VirRGINIA: Arlington, A//ard 91a (DUKE, GH, MO, NY); Bedford, 4. H. Curtiss 9868 (mo); Brunswick, Fernald & Lewis 14475 (Gu); Culpepper, Ahles & James 61644 (Ncu),; Dinwiddie, Swallen 5527 (us); Fairfax, Fosberg 30198 (GH); Fauquier, Allard 389] (Gu); Greensville, Fernald & Long 9250 (Gu), Lunenburg, AAles & James 61944 (Ncu), Nansemond, Fernald & Long 6761 (GH, Mo, NY); Norfolk, Kearney 2383 (us); Northampton, Fernald et al. 5786 oe Prince a Ahles & James 62685 (NCU); Prince Wilham, Hermann 9911 (GH, MO, NY). NoRTH CAROLINA: Anson, Correll 7091 (DUKE, GA, MICH); Beaufort, Ashe Ae Bertie, Ahles & Haesloop 52163 (Ncu); Bun- combe, Biltmore 1421 (GH, NCU, NY, US); Granville, eae 1715 (Ncu); Harnett, Boyce (Ncu); Durham, Godfrey 6729 (GH); Davidson, oe S.A, 1826 (DUKE); Cumberland, Mecklenberg, Batson 408 (DUKE); Moore, Correll 7214 (pUKE); Nash, Godfrey & Kerr 6644 (DUKE, GH); Northampton, Ah/es 52459 ae Orange, Blomquist 466 (DUKE); Pamlico, Radford 42295 (Ncu), Pender, Ashe 319 (Ncu); Person, Bowmer 147 (NcU); Richmond, Correll 7134 (DUKE, GH); Sampson, Campbell 3994 (Gu); Scotland, Godfrey 6946 (Gu); Stokes, Ashe s.n., 1896 (NCU), Transylvania, Cain 265 (TENN); Wake, Blom- quist 119 (us); Wayne, Radford 31497 (Ncu); Vance, Ahles & Leisner 20288 (NCU). SOUTH Carona: Aiken, Ahles & Crutchfield 55181 (Ncu); Allendale, Bell 5124 (Ncu, us); Bamberg, Campbell 4008 (Gu); Barnwell, Batson & Kelley s.n., 1959 (Ncu); Calhoun, Ahles 35306 (Ga, Ncu); Charleston, Ahles & Haesloop 38652 (Ncu); Dillon, Ahles 37103 (GH, NcU); Dorchester, Ahles & Haesloop 37859 (Ncu), Florence, Campbell 3967 (Gx): Hampton, Ahles & Bell 20922 (Ncu); Kershaw, Radford 30043 (NcU, NY); Lexington, Radford 29877 (Ncu, vpp); Orangeburg, Ahles 35136 (Ncu); York, Ahles 34490 (NCU). 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 213 Georaia: Baker, Godfrey 67566 (Fsu, Us); Bartow, Duncan et al. 12110 (Ga), Berrien, Celarier A-2588-I (mo); Brantley, Campbell 4026 (Gu); Butts, Campbell 4143 (GH); Calhoun, Thorne 6863 (Ga); Charlton, Campbell 4139 (Gu); Clarke, Miller s.n., 1941 (us); Decatur, Godfrey 74097 (Fsu, Ncu); Dougherty, Eyles 7666 (puKE); Lincoln, Pyson & pa 92 (Ga); McIntosh, Bozeman & Ahles 2476 (Ncu); Oglethorpe, Harper 656 (DUKE, GA); Screven, Eyles 7569 (puKeE); Tift, Celarier A- 2608-I (Mo); Toombs, Duncan & Hardin 14522 (Ncu); Thomas, Campbell 4180 (GH). Ftoripa: Alachua, Swallen 5626 (us); Bay, Godfrey 61605 (Frsu), Brevard, Fredholm 6134 (Gu, us), Calhoun, Godfrey 61601] (DUKE, FSU, US, USF, vpB); Clay, Campbell & Godfrey 4139 (Gu), Collier, Deam 65368 (us); Columbia, DeSelm s.n., 1969 (TENN); Dade, cee 1903 (GH, US, VT); Duval, A. H. Curtiss 3636a (FSU, GH, MICH, MISSA, MO, us); Franklin, Campbell 3932 (Gu); Gadsden, Godfrey 64939 (Fsu, vDB); Gitnese conn 65098 (Fsu); Gulf, Silveus 6732-A (GA, TAES); Highlands, Brass 15647 (Gu, us); Hillsborough, Lakela 23700 (Fsu, US, USF); Jackson, Campbell 3782 (Gu); Jefferson, Campbell 4069 (Gu), Lake, Silveus 6689 (TAEs); Lee, Kellogg 27041 (mo); Leon, Kral 1805 (DUKE, FSU, NY); Levy, Campbell 4095 (Gu); Liberty, Godfrey 73140 (rsu); Madison, Kral 6188 (DUKE, FSU, GH, VDB); Martin, Campbell 3872 (Gu); Okaloosa, Godfrey 76203 (Fsu); Pinellas, Tracy 7768 (Gu, MO, NY, wis); Polk, Lakela 23588 (UsF, vpB); Putnam, Godfrey & Morrill 52629a (DUKE, FSu); Orange, Baker 63 (us); Osceola, Fredholm 6098 (Gu); ae , DeSelms.n., 1969 (TENN); Wakulla, Godfrey & Morar 61587 (DUKE, FSU, Us), Walton, Godfrey 55250a (FSU, GH, NCU, USF). OuI0: Highland, Braun s.n., 1961 (us); Teen Bartley & Pontius 814 (ny, us); Jefferson, Jones s.n., 1935 (Ny); Pike, Bartley 2989 (Ny), Vinton, Musgrove 770 (NcuU). INDIANA: Clark, Seaver 26865 (Gu, us); Floyd, Deam 51650 (us); Lawrence, Kriebel 1441 (DUKE); Orange, Deam 3339512 (us); Perry, Deam 39976 (us), Vanderburgh, Deam 33104 (us). Kentucky: Barren, Braun 3636 (us); Christian, Labisky 279 (wis); Edmonson, Braun 3563 (us); Fleming, Braun 1743 (Gu, us); Lyon, Athey 3877 (Fsu); McCreary, Braun 482 (us); Meade, Sargent 100 (Gu), Wayne, Braun 2834 (us). TENNES- sEE: Blount, Thomas s.n., 1964 (TENN); Coffee, Gattinger s.n., 1880 (micH), Cumberland, Jennison 3394 (TENN); Davidson, Gattinger s.n., s.d. (NY), Fentress, Rogers 41376 (TENN), Hickman, Kral 4357 (vps); Knox, Ruth s.n., 1894 (Ny); McNairy, Shanks et al. 14704 (TENN); Montgomery, Shanks 1077 (TENN); Morgan, Rogers 41376 (NCU); Roane, Rogers 41402 (NCU, TENN, vp); Stewart, Ellis & Clebsch 930 (Ncu); Van Buren, Jltis 3405 (TENN); White, DeSelm 218 (TENN). ALABAMA: Baldwin, Kral 29768 (vps); Barbour, Kral 38017 (vps); Bibb, Campbell 3959 (Gu); Coffee, Kral 41685 (vps); Covington, Kral 33656 (TENN, vpB); Cullman, Eggert s.n., 1898 (mo); ae ales (vps); Lau- derdale, Kral 29353 (Ga, vps); Lee, Earle & Baker s.n., 7 (MIC NCU, NY); Marengo, Kral 29588B (Fsu, vps); Mobile, Mohr s.n., ee aan St. Clair, Kral 37930 (vps); Washington, McDaniel 9915 (GA, vDB). Mississippi: Amite, Ray 5471] (MISSA): Forest, Rogers 7045 (NCu, TENN); Harrison, Tracy 8394 (GH, MO, NY, US, “wis); Lafayette, Reed 34 (rsu, vpb); Wayne, Kearney 163 (us). ILLinots: Hardin, ee ne (MO, WIS); Saline, Evers 67370 (Ncu, wis). Missourt: Barry, Mackenzie s.n., 6 (Ny); Franklin, sie 299 (mo); Jefferson, Eggert s.n., ey a Peps Stevermark Frans (Mo); Shan- n, Stevermark 16339 (GH, Mo); St. Gene 0 (GH, Mo); Washington, Reese 77858 (Us). ARKANSAS: rece “Tits 25312 (wIs); Conway, Moore 321046 (wis); Izard, Robinson 2134A (Gu, Ny, us); Lawrence, Robinson 2223 (Ny), Pope, Lawson s.n., 1949 (NcU); Pulaski, Engelmann 78 (Mo); Searcy, Robinson 2184 (us); Stone, Hatch- er s.n., 1951 (Ncu). LoursiANna: Beauregard, Heyne 1381 (us), ea re 17506 (mo); LaSalle, Swallen 1050] (us); Soin, Kral 16119 (Fsu, vps); R Cassady 5$.n., 1952 (Tags); Sabine, Kral 16219 (Fsu, vpB). OKLAHOMA: Le Flore, aes 9848 (us). Texas: Brazos, Sperry 2566 (Tacs); Erath, Hancock 64-8 (Tags); Galveston, Waller & Bauml 3208A (Gu, TAES); Robertson, Lincecum 31 (rags); Walker, McCleod s.n., 1957 (rags). Belize. El Cayo Distr., San Augustin, Lundell 6727 (micu, NY). Honduras. El Paraiso, Standley 29035 (F). Cuba. Las ViLtas: Santa Clara, Sabana de Motembo, Len 11343 (GH) 216 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 I include three previously described varieties in the common variant, so called because of its high frequency of occurrence in nature. Scribner distin- guished his Andropogon elliottii var. laxiflorus from A. elliottii Chapman var. elliottii on the basis of the upper stem sheaths of the former not being inflated and overlapping. In addition, he noted the long rachis internodes (‘‘joints’’) of his variety. In Florida, Mississippi, Louisiana, and South Carolina there are at least eight populations that resemble Scribner’s type in having all the inter- nodes of the upper part of the stem elongate. While these elongations make the plants quite distinctive, they appear to represent merely an abnormality of plants of the common variant of A. gyrans var. gyrans. One of the specimens (Nash 1759 (wy!)) cited by Nash in his protologue of 4. campyloracheus (based on A. elliottii var. laxiflorus) as being from the type locality has two stems that are connected at the base. One of these has the elongate internodes of Scribner’s A. elliottii var. laxiflorus, and the other has the overlapping, inflated upper stem sheaths characteristic of the common variant of 4. gyrans var. gyrans. The cause of this possible abnormality 1s unknown, but two observations may be of importance. First, all these abnormal plants flower earlier than is normal for the common variant. Second, half of the populations contain some plants that are viviparous, the flowers being replaced by small plantlets. It is not known whether or not these plantlets are effective in reproducing the parent plant. Their frequency of occurrence in nature in the virginicus complex as a whole is extremely low. Fernald and Griscom named a pane var. projectus without appreciating the variability of peduncle lengt e single distinguishing char- with short peduncles and entirely hidden (and cleistogamous-flowered) ra- cemes, and plants with some racemes hidden and some exposed on long pe- duncles. Very infrequently, some plants may have all or many of the racemes exserted above the apex of the raceme sheaths. Moreover, the claim that this taxon has a “notable geographic segregation” (Fernald & Griscom, 1935) is unsupportable since there does not appear to be any geographic pattern to variation in peduncle length. Andropogon elliottit var. gracilior was based solely on overall plant size. | I observed them. Their size is apparently not a phenotypic response to poor soil conditions since I have seen short and tall plants within 50 m of one another in a uniform flatwoods locality in southern Florida. Regardless of these observations, plant size is far too variable in the common variant to warrant recognition of this variety. 2a. 2. Tenuous variant Ficures 11, E, F; 31. DiaGnosis. Raceme sheaths visible after anthesis. DisTRIBUTION. Sandy, moist or dry soil. Florida and Mississippi (Map 10). REPRESENTATIVE SPECIMENS. United States. FLoripa: Brevard, Campbell 3765 (Gx); Char- lotte, Hitchcock 440 (Gx), Collier, Lakela 31196 (Gu), Franklin, Godfrey 75579 (Fsu), 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 217 Highlands, Campbell 3746 (Gu); one ae Lakela sols (Gu); Lee, Hitchcock 440 (GH, MO, Ny, US); Manatee, Perdue 1789 , GH, NCU, TAES, USF); Martin, Campbell 3873 (GH); Pinellas, Lakela 27621 (usF); Volusia, Shuey & Pepe. 1544 (USF). MIs- sissippi: Harrison, Tracy 470] (GH, TAES, Us); Jackson, Tracy 2243 (Ny). This taxon includes two kinds of plants. One has both long- and short- peduncled racemes—and hence both chasmogamous and cleistogamous flow- ers—on the same stem. I have seen this kind only once (Florida, Martin Co., Campbell 3873) in the field. There are two other populations in southern Mississippi (Map 10), one of which is the source of the lectotype of A ndropogon subtenuis Nash (Tracy 2243 (Ny), Hitchcock, 1951). The second kind has only short peduncles and predominantly cleistogamous flowers. It has neither been described formally nor discussed in any previous publications concerning the virginicus complex. Invariably it has been identified as 4. virginicus. As its name implies, the tenuous variant is difficult to circumscribe. It differs from the common variant quite strikingly in the one character provided in their diagnoses. Nevertheless, there are plants of intermediate morphology that are not readily placed into either taxon. The best way to distinguish the tenuous variant from Andropogon gyrans var. stenophyllus is by comparing the ligules; the latter taxon also occurs more frequently and ranges farther in more poorly drained soils. 2b. Andropogon gyrans Ashe var. stenophyllus (Hackel) Campbell, comb. Ficures 11, C, D; 32; 33. BASIONYM: Andropogon se L. subvar. seh Hackel in DC. Monogr. Phanerog. 6: 411. 1889. Types’ Florida, Chapman s.n., s.d. (lectotype, w!; isolec- totype, w!). Andropogon virginicus L. var. hn se (Hackel) Fern. & Griscom, Rhodora 37: 142. 1935. Andropogon perangustatus Nash in Small, Fl. SE. U.S. 62. 1903. Based on Andropogon virginicus L. subvar. stenophyllus Hackel. Not Andropogon stenophyllus Roemer & Schultes, 1817 Diacnosis. Ligules long, with marginal ciliations short. DistRIBUTION. Ditches, bogs, savannas, and pond margins. Coastal Plain from North Carolina to eastern Texas (Map 11) REPRESENTATIVE SPECIMENS. United States. NortH CAROLINA: Robeson, Ahles 37246 (ncu); Wilson, Radford 40687 (Ncu). SouTH CAROLINA: Saluda, Radford 30369 (Ncv). orGia: Berrien, Harper 1707 (Gu, Mo); Charlton, Campbell 4138 (Gu); Colquitt, Godfrey 76086 (FLAS): Emanuel, Plummer & Pullen 1962 (GA); Thomas, ees (GH). FLoripa: Brevard, — 6193 (us); are Campbell 4137 (GH); Duval, A. H. Curtiss 6016 (FSU, GA, GH, , NCU, NY, TAES, US, VT); Franklin, Campbell 3822 (cu): ea Chapman s.n. (GH, Highlands, Swallen 5368; Hillsborough, Fredholm 6427 - Indian River, Tracy 9298 (Taes, us); Jackson, Campbell 3813 (Gu); Jefferson, Campbel 4070 (Gu); Liberty, Campbell 3829 (Gu); Osceola, Fredholm 6072 (us); Pasco, Ray 9594 (us, usr); Putnam, Godfrey 76892 (Fsu); St. Lucie, Si/veus 6664 (TAES); Volusia, Silveus 6529 (DUKE, GA, TAES); Wakulla, Lamp 17 (rsu); Walton, A. H. Curtiss 6928 3Hackel cited ‘Florida (Chapman) et Alabama (Mohr).”’ There are two Chapman sheets of a Florida collection and one Mohr collection (Alabama, Mobile County, s.n.) in the Hackel herbarium at w. One of the Chapman specimens is here designated the lectotype. 218 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 (GA, GH, MO, NY, US). ALABAMA: Baldwin, Kral 38228 (Gu, us, vps); Geneva, Kral 52006 (vDB). MISSISSIPPI: Harrison, Tracy 3830 (MICH, MO, NY); Jackson, Earle 1890 (TAEs); McKellar 272 (us); Rapides, Duncan 1956 (Tags). TExAs: Galveston, Waller & Bauml 3325 (GH, TAES); Robertson, Hatch 2108 (TAEs). Like the tenuous variant of Andropogon gyrans var. gyrans, A. gyrans var. stenophyllus consists of two kinds of plants differing in flowering mode. The short-peduncled, cleistogamously flowering plants tend to grow in relatively disturbed sites such as roadside ditches and have a much wider range than the mostly long-peduncled and often chasmogamous plants. I have seen only three populations of the latter kind, all in infrequently disturbed boggy sites. Two of the populations contained predominantly chasmogamous-flowered plants. In the third locality, a rather wet flatwoods in Franklin County, Florida, was a mixture of the two flowering modes (Campbell 3822), with the chasmogamous type predominating. The two types of plants differ in three ways (see TABLE 7), which hold for nonmixed populations of both kinds of plants. It is not known whether or not these two kinds are genetic segregates of one another. They are not separated as taxa because they do not differ morphologically in any ways other than the three usually correlated characters, and these characters vary within a single population. — 3. Andropogon tracyi Nash, Bull. New York Bot. Gard. 1: 433. 1900. Type: Mississippi, Lowndes County, Tracy 3083, 1895 (holotype, Ny!). Ficures 11, T; 35. Diacnosis. Stems short; leaves narrow; inflorescence units few per stem: pe- duncles, spikelets, and anthers long. DistTRIBUTION. Sandhills, sandy pinelands and scrublands. Coastal Plain from North Carolina to Mississippi (Map 8 REPRESENTATIVE SPECIMENS. United States. NortTH CAROLINA: Gates, Ahles & Duke 51584 (Ncu); Granville, Dayton 1267 (Ncu), Harnett, Radford 8762 (Ncu); Johnston, Radford 29156 (NcU); Pender, Wells s.n., 1925 (Nv); Sampson, Campbell 3993 (Gu); Scotland, Ahles 36966 (Ncu). SoutH Carona: Aiken, Miller s.n., 1966 (NcU); Allendale, Camp- bell 3716 (GH); Bamberg, Campbell 4007 (Gu); Beaufort, Cuthbert s.n., 1902 (FLAas); Charleston, Ahles & Haesloop 38051 (NCU). Savas Chatham, Eyles 6645 (us); Screv- en, Campbell 4203 (Ga); Talbot, Jones s.n., 1959 (GA); Toombs, Hardin & Duncan 14526 (GA). FLORIDA: a Kral 52173 (vps); Dade, pen et al. 897 (Ny); Calhoun, Grelen 1 1/ 58-7 (Fsu); Duval, A. H. Curtiss 4012a (Ny); Highlands, Davis s.n., 1941 (FLAS); Hills- borough, pare 6423 (us); Orange, ee 329 (s); Putnam, Campbell 4100 (Gu); Walton, McDaniel 7145 (Fsu, vps). ALABAMA: Baldwin, Kral 29797 (us, vps); Clarke, 7047 (NCU, vpB); Harrison, Earle 1899 (ny); Jasper, MeDaniel 2796 (Mo, Ny); Lowndes, Tracy s.n., 1895 a ee River. Sargent 8668 (vps). Earlier workers have noted a similarity between Andropogon tracyi and A. longiberbis, and the two species are indeed similar in general aspect. They and in length of the peduncle, spikelet, and anther. Andropogon tracyi also has 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 219 TABLE 7. Comparison of the morphology of two kinds of plants* of Andropogon gyrans var. stenophyllus. EAN (+ S.E.) PERCENT CHAS- SAMPLE PEDUNCLE SPIKELET ANTHE MOGAMY _ SIZE LENGTH (mm) LENGTH (mm) LENGTH (mm) 73 (8)t 9 (8) 12—22(-45) (4.4-)4.7-5.2(-5.5) (1.1-)1.4-1.5(-1.7) 30 (6) 4 (4.5-)5—5.5(-6) (4-)4.2-4.5(-5.2) (0.9—-)1-1.2(-1.3) *Both kinds from Campbell 3822, Franklin County, Flori +For the meaning of the measurements in this table, see saunas in Character States section. a wider, more northerly range. It is possible that a shift in flowering mode toward cleistogamy in the ancestral stock of A. tracyi may have given rise to ancestors of A. /ongiberbis. However, in the other instances in the complex where such a change in flowering mode can be inferred, there is a pronounced similarity in other specialized characters that strongly tie the taxa to one another. No such specialized characters in A. tracyi and A. longiberbis suggest a common ancestry. The resemblance of Andropogon tracyi and A. gyrans, particularly the ten- uous variant of var. gvrans, is also great. The tenuous variant has peduncles that are either variable or all short on an individual plant, smaller spikelets, and shorter anthers. 4. Andropogon liebmannii Hackel, Flora 68: 132. 1885. Type:* Mexico, Chi- nantla, Liebmann 77 (lectotype, c!; isolectotypes, c!, GH (two sheets)!, MICH!, UC (two sheets)!, Us (three sheets)!). Sorghum liebmannii (Hackel) Kuntze, Rev. Gen. PI. 2: 792. 1891. Anatherum virginicum oad subvar. /iebmannii (Hackel) G. Roberty, Boissiera 9: 213. Driacnosis. Peduncles all long, usually bearing 4 or more racemes; raceme sheaths long; spikelets broad. DISTRIBUTION. eee moist ground. Mexico and Coastal Plain of south- eastern United States The two varieties of Andropogon liebmannii have been treated as distinct species by most American authors. However, the morphological distance of 9 separating them (TasLe 6) is based primarily on differences of size, which possibly arise either from dissimilar growing conditions in Mexico and the ‘Although Hackel cited Liebmann 590 from Chinantla, Mexico, A. F. Maule of ne Botanical s numbers were not collector’s numbers ese has been chosen as the lectotype. Hackel also cited Bourgeau 2376, but the only specimen of this collection that I have examined (Mexico, Veracruz, 1865-1866 (GH!)) bears no indication that Hackel saw it. 720 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 southeastern United States or from pleiotropy of one or more genes controlling size. The two varieties share a character state (raceme number—character 22 of TABLE 4) found only in these two taxa, and they are strongly similar in most qualitative characters, flowering mode, and preference for wet soils. 4a. Andropogon liebmannii Hackel var. liebmannii FiGure 36. Andropogon liebmannii Hackel subvar. raripilus Hackel in DC. Monogr. Phanerog. 6: 413. 1889. Tyee: Mexico, Jalisco, Palmer 227, 1886 (holotype, not seen; isotypes, GHI, uc!, vT!). DiaGnosis. Small (in stem height, leaf-blade length, maximum inflorescence branching and sympodia, and inflorescence units per stem); leaves often only sparsely pubescent. DisTRIBUTION. Swales, banks, fields, and pinelands to 2200 m alt. Central volcanic belt of Mexico (Map 7). REI E SPECIMENS. Mexico. NAYARIT: Tepic, Rose 2222 (Gu), McVaugh et al. 16434 re Jats o: Los Colomos, Villarreal de Puga 4061 (micn); near Guadalajara, Palmer 227 (Gu, uc, vt), Pringle 11212 (F, GH, vt). MICHOACAN: near Ciudad Hidalgo, Gould 9613 Ce “Kral 25539 (vps); Uruapan, Hitchcock 263 (F, GH, UC). MEXICO: Tomasealtepes, Hinton 1413 (Gu). Puesia: Chinantla, Liebmann s.n., 1841 (c, GH, MICH, uc). Hipatco: Agua Blanca, Moore 2064 (GH). VERACRUZ: Tzuatlanchillo, Bourgeau 2376 (GH). Andropogon liebmannii var. liebmanaii flowers during August and Septem- ber, and infrequently in May and June. It is either a rather rare or an under- collected taxon presently known from only 12 localities. 4b. Andropogon liebmannii var. pungensis (Ashe) Campbell, comb. nov. Ficures 11, J; 37; 38. BasionyM: Andropogon mohrii (Hackel) Vasey var. pungensis Ashe, Jour. Elisha Mitchell Sci. Soc. 15: 113. 1899. Type: North Carolina, Washington County, Ashe 1598 (holotype, Ncvu!). ees liebmannii Hackel subvar. mohrii Hackel in DC. Monogr. Phanerog. 6: 1889. Type: Alabama, Mobile County, Mohr s.n., 1888 (holotype, us!). An- ogon mohrii (Hackel) Vasey, Contr. U. S. Natl. Herb. 3: 11. 1892. Anatherum heen (L.) subvar. mohrii (Hackel) G. Roberty, Boissiera 9: 213. 1960. Diacnosis. Large (relative to var. /iebmannii in characters listed in diagnosis of that taxon); leaves usually densely pubescent. DisTRIBUTION. Bogs, swamps, savannas, and flatwoods. Coastal Plain from Virginia to Louisiana; relatively infrequent in Florida (Map 6) REPRESENTATIVE SPECIMENS. United States. VirGinia: Dinwiddie, Swallen 5541 (us); Prince George, Fernald et al. 6758 (GH). NorTH CAROLINA: Pender, Blomquist 10064 (DUKE, GH, NY, US); Robeson, Ahles 37317 (Ncu); Washington, Ashe s.n., 1898 (NCU). sea Plummer & Pullen s.n., Hee ee Wheeler, Plummer & Pullen s.n., 1962 (GA). LORIDA: Escambia, Campbell 3948 (Gu); Franklin, Godfrey 79254 (Fsu), Jackson, ae, 3913 loa Liberty, rae 78259 (Fsu), Washington, Godfrey 80130 (Fsu). 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 22) Lshtlitititierln ibetidedadi aa Fisher Scientific Company SFE «206 Fisher Scientific Company dF CENTIMETERS 38 = x u 2 FIGURES 36-41. 36, Andropogon liebmannii var. liebmannii (Kral 25539, Mexico), portion of Pfloieseence. 37, 4. liebmannii var. pungensis (Campbell 3948): 37, portion of inflorescence; 38, stem. 39, 4O, . Se aa eee a 39, basal clump of leaves: 40, inflorescences at different stages (left, mature fruit; middle, young fruit; right, anthesis). 41, 4. /ongiberbis (C paola ll 4098), two inflorescences. Scale = 222 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ALABAMA: Baldwin, Tracy 8604 oo MO, TAES, US); Butler, engeenelens Escambia, Kral 44845 (vps); Mobile, Mohr s.n., s.d. (MISSA, NCU, NY, T us); Washington, Kra/ 29542 (NCU, VDB). MISSISSIPPI: Hancock Rogers 2498 (NCU, TENN, i DB): Harrison, Tracy 3895 (DUKE, GH, NCU, NY, US, W); Jackson, Demaree 34499 (GA, MISSA, TAES, US, VDB); Pearl River, Sargent 9027 (GA, ee Stone, DeSelm 1969 (TENN). LOUISIANA: ee Drummond s.n., 1832 (Ny); Calcasieu, Drummond s.n., 1833 (us). Andropogon liebmannii var. pungensis forms small, infrequent populations in little-disturbed boggy sites. The often densely pubescent stem sheaths, the number of racemes per inflorescence unit, and the large spikelets make this variety easily identifiable. Some botanists have confused it with the old-field variant of A. virginicus var. virginicus (e.g., Kral’s (1976) report of this taxon from Franklin County, Florida, is based on Kra/ 52387, which is actually the old-field variant of A. virginicus var. virginicus). These two taxa are similar in their densely pubescent stem sheaths and numerous racemes per inflorescence unit. They are best distinguished by peduncle length and spikelet dimensions. De ei net floridanus Scribner, Bull. Torrey Bot. Club 23: 145. 1896. Type: a, Lake County, Nash 1572, 1894 (holotype, us!; isotypes, GH!, US a sheets)!). Anatherum virginicum (L.) Sprengel subvar. floridan- um (Scribner) G. Roberty, Boissiera 9: 212. 1960. Figures 11, H; 39; 40. Andropogon bakeri Scribner & Ball, Bull. U. S. D. A. Div. Agrost. 24: 39. 1901. Type: Florida, Orange County, Baker 58, 1897 (holotype, Us, not seen; isotypes, GH!). Diacnosis. Leaves long, glabrous; peduncles long; hairs of rachis internode uniformly distributed, not becoming sparse toward base. DISTRIBUTION. Sandy soils in southeastern Georgia and Florida; in Pinus clausa (Chapman) Vasey scrublands in eastern Florida (Map 4). REPRESENTATIVE SPECIMENS. United States. GeorGiA: Long, Bozeman 1979 (GA, NCU). FLoripa: Alachua, Gould 6671 (Tags); Brevard, Fredholm 6008 (Gu, Mo, us); Citrus, Combs 976 (GH, us); Collier, Lakela 3119] (Gu, usr); Franklin, Si/veus 6503 (TAES, us): Hernando, Ray 9530 (GH, NCU, Us, vpB); Highlands, Campbell 3887 (Gu); Hillsborough, Garber 1877 (Gu, vt, w); Lake, Nash 1681 (Gu, Ny, us); Lee, Chase 4176 (us); Leon, Kral 1801 (puKE); Levy, Cooley et al. 7174 (NCU, USF, WIS); Manatee, Combs 1291 (us); Marion, Campbell 4194 (Gu); Martin, Campbell 3754 (Gu); Orange, Campbell 3860 (GH); Osceola, Fredholm a (GH); Palm Beach, Hitchcock 2264 (us); Polk, Lakela 273 (GH, NCU, USF); Putna oe 1960 (us); St. John, Si/veus 6744 (us); St. Lucie, Si/veus 5294 (TAES); ee Kral 5197] (vps); Seminole, Beardslee 41 (us); Volusia, Hood 32 (us). Plants of this species are easily distinguished from the rest of the virginicus complex by morphology and ecological preferences. The uniform distribution f pubescence on the rachis internodes is a unique and constant character. Laessle (1958) found Andropogon floridanus to be a good indicator of Pinus clausa scrub vegetation, an association confined to Florida. Other broomsedges may sometimes grow with P. clausa, but they characteristically invade more disturbed sites. Andropogon floridanus usually occurs in small populations of scattered individuals, but several stands of hundreds of individuals have been observed in clear-cut scrub pine (P. c/ausa) timberlands in central Florida. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 229 6. Andropogon longiberbis Hackel, Flora 68: 131, 132. 1885. Type: Florida, Garber 1877 (lectotype, w!; isolectotypes, Mo!, Ny!, Us!, w (two sheets)!). Sorghum longiberbe (Hackel) Kuntze, Rev. Gen. Pl. 2: 792. 1891. An- atherum virginicum (L.) subvar. longiberbe (Hackel) G. Roberty, Bois- siera 9: 213. 1960. Ficures 11, I, J; 41 Diacnosis. Leaf pubescence mostly appressed; raceme sheaths rather broad; spikelets long; callus hairs long. DisTRIBUTION. Sandy or rocky soils of roadsides, dunes, sandhills, vane and fields. Southern South Carolina to Florida and Bahamas (Map 2); commo only in Florida. REPRESENTATIVE SPECIMENS. United States. NorTH CAROLINA: New Hanover, Schallert 1935 (DUKE). SoutH CaRoLina: Beaufort, Bell 1956 (NCU). GEORGIA: McIntosh, Duncan 20605 (DUKE, FLAS, MISS, TENN, WIS). FLORIDA: Alachua, Campbell 3854 (Gu); Brevard, Campbell 3764 (Gu), Small et al. 3352 (Ny); Charlotte, Parrott 8] (DUKE); Clay, Campbell 4134 (Gu); Collier, Deam 60590 (pUKE); Columbia, Godfrey 76903 (Fsu); Dade, Campbell 3729 (GH); Duval, A. H. Curtiss 5571 (GA, GH, MO, NCU, NY); Flagler, Young s.n., 1940 (Ga); Franklin, Godfrey 77349 (FSU); ): Gilchrist, Campbell 3910 (Gu), Hernando, Correll 5 (GH, us); Hillsborough, Lakela 24138 (Fsu, GH, NCU); Jackson, Campbell 4032 (Gu); Lake, Nash 645 (Gu, Ny, TAES, US); Lee, Standley 169 (Gx, Ny, us); Leon, Kral pe (cu); Levy, Campbell 4097 (Gx); Madison, Godfrey 75784 (rsu); Manatee, Rugel 2 (NY, US, USF); Marion, Campbell 4108 (Gx), Monroe, Campbell 4225 (Gu); Nassau, ae 74687 (FLAS), Okeechobee, Si/veus 5776 (us), Orange, pee 49 (GH, Ny); Os- ceola, DeSelm 1969 (TENN), Pinellas, Tracy 7185 (GH, NY, TAES, US, W, WIS); Polk, Cooley et al. 8251] (GH, usF); Putnam, Campbell 4089 (Gu), Seminole, REA 20854 (wis); Suwannee, Campbell 3847 (Gu); Taylor, Godfrey 75778 (rsu); Volusia, Hood 41 (us); Wakulla, Campbell 4063 (Gu). Bahama Islands: Grand Bahama, Correll & Kral 43036 (vDB). Andropogon longiberbis is related to several other members of the virginicus complex. It apparently hybridizes with both the old-field variant of A. virginicus var. virginicus and the robust variant of A. glomeratus var. pumilus (see section on hybridization), suggesting some affinities. The morphological gap of 6.5 between A. longiberbis and A. virginicus is the smallest of all interspecific gaps. Furthermore, the gaps between A. /ongiberbis and two of the three variants of var. virginicus are only 7 and 6.5. However, there are three morphological differences that always hold: leaf-pubescence orientation, callus-hair length, and spikelet length. These differences and the abundance of A. longiberbis in southernmost Florida (where A. virginicus is quite sparsely distributed) justify maintaining A. /ongiberbis as a distinct species. Andropogon longiberbis and the robust variant of A. glomeratus var. pumilus differ in many ways, the most reliable being peduncle length, raceme- sheath width, spikelet dimensions, callus-hair length, and scabrousness of the lower glume keels. SHackel cited “Florida leg. Garber, Curtiss.” He apparently had at least three sheets of the first collection and one of the second (Florida, Duval County, 4. H. Curtiss 3638, w!, FLAS!, us!) in his herbarium (the sheets are stamped “Herbarium E. Hackel”). One of the Garber heets is here selected as the lectotype. 224 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 The possibility that Andropogon longiberbis is a cleistogamous derivative of ancestral stock of 4. tracyi is discussed under that species. About 10 percent of all the specimens of Andropogon longiberbis examined in this study bear vernal-flowering stems. Over 90 percent of the 29 plants flowering between February and July had charred basal leaves, indicating a recent fire and the possibility that burning stimulates flowering. 7. Andropogon virginicus L. Sp. Pl. 1046. 1753, not Sp. Pl. ed. 2. 1482. 1763 (= Andropogon leucostachyus HBK.). Type: America (holotype, LINN 1211.12, photo Gu!). Holcus virginicus (L.) Steudel, Nom. Bot. ed. 2. 1: 773. 1840, pro syn. Sorghum virginicum (L.) Kuntze, Rev. Gen. PI. 2: 792. 1891. DiaGnosis. Ligules short, usually brownish and prominently ciliate margined: peduncles and spikelets short; anthers often marcescent within spikelet. DISTRIBUTION. In more or less full sun and in all but poorest soils; extremely weedy. Massachusetts to Ontario and Michigan, south to Florida and Texas; Mexico and Central America; West Indies; Colombia; apparently naturalized in California, Hawaii, Japan, and Australia. Andropogon virginicus is the most ubiquitous, weedy, and taxonomically complex broomsedge in the eastern United States. The old-field variant is an extraordinarily successful colonizer of moist, cleared ground at low elevations. The great variability of A. virginicus ties it to four of the other species of the complex: one or more of its variants resemble eight of the lowest-ranking taxa of these four species. The ancestral stock of A. virginicus is hypothesized to be A, brachystachyus, and its closest relative from a morphological standpoint is A. longiberbis. Finally, certain of its variants have been confused or lumped with three taxa of 4. glomeratus and two of A. gyrans. These ties are considered in more detail in the discussion of these four species. At the subspecific level there are two varieties of Andropogon virginicus that differ in leaf color (green vs. glaucous). At a still lower level the three variants of var. virginicus and the two of var. glaucus are so closely related to one another and are separated by such variable characters that to bring them into the formal nomenclatural scheme of the group is impractical. The subspecific variation in this species reflects rapid and probably recent evolution. With the acquisition of the capability for cleistogamous flowering and the tremendous disturbance of the native vegetation by man has come an explosion of broomsedge populations. A possible phylogeny of this species and its closest relative, Andropogon brachystachyus, is discussed under the latter species. Within Andropogon virginicus all of the distances between the five variants lie between 3 and 8.5 (TABLE 6). Due to the smallness of the morphological gaps between the variants of A. virginicus, they have all been combined into one species. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 220 Key to the Subspecific Taxa of Andropogon virginicus 12, WAVES CLCCN et irks ye od ne eins 2A ee A. virginicus var. ViIrginicus. 2. Stem ape green; leaves usually pubescent, at least on margin near collar. . Raceme sheaths (1.7-)2.4-3.1(-4) mm wide; racemes (1.3-)1.5-2.3(-3) cm long: usually at least some peduncles more than 12 mm long. ............. 7a. 1. Deceptive variant. 3. Raceme sheaths usually more than 3.1 mm wide; some racemes longer than cm; peduncles less than 12 mm long. ........... 7a. 2. Old-field variant. 2. Stem internodes glaucous; leaves glabrous. ............ 7a. 3. Smooth variant. 1. Leaves glaucous. .......... 0.666. e eee eee 7b. A. virginicus var. glaucus. 4. Pubescence below raceme sheath absent; raceme sheaths (2.1-)2.6-3.8(-4. e cm long; spikelets (2.6-)3.2-3.5(-3.9) mm long; racemes (1. ae 7-2.4(-3.2) cm long. ORM eR hdc choos gaat on tent es dens AN eae Le . Drylands variant. 4. Pubescence below raceme sheath sparse to dense; raceme sheaths (2.4-)3.2-4.8 (-6) cm long; spikelets (3—-)3.5-3.9(—4.4) mm long; ne A 5—)2—3(-4) cm long. 2. Wetlands variant. 7a. Andropogon virginicus L. var. virginicus Cinna ae Walter, Fl. Carolin. 59. 1788. Type: presumably South Carolina (ho- lotype photo, GH Preah Gisions Michaux, Fl. Bor. Am. 1: 57. 1803 (as dissitiflorum, nomen superfl. for Cinna lateralis Walter and with the same type Andropogon vaginatus Ell. Bot. S. Carolina Georgia 1: 148. 1816. TYPE: no indication of origin (holotype, CHARL®). Andropogon virginicus L. var. vaginatus (Ell.) Wood, Class-book sa 1861. Dimeiostemon vaginatus (Ell.) Jackson, Index Kew. 1: 760. 1893, p Andropogon re Ell. Bot. S. Carolina Georgia 1: 150. 1816. oe South Carolina, Charleston (holotype, CHARL; photo, Gu!). Andropogon virginicus L. var. tetrastachyus (Ell.) Hackel in DC. sey Phanerog. 6: 411. 1889. ee ern tetrastachyum (Ell.) Jackson, Index Kew. 1: 760. 1893, pro syn. (as tetrastachys). Anatherum virginicum (L.) Sprengel ue tetrastachyum (Ell.) Roberty, Boissiera 9: 213. 1960. Andropogon eriophorus Scheele, Flora 27: 51. 1844, ex char. Not Andropogon erio- phorus Willd., 1805. Type: West Virginia, Charles Town (holotype, not seen). AUgrOpOROn curtisianus Steudel, Syn. Pl. Glum. 1: 390. 1854. Type: M. A. Curtis s.n. (CAEN, P,’ not seen). Andropogon virginicus L. subvar. genuinus Hackel in DC. Monogr. Phanerog. 6: 410. 1889. Andropogon virginicus L. var. genuinus Fern. & Griscom, Rhodora 37: 142. 1935. Both of the above names are based on Andropogon virginicus L DiAcnosis. Leaves green (sometimes slightly glaucous in smooth variant); ra- cemes often more than 2 per inflorescence unit. DISTRIBUTION. Same as species. ‘Scribner (1901) and Weatherby a placed this specimen in the taxon here recognized as Andropogon virginicus L. var. virgin 7Chase (1937) placed both i cae in the taxon here recognized as Andropogon virginicus L. var. virginicus 226 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 acne 42-46, inflorescences: 42, deceptive variant of . BTA EE a r. virginicus (left, Campbell 3747) and ee variant of A. wirginicus var. glaucus (Campbell 3898); 43, drylands variant of A. virginicus var. glaucus (Campbell 386 5); 44, prachystachyus (Campbell 3884); 45, old-field variant of 4. virginicus var. virginicus Campbell 4144), 46, smooth variant of A. virginicus var. virginicus (Campbell 3902). 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 22) 7a. 1. Old-field variant Ficures 11, L, M; 45. Diacnosis. Raceme sheaths wide; racemes usually 2 or more per inflorescence unit. Distrisution. Rapidly colonizing openings in mature vegetation created by disturbance; wide variety of soils, with some capacity for ecotypic differentia- tion (Chapman & Jones, 1975); to about 1000 m alt. in mountains. Massa- chusetts to Ontario, Michigan, and Iowa, south to northern Florida and Texas; distributed sporadically south of United States; naturalized in California, Ha- waii, Japan, Australia, and perhaps elsewhere (Map 12). REPRESENTATIVE SPECIMENS. Canada. ONTARIO: Kent, Catling et al. s.n., 1977 (MICH). United States. MASSACHUSETTS: Barnstable, Fernald 435 (DUKE, GH, MICH, MISSA, MO, ncu (2 sheets), NY, wis); Dukes, Kennedy 5.n., 1896 (GH (2 sheets)); Hampshire, Ah/les 84864 (cas); Middlesex, French s.n., 1888 (mo); Nantucket, Bicknell s.n., 1904 (Ny). Ruope IsLAND: Washington, Celarier A-2620-I (Mo), Newport, Fernald et al. 8480, 8487 GH). CONNECTICUT: Fairfield, Eames 8723 (usr); Hartford, Bissell s.n., 1901 (GH, Ny); iddlesex, Gould 8832 (rars); New Haven, Allen s.n., 1877 (GH). New York: Bronx, Bicknell 9656 (Ny); New York, Burnham 597 (Gu), Suffolk, Latham 29538 (Ga). New Jersey: Atlantic, Wagner 500 (mMicH); Burlington, Koster 05-53-2 (MICH); Cape May, Mackenzie 6331 (DUKE, MO, NY); Middlesex, Churchill s.n., 1889 (GH, MO); Ocean, Mackenzie 2369 (mo, Ny), Union, Griscom 1500 (UsF). PENNSYLVANIA: Adams, Lough- ridge 2767 (Taks); Berks, Stoudt & Hermann 2728 (micu); Dauphin, Berkheimer 15262 (Gu); Huntington, Westerfield 19186 (Ncu); Indiana, Feduska s.n., 1958 (vt); Mont- gomery, Dreisbach 1150 (micu); Philadelphia, Brenner 8351 (Gu); York, Pohl 1463 (wis). DELAWARE: Kent, Larsen 309 (Gu); New Castle, Canby s.n., 1899 (Gu); Sussex, Larsen 465 (DUKE, GH, MO). MARYLAND: Allegany, Brown S.n., 1971 (Ncu); Anne Arundel, Smith s.n., 1879 (us); Harford, Smith s.n., 1879 (us), Kent, Campbell 4219 (GH); Queen Annes, Campbell 4218 (Gu); Talbot, Earle 3734 (DUKE, GH). DISTRICT OF CoLumMBIA: Chase 273 (GH, MO, NY, US (2 sheets)). WEST Vircinia: Barbour, Moore 2546 (GH, MicH); Cabell, Gilbert 807 (GH, NY, TENN, wis); Greenbrier, Hunnewell 7222 (Gu); Hampshire, Downs 8980 (Ncu); Kanawha, Dennison s.n., 1966 (TAEs); Monongalia, Millspaugh 834 (Ny); Randolph, Greenman 79 (GH). VirGiniaA: Arlington, Allard 76 (mo); Bedford, A. H. 7044 (DUKE, GH); Haywood, Ahles & Duke 50305 (miss, Ncu); Hoke, Ahles 36355 (NCU, wis); New Hanover, Schallert s.n., 1935 (DUKE, NCU); Rowan, Small & Heller 349 (mo, NY). SouTH Carona: Aiken, Ahles & Crutchfield 55214 (Ncu); Allendale, Bell 5204 (DUKE, NCU); Beaufort, Bell 5249 (DUKE, NCU); Florence, Campbell 4003 (Gx); Jasper, Bell 5293 (DUKE, NCU); Kershaw, Radford 29927 (Gx, NCU); Lee, Radford 29382 (NCU, ny); Lexington, Radford 29798 (GH, NCU). GEORGIA: Baker, Thorne 6975 (GA); Bartow, 47, A. brachystachyus (left, Campbell 3884) and drylands variant A. wrginicus var. glaucus (Campbell 3898), basal clumps of leaves. 48, old-field (left) and smooth variants of A. virginicus var. virginicus (Campbell 4133a, b, St. Johns Co., Florida), stems. Scale = 15 cm. 228 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Map 12. Distribution of old-field variant of Andropogon virginicus var. virginicus. Greear 63371 (Ga); Bullock, Hall 221 (Ncu); Butts, Campbell 4144 (Gu); Carroll, Camp- bell 4147 (Gu), Clarke, Cronquist 4208 (Ga, GH, Ny); Coweta, Campbell 4145 (Gu); Dade, Cronquist 4832 (GA, GH, Ny); Evans, Hardin & Duncan 14660 (GA, micH); McIntosh, Duncan 20605 (DUKE, GH, US, USF, Wis); Putnam, Cronguist 4858 (GA, GH, MICH, MO, Ny); Thomas, Campbell et al. 4057 (Gu); Walker, Cronquist 4802 (GA, GH, MICH, NY); Wayne, Campbell 4019 (Gu). FLoripa: Alachua, Swallen 5558 (us); Baker, Campbell 3849 (Gu); Bay, Godfrey 61631 (DUKE, Fsu, USF); Calhoun, Godfrey 61600 (FSU, USF); Collier, Cooley et al. 9010 (Fsu, GH, TENN, USF); Dade, Atwater GS 156 (Fsu, USF); Dixie Godfrey 56153 (Fsu, Ny); Duval, A. H. Curtiss 3636 (DUKE, GA, MISSA, MO, NCU, a; Escambia, Campbell 3949 (Gu); aS Kral 52387 (us); Gilchrist, Godfrey 74119 (Fsu, vDB); Highlands, Ray et al. 10370 (Ncu, UsF, vpB); Hillsborough, Ray et al. 10126 (NCU, TAES, USF); Jackson, Campbell 3770 (Gu); Jefferson, Campbell 3843 (Gu); Lee, Lakela et al. 30541 (NCU, USF); Liberty, eats 4075 (Gu); Martin, Campbell 3896 GH); Pasco, Ray 9607 (Gu, NCU, US, USF); Pinellas, Tracy 7375 (mo, Ny, w); Wakulla, ans NCU, VDB); Walton, Godfrey 55249 (DUKE, FSU, GA, GH, NY). MICH- IGAN: Allegan, Van Schaack s.n., 1945 (us); Berrien, Parmelee 3017 aa: Livingston, Kilburn 371 (micu); Muskegon, Bourdo 20 (micn), Washtenaw, Bartlett s.n., 1956 (MICH). Ouio: Athens, Si/by s.n., 1896 (Ny); Cuyahoga, Jones 1397 (TENN); Greene, Demaree 11850 (mo, wis); Hamilton: Stephenson s.n., 1930 (GA, Mo); Huron, Jones 1270 (su, ncu); Pike, Crow/ s.n., 1937 (Ny); Stark, Brown s.n., 1940 (NY). INDIANA: Crawford, Deam 30268 (us), Dariess, Deam 7623 (us); Dearborn, Deam 42717 (us); Franklin, Deam 35304 (us); Lawrence, Kriebel 1428 (puKE); Monroe, Duncan 27] (GA); Owen, Deam 35013 (us); Pike, Deam 35076 (micn); Switzerland, Deam 41098 (us). KENTUCKY: Bath, Gleason & Griffiths G509b (micu); Butler, Nicely 3136 (NCU); Kenton, Braun 3752 (Ny, Us); Lincoln, Wharton G584 (mo); Montgomery, Wharton 6191 (micH, MO, NY); Powell, Gleason & Griffiths G173a (MICH, NY, TENN). TENNESSEE: Blount, Thomas SM, 1965 (TENN); Coffee, DeSelm 1994 (TENN); Fayette, DeSelm s.n., 1972 (TENN); Graniger, Cain s.n., 1938 (FSU, TENN, wis); Knox, Ruth 751 (Ny); Monroe, Sharp et al. 17048 (Ncu, 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 229 h (TENN). ALABAMA: Baldwin, Campbell 3803a (Gu), Covington, Kral 44748 (vps); Hous- ton, Campbell 4155 (Gu); Lauderdale, Kral 29355 (GA, TENN, vps); Lee, Ea ny); Mobile, Mohr s.n., 1884 (Ny, us); Montgomery, Campbell 3956 (Gu); Randolph, Campbell 4149 (Gx); Sumter, Campbell 3955 (GH). MIssissipPr: Bolivar, Ray 5990 (MISSA, usF); Covington, Jones 10878 (miss), Forrest, Rogers & Robbins 4885 (NCU, vpDB); Har- rison, Tracy 4698 (GH, MISS, MO, NCU, w); Jackson, Tracy 3787 (mo, Ny); Lafayette, Pullen 641532 (Ga, Miss, NCU); Lamar, Jones 2545 (FsU, MISS, NCU); Oktibbena, Tracy 1398 (Mo, Ny); Pearl River, Sargent 8489 (MICH, MISS, us); Tishomingo, Ray 7560 (Missa, derson 62-1056 (Fsu), Pike, Evers 67516 (Ncu), Williamson, Voigt S.n., 1950 (NcU). Iowa: Wapello, Hayden 8446 (us). Missourr: Boone, Rickett 1137 (DUKE); Butler, Eggert s.n., 1893 (mo); Cape Girardeau, Stevermark 64022 (mo); Franklin, Croat 3912 (mo); Howell, Steyermark 20034 (mo); Madison, Steyermark 1433 (Mo), New Madrid, Stey- ermark 83438 (Ga), Shannon, Redfearn et al. 844 (NCU); St. Clair, Henderson 67-1951 (FSU, VDB); St. Louis, Steyermark 638 (Mo). ARKANSAS: Baxter, Robinson 2242 (Ny); Benton, Plank s.n., 1900 (mo); Calhoun, Demaree 22670 (Mo, Ny); Clark, Demaree 21760 (mo, NCU, US); Fulton, Robinson 2285 (NY); Grant, Demaree 16571 (MICH, MO, NY, US); Howard, Demaree 9706 (Mo, Ny); Izard, Robinson 2135 (NY); Jefferson, Demaree 23398 (mo, NCU, NY); Miller, Heller 4233 (Mo, NY); Randolph, Robinson 2310 (Ny); St. Francis, Demaree 59675 (Ncu); Stone, Robinson 2328 (Ny). LOUISIANA: Allen, Ball 233 (ny); Aroyelles, Harvey 8099 (micu); Beauregard, Thieret 28136 (DUKE); Calcasieu, Shinners 22124 (wis); Jackson, Thomas & Cicala 31821 (NCU, TENN); Lincoln, Garrett 87 (wis): Morehouse, Thomas & DePoe 440 (vps), Ouachita, Thomas & Jones 895 (USF, YDB); Orleans, Hooker s.n., s.d. (NY); St. Tammany, Arséne et al. 11151 (Ny); Vermilion, Reese Los Indios, Leén & Victorin 17868 (Gu, us). Las Villas: Helechales, Leén 5407 (Gu); Soledad, Jack 6193 (A, CAS). Oriente: Loma Mensara, Ekman 321] (TEX, US). JAMAICA: Appleton, Hitchcock 9656/2 (us); Bull Head Mtn., Hitchcock 9542 (us); Cinchona, Harris 11268 (Gu, us); Claremont, Hitchcock 275 (GH, TEX, w); Montego Bay, Hitchcock 9680 230 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 us). DomINICAN Repus.ic: Monte Cristi, ee 12771 (GH, TEX); Santo Domingo, Trujillo, Allard 13194 (Gu). Lesser Antilles. TRINIDAD AND Tosaco: Broadway 4044 (GH). Colombia. CHocé: near Quidbo, Gentry & ee 24517 (Mo). Japan. KyusHu: Hondo, Togasi s.n., 1952 (GH, w, wISs). This variant 1s named for its prominence in old-field succession in the eastern United States. The type in the Linnaean herbarium apparently represents one part of this variant, and it is the basis of the common concept of Andropogon virginicus. The variant includes two kinds of plants that are so close they cannot be distinguished either in the field or in the herbarium with great confidence, although their extremes differ somewhat in height, inflorescence thickness, and number of racemes per inflorescence unit. I have seen the two kinds growing together in northern Florida and in North Carolina, where they are barely distinguishable. The deceptive variant may also closely resemble the old-field variant. The former has narrower raceme sheaths and a peduncle that occasionally exceeds 10 mm in length. These two variants are more easily separated in the field than the two kinds of plants of the old-field variant, but a similar uncertainty about herbarium material still prevails. 7a. 2. Deceptive variant Ficures 11, N; 42. Diacnosis. Raceme sheaths narrow; peduncles sometimes greater than 10 mm long; racemes usually 2. DISTRIBUTION. Flatwoods, scrublands, disturbed sites (e.g., cleared timber- lands). Coastal Plain from North Carolina to Florida (Map 16) REPRESENTATIVE SPECIMENS. United States. VirGINniA: Isle of Wight, Fernald & Long 12568 (Gu, us); Nansemond, Fernald & Long 10943 (DUKE, GH, US). NoRTH CAROLINA: Bertie, Campbell 4223 (Gu); Bladen, Campbell 4208 (Gu), Dare, Fosberg 17875 (micn); Duplin, Blomquist & Correll 4819 (Gu), Martin, Radford 41798 (wis); Nash, Godfrey & Kerr 662] (DUKE). SOUTH CaROLina: Beaufort, Cuthbert s.n., 1886 (FLAS); Berkeley, Ahles 35450 (Ncu); Clarendon, Campbell 4222 (Gu); Jasper, Bell 4813 (Ncu). GEorGia: Brant- ; ; Long Bozeman 1944 (Ga), McIntosh, Duncan 20643 (wis), Thomas, Campbell 3922 (cu), Tift, Shepherd 341 (rags). FLoripa: Alachua, Campbell 4186 (Gu); Baker, Godfrey 74698 (Fsu);, Bay, Billington s.n., 1921 (micu);, Brevard, Fredholm 6193 (us); Calhoun, Godfrey 75791 (esu); Charlotte, Lakela 24675 (FLAS, GH); Clay, Swallen 5583, 5598, 5614 (us); Collier, Lakela 31155 (GA, usF); Dade, Silveus 6620 (DUKE); Duval, 4. H. Curtiss 3639d (Gu); Franklin, Campbell 3931 (Gu), Gulf, Chapman s.n., s.d. (Gu); Highlands, Campbell 3747 (Gu); Hillsborough, Lakela 26662 (Gu); Indian River, Tracy 9257 (TAES, us); Jack- son, Campbell 4081 (Gu); Jefferson, Campbell & Godfrey 4068 (Gu); Lake, Campbell 4197 (Gu), Lee, Hitchcock 441 (mo), Leon, Godfrey & Campbell 4230 (Gu); Liberty, Campbell et al. 4168 (Gu); Manatee, Tracy 7107 (mo, Ny, w); Martin, Campbell 3870 (GH); Nassau, Godfrey 74699 (FLas); Okeechobee, West 23 (FLAS); Pinellas, Davis s.n., s.d. (FLAS); Polk, Lakela 23585 (Gu, us); Putnam, Godfrey 76896 (Fsu); St. Lucie, Silveus 6665 (rags); Sarasota, McFarlin & Van Dyne 12037 (us), Seminole, Chase 4137 (us); Walton, Godfrey 75763 (Fsu). This variant is so named because of its deceiving similarity to three other taxa, the old-field variant of Andropogon virginicus var. virginicus, A. glo- meratus var. hirsutior, and the robust variant of A. glomeratus var. pumilus. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 231 Maps 13-16. Distribution of some subspecific taxa of Andropogon virginicus: 13, var. ee wetlands variant; 14, var. g/aucus, drylands variant; 15, var. virginicus, smooth variant; 16, var. virginicus, deceptive variant. Discussion of the distinguishing characteristics of the deceptive variant is pro- vided under each of these taxa. The deceptive variant does not correspond to any taxon for which a name has been published. Its discovery has been crucial to an understanding of 232 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Andropogon virginicus because the variant appears to be both the primitive member of the species and the evolutionary link to 4. brachystachyus. 7a. 3. Smooth variant FiGureE 48. Diacnosis. Like old-field variant but with glaucous stems and glabrous leaves. DIsTRIBUTION. Poorly drained soils (e.g., shallow ponds, swales, cut-over flat- woods); much less common than old-field variant. Coastal Plain from North Carolina to Mississippi (Map 15). RE E SPECIMENS. United States. NorTH CAROLINA: Bladen, Campbell 4213 (an), Gaoren 1A: “Brantley, Campbell 4202 (Gu); Wayne, Campbell 4228 (Gu). FLORIDA: Alachua, Campbell 4232 (Gu); Brevard, Fredholm 6109 (Gu, us); Franklin, Godfrey 77363 (Fsu); Gulf, Chapman s.n., s.d. (GH); Highlands, Campbell 3902 (Gu); Indian River, Tracy 9759 (TAes); Jackson, Campbell 3785 (Gu); Leon, Godfrey 74595 (Fsu, vpB); Ae eae! 75790 (Fsu); Martin, ace 4129 (Gu); St. Johns, Campbell 4227 (GH); Volusia, Si/veus 6736 (vaes); Waku Godfrey & Morar 61570 (Fsu), Walton, Camptel 3787 —- ALABAMA: Baldwin, ered: 3803b (GH). Mississippi: Jackson, Tracy 2276 (NY The smooth variant does not correspond to any taxon for which a name has been published. It is a glaucous-stemmed, glabrous-leaved derivative of the old-field variant that has a narrower tolerance for soil moisture conditions. From a limited number of observations of progeny of these two taxa, there is no evidence that they are genetic segregates of one another. 7b. Andropogon ela sates L. var. glaucus Hackel in DC. Monogr. Phanerog. 6: 411. 1889. Type: Florida, Duval County, A. H. Guilin 3638b (ho- lotype, w!; eee GA (two sheets)!, GH!, mIssA!, Mo!, US Ficures 11, O: 42: 43: 47, pra virginicus L. var. dealbatus Hackel* in DC. Monogr ene: 6: 411. 889. Type: Alabama, Mobile County, Mohr 1884 ewe gs Peccoa capillipes Nash, Bull. New York Bot. Gard. 1: en Based on An- dropogon virginicus L. var. glaucus Hackel. Not . es glaucus Retz., 1789, or Andropogon glaucus Muhl., 1817. DiAGnosis. Leaves glaucous; racemes 2, rarely 3. DisTRIBUTION. Moist or dry soils on Coastal Plain. Southern New Jersey to eastern Texas (Maps 13, 14). The diagnoses and distributions of the drylands and wetlands variants of this taxon are included in the comparative discussion of the two taxa. R :SENTATIVE SPECIMENS. Dnjlands variant. United States. NortTH CAROLINA: Brunswick, Blomquist 10442 (DUKE); New Hanover, Chase 4582 (us). SourH CAROLINA: Allendale, Campbell 4015 (Gn). GeEorGIA: Brantley, Carapbell 4025 (Gu); Echols, DeSelm 1969 (TENN); Ware, Campbell *Hackel’s Andropogon virginicus L. vars. glaucus and dealbatus are united here for the first time. The former varietal name is used here to preserve common usage. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX PS 3708 (GH); Wayne, Campbell 4020 (Gu). FLoripa: Bay, Godfrey 76131 (Fsu); Brevard, Fredholm 6128 (GH, Mo, us); Calhoun, Ford 5642 (FLAS, NCU, w); Citrus, DeSelm 1969 (TENN); Collier, Lakela 31192 (GH, Mo, Ncu); Columbia, Silveus 6749 (DUKE, GA, TAES); Dade, Silveus 5284 (TAES, VT); DeSoto, DeSelm s.n., 1969 (TENN); Dixie, Godfrey 56152 (FSU, GH, NY, USF); Duval, A. H. Curtiss 36386 (GA, GH, MISSA, MO, us); Franklin, Godfrey 74585 (FSU, NCU, vpB); Glades, McCart 11190 (usr); Gulf, Silveus 6742a (GA, TAES, US); Hamilton, Godfrey 74648 (Fsu, NCU, vpB); Highlands, Brass 15632 (Gu, Us); Hillsbor- ough, Lakela 23614 (us, usF); Jackson, Campbell et al. 4165 (Gu); Jefferson, Godfrey 75828 (rsu); Lafayette, Godfrey 76906 (rsu); Lake, Ray 10543 (Gu, us); Lee, Stanley 12536 (us); Liberty, Campbell 3827 (Gu); Marion, Campbell 4196 (Gu); Martin, Camp- bell 3898 (Gu); Okaloosa, Godfrey 68900 (Fsu); Okeechobee, McCart 11120 (use); Or- ange, Anderson 4085 (Fsu); Osceola, Campbell 3865 (Gu); Palm Beach, Davis s.n., 1941 (Gu); Pinellas, Tracy 7377 (GH, MO, NY, TAES, US, W, WIS); St. Johns, Godfrey 747 16 (Fsu); St. Lucie, Silveus 6660 (raes, us); Seminole, Beardslee 43 (us); Wakulla, Godfrey 74591 (FLAS, FSU, NCU, VDB). ALABAMA: Baldwin, Kral 38250 (Ga, vps); Covington, Kral 44730 (vpDB). Wetlands variant. United States. NEw Jersey: Cape May, Long 5145 (Ny). NorTH CAROLINA: Bladen, Ashe s.n., s.d. (NCU, NY); Brunswick, Blomquist 405 (DUKE, Ny); Cumberland, Ahles 36571 (Ncu). GeoraiA: Bullock, Campbell 4018 (Gu); Colquitt, God- frey 76085 (rsu), Harris, Jones 22259 (Ga), Long, Bozeman & Radford 1943 (NCU). Fioripa: Bay, Godfrey 76135 (Fsu); Bradford, Conde s.n., 1977 (FLAS); Charlotte, DeSelm s.n., 1969 (TENN); Clay, Campbell & Godfrey 4136 (Gu); Collier, Cooley et al. 9062 (Fsu, GH, USF); Columbia, Ashe s.n., 1929 (Ncu); Dixie, Godfrey 69228 (Fsu); Duval, A. H. Curtiss 6055 (GH, TAES); Escambia, Silveus 6756 (TAES, VT); Gulf, Chapman s.n., s.d. (GH, MO, NY); Hillsborough, Combs 1346a (us), Indian River, Tracy 9284 (TAES, US); Jackson, Campbell 3912 (Gu); Lee, Wunderlin et al. 5399 (usr), Leon, Godfrey 75786 (rsu); Liberty, Campbell 3820 (Gu); Pasco, Ray 9596 (usr), Putnam, Godfrey 74721 (FSU, NCU, VDB); Santa Rosa, Godfrey 76185 (rsu); Walton, Campbell 3940 (Gu); Wakulla, Godfrey & Morar 61570a (DUKE, FLAS, FSU, USF); Washington, Campbell 3938 (Gu). ALABAMA: Baldwin, Campbell 3800 (Gu); Butler, Kral 44702 (vps); Geneva, Kral 41714 (GH, usr, vpB); Mobile, Mohr s.n., 1884 (Ny, US, Ww). Mississippi: Forrest, Rogers & Robbins 4884-B (miss), Harrison, Tracy 3897 (Mo, Ny, TAES); Jackson, Tracy 3814 (TAES). Louisiana: Calcasieu, Thieret 27991 (su); Livingston, Rogers 2419-A (miss, NCU); Rap- ides, Duvall 6019 (raks); St. Mary, Silveus 5406 (TAkES). Texas: Galveston, Waller 3316 (GH, TAES); Houston, Lindheimer s.n., 1841 (Mo). The pattern of small morphological differences associated with ecological differences of Andropogon virginicus var. virginicus reappears in the pair of taxa comprising A. virginicus var. glaucus. The drylands variant of A. virginicus var. glaucus produces generally shorter raceme sheaths, racemes, and spikelets; its flowers are more frequently chasmogamous, and unlike the wetlands vari- ant, it has no hairs below the raceme sheath. In addition, it grows in better- drained soil and has a narrower geographic range. In the majority of instances, these morphological and soil-moisture differ- ences are correlated. I have seen these taxa growing within one to three meters of one another at three localities in northwestern Florida. At only one of these was there difficulty in classifying any individual: a single plant on a slope between a bog inhabited by the wetlands variant and a roadside lined with the drylands variant combined the morphological features of these two taxa. The cause of this morphological intermediacy 1s unknown. The strong overlap in the characters separating the drylands and wetlands variants and the breakdown in the correlation of these characters in about five 234 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 percent of all populations examined in the herbarium are the reasons these two taxa are not recognized nomenclaturally. Hackel (1889) suggested that the drylands variant of Andropogon virginicus var. glaucus 1s closely related to (“vergit ad’’) A. brachystachyus. In inflores- cence morphology the two are sometimes similar because larger plants of the drylands variant usually produce arching branches (FiGure 43); they also share a preference for well-drained soils. They differ in stem height, in glaucousness, in peduncle, spikelet, and anther length, and in anther color. It is possible that A. brachystachyus stock is ancestral to the drylands variant, but the deceptive variant of A. virginicus var. virginicus 1s a more likely ancestor of the drylands variant because there 1s an even closer resemblance between these two. Apart from glaucousness, they differ in raceme-sheath length and in pubescence below the raceme sheat Since the deceptive and wetlands variants differ from the drylands variant in the same characters, it is not surprising that the first two taxa are morpho- logically alike. In addition to glaucousness, the wetlands variant differs from the deceptive variant in its larger raceme sheaths, wider range, and tolerance of more ee moisture Leaves of the wetlands variant are sometimes faintly glaucous, even on erie fae If such plants are not examined carefully, they may be mistaken fora plant of the smooth variant with inflorescence units with only two racemes. 8. Andropogon brachystachyus Chapman, FI. So. U.S. ed. 2. 668. 1883. TyPe:° Florida, Duval County, 4. H. Curtiss 3632 (lectotype, us!; isolectotypes, DUKE!, FsU!, GA!, GH!, MO (two sheets)!, Ncu!, Us!). Sorghum brachy- stachyum (Chapman) Kuntze, Rev. Gen. Pl. 2: 791. 1891. Anatherum brachystachyum (Chapman) G. Roberty, Boissiera 9: 212. 1960. FiGures 11, G; 44; 47. D1acnosis. Stems tall; leaves long, sparsely pubescent; inflorescence branches long, arching (in herbarium specimens inflorescences appear rather open); ra- cemes and awns short; spikelets and anthers long. DIsTRIBUTION. Sandy, often seasonally wet soils of flatwoods, savannas, pond margins, and scrublands. Southern Georgia and Florida (Map 3). REPRESENTATIVE SPECIMENS. United States. GeorciaA: Brantley, Campbell 4022 (GH); Chatham, Eyles 6652 (us); Long, Bozeman 1908 (Ncu); Wayne, Campbell 4021 (Gu). Froripa: Alachua, Si/veus 6519 (TAEs, us); Baker, Ashe s.n., 1928 (NCU); Brevard, Fred- holm 5558 (GH, Mo, us); Clay, Harper 40 (GH, Mo, us); Collier, Lakela 30327 (GH, NCU, NY, a Columbia, DeSe/m s.n., 1969 (TENN); Dixie, Godfrey 56183 (FSU, GA, GH, USF); Duval, 4. H. Curtiss 5338 (Ga, Gu, us); Highlands, Campbell 3884 (Gu); Indian River, Tracy Ae (TAES, US); anaes re 74602 (FSU, NCU, vpB); Lafayette, Godfrey 74632 (Fsu, vpB); Lake, Nash 1193 (GH, MO, NY, TAES, US); Leon, Godfrey 74594 (Fsu, vps); Levy, Godfrey et al. 64751 qs FSU, VDB); Madison, Godfrey 75831 (Fsu); Marion, Campbell 4195 (Gu); Martin, Campbell 3897 (Gu); Nassau, Godfrey 74680 (FSU, NCU, vpB); Okeechobee, McCart 11121 (usr); Orange, Baker 39 (Gu, Ny); Osceola, Campbell °The lectotypification of Hitchcock (1951) is followed here. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 239 3864 (GH); Pasco, Ray 9588 (GH, NCU, USF, vpB); Pinellas, Tracy 7376 (GH, MO, NY, TAES, us); Polk, Lakela 23584 (us, usF); Putnam, Godfrey 74720 (Fsu, NCU); Seminole, Beards- lee 46 (us), St. Johns, Godfrey 74719 (Fsu, NCU, vDB); Taylor, Godfrey 74634 (FSU, NCU, vps); Volusia, Hood 65 (us). The combination of elongate, arching inflorescence branches, long peduncles, and short racemes makes Andropogon brachystachyus very distinctive. Its clos- est relative is the deceptive variant of A. virginicus var. virginicus. These two taxa are separated by a morphological distance of 8 (TABLE 6). The deceptive variant may be a cleistogamous derivative of ancestral stock of A. brachy- stachyus. One population of Andropogon brachystachyus in Long County, Georgia (Campbell 4021), on the northern edge of this taxon’s range, suggests that one of its ancestors might have given rise to the deceptive variant. Most of the plants in this population are characteristic of A. brachystachyus: they are tall (average, 2 m; range, 1.57-2.17 m); the branches arch; the peduncles range from 20 to 40 mm long; the spikelets exceed 4 mm in length; and the anthers are more than 1.9 mm long. Two individuals, however, are shorter (average, 1.4 m) and have erect to only slightly arching branches, peduncles less than 13 mm, spikelets less than 3.8 mm, and anthers less than 1.4 mm. As peduncle, spikelet, and anther length suggest, there is a greater frequency of cleistogamous flowers at the base of the racemes of these two individuals than in the rest of the population. These changes may be the product of mutation in regulatory genes hastening sexual maturity. The importance of these plants is that they are strikingly similar to the deceptive variant of A. virginicus var. virginicus. Because of its smaller size, erect branches, shorter peduncles, and longer ra- cemes, the deceptive variant has a much different appearance than plants of A. brachystachyus (except the anomalous individuals of the Long County pop- ulation). In less conspicuous ways, however, they are very alike. They have the same ligule morphology, amount and distribution of leaf pubescence, and stem internode color. They differ, however, in three nonmorphological ways. First, in peninsular Florida the deceptive variant flowers two to four weeks before A. brachystachyus, although at the northern limit of the range of A. brachystachyus, their flowering periods overlap. Second, the deceptive variant is more frequently cleistogamous. Third, 4. brachystachyus may form large, dense populations, but it does not invade disturbed sites as does the deceptive variant. It seems plausible that the deceptive variant arose from A. brachy- stachyus stock as a precociously flowering form. This paedomorphosis (Gould, 1977) may be reflected in the earlier flowering time of the deceptive variant and, through the introduction of cleistogamy, may have provided the deceptive variant with a breeding system that is adaptive for a colonizer. 9. Andropogon glomeratus (Walter) B.S.P. Prelim. Catal. Anthophyta Pteri- dophyta New York, 66. 1888. Based on Cinna glomerata Walter, FI. Carolin. 59. 1788. Type: South Carolina (holotype, BM, not seen; frag- ment, NY!; photo, GH!). Sorghum glomeratum (Walter) Kuntze, Rev. Gen. Pl. 2: 790. 1891. Anatherum virginicus (L.) Sprengel subvar. glo- meratus (Walter) G. Roberty, Boissiera 9: 212. 1960. 236 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Andropogon macrourus Michaux, Fl. Bor. Am. 56, 57. 1803, nomen superfl. for Cinna glomerata Walter and with same type. Andropogon spathaceus Trin. Fund. Agrost. 186. 1820, nomen nudum, placed in synonymy of Andropogon macrourus Michaux by Steudel, Nomencl. Bot. ed. 2. 1: 93. 1840. Anatherum macrourum (Michaux) Griseb. Mem. Am. Acad. II. 8: 534. 1864, as macrurum. Dimeiostemon macrourus (Michaux) Jackson, Index Kew. 1: 760. 1893, pro syn., as macrurum DiaGnosis. Stem sheaths scabrous (smooth in taxon 9c, often in 9d): leaf blades long; ligules long, light brown, short-ciliate (often short and with long ciliations in taxon 9d); inflorescences oblanceolate to obpyramidal, usually with at least 3 branches at 1 or more stem nodes; inflorescence units numerous. DISTRIBUTION. Poorly drained soils; throughout entire range of virginicus com- plex. Only Andropogon glomeratus var. pumilus extends beyond eastern United States to western United States, Mexico, Central America, West Indies, and South America. Of the five taxa included in Andropogon glomeratus, three (vars. glomeratus, hirsutior, and glaucopsis) are particularly close to one another morphologically and ecologically. Besides the characters given in the species diagnosis, these three varieties have ligules that are identical in their length, short-ciliate mar- gins, and color. Andropogon glomeratus var. glomeratus appears to be the most primitive of the three because of its frequent chasmogamous flowering (as expressed in the long peduncles and large anthers). It also generally has larger spikelets than the mostly cleistogamous vars. hirsutior and glaucopsis. Finally, inflorescence shape can be used to separate var. glomeratus from the other two taxa. Although the distance between vars. glomeratus and hirsutior is only 4 (TABLE 6), the characters separating them are usually consistent and clear. In Virginia and the Carolinas, where individuals of vars. glomeratus and hirsutior may infrequently have similar peduncle lengths, one must rely upon inflorescence shape, spikelet and anther length, and flowering mode for iden- tufication While ‘hie stem sheaths of vars. glomeratus and hirsutior are scabrous, pu- bescent (at least near the collar), and green, those of var. glaucopsis are smooth, glabrous, and glaucous. Glaucousness (uncommon in the complex, in sect. LEpTOPOGON, and in the genus as a whole) is the basis for the conjecture that var. glaucopsis 1s derived, probably from var. hirsutior stock. FiGURE 49 shows the hypothesized phylogeny of these three taxa. The greatest problem within Andropogon glomeratus is the affinity of the robust variant of var. pumilus. In the past either it has not been recognized or it has been given the status of species, variety, or form. Morphologically, it is clearly farther from any of the first three taxa of the species than each of these is from any other. As discussed for vars. glomeratus and hirsutior, the robust variant may often appear very similar to them in general appearance. Characters of stem-sheath and glume-keel scabrousness and ligule morphology distinguish the robust variant and place the southwestern variant of var. pumilus in between the robust variant and the rest of 4. glomeratus. The morphological inter- 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 237 S » S A N SS se ro Ficure 49. Hypothesized phylogeny of Andropogon glomeratus vars. glomeratus, hirsutior, and glaucopsis. Numbers below names indicate mean percent chasmogamy (character 33 of TABLE 6). mediacy of the southwestern variant unites the robust variant and vars. glo- meratus, hirsutior, and glaucopsis. Key to the Subspecific Taxa of Andropogon glomeratus 9c. A. glomeratus var. glaucopsis. — . Leaves glaucous. ............000. 000s e eee eee . Leaves green. 2. Raceme sheaths usually less than 2.5 mm wide; stem sheaths often smooth; ligules usually less than | mm long, ee with long-ciliate margins. ................. 9 Robust variant of 4. glomeratus var. pumilus. 2. Raceme ance usually more a 2.5 mm wide; stem sheaths often rough; ligules usually more than | mm long, with short-ciliate margins. 3. Keels - ce glume scabrous to well below middle; plants of southwestern United States and northwestern Mexico. ...........00. 000000 e eee ee 9d. 2. Southwestern variant of A. glomeratus var. pumilus. 3. Keels of lower glume scabrous only above middle; plants of eastern United States. 4. Inflorescences (linear-)oblong; spikelets usually less than 4 mm long; anther usually marcescent within spikelet; peduncles less than 10 mm long; plants of Coastal Plain from Maryland to Mississippi. .............-0002+-005- RR BAS is 2s en aerate x eas ai _ A. glomeratus var. hirsutior. 4. Inflorescences oblong to obpyramidal; spikelets usually more than 4 m long; anther usually not marcescent within spikelet; usually some mature — 238 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 peduncles more than 10 mm long; plants of Coastal Plain and well inland from Massachusetts to Mississippi. ..9a. 4. glomeratus var. glomeratus. 9a. Andropogon glomeratus (Walter) B.S.P. var. glomeratus FIGURES Lis: 33: Andropogon macrourus Michaux var. abbreviatus Hackel in DC. Monogr. Phanerog. 6: 408. 1889. Type:'® Carolina, Rugel s.n., 1841 (lectotype, w!). Andropogon glo- meratus (Walter) B.S.P. var. abbreviatus (Hackel) Scribner, U.S. D. A. Div. Agrost. ull. 7(ed. 3): 15. 1900. Andropogon corymbosus nea a var. abbreviatus (Hackel) Nash i in Britton, Man. Fl. No. States Canada, 70. Andropogon macrourus Michaux var. cor rymbos Hackel in aa. on Phanerog. 6: 409. 1889. Type: Florida, Duval County, HT. Curtiss 3639c (holotype, w!; isotypes, FLAS!, GH (2 sheets)!, NcU, Us!) (NB. ae 3639¢ of GA! and Missa! are taxon oa ae as tee B.S. - var. corymbosus (Hackel) Scrib- . A. Div. l. 7(ed. 3): 1900. Andropogon corymbosus (Hackel) he in Britton, A 7 No. States a. 69. 1901. Andropogon vir- ginicus L. var. corymbosus (Hackel) Fern. & Griscom, Rhodora 142. 1935 Diacnosis. Inflorescences oblong to obpyramidal; peduncles and spikelets usu- ally long; stamens not regularly marcescent within spikelets. DISTRIBUTION. Bogs, swamps, savannas, flatwoods, and ditches. Massachusetts to Kentucky and Arkansas, south to Florida and Louisiana (Map 17) REPRESENTATIVE SPECIMENS. United States. MAssAcuusetTtTs: Barnstable, Fernald et al. 14984 (Gu); Bristol, Sturtevant s.n., 1888 (mo); Dukes, Seymour 1441 (DUKE, GH, NY): Nantucket, William s.n., 1894 (GH); Plymouth, Blake 11425 (micu, us). RHODE ISLAND: Washington, Collins & Lownes s.n., 1923 (GH). New York: Nassau, Bicknell 9643 (Ny); Richmond, Britton s.n., 1879 (Ny); Suffolk, Smith 3495 (mo, wis). New Jersey: Atlantic, Letterman s.n., 1887 (MICH, Mo); Burlington, MacE/lwee 1571 (GH, Mo, Ny); Camden Stewart 3052 (Ny); Gloucester, Foshberg 14488 (DUKE); Middlesex, Miller 1077 (nv): Monmouth, Britton s.n., 1883 (wy); Ocean, Morton 40931 (vps); Passaic, Nash s.n., 1899 (NY). PENNSYLVANIA: Berks, Br umbach 353-34 (GH); Chester, Pennell 8890 (Ny); Fayette, Boardman s.n., 1941 (vr); Lancaster, Carter s.n., 1910 (Ny); Philadelphia, Van Pelt s.n., 1906 (GH). DELAWARE: Kent, aes bare (GH); Sussex, Churchill s.n., 1908 (Mo). MARYLAND: Baltimore, Freeman s.n., 4 (wis); sina Wilkins 5709 (Gu); , Foreman s.n., 1873 (Ny); Prince ns True 2879 (DUKE, GA, US); Worcester, Campbell 3983 (GH). District OF CoLumBIA: Hitchcock 256 (GH, MO, NY, USF, W). WEST VIRGINIA: Cabell, Millendor s.n., 1939 (us); Raleigh, Berkley 2253 (us). VIRGINIA: Ac- (NcU); Isle of Wight, Fernald & Long 5759 (Gu); James City, Grimes 3874 (NY); Northampton, Fernald & Long 5183 (Gu); Southampton, Campbell 4217 (Gu); Surry, Terrell 4368 (Ncu). NortTH Carouina: Allegheny, Blomquist & Anderson 925 (DUKE); Bertie, Ahles & Haesloop 52066 (Ncu); Bladen, Ahles 37431 (Ncu); Brunswick, Bartram s.n., 1922 (GH); Buncombe, Biltmore 920° (Mo, NCU, NY); Cherokee, Radford 17577 (NcuU); Chowan, Ahles & Duke 51029 (NCU); Columbus, Bell 15740 (Ncu); Craven, God- Srey & White 6835 (GH); Cumberland, Ahles & Leisner 33534 (Ncu); Currituck, Blomquist 14382 (DUKE); Davidson, Denke s.n., 1826 (DUKE); Gates, Godfrey 7037 (Gu); Halifax, 'Hackel cited ““New Jersey (Gray); South Carolina (Rugel).”” The 1841 collection of Rugel from “Carolina” is here designated as the lectotype rather than Rugel’s 1842 collection from ‘‘Carolina” (w!). I have not seen material of Gray from New Jersey that Hackel would have seen 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX ZA9 vet 4F I > Cc o a E ‘2 ] Oo 2 = € 2 . @ <= 2 irs 4 Ficures 50-55. 50-54, inflorescences: 50, Andropogon glomeratus var. hirsutior (Campbell 3851); 51, robust variant of A. ae var. pumilus (Campbell 4030); 52, southwestern variant of 4. glomeratus var. pumilus (Crampton 6710, Yolo Co. , Cali- fornia (uc)); 53, A. eae glomeratus (Campbell 3915), 54, A. glomeratus var. ee) (Campbell 3812). 55, A. glomeratus var. glaucopsis (Campbell 3812), basal umps of leaves. Scale = 15 cm YY 240 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Maps 17-19. Distribution of some subspecific taxa of Andropogon glomeratus: 17, var. glomeratus; 18, var. hirsutior, 19, var. glaucopsis. Ahles & Leisner 20846 (Fsu, Ncu); Harnett, Laing 38] (Ncu); Henderson, Hunnewell 10084 (GH); Hertford, Ahles & Haesloop 52210 (Ncu), Hoke, Correll 7184 (DUKE, GH); Jackson, Correll 7903 (pUKE); Johnston, Radford 29217 (Ncu); Jones, Radford 40022 (Ncu); Lee, Stewart s.n., 1958 (NCU); Macon, Anderson s.n., 1953 (DUKE, FSU, GH); Martin, 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 241 Radford 41831 (Ncu); Moore, Blomquist 277 (DUKE), Nash, Ahles & Leisner 21259 (NCU, vps); Onslow, Wilder 17676 (puKE); Orange, Ashe s.n., s.d. (Ny); Pa mlico, Radford 42286 (ncu); Pender, Wells s.n., 1925 (ny); Pitt, Radford 41604 (Ncu); Richmond, Radford 19355 (Ncu); Robeson, Britt 2506 (wis); Sampson, Campbell 4000 (Gu); Scotland, God- frey 6964 (GH); Transylvania, egy 56938 (ncu); Tyrrell, Radford 42535 (Ncu); Wake, Ashe s.n., s.d. (Ncu); Warren, Ahles & Bell 21837 (Ncu); Washington, Radford 42352 (ncu); Wayne, Burk s.n., 1958 (cu); Yancey, Freeman 58353 (NCU). SOUTH Carouina: Aiken, Ravenel s.n., s.d. (Ny); Anderson, Davis 7919 (Mo); Barnwell, Batson & Kelley s.n., 1952 (cv); Berkeley, Ah/es 35461] (Ncu); Clarendon, Sees 4005 (Gu); Dorchester, Ahles & Haesloop 37783 (Ncu); Hampton, Bell 5351 (Ncu);, Lexington, Radford 29889 (Fsu, Soe Pickens, Rodgers 246 (DUKE); Sumter, Holdaway 46 (DUKE). GeEorGIa: Bartow, Duncan 13268 (GA, GH); Emanuel, Plummer & Pullen s.n., 1962 (NCU); poe Duncan 9028 ee Irwin, Harper 1709 (Gu, Mo, NY); Rabun, Duncan 1047 (GA); Thomas, Campbell 3921 (Gu); Toombs, Hardin & Duncan 14585 (Ncu); Ware, Campbell 4142 (Gu). Froripa: Alachua, Godfrey & Morrill 52620 (Fsu); Bay, Godfrey 61628 (Fsu, vpB); Brevard, Fredholm . (Gu, us); Columbia, Ashe s.n., 1929 (NcU); Duval, A. H. Curtiss 3639c (FLAS, GH, NCL w); Escambia, Campbell 3950 (GH); Gulf, Chapman s.n., 1893 (Ny); Jackson, Campbell 3915 (Gu); Jefferson, Godfrey 75826 (Fsu); Lake, Campbell 4109 (Gu); Leon, Kral 1801 er GH); Liberty, Campbell 4073 (Gu); oe Meislahn 53 (us); Polk, Jennings s.n., 1 (usr); Putnam, Godfrey 76895 (Fsu); Santa Rosa, Godfrey 76192 (Fsu); Wakulla, ae 72248 (FSU). KENTUCKY: Casey, Braun 2680 (us); Laurel, Braun s.n., 1933 (us); eas Rogers 37 (GA, GH, MICH, MO ny); Montgomery, Wharton 5375 (Gu, Mich, NY); Rowan, Braun 2085 (us). TENNESSEE: Bledsoe, Shanks et al. 3555 (NCU, TENN); Blount, Sharp et al. 32416 (TENN, VpB); C DeSelm et al. s.n., 1963 (Ncu); Cumberland, Shanks & Norris 7304 (TENN, US), oe Shanks 3076 (TENN, vpb); Grundy, Kral 44545 (vps); Morgan, Shanks 3062 (TENN); Polk, Clebsch 20217 (NCU, TENN). ALABAMA: Baldwin, Campbell 3805 (Gu); Cullman, Eggert s.n., 1897 (mo, ny, us); Jackson, Chase 4488 (us), ee Earle & Baker s.n., 1897 (mo); Mobile, Sargent s.n., 1941 (us); Pike, Leland s.n., s.d. (GH). Mississippi: Forrest, Rogers 4735 (miss), Greene, Rogers 2509 (TENN); ees Tracy 4699 (MICH, NCU, NY TAES, W); Jackson, Tracy cian mo); Lamar, Sargent 144 (Gu); Pearl River, Amack- -n., 1938 (Missa); Tishomingo, Anonymous s.n., 1937 (MISSA). ARKANSAS: Pulaski, Engelmann 80 (mo); Saline, Moore 321117 (wis). Andropogon glomeratus var. glomeratus is not as aggressive a colonizer as its two hypothetical derivatives, A. glomeratus vars. hirsutior and glaucopsis. Plants in the northern part of the range of var. glomeratus tend to be of shorter stature and to have shorter peduncles. Andropogon glomeratus var. glomeratus and the robust variant of var. pu- milus have inflorescences so similar in shape that most previous workers have united them and have overlooked the differences between them. The robust variant is taller, usually with rather smooth sheaths and with shorter, more ciliate, and darker ligules, narrower raceme sheaths, and lower glume keels that are scabrous below the middle. Although both taxa grow in wet sites, the robust variant is weedier, shows a greater tolerance for drier conditions and various soil types, and has a wider geographic range. 9b. Andropogon glomeratus (Walter) B.S.P. var. hirsutior (Hackel) Mohr, Bull. Torrey Bot. Club 24: 21. 1897. Based on . Andropogon macrourus Mi- chaux var. hirsutior Hackel in DC. Monogr. Phanerog. 6: 409. 1889. Tyre: Alabama, Mobile County, Mohr s.n., 1884 (holotype, w!). An- dropogon virginicus L. var. hirsutior (Hackel) Hitche. Jour. Wash. Acad. 242 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Sci. 23: 456. 1933. Andropogon virginicus L. var. tenuispatheus (Nash) Fern. & Griscom f. Airsutior (Hackel) Fern. & Griscom, Rhodora 37: 142. 1935. FiGures 11, Q; 50. Diacnosis. Inflorescences (linear)-oblong; peduncles and usually spikelets short; stamens usually marcescent within spikelet DistriBuTION. Ditches, swales, bogs, flatwoods, and savannas, often forming large populations in cleared, low ground. Coastal Plain from Maryland to Mississippi (MAP 18) REPRESENTATIVE SPECIMENS. United States. MARYLAND: Somerset, Hermann 9973 (Gu, Ny). VinGINIA: Greensville, Kra/ 14297 (usF); Isle of Wight, Fernald & Long 6760 (Gu, 16247 (Ncu); Chowan, Ahles & Duke 51035 (us); Craven, Godfrey & White 6768 (Gu); Cumberland, Ahles 36560 (Ncu); Duplin, Ahles 35747 (Gu, Ncu); Greene, Radford 40434 (NCU, vDB); Halifax, Ahles & Leisner 20845 (Fsu, NCU, USF); Hertford, Ah/es & Haesloop 52210 (Ncu);, Hoke, Correll 7181] (uc); Hyde, Fosberg 17744 (micu); Johnston, Radford 29144 (NcUu, USF); Lenoir, Radford 31591 (Ncu),; Nash, Godfrey & Kerr 6623 (GH, Mo); Pitt, Radford 41741 (Ncu); Richmond, Radford 19335 (Ncu); Robeson, Ahles 37253 (Ncu); Sampson, Campbell 3996 (Gu); Tyrrell, Radford 42553 (Ncu); Wayne, Burk s.n., 1958 (Ncu); Wilson, Radford 40781 (Ncu). SoUTH CAROLINA: Bamberg, Afhles 37658 (Ncu); Chesterfield, Radford 18679 (Ncu); Clarendon, Radford 30899 (Ncu); Darlington, Coker s.n., 1909 (Ncu); Jasper, Campbell 3964 (GH). Georaia: Berrien, Celarier s.n., 1953 (mo); Brantley, Campbell 4024 (Gu), Colquitt, Godfrey 76081 (Frsu); Grady, Ko- marek s.n., 1977 (Fsu); Terrell, Duncan 1772 (Ga); Thomas, Campbell et al. 4058 (Gu); Ware, Campbell 4141 (GH). FLoripa: Alachua, Chase 4216 (micu); Baker, Campbell 3851 (Gu); Clay, Campbell 4199 (Gu); Duval, A. H. Curtiss 3639b (GA, MISSA, MO); Hillsboro, Combs 1358 (us); Jackson, Campbell 3810 (Gu); Jefferson, Godfrey 75830 (rsu); Lake, Campbell 4131 (Gu), Liberty, Campbell 4074 (Gu); Leon, Silveus 6710-B (GA, TAES); Madison, Godfrey 75833 (Fsu); Martin, Campbell 3875 (Gu); Orange, Combs (Fsu). ALABAMA: Baldwin, Campbell 3804 (Gu); Mobile, Mohr s.n., 1896 (NcU); Wash- ington, Kral 49069 (vps). Mississippi: Harrison, Tracy 4699 (mo). Earlier workers failed to appreciate the great similarity of Andropogon glo- meratus vars. hirsutior and glaucopsis probably because they overlooked the numerous similarities in overall size, inflorescence appearance, and ligule mor- phology, as well as in morphological characters associated with the predomi- nantly cleistogamous flowering mode. The varieties are also remarkably alike in geographic distribution and ecological preferences. They colonize exten- sively, much more so than their closest relative, 4. glomeratus var. glomeratus, and form very dense populations in the moist, cleared gro ound of recently harvested timberlands. Often the two grow together in pop of thousands of individuals. Because they grow together so frequently and are morphologically so alike, the possibility that they are not distinct taxa but merely genetic segregates of one another has been carefully considered. Based on observations of several hundred seedlings grown from seeds from both taxa (growing together in na- 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 243 ture), there is no evidence for genetic segregation. The glaucousness/greenness and pubescence/glabrousness of the stem sheaths are discernible in the seedlings within a few weeks of germination. (Stem-sheath scabrousness often does not develop in the greenhouse in var. hirsutior.) The seedlings consistently match their parents in these two characters. These taxa are unusual in the virginicus complex because their morphological differences are not associated with differences in either ecology or geography. The two taxa are recognized as varieties within the virginicus complex because they are more easily distinguished from one another than are the variants. Andropogon glomeratus var. hirsutior resembles three other taxa besides A. glomeratus vars. glomeratus and glaucopsis. It has long been confused with individuals of the robust variant of 4. glomeratus var. pumilus with a narrow inflorescence. The two can be distinguished by sheath scabrousness, ligule morphology, raceme-sheath width, peduncle length, and lower-glume-keel sca- brousness. Andropogon glomeratus var. hirsutior is narrower in both ecological preference and geographic range than the robust variant. A closer morphological similarity exists between var. hirsutior of the southeastern United States and the southwestern variant of var. pumilus. Indeed, the differences (ligule margin, lower-glume scabrousness, and sometimes inflorescense denseness) are small and are not always easily discerned. Finally, in general aspect, var. Airsutior and the deceptive variant of A. virginicus var. virginicus are very close. They are separated by stem-sheath scabrousness and ligule morphology. 9c. Andropogon glomeratus (Walter) B.S.P. var. glaucopsis Mohr, Bull. Torrey Bot. Club 24: 21. 1897. Based on Andropogon macrourus Michaux var. glaucopsis Ell. Bot. S. Carolina Georgia 1: 149, 150. 1816. Type: South Carolina, Elliott 181 (holotype, CHARL, not seen; isotype,'! PH!). Andro- pogon glaucopsis (Ell.) Nash in Small, Fl. SE. U. S. 63. 1903. Andropogon virginicus L. var. glaucopsis (Ell.) Hitche. Am. Jour. Bot. 21: 139. 1934. Figures 11, R; 54; 55. see eee glaucus Muhl. Descr. uber. Gramin. 278. 1817. Not Andropogon glaucus , 1789. Type: South Carolina, Elliott 18] (holotype, CHARL, not seen faa type, ae “Cymbopogon glaucus (Muhl.) Schultes, Mant. Syst. Veg. 2: 459. 1 Diacnosis. Stem sheaths smooth; leaves glabrous, glaucous; racemes short. DIsTRIBUTION. Flatwoods, bogs, ditches, swamps, pond margins, and swales. Coastal Plain from southern Virginia to Mississippi (Map 19). REPRESENTATIVE SPECIMENS. United States. VIRGINIA: Princess Anne, Fernald & Griscom 2765 (GH, USF). NorTH CAROLINA: Bertie, Campbell 399] (Gx), Bladen, Ahles 37350 (ncu); Carteret, Phipps et al. 3688 (Ncu); Craven, Ahles & Duke 51035 (Ncu); Dare, Blomquist 8081 (Gu); Greene, Radford 40373 (Ncu, uc); Jones, Radford 39837 (Ncv); New Hanover, Canby s.n., 1867 (Ny); Onslow, Moldenke 122 (Ny); Pamlico, Godfrey & He ia eae Pender, Ahles 36208 (ncu); Sampson, Campbell 3995 (Gu), Wake, my - Wayne, Radford 31503 (Ncu). SouTH CAROLINA: Chesterfield, ne 18679 ee u): Dorchester, Ahles & Haesloop 37784 (Ncu); Georgetown, Radford \\Smith (1962) equated the Elliott specimen at PH to the specimen with the same number at CHARL. 244 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 31379 (GA, Ncu); Horry, Duke 0064 (NcU); Jasper, Campbell 3963 (GH). GEORGIA: Brantley, Campbell 4023 (GH); Camden, Duncan & Hardin 14390 (NcU); Charlton, Harper 693 (Nv), Chatham, Mellinger s.n., 1958 (GH); Lowndes, Quarterman 5345 (vps); McIntosh, Duncan 20673 (GH, MICH, NCU, USF, Wis); Thomas, Campbell 4229 (Gu); Tift, Shepard 220 (TAES); Ware, Silveus 5368 (TAES); Wayne, Campbell 4205 (GH). FLORIDA: Alachua, Godfrey & Morrill 52618 (Fsu); Baker, Godfrey 74664 (FSU, NCU, vDB); Bay, Godfrey 76141 (rsu); Brevard, Shuey & Poppleton 1540 (usr); Clay, Campbell 4200 a, Columbia, Combs & Rolf 128 (us); Dixie, Godfrey 56176 (FSU, GA, NY, USF); Duval, . H. Curtiss 6077 (GA, GH, MO, NY); Franklin, Godfrey 71208 (su); Highlands, Pras 14612 (Gu); Indian River Tracy 9255 (TAES); Jackson, Campbell 3812 (Gu); Jeffers Godfrey 74599 (Fsu, NCU, vDB); Lafayette, Godfrey 74631 (Fsu, NCU); Lake, Campi 4132 (Gu), Leon, Campbell & Godfrey 4062 (Gu); Liberty, Campbell 4072 (Gu); Madiso Kral 3760 (Fsu, GH, NCU); Manatee, Tracy 7735 (GH, MO, NY, TAES, W, WIS); Mania. Campbell 4128 (GH); Orange, Campbell 3908 (Gu); Osceola, Ray et al. 10493 (usr): TAES); Santa Rosa, Godfrey 76798 (esu): Taylor, Kral 52127 (vps); Volusia, Hood s.n., 1911 (Ga); Wakulla, Godfrey 64977 (Fsu, vps). ALABAMA: Baldwin, Campbell 3806 (Gu); Mobile, //tis et al. 21367 (wis). Mississippr: Harrison, oe 36240 (FSU, MISSA, TAES, vps); Jackson, Caldwell 368 (Fsu). Andropogon glomeratus var. glaucopsis and both the drylands and wetlands variants of A. virginicus var. glaucus have been combined by some authors (Nash, 1912) and confused by many botanists. Andropogon glomeratus var. glaucopsis differs from A. virginicus var. glaucus in its longer ligules and leaves and can be distinguished from the drylands variant by its pubescence below the raceme sheath, its inflorescence shape, and its habitat. Although these three differences do not hold for var. g/aucopsis and the wetlands variant, the former has shorter raceme sheaths and racemes than the latter. 9d. Andropogon glomeratus (Walter) B.S.P. var. pumilus Vasey, Bot. Gaz. 16: Type: Texas, Val Verde County, Nea//y 256, 1890 (holotype, US, not seen; isotype, Uc!, w!). Andropogon glomeratus (Walter) B.S.P. var. eyes Nash in Small, Fl. SE. 61. 1903. Type:'? Florida, Duval County, A. H. Curtiss 5337, 1894 (lectotype, Ny!; isolectotypes, GA (two sheets)!, Gu!, Ny!, ae Andropogon tenuispatheus (Nash) a : > 113.1912. Andropogon virginicus L. var. tenuispatheus (Nash) Fer Gris Phodor 37: 142. none pause virginicus L. var. hirsutior ate Hitehe. £ tenuispatheus (Nash) Fern. & Griscom, Rhodora 42: 416. 1940. Anatherum virginicum (L.) Sprengel ae Peer ies (Nash) G. Roberty, Bois- siera 9: 213. 1960. Diacnosis. Keels of lower glumes often scabrous to below middle. DistriBuTION. Moist, disturbed sites (e.g., roadsides, fresh or brackish swamps, swales, moist woods, and fields). Virginia to California and south; extremely common through Mexico and Central America to northern South America: common throughout West Indies (Map 20). "One of the many specimens annotated by Nash and at ny has been selected as the lectotype. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 245 Map 20. Distribution of Andropogon glomeratus var. pumilus (dots, robust variant: stars, southwestern variant). 9d. 1. Robust variant Ficures 11, P; 51. DriaGNnosis. Stem sheaths usually smooth; ligules usually short; raceme sheaths narrow DISTRIBUTION. See distribution under variety for ecological preferences of this variant. One of most aggressive and ubiquitous weeds in North America. Throughout distribution of variety except southwestern United States and northwestern Mexico (Map 2 REPRESENTATIVE SPECIMENS. United States. VIRGINIA: Isle of Wight, Fernald & Long 13887 (GH, MO, Ny); Middlesex, Hermann & Martin s.n., 1939 (MICH, NY, TENN); Nan- semond, Fernald et al. 15197 (Gu, NY); Norfolk, Fernald & Griscom 2766 (GH, USF). Jasper, Campbell 3965 as Kershaw, Radford 29957 (NC ie arene Bell 10105 (Nc a McCormick, Radford 30717 (Gu, Ncu); Spartanburg, Bell 10372 (Ncu). GEorGIA: Ber- rien, Celarier A-2600-I (mo, uc); Clarke, Duncan 286 (Ga); Clinch, Faircloth & Cribbs 5003 (GA, Mo, NCU); Harris, Jones 22345 (Ga, Gu); McIntosh, Duncan 20634 (GH, NCU, USF, wis); Pike, Duncan 3064 (GA, MICH, USF); Putnam, ected 4759 (GA, GH, MO, ny); Thomas, Campbell et al. 4055 (Gu); Sumter, Harper 651 (Ny), Washington, Dunc 4342 (GA, MISS). FLoriDA: Alachua, D’Arcy 2166 (GA, vps, wis), Baker, Goel 5850 (Gu); Broward, Stimson 852 (Fsu, NCU, USF); Citrus, Godfrey 65108 (Fsu); Collier, Lakela 27803 (GA, usF); Dade, Gill s.n., 1970 (GH, Mo, UsF); Duval, 4. H. Curtiss 6078 (GA, GH, 246 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 MO, NCU, NY); Escambia, Hansen 2349 (Fsu, wis); Franklin, clea (Fsu); High- lands, Campbell 4117 (cu); Hillsboro, Perdue 1748 (FSU, GH, NCU, TAES, UC, USF); Jackson, Campbell 4030 (GH); Leon, Wooten 2320 (Fsu, vps); Liberty, Goutrey 55529 (Fsu); Madison, Godfrey 74115 (FSU, NCU, vpB); Manatee, Perdue 1788 (FSU, GH, NCU, TAES, uc); Monroe, Campbell 3730 (Gu); Nassau, Godfrey 74691 (Fsu, vpB); Osceola, gee 6106 (GH, Mo, US); Palm Beach, Kra/ 5690 (Fsu, vps); Pinellas, Deam 2767 (mo, Ny); Santa Rosa, Vee & Roe 912 (Ncu, wis); St. Johns, Ward 2305 (Fsu, GH, NCU, USF, vps); Taylor, Godfrey 74145 (Psu, NCU, VDB). ALABAMA: Baldwin, Campbell 3801 (Gu); arbour, Campbell 4152 (Gu); Lee, Jones s.n., 1959 (GA, Ncu); Marshall, Golden s.n., 1974 (vps); Mobile, Tracy s.n., 1896 (Missa); Montgomery, Campbell 3958 (Gu); Sumter, Jones 1730 (NcU). MISSISSIPPI: ve Temple 4669 (miss, NCU); Clarke, Jones 10711 (miss); Hancock, pealad mea GA, GH, MISSA, NCU, ioe Bibs Harrison, Demaree 34532 (NCU, USF, VDB); J on, Dem 4129 Ge. NCU, , USF, VpB); Jasper, Jones 10851 (miss); Jefferson, rae 1171 (us); — "Campbell 3954 (GH); Oktib- a, Tracy s.n., 1890 (miIssA, NY, TAES); Warren, Co/vin s.n., 1938 (missa); Wayne, Campbell 3952 (GH). ARKANSAS: Ashley, ee 18562 (GH, MO, Ny, wis); Bradley, Demaree 21820 (GA, Mo, NY); Desha, Demaree 21627 (NY, TENN, UC, wis); Garland, Demaree 46698 (tars); Howard, Demaree 45229 (rats); Phillips, Palmer 26645 (GH, Mo); Pike, Demaree 9979 (GH, MO, NY, UC, USF); Saint Francis, Demaree 59675 (TENN, vbB). LOUISIANA! aes Sharpe 480 (USF): Caddo, Thomas & Overby 32838 (NCU), Concordia, Thomas & DePoe 510 (NCU, TENN); East Baton Rouge, McCoy s.n., 1952 NY); Iberia, Delos 158 (vps); Jackson, Kral 16073 (Fsu, vps); Natchitoches, Palmer 8865 (Mo, Ny); Orleans, Ewan 18792 (GH, Mo); West Carroll, Demaree 14096 (GH, MO, NY); Vermilion, Reese 3906 (FSU, GH, NCU). OKLAHOMA: Cleveland, Goodman 2362 (GH, MO, NY); Johnston, Waterfall 5685 (mo, Ny); Marshall, Goodman 7389 (uc, wis); McLain, Massey, & Hoisington 1491 (Ncu); Murray, Hopkins 1121] (mo); Payne, Gay 130 (usF), Pontotoc, Robbins 3202 (Ny, TAES, UC). TExAs: Anderson, Gould 7285 (TAES); Austin, Parks & Coy 17704 (GH, ee on Bend, ee wales n., a (TAES): Tharp & Barkley 13006 (GH, NY, UC, vDB); Gregg, York s.n., 1941 (GH, Ny); Hidalgo, Clover 462 (micH); Karnes, Johnson 1107 (raes); Kerr, Cory 52397 (GH, MICH, NY, UC); Motley, Parks & Cory 15964 (GH, TAES); Real, Cory 24353 (Gu); Tarrant, Whitehouse 17320 (micH, NY); Tom Green, Cory 5/03 (GH); Uvalde, Parks & Cory 750] (TAEs); Walker, Cory 50637 (MICH, NY); heen Palmer p02 (MICE MO). Mexico. CHIHUAHUA: Rio Bonito, LeSueur 0100 (CAS, GH, TEX, UC). COAHUILA: Las Delicias, Stewart 2818 (GH). Nuevo Leon: Cola de Caballo, "Bécile M- 414 (FLAS); Monterrey, Smith M593 (TEX). TAMAULIPAS: El Limon, Kenoyer & Crum 3611 (GH, MicH); Gd. Wanteon, Harvey & Witherspoon 9214 (micu); Jaumave, Stanford et al. 2359 (cas, GH). DURANGO: Coyotes, Maysilles 8294 (MICH, TEX, UC); Durango, Palmer 251 (F, GH, MICH, UC). SAN Luis Potosi: Tamazunchale, Edwards 944 (TEx), Fisher 37133 (Gu), Kenoyer A597 (F, MICH). JALISCO: Guadalajara, Palmer 466 (GH, Uc, w); Villa Corona, McVaugh 14415 (micH). MICHOACAN: Volcan Paricutin, Hakala s.n., 1946 (micu). HipALGo: Jacala, Kral 24887 (vps). MEXIco: Ixtapan de la Sal, Harvey 8667 (micH). PUEBLA: Cotimehuaca, Arséne 3542 (Gu); El Carrizal, Pineda s.n., 1968 (CAS, MICH), Arséne 1433 (Gu), Nicolas s.n., 1909 (GH). VERA- cruz: Cordoba, Matuda 330 (micn, vt), Bourgeau 1666 (Gu), Cyasto 330 (vt); Jalapa, Gould 926] (micH, TEX); Nautla, Beetle M-1265 (wis); Papantla, Gutierrez s.n., 1967 (CAS, MICH); Orizaba, Seaton 1/1 (F, GH), Miller 2033 (w); San Salvador de Acajete, Sharp 45622 (GH). OAXACA: San Antonio, Smith 960 (F), Pringle 5565 (Gu, vt). CHIAPAS: Chanal, Kaplan 126chS7 (F); Ixtapa, Laughlin 1076 (cas); La Trinitaria, Breedlove 41938 (cas); Ocozocoantla de Espinosa, Breedlove 37797 (cas); Pueblo Nuevo Solistahuacan, Lathrop 5215 (cas), Breedlove 19931 (cas), Rayon, Breedlove 10159 (cas, F); San Cris- tobal, Breedlove 11915 (F, MicH); Tenejapa, Breedlove 10926 (micH, TEX), Ton 1041 (F, 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 247 MICH); Teopisca, Breedlove 10545 (cas, TEx); Venustiano Carranza, Laughlin 1989 (cas); Yerba Buena, Mi// 572 (cas); Zinacantan, Laughlin 2258 (cas). CAMPECHE: Bolonchen de Rejon, Gould 12626 (uc), Carmen, Kral 25422 (micn, vpB), Rzedowski 26384 (cas, F, MICH, TEX). YUCATAN: Progreso, Swallen 2911 (micu); Sisal, Gould 12641 (uc); Ti- zimin, Gould 12652 (uc). QUINTANA a Lake Chichancanab, Swallen 2729 (micuH); Coba, Lundell 7835 (micu); Payo Obispo, Dampfs.n., 1925 (F). Belize: Buttonwood Cay, Fosberg & Spellman 54411 (F); cen ‘Lundell 4902 (CAS, GH, MICH, TEX), Gentle 134 (micH); Honey Camp, Lundell 428 (F, GH, us); Southwest Cay Is., Fosberg & Stoddard 53860 (us); Tower Hill Estate, Karling 55 (r). Guatemala. ALTA VERAPAZ: Coban, Molina s.n., 1963 (F); Tamahu, Standley 70928 (r). EscuintLa: between Escuintla and Santa Bi alee ooaa Standley 63414 (fF). HUEHUETENANGO: Cuilco, Steyermark 50800 (F, us); Ixcan, Steyermark 49337a (F). IzaBAv: Cristina, Blake 7570 (us), Stey- ermark 38479 (F). E Petén: La Libertad, Lundell 3482 (Gn, Micu, us); Tikal, Contreras 86 (F, TEX). SACATEPEQUEZ: Antigua, Standley 64695 (F). SoLoLA: Volcan Atitlan, Stey- ermark 47475 (F); Volcan San Pedro, Stevermark 47187 (fF). El Salvador: Matapan, Boden s.n., 1900 (mo). Nicaragua: Managua, Garnier 818 (micH); Puerto Cabezas, Ped- erson S.n., 1968 (wis). Costa Rica: Cartago, Weston et al. 3471 (uc); Puntarenas, Pohl & Davidse 10794 (uc); San José, Hitchcock 8465 (micu), Pohl & Davidse 11433 (uc). Panama: Barro Colorado Is., Woodworth & Vestal 542 (Fr, GH); Canal Zone, Hitchcock 8033 (us), Standley 31228 (us); Chiriqui, McCorkle C-53 (Fsu). Bermuda: Manuel 776 (Gu). Bahama Islands: Andros, Northrup 659 (Gu); Cat Is., Byrne 297 (Gu, wis); Grand Bahama, Correll 40635 (TEx), Correll & Kral 42899 (vps); Great Inagua, Dunbar 234 (Gu); South Bimini, Howard 10208 (Gu). Turks and Caicos Islands: Caicos Is., Correll 43117 (Gu). Greater Antilles. Cuspa. Habana: Cajio, Leén 14697 (Gu); Santiago de las Vegas, Wilson 2207 (mic, w). Las Villas: Manajanabo, Ledn 5302 ee Sancti-Spiritus, Leon 3024 (Gu); Soledad, Howard 6230 (Gu, us). Isla de Pinos: Nue A. H. Curtiss 294 a Bed 4453] (us). Camagiiey: Cayo Romano, Shafer 2614 (on). Oren Moa, Brist 1 (Gu); Monte Verde, Wright 1555 (Gu); Sierra de Nipe, Ekman 6376 (MICH). Pee Manchester, Webster & Proctor 5252 (GH, MIcH); Port Antonio, Maxon & Killip 288 (us); St. eae Crosby et al. 825 (GH, TEX, UC). HAITI: Ile de la Tortue, Ekman 4263 (Gu, us); Morne des Commissaires, Holdridge 1324 (MICH, TEX); Port au Prince, Beech 2048 (us), Pater 5008 (Gu); Sur Cayer, Ekman 72 (us). DOMINICAN Repus.ic: Barahona, Howard 8584 (Gu), Fuentes 128] (Gu); Leybo Province, Puenta Icacos, Ekman 15787 (GH, TEX). PUERTO Rico: Isla Verde, ree 11758 (uc). GRAND CayMAN: Brunt 1656 (FLAS, NCU). Lesser Antilles. St. Kitts: Hitchcock 16361 (us). AnTIGUA: Box 128 (us). GUADELOUPE: Questel 1422 (us), nae 3548 (us). MARTINIQUE: Duss 1301 (us). Colombia: San Andros Is., Gentry s.n., 1967 (wis). Venezuela. FALCON: Wingfield 5417 (mo). The robust variant is a remarkably aggressive colonizer in a wide variety of soils. It does not characteristically invade old fields and cleared timberlands with the regularity of Andropogon virginicus or other taxa of A. glomeratus. Instead, it prefers some flow of water in the soil and often chokes roadside ditches with its dense growth (FIGURE 27 In the greenhouse and the experimental garden it has consistently been the fastest-growing member of the virginicus complex. In the field it produces the thickest stems and the largest leaves. It appears to have some adaptations for very rapid growth, but at the same time it is phenotypically plastic in plant height. Under poorer conditions its stems may be very short, although the inflorescence remains densely and profusely branched. The robust variant tolerates a greater range of soil salinity and pH than do 248 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 other members of the virginicus complex. It grows in either brackish or fresh water and in soils derived from either acidic or alkaline rocks. The uniquely wide ecological preferences and geographic range and the dis- tinctive morphology of this variant argue for species ranking. On the other hand, certain individuals are very similar in general appearance to Andropogon glomeratus vars. glomeratus and hirsutior. The ways in which the variant differs from these two varieties have been discussed under the varieties. Furthermore, the next taxon forms a bridge between the robust variant and the rest of A. glomeratus, which makes the species large but coherent. Small or shaded plants of the robust variant may resemble the deceptive variant of Andropogon virginicus var. virginicus in overall size and inflorescence shape. The two taxa differ in stem-sheath pubescence, peduncle length, and lower-glume-keel scabrousness. 9d. 2. Southwestern variant FIGuRE 52. Diacnosis. Stem sheaths always scabrous: ligules long: spikelets often exceeding 4 mm in length. DIsTRIBUTION. Moist soils of seepage slopes and edges of springs. Utah to California, south to northwestern Mexico (Map 2 REPRESENTATIVE SPECIMENS. United States. New Mexico: Grant, Wright 2700 (GH, Mo, Ny, UC). ARIZONA: Santa Cruz, Barr 63-504 (UsF). NEvADA: no locality given, Wheeler g.n., 1871 (GH). CALIFORNIA: Inyo, Thorne & Tilforth 42512 (Ny); Los Angeles, Moxley 632 (Mo, UC); Marin, Howells.n., 1939 (ny, uc); Placer, Crampton 5802 (TAES); Riverside, Reed 2901 (CAL, LA); San Bernardino, Roos 5023 (Ny, Uc); Shasta, Bacigalupi ied (Gu); Ventura, Pollard s.n., 1946 (mo, Ny, Uc); Yolo, Beetle 4691 (TAES, UC, Wis). Mexico. BAJA CALIFORNIA Norte: Cerro la Encantada, Chambers 544 (cas, uc); Santa au Moran 11492 (cas, Uc). BAIA CALIFORNIA Sur: Mission de San Pedro Martin Valley, Wiggins 9036 (GH) In the past this variant has been called Andropogon glomeratus. It is distin- guished here for the first time as a morphologically and geographically isolated group of populations of 4. glomeratus var. pumilus. Its chief taxonomic sig- nificance is that it shares some features (sheath scabrousness and ligule length) with the first two varieties of 4. g/omeratus, and others (peduncle length and scabrousness of the lower glumes) with the robust variant; it thus holds the species together. It appears to be closest to the robust variant because the two variants share the unique feature of scabrousness of the lower glume keels, a character generally more conspicuous in the southwestern variant. Their oc- currence in the western United States and their preference for seepage condi- tions are further bases for combining them into one variety. Finally, while United States populations are fairly easily separated from the rest of var. pumilus, some populations in northern Mexico are more or less intermediate etween the two variants. More collections and especially more field work in this area will be necessary to assess how these variants interact. ACKNOWLEDGMENTS It isa pleasure to acknowledge the following people for help through exchange of ideas or critical readings of parts of this work: P. D. Cantino, E. A. K. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 249 Coombs, M. J. Donoghue, M. W. Frohlich, C. W. Greene, C. H. Haufler, W. S. Judd, N. G. Miller, R. C. Rollins, B. G. Schubert, E. A. Shaw, S. A. Spongberg, P. F. Stevens, E. L. Taylor, R. M. Tryon, and C. E. Wood, Jr. I also wish to thank Mike Canoso, Walter Kittredge, and the staff of the Harvard University Herbaria for the administration of herbarium loans, and the curators of the following herbaria (abbreviations follow Holmgren & Keu- ken, 1974) for sending loans: Botanical Museum and Herbarium, Copenhagen (c); Charleston Museum (CHARL); California Academy of Sciences (cAs); Duke University (DUKE); Field Museum of Natural History (F); University of Florida (FLAS); Florida State University (rsu); University of Georgia (GA); University of California at Los Angeles (LA); University of Michigan (mMicH); University of Mississippi (miss); Mississippi State University (Missa); Missouri Botanical Garden (mo); University of North Carolina (Ncu); New York Botanical Garden (Ny); Philadelphia Academy of Natural Sciences (pH); Texas A&M University (TAES); University of Tennessee (TENN); University of Texas (TEX); University of California at Berkeley (uc); U. S. National Herbarium (us); University of South Florida (usF); Vanderbilt University (vps); University of Vermont (vT); Naturhistorisches Museum, Vienna (w); University of Wisconsin (wIs). Dr. Robert K. Godfrey, of the Tall Timbers Research Station, helped by sharing his observations about andropogons. I wish to thank Bob and his wife, Nell, for making a home for my wife and me for many days. Thanks are also due to Mr. and Mrs. Angus K. Gholson, Jr., of Chattahoochie, Florida, and to Dr. and Mrs. Walter S. Judd, of Gainesville, Florida, for their hospitality. Angus and Walt pointed out many good collecting sites. The six months of field work in the southeastern United States was funded by the Anderson and Fernald funds of Harvard University and by the National Science Foundation (Dissertation Improvement Grant number DEB-77-17317). The skill and diligence of my brother Philip are reflected in the photographic plates. Finally, for her assistance in the field, support at home, and perseverance through the years, I thank my wife, Pegze. LITERATURE CITED ALLARD, R. W. 1975. The mating system and microevolution. Genetics 79: 115-126. Bazzaz, F. A. 1975. Plant Sa diversity in old-field successional ecosystems in southern Illinois. Ecology 56: 4 8. Bootu, W. E. 1964. Agrostology. 222 pp. Edwards Brothers, Ann Arbor, Michigan. Brown, W. V. 1949. A cytological study of cleistogamous Stipa leucotricha. Madrofio 10: 97-107 1950. A cytological study of some Texas Gramineae. Bull. Torrey Bot. Club -76. Cimeeen, C. S. 1980. Biosystematic studies in the alae eee complex (Gramineae). 217 pp. Unpubl. Ph.D. thesis, Harvard Unive 1982a. Hybridization between Andropogon cus var. r. pumilus and A. longiberbis (Gramineae) in central Florida. Brittonia 34: 146-150. 1982b. Cleistogamy in Andropogon (Gramineae). Am. Jour. Bot. 69: 1625— 1635. 1983. Wind aispeisal of some North American species of Andropogon (Gra- miineae): Rhodora —72 CARNAHAN, H. L., & H. D. Hick, 1961. Cytology and genetics of forage grasses. Bot. Rev. 27: 1-162. 250 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 CELARIER, R. P. 1956. Cytotaxonomy of the Andropogoneae I. Subtribes Dimeriinae and Saccharinae. Cytologia 21: 272-291. . Cytotaxonomy of the Andropogoneae II. Subtribes Ischaeminae, Rott- boelliinae, and the Maydeae. /bid. 22: 160-183. 1958. Cytotaxonomy of the Andropogoneae III. Subtribe Sorgheae. /bid. 23: 395-41 8. 1959. Cytotaxonomy of the Andropogoneae IV. Subtribe Sorgheae. bid. 24: 285-303. CHAPMAN, R.H., &S. B. JONes. 1975. Se ra aaa Andropogon virginicus (Gramineae). Bull. Torrey Bot. Club 102: 166- Cuase, A. 1937. Notes on types of North nee grasses. Am. Jour. Bot. 24: & C. D. Nites. 1962. Index to grass species. 3 vols. 1743 pp. G. K. Hall, Boston. CHAUDRA, N., & N. SAXENA. 1964. Morphological studies in the Gramineae 5. Vascular anatomy of the spike and spikelets of the Andropogoneae. Proc. Indian Acad. Sci. B. 59: 457-474. Cuurcu, G. L. 1936. Cytological studies in the Gramineae. Am. Jour. Bot. 23: 12-15. . 1940. Cytotaxonomic studies in the Gramineae. [bid. 27: 263-271. CLayTon, W. D. 1964. Studies in the Gramineae: V. New species of Andropogon. Kew Bull. 17: 465-470. 1967. Andropogon pterophilis Hook. Ic. Pl. 37: t. 3644. —. 1969. A revision of the genus Hyparrhenia. Kew Bull. Add. Ser. 2: 1-196. . 1972. Studies in the Gramineae: XX XI. The awned genera of Andropogoneae. Kew Bull. 27: 457-474. CLiFFORD, H. T. 1961. Floral evolution in the family Gramineae. Evolution 15: 455- Connor, H. E. 1979. Breeding systems in grasses: a survey. New Zealand Jour. Bot. 7: 547-574. Cotton, R., & C. A. Stace. 1977. Morphological and anatomical variation of Vulpia (Gramineae). Bot. Not. 130: 173-187. Davipse, G., & R. W. Pout. 1972. Chromosome number, meiotic behavior, and notes on some grasses from Central America and the West Indies. Canad. Jour. Bot. 50: 1441-1452. DusarpIn, M. 1978. Chromosome numbers of some tropical African grasses from western Zaire. Canad. Jour. Bot. 56: 2138-2152. FERNALD, M. L., & L. Griscom. 1935. Three days of botanizing in southeastern Virginia. Rhodora 37: 129-157, 167-189. Go..ey, F. B. 1965. Structure and function of an old-field broomsedge community. Ecol. Monogr. 35: 113-137. GouLp, F. W. 1956. Chromosome counts and cytotaxonomic notes on grasses of the tribe Andropogoneae. Am. Jour. Bot. 43: 395-404. | . The grass genus Andropogon in the United States. Brittonia 19: 70-76. & : SODERSTROM. 1970. Chromosome numbers of some Mexican and Colombian grasses. Canad. Jour. Bot. 48: 1633-1639 GouLp, S. J. 1977. Ontogeny and ae 501 pp. Belknap Press, Harvard Uni- versity, Cambridge, Massachuse GraHaM, A. 1973. History of the arborescen temperate element in the northern Latin American biota. Pp. 301-314 in A. GRAHAM, ed., Vegetation and vegetational his- tory in northern Latin pene Elsevier Publ. Co., New York. GuTTIERREZ, M., V. E. GRACEN, & G. E. EpwArps. 1974. Biochemical and cytological relationships ; in C, plants. Planta 119: 279-300. HackeL, E. 1889. Seo aad In: A. L. P. P. pe CANDOLLE & C. DE CANDOLLE, Monogr. Phanerog. 6: 1-716. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 251 Hartiey, W. 1950. The global distribution of the tribes of the Gramineae in relation to historical and environmental factors. Austral. Jour. Agr. Res. 1: 355-373. 1958. Studies on the origin, evolution, and oe of the Gramineae. I. The tribe Andropogoneae. Austral. Jour. Bot. 6: —-128. HesLop-HARRISON, J. 1961. The function of the ce = and the control of cleistogamy in Bothriochloa decipiens (Hack.) C. E. Hubbard. Phytomorphology 11: 378-383. Hitcucock, A. S. 1951. Manual of the grasses of the United States. ed. 2 (revised by A. Cuase). 1054 pp. U. S. Dept. Agr. Misc. Publ. 2 Hotmaren, P. K., & W. KEUKEN. 1974. Index herbaniorume Part I. The herbaria of the world. Reg. Veg. 92: 1-397. KANNENBERG, L. _& R. W. ALLARD. 1967. Population studies in predominantly ee species. VII. Genetic variability in the Festuca microstachys com- plex. Evolution 21: 227-240. KeEEveR, C. 1950. cae of succession on old fields of the Piedmont, North Carolina. Ecol. Monogr. 20: 229-250. Keno, Y. L. 1939. The gross morphology of the Andropogoneae. Sinensia 10: 274— 3 Krat, R. 1976. Additions to some notes on the flora of the southern states, particularly Alabama and middle Tennessee. Rhodora 78: 438-456. Laess_e, A. M. 1958. The origin and successional relationships of sand-hill vegetation and sand-pine scrub. Ecol. Monogr. 28: 361-387. LeLonc, M. G. 1965. Studies on the reproduction and variation in some Panicum subgen. Dichanthelium. 226 pp. Unpubl. Ph.D. thesis, lowa State University. Linnaeus, C. 1753. Species plantarum. Ray Society facsimile of the first ed. 2 vols. 1200 pp. B. Quaritch, London. 1957, 1959. Lone, R. W., & O. LaKkeLa. 1971. A flora of tropical Florida. 962 pp. Univ. Miami Press, Cora eee Masry, T. J., K. R. MARKHAM, & M. B. THOMAS. oo The systematic identification of eee. 354 pp. Springer-Verlag, New Mercacre, C. R. 1960. Anatomy of the anes eine I. Gramineae. 731 pp. Clar- endon Press, Oxford. Mou Lensrock, R. H. 1973. The illustrated flora of Illinois. Grasses. eo to Danthonia. 378 pp. Southern Illinois Univ. Press, Carbondale, Illinoi Murray, B.G. 1974. Breeding systems and floral biology in the genus Bry Heredity Nasu, G. V. 1903. Poaceae. Pp. 60-65 in J. K. SMALL, Flora of the southeastern United States. Published by the author, New Yo rk. Paropi,L.R. 1949. Lasg lgé Deschampsia. Darwiniana 8: 415-475. PoHL, R. W. ae How to know the grasses. ed. 3. 200 pp. Wm. C. Brown Co., Dubuque, low. &G. ee 1971. Chromosome numbers of Costa Rican grasses. Brittonia 23: 293-324. Raprorpb, A. E., H. E. AHLES, & C. R. Bett. 1964. Manual of the vascular flora of the Carolinas. 1183 pp. Univ. North Carolina Press, Chapel Hi Roperty, G. 1960. Monographie systématique des Andropogoneae du globe. Boissiera 9: 1-455. SCRIBNER, F. L. 1901. The grasses in Elliott’s “Sketch of the botany of South Carolina and Georgia.” U.S. D. A. Div. Agrost. Circ. 29: 1-1 SmitH, C. E., Jk. 1962. Henry Muhlenberg— botanical pioneee Proc. Am. Philos. Soc. 106: 4 443-460. StapF, O. 1919. Gramineae. /n: D. Prain, ed., Fl. Trop. Africa 9: 208-265. STEBBINS, G. L. 1957. Self-fertili and population variability in higher plants. Am. Nat. 91: 337-354. Zo2 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Flowering plants. Evolution above the species level. 379 pp. Belknap Press, Harvard University, Cambridge. TATEOKA, T. 1962. A cytological study of some Mexican grasses. Bull. Torrey Bot. Club 89: 77-82. Tompsett, P. B. 1976. Factors affecting the flowering of Andropogon gayanus Kunth. Responses to photoperiod and growth regulat ors. Ann. Bot. 40: 695-705. WEATHERBY, C. A. 1942. A list of type specimens in Elliott’s herbarium. Rhodora 44: 249_ WEATHERWAX, P. 1929. Cleistogamy in Poa chapmaniana. Torreya 29: 123, 124. Wet, J.M.J.pe. 1954. Chromosome numbers ofa few South African grasses. Cytologia 19: . 103. 60. Chromosome numbers and some morphological attributes of various oa African grasses. Am. Jour. Bot. 47: 44-49. APPENDIX A. Glossary. CALLUS—the very base of the spikelet, covered with hairs in all members of the vir- ginicus complex (Ficures |, 2, 8, and 9). COLLAR—the region of the leaf at he junction of the blade and the sheath. DISPERSAL UNIT—the spikelet and the attached rachis internode and pedicel (FIGURES , 8, and 9). INFLORESCENCE— the portion of the stem producing reproductive structure INFLORESCENCE UNIT—the racemes, the peduncle, and the subtending raceme sheath (FIGuRE I1) LIGULE—a membranaceous flange on the adaxial surface of the leaf at the collar (Fic- URE PEDICEL—a Short axis attached at its base to the rachis internode and with or without a vestigial spikelet (FIGURE PEDICELED SPIKELET— the spikelet at the apex of the pedicel; nonfunctional and reduced or absent in the virginicus complex (FiGure 8) PEDUNCLE—the axis dela by the raceme sheath and bearing at its apex two or more racemes (FIGURE 11) RACEME— the linear series of spikelets connected by the rachis internodes (FiGure I 1). “Rame,” designating a structure with sessile and pediceled spikelets, has been used in Andropogon (Pohl, 1978) RACHIS INTERNODE—the axis connecting the spikelets into racemes (FIGURE 8). APPENDIX B. Rules for determining distance between character states.* I. For characters of type a ((low range) mean (high range); characters 1, 7, 8, 9, 29, and 31), type b (low range—-high range; characters 10, 26, and 27), and type c ((low range) low mean-high mean (high range); characters 19, 20, 23, and 25), compare the ranges of types a and b and the intervals between the low and high means of type c, then assign a value based on the following conditions: O, if the range of each overlaps more than half the range of the other, or the range of one is totally contained within the range of the other. Y, if the range of one is less than or equal to half the range of the other. 1, if the range of each overlaps that of the other by less than hal II. For yee ee with two or three states (characters 2, 3, 5, 6, 11, 13, 18, 24, 28, 30, and 32 0, if identical for all states. ‘if different in one or more but not all states. 1, if no states are shared. *See TaBLes 4-6. 1983] CAMPBELL, ANDROPOGON VIRGINICUS COMPLEX 253 Ill. For the two pubescence ase (4 and 17): O, if identical in half or more of their states but different in less than two. '», if identical in less ieee half of their states and different in two or or more. 1, if no states are shared. : For character 12: 0, if identical in all stat Y, if identical in some aie but different in others. 1, if not identical 1 in any States. . For character 0, 1f overlap is “half the total range or mo %, if overlap is less than half the total rang or if 1-2 vs. 2-3 or 2-3 vs. 3-4. 1, if no overlap, or if 1-2 vs. 2-5 or 2-3 v VI. For character 15: 0, if overlap is half the ae, a or more, or if the range of one lies entirely within the range of the ‘4, if overlap is less than i a more than 0 or if Sa ti of two ranges are a5) < < equal (e.g., 1-2 vs. 2-3, 2-4, 2-5, or 2-6; 2-3 v 1, if'no overlap, or - 3 vs. 3-7, 11. VII. For character 16: 0, if the mean of each falls within me range of the other. VIII. For character 22 there are three classes: A is 2 (3 or 4); B is 2-3, 2-4, or 2-5(-7); d C is 2-9 or 2-13. Distance is based on the following conditions: 0, if classes are acl %, if class A v 1, if class A or 'B. vs. C. . For character 21, there are three classes: A, if Mae of means is less than 10 (e.g., (2)4-6(12)); B, if range of means includes 10 (e.g., (1)6-14(60)); and C, if range of means is — than 10 (e.g., (9)26-66(15)). Ae is based on the following condition 0, if cee are identical. % if class A vs. B or B vs. C. 1, if class A v . For character 33, ‘there are three classes: A, mean less than 10%; B, mean 10-50%; and C, mean greater than 50%. Distance is based on the following conditions: 0, if classes are identical. % if class A vs. B or B vs. C. 1, ifclass A vs. C. x * APPENDIX C. Ambiguous names. ene belvsii Desv. Opusc. Sci. Phys. Nat. 67. 1831. Type: no locality cited: Les e based on an unnamed figure (f. 23, fig. 4) of Palisot de Beauvois, Essai : eae 1812. Although Jackson (Index Kew. 1: 124. 1893) equated this with ie. pogon argyreus Schultes, and Hitchcock (1951) equated it with Andropogon ternarius Michaux, the illustration alone is not enough to determine the placement of this name. Andropogon loui. Steudel, he ies Glum. 1: 383. 1854. The description (““spica solitaria’ ”) does not fit: ihe sae Opogon \ <. Although Chase (1937) said the type of this name (P) is a “small, over- emanie sin oe A. virginicus,” in Hitchcock (1951) the name was doubtfully referred to 4. wirginicus L. Andropogon virginicus L. subvar. ditior Hackel in DC. Monogr. Phanerog. 6: 411. 1889. Types: Florida, Duval County, 4. H. Curtiss 3639d (Curtiss distributed three species under this number: Andropogon glomeratus (Walter) B.S.P. var. pumilus Vasey, the 254 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 robust variant (FLAS!, GA, MISsA!), A. virginicus L. var. virginicus, the deceptive variant (GH!), and A. longiberbis Hackel ee NY (two sheets)!), none of which bears any indication that Hackel used them); Mexico, Veracruz, Schiede 937 (not seen; Chase & Niles (1962) lege this collection in Andropogon bourgaei Hackel); and Mexico, Liebmann 72 (not se Hackel’s description could apply to taxa of Andropogon glomeratus, A. virginicus, or A. longiberbis APPENDIX D. Synonymy of Andropogon ternarius Michaux. Andropogon ternarius Michaux, Fl. Bor. Am. 57. 1803. Type: Carolina (holotype, p, not seen; photo, GH!). a. Andropogon ternarius Michaux var. ternarius Andropogon argenteus Ell. Bot. S. Carolina Georgia 1: 148. 1816. Type: no locality given (CHARL, not seen, placed in this taxon by Scribner (1901)). Sor- ghum argenteum (Ell.) Kuntze, Rev. Gen. Pl. 2: 790. 1891. Andropogon argyreus Schultes, Mant. Syst. Veg. 2: 450. 1824, nomen superfl. for Andropogon argenteus Ell. and with the same type. Anatherum argyraeum (Schultes) G. Roberty, Boissiera 9: 209. 1960. Andropogon muhlenbergianus Schultes, Mant. Syst. Veg. 2: 455. 1824, ex char. Type: South Carolina, not seen. Andropogon elliottii Chapman, Fl. So. U. S. 581. 1860. Based on Andropogon argenteus Ell., 1816. Not pe ieee DC., 1813. Sorghum elliottii (Chapman) Kuntze, Rev. Gen. | l. Andropogon argyreus Schultes var. orate va. Contr. U. S. Natl. Herb. 3: 12. 1892, ex char. Type: Texas, not see Andropogon argyreus enue var. macra Scribner, U. S. D. A. Div. Agrost. Bull. 1: 20. 1895. Tyre: Florida, Duval County, A. H. Curtiss 4592, 1894 (holotype, not seen) nag? a a in synonymy of Andropogon ternarius Michaux by Hitchcock, ge elliottii eee var. glaucescens Scribner, Bull. Torrey Bot. oe 23: | 1896. Type: Florida, Lake County, Nash 473 (holotype, us, not s es ak Andropogon ternarius Michaux var. glaucescens (Scribner) Fern & Griscom, Rhodora 37: 137, | ; ele scribnerianus Nash, Bull. New York Bot. Gard. 1: 432. 1900. Based Andropogon elliottii Chapman var. glaucescens Scribner. Not Andropogon es HBK., 1815. Andropogon mississippiensis Scribner & Ball, U. S. D. A. Div. Agrost. Bull. 24: 41.1901, ex char. Type: Mississippi, Harrison County, 7racy 38/8 (holotype, not seen). b. Andropogon ternarius Michaux var. cabanisii (Hackel) Fern. & Griscom, Rhodora 1935. Based on Andropogon cabanisii Hackel, Flora 68: 133. 1885. ED* Fon da, Gulf County, Chapman s.n., 1884 as w!). Sorghum ca- ae (Hackel) Kuntze, Rev. Gen. Pl. 2: 791. 189 DEPARTMENT OF BOTANY AND PLANT PATHOLOGY UNIVERSITY OF MAINE Orono, Maine 04469 *Hackel cited ‘Pennsylvania et Florida leg. Cabanis (in Herb. reg. Berolin.); Florida prope Ap- palachicola leg. Chapman.” A specimen of the latter description from the Hackel herbarium is here selected as a lectotype. 1983] HOWARD, AUBLET Za) THE PLATES OF AUBLET’S HISTOIRE DES PLANTES DE LA GUIANE FRANCOISE RICHARD A. HOWARD IN THE COURSE Of work on an unpublished manuscript of Alexander Anderson (a “hortus” of the botanical garden on St. Vincent, West Indies), reference to the drawings of Aublet’s Histoire des Plantes de la Guiane Francoise was frequently necessary. Anderson, the second director of the botanical garden (1785-1811), collected in Trinidad and Tobago (1787, 1790) and the Guianas (1791). In describing the specimens or the plants later grown from seeds, he made frequent references to and comparisons with the work of Aublet. I have been unable to find any list of modern equivalents for the taxa in Aublet’s work. The alphabetic list that follows was compiled from the existing literature in which Aublet names are mentioned or his specimens are cite In the second edition of Taxonomic Literature (1976), Stafleu and Cowan supplied several references to the life of Aublet and to the publication of his Histoire des Plantes. The commentaries of Lanjouw and Uittien (1940), Souili- jaert and Stafleu (1953), Leandri (1968), and Bernardi (1976) are most useful. Froidevaux (1897), ina paper generally overlooked, has given Aublet’s itinerary together with supporting documentation, both in considerable detail. Aublet’s herbarium was purchased by Sir Joseph Banks, along with his drawings and notes. These are now all in the British Museum (Natural History); the notes and drawings have been bound separately and are housed in the library. Banks apparently shared the specimens with Linnaeus fi/ius, whose collections are in the J. E. Smith herbarium at the Linnean Society. Additional collections are in several herbaria in Paris (Adanson, Jussieu, Rousseau) as well as at Stock- holm, Copenhagen, and Vienna, with fewer specimens at Brussels, the Field Museum, Liverpool, Missouri, Uppsala, and possibly Geneva. In designating types, holotypes, or lectotypes most recent authors have cited specimens at the British Museum, although a few holotypes have been attributed to Paris, Vi- enna, and Geneva. Herbarium specimens have been sought (BM, LINN, P) as far as time permitted and are cited for each entry, either as supporting or typotype specimens or as designated in the literature or on annotation labels. It is apparent from the articles studied that many of the drawings published by Aublet are mixtures, and in some cases the descriptions are composites of discordant elements. Many of Aublet’s descriptions, therefore, have been emended by later authors, usually by the exclusion of the fruits or flowers. In a few cases, however, the illustration and description of the fruit represent the emended species description. Cuatrecasas, Johnston, Niedenzu, Sandwith, and © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 255-292. pee 1983. 