JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Che New England Botanical Club, Inc. 22 Divinity Avenue, Cambridge, Massachusetts 02138 RAODORA GORDON P. DEWOLF, JR., Editor-in-Chief Associate Editors DAVID S. BARRINGTON RICHARD A. FRALICK . LINN BOGLE GERALD J. GASTONY DAVID E. BOUFFORD C. BARRE HEI QUIST CHRISTOPHER S. CAMPBELL MICHAEL W. LEFOR WILLIAM D. COUNTRYMAN ROBERT T. WILCE GARRETT E. CROW RHODORA (ISSN 0 0035-4902). Published four times a year (Jan i ns hcg ee ober) by The New England Botanical fee », 23 ivinity Av , MA 02138 and spr by Allen ‘ Ine. 1041 New Hampshire St., Lawrence, KS 66044. Si Ease oa é A and at additio tio} onal mang offices. RE DOR North pg eteoges Scienti ‘fic papers and: notes relating to this ae area an floristically related areas, an red w aa) plorracigartay in their oader implications, will be consid- ORA assesses page SUBSC j one $40.00 00 per calendar year, net, posta, i in funds pavable at par in the Unie States currency at Boston. Remittances Payal to RHODO Send to Fi reasurer, we Divinity Ave, Cam- Tbodora URNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 95 January 1993 No. 881 RHODORA, Vol. 95, No. 881, pp. 1-5, 1993 SCHIEDEELLA ROMEROANA (ORCHIDACEAE, SPIRANTHINAE), A NEW AND INTERESTING SPECIES FROM MEXICO Dariusz L. SZLACHETKO ABSTRACT Schiedeella romeroana Szlach., sp. nov., is newly described. It is distinguished by its lack of both rostellum and viscidium. The lip is distinctly divided into a rectangular hypochile with reduced basal auricles and a rounded epichile. It is compared with S. valerioi (Ames & Schweinf.) Szlach. & Scheviak, S. amesiana Garay and S. /laveana (Lindl.) Schlecht. Key Words: Schiedeella, Orchidaceae, new species, Mexico The genus Schiedeella Schlecht. embraces 10 species found in Mesoamerica—from the southern states of the U.S.A. to El Sal- vador and Costa Rica. The principal characteristics of these spe- cies are the structure of the flower segments and gynostemium. In the majority of these species the flowers are very small, with the lip set on a short, narrow claw, divided more or less distinctly into hypochile and epichile, with basal auricles of various size and shapes. The column foot is short and greatly reduced, the rostellum with a narrow base is linear-lanceolate, and after re- moving the semi-sheath-like viscidium it remains subulate with 1 or 3 indentations at the apex. Floral and cauline bracts are herbaceous (Szlachetko, 1992). While conducting studies on the subtribe Spiranthinae for the Flora Mesoamericana, I found a new, distinctive plant among the herbarium material from the Mexican state of Oaxaca. It differs from those previously described in its total lack of both rostellum 1 MISSOUR! BOTANICAL JUN 08 1993 y Rhodora [Vol. 95 and viscidium. The middle stigma lobe, which usually becomes the rostellum in orchids, is here membranous and transparent. Together with the terminal edges of the clinandrium, it stretches between the fairly stiff, resilient protruberances of the lateral stig- ma lobes. It does not thus constitute a barrier for the pollen mass, nor does it prevent autogamy. In this new species of Schiedeella, the pollinia protrude distinctly above the thin middle lobe of the stigma. Schiedeella romeroana Szlach., sp. nov. (Figure 1). Species Schie- deellae llaveanae et S. valerioi similis, sed ab omnibus con- generibus rostello et viscidio nullo iam diversa. Roots 5, up to 4 cm long, about | cm in diameter, clustered, fleshy. Stem 3.5 dm tall, 4 mm in diameter at base and about 1 mm in diameter under inflorescence, delicate, erect, tinged red- dish, sparsely glandular under and within inflorescence. Cauline bracts 9, herbaceous with hyaline margins, acute, the middle and upper shorter than internodes. Leaf single, petiolate; petiole about 4 cm long, narrow; blade 5 cm long, 2.5 cm wide, ovate-lanceolate, acute. Inflorescence 9 cm long, about 20-flowered, lax. Flowers horizontal, tiny. Floral bracts 6-7 mm long, ovate-lanceolate, acute, semi-transparent, herbaceous, with hyaline margins, 1-nerved. Ovary about 6 mm long, slightly twisted, sparsely glan- dular in the upper part. Dorsal sepal 5 mm long, 1 mm wide, oblong-lanceolate, obtuse, thin, semi-transparent, l-nerved, sparsely glandular outside near the base. Lateral sepals 6 mm long, 1.3 mm wide, oblong, subobtuse, asymmetric, otherwise as the dorsal one. Petals 4.9 mm long, .6 mm wide, oblong-spathu- late, obtuse, very thin, 1-nerved. Lip on short and narrow claw, divided into a hypochile and epichile; hypochile 4 mm long, 2.2 mm wide, rectangular, slightly thickened in the middle, the basal auricles rounded; epichile 2 mm long, 1.9 mm wide, rounded, minutely dentate, papillate, subobtuse. Gynostemium 3.5 mm long, slender, slightly arcuate, lacking rostellum and viscidium. Anther about 1 mm long, ovate-cordate. Pollinia with attenuate apices and rounded bases. Type. Mexico. Oaxaca: km 6.3 from Ixtlan, on brecha N from km 5.8 on Natividad road, elev. 2100-2400 m. Bosque de pino y encino, con Arbutus. In shade, in bud, flower and unripe capsule. 1993] Szlachetko—Schiedeella romeroana 3 GREENWOOD 1185 Figure 1. Schiedeella romeroana Szlach. a. flower and floral bract. b. dorsal sepal. c. petal. d. lateral sepal. e. lip. f. gynostemium. g. stigma. Very few spikes seen, ca. 10, versus ca. 200 leaves in October 1983. 18 Mar 1984. Greenwood 1185 (HOLOTYPE: AMO). EtyMo.Locy. [I dedicate this species to Dr. Gustavo A. Rome- ro, Curator of the Oakes Ames Orchid Herbarium, Harvard Uni- versity, in gratitude for his invaluable help in my work on the revision of the subtribe Spiranthinae. 4 Rhodora [Vol. 95 Figure 2. Comparison of the rostellum remnant (above), lip form (middle) and flower (below) between Schiedeella valerioi (Ames & Schweinf.) Szlach. & Sheviak (a-c), S. /laveana (Lindl.) Schlecht. (d-f), S. amesiana Garay (g-i) and S. romeroana Szlach. (k—m). Schiedeella romeroana seems to be closely related to S. /laveana (Lindl.) Schlecht., S. amesiana Garay and S. valerioi (Ames & Schweinf.) Szlach. & Sheviak (Figure 2). It is similar in the lip form to the first species and to the last in the proportion of length of lip to remaining perianth segments. It has horizontal flowers instead of the erect flowers of S. amesiana. ACKNOWLEDGMENTS I am grateful to Prof. Dr. Hab. Ryszard Ochyra for translation of the Latin diagnosis, and the Curator of AMO for loan of her- barium materials. 1993] Szlachetko— Schiedeella romeroana 5 LITERATURE CITED SZLACHETKO, D. L. 1992. Genera and species of the subtribe Spiranthinae (Or- chidaceae). 2. A revision of Schiedeella. Fragmenta Floristica et Geobotanica 37(1): 157-204. DEPARTMENT OF PLANT ECOLOGY AND NATURE PROTECTION GDANSK UNIVERSITY LEGIONOW 9, PL 80-441 GDANSK POLAND RHODORA, Vol. 95, No. 881, pp. 6-10, 1993 SARRACENIA PURPUREA L. SSP. VENOSA (RAF.) WHERRY VAR. BURKII SCHNELL (SARRACENIACEAE)—A NEW VARIETY OF THE GULF COASTAL PLAIN DONALD E. SCHNELL ABSTRACT A new variety of Sarracenia purpurea L. ssp. venosa (Raf.) Wherry is herein described as var. burkii. The variety is confined to the Gulf coastal plain and is characterized by pale pink to lavender petals and a white to pale green-white carpel, the umbraculate portion of the style being most apparent in this respect. Key Words: oe purpurea ssp. venosa, Sarracenia purpurea ssp. venosa ar. burkii, Gulf coastal plain, southeastern U.S. Sarracenia purpurea L. ssp. venosa (Raf.) Wherry is a carniv- orous pitcher plant found mainly in the coastal plain of the south- eastern United States, with mountain extensions in the Carolinas. There is a gap in its range in southern Georgia, and populations occur again in the central Gulf coastal plain (Wherry, 1933; McDaniel, 1966; Schnell, 1976). The Gulf coastal populations are addressed in this paper. In 1933, Edgar Wherry (p. 4) commented on what he described as a “mutant” of Sarracenia purpurea with an almost white um- braculate portion of the style and pale rose-pink petals, collected near Theodore, Alabama by Frank Morton Jones in 1910. Wherry found additional plants in 1932 (he did not specify exactly where, but implied southern Alabama) and sent them to Louis Burk of Philadelphia, an apparently well-regarded horticulturist of the day. The plants flowered true to the original description in Burk’s greenhouse the following year. Wherry (1933) suggested the pos- sibility of naming a horticultural variety in honor of Mr. Burk, but never did. Over a period of thirty-five years of botanizing the Sarracenia range, I have examined characters of North American pitcher plants in a series of papers (Schnell, 1978a, 1978b, 1979, 1981) in which S. purpurea ssp. venosa in the central Gulf coastal plain was mentioned. Gradually, I became impressed with the fact that nearly all specimens of the subspecies were similar to those de- scribed by Wherry; rather than being a rare mutant, it is essentially the only form present. The floral characters described, their con- 6 Schnell— Sarracenia 1993] ‘yuid syed are sjejod wuojLinped AjopiM ay} afTyM poi doop kv ase sfedag ‘uado oy) UI BUIMOIS ‘IOMOY Ul MyANG “IBA DSOUad “dss vaandind viuarDIDS “| VINBILz 8 Rhodora [Vol. 95 Figure 2. Sarracenia purpurea ssp. venosa var. burkii. Note nearly white um- braculate portion of the style. Light colored petals are pink to lavender in this example. sistency among populations, and partial isolation because of the southern Georgia “gap” indicate sufficient circumscription to des- ignate a variety of the cinco ee Sarracenia purpurea L. ssp. venosa (Raf.) Wherry var. burkii chnell, var. nov. Petalis pallide roseis vel lavandulaceis, stylis umbraculatis ovariisque albis vel pallide subviridulis, et sepalis diverse viridi- rubris vel purpurascentibus praediti. Petals pale pink or lavender, umbraculate style and ovary white to very pale green, sepals variably green-red to purple TYPE LOCALITY. United States. Alabama. Washington Coun- ty, off U.S. 45 near Deer Park: Open seep bog. 4 April 1992. (HOLOTYPE: NCU, pressed plants and color photos.) EryMoLocy. Named after early twentieth century horticul- turist Louis Burk since Dr. Wherry expressed interest in honoring r. Burk. 1993] Schnell— Sarracenia 9 Figure 3. Sarracenia purpurea ssp. venosa var. burkii. In this example, the ovary and inferior portion of the style are nearly white, while the umbraculate portion is green-white. RANGE. Central Gulf coastal plain: Florida panhandle from Liberty County west through the lower quarter of Alabama and Be A AB Open or moderately shaded pine savannas, seep bogs na "Mone slow, shallow meandering streams. Common among other Sarracenia spp. Plants grown in my greenhouse and outdoors clearly indicate that these characters are fixed and contrast markedly with the red to maroon petals and green (occasionally suffused with red) carpels of Atlantic coastal and mountain populations of ssp. venosa. Interestingly, plants growing with maximum sun exposure have the palest carpels (almost pure white), while more shaded plants tend to have some very pale green in the carpel. Petal color does not vary with light exposure, and sepals do vary in depth of color. Sarracenia purpurea ssp. venosa var. burkii grows often in large dense populations composed of many species of the genus, and all these individuals tend to hybridize readily one with the other, with many backcrosses in evidence. Care should be exercised 10 Rhodora [Vol. 95 when the occasional plant of S. purpurea with red or maroon petals and green carpels is found. Given the fecundity of Sarra- cenias in this range, one might reasonably speculate that such plants may be the product of hybridization with back-crossing and possible macrorecombination. Selfing of plants of var. burkii under greenhouse conditions results in abundant seedset and greater than 95% germination with typical adult plants as the eventual outcome. LITERATURE CITED McDanrcEL, S. T. 1966. A taxonomic revision of Sarracenia (Sarraceniaceae). Ph.D. dissertation, Florida State Univ. (University Microfilms, Ann Arbor, MI 48106, No. 67-345). SCHNELL, D. E. 1976. Carnivorous Plants of the United States and Canada. John F. Blair, Winston-Salem, NC. ——. 1978a. Sarracenia i. petal extract chromatography. Castanea 43: 107- TES; ——. 1978b. Systematic flower studies of Sarracenia L. Castanea 43: 211- 20. 1979. A critical review of published variants of Sarracenia purpurea L. Castanea 44: 47-59, 1981. Sarracenia purpurea L. ssp. venosa (Raf.) Wherry: variations in the Carolinas aggien plain. Castanea 46: 225-234. Wuerry,E. T. 1933. L f Sarracenia purpurea. Bartonia 15: 1-6. RT. 1, BOX 145C PULASKI, VA 24301 RHODORA, Vol. 95, No. 881, pp. 11-20, 1993 RHODODENDRON ALBIFLORUM HOOK. (ERICACEAE): ONE TAXON OR TWO? MEREDITH A. LANE, ZHONGREN WANG, CHRISTOPHER H. HAUFLER, PHILIP A. THOMPSON, YOUNGJUNE CHANG, CANDICE L. BREGIN, MICHAEL T. CAMPBELL, C. LUKE GARD, Gary ScuHoTtT, GREG SPIELBERG, KATHRYN E. STONER, Mark B. TAYLOR, AND ILAN YAROM ABSTRACT Populations of Rhododendron albiflorum Hook. in Colorado are disjunct by ca. 1470 km from populations in the Pacific Northwest. Morphological data showed that plants from the two sets of populations differed in length and width of petals and length of stamens. Genetic identity figures based on analysis of i isozyme electrophoresis found Colorado and Pacific Northwest populations to have an shi value of .76. These morphological, geographical and isozymic data support ognition of the Colorado plants as Rhododendron albiflorum Hook. var. satin (A. Nelson) M. A. Lane, comb. & stat. nov.; the Pacific Northwest plants are Rhododendron albiflorum Hook. var. albiflorum. Key Words: Rhododendron albiflorum, morphology, a electrophoresis, isjunct populations, Colorado, Pacific Northwe INTRODUCTION Rhododendron albiflorum Hook. (“Cascades Azalea” or “White Rhododendron’’) is a small shrub of subalpine areas in the Cas- cade Mountains of northern California and Oregon, the Olympic Mountains of Washington, and the Rocky Mountains of British Columbia, Alberta and Montana, with disjunct (by ca. 1470 km) populations in the Rocky Mountains of Routt and Jackson Coun- ties, Colorado (Figure 1). The plants have grayish bark and short- petiolate, narrowly elliptic, sparingly pubescent leaves. White flowers are borne in small axillary clusters; corollas are broadly Campanulate and up to 2 cm wide. Though the plants are very attractive, they are extremely difficult to cultivate, and so the Species is not well known among azalea fanciers. Named by Hooker (1838), Rhododendron albiflorum is based On a specimen from “alpine woods of the Rocky Mountains” collected by Drummond, probably in Alberta or southeastern British Columbia during his journey there in 1825-27 (McKelvey, 1956). In 1900, Rydberg included Hooker’s plant in Rhododen- 11 12 Rhodora [Vol. 95 Figure 1. Distribution of Rhododendron albiflorum Hook. Each dot may rep- resent one or more collections. The Colorado populations (Three Island Lake, Lake Diana, Lake Katherine, Lone Pine Creek, Gilpin Lake, Ute Pass and Sla- vonia) all lie within a 50 sq. mi. area in the Mount Zirkel Wilderness Area of Routt National Forest in Routt and Jackson Counties, Colorado. dron subg. Azaleastrum, and elevated the subgenus to generic status. He contended that the deciduous leaves of Azaleastrum albiflorum (Hook.) Rydberg distinguished it from Rhododendron, and that its large sepals, lateral flowers, peltate stigmas and cam- panulate corollas separated it from Rhodora and Azalea. Nelson (1913) accepted Rydberg’s generic treatment of the Pacific North- west plants, but considered the Colorado populations sufficiently different to warrant specific status, and named them Azaleastrum warrenii A, Nelson. Most authors of floras (e.g., Harrington, 1954; Peck, 1961; Hitchcock and Cronquist, 1973) have accepted nei- ther Rydberg’s genus nor Nelson’s species, though Weber (1990) and Weber and Wittmann (1992) use A. albiflorum (Hook.) Ryd- berg for the Colorado populations. A recent monographic treat- 1993] Lane et al.— Rhododendron albiflorum 13 ment (Philipson and Philipson, 1986) considered R. albiflorum to be the sole member of Rhododendron subg. Candidastrum, which nonetheless clusters with sect. Azaleastrum of subg. Aza- leastrum in phenetic analyses (Palser et al., 1991). In this study, we asked two questions: 1. Do the populations from the Pacific Northwest and those from Colorado constitute one taxon or two? 2. If there is more than one taxon present, should they both be recognized at the species level, or both as varieties of a single species? We addressed these questions using morphometric and isozyme electrophoretic techniques. MATERIALS AND METHODS Collection data obtained from a total of 712 specimens bor- rowed from 11 herbaria were used to construct a distribution map for the taxon (Figure 1). Pollen from a selection of these specimens was studied using both light and scanning electron microscopy. Eight quantitative characters, referred to below as variables five through twelve (VARS = leaf length, VAR6 = leaf width, VAR7 = number of flowers per cluster, VAR8 = sepal length, VAR9 = sepal width, VAR10 = petal length, VAR11 = petal width, and VAR 12 = stamen length), were measured on representative spec- imens from the Pacific Northwest (n = 51) and from Colorado (n= 7). Variables one through four were geographic and specimen identifiers. Statistical analysis of these characters was carried out ona Macintosh IIsi microcomputer using SPSS for the Macintosh (SPSS, Inc., 1990). Field observations and collection of leaf tissue for isozyme electrophoresis were carried out in the following localities (pop- ulation acronym and number of individuals in parentheses): Col- orado: Routt County, Routt National Forest, Mount Zirkel Wil- derness Area, Three Island Lake (TIL, 20 individuals); Oregon: Marion Co., Mount Hood National Forest, Breitenbush Lake Road (BLR, 40 individuals); Washington: Clallam County, Olym- pic National Park, Hurricane Hill trailhead (ONP, 20 individu- als); and British Columbia: Vancouver Island, Grouse Mountain Resort (GMR, 40 individuals). Voucher specimens were depos- ited at COLO and KANU. Leaf samples in individual plastic bags were transported to the laboratory on wet ice, ground in buffer as described by Mitton et al. (1979), and stored at —80°C in microfuge tubes until use. 14 Rhodora [Vol. 95 ii VAR 12 —— VAR 11 ene VAR 9 aT Geme —_ VAR 8 VAR eauceaiaggags : =a VAR 6 — VAR 5 C08 Pee Ce ae: aE a ee 8 28 Figure 2. Graphic representation of the statistical analysis of eight meristic characters (see Materials ¢ and Methods for ‘explanation n) of Rhododendron albiflo- 57) and Colorado (n = 7). Numbers on x- axis are cm for variables 5—6 and mm for variables 7-12. Thin line = range, boxed area = mean + one standard deviation (hatched fill = Pacific Northwest, solid fill = Colorado), vertical bar = mean. Note that while the means for some characters (e.g., petal length, VAR10, and petal width, VAR11) are significantly different (Table 1), the range for the Colorado populations is included within (VAR11) or greatly overlaps (VAR10) that of the Pacific Northwest populations, reducing their usefulness as taxonomic characters. Starch gels were prepared according to protocols of Soltis et al. (1983). Interpretable banding patterns were obtained from the enzymes glucose-6-phosphate isomerase, phosphoglucomutase, leucine amino peptidase, aspartate amino transferase and triose- phosphate isomerase when run in buffer system 6 or a modifi- cation of buffer system 8 (Haufler, 1985). The clearest banding patterns for isocitric dehydrogenase, shikimate dehydrogenase, malate dehydrogenase, and 6-phosphogluconate dehydrogenase were obtained using buffer system 11 of Soltis et al. (1983) ora pH 7.5 modification of the morpholine system of Odrzykoski and Gottlieb (1984). Data from banding patterns were analyzed using the program BIOSYS-1 (Swofford and Selander, 1981) ona Zenith 159 microcomputer. 1993] Lane et al.— Rhododendron albiflorum 3 Table 1. Comparison of means and standard deviations (SD) for eight mea- sured characteristics (VAR5 through VAR12, as explained in text) of Colorado and Pacific Northwest (Oregon, Washington, British Columbia, Alberta, Montana and Idaho) specimens of Rhododendron albiflorum. arenes were taken from herbarium sheets; flowers were rehydrated in Tween Mean (SD) Colorado Pacific Northwest Character (n= 7) = 51) P VARS Leaf L (cm) 3.8714 (1.378) 4.8314 (1.150) <.05 VAR6 Leaf W (cm) 1.4571 (0.469) 1.8196 (0.444) <.05 VAR7 Flowers/cluster 2.8571 (1.069) 4.4118 (2.376) NS VAR8 Sepal L (mm) 7.2857 (1.380) 8.7549 (1.650) <.05 VAR9 Sepal W (mm) 3.2857 (0.488) 3.8333 (1.109) NS VARIO Petal L (mm) 10.2857 (1.254) 14.1373 (2.764) .001 VARI11 Petal W (mm) 5.8571 (0.900) 7.8725 (1.466) 001 VARI12 Stamen L (mm) 7.4286 (1.512) 9.4706 (1.669) <.01 RESULTS AND DISCUSSION The original descriptions provided by Hooker (1838) and Nel- son (1913) differ in language and length, but little in content. In our examination of the plants, we found no qualitative characters, such as color or shape of corollas or shape of leaves, that could be used to distinguish two different species. Likewise, pollen from plants of the two areas was identical. Although Colorado plants tend to be shorter (usually <1 m) than Pacific Northwest plants (1-2 m), and to have smaller leaves and fewer and smaller flowers, these differences may be attributable to phenotypic plasticity in response to the severity of the habitats in which they grow. Ele- vational range for the Pacific Northwest populations is 1160- 2700 m above sea level and weather patterns have an oceanic influence, while for the Colorado populations, the elevational range is 3000-3500 m above sea level and the climate is distinctly continental. Analysis of the quantitative characters indicated that the means of measurements of leaf length and width, sepal length, petal length and width, and stamen length of the Colorado populations (Table 1) were significantly smaller than the means of the same measurements for the Pacific Northwest populations. However, correlations (Table 2) among the characters that have significantly different means (Table 1) reduce the taxonomic value of those [Vol. 95 Rhodora 16 "PAZ [0° OY} 1B JUBOYTUBIS 4» "PA CO’ OY} 1B JUROYTUBIS , 0000'T TIUVA *«P6LY" 0000°T TTUVA *9S19 *x6 109° 0000°T OlIUVA ¢LOC PST PL8l 0000°T 6uVA *9TOP" «x09 xa SPECS" xa 1 88° 0000'°1 SUVA 7980° *610€°— 0760" ceOl = OCGE: 0000°T LUVA POPC 8¢L0 2:10) 8c9oT «L187 *I7LC 0000°T OUVA (4500 1€80° 8190 esol’ PIPc STCT «a 88" 0000'T SUVA TIUVA [ITUVA OIUVA 6uVA SUVA LUVA 9UVA SUVA ‘yigug] usWIRIS = TIUVA ‘yypim yeiod = TTUVA ‘YIdug] fered = OTUVA “UIPIM jedos = 6UVA ‘WIBug] edas = SUVA “JoIsnyo Arey[ixe Jad siomoy JO Jaquinu = LYWA ‘UIPIM Jeo] = 9UVA ‘YISUI] Jeo] = CUVA ‘SPPOo UONROyTUSp! Ayipeooy pue usurtoads a1aM p—] SOTQeLIeA ‘(g¢ = u) wnuosfiqin uo4puapopoyy JO susUIIDeds UO PaINsBIUT so[qeLBA ZuUOUIL S]UDSTOYJIOI UONRIIIOD “7% IQeL 1993] Lane et al.— Rhododendron albiflorum 7 SIMILARITY 0.68 0.72 0.76 0.80 0.84 0.88 0.92 0.96 1.00 ES L ! | 1 i } Colorado ste Oregon Washington : British Columbia Figure 3. Dendrogram produced by cluster analysis calculated by the un- weighted pair group method (Swofford and Selander, 1981) based on unbiased genetic identity (Nei, 1978). differences. For example, it is reasonable for a shorter petal also to be narrower (VAR10 is correlated at the .01 level with VAR11) or for shorter leaves also to be narrower (the correlation between VARS and VAR6 is significant at the .01 level). Even though a discriminant function analysis of the Colorado populations versus the Pacific Northwest populations indicated that petal length and width, and stamen length, accounted for most of the distinction between the groups, the ranges of measurements of these char- acters for the Colorado populations are either completely within or significantly overlapping with those for the Pacific Northwest populations (Figure 2). Finally, a cluster analysis (not illustrated) performed on all 58 specimens failed to group the 7 Colorado specimens; rather, these populations were scattered among those from Montana, Alberta and Washington. Thus, we conclude that the morphological evidence does not support the recognition of the Colorado plants as a species distinct from that of the Pacific Northwest. Results from isozymic analysis showed the Colorado popula- tions to be distinct but failed to provide compelling evidence for recognizing these plants at the specific level. As summarized in Table 3 and Figure 3, genetic identity values among the Pacific Northwest populations were consistently above .90 while those between the Colorado and Pacific Northwest populations ranged from .729 to .722. Given that the average genetic identity (Nei, 18 Rhodora [Vol. 95 e 3. Matrix of similarity and distance coefficients (Nei, 1978) based on Po Pie electrophoretic analysis of 10 enzyme systems. Population acronyms are explained in Materials and Methods. Above diagonal: unbiased genetic dis- tance; below diagonal: unbiased genetic identity. Population 1 Bo 3 4 1 TIL (Colorado) — nin 258 316 2 BLR (Oregon) 761 .118 .090 3 ONP (Washington) my i fe 913 = .000 4 GMR (British Columbia) .729 .939 1.000 — 1978) among congeneric angiosperm species is .67 (Crawford, 1983), our results might be used to suggest that the Colorado populations should be elevated to specific status. However, it is probable that the Colorado populations represent glacial relics (Davis, 1983) that have been isolated for sufficient time to diverge at the isozyme level, but there has been no corresponding mor- phological divergence except in size. Thus, the combined data suggest that the most reasonable tax- onomic treatment of the two sets of populations is the recognition of two varieties of a single species. The Pacific Northwest plants are referred to R. albiflorum Hook var. albiflorum. Rhododendron albiflorum Hook. var. warrenii (A. Nelson) M. A. Lane, comb. & stat. nov. Azaleastrum warrenii A. Nelson. Type: Colorado, Jackson Co., lower slope of Mt. Zirkel, 9275 ft., 14 Jul 1911, E.R. Warren 16 [HOLOTYPE: RM!]. ACKNOWLEDGMENTS We thank William A. Weber for suggesting the problem, Eric Rabe for assistance with graphics, curators of ALTA, GH, MONTU, MO, NY, RM, OSC, UC, US, WS, and wsc for the loan of specimens, and two anonymous reviewers for their thoughtful comments. This study was conducted in part as a class project of Biol. 603 (Systematic Botany) at the University of Kansas in the fall se- mester of 1 LITERATURE CITED CRAWFORD, D. J. 1983. Phylogenetic and systematic inferences from electro- phoretic studies, pp. 257-287. In: S. O. Tanksley and T. J. Orton, Eds., 1993] Lane et al.— Rhododendron albiflorum 19 Isozymes in Plant Genetics and Breeding, Part A. Elsevier Press, Amsterdam, Davis, M. E. 1983. Quaternary history of deciduous forests of eastern North America and Europe. Ann. Missouri Bot. Gard. 70: 550-563. HARRINGTON, H. D. 1954. Manual of the Plants of Colorado. Sage Books, Denver, CO HAuFier, C.H. 1985. Enzyme variability and modes of evolution in Bommeria (Pteridaceae). Syst. Bot. 10: 92-104. Hitcucock, C. L. AND A. CRONQUIST. athe Flora of the Pacific Northwest. yy of Washington Press, Seat Hooker, oe 1838. Flora Boreali pies Si Vol. 2. H. G. Bohn, London. p. 4 OF teed S. D. 1956. Botanical Exploration of the Trans-Mississippi West 1790-1850. p. 486. The Arnold Arboretum of Harvard University, Jamaica Plain, MA. saan Be Y. B. LINHART, K. B. STURGEON AND J. L. HAMRICK. 1979, r ab Heredity 70: 86-89. Nei, M. 1978. Estimates of average heterozygosity and genetic distance from a small number of individuals. Genetics 89: 583-590. Netson, A. 1913. Contributions from the Rocky Mountain Herbarium XIII. Bot. Gaz. (Crawfordsville) 56: 63-68. OprzykoskI, I. J. AND L. D. Gottiies. 1984. Duplications of genes encoding 6- shsaeiogieiatnate dehydrogenase in Clarkia (Onagraceae) and their phy- logenetic implications. Syst. Bot. 9: 479-489. PA.ser, B. F., W. R. PHILIPSON AND M.N. PHILIPSON. 1991. Characteristics : ovary, ovule, and mature megagametophyte in Rhododendron L. and thei taxonomic significance. Bot. J. Linn. Soc. 105: 289-390. Peck, M. G. 1961. A Manual of the Higher Plants of Oregon. Oregon State University Press, Portland, OR PHILipson, W. R. AND M.N. Paco, 1986. A revision of Rhododendron. Il. Subgenera Azaleastrum, sl aaa oo and Therorhodion. Notes Roy. Bot. Gard. Edinburgh 44: RypserG, P. A. 1900. Rocky Mountain tee Mem. New York Bot. Gard. 1: 297 Soxtis, D. E., C. H. HAUFLER, D. C. DarRow AND G. J. Gastony. 1983. Starch gel electrophoresis of ferns: a compilation of grinding buffers, gel and elec- trode buffers and staining schedules. Amer. Fern J. 73: 9-27. SPSS, Inc. 1990. SPSS for the Macintosh, Version 4.0. SPSS Inc., Chicago, IL. WOFFORD, D. AND R. B. SELANDER. 1981. BIOSYS-1: A Computer Program for the Analysis of Allelic Variation in Genetics. University of Illinois, Ur- bana-Champaign, IL. Weser, W. A. 1990. Colorado Flora: Eastern Slope. University Press of Colo- rado, Niwot, me Wexner, W. A. AND R. C. Wittmann. 1992. Catalog of the Clorado Flora: A Biodiversity Baseline. University Press of Colorado, Niwot, CO. 20 Rhodora MAL. 2. W., CEL, Yi. CLE, MLC, CLG, GS. GS, KES, MET. AND LY. DEPARTMENT OF BOTANY AND R. L. MCGREGOR HERBARIUM UNIVERSITY OF KANSAS LAWRENCE, KS 66045 DEPARTMENT OF E.P.O. BIOLOGY UNIVERSITY OF COLORADO BOULDER, CO 80309-0334 [Vol. 95 RHODORA, Vol. 95, No. 881, pp. 21-24, 1993 SNOW ALGAE IN NORTHERN NEW ENGLAND BRIAN DUVAL Algae from the genus Chloromonas are known to live in snow in alpine and arctic terrain (Kol, 1968). These cryophilic organ- isms can be seen as green patches near the surface of snow, typ- ically under the limiting sunlight of a coniferous canopy. Forest litter from spruce and fir may provide nutrients essential for algal growth (Hoham, 1976). Its proliferation is generally sporadic in what often appears to be uniform snowbank habitat. Reports of snow algae in the northeastern United States have all been from the Adirondack Mountains of New York state (Hoham et al., 1989). In May 1992, green snow algae were collected from several peaks in northern New England; this is the first report of snow algae from New England. Warm spring days and snow meltwater trigger germination and migration of algal cells towards the snowbank surface. In the early morning and evening hours, light conditions are best suited to observe the algae (Hoham, 1975). In May 1992, green and orange snow algal populations were found on Mt. Mansfield and Jay Peak in Vermont, Mt. Washington in New Hampshire, and Mt. Katahdin in Maine. Collections were separately contained in Whirl-pak® bags packed in snow and transported to the labo- ratory in a cooler. Parameters such as snowbank depth, elevation and solar exposure were recorded. Light microscopy showed these cells to be Chloromonas brevispina Fritsch, or a mix of C. brevispi- na and spindle-shaped cells formally known as Scotiella cryophila (Fritsch). The latter are now believed to be asexual spores in the life cycle of a Chloromonas that may belong to the species complex of Chloromonas polyptera Fritsch (Hoham, 1983, 1992). In March 1992, algal cells collected from snow on Mt. Greylock, Massachusetts were clearly not members of cryophilic snow algae. However, a complex community of microorganisms including cyanobacteria, fungi, rotifers, ciliates and other protists have been noted on Mt. Greylock. From Mt. Washington, ciliates engorged with green snow algal cells were an interesting component of the snowbank ecosystem. Fungi belonging to the genera Chionaster and Selenotila were present in certain snow samples. Cohabitation of snow by cryo- philic algae and the fungus Phacidium may occur via adherence 21 ae Rhodora [Vol. 95 Figure 1. Phase-contrast photomicrograph of orange snow algae, unidentified species, av. 15 x 25 wm (scale bar = 40 wm). Collected from snow (Killington, VT, May 1992). of algal cells to the web-like filaments of the fungus (Hoham et al., 1989). Snow collected in the Sierra Nevada and Cascade Rang- es of California demonstrated this close spatial association (Du- val, unpubl.); however, no such relationship was noted in New England snow. Algal cells with orange carotenoids were found on man-made snow at four New England ski slopes. Secondary pigments serve to protect individual cells from visible and ultra-violet light (Har- dy and Curl, 1972). Orange algal populations were found in faint suncups (small depressions in the snow) at the edge of ski trails in relatively open to partially shaded exposures. This organism is probably a new species not reported previously from snow (Figure 1). Absorption spectra of solvent-extracted pigments from the or- ange New England snow algae were analyzed. Preliminary analysis of the pigments contained in the cells show absorption only at the wavelengths of 440, 460, and 640 nm; those absorption peaks are typical of Chloromonas cells collected in the western United States by the author. Red-pigmented cells of the genus Chlamy- domonas from western snow have an entirely different spectral 1993] Duval—Snow Algae a3 profile. Peak light absorption near the 470 nm wavelength region indicates that carotenoids and xanthophylls are most predomi- nant in the red cells. Thus absorption properties of these orange algal cells from New England appear similar to typical Chloro- monas (green) rather than Chlamydomonas (red) snow algae. Spectrophotometric analysis suggested an ability of certain cry- ophilic algae to modify their photosynthetic pigments in response to solar exposure (Iturriaga and Sullivan, 1989). However, other studies suggest that nitrogen depletion in snow causes a shift towards secondary carotenoids that protect cells from ultraviolet irradiation (Czygan, 1970; Bidigare et al., 1992 All snow algal samples are presently being examined and cul- tured in the laboratory of R. Hoham at Colgate University. This note is a partial report on the distribution of snow algae in New England. ACKNOWLEDGMENTS I thank R. Hoham of Colgate University for use of his labo- ratory, guidance and accompaniment on these field trips. I also thank the Planetary Biology Internship Program (MBL), C. P. Mckay and NASA Ames for their support. LITERATURE CITED BipicareE, R. R., M. E. ONpRuSEK, M. C. Kennicutt, R. ITurriaGA, H. R. H Y, R. W. HoHaM AND S. A. Macko. 1992. A photoprotectve func- tion for secondary carotenoids of snow algae. J. Phycol. (in pres CzyGan, F.-C. 1970. Blutregen und Blutschnee: nei ane ah ay ee von Haematococcus pluvialis und Chlamydomonas nivalis. Arch. Mikrobiol. 74: 9-76 Harpy, J. T. anp H. orn 1972. The candy-colored, snowflaked alpine biome. Nat. Hist. 81(9): 74— Honam, R. W. 1975. The fe history and ecology of the snow alga Chloromonas pichinche. Phycologia 14: 213-224 The effect of coniferous litter and different snow meltwaters upon the growth of two species of snow algae in axenic culture. Arctic Alpine Res. 8: 377-386. . 1983. The life history and ecology of the snow alga Chloromonas po- lyptera comb. nov. (Chlorophyta, Volvocales). Canad. J. Bot. 61: 2416-2429. 992. Environmental influences on snow algal microbes. Jn: B. Shafer, Ed. Proceedings of the Sixtieth Annual Western Snow Conference. Jackson Hole, Wyoming. pp. 78-83. ; ——, C. P. Yatsko, L. GERMAIN AND H. G. Jones. 1989. Recent discoveries 24 Rhodora [Vol. 95 of snow algae in upstate New York and Quebec Province and preliminary reports on related snow chemistry. Jn: Lewis, J., Ed. Proceedings of the Forty- sixth Annual Eastern Snow Conference. Quebec City, Quebec, pp. 196-200. ITuRRIAGA, R. AND C. W. SULLIVAN. 1989. Spectral light absorption character- istics of individual sea-ice microalgae from McMurdo Sound, Antarctica. Antarc. J. U.S. 1989. Review Vol. XXIV. No. 5. Koi, E. 1968. rialbyetons ge Biologie und Limnologie des Schnees und Eises I. Kryovegetation. Jn: H. J. Elster and W. Ohle, Eds., Die oe Vol. 24. E. Schweizerbartsche Verlagsbuchandlung, Stuttgart. 216 p DEPARTMENT OF MICROBIOLOGY UNIVERSITY OF MASSACHUSETTS AMHERST, MA 01003 RHODORA, Vol. 95, No. 881, pp. 25~37, 1993 A TAXONOMIC STUDY OF ARISTIDA STRICTA AND A. BEYRICHIANA ROBERT K. PEET ABSTRACT Grasses traditionally assigned to Aristida stricta Michx. include two morpho- logically and eeberapmealy nepeente —— The name Aristida stricta Michx, Ol f counties in South Carolina. A second species, Aristida saab ie) Trin. & Rupr., occurs from southern South Carolina south throughout Florida and west to Mississippi. Key Words: eter Aristida beyrichiana, Aristida stricta, Poaceae, coastal n, North America, southeastern U.S. INTRODUCTION The name Aristida stricta Michaux has long been applied to those southeastern plants colloquially known as “‘wiregrass” which dominate the understories of most frequently-burned Pinus pa- lustris Mill. and P. elliottii Engelm. woodlands and savannas from central North Carolina south to southern Florida and west to eastern Mississippi. No intraspecific variation has previously been described for this taxon. However, the range of A. stricta, s./. has a conspicuous gap in central South Carolina. Further, differences in sheath and blade indument allow plants from north of the gap to be readily distinguished from those south of the gap. In this paper I examine the morphological characteristics, distribution, and taxonomic status of these two types of Aristida stricta, s.l. For reasons explained below in the section on taxonomy, I refer to the northern plants as A. stricta Michaux and to the southern plants as A. beyrichiana Trinius & Ruprecht. MORPHOLOGICAL DIFFERENCES The two variants of Aristida stricta, s.l. can be readily and unambiguously differentiated by examination of leaf indument. On plants of the southern A. beyrichiana, a densely woolly or villous bearding or tuft of indument is present at the base of the leaf blade (Figure 1). These prominent hairs vary from being distributed on the upper surface of the blade and at the corners of the collar to also (and much more commonly) surrounding the blade, collar, and uppermost portion of the sheath. With age the 25 26 Rhodora [Vol. 95 Figure 1. Characteristic blade and sheath pubescence of (a—c) Aristida beyri- chiana Trinius & Ruprecht and (d-f) Aristida stricta Michaux. In each case the first (a and d) of the three figures is the most typical and the other two illustrate some of the range of variation. d hairs can be partially deciduous, but they are always evident on the younger foliage and, together with the involute leaves, allow A. beyrichiana to be distinguished in vegetative condition from all other southeastern North American grasses. Individual plants can differ substantially in amount of indument, but ata minimum, plants of A. beyrichiana have numerous hairs protruding from about the throat and corners of the collar. In contrast, the northern A. stricta lacks this localized bearding or tuft. At most, there may be a few prominent hairs at the corners of the collars of the most hairy plants, and on most plants there are no hairs except for those on the back of the blade as described below. The northern Aristida stricta, s.s. is readily distinguished in vegetative condition from all other native grasses by the presence of a line of villous indument (hairs .6-1.5 mm long) adjacent to and on each side of the midrib along the length of the back of the involute blade (Figure 1). The hairs can be deciduous with age, but are always present on the younger foliage. The foliage of the southern A. beyrichiana usually lacks indument on the backs of the leaves except for the characteristic woolly to villous tufts or bearding at the bases of the blades. The one exception is that seven percent of the Florida specimens examined (17 of 251) had foliar indument along the midrib; the trait was not observed in collections of A. beyrichiana from any other state. In those few Florida plants where the lines of villous indument were present on the foliage, the diagnostic wooly tuft or bearding at the base of the blade was particularly conspicuous. 1993] Peet— Aristida 27 GLI Ess U manlapytee: ESET ARNE RASTA SCS SEs YT a. £7 SADA KEL YT ATS CATT TO AOS POX OT TASTES ee esis Messe oaometel peore) | aaah PRD AS R TRY KO} QS cme FOB a ay Oy AS ©) OK Buy) ee, KA CST aGololoye ion 4 ieeais TOAD a ae a egssley ES a Figure 2. County distribution of Aristida stricta Michaux (O) and Aristida beyrichiana Trinius & Ruprecht (@). Unverified reports are indicated by (e). Other differences between the taxa can be observed, but they are generally minor and often statistical. For example, the average ligule length is less for Aristida stricta, as is culm width, and, on average, the glumes are more unequal in A. beyrichiana. DISTRIBUTION I examined all specimens labeled as Aristida stricta Michx. held by the following herbaria: CLEMS, DUKE, FLA, FLAS, GA, GH, L, MO, NCSC, NCU, NY, P, UAL, US, USCH, vpB and Southern Mississippi. County range maps were drawn using data on the 564 specimens examined that were assignable to county of origin (Figure 2). Nine counties within the range of Aristida beyrichiana for which I could not locate specimens to authenticate reliable reports are indicated on the map by small dots: Sumter County, Georgia 28 Rhodora [Vol. 95 (Jones and Coile, 1988); Camden, Candler, Pierce and Richmond Counties, Georgia (Bozeman, 1971); Harrison County, Missis- sippi (Eleuterius and Jones, 1969); and Barnwell and Charleston Counties, South Carolina (Radford et al., 1968). A record from Oglethorpe County, Georgia, in Jones and Coile (1988) is based onan incorrectly identified collection (GA No. 133102). Blomquist (1948) indicated a record of Aristida stricta for Pasquotank Coun- ty, North Carolina. Despite my examination of all of the herbaria Blomquist reported consulting, I have found no specimen to sup- port that record. Because Pasquotank County is located well north of the known range of the species, I have not indicated this record on the map, although appropriate habitat does occur within the county. The ranges of Aristida stricta and A. beyrichiana are not only distinct, but clearly separated (Figure 2). Neither species has been found in central South Carolina. The one near intrusion of the range of A. beyrichiana into the central part of South Carolina is in Berkeley County, where the species is known only from a single locality from which it appears to have been extirpated. Exami- nations of likely habitats in central South Carolina, including the sandhills around Columbia and the pine flatwoods of the Francis Marion National Forest nearer the coast failed to locate either species. Aristida stricta Michx. is almost a North Carolina endemic, occurring only in North Carolina and the northern most counties of South Carolina. In this area the species is particularly common on coarse sands dominated by Pinus palustris, both in the fall- line sandhills and on the flatlands of the outer coastal plain (Frost et al., 1986; Peet and Allard, 1993). Examination of the distri- bution of the southern wiregrass, Aristida beyrichiana, shows that in the northeastern portion of its range it is largely concentrated in those counties that contain the dune systems characteristic of the northeast sides of major rivers (e.g., the Savannah and Al- tamaha Rivers). In the western portion of its range, the species is confined to the southernmost tier of counties along the Gulf of Mexico. The ranges of the two species that make up the Aristida stricta, s.l. complex are not atypical of the ranges of other species of the southeastern pinelands. Indeed, so many species have similar ranges that it seems likely that there were two main centers of persistence of the pineland flora during the last full glacial: one 1993] Peet—Aristida 29 near but perhaps east of what is now southeastern North Carolina, and another along the Gulf coast. Species with ranges similar to that of A. stricta, s.s. and essentially endemic to the pinelands of North Carolina and adjacent South Carolina include Dionaea muscipula Ell., Kalmia cuneata Michx., Lysimachia asperulae- folia Poir., Lysimachia loomisii Torr., Pyxidanthera brevifolia Wells, Solidago pulchra Small, Solidago verna M. A. Curtis, Spo- robolus, sp. nov. (aff. teretifolius Harper; a dominant of moist savannas), Tofieldia glabra Nutt., Vaccinium crassifolium Andr. and Zenobia pulverulenta (Bart. ex Willd.) Pollard. In addition, Calamovilfa brevipilis (Torr.) Scrib., Gentiana autumnalis L. and Pyxidanthera barbulata Michx. have similar ranges, but also have disjunct populations in the pine barrens of New Jersey. Among species that, like Aristida beyrichiana, have ranges that extend from the Gulf coast north to southernmost South Carolina are Asclepias cinerea Walt., Astragalus villosus Michx., Baptisia lan- ceolata (Walt.) Ell., Gaylussacia tomentosa (Gray) Small, Gay- lussacia mosieri Small, Helianthus radula (Pursh) Torr. & Gray, Kalmia hirsuta Walt., Liatris gracilis Pursh, Pinus elliottii En- gelm., Sabatia bartramii Wilbur, Sabatia brevifolia Raf., Serenoa repens (Bartr.) Small, Sorghastrum secundum (Ell.) Nash and Vac- cinium myrsinites Lam. TAXONOMY The two taxa included in Aristida stricta, s.l. are quite similar and are obviously closely related. Because they are readily and consistently distinguishable morphologically and are geographi- cally separated as well, I have chosen to treat them as separate species. Michaux’s original description of Aristida stricta [F]. Bor. Amer. 1: 41. 1803]. reads “A. culmis foliisque stricte erectis: foliis pu- bescentibus: racemo in spicam longam, angustam, non ita con- fertam coarctato: aristis gluma longioribus. Obs. Variat foliis nee et convoluto-filiformibus. Hab. in Carolina inferi- ” The reference to “‘foliis sbiceaanan seit is certainly sugees- re of the northern species with it the back sides of the leaf blades, but ‘the description alone is insufficient to determine definitively which taxon Michaux col- lected and described. Although Michaux (1803) reported that his type collection of 30 Rhodora [Vol. 95 Aristida stricta was from “Carolina inferiore,’’ Michaux’s anno- tation on the label reads only “Hab. in Carolina.” L. M. C. Rich- ard is widely believed to have anonymously written Michaux’s text, some of it after Michaux’s death (Hitchcock, 1908; Stafleu and Cowan, 1981; Uttal, 1984) and many of the habitat locations in the text differ from those on the specimen labels. The basis for designating South Carolina rather than simply Carolina as the collection locality in Michaux (1803) remains unclear. Michaux described in his journal (Sargent, 1889) numerous trips through pine barrens between Charleston, South Carolina, and Charlotte, North Carolina, and along the coast past Wil- mington, North Carolina, both areas which today are dominated by the northern Aristida stricta. Michaux is also known to have collected widely in the regions where the southern A. beyrichiana dominates. In short, it does not appear possible to determine from published records where in the Carolinas Michaux collected his Aristida stricta. The type of Aristida stricta is in the Michaux Herbarium at Paris. The Michaux Herbarium contains two specimens labeled Aristida stricta. However, only one has the annotation “Hab. in Carolina.” As Michaux (1803) reported the type as being from “Carolina inferiore,” this specimen appears to be the type, a conclusion also reached by Hitchcock (1908) and Henrard (1928). The specimen is a mixed collection. The larger culm bases have had their inflorescences broken off and appear to represent some species other than A. stricta or A. beyrichiana. These culm bases have ligules 1.5 to 2.0 mm long, significantly longer than the .02 to .2 mm range normally encountered in Aristida stricta and A. beyrichiana, and longer than any I observed in the over 650 herbarium specimens I examined. In addition, they differ from both species in that they are almost entirely devoid of indument. However, there are two separate young, sterile culms (with ligules .1 mm long) and several separate but older leaf blades that have the distinctive villous indument (‘‘foliis pubescentibus,” Mi- chaux, 1803) along the sides of the midrib that characterize the northern Aristida stricta. They also lack the tuft of hairs at the throat and corners of the collar typical of the southern species. Because Aristida stricta and A. beyrichiana have similar flowers and inflorescences, I cannot be absolutely certain which species the fertile portion of the type collection represents. The combined collection was interpreted as the type by Hitchcock (1908) and 1993] Peet—Aristida 31 Henrard (1928), but the pubescent foliage is the only portion unambiguously assignable to a single species. Thus, the pubescent young culms and pubescent older foliage on this specimen should probably be taken as the type of A. stricta. The other specimen in the Michaux Herbarium labeled Aristida stricta is a flat-bladed species, almost certainly Aristida purpuras- cens Poir. The original Michaux description of Aristida stricta has the confusing line “‘Variat foliis planiusculis et convoluto-filifor- mibus” which suggests that the description mixes the leaf blade characteristics of the two species represented by the two Michaux Herbarium specimens labeled Aristida stricta. The ‘“‘convoluto- filiformibus” foliage trait (unlike “‘planiusculis” foliage) is char- acteristic of a single species in the region, and “‘foliis pubescen- tibus” is consistent with that species, which provides further reason to apply the name A. stricta to the species with involute, pubescent leaves. There is a third Michaux collection labeled Aristida stricta in the General Herbarium at Paris which has been referred to by Henrard (1928) as an isotype from the Richard Herbarium. This collection is more complete than the collection in the Michaux Herbarium, has conspicuous villous indument along the length of the midvein, and is unambiguously the northern form. My suspicion is that this specimen was also examined when Richard or Michaux wrote the original description of Aristida stricta for Michaux’s 1803 text, simply because it is the only specimen with all the leaves conspicuously hairy. In 1849, Trinius and Ruprecht described a new species, Aristida beyrichiana Trin. & Rupr. (Mém. Acad. St. Pétersb. VI. Sci. Nat. 7(2): 104. 1849), which they recognized as distinct from Aristida stricta. However, their description of A. beyrichiana has not been adequate for others to make a distinction; as a consequence, all subsequent authors have viewed A. beyrichiana as a synonym of A. stricta. Further, because the description only addresses the inflorescence, it is not possible to assign the description to either of the taxa recognized in the present paper. In his monograph of North America Aristida, Hitchcock (1924) stated that he thought the type of Aristida beyrichiana to be an immature A. Stricta, although he admitted uncertainty. He had examined a fragment of the type in us (No. 81011), but apparently had not seen the holotype in Le. Henrard (1926) reported in his monograph of Aristida that “‘This is a plant with awns not yet fully developed, 32 Rhodora [Vol. 95 but agrees for the rest with A. stricta Michx., having the striking villous tuft at the throat of the sheaths,” a sure indication that he was examining the southern species. I have examined the fragment at us (No. 81011), and a Beyrich collection at L (No. 908.83-1016) that Henrard (1926) referred to as an isotype. I found them both consistent with Henrard’s de- scription. The villous tufts are present on the us fragment, in- dicating that it is Aristida beyrichiana, although the tufts are small- er than average, and the blades are both a little wider and not quite as closely involute as is normally encountered. However, the L specimen is unambiguously the southern species with villous bearding at the leaf base extending onto the collar, and with typical leaf dimensions. The spikelets on both specimens are, as Hitch- cock (1924) and Henrard (1926) reported, not fully mature. The type locality for Aristida beyrichiana remains uncertain, but is probably Georgia. Trinius and Ruprecht reported that the Beyrich specimen they described originated “In pinetis Georgiae et in territorio Arkansas.”’ However, this description may simply represent the general area in which Beyrich collected; he is also known to have collected in both North and South Carolina (Sayre, 1975). Henrard (1926) reported that the label on the type reads simply “Georgia in pinetis (Beyrich),”’ but this observation almost certainly derives from the label on the putative isotype at L, which it matches exactly. Virtually all modern descriptions of Aristida stricta match the species I call 4. beyrichiana. For example, Hitchcock (1924, 1935), Henrard (1932), Godfrey and Wooten (1979) and Allred (1986) all make reference to the distinctive villous tuft at the base of the leaf blades as characteristic of the species. Interestingly, Henrard (1932) reported the range of Aristida stricta as South Carolina to Florida and Mississippi, exactly the range of A. beyrichiana. There are no specimens of Aristida stricta, s.s. in the Leiden Herbarium where Henrard worked that are old enough to have been in the collection when he was active. Except for the type in Paris, there is no evidence that Henrard ever examined specimens of the true A. stricta. Trinius and Ruprecht (1849) listed Aristida lanuginosa Clarion (Trinius and Ruprecht, Mem. Acad. St. Petersb. VI. Sci. Nat. 4(2): 46. 1838) as a synonym of A. stricta. The original 1838 reference contains the name only, and a formal description first appeared in Trinius and Ruprecht (1849) where reference was 1993] Peet—Aristida 33 made to a specimen in the Mertens Herbarium (LE) with the name A. lanuginosa Bosc. (thus the appropriate citation is A. Januginosa Bosc ex Trin. & Rupr.). Hitchcock (1924) reported that there is a Bosc specimen from South Carolina in the Padua Herbarium labeled Aristida lanu- ginosa, which is really A. Janosa Muhl. ex Ell. In 1924, Hitchcock declared A. /anuginosa to be a synonym of A. /anosa, but there is no evidence that he ever examined the type specimen. Like Hitchcock, Henrard (1927) stated that Bosc’s plant is unambig- uously the previously described Aristida lanosa, a conclusion that probably derives from observation ofa Bosc collection in p labeled A. lanuginosa (probably an isotype), but which is certainly A. lanosa. Thus, it is extremely unlikely that the name A. lanuginosa refers to the southern species of Aristida stricta, s.1., but if it did it would be the oldest name other than A. stricta to be applied to the southern taxon. However, there has been described an A. lanuginosa Burch. (Burchell, 1824), which is a later synonym of A. vestita Thunb. As the 1824 name is validly published (see Henrard, 1927: 287), A. lanuginosa Bosc ex Trin. & Rupr. is a later homonym and could not be applied to the 4. beyrichiana material, even if the type proved to be this taxon. KEY AND SPECIES CHARACTERIZATION Base of blade, collar, and upper sheath lacking a conspicuous tuft or bearding of woolly to villous indument; current-year leaves with villous indument along the sides of the midrib on the lower surface for most of the length of the blade ........... Tee ae ee ee ee eer 1. Aristida stricta Base of blade and collar (and often upper sheath) with conspicuous tuft or bearding of woolly to villous indument, always present on current-year foliage but sometimes deciduous on older fo- liage; leaves usually glabrous above the basal 2 cm of the blade (tsb vie cee eee oe 2. Aristida beyrichiana 1. Aristida stricta Michaux, Fl. Bor. Amer. 1: 41. 1803. Chaetaria stricta (Michx.) Beauv. Ess. Agrost. 30, 152, 158. 1812; not Aristida stricta Muhlenberg, Descriptio uberior Graminum. 174. 1817. Type: Carolina (HoLoTyPE: Pp, Michaux Herbar- ium!; fragment us No. 81246, but insufficient material for determination; Isotype: p, General Herbarium!). 34 Rhodora [Vol. 95 Plants perennial; culms cespitose, erect, 60-120 cm; sheaths glabrous, occasionally with a few prominent hairs at the corners of the collars but typically absent with the collars glabrous; blades closely involute, .3-1.0 mm thick when rolled, to 50 cm long, mostly emerging near base, firm and somewhat flexuous, upper surfaces scabrous with some hairs up to 1.0 mm long, hairs on the upper surface hidden within the rolled leaf or with a few emerging along the lower 5 cm, villous on the lower surface along sides of midrib with hairs .6-1.5 mm (i.e., often over twice the width of the leaf) all along the length of the blade or at least along the lower 20 cm, hairs sometimes deciduous with maturity but always present on young foliage; ligules minute to nearly absent, membrane to .1 mm, lacerate or composed of flat hairs to .1 mm; panicles long and slender, 20-35 cm long, branches appressed, floriferous from the base; glumes somewhat unequal, awns 1.5— 2.5 mm, awns usually emerging from bifid tips; first glume 7-10 mm, |-nerved or with an additional nerve on one side, scaber- ulous on the keel but otherwise glabrous; second glume 9-12 mm, 1-nerved, glabrous, if scabrous on the keel then only near the summit; lemma 6-9 mm, shorter than the second glume, glabrous except for the densely short-pilose callus, callus .4-.6 mm; awns of lemma 3, somewhat unequal, sharply divergent with matura- tion, central awn 10-15 (22) mm, lateral awns usually 1-2 mm shorter. RANGE. Coastal plain of the Carolinas from the Pamlico River on the north to the northern tier of counties in South Carolina. 2. Aristida beyrichiana Trinius & Ruprecht, Mém. Acad. St. Pé- tersb. VI. Sci. Nat. 7(2): 104. 