256 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 others have rejected Aublet’s names as nomina dubia or because the supporting specimens were mixtures. This procedure is no longer permissible under the tape Code of Botanical Nomenclature. In many cases these names can r have been typified by either excluding or accepting particular portions of aa or the plate. In some instances designation of type material has resulted in new combinations being proposed in this paper—e.g., Cordia tet- raphylla Aublet = Buchenavia tetraphylla (Aublet) Howard, replacing Buche- navia capitata (Vahl) Eichler. The suggestion that Aublet used native collectors and then described what was presented to him may well account for the mixture of materials. As will be indicated later, the error may also have occurred in the exchange between the artist and the engraver. Stafleu (1971) noted that Aublet differed from many of the Linnaean school in his use of native names for the genera he proposed. He noted that other botanists almost immediately substituted more acceptable names for those proposed by Aublet, and that 27 of Aublet’s names appear in the list of nomina conservanda. This can hardly be considered criticism of Aublet’s work. Of the 576 genera in the publication, six of Aublet’s proposed generic names are conserved: Ouratea unnecessarily, Houmiri and Vochy with orthographic changes, and only Bacopa, Rourea, and Simarouba on the basis of priority of publication. Of the 21 rejected names, one (Cananga) is a homonym: all the others apparently resulted from the proposer’s preference for a better-known or more widely used name. The substitute names suggested by Scopoli (1777) and Schreber (1789, 1791) were adopted by a few then-contemporary authors. The specimens in the J. E. Smith herbarium, for example, bear the names used by Schreber. The equivalence of the Scopoli and Schreber names to those used by Aublet will be considered in a separate paper by Dr. James Zarucchi. Stafleu and Cowan (1976) note that the publication of the Histoire was announced in March, 1775, and that copies were issued in June, 1775, but they suggest that publication extended over a period of time. In Index Nomi- norum Genericorum Plantarum (Farr et al., 1979) the date of publication of June, 1775, is assumed. Gleason, Williams, and Henrey (see entries Fothergilla and 7amonea) have called attention to the fact that in some copies of the work the name Tamonea was used twice, while in others one entry was corrected to Fothergilla, clearly indicating that not all copies resulted from a single printing. Henrey (1975) has called attention to marks for the binders that indicate re- placement pages. Indexes to Latin and French names appear in volume two of the Histoire following the main body of the text and preceding the supplement in which additional species are described. Reference to the taxa described in the sup- plement appears in the index, however. The plates of the supplement, which are numbered 368-392, complete the second volume of plates. In the list that follows there are many examples of Aublet’s specific name for a taxon differing between the text, the index, and/or the plates. Most com- monly, these involved the name of the country. Aublet was consistent in his spelling of “guianensis” in the text, and this has been adopted here as the correct spelling of that specific name. However, on the plates the same specific name has in most cases been spelled “‘guyannensis,” and that spelling has been 1983] HOWARD, AUBLET 257 adopted in some modern floras or monographs. Other orthographic or typo- graphic variations on “‘guyannensis” are reported. The majority of the genera described by Aublet are monotypic. When several species were listed, Aublet gave a specific name in the text for all of them, except in the genera Aeschynomene, Apocynum, Jussiaea, Myrtus, Periploca, and Tragia, where species were often described in the text with supporting literature citations but without a specific epithet. For most of these unnamed taxa, a specific epithet is given in the index. Urban (1919, 1920) examined these and made the necessary additions of author or combination, as required. Thus, Anechites nerium (Aublet) Urban, Eugenia pomifera (Aublet) Urban, Gonolobus scandens (Aublet) Urban, Myrcia citrifolia (Aublet) Urban, and Seshania emerus (Aublet) Urban have basionyms in which the specific epithet appears in the index in Aublet’s work but not in the text. None of these taxa is illustrated. Periploca umbellata Aublet is antedated by Echites corymbosa Jacq. and is known today as Forsteronia corymbosa (Jacq.) Meyer. One name remains obscure: Tragia scandens Aublet (based on Plumier’s Nova Pl. Gen. Amer. 14. 1703 and Pl. Amer. (Burmann, ed.), 250, ¢. 252, fig. 1. 1755) was not placed by Urban (Symb. Antill. 8: 363. 1920). A number of Aublet’s taxa have not been satisfactorily placed. Among the new genera he described, the following do not have satisfactory modern place- ment: Courimari (Elaeocarpaceae?), Macahanea (Celastraceae?), Managa (Ce- lastraceae?), Meborea (Euphorbiaceae?), Senapea (Passifloraceae?), Singana (Leguminosae?), and Tontelea (Celastraceae?). Aublet’s names for the monotypic genera Aberemoa (Annonaceae), Evea (Rubiaceae), and Ropourea (Flacourtiaceae) are nomina rejicienda—a desig- nation that should be reconsidered on the basis of modern identifications of the taxa. In compiling the following list of modern equivalents for the taxa illustrated in Aublet’s plates, I have examined the comprehensive studies by Hallier (1918), Lanjouw and Uittien (1940), and Sandwith (1937, 1942, 1962, 1963). The many treatments in Pulle’s Flora of Suriname (1932-1975) offered quick initial reference and were further checked in Lemée (1952-1956). More recent works of value in this search have been the several monographs in the Flora Neotropica series, as well as treatments by Maguire and collaborators in the “Botany of the Guayana Highland” series in the Memoirs of the New York Botanical Garden (1953-1978)—especially Steyermark’s study of the Rubi- aceae (Maguire, 1972), which differs so drastically from Bremekamp’s earlier work in Pulle’s Flora of Suriname. Other monographs and special papers are referred to in the text. The following alphabetic listing of species in Aublet’s Histoire provides reference to their description in the text (volumes | and 2, or the supplement) and their illustration in the plates. When the spelling of the specific name in the index or on the plate is different from that in the text, the variant spelling is given in parentheses. ““NG” indicates that a new genus is described, and “TS” a type species that has been determined either by monotypy or by des- ignation in Index Nominum Genericorum Plantarum (Farr et al., 1979) or the cited reference. The currently acceptable family designation is followed by the assignment used in a current flora, monographic treatment, or general reference, 208 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 as cited. Generally the author making the combination can be the point of further reference, with the place of publication available in /ndex Kewensis if not indicated. The herbarium specimens I have seen are cited BM, LINN-SM with a number, p-A with a number (Adanson), p-J with a number (Jussieu), and p-R with a volume and folio number (Rousseau). A few of the specimens attributed in the literature to the British Museum were on loan and could not be verified. I was able to examine the complete Rousseau herbarium and those bundles of the Jussieu herbarium from which Aublet specimens were cited in the index, but only a very few bundles of the Adanson herbarium. The last often contained the most complete specimens, occasionally with extensive holographic notes. The Adanson collection (loose specimens in folders in bundles) should be consulted before any other specimen is automatically considered for lectotyp- ification. The Aublet specimens appeared to be in folders containing an un- deciphered numbering system on the upper right-hand corner of the face of the folder. The extensive spirit collection at the British Museum was not ex- amined but is known to contain fruiting material associated with some of Aublet’s specimens. Aublet’s notes obtained by Sir Joseph Banks and bound separately are not complete; the missing ones should be sought in the Adanson herbarium. The original drawings (BM) are also bound separately and do not constitute a complete set. The drawings do not always match the final engraved plate, and the fruits, flowers, or dissections may be on separate sheets of paper, suggesting that the mixtures reported for some plates may be due to errors in supplying drawings to the engravers. ALPHABETIC LISTING OF THE AUBLET PLATES ABEREMOA GUIANENSIS (1: 610. t. 245 (guyvannensis))— NG, TS—Annonaceae = Guatteria guianensis (Aublet) R. Fries. Duguwetia St. Hil. is a conserved generic name (#2680), with the type (D. lanceolata St. Hil.) not conserved; Aberemoa Aublet is listed as rejected. Fries transferred the single species Aberemoa guianensis Aublet to Guatteria but has also described Duguetia guianensis R. Fries as a distinct species. In a key Hutchinson (1964, p. 78) distinguished Duguetia as having the carpels sessile or nearly so in fruit and the indumentum stellate or lepidote (as appears to be the case in Aublet’s plate 245), while Guatteria has the carpels more or less stipitate in fruit. If Aberemoa guianensis is a species of Guatteria, then Aublet’s name need not be listed as a nomen rejiciendum. A sterile specimen (BM) cannot be placed to genus. However, when Dunal (Monogr. Annon. 126. 1817) published Guatteria aberemoa based on the Aublet description and plate, he stated that a specimen at p was the Aublet species but with a smaller, short, stipitate fruit. If Aublet’s Aberemoa guianensis is indeed a species of Duguetia, a new specific name is required. ABLANIA GUIANENSIS (1: 585. ¢. 234 (guyannensis))—NG, TS—Elaeocarpa- ceae = Sloanea guianensis (Aublet) Bentham. Bentham did not cite spec- imens in making the transfer to S/oanea. BM, LINN-SM 962.1. 1983] HOWARD, AUBLET 259 ABUTA AMARA (1: 620. t. 25/)—Aristolochiaceae = Aristolochia glaucescens H.B.K. Jackson (Index Kew. 1: 4. 1893) listed Abuta amara as equaling Aristolochia glaucescens. The Aublet plate illustrates a sterile specimen, seemingly an Aristolochia but clearly not a species of Abuta (Menisper- maceae). Lemée (1955, p. 549) accepted this assignment credited to “I.K.” and later (1956, p. 32) noted that he had seen a herbarium specimen. The Aublet name “‘amara” is older, and a transfer and a new combination in Aristolochia are necessary if herbarium material can be located and does verify the identification. Specimens could not be located at BM or P. ABUTA RUFESCENS (1: 618. ¢. 250) —NG, TS—Menispermaceae = Abuta rufes- cens Aublet. For notes on the lectotypification of Aublet specimens, see Barneby and Krukoff (Mem. New York Bot. Gard. 22(2): 37-40. 1971). Specimen at BM marked “holotype” by Barneby and Krukoff, LINN-sM 1554.16; p-J 10831. ACIOA GUIANENSIS (2: 698. ¢. 280 (guyannensis))—NG, TS—Chrysobalana- ceae = Acioa guianensis Aublet. BM. ACOUROA VIOLACEA (2: 753. t. 301)—NG, TS—Leguminosae. Dalbergia L. f. is a conserved name (#3821), with the type species D. lanceolaria L. f. (not conserved; from Sri Lanka); Acouroa Aublet is listed as rejected. No modern disposition of the Guianan Acouroa violacea Au- blet has been located. BM (not found). ADIANTUM GUIANENSE (2: 963. ¢. 365 (guyannense))— Lindsaeaceae = Lindsaea guianensis (Aublet) Dryander. Specimen at BM marked “‘type.” ADIANTUM SAGITTATUM (2: 964. t. 366 (sagitatum))—Lindsaeaceae = Lindsaea sagittata (Aublet) Dryander. Kramer (Acta Bot. Neerl. 6: 260. 1957) an- notated a specimen at BM as isotype but did not see a specimen at Pp that he felt would be the holotype. AIOUEA GUIANENSIS (1: 311. ¢. 120 (guyannensis))—NG, TS—Lauraceae = Aiouea guianensis Aublet. In 1973 Kostermans annotated p-j 4260 as the isotype. Renner (Fl. Neotrop. 31: 110. 1982) cited “holotype, BM.” \MAIOUA (Suppl. 13. t. 375 (guyannensis))—NG, TS— Rubiaceae = Amaioua guianensis Aublet. Specimen at BM labeled ‘‘type.’ AMANOA GUIANENSIS (1: 256. ¢. /0/ (guyannensis))—NG, TS—Euphorbi- aceae = Amanoa guianensis Aublet. BM, P-R 6: 229. AMBELANIA ACIDA (1: 266. ¢. 104)—NG, TS— Apocynaceae = Ambelania acida Aublet. Specimen at Bm marked “holotype”; LINN-SM 440/1; p-R 7: 242. ANIBA GUIANENSIS (1: 327. t. 126 (guyannensis))—NG, TS— Lauraceae = Aniba guianensis Aublet. Kubitzki (Fl. Neotrop. 31: 45. 1982) cited “holotype BM, isotype Cc.” ANNONA AMBOTAY (1: 616. ¢. 249)—Annonaceae = Annona ambotay Aublet. BM, LINN-SM 983.5. ANNONA LONGIFOLIA (1: 615. ¢. 248)— Annonaceae = Fusea longifolia (Aublet) Saff. Bm, P-J 10770. 260 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ANNONA PALUDOSA (1: 611. ¢. 246)—Annonaceae = Annona paludosa Aublet. BM, LINN-SM 983.4, p-J 10771. ANNONA PUNCTATA (1: 614. ¢. 247)—Annonaceae = Guatteria punctata (Au- blet) Howard, comb. nov. Robert Fries (Acta Horti Berg. 10: 314. 1931) suggested Annona punctata Aublet was a Guatteria, but he did not make the combination. Later (bid. 12: 434. 1939) he listed “Annona punctata Aublet PI. Guian. franc. 1 (1775) 614, tab. 247 (fruct. excl.)” as a synonym of Guatteria chrysopetala (Steudel) Miq., which is based on A. chrysopetala Steudel (Flora 26: 754. 1843). Lemée (1956, p. 34) also accepted this assignment. A new combination in Guatteria is required for the earlier basionym. BM. APALATOA SPICATA (1: 383. ¢. 147 (Opalatoa))—NG, TS— Leguminosae = Cru- dia spicata (Aublet) Willd. Robert Williams (Fl. Trinidad Tobago 1: 256. 1931) determined that “‘Aubl. t. 147, F 56 [error for 5, 6] is Pterocarpus rohrii Vahl. Crudia spicata (Aublet) Willd. is therefore typified by the flowering specimen in the plate, excluding the fruit and seed shown. BoM, LINN-SM 743.1 APEIBA ASPERA (1: 545. f¢. 2/6)—Tiliaceae. According to Uittien (in Pulle, 1932, 3(1): 52, 53, 440), the fruit (fig. 2) is that of Apeiba echinata Gaertner, and the flowering branch is referred to Apeiba glabra Aublet. Farr et al. (1979, p. 107) state that a type species has not been designated for Apeiba. The genus is currently under study by Meyer, who will make a designation. BM. APEIBA GLABRA (1: 541. ¢. 214 (/aevis))—Tiliaceae = Apeiba glabra Aublet. pm, LINN-SM 929.6 APEIBA PETOUMO (1: 543. t. 2/5)—Tiliaceae = Apeiba petoumo Aublet. Uittien (in Pulle, 1932, 3(1): 51) cited A. petoumo Aublet “but for the fruit” and A. aspera Aublet “as to the fruit” in the synonymy of Apeiba echinata Gaertner (Fruct. Semin. Pl. 2: ¢. 12/7. 1791). Lemée (1952, p. 348) had a similar treatment. Sandwith (1931, p. 180) accepted Apeiba petoumo Au- blet (excl. fructu) and placed 4. echinata Gaertner emend. Uittien in syn- onymy. BM. APEIBA TIBOURBOU (1: 538. ¢. 273)—Tiliaceae = Apeiba tibourbou Aublet. pM, LINN-SM 929.3. APOcYNUM Acoucl (1: 274. ¢. 107 (Apocinum))— Apocynaceae = Forsteronia acouci (Aublet) DC. BM, p-R 10: 343. APOCYNUM UMBELLATUM (1: 275. ¢. 108)— Apocynaceae = Forsteronia umbel- lata (Aublet) Woodson. p-R 10: 343 AROUNA GUIANENSIS (L: 16. ¢. 5 (guyannensis))—NG, TS— Leguminosae = Di- alium guianensis (Aublet) Sandw. Specimen at BM labeled “type speci- men’; LINN-SM 69.1; P-R 1: 33. ARUBA GUIANENSIS (1: 294. t. 115 (guyannensis))—NG, TS—Simaroubaceae = Simaba guianensis Aublet. Bm, p-R 11: 349 BACOPA AQUATICA (1: 129. ¢. 49 jane NS. TS—Scrophulariaceae = Bacopa aquatica Aublet. BM, LINN-sM 279, ] 1983] HOWARD, AUBLET 261 BAGASSA GUIANENSIS (Suppl. 15. t. 376 (guvannensis))—NG, TS— Moraceae = Bagassa guianensis Aublet. Bo. BAILLIERIA ASPERA (2: 804. ¢. 317)—NG, TS—Compositae = Clibadium suri- namense L. BM, LINN-SM 1340.3. BANARA GUIANENSIS (1: 548. ¢. 2/7 (guyannensis))— NG, TS—Flacourtiaceae = Banara guianensis Aublet. Specimen at BM annotated “‘holotype”’ by Sleu- mer; LINN-SM 852.1. BANISTERIA QUAPARA (1: 464. ¢. 186)—Malpighiaceae = Hiraea quapara (Au- blet) Sprague. Niedenzu (Pflanzenr. IV. 141: 132. 1928) used the name Hiraea multiradiata Juss., rejecting the Aublet name on the basis of a mixed plate and excluding the fruit. The flowering branch illustrated by Aublet therefore typifies the species, and Sprague’s combination is to be used. BM, LINN-SM 824.3. BANISTERIA SINEMARIENSIS (1: 462. ¢. 185)—Malpighiaceae = Mascagnia si- nemariensis (Aublet) Griseb. Niedenzu (Pflanzenr. IV. 141: 113. 1928) noted that the fruit in Aublet’s plate had to be excluded. He rejected the Aublet name on the basis of a mixed plate and used the name Mascagnia volubilis (Sims) Niedz. (based on Malpighia volubilis Sims (1805)). Since Niedenzu typified the Aublet name by excluding the fruit, the combination by Grisebach is appropriate. BM, LINN-SM 824.5. BASSOVIA SYLVATICA (1: 217. ¢. 85)—NG, TS—Solanaceae = Solanum sylva- ticum (Aublet) Bitter. Lemée (1953, p. 395) accepted Solanum aubletii Pulle, although this illegitimate name was placed in synonymy by Bitter in 1921. BM, P-R 5: 189. BAUHINIA GUIANENSIS (1: 377. ¢. 145 (guyannensis))— Leguminosae = Bauhinia guianensis (Aublet). BM. BAUHINIA OUTIMOUTA (1: 375. t. 144)—Leguminosae = Bauhinia guianensis Aublet. Lemée (1952, p. 85) accepted Bauhinia rubiginosa Bong., noting as synonyms “B. coronata Benth., peut-étre B. outimoutou Aubl. selon J. Amshoff: S. Amer. Papil. 31 [1939].” Wunderlin (pers. comm.) considers it to be the same as B. guianensis Aublet and designated a ‘“‘holotype”’ (w) and an isotype (BM). Wunderlin considers Bauhinia rubiginosa Brongn. to be a much different plant. BM. BEGONIA GLABRA (2: 916. ¢. 349)—Begoniaceae = Begonia glabra Aublet. Le- mée (1953, p. 98) accepted Begonia scandens Sw. (when this name was published in 1788, B. glabra Aublet (1775) was included as a synonym). BM. BEGONIA HIRSUTA (2: 913. ¢. 348)— Begoniaceae = Begonia hirsuta Aublet. BM. BERTIERA GUIANENSIS (1: 180. ¢. 69 (guyannensis))—NG, TS— Rubiaceae = Bertiera guianensis Aublet. Specimen at BM labeled “‘type’’; P-R 8: 272. BESLERIA COCCINEA (2: 632. ¢. 255)—Gesneriaceae = Drymonia coccinea (Au- blet) Wiehler. LINN-sM 1058.3. BESLERIA INCARNATA (2: 635. ¢. 256 (Beslaria))— Gesneriaceae = Rechsteineria incarnata (Aublet) Leeuwenberg. Specimen at BM labeled ‘‘holotype” by 262 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Leeuwenberg but annotated Sinningia incarnata (Aublet) Denham by Wiehler in 1976. BESLERIA VIOLACEA (2: 630. t. 254 (Beslaria))— Bignoniaceae = Schlegelia vio- lacea (Aublet) Griseb. BM. BIGNONIA ALBA (2: 653. f. 266)—Bignoniaceae = Memora alba (Aublet) Miers. Sandwith (Candollea 7: 251. 1937) called Bignonia alba a “nomen du- bium” and concluded that it “‘must certainly not be identified with Memora bracteosa.”” Jackson (Index Kew. 1: 303. 1893) had referred Bignonia alba to Spathodea bracteosa DC., which was transferred to Macfadyena brac- teosa by Bentham and Hooker (Gen. Pl. 2: 1035). Memora bracteosa (DC.) Bur. ex K. Schum. was accepted by Pulle (Fl. Suriname 4(2): 51. 1938), without the citation of B. alba. Alwyn Gentry (pers. comm.) accepts Me- mora alba (Aublet) Miers as a “perfectly good species closely related to M. flavida. There is even a type specimen at the British Museum. It consists only ofa leaf but looks like a good match for the illustration. It is probable that Sandwith had never seen material of this taxon, but it has recently been re-collected twice by Oldeman.... In my opinion there is really nothing dubious about it at all.”” Specimen at BM labeled “type” by A Gentry. BIGNONIA COPAIA (2: 650. ¢. 265)—Bignoniaceae = Jacaranda copaia (Aublet) . Don. BM material on loan. BIGNONIA ECHINATA (2: 648. tt. 263, figs. I, 2; 264)—Bignoniaceae. Alwyn Gentry (Ann. Missouri Bot. Gard. 60: 910. 1973) noted that the fruit is Pithecoctenium echinatum (Jacq.) Baillon, while the leaves and flowers are Arrabidaea inaequalis (DC. ex Splitg.) K. Schum. BM. BIGNONIA FLUVIATILIS (2: 655. t. 267)—Bignoniaceae. Sandwith (Rec. Trav. Bot. Néer. 34: 225. 1937) noted, ‘“‘the wisest course is to treat his name [B. fluviatilis] as nomen dubium.” However, Sandwith (in Pulle, 1938, 4(2): 69) listed B. fluviatilis Aublet in the syn- onymy of Tabebuia aquatilis (E. Meyer) Sprague and Sandwith “‘partim.” No specimen located at BM. BIGNONIA INCARNATA (2: 645. tt. 261, 262)—Bignoniaceae = Mansoa kerere (Aublet) A. Gentry var. incarnata (Aublet) A. Gentry. Sandwith (Meded. Bot. Mus. Utrecht 40: 205. 1937) discovered discordant elements in the description, plates, and specimens of Aublet and recommended a status of nomen dubium. He listed (in Pulle, 1938, 4(2): 28) B. incarnata as a synonym of Cydista aequinoctialis (L.) Miers. Alwyn Gentry (Ann. Mis- souri Bot. Gard. 60: 890. 1973) listed the Aublet name as a synonym of Pachyptera kerere (Aublet) Sandwith but later (/bid. 66: 783. 1979) rec- ognized the variety. He did not see a type specimen. BIGNONIA KERERE (2: 644. ft. 260 (cherere))—Bignoniaceae = Mansoa kerere (Aublet) A. Gentry. Sandwith (Meded. Bot. Mus. Utrecht 34: 219. 1937) discussed the “faulty and deficient” description of the Aublet species and excluded the description of the fruit and figures I-4 of plate 263. Alwyn Gentry (Ann. Missouri Bot. Gard. 66: 783. 1979) proposed the assignment 1983] HOWARD, AUBLET 263 recognized here and cited Aublet s.n. (BM, P-A 12271, w) without desig- nating a lectotype. LINN-SM 1048.5. BLAKEA QUINQUE-NERVIA (1: 525. t. 2/0)—Melastomataceae = Bellucia gros- sularioides (L.) Triana. BM, LINN-SM me BOCOA PROUACENSIS (Suppl. 38. ¢. 397 (provassensis))—NG, TS—Legumino- sae = Bocoa prouacensis Aublet. Richard Cowan (Proc. Biol. Soc. Wash- ington 87: 95-128. 1974) accepted the genus Bocoa and published a re- vision. He indicated that the “type collection” 1s Aublet s.n. (BM). Such a specimen (BM) is annotated “holotype” by Cowan. BoMBAX GLOBOSA (2: 701. ¢. 28/)—Bombacaceae = Eriotheca globosa (Aublet) Robyns. Robyns (Bull. Jard. Bot. Bruxelles 33: 124. 1963) cited the Aublet plate as the holotype. No specimens have been located. BUTTNERIA SCABRA (1: 241. ft. 96)—Sterculiaceae = Byttneria scabra L. BM, P-J 12464, p-R 5: 171 CABOMBA AQUATICA (1: 321. t. 124)—NG, TS—Cabombaceae = Cabomba aquatica Aublet. p-) 3194. CACAO GUIANENSIS (2: 683. f. 275 (guyannensis))—Sterculiaceae = Theobroma guianense (Aublet) Gmelin. Cuatrecasas (Contr. U.S. Natl. Herb. 35: 379- 614. 1964) published a taxonomic revision of the genus Theobroma. Under rejected names and excluded species (pp. 586, 587) he summarized earlier notes in his paper, stating, “Consequently, Cacao guianensis Aublet is not a true species, but a mixture of three species. Therefore the Aublet ‘species’ and name has to be rejected as ‘nomen confusum’ (articles 63 (3), 65, and 66 of the Code of Nomenclature). The name Theobroma guianense has never been consistently used in monographs and general books for any known species.” He had noted (Jbid., p. 586), ““The flowers were described from specimens of Theobroma cacao L.; the branches and leaves from T. subincanum Martius. ... The illustration of this fruit (P/. 275, figs. 16 & 17) agrees ainmistakably well with only one recent collection of Theobroma fruits, namely that from French Guiana by Benoist, which is the type of T. velutinum Benoist.... At the British Museum there are preserved Au- blet specimens and among them type specimens of Cacao guianensis which are syntypes. In 1954, I could identify by close examination an herbarium sheet (with foliage, one flower and an immature fruit) as belonging to 7. subincanum Mart.. this specimen agrees with the description and drawings given by Aublet for the leaves and branchlets; this evidence could easily incline us to use this specimen as lectotype for C. guianensis. But in the carpological collection of the British Museum there is a fragmentary fruit, also a syntype of Aublet’s species, which belongs to 7. cacao.” Cuatrecasas has clearly lectotypified Cacao guianensis Aublet by the leaves and branch- es of the specimen at BM and has excluded the flowers and fruit of Aublet’s plate 275. The name Theobroma guianense (Aublet) Gmelin should be used. CACAO SYLVESTRIS (2: 687. ¢. 276)—Sterculiaceae = Theobroma sylvestris (Au- blet) Martius. Cuatrecasas (/bid., p. 387) stated, “Cacao sylvestris was 264 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 pictured from concordant parts of foliage and fruit which agree well with the species at present known as 7. subincanum, but the flowers mentioned in his description were taken from 7. cacao.” Later (pp. 563, 564) in the same work, Cuatrecasas considered Theobroma subincanum Martius (in Buchner, Repert. Pharm. 35: 23. 1830) and listed as synonyms Cacao sylvestris Aublet and Cacao guianensis Aublet “pro parte (tantum folia).” It appears that Cacao sylvestris can be typified by the foliage and that the flowers should be excluded; the specific epithet “sylvestris” should be used. BM, LINN-SM 1237.2. CACOUCIA COCCINEA (1: 450. t. 179) —NG, TS—Combretaceae = Combretum cacoucia (Baillon) Exell, not Combretum coccineum Lam. (1785). BM, P-J 13637 CALINEA SCANDENS (1: 556. ¢. 22/)—NG, TS—Dilleniaceae = Doliocarpus spraguei Cheeseman, not D. scandens Poiret (1812). BM, LINN-sM 911.1. CAMERARIA TAMAQUARINA (1: 260. ft. 102)—Apocynaceae = Malouetia tama- quarina (Aublet) DC. BM, LINN-SM 448.1, P-R 6: 225, CANANGA OUREGOU (1: 608. ¢. 244)—NG, TS—Annonaceae = Guatteria oure- gou (Aublet) Dunal. Cananga (A. P. de Candolle) Hooker & Thomson is a conserved generic name (#2684), and Cananga Aublet is rejected. Robert Fries (Acta Horti Berg. 12: 352. 1939) accepted Dunal’s combination. Specimen at BM annotated as Guatteria podocarpa Dunal, which was pub- lished later. CARAIPA ANGUSTIFOLIA (1: 562. ¢. 224, fig. 4 (angustifolia))—Chrysobalana- ceae = Couepia spp. Kubitzki (Mem. New York Bot. Gard. 29: 131. 1978) noted that Prance suggested the correct generic placement, but neither was able to place the Aublet name to species. Specimen at BM annotated Coue- pia sp. aff. carvophylloides R. Br. by Prance. CARAIPA LATIFOLIA (1: 561. ¢. 224, fig. 3)—Chrysobalanaceae = Licania mem- branacea Sagot ex Lanessan, not L. /atifolia Bentham. Kubitzki (Mem. New York Bot. Gard. 29: 131. 1978) suggested this placement. BM. CARAIPA LONGIFOLIA (1: 561. ¢. 223, fig. 3)—Chrysobalanaceae = Licania alba (Bernoull1) Cuatr., not Licania longifolia Benoist. Kubitzki (Mem. New York Bot. Gard. 29: fous oo placement. See also Sandwith (Kew Bull. 1931: 373; 1942: 53). B CARAIPA PARVIFOLIA (1: 561. ¢. 223, fig. I )—NG, TS—Bonnetiaceae = Caraipa parvifolia Aublet. Kubitzki (Mem. New York Bot. Gard. 29: 122. 1978) designated the specimen with fruit at p as the holotype; the other specimens he saw (BM, G) were only leaves. CARAPA GUIANENSIS (Suppl. 32. t. 387 (guyannensis))—NG, TS— Meliaceae = = Carapa guianensis Aublet. Specimen at 8m marked “holotype.” CARAPICHEA GUIANENSIS (1: 168. ¢. 64 (guyannensis)) NG, TS—Rubiaceae = Psychotria ligularis (Rudge) Steyerm. var. carapichea (Poiret) Steyerm., not Psychotria guianensis Raeuschel. Cephaélis Sw. is a conserved generic name (#8411), and Carapichea Aublet is rected: Steyermark considered Cephaélis a synonym of Psychotria. BM, P-R 8: 265. 1983] HOWARD, AUBLET 265 CARICA SPINOSA (2: 908. t. 346)—Caricaceae = Jacaratia spinosa (Aublet) DC. BM. CASSIA APOUCOUITA (1: 379. ¢. 146)—Leguminosae = Cassia apoucouita Au- blet. Specimen at BM marked “type.” CASSIPOUREA GUIANENSIS (1: 529. ¢. 2/1 (guyannensis))—NG, TS—Rhizo- phoraceae = Cassipourea guianensis Aublet. BM. CATINGA AROMATICA (1: 512. Index, p. 6 (Catringa). t. 203, figs. 2, 3)—Myr- taceae = Eugenia feijoa Berg. See McVaugh (Mem. New York Bot. Gard. 18: 181-183. 1969) for discussion. No supporting specimens were located. CATINGA MOSCHATA (1: 511. Index, p. 6 (Catringa). t. 203, fig. I)N—NG, TS— Myrtaceae = ? Calycorectes bergii Sandwith. For discussion of this taxon, see Sandwith (Kew Bull. 1932: 212. 1932) and McVaugh (Mem. New York Bot. Gard. 18: 222. 1969). No supporting specimens were located. CHRYSOPHYLLUM MACOUCOU (1: 233. t. 92)—Sapotaceae. Although Lemée (1953, p. 256; 1956, p. 95) accepted this species, Cron- quist (Bull. Torrey Bot. Club 73: 311. 1946) had called ita nomen confusum based “‘on fruit of Chrysophyllum cainito and foliage of some other plant.” BM (sterile specimen clearly not a Chrysophyllum). CIPONIMA GUIANENSIS (1: 567. t. 226 (Siponima guyannensis))—NG, TS— Symplocaceae = Symplocos guianensis (Aublet) Gitircke. BM, LINN-SM 1245.3. CrpuRA PALUDOSA (1: 38. t. 13) —NG, TS—Iridaceae = Cipura paludosa Au- blet. BM material on loan, but a specimen was photographed by Meyer (#4059); p-R 1: 52. CLEOME GUIANENSIS (2: 675. ¢. 273 (guyannensis))—Capparaceae = Cleome gui- anensis Aublet. p-J 11227. COFFEA GUIANENSIS (1: 150. ft. 57 (guyannensis))—Rubiaceae = Faramea sp.” Lemée (1953, p. 540) listed the Aublet name as a synonym under a ques- tionable and undesignated species of Faramea. This cannot be Faramea guianensis (Aublet) Bremek., which has as a basionym Evea guianensis Aublet. BM, P-R 6: 211 labeled “type.” COFFEA PANICULATA (1: 152. ¢. 58)—Rubiaceae. Lemée (1953, p. 542) listed this as “Coussarea paniculata ?,” but he also cited Coussarea paniculata (Vahl) Standley. The Aublet name is not placed satisfactorily. BM, P-R 6: 211. COMBRETUM LAXUM (1: 351. ¢. 1/37)—Combretaceae = Combretum rotundifo- lium Rich., not Combretum laxum Jacq. BM. COMMELINA HEXANDRA (1: 35. ¢. /2)—Commelinaceae = Dichorisandra hex- andra (Aublet) Standley. p-R 1: 49. CONAMI BRASILIENSIS (2: 927. ¢. 354)—NG, TS—Euphorbiaceae = Phyllanthus brasiliensis (Aublet) Oken. BM. CONCEVEIBA GUIANENSIS (2: 924. Index, p. 9 (Couceveiba). t. 353 (Couceveiba guyannensis))—NG, TS—Euphorbiaceae = Conceveiba guianensis Au- blet. BM. 266 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 CONOBEA AQUATICA 639. t. 258)—NG, TS—Scrophulariaceae = Conobea aquatica Aublet. CONOHORIA FLAVESCENS a. 239. t. 95)—NG, TS—Violaceae = Rinorea flaves- cens (Aublet) Kuntze. See Sandwith (1931, p. 172). Specimen at BM labeled “holotype”; p-R 5: 172 labeled “type” by Hekking. CONVOLVULUS GLABER (1: 138. ¢. 53)—Convolvulaceae = Merremia glaber (Aublet) Hallier f. p-r 6: 197. CONVOLVULUS GUIANENSIS (1: 136. ¢. 52 (guyannensis))—Convolvulaceae = Jacquemontia guianensis (Aublet) Meisner (as guyvanensis). No supporting specimens located. CoRDIA COLLOCOCA (1: 219. ¢. 86 (collo-cocus))—Boraginaceae = Cordia no- dosa Lam., not C. collococca L. Lamarck cited Aublet, ¢. 86, and indicated that material from Richard had been included in the original description of Cordia nodosa. BM, LINN-SM 374.5, P-R 5: 181 CORDIA FLAVESCENS (1: 226. ¢. 89)—Lauraceae. Ivan Johnston (Jour. Arnold Arb. 16: 44. 1935) noted that this species was based on mixed material of Cordia (Boraginaceae) and Ocotea (Laura- ceae) and rejected the name. Figures 1-3 represent a Cordia. The specimens he saw (BM, S) represented only Ocotea. Clearly this name represents some species of the Lauraceae not used in Pulle or Lemée. The Bm specimen was not located, although it was photographed by Meyer (#3784). LINN-SM 374.6 consists of foliage plus one fruit and several calyces, which might be identified by a specialist on the Lauraceae. CORDIA TETRANDRA (1: 222. ¢. 87)—Boraginaceae = Cordia tetrandra Aublet. CORDIA TETRAPHYLLA (1: 224. ¢. 88)—Combretaceae = Buchenavia tetraphylla (Aublet) Howard, comb. nov. Ivan Johnston (Jour. Arnold Arb. 16: 45. 1935) stated that the description is based on a mixture of “flowers of some species of Cordia and the leafy fruiting branches of Buchenavia capitata (Combretaceae) and so the name is to be rejected.”” The plate clearly illustrates a stem of Buchenavia capitata showing sympodial branching. Only figures 1-3, showing flowers “‘of some species of Cordia,” need to be excluded. Buchenavia capitata (Vahl) Eichler was based on Bucida capitata Vahl (1796). Regrettably, no material was located, so the Aublet plate (excluding the flowers) represents the lectotype. CoRDIA TOQUEVE (1: 228. ¢. 90)— Boraginaceae = Cordia toqueve Aublet. Spec- n at BM annotated by I. M. Johnston as Cordia heterophylla Willd. and filed under that name. Later (Jour. Arnold Arb. 16: 21. 1935) Johnston recognized Cordia toqueve Aublet and indicated that C. heterophylla Willd. was a synonym. p-R 5: | COUBLANDIA FRUTESCENS (2: 937. f. 356) —NG, TS— Leguminosae = Muellera frutescens (Aublet) Standley. Muellera L. f. (1781) is a conserved generic name (#3837), and Coublandia Aublet is rejected. However, the type species of Muellera monoliformis L. f. is not conserved, and the necessary com- bination was made by Standley in 1933. This correction is not in the list 1983] HOWARD, AUBLET 267 of conserved names in the Rules of Nomenclature. Bentham (Jour. Linn. Soc. Bot. 4(Suppl.): 117. 1860) stated, “Coublandia frutescens Aublet PI. Guian. t. 356 & fr. in t. 300 sub Dugueliae pictus.” I doubt if the fruit drawn in plate 300 is indeed Muellera frutescens. BM, LINN-SM 1225.2. COUEPIA GUIANENSIS (1: 519. ¢. 207 (guvannensis))—NG, TS—Chrysobalana- ceae = Couepia guianensis Aublet. BM COUMA GUIANENSIS (Suppl. 39. ¢. 392 (guyannensis))—NG, TS—Apocyna- ceae = Couma guianensis Aublet. BM COUMAROUNA ODORATA (2: 740. ¢. 296 (odora))—NG, TS—Leguminosae = Dipteryx odorata (Aublet) Willd. Dipteryx Schreber is a conserved generic name (#3845), and Coumarouna Aublet is rejected. The conserved type species is Dipteryx odorata (Aublet) Willd. BM, LINN-sM 1177.2. Coupout AQUATICA (Suppl. 16. t. 377)—NG, TS—Rubiaceae = Duroia aqua- tica (Aublet) Bremek. The generic name Duroia L. f. is conserved (#8316) against Pubeta L. Aublet did not give a specific epithet in the description of the genus but did in the Index (p. 9) and on the plate. Miers (Apocyn. S. America, p. 15. 1878) recognized the discordant elements in the protero- logue of Aublet and proposed the name Cupirana aubletiana Miers for the fruit. Bremekamp (Rec. Trav. Bot. Néerl. 31: 268-272. 1934), in trans- ferring the specific name to Duroia, also discussed the Aublet material and illustration. BM. COURATARI GUIANENSIS (2: 724. ¢. 290 (guyannensis))—NG, TS—Lecythida- ceae = Couratari guianensis Aublet. No material was located. COURIMARI GUIANENSIS (Suppl. 27. ¢. 384 (guyannensis))— NG, TS—Elaeocar- paceae = Sloanea sp. (?). Hallier (1918, p. 24) suggested the identification but did not make the combination. This is not S/oanea guianensis (Aublet) Bentham (Jour. Linn. Soc. 5(Suppl. 2): 69. 1861), which is based on Ablania guianensis Aublet. Although C. E. Smith did not consider the Aublet name in his monograph of the New World species of S/oanea (Contr. Gray Herb. 175: 1-112. 1945), he stated (pers. comm.) that the flowers (figs. /, 2), the fruit section (fig. 3), and the infructescence are not of a species of Sloanea. The material at BM is sterile, and the leaves do not appear to represent a Sloanea. COUROUPITA GUIANENSIS (2: 708. ¢. 282 (guyannensis))—NG, TS—Lecythi- daceae = Couroupita guianensis Aublet. No supporting material was lo- cated. COUSSAPOA ANGUSTIFOLIA (2: 956. t. 363)—Moraceae = Coussapoa angustifo- lia Aublet. Specimen at Bm annotated “holotype” by Akkermann. COUSSAPOA LATIFOLIA (2: 955. f. 362)—NG, TS— Moraceae = Coussapoa la- tifolia Aublet. Specimen at Bm annotated “holotype” by Akkermann. COUSSAREA VIOLACEA (1: 98. f. 38) -NG, TS— Rubiaceae = Coussarea violacea Aublet. BM, P-R 4: 13 COUTAREA SPECIOSA (1: 314. ¢. 122)—NG, TS—Rubiaceae = Coutarea hex- andra (Jacq.) Schum. p-s 9927. 268 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 COUTOUBEA RAMOSA (1: 74. ¢. 28)—Gentianaceae = Coutoubea ramosa Aublet. P-R 3: 114. COUTOUBEA SPICATA (1: 72. f. 27)—NG, TS—Gentianaceae = Coutoubea spi- cata Aublet. p-r 3: | CRENEA MARITIMA (1: 523. ¢. 209)—NG, TS—Lythraceae = Crenea maritima Aublet. BM, LINN-SM 851.1. CROTALARIA GUIANENSIS (2: 761. ¢. 305 (guyannensis))— Leguminosae = Cli- toria guianensis (Aublet) Bentham. Specimen at BM annotated “‘lectotype” by Fantz. CROTON GUIANENSE (2: 882. ¢. 339 (guyannense))—Euphorbiaceae = Croton guianensis Aublet. BM. CROTON MATOURENSE (2: 879. ¢. 338)—Euphorbiaceae = Croton matourense ublet. BM CURATELLA AMERICANA (1: 579. ¢. 232)—Dailleniaceae = Curatella americana L. BM, LINN-SM 960.1. CyYTISUS VIOLACEUS (2: 766. ¢. 306)— Leguminosae = Eriosema violaceum (Au- blet) G. Don. BM. DEGUELIA SCANDENS (2: 750. ¢. 300) NG, TS—Leguminosae = Derris ptero- carpus (DC.) Killip, not Derris scandens Bentham. Derris Lour. 1s a con- served generic name (#3838), and Deguelia Aublet is rejected. Bentham’s comment on the fruit (Jour. Linn. Soc. Bot. 4(Suppl.): 117. 1860) does not seem valid. BM. DRACONTIUM PENTAPHYLLUM (2: 837. Index, p. 10 (hexandra). t. 326 (Dra- cuntium hexandra))— Araceae = Anthurium pentaphyllum (Aublet) G. Don. No supporting specimen located. ENOUREA CAPREOLATA (1: 587. t. 235) —NG, TS—Sapindaceae = Paullinia cap- reolata (Aublet) Radlk. No supporting specimen located. EPERUA FALCATA (1: 369. ¢. 142)—NG, TS—Leguminosae = Eperua falcata Aublet. Richard Cowan (Smithson. Contr. Bot. 28: 26. 1975) has com- mented on the lectotypification. Lectotype at Bm (detached inflorescence only). EUGENIA COUMETE (1: 497. ¢. 196 (comete))—Myrtaceae = Myrcia coumete (Aublet) DC. BM, LINN-sM 883.9. EUGENIA GUIANENSIS (1: 506. ¢. 20/1 (guyvannensis))— Myrtaceae = Myrcia gui- anensis (Aublet) DC. BM, LINN-SM 883.22, p-J 13882. EUGENIA LATIFOLIA (1: 502. ¢. 199)— Myrtaceae = Eugenia latifolia Aublet. See Sandwith (Kew Bull. 1932: 212). Bm, p-s 13920. EUGENIA MINI (1: 498. ¢. 197)— Myrtaceae = Eugenia biflora (L.) DC. Bo. EUGENIA MONTANA (1: 495. ¢. 195)—Myrtaceae = Marlierea montana (Aublet) mshoff. B EUGENIA SINEMARIENSIS (1: 501. ¢. 198)—Myrtaceae. Lemée (1953, p. 155) listed this taxon in the synonymy of Eugenia cof- feifolia DC. with a question mark. McVaugh (Mem. New York Bot. Gard. 1983] HOWARD, AUBLET 269 18(2): 219. 1969) was unable to place the Aublet name without locating a specimen and thought it “unwise to take up the name for the species now called E. coffeifolia.” Specimen at BM annotated by Amshoff as Eugenia arivoa Aublet and filed in that folder; LINN-sM 883.12. EUGENIA TOMENTOSA (1: 504. ¢. 200)—Myrtaceae = Myrcia tomentosa (Aublet) DC. BM EUGENIA UNDULATA (1: 508. ¢. 202)—Myrtaceae = Eugenia undulata Aublet. BM, LINN-SM 883.20, p-J 13921. EUPATORIUM PARVIFLORUM (2: 797. t. 315)—Compositae = Mikania parviflora (Aublet) Urban. Koster (in Pulle, 1938, 4(2): 126) listed E. parviflorum Aublet as a synonym of Mikana amara (Vahl) Willd., which is based on Eupatorium amara Vahl (1794). Urban (Symb. Antill. 5: 223. 1907) re- ferred the Barbados plant cited by Vahl to Eupatorium latifolia Sm. and accepted Aublet’s Guiana plant as distinct. BM. EUPATORIUM TRIFLORUM (2: 795. ¢. 314)—Compositae = Piptocarpha triflora (Aublet) A. W. Bennett ex Baker. A type specimen was photographed by Meyer (#3672) but was on loan from the BM in 198 EVEA GUIANENSIS (1: 100. t. 39 (guyannensis))—-NG, TS— Rubiaceae = Fara- mea guianensis (Aublet) Bremek. Cephaélis Sw. is a conserved generic name (#8411), and Evea Aublet is rejected. Since Evea guianensis Aublet is regarded as a species of Faramea, it is not necessary to list Evea as a rejected name under Cephaélis. BM, LINN-SM 340.7, P-R 4: 131. EXACUM GUIANENSE (1: 68. ¢. 26, fig. 1 (guyannensis))—Gentianaceae = Schul- tesia guianensis (Aublet) Malme (as guyanensis). P-R 3: 113. EXACUM TENUIFOLIUM (1: 70. ¢. 26, fig. 2)—Gentianaceae = Curtia tenuifolia (Aublet) Knobl. Jonker (in Pulle, 1935, 4: 413) said that the detail figures should be excluded. p-R 3: 113 FAGARA PENTANDRA (I: 78. ¢. 30)—Rutaceae = Zanthoxylum pentandrum (Au- blet) Howard, comb. nov. Lemée (1952, p. 180) accepted the name Fagara pentandra and cited Xanthoxylum hermaphroditum Willd. (Sp. PI. 4: 756) in synonymy. The combination Zanthoxylum pentandrum (Aublet) How- ard is needed if Fagara is not recognized as a distinct genus. P-R 3: 120. FARAMEA CORYMBOSA (1: 102. ¢. 40, fig. 1) —NG, TS—Rubiaceae = Faramea corymbosa Aublet. Steyermark designated Faramea corymbosa Aublet as the We species for the genus. He apparently did not see any material. BM, P-R FARAMEA SESSILIFLORA (1: 104 (sessiflora). Index, p. 12 (sessiliflora). t. 40, fig. 2 (sessiliflora))— Rubiaceae = Faramea sessiliflora Aublet. BM, LINN-SM 340.9, p-r 4: 130 FEROLIA GUIANENSIS (Suppl. 7. ¢. 372 (guyannensis))—NG, TS— Moraceae = Brosimum rufescens Taub., not B. guianense (Aublet) Huber, which is based on Piratinera guianensis Aublet. Brosimum Sw. is a conserved ge- neric name (#1957), and Ferolia Aublet is rejected. BM. FOTHERGILLA MIRABILIS (1: 441. Index, p. 13 (admirabilis). t. 175 (mirabilia))— 270 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Melastomataceae = Miconia mirabilis (Aublet) L. O. Williams. Gleason (in Pulle, 1935, 3(1): 243) noted that some copies of Histoire des Plantes de la Guiane Francoise used Tamonea guianensis, and he accepted Mi- conia guianensis (Aublet) Cogn. without indicating that other copies used Fothergilla mirabilis in the text and on the plate. Louis Williams (Fieldiana Bot. 29: 574, 575. 1963) noted, “In the same work (2: 659, t. 268. 1775) he [Aublet] published another Tamonea in the Verbenaceae. It seems probable that he ‘discovered’ that he had described two genera with the name 7amonea and changed one of them while the work was going through the press. It seems quite clear that he changed the name of the Melastome to Fothergilla mirabilis and that he should be followed in this. Indicative that his intent was this is that the name Tamonea guianensis does not appear in the index of scientific names but Fothergilla mirabilis does—as F. admirabilis.” Henrey (1975, p. 6) noted that several pages of the original printing were changed, as indicated by the asterisk used in the binder’s instructions. This resulted from the replacement of the first Tamonea with the name Foth- ergilla, which proves to be a later homonym. Unaltered copies are in the libraries of the British Museum (Natural History) and the Royal Botanic Gardens, Kew, and presumably some are to be found in Paris. Aublet did not change the use of the name 7amonea for a member of the Verbenaceae. Schreber (1789, p. 19) substituted the name Ghinia for the verbenaceous plant. This Moldenke (Phytologia 47: 409. 1981) has adopted with the argument, “Regardless of Aublet’s later attempt to ‘cancel out’ the earlier publication by substituting ‘Fothergilla’ for it and maintain the name for the verbenaceous group, under the present Code of Botanical Nomencla- ture he was not justified in so doing. Without formal conservation, the later homonym must be rejected, even though Miconia Ruiz & Pavon (1794) has since officially been conserved over the melastomataceous Ja- monea of Aublet (1775) and in spite of the assertion by Green (1935) that ‘The name 7amonea Aubl. (Verbenaceae) stands without conservation.’”’ It can be argued that since 7amonea (Melastomataceae) is a rejected name, this clears the way for Tamonea to be used in the Verbenaceae, as Aublet clearly intended—or that the name Tamonea (Melastomataceae) was not validly published because it was “‘not accepted by the author in the original publication” (Article 34.1 of the Code). Also overlooked is the fact that Buc’hoz (Pl. Nouv. Découv. p. 9. ¢. 7. 1779) recognized early that Foth- ergilla was a later homonym and proposed the name Lieutautia mirabilis (Aublet) Buc’hoz for that species. BM, LINN-SM 782.51. GALIPEA TRIFOLIATA (2: 662. ¢. 269)—NG, TS—Rutaceae = Galipea trifoliata Aublet. BM. GOuUPIA GLABRA (1: 296. ¢. 1/6)—NG, TS—Goupiaceae = Goupia glabra Au- blet. BM, LINN-SM 551.1, p-R 12: 368. GUAPIRA GUIANENSIS (1: 308. ¢. 1/9 (Quapira guyannensis))—NG, TS—Nyc- taginaceae = Pisonia guianensis (Aublet) Howard, comb. nov. Current flo- 1983] HOWARD, AUBLET 21 ras vary in the acceptance of Guapira as distinct from Pisonia L. P-J 5170 (a poor match for the plate). GUETTARDA COCCINEA (1: 317. t. 123 (cocinea))— Rubiaceae = Isertia coccinea (Aublet) Gmelin. p-5 9991. HERNANDIA GUIANENSIS (2: 848. ¢. 329 (guyannensis))—Hernandiaceae = Her- nandia guianensis Aublet. Kubitzki (Bot. Jahrb. 89: 147. 1969) indicated the type at Bm, with additional specimens at G and NY. HEVEA GUIANENSIS (2: 871. ¢. 335 (peruviana))—-NG, TS—Euphorbiaceae = Hevea guianensis Aublet. BM. HEYMASSOLI SPINOSA (1: 324. t. 125)—NG, TS—Olacaceae = Ximenia ameri- cana L. BM, P-J 11894. HIRTELLA AMERICANA (1: 247. t. 98 (Hitella))—Chrysobalanaceae = Hirtella racemosa Lam., not H. americana L. p-R 6: 214 marked “type of H. americana Aubl.” HoumiIRI BALSAMIFERA (1: 564. ¢. 225)—Humiriaceae = Humiria balsamifera (Aublet) St.-Hil. Humiria Aublet is conserved (#3953), with the spelling correction made by Saint-Hilaire; Houmiri Aublet is rejected. BM, LINN-SM 940.1 HyDROLEA SPINOSA (1: 281. ¢. 1/0)—Hydrophyllaceae = Hydrolea spinosa L. LINN-SM 469.3. HYPERICUM GUIANENSE (2: 784. ¢. 3// (Hipericum guyannense))—Guttiferae = Vismia guianensis (Aublet) Choisy. BM. HYPERICUM LATIFOLIUM (2: 787. t. 312, fig. 1)—Guttiferae = Vismia latifolia (Aublet) Choisy. BM. HYPERICUM SESSILIFOLIUM (2: 787. ¢. 312, fig. 2)—Guttiferae = Vismia sessi- lifolia (Aublet) DC. BM. ICACOREA GUIANENSIS (Suppl. 1. ¢. 368 (guyannensis))—NG, TS—Myyrsina- ceae = Ardisia guianensis (Aublet) Mez. The name Ardisia Sw. is con- served (#6285), as is the type species A. tinifolia Sw., Icacorea Aublet is rejected. Recently, C. L. Lundell (Phytologia 49: 346. 1981) has reestab- lished Icacorea Aublet as a genus distinct from Ardisia Sw. BM. IcICA ALTISSIMA (1: 342. t. 132)—Burseraceae = Tetragastris altissima (Aublet) Swart. No supporting specimen was located. IcICA ARACOUCHINI (1: 343. ¢. 133)—Burseraceae = Protium aracouchini (Au- blet) Marchand (as aracouchili). BM. IcICA DECANDRA (1: 346. ¢. 135 (pentandra))—Burseraceae = Protium decan- drum (Aublet) Marchand. BM. ICICA ENNEANDRA (1: 345. ¢. 134)—Burseraceae = Protium decandrum (Aublet) Marchand. BM. IcICA GUIANENSIS (1: 340. ¢. 137 (guyannense))—Burseraceae = Protium gui- anense (Aublet) Marchand. BM. ICICA HEPTAPHYLLA (1: 337. ¢. 130)—NG, TS—Burseraceae = Protium hepta- Zi JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 phyllum (Aublet) Marchand. Swart (Rec. Trav. Bot. Néerl. 39: 301. 1942) reported that he could not locate a specimen and regarded the plate as the type. BM. IROUCANA GUIANENSIS (1: 329. ¢. 127 aaa NG, TS—Flacourti- aceae = Casearia guianensis (Aublet) Urban. IVIRA PRURIENS (2: 695. ¢. 279 (Juira))—NG, TS—Sterculiaceae = Sterculia pruriens (Aublet) Schum. BM, LINN-sM 1498.15 JUSTICIA COCCINEA (1: 10. ¢. 3)—Acanthaceae = Pachystachya coccinea (Au- blet) Nees. BM. JUSTICIA VARIEGATA (1: 12. t. 4)—Acanthaceae = Odontonema variegata (Au- blet) Kuntze. Specimen at BM annotated by V. M. Baum: “‘Status uncertain, to be removed from Odontonema’’, p-R |: 20. LECYTHIS AMARA (2: 716. ¢. 286)—Lecythidaceae = Eschweilera amara (Au- blet) Niedz. Scott Mori (pers. comm.) accepts this placement, noting, ‘““This is a widespread species to which the name &£. odora (Poeppig ex Berg) Miers has been most widely applied.” No supporting specimen located. LECYTHIS GRANDIFLORA (2: 712. tt. 283, 284)—Lecythidaceae = Eschweilera grandiflora (Aublet) Sandwith. Scott Mori (pers. comm.) typifies this species on the habit and flowers as illustrated by Aublet: “‘“However, the fruit he describes as of his L. grandiflora are really those of Lecythis tumefacta Miers.”’ No supporting specimen located LECYTHIS IDATIMON (2: 721. f. 289)—Lecythidaceae = Lecythis idatimon Au- blet. Scott Mori (pers. comm.) regards this as ‘ta valid name that applies to a species most often misidentified as Eschweilera amara (Aublet) Nie- denzu.”’ No supporting specimen located. LECYTHIS PARVIFLORA (2: 717. t. 287)—Lecythidaceae = Eschweilera parviflora (Aublet) Miers. Scott Mori (pers. comm.) accepts this species in Eschwei- lera. BM. LECYTHIS ZABUCAJO (2: 718. £. 288)—Lecythidaceae = Eschweilera sp. Scott Mori (pers. comm.) will publish the transfer of this species to Eschweilera, adding other species to the synonymy. No supporting specimen located. LICANIA INCANA (1: 119. ¢. 45)—NG, TS—Chrysobalanaceae = Licania incana Aublet. BM, p-R 6: 204 LICARIA GUIANENSIS (1: 313. ¢. 12] (guyannensis))—NG, TS—Lauraceae = Li- caria guianensis Aublet. p-) 17237 marked “type” in annotation by Kos- termans. LIMODORUM GRANDIFLORUM (2: 818. ¢. 32/)—Orchidaceae = Cleistes grandi- florum (Aublet) Schlechter. No supporting specimen located LIMODORUM PENDULUM (2: 819. ¢. 322)— Orchidaceae = Dichaea pendula (Au- blet) Cogn. Material on loan from BM. LISYANTHUS ALATUS (1: 204. t. 80)—Gentianaceae = Chelonanthus alatus (Au- blet) Pulle. Maas (pers. comm.) reported the “type” to be at BM. Material of the genus was on loan from BM; LINN-SM 291.4; p-R 7: 258 1983] HOWARD, AUBLET 273 LISYANTHUS CAERULESCENS (1: 207. ¢. 82)—Gentianaceae = Irlbachia caeru- lescens (Aublet) Griseb. Maas (pers. comm.) reported the “type” to be at BM. LINN-SM 291.5, P-R LISYANTHUS GRANDIFLORUS eel (1: 205. t. 8/)—Gentianaceae = Chelo- nanthus alatus (Aublet) Pulle. Maas (pers. comm.) suggested this synon- ymy, with the “type” at BM. LINN-SM 291.6, P-R 7: 258. LISYANTHUS PURPURASCENS Aublet (1: 201. ¢. 79)—Gentianaceae = Chelo- nanthus sp. Maas (pers. comm.) indicated that he and a colleague will publish a new combination of the Aublet species in the genus Chelonan- thus. The “type” is at BM. P-R 7: 258, p-1 6994. MaseA PIRIRI (2: 867. ¢. 334, fig. 1) -NG, TS—Euphorbiaceae = Mabea piriri Aublet. BM. MABEA eo (2: 870. t. 334, fig. 2)-Euphorbiaceae = Mabea taquari Au- blet. B MACAHANEA GUIANENSIS (Suppl. 6. ¢. 377 (guyannensis))—NG, TS. This taxon remains unplaced. Albert C. Smith (Brittonia 3: 552. 1940) concluded that this was not a member of the Hippocrateaceae, which he was monographing. He noted that Jussieu had suggested the Guttiferae, “which seems at least as likely a position as the Hippocrateaceae. Miers (Trans. Linn. Soc. 28: 378. 1872), on the basis of the type specimen in the herbarium of the British Museum, placed the species in Clercia. He should have used Aublet’s name, which was earlier than Vellozo’s, for his generic group.” Lemée (1952, p. 311) used “Salicia guyanensis ? (Clercia g. Miers, Macahanea g. Aubl. d’aprés I. k.).” Airy Shaw (in Willis, 1973, p. 698) was also uncertain, stating, “Macahanaea Steud. = Macahanea Aubl. = ? Salacia L. (Celastraceae).” BM MACOUBEA GUIANENSIS (Suppl. 18. ¢. 378 (guyannensis))—NG, TS—Apocy- naceae = Macoubea guianensis Aublet. BM MACOUCOUA GUIANENSIS (1: 88. ¢. 34 (guyannensis))—NG, TS— Aquifoliace- ae = Ilex guianensis (Aublet) Kuntze. BM; P-R (material was seen, but ironically I failed to record the volume and folio numbers, as did Lanjouw and Uittien). MAHUREA PALUSTRIS (1: 558. ¢. 222)—NG, TS—Bonnetiaceae = Mahurea pa- lustris Aublet. BM. MAIETA GUIANENSIS (1: 443. ¢. 176 (Mayeta guyannensis))—NG, TS—Melas- tomataceae = Maieta guianensis Aublet. BM, LINN-SM 7 MALANEA SARMENTOSA (1: 106. t. 41)—NG, TS—Rubiaceae = Malanea sar- mentosa Aublet. Bm, p-R 4: 129 MALPIGHIA ALTISSIMA (1: 455. t. 181)—Malpighiaceae = Byrsonima aubletii Kosterm., not M. altissima Jacq. BM, LINN-SM 822.4 MALPIGHIA CRASSIFOLIA (1: 457. t. 182)—Malpighiaceae = Byrsonima crassi- folia (L.) Kunth. BM MALPIGHIA MOUREILA (1: 459. ¢. 183)—Malpighiaceae = Byrsonima crassifolia (L.) Kunth, BM, LINN-SM 822.3 274 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 MALPIGHIA VERBASCIFOLIA oo 460. t. 184)—Malpighiaceae = Byrsonima ver- bascifolia (L.) DC MANABEA ARBORESCENS a 64. ¢t. 24)—Verbenaceae = Aegiphila integrifolia (Jacq.) Jackson. The genus Manabea is credited to Aublet, although no type species was or has been designated. Specimen at BM annotated “‘type”’ by Moldenke; p-Rr 3: 109. MANABEA LAEVIS (1: 66. ¢. 25)— Verbenaceae = Aegiphila laevis (Aublet) Gme- lin. Specimen at BM annotated “type” by Moldenke; LINN-sM 215.8; P-R 3: 109. MANABEA VILLOSA (1: 62. ¢. 23)—Verbenaceae = Aegiphila villosa (Aublet) Gmelin. Specimen at BM annotated “‘type” by Moldenke; LINN-sM 215.6. MANAGA GUIANENSIS (Suppl. 2. f. 369 (guyannensis))—NG, TS. This taxon has not been placed satisfactorily. Albert C. Smith (Brittonia 3: 553. 1940) stated, ““Hallier (Meded. Herb. Rijks. Leiden 35: 21. 1918) believes that this monotypic genus may be referred to Sa/acia, but I see little reason for such a belief in the original description and plate (which is repeated in Buchoz, Grand. Jard. Univ. p/. 59. 1785). Although I am unable to suggest an alternative position for the genus, I believe that it may definitely be excluded from the Hippocrateaceae.”” Airy Shaw (in Willis, 1973, p. 710) refers the genus to Salacia L. (Celastraceae) without any indication of doubt. No supporting specimens located. MAPANIA SYLVATICA (1: 47. ¢t. 17) —NG, TS—Cyperaceae = Mapania sylvatica Aublet. BM, LINN-SM 108.1, P-R 1: 50. MAPOURIA GUIANENSIS (1: 175. ¢. 67 (guyannensis))—NG, TS—Rubiaceae = Psychotria mapourioides DC., not P. guianensis Raeuschel. Specimen at BM annotated “holotype’’; p-R 8: 269. MAPROUNEA GUIANENSIS (2: 895. t. 342 (Mapronnea guyannensis)—NG, TS— Euphorbiaceae = Maprounea guianensis Aublet (Jablonski, Mem. New York Bot. Gard. 17: 180. 1967, as “guyanensis”). BM (two sheets, one marked “type’’). MAQUIRA GUIANENSIS (Suppl. 36. ¢. 389 (guyannensis))— NG, TS— Moraceae = Maquira guianensis Aublet. Bo. MARIPA SCANDENS (1: 230. ¢. 97) —NG, TS—Convolvulaceae = Maripa scan- dens Aublet. Specimen at BM annotated by Austin; LINN-SM VIII-22; p-r 5: 180. MATAYBA GUIANENSIS (1: 331. ¢. 128 (Mataiba guyannensis))—NG, TS—Sap- indaceae = Matayba guianensis Aublet emend. Radlk. (excluding the fruit, which is Swartzia alatum Willd. = S. guianensis (Aublet) Urban). Reitz (FI. Il. Catarin. Sapindaceae, pt. 1, p. 133. 1980) stated that the type was at p. BM, P-J 11365. MATELEA LATIFOLIA (1: 278. ¢. 109, fig. 2)—Asclepiadaceae = Matelea palustris Aublet. BM, P-R 9: MATELEA PALUSTRIS (1: 278. ¢. 109, ie a NG, TS—Asclepiadaceae = Ma- telea palustris Aublet. Bm, p-R 10: 3 1983] HOWARD, AUBLET 215 MATOUREA PRATENSIS (2: 642. ¢. 259 (guyannensis))—NG, TS—Scrophulari- aceae = Stemodia pusilla Bentham. Lemée (3: 416) used this name, placing Aublet’s taxon in synonymy, and attributed it to Index Kewensis. No supporting specimen was located. MAYACA FLUVIATILIS (1: 42. ¢. 15 (Maiaca fluvialis)—NG, TS—Mayacaceae = Mayaca fluviatilis Aublet. BM; LINN-sM XVIII-13; P-R 1:51 labeled “‘type.” MAYEPEA GUIANENSIS (1: 81. Index, p. 18 (Mayepa). t. 31 (guyannensis))—NG, TS—Oleaceae = Chionanthus guianensis (Aublet) Stearn. Linociera Sw. ex Schreber is a conserved generic name (#6428), and Mayepea Aublet is rejected. Stearn (Ann. Missouri Bot. Gard. 63: 355. 1976) discussed the union of genera under Chionanthus. BM, P-R 3: 103. MAYNA ODoRATA (2: 921. ¢. 352)—NG, TS—Flacourtiaceae = Mayna odorata Aublet. Sleumer (Fl. Neotrop. 22: 28. 1980) designated a BM specimen as the holotype. MEBOREA GUIANENSIS (2: 825. t. 323 (guyannensis))—NG, TS—Euphorbiaceae. Lanjouw (Euphorb. Surinam, 9. 1931) concluded this is not Phyllanthus guyanensis of Mueller-Arg. and stated, “Whether Aublet’s plant isa gen- uine Phyllanthus is very difficult to say and must be left undecided.” According to Grady Webster (pers. comm.), “The plant seems definitely Euphorbiaceous and probably belongs in the tribe Phyllantheae, but may not belong in Phyllanthus. ... Apparently it has never been re-collected, and until it is the mystery will apparently continue.” BM MELASTOMA AGRESTIS (1: 425. ¢. 166)—Melastomataceae = Leandra agrestis (Aublet) Raddi. BM. MELASTOMA ALATA (1: 410. ¢. 158)—Melastomataceae = Miconia alata (Au- blet) DC. BM MELASTOMA AQUATICA (1: 430. t. 169)—Melastomataceae = Nepsera aquatica (Aublet) Naudin. BM, LINN-SM 654.11 MELASTOMA ARBORESCENS (1: 420. t. 163)—Melastomataceae = Loreya arbo- rescens (Aublet) DC. BM, LINN-SM 782. MELASTOMA BIVALVIS (1: 404. ¢. 155, fig. a)—Melastomataceae = Acisanthera bivalvis (Aublet) Cogn. BM. MELASTOMA CACATIN (1: 437. ¢. 173)—Melastomataceae. Cogniaux (in Martius, Fl. Brasil. 14: 518. 1888) listed this as a poorly known species. Aublet’s plate shows a sterile specimen. Wurdack (pers. comm.) wrote, “Melastoma cacatin is a most frustrating and vegetatively very distinct species. I have seen the Aublet (sterile) material (BM, P) and have modern sterile and fruiting collections from Surinam and French Guiana. The species is either a Miconia or Tococa (with fair-sized terminal inflorescences), but flowers are needed.” BM, LINN-SM 782. MELASTOMA ELEGANS (1: 427. t. 167)—Melastomataceae = Clidemia hirta (L.) D. Don var. elegans (Aublet) Griseb. BM, LINN-SM 782.31. MELASTOMA FLAVESCENS (1: 423. t. 164)—Melastomataceae = Henrietella fla- vescens (Aublet) Triana. BM. 276 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 MELASTOMA GRANDIFLORA (1: 414. ¢. 160)—Melastomataceae = Rhynchan- thera grandiflora (Aublet) DC. BM, LINN-sM 782.30, p-s 14109. MELASTOMA LAEVIGATA (1: 412. t. 159)—Melastomataceae = Miconia prasina (Sw.) DC., not Miconia laevigata (L.) DC. BM, LINN-sM 782.17. MELASTOMA LONGIFOLIA (1: 432. ¢. 170)—Melastomataceae = Miconia longi- folia (Aublet) DC. BM, LINN-sM 782.16 MELASTOMA PARVIFLORA (1: 433. ¢. 17/)—Melastomataceae = Miconia prasina (Sw.) DC. BM. MELASTOMA PURPURASCENS (1: 402. ¢. 154)—Melastomataceae = Aciotis pur- purascens (Aublet) Triana. BM, LINN-SM 782. MELASTOMA RACEMOSA (1: 406. ¢. 156 (racemola))—Melastomataceae = Mi- conia racemosa (Aublet) DC. BM, LINN-sM 782.4 MELASTOMA RUBRA (1: 416. ¢. 16/)—Melastomataceae = Clidemia rubra (Au- blet) Martius. BM, LINN-sM 782 MELASTOMA RUFESCENS (1: 408. ¢. /57)—Melastomataceae = Miconia rufes- cens (Aublet) DC. BM, LINN-smM 782.33 MELASTOMA SCANDENS (1: 435. ¢. 172)—Melastomataceae = Adelobotrys scan- dens (Aublet) DC. BM, LINN-smM 782.20 MELASTOMA SPICATA (1: 423. ¢. 165)— Melastomataceae = Clidemia capitellata (Bonpl.) D. Don var. dependens (D. Don) Macbr., not C. spicata D. Don. BM, P-J 14078. MELASTOMA SUCCOSA (1: 418. ¢. /62)—Melastomataceae = Henriettea succosa (Aublet) DC. BM, LINN-sM 782.14. MELASTOMA TRIVALVIS (1: 406. ¢. 155, fig. b)—Melastomataceae = Acisanthera bivalvis (Aublet) Cogn. BM, LINN-SM 654.3 MELASTOMA VILLOSA (1: 428. ¢. 168)—Melastomataceae = Desmoscelis villosa (Aublet) Naudin. BM, LINN-sM 654.4, 654.5. MIMOSA BOURGONI (2: 941. ¢. 358)—Leguminosae = Inga bourgoni (Aublet) DC. (as Burgoni). BM (two sheets: one with flowers, one with fruit). MIMOSA GUIANENSIS (2: 938. ¢. 357 (guyvannensis))—NG, aoa Leguminosae = Stryphnodendron guianensis roe Bentham. B MINQUARTIA GUIANENSIS (Suppl. 4. ¢. 370 oe NG, TS—Olaca- ceae = Minquartia guianensis ee Specimen at BM annotated “holo- type” by Sleumer MONIERA TRIFOLIA (2: 731. ¢. 293)—Rutaceae = Moniera trifolia Loefl. Bo. MONnNTIRA GUIANENSIS (2: 637. f. 257 (guyannensis))—NG, TS—Loganiaceae = Spigelia guianensis (Aublet) Lemée. Although this combination is attrib- uted to Lemée in Index Kewensis (Suppl. 12, p. 135. 1959), Lemée (1953, p. 271) cited “S[pigelia] guianensis ? (Montira g. Aublet, d’aprés I. k.).” No supporting specimen located. MOoQulILEA GUIANENSIS (1: 521. ¢. 208 (guyannensis))—NG, TS—Chrysobala- naceae = Licania guianensis (Aublet) Griseb. BM. 1983] HOWARD, AUBLET ye ay MOoRONOBEA COCCINEA (2: 789. ¢. 313)—NG, TS—Guttiferae. The illustration of habit and figures 1-9 show Moronobea coccinea Aublet. BM. Figures a—j are Symphonia globulifera L. f. MourerA FLUVIATILIS (1: 582. ¢. 233 (fluviatili))—NG, TS—Podostemona- ceae = Mourera fluviatilis Aublet. BM, LINN-SM 963.1, P-A 21473. MourIrI GUIANENSIS (1: 453. ¢. 180 (guyannensis))—NG, TS—Melastomata- ceae = Mouriri guianensis Aublet. For a discussion of this material, see Morley (Fl. Neotrop. 15: 171, 172. 1976), who designated as lectotype the fragment in the lower right-hand corner of the sheet (BM); LINN-SM 784.1 is presumed to be an isotype. MOouROUCOA VIOLACEA (1: 142. Index, p. 20 (Mouroucoua). t. 54)—NG, TS— Convolvulaceae = Maripa violacea (Aublet) Ooststr. ex Lanjouw & Uit- tien. Specimen at BM annotated “type” by Austin; p-R 6: 194. MouTouBEA GUIANENSIS (2: 680. ¢. 274 (guyannensis))—NG, TS—Polygala- ceae = Moutoubea guianensis Aublet. BM MOoUTOUCHI SUBEROSA (2: 748. t. 299) NG, TS— Leguminosae = Pterocarpus officinalis Jacq. Specimen at BM annotated “holotype” by Rudd. NACIBEA ALBA (1: 98. ¢. 37, fig. 2)—Rubiaceae = Manettia alba (Aublet) Wern- ham. BM, P-R 4: 133 NACIBEA COCCINEA (1: 96. ¢. 37, fig. I)-NG, TS—Rubiaceae = Manettia coc- cinea (Aublet) Willd. BM, LINN-SM 234.2, P-R 4: 133. NAPIMOGA GUIANENSIS (1: 592. ¢. 237 (guyannensis))—NG, TS—Flacourti- aceae = Homalium guianense (Aublet) Oken. Sleumer (Fl. Neotrop. 22: 60. 1980) designated a specimen at BM as holotype. p-A 16427, NISSOLIA QUINATA (2: 743. ¢. 297)—Leguminosae = Machaerium quinatum (Aublet) Sandwith. Aublet did not supply the specific epithet with the generic description, but the epithet is in the index (p. 20) and on the plate. Specimen at BM annotated “holotype?” by Rudd. NONATELIA LONGIFLORA (1: 185. f¢. 77)—Rubiaceae = Palicourea longiflora (Aublet) Rich. BM, LINN-SM 333.31, P-R 8: 271. NONATELIA LUTEA (1: 190. ¢. 74)—Rubiaceae = Palicourea longiflora (Aublet) ich. BM, LINN-SM 342.2. NONATELIA OFFICINALIS (1: 182. ¢. 70, fig. 1) —-NG, TS—Rubiaceae = Psycho- tria officinalis (Aublet) Sandwith. BM, P-R 8: 271 NONATELIA PANICULATA (1: 184. ¢. 70, fig. 2)—Rubiaceae = Psychotria pani- culata (Aublet) Raeuschel. Bm (specimen filed under P. flexuosa). NONATELIA RACEMOSA (1: 186. f. 72)—Rubiaceae = Psychotria racemosa (Au- blet) Raeuschel. p-R 8: 271. NONATELIA VIOLACEA (1: 188. ¢. 73)— Rubiaceae = Psychotria capitata Ruiz & Pavon subsp. amplifolia (Raeuschel) Steyerm., not P. violacea Aublet. P-R SATA NORANTEA GUIANENSIS (1: 554. ¢. 220 (guyannensis))—NG, TS—Marcgravi- ceae = Norantea guianensis Aublet. BM 218 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 OCOTEA GUIANENSIS (2: 781. 4 3/0 (guyannensis))—NG, TS—Lauraceae = Ocotea guianensis Aublet. BM. OMPHALEA DIANDRA (2: 843. ¢t. 328)—Euphorbiaceae = Omphalea diandra L. ORELIA GRANDIFLORA (1: 271. ¢. 106)—NG, TS—Apocynaceae = Allamanda cathartica L. LINN-sM 441.1, P-R 6: 223. OURATEA GUIANENSIS (1: 397. t. 152)—NG, TS—Ochnaceae = Ouratea gui- anensis Aublet. LINN-sM 931.9. OUROUPARIA GUIANENSIS (1: 177. ¢. 68 (guyannensis))— NG, TS— Rubiaceae = Uncaria guianensis (Aublet) Gmelin. Uncaria Schreber is a conserved generic name (#8228), with the type species designated as U. guianensis (Aublet) Gmelin; Ourouparia Aublet is rejected. BM, P-R 8: 270. OUTEA GUIANENSIS (1: 29. ¢. 9 (guyannensis))—NG, TS— Leguminosae = Mac- rolobium guianense (Aublet) Pulle. Macrolobium Schreber is a conserved generic name (#3517), and Vouapa Aublet (q.v.) and Outea Aublet are both rejected. The conserved type species is Macrolobium bifolium (Au- blet) Pers. (based on Vouapa bifolia Aublet). Specimen at BM annotated “isotype” by R. Cowan; LINN-SM 77.1; P-R 1: 45. PACHIRA AQUATICA (2: 726. tt. 291, 292)—NG, TS—Bombacaceae = Pachira aquatica Aublet. BM. PACOURIA GUIANENSIS (1: 269. ¢. 105 (guyvannensis))— NG, TS—Apocynaceae = Pacouria guianensis Aublet. Landolphia Beauv. is a conserved generic name (#6562), and Pacouria Aublet is rejected. Pulle (Enum. Pl. Surinam, p. 379. 1906) published the combination Landolphia guianensis (Aublet) Pulle (as guyanensis). Pichon (Monogr. Landolphiées, Mém. I’Inst. Fr. d’Afr. Noire 35: 245. 1953) reestablished Pacouria as a segregate of Lan- dolphia and stated the type was “JiR [P].”’ BM, LINN-SM 440.2, p-R 7: 232. PACOURINA EDULIS (2: 800. ¢. 316)—NG, TS—Compositae = Pacourina edulis Aublet. BM. PAGAMEA GUIANENSIS (1: 113 (Pagama). Index, p. 21 (Pagamea). t. 44 (guy- annensis))—NG, TS—Rubiaceae = Pagamea guianensis Aublet. Bm, LINN-SM 340.8, P-R 4: 143. PALICOUREA GUIANENSIS (1: 173. t. 66 (guyannensis))—NG, TS—Rubiaceae = Palicourea guianensis Aublet. BM, P-R 8: 268. PALMA MARIPA (2: 974. frontisp. I, 2)—Arecaceae = Attalea maripa Martius. Orator Cook (Jour. Wash. Acad. Sci. 30: 297. 1940) proposed the new genus Ethnora for this taxon, with the type species Ethnora maripa (Mar- tius) Cook. Since the generic description, although diagnostic, is in English, and there is neither a Latin description nor reference to a previously published one, the name Ethnora is illegitimate. The name Attalea maripa Martius is validly published. In the index Aublet grouped other names in the genus Pa/ma as “‘spe- cificum seu Triviale’” names, citing a page reference and the numbers within the frontispiece. In the text, however, the specific names of the 1983] HOWARD, AUBLET pide) index are given only as colloquial names of the Caraibes and are not in parentheses, as was Aublet’s custom with other specific names. Cook (/bid., p. 299) stated, “The four other palms, ‘comon,” ‘avoira,’ “bache,’ and ‘zaguenete,’ are identified respectively as Oenocarpus, Astrocaryum, Maur- itia and Manicaria. No other palms were illustrated in Aublet’s Histoire, though several were briefly described without botanical names.” In con- trast, the specific name for Myrtus citrifolia Aublet is given only in the index, while the text contains a description and reference to a previously published illustration that has been accepted as the type of the species. I cannot accept the several entries under Pa/ma as a comparable case and do not believe that Cook’s “identifications” to other genera should con- stitute publication. Urban (1920, pp. 139-141) has named some of the palms associated with the Plumier references Aublet supplied. Aublet’s work is of questionable use in typifying these names. PALOUE GUIANENSIS (1: 365. Index, p. 21 (Palove). t. 141 (guyannensis))—NG, TS—Leguminosae = Paloue guianensis Aublet. Although Meyer photo- graphed an Aublet specimen (#2291), it could not be found at BM. PAMEA GUIANENSIS (2: 946. ¢. 359 (guyannensis))— NG, TS—Combretaceae = Terminalia pamea DC., not 7. guyanensis Eichler. De Candolle (Prodr. 3: 13. 1828) cited Pamea guianensis but did not transfer the Aublet species. The specimen at BM is annotated ‘“‘This is a Buchenavia” by C. A. Stace and is filed under that genus. The new combination will be published in his forthcoming revision (pers. comm.). PANAX MOROTOTONI (2: 949. ¢. 360 (wndulata))— Araliaceae = Didymopanax morototoni (Aublet) Decne. & Planchon. BM. PARALEA GUIANENSIS (1: 576. ¢. 231)—NG, TS—Ebenaceae = Diospyros gui- anensis (Aublet) Giircke. Specimen at BM annotated “isotype?” by F. B. White PARIANA CAMPESTRIS (2: 877. t. 337)—NG, TS—Gramineae = Pariana cam- pestris Aublet. BM. PARINARI CAMPESTRIS (1: 517. t. 206)—NG, TS—Chrysobalanaceae = Parinari campestris Aublet. BM, LINN-SM 648.2, p-J 2786. PARINARI MONTANA (1: 514. ¢t. 204, 205)—Chrysobalanaceae = Parinari mon- tana Aublet emend. Ducke (as to fruit, tf. 205). The branches and leaves of plate 204 are Parinari rodolphii Huber. Prance (Fl. Neotrop. 9: 179, 182, 183. 1972) discussed the confusion regarding the Aublet plates and descriptions. BM. PARIVOA GRANDIFLORA (2: 757. ¢. 303)—NG, TS—Leguminosae = Eperua carl (Aublet) Bentham. Specimen at BM annotated “holotype” by R. Cowan; p-J 14646B PARIVOA TOMENTOSA (2: 759. ¢. 304)—Leguminosae = Crudia tomentosa (Au- blet) Macbr. Bm. PASSIFLORA COCCINEA (2: 828. t. 324)—Passifloraceae = Passiflora coccinea Aublet. p-A 21430, p-s 16665. 280 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 PASSIFLORA STIPULATA (2: 830. ¢. 325)—Passifloraceae = Passiflora stipulata Aublet. p-A 21403. PASSOURA GUIANENSIS (Suppl. 21. ¢. 380 (guvannensis))—NG, TS— Violaceae = Rinorea pubiflora (Bentham) Sprague & Sandwith, not R. guianensis Au- blet. No supporting specimen located PATABEA COCCINEA (1: 111. £43) —NG, TS— Rubiaceae = Ixora davisii Sand- with, not J. coccinea L. BM, P-R 3: 127. PATIMA GUIANENSIS (1: 197. t. 77 (guyvannensis))—NG, TS—Rubiaceae = Pa- tima guianensis Aublet. BM, LINN-SM 341.1, P-R 7: 257. PAYPAYROLA GUIANENSIS (1: 249. t. 99 (guyannensis))—NG, TS—Violaceae = Paypayrola guianensis Aublet. BM, P-R 6: PEKEA BUTIROSA (1: 594. £. 238)—NG, TS—Caryocaraceae = Caryocar villo- sum (Aublet) Pers. Prance (Fl. Neotrop. 12: 33. 1973) noted that the iden- tification applied to the fruit and flowers. He was not able to locate a specimen but concluded that the leaves illustrated were not of Caryvocar. No supporting specimens were located. PEKEA TUBERCULOSA (1: 597, ¢. 239)— Sere = Caryocar nuciferum L. as to fruit only. Prance (Fl. Neotrop. 12: 56. 1973) suggested the leaves are like those found in the poe or Sievouliaceae: BM, LINN-SM 968.1, 968.2. PERAMA HIRSUTA (1: 54. ¢. 18 (Perana))—NG, TS— Rubiaceae = Perama hir- suta Aublet. BM, P-R 117 PEREBEA GUIANENSIS (2: 953. ¢. 36] (guyvannensis))—NG, TS— Moraceae = Pe- rebea guianensis Aublet. No supporting specimens were located. PIPAREA DENTATA (Suppl. 31. ¢ 386)—NG, TS—Flacourtiaceae = Casearia commersoniana Camb., not C. dentata Sessé & Mog. BM PIRATINERA GUIANENSIS (2: 888. ¢. 340 (guyannensis))—NG, TS— Moraceae = Brosimum guianense (Aublet) Huber. Brosimum Sw. is a conserved generic name (#1957), and Piratinera Aublet is rejected. BM. PIRIGARA HEXAPETALA (1: 490. ¢. 193)—Lecythidaceae = Gustavia hexapetala (Aublet) Sm. No supporting specimens were located PIRIGARA TETRAPETALA (1: 487. ¢. 192)—NG, TS—Lecythidaceae = Grias tet- rapetala ane Niedz. rs by Lemée (1953, p. 115). LINN-sM 1156.1. PIRIPEA PALUSTRIS (2: 628. ¢. 253)—NG, TS—Scrophulariaceae = Buchne palustris (Aublet) eer Sea at BM annotated “‘type” by Philoox. PIRIQUETA VILLOSA (1: 298. ¢. 177) —NG, TS—Turneraceae = Piriqueta viscosa Griseb. p-j 12802 annotated by Arbo; p-R 12: 367. PITUMBA GUIANENSIS (Suppl. 29 (Pitumba). Index, p. 24 (Pitumba guianensis). t. 385 (guyannensis))— NG, TS—Flacourtiaceae = Casearia pitumba Sleu- mer, not C. guianensis (L. C. Rich.) Urban. Specimen at Bm designated the holotype by Sleumer (Fl. Neotrop. 22: 351. 1980). POLYGALA TIMOUTOU (2: 737. t. 295)—Polygalaceae = Polygala timoutou Au- blet. BM 1983] HOWARD, AUBLET © 281 POLYGALA VIOLACEA (2: 735. t. 294)—Polygalaceae = Polygala violacea Aublet. LINN-SM 1176.31. PORAQUEIBA GUIANENSIS (1: 123. ¢. 40 (SUVEn TER r= NG, TS—Icacinaceae = Poraqueiba guianensis Aublet. BM, P-R 76. POSOQUERIA LONGIFLORA (1: 134. ¢. 5 )—NG. TS— Rubiaceae = Posoqueria longiflora Aublet. BM, p-J 9931, p-R 6: 199. POsSIRA ARBORESCENS (2: 934. t. 355)—NG, TS— Leguminosae = Swartzia ar- borescens (Aublet) Pittier. Swartzia Schreber is a conserved generic name (#3574), and Possira Aublet is rejected. Richard Cowan (FI. Neotrop. 1: 181. 1968) cited: “Type coLtection. F. Aublet s.n. (isotype BM), near source of Galibi R., French Guiana.” BM POTALIA AMARA (1: 394. ¢. 151) —NG, TS—Potaliaceae = Potalia amara Au- blet. The family Potaliaceae has been placed in the Gentianaceae by Fos- berg (Smithson. Contr. Bot. 45: 18, 19. 1980). Specimen at BM annotated “holotype” by Leeuwenberg; p-J 7021. POUROUMA GUIANENSIS (2: 892. t. 34] (guyannensis))—NG, TS— Moraceae = Pourouma guianensis Aublet. Specimen at BM photographed by Meyer (#3552) but not located. POUTERIA GUIANENSIS (1: 86. ¢. 33 (guyvannensis))—NG, TS—Sapotaceae = Pouteria guianensis Aublet. The fruit is a species of S/oanea (Elaeocar- paceae), while the flowers illustrated have been referred to Labatia pe- dunculata (Sapotaceae). Sandwith (1931, p. 477) and Baehni (Candollea 1: 264. 1942), among many others, have discussed P. guianensis. The type specimen consists of a branch with three leaves. BM PsIDIUM AROMATICUM (1: 485. ¢. 19/)— Myrtaceae = Campomanesia aromatica (Aublet) Griseb. LINN-SM 881.3. PsIDIUM GRANDIFLORUM (1: 483. ¢. 190)—Myrtaceae = Campomanesia gran- diflora (Aublet) Sagot. LINN-SM 881.4 PSYCHOTRIA VIOLACEA (1: 145. ¢. 55)—Rubiaceae. Steyermark (in Maguire, 1972, p. 399) stated that “without an examination of the Aublet specimen, it is not possible to state whether Psychotria violacea belongs to Coccocypselum or Geophila.” Lanjouw and Uittien (1940, p. 156) referred to a specimen at BM that I could not locate and to p-R 8: 282, which they felt was a good species of Coccocypselum. The final decision should be made by a specialist. QUALEA CAERULEA (1: 7. t. 2)— Vochysiaceae = Qualea caerulea Aublet. Spec- imen at BM annotated “holotype” by Stafleu. QUALEA ROSEA (1: 5. ¢. J) —NG, TS—Vochysiaceae = Qualea rosea Aublet. Specimen at BM annotated “holotype” by Stafleu. QUAPOYA PANA-PANARI (2: 900. ¢. 344)—Guttiferae = Clusia pana-panari (Au- blet) Choisy. BM. QUAPOYA SCANDENS (2: 898. ¢. 343)—NG, TS—Guttiferae = Quapoya scan- dens Aublet. BM 282 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 sri GUIANENSIS (2: 692. f. 278 (guyannensis))—NG, TS—Bombaca- ceae = Quararibea guianensis Aublet. 3m material on loan. Quran GUIANENSIS (2: 838 (Quebitea). Index, p. 25 (Quebitea guianensis). 327 (guyannensis))—NG, TS—Piperaceae = Piper humistratum G6rts : Kramer, not P. guianensis DC. Steyermark (pers. comm.) suggested this identification. The Aublet name was not mentioned by Yuncker in his comprehensive The Piperaceae of Northern South America or by the authors of P. humistratum. Aublet’s specimen (BM) Is sterile, but it and the illustration now seem to be properly placed. QUIINA GUIANENSIS (Suppl. 19. ¢. 379 (guyannensis))— NG, TS—Quunaceae = Quiina guianensis Aublet. Bm RACARIA SYLVATICA (Suppl. 24. ¢. 382)—NG, TS—Sapindaceae = Talisia syl- vatica (Aublet) Radlk. Bo. RACOUBEA GUIANENSIS (1: 590. ¢. 236)—NG, TS—Flacourtiaceae = Homalium guianense (Aublet) Oken. BM. RAPANEA GUIANENSIS (1: 121. ¢. 46 (guyannensis))—NG, TS—Myrsinaceae = Mpyrsine guianensis (Aublet) Kuntze. BM; p-R 6: 203 annotated “‘lectotype”’ by Gillis. RAPATEA PALUDOSA (1: 305. ¢. 1/85) —NG, TS—Rapateaceae = Rapatea palu- dosa Aublet. p-J 3173. RAPUTIA AROMATICA (2: 671. t. 272)—NG, TS—Rutaceae = Raputia aromatica Aublet. BM. REMIREA MARITIMA (1: 45. ¢. 16) —NG, TS—Cyperaceae = Mariscus pedun- culatus (R. Br.) T. Koyama, not Mariscus maritimus Maiq. P-R RHEXIA LATIFOLIA (1: 336. t. 129, fig. 2)—Melastomataceae = Comolia latifolia (Aublet) Cogn. No supporting specimen located RHEXIA VILLOSA (1: 334. ¢. 129, fig. 1)—Melastomataceae = Comolia villosa (Aublet) Triana. BM, LINN-SM 654.15. RIANA GUIANENSIS (1: 237. ¢. 94 (guyannensis))—NG, TS—Violaceae = Ri- norea riana Kuntze, not R. guianensis Aublet. BM; P-R 5: 173 annotated “type” by Hekking. RINOREA GUIANENSIS (1: 235. f¢. 93 (guyannensis))—NG, TS—Violaceae = Ri- norea age Aublet. Specimen at BM annotated ‘holotype’; LINN-SM 409.3; p-R 5: 169 annotated “‘type’”” by Hekking, who has proposed the ene ae of Rinorea Aublet against pete Aublet (Taxon 31: 754. 1982). ROBINIA NICOU (2: 771. ¢. 308)— Leguminosae = Lonchocarpus rufescens Ben- tham. Bentham (Jour. Linn. Soc. Bot. 4(Suppl.): 99. 1860) noted that ‘‘the single leaf of Aublet’s Robinia nicou preserved in the Banksian herbarium resembles this more than any other species.” Although flowers are illus- trated along with foliage, a specialist on the Leguminosae should decide if a new combination is justified. BM. ROBINIA PANACOCO (2: 768. ¢. 307)— Leguminosae = Swartzia panacoco (Au- 1983] HOWARD, AUBLET 283 blet) R. Cowan. Sandwith (Kew Bull. 1934: 358, 359) discussed this prob- lem in considerable detail: Aublet described and figured his Robinia panacoco from the leaves and stipules of a Swartzia and from the flowers and fruit of another genus, probably—as suggested by de Candolle—of a Lonchocarpus. Willdenow’s description of Ro- (Aublet’s) figure, and there is no evidence that he suspected the composite nature of Robinia panacoco; the name Robinia tomentosa was theref perfl d illegitimate. It was de Candolle who first identified the leaves and stipules with those of flowering material of a Swartzia collected in French Guiana by Patris, reasonably rejected the name Panacoco for the species, and accepted Willdenow’s epithet which applied to the Swartzia foliage. It is not clear whether de Candolle saw Aublet’s dried material or merely matched Patris’ specimens with the figure and description. At any rate he was correct, for the Parris foliage (kindly lent from Geneva) agrees excellently. . .with that of two sheets collected by Aublet, and corresponds sufficiently to the figure and description of the leaves of Robinia panacoco in the British Museum Herbarium. The name Swartzia tomentosa DC. should therefore be accepted for this foliage, but the Patris collection should be regarded as the type of the species dating from 1825, Robinia panacoco Aubl. being treated as a nomen confusum and Robinia tomentosa Willd. as an illegit- imate name. Unfortunately, nomina confusa are no longer recognized in the Code of Botanical Nomenclature. Sandwith clearly typified the Aublet name with the foliage of a Swartzia in the herbarium at BM. Richard Cowan published the combination Swartzia panacoco (Aublet) R. Cowan in Flora Neotropica (1: 32. 1968) and lectotypified it (p. 38) by F. Aublet s.n. (3M). He noted, “A second Aublet sheet at the British Mu- seum may be an isolectotype but it has incomplete leaves. Neither sheet bears reproductive parts.” Cowan offers further discussion on page 39. BM. RONABEA LATIFOLIA (1: 154. t. 59 (Ronobea))—NG, TS—Rubiaceae = Psy- chotria erecta (Aublet) Standley & Steyerm., not P. /atifolia Humb. & Bonpl. The basionym is Ronabea erecta Aublet (1: 156; not illustrated). BM, P-R 8: 289. ROPOUREA GUIANENSIS (1: 198. ¢. 78 (guyannensis))—NG, TS—Ebenaceae = Diospyros martinii Benoist ex Amshoff, not Diospyros guianensis (Aublet) Giircke. Ropourea Aublet is listed as a nomenclatural synonym and 1S rejected for the conserved Idesia Maxim. (#5331). It appears that Idesia Scopoli (1777) was illegitimate when published. p-r 8: 276. ROUHAMON GUIANENSIS (1: 93. ¢. 36 (guyannensis))—NG, TS—Loganiaceae = Strychnos guianensis (Aublet) Martius. In various publications on Strych- nos, neither Monachino nor Krukoff has indicated a type specimen. Lan- jouw and Uittien (1940, p. 157) suggested that p-R 4: 134 is not this species. p-A 7211 is a better match. 284 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 ROUPALA MONTANA (1: 83. ¢. 32) —NG, TS—Proteaceae = Roupala montana Aublet. BM, LINN-SM 157.1, P-R 3: 102. ROUREA FRUTESCENS (1: 467. ¢t. 187)—NG, TS—Connaraceae = Rourea fru- tescens Aublet. pm. Schellenberg’s annotation on this sheet states that the type is at P. RUELLIA RUBRA (2: 666. ¢. 270)—Acanthaceae = Arrhostoxylum rubrum (Au- ~ blet) Nees. BM. RUELLIA VIOLACEA (2: 668. ¢. 277)—Acanthaceae = Arrhostoxylum violaceum (Aublet) Nees. Lemée (1953, p. 484) used “D[ipteracanthus]? violaceus.” BM. SABICEA ASPERA (1: 194. t. 76)—NG, TS— Rubiaceae = Sabicea aspera Aublet. This taxon was designated the lectotype species by Steyermark (Mem. New York Bot. Gard. 17(1): 307. 1967). BM, LINN-SM 342.2, P-R 8: 274. SABICEA CINEREA (1: 192. ¢. 75)—Rubiaceae = Sabicea cinerea Aublet. BM, P-R 8: 274 SAGONEA PALUSTRIS (1: 285. ¢. 1/1)—NG, TS— Geka aaa = Hydrolea palustris (Aublet) Racusohel, LINN-SM 469.3, P-R II: SALVINIA AURICULATA (2: 969. ft. 367)—Salviniaceae = Salvinia auriculata Au- blet. No supporting specimen located SAQUARI GLABRA (1: 599, ¢. 240)—NG, TS—Caryocaraceae = Caryocar gla- brum (Aublet) Pers. Prance (Fl. Neotrop. 12: 40. 1973) designated a spec- imen at BM as the type. LINN-SM 968.3, P-J 11425B. SAOUARI VILLOSA (1: 601. ¢. 247)—Caryocaraceae = Caryocar villosum (Aublet) Pers. BM, P-J 11426 SAPINDUS ARBORESCENS (1: 357. ¢. 139)—Sapindaceae = Matayba arborescens (Aublet) Radlk. Bm. SAPINDUS FRUTESCENS (1: 355. ¢. /38)—Sapindaceae = Pseudima frutescens (Aublet) Radlk. A specimen at Bm was photographed by Meyer (#3576) but was not located. SAUVAGESIA ADIMA (1: 251. ¢. 100, fig. a)—Ochnaceae = Sauvagesia erecta L. P-R 6: 212 SAUVAGESIA ERECTA (1: 254. ¢. 100, fig. b)—Ochnaceae = Sauvagesia sprengelii t. Hil. p-A 20653, p-r 6: 212 SENAPEA GUIANENSIS (Suppl. 22. ¢. 38) NG. TS. Airy Shaw (in Willis, 1973, p. 1059), apparently following Hallier (1918, p. 23), referred this genus to Passiflora with a question mark. The name was considered neither by Killip in his monograph of the Passifloraceae nor by Lemée. Tillett, who is working on the genus Di/kea (which seemed a possibility), has examined a photocopy of the specimen (BM) and is convinced that it is not a member of the Passifloraceae. If the fruit can be located, an identification might be possible (pers. comm.). SERAPIAS CARAVATA (2: 816. £. 320)—Orchidaceae = ENeanthus caravata (Au- blet) Reichenb. f. No supporting specimen located. 1983] HOWARD, AUBLET 285 SIMABA GUIANENSIS (1: 400. ¢. 153 (guyannensis))— NG, TS—Simaroubaceae = Simaba guianensis Aublet. p-s 13010 is indicated as “ex herb. Candolle.” SIMAROUBA AMARA (2: 860. tt. 331, 332)—NG, TS—Simaroubaceae = Sima- rouba amara Aublet. Simarouba Aublet is a conserved generic name (#4111), with the type species Simarouba amara Aublet, although Jansen- Jacobs (in Stoffers & Lindeman, eds., Fl. Suriname 5: 326. 1979) places this taxon in the synonymy of Quassia simaruba L. f. BM. SIMIRA TINCTORIA (1: 170. ¢. 65)—NG, TS—Rubiaceae = Simira tinctoria Au- blet. Bremekamp (Acta Bot. Neerl. 3: 150-153. 1954) discussed this genus in detail but did not designate a type. BM, P-R 8: 267. SINGANA GUIANENSIS (1: 574. ¢. 230 (guyannensis))—NG, TS—Leguminosae. Airy Shaw (in Willis, 1973, p. 1069) stated ““? Leguminosae (inc. sed.),” while Farr et al. (1979, p. 1617) have “Leguminosae, Papilionaceae,” attributed to R. Cowan. BM. SIPANEA PRATENSIS (1: 148. ¢. 56) —NG, TS—Rubiaceae = Sipanea pratensis Aublet. BM, P-R 8: 2 SIPARUNA GUIANENSIS (2: 865. ¢. 333 (guyannensis))—NG, TS—Monimi- aceae = Siparuna guianensis Aublet. BM SLOANEA SINEMARIENSIS (1: 534. ¢. 2/2)—Elaeocarpaceae = Sloanea sinemari- ensis Aublet. BM, LINN-SM 928.4. SOLANUM TEGORE (1: 212. ¢. 84)—Solanaceae = Cyphomandra tegore (Aublet) Sendt. ex Walp. Specimen at BM annotated “Cyphomandra hartwegii sp. coll. A. Childs in 1974.” SORAMIA GUIANENSIS (1: 552. ¢. 219 (guyvannensis))—NG, TS—Dilleniaceae = Doliocarpus guianensis (Aublet) Gilg. BM. SOUROUBEA GUIANENSIS (1: 244. ¢. 97 eagle NG, TS—Marcgravi- aceae = Souroubea guianensis Aublet. BM, P-R SPERMACOCE ALATA (1: 60. t. 22, fig. 7)—Rubiaceae = Borreria alata (Aublet) DC. Work in press by F. R. Fosberg may clarify this and some of the following taxa. Recent work has suggested that Borreria is synonymous with Spermacoce. All of Aublet’s taxa of Spermacoce require careful study of the actual specimens, for few confident conclusions can be drawn from the descriptions or illustrations. p-R 3: 118 annotated by Sandwith. SPERMACOCE ASPERA (1: 59. ¢. 22, fig. 6)—Rubiaceae = Mitracarpus hirtus (3) DC. Sandwith (1963, p. 260) commented on this taxon. P-R 3: 118 an- notated by Sandwith. SPERMACOCE CAERULESCENS (1: 57. ¢. 19, fig. 2)— Rubiaceae = Borreria latifolia (Aublet) Schum. p-r 3: 118 annotated by Sandwith. SPERMACOCE HEXANGULARIS (1: 61. ¢. 22, fig. 8)— Rubiaceae. Lemée (1953, p. 569) called this taxon “Spermacoce sexangularis Aublet,” citing S. hexangularis Aublet as a synonym. No supporting specimens locate SPERMACOCE LATIFOLIA (1: 55. t. 19, fig. 1)—Rubiaceae = Borreria latifolia (Aublet) Schum. No supporting specimen located. 286 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 SPERMACOCE LONGI-FOLIA (1: 58. ¢. 2/)—Rubiaceae = Mitracarpus hirtus (L.) DC. See Sandwith (1963, p. 260). p-R 3: 118 annotated by Sandwith. SPERMACOCE PROSTRATA (1: 58. ¢. 20, fig. 3)—Rubiaceae = Spermacoce pros- trata Aublet. Fosberg rane comm.) stated that the type is in the Delessert herbarium at G. p-R 3: 118. SPERMACOCE RADICANS ie: 58. ¢. 20, fig. 4)— Rubiaceae = Diodia sp. nov. ? See Sandwith (1963, p. 260). p-R 3: 118. TABERNAEMONTANA ECHINATA (1: 263. ¢. 103)—Apocynaceae. Allorge (pers. comm.) has suggested that the foliage shown in plate 103 is to be called Peschira echinata (Aublet) DC. (Anacampta echinata (Aublet) Markgr.), and the fruit (figs. 7, 8) Stenosolen heterophyllus (Vahl) Marker. No specimens were seen. TACHIA GUIANENSIS (1: 75. ¢. 29 (guyannensis))—NG, TS—Gentianaceae = Tachia guianensis Aublet. Maguire and Weaver (Jour. Arnold Arb. 56: 113. 1975) cited the type as “p (IDC 6213.11: 17). p-R 3: 115 TACHIBOTA GUIANENSIS (1: 287. ¢. 1/2 (guyannensis))—NG, TS—Chrysoba- lanaceae = Hirtella racemosa Lam., not H. guyanensis (Fritsch) Sandwith. Prance (Fl. Neotrop. 9: 326. 1972) cited Aublet specimens at BM and p-R (11: 352) without designating a lectotype. TACHIGALI PANICULATA (1: 372. t. 143, fig. 1)—NG, TS—Leguminosae = Tachi- gali paniculata Aublet. Although Meyer photographed a specimen at BM (#2297), 1t was not located. TACHIGALI TRIGONA (1: 374. ¢. 143, fig. 2)—Leguminosae = Tachigali pani- culata Aublet. BM, LINN-SM 737.1. TALIGALEA CAMPESTRIS (2: 625. ft. 252)—NG, TS—Verbenaceae = Amasonia campestris (Aublet) Moldenke. Amasonia L. f. isa conserved generic name (#7156), and Taligalea Aublet is rejected. BM. TALISIA GUIANENSIS (1: 349. f¢. i (guyannensis))—NG, TS—Sapindaceae = Talisia guianensis Aublet. B TAMONEA GUIANENSIS (1: Pian = Miconia mirabilis (Au- blet) L. O. Williams. The change in name in the text and apparently on the plates of some copies is discussed under fothergilla (q.v.). BM. TAMONEA SPICATA (2: 660. ¢. 268)—NG, TS— Verbenaceae = Tamonea spicata ublet. BM, LINN-SM 55.1. TAMPOA GUIANENSIS (Suppl. 35. ¢. 388 csannensts))— NG, TS—Celastraceae. Albert C. Smith (1940, pp. 553, 554) stated, “This monotypic genus is certainly not related to Sa/acia, as ae by Hallier (Meded. Herb. Rijks Leiden 35: 24. 1918).”’ Lemée (1952, p. 311) recognized the genus, while Airy Shaw (in Willis, 1973, p. 1129) stated without question that the genus equaled Sa/acia L. Bn. TANIBOUCA GUIANENSIS (1: 448. ¢. 778 (guyvannensis))—NG, TS—Combreta- ceae = Terminalia dichotoma G. F. W. Meyer, not 7. guianensis Eichler. BM. TAONABO DENTATA (1: 569. ¢. 227)—NG, TS—Theaceae = Ternstroemia den- tata (Aublet) Sw. BM. 1983] HOWARD, AUBLET 287 TAONABO PUNCTATA (1: 571. t. 228)—Theaceae = Ternstroemia punctata (Au- blet) Sw. BM. TAPIRIRA GUIANENSIS (1: 470. ¢. 188 (guyannensis))—NG, TS—Anacardi- aceae = Tapirira guianensis Aublet. LINN-SM 829.1. TAPOGOMEA ALBA (1: 164. ¢. 62, fig. 4)—Rubiaceae = Psychotria ulviformis Steyerm., not P. alba Ruiz & Pavon. BM. TAPOGOMEA GLABRA (1: 165. ¢. 63)—Rubiaceae = Psychotria blepharophylla (Standley) Steyerm., not P. glabra (Turrill) Fosberg. BM, LINN-SM 340.5, P-R 8: TAPOGOMEA PURPUREA (1: 162. f¢. 62, fig. 3) Rubiaceae = Psychotria variegata Steyerm., not P. purpurea Merr. & Perry. BM, P-R 8: 265 TAPOGOMEA TOMENTOSA (1: 160. ¢. 6/)—Rubiaceae = Psychotria poeppigiana Mueller-Arg., not P. tomentosa Hemsley. BM, LINN-SM 340.4, P-R 8: 265. TAPOGOMEA VIOLACEA (1: 157. t. 60)—NG, TS— Rubiaceae = Psychotria apoda Steyerm., not P. violacea Aublet. The generic name Cephaélis Sw. is con- served (#8411), with the type species (also conserved) C. muscosa (Jacq.) Sw.; Tapogomea Aublet is listed as rejected. Steyermark (in Maguire et al., 1972, p. 406) has reduced Cephaélis to a section of the genus Psychotria. LINN-SM 340.2, P-R 8: 265. TAPURA GUIANENSIS (1: 126. ft. 48 (guyannensis))— NG, TS—Dichapetalaceae = Tapura guianensis Aublet. Prance (FI. Neotrop. 10: 68. 1972) cited spec- imens at BM and P but did not designate a lectotype. BM, P-R 6: 209. TARALEA OPPOSITIFOLIA (2: 745. ¢. 298)—NG, TS—Leguminosae = Dipteryx oppositifolia (Aublet) Willd. Dipteryx Schreber is a conserved generic name (#3845), with the type species conserved as Dipteryx odorata (Au- blet) Willd. (based on Coumarouna odorata Aublet): Taralea Aublet is rejected. BM. TARIRI GUIANENSIS (Suppl. 37. t. 390 (guyannensis))—NG, TS—Simarouba- ceae = Picramnia guianensis (Aublet) Jansen-Jacobs. Picramnia Sw. 1S a conserved generic name (#4131), and Tariri Aublet is rejected. De Can- dolle (Prodr. 2: 66. 1825) published the name **Plicramnia] ? tariri (Aubl.) DC.” based on Tariri guianensis Aublet. It is, of course, illegitimate. Aublet noted that his specimen was sterile. There are specimens at BM and p that are sterile, but p-1 16001 resembles the plate and is clearly a Picramnia. THOA URENS (2: 874. t. 336) —NG, TS—Gnetaceae = Gnetum urens (Aublet) Blume. Markgraf annotated an Aublet collection (BM) as Gnetum panti- culatum Spruce, stating that it is not the plant figured by Aublet. This information is given (Bull. Jard. Bot. Buitenzorg, III. 10: 455. 1930) with no citation of other Aublet material, suggesting that the plate is the type. TIBOUCHINA ASPERA (1: 446. ¢. 177) —NG, TS— Melastomataceae = Tibouchina aspera Aublet. BM, LINN-SM 654.6. TICOREA FOETIDA (2: 689. t. 277)—NG, TS—Rutaceae = Ticorea foetida Au- blet. BM TIGAREA ASPERA (2: 918. ¢. 350)—NG, TS—Dilleniaceae = Tetracera tigarea DC., not T. aspera Raeuschel. Kubitzki (Mitt. Bot. Staatssam. Miinchen 288 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 9: 94. 1971) noted that, while Davilla aspera (Aublet) Benoist was based on /igarea aspera Aublet, the specimen Benoist saw in the Lamarck her- barium was Davilla kunthii St.-Hil. The original Tigarea aspera specimen at BM is Tetracera tigarea DC., as was indicated by Kubitzki earlier (Mitt. Bot. Staatssam. Miinchen 8: 67-69. 1970). Specimen at Bm annotated “holotype” by Kubitzki. TIGAREA DENTATA (2: 920. ¢. 35/)—Dilleniaceae = Doliocarpus dentatus (Au- blet) Standley. BM. TOCOCA GUIANENSIS (1: 438. t=. 174 (guyannensis))—NG, TS—Melastomata- ceae = Tococa guianensis Aublet. LINN-sM 782.36, 782.37; p-) 14113. TOCOYENA LONGIFLORA (1: 131. 4 50)—NG, TS—Rubiaceae = Tocoyena lon- giflora Aublet. BM, p-R 6: 200 TONINA FLUVIATILIS (2: 857. f. 330) —NG, TS—Eriocaulaceae = Tonina fluvia- tilis Aublet. BM. TONTANEA GUIANENSIS (1: 108. ¢. 42 (guvannensis))—NG, TS—Rubiaceae = Coccocypselum guianensis (Aublet) Schum. p-J 9971, p-R 3: 128 TTONTELEA SCANDENS (1: 31. ¢. JO) —NG, TS—Celastraceae = Tontelea scandens Aublet. Albert C. Smith (1940, pp. 484, 485) recognized Tontelea scandens Aublet and cited for illustrations “Aubl. Pl. Guian. 3: p/. 10 (exclu. f. 10). 1775.” Such a figure is not numbered in the copies of Aublet I have had available for study. Smith (/bid., p. 484) stated, “It seems that Miers (20: 382) [Trans. Linn. Soc. 28: 382] is quite correct in taking Aublet’s plate to be a mixture and also in interpreting the species as he does. The Gerad leaf was taken by Miers as his Tontelea Aubletiana, which I hav \ referred to Cheiloclinium cognatum. Aublet’s description of the eoimens as attached within a cupular disk and the style as short (not lacking) indicates that his flowers were not to be associated with the detached leaf but probably were correctly associated with the illustrated branchlet.” Under Cheiloclinium cognatum (Miers) A. C. Sm., A. C. Smith (1940, p. 529) cited in synonymy Tontelea aubletiana Miers, Trans. Linn. Soc. 28: 383. 1872, for the detached leaf of Tontelea scandens in Aublet plate 10, figure 10. Subsequently, Lemée (1952, p. 310) recognized Salicia scandens (Aublet) Griseb., unaware of the confusion of sheets with discordant ele- ments. The specimen at BM, LINN-SM 87.1, and p-R 1: 54 represent the mixtures in this collection. TOPOBEA PARASITICA (1: 476. t. 189)—NG, TS—Melastomataceae = Topobea parasitica Aublet. BM. TOUCHIROA AROMATICA (1: 385. ¢. 148) —NG, TS— Leguminosae = Crudia aro- matica (Aublet) Willd. Crudia Schreber is a conserved generic name (#3495), with the conserved type species Crudia spicata (Aublet) Willd. (based on Apalatoa spicata Aublet),; Touchiroa Aublet is rejected. BM. TOULICIA GUIANENSIS (1: 359. t. 140 i a a TS—Sapindaceae = Toulicia guianensis Aublet. pm TOUNATEA GUIANENSIS (1: 550. ¢. 218 (guyannensis))—NG, TS—Legumino- sae = Swartzia guianensis (Aublet) Urban. Swartzia Schreber is a con- 1983] HOWARD, AUBLET 289 served generic name (#3574), and Tounatea Aublet is rejected. Richard Cowan (FI. Neotrop. 1: 144. 1968) discussed the typification of this taxon and concluded that an unnumbered Aublet collection at Bm was the “isotype.” BM, LINN-SM 958.4 TOUROULIA GUIANENSIS (1: 492. ¢. 194 (guyannensis))—NG, TS— Quiinaceae = Touroulia guianensis Aublet. No supporting specimen located TovoMITA GUIANENSIS (2: 957. t. 364 (guyannensis))—NG, TS— eanere = Tovomita guianensis Aublet. Sandwith (1931, p. 176) discussed this taxon and referred to a specimen (BM), which was not located. TRIFOLIUM GUIANENSE (2: 776. t. 309 (guyannense))— Leguminosae = Stylo- santhes guianensis (Aublet) Sw. In a monograph of the genus (Ann. Mis- souri Bot. Gard. 44: 332. 1956), Mohlenbrock used ““guyanensis”’; he saw only a photo ofan Aublet specimen. Specimen at BM annotated “‘lectotype”’ by tMannetjya. TRIGONIA LAEvis (1: 390. ¢. 150)—Trigoniaceae = Trigonia laevis Aublet. BM. TRIGONIA VILLOSA (1: 388. ¢. 149)—NG, TS—Trigoniaceae = Trigonia villosa Aublet. BM, P-J 12036 TRIPLARIS AMERICANA (2: 910. t. 347)— Polygonaceae = Triplaris surinamensis ham. (Brandbyge, pers. comm.). BM specimen may be on loan. LINN-SM 148.3. TURNERA FRUTESCENS (1: 290. ¢. 113, fig. 2)—Turneraceae = Turnera rupestris Aublet var. frutescens (Aublet) Urban. Specimen at BM labeled holotype; LINN-SM 540.9; p-A 16418, 16419; p-R 11: 354. TURNERA GUIANENSIS (1: 291. ¢. 114 (guyannensis))—Turneraceae = Turnera guianensis Aublet. BM, LINN-SM 540.7, P-R 11: 354. TURNERA RUPESTRIS (1: 289. ¢. 113, fig. 1)—Turneraceae = Turnera rupestris Aublet. Specimen at BM labeled “holotype’’; LINN-SM 540.8 VANTANEA GUIANENSIS (1: 572. t¢. 229 (guyannensis))—NG, TS—Humiri- aceae = Vantanea guianensis Aublet. BM VATAIREA GUIANENSIS (2: 755. ¢. 302 (guyannensis))— NG, TS—Leguminosae = Vatairea guianensis Aublet. BM VIOLA HYBANTHUS (2: 811. ¢. 379)— Violaceae = Corynostylis arborea (L.) Blake. VIOLA ITOUBOU (2: 808. t. 318)— Violaceae = Hybanthus calceolaria (L.) Schulze. P-J VIROLA SEBIFERA (2: 904. t. 345)—NG, TS—Myristicaceae = Virola sebifera Aublet. I can find no published comment on the several fruits illustrated with this species, which clearly represent other species or genera. Albert C. Smith (Brittonia 2: 393. 1957) designated figures 1-5 for this species. No supporting material located. VOCHY GUIANENSIS (1: t. 6 (guyannensis))—-NG, TS—Vochysiaceae = Vochysia guianensis ae Vochysia Aublet is a generic name conserved (#4266) with the corrected spelling of Poiret in Lamarck and the type 290 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 species indicated as V. guianensis Aublet. Stafleu (Rec. Trav. Bot. Néerl. 41: 452. 1948) designated the specimen at BM as the type but cited ‘‘Herb. Linn.” and a reference to Lanjouw and Uittien. BM and LINN-sM 15.1 are annotated “typi duplum.” One Aublet specimen at p (p-J 14135) has been annotated as V. speciosa Warming by Stafleu. VOTOMITA GUIANENSIS (1: 91. ¢. 35 (guyannensis))—NG, TS—Melastomata- ceae = Votomita guianensis Aublet. For discussion of this taxon, see Mor- ley (Bull. Torrey Bot. Club 90: 1-16. 1963), who cited p-r 4: 139 as the holotype. BM. VOUACAPOUA AMERICANA (Suppl. 9. t. 373)—NG, TS—Leguminosae = Voua- capoua americana Aublet. Andira Juss. is a conserved generic name (#3841), with the type species listed as ‘4A. racemosa Lamarck ex Jaume Saint- Hilaire (Dict. Sci. Nat. 2: 137. 1804).” The genus is credited to Jussieu (Gen. PI. 363. 1789 sem. 2). In fact, Lamarck published the name Andira racemosa (Encycl. Méth. Bot. 1: 171) in 1783. The monotypic Vouacapoua Aublet is incorrectly listed as a nomen rejiciendum. Baillon (Adansonia 9: 206-212. pl. 4. 1868-70) discussed this taxon, as did Huber (Bol. Mus. Paraense Hist. Nat. 6: 221-225, 1910), and Amshoff (in Pulle, 1939, 2(2): 88) has recognized this genus and species. Lamarck (/oc. cit.) expressed doubt when he listed Vouacapoua Aublet as a variety in his description of A. racemosa. Presumably it was this listing that suggested the conser- vation of Andira. Urban (Symb. Antill. 8: 300. 1920) noted that the Aublet genus differs from Andira in having a dehiscent fruit. BM, P-) 15671. VOUAPA BIFOLIA (1: 25. t. 7) -NG, TS— Leguminosae = Macrolobium bifolium (Aublet) Pers. Macrolobium Schreber is a conserved generic name (#3517), with the conserved type species Macrolobium bifolium (Aublet) Pers. See also Outea Aublet. Specimen at BM annotated “isotype” by R. Cowan; LINN-SM 77.2, P-R 1: 47. VOUAPA SIMIRA (1: 27. ¢. 8)— Leguminosae. Although Vouapa Aublet was a nomen rejiciendum with the conservation of Macrolobium, R. Cowan (Mem. New York Bot. Gard. 8: 335, 336. 1953) saw an isotype at Bm and concluded, “It represents no recognized species of Macrolobium.” He placed Macrolobium simira (Aublet) Gmelin in the synonymy of Vouapa simira, which is not placed. BM, p-R 1: 47. VOUARANA GUIANENSIS (Suppl. 12. t. 374 (guyannensis))—-NG, TS—Sapin- daceae = Vouarana guianensis Aublet. Bo. VOYARA MONTANA (Suppl. 26. 4. 383)—NG, TS—Capparaceae = Capparis montana (Aublet) Lemée. Lemée (1955, p. 666) used “‘Capparis mon- tana? (Voyara m. Aubl. selon H. Hallier dans Meded. Riks. Herb. Leiden XXXV, 1918, 23).” The combination in Capparis is credited to Lemée in Index Kewensis (Suppl. 13, p. 23. 1966); however, in Supplement 9, p. 298 (1938), the genus Voyara was referred to Capparis. BM. VoyRIA CAERULEA (1: 211. ¢. 83, fig. 2)—Gentianaceae = Voyria caerulea Au- blet. Maas (pers. comm.) indicated that the “type” is at Pp. P-R 8: 262. VoyRIA ROSEA (1: 209. ¢. 83, fig. 1)-NG, TS—Gentianaceae = Voyria rosea Aublet. Maas (pers. comm.) indicated that the “type” is at p. p-R 8: 262. 1983] HOWARD, AUBLET 291 WARIA ZEYLANICA (1: 605. t. 243)—Annonaceae = Uvaria zeylanica L. Waria Aublet is regarded as an orthographic variant of Uvaria L. Aublet cited “L. Sp. 756,” which refers to the second edition of Species Plantarum. No supporting specimens were located. XIPHIDIUM COERULEUM (1: 33. ¢. 11) —NG, TS—Haemodoraceae = Xiphidium coeruleum Aublet. No supporting specimens were located XYLOPIA FRUTESCENS (1: 602. ¢. 242)—Annonaceae = Xylopia frutescens Au- blet. No supporting ee were located. XYRIS AMERICANA (1: 40. ¢. /4)—Xyridaceae or Abolbodaceae = Abolboda americana (Aublet) ee Specimen at BM annotated “‘type’’; P-R 1: 53. ACKNOWLEDGMENTS The work on the manuscript of Alexander Anderson that led to this study has been supported by a grant from the Stanley Smith Horticultural Trust, for which I am deeply appreciative. I am grateful for many kindnesses and the cooperation of the officers, the curatorial staff, and the librarians of the British Museum (Natural History), the Linnean Society, and the Muséum d’Histoire Naturelle of Paris. Many colleagues have replied to questions regarding Aublet names and specimens. In 1950 Dr. Fred Meyer, then of the Missouri Botanical Garden, photographed most of the Aublet specimens in the British Museum. A list of many of these, supplied from the records of photographic negatives at the Missouri Botanical Garden, was most helpful in the search for supporting specimens. LITERATURE CITED AuBLeT, F. 1775. Histoire des plantes de la Guiane frangoise. 4 vols. Didot, Paris. BERNARDI, L. 1976. J.-B. C. Fusée-Aublet, le brave botaniste de la onziéme heure. Mus. Genéve 169: 2-10. Farr, E. R., J. A. Leussink, & F. A. StArLeu. 1979. Index nominum genericorum n m. Reg. Veg. 100: 1-1896. FROIDEVAUX, "HL 1897. Etude sur les recherches scientifiques de Fusée Aublet a la Guyane francaise (1762-1764). Bull. Géogr. Hist. Descr. 1897: 425 Hauer, H. 1918. Uber Aublet’s Gattungen unsicherer oder unbekannter Stellung und iiber pflanzengeschichtliche Beziechungen zwischen Amerika und Afrika. Meded. Rijks-Herb. Leiden 35: 1-83. Henrey, B. 1975. British botanical and horticultural illustrations before 1800. Oxford Univ. Press, Oxford, Englan HurtcuHInson, J. 1964. The genera of flowering plants. Vol. 1. Dicotyledones. Clarendon Press, Oxford, England pee J., & H. Urrrien. 1940. Un nouvel herbier de Fusée Aublet découvert en ce. Rec. Trav. Bot. Néerl. 37: 135-170. eee J. 1968. Un pharmacien provencal découvre une partie de la flore du nouveau monde: J. B. C. naee ee et I’“‘Histoire des plantes de la Guyane frangaise.” Adansonia, II. 8: 6. LeMEE, A. 1952- Gs aS de la Guyane francaise. 4 vols. Librairie Lechevalier, ris. MaaurrE, B., et al. 1972. The botany of the Guayana Highland, part IX. Mem. New York Bot. Gard. 23: 1-832. 292 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Putte, A. 1932-1975. Flora of epee Vereen. Kol. Inst. Amsterdam, Meded. 30, Afd. Handelmus. 11. 6 vols., 11 part SANDWITH, N.Y. 1931. Newand aos species from British Guiana, Dilleniaceae— ee Kew Bull. 1931: 170-188. 37. Notes on tropical American Bignoniaceae. Meded. Bot. Mus. Utrecht 40: oye 232. 1942. On the identifications of Aublet’s four species of Caraipa. Jour. Bot. London 80: 51-54. 62. Notes on Bignoniaceae XXV. Proposed lectotypes of certain genera. Kew Bull. 15: 453-466. 63. Notes on some Aublet types in the Paris herbarium. /bid. 17: 257- te SCHREBER, J.C. D. von. 1789, 1791. Carolia Linné... Genera plantarum ... editi octava. 2 vols. Varrentrapp & Wenner, Frankfurt. Scopo.tt, G. A. 1777. Introductio ad historiam naturalem. Wolfgang Gerle, Prague. SmitH, A. C. 1940. The American species of Hippocrateaceae. Brittonia 3: 341-554. SOUILUAERT, G., & F. A. SraFLEU. 1953. The Tristan herbarium in Orléans. Taxon 2: 3-25. STAFLEU, F. A. 1971. Linnaeus and the Linnaeans. A. Oosthoek, Utrecht. & R.S. Cowan. 1976. Taxonomic literature. ed. 2. Vol. 1. Bohn, Scheltema & Holkema, Utrecht. STEYERMARK, J. A. 1974. Rubiaceae. /m: T. LAsser, ed., Fl. Venezuela 9: 1-593, LLO1- 207 Urpan, I. 1919. Sertum antillanum IX. Repert. Spec. Nov. 16: 149-151. —. 1920. Plumiers Leben und Schriften. Repert. Spec. Nov. Beih. 5: 1-196. Wi us, J.C. 1973. A dictionary of the flowering plants and ferns. ed. 7 (revised by H. K. Airy SHAw). University Press, Cambridge, England. Wurback, J. J. 1973. Melastomataceae. Jn: T. Lasser, ed., Fl. Venezuela 8: 1-819. ——. 1980. Melastomataceae. Jn: G. HARLING & B. SpARRE, eds., Fl. Ecuador 13: 406. ARNOLD ARBORETUM 22 Divinity AVENUE CAMBRIDGE, MASSACHUSETTS 02138 1983] WOOD, BURMANNIACEAE 293 THE GENERA OF BURMANNIACEAE IN THE SOUTHEASTERN UNITED STATES! CARROLL E. Woop, JR. BURMANNIACEAE Blume, Enum. PI. Javae 1: 27. 1827, nom. cons. (BURMANNIA FAMILY) Annual (or perennial), autotrophic, partly autotrophic, or achlorophyllous “saprophytic” herbs. Leaves alternate, simple, entire, exstipulate, mostly re- duced to scales, the photosynthetic species leafy stemmed, with or without a basal rosette of small, linear leaves. Flowers perfect, regular (actinomorphic) [or zygomorphic or rarely 2-lipped]. Inflorescences mostly monochasia or dis- chasia, often reduced to a single terminal flower or condensed and headlike. Perianth petaloid, white or colored, of 2 whorls of 3 tepals united to form a tube, the 3 inner usually smaller than the outer in Burmannieae [or the 3 outer smaller, often almost lacking in Thismieae]; perianth tube cylindrical or 3-angled, often 3-winged [or 3- or 6-ribbed]. Anthers 3 and opposite the inner perianth lobes [or in Haplothismia and most Thismieae 6], sessile or subsessile in the perianth throat [or with short recurved filaments], 4-loculate, latrorse [or in- trorse], connective broad, often appendaged at top and/or base; pollen grains single or in tetrads [or dyads], 1 - or 2-porate [4-porate or 1-colpate]. Gynoecium 3-carpellate, syncarpous; style single, filiform [or short-cylindrical or conical in Thismieae], tipped with 3 short branches, each with a single stigma [or with 3 sessile or connate stigmas]; ovary inferior, 3-locular with axile placentae or 1-locular with 3 parietal placentae [in some genera with a large, globose gland inside the ovary on each side of the top of each placenta!]; ovules numerous, 'Prepared for the G ic Fl fthe Sout! United States, a project of the Arnold Arboretum currently made Peace through the ce of the National Science Foundation, under Grant DEB- 81- 11520 (Carroll E Wood, Jr., and Norton G. | em ae investigators). This treatment, the rst rb. 39: The descriptions are based primarily on the plants of this area but with information about extra- regional members of a family or genus in brackets [ ]. References that I have not verified are marked with an asterisk. I am indebted to my coe Newon oP Nene Saas K. Rogers, and Barbara Nimblett for their help with hristopher S. Campbell for Kodachrome pictures and alcohol- -preserved material of Apteria and of Bum annia biflora; and to Walter S. Judd for alcohol- -prese mannia capitata and fruits of Apteria. The illustration of Apteria was drawn by Margaret van Montfrans from my dissections of the specimens sent by Drs. Campbell and Judd © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 293-307. April, 1983. 294 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 minute, anatropous, 2-integumented, tenuinucellar; female gametophyte of the Polygonum or the Allium type. Perianth persistent on the fruit [or the upper part bearing the stamens deciduous, only the lower part persistent, or in This- mieae the entire perianth circumscissile, leaving only a basal thickened ring]. Fruit capsular [sometimes fleshy in Thismieae], dehiscing irregularly or splitting transversely [or at the top] or sometimes by valves. Seeds numerous, small: embryo minute, undifferentiated, sometimes composed of only a few (3-6) cells; endosperm present, usually consisting of relatively few, large cells. (In- cluding Thismiaceae J. G. Agardh, nom. cons.; excluding Corsiaceae Beccari, nom. cons.) TYPE GENUS: Burmannia L. A small family of 17 or 18 genera and about 135 species of perennial or annual mycotrophic herbs, varying from leafy-stemmed, green autotrophs to scale-leaved, achlorophyllous “saprophytes.’’ The family occurs primarily in the tropics and subtropics of North and South America, Africa, Asia, and Australia, but with extensions northward to Virginia and Illinois in the United States and to southern Japan, and southward to Brazil and Argentina; Angola, South Africa, and Madagascar; and southeastern Australia, Tasmania, and northern New Zealand. The distribution of Burmannia L. (q.v.) is nearly that of the family; Gymnosiphon Blume occurs in Central and South America, the West Indies, tropical Africa, Madagascar, Malaysia, Indonesia, and New Guinea; and Thismia occurs in the Americas, tropical Asia, and Australia-New Zealand. Twelve genera are indigenous in South America, and six of these extend into North America, with Burmannia L. and Apteria Nutt., of tribe Burmannieae, and Thismia, of tribe Thismieae, reaching the United States. Although most genera are confined to the Americas, to Africa, or to Asia-Australia, there are some truly extraordinary disjunctions. Glaziocharis Taub. ex Warm. has two species: G. macahensis Taub. ex Warm., known only from Rio de Janeiro, Brazil, and G. Abei Akasawa, known from Shikoku and southern Kyushu, Japan. Thismia Griff. sect. Ropwaya Schlechter includes only two closely related species: 7. Rodwayi F. Mueller (Bagnisia Hillii Cheeseman), of Aus- tralia (eastern Victoria and Tasmania) and New Zealand (North Island), and T. americana N. E. Pfeiffer, known from only one colony discovered on low prairies near Chicago in 1912 and presumably now extinct. A more familiar pattern is seen in Burmannia, in which there are several tropical American— tropical Asian disjunctions. The family is generally regarded as being composed of two or three groups of genera that have been ranked by various authors as tribes, subfamilies, or families (see Jonker, 1938; Lawrence). Bentham & Hooker, Baillon, and Engler, among others, considered the Burmanniaceae to consist of tribes Burmannieae (Euburmannieae), Thismieae, and Corsieae. Thorne recognized these same groups at the rank of subfamily, while Hutchinson treated them as distinct families. Jonker, Takhtajan, Cronquist, and others have maintained Corsiaceae and Burmanniaceae (with tribes Burmannieae and Thismieae) as distinct fam- ilies. Airy Shaw established a new tribe, Haplothismieae, to accommodate Haplothismia annulata Airy Shaw (of South India) in Burmanniaceae, kept tribes Burmannieae and Thismieae, and maintained Corsiaceae as a separate family. 1983] WOOD, BURMANNIACEAE 295 The Burmanniaceae have been placed variously in relation to other families of monocotyledons. They have often been associated with the Orchidaceae on the basis of their minute seeds that supposedly lack endosperm. Endosperm, although scant, does occur, however, and Jonker (1938, p. 6) wrote, “The presence of endosperm and the construction of the flower makes relationship with the Liliiflorae highly probable. Of this order Amaryllidaceae, Iridaceae and Taccaceae seem to be most closely related.” Hutchinson (1973, p. 838) placed the Burmanniales (Burmanniaceae, Thismiaceae, Corsiaceae) “next to Haemodorales, and especially near Hypoxidaceae (Curculigo).’’ He remarked, however, that “the affinity with Orchidaceae is probably closer than at first appears, the two groups having developed on parallel lines with regard to certain characters such as saprophytism and minute seeds.” At present, most authors seem to favor a position for Burmanniaceae in or near the Liliales, but Cron- quist maintains the order Orchidales to include Orchidaceae, Geosiridaceae (removed from Burmanniaceae by Jonker), Burmanniaceae, and Corsiaceae. Melchior, Thorne, and Takhtajan all have favored Liliales (Liliiflorae), Mel- chior placing Burmanniaceae and Corsiaceae in suborder Burmanniineae and Thorne linking Burmanniaceae (with subfamilies Burmannioideae, Thismioi- deae, and Corsioideae) and Iridaceae (with subfamilies Iridoideae and Geo- siridoideae) in suborder Iridineae. Takhtajan thinks the Burmanniaceae to be closely related to Liliales, especially Iridaceae, and evidently derived from them. Many species of Burmanniaceae are known only from single collections or localities. As a result, although general morphology is the best-known feature of the group, the detailed morphology, mycorrhizal relationships, embryology, and palynology are known for only relatively few species, and the physiological ecology, floral biology, cytology, and chemistry are nearly untouched. The species of Burmannia, the largest genus of the family, range from leafy, green autotrophs to achlorophyllous “saprophytes.”” The other members of the family appear to be achlorophyllous. Those plants that have been studied are endomycorrhizal, and it has long been assumed that the achlorophyllous Bur- manniaceae, most of which ‘“‘occur almost exclusively on decaying leaves, wood and roots in the deep shade of tropical, wet, primeval forests” (Jonker, 1938, p. 13) are saprophytes. However, in view of research on Monotropa (Ericaceae subfam. Monotropoideae) and Orchidaceae (see Furman & Trappe), it seems much more probable that these, along with similar achlorophyllous plants such as Voyria (Leiphaimos) (Gentianaceae) (see Wood & Weaver), Sciaphila (Triu- ridaceae), and Epirrhizanthes (Polygalaceae) that sometimes grow intermixed with them, are instead epiparasites that parasitize the roots of vascular plants, although not directly attached to them, via a shared mycorrhizal fungus. Jonker (1938, p. 12) remarked about Thismia americana: “‘It 1s hard to believe that Chicago is the normal area for this species, but I cannot give a satisfactory explanation why it occurs there. The habitat, an open prairie among the moss, is very different to the habitats of other Thismias, usually growing saprophyt- ically in primeval forests.’’ Epiparasitism of the Monotropa type makes such a habitat both quite possible and believable. The family is of no economic importance. 296 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 REFERENCES: Airy SHAw, H. K. A new genus gears of Burmanniaceae. Kew Bull. 7: 277-279. 1952. [Haplothismia exannulat ry Shaw from South India, Travancore-Cochin State; new tribe, Haplo ie co Illustrated. “‘The floral structure of Haplothismieae represents the least specialised type found in the Burmanniaceae.”’] AgsiLvsGi, G. Wild flowers of the Big Thicket, East Texas, and agen Louisiana. xil + 361 pp. College Station, Texas. 1979. [Burmannia biflora, 122, 123 (colored illus., 55); Apteria aphylla, 123 Cee illus., 56).] AKASAWA, Y. A new species of Glaziocharis (Burmanniaceae) found in Japan. (In English, Latin, Japanese.) Jour. Japan. Bot. 25: 193-196. pls. 1, 2. 1950. [G. Abei, Awa Prov hikoku; genus previously known from one collection from Brazil: see also C. Ane ibid. 197-199, in Japanese. ] AoyaAMaA, M., K. KArAsAwa, & R. TANAKA. Chromosomes of oo Abel, a saprophyte. Chromosome Inform. Serv. 25: 34, 35. 1978. [2n 2.] BAILLOoN, H. Burmanniacées. Hist. Pl. 13: 170-182. 1894. BENTHAM, G., & J. D. Hooker. Burmanniaceae. Gen. Pl. 3: 455-460. 1883. BOLKHOVSKIKH, Z., V. Grir, T. MATVEJEvA, & O. ZAKHARYEVA. Chromosome numbers of flowering plants. A. A. FEpeRov, ed. 926 pp. Acad. Sci. USSR. V. L. Komarov Bot. Inst. ee hae 1969. [Burmanniaceae, 181. Bowpben, B. N. Burmanniaceae. Pp. 320, 321 in V. H. Heywoop, consultant ed., Flow- ering plants on the world. New York. 1978. [Includes inaccurate distribution map; Burmannia coelestis, Afrothismia sp., Haplothismia exannulata ie - d. Brape, A. C. Sinopse das “Burmanniaceae” da flora do Brasil. Arq. Jar ot. Rio a 7: 11-42. [10 as _ folded map not paged.] 1947 (publ. ee (aa 13 genera; 11 spp. Burmannia, 4 spp. Apter BREWBAKER, J. L. The distribution and ee eee of binucleate and trinu- cleate pollen grains in . Am. Jour. Bot. 54: 1069-1083. 1967. [Data on 1908 genera, approximately half studied by ihe oe CHAKRAPANI, P., & B. Ras. Pollen morphological studies in Burmanniaceae. Grana 11: 164-179. 1971. (27 species in 8 genera studied.] — T. F. Bagnisia Hillti, Cheesem. A new species of Burmanniaceae from New Zealand. Bull. Misc. Inf. Kew 1908: 419-421. 1908. [= Thismia Rodwayi F. Mueller fide Jonker. ] . Notice of the discovery of a species of Burmanniaceae, a family new to the New Zealand flora. Trans. New Zealand Inst. 41: 141-143. 1908. [Bagnisia Hillii, North Island of New Zealand. = Thismia Rodwayi.] Cirerri, R. L’ “habitat” e¢ la micorrizia di alcune Burmanniaceae della Repubblica Dominicana. Atti Ist. Bot. Univ. Pavia, V. 7: 25-34. 1946.* [Includes key and new formae. ay oe D. G. Sarcosiphon Rodwayi in Australia. Victorian Nat. 52: 163-166. p/. 19. 936. [Thismia Rodwayi.] . Further notes on “fairy lanterns.” [bid. 67: 167, 168. p/. 18. 1941. [T. Rodwayi.] Corre, D. S., & H. B. Correct. Aquatic and wetland plants of southwestern United States. xvi + 1777 pp. Environmental Protection Agency, Washington, D. C. 1972. (Reissued in 2 vols. by Stanford Univ. Press. 1975.) [Burmanniaceae, 686, 688, 689: Burmannia capitata, B. biflora, Apteria aphylla; illustrated. See also D. S. CorRELL M. C. JoHNston, Manual of the vascular plants of Texas, 1970: Burmanniaceae, 433, 434] Cronguist, A. An integrated system of classification of flowering plants. 1262 pp. New York. 1981. [Orchidales: Geosiridaceae, Burmanniaceae, Corsiaceae: Burmanni- aceae, 1236, 1237. Davis, G. L. Systematic embryology of the angiosperms. x + 528 pp. New York. 1966. [Burmanniaceae, 63. EIcHLer, A. W. Bliithendiagramme. Erster Theil. 348 pp. Leipzig. 1875. (pennaneee 178, 179.] 1983] WOOD, BURMANNIACEAE 297 ENGLER, A. Burmanniaceae. Nat. Pflanzenfam. II. 6: 44-51. 1888; Nachtrage zu II. 6: 96. 1897. Prana G. Pollen morphology and plant taxonomy. Angiosperms. 539 pp. Stockholm. 952. (Corrected reprint with addendum [pp. 541-553]. New York. 1971.) [Bur- manniaceae, 84, 85. Ernst, A., & C. BERNARD. Beitrage zur Kenntnis der Saprophyten Javas. I-XV. Ann. Jard. Bot. Buitenzorg 23: 20-61. pls. 9-17. 1910; 24: 55-97. pls. 8-17. 1911; 25: 161-188. pls. 13-17. 1912; 26: 219-257. pls. 17-22. 1912; 28: 99-124. pls. 14-19. 1914. Titles and references to individual papers I-XV follow: [Introduction.] /bid. 23: 20-31. I. Zur Systematik von Thismia javanica (by J. J. SMitH). 23: 32-35. pis. 9, 10. U1. Aussere und innere Morphologie von Thismia javanica J.J.S. 23: 36-47. pls. 11-13. Ii. Embryologie von Thismia javanica J.J.S. 23: 48-61. pls. 14-17. IV. Zur Systematik von Thismia clandestina Miq. und Thismia Versteegii J. J. Sm. 24: 55-60. pis. 8, 9. V. Anatomie von Thismia clandestina Mig. und Thismia Versteegii Sm. 24: 61-69. pls. 10-12. VI. Beitrage zur Embryologie von Thismia clandestina Mig. und Thismia Versteegii Sm. 24: 70-78. pls. 12, 13. VII. Zur Systematik von Burmannia candida Engl. und Burmannia Championii Thw. (by J. J. Smity). 24: 79-83. pls. 14, 15. VULL. Aussere und innere Morphologie von Burmannia candida Engl. und Burmannia Championii Thw. 24: 84-97. pls. 16, 17. IX. Entwicklungs- Sper des Embryosackes und des Embryos von Burmannia candida Eng]. und B. Championii Thw. 25: 161-188. pls. 13-17. X. Zur Systematik von Burmannia Pee Don (by J. J. SmitH). 26: 219-222. pi. 17. XI. Aussere und innere Mor- phologie von Burmannia coelestis Don. 26: 223-233. pl. 18. XII. Entwicklungsge- schichte der Embryos und des Endosperme von Burmannia coelestis Don. 26: 234- Becc. 28: 102-120. pls. 15-18, 18*. XV. Embryologie von Burmannia tuberosa Becc. 28: 121-124. p/. 19 Espinosa B., M. R. Algunas ee sobre la Arachnites uniflora. Phil. Revista Chilena Hist. Nat. 30: 299-303. ; FuRMAN, T. E., & J. M. TRAPPE. ane and ecology of mycotrophic i ag angiosperms. Quart. Rev. Biol. 46: 219-225. 1971. [Burmannia, Glaziocharis, Gym nosiphon, pete Thismia listed Anes genera of mycotrophie achlorophy!- M a per pager (as By laead): ] GODFREY, , & J. W. Wooten. Aquatic and wetland plants of southeastern United States. Sere citer xi + 712 pp. Athens, Georgia. 1979. [Burmanniaceae, 623- 625; illustrations of Apteria aphylla, Burmannia biflora, B. capitata (from Correll & Corre Shae K., & K. SUssENGUTH. Beitrage zur Kenntnis der siidamerikanischen Burman- een. Flora 117: 55-90. pls. 2, 3. 1924. [Morphology and anatomy; Thismia ees, Cymbocarpa Urbanii illustrated. Groom, P. Thismia Aseroe (Beccari) and its mycorhiza. Ann. Bot. 9: 327-361. pis. 13, 14, 1895 HATSCHBACH, cc? & O. GuimarAes. Burmanniaceas do Estado do Parana. Bol. Mus. Bot. Munic. Curitiba 5: 1-17. p/s. 1-4. 1972. [Brazil. Includes Burmannia bicolor, i capitata, B. alba, Cymbocarpa refracta, Gymnosiphon tenellus, Apteria genti- noides, Dictyostega orobanchoides, Miersiella umbellata. Paar ee J. The families of flowering plants. ed. 3. xx + 968 pp. Oxford. 1973. 691), the order Burmanniales is composed of Burmanniaceae, Thismaiaceac: and Corsiaceae Jouow, F. Die chlorophyllfreien Humusbewohner West-Indiens, biologisch-morpholo- gisch dargestellt. Jahrb. Wiss. Bot. 16: 415-449. pls. 16-18. 1885. [Includes Apteria aphylla (as A. setacea), Burmannia capitata.] 298 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 — hl lif ach ihren biologischen und anatomisch- Ae veniam mech Verhlisisen. Ibid 20: 475-525. 1889. [Includes Bur- mannia capitata, Gymnosiphon, Dictyos Jones, S. B. Mississippi flora. VI. ee families. Castanea 41: 189-212. 1976. [Burmanniaceae, 195, 196; Apteria aphylla, Burmannia biflora, B. capitata; Apteria and 8B. capitata faapeed: 196.] JonKER, F. P. A monograph of the Burmanniaceae. Meded. Bot. Mus. Utrecht 51: i-v, 1-279. 1938. [Introduction, 1-3; General part, 4-14; Critical part, 15-47; Taxo- nomic part, 49-273; Index, 275-279. Burmannia, 18-26, 57-163; Apteria, 35-37, 205- et Les Géosiridacées, une nouvelle famille de Madagascar. Rec. Trav. Bot. Néerl. 36: 473-479. 1939. (Meded. Bot. Mus. Utrecht 60.) [Geosiridaceae, fam. nov., with Geosiris aphylla Baillon, excluded from Burmanniace urmanniaceae. Jn: R. E. Woopson, Jr., & R. W Sc CHERY & COLLABORATORS. Fl. ‘Panama II(1). Ann. Missouri Bot. Gard. 32: 42-47. 1945. [Apteria aphylla, Burmannia capitata, Gymnosiphon panamensis, G. suaveolens, Thismia panamen- sis. . Burmanniaceae. Jn: C. G. G. J. VAN STEENIS, Fl. Malesiana I. 4: 13-26. 1948; 5: 553, 554 (corrections). 1958. [15 spp. Burmannia, 7 spp. Gymnosiphon, 14 spp. Thismia, | sp. Geomitra, 2 spp. Scaphiophora.| Kores, P., D. A. Wuite, & L. B. THEN. Chromosomes 2 Corsia (Corsiaceae). Am. Jour. Bot. 65: 584, 585. 1978. [C. cornuta, C. clypeata, = 18.] Lawrence, G. H. M. Taxonomy of vascular plants. xiii ‘i+ 823 pp. New York. 1951. [Burmanniaceae, 431, 432; Burmannia biflora illustrated.] Lussock, J. A contribution to our knowledge of seedlings. Vol. 2. 646 pp. London. 1892. [Burmanniaceae, 561, 562; Apteria aphylla, Burmannia capitata.] Maas, P. J. M. Neotropical saprophytes. Pp. 365-370 in K. LARSEN & L. B. HOLM-NIELSEN, eds., Tropical botany. London, New York, San Francisco. 1979. [Brief accounts of members of Gentianaceae, Triuridaceae, Burmanniaceae. ] McLennan, E. I. Thismia Rodwayi F. Muell. and its endophyte. Austral. Jour. Bot. 6: 25-37. pls. 1-4. 1958. [Excellent account nantes mycorrhizal association. | Meccuior, H. Unterreihe Burmannineae (Burmanniales). 7n: H. MELCHIOR, Engler’s Syllabus der Pflanzenfamilien, ed. 12. 2: 538, 539. 1964. ee Corsiaceae assigned to suborder Burmannineae of the order pees (Liliales Meyer, K. J. Parthenogenesis bei Thismia javanica im Lichte der Haberlandtschen _ Miers, J. On some new Brazilian plants allied to the natural order Burmanniaceae. Trans. Linn. Soc. London 18: 535-556. pls. 37, 38. 1841. [Gonyanthes, Dictyostega, Cym- bocarpa, es - vars described, A. aphylla (as A. setacea) discussed. Netouitzky, F. Ana e der Angiospermen-Samen. Handb. Pflanzenanat. II. Arche- 0. vi+ se pp. 1926. [Burmanniaceae, 91, 92.] PepiereR. N. E. Morphology of Thismia americana. Bot. Gaz. 57: 122-135. pls. 7-11. 1914. —_— € sporangia of Thismia americana. Ibid. 66: 354-363. pl. 16. eee Raion DE LA Sota, E. Mamorea Singeri, un nuevo género y especie de Burmanniaceae Darwiniana 12: 43-47. 1960. [From Bolivia; related to ene tribe Thismie ae.] RaAmspoTtToMm, J. Orchid mycorrhiza. Trans. Brit. Mycol. Soc. 8: 28-61. 6 unnumbered pls. 1922. [Includes Burmanniaceae, Ericaceae, Pyrolaceae, Gentianaceae, and Lol- ium (Gramineae). | Rao, V. S. Certain salient features in the floral anatomy of Burmannia, Gymnosiphon and Thismia. Jour. Indian Bot. Soc. 48: 22-29. 1969. [B. disticha, B. nepalensis, G. cymosus, G. aphyllus, T. Aseroe.] RENDLE, A. B. The classification of flowering plants. Vol. 1. Gymnosperms and mono- cotyledons. ed. 2. xvi + 412 pp. Cambridge, England. 1930. [Burmanniaceae, 344, 345, 393.] 1983] WOOD, BURMANNIACEAE 299 oes H. W. Wildflowers of the United States. Vol. 2. The Southeastern States. Part + 322 pp. New York. 1966. [Burmanniaceae, 78, ie el oe 25, 26. a mannia biflora, B. capitata, Apteria aphylla illustrated.] V 3. Texas. Part 1. 274 pp. 1969. [Burmanniaceae, 50; Burmannia capitata, yee eiile 57 Ol. 7) (illustration same as in Vol. 2.] Rip.ey, H. N. The Burmanniaceae of the Malay Peninsula. Jour. Straits Branch Roy. Asiat. Soc. 22: 331-339. 1890. [Burmannia longifolia, B. disticha, B. coelestis, B. tuberosa, B. gracilis, Gymnosiphon borneense, Thismia Aseroe, T. fumida; B. tu- berosa, B. gracilis said to be saprophytic Royen, P. van. Corsiaceae of New Guinea and surrounding areas. Webbia 27: 223-255. 1972. [23 spp.; Corsiaceae consisting of ca. 25 spp. in New Guinea, | in Australia, 2 in the Solomon Islands, and | (of -Arachnitis) i in Chile. SCHLECHTER, R. Die Thismieae. Notizbl. Bot. Gart. Berlin 8: 31-45. 1921. [10 genera; 6 later recognized by Jonker (1938), the others combined with Thismia.] SeuBERT, M. Burmanniaceae. In: C. F. P. von Martius, Fl. Brasil. 3(1): 53 60. 1847. {Includes Apteria lilacina, Burmannia bicolor, B. capitata, B. flava. SHIN, T. Two species of the genus Glagiocharis [sic] (Burmanniaceae) from southern Kyushu. (In Japanese.) Jour. Japan. Bot. 49: 3-6. 1974. [Glaziocharis macahensis (described from Brazil) and G. Abei; generic name spelled Glagiocharis throughout. ] SmiTH, J. J.: see Ernst, A., & C. BERNARD, Beitrage zur Kenntnis der Saprophyten Javas, I, VII, X, XIII. Stone, B. C. Rediscovery of Thismia clavigera (Becc.) F. v. M. (Burmanniaceae). Blumea 26: 419-425. 1980. [Geomitra clavigera transferred to Thismia; includes key to the nine spp. of Thismia known from Malaya Swamy, B. G. L., & K. V. KRISHNAMURTHY. The helobial endosperm. A d ial Phytomorphology 23: 74-79. 1973 (publ. 1974). (Burmanniaceae, 75; development of endosperm in Burmannia coelestis and B. pusilla conforms to A- form of helobial TAKHTAJAN, A. L. Outline of the classification of flowering plants (Magnoliophyta). Bot. Rev. 46: 225-359. 1980. [Burmanniales (Burmanniaceae, Corsiaceae), 313, 358, closely related to Liliales, especially Iridaceae, and evidently derived from them.] THorne, R. F. A phylogenetic classification of the Angiospermae. Evol. Biol. 9: 35-106. 1976. [Order Liliales composed of suborders Liliineae, Iridineae, Orchidineae; Iri- dineae composed of Iridaceae and Burmanniaceae, 64, 65, 96, 97. Warp, D. B. Checklist of the vascular flora of Florida. Part 1. Univ. Florida Agr. Exper. Sta. Tech. Bull. 726. 76 pp. 1968. [Apteria aphylla, Burmannia biflora, B. capitata, B. flava, 57 Wiiuams, L. O. Notes on the family Corsiaceae. Harvard Univ. Bot. Mus. Leafl. 12: 179-183. pl. 24. 1946. [Includes 2 new spp. of Corsia.] Wiis, J. C. A dictionary of the flowering plants and ferns. ed. 7. Revised by H. K. SHAW. xxii + 1214 pp. + key to families of flowering plants (pp. i—lii). Cam- bridge, England. 1966. [Burmanniaceae, 166, 167, key ix; Corsiaceae, 286, 287.] Woop, C. E., Jr., & R. E. WeAver, Jr. The genera of Gentianaceae in the southeastern United States. Jour. Arnold Arb. 63: 441-487. 1982. [Voyria, 483-487.] KEY TO THE GENERA OF BURMANNIACEAE IN THE SOUTHEASTERN UNITED STATES General characters: Small, erect, green or purplish herbs with solitary or cymose flow- ers; perianth petaloid, regular, the tepals in 2 whorls of 3, united; anthers 3, nearly sessile n the perianth below the inner whorl of tepals, opening horizontally, the connective appendaged; gynoecium 3-carpellate, syncarpous, the ovary inferior; fruit a capsule topped 300 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 by the persistent perianth; seeds numerous, minute, the embryo undifferentiated, the en- dosperm scanty Ovary 3-locular, with axile placentae; flowers with or without wings; capsule opening by horizontal fissures or irregular splits; plants green, with or without a basal rosette of linear leaves and with small linear to lanceolate cauline leaves. ..... 1. Burmannia. Ovary |-locular, with 3 parietal soars flowers wingless; capsule opening from base by 3 upward-curling valves (see Fic. 1, k); plants without basal leaves, stem and scalelike leaves purplish or violet. .....00.00.0000.0 000 ccc cece cece ee 2. Apteria. Tribe BURMANNIEAE Subtribe Burmanniinae 1. Burmannia Linnaeus, Sp. Pl. 287. 1753: Gen. Pl. ed. 5. 139. 1754. Erect, terrestrial [rarely pendent, epiphytic], annual [or perennial], autotro- phic [or “‘saprophytic”’ (more probably epiparasitic)], mycorrhizal herbs. Stems 3-60 cm (ours usually less than 25 cm) tall, usually Dp branched) [in sect. FOLIOSA with conspicuous grasslike ee. veined decurrent leaves to 25 cm long or] with variously reduced, often scalelike leaves and with or without a rosette or a few linear-lanceolate to linear or subulate leaves at the base. Flowers small (in ours 2-12 mm long), blue, pink, white, yellow, or greenish, erect [or rarely pendulous], winged or wingless, usually in the axil of a small bract, basically in bifid cymes (monochasia or dichasia) of 2 to 15 (to many) flowers, but by reduction solitary at the stem tip or in some (e.g., B capitata) the cymes contracted into a head. Perianth with 6 small, erect [rarely spreading] lobes, these often with inrolled margins, the 3 outer lobes smaller than the 3 inner, the tube + cylindrical, 3-angled or 3-winged. Anthers 3, nearly sessile on the perianth below the inner perianth lobes, the connective broadly triangular in ours, bearing the 4 locules at its margins and in most species with 2 apical appendages or crests at its apex and in some species (e.g., B. flava) also with a median appendage at the base; locules dehiscing horizontally; pollen grains single [or in dyads or tetrads], |-porate, 1- or 2-porate, [or 2-porate]. Style filiform, divided into 3 short branches at the top, each branch with a funnel-shaped, bowl-shaped, or peltate stigma [or stigmas sessile at top of style]. Ovary 3-angled, with or without wings, 3-locular, the placentae axile, with numerous minute ovules. Perianth persistent and drying atop the winged or wingless capsule; capsule opening by irregular horizontal fissures. Seeds very numerous, minute, oblong to ellipsoid; embryo undifferentiated; endosperm scarce. (7ripterella Michaux) Type species: B. disticha L., 2n = ca. 136; effec- tively typified by Linnaeus (see below); see also Britton & Brown, Illus. Fl. No. U.S. & Canada. ed. 2. 1: 547. 1913. (Name commemorating the Dutch botanist Johannes Burman, 1706-1779. Linnaeus wrote [in Critica Botanica, 1737: transl. Sir Arthur Hort, 1938], “Burmannia is a plant of Ceylon [Sri Lanka] with a double spike, seeing that Burman collaborated with Hermann in writing of the plants of that country.’’) A genus of pantropical distribution with extensions into cooler areas in the southeastern United States (to southeastern Virginia and southeastern Okla- 1983] WOOD, BURMANNIACEAE 301 homa), southern Brazil (Rio Grande do Sul), Argentina (Corrientes), South Africa (Transvaal and Natal), Madagascar, southeastern Australia, and south- ern Japan (Honshu). Jonker (1938) recoenized 57 species in two sections: Fo.iosa Jonker, including five 1 with cauline, grasslike, parallel- veined leaves; and BURMANNIA, the remaining species, both chlorophyllous and achlorophyllous, with basal leaves (if present at all) reduced to a few linear leaves and stem leaves scalelike. With the exception of Burmannia longifolia Becc., which is widely distributed in Malaya, Indonesia, the Philippine Islands, and New Guinea, the members of sect. FoLIosA are indigenous to the Western Hemisphere. Jonker recognized | 2 species of sect. BURMANNIA in the Americas, 11 in Africa, and 30 in Asia and Australia. Subsequently, two species of sect. Fouiosa (both leafy epiphytes) and three of sect. BURMANNIA have been de- scribed from South America. Several of the species of Burmannia have very wide continental distributions, and Jonker has pointed out, in addition to the geographic disjunction in sect. Fouiosa, close relationships between the West Indian-South American B. bi- color Martius, the African B. /atialata Hua, and the Asiatic B. coelestis G. Don, 2n = ca. 32, which are hardly distinguishable, and between B. tenella Bentham, the only American achlorophyllous species, and B. /utescens Becc., of Malaya. Three chlorophyllous species of sect. BURMANNIA occur in the southeastern United States. Burmannia biflora L., characterized by lack of a basal rosette of leaves and by small, bright blue to violet (to white), strongly three-winged flowers either solitary at the tip of the stem or in 1-15-flowered bifid cymes with a central terminal flower, is distributed sporadically in peaty bogs, in savannas, and on grassy pond shores from Collier County, Florida, northward on the Coastal Plain to southeastern Georgia, the Carolinas, and the south- easternmost counties in Virginia, and westward to western Florida, southern Alabama and Mississippi, northern Louisiana (Ouachita Parish), and south- eastern Texas (Houston and Hardin counties). The tropical American Burmannia capitata (J. F. Gmelin) Martius (7rip- terella capitata Michaux), which has small cauline leaves and only a few basal leaves or none at all and small, pinkish, purple (rarely in our area), greenish, or nearly white, wingless flowers in contracted, terminal, headlike inflores- cences, occurs in a variety of moist to wet acid habitats, including pinelands, seepage slopes, gravel pits, bogs, and pondshores from Lee and Highlands counties, Florida, northward to the central Piedmont of Georgia, and—on the Coastal Plain—to Lenoir and Harnett counties, North Carolina. It extends westward through Florida, southern Alabama, and southern Mississippi, to central Louisiana and eastern Texas (Smith, Robertson, and Gonzales coun- ties), and it occurs disjunctly in southeastern Oklahoma (Pushmahata County). It is widely distributed southward through the West Indies and Central and South America: Cuba, Hispaniola, Puerto Rico, southern Mexico (Chiapas), Belize, Nicaragua, Panama, Colombia, Venezuela, Guyana, Surinam, French Guiana, Trinidad, Brazil, Bolivia, Paraguay, and Argentina (Corrientes). Burmannia flava Martius, also of wide distribution, reaches the United States only in southern Florida, where it has been collected only twice (see Ward, 1979): near Fort Myers, Lee County, in 1916, and in the Fakahatchee Strand 302 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 in central Collier County, in 1946. It occurs south of our area in Cuba, southern Mexico (Chiapas), Belize, Costa Rica, Colombia, Venezuela, Guyana, Surinam, Brazil, Bolivia, Paraguay, and Argentina (Misiones and Corrientes). It is char- acterized by linear-subulate rosulate and cauline leaves, and by greenish flowers with narrow yellow wings and yellow perianth lobes. The anthers of all three species have two apical crests. Anthers of Burmannia capitata are unappendaged at the base, while those of B. biflora have a short, obtuse median appendage, and those of B. flava an acute, pendulous one. At least Burmannia biflora has been found growing with Apteria in our area, and there are records of achlorophyllous species of Burmannia growing with other burmanniaceous genera (see Jonker, 1938, p. 13), but there seem to be no records of two species of Burmannia occurring together. The wide range of ecophysiological types found in the genus—from leafy- stemmed autotrophic epiphytes, through reduced, scale-leaved but chloro- phyllous, terrestrial types, to achlorophyllous “‘saprophytes’’— merits investi- gation. The range within the single genus is perhaps comparable to that ex- emplified by the gentianaceous genera Obolaria-Bartonia-Voyria (see Wood & Weaver) or by the orchidaceous series Listera-Goodyera-Myrmechis-Zeuxine (Z. purpurascens)-Corallorhiza arranged by Montfort & Kiisters (see Furman & Trappe). Although the mycorrhizae and the associated fungus have been described for several species of Burmannia, Van der Pijl’s study of the endo- phytic fungus of B. candida Griff. ex J. D. Hooker, 2n = 12, which he found to be a phycomycete, probably a member of the Peronosporaceae, appears to be the only attempt to identify the fungus associated with a species of Bur- mannia and to determine whether the same fungus is shared by different ‘“‘sap- rophytes.” The floral biology and breeding system do not seem to have been investigated for any species of Burmannia; there are only a few recorded chromosome counts; and there seem to have been no investigations of the chemistry. To judge from the limited liquid-preserved material available, it appears that the flowers of Burmannia capitata are self-pollinated, for the pollen seems to be shed before the flower bud opens, and the stigmas are covered with germinating grains. In B. biflora the small inner tepals are curved inward above the anthers, hiding them from sight, and the three stigmas protrude between the inner tepals. Whether or not our species actually are annuals is not well established. Pollen of 19 species of Burmannia was studied by Chakrapani & Raj, who found that most species have one-porate grains, while B. dasyantha Martius has atreme grains, several species have one- or two-porate grains, and B. congesta (Wright) Jonker has two-porate grains. Pollen grains are single (or in tetrahedral tetrads [or dyads] in B. aprica (Malme) Jonker). Brewbaker found Burmannia to be one of only 10 genera (out of 1908) in which both two- and three-nucleate pollen have been more or less reliably reported. He confirmed the three-nucleate condition of pollen of B. biflora, but the two-nucleate pollen reported for B. Championii Thw., 2n = 12, and B. candida should be recon- firmed. Both the Polygonum and the Allium types of megagametophyte de- 1983] WOOD, BURMANNIACEAE 303 pean sane been described in Burmannia. Burmannia coelestis is apospor- with an unreduced megagametophyte. Embryos are rudimentary and eee Gases four-celled in B. coelestis and ten-celled in B. cap- itata). Endosperm ontogeny is A-form helobial in B. coelestis and B. pusilla (see Swamy & Krishnamurthy). Burmannia appears to be most closely related to the South American genera Campylosiphon Bentham (C. purpurascens Bentham) and Hexapterella Urban (H. gentianoides Urban). REFERENCES: Under family ref AJILVSGI, BAILLON, BENTHAM & HOOKER, BOWDEN, BRADE, BREWBAKER, BOLKHOVSKIKH ef al., CHAKRAPANI & RAJ, CIFERRI, CORRELL & CORRELL, Davis, EICHLER, ENGLER, aa ERNST 7 BERNARD (VII-XV), FURMAN & TRAPPE, Goprrey & Wooten, HATSCHBACH & GUIMARAES, HUTCHINSON (1934, 1959, 1973), ans (1885, 1889), Jones, Fo ece (1938) eee) pp. 18-26, 57-163; 1945, pp. 43, 44; 1948), Lawrence, Luppock, MAaAs, MELCHIOR, NETOLITZKY, RAMSBOTTOM, RAo, RENDLE, RICKETT, RIDLEY, SEUBERT, SWAMY & KRISHNAMURTHY, and WARD. AREKAL, G. D., & S. N. RAMAsSwamy. Embryology of Burmannia pusilla (Wall. ex Miers) THW. [sic] and its taxonomic status. (In English; German abstract.) Beitr. Biol. Pflanzen 49: 35-45. 1973. Duncan, W. H., & L. E. Foore. Wildflowers of the southeastern United States. vii + 6 pp. Athens, Georgia. 1975. [B. capitata, 268, colored illus., 269; B. biflora mentioned, 268.] Ernst, A. Apogamie bei Burmannia coelestis Don. Ber. Deutsch. Bot. Ges. 27: 157- 168. pl. 09. Josepn, J., R. ANSARL: & C. N. MoHANAN. Burmannia Championii Thw.—an addition to the flora of South India. Jour. Bombay Nat. Hist. Soc. 76: 552, 553. 1979 (publ. 1980). [Morphology, distribution, illustration.] LARSEN, K. Studies in the flora of Thailand. 14. Cytological studies in vascular plants of Thailand. Dansk Bot. Ark. 20: 211-275. 1963. [Burmanniaceae, 262, 263: Bur- mannia coelestis, 2n = ca. 32; B. Wallichii, n = 16; B. disticha, 2n = ca. 136. Maas, P. J. M. Notes on New World saprophytes II. Acta Bot. Neca 30: 139-149. 1981. [New species in Sciaphila (Triuridaceae); Burmannia, Gymnosiphon, Dictyostega (Burmanniaceae); new combinations in Voyria (Gentianaceae). Burmannia Jonkeri, B. vaupesiana Van Benthem & Maas, new spp., illustrated.] Martius, C. F. P. von. Burmannia. Nov. Gen. Sp. Pl. Brasil. 1: 9-12. pi. 5. 1823. [B. bicolor, B. flava, B. dasyantha.] Mo -Fino, J. F. Una Burmaniacea interesante para Misiones: “‘Apteria lilacina” Miers. Physis 6: nial pl. 1923. [B. flava noted as collected near Paggi, Alto Uruguay, Argentin Pal, R. M. Shidies in the floral morphology and anatomy of the Burmanniaceae. I. Vascular anatomy of the flower of Burmannia pusilla (Wall. ex Miers) Thw. Proc. Indian Acad. Sci. B. 63: 304-308. 1966. [Cf. RAo under family references.] Put, L. vAN DER. Die Mycorrhiza von Burmannia und Epirrhizanthes und die Fort- pflanzung ihres Endophyten. Rec. Trav. Bot. Néerl. 31: 761-779. 1934. [B. candida Engler, B. coelestis G. Don, E. elongata Blume; mycorrhizal fungus of B. candida a phycomycete, cf. eu. RApForp, A. E., H. E. Antes, & C. R. BELL. Manual of the vascular flora of the Carolinas. Ixi + 1183 pp. Chapel Hill, North Carolina, 1964, [Burmanniaceae, 330, 331; B biflora, B. capitata; by H. E. AHLEs. Rao, V. S. The stamens of Burmannia. Jour. Univ. Bombay Sci. 36: 25-27. 1967.* 304 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 . The orientation of the ventral bundles in the ovary of Burmannia. Sci. Cult. 34: 253. 1968.* SCHININI, A. Notas sobre Burmanniaceas Argentinas. (English summary.) Bol. Soc. Ar- gent. Bot. 16: 349-354. 1975. [B. flava recorded from Prov. Misiones and Prov. Corrientes, extending range to 28°20’ S lat.; B. capitata recorded from Prov. Cor- rientes, extending range to 28°15’ S lat.] STEYERMARK, J. A. Una nueva Burmannia de seccion Foliosa Jonker. Acta Bot. Venez. 6: 89-91. 1971 (publ. 1972). [B. Carrenoi Steyerm. described from Estado Bolivar, Venezuela; closest to B. Wercklei Schlechter, of Costa Rica. Stems leafy, to 30 cm long, flowers hanging, as in B. /ongifolia, the only Asiatic species assigned to this section by " onker. TAYLOR, R. J., & C. E. TAytor. Additions to the vascular flora of Oklahoma. Sida 7: 361-368. 1978. [B. eee 364; bog 5. 5 mi W of Antlers, Pushmahata Co. Tuomas, R. D. Burmann d Bowlesia incana (Umbelliferae) in Ouachita Parish, ia Sida 3: 183, 184, 1967. Warp, D. B. The genus Anonymos and its nomenclatural survivors. Rhodora 64: 87— 92. 1962. [B. capitata (J. F. Gmelin) Martius, 88; cf. WiLBurR.] , ed. Rare and endangered biota of Florida (P. C. H. PrircHarp, series ed.). Vol. 5. Plants. xxix + 175 pp. Gainesville, Florida. [1979.] [Fakahatchee Burmannia, Burmannia flava Martius, prepared by D. B. Warp, pp. 133, 134; includes account of collection by L. J. Brass, January 14, 1946. Only other collection known from Florida made by J. P. Standley, October 12, 1916.] Wicsur, R. L. The identity of Walter’s species of Anonymos. Jour. Elisha Mitchell Sci. Soc. 78: 125-132. 1962. [B. capitata “ F. Gmelin) Martius the correct name for Anonymos capitata Walter; cf. WARD Subtribe Apteriinae Jonker 2. Apteria Nuttall, Jour. Acad. Nat. Sci. Philadelphia 7: 64. pl. 9, fig. 2. 1834. Small, erect, presumably annual mycotrophic herbs lacking chlorophyll. Roots short and thin. Stem simple (sometimes branched above), violet to purple, with | to few (to ca. 6, rarely to 10) flowers. Leaves small, scalelike, sessile. Flowers about 1[—2] cm hs the perianth blue, violet, or purplish, rarely white [or blue and white], narrowl [or salver or funnel shaped]; perianth lobes 6, all the same eae the outer uate: the inner much narrower, the united part of the perianth more than 3 times the length of the lobes. Stamens adnate to perianth tube below the 3 inner segments, each subtended by a broadly V-shaped, almost saclike, vasculated ridge (cf. Ficure 1, d—f); filament short and thick, bearing on the abaxial side a large, 2-lobed wing and at the top expanded into a broad horizontal connective; locules borne horizontally and opening latrorsely, exposing the large pollen grains (in tetrads) in 2 masses. Ovary ovoid, abruptly tapering into the filiform style which is expanded into 3 short branches at the top, each branch terminated by a dish-shaped stigma at the level of 2 of the anther locules of 2 adjacent Staniens, ovary |-locular with 3 | gz tal pl acentae bear Ing numerous minute ovules. Ca psule crowned by the persistent perianth, opening ‘from the base by 3 upward-curling valves, the placentae persistent. Seeds minute, very numerous, + ellipsoid; embryo undifferentiated; endosperm consisting of 50-80 large cells. Type sPECIEs: A. setacea Nutt., nomen superfluum = 4. aphylla (Nutt.) Barnh. ex J. K. Small (Lobelia aphylla Nutt., 1822). (Name from Greek, a, without, and pferon, 1983] WOOD, BURMANNIACEAE 305 wing—wingless, in contrast with the three-winged flower of Burmannia bicolor L., which Nuttall knew as Tripterella caerulea Ell.) — NODDING NIXIE (see J. K. Small, Man. SE. Flora, 1933), a vernacular name of unknown origin, possibly coined, but not very appropriate since the flowers seldom nod. Three closely related species of the warmer parts of the Western Hemisphere: A. aphylla, which ranges from the southern United States to Brazil and Bolivia; A. lilacina Miers, known from Colombia, Brazil, and Argentina; and A. gen- tianoides Jonker, of Brazil, Paraguay, and Bolivia. The three differ in the shape and size of the perianth and in the structure of the stamens. Our single species, A. aphylla (A. setacea Nutt., A. boliviana Rusby, A. hymenanthera Miquel, A. Ulei Schlechter), is distributed from northeastern Florida and southwestern eorgia, westward through southern Alabama, Mississippi, and Louisiana, to southeastern Texas, and southward through Florida to Cuba, Hispaniola, Puer- to Rico, Jamaica, Dominica, Guadeloupe, and Trinidad, as well as from Mexico and Belize southward through Costa Rica and Panama to Colombia, the Gui- nas, and Brazil and Bolivia. Jonker reduced A. hymenanthera to A. aphylla var. hymenanthera (Miquel) Jonker but noted that “‘intermediate forms occur between the species and its variety especially in the West-Indian Islands.” Other workers may well decide on still further reduction. A confusion that began with Nuttall himself is shown in Jonker’s distribution map (1938, p. 37), which shows Apteria aphylla as occurring northward in the Mississippi Embayment past St. Louis, Missouri, and in the citation of four collections from the coast of Mississippi (p. 205) as being from Missouri. Nuttall (1834, p. 65) wrote, “Found in the vicinity of St. Louis, Missouri, by Mr. L. C. Beck; in Alabama by Doctor Gates; and discovered in East Florida by Mr. Ware, probably on the margins of shallow grassy ponds, the situations mostly affected by the Tripterellas.”” Whether or not Nuttall confused “St. Louis” with either Bay St. Louis or St. Louis Bay on the coast of Mississippi, or whether a label mistake was involved, Apteria occurs only in the southernmost parts of both Mississippi and Louisiana and is not known northward in either Ar- kansas or Missouri. Little seems to be recorded about the biology of any of the species of Apteria. All are presumed to be annuals and saprophytes via their mycorrhizal asso- ciation, although the habitats of at least A. aphylla (acid, swampy woods, at the edges of bogs and hammocks, and in pine flatwoods, often in the company of Burmannia) do not seem to be favorable for a saprophyte. It seems more likely that these are epiparasites (via their mycorrhizae) of the Galeola (Or- chidaceae) or Monotropa (Ericaceae) type. Uphof, who studied A. aphylla at Orlando, Florida, did not observe insect visitors to the flowers, and on the basis of the presence in the flowers of pollen-eating mites covered with pollen and crawling over the stigmas, the position of the stigmas (between the pollen masses of adjacent anthers), and the apparent adnation of anther, pollen, and stigma, concluded that the flowers are mostly self pollinated. The essentially smooth pollen is borne in tetrahedral tetrads ca. 47 um in diameter (Chakrapani & Raj) and is three-nucleate at the time it is shed (Brewbaker). The megaga- metophyte is of the Polygonum type. The embryo is only four-celled at the time the seed is shed. 306 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 Mwm. Ficure |. Apteria. a-l, A. aphylla: es plant with immature fruit, two flowers at anthes is, and flower bud, x 4; b, flower bud, x 3; c, flower and immature fruit with stamen subtended by a V-shaped, saclike ridge, X 6; e, stamen, adaxial side, with ad- joining parts of perianth, showing broad horizontal connective with masses of pollen tetrads at both sides and two-lobed, winglike appendage of filament behind, edges of wing inrolled, apex of wing lobes folded toward sides, x 12; f, same, oblique view, pollen removed to show two of four anther locules, < 12; g, stamen, abaxial side, showing two- lobed, winglike appendage of filament, anther connective and locules invisible behind it, perianth partially cut away to expose stamen, X 12; h, gynoecium, perianth cut away from top of ovary, X 6; i, stigmas and style branches from above, x 12; j, diagrammatic cross section of ovary to show placentation, x 12; k, capsule crowned by persistent rianth, three valves Sona away from placentae and curling upward, a few seeds persistent on placentae, < 6; 1, seed, < 50. 1983] WOOD, BURMANNIACEAE 307 The mature fruits are almost never collected and are little known. They should be looked for in localities where the plants have been found in flower. The capsule opens by three upward-curling valves, and the three placentae are persistent as shown in Ficure |, k. REFERENCES: Under family references see AyILVSGI, BAILLON, BENTHAM & Hooker, BRADE, BREW- BAKER, CHAKRAPANI & RAJ, CIFERRI, CORRELL & CORRELL, Davis, ENGLER, ERDTMAN Goprrey & Wooten, HATSCHBACH & GuIMARAEs (pl. 3), JoHow (1885), JONEs, JONKER (1938, especially pp. 35-37, 203-211; 1945, pp. 46, 47), ee Maas, MELCHIOR, Miers (pp. 545-548), Nerouitzky, RICKETT, SEUBERT, and W Hooker, W. J. Apteria orobanchoides. Ic. Pl. 3: pl. 254. 1840. Apteria setacea. Ibid. 7: pl. 660. ane Mo .Fino, J. F. a Burmaniacea interesante para Misiones: Apteria lilacina Miers. Physis 6: ae 330. pl. 1923. [Northeasternmost province of Argentina.] NuttA.t, T. A description of some of the rarer or little known plants indigenous to the United States, from the dried specimens in the herbarium of the Academy of Natural Sciences in Philadelphia. Jour. Acad. Nat. Sci. Philadelphia 7: 61-115. pls. 9-12. 1834. [Apteria, 61-66, pl. 9, fig. 2.] Upnor, J.C. T. Apteria aphylla. Beitrage zur Kenntnis der Burmanniacee Apteria aphylla (Nutt.) Barnhart. Osterr. Bot. Zeitschr. 78: 71-80. 1929. [Habitat, morphology, pollination, anatomy of stem, leaf, root, and seed.] ARNOLD ARBORETUM NUE CAMBRIDGE, MASSACHUSETTS 02138 STONE, FREYCINETIA AND PANDANUS 309 STUDIES IN MALESIAN PANDANACEAE, 19. NEW SPECIES OF FREYCINETIA AND PANDANUS FROM MALESIA AND SOUTHEAST ASIA BENJAMIN C. STONE DURING A STUDY VIsiT of several months in the Harvard University Herbaria made possible through the courtesy of the Director of the Arnold Arboretum, several new taxa of Freycinetia and Pandanus were recognized. Their descrip- tions, presented here, form a part of the background necessary for the treatment of the Pandanaceae in Flora Malesiana, which is in preparation. I am grateful to the curators of the various herbaria listed herein for making these materials available for study. Freycinetia Gaudich. Sect. BLUMEELLA Stone Freycinetia impudens B. C. Stone, sp. nov. Liana modice robusta caulis foliosis usque ad 15 mm diametro. Folia coriacea usque ad 60-90 x 2.5—4 cm, lineari-lanceolata, sensim attenuato-acuminata, marginibus in basi breve tractu spinuloso-denticulatis, et apicem versus mani- feste denticulatis (denticulis vix 1 mm longis, acicularibus, 2-5 mm sese se- paratis); costa media infra breviter carinato, basi inerme, apicem versus spi- nuloso-denticulata; pagina adaxiale minute obscureque striatula, abaxiale crebre venoso-striatula et subevidenter tessellata, reticulationibus ca. x 0.4 mm, minutissime per stomata punctulatis,; auriculis 7-12 x 8-11 cm, subrigide membranaceis, in sicco pallide brunneis, apice incurvato- subtruncatis, deinde abrupte angustato-foliaceis, usque ad 15 cm longis, basi ca. 8 cm lato. Cephalia cylindracea, pedunculis usque ad 5 cm longis et 6 mm diametro, se (vel distaliter sparse minuteque hispidulis); cephalia immatura ca. 6-8 X mm, carpidiis plerumque 2-4-natis, ca. 5 X 1 mm, pileo probabiliter 2.5 mm longo, vertice truncato, stigmatibus 2—4(-6) confluentibus, annulo angusto cincto. Spadix masculus ca. 12 cm longus, pedunculo glabro 5-7 cm longo et ca. 5 mm diametro, parte florifero ca. 7 cm longo et usque ad 9 mm diametro, staminibus congestis ca. 2-2.5 mm longis, filamentis 1.5-2 mm longis, antheris breve-oblongis ca. 0.8-1 mm longis obtusis. Cephalium bisexuale in specimine singulo (Docters van Leeuwen 9986) visum: carpidia et stamina intermixta, © President and Fellows of Harvard College, Journal of the Arnold Arboretum 64: 309-324. yee 1983. 310 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 filamentis staminorum 3.5—4 mm longis, antheris ut videtur normalibus, stam- inibus carpidium aequantibus vel leviter superantibus. Type. New Guinea, Papua, Fly River 30 mi above D’Albertis junction, Oro- ville Camp, in rainforest, Aug. 1936, L. J. Brass 7426 (holotype, A!). ADDITIONAL SPECIMENS EXAMINED. New Guinea. IRIAN BARAT: Rouffaer R., 175 m alt., forested hills, W. Mf. Docters van Leeuwen 9986 (Bo!, L!); between Hollandia and Hol- landia-Binnen, 80 m alt., ae forest, BW 3612, C. Kalkman (L!) Brass 7426 is one of the two specimens tentatively identified by Merrill and Perry as Freycinetia pseudo- insignis Warb. They remarked (J. Arnold Arb. 20: 158. 1939): “Somewhat hesitantly we add here the following collection: Brass 7426.... This differs from the other collections in its obviously narrower leaves. The anthers of the ¢ inflorescence are about 3 mm. long [sic].”’ Martelli and Merrill and Perry have all mingled two species under the name F. pseudo- insignis, at least one of them (that now named above as F. impudens) and possibly the other cannot be equated with Warburg’s species. Warburg’s protolog calls for staminate spadices 8 cm long, the floriferous part 4 cm by 7 mm, and filaments 1.5 mm long. In F. impudens the spadices are about this size but the stamens are longer. Another very similar species is F. inouei Kanehira, also described from Irian Barat. Merrill and Perry thought this species to be the same as their F. archboldiana, and the two do belong together in Sect. BLUMEELLA. However, since the inflorescence in F. inouei appears to be regularly ternate (rather than multicephalic, as is the norm in F. archbol- diana), it seems better to keep them distinct. On the other hand, it is possible that F. inouei, which was typified by a pistillate specimen, is the same taxon as F. pseudo-insignis, which was typified by a staminate one. In any case, F. inouel differs from F. impudens by its somewhat narrower and shorter leaves with less acuminate apices, smaller leaf prickles, and shorter, narrower spadices. The berries of F. inouei are mostly bistigmatic, while those of F. impudens are more frequently 3- to 5-stigmatic. It is interesting to note the occurrence of bisexual cephalia in this species, a phenomenon rarely seen but known to date in Freycinetia imbricata Blume (Biinnemeijer 5031, from Sumatra), F. negrosensis Merr. (Stone 12911, from the Philippines), F. scabripes Warb. (PNH 117425, Gutierrez et al., from the Philippines), and F. reineckei Warb. (P. A. Cox, several collections from Sa- moa). Only in the case of F. reineckei, however, is it known that both the male and the female organs in the cephalium are functional. Cox (pers. comm.) suggests that such bisexuality is an adaptive response to promote fertilization through bat pollination, since the bats are more attracted to polliniferous than to pistillate cephalia. Sect. DEVRIESEELLA Stone Freycinetia micrura B. C. Stone, sp. nov. FIGURE I. Liana habitu valde similis Freycinetia sphaerocephalo Gaudich. sed pedun- culo foemineo densiter hispidulo, foliis parce angustioribus differt. Folia parva 7-10 cm X 4-7 mm, anguste elliptico-suboblanceolata, apice basinque angus- tata, apice sensim attenuato-caudato in caudo ca. 15-20 mm longo desinentio: 1983] STONE, FREYCINETIA AND PANDANUS 311 Balgocy 2276 ei Sele oe Lae — —=- ase wr a : : JeP SIT Cc | a oe 2mm Figure |. Freycinetia micrura (from Van Balgooy 3216, type collection): a, habit of fruiting branchlet; b, leaf, abaxial surface; c, portion of leaf margin, showing veins and marginal prickle; d, staminate spike; e, stamen, enlarged; f, berry, top view, showing 3 stigmas; g, berry, side view, seed beside it; h, seed, enlarged; i, auricle of leaf base, showing spinular margin. vagina appresso, auriculis 12-14 x 2-3 mm, apice curvatis, marginibus spi- nuloso-pectinatis, spinulis angustis subrectis 0.8-1.8(-2) mm longis; lamina foliorum marginibus basi apiceque spinulosis (parte medio tertio inerme) vel, in foliis parvioribus, e basi ad apicem spinulosis; spinulis antrorsis vulgo | mm longis, apicem versus brevioribus (ca. 0.5 mm longis); costa media in parte dimidia distali spinulosis, spinulis ad eos in margine adjacentio similis, vulgo 1-2 mm sese separatis. Inflorescentia terminalis, ternata (vel quaternata), brac- teis late ovatis, usque ad 30 X 15 mm, plus minusve caudatis, marginibus minute serrulato-spinulosis; pedunculo foemineo usque ad 20 mm longo, 2- 312 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 2.5 mm crasso, subcompresso, densiter hispidulo e basi ad apicem spinulis subappressis congestis 0.5 mm longis; cephalio ay eae subgloboso (ubi immaturo ca. 17 X 10 mm); gynoecio breve 4-5 X 1. mm, apice ca. | mm alto, late pyramidato-rotundato, areola centrali bi- al quinque- estiematticno ca. 1-1.4 mm lato instructo; seminibus immaturis ca. | mm longis. Cephalia mascula ca. 8—9 mm longae, pedunculo masculo laevi, glabro, ca. 15 mm longo; staminibus vix numerosis, filamentis 1.75 mm longis, antheris subgloboso- oblongis ca. 0.5 mm longis. Type. Sulawesi, Mt. Roroka Timbu, western slope at 2000 m alt., in montane forest dominated by Agathis 40 m tall; climber with green fruit, bracts white, purple tipped; May 1979, M. J. van Balgooy 3216 (holotype, L!). A very distinct little plant evidently related to Freycinetia sphaerocephala Gaudich. of the Philippines, but differing markedly in the very densely his- pidulous peduncles of the pistillate cephalia, and in the narrower leaves; the spinules of the leaf and auricle sate tags be up to 2 mm long. The fruit described is unfortunately not yet matu I am grateful to Dr. van Balgooy for ‘his kindness in allowing me to study the materials he collected, in advance of their distribution. Pandanus L. ex Stickman Sect. ACROSTIGMA Kurz Pandanus ketele B. C. Stone, sp. nov. FIGURE 2. Frutex gracilis, stipite erecto usque ad 3 m alto, 2-5 cm diametro, laevi, radicibus aereis basalibus, parvis, brevibus. Folia ca. 160 < 3.7-4.1 cm elon- gato-linearia, apice acutata, supra nitidam viridem, infra subglaucam, nervis longitudinalibus ca. 100 per foliam congestis; marginibus in basi (parte vagi- nanti ca. 7 cm longo excepto) spinuloso-denticulatis, denticulis patento-an- trorsis Curvatis usque ad 2 mm longis, 2-6 mm sese separatis, in medio, den- ticulis appressiter antrorsis, ca. 0.5 mm longis, 2-4 mm sese separatis, apicem versus, denticulis ca. 0.2 mm longis, ca. | mm sese separatis; costa media infra in basi denticulis unguiculiformibus usque ad 3 mm longis et 5-9 mm sese separatis, in medio denticulis parvis vel costa inerme carinata, deinde denticulis parvis vix 0.4 mm longis, 5-15 mm sese separatis, apicem versus denticulis parvissimis vix 0.2 mm longis, ca. 1 mm sese separatis; plicis binis apicali- ventralibus in parte distale ca. 30 cm longo denticulatis, denticulis antrorsis parvis subirregulariter dispersis. Inflorescentia lateralis, gracilis, suberecta, ca. 64 cm longa, 14 mm diametro, parte aeees fertili ca. 12 cm longo, capitulis congestis vulgo 7 aggregatis, parvis, ovato-subglobosis, subplano-convexis, 3.5— 4.5 cm longis, ie rane aurantiaco- rubris, e pluribus drupis com- positis. Drupa ca. 2 cm longa (stylo incluso), pileo ca. 8 mm longo, 3 mm crasso, ees n stylo spiniformi antrorse curvato 5— 6 mm longo producto; parte seminifero ca. 8 mm longo. Type. Papua New Guinea, Southern Highlands Province, Tari valley, Itipu village, ca. 1640 m alt., in swampy areas with secondary growth, Sept. 1981, B.C. Stone & C. Rose 15005 (holotype, KLU!). Figure 2. Pandanus ketele: leaves and infructescences of Stone & Rose 15005 (type aieeien) photographed i in the field. Ruler is 30 cm lon This species clearly belongs with Pandanus adinobotrys Merr. & Perry, P. setistylus Warb., P. angiensis Kanehira, and P. pseudosyncarpus Kanehira, all f New Guinea (including Irian Barat), but differs from all of them in its much narrower leaves and its very long pistillate peduncle bearing small, crowded cephalia. Like these species it is characterized by lateral inflorescences—1.e., axillary fertile shoots that bear scale leaves (and occasionally some reduced foliage leaves) and floral bracts—that are clearly differentiated in function from the strictly foliage-bearing leaf crowns. All these species occur in the interior and mainly at intermediate altitudes, usually between 600 and 1800 m. Pan- danus ketele occurs in small swamps and is sometimes cultivated as a hedge plant. The ripe drupes are dull vermilion, as is true for most (perhaps all) of the above-mentioned related species. The collectors thank the children of Itipu village for help in collecting the material. Pandanus lustrorum B. C. Stone, sp. nov. FIGURE 3. Frutex erectus, stipite usque ad 4 m alto, ramis fertilibus lateralibus brevissi- mis, foliaceis. Folia anguste lineares acutissima usque ad 107 x 2 cm, in ramis fertilibus breviora, rigide coriacea, basi in vaginam brevem expansa, vagini Ca. 2.5 cm longo striatulo-venoso; laminis obscure longitudinaliter venosis; mar- ginibus in basi dentibus patenti-antrorsis ca. 2 mm longis, 1.5-8 mm sese separatis, in medio dentibus valde appressis vel antrorsis 0.6—1 mm longis, 4- 8 mm sese separatis, apicem versus denticulis 0.4-0.7 mm longis, 0.5-0.9 mm sese separatis; costa media infra in basi dentibus reflexis usque ad 4 mm longis, 3-9 mm sese separatis, in medio denticulis antrorsis vix | mm longis, 3-10 mm sese separatis, apicem versus denticulis brevioribus denticulis marginorum similibus; plicis binis apicalibus denticulatis. Inflorescentia terminalis, pedun- culo 30-35 cm longo, 5-6 mm diametro (in sicco), bracteato, distaliter infra cephalia modice sinuato; bracteis non vidi, delapsis. Cephalium ovoideum, 5.2 314 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 FLORA if NEW Gl NEA Botanical Collections of the Division of Botany, Department of Porests, Lae NGF G871a J, « Wome ' * March, 197 Locality: Tele! A, F Dist, Wet Sepik, Alt, 16 Lat Long. | Habitat: Wet svamp. pandans which inc lude and Freycinettia yaa Lic Luss i two species Annot.: Erect pandan, height with many short side branches. Leaves grey, Fruit pale orange brown, solitary. Fam,: Pandanaceae Name: Pandanus Dupl. sent to: L. Bri. Canb. A. -K. «. Sing. Syd. Bish. PNH. US. ALE Lg cm FiGure 3. Pandanus lustrorum (NGF 4871/8): isotype in LAE with monocephalic infructescence. 1983] STONE, FREYCINETIA AND PANDANUS bi be) x 4.8 cm, ca. 600 ad 800 drupis compositum, aurantiaco-brunneum, drupis angustis subfusiformibus, 13-19 mm longis (stylo incluso), parte basilari 6- 8(-9) mm longo, pileo acutissime conico-pyramidato 5-7 mm longo apice in stylo antrorse deflecto spiniformi 4 mm longo producto. Mesocarpium superum medulloso-fibrosum, inferum fibrosum brevissimum. Endocarpium ca. 7 mm longum in parte dimidio basali situatum anguste obconicum, pariete ca. 0.15 mm crasso. Cetera ignota. Tyre. Papua New Guinea, West Sepik Province, Telefomin, 1600 m alt., erect pandan in wet swamp, March 1975, NGF 48718, Womersley & Waikabu (holotype, KLU!; isotypes, BRI, L, LAE). ADDITIONAL SPECIMENS EXAMINED. New Guinea. PapuA New GuINEA. West Sepik Prov- ince: Telefomin, 1600 m alt., wet swamp, NGF 48716, Womersley & Waikabu (ku, LAE), NGF 20829, Henty (LAe!). This species is very much like Pandanus adinobotrys, and previously the above-mentioned specimens were assigned to that species. However, the spec- imens differ in their narrower leaves, often solitary fruit-head, and shorter styles, and they therefore appear to represent a somewhat cryptic species. This species, together with P. ketele (described above), forms an addition to the cluster of New Guinea species centered around P. setistylus. The limits of variation within members of this group are still not very well understood, and further field study of them is desirable. Sect. MARKGRAVIDENDRON Huynh & Stone Pandanus sulawesicus B. C. Stone, sp. nov. FIGURE 4. Arbor, stipite in basi radicibus aereis rigidis emittens. Folia lineari-loriformia, apice abrupte acuminato breve-caudata, basi sensim angustata, profunde cana- liculato; marginibus in basi spinulosis, spinulis curvato- patentibus curvatis acutis, ca. 4 mm longis, rariter duplicatis, 2-5 mm sese separatis, in medio cum spinulis 2.5 mm longis, 7-9 mm sese separatis, apicem versus spinulis ca. | mm longis, 0.6—3 mm (vel usque ad 12 mm) sese separatis; costa media in basi spinulis reflexis vel recurvatis ca. 4 mm longis, 2-5 mm sese separatis, in medio inermis, profunde et acute carinata, vel sparse et irregulariter pauci- spinulosa, spinulis antrorsis, apicem versus, spinulis antrorsis ca. 1 mm longis, confertioribus, ca. mm sese separatis, in cauda densiter congestis, vix | mm sese separatis; ceeae longitudinalibus ca. 0.75-0.9 mm sese separatis, infra leniter prominulis, ca. 86 per foliam (costa media excluso), apicem versus cum venulis transversis ca. 2-3 mm remotis connectis. Inflorescentia foeminea pedunculata; cephalium solitarium, oblongo-cylindraceum, ca. 19 X 9cm, plu- rimis drupis compositum. Drupa irregulariter cuneata, apice truncato, (33-) 6-40 mm longa, pileo indistincto integri, quinque- vel sexangulato, ca. 15 mm lato, vertice plano vel subdepresso, stylo perbreve ad latum areolis verticis posito, sursum spectantio; stigma ca. 3 X 3 mm, oblique vel verticaliter situato, aperturo centrali, margine minute denticulato, denticulis superioribus 3 vel 5, inferioribus 5 vel 8. Mesocarpium apicale ca. 8 mm longum cavernosum; basale 316 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 b FiGure 4. Pandanus sulawesicus (from Van Balgooy 3107, type collection). a—c, leaf: commussures. d-f, drupe: d, side view; e, top view; f, longitudinal section. g, stigma, enlarged, frontal view. 1983] STONE, FREYCINETIA AND PANDANUS S17 fibrosum ca. 12-15 mm longum. Endocarpium centrale ca. 20 mm longum uniloculatum apice rotundato basi truncato, pariete ca. 3 mm crasso parte medio toruloso 4-5 mm crasso excepto. Cetera ignota. Tyre. Indonesia, Sulawesi, ca. lat. 1°S, long. 120°E, Sopu Valley, 85 km SSE of Palu, 1000 malt., rainforest on flat, very wet alluvial soil; stilt-rooted pandan, fruit orange; May 1979, Van Balgooy 3107 (holotype, L!). This is the second member of sect. MARKGRAVIDENDRON, formerly a mono- typic section consisting only of Pandanus navicularis Stone (endemic to New Ireland in the Bismarck Archipelago). In its leaves, Pandanus sulawesicus is reminiscent of P. cheilostigma Stone and P. brevistipes Martelli, both of sect. CHEILOSTIGMA Stone, but the fruit characters are quite different. Sect. CHEILOSTIGMA is endemic to the Philippines. Sect. RYKIA (DeVriese) Kurz Pandanus albifrons B. C. Stone, sp. nov. FiGuReEs 5-7. Frutex simplex vel sparse ramosus usque ad 2-3.5 m altus, stipite angustl radicibus aereis paucibus brevibusque emittenti. Folia perlata, arcuata, usque ad 90-100 X 12-14 cm, acuminato-brevicaudata, basi conspicue angustata in- fra albo-glaucescentia; vaginis auriculatis, auriculis usque ad 10-11 x | cm; marginibus in basi unguiculatis dentibus antrorsis patentibus curvatis ad 3-4 mm longis, 5-20 mm sese separatis, in parte medio dentibus remotissimis |- 2 mm longis, (20-)30-50(-60) mm sese separatis, depressiter antrorsis, in apice denticulis vix 1 mm longis, 1-5 mm sese separatis; costa infra in parte brevi basilari retrorse hamato-spinosa spinis 3-4 mm longis, 5-15 mm sese separatis, costa deinde pro maxima parte mediani apicem versus et caudem excepta exarmata obscure carinata, in apice denticulatis, denticulis antrorsis vix | mm longis, 2-4 mm sese separatis; pagina laminae utrinque apicem versus minute tessellata, nervis longitudinalibus ca. 100 ad 105 per foliam, venulis transversis ca. 0.6-0.7 mm sese separatis. Inflorescentia terminalis, pedunculo foemineo ca. 15 cm longo, 7 cm diametro, bracteatis, bracteis albidis exterioribus usque ad 23 X 6-7 cm, interioribus ovato-navicularibus breviter caudatis ca. 8 x 4 cm. Cephalium solitarium ellipsoideum e 90 ad 100 drupis compositum, ca. 12 X 7.7 cm, flavidum. Drupa obovoideo-cuneiforma ca. 25 x 14 x | 1-12 mm, pileo depresse-pyramidato hexagonato, stylo brunneo, 4-5 mm longo, central1, antrorse deflecto, 1- vel 2-cornulato, cornulis ca. 1-2 mm longis stigmate nigro. Pericarpium subcarnosum flavidum. Mesocarpium superum medulloso-fibro- sum ca. 8-10 mm longum; inferum fibroso-carnosum ca. 5-7 mm longum. Endocarpium inverso-cupuliforme apice in prolongatione centrali erecti, basi leniter angustato, infra centro positum, ca. 8 mm longum (corno centrali ex- cepta), atro-brunneum, pariete ca. | mm crasso. Cetera ignota. Type. Malaysia, Kepong Forest Research Institute, plants introduced from Pasoh Forest Reserve (Negri Sembilan); fruiting; November 1978, Stone & Ng 14003 (holotype, KLU!; isotypes, BISH, K, KEP, L, SING, US). This handsome species belongs to the group including Pandanus crinifolius 318 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 hi % Ficures 5-7. Pandanus albifrons: 5, type tree; 6, bracts and 2 cephalia from type tree; 7, nee tree with exposed cephalium (note new shoot with white leaf bases in upper center). 1983] STONE, FREYCINETIA AND PANDANUS 319 Martelli (Malaya), P. /euconotus Stone (Borneo), and P. nervosus Stone (Su- matra). It is most similar to P. /euconotus, both have remarkable white leaf undersurfaces that are particularly conspicuous at the base of the leaf. The new species differs from P. crinifolius in its whitish leaf-base undersurfaces, white or off-white floral bracts, less crispate blades, and longer styles; from P. /eu- conotus in its more abruptly rounded-acuminate leaf tips, white or off-white spathes, and broader laminae; from P. nervosus in its wider leaves, more el- lipsoid cephalia, and whiter spathes; and from all these species in its very remote marginal prickles along the median part of the lea Although clearly related to such other species as Pandanus unguifer Hooker f., P. tonkinensis Stone, P. dictyotus St. John ex Stone, and P. huynhii Stone, P. albifrons is distinguished by several characters, especially the very broad, loriform leaves with the undersurfaces white near the base and broad auricular basal margins, and the white (vs. pink or yellow) floral bracts. The staminate plants of P. albifrons are still unknown, but they should strongly resemble those of P. leuconotus, undoubtedly the most closely related species. Pandanus dictyotus St. John ex Stone, sp. nov. Frutex humilis, habitu Pandanus crinifolio vel parviore. Folia lineari-lori- formia, apice acuminata ieee: caudata, basi leniter angustata; vaginis purpurascentibus; laminis ca. 8 cm, striatulo-nervosis, apicem versus evidenter tessellatis, cauda ae filiforma ca. 5 cm longa terminatis; mar- ginibus in basi cum denticulis patentibus leniter arcuatis 2-3 mm longis, vulgo 10-25 mm sese separatis, in medio subinermis, apicem versus denticulis an- trorse curvatis ca. 1 mm longis, 1-3 mm sese separatis; costa media dorso in basi paucidenticulatis, denticulis retrorsis, 2-3 mm longis, deinde erectis, sub- inermis, in medio inermis, apicem versus denticulis cis in margine adjacentio similibus. Inflorescentia terminalis, brevis, pedunculo foemineo ca. 7.5 cm ongo, 5 mm crasso, apice incrassato, bracteato, bracteis superioribus conges- tibus ovato-ellipticis brevicaudatis, marginibus minute serrulatis; cephalio sub- ellipsoideo ca. 7 X 4.5 cm, drupis simplicibus composito; drupa ca. 18-20 mm longa, pileo hemispherico pyramidato, stylo centrali antrorse curvato, subulato, ca. 4 mm longo, 1- vel 2-cornato, cornulis brevibus vix | mm longis. Type. Sarawak, Bako National Park, March 1961, S 13017, Rashid b. Taggoi (holotype, SAR!). A small shrub of lowland forest. For taxonomic comments, see under Pan- danus albifrons. Pandanus leuconotus B. C. Stone, sp. nov. FIGURE 8. Frutex; stipite erecto usque ad 2.5 m alto e basi radicibus brevibus paucibus emittenti. Folia late loriformia arcuata, basi modice attenuata ad apicem acu- minato-caudata usque ad 120 x 9.5 cm, tessellato-reticulata (in sicco), infra in parte basilari conspicue glauco-alba, marginibus toto denticulatis, denticulis maximis ad 7 mm longis, antrorsis; costa infra in basi denticulis reflexis armata. a20 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Inflorescentia terminalis, cephalio druparum breve oblongo 10-12.5 x 6-7 cm, obscure trigonato, drupis 25-30 x 9-11 mm, pileo truncato-pyramidato, stylo castaneo uni- vel rariter bicornato 4-5 mm longo antrorse valde deflecto, a spathis fertilibus 7 usque ad 35 cm longis, spicibus oblongis albis 5-7 x 1.5 cm, phalangibus staminorum umbellatis columnis 3—5 mm longis, filamentis ca. | mm longis, antheris 1-1.2 mm longis apiculatis, apiculo 0.9-1.1 mm longo. Type. Borneo, Sabah, Ranau District, Poring, base of Mt. Kinabalu, 330 m alt., in forest near Hot Springs, April 1975, B. C. Stone et al. 12906 (holotype, KLU; Isotypes, BISH, K, L, SAN). ADDITIONAL SPECIMENS EXAMINED. Borneo. SABAH: Mt. Kinabalu, Dallas, ca. 900 m alt., M.S. Clemens 27751 (uc!), J. & M. S. Clemens 26857 (a!, Bo!); Mt. Kinabalu, 330 m ‘-alt., Stone 12905 (kLUu). This species 1s similar in overall appearance and many particulars to Pan- danus albifrons (described above), of the Malay Peninsula, and shares the white coloration of the leaf-base undersurface with that species, which is manifestly its nearest relative. It 1s also related to P. crinifolius and, a little more distantly, to P. tonkinensis Stone. For further details, see the comments after P. albifrons. Plants of Pandanus leuconotus have been brought by Kinabalu National Park workers from nearby places and planted in the Poring Hot Springs area, where they thrive, flowering and fruiting regularly. Pandanus nanofrutex B. C. Stone, sp. nov. Frutex nanus. Corona foliorum compacta. Folia linearia angusta usque ad 95 x 2.8 cm, apice acuminato vel breve-caudato, basi leniter angustata; mar- ginibus denticulatis, in basi denticulis 2-3 mm longis, 5-12 mm sese separatis, in medio denticulis ca. | mm longis, appressis, antrorsis, ca. 5-14 mm sese separatis, apicem versus denticulis 0.5-0.75 mm longis, vulgo 1-2 mm sese separatis; costa media basem versus denticulis retrorsis 2-3 mm longis, vulgo 10-20(-35) mm sese separatis, in medio (parte tertio folii) inermis, carinatis, apicem versus denticulis antrorsis | mm longis, 5 mm sese separatis, et in cauda vix 2 mm sese separatis; venis longitudinalibus angustis, prominulen- tibus, 0.5-0.6 mm sese separatis, ca. 50 per foliam; venulis transversalibus prominulentibus, reticulationes coe fete formantibus, areolis vix | mm longis. Inflorescentia terminalis, erecta, cephalio solitario oblato- -globoso ca. 60 X 50-55 mm, pedunculo ad 14 cm longo angusto, apicem versus in- ca. 20 X 20 mm. Drupae ca. 85 ad 95 per cephalio, 18-22 mm longae stylo excluso (ad 27 mm longo stylo incluso), 10-12 mm latae; pileo subhemispheri- co-pyramidale, 3-5 mm alto; stylo centrali, erecto vel antrorse deflecto, haud vel minime furcato, subtereti vel compresso, 5-6 mm longo, deltoideo-acuto, furcis ca. 1 mm longis; stigmate ca. 4 mm longo, ovato-elliptico. Mesocarpium superum medullosum, inferum fibrosum subcarnosum. Endocarpium central- 1983] STONE, FREYCINETIA AND PANDANUS BA | Bcs-1972 CLEMENS 27751 FiGure 8. Pandanus leuconotus (from Clemens 27751), drupe: a, side view; b, top view; c, longitudinal section. ium ca. 8 mm longum, apice conico-elongato, atrobrunneum, osseum, pariete ca. | mm crasso, loculo singulo. Type. Tonkin [Vietnam, Ha Son Binh Province], Mt. Ba Vi, vallée du Lankok, dans les forets, Janvier 1889, Balansa 4165-bis (holotype, P!). This species stands out among the members of sect. RyK1A because of its dwarf, shrubby stature, its narrow leaves with large marginal prickles, and its small fruit-heads. Martelli had determined the type specimen previously as Pandanus stenophyllus Kurz (= P. nitidus (Miq.) Kurz), a Javanese species of sect. ASTERODONTIA Stone and not very closely related. Instead, P. nanofrutex belongs to a small group of species within sect. RyKIA that includes P. unguifer, P. bicornis Ridley, P. huyhnii, and some others that occur in southeastern Asia and western Malesia. It is quite similar to P. bifidus St. John, of Thailand, but differs in its smaller cephalia and drupes, its shorter style-forks, and its con- siderably shorter leaves. Pandanus nervosus B. C. Stone, sp. nov. Frutex. Folia usque ad 100 x 11 cm, versus apicem perabrupte rotundata, apice acuminato-caudata, caudis ca. 9 cm longis,; marginibus valde armatis, dentibus usque ad 3-4 mm longis, in caudo dense congestibus multo minoribus ca. 1 mm longis; pagina utrinque valde reticulato-tessellata; costa basi dentibus retrorsis usque ad 5 mm longis, 1-2 cm sese separatis, dispositis. Inflorescentia terminalis, foemineo solitario, cephalio subgloboso, ca. 7 X 5 cm, e ca. 80 ad 90 drupis composito, pedunculato; drupis uniloculatis, 15-18 x 6-9 mm vel usque ad 22 x 13 mm, 5- vel 6-angulatis, apice subtruncato-subdepresse pyra- Bee JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 midato, stylo 1- vel 2-cornuto, ca. 3-5 X 4-6 mm, cornis acicularibus 1-2 mm longis divergentibus vel valde divaricatis; endocarpio cupulato apice in acu- mine anguste conico breviter prol 15-16 mm longo terminato. Cetera ignota. E. Sumatra, east coast, vale of Tangga, Asahan, May 1927, H. H. Bartlett pe (holotype, us!). ADDITIONAL SPECIMENS EXAMINED. Indonesia. SUMATRA: Asahan, Goerach Batoe, swampy places in forest, Yates 1653 (uc!); west coast, Bukit Baloe Banting, 1100 m alt., Jacobson 195 (po!); Asahan, vic. of Taloen Djoring, Silo Maradja, Dec. 1927, Rahmat si Toroes 39 (A!, MICH). This is a very characteristic species known so far only from Sumatra. The Jacobson collection included living plants (or seeds) that were subse- quently grown in the Botanical Gardens, Bogor (now known as Kebun Raya, Indonesia). Specimens were later collected (in September 1932) from the cul- tivated plants. There is a whole fruit-head preserved in alcohol in the carpologi- cal collection there. Pandanus regalis B. C. Stone, sp. nov. Ficures 9, 10. A Pandanus furcato affini sed cephalio perelongato trigono-subcylindrico usque ad 100 * 25 cm, plurimis drupis composito; drupis ca. 38-40 x 10 mm pileis pyramidato-rotundatis; stylo aceroso vel furcato; mesocarpio supra fi- broso-medulloso, infra fibroso-subcarnoso; endocarpio osseo atrobrunneo-fer- rugineo apice acute pyramidato base truncato fere totam corpam drupae oc- cupanti, ca. 27 mm longo, pariete ca. 2 mm crasso. Semen unicum. Type. Sumatra, Simaloer Island, Feb. 1919, Achmad 909 (Bo!). No other material has been seen of what is certainly an imposing species. Although the leaf specimens in particular are only fragmentary, they give the impression of being from a very large plant. The middle portion of an adult leafis 10.5 cm wide, with marginal prickles ca. 4 mm long and about 5-6 mm apart; another part of the leaf closer to the apex is nearly 8 cm wide, narrowing distally to 6.5 cm, with similar but closer clydoniform prickles. The fruit-head is longer than that of almost any other Pandanus species known to me, and the drupes are stocky and almost parallel sided. Certainly this is a very distinctive plant. Nevertheless, it 1s clearly among the species closely related to Pandanus furcatus Roxb. Pandanus tonkinensis B. C. Stone, sp. nov. Frutex humilis, stipite usque ad 15 mm diametro; corona foliorum compacta. denticulis retrorsis 4-5 mm longis, 10-15 (deinde ad 30) mm sese separatis, Ficures 9, 10. Pandanus regalis (from Achmad 909, type collection): 9, entire in- fructescence; 10, detail of same, showing drupe apices and forked styles. in parte tertio medio inermi, apicem versus denticulis antrorsis ad eos in margine adjacentio similibus, in caudo extremo remotioribus; nervis longitu- dinalibus manifestis, cum venulis transversalibus reticulationibus tessellatis formantibus; lamina vulgo 48- ad 50-nervatis, nervis ca. 0.8-0.9 mm sese separatis. Inflorescentia terminalis, pedunculo ad 14 cm longis, apice 9 mm crasso, triquetro, bracteato; cephalium solitarium globosum maturitate ca. 55 X O mm, e ca. 70 drupis compositum; receptaculo ca. 15 X 10 mm; drupis obovoideis, ca. 25-27 X 10 mm, drupis basalibus exceptis (eis ca. 17 X 10-11 mm); pileus 3-5 mm altus, depresse pyramidatus, stylo centrale erecto vel antrorse deflecto rubro-brunneo nitido 4-5 mm longo subtereti vel compressi 1- vel 2-cornato instructo; furcibus stylorum vix | mm longis, acutis; me- socarpium superum medullosum, inferum fibroso-subcarnosum; endocarpium ca. 10 mm longum, apice in rostro subconico-centrale producto, |-loculatum pariete ca. | mm crasso, osseo, atroferrugineo. Type. [Vietnam], Dac Lac [Darlac] Province, SE of Poste du Lac, 1200 m alt., April 1921, Poilane 32462 (holotype, P!). ADDITIONAL SPECIMENS EXAMINED. Vietnam. HA Son BINH PRoviNcE: Mt. Ba Vi, 1000 m alt., Balansa 4165 (p!); Broyi, Hayata 5211 (p!). NHA TRANG [PHU KHANH] PROVINCE: Nui-han-heo, Poilane 4888 (P!). QUANG TRI [BINH TRI THIEN] PRoviNcE: Col d’Ailao et Dent du Tigre, 1500 m alt., Poilane 10403 (P!). This taxon was segregated first by Martelli and was described, but only in 324 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 French, in H. Lecomte (Fl. Gén. Indo-Chine 6: 1063. 1937). No Latin diagnosis was ever published. Because Pandanus nanofrutex (described above) was among the specimens ascribed to P. tonkinensis by Martelli, it seems preferable to publish the latter as a new species here. Pandanus tonkinensis is closely related to certain species of sect. RYKIA with small or even dwarf stature, small fruit-heads, leaves often noticeably con- tracted toward the base and usually abruptly acuminate-caudate at the apex, and drupes with usually forked styles. This group includes such species as P. unguifer, P. bicornis, P. huyhnii, and P. nanofrutex. These, and perhaps some others from Thailand, form a complex in which species distinctions are often obscure, perhaps mostly because of the very few and often inadequate collec- tions made. For most, the staminate plants remain unknown. Basic to any improvement in our knowledge of this group, therefore, is a continued effort to obtain good collections supplemented by field observations and photographs. DEPARTMENT OF BOTANY UNIVERSITY OF MALAYA KuALA Lumpur, MALAYSIA 1983] WOFFORD, LINDERA a25 A NEW LINDERA (LAURACEAE) FROM NORTH AMERICA! B. EUGENE WOFFORD Tue Genus Lindera Thunb., composed of upwards of 100 species, is primarily of eastern Asia (Wood, 1958), with only three species in North America. Lin- dera benzoin (L.) Blume (including L. benzoin var. pubescens (Palmer & Stey- erm.) Rehder), the most common and widespread North American taxon, ranges generally throughout eastern North America from Maine, southern On- tario, and southern Michigan southward, and westward into Kansas, Okla- homa, and Texas. It occurs in mesic habitats along shaded streambanks and in damp woods and bottomland swamps. Lindera melissifolia (Walter) Blume, a rare and local species, inhabits mesic to hydric sites (i.e., bottomland hard- woods, depressions, and margins of sandy sinks and ponds). Extant populations are presently known from the Coastal Plain of North Carolina, South Carolina, Georgia, Mississippi, and along the Arkansas-Missouri state line (Map 1). Populations from Alabama (Wilcox Co., Buckley s.n., GH!, Ny!, Us!), western Florida (without specific locality, Chapman s.n., Ny!), and Louisiana (without specific locality, Hale s.n., GH!) have not been seen in over a century and are probably now extirpated from these states. The nomenclatural history, ecology, distribution, and morphological features for separating L. melissifolia from L. benzoin were elucidated by Steyermark (1949). During herbarium studies in preparation for a forthcoming treatment of Lauraceae for the Vascular Flora of the Southeastern United States (J. R. Massey, ed.), I encountered great difficulty in placing a number of collections of Lindera within the circumscription of either L. benzoin or L. melissifolia. Most of these collections were made from the Mississippi Gulf Coastal Plain by Drs. Sidney McDaniel and Ken E. Rogers. Subsequent field and herbarium studies have led me to conclude that these specimens are representative of a new species, described here. Lindera subcoriacea B. E. Wofford, sp. nov. FiGuRE 1. Frutex parvus, ad 2 m alta, dioecious. Folia subcoriacea, vix aromatica, elliptica vel oblanceolata, 4-7.5 cm longa, 2—3.5 cm lata, apice obtusa vel rotunda. Drupa vivida rubra, ca. 10 mm longa, elliptica. Dioecious, deciduous shrub to 2 m. Stems smooth, light gray and pubescent when young, becoming darker gray and glabrous with age. Leaves with petiole canaliculate, 3-10 mm long, pubescent; blade subcoriaceous, horizontal to ascending, elliptic to oblanceolate, 4-7.5 by 2-3.5 cm (lower stem leaves re- ‘Contributions from the Botanical Laboratory, The University of Tennessee, N.S. No. 543. © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 325-331. April, 1983. 326 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 e L. benzoin % L. melissifolia @ L. subcoriacea Pa Map |. Distribution of Lindera in the southeastern United States. duced), the apex obtuse to rounded, rarely slightly acuminate (mucronulate when young), the base cuneate, the margin entire (ciliate when young), the venation eucamptodromous, the lower surface pale green, glaucescent, mod- erately pubescent, the upper surface darker green, pubescent when young, be- coming essentially glabrous with age. Flowers appearing before leaves in axils of preceding year’s leaves on stout, supraaxillary branches |.5 mm long and terminated by vegetative bud; inflorescences of | to 4 umbellike cymose clus- ters, each cluster 3- or 4- (rarely 5-)flowered, subtended by 2 pairs of concave, decussate, coriaceous, deciduous bracts, with outer pair 2.5 by 3 mm, inner pair 3 by 3.5 mm, the secondary bracts caducous, tepaloid, reduced. Flower buds globose, ca. 2.5 mm wide; flowers imperfect, regular; tepals 6, 2.2 mm by 1.8 mm, glabrous, pellucid-punctate, perianth tube short. Staminate flowers on pedicels to 4 mm long; stamens 9, introrse, 2-locular, yellow, those of series 1983] WOFFORD, LINDERA a2? I and II similar, 2.5 mm long, with anther | by | mm, those of series III 2.5 mm long, filament broadened, each with pair of conspicuous glands at base, those of series IV lacking; pistillodium 1.2 mm long. Pistillate flowers on pedicels to 1.5 mm long; tepals slightly smaller than those of staminate flowers; stamens variously developed, often reduced to glands resembling those at base of series III stamens of staminate flowers; stigma papillose, style 1 mm long, the ovary elliptic, 1 by 0.6 mm. Drupes bright red, ca. 10 mm long, elliptic, borne on pedicels to 4 mm long. Type. Mississippi, George Co., hillside bog | mi NW of Escatawpa River and 100 yd SW of Hwy. 98, 27 July 1982, B. E. Wofford 82-121, with D. L. Scott & M. G. Lelong (holotype, A; isotypes, IBE, NY, RSA, TENN, VDB). DISTRIBUTION AND PHENOLOGY. Open areas of permanently wet, peaty, ever- green shrub bogs on Gulf Coastal Plain of Mississippi and Louisiana (Map 1). Flowering in mid-March; fruits maturing in late fall. SPECIMENS EXAMINED. Louisiana. WASHINGTON ParisH: swampy woods beside La. 10 and Lawrence Creek, 2 mi E of a eee sect. 29, T. 2S, R. 11 E, Thomas et al. 32367 (NLU). Mississippi. FORREST Co.: 100 yd W of Hwy. 49,8 or Hattiesburg ca. 7 mi from ject. of hwys. 98 and 49, S Bae, of artificial pond, Rogers 4932 (miss). Gerorce Co.: 5.5 mi S of Agricola, common shrub in quaking bog, McDaniel 4165 (Fsu, 1BE); 5.5 mi S of Agricola, McDaniel 10345 (Fsu, 1BE, Miss. Mus. Nat. Sci., smu), 10348 (FSU, IBE); quaking bog 5.5 mi S of Agricola, W side of Hwy. 613, sect. 32, T. 3 S,R. 5 on open hillside along S side of Hwy. 98, ca. 0.5 mi W of Alabama-Mississippi state line, NW'4 sect. 20, T. 2S, R. 4 W, Lelong 5199 (Ncu); at Mississippi-Alabama line off Hwy. 98 ca. 200-300 yd S of hwy., Rogers 2902 (miss, MO, NCU); bog in longleaf pine woods S of U. S. 98 just W of Alabama state line, Thomas et al. 43060 (Lsu, NLU); Shipman bog, Wilmer Quad., sects. 18 (SW), 19 (NW'4), 13 (SE'4), T. 1S, R. 4 W, Rogers 45917 (TENN); 0.8 mi NW of Escatawpa R. off Hwy. 98 to Lucedale, Lelong s.n., 3 April 1981 (TENN); ca. 0.8 mi W of Nespas ag Alabama state line on Hwy. 98, SE's sect. 13, T. 2S, R. 5 W; SW sect. 18, T. 2S, R. 4 W, Gordon et al. 2552 (Miss. Mus. Nat. Sci.); hillside bog ca. 0.8 mi W of Mississippi- Alabama line S of Hwy. 98, Gordon & Burris 2749 (Miss. Mus. Nat. Sci.); Wilmer Quad., | mi NW of Escatawpa R. then 100 yd SW of Hwy. 98, Wofford et al. 81-47 (A, TENN); upland bog ca. | mi W of Escatawpa R. and 100 yd SW of Hwy. 98, Wofford et al. 82-1 (TENN). GREENE Co.: acid bog 2 mi SE of Leakesville, Hwy. 63, Jones 8484 (su, miss); bluff SE of Chickasawhay me at Leakesville, Jones 11225 (Fsu, Miss). JACKSON Co.: evergreen shrub bog, ca. 1.5 W of LaRue and ca. 100 yd W of dirt road, sect. 23, T. 5 S, R. 9 W, Wofford 82-2 Geigy: STONE Co.: Toc-O-Leen Lake, sect. 20, T. 2S, R. 13 W, Rogers 8769 (mo, NCU, NY, SMU, USF) Salient features for recognizing Lindera subcoriacea include its subcoria- ceous, elliptic to oblanceolate leaves (obtuse to rounded at the tip) that are less than 7.5 cm longand only faintly aromatic. Although several primarily northern deciduous woody taxa tend to produce more coriaceous or semievergreen leaves near the southern limit of their ranges, the thickened leaves of L. subcoriacea appear to be a genetic divergence; L. benzoin leaves remain membranaceous at roughly the same latitude in Florida, southwestern Mississippi, and Loui- siana. Other characters that can be used to separate the three North American taxa are summarized in TABLE |. JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 PLANTS OF MI FiGure 1. Holotype of Lindera subcoriacea Wofford. Fragrance ruit Length (mm) Pedicel Horizontal to as- cending Faint, piny lem- on 10 Not thickened at apex; decidu- ous Horizontal to as- cending Strong, “spicy” 10 Not thickened at apex; decidu- ous 1983] WOFFORD, LINDERA 329 Tas_e 1. Character comparisons of the North American taxa of Lindera. TAXON CHARACTER L. subcoriacea L. benzoin L. melissifolia Leaf Texture Subcoriaceous Membranaceous Membranaceous Length X width (cm) 4-7.5 X 2-3.5 6-15 X 3-6 8-16 X 3-6 Shape Elliptic to oblan- Obovate Ovate to elliptic ceolate Apex Obtuse to round- Acuminate Acute ed Base Cuneate Cuneate Widely cuneate to rounded Pubescence (abaxial side) Present Present or absent Present Orientation Drooping Strong, sassafras- like 12 Thickened at apex; persis- tent Lindera subcoriacea is restricted to open areas of permanently wet, peaty, evergreen shrub bogs on the Gulf Coastal Plain of Mississippi and Louisiana. These highly organic, acidic microhabitats are fire dependent and offer abun- dant sunlight, moisture, and few competitors. In small (up to ca. 5 m in di- ameter) areas of deep peat accumulation, they simulate the ‘“‘quaking”’ nature usually associated with the more familiar peat bog formations of cooler regions of the Northern Hemisphere. Dominant woody plant associates include Clif- tonia monophylla (Lam.) Britton ex Sarg., Magnolia virginiana L., Persea pa- lustris (Raf.) Sarg., Myrica heterophylla Raf., M. cerifera L., Ilex glabra (L.) A. Gray, Hypericum brachyphyllum (Spach) Steudel, and Smilax laurifolia L. Herbaceous associates are more variable in frequency and reflect the seral stage of each individual bog. Some of the more common of these include Sarracenia alata (Wood) Wood, Oxypolis filiformis (Walter) Britton, Lophiola americana (Pursh) Wood, Sabatia macrophylla Hooker, Dichromena latifolia Baldwin, Zigadenus glaberrimus Michaux, Rhexia alifanus Walter, R. petiolata Walter, Polygala cruciata L., P. cvmosa Walter, Fuirena squarrosa Michaux, Cacalia lanceolata Nutt., Muhlenbergia expansa (DC.) Trin., Panicum spretum Schultes, P. scabriusculum Ell., P. nudicaule Vasey, Stokesia laevis (Hill) Greene, and Eupatorium recurvans Small. Conversely, both Lindera benzoin and L. melissifolia occur on mostly shad- ed, mesic or slightly hydric sites but never in areas that are saturated throughout the growing season. In the southeastern United States L. benzoin is associated with taxa characteristic of shaded alluvial woods and eastern deciduous forests 330 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 (e.g., Acer saccharum Marsh., Liriodendron tulipifera L., Fagus grandifolia Ehrh., Aesculus glabra Willd., A. octandra Marsh., Sanguinaria canadensis L., Erythronium americanum Ker-Gawl., Claytonia virginica L., Polystichum ac- rostichoides (Michaux) Schott, Dentaria spp., and Trillium spp.). Similarly, L. melissifolia occurs on mesic or slightly hydric sites that are occasionally in- undated for several months during the growing season. Along the Mississippi Embayment in Mississippi and Arkansas, I have observed this taxon in low depressions of bottomland forests dominated by species of U/mus, Liquid- ambar, Celtis, Fraxinus, and Quercus. Such forests are somewhat transitional in species composition between swamp forests and eastern deciduous forests. These local depressions, however, are often filled with dense clonal stands made up almost exclusively of L. melissifolia with essentially no characteristic her- baceous associates. On the Atlantic Coastal Plain L. melissifolia occurs around the margins of limestone sinks and pineland depressions. Competition with herbaceous associates, with the exception of Woodwardia virginica (L.) Sm. and occasional Xyris and Eupatorium spp. (R. Porcher, pers. comm.), is also minimal here. Distributional and phenological data provide additional evidence for the recognition of Lindera subcoriacea as a distinct species. On the Gulf Coastal Plain it is disjunct from the nearest known population of L. benzoin by ca. 75 miles, and from the nearest population of L. melissifolia by ca. 150 miles (see Map 1). Field studies in the spring of 1982 showed that L. subcoriacea reached peak anthesis on March 18-20, while the Mississippi population of L. melis- sifolia 150 miles to the north had flowered about seven to ten days earlier. More importantly, by March 21 nearby populations of L. benzoin in West Feliciana Parish, Louisiana, and Wilkinson Co., Mississippi (both at roughly the same latitude as L. subcoriacea), had half-expanded leaves, young fruits, and only a few scattered remnants of withered staminate inflorescences. Al- though more detailed phenological studies are needed, it is safe to conclude that existing geographic and phenological barriers preclude opportunities for gene exchange between L. subcoriacea and either L. benzoin or L. melissifolia. Fragrances from crushed leaves and stems provide additional characters for species delimitation. Lindera benzoin emits a distinct and familiar ‘‘spic fragrance. Lindera melissifolia exudes an odor similar to that of sassafras (Stey- ermark, 1949; Godfrey & Wooten, 1981). Both of these fragrances are easily detected in fresh material throughout the growing season. Lindera subcoriacea, on the other hand, emits a scarcely detectable piny lemon fragrance when stems and leaves are young but becomes essentially odorless during late summer and fall. Analyses of these compounds are currently underway. Additional prelim- inary data from flavonoid studies show distinct interspecific chromatographic profiles from leaf extracts, with similar profiles from flower extracts. These data will be presented later. In summary, Lindera subcoriacea can be separated from both L. benzoin and L. melissifolia on the basis of morphological, ecological, distributional, and phenological data. In morphological features (TABLE 1) it appears to be more closely related to (and perhaps shares a common ancestor with) L. ben- zoin. In ecological characters, however, its affinities differ: it shares a preference 1983] WOFFORD, LINDERA 33! for more hydric sites with L. melissifolia, yet it requires open, sunny sites unlike those of either L. benzoin or L. melissifolia. It appears equally distinct from both taxa in distribution, phenology, and chemistry. Therefore, additional field studies along with chemosystematic and cytological data are needed before more definitive phylogenetic relationships can be theorized. Its closest rela- tionships may prove to be among the numerous species of eastern Asia. Finally, Lindera subcoriacea appears to be quite rare and in need of protection and habitat management. Additional plants have not been located at sites of several of the earlier collections, and some of the known populations are pres- ently threatened by construction projects or occur on unprotected private land. Bog habitats, in general, are rapidly disappearing subsequent to man’s control of naturally occurring seasonal fires combined with the construction of drainage ditches, farm ponds, and highways. Folkerts (1982) recently estimated that approximately 97 percent of the former Gulf Coast bog communities have been destroyed or seriously altered since pre-Columbian times. ACKNOWLEDGMENTS Field work was supported by a grant from the Johnson Fund of the American Philosophical Society. Thomas S. Patrick provided the photograph of the type and carefully reviewed the manuscript. Curators of the following herbaria pro- vided specimens or distribution data: FSU, GA, GH, IBE, LSU, MARY, MISS, MISSA, MO, NCU, NLU, NY, SMU, US, Mississippi Museum of Natural History, and the Reed Herbarium. LITERATURE CITED FoLKerts, G. W. 1982. The Gulf Coast pitcher plant bogs. Amer. Sci. 70: 260-267. Goprrey, R. K., & J. W. Wooten. 1981. Aquatic and wetland ae of southeastern United States: dicotyledons. Univ. Georgia Press, Athens, Georg STEYERMARK, J. A. 1949. Lindera Peni as Rhodora 51: 16D. Woop, C.E., Jr. 1958. The a of woody Ranales in the southeastern United States. J. Arnold Arbor. 39: 296-346 DEPARTMENT OF BOTANY UNIVERSITY OF TENNESSEE KNOXVILLE, TENNESSEE 37996 JOURNAL OF THE ARNOLD ARBORETUM INSTRUCTIONS FOR AUTHORS General policy The Journal of the Arnold Arboretum is primarily a staff journal, and staff papers have priority. Other papers are accepted, as space permits, from former staff or former students, and from other botanists who have worked on our collections or who have done research on a plant group or in a geographic area of interest to the Arboretum. Submission of manuscripts Manuscripts should be submitted in duplicate to Ms E. B. Schmidt, Managing Editor, Journal of the Arnold Arboretum, 22 Divinity Avenue, Cambridge, Massachusetts 02138. A copy of the manuscript should be retained so that when reviews and/or editorial suggestions are received, any necessary correc- tions can be made and the appropriate portions of the paper resubmitt For ease of editing, an outline of the paper (not to be published) ee the basic structure of the manuscript should be included. Preparation of manuscripts Papers should be triple spaced throughout (including title, text, citation of specimens, footnotes, acknowledgments, bibliography, and figure legends), on bond (not erasable) paper, with wide margins on all four sides. Nothing should be underlined except generic and infrageneric scientific names, italics when present in a quotation, and the collector and collection number of all specimens cited. FORM AND STYLE. This can be determined from a recent issue of the Journal. The title should be as short as possible; it should usually contain the name of the family concerned but not authorities of scientific names. Abstracts are enerally not used. Abbreviations should be employed only when two or more letters will be saved, and with the exception of units of measure, compass directions, and herbarium designations, they should always be followed by a period. Metric measurements should be used when possible, but information on specimen labels should not be changed. Authority names should be given for all generic and infrageneric taxa the first time they are mentioned in the text unless they are included in the formal taxonomic treatment. Acknowledgments should be placed at the end of the paper before the bib- liography; the author’s current address should follow the bibliography. Foot- notes should be kept to a minimum and should be numbered consecutively throughout the paper, with the exception of those appearing in tables. Here either standard symbols or lower-case letters should be used In case of question, the latest edition of Words into Type, by Marjorie E. Skillin et a/. (Prentice-Hall), should be consulted. CITATION OF SPECIMENS. Currently accepted geographic should be used, with spelling according to a standard source. Names of countries should be in English and should be typed in regular capital and lower-case letters. Below the country level, names may be in the language of the country involved. If this option is taken, careful attention should be paid to consistency, spelling, and accent marks. Arrangement of areas should be consistent (geographic or alphabetic, preferably the former) within a paper. information should not be changed unless it is obviously wrong or lacking no information; in this case, additions or corrections should be bracketed. When a collector has a common surname, his initials should be given. With specimens from the Indo-Malesian area, care should be taken to de- termine whether collections are institutional or not. In the case of an institu- tional collection, the institutional series and number (e.g., LAE 20257) should be given. This, in addition to the location, is all the information needed for a brief, unambiguous citation. If it is desirable to include the collector, this information should be placed after the institutional series and number (e.g., LAE 20257, Foreman or LAE 20257 (Foreman)). The list of exsiccatae should be arranged by institutional numbers, where applicable, for ease of use. ILLUSTRATIONS AND LEGENDS. Reference must be made in the text to all maps, figures, and plates. Insofar as possible, their sequence should be determined by the order in which they are mentioned. Plates (illustrations grouped together at the end of an article) and figures (illustrations scattered through the text of an eect should be prepared with Journal page proportions in mind. The ximum size after reduction is 4.25 by 6.5 inches (10.8 by 16.5 cm) for aon and 4.25 by 6.75 inches (10.8 by 17.1 cm) for plates. A figure may occupy any portion of the length of a page: plates should be more or less full- page size. To facilitate mailing and handling, mounted illustrations must be of a manageable size. Line drawings showing habit and plant parts should be lettered in the same order (e.g., habit, leaf, inflorescence, whole flower, calyx, corolla) on each il- lustration throughout a paper. Photographs should be trimmed, grouped appropriately, and mounted with no space between them on stiff white cardboard with a margin of at least | inch left on all four sides. The author’s name and the figure number(s) should be noted on the back of each illustration. To prevent bending or other damage, the art work should be wrapped carefully and shipped flat. A clear copy of each illustration should be included for review purposes Legends should be written in telegraphic style (see back issues for examples). They should be grouped in numerical sequence on a separate page, rather than placed below each figure or plate. Hlustrations of each type (i.e., figures, plates, or maps) should be numbered consecutively and separately, figures and maps with Arabic numerals and plates with Roman numerals. For example, a paper could include figures 1-3, maps 1-3, and plates I-VII. If the illustrations are to be figures, each figure should be numbered separately, with the numbers running in order through the text. Subdivisions of the figures should be indi- cated with letters. Thus, figures 1-5 may appear on one page, figures 6-8 on the following page, and figure 9, A-C, on the next. In the case of plates, each page of photographs is a separate plate; the individual photographs comprising the plates are numbered or lettered consecutively. [lustrations with dark back- grounds should have white letters or numbers and vice versa. Whenever pos- sible, scales should be included in the illustrations; any magnifications necessary in the legends should be calculated to include reduction of illustrations to our page size. Illustrations are not returned to the author after publication unless this is requested. TABLES. Titles for tables should be short, with all explanations placed in footnotes. Tables should be as simple as possible and must be neatly typed. Long and/or complicated tables can be photographed directly if they are in good order and the copy is clear enough (in this case the copy should not be triple spaced, but should have spacing appropriate to contents and headings); this eliminates the chance for error and the need for proofreading ables should be numbered consecutively using Arabic Ramenls. Each table should be cited in the text. BIBLIOGRAPHY. The Guide to Citation of Botanical Literature in the Inter- national Code (all editions through 1972) should be followed. When possible, reference should be made to past issues of the Journal for form. Titles should be abbreviated according to Journal (Schwarten & Rickett, Bull. Torrey Bot. Club 85: 277-300. 1958) or Botanico-Periodicum-Huntianum style; the two styles should not be mixed within a paper. Runovers should be indented. Except in cases where confusion would result, only authors’ initials are used in place of first and middle names. All typing should be done in regular capital and lower-case letters, and nothing should be underlined except generic and infra- generic scientific names. Titles of articles and books should not be capitalized except for the first word, scientific names, and proper nouns and adjectives. AUTHOR’S ALTERATIONS. Charges for extensive alterations in proofs will be billed to the author. Page charges Authors are requested to help defray printing costs. Although actual printing costs are much higher, $20.00 is the customary charge, and authors expected to make every effort to pay. Under special circumstances the fee may be reduced or waived altogether, if this is agreed upon in advance. Ability or inability to pay will in no way affect acceptance or handling of a manuscript. LEON CAMILLE MARIUS CROIZAT(-CHALEY) Turin, Italy—July 16, 1894 Coro, Venezuela— November 30, 1982 We record with regret the death of Dr. Léon Croizat, member of the staff of the Arnold Arboretum from 1938 to 1947 and student of the Euphorbiaceae. Journal of the Arnold Arboretum April, 1983 CONTENTS OF VOLUME 64, NUMBER 2 Systematics of the Andropogon virginicus Complex (Gramineae). CHRISTOPHER 8, CAMPBELL .... 0... 0... cece eee ee eg eeueeen 171 The Plates of Aublet’s Histoire des Plantes de la Guiane Francoise. RICHARD Aa TIOWARD 6 5 o-ad404 4494064560445 wei eeeeesoREs Zoo The Genera of Burmanniaceae in the Southeastern United States. CARROL be WOO: Ihe. fic Ach egg a Dod Woda a's ouewekdualekaneway 293 Studies in Malesian Pandanaceae, 19. New Species of Freycinetia and Pandanus from Malesia and Southeast Asia. BENJAMIN C. STONE 3406.2 050604 0540006 dd ews oa obo bibwd eed he 309 A New Lindera (Lauraceae) from North America. B. EUGENE WOFFORD ......... 0.0 cece ccccuccucueeuceccucuas 325 Volume 64, Number I, including pages 1-169, was issued January | 1, 1983. JOURNAL OF THE ARNOLD ARBORETUM HARVARD UNIVERSITY VOLUME 64 NUMBER 3 US ISSN 0004-2625 Journal of the Arnold Arboretum Published quarterly in January, April, July, and October by the Arnold Arboretum, Harvard University. Subscription price $30.00 per year. Sul 1 I 1 be sent to Ms E. B. Schmidt, Arnold Arboretum, 22 Divinity Avenue, Cambridge, Massachusetts 02138, U. S. A. Claims will not be accepted after six months from the date of issue. POSTMASTER: send address changes to Ms E. B. Schmidt, Arnold Arboretum, 22 Divinity Avenue, Cambridge, Massachusetts 02138, U.S. A. Volumes I-51, reprinted, and some back numbers of volumes 52-56 are available from the Kraus Reprint Corporation, Route 100, Millwood, New York 10546, Lash EDITORIAL COMMITTEE S. A. Spongberg, Editor E. B. Schmidt, Managing Editor P. S. Ashton K. S. Bawa P. F. Stevens C. E. Wood, Jr. Printed at Allen Press, Inc., Lawrence, Kansas COVER: The stylized design appearing on the Journal and the offprints was drawn by Karen Stoutsenberger Second-class postage paid at Boston, Massachusetts, and additional offices. JOURNAL OF THE ARNOLD ARBORETUM VOLUME 64 Juty 1983 NUMBER 3 A REVISION OF THE GENUS ALSOPHILA (CYATHEACEAE) IN THE AMERICAS David S. CONANT Tue Genus Alsophila R. Br. is here redefined based on the recent biosyste- matic studies of Conant (1975, 1976) and Conant and Cooper-Driver (1980). A major contribution to the understanding of the classification of the Cyathea- ceae has been Tryon’s (1970) separation of the scaly tree ferns into three natural alliances: Sphaeropteris Bernh. (with conform petiole scales; see Tryon, 1970, for definition of scale types); A/sophila sensu Tryon and Nephelea Tryon (with setate, marginate scales); and Trichipteris Presl, Cyathea Sm. sensu Tryon, and Cnemidaria Pres| (with nonsetate, marginate scales). Based on this system, recent revisionary studies of the Cyatheaceae by Tryon (1971a, 1976), Gastony (1973), Stolze (1974), Windisch (1977, 1978), and Barrington (1978) have provided excellent data against which the present systematic conclusions have been compare Holttum (1961) divided the scaly tree ferns into two genera, Cyathea sensu lato and Cnemidaria. Cnemidaria sensu Holttum is equivalent to Cnemidaria sensu Tryon, whereas Cyathea sensu Holttum includes the rest of Tryon’s scaly genera. Cyathea sensu lato was divided into subg. SPHAEROPTERIS (with setif- erous (conform) petiole scales) and subg. ALSOPHILA (with flabelloid (marginate) scales). Subgenus ALSOPHILA included sects. ALSOPHILA (indusiate, axes green- ish, fertile and sterile pinnules monomorphic) and GyMNOSPHAERA (exindu- siate, axes dark, chocolate brown, fertile and sterile pinnules dimorphic) (Holt- tum, 1974). The present genus A/sophila includes Tryon’s Alsophila and Nephelea in the New World and Holttum’s Cyathea subg. ALSOPHILA in the Old World. (All Old World species with marginate petiole scales also have setae. Species with nonsetate, marginate scales are restricted to the New World.) The extensive hybridization documented by Conant (1975, 1976) and Conant and Cooper- Driver (1980) is evidence that A/sophila sensu Tryon and Nephelea are more © President and Fellows of Harvard College, Journal of the Arnold pees 64: 333-382. Ge 1983. 334 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 closely related than they are distinct and separable genera. Furthermore, Conant and Cooper-Driver (1980) have presented evidence that new species have arisen from these “‘intergeneric” hybrids. These species of hybrid origin can be as- signed to a genus only on an arbitrary basis—1.e., whether genes for squaminate spines are inherited and expressed or not. Gastony (1973, 1974) has demon- strated that most species of Al/sophila sensu Tryon and Nephelea have 16- spored sporangia, a feature that distinguishes them from other scaly Cyathea- ceae. Gastony (1973) and Gastony and Tryon (1976) have shown that species of A/sophila sensu Tryon cannot be distinguished from Nephelea by their spores, lending additional support to their inclusion in the single genus A/sophila. Species of A/sophila (in the present sense), however, are not known to hy- bridize with species of the nonsetate, marginate alliance (Trichipteris, Cyathea sensu stricto, Cnemidaria). The setate and nonsetate groups with marginate scales appear to be morphologically distinct and reproductively isolated from each other. Therefore, I cannot accept the conclusion of Holttum (1974) that these groups are elements of the single genus Cyathea. This redefinition of A/sophila obviates the need for a reclassification of the family. Such a task is best left until there have been further biosystematic studies of the species with nonsetate, marginate scales and until the relation- ships of the species with marginate scales to those with conform scales are more fully understood. DISTRIBUTION AND ECOLOGY Alsophila is a genus of about 235 species distributed throughout the wet tropics of the Old and New Worlds. In the Old World, there are about 205 species divided between Africa and Madagascar (about 50 species—Schelpe, 1970; Holttum, 1981; Tardieu-Blot, 1951, 1953), the Indian subcontinent east to China and Japan (about 15 species—Holttum, 1965), Malaysia (about 114 species— Holttum, 1963), and Australasia and the Pacific (about 26 species— Holttum, 1964). There are 30 species in the New World, of which 16 are centered in the Greater Antilles, 14 occur in Mexico and Central and South America, and | extends northward into the Lesser Antilles from Venezuela (see Map 1). The center of diversity is in the Greater Antilles, where 16 species are endemic; Hispaniola, with 8 species and 5 endemics, is the island with the highest diversity. The American species of 4/sophila are mostly plants of the cloud forest- elfin forest type of habitat. Exceptions to this are 4. jimeneziana, A. cuspidata, and A. sternbergii, which can often be found at lower elevations in gallery forest. A/sophila is primarily a genus consisting of understory, shade-loving species. Heliophilic plants like those found in the genera Sphaeropteris, Tri- chipteris, and Cyathea sensu Tryon are not found in American Alsophila and apparently do not occur in eastern Asian 4/sophila either (Holttum, 1963). MORPHOLOGY Stem. Stem habit and size provide good diagnostic characters for some species. In Alsophila nockii and A. rupestris the stem is short, prostrate, and less than 1983] CONANT, ALSOPHILA 335 as 100 90 80 70 60 50 40 30 20 \ \ \ Map 1. Distribution of the genus 4/sophil/a in the Americas. 2 cm in diameter; in 4. brooksii and A. urbanii it is short, prostrate (sometimes decumbent in A. urbanii), and to 6.5 cm in diameter. Among species with erect stems, most are greater than 6 cm in diameter and 2 m in length. Exceptions are A. paucifolia, which has an unusually slender (less than 2 cm in diameter) stem of unknown length, and 4. amintae and A. abbottii, which generally have stems less than 5 cm in diameter and 1.5 m in length. Squaminate spines may or may not be present on the stem. In general, spines are restricted to the species with tall, stout stems, although reduced spines are present in A/sophila brooksii and in certain hybrids between spiny and spineless species. JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 wy Yt oa Wd Alsophila br is oe 2265), grad- , A. woodwardioides (Conant —— dually tapering apex, 3, A. ane (Haught 2001), aarp as apex, x. 2263), des tapering base, x .18: 5, 18. Ficures 1-5. — ceDIEeS and bases: 1, ually tapering apex, x .09; vi bryophila (Conant urbanit (Congni i abruptly reduced base Adventitious buds occur in six of the nine species with squaminate spines in Mexico and Central and South America, as well as in Alsophila woodwardi- oides, of Jamaica, Hispaniola, and Cuba. Alsophila sternbergii is reported to produce stolons at the base of the stem that may facilitate vegetative repro- duction in this species (Brade, 1971; as Cyathea sampaioana). Stolons are not known elsewhere among American species of the genus but are well documented (Hallé, 1966) in the African A. manniana The stem apices of all species are protected by deltoid to lanceolate scales with either a single apical seta or both apical and marginal seta Lear. Although the petiole and the lamina are usually described separately, many of the characters of the petiole are present on the rachis in a reduced form, reminding us that the division of the leaf into petiole and lamina is arbitrary. This arbitrariness is also apparent in the treatment of aphlebiae and 1983] CONANT, ALSOPHILA 307 disjunct basal pinnae. Although these may more properly be parts of the lamina, they are treated here as belonging to the petiole The terminology for the parts of the fern leaf fe been somewhat standardized (Tryon, 1960), but it is still confusing to compare fern species having complex (bipinnate-pinnatifid) leaves and those having simpler (pinnate-pinnatifid) ones. A factor contributing to the confusion is that some terms connote homologies whereas others relate more to size: petioles, rachises, pinna-rachises, pinnule- rachises, pinnae, and pinnules of bipinnate-pinnatifid and pinnate-pinnatifid leaves are homologous structures; costae, costules, and penultimate and ulti- mate segments are nonhomologous. LAMINA. Lamina dissection varies from pinnate-pinnatifid to tripinnate (oc- casionally quadripinnate). Ten of the twelve species with pinnate- -pinnatifid to bipinnate laminae occur in the Greater Antilles. A/sophila rupestris (Colombia) and A. paucifolia (Ecuador) also have the pinnate-pinnatifid architecture. Bi- pinnate-pinnatifid lamina dissection occurs in sixteen species distributed throughout the range of the genus in the Americas. Two species, A. salvinil (Mexico, Guatemala, Honduras) and 4. polystichoides (Costa Rica and Pana- ma), have tripinnate to quadripinnate laminae. The lamina apex is one of two types. Most species have an acuminate apex in which the more distal pinnae are gradually reduced (Ficures 1, 2). Nine species from Mexico, Central and South America, and the Lesser Antilles have a distinctly pinnalike imparapinnate lamina apex (FIGURE 3). The lamina may be gradually narrowed at the base as in A/sophila amintae and mature leaves of A. bryophila (Figure 4) or abruptly reduced as in A. urbanii (FIGURE 5). Some species (e.g., 4. portoricensis) are variable with respect to the lamina base. Pinna shape is often a useful character. A measure of pinna shape is width divided by length: long, narrow pinnae will have low width/length quotients; broader ones will have higher quotients. PeTIOLE. Petiole length is generally correlated with lamina base. Species with gradually narrowed lamina bases are usually short petiolate, whereas those with abruptly reduced lamina bases tend to be long petiolate. Squaminate spines may or may not be present on the petiole. The presence of petiole spines is precisely correlated with the presence of stem spines. Re- duced petiole spines may be ee in Alsophila brooksii and on certain hybrids between spiny and nonspiny speci Tardieu-Blot (1941) found that = aphlebioid! species of Madagascar were of three basic types. The first, including A/sophila decrescens (Kuhn) Tryon (as Cyathea decrescens), has gradually tapering lamina bases in which pinnae grade into basal subaphlebioid pinnae. American species belonging to this group include 4. amintae, A. abbottii, A. minor, A. auneae, A. grevilleana, and A. setosa. (In A. minor the basal subaphlebioid pinnae may be disjunct.) The WW bat lahj Acanhiebiaid e technically i t, as suggested by Holttum (1981), they are ‘ieintanie d here because these terms have historically been consistently used in describing skeletonized basal pinn 338 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ") Uy ( ge Ficures 6-11. Lamina scales. 6, 7, A/sophila nockii: 6, Hart 13, bullate scale, x 100; 7, Underwood 1355, bicolorous, lanceolate scale with single apical seta, X 65. 8, A. brooksii (Conant 677), lanceolate scale with marginal and apical setae, X 68. 9, A. salvinii (Stone & Broome 2814), acaroid scale, x 105. 10, 11, A. engelii (Pennell 9695): 10, setate, flabellate scale, x 28; 11, highly dissected scale, « 50. second type, including 4. dregei (Kunze) Tryon (as C. dregei), has nonaphle- bioid basal pinnae that are separated from the rest of the pinnae by a large gap. Although 4. bryophila has disjunct basal pinnae of this type in some juvenile plants, its pinnae are pinnatifid and those of 4. dregei are pinnate- pinnatifid. The third type, including 4. boivinii Ettingsh., has highly skeleton- 1983] CONANT, ALSOPHILA 339 ized aphlebiae that are separated from the rest of the pinnae by a gap. Three species, A. capensis, A. salvinii, and A. urbanii, have aphlebiae of this type. INDUMENT. The abaxial surface of the axes and veins of the lamina are invested with various kinds of scales, squamules, trichomes, and trichomidia that are of considerable taxonomic importance. The indument on the adaxial surface of the axes and veins is more or less similar in all species and is therefore usually not described. Bullate scales (FiGURE 6) are nearly orbicular to elongate, inflated, and with or without setae. If they are nearly flat, they are termed “subbullate.” Bullate scales are present on the costae or costules of 14 of 16 species in the Greater Antilles (all except Alsophila brooksii and A. urbanii) and of 4 (A. salvinil, A, tryoniana, A. engelii, and A. capensis) of 14 continental species. Bicolorous (dark bodied, light margined), lanceolate scales with a single apical seta (FIGURE 7) (or sometimes with a few setae at the base) are found in four related species of the Greater Antilles: Alsophila amintae, A. abbottii, A. nockii, and A. tussacii. Alsophila imrayana, of South America and the Lesser Antilles, has often concolorous, lanceolate scales with a single apical seta. A/lsophila paucifolia, of Ecuador, has bicolorous scales with a very dark, indurated body and a single apical seta; however, these tend to be deltoid in shape. Alsophila woodwardioides, A. salvinii, and A. capensis have bicolorous, lanceolate scales with one to several apical setae. Concolorous to bicolorous, elongate to lanceolate scales with marginal and apical setae (FiGURE 8) are found in many species. In the Greater Antilles they are present in Alsophila hotteana, A. minor, A. brooksti, A. urbanii, and A portoricensis. Four continental species, 4. tryoniana, A. polystichoides, A. stern- bergii, and A. firma, also have this type of scale. Acaroid scales (FIGURE 9) are small and orbicular, usually with an indurated body and many dark, marginal setae; they may be reduced to tiny, setate squamules nearly without a scale body. Species in the Greater Antilles with acaroid scales include A/sophila hotteana, A. minor, A. brooksii, A. balanocar- pa, A. urbanii, and A. portoricensis. Among continental species, such scales occur in A. salvinii, A. setosa, A. sternbergii, A. cuspidata, and A. erinacea. Alsophila engelii, of Colombia and Venezuela, has two additional scale types not found in other American species of the genus. One is flabellate and setate (Ficure 10) (this type may be reduced to tiny, setate squamules with unsclerified bodies). The second is highly dissected and setate (FiGurE | 1), giving an axis the appearance of being pubescen Young axes may be densely decduausly scurfy with tiny, appressed squamules; this character was not found to be of taxonomic significance, however. Most species of Alsophila have trichomes on the abaxial surface of their smaller axes. Alsophila bryophila and A. auneae are characterized by a partic- ularly dense pubescence, while 4. amintae, A. nockil, A. paucifolia, A. salvinii, and A. capensis lack trichomes on their axes. A/sophila cuspidata and A. stern- bergii have stellate trichomes on their veins in addition to unbranched ones on their larger axes. Tiny, antrorse trichomidia may or may not be present on the abaxial surface 340 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 6 & 13 ' | >). oe Se de P wa ern $f ae Pes TEAL. 4 StU) FiGures 12-15. Chromosomes and spores of Alsophila amintae. 12-14, Conant & Kitfield 2274: 12, sporangium with 4 spore mother cells, x 200; 13, spore mother cell in diakinesis showing n = ca. 69 pairs, X 1850; 14, sporangium with 16 spores (arrow indicates spore partially obscured by sporangium), X 200. 15, Conant 595, spores, x 1500. of the veins. These are usually golden in color, but in A/sophila engelii they are whitish. Sori. The sori are exindusiate or with a hemitelioid (FiGuRE 48), meniscoid (FiGuRE 25), cyatheoid (FiGure 18), or sphaeropteroid (FIGURE 38) indusium. Sporangia are clustered about a usually columnar receptacle, resulting in a semiglobose to nearly globose sorus. They have an oblique annulus and a short 1983] CONANT, ALSOPHILA 341 stalk that meets the capsule to one side of the basalmost portion of the annulus. Gastony (1973, 1974) has shown that the archesporial cell typically divides twice mitotically, producing 4 spore mother cells (FiGuRE 12), and that each of these undergoes meiosis (FIGURE 13), producing a total of 16 spores per sporangium (FiGurE 14). Gastony and Tryon (1976) reported several species of Alsophila (including the American A. sa/vinii and A. capensis) with 64-spored sporangia. Spores of A/sophila are trilete and have a reticulate pattern of ridges (Ficure 15) (Gastony, 1973, 1974; Conant, 1975; Gastony & Tryon, 1976; Conant & Cooper-Driver, 1980). CYTOLOGY Chromosome counts in the genus 4/sophila have been n = 69 or n= ca. 69 pairs of bivalents. Brownlie (1958, 1965) reported n = 69 for four species from Australasia: 4. alata Fourn., A. colensoi Hooker, A. smithii (Hooker) Tryon, and A. tricolor (Colenso) Tryon (all as species of Cyathea). From the areas of India and Sri Lanka, Manton and Sledge (1954), Mehra and Singh (1955), and Ghatak (1962) have reported n = 69 or ca. 69 for eight species: Alsophila gigantea Hooker, A. hookeri (Thwaites) Tryon, A. khasiana Kuhn, A. /atebrosa Hooker, A. ornata Bedd., A. sinuata (Hooker & Grev.) Tryon, A. spinulosa (Hooker) Tryon, and A. walkerae (Hooker) J. Sm. (all as species of Cyathea). Manton (1954) cited n = 69 for Alsophila latebrosa (as Cyathea latebrosa), from Malaysia. Roy and Holttum (1965) cited n= ca. 69 for A. podophylla Hooker (as Cyathea podophylla), from southern China. Manton and Sledge (1954) reported n = 70 for A. capensis (L. f.) J. Sm., and Manton (1959) re- ported n = ca. 69 for A. camerooniana (Hooker) Tryon (both from Africa and both as species of Cyathea). The report of n = 70 for A. capensis was from a single plant at Kew Gardens. Since other species in the family have n= 69, cytological reexamination of this species in the field is desirable. Walker (1966) reported n = 69 for three Jamaican species, Al/sophila auneae, A. grevilleana, and A. tussacii (all as species of Cyathea). Gomez-Pignataro (1971) reported n = 69 for two Costa Rican species, 4. erinacea and A. firma (both as species of Cyathea). Conant (1976) and Conant and Cooper-Driver (1980) cited n = ca. 69 for 4. amintae, A. bryophila, A. amintae X A. porto- ricensis, and A. bryophila x A. portoricensis, all from Puerto Rico. Perhaps the most unusual aspect of the cytology of the genus A/sophila is that the two hybrids that have been examined are fertile diploids. Spores are well formed and germinate readily when sown on a nutrient agar medium. SPECIATION Ald Pg f ] The genus originated in the Old World, where 87 percent of the own taxa are ‘found. The American species of Alsophila are most closely related to the African and Madagascan elements of the genus (Holttum, 1981), and the presence of A/sophila in the New World is undoubtedly the result of long-distance dispersal from the east. Dust from the Sahara Desert 342 JOURNAL OF THE ARNOLD ARBORETUM [vVOL. 64 encountered by aerobiologists sampling air in the Caribbean provides evidence that wind patterns favor such long-distance dispersal (Prospero, 1968; Prospero et al., 1970). At least four separate migrations are believed to have established the basic geographic elements of American A/sophila. In some cases migration may have been recent, resulting in little morphological differentiation of the immigrant. Thus, A. capensis in southeastern Brazil is only slightly different from that species in southern Africa and was assigned subspecific rank in accordance with the degree of differentiation. A/sophila salvinii, of Mexico, Guatemala, and Honduras, is a distinct American species clearly related to species of Holttum’s sect. GYMNOSPHAERA (Holttum, 1974, 1981), of Madagascar and Borneo. In other cases, migrations may have occurred in the more distant past, allowing time for more morphological differentiation and resulting in new American species groups. It is more difficult to determine the relations between American and Old World species groups than it is to relate the above two American taxa to their Old World relatives. This may be a direct result of the diversification process. Although a few characters may still be held in common, shifts in gene frequencies result in a situation where characters that are obscure in the Old World group may become common and predominant in the New World group. Such may have been the case with the development of the squami- nate spine in the New World. Holttum (1981) suggests there are about 20 species with complex leaves and squaminate spines (excluding A/sophila setosa, as Cyathea setosa) that form a closely allied group. Gastony (1973) recognized only 15 such species (all as Nephelea, and with one suggested as a probable hybrid) and clearly demon- strated that they fall into two distinct groups, which are referred to below as the A. cuspidata and the A. woodwardioides groups. The South American species with imparapinnate lamina apices, ovate to elliptic pinnae (width/length quotient (0.23—)0.27-0.31), and sphaeropteroid indusia include A/sophila cuspidata, A. sternbergii, A. erinacea, A. incana, A. imrayana, and A. firma and are hereafter referred to as the A. cuspidata group. This appears to be related to the A. serratifolia group of Madagascar, which includes A. serratifolia (Baker) Tryon, A. marattioides (Kaulf.) Tryon, A. per- rieriana (C, Chr.) Tryon, and A. tsi/otsilensis (Tardieu-Blot) Tryon (all as species of Cyathea) (Tardieu-Blot, 1951) and is characterized by the imparapinnate lamina apex, a similar pinna size and shape, and (usually) a sphaeropteroid indusium. In the A. cuspidata group the lamina is slightly more complex (bi- pinnate-pinnatifid) and the petioles and stems are armed with conspicuous squaminate spines, whereas in the A. serratifolia group the lamina is bipinnate and the spines are inconspicuous or absent. The Greater Antillean species with a gradually tapering lamina apex, long, narrow pinnae (width/length quotient 0.12—0.17(-0.24)), and pinnate-pin- natifid to bipinnate lamina dissection (including A/sophila minor, A. bryophila, A, hotteana, A. amintae, A. abbottii, A. nockii, A. brooksii, and A. urbanii) are hereafter referred to as the A. minor group. Holttum (1981) comments that these Antillean species appear to be related to the group of African and Mad- 1983] CONANT, ALSOPHILA 343 agascan species (here referred to as the A. decrescens group) with a similar lamina apex, pinna shape, and lamina dissection. African species include 4. mossambicensis (Baker) Tryon, 4. camerooniana (Hooker) Tryon, and 4. wel- witschii (Hooker) Tryon (all as species of Cyathea). Madagascan species with these characters, too numerous to list in their entirety, include 4. madagas- carica Bonap., A. approximata (Bonap.) Tryon, A. zakamenensis (Tardieu- Blot) Tryon, and A. decrescens (Kuhn) Tryon (Tardieu-Blot, 1951; all as species of Cyathea). In the A. minor group the cyatheoid indusium is predominant, whereas in the 4. decrescens group the sphaeropteroid indusium 1s more com- mon among Madagascan species. Both the 4. minor and the A. decrescens groups include several species with aphlebioid to subaphlebioid basal pinnae; A. urbanii and A. zakamenensis are notably similar in this respect Once established, the South American element may have diversified to some extent, forming some or most of the species in the A/sophila cuspidata group. The sequence of species from southeastern Brazil north through the Andes to Mexico suggests that a process such as evolutionary migration (Tryon, 1971a, 1971b) may have been involved in forming this species group. Similarly, the Greater Antillean immigrant may have undergone some adaptive radiation, as frequently occurs in island archipelagoes, forming some or most of the species in the A. minor group. These geographic processes, however, are thought to account for only half to two-thirds of the species diversity within present-day American A/sophila. The remainder is believed to have arisen via allohomo- ploid speciation. Conant and Cooper-Driver (1980) have discussed the origins via allohomoploidy of 4. setosa, A. trvoniana, A. polystichoides, A. balano- carpa, and A. auneae (as species of Nephelea). The six Greater Antillean species with bipinnate-pinnatifid leaf architecture, spines, a gradually tapering lamina apex, and bullate scales (A. woodwardioides, A. jimeneziana, A. fulgens, A. grevilleana, A. portoricensis, and A. tussacii; hereafter referred to as the A. woodwardioides group) are interesting to consider in relation to the allohomo- ploid pathway. Gastony (1973) viewed them as perhaps being derived via divergent geographic speciation from an ancestral plexus similar to the spiny, pinnate-pinnatifid 4. auneae and A. balanocarpa (as Nephelea pubescens and N. balanocarpa), which presumably were related to the A. minor group. Al- sophila tryoniana, the only spiny continental species with bipinnate-pinnatifid leaf architecture, a gradually tapering lamina apex, and bullate scales, was interpreted as being related to the Antillean 4. woodwardioides group, and as perhaps being intermediate between the Antillean and the continental A. cus- pidata groups. Conant and Cooper-Driver (1980) have proposed a different origin for A. tryoniana (as Nephelea tryoniana), suggesting that the gradually tapering lamina apex and the bullate scales were inherited from 4. sa/vinii via allohomoploid speciation. Reappraisal of the A/sophila woodwardioides group in view of the allohomo- ploid pathway results in an interpretation different from that of Gastony (1973). The numerous instances of hybridization between the 4. minor and the A. woodwardioides groups (Conant, 1975, 1976; Conant & Cooper-Driver, 1980) and the extent to which they share unusual characters indicate an intimate, reticulate relationship of the groups. Characters such as the pubescent indusium 344 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 in A. minor, A. brooksti, and A. portoricensis, the bicolorous, lanceolate laminar scales with a single apical seta in 4. abbottii, A. amintae, A. nockii, and A. tussacii, and the subaphlebioid basal pinnae in A. amintae, A. abbottii, A. minor, and A. grevilleana result in a reticulate pattern of species relationships. The only significant differences between the 4. minor and the 4. woodward- ioides groups are that members of the former tend to be pinnate-pinnatifid and spineless whereas those of the latter are bipinnate-pinnatifid and spiny. The origin of the latter two characters may have been in the 4. cuspidata group. Migration of a member of this South American group into the Greater Antilles, subsequent hybridization with a member or members of the A. minor group, and later allohomoploid speciation could account for the bipinnate-pinnatifid leaf architecture, the spines, and the gradually tapering lamina apex of the 4. woodwardioides group. This would also explain the sphaeropteroid indusium in 4. fulgens and the adventitious buds in A. woodwardioides, characters oth- erwise restricted to continental species of the 4. cuspidata group. Following introduction of Andean characters into the Greater Antilles, allohomoploid speciation could account for all species of the A. woodwardioides group and for A. auneae and A. balanocarpa. HYBRIDS Hybrids in the genus A/sophila can be divided into two groups. The first comprises species of hybrid origin in which reproduction has been stabilized by allohomoploidy. Conant and Cooper-Driver (1980) have discussed the origins of A. auneae, A. balanocarpa, A. tryoniana, A. polystichoides, and A. setosa (as species of Nephelea), dealt with above in the section on Speciation. Holttum (1981) also suggests that A. sefosa is the result of hybridization between 4A. capensis and A. sternbergii (all as species of Cyathea). The second group consists of hybrids that are not reproductively stable or are too poorly known to be admitted to specific status. Diagnoses of these follow the enumeration of species in the Systematic Treatment. Comments below pertain to this latter group. The detection of hybrids has been facilitated by the presence of certain morphological markers. Lamina dissection, pinna shape, and presence of squaminate spines have been particularly useful. In the Greater Antilles six species (the Al/sophila woodwardioides group) have bipinnate-pinnatifid lami- nae, ovate to elliptic pinnae (width/length quotient 0.29 or more), and long, squaminate spines. Eight species (the 4. minor group) have pinnate-pinnatifid to bipinnate laminae, long, narrow pinnae (width/length quotient 0.17 or less, except 0.24 or less in A. urbanii) and are without spines. Crosses between the two groups can be recognized in the field by the combination of a bipinnate- lobed to bipinnate-pinnatifid lamina dissection with a longer and narrower (width/length quotient 0.17-0.29) pinna shape than is normally encountered in the more complex leaves. Plants with this intermediate lamina also usually have short spines, an intermediate condition in a cross between long-spined and spineless species. Crosses within groups are difficult to recognize and have been detected only by intensive field or herbarium study. 1983] CONANT, ALSOPHILA 345 Field observations have been made on six of the nine hybrids listed at the end of the Systematic Treatment. All were growing in or near ecologically disturbed areas. In some cases the disturbances were due to the activities of man, such as road building or the clearing of land for agricultural purposes. Three of the hybrids were found along the upper reaches of streams remote from any disturbance by man, and two were encountered numerous times in apparently virgin cloud forest. Small landslips are common along stream banks and in the water-soaked soil of the ie hea such disturbances provide places where natural hybridizations can oc It is difficult to determine whether the hybrid plants encountered in the field are F, or later-generation hybrids. Some tend to be rare and intermediate in morphology, suggesting they are the first generation. Others are very close to one parent or the other, indicating either a backcross or F, segregate origin. Conant and Cooper-Driver (1980) have demonstrated that Alsophila amin- tae X A. portoricensis (as A. dryopteroides X Nephelea portoricensis) 1s more abundant than either of its parents in one population and appears to be re- producing itself. These later-generation plants are morphologically uniform (see fig. 2 in Conant & Cooper-Driver, 1980), suggesting that this hybrid may be an incipient allohomoploid species. Recognition at specific rank, however, will await laboratory analysis of its reproductive stability. HERBARIA In addition to the herbaria listed in Lanjouw and Stafleu (1964) and cited by the recommended abbreviations, the following are also mentioned: her- barium of the El Verde Field Station of the Center for Energy and Environ- mental Research, Rio Piedras, Puerto Rico (Herb. Evrs), and the herbarium of the Jardin Botanico Rafael M. Moscoso, in Santo Domingo, Dominican Republic (Herb. JBRMM). SYSTEMATIC TREATMENT Alsophila R. Br. Prodr. Fl. Novae Holland. 158. 1810. Type: Alsophila australis R. Br Me aaa Blume, Enum. Pl. Javae, 242. 1828. Type: Gymnosphaera glabra Blume = Alsophila glabra (Blume) Hooker. Pees Gardner, London J. Bot. 1: 441. 1842. Lectotype: Amphicosmia riparia (Willd.) Gardner (Cyathea riparia Willd.) = Alsophila ee (L. f.) J. Sm. (for lectotypification, see R. M. Tryon, Contr. Gray Herb. 200: 2 Dichorexia Presl, Gefassbiindel Stipes der Farrn, 36. 1847 eee from Abhandl. eae Bohm. Ges. Wiss. V. 5: a 1848). Type: Dichorexia latebrosa (Hooker) esl = Alsophila latebrosa Hooke Phssanobot rya Alderw. Bull. Jard. Bot. Buitenzorg, II. 28: 66. 1918. Type: Thysa trya arfakensis (Gepp) Alderw. (Polybotrya arfakensis Gepp) = Alsophila bi- ene Rosenstock. Nephelea Tryon, Contr. Gray Herb. 200: 37. 1970. Type: Nephelea polystichoides (Christ) Tryon = Alsophila polystichoides Christ. 