1849. Type: Georgia? (“In pinetas Georgiae et in territorio Arkansas.”’) (HOLOTYPE: LE, Trinius Herbarium, not seen, fide Henrard, 1926; fragment us! No. 81011; Isotype: L!, No. 908.83-1016, see Henrard, 1926). Plants perennial; culms cespitose, erect, 40-120 cm; sheaths glabrous except at the summit, copiously to somewhat villous on the upper .5 cm, especially about throat and collar, always with prominent protruding hairs at the corners of the collar; blades closely involute, .3-1.0 mm thick when rolled, to 20-40 cm long, mostly emerging near base, firm and somewhat flexuous, upper surfaces scabrous with few to many hairs protruding near base 1993] Peet —Aristida 35 up to 1.0 mm, lower surface somewhat to copiously villous to woolly at the base, but usually glabrous beyond the lowest 1-3 cm (at maturity, villous indument is sometimes deciduous, but young culms always possess this woolly tuft), those few plants with indument somewhat persistent along sides of midrib on underside along lower 20-30 cm of blade as in A. stricta are always copiously hairy at the bases of blades and on the sheaths, and are woolly to villous around the collars; ligule membrane .05-.3 mm, behind which are often conspicuous hairs; panicles long and slen- der, 15-30 cm long, branches appressed, floriferous from the base; glumes subequal, with awns 1.5—2.5 mm from bifid tip; first glume 7-10 mm, 1-nerved or with an additional nerve on one side, glabrous on the back but keel scaberulous; second glume 8.5-1 1 mm, |-nerved, glabrous, scabrous only on keel or not at all; lem- mas 6-8 mm, shorter than the second glume, glabrous except for densely short-pilose callus, callus .4-.6 mm; awns of lemma 3, somewhat unequal, sharply divergent with maturation, central awn 7-14 mm, lateral awns usually 1-2 mm shorter. RANGE. Southeastern coastal plain from southern South Car- olina south to Dade County, Florida, and west along the coastal counties to at least Jackson County, Mississippi. ACKNOWLEDGMENTS I thank K. N. Gandhi, John Kartesz, Jim Massey, Rogers McVaugh, Alan Weakley and Jo Willems for advice and encour- agement during various phases of this project, Dorothy Allard, Kelly Allred, Julia Larke, Jim Massey, Rogers McVaugh and Peter White for helpful comments on the manuscript, Susan Whitfield for drawing the illustration, and the directors and staff of CLEMs, DUKE, FLA, FLAS, GA, GH, L, MO, NCSU, NCU, NY, P, UAL, US, USCH, vps and the University of Southern Mississippi Herbarium for assistance and advice during my visits, or for loans of specimens. LITERATURE CITED LLRED, K. W. 1986. Studies in the Aristida (Gramineae) of the southeastern United States. IV. Key and conspectus. Rhodora 88: 367-387. Biomouist, H. L. 1948. The Grasses of North Carolina. Duke University Press, Durham, NC. : ; Bozeman, J. R. 1971. A sociologic and geographic study of the sand ridge 36 Rhodora [Vol. 95 vegetation of the coastal plain : Georgia. Ph.D. dissertation, University of North Carolina, Chapel Hill, N BuRCHELL, W. J. 1824. Travels in : Interior of Southern Africa, Vol. 2, p. 226. Longman, Hurst, Rees, Orme and Brown, London. Eveuterius, L. N. AND S. B. Jones, Jr. 1969. A floristic and ecological study of pitcher ay bogs in south Mississippi. Rhodora 71: 29-34. Frost, C. C., J. W. WALKER AND R. K. Peer. 1986. Fire-dependent savannas and prairies af the Southeast: original extent, preservation status and man- agement problems, pp. 348-357. In: D. L. Kulhavy and R. N. Conner, Eds., Wilderness and Natural Areas in the Eastern United States. Center for Ap- plied Studies, Nacogdoches, TX. Goprrey, R. K. AND J. W. Wooten. 1979. Aquatic and Wetland Plants of the 5 Sagat United States. Monocotyledons. Univ. Georgia Press, Athen HENRARD 1 ie 1926-33. A critical revision of the genus Aristida. Meded. Rijks Herb. Leiden 54: i—viii, 1-220, 1926; 221-464, 1927; 465-701, 1928; 703- 747, 1933. 1932. A monograph of the genus Aristida. Meded. Rijks Herb. Leiden Poy Hircucock, A. 8S. 1908. Types of American grasses: a study of the America species of grasses described by Lin — Gronovieus, Sloane, Swartz, and Michaux. Contr. U.S. Natl. Herb. 12: 113-158. . 1924. The North American shane of Aristida. Contr. U.S. Natl. Herb. 22: 517-586. . 1935. Aristida. North American Flora 17: 376-406. Jongs, S. B., JR. AND N. C. Comte. 1988. The Distribution of the Vascular Flora of Georgia. Department of Botany, University of Georgia, Athens, GA. MicHAux, A. 1803. Flora Boreali-americana. Jouanaux, Paris. Peet, R. K. AND D. J. ALLARD. 1993. Longleaf pine eeatotion of the Southern Atlantic and Eastern Gulf Coast regions: a preliminary classification. Proc. 18th Tall Timbers Fire Ecology Conf. (in press). RADFORD, A. E., H. AHLES AND C. R. BELL. 1968. Manual of the Flora of the Carolinas. Univ. North Carolina Press, Chapel Hill, NC SARGENT, C.S. 1889. Portions of the journal of Andre Michaux, botanist, written during his travels in the United States and Canada, 1785-1796, with an introduction and explanatory notes. Proc. Amer. Philos. Soc. 21: 1-146. Sayre, G. 1975. Cryptogamae exsiccatae: an annotated bibliography of exsic- catae of algae, lichenes, hepaticae, and muscii. Mem. New York Bot. Gard. 19: 277-423 STAFLEU, F. A. AND R. S. Cowan. 1981. Taxonomic literature. International Bureau for Plant Taxonomy and Nomenclature, Vol. 3. Utrecht, The Neth- erlands. Trintus, C. B. AND F. J. Ruprecut. 1838. Graminum in hisce saaeighelpoegohis generibus ac speciebus. Supplementia. Mem. Acad. St. Petersb. VI. Sci. N 4(2): 1-107. [The first page of the article indicates ““Conventui 1835.” The volume has two virtually identical title pages, one indicating Tome IV. conde partie (corresponding to the spine of the volume consulted), and a other page Tome II, no partie indicated. The running footnote on the pages reads “Tom. III. 2ds p.” Index Kewensis cites this reference as 1838 4(2), 1993] Peet— Aristida 37 Hitchcock 1924 as 1836 2(1), and Stafleu and Cowan as 1836 3(2). The pagination is consistent with 4(2). AND . 1849. Agrostidea III, Callus obconicus (Stipacea). Mém. Acad. St. Pétersb. VI. Sci. Nat. 7(2): 1-179. [The first page of the article indicates ““Conv. exhib. 1842.” This volume is bound as 1849 Tome 7, partie 2, but has a title page indicating 1849 Tome V, no partie indicated. The running footnote reads “Tom. V.” Hitchcock 1924 cites this as 5(1), but Index Kewensis cites it as Vol. 7. Stafleu and Cowan indicate that this is 1849 7(2), but note that there is an 1843 preprint.] Urtat, L. J. 1984. The type locations of the Flora boreali-americana of André Michaux. Rhodora 86: 1-66. DEPARTMENT OF BIOLOGY, CB#3280 UNIVERSITY OF NORTH CAROLINA CHAPEL HILL, NC 27499-3280 USA RHODORA, Vol. 95, No. 881, pp. 38-51, 1993 A REVISION OF THE CHAMAESYCE DELTOIDEA (EUPHORBIACEAE) COMPLEX OF SOUTHERN FLORIDA ALAN HERNDON! ABSTRACT Morphological investigation of the Chamaesyce deltoidea complex demon- strates a lack of sharp boundaries between the taxa. In combination with distri- butional data, the observed patterns of morphological variation suggest that the complex is best treated as a single polymorphic species. New combinations are proposed for Chamaesyce pinetorum and C. adhaerens to provide a consistent treatment for the morphological groups within the complex. The current conser- vation status of each morphological group is reviewed. Key Words: ieee deltoidea, subspeci plex, end d ies, ZeO- phical distinction, south ‘Florida L SP ESS | (CC 1 \ deltoide Small represents a complex of four closely related taxa found i in the rocky pinelands south of Miami and similar pinelands on the lower Florida Keys (Figure 1). Three of the taxa are currently listed as endangered (Bentzien, 1985; listed under the name Euphorbia deltoidea), and a clear understanding of taxonomic limits within the complex is required for proper application of provisions of the Endangered Species Act. However, current treatment of the group (Burch, 1965, 1966) does not satisfactorily account for patterns of variation found within the complex. Euphorbia deltoidea Engelmann ex Chapman was the first member of the complex to be described. It was included in the appendix (p. 647) of Chapman (1883). This species was trans- ferred to the genus Chamaesyce by Small (1903, p. 710, 1033). During the following three decades, Small proposed three new species closely related to C. deltoidea: C. pinetorum (Small, 1905), S. serpyllum (Small, 1913, p. 81, 155), and C. adhaerens (Small, 1927). Members of the complex are characterized by wiry stems radiating from a woody taproot, thick leaves with shallowly cor- date bases, cyathia on short peduncles solitary in leaf axils, and minute or absent gland appendages. These four taxa are endemic to southern Florida. Several West Indian species, particularly C. 1p A Ty ai eat Te 4 1c2: versity, Miami, FL 33199. Florida International Uni- 38 1993] Herndon— Chamaesyce deltoidea 39 Rn == qo /> deltoidea , rs fe Sie ee pinetorum Ny serpyllum Big Pine Key Figure 1. Map of southern Florida showing geographical distribution of the four 1 . Ps te land} Pe oe i 1 oe Ae diese te : s Gagne lengths for the three mainland subspecies. Each bar represents a separate popu- lation; length of bar is proportional to length of stem and leaf trichomes in the population. Only populations with accurately known locations are mapped. turpinii (Boissier) Millspaugh [including C. puertoricensis (Urban) Millspaugh] and C. centunculoides (H.B.K.) Millspaugh are close- ly related to the C. deltoidea complex, but are easily distinguished by possession of gland appendages equalling or exceeding the gland in width. In Small’s final treatment of the complex (Small, 1933), each member was considered to represent a distinct species. Major characters used to define the species were habit and pubescence. Minor distinctions were also noted in seed and cyathium char- acters (Table 1). More recently, Burch (1966) grouped the four members of the complex into two species distinguished from each other by habit. Chamaesyce deltoidea included all populations having prostrate stems (Small’s C. adhaerens, C. deltoidea and C. serpyllum) while C. pinetorum included all plants with erect or ascending stems. This treatment was adopted by Long and Lakela (1971). The most recent study of this complex was that of Remus (1979), and it purported to support Burch’s treatment. 40 Rhodora [Vol. 95 Table 1. Characters used by Small (1933) in distinguishing among the taxa of the Chamaesyce deltoidea complex. Character C. adhaerens C. deltoidea CC. pinetorum C. serpyllum Habit prostrate prostrate ascending _— prostrate Pubescence villous- glabrous villous- puberulent- hirsutulose hirsute canescent Leaf as long as as long as ———* longer than shape wide wide wide Cyathium about 1 mm less than 1 mm less than length 1 mm 1 mm Seed about 1 mm less than 1 mm less than length 1 mm 1 mm * This character not reported by Small. However, Remus used the name Chamaesyce adhaerens for dif- ferent plants than Small, and his conclusions actually imply a closer relationship between Small’s C. adhaerens and C. pineto- m than between C. adhaerens and C. deltoidea. In this study, a modified key for identification of the taxa is presented and relationships between taxa are discussed. The four taxa of Small are all treated as subspecies of Chamaesyce delto- idea. Only the subspecies designations are referred to in the text below. MATERIALS AND METHODS A survey of all extant potential habitat was carried out during the years 1980-81. Materials from throughout the current range of the taxa were collected for analysis during this study. Additional herbarium specimens were borrowed to supplement these collec- tions. A total of 277 specimens was examined. Several characters were measured on dried specimens. Leaf length was measured along the midvein and leaf width was mea- sured perpendicular to the midvein at the widest point of the leaf on up to five fully-exposed, flat leaves per sheet. Cyathium length and width (measured across the glands) were measured for up to seven cyathia per sheet. When available, cyathia with staminate or pistillate flowers at anthesis were measured, otherwise cyathia with young or mature fruit were measured. Seed length was mea- sured along the long axis of the seed using an optical comparator. Trichome length was measured or, when the trichomes were curved or bent, estimated on stem, leaves and cyathia using an optical 1993] Herndon—Chamaesyce deltoidea 41 comparator. Trichome density on stems and leaves was scored following the method of Isley (1953) where samples were assigned to density classes on the basis of the ratio between average tri- chome length and the average distance between trichomes. The youngest fully expanded leaves and the stem sections immediately below them were used for trichome density estimates. Trichome length was most conviently measured near the tip of a stem. Specimens preserved in Formalin—Acetic Acid—Alcohol or 95% Ethyl Alcohol were available for several populations covering the entire current range of the complex. Cyathium length and width measurements were taken from these specimens to verify and supplement results obtained from dried specimens. RESULTS Bar graphs of seed length (Figure 2) show a large degree of overlap among all four taxa. Seed length in ssp. serpyllum averages lower than in the other three subspecies, but the significance of this difference is obscured by the large degree of overlap with other subspecies. In the other characters used by Small, a similar picture is presented. In the few cases where average differences are found between subspecies, such as leaf shape, where ssp. ser- pyllum has narrower leaves in general than other taxa, overlap between taxa is extensive. Cyathium width proved to be an unsatisfactory character to measure from dried specimens due to variations in the amount of pressure on different cyathia during drying. Measurements taken from liquid-preserved material (Figure 3) showed that variation within populations was always high and often a large fraction of the total variation found within the complex. No consistent dif- ferences in cyathium size or shape were found to separate the taxa. Populations of ssp. de/toidea, ssp. pinetorum and ssp. serpyllum are easily identifiable, each having a distinct set of morphological characteristics. Subspecies deltoidea is characterized by prostrate stems, a predominantly short-shoot growth form with internodes shorter than leaf width, stems and leaves glabrous or very spar- ingly pubescent, and appressed uncinate trichomes. Subspecies pinetorum has erect or ascending stems, a predominantly long- shoot growth form with internodes usually 1.5 to 2 x leaf width, and a dense covering of long, straight, spreading hairs. Subspecies 42 Rhodora [Vol. 95 adhaerens 40 | deltoidea pinetorum PERCENT OF CASES 5 serpyllum 0:75 0.80 0.85 > 0.90: ~ 6:95 -1.00° 1.05 1.40- 1:15) 1.20 SEED LENGTH nnd Figure 2. Dp. L s eid bay! -_ 4] 4 sh. bspecies of the anette deltoidea complex (based on measurements of 35 pean for De each subspecies). Degree of overlap demonstrates that this character cannot be used to distinguish among the taxa. serpyllum is characterized by prostrate stems, a predominantly long-shoot growth form, and a very dense pubescence (which gives the plant a silvery appearance) of appressed, irregularly twisted trichomes. Subspecies adhaerens is here defined in terms of pubescence characters alone. It has appressed uncinate trichomes on leaf sur- faces and twisted, ascending to spreading, trichomes on stem surfaces. Trichome length, trichome density and plant habit vary in a Clinal fashion across the range of ssp. adhaerens. At the 1993] Herndon—Chamaesyce deltoidea 43 i ar _ = oo) on T T T CYATHIUM WIDTH (mm) © wo T 1.5 1S CYATHIUM LENGTH (mm) +t x Pe | +. A A+} 1 yw 5 i | preserved cyathia. Curves enclose measurements from four populations. Vari- ability within single populations is often a large fraction of the total variability observed in the complex. northern end of its range it approaches ssp. deltoidea in appear- ance, and at the southern end, it grades into ssp. pinetorum. DISCUSSION Subspecies pinetorum is not isolated from other members of the complex as the current classification (Burch, 1966) implies; it forms an evolutionary unit equivalent to ssp. deltoidea or ssp. serpyllum. The taxonomic rank of the three groups of populations is hard to define. They are more strongly differentiated than most subspecific taxa, and, if ssp. adhaerens were not present, could easily be considered weakly separated species. But, for the reasons given below, I treat them here as members of a single polymorphic species. The differences between all four members of the complex are vegetative and no differences have been found in their life his- tories or modes of reproduction. The vegetative characters used to define the differences are found to vary in a clinal fashion in the mainland populations, and these populations seem to inter- 44 Rhodora [Vol. 95 grade along broad zones of contact. Differences between the pop- ulations of ssp. serpyllum on the Florida Keys and the mainland subspecies are no different in kind than the differences between the extreme forms of the mainland subspecies. Finally, there is no geographical overlap of different morphological types to sug- gest barriers to interbreeding. One specimen of ssp. adhaerens preserved at ny (Small & Carter 860) indicates a collection lo- cality (by an arrow drawn on a map of Dade Co.) within the range I have defined for ssp. pinetorum. Another specimen of ssp. del- toidea at Ny (Small & Wilson 197 9) similarly indicates a collection locality outside of the range shown for ssp. deltoidea in Figure 1. No detailed information on the locality of either collection has been located nor has any reference by Small to finding any sub- species growing together. Burch (1965) stated that ssp. pinetorum is broadly sympatric with ssp. adhaerens and ssp. deltoidea, but this assertion is not supported by any specimens I have seen. Since neither the precisely located collections of Burch nor my own field survey produced any examples of two members of the complex growing together, I do not consider these suggestions of range overlap credible. In addition to the arguments presented above, it is relevant that qualitative differences in pubescence can be found within other species of Chamaesyce. In particular, C. turpinii, as pres- ently understood, has essentially the same range of pubescence types as do pubescent forms of C. de/toidea. In a different group, the current conception of C. polycarpa (Bentham) Millspaugh encompasses both glabrous and pubescent plants. Populations of ssp. adhaerens bridge the morphological gaps between the other three taxa. Stem habit within this taxon ranges from appressed to ascending and a mixture of long-shoots and short-shoots is often found on a single plant. The only charac- teristic that consistently separates ssp. adhaerens from ssp. del- toidea on one hand and ssp. pinetorum on the other is the oc- currence of irregularly twisted hairs in stem pubescence. It differs from ssp. serpyllum in having less dense pubescence with only uncinate trichomes on the leaf surfaces. Geographically, ssp. ad- haerens occupies the area between ssp. deltoidea and ssp. pine- torum (Figure 1). When trichome length, plant habit, or pubes- cence density are considered separately, the values found in populations of ssp. adhaerens smoothly bridge the gap between ssp. deltoidea and ssp. pinetorum. Close inspection of Figure | 1993] Herndon—Chamaesyce deltoidea 45 suggests that there is an abrupt change in trichome length at the boundary between ssp. deltoidea and ssp. adhaerens, but speci- mens with intermediate trichome length exist (Lakela & Almeda 30450 and Lakela & Pardue 31583) that are not mapped because their locations are not accurately known. The phylogenetic relationships of ssp. adhaerens within the complex are not fully clear. Currently, ssp. adhaerens is repre- sented by a small series of populations intermediate in morpho- logical features between ssp. deltoidea and ssp. pinetorum. Also, ssp. adhaerens is much more uniform morphologically within populations and has larger differences between populations than the other three taxa of the complex. This pattern of variation suggests that relatively recent hybridization between ssp. de/toidea and ssp. pinetorum might have given rise to ssp. adhaerens. Such a scenario is also supported by the apparent intergradation of ssp. adhaerens with those two along its borders. However, hybridiza- tion would only be possible if the putative parents were growing in much closer proximity than they are currently. Alternatively, it is possible that the pattern of variation reflects local adaptation to varying environmental factors, such as differences in soil type or water availability. Changes in substrate conditions are at least loosely correlated with the changes in morphological pattern. The range of ssp. deltoidea corresponds to that area of the Miami Rock Ridge overlain with a thin layer of white sand (Snyder et al., 1990). The ssp. adhaerens occurs in the area with a reddish sandy loam while ssp. pinetorum is largely found in the lower, wetter pinelands at the southwestern end of the Miami Rock Ridge. Unfortunately, the pinelands habitat of the complex has been so fragmented and reduced over the past decades that it is no longer possible to make detailed studies to compare these two hypoth- eses. In particular, the zones where the taxa would have met historically have been obliterated, so the abruptness of the tran- sitions cannot be assessed. It is also possible that the patterns of variation seen in present populations of ssp. adhaerens may reflect accidents of preservation more than historical patterns. Despite the uncertainty, I consider the hypothesis of local differentiation more probable on the basis of current evidence. As noted above, ssp. adhaerens does not have the morpholog- ical distinctiveness of the other members of the complex. Still, the populations are in most cases easily distinguished from the other taxa and occupy a significant portion of the range of the 46 Rhodora [Vol. 95 complex (Figure 1). The combination proposed by Burch in 1966 (Chamaesyce deltoidea ssp. deltoidea var. adhaerens) does not adequately describe the position of ssp. adhaerens within the complex. In particular, there is no reason to believe that ssp. adhaerens is more closely related to ssp. de/toidea than it is to ssp. pinetorum. For these reasons, and in accordance with the hypothesis that the various morphological types all arose inde- pendently through adaptation to local conditions, I use the rank of subspecies for all taxa in the complex. TAXONOMIC TREATMENT Chamaesyce deltoidea (Engelmann ex Chapman) Small Plants herbaceous, stems 10-15 cm long, wiry, radiating from a woody taproot, appressed to the ground surface or ascending to erect. Leaves reniform to ovate; thick, obscuring secondary venation; bases strongly inequilateral, commonly shallowly cor- date; margins entire, thickened, usually inrolled. Stipules mem- branaceous; usually joined to base and deltoid in outline, often lacerate or lobed, sometimes split fully to the base; white to red- dish; inconspicuous. Inflorescences solitary in leaf axils. Peduncle shorter than to slightly exceeding the leaf. Cyathia glabrous to densely pubescent, involucre campanulate, .9 to 1.9 mm long, .5 to 1.7 mm wide, lobes triangular, equalling to slightly exceeding the glands. Glands elliptic, .4 mm long, .3 mm wide. Gland ap- pendages minute or lacking. Capsule glabrous to densely pubes- cent, about 1.3 mm long. Seeds about 1 mm long, quadrate, brown, the flat surfaces usually having 3-4 shallow (sometimes inconspicuous) ridges. KEY TO SUBSPECIFIC TAXA OF CHAMAESYCE DELTOIDEA 1. Stems erect or ascending 2. Pubescence of leaves spreading, .6-.7 mm long, straight; pubescence of stems same .......... 3. ssp. pinetorum 2. Pubescence of leaves appressed, .1-.25 mm long; pubes- cence of stem spreading or ascending, .1-.25 mm long, irregularly twisted. .<.......<....... 2. ssp. adhaerens 1. Stems prostrate or appressed to ground surface . Pubescence of leaves sparse, appressed; stems glabrous or thinly pubescent; hairs on leaves uncinate; hairs on stems WANE i, 1. ssp. deltoidea 1993] Herndon—Chamaesyce deltoidea 47 3. Pubescence of leaves dense, appressed to spreading; pu- bescence of stems dense; hairs on leaves uncinate or ir- regularly twisted; hairs on stems irregularly twisted 4. Pubescence of leaves composed of irregularly twisted hairs; plants appearing silver or gray ............. PERERA CY Sp Rye eT Td ce hy 5 Ae epee es 4. ssp. serpyllum 4. Pubescence of leaves uncinate; plants bright green ... seg Laan etn re eae ed ae 2. ssp. adhaerens 1. Chamaesyce deltoidea (Engelmann ex Chapman) Small ssp. deltoidea, Fl. Southeastern U.S., p. 710, 1903. E ee deltoidea Engelmann ex Chapman, Flora of the Southern U. i es zy 47, 1883. Type: United States. Florida [Dade Co.], rocky pin q a Biscayne, June, Curtiss 2474* (LECTOTYPE: MO (1792963)! fern des- ignated); Isorypes: cu!, GH!, MIN!, NcBS!, NY [3 sheets]!, pH [2 sheets]!, us [2 sheets]!) Plants forming small mats with stems and leaves appressed to ground surface, bright green. Margins of adjacent leaves on most stems typically overlapping. Stems glabrous or very sparsely pu- bescent with appressed, uncinate trichomes .10-.25 mm in length. Leaves usually glabrous above, often sparsely pubescent beneath with appressed, uncinate trichomes .10-.25 mm in length. Cyathia and capsules glabrous to sparsely pubescent. Although Chapman credited Engelmann with this species, com- parison of the published description (Chapman, 1883) with the specimen in the Engelmann herbarium (Mo 1792963) and with a manuscript description by Engelmann (mo archives, Engelmann collection), shows that Chapman’s description was derived in- dependently. The Engelmann name was known to Chapman from one Or more specimens in the exsiccata series North American Plants distributed by A. H. Curtiss (Curtiss 2474*). All replicates of Curtiss 2474* that could be located in this country were ex- amined, but none was found that could be identified with the specimen Chapman used in drawing up his description. Burch annotated the specimen in the Engelmann herbarium as lectotype for the species, and I accept this choice. The remaining replicates of Curtiss 2474* have traditionally been treated as isotypes. Exsic- cata numbers in this series are not equivalent to collection num- bers, but * and ** are appended to exsiccata numbers elsewhere in the series to denote different collections of the same species. 48 Rhodora [Vol. 95 Also, specimens of Curtiss 2474* I have seen are consistent with replicates from a single gathering, so I acccept them as isotypes. The year of collection of Curtiss 2474* remains uncertain, but annotations on sheets received by Engelmann and the U.S. De- partment of Agriculture suggest that the specimens were collected in 1880. This is the most northerly taxon in the complex. Its historical range was in the dry Miami Rock Ridge Pinelands from about the center of Miami to some 13 miles south. All but a few rem- nants of this pineland area have been converted to commerical or residential use. In those remnants, the Chamaesyce grows in a thin layer of white sand over limestone bedrock. It is most abundant in areas free of shrubby undergrowth. Most of the re- maining populations are found on lands owned by the federal, state or local governments, and, given protection from develop- ment as well as periodic burns to keep the pinelands open, should persist indefinitely. Management regimes imposed in these pine- land remnants, however, could affect survival of the populations in unforeseeable ways. Continued population monitoring will be necessary to provide sufficient warning of potential adverse ef- fects. REPRESENTATIVE SPECIMENS: UNITED STATES. Florida: Dade Co., pinelands north of Goulds, 10 Aug 1963, Burch 144a (FLAS, MO, NY); rocky pine woods near Miami, 11 July 1895, Curtiss 5468 (F, FLAS, GH, MIN, NY, US); Miami, July 1877, Garber sn (F, FLAS [2 sheets], GH, Ny, Us [2 sheets]); in dry sandy pineland, Buena Vista, 26 Dec 1929, Moldenke 29] (Mo, NY, PH, US); Miami, 13 May 1904, Tracy 9114 (CU, F, GH, MIN, MO, NY, PH, US). 2. Chamaesyce deltoidea ssp. adhaerens (Small) Herndon, comb. et stat. nov. Chamaesyce adhaerens Small, Torreya 27: 104. 1927. Type: United States. Flor- ida, Dade Co., pinelands between Peters Prairie and Homestead, 10 Nov 1906, Small & Carter 2531 (HOLOTYPE: NY!). Stems and leaves appressed to ground surface and plants form- ing small mats or stems ascending and plants forming small tufts. Margins of adjacent leaves on most stems overlapping. Stems pubescent with ascending to spreading, irregularly bent trichomes .25-.55 mm in length. Leaf trichomes nearly the same length but appressed, uncinate. As discussed in the text, this taxon occupies the gap between 1993] Herndon—Chamaesyce deltoidea 49 ssp. deltoidea and ssp. pinetorum both geographically and mor- phologically. It is found in the Redlands area of Dade Co. where it grows on a fine, reddish sandy loam over limestone bedrock in remnants of the Miami Rock Ridge Pinelands. It is the rarest taxon in the complex, and the one most likely to be lost because no known populations are found on protected lands. REPRESENTATIVE SPECIMENS: UNITED STATES. Florida: Dade Co., badly burned pinelands on south side of Bauer Dr., east of Krome Ave., 3 May 1978, Avery 1884 (Everglades National Park, FLAs, USF); open pineland near Homestead, 16 Apr 1964, Burch & Ward 271 (F, GH, MO); open pine- i age cng aacaapieis 9 Aug 1980, Herndon 402 (FLAS, FTG; also distributed to sBsp, Pinelands, near Murden Hammock, 22 June 1915, Small, Mosier ‘ Small 645 i (FLAS, GH, NY). 3. Chamaesyce deltoidea ssp. pinetorum (Small) Herndon, comb. et Stat. nov. il sn pinetorum Small, Bull. N.Y. Bot. Gard. 3: 4. 1905. Type: United Florida, Dade Co., Pinelands near the homestead road, between iter and Longview Camp, 9-12 Nov 1903, Small & Carter 836 (HOLOTYPE: NY!; Isotype: pH!). Euphorbia smallii Oudejans, Phytologia 67: 236, 1989. Plants forming small tufts. Stems ascending to erect, usually red. Margins of adjacent leaves on most stems not overlapping. Stems, leaves, and capsules villous, with straight, spreading tri- chomes .6-.7 mm in length. In addition to the easily identified holotype and isotype, there are a few more potential isotypes which should be noted. These specimens have the same label data as Small & Carter 836, but do not have a collection number. One such sheet at US was an- notated as an isotype by Burch. Strongly upright, often red stems give this plant a highly dis- tinctive appearance. It grows in rarely flooded pinelands, although it seems to grow in slightly wetter areas than the other three taxa. This subspecies is currently the most abundant of the complex, primarily because almost a quarter of its original range falls within the boundaries of Everglades National Park. It is protected from direct habitat destruction, but its survival may be dependent upon the specific fire regime imposed. REPRESENTATIVE SPECIMENS: UNITED STATES. Florida: Dade Co., pinelands near Long Prairie, 24 Mar 1904, Britton 193 (F, Ny); open pine wood, Princeton, 30 Nov 1963, Burch 232 (FLAS, NY, US); open rocky pinelands, Homestead, 13 50 Rhodora [Vol. 95 Apr 1980, Herndon 348 (Fra; also distributed to asp, NCU, RSA, TEX); open rocky pinelands, Long Pine Key, Everglades National Park, 5 Jun 1981, Herndon 514 (FLAS, FTG; also distributed to yBsD, NCU, RSA, TEX); Pinelands about Sykes Ham- mock, 2 Jul 1915, Small, Mosier & Small 6761 (FLAS, GH, NY). 4. Chamaesyce deltoidea ssp. serpyllum (Small) Burch, Ann. Mis- souri Bot. Gard. 53: 90-99, 1966. —— serpyllum Small, Flora Florida Keys 81. 1913. Type: United States. , Monroe Co., Big Pine Key, 17 Nov 1912, Small 3811 (HOLOTYPE: me Euphorbia deltoidea ray rer ex Chapman var. serpyllum (Small) Oudejans, Phytologia 67: 45, Stems and leaves appressed to ground surface, forming small silver-green to white mats. Margins of adjacent leaves on most stems not overlapping. Stems, leaves and capsules densely pu- bescent with appressed to ascending, irregularly bent trichomes .1-.25 mm in length. The dense, silvery pubescence of this plant blends in amazingly well with the surface of the coral rock it grows on and probably also provides considerable protection against desiccation in the hot, dry pinelands of the lower Florida Keys. Much of the original range of this subspecies lies within the boundaries of the National Key Deer Refuge and is protected from direct habitat destruction. Continued commercial and res- idential development in the Florida Keys, however, could ad- versely affect the subspecies through changes in availability of fresh water or changes in fire regime. Currently, the subspecies is widespread in the pineland areas of Big Pine Key, but the pop- ulations are scattered, and the total number of plants does not appear to be large. REPRESENTATIVE SPECIMENS: UNITED STATES. Florida: Monroe Co., occa- sional in pine woods, Big Pine Key, 10 Aug 1963, Burch 139 (FLAS, GH [2 sheets], MO, NY); Big Pine Key, 8 Feb 1977, Correll & Popenoe 48057 (Frc, Ny); pine woods, Big Pine Key, 1-17 Feb 1937, Killip 32062 (Gu, Mo, Us); Pinelands, Big Pine Key, 12 Nov 1912, Small 3768 (F, Ny [2 sheets]); Pinelands, Big Pine Key, 17 Dec 1913, Small & Small 5027 (F, Ny). ACKNOWLEDGMENTS The courtesy of curators at AA, BH, CU, F, FLAS, FTG, GH, MIN, MO, NCBS, NY, PH, US, USF, Archbold Biological Sta- tion, and Everglades National Park in providing facilities for study during visits to their institutions or in loaning specimens for this 1993] Herndon—Chamaesyce deltoidea Si oer is greatly appreciated. I also thank the curators at IA, MICH WELC who graciously searched their collections for examples iv Curtiss 2474*. Useful reviews of this study and suggestions for improvement were provided by R. Sanders, M. P. Coons and W. Judd LITERATURE CITED BENTZIEN, M. M. 1985. Endangered and threatened wildlife and plants; endan- gered and threatened status for five Florida Pine Rockland plants. Federal Register 50: 29345-29349. Burcu, D. 1965. A taxonomic revision of the genus Chamaesyce (Euphorbi- aceae) in the Caribbean. Ph.D. dissertation, Univ. of Florida, Gainesville. 1966. Two new species of Chamaesyce (Euphorbiaceae), new combi- nations, and a key to the Caribbean members of the genus. Ann. Missouri Bot. Gard. 53: 90-99 CHAPMAN, A. W. 1883. Flora of the Southern United States, 2nd ed. Ivison, Blakeman, Taylor and Co., New York. Istey, D. 1953. Desmodium paniculntam (L.) DC. and D. viridiflorum (L.) DC. mer. Midl. Naturalist 49: 920-933. Lona, R. W. AND O. LAKELA. 1971. A Flora of Tropical Florida. University of Miami Press, Miami, FL. Remus, J. Y. 1979. She of two endangered Florida species of Chamaesyce (Euphorbiaceae). Florida Scientist 42: 130-136 SMALL, J. K. 1903. Flora of the Southeastern United States. University of North Carolina Press, Chapel Hill Additions to the flora of subtropical Florida. Bull. New York Bot. Gard. 3: 419-440. 1913. Flora of the Florida Keys. Privately printed, New Yor . 1927. A new Chamaesyce from tropical Florida (Chamaesyce adhaer- ens). Torreya 27: 104. 1933. Manual of the Southeastern Flora. University of North Carolina Press, Chapel Hill. SNYDER, J. R., A. HERNDON AND W. B. Ropertson. 1990. South Florida rock- land. Jn: Myers, R. L. and J. J. Ewel, Eds. Ecosystems of Florida. Univ. of Central Florida Press, Orlando. SOUTH FLORIDA RESEARCH CENTER EVERGLADES NATIONAL PARK HOMESTEAD FL 33030 and FAIRCHILD TROPICAL GARDEN 11935 OLD CUTLER ROAD MIAMI, FL 33156 RHODORA, Vol. 95, No. 881, pp. 52-75, 1993 THE INTERPLAY OF BOTANISTS AND TEM - D STEWARDSHIP OF A RARE SPECIES CHARLES V. COGBILL ABSTRACT shone bl Oakes ex Rydb. was first collected ca. 1824 by Thomas nroe Flats in the Presidential Range of New Hampshire. Partly os ause the sea has always been rare, botanists have searched for, collected, studied, hiked past, propagated, and protected it. This paper documents botanical exploration, taxonomic interpretations, and historical status of P. robbinsiana populations. In the last 160 years, human interplay with P. robbinsiana has gone through roughly three phases of intense interest: discovery, collection and stew- ardship. In each period, human perceptions of, and reactions to, the rarity of this species have changed. The most significant human impact on the Monroe Flats plants was trail disturbance and zealous herbarium collection in the late 1800’s. Even the cumulative removal of over 850 plants (equivalent to approximately half of the population) has not permanently affected the population size or dis- tribution on Monroe Flats. Active conservation over the past 75 years has sup- plemented the natural resilience of the species. A second tiny and imperiled population on Franconia Ridge, discovered by Edward Tuckerman, Jr. ca. 1840, has never drawn much management or botanical interest. Key Words: Potentilla robbinsiana, historical ecology, conservation biology, en- angered species, alpine zone, herbarium collections, Thomas Nut- tall, William Oakes, Edward Tuckerman, Jr., White ae, New Hampshire INTRODUCTION Potentilla robbinsiana Oakes ex Rydb. (dwarf cinquefoil), the only plant species endemic to the White Mountains of New ampshire, is singularly relict. It has persisted—perhaps at only two spots—since the alpine region was separated from the arctic some 9000 years ago (Spear, 1989). The nearest related species, Potentilla hyparctica Malte (Love and Léve, 1965, 1966), occurs km to the north at four sites in the western Shickshock Moun- tains of the Gaspé, Québec, Canada. The latter alpine plants are themselves equally isolated, 800 km from arcti (Hul tén, 8). Native Potentilla robbinsiana presently grows in two restricted locations. The primary population of 1500-2000 flowering in- dividuals occurs on Monroe Flats (Sargent’s Purchase, Coos a2 = x 1993] Cogbill—Potentilla History 53 County), 2 km southwest of the summit of Mount Washington and just south of Lakes of the Clouds. The area (1548 m elevation) is at the head of Oakes Gulf, just above treeline in the col between Mt. Monroe to the west and the south shoulder (Bigelow Lawn) of Mount Washington to the east. Here a low dome of till supports the most well-developed turf-banked terraces in the White Moun- tains. Most P. robbinsiana plants grow on the periphery of this knoll, primarily southeast of the barren crown. Some 30 km to the west, a second tiny population is found in the small alpine zone on Franconia Ridge (Franconia, Grafton County). This sta- tion of less than ten individuals is on a west-facing cliff (1460 m elevation) just south of Mt. Lincoln. Potentilla robbinsiana was listed as federally endangered in 1980 (Cook, 1980). Although a “classic rarity,” with both highly re- stricted geographic range and habitat specificity (Rabinowitz, 1981), the species is locally abundant at one locality and occurs in a distinctly different habitat at a second (Graber, 1980; Fitz- gerald et al., 1990). Further dimensions of P. robbinsiana’s rarity include close human contact and taxonomic ambiguity. Despite recent population studies and political concern (Steele, 1964; Gra- ber, 1980; Graber and Brewer, 1985; Fitzgerald et al., 1990), the botanical history of P. robbinsiana has been little known except for herbarium collections. Herbarium Collections Herbaria provide an unparalleled database on historic collec- tion locations, collector activity, and taxonomic treatment. Of the 96 public herbaria contacted for information on Potentilla robbinsiana and its synonyms, 59% (57) had some plant material (Table 1). Specimens are found today in herbaria from St. Johns- bury, Vermont to Grahamstown, South Africa. Herbaria typically contain massive collections from the late 19th century. Although large New England herbaria have traditionally been tapped for local information, they contain less than 20% of the more than 235 existing sheets of P. robbinsiana. Previous studies using 21 to 50 sheets (Pease, 1924; Steele, 1964; Graber, 1980; New Hamp- shire Natural Heritage Inventory files) have given a limited pic- ture of collection history. Significantly, most early and most Fran- conia collections are found in European herbaria. The extensive herbarium material represents one of the most 54 Rhodora [Vol. 95 Table 1. Number of separately labelled collections of Potentilla robbinsiana known from various public herbaria. Fran- Monroe Flats conia Location 1824—1858-—1916— 1840— Total Sheets Plants Herbaria 47 1909 83 1915 (#4) (#) ~~ () Large New England (GH, NEBC, HNH, NHA, VT) 8 49 1 2 60 42 220 Other New England (in 10 of 15 checked) 9 aT 6 2 44 41 135 Large American (US, NY, PH, F, MO, CAN) 5 54 3 3 65 56 210 Other American (in 19/42) 2 57 ij 1 67 61 201 Outside N. America (in 17/28) 28 6 4 5 43 35; 113 All known (n = 57/96) 52 OS 6 421 12.2738. 233-48 intensive collections ofa single species yet documented. It is even more impressive considering most collections are from a single site. Herbaria contain 279 separate labels of 879 individual plants collected from ca. 1824 to 1983 (Table 1). Forty-eight collectors are represented, but over half of the plants were taken by only six botanists (W. Oakes—20 sheets; E. Tuckerman, Jr.— 23 sheets; C. G. Pringle—28 sheets; Faxon brothers—17 sheets; W. W. Eg- gleston—55 sheets; and E. F. Williams— 13 sheets). In addition, many specimens were not deposited (e.g., are in personal herbaria, used for transplants or cultivation, or lost). Since the 846 known specimens collected at Monroe Flats represent a minimum num- ber, it is probable that collectors have removed well over a thou- sand plants. DISCOVERY The dates, locations, nomenclature, and even origin of most early reports of Potentilla robbinsiana are unclear. Conflicting accounts credit at least five botanists, in addition to Robbins, with its discovery. The original publication of Potentilla robbin- siana cites material from “Nuttall! Oakes! Tuckerman!” (Torrey and Gray, 1840). While Thomas Nuttall apparently collected it in 1824, the earliest dated herbarium specimen (ups) is from William Oakes and Charles Pickering in 1825. In 1840, Edward 1993] Cogbill— Potentilla History 53 Tuckerman, Jr. wrote that ““Mr. Oakes who discovered it at this station had it only in fr[uit]” (CuCo 3/1840). In a letter written in October 1840, Tuckerman acknowledged that both Nuttall and Oakes “had only previously found the plant,” but intriguingly added that “Mr. Greene had also found it” (Qk). Benjamin D. Greene was at Lakes of the Clouds in 1823 when he found at least two other rare species (Pease, 1917, 1964), but P. robbinsiana is notably absent from Greene’s collections cited in Bigelow (1824). Without a specimen, field notes or substantiating account, any reports before 1824 are uncertain. Thomas Nuttall Either Asa Gray or John Torrey had “seen an authentic spec- imen” of Potentilla robbinsiana from Nuttall. Retaining chro- nological priority, they listed Nuttall first (Torrey and Gray, 1840). Although no dated material survives, Thomas Nuttall almost certainly collected Potentilla robbinsiana in August 1824. Notes on herbarium sheets of P. robbinsiana indicate that in September 1839, Nuttall told Edward Tuckerman, Jr., “that he also found it” (UPS, Qk). Charles Pickering (1879) even wrote that in 1824 “Nuttall [was] visiting the White mountains, meeting with . . . on the alpine portion of the mountains, Potentilla frigida [= P. rob- binsiana], Gnaphalium supinum ... and Salix repens.” Nuttall made only one trip to the White Mountains, obviously collecting alpine plants in August 1824 when he signed Crawford’s register (CrPa, 12/8/1824; Pease, 1918; Graustein, 1967). A sheet of S. uva-ursi (= “S$. repens’’) from Nuttall’s herbarium (BM) is indeed labeled ‘“‘White Mts., 1824...” There are no collections of P. robbinsiana now in Nuttall’s herbarium, but one undated speci- men of “Potentilla brauniana Hoppe” from the “White Moun- tains” attributed to Nuttall is in Hooker’s herbarium at Kew. This sheet (k) thus appears to be part of Nuttall’s original collection and the earliest surviving collection of Potentilla robbinsiana. William Oakes was competitive and thus skeptical of Nuttall’s rarities. He referred to Gnaphalium supinum as “‘Nuttall’s doubt- ful” species (OaCoGH, 4/1829) and summarily dismissed Nut- tall’s Carex atrata as variously misidentified (e.g., TuPa, 8/1839). Apparently in reaction to Torrey’s publication acknowledging Nuttall’s collections, Oakes wrote to Torrey that Nuttall had told him that Nuttall “*. .. did not find them [Potentilla minima and Arnica mollis] himself.” Oakes asserted that ““Nuttall’s specimens 56 Rhodora [Vol. 95 ... were given him ... by Dr. Holmes of Maine, who was pos- terior to P[ickering] and myself...’ (ToCo, 4/1841). In fact, Ezekiel Holmes’ only known trip to Monroe Flats was on 27 July 1825, when he “coasted along by blue pond and Mt. Monroe...” in the company of Oakes (CrPa, 7/1825). Confusingly Oakes must have been responsible for any collection by the young Holmes. It seems likely that Oakes’ self-serving story was simply trying to obfuscate the discovery of P. robbinsiana by the reticent Nuttall (cf. Graustein, 1964, 1967). William Oakes The major early collections of Potentilla robbinsiana are at- tributed to William Oakes. During several weeks in 1825 Oakes made “tours around the mountains” collecting “‘some rare [plants] . not to be found elsewhere in America” (Crawford, 1846). Oakes, together with Charles Pickering, made collections of a dozen new alpine species (Oakes, 1841; Pickering, 1879; Pease, 1964). They apparently found P. robbinsiana at this time, as the only dated herbarium sheet was collected by “‘Pickering & Oakes, July 1825” (ups). In 1827 both Oakes and Pickering returned to Ethan Crawford’s house (CrPa, 18—28/7/1827). Oakes collected Potentilla robbin- siana in July 1827, as he wrote to John Torrey that he had “‘just returned from the White mountains and had found 3 more flow- ering specimens of the little Potentilla . . . [with] several hundred specimens observed” (ToCo, 8/1827). In addition, Tuckerman stated (ups) that Oakes and Pickering found P. robbinsiana in 1827 and a collection (Qk) attributed to Pickering alone must be from either 1825 or 1827 (Pickering, 1879; Pease, 1964). In April 1828, Oakes sent Torrey one flowering specimen of ‘“‘Potentilla frigida” (ToCo, 15/4/1828). A single specimen in a small packet labeled “‘Potentilla frigida, Vill., Wh. Mts., Oakes” (Ny) fits this description and is apparently contemporary, albeit not obtained directly from Torrey’s herbarium. Other Early Collections Before its first published mention in 1840, Potentilla robbin- siana was known by many. In 1829, Ethan Crawford helped “some 1993] Cogbill— Potentilla History 57 botanists from Boston . . . [collect] an assortment . . . [of] delicate mountain flowers” to send to Europe (Crawford, 1846). These plants probably included P. robbinsiana (pr) and the type of Pre- nanthes bootii (G; Pease, 1917; Graustein, 1964) collected by J. W. Boott in 1829, as well as the P. robbinsiana (pr) attributed to Francis Boott’s herbarium (Lehmann, 1851). Oakes and Tuck- erman both made multiple collections of P. robbinsiana, and at least four other botanists had herbarium collections by 1840. Some of Oakes’ collections come from his later explorations, 1843-47 (Pease, 1964). For example, two fruiting specimens, “‘Le- git Oakes” (Ny), are from the Princeton (perhaps Torrey’s) her- barium, and the nomenclature ‘“‘Potentilla minima” dates them in the 1840’s. Oakes certainly got P. robbinsiana in 1843 when he collected all known rare species in the White Mountains (Pease, 1917). In 1844, Oakes first discovered Euphrasia oakesii at an unknown location, but probably on Monroe Flats where it has always grown together with P. robbinsiana (Oakes, 1847; Rob- inson, 1901). Despite contemporary knowledge of Nuttall’s, Oakes’ and Boott’s early collections, they were later overlooked. For example, Asa Gray (1867) only cited collections attributed to “Robbins, Tuckerman, & c....” Following Gray, researchers (e.g., Allen, 1888; Rydberg, 1898; Fernald, 1950; Graber, 1980; Lanier and Hill, 1983) have erroneously believed that James W. Robbins discovered Potentilla robbinsiana. In 1829 William Oakes, the source for the specific name, sent Robbins on a highly successful collecting trip to the White Mountains (Oakes, 1841; Pease, 1917). Although Oakes had mailed him a specimen of P. robbinsiana in April 1829 (OaCoGH, 4/1829) and “directed him exactly . . . [to] the locality of the rarer plants[,] . . . Robbins found none of [them]” (TuCo, 4/1841). Neither herbarium sheets nor any report of P. robbinsiana by Robbins has been found. Ironically, it appears that Robbins was one of the few botanists who visited the area before 1830 who did not collect his namesake. Furthermore, the possessive form of the colloquial name, Robbins’ (sic) cinquefoil, is a misnomer. The Latin adjectival suffix (-iana) implies that Robbins was simply being complimented (Stern, 1966). Although he admittedly did “not care much for the Lindley rule” (TuCo, 12/1843), Oakes clearly recognized discoverers with a genitive (-ii suffix) specific epithet. 58 Rhodora [Vol. 95 Locale of First Collections Until 1840, Potentilla robbinsiana’s location was “obscure & out of the way of travellers” (CuCo 3/1840). No P. robbinsiana collected before 1839 bears an exact location, but circumstantial evidence indicates that all came from Monroe Flats. Early ex- plorers (e.g., Bigelow and F. Boott in 1816) used an eastern route up the White Mountains (OaCoAAS, 10/1844), and none reported P. robbinsiana. Collections apparently started only after the first Crawford Path was cleared on the west side in 1819 (Crawford, 1846). This route passed near Mt. Monroe, but after 1821 most trips up Mount Washington were made over a more northerly second Crawford Path (Waterman and Waterman, 1989). The latter route was a “poor one for botanizing” (Southwick, 1842), and botanists obviously explored interesting areas near “E. A. Crawford’s old path” (Oakes, 1841). Both Nuttall and Oakes visited the Monroe region well before a bridle path first crossed Monroe Flats ca. 1840 (Waterman and Waterman, 1989). In 1824 Nuttall found Gnaphalium supinum in a “dry ravine of the Amo- noosuck” (Torrey and Gray, 1840), conceivably the edge of Mon- roe Flats. Similarly, Nuttall’s collection of the rare Carex atrata (Beck, 1833) is consistent with a site near Oakes Gulf. In 1825, Oakes first visited the “fertile locality” of Oakes Gulf (Pease, 1917). All of Oakes’ sheets in 14 herbaria are simply from “Alpine Region, White Mountains, New Hampshire,” but Tuckerman definitively states that Oakes found P. robbinsiana at only one station (CuCo 3/1840; ak). Significantly, Tuckerman wrote that P. robbinsiana was found “by Mr Oakes & Dr. Pickering at the same station where I found it” (Ups). This location, unquestion- ably Monroe Flats, was the “peculiar home of the alpine cinque- foil’ (Tuckerman, 1860). The flora of the Franconia Range was “very poor” (OaCoAAS, 10/1844), but Potentilla robbinsiana was found there at first ex- ploration. Although the new site was never publicized, Edward Tuckerman, Jr. collected five undated herbarium sheets (7 spec- imens) from the “Franconia Mountains” (Ny, K, UPS, E). One sheet (ups) elucidates the location as “Great Haystack”” where Tuck- erman first recorded alpine species in 1840 (TuCo, 10/1840; Tuckerman, 1843). Tuckerman was distributing P. robbinsiana in Europe in 1841-42 where most of his Franconia specimens are now found. In the late summer of 1844, Tuckerman returned to 1993] Cogbill— Potentilla History 59 Franconia Ridge for a “‘very thorough exploration of the Gt. Haystack.” He then stated that ‘“*Potentilla (minima) is limited to the spot where I first found it & here only grows Polygonum vivip. ...” (OaCoAAS, 10/1844). Thus his Franconia discovery was presumably ca. 1840 from a single station. Interestingly, Tuckerman’s collection of Polygonum viviparum (NY) was the only time this species has ever been recorded from Franconia. SYSTEMATICS Early Taxonomic Uncertainty Early American botanists were initially confused about the tax- on to which their alpine Potentilla belonged. Without an Amer- ican counterpart, most early collections of the dwarf Potentilla from the White Mountains were compared to various European taxa. For example, Nuttall’s ca. 1824 specimen (k) is labeled “‘P. Brauniana’’; F. Boott’s 1829 plant (PR; Lehmann, 1851) has the undeterminate “Pot. sp.” annotated to P. nivea?; and J. W. Boott’s 1829 companion collection (PR) is identified as P. villosa. While there was some early agreement that the plants belonged to P. brauniana, Oakes initially decided it belonged to another Euro- pean alpine species. In 1827, Oakes simply called it the “little Potentilla’’ and stated that the specific taxon ‘“‘cannot be settled until European botanists will make up their minds” (ToCo, 8/1827). Soon afterward, Oakes called it “Potentilla frigida” when he sent specimens to John Torrey (ToCo, 4/1828). Ironically, Torrey’s reply asked if Oakes was “convinced that this cannot be the species we first considered it (P. nana Lehm.)” (OaCoAAS, 4/1828). Prophetically, Torrey’s early determination placed it with the then little known arctic species to which it is now considered most closely related. Early collectors had trouble recognizing the species without a name or description, yet taxonomists were dependent on their collections. For instance, Tuckerman was still unaware of Poten- tilla robbinsiana 15 years after its discovery and he failed to recognize it as he first crossed Monroe Flats. In June 1839 he collected a “‘Pl[anta] incog[nita] near snow Mt. Monroe” (TuPa, 24/6/1839). Two days later, he refined his identification of the unknown plant to “Planta Rosacearum.” Later that summer, Tuckerman had identified the rose’s genus (TuPa, 8/1839), and 60 Rhodora [Vol. 95 specifically discussed the Potentilla collections and taxonomy with both Oakes and Nuttall. After an evening talk with Oakes, Tuck- erman recorded their uncertainty: “Potentilla -—- Mr Oakes doubtful about” (TuPa, 2/8/1839). Oakes, nevertheless, must have developed the idea of a distinct species, as two weeks later Tuck- erman annotated a specimen collected earlier that summer: “‘Po- tentilla Robbinsiana Oakes! MSS.” (NEBc). In September 1839, Nuttall also told Tuckerman that he still “considered it distance [sic] from P. Brauniana or P. verna or nivea”’ (UPS). Although Oakes coined the name “Robbinsiana,” he never pub- lished a description and apparently only communicated the name privately. The first published description of the taxon listed “‘Po- tentilla Robbinsiana, Oakes! mss.” as a synonym for an unnamed variety of the species P. minima (Torrey and Gray, 1840). Sig- nificantly, one 1825 collection by Oakes has an original label with the same nomenclature: “‘Potentilla Robbinsiana Oakes MSS.” (ups). All Oakes’ other specimens or manuscript material referring to “Robbinsiana” have apparently been relabeled or lost. Con- fusingly, most of Oakes’ herbarium sheets have printed labels with Potentilla minima and P. Robbinsii (sic) as a synonym. These labels, however, were added by Asa Gray after Oakes’ death in 1848 (Gray, 1849). They were part of sets of “‘most of the rarer plants found in New England” which were sold (at $20 per 500 species) accounting for the wide distribution of Oakes’ herbarium. Significantly, Oakes seldom, if ever, wrote the name “‘Robbin- siana” himself, rather choosing “‘frigida” on both surviving her- barium material and correspondence before 1840 (e.g., Ny; ToCo, 3/4/1828; OaCoGH, 4/1829). After Torrey and Gray submerged P. Robbinsiana under P. minima, Oakes changed his designation and refered to the species as “‘Potentilla minima” (e.g., ToCo, 4/1841; TuCo, 12/1843). Curiously, even after receiving credit for the species name Robbinsiana, Oakes rejected it. Taxonomic Revision The long lack of recognition, due in part to confused taxonomy, was finally settled by Asa Gray. Working on the Roseaceae for Torrey’s Flora of North America in the 1830’s, Gray examined Potentilla material, including Lehmann’s herbarium (GrCo, 10/ 1839; Torrey and Gray, 1840). Torrey and Gray (1840) decided that the White Mountains species “‘agrees well with the character 1993] Cogbill— Potentilla History 61 of P. minima’ Haller, the accepted synonym of Potentilla braun- iana Hoppe (Koch, 1837). They relegated ““P. Robbinsiana”’ to a variety (Torrey and Gray, 1840). Interestingly, Gray only saw plants in the field in September 1842, while on a trip with Tuck- erman (GH, OXF). The first edition of Gray’s Manual (1848) continued to recognize P. minima with the short-stemmed, small- flowered American variety. Although Potentilla expert Lehmann (1851) concurred exactly with this designation, Gray (1856) later transferred the taxon to P. frigida Villars. After being accepted as a variety of various European species for over 70 years, P. robbinsiana was validly given full species status by Rydberg (1896). Others (e.g., Britton and Brown, 1897; Rydberg, 1898; Wolf, 1908; Robinson and Fernald, 1908) concurred with this determination. The genus Potentilla is known for its “devilish combination of apomixis, hybridization, and morphological plasticity” (Polunin, 1959). Due in part to the obscure evolutionary history of this “anomalous species” (Fernald, 1950), the taxonomic status of P. robbinsiana is still unclear (e.g., Fernald, 1931; Léve and Love, 1965, 1966). One chromosome count from Monroe Flats showed an odd number polyploid 2n = 49, which would indicate an obligate apomict (Léve and Live, 1966). Potentilla robbinsiana is indeed agamospermous (T. D. Lee, pers. comm.). Although Love and Love (1966) concluded that P. robbinsiana “‘is appar- ently an apomictic segregate of Potentilla hyparctica,” the taxon could also be a hybrid conserved by apomixis (Steele, 1964). Potentilla hyparctica itself is not well defined, but is typically 2 = 42 (Malte, 1934; Live and Léve, 1965; Dansereau and Steiner, 1956; Fernald, 1943; Moore, 1982). A distinct ancient hybrid between P. hyparctica and a 2n = 56 taxon (known in other arctic species in the genus) producing the 2n = 49 Monroe Flats’ plant seems plausible. While being related to P. hyparctica (Léve and Live, 1965, 1966; Crow, 1982), this hybrid would explain the enigmatic nature of P. robbinsiana. Potentilla robbinsiana is morphologically distinctive, probably a direct result of its limited genetic variability. For example, flow- ering stem lengths of plants from the White Mountains overlap only slightly with those of P. hyparctica (Table 2). Even alpine P. hyparctica from the Gaspé can be easily distinguished from the White Mountain population (Fernald, 1950; D. Marris, mT her- barium note 1972). When grown in the greenhouse, however, Monroe Flats plants take on some of the characteristics of their 62 Rhodora [Vol. 95 Table 2. Measurements of fruiting stem length from late-July and August herbarium collections of Potentilla from the White Mountains, New Hampshire, the mountains of the Gaspé and the low arctic in Québec, Canada. Plants bid ueniannaiel Fruiting stem length (mm) Collection (n) Mean* +SD Range Monroe Flats Nuttall ca. 1824 1 18.10 18.1 Oakes pre-1848 19 16.74b S61 7.2-26.5 Tuckerman ca. 1840 11 22.91la 6.42 13.1-33.1 Since 1880 194 22.74a 5.88 6.3-39.6 Franconia Tuckerman ca. 1840 6 45.38c 11.65 26.0-57.7 North 1897-1908 2 18.50 8.34 12.6-24.4 Fernald 1915 1 20.30a 4.09 13.1-26.2 Québec (P. Ayparctica) Gaspé a7 69.53d 22.95 35.1-103.7 Ungava 24 68.26d 20.78 31.8-123.7 * Different letters indicate a significant (t-test, P < .05) difference between means. arctic relatives (Love and Live, 1965; K. D. Kimball, pers. comm.). Nevertheless, the geographic isolation, morphological differences, narrow ecological range and asexual breeding system all argue for the continued species-level separation of P. robbinsiana. Collections from the Franconia Range in the 1840’s are different from either Monroe Flats or recent Franconia plants. Their hab- itat, presumably damp mats with Polygonum, was also odd. Tuck- erman labeled the specimens “forma aestivalis” and on one sheet indicated the character of “‘pedunc. elongatis” (Ny). Long fruiting stems (unfortunately there are no flowers), place them between P. robbinsiana and P. hyparctica (Table 2). These remarkably different plants perhaps represented genetically distinct stock. PRESENT DISTRIBUTION Population on Monroe Flats In 1839 Potentilla robbinsiana was found on a “stony alpine moor on Mount Monroe” (TuPa, 6/1839; Tuckerman, 1843). Its distribution was then described as “very limited & very [struck out] obscure” (CuCo 3/1840). Since 1865, herbarium records 1993] Cogbill—Potentilla History 63 have cited it as specifically along the Crawford Path (Gu). The original Crawford Path skirted the southeast edge of the knoll within the upper edge of the present range of P. robbinsiana (Fig- ure 1). Reminiscences from the 1960’s also place the majority of the P. robbinsiana plants on the southeast side of the knoll and figuratively ““moving toward” the outcrop (C. Harris, pers. comm.). A survey of this main population in 1967 revealed a distribution east of the trail, very similar to the present range (W. N. Tiffney, Jr. and B. K. Fowler, pers. comm.). The .11 ha area occupied by the main population has changed little from 1973 to the present (Graber, 1980; Graber and Brewer, 1985; Figure 1). In addition to the main congregation of Potentilla robbinsiana near the rock outcrop at the south edge of knoll, there has been a handful of small outlier populations on Monroe Flats (Figure 1). These sub-populations, scattered over the .85 ha barren area, are in marginal habitats and may be ephemeral. Currently, three peripheral populations, all “discovered” since 1983, are scattered to the north of the main population. In 1985, these outliers had a total of 15 flowering individuals; this number decreased to nine plants in 1992. For over 40 years, a notable outlier population was found west of the post-1918 Crawford Path. About 1900, Potentilla robbin- siana grew there “in dry stony soil’ together with Euphrasia oakesii (Robinson, 1901). In 1946, Steele (herbarium notes) still found both species ‘ton the southwestern side of a shaly hump.” This P. robbinsiana sub-population contained a limited number of plants (i.e., 12 to 50 individuals as estimated by F. Steele, C. Harris, W. N. Tiffney, Jr.) and was last seen about 1969. Inter- estingly, the only extant colonies of Euphrasia are now east of the trail just beyond Potentilla’s range; apparently both of these species have been extirpated from the west side of Monroe Flats. Steele’s often-quoted conclusion (e.g., Cook, 1980; Graber, 1980; Lanier and Hill, 1983) that P. robbinsiana now occupies one- fourth of its 1934 area is based on the disappearance of the re- stricted sub-population west of the trail (Steele, personal herbar- ium notes, 1977). A colony 30 m south of the outcrop similarly decreased from 40 individuals in 1973 (Graber and Brewer, 1985) to none in 1988. Overall, the small sub-populations seem prone to extirpation while the main population has remained viable for centuries. In addition to native plants, there are currently several trans- 64 Rhodora [Vol. 95 er Tm — Z Figure 1. Map of Monroe Flats, Presidential Range, New Hampshire showing the current range of Potentilla robbinsiana. Shaded area is the extent of the main 1993] Cogbill—Potentilla History 65 planted populations of P. robbinsiana established in the White Mountains. Three programs (i.e., Harris, 1964; Graber, 1974-85; AMC, 1986-89) have transplanted more than 500 individuals, grown from Monroe Flats stock, into at least six locations (i.e., Camel Patch, Oakes Gulf, top of Monroe Flats, Lakes Hut, Boott’s Spur, North Lafayette). Since the 1820’s, Potentilla robbinsiana has been common on Monroe Flats. In August 1827, William Oakes observed “several hundred specimens” (ToCo, 8/1827). Considering its inconspic- uous nature after flowering, Oakes’ casual record is certainly a gross underestimate. Between 1824 and 1847, botanists removed more than 96 plants from the site, including at least 50 by Oakes himself. In several years at the turn of the century, hundreds of plants were collected at once. A minimum population of many hundreds must have been maintained as this heavy and sustained collecting occurred without comment or extirpation. In 1977, a casual estimate of the “population of ca. 300 plants” (NHA: Crow #2449 & Hellquist) was remarkably similar to that of 1827. Much of the population, however, was again overlooked as Graber (1980) had counted 1801 large plants in 1973. Some 1701 flowering plants were counted in 1992 (M. Iszard-Crowley, pers. comm.) and the population on Monroe Flats seems to have been relatively stable over the past 20 years (Graber and Brewer, 1985; Fitzgerald et al., 1990). For the last 165 years the population has certainly been in the hundreds, and probably had been in the thousands. Population on Franconia Ridge The first known Franconia Potentilla robbinsiana were near Mt. Lafayette. Tuckerman’s 1840’s collections were most likely from the accessible summit of “‘“Great Haystack,” the early name for Mt. Lafayette. On 10 July 1897 during an Appalachian Mountain Club (AMC) field meeting (Anon., 1898), Frederic Endicott col- lected a single P. robbinsiana from “‘near north summit of Mt. Lafayette” (GH). The third and last known collection was on 6 — of abandoned trails, specifically: A—the original Crawford Path (ca. 1840-191 8); B—the Crawford Path (1915-83); and C—the Dry River Trail (ca. 1938-65). Large rocks and outcrops are solid. A scree wall built in 1983 to delimit the restricted area is shown as a chain. 66 Rhodora [Vol. 95 July 1908 by an unknown “Hunt” on “Mt. Lafayette” (CONN). This collector was perhaps William S. Hunt, an amateur who collaborated with Endicott in producing a list of alpine plants collected on AMC trips (Fuller et al., 1907). The Mt. Lafayette station was apparently extant in 1908, but extirpated by 1915. Since 1915, all Potentilla robbinsiana from Franconia have been found at a station well south of Mt. Lafayette. On 18 July 1915 during a weekend trip over Franconia Ridge, M. L. Fernald and his graduate student, F. J. Smiley, found P. robbinsiana on “damp gravel and crevices about the low north dome of Mt. Lincoln” (Gu; FeCo, 8/1915). Fernald returned in August 1915 to scour the ridge, finding rare plants elsewhere, but P. robbinsiana was apparently restricted to the one site. On 13 June 1963, Fred Steele found P. robbinsiana still on Mt. Lincoln. Two plants, including one with 50 flowers which was the largest Steele had seen, were growing with dead moss “on sloping rock” or “sort of ledge” along the trail north of Mt. Lincoln (Steele, pers. comm.). Botanists, including Steele, saw three plants in July 1965 (Steele, personal herbarium notes), but were unable to locate any in the 1970’s. On 24 June 1984, AMC hutman Rick Boyce, found two living P. robbinsiana when he “just happened to see a yellow flower” while hiking just south of Mt. Lincoln. The following year Charles Cogbill and Roger Collins found four additional flowering individuals in crevices on the cliff directly below these plants. Sketchy records of Potentilla robbinsiana indicate several small historic populations scattered along Franconia Ridge. All Fran- conia reports total only 51 individuals. Their extremely limited numbers, spotty occurrence, and habitat on gravel, ledges, or cliffs, set them apart from those on Monroe Flats. Some 38 plants are known to have been collected from Franconia and 29 (57% of the plants ever known there) are from a single collection by M. L. Fernald (11725) in 1915. None has been collected since 1915. Only 13 separate flowering plants have been seen in thorough searches over the last 30 years (e.g., Steele, 1964; Crow, 1982; Cogbill, pers. obs.). The Franconia population was never large; its current (1991) population of five flowering individuals is im- periled at the known remaining station. Discounted Locations Several herbarium collections from Mt. Washington sites are mislabeled. Tuckerman’s 1839 herbarium labels (ups, MASS, NEBC) 1993] Cogbill—Potentilla History 67 read: ““Mt. Washington, N.E. [sic] side of the Peak.” Despite incorrect dates and directions on the labels, his notes are explicit that he collected from Monroe Flats, “‘S.E. [sic] of the peak” (TuPa, 6/1839) and “not elsewhere” (ups). Beaudry’s collection from “Mt. Washington: 7-mile post” on 26 July 1958 (mT) was also mislabeled. His co-collectors that day did not find it there, actually collecting P. robbinsiana only from Monroe Flats (A. L6éve, pers. comm.; Louis-Marie, QFA field notes). Two odd col- lections from ““Tuckerman’s Ravine”’ (NEBC: no collector and no date, but a H. G. Jesup 1890's label) and “Alpine Garden” (BSN: collector J. H. Emerton, 1907) have no other substantiation. Some herbarium sheets have indefinite locations such as “alpine region, New England,” “White Mountains,” or “Mt. Washington’’; all except those from Franconia apparently are from the single Mon- roe Flats site. No verifiable record of Potentilla robbinsiana exists outside New Hampshire. Speculation about a Vermont site is based on a specimen (NEBC) labeled “‘Mansfield Mt. E.T.” found in 1979 among Robbins’s material. Considering the notation, it has been interpreted as an Edward Tuckerman collection. Interestingly, Tuckerman’s only known trips to Mount Mansfield, Vermont were in 1839 and 1840, and his field notes explicitly indicate that he did not find Potentilla then (TuPa, 7/1839; 7/1840). Appar- ently the label is not in Tuckerman’s hand and was written after 1842 (Oakes, 1842). This enigmatic collection credited to Ver- mont (e.g, Crow, 1982; Gleason and Cronquist, 1991) remains dubious. Recent references (e.g., Countryman, 1980; Graber, 1980; Gleason and Cronquist, 1991) to an unvouchered sighting on a Vermont mountain from 1970 are based on a misidentification of P. norvegica from Mt. Equinox (Graber and Brewer, 1985). PERCEPTIONS AND REACTION TO RARITY Discovery Era: Potentilla minima 1824-60 Much of the early botanical exploration of the White Mountains was driven by the search for rare alpine species, including several found near Monroe Flats (Tuckerman, 1843; Bigelow, 1824; Oakes, 1847; Pickering, 1879; Pease, 1917, 1964). By the 1840’s, Oakes considered a couple of local species ‘approaching exhaustion,” presumably due to over-collection of rarities (Pease, 1917). Sig- nificantly, Potentilla minima never received the same attention 68 Rhodora [Vol. 95 as these other species and was not even listed as “rare” (Spaulding, 1855). Lack of prominence of this species was exemplified by the fact that no one took credit for its discovery and by the failure of anyone, especially Nuttall or Oakes, to publicize or publish any description Early interest: in Potentilla robbinsiana was driven by taxonom- ic considerations. Collectors were uncertain what taxon they found and were selectively exchanging specimens and intelligence (e.g., OaCoAAS, 4/1828; TuPa, 8/1839; TuCo, 12/1843). Even Tuck- erman’s early collections must have been limited as he was very concerned over the loss of some of his “rare” flowering P. robbinsii (sic) specimens (CuCo, 3/1840; 9/1840) and indicated that “my specimens are exhausted” (Qk, 10/1840). Oakes’ and Tucker- man’s extensive and widely dispersed collections from Monroe Flats were fundamental in taxonomic work. Once Gray deter- mined that it was part of a European alpine species (Torrey and Gray, 1840), it was no longer a “new” species and collections and interest diminished. Its well-known location, albeit obscure and disjunct from the Alps, rendered the White Mountain P. minima less of a rarity (TuCo, 12/1843; Tuckerman, 1860). Collection Era: Potentilla frigida 1860-1910 Following an increased public interest in natural history, vast botanical collections were made between 1860 and 1910 (Table 1). Completion of the carriage road to the summit of Mount Washington in 1861 gave convenient access to the long-known site at Monroe Flats. In June 1862, three collectors began an era of group collecting by taking 34 plants (17 sheets). During the collection peak of the Victorian era, botanists sought to document carefully the alpine flora by making regular collecting trips to Monroe Flats (Pease, 1917). For example, “in the last week of May [18787], Edwin Faxon climbed from Crawford Notch “‘to gather Potentilla frigida in flower in its alpine home, . . . the little plain by the ice-clad Lake of the Clouds” (Kennedy, 1900). Zeal- ous botanists made mass collections on specific forays (e.g., 60 plants by 5 collectors on 5 June 1878; 59 plants on 15 August 1898; and 75 plants on 2 June 1903). In 1898 alone, 143 known individuals were collected. In all, some 73% of P. robbinsiana collections date from 1875 to 1909 with more than 403 individ- uals taken during just 9 years at the turn of the century. At its 1993] Cogbill— Potentilla History 69 maximum, collection pressure was significant, removing an equivalent of perhaps one-third of the present population in a decade. In the late 1800’s, botanists expressed little concern about Po- tentilla frigida. Monroe Flats continued to supply unrestricted collections as many sought rarities to expand their personal her- baria (e.g., Northrop and Northrop, 1910). The two most prolific collectors, Pringle in the 1870’s and Eggleston in the 1890’s, were even selling alpine plants, specifically including P. robbinsiana, to other collectors at 10 cents per sheet (Anon., 1878; EgPa, 1901). A few such as G. M. Allen (1888) recognized the “incursions of botanists,” but he still “took but a tiny spray ... [of P. robbin- siana] ... and left the remainder of the plant.” The western edge of Potentilla robbinsiana habitat has histor- ically been close to trails (Figure 1), in particular part of the original Crawford Path (ca. 1840-1918) and the Dry River Trail (ca. 1938-65). For over 150 years, collectors, horses, and hikers passed through the population on a cleared topographic treadway (Waterman and Waterman, 1989). Improvement of the Crawford “bridle” Path, trampling by its users, and easy access for collectors have arguably been the most significant human impact on the plants. Nevertheless, this trail section, abandoned in 1918, now supports a vigorous population of at least 71 flowering P. rob- binsiana. Collecting and trail activities from 1840 to 1918 ap- parently did not permanently affect the population. Stewardship Era: Potentilla robbinsiana 1910-83 In 1915 Fernald made the last mass collection of Potentilla robbinsiana, nearly eliminating the population on Franconia Ridge. This was Fernald’s only collection of this species, and reflected earlier unrestricted practice. Significantly, only 21 collections con- sisting of 56 specimens have been made at Monroe Flats since 1910 (Table 1). There were at least 15 individual collectors, but many explicitly showed restraint. For example, despite long-term interest in the plant, S. K. Harris collected only 3 plants for herbaria (1927-30) and F. Steele collected only 5 single plants (personal herbarium, 1934-56). In contrast, botanists associated with field excursions of the International Botanical Congress in Montreal in 1959, despite being asked not to collect it (Léve and Léve, mimeographed field guide, 1959), made at least six collec- 70 Rhodora [Vol. 95 tions (19 plants). In the past 30 years, only three sheets of single plants have been collected, all by researchers working on the plant. Recently, plants have been used for chromosome counts (COLO: Love & Love, 1962), greenhouse cultivation (A. L6ve, pers. comm.), transplants (Harris, field notes, 1964: 6 plants), vouchers (NHA: Crow and Hellquist, 1977), and “‘collected accidentally during a survey of the population” (msc: Brewer, 1983). In the mid-1900’s, an emerging st hic spread among botanists and the hiking community. The very restricted distri- bution of Potentilla robbinsiana was documented and it was con- sidered “probably the rarest plant to be found on the Presidential Range” (Harris, 1930, 1941). Botanists told their colleagues the plant is “scarce . . . please do not collect!’ (L6ve and Léve, mim- eographed field guide, 1959), and asked the public to “‘please help protect this rare plant’ (Bliss, 1963). Responding to increased curiosity about the plant, Harris and others in the 1960’s showed visitors the small population west of the trail, rather than disturb the main population (C. Harris, pers. comm.). The Dry River Trail was even rerouted by those (i.e., S. K. Harris, W. N. Tiffney) who were concerned about its effect on erosion and several rare species (C. Harris, pers. comm.). Although recent human tram- pling has been considered a crucial impact (Graber, 1980; Graber and Crow, 1982; Graber and Brewer, 1985), effects associated with the post-1918 Crawford Path are minor in contrast to those in the past. Due to efforts by interested botanists (e.g., F. Steele, R. Graber, G. Crow), the species was determined to be federally endangered. Following a recovery plan, Monroe Flats was closed to access in 1983 (Lanier and Hill, 1983). After nearly 75 years of informally decreasing human pressure, the species was legally protected from disturbance or collection. Recent human activities have been di- rected toward augmenting populations through trail relocation, access restriction, population monitoring and transplant pro- grams (Fitzgerald et al., 1990). After 160 years of human influ- ence, Potentilla robbinsiana is still tied to human intervention in the management era. ACKNOWLEDGMENTS Colleagues K. D. Kimball, T. D. Lee, B. T. Fitzgerald, R. Paul and J. Doucette were part of a larger project on the population 1993] Cogbill— Potentilla History 71 dynamics, genetics and management of Potentilla robbinsiana, and helped me in many ways, including a review of earlier ver- sions of the manuscript. F. Steele, C. Harris, A. Love and W. N. Tiffney, Jr. all eagerly shared data and their memories of previous work. The following archives gave access to manuscript material: American Antiquarian Society; Amherst College; Dartmouth Col- lege; Gray Herbarium, Harvard University; New England Botan- ical Club; New York Botanical Garden; University of North Car- olina, Chapel Hill and University of Vermont. Many thanks to the curators of the following herbaria (acronyms follow Holmgren et al., 1990) who kindly provided information, access to their collections, and/or photocopies: B, BEDF, BKG, BM, BP, BR, BRU, BSN, BUF, C, CAN, CAS, CGE, CHRB, CLM, CM, COLO, CONN, CU, DAO, DBL, DUKE, DWC, E, F, FI, G, GA, GH, GRA, H; HNHAILE; IND; BK: KSC,.L, LE, LIV, LU, LY, 12, M, MAINE, MANC, MASS, MICH, MIN, MO, MSC, MT, MTMG, MU, NACS, NCU, NEB, NEBC, NHA, NY, NYS, OKL, OSU, OXF, P, PAC, PH, PM, POM, PR, QFA, QK, S, SCNH, SFS, SJFM, SYRF, TENN, TEX, TRT, TUFT, U, UBC, UC, UCS, UNB, UPS, US, USF, VINS, VPI, VT, W, WELC, WIS, WTU and YU. The U:S. Forest Service, Ammonoosuc District of the White Mountain National Forest, facilitated access to the closed habitat, and the Appalachian Mountain Club provided logistical support. Partial financial support for this study came from contracts with the Appalachian Mountain Club and the U.S. Fish and Wildlife Service. MANUSCRIPTS CITED Citations to unpublished manuscripts are abbreviated: CrPa—Crawford’s register, 1823-35, Lucy Crawford Papers, MS 626 (3), Dartmouth College Special Collections CuCo—letters: Tuckerman to Curtis 9/3/1840, 11/9/1840, Moses Ashley Curtis Papers, #199, Southern Historical Collection, University of North Carolina Library, Chapel Hill EgPa—printed sheet [1901], Box 1, folder 10, W. W. Eggleston papers, MS-28, Dartmouth College Special Collections FeCo—postcard: Smiley to Fernald, 15/8/1915, M. L. Fernald correspondence, Gray Herbarium GrCo—letter: Gray to Hooker, 13/10/1839, A. Gray correspon- dence, Gray Herbarium a2 Rhodora [Vol. 95 OaCoAAS~— letters: Torrey to Oakes 24/4/1828, folder 18; and Tuckerman to Oakes 10/1844, folder vs W. Oakes Corre- spondence, American Antiquarian Societ OaCoGH— extracts of letters: Oakes to Robbins 4/1829, 24/7/ 1829; and Oakes to Torrey 9/6/1834, folder 31; Biog 2 Oa 4, Box W, W. Oakes correspondence, Gray Herbarium ToCo—letters: Oakes to Torrey: 3/8/1827, 3/4/1828, 15/4/1828, 3/4/1841; J. Torrey correspondence, New York Botanical Garden TuCo—letters: Oakes to Tuckerman: 16/4/1841, vol. 3; 30/12/ 43, vol. 5; Torrey to Tuckerman, 8/10/1840, vol. 3; E. Tuckerman, Jr. correspondence, American Antiquarian So- ciet TuPa—botanical journals, 1838-39, Box 1, folder la; 1840, Box , folder 1; Tuckerman Botanical Papers, Amherst College Archives HERBARIUM RECORDS CITED Citation of herbarium acronym (Holmgren et al., 1990) indi- cates data or quotations from material contained on labels, an- notations, or notes attached to herbarium sheets from that her- rium. LITERATURE CITED — G.M. 1888. American cinque-foils. Popular Science Monthly 32: 189- 99. pre 1878. Pringle’s plants. Bull. Torrey Bot. Club 6: 199. ———. 1898. Report of the excursion committee Appalachia (1898): 397-401. Beck, L. C. 1833. Botany of the Northern and Middle States. Webster and Skinners, Albany, NY. Buiss, L. C. 1963. Alpine zone of the Presidential Range. Privately published, Edmonton, Alberta BiGELow, J. 1824. Florula Bostonensis, 2nd ed. Cummins and Hillard, Boston, MA Britton, N. L. AND A. Brown. 1897. Illustrated Flora of the Northeastern United States and Adjacent Canada. Charles Scribners and Son, New York. Cook, R. S. 1980. Determination of Potentilla robbinsiana to be an endangered species with critical habitat. Federal Register 45(182): 61944-61947 CouNTRYMAN, W. D. 1980. Vermont’s endangered plants and the threats to their survival. Rhodora 82: 163-171. 1993] Cogbill— Potentilla History 73 CRAWFORD, L. 1846. The White Hills, reprinted in 1966 as Lucy Crawford’s History of the White Mountains. S. Morse, Ed. University of New England Press, Hanover, N Crow, G. E. 1982. So England’s Rare, Threatened, and Endangered Plants. GPO, ype DC. DANSEREAU, P. AN E. STEINER. 1956. Studies in Potentillae of high latitudes and altitudes. - ‘Central Baffin Island populations. Bull. Torrey Bot. Club 83: 113-135. FERNALD, M. L. 1931. Specific segregations and scaseiapslh in some flowers of eastern North America and the Old World. Rhodora 9. ——. 1943. Two later homonyms. Rhodora 45: 111-1 12. ——. 1950. Gray’s Manual of Botany, 8th ed. American Book Co., New York, NY. FITZGERALD, B. T., K. D. KIMBALL, C. V. CoGBILt AND T. D. Lee. 1990. The biology and management of Potentilla robbinsiana, an endemic from New Hampshire’s White Mountains. New York State Mus. Bull. 471: 163-166. FULLER, T. O., F. Enpicotr, W. S. HUNT AND W. R. Davis. 1907. The flowers of Mount Washington. Among the Clouds, 17 July 1907. p. 3. GLEASON, H. A. AND A. CRoNnquisT. 1991. Manual of Vascular Plants of North- eastern United States and Adjacent Canada, 2nd ed. New York Botanical Garden, Bronx, NY Graber, R. E. 1980. The life history and ecology of Potentilla robbinsiana. Rhodora 82: rode WER. 1985. Changes in the population of the rare and endangered ek Potentilla robbinsiana Oakes during the period 1973 to 1983. Rhodora 87: 449-457. G. E. Crow. 1982. Hiker traffic on and near the habitat of Robbins cinquefoil, an endangered plant species. New Hampshire Agric. Sta. Tech Bull. 522. GrausTEIN, J. E. 1964. Early scientists in the White Mountains. Appalachia 30(7): 44-63. ——.. 1967. Thomas Nuttall, naturalist, explorations in America, 1808-41. Harvard Univ. Press, Cambridge, MA. Gray, A. 1848. A Manual of the Botany of the Northern United States. J. Munroe and Co., Cambridge, MA. ———. 1849. Herbaria of New England plants. Amer. J. Sci. 8: 293. ——. 1856. A Manual of the Botany of the Northern United States, 2nd ed. G. P. Putnam and Co., New York. 1867. A Manual of the Botany of the Northern United States, 5th ed. Ivison, Phinney, Blakeman, and Co., New York. ssa: S. K. 1930. Some characteristic plants of Mount Washington. Bull. ston Soc. Nat. Hist. 55: 5-11. . 1941. More plants of the Presidential Range, the saxifrage and rose families. Appalachia 7(91): 320-328. Ho.tmaren, P. K., N. H. HOLMGREN AND L. C. BARNETT. 1990. Index Her- a Part I: The Herbaria of the World. New York Botanical Garden, Bronx, NY. ieee E. 1968. Flora of Alaska and Neighboring Territories. Stanford Uni- versity Press, Palo Alto, CA. pp. 607-616 74 Rhodora [Vol. 95 Kennepy, G. G. 1900. Edwin Faxon. Rhodora 2: 107-1 Kocn, W. D. J. 1837. Synopsis florae germanicae et ese F. Wilmans, Frankfurt am Main, German LANrER, J. AND B. J. Hitt. 1983. Robbins Seige recovery plan. U.S. Fish and Wild. Serv. and U:S. For. Serv., Fr, © LEHMANN, J. G.C. 1851. Pugillus 9: a Love, A. AND D. Léve. 1965. Taxonomic remarks on some American alpine net Univ. Colorado Studies Ser. in Biol. 17: 23-24. 1966. Cytotaxonomy of the alpine vascular plants of Mount Washington. Univ. Colorado Studies Ser. in Biol. 24: 37. Matte, M. O. 1934. Critical notes on plants of arctic America. Rhodora 36: 172-179. Moore, D. M. 1982. Flora Europea Check-list and Chromosome Index. Cam- bridge Univ. Press, Cambridge, Englan Norturop, J. I. AND A. R. NorRTHROP. 1910. Notes on the plant distribution of Mt. Washington, N.H., pp. 255-258. In: H. F. Osborn, Ed., A Naturalist in the Bahamas. AMS Press, New York. Oakes, W. 1841. Notice of some of rare plants of New England with descriptions of new species. Hovey’s Mag. Hort. 7: 178-186. 1842. Catalogue of plants. Jn: Z. Thompson, Ed., Natural History of Vermont, Z. Thompson, Burlington . 1847. Notice of some of the slants of New England. Hovey’s Mag. Hort. 13: 217-218. Pease, A. S. 1917. Notes on the botanical exploration of the White Mountains. Appalachia 24(2): 158-178. ——. 1918. Nuttall and Pickering in the White Mountains. Rhodora 20: 39. . 1924. Vascular flora of Coéds County, New Hampshire. Proc. Boston Soc. Nat. Hist. 37: 39-388. 1964. A Flora of Northern New Hampshire. New England Botanical Club, Cambridge, MA. PICKERING, C. 1879. Chronological History of Plants. Little and Brown Co., Boston, MA. Potunin, N. 1959. Circumpolar Arctic Flora, pp. 276-277. Clarendon Press, Oxford, England. RaBinow!Tz, D. 1981. Seven forms of rarity, pp. 205-217. In: H. Synge, Ed., The Biological Aspects of Rare Plant Conservation. J. Wiley and Sons, New York. Rosinson, B. L. 1901. The North American Euphrasias. Rhodora 3: 270-272. AND M. L. FERNALD. 1908. Gray’s Manual of Botany, 7th ed. Amer. Book Co., New York. RypserG, P. A. 1896. Notes on Potentilla—III. Bull. Torrey Bot. Club 23: 301- 307. 1898. A monograph of the North American Potentilleae. Mem. Dept. Bot. Columbia Univ. 2: 1-223. ck, E. E. 1842. Notes of a tour to the White Hills in July 1841. (in- cluding) List of plants collected by Mr. E. W. Southwick, on the White Mountains of New Hampshire, “hos 15, 1841, with notes and aivtoaces by J. Barratt. The Classic 2: 174-18 SPAULDING, J. H. 1855. Plants of “ alpine regions of the White Mountains, 1993] Cogbill—Potentilla History 75 pp. 94-97. In: Historical Relicts of the White Mountains. Nathaniel Noyes SPEAR, R. W. 1989. Late-Quaternary history of high-elevation vegetation in the White Mountains of New Hampshire. Ecological Monographs 59: 125-151. STEELE, F. L. 1964. Potentilla robbinsiana in the White Mountains of New Hampshire. Rhodora 66: 408-411. STERN, W. T. 1966. Botanical Latin. Hafner, New York. pp. 294-298. TorreEY, J. AND A. Gray. 1840. A Flora of North America 1(3): 361-544. Wiley and Putnam, New York. TUCKERMAN, E., Jr. 1843. ee on some interesting plants of New England. aie. J. Sci. 45: 27-49. 1860. The vegetation the White Mountains, pp. 230-241. Jn: T. S. King, Ed., The White Hills. Crosby, Nichols, Lee, and Co., Boston, MA. WATERMAN, L. AND G. WATERMAN. 1989. Forest and Crag. Appalachian Moun- tain Club, Boston, MA. Wo r, T. 1908. Monographie der Gattung Potentilla. Biblioth. Bot. 71: 1-714. RR2 BOX 160 PLAINFIELD, VT 05667 and RESEARCH DEPARTMENT APPALACHIAN MOUNTAIN CLUB GORHAM, NH 03845 RHODORA, Vol. 95, No. 881, pp. 76-78, 1993 SALIX VESTITA PURSH AND SAXIFRAGA OPPOSITIFOLIA L.: ARCTIC-ALPINE SPECIES NEW TO NOVA SCOTIA RENE J. BELLAND AND W. B. SCHOFIELD During the course of bryological field work carried on for the Canadian Parks Service (Cape Breton Highlands National Park), we noted rare vascular plants but made no comprehensive col- lections. On the cliffs above Corney Brook, however, we discov- ered two arctic-alpine species not reported previously for the flora of Nova Scotia. Salix vestita Pursh and Saxifraga oppositifolia L. are here reported as new to the flora of the province: Inverness County, Cape Breton Highlands National Park, Corney Brook gorge, 46°43’N, 59°53’W, 6 July, 1991. Elevation ca. 280 m. Both species were extremely rare, each represented by a single plant. The single Salix vestita (Belland, Schofield & Moniz de Sa 16587, CAN) grew in a large, sheltered, deep crevice as a shrub ca. .75 m tall. The specimen was pistillate and appeared to be healthy; peeehae nan were immature. The single Saxifrag tin an intermittent seep on a | partly ue rock face. The plant was small, lacked flowers or fruit, and appeared somewhat etiolated. Because of the scarcity of S. oppositifolia at the site, no vouchers were taken of this species. A clear photograph (CAN) was made and careful ob- servations confirmed its identify. The distinctive ‘‘four-ranked opposite bristle-ciliate dorsally keeled . . . leaves” (Fernald, 1950) readily identify this species even when etiolated. We are con- vinced that other populations exist in the Cape Breton Highlands National Park, which await discovery through careful exploration of cliffs in upper reaches of watercourses. The Corney Brook site is rich in arctic-alpine and boreal dis- junct plants. On the same cliff we found Achillea borealis Bong., Carex atratiformis Britt., C. capillaris L. var. major Blytt., C. scirpoidea Michx., Cystopteris fragilis (L.) Bernh., Draba arabi- sans Michx., Erigeron hyssopifolius Michx., Galium kamtschati- cum Steller, Lycopodium selago L., Pinguicula vulgaris L., Pri- mula mistassinica Michx., Saxifraga aizoides L., S. aizoon Jarg. 76 1993] Belland and Schofield— Salix and Saxifraga 77 var. neogaea Butters, Scirpus caespitosus L. var. delicatulus Fern., Sedum rosea (L.) Scop., Senecio pauperculus Michx., Solidago multiradiata Ait. and Polystichum lonchitis (L.) Roth. Among the res lam are Cyrtomnium hymenophylloides (Hueb.) Nyh. ex. p., Entodon concinnus, Orthothecium strictum Lor., Plagiobry- um zierii (Hedw.) Lindb. and Pohlia filiformis (Dicks.) Andr. These rare plants occurred on humid, north-facing cliffs. The site possesses many of the characteristics described by Hounsell and Smith (1968) for habitats in which arctic and boreal disjunct plants occur in Nova Scotia. The cliffs are sheltered in many places by tall trees. This effect, combined with the aspect, would create a cool microclimate throughout the growing season and one which would experience considerable snow accumulation during the winter. Although many of the plants at the Corney Brook site are cal- cicoles, there is no evidence of calcareous rock (e.g., limestone, dolomite, basalt) at the site. In fact, many of the bryophyte species present grow only on noncalcareous rock (e.g., Andreaea rupestris Hedw., Dicranodontium denudatum (Brid.) Britt. ex Williams, Diplophyllum taxifolium (Wahlenb.) Dum., Paraleucobryum lon- gifolium (Hedw.) Loeske). The population of Salix vestita in Cape Breton represents its southernmost locality in eastern North America. The species is best represented in the Rocky Mountains, and occurs disjunct in the Hudson Bay area and in the Gulf of St. Lawrence region (Meusel et al., 1965) where it extends northward along the Lab- rador coast as far as 59°N (Nachvak). The Canadian Rocky Moun- tain distribution has been mapped recently by Brayshaw (1976) and Moss and Packer (1983). The world distribution is mapped by Meusel et al. (1965). The station for Saxifraga oppositifolia also marks an unusual record; it is widely distributed elsewhere at calcareous sites in the Gulf of St. Lawrence region, occurring along the entire west coast of Newfoundland, and along the North Shore of Québec as far west as the Mingan Archipelago. The southernmost populations in eastern North America are in New England, on Mt. Mansfield, and the Willoughby Cliffs and Job Pond Cliffs, Vermont (Crow, 1982; Seymour, 1982) where it is local on the cliffs. These two taxa, each represented by a single individual, rep- resent very rare native species in the province; their protection by virtue of their location in the National Park is assured. 78 Rhodora [Vol. 95 ACKNOWLEDGMENTS We thank Mario Moniz de Sa for his able assistance in the field and J. Bridgland of the Canadian Parks Service for providing invaluable assistance in arranging logistic support. Financial support for this work was provided by a Canadian Parks Service contract (No. 1632-89-023, to RJB). Additional funding was provided by a Natural Sciences and Engineering Re- search Council of Canada grant (No. A-1217, to WBS). Reprint requests should be directed to W. B. Schofield. LITERATURE CITED BraysHAw, T.C. 1976. Catkin bearing plants of British Columbia. Occas. Pap., British Columbia Prov. Mus. Series No. 18. Crow, G. E. 1982. New England’s Rare, Threatened, and Endangered Plants. U.S. Government Printing Office, Washington, FERNALD, M. L. 1950. Gray’s Manual of Botany. American Book Co., New York HounsELL, R. W. AND E. C. SmitH. 1968. Contributions to the flora of Nova Scotia. IX. Habitat studies of arctic-alpine and boreal disjuncts. Rhodora 70: 176-192 MEusEL, H., E. JAGER AND E. WEINERT. 1965. Siieiaaal Chorologie der Zentraleuropaivchan Flora. Gustav Fischer Moss, E. H. AnD J.G. PAcKER. 1983. Flora of ee, 2nd ed. Univ. of Toronto Press, Toronto, Ontario. Seymour, F. C. 1982. The Flora of New England, 2nd ed. Phytologia Memoi R.J.B. 2260 CORAL REEF PLACE VANCOUVER, B.C. V5P 4T3 CANADA DEPARTMENT OF BOTANY UNIVERSITY OF BRITISH COLUMBIA VANCOUVER, B.C. V6T 1Z4 CANADA RHODORA, Vol. 95, No. 881, pp. 79-81, 1993 BOOK REVIEW Stuckey, Ronald L. 1992. Women Botanists of Ohio: Born before 1900. 67 pp. Columbus, Ohio: Ohio State University, pb. Copies available from Ronald L. Stuckey, RLS Creations, P.O. Box 3010, Columbus, OH 43210. (Priced $11.50, in- cludes mailing.) Dr. Ronald L. Stuckey and the late Dr. Emmanuel D. Rudolph, botanists at Ohio State University, are interested both in botanical history and in the role of women in botanical science. Dr. Rudolph has written about the broader aspects of this subject; Dr. Stuckey is an expert on early mid-west botanical history and the part women have played. This collection of biographies was published for the May, 1992 meeting of The Council on Botanical and Horticultural Libraries held at Columbus, Ohio. Librarians and historians of science, especially, appreciate such a gathering of scattered source material under one cover. Stuckey started to gather data on early Ohio botanists when he came to Ohio State in 1965. In 1973, he wrote a twenty-page biography and bibliography for the Michigan Botanist of Ohio’s Dr. Emma Lucy Braun, one of our country’s most eminent ecol- ogists and botanists. The theme for the Ohio Academy of Science meeting in 1982 was ‘“‘Women in Science.” Dr. Stuckey presented a paper on fourteen women botanists born before 1900 who had made significant contributions. His continued research on this subject culminates in this book which is dedicated to the memory of two younger Ohio women botanists with distinguished teaching careers: Dr. Janice Carson Beatley (1919-87) and Dr. Jane Cyn- thia (McLoughlin) Decker (1935-88). In the United States, before the beginning of the twentieth century, some 1454 women were active in various aspects of botany, according to Dr. Rudolph. Of fifty-five Ohio women in this list, Dr. Stuckey has selected eighteen for his book. His sketch- es include a picture, if available, biographical information, edu- cation, and mention of their most important publications and awards. Some of the women were professional botanists, some were amateurs, but all published papers and/or books on the results of their studies. Many were students of William Kellerman who came to Ohio State in 1891 (died 1908), and John H. Schaff- ner who was on the faculty from 1897 to 1931. Two of the women 79 80 Rhodora [Vol. 95 married these professors. Stella Victoria (Dennis) Kellerman (1855-1936) assisted her husband with illustrations for his books, and in research. Independently she published her own research on the morphology of plants and flowers, notably on the tassel of the corn plant. Mabel (Brockett) Schaffner (1869-1906) assisted her husband and also did independent research. Her paper on the embryology of shepherd’s purse (Capsella bursa-pastoris) is a clas- sic. The first woman botanist in Ohio was Eliza Griscom Wheeler (1817-50) who married William Sterling Sullivant, the distin- guished bryologist, in 1834. She worked alongside her husband in the field, in the herbarium and in the office, and illustrated many of his books. By the mid- to late 1800’s a number of women had become involved in collecting and in publishing local floras. The best known of this group was Harriet Louise Keeler (1844— 1921), a teacher for forty years who wrote seven semi-popular books on northeastern native plants including The Wild Flowers of Early Spring, Our Native Trees, Our Northern Shrubs, and The Wayside Flowers of Summer. Emily Katherine (Dooris) Sharp (1846-1935) often accompa- nied her doctor husband on his medical rounds by carriage in Madison County and developed a keen interest in collecting and studying plants and their habitats. She published her observations in a small book, Summer in a Bog, in 1913. Her chapter on the activities of earlier women botanists in Ohio was the only real source of such information until Dr. Stuckey’s book. She added some 500 specimens to the herbarium under Dr. Kellerman’s direction. Dr. Clara Gertrude Weishaupt (1898-1991) was an outstanding teacher of general botany and local flora at Ohio State. She de- veloped The Vascular Plants of Ohio (3rd ed., 1970), for her courses. She contributed the section on Gramineae in Braun’s Vascular Flora of Ohio, The Monocotyledonae, Vol. 1. Concur- rently with her teaching she was curator of the Ohio State Uni- versity Herbarium, and was instrumental in reorganizing and revitalizing it after World War II. Dr. Stuckey concludes his book with a six-page tribute to Dr. E. Lucy Braun, most famous for her pioneering studies in forest ecology and author of Deciduous Forests of Eastern North America (1950), Woody Plants of Ohio (1961) and Vascular Flora of Ohio: The Monocotyledonae, Vol. 1. There is an eleven-page bibliog- 1993] Book Reviews 81 raphy, a summary and additional statistics. Besides the botanical information in the book, Dr. Stuckey has included a set of the fourteen calendars which make up the Gregorian Calendar. These calendars, seven for even years and seven for odd years, are print- ed on the same pages as the portraits of the botanists. An expla- nation of the calendars is given in pages vili and ix of the Intro- duction. They are included because “TIME is the most important thing in your life.”” One might add that all eighteen of these women obviously made the very best use of time in their very full and accomplished lives. MARY M. WALKER 14 CHESTNUT ST. CONCORD, MA 01742 RHODORA, Vol. 95, No. 881, pp. 82-84, 1993 BOOK REVIEW Robert R. Ireland and Linda M. Ley. 1992. Atlas of Ontario Mosses. Syllogeus 70, Canadian Museum of Nature (v + 138 pp., including 490 maps and 2 figs.). Available from the Canadian Museum of Nature, Direct Mail Section, P.O. Box 3343, Station D., Ottawa, Ontario, Canada K1P 6P4. (No charge.) Due to the efforts of Dr. Robert R. Ireland as curator of the bryophyte collections at the Canadian Museum of Nature, an enviable body of bryophyte data has been accumulated there. This atlas is another in a continuing series of publications of distributional data regarding the Canadian moss flora that has been accumulating for a long time. As Chairman of the Canadian Botanical Association Bryophyte Checklist Committee, Ireland organized, with other Canadian bryologists, a Checklist of the Mosses of Canada (Ireland et al., 1980), which effectively pro- vided species lists for every province and territory in Canada. The checklist was revised seven years later (Ireland et al., 1987). Ireland became familiar with the moss flora of western Canada while studying for his doctorate under Elva Lawton at the Uni- versity of Washington, Seattle, making contributions to Lawton’s Mosses of the Pacific Northwest, an important flora of the north- western United States and adjacent Canada (Lawton, 1971). Upon becoming curator in the Bryology Section of the Canadian Mu- seum of Nature at Ottawa, Ireland proceeded with taxonomic revisions in the Bryophyta, including treatment of neotropical bryophytes, and also concentrated on floristic research in the Maritime Provinces, Quebec and Ontario. In 1975, he, together with Ray F. Cain, produced a checklist of the mosses of Ontario detailed by political subunit (county, etc.) (Ireland and Cain, 1975), and in 1982, Ireland produced a major flora with keys, descrip- tions and illustrations of the moss flora of the Maritime Provinces (Ireland, 1982). In 1987, with Gilda Bellolio-Trucco, he wrote a useful, illustrated guide to many of the hornworts, liverworts, and mosses of eastern Canada for amateur botanists or professionals with little experience in the Bryophyta. Throughout the course of research activity, he has made skillful and generous use of the price and herbarium assistants with whom he has collabo- rated. 82 1993] Book Reviews 83 With the Atlas of Ontario Mosses, written with Linda M. Ley, Ireland adds to the understanding of the Canadian flora by ex- tending knowledge of the Ontario moss flora with 17 pages of phytogeographic discussion, two figures showing general phys- iographic and vegetational features, and 490 distribution maps of the known moss flora of that province. Textural topics include general essays on provincial physiography and geology, climate, vegetation, collectors, distribution patterns of Ontario mosses, and a short bibliography. What is new and valuable about this publication is that each species has its own dot map, showing, in Many species, distinctive distributional patterns, and generally, greater biodiversity in the south, reflecting the situation with vas- cular plants. Such patterns are not much discussed, however, other than assignment to the categories ““Widespread, Southern, North- ern, Eastern, and Western,” which are not referred to other phy- togeographic studies. For instance, although the Niagara Escarp- ment is mentioned in the discussion of geology as “‘a northeasterly facing cuesta of Silurian carbonates” extending from Niagara Falls through the Bruce Peninsula and Manitoulin Island, there is no mention of the common and distinctive correlation of this im- portant Ontario feature with the distributions of calciphilic spe- cies. Perhaps Ireland plans more detailed discussion of geographic patterns of Ontario mosses in a future work. The species maps include river systems as details but no po- litical, physiographic or vegetational subunits. There are no over- lays provided; one can try to mentally overlay the political map included with the 1975 Ontario checklist or leaf back and forth to the physiographic and vegetational maps in Figures | and 2. For some readers, however, this lacks exactitude, which is es- pecially unfortunate in that considerable collecting since the 1975 checklist is reflected in the Atlas, and additions to counties and other subprovincial political areas are very difficult to locate and assign. The maps are of little value to local workers interested in local floras. Thus, the Atlas cannot be considered an “extension” to the 1975 checklist in the sense of being an update, but is something else entirely. The maps have value in showing the relative success of partic- ular species in the various temperate and boreal zones of Ontario, and will be valuable in helping to determine range limits for species in the Bryophyte volume of the ongoing Flora North America project. Thus, although the major criticism of this work 84 Rhodora [Vol. 95 is that Ireland and Ley missed an opportunity to interpret the actually quite interesting distributional patterns of the Ontario moss flora, this work is valuable in its contribution to the un- derstanding of moss distributional patterns, which are often sur- prisingly wide and curiously different in many ways from those of vascular plants at the same taxonomic level. We look forward to additional contributions from Dr. Ireland and the Bryology Section at the Canadian Museum of Nature. LITERATURE CITED IRELAND, R.R. 1982. Moss Flora of the Maritime Provinces. National Museums of Canada, Publications in Botany 13, Ottawa, Ontario. RASSARD, W. B. SCHOFIELD AND D. H. Vitt. 1980. Checkin of i Mosses of Canada. National Museums of Canada, Publi- cations in mera 8, Ottawa, Ontario ARD, W.R R Sctormip AND D. H. Vitt. 1987. Checklist of the mosses we Canada II. Lindbergia 13: 1-62. AND R. F. Cain. 1975. Checklist of the Mosses of — National Museums of Canada, Publications in Botany 5, Ottawa, Ont Lawton, E. 1971. Moss Flora of the Pacific Northwest. Hattori cea Lab- oratory, Nichinan, Japan. P. M. ECKEL CLINTON HERBARIUM BUFFALO MUSEUM OF SCIENCE BUFFALO, NY 14214 Vol. 94, No. 880, including pages 327-409, was issued December 16, 1992. INFORMATION FOR CONTRIBUTORS TO RHODORA Submission of a manuscript implies it is not being considered for publication simultaneously elsewhere, either in whole or in part. Manuscripts should be submitted in triplicate (an original and two xerographic copies) and must be double-spaced (at least 4”) throughout including tables, figure legends, and literature cita- tions. The list of legends for figures and maps should be provided on a separate page. Footnotes should be used sparingly. Do not indicate the style of type through the use of capitals or under- scoring, particularly in the citation of specimens. Names of genera and species may be underlined to indicate italics in discussions. Specimen citations should be selected critically, especially for common species of broad distribution. Systematic revisions and similar papers should be prepared in the format of ““A Monograph of the Genus Malvastrum,” S. R. Hill, Rhodora 84:1-83, 159- 264, 317-409, 1982, particularly with reference to indentation of keys and synonyms. Designation of a new taxon should carry a Latin diagnosis (rather than a full Latin description), which sets forth succinctly just how the new taxon is distinguished from its congeners. Papers of a floristic nature should follow, as far as possible, the format of “Annotated List of the Ferns and Fern Allies of Arkansas,” W. Carl Taylor and Delzie Demaree, Rho- dora 81: 503-548, 1979. For bibliographic citations, refer to the Botanico-Periodicum-Huntianum (B-P-H, 1968) which provides standardized abbreviations for journals originating before 1966. All abbreviations in the text should be spade e by a period, except those for standard units of measure and directi For standard abbreviations and for guidance in ‘other matters of biological writing style, consult the CBE Style Manual, 5th ed. (original title: Style Manual for Biological Journals). In preparing figures (maps, charts, drawings, photos, etc.) please remember that the printed plate will be 4 x 6 inches; be sure that illustrations are proportioned to reduce correctly, and indicate by blue pencil the intended limits of the figures. (Some “turn-page” figures with brief legends will be 32 x 6 in.) Magnification/reduction values given in text or figure legends should be calculated to reflect the actual printed size. An Abstract and a list of Key Words should be supplied at the beginning of each paper submitted, except for a very short article or note. All pages should be numbered in the upper right-hand corner. Brevity is urged for all submissions. RHODORA January 1993 Vol. 95, No. 881 CONTENTS Schiedeella romeroana (Orchidaceae, Spiranthinae), a new and interesting species from Mexico Paaries 1 he ee ee sys hate ta oe oe 1 Sarracenia purpurea L. ssp. venosa (Raf.) Wherry var. apie Schnell (Sar- raceniaceae)—a new variety of the Gulf coastal pla I GC a see eae eRe ee wees 6 Rhododendron albiflorum Hook. (Ericaceae): one taxon or two? Meredith A. Lane, Zhongren Wang, Christopher H. Haufler, Philip A. Thompson, Youngjune Chang, Candice L. Bregin, Michael T. Camp- bell, C. Luke Gard, Gary Schott, Greg Spielberg, Kathryn E. Stoner, eee BT Oe Fs FE oe ne ec oe eee 11 Snow algae in northern Engla I a a as Fes oe oe ceed Boo 21 A taxonomic study of Aristida stricta and A. beyrichiana ee a aia eee 25 A revision of the Chamaesyce deltoidea (Euphorbi ) complex of south- Alan Herndon 38 = Pega of botanists and Potentilla robbinsiane discovery, system- cs, collection, and stewardship of a rare " Charles V. Cogbill 52 Salix Saxij Pn ae a, NR a ee toc NeW to Nova Scotia Rent J. Bella and WE Schild s,s 76 BOOK REVIEW Women Botanists of Ohio: Born before 1900 Mary Mt. Welt a oe oe as 79 Atlas of PORES: Mosses Hovdora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB 7 ~ Uy) et: i wipe) Y \ Kp) ¥ sax) SSS f off SAS < a a Wa ~¥ Bye ey Uy, iw BOP Cr ay, » ip cg Ee) Ke p/ eS A on SOR | i EC wt 2 Y i: Rp 074 hs ; { Vol. 95 April 1993 No. 882 CC IEUEEUEESSSS'SS;' a a eae ih Che New England Botanical Club, Inc. 22 Divinity Avenue, Cambridge, Massachusetts 02138 RHOBORA GORDON P. DEWOLF, JR., Editor-in-Chief Associate Editors DAVID S. BARRINGTON RICHARD A. FRALICK A. LINN BOGLE GERALD J. GASTONY DAVID E. BOUFFORD C. BARRE HELLQUIST CHRISTOPHER S. CAMPBELL MICHAEL W. LEFOR WILLIAM D. COUNTRYMAN ROBERT T. WILCE GARRETT E. CROW RHODORA (ISSN 0035- ie Published four times a year (January, April, ey and October) by The New England Botanical Club, 810 East 10th St., Lawrence, KS 66044 and printed by Allen Press, Inc., 1041 New Hampshire St., Lawrence, KS 66044. Second class postage pending at Lawrence, KS. POSTMASTER: Send address changes to RHODORA, P.O. Box 1897, Lawrence, KS 66044. RHODORA is a journal of botany devoted primarily to the flora of North America. Scientific papers and notes relating to this area and floristically saa areas, and articles concerned with systematic bot- any and cytotaxonomy in their broader implications, will be consid- ered. Aiticles are agave to peer review. RHODORA assesses page charges. SUBSCRIPTIONS: $40.00 per calendar year, net, postpaid, in funds payable at par in the United States currency. Remittances payable to RHODORA. Send to RHODORA, P.O. Box 1897, Lawrence, KS 66044-8897. irises: Regular $35; Family $45. Application form printed ear BACK VOLUMES AND SINGLE COPIES: Some available; infor- mation and prices will be furnished upon request to the Treasurer. ADDRESS CHANGES: In order to receive the next number of RHO- DORA, changes must be received by - apace office prior to the first day of January, April, July, or Oct INFORMATION FOR cee Inside back cover, January and April. MANUSCRIPTS: Send to: Gordon P. DeWolf, Jr. Managing Editor, RHODORA 125 Long Hill Road West Brookfield, MA 01585 Cover Illustration Amelanchier nantucketensis Bickn., the Nantucket Shadbush, is found only on Ty moors, pine-barrens and pond margins on Nantucket and Martha’s Vineyard Islands. A surculose, ae colonial shrub to 2 m, this member of the — y Alison Dib This paper meets the requirements of ANSI/NISO 239.48-1992 (Permanence of paper). Tbodora URNAL OF T NEW ENGLAND BOTANICAL CLUB Vol. 95 April 1993 No. 882 RHODORA, Vol. 95, No. 882, pp. 85-96, 1993 THE HYBRID ORIGIN OF ELEOCHARIS MACOUNII PAUL M. CATLING AND STUART G. HAY ABSTRACT For nearly a centu d Eleocharis macounii Fernald has been known with certainty ont from the type station i in ee It has been treated as a synonym E. obtusa (Willd.) Schultes, E. ovata (Roth) R. & S. or E. palustris (L.) R. & S., or as a hybrid. Various quantitative and qualitative morphological features suggest that it is a hybrid involving E. sinuililainese 6 series Palustriformes and E. obtusa in series Ovatae. Eleocharis x macouni by its rhizome internodes less than 2 mm a presence of many culms of vatying Leeigilis arising from a single tuft and pale greenish-brown achenes with prominent keels and a rough cellular surface. This natural hybrid is of particular interest since it is the first reported in the genus Eleocharis in North America, and it involves different species groups within the genus. Key Words: Eleocharis x macounii, Cyperaceae, hybrid, Québec, North America INTRODUCTION The genus Eleocharis (Cyperaceae), commonly known as the spikerushes, is receiving increasing attention in North America because of the potential use of several species in aquatic weed management in irrigation systems and in pollution abatement (e.g., Sutton, 1984). The classification and identification of many of the species remain problematic. One of the problems is the status of Eleocharis macounii Fernald, a plant described in 1899 from specimens collected by J. Macoun at North Wakefield, Qué- bec Marie-Victorin (1935) treated Eleocharis x macounii as a spe- cies, noting that it was known only from the Gatineau River. Svenson (1937) also treated it as a species but suggested that it 85 MISSOUR! BOTANICAL DEC 2 1993 meahrnral 1 @eRAT. 86 Rhodora [Vol. 95 was problematic, close to E. intermedia Schultes, and also had characteristics of the European E. multicaulis (Sm.) Desv. It was treated as a species in the 7th and 8th editions of Gray’s Manual (Robinson and Fernald, 1908; Fernald, 1950). Fernald noted that E. macounii was similar to the North American E. intermedia Schultes and the European E. carniolica Koch. Later Svenson (1953) noted that E. macounii was best treated as ‘“‘an estuarine modification of E. obtusa (Willd.) Schultes, and he treated it as a synonym of E. obtusa in his (1957) monograph suggesting that it was “‘a variant with elongated style base somewhere between E. obtusa and E. ovata (Roth) R. & S., in which some of the achenes are slightly trigonous.”’ Svenson (in Kartesz and Kartesz, 1980) placed E. macounii in synonymy with E. obtusa var. obtusa. However, Hines (1975) in his monograph of the E. ovata complex (which includes E. obtusa) did not mention E. macounii under either E. obtusa or E. ovata. Gleason and Cronquist (1965) placed Eleocharis macounii in synonymy with E. intermedia, and it is treated in the same way in the United States national list of scientific plant names (Anon., 1982). With respect to Canadian literature, Eleocharis macounil was treated as a questionable synonym of E. intermedia by Scoggan (1978), omitted by Boivin (1967), and is not mentioned in the listing of the rare plants of Québec (Bouchard et al., 1983). In his recently published Cyperaceae of eastern Canada, Boivin (1992) placed E. macounii in synonymy with E. ovata sensu lato (i.e., including E. obtusa). The only annotation on the type at Gu is that of Boivin (dated 1979) which indicates his view that the material is referable to E. palustris (L.) R. & S. sensu lato, but with the rhizomes, a distinctive feature of the latter species, broken off. On the isotype sheet at CAN, S. G. Hay noted in 1988 that E. macounii is probably a partly sterile hybrid between E. intermedia and E. obtusa vat. jejuna Fernald. Hay also uncovered another specimen of FE. ma- counii, misidentified as E. intermedia, collected from Cascades (Malte 7.8.1913, can), a locality situated close to the type station. In summation Eleocharis macounii has been regarded as a spe- cies, as part of the variation within one of four other taxa, or as a hybrid. The purpose of the present paper is to examine critically each of these possibilities, and particularly to consider the hy- pothesis that E. macounii is of hybrid origin. 1993] Catling and Hay— Eleocharis 87 METHODS The percentage of fruit development was determined on 15 mature spikes from the type material of Eleocharis macounii at GH and CAN. Pollen from five different spikes, also from the type material, was stained using the phloxine and fast green technique (Owczarzak, 1952) to assess fertility. In order to determine whether E. macounii was intermediate in quantitative characters, 10 characters (Table 1) were measured in thirty specimens of each of E. intermedia, E. obtusa and E. ovata. These were selected to represent variation over a broad area of eastern Canada. For E. macounii two culms from the right hand group of tufts and two from the left hand group on the holotype were measured. Sheath length (mm) was measured from the base of the culm to the top of the sheath. Sheath width was measured across the mouth. Culm width was measured 1 cm below the inflorescence. Culm height is the distance from the tip of the spike to the base of the culm. Scales (also frequently referred to as bracts) measured were in all cases taken from the middle of the inflorescence. The length of the scarious margin of the scale was measured from the apical margin back to where it was no longer translucent. The length of the achene was measured from the base of the basal portion below the bristles to the base of the tubercle. Tubercle width was used to separate E. ovata and E. obtusa since it is more easily determined than stamen number and sep- arates the species readily (Hines, 1975). Here plants with a tu- bercle less than .5 mm wide are treated as E. ovata, while plants with a tubercle more than .51 mm wide are treated as E. obtusa. RESULTS AND DISCUSSION Fernald (1899) noted with his original description of Eleocharis macounii that it resembled the European E. carniolica Koch in habit and inflorescence. He correctly noted, however, that E. macounii has a much broader tubercle (Figure 1B) than E. car- niolica. Furthermore the surface of the achene of E. carniolica is not clearly cellular, but rather obscurely wrinkled longitudinally and the tubercle of E. carniolica has a much more distinct neck. The European Eleocharis multicaulis (Sm.) Desv. is similar to E. macounii in some respects, but is a much more robust plant [Vol. 95 Rhodora 88 Ir'0 tro 80 £90 PLO s'0 L¢°0 pr'0 aml) 070 770 810 WIpiM sposeqn | IPO ctr'0 Or'0 £¢'0 tr'0 970 1¢'0 80 970 sO bro 970 iystay spou9qn | ram) 9L°0 89°0 6L°0 93°0 0L'0 89°0 9L°0 09°0 ¢9°0 040 09°0 YIPIM dusty sO'l Ol 00°! 