346 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Stem prostrate, decumbent, or erect, to 15 m tall and 15 cm in diameter (excluding adventitious roots), with or without squaminate spines. Leaves to 4 y 1.6 m, exstipitate to long-petiolate; petiole brown to black, smooth to tuberculate or with squaminate spines, the base with uniformly tan to bicol- orous, deltoid to lanceolate, marginate scales (these fleshy in 1 species) having single (usually dark) apical seta or both apical and marginal setae, with or without aphlebiae or subaphlebioid basal pinnae, occasionally with gap between basal aphlebiae, subaphlebioid basal pinnae, or basal pinnae and next higher pinna pair; lamina pinnate-pinnatifid to rarely quadripinnate, the base grad- ually tapering or abruptly narrowed, the apex gradually lapering and acute to scales, or glabrous; pinnae sessile to stalked, narrower to broader at base, apex blunt tipped to attenuate, the pinna-rachis with various combinations of tri- chomes, scales, and/or squamules, the veins free, 6 to 24 pairs, often with tiny, antrorse, golden trichomidia beneath; indusia lacking or hemiteloid, meniscoid, cyatheoid, or sphaeropteroid, glabrous or pubescent; sporangia with oblique annulus uninterrupted by stalk, usually with 16 trilete spores. (Description applies especially to American elements of genus; all variations of nondiagnos- tic characters of Old World species not included.) USE OF THE KEY The following is a key to complete and mature specimens; it includes char- acters of the lamina apex, middle pinnae, petiole, and basal pinnae, as well as data on size, posture, and armament of the stem, provenance, and ecology. Incomplete specimens can be identified if one has some knowledge of the geography of certain character combinations discussed above in the sections on Morphology and Speciation. Juvenile specimens may be problematic. For example, young plants of Alsophila bryophila and A. minor may have a con spicuous gap between the basal one to three pairs of pinnae or sjbiehichon pinnae and the main portion of the lamina. If the disjunct basal pinnae are reduced or lost, specimens often appear to have a long petiole, which may cause confusion at couplet 3 in the key. The descriptions of lamina complexity apply to the mid-portions of the middle pinnae. The user should beware of specimens from the Greater Antilles that are bipinnate-lobed to bipinnate-pinnatifid and have long, narrow pinnae (width/length quotient 0.17—-0.29). These may be hybrids and are discussed above in the section entitled Hybrids. They are enumerated at the end of the Systematic Treatment (nos. 31-39). Descriptions of indument and scales apply to the abaxial surface of axes. KEY TO SPECIES OF ALSOPHILA 1. Stem and petiole without pi ] te sp duced spi ionall present in 4. brooksii). 2. Lamina pinnate-pinnatifid to bipinnate; stem prostrate, decumbent, or erect. 1983] CONANT, ALSOPHILA 347 3. Lamina sessile to ae petiolate, gradually narrowed at base (abruptly nar- rowed in A. rupestr 4. Indusia eco. 5. Large, erect plants; stem always more than 2.5 cm in diameter 6. Indusia glabrous; stem less than 6 cm in diameter, less than 2 m tall; scales of pinna-rachis and costule mostly without lateral setae. 7. Pinna-rachis sparsely to moderately pubescent; basal pinnules often completely overlapping and clasping rachis; aoe and rachis tan to brown. [Hispaniola]. ............. 3. A. abbottii. . Pinna-rachis glabrous; basal pinnules not or only partly overlap- ping rachis; petiole and rachis dark purplish-brown. [Puerto Rico]. STARE Ah ube Tenia shakeela aaah te ban Glamegle eee ae 2. A. amintae. 6. Indusia pubescent; stem to 11 cm in diameter and 7 m tall; scales of pinna-rachis and costule with many conspicuous lateral setae. [Hispaniola, Cuba]. 6. A. 5. Small, prostrate plants; stem never more than 2.5 cm in pes 8. a iole and rachis black; pinna-rachis with trichomes only. [Colom- Se ese a alas ihe aes oad a aes eae ahaa ea 9. A. rupestris. Petiole brown; rachis stramineous; pinna-rachis with whitish to bi- colorous, lanceolate scales with single apical seta, and with trichomes 1 ~— oo and bullate scales. [Jamaica]. 4. Indusia meniscoid. 9. Pinna-rachis densely pubescent and with few nonsetate to setate, whit- ish, bullate scales; petiole and rachis purplish-brown. [Puerto Rico]. Mead osc) pea nate ae hee te a ee ee 5. A. bryophila. 9. Pinna-rachis sparsely to moderately pubescent and with acaroid scales and dark, ovate scales with apical and lateral setae; petiole brown; rachis PAM PELAIG |S. ae ceeeitnr arte tytaes alata eae a much eee setae eth 4. A. hotteana. 3. Lamina long-petiolate, ae narrowed at base. 10. Pinnules less than 5 mm wide; petiole never with aphlebiae. 11. Indusia Sac variously rupturing at maturity, glabrous; petiole and rachis black; stem erect. [Ecuador]. .. 10. 4. paucifolia. 11. Indusia anes intact at maturity, pubescent: petiole and rachis brown t stem prostrate. [Puerto Rico, Hispaniola, Cuba]. Mpls coke een Aasly eg death ah ede aes eect hetero enteg e 7. A. brooksii. 10. Pinnules more than 5 mm wide; indusia cyatheoid, occasionally with | or 2 clefts, glabrous (occasionally slightly pubescent); petiole with or with- out aphlebiae at base; stem prostrate to decumbent. [Dominican Re- DOU NG) hai eae chink tpt dex ey arte sine sev ee 8. A. urbanii. 2. Lamina cence pinnatifid to tripinnate (rarely quadripinnate); stem erect. 12. ade indusiate; lamina bipinnate-pinnatifid; petiole and rachis tan to brown. 13. Indusia hemitelioid; petiole with 1 to few pairs of aphlebiae at base: middle pinnae with width/length quotient ca. 0.36. [SE nape er eee er Terese . A. nockit. 13. Indusia deeply cyatheoid to sphaeropteroid; petiole ee aphlebiae; middle pinnae with width/length quotient ca. 0.27. [Colombia, Vene- 7 AWL=} E\ Eegge ie Dee renee ner See eae ee Ee RE eM Ae I. A. engelii. : late; lamina tri l ; petiole and rachis shiny black; petiole with several pairs of aphlebiae. Mens. Guatemala, 12. A on SE ee atsnese Ghee eae ethan greene ie diaeaee wees aut _ A. salvinii. 1. Stem and petiole with conspicuous squaminate spines. 14. Lamina pinnate-pinnatifid to bipinnate. 15. Pinnules sessile and often basally auriculate; basal pinnae reduced but not 348 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 subaphlebioid; middle pinnae with width/length quotient . 0.19. [Cuba]. fase adem aueae dean hs ee he cay cay wey aan a eh ape este Rens 15. A. balanocarpa. 15. Pinnules adnate and without basal auricles; basal pinnae reduced and often witl 2. [Jamaica]. Be reese teeie ead pea et ee ee et we oe ar A, auneae. 14. Lamina bipinnate-pinnatifid to tripinnate (rarely quadripinnate). mina apex gradually tapering. [Greater Antilles, except 4. tryoniana]. 17. Sorus indusiate; scales and trichomes of costae and costules various. 18. Indusia meniscoid to deeply cyatheoid, glabrous or pubescent; leaf tissue papyraceous to subcoriaceous. 19. Veins and indusia glabrous. 20. Basal pinnae reduced or not, not subaphlebioid; scales and trichomes of costae various 21. Costa usually sparsely squamose, the scales brown to blackish and with apical and lateral setae. 22. Pinna-rachis usually narrowly alate distally, the wings green; costa glabrous to sparsely pubescent; stem spines less than 5 mm a [Hispaniola]. itty, Baad tele ee eee 7. A, jimeneziana. . Pinna-rachis not alate isl costa pubescent, often densely so; stem spines to 15 mm long. [Jamaica, Hispaniola, Cuba. et ener fanaa Sale fae een eae 20. A. woodwardioides. . Costa usually densely squamose, the scales tan to bi- colorous and with a single apical seta. Lape Beato hese ana yonder et esta nee ee ussacil. . Basal pinnae often highly reduced and eee cos- ta squamose to furfuraceous and with dark-setate, bullate scales; pinna-rachis usually narrowly alate distally, with wings green. [Jamaica]. ............ 16. A. grevilleana. 19. Veins and indusia pubescent. [Puerto Rico]. .............. a tenes ars Sue ea iss ee me ace as eee meee 19. A. portoricensis. 18. Indusia sphaeropteroid, glabrous; leaf tissue coriaceous. bere IGAl cata die ssa bene cei teen eee wes 18... eee 17. Sorus exindusiate: costules sparsely pubescent and with bullate to sub- bullate scales. [Guatemala, Nicaragua, Honduras]. . 23. 4. tryoniana. 16. repo imparapinnate. [Mexico, Central and South America, Lesser An- tilles]. 23. Indusia meniscoid to sphaeropteroid, fou pinnae not subaphlebioid. 24. Lamina bipinnate-pinnatifid; indumentum of veins various; in- dusia various; width/length quotient of middle pinnae 0.23-0.32. 25. Veins with whitish, stellate trichomes or minute, white, stellate squamules. 26. Leaf tissue papyraceous; veins with white, stellate tri- chomes; petiole scales with as fag and with or with- out lateral setae. [Below 1200 m 27. Veins ae saa with stellate trichomes mile 3 or 4 more or less str aigh usia a rown, persistent mostly subsphaeropteroi [Nicaragua, Costa Rica, Panama, Colombia, Frenc Guiana, Ecuador, Ree Brazil, Peru, ee Parag Ways cur erases ese 6. A. cuspidata. . Veins and indusia with stellate trichomes having ir- regular arms; indusia firm, brown, and persistent (or No No in) NO oO NO ~ 1983] CONANT, ALSOPHILA 349 indument of indusia and veins less conspicuous and indusia fugacious). [Brazil, Paraguay]. ............ 27. A. sternbergii. 26. Leaf tissue coriaceous; veins with occasional minute, white, stellate squamules; indusia glabrous to squamulate; petiole scales with single apical seta. [700-2800 m alt.; Costa Rica, Panama, Venezuela, Colombia, Ecuador, Peru, Bolivia]. 24. A. erinacea. 25. Veins mostly without trichomes or with unbranched trichomes or with squamules not white or stellate. 28. Indusia subsphaeropteroid to seein glabrous or with few minute squamules; petiole spines t 29. Petiole scales with single apical seta; a pinnae not reduced; middle pinnae with width/length quotient ca. 0.30; pinna rachis not alate distally. [Costa Rica, eee Venezuela, Lesser Antilles, eae Ec- 2 ayana. 29, Patiole scales with apical and numerous oe setae; basal pinnae reduced; middle pinnae with width/length quotient ca. 0.23; pinna-rachis usually alate distally, with wings green. [Mexico, Guatemala, El Salvador Honduras, Nicaragua, Panama, Ecuador]. ......... 29, ma. . Indusia meniscoid, cyatheoid, or subsphaeropteroid, the latter type pubescent; petiole scales with single apical seta; petiole spines to 4 mm. [Colombia, Ecuador, Peru, Boliv- SATOONTING): apace Garage yen ty see ee 28. A. incana. 24. Lamina tripinnate (rarely quadripinnate); veins glabrous to sparsely squamulose; indusia sphaeropteroid, glabrous; width/length quo- tient of middle pinnae ca. 0.34. [Costa Rica, Panama]. ......... 30. A. polystichoides. 23. Indusia hemitelioid: basal pinnae often subaphlebioid. oe a zil, a gentina]. . seto No [oe) 1. Alsophila nockii (Jenman) Tryon, Contr. Gray Herb. 200: 29. 1970. Ficures 16-18. Cyathea nockii Jenman, J. Bot. 17: 257. 1879. Type: Jamaica, Parish St. Andrew, Cinchona, Nock 107 (holotype, presumably at K, not seen). PossIBLE ISOTYPE: Ja- maica, Nock s.n., s.d. (us!). Stem prostrate, to 20 cm long and 2 cm in diameter. Leaves 0.3-1.2 m by 11-26 cm; petiole 1-3 cm long, brown, base with persistent, tan to brown, often light-margined, lance-ovate to lanceolate scales with | apical seta; lamina pinnate-pinnatisect to bipinnate, the base gradually tapering, the apex gradually tapering, acuminate, the rachis stramineous, invested with lanceolate, flexuous, whitish to bicolorous scales with | apical seta (occasionally few marginal setae) and many rusty-brown trichomes beneath; pinnae 5.5—13.5 by 1-2.4 cm, sessile, conspicuously broader at base, gradually tapering at apex, adjacent ones widely paced, the pinna-rachis with few stiff, brown trichomes beneath, base with lanceolate, flexuous, whitish to bicolorous scales with | apical seta (rarely with ew marginal setae), these grading into whitish, setate, bullate scales apically; pinnules 6-15 by 2.5-3.5 mm, margin becoming slightly serrate at apex, the 350 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ne ‘ ue Ge tomes rane 1a i nO wu ~ A ee Ma 19 £ [lneese™ he AN = =< = Ficures 16-22. 16-18, Alsophila nockii. 16, 17, Nock s.n. (us 1,421,983): 16, central pinnae, < .5; 17, central pinnules of central pinnae, X 2. 18, Hart 13, indusium, X 15. 19, 20, A. amintae (Conant 1852): 19, central pinnae, x .5; 20, central pinnules of central pinnae, x 2. 21, 22, A. abbottii: 21, Conant 2118, central pinnae, X .5; 22, Conant 2121, central pinnules of central pinnae, x 2. basal ones much larger and more dissected, free to partially overlapping rachis, the pinnule-rachis with indument on undersurface similar to that of pinna- rachis, the veins 6 to 9 pairs, with tiny, antrorse, golden trichomidia beneath; indusia cyatheoid, glabrous. DISTRIBUTION AND ECOLOGY. Jamaica, Blue Mountains (Map 2). Steep, forested slopes; 910-1830 m alt. REPRESENTATIVE SPECIMENS. Jamaica. No further locality: Hart 13 (us), Hart 2838 (mo), ate September, 1879 (k), Jenman s.n., 1874-79 (Ny), J. P. 42 (us). Blue Moun- s: Moore s.n., 1896 (GH). St. THOMAS: Vinegar Hill, near Bath, Maxon 279] (us), nee 1355 (Ny). St. ANDREW: Cinchona, Harris 7322 (Ny). CONANT , ALSOPHILA 351 Se es aa oe “4 che ee era 2 - hd he a Donrins 3 Z at Se ee? bee. a pry, ome | ON 8 Maps 2-9. Distribution of species of A/sophila in the Greater Antilles: 2, 4. nocki; 3, A. amintae, 4, A. abbottii, 5, A. hotteana, 6, A. bryophila, 7, A. minor, 8, A. brooksit; 9, A. urban Alsophila nockii is the smallest and rarest species of Alsophila in the Greater Antilles. Jenman (1898, p. 49) reported it as ““common on a limited area of the disintegrated acclivous forest near Vinegar Hill collected since then. but it has rarely been Sey JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 2. Alsophila amintae Conant, Rhodora 83: 149. 1981. Figures 19, 20. Cyathea dryopteroides Maxon, Amer. Fern J. 14: 99. 1925. Type: Puerto Rico, near , Monte Cerrote, Britton & Brown 5424 (holotype, us!; isotypes, GH!, Mo!). Abophila dryopteroides (Maxon) Tryon, Contr. Gray Herb. 200: 29. 1970, not A/- sophila dryopteroidea Brause (= Thelypteris dryopteroidea (Brause) Reed), or A/- sophila dryopteridoides Domin (= Trichipteris dichromatolepis (Fée) Tryon). Stem erect, to 1.3 m tall and 5.1 cm in diameter. Leaves 0.5-1.6 m by 16- 43 cm; petiole 1.3-3.5 cm long, purplish brown, the base with deciduous, golden-brown, concolorous to occasionally lighter-margined, lanceolate scales with | apical seta; lamina pinnate-pinnatifid, the base gradually tapering, the apex gradually tapering and acuminate, the rachis purplish brown, deciduously puberulous beneath; pinnae 8-20 by 1|.3-3 cm, sessile or nearly so, broadest at base, gradually tapering at apex, adjacent ones widely spaced, lowermost reduced and subaphlebioid, the pinna-rachis with scales on undersurface bi- colorous, lanceolate, with | apical seta; pinnules 10-17 by 2.5-5 mm, margin becoming crenate to serrate at apex, the basal ones slightly larger and more dissected, partially overlapping rachis, the pinnule-rachis with scales on un- dersurface of base few, tan, bullate, with or without apical setae, the veins 7 to 9 pairs, with tiny, antrorse, golden trichomidia beneath; indusia cyatheoid, glabrous DISTRIBUTION AND ECOLOGY. Puerto Rico, Cordillera Central (MAp 3). Shaded ravines and cloud forest; 1000-1200 m alt REPRESENTATIVE SPECIMENS. Puerto Rico. Cordillera Central: Monte Jayuya, near Cerro de Punta, ca. 65 km WSW of San Juan, Conant 595 (GH, Ny, US), 682 (GH, Herb. Evers, rb. JBRMM, UJ, K, NY), 683 (F, GH, NY, SV), 689 (GH, Herb. JaRMM), Conant & eae 2256 (sv), 2301 (GH), 2309 (Us), Sargent 3141 (us); Monte Guilarte, W of Adjuntas, ca. km WSW of San Juan, Conant 1700 (us), 1852 (k), 1853 (F), 1856 (Ny), 185 : (GH). The name A/sophila amintae appears here in its corrected form (with the epithet in the genitive); the error was kindly pointed out by K. U. Kramer (pers. comm. The bac sei transition from normal to subaphlebioid pinnae in A/sophila amintae is similar to that found in the 4. decrescens group of Madagascar. This species hybridizes with A/sophila bryophila and A. portoricensis on Monte Jayuya and Monte Guilarte, Puerto Rico. 3. Alsophila abbottii (Maxon) Tryon, Contr. Gray Herb. 200: 29. 1970. FiGures 21, 22. Cyathea abbottii Maxon, Proc. Biol. Soc. Wash. 37: 98. 1924. Type: Dominican Re- public, Prov. Duarte [Pacificador], near San Francisco de oe Loma Quita Espuela, Abbott 205] (holotype, us (3 sheets)!: isotypes, GH!, NY!). Stem erect, to 1.6 m tall and 5.1 cm in diameter. Leaves 1.1-1.5 m by 30- 42 cm; petiole 1.5-3.8 cm long, brown, base with deciduous, golden-brown, concolorous to occasionally lighter-margined, lanceolate scales with | apical seta; lamina pinnate-pinnatifid, the base gradually tapering, the apex gradually tapering, acuminate, the rachis brown, deciduously puberulous beneath; pinnae 1983] CONANT, ALSOPHILA 253 14-21 by 2.1-3.2 cm, sessile, broadest at base, gradually tapering at apex, the adjacent ones contiguous, the lowermost ones reduced and subaphlebioid; the undersurface of pinna-rachis having stiff, erect trichomes, bicolorous, lanceo- late scales with | apical seta, and tiny, brown squamules with or without apical setae; pinnules 10-17 by 3—-4.5 mm, margin becoming serrate at apex, the basal ones much larger and more dissected, overlapping rachis, the pinnule-rachis with scales on undersurface of base rusty brown, bullate, with | apical seta, the veins 9 to 11 pairs, with tiny, antrorse, golden trichomidia beneath; indusia cyatheoid, glabrous. DISTRIBUTION AND ECOLOGY. Hispaniola (Map 4). Shaded montane rainforest; 725-1200 m alt. REPRESENTATIVE SPECIMENS. Haiti. DEp. bv Norp-Quest: Haut-Piton, near Port de Paix, Ekman H3728 (us). Dominican Republic. LA Veca: Loma de la Campana, Ekman H11515 (ny, us); between Bonao and El Rio, Jiménez & Marcano 3062 (us); Palo Aboqueteado, SW of Bonao, Liogier 14806 (Herb. J8RmMM); Firma de ania a Piedra Blanca, Liogier & Liogier 19943 Siac JBRMM); Loma de la Sal, 25 km Constanza, Judd 1386-1389 (Gu). DUARTE: Loma Quita Espuela, NE of San a. de Macoris, Abbott 2/45 (us), Conant ir (GH), 2117 (us), 2118 (Ny), Ekman H1227 (us). Alsophila abbottii is most closely related to 4. amintae, of Puerto Rico. These species are similar in size and appearance, and on the undersurface of the pinna-rachis both have bicolorous, lanceolate scales with one apical seta. A relationship also exists between these species and 4. nockii and A. tussacti, of Jamaica, which are the only other species of A/sophila in the Greater Antilles with this type of laminar scale. There is a gradual basipetal transition from normal to subaphlebioid pinnae in Alsophila abbottii—a trend similar to that found in the A. decrescens group of Madagascar. Several collections from the Dominican Republic, Prov. La Vega, Loma de la Sal (e.g., Judd 1388) are unusual in that the pinnae are widely spaced. Howard & Howard 8545 (a) is unusually pubescent and may represent a hybrid of Alsophila abbottii with A. minor. Jiménez & Marcano 3062 has leaves that are more dissected than usual and may be a hybrid involving a species of A/sophila with more complex leaves. 4. Alsophila hotteana (C. Chr. & Ekman) Tryon, Contr. Gray Herb. 200: pie 1970. FiGures 23-25. Cyathea ee C. Chr. & Ekman, Kongl. Svenska Vetenskapsakad. Handi. III. 16(2); 12. 1937. Type: Haiti, Dép. du Sud, near Camp Perrin, SE of Jérémie, Morne lV Etang, ote H10383 (holotype, presumably in Herb. Ekman, Port-au-Prince, not seen; isotype, vus!). Stem erect, to ca. 2 m tall and 8 cm in diameter. Leaves ca. 2 m by 50 cm, petiole 1-4 cm long, brown, the base with deciduous, brown, lanceolate scales with 1 apical seta; lamina pinnate-pinnatifid to pinnate-pinnatisect, the base gradually tapering, the apex gradually tapering and acuminate, the rachis tan, deciduously puberulous beneath; pinnae to 24 by 3.5 cm, sessile, slightly broad- 354 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ~ e 4 fee 7) = wm 7 ee < ° i] ea) ° 7 3 Qa oO jan oa oO Q ae) = eh oO =) io) me 5 = 3 a 28 oO 3 a fo] a oc S = je) = oO twa i=" f> 3 ON lateral faces of achenes usually with 1 or more ribs and/or glands; inflorescences usually either unbranched or branched only at lowest node(s) (Echinodorus Berteroi sometimes with panicles as in Alisma). B. Flowers — inflorescences (usually) with more than 3 oe per node umbelliform. 0.000.000.0000 cette ee Ee inedenis B. Flowers aa imperfect; inflorescences (usually) with only 3 i per node. 3. Sagittaria. 394 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 1. Alisma Linnaeus, Sp. Pl. 1: 342. 1753; Gen. Pl. ed. 5. 160. 1754. Perennial (or sometimes annual?), glabrous (or nearly so), usually emergent, sometimes submersed [or terrestrial] herbs with rosettes and scapes arising apically from upright, fleshy, bulb-shaped stems, these sometimes forming a series by growth of axillary buds. Leaves highly variable in shape and size, earliest leaves linear, later leaves [sometimes remaining submersed and linear], sometimes floating, then narrow-lanceolate [to linear], blade of emergent or terrestrial leaves elliptic, lanceolate, or ovate, obtuse to acuminate, often cus- pidate at the apex, cuneate to cordate at the base, with a conspicuous midvein, the longitudinal nerves diverging from the base of the blade or from the mid- vein, these interconnected by fairly regularly spaced, parallel, ascending vein- lets; petiole usually longer than the blade. Inflorescences 1 to several, erect [sometimes bent in 4. gramineum], usually taller than the leaves [except on submersed plants], pyramidal, compound panicles with long peduncles and (including the long pedicels) [1 or] 2 or 3(—5) orders of branching, sometimes bearing several hundred flowers, branches and pedicels often mixed, in (pseu- do-)whorls of up to 7(—10) [or more] members, with [1—]3—8(-10) nodes, each branch terminating in a flower, this often appearing as a member of an um- belliform cluster; bracts sometimes basally connate, usually subulate to lan- ceolate (sometimes oblong or ovate). Flowers perfect, with all parts separate (except for varied connate and adnate relationships at their bases). Sepals with hyaline margins, concave, persistent. Petals white (in 4. subcordatum with a faint yellowish spot near the base when fresh) [to pink or purplish], persisting only | day, with entire or slightly wavy margins [to coarsely notched or lobed toward the apex], usually rounded, about as long as or longer than the sepals. Stamens 6, originating as 3 antepetalous pairs (but this relationship not re- maining obvious), approximately as long as or longer than the carpels, anthers oblong, elliptic, [or nearly orbicular], filaments broadened toward the base, variably inserted on a ring of tissue above the receptacle, longer than the anthers; pollen grains with granulate, circular (or nearly so) pores, exine tectate, granular to spinulose. Gynoecium a ring of many nearly free carpels on a flat receptacle, these elliptic to obovate (tending to have the adaxial edges straighter than the abaxial edges), or D-shaped, not completely closed at anthesis, with nectaries at the basal adaxial edges, each with an outwardly curled style shorter than the length of the ovary [or the style straight and up to twice the length of the ovary], style inserted at or slightly below the adaxial edge of the broad summit of the ovary [sometimes inserted nearly centrally at the apex to below the middle of the adaxial side]; ovule anatropous or amphianatropous. Achenes ca. [10-]13- 20[-35], flattened, obovate to elliptic, sometimes ovate, with rounded apices, the adaxial edges often straight below the remnants of the styles (when present), with | or 2 abaxial grooves. Seed flattened, elliptic or obovate-oblong in outline (A. subcordatum). LECTOTYPE SPECIES: Alisma Plantago-aquatica L., the only one of Linnaeus’s original species not removed to other genera; see also J. K. Small, N. Am. Fl. 17: 43. 1909. (Name ancient, adopted by Linnaeus from Dioscorides.) — WATER-PLANTAIN. 1983] ROGERS, ALISMATACEAE 295 A genus of nine species (as recognized by Bjérkqvist, 1968) distributed nearly throughout the temperate Northern Hemisphere, extending northward slightly beyond the Arctic Circle (in the Old World), and to the south mostly above the Tropic of Cancer (with some notable transgressions in eastern Asia). Alisma only rarely occurs south of the equator: along eastern Africa south to Zimbabwe, in South Africa, Chile, Australia, and New Zealand. Alisma Plantago-aquatica accounts for most of these southern outposts, although 4. /anceolatum With. grows in Australia and Chile. Other than in eastern Africa, these occurrences probably stem from introduction by humans, although this matter is not settled in every case, with some authors especially prone to accept A. Plantago-aquat- ica as possible native to Australia (cf, Aston; Bjérkqvist, 1967; Hultén, 1962; Samuelsson, 1932). Alisma triviale Pursh and A. gramineum Le}. are widespread in the United States, but north and west of the range of the Generic Flora.? Alisma subcor- datum Raf. (A. Plantago-aquatica L. var. parviflora (Pursh) Torrey, A. parvi- flora Pursh, A. Plantago-aquatica L. subsp. subcordatum (Raf.) Hultén) grows in every state (except Louisiana?) east of a line connecting Texas, Nebraska, and South Dakota, in southern Canada (perhaps as far west as Saskatoon), and possibly in Arizona and/or in northern Mexico. Rubtzoff reported it (as prob- ably an introduction) in the Yosemite Valley of California. Alisma subcordatum shares the northern part of its range with the similar A, triviale. Although Samuelsson (1933) found intermediates not to be formed between these two species despite the wide geographic overlap (his 4. Plantago- aquatica subsp. brevipes (Greene) Samuels. corresponding to A. triviale as used here), other authors (e.g., Hellquist & Crow; Voss, 1972) have noted difficulties in distinguishing them. All of these authors (and Hendricks) included one or both in A. Plantago-aquatica, which (as circumscribed by Bjérkqvist, 1968) 1s native only to the Old World. That the three are distinct has been advocated by Bjérkqvist (1968), Fernald (1946), and Pogan (1963, 1964). Alisma sub- cordatum differs from A. triviale in being diploid (vs. tetraploid), and in a subtle set of characters probably largely related to the levels of ploidy: A. subcordatum has smaller floral parts, pollen grains, stomata, and achenes, and its styles are much shorter than the ovary (vs. about as long). (For detailed comparison see Bjorkqvist, 1967, 1968; Fernald, 1946; Pogan, 1964.) Not surprisingly, the two species are intersterile. Alisma is distinguished by its six stamens, its usually complex, highly branched panicles, its many carpels in a flat ring, and its achenes with one or two pronounced abaxial furrows. Although in his recent revision and related studies Bjorkqvist (1967, 1968) thought A/isma to be closest to Echinodorus, Baldellia Parl., Luronium Raf., Caldesia Parl., and Damasonium, he (1968, p. 98) found these genera “clearly distinguished from each other by many different mor- phological characteristics” and, on the basis of crossing experiments, asserted 3Contrary to Hendricks’s (p. 484) mention of Alisma Plantago-aquatica L. var. americanum (= A. triviale) in Arkansas, I have identified Demaree 17866 (GH), a duplicate of the collection he cited, as A. subcordatum sensu Bjérkqvist. 396 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 that Alisma is “isolated from all other genera by absolute sterility barriers.” (However, see Echinodorus for discussion of a possible intergeneric hybrid involving Alisma.) Alisma, Caldesia, and Limnophyton Miq. comprised Pi- chon’s subtribe Alismatinae. Except for the well-known Alisma Plantago-aquatica, the several species included in A/isma by Linnaeus and Micheli have since been removed to other genera. (Indeed, most species assigned to this genus by the authors who de- scribed them have since been placed elsewhere.) Buchenau (1903), in the first revision since that of Micheli, narrowed the generic circumscription by treating A. Plantago-aquatica as the sole species. His several varieties and forms of this species reflected the modern circumscription of the genus but not the trend by later botanists toward recognition of multiple species within this assemblage. Misleading environmental plasticity within species in contrast with relatively low variation in the genus as a whole, coupled with a worldwide distribution and a paucity of type material, has contributed to a confusing lack of accord in the delimitation and nomenclature of species in the three latest revisions. In the first of these, Samuelsson (1932) examined more specimens than his predecessors, recognized six species (one new), and provided particularly useful details of distribution and floral morphology. The revision by Hendricks, who accepted only four species, contributed a unique North American emphasis. His opinions differ radically from Samuelsson’s and from Bjérkqvist’s, partly in that his specific delimitations are least correlated with geographic areas. Nomenclatural and other problems with this treatment were outlined by Voss (1958). A detailed historical survey is available in Bjérkqvist’s revision (1967, 1968), which is buttressed by broadly based and clearly presented anatomical-mor- phological studies, new experiments to evaluate environmental modification, breeding experiments, and new cytological studies. Nine species emerged as reproductively and morphologically distinct, although as Bjérkqvist stated, nomenclatural research on these remains incomplete. Spontaneous hybridization between Alisma Plantago-aquatica and A. lan- ceolatum appears to be rare (Bjérkqvist, 1968; Kloos; Pogan, 1965). Some additional species can be crossed in the laboratory, but the resulting hybrids are sterile or reduced in fertility. BjOrkqvist detected no sterility barriers be- tween populations of the same species. Alisma has been the subject of numerous cytological studies, culminating in that by Bjérkqvist (1968), who reviewed and criticized the considerable pre- vious work. His determinations of chromosome number and morphology are backed by a broad sampling of specimens from each of the species in his revision. The base number in the genus is 7, with 4. Plantago-aquatica, A. gramineum, A. Wahlenbergii (Holmb.) Juz., A. subcordatum, and A. orientale (Samuels.) Juz. all being diploids, with 27 = 14. Alisma lanceolatum (2n = 26, 28), A. triviale (2n = 28), and A. rariflorum Samuels. (2n = 26) are tetraploids. Alisma canaliculatum Braun & Bouché is hexaploid (2 = 42). However, these figures (as well as the observations on the chromosomal morphology below) are not consistent throughout the cytological literature, with nomenclatural, taxonomic, and technical difficulties contributing to the inconsistencies. As was 1983] ROGERS, ALISMATACEAE aod already mentioned, A. subcordatum is not free from such problems. Bjdrkqvist (1968) found 2 = 14 for this species in 60 specimens. This number was also reported by Baldwin & Speese and Pogan (1963, 1964). Reports of 2n = 28 by Bjorkqvist (1961, retracted in 1968), Brown, and Heiser & Whitaker appear to be based upon misidentified material. The haploid chromosomal sets in each of the nonaneuploid species are fairly uniform: five chromosomes with median or submedian centromeres and two shorter chromosomes with subterminal centromeres. Consolidation between the two short types via reciprocal translocation in tetraploid plants is thought to account for the deficiency in number of chromosomes and the peculiar appearance of one pair in the aneuploids. (For elaboration see BjSrkqvist, 1968; Castro & Noronha Wagner; Erlandsson; Frey; Hendricks; Mikkola; Oleson; Pogan, 1962, 1967; Sharma; Sharma & Mukherji; Tschermak-Woess; and Wulff.) Alisma subcordatum grows on wet or periodically flooded soil or in shallow water in bogs, marshes, ditches, ponds, and streams. This species usually does not grow submersed and tends to have the broader forms of leaves, with linear leaves formed only by the seedling. When grown underwater, it produces lax floating leaves with narrow blades (Rhoades). Alisma gramineum and A. Wah- lenbergii, on the other hand, are adapted for growth completely submersed (the former tolerating depths to 4 m, according to Gliick & Kirchner) and conse- quently have predominantly linear leaves. They do not form floating leaves. When these species are grown out of water, the blade is narrow and tapered at both ends. (For additional information on heterophylly in A/isma, consult Arber, 1920; Bjérkqvist, 1967; Gliick, 1905; Gliick & Kirchner; and Scul- thorpe.) Alisma gramineum (sometimes) and 4. Wahlenbergii (characteristically) flower and set fruit underwater; the flowers of both have been described as cleistogamous or not opening underwater (Bjérkqvist, 1968; Wendt). The most common floral visitors to species of Alisma at some European and North American localities have been a variety of dipterans and short-tongued bees. Daumann (1964, 1965; see also Gliick & Kirchner, p. 607) found the pollen to be less cohesive than usual for entomophilous plants and demon- strated transfer of pollen by wind. Absence of agamospermy was shown by Bjorkqvist (1968), who also found every species to be self compatible, a con- dition of interest in connection with the several disjunct stations for some species. Since Alisma gramineum is particularly well suited to submerged growth, it is useful as an aquarium plant (see Stodola, Wendt). Other species are nuisances as weeds in rice and wild rice (Zizania) fields and in drainage ditches (Meeklah & Darwin; Ransom & Oelke; Samuelsson, 1932; Sculthorpe, p. 457). Members of the genus are used only rarely as food (Rickett, Wood ef al.), probably because of noxious and bioactive compounds in the plants. Several reports indicate toxicity to humans and other mammals, including irritation to human skin (Mitchell & Rook, Wood et a/.). By far the greatest economic significance of Alisma is in medicine, with a history extending from modern pharmacology back to A.p. 200 in China. Alisma species (probably mainly A. orientale) are cultivated in eastern Asia chiefly for the rhizome, which is sometimes sold 398 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 sliced or powdered to be used alone or mixed with other drugs. The sundry properties attributed to “A/ismatis Rhizoma,” too many to list here, are enu- merated in Perry, Stuart, and other references. Most of the pharmacological study concerning A/isma involves the ability of the crude drug or compounds isolated from it to diminish concentrations of cholesterol (while altering the balance of other lipids) in the liver and bloo of laboratory animals fed certain diets (Imai ef a/.; Kobayashi; Murata et al., 1970a). The lipotropic agent choline was detected by Kobayashi (1960a), but most work has been centered on a group of triterpenes (alisols) that reportedly have diuretic and antiinflammatory activity in rats (Murata et al., 1970a) beyond their effects on metabolism of lipids. Isolation and determination of structures of the alisols have been pursued by Murata and collaborators (see also Kamiya et a/.). Imai et a/. refined observations on the hypocholesterolemic activity of alisols (particularly one), compared their efficacies (see also Murata et al., 1970a), discussed the relationship between structures and activities, and mentioned that the alisol they studied most seemed to interfere with the ab- sorption of cholesterol by the intestine. REFERENCES: Under family references see ArBER (1920, 1925), ARGUE (1974, 1976), Aston, BALD- WIN & SPEESE, BEAL (1960, 1977), BOHDANOWICZ, BROWN, BUCHENAU (1857, 1868, 1882, AY EIS ; Q Tees. JOHRI (1936), Kau (1976, 1978), Krutzscu, Lieu (1979a), Love & Lave. Lusspock, MAHESHWARI (1950, 1964), Mever (1935a, b), MiCHELI, MITCHELL & ROOK, MuLLer, NETOLITZKY, OLESON, PAMMEL, PERRY, PETELOT, PICHON, POGAN, PUNT & REUMER, RICKETT, RIDLEY, ROBERTSON, SAL ISBURY, SATTLER & SINGH, SCHAFFNER (1896), SCHAUMANN, SCHILLING, SCULTHORPE, SHARMA, SINGH, SINGH & SATTLER (1972), SMALL, TANT, STICH, STODOLA, STUART, TOMLINSON, VISSET, Voss, WENDT, WILDER, and WoDE- HOUSE. ALEXANDROVA, L. A. A contribution to the taxonomy of the species of Alisma L. (In Russian.) Bot. Zhur. 52: 362-370. 1967. Best, R. G., M. E. WeHbe, & R. L. Linper. Spectral reflectance of hydrophytes. Remote Sensing of Environment 11: 27-35. 1981. [Spectral reflectance data on Alisma “Plan- tago” and other aquatic plants.] ByOrKovist, I. Chromosome numbers of the genus Alisma L. (In Swedish; English summary.) Bot. Not. 114: 281-299. 1961. [Updated in following papers. ] . Studies in Alisma L. I. Distribution, variation and germination. Op. Bot. 17: 1-128. 1967. . Studies in Alisma L. I]. Chromosome studies, crossing experiments and tax- ee Tbhid. 19: 1-138. 1968. [Includes a taxonomic revision. ] Buu, H. P., 8S. BapGerR, & B. S. SEKHON. A rapid screening of Mississippi aquatic and wetland plants for antimicrobial activity. ie Miss. Acad. Sci. 23(Suppl.): 79. 1978. {Extract from A. triviale (probably A. subcordatum) with antimicrobial activity.] Castro, D., & M. NoRONHA WAGNER. Preliminary observations on the cytology of the genus Alisma L. in Portugal. Genét. [bér. 2: 75-82. pls. 1, 2. 1950. CHUANG, C. K., W. P. CHANG, & H. F. SHEN. A preliminary report on the chemical constituents of the Chinese drug tse-hsieh. Trans. Sci. Soc. China 7: 207-215. 1932. [Abstract in Lingnan Sci. Jour. 12: 153. 1933. Notes on uses and chemistry. Fur- 1983] ROGERS, ALISMATACEAE 399 furaldehyde, asparaginine, a mixture of fatty acids, and other compounds were oun Goch ae ._D. Alisma gramineum in Vermont. Rhodora 70: 577-579. 1968. eee in North America; collected on shores of Lake Champlain.] DAUMANN, E. Zur Morphologie der Bliite von Alisma Plantago-aquatica L. Preslia 36: 226-239. 1964. [Chiefly concerned with the ‘‘Diskusring” around the inside base of the floral parts; nectaries and floral teratologies described.] . Insekten- und a bei Alisma Plantago-aquatica L. Osterr. Bot. Zeitschr. 112: 295-310. ERLANDSSON, S. eines studies of three Alisma species. Sv. Bot. Tidskr. 40: 427- 435. 1946 Espinosa Bustos, M. R. El género Alisma en Chile. Revista Chil. Hist. Nat. 31: 156, 157. 1928. [See also GUNCKEL.] FERNALD, M. L. Extreme variations of Alisma Plantago. Rhodora 3: 206. 1901. [Reports Alisma Plantago L. var. lanceolatum Hoffm. from No te American localities and A. Plantago L. var. graminifolium Wahlb. from North D ota. The North American representatives of Alisma Pine aquatica. Ibid. 48: 86- 88. 1946. Funta, M. On the drug “‘takusya” from Sinano in Japan. (In Japanese.) Jour. Jap. Bot. 17: 298-302. 1941. [A. orientale, includes photographs of rhizomes and plants in cultivation.] GUNCKEL, H. Dos especies del género A/isma encontradas en Chile. Revista Chil. Hist. Nat. 36: 42, 43. 1932. [Alisma Plantago-aquatica, A. lanceolatum; see also EsPINOSA Bustos. ] GUNTHER, R. T. The Greek herbal of Dioscorides. Frontisp. + ix + 701 pp. Oxford. 1934. [Alisma, 393; illustration and description of questionable identity.] Henpricks, A. J. A revision of the genus A/isma (Dill.) L. Am. Midl. Nat. 58: 470- 3. 1957 Hooper, D. On Chinese medicine: drugs of Chinese pharmacies in Malaya. Gard. Bull. Straits Settl. 6: 1-163. 1929. [A/isma, 7; preparation of drug, uses.] Hu tén, E. The amphi-Atlantic plants and their phytogeographical connections. Sv. Vet.-Akad. Handl. IV. 7(1): 1-340 (+ map). 1958. [Alisma gramineum, 19, 278; map, 259.] . Thecircumpolar plants. I. Vascular cryptogams, conifers, monocotyledons. /bid. IV. 8(5). 275 pp. 228 maps. 1962. [Alisma Plantago-aquatica sensu lato, A. sub- cordatum, 160, map no. 151; localities for A. Plantago-aquatica outside of its usual range; mentions fossil pollen from Pliocene; references to other maps. Imai, Y., H. MatsumMuRA, & Y. ARAMAKI. Hypocholesterolemic effect of alisol A-24- monoacetate and its related compounds in rats. Jap. Jour. Pharm. 20: 222-228. 1970. [Alleviation of effects of atherogenic diet.] ae T. Uber das Vorkommen von schuppen- oder haarférmigen Gebilden inner- halb der Blattscheiden bei monokotylischen Gewachsen. Bot. Zeit. 16: 177-179. 1858. [“Squamulae intravaginales” of 4. Plantago- ee compared with those of Butomus umbellatus. IsH1DoYA, T. Chinesische Drogen. Part 2. (2+) 147 pp. (+ index). Keyo, Japan. 1934. Aline 98, 99] Kamiya, K., T. Murata, & M. NisHikAwa. Biological-active triterpenes of — eee Ill. The X-ray crystallography of alisol A (23, 24)-acetonide | 1-mono- bromoacetate. Chem. Pharm. Bull. 18: 1362-1368. 1970. Pathe HEHE oa uatica var. orientale Samuels.; for other parts of this series, see papers by Murata et al.] Kimura, K. Heilpflanzen und ihre Anwendung in Japan. Therapiewoche 23: 4952. 1973. Kioos, A. W. Aanwinsten van de Nederlandse flora in 1945, 1946 en 1947. Nederl. Kruidk. Arch. 57: 199-243. 1950. [Alisma x ee Schotsm. = A. lanceo- latum With. x A. Plantago-aquatica L., 211, 212.] 400 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 KosayaSsHI, T. Studies on lipid metabolism. Il. Manufacture of lipotropic substance m Alisma Plantago L. (“takusha’’). Jour. Pharm. Soc. Japan 80: 1456-1459. 1960a. III. Lipotropic effect of Alisma Plantago L. (“‘takusha’’) extract. [bid. 1460- 1465. 1960b. IV. Several biological criteria of Alisma Plantago L. (“takusha’’) ex- tract. [bid. 1465-1468. 1960c. Lous ey, J. E. Alisma gramineum in Britain. Proc. Bot. Soc. Brit. Isles 2: 346-353. pl. 6. 1957. [Includes map showing distribution in Europe and Africa; comparison with other European species. ] Love, J. E. A method for obtaining mitotic figures in seedlings of Alisma [Dill.] L. Trans. Illinois Acad. Sci. 49: 43-46. [1956] 1957. [Includes drawings of seedlings and early growth. ] LUNELL, J. The genus 4lisma in North Dakota. Bot. Gaz. 43: 209-213. 1907. [Alisma Plantago-aquatica, A. arcuatum Mich., eases J & MEEKLAH, F. A., & R. K. Darwin. Control of water plantain in Otago and Southland. New Zealand Jour. Agr. 106: 145. 1963. [- see Plantago-aquatica occludes drain- age ditches. ] Meyer, F. J. Beitrage zur Anatomie der Alismataceen. V. Die Gattungen Damasonium und Alisma in Lichte der Anatomie. Beih. Bot. Centralbl. 54A: 156-169. 1935. [Only slight anne oun d.] Mikko La, L. Th m e number and chromosomes of Alisma Plantago-aquatica L. in Finland. Arch. Soc. Zool. Bot. Fenn. 15: 43-47. [1960] 1961. [2n = 14; pho- tographs and drawings of chromosomes; discussion of technical difficulties that may have led other authors to obtain different results. Morton, J. D. Water plantain (Alisma Plantago-aquatica): opening of flowers. Glasgow Nat. 18: 454. 1966. Murata, T., Y. Imai, T. Hirata, & M. Miyamoto. Biological-active triterpenes of a ie Rhizoma. 1. ere of the alisols. Chem. Pharm. Bull. 18: 1347-1353. _ SHINOHARA, & M. Miyamoto. Biological-active triterpenes of Alismatis pra IV. The structures of alisol B, alisol B monoacetate and alisol C mo acetate. Some reactions of the a-hydroxy epoxide of the alisol B derivatives. Chem. Pharm. Bull. 18: 1369-1384. 1970b. Nacao, Y., 8. Fuyioka, F. KAWANISHI, & T. MATSUOKA. Studies on the quality of the Chinese drug “‘zexie” and on the cultivation of Alisma orientale. I. On the culti- vation of the A/isma plant. (In Japanese; English summary.) Jour. Takeda Res. Lab. 34: 449-454. 1975.* Parrick, F. M., & M. W. Loutir. The uptake of heavy metals by epiphytic bacteria on Alisma Plantago- lee Water Res. 11: 699-703. 1977. [Bacteria on leaves accumulate certain heav tals. ] — E. Karyological adios in Alisma L. The ae and the eye evolution of . lanceolatum With. Acta Biol. Cracov. Bot. 5: 1-28. pls. 1-5. 1962. [Criticized . ByORKOVIST, 1968: table of chromosome ee and species; idiograms; lit- erature review concerning 4. /anceolatum.] Taxonomical value of . ekg triviale Pursh and Alisma subcordatum Rafin. Canad. Jour. Bot. 41: 1011-1013. 1963. . Taxonomical value of A/isma a ake and Alisma triviale Pursh. Acta Biol. Cracov. Bot. 6: 185-202. pl. 9. [1963] 1 Cytotaxonomical remarks on two Asiatic cee es of Alisma L. [bid. 10: 185- 193. pl. 26. 1967. [A. Plantago-aquatica subsp. orientale, A. canaliculatum; map of distribution in Asia; discussion of possible origin a taxonomic rank of ea orientale, these taxa compared with each other and with 4. Plantago-aqua- tic oe J. K., & E. A. Oc_tke. Common waterplantain (A/isma triviale) saa eeiE with ae rice (Zizania palustris). Weed Sci. 30: 10-14. 1982. [4. triviale a pest 1 1983] ROGERS, ALISMATACEAE 401 the eee of wild rice in Minnesota; yield losses avoidable if water plantain is controlled ea RHOADES, R. W. ae aquatic form of Alisma subcordatum Raf. Rhodora 64: 227-229, 1962. [Plants of this species can produce floating leaves if submerged at any time in their development. ] Rustzorr, P. Notes on the genus Alisma. Leafl. West. Bot. 10: 90-95. 1964. [Criticism of revision by HENDRICKS; . anceolatum, A. Plantago-aquatica, A. subcordatum in Northwestern United Stat SAMUELSSON, G. Die Arten der eee Alisma L. Ark. Bot. 24A(7): 1-46. pls. 1-6. 1932. [Alisma “‘trivialis” Pursh treated as a synonym of A. ea includes SE ane of individuals of some species.] Alism ee 3: 88-90. as 73-75. 1933. [Includes distribution ape s for Fs SATTLER, R. Geared of flowers. A photographic text-atlas. xxvi + 207 pp. Toronto & Buffalo. 1973. [A. triviale, 168-171; includes bibliography and discussion of pre- vious work SEIDEL, K. Physiologische Leistung von Alisma Plantago L. (Froschloffel). Naturwis- senschaften 58: 151. 1971. [This species grown under a range of concentrations of 109-118. 1974. [Review of literature concerning the gynoecium; anatomical illus- trations of pedicel, flower, and carpel; description of germination; drawings of seed- lings and inflorescence. ] Suara, A. K., & R. N. MuKHerRsI. Cytology of two members of Alismaceae. Bull. Bot. Soc. Bengal 9: 32-35. [1955] 1956. [A. Plantago, 2n = 16; Limnocharis flava, 2n = 20.] TsCHERMAK-Woess, E. Zytologische Untersuchungen an den Alisma-Arten der Um- gebung Wiens. Osterr. Bot. Zeitschr. 95: 270-276. 1948. Voss, E. G. Confusion in Alisma. Taxon 7: 130-133. 1958. [A review of the revision by HENDRICKS. ] Woop, H. C., C. H. LAWALL, H. W. YouNGKEN, A. OsoL, I. GriFFITH, & L. GER SHENFELD. The eee of the United States of America. ed. 22 (with supple- ment). xix + 1894 + 76 pp. Philadelphia & London. 1940. [Alisma “Plantago L.,’ Wu trr, H. D. Chromosomenstudien an der schleswig-holsteinischen Angiospermen- Flora. V. Ber. ree Bot. Ges. 63: 64-70. 1950. ee Plantago-aquatica L., t).] Wyp er, H. Morphologische Mittheilungen. Flora 46: 81-90, 97-105. pls. 1-4. 1863. [Alisma “Plantago L.,” 87-90, 97-100; a detailed study of the shoot relations and branching pattern in the inflorescence of this species. ] 2. Echinodorus Engelmann in Gray, Man. Bot. 460. 1848. Annual or perennial, submersed or emergent aquatic herbs, glabrous [or with sparse, single, stellate, or tufted hairs on stems, leaves, petioles, axes of the inflorescence, calyces, and bracts]. Leaves sometimes polymorphic (in our area true of E. tenellus and especially E. Berteroi), submersed leaves tending to be ribbonlike and emersed leaves petiolate, the blade highly variable, linear to elliptic, lanceolate, deltoid, or cordiform, [retuse or] rounded to long-acuminate at the apex, cuneate to cordate at the base, often with translucent markings, 402 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 FiGure |. Alisma and Echinodorus. a-l, 4. subcordatum: a, small plant with fruits, note base of second plant of axillary origin, x 4; b, leaf blade, x 34; c, bract from lower whorl of main axis of scape, x 1; d, flower, showing separate carpels in a ring, X 10; e, flower in vertical section slightly off center, carpel opened to show the single ovule, x 15; f, stamens, petals, and sepals from above, x 12; g, cluster ofachenes enclosed by persistent sepals, X 10; h, achene, remnant of style at upper left, x 10; 1, diagrammatic transverse section of achene cut near the middle, adaxial edge at left, Bericarp unshaded, the 2 arms of the bent embryo at center (stippled) surrounded by thin seed coat, endosperm absent, near apex, x 6; n, £. cordifolius, resin glands on side, x 6; 0, E. tenellus var. parvulus, without resin glands, x 12 1983] ROGERS, ALISMATACEAE 403 lateral nerves usually originating from the base of the blade (sometimes from the midvein), extending to the apex or margin, interconnected by usually par- allel and conspicuous veinlets (in some narrow leaves the veinlets arising pinnately from the midvein and extending to or nearly to the margin, or veinlets sometimes inconspicuous or absent). Scapes single or clustered, erect to pros- trate, generally longer than the leaves, racemose, paniculate, or umbelliform, bearing at the nodes (pseudo-)whorls of sometimes basally connate, subulate to ovate [or oval] bracts, these subtending a variable number of pedicels and/ or branches, often enclosing a cluster of smaller inner scales. Flowers perfect. Sepals with conspicuous longitudinal nerves, usually concave. Petals white or sometimes pink [rarely yellow], longer than the sepals, in our species rounded or emarginate at the apex; nectaries adaxial on the bases of petals and on carpels. Stamens (6-)9-30 [or more], with basifixed or versatile, latrorsely dehiscing anthers; pollen grains usually more or less spheroidal, with circular or irregularly shaped, granular pores, exine granular to spinulose (grains ovate to reniform with the exine reticulate in £. Berteroi fide Argue, 1976). Carpels numerous and free on a usually convex receptacle, variably elongate, each with an apical or lateral style. Fruit a usually spinose, aggregate head of beaked (beak small or absent in F. tenellus var. parvulus), ribbed [ribs rarely absent], compressed, elongate achenes, elliptic or widest above the middle, often slightly curved or roughly D-shaped, usually with | or more yellowish (dry) glands (absent in E. tenellus). Seed with a smooth or spotted integument. (Including Helianthium (Engelm. ex. J. D. Hooker) J. G. Smith in Britton, Manual, ed. 2. 54. 1905. Type species: H. tenellum (Martius) Britton = E. tenel/us (Martius) Buch.) Lectotype species: Alisma rostratum Nutt. = Echinodorus Berteroi (Sprengel) Fassett; see J. K. Small, N. Am. Fl. 17: 46. 1909. (Name probably from Greek, echinos, hedgehog, and doros, a leather bag, in reference to the spinose achenes.) — BURHEADS. Nearly 50 species in two subgenera and 13 sections distributed from the northern United States to Patagonia. Inasmuch as most of the species are from Central and South America and are known from meager collections (many by only one, and some only from cultivated plants), the genus as a whole remains poorly known. Over half of the species have been described within the last three decades, and more will surely be discovered with continued exploration in the neotropics. Hence generalizations about Echinodorus—especially those made before the efforts of Fassett and Rataj—still rest upon a superficial sam- pling. Three species of three sections from the two subgenera are indigenous to the United States, all of them occurring in the Southeast. Subgenus HELIANTHIUM (Engelm. ex J. D. Hooker) Fassett (carpels 20 or fewer in a loose head, anthers basifixed) is represented by Echinodorus tenellus (Martius) Buch. (Alisma tenellum Martius, E. parvulus Engelm., Helianthium parvulum (Engelm.) Small). This member of sect. TENELLI Fassett (leaf blades linear-lanceolate to elliptic, ribs of achenes not crested, inflorescences with one or few whorls) ranges from the northern United States to southern Brazil. Fassett recognized four varieties; Rataj (1975), only two (he reduced one to synonymy with var. tenel/us and elevated the other to the rank of species, 404 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 thereby eliminating most of the West Indies from the range of FE. tenellus). The variety in our area, FE. tenellus var. parvulus (Engelm.) Fassett, treated by some recent authors as a distinct species, grows in Cuba and Mexico, and sporadically in the area defined by Texas, Kansas, Michigan (see Voss), Massachusetts (possibly extirpated in this state), and Florida (not in the Appalachians). Reports from the northern shore of Lake Superior remain questionable (see Agassiz, Parry, Scoggan, Rosendahl & Butters). These small plants are distinguished from the other two species in the Southeast by having (6-)9 (vs. 12 or more) stamens, no pellucid lines in the leaves, umbelliform inflorescences (vs. racemes or panicles), and achenes lacking glands and conspicuous beaks, and by their pseudostoloniferous habit (because of which this species sometimes forms mats). Subgenus ECHINODORUS (carpels many 1n a dense head, anthers versatile) is represented in our area by two species from two sections. The sole species of sect. ECHINODORUS (sect. Berteroii Rataj) (strongly heterophyllous, stamens (9-)12, achenes rostrate, these with one gland or none), Echinodorus Berteroi (Sprengel) Fassett (A/isma Berteroi Sprengel, A. rostratum Nutt., E. rostratus (Nutt.) Engelm.), has often been confused with or included in our other species, E. cordifolius (L.) Griseb. (Hence “FE. cordifolius” in literature before clarifi- cation by Fernald and Fassett should be interpreted with care.) Echinodorus Berteroi var. Berteroi (as circumscribed by Rataj) (2m = 22?; see Heiser & Whitaker) is distributed along the southern edge of the United States from mid-California to Georgia, and southward to southern Mexico; in much of the drainage of the Mississippi River in the area defined by Ohio, South Dakota, Oklahoma, and Arkansas; and throughout the West Indies. (See Fassett for discussion of doubtful reports that would broaden the range.) A disjunct variety (var. patagonicus Rataj) grows only in Argentina. Echinodorus Berteroi var. lanceolatus (Engelm.) Fassett (E. cordifolius var. lanceolatus Mack. & Bush) was found in Rataj’s (1975) experiments to differ from var. Berteroi merely as a result of envi t and was accordingly reduced to synonymy. This highly variable species produces at least three distinct forms of leaves and sometimes occurs as dwarf individuals only a fraction of the usual size. Echinodorus Berteroi differs from F. cordifolius by having upright, often compound inflorescences; sepals with smooth (vs. papillose) abaxial nerves: glands (when present) extending farther toward the apex of the achene; and usually 12 (vs. over 20) stamens. One of eight species of sect. CorpiroLi Rataj (stamens 24-30, leaves with nonreticulate pellucid lines or spots), Echinodorus cordifolius (A. cordifolia L., S. radicans Nutt., FE. radicans (Nutt.) Engelm.), 27 = 22, is distributed along the Coastal Plain from southern Texas (and according to some reports in Mexico) to the vicinity of Washington, D. C., and in the drainage of the Mississippi River to mid-Tennessee in the east, extending westward to Kansas and Oklahoma and ranging north to central Illinois. Rataj (1975) reported this species in Venezuela. Echinodorus cordifolius differs conspicuously from the other two species in having young plantlets mixed with flowers at the nodes on arching or procumbent inflorescences. (Note, however, that De Wit reported that FE. Berteroi under cultivation in deep water also forms plantlets on the inflorescence.) 1983] ROGERS, ALISMATACEAE 405 Plantlets arise from vegetative buds on upright and horizontal inflorescences in several species of subg. ECHINODORUS, where such buds terminate lateral, bostrycoid, flower-bearing complexes and sometimes also occur singly in axils of bracts. In Echinodorus tenellus, E. quadricostatus Fassett emend. Rataj, and probably other species of subg. HELIANTHIUM, modified prostrate inflorescences on submersed individuals form only plantlets, never flower buds. These in- determinate, runnerlike pseudostolons resemble typical alismataceous inflo- rescences in bearing pseudowhorls of scale leaves at the nodes; on the pseu- dostolons the only other structure at each node is a single vegetative bud (Charlton, 1968, 1973). Richard published the name Echinodorus in 1815 accompanied only by the scarcely informative ‘“‘alismae polyandrae” long before Engelmann’s generic description in the first edition of Gray’s Manual. Buchenau (1868) listed 15 specific names and discussed the application of some of them. Micheli mono- graphed Echinodorus in 1881 and included 17 species, about half of which have since been removed to other genera or have otherwise undergone changes in name. Buchenau contributed comments the following year, and a revision including 20 species in 1903. Fassett’s treatment of the species in North Amer- ica and the American tropics contains useful explanations of decisions con- cerning taxonomy and nomenclature. In a revision of the entire genus that followed preliminary papers, Rataj (1975; see also 1970, 1973, 1974, 1978), agreeing with Pichon and apparently with Fassett, diverged from Buchenau and Micheli by excluding all species in the Old World. Rataj’s infrageneric classification corresponds closely to Fassett’s, except that several species were added to both subgenera and subg. EcHINopoRUS was subdivided into 11 sections. Echinodorus and Sagittaria, both in Pichon’s subtribe Sagittariinae, are more similar to each other than either is to A/isma. They both usually have achenes with glands or resin ducts on the faces, carpels crowded onto a dome-shaped receptacle, spherical pollen (vs. polyhedric in A/isma), and the diploid chro- mosome number of 22 (vs. multiples of x = 7 in Alisma; note, however, that the chromosome number is established for only one species of Echinodorus). Echinodorus differs from Sagittaria in having perfect (vs. usually imperfect or a combination of perfect and imperfect) flowers; usually plumper carpels and achenes; a stronger tendency toward bostrycoid complexes in the inflorescence (and thus more than three flowers per node); pollen grains with smaller spinules (Argue, 1976); one pair (vs. two or three pairs) of chromosomes with nearly median centromeres; and two (vs. no) pairs of chromosomes with satellites (see Baldwin & Speese; Beal, 1960). Although similar in appearance to the terminal pseudostolons produced by species of Echinodorus, stolons of Sag- ittaria are axillary branches. Differences in the circumscription of Echinodorus complicate comparison of generalizations made about it by different authors. After the first appearance of the generic name but before the monographic works enumerated above, Echinodorus was usually included in A/isma. Baillon placed it under Sagittaria. Several species previously regarded as components of Echinodorus by one or more botanists have been the bases of other genera of Alismataceae. Among 406 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 these are Baldellia (containing the often-mentioned Echinodorus ranunculoides (L.) Engelm.), Ranalisma, Burnatia M. Mich., Rautanenia Buch., Albidella Pichon, and Helianthium. Except for the last two, all are limited to the Old World. Echinodorus is compared with some of these genera in Argue (1976), Charlton (1973), Pichon, and Rataj (1975). Even as Echinodorus is circum- scribed by modern authors, a troublesome heterogeneity has repeatedly been mentioned. In 1974 Argue believed that palynological evidence substantiated the sep- aration of Baldellia from Echinodorus. Similarly, Wodehouse thought that Baldellia (Echinodorus) ranunculoides, together with “FE. cordifolius” (probably E. Berteroi), comprised a palynological type apart from five other species of Echinodorus. In a later survey involving 20 species of Echinodorus, Argue (1976) found that evidence from palynology “might ultimately provide some clues for its dismemberment” and assigned the pollen of Baldellia and Echi- nodorus to different morphological categories (acknowledging intermediacy), but he did not repeat his assertion of 1974. On the other hand, Charlton (1973) found Baldellia ranunculoides to show “strong similarities” to Fassett’s sect. TENELLI (but not, he emphasized, subg. EcHiNoporus sensu Fassett) in having pseudostolons and a determinate, morphologically similar inflorescence. Bal- dellia differs from Echinodorus in its 6 (vs. usually more) stamens and tends to have more terminal styles Although Ranalisma likewise resembles species of Echinodorus in producing pseudostolons, those of Ranalisma (humile) differ in being sympodial (vs. monopodial) and in having bracts in pairs as opposed to pseudowhorls of three (Charlton, 1968, 1973; Charlton & Ahmed, 1973b). That the aerial inflores- cences of Ranalisma are sympodial further separates Ranalisma from Echi- nodorus and from the Alismataceae in general (Charlton, 1968, 1973; Charlton & Ahmed, 1973b). Charlton & Ahmed (1973a, b) concluded that the sum of evidence from floral and vegetative structures indicates an isolated position for Ranalisma within the Alismataceae (cf. Gagnepain). Development of secretory canals, a familial characteristic, reaches an extreme in leaves of species of Echinodorus. Sometimes branched, the canals may accompany veins or be free in the mesophyll. Appearing as translucent lines or spots or as a reticulum, these pellucid markings are helpful in recognizing the genus, and they vary sufficiently in shape, size, and distribution to serve as taxonomic characters for species. Elaboration on their structure and distri- bution can be found in Meyer’s (1932) anatomical survey of leaves from several species. All three species of Echinodorus in the Southeastern United States grow in permanently or periodically wet mud or sandy soil, usually in (or on the shores of) streams, ponds, temporary pools, ditches, marshes, and swamps. Echino- dorus Berteroi tolerates “‘almost dry places”’ as well as brackish water according to Rataj (1975). Rand noted that FE. tenellus seemed sometimes to bloom in Massachusetts while completely underwater. Rataj (1975) indicated that hybridization occurs among several of the ex- traregional species and also noted alleged hybrids between Echinodorus tenellus var. tenellus and var. parvulus. Such hybrids were collected even in Texas and 1983] ROGERS, ALISMATACEAE 407 Florida, states far from the geographic range he provided for var. tenellus. Certain collections suggested “‘introgression”’ between FE. Berteroi and E. cor- difolius to Fassett The intergeneric hybrid between Echinodorus (Baldellia) ranunculoides and Alisma Plantago-aquatica has been reported not entirely independently by Durand & Pittier, Gliick (1913), Knobloch, and Wehrhahn. Ina comprehensive comparison of supposed hybrids and parental species, Gliick encountered nu- merous points of intermediacy, the most impressive in the anatomy of the fruit. Wehrhahn assigned the name A/ismodorus Muretii to the plant he iden- tified as resulting from the cross under consideration. However, after artificially pollinating over 100 flowers, BjGrkqvist (1968, see Alisma) failed to produce this hybrid and suggested alternative identifications for specimens and drawings considered by others to represent it. Because of their attractive foliage (which is varied in color, texture, and shape), hardiness, ease of cultivation, and in some cases vegetative propagation from the inflorescence, species of Echinodorus are popular for cultivation in aquaria; at least 20 have been used this way, among them the three species native to the United States. Confusing disharmony between specific names applied by aquarists and those applied by taxonomists is partly alleviated in Rataj’s revision. The plants are usually marketed as ‘““Amazon swordplants or under other names with “swordplant” a pea sometimes also as “cellophane plants” (e.g., &. Berteroi, with its membranaceous submersed leaves). (For information on Echinodorus in the aquarium see De Wit, Klee, Stodola, and Wendt.) A phycomycete tentatively identified as Aphanomyces euteiches Drechsler caused severe loss of “FE. brevipedicellatus Buch.” (E. amazonicus Rataj?) at a nursery in Florida during 1970 and 1971. Four other species of Echinodorus appeared to be resistant to the disease (Ridings & Zettler). In South America and the West Indies, species of Echinodorus are attributed with impressive medicinal benefits. Tea from the leaves, extract from under- ground parts, and other preparations have been used as an invigorating bev- erage, a diuretic, a laxative, and an astringent gargle. They have been used to clean and heal the skin, to “purify the blood,” and to counter edema, various ailments of the kidneys and liver, and other afflictions (Corréa, Peckolt, Penna, Roig y Mesa). As reported by Grosourdy, chopped roots (and, to a lesser extent, leaves) from FE. cordifolius (E. Berteroi?) rubbed onto the skin cause blistering. “Tubers” of E. grandiflorus (Cham. & Schlecht.) M. Mich. are boiled and consumed by the Mataco Indians in Argentina (Steward). According to Torrey, Mohave Indians used “‘seeds” of E. rostratus (E. Berteroi?) as food. Brazilian species are used as a source of dye for textiles (Peckolt). REFERENCES: Under family references see ARGUE (1974, 1976), BAILLON, BALDWIN & SPEESE, BEAL (1960, 1977), BENTHAM & Hooker, BUCHENAU (1868, 1882, 1903), CHARLTON (1973), CHARLTON & AHMED (1973a, b), CURRY & ALLEN, DAHLGREN (1934), DAHLGREN & CLIFFORD, DE WIT, EBER, EICHLER, HEISER & WHITAKER, KAUL (1976, 1978), LeEIns & STADLER, MICHELI, PECKOLT, PICHON, RATAJ (1970, 1978), RickETT, RIDLEY, SATTLER & SINGH, SCOGGAN, SCULTHORPE, SMALL, STODOLA, Voss, WENDT, and WODEHOUSE. 408 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Aaassiz, L. Lake Superior: its physical character, a and animals, compared with those of other and similar regions. Frontisp. + x + pp. 10-428 + map + 8 pls. Boston. 1850. [P. 7 E. subulatus Engelm. (= E. aes in a list of vegetation of the northern shor CHARLTON, W. A. Studies in the Alismataceae. I. eas morphology of Echi- nodorus ee Canad. Jour. Bot. 46: 1345-1360. pis. / 1968. n the Alismataceae. V. Experimental ne ene of Ai sea in bea of Echinador tenellus by means of growth inhibitors. /bid. 52: | 1142. pls. 1-3. . Studies in - legaeiieek VII. Disruption of phyllotactic and organogenetic patterns in pseudostolons ei Echinodorus tenellus by means of growth-active sub- stances. [bid. 57: 215-222. 1979. . Studies in the eee VIII. Experimental modification of organogenesis in Ranalisma humile. Ibid. 223-232. 1979. [Results compared with those of similar treatment applied to EF. tenellus.] Corréa, M. P. Diccionario das plantas uteis do Brasil e das exoticas cultivadas. Vol. 2. Frontisp. + xxii + 707 pp. Rio de Janeiro. 1931. [Echinodorus, 205, 209, 214, description, uses, other notes. DURAND _ Prrtier. Catalogue de la flore Vaudoise. Bull. Soc. Bot. Belg. 21: 197- 328. 1882. [Alisma Plantago-aquatica X Echinodorus (Baldellia) ranuncu- loides, 243.] Fassett, N. C. Echinodorus in the American tropics. Rhodora 57: 133-156, 174-188, 202-212. 1955. FERNALD, M. L. Additions to and subtractions from the flora of Virginia. hee 49: 85-115, 121-142, 145-159, 175-194. pls. 1056-1085. 1947. [Pp. 107, 108; E. radicans (Nutt.) Engelm. synonymous es cordifolia L. and oe from E. rostratus Nutt.; Sagittaria planipes, GAGNEPAIN, F. Ranalisma Stapf devient ee Rich. Bull. Soc. Bot. France, V. 5: 274-276. 1929. Guiick, H. Gattungs-Bastarde innerhalb ey Familie der Alismaceen. Beih. Bot. Cen- tralbl. 30(Abt. 2): 124-137. pls. 12, 13. 1913. Grosourpy, D. R. pe. El médico botanico an Vol. 3. 416 pp. Paris. 1864. [Echi- aaa C.L., A. Cronauist, & M. Ownsey. Vascular plants of the Pacific North- t. Part 1. 914 pp. Seattle & London. 1969. [P. 142, a plant from Washington eee eet identified as E. (Baldellia) ranunculoides (L.) Engelm.] sie NIELSEN, L. B. The identity of Alisma boliviana Rusby (Alismataceae). Brittonia : 276— 278. 1979. [E. bolivianus (Rusby) Holm-Nielsen, not £. austroamericanus Rat aj. Kee, A. i sear of swordplants. fees Jour. 31: 230- nae 235, 236. 1960. [Species ae in aquaria, common names, descriptive informatio Knos_iocu, I. W. acumen eee in flowering plants. ee 21: 97-103. 1972. ee 98.] LEONARD, S. W. New records and notes on the flora of the Carolinas. Jour. Elisha Mitchell Sci. Soc. 88: 265, 266. 1972. [E. parvulus (tenellus) in Aiken Co., South Carolina. Lipscoms, B. Echinodorus tenellus var. parvulus (Alismataceae), in the Ozarks of Ar- kansas. Castanea 42: 254, 255. 1977. [Lipscomb 992 (smu), from Baxter Co., Ar- kansas. ] McGreacor, R. L., coordinator, & T. M. BARKLEY, ed. (& several collaborators). Atlas of the flora of the Great Plains. xii + map + 600 pp. Ames, Iowa. 1977. [Echi- nodorus, 416, 573; distribution maps.] Meyer, F. J. Beitrage zur Anatomie pies Alismataceen. Beih. Bot. Centralbl. 49(Abt. 1): 309-368. 1932. [Based entirely on Echinodorus, emphasizing E. macrophyllus (Kunth) 1983] ROGERS, ALISMATACEAE 409 M. Mich., with detailed anatomical observations on the leaves of several species; Meyer’s circumscription of Echinodorus considerably different from Rataj’s recent revision. ee R.S.,&C. J. ae Rare plants of New York State. vii + 96 pp. Albany. 981. [E. ane ee, CC. eae catalogue of plants of Wisconsin and Minnesota. Jn: D. D. Owen et al., Report of a geological survey of bigs Iowa, and Minnesota; and incidentally of a portion of Nebraska Territory. Fro . + 638 pp. + 27 pls. + 21 maps and charts. Philadelphia. 1852. [E. subulatus Fees 619.] PENNA, M. Dicionario Brasileiro de plantas pacer ed. 3. 409 pp. Rio de Janeiro, Sao Paulo. 1946. LE. macrophyllus, 155, several uses.] RAND, E. L. Observations on Echinodorus parvulus. Bidets! 83-85. pl. 45, figs. 4, 5. 1903. [Living plants in Massachusetts. Ratas, K. Echinodorus in nature and the aquarium. (In Czech.) Ziva 21: 88-90, 133- 135, 174-177, 214-217. 1973. 14-16, 49-51, 94-97, 134-136, 176-178, 206, 207. 1974. [Includes photograp . Revizion of the genus Echinodors Rich. Ceskosl. Akad. Véd. 156 pp. Prague. 1975. [The most extensive taxonomic treatment of Echinodorus.] Ripincs, W. H., & F. W. eat Aphanomyces blight of cee sword plants. Phytopathology 63: 289-295. 1973. Rosinson, B. L. The generic position of Echinodorus ee Rhodora 5: 85-89. pl. nodorus was renamed E. parvulus Engelm. This is probably synonymous with Echi- nodorus tenellus (Martius) Buch.] Roic y Mesa, J. T. Plantas medicinales, aromaticos o venenosas de Cuba. Parte 1. 448 pp. Havana. 1945. [E. cordifolius (L.) Griseb. (probably E. Berteroi), 423, 424; common names, description, uses, notes on chemistry, bibliography. ] Rotanpb, F. Etude de ultrastructure des apertures: III, Compléments fournis par le microscope électronique a balayage. Pollen Spores 11: 475-498. 1969. LE. humilis (Ranalisma humile), 480, 482, 485; SEM views of pollen.] ROSENDAHL, C. O., & F. K. Butters. Reputed Minnesota plants which probably do not occur in the state. Minn. Bot. Stud. 4: 461-473. 1916. [P. 467, E. tenellus (Martius) Buch. does not occur, probably Sagittaria misidentified. ] STEWARD, J. H., ed. Handbook of South American Indians. Vol. 1. xix + 624 pls. Washington, D. C. Smithson. Inst. Bur. Am. Ethnol. Bull. 143. 1946. ea ede. 247. THorNnE, R. F. Vascular plants previously unreported a Georgia. Castanea 16: 29- 48. 1951. ms rostratus (Nutt.) Engelm. (£. Berteroi), 3 Torrey, J. Description of the general botanical ee Pp. 59-182, pls. 1-25 in Reports of explorations and surveys, to ascertain the most practicable ‘and econom- ical route for a railroad from the Mississippi River to the Pacific Ocean, Vol. 4, Part 4, Washington. 1857. [Alismataceae, 142, 143. Viuct, P. J. VAN DER. De cellofaanplant. Aquarium 44: 290-295. 1974. [E. Berteroi.]* WEHRHARN, H. R. Die Gartenstauden. Vol. 1. v + 624 pp. Berlin. 1931. [See pp. 6-9.] 3. Sagittaria Linnaeus, Sp. Pl. 993. 1753; Gen. Pl. ed. 5. 429. 1754. Annual or (usually) perennial, often amphibious ae Usually glabrous (S. latifolia var. pubescens (Muhl.) J. G. Sm. and sometimes S. guayvanensis HB pubescent on most organs, other species sometimes a pubescent Alaments), Roots usually conspicuously septate. Most species producing one to many thin axillary stolons, these sometimes branching, sometimes forming a chain of 410 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 bl | ne, YF FiGure 2. Sagittaria. a-m, S. /atifolia: a, flowering plant with stolons, plant mon- oecious, the inflorescence with carpellate flowers below and staminate flowers above, x 4: b, staminate flower, bracts and pedicels of 2 other flowers below, x 2; c, central 1983] ROGERS, ALISMATACEAE 411 plantlets, the terminal bud often distal to an egg-shaped or subcylindrical stor- age organ comprised of 2 or more thickened internodes, this “tuber” sheathe in scale leaves, sometimes forming a plantlet or (in the year it was formed or after overwintering) the bud distal to the tuber growing into a new stolon, this sometimes very short and becoming vertical before terminating in a young plant. Plants usually with short upright stems sheathed (at least at the top) in leaf bases, some species with a horizontal rhizome bearing leaves along its sagittate-hastate, sometimes spongy), floating (then assuming almost the entire range of shapes, commonly elliptic or lanceolate), or growing out of water, then the blade linear to sagittate, usually acute to long-acuminate, sometimes round- ed or obtuse at the apex, with a midrib and lateral nerves, these usually inter- connected by more or less regular, parallel veinlets; petioles often spongy, especially at the broadened bases. Plants usually monoecious, with carpellate flowers toward the base of the scape, sometimes polygamous (then usually with staminate flowers distal to perfect flowers), sometimes dioecious. Scape(s) | to several per plant, erect or lax and floating, characteristically racemes, sometimes once branched at the basal node, on some small plants umbelliform or with only | flower, the main axis with up to 10(-18) nodes, each bearing a (pseu- do-)whorl usually of 3 flowers subtended by usually basally connate, persistent or scarious, (often broadly) ovate to linear bracts. Flowers pedicellate (infre- quently sessile). Petals white [rarely reddish or yellowish], sometimes with a dark spot at the base, larger than the sepals, usually with irregular margins. Nectaries on filaments, staminodes, and carpels, these sometimes modified. Staminate flowers with the sepals usually reflexed, sometimes with abortive carpels, stamens many ina dense cluster on the receptacle, the filaments tending to be dilated at or above the base, the anthers [linear or] narrowly to broadly elliptic-oblong, basifixed or dorsifixed near the base, dehiscence extrorse or ed on an expanded, rounded receptacle, the tapered styles shorter or longer than the ovaries, apical or adaxial; ovules anatropous. Perfect flowers appearing part of staminate flower in vertical oS (note sterile carpels in center), X 3; d, abaxial side of stamen, anther dehiscing, x 6; e, adaxial side of 2 stamens and sterile carpel, x 6: f, vertical section of carpellate Pa immediately after fall of petals, x 3; g, carpel from carpellate flower, x 12; h, anatropous ovule, X 25; 1, fruiting “head,” an aggregate of achenes, x |; j, k, opposite sides of mature achene (note crest on side in “*k,” and resin duct on both sides), X 6; 1, seed removed from achene, oriented as in aie in “*k.”? x 12;m, embryo from soaked seed, oriented as in seed in “1,” X 12. n—-p, S. australis: n, 0, opposite sides of achene, showing crest on both sides and i of resin ducts, X 6; p, seed removed from achene, oriented as in achene in “0,” x 12. q, S. lancifolia: achene with | resin duct, X 12. r, S. graminea: achene with 1 resin duct above crest and another below it, X 12. s, S. Engelmanniana subsp. Engelmanniana: achene with crest and several resin ducts, X 6 412 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 like carpellate flowers but with stamens. Achenes usually strongly compressed, mostly obovate or dolabriform, the margins drawn more or less out into wings, the abaxial margin often broadest and often with an irregular edge (or both edges sometimes irregular), often with facial glands (resin ducts) and ribs, these frequently with irregular, even ornamented edges, the remnant of the style projecting laterally or apically from the apex or from the upper half of the adaxial edge. x = 11. (Including Lophotocarpus T. Durand.) LECTOTYPE SPECIES: Sagittaria sagittifolia L.; see J. K. Small, N. Am. Fl. 17: 51. 1909. (Name from Latin sagitta, arrow, from the shape of some leaves, adopted by Linnaeus from earlier use.) — ARROWHEAD, WAPATO(O), DUCK-POTATO, SAGITTARIA. Approximately 35 species, with subgenus SAGITTARIA distributed throughout the Northern Hemisphere, primarily in North America, from subarctic to sub- tropical and some tropical regions. The most widespread species in the New World, S. /atifolia Willd., ranges from Canada to northern South America. Sagittaria lancifolia L. extends as far south as Brazil, perhaps beyond. Most species of subgenus LopHOTOCARPUS are native to the New World between southeastern Canada and southern South America; S. guayanensis 1s pantrop- ical. Disjunct populations of a number of species occur at great distances from their apparent natural ranges; for example, according to Aston, none of the four species in Australia is native. There are roughly 20 or more species in the United States, approximately three quarters of them in the Southeast. Taxonomic confusion and disagreement make a definitive listing of species impossible. The enumerations below include the species within the range of the Generic Flora that were recognized by Bogin, plus others added in accor- dance with subsequent work. Bibliographic references and comments are pro- vided to facilitate further efforts toward clarification and enumeration of the extensive synonymy. Only the most noteworthy synonyms are given. Subgenus LopHotocarpus (T. Durand) Bogin (carpellate flowers with sepals appressed or spreading, pedicels recurved and thickened in fruit, flowers often perfect) is represented by Sagittaria calycina Engelm. (Beal, 1960), S. guaya- nensis (a weed in rice fields; Thieret, 1969), S. montevidensis Cham. & Schlecht. (adventive), and S. subulata (L.) Buch. (including or not S. Kurziana Glick and S. stagnorum Small). Subgenus SAGITTARIA (carpellate flowers with sepals reflexed, pedicels typ- ically ascending, or recurved but not thickened, flowers rarely perfect) is na resented by S. australis (J. G. Sm.) Small (S. /ongirostra auct. non : see Beal, Hooper, & Rataj), S. brevirostra Mack. & Bush (Beal, Wooten, & Kaul), S. Engelmanniana J.G.Sm., S. falcata Pursh (Beal, 1960), S. fasciculata Beal (1960), S. graminea Michx., S. lancifolia, S. latifolia (S. sagittifolia L. var. longirostra M. Mich.), S. papillosa Buch., S. platyphylla (Engelm.) J. G. Sm. (Wooten, 1973b), S. rigida Pursh (probably in Tennessee), and 5S. secun- difolia Kral. Whether or not S. teres S. Watson and especially S. isoetiformis J. G. Sm. are distinct from S. graminea, and whether S. teres ranges as far south as the Carolinas, are points of disagreement (see Beal, 1960, 1977, es- pecially p. 60; Bogin; Godfrey & Adams; Godfrey & Wooten) Probably only one of the three species originally assigned to Sagittaria by 1983] ROGERS, ALISMATACEAE 413 Linnaeus, S. sagittifolia, belongs to the genus in its modern sense. Even though S. trifolia L. frequently appears in modern publications, the illustration of it that Linnaeus cited either is very inaccurate or represents a plant referable to some other genus (Bogin thought Ranunculus). The 13 species tentatively rec- ognized by Micheli in 1881 reflect both a generally modern circumscription of Sagittaria and trouble with the long-standing question of its relationship with Lophotocarpus. Smith (1895, 1900), due in large part to discovery of new entities, recognized 23 species in North America alone. He differed from Mi- cheli in not including any of the New World taxa within S. sagittifolia, a view upheld in all subsequent revisions. Buchenau (1903) included most of Smith’s species among the 31 in Das Pflanzenreich. Stressing that examination of extensive materials revealed new intermediacy between previously recognized taxa, Bogin (1955) reduced the number of species to 20 (including those added by treating Lophotocarpus as a subgenus of Sagittaria). Rata) (1972a, b) coun- terbalanced the North American emphasis of the preceding studies by revising the species of the Old World, the West Indies, and Central and South America. With a small number of exceptions, his species and subgenera are congruent with those of Bogin. The predominant problem concerning the delimitation of Sagittaria is its relationship with Lophotocarpus (Lophiocarpus Miq.), which appears as a sub- genus of Sagittaria in recent revisions. Lophotocarpus has been thought to differ in being annual and polygamous (vs. without perfect flowers), in having hypogynous stamens (vs. stamens erroneously perceived as inserted above the receptacle in Sagittaria), and in having three (vs. two) pairs of chromosomes with nearly median centromeres (see Smith, 1895, 1900; Baldwin & Speese; and the other papers cited in this paragraph). Mason, who provided a taxonomic history of Lophotocarpus, argued that some distinctions have been inaccurately and unclearly described, that they are of insufficient character to allow un- ambiguous identifications, that conditions ascribed to one genus appear in the other, and that at least one species assigned to Sagittaria based on some of these characters is obviously closer to Lophotocarpus in other regards. Beal (1960) doubted the significance of the cytological difference (which was indeed based on only one species of Lophotocarpus). Evidence from floral vasculature and development and from palynology reinforce merging the genera (Argue, 1976; Kaul, 1967). After Pichon advocated synonymy, Bogin broadened the circumscription of Lophotocarpus and reduced it to a subgenus of Sagittaria, emphasizing the nature of the sepals and pedicels rather than the presence of perfect flowers. (Other distinguishing features of Sagittaria are presented in its comparison with the similar Echinodorus in the treatment of that genus.) Bogin treated problematical complexes as species composed of infraspecific taxa. The largest, centered around Sagittaria graminea, has been studied by Wooten (1970; 197 1a, b; 1973a, b). Drawing data from transplant experiments, experiments to assess environmental plasticity, chromatography, and analysis of edaphic factors, she established three varieties within this species, and with emphasis on edaphic considerations, observations in the field, and crossing studies, she recognized several additional species. (Note that S. secundifolia subsequently described by Kral also appears to be closely related to this as- 414 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 semblage.) Adams and Godfrey, employing observations on living populations and transplant experiments, did not think taxonomic recognition of any of the components of the S. subu/ata complex to be justified but advocated further studies of which Houk undertook one. (Later, Godfrey & Wooten recognized S. stagnorum and S. Kurziana as species aaa from 5. Seas ») Robins experiments, analysis of ecological parameters, and rvations contributed to the decision by Beal, Wooten, & Kaul to elevate sie s sub- species of the ““S. Engelmanniana complex” to species Hybridization occurs, but to an as yet inadequately investigated degree. The plasticity of the plants and related taxonomic confusion make recognition of hybrids difficult. Crossing experiments have shown Sagittaria graminea to be interfertile with four other species (Wooten, 1973b), with other possibilities suspected (see Bogin). On the other hand, Beal ef a/. (1982) tried and failed to demonstrate interfertility among S. australis, S. brevirostra, S. cuneata Shel- don, S. Engelmanniana, and S. latifolia. If a small number of possibly incorrect deviating reports are disregarded, Sagittaria appears to have the uniform chromosome number 2n = 22 (Baldwin & Speese; Beal, 1960; Bloom; Bogin; Brown, 1946; Larsen; Léve & Léve; Oleson; Sharma). The only indication of polyploidy is the ca. 44 count by Bogin for S. subulata var. gracillima(S. Watson) J. G. Sm. Species of subg. SAGITTARIA have one long pair of chromosomes with nearly median centromeres, nine pairs of intermediate length with subterminal centromeres, and one short pair with submedian centromeres (Brown, 1946; Baldwin & Speese: but see Oleson for a somewhat different report). Sagittarias grow in diverse aquatic habitats, commonly in shallow water or on wet banks in or bordering on streams, ponds, swamps, marshes, and ditches, sometimes in tidal areas, sometimes constantly submersed, and sometimes on sites that dry periodically. The degrees of submersion or drying tolerated by different species vary widely and are closely related to the diversity of their foliage. The alismataceous propensities for submersed and juvenile leaves to be straplike phyllodia and for aerial leaves to have expanded blades reach extremes in Sagittaria. Sagittaria (subulata var.) Kurziana grows underwater and has linear leaves, these reportedly sometimes exceeding 3 m in length (said by one collector to be as long as 50 ft). At the other extreme (according to Bogin), S. longiloba Engelm. ex Torrey invariably forms emergent, sagittate leaves. Foliar variation in several species, among them S. /atifolia and S. sagittifolia, spans most of the range in the genus. The relative roles of ecological and genetic control vary among the taxa, a matter in need of continued investigation. Within the Sagittaria graminea complex, Wooten (1970) demonstrated that differences in the forms of emergent leaves and phyllodia between seven pop- ulations (of three varieties and four ecotypes) are attributable more to genetic differences than to adjustment to depth of water. Generalization of Houk’s frey on populations of this complex. They showed great diversity in the forms of phyllodes at certain localities (in some populations ranging between linear 1983] ROGERS, ALISMATACEAE 415 and sagittate), with the shapes obviously related to the water regime and even changing from year to year. (Supplementary references dealing with foliar vari- ation in Sagittaria are Arber, 1920; Gliick, 1905; Gliick & Kirchner; Hroudova; Schanderl; and Sculthorpe.) Although usually described as monoecious, plants of subg. SAGITTARIA are at times entirely staminate or carpellate and often bear perfect flowers. In Sagittaria latifolia, staminate and carpellate flowers are of variable proportions in individual inflorescences, but of nearly equal overall frequency (Schaffner, 1924, 1929). After observing cultivated plants and wild populations, Wooten (1971b) concluded that whether the flowers are staminate or carpellate in this species was not affected by environmental fluctuations to which plants were subjected, and is therefore under genetic control; that monoecious plants are self fertile; that germination of achenes from dioecious plants is especially inhibited: that reproduction in dioecious populations (which sometimes form “unisexual” stands) may be mostly asexual; and that dioecious populations generally seem to be distributed along major rivers, which could convey their propagules. In S. brevirostra, Kaul (1979) found a ratio of more than three staminate flowers per carpellate flower to remain fairly constant over a decade in one lake, despite changes in environmental conditions. In Europe and North America, floral visitation by an array of insects— mostly various flies and short-tongued bees—has been observed (Gltick & Kirchner, Lovell, Robertson, Turner). In Sagittaria subulata completely submersed car- pellate flowers possibly produce achenes (Adams & Godfrey). Winged, sometimes ornamented achenes (as discussed under the family) are not the only disseminules: dispersal is also linked to asexual reproduction. In extensive, probably highly clonal stands large numbers of plants are sometimes connected by stolons (e.g., in the S. subulata complex). Flotation of tubers, rhizomes, and entire plants with buds at the base (described by Lohammar) surely brings about effective relocation. Moreover, as in species of Echinodorus, plantlets form at nodes within inflorescences of S. subulata (Adams & Godfrey; Buchenau, 1903). Multiple proembryos in an embryo sac of Sagittaria graminea were inter- preted by Johri (1936) probably to have resulted from fertilization of synergids by a second pollen tube. A number of species of Sagittaria are cultivated as ornamentals. Doubled forms have been designated S. japonica Hort. (Porterfield, W. Smith). Several species (including S. subulata) producing attractive band-shaped leaves when submersed are grown in aquaria, sometimes under the name S. natans (see especially Wendt). Some serve as green manure (and as bothersome weeds) in rice fields, and as oxygenators in ponds where fish are raised. Preparations involving sagittarias have been attributed with diverse medic- inal benefits, primarily in eastern Asia, but also by North and South American Indians and others. Most commonly mentioned are applications to soothe and cleanse afflictions of the skin. In this connection it is noteworthy that Sharma and colleagues (1975b), during screening of Indian plants over a wide range of biological activities, observed antiinflammatory activity in an extract from S. sagittifolia from which they isolated a new diterpene, sagittariol. Conversely, 416 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 contact with tubers or extracts from some species has caused dermatitis in humans (Mitchell & Rook, Morton) Throughout most of its range Sagittaria is a source of food. Tubers of several species, produced abundantly toward the end of the growing season, are com- monly compared to and prepared like potatoes. Baked, boiled, fried, ground into flour, or cooked in more elaborate fashion, the tubers are highly esteemed, but when raw they are acrid and likely to be toxic (Pammel, Stuart). Sagittaria is cultivated and the tubers marketed in eastern Asia, and it is sold among Chinese foods in the United States. American Indians, who acquired caches from the homes of muskrats and beavers, candied Sagittaria tubers with maple sugar or dried them for long-term storage. Not surprisingly, it has been sug- gested that human activity has historically played a role in the distribution of Sagittaria. (The tubers presumably were transported by ancient peoples, and the achenes from sagittarias growing as weeds probably contaminated rice.) The tubers remain popular among modern enthusiasts for edible wild plants. (Selected extra references concerned with Sagittaria as a source of remedies and nourishment are Fernald & Kinsey; Gibbons; Harrington; Peckolt; Perry; Porterfield; Smith, 1932, 1933; and Winton & Winton.) REFERENCES: Under family references see ARBER (1920), ARGUE (1974, 1976), ASTON, BAILLON, BALDWIN & SPEESE, BEAL (1960, 1977), BLoom, BRown, BUCHENAU (1903), CHARLTON, CHEADLE, Cook, CurRY & ALLEN, DAHLGREN (1934), DE Wit, EICHLER, EwArT, GLUCK (1905, 1936), GLtick & KIRCHNER, GODFREY & Wooten, Haynes, HEGNAUER, HELLQuist & Crow, HEsseE, JOHRI (1934; 1935a, b,c; 1936), KAuL (1967, 1978), KRuUTzscH, Lens & STADLER, Lieu (1979a, b), LOveE & L6ve, MICHELI, MITCHELL & ROOK, NETO- LITZKY, OLESON, PAMMEL, PECKOLT, PERRY, PETELOT, PICHON, PUNT & REUMER, RATAJ (1970, 1978), RickeTT, RIDLEY, ROBERTSON, SATTLER & SINGH, SCHAFFNER (1897), SCHAUMANN, SCHILLING, SCOGGAN, SCULTHORPE, SHARMA, SINGH, SINGH & SATTLER (1973, 1977), SticH, STODOLA, STUART, Swamy, VISSET, WENDT, and WODEHOUSE. ADAMS, P., & R. K. GopFrrey. Observations on the Sagittaria subulata complex. Seog dora 63: 247-266. 1961. [Studies on a aa of the “‘subulata,”’ “‘stagnorum,’ and ‘‘kurziana”’ phases; cf’ GopFREY & WoorTen, Houk.] ANGIER, B. Free for the eating. 191 pp. asasbure Pennsylvania. 1966. [Sagittaria, 32-135.] ARGUE, C. L. Pollen of the Alismataceae and Butomaceae. Development of the nexine in Sagittaria lancifolia L. Pollen Spores 14: 5-16. 1972. [Includes TEM micrographs and comparison with eas, BEAL, E. O.,S.S. Hooper, & K. Ratay. Misapplication of the name Sagittaria longirostra (Micheli) J. Smith (Alismatacear) to S. australis (J. G. Smith) Small. Kew Bull. 35: 369-37 JJ. W. i & R.B. KAuL. Review of the Sagittaria Engelmanniana complex (Alismataceae) with environmental correlations. Syst. Bot. 7: 417-432. 1982. [Dis- oe tion maps, environmental parameters related to distributions, crossing exper- ments; S. see S. brevirostra, S. Engelmanniana, S. latifolia, S. cuneata;, see me BEAL et al., 1980. BLoepeL, C. A., & A. M. Hirscu. Developmental studies of the leaves of Sagittaria latifolia and their relationship to the leaf-base theory of monocotyledonous leaf morphology. Canad. Jour. Bot. 57: 420-434. 1979. [Lamina and petiole of mature 1983] ROGERS, ALISMATACEAE 417 leaf derived from primordial upper leaf zone, not leaf-base region, cf. discussion of phyllode theory in ArBerR, 1920. Boain, C. Revision of the genus Sagittaria (Alismataceae). Mem. N. Y. Bot. Gard. 9: 179- 233. 1955. [Comparison with other genera, historical survey, cytology, clado- gram, uses, distribution maps. Brown, W. V. A note on Sagittaria eit Rhodora 44; 211-213. 1942. [Achenes illustrated; S. Kurziana X S. stagno CHADHA, Y. R., chief ed. The wealth oe India . Vol. 9. Raw materials. Frontisp. + XXXviii + 472 + xiv pp. New Delhi. 1972. [Sagittaria, 167, 168. CLAUSEN, R. T. Sagittaria eis anew species from the New Jersey a barrens. Rhodora 39: 29-31. pl. 454. 7. [= S. graminea, according to Bogi . The variations of nies subulata Torreya 41: 161, 162. 1941. Cruz, A. A. DE LA, & W. E. Por. Amino acid content of marsh plants. Estuarine and Coastal Raa Sci. 3: 243-246. 1975. [Tabulation of amino acids found in S. graminea and S. lancifolia in Mississippi.] FARNSWorTH, N. R., R. N. Blomster, M. W. Quimsy, & J. W. SCHERMERHO The Lynn index. A bibliography of te FT V2 AS wor les: eat f. sinensis; list of sugars, asparagine, unidentified alkaloid, phytosterol, phytosterolin, and fatty acids. Fassett, N. C. Lophotocarpus on the north-eastern river-estuaries. Rhodora 24: 71-73. pl. 137. 1922. [L. calycinus var. spongiosus. FERNALD, M. L., & A. C. Kinsey. Edible wild plants of eastern North America. xiv + 452 pp. Cornwall-on-Hudson, New York. 1943. [Sagittaria, 86-89. FERREN, W. R., Jr. Range extensions of Sagittaria montevidensis in me Delaware River system. Bartonia 42: 1-4. 1974. [Notes on localities and habita GarteEN, C. T., & D. Paine. A mult ivariate analysis of factors ire radiocesium uptake by Sagittaria latifolia in coastal plain environments. Jour. Environ. Qual. 6: 78-82. 1977.* Gispons, E. Stalking the wild asparagus. Field guide ed. x + 303 pp. New York. 1970. [Sagittaria, 21-24.] Griick, H. A new Sagittaria from Florida: Sagittaria ee Bull. Torrey Bot. Club 54: 257-261. 1927. [S. subulata var. Kurziana (Glick Goprrey, R. K., & P. Apams. The identity of fore soetformis (Alismataceae). Sida 1: 269-273. 1964. [S. teres, S. isoetiformis, S. graminea all distinct species; suggests that Beal’s reports of S. teres in the Carolinas reat from misidentification of S. isoetiformis (cf. BEAL, 1977).] GovINDARAJALU, E. Further contribution to the anatomy of the Alismataceae: Sagittaria ensis H. B. K. ssp. lappula (D. Don) Bogin. Proc. Indian Acad. Sci. B. 65 142-152. 1967. [Anatomy of leaves, petioles, peduncles, and roots; includes illus- trations of hydropoten and vessel elements.] HARRINGTON, H. D. Edible native plants of the Rocky Mountains. Map + 392 pp. ee 1967. [S. latifolia, 206-210.] Houk, R. D. experimental taxonomic study of the Sagittaria subulata complex. Diss. Abstr. B. 27: 1048. 1966. [Distinguished 3 genecodemes. ] Hroupova, Z. Occurrence of Sagittaria sagittifolia at different depths of water. Folia Geobot. Phytotax. 15: 415-419. 1980. [Visited 100 localities; proportions of leaf types plotted against depth of water.] Kau, R. B. Inflorescence architecture and flower sex ratios in Sagittaria brevirostra (Alismataceae). Am. Jour. Bot. 66: 1062-1066. 1979. [Pollination probably geito- nogamous and xenogamous.] Krat, R. A new phyllodial-leaved oe (Alismaceae) from Alabama. Brittonia 34: 12-17. 1982. [S. secundifolia, sp. nov.; includes key to o “graminea” type sagittarias. | Kuzniewskl, E. Sagittaria L. genus in the light of the “Wroclaw Taxonomy.” (Translated from Polish by H. M. Massey.) Available from U. S. Dept. of Commerce, Clear- 418 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 inghouse for Federal Scientific and Technical Information. Springfield, Virginia. Originally published in Acta Soc. Bot. Polon. 25: 275-284. 1956. [Quantitative dendritic method applied to Sagittaria, results compared with subgeneric classifi- cation by Buchenau (1903); suggestion that Sagittaria originated in North America and migrated to the Old World across both the Atlantic and Pacific oceans. ] Larss_e, A. M. The use of root characteristics to separate various ribbon-leaved species of Sagittaria from species of Vallisneria. Turtox News 31: 224, 225. 1953. [Sagittaria with branched roots; see also THIERET, 1971.] Larsen, K. Studies in the flora of Thailand. 14. Dansk Bot. Ark. 20: 205-275. 1963. [Sagittaria, 248. Leavitt, R. G. Reversions in Berberis and Sagittaria. Rhodora 2: 149-155. pl. 19. 1900. [S. montevidensis, illustration of seedling. LOHAMMAR,G. Sagittaria natans X sagittifolia. Sv. Bot. Tidskr. 67: 1-4. 1973. [Forming large colonies in Lapland; overwintering buds sometimes not set off from plant by sy Love.L, J. H. Three fluvial flowers and their visitors. Asa Gray Bull. 6: 60-65. 1898. [S. latifolia; re described and visitors listed— mostly Syrphidae, some beetles, bees, additional dipterans. ] MACcKENzIE, K. K. Notes on Sagittaria. Torreya 9: 30-32. 1909. [Found (probably) S. ther.] Mason, H. L. The status of Lophotocarpus in western North America. Madrofio 11: 263-270. 1952. MatsuTAnl, H., 8. Kusumoro, R. Korzumt, & T. Sapa. N(carboxymethyl)-L- in the acid idole of Sagittaria pygmaea. Phytochemistry 18: 661, 662. 1979 [Also found diamenopimelic acid and ‘“‘usual amino acids.” Meyer, F. J. Beitrége zur Anatomie der Alismataceen. III und IV. Die Blattanatomie von Lophotocarpus und Limnophyton. Beih. Bot. Centralbl. 52(B): 96-111. 1934. [Studied Lophotocarpus guayanensis and L. Seubertianus.] Morton, J. F. Atlas of medicinal plants of Middle America. Bahamas to Yucatan. xxvill + 1420 pp. Springfield, Illinois. 1981. [S. /ancifolia, 22.] Mounrter, J. Beobachtungen tiber besondere Eigenthiimlichkeiten in der Fortpflanzung- weise der Pflanzen durch Knospen. 3. Uber die Knospen der Sagittaria sagittaefolia L. Bot. Zeit. 3: 689-697. 1845. Nocte, E. F. Botanische Bemerkungen tiber Stratiotes und Sagittaria. 44 pp. 2 pls. Copenhagen. 1825. [S. sagittifolia, detailed description and illustrations of under- ground organs; reported tubers on runners in “‘A/isma Plantago,” but this clearly based on misidentified material (see comments in MUNTER).] PALIWAL, S. C., & G. S. LAvANnia. Epidermal structure and distribution of stomata in Sagittaria guayanensis H. B. & K. Curr. Sci. Bangalore 47: 553-555. 1978. [Includes illustration of hydropoten; stomata sometimes on both surfaces of floating ete q PORTERFIELD, W. M. The arrowhead as a food among the Chinese. Jour. N. Y. Gard. 41: 45-47. 1940. [Descriptions of tubers and plants, S. sinensis, S. Lae references to horticultural information on doubled forms, see also W. Smitu.] PRANCE, G. T. New dasa : phanerogams from Amazonia. Brittonia 23: 438-445. 1971. [S. purusana, sp. nov.] Ratal, K. Revision of the genus Sagittaria. Part I. (Old World species). Annot. Zool. Bot. Slov. Narod. Muz. Bratislava 76. 31 pp. 6 pls. 1972a. [Includes brief comparison of Sagittaria with other genera . Revision of the genus Sane Part II. (The species of West Indies, Central and South America). /bid. 78.61 pp. 1972b. [Includes 2 new species, but see RATAJ, 1978.] Rustsov, N. I. Sagittaria platyphylla. A new alien species in the flora of the European part of the U.S.S.R. (In Russian.) Bot. Zhur. 60: 387, 388. 1975 1983] ROGERS, ALISMATACEAE 419 SCHAFFNER, J. H. Expression of the sexual state in Sagittaria latifolia. Bull. Torrey Bot. Club 51: 103-112. 1924. Fluctuation of the point of sex reversal in Sagittaria latifolia. Am. Jour. 16: 191-195. 1929. [Sampled over 1000 inflorescences from several poculations see also WooTEN, 1971b. SCHANDERL, H. Sagittaria le ee als Kompasspflanze. Planta 7: 113-117. 1929. (Uncrowded leaves in sunny places orient in accordance with intensity of illumi- n SERBANESCU-JITARIU, G. Untersuchungen iiber das Gynézeum, die Frucht und die Kei- mung der Samen von Sagittaria sagittifolia L. An. Univ. Bucuresti Biol. Veg. 22: 17-24. 1973. [Describes carpels in a spiral arrangement; cf. SINGH & SATTLER, 1973.] SEVERIN, C. F. Origin and structure of the secondary roots of Sagittaria. Bot. Gaz. 93: 93-99. 1932. [Studied S. /atifolia.] SHARM C., Y. N. SHUKLA, & J. S. TANDON. Constituents of Colocasia formicata, niveum, Paspalum scrobiculatum, Mundulea sericea and Duabanga sonneratiodes. Phytochemistry 11: 2621-2623. 1972. [Found hentriacontanone, sitosterol in Sag- ittaria.] ——. Alkaloids and terpenoids of Ancistrocladus heyneanus, Sag- ittaria ee Leni formosa and Hedychium spicatum. Ibid. 14: 578, 579. ne : on, & M. M. Duar. Sagittariol: a new diterpene from Sagittaria Bee Phytochemistry 14: 1055-1057. 1975b. ee of structure. ] SMALL, J. K. Sagittaria latifolia. Addisonia 2: 27, 28. pl. 54. 1917 SmitH, H. H. Ethnobotany of the Ojibwe Indians. Bull. Milwaukee Public Mus. 4: 327- 525. pls. 46-77. 1932. [Sagittaria, 353, 396.] Ethnobotany of the forest Potawatomi Indians. /bid. 7: 1-230. pls. 1-38. 1933. [Sagittaria, 37, 94, 95.] Smit, J.G. A revision of the North American species of Sagittaria and Lophotocarpus. Annual Rep. Missouri Bot. Gard. 6: 27-64. pls. /-29. 1895. [Reprint issued in 1894 with different pagination; includes defense of retaining oer as a distinct genus and a list of species that produce tubers and rhizom _ Revision of the species of Lophotocarpus of the United States: and description of a new species of Sagittaria. Ibid. 11: 145-151. 1900. [Reprint issued in 1899 with different pagination; a of these species of Lophotocarpus treated as synonyms of S. montevidensis by Bogin.] SmitH, W. Illustration of Sagittaria japonica. Gard. Chron. III. 30: 170, 171. 1901.[A ed form; see also una Su, K. L., Y. ABut-Hays, & E. J. SraBa. Antimicrobial effects of aquatic plants from Minnesota. Lloydia 36: 80-87. 1973a. [Includes S. cuneata, S. latifolia.] Sapa. Toxicity, anti-neoplastic, and coagulation effects of aquatic plants from Minnesota. Lloydia 36: 99-102. 1973b. _ ApuL-Hagys. Preliminary chemical studies of aquatic plants from Minnesota. Lloydia 36: 72-79. 1973c. [Tested for alkaloids, flavonoids, steroids, and lipids; S. latifolia, S. cuneata, with flavonones, flavonols, saponins, unusually high eee content. Tuieret, J. W. Sagittaria guayanensis (Alismaceae) in Louisiana: new to the United States. Sida 3: 445. 1969. Observations on some aquatic plants in northwestern Minnesota. Mich. Bot 10: 117-124. 1971. [Vegetative comparison of Sagittaria with Vallisneria, see also Turner, C. E. Reproductive biology of Sagittaria montevidensis Cham. & Schlecht. ssp. calycina (Engelm.) Bogin (Alismataceae). Diss. Abstr. B. 43: 1358, 1359. 1982. [Studied populations in California—plants self-compatible but apparently strongly 420 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 outcrossed, agamospermy lacking, ee a of perfect flowers not increased with increase in nutrients; a weed in rice Ss. WILSON, P. Sagittaria lancifolia. Addisonia 12: 43. pi. 406. 1927. Winton, A. L., & K. B. Winton. The structure and composition of foods. Vol. 2. xiv + 904 pp. New York & London. 1935. [Sagittaria, 120-122.] Wooten, J. W. Experimental investigations of the Sagittaria graminea complex: trans- plant studies and genecology. Jour. Ecol. 58: 233-242. 1970. [3 varieties. ] . Chromatographic studies in the Sagittaria graminea complex. Canad. Jour. Bot. 49: 1793-1797. 1971a. [Patterns of phenolic compounds from submersed leaves differ from patterns from emersed leaves; interpopulational variation correlated with previously i aor ecotypes.] The monoecious and dioecious conditions in Sagittaria latifolia L. (Alisma- taceae). Evolution 25: 549-553. 1971b. [Cf SCHAFFNER, 1924, 1929. . Edaphic factors in species and ecotype differentiation of Sagittaria. Jour. Ecol. 61: 151-156. 1973a. [Sampled soils for several compounds and found significant differences between localities for S. platyphylla, S. cristata, and 3 vars. of S. gra- minea Taxonomy of seven species of Sagittaria from eastern North America. Brittonia 25: 64-74. 1973b. C. E. Lamotte. Effects of photoperiod, light intensity, and stage of devel- opment on flower i a in Sagittaria graminea Michx. (Alismataceae). Aquatic Bot. 4: 245-255. ARNOLD ARBORETUM HARVARD reel 22 Diviniry AVEN CAMBRIDGE, eee HUSETTS 02138 1983] KRAL, XYRIDACEAE 42] THE XYRIDACEAE IN THE SOUTHEASTERN UNITED STATES! ROBERT KRAL , aeaicesie Agardh, Aphor. Bot. 158. 1823, rideae,”” nom. cons. ses etes GRASS FAMILY) Annual or perennial, stemless to caulescent, usually rosulate, scapose ter- restrial herbs of high-hydroperiod soils [rarely aquatic]. Roots mostly slender, iffuse-fibrous, with root hairs. Axis sympodial [or monopodial]. Leaves al- ternate, distichous or spiral, ligulate or eligulate, the bases broad, open-sheath- ing, frequently equitant and keeled, the blades laterally [to dorsiventrally] com- pressed, usually flattened, often terete or ventrally sulcate, the indument usually of uniseriate-glandul I hed [to branched] trichomes [or absent]. Inflorescence(s) lateral [to a the scapes | to few, arising from axils of scape sheaths [or inner leaves], naked [to short-bracteate], each bearing apically 1 [or more] imbricate-bracted spikes or heads [or a paniculate arrangement of same]. Flowers perfect, 1 to many, solitary and subsessile to pedicellate in the axils of chaffy, leathery, or scarious bracts. Perianth of 2 differentiated whorls. Sepals 3, the anterior (inner) one usually membranaceous and wrapped arcund the corolla, abscising as the flower opens, the other 2 subopposite, connivent [to basally connate], chaffy, boat shaped, usually keeled, clasping the mature capsule. Petals 3, equal [to unequal], distinct [to united and salverform], strong- ly clawed, the spreading blades broad, yellow to white [or blue]. Stamens usually 3, epipetalous. Staminodia 3, distinct, clawed as in petals and distally 2-armed, also moniliform-hairy or reduced [or absent]. Anthers 4-sporangiate, bilocular at anthesis, introrsely or laterally dehiscent, dehiscing longitudinally; pollen monosulcate or inaperturate. Gynoecium 3-carpellate, the ovary |-locular [to incompletely 3-locular], the placentation marginal or parietal [basal, free-cen- tral, or axile]; style terminal, tubular, apically 3-branched; stigmas 3, truncated, glandular-hairy; ovules mostly numerous, anatropous, 2-integumented. Fruit ‘Prepared for the Generic Flora of the Southeastern United States, a project of the Arnold Ar- boretum currently made possible through the support of the National Science Foundation, under Grant DEB-81-11520 (Carroll E. Wood, Jr., principal investigator). This treatment, the 99th in the series, follows the format established in the first paper (Jour. Arnold Arb. 39: 296-346. 1958) and continued to the present. The area covered by the Generic Flora includes North and South Carolina, Georgia, Florida, Tennessee, Alabama, Mississippi, pUANSAS, aad Loursiaia: The descriptions are based primarily on the plants of this area, with informatio | members of a family or genus in brackets [ ]. References that I have not vented are marked with an asterisk. The illustration was drawn by Karen S. Velmure from dissections by K. R. Robertson and C. E. Wood, Jr., of material ee by Norton G. Miller in North Carolina. © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 421-429. July, 1983. 422 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 capsular, mostly loculicidal; seeds usually numerous, mostly under 2 mm long, with strong longitudinal ridges and finer cross lines, translucent or farinose- opaque, the embryo small, situated at base of an abundant mealy endosperm. Type GENUS: Xyris L. A pantropical family of four genera and nearly 300 species, most of them in Xyris, the only genus to range into the North Temperate Zone. The remaining genera (Orectanthe Maguire, Achlyphila Maguire & Wurdack, and Abolboda Humb. & Bonpl.) are small and are confined to northern South America. The family is mainly distinguished by its combination of rosulate and scapose habit; it is also characterized by vessels with simple perforation plates in all vegetative organs; leaves with open sheaths and with narrow blades having parallel or uninerved venation; flowers solitary in axils of (usually) chaffy, imbricate bracts in conelike spicate or capitate inflorescences; perianth tri- merous, heterochlamydeous, the sepals with the inside (anterior) member fu- gacious, membranaceous, and covering the rest of the flower in bud; androe- cium of 3 epipetalous stamens with anthers opening longitudinally and (usu- ally) 3 staminodia; gynoecium tricarpellate, fruit capsular, typically loculicidal; and seeds small, with copious farinaceous endosperm. Engler’s placement of the Xyridaceae in the large and artificial order Fari- nosae together with 12 other families showing mealy endosperm and a usually compound superior ovary met with early opposition from many phylogenists. However, there seems to be general agreement in most recent studies that the family is closely related to the Rapateaceae, and both are placed either in the order Xyridales (Hutchinson, 1973) or, more popularly, in the Commelinales (Takhtajan, 1980; Cronquist, 1981; Thorne). The four genera of Xyridaceae break into two groups (Abolboda and Orec- tanthe vs. Achlyphila and Xyris) on the basis of foliar and pollen characters. Abolboda and Orectanthe have spinose pollen, appendaged styles, and polys- tichous leaves with dorsiventral structure and commonly with a hypodermis of colorless cells under both surfaces. Abolboda (ca. 20 species) has an interior sepal that is reduced so that often only the lateral sepals are evident; a corolla that is regular and usually blue; staminodia that, if present, are simple and unbranched; and either styles that are apically three-branched or stigmas that are strongly hairy and trilobed. Orectanthe (O. ptaritepuiana (Steyerm.) Ma- guire and O. sceptrum (Oliver) Maguire) has three sepals, but the corolla is irregular and yellow, usually strongly curved outward with the interior lobe enfolding the other two lobes in bud; staminodia are lacking; and the style is simple, with the stigma subcapitate. Achlyphila and Xyris have obviously or obscurely distichous leaves that are usually laterally compressed and lacking the colorless hypodermal layers. Their pollen lacks spines, their styles are unappendaged, and their corollas are regular. The monotypic Achlyphila (A. disticha Maguire & Wurdack) has an elongate, creeping, scaly rhizome from which arise erect, distichously leafy stems; it lacks staminodia and its styles are undivided. Members of Xyris are mostly short stemmed, have an outer sepal much different from the inner two, produce apically bipartite staminodia (in most), and have styles prominently three- 1983] KRAL, XYRIDACEAE 423 branched above the middle. It is the only genus of the four that produces hairs on the leaves. Tomlinson believes that, while 4bo/boda and Orectanthe are evidently closely related, Achlyphila may be a link between Abolboda and Xyris, thus making it more difficult to consider Abolbodaceae a distinct family, as did Nakai. Before the discovery of Achlyphila, Nakai’s lead was temporarily followed by Takh- tajan (1959) The family is of little economic importance. Some species of Xyris are an important food for the wild turkey (Meleagris gallopavo) in the southeastern United States, some others are occasionally used as aquarium plants, and a few species with long scapes and conspicuous spikes are harvested for use in dried plant displays. REFERENCES: ArBerR, A. Leaves of the Farinosae. Bot. Gaz. 74: 80-94. 1922. [Xyridaceae, 83, 84, pi. Ly, ) I study. Frontisp. + 258 pp. Cambridge. 1925. o ahead 7 89, os ae BAILLon, H. Xyridacées. Hist. Pl. 13: 224-227. 1894. BENTHAM, G., & J. D. HooKER. Xyridaceae. Gen. Pl. 3: 841-843. 1883. CarLouisT, S. Anatomy of Guayana Xyridaceae: Abolboda, Orectanthe, and Achlyphila. Mem. N. Y. Bot. Gard. 10(2): 65-117. 1960. CASTELLANOS, A. Xyridaceae. Jn: H. R. Descoe, ed., Genera et species plantarum Argentinarum. Vol. 3. 383 pp. /33 pls. Buenos Kanes. 1945. [Xyridaceae, 41-66, pls. 6-13, maps; useful information on distributions. ] Cook, C. D. K., B. J. Gut, E. M. Rix, J. SCHNELLER, & M. Seitz. Water plants of the world. vil + 561 pp. The Hague. 1974. [Includes Abolboda, Xyris.] CronquisT, A. The evolution and classification of flowering plants. x + 384 pp. Boston. 1968. [Xyridaceae, 333, 334.] An integrated system of classification of flowering plants. xviii + 1262 pp. New York. 1981. [Commelinales, 1107-1115.] ENDLICHER, S. Genera plantarum. Ix + 1483 pp. Vienna. 1836-1840. [Xyrideae, in- cluding Mayaca, 68-70.] ENGLER, A. Xyridaceae. Nat. Pflanzenfam. I]. 4: 18-20. 1887. ErRpDTMAN, G. Pollen morphology and plant taxonomy, angiosperms. xii + 539 pp. Wal- tham, ees 1952. [Xyridaceae, 453, 454; case made for distinguishing Abolbodacea : Handbook of palynology. 486 pp. New York. 1969. [Xyridaceae, 89.] elteres tiber Merkmalsbestand und Verwandtschaftsbezichungen der “Fari- nosae.”” Ibid. 3: 169-207. 1962. Commelinales. 7a: H. Mevccuior, A. Engler’s Syllabus der Pflanzenfamilien. ed. 12. 2: 549-561. 1964. [Xyridaceae, 552, 553.] Hoes J. The families of flowering plants. ed. 3. xvili + pp. Oxford. 1973. [Xyridaceae, 705, treated as part of Xyridales with Rapateacea Iproso, J. M. Xiridaceas de Colombia. Caldasia 6(29): 184— 260. 1954. [Illustrated taxonomy of Colombian Xyris.] Kunth, C. S. Xyridaceae. Enum. Pl. 4: 1-29. 1843. [Several new species treated, in- cluding XY. ambigua Beyr.] Lanjouw, J. Xyridaceae. Im: A. PULLE, ed., Fl. Suriname 1(1): 225-248. 1938. 424 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 MacuirE, B. Xyridaceae. Jn: B. MAGuire, J. J. WURDACK, & COLLABORATORS, Botany of the Guayana Highland—part HI. Mem. N. Y. Bot. Gard. 10(1): I- 19. 1958. [Orectanthe described; 17 species of Abolboda treated.] & L.B.Smitu. Xyridaceae. Jn: B. MAGuIRE, J. J. WURDACK, & COLLABORATORS, Botany of the Guayana Highland—part V. /bid. 10(5): 1 37. 1964. [A good treat- ment, by key (72 species of Xyris); illustrations of Venezuelan and ena species. ] & J. J. Wurpack. Xyridaceae. In: B. MAGuire, J. J. WurRDAcK, & COLLA- RATORS, Botany of the Guayana Highland—part WV. Ibid. 10(2): 11-15. 1960. Lichlyphila described; key to the genera of Xyridaceae.] Mate, G. O. A. Beitrage zur Anatomie der Xyridazeen. Sv. Bot. Tidskr. 3: 196-209. 1909. —. Xyridologische Beitrage. Ark. Bot. 19(13): 1-8. figs. 1-3. 1924. [Xyris and Abolboda.] ———. Xyridaceae. Nat. Pflanzenfam. ed. 2. 15a: 35-38. figs. 1/-15. 1930. Martius, C. F. P. von. Xyridaceae. Herb. = Brasil. 24(2): 56, 58. 1841. Nakal, T. Ordines, familiae, tribi, genera, sectiones, species, varietates, formae et com- binationes novae a Prof. Nakai-Takenoshin adhuc et novis edita. vil + 256 pp. Tokyo. 1943. [Abolbodaceae, Xyridaceae, 221. NILSSON, i pundien iiber die Xyrideen. Sv. Vet.-Akad. Handl. 24(14): 1-74. 1892. natomische Eigenthiimlichkeiten der Gattung Yyris. Bot. Centralbl. 53: 347, 348. 1893. Pita, S. K., & A. Pittat. Root apical organisation in monocotyledons— Xyridaceae. Proc. Indian Acad. Sci. B. 54: 234-240. 1961. Seubert, M. Xyrideae. Jn: C. F. P. von Martius, Fl. Brasil. 3(1): 209-224, 231, 232. pls. 22-30. 1855. [Xyris and Abolbo da. Smitn, L. B., & R. J. Downs. Xyridaceae from Brazil. Jour. Washington Acad. Sci. 44: 3 —. Xyridaceae from Brazil—II. Proc. Biol. Soc. Washington 73: 245- 260. 1960. & _ Xiridaceas. /n: P. R. Reitz, ed., Fl. illus. Catarinense. Parte I, Fasc. XIRI. 54 pp. 1965. [17 species.] & Xyridaceae. /n: F. C. HOEHNE, = Brasilica 9(2): 1-215. 1968. [The definitive study on South American Xyridacea SoLEREDER, H., & F. J. Meyer. Xyridaceae. Syst. i. at. Monokot. 4: 36-50. STAUDERMANN, W. Die Haare der Monokotylen. Bot. Arch. 8: 105-184. on ods daceae, 124, 125, pl. 9, fig. 7 aes K., & R. BEvERIE. Uber die Xyridaceengattung Abolboda Humb. et 1. Bot. Jahrb. 67: 132-142. 1935. Se A. Die Evolution der Angiospermen. vii + 344 pp. Jena. 1959. [Superorder Farinosae, order Commelinales, 272, 273. Outline of the classification of flowering plants (Magnoliophyta). Bot. Rev. 46: 295. 359. 1980. [Superorder V. Commelinanae, order 11. Commelinales (including Rapateaceae, Xyridaceae (including een r ees Commelinaceae, Mayacaceae, 316, 317.] Tuorne, R. F. A phylogenetic classification of the Angiospermae. Evol. Biol. 9: 35— 106. 1967. [Commelinales, 100, 101. TomLinson, P. B. Commelinales—Zingiberales. Jn: C. R. METCALFE, ed., Anato monocotyledons. Vol. 3. xx + 446 pp. Oxford. 1969. [Xyridaceae, 92- 27, Excellent overview of anatomy of family with suggestions as to taxonomic application. ]} 1. Xyris Linnaeus, Sp. Pl. 1: 41. 1753; Gen. Pl. ed. 5. 25. 1754. Annual or perennial, short- [to prominently] caulescent, usually rosulate, scapose herbs with diffuse fibrous roots, usually perennating by lateral offshoot 1983] KRAL, XYRIDACEAE 425 Si fe? to ee i Figure 1. Xyris.a—n, X. fimbriata: a, habit, x Ys; b, inflorescence, with | open flower and marcescent corollas of earlier flowers, x 2; c, inner (anterior) sepal, < 6; d, flower from above, showing the 2 d the subtending bract below (behind flower), < 5; e, petal with filament of fertile. anther adnate to claw, and staminodium with free filament and much-branched tip, x 6; f, anther before dehiscence, x 6; g, tip adaxial side of 1 placenta with orthotropous ovules, * 12; m, dehiscing capsule, x 6; n, seed, X 40. 