66°0 r0'l 06'0 98°0 960 080 £0'l 4 ei 860 yisua] suayoy L10 810 910 070 $70 710 810 ¢7'0 0) te) L0'0 rant) 700 yipim urasew [BoIde SNOLIBDS a]BIS LS'0 89°0 8r'0 19°0 cL°0 9r'0 SO 0L°0 Or'0 t70 Or'0 910 xode wosj WU 7°() YIP a[Bog PIT 871 80'1 SIT Or'l 00° £01 0c'T 060 60'1 871 660 YIPIM afBog Oe 7 9CT 9TT 961 O£'7 Pll 68'1 OTT OL'T 161 OV? OL’! yiduay ajeog cc 087 OT og 00°F 067 LG’ 00°F 09°T OT O87 06'1 YIpim ayxids SsTOl OS'II Os’8 oC L Os'Tl OS'b 18s 00°6 00°F sos 00°L 06°£ yisua] ayidg SwT6tI OO89T O00S6 O007Z ODSSh ODTEI Le6st 00987 O06F OOOZTI ODL8I O0'0S 1ysiay WIND 8¢°0 vL'0 os'0 LL0 {on 9¢°0 9¢°0 73°0 9¢°0 re'0 OF'0 o£'0 yipim wn) €L°0 838°0 690 Ol 9¢'l 8240 30 orl 8S°0 £¢°0 810 tr'0 yipim yieeys SLiilc 00°LT OO'8I eS PE 00°09 0017 le 4 OO'LE OO'rI O88! 00°07 00°L yisuay yieays ues “xe “Ul ueoW “xe “UIW uBoW “xeW uI-W ubOW xXEW “UIC: Jajoviey) munoospu x “F psniqo “q DIDAO “FZ DIpauLsaqul “7 “WW UT OJP S]USUIOINS -BaW [JY JeuUseW a[qQei{Ieae [je UO paseq UONdUDsap SuIAUedWIODOR Ose 398 ING ‘Pp = u MuNnOIDUx “F104 “YE =u ‘DIDAO “J pure Dsnigo “F ‘pIpautiaqul “J 104 “WXe) SUDYIOI]Y URIUIUTY YON WIISedYLIOU JNO} UI SJo1OBIBYD p] JOJ SONjeA URIW PUR WINWIXeW ‘WNUTUTYy “T AQEL 1993} Catling and Hay— Eleocharis 89 with much longer scales and broader spikes. It also has a slightly longitudinally wrinkled achene surface unlike E. macounii. Thus E. macounii is not a European introduction. Boivin suggested on his annotation label that the type material of Eleocharis macounii is referable to E. palustris with rhizomes broken off. This suggestion has to be rejected for several reasons: the cellular achene surface (Figure 1B), the presence of culms of varying lengths from very short to very long (Figure 2A), the fact of many culms arising from the same point on an abbreviated rhizome (Figure 2A), the elliptic scales (Figure 2C), the relatively broad tubercle (Figure 1B) and the relatively pale greenish-brown mature fruit are all characters lacking in E. palustris sensu lato (Figures 1D, F, 2E, H), including E. erythropoda Steudel (E. calva, and E. smallii Britton). The type material of E. macounii does not represent a freak, because the plants do not show any abnormal growth or infection. Only 10-30% of the flowers on spikes of E. macounii have well developed fruit, the remainder being aborted. Many Eleocharis species including E. intermedia, E. ovata, and E. obtusa invariably have 95-100% of the flowers ripening into achenes, in the absence of obvious fungus infection or insect attack. The high level of fruit abortion in E. macounii suggests that it may be a hybrid. The hybrid concept is further supported by pollen infertility since only 10% of the pollen from one spike and 40% of the pollen from another spike, both from the holotype of E. macounii, took up the stain. In contrast more than 90% of pollen grains from E. intermedia, E. ovata, and E. obtusa stained well. Even in the E. macounii pollen that did stain, the stain was often pale or confined to a small part of the grain. The cellular, three-sided and greenish surface of the achenes on the holotype plants of Eleocharis macounii (Figure 1B) is sug- gestive of E. intermedia (Figure 1A), although these characters are less pronounced than in the latter species. If E. intermedia is one parent, as evidenced by the cellular achene surface, then the other parent may be expected to possess certain features expressed in E. macounii, but not in E. intermedia. Among these characters that are to be expected is a broad tubercle (Figure 1B), a keeled and lenticular achene, a relatively long spike, and blunt, broad- tipped scales that have. a relatively broad hyaline margin apically. There are t n: E. obtusa (Figures IC, 2D) and E. ovata (Figures 1E, 2G). Eleocharis erythropoda 90 Rhodora [Vol. 95 i " ie ai, reg if cht Sia ~ 4 ak! ¥ it ra?) ay Gn = 4 ee ee oy PP he a ay or eye 2 Cg ate a AS “ : = ‘4 pa ae ¢ 1 irr! J He. 1993] Catling and Hay— Eleocharis 91 (previously widely known as E. calva) and the E. smallii-palustris complex are eliminated as a source of putative parents because a hybrid of either of them with E. intermedia would be expected to have more pointed and broad-based scales and an achene with an elongate and fleshy tubercle and lacking prominently keeled edges, and rhizomes developed to a greater or lesser extent. E/e- ocharis macounii on the other hand has broadly rounded scales (Figure 2C), achenes with a relatively short tubercle and two or three prominently keeled edges (Figure 1B), and lacks an elongated rhizome. Eleocharis obtusa and E. ovata, however, both possess many of the characters to be expected in the other parent, notably the prominent keels on the achene, especially just below the tu- bercle. Both E. obtusa and E. ovata possessed values for various quantitative characters, such as tubercle width (Figure 1B) that were similar to E. macounii or intermediate with respect to E. intermedia. With regard to these two species, E. ovata has rela- tively short spikes and scales, and smaller achenes (Table 1), whereas relatively longer spikes and scales and larger achenes are predicted in the other parent if E. macounii were to be inter- mediate in these characters (Table 1). Since intermediacy is likely in many features of hybrids, E. obtusa is a more likely parent. It seems undesirable to recognize the var. jejuna of the latter species (e.g., Hines, 1975). The local rarity of E. ovata (Di Labio and Brunton, 1989; see also M. O. Malte, 4 Sept. 1922, sub E. inter- media, CAN) and the fact that it has not been found at Wakefield, where E. obtusa is common, also suggests that the latter is more likely to be the other parent. Eleocharis x macounii appears to be the first hybrid reported in the genus Eleocharis in North America, and it should at least raise the question of whether hybridization is more widespread in the genus than currently recognized, and associated with some of the taxonomic difficulties. Natural hybrids in Eleocharis have been reported from Japan (e.g., Koyama, 1961), and hybrids in Eleocharis subseries Palus- tres in Europe have been well documented (Strandhede, 1965) Figure 1. Representative fruits of A agngt seadinna (Catling sige DAO). B E. x macounii (Macoun 7552, Gu). C E. obtusa (Cody 6579, Dao). D palustris (Gillet 11765, pao). E E. ovata (Dore bese DAO). F E. peer hos (Rolland-Germain 10284, pao). Drawings by Marcel Jomphe. 92 Rhodora [Vol. 95 Wesel frag, ss Figure 2. A-C Eleocharis x macounii (Macoun 7552, Gu). A habit. B spikelet. Cc lower inflorescence scale. D-H lower inflorescence scales of various Eleocharis species. D E. obtusa (Cody 6759, pao). E E. erythropoda (Rolland-Germain 10284, DAO). F E. intermedia (Catling 7/09/87, pao). G E. ovata (Dore 24360, Dao). H “ ; sineng (sub E. palustris var. major, Gillet 11765, pao). Drawings by Marcel omphe. 1993] Catling and Hay—Eleocharis 93 with cytological studies. Eleocharis mamillata subsp. mamillata x E. palustris subsp. palustris has greatly reduced fertility. It is to be expected that a hybrid between quite different species in different series (as treated by Svenson, 1957), like E. x macounii, would also have reduced fertility. The chromosome number of E. obtusa is 2n=10, whereas E. intermedia has been reported with 2n=22 (Schuyler, 1977), so that if living plants of E. x macounii are discovered, it should be possible to support the putative par- entage with both chromosome number and meiotic anomalies. Unfortunately repeated attempts by the authors and other bota- nists to locate living material of E. x macounii at the type station and elsewhere have been unsuccessful. The following key and description, based on the type material and one other collection, will facilitate the identification of E. x macounii. KEY TO THE ELEOCHARIS SPECIES OF SOUTHERN ONTARIO AND QUEBEC WITH DIFFERENTIATED TUBERCLES, FIRM SHEATHS AND ACHENES LACKING LONGITUDINAL RIDGES 1. Rhizome internodes more than 2 mm long, tufts with most Cline OF Gas Ee a Le 2 2. Achenes with a roughened cellular or reticulate surface, et eee eae . perennials of series Pa/ustriformes subseries Trunca- tae including E. compressa, E. elliptica and E. nitida 2. Achenes essentially smooth and shiny, at most only ob- scurely cellular or reticulate, 2-sided ................ .. perennials of series Palustriformes subseries Palustris including E. erythropoda and E. smallii 1. Rhizome internodes ae than 2 mm long, tufts with culms of many different ages and heights ..................... 3 3. Achenes with a roughened cellular or reticulate surface 4 4. Longer spikelets 5-7 mm long; achenes becoming dark green, without prominent keels, subterete to equilat- —, triangular in cross section; tubercles 0.18-0.22 WIE occ Seles Pe ob ews sao E. intermedia 4. se spikelets 8-11 mm long; achenes pale greenish- brown, with 2 or 3 keels very prominent below tu- 94 Rhodora [Vol. 95 bercle, more or less lenticular in cross section; tuber- cles (0.2)0.3-0.5 mm wide ......... E. x macounlii 3. Achenes with a smooth surface .......26...006 254555 5 5. Tubercle less than 0.5 mm wide ........... E. ovata 5. Tubercle 0.51 mm or more wide .......... E. obtusa Eleocharis < macounii Fernald (pro sp.) (E. intermedia Schultes x E. obtusa (Willd.) Schultes), Proc. Amer. Acad. 34: 497, Figure 26a. 1899. TYPE: Québec, sneha Wakefield, 13 Sept. 1893, J. Macoun 75.5.2 (HOLOTYPE— GH! ISOTYPE—CAN)). Annual herb; rhizome internodes less than 2 mm long, culms red at the base, green above, erect to divergent or bending down- ward, numerous, 20-230 mm long and of greatly varying lengths within a tuft, 0.3-0.8 mm wide 1 cm below the inflorescence; uppermost sheaths 5-35 mm long, 0.3-0.9 mm wide at the acute apex. Spikelets lanceolate, acute, densely 5—52-flowered, 3-11 mm long, 1.5—3 mm wide; scales reddish-brown with green mid- rib, ovate, obtuse, those in the middle of the spikelet 1.8-2.5 mm long, 1-1.3 mm wide, with a scarious margin 0.1-0.2 mm wide at the apex. Stamens 3, filaments 0.9-2 mm long, anthers 0.2- 0.4 mm long, not reaching to the top of and concealed by the subtending scale, pollen largely infertile. Styles mostly trifid, oc- casionally bifid, the branches 0.5—0.9 mm long, extending beyond the tip of the subtending scale. Achenes mostly aborted, those maturing pale greenish-brown when fully mature, obovoid, 0.9- 1.1 mm long (excluding tubercle), 0.6-0.8 mm wide, with mostly 3 (sometimes 2) keels becoming very prominent toward the apex, the surface rough-cellular, i.e., with circular or rectangular de- pressions with elevated walls; tubercle deltoid 0.3-0.4 mm high, 0.2-0.5 mm wide; bristles (4)5-6(7), brown, retrosely barbed, mostly 1.2—2.1 mm long, much exceeding the achene and tubercle. ; = na material examined: Québec: Cascades, 7-8-1913, M. O. Malte s.n. CAN). Distribution: Collected only from Wakefield and Cascades, both localities along the Gatineau River of Québec. Comments: Eleocharis x macounii lacks elongate rhizomes and has the tufted or caespitose growth habit of E. obtusa and E. intermedia, as well as flowering/fruiting culms of many lengths as in these species. This latter feature, as well as the more or less 1993] Catling and Hay— Eleocharis 95 diverging culms, readily distinguishes it from E. erythropoda, which it otherwise resembles. There is reason to suspect that Eleocharis x macounii may have culms longer than 230 mm and more than | mm wide, with larger spikelets than noted above, based on broken culms on tufts oth the holotype and isotype sheets, but these tufts cannot be identified with certainty. The material collected by Malte more closely resembles E. intermedia by its shorter culms and achenes with narrower tubercles. ACKNOWLEDGMENTS Dr. J. Cayouette kindly provided information, and both he and Dr. E. Small made useful comments as reviewers. Mr. M. Jomphe prepared the illustrations. LITERATURE CITED ANonyMous. 1982. National List of Scientific Plant Names, Vols. | & 2. U.S.D.A., Washington, D.C. Borvin, B. 1967. Enumération Pag apes du Canada. V—Monopsides (lére partie). Naturaliste Canad. 94: -157. . 1992. Les Cypéracées de A du Canada. Provancheria 25: 1-230. BoOuCcHARD, A., D. BARABE, M. Dumais and S. Hay. 1983. The rare vascular sissies of Québec. National Museum of Natural Sciences, Syllogeus 48. Di Lasio, B. M. and D. F. BRUNTON. 1989. The oval spike-rush (Eleocharis ovata) new to the Ottawa district and Pontiac County, Quebec. Trail and Landscape 23: 15-17. FERNALD, M. L. 1899. Eleocharis ovata and its North American allies. Proc. Amer. Acad. Arts. 34: 483-497. . 1950. Gray’s Manual of Botany, 8th ed. American Book Co., New York. GLeason, H. A. and A. Cronquist. 1965. Manual of Vascular Plants of North- eastern United States and Adjacent Canada. D. van Nostrand Co., Princeton, aia D. M. 1975. A graph of the Eleocharis lex (Cyperaceae) n North America. Ph.D. thesis, University of Michigan. ae J. T. and R. Kartesz. 1980. A synonymized checklist of the vascular flora of the United States Canada, and Greenland. University of North Carolina Press, Cha Koyama, T. 1961. ota of the family Cyperaceae. Part II. A revision of the species of the Cyperaceae in Japan and its neighbouring regions. J. Fac. Sci. Univ. Tokyo, Sect. 3, Bot. 8(3): 81-148. Marie-Victorin, F. 1935. Flore Laurentienne. Les Presses de l'Université de Montréal. 917 pp. 96 Rhodora [Vol. 95 Owczarzak, A. 1952. A rapid method for mounting pollen grains. Stain Tech- nology 27: 249-251. Rosrnson, B. L. and M. L. FERNALD. ibe Gray’s Manual of Botany, 7th ed. American Book 5 New York. 926 p SCHUYLER, A. E. 1977. tern North Amer- ican Eleocharis Mcaheee Brittonia 29: 129-133. ScoGGAN, H 1978. The Flora of Canada, part 2. National Museum of Natural Sloe: Poblicaicns: in Botany 7: 93-545. STRANDHEDE, S.-O. 1965. Chromosome studies in Eleocharis, subser. Palustres. III. Observations on western European Taxa. Opera Bot. 9(2): 1-86. Sutton, D. L. 1984. Spikerushes. Aquatics 6: 4, 9, 10. Svenson, H. K. 1937. Monographic studies in the genus Eleocharis. 1V. Rhodora 39: 210-231, 236-273. ——. 1953. The Eleocharis obtusa-ovata complex. Rhodora 55: 1-6. ——. 1957. Poales-Cyperaceae-Scirpeae. N. Amer. Flora 18(9): 505-556. AGRICULTURE CANADA BIOSYSTEMATICS RESEARCH DIVISION CENTRE FOR LAND AND BIOLOGICAL RESOURCES RESEARCH WM. SAUNDERS BUILDING, CENTRAL EXPERIMENTAL FARM OTTAWA, ONT., CANADA KIA 0C6 INSTITUT DE RECHERCHE EN BIOLOGIE VEGETALE UNIVERSITE DE MONTREAL 4101 EST RUE SHERBROOKE MONTREAL, QUEBEC, CANADA H1X 2B2 RHODORA, Vol. 95, No. 882, pp. 97-112, 1993 THE MARINE AND BRACKISH WATER SPECIES OF VAUCHERIA legen area CHRYSOPHYTA) OM CONNECTICUT CRAIG W. SCHNEIDER, LAURA A. MACDONALD, JAMES F, CAHILL, JR.,! AND SARAH W. HEMINWAY ABSTRACT A more than decade-long field investigation of the genus Vaucheria in Con- necticut salt marshes and estuarine habitats has revealed the presence of nine species, including five species and a variety not previously known for the area: V. arcassonensis, a compacta bans koksoakensis, hg: coronata, x intermedia, KF. minuta, and V. It of gametangia, prorat for a games of mostly vegetative field-collected sam- ples in their native muds. All marine and brackish water species of Vaucheria previously reported from other aa of New England are now known from Long Island Sound. Key Words: Vaucheria, Vaucheriaceae, Tribophyceae, Long Island Sound, Con- necticut INTRODUCTION In his comprehensive flora of the green algae of North America, F. S. Collins (1909) included the then poorly known family Vau- cheriaceae, listing just three marine species of Vaucheria from New England. Since that time, a number of studies has greatly increased our understanding of the family, culminating in the publication of a seminal monograph on the North American Vaucheriaceae by J. L. Blum (1972). This work prompted regional marine studies of Vaucheria in localities on the continent where the genus was poorly known, including Ott and Hommersand (1974) in North Carolina, Bird et al. (1976) in Nova Scotia, Pecora (1976, 1977, 1978, 1980) in Louisiana and Texas, Gallagher and Humm (1981) on the Florida Gulf coast and Garbary and Fitch (1984) in British Columbia and Washington. However, areas re- main in North America, including much of New England, where the genus has not been investigated systematically. When com- piling the list of seaweeds for Connecticut (Schneider et al., 1979), the first author recognized the paucity of species records in the ' Present address: Ecology Group, Department of Biology, University of Penn- sylvania, Philadelphia, PA 19104. 97 98 Rhodora [Vol. 95 —+ 10 km Connecticut Connecticut R. Thames R. Hudson R. Housatonic R. \ if Atlantic Ocean i 73 00 W 72° 00 W | Figure 1. Map of the study area. For specific collecting site codes, see Table 1. state relative to other areas in New England, in particular Mas- sachusetts, and initiated the study reported here. At the time, only four marine species of Vaucheria had been reported from Con- necticut: V. compacta (Collins) Collins ex W. R. Taylor (as V. piloboloides), V. litorea C. Agardh, V. velutina C. Agardh (as V. thuretii, Collins, 1900), and V. vipera Blum (Collins, 1900; Blum, 1972). In part, the relatively small number of species known at that time for Connecticut, a state with a 160 km coastline divided by dozens of rivers and streams entering Long Island Sound, reflected the few collections made in the area. Furthermore, most collec- tions were taken in the drier summer season, not during the au- tumn-winter-spring period of optimal growth and reproductive conditions for most Vaucheria species (Blum and Conover, 1953; Blum, 1966). Although Vaucheria is a common constituent of salt marsh and estuarine muds, it is most often found in the vegetative state, thereby proving impossible to distinguish at the species level. Previous workers (e.g., Ott and Hommersand, 1974; Pecora, 1976) found that suitable laboratory culture conditions would first allow luxuriant vegetative growth from the mud plat- form and then gametangial production, giving a more complete floristic assessment of Vaucheria species within a habitat. 1993] Schneider et al.— Vaucheria 99 Table 1. Long Island Sound collection sites. Fairfield Co., Greenwich Point Park salt marsh, Greenwich (GPP) Fairfield Co., Cove Island Park salt marsh, Stamford (CIP) Fairfield Co., mouth of Silvermine River, east bank, East Norwalk (SLV) Fairfield Co., Sherwood Island State Park salt marsh, Westport (SHI) Fairfield Co., Pride’s Point, tidal channel vic. Housatonic River, Stratford PPS) MOA Pp New Haven Co., Milford Point salt marsh, Milford Harbor, Milford (MLH) New Haven Co., Morris Creek salt marsh, Lighthouse Point, East Haven (MOR) New Haven Co., Mansfield Point salt marsh, East Haven (MNS) New Haven Co., Indian Neck Inlet salt marsh, Branford (INI) New Haven Co., Island Bay salt marsh, Leetes Island, Guilford (LTS) New Haven Co., Joshua Cove salt marsh, Sachem Head, Guilford (JCV) New Haven Co., Chaffinch salt marsh, Guilford Harbor Rutt ) New Haven Co., Seaview Beach salt marsh, Madison (SV New Haven Co., Hammonasett State Park salt marsh, janie (HSP) Middlesex Co., Meig’s Point salt marsh, mouth of Hammonasett River, Clinton (HRC) Middlesex Co., mouth of Indian River, east bank, Clinton (IRC) Middlesex Co., Clinton Beach salt marsh, Clinton (CBC) Middlesex Co., Menunketesuck River, river banks, Grove Beach (GRV) New London Co., Saybrook Point salt marsh, Old Saybrook (OSP) New London Co., Blackhall River, river banks, Old Lyme (BOL) New London Co., Rocky Neck State Park salt marsh, East io (RNS) New London oo , Pleasure. Beach salt marsh, Waterfor d (P New London C ies (HMW) OZZrACHm at S< CH“ ROW MATERIALS AND METHODS Since the initiation of this study in 1980, we have visited more than 20 salt marsh and estuarine habitats along the Connecticut coast of Long Island Sound from Greenwich to New London (Figure 1), often revisiting the same sites in different seasons as well as years. The distance to collecting sites from Hartford ac- counted for the unevenness in visits to different sites. For ex- ample, we have taken nearly a hundred samples from over 20 visits to Harkness Memorial State Park, the most frequently vis- ited salt marsh, but only visited the most distant locality, Green- wich Point Park, once. Few collections were g summer, a period when typically high nacnees high light intensities and lower rates of precipitation do not foster obvious growth o Vaucheria in Long Island Sound salt marshes. Codes identifying the collecting sites are given in Table 1. At each collecting site in close proximity to a tidal channel or stand- 100 Rhodora [Vol. 95 ing water, quadrats of mud 42.25 cm? and 2-3 cm thick (size compatible with culture vessels) with visible Vaucheria siphonous filaments or mats, were cut with a serrated knife and removed with a broad spatula. Mud samples without obvious Vaucheria were taken only when siphons could not be located within a given marsh. Vascular plants within the quadrat were cut off at mud level. Two to six mud samples were taken from different areas within each salt marsh or estuarine bank and transported back to the laboratory on ice. In the laboratory, each sample was observed under the micro- scope for gametangia. Then, each sample was subdivided into quarters, placed into deep culture vessels and filled to one-half of the mud sample thickness with culture media. During the early years of the study, the divided crude cultures were placed in enriched seawater Grund medium (von Stosch, 1969) and then incubated in Hotpack No. 352642 growth chambers with set pho- toperiods of 8L:16D, 12L:12D or 16L:8D at 5° or 15°C temper- atures. Because all of the collected species produced gametangia at least during the longest daylength and warmer temperature, beginning in 1987 we standardized culture conditions with a pho- toperiod 16L:8D, light intensity ca. 125 wmol/m2/sec., and tem- peratures of 15°C during the light cycle and 13°C in the dark cycle. We also varied the salinity in the crude cultures (0.3, 2.0, 4.0, 6.5, 16.0, 30.0 ppt) using modified ASP-V media (Entwisle, 1988). Additionally, mud cores of some of these cultures were subcul- tured into the same salinity media with 1.75% agar added. Agar cores were removed with a 0.5 cm cork borer and replaced with the same sized mud-Vaucheria core. The Petri dishes were sealed with Parafilm®. Microscopical observations on each crude and agar culture were made every three to four days. Drawings were made with the aid of a Zeiss camera lucida, and vouchers of each of the species are deposited in Herbarium C. W. Schneider at Trinity College, Hartford. RESULTS AND DISCUSSION All nine species and one variety of Vaucheria presently reported from New England have been found in Connecticut marine an brackish water environments along Long Island Sound, with sev- eral of the species appearing together in the same small quadrat 1993] Schneider et al.— Vaucheria 101 of mud collected from a particular salt marsh or estuarine bank. Many of the mud samples taken without visible Vaucheria were found to produce siphons after incubation in culture media. Either siphons existed within the mud below grade, or dormant zygotes or asexual spores were coaxed into germinating under favorable culture conditions. Few of the crude cultures produced vegetative siphon growth without ultimately developing gametangia, and many cultures produced two to four species over time in a single quadrat. Several collecting sites, in particular those closest to large industrial cities such as Norwalk, Bridgeport, and New Haven, never produced Vaucheria; but on the whole, the genus is readily collected throughout the year and the coastal region of the state. None of the species presented characteristics which differ from Massachusetts populations, so the reader is referred to Blum (1972) or Christensen (1987) for descriptions including pertinent mea- surements. New records for Connecticut are noted with asterisks (*) VAUCHERIACEAE DumonrtTigR 1822, p. 71 *V AUCHERIA ARCASSONENSIS P. J. L. Dangeard 1939, p. 216. Figures 2a, b This species was previously known in New England only from Maine and Massachusetts (Blum and Conover, 1953; Webber, 1968). In Connecticut, it was one of the two most frequently encountered species. To the south, Vaucheria arcassonensis is the most abundant Vaucheria species in North Carolina (Ott and Hommersand, 1974), and despite extensive investigations in the Gulf of Mexico, it has been found only at a single locality in Wakulla Co., Florida (Pecora, 1980). We find V. arcassonensis throughout the state, fruiting during the winter-spring period, very commonly in the same samples with V. coronata and on occasion with V. intermedia, V. minuta, and V. velutina. Similarly, in the British Isles V. arcassonensis is generally mixed with V. inter- media and often associated with V. coronata and V. velutina (Christensen, 1987) TYPE LOCALITY: France DISTRIBUTION: Nova Scotia, New England, North Carolina, Flor- ida Gulf coast, Europe, Australia. 102 Rhodora [Vol. 95 Figure 2. Connecticut marine and brackish water Vaucheria species, all scale bars = 100 um: a-b. V. arcassonensis; a, JCV coll. Sept. 1982; b, HMW coll. Jan. 1992. c-f. V. compacta, OSP coll. Oct. 1991; c-d, antheridia, e-f, oogonia. g—h. V. compacta var. koksaokensis, BOL coll. Feb. 198 1; g, oogonium; h, antheridium. i-j. V. coronata; i, HMW coll. Jan. 1992; j, PBW coll. Apr. 1983. k-n. V. inter- 1993] Schneider et al.— Vaucheria 103 COLLECTIONS: Fairfield Co.—23 Mar. 1981, CIP, 7. Roche, fertile in culture 20 Apr. 1981; 23 Mar. 1981, GPP, 7. Roche, fertile 8 Apr. 1981. New Haven Co.— 23 Sept. 1982, JCV, S. Heminway, fertile 28 Feb. 1983; 3 Apr. 1983, JCV, S. Heminway, fertile 18 Apr. 1983; 1 Mar. 1992, CGH, J. Cahill, fertile in field. New London Co.—30 Nov. 1982, PBW, S. Heminway, fertile 20 Apr. 1983; 29 Jan. 1992, HMW, L. MacDonald, fertile 6 Feb. 1992; 27 Feb. 1992, PBW, J. Cahill, fertile 1 Mar. 1992; 17 Mar. 1992, PBW, L. MacDonald, fertile in field; 6 May 1992, HMW, C. Schneider, fertile 13 May 1992; 6 May 1992, PBW, J. Cahill, fertile 13 May 1992. VAUCHERIA COMPACTA (Collins) Collins ex W. R. Taylor 1937, p. 226. Figures 50-f This species was previously reported in New England from Massachusetts (Blum and Conover, 1953; Webber, 1975) and Connecticut (as Vaucheria piloboloides, Collins 1900). We com- monly found V. compacta in the same mud samples as V. velutina and occasionally with V. intermedia and V. minuta. In our cul- tures, V. compacta was fertile from 0.3-30 ppt salinity. This spe- cies is only known from very low salinity brackish water in Lou- isiana (2.2 ppt; Pecora, 1977), but elsewhere is found in higher salinity environments as well (Blum, 1972; Simons, 1974; Chris- tensen, 1988). TYPE LOCALITY: Massachusetts DISTRIBUTION: Nova Scotia, New England, Louisiana, Europe. COLLECTIONS: New Haven Co.—29 Sept. 1984, CGH, P. Renaud, fertile in culture 1 Mar. 1985. Middlesex Co.—5 Mar. 1987, GRV, B. Bothwell, fertile 4 Apr. 1987; 22 Sept. 1991, GRV, J. Cahill, fertile 1 Oct. 1991. New London Co.—19 Feb, 1981, RNS, T. Roche, fertile 11 June 1981; 18 Aug. 1984, HMW, P. Renaud, fertile 11 Sept. 1984; 29 Sept. 1984, HMW, P. Renaud, fertile 26 Oct. 1984; 28 Oct. 1984, HMW, P. Renaud, fertile 22 Jan. 1984; 15 Oct. 1991, OSP, L. Mac- Donald, fertile 2 Nov. 1991; 12 Sept. 1992, HMW, H. Miller, fertile 4 Nov. 1992. *V AUCHERIA COMPACTA Var. KOKSOAKENSIS Blum et Wilce 1958, 286. Figures 2g, h This variety was differentiated from the typical form of Vauche- ria compacta due to its greatly elongated oogonia (Blum and Wilce, — media; k, GRV coll. Sept. 1991; Lm, HMW coll. Oct. 1980; n, RNS coll. Oct. 1980. 104 Rhodora [Vol. 95 1958). To date, we have discovered this taxon only along the muddy banks at the mouth of the Blackhall River in Old Lyme, a habitat physically if not environmentally similar to that of the type locality at the mouth of the Koksoak River near Ungava Bay. Only one previous collection of V. compacta var. koksoak- ensis has been reported for New England, that from an Ipswich, Massachusetts salt marsh (Webber, 1968), thus our record rep- resents a new southern limit of distribution for the entity. TYPE LOCALITY: Quebec DISTRIBUTION: Eastern Canada, Massachusetts, Connecticut. COLLECTION: New London Co.—18 Feb. 1981, BOL, 7. Roche, fertile in culture 10 June 1981. *V AUCHERIA CORONATA Nordstedt 1879, p. 177. Figures 21, j This species was previously known only from Maine and south- ern Massachusetts in New England, (Blum and Conover, 1953). We have collected this alga with its diagnostic “‘crowned oogonia” from several localities across the state, commonly in the same mud quadrats as Vaucheria arcassonensis, and occasionally with V. intermedia and V. minuta. It has been collected fertile in the field from March—-May. Although it thrives in higher salinities (Christensen, 1988), at its southern-most habitat for North Amer- ica in Louisiana, Pecora (1980) only found V. coronata in 3.0 ppt salinity. TYPE LOCALITY: Sweden DISTRIBUTION: Greenland, Nova Scotia, New England, Virginia, North Carolina, Louisiana, Europe. COLLECTIONS: Fairfield Co.—23 Mar. 1981, CIP, 7. Roche, fertile in culture 20 Apr. 1981. New Haven Co.—1 Mar. 1992, CGH, J. Cahill, fertile 13 Mar. 1992. New London Co.—30 Nov. 1982, PBW, S. Heminway, fertile 20 June 1983; 1 May 1983, HMW, S. Heminway, fertile in field; 29 Jan. 1992, HMW, L. Mac- Donald, fertile 19 Feb. 1991; 27 Feb. 1992, PBW, L. MacDonald, fertile 1 Mar. 1992; 17 Mar. 1992, PBW, C. Schneider, fertile in field. *VAUCHERIA INTERMEDIA Nordstedt 1879, p. 179. Figures 2k—n This alga has previously been collected from New England only in Massachusetts (Blum and Conover, 1953; Webber, 1968). We 1993] Schneider et al.— Vaucheria 105 have found Vaucheria intermedia only on the eastern half of the Connecticut coastline, from Grove Beach to Waterford, where it is often a dominant species, at times intermixed with V. arcas- sonensis, V. coronata and/or V. velutina. Blum and Conover (1953) found V. intermedia in a habitat with a known wide range of salinities in Massachusetts from fresh water to full-strength sea- water. TYPE LOCALITY: Sweden DISTRIBUTION: Nova Scotia, New England, North Carolina, Cal- ifornia, Washington, Europe. COLLECTIONS: Middlesex Co.—22 Sept. 1991, GRV, /. Cahill, fertile in culture 7 Oct. 1991. New London Co.—2 Oct. 1980, PBW, 7. Roche, fertile 16 Feb. 1981; 2 Oct. 1980, RNS, T. Roche, fertile 18 Feb. 1981; 18 Feb. 1981, BOL, 7. Roche, fertile 10 June 1981; 19 Feb. 1981, RNS, 7. Roche, fertile 11 June 1981; 30 Nov. 1982, PBW, S. Heminway, fertile 8 Dec. 1982; 12 Sept. 1992, HMW, H. Miller, fertile 4 Nov. 1992; 9 Dec. 1992, PBW, B. Stockton, fertile 22 Dec. 1992. VAUCHERIA LITOREA C. Agardh 1823, p. 463. Figures 3a—c Although this species was previously reported for Connecticut as early as the turn of the century (Collins, 1900), we found Vaucheria litorea to be one of the more rarely collected species, only finding it within the last two years of the study in Saybrook Point salt marsh at the mouth of the Connecticut River. Here it was found in the same mud samples with V. compacta and V. velutina. In neighboring Rhode Island, only trace amounts of this alga were found in open ocean and estuarine habitats (Villalard- Bohnsack et al., 1988). Along the Atlantic coast, Vaucheria litorea is established in salt marshes and brackish ponds (Blum, 1972; Ott and Hommersand, 1974), but in the Gulf of Mexico it is characteristic of low salinity areas and habitats that are only infrequently flooded with brackish water, areas with 1.0-2.6 ppt (Pecora, 1977). TYPE LOCALITY: Denmark DISTRIBUTION: Nova Scotia, New England, New York, New Jer- sey, North Carolina, Gulf of Mexico, Washington, Europe, Aus- tralia. COLLECTION: New London Co—15 Oct. 1991, OSP, L. MacDonald, fertile in culture 2 Nov. 1991. 106 Rhodora [Vol. 95 Figure 3. Connecticut marine and brackish water Vaucheria species, scale bars = 100 um unless otherwise noted: a—c. V. litorea, OSP coll. Oct. 1991; a, anther- idium; b-c, oogonia; c, scale = 500 um. d-e. V. minuta, JCV coll. Sept. 1982, scales = 50 um. f-g. V. nasuta, PBW coll. Feb. 1992. h-i. V. velutina; h, GRV 1993] Schneider et al.— Vaucheria 107 *VAUCHERIA MINUTA Blum et Conover 1953, p. 399. Figures 3d, e Originally described from Great Pond, Falmouth, Massachu- setts, this alga was otherwise only known in New England from Maine (Blum and Conover, 1953; Webber, 1968). In Connecticut, we only infrequently found Vaucheria minuta along the eastern half of the coastline. Because of its small size, Ott and Hom- mersand (1974) suggested that this alga was perhaps overlooked and overshadowed by larger Vaucheria species in the same hab- itats. This could help account for the small number of collections in our study, as each collection of V. minuta in our mud quadrats was associated with V. arcassonensis, V. compacta, V. coronota, and/or V. velutina. Blum and Conover (1953) discovered fruiting in a narrow range of natural marsh salinities 26-30.2 ppt. In culture, we found Vaucheria minuta with gametangia throughout the range of sa- linities tested, 0.3-30.0 ppt. Christensen (1988) had reported growth of this species in European populations between 2.5—40 ppt in liquid culture, with gametangia from 12.5-25 ppt. In that our cultures presumably included residual salts within the crude mud samples, we cannot state that the salinities in the muds from which our plants grow were in fact the same as the media applied to them (see following discussion on Salinity Culture Studies), thus fertility in the lowest salinity media may be misleading. TYPE LOCALITY: Massachusetts DISTRIBUTION: New England, North Carolina, Europe. COLLECTIONS: New Haven Co.—23 Sept. 1982, JCV, S. Heminway, fertile in culture 2 Mar. 1983. New London Co.—29 Sept. 1984, HMW, P. Renaud, fertile 15 Apr. 1985; 29 Jan. 1992, HMW, J. Cahill, fertile 19 Feb. 1992; 6 May 1992, HMW, C. Schneider, fertile 13 May 1992. *VAUCHERIA NASUTA W. R. Taylor et Bernatowicz 1952, p. 408. Figures 3f, g Prior to this study, Vaucheria nasuta was reported from New England only in Barnstable Marsh on Cape Cod (Bernatowicz, a coll. Sept. 1991; i, RNS coll. Oct. 1980. j-k. V. vipera, JCV coll. Sept. 1982, scales = 25 um. 108 Rhodora [Vol. 95 1958) and two sites in Gloucester, Massachusetts (Webber, 1976). Webber (1976) found this species growing on the Spartina leaf litter at the separation of the lower and upper salt marsh. We found this species in one salt marsh; a small, protected habitat along the bank of a tidal channel behind Pleasure Beach, Water- ford, underneath Spartina and intermixed with populations of V. arcassonensis, V. coronata, and V. velutina in the same mud quad- rat. Although V. nasuta is apparently uncommon in New England, it is the second most abundant species along the North Carolina coast after V. arcassonensis (Ott and Hommersand, 1974). Blum reported this alga from California in 1971, the only known non- Atlantic Ocean site. TYPE LOCALITY: Bermuda DISTRIBUTION: New England, North and South Carolina, Ber- muda, Gulf of Mexico, California. COLLECTIONS: New London Co.—27 Feb. 1992, PBW, J. Cahill, fertile in culture 1 Mar. 1992; 17 Mar. 1992, PBW, L. MacDonald, fertile in field. VAUCHERIA VELUTINA C. Agardh 1824, p. 312. Figures 3h, i In New England, this alga is known from all of the coastal states except New Hampshire (as Vaucheria thuretii, Farlow, 1881; Col- lins, 1900; Blum, 1972; Villalard-Bohnsack et al., 1988). V. vel- utina is widespread in North America and the most commonly encountered species in the Gulf of Mexico, where it grows in salinities from 2-35 ppt (Pecora, 1977, 1980). In Connecticut, we also find it more frequently in salt marshes throughout the state than any other Vaucheria species, and at the full range of salinities tested. V. velutina was nearly always found growing in the same mud quadrats with V. compacta, and also with all of the other species except V. litorea, thus paralleling intrageneric associations in Europe (Christensen, 1987). TYPE LOCALITY: France DISTRIBUTION: New Brunswick, New England, New Jersey, North Carolina, Florida, Gulf of Mexico, Washington, British Colum- bia, Iceland, Europe, West Africa, Indian Ocean, Asia, Hawaii, Australia. COLLECTIONS: Fairfield Co.—23 Feb. 1985, SHI, P. Renaud, fertile in culture 30 Apr. 1985. New Haven Co.—23 Sept. 1982, JCV, S. Heminway, fertile 11 Oct. 1993] Schneider et al.— Vaucheria 109 1982; 29 Sept. 1984, CGH, P. Renaud, fertile 29 Jan. 1985; 5 Mar. 1987, SVM B. Bothwell, fertile 12 Mar. 1987; 1 Mar. 1992, CGH, L. MacDonald, fertile 17 Mar. 1992. Middlesex Co.—5 Mar. 1987, IRC, B. Bothwell, fertile 4 Apr. 1987; 5 Mar. 1987, GRV, B. Bothwell, fertile 4 Apr. 1987; 22 Sept. 1991, GRV, J. Cahill, fertile 1 Oct. 1991; New London Co.—2 Oct. 1980, HMW, 7. Roche, fertile 10 Mar. 1981; 2 Oct. 1980, RNS, 7. Roche, fertile 18 Nov. 1980; 18 Aug. 1984, HMW, P. Renaud, fertile 7 Sept. 1984; 29 Sept. 1984, HMW, P. Renaud, fertile 5 Oct. 1984; 3 Mar. 1987, HMW, B. Bothwell, fertile 4 Apr. 1987; 27 Feb. 1992, PBW, J. Cahill, fertile 19 Mar. 1992; 12 Sept. 1992, RNS, B. Stockton, fertile 21 Sept. 1992. VAUCHERIA VIPERA Blum 1960, p. 298. Figures 3j, k Previously reported from Bridgeport, Connecticut along with the type collection from Barnstable Harbor, Cape Cod and South Essex, Mass. (Blum, 1960), Vaucheria vipera has not been re- ported since in North America, although it has been discovered in Europe and Asia (Simons and Vroman, 1968). We collected it only once, in a salt marsh off Trolley Road, Sachem Head at Joshua Cove, Guilford, from the high intertidal banks of the tidal inlet approximately 17 m from the cove on compacted sediment covered with cyanobacteria and sparse Spartina. This habitat would have been inundated at high tide with full-salinity seawater, especially during times of low precipitation. Blum (1960) found this delicate species mixed with either V. compacta and V. velutina (as V. thuretii) and the Joshua Cove population shares its habitat with V. minuta and V. velutina. TYPE LOCALITY: Massachusetts DISTRIBUTION: New England, Europe, Japan, Hong Kong. COLLECTION: New Haven Co.—23 Sept. 1982, JCV, S. Heminway, fertile in culture 8 Oct. 1982 SALINITY CULTURE STUDIES All of the mud quadrats collected within the last two years were subdivided and cultured in media with salinities of 0.3, 2.0, 4.0, 6.5, 16.0, and 30.0 ppt. Because the technique consisted of placing estuarine mud samples in culture vessells and then subjecting them to culture media with various salinities, the effective salinity of a culture did not take into account the native salts within each mud sample. Nevertheless, the wide range in salinity of the media 110 Rhodora [Vol. 95 into which the mud was inoculated was expected to account for some differences in a particular species’ ability to reproduce. Al- though some species from local populations indeed did not pro- duce gametangia in culture at all salinities, at least one population of Vaucheria arcassonensis, V. compacta, V. litorea, V. nasuta, and V. velutina was found reproductive in all of the salinity media. Although they produced gametangia during the earlier years of the study in the enriched seawater Grund medium, V. coronata and V. intermedia were never identified in the 30.0 ppt salinity ASP-V medium despite producing abundant reproductive struc- tures in subdivided samples at all of the other salinities. Chris- tensen (1988) successfully cultured both species at a wide range of salinities and reported gametangia for V. intermedia between 0-40.0 ppt and V. coronata between 7.5-30.0 ppt. V. minuta was identified in all except the 4.0 ppt salinity medium, but this was considered an artifact of data collection as this very delicate spe- cies was found in all lower and higher salinities and, because of its size, could easily be overlooked in cultures of mixed species. Subcultures from the crude samples invariably produced agar- penetrating siphons within the Petri dishes, with excellent growth at all salinities. After more than a week, all of the cultures initiated the production of numerous aplanospores, and many of these quickly germinated in situ. In the third or fourth week after in- oculation into the agar media, gametangia were finally observed for three species, Vaucheria arcassonensis, V. compacta, and V. coronata. Because more than one species was found in most of the agar dishes and all three species have overlapping siphon diameters (Blum, 1972), we cannot, therefore, presume that aplanospores were produced by all three species. Christensen (1987), however, reported aplanospores of V. compacta forming under culture conditions, yet he rarely collected them in the field together with the identifying gametangia. Christensen also added that asexual spores of V. arcassonensis and V. coronata were unknown. ACKNOWLEDGMENTS We thank Betsy Bothwell, Geoffrey Dimmick, Alisa Furman, Hetal Lakhani, Lanny Miller, Paul Renaud, Thomas Roche, Britt Stockton, and Stephanie Wood for their assistance on this project 1993] Schneider et al.— Vaucheria 111 both in the field and in the lab. We acknowledge a student research grant from the Hartford Chapter of Sixma Xito L.A.M. and J.F.C. LITERATURE CITED AGARDH, C. A. 1822-3. Species Algarum, Vol. 1, Part 2. Lund. 1824. Systema Algarum. Lund. BERNATOWICz, A. J. 1958. A Bermudian marine Vaucheria at Cape Cod. Biol. Bull. Mar. Biol. Lab. Woods Hole 115: 344. Birp, C. J., T. EDELSTEIN AND J. MCLACHLAN. 1976. Investigations of the marine algae of Nova Scotia. XII. The flora of Pomquet Harbour. Canad. J. Bot. 54: 2726-2737 Bium, J.L. 1960. A new Vaucheria from New England. Trans. Amer. Microscop. Soc. 79: 298-301. . 1966. Sexual periodicity of North American species of Vaucheria. Phy- kos 5: 68-71 ——. 1971. Notes on American Vaucheriae. Bull. Torrey Bot. Club 98: 189- 194. ———. 1972. Vaucheriaceae. North American Flora. Series II, Part 8, pp. 1-63, New York Botanical Garden, New York. . Conover. 1953. New or noteworthy Vaucheria from New age salt marshes. Biol. Bull. Mar. Biol. Lab. Woods Hole 105: 395-401. DR. T. Witce. 1958. Description, distribution, and ecology of three iene of Vaucheria previously unknown from North America. Rhodora 60: CHRISTENSEN, T. 1987. Seaweeds of the British Isles, ieee : ao (Xan- thophyceae). British Museum (Natural History), Lo ——. 1988. Salinity preference of twenty species of acusmee (Tribophyceae). J. Mar. Biol. Assoc. U.K. 68: 531-545. Couns, F.S. 1900. Preliminary lists of New England plants,— V. Marine algae. Rhodora 2: 41-52. . 1909. The green algae of North America. Tufts Coll. Stud., Sci. Ser. 2: 79-480. DANGEARD, P. J. L. 1939. Le genre Vaucheria, spécialement dans la région du sud-ouest de la France. Botaniste 29: 183-265. oe = -C. 1822. Commentationes Botanicae. Observations Botaniques. Tourn ENTWISLE, T. 7 1988. An evaluation of taxonomic characteristics in the Vauche- ria prona (Vaucheriaceae, Chrysophyta). Phycologia 27: 183-200. FARLow, W.G. 1881. The marine algae of New England. Jn: Report of the U.S. Commission of Fish and Fisheries. Report of the Commissioner for 1879, Appendix A. Natural History. Govt. Printing Office, Washington GALLAGHER, S. B. AND H. J. HumM. 1981. Vaucheria of the central Florida Gulf coast. Bull. Mar. Sci. 31: 184-190. Garpary, D. J. AND R. S. Fitcu. 1984. Some brackish species of Vaucheria (Tribophyceae) from British Columbia and northern Washington. Natur- aliste Can. (Rev. Ecol. Syst.) 111: 125-130. 112 Rhodora [Vol. 95 Norpstept, C. F. O. 1879. Algologiska smasaker. 2: Vaucheria-studier 1879. Bot. Not. 1879: 177-190. . W. AND M. H. HomMMeErRSAND. 1974. Vaucheriae of North Carolina. I. ’ Marine and brackish water species. J. Phycol. 10: 373-385. Pecora, R. A. 1976. Vaucheria synandra (Xanthophyceae, Vaucheriales): first record for North America. J. Phycol. 12: 131-133. . 1977. Brackish water species of Vaucheria (Xanthophyceae, Vaucheri- ales) from Louisiana and Texas. Gulf Res. Rep. 6: 25-29. 1978. Occurrence of Vaucheria adela, V. lii, and V. nasuta (Xantho- phyceae 7 brackish marshes of the northern Gulf of Mexico. NE. Gulf Sci. 2: 137- : a peers on the genus Vaucheria (Xanthophyceae, Vaucheri- ales) from the Gulf of Mexico. Gulf Res. Rep 87-391. SCHNEIDER, C. W., M. N. SUYEMOTO AND C. aa 1979. An annotated check- list of Connections a a irises State Geol: sah ign Ae 1-14. Simons, J. 1974. Vaucheria compacta: gal species. Acta Bot. Neerl. 23: 613-626. AND M. VROMAN. 1968. Some remarks on the genus Vaucheria in the Netherlands. Acta Bot. Neerl. 17: 461-466. TAYLOR, W.R. 1937. Notes on North Atlantic marine algae. I. Pap. Mich. Acad. Sei, 22: 225-233. . BERNATOWICz. 1952. Marine species of Vaucheria at Bermuda. Bull. Mar. Sci. Gulf eta 2: 405-413. VILLALARD-BOHNSACK, M., P. PECKOL AND M. M. HARLIN. 1988. Marine mac- roalgae of Narragansett Bay and adjacent sounds. Jn: R. G. Sheath and M. M. Harlin, Eds., Freshwater and Marine Plants of Rhode Island. Kendall/ Hunt, Dubuque, I VON Stoscu, H. A. 1969. Observations on Corallina, Jania and other red algae in culture. Proc. Int. Seaweed Symp. 6: 389-399. Wesser, E. E. 1968. New England salt marsh Vaucheriae. Rhodora 70: 274— 277 ———. 1975. Phycological studies from the Marine Science Institute, Nahant, Massachusetts. I. Introduction and preliminary tabulation of species at Na- hant. Rhodora 77: 149-158. 1976. Rediscovery of Vaucheria nasuta in Massachusetts. Rhodora 78: 556-559. C.WS., LAM, IFC. DEPARTMENT OF BIOLOGY TRINITY COLLEGE HARTFORD, CT 06106 S.W.H. GEOTOXI ASSOCIATES INC. 2377 MAIN ST. GLASTONBURY, CT 06033 RHODORA, Vol. 95, No. 882, pp. 113-118, 1993 VAUCHERIA SUBSIMPLEX CROUAN FROM A MASSACHUSETTS SALT MARSH: FIRST UNITED STATES RECORD EDGAR E. WEBBER ABSTRACT Vaucheria subsimplex Crouan has been collected from a salt marsh at Glouces- ter, Massachusetts. This report appears to be the first citation of this species in the United States, and it is the second in North America. At Gloucester, Vaucheria subsimplex and V. velutina occur as a “zone” at MHW in the mud of tidal creek banks, these algae forming gametangia throughout the summer. Ecological ob- servations of the Gloucester material compare favorably with similar data about V. subsimplex from European and Scandinavian salt marshes. Key Words: Vaucheria, salt marsh, Massachusetts INTRODUCTION Prior to 1960, the only paper dealing with the botany of Mas- sachusetts salt marshes was that of Chapman (1940), which em- phasized the biology of seed plants at a marsh in the town of Saugus. Indeed, field studies of marine algae at that time had not dealt specifically with the salt marsh habitat, but centered about the vegetation of open coasts in temperate regions of the world. Such paucity of algal research in salt marshes prompted field and laboratory studies emphasizing benthic algae in an extensive salt marsh-estuary complex at Ipswich, MA (e.g., Wilce et al., 1970; Webber and Wilce, 1971). Vaucheria species are among the most conspicuous algae in New England salt marshes. Their massed threads commonly form green to greenish-black “turfs” on mud and marine peat substrata. Despite their prevalence, only 4 species of Vaucheria were known from New England salt marshes before 1953. At this time, reports by Blum and Conover (1953), and later by Blum (1960) and Webber (1968), resulted in 10 Vaucheria taxa recognized from Massachusetts salt marshes. The taxa are: V. compacta (Collins) Collins; V. litorea C. Ag; V. piloboloides Thur.; V. velutina C. Ag.; V. arcassonensis Dangeard; V. coronata Nord.; V. intermedia Nord.; V. minuta Blum and Conover; V. vipera Blum; V. com- Pacta (Collins) Collins var. koksoakensis Blum and Wilce. An additional species, V. nasuta Taylor and Bernatowicz, was col- lected from a salt marsh in Gloucester (Webber, 1976). I have 113 114 Rhodora [Vol. 95 since found V. nasuta in abundance at a Yarmouthport salt marsh (Cape Cod), and recently it was located for the first time in Con- necticut (Schneider et al., p. 108). MATERIALS AND METHODS During August, 1987, collections of Vaucheria were made in the Cape Ann region of Massachusetts. The type locality for V. vipera in Essex, and a small salt marsh immediately behind Gloucester Harbor were chosen as collecting sites. At Essex, sam- ples were taken from the mud banks of a tidal creek, as well as from a poorly drained section of the marsh populated by Sali- cornia europaea L. From the marsh at Gloucester, Vaucheria was evident both on creek bands and on the muddy soil beneath Spartina alterniflora Loisel. Collections were returned to Keuka College, Keuka Park, NY, and kept in petri plates under ambient day/night periods for sev- eral additional days until green “tufts” arising from the mud were apparent. At this time, small portions of this new growth were placed in sterile petri dishes containing Erd-Schreiber medium (Provasoli et al., 1957). The algae were then grown in a culture chamber at 16°C under constant light, the dishes being approxi- mately 10 cm below a 20 W cool fluorescent light. Fresh medium was added at weekly intervals, and all cultures were examined microscopically at this time. RESULTS AND DISCUSSION Nine days after inoculation, antheridia and oogonia formed on the Vaucheria from the Essex creek banks, and two species, V. vipera and V. velutina, were identifiable. Collections of mixed Vaucheria species are common. Because of this mixing, it is important to maintain cultures until all the algae have reached a sexual reproductive condition, i.e., until all species from the mix are identified. For example, in a Gloucester site collection, two weeks after initial inoculation, V. /itorea was abundantly reproductive. The algae were removed, preserved in a 3% formalin-sea water mix- ture, and the vegetative material was left in the petri dish. On September 15, 1987, 45 days after initial inoculation, these Vaucheria threads became reproductive. Upon microscopic ex- 1993] Webber— Vaucheria 115 Figure _ _Vaucheria subsimplex Crouan showing the typical terminal game- tangia wit . Material collected from a salt marsh, Gloucester, MA, in summer, 1987. Drawing by Dr. John Blum. Magnification 150 x amination, the algae were identifiable as Vaucheria subsimplex Crouan. At this time, other cultures from the same field location also contained V. subsimplex (Figure 1 Vaucheria subsimplex is of common occurrence in European Salt marshes. The Gloucester, MA record appears to be the first record of this species in the United States, and it is the second in North America. The first report of V. subsimplex in North Amer- ica was based on material from Ungava Bay, Quebec (Blum and Wilce, 1958). The Gloucester and Essex, MA locations w were re-visited in early June, 1992, and additional Vaucheria collections were taken. Cul- ture conditions for these collections were the same as described in METHODS. On June 18, V. subsimplex was located in one collection from the Gloucester tidal creek banks. Mixed with this species was V. velutina, which also appeared in other collections from this site. Thus, a V. subsimplex/V. velutina “zone” exists on the Gloucester mud creek banks at MHW level. Small amounts of V. nasuta also appeared in these and additional collections from this “zone. 116 Rhodora [Vol. 95 The prevalence of a summer reproductive Vaucheria subsim- plex/V. velutina mixture on tidal creek banks at MHW has been reported from the Netherlands (e.g., Nienhuis and Simons, 1971; Polderman, 1975b). In another study, Polderman (1975a) cited the winter and autumn as prime reproductive times for V. sub- simplex. Also, according to Simons and Vroman (1973), Vauche- ria subsimplex and V. velutina are ““summer species” growing mostly on the seaward edges of shores on hydrogen sulfide con- taining substrata where soil moisture and salinity are high. These same authors (1968) stated that V. subsimplex may also reproduce sexually to a lesser degree i in winter and spring. Winter data from Gloucester tion of this species is not avail- able, but V. subsi mplex readily forms gametangia during the sum- mer in Massachusetts Salinity levels at the Gloucester site appear high. Salinities from a site approximately 100 meters seaward from the above location show a range from 22%o to 29%o, with extremes of 17%o and 31%o (Mass. Audubon Soc., 1990). In laboratory experiments (Chris- tensen, 1988) Vaucheria subsimplex showed maximum growth at 22%o, and gametangia were formed between 10 to 30%, thus supporting his field observations that V. subsimplex occurs in sites of “constant and high salinity.” In addition to being a component of the Vaucheria subsimplex “zone” at Gloucester, V. velutina is common also at Essex and deserves further mention. Here, it is mixed with V. vipera and V. compacta and inhabits exposed sites of wet black muck in the upper littoral, subject to sea water dilution by rainfall and flooding by spring tides. Such observations suggest a euryhaline nature for V. velutina, which is confirmed by culture studies and its occur- rence with V. compacta “high up in estuaries” (Christensen, 1988). Further, V. velutina appears to increase its vegetative mass under field conditions through abundant aplanospore production and germination (Knutzen, 1973; Simons and Vroman, 1973; Simons, 1975). Asexual reproduction by aplanospores was common in several of my V. velutina cultures. CONCLUSIONS The first United States record for Vaucheria subsimplex Crouan is reported from a salt marsh at Gloucester, MA. Vaucheria sub- simplex and V. velutina, admixed with small quantities of V. 1993] Webber— Vaucheria 117 nasuta, form a “zone” at MHW on tidal creek banks. Such an occurrence appears related to high moisture and salinity levels of their mud substrata. Vaucheria subsimplex readily forms gam- etangia throughout the summer; winter data concerning the re- production of this species is not yet available. Ecological obser- vations of V. subsimplex at Gloucester compare favorably with relevant data about this species from various salt marshes in Europe and Scandinavia. ACKNOWLEDGMENTS I am most grateful to Dr. John L. Blum for his support and assistance during the course of this study. Sincere thanks are given to: Mr. James Dominick, Gloucester, MA for granting me access to the salt marsh where Vaucheria subsimplex was discovered; Dr. Donald Pfister, Curator, The Farlow Herbarium, for loaning type material of Vaucheria sphaerospora Nord.; Dr. William Dress, Prof. Emeritus, Bailey Hortorium, Cornell University; and to Dr. Robert Buchsbaum, Massachusetts Audubon Society. Particular appreciation is expressed also to Drs. Robert Wilce and Crag Schneider for their constructive critiques of this manuscript. LITERATURE CITED Bium,J.L. 1960. Anew Vaucheria from New England. Trans. Amer. Microscop. Soc. 79: 298-301. AND J. T. Conover. 1953. New or noteworthy Vaucheriae from New England salt marshes. Biol. Bull. 105: 395-401. Wiice. 1958. Description, distribution, and ecology of three species of Vaucheria previously unknown from North America. Rhodora 60: 283-288. CHAPMAN, V. J. 1940. Studies in salt marsh ecology, VI and VII. J. Ecol. 28: 118-152 res a T 1988. Salinity preference of twenty species of Vaucheria (Tri- bophyceae). J. Mar. Biol. Assoc. U.K. 68: 531-545. _ a KnutTzen, J. 1973. Marine species of Vaucheria (X phy ) way. Norw. J. Bot. 20: 163-168. ot MASSACHUSETTS AUDUBON SociETY. 1990. Gloucester Harbor Monitoring Re- port. Gloucester, MA 01930. Nienuuts, P. H. AND J. Srmons. 1971. Vaucheria species and some other algae in a Dutch salt marsh, with ecological notes on their periodicity. Acta. Bot. Neerl. 20: 107-118 th Nor- 118 Rhodora [Vol. 95 POLDERMAN, P. J. G. 1975a. The algal communities of the northeastern part of the salt marsh ““De Mok” on Texel (The Netherlands). Acta. Bot. Neerl. 24: 361-378 . 1975b. Some notes on the algal vegetation of two brackish polders on Texel (The Netherlands). Hydrobiolo. Bull. 9: 23-34. Provasout, L., J. MCLAUGHLIN AND M. Droop. 1957. The development of artificial wedi for marine algae. Archiv. Microbiol. 25: 392-428. SCHNEIDER, C. W., L. A. MACDONALD, J. F. CAHILL AND S. W. HEMINWAY. 1993. The marine ny brackish water species of Vaucheria from Connecticut. Rho- dora 95: 97-112. Stmons, J. 1975. animale species from estuarine areas in The Netherlands. Netherlands J. Sea R — AND M. VRoMaAn. 1968. Some remarks on the genus Vaucheria in the Netherlands. or a Neerl. 17: 461-4 AND . Vaucheria species rani the Dutch brackish inland ponds “De Fhe 3 ee Bot. Neerl. 22: 177-192. Wesser, E. E. 1968. New England salt marsh Vaucheriae. Rhodora 70: 274— RUT ——. 1976. Rediscovery of Vaucheria nasuta in Massachusetts. Rhodora 78: 556-559. AND R. T. Witce. 1971. Benthic salt marsh algae at Ipswich, Massachu- setts. Rhodora 73: 262-291. Witce, R. T., E. E. WEBBER AND J. R. SEARS. 1970. Petroderma and Porterinema in the new World. Mar. Biol. 5: 119-135 DIVISION OF NATURAL SCIENCES KEUKA COLLEGE KEUKA PARK, NY 14478 RHODORA, Vol. 95, No. 882, pp. 119-121, 1993 THE IDENTITY OF ERYTHRONIUM BRACTEATUM (LILIACEAE) BRIAN MATHEW Whilst trying recently to account for all the epithets which have been published in the genus Erythronium I encountered the ques- tion of the identity of ‘Erythronium bracteatum’ Bigelow. Homer D. House, in his Annotated List of the Ferns and Flowering Plants of New York State (1924) gives FE. bracteatum as a synonym of E. americanum and, although I had no particular reason to doubt this opinion, I felt that a sight of the type specimen would be of interest, in view of this curious epithet in a genus which does not normally possess bracts on its inflorescences. The entry in Index Kewensis for this epithet gives the author and place of publication as Bigelow in Beck, Bot. 365 = albidum. This refers to Lewis C. Beck, Botany of the Northern and Middle States (1833), and a check on page 365 reveals that the plant was collected on ‘high mountains, Ver.’ Beck did provide a description of the plant and attributed the name to Bigelow, but without a literature reference to Bigelow’s involvement with the name. The obvious place to start a search for any such involvement was in Bigelow’s Florula Bostoniensis. The first edition proved to be of no help but the second, published in 1824, revealed that the species was in fact validly published with Latin and English de- scriptions and a type citation; Bigelow acknowledged that Boott actually provided the name: ‘discovered by Mr. Boott — — — and by him designated by the foregoing name.’ The full citation for the species is: Erythronium bracteatum Boott ex Bigelow, Fl. Boston. ed. 2:136(1824). Type: USA, Vermont, Camel’s Rump moun- tain, June, Boott (holotype K!). Bigelow provides the following description: “Leaves very un- equal, the primary one being two or three times as large as the secondary. Scape shorter than the leaves with a narrow lanceolate bracte about an inch long, situated an inch or two below the flower. Corolla yellow, half as large as in the preceding species fie. E. americanum], petals gibbous at base. Stamens half as long as the corolla. Style clavate; stigmas united?’ 119 120 Rhodora [Vol. 95 The description given by Beck is almost a straight repetition of that provided by Bigelow. W. J. Hooker was also aware of this plant, although in Flora Boreali-Americana 2:182(1839) he gives no real indication that he knew of the earlier work by Bigelow, in spite of the fact that the Kew Library copy of the Florula Bostoniensis was donated to ‘Professor Hooker—with the author’s respects.’ In a footnote to E. albidum Hooker states: ‘I possess from Dr. Boott a very remarkable state of this plant, or a distinct species, from Vermont, U.S., E. bracteatum Boott, mst. The leaves are lanceolate, and taper gradually into the foot-stalk. The flower is small (34 of an inch long), greenish, and there is a bractea on the peduncle, in- dicating a disposition to bear more than one flower. Only one specimen was found.’ This Hooker connection provided hope that the type may in fact be in existence in the Kew herbarium, and a search through all the Erythronium folders duly revealed a specimen labelled Erythronium bracteatum Bigelow; the reverse of the sheet reads: ‘Discovered on Camel-Rump’s [presumably a mistake for Camel’s Hump!] mountain near Waterbury-Vermont- U.S.A. July 14. 1817 [signed] Francis Boott. This was the only specimen I found in flower & this was faded. I was too late to find it in perfection. An accurate drawing of this was made by my friend Mrs Hardcastle which I sent to Sir J. E. Smith. [ini- tialled] F.B.’ The Kew specimen consists of two leaves without a rootstock, and an inflorescence bearing one flower; the scape does bear a bract just beneath the flower, as observed by Bigelow, Beck and Hooker. The venation of the leaves, and of the perianth segments, is typical of Erythronium americanum and in the absence of any strong contenders it seems likely that E. bracteatum is an aberrant variant of this. The flower is not in good condition so that the internal features are difficult to observe, but it is smaller than that of E. americanum with segments under 2 cm in length. The flow- ering time, given as July 17th by F. Boott, is extraordinarily late for this species and it seems possible that something had happened to this particular individual, such as being buried by a slip of soil or rocks, or beneath a very late-melting snowpatch, causing its abnormal flowering time and morphological appearance; even the ‘bract’ is not very convincing and appears to be somewhat de- formed. On the evidence available I am inclined to agree with Homer 1993] Mathew—Erythronium 121 D. House, that E. bracteatum is a synonym of E. americanum Ker-Gawler. ACKNOWLEDGMENTS I would like to thank Prof. David Barrington of the Department of Botany, University of Vermont, for checking and commenting upon the manuscript. LITERATURE CITED Beck, L. C. 1833. Botany of the Northern and Midland States. Webster and Skinners, Albany, BIGELow, J. 1824. Florula Bostoniensis. Ed II. Cummings, Hilliard & Co., Bos- A , MA. Hooker, W. J. 1839. Flora Boreali-americana. G. Bohn, London House, H. D. 1924. Annotated list of the ferns and oo plants of New York State. New York State Mus. Bull. 254: 5-7 ROYAL BOTANIC GARDENS KEW, SURREY TW9 3AE ENGLAND RHODORA, Vol. 95, No. 882, pp. 122-128, 1993 ISOETES PROTOTYPUS (ISOETACEAE) IN THE UNITED STATES DANIEL F. BRUNTON AND DONALD M. BRITTON ABSTRACT Isoetes prototypus Britton, an endemic of the Atlantic Maritime Ecozone of Northeastern North America, is reported for the first time in the United States from Acadia National Park, Maine. It is a rare species throughout its range, being known elsewhere (in Atlantic Canada) from only five populations. For reasons of site requirements and possibly dispersal limitations, /soetes prototypus is unlikely to be found in the United States away from the coastal region of Maine. It warrants designation as Critically Imperiled (G1S1) throughout its range. Key Words: Isoetes prototypus, Isoetaceae, Maine, Acadia National Park, distri- bution, conservation INTRODUCTION The rarest member of the pteridophyte genus /soetes in north- eastern North America is the Prototype Quillwort, J. prototypus Britton (Britton and Goltz, 1991). It is one of the few northeastern North American diploid Isoetes species (2n = 22), increasing its value for evolutionary investigations of a group in which allo- ploidy and autoploidy are considered to be important for speci- ation (Taylor and Hickey, 1992). Isoetes prototypus also has a variety of morphological characteristics which are rare and in some cases, unique in North American Jsoetes (Britton and Goltz, 1991) (Figures 2 and 3). Isoetes prototypus has been previously reported from only four lakes, all in Atlantic Canada.' A specimen representing an ad- ditional locality and constituting the first record from outside Canada was discovered in material taken in Acadia National Park, Maine. The following describes that population and considers aspects of the ecology, geography and conservation significance of the species. ' A fifth Canadian population has recently been rediscovered at Sandy Lake, Annapolis Co., Nova Scotia (D. M. Britton, A. Anderson and R. Newell 12,506, 23 August 1990, oac). 122 1993] Brunton and Britton— Jsoetes 123 METHODS A recent review ofa set of Isoetes specimens collected in Acadia National Park, Maine resulted in the discovery of a specimen of I. prototypus. The site was subsequently visited and investigations of Isoetes population size and site ecology were undertaken. Com- parison of these data with those gathered in earlier studies (e.g., Britton and Goltz, 1991) was undertaken in order to assess eco- logical and geographical similarities between American and Ca- nadian populations. RESULTS Despite recent surveys of most northeastern /soetes material for several studies (including Kott and Britton, 1983; Taylor et al., 1993 and pers. obs. (1989-1992) in maritime Canada and New England), J. prototypus remained unrecorded in the United States until we identified a collection from the north end of Bubble Pond, Acadia National Park, Maine (C. W. Greene 1782, 18 August 1987, HCOA, OAC). A subsequent search of the site uncov- ered a large population of J. prototypus there (D. F. Brunton and K. L. McIntosh 11,347, 19 September 1992, OAC, DFB, CAN, DAO, MIL, BM). Bubble Pond is a small, spring-fed lake at 101 m asl—a rela- tively high elevation along the coastal plain. The lake is situated in a broad Palaeozoic volcanic and intrusive bedrock trough. Its profile is characterized by a broad, shallow shelf of silty-sand/ coarse sand extending 4 to 40 m out from shore which pitches steeply into much deeper, noticeably darker water (Figure 1). Isoetes prototypus was found growing through dense Eriocaulon septangulare mats in a 20-30 cm thick layer of loose, silty ooze Over sand in deep cool, oligotrophic water at the top of the steep slope to deeper water. The plants were not observed in water less than 1.5-> 2.0 m deep, although label data from the 1987 col- lection reports that those plants were collected in water ca. 1.0 m deep. Bubble Pond J. prototypus has unusually long leaves— 12.5 cm (average of 15 plants)—compared to Canadian material—6.4 cm (average for 22 plants from Nova Scotia and New Brunswick). It is otherwise typical of the species (cf. Britton and Goltz, 1991) viz, extremely straight, stiff, sharp-pointed, lustrous, dark-green 124 Rhodora [Vol. 95 Figure 1. North end of Bubble Pond, Acadia National Park, Maine. /soetes prototypus is abundant well out from shore, from the point where water depth > 1.5 m is achieved (dark area). leaves with mahogany-coloured bases and a velum completely covering the sporangium. Megaspores are small (< 500 wm) and ornamented by an obscure, reticulate network of low, broad, ‘meandiform,’ mound-like markings on an essentially smooth surface. These are typical of the megaspores of J. prototypus (Fig- ure 2). I. tuckermanii A. Br. occurs commonly at the north end outlet of Bubble Pond in 2 to 10 cm deep fresh, warm water on silty- sand/coarse sand flats; none was observed in deeper water. The presence of a third [soetes species in Bubble Pond, J. macrospora Dur., is indicated by the single up-rooted plant of the hybrid Tsoetes x harveyi A. A. Eat. (= I. macrospora x tuckermanii). It is acommon sterile hybrid within mixed populations of the parent species and always occurs in the company of both parents (Britton, 1991). Both of these species (and their hybrid) can be readily separated from J. prototypus by their more delicate, recurved, reddish-green leaves and larger, intricately ornamented mega- spores (high-walled, densely reticulate pattern with a band of small, more or less distinct spines along the equator—cf. Kott and Britton, 1983). 1993] Brunton and Britton—Jsoetes 125 Figure 2. Isoetes prototypus megaspore with irregular pattern of tubercles and low, meandiform ridges (TOPOTYPE/PARATYPE, York Co., New Brunswick; Hinds 77-41 UNB 28537) (bar = 100 um). This habitat is comparable to Canadian sites. First evidence of I. prototypus is usually offered by plants washed up on shore because the typically abundant deep water populations are not readily apparent from the surface (Britton and Goltz, 1991). Sim- ilarly, I. prototypus was not apparent during an unaided search from the surface of Bubble Pond. We examined that population with the aid of a glass-bottomed box to improve underwater viewing and by swimming down to the plants. We have had success in obtaining plants at other sites by dragging an anchor across the lake bottom and collecting the dislodged plants which readily float to the surface thereafter. Evidence of J. prototypus was not detected in a brief survey of Other apparently suitable lakes in Acadia National Park in Sep- tember 1992; nearby Jordan Pond appears to be a particularly likely candidate but has received considerable botanical attention for over a century without providing evidence of this species. 126 Rhodora [Vol. 95 Oo ao pe #e.8 heey > eee ae as | yy, Na s = ? 4a Pb - \, “J %,, « a F v . * oi hy &. ‘ : , Figure 3. Densely fine-spinulose Isoetes prototypus microspore (TOPOTY PE/ PARATYPE, York Co., New Brunswick; Hinds 77-41 UNB 28537) (bar = 10 um). DISCUSSION Few washed-up plants were evident at Bubble Pond, in contrast to other J. prototypus lakes we have examined. Boating and swim- ming are prohibited in this lake, perhaps lessening the disturbance to the bottom by motors, fishing lines and swimmers. We also suspect that bottom-feeding waterfowl and Muskrats, wintering reptiles and amphibians, and/or spawning fish may be responsible for producing massive shore ‘wracks’ (accumulations of washed- up vegetation) consisting of various species of Jsoetes plants, but this has not been confirmed. Such Jsoetes wracks are regularly encountered along the Atlantic coast and in western North Amer- ican cordillera—pers. obs. The Bubble Pond station of I. prototypus constitutes the south- ernmost population for this endemic of the Atlantic Maritime Ecozone (cf. Wiken, 1986). Its occurrence in a deep, cold, nutrient- poor lake may reflect a particularly strict set of site requirements and/or the species’ inability to tolerate significant competition. 1993] Brunton and Britton—J/soetes 127 We typically find Isoetes prototypus in such sites either growing alone or with very few associates (such as Eriocaulon septangulare With., Isoetes anise or Isoetes x heterospora A. A. Eat.). Such a distinctive and restrictive habitat requirement likely also restricts the distributional potential of J. prototypus, contributing to its global rarity. Dispersal to such sites may be limited by the existence of its complete sporangial velum which reduces the ease of separation and rupture of the sporangium from the sporophyll, perhaps reducing the availability of free microspores and mega- spores for transport by water currents, animal activity or other dispersal agents. It seems likely that J. prototypus will remain a rare species in the United States, found only in the cold, clear Palaeozoic-bed- rock based lakes in the Atlantic Maritime Ecozone along and near the Atlantic coastal plain in Maine. Given the greater intensity of botanical investigation in New England than in Atlantic Can- ada, it seems unlikely that evidence of commonly distributed American populations would have been missed when documen- tation of comparable Canadian populations dates back to the early years of this century. The conservation status of L. prototypus meets the high priority designation of the Nature Conservancy ranking scheme (Morse, 1987; Argus and Pryer, 1990); the species likely warrants Gl (Critically Imperiled) status. Similarly, it warrants Critically Im- periled Provincial/State status (S1) for Nova Scotia, New Bruns- wick and Maine. The rarity of populations of this species and the general vul- nerability of aquatic Jsoetes to water pollution strongly suggests that J. prototypus warrants Endangered (or at least Vulnerable) Species designation in both Canada and the United States. ACKNOWLEDGMENTS Our thanks to congenial and expert field associates Allan An- derson and Karen L. McIntosh who have contributed a great deal to our on-site investigations throughout the range of the species. We also appreciate the opportunity of examining the Acadia Na- tional Park material afforded by Acadia National Park botanist Linda Gregory. The second author thanks the National Sciences and Engineering Research Council of Canada for a grant in aid of research. 128 Rhodora [Vol. 95 LITERATURE CITED Arcus, G. W. AND K. M. Pryer. 1990. Rare Vascular Plants in Canada: Our Natural Heritage. Botany Division, Canadian Museum of Nature, Ottawa. Britton, D. M. 1991. A hybrid Isoetes, f. x harveyi, in northeastern North America. Can. J. Bot. 69: 634-640 . P. Gottz. 1991. Isoetes prototypus, a new diploid species from eastern Canada. Can. J. Bot. 69: 277-281. Kort, L. AND D. M. Britton. 1983. Spore morphology and taxonomy of Jsoetes 163. Morse, L. E. 1987. Rare Pr protection, Conservancy style. The Nature Con- servancy Magazine 37(5): 10-15. TAyYLor, W. C., N. T. LUEBKE, D. M. Britton, R. J. HICKEY AND D. F. BRUNTON. 1993. Isoétaceac H. G. L. Reichenbach—Quillwort Family. Jn: N. Morin, Ed., Flora of North America, Vol. 1. Oxford University Press (in press). AND J. Hickey. 1992. Habitat, evolution, and speciation in Jsoetes. Ann. Missouri Bot. Gard. 79: 613-622. Wiken, E. 1986. Terrestrial Ecozones of Canada. ee Land Classification Series No. 19, Environment Canada, Ottawa 216 LINCOLN HEIGHTS ROAD OTTAWA, ONTARIO K2B 8A8 DEPARTMENT OF MOLECULAR BIOLOGY AND GENETICS UNIVERSITY OF GUELPH, GUELPH, ONTARIO NIG 2W1 RHODORA, Vol. 95, No. 882, pp. 129-136, 1993 CAREX CASTANEA x C. DEBILIS, ANEW NATURAL HYBRID FROM ONTARIO P. M. CATLING ABSTRACT A high level of —— — non- Heasented aaihess, Li intermediacy in qualitative and quantitative plants from the shore of Wendigo Lake in vaomlees Ontario as Carex castanea x CC debilis, anew natural hybrid. This nw taxon has prominent sn pubescence like C. castanea, t relati mm of C. debilis). The cfc density of the spike and many other characters 2 are more or less intermediate. This hybrid is probably scarce because of the evident edaphic isolation of its parents. A key and description are included. Key Words: Carex section Sylvaticae saci castanea, Carex debilis, hybrid, On- tario, Canada, North Am INTRODUCTION Carex x knieskernii Dewey, the hybrid of C. arctata Boott and C. castanea Wahl., both species in section Sy/vaticae, is a wide- spread taxon and one of the better known hybrids in the genus Carex in North America (Cayouette and Catling, 1992), but few other hybrids involving C. castanea have been reported. Since C. castanea is conspicuously pubescent, unlike most North Ameri- can species of Carex, and this character may be expected in its natural hybrids, the hybrids involving Carex castanea should be easy to recognize. Several plants of Carex found along the upper, semi-wooded shore of Lower Wendigo Lake in Larder River Provincial Park, northern Ontario had leaves 1.5-2.5 mm wide and were clearly not referable to C. castanea, the wider leaves of which are 4.5-8 mm wide, yet they had distinctly pubescent leaves and sheaths, Ovate perigynia and drooping pistillate spikes, thus suggesting a hybrid origin involving C. castanea. The non-exserted, poorly developed anthers and development of only four achenes from 700 pistillate flowers also suggested a hybrid origin. The purpose of the work reported here was to assess the likelihood of a hybrid Origin and to provide evidence for the parentage. $29 130 Rhodora [Vol. 95 METHODS Fifteen morphological characters (Table 1) were selected and measured in 10 specimens of each of Carex arctata, C. castanea and C. debilis from throughout Ontario. The same 15 characters were measured in 4 specimens of C. x knieskernii from four wide- ly separated localities in eastern North America and in four spec- imens of the putative C. castanea x debilis hybrid from Wendigo Lake. The length of the longest hair on a lower leaf was measured along the lower edge of the leaf. The total culm height was mea- sured from the roots to the tip of the staminate spike. The length of the lowest peduncle was measured from the culm to the base of the spike. The lowest pistillate spike length was measured from the base of the lowest scale to the tip of the longest perigynium. Although the achenes of the putative hybrids C. castanea x debilis and C. x knieskernii were mostly aborted, they were still partially developed and stalks could be measured. Material measured included: for Carex arctata, Calder 788, 7111, Cody 6464, Dickson & Martin 1656, Dore & Koyama 19938, Laird 25 May 1952, Soper & Lindsay 9659, Varga 5 May 1982, Wood 29 May 1934, Zgierska 27 May 1985; for Carex castanea, Brunton 6022, Calder 3106, 3132, Garton 2073, 7329, 11887, Jennings 14241, Lepage 38019, Lloyd 20 June 1946, Mayall & Cormack 151; for Carex castanea x debilis, Catling & Catling 91-224; for Carex debilis, Barkworth 2041, Calder 1952, 1969, 6505, Garton 4836, Gillet & Dore 7636, Hart 1490, Jenkins 4708, Moore 2609 and Watson 3859; for Carex x knieskernii, Calder 3107, Johnson 28 June 1981, Lepage 36308, Williams, Collins & Fernald 18 July 1902. Means and standard deviations of the characters for the different groups were compared. Additional material of all groups was examined to ensure that the samples were representative, although not necessarily fully comprehensive with respect to the variation within taxa. The key and description are based on a larger sample than that providing the basis for Table 1. RESULTS AND DISCUSSION Although the sample sizes employed are small, they do cover a broad area and examination of many additional specimens sug- gested that the data in Table 1 were indeed representative of the 5 taxa, especially in the northern Ontario portions of their ranges. 1993] Catling— Carex hybrid 131 While both C. arctata and C. debilis Michaux have inconspic- uously scabrous-pubescent sheaths and C. debilis also often has scabrous-pubescent leaves, the pubescence of the putative hybrid was much more prominent, being 0.5—0.6 mm long on the lower leaf surface (Table 1). Of the Carex species occurring in the local area where the putative hybrids were found, only C. castanea, with longer pubescence on the lower leaf surface 0.5~0.7 mm long is comparable, and consequently C. castanea is almost certainly one of the parents. Among the species in the neighbourhood that possessed some characteristics of the putative hybrids were Carex arctata, C. capillaris L. and C. debilis. Considering ways in which the putative hybrids differed from C. castanea, the other parent would have to have relatively narrow leaves, relatively long pseu- doculms, relatively long spikes, more rounded scales, relatively narrow perigynia, and relatively long-stalked achenes. The only species that satisfies all of these criteria is Carex debilis. Carex arctata has leaves too broad and acuminate scales (Table 1), whereas C. capillaris has spikes that are too short (< 2 cm long). Both C. arctata and C. capillaris have relatively short pseudo- culms and achenes without stalks. As is characteristic of hybrids, the four plants of putative C. castanea x C. debilis have some characteristics of one parent, some of the other, some unique characteristics, and many char- acteristics that are expressed as intermediate between the putative parents (Table 1). In leaf pubescence the putative hybrids are like C. castanea, whereas in sheath pubescence they tend more to- wards intermediacy. The perigynia are narrower than in either putative parent, possibly because of the lack of distension by well developed achenes. In the length of the pistillate spike and width of tips of the pistillate bracts, the putative hybrids are closer to C. debilis (Table 1). Intermediacy is apparent to a greater or lesser extent in flower number in the pistillate spike, pistillate bract length and width, achene stalk length, length and width of the staminate scale, and width of the staminate scale near the tip (Table 1). Although the short stalks of the achenes, whether the latter are developed or not, are clearly intermediate, they could be viewed as a developmental anomaly. For this reason and be- cause developed achenes where the stalk length is most easily assessed are scarce, the intermediate stalk length is of limited value in the recognition of the hybrid. Both putative parents occurred within 15 m of the hybrid plants. 9°0 C910 98°0 ig | L0 (ww) dy wo Ww 9 YIpIM “pay “¢] £810 Ov'l 9'I (aa Isl0 vel Sl | (WIL) YIPIM V[BIS BIBUTWTLIS ISOMO] PITY] “PI OST'O ce'P 8 cr soe'0 cS % cP Lt (Wit) YIsUZ] speos oJeUTUNEIS ISAMO] PIIYL “€] 1p0'0 b7'0 £0 70 cel‘ 690 60 ¢0 (wuz) aseq areiidns Jo ysug] ‘ousYyoy “71 L£S0'0 $0'0 10 00 zS0'0 Am 70 10 (wu) aseq ajeiidns jo yIBua] “wintUAsLIOg *] | €L1'0 SLv 6% cP TL8°0 cH’s VL cv (Ww) YIsUI] WNTUASLIIg “O] 6710 $9°0 80 ¢'0 L710 $90 60 $0 (ww) di wo WUT ¢°Q YIPIM “pry 6 960°0 cel vl cl a ie) tll Cl 60 (WW) YIPIM JoRIg ajeI[NSId JsoMmO] PITY “8 9170 O87 Oe $c OLE°0 VEC Ge yrs (Ww) YIBUZ] 19vIq aILT[NSId samo] PITY *L 916'¢ 00°91 Oe 4 OCI CLOE 00°71 00% OL (ou) Joquinu Jamoy ‘axids areyusid isamo7T “9 961°S OS’ 67 O'SE Oe 34 8c0'Tl Ov'ce 0'6S O81 (iw) YIBua] o¥Ids aIRI[MISId IsaMOT “¢ €7L'I 69 c6 Cs 169°C CSL Gel SP (Wid) YyIsua] ajounped IsamoT “p 8S8'P COLT CVE CCC OLY II 19°9€ 0°09 O17 (1d) 1YyBIaY WYN apNJaj [RIO], “¢ 0S0°0 zs'0 9°0 $0 €S0°0 $00 10 0°0 (WILL) Jeo] JAMO] UO ITeY ISaBUO] Jo YIsUIT *Z Sbc 0 O8'l e4 91 evil Ott cs ae | (UI) YIPIM Jeo] WINUIXR, “| ds uray “xR ‘UL as uray xe ul (situn) Ja}9RIeYyD syjiqap x vauDisv2 “D syiqap ‘D ‘sjigap x vauvjsv2 ‘Dd daneind pur musaysauy x “2 ‘Sigap ‘D ‘bauvjispo “> ‘vIDIIAV XadMD Ul S19\IBALY p] JO} UONIADP prepues pue URI “WINUTXeU ‘WUNUTTUTPY “| 2IQe sst0 90 60 v0 @8I0 80 Pee. 9010 Le0 8h £0 (ww) dn wo wu ¢°Q yp “pry “CT iveto Fb £29 0 FT. ee es SCO St Oe (WU) YIPIM s[BOS oWBUTLULIS SOMO] PITY L “pI coe SUP Sr Se. GIO Sie. Pe ae Sts0 eer «FE CCE (WU) YISUI] B[eOS aIVUTWILIS IsSaMO] PITY “€] cvo'0 cOO 810 0 Osoo0 400 10 0 7ceoo0 600 10 0 (WIL) aseq aye1IdHs Jo yIBud] ‘susYyOW “Z| 040;0 SEO 60 670) 69600 TO CUEOhCUTG BHO'O EI'0 )6=O70 10 (WI) aseq aieiidns jo yidus] “wNtUAsLIOg *[ | Osv'O SBE FF TE O880 Toh Oh ee. Cer0 Cie USCS (Wut) YIBUI] WINIUABIIOg “(| £900 ci0 8&0 TO €E8I0 060 SO -% S800 Seo sd ZD (ww) diy wor WU ¢°Q YIpIM “pry “6 esto Sst 6FF USE 6 UCR OST. OE er Pesied Ol LT cl (WWW) YIPIM JovIq o1eI[NSId somo] PITY, “8 ser0 9St WF Qe 6600 COE PE cee -OFe0 OE GCE OF (wu) YIUI] J9BIg aIRI[NSId IsoMmo] PITY], *L 696L OLE Obb OST OF8F OO'ST O6T OBI O9F9 OTST OBE OLI (OU) JoqQuinuU JoMoOY ‘oxIds oJeI[NSId samo] “9 IpOL OCSb O6S OLE 9F69 SLIE O6E OZ H78E O80Z O87 O9I (wut) YSU] oy¥Ids a1eI[NSId IsaMOT *¢ [dsc 806 6S'ST €9 «68180 TOE 6S UGE CUCIENO «668SUCHC‘i«iE (wis) yidua] ajounpod 1samoT “p SISS €7lLp TLS O6E B8SEL STEP OPS LLE PIT9I SOSS S'h6 8'OF (Wud) 1YysI9Yy WINS INJoj [BIOL *¢ 00 00 OO OO oOs00 480 v0 %ZO 800 950 LO $0 (WU) Jed] JIMO] UO JIeY ISoBUO] Jo YIBUST *Z csoTt 659 S6 SY O80 tr 2 we e080 SOS Ch cs (WW) YIPIM Jeo] WINWIxeY *] GS ur “xeW ‘UW GS uBR; “xeW ‘UW GS URW “xeW UI (st1un) s919RIeYD DIDIID “> Mudaysaiuy x “D DIUDISDI “D ‘ponunuoy “| IqeL 134 Rhodora [Vol. 95 Thus the circumstantial evidence for the hybrid nature of the evidently intermediate plants was strong. In addition to characters noted in the key, this new hybrid may be distinguished from C. x knieskernii by shorter fertile culms 22-34 cm high, and longer lower peduncles 5-9 cm long. Carex x knieskernii has fertile culms up to 54 cm tall, and lower pe- duncles 2.9-4.8 cm long (Table 1). Over much of its northeastern range, C. castanea is a plant of alkaline soils (pH above 6.5), whereas C. debilis is a plant of acid substrates, usually sand or shallow soil over granitic rock. The waters of the Larder River running into Wendigo Lake have a relatively high pH (7.1 in July), but away from the shore substrates are characteristically acid, being either sand or shallow soil over bedrock. Along the shoreline, which is subject to seasonal fluc- tuation, many plant species occur that are characteristic of al- kaline substrates, but they are evidently limited to the region below the level of the highest water. Examples include Carex buxbaumii, Carex flava, Equisetum scirpoides, Platanthera psy- codes, and Rhamnus alnifolia. The C. castanea x C. debilis hy- brids occurred with Rubus pubescens and the rare Scirpus clintonii in a transition from shoreline thicket dominated by A/nus rugosa and Myrica gale to mixed woods with Abies balsamea, Picea mariana, Thuja occidentalis, and Betula payrifera. Carex castanea occurred in the shrub thickets. Above this level is a characteristic acid ground flora which included C. debilis as well as Clintonia borealis, Cornus canadensis, Epigaea repens, Lycopodium spp., and Vaccinium myrtilloides. Consequently a rather unusual sit- uation existed where acid substrates were adjacent to alkaline substrates, thus allowing C. castanea and C. debilis, species oth- erwise edaphically isolated, to come into close contact. The following key and description of the new hybrid will enable it to be differentiated from related taxa. KEY TO CAREX SECT. SYLVATICAE IN ONTARIO la. Leaf sheaths and especially blades glabrous or scabrous, the longer hairs less than 0.1 mm 2 lb. Leaf sheaths and blades conspicuously hairy, the hairs 0.2- 0.7 mm lon 70 0 We ee, ee a we ae Re ce wm) im eo meee) is So we ee Le Oe 1993] Catling— Carex hybrid 135 SES MT GY ah Se LaF a ar set SS ea at Cote a ata eile Sh ree eee Gop als fare seciep ice? wt ves Cates 3a. Achenes mostly well developed ............. C. castanea 3b. Achenes mostly more or less aborted ................. 4 4a. Leaves 1.5-2.5 mm wide ........ C. castanea x C. debilis ei. Leaves 4-72 Gipiy Wide 5 osc cde Sea oe weeks cum. Carex castanea x C. debilis Caespitose perennial. Fertile culms 22-34 cm tall, with reddish bladeless sheaths at the base, these becoming filamentous with age. Leaves 2-5 per fertile culm, mostly in the basal third, the blades 1-8 cm long, 1.5-2.5 mm wide, pubescent with hairs 0.1- 0.6 mm long on both surfaces; sheaths 3-8 cm long, pubescent with hairs 0.1-0.4 mm long, the inner band brownish-translucent, sometimes with minute reddish dots upwardly; ligule of upper- most fertile culm leaf rounded or triangular, 2.5-4 mm long, 0.5- 0.8 mm wide, brownish-hyaline usually with minute red dots and a pubescent margin. Pseudoculms 30-45 cm long including blades 1~40 cm long, 1.5—2.5 mm wide, with sheaths to 10 cm long, ligules 3-5 mm long, as in fertile culms. Infructescence 12-20 cm long; spikes 3-4, peduncled, the lower peduncles 5.2-10.1 cm long; pistillate spikes 2-3, 23-35 mm long, with 12-21 flowers; lowermost bract with blade 6-11 cm long, 1-1.3 mm wide, sheath 30-40 mm long, the upper bracts much reduced; staminate spike solitary, terminal, 20-27 mm long, ca. 20-30-flowered. Lower pistillate scales whitish-brown or pale brown centrally with whit- ish-hyaline margins, ovate, acute, 2.5-3 mm long, 1.2-1.4 mm wide, 0.5—0.8 mm wide 0.5 mm from tip, more or less pubescent on the margins. Lower staminate scales pale brown centrally with broad white hyaline margins, oblong, rounded at the tip, 4.2-4.5 mm long, 1.2-1.6 mm wide, 0.6-0.8 mm wide 0.5 mm from tip, pubescent on the margins. Perigynia ascending, sessile or with a stalk to 0.1 mm long, brown when mature, often with obscure reddish dots and flecks, especially basally, glabrous, 4.5-4.9 mm long, 1~1.4 mm wide, with two prominent lateral veins and 3-5 less pronounced veins, the body ovate tapering to an obscurely bidentate beak 0.9-1.3(1.5) mm long with teeth 0.1-0.2 mm long. Achenes rarely developing, pale brown with lighter brown edges, trigonous, obpyriform and slightly inflated above the stalk, the 136 Rhodora [Vol. 95 body 1.3-1.5 mm long, 0.7—1 mm wide, the basal stalk 0.2-0.3 mm. Styles indurate in lower third, brittle and shrivelled above, straight above achene; stigmas 3. Anthers 3, 0.9-1.2 mm long, not fully developed, nor exserted. Pollen mostly not developed or malformed. MATERIAL EXAMINED: ONTARIO: Timiskaming District: Bayly Twp.; wooded east shore of Lower Wendigo Lake, 47°52’00’N.L., 79°41'19”W.L., Map 31 M/13 983024, 11 August 1991, V. R. Catling & P. M. Catling 91-224 (DAO, MICH). ACKNOWLEDGMENTS Dr. M. J. Waterway of McGill University kindly commented upon the manuscript and examined the DAO specimen agreeing with its putative hybrid identity. Dr. J. Cayouette and Mr. W. J Cody, both of Agriculture Canada, provided some useful sugges- tions. LITERATURE CITED CaYoueTTE, J. AND CATLING, P. M. 1992. Hybridization in the genus Carex. Botanical Review 58: 351-438. BIOLOGICAL RESOURCES DIVISION CENTRE FOR LAND AND BIOLOGICAL RESOURCES RESEARCH AGRICULTURE CANADA, WM. SAUNDERS BLDG. CENTRAL EXPERIMENTAL FARM OTTAWA, ONTARIO, CANADA KIA 0C6 RHODORA, Vol. 95, No. 882, pp. 137-154, 1993 DR. THOMAS F. LUCY: EARLY BOTANIST OF THE CHEMUNG RIVER VALLEY, NEW YORK CAROL L. KELLOFF! AND LEE B. KAss ABSTRACT Thomas Francis Lucy (1844—1906) was an eclectic medical doctor who devoted most of his life to studying the plants of the Susquehanna Valley, specifically the Chemung River Valley, Chemung County, New York. He also spent his later life collecting and exchanging specimens of North American plants. The main col- lection of the Flora of the Upper Susquehanna that he prepared for the Elmira Academy of Sciences is currently housed at EcH. Other specimens collected by Lucy are located at BH-CU, NYS, NY, F, US, and MIN. After his death, many of his North American specimens were deposited in Bur. These are currently being curated at ECH. Key Words: ise Francis Lucy, Susquehanna Valley, ' Chemung County, Up- usquehanna Flora, Elmira Academy mira College perce New York; Buffalo Academy of Sciences, New York INTRODUCTION This biography summarizes the life and work of Dr. Thomas F. Lucy (1844-1906, Figure 1, Table 1), who lived and collected botanical specimens in the Chemung River valley, New York, for more than 30 years (1868-1900; Clute, 1898; Smithsonian In- stitution Archives, Record Unit 221). Lucy’s herbarium and the records of other local botanists provided the framework for Clute’s (1898) Flora of the Upper Susquehanna and Its Tributaries. Clute’s book was important during this period, for it completed a “chain of local floras extending from eastern Pennsylvania to Lake On- tario” (Clute, 1898). In the late 1890’s, Lucy donated more than 1500 beautifully mounted and meticulously labeled botanical specimens to the Elmira Academy of Sciences (Kelloffet al., 1990). Today this collection forms the nucleus of the Elmira College Herbarium (EcH). PERSONAL LIFE AND EDUCATION Thomas Francis Lucy was born on 7 June 1844 in Bredons Nortin Hall parish, Worcester, England. He was the son of Henry ' Current address: Department of Botany, NHB#166, Smithsonian Institution, Washington, D.C. 20560. 137 138 Rhodora [Vol. 95 Figure 1. Photo of Dr. Thomas Francis Lucy from a postcard written ca. 1900 to his daughter Maud (Mrs. Hattie ““Ma oa Seaman), who donated it to the New York State Museum at Albany, NY (193 1993] Kelloff and Kass—Lucy 139 Table 1. Chronology of the life of Thomas Francis Lucy M.D. (1844-1906). 1844 eee in Worcester, spies 1850’s Lived in Corning, New 1865-66 habe Hobart College, ‘atten New York; Member Chi Phi Fraternity 1868 Naturalized Citizen of USA 1870-77 Married Josephine Searles, had two daughters; Lived in Ashland, New York 1879 First main botanical collections 1881 M.D. Eclectic Medical College of the City of New York a ee of Botany, Elmira Academy of Sciences rst botanical publication 1880’s—1900’s nee plants for ‘“‘Upper Susquehanna Flora” 1882 Elected to American Society of Microscopists 1883-1900 Lived in Elmira, New York 1888 Listed in Cassino’s International Scientist’s Directory 1891 Corresponding member Torrey Botanical Club 1892-1900 Prepared Herbarium ih “Upper Susquehanna Flora” for the Elmira Academy of Sci 1895 Listed in Cassino’s Nasa? 's Directory of the United States and ana 1898 Publication of Clute’s Flora of the Upper Susquehanna and It’s Tributaries 1899 Advertised ay of duplicate specimens for “Flora of the Upper Susquehann 1900 Moved to Raiile New York 1906 Died in Buffalo, N Lucy’s duplicate plant specimens donated to Buffalo Academy of Sciences Lucy, an English barrister, and Sarah Hook Lucy (Bureau of Vital Statistics— Buffalo, 1906; New York State Museum at Albany, 1938a). The Lucy family came to America in the mid-1850’s, settling in Corning, New York (New York State Census, 1855). Thomas F. Lucy and his younger siblings, Ellen S. Lucy and Harry W. Lucy, lived in Corning with their mother and attended school (Federal Census Records, 1860). Henry Lucy, apparently ati to find work in the Corning area, became a merchant in York City (New York State Census Records, 1865). In 1865, at age 21, Thomas F. Lucy enrolled at Hobart College, Geneva, New York (Catalogue of Hobart College, 1865-66) and was initiated into the Upsilon Chapter of the “Secret Order of Chi Phi’ on March 14 (Chi Phi Centennial Memorial Volume, 140 Rhodora [Vol. 95 1924; Hobart College, 1865, 1867; Robson, 1977). During the late 1800’s it appears to have been the custom for college au- thorities to grant charters to secret societies, wherever established. The Alpha Omega Fraternity of Hobart College was established in 1866. Lucy is listed as President and provider of this society (Hobart College, 1866). The objective and number of society members are unknown. The Alpha Omega Fraternity dropped into obscurity as quickly as it was established. During his soph- omore year at Hobart, Lucy was the class historian (Hobart Col- lege, 1866). It was noted in the Catalogue of Hobart College (1866- 67) that Lucy was not in residence when the Catalogue went to press. According to the President’s Report (1909-1911), Lucy was listed as being a “non-grad.” Sometime prior to September 1868, Lucy submitted his ap- plication and petition for naturalization to the Steuben County Courthouse, Bath, New York. On 28 September 1868, as wit- nessed by Frank Brown and George Graves of Corning, New York, Lucy became a citizen of the United States (pers. comm. — Marianne Springer, County Clerk’s Office, Bath, New York). Sometime after 1870, Lucy married Josephine Searles, born 1849 (Chemung County Historical Society, 1976) and daughter of Alfred Searles, farmer and mill-wright (Child, 1868) and Sarah Baker Searles of Pennsylvania (Bureau of Vital Statistics— Elmira, 1890). Josephine was a school teacher in Ashland, New York (Federal Census Records, 1870). They lived on the Searles farm in Ashland and had two daughters, Anna May, born 1876, and Hattie Maud, born November 1877 (Federal Census Records, 1880a, 1880b, 1900a). In 1881 Lucy received his Doctor of Medicine degree from the Eclectic Medical College of New York City (Barnhart, 1965). According to the Eclectic Medical College of the City of New York Annual Announcement (1867), the goal of “eclecticism [as practiced at the college] aim[ed] to enlarge and improve the most important portion of practical medicine, . . . in exploring our in- digenous medical botany, and obtaining the medicinal principles isolated from each plant, so as to administer in the smallest dose and most agreeable form.” In addition, “. . . no medical treatment should be allowed that permanently impairs the vital powers.” The candidates were required “‘to have attained the age of 21” and be “‘of good moral character.” Requirements for graduation included the stipulation that candidates must have studied med- 1993] Kelloff and Kass—Lucy 141 icine with a “respectable practitioner” for 2-4 years and must have attended several “full courses of lectures” at an accredited college. Since it was required for the candidate to attend the last lectures at the Eclectic Medical College, Lucy probably resided in New York City during this time. The Eclectic Medical College of the City of New York closed its doors in 1913 due to an inability to meet American Medical Association supported guidelines (Let- ter from W. C. Black, Archivist, Lloyd Library and Museum, to L. Kass and C. Kelloff, 4 August 1992). We believe that Lucy’s interest in botany was reinforced by his education in Eclectic Medicine. Although Lucy is listed as a phy- sician in the 1887 Elmira City Directory and the 1901 Buffalo City Directory, we were not able to find that he had any affiliation with local hospitals or local or regional Medical Associations. If he was a practicing physician he may have initially collected plants for use in treating his patients. It appears, however, considering his affiliations with scientific societies, his interests in medicinal plants led to his life’s work of studying the natural history of the plants of Chemung County and surrounding regions. He may have been able to devote time to his interest in botany due to the financial support provided by his wife. In August 1882, the American Society of Microscopists held their fifth annual meeting in Elmira, New York. It was on the 15th of August that, ‘“‘on recommendation of the Executive Com- mittee,” Thomas F. Lucy was “elected to membership in the Society” (Proceedings of the Amer. Society of Microscopists, 1882). Lucy was dropped from the rolls in 1889 when the society had failed to receive correspondence from him in the preceding three years. Lucy and his family initially lived in the New York towns of Wellsburg, Lowman, and Chemung. Lucy botanized these areas, as his herbarium labels revealed (Figures 2A and B). From Lucy’s (1883a, 1883b) “Notes from Chemung county, N.Y.,”” published in the Bulletin of the Torrey Botanical Club, we learned that Lucy had moved to Elmira, New York, where he subsequently devel- oped an affiliation with the Elmira Academy of Sciences. On 14 April 1885 Josephine Lucy purchased a house for $600 at 215 Mt. Zoar Street, Elmira, from Charles and Aurelia Davis (Che- mung Deeds, 1885). Josephine Lucy died on 21 January 1890 after a long illness (Bureau of Vital Statistics—Elmira, 1890). In her Will (Chemung 142 Rhodora [Vol. 95 ae ane ene ene nee ++ HERBARIUM AMERICANUM, FLORA OF NEW YORK. ‘Powe, Bet. Syuctsterm. Scie y Nom. Be nu trtaenare ry Zaeen & be Nom. ¥ul om. Vul. ee Shon Chums Ce Sie ee deh Lined - Chemmng fey Diet. Geo Dist Ger seer Observ Pnnppeent Ouserv. jetleres - Dies horn err . a dat oe fs cad syaniad Die 5 ges ie LAL es iotavcinn Tad... 1s 2 éf3 Ex He rbarium, Dr T. F. Lae aey- is: Herbarium, Dr. T. F. Luey So ne ee a eee Fiora or tae Urrer Waters OF THE SusquEHANNA. or HE Pees eS seit. senate een nc iE whle meat |) De pals fusbiss nk L. ake Wir ae ah 3 Ex Berby aa es SLs Coleg. sectos Cham, %, hy. F-test Le No. No. (oe ' ry /3 eet Be ce ee ee ASE Cc Figure 2. Examples of the various types of herbarium labels used by Thomas Francis Lucy. These labels are less documented compared with the labels Lucy prepared for es Elmira Academy of Sciences “Upper Susquehanna Flora (Kelloff et al. 1990). A. and B. Unmounted labels from Lucy’s collection given to BUF after his death in 1906. C. Label from herbarium specimen at us obtained in an exchange with ny. D. Label from herbarium specimen at cu in the herbarium of Stewart H. Burnham. Deeds, 1889), Josephine left “‘all of [her] estate real and personal” to her two daughters ‘““May Lucy” [14 yr.] and “Maud Lucy” [12 yr.]. To her husband she left “all monies accruing from [her] shares in the ... Chemung Valley Mutual Loan Association over and above sufficient to pay all [her] obligation hereto. . and “one complete set of the Encyclopaedia ene [sic] to en [her] last wishes as a gift intended for After Josephine’s death, Lucy ed beats “Emma” Shirey, and remained at Mt. Zoar Street with his two daughters, Anna May and Hattie M. [Maud] (Chemung County Census Records, 1892). Emma was born December 1862 (Federal Census Records, 1900a) in Clearfield, PA (New York State Museum at Albany, 1938a). By 1896, Emma’s brother, Wallace Shirey, a Linotype operator for the Gazette office in Elmira, also boarded at Mt. Zoar Street (Hanford, 1896). The Elmira City Directory for 1896 (Han- ford, 1896) does not list Anna May as a resident at Mt. Zoar. According to the Indenture (Bill of Sale) for the Mt. Zoar property 1993] Kelloff and Kass— Lucy 143 (Chemung Deeds, vol. 102, p. 432), Anna May was married to a Mr. Smith (date unknown) and lived in New York City. In a postcard written ca. 1900 (New York State Museum at Albany, 1938b), Lucy expressed his concerns to Maud about his older daughter: “wrote to May at Bufffalo] [but have] not heard anything from her since she wrote before Xmas.” He asked Maud how she liked the photo of him (Figure 1), which was printed on the card, and sent his love to the “babes The Indenture (Chemung Deeds, 1899) also revealed that Maud had married a Mr. Seaman and was living in Elmira. The Federal Census Records (1900b) revealed that Maud was married to Wil- lice M. Seaman, lived at 507 Baldwin, Elmira, and had a daughter, Irene D., born May 1900. They later moved to 153-'2 Steuben Street, Utica, New York, where Willice worked for Martin & Seaman (Utica City Directory, 1904; Anon., 1906). From the file to which the above mentioned postcard is attached, we learned that in 1938, Mrs. Hattie ‘‘“Maud” Seaman, the donor of this postcard, resided in Little Falls, New Jersey. Lucy purchased the house on Mt. Zoar Street from his two daughters, Anna May Smith and Hattie M. Seaman, on 27 May 1899 for the sum of $1500.00 (Chemung Deeds, 1899). In a letter to New York State Botanist Charles Peck (1833-1917), 6 October 1900, Lucy stated that he had recently moved to Buffalo, New York (New York State Museum at Albany, 1900). He may have made this move to be closer to his daughter May and his wife’s sister Mrs. Lewis C. Lanich and her brother Wallace, who boarded with Lanich (Federal Census Records, 1900a). Lucy died in Buf- falo on 29 October 1906 of Bright’s Disease (Bureau of Vital Statistics— Buffalo, 1906). Emma is believed to have died only a few months later in February 1907 (New York State Museum at Albany, 1938a). EARLY INTERESTS IN NATURAL SCIENCES Thomas F. Lucy had a broad interest in the natural sciences. In 1878, he wrote to Spencer F. Baird (1823-1887), Secretary Smithsonian Institution, Washington, D.C. requesting informa- tion and an assortment of publications. Baird was pleased to write back (Smithsonian Institution Archives, Record Unit 33) that he was able to “supply [Lucy] with a copy of Morris’ [1862] Synopsis 144 Rhodora [Vol. 95 of Lepidoptera, as requested.” Lucy had also offered to collect birds and their eggs for the National Museum, but Baird stated that the Museum would only be interested in “‘specimens of the rarer species.” In this same correspondence, Baird referred Lucy to J. M. Emerton of Salem, Mass., as an authority on spiders. In other letters to Baird, Lucy requested copies of H. Loew (1862, 1864, 1873) and R. Osten Sacken’s (1869) Monographs of the Diptera of North America, parts 1-4; Classification of the Cole- optera of North America (Le Conte, 1862, 1873), and also re- quested ““M.O. Vol. 3¢ 1862” (Smithsonian Institution Archives, Record Unit 28). Lucy also requested the Toner lectures I-VII (Woodward, 1873; Brown-Sequard, 1877; DaCosta, 1874; Wood, 1875; Keen, 1876; Adams, 1877; Shakespeare, 1879), which were given in Washington, D.C. on medical subjects. At Elmira College we located a large 1034 x 16% x 1% inch leather-bound book in which Lucy kept notes on his readings and research in natural science. This notebook is currently housed in the Elmira College Herbarium (EcH) and will be deposited in the Elmira College Archives. Among his first entries were tables, notes and descriptions of “Coleoptera” [beetles] ‘““common to the East- ern and Western Continents” as well as those “‘peculiar to Amer- ica, concentrating on Russian America” and species found in Oregon and California. In addition to beetles, Lucy kept notes on fishes from California, Rocky Mountains and the Oregon Terri- tory as reported by ““Hayden’s Survey of Montana and adjacent Territories 1891.” Other “‘orders” of organisms he listed from Hayden’s survey are “Odonata” [dragonflies], “Hemiptera” [true bugs], “Lepidoptera” [butterflies and moths], “Orthoptera” [grasshoppers] and “Plants.” Lucy listed 384 plants including their localities and abundance from “No. 3 Catalogue of Plants collected in Washington Terri- tory Vol. 12 Part 2%. Plants Collected from the Summit of the Cascade Mts. Eastward to the upper Columbia River, and North- ward to the 49" Degree of Latitude between July & Sept. 1853 [No author cited].” In one 38-page section of his notebook Lucy listed a “Flora New York State,” beginning with “Order I Ranunculaceae” and ending with “Order 120 Lycopodiaceae.” Using red ink he listed genera and species followed in black ink by counties where the plants should be found and often a date when he apparently located them. 1993] Kelloff and Kass— Lucy 145 i.e., “Coptis trifolia [in red ink] Salisb. Cohocton Stub. 872 1 [Steuben] Co. Aug. 3. Bogs. Chemung Co. [underlined in red] Lowmans Swamp.” Lucy made notes on “‘the Geographical Distribution of British Mosses” and “Outlines of British Fungology” by the Rev. N. J. Berkley; an “Introductory Essay to the Flora of Tasmania” by J. D. Hooker, and various plant classification systems including an extensive listing of “Bentham and Hooker’s Classification. Reg.[num] Veg.[etabilia].”’ Tables of fossil animals and plants are listed throughout the notebook. It is obvious from his notes and letters to Dr. Lucy was interested in all aspects of natural history. LUCY’S BOTANICAL INTERESTS Although Lucy’s initial interest in natural history was ento- mological, the first record of his interest in botany was published in 1881, the year he graduated from medical school, as a note in the Bulletin of the Torrey Botanical Club (Lucy, 1881). Here, he commented on finding a single specimen of “Cacalia atriplicifolia L. at the side of the railroad at Wellesburg [sic]” in 1874. Having not seen the plant again, Lucy wondered if the seed had been dropped by a passing train. Lucy reported (1891) that his first main botanical collections were made in 1879. Between 1895 and 1898 Lucy sent 266 of his specimens to the University of Min- nesota Herbarium (MIN, letter from B. G. Ownbey to C. Long, 4 Nov. 1986). Clute (1898) reported that Lucy had probably been botanizing since ca. 1868. From Lucy’s notes (Lucy 1883a) and herbarium labels (Figure 2) we know that he also made collections in the New York counties of Livingston and Steuben. But Lucy concentrated his collecting in the small County of Chemung, New York (Lucy, 1882, 1883b, 1891; Millspaugh, 1887). Most of the specimens reported by Clute (1898) for Chemung County were collected by Lucy. His herbarium labels indicate that Lucy was also interested in plants of North America (Figure 2A and C). In a 25 June 1890 letter to the specialist in grasses George Vasey (1822-1893); see Ewan, 1969, and (Smithsonian Institution Archives, Record Unit 220), Lucy thanked Vasey for sending him the No. 1 Contributions of the United States National Herbarium (Vasey and Rose, 1890a, 1890b). He also stated that he was “par- ticularly interested in the Solidagos and Asters” and wanted spe- 146 Rhodora [Vol. 95 cies from the southern states for comparison in order to complete his herbarium. He was willing to exchange specimens with any southern botanist. In the same letter, Lucy commented that Mills- paugh said he, Lucy, had a finer collection than Cornell Univer- sity. He also told Vasey that his collection was filled with north- western species and that Dickenson [whom we believe to be J. Dixon] & [D.] Drake of Portland, Oregon, “say I send them the best specimens they receive.” Lucy was listed in the Bulletin of the Torrey Botanical Club (1891) as a corresponding member, was listed in the Naturalist’s Directory of the United States and Canada (Cassino, 1895) and the International Scientist’s Directory (Cassino, 1888). THE ELMIRA ACADEMY OF SCIENCES— UPPER SUSQUEHANNA FLORA Lucy’s botanical career advanced after he moved to Elmira and became affiliated with the Elmira Academy of Sciences. This so- ciety was established in 1861 for the “pursuit of astronomical and scientific studies” (Elmira Academy of Sciences, 1881) and was housed in the Observatory opposite the Elmira Female Col- lege (name changed to Elmira College in 1890—Barber, 1955). The Observatory was acquired by the college in 1881 (Barber, 1955; Kelloffet al., 1990). Lucy is listed as Chairman of the Elmira Academy of Sciences’ Section of Botany (Elmira Academy of Sciences, 1881, 1897) and a member of the Council of Admin- istration for the year 1886 (Elmira Academy of Sciences, 1881). As reported by Clute (1898) in the preface to the Flora of the Upper Susquehanna, “Dr. Lucy has also presented to the Elmira Academy of Sciences a nearly complete collection of the plants of our region.” This collection is currently housed in the Elmira College Herbarium (Ecu, Kelloff et al., 1990). Flora of the Upper Susquehanna was part of a “general plan for an extended study of the flora of the headwaters of the Susque- hanna river” (Clute, 1898). Based in Binghamton, Broome Coun- ty, New York, Clute (1869-1950) found that very little had been published on the flora of this region. The collections and notes of local botanists provided the needed information for this pro- ject. As described in Clute’s (1898) preface these botanists were “Mr. Frederick V[ernon]. Coville [1867-1937], Botanist of the 1993] Kelloff and Kass— Lucy 147 United States Department of Agriculture, [who] made a very careful examination of the botany of central Chenango county. Dr. C[harles]. F[rederick]. Millspaugh [1854-1923], now curator [appointed 1894] of the Field Columbian Museum, Chicago, made many observations upon the flora in the vicinity of Binghamton and Waverly during ten years’ practice of medicine [1881-1891] in these places. It was in this region that most of the plants figured in his “American Medicinal Plants” [1884] were studied and depicted [in 1920, Millspaugh published The Bahama Flora with N. L. Britton (1854-1923), curator at Ny]. Mr. David F[letcher]. Hoy [1863-1930], Registrar of Cornell University, has collected several seasons in the Susquehanna valley in Delaware county. Dr. Thomas Francis Lucy [1844-1906] has devoted much of his time to the botany of the Chemung valley for the past thirty years. Mr. James A[nsel]. Graves [1828-1909], has botanized exten- sively in the lower Chemung valley, and for the past twenty years has made careful observations on the flora in the vicinity of Sus- quehanna borough, his home. Prof. Frank E[rnest]. Fenno [d. 1920], in the past three years has been indefatigable in investi- gating the botany of Barton, Apalachin and intermediate points. Mr. W[illiam]. Clay]. Barbour (1877-1933), has contributed many notes on the plants of Sayre, Athens and Waverly, and Miss S. A. Brown, has favored us with a very complete list of plants with notes from the vicinity of Unadilla Forks. The author’s knowledge of the flora was acquired during a ten years’ residence at Bing- hamton, in which time the greater part of the region came under his observation (Clute, 1898; Ewan, 1969; Barnhart, 1965; Stafleu and Cowan, 1979).” In a letter to New York State Botanist Charles Peck (New York State Museum at Albany, 1900), Lucy reported: “I have just completed my work for the Elmira Academy of Sciences, which I commenced in 1892.” It was during this time that Lucy began to advertise the sale of his duplicates for the “Flora of the Upper Susquehanna . .. at the rate of Five Dollars per 100 species,” or priced according to their rarity, in a flyer dated December 1899 (Smithsonian Institution Archives, Record Unit 221). This flyer sent to the Smithsonian, stated that “lists of species per Patter- son’s [1892] or Heller’s [1898] check-list could be purchased.” In response to receiving Lucy’s flyer, F. V. Coville (1867-1937) Bot- anist, United States Department of Agriculture, asked J. N. Rose (1862-1928), Assistant Curator, Department of Botany, Smith- 148 Rhodora [Vol. 95 sonian Institution, his thoughts on this flyer. Rose replied to Coville on 25 January 1900 (Smithsonian Institution Archives, Record Unit 221), “Mr. Pollard (1872-1945) (Curator of Plants, Department of Botany, Smithsonian Institution (1895-1903)] thinks that Dr. Lucy’s collection from the Susquehanna would be of undoubted value,” although the National Herbarium must already “have abundant material” from this area. Coville again asked Rose what he recommended. Rose then replied “simply make a selection of the most desirable of these plants.”” On the back of the memorandum, C. L. P{ollard] commented to Rose, “this set would be of undoubted value [since] we have very little from the Chemung Valley.” It was noted on this same memo- randum that a letter was sent to Dr. Lucy on 6 February 1900. Lucy replied to Coville on 12 February 1900: “‘Herewith I send you a list of my Susq. Flora species, numbered according to Pat- terson’s check-list for all those I can now supply. Please check those which you may desire and return same.” We found no further correspondence in the files. According to the National Museum of Natural History Registrar’s Office, the Smithsonian never purchased nor received directly Lucy’s collections from the Susquehanna Flora. However, in that same year Lucy donated 100 specimens from the Susquehanna Valley, New York, to the newly established herbarium of the New York Botanical Garden (Journal of the New York Botanical Garden, 1900). Additionally, another 200 specimens from western New York were sent on exchange to the herbarium of the New York Botanical Garden (Bulletin of the New York Botanical Garden, 1900). The Smith- sonian eventually acquired some of Lucy’s specimens in an ex- change with the herbarium of the New York Botanical Garden (Figure 2C). In 1990, at the New York State Museum herbarium (NYS), we located a number of Lucy specimens collected during the 1880’s. Some of these specimens were obtained in exchange with Ny. The curators at Nys informed us that they do not have the acquisition records or details of their holdings of Lucy’s spec- imens. Dr. Lucy’s specimens have been reported by Chaudhri et al. (1972) as being located at F, MIN, NY, Ns, and BUF. We have also located specimens at Us, on exchange with ny as noted above, and at BH-cu (Figure 2D) probably in exchange with Stewart Burnham (1870-1943), Assistant Curator at cu, whose specimens are in the Lucy collection at Ec. 1993] Kelloff and Kass— Lucy 149 LUCY IN BUFFALO, NEW YORK In a letter to Charles Peck of 6 Oct 1900, sent from Niagara Street, Buffalo, N.Y., Lucy expressed an interest in getting a po- sition that he had learned of from Miss Sletron of the Buffalo Academy of Sciences. The position was on the State Botanical Survey. We do not know whether Dr. Lucy received this appoint- ment. Lucy was also interested in Salix of northwestern United States. In March 1904 he wrote to C. V. Piper (1867-1926) (Office of the Agrostologist), requesting specimens of Salix from that region (Hunt Institute for Botanical Documentation, 1904a). Lucy re- ceived a reply from Carlton R. Ball (1873-1958), Assistant Agro- stologist, informing him that he would be “glad to negotiate ex- change for material from the Southern and Eastern United States” (Hunt Institute for Botanical Documentation, 1904b). Lucy re- plied to Ball’s letter on 3 May 1904 thanking him for his infor- mation (Hunt Institute for Botanical Documentation, 1904c). In this same response he informed Ball that he now had ‘7583 sheets” of North American species and that he would be glad to send some of them in exchange, if he received a list of required species listed by Patterson’s [1892] or Heller’s [1898] check-list numbers. CURRENT STATUS OF LUCY’S PERSONAL HERBARIUM After Lucy’s death in 1906, his botanical collection, still in newspapers, was donated to the Buffalo Academy of Sciences by Emma’s sister, Mrs. L. C. Lanich (New York State Museum at Albany, 1938a; letter from R. Zander to C. Long and L. Kass, 21 January 1986). In 1986, when it was shown to us, the collection was in the original condition; except that some specimens having complete label information had been filed in the BuF collection. Until that time BUF curators had no way to identify most of the specimens by the numbers that were written on the newspapers. However, by using the Lucy specimens at Elmira, we determined that the numbers were either Patterson or Heller check-list numbers (Kel- loff et al., 1990). Apparently the donated specimens had not been accompanied by Lucy’s catalogues or field notebooks. In 1988 arrangements were made for these duplicate specimens to be cur- 150 Rhodora [Vol. 95 ated. The Lucy duplicate collection from BUF is currently being compared with the original Lucy collection at ECH. ECH will pro- vide full labels and a complete set of duplicates to be sent back tO BUF. CONCLUSIONS Although Thomas F. Lucy pursued a degree in medicine, he apparently did not practice this profession. Prior to his pursuit of medicine, he seems to have been widely interested in natural history, concentrating initially on insects and birds and subse- quently on invertebrates, vertebrates and then plants. The year he graduated from medical school his interests in botany seemed to swell and he became the Chairman of Botany for the Elmira Academy of Sciences. It appears that he had time to pursue his botanical interests due to the financial support provided by his wife. Because he prepared a meticulously labeled plant collection for the Elmira Academy of Sciences and kept extensive notes on his research and readings, it is difficult to believe that he did not have a carefully kept field notebook or catalogue of his collections. Unfortunately, we have not located such a book and believe that it may have been lost prior to his sister-in-law’s donating his private herbarium to the Buffalo Academy of Sciences. It pleases us to have been able to uncover so much information on Dr. Lucy and his contributions to the flora of New York State. This in spite of the amount of time that has passed since he began his studies of the flora of one small county. ACKNOWLEDGMENTS We thank the following people for their help and encouragement during this project. Dr. Steve Clemants for locating the photo- graph of Dr. Lucy at the New York State Museum, Albany; Robert Dirig for continued encouragement and professional assistance; Richard Hurley for locating information on Chemung County; Anita Jacobs for locating Dr. Lucy’s obituaries in Buffalo; Anita Karg for bringing to our attention the Carlton Ball Letters at the Hunt Institute for Botanical Documentation; Jan Kather for pre- paring Dr. Lucy’s photograph; Dr. Malcom Marsden for ms. proofreading; Dr. Bryan Reddick for making available financial support from Elmira College; and Dr. Richard Zander for help with the Lucy Collection at BuF. 1993] Kelloff and Kass—Lucy 151 LITERATURE CITED Apams, W. 1877. Toner Lecture VI: Subcutanious surgery, its principles and its recent extensions in practice. Smithsonian Misc. ie n #302. Anon. 1906. Obituary: Lucy. The Buffalo Enquirer. 30 Oct BARBER, W. C. 1955. Elmira College: The First 100 Years. Veitere Book BARNHART, J. H. 1965. Biographical Notes Upon Botanists, Vol. 1-3. G. K. Hall and Co., Boston, MA. BROWN-SEQUARD, C. E. 1877. Toner Lecture II: Dual character of the brain. Smithsonian Misc. Publication #291 BULLETIN OF THE NEw YorK BOTANICAL Awa. 1900. Report of the Curator of the Museum. I(5): 318 BULLETIN OF THE TORREY BOTANICAL CLUB, 1891. List of Members Revised and Corrected. Vol. 18. BUREAU OF VITAL STATISTICS—BUFFALO, NY. 1906. Certificate and record of death. Thomas Francis Lucy. BUREAU OF VITAL STATISTICS~ELMIRA, NY. 1890. Verified Transcript. Death Record. County of Chemung, State of New York. Josephine Lucy. CASSINO, - E. 1888. The International Scientist’s Directory. S. E. Cassino, Bosto : (308. The gray Directory of the United States and Canada. S. E. Cassino, Publ., Bos CATALOGUE OF HOBART Ae Geneva, NY. 1865-66. Edgar Parker, Printer. 1866-67. Edgar Parker, Printer. EWAL. 1972. Index Herbariorum: Part Il (3) Collectars A. Oosthoek’s Uitgeversmaatschappij N. V., Utrecht. CHEMUNG County CENsus RecorpDs. 1892. County Clerk— Benartment of Re- cords, Elmira, as CHEMUNG CouNTy HisTorIcAL Society. 1976. Grave transcriptions: Elliott Searles Cemetery. CHEMUNG Deeps. 14 April 1885. Vol. 85, p. 606. Indenture. Charles B. Davis and Aurelia Davis to Josephine Lucy . 27 May 1899. vs 102, p. 432. iodediune, Anna M. Smith and Ano to Dr. Thomas F. Luc . 23 September a. Vol. 121, p. 402. Will of Josephine Lucy. Cur Put CENTENNIAL MEMORIAL VOLUME. 1924. Upsilon Chapter Cuitp, H. 1868. Gazetteer and Business Directory of Chemung and Schuyler Counties, New York for 1868-9. Journal Office, Wadeaaus Street, Syracuse, New York. Ciute, W.N. 1898. Flora of the Upper Susquehanna and Its Tributaries. Willard N. Clute and Co., Binghamton, NY. Dacosta, J.M. 1874. Toner Lecture III: On strain and over-action of the heart. Smithsonian Misc. Publication #279. Ectectic MEDICAL COLLEGE OF THE City OF New York. 1867. Annual An- E_mira ACADEMY OF SCIENCES. 1881. Charter, constitution and by-laws: with a list of the members etc. O. H. Wheeler Printer, Elmira, NY 152 Rhodora [Vol. 95 —. 1897. Charter, constitution and by-laws: with a list of the members etc. Gazette marek csi Ewan, J. 1969. H y of f Botany in the United Stat Hafner Publishing Company, New York. FEDERAL CENSUS REcorDs. 1860. Sarah Lucy. nee: Co., Corning, NY. . 1870. Alfred Searles. Chemung Co., Ash ——. 1880a. Thomas F. Lucy. (Soundex on Pe Co., Ashland, New York. vol. 10, ed. 63, sheet 4, line 28. . 1880b. Alfred Searles. eee S642) Chemung Co., Ashland, New York. vol. 10, ed. 63, sheet 4 —. 1900a. Thomas F. buck. Sound L200) Chemung Co., Elmira, New York. vol. 22, ed. 30, sheet 14, line 48. ——. 1900b. W. M. Seaman. nee $550) Chemung Co., Elmira, New York. vol. i a. 15, sheet 9, line 21. Hanrorp, G. 1896. Hanford’s — _ and Elmira Hts. Directory. Gazette Co., Book & Job Printers, Elm: HELLER, A. A. 1898. Catalogue of ah American Plants North of Mexico: Exclusive of the Lower Cryptogams. No publisher listed Hospart CoLieGce. 1865. The Echo of the Seneca. VI(1): 10. . 1866. The Echo of the Seneca. VII(1): 12, 18, & 25. . 1867. The Echo of the Seneca. VIII(1): 12 & 32. Hunt INSTITUTE FOR BOTANICAL DOCUMENTATION. 1904a. Carlton Roy Ball papers. Letter to C. V. Piper from T. F. Lucy dated 31 March 1904. . 1904b. Carlton Roy Ball papers. Letter to T. F. Lucy from C. R. Ball ete 25 April 1904. . 1904c. Carlton Roy Ball papers. Letter to C. R. Ball from T. F. Lucy dated 3 May 1904. JOURNAL OF THE New YorkK BOTANICAL GARDEN. 1900. Accessions. I(4): 6 KEEN, W. W. 1876. Toner Lecture V: On the surgical complications and sm of the inued fever, with a bibliography of works on diseases of the joints, bones, larynx, the eye, gangrene, haematoma phlegmasia, etc. Smithsonian Misc. Publication #300. KeLtorr, C. L., L. B. Kass AND A. J. KowAski. 1990. Thomas F. Lucy’s “Upper Susquehanna Flora” Herbarium Rediscovered. Rhodora 92(869): 1-10. Le Conte, J. L. 1862. Classification of the Coleoptera of North America. Part 1. Smithsonian Misc. Publication #136. vol. 8: 279-348. . 1873. Classification of - Coleoptera of North America. Part 2. Smith- sonian Publication #265. vol. 8. Loew, H. 1862. Monographs ~ the Diptera of North America. Part 1. Smith- sonian Misc. Publication #141. Smithsonian Institution. Washington, D.C. . 1864. Monographs of the Diptera of North America. Part 2. Smithsonian Misc. Punication whit. Smithsonian oe Washington, D D.C. ——.. 1873. merica. Part 3. Smithsonian Misi, Publication #256. Smithsonian Jeon Washington, D.C. Lucy, T. F. 1881. Notes from Chemung County, New York. Bulletin of the Torrey Botanical Club 8(10): 115. 18 Notes from ee County, New York. Bulletin of the Torrey Roti! Club 9(5): 7 1993] Kelloff and Kass—Lucy 153 1883a. Notes from Chemung County, New York. Bulletin of the Torrey Rotenned Club 10(1): 8-9. - 1883b. Notes from Chemung County, New York. Bulletin of the Torrey Botanical Club 10(5): 70. 91. The Chemung County Flora—its relation to that of the Southern Tier Counties, and a brief comparison with other portions of New York State. In: Proceedings of the Elmira Academy of Sciences, June Vol. 1, No. 1. The Academy. Elmira, New York. MiLispauGu, C. F. 1887. Notes on the Flora of Cayuta Creek. Bulletin of the Torrey Botanical Club 14(9): 83-186. Morris, J. G. 1862. Synopsis of the described Lepidoptera of North America. Part 1: Diurnal & ise reco Lepidoptera. Smithsonian Misc. Publication. New York STATE Census Recorps. 1855. Steuben Co., Bath, NY. . 1865. Seen Co., Bath, NY. New York STATE Museum AT ALBANY. 1900. Department of Botany Archives. Charles H. Peck letters, 6 October. ——. 1938a. Department of Botany Archives—Biography file: Lucy, Thomas Francis (1844-1906), 4” x 6” index card. Information provided by Mrs. Hattie Seaman 8b. Department of Botany Archives. Post card attached to Thomas OsTEN SACKEN, R. 1869. Monographs of the Diptera of North America. Part 4. Smithsonian Misc. Publication #219. Smithsonian Institution. Washington, PATTERSON, H. N. 1892. Patterson’s Numbered Check-list of North American Plants North of Mexico. H. N. Patterson, Pub., Oquawka, IL. PRESIDENT’S REPoRT. 1909-1911. Hobart College Bulletins 1911. IX (No. 2, PROCEEDING OF THE Anainican: Society OF Microscopists. 1882. Fifth Annual Meeting, Elmira, New York. Steam printing House, Buffalo, Rosson, J., ED. 1977. Baird’s Manual of American College Fraternities. 19th ed. Baird’s Manual Foundation, Inc., Menasha, WI. SHAKESPEARE, E. O. 1879. Toner Lecture VII: The nature of respiratory inflam- mation in arteries after ligature, acupressure and torsion. Smithsonian Misc. Publication #321. SMITHSONIAN INSTITUTION ARCHIVES. Record Unit 28. Office of the Secretary, 1879-1882. Spencer F. Baird. Incoming Correspondence, p. 1479. Letter to Spencer F. Baird from T. F. Lucy, April 1, 1879. —. Record Unit 33. Office of the Secretary, 1865-1891. Joseph Henry, Spencer F. Baird, Samuel P. Langley. Outgoing Correspondence. v. 70, p. 11. Letter to T. F. Lucy, June 28, 1878 . Record Unit 220. United States National Museum, Division of Plants, 1870-1893 Records. Box 9. Letter to G. Vassey from T. F. Lucy, 25 June —. Heckd Unit 221. United States National Museum, Division of Plants, 1886-1928 Records. Box 14. “Flora of the Upper Susquehanna” flyer, De- 99 STAFLEU, F. A. AND R. S. Cowan. 1976-1986. Taxonomic Literature, 2nd ed., Vols. 1-7. Bohn, Scheltema & Holkema, Deventer, Netherlands. 154 Rhodora [Vol. 95 Utica City Directory. 1904. Utica Directory Publishing, Utica, NY. Vasey, G. & J. N. Rose. 1890a. List of Plants collected by Dr. Edward Palmer in 1888 in southern California. Jn: United States National Herbarium Con- tributions, No. 1. Smithsonian Institution. Washington, D.C. 1890b. List of plants collected by Dr. Edward Palmer in 1889 at 1) Lagoon, 2) Cedros Island, 3) San Benito Island, 4) Guadalupe Island, 5) Head of the Gulf of California. Jn: United States National Herbarium Contribu- tions, No. 1. Smithsonian Institution. Washington, D.C. Woop, H.C. 1875. Toner Lecture IV: A study of the nature and mechanism of fever. Smithsonian Misc. Publication #282. WoopwarbD, J. J. 1873. Toner Lecture I: On the structure of cancerous tumors and the mode in which adjacent parts are invaded. Smithsonian Misc. Pub- lication #266. DIVISION OF MATHEMATICS AND NATURAL SCIENCES ELMIRA COLLEGE ELMIRA, NY 14901 RHODORA, Vol. 95, No. 882, pp. 155-157, 1993 DISCOVERY OF SUBULARIA AQUATICA L. IN COLORADO AND THE EXTENSION OF ITS RANGE Jopy K. NELSON AND WILLIAM E. HARMON ABSTRACT A new record for the vascular flora of Colorado is reported. Subularia aquatica L. was found growing on a muddy, subalpine lake bottom in north-central Col- orado during the field season of 1992. This extends its previously known range in the Southern Rocky Mountains by approximately 220 miles (354 km) to the east from the Utah localities and 270-280 miles (435-450 km) southeastward from its Wyoming localities. Key Words: Subularia aquatica L., Colorado, vascular flora, range extension Field work during the summer of 1992 has added a new record to the vascular flora of Colorado. Treatments by Coulter (1885), Rydberg (1917), Fernald (1950), Gleason (1952), Harrington (1954), Gleason and Cronquist (1963), Mulligan and Calder (1964), Dorn (1988), Weber (1976, 1987, 1990), and Weber and Wittman (1992), do not indicate that Subularia aquatica L. occurs in Col- orado. Subularia aquatica L. is a dwarf, aquatic or littoral mustard, in which the flowers are either self-pollinating or when submerged are cleistogamous. Subula from the Latin refers to the awl-shaped leaves which lend to the common name of awlwort. Often found in clear, cold water lakes at higher elevations (Mulligan and Cal- der, 1964), the plants grow on muddy bottoms of lake margins (Gleason, 1952). The discovery of S. aquatica in Colorado was made in the Neota Wilderness Area, Larimer County, (R75W, T7N, Sec. 33), in a small unnamed lake at an elevation of 10,354 ft. Mulligan and Calder (1964) report that S. aquatica is often found growing in conjunction with other plants which have similar awl or grass- like leaves. They give Eleocharis acicularis (L.) R. & S. and similar species of Limosella as examples. The Colorado specimens were found growing abundantly on the muddy lake bottom in associ- ation with Jsoetes bolanderi Engelmann another abundant species there. The discovery of S. aquatica L. in Colorado significantly ex- pands its previously known range in the Southern Rocky Moun- tains. Widely distributed across the northern hemisphere from 155 156 Rhodora [Vol. 95 Greenland, Iceland, northern Europe to the Urals, then disjunct to Kamchatka, and then extending from Alaska to Newfoundland, its furthest distribution southward in North America is into Cal- ifornia, Utah, and Wyoming (Mulligan and Calder, 1964). The closest collections to the Colorado location are from Mirror Lake, Duchesne Co., Utah and New Fork Lakes, Sublette ma Dinwoody Creek, Fremont Co., and Yellowstone Lake, Yello stone National Park, all in Wyoming (Mulligan and Calder, 1964), This discovery of S. aquatica L. in Colorado extends its range eastward from Utah by approximately 220 miles (354 km) and its range southeastward from the Wyoming localities by approx- imately 270-280 miles (435-450 km) Voucher specimens are being deposited at the University of Northern Colorado Herbarium at Greeley (GREE), the University of Colorado Herbarium at Boulder (coLo), and the Rocky Moun- tain Herbarium at Laramie (RM). Special thanks are given to Dan Bach and John Moore for their assistance. LITERATURE CITED CouLTErR, J. M. 1888. Manual of Botany of the Rocky Mountain Region. Ivison, Blakeman, Taylor, and Co., New York, N.Y. 4 Dorn, R. D. 1988. Vascular Plants of Wyoming. WMonintiin West Publ., Chey- FERNALD, M. L. 1950. ied s Manual of Botany, 8th ed. American Book Co., New York, N.Y. 1632 p Gueason, H. A. 1952. The, new Britton and Brown Illustrated Flora of the Northeastern United States and Adjacent Canada. Lancaster Press, Lancas- ter, Pa. 655 pp. AND A. Cronquist. 1963. Manual of Vascular Plants of Northeastern United wy and Adjacent Canada. D. Van Nostrand Co., Inc., New York, N.Y. 810 p HARRINGTON, i DB 1954. Manual of the Plants of Colorado. Sage Books, er: 666 pp. Mu .uican, G. A. AND J. A. CALDER. 1964. The genus Subularia (Cruciferae). Rhodora 66: 127-135. RypserG, P. A. 1917. Flora of the Rocky nes and Adjacent Plains. Steinman and Foltz, Lancaster, Pa. 1110 p Weser, W. A. 1976. Rocky mountain flora. yore Associated University Press. Boulder, Co. 479 pp. . 1990. Colorado Flora: Western Slope. Colorado Associated University Press, Boulder, Co. 530 pp. ——. 1990. Colorado Flora: Eastern Slope. University Press of Colorado, Niwot, Co. 396 pp. 1993] Nelson and Harmon— Subularia ie 1992. Catalog of the Colorado Flora: a Biodiversity Baseline. University Press of Colorado, Niwot, Co. 215 pp. DEPARTMENT OF BIOLOGICAL SCIENCES ROSS 228 UNIVERSITY OF NORTHERN COLORADO GREELEY, CO 80639 RHODORA, Vol. 95, No. 882, pp. 158-165, 1993 FIRE AND CUSCUTA GLOMERATA CHOISY IN OHIO: A CONNECTION? JAMES S. MCCORMAC AND JENNIFER L. WINDUS ABSTRACT Cuscuta glomerata Choisy is a distinctive species largely confined to the central and midwestern prairie region. It was noted as extirpated in Ohio in 1988 (the last collection dating from 1933). Since 1989 ras stations for the species in Ohio have been located. All are associated with prairie remnants which have, in recent years, been subjected to regular controlled burning. It is suggested that fire is essential to rupture the seed coat and stimulate germination. Key Words: Cuscuta glomerata, Ohio, prairie, fire Cuscuta glomerata Choisy (Cuscutaceae), colloquially known as Rope-dodder (Gleason and Cronquist, 1991), is a distinctive species in a genus noted for difficulty in species separation (Austin, 1979). While some species are partly autophytic, many species lack chlorophyll and appear to be entirely parasitic. Rope-dodder is easily identified by its dense, elongate, continuous flower clus- ters which, when in flower, resemble a thick orange-colored pipe- cleaner twining among the vegetation (Figure 1). Other characters which separate C. glomerata from other eastern North American species are its capitate stigmas, distinct sepals, and sessile flowers subtended by several loosely ascending bracts with recurved tips. Rope-dodder exhibits a tendency to parasitize species of Aster- aceae, such as Ambrosia, Aster, Helenium, Helianthus, Liatris, Silphium, Solidago and Vernonia (Yuncker, 1965). Members of these genera are commonly found in prairies, and the range of C. glomerata is largely confined to the central and midwestern prairie region (Fernald, 1950). Fire is known to play an important role in prairie ecosystems by stimulating and maintaining populations of certain prairie species (Anderson, 1982). Historical accounts by early pioneers settling mid-western and central prairie regions report on the ferocity and frequency of these blazes (Brown, 1883). As prairies diminished in size and became increasingly fragmented with the advance of civilization, fire suppression became the norm and thus prairie fires were virtually eliminated (Curtis, 1959). In recent years natural-areas managers have begun to use fire as a tool for the management of prairie communities. In 1983, the Ohio De- 158 1993] McCormac and Windus— Cuscuta 159 ure 1. Cuse itizing Solidago canadensis in Milford Cen- o ter le Union pS hai Ohio. partment of Natural Resources (ODNR) began to experiment with controlled burning on state-owned prairie areas. Since that time DNR’s controlled burning activities have become a vital part of the state’s natural areas management program. In the springs 160 Rhodora [Vol. 95 Table 1. Correlation between burning history and discovery of Ohio Cuscuta glomerata populations. Year County Discovered Fire History Madison 1920 is Pickaway 1930 ? Madison 1933 ze Logan 1989 1984, 1985, 1986, 1987, 1988, 1989 ogan 1989 4 Wyandot 1990 a Union 1991 1987, 1988, 1989, 1990, 1991 Wyandot 1992 1992 ie 1 a 1 1 . - 4 1 times in the last decade, but they were unsure of which years. of 1991 and 1992, ODNR conducted 17 prescribed burns on 10 prairies. In Ohio, most of the prairie areas which existed prior to Eu- ropean settlement have been eliminated; only small, isolated frag- ments remain (Cusick and Troutman, 1978). Since natural forces such as fire no longer maintain these areas, prairie remnants in many cases have been further reduced by the invasion of non- native and woody species. This successional process and destruc- tion of habitat has resulted in several prairie species being listed as extirpated in Ohio, such as Besseya bullii (Eaton) Rydb. (Kit- tentails), Perideridia americana (Nutt.) Reichb. (Perideridia), Viola pedatifida G. Don (Prairie Violet), Dalea purpurea Vent. (Purple Prairie-clover), and Oenothera triloba Nutt. (Stemless Evening-primrose). Other prairie plants are listed as endangered or threatened, including Si/phium laciniatum L. (Compass-plant), Gentiana puberulenta J. Pringle (Prairie Gentian), Valeriana cil- iata T. & G. (Prairie Valerian), Baptisia lactea (Raf.) Thieret (Prairie False Indigo), Sporobolus heterolepis A. Gray (Prairie Dropseed), Platanthera leucophaea (Nutt.) Lindl. (Prairie Fringed Orchid), and Cypripedium candidum Muhl. (White Lady’s-slip- per) (Division of Natural Areas and Preserves, 1992). Cuscuta glomerata was listed as extirpated in Ohio until 1988 (Division of Natural Areas and Preserves, 1988), as the last collection was from Madison County in 1933. On August 12, 1989, the species was rediscovered in a prairie in Logan County (Peskin, 1990). Four other populations have since been located (Table 1). The first site, known as Liberty Fen 1993] McCormac and Windus— Cuscuta 161 State Nature Preserve, was acquired by ODNR in 1981. The area of the preserve where the Dodder occurs has been subjected to spring burns annually since 1984. It is unlikely that C. glomerata was present prior to its discovery in 1989, except possibly in a dormant state. Rope-dodder is a conspicuous, easily detected spe- cies, and botanical research has been conducted at this site an- nually since ODNR assumed ownership. Approximately 75 plants were parasitized by the dodder, mostly Helianthus grosseserratus Martens (Sawtooth Sunflower), with a few Solidago canadensis L. (Canada Goldenrod) affected. Associates in this area include Solidago uliginosa Nutt. (Bog Goldenrod), Aster puniceus L. (Bris- tly Aster), Carex trichocarpa Muhl. (Hairy-fruited Sedge), Phy- sostegia virginiana (L.) Benth. (Obedient-plant), Solidago riddellii Frank. (Riddell’s Goldenrod), Carex sartwellii Dewey. (Sartwell’s Sedge), and Pedicularis lanceolata Michx. (Swamp Lousewort). On September 21, 1989, C. glomerata was located at another site in Logan County, also a moist prairie association bordering a fen. Approximately 100 host plants were infested with Rope- dodder, all of them Solidago canadensis. Other plant taxa in the vicinity of the dodder included Agalinis tenuifolia (M. Vahl.) Raf. (Common Agalinis), Lobelia siphilitica L. (Great Lobelia), Che- lone glabra L. (Turtlehead), Solidago riddellii and Physostegia virginiana. On August 25, 1990, a third population of C. glomerata was discovered, at Killdeer Plains Wildlife Management Area in Wy- andot County. Killdeer Plains is an 8,637-acre tract located in the former Sandusky Plains, a large wet prairie which covered much of Marion and Wyandot counties. Small prairie remnants still persist in and around the wildlife area, which is owned and managed by ODNR, Division of Wildlife (DOW). Large sections of Killdeer Plains are burned on a rotating basis, and DOW per- sonnel indicated that the site where the Dodder grows has been burned several times in the last fifteen years. Approximately 150 plants of Helianthus grosseserratus were parasitized by Rope- dodder. Associate species include Andropogon gerardi Vitman (Big Bluestem), Silphium terebinthinaceum Jacq. (Prairie-dock), and Liatris spicata (L.) Willd. (Spiked Blazing-star). The fourth modern population of Cuscuta glomerata was \o- cated on July 26, 1991, in Union County. This population occurs in the largest prairie remnant associated with the former Darby Plains, which covered ca. 385 square miles in west-central Ohio 162 Rhodora [Vol. 95 Figure 2. Ohio distribution of Cuscuta glomerata. x = pre 1934. O = post 1988. O = original prairie areas of central Ohio. (Adapted from unpublished map by E. N. Transeau.) (King, 1981). This site, known as Milford Center Prairie, is a narrow strip ca. 1.3 miles long on a power transmission line right- of-way owned by Dayton Power and Light. It is managed as a prairie preserve by ODNR, Division of Natural Areas and Pre- serves (DNAP). A number of rare or uncommon prairie plants persist here, including Silene regia Sims (Royal Catchfly), Rosa blanda Aiton (Smooth Rose), Solidago rigida L. (Stiff Goldenrod), Melica nitens Nutt. (Three-flowered Melic), and Lathyrus venosus Muhl. (Wild Pea). The Rope-dodder was found on approximately 30 host plants of two species, Helianthus grosseserratus and Sol- idago canadensis, in an area of the prairie regularly subjected to 1993] McCormac and Windus— Cuscuta 163 fire. DNAP personnel have conducted spring burns at Milford Center Prairie annually since 1987. It seems unlikely that C. glomerata bloomed here in recent years, as this prairie is fre- quently visited by botanists, and is easy to examine due to its narrow width. A fifth population of Rope-dodder was discovered on August 16, 1992, in a roadside prairie remnant near Killdeer Plains Wild- life Management Area in Wyandot County. About 75 plants of Helianthus grosseserratus were infested by dodder. This prairie is drier than the other Cuscuta glomerata sites, and associates include Andropogon gerardi, Veronicastrum virginicum (L.) Farw. (Culver’s Root), Spiraea alba Duroi (Meadowsweet), Ratibida pinnata (Vent.) Barnhart (Prairie Coneflower), Pycnanthemum virginianum (L.) Durand & B. D. Jackson (Virginia Mountain- mint), Silphium trifoliatum L. (Whorled Rosin-weed), and Mo- narda fistulosa L. (Wild Bergamot). Many of the roadsides in this area are regularly burned in the spring by local farmers, and a check with the owner of this site revealed that it has been regularly burned in recent years, most recently in the spring of 1992. Historical collections of Cuscuta glomerata in Ohio are known from three sites in two counties: Madison in 1920 and 1933, and Pickaway in 1930. No habitat information is included on the labels from these collections, but both of these counties lie within former large prairie areas (Figure 2). The Pickaway Plains area extended from three to seven miles across Pickaway County (Bradley 1835), but has been almost entirely destroyed. The Darby Plains included large areas of Madison County. Prior to European settlement, prairies covered between 700 and 1500 square miles within the borders of present-day Ohio (Cusick and Troutman, 1978). Now, only small isolated fragments exist. Because these areas are of interest to naturalists, surviving prairie remnants are well known and studied. The absence of records of Cuscuta glomerata in Ohio from 1933 to 1989, despite its conspicuous appearance, utilization of common host species, and occurrence in frequently studied habitats, suggests that eco- logical conditions necessary for the growth of Rope-dodder were lacking during this period. At least some species in the genus Cuscuta have hard, impermeable seed coats which allow for an extended period of dormancy, and scarification or acid treatments have been necessary in laboratory conditions to induce germi- nation (Tingey and Allred, 1961). It is possible that heat, such as 164 Rhodora [Vol. 95 is generated by prairie fires, may rupture seed coats and stimulate germination. The recent records of this species correspond with the re-introduction of fire into Ohio’s prairie ecosystems by nat- ural-areas managers. Since all extant Ohio populations of Cuscuta glomerata have appeared in areas recently burned, it appears fire may be essential to the stimulation and growth of Rope-dodder in Ohio prairies. Voucher specimens of Cuscuta glomerata collected as part of this survey have been deposited in the following herbaria: CLM, KE, MICH, MU, NY, and os (Holmgren et al., 1990). ACKNOWLEDGMENTS We thank Cecelia Johnston and Guy Denny for providing data on populations of Rope-dodder in Union and Wyandot counties, as well as Greg Schneider and Allison Snow for their helpful comments on the manuscript. Special thanks to Jerry and Carol Baskin and Nancy Strayer, for their assistance. LITERATURE CITED ANDERSON, R. C. 1982. An evolutionary model summarizing the roles of fire, climate, and grazing animals in the = and maintenance of grasslands: an end paper, pp. 297-308. In: J. R. Estes, R. J. Tyrl and J. N. Brunken, oo a rasses and grasslands: iaeraerrns es ecology. Norman, Oklahoma linea D. F. 1979. Comments on Cuscuta—for collectors and curators. Bull. Torr. Bot. Club 106(3): 227-228. BRADLEY, C. P. 1835. Journal of Cyrus P. Bradley. In Anonymous. 1906. Ohio Archaeol. Hist. Public. Vol. 15. Fred I. Heer, Columbus, OH. 514 pp. Brown, R. C. 1883. The history of Madison County, Ohio. W. H. Beers and Co., Chicago, IL. 1165 CuRTIS, J. it Bs, Vegetation of Wisconsin. University of Wisconsin Press, Madison, WI. 657 pp. Cusick, A. AND K. R. TRourMAN. 1978. The prairie survey project—a summary of data to date. Ohio Biol. Surv. Inform. Circ. No. Division OF NATURAL AREAS AND PRESERVES. 1988. Rare native Ohio vascular plants. 1988-89 status list. Ohio Department of Natural Resources, Colum- bus, OH. 19 pp. 1992. Rare native Ohio plants. 1992-93 status list. Ohio Department of Natural Resources, Columbus, OH. 20 pp. FERNALD, M. L. 1950. enh s manual of botany, 8th ed. American Book Com- pany, New York. 1632 p GLeason, H. A. AND A, ae. 1991. Manual of vascular plants of north- 1993] McCormac and Windus—Cuscuta 165 eastern United States and adjacent Canada. The New York Botanical Garden, O pp. HoLMGREN, P. K., N. H. HOLMGREN AND L. C. BARNETT. 1990. Index herbar- iorum, Part 1: the herbaria of the world. Ed. 8. Regnum Veg. Kina, C. C. 1981. Prairies of the Darby Plains in west-central Ohio, pp. 108- 127. In: R. L. Stuckey and K. J. Reese, Eds., Proceedings of the Sixth North American Prairie Conference. Ohio Biol. Surv. Biol. Notes No. 15, Colum- bus, OH. Peskin, P. K. 1990. Knotted Dodder b Syaieas glomerata: Convolvulaceae) in Tinacey, D. C. AND K. R. ALLRED. oa ee dormancy in seeds of Cuscuta approximata. Weeds 9: 429-436. YUNCKER, T. G. 1965. Cuscuta. North American ies Series II, Part 4. The New York Botanical Garden, Bronx, NY. 51 p OHIO DEPARTMENT OF NATURAL RESOURCES DIVISION OF NATURAL AREAS AND PRESERVES 1889 FOUNTAIN SQUARE, BUILDING F COLUMBUS, OHIO 43224 RHODORA, Vol. 95, No. 882, pp. 166-183, 1993 THE BIOLOGY AND TAXONOMY OF THE PORTULACA OLERACEA L. (PORTULACACEAE) COMPLEX IN NORTH AMERICA JAMES F. MATTHEWS, DONNA W. KETRON, AND SANDRA F, ZANE ABSTRACT Portulaca oleracea L. is a polyploid, cosmopolitan weed with a broad physio- logical adaptability. It is morphologically variable, grows rapidly, is self-compat- ible producing large numbers of seeds that have a long survivability. The vari- ability is expressed in a series of local populations that exhibit clinal variation geographically. Data obtained from the literature on ne biology of the species are compared to er to develop a more func- tional taxonomic treatment of this polymorphic species. Key Words: Portulaca oleracea, taxonomy, species biology, cytology, seed mor- phology INTRODUCTION Portulaca oleracea L., purslane, is one of the world’s most aggressive weeds, having been noted as the eighth most frequent plant (Coquillant, 1951) and one of the ten most noxious weeds (Singh and Singh, 1967). It iscommon in cultivated fields, gardens and lawns, as well as waste areas, driveways, eroded slopes and bluffs, occurring from sea level to 2600 m (Vengris et al., 1972). It is most common in the temperate and subtropical regions, although it grows in the tropics and into the higher latitudes. Legrand (1962) reports it from 45° north latitude and 40° south latitude, while Zimmerman (1977) extends this to 51° on the American continent and 54° on the European continent. Recently this had been extended to 58° in the Canadian province of Alberta (J. G. Packer, pers. comm.). The origin of the species is uncertain and new data and inter- pretations have made the accepted theories questionable. De- Candolle (1886) considered it to be Old World (India) spreading into the New World. Ridley (1930) suggested that P. oleracea is a native of deserts or desert-borders of North Africa, and Hagerup (1932) cited the Sahara as its center of origin. Although not citing this species specifically, Geesink (1969) concluded that the an- cestral lineage of the genus is in Australia. In North America this species has been considered as introduced from the Old World. 166 1993] Matthews et al.— Portulaca 167 It is included in Dioscorides’ first century herbal (Gunther, 1959). Gray and Trumbull (1883a) summarized the evidence for a pre- Columbian existence of purslane in North America by noting that Columbus’ diary (1492) had an entry reporting purslane in Cuba. Also Nuttall in Missouri and Long in what is now Colorado, reported purslane before settlers could have transported it from the east. Gray and Trumbull (1883b) allude to a possible Viking entry, possibly from Iceland into Greenland and finally into New- foundland. There is no current report of this species occurring in either Iceland or Greenland. Additionally, Scott (pers. corresp. 1992) reports that purslane is not currently part of the flora of Newfoundland. These data support the latitudinal limits pre- sented above. Byrne and McAndrews (1975) report on the finding of P. olera- cea pollen and seeds in sediment samples ofa lake in southwestern Ontario, Canada, dating from the period 1430-89 A.D., with more intensive analyses extending the range from 1350 to 1539 A.D. Their explanation of this occurrence is based on man’s use of the plant. Indians living in the vicinity of the lake possibly collected plants from fields and washed them in the lake, depos- iting the pollen and seeds in the water. In addition to the purslane residual, maize pollen (Zea mays) and pollen and seeds of sun- flowers (Helianthus annuus) were also found in the same layer of sediment. Furthermore purslane seeds have been recovered from archaeological sites in Kentucky which date back to the first mil- lenium B.C. (Watson, 1969) and from southern Louisiana dating back 1500 years (Walker, 1936). These data indicate an indige- nous development or movement of the species across the conti- nents during prehistoric times (Chapman et al., 1974). Richard Yarnell (pers. comm. 1992) suggests that P. oleracea has been in North America perhaps since 7000 B.C. Further insight on the success of this species can be gained by examining the biology of P. oleracea reported on below from the work of several investi- gators MATERIALS AND METHODS Herbarium specimens were obtained from: ARIZ, ASC, ASTC, ASU, BRY, CAS, COLO, DAV, DES, DUR, F, FLAS, FSU, FWM, GH, HPC, ISC, JEPS, KANU, KS, KSC, KSP, LAF, LL, LSU, MICH, MNA, MO, MU, NLU, NMC, NO, NY, OKL, OKLA, PAUH, RSA/POM, SD, SMU/BRIT, TAES, 168 Rhodora [Vol. 95 TAMU, TEX, UARK, UC, UCSB, UMC, UNCC, UNLV, UNM, US, USF, UT, UTC, UTEP, and uvsT. Live specimens were collected from Cali- fornia, San Bernardino Co. (Valley between Granite and Van Winkle Mts. on Kelbaker Rd., 7 mi. N of I-40; Pass between Vontrigger Hills and Hackberry Mts., E side of Lanfair Rd., 11 mi. N of jct. Goffs Rd.); New Mexico, Bernalillo Co. (base of light pole, San Felipe St., Old Town, Albuquerque); Illinois, Jackson o. (Lawn, Pyramids Complex, S. Rawlings St., Carbondale); North Carolina, Mecklenburg Co. (Garden, 5724 Doncaster Dr., Charlotte); and South Carolina, Beaufort Co. (Along sidewalk, Coligny Plaza, Hilton Head Island). Chromosome squashes were obtained by placing root tips in 0.1% colchicine for 1.5 hr, fixing in 3:1 alcohol-acetic acid overnight, hydrolyzing in 1 N HCl for 12 min. at 60°C, placing in Feulgen stain for four hours and squashing in dilute Aceto-carmine stain to provide cellular detail. Four plants from each site were studied. Photographs were taken using a Zeiss Axioskop equipped with Normarski differential in- terference contrast optics. Seed surfaces were examined and pho- tographed, after sputter coating with gold-palladium on a Hum- mer V, with a Jeol JSM-35CF scanning electron microscope. THE SPECIES BIOLOGY OF PURSLANE The — is mostly an annual, but may be perennial in the tropics. The 1 te or opposite, obovate to spatulate with an obtuse or truncate-emarginate apex. The leaves may range from 40 mm by 15 mm up to 60 mm by 25 mm in fertile soils. Apical verticils have 2-5 leaves, with usually 4. Axillary hairs are missing, inconspicuous, or barely visible. Flowers are in a group at the end of the stem subtended by 4 leaves. The two sepals are fused to the base of the ovary and may grow into a wing-like carina of 3-4 mm long that can cover the ripening fruit. There are (4)5(6) yellow petals ranging from 3-10 mm long by 2-8 mm wide with 6-15 (3-20) stamens. The style branches are 3-6, the capsule ranges from 4-9 mm, dehiscing at or just below the middle. Seeds are black, 0.6—1 mm, usually with granulate (Figure 1) to flat stellate (S-undulate) (Figure 2) surfaces and small tubercles along the dorsal curvature. However the seed surface can also be raised stellate (S-undulate) with tubercles on the lateral seed surfaces (Figure 3) or exhibit intermediate patterns (Legrand, 1962; Geesink, 1969; Kim, 1989). 1993] Matthews et al.— Portulaca 169 Zimmerman (1976) reported on the growth characteristics of weediness in P. oleracea, in comparison with P. pilosa and P. grandiflora. Common purslane branches almost immediately af- ter germination and therefore spreads out in its growth pattern quicker than the other two species. It produced flowers in day lengths from 4-24 hours, with a direct correlation between in- creased illumination and flower production (except for 20 hr.). There was no flowering photoperiod. Capsule production and overall plant growth (dry matter) also increased with increasing day length. When tested with nonvarying light and varying tem- peratures, ranging from 16°/11°C (day/night) up to 28°/22° the number of capsules increased by a factor of 1.3. When tested for productivity in varying moisture regimes, purslane was able to mature capsules in soils of high or low moisture. Zimmerman (1977) reported on thet tem and seed physiology. Purs- lane flowers did not open on “cloudy days or days below oA a When opened, they remained open for 4 hours. The flowers were self-fertile, and under experimentation did not exhibit apomixis. No insects visited the flowers. Mulligan (1972) saw no insect visitors to purslane flowers over a three-year study. These same features of autogamy, no apomixy and lack of insect pollinators were supported by Kim and Carr (1990a). Miyanishi and Cavers (1980) reported evidence of at least 5% outcrossing in purslane and Mulligan (1972) and Zimmerman (1976) proposed that out- crossing is accomplished by wind. Our examination of the flowers showed the pollen to be sticky, not a characteristic of a wind borne process. Kim and Carr (1990b) reported a natural hybrid in which P. oleracea was one of the putative parents. In Zimmerman’s study (1977), seed germination remained con- stant, at over 90%, over a 2.5 yr. period. Vengris et al. (1972) reported the following percentages of germination for seeds: 0.2 year-old, 39%; 1 year-old, 78%; 7 years-old, 59%; 14 years-old, 59%. Muenscher (1955) reported that purslane seeds were viable for at least 40 yr. Zimmerman (1977) found over 60% seed via- bility after passing them through a house sparrow (Passer do- mesticus). Light was required for germination, but the tempera- ture requirements were variable. Purslane seeds can germinate at 10°C and ina test of overwintering, over 82% of the seed survived a winter in Michigan. Singh (1968) showed that seeds collected in India germinated over a range of 10°-40°C, but did not ger- minate at 50°C. Vengris et al. (1972) noted that a purslane plant 170 Rhodora [Vol. 95 Figure 1. Granulate pattern in P. oleracea. Figure 2. Stellate (S-undulate) pattern in P. oleracea. Figure 3. Stellate (S-undulate) tuberculate pattern of a specimen labelled P. retusa that matches Engelmann’s lectotype. 1993] Matthews et al.— Portulaca 171 of 28 cm produced 6723 seeds after 5-6 weeks of growth (e.g., the first flush of flowers). Zimmerman (1976) estimated that over an entire season a purslane plant can produce 101,625 to 242,540 seeds. From the above, Portulaca oleracea can be characterized as a weed of open, disturbed habitats with nearly a world-wide adapt- ability and distribution. It grows rapidly, producing flowers, fruits and a large number of seeds within six weeks of germination. It has a wide tolerance of photoperiod, light intensity, temperature, moisture and soil type. Seeds germinate under conditions that enhance the survival of seedlings. The self-compatible breeding system, longevity of seeds, resistance of the seeds to the digestive processes of animals, and the overwintering capacity of the seeds help ensure its survival and distribution. These characteristics have a bearing on the taxonomy of the species as described below. TAXONOMY In an ancient, cosmopolitan species in which autogamy is the rule, local populations possibly reflecting variable morphological and physiological traits can be expected. The fact that the species has not been split into a plethora of microspecies (as defined by © Stace, 1989) is surprising. An examination of the two most recent comprehensive monographs, Legrand (1962) and Geesink (1969), show the following: Legrand’s treatment of the American species has only three synonyms at the specific level under P. oleracea. Geesink’s treatment of the Indo-Australian and Pacific species places only two in synonymy. There thus appears to have been a reticence to subdivide P. oleracea. Whether this is due to a lack of understanding of the genus or of an appreciation of the vari- ability of this species is unknown. In reality, both reasons probably are contributors. An examination of the works of Legrand and Geesink, cited above, shows several examples of taxonomic/no- — Figure 4. Raised stellate (S-undulate) pattern without tubercles labelled as P. retusa. Figure 5. Raised stellate (S-undulate) pattern with incipient tubercles, labelled as P. retusa. Figure 6. Granulate pattern with tubercles labelled as P. retusa. 172 Rhodora [Vol. 95 menclatural problems in species less widely distributed than P. oleracea. Gorske et al. (1979) provide some insight into the world-wide variability of P. oleracea in their numerical taxonomic study. Forty-four ecotypes representing both weedy and cultivated forms from 18 countries were analyzed, using 36 morphological char- acteristics. The cultivated forms, used for food, were obtained from commercial seed sources in Egypt, France, Norway and the United States. No physiological or cytological characters were included in the study. Three different methods were used in the analysis: the unweighted pair group method of clustering using arithmetic averages, principal component analysis and stepwise multiple discriminant analysis. All three analyses gave similar results and show that the cultivated forms can be separated from the weedy populations. The weedy populations can be divided into three distinct groups: cool temperate, warm temperate to subtropic and humid subtropic to tropic. The Canadian popu- lations fit into the cool temperate. The populations from the upper USS. fit into the warm temperate to subtropic group while those from the lower U.S. fit into the humid subtropic to tropic group. The effects of selection for cultivation (forming a separate group) are apparent, as well as are the lack of smaller groupings over large geographic areas. That this study over such a large geograph- ic distribution with a large representation of ecotypes showed only three groups that fit a clinal variation from warm humid subtropic to cool temperate, and did not segregate into groups that could be considered as microspecies supports our conclusion that P. oleracea is best viewed as a single polymorphic species. As dis- cussed below under the cytology, within the U.S., three cytotypes occur. There is no well-defined cytological correlation with the cool temperate, warm temperate to subtropic and humid sub- tropic to tropic groups. Portulaca oleracea is a weed that exhibits wide geographical, morphological, physiological and cytological plasticity, an obvious reason for its success. Several names have been proposed as segregates within this complex. One such is P. neglecta Mackenzie and Bush, based on a specimen collected by Bush in Courtney, MO in 1900. Type: U.S.A. Missouri, Jackson Co., 24 Aug 1900, B.F. Bush s.n. (Ho- lotype: Mo!, Isotype: ny!). Courtney no longer exists, having been eliminated by the growth of Kansas City. An examination of the type specimens and seven other specimens identified as P. neg- 1993] Matthews et al.— Portulaca 173 lecta (the only ones received from all loan requests) from Mo, KS and mi have vegetative features and seed surfaces typical of P. oleracea. The leaves are large, 40 mm by 15 mm; but this is typical for large specimens of P. oleracea. In the original descrip- tion of P. neglecta, Mackenzie and Bush (1902) differentiated the new species from P. oleracea on the basis of earlier opening flow- ers; larger size of the plant; larger, broader and thinner leaves and more numerous stamens. He said that P. neglecta occurred in MO, KS and AR and is probably what has been called P. retusa Engelm. in MN, MO and KS. In correspondence from Aug to Nov 1909, from ny, attached to a herbarium sheet without a plant, N. L. Britton asked Bush for fresh material for study. Bush’s reply was that the species grew in rich, sandy bottoms and the floods of the past two years had apparently destroyed the popu- lations. Bush also added that the other form, typical P. oleracea, grew in pastures and looked very different. From this exchange, we conclude that the large specimens were either escaped culti- vated forms or represented local populations from wet habitats in which the plants grew larger than those in drier pastures. All of the features Mackenzie used to differentiate P. neglecta are easily modifiable vegetative features. No specimens have been reported since these initial collections. We conclude that P. neg- lecta Mackenzie and Bush is P. oleracea L. and that the former name should be in synonomy under the latter. In another study of this species, Danin et al. (1978) proposed nine subspecies of Portulaca oleracea, based on seed size, seed surface and chromosome number, with the cytotypes having dis- tinct geographic separation. We recognize the need to express infraspecific variation (Matthews et al., 1991, 1992a); and as Stace (1989) concluded, “the provision of a name attracts attention to a taxon.” In the genus Portulaca each species must be studied to determine which characters are consistent and persistent, a dif- ficult process. Due to the almost cosmopolitan distribution of P. oleracea with its adaptability to many habitats, the propensity for self pollination, long term seed viability and the lack of specialized seed dispersal mechanism, this species exists in a myriad of local populations that exhibit morphological and/or physiological vari- ability peculiar to each deme. Since the species is succulent, tax- onomic characters have been sought in the more durable struc- tures, such as seed surface. In some species of Portulaca these surfaces appear to be consistent and predictable but in the wide- 174 Rhodora [Vol. 95 spread P. pilosa the seed surface morphology was neither consis- tent nor predictable over the geographical range of the species (Matthews and Levins, 1985, 1986; Matthews et al., 1992a, 1992b). Legrand (1962) stated that seed surface characters, particularly in P. oleracea, are not dependable because the variation is unpre- dictable. Each population often presents a different pattern. He concluded that seed surface cannot be used to separate even va- rieties because infinite transitions are produced between non- tuberculate and tuberculate and between rugose and stellate sur- face patterns. We find this same situation in P. oleracea where there is a mosaic of seed surface morphotypes. We do not think that these morphotypes should be treated as subspecies. We tend to follow in principle the concept of subspecies as defined by Stuessy (1989) where a subspecies should have several conspicuous morpholog- ical differences, should be cohesive geographically and largely allopatric, with multigenic control of the differences. Additionally, there is the possibility of hybridization along contact zones but with markedly reduced fertility of hybrids. In a subsequent study by Danin and Anderson (1986) on the subspecies of P. oleracea in Florida, five subspecies representing three cytotypes were identified and their distributions mapped. Most of these five overlapped both in geographic distribution and in chromosome number. In addition there were several specimens that could not be assigned to any subspecies, implying more vari- ation than had been circumscribed in the nine subspecies already described. These were suggested to be potential hybrids. This does not mean that seed surface morphology is worthless as a taxonomic character. In some genera it has proven useful in defining subspecies. Walters (1964) in Flora Europaea recognized four subspecies of Montia fontana (Portulacaceae) based on seed surface morphology and very sharply distinct geographic distri- butions. Here the morphological character is enhanced by allo- patric distributions, a situation not present in the study by Danin and Anderson (1986) of P. oleracea in Florida. We conclude that P. oleracea exists as a polymorphic species and is not divisible into subspecies based on seed surface as the primary morphological trait. Another part of the P. oleracea complex is P. retusa Engelm.., described in 1850 (Gray, 1850) from specimens collected by Lind- heimer in the granitic region of the Liano (sic), Llano of western 1993] Matthews et al.— Portulaca 75 Table 1. Herbarium sheets of P. retusa Engelm. considered as lectotypes. 1. P. retusa Engelm., Aug 1849. Engelmann s.n. (MO, 122802), specimens num- bered 8871 and 8872. 2. P. retusa Engelm., no date. Engelmann s.n. (MO, 122803). P. retusa Engelm., no date. Engelmann s.n. (MO, 122804). P. retusa Engelm., 1849. Engelmann s.n. P. retusa Engelm., 1849. Engelmann s.n. (NY). Obtained from Princeton in 1945. AR ool 6. P. retusa Engelm., 1849. Engelmann s.n. 7. P. retusa Engelm., Aug 1849. Englemann s.n. (GH). Texas. Engelmann grew plants from seeds sent to St. Louis and distributed specimens to Gu and ny. He distinguished P. retusa from P. oleracea by the former having broader retuse leaves, broader calyx, distinctly tuberculate pale seeds, and larger style with 3-4 short stigmas. He noted that the number of stamens was variable, up to a maximum of 15-17 and that the flowers opened earlier in the day than those of P. oleracea. With the difficulty in assessing herbarium specimens in this genus and with the morphological variability of P. oleracea as reported above, the floristic treatments for areas in which both species were reported have relied primarily on the tuberculate seed surface and 3—4 stigmas to distinguish P. retusa (Wooton and Standley, 1915; Wilson, 1932; Kearney and Peebles, 1942, 1951; Shreve and Wiggins, 1964; Harrington, 1964; Correll and Johnston, 1970; McDougall, 1973; Martin and Hutchins, 1980; Barkley, 1986; Henrickson, in prep.; Felger, in prep.). In an earlier study, Matthews and Levins (1986) published an SEM of the seed surface of P. retusa showing a tuberculate, S-un- dulate (raised stellate) pattern as typical for this species. This illustration is reproduced as Figure 3. In the preparation of this treatment for North America, we gathered 121 specimens labelled P. retusa. We wanted to compare these to the holotype, but found that Engelmann had failed to designate such and there have been no lectotypes designated. There are seven specimens which qual- ify for consideration as type material, Table 1. Sheet one has two plants. One is from the Engelmann collection of Aug 1849 with Engelmann’s personal label: “From Texas seeds, cultivated St. Louis.” Seeds are present on the specimen, which is noted on the sheet as no. 8871. The other plant on the sheet, with the no. 8872, was collected by Wright in Oct 1849 in western 176 Rhodora [Vol. 95 bs @i. * ge are Figure 7. Stellate (S-undulate) pattern with definite tubercles, labelled P. ole- acea. Figure 8. Root-tip squash showing 2” = 36 in P. oleracea. Texas. This plant is also P. retusa, but the point at which it was attached to this sheet is uncertain. Sheet 2 has several pieces of plants, including seeds, with Engelmann’s handwriting: “From Texas seeds, cultivated St. Louis.”’ Sheet 3 has one large top of a plant, with Engelmann’s handwriting: “From Texas seeds, cul- tivated St. Louis.” Seeds are present. Sheet 4 has one large top of a plant, with a label not written by Engelmann: “Texas seeds.” Seeds are present. Sheet 5 has one large top of a plant, has an Englemann handwritten label: “From Texas seeds, cultivated.” Originally the specimen was at Princeton University but was ac- quired by NY in 1945. There are no seeds on this specimen, which Legrand annotated as var. retusa in 1949. Sheet 6 is a mixed collection. At the top are two small, whole plants annotated by Legrand as P. lanceolata. Below are the tops of two plants of P. retusa, below which is Engelmann’s handwritten label: “From Texas seeds, cultivated.”’ There are two packets of seeds, one at the top and one at the bottom. Sheet 7 has the tops of several plants, has a personal label in Engelmann’s handwriting: “From Lindheimer’s Texas seeds, cultivated.’ Seeds are present. Sheets 1 and 7 are the most documentable specimens. Sheets 2 and 3 have no dates, although the date can be implied due to the se- quence of numbers. Sheet 4 does not have an Engelmann personal label. Sheet 5 does not have any seeds. Sheet 6 is a mixed col- lection. Since Engelmann’s work was in St. Louis, we designate 1993] Matthews et al.— Portulaca 177 Table 2. A comparison of the number of specimens showing the overlap of secondary morphological characteristics used to separate P. retusa from P. oler- acea. Seed Size Seed Size Specimens 0.9 mm Less Than Winged Unwinged Labeled or More 0.9 mm Sepals Sepals P. retusa 83 38 103 18 P. oleracea 58 222 Lis 169 the first specimen, numbered 8871, as the lectot STATES, from Texas seeds, cultivated St. Louis, Aug 1849, E i hat $.n., MO!, isolectotypes: Mo!, GH!, NY!. Notwithstanding this lack of a holotype, this suite of six spec- imens (one had no seeds) had the same seed surface character- istics. From our group of 121 specimens labelled P. retusa, 59 specimens had the combined raised stellate and tuberculate con- dition. The others showed various combinations of transitional morphologies, ranging from raised stellate with no tubercles to tuberculate without the stellate pattern. As we examined over 1300 specimens labelled P. oleracea, we found 280 specimens with seed surfaces that approached the concept of P. retusa. We looked at all specimens with tuberculate or stellate surfaces in an attempt to correlate seed size larger than 0.9 mm, retuse leaf apex, winged (carinate) sepals around the capsule, obtuse vs. acute sepal tips and no. of stigmas/styles, other morphological features that have been used as secondary identifying characteristics in the various treatments cited above. In this broad-leaf group of species, there are many specimens that were well prepared during drying, with packets containing leaves and fruits. We were able to com- pare the characters enumerated above with the seed surface pat- terns. The retuse leaf apex is completely unreliable. To illustrate the lack of consistency in the other two characteristics, Table 2 shows the comparison of these features. Many of these 280 spec- imens were identified originally as either P. oleracea or P. retusa and later annotated the reverse. We could find no consistency in the shape of the sepal tips nor a definitive number of stigmas/ Styles. Checking for stigmas/styles in dried portulacas is not very accurate however. Figures 4, 5, 6 and 7 illustrate some of the intermediate patterns of the seed surfaces, which bridge the gaps between Figures 1, 2 and 3 (Table 3). Figures | and 2 from the 178 Rhodora [Vol. 95 Table 3. Voucher data for representative specimens of Portulaca used for SEM of seed surfaces. Fig. 1. P. oleracea L. Louisiana: Rapides Parish, 19 Jul 1979, Pias 4345 U). Fig. 2. P. oleracea L. Alabama: Mobile Co., 26 Aug 1965, Deramus D732 U). Fig. 3. P. retusa Engelm. Texas: Williamson Co., 16 Aug 1961, Walker 73 X). Fig. 4. P. retusa Engelm. Arizona: Apache Co., 2 Sep 1984, Mohlenbrock 2434 U Fig. 5. P. retusa L. Arizona: Navajo Co., 1897, Zuch s.n. (US). Fig. 6. P. retusa Engelm. Arizona: Pinal Co., 8 Sep 1929, Kearney 6002 IZ ARIZ). Fig. 7. P. oleracea L. Texas: Garza Co., 29 Jul 1967, Hutchins 1354 (SMU). southeast show the granulate and obscurely stellate patterns re- spectively. Figure 3 is the typical “P. retusa” type. Figure 4 shows the beginning of the raised stellate pattern, with Figure 5 showing an increase in height. Figure 6 shows a tuberculate pattern emerg- ing from a granulate surface, while Figure 7 shows the beginning of a tuberculate pattern from a completely stellate (no granules), which then merges into Figure 3. Other transitional surfaces can be found, but these serve to show the clinal variation. Legrand (1962) specifically discussed the concept of P. retusa. He dismissed all of the morphological features (seed size, winged sepals and number of stamens) as being influenced by the envi- ronment and the season at the time of development. He also noted that there were t t 1 seed surface patterns between P. olera- cea and P. retusa, saying that he had seen these same forms in plants from the provinces of Salta and Cordoba in Argentina. We observed on specimens he annotated early in his study that he considered P. retusa to be a variety of P. oleracea. A specimen from NY was annotated in 1949 as var. retusa, but by the time he published his monograph in 1962 he had lost confidence in the seed surface characteristic and failed to recognize P. retusa as a variety. Since Engelmann originally cited the granitic region of west Texas as the type locality, we checked the 121 specimens with well defined stellate-tuberculate seed surfaces for habitat and geo- graphic distribution. There were no habitat correlations, the plants being associated with alluvial soil, clay meadows, rocky slopes, granite, rhyolite, limestone, gypsum and dolomite. Geographi- 1993] Matthews et al.