0, p, X. brevifolia: 0, habit, x 2; p, seed, x 40 426 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 buds. Axis sympodial. Foliage externally smooth to variously papillose or rough [to pilose], often bearing uniseriate secretory trichomes on inner surfaces of leaf sheaths. Outer leaves often scalelike; principal leaves mainly linear, equi- tant, distichous, the sheathing bases open, usually strongly clasping and dilated distally, narrowing gradually or abruptly to junction with blade, there ligulate or eligulate, the sheath edges convergent to form one side, the midrib area the other, of a laterally flattened to terete, angulate or sulcate, usually linear, entire or scabrid [or ciliate] blade. Scapes single or few to a shoot, linear, terete to variously compressed, costate or ancipital, mostly overtopping leaves, each subtended by a sheath leaf that is closed and involute-tubular proximally, open and variously bladed [or bladeless] distally. Inflorescence a terminal, conelike head or spike of tightly to loosely spirally or distichously imbricate chaffy bracts, the lowest usually sterile and smaller than the fertile [less frequently, larger than the fertile and with cusps or blades overtopping the inflorescence], the fertile each with a subsessile axillary flower, sometimes the uppermost again barren. Sepals unequal, the 2 lateral opposing, connivent around the floral base, distinct (or variously connate), chaffy, boat shaped, variously keeled, equilateral or not, the anterior (inner) sepal scarious and enclosing the corolla in bud, fugacious. Petals free [to basally connate], subequal, long clawed, the blades usually broad, yellow [or orange], rarely white, spreading. Stamens equal, the short, fleshy filaments departing just above the petal claw, the anthers basifixed, 2-locular, the pollen monosulcate; staminodia 3, subequal, mostly 2-armed apically, the branches penicillate, with moniliform hairs [rarely glabrous or vestigial]. Ovary 1-locular [to imperfectly 3-locular]; ovules several to man the placentae marginal or parietal [axile, basal, free-central]; style terminal, slender, tubular, strongly 3-branched above the middle, each branch termi- nating in a truncate, U-shaped or funnelform stigma fringed with glandular hairs. Capsule usually thin walled, 3-valved, dehiscing along 3 lines alternate with placentae. Seeds small, translucent or farinose, with copious mealy, starchy, proteinaceous endosperm, the embryo small, basal-lateral, the seed coat mostly longitudinally ribbed and cross lined. (Kotsjiletti Adanson, 1763; Xuris Adan- son, 1763; Xyroides Thouars, 1806; Jupica Raf., 1836; Ramotha Raf., 1836: Schizmaxon Steudel, 1856.) Type species: Xyris indica L. (Name Greek, in reference to a plant with two-edged leaves, from xyron, a razor.) — YELLOW-EYED GRASS. There are three sections: PoMATOXxyYRIS Endl., with possibly 20 species, 1s characterized by axile placentation and is confined to Australia; Xyris (sect. Euxyris Endl.), with ca. 100 species, is distinguished by marginal or parietal placentation and 1s predominantly eae ae and North American; NEMATOPUS Seub., by far the largest section and m lly the most diverse, is char- acterized by basal or free-central Seay and ‘centers in South America. No species are known from Europe, continental Asia, or (except for two species— one North American, one Asian—naturalized in Hawaii) the central or northern Pacific islands. In the most recent treatment of North American Xyris (Kral, 1966), 19 species and three varieties were recorded for continental North America north 1983] KRAL, XYRIDACEAE 427 of Mexico. One new species has been described since (Kral, 1978), bringing the total to 23 taxa, all of them indigenous and all confined in North America to the eastern United States and Canada. Most are limited to the Coastal Plain, with all but one either restricted to or occurring in the southeastern United States. Only three taxa grow in the wetlands of the Canadian Shield, and of these only YX. montana Ries is confined to glaciated portions of the continent. Ten species are restricted to the southeastern United States, some as narrow endemics. Two are weeds of New World wetlands; seven center in the lower Coastal Plain terraces of Florida but also occur in the Caribbean (primarily western Cuba) and the pinelands or wet savannas of Belize, Honduras, Mexico, and Nicaragua. Fifteen additional species (five in sect. NEMATOPUS, ten in sect. Xyris) are found in Mexico, Central America, and the Caribbean, with the majority endemic. Gustaf O. A. Malme, European student of the Xyridaceae and in the early part of the century the recognized authority on Xyris, was the last to attempt formal subdivision of sect. Xyris for North America (1937b). He delimited seven subsections, three monotypic and only one (Mexicanae) not represented north of Mexico. His artificial classification was based primarily upon char- acters of the lateral sepals (keels scabrid, ciliate, fimbriate, or villous, vs. keels entire), and secondarily on vegetative characters (e.g., leaf width, leaf length vs. scape-sheath length, degree of fleshiness of leaf-sheath bases). In contrast, the study of Brazilian xyrids by L. B. Smith and R. J. Downs (1968) pays far more attention to vegetative morphology—particularly foliar anatomy and inflorescence characters. Relatively little has been done with the cytology of North American Xyris beyond some analyses of microsporocytes (Lewis, 1961; Kral, 1966). Twenty taxa have been so studied, and in all the counts have consistently been n = 9. A few root-tip studies (Kral, not yet published) have shown some genomal distinctness. The flowers of Xyris are ephemeral, as is often the case in the Commelinales, with corollas normally not expanded for more than a few hours. Even in the southeastern United States, where several species may cohabit within a small area of wetlands, there may be wide differences between species or species complexes as to time of flowering within a 24-hour period. No nectaries have been found in any species in the genus, and few insects have been observed on the flowers. Only pollen-gathering bees (Bombidae, Andrenidae) have been noted, and to judge from the infrequency of even these visitors to Xyris flowers, it has to be concluded that North American xyrids are not obligately insect pollinated. Malme and others have speculated that the primary agent of pol- lination is wind, but there is considerable evidence (as yet to be tested) that much seed is set apomictically. REFERENCES: See all family references. BEAL, E. O. A manual of marsh and aquatic vascular plants of North Carolina, with habitat data. N. Carolina Agr. Exper. Sta. Tech. Bull. 247. iv + 298 pp. 1977. LXyris, 147, 148.] 428 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Biomouist, H. L. The genus Yyris L. in North Carolina. Jour. Elisha Mitchell Sci. Soc. 71: 35-46. 1955. [Key to North Carolina species, nomenclatural notes.] CHAPMAN, A. W. Flora of the southern United States. xxxviii + 621 pp. Cambridge, Massachusetts. oe [Xyridaceae, 499-501; several new species described. CorreLL, D. S., & H. B. CorreLt. Aquatic and wetland plants of southwestern United States. Frontisp. ff xvi + 1777 pp. Washington, D. C. 1972. (Reissued in 2 vols. by Stanford Univ. Press. 1975.) [Xyridaceae, 578, 580-588.] & M. C. Jounston. Manual of the vascular plants of Texas. Frontisp. + xv + 1881 _ + map. Renner, Texas. 1970. [Xyridaceae, 348-352.] ELLIOTT, a A sketch of the botany of South-Carolina and Georgia. Vol. 1. iv + 606 I pls. Charleston. 1821. [Xyridaceae, 51-53; X. flexuosa, X. fimbriata, X. juncea eee d. Fassett, N. C. Manual of aquatic plants. vii + 382 pp. York, Pennsylvania. 1940. [Xyridaceae, 169, 170, 173; X. papillosa Fassett.] FERNALD, M. L. Technical studies on North American plants. Rhodora 48: 56-59. pls. 1007, 1008. 1946. [Discussion of ¥. Bayardii Fern. and sect. Brevifoliae.] —, Gray’s Manual of Botany. ed. 8. Ixiv + 1632 pp. Boston. 1950. [Xyridaceae, 387-390.] GLeAsoN, H. A. The new Britton and Brown illustrated flora of the northeastern United States and adjacent Canada. Vol. 1. Ixxv + 482 pp. New York. 1952. [Xyridaceae, 374-377 GopFREY, R. K., & J. W. Wooten. Aquatic and wetland plants of southeastern United States. Monocotyledons. xi + 712 pp. Athens, Georgia. 1979. [Xyridaceae, 479- 502; 19 species, 17 illustrated.] Grisepacn, A. Catalogus plantarum Cubensium. iv + 301 pp. Leipzig. 1866. [Xyri- daceae, 223, 224. Harper, R. M. Two misinterpreted species of Xyris. Torreya 5: 128-130. 1905. [Xyris torta Sm. vs. X. flexuosa Muhl.] Botanical explorations in Georgia during summer of 1901. Bull. Torrey Bot. Club 30: 325, 326. 1905. LX. scabrifolia Harper.] HELLouist, C. B., & G. E. Crow. Aquatic vascular plants of New England: part 5. raceae, Lemnaceae, Xyridaceae, Eriocaulaceae, and Pontederiaceae. New Hamp- shire Agr. Exper. Sta. Bull. 523. iii + 48 pp. 1982. [Xyridaceae, 23-34; includes illustrations, distribution maps, and SEM photographs of seeds of X. torta, X. mon- oe X. difformis, and X. Smalliana.] KRAL, The genus Yyris in Florida. Rhodora 62: 295-319. 1960. [16 species treated a illustrated. ] Identity of site caroliniana Walter (Xyridaceae). Sida 2: 171, 172. 1965. ee hia Xyris eee of continental United States and Canada. /bid. 177-260. [The most recent treatment, including keys, descriptions, maps, chromosome counts for 19 species, figures, some new species.] _ Observations on the flora of the southeastern United States with special reference to ‘northern Louisiana. [bid. 395-408. 1966. [Xyris, remarks on distribution of six species, 400-402.] . Some notes on the flora of the southern states, particularly Alabama and middle Tennessee. Rhodora 75: 366-410. 1973. [Distributional notes on six species, 381, 382.] — ew species of Xyris (sect. Xyris) from Tennessee and northwestern Georgia. Ibid. 80: 444-447, 1978. LX. tennesseensis Kral.] & L. B. Smitu. Typification of Xyris macrocephala Vahl. Taxon 21: 651, 652. 1972. [Nomenclatural note. } Lewis, W. H. Chromosome numbers for three United States species of yris (Xyrida- 1983] KRAL, XYRIDACEAE 429 ceae). Southwestern Nat. 6: 99, 100. 1961. [” = 9 for X. difformis, X. platylepis, X. torta.] _H. L. Srripiinc, & R. G. Ross. Chromosome numbers for some angiosperms ef the southern United States and Mexico. Rhodora 64: 147-161. 1962. [n = 9 for X. Elniottii, X. torta.] Mate, G. O. A. Beitriige zur Xyridaceen—Flora Sudamerikas. Bihang Sv. Vet.-akad. Hand. = 1-18. 1901. , Untergattung Nematopus (Seubert). Entwurf einer Gliederung. Ark. Bot. 130): iC 103. 1913a. [A thorough taxonomic review of the section.] Die amerikanischen Spezies der Gattung Yyris L., Untergattung Euxyris (End- licher). Ibid. 13(8): 1-32. 1913b. [A taxonomic, nomenclatural, and anatomical overview of species of sect. Xyris.] Die Xyridazeen der Insel Cuba. [bid. 19(19): 1-6. 1925. [Some U. S. species treated. ] . De nonullis speciebus Xyridis praecipue australiensibus et asiaticis. (In German and Latin.) Sv. Bot. Tidskr. 21: 381-396. 1927 . Beitrage zur Kenntnis der sudamerikanischen Xyridazeen. Ark. Bot. 25A(12): 1-18. 1933. LX. Drummondii Malme.] Notes on North American Xyridaceae. Bull. Torrey Bot. Club 64: 45. 1937a. [Three new varieties. ] Xyridaceae. N. Am. Fl. 19: 3-15. 1937b. [Treatment to “‘subsections” of Xx ris, but at not given.] Poutsen, V. A. Anatomiske studier over ad an vegetative organer. Vidensk. Medd. Naturh. For. Kjabenhavn 2: 133-152. 1892. RENDLE, A. B. Notes on Xyris. Jour. Bot. London 37: 397-399. 1899. Laer. Rickett, H. W. Wildflowers of the United States. Vol. 2. The southeastern states. 689 pp. New York. 1966. [X ae 83, 84, . Ibid. Vol. 3. Texas. 556 pp. New York. 1969. [Xyridaceae, 41, p/. 13.] Ries, H. Review of the North American species of the genus Xyris. Bull. Torrey Bot. Club 19: 38-43. 1892. [Includes ¥. montana Ries.] Scuuttes, J. A. Xyris. In: J. J. Roemer & J. A. ScHuctes, Linné Mant. Syst. Veg. 1: 350-352. 1822. LX. Baldwiniana Schultes, p. 351.] SMALL, J. K. Flora of the southeastern United States. xii + 1370 pp. New York. 1903. LXyris, 231-234. Manual of the southeastern flora. xxii + 1554 pp. illus. New York. 1933. [Xyris, 251-255.] DEPARTMENT OF eae BIOLOGY VANDERBILT UNIVERSIT NASHVILLE, aaoeeen 37235 WOOD, MENYANTHACEAE 43] THE GENERA OF MENYANTHACEAE IN THE SOUTHEASTERN UNITED STATES! CARROLL E. Woop, JR. MENYANTHACEAE Dumortier, Anal. Fam. Pl. 20, 25. 1829, ‘““Menyanthideae.” Marsh, wetland, or aquatic perennial [or annual] herbs with alternate, simple or 3-foliolate, exstipulate leaves with sheathing petiole bases. Inflorescences various, monopodial or sympodial. Flowers regular (actinomorphic), perfect and often distylous, to imperfect. Calyx of 5 free or united sepals. Corolla sympetalous, 5-lobed, the lobes valvate or induplicate-valvate in bud, often with the margins and sometimes the inner surface fringed or bearded. Stamens 5, epipetalous, alternate with the corolla lobes, the anthers versatile, sagittate stigma 2-lobed; ovary |-locular, with numerous ovules on 2 (or 3) parietal placentae. Fruit a 2-4-valved capsule, or + fleshy and irregularly dehiscent or indehiscent. Seeds numerous to few, usually smooth (or variously ornament- ed), with copious endosperm and a cylindrical embryo. (Tribe Menyantheae Griseb.) Type GENUS: Menyanthes L. A small family of five genera and about 45 species: Menyanthes L., Fauria Franchet (Nephrophyllidium Gilg), and Liparophyllum J. D. Hooker, each with a single species; Vil/arsia Ventenat, with 12 species in Australia and one in South Africa; and Nymphoides Séguier, with about 30 species in Africa, Aus- tralia, tropical America, eastern North America, and Europe. Menyanthes tri- ‘Prepared for the Generic Flora of the Southeastern United States, a project of the Arnold Ar- Georgia, Florida, Tennessee, Alabama, Mississippi, Arkansas, and Louisiana. The descriptions are based primarily on the plants of this area, with information about bien members of a family or genus in brackets [ ]. References that I -have not verified arked with an asterisk. ] am indebted to Dr. Norton Miller for his editorial and cen help, and to Dr. Robert L 1 1: uff, who has kindly I eg The illustration was prepared by the late Dorothy H. Marsh in 1957, long before Dr. Ornduff showed that Nymphoides cordata is dioecious. The living material came from Pay s Pond in Concord, Massachusetts, where only the etna plant occurs, and the other from specimens in the Gray Herbarium from Florida (e, f, Godfrey 55096 & Kral), South Carolina (g, Godfrey & Tryon 405), and New York (h, ee 3541) © President and Fellows of Harvard College, Journal of the Arnold Arboretum 64: 431-445. ae 1983. 432 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 foliata L. is circumboreal in distribution. Fauria Crista-galli (Menzies ex Hook- er) Makino is bicentric, with subsp. Crista-ga/li, 2n = 102, in western North America (coastal Washington and British Columbia, northward to the Gulf of Alaska) and subsp. japonica (Franchet) Gillett, 2 = 68, in eastern Asia (south- ern Kuril Islands, the Japanese islands of Hokkaido and Honshu, and southern Korea). Liparophyllum Gunnii J. D. Hooker is restricted to Tasmania and New ealand. Variously recognized as a tribe of the Gentianaceae (Grisebach, Bentham & Hooker), as a subfamily (Gilg, Rendle), or as a family (Cronquist, Hutchinson, Lindsay, Takhtajan, Thorne, Wagenitz), the Menyanthaceae differ from the Gentianaceae in their marsh or aquatic habitat; alternate, petiolate leaves; collateral vascular bundles; valvate or induplicate-valvate aestivation of the corolla lobes; fused lateral corolla traces and + bilaterally symmetrical vascular plan in the flowers; tenuinucellar ovules with a single integument and integ- umentary tapetum; cellular endosperm development; and chemistry (loganin present, gentiopicrin and L-(+)-bornesitol absent). The Menyanthaceae are generally agreed to be related to the Gentianaceae but were placed in the Polemoniales by Cook and in the Solanales by Cronquist. Nilsson has distinguished two pollen types in the family: the Menyanthes type (Menyanthes, Fauria), with ‘‘3-colporate, generally subprolate to prolate, striate to rugulose” grains; and the Vi//arsia type (Villarsia, Nymphoides, Li- parophyllum), with ““3-colporate, generally oblate to suboblate, parasyncolpate, striate, rugulose, spinulose, verrucose or + smooth” grains. He found that pollen morphology supports an alliance of Menyanthaceae with Gentianaceae, Menyanthes and Fauria having pollen “similar to such genera as Gentiana, Crawfurdia, etc.” Heterostyly and associated self-incompatibility occur in Fauria, Menyanthes, Nymphoides, and Villarsia. Liparophyllum is monomorphic and self-compat- ible (Ornduff, 1982). Both homostylous self-compatible and heterostylous self- incompatible species occur in Vil//arsia (Ornduff, 1974), and dioecious species occur in Nymphoides (Ornduff, 1966). Gynodioecy has been recorded in N. cristatum (Vasudevan Nair). In heterostylous species pollen grains from short- styled flowers are larger than those from long-styled ones. The family is of littke economic consequence. Menyanthes has been used rarely in medicine. Several species of Nymphoides (especially N. peltata) are grown to a limited extent as ornamental, floating-leaved aquatics. REFERENCES: AGABABIAN, V. S., & K. T. TUMANIAN. Palynomorphological study of the family Gen- tuanaceae. IV. (In Russian.) Biol. Zhur. Armenii 30(8): 43-53. 1977. [Includes four genera of Menyanthaceae.] [See also /hid. 33: 466-471. pis. 1-4. 1980. Aston, R. E., & B. L. TURNER. Biochemical systematics. 404 pp. Englewood Cliffs, New Jersey. 1963. [Menyanthaceae reported to have loganin, Gentianaceae do not; Menyanthaceae lack gentiopicrin, which Gentianaceae have.] ANDREAS, B. K., & T.S. COOPERRIDER. The Gentianaceae and ea lees seas of ee Castanea 46: 102-108. 1981. [AM/envanthes extant in Licking, Summit, a counties, once distributed from central to northeastern Ohio; Nymphoides a recorded from Ashtabula.] 1983] WOOD, MENYANTHACEAE 433 Arper, A. Water plants. Frontisp. + xvi + 436 pp. Cambridge, England. 1920. [Menyan- thes, 199, 205, 313; Nymphoides peltatum, see especially 39-41, 240-243.] ASTON, H. I. Aquatic plants of Australia. xvi + 368 pp. Melbourne. 1973. [Menyan- thaceae, 105-133; Nymphoides, 12 spp., 106-117; Villarsia, 12 spp., 118-132; (Li- parophyllum, 133).] BalLey, L. H., E. Z. BAtLey, & STAFF oF L. H. BArLey Hortorium. Hortus third. xvi + 1290 pp. New York & London. 1976. [Gentianaceae, 503; Fauria, 472; Menyanthes, 728; Nymphoides, 773, 774. BAILLon, H. Gentianacées. Hist. Pl. 10: 113-145. 1889. [Série des Ményanthes, 121- 123; Ne Cele 144, 145. BeaL, E. O. A manual of marsh and aquatic vascular plants of North ea bey habitat data. N. Careline Agr. Exper. Sta. Tech. Bull. 247. iv + 298 p [Gentianaceae, 231-235; Nymphoides, 234; Menyanthes, 235.] BENTHAM, G., & J. D. Hooker. Gentianeae. Gen. Pl. 2: 799-820. 1876. [Menyantheae, BOLKHOVSKIKH, Z., V. GriF, T. MATVEJEVA, & O. ZAKHARYEVA. Chromosome numbers of flowering plants. A. A. Feperov, ed. 926 pp. Acad. Sci. USSR. V. L. Komarov Bot. Inst. Leningrad. 1969. [Menyanthaceae, 432.] BREWBAKER, J. L. The distribution and phylogenetic significance of binucleate and trinucleate pollen grains in the angiosperms. Am. Jour. Bot. 54: 1069-1083. 1967. [Includes 1908 genera, approx. half of these studied by the author, Gentianales, 1075.] Cook, C. D. K. Menyanthaceae. Pp. 231, 232 in V. H. Heywoop, consultant ed., Flowering plants of the world. New York. 1978. [Includes illustrations of Menyan- thes, Liparophyllum, Nymphoides peltata; family placed in Polemoniales, but text relates Menyanthaceae to Gentianaceae Cronaulst, A. An integrated system of classification of flowering plants. 1262 pp. New York. 1981. [Refers Menyanthaceae to Solanales rather than to Gentianales.] Davis, G. L. Systematic embryology of the angiosperms. x + 528 pp. New York. 1966. [Gentianaceae, 126, 127; Menyanthaceae, 176.] EICHLER, A. W. Bliith endiagramme. Erster Theil. 348 pp. Leipzig. 1875. [Gentianaceae (incl. Menyanthaceae), 245-249; fig. 134C, Menyanthes.] ERDTMAN, G. Pollen morphology and plant taxonomy. Angiosperms. 539 pp. Stock- holm. 1952. (Corrected reprint with addendum [pp. 541-553]. New York. 1971.) [Gentianaceae (incl. Menyanthaceae), 183-185. Ganpers, F. R. The biology of heterostyly. New Zealand oe ca 17: 607-635. 1979. [Heterostyly in Fauria, Menyanthes, Nymphoides, Villars GASTALDO, P. Compendium of the Italian medicinal flora. S ieaee 46(2): 57-82. 1975.* [Includes ee Gita, E. Gentianaceae. Nat. Pflanzenfam. IV. 2: 50-108. 1895. [Menyanthoideae, 105- 108 Gittett, J. M. The gentians [Gentianaceae] of Canada and Greenland. 99 pp. Research Branch, Canada Dept. Agr. Ottawa. 1963. [Menyanthaceae: Menyanthes, Fauria, Nymphoides, 82-90. Keys, nomenclature, descriptions, comments, illustrations, dis- tribution sca ] ———. The sata of the Asiatic and American populations of Fauria Crista-galli ‘(Menyanthaceae). C nad. Jour. Bot. 46: 92-96. 1968. Gtutck, H. Biologische oe morphologische Untersuchungen tiber ee co Sumpf- gewachse. IV. Untergetauchte und Schwimmblattflora. 746 pp. Jen Grisepacu, A. H. R. Genera et species Gentianearum. vill + 364 pp. ne & Tii- bingen. 1838. [Tribus Menyanthideae, 336-348] Gentianaceae. [n: A. DE CANDOLLE, Prodr. 9: 38-141. 1845. [Tribus II. Menyan- theae, 136-141.] GufGuEN, F. Anatomie comparée du tissu conducteur du style et du stigmate des 434 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Phanérogames (II. Dicotylédones, apétales et gamopétales). Jour. Bot. Morot 16: 48-65. 1902. [Gentianées, 64, 65; includes Menyanthes and Nymphoides.] GuERIN, P. Recherches sur le ere et la structure anatomique du tégument séminal des Gentianacées. Jour. Bot. Morot 18: 33-52, 83-88. 1904. [Ményan- SS 83-85. Le développement de Il’anthére chez les Gentianacées. Bull. Soc. Bot. France 73: 5- 18. 1926. [Limnanthemum (Nymphoides), Menyanthes, 15-17. Guppy, H. B. Water-plants a their ways. Their dispersal and its observation. Sci.- Gossip, II. 1: 145-147. Hara, H. Contributions to the — of variations in the Japanese plants closely telated to those of Europe or North America. Part 2. Jour. Fac. Sci. Univ. Tokyo Bot 343-391. 1956. [Menyanthes, 360; Nymphoides peltata, 361.] Hea, G. Gentianaceae. Illus. Fl. Mittel-Europa 5(3): 1953-2042. pls. 214-217. 1926; 2043-2047. 1927. (Nachdruck der |. Aufl. ees 1957-1960, p/. 214, fig. 1; Nymphoides peltata, 1961-1964, pl. 214, fig. Howarp, R. A. Some observations on the nodes ee plants with special reference to the neers of the split-lateral versus the common gap. Pp. 195-214 in N. K. B. Rosson, D. F. Cutter, & M. Grecory, eds., New research in plant anatomy. New York & London. 1971. (Bot. Jour. Linn. Soc . London 63(Suppl. 1). 1970.) aye 204; quotes Sinnot re Menyanthes, 3-5 bundles enter leaf from as many gap HUTCHINSON, 1 The families of flowering plants. Vol. 1. Dicotyledons. xiv + 328 p London. 1926. [Gentianaceae, 288, 290.] ed. 2. xv + 510 pp. Oxford. 1959. (Gen. tianaceae, 450, 451; Menyanthaceae, 451-453, ua illustrated.] ed. 3. xx + 68 pp. Oxford. 1973. [Gentianaceae, 554, 555; Menyanthaceae, 555-557; Menyan- thes illustrated.] Evolution and phylogeny of flowering plants. xxv + 717 pp. London & New York. 1969. [Gentianales (Gentianaceae, Menyanthaceae), 546-552; Menyantha- .] ceae, 547, 552 Inouye, H., S. Uepa, & Y. NAKAMURA. Biosynthesis of the bitter glycosides of Gen- tianaceae, ore oe and sweroside. (In German.) Tetrahedron Lett. 1967: 3221-3226. JENSEN, S. R., B. J. NIELSEN, re Dien: Iridoid compounds, their occurrence and systematic importance in the angiosperms. Bot. Not. 128: 148-180. 1975. [Gen- tianales, 165-167; families of Gentianales mainly shin anes by occurrence of seco-iridoids (reported in both Gentianaceae and Menyanthace KNOBLAUCH, E. Beitrige zur Kenntniss der Gentianaceae. Bot. Centralbl. 60: 321-334, 353-363, 385-401. 1894. [Tribe Menyantheae, 398-401; Menyanthes, Villarsia, Nymphoides.] KrisHna, G. G., & V. Puri. Morphology of the flower of some Gentianaceae with special reference to A eee Bot. Gaz. 124: 42-57. 1962. sie G. H. M. Taxonomy of vascular plants. x11 + 823 pp. New York. | ntianaceae, including h 670-672; Nymphoides peltata illustrated ] ee P., & M. P. DANGy-CaAye. Biochemical contribution to the taxonomic stud of Gentianaceae. Pl. Med. Phytotherap. 7(2): 87-94. 1973.* [Includes Menyanthes and Nymphoides.] Lersten, N. R. A review of septate microsporangia in vascular plants. Iowa State Jour. Sci. 45: 487-497. 1971. [Includes Gentianaceae. ] Linpsey, A. A. Anatomical evidence for the Menyanthaceae. Am. Jour. Bot. 25: 480- 485. 1938. [Concludes that Menyanthaceae ‘“‘merits full family status.’ Maueswar! Devi, H. Embryological studies in the Gentianaceae: Gentianoideae and Menyanthoidese Proc. Indian Acad. Sci. B. 56: 195-216. 1962.* Martin, A. C. comparative internal morphology of seeds. Am. Midl. Nat. 36: 513- 660. 1946. esas 588, 627; Menyanthaceae, 574, 575.] 1983] WOOD, MENYANTHACEAE 435 ean W.C. Aquatic plants of the United States. x + 374 pp. Ithaca, New Yo rk. 944, [Menyanthaceae, 301-306; Menyanthes, 3 spp.; Nymphoides illustrated; seed- mae of N. cordatum and N. peltatum illustrated MuHLBERG, H. The complete guide to water plants. 392 pp. (including 109 color + 112 black-and-white photographs). EP Publishing. 1982. [Menyanthaceae, 187-190, Menyanthes, color photo 85, Nymphoides peltata, color photos 86, 87 Netouitzky, F. Anatomie der Angiospermen-Samen. Handb. Pflanzenanat. II. Arche- gon. 10. vi + 365 pp. 1926. [Gentianaceae, 263-266; Menyanthoideae, 264.] Nitsson, $. Menyanthaceae Dum. Taxonomy by R. OrnbuFF. World Pollen and Spore ee . 19 pp. Stockholm. 1973. [Includes all genera.] cr, R. Heterostyly in South African flowering plants: a conspectus. (Afrikaans ge eeon gee S. Afr. Bot. 40: 169-187. 1974. [Gentianaceae, 173, 175; Menyan- thaceae, 179— distyly 1 in Villarsia capensis, Nymphoides Thunbergiana.| Perrot, E. Griseb.). Bull. oc. Bot. France 44: 340- 354. pl. 12. 1897. [Menyanthes, Fauria, Villarsia, Lim- nanthemum (Nymphoides), Liparophyllum Anatomie comparée des Gentianacée 5, An nn. Sci. Nat. Bot. VIII. 7: 105-292. pis. 1-9. 1899, (Reprinted as Thése, Fac. Sci Paris, 1899.) [Maintains Gentianaceae ProcEL. A. Gentianaceae. /n: C. F. P. von Martius, FI. Brasil. 6(1): 197-248. pls. 54- 66. 1865. [Limnanthemum, 243, 244: L. Humboldtianum, L. microphyllum.] RENDLE, A. B. The classification of flowering plants. Vol. 2. Dicotyledons. xx + 640 pp. Cambridge, England. 1938. [Gentianaceae, including subfams. Gentianoideae and ai ert 463-468.] Rickett, H. W. Wildflowers of the United States. Vol. 1. The northeastern states. 560 pp. New York. 1966. [Gentianaceae, 306-314; Menyanthes and Nymphoides, 314, pl. 96.] Vol. 2. The southeastern states. 689 pp. New York. 1966. [Gentianaceae, 389-397; Nymphoides, 397, pl. 144.] Vol. 3. Texas. a pp. New York. [Gentianaceae, 281-284; Nymphoides aquatica, 284, pl. 8 Riptey, H. N. The dispersal of plants throughout the Lave xx + 744 pp. Ashford, Kent. 1930. eee 218, 373, 464, 490, 491, 517, 546; Nymphoides (as Lim- nanthemum), 218.] Rork, C. L. Beet studies in the Gentianaceae. Am. Jour. Bot. 36: 387-401. 1949. SCHILLING, N. Distribution of L-(+)-bornesitol in the Gentianaceae and Menyanthaceae. Phytochemistry 15: 824-826. 1976. [In 23 of 32 genera investigated; absent only in subtribe Exacinae and in Menyanthaceae.] ScoccaN, H. J. Gentianaceae. Fl. Canada 4: 1234-1246. 1978. SCULTHORPE, C. D. The biology of aquatic vascular plants. xviii + 610 pp. London. 1967. [Includes numerous references to Menya nthes and Nymphoides.] Stott, K. A. H. Zur Embryologie der Gentianaceen und Menyanthaceen. Sv. Vet.- Akad. Hand. II. 61(14): 1-56. 1921. TAKHTAJAN, A. L. Outline of the classification of flowering plants (Magnoliophyta). Bot. Rev. 46: 225- 359. 1980. [Gentianales, 292, 293.] Tuorne, R. F. A phylogenetic classification of the Angiospermae. Evol. Biol. 9: 35- 106. 1976. ck a Gentianiflorae, 90-92.] Tutin, T. G. Menyanthaceae. In: T. G. TuTin et al., eds., FI. Europaea 3: 67, 68. 1972. ee trifoliata, Nymphoides peltata.] VUAYARAGHAVAN, M. R., & U. PADMANABAN. Morphology and embryology of Cen- taurium ramosissimum Druce and affinities of the family Gentianaceae. Beitr. Biol. Pflanzen 46: 15-37. 1969. WaceEni1z, G. Gentianales (Contortae, Loganiales, Apocynales). [n: H. ME cuior, A. Engler’s Syllabus der Pflanzenfamilien. ed. 12. 2: 405-424. 1964. [Menyanthaceae, ] > 436 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 WETTSTEIN, R. Handbuch der Les Botanik. ed. 3. viii + 1081 pp. Leipzig & Vienna. 1924. [Menyanthaceae, 811 Wiis, J.C. A dictionary of the ene plants and ferns. ed. 7. (Revised by H. K. Airy SHAW). xxil + 1214 pp. + key to the families of flowering plants (liii pp.). Cambridge, England. 1966. [Menyanthaceae, 714, 715]. KEY TO THE GENERA OF MENYANTHACEAE IN THE OUTHEASTERN UNITED STATES General characters: Marsh or aquatic herbs with simple or compound alternate leaves; corolla regular, sympetalous, the 5 stamens alternate with the 5 lobes; gynoecium syn- carpous, 2-carpellate, adnate at the base to the perianth, the ovary I-locular with 2 parietal placentae; fruit a dehiscent or indehiscent capsule. Leaves 3-foliate, emersed; palustrine herbs with a creeping rhizome; inflorescences racemose, emersed; corolla lobes valvate in bud; petals bearded on the inner surface: fruits borne above water, dehiscent, 2-valved. .................... 1. Menyanthes. Leaves simple, cordate to reniform, floating or submersed; aquatic herbs; inflorescences cymose, the flowers opening singly above water; corolla lobes induplicate-valvate in bud; petals not bearded on the inner surface; fruits ripening under water, indehiscent OF OPEMING INTEC UAT] Vig a cic w ed whe awh Ora Pa cea Sha adadepedie wae 2. Nymphoides. 1. Menyanthes Linnaeus, Sp. Pl. 1: 145. 1753; Gen. Pl. ed. 5. 71. 1754. Glabrous perennial aquatic herbs with creeping sympodial rhizomes bearing adventitious roots and a cluster of leaves at the apex; leaves and flowers emer- gent from water. Leaves alternate, ternate, petiolate, with large, membrana- ceous, sheathing bases. Inflorescences racemose on a leafless scape from the terminal bud of the rhizome. Flowers distylous (heterostylous, dimorphic), each subtended by a small bract. Calyx 5-lobed, synsepalous, the lobes partly recurved. Corolla deeply 5-lobed, pink outside, [pink to] white inside, the tube short, adnate to base of ovary, the lobes 3-veined, valvate in bud, the inside of lobes bearded with white fimbriae. Stamens at sinuses of lobes, the anthers basifixed, introrse. Style terminal, the stigma 2-lobed; ovary globular, the pla- centae parietal. Fruit a 2-valved loculicidal capsule, opening along lateral su- tures, with numerous (ca. 20) seeds. Seeds light brown, shining, elliptical, slightly compressed, with abundant endosperm and a cylindrical embryo. Lectotype species: M. trifoliata L.; typified by the transfer from Menyanthes to Villarsia of two of the three Linnaean species (Ventenat, Choix de Plantes, pl. 9. 1803); see also Britton & Brown, Illus. Fl. No. U. S. Canada. ed. 2. 3: 17. 1913. (Name used by Theophrastus, from Greek menyein, disclosing, and anthos, flower; later applied to this genus with flowers expanding in succession in the raceme.) — BOG BEAN, BUCK BEAN. The single species, Menyanthes trifoliata, 2n = 54, is circumpolar in distri- bution in cold bogs, boggy meadows, fens, and shallow waters of pond margins, mainly between lat. 40°N and the Arctic Circle. It occurs in most of Europe but is rare in the Mediterranean region (Tutin), and it is distributed eastward to eastern Central Asia, Siberia, the Himalayas, Tibet, western China and Manchuria, Korea, Sakhalin, and Japan (Hokkaido, Honshu, and northern Kyushu) (Hara). In North America it is found from Greenland, Labrador, and 1983] WOOD, MENYANTHACEAE 437 Newfoundland, west to Alaska, and south to the limit of the Wisconsinan glaciation, with scattered, often disjunct localities in the eastern United States in New Jersey, Delaware, and Pennsylvania, northern Virginia (Madison and Augusta counties), eastern West Virginia (Pocahontas County), northwestern- most North Carolina (Wautauga County), Ohio, Indiana, Illinois, Missouri (Reynolds County, in the southeastern Ozarks), and Nebraska. In the western United States it occurs mostly at higher altitudes south to Colorado, Montana, of corolla size, color, and bearding of the upper surface of the lobes. The species is distylous (or possibly monomorphic in some localities; see Avebury), but collectors seldom indicate whether both long- and short-styled forms are present in a given population. Long-styled plants of var. trifoliata have smaller pollen grains than short-styled ones (see Nilsson, Fossel & Vor- wohl), but the situation is inconsistent in var. minor (Nilsson). According to Guppy, seeds of Menyanthes float for two months. Ravn noted the buoyant seed coat. Ridley recorded that the seeds have been found in the excreta of reindeer, and in the crops of the European wild duck and the Amer- ican mallard duck. Hochreutiner, who fed the seeds to three species of fish, found that nearly all germinated after passage (1-3 days) through the digestive tract. On the basis of morphology, Menvanthes is presumably most closely related to Fauria, a relationship supported by pollen morphology. The dried leaves of Menyanthes have been used as a substitute for hops in brewing. The very bitter juice has been used as a remedy for bowel trouble and dyspepsia; in large doses the effect is purgative and emetic. It has also been used as a tonic, an astringent, an antirheumatic, and a febrifuge. REFERENCES: Also see family references, especially AGABABIAN & TUMANIAN; ANDREAS & Coo- PERRIDER; ARBER; BAILEY, BAILEY ef al.; BAILLON; BEAL; EICHLER; ERDTMAN; GANDERS; NETOLITZKY; ime PERROT; RENDLE; RICKETT, 1966; Rites Rone Soccar SCUL- THORPE; STOLT; and TUTIN Avesury, Lorp [J. L. Luspock]. Notes on the life history of British flowering plants. xxill + 452 pp. London. 1905. [Menyanthes, 288, 289. CLARKSON, R. B. The vascular flora of the Monongahela National Forest, West Virginia. Castanea 21: 1-120. 1966. [Cranberry Glades, Pocahontas Co., locality for Menyan- thes at its southern limit in West Virginia, 26-69. Also mentions population at Big Meadows Bog in Shenandoah National Park, Virginia.] CouLT, D. A., & K. B. VALLANCE. Observations on the gaseous exchanges which take place between CR anInES trifoliata L. and its environment. Jour. Exper. Bot. 9: 384-402. 1958. DOROFEYEV, a I. The Miocene flora from the environs of the village Yurovskoye on Irtysh. (In Russian.) Bot. Zhur. 51: 1480-1489. 1966. [Includes M. parvula Nikit.] FERNALD, M. L. Menyanthes trifoliata, var. minor. Rhodora 31: 195-198. 1929. [Var. minor Michx. ex Raf. the taxon in Atlantic North America.] 438 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Fosse , A., & G. VoRwOHL. Honigbienen und Fieberklee (Menyanthes trifoliata). Phyton Austria 16: 49-55. pl. 3. 197 a G., & P. SIMCHEN. Incorporation of ace into gentiopicroside. Zeit. Naturf. B: 356, 357. 1969.* [Swertia spp., Menyant aa D. G. Biological flora of the British Isles. Menmiie trifoliata L. Jour. Ecol. 52: 723-735. 1964. [Detailed account.] HocureutTINer, G. Dissémination des graines par les poissons. Bull. Herb. Boiss. 7: 459-466. 1899. [Menyanthes, 461, 462. House, H. D. Wild flowers of New York. Part 2. ii + pp. 187-362. pls. 144-264. Albany, New York. 1920. (Popular ed. in 1 vol. 362 pp. + 264 pls. 1934.) [M. trifoliata (var. “a H., T. Yosuipa, S. Tosita, & M. OxiGAwA. Studies on monoterpene glucosides, rt 9. Chemical correlation between asperuloside and loganin. Tetrahedron 1970: 3005 3915. 1970.* [Includes Menyanthes.] JENTYS-SZAFEROWA, J. Seeds of ie ae L. from the Pliocene to the present time. Prace Inst. Geol. Warsz. 10. 1953.* . Importance of Quaternary materials for research on the historical evolution of plants. Veréff. Geobot. Inst. Rubel Ziirich 34: 67-73. 1958.* [Carpinus, Menyan- thes KEELEY, s. L., Jr., & R. W. Koskotcu. Alkaloids and other natural products. Pp. 274- 283 i nc. K. ‘CAIN: ed., Annual reports in medicinal chemistry. 1970. New York. 1971. ‘ [Includes Menyanthes.] Kokawa, 8. Some tentative methods for the age-estimation by means of morphometry of Menyanthes remains. Jour. Inst. Polytech. Osaka City Univ. D. Biol. 9: 111-118. 1958.* —. On the discrete distributi f pl ic values of Menyanthes seed remains in Japan. Ibid. 119-123.* Morphometric reconstruction of the compressed seed remains of Menyanthes in Japan. Ibid. 11: 79-89. 1960.* . Distribution and Da ytoswalieraphy of Menyanthes remains in Japan. Jour. Biol. Osaka City Univ. 12. 1961.* . Age effect on the morphometric values of the fossil de gagee seeds in Japan Bates by Szaferowa’s graphic method. /bid. 13: 87-98. w localities of fossil Menyanthes in Japan with a of its mor- joxeae value ayes Ibid. 14: 97-106. 1963.* Korcuaaina, I. A. Early Quaternary seed floras in the lower course of the Irtish. (In Russian; English summary.) Bot. Zhur. 43: 1121-1134. pls. 1-4. 1958. [Includes Menyanthes LENDRACK, K. Beit itrag zur Kenntniss der Bestandteile von Menyanthes trifoliata und eee Centaurium. 27 pp. Berlin. LieBeLT, O. Ueber der Bitterstoffe des Bitterkles Heda a und der Bar- H ae rom the Neogene Tertiary lignites of central Jutland, part IL a Dansk Vidensk. Selsk. Biol. Skr. 20(9): > ae are = fee tertiaria, ieee: PAMAKSTYTE-JUKNEVICIENE, G. S oe aeons Calamus L.) and ee porbean (Menyanthes Cpe L). (in Russian.) Bot. Sada Pribaltiki 445-450. PaTEL, R. C., J. A. INAMDAR, : As Rao. Structure and ontogeny of stomata in some Gentianaceae and Menyanthaceae complex. Feddes Repert. 92: 535-550. 1981. [Structure and development of stomata in leaves of 8 genera, 12 spp.: includes Menyanthes.] Ponzo, A. Sulla fillotassi: i nomofilli di Menyanthes trifoliata L. Nuovo Giorn. Bot. Ital. II. 44: 201-222. 1937. 1983] WOOD, MENYANTHACEAE 439 Ravn, F. K. Sur le faculté de flotter chez les graines de nos plantes aquatiques et marécageuses. (In Danish; also in French.) Bot. Tidsskr. 19: 143-188. 1894. [Meny- anthes, 157, 182.] SHELDON, J. L. Menyanthes trifoliata in West Virginia. Rhodora 12: 11, 12. 1910. SoueGes, R. Embryogénie des Gentianacées. Développement de embryon chez le Menyanthes Pie. Compt. Rend. Acad. Sci. Paris 217: 488-490. 1943. SpeTA, F. Proteinkérper in Zellkernen: neue Ergebnisse und deren Bedeutung fiir die Gefasspflanzensystematik nebst einer Literaturiibersicht fiir die Jahre 1966-1976. (In German; French and English abstracts.) Candollea 32: 133-163. 1977. [Includes Menyanthes.| STOUSLAND, C. Menyanthes trifoliata L. Pharmakognostische Monographie. Diss. 135 pp. Basel. 1930.* Sykes, W. R. Checklist of dicotyledons naturalized in New Zealand. 10. Polemoniales and Boraginaceae. New Zealand Jour. Bot. 19: 311-317. 1981. [Menyanthes, 313, at Canterbury and Darfield.] THUNBERG, K. P. Dissertatio de usu Menyanthidis trifoliatae. Upsala. 1797.* TRUCHANOWICzOWNA, J. Fossil seeds of the genus Menyanthes in Eurasia. (In Polish; English summary.) Acta Palaeobot. 5(1): 25-53. 2 pls., 7 tables. 1964. . Seeds of the genus Menyanthes from the Polish Miocene. (In Polish; English summary.) /bid. 8(1): 31-52. J pl. 1967. VALLANCE, K., & D. A. CouLt. Observations on the gaseous exchanges which take place between Menyanthes trifoliata L. and its environment. I. Jour. Exper. Bot. 2: 212- 222. 1951.* Watts, W. A. The identity of Menyanthes microsperma from the Gort Interglacial, Ireland. New Phytol. 70: 435, 436. 1971. [Nymphoides cordata.] YAMAGATA, O. A palynological study of a Menyanthes bed from Nagano Prefecture, Japan. Ecol. Rev. 14: 267, 268. 1957. [Peaty lignite with Menyanthes seeds; no Menyanthes pollen.] YORDANOV, D., S. DENcev, & N. NrkoLtov. New chorological data of several species of higher plants. Izv. Bot. Inst. Bulg. Akad. Nauk 25: 211-215. 1974.* [Includes M. trifoliata.] 2. Nymphoides J. F. Séguier, Pl. Veron. (Stirp. Agro Veron. Reper.) 3: 121. Perennial rhizomatous ETDS bes prea rounded to cordate leaves (su- perficially resembling ymphaea or Nuphar). Flowers borne in cymose, umbellike clusters (in our iio slack species at the summit of a slender, petiolelike internode with the petiole of the single floating leaf an apparent continuation of the stem), often with clusters of spurlike or “‘banana- white or yellow, delicate, 5-lobed, bearing 5 glandular, staminodelike tufts or fringes of trichomes near the base, | opposite each lobe; lobes induplicate- valvate in bud. In heterostylous species (e.g., N. peltata) flowers with styles either long and with well-developed 2-lobed stigmas and anthers borne below the level of the stigma on short filaments, or short and with smaller stigmas and anthers borne above the stigmatic level on long filaments; anthers dehiscing introrsely; pollen of long-styled flowers smaller than that of short-styled ones. In dioecious species gynoecium of staminate flowers lacking a style, the stigmas undeveloped, the ovules as large as those of carpellate flowers but nonfunc- 440 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Ficure 1. Nymphoides. a-f, N. cordata: a, floating leaf, inflorescence with spurlike adventitious roots, and small branch with unexpanded leaf and inflorescence, x '; b, staminate flower, showing staminodiumlike appendages at base of petals and stamens alternating with petals, x 3; c, staminate flower, corolla removed to show sterile gynoe- cium and 4 of 5 nectar glands at base of ovary, x 6; d, cross section of ovary of staminate flower showing nonfunctional ovules, x 8; e, fruit, x 4; f, seed, x 10. g, N. aquatica: seed, X 10. h, N. peltata: seed, X 4. tional: in carpellate flowers the style short, the stigmas well developed, fleshy Five nectar glands present at base of ovary opposite base of corolla lobes. Fruits maturing under water, small, indehiscent or irregularly dehiscent with small, plump, nearly smooth to papillate seeds, or (in N. peltata) the capsules larger, resembling those of Frasera, the seeds larger, flat, their margins coarsely ciliate. (Limnanthemum S. G. Gmelin.) Tyre species: Nymphoides aquis innatans Tournefort = N. peltata(S. G. Gmelin) Kuntze. (Name from Greek, nymphaia, and eidos, resemblance, from the similarity of the leaves of NV. peltata to those of the water lily, Nymphaea.) — FLOATING-HEART. A genus of 30-35 species of floating-leaved aquatics, primarily of the tropics, best represented in the Old World (13 species in Africa and Madagascar, 12 species in Australia), with about five species in tropical America and two indigenous to eastern North America. Nymphoides peltata, the single Eurasian species, is often cae as an ornamental plant and is sporadically natu- ralized in the United Sta Nymphoides cordata te Fern. (N. lacunosa sensu Fernald), 2 = 36, is distributed in quiet ‘“‘soft’” (noncalcareous) waters from Louisiana and western Florida, northward on the Coastal Plain to New England, Nova Scotia, and Newfoundland, southwestern Quebec, and Ontario. It has thin, cordate-ovate floating leaves that are nearly smooth beneath and not densely dark pitted, capsules to 1.5 times the length of the calyx, and small seeds with a smooth to papillate surface. Nymphoides aquatica (J. F. Gmelin) Kuntze, banana float- ing-heart, 2n = 36, a larger, coarser plant with ovate to reniform leaves with the lower surface usually thickened and densely dark pitted, the petiole and stem densely covered with dark glands, a capsule much larger than the calyx, and larger seeds with a tuberculate or papillate surface, is found on the Coastal 1983] WOOD, MENYANTHACEAE 44] Plain from southern Florida west to eastern Texas, and north to Delaware and southern New Jersey. The corollas of N. aquatica are almost twice as large as those of N. cordata. However, corollas press so poorly (waxed paper or facial- or toilet-tissue should be used) that they are seldom adequately preserved in herbarium specimens of either these or other members of the genus. Nymphoides peltata (Limnanthemum peltatum S. G. Gmelin, L. Nym- phoides (L.) Hoffmans. & Link), yellow floating-heart, 2 = 54, is indigenous to most of Europe north to Sweden and north-central Russia and is distributed eastward through the Caucasus to Iran, India, Taiwan, Mongolia, Manchuria, southern Siberia, Korea, and Japan. It is sporadically naturalized in the United States in New England, New York, New Jersey, Pennsylvania, Maryland, Ohio, Indiana, Illinois, Missouri, Arkansas, Mississippi, Louisiana, Oklahoma, Tex- as, Arizona, and Washington (see Stuckey). It 1s the best-studied species of the genus (see references). Sometimes forming dense stands in quiet waters, it overwinters by short-shoots that form new leaves and long-shoots in the spring. The flowering stems develop from a leaf axil of a long-shoot, and the flowers are produced sympodially in cymose, umbellike inflorescences. Each golden- yellow flower (ca. 3.5 cm across) lasts a single day, as in other species. The corolla absorbs ultraviolet light centrally and reflects it peripherally (Van der Velde & Van der Heiden), as do the yellow flowers of the Australian N. gemi- nata (R. Br.) Kuntze (see Ornduff & Mosquin). Distyly 1s coupled with a weak self-incompatibility system. Van der Velde & Van der Heijden recorded 43 species of insects (mostly Apidae, Syrphidae, and Ephydridae) as visitors to the flowers in the Netherlands. Megagametophyte development is of the Po- lygonum type. The capsules mature under water one to two months after flowering, then burst open at the base. The flesh becomes soft and decays. The seeds float and are dispersed over the water surface by wind; they also adhere to water birds. In both of our white-flowered indigenous species, the long, slender first internode of the inflorescence (see Goebel) is often misinterpreted as the long petiole of a floating leaf. Actually, each elongated stem bears a single floating leaf and is terminated by a flower; the cymose inflorescence develops sym- podially as in Nymphoides peltata and is supported near the surface of the water by the floating leaf-blade (see FiGure 1). In both species, fleshy spurlike (N. cordata) or somewhat banana-shaped (N. aquatica) adventitious roots are produced on either side of the inflorescence. These clusters function as over- wintering vegetative reproductive structures. Both species are dioecious (Orn- duff, 1966); collectors should note the occurrence of staminate and carpellate plants. The ultraviolet reflectance patterns have not been studied, but in the similarly white-flowered N. indica (L.) Kuntze the corolla absorbs ultraviolet light uniformly (Ornduff & Mosquin). The genus includes diploids, tetraploids, and hexaploids (2” = 18, 36, 54). The American species studied thus far are tetraploids, and Ornduff (1969) found that the white-flowered tropical American tetraploids generally known as Nymphoides Humboldtiana (HBK.) Kuntze are conspecific with the Old World white-flowered diploid N. indica. Both homostylous self-compatibility and distylous self-incompatibility, as well as dioecism, occur in the genus. Ornduff 442 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 (1970) concluded that “dioecism has evolved at least two and perhaps three times in Nymphoides” and that it is “evident that tetraploidy has arisen in- dependently in the genus more than Nymphoides is closely related to ne Vent., which has about 12 species in Australia and one in South Africa. In comparing the Australian members of the two genera, Aston (1969, 1973) pointed out that in Nymphoides the plants are true aquatics, the flowers are in nonpaniculate inflorescences, the pedicels bend downward after flowering, and the fruits ripen under water and are indehiscent or break up irregularly. In Vi//arsia the plants are mostly wetland herbs with erect paniculate (rarely condensed and capitate) inflorescences, the capsules are not carried under water for ripening, and the fruits are usually valvate capsules. Nymphoides peltata, N. aquatica, and N. indica (including N. Humboldtiana) are cultivated in the United States to limited degrees for their attractive flowers and floating leaves. The somewhat bananalike adventitious roots from the inflorescences of N. aquatica are sold as an aquarium oddity, underwater ba- nana plant. REFERENCES: Also see family Ba especially AGABABIAN & TUMANIAN,; ANDREAS & Coo- PERRIDER; ARBER; ASTON; BAILEY, BAILEY et al.; COOK; GANDERS; GILLETT; GRISEBACH; GUEGUEN; GUERIN; Hara; HEGI; Costa CH, LAWRENCE; LEBRETON & DANGY-CAYE; MUENSCHER; anne necans NILSSON: ORNDUFF; PERROT; PROGEL; RICKETT; RORK; SCUL- THORPE; STOLT; and Tu p’ALMEIDA, J. F. R. On the shoot morphology of Limnanthemum. Jour. Indian Bot. Soc. 7: 1-11. 1928. [N. cristata, N. indica.] Anonymous. Banana plant, without bananas. Aquarium 7: 173. 1939.* Aston, H.I. The genus Vil/arsia (Menyanthaceae) in Australia. Muelleria 2: 3-63. 1969. Barrett, S.C. H. Dimorphic incompatibility and gender in Nymphoides indica (Meny- anthaceae). Canad. Jour. Bot. 58: 1938-1942. 1980. [Distylous; lower Amazon, Brazil; strong self- and intramorph-incompatibility system demonstrated among 20 individuals tested.] Boynton, K. R. Nymphoides oe Addisonia 11: 25, 26. pl. 365. 1926. [Illustrates plant from Puerto Rico. CLARK, O. M. Spread of ener peltatum in Lake Messina. Proc. Oklahoma Acad. Sci. 18: 21, 22. 1938. [In newly created lake west of Bristow.] Conarp, H. S. The banana floatingheart (Nymphoides aquaticum). Proc. lowa Aca Sci. 1937. 44: 61-64. 1938. [Comparison of hibernating bodies of N. aquatica witht those of Nymphaea mexicana, important as a duck food.] CorreELL, D. S., & H. B. CorreLt. Aquatic and wetland plants of southwestern United eee Frontisp. + xvi t+ 1777 aa Environmental Protection Agency, Washington, 1972. (Reissued in 2 vols. by Stanford Univ. Press. 1975.) [Gentianaceae, ae 1332; N. aquatica (illustrated), N. peltata, 1331, 1332.] CoUNTRYMAN, W. D. History, spread and present distribution of some immigrant aquat- ic wee a in New England. Hyacinth Control Jour. 8(2): 50-52. 1970.* [Includes N. peltat CRETE, P. Senate Développement de l’embryon chez le Limnanthemum Nym- phoides Hoffg. et Link. Compt. Rend. Acad. Sci. Nat. Paris 242: 3110-3113. 1956.* Date, H. M., & G. E. Mitter. Changes in the aquatic macrophyte flora of Whitewater 1983] WOOD, MENYANTHACEAE 443 Lake near Sudbury, Ontario, Canada from 1947 to 1977. Canad. Field Nat. 92: 264— 270. 1978. [N. cordata.] cae ae E. Unexpected new finds on Taiwan. Taiwania 10: 151-154. 1964. [N. na.] meee WW. J. The identity of the aquatic “banana plant.” Baileya 2: 19-22. 1954. [““Underwater banana plant” = overwintering adventitious tuberous roots produced at base of inflorescences of N. aquatica.] Exias, T. 8S. Menyanthaceae. 7n: R. E. Woopson, Jr., et al., eds., Fl. Panama. Ann. Missouri Bot. Gard. 56: 29-32. 1969. [N. Humboldtiana.] FauTtH, A. Beitrage zur Anatomie und Biologie der Friichte und Samen einiger ein- heimischer Wasser- und Sumpfpflanzen. Beih. Bot. Centralbl. 14: 327-373. 1903. (Includes fruit and seeds of Nymphoides peltata and Menyanthes.] Fensom, D. S., & D. C. SPANNER. Electro-osmotic and biopotential measurements on phloem strands of Nymphoides peltatum. Pl. Arch. Wiss. Bot. 88: 321-331. 1969.* GmELIN, S. G. Lychnanthes volubilis et ee peltatum. Nov. Comm. Acad. Sci. Imp. Petropol. 14(1, 1769): 525-530. p/. 17. 1770.* GoprFRrEY, R. K., & J. W. Wooten. Aquatic and eee plants of southeastern United States. Dicotyledons. x + 933 pp. Athens, Georgia. 1981. [Menyanthaceae: Nym- phoides, 537-540; N. peltata, N. cordata, N. aquatica.] GoEBEL, K. Morphologische und biologische Studien. VI. Limnanthemum. Ann. Jard. Bot. Buitenzorg 9: 120-126. p/. 76. 1891. [Includes morphology of inflorescence; N. aurantiaca, N. cristata, N. Humboldtiana, N. indica, N. peltata.] HATSCHBACH, G., & N. IMAGuIRE. Meniantaceas do Estado do Parana, Brasil. Bol. Mus. Bot. Munic. ‘Curitiba 9: 1-5. 1973. [N. indica.] Hazarnavis, L. Microsporogenesis in Limnanthemum indicum Thw. Jour. Biol. Sci. 2: 118, 119. 1959.* Hensius, H. W. Eenige waarnemingen en beschouwingen over de bestuiving van bloe- men der Nederlandsche flora door insecten. Bot. Jaarb. 4: 54-144. 1892. [Heterostyly and pollen dimorphism in N. peltata.] House, H. D. A new Lae aa the Hudson River flora. Torreya 37: 80-82. 1937. [N. peltata, as N. nymphae Jones, W.N. Obs eae on he response of leaves of Limnanthemum and Tropaeolum to light and gravity. Ann. Bot. I. 2: 819-825. 1938. [N. peltata.] KaAuL, R. B. Anatomical nes apes on floating leaves. Aquatic Bot. 2: 215-234. 1976. [24 genera; Nymphoides, 226, Keppy, P.A. Vegetation with ee coastal plain affinities in Axe Lake, near Georgian Bay, oauag Canad. Field Nat. 95: 241-248. 1981. [Includes N. cordata.] Mayumpar, G. P. A preliminary note on polystely in Limnanthemum cristatum and Ottelia alismoides. Curr. Sci. Bangalore 6: 383-385. 1938.* Menta, A. S. A study of the primary phloem of the petiole of ag peltatum (Gmel.) O. Kunze. Jour. Indian Bot. Soc. 43: 257-261. pls. 1, 2. 1964 MookerJEA, A. Cytology of Limnanthemum cristatum Griseb. oo Sci. Bangalore 20: 328, 329. 1951.* [2n = 18.] OrnbwuFF, R. The origin of dioecism from heterostyly in Nymphoides (Menyanthaceae). Evolution 20: 309 314. 1966. Neotropical Nymphoides the case for taxonomic provincialism. (Abstr.) Am. Jour. a 55: 736, 737. 8. —. Neotropical Nym ae (Menyanthaceae): Meso-American and West Indian speciee Beer 346-352. 1969. [1970.] . Cytogeography of Nymphoides (Menyanthaceae). Taxon 19: 715-719. 1970. ; le arian observations on Vi//larsia (Menyanthaceae). Austral. Jour. Bot. 22: 513-516. 1974 Heterostyly and incompatibility in Villarsia capitata (Menyanthaceae). Taxon 31: 495-497. 1982. 444 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ——— & T. Mosquin. Variation in the i ciate qualities of flowers in the sk pcan possible adaptive significance. Canad. Jou Bot. 48: 603- 605. pl. J. 1970. Pinkava, D. J., E. Lento, & D. Kem. Plants new to Arizona flora, part 2. Jour. Ariz. Acad. Sci. 5: 226. 1969.* [N. peltata, Catalpa bignonioides.] Ponce DE LEON y AyMé, A. Joyas de la flora cubana. Una als ninfa (Nymphoides Grayanum (Griseb.) Arthur). Revista Soc. Cuba. Bot. 15: 77. Ras, T. R. N., & K. M. Ponnappa. Some interesting fungi oe on n aquat ic weeds and Striga species in Pee Jour. Indian Bot. Soc. 49: 64-72. 1970. [Cercospora nymphaeacea on N. indicu Rao, C. G. P. Anatomical ane on 1 fungal galls. I. Preliminary observations on Phy- soderma on Limnanthemum indicum Thw. Phytomorphology 12: 201-204. 1962. Rayna, A. Répartition geographique des Nymphoides (Menyanthaceae) africains et malgaches. (English summary.) Mitt. Bot. Staatssam. Munchen 10: 122- 134. 1971. [Twelve spp., one described as n Le genre Nymphoides (Menyanthaceae) en Afrique et a ot It partie: morphologie. (English summary.) Adansonia II. 14: 227-270. 1974 bid. 2° partie: taxonomie. (English summary.) [bid. 405-458. [13 spp. ] RAYNAL, A &A _ RAYNAL. Un exemple d’application du traitement électronique de fiaformation 4 la construction des clefs dichotomiques. Adansonia II. 14: 459-467. 1974. Reppy, N. P., & B. BAHADUR. Heterostyly in Nymphoides indica. Jour. Indian Bot. Soc. 55: mre 1976.* Royven, P. vAN. Sertum Papuanum 10. Gentianaceae. Nova Guinea Bot. 17: 369-416. pl. 37. 1964. [N. ‘indica, N. narvifolia. N. hydrocharoides, 410-414.] SCHILLING, A. J. Anatomisch- eas Untersuchungen tiber die Schleimbildung der asserpflanzen. Flora 78: 280-360. 18 Se Y., & R. TANAKA. Karyomorphological — in three species of Nyim- phoides in Japan. Jour. Jap. Bot. 55: 20-24. 1980. [N. coreana, N. indica, N. Sincu, S. P., & R. SAHA. Effect of certain abiotic factors on the development of root tubers of Nymphoides indica in a natural lake. Geobios 1: 181, 182. 1974.* Situ, L. B. A new Nymphoides from Colombia. Jour. Washington Acad. Sci. 42: 160, 161. 1952. [N. flaccida.] SPANNER, D. C., & R. L. Jones. The sieve tube wall and its relation to translocation. Pl. Arch. Wiss. Bot. 92: 64-72. 1970.* [Phaseolus vulgaris and N. peltata.] J. N. Preps_e. The movement of tracers along the petiole of Nymphoides peltatum. I. A preliminary study with '*’Cs. Jour. Exper. Bot. 13: 294-306. 1962. ‘i SRINIVASAN, A. Cytomorphological anes of Limnanthemum cristatum Griseb. and Enicostemma littorale Blume. Proc. Indian Acad. Sci. B. 14: 529-542. 1941.* Srivastava, M. G. Chromosome number in genus Limnanthemum. Sci. Cult. 21: 215. 1955.* Stover, E. L. Life history of Nymphoides peltatum. Bot. Gaz. 93: 474-483. 1932. {[Embryological study.] Stuckey, R. L. The introduction and distribution of Nymphoides peltatum (Menyan- thaceae) in North America. Bartonia 42: 14-23. 1974. [Includes map VASUDEVAN, R. A new species of Nymphoides (Menyanthaceae) from South India. Kew Bull. 22: 101-106. 1968. [N. macrosperma; dioecious.] VASUDEVAN Nair, R. Heterostyly and breeding system of Na cristatum (Roxb.) eee Jour. Canes ay Nat. Hist. Soc. 72: 677-682. VELDE, G. VAN DER. mphoides peltata (Gmel.) O. Kuntze ie abies as a food plant es ae lemnata (L.) (Lepidoptera, Pyralidae). Aquatic Bot. 7: 301-304. 1979 _T. G. Giessen, & L. VAN DER HEIDEN. Structure, biomass and seasonal changes 1983] WOOD, MENYANTHACEAE 445 in biomass of Nymphoides peltata Cae : O. Kuntze (Menyanthaceae), a prelimi- nary study. Aquatic Bot. 7: 279-300. , & VAN DER HEIDEN. The ao ee and seed production of Nymphoides peltata (Gmel.,) O. Kuntze (Menyanthaceae). Aquatic Bot. 10: 261-293. 1981. Waaner, R. Die Morphologie des Limnanthemum Nymphaeoides (L.) Lk. Inaug. Diss. Fac. Kaiser-Wilhelms-Universitat, Strassburg. 19 pp. 1895. (Bot. Zeit. 53(1): 189- 205. pl. 8. 1895.) [N. peltata, development, branching, etc.] Wana, D. T. Karyokinetic study of Limnanthemum Ae ee Hoffmgg. et Link. Bull. Fan Mem. Inst. Biol. Bot. 10: 113-115. p/. 6. 194 Warp, D. B. The genus Anonymos and its nomenclatural survivors. Rhodora 64: 87- 92. 1962. [N. aquatica (J. F. Gmelin) Kuntze, 89, Watts, W. A. The full-glacial vegetation of northwestern oe Ecology 51: 17-33. 1970. [Macrofossils include seeds determined to be N . The identity of Menyanthes microsperma n. sp. aes ie it Gort Interglacial, Ireland. New Phytol. 70: 435, 436. pl. J. 1971. [Seed remains identified as a species of Nymphoides, probably N. cordata.] Wicks, J. C. Some morphological aids in distinguishing Nuphar microphyllum from similar aquatics. Rhodora 75: 65-74. 1973. [N. cordata.] Wixpur, R. L. The identity of Walter’s species of Anonymos. Jour. Elisha Mitchell Sci. Soc. 78: 125-132. 1962. [N. aquatica (J. F. Gmelin) Kuntze Wit, H.C. D. pe. Aquarium plants. aba aa +255 pp. + 29 photographs. London. 1964. aes T4178: N ’ aquatica, N. Humboldtiana, N. indica, N. peltata.] ARNOLD ARBORETUM 22 Divinity AVENUE CAMBRIDGE, MASSACHUSETTS 02138 ENDRESS & SAMPSON, TRIMENIACEAE 447 FLORAL STRUCTURE AND RELATIONSHIPS OF THE TRIMENIACEAE (LAURALES) PETER K. ENDRESS AND F. B. SAMPSON THE FAMILY Trimeniaceae is a small western Pacific group of the Laurales, formerly included in the Monimiaceae but separated as an independent family by Gibbs (1917) and, with a detailed discussion, by Money, Bailey, and Swamy (1950). It comprises two genera, 7rimenia Seem. (three to seven species) and Piptocalyx Oliver (two species) (Perkins, 1925; Rodenburg, 1971; Smith, 1978), occurring in eastern Australia, Celebes, the Moluccas, New Guinea, New Brit- ain, Bougainville, New Caledonia, Fiji, Samoa, and the Marquesas Islands (Perkins, 1925; Rodenburg, 1971; Van Balgooy, 1975). This study, undertaken because the Trimeniaceae were poorly known and had not yet been studied comparatively, is an investigation of the floral structure of species of Trimenia and Piptocalyx. In addition, systematic relationships, both within the family and with other groups, are discussed. MATERIAL AND METHODS Floral material fixed in FAA was studied from the following species and collections: Trimenia papuana Ridle Endress 4066 (cited in the text as E 4066), Aug. 1977, eastern Papua New Guinea (buds, flowers, fruits). Endress 4087, Aug. 1977, eastern Papua New Guinea (buds). Womersley s.n., Sept. 1970, eastern Papua New Guinea (buds, flowers). Trimenia neocaledonica Baker f. McPherson 4044, Aug. 1981, New Caledonia (buds). Endress 6315, Sept. 1981, New Caledonia (buds, flowers). Trimenia weinmanniifolia Seem. Siwatibau s.n., July 1971, Fiji (buds, young fruits). Piptocalyx moorei Oliver Endress 4005, July 1977, Coffs Harbour, New South Wales, Australia (buds). Endress 4367, Sept. 1977, Coffs Harbour, New South Wales, Australia (buds, flowers). WELTU 13764 (Sampson), Aug. 1981, near Armidale, New South Wales, Australia (buds). Plants of Trimenia papuana, T. neocaledonica, and Piptocalyx moorei were observed at anthesis in the field. Anatomical investigations were carried out © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 447-473. July, 1983. 1-8. Flowers and buds. 1-3, flowers at anthesis: 1, Trimenia papuana (E 4066), hermaphroditic; 2, 7. neocaledonica (E 6315), staminate; 3, Piptocalyx moorei (E 4367), hermaphroditic. 4-6, mature flower buds: 4, 7. papuana (E 4066); 5, T. 1983] ENDRESS & SAMPSON, TRIMENIACEAE 449 with serial microtome sections stained with safranin and astra-blue or with Heidenhain’s hematoxylin. In addition, pollen formation and development were studied by means of squashes made in iron aceto-carmine. Gynoecial structure and development were examined with the aid ofa Philips 505 SEM at Victoria University of Wellington, New Zealand. OBSERVATIONS Tue FLOWERS AT ANTHESIS TRIMENIA PAPUANA. Trimenia papuana is a tree of tropical mountain forests. The paniculate inflorescences are produced in leaf axils and at the end of leafy shoots. They consist of many small, inconspicuous, scentless flowers. In our collections most flowers were hermaphroditic, but a few were staminate, with the gynoecium reduced or lacking. Whitish stamens with long, narrow anthers surround a normally single styleless carpel consisting of a green ovary and a terminal, tuftlike, white, dry stigma (Ficure 1). The pollen is dry and is easily blown from open anthers. No nectar is produced. Flower visitors were not seen. These features give the impression that wind plays a major role in pol- lination. A peculiar feature is that before anthesis the outermost tepals fall off, and at full anthesis all the inconspicuous, brownish tepals have been shed. OTHER SPECIES. The flowers of Trimenia neocaledonica (a tree) and Piptocalyx moorei (a vine), both occurring in subtropical rain forests, do not differ from those of T. papuana in the features described above (Ficures 2, 3, 35).! Al- though the flowers have broader anthers and are thus slightly more showy, they too are probably predominantly wind pollinated. A difference is that in both of these species the inflorescences are fewer-flowered botryoids (racemelike panicles; for terminology see Troll, 1964). Both species are andromonoecious, with more male flowers than are present in 7. papuana. PHYLLOTAXY AND NUMBER OF FLORAL PARTS TRIMENIA PAPUANA. Phyllotaxy was determined in four flowers (two from each of two collections). The mean divergence angles of all floral parts were Lat6: and 138.7° in specimen EF 4087, and 138.2° and 138.4°in W, 1970. The average of all 136 divergence angles present in the four flowers was 138.2°. The most been published, some of them in remote places. Therefore, it seems reasonable to add the following Swamy, 1950; Smith, 1978), 7. weinmanniifolia (Seemann, 1871; Perkins & Gilg, 1901; Hutchinson, 1973; Smith, 1978, 1981), Piptocalyx moorei (Maiden & Baker, 1895; Oliver, 1895; Perkins, 1911; Beadle, 1972), and P. macrurus (Gilg & Schlechter, 1917, 1923). OO neocaledonica (M 4044), 6, P. moorei (W 13764). 7, 8, T. papuana (E 4087), sections of flower bud at time of initiation of inner stamens: 7, transverse; 8, longitudinal (dotted areas, procambium, with phloem in differentiated vascular bundles; black areas, xylem). 450 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 regular values were obtained from flower buds in which the innermost floral parts had just been initiated. The phyllotaxy of the flowers is spiral, with more or less constant divergence angles approaching the “limiting divergence” (137.5°) of the Fibonacci series (Ficures 7, 41, 42). The constancy is also maintained at the transition from the perianth to the androecium and that from the androecium to the unicar- pellate gynoecium (FiGurE 53). In terminal flowers the outermost “tepals” (ca. 6) are often arranged in decussate pairs before the spiral phyllotaxy starts, thus continuing the decussate phyllotaxy of the bracts on the inflorescence axis. We call them tepals because there is no abrupt change between these organs in phyllotaxy, shape, or internode length. In lateral flowers the first two phyllomes take the position of prophylls. They are situated transversely, with a somewhat larger angle to the anterior than to the posterior. It is also useful to call these organs tepals because they are not distinctive in shape or internode length from the following ones (tepals). Thirty flowers (ten from each of three collections) revealed the following range in the number of floral parts: perianth (P), (13 to) 17 to 25 (to 28); androecium (A), 14 to 25; and gynoecium (G) (0 or) | (or 2). Thus, the range given by Rodenburg (1971) is extended by our collections, This result reem- phasizes the wide range of variation in number of floral parts in flowers with a spiral phyllotaxy. Because the stamens have a narrow base, they can be distinguished from tepals even at early stages in development, in spite of the unaltered phyllotaxy and similar marked plastochrons (Ficures 7, 41). Toward the end of androe- cium initiation, the roundish floral apex decreases in size and becomes more or less five-angled in outline due to the effect of the adjoining stamens (FIGURE 53). OTHER SPECIES. In Trimenia weinmanniifolia and Piptocalyx moorei the floral phyllotaxy is the same as in 7. papuana. The flowers of these species are spirally arranged, with the mean divergence angle approaching 137.5°. Divergence an- gles were not measured in 7. neocaledonica because young enough buds were not available, but the arrangement is clearly spiral. Trimenia neocaledonica differs from 7. papuana and P. moore in often having the first two (transverse) phyllomes at some distance from the flower at the base of the pedicel (FIGURES 4-6). (Such flowers were not included in the count of organ numbers.) Thus, the limits between “‘prophylls” and ‘tepals’ are not clear-cut in this group. The range in number of floral organs in the three species (from twenty flowers of each) is: Trimenia weinmanniifolia (S, 1971) P, 14 to 23 A, 7 to 12 G, 0 or | T. neocaledonica (M 4044) P..13 16:2) A, 11 to 15 G, 0 or | Piptocalyx moorei (E 4005, W 13764) Pz to. A, 7 to 16 G, 0 or | (or 2) 1983] ENDRESS & SAMPSON, TRIMENIACEAE 451 PERIANTH TRIMENIA PAPUANA. The 13 to 28 tepals gradually change in size and shape from the smaller, roundish or depressed-obovate outermost ones to the larger, obovate to spathulate inner ones (Ficures 9, 33). The spathulate tepals form hoods over the androecium by virtue of their apically broadened and curved flanks (FiGurE 9, nos. 11-14). The outermost tepals have thick bases, with a basal extension resembling a hypopeltate bract on the dorsal side (FIGURES 8; 9, nos. 1-9; 39; 40; cf. Endress, 1975). Small-celled abscission tissue is differ- entiated at the base of each tepal (Ficure 40). Abscission of a tepal starts adaxially. Toward the center of the flower the tepals become thinner, and marginal parts may be reduced to a single cell in thickness. All tepals have a single unbranched vein. The outer tepals bear uniseriate, tricellular hairs on the margins that are exposed in the bud. Each hair has an elongated apical cell. The inner tepals are glabrous. Oruer species. In Trimenia weinmanniifolia the 14 to 23 tepals do not differ from those of 7. papuana described above. In Trimenia neocaledonica all 15 to 21 tepals are roundish. The outermost ones are much smaller than the inner. The basal extension is less marked than in T. papuana. The veins are unbranched in the outer tepals and laterally branched in the inner ones; the innermost tepal again has simplified vasculature (FiGureE 10). All tepals are glabrous. The 2 to 11 tepals of Piptocalyx moorei are ovate to obovate and lack a marked basal extension. The veins are branched in the outer tepals and simple in the inner ones (FiGuRE 11). However, as in the other taxa, there is always only one trace to each tepal. The outer surface of the outer tepals is pilose where it has not been covered by other tepals or by the subtending bract in the young bud. The tricellular hairs resemble those of Trimenia papuana. ANDROECIUM In Trimenia papuana the 14 to 25 stamens are ca. 4 mm long and slender (Ficures 13, 14, 34). The anther and the filament are equal in length. The mature anther is latrorsely dehiscent, opening by two lateral slits (FiGURE 17). It is capped by a short, linguiform connective apex. One collateral vascular bundle serves the stamen, ending in the connective apex in a group of short, tracheidlike cells. In a few flowers transitional forms between tepals and sta- mens have been seen: these were spathulate organs with a rudimentary pollen sac situated toward the base (FiGuRE 12). Although the 11 to 15 stamens of 7rimenia neocaledonica and the 7 to 16 of Piptocalyx moorei are of about the same length as those of 7. papuana, they have much broader filaments and broader, extrorsely dehiscent anthers (FIGURES 15, 16, 18, 19, 43). In all three species there is a considerable increase in filament length from the time tepals are detached to when pollen is shed. The anthers are tetrasporangiate in all species. Both genera have microspo- rangia with large, persistent epidermal cells (smallest in 7rimenia papuana, largest in T. neocaledonica) and an endothecium that develops the characteristic 452 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 PODODOOG QDO00G! een ales EY, Me ce <>" 12 __imm 9-1] 15 nOG0000 —- 9- : 1. Tepals of mature flower buds (each sequence from one bud), adaxial w: 9, Trimenia papuana (E 4066), outermost tepals not drawn; 10, 7. neocaledonica (E 6315) 11, ies moorel (E 4367). fibrous radial bars of thickening. The endothecium is restricted to the external part of each theca and is absent in the connective (FIGURE 43), which has a high tannin content. There is an irregularly one- or two-layered middle wall region that becomes crushed before anthesis. FiGure 44 illustrates a stage in 1983] ENDRESS & SAMPSON, TRIMENIACEAE 453 Figures 12-19. Stamens. 12, Trimenia papuana (£ 4066), innermost tepal with rudimentary pollen sac, adaxial view, from mature bud. 13, 7. papuana (E 4066), stamen at anthesis, lateral view (right side = adaxial). 14-16, stamens at anthesis, abaxial view: 14, T. papuana (E 4066); 15, T. neocaledonica (E 6315), 16, Piptocalyx moorei (E 4367). 17-19, anthers shortly before dehiscence, transverse sections, adaxial side above, en- dothecium hatched: 17, 7. papuana (E 4066), 18, T. neocaledonica (E 6315); 19, P. moorel (E 4367). the development of the microsporangium of Piptocalyx moorei after the epi- dermis, endothecium, and middle wall layer have been differentiated. A single- layered tapetum encl porog tissue that is still undergoing cell division (not visible in the photograph). The tapetum is glandular, and by the time pollen mother cells have reached the early stages in prophase I of meiosis, most tapetal cells are binucleate (Trimenia papuana, Piptocalyx moorei). Microsporogenesis was observed in T. papuana. Cytokinesis is of the successive type, with wall formation by a centrifugal cell plate occurring at the end of meiosis I and IJ (Ficures 45, 46). 454 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Ficures 20-26. Trimenia papuana, carpel development, median longitudinal sec- tions (left side = adaxial): 20, origin of ventral ‘“‘cross zone” (E 4087); 21, chairlike stage (W, 1970); 22, early epidermal periclinal divisions in future stigmatic region (cell rows 1983] ENDRESS & SAMPSON, TRIMENIACEAE 455 The resulting tetrads are isobilateral, decussate, or intermediate between these types. In the limited material examined (several buds from a single collection), division stages were synchronized in each microsporangium throughout mei- osis—an unusual feature. Although early stages in the division of pollen mother cells are normally synchronous due to linkage by cytoplasmic channels, exact synchroneity is usually subsequently lost when a callose wall forms around each meiocyte (Heslop-Harrison, 1966). Individual microspores within a tetrad are bilaterally symmetrical (FiGURE 47). Due to dissolution of callose wall material, microspores have a plano-convex shape for a time after they have separated; this reveals their orientation when they are in tetrads. Thus in FIGURE 48 the center of the curved edge of the microspore is at the distal pole. With maturation the pollen grains of 7. papuana become more rounded, but they retain an ellipsoid (rather than spherical) shape. When pollen mitosis occurs, the generative cell is cut off so that it lies midway between the two more pointed ends of the grain. This is also the case in 7. neocaledonica (FiGuRE 49) and P. moorei. In these two species pollen is biaperturate, with the centers of the apertures located at the two more pointed ends of each grain. This is in contrast to 7. papuana, which in our collections had inaperturate pollen (Sampson & Endress, in press). Polyforate pollen has been reported in some collections of this species (Walker, 1976) and in 7. weinmanniifolia (Money, Bailey, & Swa- my, 1950). Although our material of 7. neocaledonica and P. moorei did not contain stages with pollen in tetrads, we assume (by analogy with 7. papuana) that the more pointed ends of the grains are equatorially aligned and therefore that the pollen is disulculate (in the terminology of Walker, 1976). Because the nascent generative cell is equidistant from the two apertures, it obeys Huynh’s “law of the longest distance” (Huynh, 1972; Sampson, 1982). The generative cell is soon detached from the wall of the grain to lie within the cytoplasm of the generative cell. In all three taxa the pollen is shed in the two-celled stage (Figures 50, 51). The septum between each pair of microsporangia breaks down before de- hiscence (FiGurE 43). Lateral longitudinal slits are formed, each by separation of small cells along a line of weakness (stomium) at the outermost part of the wall that separates radially adjacent microsporangia, and pollen is shed. GYNOECIUM TRIMENIA PAPUANA. The gynoecium consists of one (very rarely two) carpels. At anthesis the carpel is ca. 2 mm long and | mm broad. The barrel-shaped, of epidermal origin indicated), inception of inner integument (FE 4087); 23, rotation of with meiocytes in ovule, integumentary vascular bundle still procambial (cf. Figure 26) (W, 1970); 25, ovule at anthesis, elongated nucellus and long, tubular embryo sac (W, 1970); 26, nucellus with numerous sane ke (enlarged from FiGureE 24) (W, 1970). (Dotted areas: meristematic tissue; procambium; phloem in differentiated vascular bun- dles. Black areas: xylem in differentiated ee bundles. Plus signs: tanniferous tissue. Black lines in stigmatic region: cell rows. Ficures 27-29. Transverse sections of carpels at anthesis Sener side up). 27, 7r/- menia papuana (E 4066), series from | carpel: a, stigma, “cross zone’; b, upper extension of dorsal bundle and 2 adjacent bundles; c, vascular bundles rane ring; d, e, top of 1983] ENDRESS & SAMPSON, TRIMENIACEAE 457 Ficures 30-32. Median longitudinal sections of carpels at anthesis (left side = adax- ial): 30, Trimenia papuana (W, 1970): 31, T. neocaledonica (EF 6315); 32, Piptocalyx moorel (E 4367). (Dotted areas: phloem. Black areas: xylem.) slightly asymmetric ovary is topped by a capitate, tuftlike stigma; there 1s no distinct style (FiGurEs |, 36, 63). The carpel is remarkable in being extremely utriculate (FIGURE 30). At anthesis the ventral and dorsal regions are nearly the same length. The ventral “‘cross zone” is situated at the top of the carpel in the stigmatic region (FiGurE 30). The “‘stylar’” canal remains a narrow, transverse slit resulting from thickening but not widening of the carpellary wall (FIGURE 27, a—d). Longitudinal shallow furrows on the ovary surface mark the position of the anthers in bud. On the ridges between these depressions, rows ovary; f, tip of outer (oi) and inner (ii) integuments; g, middle region of ovary (raphe (r) , base of nucellus and integuments; i, base of ovarian cavity; k, 1, union of vascular bundles into | dorsal (d), 1 ventral (v), and 2 laterals (1); m, base of ovary, rearrangement of carpellary bundles toward stele; n, se of androecium, stele with stamen traces. 28, T. neocaledonica (E 6315), middle region of ovary. 29, Piptocalyx moorei (E 4367), middle region of ovary, slightly earlier os (Dotted areas: phloem. Black areas: xylem. Small circles at periphery of sections: mucilage cells 458 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Ficures 33-38. Bud, flowers, and gynoecia, x 10. 33, Trimenia papuana (E 4066), mature bud, outermost tepals fallen. 34, 7. papuana (E 4066), flower shortly before anthesis, tepals removed. 35, Piptocalyx moorei (E 4367), bicarpellate flower at early anthesis. 36-38, gynoecia after anthesis: 36, 7. papuana (E 4066), 37, T. neocaledonica (E 6315), 38, P. moorei (E 4367). of hooklike, upwardly directed, unicellular hairs are developed (FiGures 60, 61). The carpel contains one ovule in a ventral-median position. The ovule is crassinucellar, bitegmic, anatropous, and pendent, with the micropyle directed upward (FiGures 25, 27, 30, 64). 1983] ENDRESS & SAMPSON, TRIMENIACEAE 459 At anthesis a number of branching and anastomosing bundles (up to ten on each side) extend through the ovary wall between the dorsal and the ventral vascular bundles (FiGureE 27, g). They are more numerous than as described by Money, Bailey, and Swamy (1950, table 3, p. 389)—20 or more compared to four or five! The dorsal bundle terminates immediately below the stigma, and the other bundles are slightly shorter (FiGuRE 27, b). The bundles all join near the top, forming a ring of short, tracheidlike cells (FIGURE 27, c). In the upper region of the carpel, each bundle has several tracheids on its outer side, while the normal xylem on the inner side is smaller (FiGuREs 27, d; 30). Toward the base of the carpel, the lateral bundles anastomose, so that four more or less clear-cut complex bundles remain: a dorsal, a ventral, and two laterals (FiGurRE 27, h-m). There is no obvious reorientation of the bundles demarcating the transition between the gynoecium and the floral stele, in contrast to the situation in many other plants A strand from the ventral bundle supplies the ovule. The ovular bundle does not terminate at the chalaza but extends to the top of the outer integument near the micropyle (FiGures 25, 27). Here, the outer integument has a very short, discrete tip (FiGuRE 27, f). In the raphe and the outer integument (an- tiraphe) the vascular bundle occupies the peripheral ie which suggests that thickening of these parts occurs only on their inner si The inner integument is three or four cells thick. The outer integument is about six or seven cells thick, except at the top (ten to fifteen cells) and in the vascularized median region. There are three histologically conspicuous cell layers: the very small celled, ‘‘meristematic” inner epidermis of the ovary wall; the outer epidermis of the outer integument, with radially much elongated cells that develop into the stony layer of the seed; and the inner epidermis of the inner integument, which is tanniferous (FIGURE 25). The ovule almost completely fills the ovary; the small gaps in between contain a mucilaginous secretion. In the middle region of the ovary, deep furrows separate the raphe and the antiraphe from the central portion of the ovule (FiGurE 27, g, h). The inner integument forms a massive cap over the micropyle, with the center containing a tanniferous zone (FicuRES 25, 64). The outer integument is also involved in micropyle formation. The long, narrow embryo sac extends throughout the entire length of the nucellus, almost growing out of the top of it (FiGureE 25). The plumose stigma consists of hice pluricellular, pluriseriate, tanniferous hairs (FIGURE 62). The periphery of the carpel large mucilage cells and—in the uppermost part—some groups of weakly lignified cells; the upper part of the center of the carpel (and ovular base) has especially thick- walled (but not lignified) cells with irregular thickenings. It seems unlikely, however, that this is pollen tube transmitting tissue because the cells are iso- diametric, not elongated. Pollen tubes probably grow only through the narrow “stylar” canal. Together with the ovular surface, this tissue becomes lignified, beginning in its center, during fruit development. Soon after anthesis, the ovary wall interior to the vascular bundles becomes tanniferous. There is no further lignification in the ovary wall. The groups of 460 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 meats a A - ny lt agin eet ay agi A ae ig ws 4 FiGures 39-42. 4087), median longitudinal section, x 40 (cf. FiGure 8). 40, abscission of outer tepal (E from mature bud (E 4066), median longitudinal section (note dorsal extension), x 100. 41, 42, floral buds at time of initiation of inner stamens (£ 4087), note spiral phyllotaxy: 41, tepals and few outermost stamens removed, x | 20; 42, tepals and many outer stamens Trimenia papuana. 39, floral bud at time of androecium initiation removed, x 200. <7. <6 11 SP WS Figures 43-51. Anther and pollen development. 43, Trimenia neocaledonica (E 6315), transverse section through nearly mature anther, x 100. 44, Piptocalyx moorei (E 4005), portion of transverse section through young anther, showing microsporangium with sporogenous tissue, X 375. 45-48, T. papuana(W, 1970), x 800: 45, pollen mother cells at prophase II; 46, pollen mother cells at telophase II; 47, young tetrad with callose wall material enclosing microspores (only 3 visible); 48, young microspore soon after separation from tetrad. 49, 7. neocaledonica (M 4044), pollen grain soon after formation of vegetative cell and generative cell (at left), x 800. 50, P. moorei (E 4367), mature pollen grain, x 800. 51, 7. papuana (F 4066), mature pollen grains, 800. 462 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 FiGures 52-57. Trimenia Se gs (E 4087). 52, 53, floral apex at time of carpel initiation: 52, lateral view, X 200, carpel primordium at left; 53, from above, X 330, el primordium at right, 54, 55, carpel primordium with ventral ‘“‘cross zone’’ dif- Pee ee 54, lateral view, X 360; 55, from above, < 410, note residual floral apex at left with convex cell surfaces. 56, 57, carpel primordium at chairlike stage: 56, lateral view, X 360; 57, from above, X 410. 1983] ENDRESS & SAMPSON, TRIMENIACEAE 463 weakly lignified cells at the upper periphery of the carpel do not differentiate further. The carpel is initiated at a normal divergence angle from the last stamen, to one side of the apex (FiGurE 53). Since a small portion of the ventral part of the apex is not used up during carpel formation, the carpel is not truly terminal. In SEM preparations the residual apex is distinguished by 1ts convex cell surface; on the fast-growing carpel primordium the cell surface is flatter (FIGURES 52- 57). Histologically, the residual apex soon becomes tanniferous (FiGuREs 20, 21). The carpel primordium is thicker than a stamen primordium even at a morphologically undifferentiated stage (FIGURE 53). The carpel primordium soon becomes angled because of the protruding neighboring stamen primordia (Figures 55, 57). Halfway up its ventral side a platform (“cross zone’’) arises (FiGures 20, 54), and the carpel primordium becomes chair shaped (FIGURES 21, 56). At this stage the carpel bears considerable resemblance to those of the Lauraceae (Endress, 1972) and certain Monimioideae of the Monimiaceae (Sampson, 1969a; Endress, 1980a). Unlike in the Lauraceae and the Monimi- oideae, however, during further development by extensive intercalary elon- gation, the relative difference in length between dorsal and ventral parts de- creases, and the carpel becomes extremely utriculate (FIGURES 22-24, 58-60). There is an early inception of the pluricellular stigmatic papillae, coupled with (and even preceded by) extensive periclinal cell divisions of the epidermis in the apical part of the carpel (FiGures 22, 23, 60). The stigma is conical in bud; only at anthesis does it expand and become capitate (FiGurEs 61, 63). By anthesis, the ovule has rotated almost 360° from the original ventral cross zone where it originated (FIGURES 21-25, 30). The inner integument appears before the outer, as is normal in bitegmic ovules (FiGuRE 22). The nucellus undergoes relatively little growth until meiosis and is soon overtopped by the inner integument. This leaves only a narrow micropyle, which becomes indis- tinct even before anthesis (FIGURE 24). At the time of integument initiation, the nucellar tissue consists of relatively small cells and an archesporium is not yet distinct. Later, numerous meiocytes differentiate at the base of the nucellus (FiGurREs 24, 26, 64). Thus, the arche- sporium is multicellular. One or several embryo sacs start growing after meiosis, but only one reaches maturity. It grows like a narrow tube toward the apex of the now fast-elongating nucellus and stops immediately at the border of the nucellar apex, near the cap over the nucellus formed by the inner integument (FiGuRE 25). There it becomes more voluminous. The egg apparatus and polar nuclei are situated near the nucellar apex, while the antipodals probably remain at the base of the embryo sac tube but are hard to trace. It appears as if the tanniferous hypostase and the tanniferous cap over the nucellus formed by the inner integument prevent further elongation of the intrusive embryo sac apices at both ends of the nucellus. OTHER SPECIES. The gynoecial structure of Trimenia neocaledonica and Pip- tocalyx moorei closely resembles that of 7. papuana (Ficures 31, 32). There- fore, only differences will be mentioned. 464 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Ficures 58-61. is et olin 58,59, intercalary longitudinal growth of middle and basal regions, X 220 (note ey cell “‘families’’): 58, lateral view; 59, from adaxial side. 60. young carpel at inception of stigma differentiation (W, 1970), X 80. 61, carpel with conical stigma, from older bud (£ 4066), x 30 In Trimenia neocaledonica the carpel at anthesis is longer (ca. 3 mm) than in 7. papuana (Ficures 31, 37). Histologically, the whole carpel is richer in tannins; the mucilage cells at the carpel periphery are larger, and their mucilage content seems to be higher (FIGURE 28). The ovule is thicker than in 7. papuana. In Piptocalyx moorei the carpel is ovoid and thicker (ca. 1.5-2 mm) but not i 2 ee F th “% at BX O72 e8 °9 FiGurREs 62-65. Stigmas and ovules. 62-64, Trimenia papuana: 62, stigma from old bud (E 4066), longitudinal section, x 100 (note pluricellular papillae); 63, stigma at anthesis (E 4066), X 30; 64, ovule with several meiocytes (W, 1970), median longitudinal section, X 160 (note tanniferous tissue in top of inner integument and hypostase). 65, Piptocalyx moorei (E 4367), ovule at anthesis, longitudinal section, x 120, nucellar apex depressed and with papilla, embryo sac apex inside papilla. longer than in Trimenia papuana (Ficures 32, 38). Histologically, the tannin content is lower. There are fewer (but larger) mucilage cells (FiGuRE 29). In the upper carpel wall the outer tracheid zone of the vascular bundles is even more developed than in 7. papuana. The most obvious difference in the ovular structure at anthesis is that the nucellar apex is depressed, and in the center of this depression the tip of the ripe embryo sac is situated inside a papilla (FIGURES 32, 65). The inner integument contains conspicuously elongated cells near the 466 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 depression. In both species the ovary walls are glabrous, in contrast to those of 7. papuana. DISCUSSION STRUCTURE OF FLORAL ORGANS PHYLLOTAXY, NUMBER, AND INITIATION OF FLORAL PARTS. The floral organs have a spiral phyllotaxy in all four species examined. This has not been rec- ognized in earlier studies. While some authors (Perkins & Gilg, 1901; Gibbs, 1917; Perkins, 1925; Rodenburg, 1971) report a spiral perianth for 7rimenia species, only Morat and MacKee (1977) state that the androecium has a spiral phyllotaxy (in 7. neocaledonica). The reason is that the spiral 1s obscured during later ontogeny because of differential expansion of the flower base caused by the extensive enlargement of the carpel. In Piptocalyx spiral floral phyllotaxy has not previously been recognized. The analysis of floral parts in our material has extended the previously recorded range of variation in stamen number in all species, and in tepal number in all species except 7. weinmannilfolia (Rodenburg, 1971; Beadle, 1972; Morat & MacKee, 1977). Trimenia neocaledonica has invariably been described as having 12 stamens, even in the recent studies of Rodenburg (1971) and Morat and MacKee (1977). In our material it varied from 11 to 15, with an average of 13.2. Thus, the family Trimeniaceae is an additional one within the Magnoliidae to have spiral floral phyllotaxy (cf., for example, Endress, 1980a, 1980b). In all these cases there tends to be an obvious correlation between spiral phyl- lotaxy, marked plastochrons, and considerable variation in the number of floral arts PERIANTH. In the Trimeniaceae, as in other Magnolidae with spiral floral phyllotaxy, there is neither a separation into perianth members and bracts toward the floral periphery nor a clear differentiation of two different kinds of perianth organs—sepals and petals (see also Hiepko, 1965; Endress, 1980a, 1980b). ANDROECIUM. Stamen structure 1s unremarkable. The stamens have neither the valvular dehiscence and paired appendages typical of the Laurales nor a massive construction as in many Magnoliales; they are differentiated into fil- ament and anther. The shape of the long, slender anthers and thin filaments of some 7rimenia species 1s unusual and resembles that of many wind-polli- nated groups in various angiosperm families (e.g., Eupteleaceae; cf. Endress, 1969). The Trimeniaceae differ from the Chloranthaceae (Vijayaraghavan, 1964), the Lauraceae (Bhandari, 1971), the Monimiaceae (Sampson, 1969a, 1969b; Bhandari, 1971), and most other members of the woody Ranales in having a middle wall layer only one or two cells in thickness. In having a glandular tapetum the Trimeniaceae resemble the Calycanthaceae (Bhandari, 1971), the Chloranthaceae (Vijayaraghavan, 1964), Cassytha of the Lauraceae (Sastri, 1963), and the Monimiaceae. This contrasts with other members of the Laura- 1983] ENDRESS & SAMPSON, TRIMENIACEAE 467 TABLE 1. Similarities between four species of Trimeniaceae. Species Trimenia T. weinmann- T. neocale- Piptocalyx Character papuana tifolia donica mooret Hasit Tree Vine TEPAL NUMBER =e sill Nese APEX Not depressed Depressed LEAF MARGINS thed Entire INFLORESCENCE TYPE a ea botryoids (panicles) |\Simple botryoids FLOWER BUD SHAPE Elongat lobular EPALS VASCULARIZATION One bundle Bundle branching INNERMOST, SHAPE Spathulate ovate OUTER, SHAPE \Markedly hypopeltate Weakly hypopeltate ANTHER DEHISCENCE Latrorse Extrorse POLLEN Polyforate or inaperturate Disulculate ceae, which have an amoeboid tapetum (Sastri, 1963). Successive cytokinesis, recorded in microspore mother cells of the Trimeniaceae, has also been found in some Monimiaceae (Sastri, 1963; Sampson, 1969a, 1977) and in the Lau- raceae (Bhandari, 1971); simultaneous cytokinesis has been found in another member of the Monimiaceae (Sampson, 1969b) and in the Calycanthaceae (Bhandari, 1971) and Chloranthaceae (Vijayaraghavan, 1964). As is charac- teristic of the more primitive groups of angiosperms (Brewbaker, 1967), the Trimeniaceae shed pollen in the two-celled condition, although three-celled pollen has been found in one genus of the Lauraceae (Gardner, 1974) and in one member of the Monimiaceae (Sampson, 1969b). GYNOECIUM AND FRUIT. The mature carpel is totally utriculate up to the stigma. This was inferred by Leinfellner (1969) from the description of the vasculature in Money, Bailey, and Swamy (1950) and is confirmed in this study. However, the vasculature of the carpel at anthesis is more complicated in Trimenia papuana and the two other species investigated than was described by Money, Bailey, and Swamy. F ee they did not report the presence of a vascular bundle in the outer integument. A remarkable See feature is that in spite of the fully utriculate condition of the carpel at anthesis, the early developmental stages are relatively less utriculate and thus more closely resemble those of the Lauraceae and some Monimiaceae (Sampson, 1969a; Endress, 1972, 1980a). The Trimeniaceae are therefore one more example of a family in the Magnoliidae with extremely utriculate carpels (see, for example, Endress, 1980b he ovule of the Trimeniaceae has several peculiar lauralian traits that it shares especially with certain Lauraceae, Hernandiaceae, and Calycanthaceae, but (interestingly) less with the Monimiaceae. However, only a few represen- tatives of these families have been studied so far (see TABLE 2, and Endress, 1972), and it is not known how widely distributed these traits are within the Laurales. Such features are the differentiation of many meiocytes in the nucellar base, the marked elongation of the nucellus after meiosis, the apical depression of the nucellus, and the tubelike, intrusive growth of the embryo sac toward TABLE 2. Occurrence in related families of characters found in the Trimeniaceae.* RELATED FAMILIES CHARACTER OF TRIMENIACEAE Lauraceae Chloranthaceae Hernandiaceae Monimiaceae, s.1. Climber (Piptocalyx) (Cassytha) — (+) (Palmeria) Leaves opposite (+) + _ + Node unilacunar, 2-trace + 1 = = Mucilage cells present (L) + — (+) - Anemophily = (+) oa (Hedycarya, pp.) Floral phyllotaxy spiral ae - (+) Stamen without appendages = ~ (+) (+) Anther dehiscence longitudinal - + — (+) Tapetum glandular (Cassytha) + ? _ Pollen meiosis successive + _ ve (+) Pollen inaperturate (7rimenia, p.p.) + (+) + + (Walker, 1976) (L) Exine tectate-columellate (Sampson & — + — (Laurelia) Endress, in press; Kubitzki, 1981) 89P WOLAYOUUV ATONUV AHL JO TYVNANOL p9 “10A] Carpel | per flower Carpel extremely utriculate at anthesis Carpel development including chairlike sta e Ovule | per carpel, in ventral-median position (L) Ovule anatropous, pendent (L) Outer integument with vascular bundle (Van Heel, 1971) ee palisade-shaped cells (Corner, 1976) Fruit: i erry Nucellus elongated (L) Nucellar apex depressed (Piptocalyx) Meiocytes several per ovule (L) Embryo sac narrow, tubular (L) Chromosome number v = 8 (Goldblatt, 1974; Goldblatt & Briggs, 1979; Ehrendorfer, 1976; Raven, 1976) +++ (Cassytha and other genera) (+) + (Cassytha) (Cassytha) + BE at (Xymalos) (ir) a _ + 2 (Monimioideae) ~ - ~ - + (Monimioideae) = (Hernandia) _ — (+) — = (Gyrocarpus) = > ? (Siparuna) — +(?) (Siparuna) (Hedyosmum) = = *References are not given for characters reviewed in Money, Bailey, and Swamy (1950) and Endress (1972) or ene elsewhere in this paper. Explanation of symbols: L, characteristic “lauralian” feature; +, present; (+), sometimes present; —, absent; ?, unknow AVAOVINAWNIUL ‘NOSAIWNVS 8 SSHUANA [£861 69P 470 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 the nucellar apex (and eventually into the micropyle in other Laurales). These features seem to be less prominent in the Magnoliales. The fruits of the Trimeniaceae have variously been described as berries (e.g., Rodenburg, 1971) or drupes (e.g., Morat & MacKee, 1977). This may be partly because the term “berry” is not always used in the same sense, but it is also due to misinterpretation of the fruit. The fruit contains a hard layer that is grooved or ridged on the outer surface in some species. Although the hard layer looks like the endocarp of a drupe, it clearly corresponds to the periphery of the seed. The fruit is, therefore, a one-seeded berry. SYSTEMATIC CONCLUSIONS RELATIONSHIPS WITHIN THE FAMILY. 7rimenia and Piptocalyx were last com- pared by Gilg and Schlechter (1923). According to their key, Trimenia has 15 to 20 spiral tepals, Piptocalyx 4 to 6 biseriate ones. The present study has shown that the two genera are less distinctive in these features. Both have spiral tepals, and in rare cases Piptocalyx may have as many as | 1, therefore reaching the lower limit of the tepal range of 7. papuana, as given by Rodenburg (1971). Another surprising result is that Trimenia neocaledonica shares some fea- tures with Piptocalyx that it does not share with the other Trimenia species. In particular, the discovery that the pollen of 7. neocaledonica 1s disulculate as in Piptocalyx (not polyforate or inaperturate as in the other Trimenia species) is of interest in this context (Sampson & Endress, in press). Thus, 7. neo- caledonica clearly occupies a central position within the family. This is shown by TABLE It seems, therefore, that the only obvious differences left between Trimenia and Piptocalyx are growth form (tree vs. vine) and tepal number (= 11 vs. From these results three possibilities for a classificatory consequence arise: the transfer of Trimenia neocaledonica to the genus Piptocalyx, the merger of Piptocalyx with Trimenia, or no change. The justification for following any of these three possibilities is about equal. We prefer the third possibility because it avoids unnecessary nomenclatural changes. On the whole, the Trimeniaceae appear to be a rather uniform family. Tyi- menia neocaledonica is possibly the most primitive species. RELATIONSHIPS WITH OTHER FAMILIES. Until the classic study on the Monim- iaceae by Money, Bailey, and Swamy (1950), Trimenia and Piptocalyx were usually included in that family. Money, Bailey, and Swamy discussed the rea- sons for segregating these two genera into a separate family (following Gibbs, 1917) and found relationships with the Lauraceae, Austrobaileya, and the Chloranthaceae, in addition to the Monimiaceae. Thorne (1976) pointed to rather close relations with the Chloranthaceae; Takhtajan (1980), with the Amborellaceae. Dahlgren (1980) listed the Trimeniaceae between the Austro- baileyaceae and the Monimiaceae, and Cronquist (1981) placed them between the Amborellaceae and the Monimiaceae. The results of our study underline the isolated position of the Trimeniaceae. The family shows relationships with the Lauraceae (especially the Cassythoi- 1983] ENDRESS & SAMPSON, TRIMENIACEAE 47] deae), the Chloranthaceae, and the Monimiaceae, but it is only distantly related to the Austrobaileyaceae. The Amborellaceae are still not well enough known for a comparative evaluation. The Hernandiaceae also share a number of special characters with the Trimeniaceae. Those of particular interest are listed in TABLE 2. That the Trimeniaceae belong to the Laurales is shown by some characteristic “Jauralian” features (see TABLE 2). The core of the Laurales consists of the Monimiaceae, Lauraceae, Hernandiaceae, and Gomortegaceae. However, several characters deviating from those generally found in the Laurales point to the isolated position of the Trimeniaceae. These characters include absence of a floral cup; spiral floral phyllotaxy—also present in Go- mortegaceae and some Monimiaceae; caducous tepals; stamens without flaps and appendages—also present in some Monimiaceae; extremely utriculate car- pels; polyforate pollen; tectate-columellate exine—also present in Laurelia; capitate stigma with multicellular papillae; outer integument with vascular bundle—also present in Hernandia; chromosome number of 1 = 8. Although the Trimeniaceae share a few of these characters with the Magnoliales, their affinities to the Laurales are still stronger, in spite of their isolation. A closer relationship with the Austrobaileyaceae, as supposed by some authors, can be excluded because the common shared features (vine, leaves opposite, node unilacunar and two-trace, floral phyllotaxy spiral, carpels extremely utriculate, outer integument with vascular bundle) are too scattered. The Austrobailey- aceae are of clearly magnolialian (not lauralian) affinity (Endress, 1980b). Thus the Trimeniaceae are a very homogeneous, isolated relict family of the Laurales with closest relationships to the Lauraceae, Monimiaceae, Chloran- thaceae, and Hernandiaceae; at the same time they are somewhat distant in certain characters from the core of the Laurales. — ACKNOWLEDGMENTS The senior author wishes to thank the Georges-und-Antoine-Claraz Schen- kung for financial support for the trips to Papua New Guinea, New Caledonia, and Australia. Thanks are also due to the late Dr. J. L. Gressitt (Ecology Institute, Wau, Papua New Guinea), Dr. P. Morat (O.R.S.T.O.M., Noumea, New Caledonia), Dr. G. McPherson (Nouméa, New Caledonia), P. Burgess (C.S.LR.O., Coffs Harbour, N.S.W.), and D. B. Foreman (University of New England, Armidale, N.S.W.) for their support during field work. We thank Dr. J. S. Womersley, Lae, Papua New Guinea, and S. Sivatibau, Fiji, for providing fixed floral material. Some of the microtome sections were prepared by Mrs. R. Jacob and Mr. A. Verde, University of Ziirich, and their assistance is grate- fully acknowledged. LITERATURE CITED BaLcooy, M. M.J. vAN. 1975. Pacific plant areas 3: 262, 263. Rijksherbarium, Leiden. BEADLE, N. C. W. 1972. Students flora of north eastern New South Wales. Vol. 2. University of New England, Armidale, N.S.W. 472 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 BHANDARI, N. N. 1971. Embryology of the Magnoliales and comments on their rela- tionships. Jour. Arnold Arb. 52: 1-39, 285-304. BrewBakeR, J. L. 1967. The distribution and phylogenetic significance of binucleate and frinucleaté pollen grains in the angiosperms. Am. Jour. Bot. 54: 1069-1083. Brown, F. B. H. 1935. Flora of southeastern Polynesia. Il. oe Bishop Mus. CORNER, E J. H. 1976. The seeds of dicotyledons. Vols. 1, 2. Cambridge University Press, Cambridge, England. Cronauist, A. 1981. An integrated system of classification of the flowering plants. Columbia University Press, New York. DAHLGREN, R. M. T. 1980. A revised system of classification of the angiosperms. Bot. ur. Linn. Soc. 80: 91-124. aay As F. 1976. Evolutionary significance of chromosomal differentiation pat- in gymnosperms and primitive angiosperms. Pp. 220-240 in C. B. Beck, ed., Origin and early an of angiosperms. Columbia University Press, New York. Enpress, P. K. 1969. Gesichtspunkte zur systematischen Stellung der Pupieleaceen (Magnoliales). Ber. Schweiz. Bot. Ges. 79: 229-278. —.. 1 . Zur vergleichenden Entwicklungsmorphologie, Embryologie und Sys- tematik bei Laurales. Bot. Jahrb. 92: 331-428. 1975, aera Formbeziehungen mit Hillfunktion im Infloreszenz- und Bliitenbereich. Ibid. 9 4. 1980a. Floral nee and relationships of Hortonia (Monimiaceae). Pl. Syst. Evol. 133: 199-221. 1980b. The reproductive structures and systematic position of the Austro- baileyaceae. Bot. Jahrb. 101: 393-433. GarpneR, R. O. 1974. Trinucleate pollen in Beilschmiedia Nees (Lauraceae). New Zealand Jour. Bot. 12: 243, 244. Gisss, L. S. 1917. A contribution to the phytogeography and flora of the Arfak moun- tains etc. Taylor & Francis, London GILG, E., & R. SCHLECHTER. 1917. Ueber zwei pflanzengeographisch interessante Mo- nimiaceen aus Deutsch-Neu-Guinea. Bot. Jahrb. 55: 195-201. & 23. Die Monimiaceen-Gattung /denburgia. Ibid. 58: 244-248. GOLbDBLATT, P. 1974. 7 to the knowledge of cytology in Magnoliales. Jour. mold Arb. 55: 453-4 G. Briccs. a Chromosome number in two primiti ts, Xymalos monospora (Monimiaceae) and Piptocalyx moorei (Trimeniaceae). Ann. Missouri Bot. Gard. 66: 898, 899. Hee., W. A. van. 1971. The labyrinth seed of Hernandia peltata Meissn. in DC. Proc. Kon. Nederl. Akad. Wet. Amsterdam, ser. C. 74: 4 HeEs.top-HArRIson, J. 1966. Cytoplasmic continuities dune spore formation in flow- ering plants. Endeavour 25: 65-72. at Hutcuinson, J. 1973. The families of flowering plants. Clarendon Press, Oxford. Huynu, K.-L. 1972. The original position of the generative nucleus in the pollen tetrads of Agropyron, Itea, Limnanthes, ae Onosma, and its phylogenetic significance in the angiosperms. Grana 12: 105- i! 2. KANEHIRA, R., & S. Hatusima. 1942. The Kanehira-Hatusima 1940 collection of New i 262. Kusirzki, K. 1981. The tubular exine of Lauraceae and Hernandiaceae, a novel type of exine structure in seed plants. Pl. Syst. Evol. 138: 139-146. a W. 1969. Uber die Karpelle verschiedener es ere VIII. Uberblick alle Sie der Ordnung. Osterr. Bot. Zeitschr. 117: ve Mine J.H.,&R. T. BAKER. 1895. Botanical notes from the ee Museum. Proc. ae Soc. New S. Wales, II. 10: 512-518 1983] ENDRESS & SAMPSON, TRIMENIACEAE 473 Money, L. I., I. W. BaiLey, & B. G. L. Swamy. 1950. The morphology and relationships of the Montanacese: Jour. Arnold Arb. 31: 372-404. Morar, P., & H.S. MacKee. 1977. Quelques précisions sur le Trimenia neocaledonica Bak. f. et la famille des Triméniacées en Nouvelle-Calédonie. Adansonia, IJ. 17: 205-213. O.iverR, D. 1895. Piptocalyx moorei Oliv. In: W. J. Hooker, ed., Ic. Pl. IV. 4: pl. 2367. Perkins, J. 1911. Monimiaceae (Nachtrage). Jn: A. ENGLER, ed., Pflanzenr. IV. 101(Heft 49): 1-67. —. 1925. Ubersicht iiber die Gattungen der Monimiaceae. Engelmann, Leipzig. & E. Gitc. 1901. Monimiaceae. Jn: A. ENGLER, ed., Pflanzenr. IV. 101(Heft 4): 1-122. RAVEN, P. H. 1976. re bases of angiosperm phylogeny: cytology. Ann. Missouri Bot. Gard. 62: 724-7 RENDLE, A. B., E. G. oo & S. L.M. Moore. 1921. A systematic account of the plants collected i in New Caledonia and the Isle of Pines by Prof. R. H. Compton in 1914—part I. Flowering plants (angiosperms). Jour. Linn. Soc. Bot. 45: 245-417. RopeNBuRG, W. F. 1971. A revision of the genus 7rimenia (Trimeniaceae). Blumea 19: 3-15. Sampson, F. B. 1969a. Studies on the Monimiaceae. I. Floral morphology and ga- metophyte development of Hedycarya arborea J. R. et G. Forst. (subfamily Mon- eae Austral. Jour. Bot. 17: 403-424. 1969b. Studies on the Monimiaceae. III. Gametophyte development of Laurelia novae-zelandieae A. Cunn. (subfamily Atherospermoideae). Ibid. 17: 425-439. 1977. Pollen tetrads of Hedycarya arborea J. R. et G. Forst. (Monimiaceae). Grana 16: 61-73. ——. 1982. Variation 2 ea of the nascent g tive cell in pollen of Hedycarya (Monimiaceae). [bid. P. K. ENDRESS. i morphology i in the Trimeniaceae. Grana (in press). SastrRi, R. L. N. 1963. Studies in ce Lauraceae. IV. Comparative embryology and phylogeny. Ann. Bot. n.s. 27: 4 SEEMANN, B. 1871. Flora Vitiensis. are Lon Smitu, A. C. 1978. A precursor to a new flora of Fiji i. Allertonia 1: 331-4 —. 1981. Flora Vitiensis nova. Vol. 2. Pacific Tropical Botanical ee Lawai, Kauai, Hawaii. TAKHTAJAN, A. 1980. Outline of the classification of flowering plants (Magnoliophyta). Bot. Rev. 46: 225-359. THORNE, R. F. 1976. A phylogenetic classification of the Angiospermae. Evol. Biol. 9: -106. TROLL, W. 1964. Die Infloreszenzen, 1. G. Fischer, Stuttgart. VIJAYARAGHAVAN, M. R. 1964. Morphology and embryology of a vesselless dicotyle- don— Sarcandra irvingbaileyi Aor and systematic position of the Chlorantha- ceae. Phytomorphology 14: 429-44 WALKER, J. W. 1976. Comparative ick morphology and phylogeny of the ranalean complex. Pp. 241-299 in C. B. Beck, ed., Origin and ~~ evolution of angiosperms. Columbia University Press, New York and London .K. E. F. B.S. INSTITUT FUR SYSTEMATISCHE BoTa- Botany DEPARTMENT NIK DER UNIVERSITAT VICTORIA UNIVERSITY OF ZOLLIKERSTRASSE 107 WELLINGTON 8008 ZURICH,

1983] ROLLINS, CRUCIFERAE 497 Ficure 2. Arabis tiehmii (Ziehim 756/), showing elongated, thickened caudex with old leaf-bases. Type. Nevada, Washoe Co., Sierra Nevada, Carson Range, ridge N of Mt. Rose, ¥4 air mi NNW of peak. near rock outcrops on decomposed granite, 19 Aug. 1982, 4. Tiehm 756] (holotype, GH; isotypes to be distributed). Perennial: caudex simple or closely branched, thickened with old leaf-bases. Basal leaves tufted, erect, petiolate, 1.5-2.5 cm by 4-6 mm; blade spatulate to oblanceolate, acute and often apiculate with large simple trichome, entire, grayish green, 1-nerved from base to apex. glabrous except for marginal forked or simple trichomes on petiole margins. Cauline leaves 3 to 5, sessile, oblong, 8—12 mm long, acute, the lower ones cuneate at base, glabrous or with occasional 498 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 trichomes along margin toward base, the upper ones cuneate at base or some- times with minute auricles. Stems slender, erect to somewhat decumbent, often flexuose, 8-17 cm long, glabrous. Fruiting pedicels straight, divaricately as- cending to erect, slender, 3-5 mm long, glabrous; siliques erect to slightly divaricate, flattened parallel to septum, 16-22 by |.5-2 mm, acute, glabrous, l-nerved below; style barely evident to less than 0.5 mm long. Seeds oblong, plump, wingless on sides but with abbreviated wing distally, ca. 2 by | mm: cotyledons accumbent. In growth habit, Arabis tiehmii is more similar to A. davidsonii E. Greene than it is to any other species of Arabis L. The caudex is similarly thickened with old leaf-bases but is not nearly as heavy as in that species (see FiGure 2). In such characters as the tufted basal leaves, thickened caudices, flexuose stems, and ascending siliques, 4. tiehmii is much like Smelowskia holmegrenii Rollins. This resemblance prompted me to examine S. ho/mgrenii again, with the thought that it might be better placed in 4rabis than in Smelowskia C. A. Meyer. The nearly terete siliques, incumbent cotyledons, and unguiculate pet- als, however, preclude it from the genus 4rabis. Since these are features of Smelowskia, 1t seems best to leave S. holmgrenii in Smelowskia even though it is clearly not closely related to other known species of the genus. In a comparison of Arabis tiehmii and A. davidsonil, differences in the length of the siliques, the length of the basal leaves, and the nature of the seeds stand out. Both the siliques and the basal leaves of 4. davidsonti are more than twice as longas those of 4. tiehmii. Also, the petiole of the basal leaves of A. davidsonii is long, while that of A. t/ehmii is scarcely defined, with the blades merely tapering to the point of insertion. The seeds of 4. tiehmii are narrowly oblong, with an abbreviated thickish wing only at the distal end: those of 4. davidsonii are broadly oblong to orbicular and narrowly winged all around. Arabis davidsonii was at one time confused with 4. /yallii S. Watson and was even made a variety of 1t by Smiley (1921). However, neither 4. davidsonii nor 4. tlehmii is considered to be closely related to A. /yallii. Now that there are two known species having much-thickened caudices with numerous leaf- bases (see FiGurE 2) (vs. the relatively slender, nearly naked caudex branches of A. /yallil), it is easy to see the evolutionary trend that resulted in these distinctive structures. With 4. tiehmii as a near relative, 4. davidsonii is no longer a somewhat anomalous species in the genus Arabis. J. W. Congdon C10 (Gu) is a fragmentary specimen, consisting of one fertile stem, a separate infructescence, and a few basal leaves, which I previously referred doubtfully to Arabis davidsonii. It was collected at Mt. Warren Pass, Mono Co., California, on 21 August 1894. The siliques are short like those of A. tiehmii, and the seeds are oblong with more of a wing. I believe this specimen is closer to A. tiehmii than to A. davidsonti, but there is still room for doubt because of the inadequacy of the material. Arabis rigidissima Rollins var. demota Rollins, var. nov. Herba perennis; siliquis valde |-nervatis, obtusis, 4-6 cm longis, ca. 3 mm latis: seminibus late oblongis vel prope orbicularibus, alatis. 1983] ROLLINS, CRUCIFERAE 499 Type. Nevada, Washoe Co., Sierra Nevada, Carson Range, Galena Creek, TI7W, R19E, $17, 1.5 air mi ESE of Mt. Rose peak, 7900 ft alt., rocky areas at the edge of aspen groves, 26 Aug. 1982, A. Tiehm 7572 (holotype, GH; isotypes to be distributed). Stems one or few from branched or simple, ligneous or subligneous base, usually branched, stiff, with branches rigidly ascending. Leaves narrowly pet- iolate: blade oblanceolate, pubescent with 3 or 4 branched dendritic trichomes, tufted. Pedicels and siliques divaricately ascending, glabrous; siliques few, re- mote, straight to slightly curved, 4-6 cm by ca. 3 mm, obtuse at apex, the valves veiny, strongly l-nerved nearly to apex, margins slightly uneven, the style absent or very short. Seeds in single row, flattened, winged, broadly oblong to nearly orbicular, 2.5-3 mm long or in diameter, wing ca. 0.5 mm wide; cotyledons accumbent. True Arabis rigidissima has only rarely been collected, and to my knowledge it is restricted to Trinity and Humboldt counties in California, 200 or more miles northwest of the site where var. demota was found. The differences between var. rigidissima and var. demota are minor but consistent and show that the populations have diverged evolutionarily enough to merit nomencla- tural recognition. In var. rigidissima there is a definite style present and the siliques are acuminate toward the apex, while in var. demota there is no style or only the barest suggestion of one and the silique apex is obtuse or at most acute. In var. demota the valves are prominently veined and a strong nerve beginning at the base reaches to near the tip, while in var. rigidissima the valves hardly show any veininess and the much-less-distinct nerve reaches only just above the middle. The seeds of var. rigidissima are more nearly orbicular, and the winging surrounds the entire seed; in var. demota the seeds are usually broadly oblong with the winging only at the distal position and along one side. Descurainia torulosa Rollins, sp. nov. Herba multicaulis, humifusis; caulibus tenuibus, procumbentibus, dense pu- bescentibus, 6—10(-13) cm longis; foliis radicalibus petiolatis, lobatis, 2-3 cm longis; foliis caulinis does reductis: floribus minutis; pedicellis fructiferis ap- pressis, erectis, 1.5-2.5 mm longis; siliquis linearibus, teretibus, dense pubes- centibus, erectis, torulosis, 8—12(—15) mm longis, ca. | mm latis; stylis tenuibus, ca. 0.2 mm longis; seminibus anguste oblongis, exalatis, ca. 1.3 mm longis, ca. 0.5 mm latis; cotyledonibus incumbentibus. Type. Wyoming, Fremont Co., Wind River Range, near Brooks Lake, rocky slopes at the base of cliffs 2 mi NW of the lake, 10,000 ft alt., 8 July 1966, K. W. Scott 761 (holotype, GH). Biennial or possibly perennial with well-developed taproot and unbranched crown; pubescent throughout with dendritic, few- to many-branched trichomes, stems several to numerous, procumbent, arising from dense cluster of crown 500 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 leaves, simple or branched, 6—! 0(—13) cm long. Crown leaves with petiole short: blade pinnately lobed, 2-3 cm long. primary lobes simple or with | or 2 subsidiary lobes. Cauline leaves few, similar to crown leaves. Flowers minute; sepals oblong, nonsaccate, slightly more than | mm long, with hyaline margin all around, pubescent; petals spatulate, ca. 1.5 mm long, whitish. Fruiting pedicels erect, appressed to rachis, 1.5-—2.5 mm long; siliques erect and ap- pressed to rachis, curved outward to nearly straight. terete. linear. torulose, tapered below and toward apex, 8-1 2(-—15) mm long, slightly more than | mm in diameter, the style slender, ca. 0.2 mm long. smaller in diameter than stigmatic area, usually pubescent. Seeds in single row, narrowly oblong, plump. ca. 1.3 mm long, wingless, ca. 0.5 mm in diameter, dark brown; radicle slightly longer than cotyledons; cotyledons incumbent. As Detling (1939) pointed out, there are two principal growth types found in North American Descurainia Webb & Berthelot: one in which the single stem branches only well above the middle, resulting in a wandlike appearance: the other with branching beginning near the base of the single stem and with the numerous erect branches nearly equal to the leader, giving a bushlike aspect to the plants. A third growth type, found in a number of South American species, 1s characterized by several to many more or less equal stems arising from the crown of a taproot. The stems are erect in some of these species and procumbent in others. Several of the South American species are definitely perennial, whereas all heretofore-known North American species are annuals or biennials. In habit, Descurainia torulosa, unlike any known North American species, shares the characteristics of those South American species with pro- cumbent stems. This species 1s probably perennial rather than biennial: the taproot with its crown 1s well developed. and the available specimens look as though they could be perennial, but one cannot be sure from the specimens alone. In addition to its distinctive branching habit, D. toriu/osa grows so low that 1t does not resemble any other North American species. Also, the closely appressed but flaring torulose siliques and the extremely short pedicels char- acteristic of D. toru/osa are not matched in material of other species. Draba hitchcockii Rollins, sp. nov. Herba perennis, caespitosa; caudicibus ramosis; caulibus scapiformibus. erectis, hirsutis, (2-)3-7(-9) cm _ altis: foliis rosulatis, spathulatis vel oblongis, cilatis, (3—)4-7(-10) mm longis, (1—-)1.5—2 mm latis: sepalis late oblongis, hirsutis, ca. 3 mm longis; petalis late obovatis. albis. 5-6 mm longis. ca. 4 mm latis: pedicellis rectis, hirsutis, adscendentibus, (4-)6-10(-12) mm longis; siliquis erectis, ellipticis vel late oblongis, pubescentibus (4—)5—7(-8) mm longis, (3-)3.5-4.5(-5) mm latis; stylis 1.5-2 mm longis; seminibus ob- longis, ca. 1.8 mm longis; cotyledonibus accumbentibus. Type. Idaho, Butte Co., S Lost River Range, T7N, R26E, $17, N side of Elbow Canyon, 23 May 1982, 4. F. Cholewa 857 (holotype, GH: isotypes. 1p, NY, and others to be distributed). 1983] ROLLINS, CRUCIFERAE 501 Perennial; caespitose; caudex loosely branched but forming clusters up to 1 dm in diameter; branches covered with remnants of old leaves and leaf-bases, terminating in dense, recent leaf-clusters. Leaves nonpetiolate, spatulate to oblong, (3-)4-7(-10) by (1-)1.5-2 mm, obtuse, |-nerved, ciliate on margins and upper surface with large, simple (rarely forked) trichomes, marginal tri- chomes often as long as leaf width. lower surface with coarse, forked to 4-branched trichomes. Scapes slender, erect or outer slightly decumbent at base, (2-)3-7(-9) cm tall, leafless, hirsute with mixture of long, simple or forked trichomes and shorter, many-branched ones; sepals broadly oblong, ca. 3 mm long, hirsute with simple or forked trichomes; petals broadly obovate, 5-6 by ca. 4 mm, pure white, blade sharply constricted to narrow claw; stamens sub- equal, filaments slightly dilated at base. Infructescences racemose, subcorym- bose to more elongated; pedicels straight, ascending to divaricately ascending, (4-)6-10(-12) mm long, hirsute; siliques erect, elliptic to broadly oblong or nearly ovate, (4-)5-7(-8) mm by (3-)3.5-4.5(-5) mm, densely pubescent with simple or forked trichomes, the styles 1.5-2 mm long, the ovules 4 to 6 in each loculus, the funiculi less than 0.5 mm long. Seeds oblong, plump, wingless, ca. 1.8 mm long; cotyledons accumbent. REI \TATIVE SPECIMENS. Idaho. Butte Co. Lost River Range: Anderson Canyon ca. 11 mi iN of Arco, S. & P. Brunsfeld 859 (Gu'), Arco Hills area, Reese 138, 147, 148 (GH); Elbow Canyon, Henderson 4959 (Gu); Elbow Canyon ca. 12 mi E of Mackay, : & P. Brunsfeld 810 (Gu, Ny); Cedarville Canyon, north fork, S. & P a unsfeld 1102 (Gu); E slope just above bottom of Jaggles Canyon, ca. 14 mi above entrance, Henderson 4993 (Gu); Maddock Canyon, S. & P. Brunsfeld 821 (Gu, Ny), aie 4979 (GH of Sands Canyon, B. Anderson 29 (Gu). Custer Co.: Lost River Range, Waddoup’s Canyon, Reese 172 (Gu). This species was named for C. L. Hitchcock, monographer of western North American Draba and long-time student of the flora of the Pacific Northwest. Although Draba hitchcockii has white flowers, it is undoubtedly most closely related to the yellow-flowered D. paysonii Macbr. var. paysonii, which occurs in Montana and Wyoming. It also differs from that species in having longer styles, more uniformly large simple trichomes on the leaf margins and the upper leaf-surfaces, and usually longer, more slender scapes. Draba hitchcocki differs even more from D. paysonii var. treleasii (Schulz) C. L. Hitche. The latter is more widespread than var. paysonii and occurs in the general area of Idaho where D. hitchcockii is found, but usually at higher elevations. Although flower color is of little or no taxonomic significance in many genera of the Cruciferae, it isan important taxonomic character in Draba L. When correlated with other features, it can be utilized to distinguish taxa at the specific and infraspecific levels. The fruits of Draba hitchcockii and D. paysonii var. paysonii are similar in size and in the pubescence on the valve surfaces, with the trichomes either simple (especially on the silique margins) or a combination of simple and orked. In some plants of D. hitchcockii, only simple trichomes are found on \Most, if not all, of the collections cited are probably also represented at the University of Idaho Herbarium (1p), but this has not been verifie 502 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 the entire silique. The shape of the siliques is different in the two taxa: in D. hitchcockti they are usually elliptic (rarely narrowly elliptic or nearly broadly ovate); In var. pavsonil they are always broadly ovate. I have not seen any material of var. paysonil with anything other than a short subcorymbose in- fructescence, whereas in D. hitchcockii the infructescence is almost always racemose and 1s often somewhat elongated. The short stout scape and strongly condensed habit in var. paysonii contrast with the slender elongated scapes and the loose habit in D. hitchcockii. I am indebted to Dr. Douglass Henderson. of the University of Idaho, for the following information on the plant associates and habitat of Draba hitch- cockii. All populations are on thin soil in rocky limestone areas that range from steep to gently sloping. ee species include Leucopoa kingii (S. Watson) W. A. Weber, 4gropyron spicatum (Pursh) Scribner & Sm., Cerco- carpus ledifolius Nutt. var. intercedens C. Schneider, Pseudotsuga menziesii (Mirbel) Franco, Artemisia tridentata Nutt.. 4. arbuscula Nutt. var. nova (A. Nelson) Cronq., Phlox hoodii Richardson, Draba oreibata Macbr. & Payson. and D. densifolia Nutt. Cercocarpus and Artemisia are the most important woody genera. Although Drahba hitchcockti occurs at elevations from 5500 to ca. 7200 ft, most populations grow between 6000 and 6900 ft. Draba pennellii Rollins, sp. nov. Herba perennis, caudicibus ramosis; foliis radicalibus, pubescentibus, ob- longis, acutis, 5-8 mm longis, ca. 2 mm latis: foliis caulinis 2-5, sessilibus, pubescentibus; caulibus tenuis, erectis, pubescentibus. 4-7 cm altis; pedicellis rectis, divaricatis, pubescentibus, 5-8 mm longis: sepalis oblongis. pilosis; pe- talis albis, obovatis, 4.5-6 mm longis: siliquis anguste ovatis, pubescentibus, ca. 5 mm longis; stylis ca. 2 mm longis; seminibus ignotis. TYPE. Nevada, White Pine Co., Shell Creek Range, at head of South Fork of Berry Creek, on rock aoe 10,000 to 10,500 ft alt.. 13 July 1938, FLOW. Pennell 2 R. L. Shaeffer, Jr. 22977 (holotype, Gu). Perennial with loosely branching caudex; caudex branches elongated, par- tially covered with old leaves and leaf-bases, both sterile and fertile branches present in each clump; leaf-clusters elongated. Leaves nonpetiolate. oblong, 5— 8 by ca. 2 mm, acute, entire, 1-nerved, densely covered with stalked, 4- or 5-branched, coarse trichomes, upper surface sometimes nearly glabrous. Flow- ering stems slender, simple, erect, 4-7 cm tall, densely pubescent with spreading dendritic, forked, or occasionally simple trichomes, simple trichomes increas- ing in frequency toward inflorescence; pedicels divaricately ascending, straight. densely pubescent with coarse, mostly dendritic and forked but occasionally simple trichomes; sepals oblong, ca. 3 by ca. 1.2 mm, greenish then turning purplish and with hyaline margin in age, sparsely covered on exterior with coarse, simple or forked trichomes; petals obovate, 4.5-6 by 3-4 mm, abruptly narrowed to short claw less than | mm long, truncate to shallowly retuse, white; stamens subequal, filaments dilated toward base. Siliques flattened, ovate to nearly oval, sometimes slightly asymmetric, ca. 5 by ca. 3 mm, moderately 1983] ROLLINS, CRUCIFERAE 503 pubescent with minute forked (or simple) trichomes; styles ca. 2 mm long. Mature seeds not available. SPECIMENS EXAMINED. Nevada. WHITE PINE Co.: 20 mi N of Ely, North Shell Peak, 11,800 ftalt., Lavin 4209 (Gu); Shell Creek Range, Steptoe Creek road, 9.3 m1 Eof U.S. Highway 93, H. K. Sharsmith 4823 (Gu). Because cauline leaves are present on the flowering stems of both species, it is tempting to suggest that Draba arida C. L. Hitche. is the closest known relative of D. pennellii. However. D. arida has rosulate basal leaves, closely branched or unbranched caudices, and only one or a few flowering stems per plant. Plants of D. pennellii are entirely different, consisting of numerous elon- gated caudex branches and basal leaves that are in elongated tufts. The number of flowering stems per plant ranges from several to many, but it is difficult to assess this feature accurately from specimens alone because the plants are tufted in cracks of rocks, making collection of a complete plant difficult. Collectors usually break off individual caudex branches. The habit of growth is much like that of D. sphaeroides Payson, and I am inclined to associate D. pennellii with that species. In both species there is stem elongation, forming well-defined internodes both between individual leaves and between tufts of leaves. Often new branches arise in the axils of individual leaves, and either these terminate in a cluster of sterile leaves, or a fertile branch arises at the terminal end of the cluster, forming an inflorescence toward its apex. The flowers of both D. sphaeroides and D. arida are yellow, while those of D. pennellii are pure white. The trichomes of all three species are fairly similar and cannot be relied upon as identifying characters. In D. pennellii the petal is distinctively shaped. with the broad, full blade narrowing abruptly to a very short claw. Neither in D. sphaeroides nor in D. arida is there a sharp differentiation into blade and claw: the blade narrows very gradually from the widest point at the apex to the point of insertion. As compared to Draba sphaeroides, D. pennellii has fertile stems with 2 to 5 leaves (vs. a leafless scape); white. obovate petals that are sharply differen- tiated into blade and claw (vs. yellow. spatulate. and gradually tapered from blade to claw); and dense pubescence extending up the fertile stems to include the pedicels of the infructescence (vs. the upper portion of the fertile stems and pedicels glabrous). In both species the siliques are either pubescent with minute forked or simple trichomes, or glabrous. Lesquerella goodrichii Rollins, sp. nov. FIGURE 3. Herba perennis; caudicibus rz is densis:; caulibus decumbentibus vel erectis, tenuibus, 8-14 cm longis: foliis radicalibus erectis. dense pubescentibus, ar- genteis, petiolatis, (1.5—)2—4(-6) cm longis. (3-)4—7(-10) mm latis: foliis caulinis integris, anguste spathulatis. cuneatis: sepalis viridis vel argenteis. pubescen- tibus, oblongis, nonsaccatis. ca. 5 mm longis, ca. 1.8 mm latis; petalis spathu- latis, aureis, 7-9 mm longis. 2.5-3 mm latis: pedicellis divaricatis. sigmoideis vel fere rectis, pubescentibus. 4-6 mm longis: siliquis erectis, compressis. cl- lipticis. 4-5 mm longis. ca. 3 mm latis: stylo 3-4 mm longo: loculis 2-4 ovulatis: seminibus immaturis exalatis: cotvledonibus accumbentibus. 504 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ae ¥ v Ss 16 Fy 1Q@KVY A256 166U 664 11683 MC2 Figure 3. SEM photos of trichomes of Lesquerella goodrichii (Goodrich 16951): above, spacing of trichomes on leaf-surface; below, enlargement of single trichome. 1983] ROLLINS, CRUCIFERAE 505 Tyee. Utah, Millard Co., Desert Experimental Range, Tunnel Springs Mts., 21 mi SE of Garrison, T24S, R17W, S9, steep slopes and ridges, mostly N ex- posure, 8000 ft alt., 16 June 1982, S. Goodrich 16951 (holotype, GH: isotypes, BRY, NY, UTC). Perennial, densely covered throughout (except on petals and styles) with appressed radiate trichomes; caudex closely branched, branches usually covered with old leaf-bases; stems several to many, slender, simple, mostly decumbent, 8-14 cm long. Basal leaves numerous, erect, (1.5—)2—4(-6) cm by (3-)4—7(-10) mm: petiole to 3 cm long, slender; blade oblanceolate and entire to deltoid with broad tooth on each side. Sepals oblong, nonsaccate, ca. 5 by ca. 1.8 mm. greenish to silvery; petals erect, spatulate, not differentiated into blade and claw, 7-9 by 2.5-3 mm, yellow. Fruiting pedicels divaricately ascending to widely spreading, sigmoid to nearly straight, 4-6 mm long; siliques erect, el- liptic, strongly compressed parallel to septum, 4-5 by ca. 3 mm, the valves densely pubescent on exterior. glabrous on interior, the septum entire, the replum elliptic, acute at apex, the style 3-4 mm long, slender, with expanded stigma, glabrous or with | or 2 trichomes near base; ovules 2 to 4 in each loculus, attached above middle of replum. Immature seeds wingless, often crowded and misshapen; cotyledons accumbent. SPECIMENS EXAMINED. Utah, MILLARD Co.: Desert Experimental Range, 20 mi SE of Garrison, T24S, R17W, $3, 7200 ftalt., Goodrich 16539 (BRY, GH); Canyon Mts., Lyman Canyon, 7.75 mi E of Oak City, T17S, R3W, S16, 8000 ft alt.. Goodrich 16883 (Bry, GH, NY, UTC), Goodrich 17949 (BRy). The strongly flattened siliques of Lesquerella goodrichit indicate a close re- lationship to L. occidentalis S. Watson, and these two species are indeed similar in many respects. However. the trichomes are decidedly different: in L. occi- dentalis they are stellate with radiating branches, have slender, elongated, very gradually tapering tips, and are free to their bases or nearly so; in L. goodrichii they have comparatively massive branches and branch bases, the tips contract abruptly to a point, and there is considerable fusion toward the center of each trichome. (See FiGure 3. For comparison, see plates 2 and /6 in Rollins & Banerjee, 1975—SEM pictures of trichomes of L. occidentalis subspp. occi- dentalis and cinerascens.) Plants of Lesquerella occidentalis are more robust and have thicker stems and longer infructescences than those of L. goodrichii, which are usually del- icate, with slender stems and narrow, short infructescences. The pedicels of L. goodrichii tend to be weakly sigmoid and erect or at least ascending, while those of L. occidentalis are strongly sigmoid and are more at right angles to the rachis or even descending. The long, slender petioles and relatively short blades of the numerous basal leaves of L. goodrichii are distinctive features of this species. It is with some hesitation that I have cited the two collections from the Canyon Mountains under Lesquerella goodrichii. Although these plants have the same trichome type as those from the Desert Experimental Range, the leaf- blades are much broader and of a different shape, and they usually have two broad teeth instead of being entire. The two localities are at some distance 506 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 from each other even though they are in the same county. Additional material with mature fruits is required from the Canyon Mountain area to see whether a minor taxon should be recognized under L. goodrichii. Lesquerella parviflora Rollins, sp. nov. Herba perennis multicaulis; caulibus procumbentibus vel adscendentibus, 1-3 dm longis; foliis radicalibus late obovatis, petiolatis, argenteis, sparse den- tatis vel integris, |-2 cm longis, 6-10 mm latis; foliis caulinis cuneatis, oblan- ceolatis vel oblongis, ea dame oblongis, nonsaccatis, dense pubescen- tibus, 3-3.5 mm longis, ’ m latis; petalis spathulatis, flavis, 5-6 mm longis, 2—2.5 mm latis; nedicellis Gee recurvatis, 6—8(-12) mm longis: siliquis ellipticis vel prope globosis, dense pubescentibus, 3-4 mm longis, 2.5- mm latis; stylo glabro, ca. 3 mm longo: loculis 2-ovulatis; seminibus late oblongis vel prope orbicularibus, 2.5—3 mm longis, ca. 2.5 mm latis:; cotyle- donibus accumbentibus. Type. Colorado, Rio Blanco Co., USGS Black Cabin Gulch Quadrangle, T3S, Seco S13, %4 mi S of spring in Tommy’s Draw, 2595 m alt., 21 July 1982, V. Baker & S. Sigstedt 82-308 (holotype, GH; isotype, cs). Perennial, silvery from dense covering of stellate trichomes; stems several to numerous, prostrate to ascending, simple or rarely branched above, 1-3 dm long. Basal leaves tufted, petiolate, 1-2 cm by 6-10 mm: blade broadly obovate, rounded to obtuse, entire or with | or 2 broad teeth, silvery. Cauline leaves oblanceolate to nearly oblong, cuneate at base, entire. Sepals oblong, nonsac- cate, 3-3.5 by ca. 1.5 mm, densely pubescent; petals erect, spatulate, scarcely differentiated into blade and claw, 5-6 by 2-2.5 mm, yellow; anthers nearly oval, ca. | mm long. Infructescences rather dense, '3—'4 of stem length; fruiting pedicels recurved, 6-8(—12) mm long, densely pubescent; siliques mostly pen- dent, elliptic to nearly globose, usually slightly compressed parallel to septum, 3-4 by 2.5-3 mm or ca. 3 mm in diameter, densely pubescent on exterior, the valves glabrous or with few scattered trichomes on interior, the styles ca. 3 mm long, noncapitate, glabrous, the ovules 2 per locule, attached toward apex of replum, the septum entire, the replum nearly oval to obovoid in outline. Seeds oval to slightly longer than broad, somewhat compressed, wingless, 2.5— 3 by ca. 2.5 mm: cotyledons broadly oblong to nearly orbicular, accumbent. SPECIMENS EXAMINED. Colorado. Rio BLANCO Co.: USGS Segar Mountain Soca as TIS, R96W, SI, 1 mi SSW of junction of Hay Gulch and White R., 1905 m alt., Baker & Nate 82-231 (cs, GH); USGS Philadelphia Creek Quadrangle, T1S, RIOOW. S21, 2.45 mi SSE of junction of West Fork Spring Creek and McDowell Sha 2316 malt.. Kelley a Naumann 82-155 (cs, GH); Cathedral Bluff Rim Road, Piceance Basin, lat. 39°46’36” N, long. 108°33’38” W, Peterson et a 1141 (cs), Painter et ri 132 (cs); USGS Jessup Gulch Quadrangle, T2S, R96W, S21, 2.2 mi NNE of junction of Collins ines and Piceance Creek, Wa/ker & Naumann 82- 213 (cs); T3S, RIOOW, S24, NE 4, 1.8 NE of confluence of Lake Creek and Cathedral Creek, Parachute Creek Member of Green River Formation, Wilken 13866 (COLO, CS, GH, RM). The recurved pedicels of Lesquerella parviflora are similar to those of L. 1983] ROLLINS, CRUCIFERAE 507 arenosa (Richardson) Rydb. and L. /udoviciana (Nutt.) S. Watson, and the siliques of these three species are somewhat alike. The most fundamental differences between L. parviflora and the other two species are in ovule number and in the position of the funiculi on the replum of the silique. In L. parviflora there are only two ovules per locule and the two funiculi are restricted to an area near the apex of the replum; in both L. arenosa and L. ludoviciana the number of ovules is variable, ranging from four to eight in each locule, and the funiculi are arranged not only toward the apex of the replum but down the sides as well. Although the trichomes of all three species are stellate, with four to six primary rays (each of which is forked or bifurcate to provide anywhere from ten to twenty free tips), those of L. parviflora are smaller, with shorter, more massive, more tapered ultimate branches and a greater amount of fusion toward the center of the trichome than is found in the other two species. Characteristically, the trichomes are spreading on the siliques of L. arenosa and L. /udoviciana, while they are closely appressed in L. parviflora. There 1s a sharp difference in the shape of the basal leaves: in L. arenosa and L. lu- doviciana the inner basal leaves are linear or nearly so with a scarcely differ- entiated petiole, but in L. parviflora all of the basal leaves are distinctly dif- ferentiated into a comparatively broad blade and a short petiole. A significant feature of Lesquerella parviflora is the presence in the inflores- cences of short shoots subtended by leafy bracts. This is particularly evident in Wilken 13866, where many of the specimens are in flower. The only other species of Lesquerella known to have short shoots in the inflorescences are L. argyrea (A. Gray) S. Watson subsp. diffusa (Rollins) Rollins & E. Shaw and L. peninsularis Wiggins, both of which occur exclusively in Mexico Among the specimens of the seven collections of Lesquerella par wore cited, there is considerable diversity, which is probably attributable in part to the habitat. Baker & Sigstedt 82-308 was collected on shale ledges, implying an exposed habitat. Quite compact in the caudex area, these specimens have prostrate stems, narrow, dense infructescences, and short pedicels. The siliques are uniformly elliptic in outline and slightly compressed parallel to the septum. The specimens of Baker & Naumann 82-231, collected “on shale with Ribes aureum and Pinus,’ show both decumbent and upright stems, looser infruc- tescences, slightly longer pedicels, and elliptic to subglobose siliques. Kelley & Naumann 82-155, ““common on slopes of canyon with Acer, Prunus, Betula, and Agropyron,” is made up of sprangly specimens with ascending, sparsely branched stems, loose infructescences, long pedicels, and elliptic siliques that are noticeably compressed parallel to the septum. From my experience and observations in the field involving other species of Lesquerella, it appears that the differences in habit shown by these collections probably reflect responses to habitat: in the first instance, a dry, open, possibly windswept area; in the second, a somewhat more protected locale; and in the third, probably a shady situation with deep soil and other conditions favoring more rapid growth than normally takes place in the more open habitats. The variation in silique shape noted in Baker & Naumann 82-231 is probably genetic and reflects the type of diversity found between (not within) populations in other species of Les- querella. This diversity cannot be explained from the material available. — 508 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Polyctenium williamsiae Rollins, sp. nov. Herba perennis, caespitosa; caudicibus ramosis; foliis pectinatis, sparsim pubescentibus; caulibus erectis vel decumbentibus, simplicibus, 4-10 cm lon- gis; inflorescentiis corymbosis; petalis obovatis vel spathulatis, truncatis, 3-3.5 mm longis, 1.8—2 mm latis, albis vel purpureis: infructescentis elongatis; pedi- cellis adscendentibus vel divaricatis, rectis, pubescentibus, 3-5 mm longis: siliquis late oblongis, glabris, 3-4 mm longis, 2—2.5 mm latis; stylo ca. 0.5 mm longo; seminibus immaturis, oblongis, immarginatis. Type. Nevada, Washoe Co., Virginia Range. TI7N, R20E, $16, foothills E of Little Washoe Lake, edge of a vernal pond, 5680 fi alt., 4. Tiehm 7135 & M. Williams (holotype, GH; isotypes to be distributed). Perennial with taproot and with occasional underground root branches bear- ing retohos along their length; caudex loosely branched. Basal leaves densely tufted, stiff, pectinate, pubescent with mostly dendritically branched trichomes, these interspersed with larger, simple or forked trichomes; linear leaf segments often tipped with large, simple or forked trichomes. Cauline leaves similar to basal, sessile. nonauriculate. Inflorescences densely corymbose; sepals broadly oblong, ca. 2 mm long, hyaline margined, greenish to purplish, sparsely pu- bescent; petals obovate to broadly spatulate, truncate at apex, gradually tapering to point of insertion, 3-3.5 by 1.8—2 mm, whitish to purplish; anthers nearly orbicular, less than 0.5 mm long. Infructescences elongated: pedicels ascending to divaricately ascending, straight. 3-5 mm long, sparsely pubescent; siliques crowded, divaricately caer eee oblong to nearly oval in outline, rounded above and below, 3-4 5 mm, compressed (usually unevenly) in plane opposite to that of ie. paste glabrous, the styles thick, ca. 0.5 mm long, the stigma slightly expanded beyond diameter of style, the ovules 15 to 20 per locule, the funiculi slender, ca. 0.5 mm long. Immature seeds broadly oblong, plump, wingless. Greene (1912), in setting up Po/yctenium E. Greene as a genus distinct from Smelowskia, believed that it consisted of three species, P. bisu/catum E. Greene, P. fremontii (S. Watson) E. Greene, and P. glabellum E. Greene. However, it was later shown (Rollins, 1938) that all of the material then available, including the types of Greene’s species. was referable to a single species consisting of two varieties. The first incontrovertible evidence that Po/vcfenium is not monotypic comes from the newly described P. wi/liamsiae. Comparison of this species with the only other known species in the genus, P. fremontii, yields several differences. The siliques of P. williamsiae are broadly oblong to nearly oval— much shorter and wider and thus of a very different shape than the linear, much longer, narrower siliques of P. fremontii. The flowers of P. williamsiae are considerably smaller than those of P. fremontii. The styles of P. williamsiae are about half the length of those of P. fremontii. The dried leaves, stems, and siliques are greenish in P. williamsiae and gray in P. fremontil. The valves of the siliques in P. fremontii are smooth and rigid, while those of P. williamsiae are uneven, showing depressed areas that indicate a lack of rigidity in the valve wall. — 1983] ROLLINS, CRUCIFERAE 509 The generic status of Po/yctenium has been considerably strengthened by the discovery of a second species. Although Jepson (1936) and Abrams (1944) did not accept Polyctenium, the genus has been recognized in the more recent floras (e.g., Munz, 1959; Hitchcock & Cronquist, 1973) that cover the area where it occurs. ACKNOWLEDGMENTS Research leading to this publication was supported by National Science Foundation Grant DEB78-08766. I am especially indebted to Douglass Hen- derson and his associates, of the University of Idaho, for providing extensive material and field notes on Draha hitchcockti. | also want to thank J. Scott Peterson, of the Colorado Natural Heritage Inventory in Denver, who has been especially helpful in sending me interesting material from the Piceance Basin survey, and Dieter Wilken, Colorado State University, for his interest and for forwarding specimens from that survey. In addition, I am pleased to acknowl- edge the continuing help in Nevada of Arnold Tichm (Reno) and Margaret Williams (Sparks), and in Utah of Sherel Goodrich (Intermountain Forest and Range Experiment Station, Provo). It is a pleasure to name 4rabis tlehmii, Polyctenium williamsiae, and Lesquerella goodrichii, respectively, in their hon- or. LITERATURE CITED pages _ 1944. or ali flora of the Pacific states. Vol. 2. viii + 635 pp. Stanford Univ. Press, Stan DETLING, ot E. 1939. a revision of the North American species of Descurainia. Amer. Midl. Naturalist 22: 481-520. GreENE, E. L. 1912. Certain cruciferous types. Leafl. Bot. Observ. Crit. 2: 219, 220. Hitrcucock, C. L., & A. Cronguist. 1973. Flora of the Pacific Northwest: an illustrated manual. xix + 730 pp. Univ. i Press, Seattle. Jepson, W. L. 1936. Cruciferae. Pp. 17-17 pole 2, pt. 2, in A flora of California. Assoc. Students Store. Univ. California, oe ley Macuire, B.. & A. H. HormGren. 1951. Botany of the Intermountain Region II. Lesquerella. Madrono 11: 172-184. mae G. A. 1968. Transfers from Phiysaria to Lesquerella (Cruciferae). Canad. J. Bot. 46: 527-530. Munz, 2 - 1959. A California flora. 1681 pp. Univ. California Press, Berkeley and Payson, E. B. 1922. A monograph of the genus Lesquerella. Ann. Missouri Bot. Gard. 8: 103 —236. ROLLINS, > C. 1938. alee ale Polyctenium. Rhosora 40: as 305. —. 1939. The cruciferous s Physaria. Rhodora 41: 392-41 : nae Studies on ae cat American Cruciferae. Contr. Gia Herb. 171: 42-53. 1981. Studies in the genus PAysaria (Cruciferae). Brittonia 33: 332-341. 1982. A new species of the Asiatic genus neil: (Cruciferae) for North America and its biogeographic implications. Syst. . 7: 214-220. & U. BANERJEF. 1975. Atlas of the ae of Lesquerella (Cruciferae). 48 pp. Bussey Inst. of Harvard Univ., Cambridge, Massachusetts. 510 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 & E. A. SHaw. 1973. The genus Lesquerefla (Cruciferae) in North America. 288 pp. Harvard Univ. Press, Cambridge, Massachusetts. SMILEY, F. J. 1921. A report rd re the boreal flora of the Sierra Nevada of California. Univ. Calif. Publ. Bot. 9: 23. GRAY HERBARIUM HARVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS 02138 1983] HOWARD & STAPLES, CATESBY’S PLANTS pee THE MODERN NAMES FOR CATESBY’S PLANTS RICHARD A. HOWARD AND GEORGE W. STAPLES Mark Catessy (1682-1749) collected plants in the southeastern United States from Maryland to Florida between 1712 and 1719 and again between 1722 and 1725 (Reveal, 1983). During part of 1725, before his return to England in 1726, he collected in the Bahama islands of New Providence, Abaco, Andros, and Eleuthera (Britton & Millspaugh, 1920). Dried specimens, as well as seeds and possibly living plants, were sent to correspondents and sponsors in England and Europe (Allen, 1937). Plants and animals were sketched in the field, and drawings were occasionally done from plants grown from seeds or under cul- tivation in Europe. Catesby (1731, p. xi) said, ““As I was not bred a Painter I hope some faults in Perspective, and other Niceties may be more readily ex- cused... .’’ Unfortunately, two drawings are so inaccurate that they still can- not be identified. The Natural History of Carolina, Florida and the Bahama Islands, termed ‘‘a pioneer contribution to the natural history of the West Indies as well as to that of the southern United States” (Stearn. 1958, p. 328), was published in parts between 1730 and 1743, with an appendix appearing in 1747. The work was soon cited by J. F. Gronovius in his Flora Virginica. Linnaeus cited many but not all of Catesby’s plates in Species Plantarum (ed. 1, 1753; ed. 2, 1762, 1763), Systema Naturae (ed. 10, 1759), Centuria I. Plantarum (1756; contains single reference to Vinca lutea), Mantissa Plantarum (1767), and Mantissa Plantarum Altera (1771) Catesby’s plates appear to be the types of twenty-five recognized taxa, of which twenty-one were described by Linnaeus and four by subsequent authors. They are also the types of an additional twelve synonymous names. Typotype'! specimens from the Sloane herbarium have been indicated for some species, primarily by Dandy (1958). Linnaeus may have seen some of Catesby’s col- lections when he visited Gronovius (1735) or Sherard (1736). It is generally stated that the plate is the type, and supporting specimens “in the Sloane set can be associated with certain of the plates” (Stafleu, 1967, p. 78). Not all of Catesby’s plates have been satisfactorily identified, in spite of existing listings. In a search of the Sloane herbarium (BM) and the Sherard and Du Bois herbaria (OXF), NO specimens were found to aid in the identification of plates 30, 45, or 52 in volume two. Catesby’s specimens are also known to occur in the general herbarium at the British Museum (Natural History). These are primarily from IT . } lick 1°1) ; q eet | i ly de sienated as ‘the type ofa valid y y published name Ty poty pe specimens have r no nomenclatural standing but may be considered in the f ambiguous or incom- plete type renee or iueeatione. © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 511-546. an 1983. oid JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 the Gronovius and the Samuel Dale herbaria. Sir Joseph Banks had acquired the Gronovius herbarium, and the Dale herbarium had been presented to the Society of Apothecaries at the Chelsea Physic Garden. Both herbaria were acquired by the British Museum (Natural History). Catesby’s own holographic notes, although commonly attributed to the Oxford collections, also occur on various specimens in London (BM). A second edition of Catesby’s Natural History was edited by George Edwards and published in 1754. A third edition, published in 1771, contained as an appendix credited to Edwards “A Catalogue of the Animals and Plants Rep: resented in Catesby’s Natural History of Carolina: with the Linnaean names.’ Also in 1771 Johann Reinhold Forster published a translation of J.-B. Bossu’s Travels Through that Part of North America Formerly Called Louisiana, to which he added (Volume 2, pp. 17-67) ““A Systematic Catalogue of all the Known Plants of English North-America, or, a Flora Americae Septentriona- lis.’ Forster cited many but not all of Catesby’s plates, with references to Linnaeus’s works. It is possible that both Edwards and Forster obtained much of the information from the second edition of Gronovius’s Flora Virginica (1762). where more of Catesby’s plates are cited than in edition one and ref- erences are given to Linnaeus’s Species Plantarum. Occasional references as- sociating Catesby’s plates with Linnaean names are also found in Adanson’s Familles des Plantes (1763). Modern names for Catesby’s plants and plates have been suggested by various twentieth-century authors. The first comprehensive treatment was that by Brit- ton and Millspaugh in The Bahama Flora (1920). where all but one of Catesby’s plants from the Bahama Islands were identified under the then-accepted names. Regrettably, Catesby’s polynomials were not listed in the index, and some of the suggestions of Britton and Millspaugh have therefore been overlooked by subsequent authors. In 7he Sloane Herbarium Dandy (1958) identified a few of Catesby’s polynomials and plates and associated them with typotypes in that herbarium. Stevenson (1961) in a treatment of Catesby in volume two of the Catalogue of Botanical Books in the Collection of Rachel MeMasters Miller Hunt, suggested a few identifications. In 1974 the Beehive Press, Savannah, yeorgia, issued a partial facsimile of Catesby’s work, with reproduction of a number of the colored plates together with the related text. An accompanying volume contains an historical introduction by George Frick. notes by Joseph Ewan, and black and white reproductions of all the plates. Ewan suggested determinations of all of the plants and birds illustrated by Catesby. but we do not agree with all of his names. Since the facsimile reproduction and the text have not been widely distributed and no reprints of the folio-sized text are available, we believe that the following lists and comments should be useful. Catesby’s collections (8M, OXF) (Clokie, 1964) are more extensive than those illustrated in his Natural History. These have not been cited, to our knowledge, and have no bearing on modern nomenclature, but they are of interest since some represent introduced taxa or give clues to his collecting localities. The search for typotype specimens of the species illustrated involved all of the specimens in volumes 212 and 232 of the Sloane herbarium (BM), and are so cited. Commonly mounted with other collections, the specimens are small or fragmentary and may occur on several pages. The Sherard and Du Bois herbaria (OXF) are retained as historic herbaria, often without separation to modern 1983] HOWARD & STAPLES, CATESBY’S PLANTS om Be) families or genera. A large number of Catesby’s specimens with holographic notes were found in the unidentified material at Oxford; these have been an- notated, with a decision pending where they will be filed. This work began with the need for the modern nomenclature of Catesby’s illustrations. These were cited in an unpublished manuscript of Alexander Anderson. the second director of the St. Vincent Botanical Garden (1785- 1811), who introduced Bahama plants to that garden and supplied binomials himself. The work on Anderson, including visits to London and Oxford. was supported by a much-appreciated grant to the senior author by the Stanley Smith Horticultural Trust. We are also grateful to the curatorial staff at the British Museum (Natural History) and the Fielding-Druce Herbarium at Ox- ford, We particularly value the comments of Charles Jarvis (BM, LINN), who 1S endeavoring to typify the Linnaean taxa in Species Plantarum. Our cooperative colleague, Dr. Leslie Garay, supplied the identifications of the Orchidaceae. In the following notes, the plants that Catesby described and illustrated have been arranged alphabetically by families and genera. All names that we have determined to be based on Catesby’s illustrations are marked with an asterisk. The plants are cited as plates by volume and number —for example (1: 7. //). Catesby’s descriptive phrases in Latin and his common names are given, with spelling, punctuation, and capitalization as in the first edition of Natural His- tory. Comments are made on the typification. Associated specimens are cited for the Sloane herbarium (H.S.), the general herbarium of the British Museum of Natural History (BM), the Dale herbarium (sm-Dale), the Sherard herbarium (oxF-s), and the Du Bois herbarium (OxF-pB). A list 1s supplied. numerically arranged, with our suggested identifications of plates in volumes | and 2, including the Appendix. GY MNOSPERMAE TAXODIACEAE Taxodium distichum (L.) Rich. (1: ¢. //) Cupressus Americana. Cypress of America. Cupressus disticha L. Sp. Pl. 1003. 1753. Catesby, Hortus Upsal., Gro- novius. and others are cited. (H.S. 212. f. 4 (left); 232, f. 69, f. 85.) ANGIOSPERMAE — MONOCOTYLEDONEAE AMARYLLIDACEAE Hymenocallis caroliniana (L.) Herb.* (2(App.): ¢. 5) Lilio-Narcissus Polianthus, flore albo. Pancratium carolinianum L. Sp. Pl. 291. 1753. Linnaeus cited only Catesby. Dandy (1958, p. 112) noted that the plate is the type. Sealey (Kew Bull. 1954: 234-236) commented on the illustration and recognized the species under Hymenocallis, but noted that it had not been re-collected in the area where Catesby is known to have traveled. It is represented by only a single collection. No current floras of the southeastern United States recognize the species, but it is listed in Correll and Johnston (Manual of the Vascular Plants of Texas, 1970) for Texas. 514 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Zephyranthes atamasco (L.) Herb. (2(App.): ft. /2) Lilio narcissus Virginiensis. Attamusco Lily. Amaryllis atamasca L. Sp. Pl. 292. 1753. Catesby, Hortus Cliffort., Gro- novius, and others were cited. (Bm-Dale.) ARACEAE Orontium aquaticum L. (1: ¢. $2 Arum aquaticum minus; S. Arisarum fluitans pene nudo Virginianum. Orontium aquaticum L. Sp. Pl. 324. 1753. Catesby was not cited. Peltandra virginica (L.) Schott & Engler (1: ¢. 83) Arum Sagitariae folio angusto, acumine & auriculis acutissimis. Ewan (1974, p. 94) identified this illustration as P. sagittaefolia (Michx.) Morong, which has a white spathe and red berries. Symplocarpus foetidus (L.) Nutt. (2: ¢. 7/) Arum Americanum, Betae folio. Scunk Weed. Dracontium foetidum L. Sp. Pl. 967. 1753. Catesby was not cited. Unidentified A/ocasia or XNanthosoma (2: t. 45) Arum maximum Aegyptiacum, quod vulgo Colocasia. The plate shows an inflorescence and a folded leaf. neither of which rep- resents Colocasia esculenta (L.) Schott, the identification suggested by Ewan (1974). Catesby stated: The roots of this Plant are tuberous, with many small Fibres growing from them; some of them weigh six or eight Pound, of an irregular Form, the Outside of a rusty brown colour, the Inside white. The Leaves grow out of the Earth, with only their Foot-stalks, to the Height of four or five Feet; they are shaped somewhat like a Heart, of a pale Green, very ample, some of them being two Feet wide, and more in Length. The Flower in Form resembles that of the common 4rum, tho’ in Colour different, the Hood is green without, and of a light yellow within; the Pistil is long and slender, of a light purple Colour. Sir Hans Sloane, has so amply treated of this useful Plant, that I shall ask Leave only to add a few Remarks more. It is a Tropick Plant, not caring to encrease much in Carolina, and will grow no where North of that Colony; yet the Negro’s there (who are very fond of them) by annually taking up the Roots to prevent rotting, Carolina, there was introduced a new Kind, wholly without that bad Quality, and requiring no more than common Time to boil them, ~ may be eat raw, without offending the Throat or Palate; this was a welcome Improv among the Negro’s, nd was esteemed a Blessing; they being delighted Wie all their pee Food, particularly this, which a great Part of Africa subsists much on This later introduction may be Colocasia esculenta (L.) Schott. BROMELIACEAE Catopsis berteroniana (Schultes) Mez (2: r. 77) Viscum Caryophylloides, Aloés foliis viridibus acuminatis, floribus race- mosis luteis. Tillandsia balbisiana (Schultes) Roemer & Schultes (2: 7. 89) Viscum Cariophylloides angustifolium, Floribus longis tubulosis caeruleis, ex spicis squamosis caeruleis erumpentibus. Wild Pine 1983] HOWARD & STAPLES, CATESBY’S PLANTS 515 COMMELINACEAE Commelina virginica L. (2: f. 62) Pseudo-Phalangium ramosum. Commelina virginica L. Sp. Pl. ed. 2.61. 1762. Catesby was not mentioned. (ALS: 212 i621 57) GRAMINEAE Oryza sativa L. (1: ¢. /4) Rice. Oryza sativa L. Sp. Pl. 333. 1753. Catesby, Hortus Cliffort., and others were cited oe paniculata L. (1: ¢ amen My loebolior on Oxphaiton Carolinianum, &c. Sea-side Oat. ae paniculata L. Sp. Pl. 71. 1753. Catesby, Hortus Cliffort., and Gro- novius were cited. HyYDROCHARITACEAE Thalassia testudinum Konig (2: 4. 38 Alga marina, graminea angustissimo folio. Catesby was not cited by Kénig. (H.S. 232, f. 13.) Hy POxIDACEAE Hypoxis sp. (1: 7. 33) Ornithogalum luteum parvum folijs gramineis glabris. Little yellow Star- ower. Ewan (1974, p. 92) cited this illustration as Hypoxis hirsuta (L.) Cav., which is based on Ornithogalum hirsutum L. (Sp. Pl. 306. 1753). Linnaeus did not cite Catesby. No ce on could be found, and the plant may well be Hypoxis juncea J. E. LILIACEAE Lilium canadense L. (2(App.): ¢. //) Lilium sive Martagon Canadense, floribus magis flavis, non reflexis. Lilium canadense L. Sp. Pl. 303. 1753. Catesby, Morison, and others were cited. Lilium catesbaei Walter (2: 7. 58) Lilium Carolinianum, flore croceo punctato, petalis longioribus & angus- tioribus. Red Lill ‘atesby was not mentioned in Walter’s description (Fl. Carolin. 123. 1788). Lilium philadelphicum L. (2(App.): ¢. 8) Lilium angustifolium, flore rubro singulari. Lilium philadelphicum L. Sp. Pl. ed. 2. 435. 1762. Catesby was not men- tioned in the original pee: and the type is presumably Miller, Fig. PI. Gard. Dict. 2: ¢. 165, fig. 1. 1760. 516 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 Lilium superbum L. (2: 7. Lilium, sive pies ae nse, flore luteo punctato. Lilium superbum L. Sp. Pl. ed. 2. 424. 1762. Miller. Trew, Ehret, and Catesby were cited. Ewan (1974, p. 97) called this illustration Lilium mi- chauxti Poiret. Trillium catesbaei Ell.* (1: 4. 45) Solanum triphylion flore hexapetalo carneo. atesby was not mentioned in the original description (Sketch Bot. S.-Carolina Georgia 1: 429. 1817). Dandy (1958, p. 112) stated that the plant is Trillium catesbaei Ell., with a typotype specimen H.S. 212, f£. 59 (flower, lower row, second from right). See also A. B. Rendle (Jour. Bot. London 39: 322-325. 1901) Trillium maculatum Raf.* (1: 7. 50) Solanum triphyllon flore hexapetalo tribus petalis purpurejs erectis coeteris viridibus reflexis. Linnaeus cited Catesby, Gronovius, and Plukenet in the protologue of Trillium sessile L. (Sp. Pl. 340. 1753). Dandy (1958, p. 112) stated that Catesby’s illustration is Trillium maculatum Raf., with a typotype H.S. 212. f. 59 (flower, lower right), and this selection has been confirmed by J. D. Freeman (Brittonia 27: 27. 1975). ORCHIDACEAE Cleistes divaricata (L.) Ames (1: ¢. 58, above) Helleborine Lilijfolio caulem ambiente, flore unico hexapetalo, tribus pe- talis longis angustis obscure purpureis, caeteris brevioribus roseis. Lilly-leaf'd Helebore. Arethusa divaricata L. Sp. Pl. ed. 2. 1346. 1763. Catesby and Gronovius were cited. Cypripedium acaule Aiton (2: ¢. 72) Helleborine. Lady’s Slipper of Pensi/vania. Ewan (1974, p. 98) identified this illustration as Cypripedium calceolus L. The plate 1s also the type of Cypripedium vittatum Vell. var. planum Raf. (Herb. Raf. 44. 1833). Cypripedium acaule Aiton (2(App.): f. 3) Calceolus, flore maximo sy tee purpureis venis notato, foltis amplis hir- suds crematis, radice dentis canini. This illustration is the type of Cypripedium catesbianum Raf. (Herb. Raf. 44. 1833). Rafinesque cited the third edition of Catesby (1771), with this plate as “r. /03.” Cypripedium pubescens Willd. (2: 7. 73) Calceolus Marianus glaber, Petalis angustis. Yellow Lady’s Slipper. Ewan (1974, p. 98) referred this illustration to Cypripedium calceolus L. var. pubescens (Willd.) Correll. The plate is also the type of Cypripedium vittatum Vell. var. tortile Raf. (Herb. Raf. 44. 1833). Epidendrum boothianum Lindley (2: 7. 74) Viscum Carvophilloides, floribus parvis luteis punctatis. — 1983] HOWARD & STAPLES, CATESBY’S PLANTS 517 Ewan (1974) identified this illustration as Epicladium boothianum (Lind- ley) Small. Epidendrum cochleatum L. (2: ¢. 88, right) Viscum Caryophylloides, Lilti albi foliis, Floris labello brevi purpureo, cete- ris Petalis ex luteo virescentibus. Epidendrum cochleatum L. Sp. Pl. ed. 2. 1351. 1763. Sloane, Catesby, and Plumier were cited. The Plumier reference (Burman, Pl. Am. 180. ¢. /85, fig. 2. 1759) has been chosen as the lectotype. Epidendrum nocturnum Jacq. (2: ¢. 63) Viscum Carvophylloides, foliis longis in apice incisis, floris labello albo trifido, petalis luteis, longis angustissunis. Epidendrum nocturnum L. Sp. Pl. ed. 2. 1349, 1763. Catesby and Jacquin (Select. Stirp. Am. Hist. t. /39) were cited. Jacquin’s figure has been chosen as the lectotype. Epidendrum plicatum Lindley (2: ¢. 88, left) Viscum radice bulbosa; Floris labello carneo, ceteris sordide luteis. Polystachya concreta (Jacq.) Garay & Sweet (2: ¢. 55) Viscum Caryophylloides ramosum, floribus minimis albis. Ewan (1974, p. 97) called this illustration Pol/vstachya minuta (Aublet) Britton. Vanilla mexicana Miller (2(App.): ¢. 7) Volubilis siliquosa mexicana pane folio. Epidendrum vanilla L. Sp. Pl. 952. 1753. Catesby and others were cited. Ewan (1974, p. 99) referred this illustration to Vanilla planifolia Andr. SMILACACEAE Smilax lanceolata L. (2: 1. 84, below) Smilax non Spinosa baccis rubris. Smilax lanceolata L. Sp. Pl. 1031. 1753. Catesby was not cited. (OXF-DB, OXF-S.) Smilax laurifolia L. (1: ¢. 75) Smilax laevis, Lauri folio, baccis nigris. Bay-leaved Smilax. Smilax laurifolia L. Sp. Pl. 1030. 1753. Catesby and Gronovius were cited, (OXF-S.) Smilax pumila Walter (1: 7. 47) Smilax non spinosa, humilis, folio Aristolochiae, baccis rubris. Catesby was not cited in the original description (Fl. Carolin. 244. 1788.) Ewan (1974, p. 93) referred this illustration to Smilax herbacea L. Smilax tamnoides L. (1: 7. 52) Smilax Brvoniae nigrae folijs caule spinoso, baccis nigrts. Linnaeus (Sp. Pl. 1030. 1753) cited only Catesby; however, a specimen obtained by Kalm (LINN 1132.10) is preferable as a lectotype. Ewan (1974, p. 93) referred this illustration to Smilax bona-nox L. (A HeSa232. f 31 OXF-DB; OXF-S.) 518 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ANGIOSPERMAE — DICOTYLEDONEAE ACERACEAE Acer rubrum L. (1: 7. 62 Acer Virginianum, folio majore, subtus argenteo, supra viridi splendente. Red flowring Maple. Linnaeus (Sp. PI. 10 and others. (H.S. 212. ao ) cited Catesby. Hortus Upsal., Gronovius, f, 14: 232.1. 323) ANACARDIACEAE Anacardium occidentale L. (2(App.): f. 9) Pomifera, seu potius prunifera Indica, nuce viniformi summo pomo in- nascente, Cajous vel Acajous dicta. Cushew Tree. Linnaeus (Sp. Pl. 383. 1753) cited Catesby. Hortus Cliffort., and others. Metopium toxiferum (L.) Krug & Urban (1: 1. 40) Toxicodendron folijs alatis fructu purpureo Pyri formi sparso. Poison-Wood. Amiyris toxifera L. Syst. Nat. ed. 10. 1000. 1759. Linnaeus cited Catesby and Hortus Cliffort. Rhus glabra L. (2(App.): ¢. 4) Rhus glabrum Panicula speciosa coccinea. Linnaeus (Sp. Pl. 265. 1753) cited Gronovius and others, but not Catesby. ANNONACEAE Annona glabra L.* (2: ¢. 64) Anona maxima, foliis latis fructu maxuno luteo conoide, cortice glabro. Linnaeus (Sp. Pl. 537. 1753) cited only Catesby but supplied a new phrase name. No supporting specimens have been located. Dandy (1958. p. 112) stated that the plate was the type. Annona glabra L. (2: ¢. 67) Anona fructu viridi laevi, Pyri inversi forma. This illustration has been referred to Annona palustris L. (Sp. Pl. ed. 2. 757. 1762). Catesby was not cited, and the species is regarded as a synonym of Annona glabra L. (1753). Ewan (1974, p. 97) called this illustration Annona cherimolia Maller. No associated specimens have been located to verify the identification either way. Annona reticulata L. (2: 7. 86) Anona maxima, foliis oblongis, angustis; fructu maximo luteo conoide; cortice glabro in areolas angulares distincto. Linnaeus (Sp. Pl. 537. 1753) cited Catesby, Sloane, and Ray. Fawcett and Rendle (Fl. Jamaica 3: 198. 1914) designated “Sloane Herb. vii. 94” as the type specimen. Asimina triloba (L.) Dunal* (2: ¢. S5) Anona fructu lutescente, laevi, scrotum Arietis referente. Annona triloba L. Sp. Pl. 537. 1753. Only Catesby was cited, although a new phrase name was used. Dandy (1958, p. 112) stated that the plate is the type. 1983] HOWARD & STAPLES, CATESBY’S PLANTS S519 APOCYNACEAE Echites umbellata Jacq. (1: ¢. 58, below) Apocynum Scandens folio, cordato flore albo. Dogs-bane. Tabernaemontana echites L. Syst. Nat. ed. 10. 945. 1759. Only Sloane was cited. Plumeria obtusa L.* (2: ¢. 93, above) Plumeria flore niveo, foliis brevioribus obtusis. Linnaeus (Sp. Pl. 209. 1753) cited only Catesby. Dandy (1958, p. 112) stated that the plate is the type. Plumeria rubra L. (2: ¢. 92) Plumieria Flore Roseo odoratissimo. Linnaeus (Sp. Pl. 209. 1753) cited Catesby, Hortus Cliffort., and others. Urechites lutea (L.) Britton* (2: 4. 53) Apocynum Scandens, Salicis folio, flore amplo pleno. Vinea lutea L. Cent. II. Pl. 12. 1756. Dandy (1958, p. 112) called the plate the type. Echites catesbaei G. Don (Gen. Syst. 4: 74. 1838), now regarded as a synonym of Urechites lutea, is also based on Catesby’s plate. AQUIFOLIACEAE Ilex cassine L. (1: ¢. 31 Agrifolium Carolinense folijs dentatis baccis rubris. Dahoon Holly. Linnaeus (Sp. Pl. 125. 1753) cited Catesby and Hortus Cliffort. (H.S. 212, f. 65. (upper left).) Ilex vomitoria Aiton (2: ¢. 57) Cassena vera Floridanorum, Arbuscula baccifera Alaternifacte, foliis al- ternatim Sitis, tetrapyrene. Ilex cassine var. B, L. Sp. Pl. 125. 1753. The variety remains unnamed, and Linnaeus cited Catesby and Plukenet. The type of Aiton’s taxon (Hortus Kew. 1: 170. 1789) is presumably a cultivated plant. Rehder (Jour. Arnold Arb. 3: 212-215. 1922) discussed the fact that the older name, Cassine Paragua Miller (Gard. Dict. ed. 8. no. 2. 1768), based on Catesby’s plate, may be considered a later homonym of Cassine Peragua L. (Sp. Pl. 268. 1753). ARALIACEAE Panax quinquefolius L. (2(App.): ¢. /6) Aureliana Canadensis R. P. Lafiteau. Ginseng. Linnaeus (Sp. Pl. 1058. 1753) cited Catesby, Gronovius, and others. ARISTOLOCHIACEAE Aristolochia serpentaria L. (1: 7. 29) Aristolochia pistolochia seu Serpentaria Virginiana caule nodoso. Snake- Root of Virginia. Linnaeus (Sp. Pl. 961. 1753) cited Catesby and Gronovius. 520 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 BERBERIDACEAE Podophyllum peltatum L. (1: 1. 24) Anapodophyllon Canadense Morini. May Apple. Linnaeus (Sp. Pl. 505. 1753) cited Catesby, Hortus Cliffort., and Hortus Upsal. (H.S. 212, f. 63 (upper left).) BIGNONIACEAE Bignonia capreolata L. (2: ¢. 82) Bignonia Americana, capreolis donata, siliqua breviore. In the original description (Sp. Pl. 624. 1753) Linnaeus cited Hortus Clif- fort.. p. 317, but not Catesby. Further discussion was presented by Sprague (Jour. Bot. London 60: 236-238. 1922). Wilbur (Taxon 29: 299-303. 1980) has proposed conserving Bignonia capreolata L. as the lectotype of the genus Bignonia L. Campsis radicans (L.) Seem. (1: ¢. 65) Bignonia, Fraxini folijs, coccineo flore minore. Trumpet-Flower. Bignonia radicans L. var. 8, L. Sp. Pl. 625. 1753. Only Catesby was cited. Dandy (1958, p. 112) stated that the plate is the type of Bignonia radicans var. 8. The unnamed variety was not recognized by Rehder (Bibliogr. Cult, Trees Shrubs, 1949). The color variation in various copies of the Natural History suggests that the variety is not worthy of recognition. Catalpa bignonioides Walter (1: 1. 49) Bignonia Urucu folijs flore sordidé albo, intus maculis purpuretjs & luteis asperso, Siliqua longissima & angustissima. Catalpa Tree. Bignonia an 7 Sp. Pl. 622. 1753. Catesby and Hortus Cliffort. were cited. (H.S. 212, » 232, f. 51.) Jacaranda ae Grebe (1: 4. 42) Arbor Guajaci latiore folio, Bignoniae flore caeruleo, fructu duro in duas partes disiliente, seminibus alatis imbricatim positis. Broad leaf'd Guaicum, with blew Flowers. Bignonia caerulea L. Sp. Pl. 625. 1753. Only Catesby was cited. Dandy (1958, p. 112) stated that the plate is the type. and H.S. 232. f. 12 (upper right) 1s the typotype specimen. Tabebuia bahamensis (Northrop) Britton (1: ¢. 37) Bignonia arbor pentaphylla flore roseo majore siliquis planis. Bignonia. r Bignonia pentaphvila L. (Sp. Pl. ed. 2. 870. 1763) Linnaeus cited Catesby and others. This illustration is not Tabehuia pentaphylla (Juss.) Hemsley. (H.S. 232, f. 45 (lower right).) BORAGINACEAE Bourreria ovata Miers (2: ¢. 79) Pittoniae similis, Laureolae foltis, floribus albis, baccis rubris. In the protologue of Fhretia bourreria L. (Sp. Pl. ed. 2. 275. 1762), Linnaeus cited Catesby, Browne, Sloane, and others. In 1869 Miers (Ann. Mag. Nat. Hist. IV. 3: 207) cited Catesby “il. tab. 79” in his description of the new species Bourreria havanensis Miers. In 1911 O. E. Schulz (in Urban. Symb. 1983] HOWARD & STAPLES, CATESBY’S PLANTS 521 Antill. 7: 60) excluded the Catesby citation when he referred B. havanensis Miers to the synonymy of Beureria succulenta var. revoluta (Kunth) Schulz. Schulz (op. cit., p. 55) referred Catesby’s illustration to Beureria ovata Miers, although it was not included in the original description of that species. (H.S. 232, f. 45 (upper right).) Cordia sebestena L. (2: 1. 9/, above) Caryophyllus spurius inodorus, Folio subrotundo scabro, flore racemoso hexapetaloide coccineo Speciosissimo. Linnaeus (Sp. Pl. 190. 1753) cited Catesby, Sloane, Dillenius, and Ray. (H.S. 232, f. 24, f. 33.) BURSERACEAE Bursera simaruba (L.) Sarg. (1: ¢. 30) Teribinthus major Betulae cortice, fructu triangulari. Gum-elimy Tree. Pistacia simaruba L. Sp. Pl. 1026. 1753. Linnaeus cited Sloane and Catesby. Fawcett and Rendle (FI. Jamaica 4(2): 206. 1920) stated, “Type in Herb. Sloane in Herb. Brit. Mus.” Ewan (1974, p. 92) cited this plant under the synonymous name Llaphrium simaruba (L.) Rose. CALYCANTHACEAE Calycanthus floridus L. (1: 1. 46) Frutex corni folijs conjugatis, floribus instar Anemones stellatae, petalis crassis, rigidis, colore sordidé rubente; cortice aromatico. Linnaeus (Syst. Nat. ed. 10. 1066. 1759) cited Catesby, Ehret, Duhamel, and Miller. A lectotype has not been chosen. (H.S. 212, f. 16; 3mM-Dale.) CANELLACEAE Canella winterana (L.) Gaertner (2: f. 50) Arbor baccifera, laurifolia, aromatica, fructu viridi calyculato racemoso. Winter’s Bark. Laurus winterana L. (Sp. Pl. 371. 1753). Linnaeus cited Catesby, Hortus Cliffort., and Sloane. CARYOPHYLLACEAE Silene virginica L. (2: ¢. 54) Lychnis viscosa, Virginiana, flore amplo coccineo: seu Muscipula Regia. Linnaeus (Sp. Pl. 419. 1753) cited only Gronovius. Specimens collected by Kalm and Pursh are in the Linnean herbarium. (HLS. PAD sie al'§.-) CHRYSOBALANACEAE Chrysobalanus icaco L. (1: f. 25) Frutex cotini fere folio crasso, in summitate deliquium patiente, fructu ovali coeruleo ossiculum angulosum continente. Cocoa Plum. Linnaeus (Sp. Pl. 513. 1753) did not cite Catesby. Prance (Fl. Neotrop. 9: 15. 16. 1972) stated that the type was LINN 641.1. SPH) JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 CLETHRACEAE Clethra alnifolia L. (1: f. 66 Alni folia Americana serrata, floribus pentapetalis albis, in spicam dispo- sitis. Linnaeus (Sp. Pl. 176. 1753) cited Catesby and Gronovius. (H.S. 212, f. 50° 232, f. 35; CLUSIACEAE Clusia rosea Jacq. (2: 1 Cenchramidea Arbor saxis adnascens, obrotundo pingui folio, fructu pomi- formi, in plurimas capsulas granula ficulnea, stilo columnari octogono prae- duro adhaerentia continentes, diviso; Balsamum fundens. Balsam-Tree. Linnaeus (Sp. Pl. ed. 2. 1493. 1763) cited Jacquin and Catesby. (H.S. 232 f. 8 (one leaf).) COMBRETACEAE Conocarpus erectus L. (2: ¢. 33, above) Manghala arbor Curassavica, foliis Salignis. Button-Wood. Linnaeus (Sp. Pl. 176. 1753) cited Hortus Cliffort., Sloane, and others, but not Catesby. (H.S. 232, f. 24.) Laguncularia racemosa (L.) Gaertner (1: ¢. 86) An Thymelaea foliis obtusis. Conocarpus racemosus L. Syst. Nat. ed. 10, 930. 1759. Linnaeus did not cite Catesby. Britton and Millspaugh (1920, p. 646) were unable to identify this illustration. Although the plant described and illustrated by Catesby has alternate leaves, in all other characters it is clearly the white mangrove of the Bahamas. COMPOSITAE Borrichia arborescens (L.) DC. (1: ¢. 93) Chrysanthemum Bermudense Leucoji foltis virentibus crassis. Buphthalmum frutescens L. Syst. Nat. ed. 10. 1227. 1759. Linnaeus cited only Plumier. Sasa purpurea (L.) Moench (2: ¢. 59) Chrysanthemum Americanum, Doronici folio, flore Persici coloris, umbone magno prominente ex atro purpureo, viridi, & aureo fulgente. Rudbeckia purpurea L. Sp. Pl. 907. 1753. Linnaeus cited Catesby, Gro- novius, and others. Salmea petrobioides Griseb. (1: /. 72) Arbor maritima, foliis conjugatis pyriformibus apice in summitate instruc- tls, floribus racemosis luteis he type of the Grisebach species is a collection by Swainson from the Bahamas. ny incorrectly described and illustrated the florets as yellow instead of whi 1983] HOWARD & STAPLES, CATESBY’S PLANTS 23 Wedelia bahamensis (Britton) O. E. Schulz (1: 4. 92) Although this plant is illustrated, there is no accompanying text. “Chry- santhemum &c.” appears on the plate in the lower right corner. The type of the basionym Stemmodontia bahamensis Britton is Britton and Brace 302, from the Bahamas. Catesby’s plate was cited by Britton and Millspaugh (1920, p. 451). CONVOLVULACEAE Ipomoea batatas (L.) Lam. (2: ¢. 60) Convolvulus Radice tuberoso esculento. Virginian Potato. Convolvulus batatas L. Sp. Pl. 154. 1753. Linnaeus cited Catesby, Hortus Cliffort., and others. Ipomoea carolina L.* (2: 4. 9/7, below) Convolvulus minor Pentaphyllos, flore purpureo minore. Linnaeus (Sp. Pl. 160. 1753) cited only Catesby. Dandy (1958, p. 112) stated that the plate is the type. (Bm-Dale. Ipomoea microdactyla Griseb. (2: 1. 87, below) Convolvulus foliis variis, inferioribus trifariim divisis, superioribus sagit- tatis; floribus ex rubro purpureis. Ipomoea sagittata Poiret (1: 4. 35) Convolvulus Caroliniensis angusto sagittato folio, flore amplissimo_ pur- pureo, radice crassa. Purple Bind-weed of Carolina. (OXF-DB.) CORNACEAE Cornus florida L. forma rubra (Weston) Schelle (1: 4. 27) Cornus mas Virginiana, flosculis in corymbo digestis perianthio tetrapetalo albo radiatim cinctis. Dogwood Tree. Linnaeus (Sp. Pl. 117. 1753) cited Catesby, Gronovius, Hortus Cliffort.. and Hortus Upsal. (H.S. 212. f. 5 (upper left, buds only); 233, f. 89 (buds); BM-Dale.) EBENACEAE Diospyros virginiana L. (2: f. 76) Guajacana. Linnaeus (Sp. PI. 1057. 1753) cited Catesby, Gronovius, and Hortus Clif- fort. ERICACEAE Kalmia angustifolia L. (2(App.): ¢. 17, /eft) Chamaedaphne semper virens, foliis oblongis angustis, foliorum fasciculis oppositis é foliorum alis. Linnaeus (Sp. Pl. 391. 1753) cited Catesby, Gronovius, and Plukenet. Kalmia latifolia L. (2: ¢. 9S) Chamaedaphne foliis Tint, floribus bullatis umbellatis. 524 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Linnaeus (Sp. Pl. 391. 1753) cited Catesby, Gronovius, and Plukenet. (H.S. 212, f. 64 (lower right, flowers); 232, f. 54 (fruit).) Leucothoé racemosa (L.) A. Gray (2: Frutex foltis serratis, Aanibus eee us spicatis subviridibus, capsula pen- tagona. Linnaeus (Sp. Pl. 394. 1753) cited Catesby and others in the protologue of Andromeda paniculata L. Rehder cited “Andromeda cle L. p.p max.” in the synonymy of Leucothoé racemosa. The type Jeena racemosa L. (Sp. Pl. 394. 1753), the basionym, is a specimen eel by Kalm (LINN 563.15). Oxydendrum arboreum (L.) DC.* (1: 7. 7/) Frutex folijs oblongis acuminatis, floribus spicatis unoversu dispositis. Sor- rel-Tree. Andromeda arborea L. Sp. Pl. 394. 1753. Linnaeus cited Catesby and “Gronovius virg. 48. The latter reference is to Catesby, and the illustration may well be the type. (H.S. 212. f. 60; 232, f. 57: sm-Dale. No material is in the Linnean herbarium.) Rhododendron maximum L. (2(App.): ¢. /7, right) Chamaerhododendros lauri-folio semper virens, floribus bullatis corym- bosis. Linnacus (Sp. Pl. 392. 1753) cited Catesby and “*“Amoen. Acad. 2. p. 201.” Rhododendron viscosum (L.) Torrey var. aemulans Rehder (1: ¢. 57) Cistus Virginiana, flore & odore Periclymeni. Upright Honysuckle. Azalea viscosa L. (Sp. Pl. 151. 1753). Linnaeus cited Catesby, Gronovius, and others. A specimen collected by Kalm is in the Linnean herbarium. EUPHORBIACEAE Croton eluteria (L.) Sw.* (2: 7. 46) An Ricinoides Aeleagni folio? \lathera Bark. Linnaeus (Sp. Pl. 1042. 1753) ended his description of Clutia cascarilla L. with a dagger symbol, suggesting an imperfectly known species or some doubt or uncertainty; he cited only Catesby. Dandy (1958, p. 112) stated that the plate was the type. Millspaugh (Publ. Field Mus. Bot. 2: 306-308. 1909) concluded that this was the same as C/utia e/uteria L., for which Linnaeus (Sp. Pl. 1042. 1753) cited Hortus Cliffort. and others. Millspaugh combined the taxa on the basis of page priority and accepted the name Croton eluteria (L.) Sw. In reaching this conclusion, he also noted the inaccuracy of Catesby’s drawing and the fact that Linnaeus later (Sp. Pl. ed. 2. 1424. 1763) changed the description of Clutia cascarilla to apply to a taxon known as Croton linearis Jacq. Ewan (1974, p. 96) accepted Croton eluteria (L.) Sw. for Catesby’s plate and stated that the typotype was “‘H.S. 232. f. 24 [upper left].” Hippomane mancinella L. (2: ¢. 95, above) Mancanilla Pyri facie. Mancaneel-Tree. Linnaeus (Sp. Pl. 1191. 1753) cited Catesby, Hortus Cliffort., Sloane, and others. (H.S. 232, f. 5 1983] HOWARD & STAPLES, CATESBY’S PLANTS a2 GN Phyllanthus epiphyllanthus L. (2: 7. Phylanthos Americana Planta a es ad foliorum crenas proferans. Linnaeus (Sp. Pl. 981. 1753) cited Catesby, Hortus Cliffort., and others. Webster (Jour. Arnold Arb. 39: 109. 1958) stated that the type was Hortus Cliffort. 439.1 (Bm). Ewan (1974, p. 96) used the synonymous name Xy/o- phylla epiphyllanthus (L.) Britton for this illustration. icrodendron baccatum (L.) Krug & Urban (2: ¢. 32) Tapia trifolia fructu majore oblongo. Juglans baccata L. Syst. Nat. ed. 10. 1272. 1759. Linnaeus cited “Brown jam. 346. n. 2.” Ewan (1974, p. 96) suggested the synonymous name Picro- dendron macrocarpum (Rich.) Britton. FAGACEAE Castanea pumila (L.) Miller (1: 7. 9) Castanea pumila Virginiana, fructu racemato parvo in singulis capsults echinatis unico. Chinkapin Fagus pumila L. Sp. Pl. 998, 1753. Linnaeus cited Catesby, Gronovius, and others, indicating with an asterisk that a good description was in Gro- NOvVIUS. Quercus alba L. (1: 7. 2/, /efi) Quercus alba Virginiana. White Oak. Linnaeus (Sp. Pl. 996. 1753) cited Catesby and Gronovius. Quercus incana Bartram (1: ¢. 22 Quercus humilior salicis folio breviore. Highland Willow Oak. Dandy (1958, p. 112) stated that the plate is the type of Quercus phellos L., with a typotype specimen H.S. 212. f. 78. However, Linnaeus cited only Catesby’s ¢. 22 for his Quercus phellos var. y (Sp. Pl. 994. 1753). Michaux (Hist. Chénes Am. 8. ¢. /4. 1801) cited Catesby when he described Quercus cinerea Michx. Sargent (Silva 8: 172. 1895) raised an earlier variety to specific level as Quercus brevifolia Sarg.. noting that it was first illustrated and de- scribed by Catesby. Bartram’s name (Travels N. S. Carolina, 378. 1791) is the oldest and is currently accepted in many floras but does not appear to have been typified. Quercus laevis Walter (1: 7. 23 Quercus Esculi divisura foliis amplioribus aculeatis. Red Oak. The Catesby reference and others were cited by Linnaeus under his Quercus rubra L. (Sp. Pl. 996. 1753). The plate is also the type of Quercus catesbaei Michx. (Hist. Chénes Am. 17. tt. 29, 30. 1801), now considered a synonym of Q. laevis Walter (Fl. Carolin. 234. 1788), for which no specimen or ref- erence was cited in the original description. Ewan (1974, p. 92) identified the illustration as Quercus rubra L. Quercus marilandica Muenchh.* (1: ¢. 19) Quercus (forte) Marilandica, folio trifido ad Sassafras accedente. Black ak. Linnaeus (Sp. Pl. 996. 3) cited only Catesby under his Quercus nigra L. var. 6. Dandy (1958, p. i 12) stated that the plate is the type of Quercus — 526 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 nigra L. var. 8, with the typotype H.S. 232, f. 93. Muenchhausen (Hausvater 5: 252. 1770) cited only Catesby’s plate. Quercus nigra L. (1: ¢. 20, above) Quercus folio non serrato, In summitate quasi triangulo. Water-Oak. Linnaeus (Sp. Pl. 995. 1753) cited Catesby and Gronovius. Quercus phellos L. (1: 7. /6) aes An potius Ilex Marilandica folio longo angusto salicis? Willow ak iawadens (Sp. Pl. 994. 1753) cited Catesby, Gronovius, and Ray for the typical variety. (H.S. 212, f. 77 (upper); 232, f. 98.) Quercus prinus L. (1: ¢. /3) Quercus castaneae follis, procera arbor virginiana. Chesnut-Oak. Linnaeus (Sp. Pl. 995. 1753) cited Catesby. Hortus Cliffort., Gronovius, and others. A Hortus Cliffort. specimen is in the Linnean herbarium. (H:S. 232, f. 14.) Quercus rubra L. (1: ¢. 2/7, right) Quercus Caroliniensis virentibus venis muricata. White Oak, with pointed Notches innaeus (Sp. Pl. 996. ae cited Catesby, Hortus Cliffort., and others for Quercus rubra var. 8. (H.S. 212, f. 5 (right). Quercus virginiana Miller (1: ¢. Quercus sempervirens, foliis oblongis non sinuatis. Live-Oak. For Quercus phellos var. 3, Linnaeus (Sp. Pl. 994. 1753) cited only Catesby. Dandy (1958) stated that the plate 1s the type. with the typotype H.S. 212. f. 81 (lower). Trelease (Mem. Natl. Acad. Sci. 20: 157. ¢. 8. 1924) referred this variety to the synonymy of Quercus virginiana. FLACOURTIACEAE Banara minutiflora (A. Rich.) Sleumer (2: 4. 42, right) Frutex foltis oblongis serratis alternis, Acactae floribus luteis, fructu brevi, calyculato viridi. Britton and Millspaugh (1920, p. 284) and Ewan (1974. p. 96) referred Catesby’s illustration to Banara reticulata Griseb. (1860). Sleumer (Fl. Neo- trop. 22: 88. 1980) proposed the new combination based on the older (1845) Ilex minutiflora A. Rich. — GENTIANACEAE Gentiana catesbaei Walter (1: ¢. 70) Gentiana Virginiana, Saponariae. fae Hore coeruleo longiore. For Gentiana saponaria L. (Sp. Pl. 228. 1753) Linnaeus cited Catesby, Gronovius, and Morison. We are ne to find any information on the typification of this name, although a specimen collected by Kalm is in the Linnean herbarium. Catesby’s illustration appears to have the partially opened flowers of G. saponaria but was identified as G. cateshaci Walter by Ewan (1974, p. 94). Fernald (Rhodora 49: 175, 176. 1947) recognized G. cateshaei Walter after that name had been in the synonymy of G. saponaria L. for — 1983] HOWARD & STAPLES, CATESBY’S PLANTS S27 many years. Fernald saw a specimen in Walter’s herbarium, which he illus- trated. He commented: “Although Walter did not cite Catesby’s plate, his intent in giving the name of G. Catesbaei is pretty obvious.” Fernald quoted Curtis (Boston Jour. Nat. Hist. 1: 128. 1835) as having said “Tab. 70 of Catesby’s Carolina, represents it” and concluded, “There should be no ques- tion about the identity of G. Catesbaei.” A possible typotype specimen, H.S. 212, f. 87 (lower right), was not dissected, rendering impossible the use of corolla characters given in local floras. GOODENIACEAE Scaevola plumieri (L.) Vahl* (1: ¢. 7%) Lobelia frutescens, Portulacae folio. Lobelia plumierii L. Sp. Pl. 929. 1753. Linnaeus cited “Fl. Zey. 313, Plum. gen. 21,” and Catesby. Jeffrey (Kew Bull. 34: 543. 1980) designated Catesby’s plate as the lectotype. HAMAMELIDACEAE Hamamelis virginiana L. (2(App.): ¢. 2) Hamamelis. Linnaeus (Sp. Pl. 124. 1753) cited Catesby, Gronovius, and others. A specimen collected by Kalm is in the Linnean herbarium. (H.S. 212, f. 4 (right).) Liquidambar styraciflua L. (2: ¢. 65) Liquid-ambari Arbor, seu Styraciflua, Aceris folio, fructu Tribuloide, Le. Pericarpio orbiculari ex quam plurimis apicibus coagmentato, semen recon- dens. Sweet Gum-Tree. Linnaeus (Sp. Pl. 99. 1753) cited Catesby, Gronovius, Hortus Cliffort., Hortus Upsal., and others. A specimen collected by Kalm is in the Linnean herbarium. (H.S. 212, f. 79; 232, f. 34; BM-Dale.) JUGLANDACEAE Carya alba (L.) K. Koch (1: ¢. 38) Nux Juglans alba Virginiensis. Hiccory Tree. Following Rehder’s proposal (Jour. Arnold Arb. 26: 483. 1945) to consider Carva alba L. a nomen ambiguum, this species has generally been known as Carya tomentosa (Poiret) Nutt. As Rehder indicated, Linnaeus had two taxa in the original description of Juglans alba (Sp. Pl. 997. 1753), in which Catesby, Gronovius, and Plukenet were cited. However, Crantz (Inst. Rei Herb. 1: 157. 1766) listed only Catesby in his description of Juglans alba, thus typifying the Linnaean name by excluding the Plukenet and Gronovius references. Ewan (1974, p. 93) identified the figure as C. tomentosa (Poiret) Nutt. and stated ‘This identification applies to the fruit but the leaves are those of Carya glabra Mill. according to Dr. Donald E. Stone who says “if one added trichomes to the drawing of the shoot it would pass for mockernut. C. tomentosa is what George Edwards must have been referring to as Juglans 528 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 alba Lo” This unresolved problem we me specialists, who should examine a possible supporting specimen, H.S. 212, f. 3. Carya cordiformis (Wangenh.) K. Koch (1: 4. 38, aes fruit) Nux Juglans Carolinensis fructu minimo putamine levi. Pignut. Ewan (1974, p. 93) identified this illustration, stating, ““The description of the fruit fits this species according to Dr. Stone.” Juglans nigra L. (1: 4 67) Nux juglans nigra Virginiensis. Black Walnut. Linnaeus (Sp. Pl. 997. 1753) cited Catesby and Hortus Cliffort. (H.S. f. 94, f. 97.) i) wo bo LAURACEAE Ocotea coriacea (Sw.) Britton (2: ¢. 28, above) Cornus, foliis Salicis Laureae acuminatis; floribus albis; fructu Sassafras. Laurus coriaced Sw. (Prodr. 65. 1788) was based on material from Jamaica but occurs in the Bahamas (probably the source of Catesby’s material). Catesby’s plate 1s the type of Laurus cateshyana Michx. (Fl. Bor. Am. 244. 1803), transferred by Sargent to Ocotea (Silva N. Am. 7: 11. 1895) and later to Nectandra (Garden Forest 2: 448. 1899), and now considered to be a synonym of Ocotea coriacea (Sw.) Britton. Persea borbonia (L.) Sprengel* (1: 7. 63) Laurus Carolinensis, folijs acuminatis, baccis caeruleis, pediculis longis rubris insidentibus. Red Bay Laurus borbonia L. Sp. Pl. 370. 1753. Linnaeus cited Catesby, Hortus Cliffort.. and Gronovius. Catesby’s plate was designated the lectotype by Kopp (Mem. N. Y. Bot. Gard. 14: 44. 1966). (H.S. 212, f. 1.) Sassafras albidum (Nutt.) Nees var. molle (Raf.) Fern. (1: ¢. 55) Cornus mas odorata, folio trifido margine plano, Sassafras dicta. Laurus sassafras L. Sp. Pl. 371. 1753. Linnaeus cited Catesby, Hortus Cliffort., and others. (Bm-Dale.) Ts LEGUMINOSAE Acacia tortuosa (L.) Willd. (2: 1. 44) Arbor foltis pinnatis, spica pendula sericea. Mimosa tortuosa L. Sp. Pl. ed. 2. 1505. 1763. The type is a specimen collected by Browne in Jamaica (LINN 1228.27). Caesalpinia bahamensis Lam. (2: ¢. 5/, above) Pseudo-Santalum croceum. Brasiletto. Linnaeus (Sp. Pl. 380. 1753) cited Catesby, Plumier, Ray, and Sloane in the protologue of Caesalpinia brasiliensis. Urban (Symb. Antill. 2: 286. 1900) noted that three species were involved, one being “Caesalpinia bahamensis Lam. (ex Catesb. Car. t. 51. cum patria erronea Carolina). Lamarck (Encycl. Meth. Bot. 1: 461. 1785) had cited Miller, Sloane, and Catesby. Caesalpinia brasiliensis L.1s typified by a Plumier reference and is endemic to Hispaniola. The Sloane description given by Linnaeus is Peltophorum brasiliense Urban (Symb. Antill. 2: 286. 1900), which was proposed as a combination with the 1983] HOWARD & STAPLES, CATESBY’S PLANTS 329 basionym Caesalpinia brasiliense Sw. (not L.) (Obs. Bot. 166. 1791) but should be considered a new name credited to Urban. Fawcett & Rendle (FI. Jam. 4: 90. 1920) referred to a Browne specimen named by Linnaeus in the Linnean herbarium: however, cither the Sloane reference (Voy. Is. Nat. Hist. ll, 184. ¢. 231, figs. 3, 4) or the supporting specimen (if located) is more appropriate for typification. Dalbergia ecastophyllum (L.) Taub. (2: 4. 24) Arbor, Populi nigrae foltis, fructu reni-formi mono-spermo. Hedysarum ecastophyllum L. Syst. Nat. ed. 10. 1169. 1759. Ewan (1974, p. 95) identified this illustration as Ecastophyllum brownei Pers., which 1s distinguished on the basis of several characters that cannot be determined from Catesby’s description or illustration. No supporting specimen has been located, and the problem is unresolved. Erythrina herbacea L. (2: ¢. 49) Corallodendron humile, spica florum longissima coccinea, radice crassis- Simo. Linnaeus (Sp. Pl. 706. 1753) cited Catesby (f. 49), Royen, Dillenius (¢. 90), and Hortus Cliffort., with references to Carolina and “Missipi.” Krukoff (Brittonia 3: 284. 1939) cited Catesby and stated the type location to be Carolina. He noted that the specimens in the Linnean herbarium appeared in 1767 and therefore cannot be the type. “The description, the plates cited by Linnaeus and the fact that only one species of Erythrina is found in Carolina provide sufficient evidence for interpretation of the species” (Kru- koff, foc. cit.). Galactia rudolphioides (Griseb.) Bentham & Hooker (2: 1. 28, below) An Phaseolus minor lactescens flore purpureo. Dioclea rudolphioides Griseb. Mem. Am. Acad. II. 8: 178. 1860. The type is Wright 1187 from Cuba. Gleditsia aquatica Marsh. (1: ¢. 43) Acacia Abruae folijs, triacanthos, capsula ovali unicum semen claudente. Acacia. Gleditsia triacanthos var. 6, L. Sp. Pl. 1057. 1753. Linnaeus cited only Catesby. Marshall (Arbust. Am. 54. 1785) did not cite Catesby in the original description. (H.S. 212, f. 61 (upper).) Haematoxylon campechianum L. (2: /. 66) Lignum campechianum, species quaedum Brasil. Logwood. Linnaeus (Sp. Pl. 384. 1753) cited Catesby, Hortus Cliffort., and Sloane. The type is LINN 538.1. (H.S. 232, f. 81, f. 90.) Lysiloma latisiliquum (L.) Bentham (2: ¢. 42, left) Acacia, Buxi foliis rotundioribus, floribus albis, siliqua lata compressa. The discussions of Gillis and Stearn (Taxon 23: 185-191. 1974), De Wit (Taxon 24: 349-352. 1975), and Polhill and Stearn (Taxon 25: 323. 1976) are pertinent, although Linnaeus did not cite Catesby. Pithecellobium bahamense Northrop (2: ¢. 97) Acacia foliis amplioribus, siliquis cincinnatis. Linnaeus described Wimosa circinalis L. (Sp. Pl. 517. 1753) and cited only Plumier (Spec. 17) for a plant from Hispaniola. Later (Sp. Pl. ed. 2. 1500. 530 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 1763), he added references to Catesby and Browne. This taxon as Pithecel- lobium circinalis (L.) Bentham is considered endemic to Hispaniola. Britton and Millspaugh (1920, p. 156) referred Catesby’s illustration to Pithecolo- bium mucronatum Britton ex Coker, but this species has been considered a synonym of Pithecellobium bahamense Northrop in the Corrells’ Flora of the Bahama Archipelago (1982). Robinia hispida L. (2(App.): ¢. 20) Pseudo Acacia ee ene roseis. Linnaeus (Mant. PI. 101. 1767) cited Catesby, Jacquin, and Miller. LOGANIACEAE Gelsemium sempervirens (L.) Aiton (1 Gelseminum sive Jasminum luteum ordoratum Virginianum scandens, semper VIFens. Bignonia se ileal L. Sp. Pl. 623. 1753. Linnaeus cited Catesby, Ray, Royen, and P net. Spigelia ie (L.) L. (corrected reference) (2: ¢. 78) Gentiana forte? quae Periclymeni Virginiani flore coccineo, Planta Mari- landica spicata erecta, foliis conjugatis. Indian Pink Lonicera marilandica L. (Syst. Nat. 12: 166. 1767). No specimens or references were cited. In the same volume (p. 734), Linnaeus transferred the taxon to Spigelia, a combination generally overlooked in current floras. There are no specimens in the Linnean herbarium, and a neotype should be des- ignated. (H.S. 212, f. 33 (middle).) LORANTHACEAE Dendropemon purpureum (L.) Krug & Urban* (2: 7. 95, below) Viscum foliis latioribus,; baccis purpureis pediculis incidentibus. Viscum purpureum L. Sp. Pl. 1023. 1753. Linnaeus cited only Catesby. No supporting specimen has been located. Phoradendron rubrum (L.) Griseb.* (2: 7. 8/, below) Viscum foliis longioribus baccis rubris. Viscum rubrum L. Sp. Pl. 1023. 1753. Linnaeus cited only Catesby, and Dandy (1958, p. 112) stated that the plate is the type. (H.S. 232, f. 2 (lower right), f. 8 (upper left specimen marked lectotype), f. 16 (upper left).) MAGNOLIACEAE Liriodendron tulipifera L. (1: 4. 438) Arbor Tulipifera Virginiana tripartito aceris folio, media lacinia velut ab- SCISSQ. Linnaeus (Sp. Pl. 535. 1753) cited Catesby, Hortus Cliffort., and others. (HS. 2 121, 80.) Magnolia acuminata (L.) L.* (2(App.): ¢. 75) Magnolia flore albo, folio majore acuminato 2 albicante. Magnolia virginiana var. « acuminata Sp. Pl. 536. 1753. Catesby and Gronovius are cited. 1983] HOWARD & STAPLES, CATESBY’S PLANTS 531 Magnolia acuminata (L.) L. Syst. Nat. ed. 10. 2: 1082. 1759. Linnaeus cited Catesby and Kalm. Dandy (1958, p. 112) stated that the plate is the type of Magnolia acuminata (L.) L. Magnolia grandiflora L. (2: ¢. 6/) Magnolia altissima, flore ingenti candido. Laurel Tree of Carolina. Linnaeus (Syst. Nat. ed. 10. 1082. 1759) cited Catesby, Trew, and others. Magnolia tripetala (L.) L.* (2: ¢. 80) Magnolia, amplissimo flore albo, fructu coccineo. Umbrella Tree. Magnolia virginiana var. tripetala L. Sp. Pl. 536. 1753. Linnaeus cited only Catesby. Magnolia tripetala (L.) L. Syst. Nat. ed. 10, 2: 1082. 1759. Linnaeus cited only Catesby, and Dandy (1958, p. 112) stated that the plate is the type. Ewan (1974, p. 98) referred this illustration to Magnolia macrophylla Michx., a species with a very different leaf shape. Magnolia virginiana L. (1: ¢. 39) Magnolia Lauri folio, subtus albicante. Sweet flowring Bay. Magnolia virginiana L. var. glauca L. Sp. Pl. 535. 1753. Linnaeus cited Catesby, Dillenius, and Plukenet. Magnolia glauca (L.) L. Syst. Nat. ed. 10. 2: 1082. 1759. Here Linnaeus cited Trew, Ehret, Catesby, Dillenius. and Plukenet. Dandy (1958, p. 112) stated that the plate is the type of Magnolia glauca (L.) L.* MALVACEAE Hibiscus tiliaceus L. (2: ¢. 90) Ketmia, amplissimo Tiliae folio, subtus argenteo, Flore magno luteo. Maho- ree. Linnaeus (Sp. Pl. 694. 1753) did not cite Catesby. Ewan (1974, p. 99) incorrectly referred this illustration to Thespesia populnea (L.) Solander. Fryxell (pers. comm.) recognizes the native western hemisphere plant as Hibiscus pernambucensis Arruda. The supporting Catesby specimen, HS. 232, f. 120 (lower), agrees with the description of this species. Phymosia abutiloides (L.) Ham. (1: ¢. 77 Ketmia frutescens glauca, Aceris majoris folio longiore, serrato, flore car- neo. Valva abutiloides L. Sp. Pl. ed. 2.971. 1763. Linnaeus cited only Dillenius, “elth. 1, t. 1, f 1.° (HS. 232, f. 120 (upper right); BM-Dale.) MELIACEAE Swietenia mahagoni (L.) Jacq. (2: ¢. 8/, above) Arbor foliis pinnatis, nullo impari Alam claudente, nervo ad latus unum excurrente, fructu anguloso magno, semine alato instar Pinus. Mahogany Tee. Cedrela Mahag. L. Syst. Nat. ed. 10. 2: 940. 1759. Linnaeus cited only Catesby, and Dandy (1958, p. 112) stated that the plate is the type of Cedrela mahagoni L. Pennington (in Pennington & Styles, Fl. Neotrop. 28: 401. 1 981) agrees. a32 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 MENISPERMACEAE Cocculus carolinus (L.) DC. (1: ¢. 57) Smilax (forte) lenis, folio anguloso hederaceo. Menispermum carolinum L. Sp. Pl. 340. 1753. Linnaeus cited no references for this taxon, and no plant material on which it was based Is still in existence. However, under Cissampelos smilacina L. (Sp. Pl. 1032. 1753) Linnaeus cited only Catesby, so the illustration is the type of that specific name. The latter taxon has been of uncertain placement. Asa Gray (Synopt. Fl. N. Am. 1: 64-66. 1895) cited Catesby’s plate and referred C. smilacina L. to Cocculits carolinus (L.) DC. However, according to both the entry in pa Kewensis and Diels (in Pflanzenr. IV. 94: 257. 1910). C. smilacina is Menispermum canadense L. Ewan (1974, p. 93) stated, “Though Cocculus carolinus (L.) . occurs in the region visited by Catesby there is little doubt this is Menispermum [canadense L.].”” However, Catesby illustrated and described red fruits (characteristic of Cocculus), while eae eae has black fruits. Among possible supporting specimens, H.S. ae 4] ae Te 104 are Cocculus, but H.S. 212, f. 21 is a Menispermum. i wo that either a neotype should be designated for Cocculits ceo carolinus or, less desirably, that the name of this well-known species should be changed to utilize the basionym Cissampelos smilacina, which is typified by Catesby’s plate. MONOTROPACEAE Monotropa uniflora L. (1: 7. 36) Orobanche Virginiana flore pentapetalo cernuo. Broom-rape. Linnaeus (Sp. Pl. 387. 1753) cited Catesby and Gronovius. MORACEAE Ficus citrifolia Miller (2(App.): ¢. 7) Ficus citrii folio, fructu parvo purpureo. Ficus indica L. var. B., L. Sp. Pl. 1060. 1753. Linnaeus cited Catesby, Hortus Cliffort., Sloane. and others. According to De Wolf (i Woodson, FI. Panama, IV. 2: 160. 1960), the taxon is typified by ‘ta specimen in Herb. Banks.” Ewan (1974, p. 100) identified this illustration as Ficus brevifolia Nutt., a name now considered to be a synonym. MyYRICACEAE Myrica cerifera L. (1: ¢. 69%) Myrtus, Brabanticae similts, Carolinensis, He ae fructu racemoso sessilt monopyreno. Narrow-leaved Candle-berry Myrtle Linnaeus (Sp. Pl. 1024. 1753) cited ena Gea Cliffort.. Hortus Up- sal., Gronovius, and others. (BM-Dale.) a 1983] HOWARD & STAPLES, CATESBY’S PLANTS a33 Myrica pensylvanica Loisel (1: ¢. /3) Myrtus Brabanticoe similis Caroliniensis humilior; foliis latioribus & ma- gis Serratis. Broad-leaved Candle-berry Myrtle. Linnaeus (Sp. Pl. 1024. 1753) cited only Catesby under W. cerifera L. var. 3. Loisel (in Duhamel, Traité Arbres Arbustes, ed. augm. 2: 190. ¢. 55. 1804) cited Catesby but also published a plate. The type is then either Catesby’s or Loisel’s illustration. (H.S. 232. f. 50.) NYSSACEAE Nyssa aquatica L.* (1: ¢. 60) Arbor in aqua nascens, folijs latis acuminatis & dentatis, fructu Eleagni majore. Water- Tupelo. Linnaeus (Sp. Pl. 1058. 1753) cited Catesby, tt. 47 and 60, as well as Hortus Cliffort.. Gronovius, and Plukenet. Eyde (Taxon 13: 131. 1964) dis- cussed the complicated problem. We accept his implication that Catesby’s t. 60 is the best option for the type. Supporting specimens are H.S. 212, f. 67 (left), and 232, f. 52 (right). Ewan (1974, p. 94) has identified this illus- tration as Nyssa ogeche Bartram, a very different plant. Nyssa sylvatica Marsh. (1: ¢. 4/) Arbor in aqua nascens, folijs latis acuminatis & non dentatis fructu Eleagni minore. Tupelo Tree. As indicated above, Linnaeus (Sp. Pl. 1058. 1753) cited this plate under Nyssa aquatica. Eyde (Taxon 13: 130, 131. 1964) stated, ‘in the absence of a specimen, Marshall’s description is the type.” However, Catesby’s ¢. 4/ must be the lectotype—not Marshall’s “description.” Ewan (1974, p. 93) identified this illustration as Nyssa aquatica L. OLEACEAE Chionanthus virginicus L. (1: 7. 68) Amelanchior Virginiana, Lauro cerasifolio. Fringe Tree. Linnaeus (Sp. Pl. 8. 1753) cited Catesby and Hortus Cliffort. Fraxinus americana L. (1: ¢. 80) Fraxinus Carolinensis, folijs angustioribus utrinque acuminatis, pendulis. Linnaeus (Sp. Pl. 1057. 1753) cited Catesby, Gronovius, and Ray. A spec- imen collected by Gronovius is in the Linnean herbarium. (H.S. 212, f. 11 (lower).) Osmanthus americanus (L.) Gray (1: 7. 6/) Ligustrum Lauri folio, fructu violaceo. Purple-berried Bay. Olea americana L. Mant. Pl. 24. 1767. Linnaeus cited only Catesby, but he indicated a specimen from *“D. Guarden.” Peter Green (Notes Roy. Bot. Gard. Edinburgh 22: 463. 1958) has designated an unmarked sheet in the Linnean herbarium from Dr. Garden as the type specimen. The combination in Osmanthus has been incorrectly attributed to Bentham and Hooker (Gen. Pl. 2: 677. 1876), who only implied it by listing the species under Osmanthus. (H.S. 212, f. 22 (upper).) 534 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 PASSIFLORACEAE Passiflora cuprea L.* (2: ¢. 93, below) Grandilla, foliis Sarsaparillae trinerviis; flore purpureo, fructu Olivaeformi caeruleo. Linnaeus (Sp. Pl. 955. 1753) cited Amoen. Acad. (1: 218. fig. 2. 1749), Dillenius, Martens, and Catesby. Theoretically, any of the plates is available for typification. Ewan (1974, p. 99) noted that Catesby’s plate 1s “‘cited in Sp. pl. where three additional references are cited but the ‘Habitat’ is given only as ‘Providentia, Bahama.’ and thus fixes the type in the present-day type concept for the Linnaean name.” (H.S. 232, f. 22 (lower center).) Passiflora suberosa L. (2: ¢. 51, below) Flos Passionis minimus, trilobatus flore sub-caeruleo. Linnaeus (Sp. Pl. 958. 1753) did not cite Catesby. Ewan (1974, p. 97) identified this illustration as Passiflora pallida L. and cited Killip (Publ. Field Mus. Bot. 19: 88-97. 1938), although Killip in that publication accepted Passiflora suberosa L. and listed P. pallida L. as a synonym. Killip (p. 97) rejected “rules of page priority [that] would require the use of the name P. pallida” and noted but did not cite as a type the specimen in the Linnean herbarium that Linnaeus had in 1753. (Bm-Dale.) PLATANACEAE Platanus occidentalis L. (1: ¢. 56 Platanus Occidentalis. Western Plane Tree. Linnaeus (Sp. Pl. 999. 1753) cited Catesby and Hortus Cliffort. (H.S. 212, f. 68 (upper).) POLYGONACEAE Coccoloba diversifolia Jacq. (2: t. 94) Cerasus latiore folio; fructu racemoso purpureo majore. Pigeon-Plum. In the second edition of Species Plantarum (1: 677. 1762), Linnaeus cited Catesby’s r. 94, Gronovius, and Royen in the protologue of Prunus virginiana (q.v.), but not Catesby’s 1: 4. 28 or Plukenet as he had done in the first edition. Coccoloba uvifera (L.) L. (2: £. 96) Prunus maritima racemosa, folio rotundo glabro; fructu.minore purpu- reo. Mangrove Grape Tree. Polygonum uvifera L. Sp. Pl. 365. 1753. Linnaeus cited Catesby and Hortus Cliffort. (H.S. 232, f. 8 (lower right).) PRIMULACEAE Dodecatheon meadia L.* (2(App.): ¢. /) Meadia Linnaeus (Sp. Pl. 144. 1753) cited only Catesby and Plukenet (Alm. Bot. 62. t. 79. fig. 6. 1696). Dandy (1958, p. 112) stated that the plate is the type, thus designating a lectotype by choosing Catesby and rejecting Plukenet. 1983] HOWARD & STAPLES, CATESBY’S PLANTS 535 RHAMNACEAE Colubrina elliptica (Sw.) Briz. & Stern (1: 4. /0) rutex Lauri folio pendulo, fructu tricocco, semine nigro splendente. Red- Wood. Ewan (1974, p. 92) identified this illustration as Colubrina reclinata (L’Hér.) Brongn., which is now regarded as a synonym of C. elliptica. Reynosia septentrionalis Urban (1: ¢. 75) Prunus Buxi folio cordato, fructu nigro rotundo. Bullet-Bush. Sargent (Silva N. Am. 2: 21. ¢. 56. 1891) noted that Catesby’s plate was the earliest account of Reynosia /atifolia Griseb. Urban (Symb. Antill. 1: 356. 1899) separated the Bahamian and the Cuban plants, although both species are recognized in the current flora of Cuba. RHIZOPHORACEAE Rhizophora mangle L. (2: ¢. 63) Candela alleles foliis Laurinis, flore tetrapetalo luteo, fructu angus- tiore. Mangrove Tre Linnaeus (Sp. Pl. ee 1753) did not cite Catesby. ROSACEAE Prunus virginiana L. (1: ¢. 23) Cerasi similis arbuscula Mariana. Poedi folio, flore albo parvo racemoso. Cluster’d black Cherry. Linnaeus (Sp. Pl. 473. 1753) cited Catesby’s ¢. 28, Gronovius, Royen, and “Pluk. mant. 43. t. 339.” He later (Sp. Pl. ed. 2. 1: 677. 1762) changed the references, adding Catesby 2: t. 94 (Coccoloba diversifolia Jacq.) and eliminating mention of ¢. 28. (OXF-S.) RUBIACEAE Casasia clusiifolia (Jacq.) Urban (1: ¢. 59) Arbor Jasmini, floribus albis, dis Cenchranmideoe, fructu ovali, semi- nibus parvis nigris mucilagine involutis. Seven Years Apple Catesbaea spinosa L.* (2: ¢. /00) Frutex Spinosus Buxi foliis, plurimis simul nascentibus, flore tetrapeta- loide, pendulo, sordide flavo, tubo longissimo; fructu ovali croceo, semina parva continente. Linnaeus (Sp. Pl. 109. 1753) cited only Catesby. Dandy (1958, p. 112) stated that the type of C. spinosa L., H.S. 232, f. 21 (upper left and lower right), is thus the type of the specific and generic names. The generic name is correctly attributed to Linnaeus (Farr et a/., Index Nom. Gen. 1: 303. 1979), although Catesby stated “Catesbaeat. Lycium Catesbeil, Authore D Gronovio” and, in a footnote, “It is not without Reluctancy, that I here exhibit a Plant with my own Name annexed to it; but the Regard and Obligations I owe to my learned Friend Dr. J. F. Gronovius of Leyden, who 536 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 was pleased some Years since to honor me, tho’ undeservedly, with the Title of this Genus, obliges me not to suppress it.” The binomial ““Lycium Catesbeii” has not been located in Gronovius’s publications. In addition to the above- mentioned specimens in the Sloane herbarium, there is a better one in the general herbarium at BM from Gronovius. Mitchella repens L. (1: ¢. 20, below) Myrti subrotundis folijs, floribus albis gemellis ex provincia Floridana. Sy- ringa Baccifera. Linnaeus (Sp. Pl. 111. 1753) cited Catesby. Gronovius, and others. RUTACEAE Amyris elemifera L. (2: 1. 33, below) Frutex trifolius resinosus, floribus tetra-petalis albis racemosis. Linnaeus (Syst. Nat. ed. 10. 1000. 1759) cited Catesby and Hortus Cliffort. Fawcett and Rendle (Fl. Jamaica 4: 191. 1920) suggested that a specimen in the Linnean herbarium collected by Browne was the type. Ptelea trifoliata L. (2: 7. S3) “rutex Virginianus trifolius, Ulmi Samaris; Banisteri. Linnaeus (Sp. Pl. 118. 1753) did not cite Catesby. (H.S. 212. f. 66 (upper left): 232, f. 539 Zanthoxylum clava-herculis L. (1: 6. 26) Zanthoxylum spinosum, Lentisci, longtoribus folijs Euonimi fructu. cap- sulart ea Insula Jamaicensi. Pellitory, or Tooth-ach Tree Linnaeus (Sp. Pl. 270. 1753) cited Catesby and Hortus Cliffort. in the protologue and gave the distribution as Jamaica, Carolina, and Virginia. We suspect that Catesby’s reference to Jamaica is in error. The only associated specimen (BM-Dale) 1s Z. clava-herculis. Zanthoxvlum clava-herculis Sw. (Obs. Bot. 375. 1791), not L. (1753), from Jamaica has been referred to the synonymy of 4. martinicense DC. SALICACEAE Populus heterophylla L. (1: 7. 34) Populus nigra folio maximo gemmis Balsamum odoratissimum funden- tibus. Black Poplar of Carolina. For Populus balsamifera L. (Sp. Pl. 1034. 1753) Linnaeus cited Catesby and Hortus Cliffort. Rouleau (Rhodora 48: 103-110. 1946) concluded that Catesby illustrated Populus heterophylla but described a mixture of P. het- erophylla and P. deltoides. Rouleau cited a specimen in the British Museum, which may have been HS. 212, f. 11 (upper) or 232, f. 52 (left, 1 leaf). SAPOTACEAE Manilkara bahamensis Lam & Mecuse (2: ¢. 87, above) Anona foltis Laurinis, in summittate incisis; fructu compresso scabro fusco, in medio acumine longo. Sappadillo Tree. For Sfoanea emarginata L. (Sp. Pl. 512. 1753) Linnaeus cited only Cates — oy 1983] HOWARD & STAPLES, CATESBY’S PLANTS 537 but incorrectly attributed the plant to Carolina. Dandy (1958, p. 112) stated that the plate was the type of S/oanea emarginata L.*, with the typotype ES. 232. f..15. Achras bahamensis Baker (in J. D. Hooker, Ic. Pl. 18. ¢. 7795. 1888) was typified by Eggers 3837, from Fortune Island, and Catesby’s plate was cited, but Baker was unaware of Linnaeus’s description of the plant in the genus Sloanea. Manilkara bahamensis Lam & Meeuse (Blumea 4: 354. 1941) ts therefore to be treated as a new name and not a new combination (Art. 72, Note 1). The Linnaean specific epithet cannot be transferred as Manilkara emarginata (L.) Britton & Wilson (Sci. Survey Porto Rico Virgin Is. 6: 366. 1926) because it would be a later homonym of VW. emarginata Lam (Bull. Jard. Bot. Buitenzorg, III. 7: 241. 1925), a plant from the Hawaiian Islands. Mastichodendron foetidissimum (Jacq.) Lam (2: ¢. 75) Cornus, foliis Laurinis, fructu majore luteo. Mastic Tree. Ewan (1974, p. 98) suggested Sideroxvion foetidissimum Jacq. for this illustration. That name serves as the basionym. SARRACENIACEAE Sarracenia X catesbaei (Ell.) Bell* (2: 6. 69) Sarracena, foliis longioribus & angustioribus; Bucanephyllon elatius Vir- ginianum, &c. atesby, Hortus Cliffort., and Gronovius were cited under Sarracenia flava L. (Sp. Pl. 510. 1753). Elliot (Sketch Bot. S.-Carolina Georgia 2: 11. 1824) cited Catesby. Bell (Jour. Elisha Mitchell Sci. Soc. 68: 60. 1952) discussed the hybrid status suggested for Catesby’s plant. (H.S. 212, f. 20, f. 21, f 45.) Sarracenia purpurea L. (2: /. 70 Sarracena, foliis brevioribus latioribus. Sarracena Canadensis, foliis cavis & auritis. Linnaeus (Sp. Pl. 510. 1753) cited Catesby, Hortus Cliffort., and Gro- novius. (BM-Dale.) SAXIFRAGACEAE Philadelphus inodorus L.* (2: ¢. 84, above) Philadelphus flore albo majore inodoro. Linnaeus (Sp. Pl. 470. 1753) cited only Catesby. Dandy (1958, p. 112) stated that the plate is the type. with the typotype H.S. 212, f. 16 (upper right). (OXF-DB.) STERCULIACEAE Theobroma cacao L.* (2(App.): ¢. 6) Cacao Arbor. Cacao Tree. Linnaeus (Sp. Pl. 782. 1753) cited Catesby, Hortus Cliffort., and others. Cuatrecasas (Contr. U. S. Natl. Herb. 35: 496. 1964) selected H.S. 5, f. 59 as the lectotype. 538 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 STYRACACEAE Halesia tetraptera Ellis (1: ¢. 64) Frutex, Padi folijs non serratis, floribus monopetalis albis, campani-for- mibus, fructu crasso tetragono. Halesia carolina L. (Syst. Nat. ed. 10. 1044. 1759). Linnaeus cited Catesby and material collected by Ellis. Reveal and Seldin (Taxon 25: 123-140. 1976), discussing in detail the identity of Linnaeus’s species, concluded (p. 127) that it was different than the one illustrated by Catesby, and that Linnaeus had based his description of #/. carolina “only on the material sent to him by Ellis. The Catesby illustration is somewhat difficult to associate with a given species only because it 1s not entirely accurate. However, it appears to rep- resent the Coastal Plain phase of Halesia carolina, sensu authors, which is now infrequently encountered near Charleston. Little can be gleaned from Catesby’s rather long description.” SYMPLOCACEAE Symplocos tinctoria (L.) L’Hér. (1: 4. 54) Arbor lauri folio, floribus ex foliorum, alis pentapetalis, pluribus staminibus donatis. Hopea tinctoria L. (Mant. Pl. 105. 1767; Syst. Nat. ed. 12. 509. 1767). The two references are considered to have been published simultaneously. Linnaeus cited Catesby and appears to have credited the genus to Garden but called the illustrated plant “mala.” A specimen in the Linnean herbarium (942.1) may be from Dr. Garden and is in fruit. A specimen collected by Catesby and in the Gronovius herbarium (BM) is in flower. Although Linnaeus described both flowers and fruit, we designate the specimen in his herbarium as the lectotype. THEACEAE Gordonia lasianthus (L.) Ellis (1: 7. 44) Alcea Floridana quinque capsularis Laurinis folijs, leviter crenatis, semi- nibus coniferarum instar alatis. Loblolly Bay. In the protologue of Hypericum lasianthus L. (Sp. Pl. 783. 1753), Linnaeus cited Catesby, Hortus Cliffort., and others. Later, Linnaeus (Mant. Pl. 2: 570. 1771) used Gordonia lasianthus and cited the second edition of Species Plantarum (1763, p. 1101). The generic name is attributed to Ellis (Philos. Trans. 60: 520. 1771) and is conserved (#5148). (H.S. 212, f. 13 (lower); 232, f. 50 (right); BM-Dale; OxF-s.) Stewartia malacodendron L. (2(App.): ¢. /3) Steuartia. Linnaeus (Sp. Pl. 698. 1753) cited Catesby and “Act. Ups. 1741, 79, t. 2 THEOPHRASTACEAE Jacquinia keyenis Mez (1: ¢. 98) Frutex Buxi foltis oblongis, baccis pallide viridibus apice donatis. Soap- Wood. Catesby’s illustration was cited by Britton and Millspaugh (1920, p. 317). FiGure l. ks Lins de Catesby’s illustration of ““Frutex aquaticus, floribus luteis. fructu rotundo quinque-capsulari” (Vol. 2. ¢. 30). Unidentified. SLNV 1d S.APSALVO ‘SATAIV.LS ¥ GUVMOH [E861 cS 540 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 VERBENACEAE Avicennia germinans (L.) L. (1: ¢. 35) Frutex Bahamensis foltis oblongis succulentis, fructu subrotundo unicum nucleum continente. Bontia germinans L. (Syst. Nat. ed. 10. 1122. 1759; Sp. Pl. ed. 2. 891. 1763). Lectotype: Browne (LINN 813.2), according to Stearn (Kew Bull. 1958: 34. 1958). Ewan (1974, p. 95) identified this illustration as Avicennia nitida Jacq., a synonymous name. Callicarpa americana L. Frutex baccifer, erie ans: folijs scabris latis dentatis, & conjugatis; bac- cis purpuris dense congestts. Linnaeus (Sp. Pl. 111. 1753) cited Catesby and Gronovius. A specimen collected by Gronovius is LINN 136.1. VITACEAE Cissus tuberculata Jacq. (2: 1. 43) Frutex Rubo similis, non spinosus, capreolatus, fructu racemoso caeruleo Mori-formi. UNPLACED ILLUSTRATIONS 2: ¢. 30, above. Frutex aquaticus, floribus luteis, fructu rotundo quinque-cap- sulari. Edwards and Forster were unable to identify this illustration (see FiGure 1). Ewan (1974, p. 96) suggested with reservations Samo/lus ebracteatus HBK. (Primulaceae), stating that, “The inflorescence as drawn bears a striking re- semblance.” Catesby (2: 30) described this plant as follows: ese Plants grow usually about twelve Feet high, arising with innumerable small Stems, alternately bent, from which shoots forth smaller Twigs, set with small pointed smooth Leaves; the Flowers grow on the Tops of the Branches before they open, being inclosed in small brown pointed Periantheums, set on short Foot-stalks; are hexapeta- lous, of a deep yellow Colour, with long pices of the same Colour: They grow in shallow, fresh Water Ponds, in the Woods of Virginia and Carolina, a in the beginning of February adorn the Woods. when few other Plants appear in Blossom: The Flowers are succeeded by small round Capsula’s, which in March and Apri/ divide into four Parts and disclose their Seeds, which are very small, and being dispersed by the Wind into watery Places, spring up very thick and blossom in a short Time. As drawn. the flowers appear to resemble an Hfypericum, and the habit perhaps Hypericum densiflorum Pursh. However, the alternate leaves, the flow- ering time, and the description of the fruit negate this suggestion. No supporting specimens were located. 2: ¢. 52, plant. Frutex Lauri longiore folio. Ewan (1974, p. 97) suggested Decumaria barbara L. for this illustration (see FiGureE 2), recognizing aberrancies in the habit, the leaves, and the shape of the flower. Catesby (2: 52) described the plant as follows: FIGURE 2. Catesby’s illustration of “Frutex Lauri longiore folio” (Vol. 2, 4. 52). Unidentified. SLNV Id S.APSALVO ‘SATAIVLS ¥ CGUVMOH [£861 vs 542 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 This Shrub is a Native of Virginia, and grows in wet Swamps and standing Waters; it rises from the Ground, with many Stems to the Height of eight or ten Feet, of a reddish Colour. The Leaves are placed alternately an Inch from one another, and are in Shape like those of a Bay, stiff and shining: at the Pedicles of the Leaves grow the Flowers, which are tubulous, of a pale red Colour, and set on Stalks two and three Inches long; these Flowers are succeded by small conic Seed-Vessels about the Size of large Peas, which when ripe, open in two Parts and display many small black Seeds. It retains the Leaves all the Winter. If the bilobed capsule, as drawn, is associated with what appears to be a tubular corolla, the plant may be in the Apocynaceae. We know of no plant that agrees with the characters given in the description. We doubt that it is Decumaria. Again, no supporting specimens were locate LITERATURE CITED feta M. 1763[-1764]. Ne des plantes. Vols. 1, 2. Paris. Auten, E.G. 1937. New light on Mark Catesby. Auk 54: ne 363. Bossu, J.-B. 1771. Travels TaN that part of poe America formerly called Loui- siana. (Translated by J. R. Forster.) Vols. 1, on. Britton, N. L., & C.F. MitispauGu. 1920. The eat flora. New Era Printing Co.., La em Pennsylvania. CATESBY, {1730- ereads 1747]. aa natural history of Carolina, Florida and the ae Islands. Vols. 1, 2. Lon Crokir, H. M. 1964. An account of the = ne es Department of Botany in the University of a Oxford Univ. Press, Londor Corrett, D. S., & H. B. Corrett. 1982. Flora ae Bahama Archipelago. J. Cra- mer, Vaduz, an Danby, J. E. 1958. The Sloane Herbarium. British Museum, London. Ewan, J. 1974. Notes. Pp. 89-100 in The natural history of Carolina, Florida and the Bahama ee ee the late Mark Catesby. Beehive Press, Savannah, Georgia GRONovIUS, J. F. , 1743. Flora an Leiden 1762. ae earn ed 2. Leider Linnaeus, C, 1753. Species plantarum. Vols. 1, 2. Stockholm. 1756. Centuria II. Plantarum. eae: —. 1762, 1763. Species olan acim: ed. 2. Vols. 1, 2. Stockholm. ieee ape plantarum altera. Stockho Piumier, C. 1755-1760. Plantarum americanarum. tt BURMANN, ed.) Amsterdam, Leiden. Reveat, J.L. 1983. Significance of pre-1753 botanical explorations in temperate North America on Linnaeus’ first edition of Species Plantarum. Phytologia 53: 1-96. SLOANE, H. 1707, 1725. aos to the islands Madera, Barbados, Nieves, S. Chris- tophers and Jamaica. Vols. |, 2. London. StarLeu, F. A. 1967. Taxonomic cones Reg. Veg. 52: STEARN, W. T. 1958. Publication of Catesby’s Natural History of Carolina. Jour. Soc. Bibliogr. Nat. Hist. 3: 328. STEVENSON, A. L961. Sala orue of botanical books in the collection of Rachel Mc- Ma sters Miller Hunt. Vol. 2, part 2. Hunt Botanical Library, Pittsburgh, Pennsyl- 1983] HOWARD & STAPLES, CATESBY’S PLANTS 543 9. 10. Be [3 14. 15. 16. 17. 18. Lo: 20, 20, 21. rae a 23. 24. 25: 26. pie 28. 29 30. a a2; 33. 34. ao 36. one 38. 32; 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. SO. 51. ape 53. 54. 55. AppENp1x. Determinations of Catesby’s Illustrations of Plants VOLUME 1 Castanea pumila (L.) Miller— Fagaceae Colubrina elliptica (Sw.) Briz. & Stern—Rhamnaceae Taxodium distichum (L.) Rich. eee Myrica pensylvanica eae icaceae Oryza sativa L.— Gramineae Smilax laurifolia L.— Smilacaea Quercus phellos L.— Fagacea Quercus virginiana Miller—Fagacsi Quercus prinus L.—Fa Quercus marilandica Vat eee above. Quercus nigra L.— Fagaceae below. Mitchella repens L.— Rubiaceae lefi. Quercus alba L.— Fagaceae right. Quercus rubra L.—Fagaceae Quercus incana Bartram (Quercus cinerea Michx.*)—Fagaceae Quercus laevis Walter (Quercus catesbaei Michx.*)— Fagaceae Podophyllum peltatum L.—Berberidaceae Chrysobalanus icaco L.—Chrysobalanaceae Zanthoxylum clava-herculis L.—Rutaceae Cornus florida L. forma rubra (Weston) Schelle—Cornaceae Prunus virginiana L.— Rosaceae Aristolochia serpentaria L.—Aristolochiaceae Bursera simaruba (L.) Sarg.— Burseraceae Ilex cassine L.— Aquifoliacea Uniola aan L.—Gramineae Hypoxis sp.— Hypoxidaceae padi: Ee vophelia L.—Salicaceae Ipomoea sagittata eRe eke Monotropa uniflora L.—Monotropacea Tabebuia bahamensis (Northrop) Beane Bignoniaceae Carya alba (L.) K. Koch—Juglandaceae; Carya cordiformis (Wangenh.) K. Koch— Juglandaceae (smaller fruit) Magnolia virginiana L. (Magnolia virginiana var. glauca L.*)—Magnoliaceae Metopium toxiferum (L.) Krug & Urban—Anacardiaceae Jacaranda caerulea ue ) Griseb.* (Bignonia caerulea L.*)—Bignoniaceae Gleditsia aquatica Marsh.— Leguminosae Gordonia lasianthus (L .) Ellis— Theaceae Trillium catesbaei Ell.*— Liliaceae Calycanthus floridus L.—Calycanthaceae Smilax pumila Walter—Smilacaceae Liriodendron tulipifera L.—Magnoliaceae Catalpa bignonioides Walter— Bignoniaceae Trillium maculatum Raf.*—Liliaceae Cocculus carolinus (L.) DC. (Cissampelos smilacina L.*)—Menispermaceae Smilax tamnoides L.—Smilacaceae Gelsemium sempervirens (L.) Aiton—Loganiaceae Symplocos tinctoria (L.) L7Hér.—Symplocaceae Sassafras albidum (Nutt.) Nees var. molle (Raf.) Fern.— Lauraceae *Names determined to be based on Catesby’s illustrations. . Croton eluteria (L.) Sw.* . Callicarpa americana L.— Verbenaceae JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 . Platanus occidentalis L.— Platanaceae . Rhododendron viscosum (L.) Torrey var. ners Rehder— Ericaceae above. Cleistes divaricata (L.) Ames—Orchidace , below. Echites umbellata Jacq.— Apocynaceae . Casasia clusiifolia (Jacq.) Urban— Rubiaceae ok . Nyssa aquatica L.*—Nyssaceae . Osmanthus americanus (L.) Gray—Oleaceae Acer rubrum L.— Aceraceae . Persea borbonia (L.) Sprengel* (Laurus borbonia L.*)— Lauraceae Halesia tetraptera Ellis—Styracaceae mpsis radicans (L.) Seem.— Bignoniaceae . Clethra alnifolia L.—Clethraceae Juglans nigra L.—Juglandaceae . Chionanthus virginicus L.—Oleaceae . Myrica cerifera L.—Myricaceae . Gentiana catesbaei Walter—Gentianacecae . Oxydendrum arboreum (L.) DC.*—Ericaceae . Salmea petrobioides Griseb. —Compositae 5. Reynosia septentrionalis eae nas . Phymosia abutiloides (L.) Ham.—Malva . Scaevola plumieri (L.) Vahl* na Spee L.*)— Goodentaceae . Fraxinus americana L.—Oleac . Orontium aquaticum ie . Peltandra virginica (L.) Schott & Engler— Araceae Avicennia germinans (L.) L.— Verbenaceae . Laguncularia racemosa (L.) Gaertner—Combretaceae . Wedelia bahamensis (Britton) Schulz—Compositae 3. Borrichia arborescens (L.) DC.—Compositae Jacquinia keyensis Mez— Theophrastaceae VOLUME 2 . Dalbergia ecastophyllum (L.) Taub. — Leguminosae . Phyllanthus erent L.—Euphorbiaceae . above. Ocotea coriacea (Sw.) Britton (Laurus catesbyana Michx.*)— Lauraceae . below. Galactia Sidsides (Griseb.) Bentham & Hooker—Leguminosae Unidentified . Picrodendron baccatum (L.) Krug & Urban—Euphorbiaceae . above. Conocarpus erectus L.—Combretaceae . below. Amyris elemifera L.—Rutaceae Thalassia testudinum KGnig—Hydrocharitaceae . left. Lysiloma latisiliquum (L.) Bentham — Leguminosae . right. Banara minutiflora (A. Rich.) Sleumer—Flacourtiaceae . Leucothoé racemosa (L. . Acacia tortuosa (L.) Willd. — Leguminosae tee Gray —Ericaceae Alocasia sp. or Xanthosoma sp.— Araceae — : Erythrina herbacea L.— Leguminosae . Canella winterana (L.) Gaertner—Canellaceae . above. Caesalpinia bahamensis Lam.— Leguminosae . below. Passiflora suberosa L.—Passifloraceae Unidentified Clutea eluteria L.*, Clutia cascarilla L.*)— Euphorbiaceae 1983] HOWARD & STAPLES, CATESBY’S PLANTS 545 an Loa) — WwW 74. 75. 76. dale 78. 1D: 80. 81. 81. 82. 83. 84. 84. 85. 86. 87. b 87. 88. 88. 89. 90. 91. 91. 92. 93. 93. 94. 95: _ Urechites lutea (L.) Britton* (Vinca lutea L.*, Echites catesbaei G. Don*)—Apo- cynaceae Silene virginica L.—Caryophyllaceae . Polystachya concreta Uae : an & Sweet— Orchidaceae . Lilium superbum L.—Lilia . Ilex vomitoria L. (Cassine mn Miller*)— Aquifoliaceae _ Lilium catesbaei Walter— Liliaceae _ Echinacea purpurea (L.) Moench—Compositae . Ipomoea batatas (L.) Lam.—Convolvulaceae . Magnolia grandiflora L.—Magnoliaceae . Commelina virginica L.—Commelinaceae . Rhizophora mangle L.— Rhizophoraceae Annona glabra L.*—Annonaceae . Liquidambar styraciflua L.—Hamamelidaceae _ Haematoxylon campechianum L.— Leguminosae Annona glabra L.— Annonaceae _ Epidendrum nocturnum Jacq.— Orchidaceae _ Sarracenia X catesbaei (Ell.) Bell* —Sarraceniaceae . Sarracenia purpurea L. Be uae ces . Symplocarpus foetidus (L.) Nutt. 1PM Cypripedium acaule Aiton (C ee vittatum var. planum Raf.*)—Orchida- . Cypripedium pubescens Willd. (Cypripedium vittatum var. tortile Raf.*)—Orchi- daceae Epidendrum boothianum Lindley— Orchidaceae Mastichodendron Ai gee a oa Lam —Sapotaceae Diospyros virginiana L.—Ebenaceae Catopsis berteroniana aes Mez—Bromeliaceae Spigelia marilandica (L.) L.—Loganiaceae Bourreria ovata Miers— Boraginaceae Magnolia tripetala (L.) L.* (A/agnolia virginiana var. tripetala L.*)— aac above. Swietenia mahagoni (L.) Jacq. (Cedrela mahag. L.*)—Melia below. Phoradendron rubrum (L.) Griseb.* (Viscum rubrum fe a aicece Bignonia capreolata L.— Bignoniaceae Ptelea trifoliata L.—Rutaceae above. Philadelphus inodorus L.* = ee ated below. Smilax lanceolata L.—Smilacaceae Asimina triloba (L.) Dunal* (4nnona triloba L.*)—Annonaceae Annona reticulata L.—Annonaceae above. Manilkara bahamensis Lam & Mecuse (Sloanea emarginata L.*)—Sapo- taceae below. Ipomoea microdactyla Griseb.—Convolvulaceae left. Epidendrum plicatum Lindley— Orchidaceae right. Epidendrum cochleatum L.— Orchidaceae Tillandsia balbisiana (Schultes) Roemer & Schultes—Bromel iaceae Hibiscus tiliaceus L.— Malvaceae above. Cordia sebestena L.— Boraginaceae below. Ipomoea carolina L.*—Convolvulaceae Plumeria rubra L.—Apocynaceae above. Plumeria obtusa L.*—Apocynaceae below. Passiflora cuprea L.*—Passifloraceac Coccoloba diversifolia Jacq. —Polygonaceae above. Hippomane mancinella L.— Euphorbiaceae 546 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 95. below. Dendropemon purpureum (L.) Krug & Urban* (Viscum purpureum L.*)— Loranthaceae 96. Coccoloba uvifera (L.) L.—Polygonaceae 97. Pithecellobium bahamense Northrop— Leguminosae 98. Kalmia latifolia L.— sven 99. Clusia rosea Jacq. —Clusiaceae 100. Catesbaea spinosa L.* (C ae L.*)— Rubiaceae APPENDIX 1. Dodecatheon meadia L.*—Primulacea 2. Hamamelis virginiana L.— Hama =e 3. Cypripedium acaule Aiton es al catesbianum Raf.* 4. Rhus glabra L.—Anacardiacea i — Orchidaceae 5. Hymenocallis caroliniana (L.) Herb.* (Pancratium carolinianum L.*)—Amarylli- daceae 6. Theobroma cacao L.*—Sterculiaceae a 9. Anacardium occidentale L.—Anacardiaceae 11. Lilium canadense L.— Liliaceae 12. Zephyranthes atamasco (L.) Herb.—Amaryllidaceae 13. Stewartia malacodendron L.—Theaceae 15. de acuminata (L.) L.* yeas virginiana var. acuminata L.*)— liace 16. Panes quinquefolius L.—Araliaceae 17. left. Kalmia angustifolia L.—Ericaceae 17. right. Rhododendron maximum L.—Ericaceae 18. Ficus citrifolia Miller— Moraceae 20. Robinia hispida L.— Leguminosae ARNOLD ARBORETUM 22 Divinity AVENUE CAMBRIDGE, MASSACHUSETTS 02138 Magno- 1983] WOOD, INDEXES TO GENERIC FLORA 547 INDEXES TO PAPERS | TO 100 PUBLISHED AS PARTS OF THE GENERIC FLORA OF THE SOUTHEASTERN UNITED STATES CARROLL E. Woop, JR. WITH THE PUBLICATION of the one-hundredth paper in the continuing series dealing with the genera and families of seed-bearing plants in the southeastern United States, it seems appropriate, especially in response to numerous in- quiries, to provide indexes to the papers themselves, their authors, the families and genera treated, and the published illustrations of genera. This information is given here in four parts: I. Chronological List of Papers Published in the Generic Flora II. Authors and Titles of Papers Published in the Generic Flora III. Index to Family Treatments Published in the Generic Flora IV. Index to Generic Treatments and Illustrations Published in the Generic Flora All of the indexed papers have appeared in the Journal of the Arnold Ar- boretum, beginning in July, 1958 (Volume 48, Number 3), and continuing to the present. Thirty-three authors have contributed to the series. All of the papers have been edited by C. E. Wood, Jr., and the last thirteen have been edited by Norton G. Miller, as well. Bernice G. Schubert, Elizabeth B. Schmidt, and Stephen A. Spongberg have contributed greatly in their roles as editors of the Journal. Barbara Nimblett and George K. Rogers have helped in various ways with the present indexes The hundred papers include 130 families, of which the Compositae and Leguminosae have not yet been treated in full. As published in these papers, 475 genera have been recognized; subsequent taxonomic realignments have suggested that a few additional segregate genera should be recognized, however, and these have been indexed as such. In a few instances conservation or a more inclusive generic concept has changed a generic name (e.g., Chamaedaphne conserved over Cassandra, Myrsine, including Rapanea). The principal synonyms of families and genera are included in the respective indexes. In addition, all familial and generic names recognized by John K Small in his Manual of the Southeastern Flora (New York, 1930; republished by University of North Carolina Press, Chapel Hill, 1961) have been included, along with their equivalents, since the concepts of family and genus used in the Generic Flora generally are broader than those of Small. From the beginning of the series, publication of completed familial and generic treatments has not been delayed pending completion of suitable illus- trations, although the ultimate goal has been to have at least one illustration © President and Fellows of Harvard College, | ae of the Arnold Arboretum 64: 547-563. en 1983. 548 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 for each family. One hundred ninety-five genera are illustrated in the published papers. Subsequently, illustrations of 98 others treated in these papers have been prepared, and it is hoped that all of these can be made available eventually. ee -one of the 98 are included (with captions directly on the illustrations) in C. E. Wood, Jr., A Student's Atlas of Flowering Plants: Some Dicotyledons of Eastern North America, or in A. E. Radford et al., Vascular Plant Systematics (both published by Harper & Row, New York, 1974). I. CHRONOLOGICAL LIST OF PAPERS IN THE GENERIC FLORA OF THE SOUTHEASTERN UNITED STATES The papers are numbered 1n the sequence of their publication, although the position of each in the series was not designated in most of the earlier ones. The family (or families) treated in each paper is listed. along with the surname of the author(s) and the volume, inclusive pages. and year of publication in the Journal of the Arnold Arboretum. Full titles of papers and initials of authors are given in the index of authors and papers. Jour. ARNOLD ARB. PAPER NUMBER AND FAMILY VOLUME-PAGES- OR FAMILIES TREATED AUTHOR(S) AR 1. Magnolhiaceae, Annonaceae, Wood 39: 296-346. 1958 Hliciaceae. Schisandraceae,. Canellaceae, Calycanthaceae, Lauraceae 2. Nymphaeaceae, Ceratophyllaceae Wood 40: 94-112. 1959 3. Empetraceac, Diapensiaceae Wood & Channell 161-171. 4. Theophrastaceae. Myrsinaceae. Wood & Channell 268-288. Primulaceae 5. Oleaceae Wilson & Wood 369-384. 6. Plumbaginaceae ’ — 1 & Wood 391-397. 7. Theaceae od 413-419. 8. Sapotaceae, Ebenac “ ae & Channell 41: 1-35. 1960 Styracaceae, ae 9. Araceae. Lemnaceae Wilson 47-72. 10. Sarracentaceae, Droseraceae Wood 152-163 11. Hydrophyllaceae. Sener Wilson 197-212. 12. Myrtaceae Wilson 270-278. 13. Convolvulaceae Wilson 298-317. 14. Ericaceae Wood 42 10-80 1961 15. Cyrillaceae, Clethraceae Thomas 96-106. 16. cree Passifloraceae Brizicky 204-218. 17. Violaceae Brizicky 321-333. I Brizicky 43 -22 1962 19. Simaroubaceae. Burseraceae Brizicky 173-186. 20. Papaveraceae. Fumariaceae Ernst 315-343. 21. Anacardiaceae Brizicky 359-375. 22. Thymelaeaceae Nevling 428-434. 23. Leitneriaceae Channell & Wood 435-438. 24. Capparaceae. Moringaceae Ernst 4: 81-95 1963 25. Loasaceae Ernst & Thompson 138-142. 1983] WOOD, INDEXES TO GENERIC FLORA 549 Jour. ARNOLD ARB, PareR NUMBER AND FAMILY VOLUME-PAGES- OR FAMILIES TREATED AUTHOR(S) é 26. Hamamelidaceae, Platanaceae Ernst 44: 193-210. 1963 27. Tribes of Compositae Solbrig 436-461. 28. ee Aceraceae, Brizicky 462-S01. Hippocastan 29. See eee Ernst 45: 1-35. 1964 Menispermaceae 30. Celastraceae, Hippocrateaceac, Brizicky 206-234. Aquifoliaceae 31. Lythraceae Graham 235-250. 32. Elaeagnac Graham 274-278 33. Rhirphoracas Combretaceae Graham 285-301. 34. Cistacea Brizicky 346-357. 35. eee Brizicky 439-463. 36. Vitaceae Brizicky 46: 48-67. 1965 37. Polygonaceae Graham & Wood 91-12). 38. Valerianaceae, Dipsacaceac Ferguson 218-23). 39. Tiliaceae. Elaeocarpaceae Brizicky 286-307. 40. Caprifoliaceae Ferguson 47: an 1966 41. Sterculiaceae Brizicky 42. Cornaceae Ferguson va 116 43. Nyssaceae Eyde 125 44. Araliacea Graham 126-136 Goodenia Brizicky 293-300 46. Euphorbiaceae Webster 48: 303-430. 1967 47. Compositae-Senecioneae Vuilleumter 50: 104-123. 1969 48. Portulacaceae, Basellaceae Bogle 566-598. 49. Compositae-Mutsieae Vuilleumier 620-625. 50. Ulmaceae Ehas 51 18-40 1970 51. Cannabaceae Miller 185-203 52. Acanthaceae Long 257-309 53. Molluginaceae, Aizoaceae Bogle 431-462 54. Chrysobalanaceae Prance 521-528. 0% Urticaceae Miller 52: 40-68. 1971 56 Elia 159-195 57. Staphyleaceae ses 196-203 58 a laceae iller 267-284 59 ae a 305-318 60 ae Thieret 404-434 61. Saururaceae Wood 479-485. 62. Linaceae Robertson 649-665. 63. Juglandaceae ‘has 53: 26-51 1972 64. Malpighiaceae Robertson 101-112. 65. Geramiaceae Robertson 182-201 66. Phrymaceae Thieret 226-233 67. Saxifragace Spongberg 409-498 68. Zygophyll Porter 531-552 69. Compositae-Lactuceae Vuilleumier 54: 42-93. 1973 70. Krameriaceae Robertson 322-327. 71. Nyctaginaceae Bogle 55: 1-37. 1974 72. Leguminosae-Mimosoideae lias 67-118. 73. Rosaceae Robertson 303-332, 344-401 550 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Jour. ARNOLD ARB. Paper NUMBER AND FAMILY VOLUME-PAGES- OR FAMILIES TREATED AUTHOR(S) EAR 74. Oxalidaceae Robertson 56: 223-239. 1975 75. Mayacaceae Thieret 248-255 76. Bromeliaceae Smith & Wood 375-397 cae sa eied Wood 413-426 78. Podostemacea Graham & Wood 456-465 79. Leguminosae- asalpinions Robertson & Lee 57: 1-53. 976 80. Guttiferae lus Adams & Woo 74-90. 81. Haemodorac Robertson 205-216 82. eae Thieret 58: 25-39 1977 83. Najadace Haynes 161-170 84. a ce Hayne 59: 170-191. 1978 85. Crassulaceae Spongberg 197-248 86. Amaranthaceae Robertson 62: 267-314. 1981 87. Stemonaceae ogers 63: 327-336. 1982 88. Sparganiaceae Thieret 341-356 89. Casuarinaceae Rogers 357-373 90. Bataceae Rogers 375-386 91. Olacaceae Robertson 387-399 92. Viscaceae uyt 401-410 93. vee aceae Miller 411-427 94, Melastomataceae Wurdack & Kral 429-439 95. ee Wood & Weaver 441-487 96. Compositae-Vernonieae Jones 489-507. 97. Burmanniaceae Wood 64: 293-307. 1983 98. Alismataceae Rogers 387-424 99. Xyridaceae Kral 425-433 100. Menyanthaceae Wood 435-449 Il. AUTHORS AND TITLES OF PAPERS PUBLISHED IN THE GENERIC FLORA OF THE SOUTHEASTERN UNITED STATES Under each author or pair of authors, papers are arranged chronologically. Volume numbers, pages, and dates refer to the Journal of the Arnold Arboretum. AbaAMs, P., & C. E. Woo The genera of Goa (Clusiaceae) in the southeastern United States. 57: 74-90. 1976. Boc ie, A. L. The genera re Portulacaceae and Basellaceae in the southeastern United States. 50: 566-598. 1969. The genera a Molluginaceae and Aizoaceae in the southeastern United States. 51: 431-462. 1970. The genera Se cadneces in the southeastern United States. 55: 1-37. 1974. Brizicky, G. K. The genera of Turneraceae and Passifloraceae in the southeastern United States. 42: 204-218. 1961. The genera of Violaceae in the southeastern United States. 42: 321-333. 1961. The genera of Rutaceae in the southeastern United States. 43: 1-22. 1962 1983] WOOD, INDEXES TO GENERIC FLORA beg The genera of Simaroubaceae and Burseraceae in the southeastern United States. 43: 173-186. 1962. The genera of Anacardiaceae in the southeastern United States. 43: 359-375. 1962. The genera of Sapindales in the southeastern United States. 44: 462-501. 1963. The genera of Celastrales 1 in the southeastern United States. 45: 206-234, 1964. The genera of Tiliaceae and Elaeocarpaceae in the southeastern United States. 46: 286-307. | The genera of Sterculiaceae in the southeastern United States. 47: 60-74. 1966. The Goodeniaceae in the southeastern United States. 47: 293-300. 1966. CHANNELL, R. B., & C. E. Woop, Jr. (see also Woop & CHANNELL) The genera of the Primulales of the southeastern United States. 40: 268-288. 1959. The genera of Plumbaginaceae of the southeastern United States. 7 eg 397. 1959, The Leitneriaceae in the southeastern United States. 43: 435-438. ELIAS, The genera of Ulmaceae in the southeastern United States. 51: 18-40. 1970. The genera of Fagaceae in the southeastern United States. 52: 159-195. 1971. The genera of Myricaceae in the southeastern United States. 52: 305-318. is The genera of Juglandaceae in the southeastern United States. 53: 26-51 The genera of Mimosoideae (Leguminosae) in the southeastern United ee 55: 67- 974, Ernst, W. R. The genera * eee and Fumariaceae in the southeastern United States. 43: 315-343. The genera ea pacers and Moringaceae in the southeastern United States. 44: 81-95. 1963. The genera of ae and Platanaceae in the southeastern United States. 4: 193-210. 3. The genera of a aes a) ali and Menispermaceae in the south- eastern United States. 45: 1-35. | & H THOMPSON The Loasaceae in the southeastern United States. 44: 138-142. 1963. Eype, R. H. The Nyssaceae in the southeastern United States. 47: 117-125. 1966. FERGUSON, I. The genera of Valerianaceae and Dipsacaceae in the southeastern United States. 46: 218-231. 1965. The genera of Caprifoliaceae in the southeastern United States. 47: 33-59. 1966. The Cornaceae in the southeastern United States. 47: 106-116. 1966. GRAHAM, S. A. The genera of Lythraceae in the southeastern United States. 45: 235-250. 1964. The Elaeagnaceae in the southeastern United States. 45: 274-278. The genera of oo and Combretaceae in the southeastern United States. 45: 285-301. 1964. The genera of Artie in the southeastern United States. 47: 126-136. 1966. Cc. The genera of Senter in the southeastern United States. 46: 91-121. 1965. The Podostemaceae in the southeastern United States. 56: 456-465. 1975 Haynes, R. R. The Najadaceae in the southeastern United States. 58: 161-170. 1977. The Potamogetonaceae in the southeastern United States. 59: 170-191. 1978. o52 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Jones, S. B., JR. The genera of the Vernonieae (Compositae) in the southeastern United States. 