— Portulaca 179 cally the specimens were distributed from central TX up through the panhandle of TX into OK, southern KS and CO and westward into southern NM and AZ, with transitional forms into the Mo- jave Desert of CA. Because of the lack of predictable and persistent morphological features, including seed surface, we conclude that P. retusa is a part of the polymorphic P. oleracea. CYTOLOGY Portulaca is characterized by a sequence of polyploids, partic- ularly in the more geographically widespread species (Matthews etal., 1992a, 1992b). Portulaca oleracea shows this sequence also, from a base of X = 9. There are diploid races (2n = 18) in Africa (Hagerup, 1932), Central and North America (Danin et al., 1978); tetraploid races (2m = 36) in India (Khullar and Dutta, 1973), Central and North America (Danin et al., 1978) and hexaploid races (2n = 54) in India (Khullar and Dutta, 1973; Sanjappa, 1978), Africa (Hagerup, 1932; Bouharmont, 1965; Boquar, 1986; Nyananyo and Okoli, 1987), in Europe (Walters, 1964; Bouhar- mont, 1965) and in North America (Cooper, 1935; Steiner, 1944; Heiser and Whittaker, 1948; Mulligan, 1961; Danin et al., 1978) and Hawaii (Kim and Carr, 1990). In addition, Sharma and Bhat- tacharyya (1956) reported 2n = 45 from India and Sugiura (1936) reported 2n = 52 from Japan. The North American counts of 2n = 54 have been scattered geographically from Wisconsin (Cooper, 1935), Virginia (Steiner, 1944), California (Heiser and Whittaker, 1948; Danin et al., 1978), Ontario (Mulligan, 1961). In addition, Danin et al. (1978) reported 2 = 36 from California. Danin and Anderson (1986) reported in their study of the subspecies of P. oleracea in Florida the occurrence of 2n = 18, 36 and 54. No counts were made for this study but numbers were projected based on the previous report by Danin et al. (1978) correlated with the identification of the subspecies. We wanted to determine the chromosome number from a va- Tiety of geographic areas to determine if a correlation existed between geography and cytology, when compared to the above data. We also wanted to publish a photograph of the chromosomes for a North American collection. Photographs of the chromo- somes of P. oleracea are scarce in the literature. Some early reports were accompanied by camera lucida drawings (Cooper, 1935; 180 Rhodora [Vol. 95 Steiner, 1944). There is a photograph of a meiotic figure by Kim and Carr (1990) from Hawaii. Danin et al. (1978) provided the most comprehensive study of the North American distribution, but published no photographs to document the counts. We can understand the reason for this. After a survey of 20 plants from CA, NM, IL, NC and SC involving over 250 root tip preparations we have been unable to achieve a high quality figure for photog- raphy. Baquar (1986) stated, ““The chromosomes in general are tiny, threadlike, forming clusters and are difficult to spread for karyological studies.” Our results are not different from these other investigators. Only two other papers have photographs of the chromosomes of P. oleracea; Khullar and Dutta (1973) from India and Nyananyo and Okoli (1987) from Africa. We were able to obtain 10 counts of 2” = 54 from IL. Ninety-one counts from CA, NM, NC and SC were 2n = 36. There appears to be a tendency for the hexaploid (2” = 54) to occur in the northern part of North America. However before any conclusions can be drawn, further counts from the United States and Canada would have to be made, and these should be compared to the counts from upper latitudes in Europe. Counts of 2” = 54 have been reported from Africa and India. The best figure we produced is shown in Figure 8, a 2n = 36 from SC. Voucher specimens are on file at UNCC. ACKNOWLEDGMENTS We wish to thank the McMillan Greenhouse and Conservatory at UNC Charlotte for the maintenance of live plants and S. Boyd (RSA) and M. Basinger (SIU) for collection of live material. Frank Machado, J. Arvey, S. Chapat and Dr. G. Windholz assisted in translation. LITERATURE CITED BAQUAR, ae R. 1986. Cytotaxonomic studies of the family Portulacaceae from Nigeria. La Kromosoma II 41: 1255-1262. ieee T. M. 1986. Flora of the great plains. Univ. Press of Kansas, Lawrence. BOUHARMONT, J. 1965. Note sur la cytologie de quelques especies de Portulaca. Bull. Soc. Roy. Bot. Belgique. 98: 175-188. Byrne, R. AND J. H. McANprews. 1975. Pre-Columbian purslane (Portulaca oleracea L.) in the new world. Nature 253: 726-727. CANDOLLE, A. P. DE. 1886. Origin of Cultivated Plants. 1959 reprint by Hafner, New York. CHAPMAN, J., R. B. STEWART AND R. A. YARNELL. 1974. Econ. Bot. 28: 411- 412. 1993] Matthews et al.— Portulaca 181 Cooper, D. C. 1935. Microsporogenesis and the development of the male ga- metophyte in Portulaca oleracea. Amer. J. Bot. 22: 453-457. CoQuILLANT, M. 1951. Sur les plantes les plus communes a la surface du globe. Bull. Mens. Soc. Linn. Lyon 20: 165-170. CorreLL, D. C. AND M. C. JoHNsTON. 1970. Manual of the Vascular Flora of Texas. Texas Research Foundation, Renner, TX. L. C. ANDERSON. 1986. Distribution of Portulaca oleracea L. (Portulucues!) subspecies in Florida. Sida 11: 318-324. , 1. BAKER AND H. G. BAKER. 1978. Cytogeography and taxonomy of the Portulaca oleracea L. complex. Israel J. Bot. 27: 177-211. FeLcer, R.S. In prep. Flora of the Northwestern Sonora, Mexico. GeesINK, R. 1969. An account of the genus Portulaca in Indo-Australia and the Pacific (Portulacaceae). Blumea 17: 274-301. Gorskg, S. F., A. M. RHODES AND H. J. Hoppen. 1979. A numerical taxonomic study of Portulaca oleracea L. Weed Sci. 27: 96-102. Gray, A. 1850. Plantae Lindheimerianae. Bost. J. Nat. Hist. VI: 141-240. —— AND J. H. baer 1883a. Am. J. Sci. 125(148): 241-255. Am. J. Sci. 126(152): 128-138. GUNTHER, R. T. 1959. it Greek Hebal of Dioscorides. Hafner, New York, N.Y. (reprint of 1933 ed.). HaGeRup, O. 1932. Uber Polyploidie in Beziehung zu Klima, Okologie und Phylogenie ign ag aus Timbuktu). Hereditas 16: 19-40. HARRINGTON, H. D. 1964. Manual of the Plants of Colorado. Sage Books, Denv ver. Henrickson, J. In Prep. Chihuahuan desert flora. HEtser, C. B. AND T. W. WHITTAKER. 1948. Chromosome number, polyplidy and growth habit of California weeds. Amer. J. Bot. 35: 179-186. KEARNEY, T. H. ANDR. H. PEeBLes. 1942. Flowering plants and ferns of Arizona. U.S.D.A. Misc. Pub. 423. U.S. Govt. Printing Off. AND 951. Arizona Flora. Univ. of California bie Berkeley. KHULLAR, S. P. AND M. Dutta. 1973. Cytotaxonomic studies on the genus nd a from Chandigarh (North India). Bangladesh J. Bot. >. 95-102. Kim, I. 1989. Biosystematics and ultrastructure of Portulaca in Hawaii. Ph.D. dissertation, University Microfilms, Ann Arbor, MI. ‘ARR. 1990a. Reproductive biology and uniform culture of Portulaca in Hawaii. Pacific Sci. 44: 123-129 . 1990b. Cytogenetics and hybridization of Portulaca in Ha- waii. Syst. Bot. 15: 370-377 EGRAND, . 1962. Las taciie Americanas de Portulaca. Anales del Museo de Historia Nat. de Montevideo II Mackenzie, K. K. AND B. F. Busu. 1902. New plant from Missouri. Trans. Acad. St. Louis XII: 81. Martin, W. C. AND C. R. Hutcuins. 1980. A Flora of New Mexico. J. Cramer, Vaduz, Liechtenstein. Mattuews, J. F. AND P. W. Levins. 1985. Portulaca egy p P. mundula I. M. Johnst. and P. parvula Gray in the southwest. Sida 11: 45-61. . 1986. The systematic significance of oe morphology in oa Asset g py. Syst. Bot. 11: J} 182 Rhodora [Vol. 95 AND D. W. KeTRON. 1991. Two new combinations in Portulaca (Por- tulacaceae). Castanea 56: 304-305. , AND S. F. ZANE. 1992a. Portulaca umbraticola Kunth (Portu- lacaceae) in ee United States. Castanea 57: 202-208. The reevaluation of Portulaca pilosa and P. mundula ee: Sida 1527 McDoucLALL, W. B. 1973. Seed plants of Northen Arizona. Museum of North- ern Arizona, Flagstaff. MryaAnisHI, K. AND P. B. Cavers. 1980. The biology of Canadian weeds. 40. Portulaca oleracea L. Canad. J. Plant. Sci. 60: 953-963. MUENSCHER, W. C. 1955. Weeds, 2nd ed. The MacMillan Co. New York, N.Y. Mutuican, G. A. 1961. Chromosome numbers of Canadian weeds. III. Canad. J. Bot. 39: 1057-1066. Autogamy, allogamy and pollination in some Canadian weeds. Canad. Fi Bot. 50: 1767-1771. NyAnanyo, L. AND B. E. Oxout. 1987. Cytological and morphological studies on Nigerian species of Portulaca (Portulacaceae) in relation to their taxon- omy. Fedd. Rep. 98: 583-587. Riptey, H. N. 1930. Dispersal of Plants throughout the World. L. Reeve and Co., Ashford, England. SANJAPPA, M. 1978. IOPB Chromosome number reports LXI. Taxon 27: 375- 392. SHARMA, A. K. AND N. K. BHATTACHARYYA. 1956. Cytogenetics of some mem- bers of Portulacaceae and peta families. eatyoes 8: 257-27 SHREVE, F. ANDI. L. Wicacins. 1964. d Flora of the Sonoran Desert. Stanford Univ. Press, Stanford. Stncu, J. S. AND K. P. SinGH. 1967. Contribution to the ecology of ten noxious weeds. J. Ind. Bot. Soc. 46: 440-451. SINGH, ee P. 1968. Thermoresponse of Portulaca oleracea seeds. Current Sci. 6-507. STACE, i 1989. Plant Taxonomy and Biosystematics, 2nd ed. Edward Arnold, London. STEINER, E. 1944. Cytogenetic studies on Talinum and Portulaca. Bot. Gaz. 105: 374-379. Stugssy, T. F. 1989. Plant Taxonomy. Columbia Univ. Press. New York, N.Y. Sucrura, T. A. 1936. A list of rue numbers in angiospermic plants. If. Proc. Imp. Acad. Tokyo 12 VENGRIS, J., S. DUNN ann Df Stce¥nce SorNCAES. 1972. Life history studies s related to the northeast. 7. common purslane. Mass. Agric. Exp. Stn. Bull. 598: 1-44. Wa ker, W. M. 1936. The dieg Mounds, Catahoula Parish, Louisiana. Bur. Am. Ethol. Bull. 113, 73 p Watters, S. M. 1964. Portulaca a In: T. G. Tutin, V. H. Heywood, N. A. Gurges, D. H. Valentine, S. M. Walters and D. A. Webb, Eds., Flora Eu- dge. Watson, P. J. 1969. The prehistory of Salts Cave, Kentucky. oe of inves- tigation, No. 16. Illinois State Museum, Springfield, IL. 86 p Witson, P. 1932. Portulaca. In: North American Flora, 21(4): 328-336. 1993] Matthews et al.— Portulaca 183 Wooton, E. O. AND P. C. STANDLEY. 1915. a - New Mexico. Contrib. U.S. Nat. Herb. 19. U.S. Govt. Printing Off., W ZIMMERMAN, C. A. 1976. Growth rege ania Btn of weediness in Portulaca oler- acea L. Ecology 57: 964-974. 19 A comparison of breeding systems and seed physiologies in three ates - Portulaca L. Ecology 58: 860-868. DEPARTMENT OF BIOLOGY UNIVERSITY OF NORTH CAROLINA CHARLOTTE, NC 28223 RHODORA, Vol. 95, No. 882, pp. 184-187, 1993 PORPHYRA AMPLISSIMA (KJELLMAN) SETCHELL ET HUS: NEW RECORDS OF AN “ARCTIC” SEAWEED IN SOUTHERN MAINE, NEW HAMPSHIRE AND NORTHERN MASSACHUSETTS! EDWARD J. HEHRE? AND ARTHUR C. MATHIESON The occurrence of Porphyra amplissima (Kjellman) Setchell et Hus in the southern portion of the Gulf of Maine represents a significant extension of its previously known range. Recent ac- counts limit this species to Norway and Spitzbergen in the Eastern Atlantic (South and Tittley, 1986) and to the Arctic, Maritime Provinces of Canada, and coastal Maine (Bird and McLachlan, 1992). Taylor (1957) included this species as Porphyra miniata (C. Agardh) C. Agardh var. amplissima (Kjellman) Rosenvinge, citing the arctic and Maine coasts as its range in North America. However, Taylor’s vague description of var. amplissima as a dis- tinct entity and the lack of any published references pertaining to the Maine records of this taxon have previously cast doubt as to its occurrence in this region. The recent account (Bird and McLachlan, 1992) of this species from the Atlantic Coast of Nova Scotia and lower Bay of Fundy has provided the impetus to search for its presence in the Gulf of Maine. The distribution of Porphyra amplissima in New England and the southern part of the Bay of Fundy is substantiated by an examination of P. miniata specimens deposited in the Farlow Herbarium, Harvard University (FH) and the Hodgdon Herbarium, University of New Hampshire (NHA). The majority of our collections of Porphyra miniata from Cam- pobello (Stone et al., 1970) and the Wolf Islands (Hehre et al., 1970), both of which are in the southern Bay of Fundy, are without a doubt P. amplissima. In addition, we have records of P. am- plissima from the nearshore open coast of Maine (from Cutler to York), New Hampshire, and as far south as Cape Ann, Massa- chusetts. We have also recorded P. amplissima from several off- ' Scientific Contribution Number 1815 from the New Hampshire Agricultural Experiment Station; also issued as Contribution Number 277 from the Jackson Estuarine Laboratory/Center of Marine Biology. ? Present address: 26 Park Street, South Berwick, Maine 03908. 184 1993] Hehre and Mathieson—Porphyra 185 shore islands in Maine (e.g., Boon, Hurricane, Smuttynose, and Two Bush) and within estuarine sites near North Lubec, Maine and Dover-Portsmouth, New Hampshire. The New Hampshire sites are within the outer reaches of the Great Bay Estuary System where strong tidal currents occur, simulating open coastal envi- ronments (cf. Mathieson et al., 1977, 1983; Reynolds and Ma- thieson, 1975). The estuarine tidal rapid at Dover Point (ap- proximately 8.5 miles inland) represents the innermost estuarine collection. The presence of P. amplissima from Pigeon Cove on Cape Ann (Massachusetts) is of interest for two reasons: this is the apparent southern limit of the taxon in the Western Atlantic; our collections include beautiful specimens collected by Mary Hitchcock, ca. 1900. Both Porphyra amplissima and P. miniata are primarily “arc- tic” species whose southern distributional limits reach northern Massachusetts. They are both annuals (April-November), with identical reproductive phenologies (May—November). The plicate nature of the thallus and the intermixed mosaic of carposporangial and spermatangial tissue on the margins of the blade serve to distinguish P. amplissima from P. miniata (cf. Bird and Mc- Lachlan, 1992). The latter criterion had been used previously to describe P. miniata (Taylor, 1957). Our sublittoral data for P. amplissima (i.e., 0-30 feet) agree with those of Bird and Mc- Lachlan (1992). The marine algal flora of New Hampshire has been studied in detail during the past 25 years (see Mathieson and Hehre, 1986). Based upon these studies we believe that Porphyra amplissima is an uncommon species that occurs sporadically in this part of its range. Porphyra miniata is definitely the more common of the two species in the southern Gulf of Maine, but P. amplissima becomes increasingly more prevalent as one goes east along the Maine coast towards the Bay of Fundy. Further studies are needed before we can ascertain the importance of P. amplissima to the marine algal flora of Maine. ACKNOWLEDGMENTS We would like to express our sincere appreciation to the per- sonnel of the Farlow Herbarium of Harvard University for their assistance and loan of herbarium specimens. 186 Rhodora [Vol. 95 REPRESENTATIVE SPECIMENS: CANADA: New Brunswick, South Wolf Island—R. A. Stone & E. J. Hehre, 4 June, 1968 (NHA 13,500); E. J. Hehre & J. R. Conway, 13 August, 1968 (NHA 14,051 & 14,052). Campobello Island—East Head Harbor, E. Hehre, 21 May, 1966 (NHA 1,258); Harbour de Loutre, R. A. Stone & E. J. Hehre, 14 June, 1967 (NHA 9,141); J. R. Conway, E. J. Hehre, R. A. Stone & R. A. Fralick, 18 May, 1968 (NHA 13,349); Herring Cove, E. J. Hehre, 20 May 1966 (NHA 4,626); J. R. Conway & E. J. Hehre, 11 August, 1968 (NHA 14,099); Little Duck Pond, R. A. Stone, E. J. Hehre and R. B. Pike, 5 June, 1968 (NHA 13,350); E. J. Hehre & J. R. Conway, 12 August, 1968 (NHA 14,149 & 14,150), Mulholland’s Bend, R. A. Stone & E. J. Hehre, 15 June, 1967 (NHA 9,261 & 9,270); E. J. Hehre & J. R. Conway, 9 August 1968 (NHA 14,125); Raccoon Point, E. J. Hehre & R. A. Stone, 16 June, 1967 (NHA 9,175); E. J. Hehre & J. R. Conway, 10 August, 1968 (NHA, 14,177); Whiterock Cliffs, R. A. Stone & E. J. Hehre, 17 June, 1967 (NHA 9,313). UNITED STATES. Maine (nearshore)— Pirates Creek, North Lubec, D. Lavoie, 21 May, 1966 (NHA 1243); Cutler, F. S. Collins, 6 July, 1902 (FH); East Machias, G. Vagenas, 21 May, 1966 (NHA 1,245); Long Sands, York, J. Kotheimer, 6 June, 1966 (NHA 1,255 & 1,256); (offshore islands)— Hurricane Island, Penobscot Bay, J. Hambrook, 29 May, 1986 (NHA 49,943); J. Hambrook (— 10 to —20 feet), 1 June, 1986 (NHA 49,935); Two Bush Island, near Hurricane Island, Penobscot Bay, J. Hambrook, 12 June, 1985 (NHA 49,164); Boon Island, York (—20 to —30 feet), 4. C. Mathieson, 30 June, 1976 (NHA 27,897 & 27,932), Smuttynose Island, Kittery, A. C. Mathieson, 19 November, 1976 (NHA 39,505). New Hampshire (nearshore)— Jaffrey Point (Fort Stark), New Castle, E. J. Hehre & A. C. Ma- thieson, 9 August, 1967 (NHA 10,321); Rye Ledge, Rye, J. Ship- man & J. Conway, 21 June, 1966 (NHA /,275); (estuarine)— Fort Constitution (Fort Point), New Castle (0.8 miles inland), J. R. Conway & J. Shipman, 22 June, 1966 (NHA 1,277); E. J. Hehre, 29 May, 1967 (NHA 9,035); E. J. Hehre, 23 June, 1967 (NHA 8,957); Bridge at Rte. 1 Bypass, Portsmouth, west bank (4.0 miles inland), E. J. Hehre, 29 May, 1967 (NHA 9,036); Dover Point (Hilton Park), Dover, junction of Little Bay and Piscataqua River (8.6 miles inland), E. J. Hehre, 23 June, 1967 (NHA 9,392); N. B. Reynolds, 15 May, 1969 (HA 15,491); N. B. Reynolds, 17 June, 1969 (NHA 15,492). Massachusetts (nearshore)—Pigeon Cove, Cape Ann, E. J. Hehre, 23 June, 1966 (NHA 1,269); M. Hitchcock, 1993] Hehre and Mathieson—Porphyra 187 ca. 1900 (NHA 1,244, 1,246, 1,247); Halibut Point, Cape Ann, /. M. Lamb and M. Zimmermann, 17 June, 1962 (FH, A-455); Na- hant, F. S. Collins, 24 June, 1882 (FH). LITERATURE CITED Birp, C. J. AND J. L. McLACHLAN. 1992. Seaweed Flora of the Maritimes 1. Rhodophyta— The Red Algae. Biopress Ltd., Bristol, England. Henre, E. J.,, J. R. CONWAY AND R. A. Stone. 1970. Flora of the Wolf Islands. Part III: the marine algae. Rhodora 72: 115-1 18. Martuigson, A. C. AND E. J. Henre. 1986. A synopsis of New Hamsphire Martuieson, A. C., C. NEEFUS Pee EMERICH PENNIMAN. 1983. Benthic ecology in an estuarine tidal rapid. Bot. Mar. 26: 213-230. Matuieson, A. C., E. Tveter, M. DALy AND J. Howarp. 1977. Marine algal ecology in a New a tidal rapid. Bot. Mar. 20: 277-290. REYNOLDs, N. B. AND A. C. MATHIESON. 1975. Seasonal occurrence and ecology of marine algae in a New Howrah tidal rapid. Rhodora 77: 512-533. Soutn, G. R. aANp I. Titttey. 1986. A Checklist and Distributional Index of the Benthic Algae of the North Atlantic Ocean. Spec. Publ. Huntsman Marine Lab. and British Museum (Natural Hist). Stone, R. A., E. J. Henre, J. R. Conway AND A. C. Matuieson. 1970. A preliminary checklist of the marine algae of Campobello Island, New Bruns- wick, Canada. Rhodora 72: 313-338. TayLor, W. R. 1957. The Marine Algae of the Northeastern Coast of North America. Univ. Michigan Press, Ann Arbor. DEPARTMENT OF PLANT BIOLOGY JACKSON ESTUARINE LABORATORY UNIVERSITY OF NEW HAMPSHIRE DURHAM, NH 03824 RHODORA, Vol. 95, No. 882, p. 188, 1993 NEW ENGLAND NOTE SOME NEW RECORDS FOR NEW ENGLAND STATES. III. Ray ANGELO These previously unreported records for New England states have come to the attention of the Club’s Curator of Vascular Plants in the course of executing his curatorial duties. The first and second such reports appeared in Rhodora 91: 342-344 (1989) and 92: 294-296 (1990). These records are intended to represent the first published ev- idence of these species in the state, not necessarily the first col- lection. The records cited here are for specimens that have been accessioned by the Club herbarium (NEBC). Their newness has been determined by making use of the most current floras, check- lists and other readily available published sources which are based upon public herbarium records. NEW RECORD FOR MAINE Glyceria acutiflora Torr. [Poaceae] Limington, ME (York Co.); Samuel N. F. Sanford, August 6, 1925; *“*Peqaket Lake. Jn shallow water. NEW RECORD FOR RHODE ISLAND Celastrus orbiculatus Thunb. [Celastraceae] Block Island, RI (Washington Co.); Robert Marks; August 31, 1974. [This specimen, originally identified as Celastrus scandens L., has been reidentified by R. Angelo. This species also col- lected from Warwick, RI (Kent Co.), David E. Boufford (#25287) & Ray Angelo (1990).] NEW ENGLAND BOTANICAL CLUB HERBARIUM 22 DIVINITY AVENUE CAMBRIDGE, MA 02138 188 RHODORA, Vol. 95, No. 882, pp. 189-190, 1993 A TAXONOMIC STUDY OF ARISTIDA STRICTA AND ARISTIDA BEYRICHIANA; ADDENDUM! ROBERT K. PEET I have discovered an additional isotype of Aristida beyrichiana Trinius & Ruprecht held at Mo (2874410). The label includes the specific date and location of the collection, information lacking on the labels of the other known type material. The specimen comes from the Bernhardi Herbarium (see D’Arcy, 1971) and is mounted together with a collection of Aristida stricta Michaux, apparently collected by Nuttall. The label includes the initials CB, and J. M. Greenman annotated the collection as that of Charles Beyrich. This Beyrich collection has the same degree of spikelet immaturity as the isotypes in L and us. The label reads "6/1 33,” which is consistent with the fact that Beyrich collected in the southeastern states only in 1833 (Sayre, 1975), and the June 1 date is consistent with the immaturity of the spikelets. The label indicates that the collection is from the area near Ebenezer. Pre- sumably this is the Ebenezer located in Effingham County, Geor- gla, approximately 25 miles north of Savannah. The isotype in L indicates the type collection is from Georgia. Ebenezer, while already a declining town in 1854 (Baldwin and Thomas), is one of the oldest villages in Georgia and, being on the main route of travel up the Savannah River, was visited by traveling botanists such as William Bartram (Harper, 1958). Aristida beyrichiana is known for Effingham County, Georgia (EHCV 571). There was in the early 1800’s another village of Ebenezer, Georgia in Morgan County (Baldwin and Thomas), but this location is out of the range of A. beyrichiana and is much less likely to have been visited by Beyrich ' Peet, R. K. 1993. A taxonomic study of Aristida stricta and A. beyriciana. Rhodora 95: 25-37. LITERATURE CITED BALDwin, T. AND J. THomas. 1854. New and Complete Gazetteer of the United States. Lippincott, Grambo & Co., Philadelphia, PA. D’Arcy, W.C. 1971. Mysteries and treasures in Berlandi’s Herbarium. Bulletin Missouri Botanical Garden 59: 20-25 Harper, F., Ed. 1958. The aig of William Bartram: Naturalist’s Edition. Yale Univ. Press, New Haven, CT. Sayre, G. 1975. Cryptogamae panes an annotated bibliography of exsic- 18 190 Rhodora [Vol. 95 catae of algae, lichens, hepaticae, and muscii. Mem. New York Bot. Gard. 19: 277-423. DEPARTMENT OF BIOLOGY, CB#3280 UNIVERSITY OF NORTH CAROLINA CHAPEL HILL, NC 27599-3280 USA RHODORA, Vol. 95, No. 882, p. 191, 1993 THIRD ANNUAL RICHARD AND MINNIE WINDLER AWARD WINNER ANNOUNCED On April 16th, at the annual meeting of the Association of Southeastern Biologists (ASB), the third annual Richard and Min- nie Windler Award was presented to Warren Frank Lamboy. Dr. Lamboy is a Research Associate in the germ plasm lab of the Plant Genetic Resources Unit of the New York State Agricultural Experiment Station, Cornell University in Geneva, NY. His pa- per, “The Taxonomic Status and Probable Origin of ASTER CHLOROLEPIS, a Southern Appalachian Endemic,” was published in the March 1992 issue of CASTANEA, the quarterly publication of the Southern Appalachian Botanical Society (SABS) In the research described in this paper, the author uses tradi- tional, phytogeographical, cytological and numerical techniques to clarify the taxonomic status and explore the relationship be- tween two species of the genus Aster, the endemic Aster chlorolepis and its close relative, Aster divaricatus L., a species ranging from Northern Alabama and Georgia to Southern Canada. Dr. Lam- boy’s data lead him to the conclusion that A. ch/orolepis is an autotetraploid derivative of A. divaricatus. The publication rep- resents a portion of a Ph.D. dissertation submitted to the Uni- versity of Illinois, Urbana. The Richard and Minnie Windler Award was established in April of 1990 at the annual meeting of the Southern Appalachian Botanical Society in Towson, Maryland, by Dr. Donald R. Wind- ler as a memorial to his parents. The award is presented annually to the author(s) of the best paper in systematic botany published in the previous year’s volume of CASTANEA. The eligible papers fall into the broad area of plant systematics, including floristic, experimental, revisionary, and nomenclatural studies. This year’s awardee received a framed certificate and a monetary award of $200, funded by an endowment administered by the Society. The winner is selected by a committee comprised of the three most recent systematics editors of the journal, after their terms as editor have expired. The present committee is chaired by Dr. David Whetstone of the University of Jacksonville. The other committee members are Dr. James Matthews of the University of North Carolina at Charlotte and Dr. Robert Haynes of the University of Alabama at Tuscaloosa. 191 1994 NEBC AWARD FOR THE SUPPORT OF BOTANICAL RESEARCH The New England Botanical Club will offer an award of $1,000 in support of botanical research to be conducted in relation to the New England Flora during 1993. This award is made to stim- ulate and encourage botanical research on the New England flora, and to make possible visits to the New England region by those who would not otherwise be able to do so. The award will be given to the graduate student submitting the best research pro- posal dealing with field studies in systematic botany, biosyste- matics, plant ecology, or plant conservation biology, although proposals for research in other areas of botany will also be con- sidered. This award is not limited to graduate students at New England institutions, nor to members of the New England Bo- tanical Club. Papers based on this research must acknowledge the NEBC’s support, and it is encouraged that they be submitted to Rhodora for possible publication subject to standard review pro- cesses. The 1993 Award recipient was Ms. R. Deborah Overath of the University of Georgia, in support of her work on the effects of apomikxis on genetic variation in Amelanchier laevis. Applicants must submit a proposal of no more than three dou- ble-spaced pages, a budget (the budget will not affect the amount of the award), a curriculum vitae, three copies of the application, and two letters in support of the proposed research, one from the student’s thesis advisor. Proposals and supporting letters must be received no later than March 1, 1994. The recipient of the award will be notified by April 30, 1994. Proposals should be sent to: AWARDS COMMITTEE, THE NEw ENGLAND BOTANICAL CLUuB, 22 Divinity AvE., CAMBRIDGE, MA 02138. 192 THE FLORA OF NEW ENGLAND A Manual for the Identification of All Vascular Plants including Ferns and Fern Allies Growing without Cultivation in New England by Frank Conkling Seymour illustrations by Andrea Robbins Second Edition 1982 Fourth Printing 1993 AVAILABLE FROM: Lubrecht and Cramer, Inc. 38 County Route 48 Foresturgh, New York 12777 193 THE NEW ENGLAND BOTANICAL CLUB 22 Divinity Avenue Cambridge, MA 02138 The New England Botanical Club is a non-profit organization that promotes the study of plants of North America, especially the flora of New England and adjacent areas. The Club holds regular meetings, has a large herbarium of New England plants, anda library. It publishes a quarterly journal, RHODORA, which is now in its 94th year and contains about 400 pages a volume. Membership is open to all persons interested in systematics and field botany. Annual d 35.00, including a to RHODORA. Members living within about 200 miles of Boston receive notices of the Club meetings. To join, please fill out this membership application and send with enclosed dues to the above address. Regular Member $35.00 Family Rate $45.00 For this calendar year For the next calendar year Name Address City & State Zip Special interests (optional): Vol. 95, No. 881, including pages 1-84, was issued May 4, 1993. 194 INFORMATION FOR CONTRIBUTORS TO RHODORA Submission of a manuscript implies it is not being considered for publication simultaneously elsewhere, either in whole or in part. Manuscripts should be submitted in triplicate (an original and two xerographic copies) and must be double-spaced (at least 3”) throughout including tables, figure legends, and literature cita- tions. The list of legends for figures and maps should be provided on a separate page. Footnotes should be used sparingly. Do not indicate the style of type through the use of capitals or under- scoring, particularly in the citation of specimens. Names of genera and species may be underlined to indicate italics in discussions. Specimen citations should be selected critically, especially for common species of broad distribution. Systematic revisions and similar papers should be prepared in the format of “A Monograph of the Genus Malvastrum,” S. R. Hill, Rhodora 84:1~-83, 159- 264, 317-409, 1982, particularly with reference to indentation of keys and synonyms. Designation of a new taxon should carry a Latin diagnosis (rather than a full Latin description), which sets forth succinctly just how the new taxon is distinguished from its congeners. Papers of a floristic nature should follow, as far as possible, the format of “Annotated List of the Ferns and Fern Allies of Arkansas,” W. Carl Taylor and Delzie Demaree, Rho- dora 81: 503-548, 1979. For bibliographic citations, refer to the Botanico-Periodicum-Huntianum (B-P-H, 1968) which provides standardized abbreviations for journals originating before 1966. All abbreviations in the text should be followed by a period, except those for standard units of measure and direction (compass points). For standard abbreviations and for guidance in other matters of biological writing style, consult the CBE Style Manual, 5th ed. (original title: Style Manual for Biological Journals). In preparing figures (maps, charts, drawings, photos, etc.) please remember that the printed plate will be 4 x 6 inches; be sure that illustrations are proportioned to reduce correctly, and indicate by blue pencil the intended limits of the figures. (Some “turn-page”’ figures with brief legends will be 312 x 6 in.) Magnification/reduction values given in text or figure legends should be calculated to reflect the actual printed size. An Abstract and a list of Key Words should be supplied at the beginning of each paper submitted, except for a very short article or note. All pages should be numbered in the upper right-hand corner. Brevity is urged for all submissions. 195 THE NEW ENGLAND BOTAN 2 Elected Officers and Council Men seh rT Pema ce mike President: Leslie J. Mehrhoff, Connecticut Geology ar ature ee : nee Satay i RHODORA April 1993 Vol. 95, No. 882 CONTENTS a 85 "i. there /Paul M. "Catling and Stuart G. Hay ee a, Vas ae Pp f Vaucheria ‘tribeghvcens, Chry- ir ira A. MacDonald, James F. Cahill, a ; a ee E. JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Che New England Botanical Club, Inc. 22 Divinity Avenue, Cambridge, Massachusetts 02138 RHODORA GORDON P. DEWOLF, JR., Editor-in-Chief Associate Editors DAVID S. BARRINGTON RICHARD A. FRALICK A. LINN BOGLE GERALD J. GASTONY DAVID E. BOUFFORD C. BAR UIST STOPHER S. CAMPBELL MICHAEL W. LEFOR WILLIAM D. COUNTRYMAN ROBERT T. WILCE TMASTER: Sora changes 1897, Faience, KS 66044. y / devoted Raat it to the _ of gf -gonAr ‘SHIPS: + Repl S35 Family $45. Applicatio form printed prior to eet eanse ER Sew Missour; BOTAN ICAL MAY 1 4 1994 TRhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 95 July/October 1993 No. 883/884 GARDEN LiBRaRy RHODORA, Vol. 95, No. 883/884, pp. 197-213, 1993 TAXONOMIC REVISION OF ARCTOMECON TORR. & FREM. DEANNA R. NELSON AND STANLEY L. WELSH ABSTRACT The genus Arctomecon Torr. & Frem. is reviewed, its three species, A. humilis Coville, A. californica Torr. & Frem., and A. merriamii Coville, are provided with a diagnostic key, described, and discussions of their peculiar habitat requirements and other pertinent information are included. Key Words: Arctomecon, A. humilis, A. californica, A. merriamii, habitats, ge- ography, Mohave Desert INTRODUCTION The genus Arctomecon Torr. & Frem. (Papaveraceae) is com- posed of three distinctive species: Arctomecon californica Torr. & Frem., A. humilis Coville, and A. merriamii Coville. All are gypsophilous species endemic to northern portions of the Mojave Desert regions of the southwestern United States, where their distributions are restricted to peculiar edaphic situations. Plants of all of the species are perennial herbs with long taproots, woody caudices, and tufts of basal leaves. Flowers are large and showy with petals white or yellow. All three species are considered to be rare, but only A. humilis is presently listed under stipulations of the Endangered Species Act of 1973 as amended (USFWS, 1979). Recent inventories of populations of A. californica and A. merriamii revealed losses in habitat, due primarily to develop- ment and urban expansion in the Las Vegas area, which has resulted in elimination of entire subpopulations. Need for pro- 197 198 Rhodora [Vol. 95 tection of all species in the genus under the Endangered Species Act is indicated. The generic name Arctomecon was regarded as neuter by Tor- rey, hence the name californicum. It is regarded here as feminine, following the example for Greek nouns with the “-on” termina- tion cited in Stern’s Botanical Latin (1983, p. 84), and the ter- mination of the specific epithets agree with that gender. When included in quotes the original form, as treated by Torrey and others, is followed. HISTORY The type of the genus Arctomecon (Greek arktos, of bear; and mekon, a poppy, from the fancied resemblance of the hairy leaves with the lobes armed with a long trichome to a bear’s foot), and its initial species californica (of Alta California, then a portion of Mexico), was collected by John Charles Fremont on his second expedition of exploration of the American West. On 3 May 1844 they reached the springs at ‘“‘/as Vegas [The Meadows]—a term which the Spaniards use to signify fertile or marshy plains, in contradistinction to //anos, which they apply to dry and sterile plains. Two narrow streams of clear water, four or five feet deep gush suddenly, with a quick current, from two singularly large springs; these and other waters of the basin, pass out in a gap to the eastward. The taste of the water is good, but rather too warm to be agreeable; the temperature being 71° in the one, and 73° in the other. They, however, afforded a delightful bathing place” (Fremont, 1845, p. 266). There he collected the plant that became the type specimen “‘on the banks of a creek.” The water from the two springs coalesced into a stream—the only one in the vicinity (Welsh, 1993). In 1874, Charles Christopher Parry collected, near St. George, Utah, a “showy Papaveraceous plant with nodding white flowers” (Parry, 1875). Parry identified this plant as A. californica, but recognized that it differed from Fremont’s earlier, solitary collec- tion, in its “less hairy leaves, four (not six) valved capsule and more caespitose habit.” Parry forwarded a portion of his collec- tion (he took plants in sets for later distribution through exchange or sale) to Dr. Asa Gray at Harvard University. Dr. Gray also identified the Parry materials as A. californica (Gray, 1877). Nei- ther Parry nor Gray had noted that the Fremont materials were 1993] Nelson and Welsh—Arctomecon 199 described as having yellow petals. Parry’s materials were noted as distinctive by Coville (1892) who published them as belonging to a newly erected species, A. humilis Coville, but not until two years following the death of Parry (Welsh, 1988). In that same publication (Coville, 1892) a second species with white flowers, A, merriamii Coville, was proposed for the genus. It was based on specimens taken by Vernon Bailey west of Vegas Ranch in Lincoln County, Nevada in 1891 (Mozingo and Williams, 1980). Thus, the genus, as it is currently recognized, was fully constituted by 1892. TAXONOMIC AFFILIATIONS Arctomecon differs from other papaveraceous genera present in the Mojave Desert (Argemone, Platystemon) in general habit and gross morphologic features. Its species are the only members of the poppy family, in North America at least, recognized as gypsum restricted. The three species share ample characters to readily demonstrate their affinity, while the differences between the species are substantial enough to provide clear taxonomic units. All three species can be readily recognized by their foliage: the dense rosette of relatively long, broad, green leaves appears very “extravagant” in comparison to that of most plants of the warm desert regions. Affinities of the genus to others of the Amer- ican southwest have not been resolved. The species differ morphologically from each other on the basis of arrangement of flowers in the inflorescence, flower color, status of the petals following anthesis, number of locules per ovary, Outline of apical margins of the leaf, amount and kind of foliar vestiture, and the overall stature of the plant. Distinct morpho- logical differences, when taken with the geographic isolation of the species from each other, suggest that the separation of the taxa was not a recent event. Raynie et al. (1990) report alkaloid content of the three Arc- tomecon species. Total number of alkaloids was greatest for A. humilis, with 14 unique alkaloids. Eight were detected in the tissues of A. californica and five in A. merriamii. Three alkaloids were found to be common to all species of the genus. Allocryp- topine and protopine were major alkaloids in all species, while 12-methoxyallocryptopine (a new protopine-type alkaloid isolat- ed during the study and presently known only from Arctomecon) 200 Rhodora [Vol. 95 was found in lesser quantities. It is uncertain what clues alkaloid composition can provide to understanding the evolutionary re- lationships between this genus and other Papaveraceae, and be- tween the species in the genus. The larger numer of alkaloids in A. humilis may suggest that the species is ancestral to the others (Raynie et al., 1990). To date, the chemical and morphological criteria have not provided evidence of a close relationship of this genus to others in the family. DISTRIBUTION All species of Arctomecon are endemic to the northern margin of the Mojave Desert regions of the southwestern United States (Figure 1). The range of A. humilis is the most restricted of the three. It is presently known to be limited to an archipelago of less than a dozen sites, all within a few miles of St. George, Utah. Plants are strictly limited to gypsic outcrops of the Moenkopi Formation, primarily to the Shnabkaib Member, but occasionally extending downward to the Middle Red Member and upward to the Upper Red Member. The low, rolling terrain formed by these gypsum outcrops makes these areas popular for off-road vehicle use. All accessible sites are heavily impacted by off-road vehicular traffic which rapidly destroys cryptogamic soil crusts and com- pacts the substrate. It is noted that Kearney and Peebles (1960) cite A. humilis from northern Arizona, and an equivocal specimen cited from there is present at Bry, but probably it came from Utah. Recent investigations, including those by the authors, have not located this species in Arizona, although it has been found within one km of the Utah-Arizona boundary. The ranges of A. californica and A. merriamii appear to overlap in the Las Vegas area, but the species do not occur together, probably due to differences in habitat requirements. In the Las Vegas area A. californica occurs on Quaternary Alluvium, often on ancient gypsic playa remnants. To the east, near Lake Mead, it is on other alluvium, but also on the Moenkopi Formation, and, in adjacent Arizona, it evidently occurs on gypsic limestones. In comparison, A. merriamii seems to prefer the gypsic limestones and gravels, and mainly occurs in numerous, small, disjunct sub- populations. Many habitats that appear to be habitable by the species of this genus do not support any of the plants. The reasons for the paucity of the plants are unknown. Several known sites Figure 1. Habit sketch of Arctomecon humilis Coville. of former habitation of both A. californica and A. humilis have been obliterated by construction, with pavement and structures now occupying the sites. It is believed that some other populations may have recently disappeared, especially in the Las Vegas area. The California bearclaw poppy occurred as recently as the mid 202 Rhodora [Vol. 95 1980’s in a vacant lot in a subdivision along Craig Road in Las Vegas, growing on an evidently saline playa remnant. The range of Arctomecon merriamii extends from Death Valley on the west to the Spring Mountains of Nevada on the east. The species has been reported from approximately 20 sites within Death Valley National Monument, and from three or four moun- tain ranges in Nevada and near Ash Meadows, Nevada. One population is known from within the Las Vegas, Nevada, city limits. This population has been closely monitored by the authors and presently includes 15 individuals. The site is heavily impacted by construction on adjacent lots and its future is in doubt. ECOLOGY All three species of Arctomecon have been observed on heavily gypsic substrates. Since all records of occurrence of A. humilis and A. californica have been from such substrates, it seems rea- ms to assume that this affinity to gypsum is obligate for these species. The situation with A. merriamii is not so clear. It is known from both definitely gypsic sites and from limestones where the presence of gypsum has not been demonstrated. Thus, it might be considered a facultative gypsophile. Gypsum content of soils at four sites occupied by A. humilis was determined to be from 27 to 51 percent by weight (Nelson, 1989) and at five A. californica sites ranged from 36 to 69 percent (Meyer, 1987). Gypsum content of soils at four sites occupied by A. merriamii was determined to be from 2 to 8 percent by weight. The soils on sites occupied by A. humilis and A. californica tend to have very low bulk densities due to the presence of sponge gypsum and often have a heavy cover of cryptogams (Meyer, 1987; Nelson, 1989). It has been suggested that the unique chem- istry of gypsic sites, especially the high sulfur content, is a primary factor responsible for the evolution of gypsophile endemics (Par- sons, 1976). Meyer (1986) found gypsophiles to occur only where gypsum content was sufficiently great to effect physical properties of soil, while much smaller amounts of gypsum would still create high sulfate levels. Because of their apparent obligate affiliation with gypsic soils, both A. californica and A. humilis are not found growing in close association with the creosote bush-dominated desert shrub com- 1993] Nelson and Welsh—Arctomecon 203 munity typical of the Mojave Desert (Meyer, 1987; Nelson, 1989). They are, instead, found with a complement of other gypsum- tolerant species, including shadscale [Atriplex confertifolia (Torr. & Frem.) Wats.], burrobush (Hymenoclea salsola T. & G.), indigo bush [Psorothamnus fremontii (Torr.) Barneby], ephedra (Ephed- ra torreyana Wats.), and Anderson wolfberry (Lycium andersonii Gray). Plant species associated with A. merriamii are much more diverse. It has not only the widest geographic range, and the broadest tolerance with regard to substrate, but it also has the broadest elevational tolerance (610 to 1710 m). The elevational range of A. humilis is 750 to 1070 m, and that of A. californica is between 550 and 950 m. A solitary record of A. californica is known from Utah, from the Moenkopi Formation, on the Coal- pits Wash road, where it might have been recently introduced. All of the species flower in April and May, with A. humilis reaching the peak of flowering in early May, somewhat later than for the other species. Pollinating agents observed at flowers of low bearclaw poppy include both honey bees (Apis mellifera) and native ground nesting bees. Experimental hand pollinations have shown that this species relies almost exclusively on outcrossing with neighboring plants (Nelson, 1989). Information on polli- nation ecology of the other species is lacking. Large plants of Arctomecon humilis may produce as many as 400 flowers, of which 70-75 percent are expected to mature fruit. Capsules generally produce at least 23 fully developed seeds. Pet- als of this species are persistent, surrounding the developing cap- sule as they dry. The mature capsule abscisses from the pedicel several millimeters below the mature ovary and the fruit enfolded by marcescent petals falls free. Some seeds, those falling from capsules still attached to the plant, are deposited among the leaves of the rosette and elsewhere near the plant base. Other seeds are scattered for some distance from the plant when the fruit abscisses and is tumbled across the land surface, propelled by the wind against its petal “wings.” Studies of populations of A. californica (Meyer, 1987) and A. humilis (Nelson, 1989) demonstrate drastic year-to-year fluctu- ations in population density. Data suggest that, for both species, such volatile cycles are primarily the result of a relatively short lifespan (individuals surviving the first year may be expected to live an average of four to five years) coupled with irregular oc- currence of germination events. Meyer (1987) found that years 204 Rhodora [Vol. 95 with sufficient winter rainfall to assure significant establishment of A. californica seedlings may occur as much as nine years apart. During a four year study of populations of A. humilis only a single germination event was observed (Nelson, 1989). Both species appear to rely on high reproductive output and the maintenance of a large reservoir of long-lived seed in the soil in order to persist. Separation of organic material from soil collected near A. humilis plants, using a flotation technique, revealed a large quantity of seed “entrapped” within cryptogamic soil crusts. This suggests the importance of protecting these soil surfaces in order to main- tain soil seedbanks critical to the survival of populations. Dy- namics of population densities in A. merriamii have not been documented. TAXONOMY Arctomecon Torr. & Frem. in Frem. Rep. Expl. Exped. Oreg. & N. Calif. 312. 1845. TYPE SPEciEs: Arctomecon californica Torr. & Frem. in Frem. Perennial herbs from a subligneous caudex and long taproot, subacaulescent, caulescent, or subscapose; leaves mostly in a basal rosette, some typically cauline, usually cuneate, long-hirsute, mostly toothed apically and the teeth spinulose-tipped; flowers large and showy, solitary, or borne in leafy-bracteate cymes, nod- ding in bud, becoming erect at anthesis; sepals 2 or 3, caducous; petals 4 or 6, white or yellow; anthers basifixed, oblong, opening longitudinally; pistil 1, the carpels 4 or 5 (6), the styles obsolete or united and less than 1.5 mm long, the stigmas connate, with 4-6 lobes; ovary unilocular; capsules acrocidal along 4—6 sutures; seeds strophiolate. 1. Flowers yellow, (5) 6-20, in a paniculiform cyme; plants sub- acaulescent, of Clark County, Nevada, and Mohave County, Arizona (apparently introduced in eer oe County, LIRR ks Palo bo Mae ee it ate A. californica 1. Flowers white, solitary or 2—5 (7); plants variously distributed I ee ee en aed eed Ee are eee uke TA GAR el eee ga 2 2. Petals 4; plants scapose; flowers 2-5 in a paniculiform cyme; plants of Washington County, Utah ....... A. humilis 2. Petals 6; plants subacaulescent; flowers 1 or 2 (3-7) in a greatly shortened cyme, the elongated pedicels scapose; 1993] Nelson and Welsh—Arctomecon 205 plants of Death Valley, California, and Nye, Clark, and Lincoln counties, Nevada .............. A. merriamili Arctomecon humilis Coville Proc. Biol. Soc. Wash. 7: 67. 1892. (Figure 1) Low or dwarf bearclaw poppy. Type locality: “In 1874 Dr. C. C. Parry collected in the vicinity of St. George, Utah, an Arctomecon, which Dr. Gray referred to A. californicum. The material now at hand shows that it is distinct both from the original plant of Fremont and from the species just described [i.e., A. merriamii]. It differs from the former in its smaller size throughout, less hairy leaves, fewer flower parts, white petals, dilated filaments, and the presence of a style; from A. merriamii in its smaller size and more scanty hairs, more than 1-flowered peduncles, fewer flower parts, persistent petals, and obovate, several times shorter than the capsule” (Coville, 1892). Type: “Flora of Southern Utah, &c. Arctomecon humile, Co- ville, non Arctomecon Californicum Torr. & Fremont, C. C. Parry 6, 1874” (HOLOTYPE GH; IsoTYPE NyY!; frag. Us!). Plants forming rounded clumps, 1.5—2.5 dm tall, from a branch- ing caudex, this clothed with brown to ash colored, marcescent leaf bases, caulescent to apparently subscapose; stems glabrous to scabrous with crisped trichomes; basal leaves 0.5-8 (8.5) cm long, 4-16 mm wide, cuneate to obovate, usually 3- or 4-toothed or -lobed apically, glaucous and hirtellous as well as sparingly long pilose; cauline leaves typically the shorter and wider, cuneate to narrowly obovate, commonly deeply 3-lobed apically and each with an acerose bristle 1-5 mm long; cyme subcorymbose to paniculiform, bracteate; flowers mainly 3-5, on pedicels 2-9 cm long; sepals 2-8 mm long, usually 2, glabrous; petals 4—6, white, 1.5—4 cm long, usually as wide as long, oval to obovate, persistent on and falling with the mature fruit; stamens surpassed by the stigma; anthers 2 mm long, the filaments somewhat dilated; car- pels 3 or 5, astylous or the style to | mm long; capsules obovoid, 8-13 mm high, 6-9 mm wide, acrocidal along the sutures, splitting Ys to 2 the height of the capsule; fruit soon abscissing following dehiscence; seeds shiny black, 2.5-3 mm long, ca. 1.2 mm thick, conspicuously arilate. This plant is endemic on gypsiferous outcrops of the Moenkopi Formation (primarily the Shnabkaib Member, but also the Middle Red, and Upper Red members) in Washington County, Utah. 206 Rhodora [Vol. 95 The mature capsules, with marcescent petals attached, fall from the plant soon after maturity and serve to aid dispersal by being carried by wind. As irreplaceable portions of our natural heritage all members of the genus should be regarded a national prize, as jewels of great price, and protected for future generations, whose advocacy this generation must represent. The low bearclaw poppy (A. Aumilis) is being impacted by off-road vehicles and by urban sprawl in the vicinity of St. George. It is legitimately cited as endangered under stipulations of the Endangered Species Act of 1973, as amended. Despite that listing and attempts by concerned federal and state agencies, and many private individuals, the habitat is being sys- tematically impacted by off-road traffic. Only prudent preserva- tion of the habitat will guarantee survival of this species into the future. Representative specimens: United States. Utah, Washington County, St. George, 23 June 1937, W. P. Cottam 7238 (BRY!); W side Warner Ridge, 11 April 1983, E. Neese 12896 (BRyY!); 1.5 km E of Bloomington, | May 1981, S. L. Welsh 20388 (BRyY!); ca. 5 km S of St. George, 3 May 1970, S. L. Welsh & N. D. Atwood 9695 (BRyY!); S of St. George, 18 May 1971, L. C. Higgins 4210 (BRY!); Bloomington, Price Hills, 26 May 1983, L. C. Higgins & B. Welsh 13423 (BryY!). Arctomecon californica Torr. & Frem. in Frem. Rep. Expl. Exped. Oreg. & N. Calif. 312. 1845. (Figure 2) California bearclaw poppy. Type locality: ““This remarkable plant was found in only a single station in the Californian mountains, on the banks of a creek; flowering early in May. The soil was sterile and gravelly. Although very near Papaver, it differs so much in habit and in the strophio- late seeds, as well as in other characters that it must be a distinct genus” (Fremont, 1845). Type: “Fremont’s Expedition to California, 1843-4. No. 429, May 3, 1844” (Ny!) (l.c.). Plants forming tall to rounded clumps, subscapose, 2.5-6 dm tall, from a branching caudex, this clothed with ash to straw- colored marcescent leaf bases; basal leaves, 5-14 cm long, 3-25 mm wide, cuneate, usually with 3-5 apical teeth each with an acerose bristle 4-10 (12) mm long, hirtellous as well as moderately N \ Y aN Wy a em \, WW ul! \ Uy RAY ALN Wa ha Wr) Me y | Wage Ne AN v i fe fF —= ae a Ne a ib Soa ce > mri | or ee ISS VA = - Te ce wn Nii Z \ Figure 2. Habit sketch of Arctomecon californica Torr. & Frem. 208 Rhodora [Vol. 95 long pilose with barbellate trichomes to ca. 10 mm long; stems sparsely pilose and scabrous, glabrous above; pedicels 7-17 cm long; cymes paniculiform, 5- to 20-flowered, 10-30 cm long, with only | or 2 foliose bracts, these long pilose; sepals 3, 8-15 mm long, glabrous or sparingly pilose; petals 4 or 6, yellow, 1.5-3.2 cm long, slightly longer than wide; stamens equaling or slightly surpassing the stigma; anthers 2-3 mm long, the filaments fili- form; carpels 4 or 5, astylous; capsules ovoid to obconic, 12-23 mm long, 7-10 mm wide, acrocidal along 4 or 5 sutures, splitting not more than '4 the length of the capsules; seeds shiny black, 2.5-3 mm long, ca. 1.2 mm thick, conspicuously arilate. The plants occur in gypsic substrates in Clark County, Nevada, and also in Mohave County, Arizona, east of Lake Mead. The populations fluctuate numerically, and often dramatically, in re- sponse to rainfall patterns (Meyer, 1987), “disappearing” from sites for several years and reappearing en masse (Welsh et al., 1987). The type of this specimen (LECTOTYPE NY!) consists of a drawing and a fragment envelope containing an immature fruit and por- tions of a second one. The data on the label are as follows: “‘Fre- mont’s Expedition to California, 1843-4. No. 429, May 3, 1844. Arctomecon Californicum. Torr. ‘Specimen lost by Nick Pooley— or the Lithographer. It was the only one collected by Fremont! Found ata place called Las Vegas, near the Rio Virgen in Southern Utah—Lat. 36°10’.”” The drawing has the notation, “(original drawing) by Nick Pooley—Arctomecon californicum.” It is evi- dent that Mr. Pooley was one of several artists occasionally em- ployed by Dr. John Torrey in the 1840’s to provide illustrations for publication. The surviving illustration shows a plant remark- able for its completeness and beauty, a fact not often in evidence in numerous other Fremont flood-damaged collections from the troubled second expedition (Welsh, 1993). Pooley’s illustration will serve as frontispiece in the published version of the work by Welsh (1993) to be published later. The note in Torrey’s hand demonstrates a degree of frustration that the specimen was lost. The epithet californica denotes the historical designation of the area where the plant was collected, i.e., Alta Californica, then a portion of Mexico (Welsh, 1993). Alta California included not only present California, but Nevada, Utah, Arizona, and portions of Colorado and New Mexico as well. The epithet was well chosen. Representative specimens: United States. Arizona, vicinity of 1993] Nelson and Welsh—Arctomecon 209 Lake Meade, 25 Mar. 1981, R. K. Gerrisch 4816 (BRy!). ne Clark County, 2 mi. W of Las Vegas, 19 April 1938, J. Co B. Clokey 8347 (sry!); 2 mi. S of old town of St. Ede W shore of Lake Meade, 4 May 1941, V. M. Tanner s.n. (BRyY!); do, North of Las Vegas on Hwy 91-93, milepost 38, 4 April 1970, N. D. Atwood 2387 (BRy!); do, ca. 2 mi. NE of Frenchmans Mt. and ca. 4 mi. NE of Las Vegas, 5 April 1981, J. C. Stimson 439 (BRY!); do, ca. 18 mi. S of Overton, 21 April 1982, N. D. Atwood & S. L. Welsh 8605 (Bry!); do, Jct of Cheyenne and Lamb Blvds, Las Vegas, 27 April 1978, D. Williams s.n. (BRyY!); do, North Las Vegas, Craig Rd., 7 April 1983, S. L. & S. L. Welsh 21539 (Bry!); do, North-bound lane of I-15, near Nellis Air Force Base, 30 April 1988, D. Nelson & K. T. Harper 305 (sry!). Utah, Wash- ington County, Clint Isom Ranch on Coalpits Mesa road, 20 June 1986, R. B. Warrick 2039 (BRy!). Arctomecon merriamii Coville Proc. Biol. Soc. Wash. 7: 66. 1892. (Figure 3) Merriam bearclaw poppy. Type locality: “Type specimen in the United States National Herbarium, No. 1890, Death Valley Expedition; collected May 1, 1891, a few miles west of Vegas Ranch, Lincoln County, Ne- vada, by C. Hart Merriam and Vernon Bailey” (I.c.). Type: “Vegas Valley, Lincoln Co., Nevada, between Cotton- wood Cr. & Vegas ranch, Vernon Bailey 1890, May 1, 1891” (HoLotyPe us!; IsoryPe us!) Plants clump-forming, 1-4 dm tall, subacaulescent, from a cau- dex clothed with ash-colored marcescent leaf bases; stems very short; leaves clustered at the base, 3.5-12 cm long, cuneate to obovate, the base cuneate-attenuate, with 3-12 apical teeth (or sometimes merely crenate), each tooth with an acerose bristle 5—- 12 mm long, moderately to densely long pilose, especially dense near the apex, the trichomes 1-12 mm long, barbellate; inflores- cence axis much shortened, cymose, with 1-6 scapose pedicels per stem, these (7) 15-36 cm long; sepals 2-3, 14-20 mm long, pubescent like to the leaves; petals 6, white, 2.5-4 cm long, 2-3 cm wide; anthers 4-5 mm long, the filaments somewhat dilated above; carpels usually 6, astylous; capsule linear-oblong, 2.5-4 cm long, 7-11 mm wide, dehiscing 4 the length of the capsule along 6 sutures. 5. Figure 3. Habit sketch of Arctomecon merriamii Coville. 1993] Nelson and Welsh—Arctomecon 211 UTAH NEVADA NAT'L RECREATION AREA CALIFORNIA A.californica & ARIZONA A. humilis 7 A. merriamii A Figure 4. Distribution map for species of Arctomecon in southern Nevada and adjacent portions of California and Utah. This plant grows on gypsic and limestone substrates in Death Valley (Inyo Co., California) and eastward into Nye. Clark, and southern Lincoln counties, Nevada. The label of the type specimen bears only the name of Vernon 212 Rhodora [Vol. 95 Bailey as collector, but the type is attributed to Merriam, party director, and Bailey. Is it probable that Dr. C. Hart Merriam had no hand in its collection? And, if so, why was it not named for Bailey? There is a hint of playing politics, or of personal distrust or dislike, in the selection of the specific epithet. Coville should have honored collector Bailey with the name. Representative specimens: United States. California, Inyo County, Racetrack Valley road, ca. 18 km S of Ubehebe crater, 1 April 1982, Thorne et al. 1595, 1 April 1982 (Bry!); do, Funeral Mts, road to Chloride, ca. 8 km from Beatty road, 6 April 1983, K. Thorne & B. Ratcliffe 2342 (Bry). Nevada, Clark County, South of Indian Springs, 26 April 1939, J. W. Clokey 8238 (BRy!). Lin- coln County, Rainbow Canyon road, 10 km E of Hwy 93, 22 April 1988, D. Nelson 284 (Bry!); do, 10 km E of Hwy 73, along Caliente road, 4 September 1981, E. Neese & S. Clark 10994 (BRY!); do, Meadow Valley Mts, ca. 23 mi. SW of Elgin, 19 April 1982, N. D. Atwood & S. L. Welsh 8553 (Bry!). ACKNOWLEDGMENTS The authors wish to express their appreciation to Kaye H. Thorne who provided the illustrations, and to Dr. Kimbal Harper and Dr. Susan Meyer who offered help and encouragement during this work. LITERATURE CITED Covitie, F. V. 1892. Descriptions of new plants from southern California, Ne- vada, Utah, and Arizona. Proc. Biol. Soc. Wash. 7: 65-80. CronaulisT, A. 1981. An Integrated System of Classification of Flowering Plants. Columbia University Press, New York. 1262 FREMONT, J.C. 1845. Report of the exploring eaainion to Oregon and North California in the years 1843-44. Senate Document 174. Gales and Seaton, Washington. 