63: 489-507. nee Krat, R. (see also WURDACK & KRAL) The — in the a a United States. 64: 425-433, 1983. Kur, The Viscaceae in the enn United States. 63: 401-410. 1982. Lee, Y. T. (see ROBERTSON & L Lona, R. W. The genera of Acanthaceae in the southeastern United States. 51: 257-309. 1970. Mitier, N. G. The genera of the Cannabaceae in the southeastern United States. 51: 185-203. 1970. The genera of the Urticaceae in the southeastern United plates: 52: 40-68. 1971. The Polygalaceae in the southeastern United States. 52: 267- I71. The Caricaceae in the southeastern United States. 63: 41 1-427, 1982. Neviuina, L. L., Jr. The ai in the southeastern United States. 43: 428-434. 1962. Porter, D. The genera oF Zygophyllaceae in the southeastern United States. 53: 531-552. 1972. PRANnce, G. T. he genera of C hrysobalanaceace 1n the southeastern United States. 51: 521-528. 1970. ROBERTSON, K. R. The Linaceae in the southeastern United States. 52: 649-665. 1971. The Malpighiaceae in the southeastern United States. 53: 101-112. 197 The genera of Geraniaceae in the southeastern United States. 53: 182- 201, 1972. The Krameriaceae in the southeastern United States. 54: 322-32 The genera : uaa in the southeastern United States. 55: 303 ne . 344-401, 611-662. | The eee in the southeastern United States. 56: 223-239. 1975. The genera of Haemodoraceac in the southeastern United States. 57: 205-216. 1976. The genera of Amaranthaceae in the southeastern United States. 62: 267-314. 1981. The genera of Olacaceae in the southeastern United States. 63: 387-399, 1982. & Y. T. Let The genera of Caesalpinioideae (Leguminosae) in the southeastern United States. 57: 53. 976 ROGERS The Stemonaceae in the southeastern United States. 63: 327-336. 1982. The Casuarinaceae in the southeastern United States. 63: 357-373. 1982. The Bataceae in the southeastern United States. 63: 375-386. 1982. The genera of Alismataceae in the southeastern United States. 64: 387-424. 1983. SmitH, L. B.. & C. E. Woop, The genera of Bromeliaceae in cine southeastern United States. 56: 375-397. 1975. SOLBRIG, O. The tribes of Compositae in the southeastern United States. 44: 436-461. 1963. SPONGBERG, S. A, The Staphyleaceae in the southeastern United States. 52: 196-203. 1971. The genera of Saxifragaceae in the southeastern United States. 53: 409-498. 1972. The genera of Crassulaceae in the southeastern United States. 59: 197-248. 1978. TrHiperet, J. W. The genera of Orobanchaceae in the southeastern United States. 52: 404-434. 1971. The Phrymaceae in the southeastern United States. 53: 226-233. 1972. The Sy eseenis in the southeastern United States. 56: 248-255. 1975. The Martyniaceae in the southeastern United States. 58: 25-39. The Res renee in the southeastern United States. 63: 341-356. 1982, 1983] WOOD, INDEXES TO GENERIC FLORA 553 Tuomas, J. The ee ‘of the Cyrillaceae and Clethraceae of the southeastern United States. 42: 96-106. 1961 THompson, H. J. (see Ernst & THOMPSON) VUILLEUMIER, B. S. The genera of Senecioneae in the southeastern United States. 50: 104-123. ee The tribe Mutisieae (Compositae) in the southeastern United States. 50: 620-6 1969. The genera of Lactuceae (Compositae) in the southeastern United States. 54: 42-93. 33 Weaver. R. E., JR. (see Woop & WEAVER) Webster, G. L. The genera of Euphorbiaceae in the southeastern United States. 48: 303-430. 1967. WILSON, K. A. The genera of the Arales in the southeastern United States. 41: 47-72. 1960. The genera of Hydrophyllaceae and Polemoniaceae in the oa United States. 41: 197-212. 1960. The genera of Myrtaceae in the southeastern United States. 41: 270-278. 1960. The genera of Convolvulaceae in the southeastern United States. 41: 298-317. 1960. E. Woop, Jr. The genera of Oleaceae in the southeastern United States. 40: 369-384, 1959. Woop, C. E., Jr. (see also ADAMS & — CHANNELL & Woop, GRAHAM & Woop, SmitH & Woop, and WILSON & Woop) The genera of the woody Ranales in the southeastern United States. 39: 296-346. 1958. The genera of the Nymphaeaceae and Ceratophyllaceae in the southeastern United States. 40: 94-112. 1959. The genera of Theaceae of the southeastern United States. 40: 413-419. 1959. The genera of Sarraceniaceae and Droseraceae in the southeastern United States. 41: 152-163. 1960. The genera of Ericaceae in the southeastern United States. 42: 10-80. 1961. The Saururaceae in the southeastern United States. 52: 479-485. 1971. The Balsaminaceae in the southeastern United States. 56: 413-426. 1975. The genera of Burmanniaceae in the southeastern United States. 64: 293-307. 1983. The genera of peusenacse in the southeastern United States. 64: 435-449. 1983. . CHANNELL The Empetraceae anid Diapensiaceae of the southeastern United States. 40: 161-171 9 The genera of the Ebenales in the southeastern United States. 41: 1-35. 1960. & WEAVER, JR. The genera of Gentianaceae in the southeastern United States. 63: 441-487. 1982. WuRDACK, J. l RAL The genera of Melastomataceae i in the southeastern United States. 63: 429-439. 1982. Il. INDEX TO FAMILIES PUBLISHED IN THE GENERIC FLORA SERIES Volume, number of the first page of the article. and year of publication are given for each family or part of a family listed. All refer to the Journal of the Arnold Arboretum. 554 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 oe: 51: 257. 1970 Cardueae 44: 448. 1963 Acerac 44: 481. 1963 Cichorieae = Lactuceae Aesculaceae = - aera Eupatori 44: 452. 1963 Aizoacea 443. 1970 Heliantheae 44: 453. 1963 eee ae 7 387. 1983 Inuleae 44: 456. 1963 Allionaceae = Nyctaginaceae Lactuceae 44: 450. 1963 Altingiaceae = ‘naa ee 54: 42. 1973 Amaranthaceae 63: . 1981 Mutisieae 44: 457. 1963 Amygdalaceae = Rosaceae 50: 620. 1969 Anacardiaceae 43: 359. 1962 Senecioneae 44: 459, 1963 Annonaceae 39: 309. 1958 50: 104. 1969 Aquifoliaceae 45: 227. 1964 Vernonieae 44: 460. 1963 Araceae 41: 47. 1960 63: 489. 1982 Araliaceae 47: 126. 1966 unio 41: 298. 1960 Ardisiaceae = Myrsinaceae Cornaceae 47: 106. 1966 Armeriaceae = Plumbaginaceae eenres 59: 197. 1978 Asteraceae: see Compositae eres = Convolvulaceae Balsaminaceae 56: 413. 1975 Cyrillaceae 2: 96. 1961 Basellaceae 50: 590. 1969 Daphnaceae = Thymelaeaceae Bat 63: 375. 1982 Diapensiaceae 40: 164. 1959 aa pied 45: 1. 1964 Dichondraceae = Convolvulaceae Bromeliace 56: 375. 1975 Dionaeaceae = Droseraceae Gaines sensu Small = Dipsacaceae 46: 226. 1965 Goodentiaceae Dodonaeaceae = Sapindaceae Burmanniaceae 64: 293. 1983 Droseracea 41: 156. 1960 Burseraceae 43: 183. 1962 Ebenaceae 41: 17. 1960 Byttneriaceae = Sterculiaceae Elaeagnacea 45: 274. 1964 Cabombaceae = Nymphaeaceae Elaeocarpaceae 46: 304. 1965 ie ae ae = Leguminosae Empetraceae : 162. 1959 Caesalpinioidea Epilobiaceae = Onagraceae Ge 57: 1. 1976 Ericaceae 42: 10. 1961 Calycanthaceae 39: 322. 1958 Euphorbiaceae 48: 303. 1967 Canellaceae 39: 320. 1958 Fabaceae: see Leguminosae Cannabaceae 51: 185. 1970 Fagaceae 52: 159. 1971 Capparaceae 44: 81. 1963 Frangulaceae = Rhamnaceae Caprifoliaceae 47: 33. 1966 Fumariaceae 43: 333. 1962 Caricaceae 63: 411. 1982 Galacaceae = Diapensiaceae Cassiaceae = Leguminosae Gentianaceae 63: 441. 1982 Cassythaceae = Lauraceae Gerania 53: 182. 1972 Casuarinaceae 63: 357. 1982 Goodeniace 93. 1966 Celastraceae 45: 206. 1964 Grossulariaceae = Sa aa Ceratophyllaceae 40: 109. 1959 Guttiferae 74. 1976 Chrysobalanaceae 51: 521. 1970 Haemodoraceae 205. 1976 Cichoriaceae = Compositae-Lactuceae Hamamelidaceae 44: 193. 1963 eee 45: 346. 1964 Hederaceae = Araliaceae Clethrac 42: 102. 1961 Hippocastanaceae 495. 1963 Cie see Guttiferae Hippocrateaceae 45: 223. 1964 Combretaceae 45: 293. 1964 Hydrangeaceae = Saxifragaceae Compositae Hydroleaceae = Hydrophyllaceae Tribes of 44: 436. 1963 Hydrophyllaceae 41: 197. 1960 Anthemideae 44: 444. 1963 Hypericaceae = Guttiferae Arctoteae 44: 445. 1963 Illiciaceae 39: 315. 1958 Asiereae 44: 446. 1963 Iteaceae = Saxifragaceae Calenduleae 44: 448. 1963 Juglandaceae 53: 26. 1972 1983] WOOD, INDEXES TO GENERIC FLORA 520 oo 54: 322. 1973 Podostemaceae 56: 456. 1975 Lactuc Polemoniaceae 41: 205. 1960 cae sitae) 54: 42. 1973 Polygalaceae 52: 267. 1971 Lardizabalaceae 45: 20. 1964 Polygonaceae 46: 91. 1965 Lauraceae 39: 326. 1958 Portulacaceae 50: 566. 1969 Leguminosae Potamogetonaceae 59: 170. 1978 Caesalpinioideae 57 1. 1976 Primulaceae 40: 273. 1959 imosoideae 55: 67. 1974 Pyrolaceae = Ericaceae Leitneriaceae 43: 435. 1962 Rhamnaceae 45: 439. 1964 nacea 41: 63. 1960 Rhizophoraceae 45: 285. 1964 Linaceae 52: 649. 1971 Rosaceae 55: 303 Loasaceae 44: 138. 1963 34 Loranthaceae: see Viscaceae 1974 Lythrace 235. 1964 Roxburghiaceae = Stemonaceae Magnoliaceae 39: 299. 1958 Ruppiaceae = Potamogetonaceae Malaceae = Rosaceae Rutaceae 43: 1. 1962 Malpighiaceae 53: 101. 1972 Sambucaceae = Caprifoliaceae Martyniaceae 58: 25. 1977 Sapindaceae 44: 463. 1963 yacaceae 56: 248. 1975 Sapotaceae 41: 2. 1960 Melastomataceae 63: 429. 1982 Sarraceniaceae 41: 152. 1960 Menispermacea 45: 23. 1964 Saururaceae 52: 479. 1971 Menyanthace 435. 1983 Saxifragaceae 53: 409. 1972 Mimosaceae = Leguminosae Schisandraceae 39: 318. 1958 imosoideae Sedaceae = Crassulaceae (Leguminosae) So 67. 1 Senecioneae Molluginaceae 431. 1970 (Compositae) 50: 104. 1969 Monotropaceae = Ericaceae Simaroubaceae 43: 173. 1962 Morinaceae = Dipsacaceae Sparganiac 63: 341. 1982 Moringaceae 44: 93. 1963 Spondiaceae = Anacardiaceae Mutisieae Staphyleaceae 96. 1971 (Compositae) 50: 620. 1969 temonaceae 63: 327. 1982 Myricaceae 922.305.1971 Sterculiaceae 47: 60. 1966 Myrsinaceae 40: 271. 1959 Styracaceae 22. 1960 Myrtaceae 41: 270. 1960 et aes = "SuTiatoubacea® Najadac 61. LO77 Symploc 41: 31. 1960 Nelumbonaceae = Nymphacaceae = Aone = Combretaceae Nyctaginace : 1. 1974 Tetragoniaceae = Aizoaceae Nymphaeaceae 40: 94. 1959 Theaceae 40: 413. 1959 Nyssaceae 47: 117. 1966 Theophrastaceae 40: 269. 1959 Olacaceae 63: 387. 1982 Thymelaeaceae 43: 428. 1962 leaceae 40: 370. 1959 Tiliaceae 46: 286. 1965 Orobanchaceae 52: 404. 1971 Turneraceae 42: 204. 1961 Oxalidaceae 56: 223. 1975 Ulmaceae 51: 18. 1970 Papaveraceae 43: 315. 1962 Urticacea 2: 40. 1971 Papayaceae = Caricaceae Vacciniaceae = Ericaceae Parnassiaceae = Saxifragaceae Valerianaceae 218. 1965 Passifloraceae 42: 209. 1961 Vernonie Penthoraceae = Saxifragaceae (Compositae) 63: 489. 1982 Phrymaceae 53: 226. 1972 Violaceae 42: 321. 1961 Pisoniaceae = Nyctaginaceae Viscaceae 63: 401. 1982 Platanaceae 44: 206. 1963 Vitaceae 46: 48. 1965 Plumbaginaceae 40: 391. 1959 Xyridac 64: 425. 1983 Podophyllaceae = Berberidaceae Zygophyllaceae 53; 3b 172 536 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 IV. INDEX TO GENERIC TREATMENTS AND ILLUSTRATIONS PUBLISHED IN THE GENERIC FLORA SERIES Asin the preceding list of families, the references to genera include the volume number, the first page of each treatment. and the year in which each was published in the Journal of the Arnold Arboretum. Ulustrated genera are marked with an asterisk. Pertinent synonyms are in italics, with a reference to the genus under which that name is treated. References in brackets pertain to genera included in a key but not admitted to the flora. *Acacia 55: 99. 1974 Anamomuis = Myrcianthes Acaciella = Acacia Ananas 56: 395. 1975 Acalypha 48: 370. 1967 Andrachne 48: 327. 1967 Acer 44: 482. 1963 *Androsace 40: 276. 1959 Acetosad = Rumex Annona 39: 313. 1958 Acetosella = Rumex *Anredera 50: 594. 1969 tchras = Manilkara Antenoron 46: 108. 1965 een 62: 294. 1981 a = Gentianopsis Acnida = Amaranthus Antigonon 46: 118. 1965 Acomastylis = Geum ae = Dyschoriste cOrUus 41: 50. 1960 Aphanes = Alchemilla Acuan = Desmanthus Aphragmia = Ruellia [Adenanthera 55: 72. 1974] Aphyllon = Orobanche Adenoropium = Jatropha Apogon = Krigia Adicea = Pilea *Apteria 64: 304. 1983 Adipera = Cassia Aralia 47; 130. 1966 Adlumia 43: 338. 1962 Ardisia 40: 272. 1959 Aesculus 44: 496. 1963 Argemon 43: 328. 1962 Agaloma = Euphorbia Argentacer = Ace arista 42: 40. 1961 Arge ntina = - lk Agrimonia 55: 390. 1974 Argyre 41: 317. 1960 Ailanthus 43: 179. 1962 oe ane 48: 364. 1967 kebia 45: 21. 1964 *Arisaema 41: 58. 1960 *Albizia 55: 109. 1976 Arnica 50: 106. 1969 Alchemilla 55: 393. 1974 ace ae ia didencite= = Cleome Aronia = S Aleurites 48: 342. 1967 Arrhosty re = Ruellia *Alisma 64: 398. 1983 Arsenococcus = Lyonia Allionia: see Mirabilis *Aruncus 55: 325. 1974 *Alternanthera 62: 302. 1981 Ascyrum = Hypericum Alvaradoa 43: 182. 1962 Asemeia = Polygala * Amaranthus 62: 282. 1981 Asimina 39: 311. 1958 Amarella = Gentianella Astilbe 53: 422. 1972 Amarolea = Osmanthus Asystasia 51: 298. 1970 *Amelanchier 55: 633. 1974 Ayenia 47: 68. 1966 oe 45: 240. 1964 Azalea: see Rhododendron Ampelopsis 46: 56. 1965 Bartonia 63: 477. 1982 Ampe iia ee = Pieris *Batis 63: 375. 1982 Amyedalus = Prunus Batodendron = Vaccinium Amyris 43: 11. 1962 Befaria 42: 18. 1961 Anac oo = Sedum Beloperone = Justicia Ana 40): 284. 1959 Benthamidia = Cornus 1983] WOOD, INDEXES TO GENERIC FLORA DoT Benzoin = Lindera *Casuarina 63: 357. 1982 Berberis 45: 9. 1964 eee = Linum Berchemia 45: 457. 1964 *Cato 392. 1975 Bicuculla = Dicentra Canisehyiliie 45: 18. 1964 Bilderdykia = Polygonum [Cayratia 46: 49. 1965] Biltia = Rhododendron Ceanothus 45: 451. 1964 Bivonea = Cnidoscolus Celastrus 45: 215. 1964 Blechum 51: 284. 1970 Celosia 62: 280. 1981 Blutaparon 62: 312. 1981 Iti 51: 32. 1970 Sea 52: 59. 1971 Centaurium 63: 453. 1982 erhavi 55: 23. 1974 Centrostachys = Achyranthes Balboa = = Oxalis Centunculus = Anagallis Bonam 1: 306. 1960 Cerasus = Prunus ae aryum = Corn *Ceratiola 40: 163. 1959 Boussingaultia: see Anredera Ceratophyllum 40: 109. 1959 Boykinia 53: 440. 1972 *Cerci 57: 48. 1976 Brachyramphus = Launaea Cerothamnus = Myrica enia 40: 103. 1959 Chaetoptelea = Ulmus Brayulinea = Guilleminea Chamaechrista = Cassia Breweria: see Bonamia *Chamaedaphne 42: 54. 1961 Bruneria = Wolfha Chamaepericlymenum = Cornus *Brunnichia 119. 1965 *Chamae 48: 420. 1967 Bryophylum = Kalanchoé *Chaptalia 50: 620. 1969 cida 45: 297. 1964 *Chelidonium 43: 325. 1962 *Bumelia 41: 7. 1960 Chimaphi 61. 1961 Burmannia 64: 300. 1983 Chiogenes = Gaultheria ursera 43: 185. 1962 Chionanthus 40: 380. 1959 Buxella = Gaylussacia *Chrysobalanus 51: 523. 1970 *Byrsonima 53: 109. 1972 Chrysophyllum 41: 10. 1960 Cabomba 40: 102. 1959 *Chrysosplenium 53: 453. 1972 Cacalia 50: 115. 1969 Cicca = Phyllanthus *Caesalpinia 57: 13. 1976 *Cichorium 4: 47. 1973 Cailleia = oo Cinnamomum 39: 334. 1958 Calonyction = Ipom Cissampel 45: 33. 1964 Calophanes = Dy shoree Cissus 46: 54. 1965 *Calycanthus 39: 323. 1958 Citrus 43: 17. 1962 *Calycocarpum 45: 28. 1964 Claytonia 50: 584. 1969 Calyptranthes 41: 273. 1960 Cleome 44: 87. 1963 Calystegia 308. 1960 *Clethra 42: 103. 1961 Camphora = Cinnamomum Cliftonia 42: 101. 1961 Canella 39: 321. 1958 *Clusia 57: 78. 1976 See 51: 188. 1970 *Cnidoscolus 48: 349. 1967 Caperoni 48: 363. 1967 *Coccoloba 46: 114. 1965 G ee = Corydalis Coccolobis = Coccoloba Capparis 44: 85. 1963 *Cocculus 5: 30. 1964 Caraxeron: see Blutaparon Colocasia 41: 56. 1960 Cc ales um 44: 468. 1963 Colubrin 45: 455. 1964 *Caric 63: 415. 1982 *Comptonia 52: 315. 1971 ‘ary 53) 321972 Conocarpus 45: 298. 1964 Cassandra = Chamaedaphne *Conopholis 52: 420. 1971 *Cassia 57: 35. 1976 Convolvulus 41: 308. 1960 Cassytha 39; 344, 1958 46: 295. 1965 Castalia = Nymphaea Corculum = Antigonon *Castanca §2: 173. 197] *Cornus 47: 109. 1966 558 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 *Corydalis 43: 339. 1962 eo eee = Ruelha Cotinus 43: 366. 1962 Dirca 1. 1962 laa see Pyracantha Ditaxis = Argythamnia *Crassul 59; 231. 1978 cal enced = - assia ee 55: 626. 1974 Dodecath 40: 277. 1959 “rep 54: 91. 1973 44: 475. 1963 Cristatella = Drejerella = Justicia Crantienn = = Helianthemum rosera 1: 160. 1960 Crookec = Hypericum Drymocatlis = Potentilla * 63: 331. 1982 rypetes 8: 329. 1967 ak oe aia 45: 220. 1964 *Duches 55: 371. 1974 ; 48: 353. 1967 Dupratzia = Eustoma Cryphiacanthus = Ruellia *Dyschoriste 51: 292. 1970 Cubeltum = Hybanthus Pelee 64: 405. 1983 “upania 74. 1963 Elaea 5: 276. 1964 Cuphea 45: 247. 1964 a vere = Euphorbia Curcas = Jatropha Elaphrium = Bursera Cuscu 41: 301. 1960 Elephant opus 63: 503. 1982 a ae - ae *Elliottia 42: 20. 1961 Cymbia = Krig Ellisia 41: 201. 1960 Cynoxylon = ee Elytraria 51: 278. 1970 Cynthia = Krigia Emoblica = Phyllanthus Cypselea 51: 457. 1970 a — = Cassia *Cyrilla 42: 98. 1961 50: 122. 1969 alibarda 55: 361. 1974 ane = Cocculus Dasiphora = Potentilla *Epifagus 52: 414. 1971 Dasystephana = Gentiana Epigaea 42: 58. 1961 Decachaena = Gaylussacia Erechtites 50: 120. 1969 Decodon 45: 238. 1964 Eniogonum 46: 96. 1965 *Decumaria 53: 495. 1972 Erodium 53: 197. 1972 Deeringothamnus = Asimina Erythraea = Centaurium elonix 57: 20. 1976 Eschscholzia 3: 327. 1962 Delopyrum = Polygonella Eubotrys = Leucothoé Dendropogon = Tillandsia Eugenia 41: 274. 1960 Dentoceras = Polygonella *Euonymus 45: 210. 1964 Desmanthus 55: 92. 1974 *Euphorbia 48: 395. 1967 Desmothamnus = Lyonia *Eustoma 63: 480. 1982 226. 1978 Evolvul 41: 305. 1960 Dianthera = Justicia Exogonium = Ipomoea uit = Dicliptera Exothe 479, 1963 Diaphoranthema = Tillandsia *Fagus 52: 166. 1971 icentra 43: 336. 1962 Filipendula 55: 349. 1974 Dichondra 41: 304. 1960 Firmiana 47: 66. 1966 Dichrophylhun = eee restiera 40: 379, 1959 panna a ys 5: 72. 1974] *Fothergilla 44: 201. 1963 Diclipt a 300. 1970 *Fragaria 55; 362. 1974 Didiplis — - ” Peplis Frangula = Rhamnus *Diervilla 47: 51. 1966 *Franklinia 40: 415. 1959 *Dionaea 41: 158. 1960 Frasera 63: 472. 1982 Diospyros 41: 18. 1960 Fraxinus 40: 371. 1959 Dipholis 41 6. 1960 Froelichia 62: 300. 1981 Diphylleia 45: 16. 1964 umaria 342. 1962 Dipsacus 46: 228. 1965 Galarhoeus = Euphorbia 1983] WOOD, INDEXES TO GENERIC FLORA 559 Pec 40: 170. 1959 Hudsoni 45: 354. 1964 Gal 51: 449. 1970 ae Vaccinium Ga cee = Polygala Humulus 51: 196. 1970 Gatesia = Dicliptera Hybanthus 42: 323. 1961 *Gaultheria 42: 42. 1961 Hydatica = Saxifraga Gaylussacia 42: 77. 1961 *Hydrang 53: 487. 1972 tlana 63: 457. 1982 Hydrole 41: 205. 1960 *Gentianella 63: 464. 1982 Te 41: 199. 1960 Gentianopsis 63: 469. 1982 Hygrophila 51: 296. 1970 Geobalanus = Licania i ae = Sedum *Geranium §3: 189. 1972 ypelate 4: 478. 1963 Gerardia = Stenandrium Hypericur S72. °825,1970 Geum 55: 382. 1974 Hypochoeri 4: 62. 1973 Gilia: see Ipomopsis Hypopitys = Monotropa Gillenia = Porteranthus Icacorea = Ardisia *Gisekia 51: 435. 1970 ; 45: 228. 1964 Glabraria = Litsea iene 39: 316. 1958 *Gleditsia 57: 26. 1976 *Impatie 56: 416. 1975 *Glinus 51: 441. 1970 re = Oxalis Glycosmis 43: 12. 1962 Ipomoea 41: 311. 1960 ag 62: 307. 1981 Ipomopsis 41: 211. 1960 don 40: 414. 1959 *Iresine 62: 309. 1981 Gossypianthus = = Guilleminea *Ite 53: 476. 1972 ania 45: 462. 1964 Jacquemontia 41: 307. 1960 Grossularia = Ribes *Jacquinia 40: 207. 1959 uaiacum : 547. 1972 Jasminum 40: 382. 1959 uapira = Pisonia Jatropha 48: 340. 1967 Guilandina = Caesalpinia Jeffersonia 45: 17. 1964 Guilleminea : 297. 1981 *Juglans 53: 45. 1972 *Guzmania 56: 390. 1975 *Justicia 51: 302. 1970 *Gyminda 45: 221. 1964 Kalanchoé 59: 237. 1978 Gymnanthes 48: 387. 1967 Kallstroemia §3: 539. 1972 *Gymnocladus 57: 21. 1976 Kalmia 42: 23. 1961 Gymnostoma = Casuarina Kalmiella = Kalmia Gynandropsis = Cleome Kitchingia = Kalanchoé ynura 106. 1969] Krameria : 322. 1973 Gyrotheca = Lachnanthes *Krigia Bde. S12 1973 *Halesia 41: 26. 1960 Krug 45: 462. 1964 Halimum:. see Helianthemum Lachemilla = Alchemilla amamelis 44: 197. 1963 *Lachnanthes 57: 210. 1976 a = Cacalia Lactuca 54: 79. 1973 era 47: 128. 1966 Lagunculari 5: 300. 1964 ee = Licania Lapithea = Sabatia Helanthium = Echinodorus *Laportea 525 52.1971 Helianthemum 45: 351. 1964 apsana 54: 88. 1973 Hemicyclia = Drype Lasiococcus = Gaylussacia a Launaea 54: 71. 1973 Heuchera 53: 435. 1972 Laurocerasus = Prunus Hicoria = Carya Lechea 45: 356. 1964 Hi ium 54: 73. 1973 Leiophyllum 42: 26. 1961 Hippocratea 45: 225. 1964 Leiphainos = = Voyria Hippomane 48: 393. 1967 bea 43: 435. 1962 *Hottonia 40: 279. 1959 41: 67. 1960 560 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Leontodon 54: 65. 1973 oo see Manilkara Lepadena = Euphorbia rabilis 55: 10. 1974 Leptamnium = Epifagus Miianieen = Licaria Leptoglottis = Schrankia *Mitella 53: 425. 1972 *L_epuropetalon 53: 458. 1972 Mohrodendron = Halesia *Leucaena 55: 78. 1974 *Mollugo 51: 437. 1970 Leucothoé 42: 38. 1961 Moluchia = Melochia *Licania 51: 526. 1970 Momisia = Celtis Licaria 39: 338. 1958 onotrope 4 66. 1961 igustru 40: 381. 1959 Monotropsis 42: 64. 1961 a = Nymphoides Moquilea = si Limoniu 40: 395. 1959 ace - ca indera 39: 342. 1958 Moring, 93. 1963 *Linum §2: 653. 1971 Mee = = Schrankia *1 iquidambar 44: 203. 1963 Mosiera = Myrtus ba 39: 306. 1958 ee = Lactuca Lit 39: 343. 1958 Munt 46: 305. 1965 ahadiie = Rhus Wines = Arisaema *] onicera 47: 54. 1966 Muscadinia = Vitis *Lophiola 4213: 1976 Myginda = Crossopetalum Lophotocarpus = Sagitara Myrcianthes 275. 1960 Lucuma: see Pout *Myri 52: 308. 1971 Lygodesmia 54: 57. 1973 Myrsine 40): 273. 1959 onia 42: 47. 1961 Myrtu 1: 276. 1960 Lysiloma 55: 107. 1974 Nabalus = Prenanthes Lysimachia 40): 282. 1959 *Naj 58: 161. 1977 = eau 45: 242. 1964 Nama 41: 204. 1960 Magnoli : 302. 1958 Nandina 45: 8. 1964 Malachodendron = = Stewartia Nectandra 39: 336. 1958 alus = Pyr Negundo = Mangifera 43: 362. 1962 Nelumbiunm = Nelumbo Manihot 48: 345. 1967 *Nelumbo 40: 105. 1959 anilkara 41: 13. 1960 Nemophila 41: 200. 1960 Maritlaunidium = Nama Neocleome = Cleome Martynia: see Proboscidea Neopierits = Lyonia Mastichodendron 41: 5. 1960 *Neptunia 55: 95. 1974 *Mavyaca 56: 248. 1975 *Neviusi 55: 345. 1974 Maytenus 45: 218. 1964 Nintooa = Lonicera Melaleuca 41: 273. 1960 Juphar 40: 100. 1959 Melicoccus 44: 472. 1963 Nyctelea = Ellisia Melochia 47: 70. 1966 Nymphaea 40: 97. 1959 Menispermum 45: 32. 1964 *Nymphoides 64: 443. 1983 Mentzelia 44: 140. 1963 Nymphozanthus = Nuphar Menyanthes 64: 440. 1983 *Nyssa 47: 119. 1966 Menziesia 42: 28. 1961 ee _ 63: 475. 1982 Mercurialis 48: 366. 1967 *Oke 5: 20. 1974 Merremia 41: 310. 1960 Deecii see Merremia Mesadenia = Cacalia Opulaster = Physocarpus Metopium 43: 368. 1962 Oroban 52: 424. 1971 Micranthes = Saxifraga *Orontium 41: 54. 1960 Microptelea = Ulmus Osmanthus 40: 377. 1959 Micropyxis = Anagallis *Oxalis 56: 229. 1975 Mimosa §5: 82. 1974 Oxyvbaphus = Mirabilis 1983] WOOD, INDEXES TO GENERIC FLORA 561 Oxycoccus = Vaccinium Polygonum 46: 105. 1965 *Oxydendrum 42: 56. 1961 *Poncirus 43: 14. 1962 Padus = Prunus sae 55: 329. 1974 Panax 47: 132. 1966 *Portulaca 50: 571. 1969 Papaver 43: 330. 1962 i ee = Trianthema Parageum = Geum Potamogeton 59: 174. 1978 *Parietaria 52: 64. 1971 *Potentill 55: 373. 1974 Parkinsonia 57: 32. 1976 Poterium = Sanguisorba Parnassia 53: 461. 1972 Pouteri 41: 11. 1960 Parsonsia = Cuphe Prenanthes 54: 76. 1973 Parthenocissus 46: 57. 1965 Prinos = Mex *Passiflora 42: 211. 1961 *Proboscidea 58: 33. 1977 Pedilanthu 48: 427. 1967 *Prunus 55: 654. 1974 Peiranisia = Cassia Pseudo-elephantopus 63: 505. 1982 *Peltandra 41: 55. 1960 idium 41: 277. 1960 *Penthorum 530419. 1072 Psilorhegma = Cassia Peplis 45: 241. 1964 Ptelea 43: 9. 1962 ersea 39: 3305 1958 Pyracantha 55: 624. 1974 Persicaria = Polygonum Pyrola 42: 62. 1961 Phacelia 41: 202. 1960 Pyrrhopappus 54: 55. 1973 Pharbitis = Ipomoea *Pyrus 55: 640. 1974 Phenianthus = Lonicera Pyxidanthera 40: 166. 1959 *Philadelphus 480. 1972 Quamoclit = Ipomoea Philoxerus: see Blutaparon Quercus 179. 1971 Phlox 41: 207. 1960 Ramium = Boehmeria *Phoradendron 63: 407. 1982 Rapanea = Myt rsine y 53: 227. 1972 eynosia 5: 461. 1964 Sennen 48: 332. 1967 Reynoutria = Polygonum *Physocar 55: 316. 1974 Rhacoma = Crossopetalum areas = Tillandsia Rhamnus : 445. 1964 Picramnia 43: 181. 1962 *Rhexia 63: 432. 1982 cris 54: 67. 1973 *Rhizophora 45: 286. 1964 *Pieris 42: 51. 1961 Rhodiola = Sedum *Pilea S22 99, 1971 hododendron 42: 30. 1961 Pilostaxis = Polygala Rhodomyrtus 41: 278. 1960 Piriqueta 42: 206. 1961 Rhus 43: 370. 1962 *Pisonia i on lee Rhytiglossa = Justicia *Pistia 41: 61. 1960 *Ribes 53: 466. 1972 Pithecellobium a5; 115.1974 *Ricinus 48: 379. 1967 Pity otha = = Asimina Riedlea = ee Plan 31. 1970 Rivea: see Tur Plat 44: 206. 1963 Robertiella = ane eae = Polygonum *Rosa > 611. 1974 mbago 40: 393. 1959 otala 45: 240. 1964 Se laa 45: 13. 1964 Rubacer = Rubus *Podostemum 56: 461. 1975 *Rubus 55: 352. 1974 Dae = Caesalpinia *Ruellia 51: 285. 1970 Poinsettia = Euphorbia Rufacer = Acer Polanisia = Cleome Rugelia = Senecio Polemonium 210. 1960 umex 46: 99. 1965 Polycodium = Vaccinium *Ruppia 59: 185. 1978 *Polygala 52: 271. 1971 uta 43: 5. 1962 Polygonella 46: 112. 1965 *Sabatia 63: 449. 1982 562 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Saccharodendron = Acer Stuartia = Stewartia Sageretia 45: 449. 1964 Stylipus = Geum *Sagittaria 64: 413. 1983 Stylisma 41: 306. 1960 *Sambucus 47: 38. 1966 *Stylophorum 43: 323. 1962 — = Samolus *Styrax 41: 24, 1960 *Samolu 40: 286. 1959 Suriana 43: 176. 1962 ane 43: 322. 1962 Swertia: see Frasera i 396. 1974 Swida = Cornus Sanidophyllum = Hypericum Symphoricarpos 47: 49. 1966 indus 44: 470. 1963 oo 41: 51. 1960 Sapium 48: 391. 1967 *Symploc 41: 32. 1960 Sapota = Manilkara Synosma = Cacaha Sarothra = Hypericum Syringa 40: 375. 1959 Sarracenia 1: 153. 1960 *Talin 50: 578. 1969 Sassafras 39: 339. 1958 Tamala = Persea *Saururus 52: 482. 1971 Tamarindus 57: 46. 1976 avia 48: 325. 1967 araxacum §4: 83. 1973 *Saxifraga 53; 443.1972 Terminalia 45: 296. 1964 *Scaevola 47: 296. 1966 *Tetragonia 51: 459. 1970 Schaefferia 45; 222. 1964 *Tetrazygia 63: 437. 1982 Schinus 43: 365. 1962 Tetrorum = Sedum Schisandr 39: 319. 1958 Thalesia = Orobanche Schmailtzia = Rhus Therophon = Boykinia *Schoepfia 63: 397. 1982 *Thunbergia 51: 273. 1970 *Schrankia 55: 88. 1974 Thyella = Jacquemontia Schweinitzia = Monotropsis Thysanella = Polygonella Sebastiania 48: 385. 1967 *Tiarella 53: 430. 1972 si 59: 206. 1978 oe = oe Seren 50: 109. 1969 Tidestr 62: 299. 1981 Serinia = = Krigia Tilia 46: 290. 1965 *Sesuvium 51: 450. 1970 Tillaea = Crassula ae rw ee Shortia ee ae = Crassula *She 40: 167. 1959 *Tillandsi 56: 383. 1975 Shultzi = Obolaria Tiniaria = Polygonum Sibbaldiopsis — = Potentilla Sideroxylon: see Mastichodendron Sieversia = Geum Simarouba 43: 177. 1962 Sitilias = Pyrrhopappus Sonchus 54: 69. 1973 Sorbus = Pyrus *Sparga 63: 342. 1982 nium Spathyema = Symplocarpus Spatularia = Saxifraga Spiraea 55: 320. 1974 Spirodela 41: 66. 1960 *Sta 52: 199. 1971 Danan = Lysimachia *Stenandriun 51: 281. 1970 Stewartia 40: 418. 1959 *Sullingia 48: 388. 1967 *Stokesia 63: 500. 1982 Strophocaulos = Convolvulus Tithymalopsis = Euphorbia Tithymalus = Euphorbia el sensu oo = Pedilanthus orrubia is ae = Ferien noron Toxicodendron = Rhus Tracaulon = Polygonum Tragia 48: 376. 1967 Tragopogon 54: 60. 1973 *Trema 51: 37. 1970 *Triadenum 57: 88. 1976 Triadica = Sapium *Trianthema 51: 453. 1970 *Tribulus 53: 541. 1972 Triclisperma = Polygala Trientalis 40: 281. 1959 Triosteum 47: 47. 1966 Triphasia 43: 13. 1962 *Triumfetta 46: 300. 1965 1983] WOOD, INDEXES TO GENERIC FLORA 563 Tubiflora = Elytrari Wallia = ae Tulipastrum = Maen olia Walther 47: 72. 1966 Turbina 41: 316. 1960 Wolffia 41: 71. 1960 Turnera 42: 208. 1961 Wolffiella 41: 70. 1960 *Ulmus 51: 21. 1970 Xanthoxalis = Oxalis *Urtica 52: 47. 1971 es aaa = Zanthoxylum Urticastrum = Laportea *xime 63: 393. 1982 Vaccinium 42: 69. 1961 Yolisma 1a = Lyo Vachellia = Acacia Xvlophylla = Phyllanthus Valeriana 46: 220. 1965 Xylosteon = Lonicera Valerianella 46: 223. 1965 *X yris 64: 428. 1983 Vernonia 63: 494. 1982 Yeatsia = Dicliptera Viburnum 47: 41. 1966 Youngia 54: 89. 1973 Viola 42: 324. 1961 Zanthoxylum 43: 6. 1962 Vitis 46: 61. 1965 alee 42: 45. 1961 *Voyria 63: 483. 1982 Ziz 45: 459. 1964 Waldsteinia 55: 388. 1974 a ine = Euphorbia ARNOLD ARBORETUM RVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS 02138 The Generic Flora of the Southeastern United States is a long-term, comprehensive project covering the 192 families and approximately 1280 genera of seed-bearing plants of North and South Carolina, Georgia, Florida, Tennessee, Alabama, Mississippi, Ar- kansas, and Louisiana, an area of some 444,000 square miles (1.15 million square kilometers). For each family and genus original observations are synthesized with pub- lished information from botanical and other journals of the world to show much nee is and is not known about each group. The treatment for each family and its genera includes descriptions, keys, original place of publication, principal synonyms, common names, number of species (and sometimes brief descriptions and distributions of these), infrafamilial and infrageneric ranks represented, relationships, diagnostic features, eco- nomic importance, and a wide variety of notes from all branches of botany. Descriptions are based primarily on the plants of the southeastern United States, but information about extraregional members ofa family or genus 1s often included. Annotated references provide a summary of published information. Original, detailed 1 generi characteristics and various features of structure and biolog In addition to the one hundred articles indexed above, i. are fifty collateral papers and notes that have developed from this work. The 150 papers total 2664 pages. Four hundred and ninety plates illustrating 552 genera have been prepared, with one to twenty- seven plates for each of 179 families. Initially begun through the generosity of George R. Cooley (see frontispiece and ded- ication of Journal of the Arnold A: rboretum, Vol. 63, no. 4, 1982), work on the Generic currently supported by BSR-8111520 (C. E. Wood, Jr., principal investigator), under which these indexes were prepared, and by oe 8303100 (N. G. Miller, principal in- vestigator). CHENG & YANG, ASARUM 565 A SYNOPSIS OF THE CHINESE SPECIES OF ASARUM (ARISTOLOCHIACEAE) CHING-YUNG CHENG AND CHUN-SHU YANG THE GENUS Asarum L. has its center of distribution in eastern Asia, especially in China and Japan. The genus includes two distinct subgenera, each with several subcategories. Subgenus AsArum is characterized by almost distinct perianth lobes and more or less united styles with terminal stigmas, while subg. Heterotropa has an obvious floral tube and six free styles with lateral stigmas. Due to these and other differences, the genus has occasionally been divided into two, three, or even more segregate genera (e.g., Hexastylis Raf., Heterotropa Morren & Dene., 4siasarum Maekawa, Japonasarum Nakai, and Geotaenium Maekawa). The characters utilized to distinguish these genera, however, are of comparatively minor significance. Since Asari appears to be a natural and consistent taxon with close relationships among its species and salient evolu- tionary trends in its floral structures, it has been maintained by most authors (e.g., Braun, 1861; Hayata, 1915; Maekawa, 1932: Liu & Lai, 1976), as well as by the present ones. In this paper we divide Asarui into two subgenera and propose five sections and four series, essentially following Araki’s (1937) hi- erarchy but with some emendations and modifications. In China species of Asaruim that produce pungent aromatic roots are used in traditional medicine as a remedy for pain and colds. Among them, 4. heterotropoides var. mandshuricum is considered of the greatest value, although A. sieboldii is believed to have been the first species used in ancient Chinese medicine. All the remaining species, including those utilized in folk medicine, are used for the same or slightly different purposes. CONSPECTUS OF THE GENUS ASARUM IN CHINA' Asarum L. Sp. Pl. 442. 1753; Gen. Pl. ed. 5. 201. 1754. LecroryPe species: A. europaeum L., fide Britton & Brown, Ill. Fl. N. U.S. ed. 2. 1: 642. 1913. peers Raf. Neogenyton, 3 - 1825. Type species: HH. arifolia (Michx.) Raf., based Asarum arifolium Michx. (U.S. A.) ee Morren & Dene. Ann. Sci, Nat. Il. 2: 314. 4. 10. 1834. Type species: /7. asaroides Morren & es (Japan.) Japonasarum Nakai, Fl. Sylv. Koreana 21: 16. 1936. Type spectes: J. caulescens (Maxim.) Nakal, i on. Asarum caulescens Maxim. (Japan, Korea Asiasarum Maekawa in Nakai, Fl. Sylv. Koreana 21: 17. 1936. Type species: A. sieboldti Miq.) Maekawa, based on yee sieboldii Miq. (Korea.) Geotaenium Maekawa, Proc. VII Pacific Sci. Congr. 5: 217. 1953. Type species: G. epigynum (Hayata) Maekawa, based on Asarwm epigynum Hayata. (China, Taiwan.) Some species occurring in Taiwan are not included in this treatment because we have not seen adequate material. © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 565-597. October, 1983. 566 JOURNAL OF THE ARNOLD ARBORETUM [vVoL. 64 Perennial herbs; rhizomes short and oblique or long and prostrate; roots and rhizomes pungently aromatic. Leaves | or 2, usually long-petiolate, variously cordate or sagittate. Flowers solitary, axillary or between 2 opposite leaves, greenish, usually with tufts of short purplish hairs (rarely glabrous); peduncles usually short, with | or 2 cataphylls at base; perianth of 3 sepals, these separate or coalescent at base and forming a cupular, urceolate, campanulate, or fun- nelform tube, the tube often ridged or tessellate on inner surface and constricted at upper end to form orifice, the orifice usually with laminate ring: perianth lobes 3, erect, spreading, or reflexed; stamens 12 in 2 series, rarely with extra whorl of 3 aberrant stamens or staminodes, filaments evident or anthers sub- sessile; ovary inferior or half inferior (Sometimes nearly superior), 6-loculate, the placentation axile with numerous ovules, the styles 6, distinct with essen- tially terminal stigmas or united and with lateral stigmas. Fruit a fleshy, ber- rylike capsule, splitting irregularly when ripe; seeds numerous, more or less cordate, convex dorsally, plane or grooved ventrally, arillate at base. 2n = 12, 24, 26 About ninety species, mostly in the warm-temperate region of eastern Asia; a few species in North America and one endemic to Europe. Thirty-one species, three varieties, and one form occur in China and are distributed mainly along the Yangtze River. A key to the subgenera, sections, and series into which these species fall is presented below. KEY TO THE SUBGENERA, SECTIONS, AND SERIES OF ASARUM — Perianth with lobes essentially free above ovary, not forming distinct tube or only very short one; stamens with + long filaments; styles united into column, the 6 stigmas radiating, essentially terminal (subg. ASARUM). 2. Perianth with tube completely lacking. the lobes free, erect or rarely reflexed (sect ASARUM 3. Perianth lobes erect, caudate or pointed. ................ ser. Calidasarum. 3. Perianth lobes reflexed, neither caudate nor long-pointed. ................ See ae Mer eet ee ue Leet Sete eel ee ek ee ee ser. Japonasarum. 2. Perianth with tube very short; the lobes united at very base, erect or reflexed. AReip sarees Wott ee de Ae yg a te ies Sn see ee sect. BREVITUBA. Perianth with lobes united above ovary, forming tube of various shapes; stamens subsessile or with very short filaments, if with long filaments then inserted on ovary: styles 6, free or rarely united basally, often bifid at apex, stigmas lateral or rarely + terminal (subg. HET EROLEORS A). Stamens inserted on ovary, filaments longer than anthers; ovary he anaes styles short; perianth ou throughout, orifice lacking _- 1E TINGS v.40 ses Seetad fe heat aie Stace ta ee herent A ded ota diy een he ea eee he ia ae M, 4. Stamens not inserted on ovary, filaments usually very short; ovary we or half inferior, rarely almost superior, styles long; perianth with lobes often with papillate or pulvinate areas at base, orifice usually with laminate ring. 5. Perianth with tube usually 1-4 cm long, often with conspicuous orifice ring, papillate area localized at base of lobes (sect. HETEROTROPA). 6. Styles not bifid at apex, stigmas terminal or subterminal. .............. b Sepah ateatae duntea ten dul eka, foe-sen ag aur ee oon ta agie Aare ser. Achidasarum. 6. Styles bifid at apex, stigmas lateral. ...0....000...000.... ser. Bicorne. Perianth with tube usually 5-8 cm long, orifice lacking evident ring, papillae scattered from middle of lobes downward into tube and arra pera in vertical TOWS:. 2ooutiweleedeta ke taer b 62a de dhestamehete nest . LONGIFLORA. —_ wr 1983] CHENG & YANG, ASARUM 567 SUBGENUS ASARUM Asarum subg. Asarum Asarum subg. Choridasarum Araki, Acta Phytotax. Geobot. 6: 125. 1937. Perianth lobes free or united only at base and forming very short tube; stamens with filaments usually longer than anthers; styles united into column, 6-lobed at apex, stigmas mostly terminal. Type species: 4. europaeum L. Nine species and one variety in two sections and two series in China. Key to the Species of 4Asarum subg. ASARUM 1. ee mee totally free, erect or reflexed, not forming tube. 2. Perianth lobes erect and caudate or long-pointe 3. Pein lobes with abrupt tail ca. | cm lorie: lobes of styles entire, stigmas terminal. ee broadly ovate, triangular-ovate, or ovate-cordate, ere and mostly without white blotches above. ....................5. caudigerum. Leaves mostly subcordate or broadly ovate, with minute; as spaced white blotches above. ........... la. A. caudigerum var. cardiophylum. 3. Perianth lobes gradually tapering into long point to | cm long; lobes of styles notched, with stigmas at base of notches. ............. 2. 4 renicordatum. 2. Perianth lobes reflexed, with apex acute or oe never caudate or long-pointed: leaves usually 2 to 4, apparently oppos 5. Plants densely covered with long, ae hairs (black when dry); stamens and ee not —— stigmas terminal; leaves broadly ovate, acute or short- AGUIMINGALCs: cassie. ch ga eane Peed a net bob uel a ee hie Bin ele den eh ed 4. pulchellum. 5. ae eee pubescent; stamens and styles often exserted; lobes of styles notched, stigmas outside of notches; leaves cordate, acuminate. ........... 4. — Perianth lobes united at base, forming very short tube above ovary 6. Perianth lobes with short tail 1-4 mm long; leaves pubescent only along veins. qe ae ts 20-30 cm high: perianth lobes with incurved tail ca. 2mm long; stamens , connectives elongated and ligulelike. .............. 5. A. caudigerellum. 7. sere 10-15 cm high: perianth lobes acute or, if short-tailed, tail not incurved; 12 or fewer, connectives not elongated. .............. 6. A. debile. sta 6. oo ae not tailed: leaves usually with scattered pubescence on ee sur- 3. bert lobes bec styles slender, stigmas not radiating downward; duncles long and filiform. .....00.00.00.0..0...0..000.. 7. A, hi a 8. Perianth lobes one or spreading: styles Sy and thick, stigmas radiating downward; peduncles rather stout, never filif 9. Leaves cordate or broadly ovate; ovary ere ovoid, with 6 pronounced ribs; mainland China. ..........0.. 00.00.0000. 0c uu 8. A. geophyllum. 9. Leaves narrowly ovate or triangular ovate; ovary narrowly obovoid, lackin ribs; Taiwan and Hainan. ..................0020 0000 . A. epigynum. Asarum sect. Asarum Asarum sect. Euasarum Braun. Index Sem. Berol. Append. 12. 1861. Asarum sect. Ceratasarum Braun, ibid. Perianth lobes essentially free, not forming true tube. Asarum sect. ASARUM comprises two series with four species and one variety in China. 568 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Asarum ser. Calidasarum (Araki) C. Y. Cheng & C. S. Yang. stat. nov. Asarum sect. Calidasarum Araki, Acta Phytotax. Geobot. 6: 125. 1937. Perianth lobes erect. long-pointed to caudate at apex. Type SPECIES: Asaruim leptophyllum Hayata (= A. caudigerum Hance). Two species and one variety in China. |. Asarum caudigerum Hance, London Jour. Bot. 19: 142. 1881; Hemsley in Curtis’s Bot. Mag. 116: 4. 7/26. 1890. Type: Guangdong, East River. Hance 21336 (n.v., photo Pe’). Asarum leptophyllum Hayata, Ic. Pl. Formosa. 5: 147. 1915. Type: Taiwan. Arisan. favata & Ito s.n. (photo Pe). Asarum leptophyllum var. triangulare Hayata, ibid. Type: Taiwan, “Arisan.” [to & Hayata s.n. (photo pr). DistriBUTION. Zhejiang. Jiangxi, Fujian. Taiwan, Hubei, Hunan, Guangdong. Guangxi, Sichuan, Guizhou, and ra Thickets along streams and road- sides; frequently at 350-1160 m Hayata described Asariwn leptophyllum from Taiwan in 1915. He stated that itis allied with 4. caudigerum Hance and 1s distinguished only in having hght green flowers instead of reddish ones. He also considered plants with triangular leaves to constitute var. triangu/are. In fact, the reddish flowers of Asaruim are due to the presence of reddish hairs on the otherwise greenish perianth lobes. Both the amount of pubescence and the shape of the leaves are variable on individual plants. Based on comparisons of material from Fujian with the figures presented by Liu and Lai (1976), we cannot support the separation of these three taxa. la. Asarum caudigerum Hance var. cardiophyllum (Franchet) C. Y. Cheng & C. S. Yang, comb. et stat. nov. Basionym: -lsaruim cardiophyllum Franchet, Bull. Mus. Hist. ae Paris 1: 66. 1895. Syntypes: Yunnan. Long-ki, De/avay 5/50 (photo pr), 5205 (n.v.). According to Franchet, Asaruit cardiophyllum is closely related to -l. cau- digerum Hance, differing in having floral stems that bear four leaves in two pairs, and styles that are united into an undivided column. These features, however, also occur in typical 4. caudigerum. Since plants of 4. cardiophyvllum from Sichuan and northern Yunnan have mainly cordate leaves, usually with -Acronyvms a herbaria follow those outlined by Holmgren, Keuken. & Schofield in /adex HHer- bariorum, ed. 1981 (Regnum Vegetabile 106), with the following exceptions (all located in the People’s ea of China) that do not appear in that work: Anu. Anhui University, Hefer, Anht Gim, Guizhou Institute of Chinese Traditional Medicine, Guivang, Gaga HIM, Hangzhou asin of Materia Medica. Zhejiang Academy of Experimental Sanitary Sciences, Hangzhou, Zhejiang: ors. Sichuan School of Chinese Traditional Medicine. Omei, Sichuan: ptm. Beying Chinese Traditional Medicinal College. Beijing: and sir. Sichuan Institute of Chinese Traditional Medicine, Chongqing. Sichuan 1983] CHENG & YANG, ASARUM 569 fine white patches on the upper surface. we feel that these plants are best accorded varietal status under 4. caudigerum. 2. Asarum renicordatum C. Y. Cheng & C. S. Yang, sp. nov. FIGURE I. Proximum A. caudigero, sed foliis renicordatis, lobis perianthii apice longe attenuatis, lobis stigmatis obcordatis, differt Acaulescent perennial herbs from long, slender rhizomes. Leaves 2, opposite; petiole 10-14 cm long; blade renicordate or subcordate, 3-4 by 6-7.5 cm, rounded and obtuse (rarely acute) at apex, with scattered hairs above, more densely pubescent below, margin densely ciliate. Flowers solitary between leaves: peduncle ca. 2.5 cm long, densely pubescent: perianth campanulate, yellowish brown-hairy outside. the lobes slightly united at base but not forming tube, triangular-lanceolate, ca. 10 by 4 mm, usually attenuated into point 4-10 mm long at apex; stamens with filaments ca. | mm long, connectives awl shaped: styles united and 6-lobed at apex, the lobes notched, + inverted heart-shaped, the stigmas attached at notches. Fruits not seen. Tyre. Anhui, Huang-shan, mountain slope near stream, Nanjing Pharmacy College s.n. (holotype, PTM; isotype. PEM). AbDDITIONAL SPECIMEN EXAMINED. Anhui: Jui-hua-shan, 720 m alt., 42 5583 (Anu). Asarum renicordatum is closely related to 4. caudigerum Hance, but differs in having renicordate leaves, perianth lobes with gradually tapering, long- pointed apices, and notched style tips. While its leaves are similar to those of A. forbesii Hance, its flowers differ considerably. The specific name is derived from the broad, kidney-shaped leaves of the plants. Asarum ser. Japonasarum (Nakai) C. Y. vans & C.S. Yang, stat. nov. Japonasarum Nakai, Fl. Sylv. Koreana 21: 16. 1936. Asarum sect. Japonasarum (Nakai) Araki, ee Phytotax. Geobot. 7: . 1937. Perianth lobes reflexed downward. Type species: Japonasarum caulescens (Maxim.) Nakai (= Asarum caulescens Maxim.). Two species in China, others in Japan and Korea. 3. Asarum pulchellum Hemsley, Gard. Chron. HI. 7: 422. 1890; Jour. Linn. Soc. 26: 360. 1891. Type: Hubei, Henry 7800 (photo Pe). DisTRIBUTION. Anhui, Jiangxi, Hubei, Sichuan, and Guizhou. In woods in mountainous areas, often in loamy soil; abundant at 700-1700 m alt. This species is easily distinguished from near allies by its dense, long, white hairs, which become black when dry. 4, Asarum caulescens Maxim. Bull. Acad. Sci. St.-Pétersb. 17: 162. 1872. Type: Japan, Tschonoski s.n. (photo of syntype PE). Asarum caulescens var. setchuenense Franchet, Jour. Bot. Morot 12: 302. 1898. Type: SICHUAN, Tchen keou tin, Farges s.n. (n.v.). 570 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Ficure |. Asarum renicordatum: a, habit of Hone * .8; b, styles and ra- diating stigmas, < 1.6: c, anther, x 1.6; d. perianth lobe. 1983] CHENG & YANG, ASARUM oval Asarum franchetianum Diels, Bot. Jahrb. 29: 308. 1901. Type: Sichuan, Nanchuan, Bock von Rosthorn 2446 (photo PE) Japonasarum caulescens (Maxim.) Nakai, Fl. Sylv. Koreana 21: 17. 1936. DISTRIBUTION. Shaanxi, Gansu, Hubei, Sichuan, and Guizhou. This species was first described from Japan. It has not been found in the eastern provinces of China, although it has often been collected in western China. This discontinuous distribution caused Franchet to distinguish the Chinese plant as Asarum caulescens var. setchuenense, and he even contem- plated recognizing it as an independent species. Franchet distinguished the Chinese plants on the basis of their cordate style lobes and their stamens that are slightly shorter than the stylar column. However, dissections of flowers from Japanese specimens and comparisons with dissections of Chinese spec- imens (including plants from Sichuan, Hubei, and Shaanx1) indicate few if any differences between the Japanese and Chinese plants. Consequently, we include var. setchuenense in synonymy. With regard to Asarum franchetianum, Diels pointed out that the perianth tube is adnate to the ovary only halfway in the Japanese plant, while in the plants from Sichuan the adnation 1s almost com- plete. However, if the overall similarity of floral morphology is considered, 4. franchetianum seems inseparable from 4. caulescens. Asarum sect. Brevituba C. Y. Cheng & C. S. Yang, sect. nov. Perianthium lobis basi coalescentibus et tubo brevi. Type SPECIES: Asarum himalaicum Hooker f. & Thomson. Four or perhaps five species in China. 5. Asarum caudigerellum C. Y. Cheng & C. S. Yang, sp. nov. FIGURE 2. Species A. debili similis laminis foliorum in medio margineque leviter in- curvatis, base perianthiis breve tubularibus et lobis in apicem brevissime cau- datis, differt plantis altioribus. foliis apice acuminatis vel longe acuminatis, connectivis antherarum apice valde productis, longe ligulatis. Perennial herbs, 20—40 cm high; rhizomes long and horizontal with prolonged internodes; stems 2—5 cm long, inclined at tip, with | or 2 pairs of opposite leaves. Leaves with petiole 4-18 cm long, the cataphylls at its base broadly ovate, ca. 2 by 1-1.5 cm, densely ciliate; blade cordate or oblong-cordate, 3- 7 by 4-10 cm, acuminate or long-acuminate at apex, slightly incurved at margin, cordate at base. Flowers dark purple; peduncle 1-1.5 cm long; perianth with the tube very short, ca. 5 mm long, 10 mm in diameter, the lobes triangular- ovate, 10 by 7 mm, acuminate with incurved tail ca. 5 mm long, pilose on outer surface; stamens with filaments usually longer than anthers; ovary infe- rior, subspherical, 6-ribbed, pilose, the styles united, with 6 short radiating branches at apex Type. Sichuan, Mt. Omei, between 1700 and 2100 m alt., under thickets, along mountain path, K. H. Yang 54370 (PE ae JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Figure 2. Asarum caudigerellum: a, habit of flowering plant, x .65; b, ep x 1.3; c, flower, with perianth lobe cut to expose stamens, x I: d, stam stylar an. showing coherent portion and terminal stigmas, x 1.3 ens, * 1.3 1983] CHENG & YANG, ASARUM 573 ADDITIONAL SPECIMENS EXAMINED. Sichuan: Mt. Omei, 1900 m alt... C. Y. Cheng 807a (peM); Nanchuan, /. 7. Yung 90302 (re); Beichuan, Econ. ae <5 (sit); Tien- Re HH. L. Tsiang 34001 (52), Hubei: He-feng, HH. J. Li 8487 (ptm): ie -en, H. J. Li 3989 (ptm). Guizhou: Na-yung, Bi-jieh Exped. 420 (pe). Yunnan: I-lia H. T. Tsai re (PE). Asarum caudigerellum is closely related to 4. debile Franchet but differs in having a larger stature; acuminate leaves; and 12 stamens, all of which are normally developed. produce viable pollen, and have anther connectives with long-produced tips. Asaruim debile, on the other hand, is a small, weak plant with acute leaves, 12 or usually fewer stamens, and anther connectives that are not produced. The specific name. a diminutive of caudigerum, refers to the short tail on the perianth lobes. 6. Asarum debile Franchet, Jour. Bot. Morot 12: 305. 1898. Type: Sichuan, Cheng-kou, Farges s.n. (P, n.v.). DistrRiBUTION. Anhui, Shaanxi, Hubei. and Sichuan. Found occasionally in rocky, moist places near streams; | 300-2300 m alt. 7. Asarum himalaicum Hooker f. & Thomson ex Klotzsch. Monatsber. Deutsch. . Wiss. Berlin 1: 385. 1859; Duchartre 7m DC. Prodr. 15: 424. 1864: Hooker f. Fl. Brit. India 5: 72. 1890. Type: Sikkim Himalaya, J. D. Hooker s.n. («). DistRIBUTION. Shaanxi, Gansu, Hubei, Sichuan, Guizhou, Yunnan, and X1- zang, also in India. Near streams and under thickets in shady, moist places. 8. Asarum geophyllum Hemsley, Gard. Chron. II]. 7: 422. 1890; Hooker f. Curtis’s Bot. Mag. 117: 4. 7/68. 1891. Type: Guangdong. “North River,” Ford 125 (photo Pe). Asarum cavaleriei Lévl. Repert. Sp. Nov. 9: 78. 1910. Type: Guizhou, “Lofou,” Ca- valérie 3671 (kK, n.v., photo PE) DISTRIBUTION. Guizhou, Guangxi, and Guangdong. In shady forests or moist valleys; 250-700 m alt. 9. Asarum epigynum Hayata, Ic. Pl. Formosa. 5: 140. 1915; Liu & Lai in Li, Fl. Taiwan 2: 557. 1976: Maekawa. Jour. Jap. Bot. 52: 231. 1977. Type: Taiwan, without precise locality, 7. Soma s.n. (n.v.). Geotaenium epigynum (Hayata) Maekawa, se VII Pacific Sci. Congr. 5: 219. 1953: ur. Jap. Bot. 52: 251. 1977: ibid. 53: 291, 1978. DisTRIBUTION. Taiwan and Hainan. In moist valleys and forests. Maekawa removed this species from -fsarun and placed it in the genus Geotaenium on the basis of the plant’s chromosome number (27 = 12) and its flowers with a rudimentary “corolla.” Otherwise, this species is a typical Asa- 574 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 rum, and its floral structure suggests a close relationship with 4. The rudimentary “petals” appear to be staminodia, and staminodia are also present in the flowers of A. macranthum (Maekawa, 1978). An extra whorl of aberrant stamens has also been reported in the flowers of 4. caudigerum of the geophyllum. same subgenus SUBGENUS HETEROTROPA Asarum subg. Heterotropa (Morren & Dene.) Schmidt /7 Engler & Prantl, Nat. Pflanzenfam. ed. 2. 16b: 230. 1935. Heterotropa Morren & Dene. Ann. Sci. Nat. If. 2: 314. ¢. 70. 1834. Asarum subg. Gamoasarum Araki, Acta Phytotax. Geobot. 6: 125. 1937. Perianth lobes united above ovary. forming variously shaped perianth tubes: stamens with very short filaments or subsessile; styles often free, rarely short and subconnate (in Japanese species), often bifid at apex. rarely subentire. Type species: Heterotropa asaroides Morren & Dene. (= 4Asarum asaroides (Morren & Dene.) Makino). wenty-two species, three varieties, and one form in China, these falling into three sections and two series. Key to the Species of Asarum Subg. HETEROTROPA Stamens with filaments evident, longer than anthers: ovary half inferior; styles short: perianth with the tube naked at throat, lacking papillae and ring, the lobes smooth at base, lacking papillae or pulvinate areas roar 2. Perianth lobes erect or spreading, never enexed at anthesis; leaves short-acu- minate, with scattered hairs over upper surface 3. Leaves usually hairy only on veins on lower surface, petiole ree Se pee Seu Sea e ae ee aG Rh ee ue bee ae aaah balan canes 10... _sieboldii 3. Leaves usually densely pubescent on lower surface, Ws iole pubescent. ..... Se hapie Wk Hie eae ies eee ah oah Ma an eey woe ae 10a. : daelaa f. seoulense. 2. eae ae alage at anthesis: leaves fa or blunt at apex. short-pubescent 11 A ele Ue var. mandshuricum. on fat veins above. .................. cae with fimients shorter than ee ovary inferior sionally nearly superior, styles rather long; perianth having che wath ( orifice ring at throat, lobes papillate or pulvinate at base. Styles not bifid at apex, stigmas terminal or subterminal. Perianth having the tube with orifice ring, inner surface tessellate with slightly elevated bars (rarely longitudinally ridged), the lobes not reflexed, usually papillate near base. Ss long and slender, mostly creeping; leaves elliptic-ovate, acu- nate: perianth hg gradually dilated above orifice, forming neck, then 12. A. chinense. r half inferior, occa- rarely without) ON ON mses =e ‘SEE PERE ocs3a. Oo —- “A SD on om SS re, thick, erect; leaves cordate or variously ovate, obtuse: perianth tube without neck, divided immediately into lobes. ........... pe oe ee Bagens eye ees ee Rec tee ane da ea eee ee ey ein Perianth with the tube lacking orifice ring, inner surface longi ly ridge the lobes strongly reflexed laterally, pulvinate at base. .... 14. v1. ieee 4. Styles bifid at apex, stigmas attached laterally just below wen 7. Perianth having the tube generally urceolate with conspicuous constriction at oa) 1983] CHENG & YANG, ASARUM 515 throat, usually with ring, inner _ longitudinally ridged or tessellate, the lobes papillate or pulvinate at Perianth tube tessellate on inner surface. 9. Perianth tube only slightly constricted at throat, orifice ring narrow 8. 10. Perianth tube broad, cup shaped or hemispherical, the orifice — ca 1.5 cm in diameter, ring inconspicuous or very narrow; leaves with white patches on upper surface. .... 15. A - chingchengense. 10. Perianth tube narrow, obconical, de or ciate 1, the orifice small, 4 nm in diameter, ring conspic 11. Flowers ca. mi mm in diameter; perianth oa the tube cylin- drical or campanulate, the lobes broad-ovate, ca. 5 mm long, without papillae at base. ...............0-. 16. A. forbesil. . Flowers 20-30 mm in diameter; perianth with the tube ob- conical, the lobes triangular or ne 10-15 mm wide, with papillate area at bas 12. Leaves ovate, 4-5 by 3-4 cm, usually light green beneath; cataphylls ovate, | cm long: perianth — ith densely papillate area at base. ............... A. taitonense. . Leaves broad- see or r triangular- -ovate, - 10 by aes purplish beneath; cataphylls long, narrowly ovate, mm long; perianth lobes with only scattered pa pa Seba beta ate ee eee ee 18. A. infrapurpurcum _ _ NO ° 9. Perianth tube strongly constricted at throat, orifice ring broad and ev ident. 13. Stigmas elongated, each extending along notch of bifid style and + hooked at free end. 14. Leaves triangular-ovate. veins reddish beneath, petiole streaked reddish brown; stamens oblong, anther connectives thick and bilobed at apex: stigmas elongated, decidedly hooked. ...... ecu Wis Sosa cksensends Saw ttios eee ape eee 19. A. macranthum. . Leaves ovate-cordate or narrowly ovate, the veins green be- neath, the petiole green, not streaked; stamens triangular, an- ther connectives not bilobed at apex; stigmas oblong-ov slightly hooked. .................2--5 20. A. ae Ne & 13. Stigmas + ovoid, neither elongated nor hook ked. 15. Flowers to 5 cm in diameter: perianth with the tube subcylin- drical, orifice ring 2-3 mm wide, the lobes Dias ovate, ca. 3 by 3-4 cm; leaves large, 7-15 by 6-11 cm. ..21 A. delavayi. . Flowers to 2.5 cm in diameter; perianth with the ee urceolate, orifice ring ca. | mm wide, the lobes broad- or triangular-ovate, ca. 8 by 8-12 mm; leaves 6-10 by 5-7 cm. 16. Leaves dull green, white blotched above, purplish red be- low; anther connectives pointed. .................0005 Wa ikea abate ea dya ga eaee eee 22. A. porphyronotum. 16. Leaves deep, lustrous green and not es white above, light green below; < aouicl connectives rounded. ........ uinrine ee ae ded le 2a. A. pois ea var. atrovirens. — nN 8. Perianth tube longitudinally Eves on inner surface, cross bars lacking or very faintly develops 17. Perianth tube swollen ae middle or apex, forming barrel-shaped or girdlelike dilated zo 18. Perianth with the oe having pinkish, oe papillate area at base, the tube having dilated zone broad and near apex, inner surface ee pease ridges ca. 0.5 mm oak re ce with TALLOW. TU Oso ycsciseie koe aiane a des hee aed Ge ee Sates 23. A. inflatum. 576 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 18. Penianth with the lobes having white or yellowish, circular or semicircular pulvinate area at base, the tube ig inner surface of dilated zone without ridges, orifice ring abse 19. Perianth having the lobes with transverse rows of papillae below pulvinate area, the tube with dilated zone at middle aig and with orifice large, ca. 1 cm in diameter: els 1-5 cm long. ...............0.. 24. A. maximunn. 19, Perianth with the lobes lacking transverse rows of papillae below pulvinate area, the tube with dilated zone near beste orifice small; pedicels ca. 9 cm long. pote th 25. A. insigne. 17. Perianth tube enlarged gradually, not forming ee dilated zone. Perianth tube and lower leaf surface glabrou 21. Perianth having the tube constricted aise ‘at throat. lacking orifice ring. the lobes with insignificant area of papillae. ea pet Gane hee eee ee eed 26. A. nanchidnense Perianth having the tube constricted deeply at throat, with conspicuous orifice ring, the lobes with prominent papillate area al base. 22. Robust plants with short rhizomes and + thick, fleshy roots; leaves narrowly to broadly ovate, or hastate-ob- long, 15-25 by 11-14 cm: flowers usually 2: anther con- neclives pointe Cs euteecae eee _ AL sagittartoides. Slender plants with long rhizomes and fibrous, rarely + ie hy roots; leaves cordate or narrowly elliptic-ovate, 8— 4 by 5-8 cm; flowers usually ie anther connectives 21. to bo a : igulate. 20.0.2... . dongerhizomatosum. 20. Peranth ty and lower leaf surface 2 pubescent with vel- JOWISH DAIS 2 peso osc vp y ares 29. A. wulingense. 7. Perianth with the a ren not constricted at throat or only slightly so. lacking orifice ring but occasionally with ring of papillae, the lobes with papillae at base usually me and scattered, sometimes in vertical rows and extending downward into 23. Perianth tube erect or symmetrical: leaves triangular-ovate or elliptic- ovate, acute or short-acuminate; rhizomes short, erect. 24. Perianth tube 3-5 cm long, not much widened above; leaves trian- gular-ovate or subtr aneeiar, the upper surface with whitish blotches, oa a eects eet iene eon ee 30. AL. aes wn. 24. Perianth tube ca. | c lone: widened from middle: leaves ellip ovate, the upper sur sparingly pubescent, not white blotched. . oe ote eee eee eee. 30a. 4. magnificum var. dinghuense. Perianth tube often inclined to one side (slightly constricted at throat). slightly asymmetric; leaves narrowly ovate, triangular-ovate, or oblong- hastate, long-acuminate; rhizomes to 20 cm long, horizontal to Los) Asarum sect. Asiasarum (Maekawa) Araki, Acta Phytotax. Geobot. 6: 126. 1937 Astasarum Maekawa in Nakai. Fl. Sylv. Koreana 21: 17. 1936. Perianth with tube lacking orifice ring, lobes smooth throughout; stamens inserted on surface of ovary, with long filaments. Type species: sarum sieboldii Miq. (Japan). Two species and one variety in China. 1983] CHENG & YANG, ASARUM Sigs 10. Asarum sieboldii Mig. Ann. Mus. Bot. Lugduno-Batavum 2: 134. 1865. Type: Japan, Yezo, Siebold s.n. (tL, n Asiasarum sieboldii (Mig.) Maekawa in Nakai, Fl. Sylv. Koreana 21: 22. 1936. DistriBUTION. Shandong, Henan, Shaanxi, Anhui, Zhejiang, Jiangxi, Hubei, and Sichuan; also in Korea and Japan. Moist soil in forested areas; | 200-2100 m alt. 10a. Asarum sieboldii Miq. f. seoulense (Nakai) C. Y. Cheng & C. S. Yang. Stat. nov. BasionyM: Asarum sieboldii Miq. var. seoulense Nakai, Repert. Sp. Nov. 13: 267. 1914: Bot. Mag. Tokyo 28: 519. 1914. Type: Korea, Mt. Peukansan, NV. Okada s.n. r1?, n.v.). Asiasarum heterotropoides Schmidt var. seoulense (Nakai) Maekawa in Nakai, FI. Sylv. Koreana 21: 20. 1936 DistripuTION. Liaoning, Jilin, and extending northward into Korea. Habitat similar to that of the typical form. Nakai compared the Korean plant with Asarum sieboldii and established var. seoulense on the basis of the dense pubescence on the lower leaf surface. This character, however, does not seem sufficient to warrant varietal status. and we think that the taxon is better recognized at the rank of forma. 11. Asarum heterotropoides Schmidt var. mandshuricum (Maxim.) Kitagawa. Lineam. Fl. Manshur. 174. 1939 Asarum Lean mandshuricum Maxim. Mélanges Biol. Bull. Phys.-Math. Acad. Sci. St.-Pétersb. 8: 399. 1871. Type: S Manchuria, without collector or number (LE, 2.) Aslasarum Ce Dien (Schmidt) Maekawa var. mandshuricum (Maxim.) Mae- kawa in Nakai, Fl. Sylv. Koreana 21: 18. 1936. DistrRiBUTION. Heilongjiang, Jilin, and Liaoning: occasionally cultivated in the southern provinces. Moist, cool. shady forests. Asarum sect. Heterotropa (Morren & Dene.) Braun, Index Sem. Hort. Bot. Berol. Append. 13. 1 Heterotropa sect. Circinaria Mackawa, Jour. Jap. Bot. 57: 262. 1982. Perianth tubes variously shaped. usually with orifice ring: stamens subsessile or with very short filaments, never inserted on surface of ovary. TYPE SPECIES: Asarum asaroides (Morren & Dene.) Makin Eighteen species and one variety in Ge eee falling into two series. Asarum ser. Achidasarum (Duchartre) C. Y. Cheng & C. S. Yang, stat. nov. Asarum sect. Achidasarum Duchartre in DC. Prodr. 15: 426. 1864. Styles entire at apex, stigmas terminal. Type SPECIES: Asarum elegans Du- chartre = 4. parviflorum (Hooker f.) Regel (Japan, Yokohama). A small series comprising three species in China. [VOL. 64 . i Ficure 3. Asarum ichangense: a, habit of flowering plant, .65:; b, cataphyll, x 1.3: c, longitudinal section of perianth tube, showing tessellate inner surface and broad orifice ring, X 2.6; d, flower ae perianth tube removed to show androecium and gynoecium, x 1; e, stamens, x 1.3; f| apex of styles, with terminal stigmas, X 2. 1983] CHENG & YANG, ASARUM 79 12. Asarum chinense Franchet, Jour. Bot. Morot 12: 303. 1898. Type: Sichuan, Cheng-kou, Farges 1205 (Pp, n.v., photo PE). Asarum fargesii Franchet, Jour. Bot. Morot 12: 306. 1898. Type: Sichuan, Cheng-kou, Farges 966 (P, n.v., photo PE DISTRIBUTION. Hubei and Sichuan. In forested ravines; 1300-1500 m alt. Both Asarum fargesii and A. chinense were described by Franchet from material collected at Cheng-kou in northern Sichuan Province. During a special trip to the type locality of these two species to observe these plants in the field, we found that the criteria used by Franchet to distinguish plants of the two species (leaf coloration and the internal structure of the perianth tubes) do not hold. We have therefore placed 4. fargesii in the synonymy of A. chinense. 13. Asarum ichangense C. Y. Cheng & C. S. Yang, sp. nov. FIGURE 3. Affine A. chinensi, sed rhizomatis brevioribus, foliis cordatis vel ovato- cordatis apice obtusis, tubo perianthii valde constricto ad orem haud constricto ad collum, differt. Perennial herbs from short, nearly perpendicular rhizomes. Leaves with petiole 3-15 cm long, the cataphylls at its base ovate or narrowly ovate, ca. cm long, ciliate; blade cordate or cordate-ovate (rarely hastate), 3-6 by 3.5— 7.5 cm, obtuse or acute at apex, cordate at base, usually dark green or rarely with white patches along midvein above, shortly pubescent along veins above, glabrous below. Flowers purplish green to dark purple, 1-1.5 cm in diameter; peduncle ca. | cm long, recurved; perianth with the tube spherical, ca. | cm in diameter, tessellate on inner surface, strongly constricted at narrowly ringed throat, the lobes triangular-ovate, 1-1.4 cm by 9-10 mm, with small rugose, papillate area at base; anthers with connective produced, rounded with notch at apex; ovary + superior, the styles united, with 6 short, radiating arms at apex, the stigmas ovoid, terminal. — Type. Hubei, Ichang, on shady and moist mountain slopes, Yang & Xueh 74004 (holotype, PTM). ADDITIONAL SPECIMENS EXAMINED. Zhejiang: Shou-chang, in forests or in open grass land, Bot. Res. Exped. 27075 (re); Mt. ce near Kin-hua, Wigo s.n. (iBK, PE). Jiangxi: Lu-shan, M@. C. Wang 707 (Lus, pe), Wu-gung-shan, in shady, moist thickets, Jiangxi Exped. 530 (pe). Hunan: Nan-yueh, es -chi, L. H. Liu 15744 (pe); Heng-shan, 4xon- ymous 172 (PE). Asarum ichangense is widely distributed along the Yangtze River and may extend southward into Fujian, Guangdong, and Guangxi provinces. Its leaves vary considerably in shape, size, and color patterns. The close relationship of A. ichangense and A. chinense 1s evidenced by their floral structure— both have small flowers with united styles and terminal stigmas. Asarum chinense differs, however, in its long, running rhizomes, its acuminate and essentially elliptic- 580 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 oti t Uff 4 A at SO Ra) Ni sg Sy \ bh Na . Figure 4. Asarum fukienense: a. habit of flowering plant. x .65: b, habit of fruiting plant, x .65; c, cataphyll, x 1.3: d, portion of inner surface of perianth tube and reflexed lobe, X 1.3; e, stamen, x 1.3; f, styles, x 2. 1983] CHENG & YANG, ASARUM 581 ovate leaves, and its perianth tubes with a short neck above the constriction. In A. ichangense the rhizomes are perpendicular, short, and thick, the leaves are generally cordate with obtuse or acute apices, and the perianth tubes lack a neck. 14. Asarum fukienense C. Y. Cheng & C. S. Yang, sp. nov. FIGURE 4. Foliis deltoideo-ovatis, foliis subtus et pedunculisque pilis fulvis vestitis, species haec habitu A. wulingense similis, sed lobis perianthii sub anthesis marginibus reflexis, ex medio ad basin pulvinatis. stylis apice stigmatis coronatis interdum apice leviter bilobis lateraliter stigmatosis, differt. Perennial herbs from short rhizomes with clustered roots. Leaves subco- riaceous, triangular-ovate or narrowly ovate, 4.5-10 by 4-7 cm, acute or shortly pointed at apex, auricular-cordate at base, deep green with occasional white patches above. densely yellowish brown-pubescent below. Flowers purplish green: peduncle 1—2.5 cm long, pubescent with soft, yellowish hairs: perianth with the tube cylindrical, 1.5 cm long, | cm in diameter. yellowish- pubescent outside, longitudinally on inner surface, lacking constriction or orifice ring at throat, the lobes broadly ovate, 15 by 10 mm, with margins recurved, especially laterally, and with yellowish, semicircular, pulvinate area extending from base to middle: stamens with filament very short, connective pointed; ovary inferior, the styles 6, free, entire (rarely shallowly notched) at apex, the stigmas terminal or subterminal. Capsules ovate-spheroid, 7-17 mm in diameter, often with persistent remains of perianth. Type. Fujian, Wu-i-shan, shady thickets, W. J. Wang 1929 (holotype, Lus: isotype, PE). ADDITIONAL SPECIMENS EXAMINED. Anhui: Qi-men, in valley in mixed forest, 47. X. Deng 5005 (pe); same locality, at 1500 m, shady moist place, WH. D. Sun 30 (pe); same locality, at 870 m, J. S. Yueh et al. 5321 (Nas, PE); Wu-i-shan, M. J. Wang et al. 2418 (LUS, PE). Fujian: Shang-hang. Y. Lin 4/50 (pe). Zhejiang: Shuei-chang, Bot. Res. Exped. 25738, 25903 (PE). Asarum fukienense is characterized by its yellowish brown pubescence and its nonconstricted perianth tubes with the lobes strongly recurved laterally. In overall appearance, 4. fikienense is similar to A. wulingense from Guangxi Province, but the latter differs considerably in having constricted perianth tubes with spreading, nonrecurved perianth lobes, distinct styles with bifid apices, and lateral stigmas. Asarum ser. Bicorne Araki, Acta Phytotax. Geobot. 6: 127. 1937. Perianth with the tube 1-2 cm long (rarely longer), longitudinally ridged or tessellate internally, constricted at throat, orifice evident and often having laminate ring, the lobes having papillate or pulvinate area at base. TYPE SPECIES: Asarum asaroides (Morren & Dene.) Makino. Fourteen species and one variety in China. Ns ioe) ine) JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 FiGure 5. Asarum chingchengense: a, habit of flowering plant, < .6; b, cataphyll, x 1.2: c, portion of inner surface of perianth tube and lobe, x 1.5; d, flower with perianth removed, showing androecium and gynoecium, x 2.4; e, stamens, X 1.2; f, bifid style ups and lateral stigmas, x 1.2. 1983] CHENG & YANG, ASARUM 583 15. Asarum chingchengense C. Y. Cheng & C. S. Yang, sp. nov. Ficure 5. Tubis perianthii leviter constrictis, ovariis subsuperioribus haec nova species habitu A. taitinensi similis, sed tubis perianthii cupularis non profundis ad orem leviter constrictis ad maximam aperturam, foliis et petiolis pedunculique glabris, differt. Perennial herbs from long, horizontal rhizomes, internodes ca. 1.5 cm long with + thick and fleshy roots clustered at nodes. Leaves with petiole 6-18 cm long, the cataphylls at its base triangular-ovate, 2 by 1.5 cm, ciliate: blade ovate-cordate, long-ovate, or nearly hastate, 6-10 by 5-9 cm, acute at apex, deeply auriculate or subcordate at base, green, usually with white blotches and sparingly pubescent above, glabrous beneath. Flowers purplish green; peduncle ca. 2 cm long: perianth having the tube cupular or hemispherical, ca. 1.4 cm long, 2 cm in diameter, slightly constricted at throat, orifice large with ring very narrow and inconspicuous, inner surface tessellate, the lobes broad-ovate, ca. 2 by 2.5, with semicircular, rugose, papillate area at base; stamens with filament very short, connective rounded; ovary + superior, the styles 6, free. forked or notched at apex, the stigmas ovoid, lateral. Type. Sichuan, Guan-xien, moist slopes in thickets or partial shade, C. Y. Cheng & C. S. Yang 63001 (holotype, PEM). oe AL SPECIMENS EXAMINED. Sichuan: Ching- aa FT. Wang 20386 (ve): Chung-qing, Jin-yun-shan, C. Y. Cheng & C. S. Yang s.n. (pem, ptm). Guizhou: Mei- tan, J. M. Chen 307 (pe); Chishuei, X. L. Chen 005 (Gim). Asarum chingchengense is closely related to A. taitonense Hayata from Tai- wan, which differs from the Sichuan plant in having deeper perianth tubes with a narrower throat, and pubescent lower leaf surfaces, petioles, and peduncles. 16. Asarum forbesii Maxim. Bull. Acad. Sci. St.-Pétersb. 31: 92. 1887. Type: N. Zhejiang, Forbes 2056 (photo pe). DistRiBUTION. Henan, Anhui, Jiangsu, Jiangxi, Hubei, Hunan, and Sichuan. Fairly common in thickets and forests; 200-800 m alt. ee Asarum taitonense Hayata, Ic. Pl. Formosa. 5: 148. 1915; Liu & Lai in Fl. Taiwan 2: 581. 1976. Type: Taiwan, Mt. Taiton, Faurie s.n. (photo of fragment PE). Heterotropa taitonensis (Hayata) Maekawa ex Nemoto, Fl. Jap. Suppl. 165. 1936; Jour. Jap. Bot. 53: 299. 1978. DistRiBUTION. Endemic to Taiwan. In forests and thickets: low to medium altitudes. 18. esi infrapurpureum Hayata, Ic. Pl. Formosa. 146. 1915; Liu & Lai in _ Fl. Taiwan 2: 579. 1976. Type: Taiwan, Sasaki s.n. (n.v.). DistRIBUTION. Endemic to northern Taiwan. In thickets and wet places. 584 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Ficgure 6. Asarum crispulatum: a, habit of flowering plant, x .65: b, cataphyll. x 1.3: c. longitudinal section through perianth, x .65; d, stamens, * 1.3: e¢. gynoecium. x 1.3: (1.3 f. styles, = 1983] CHENG & YANG, ASARUM 585 19. Asarum macranthum Hooker f. Curtis’s Bot. Mag. 116: ¢. 7022. 1888; Liu & Lai in Li, Fl. Taiwan 2: 579. 1976; Maekawa, Jour. Jap. Bot. 53: 296. 1978. Type: Taiwan, Kelung, Ford s.n. (K, n.v.). DISTRIBUTION. Endemic to Taiwan. Forested habitats; low to medium altitudes. Although J. D. Hooker did not designate a type for Asarum macranthum, he did indicate the sources of the dried specimens and the living plants utilized in drawing up his description. The material at Kew collected by Ford should be studied if a lectotype is to be selected. 20. Asarum crispulatum C. Y. Cheng & C. S. Yang, sp. nov. FIGURE 6. Affinis A. macranthi sed lobis perianthii modice undulatis, staminibus trian- gularibus, foliis subtus viridibus non rubris nervis et petiolis non-maculatis, differt. Perennial herbs from short rhizomes with clustered roots. Leaves with petiole 6-15 cm long, pubescent with short, soft hairs, the cataphylls at its base ovate, ca. 2 by 1.3 cm, ciliate; blade ovate-cordate or narrowly ovate, 5-9 by 2-2.5 cm. acute or short-acuminate at apex, cordate or auricular-cordate at base, deep green with occasional whitish patches and scattered hairs above, light green and glabrous beneath. Flowers purplish green, ca. 3-5 cm across; peduncle ca. | cm long; perianth having the tube urceolate, ca. 1.5 cm long, 1.2-2 cm in diameter, orifice small with ring ca. 1.5 mm broad, inner surface tessellate. the lobes ovate, |.8-2.2 by 2-2.6 cm. margins + undulate; stamens triangular, the filament short, broadened at base, the connective acute or obtuse at apex: ovary half inferior, the styles 6, bifid at apex, the stigmas oblong-ovoid, inserted in notch and hanging free, free end slightly hooked. Type. Sichuan, Nan-chuan. in moist habitat, Yang & Li 72004 (holotype, pT). ADDITIONAL SPECIMENS EXAMINED. Sichuan: without further locality, Yang 78-lan-001 (ors). Cultivated: hort. Beijing Medical College, Cheng 7808 (PEM). Asarum crispulatum is allied to A. macranthum, which differs in having blotched or spotted petioles, strongly undulate perianth lobes, linear-oblong stamens with notched connectives, and strongly hooked stigmas. The specific epithet of the new species, crispulatum, refers to its undulate perianth lobes. 21. Asarum delavayi Franchet. Bull. Mus. Hist. Nat. Paris 1: 66. 1895; Jour. Bot. Morot 12: 304. 1898. Synrypes: Yunnan, Long-ki, Delavay 5105 (p, photo PE), Delavay 5205 (p, n.v.). Heterotropa splendens Maekawa, Jour. Jap. Bot. 57: 261. pl. 14. 1982. DISTRIBUTION. Sichuan and northern Yunnan. Shady, moist slopes of forests and thickets, usually of southern exposure; 800-1600 m alt. The leaves of Asarum delavayi are variable in both shape and coloration. They are generally large (9-18 by 6-14 cm), with the upper surface green and the lower surface light green. Some leaves are occasionally found with white 586 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 blotches above or reddish purple coloration below, and other leaves combine these color variations. 22. Asarum porphyronotum C. Y. Cheng & C. S. Yang, sp. nov. FiGurRe 7. Similis A. delavayi, sed differt foliis subtus semper purpureis, floribus plu- ribus minoribus urceolatis circa 2.5 cm longis et 2—2.5 cm in diametro, tubis perianthi brevioribus ovaliformibus circum oreum intus angustatis annulis instructis, connectivis antherarum apice breviori-pyramidalibus. Perennial herbs from short rhizomes. Leaves with petiole purplish or purplish brown, the cataphylls at its base usually 2, broadly ovate, lower one larger, ciliate; blade elliptic-oblong to broadly ovate (rarely subhastate), 5-15 by 6- 13 cm, acute or obtuse at apex, auriculate to cordate at base, margin ciliate, dull green and white-blotched above, purplish red below. Flowers purple, 2- 2.5 cm in diameter; peduncle ca. | cm long; perianth having the tube urceolate, ca. 12 mm long, | 1-14 mm in diameter, constricted and with narrow orifice ring, tessellate internally, the lobes broadly triangular-ovate, ca. 8 by 11 mm, densely papillate in triangular pattern at base; stamens with filament very short, connective acute; ovary + superior. the styles 6, bifid at apex, the stigmas ovoid, lateral. Type. Sichuan, without precise locality. Cheng 7507 (holotype, PEM: isotype. PTM) ADDITIONAL SPECIMENS EXAMINED. Cultivated: hort. Beyjing Medical College, Cheng 7601, 7607, 7816, 7944 (PEM). Related to 4sarum delavayi Franchet, A. porphyronotum is characterized by its urceolate perianth tubes, its very narrow orifice ring, its shortly acute anther connectives, and its leaves with persistent white patches above and purplish red coloration below. Asarum delavayi, by contrast, has larger flowers (3-4 cm long and 4-6 cm in diameter), with the perianth tubes usually constricted near the apex and then dilated, forming a collar. The specific name of the new species refers to the purplish red undersurfaces of the leaves. 22a. Asarum porphyronotum var. atrovirens C. Y. Cheng & C. S. Yang, var. nov. FIGURE 8. A varietate typica differt foliis angustioribus paribus supra atroviridibus nunquam praeditis albo-maculatis, connectivis antherarum apice haud pro- ductis. Type. Sichuan, without precise locality, Cheng 7562 (holotype. PEM: isotype, PTM). This new variety is readily separated from Asarum porphyronotum var. porphyronotum in having lustrous, deep green leaves that are never dull green or white-blotched above or purplish red below. In addition, the flowers are often smaller, and the stamen connectives are rounded. The epithet, a/rovirens, refers to the color of the leaves. 1983] CHENG & YANG, ASARUM 587 Sg Y WES 5S Figure 7. Asarum porphyronotum: a, habit of flowering plant, X .65; b, cataphyll, x .65; c, longitudinal section through perianth, showing inner surface, x 1.3: d. flower with perianth removed, showing androecium and gynoecium, x !.3; e, stamens, x 1.3; f, styles and stigmas, x 1.3. 588 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 RE 8. Asarum Dorp iyrononiin: var. atrovirens: a, habit of flowering plant, x . b, cataphyll, x 1.4: ¢. me lateral view, X 1: d. portion of perianth, showing inner surface of tube oe lobe, X I: e, flower ot perianth and androecium removed, showing gynoecium, 1.75; f, stamens, x 1.4; g, styles and stigmas, x 1.4 1983] CHENG & YANG, ASARUM 589 23. Asarum inflatum C. Y. Cheng & C. S. Yang, sp. nov. FIGURE 9. Planta tubo perianthii supra medium extus annulo crasso cincto A. maximo affinis, sed tubis perianthii valde constrictis, annulo crassiore altioreque intus longitudinaliter crassius plicato, lobis perianthii longe ovatis basi papillato- rugosis haud pulvinatis, differt. Perennial herbs with short rhizomes and numerous + fleshy roots. Leaves with petiole 7-10 cm long, the cataphylls at its base ovate, ca. 1.5 cm long, densely ciliate; blade ovate. triangular-ovate, or nearly hastate, 4-11 by 5-11 cm, acute at apex, auriculate at base with divaricate lobes, shortly pubescent on veins above, glabrous beneath. Flowers dull purple; peduncle ca. | cm long: perianth having the tube lantern shaped, with the lower portion broadly cylin- drical, ca. | em long, 1.5 cm in diameter, densely striate, and with the upper portion abruptly swollen into broad, hollow disc ca. 9 mm high, 2 cm in diameter, longitudinally ridged externally, rugosely ribbed internally, constrict- ed at apex into narrow orifice with broad, flat ring, the lobes ovate. to 3.5 by ca. 2.5 cm, with triangular, rugose, papillate area at base continuing to orifice: stamens with filament very short and anther connective obtuse; ovary half inferior, the styles 6, apices bifid and hornlike, stigmas lateral below notches. Type. Sichuan, Cang-xi, moist places in ravine and rock fissures, at 1000 to 1400 meters, 4nonymous 72N-576 (holotype, SIT). ADDITIONAL SPECIMEN EXAMINED. Sichuan: Wang-cang,. Sichuan Econ. /:xped. 4747 (si1). The flowers of Asarum inflatum, like those of A. maximum, are relatively large and prominently swollen. However, in 4. inflatum the swollen = is near the top of the perianth tube and is broad with vertical ridges. In 41. maximum, by contrast, the swollen area occurs at the center of the tube and is girdlelike, and the prominent ridges are lacking. The perianth lobes of AL inflatum are beset with rough papillae, while those of 4. maximum have only a pulvinate region. The epithet inflatum has been applied to this new species because of its swollen perianth tubes. 24. Asarum maximum Hemsley, Gard. Chron. III. 7: 422. 1890; Jour. Linn. oc. Bot. 26: 359. 1891: J. D. Hooker, Curtis’s Bot. Mag. 122: ¢. 7456. 1896. Type: Hubei, Ichang. /enry 3669 (kK, n.v., photo PE). DistRIBUTION. Hubei and eastern Sichuan. On cliff faces and in litter on forest floors: 600-800 m alt. 25. Asarum insigne Diels, Notizbl. Bot. Gart. Berlin-Dahlem 10: 855. 1930. Tyre: Guangxi, Da-yao-shan, C. Wang 39031 (isotype, sys). Asarum Seas Schmidt, oo 121. pi. 30. 1933. Type: Guang- dong, Ying-de 50476 (isotype. SYS). Asarum oe C., . a Acta a. Sinica 13(2): 19. pl. 4, figs. 3-8. 19 Type: Guangxi. Xing-an, C. fF. Liang 3416] (holotype, 1BK — DistRIBUTION. Guangdong and Guangxi. Moist, shady forests: frequent at ca. 500 m alt. 590 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 Figure 9. Asarum inflatum: a, oa of powering plant, x .62: b. cataphyll. « 1.25: eianeien a section through flower. < .62: d. flower with a removed. showing oun and gynoecium, * 135.4 e, stamen, X 1.25. 1983] CHENG & YANG, ASARUM a7] When he described Asarum long lunculatum, Schmidt stated that it dif- fered from A. insigne in having longer (to 9 cm) peduncles and obtuse anther connectives. As more material has become available, however, it has become evident that these are both variable characters and that 4. /ongepedunculatum cannot be maintained. Likewise, A. gracilipes is merely a slender form of 4. insigne, and Yang’s name is placed in the synonymy of A. insigne. 26. Asarum nanchuanense C. S. Yang & J. L. Wu.’ sp. nov. FiGure 10. Habitu species nova A. maximo similis, sed tubo perianthi extus annulatis crassis nullis, lobis perianthii basi minime pulvinatis non papillato-rugosis, foliis subtus purpureis, differt. Perennial herbs from short rhizomes and + fleshy roots. Leaves with petiole 2.5-7.5 cm long, the cataphylls at its base broadly ovate, 2 by 1.8 cm, cillate: blade cordate, 5-7 by 6-8.5 cm, the lobes at base 2-2.5 cm long. the upper surface deep green with white blotches, shortly pubescent along lateral veins, the lower surface reddish purple. ae _ os cm across; peduncle ca. 1.5 cm long; perianth with the tube cylindrical, 2 cm long, ca. 2 cm in diameter, slightly constricted, with large, naked ee ring inconspicuous, internal sur- face usually with 10 longitudinal ridges, the lobes broadly ovate, ca. 1.5 cm long. having small pulvinate areas ca. 2 mm in diameter at base; stamens with filament very short, anther connective produced into acute apex; ovary inferior, the styles 6, apically notched, the stigmas outside notches. Type. Sichuan, Nan-chuan, Cha-sha, in rock crevices in thickets, Z. L. Yang 37 (holotype, PTM). ADDITIONAL SPECIMENS EXAMINED. Sichuan: Nan-chuan, C. S. Yang 72004 (PEM, PTM). — Allied to Asarum maximum of Hubei Province, 4A. nanchuanense differs from that species in having gradually dilated perianth tubes that lack a girdlelike dilated zone, and perianth lobes with only a small pulvinate area near the base. Moreover, there is no white collar, and papillae are absent around the orifice. The specific name is derived from the type locality in Sichuan Province. 27. Asarum sagittarioides C. F. Liang, Acta Phytotax. Sinica 13(2): 23. pl. 5, figs. 7-11. 1975. Type: Guangxi, Jin-xiu, K. Y. Li 40958 (holotype, 13k). DIsTRIBUTION. Endemic to Guangxi Province. On slopes and near streams, usually under thickets; 900-1200 m alt. 28. Asarum longerhizomatosum C. F. Liang & C. S. Yang, Acta Phytotax. Sinica 13(2): 21. pl. 1, fig. 2; pl. 2, figs. 4-10. 1975. Tyre: Guangxi, Da- ming-shan, C. L. Chang O01 (isotype, PTM). DISTRIBUTION. Endemic to Guangxi. Open thickets or sunny, rocky areas. , formerly of the Beijing Chinese Traditional Medicinal College, Beijing, is a student of 3J. L. W the eee chiaceae who is collaborating with Professor Yang; he 1s now at the Sichuan Chinese Traditional Medicinal School. Ome. Gan \O in) JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Ficure 10. Asarum nanchuanense: a, habit of flowering plant. * .62:; b. cataphyll, * 1.25: ¢, portion of perianth, showing inner surface of tube and lobe, x 1.25: d. flower with perianth removed, showing androecium and gynoecium, * 1.85 1983] CHENG & YANG, ASARUM 593 29, Asarum wulingense C. F. Liang, Acta Phytotax. Sinica 13(2): 22. p/. 5, figs. 1-6. 1975. Type: Guangxi, Lung-sheng, S. L. Yu & H. F. Qin 700702 (holotype, IBK). DistTRIBUTION. Hunan, Guangdong, and Guangxi. 800-1000 m allt. This species is one of the few in Asarum that has an indumentum of soft, yellowish brown hairs. Asarum sect. Longiflora C. Y. Cheng & C. S. Yang, sect. nov. Tubis perianthii longis et infundibuliformibus non constrictis nonnunquam leviter constrictis, sine annulo ad orem, papillis remote dispositis e medio loborum ad imum tubo in seriebus verticalibus. Type species: Asarum mag- nificum Tsiang ex Cheng & Yang (Hunan). Two species and one variety in China. 30. Asarum magnificum Tsiang ex C. Y. Cheng & C. S. Yang, sp. nov. FiGuRE Il. Asarum fee Tsiang, Fl. Hupehensis 1: 208. fig. 278. 1976. Description in hinese, lacking a Latin diagnosis. Heterotropa magnifica (Tsiang) Maekawa, Jour. Jap. Bot. 57: 262. 1982. An invalid new combination based on Asarum magnificum Tsiang, Ibid. Planta floribus maximis, tubis perianthii longissimis intus sparsim longitu- dinaliter papillato-rugosis a A. petelotio similis, sed tubis perianthii erectis ad orem haud constrictis circum orem annulatis non instructis, foliis triangulari- ovatis vel oblongo-ovatis apice acutis vel breve acuminatis supra ad costam albo-maculatis, rhizomatis brevioribus, differt. Perennial herbs from short rhizomes with clustered, + fleshy roots. Leaves with petiole 6-16 cm long, the cataphylls at its base 3 or 4, ovate, ca. 1.5 cm long, densely ciliate; blade subcoriaceous, triangular- or oblong-ovate, 6-13 by 5-12 cm, acute or acuminate at apex, auriculate at base, white-blotched along midvein above (midvein often covered by short hairs), glabrous below. Flowers purplish green, large, 4-5 cm across, peduncle ca. 1.5 cm long: perianth with the tube funnelform, 3-5 cm long, 1.5 cm in diameter, not constricted (hence no definite orifice), internal surface having large papillae in vertical rows (to- ward base these often replaced by longitudinal ridges), the lobes triangular- ovate, ca. 3 by 2-2.5 cm, having purplish crescent toward middle and triangular, papillate area beneath continuous with papillae of tube; anthers with connective acute; ovary inferior, the styles 6, free, bifid at apex, the stigmas lateral. Type. Hunan, Qi-yang, in shade, Y. Tsiang & S. C. Chen 859 (holotype, 1Bsc). ADDITIONAL SPECIMENS EXAMINED, Jiangxi: Da-mao-shan, beside stream, in shade, 700 W.D. Yang 24 (Lus). Zhejiang: Chun-an, Bot. Res. Exped. 27513 (HIM). In addition to its distribution in Hunan, Jiangxi, and Zhejiang provinces, Asarum magnificum is also known to us to occur in Hubei and Guangdong 594 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Figure Il. Asarum magnificum: a, habit of flowering plant, * .65: b. cataphyll. * .65; c, portion of perianth, showing inner surface of tube and lobe, x 1: d, stamens, x 2.6; e, styles and lateral stigmas, x 2.6. 1983] CHENG & YANG, ASARUM 595 Figure 12. Asarum magnificum var. sea ee a, habit of flowering plant, < .62: b, cataphyll, x .62; c, ae ae view, ; d, portion of perianth, showing inner surface of tube and lobe, .25; e, flower sh penanth removed, showing androecium and gynoecium, X 1.25; . stamens, x 1.25; g. styles and stigmas, X 1.25. 596 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 provinces and probably also grows in Shaanxi. 4Asarum magnificum was ini- tially identified and named in the herbarium of Guangdong College of Agri- culture and Forestry by Professor Ying Tsiang, but it was never published. It is closely related to 4. pete/otii, from Yunnan southward into Indochina, and both species are characterized by their large flowers with unconstricted perianth tubes that lack an orifice ring but have large papillae on the inner surface. Asarum magnificum, however, 18 distinguished by its straight, symmetrical perianth tubes, its triangular- or oblong-ovate leaves with acute apices and white-blotched upper surfaces, and its very short, nearly perpendicular rhi- zomes. 30a. Asarum magnificum var. dinghuense C. Y. Cheng & C. S. Yang, var. nov. FiGuRE 12. A typo tubis perianthit brevioribus circum | cm longis et 8 mm in diametro et supra medium ad | cm in diametro dilatis, intus tenuiter et parce papillosis, foliis saepe ovato-oblongis ee non albo-maculatis parce breviter hirtis subtus interdum oleosipunctatis, differ Type. Guangdong, Chao-qing, Ding-hu-shan, alt. 300-700 m, in damp, shady thickets, G. Q. Ding et al. 1039 (1BSc). ADDITIONAL SPECIMENS EXAMINED. Guangdong: Ding-hu-shan, L. Teng 11066 (insc); Xin-i, C. Wang 37999 (isc). Asarum magnificum var. dinghuense differs from var. magnificum in having smaller flowers with fewer papillae on the perianth lobes, and leaves green throughout, sometimes with punctate lower surfaces. 31. Asarum petelotii Schmidt, Notizbl. Bot. Gard. Berlin-Dahlem I1: 100. 1931. Type: Vietnam, Tonkin, Péte/ot 3891 (B, n.v.) DISTRIBUTION. Southeastern Yunnan; also Vietnam. In moist, forested areas: 1100-1700 m alt. Previously unrecorded for China, this species has large flowers with the perianth tubes slightly asymmetric due to a bend to one side. ACKNOWLEDGMENTS Deep appreciation is extended to P. S. Ashton and P. F. Stevens for their encouragement in this work. We also wish to extend our sincere thanks to S. A. Spongberg for his help in reading the manuscript and in making many valuable suggestions and criticisms. Thanks are also due Wei-cheng Zung, Professor Emeritus, College of Forestry, Beying, for his illustrations of the new species. LITERATURE CITED ARAKI, Y. 1937. The species of Asarunt in the Santan District. Acta Phytotax. Geobot. 6: 122-135 1983] CHENG & YANG, ASARUM 597 1861. Index seminum Horti Botanici Berolinensis 1861. Appendix. Plan- - Braun, A. tarum novarum et minus cognitarum, quae in Horto Regio Botanico Berolinensi coluntur. 14 pp. Berlin. Hayata, B. 1915. Aristolochiaceae. Icones plantarum Formosanarum 5: 137-150 Liu, T.S., & M. J. Lat. 1976. Aristolochiaceae. /n: H. L. Li, ed., Flora of Taiwan 2: ae Maekawa, F. 1932. Alabastra diversa I. Bot. Mag. Tokyo 46: 561-586. ——. 1978. Notes on Asarum and Heterotropa from Taiwan. Jour. Jap. Bot. 53: 289-299, CY.C CS. Y. BENING CHINESE TRADITIONAL DEPARTMENT OF PHARMACEUTICAL SCIENCE BEWING MEDICAL COLLEGE BEIJING PeEopLe’s REPUBLIC OF CHINA MEDICINAL COLLEGE BEING PEOPLE'S REPUBLIC OF CHINA ZIMMERMANN & SPERRY, RHAPIS EXCELSA, IX 599 ANATOMY OF THE PALM RHAPIS EXCELSA, IX. XYLEM STRUCTURE OF THE LEAF INSERTION MARTIN H. ZIMMERMANN AND JOHN S. SPERRY STEMS OF PERENNIAL PLANTS, particularly trees, represent a considerable in- vestment in biomass. Trees can survive even under the most adverse condi- tions, but only if the hydraulic integrity of the stem 1s preserved. A very important and vulnerable part of the stem 1s the xylem. As water is pulled into the top of the tree, a period of excessive drought can drop xylem pressures to such low negative values that water columns break (cavitation). The tracheary elements in which this happens are permanently lost as functional parts unless positive pressures refill them within a very short time (perhaps hours?). We have recently found that the xylem of trees is constructed in such a way that under conditions of severe drought the hydraulic integrity of the stem 1s favored at the expense of leaves or even branches (Zimmermann, 1978). The principle of this construction works as follows: the resistance to flow in the tracheary elements is much lower in the stem than in the peripheral parts (leaves, twigs, branches). Under transpirational conditions pressures therefore drop to much lower levels in the peripheral organs than in the stem. If pressures drop to disastrously low levels. water columns break first in leaves, later in lateral axes and last in the stem. This quantitatively adjusted xylem construction 1s fefeceed to as the hydraulic architecture of the plant (Zimmermann, 1978). Although hydraulic architecture appears to be important as a mechanism for controlling cavitation in dicotyledonous and coniferous trees, such trees at least have the advantage of being able to renew lost xylem by cambial activity. In contrast, palms and other arborescent monocotyledons that have no sec- ondary growth are entirely dependent on primary vascular stem tissue through- out their entire life. Prevention of permanent embolism in the stem is therefore of paramount importance for the survival of palms. From our earlier inves- tigations of palm-stem anatomy (Zimmermann & Tomlinson, 1965), we know that the xylem that connects stem and leaves consists of very narrow tracheary elements. This indicates, at least qualitatively. that the leaf insertion represents a hydraulic bottleneck. In this paper qualitative and quantitative anatomical analysis of the stem-to-leaf vascular connection will provide the basis for dis- cussing the hydraulic architecture of Rhapis and how it may work to preserve the functionality of the stem. According to our hypothesis, the hydraulic bot- tleneck in the leaf insertion is absolutely essential for the survival of palms. In this series of papers we have used R/hapis as a model for all palms—a reasonable assumption supported by substantial comparative evidence (Zimmermann & Tomlinson, 1974). my © President and Fellows of Harvard College. 1983. Journal of the Arnold Arboretum 64: 599-609. October, 1983. 600 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 A r 2 | us a || x E a - z z || < WW 5 - o re) | 7.) 3 Ww a <| 15 = x = 9) a | | Lane I Apa Oke at Rees Ne @ +O UNO SS 26s es @ -8O,NO-- Otecat eaee yh STEM PERIPHERY mm ay a es 1983] ZIMMERMANN & SPERRY, RHAPIS EXCELSA, IX 601 MATERIALS AND METHODS Greenhouse-grown Rhapis exce/sa (Thunb.) Henry plants from the Harvard Forest in Petersham, Massachusetts, and field-grown plants from the Fairchild Tropical Garden in Miami were used in this study. The anatomy of tracheary elements in the stem-to-leaf vascular connection was studied 1n macerated material. Macerations were made from mature tissue of the peripheral stem area and the base of the leaf sheath and were examined for those elements of narrow diameter that connect stem and leaf sheath (see Ficure |, at D). The stem macerations necessarily contained both stem-to- leaf xylem and xylem from narrow peripheral axial bundles. The macerations from the leaf sheath, however, contained only xylem from the stem-to-leaf connection. From the macerated material, elements could be classified as vessel members or tracheids. Secondary-wall structure, length, and outside diameter of the elements were recorded, and many elements were photographed. Al- though inside diameter is relevant for hydraulic analysis, it is difficult to see in macerated material. The distinction between metaxylem and protoxylem in the leaf insertion was made from both structural and developmental information. We define pro- toxylem as the primary xylem that has reached maturity before the surrounding tissue has ceased elongation, and metaxylem as that maturing after elongation 1s complete, even though it may begin to differentiate much earlier. Secondary- wall structure of tracheary elernents provides only circumstantial evidence for distinguishing between protoxylem and metaxylem as thus defined. In the macerations tracheary elements with annular or helical secondary thickenings were regarded as protoxylem, and elements with pitted walls were assumed to be metaxylem. To demonstrate the distribution of metaxylem and protoxylem in the leaf insertion more conclusively, we studied the relationship of xylem maturation to leaf elongation. Rate of elongation for the sheath, petiole, and subtending stem combined was determined from repeated measurements of the distance between the insertion of the lamina on the petiole and a reference point on the mature stem. These measurements were made on the four uppermost leaves with expanded laminae. Elongation of the petiole was similarly determined from reference points marked on it in waterproof ink. Extension growth of the sheath or stem alone was impossible to measure; one cannot dissociate the extension of the leaf sheath from the extension of its associated internode because of the overlapping leaf bases. Leaf axes (sheath and petiole) or petioles that had recently ceased elongation were sectioned by hand. and the sections were stained in phloroglucino'-concentrated HCl. In a given section of this Figure |. Diagrammatic repre sentation of departure of leaf-trace bundle in Rhapis excelsa. Solid lines in curves A-B-C-A and D-E indicate presence of metaxylem vessels: dashed lines, narrow protoxylem tracheids. Transverse sections of vascular bundles ia E, right) correspond to positions A-E in curves on left; mx = metaxylem, px = pro xylem. Structure shown in diagram repeats itself axially, as indicated by 2 eae marked A 602 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 nonelongating tissue, metaxylem was indicated by immature tracheary ele- ments (those with thin and nonstaining walls). Any fully mature elements (those with thick, stained walls) in the same section must represent protoxylem. Par- tially mature elements (those with intermediate wall characteristics) in the section were assumed to be transitional between protoxylem and metaxylem. These were always located between the mature and immature elements. Vessel- and tracheid-length distributions were measured with the latex paint— infusion technique of Zimmermann & Jeje (1981). Dilute latex paint was in- jected into a cut stem for several days; paint-filled vessels were then counted with the aid of a stereo microscope on transversely cut surfaces at various dis- tances from the point of injection. Vessel-length distribution in the vessel population of the stem was calculated from these counts. In our experimental material protoxylem elements were so narrow that they could not be resolved by the stereo microscope in transverse view. We therefore prepared relatively thick transverse sections so that paint-filled elements could be counted with the compound microscope. Each transverse section was cut into ca. 2-mm- wide strips; by counting piece after piece. it was possible to keep track of which parts of the section had already been counted. Relative conductance per unit length of the xylem was determined at points along the vascular path from stem to lamina base in the following manner (see Figure 10, below). Petioles of leaves that had completed their extension were sectioned transversely from the base of the sheath to the lamina at 2-cm in- tervals. From these sections, inside diameters of all tracheary elements wider than ca. 12 um were measured (smaller-diameter elements are hydraulically insignificant). The sums of all fourth powers of inside tracheary diameters were then calculated for each transverse section to yield a figure proportional to the conductance per unit length, according to the Hagen-Poiseuille equation of laminary flow. This assumes that all conduits are ideal capillaries, which 1s not strictly true (Zimmermann, 1983). Relative conductances along the vascular path were then plotted, based on the conductance arbitrarily taken as equal to one at the point of leaf attachment. Because the xylem connection between stem and leaf is extremely complex (Zimmermann ef a/., 1982), the relative conductance value of the stem is based on the comparison of a single large metaxylem vessel of an axial bundle with a single leaf trace at the leaf insertion. This comparison is possible because we know that each leaf trace 1s ultimately continuous with a single axial bundle (see FiGure 1). We did not determine the relative conductance of the leaf traces between the base of the sheath (D in FiGure 1) and their junction with the metaxylem of axial bundles (approx- imately C to Bin Ficure |) because of the difficulty in distinguishing the total complement of these traces from other bundles in the stem. RESULTS The structure of the leaf-trace complex in the stem has been studied previ- ously (Zimmermann & Tomlinson, 1965; Zimmermann er a/., 1982) and is summarized for a single leaf-trace bundle in Figure |. A vascular bundle at A contains only metaxylem vessels (mx) and is called an axial bundle. These 1983] ZIMMERMANN & SPERRY, RHAPIS EXCELSA, IX ~ * py oan + P ¢ at ? \ isitnt Ae a Ficures 2-9. 2, transverse section of leaf trace in stem. Large metaxylem vessels (mx) continuous within ark smaller tracheids (px) connect stem with petiole (F = fibers, Ph = phloem), x 260. tips of protoxylem tracheids with helical secondary-wall thickening, x 130. 5, ae with helical secondary wall at tip and reticulate secondary ng main part of body, x 130. 6, cluster of tracheids (note matching position of their secondary walls), x 130. 7, 8, vessel element in macerate elem shown from ae ent angles: scalariform perforation plate at lower end at very tip of element, u about '4 down from tip, * 230. 9, relatively wide vessel a of leaf sheath near neces x 230. 604 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 normally containa single vessel, except where vessel ends overlap. [fone follows such an axial bundle acropetally, one eventually detects a few protoxylem elements (px). This point is located at least 10 cm below the departure of the leaf trace. At Ba few narrow px elements are present. Some of these px elements are in lateral contact with the wide mx vessel. As one moves acropetally along the bundle from B to C, the number of px elements seen in transverse section increases. This 1s true to a lesser extent of the mx vessels as well. Lateral contact between px and mx Is present to a point somewhere between B and C. At C the px is completely isolated laterally from the mx, even though both systems are within the same vascular bundle. The isolation of px from mx below C marks the beginning of the leaf trace proper. FiGure 2 shows a leaf trace in a stem transverse section corresponding to level C in FiGure |. If one follows such a leaf trace acropetally, one can see that the large metaxylem vessels (mx) continue up the stem within an axial bundle branch (A in Ficure |) and bridges (not shown in FiGureE |, but see Zimmermann & Tomlinson, 1965), and that the narrow protoxylem (px) tra- cheary elements continue as a leaf-trace bundle that enters the leaf base (at D in FiGure |) (Zimmermann et al., 1982). Thus, the leaf base is supplied by protoxylem only. This paper primarily concerns the xylem of the leaf trace in the leaf-insertion region. This is the area of the leaf trace between its beginning as isolated px in the vascular bundles of the stem (C in Figure |: FiGure 2) and its course through the basal part of the leaf sheath. The term “leaf insertion” refers to the organographic junction of leafand stem (D in FiGure |), not to the vascular junction. The vascular junction of leafand stem is the region of contact between px and mx in the vascular bundles of the stem (between B and C in FiGure 1). The xylem of the leaf trace in the leaf-insertion region contains narrow tracheary elements that suggest the presence of a hydraulic constriction for xylem sap entering the leaf. ANATOMY OF THE TRACHEARY ELEMENTS The macerated tissues from both the peripheral stem area and the lowermost part of the leaf sheath included parenchyma cells, fibers, and tracheary ele- ments—the latter the objects of our attention. Even in stem macerations one can easily recognize protoxylem elements of outgoing leaf traces by their narrow diameter and annular or helical secondary wall structure, although there may be a few additional axial elements of narrow peripheral axial bundles. In mac- erations from the basal part of the leaf sheath, we are dealing exclusively with xylem that is continuous from stem to leaf. The macerated material from both the stem periphery and the leaf sheath contained many tracheids about | mm long. The shortest one recorded was 425 wm; the longest (in the leaf base), 2.8 mm. Tracheids tended to be somewhat longer in the leaf sheath than in the stem, and their outside diameters ranged from ca. 23 toca. 40 um. Most tracheids had helical secondary walls throughout their length and were assumed to be protoxylem (Ficures 3, 4, 6). There were others, however. with helical secondary walls at their tips and reticulate walls 1983] ZIMMERMANN & SPERRY, RHAPIS EXCELSA, IX 605 in their middle sections (FIGURE 5). These tracheids are apparently transitional between metaxylem and protoxylem. No tracheids with entirely reticulate walls were found. We did find reticulate-walled vessel elements of narrow diameter, which we classified as metaxylem. Some were of narrow, long, tracheidlike shape with scalariform perforation plates (FiGURES 7, 8); others were wider. but still with scalariform perforation plates (FIGURE 9). In the stem preparations we could not be sure whether these apparent metaxylem vessels were from stem-to-leaf xylem or from narrow peripheral axial bundles. Their presence in the leaf base, however, suggested that both protoxylem and metaxylem con- tribute to the bottleneck of the leaf insertion. Extension-growth studies provided more direct evidence for the presence of metaxylem in the leaf-insertion region. As expected, elongation of the petiole ceased basipetally, and this was assumed to be the case for the sheath as well. Total elongation of the petiole, the leaf sheath, and the subtending stem ceased in the second-youngest expanded leaf. Anatomical examination of the leaf insertion of the second-youngest expanded leaf revealed narrow metaxylem elements (as indicated by their immaturity in the nonelongating tissue). Such elements are presumably destined to become those that we saw as narrow, reticulate-walled vessel members in the maceration of the mature leaf base. This metaxylem extends from the very base of the sheath into the petiole (solid line from D to E in FiGure 1). Comparative structural examination indicated that metaxylem maturation is basipetal and is not complete until the leaf is the fourth expanded one in the crown. There was no indication of metaxylem in the area of the leaf trace between the axial bundle junction and the base of the sheath (ca. C to Din FiGure 1). Length distributions of the wide metaxylem vessels of the central cylinder of the stem have been reported before (Zimmermann ef a/., 1982). In another set of latex paint-infusion experiments, we measured the length distribution of the tracheids and vessels in the peripheral stem area. We counted a total of 1843 perfused narrow conducting units (tracheids and vessels) in the stem. Counts at 1-mm intervals yielded the following length distribution: 55 percent. lengths shorter than | mm; 35 percent, 1-2 mm; and 10 percent, longer than 2 mm and up to several cm. From our maceration results 1t was obvious that the two shortest length classes (0-1 and 1-2 mm) concerned tracheids and the longer length classes concerned primarily vessels, although there are a few tracheids exceeding 2 mm in length. Length distributions of the conducting units in the sheath and petiole concerned only metaxylem vessels and were similar in both regions. Roughly 85 percent of the vessels were shorter than | cm; the remaining were mostly shorter than 5 cm, with a few as long as 9 cm. — DIMENSIONS OF THE STEM-PETIOLE CONNECTION Cursory observation revealed that there are more narrow tracheary elements at the leaf insertion than in the petiole. In a typical case the total number of tracheary elements (per transverse section) at the leaf insertion was a little over 200, but in the first 2 cm of the leaf sheath above the insertion it decreased to about half that number and then remained roughly constant along the length 606 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 yum 100 4 STEM INSIDE TRACHEARY DIAMETER wn Oo ine) lan) STEM [oa] RELATIVE CONDUCTANCE = ry EO ee ee ee 10 15 20 25 30 35 cm PETIOLE, DISTANCE FROM INSERTION Figure 10. Above: average inside diameters of all tracheary elements wider than 12 um, from insertion at stem (0 cm) along petiole (narrower ones hydraulically insignificant). Average diameter of large vessels in stem given at left for comparison. Vertical bars = standard deviations. Below: calculated relative conductances (per unit length) of xylem at points along vascular path from stem to lamina base, with conductance = | at leaf insertion (O cm). Values for petioles (including leaf sheath) from 3 different plants. Average value for stem shown on left. 1933) ZIMMERMANN & SPERRY, RHAPIS EXCELSA, IX 607 of the petiole. The diameters of the tracheary elements 1n each bundle increased from an average of 24 um at the insertion to ca. 35 wm within the petiole (FiGure 10). Absolute values of these dimensions varied from leaf to leaf. and among different shoots, but relationships within a plant were quite constant. Ficure 10 also shows the relative conductance along a typical mature stem- to-leaf connection for three leaves. Plainly, the observed bottleneck at the insertion is quantitatively significant. Conductance is highest in the stem and lowest at the insertion; it increases distally in the petiole to a value somewhat below that of the stem. The magnitude of the constriction would probably be even greater if the area of the leaf trace that is solely protoxylem (between C and D in Ficure 1) could have been measured. The bottleneck is consistently present in all leaves examined, although its magnitude varied somewhat from leaf to leaf. The meaning of this variation is unknown, but it may be correlated with leaf position. Leaves low on the stem seem to have less of a bottleneck (Sperry. unpubl.). Although based on precise measurements, FiGuURE 10 shows a somewhat theoretical result. We know that the hydraulic constriction at the leaf insertion must be greater than Figure 10 indicates. There are four reasons for this. 1) The area of the leaf trace (roughly 4 cm) between C and D in Ficure | could not be included in the calculations for technical reasons mentioned. Because this stretch is constituted wholly of protoxylem, it probably has a conductance as low as or lower than that determined for the base of the leaf sheath. 2) The region between B and C in Ficure |, where water must pass from the large metaxylem vessels to the narrow contiguous protoxylem tracheids, offers ad- ditional resistance to flow. The amount of resistance afforded by this crossover is unknown, but it may be considerable. 3) The xylem of the insertion consists primarily of tracheids, while in the stem and petiole most of the water moves through vessels. Tracheids are less efficient conductors than vessels, regardless of diameter. 4) Some of the protoxylem tracheids are certainly nonfunctional because they have been stretched and superseded by later-formed tracheids and vessels. These should not have been counted, but it was not possible to recognize nonfunctional tracheids in transverse sections. We must therefore assume that the actual hydraulic constriction at the leaf insertion is greater than is indicated by Figure 10. DISCUSSION Our observations provide evidence for a considerable hydraulic constriction in the xylem path between stem and petiole. It is caused in part by the absence of metaxylem from the area of the leaf trace between the junction of the axial bundle and the base of the sheath (the dashed line between C and D in FiGURE 1). Small metaxylem vessels, however, continue the constriction at the base of the leaf sheath (the solid line immediately above D in Ficure |). From the base of the leaf sheath to the petiole, conductance increases due to the presence of ever-wider metaxylem vessels (FiGureE 10). The fact that metaxylem con- tributes to the structure of the constriction suggests that this constriction is not 608 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 merely a consequence of a developmental constraint (the inability of the plant to make wide protoxylem), but that it is an integral part of the vascular design and has adaptive value. We suggest that this hydraulic constriction at the leaf insertion plays a critical role in confining cavitation and subsequent conduction failure to areas beyond the stem, thereby insuring the functional integrity of the stem xylem. If we assume that the likelihood of cavitation in a tracheary element increases with increasing sap tension and increasing tracheary diameter (given that the pore sizes in the tracheary walls are all the same (Zimmermann, 1983)), the quantitative anatomy presented in this paper indicates that vulnerability to cavitation is highest in the petiole, or possibly in the leaf blade. The reduced conductance at the insertion indicates a sharp pressure drop from stem to petiole when sap 1s flowing. Consequently, under transpirational conditions, xylem sap in the petiole will always be at considerably lower pressure than in the stem. Even in the absence of a constriction. xylem pressure in the petiole will be slightly lower than that in the stem, but the constriction introduces a sharp drop and thus greatly increases the pressure difference under conditions of flow. In addition to being at relatively low (negative) pressures, the sap in the petiole must flow in wide metaxylem vessels. Thus, relatively high tension is combined with wide vessels in the middle of the petiole. For this reason, we believe that 1n time of water stress, cavitations may be largely confined to the wide vessels of the petiole. Conduction in the xylem of the expendable leaf will be sacrificed to preserve the function of the xylem of the stem. The leaf-trace structure illustrated for Rhapis in FiGure | is similar to that in other palms (Zimmermann & Tomlinson, 1974). Ifthe hydraulic architecture of other palms acts as we believe it does in R/apis, it confines cavitations to the leaves. If no dependable mechanism is available to palms for the reversal of cavitation, the hydraulic constriction at the leaf base may be a necessary condition for the survival of palms. Less dramatically. it may also be respon- sible in part for observed patterns of leaf longevity in palms. Although some palms (e.g., arecoid palms with a crownshaft like Roystonea) abscise their leaves while they appear to be still healthy, when R/hapis and many other palms are grown under ideal (humid) conditions, the older leaves merely dry out. In these latter species, hydraulic architecture may be the mechanism responsible for the sequence of events. As a seedling palm grows taller, its stem must share the water supply with more and more leaves. Xylem pressures drop. Further- more, if progressively more vessels 1n the petiole cavitate as the leaf ages, older leaves will have more vapor-blocked vessels than younger ones. This will increase the resistance to flow in old leaves, which may in turn increase the likelihood of cavitation. Hydraulic constriction may therefore favor young— and discriminate against old—leaves. Xylem failure in older leaves would happen more rapidly under unfavorable (dry) conditions. This would explain the fact that while Rhapis plants grown outdoors in the sun carry only a tuft of leaves at the top of the stem, those grown in a greenhouse and watered regularly retain leaves along a much greater length of the stem. The hypothesis put forth in this paper is consistent with the available ana- tomical data and certain commonplace observations of palm growth. We hope 1983] ZIMMERMANN & SPERRY, RHAPIS EXCELSA, IX 609 that it is a promising step toward a more complete analysis of how palm xylem functions in sap conduction and palm growth. — ACKNOWLEDGMENTS We thank the staff of the Fairchild Tropical Garden for giving us access to material and laboratory space. and P. B. Tomlinson for reading the manuscript and making very helpful suggestions for its improvement. Monica Mattmuller assisted us throughout the entire project. Her help is greatly appreciated. LITERATURE CITED a M. H. ae Hydraulic architecture of some diffuse-porous trees. Canad. J. Bot. 56: 2286-2 Hera lea aici and the ascent of sap. oe series In wood science. Vol. 1. Springer-Verlag, Berlin, Heidelberg, New Y & A. A. Je 1981. Vessel-length distribution in stems of some American — plants. Canad. J. Bot. 59: 1882-189 Cue, & J. S. Sperry. 1982. “An atomy of the palm Rhapis excelsa, wae Vessel network and vessel-length distribution in the stem. J. Arnold Arbor. 63: 83-95. ToMLINSON. 1965. Anatomy of the palm Rhapis excelsa, 1. Mature vegetative axis. J. Arnold Arbor. 46: 160-180 & . 1974. Vascular patterns in palm stems: variations of the Rhapis principle. bid. 55: 402-424 CaBot FOUNDATION HARVARD UNIVERSITY PETERSHAM, MASSACHUSETTS 01366 DARWIN, TIMONIUS 611 NEW SPECIES OF TIMONIUS (RUBIACEAE) FROM PAPUASIA STEVEN P. DARWIN Timonius DC., the largest genus of tribe Guettardeae, is confined to the paleotropics from the Seychelles eastward through Malesia and tropical Aus- tralia to Micronesia and the South Pacific, with the center of diversity in New Guinea. During recent studies of the genus, several new species have been discerned from herbarium material available for examination. Although a re- visionary study of the entire genus is planned, the new species are proposed here since the monographic treatment will probably require a concerted effort of several years. I am grateful to the directors and curators of the cited herbaria for making specimens available for loan. This research was partially supported by grant DEB-8102781 from the National Science Foundation. The delimi- tation and sequence of Papuasian geographic regions employed in this study follow those of Coode (Brunonia 1: 131-302. 1978). Timonius zuckianus S. Darwin, sp. nov. Ex affinitate speciebus foliis paxillatis praesertim T. longifolii Valeton sed floribus longioribus, fructibus magnis, foliorum laminis apice abrupte acute acuminatis et cymis glabris differt. Shrub or tree to 20 m tall. Branchlets 2-5 mm broad toward apex, compressed but becoming terete, gray, glabrous except puberulent above stipule scars; stipules imbricate, lanceolate, acuminate, to 130 by 15 mm, glabrous without, glabrous within except for slender colleters toward base. Leaves with petiole stout, semi- to sub-terete and somewhat winged by leaf blade, 12-55 mm long, ca. 2mm broad at middle, glabrous; blade elliptic, 7-13 by 3-7.5 cm, abruptly and sharply acuminate at apex, obtuse at base or decurrent onto petiole, co- riaceous, glabrous on both surfaces, the costa plane or canaliculate above, prominent beneath, the lateral nerves indistinct, the veinlets paxillate, pro- minulous-striate above, prominulous or subplane beneath, the domatia resem- bling small pits, glabrous or minutely puberulent, dispersed over lower surface. Staminate cymes 5- to 7-flowered, at anthesis to 3 by 3 cm: peduncles stout, compressed, 6-10 by up to 3 mm, glabrous; bracts none; calyx tube cupular, truncate to undulate, (I-)2-3 mm high, glabrous, the basal portion (aborted ovary) to 1.5 mm high; corolla narrowly infundibular, the tube 8-12 by ca. 1.5 mm, glabrous inside and out, the limb 12-18 mm broad, the lobes 4, ovate, 6-8 by 3-4 mm, acute, somewhat irregular at margin; stamens 4, the anthers 5-6 by ca. 0.5 mm, glabrous, included. Pistillate cymes 2- or 3- (or 4-)flowered, ) President and Fellows of Harvard College, 1983. oe of the Arnold Arboretum 64: 611-618. Co 1983. 612 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 at anthesis to 3.5 by 2 cm, the peduncles stout, compressed, 8-18 by up to 3 mm, glabrous; bracts none; calyx tube cupular, truncate to undulate, 0.5-1 mm high, glabrous; corolla subhypocrateriform, the tube 5-7 by ca. 1.5 mm, gla- brous inside and out, the limb to 12 mm broad, the lobes 6 to 8, ovate. to 5 y ca. 1.5 mm, acuminate, somewhat irregular at margin: staminodia 6, ca. 3 by 0.5 mm, glabrous, included: ovary and disc glabrous, the style glabrous, unequally 6-branched, slightly exserted. Fruits short-cylindric to depressed- globose. 6-sulcate, 4-5 by 6-7 mm, glabrous, crowned by persistent calyx; pyrenes 35 to 45, vertically oriented and arranged in 6 radiating double files. Type. Normandy Island, Mt. Pabinama, 820 m alt., 2 May 1956, Brass 25653 2 (holotype, A!; isotypes, CANB!, LAE!). easel TION. Islands of extreme southeastern Milne Bay Province of Papua New Gui ECOLOGY AND FIELD NOTES. Common in lower montane forests and stunted mossy forests of ridge-crests from 700 to 1700 m alt. Ranging in habit from a shrub or small tree of about 2 m to a canopy tree 20 m tall. Bark rough, dark brown; wood white to pale straw-colored. Probably flowering and fruiting throughout the year. The pistillate and staminate flowers are reported as white, the fruits as black and fleshy when mature. SPECIMENS SEEN. Papua. PAPUAN ISLANDS: Goodenough Is., E slopes, Brass 24525 @ (A, CANB, LAE); Fergusson Is., Mt. Kilkerran, E slopes Digeenee, LAF 71045 4 (A, BISH, LAF); Tagula Is., Mt. Riu, W slopes, Brass 27839 2 (a, LAE), Mt. Emuwa, LAE 745209 (BRI, CANB, LAE); Rossel Is., Kwa mountain above ae NGF 27077 @ (A, BRI, CANB, LAE), Mt. Rossel, summit, Brass 28458 2 (A, BO, LAE), Mt. Rossel, S slopes, Brass 28389 3 (A, BO, LAE). Timonius zuckianus is allied to the other species of the genus in which lateral or secondary nerves are only weakly developed, with the paxillate veinlets forming a Striate pattern on the upper leaf surfaces in herbarium specimens. In addition, the calyces are truncate and stipules are imbricate in bud. The present species 1S unusual in its exceptionally large flowers and more promt- nently acuminate leaf apices. It resembles 7. /ongifolius Valeton, from the Sepik area of New Guinea, in its cvmose pistillate inflorescences (in all other species of this alliance the pistillate flowers are solitary), but its larger fruits, glabrous inflorescences, and elliptic rather than oblanceolate leaves suggest 7. singularis (F. Mueller) L. S. Sm., from Australia take pleasure in naming this distinctive specks for Dr. and Mrs. Robert K. Zuck, formerly of Department of Botany, Drew University, who inspired many college students to pursue botanical careers. — > Timonius pubistipulus S$. Darwin, sp. nov. Ex affinitate speciebus foliis paxillatis praesertim T. belensis Merr. & Perry sed stipulis intus minimum in centro vel basin versus dense pallidis sericeis differt. Epiphytic shrub or terrestrial tree to 35 m tall. Branchlets 3-4 mm broad 1983] DARWIN, TIMONIUS 613 toward apex, compressed but becoming terete, either glabrous except puber- ulent above stipule scars or conspicuously pubescent: stipules imbricate, lan- ceolate, to 80 by 15 mm, long-acuminate, glabrous without, toward center or base densely pale-sericeous within. Leaves with petiole stout, semi- to s terete, 4-20 mm long, |.5-—2 mm broad at middle, glabrous; blade elliptic to obovate, 4-12 by 1.5-5.5 cm, acute to short-acuminate at apex, acute to subat- tenuate at base, rigidly coriaceous, glabrous above and beneath, the costa can- aliculate above, prominent beneath, the lateral nerves hardly distinguishable, the veinlets paxillate, subplane to prominulous on both surfaces, the domatia resembling small pits, glabrous to minutely puberulent. Staminate cymes 3- ie 5-flowered, in bud ca. 1.5 by | cm: peduncles stout, subcompressed, 3-5 1.5-2 mm, glabrous; bracts none; calyx tube cupular, truncate to cee obtuse-denticulate, 0.5-2 mm high, glabrous; corolla narrowly infundibular. the tube to 8 by ca. 1.5 mm. glabrous inside and out, the limb to 5 mm broad, the lobes 4, lanceolate, ca. 3.5 by | mm, sharply acuminate; stamens 4, the anthers ca. 6 mm ne glabrous. Pistillate flowers solitary; peduncle com- pressed, 1-5 by 1-2 mm, in fruit to 12 mm long; glabrous: bracts none; calyx tube cupular to eoindrical: truncate to undulate, 0.5-1 mm high, glabrous; corolla broadly infundibular, the tube (3-)5-7 by 2—2.5 mm, sometimes some- what broader in throat. glabrous inside and out, the limb to 8 mm broad, the lobes 6 or 8, lanceolate, 2.4—3.5 by 1-1.5 mm, conspicuously acuminate; stam- inodia 6, ca. 2 mm long, glabrous, included; ovary subglobose, 3-3.5 by 3-4 mm, glabrous like the disc, the style 6- to 8-branched, subexserted. Fruits subglobose to broadly obovoid and somewhat laterally compressed (when dried appearing distally sulcate and proximally tuberculate), 6-9 by 7-10 mm, gla- brous, crowned by persistent calyx; pyrenes 30 to 45 in transverse section, ca. 10 in vertical section, probably up to 60 per fruit, subvertically oriented. I am readily able to distinguish two varieties within this species. Timonius pubistipulus var. pubistipulus T. avenis sensu Merr. & Perry. Jour. Arnold Arb. 26: 235. 1945, pro parte, non Valeton. T. carstensensis sensu Merr. & Perry. Jour. Arnold Arb. 26: 238. 1945, pro parte, non ernham Type. Papua New Guinea, Western Highlands Province, Min) River ne ay above the Nona River camp, in Nothofagus forest at 9000 ft alt., 30 July 1957. Pullen 268 2 (holotype, Al; isotypes, BISH!, BM!, BO!, BRI), CANB!, G!, oi DistRIBUTION. Interior of North East New Guinea and Papua. ECOLOGY AND FIELD NoTES. In montane 4Araucaria, Nothofagus, and moss- forests or swamps, 350-3400 m alt.: most collections from above 1500 m alt. Epiphytic shrubs, “climbers,” or trees to 30 m tall with trunk diameters to 0.5 m, frequently in forest edges, burnt-over areas, and other open habitats. The bark surface varies from gray and smooth to brownish and pustular or scaly; the wood is white to reddish straw-colored; the stipules frequently reddish and conspicuous. Staminate and pistillate flowers vary in color from white to green- 614 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 ish, yellowish, or rarely pinkish, with the corolla lobes often paler than the tube; mature fruits are dark red to black. Flowering and fruiting throughout the year. VERNACULAR NAMES. “Komal” (Kiliga), “Korma” (Enga), “Kuma” (Hagen, Kiliga, Togoba), ““Kume”’ (Mend), ““~Kumeh” (Onim), or “Ome” (Mendi); “Haikapula” (Jalibu); ““Abalan” (Enga): ““Moghon” (Nerenavip); “Sikil” (To- goba): * oa Bala Iara Sana sauna i a aa “Tombumpf.” or ““Tomburump” (Minj):; ““Yambasi” (Fiyugi). SPECIMENS SEEN. North East New Guinea. West Sepik: S of Nerenavip, Hindenburg Range, NGF 32144 2 (A, BRI, CANB, we NGF 32181 8 (A, BRI, CANB, LAE). MADANG: S slopes Finisterre Range to S of Mt. Abilala. Pullen 51? @ (CANB, L. LAE). MOROBE: Mannasat, Cromwell Mts., Pd 9600 @ (A, BRI, CANB, ee Sarawaket, Clemens 5758 @ (A); summit of Mt. Shongul, NGF 37434 4 (Ae); New Yamap, NGF 27552 2 (A, BRI, CANB, LAE, SING); Aseki Road, ca. 57 km from Bulolo, LAk 75226 2 (LAE, UPNG); oo ridge near Haumuga, Asceki area. Schodde & Craven 4743 2 (A, BRI, CANB, L. AE); Edie Creek road above Wau, NGF /3489 2 (pri, CANB, LAE); Bulldog Road ca. 20 i S of Wau, Fallen 550 (LAeE); above Bakaia, ca. 15 mi SW of Garaina, Hartley 12716 2 (A, BRI, CANB, L, LAE). WESTERN HIGHLANDS: Sirunki, Marapamanda, E of Laigam and Yamara, ANU 762 2 (A, CANB, LAE); N slopes of Sugarloaf complex along Wap R.. Hoogland & Schodde 7089 2 (A, BM, BRI, CANB, LAE, Zz). ca. 5 mi N of Londoli, Ambum Valley on Wabag—Maramuni Road, ee oes ANB, LAE); River Kum, Mt. Kum, LAE 50320 2 (A, CANB, LAE); Tomba, NGF 3551/2 6 & 2 (LAE); top of Mt. Oga. Saunders 677 (A, BM, LAE): Slopes of Mt. Kinkain, ke ee Saunders 741 (CANB); upper Nona R., Saunders 747 (A, BM, BRI, CANB, LAE), Saunders 748 (CANB), Minj-Nona Divide, N side of Kubor Range, Pullen 5167 ° (A, BRI, CANB, LAE); Kubor Ranges above Kuli, Robbins 524 2 (LAE). EASTERN HIGHLANDS: 2 mi SW of Koge Mission, Sinasina, Hide 309 2 (LAE), Hide 486 (LAF), Papua. SouTHERN HIGHLANDS: Mimimbipl. 20 mi from Mendi, Tambul Road, LAF 5532] 2 (A, BIsH, LAE); W slope of Mt. Giluwe above Klareg, Schodde 2015 2 (A, BRI, CANB, LAE); SW slope of Mt. Giluwe, ls 1673 6 (A, CANB, LAE); base of Mt. Giluwe, LAE 74437 @ (BRI, CANB, LAE); slop Mt. Giluwe, NGF 32502 2 (A, BRI, CANB, LAE, SING); Omin, ANU’ 20326 oe near neice Mission sawmill, N side of Pangia Road 4 mi E of Jalibu, UPNG 4224 2° (UPNG). CENTRAL: E of Mt. Tafa, Brass 4077 @ (pri); Mt. Tala, Brass 5046 9 cee ae road from Woitape to Kosipi, lowasi Swamp, NGF 20308 2 (A, BO, BRI, CANB, LA NG); E slope Lake Myola no. 2, NGF 34629 @ (A, BISH, LAE); Koitaki, Carr 12522 : in . BM, CANB, SING); Boridi, Carr 13095 2 (A, CANB, SING), Carr 13343 2 (A, BM, CANB, SING), Carr 14607 & (A, BM, CANB); Lala Valley, Carr 15856 (BM, L, SING); The Gap, Carr 15042 8 (A, BM, SING); above The Gap, Carr 138199 (A, BM, CANB, K); Sogeri region, Forbes 770 2 (A, BM, L). NORTHERN: Kokoda, Carr 16453 2 (B, CANB), Carr 16478 8 (CANB, G). MILNE Bay: junction Ugat and Mayu rivers near Mayu 1, NGF 28897 2 (A, BISH, CANB, LAE, SING); Mayu 2, Mt. Suckling, LAE 54117 2 (a, CANB, LAE); Goropu Mts.. Pumpunipon to S ridge of Goe Dendeniwa, Veldkamp & Stevens 5694 6 (BisH, L, LAr); rack, S spur to Pumpunipon, Mt. Suckling complex, LAF 54178 2 (CANB, LAE); N slopes of Mt. Dayman, Maneau Range. Brass 2 (A, CANB, LAE), Brass 22702 2 (A, CANB, LAE), Brass 22874 2 (A, CANB, LAE); Mt. Day man, a ke 1090 2 (CANB): Mt. Wadimana, NE outlier a Mt. Simpson, Pullen 7842 (A, LAE); summit area, Mt. Wadimana, Schodde 5472 6 (CANB, L, LAE); Aparamu Ridge, E oe Mt. Simpson Range, Schodde 5519 (CANB). Timonius pubistipulus var. pubescens S. Darwin, var. nov. Omnio idem ac varietas typica sed ramuli pubescentia conspicua et plus minusve persistent! differt. 1983] DARWIN, TIMONIUS 615 Type. Papua New Guinea, Morobe Province, Mt. Kaindi near Edie Creek, heavy liana in a Nothofagus a fie ft alt.. 9 May 1963. Hartley 11840 ° (holotype, A!; isotypes, c!, CANB!, G!, L!, LAE!) DISTRIBUTION. Interior of eastern West New Guinea, North East New Guinea. and Papua. ECOLOGY AND FIELD NOTES. Montane forests. 1500-2600 m alt.; mainly in swamps, myrtaceous scrub, edges of trails, ridge-tops, old landslides, and other open or successional habitats. Varying in habit from epiphytic shrubs to lianas or slender trees to 25 m tall with trunk diameters to 25 cm. The bark surface is pale gray to dark brown and smooth, the wood white to pale brown. The flowers are yellowish or brownish white, rarely dull pinkish, and the fruits are maroon to black when mature. Flowering and fruiting throughout the year. VERNACULAR NAME. “Komat” (Enga). SPECIMENS SEEN. West New Guinea. SNow Mountains: Byenkorfbivak, Helling Hell- wigberg, Pulle 773 (A, L). North East New Guinea. West Se PIR: wee second bush camp below Tamanagabip on track to Busiliin. LAE 59483 2 (A, BISH, LAE). Morost: NE slope of Mt. Rawlinson, /oogland & Craven ee 2 (A, BM, BRI, CANB, K, LAE); Ogeramnang, Clemens 4649 2 (A, B): Aseki- Pee Divide, LAE 54772 2 (A, BISH, CANB, LAE); E slope Spreader Divide. ca. 7 mi NW of Aseki, Schodde & Craven 4904 2 (A, BRI, CANB, LAE); near Wengomanga, via Oiwa, Cie & Schodde 1291 2 (CANB, LAE): along water race of N. G. Mine, NGF 21063 2 (A, BISH, BRI, CANB, LAF, SING): Edie Creek, NGF 19948 (CANB, LAE), NGF 19959 9 (A, BISH, BO, BRI, CANB, LAE, SING), Hartley 11725 8 (A, BRI, CANB. LAE); Mt. Kaindi. SE slope, ene es 195 2 (LAE); track descending SE from S summit of Mt. Kaindi. UPNG 31774 (uPNG): 4 km WSW of Wau. Mt. Kaindi. Allison NG-171 2 (LAe); Mt. Kaindi summit, NCE ae 2 (A, BO, BRI, CANB, K, LAE): Kaindi, Brass 29559 2 (A, BO, CANB, LAE), Brass 29561 6 (A, CANB, LAE); ca. 15 km SSW of Wau, Bulldog Road, WE7/ 13742 (1 = Bulldog Road, 12 km SSE of Wau, Pratt NG- 105 @ (LAE); Bulldog Road ca. 13 km SSE of Wau, Allison NG-16/] 2 (LAr); Lake Trist, NGF 29735 2 (a4, BRI, CANB, LAE, SING); 5 km E of Lake Trist, Payymans 1568 2 (CANB. LAE). WESTERN HIGHLANDS: top of ridge 3 mi S of Kompiam, ANU 2805 2 (A, CANB. &. LAE). EASTERN HIGHLANDS: confluent Warapuri and Kori rivers, Wahgi-Jimmy Divide, - Sonera NGF 18173 2 (BRI, CANB, LAE); summit Mt. Suen Jermy 5025 2 (LAE). Pap THERN HIGHLANDS: Onim, Raw 160 2 (LAE). WESTERN: Mt. Bosavi, N side, Jacobs $8599 (BISH, L. LAE). NORTHERN: E side of Lake Myola no. a LAE 61972 4 (BISH, BO, CANB, LAE, MO, SING). Timonius pubistipulus, clearly allied to other paxillate-veined species of the genus, is most closely related to 7. belensis Merr. & Perry, with which it 1s sympatric in the western part of its range. However, the present species 1s readily distinguishable from all other “paxillate” species by its stipules, which are conspicuously and densely pale-sericeous on the inner (adaxial) surface. Although the extent of the stipular pubescence varies, | have had no difficulty assigning specimens here. Most pistillate material can be recognized as be- longing to 7. pubistipulus on the basis of the subvertically arranged pyrenes: these are intermediate between the very oblique pyrenes of 7. be/ensis and the strictly vertical ones of such other species as 7. avenis Valeton. Timonius pubistipulus var. pubescens differs from the typical variety in its branchlets that are usually conspicuously brown-pubescent, at least when young. 616 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 rather than glabrous. This is a feature shared with 7° t(richanthus Merr. & Perry, although that species has pubescence extending to the outer surfaces of stipules and lower surfaces of leaf blades, as well as other distinguishing characters. On the basis of available collections, 1t appears that var. pubescens extends farther west but not so far east as var. pubistipulus, but the varieties are sympatric through most of their range and occupy similar habitats. Timonius subavenis (Valeton) S. Darwin, comb. nov. Timonius scaber Valeton var. subavenis Valeton, Bot. Jahrb. 61: 51. 1927. Epiphytic shrub, or at maturity small tree. Branchlets 2-3 mm broad toward apex, compressed but becoming terete. pubescent with scattered, spreading, brown hairs; stipules imbricate. lanceolate, to 45 by 10 mm, acuminate, abaxial- ly hirsute toward base and center with somewhat shaggy brownish hairs but usually glabrous toward margin, adaxially glabrous except for scattered colle- ters. Leaves with petiole slender, 5-15 mm long, |-2 mm broad at middle, hirsute; blade long-elliptic to oblanceolate, 4-14 by 3-7.5 cm, at apex long- acuminate to sharply acute. at base cuneate and decurrent, thinly coriaceous to chartaceous, above short-pubescent when young, soon glabrous except base of costa, beneath sparsely to densely appressed or spreading brown-pubescent, the costa plane to canaliculate above, prominent beneath, the lateral nerves obscure above, submerged beneath, the domatia absent. Staminate cymes 3- to | 1-flowered, at anthesis 1.5-3 by 1.5-2.5 cm: peduncles 2-8 by ca. 1.5 mm, densely tomentose to subglabrous; bracts absent; calyx cupular, truncate, the limb about 0.5 mm high, glabrous, the basal portion (abortive ovary) about | mm high, pubescent but at very base glabrous and constricted; corolla narrowly infundibular, the tube to 10 mm long (late bud). densely hirsute outside with distally pointing hairs, probably glabrous inside. the lobes 4. Pistillate flowers solitary; peduncle to 5 by ca. 2 mm, puberulent with scattered hairs to glabrous; bracts not seen: calyx tube cupular, truncate, about 0.5 mm high, glabrous; corolla narrowly infundibular, the tube 6-7 mm long, to 1.5 mm broad at middle, hirsute outside, probably glabrous inside, the limb 5-6 mm broad, the lobes 6, to 2 mm by ca. | mm, acute; ovary subglobose, 3-3.5 by 3-3.5 mm, pubescent with scattered long hairs to glabrous. Fruits subglobose, somewhat laterally compressed, 4—5 by 3.5-6 mm. 6- to 8-ribbed, pubescent like ovary, crowned by persistent calyx and disc; pyrenes to 45 per fruit, vertically oriented, arranged in 6 to 8 radiating double files corresponding to ribs. —_— Tyee. The identity of Valeton’s species Timontius scaber is as yet uncertain since I have not located any material of Ledermann 13086 (pistillate), the only collection cited (Bot. Jahrb. 61: 50. 1927). Variety subavenis would seem even more mysterious since no collection at all was cited, but one staminate spec- imen, Ledermann $793 (SING!). 1s annotated ““Timonius scaber var. subavenis n. sp.” ona Berlin Museum label. possibly in Valeton’s hand. This I designate the lectotype of 7. scaber var. subavenis, here elevated to specific rank. The lectotype agrees well with Valeton’s unusually brief description. It bears only the locality “Sepik-Gebiet.” but was probably collected in the vicinity of Malu and April River (Fl. Males. 1: 317. 1950). 1983] DARWIN, TIMONIUS GOL? ECOLOGY AND FIELD NOTES. Ranging in habit from an epiphyte to a liana, shrub, or small tree. The few available collections were made in primary forests and swamp forests at 50-1750 m alt. Staminate and pistillate flowers are white to reddish. Flowering and fruiting throughout the year. SPECIMENS SEEN. West New Guinea. Svow Mountains: Byenkorfbivak, Zuidhelling Hell- wigberg, Pulle 6504 (A, BO, L, SING); Pesechem-vallei, Pulle 1155 6 & 2 (Bo, L). DIAJAPURA: Rouffaer R., Docters van Leeuwen 10141 2 (A, Bo, K. L), Docters van Leeuwen 10172 6 (Bo, K, L); Bodem R., 60 km SE of Sarmi, BW 8/0] @ (aR. CANB, LAE); Bernhard bivouac, Mever Drees 499 6 (A, BO, K, L, SING), Meyer Drees 607 2 (Bo, L). Timonius subavenis is similar to 7. trichanthus Merr. & Perry but occupies a more westerly range. As in 7. trichanthus, stipules and corollas (staminate and pistillate) are pubescent outside, the fruits are sulcate with the pyrenes arranged vertically, and the staminate inflorescences have fewer than 15 flowers. The differentiating characters lie mainly in the calyces: in 7. (richanthus the basal portion (ovary or abortive ovary) 1s glabrous or nearly so (vs. pubescent in 7. subavenis) while the calyx tube is pubescent and | mm or more high (vs. glabrous and about 0.5 mm high in 7. swbavenis). In addition, the pistillate peduncles are somewhat longer in 7. trichanthus. Timonius paiawensis S. Darwin, sp. nov. Ex affinitate T. laevigati Valeton sed foliorum nervis lateralibus utrinque 5- 7 supra impressis infra prominentibus differt. Shrub or slender tree to 8 m tall. Branchlets 3-5 mm broad toward apex, compressed but becoming terete. yellow-puberulent; stipules valvate, deltoid, to 15 by 5mm, acute, minutely and densely puberulent without, densely yellow- hirsute within. Leaves with petiole rigid, semiterete, 5-20 by 1-3 mm, glabrous to subpuberulent; blade narrowly elliptic to suborbicular, 7-18 by 2.5-12 cm, gradually or abruptly sharp-acuminate at apex, narrowly to broadly cuneate at base, subcoriaceous, glabrous and glossy above (dark when dried), glabrous or with small hairs scattered over lamina and frequently denser on costa and nerves beneath, the costa plane to canaliculate above, elevated beneath, the lateral nerves spreading, 5 to 7 on each side of costa, impressed above, prom- inent beneath, the veinlets subimpressed above, reticulate, the domatia incon- spicuous in axils of lateral nerves. Staminate cymes to 20-flowered, at anthesis ca. 2 by 2-3 cm; peduncles stout, 3-15 by ca. 2 mm, yellow- to brown-pu- berulent; bracts broadly clasping, to 3 by 5 mm, rounded at apex, pubescent, caducous, probably a larger, caducous, opposite pair enclosing young inflores- cence (not seen); calyx narrowly ellipsoid to ovoid, in bud 4-7 mm high, the limb irregularly 5-toothed, often rupturing laterally; corolla (seen in bud) yel- low- to red-puberulent, the lobes 5. Pistillate flowers solitary; peduncle 4-15 (-22) by ca. 2 mm, yellow-puberulent: bracts connate, calyptrate, to 8 mm long, puberulent, caducous, affixed beneath ovary; calyx tube ca. 2 (3-5 in fruit) mm high, distally contracted and confluent with top of ovary, 6-toothed or somewhat erose; corolla narrowly infundibular, the tube ca. 8 by 2 mm, pale sericeous without, probably glabrous within, the limb ca. 10 mm broad, the lobes 6, narrowly lanceolate, ca. 4 by | mm, acute; style pubescent. Fruits 618 JOURNAL OF THE ARNOLD ARBORETUM [vVoL. 64 subglobose to aa 9-14 by 9-14 mm, densely yellow-puberulent, crowned by persistent calyx; pyrenes ca. 25 in transverse section, ca. 18 mm long. obliquely er in reddish pulp. surrounded by firm mesocarp 1-2 mm thick. Type. Papua New Guinea, Morobe Province, 10 mi up Paiawa River, low shrub or small tree, 20 October 1965, NGF 25030 @ (holotype, A!; isotypes, BISH!, BRI!, CANB!, LAE!, SING). ECOLOGY AND FIELD NOTES. Shrubs or slender trees apparently confined to riverine habitats, from sea level to 600 m alt. The bark surface is reported as ray to brown, smooth or somewhat vertically fissured; the wood is white to brown. The flowers are noted as white to greenish or brownish, and the fruits are reddish when mature. Flowering and fruiting between May and October. SPECIMENS SEEN. North East New Guinea. Morose: Baden Bay, Buso, NGF 39097 6 & 2 (A, BRI, CANB, LAE); Buso, NGF 47767 @ (A, BISH, BO, BRI, CANB, LAE); near Buso, growing on Buso R., LAK 52328 6 & 2 (A, BISH, LAE), Conn et al. 301 2 (A): ee Forestry Camp. Conn 313 2 (A, CANB); Paiawa Valley, NGF 47318 @ (Bisu. BO, BRI, CANB, GH, LAE), NGF 47326 4 (Lae); Saru R., 7 mi SE of Garaina, NGF 47993 2 (A, BISH, BO, BRI, CANB, LAE). 7imonius paiawensis 1s similar to 7. laevigatus Valeton in its coriaceous leaves, vellow-pubescent branchlets, deltoid, valvate stipules, and denticulate calyces that in pistillate specimens are more or less confluent with the top of the ovary in the young bud. The leaves of 7imonius laevigatus, however, have more numerous lateral nerves that are neither impressed above nor strongly elevated beneath. Furthermore, the two species have very different geographic ranges: 7. /aevigatus extends from the Sepik region of Papua New Guinea westward, while 7. pdiawensis seems confined to Morobe Province. DEPARTMENT OF BIOLOGY TULANE UNIVERSITY New OrLEANS, LOUISIANA 7O118 1983] KIEW, CHIONANTHUS 619 TWO UNUSUAL CHIONANTHUS SPECIES FROM BORNEO AND THE POSITION OF MYXOPYRUM IN THE OLEACEAE RUTH KieEw CHIONANTHUS Linnaeus (including Linociera Sw.—Stearn, 1976), with about 70 species described, is the largest genus of the Oleaceae in Malesia but the least well known. As study of the genus progresses, it is possible to appreciate the range of variation within the genus; for example, study of the Malayan species shows that there are intermediate species with characters of both sec- tions CERANTHUS Bentham & Hooker and EuLinocierA Bentham & Hooker (Kiew, 1979), and investigation of the Bornean species reveals a much wider range in inflorescence and fruit form than was previously described (Kiew, 1980). THE PROBLEM OF LINOCIERA MACROBOTRYS In the revision of the Sarawak species (Kiew, 1980), specimens of Linociera macrobotrys Merr. were not available for study. On the grounds of two unusual characters for the genus—its habit (it is a shrub 2 m tall) and its terminal inflorescence—the species was excluded from the genus with a note that its description closely matched that of MWyxopyrum enerve Van Steenis. In the Kew and Leiden herbaria, specimens identified as L. macrobotrys are in fact the small-leaved form of 7. enerve. The type of Linociera macrobotrys was subsequently traced to the herbarium of the Arnold Arboretum. Examination showed that the inflorescence 1s not terminal but axillary, and that the flowers are all staminate (a character not mentioned in Merrill’s description (1918)). The tendency toward unisexual flowers is seen in several Malesian genera of the tribe Oleeae Bentham & Hooker. Olea L. has polygamous species. Osman- thus Lour. generally has unisexual flowers (Green, 1958), although the Malesian species, O. scortechinii King, is hermaphrodite. The vast majority of Chionan- thus species have hermaphrodite flowers, but polygamous species do exist (for xample, the well-known Chinese fringe tree, Chionanthus retusus Lindley & Paxton, and C. enerve (see below)). There are also unisexual species such as Linociera oxycarpa Lingelsh. and L. rupicola Lingelsh., both from New Guinea, and C. macrobotrys, as yet known only from a staminate specimen from Sa- rawak. Lingelsheim (1927), in his study of the Oleaceae in New Guinea, mentioned that the polygamous and unisexual Chionanthus species, in which the filament is fused to the corolla, could be distinguished from those of Olea, in which the © President and Fellows of Harvard College. 1983 Journal of the Arnold Arboretum 64: 619-626. Seober 1983. 620 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 filament is free. In all Malesian species of Chionanthus investigated so far (both those with hermaphrodite and those with staminate flowers), the filament is always short and is fused near the base of the corolla. The connective varies in width (FiGureE |, d-1) and is apiculate in some species (FiGure |, g). With the exception of O. paniculata R. Br. (FiGuRE 1, c), in which the filament is not only free from the corolla but also versatile, the Malesian Olea species have similar stamens with short filaments fused to the corolla (FiGure I, a, b). Staminal characters can therefore not be used as a generic distinction be- tween Chionanthus and Olea Chionanthus macrobotrys has staminate flowers and oblong corolla lobes with a rounded apex, which are also found in the Olea species in Borneo, O. decussata (Heine) Kiew and Olea javanica (Blume) Knoblauch. Chionanthus macrobotrys is distinguished from these Olea species by the absence of the rounded bud characteristic of Olea and by the corolla that is divided more than halfway. These characters confirm this taxon as a species of Chionanthus. Chionanthus macrobotrys (Merr.) Kiew, comb. nov. oo macrobotrys Merr. Philipp. J. Sci. 13: 117. 1918. Type: Sarawak, Ist Di- sion, Mount Poe [Gunung Pueh], Foxworthy 369 (A). Shrub 2 m tall. Twigs terete, brown becoming grayish with age, minutely hirsute. Leaves with petiole 5 mm long; blade lanceolate, 5—5.5 by 2-2.5 cm, acute at apex, entire and recurved at margin, cuneate at base, coriaceous, minutely punctate below, glabrous: midrib impressed above, with lateral veins ca. 4 pairs, obscure above and below. Inflorescence a lax panicle with fourth- order branching, axillary, minutely hirsute, 14.5 cm long, lowest second-order branches 9 cm long; bracts oval, 2-3 mm long. Staminate flowers yellowish, 2 mm long; pedicel 2-3 mm long: calyx 0.5 mm long, deeply divided to base, lobes acute, glabrous; corolla 2 mm long, lobes | mm long, induplicate-oblong with rounded apex, joined in pairs for | mm at base; stamens 2, connate to base of corolla, the anthers almost sessile, to | mm long, connective ca. 0.5 mm wide. Ovary absent. Pistillate or hermaphrodite flowers and fruits un- known. This description differs in two important respects from that of Merrill: the inflorescence 1s axillary (not terminal as Merrill reported), and the flowers are staminate. The species 1s known from a single locality, Gunung Pueh, and Merrill cited two specimens, Foxworthy 369 (type) and 290; I have not been able to locate the latter. On the herbarium label Foxworthy noted that the plant grows in forests on the upper slopes of the mountain between 1500 and 1700 m. He gave “barungian batu”’ as the Dyak name Among the Bornean Chionanthus species, C. panoroins most resembles C. ramiflorus Roxb. in its rounded corolla lobes (FiGureE 1, e, h) and in its lax, much-branched panicle. However, the flowers of C. ramiflorus are her- maphrodite and the leaves are larger. Merrill (1918) considered the leaves of C. macrobotrys to be unusual in their small size and coriaceousness, but those of the montane form of C. /axiflorus Blume and the small-leaved form of C. enerve are very similar. Of the Bornean species other than C. macrobotrys, only C. longipetalus (Merr.) Kiew 1s a shrub. 1983] KIEW, CHIONANTHUS 621 Ficure 1. Stamen attachment in staminate and hermaphrodite flowers of Olea and Ghee is (including Linociera): a, Olea decussata (staminate, Chew 939); b, O. bra- ies (hermaphrodite. Ahmad s.n.); c, O. paniculata (hermaphrodite, Schodde & Craven 61), d, Chionanthus porcatus (hermaphrodite, Anderson 20044), e, C. ramiflorus (her- aoe Van Royen & Sleumer 6939), f, Linociera oxycarpa (staminate, NGF 42633 g L. rupicola (staminate, TGH 13667), h, C. macrobotrys (staminate, Foxworthy 369). i, C. enerve (staminate, Collenette 905). — THE PROBLEM OF MYXOPYRUM ENERVE The matching descriptions of Linociera macrobotrys (which is a species of Sa ia and Myxopyrum enerve Van Steenis and the identification of pecimens of WW. enerve as L. macrobotrys have led me to reexamine the position of ee enerve. Myxopyrum enerve is atypical of Myxopyrum Blume in that its leaves are not triplinerved and its stem 1s terete. This species also differs in its leaf anatomy from Myxopyrum coriaceum Blume and VW. nervosum Blume. In these Myxopyrum species foliar sclereids are confined to below the midrib and only brachysclereids are present; in M]. enerve there are dendrosclereids throughout the lamina (FiGuRE 2, C) as well as brachysclereids in the midrib and the petiole. In both M. coriaceum and M. nervosum the xylem forms several discrete blocks in the petiole; in M. enerve it is crescent shaped (Ficure 3). In addition, there is a difference in position of the ovules— pendulous in ©. enerve and ascending in M. coriaceum and M. nervosum (although Van Steenis (1967) reported that the specimen of M. ner- vosum he had examined had pendulous ovules). In all these characters, WMyxopyrum enerve resembles Chionanthus and not Myxopyrum. Species of Chionanthus have pinnate venation, terete stems, den- drosclereids in the lamina, crescent-shaped xylem in the petiole (Kiew & Ibra- 622 JOURNAL OF THE ARNOLD ARBORETUM [voL. 64 A FiGure 2. Cross sections of laminae: A, Myxopyrum coriaceum (S 25442, Ander- son), X 370: B, M. ae eae 32382), x 370: C, Chionanthus enerve (S 22740, Asah Luang). x I — him, 1982), corolla lobes longer than the tube, and pendulous ovules. Her- barium specimens of \V/. enerve also have black petioles, a character found in — but not in Myxopyrun. The small flowers fall within the range f flower size for Chionanthus (2-9 mm long: Kiew, 1980). Two characters eee for Chionanthus are the scandent habit and the terminal inflorescence. However, another Bornean species, C. /ongipetalus (Merr.) Kiew, is a shrubby species that has once (S 2/498, Ashton) been recorded as chmbing, and C. retusus has a terminal inflorescence. For all these reasons, \W. enerve is un- doubtedly a species of Chionanthus. Chionanthus enerve (Van Steenis) Kiew, comb. nov. My Xopyrunt enerve Van Steenis, Blumea 15: 152. fig. 3. 1967. Type: Sarawak, 4th Division, Gunung Dulit, S 22740, Asah Luang (holotype. t). Woody scandent climber 2-5 m long. Twigs terete, lenticellate. drying gray. Leaves with petiole 0.5-1.5 cm long, grooved above, drying black: blade ovate- lanceolate, 4.5-12.5 by 1.5-6.5 cm, acute to acuminate at apex, truncate to rounded at base, entire and recurved at margin, thickly coriaceous, glabrous, punctate below toward base of leaf; midrib depressed above, prominent below, with lateral veins 5 to 8 pairs, slightly impressed above and below. Inflorescence paniculate, terminal in flower but becoming displaced to lateral position in fruit, 4-12 cm long. once branched, lowest branches less than half length of iniorescence: bracts foliaceous, 4-7 by 1-3 mm. Flowers polygamous, greenish yellow, 1.5-2 mm long: pedicel 1-2 mm long: calyx 4-lobed, deeply divided to base, margin finely ciliate; corolla 1-2 mm long, lobes 4, induplicate, reaching to base, rounded at apex: stamens 2. the anthers almost sessile. 1.5 mm long, connective broad; ovary | mm long. ovoid, 2-loculate. each locule with 2 pendulous ovules, stigma 2-lobed: staminate flowers without vestige of ovary but otherwise similar, Fruits ovoid, 4-7 by 3-6 mm. minutely apiculate. green ripening purple, endocarp brittle; seed 1, endosperm copious. Habitat. Mossy forest on exposed summits or ridges, 780-2500 m: once re- corded from mossy limestone cliff (Gunung Api). VERNACULAR NAME, Batu fawi (Kelabit). 1983] KIEW, CHIONANTHUS 623 Ficure 3. Cross sections of petioles: A, Chionanthus enerve (S 22740, Asah Luang), « 12: B, Myxopyrum nervosum (Henderson 32382), x 20; C, M. coriaceum (S 25442 Anderson), X 25. SPECIMENS EXAMINED. Borneo. SABAH: Kinabalu, Clemens 28777 (Bo), 30910 (Bo, K), 31442 (Bo), 32834 (Bo), 40183 (Bo), 40704a (Kk, L), 50128 (kK), 51469 (Kk); Chew & Corner 4499 (k); Collenette 905 (kK). SARAWAK. Ist Division: Gunung Santubong, Martin S 37597 (SAR). 2nd Division: Lubok Antu, S 3383/, Chai (KEP, SAR), S 33950, Chai (SAR), S 33953, Chai (sar). 4th Division: Gunung Dulit, S 22740, Asah Luang (L); Gunung Api, S 30433, Chai (sar); Kelabit Highlands, Nooteboom & Chai 2263 (SAR). KALIMANTAN: Gunung Semedoem, Hallier 688 (Bo), 709 (Bo Van Steenis (1967) based his description ona single specimen, but this species has been collected from exposed hilltops and ridges from 780 to 2500 m throughout Sarawak and Sabah. There is considerable range in leaf size (from that of the type specimen, 7.5-12 by 3-6 cm, to small-leaved plants such as Clemens 40704a, 4.5-7 by 1.5-2.75 cm) and inflorescence length (4 cm in Clemens 40704a vs. 7.5-9 cm in the type). Among the species from Sarawak, the combination of coriaceous, lanceolate leaves and paniculate inflorescence is found only in Chionanthus laxiflorus. Chionanthus enerve can be distin- guished from this species by its scandent (vs. arboreal) habit, its terminal inflorescence with short lower secondary branches (vs. branches more than one-third the length of the inflorescence), and its elliptic (vs. obpyriform) fruit. Although the type specimen has hermaphrodite flowers, other specimens (5 30433, Chai, and Collenette 905) have staminate flowers; the species is there- fore polygamous. POSITION OF MYXOPYRUM IN THE OLEACEAE Bentham and Hooker (1876) recognized four tribes in the Oleaceae and placed Myxopyrum in tribe Oleineae together with Olea and Ligustrum L., which have induplo-valvate corolla lobes and albuminous seeds. Knoblauch (1895) recognized two subfamilies: Jasminoideae (containing the single tribe Jasmineae), with bilobed fruits and amphitropous or anatropous seeds; and Oleoideae (containing three tribes), with unlobed fruits and—except in Maye- pea Aublet and Myxopyrum—pendulous seeds. Myxopyrum was isolated 624 JOURNAL OF THE ARNOLD ARBORETUM [VoL. 64 TasLe 1. Common peer erey of Myxopyrum and ae Oleeae, compared ith those of tribe Jasminea CHARACTER MyxXopyRUM AND OLEEAE J ASMINEAE Leaf Simple Unifoliate, trifoliate, r pinnate Corolla Lobes Z Tube Medium to short or absent Long to medium Style Medium to short or absent Long to medium Aestivation Valvate (except in Imbricate Osmanthus) Fruit Not bilobed Bilobed Flavones Present Absent from the other genera in tribe Oleineae by its ascending anatropous seeds. Solereder (1891) wrote that ovule position in Mayepea (now Chionanthus) is not uniform and detailed three positions: pendulous (most species), attached laterally, and attached almost at the base. Van Steenis’s report (1967) of pen- dulous ovules in Myxopyrum nervosum (I find them to be ascending) is another indication that ovule position should not be used as a key character in delimiting the subfamilies. Johnson (1957) expanded the scope of subfamily Jasminoideae to include five tribes: Jasmineae Bentham & Hooker (the only tribe with bilobed fruits), Forsythieae H. Taylor ex L. Johnson, Fontanesieae H. Taylor ex L. Johnson, Schrebereae L. Johnson, and the monogeneric Myxopyreae L. Johnson (pro- visionally placed in the subfamily). His subfamily Oleoideae included only two tribes, Fraxineae (Fraxinus L.) and Oleeae (which in Malesia included Chion- anthus, Ligustrum, Olea, and Osmanthus). The latter is a natural group char- acterized by two pendulous ovules in each locule of the ovary and by a basic chromosome number of 23. Apart from its scandent habit and the position of its ovules, Myxopyrum has little in common with the Jasminoideae and in particular with Jasminum L., the other Malesian climbing genus. Ifa wider range of characters is examined (see TABLE 1), it is obvious that Myxopyrum has more in common with John- son’s tribe Oleeae than with the Jasmineae. If keyed out under his subfamily Oleoideae, Myxopyrum splits: species with a fleshy, urceolate corolla key out under Noronhia Stadm. ex Thou. (a similarity already noted by Blume, 1850), and those with the tube shorter than the lobes key out under Olea (Tetrapilus Lour.) and Chionanthus (Linociera). In addition, Myxopyrum nervosum has flavone (apigenin) glycosides that are present in genera of subfamily Oleoideae but absent in those of subfamily Jasminoideae (Harborne & Green, 1980). Based on a combination of leaf, flower, and fruit characters, rather than on the single character of ovule position, Myxopyrum should be transferred to subfamily Oleoideae (Kiew, 1981). However, although Myxopyrum shows sim- ilarities to genera of tribe Oleeae, it is distinct in several respects. In addition to its climbing habit, quadrangular stem, triplinerved leaves, and ascending — 1983] KIEW, CHIONANTHUS G25 ovules, it differs from all other oleaceous genera in two other characters. The separate blocks of xylem (FiGure 3) have not been recorded from other genera of the Oleaceae, which have either a medullated cylinder (reported from Chio- nanthus, Fraxinus, and Osmanthus) or crescent-shaped xylem (reported from Chionanthus, Fontanesia Labill., Jasminum, Ligustrum, Notelaea Vent., Olea, Osmanthus, Phillyrea L., and Syringa L.) (Metcalfe & Chalk, 1950; Kiew & Ibrahim, 1982). The flavones present in M. nervosum are three unidentified glycosides of apigenin not found in the other genera of Oleoideae so far in- vestigated (Harborne, pers. comm.). The chromosome number of Myxopyrum is still unknown. Because of these distinct characters, Myxopyrum should be retained in the separate tribe Myxopyreae, which is included in subfamily Oleoideae. ACKNOWLEDGMENTS This work was supported by Research Grant S/6/73 awarded by the Agri- cultural University of Malaysia. I am indebted to the curators of the Arnold Arboretum and the Rijksherbarium, Leiden, for the loan of specimens, and to the directors of Kew Gardens, Herbarium Bogoriense, the Forest Research Institute, Kepong, and the Forest Department, Kuching, for permission to work in their herbaria. LITERATURE CITED BENTHAM, G., & J. D. Hooker. 1876. Oleaceae. Gen. PI. 2: 672-680. Biume_, C. L. 1850. Oleaceae. Ann. Mus. Bot. Lugduno-Batavum 1; 310-320. Green, P.S. 1958. A monographic rev — of Osmanthus in Asia and America. Notes Roy. Bot. Gard. Edinburgh 22: 439- Harporne, J. B., & P. S. GREEN. an - chemotaxonomic survey of flavonoids in leaves GF Oleacsie: J. Linn. Soc. Bot. 81: 155-167. Jounson, L. A. S. 1957. A review of the family Oleaceae. Contr. New South Wales Natl. Herb. 2: 396-418. Kiew, R. 1979. The genus Chionanthus (Oleaceae) in Malesia. I. The Malayan species. ee Forester 42: 259-277 80. The genus C "hionanthus (Oleaceae) in Malesia. II. The Sarawak species. Ibid. ye 362-392. 1981. Myxopyrum and its position within the Oleaceae. Abstr. XIII Internat. Bot. Congr. 2 &C.S. aac 1982. Comparative study of leaf anatomy of Malaysian species of Chionanthus and Olea (Oleaceae) with special reference to foliar sclereids. J. Linn. Soc., Bot. 84: 79-101. Knosiaucn, E. 1895. Oleaceae. In: A. ENGLER & K. PRANTL, eds., Nat. Pflanzenfam. IV. 2: 1-16. LINGELSHEIM, A. 1927. Die Oleaceen Papuasiens. Bot. Jahrb. Syst. 61: 1-22. MERRILL, E. D. 1918. New species of Bornean plants. Linociera macrobotrys. Philipp. J. Sci. 13: 117. METCALEE, C. R., & L. CHALK. 1950. Oleaceae. Anatomy of dicotyledons. Vol. 2. Pp. 893-900. Clarendon Press, Oxford. SOLEREDER, M. 1891. Ueber eine neue Oleacee der Sammlung von Sieber. Bot. Cen- tralbl. 45: 398-404. 626 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 STEARN, W. T. 1976. rey of Chionanthus and Linociera Bot. Gard. 63: 355-2 STEENIS, C. G. G. J. VAN, eee Miscellaneous botanical notes XVIII. 119. The affinity of the genus Myxopyrum within the —. with description of a remarkable new species from Borneo. Blumea 15: 150- Oleaceae). Ann. Missouri — DEPARTMENT OF BIOLOGY AGRICULTURAL UNIVERSITY OF MALAYSIA SERDANG, SELANGOR, MALAYSIA 1983] GIDEON, SPERMACOCE 627 A NEW NAME IN SPERMACOCE FOR TWO SPECIES OF BORRERIA FROM NEW GUINEA OSIA GIDEON THE BORRERIA-SPERMACOCE PROBLEM has recently been the subject of much discussion by some of the keenest students of the Rubiaceae. For the last century Spermacoce L. and Borreria G. F. W. Meyer have been maintained as separate genera. The only difference between the two is whether one or both valves of the capsule dehisce; their general facies are so similar that a specimen without mature fruits cannot usually be referred to the proper genus with any certainty. I am not sure that this character is either good or constant enough to justify generic separation. Although it sounds defensible on paper, it is difficult to see; worse still, it is absent without ripe fruits. Verdcourt (Kew Bull. 30: 301. 1975) reunited Borreria with Spermacoce and gave an excellent explanation of the problem in support of this viewpoint. Interested readers should refer to his paper. Since Spermacoce sensu lato (including Borreria as a section) 1s readily identifiable, I prefer this view and will treat New Guinea species accordingly. Fosberg (pers. comm.) has for many years (even before Verdcourt’s reduction of Borreria) been using Spermacoce in its broad sense. Merrill and Perry described two species of Borreria from New Guinea, B. linearis Merr. & Perry and B. lanceolata Merr. & Perry. These species, with bilocular capsules and both valves opening from the top to expose the seeds, undoubtedly belong to Borreria (= sect. BORRERIA (G. F. W. Meyer) Verdcourt). The main characters used to distinguish these two taxa were the erect habit of B. lanceolata, its lanceolate leaves, and its slightly larger flowers with corolla lobes definitely pubescent above. Study of numerous subsequent collections has shown that these distinctions are unreliable, and that a single variable taxon is represented. The habit of Brass 11737 (the type of B. lanceolata) is not recorded on the field label; the specimen, with many long, rather slender stems branching near ground level, does not suggest an erect plant. The usual habit is decumbent to low spreading. The leaves vary little in shape: in Brass 4067 (the type of B. linearis) many leaves have dried with the margins strongly curved, but some have remained flat and are distinctly lanceolate. Most leaves have short, stiff hairs along the midrib beneath, as well as on the margin (particularly near the apex). Brass 11737 is the only collection that has no pubescence on the leaves. Flower size is rather variable and cannot be regarded as a reliable character. I have found no flowers in which the upper surfaces of the corolla lobes are glabrous. The seeds, although variable in size, are very uniform in appearance. In the absence of reliable distinctive characters, 1t seems wise to unite the two species under one name in Spermacoce. Although I have © President and Fellows of Harvard College, 1983. Journal of the Arnold Arboretum 64: 627, 628. October, 1983. 628 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 seen only the isotypes (from BRI), Mr. E. Henty, who has seen the holotypes, has kindly passed his notes on to me The specific epithets /inearis and lanceolata cannot be used in Spermacoce due to the existence of several earlier names (S. /inearis HBK. Nov. Gen. Sp. 343. 1818; 8. lanceolata Link, Enum. Hort. Berol. 1: 132. 1821; 8. lanceolata Frank ex Presl, Bot. Bemerk. 86. 1844) that would make them later homonyms. It is therefore a pleasure to name the taxon after Dr. Leonard Brass, who collected the types of the original taxa. Brass 4067 is here retained as the holotype since it is more representative of the species. Spermacoce brassii Gideon, nom. nov. de linearis Merr. & Perry, Jour. Arnold Arb. 26: 34. 1945. Type: [Papua New a, Central District,] E of Mt. Tafa, 2350 m,. Brass 4067 (holotype, A; isotype, os Borreria lanceolata Merr. & Perry, Jour. Arnold Arb. 26: 35. 1945. Type: (Indonesia, Irian Barat,] Balim River, Brass 11737 (holotype, A; isotype, BRI). DISTRIBUTION. New Guinea: along the central cordillera from Wissel Lakes (Irian Barat) to Mt. Donana (Milne Bay District), and the Saruwaged Range; 1300-3000 m alt. DIVISION OF BOTANY OFFICE OF Forests, DEPARTMENT OF PRIMARY INDUSTRY P. O. Box Lar, PAPUA New GUINEA INDEX Abacopteris penangiana, 24 Acer saccharum, 330 Abelia, — sinense, 68 — engleriana, 85 — stachyophyllum, 68 Aberemoa, 257, 258 — sutchuenense, 68 — guianensis, 258 Aceraceae, 9, 67, 518, 544 Abies, 6 Achlyphila, 422, 423 — fargesii, 4, 6, 39 Achras bahamensis, 537 Ablania guianensis, 258, 267 ere aspera, 45 ieee bee 423 — biden — americana, 291 Acioa ee nsis, 259 Ria) 291, 423 Aciotis purpurascens, 276 Abuta, 259 Acisanthera bivalvis, 275, 276 — amara, 259 Aconitum cannabifolium, 46 — rufesc 59 — hemsleyanum Acacia, 141, 149, 157 — scaposum, 47 — tortuosa, 528, 544 Acouroa, 259 Acalypha australis, 62, 126 — violacea, 259 Acanthaceae: Systematics of Holographis, | Actaea asiatica, 47 129-1 Actinidia callosa, 70 Acanthaceae, 85, 129, 131, 149, 154, 272, — — var. henry, 70 284 — chinensis, 70 subfam. Acanthoideae tribe Aphelan- — — var. hispida, 71 re 130, 132 — polygama, 7 — tribe Gendarusseae subtribe Gendarus- — — var. lecomtei, 71 seae, 129 — tetramera, 71 — tribe Justicieae subtribe Asystasieae, 129 Actinidiaceae, 9, 70 — — subtribe Eujusticieae, 12 Actinodaphne reticulata var. forrestii, 49 Acanthopanax giraldii, 74 Additions and Changes in the Neotropical — henryi, Convolvulaceae— Notes on Merremia, — leucorrhizus var. fulvescens, 74 Operculina, and Turbina, 483-489 — — var. scaberulus, 75 Adelobotrys scandens, 2 — setchuenensis, 75 Adenocaulon himalaicum, 90 — trifoliatus, 75 Adenophora axilliflora, 90 Acer, 6, 9, 115, 507 — capillaris, — sect. Negundo, 9 — wilsonil, — amplum, 67 Adiantaceae, 19 — caudatu Adiantum capillus-veneris, 19 — — var. multiserratum, 67 — davidii, | — davidi — edentulum, — erianthum, 67 — erythrochlamys, 19 — flabellatum, 67 — gulan — franchetii, 67 — myriosorum, 19 — griseum, 67 — pedatum, | — henryi, 67 — sagittatum, 259 — maximowiczil, 67 Aegiphila integrifolia, 274 — mono, 68 — laevis, _ ara 67 — villosa, 274 apilio, 67 Aeschynomene, 257 — cle eee . 8, 68 Aesculus glabra, 330 — rubrum, 115, 518, 544 — octandra, 330 630 Agathis, 166 Agave, 145, 149, 157 Agrimonia pilosa, 53 Agrostis clavata, 93 Ainsliaea gracilis, 90, 127 — triflora, 9 Ai1ouea guianensis, 259 Akebia, — trifoliata, 48, 125 , 73, 74 —_ ener eras 73, 74 — sinicum, jones 406 Alet _ stem oba, 95 Moa duclouxi, 19 — farinosa, Alisma, 383-386, 393-398, 402, 405, 407 Alismataceae in the Southeastern United States, The Genera of, 383-420 Alismataceae, 383-420 Allamanda cathartica, 278 fannie okudairai, 20 — squamifera, 20 Allium cyaneum, 96 — henry, — victorialis, 96 Alocasia, 514, 544 Alsophila (Cyatheaceae) in the Americas, 2 A Revision of the Genus, 333-38 Alsophila, 333-382 — abbotti, 335, 337, 339, 342, 344, 347, — amintae, 335, 337, aie 344, 347, 350-353, 356, 372, 3 — — X Alsophila bryophila, 372, 373 — — X Alsophila portoricensis, 341, 345, 372-374 — approximata, 343 — articulata, 372 — auneae, 337, 339, 341, 343, 344, 348, 356, 369, 370, 377 x Alsophila tussacil, 376, 377 — australis, 34 — balanocarpa, 339, 343, 344, 348, 357, 370, 376 _ oe woodwardioides, 376 — bifo rm 45 _ boivinii, 38: 366 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Alsophila ey 335, 337-339, 342, 344, . 358-360, 374-376 ——*x eee fulgens, 374, 375 — — X Alsophila portoricensis, 374 Nephelea portoricensis, 374 — bryophila, 336-339, 341, 342, 346, 347, 351, 352, 354-357, 372, 373, 377 — — X Alsophila Sy as 341, 373 — camerooniana, 341, — capensis, 339, 341, vi 344, 345, 347, 364, 366, 367 — — subsp. capensis, 367-369 — subsp. ai 364, 367-369 — colenso1, 3 — confinis, i 375 — crassa, 370 _ cuspidata: 334, 336, 339, 342-344, 348, 71 — decrescens, 337, 343, 352, 353, 356,358 — dregel, , 356 — dryopteridoides, aoe — dryopteroidea, 352 _ eae 352 x Nephelea portoricensis, 345, 372, — elongata, 363 — engeli, "338, 340, 347, 362-365 — erinacea, 339, 341, 342, 349, 371 — firma, 339, 341, 342, 349, 366, 372 — fulgens, 343, 344, 348, 359, 370, 375 — a 337, 341, 343, 344, 348, 370, 37 _ ae 34] — hotteana, 339, 342, 347, 351, 353-355, 357, 376 — — xX Alsophila woodwardioides, 375. 376 — — X Nephelea sp., 376 — imrayana, a 342, oe 37] — incana, 342, 349, — jimeneziana, 334, ve 348, 370, 375 — khasiana, 341 — latebrosa, 341, 345 — madagascarica, 343 i 6 — mexicana, 372 — minor, 337, 339, 342-344, 346, 347, 351, 353, 354, 356-358, 360, 361, 375, 376 — — X Alsophila woodwardioides, 375 — mossambicensis, 343 1983] Alsophila munchu, 365 — nockii, 334, 338, 339, 342, 344, 347, 349-351, 353 — ornata, 341 Reet 335, 337, 339, 347, 361-364, 366 — amare We — podophy — ms ca 7 339, 343-345, 349, 366, 372 — portoricensis, 339, 343, 344, 348, 352, 356, we 370, 373, 374, 377 — pubescens, 369 — et 334, 337, 347, 361-364, 366 — salvinii, 337-339, 341-343, 347, 362, — serratifolia, 342 — setosa, 337, 339, 342-344, 349, 369, 371 — sinuata, 341 34 — sternbergil, 334 336, 339, 342, 344, 349, 369, a — tricolor — eae o 343, 344, 348, 366, 371 — tryonorum, 371 — tsilotsilensis, 342 — eee 339, 341, 343, 344, 348, 353, 370, 37 — a 335-337, 339, 342-344, 347, -361 — woodwardioides, 336, 339, 342-344 348, 355, 358, 370, 375, 376 — zakamenensis, 343, 360 Amaioua gulanensis, 259 Amanoa Sa ewis 259 Amaranthac meee ilineae 513, 546 Amaryllis atamasca, 514 Amborellaceae, 470, 471 Amentotaxus ee 39 Americas, A Revision of the Genus Al- sophila Ce i in the, 333-382 Ampelopsis, — megalophylla — a 60 Amphicosmia, 345 — riparia, a 367 INDEX Amyris sou 536, 544 — toxifera, 518 eee: ae ta, 286 Anacardiaceae, 63, 287, 518, 543, 546 Anacardium occidentale, 518, 546 Anaphalis margaritacea subsp. japonica, 90 — var. japonica, 90 — sinica, 90 Anatherum argyraeum, 254 — brachystachyum, 234 — macrourum, 236 — virginicum subvar. floridanum, 222 — — subvar. glomeratum, 235 — — subvar. laxiflorum, 212 — — subvar. liebmannii, 219 — — subvar. Sea 223 | | wn fo = - pee] ian! a2 ° 220 g ete 244 — — subvar. tetrastachyum, 225 Anatomy of the Palm Rhapis excelsa, IX. Xylem Structure of the Leaf Insertion, 599-609 Andira, 2 — racem Andrachne chinensis: 62 Andromeda arborea, 524 — paniculata, 5 — racemosa, 524 Andropogon virginicus Complex (Gra- mineae), Systematics of the, 171-254 Andropogon, 171-254 — sect. Andropogon, 173 — sect. Leptopogon, 173, 174, 209 — arctatus, 177, 179-182, 185, 193, 203, 206, 207, 209-211 — arenarius, 174, 175 — argyreus, 253, 254 — bicornis, 173-17 — bourgaei, 174. n — brachystachyus, 179- 182. 184, 186, 187. 191-193, 203, 206, 207, 209, 211, 224, ee — Sabanisi — campsloracheus 212, 216 — capillipes, 2 — aindestinue: 310. 212 — corymbosus, 238 — — var. abbreviatus, 238 632 Andropogon curtisianus, 225 25 ——f. gracilior, 212 | a = a we = = - x — floridanus, 179- 182, 184, 186, 190-193, 203, 206, 207, 210, 211, 221, 222 _ earner 172, 173,199, 481, 182, — — var. glaucopsis, 181, 182, 185, 187, 188, 190, 192, 200, 201, 203, 236, 237, 239-244 — var. glomeratus, 176, 177, 181, 182, 187, 204, 205, 236-243, 248 var. hirsutior, 181, 182, 185, 187, 188, 201, 204, 230, 236, 237, 239-243, — var. pumilus, 177, 178, 181, 182, 185, 187-189, 191, 193, 202-205, 207, 223, 230, 236, 237, 239, 243-248, 253 — — var. tenuispatheus, 244 — gracilior, 212 — gracilis, 173, 174 — gyrans, 172, 173, 190, 212, 213, 219, 224 — var. gyrans, 176, 177, 180-183, 185, 187. 191, 195, 204-208, 210, 212-219 — var. stenophyllus, 180, 183, 184, 187, 191, 203, 204, 206, 209, 212, 213, 217- 219 — halli, — ieee 174, 175, 177 — lateralis, 174, 175 193, 202, 210, us ) eoilenes tacos 224 Se 181, , 202, 219, “reba ae 195, oe ort JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Andropogon liebmannii var. pungensis, 180, 181, 183, 184, 205, 211, 220-222 — subvar. raripilus, 220 — longiberbis, 179-181, 183, 184, 188, 189, 191, 193, 201, 202, 205, 211, 218, 219, — mississippiensis, 254 — mohrii, 22 pungensis, 220 — selloanus, 174, 175 t — ternarius, 173. 175, 209, 210, 253, 254 — var. cabanisii, 173, 206, 210, 254 — — var. glaucescens, 254 — — var. ternarius, 206, 254 _ tetrastachyus, 225 07 — tracyi, 179, 181, 183, 184, 190, 200, AE ee eee 75 — vaginatus, 225 — virgatus, 173 — virginicus, 172, 173,179, 181,187, 191, 201, 203, 212, 217, 223-225, 230, 232, cco amibocns. 238 — — var. dealbatus, 232 — — var. glaucopsis, 243 — — var. glaucus, 181, 183,185, 187, 190- 192, 203, 224-227, 231-234, 244 — — var. graciliformis, 212 — — var. hirsutior, 2 — — — f. tenuispatheus, 244 1983] Andropogon virginicus var. stenophyllus, 217 — — var. tenuispatheus, 244 — — — f, hirsutior, — — var. tetrastachyus, 225 — — var. vaginatus, — — var. virginicus, 171, 176-178, 181, 183-188, 190, 192, 201-207, 222-235, 243, 248, — — subvar. ditior, — — subvar. genuinus, 225 — — subvar. Sear lias 212, 217 Androsace henryi, 79 Anechites nerium, 257 Anemone hupehensis, 47 — triloba, 518, 545 Annonaceae, 257-260, 264, 291,518, 545 Anotis oe 85, 127 Anteno _ Dalen e, 45 Anthurium mane 268 Apalatoa spicata, 260, 288 , 260 Aphelandra, 130, 131, 134, Apocynaceae, 9, 259, 260, rai 267, 273, 278, 286, 519, 544, 545 — umbellatum, 260 Apteria, 294, 300, 302, 304-307 Aquifoliaceae, 63, 273, 519, 543, 545 Aquilegia oxysepala, 47 Arabis, 496, 498 — davidsonii, 498 INDEX 633 pas lyallii, 498 — — rigcissima, 499 var. demota, 498, 499 _ 75 Araliaceae, 74, ioe. 279, 519, 546 a, 613 Arethusa divaricata, 516 Arisaema consanguineum, 95 — eru ee 9 — fargesii, 95 - heteropilm, 95 — lobatum, Aaeiaiognic 259 — 43 Aristolochiaceae: A Synopsis of the Chinese Species of Asarum, 565-597 Arisiolochiaceae, as 44, 259, 519, 543 0 Arrhostoxylum rubrum, 284 — violaceum, 2 Arrowhead, 412 ea 502 _ ands var. nova, 502 — argyi, 91 — — var. incana, 9] — capillaris, 91 — lactiflora, 91 — subdigitata, 91 — tridentata, 502 Arthraxon hispidus, 93 — — var. cryptatherus, 93 Arthromeris cuneata, 36, 124 Aruba guianensis, 260 Arum, 514 Asarum (Aristolochiaceae), A Synopsis of the Chinese Species of, 565-597 Asarum, 565-597 — subg. Asarum, 565-567 634 Asarum subg. Asarum sect. Asarum, 566, 567 — — — ser. Calidasarum, 568 JOURNAL OF THE ARNOLD ARBORETUM , 569 subg. Gamoasarum, — sect. Heterotropa, 566, 577 - Achidasarum, 566, 577 sect. Japonasarum, 569 arifolium, 565 asaroides, 574, 577, 581 cardiophyllum, 568 caudigerellum, 567, 571-573 cece. 567-569, 574 — cardiophyllum, 567-569 ea 565, 567, 569, 571 — var. setchuenense, 569, 571 73 chingchengense, 575, 582, 583 crispulatum, 575, 584, 585 debile, 567, delavayi, 575, 585, 586 elegans, 577 epigynum, 565, 567, es 574 heterotropoides var. 565, 574, 577 himalaicum:; 567, 571, 573 insigne, 576, 589, 591 r. triangulare, 568 longepedunculatum, 589, 591 longerhizomatosum, 576, 591 macranthum, 574, 575, 585 magnificum, 576, 593, 594, 596 mandshuricum, Asarum magnificum var. dinghuense, 576, 595, 396 — — var, magnificum, 596 — maximum, 576, 589, 591 — nanchuanense, 576, 591, 592 i 7 — ae ee 575, 586, 587 — — var. atrovirens, 575, 586, 588 orphyronotum, 586 — sei 567, 569 — renicordatum, 567, 569, 570 — sagittarioides, 576, 591 — sieboldii, 565, 574, 576, 577 — taitonense, 575, 583 — wulingense, 576, 581, 593 Asclepiadaceae, 82, 274 Asia, New Species of Freycinetia and Pan- danus from Southeast (Studies in Male- sian Pandanaceae, 19), 309-324 Asiasarum, 565, 576 — heterotropoides var. mandshuricum, 577 7 — — var. seoulense — sieboldii, 577 Asimina triloba, 518, Asplenium incisum, 24 — sarelii, 24 — trichomanes, 24 Aster ageratoides var. oo 91 — — var. laticorymbus, — — var. micranthus, 91 — var. scaberulus, 9] _ es aibeste ns, 91 — brachyphyllus, 91 Asteropyrum, 9 Astilbe rubra, 51 Astrocaryum, 279 Asystasiella chinensis, 85 — mackinnoil, 21 — multifidum, 20 [voL. 64 1983] INDEX 635 Athyrium vidal, 21, 124 Begonia, 72 — wardii, 21 — evansiana, 72 Atropanthe sinensis, 83 — glabra, 261 Attalea maripa, 278 — hirsuta, 261 Aublet’s Histoire des Plantes de la Guiane — scandens, 261 Francoise, The Plates of, 255-292 — sinensis, 72 Aucuba, 9 Begoniaceae, 72, 261 —_ chinensis, re) Belamcanda chinensis, 98 Austin, D. F., and G. W. Strapces. Addi- — Belliolum (Winteraceae), Wood Anatomy tions and he in the Neotropical of, and a Note on Flowering, 161-169 Convolvulaceae—Notes on Merremia, Belliolum, 161-16 Operculina, and Turbina, 483-489 — crassifolium, 167 Austrobaileya, 470 — gracile, 163, 165, 166 Austrobaileyaceae, 470, 471 — haplopus, 161-163, 165, 168 Avicennia germinans, 540, 544 — kajewsku, 163, — nitida, 540 — pancheri, 161, 163-168 Azalea viscosa, 524 Bellucia grossularioides, 263 Azolla imbricata, 39, 124 Berberidaceae, 9, 48, 125, 520, 543 Azollaceae, 39, 124 Berberis circumserrata, 48 — dasystachya Bacopa, 256 — feddeana, 48 — aquatica, 260 — henryana, 48 Bagassa guianensis, 261 — julianae, 48 Baillieria aspera, 261 — mitifolia, 48 ee involucrata, 44 — sargentiana, 48, 125 — japonica, 45 — triacanthophora, 48 oe 44 — virgetorum, 48, 125 Baldellia, 395, 406 Berchemia, 115 Balsaminaceae, 68, 126 flavescens, 69 Banara guianensis, 261 — floribunda, 69 — minutiflora, 526, 544 — scandens, | — reticulata, 526 Berginia, 129, 130, 136, 140, 153, 155 Banisteria quapara, 261 — hintonii, 130, 15 — sinemariensis, 261 — palmeri, 130, 144 ,D. E. Bourrorp, A.L. — virgata, 130, 136, 139 CHANG, Z. ees T. R. Duptey, S.A. — — var. glandulifera, 130, 143 He, Y. X. Jin, Q. Y. Li, J. L. Lureyn, — Bernardia, 145 S. . SPONGBERG, S.C. Sun, Y.C. TANG, — Berteria guianensis, 261 J. X. WAN, and T. S. Yinc. The 1980 Besleria coccinea, 261 Sino American Botanical Expedition to. — incarnata, 261, 262 Western Hubei Province, People’s Re- — violacea, 262 public of China, 1-103 Betula, 6, 25, 507 gm Arash Bruce, Davip E. BouF- — albo-sinensis, 41 pb, and STEPHEN A. SPONGBERG. Me- — fargesii, 41 ce glyptostroboides—Its Present — insignis, 41 Status in Central China, 105-128 — luminifera, 41, 115 Bartonia, 302 — nigra, 115 Bassovia sylvatica, 261 — utilis, 41 Batatas tomentosa var. elongata, 484 Betulaceae, 41 Bauhinia coronata, 261 Beureria ovata, 521 — guianensis, 261 — succulenta var. revoluta, 521 — hupehana, 60 Bidens parviflora, 91 — outimouta, 261 — pilosa, 91 — rubiginosa, 261 — tripartita, 91 636 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 BiENOn 520 BERG, and BRUCE BARTHOLOMEW. Me- alba, 262 tasequoia glyptostroboides—Its Present —- caenilea, 520, 543 Status in Central China, 105-128 — capreolata, 115, 520, 545 Bourreria havanensis, 520, 521 — catalpa, 520 — ovata, 520, 521, 545 — copaia, 262 Brandisia, 9 — echinata, 262 Briza, 174 — fluviatilis, 262 Siena 514, 545 — incarnata, 26 Brosim — kerere, 262, 263 — ane 269, 280 — pentaphylla, 520 — rufescens, 2 — radicans var. 8, 520 Broussonetia papyrifera, 43 — sempervirens, 530 Bubbia, 161, 166 Bignoniaceae, 9, 262, 520, 543-545 — semecarpoides, 166 Bischofia polycarpa, 62 Buchenavia, 279 Blakea quinque-nervia, 263 — capitata, 256, 266 Blechnaceae, 27, 124 — tetraphylla, 256, 266 Blechnum eburneum, 27 Buchnera palustris, 280 Bletilla ochracea, 98 Bucida a 266 Bocoa, 263 Buck bea — prouacensis, 263 Buckley er Boehmeria diffusa, 44 Buddleja albiflora 81 — gracilis, 44 — davidii, 81 Bog bean, 436 auisongit densa, 94, 127 Bombacaceae, 263, 278,282 Buphthalmum frutescens, 522 Bombax globosa, 263 Bupleurum chinense, 75 Bonetiella, 157 — longicaule var. eo obs Bonnetiaceae, 264, 273 — — var. giraldii, Bontia germinans, 540 — longiradiatum var. porphyranthum, 76 Boraginaceae, 82, B66. 520, 545 — petiolatum, 76 Borneo, Two Unusual Chionanthus Species Burmannia Fam ily, 293 from, and the Position of Myxopyrum —_ Burmannia, 294, 295, 300-305 in the Oleaceae, 619-626 sect. Burmannia, 301 Borreria from New Guinea, A New Name__ — _ ¢ect. Foliosa, 300, 301 Burmanniaceae in the Southeastern United n Spermacoce for Two Species of, 627, 628 States, The Genera of, 293-307 Borreria, 285, 627 Burmanniaceae, 293-307 — alata, 285 — subfam. Burmannioideae, 295 =. lanceolate: 627, 628 — subiam, Corsioideae, 2 — latifolia, 285 — subfam. Thismioideae, 295 — linearis, 627, 628 — tribe Burmannieae, 293, 294, 300 Borrichia arborescens, 522, 544 — — subtribe Apteriinae, 304 Bothriochloa, 173, 17 — — subtribe Burmanniinae, 300 Botrychiaceae, 18 — tribe Corsieae, 294 Botrychium strictum, 18 — tribe Euburmannieae, 294 Bourrorb, D. E., A. L. CHANG, Z. CHENG, — Ube Haplothismieae, 294 T. R. Duptey, S. A. He, Y. X. Jin, Q. tribe Thismieae, 293, 294 Y. Li, J. L. Luteyn, S. A. SHONGHERG: Burnatia, 406 S.C. Sun, Y. C. TANG, J. X. WAN, T.S. — Bursera, 144 YING, and B. ateEoMEt The 1980 Burseraceae, 271, 521, 543 Sino-American Botanical Expedition to Butomaceae, 384, 385 Western Hubei Province, People’s Re- Buttneria scabra, 263 public of China, 1-103 Buxaceae, 9, 62, 126 Bourrorp, Davin E., STEPHEN A. SPONG- — Buxus microphylla var. sinica, 62 1983] Byrsonima aubleti, 273 — verbascifolia, 274 Byttneria scabra, 263 Cabomba aquatica, 263 Cabombaceae, 263 Cacalia ainsliaeflora, 91 — sinica, 91 , 264 Caesalpinia bahaniensi& 528, 529, 544 — brasiliensis, 528, 529 — sepiaria, 60 Caldesia, 395, 396 Calinea scandens, 264 Callicarpa americana, 540, 544 — bodinieri var. bodinieri, 82 var. giraldii, 82 Calycanthaceae, 466, 467, 521, 543 Calycanthus floridus, 521, 543 capil nase a aaa var. japonica, 90 — maximowiczu, 90 CAMPBELL, caer S. Systematics of the Andropogon virginicus Complex (Gramineae), 171-254 Campomanesia aromatica, 281 — grandiflora, 281 Campsis radicans, 520, 544 Campylosiphon, 303 Campylotropis wo 60 Cananga, 256, — ouregou, 264 Canella winterana, 521, 544 Canellaceae, 521, 544 Cannabinaceae, 43 Cannabis sativa, 43 Capparaceae, 265, 290 ana, 290 Gaprifoliacese, 9, 85, 127 INDEX 637 Caraipa angustifolia, 264 — latifolia, 26 Carapa guianensis, 264 ae ay — guianensis, 264 Cardamine eee 50 Cardiocrinum, 9 — giganteum var. yunnanense, 96 9 Carica spinosa, 265 Caricaceae, 265 CARLQUIST, SHERWIN. Wood Anatomy of Belliolum (Winteraceae) and a Note on Flowering, 161-169 pinus, 9 — Pein ie 115 — chinensis, 41 — cordata var. chinensis, 41 — fargesii, 41, 115 — hupeana, 41 — simplicidentata, 41 Carrierea calycina, 72 Cary ocaraceae, 280, Caryophyllaceae, 46, 521, 545 Caryopteris incana Casasia clusiifolia, 535, 544 — commersoniana, 280 — dentata, 28 — guianensis, 272, 280 — pitumba, 280 Cassia apoucouita, 265 Cassine paragua, 519, 545 Cassipourea guianensis, 265 — seguinii, 124 638 Catalpa — bignoni ae 520, 543 Catesbaea spinosa, 535, 536, 546 Catesby’s Plants, The Modern Names for, 511-546 Catinga aromatica, 265 — moschata Catopsis besteroniana, 514, 545 Cayratia izocinea: 69 — pseudotrifolia, 69 Cedrela mahagoni, 531, 545 Celastraceae, 65, 126, 257, 273, 274, 286, 288 Celastrus angulatus, 65 — gemmatus, 65 — glaucophyllus, 65 — orbiculatus, 65 — rosthornianus, 65 — rugosus, Celosia argentea, 46 Celtis, 330 — biondii, 43 — vandervoetiana, 43 Cephaélis, 264, 269, 287 — muscosa, 287 Cephalotaxaceae, 39 Cephalotaxus fortunei, 39, 113 — sinensis, Cercidiphyllaceae, 9 Cercidiphyllum, 8, 9 Cercidium, 141 Cercis chinensis, 60 — racemosa, 60 Cercocarpus, 502 — ledifolius var. ceeag 502 CHANG, A ; G, T. R. DuDLEy, S.A. HE, YX Joc. ln L.boneyn S.A. Sonat RG,S.C.SuN, Y.C. Tae. J. X. , 7. S. Yinc, B. BARTHOLO- MEW, ser D. E. Bourrorp. The 1980 Sino-American Botanical oe to Western Hubei Province, People’s Re- public of China, 1-103 ane chusana, 19 — duclo , 19 ares 19 — kuhnii, 19 Cheiloclinium cognatum, 288 Cheilsoria chusana, 19 Chelonanthus, 273 — alatus, 272, 273 CHENG, CHING-YUNG, and CHUN-SHU JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Yana. A Synopsis of the Chinese Species of Asarum (Aristolochiaceae), 565-597 CHENG, Z., T. R. DupLey, S. A. He, Y. X. Jin, Q. Y. Li, Jo L. Luteyn, _ A. SPONGBERG, S. C. Sun, Y. C. TANG, J. : N, T. S. Yinc, B. aor D.E. Bourroan, and A. L. CHANG. The 1980 Sino-American Botanical Expedi- tion to Western Hubei Province, Peo- ple’s Republic of China, 1-103 Chikusicholoa, 174 Chimaphila japonica China, Central, IMemsequeia glyptostro- boides—Its Present Status in, 105-128 China, People’s Republic of, The 1980 Sino-American Botanical Expedition to Western Hubei Province, 1-103 Serres Species of Asarum (Aristolochi- ae), A Synopsis of the, 565- eee Species from Borneo, Two Unusual, and the Position of Myxopy- rum in the Oleaceae, 619-626 Chionanthus, 275, 619-622, 624, 625 — sect. Ceranthus, 619 — sect. Eulinociera, 619 << — longipetalus, 620, 622 — macrobotrys, 620, 621 iginicus us, Saas 544 @hiysobalanaceae, 259) oe 267, 271, 272, 276, 279, 286, 521, 5 Chrysobalanus icaco, or 543 Chrysophyllum, 2 — cainito, me — macouco Chrsosenun cillatum, 51 — lanuginos _ nner. 6 51 — pilosum var. valdepilosum, 51 Cimicifuga acerina, 47 a, 47 — glomerata, 235, 236 — lateralis, 225 1983] INDEX 639 Ciponima guianensis, 265 Cipura paludosa, 265 Circaea alpina subsp. imaicola, 74 Codonopsis pilosula, 90 — tangshen, 90 meets bracteatum, 98 — erubescens, 74 — viride var. bracteatum, 98 — glabrescens, 74 Coffea Aer 265 _ sees 74 — paniculata, 265 — repens, 74 Colocasia esculenta, 514 ae fargesii 92 Colubrina, 144, 149, 157 — henryi, 92 — elliptica, 535, 543 _ Heraes var. intermedium, 92 Cissampelos smilacina, 532, 543 ta, 535 Odi breiaecac, ans: 279, 286, 522,544 Cissus tuberculata, 540, 544 Co mbretum cacou Claytonia virginica, 330 — coccine Cleistes Svea 516, 544 — laxum, 265 — grandiflorum, 272 _ rotundifolium, 265 ae iene 47 — lasiandra, 47 545 — ecient 47 Commelinaceae, 9, 95, 265, 515, 545 — uncinata var. coriacea, 47 Comolia latifolia, 282 — villosa, 282 Compositae, 90, 127, 261, 269, 278, 522 544, 545 Conami brasiliensis, 265 — lanosa, 71 CONANT, Davip S. A Revision of the Ge- Cleome guianensis, 265 nus Alsophila aaa in the Clercia, 273 Americas, 333-382 Conceveiba guianensis, 265 Coniogramme caudiformis, 19 — caudifrons, 19 Clerodendrum bungei, 82 —tr a oe mum var. fargesii, 82 ichotomum, 82 Gene SR 522, 544 — intermedia, 19 — fargesii, 77, 115 — robusta var. repandula, 20, 124 Clethraceae, 77, 522, 544 Ca 284 Clibadium suriname ] Conobea aquatica, 266 Clidemia capitellata var. ' dependens, 276 Conocarpus erectus, 522, 544 — hirta var. elegans, 275 — racemosus, 522 — rubra, 276 Conohoria, 282 — spicata, 276 — flavesce ns, 266 Convolvulaceae, Additions and Changes in the Neotropical — Notes on Merremia, Cliftonia monophylla, 329 Clinopodium polycephalum, 83, 127 Clintonia udensis, 96 Operculina, and Turbina, 483-489 Clitoria guianensis, 267 Convolvulaceae, 82, 266, 274, 277, 483- Clusia pana-panari, 281 489, 523, 543, 545 — rosea, 522, 546 Convolvulus alatus, 487 Clusiaceae, 522, 546 — batatas, 523 Clutia cascarilla, ie 544 glaber, 266 — eluteria, — guianensis, 26 Cnemidaria, 333 a — pentaphyllus, 486 Coccocypselum, 281 Conyza canadensis, 92 — guianensis, 28 8 — orbiculatus, 15 — flavescens, 266 — heterophylla, 266 — nodosa, 266 640 JOURNAL OF THE ARNOLD ARBORETUM Cordia sebestena, 521, 545 — tetrandra, — tetraphylla, 256, 266 — toqueve, 266 Cormophyllum capense, 367 Cornaceae, = 77, 126, 523, 543 Ea a, 77, 115 — flor sae a, 523, 543 — kou var. ane 77, 126 Corsiaceae, 294, 295 Corydalis davidii, 50 — incisa, 50 Corsopsis platypelta, 53 — sinen — aes 53, 12 Corylus ferox var. tibetica — heterophylla var. sutchuenensis, 41 — mandshurica, — sieboldiana var. mandshurica, 42 — yunnanensis, 42 Corynostylis arborea, 289 — bullat Couroupita guianensis, 267 Coussapoa angustifolia, 267 — latifolia, 26 Coussarea paniculata 265 Coussarea violacea, 267 Coutarea hexandra, 267 — speciosa, 267 Coutoubea ramosa, 268 — spicata, 268 Crassulaceae, 51 Crataegus cuneata, 55 — wilsonil, 55 Crawfurdia Crenea maritima: 268 Crotalaria guianensis, 268 41, 145 , 268 Cruciferae of Western North America, i 0 jolene japonica, 110 C a Seces var. japonica, 76 — japonica Cucubalus hacaica. 46 = lanceolata: 9, 113, 117 Cupirana aubletiana, 267 Cupressaceae Cupressus disticna: 513 — funebris Curatella americana, 268 Curculi Curtia fenutealia, 269 Cuscuta japonica, 82 Cyathea, 333, 334, 341-344 — — sect. Gy — subg. Sphaeropteris, 333 — abbott, 352 — arborea var. concinna, 376 — capensis, 366, 367 — — var. polypodioides, 369 — concinna, — confinis, 355, 375 — conquisita, 377 333 mnosphaera, 333, 342 1983] ale crassa, ‘de ata, 37 — rupestris, 361 — setosa, 342 — stubel — urbanii, 360 — — var. conferta, 360 woodwardioides, 370 Cyatheaceae: A Revision of the Genus AI- sophila in the Americas, 333-382 Cyatheaceae, 333, 334 Cyclocarya Se 41 Cyclosorus acuminatus, 24 Cydista eeea 262 Cymbopogon glaucus, 243 Cynanchum auriculatum, 82 — inamoenum, 82 Cynoglossum zeylanicum, 82 Cyperaceae, 94, 127, 274, 282 Cyphomandra hartwegil, 285 — tegore, Cypripedium acaule, 516, 545, 546 — calceolus, 516 ar. pubescens, 516 — — var. tortile, 516, 545 INDEX Cyrtocarpa, 144 rtomium fortunei, 27 — macrophyllum, 27 Cytisus violaceus, 268 Dactylis glomerata, 94 re 259 — dyeriana, 60 _ Seon ck 529, 544 — lanceolaria, 259 — mimosoides, 60 — stenophylla, 125 Damasonium, 384, 385, 395 DANIEL, THOoMas F. Systematics of Holo- graphis (Acanthaceae), 129-160 Daphniphyllaceae, 62 Daphniphyllum longistylum, 62 2 New Species of Ti- monius 's (Rubiaceae) from Papuasia, 61 1- 18 Dasylirion, 149 avidia, 8, 9 = “involucrata var. vilmoriniana, 73 ,9 Delphinium potaninii, 47 Dendranthema boreale, 92 — indica, 92 Dendrobenthamia angustata, 77 — japonica var. chinensis, 77 Dendropemon purpureum, 530, 546 Dennstaedtia hirsuta, 18 , 18 Dennstaedtiaceae, 18, 124 Dentaria, 330 Derris, 368 — pterocarpus, 268 68 9, 500 Desmodium oldhamii, 60 642 JOURNAL OF THE ARNOLD ARBORETUM Desmodium podocarpum subsp. fallax, 60 — subsp. oxyphyllum, Desmoscelis villosa, 276 Deutzia schneideriana, 51 — vilmorinae, 51 Deyeuxia sylvatica, 94 Dichanthium, 173 45 Dichorisandra hexandra, 265 Dichroa febrifuga, 5 Dichromena latifolia, 329 Dicotyledoneae, 40, 124, 518 Didymopanax morototoni, 279 Digitaria sanguinalis, 94 Dilkea, 284 Dilleniaceae, 264, 268, 285, 287, 288 Dimeiostemon macrourus, 236 — tetrastachyum, 225 — vaginatus, 225 Dioclea nado lpi cides: 529 Diodia, 286 Dioscorea giraldii, 97 — nippo nica var. rosthornil, 97 e, 97 Diospyros edienendt 279, 283 — martini, 283 _ ‘virginiana, 523, 545 oni 89 Dipteniemtinry violaceus, 284 Dipteroma, 9 Dipteronia sinensis, 68 _ eapositifelia. 287 Disporum bodinieri, 96 — cantoniense, 96 Dodecatheon meadia, 534, 546 Doliocarpus dentatus, 288 Doliocarpus guianensis, 285 —s ns, 26 — spraguei, 264 Draba, 496, 501 — densifolia, 502 — hitchcockii, 500-502 — oreibata, 502 — paysonil var. ee 501, 502 — — var. treleasu, 501 — pennellii, 502, 503 — sphaeroides, 503 Dracontium foetidum, 514 — pentaphyllum, 268 Drymonia coccinea, 261 Drymotaenium miyoshianum, 36, 124 Dryoathyrium dielsu, 21 — henryi, ee aes 21 — unifurcatum, 21 Drjoptenidaceae 27, 124 _ submarginalis, 30501 — supraimpressa, 30, 32, 124 — tokyoensis, 32 — yunnanensis, 28, 30 Duck-potato, 412 [VOL. 64 Dub ey, T. R., S. A. He, Y. X. Jin, Q. Y. Li, J. L. Luteyn, S. A. SPONGBERG, S. C. Sun, Y. C. TANG, J. X. WAN, T.S. YING, B. BARTHOLOMEW, D. E. BOUFFORD, A. L. CHANG, and Z. CHENG. The 1980 Sino- American Botanical Expedition to West- ern Hubei Province, People’s Republic of China, 1-103 Duguetia guianensis, 258 Duroia, 267 — er 267 Ebenaceae, 79, 279, 283, 523, 545 Ecastophyllum brownei, 529 Echinacea purpurea, 522, 545 1983] INDEX 643 Echinodorus, 383, 384, 386, 393, 395, 396, | Ericaceae subfam. Monotropoideae, 295 15 401-407, 413, 4 rigeron elongatus, 92 — subg. Echinodorus, 404-406 Eriocaulaceae, 95, 288 — — sect. Berteroii, 404 Eriocaulon buergerianum, 95 — — sect. Cordifolu, 404 — robustium — — sect. Echinodorus, 404 Eriophorum comosum, 94 — subg. Helianthium, 403, 405 Eriosema violaceum, 268 — sect. Tenelli, 403, 406 Eriotheca globosa, 263 Bohites catesbaei, 519, 545 Erythrina, 145, 529 — corymbosa, 257 — herbacea, 529, 544 — umbellata, 519, 544 Erythronium americanum, 330 Ehretia bourreria, 520 Eschweilera, 27 Elaeagnaceae, 72 — amara, 272 Elaeagnus henryi, 72 — grandiflora, 272 — lanceolata, 7 — odora, — umbellata, 72 — parviflora, 272 hoes 257, 258, 267, 281, 285 Ethnora, 278 Elaphrium simaruba, 521 — maripa, 278 Elatostema ichangense, 44 Eucommia, 9 — stewardi, 44 Eucommiaceae, 9 Elleanthus ae: 284 Eugenia arivoa, 269 Elsholtzia ciliata, 83, 127 — biflora, 268 — cypriani, 83 — coffeifolia, 268, 269 — flava, 83, 127 — feijoa, 265 — fruticosa, 83 — guianensis, 268 Emmenopterys, 9 — latifolia, 268 — henryi, 8 — mini, 268 Enpress, PETER K., and F. B. SAmMpson. — montana, 268 Floral Structure and Relationships ofthe | — pomifera, 257 Trimeniaceae (Laurales), 447-473 — sinemariensis, 268, 269 seen a 78 — tomentosa, 269 — serrulat 8 — undulata, 269 Enourea ie. 268 Euodia, 61 Eperua falcata, 268 — rutacarpa var. oo 61, 126 — grandiflora, 279 — — var. officina Epicladium boothianum, 51 Euonymus St aie 65 Epidendrum boothianum, an 517, 545 — alatus, 65 — cochleatum, 517, 545 — cornutus, 65 — nocturnum, 517, 545 — crinitus, 66 — plicatum, 517, 545 — elegantissimus, 66 — vanilla, 517 — fortunel, Epilobium cephalostigma, 74 — giraldi, 66 — nepalense, 74 — hamiltonianus, 66 — parviflorum, 74 — kiautschovicus, 66, 126 — sinense, 74 — maackii, 66 Epimedium, 9 — oxyphyllus, 66 —d — phellomanes, 66 Epirrhizanthes, 295 — porphyreus, 66 Equisetaceae, 17 — sanguineus, 66 quisetum arvense, 17 — verrucosoides, 66 — hyemale, 17 — — var. vividiflora, 66 — ramosissimum, 18 Eupatorium, 33 Erianthus fulvus, 94 — amara, Ericaceae, 9, 78, 127, 523, 544, 546 — chinense, 92 644 Eupatorium japonicum, 92 — latifoli 9 Euphorbia chrysocoma, 62 — hylonoma, 62 Papherbinsese 62, 126, 257, 259, 265, 268, 271, 273-275, 278, 524, 544, 545 _ tribe oo 275 — obtusifolia, Hil. 126 Euscaphis japonica, 66 Evea, 257, 269 — guianensis, 265, 269 Exacum guianense, 269 — tenuifolium, 269 Fagaceae, 42, 124, 525, 543 Fauria, 431, 432, 437 Ferohia, 269 — gulanensis, 269 Ficus, 145 — brevifolia, 532 — citrifolia, 532, 546 — sarmentosa var. ail 43 — — var. impressa, 43 Flacourtiaceae, 72, 257, 261,272, 275, 277, 280, 282, 526, 544 Floating-heart, 440 Floral Structure and Relationships of the Trimeniaceae (Laurales), 447-473 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Flourensia, 157 Flowering, Wood Anatomy of Belliolum (Winteraceae) and a Note on, 161-169 Fontanesia, 625 Forsteronia aoe 260 os Forssthia suspensa 80 Fortune Fothergilla, 256, 270, 286 — mirabilis, 269, 270 Fouquieria, 144, 149, 157 Fragaria gracilis, 55 Braxinis: 115, 330, 624, 625 — americana, 533, 544 — paxiana, 80 Freycinetia and Pandanus from Malesia and Southeast Asia, New Species of (Studies in Malesian Pandanaceae, 19), 309-324 Freycinetia, 3 — sect. Blumeella, 309, 310 — sect. DeVrieseella, 310 seudo-insignis, 310 — reinec 10 — scabripes, — sphaerocephala, 312 Fuirena squarrosa, 329 Fumariaceae, 50 Fusea longifolia, 259 Galactia rudolphioides, 529, 544 Galinsoga parviflora, Galipea trifoliata, 270 Gelsemium sempervirens, 530, 543 Genera of Alismataceae in the Southeast- States, Indexes to _Papers Published as Parts of the, 547-563 Gentiana, 432 1983] Gentiana oe 526, 527, 544 — panthaica, 81 Brees , 526 Gentianaceae, 81, 127, 268, 269, 272, 273, 281, 286, 290, 295, 432, 526, 544 Gentianopsis scabromarginata, 81 ] — epigynum, 565, 573 Geraniaceae, 61 Geranium henryi, 61 — sibiricu Gerardia digitata, 484 are 84, 261 Ghinia, 270 eG Osia. A New Name in Sperma- coce for Two Species of Borreria from New Guinea, 627, 628 Ginkgo a 113, 122 Glaziocharis, 294 Gleditsia aquatica 529, 543 — triacanthos var. 6, 529 Glyptostrobus ino 107 — pensilis, 107 Gnetaceae, 287 Gnetum paniculatum, 287 — urens, 287 Gomortegaceae, 471 Gonolobus scandens, 257 Gonostegia hirta, 44, 125 Goodeniaceae, 527, 544 Goodyera, 302 Gordonia lasianthus, 538, 543 Goupia glabra, 270 Gouplaceae, Gramineae: Systematics of the Andropo- gon virginicus Complex, 171-254 Gramineae, 93, 127, 173, 279, 515, 543 — subfam. Panicoideae, 179 — tribe Andropogoneae, 172, 177, 178, 8 Grewia biloba, 70 Grias tetrapetala, 280 Guapira, 271 — guianensis, 270, 271 Guatteria, 258, 260 Guettarda coccinea, 271 Gustavia hexapetala, 280 INDEX 645 Guttiferae, 71, 271, 273, 277, 281, 289 Gymnosphaera, 345 — glabra, 345 Gymnostemma cardiospermum, 89 — pentaphyllum Gynura crepidioides, 92 Gyrocarpus, 469 Haematoxylon capechianum, 529, 545 Haemodoraceae, Halenia elliptica var. elliptica, 81 — — var. grandiflora, Halesia carolina, 538 — tetraptera, 538, 544 Hamamelidaceae, 9, 53, 125, 527, 545, 546 Hamamelis virginiana, 527, 546 Haplothismia, 293 H Y.X. Jin, Q. Y. Li, J. L. Lureyn, S. A See S.C. Sun, Y. C. TANG, J. KX. Wan, T. S. Yinc, B. BARTHOL- w, D. E. Bourrorb, A. L. CHANG, Z. CHENG, and T. R. DubDLey. The 1980 Sino-American Botanical Expedition to Western Hubei Province, People’s Re- public of China, 1-103 Hechtia, 144, 149 Hedera nepalensis var. sinensis, 75 Hedycarya, 468 Hedyosmum, 469 Hedysarum ecastophyllum, 529 Helianthium, 403, 406 Helwingia, 9 — chinensis, 77 — japonica, 77 — — var. hypoleuca, 77 — — var. japonica, 77 Hemerocallis minor, 96 Hemiphragma heterophyllum, 84 Hemitelia capensis, 367 — firma, 372 — gardneriana, 369 — riparia, 367 Hemsleya chinensis, 89 Henrietella flavescens, 275 276 u Hecnandia 469, 471 646 Hernandia guianensis, 271 Hernandiaceae, ape ee 468, 471 Heymassoli spinosa, 271 Hibiscus pernambucensis, 531 — tiliaceus, 531, 545 Hippocrateaceae, 273, 274 Hippomane mancinella, 524, 545 Hiraea multiradiata, 261 — quapara, 261 Hirtella americana, 271 — guyanensis, 286 — racemosa, 271, 2 Histoire des Plantes de la Guiane Fran- coise, The Plates of Aublet’s, 255-292 Holboellia, 9, 125 — fargesii Holcus vireiiiou Holographis (Acanthacea®), Systematics of, l 29- Holographis, 129-160 — anisophylla, 131-135, 137, 150-153, 155 — argyrea, 131-135, 137, 145-147 — ehrenbergiana, 129, 131-137, 150, 157- 159 — hintoni, 131-135, 137, 146, 150, 153 — ilicifolia, 130-133, 135-137, 147-149 — pallida, 130-135, 137, 152, 154, 155 — parayana, ee 134, 135, 137, 139, 146, 149, 153 Be ee ie 135, 137, 146, 155-157, 159 — tamaulipica, 131-133, 135-139, 153, 159 — virgata, 131, 133, 135, 137, 139-145, 147 —— ya glandulifera, 143 — ee ph: 140-145 45 Ffomelinn guianense, 277, 282 Hopea tinctoria, 538 Hosiea sinensis, 67 Hosta, Houmiri, 256, 271 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Houmiri balsamifera, 271 Houttuynia cordata, 40, 115 Howarb, RICHARD A. The Plates of Au- blet’s Histoire des Plantes de la Guiane Francoise, 255-292 Howarpb, RICHARD A., and GEORGE W. STAPLES. The Modern Names for Cates- by’s Plants, 511-546 Hubei Province, People’s Republic of China, The 1980 Sino-American Botan- ical Expedition to Western, 1-103 Hugeria vaccinioides, 79 , 289 Humulus ae 43 — scandens, Huperzia crisp, 17 — hupehensis, 17 Hybanthus accolals 289 Hydrangea aspera subsp. robusta, 51 ubsp. chinensis, 52 Hydrocharitaceae, 384, 515, 544 Hydrolea eee 284 — spinosa, 27 Hydrophyilaceae, 271, 284 13 Hymenophyllaceae, 18, 123 ynium virgatum, 173 Hypolepidaceae, 18, 124 Hypolepis punctata, 18, 124 ypoxidaceae, 295, 515, 543 43 — juncea, 515 Icacinaceae, i. 281 Icacorea, 27 1983] Icacorea guianensis, 271 Icica altissima, 27 ge _ heptaphylla, 2722: Idesia, 283 Idria, 141 Ilex, 115 — subg. Ilex sect. Paltoria ser. Cassinoides, 64 — cassine, 519, 543 9 — pedunculosa, 63, 64 — pernyi, 63 — rockii, 64 — shennongyjiaensis, 63-65 — sugerokii, 63-65 — — var. brevipedunculata, 65 — vomitoria, 519, 545 Tilieiacene: 9, Illicium — henryi, 49 Impatiens blephorosepala, 68 — dicentra, 6 — exiguiflora, 68, 126 Inbreeding Depression in Metasequoia, 5-481 Indexes to Papers | to 100 Published as Parts of the Generic Flora of the South- eastern United States, 547-563 Ipomoea, 483, 488, 489 — sect. Eriospermum, 489 — albiflora, 484 — — var. cineria, 484 INDEX 647 Ipomoea albiflora var. divergens, 484 r. stricta, — — var. y simplicifolia, 484 — ericoides, 484 Iridaceae, 98, 265, 295 — subfam. Geosiridoideae, 295 — subfam. oo 295 Iris wilsonu, 98 Irlbachia ieee 273 Iroucana guianensis, 272 Isachne nipponensis, 94, 127 Ischnogyne, 9 Isertia coccinea, 271 Itea ilicifolia, 52 Ixora davisil, 280 gees caerulea, 520, 543 — copal Ce pincea 265 Jacquemontia guianensis, 266 4 — caulescens, 569, 571 asminum, 624, 625 — floridum, 80 — lanceolarium, 80 Jatropha, 141, — vernicosa, 145 Jin, Y. X., SG eer ae SPONGBERG, S. C. Sun, Y. C. G, J. T. S. Yinc, B. ae D. E. Bou FFORD, A. L. CHANG, Z. CHENG, T. R. Duptey, and S. A. He. The 1980 648 Sino-American Botanical Expedition to Western Hubei Province, People’s Re- public of China, 1-103 Juglandaceae, 41, 527, 543, 544 Juglans alba, 527, 528 — luzuliformis, 95 _ modicus, 95 — potaninil, 95 Juniperus ate 113 upica, 42 Jussiaea, 257 Justicia coccinea, 272 — variegata, 272 Kalimeris indica, 92 Kalmia gue 523, 546 — latifolia, 524 Kerria japon 5 Keteleeria co dina 39, 113 Krew, Rutu. Two Unusual Chionanthus Species from Borneo and the Position of Myxopyrum in the Oleaceae, 619-626 Kinostemon ornatum, 83 Kolkwi Krat, Ropert. The Xyridaceae in the Southeastern United States, 421-429 Kuser, JOHN. Inbreeding Depression in Metasequoia, 475-481 Labatia pedunculata, 281 Labiatae, 83, 127 Lactuca graciliflora, 92, 127 Laguncularia racemosa, 522, 544 Landolphia, 278 — guianensis, 278 Lardizabalaceae, 9, 48, 125 Larrea, 141, 149 Lauraceae: A New Lindera from North America, 325-331 Lauraceae, 9, 49, 125, 259, 266, 272, 278, 463, 466-468, 470, 471, 528, 543, 544 ubfam. Cassythoideae, 470 Laurelia, 468, 471 Laurus borbonia, 528, 544 — catesbyana, 528, 544 — coriacea, 528 — sassafras, 528 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Laurus winterana, 521 Leaf Insertion, Xylem Structure of the: Anatomy of the Palm Rhapis excelsa, IX, 599-609 Leandra agrestis, 275 Lecanthus peduncularis, 44 Leguminosae, 60, 125, 257, 259-261, 263, 265-268, 276-279, 281, 282, 285-290, 528, 543-546 Leiphaimos, 295 Lemaireocereus, 144 Lemmaphyllum drymoglossoides, 36 Lemna japonica, 9 Lemnaceae, 95 Leontopodium Jj japonicum, 92 ae artemisia, 83 — japonicus, Deere drymoglossoides, 36, 124 — elongata, 36 Lepidomicrosorium subhastatum, 36, 124 Lepisorus asterolepis, 36 36 Leptodermis oblonga, 85 — wilsonii, 85 Leptolepidium kuhniil, Leptopogon carinatus ok arctatus, 207 Leptopus chinensis, 62 Lespedeza Bea. 60 — cuneata, - — thun 1, 61 fee "49 | 493, 496, 507 — alpina, 493 — arenosa, 507 — argyrea subsp. diffusa, 507 — cinerea, 493 — garretti 3 — goodrichii, 503-506 — kin _ ene 507 — — subsp. cinerascens, 505 — — subsp. occidentalis, 505 1983] Lesquerella parviflora, 506, 507 Li, Q. Y., J. L. Luteyn, S. A. SPONGBERG, S. C. Sun, Y. C. Tanc, J. X. WAN T. S. Yinc, B. BARTHOLOMEW, D. E. 1980 Sino-American Botanical Expedi- tion to Western Hubei Province, Peo- ple’s Republic of China, 1-103 Licania alba, 264 — guianensis, 276 — incana, 27 — latifolia, 264 Licaria guianensis, 272 Lieutautia mirabilis, 270 Ligularia hodgsonii, 92 — veitchiana, 93 Ligusticum daucoides, 76 Ligustrum, 623-625 — acutissimum, 80 — henryi, 80 — sista 80 _ oe Sy — sin ideas "9, oe 515, 543, 545, 546 Lilium canadense, 515, 546 — catesbaei, 515, 545 — lancifolium — michauxii, 516 _ philadelphicum, 515, 546 — superbum, 516, 545 — taliense, 96 40 Limnocharitaceae, 384, 385 Limnophyton, 396 Limodorum grandiflorum, 272 — pendulum, 2 Lindera (Lauraceae) from North America, A New, 325-331 Lindera, 325-331 — benzoin, 115, 325-327, 329-331 — — var. pubescens, 325 a, 50 — melissifolia, 325, 326, 329-331 INDEX 649 Lindera obtusiloba, 50 — subcaudata, 5 — subcoriacea, 325-331 — sagittata, 259 — rupicola, 619, 621 Liparophyllum, 431, 432 Liquidambar, 330 — acalycina, 53, 115, 125 — formosana — styraciflua, 115, 527, 545 Liriodendron ,9 — tulipifera, 330, 530, 543 Liriope graminifolia, 96 Listera, 3 Lisyanthus alatus, 272 ene ele stocagpus. 42 — polystachyus, 4 Litsea ichangensis, 50 — sericea, 5 Lobelia plumierii, 527, 544 rAalnea ae 81, 276, 283, 530, 543, 545 Lomatogonium bellum, 81 Lonchocarpus, 283 — rufescens, 282 Lonicera, 86 — gynochlamydea, 85 henryi, 86 SAsiefencis, 86 — tangutica, 86 — tragophylla, 87 — trichopoda, 87 Lophiocarpus, 413 Lophiola americana, 329 Lophotocarpus, 41 Loranthaceae, 44, 530, 545, 546 Loranthus levinei, 44 Loreya arborescens, 275 650 Loxogrammaceae, 39 Loxogramme grammitoides, 39 Lunathyrium, 24 — centro- ae 24 — piraldii, _ eee 21, 23 — vermiforme, 21—23, 124 — wilsoni, 23, 124 Lundellia, 129, 130, 136, 147 45 Luteyn, J. L., S.A. pees SUN, ~S Teel: X. , 1. S. Yinc, B. D. . anaes A. L. Chins Z CHENG, T. R. eee S.A, E, Y. X. Jin, and Q. Y. Li. The 1980 eae American Botanical Se. to Western Hubei Province, People’s Re- public of China, 1-10 Lycianthes lysimachioides, 83 Lycopodiaceae, Lycopodium espa: 17, 123 _ eo ley Lyonia, 9 _ ovalifolia var. elliptica, 78 Lysiloma, 14 — latisiliquum, 529, 544 Lysimachia christinae, 79 — clethroides, 79 Lysionots paucifiorus, 84 Lythraceae, 73, 126, 268 Mabea piriri, 273 — taquari, 273 3 Macfadyena bracteosa, 262 Machaerium quinatum, 277 Macleaya microcarpa, 50 Maclura, Macoubea guianensis, 273 Macrothelypteris oligophlebia var. ele- gans, Mago acuminata, 530, 531, 546 — glauca, 531 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Magnolia ree 531, 545 — macrophyl 531 Malesia and Southeast Asia, New Species of Freycinetia and Pandanus from (Stud- ies in Malesian Pandanaceae, 19), 309- 2) 32 Malesian Pandanaceae, Studies in, 19. New Species of Freycinetia and Pandanus from Malesia and Southeast Asia, 309- 324 Mallotus oo 62 — tenuifolius, 126 Malouetia araquaes 264 — verbascifolia, 274 — volubilis, 261 MMalpichisceae: 261, 273, 274 Malus, 8, $5, 57 Maprounea guianensis, 274 Maquira guianensis, 274 1983] Marcgraviaceae, 277, 285 Maripa scandens, 274 Mariscus eiaramiuis, 282 — pedunculatus, 282 Marlierea montana, 268 Mascagnia sinemariensis, 261 — volubilis Mastichodendron foetidissimum, 537, 545 Matayba arborescens, 284 — gulanensis, 2 Matelea latifolia, 274 — palustris, 2 Matourea pratensis, 275 Matteuccia, 9 — centro- chinense, 25 — intermedia, 25 — orientalis, 25, 27 — — f. monstra, 25-27 — struthiopteris, 27 Mauritia, 279 Mayaca fluviatilis, 275 Mayacaceae, 275 Mayepea, 275, 623, 624 — guianensis, 275 Melampyrum roseum var. obtusifolium, 84 Melandrium tatarinowu, 46 Melanosciadum pimpinelloideum, 76 Melastoma agrestis, 27 ais — purpurascens, 276 — rubra, 276 — rufescens, 276 — irivalyis: 276 INDEX 651 Melastoma villosa, 276 Melastomataceae, 263, 270, 273, 275-277, 282, 286-288, 290 Meliaceae, 264, 531, Melica onoei, 94 Melicope, 61 Meliosma beaniana, 68 — dillenifolia subsp. cuneifolia, 68 ubsp. flexuosa, 68, 126 545 — pinnata subsp. barbulata var. oldhamii, 6 ] — veitchiorum, 68 Menispermum Wenvanthaceac: in the Southeastern United States, The Genera of, 431-445 Menyanthaceae, 431-445 Menyanthes, 431, 432, 436, 437 Merremia, Operculina, and Turbina, Notes on; Additions and Changes in the Neo- tropical Convolvulaceae, 483-489 Merremia, 483-48 — cissoides, 483 contorquens, 483, 484 — digitata, 483, 486 — ericoides, 483, 484 — glaber, 266 — macrocalyx, 486 — nervos — platyphylla, vie 487 — repens, 4 — weberba i Metaplexis hemsleyana, 82 — sinensis, Metasequoia glyptostroboides—Its Pres- ent Status in Central China, 105-128 Metasequoia, Inbreeding Depression in, 475-481 Metasequoia, 14, 105-128, 475-481 — glyptostroboides, 14, 40, 105-128, 475, 478 Metopium toxiferum, 518, 543 Miconia, 270, 275 — alata, 275 652 Miconia guianensis, 270 — mirabilis, 270, 286 — prasina, 276 — racemosa, 276 — rufescens, 276 Microlepia marginata, 18, 124 Microsorium fortunei, 36 Microstegium andl un 94, 127 Microtropis triflora, 66, 126 Mikania amara, 269 — parviflora, 269 Mimosa, 144, 149, 157 8 Mimulus tenellus, 84 Minquartia guianensis, 276 Mitchella repens, 536, 543 Mitracarpus hirtus, 285, 286 odern Names for Catesby’s Plants, The, 511-546 Moniera trifolia, 2 Monimiaceae, 285, 447, 466-468, 470, 471 — subfam. Monimioideae, 463, 469 Monocotyledoneae, 93, 127, 513 9 is, 27 Moquilea guianensis, 276 Moraceae, 9, 43, 125, 261, 280, 281, 532, 546 Moronobea coccinea, 277 Mortonia, 149 Morus, : “ — alba, Mosla aan 83, 12 Mourera fluviatilis, th Mouriri guianensis, 277 Mouroucoa violacea, 277 Moutoubea guianensis, 277 Moutouchi suberosa, 277 Muellera, 266 — frutescens, 266, 267 Muhlenbergia expansa, 329 — hugelii, 94 Myrcia citrifolia, 257) — coumete, 268 JOURNAL OF THE ARNOLD ARBORETUM 207, 269),.274, [VOL. 64 Myrcia tomentosa, 269 Myrica cerifera, 329, 532, 544 — — var. 6, 533 — heterophylla, 329 Myristicaceae, 289 Myrmechis, 302 Myrsinaceae, 79, 127, 271, 282 Myrsine africana, 79 — gulanensis 282 oo 265, 268, 269, 281 s, 257 yrt — citrifolia, 279 Myxopyrum in the Oleaceae, Two Unusual Chionan thus Species from Borneo and 26 — nervosum, 621-625 Nacibea alba, 277 — coccinea, 277 Nandina domestica, 49 pimoga guianensis, 277 Nectandra, 528 Neillia sinensis, 56 Neolepisorus ovatus f. deltoidea, 36 Neolitsea oa a 50 Neopringlea Neottianthe ies 98 — monophylla, 98 Nephelea, 333, 334, 342-345, 369, 377 — balanocarpa, 343, 370 — — X Nephelea woodwardioides, 376 — concinna, 369, 376, 377 0 — cuspidata, 371 fulgens, 370 — grevilleana, 370 mexicana, 372 — polystichoides, 345, 372 — portoricensis, 370 — pubescens, 343, 369 — setosa, 371 — sternbergii. 371 — tussacii, 37 — woodwardioides, 370 Nephrophyllidium, 431 1983] Nepsera aquatica, 275 Neriacanthus, 130 New Guinea, A New Name in Spermacoce for Two Species of Borreria from, 627, 628 New Lindera (Lauraceae) from North America, A, 325-331 New Name in Spermacoce for Two Species of Borreria from New Guinea, A, 627, 628 New Species of Timonius (Rubiaceae) from Papuasia, 611-618 Nissolia quinata, 277 Nodding nixie, 305 Nonatelia longiflora, 277 — lutea, — officinalis, 277 Norantea guianensis, 277 Noronhia, North Macica. A New Lindera (Laura- ceae) from, 325-331 North America, Studies in the Cruciferae 0 Nothofagus, 613, 615 Nothopanax davidii, 75, 126 Nothosmyrnium japonicum var. sut- 76 Nyctaginaceae, 270 Nymphaea, 439, 440 Nymphoides, 431, 432, 436, 439 Nyssa aquatica, 115, 533, 544 — ogeche, 53 — sinensis, 11 — sylvatica, 533, 543 — var. biflora, 115 Nvcehiceae: 9, 73, 533, 543, 544 Obolaria, 302 Ochaacede: 278, 284 Ocotea, 266, 5 Olea, 619- 621, 6236 625 — americana, 533 INDEX 653 Olea brachiata, 621 — dec ussata, ae 621 Oleaceae, Two Unusual Chionanthus Species from Borneo and the Position of Myxopyrum in the, 619-626 Oleaceae, 80, 127, 533, 544, 619-626 — subfam. Jasminoideae, 623, 624 — — tribe Forsythieae, 624 | | ee a) oy & Oo o o © a a NO ae — tribe Oleineae, 623, 624 Omphalea diandra, 278 ense, 19 aa and Turbinia, Notes on Mer- mia; Additions and Changes in the otropical Convolvulaceae, 483-489 ne 483, 487 — alata, 4 ar. pubescens, 487 | roe) => ae wn wn oo ar. mucronata, 488 — pterodes, 487 — — f. pubescens, 487 Ophiopogon bodinieri, 96 Ophiorrhiza japonica, 85 Oplismenus undulation, 94 Opuntia, 141, Orchidaceae, 9, mw 272, 284, 295, 516, 544-546 Orectanthe, 422, 423 Orelia grandiflora, 278 Oreocnide frutescens, 44 — fruticosa, 44 Origanum vulgare, 83 Ornithogalum hirsutum, 315 Orobanchaceae, 84 Orobanche coerulescens, 84 Orontium aquaticum, 514, 544 Oryza sativa, 515, 543 Osmanthus, 533, cae 625 — americanus, 533, 654 JOURNAL OF THE ARNOLD ARBORETUM [VOL. Osmanthus armatus, 80 — fragrans, 81 Ars Oxydendrum arboreum, 524, 544 Oxypolis filiformis, 329 Pachira aquatica, 278 Pachycereus, 141, 144 Pachycormus, 141, 144 Pachyptera kerere, 262 Pachysandra, 9 — terminalis, 62 Pachystachys coccinea, 272 Paederia scandens, 85 Paeonia obovata var. willmottiae, 47 Pagamea guianensis, 278 Palicourea guianensis, 278 Palmeria, 468 Paloue guianensis, 279 — — var. japonicus, 75 quinquefolius, 519, 546 Pace ain carolinianum, 513, 546 Pandanaceae, Studies in Malesian, 19. New Species of Freycinetia and Pandanus Bo Malesia and Southeast Asia, 309- 24 ee from Malesia and Southeast Asia, New Species of Freycinetia and (Studies in Malesian eee naceae, 19), 309-324 Pandanus, 309, 312, 322 — sect. eae ees 312 — sect. Asterodontia, 321 — sect. Cheilostigma, 317 — sect. Markgravidendron, 315, 317 — sect. Rykia, 317, 321, 324 — albifrons, 317-320 — angiensis, 313 Pandanus bicornis, 321, 324 — pseudosyncarpus, 313 — regalis, 322, 323 — setistylus, 313, 315 biaceae) from, 611-61 Parathelypteris nipponica, 24 Pariana campestris, 279 Parietaria debilis var. micrantha, 44 — micrantha, 44 Parinari ee 279 — montan _ rodolphii, 279 Paris polyphylla, 96 Parivoa ag 279 — tome Parnassia er a2 — wightiana, 52 Barthenium. 149 Parthenocissus henryana, 70 70 280 —_ Sabenocs. 534, 544 64 1983] INDEX 655 Passifloraceae, 257, 279, 280, 284, 534, Phryma leptostachya var. asiatica, 85 544, 545 2.85 rymaceae, 9, Passoura guianensis, 280 Phyllanthus, 275 Patabea coccinea, 2 — brasiliensis, 265 Patima guianensis, 280 — epiphyllanthus, 525, 544 Patrinia angustifolia, 89 — guyanensis, 275 — monandra, 89 Phymosia abutiloides, 531, 544 samen capreolata, 268 Physalis alkekengi var. franchetu, 83 Paulownia, 9 Physaria, 491-493, 496 Pay ete guianensis, 280 — acutifolia, 492, 493 Pedicularis holocalyx, 84 — alpestris, 492 — resupinata, 84 — chambersii, 492 — torta, — condensata, 493 Pekea butirosa, 280 — didymocarpa, 492, 493 — tuberculosa, 280 — — var. integrifolia, 492 Peltandra sagittaefolia, 514 — floribunda, 492, 493 — virginica, 514, 544 — geyeri, 492, 496 Peltophorum brasiliense, 528 — lepidota, 492, 493, 496 Pennisetum alopecuroides, 94, 127 — newberryi, 492 Penthorum, 9 — obcordata, 491, 492, 495, 496 People’s Republic of China, The 1980 Sino- — oregona, 493 merican Botanical Expedition to West- —Phytolacca, 46 ern Hubei Province, 1-103 — acinosa, 4 Peracarpa, 9 — polyandra, 46 Perama hirsuta, 280 Phytolaccaceae, 46 Peranema cathode, 27, 124 Picea wilsonil, Peranemacea Picramnia, 287 Perebea a isik 280 — guianensis, 287 Periploca, 257 — tariri — umbellata, 257 Picrasma a quassioides, 126 Peronosporaceae, 302 Picris hieracioides subsp. japonica, 93 Perrottetia racemosa, 66 Picrodendron baccatum, 525, 544 Persea borbonia, 528, 544 — macrocarpum, 525 — palustris, 329 Pieris formosa, 78 Pertya sinensis, 93 Pilea fasciata, 44 Peschira echinata, 286 — japonica, 44 Peucedanum praeruptorum, 76 — martinii, 44 Phegopteris decursive- ae 24 — plataniflora, 44 — polypodioides, 2 — sinofasciata, 44 Philadelphus incanus, 52 Pimpinella arguta, 76 — inodorus, 537, 545 — diversifolia, 76 — sericanthus, 52 — — var. stolonifera, 76 Phillyrea, 625 Pinus, 507 Phleum alpinum, 94 — armandii, 6, 39, 113 Phlomis umbrosa, 83 — clausa, 186, 222 Phlox hoodii, 502 — elliottii, 186, 209 Phoebe neurantha, 50, 125 — massoniana, 6, 12, 40 Phoradendron rubrum, 530, 545 — — var. henryi, 40 Photinia amphidoxa, 59 — palustris, 187 — beauverdiana, Piparea dentata, 280 — — var. notabilis, 56 Piper guianensis, 282 — parvifolia, 5 — humistratum, 282 — villosa, 56 Piperacea e, 282 Phryma, 9 Piptocalyx, 447, 466, 468-470 656 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Piptocalyx macrurus, 449 Polygalaceae, 61, 277, 280, 281, 295 — moorei, 447-453, 455, 457, 458. 461, — Polygonaceae, 9, 45, 125, 289, 534, 545, 463-465, 467 546 Piptocarpha triflora, 269 Polygonatum cyrtonema, 96 Piratinera, 28 — odoratum, 96 — guianensis, 269, 280 — sibiricum, 96 Pirigara hexapetala, 280 — verticillatum, 96 — tetrapetala, 280 — zanlanscianense, 96 Piripea palustris, 280 Polygonum ee var. sinense, 45 Piriqueta villosa, 280 — caespitosum, 45, 12 — viscosa, 2 — ciliinerv Pisonia, 27 — cuspidatum, 45 — gulanensis 270 — — hydropiper, 45, 125 Pistacia simaruba, 521 — multiflorum, Pithecellobium ee 529, 530,546 — — var. ciliinerve, 45 — circinalis, 530 — muricatum, 45, 125 Pithecoctenium echinatum, 262 — persicaria, 45, 125 Pithecolobium mucronatum, 530 — pilosum, 45 Pittosporaceae, 53 — strigosum var. muricatum, 45, 125 Pittosporum glabratum var. neriifolium, — thunbergii, 45, 125 53 — uvifera, 534 — rehderianum, 53 Polypodiaceae, 36, 124 — truncatum, 53 Polypodioides pseudo-amoenum, 36, 37 Pitu a gulianensis, 280 Polypodium capense, 367 Platanaceae, 534, 544 — pseudo-amoenum, 36 Platanthera hologlottis, 98 Polystachya concreta, 517, 545 Platanus occidentalis, 534, 544 — minuta, 517 Plates of Aublet’s Histoire des Plantes de — Polystichum, 6 la Guiane Francoise, The, 255-292 _ roaches, 330 Platycarya strobilacea, 41 — capens Pleione bulbocodioides, 98 — ae 32 Pleurosoriopsidaceae, 9, 25 _ oe any , 32 Pleurosoriopsis, 9 — ero 39 — makinoil, 25 —_ ies lie 32, 33, 124 Pleurospermum giraldii, 77 — longiaristatum, 33, 34 Plumeria obtusa, 519, 545 — makinoii, 32, 33 _ are 519, 545 — neolobatum, 34 74 — pteropt _ a is, 94 _ shennongense, 33-35 Podophyllum peltatum, 520, 543 — submite, 34 Podostemonaceae, 27 — tsus-simense, 36 Polybotrya arfakensis, 345 Populus, 33 Polyctenium, 496, 508, 509 — adenopoda, 115 — bisulcatum, 508 — balsamifera, 536 — fremonti, 508 davidiana, 4 — glabellum, 508 — deltoides, 53 — williamsiae, 508, 509 — heterophylla, 115, 536, 543 Polygala arillata, 61 — lasiocarpa, 40 — cruci — wilsonil, 4 — cymosa, 329 Porana racemosa, 82 — tatarinown, 62 Poraqueiba guianensis, 281 — timoutou, 280 Posoqueria longiflora, 281 — violacea, 281 Possira, 281 — wattersii, 62 — arborescens, 281 1983] Potalia amara, 281 Potaliaceae, 281 Potentilla fruticosa, 56 — leuconota, 56 Pourouma guianensis, 281 Pouteria guianensis, 281 Pratia nummularia, 90 Prenanthes tatarinowll, 93 Pringleophytum, 130, 136 — lanceolatum, 130, 136, 139 Pronephrium penangiana, 24 Proteaceae, 28 Protium aracouchini, 271 1 — heptaphyllum, 271 Prunus, 507 — brachypoda, 56 — salicina, 56 — virginiana, 534, 535, 543 Pseudima frutescens, 284 Pseudocyclosorus tsoi, 24, 124 Pseudocystopteris atkinsonu, 24 24 Pseudotsuga menziesii, 480, 502 65, Pseudowintera, 161, | 166 Psidium aromaticum, 281 — grandiflorum, 281 Psychotria, 264, 287 a, 287 poda, 287 — Bicone sophia, 287 — capitata subsp. amplifolia, 277 283 — latifolia, 2 j — violacea, O77, 281, 287 Ptelea trifoliata, 536, 545 Pteridaceae, 18, 124 INDEX Pteridium revolutum, 18 Pteridophyta, 17, 123 Pteris excelsa, 18 — nervosa, 18 — vittata, | — wallichiana, 18, 12 Pterocarpus Beanies 244 rohru, 260 6 Pyracantha crenulata, 56, 125 — fortuneana, 56, 12 Pyrola ae 78 Pyrolaceae, 7 Pyrrosia ae 37 — calvata, 3 — caudifrons, 22, 37 — drakeana Pyrus pyrifolia, 56 — serrulata, 56 quale caerulea, 281 rosea, 281 Ourpeva aa panari, 281 — scandens, 28 Se eabea. guianensis, 282 Quassia simaruba, 285 aie ete 282 oO < a Nn Wn 7) Lea) wn B Ww — marilandica, 525, 526, 543 658 Quercus myrsinifolia, 43 — nigra, 526, 543 — — var. 8, 525, 526 — oxyodon, 43 _ aes 525, 526, 543 526 _ ee 526, 543 Quiinaceae, 282, 289 Rabdosia excisoides, 83 — nervosa, 83 Racaria sylvatica, 282 Racoubea guianensis, 282 Ramotha, 426 Ranalisma, 385, 406 Randia, 149 Ranunculaceae, 9, 46, Ranunculus, 413 Rapanea guianensis, 282 125, 385 Raputia aromatic 282 Rautaneni Recheteinera incarnata, 261 Reineckea, 9 Remirea raciatligias 282 Revision of the Genus Alsophila (Cyathea- ceae) in the Americas, A, 333-382 Reynosia latifolia, 535 — iteinophyllus, 69 — leptophyllus, 69 — utilis, 69 Rhapis excelsa, Anatomy of the Palm, IX. Xylem Structure of the Leaf Insertion, -609 Rhapis, 599, 608 — excelsa, 599-609 Rhexia alifanus, 329 — villosa, 282 JOURNAL OF THE ARNOLD ARBORETUM [vOL. 64 Rhizophora mangle, 535, 545 Rhizophoraceae, 265, 535, 545 Rhododendron argyrophyllum subsp. hypoglaucum, 78 _~ fargesil, 78 — micranthum, 79 — simsil, — sutchuenense, 79 — viscosum var. aemulans, 524, 544 Rhus chinensis, — glabra, 518, 546 Rhynchanthera sasaiiios 276 Rhynchospermum verticillatum, 93 Rhynchospora eae Riana guianen Ribes, 52 — acuminatum, 52 — aureum, 507 — fasciculatum var. chinense, 52 — rlana, 282 Rivea cordata, 488, 489 Ro ee hispida, 530, 546 cou, 282 = pariaeose: 282, 283 RoGers, GEORGE K. The Genera of Alis- mataceae in the Southeastern United States, 383-420 ROLLINS, REED C. Studies in the Cruciferae of Western North America, 491-510 — gulanensis, 283 eae indica, 57 Rosa, 57 — re iae, 56 — banksiopsis, 56 — helenae, 57 1983] Rosa henryi, 57, 125 rta oe a 125, 535, 543 Rotala indica, 73, 126 Rouhamon guianensis, 283 Roupala montana, 28 Rubia cordifolia, 85 Rubiaceae: New Species of Timonius from Papuasia, 611-618 Rubiaceae, 9, 85, 127, 257, 259, 261, 264, 265, 267, 269; 271, 273, 274,277,278, 280, 281, 283-288, 535, 543, 544, 546, 27 — tribe Guettardeae, 611 Rubus amphidasys, 57, 125 — flosculosus, 57 — — var. hubeiensis, 58 — setchuenensis, 58, 125 — simplex, Nanas oe a22 Ruellia, 284 Rutaceae, 61. i 269,270, 276,282, 287, Sabatia macrophylla, 329 Sabiaceae, 68, 126 aera een 284 a, 284 Sina 383, 384, 386, 393, 405, 409- 416 — subg. Lophotocarpus, 412 — subg. Sagittaria, 412, 414, 415 Sagonea palustris, 2 Salacia, 273, 274, 286 — guyanensis, 273 — scandens, 288 Salicaceae, 40, 536, 543 Salix, 6, | — fargesii — — var. kansuensis, 40 Salmea petrobioides, 522, 544 Salvia maximowicziana, 83 INDEX Salvinia auriculata, 284 ea) 2. S = ro) O oO x) oO BR — schweriniana, 87 Samolus ebracteatus, 540 SAMPSON, and Peter K. ENDRESS. Floral Structure ane ae ree of the Trimeniaceae (Laurales), 447-473 Sanguinaria canadensis, 330 Sanicula orthacantha, 77 Santalaceae, 9 Saouari glabra, 284 — villos a, 284 de 268, 274, 282, 284, 286, 288, 290 Sapindus arborescens, 284 — frutescens, 284 Sapotaceae, 265, 281, Sarcococca eer var. humilis, 62, 126 — ruscifolia, 63 _ esbael, 537, 545 — ee - urpurea, 537, 545 Sarraceniasas 537, 545 Saruma, 9 — aes 44 Sassafras, 9 albidum var. molle, 528, 543 Saurarsten e, 40 Saururus cernuus, 115 | < 3 g. 5 ee bel) & ,9 Sauvagesia adima, 284 erecta, 28 sprengelil Saxifraga flabellifolia, 53 — giraldiana var. hupehensis, 53 oe — glaucescens, 49 — pubescens, 125 Schisandraceae, 9, 49, ae Schizachyrium, 173, | — gracile, 174 660 Schizmaxon, 426 m, 53 Schlegelia violacea, 262 Schultesia guianensis, 269 Scirpus lushanensis, 94 Scopolia sinensis, ra Scrophularia henry , 84 Seto PaMlaniAagens, 9, 34, 127, 260, 266, 275, 280 eee aizoon, 51 _ amplibracteatum, 5] — erythrostictum, 5] — filipes, — telephium ee alboroseum, 51 — verticillatum, Selaginella inv ave ens, 17 — labordei, — moellendorfli, 17 — nipponica, 17 — uncinata, 17 Selaginellaceae, 17 Senapea, 257 — gulanensis, 284 94 Sideroxylon Poetiiseinun: 537 Siegesbeckia pubescens, 93 Simaba guianenss, 260, 285 Simarouba, 2 285 — amara, ae Simaroubaceae, | 26, 260, Simira tinctoria, 285 Simmon 4] Stharunidinariannitida, 4,68 57 Ween Hubei Province, public of China, The Sinofranchetia, 9 Peo 1980, 285, 287 1-103 3 erican Botaitical Expedition to ple’s Re- JOURNAL OF THE ARNOLD ARBORETUM Sinofranchetia chinensis, 48 Sipanea pratensis, 285 Siparuna, 469 — guianensis, 2 Sloanea, 258, 267, 281, 537 — emarginata, 536, be 545 — guianensis, 258, — sinemariensis, i Smelowskia, 498, 508 ae _ ee 5] 7 — riparia var. acuminata, 97 — scobinicaulis, 97 — stans, 97 — tamnoides, 517, 543 Solanaceae, 83, 261, 285 Palani aubleti, 261 4 4 — pittosporifolium, 84 — ania — tegore, 285 Sol idago decaniens: 93, Soramia gulianensis, Sorbaria arborea, 58 Sorbus, 6, 9 — alnifohia, 38 — zahlbruckneri, 59 Sorghum Saag 207 — argenteum, 254 _ brachsstacum, 234 — elhotti, 254 [VOL. 64 1983] INDEX 661 Sorghum ie 235 CHANG, Z. CHENG, T. - Dub _ey, S. A. — hebmanni E, Y. IN,Q. Y.LiandJ.L. Luteyn. Sapa eee 223 The 1980 ee Botanical Ex- — virginicum, 224 pedition to Western Hubei Province Souroubea guianensis, 285 Southeastern United States, Indexes to Pa- pers | to 100 Published as Parts of the Generic Flora of the, 547-563 Southeastern United States, The Genera of Alismataceae in the, 383-42 Southeastern United States, The Genera of Burmanniaceae in the, 293-307 Southeastern United States, The Genera of Menyanthaceae in the, 431-445 Southeastern United a The Xyrida- ceae in the, 421-42 Spathodea bracteosa, a Spermacoce for Two Species of Borreria m New Guinea, A New Name in. 627, 628 Spermacoce, 285, 627, 628 — sect. Borreria, 627 285 — caerulescens, — hexangularis, — lanceolata, 628 — sexangularis, 285 SPERRY, JOHN S., and MARTIN H. ZIMMER- MANN. Anatomy of the Palm Rhapis ex- celsa, IX. Xylem Structure of the Leaf Sphenomeris chusana, 18 Spigelia, 530 — guianensis, 276 — marilandica, 530, 545 Spiraea chinensis, 59 — japonica var. acuminata, 59 — veitchu, 5 SPONGBERG, THOLOMEW, Metasequola Present Status in Central China, 28 STEPHEN A., Bruce BArR- and Davip E. BouFForRD. glyptostroboides—Its 105- SPONGBERG, S. A., 8. C. Sun, Y. C. TANG, X. Wan, T. SS. Yinc, B. Bar- THOLOMEW, D. E. Bourrorp, A. L. People’s Republic of China, 1-103 Stachyuraceae, 72 Stachyurus chinensis, 72 — — var. latus, 72 Staphylea bumalda, 66 — holocarpa, 66 Staphyleaceae, 66 Srapces, G. W., and D. F. Austin. Addi- tions and Changes in the Neotropical Convolvulaceae— Notes on Merremia, Operculina, and Turbina, 483-489 Staples, GeEorGE W., and RicHArRD A. Howarp. The Modern Names for Catesby’s Plants, 511-546 Stemmodontia bahamensis, 523 Stemodia pusilla, 275 Stenandrium, 129, 130 — subg. prea aa 130 Stenoloma chusa 8 Stenosolen heterophyllus, 286 372 Sterculiaceae, 263. 272, 537, 546 Stewartia malacodendron, 538, 546 — sinensis, 7 Stipa, 174 Stokesia laevis, 329 STONE, BENJAMIN C. Studies in Malesian Pandanaceae, 19. New Species of Frey- cinetia and Pandanus from Malesia and Southeast Asia, 309-324 Stranvaesia amphidoxa, 59 — davidiana, 59 var. undulata, 60 Streptolirion, 9 ubile, 95 Streptopus obtusatus, 97 Struthiopteris eburnea, 27 Strychnos, 2 — guianensis, 2 Stryphnodendron guianensis, 276 Studies in Malesian Pandanaceae, 19. New Species of Freycinetia and Pandanus from Malesia and Southeast Asia, 309- 324 Studies in the Cruciferae of Western North America, 491-510 Stylophorum, : — lasiocarpum, 50 Stylosanthes eae, 289 662 Styracaceae, 80, 538, 544 Styrax bodinieri, 115 _ oe 115 Sun, S. C., Y. C. TANG Xx. WAN, YING, i. Bain eae D. E. BowEonn A. L. CHANG, Z. CHENG, T. R. DuDLey, 5, Y. X. Jin, Q. Y. A. ShONeBERS: pedition to Western Hubei Province, People’s eee of China, 1-103 Swartzia, 281, 288 — alatum, a — arborescens, 281 — panacoco, 282, 283 — tom a, 28:2 Swertia ca pases 81 — punic an ee | Swietenia Saree 531, 545 Sycopsis, — sinensis, 53 Symphonia globulifera, 277 Symplocaceae, 80, 538, 543 , 514, 545 80 Synopsis of the Chinese ee of Asarum (Aristolochiaceae), A, Synurus deltoides, 93 Syringa, 625 — reflexa, 81 Systematics of Holographis (Acanthaceae), 129- cay of the oes virginicus mplex (Gramineae), 171-254 Tabebuia aquatilis, 262 — bahamensis, 520, 543 — pentaphylla, 520 Tabernaemontana echinata, 286 — echites, 519 Taccaceae, 295 Tachia guianensis, 286 Tachibota guianensis, 286 Tachigali paniculata, 286 a, 286 — campestris, 286 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Talisia guianensis, 286 — spicata, Tampoa guianensis, 286 TANG, Y. J. X. Wan, T. S. YING, B. BARTHOLOMEW, D. E. Bourrorp, A. L. CHANG, Z. CHENG, T. R. Dubey, S. A. He, Y. X. Jin, Q. Y. Li, J. L. Luteyn, S. A. SPONGBERG, and S. C. Sun. The 1980 Sino-American Botanical Expedi- tion to Western Hubei Province, Peo- ple’s Republic of China, 1-103 Tanibouca guianensis, 286 Taonabo dentata, 286 — punctata, 287 Tapirira guianensis, 287 Tapura gianensi® 287 Taralea, 287 — oppositifolia, 287 Tariri, 287 — guianensis, 287 Taxaceae, 39 Taxodiaceae, 40, ra 543 Taxodium, 107, 115 — distichum, ih et a Taxus chinensis, 39, Terminalia a oe _ ea, 279 serena dentata, 286 — ctata, 287 Pet erro cin 9, 46 Tetracentron, 8, ine — ruticarpum, 61 Tetragastris oe 271 Tetrapilus, Tetrasugma an 70, 126 — obtectum Thalassia testudinum, 515, 544 1983] Thalictrum przewalskii, 48 — robustum, 48 — uncinulatum, 48 Theaceae, 71, 126, oe 287, 538, 543, 546 , 263 THESphrastaceae. 538, 544 Thespesia populnea, 531 Thismia, 294 — sect. Rodwaya, 294 Thismiaceae, 294, 295 Thladiantha henryi, 89 Thysanobotrya, 345 — arfakensis, 345 Tiarella polyphylla, 53 Tigarea ie 287, 288 — dentata, 288 Tilia chinensis, 70 — oliveri var. cinerascens, 70 Tiliaceae, 70, Tillandsia oalbisiana, 514, 545 Timonius eae i om Papuasia, New _ pub supalus. 612, 613, a var. pubescens, 614 — var. pubistipulus, so 614, 616 _ pee 616 ea 616 — cone s, 612 _— aan 616, 617 — trichanthus, 616, 617 INDEX Timonius zuckianus, 611, 612 Tinospora sagittata, 49 Tococa, 27 — guianensis, 288 Tocoyena longiflora, 288 Tonina fluviatilis, 288 Tontanea guianensis, 288 — scandens, 288 atica, 288 Toulicia eee 288 Tounatea, 289 — guianensis, 288, Touroulia eaenca 289 Tovomita guianensis, 289 Trachelospermum, 9 Tragia, 257 O57 Trichipteris, 333, 334 — dichromatolepis, 352 72 — tryonorum 7 Trifolium guianense, 289 Trigonia laevis, 289 — villosa, 289 Trigoniaceae, 289 Trillium, 330 — catesbaei, 516, 543 — maculatum, 516, 543 — sessile, 51 Trimenia, 447, 466, 468, 470 — neocaledonica, 447-453, 455, 457, 458, 461, 463, 464, 466, 467, 470 — papuana, 447-467, 470 — weinmanniifolia, 447, 449-451, 466, 467 Trimeniaceae (Laurales), Floral Structure and Relationships of the, 4 3 Trimeniaceae, 447-473 Triosteum, 9 — himalayanum, 87 Triplaris americana, 289 — surinamensis, 289 Triplostegia glandulifera, 89 Tripterella, 30 455, 664 JOURNAL OF THE ARNOLD ARBORETUM [VOL. 64 Tripterospermum affine, 81, 127 Viburnum formosanum, 87 Triuridaceae, 2 — hupehense subsp. hupehense, 87, 88 Tsuga chinensis, 8, 40 — — subsp. septentrionale, 87 Tupistra, — ichangense var. atratocarpum, 88 — chinensis, 97 — — var. ichangense, 87 Turbina, Notes on Merremia, Operculina, — lobophyllum, 87, 88 and; Additions and Changes 1n the Neo- = — ovalifolium, 87, 88 tropical Convolvulaceae, 483-489 — plicatum : tomentosum, 88 Turbina, 488 — propinquum, — abutiloides, 488, 489 — Sa iochullin. 88 — cordata, 488, 489 — sargenti, 88 Turnera frutescens, 289 — setigerum, 88 — gulianensis, 289 — sympodiale, 88 — rupestris, 289 — utile, 89 — — var. frutescens, 289 — veltchi, 89 Turneraceae, 280, 289 Vicia cracca, 61 Two Unusual Chionanthus Species from — pseudo-orobus, 61 Borneo and the Position of Myxopyrum Viguiera, 145, 149 in the Oleaceae, 619-626 Villarsia, 431, 432, 436, 442 Vinca lutea, 511, 519, 545 Ulmaceae, 9, 43 Viola grypoceras var. pubescens, 72 mus, 3 — hybanthus, 28 — americana, | 15 — itoubou, 289 — multinervis, 115 — prionantha, 72 Umbelliferae. 9, 75 — stewardiana, 72 Uncaria, 278 Violaceae, 72, 266, 280, 282, 289 — gulanensis, 278 Virola sebifera, 289 Uniola, 174 Viscum sg ae 530, 546 aniculata, 515, 543 — rubru 545 Urechites lutea, 519, 545 Vismia eee a 27) Urucaceae, 44, 125 — latifolia, 271 Uvaria, 291 — sessilifolia, 271 — zeylanica, 291 Vitaceae, 69, 126, 540, 544 Vitis. 115 Vaccinium sect. Hugeria,. 9 — betulifoli: 70 — en — Hees var. pagnucci, 70 aponicum var. sinicum, 79 Vochy, 2 Wie tanaeere, 8 — oe 289, 290 Vanilla mexicana, 517, 546 Vochysia, 289 — planifolia, 517 — gulanensis, 289 Vantanea guianensis, 289 — speciosa, 290 Vatairea guianensis, 289 Vochysiaceae, 281, 289 Veratrum oblongum, 97 Votomita guianensis, 290 Verbenaceae, 82, 270, 274, 286, 540.544 Vouacapoua, 290 Veronicastrum a 84, 127 — americana, 290 Viburnum, 9, 115, 116 Vouapa, 278, 290 — betulifolium, 8&7 — bifolia, 278, 290 — cylindricum, 87, 12 — simira, 2 — erosum subsp. ichangense var. atrato- Vouarana EUIANENSIS, 290 carpum, 88 Voyara, 2 — erubescens, 87 — montana, 290 — flavescens, 87 Voyria, 295, 302 — foetidum, 127 — caerulea, 290 — — var. rectangulatum, 87 — rosea, 290 1983] Vulpia, 174 Wan, J. X., T.S. Yinc, B. BARTHOLOMEW, D. E. Bourrorp, A. L. CHANG, Z. CHENG T. R. Duprey, S. A. He, Y. X. Jin, Q. Y. Li, J. L. Lutreyn, S. A. SPONGBERG, S.C. Sun, and Y. C. TANG. The 198 Sino-American Botanical eect to Western Hubei Province, People’s Re- public of oe 1-103 Wapato, 41 Waria ae 291 in, Water-plantain Family, 383 Wedelia bahamensis, 523, 544 Weigela, 9 — japonica var. sinica, 89 Winteraceae: Wood Anatomy of Belliolum and a Note on Flowering, 161-169 Winteraceae, 161, 165, 166, 168 Worrorp, B. EuGene. A New Lindera (Lauraceae) from North America, 325- 331 Wood Anatomy of Belliolum (Wintera- ceae) and a Note on Flowering, 161 Woop, CARROLL E., Jr. Indexes to Papers ! to 100 Published as Parts of the Ge- neric Flora of the Southeastern United States, 547-5 Woop, Carro_t E., Jr. The Genera of Burmanniaceae in the Southeastern United States, 293-307 Woop, CARROLL E., Jr. The Genera of yanthaceae in the Southeastern ee States, 431-445 Woodsia polystichoides, 27 Woodsiaceae, 27, 124 Woodwardia unigemmata, 27, 12 — virginica, 33 Xanthosoma, 514, 544 Xanthoxylum hermaphroditum, 269 Ximenia americana, 27 Xiphidium coeruleum, 291 Xuris, 426 Xylem Structure of the Leaf Insertion: a oe of the Palm Rhapis excelsa, IX, ena epiphyllanthus, Xylopia frutescens, 29 INDEX 665 Xymalos, 469 Xyridaceae in the Southeastern United 429 Xyridaceae, 291, 421-429 Myris, 330, 422-427 — sect. Euxyris, 426 — sect. Nematopus, 426, 427 — sect. Pomatoxyris, 426 — sect. Xyris, 426, 427 — americana, 291 Xyroides, 426 YANG, CHUN-SHU, CHING-YUNG Cuenc. A Synopsis of the Chinese Species of Asarum (Aristolochiaceae), 565-597 Yellow-eyed grass, 42 Yellow- ae Grass Family, 42 YING, S.. B. seco nea ; Bourrorp, A. L. CH , 2. CHENG, T. R. Dubey, S. A. aN y. xX. Jr Li, J. L. Luteyn,S. A. — BERG Sun, Y. C. TANG, and J. X . The 1980 ee American epee Expedi tion to Western Hubei Province, Peo- ple’s Republic of China, 1-103 Youngia denticulata Yucca, 149, 157 Zanthoxylum armatum, 61 — clava-herculis, 536, 543 — dissitum, 61 — martinicense, 536 — undulatifolium, 61 Zelkova sinica, 43 ee atamasco, 514, 546 uxine, 302 Fisadenus glaberrimus, 329 ZIMMERMANN, MARTIN H., and JOHN S. SPERRY. Anatomy of the Palm Rhapis Xylem Structure of the Leaf -609 Zingiber mioga, 98, 127 Zingiberaceae, 98, 127 Zizania, 397 Zygogynum, 161, 165, 166 — bicolor, 166 — pomiferum, 166 = JOURNAL oF tre ARNOLD ARBORETUM HARVARD UNIVERSITY VOLUME 64 1983 Dates of Issue No No. No No . | (pp. 1-169) issued 11 January 1983. 2 (pp. 171-331) issued 8 April 1983. . 3 (pp. 333-490) issued 19 July 1983. . 4 (pp. 491-665) issued 20 October 1983. Contents of Volume 64 The 1980 Sino-American Botanical Expedition to Western Hubei Province, People’s Republic of China B. BARTHOLOMEW, D. E. BOUFFORD, A. L. CHANG, Z. CHENG, T. R. DubLey, S. A. HE, Y. X. JIN, Q. Y. Li, J. L. LUTEYN, S. A. SPONGBERG, S.C. Sun, Y. C. TANG, J. X. WAN, AND T.S. YING ............ Metasequoia glyptostroboides—Its Present Status in Central China. BRUCE BARTHOLOMEW, DAVID E. BOUFFORD, AND STEPHEN A. SPONGBERG® 266.05. bsi Grewal vas Ye eGo b de tn cs eo eine ee ae Systematics of Holographis (Acanthaccae). THOMAS F. DANIEL .......0.0.. 000000 eee eee t tenes Wood Anatomy of Bellio/um (Winteraceae) and a Note on Flowering. SHERWIN CARLQUIST .....0.00000 0000 cee cece eee e eens Systematics of the Andropogon virginicus Complex (Gramineae). CHRISTOPHER S. CAMPBELL .......0.0000000 0c cece eens The Plates of Aublet’s Histoire des Plantes de la Guiane Francoise. RICHARD A. HOWARD ............ 0000 cece eee tenet tenes The Genera of Burmanniaceae in the Southeastern United States. CARROLL E. Woon, JR. 0. een Studies in Malesian Pandanaceae, 19. New Species of Freycinetia and Pandanus from Malesia and Southeast Asia. BENJAMIN C. STONE ..... 000000000 cee beens A New Lindera (Lauraceae) from North America. B. EUGENE WOFFORD .............0 000 ccc cee ee eee ae A Revision of the Genus 4A/sophila (Cyatheaceae) in the Americas. DAVID S. CONANT 2.000000 cee es The Genera of Alismataceae in the Southeastern United States. EORGE K. ROGERS ........0. 0.0. ccc eee een es The Xyridaceae in the Southeastern United States. ROBERT KRAL 2.0... eee eee nes The Genera of Menyanthaceae in the Southeastern United States. CARROLL E. WOOD, JR. 00. eens Floral Structure and Relationships of the Trimeniaceae (Laurales). PETER K. ENDRESS AND F. B. SAMPSON .............00 000 ce eeue Inbreeding Depression in Metasequoia. JOUMRUSER: 29.08.2443503064.525 855555 2cneetae eis tee essen ae 1963 Additions and Changes in the Neotropical Convolvulaceae— Notes on Merremia, Operculina, and Turbina. DD: Fe Asti AN Gr Ws STAPLES: 4b creole sede oe eee 483 Studies in the Cruciferae of Western North America. REED C. ROLLINS ....... 0000 eee eee tens 49] The Modern Names for Catesby’s Plants. RICHARD A. HOWARD AND GEORGE W. STAPLES ...........00--- S11 Indexes to Papers | to 100 Published as Parts of the Generic Flora of the Southeastern United States. CARROLL E. Woon, JR. 1. ee 547 A Synopsis of the Chinese Species of Asarum (Aristolochiaceae). CHING-YUNG CHENG AND CHUN-SHU YANG ..........0002000005 565 Anatomy of the Palm Rhapis excelsa, 1X. Xylem Structure of the Leaf Insertion. ARTIN H. ZIMMERMANN AND JOHN S. SPERRY ...........000-- 599 New Species of 7imonius (Rubiaceae) from Papuasia. STEVEN P. DARWIN 2.0... tent eenee 611 Two Unusual Chionanthus Species from Borneo and the Position of Myxopyrum in the Oleaceae. RUTH IRIEW «oc hate Ss a ea AWS ed dn Ka Re dE EE 619 A New Name in Spermacoce for Two Species of Borreria from New Guinea. OSTAMGIDEON? oe.ctsgrne eae s 23 aero i aes ee O27 NOTICE Starting with Volume 65 (1984), the cost per volume of the Journal of the Arnold Arboretum will be $50 for domestic and $55 for foreign subscriptions. Journal of the Arnold Arboretum October, 1983 CONTENTS OF VOLUME 64, NUMBER 4 Studies in the Cruciferae of Western North America. BBE. TRIO IIIS sic dis iwrushyiktcaice estat nes earatapuvcecdodoin At Gadi arth todas abet, evel WXons vse 491 The Modern Names for Catesby’s Plants. RICHARD A. HOWARD AND GEORGE W. STAPLES ................ 51 Indexes to Papers | to 100 Published as Parts of the Generic Flora of the Southeastern United States. CARROLE Bo WOOD IR. sess ig cad dre soder-iniien woh anhom ora: dqewbeaageqoingans 547 A Synopsis of the Chinese Species of Asarum (Aristolochiaceae). CHING-YUNG CHENG AND CHUN-SHU YANG .............-00-4.. 565 Anatomy of the Palm Rhapis excelsa, IX. Xylem Structure of the Leaf Insertion. MARTIN H. ZIMMERMANN AND JOHN S. SPERRY ................ 599 New Species of 77monius (Rubiaceae) from Papuasia. PUSVeI WAR MIT ¢ neMeswaewes i kwG res audbaseuh sanpubiimeea 611 Two Unusual Chionanthus Species from Borneo and the Position of Myxopyrum in the Oleaceae. TOE ESO scan xy ir bey: sh ecoema gree at bind cy Se aden ae Sea. cela wag etn aaa echoed 619 A New Name in Spermacoce for Two Species of Borreria from New Guinea. OSIAISISEGN ope ec eens wosde- ver acuins die owrtiatniend wader eda sod ba ne 627 Volume 64, Number 3, including pages 333-490, was issued July 19, 1983.