1887. Memoirs ‘ My Life, Vol. 1. Belford, Clark, and Co., Chicago mud New York. 655 Gray, A. 1877. pie mn to the botany of North America. Proc. Amer. Acad. 12: 52-53. T.H R. H. Peestes. 1960. Arizona Flora. University of Cali- fornia Press, Berkeley. 1085 pp. Meyer, S.E. 1986. The ecology of gypsophile endemism in the eastern Mohave Desert. ape 67: 1303-1313. ——. 1987. Life history of Arctomecon california, a Mohave Desert endemic perennial herb. Unpublished manuscript, U.S. Forest Service, Provo, Utah. 1993] Nelson and Welsh— Arctomecon Z13 Mozinco, H. N. AND M. WiiuiAMs. 1980. Threatened and endangered plants of Nevis U.S. Fish and Wildlife Service and Bureau of Land Management. 268 p NELSON, S R 1989. Site characteristics and habitat requirements of the en- dangered dwarf bear-claw poppy (Arctomecon humilis Coville, Papaveraceae) and demographic and seed bank biology of Arctomecon humilis (Papavera- ceae), a short-lived perennial. Unpublished M.S. thesis, Brigham Young University, Provo, h PARRY. C.'C. 1875. Botanical observations in southern Utah—II. Amer. Nat- sraist 96: 139-146. Parsons, R. F. 1976. Gypsophily in plants—a review. Amer. Midl. Naturalist 96: 1-20. Raynig, D. E., M. L. Lee, D. R. NELson, K. T. HARPER, E. W. MEAD AND F. R. STERMITz. 1990. Alkaloids of A ). 12-Meth- ollocryptopine, a new protopine-type alkaloid. Biochem. Sys. and Ecol. 18: 45-4 STERN, W. T. 1983. Botanical Latin, 3rd ed., Revised. David & Charles, London. 566 pp. U.S. FisH AND WILDLIFE SERVICE. 1979. Determination that Arctomecon humilis is an endangered species. Federal Register 44(216): 64250-64252 WE su, S. L. 1988. Utah botanical explorer Charles Christopher Parry. Great Basin Nat. 48: 9-1 1993. Botanical contributions of John Charles a . apie eanil inamuscrip, Herbarium of Brigham Young University. 6 ——.,N. D. Atwoop, S. GoopRICH AND L. C. HIGGINs. 1987. "i Utah flora. Mem. Great Basin Nat. 9: 1-894. DEPARTMENT OF BOTANY AND RANGE SCIENCE AND LIFE SCIENCE MUSEUM BRIGHAM YOUNG UNIVERSITY PROVO, UT 84602 RHODORA, Vol. 95, No. 883/884, pp. 214-224, 1993 ACHENE MICRO-MORPHOLOGY AS A SYSTEMATIC AID TO THE SERIES PLACEMENT OF SVENSON’S UNDESIGNATED ELEOCHARIS (CYPERACEAE) SPECIES FRANCIS J. MENAPACE! ABSTRACT Eleocharis species unassigned by Svenson to the series level at the completion of his worldwide baieiiasiaeiis hie! examined utilizing scanning signs iia Py periclinal cell revealed mictoumaiolieieal hacen) in support of assigning E. squamigera and E. subarticulata to the series Palustriformes, subseries Truncatae. Although the results failed to delegate unequivocally E. albida and E. eee addi- btained. An assessment of E. minarum proved to be impossible due to the. absence es mature achenes. Key Words: Cyperaceae, Eleocharis, micro-morphology, SEM, achene INTRODUCTION The last monographic study of Eleocharis R. Br., on a cos- mopolitan basis, was conducted by Svenson (1939a). Svenson (1929, 1939a) partitioned Eleocharis at the series and subseries levels, based on macro-morphological characters of the culm, floral scales, and fruit, with the achenes being of particular im- portance. Of the 146 species recognized by Svenson (1939a), five were left unassigned to the series level (E. albida Torr., E. melano- carpa Torr., E. minarum Boeck., E. squamigera Svens. and E. subarticulata (Nees) Boeck.). Svenson elected to place these plants under the heading “‘species of uncertain classification”’ due to the absence or conflicting distribution of diagnostic characters. Recent survey papers on the micro-morphology of Eleocharis achenes found sufficient epidermal characters to warrant their use taxonomically (Varma et al., 1989; Menapace, 1991). A detailed account of achene micro-characters at the series level is provided by Menapace (1991). Such characters as the configuration of the anticlinal cell walls (e.g., entire, crenate, etc.), the contour of the cell lumens (e.g., level, concave, etc.), and the presence or absence of lumen pits and depressions, were found to be systematically ' Present address: University of North Alabama, Florence, AL 35632 USA. 214 1993] Menapace— Eleocharis 215 helpful at the series and subseries ranks. SEM photomicrographs of achenes of representative species were provided in that paper (Menapace, 1991) to illustrate the characters just enumerated. Given the undesignated series status of the taxa named above, it was of interest to determine if the micro-morphology of their achene walls would add additional features from which to assess their subgeneric relationship. MATERIALS AND METHODS Mature achenes were removed from selected herbarium spec- imens. In order to evaluate infraspecific variation, two or three collections of each species were examined covering as wide a geographic distribution as possible. Data pertaining to the vouch- er specimens are provided in Table |. The cuticle and outer peri- clinal cell walls of Eleocharis achenes obscure systematically per- tinent micro-morphological features. As a consequence, these layers must be removed to permit observation of the taxonom- ically useful characters affiliated with the anticlinal cell walls and cell lumens. The cuticle and outer periclinal cell walls were elim- inated by submerging the achenes for 24 hours into a 1:9 con- centration of sulfuric acid and acetic anhydride. The achenes were subsequently removed from the acid mixture, sonicated, washed with distilled water, and coated with 30 nm of a gold-palladium alloy utilizing an International Scientific Instruments (ISI) PS2 coating unit. Micrographs were taken solely from the central por- tion of the achenes with an ISI Alpha-9 Scanning Electron Mi- croscope. RESULTS AND DISCUSSION In a prior study of Eleocharis (Menapace, 1991), X-ray spec- trometry determined that silicon is the prevailing element of the achene wall, the chemical form of which is believed to be silica. The initial micrograph (Figure 1) presents an Eleocharis achene with the cuticle and outer periclinal cell walls intact. All remaining micrographs, with the exception of Figures 2, 3, and 11, are Eleo- charis achenes in which the cuticle and outer periclinal cell walls have been removed to reveal features of the anticlinal cell walls and cell lumens. Table 1. Collection data of specimens utilized for SEM observations. Eleocharis acicularis. UNITED STATES. Illinois: heap Co., Elkville, Moh- ra big albie (stu); Nebraska: Lincoln Co., Kings Lake, Kellogg s.n. (stu); Co., Razorville, Steyermark ie Eleocharis aes BERMUDA. Pembroke Parish: Bevibrake Marsh, Moore 3164 (GH); MEXICO. Tamaulipas: Tampico, Palmer 570 (Gu); UNITED STATES: Florida: Levy Co., Waccasassa Bay, Easterday s.n. (FSU). Eleocharis elegans. CUBA. Chicks La Carmit Florida Blanca, Clements 1990 ead ECUADOR. Guayas: Tenquel, Holmgren 99 (S); JAMAICA. St. Anne: t, Proctor 34808 (1). Eleocharis elliptica. UNITED STATES. Michigan: Charlevoix Co., Beaver Is- land, Menapace, 24 (stu); Beaver Island, Menapace 25 (stv). pesos ib Cecsigeio BELIZE. Cayo: San Luis, Dwyer 338 (stu); DOMINICAN ae Trujillo: Bayaquana, a 22448 (GH); HONDURAS. Mora- sas, Williams 17054 (wi Eleocharis cena JAMAICA. [eee im Cut River, Proctor 35384 (1); eto: Pongode Manseriche, Mexia 6174 (Gu); UNITED STATES. aan Gulf Co., White City, Godfrey 75703 (Fsu). Eleocharis intermedia. CANADA. Ontario: Renfrow Co., Lyndoch Township, Dickson 1413 (cAN); UNITED STATES. Indiana: Nobel Co., Blackman Lake, Deam 21865 (IND); Massachusetts: Berkshire Co., Pontoosue Lake, Churchill s.n. (Mo). Eleocharis melanocarpa. UNITED STATES. Indiana: Porter Co., Indiana Dunes St. Park, Umbach 3817 (Mo); Texas: Leon Co., Oakwood, Palmer 13404 (mo); Massachusetts: Barnstable Co., Ostetville, Proctor 38072 (tr). Eleocharis multicaulis. FRANCE. Allier: Chassignol 903 (ny); PORTUGAL. Azores: Terceira Island, Silva 135 (Ny); SWEDEN. Blekinge: Ronneby, Broddeson 2 (ny). Eleocharis nodulosa. GUATEMALA. El Quiche: Nabaj, Proctor 25301 (u); S: Comayagua: Siguatepeque, Errazuriz 112 (stu); Morazan: nta, Espinosa 168 (stu). Bhar ods pachystyla. COSTA RICA. Puntarenas: Davidse 1269 (mo); CUBA. Moa: Oriente, Clements 3588 (Gu). Eleocharis parvula. UNITED STATES. Florida: Collier Co., Marco, Kattato 749 (sw). sisoais quinqueflora. CANADA. Saskatchewan: Cypress Hills, Breitung oe 0); UNITED STATES. Illinois: McHenry Co., Ringwood, Vassey 5.n. rail ichigan: Charlevoix Co., Beaver Island, Menapace 87 (stv). Pet atts — JAPAN Honshu: Togasi 1103 (Ny); Nara City, Kitagawa Kyushu: Lake Odako, Koyama 7012 (ny Eleocharis postelines, UNITED STATES. Arizona: Santa Cruz Co., Canelo Hills ure Preserve, Adams s.n. (stu); New Jersey: Ocean Co., Barnegat Bay, Hermann 439. 0). Eleocharis squamigera. BRAZIL. Parana: Jaguariaiva, Dusen 13276 es Santa Catarina: Campo Ere, Smith 13673 (ny); Cacador, Smith 8982 Eleocharis subarticulata. BRAZIL. Gardner 719 (Mo); Santa coats ‘Cacador, Smith 8990 (ny); Parana: Curitiba, Dusen 1 16a (Ny). Eleocharis tenuis. UNITED STATES. New Jersey: Ocean Co., Lanoka Harbor, Hirt 1387 (wy), Cape May Co., Villas, Wilson 910 (Ny); New York: Suffolk Co., Cambell 163 (Gu). 1993] Menapace— Eleocharis Zi7 In the discussion that follows the “species of uncertain classi- fication” are assessed in accordance with the micro-morpholog- ical characters of the achene wall. Eleocharis albida Torr. Eleocharis albida isa North American maritime spikerush from Maryland to central Mexico (Gleason, 1952). In his initial review of Eleocharis, Svenson (1929) assigned E. albida to the series Sulcatae, only to describe it at the completion of his study as a plant of “... unknown affinity .. .,” a species not allied to any other Eleocharis (Svenson, 1939a). In his final monographic work, covering North American Eleocharis, Svenson (1957) suggested that FE. albida may be allied to the series Pauciflorae, although a detailed justification was not provided. Plants of the Pauciflorae possess trigonous achenes that lack a constricted tubercle (e.g., E. parvula (R. & S.) Link, Figure 2, NCT). The fruit of E. albida is in fact trigonous, however, a large constricted tubercle is also present (Figure 3, CT). There is little question that the constricted condition of the tubercle persuaded Svenson to refrain from formally assigning E. albida to the Pau- ciflorae. On a micro-morphological basis, the achenes of E. albida (Fig- ures 4 and 5) bear widely crenate to dentate anticlinal cell walls, with undulating to level lumens that are rarely pitted. Attributes of this nature suggest a relationship with E. filiculmis Kunth series Sulcatae (Figure 6), E. multicaulis (Smith) Smith series Multi- caules (Figure 7), E. pellucida Presl series Multicaules (Figure 8), and E. pachystyla (Wright) C. B. Clarke series Sulcatae (Figure 9). The anticlinal cell walls of E. geniculata (L.) R. & S. (Figure 10), series Maculosae, are also crenate to dentate. However, in contrast to the trigonous achenes of E. albida, the achenes of E. geniculata are lenticular. Such a salient difference is sufficient, in the author’s judgement, to preclude a close affiliation between E. geniculata and E. albida. Eleocharis melanocarpa Torr. Although encountered sporadically within the interior of North America, E. melanocarpa is primarily a sedge of the Atlantic coastal plain from Massachusetts to Texas (Gleason, 1952). Sven- 218 Rhodora [Vol. 95 Pics ens L 4 pie asain ol S8072)) 208: goa sean fils albida (Moore 3164), Bar = 100 yu aes 1 7054, i. ousticnile (sie 135). 8. E. peor stihes nichhene 3588). 9. E. pellucida (Koyama 7012). Gap in bars = 1 ym. NCT = non-constricted tubercle, CT = constricted ef 1993] Menapace— Eleocharis 219 17, —_ - a Figures 10-17. 10. E. geniculata (Godfrey 75703). 11-13. E. melanocarpa (Proctor 38072), 14. E. quinquefolia (Menapace 87). 15. E. rostellata (Hermann 4392). 16. E. tenuis (Hirt 1387). 17. E. acicularis (Mohlenbrock 5760). Gap in bar. $=] hm. son (1929), initially assigned E. melanocarpa to its own mono- typic series, the Melanocarpeae. At the completion of his mono- graphic study, however, Svenson (1939a) abandoned the Melanocarpeae relegating E. melanocarpa to uncertain status. 220 Rhodora [Vol. 95 Later, Svenson (1957) again modified his view and formally del- egated E. melanocarpa to the series Pauciflorae, although a de- tailed explanation regarding this assessment was not given. As previously noted, the tubercles of Pauciflorae achenes are confluent with the achene body (Figure 2). There is little doubt that the highly appressed, non-constricted tubercle of E. melano- carpa (Figure 11) was a key factor in Svenson’s decision to allocate E. melanocarpa to the Pauciflorae. Although the convolute an- ticlinal cell walls of E. melanocarpa (Figures 12 and 13) do not repudiate Svenson’s decision to assign E. melanocarpa to the Pauciflorae, given the absence of achene conformity at the micro- morphological level within the Pauciflorae (e.g., E. quinqueflora (F. X. Hartm.) O. Schwarz, Figure 14, and E. rostellata Torr., Figure 15), the author questions whether the series, is in fact, a natural entity. Similar to E. albida, the convolute anticlinal cell walls of E. melanocarpa implies a relationship with either the Multicaules or Sulcatae. Eleocharis minarum Boeck. Eleocharis minarum is a taxon of questioned authenticity. Svenson (1939b), after examining the syntypes deposited at Stock- holm (S) and Copenhagen (C), discovered that E. minarum was based on immature plants, possibly allied to E. dunensis Kuek. of the series Su/catae. The author has also inspected the Stock- holm and Copenhagen plants, and concurs with Svenson’s as- sessment. A thorough search of the world’s major herbaria failed to secure a mature specimen of E. minarum. Given the absence of a plant bearing mature achenes, a micro-morphological evaluation was not possible. Eleocharis squamigera Svens. This species is a South American spikerush about which little has been published. Eleocharis squamigera has been collected from scattered localities throughout the tropical regions of the continent (Svenson, 1934). According to Svenson (1934), the general morphological ap- pearance of E. squamigera is similar to E. tenuis Schult., of the 1993] Menapace— Eleocharis 221 series Palustriformes subseries Truncatae. In fact, the type spec- imens of E. squamigera were initially assigned to E. tenuis (Sven- son, 1934). Nonetheless, the micro-morphology of the achenes differ in that the anticlinal cell walls of E. squamigera (Figures 18 and 19) are crenate to serrate, while those of E. tenuis (Figure 16) are serrulate to entire. Furthermore, the cell lumens of E. tenuis are more concave than those of E. squamigera. Clearly, the two plants are distinct, allied taxa. Svenson (1934) also suggested an affinity with the series Acicu- lares. However, achenes of the Aciculares possess highly elevated anticlinal cell walls (e.g., E. acicularis (L.) R. & S., Figure 17), a feature which is lacking from the achenes of E. squamigera. Although the achene micro-morphology of E. squamigera con- trasts somewhat with that of E. tenuis, it does approximate other taxa of the series Palustriformes subseries Truncatae (e.g., E. nod- ulosa (Roth) Schult., Figure 20, and FE. intermedia (Muhl.) Schult., Figure 21). Eleocharis subarticulata (Nees) Boeck. Eleocharis subarticulata, like much of the spikerush flora of Latin America, suffers from a lack of modern studies. Svenson (1939b) found the vegetative and achene characters to be near E. intermedia, a plant of the series Palustriformes subseries Trun- catae of North American distribution The micro-morphology of the achene wall (Figures 22 and 23) also suggests an alliance with the Truncatae. The anticlinal cell walls are repand to crenate with slightly concave lumens, attri- butes which are consistent in outline with E. elegans (H.B.K.) R. & S. (Figure 24) and to a lesser extent with E. e/liptica Kunth (Figure 25). SUMMARY Micro-morphological characters of the achene wall add cre- dence to Svenson’s (1929) conclusion that E. albida is allied to the series Sulcatae. The results are equally conducive, however, to a relationship with the series Multicaules. Consequently, a definitive decision regarding the serial placement of E. albida was not obtained. The micro-morphological attributes affiliated with the achenes 222 Rhodora [Vol. 95 Figures 18-25. 18 and 19. E. squamigera (Smith 13673). 20. E. elegans (Proc- tor 34808). 21. E. intermedia (Dickson 1413). 22 and 23. E. subarticulata (Smith 13673). 24. E. nodulosa (Proctor 25301). 25. E. elliptica (Menapace 24). Gaps in bar = 1 um 1993] Menapace— Eleocharis 223 of E. melanocarpa were also inconclusive. The convolute anti- clinal cell walls suggest a kinship with either the Multicaules or Sulcatae. Nonetheless, the evidence is once again equivocal as to which series is in fact the best alternative. Given the continued ambiguity surrounding the serial place- ment of E. albida and E. melanocarpa, it is considered best to retain their status as “‘species of uncertain classification.” With regard to E. squamigera and E. subarticulata, the author has concluded that the achene micro-morphology constitutes suf- ficient evidence, in conjunction with the overall macro-morpho- logical affinities, to assign E. squamigera and E. subarticulata to the series Palustriformes subseries Truncatae. An assessment of E. minarum proved to be impossible due to the absence of a specimen bearing mature achenes. ACKNOWLEDGMENTS The author wishes to express appreciation to Dr. Donald Ugent, Dr. Lawrence C. Matten, and the faculty and staff, Department of Plant Biology, Southern Illinois University at Carbondale (SIU- C). Special thanks to Dr. Barbara Crandall-Stotler, director of the SEM facility, Department of Plant Biology SIU-C. Gratitude is also extended to Dr. Daniel E. Wujek (CMC), Dr. Gordon P. DeWolf, Mr. John McMurray (Duke), Ms. Karen Nash (SIU), and an anonymous reviewer, for their helpful critique of the manuscript. LITERATURE CITED Gieason, H. A. 1952. The New Britton and Brown Illustrated Flora of the arpa United States and Adjacent Canada. Van Nostrand and Rein- hold Co., New York, NY. eo MENaAPACE, F. al D. E. Wusex. 1986. A systematic revision of the genus Carex SP aaeaes with respect to the section Lupulinae. Can. J. Bot. 64: 2785-2 oy . preliminary micro-morphological analysis of Eleocharis (Cy- peraceae) achenes for systematic potential. Can. 1541. Svenson, H. K. 1929. Monographic studies in the genus Eleocharis. Rhodora 31: 127-129. —. 1934. Monographic studies in the genus Eleocharis. Rhodora 36: 389. —. 1939a. Monographic studies in the genus Eleocharis. Rhodora 41: 3-4. ——. 1939b. Monographic studies in the genus Eleocharis. Rhodora 41: 104— 105. 224 Rhodora [Vol. 95 1957. Eleocharis. North American Flora 18: 509-540. VARMA, S. K., A. K. PANDEY AND A. K. SINHA. 1989. Epidermal surface patterns of achene in Eleocharis R. Br. (Cyperaceae). Curr. Sci. 58: 1374-1377. DEPARTMENT OF PLANT BIOLOGY SOUTHERN ILLINOIS UNIVERSITY CARBONDALE, IL 62901 RHODORA, Vol. 95, No. 883/884, pp. 225-233, 1993 REDISCOVERY OF THE MANY-CROWNED STRAWBERRY, FRAGARIA MULTICIPITA FERNALD PAUL M. CATLING ABSTRACT Three stations of Fragaria multicipita Fernald were rediscovered in the general area of the type locality along Riviére Ste.-Anne in Gaspé, Québec. This taxon has not been accepted as a species by a number of recent compilers and monog- raphers, but the plants retained distinctive features under greenhouse cultivation suggesting that taxonomic recognition may be appropriate, although the appro- priate rank is unclear. The distinctive features include small size, small almost stalkless leaflets and a multicipital, more or less runnerless caudex. The plants were confined to dry sandy or silty, open riverine bars. Study of living plants of F ae ed he ee ee - Pegs eee ptemies rth America. It will also contribute additional information relevant to the pe ns of strawberries, this in turn providing the information needed for substantial variation in native species of Fragaria, and in situ as well as ex situ protection is desirable. Key Words: Fragaria multicipita, strawberry, distribution, ecology, taxonomy, phytogeography, endemism, Gulf of St. Lawrence, germplasm protection, Québec, Canada Fragaria multicipita was described by M. L. Fernald in 1908 on the basis of material that he collected with J. F. Collins (230, GH) along Riviére Ste.-Anne in Gaspé, Québec. The species was never discovered anywhere else and Fernald was unable to re- locate the plants in 1927 when he found that the type locality had been used for a logging camp and the habitat there completely destroyed (Fernald, 1940). Many botanists visited the Ste.-Anne valley to search for Fragaria multicipita, but all searches were unsuccessful and it was assumed to be extinct (Drury, 1969; Scog- gan, 1978). Having been known only from the type collection for 84 years, it was rediscovered in 1992 at three separate localities along the Riviére Ste.-Anne. Pressed specimens of recently col- lected plants have been deposited at pao and mrt and living ma- terial is being propagated for further research. The purposes of the following notes are to outline the history of this taxon, and to provide some new information resulting from its rediscovery, along with an indication of its potential significance. 225 226 Rhodora [Vol. 95 TAXONOMIC HISTORY Fragaria multicipita was retained as a species by Rydberg (1908) in his monograph produced the same year that Fernald described the plant. It was one of three species of strawberries recognized in northeastern North America by Fernald in 1950, and it was accepted by Scoggan the same year in his flora of Bic and the Gaspé Peninsula. Most authors after 1950 either overlooked it, rejected it entirely, or placed it in synonymy under F. virginiana with neither expla- nation nor comment. Among the reasons for this are the two facts that it was not found again (see above) and that the most influ- ential botanists concerned with northeastern North America at the time were “lumpers”’ who considered taxa described by Fer- nald as the products of extreme splitting. Gleason (1952) placed F. multicipita in synonymy under F. virginiana var. terrae-novae (Rydb.) Fern. & Wieg. Fragaria multicipita was not mentioned by Boivin (1966) or Gleason and Cronquist (1963, 1991). Staudt (1962) in a monographic work on Fragaria did not mention F. multicipita, nor was it mentioned by him (1989) in a recent syn- opsis of the species of Fragaria. Scoggan (1978) rejected F. mul- ticipita from the flora of Canada, but he nevertheless discussed it in some depth. He correctly noted that Staudt (1962) had treated plants of F. vesca from Europe that were without runners as f. eflagellaris (Dene.) Staudt, although he (Staudt) had not men- tioned F. multicipita. Kartesz and Kartesz (1980) treated F. mul- ticipita as a synonym of F. virginiana ssp. virginiana. With this lack of acceptance it is not surprising that it was not mentioned in recent listings of the rare plants of Québec (Bouchard et al., 1983; Lavoie, 1992). Regardless of a recent lack of acceptance, F. multicipita has not been lost completely. Rice et al. (1982) listed it as one of four species of Fragaria occurring in Canada and the United States. A number of recent studies have found Fernaldian taxa to be discrete and worthy of recognition. Among these are studies of other northeastern endemic species named by Fernald such as Aster anticostensis Fernald and Aster laurentianus Fernald, both of which have recently proved worthy of species rank (Brouillet and Labrecque, 1987; Houle and Haber, 1990; Labrecque and Brouillet, 1990). Despite taxonomic recognition of some aber- rations and an untenable varietal concept, taxa described by Fer- 1993] Catling—Fragaria 227 nald are now more widely regarded as the potentially astute ob- servations of one of North America’s great taxonomists. As noted by Drury (1969); “Fernald’s judgement and ability to separate cogent from trivial differences were amazing.’ DISTINCTIVE FEATURES Although Fernald (1908) described the plant as caespitose with the caudex divided into several to many crowded upright branches that were very rarely stoloniferous, and he named it muilticipita in reference to the multicipital caudex (i.e., the root top or stem base with many crowns or heads), it was not until 1950 in the 8th edition of Gray’s manual that it became completely clear that this is what he regarded as one of the most distinctive characters. Additionally Fernald (1950) used the relatively small leaves (less than 3 cm long) with sessile leaflets possessed by F. multicipita to distinguish it from F. virginiana. Rydberg (1908) also used the lack of runners and the small sessile leaflets to distinguish F. multicipita from other species, but he also employed its short petals less than 5 mm long to distin- guish it from his Fragaria terrae-novae Rydb., which most sub- sequent authors placed with F. virginiana, either as a synonym or as a variety (e.g., Fernald, 1950). Scoggan (1950) utilized Ryd- berg’s (1908) key characters in his flora of Bic and the Gaspé Peninsula. Fragaria multicipita was basically distinguished as a small-leaved bushy strawberry lacking runners (Figure 1) by those taxonomists who recognized it. REDISCOVERY AND RECENT OBSERVATIONS Much of the lower valley of Riviére Ste.-Anne is a broad and more or less flat floodplain. Although the river is relatively fast flowing, it meanders through riverine deposits of gravel, sand and silt. The disturbance and erosion from periodic torrential flooding accompanied by ice-scouring in spring, and later deposition re- sults in extensive open areas near the river. Substrates vary in amounts of disturbance as well as composition and wetness, the latter a consequence of height and slope. Although some of the deposits are without vegetation cover, where disturbance is lim- ited or deposition less recent, a vegetation cover has developed. Fragaria multicipita was rediscovered on dry, open, gentle slopes 228 Rhodora [Vol. 95 Fragaria virginiana Duch, 38p. glauca (Wats.) Staudt Annotated by G, Staudt 1960 Ottawa a PROVINCE OF QUEBEC No. 23¢ Pragarca grub ben pla rey Figure 1. The holotype of F. multicipita (Fernald & Collins 230, GH). on bars composed of silt, fine sand, or gravel (Figure 2) where few other plants occurred and the vegetation cover was sparse. At 4 km SSE of Ste.-Anne-des Monts, 20 plants were found. At approx. 9 km SSE of Ste.-Anne-des-Monts, 52 plants were found. 1993] Catling— Fragaria 229 Figure 2. A, Riviére Ste.-Anne, Gaspé, showing extensive open, silty sandy and gravel bars where Fragaria multicipita still occurs. The habitat extends across the center of the photograph between the barren, rocky rivershore and the en- croaching forest. At approx. 13 km SSE of Ste.-Anne-des-Monts, 15 plants were found. Ten of the plants found at the station 9 km SSE, growing on an isolated bar differed from the type material only in having oo instead of appressed trichomes on the leaf petioles. ther species growing with or near to Fragaria multicipita in- cluded Achillea millefolium L. ssp. borealis (Bong.) Breitung, Bro- mus ciliatus L., B. inermis Leyss., Chrysanthemum leucanthemum L., Dryas drummondii Richards., infrataxa of E/ymus trachycau- lus (Link) Gould in Shinn., Elytrigia repens (L.) Nevski, Fragaria virginiana Miller, Galium mollugo L., Oenothera parviflora L., Poa compressa L., P. pratensis L., Rubus pubescens Raf. and Silene vulgaris (Moench) Garcke. The variation in Fragaria along Riviére Ste.-Anne is remark- able. In addition to F. multicipita with either spreading or ap- pressed trichomes, F. vesca L. ssp. americana (Porter) Staudt occurs in moist and more shaded places. Finally, plants referable to F. virginiana Miller but of different heights, had divergent or 230 Rhodora [Vol. 95 appressed ascending trichomes on the petioles, large or small leaflets of various shapes and numbers with long or short petioles, and stolons of various lengths, colors and thicknesses. One ob- vious potential explanation for the variation is the remarkable range of adjacent habitats, but a mingling of glacial and preglacial relicts (Drury, 1969; Fernald, 1925) with more recent invaders is also a possibility. The Ste.-Anne rivershore environment includes some inter- esting subarctic or alpine elements, in addition to Dryas drum- mondii, such as Carex atratiformis Britton, C. capillaris L., and C. media R. Br. These elements present a striking contrast to the surrounding forest and are only a few of the species included in the distinct and restricted rivershore flora of the region (Labrecque and Brouillet, 1990; Scoggan, 1950). Some of these elements, such as the arctic and cordilleran Lathyrus venosus Muhl. var. intonsus Butters & St. John, are known from only one, or a very few, eastern stations. Other well established northeastern endemics, such as Aster anticostensis Fernald and Pedicularis furbishiae S. Watson, are also species of disturbed rivershores. The Gaspé Peninsula and the general region of the Gulf of St. Lawrence have been identified as an evolutionary center (Argus and McNeill, 1974) and an area of relatively high endemism (e.g., Marie-Vic- torin, 1938; Morisset, 1971; Argus, 1977; Bouchard et al., 1985). Fragaria multicipita appears to be similar to many of the Gulf of St. Lawrence endemics in being a plant of open, disturbed habitats that is rare and characterized by small populations (Marie-Vic- torin, 1938; Morisset, 1971; Wynne-Edwards, 1939). MATERIAL EXAMINED: CANADA: Québec: Gaspé Ouest Mu- nicipality: Cap Chat Township: River Ste. Anne-des-Monts, 14—- 17 July 1906, M. L. Fernald & J. F. Collins 230 (HoLotyPE-GH!, IsOTYPE-GH!); 9 km SSE of Ste.-Anne-des-Monts (hwy 132) beside Riviére Ste.-Anne, 49°04'40’N, 66°29'55”W, 16 July 1992, P. M. Catling & V. R. Catling 13251 (DAO, MT). SIGNIFICANCE The rediscovery of F. multicipita is significant in many respects. It has already resulted in additional information relevant to clas- sification. When portions of 50 plants representing variation in Fragaria along the Ste.-Anne rivershore were grown beside each 1993] Catling—Fragaria 231 other in a greenhouse in similar pots and similar soil mix, all retained the distinctive morphological features seen in the field and in some cases these features became even more pronounced. For example, plants referable to Fragaria multicipita grew into large runnerless “‘bushes”’ with 50-100 ramets, but retained their small size and small nearly stalkless leaflets, and either did not produce runners or produced a few very short runners to 10 cm long. The evidence of localized genetic differentiation, as well as a rather distinct habitat, are consistent with species rank, but the more important point is that there is now living material available for the further assessment of rank and the study of variation. Fragaria multicipita was one of the species used by Fernald (1925) to support his “‘persistence” or “‘nunutak”’ hypothesis, and it was also one of 45 species, potentially supporting the hypothesis, that were screened by Drury (1969). The explanation for, and nature of endemism and unusual patterns of plant variation in the Gulf of St. Lawrence region has still to be thoroughly explored, although recent studies (e.g., Brouillet and Labrecque, 1987; La- brecque and Brouillet, 1990; Houle and Haber, 1990) have con- tributed much useful information. As a non-halophytic and non-serpentine taxon, F. multicipita is especially relevant to the persistence concept and to an understanding of northeastern ende- mism. Drury (1969) noted that, to advance knowledge in these areas, what was needed most of all were detailed studies of ecol- ogy, natural history, and variation of some of the most critical endemic taxa, such as F. multicipita. An understanding of variation in native wild strawberries and the protection of this variation has become more relevant to agriculture over the past several years for two reasons. Firstly, it has become more widely recognized that North American straw- berry production has a narrow germplasm base (e.g., Luby et al., 1992). Secondly, the use of native wild strawberry germplasm to increase genetic diversity of cultivated strawberries is increasingly important because of the need to reduce the use of chemicals for disease and pest control, while at the same time increasing pro- duction efficiency (e.g., Daubeny, 1990). Two additional points arise from these considerations. One is that there are in North America important centers of crop-relative diversity where in situ protection is desirable, the valley of Riviere Ste.-Anne being an example. Secondly information on patterns of variation within species, that would be useful for purposes of 2352 Rhodora [Vol. 95 germplasm protection, are inadequate, even for some of our most familiar native plants such as strawberries ACKNOWLEDGMENTS Dr. J. Cayouette, Dr. E. Small and Mr. W. J. Cody, all of the Biological Resources Division of Agriculture Canada’s Centre for Land and Biological Resources Research, provided useful com- ments. LITERATURE CITED Arcus, G. W. 1977. The conservation of ie rare and endangered plants, pp. 139-143. In: T. Mosquin and C. Suchal, Eds., Canada’s Threatened Species and Habitats. Canadian Wildlife ee Ottawa. cNen. 1974. Conservation of evolutionary centers in Canada, pp. 131-141. Jn: J. S. Maini and A. Carlisle, Eds., Conservation in Canada, A Conspectus. Canadian Forestry Service Publication 1340. Borvin, B. 1966. Enumération des plantes du Canada II—Lignidées. Naturaliste Can. 93: 371-437. BOUCHARD, A., D. BARABE, Y. BERGERON, M. DuMAIS AND S. Hay. she tigives dil des plantes vasculaires rares du Québec. Naturaliste Can. 112: 283-300. , M. Dumais AND S. Hay. 1983. Les plantes ae ara rares du Québec. Sviloseus (National Museum of Canada) No. 4 BROUILLET, L. AND J. LABRECQUE. 1987. Aster gaspensis serecit nombre chro- mosomique et hybridation naturelle avec |’A. novi-belgii L. Naturaliste Can. 114: 159-165. Dauseny, H. A. 1990. Strawberry breeding in Canada. Hortscience 25: 893- 894 Drury, W. H. 1969. Plant persistence in the Gulf of St. Lawrence, pp. 105- 148. In: K. N. H. Greenidge, Ed., Essays in Plant Geography and Ecology. Nova oe di Museum, Halifax FERNALD, M. 1908. Notes on some plants of northeastern North America. Rhodora ie 46-55. . 1925. Persist f iated f Boreal America. Mem. Amer. Acad. Arts Sci. 15: 238-34 a . 1940. The problems of conserving rare native plants. Smithsonian Re- port 1939: 375-391. ——. 1950. Gray’s Manual of Botany, 8th ed. American Book Co., New York. Gieason, H. A. 1952. The New Britton and Brown Illustrated Flora of the secre United States and Adjacent Canada. Hafner Press, New York. —— AnD A. Cronauist. 1963. Manual of Vascular Plants of ncnaranel United Ses and Adjacent Canada. D. van Nostrand Co., New AND 1991. Manual of vascular plants of northeastern "United States and idincent Canada, 2nd ed. New York Botanical Garden, Bronx, New York. 1993] Catling —Fragaria 2a3 Hou.e, F. AND E. HaBer. 1990. Status of the Gulf of St. Lawrence Aster, Aster laurentianus Seong in Canada. Can. Field-Nat. 104: 455-459. ESZ, J. AND R TEsz. 1980. A synonymized checklist of the vascular flora of the Tinieed States, Canada, and Greenland, Vol. II. The Biota of North America. University of North Carolina Press, Chapel Hill. 498 pp. LABRECQUE, J. AND L. BRoumLLeT. 1990. Aster anticostensis, an endemic of northeastern North America: biology and conservation. Rhodora 92: 129- 141. Lavorg, G. 1992. Plantes vasculaires susceptibles d’étre vip aid menacées ou vulnérables au Québec. Ministére de Environnement. Québe d Lusy, J. J., J. F. HANcock, JR. AND J. R. BALLINGTON. 1992. Patlecdon of native strawberry germplasm in the Pacific seienee and northern Rocky Moun- tains of the United States. Hortscienc Marie-VicTorIn, F. 1938. Phytogsowraphica problems of eastern Canada. Amer. Midl. Nat. 19: 489-558 MorIsseET, P. 1971. Bnasieisnt in vascular plants of the Gulf of St. Lawrence region. a oe. Can. 98: 1 Je Rice, W. E. et al. Compilers. soak Sade list of scientific plant names, 2 vols. ay Cons. Serv. Publ. SCS-TP-159. RybserG, P. A. 1908. Fragaria. North American Flora 22: 356-365 ScoGcGan, H. J. 1950. Flora of Bic “a the Gaspé Peninsula. National Museum of Canada Bulletin No. 115: 399 p The flora of Canada, wae 3 api Museum of Natural Sciences Publications } in Botany 7(3): 547-1 Staupt, G. 1962. Taxonomic ey in Oe genus Fragaria. Can. J. Bot. 40: 8 6 1989. The species of Fragaria, their taxonomy and geographic distri- bution. Acta Hort. 265: Wynne-Epwarps, V.C. 1939. S in the isolati f Ipine plant Trans. Roy. Soc. Can. Sect. 5: 35-41. AGRICULTURE CANADA, BIOLOGICAL RESOURCES DIVISION CENTRE FOR LAND AND BIOLOGICAL RESOURCES RESEARCH SAUNDERS BUILDING, CENTRAL EXPERIMENTAL FARM OTTAWA, CANADA KIA 0C6 RHODORA, Vol. 95, No. 883/884, pp. 234-253, 1993 . Ses aan NUMBER DETERMINATIONS IN M. COMPOSITAE, TRIBE ASTEREAE. V. a ren NORTH AMERICAN TAXA JOHN C. SEMPLE, JIE ZHANG AND CHUNSHENG XIANG ABSTRACT Chromosome numbers are reported for 333 individuals of 109 taxa and two interspecific hybrids from among 11 genera. The vast majority of the reports are for Aster sensu lato and Solidago. Many counts confirm previous reports for the taxa; approximately one third of the counts are first reports for a taxon, or ploidy level of a taxon, in one or more of the 26 states of the eastern United States and one Canadian province sampled. The cytogeography and cytotaxonomy of the Aster pilosus, Solidago stricta and S. uliginosa complexes are discussed. The fol- lowing first reports are included: Aster attenuatus, 2n = 48; A. surculosus, 2n = 36; Solidago latissimifolia, 2n = 18,, 2n = 36, 2n = 27, 2n = 54; S. patula var. strictula, 2n = 9,; S. pulchra, 2n = 18,. Key Words: pag sine Astereae, chromosome numbers, Aster, Boltonia, Co- a, Croptilon, Chrysopsis, Euthamia, Gutierrezia, Heterotheca, Hause Solidago, Vingulus, cytogeography, Canada, United States INTRODUCTION This is the fifth in a continuing series of reports by our labo- ratory and collaborators on chromosome numbers in the tribe Astereae (Semple, 1985; Semple and Chmielewski, 1987; Semple, Chmielewski and Lane, 1989; Semple, Chmielewski and Xiang, 1992). Determining the distribution patterns of cytotypes requires numerous counts from the range of a taxon. The counts listed in this paper are reported as further contributions to such studies. In this paper, our focus is on taxa native to the eastern and central United States and adjacent Canada. MATERIALS AND METHODS Meiotic counts were made from pollen mother cells dissected from buds fixed in the field in 3:1/EtOH: glacial acetic acid and subsequently stored under refrigeration in 70% EtOH. Mitotic counts were made from root tip cells taken from transplanted wild rootstocks or from seedlings grown from achenes collected in the wild. Root tips were pretreated in 0.01% colchicine or saturated paradichlorobenzene (PDB) for 2-3 hr., fixed in either Modified 234 1993] Semple et al.—Chromosomes tao Carnoy’s Fixative (4:3:1/chloroform : EtOH: glacial acetic acid) or Acetic Alcohol Fixative (3:1/EtOH: glacial acetic acid) and hydrolyzed in 1 N HCl for 30 min. at 60°C before squashing. Both anther sacs containing pollen mother cells and meristematic root tips were squashed in 1% acetic orcein, and counts of chro- mosomes were made from freshly prepared material. Permanent slides were made in most cases as described by Semple, Brammall and Chmielewski (1981) and remain in Semple’s possession. All counts were made from sporophyte tissue cells and are reported as “2n =.” Those counts made from late prophase (diplo- tene and diakinesis) and first metaphase in pollen mother cells undergoing meiosis are followed by “,,” indicating that pairs of chromosomes were counted, e.g., 27 = 18,. Many of the taxa studied include polyploids and could be indicated by additional notation, e.g., 2” = 4x = 32. However, since nearly all, if not all of these cytotypes appear to have undergone diploidization, counts for them are reported without indicating ploidy level. Vouchers for all counts are deposited in wAT with duplicates distributed to various other herbaria. Identifications were made by Semple. In some cases, the voucher specimens did not agree with published descriptions in one or more minor traits, such as the amount of pubescence or floret color; these cases are indicated by the qualifier ‘“‘aff.”’ in Table 1. RESULTS AND DISCUSSION Chromosome number determinations from 333 individuals representing 109 taxa and two interspecific hybrids from among 11 genera are reported in Table 1. All location and voucher data are published here for the first time. Populations were sampled in one province in Canada and in 26 states in the United States. In total, 308 of these reports are for asters and goldenrods (Aster, Virgulus, Solidago and Euthamia). Two-thirds of the counts con- firm previous reports for the taxa and most are presented without comment. Included in Table | are the following first reports for these taxa: Aster attenuatus Lindl., 2n = 48 (one population) from Texas; A. surculosus Michx., 2n = 36 (one population) from North Carolina; Solidago latissimifolia Mill. (one population each), 2n = 18,, from South Carolina, 2n = 36 from Massachusetts, 21 = 27,, from Massachusetts and Rhode Island, 2n = 54 from Mas- 236 Rhodora [Vol. 95 Table 1. cage number ie nado of Astereae from Canada and the United State Ah = Tufail Ahmed; Ch = J. G. Chmie- lewski; S = J. C Semple: Su = = Bambang Agus Suripto; ¥g = ChunSheng Xiang. When voucher specimens did not fit a published taxon description in one or more minor traits, the qualifier “aff.” is used. State highway numbers are indicated by the standard postal code abbreviation for the state followed by the number. ' First report for the taxon; ? first report(s) for the taxon or ploidy level for the particular province or state. Aster acuminatus Michx. 2n = 9,,. Maine. York Co.: Lyman, ME-111 0.1 km E f ME-35, S & Su 9587. A, anomalus Englemann in Torr. & Gray. 2n = 16. Arkansas.” Carrol Co.: US- 62 W ~ ipl 6.9 km E of AR-23, S & Su 9950. Illinois. Union Co.: Rd., woods, S & Su 9877 A. ee (L.) MacMillan ig and - paternus Cronq.) 2n = 18. North Carolina. Cherokee Co.: W of Ranger, US-64 7.8 km W of US-129, S & Jeff Semple 10353 (2 plants). A. attenuatus Lindl. in Hook.' 2n = 48. Texas. Jasper Co.: E of Woodville, US- 190 13.5 km E of US-69, S & Su 10055 A. avitus Alex. 2n = 18. Georgia. Dekalb Co.: base of Stone Mt. near access rd to top of mountain, S & Jeff Semple 10346. A. borealis (Torr. & Gray) Prov. 2n = 64. Ontario. Kenora Dist.: Peawanuck (S of old Winisk), Clements s.n. A. chlorolepis Burgess in Small. 2n = 36. North Carolina. Yancey Co.: Mt. Mitchell, midway along rd to park, S & Su 9684. A. ciliolatus ee 2n = 48. Ontario. Kenora Dist.: Peawanuck (S of old Win- isk), Clement. A. cordifolius “46 a = 16. Illinois. Union Co.: Trail of Tears State For., S & Su 9882. Kentucky. Allen Co.: US-31 a few mi. S of Petroleum, S & Ch 9110. 2n = 32. Kentucky. Adair Co.: NW of Toria, Cumberland Pkwy (MP38.1), S & Su 9447. Pulaski Co.: KY-1392 ca. 10 km S of Livingston, bluff of Rock- castle R., S & Su 9603. eoaereysl Flore Co.: US-59 just N of Ouachita Nat'l Fon boundary, S&S A. aff. cordifolius L. x A. pee Willd. 2n = 32. Ontario. Waterloo Co.: Wellesley Twp, S & Ch 9097A. A. depauperatus (Porter) Fern. 2n = 16. North Carolina. Granville Co.: NE of Butner, along power line right-of-way, S & Su 9724. A. divaricatus L. 2n = 9,,. apg enn Perry Co.: PA-74 3 km NW of Ickes- berg, S & Su 9490. 2n = 18. Massachusetts. Bristol Co.: N of Somerset along Segreganset Estuary, S 9/83. A. drummondii Lindl. 2n = 16. Arkansas. Carrol Co.: N edge of Berryville, AR-21, corner of Julie St. and Standley Ave. NE, S & Su 9948. Louisiana.’ Caddo Parish: LA-2 8.0 km E of Vivian (LA-1), S & Su 10031. Texas.’ La- mar Co.: Brackeen Cemetery Rd. off TX-19, S & Su 10003. Newton Co.: NE of Burkeville, TX-63 2.5 km W of Sabine R., S & Su 10049. 2n = 32. Illi- nois. Kakankee Co.: NW of Bradley, Kakankee State Park, along Rock R., S & Su 9867. St. Clair Co.; Fairview Heights, community park, S & Su 9406. Mississippi.” Adams Co.: SE of Natches, Fire Tower Rd. just E of US-91, S 1993] Semple et al.—Chromosomes 237 Table 1. Continued. & Su 10087. Texas.’ Anderson Co.: US-84 17.6 km E of Palestine city limits (Loope Rd-56), S & Su 10016. A. dumosus L. 2n = 8,,. North Carolina. Bertie Co.: SE of Bertie, NC-308 9.8 km SE of US-17, S & Su 9742. Yadkin Co.: WSW of Yadkinville, US-421 ca. 1 km E of US-21, S & Su 9708 & S & Su 9709. 2n = 16. Kentucky. Pulaski Co.: KY-1392 ca. 10 km S of Livingston, bluff of Rockcastle R., S & Su 9604. Louisiana. Concordia Parish, W of Ferriday, US-84 6 km W of LA- 15, vicinity of picnic area, S & Su 10069. Massachusetts. Essex Co.: N of Newbury, by airport, ca. 1.5 km W of Plum Is. R., S & Su 9574. Mississippi. Amite Co.: NE of Liberty, Holmes Rd. 6.7 km SE of MS-570, S & Su 10102. n = 32. Arkansas. Franklin Co.: N of Ozark, AR-23 just S of I-40, S & Su 9966. Sebastian Co.: AR-28 4.6 km E of Cauthron, 14 km W of US-71, S & Su 9980. Louisiana. Vernon Parish: US-171 3.4 km N of Anacoco (LA-111), S & Su 10041. Oklahoma.2? McCurtain Co.; US-70 3.5 km E of Valiant, W of Millerton, S & Su 9995. South Carolina. Clarendon Co.: US-521 1.1 km SE of Wilson, S & Su 9784. Texas.? Newton Co.: NE of Burkeville, TX-63 2.5 km W of Sabine R., S & Su 10048. A, gracilis Nutt. 2n = 9,,. New Jersey. Burlington Co.: N of Chatsworth, Co. Rd.-563 0.9 km S of NJ-72, S 10365; S of Chatworth, local road to cranber- ry bog, S 10369. Ocean Co.: W of Barnegat, Co. Rd.-554 2.4 km W of Gar- den State Parkway, S & Su 9516. A. hemisphaericus Alex. in Small. 2n = 18. Mississippi. Franklin Co.: US-98 just E of MS-570, S & Su 10101 A, ‘ne L. var. laevis x A. lanceolatus Willd. ssp. lanceolatus var. lanceolatus. = 48. Ontario. Kent Co.: Hwy-3, Port Crewe, S & Ch 9138. oA: eet Willd. ssp. iene var. hirsuticaulis Semple & Chmielewski. 2n = 32. Wisconsin. Douglas Co.: US-2 E of Maple, S 9047-2. A. lanceolatus Willd. ssp. acon oe var. interior (Wieg.) Semple & Chmie- lewski. 2n = 48. Arkansas.? Logan Co.: SW of Booneville, AR-23 2.2 km E of Carolan Cemetery, S & Su 9976. Iinois. White Co.: S of Norris City, junction of IL-1 and US-45, S & Su 9423. 2n = 64. Kentucky.’ Webster Co.; KY-109 1.7 km W of Clay, S & Su 9428. A, lanceclarns Willd. ssp. /anceolatus var. latifolius Semple & Chmielewski. 2n = 64. Illinois. Franklin Co.: IL-149 1.7 km W of Zeigler (IL-148). S & Su 9875. Indiana. Putnam Co.: SE of Cloverdale, IN-42 4.1 km W of US-231, S & Su 9401. Kentucky.? Robertson Co.: KY-617 0.5 km W of Piqua, banks of Bluelick R., S & Su 9593. Webster Co.; KY-109 1.7 km W of Clay, S & Su 9. 427. A. lanceolatus Willd. ssp. lanceolatus aff. var. latifolius Semple & Chmielewski. 2n = 64. Alabama.? Montgomery Co.: SE of Montgomery, US-321 0.7 km SE of US-82 S & Su 10155. The rays are blue and the phyllaries are less graduated than —. yeh the variety). A, oe (L.) Britt. 16. Missouri. Carter Co.: US-60 1.8 km E of Van nS & Su 9907. ellis Flore Co.: US-59 just N of Ouachita Nat’l 0 neon sai iver bank, S & Su 9984. 2n = 32. Alabama. Dale Co.: E of Clayhatchee, US-84 at Little Choctawhatchee R., S & Su 10152. Mlinois. 238 Rhodora [Vol. 95 Table 1. Continued. Hamilton Co.: IL-142 1 km SE of Dahlgreen, S & Su 9419. Kentucky. Estill Co.: KY-52 E of Ravenna, S & Su 9457. Hopkins Co.: KY-109 1 km N of Charleston, S & Su 9430. Pulaski Co.: KY-1392 ca. 10 km S of Livingston, bluff slope of Rockcastle R., S & Su 9601, Missouri. Douglas Co.: MO-14 bine! Bridges (MO-181), along N. Fork of White R., S & Su 9927. Tennes- ee. Montgomery Co.: US-79 0.5 km S of Kentucky state line, S & Su 9435. Teal: Rusk Co.: US-84 6.4 km E of Reklaw, S & Su 10023. A. linariifolius L. 2n = 9,,. Louisiana. Rapides Parish: N of Forest Hill, LA-112 km W of US-165, S & Su 10090. Maine. Sagadahoc Co.: NW of Popham Beach, junction of ME-209 and ME-216, S 10378. Massachusetts. Barnstable Co.: Harwich, Queen Anne Rd 2 km NE of Old Chatham Rd., S & Su 9559. New Jersey.” Atlantic Co.: US-40 5.6 km W of Mays Landing (NJ-42), S & Su 9504. South Carolina. Berkeley Co.: SC-45 ca. 5 km E of St. Stephen, Francis Marion Nat’! For., S & Su 9794. A. nemoralis Ait. 2n = 9,,. Maine. Sagadahoc Co.: W Popham Beach, ME-209 2 km E of ME-216, S 10380. A. novi-belgii L. 2n = 48. Maine. York Co.: Wells, vicinity of Municipal Of- fices, S & Su 9581. A. ontarionis Wieg. 2n = 32. Tennessee. Marshall Co.: Henry Horton State Park, S & Ch 9117. Montgomery Co.: NE of Clarkville, US-79 0.5 km S of Kentucky state line, S & Su 94 A. oolentangiensis Riddell. 2n = 32. Illinois. St. Clair Co.; Fairview Heights, community park, S & Su 9407. Missouri. Oregon Co.: MO-19 16.6 km S of US-60, N of Greer, S & Su 991]. Wayne Co.: Co. Rd.-F 0.1 km E of US-67, 7.8 km S of Greenville, S & Su 9894. Oklahoma. McCurtain Co.: US-259 20.9 km N of Broken Bow, S & Su 9990. A. aff. parviceps (Burgess) Mackenzie & Bush. [See text for discussion of these collections.] 27 = 32. Missouri. Washington Co.: MO-8 6.6 km W of Potosi, S, Su & Ah 9393. A. aff. pilosus Willd. var. pringlei (A. Gray) Blake. [See text for discussion of these collections.] 2” = 48. Missouri.? Oregon Co.: N of Greer, MO-19 16.6 km S of US-60, Mark Twain Nat’l For., vicinity of For. Rd.-3174, S & Su 9912. Taney Co.: MO-76 12.6 km W of county line, cedar glade, S & Su 9938. Wayne Co.: Co. Rd.-F 0.1 km E of US-67, 7.8 km S of Greenville, S & Su 9892. A. pilosus Willd. var. pilosus. 2n = 32. Ontario. Essex Co.: Colchester N. Twp, Essex-12 5 km SW of Gesto, SW of Canard R., S & Ch 9145. Arkansas. Madison Co.: S of Eureka Spring, AR-23 1.8 km S of county line, S & Su 9955. IMinois. 7.5 km E of McLeansboro at Co. Rd.-1400E, S & Su 9422. Kentucky. Nicholas Co.: Bluelick Battlefield State Park, S & Su 9596. Warren Co.: US-68 at Logan Co. line, vicinity of quarry, S & Su 9441. North Caroli- na. Avery Co.: E of Linville, top of Grandfather Mt., edge of parking lot, S & Su 9672. 2n = 48. Illinois. Hamilton Co.: IL-142 1.0 km SE of Dahlgreen, S & Su 9417. meanest McCurtain Co.: W of Millerton, US-70 3.5 km E of Valiant, S & Su 9 A, pilosus Willd. aff. var. : prinighet (A. Gray) Blake. 2n = 48. Massachusetts Barnstable Co.: East Brewster, along Bike Trail W of Nickerson State Park, 1993] Semple et al.— Chromosomes 239 Table 1. Continued. near MA-6A, S & Su 9551. (The plant is more hairy than typical but not as hairy as var. pilosus.) . praealtus Poir. var. praealtus. 2n = 32. Mississippi. Amite Co.: NE of Liber- ty, Holms Rd., S & Su 10105. Missouri. Greer Co.: SE of Walnut Grove, MO-123 at Farm Rd-56, S & Su 9946. Oklahoma. McCurtain Co.: US-70 E edge of Idabel, S & Su 9993. 2n = 48. Texas. Lamar Co.: US-271 S of Pow- derly, N of Paris, N of Loop-Rd.-287, S & Su 9999. A. prenanthoides Muhl. 2n = 32. Kentucky. Laurel Co.: along Rockcastle R. above mouth of Blair Branch Creek, S & Su 9854. Menifee Co.: Boone Nat’l For., KY-715 along river, S & Su 9465. Tennessee.? Unicoi Co.: E of Unicoi, TN-107 5.2 km E of TN-173, S & Su 9632. . priceae Britt. 2n = 64. Kentucky.’ Logan Co.: Russellville, US-68 just N of US-431, glade, S & Su 9438. Warren Co.: US-68 at Logan County line, by quarry, S & Su 9440-1. . puniceus L. 2n = 16. Georgia.” Rabun Co.: US-76 ca. 3 km W of Clayton, S$ & Su 9830. Minnesota. Pine Co.: MN-18 E of N end of Big Pine Lake, near county line, S 9069. North Carolina. Avery Co.: W of Linville, roar Mt., near Black Rock Hiking Trail, S & Su 9673. neared Unicoi Co.: E of Unicoi, TN-107 0.3 km W of NC state line, S & Su 964 A. racemosus Ell. Spal treated in most floras under the isp name A. virnineus Lam.] 2n = 8,,. North Carolina. Chowan Co.: NC-37 just N of Albermarle Sound Causeway, S & Su 9747. 2n = 16. Alabama.- ‘visio mery Co.: SE of Montgomery, US-231 0.7 km SE of US-82, S & Su 10/56. Tusca- teeiae Co.: US-82 2.4 km W of US-43, W of Northport, NW of Tuscaloosa, S$ & Su 10162. Arkansas. Washington Co.: US-71, rest area N of Brentwood, S$ & Su 9959. Kentucky.? Webster Co.: KY-109 1.7 km W of Clay. S & Su 9429. Vouisiana.? East Feliciana Parish: S of Felps, LA-67 2.5 km S of LA- 42? § & Su 10107. Red River Parish: N of Edgefield, LA-7 11.2 km S of county line, S & Su 10036. Mississippi. Franklin Co.: NW of Roxie, US-84 1.5 km W of MS-33, S & Su 10098. Monroe Co.: S of Becker, MS-25 0.9 km N of MS-8, S & Su 10167. Missouri.? Wayne Co.: Co. Rd.-F 0.1 km E of US-67, 7.8 km S of Greenville, S & Su 9895. Texas. Anderson Co.: US-84 17.6 km E of Palestine city limit (Loop Rd-56), S & Su 10017. Grimes Co.: TX-105 2.6 km W of Plantersville, S & Su 10064. San Jacinto Co.: TX-150 3.6 km E of Shepherd, S & Su 10062. 2n = 32. North Carolina. Durham Co.: N of Durham, by Eno R., Penny’s Head Nature Res., S & Su 9718. A. reticulatus Pursh. 2n = 18. Florida. Leon Co.: S of Bloxham, Appalachicola N.F., For. Rd.-320 just E of Co. Rd.-375 (TIS, R4W, Sec. 32), S & Jeff Sem- ple 10342. A. retroflexus Lindl. in DC. 2n = 48. North Carolina. Mitchell Co.: fo! 226 0.5 m E of ogee state line, S & Su 9648. Watauga Co.: W o Rock, o & Su 9681. Tennessee. Unicoi Co.: TN-107 17 km E of Calcot S& Su 964 A. seric arpa (Small) K. Schum. 2n = 18. South Carolina.’ neat st NE of entrance to Cheraw Fish Hatchery, S & Su 9780. A. shortii oor in Hook. 2n = 16. Illinois. Kakankee Co.: NW of ment Kakankee R. State Park, woods by Rock R., S & Su 9868. Kentucky. Adair aN nN k ae, 240 Rhodora [Vol. 95 Table 1. Continued. Co.: NW of Toria, Cumberland Parkway (MP38.1) W of KY-80, S & Su 9449. 2n = 32. Indiana. Putnam Co.: SE of Cloverdale, IN-42 4.1 km W of US-231, S & Su 9402. Kentucky. Christian Co.: KY-109 SE of Dawson Spring, Pennyrite For. State Park, S & Su 9433. Warren Co.: S of Alvaton, US-231 2.8 km S of Co. Rd.-961, S & Ch 9105. A. i i Michx. 2” = 18. North Carolina. Cherokee Co.: US-64 7.8 km of Ranger (US-129), S & Jeff Semple 10352 (3 plants). A. ee Ait. 2n = 36,,. Massachusetts. Plymouth Co.: W of Cedarville, Long Pond Rd. 0.3 km N of Carters Bridge Rd., N of Great Herring Lake, S Su 9567 A. subulatus L. var. subulatus. 2n = 5,,. New Jersey. Ocean Co.: Tuckerton, cedar marsh E of town center, S & Su 9525 A. surculosus Michx.! 2n = 36. North Carolina.? Cherokee Co.: SE of Ranger, 2.35 km S of US-64, W of Ranger, S & Jeff Semple 10350. A. texanus Burgess in Small. 2n = 32. Texas. Grimes Co.: TX-105 2.6 km W of Plantersville, S & Su 10065. A. aff. texanus Burgess in Small. 2n = 32. Texas. Jasper Co.: US-190 at Farm Rd-256, W of TX-63, S & Su 10054. A. tortifolius Michx. 2n = 9,,. South Carolina. Jasper Co.: SC-652 14.9 km SE of Pineland, S & Su 9804. A. turbinellus Lindl. 2n = a 96. Missouri. Washington Co.: MO-8 6.6 km W of Potosi. S, Su & A A. umbellatus Mill. var. aoe 2n = 9,,. Connecticut.2 New London Co.: Waterford, Waterford Conservation Area, Nevins Brook, S 9/9]. Massachu- setts.> Essex Co.: N of Newbury, ca. 1.5 km W of Plum Is. R., opposite air- port, S & Su 9573. A, undulatus L. 2n = 9,,. South Carolina. Berkeley Co.: SC-45 ca. 5 km E of St. Stephen, Francis Marion Nat’l For., S & Su 9793. 2n = 32. Kentucky. Adair Co.: NW of Toria, Cumberland Parkway (MP38.1) W of KY-80, S & Su 9448. North Carolina. Watauga Co.: US-421 at Wilkes Co. line, S & Su 970 ji A. urophyllus Lindl. in DC. 2n = 16. Arkansas.’ Carrol Co.: W of Berryville, km E of AR-23, S & Su 9949. Minnesota. Hennepin Co., Eden a, Seta Rd. just W of county airport, S et al. 885]. Missouri. Carter of US-60, S & Su 9903. Hickory Co.: E of Wheatland, US-54 0.9 km E of Hicomo Picnic Area, S & Su 9943. Wayne Co.: E of Williamsville, MO-49 4.4 km E of Co. Rd-A, S & Su 9901. Boltonia asteroides (L.) L’Her. var. recognita (Fern. & Griscom) Cronq. 2n = 18. Illinois. Hamilton Co.: IL-142 1 km SE of Dahlgreen, S & Su 9415; IL- 142 8.5 km SE of Dahlgreen, S & Su 9420. C onyza canadensis (L.) Cronq. var. canadensis. 2n = 18. Ontario. Waterloo Re- onal Mun.: Waterloo, University of Waterloo, north campus greenhouse, weed in pot, S /038x. Croptilon divaricatum (nutt.) Raf. 2n = 4,,. North Carolina. Pitt Co.: W of Bethel, US-64 0.2 km E of Edgecombe Co. line, S & Su 9736 1993] Semple et al.—Chromosomes 241 Table 1. Continued. —— gossypina (Michx.) Ell. ssp. gossypina f. gossypina. 2n = 9,,. North a. Pender Co.: Hampstead, NC-210 0.9 km W of US-17, S & Su C. gossypina (Michx.) Ell. ssp. gossypina f. trichophylla (Nutt.) Semple. 2n = 9,,. North Carolina.” Bladden Co.: NC-53 20 km SE of US-701, S & Su C. mariana (L.) Ell. 2n = 12,,. New Jersey. Atlantic Co.: US-40 5.6 km W of Mays Landing (Hwy-52), S & Su 9503. Louisiana. Beauregard Parish: LA-12 11.8 km W of Ragley (US-171), 1.4 km W of Gordon Fire Tower, S & Su 10089. North Carolina. Bertie Co.: SE of Bertie, NC-308 9.8 km SE of US- 17, S & Su 9742. South Carolina. Colleton Co.: NW of Jacksonboro, NC-64 ca. 2.5 km NW of US-17, S & Su 9799 Euthamia leptocephala (Torr. & Gray) Greene. 2n = 18. Texas. Navarro Co.: W of Eureka, US-287 3.3 km W of Farm Rd.-637, S & Su 10010. Newton Co.: TX-63 2.5 km W of Sabine R., NE of Burkeville, S & Su 10050. Polk Co.: E of Livingstone, US-190 1.8 km E of Soda, 16.4 km E of TX-146, S & Su 10061. E. anor (Pursh) Nutt. 2” = 18. North Carolina. Franklin Co.: NC-56 4.2 county line, S & Su 9731. South Carolina.2 Hampton Co.: SC-68 11 km Ww of Yemasee, S & Su 9802 Gutierrezia texana (DC.) Torr. & Gray var. texana. 2n = 4,,. Texas. Dallas Co.: Cedar Hill, Cedar Hill Rd. opposite lot No. 1110, RR right-of-way, S & Su 10008 seeds secession (Greene) Shinners var. glandulissimum Semple. 2n a.? Limestone Co.: Littleville, US-43 at N city limits, 5 & Su 10178. eae Carrol Co.: NE of Oak Grove, AR-21 just S of Blue Eye, S & Su 9947. Missouri.2 Christian Co.: Ozark, MO-14 just E of US-65, S & Su 9939. H. subaxillaris Lam. 2n = 9,,. New Jersey.? Salem Co.: W of Malaga, NJ-55, § Su 9498 Pityopsis falcata (Pursh) Nutt. 21 = 9,,. Massachusetts. Barnstable Co.: Cape od, East oo by Bike Trail W of Nickerson State Park, near MA-6A, S & Su 9762. P. graminifolia (Michx.) Nutt. var. latifolia (Fern.) Semple & Bowers. 2n = 18,,. North Carolina. Chowan Co.: NC-37 just N of Albermarle Sound Causeway, SE of Edenton, S & Su 9746. Pender Co.: Hampstead, NC-210 ca. 4.1 km W of US-17, S & Su 9762. P. graminifolia (Michx.) Nutt. var. tenuifolia (Torr.) Semple & Bowers. 2n = 9,,. Louisiana.? Beauregard Parish: LA-12 SE of DeQuincy, S & Su 10083. East Feliciana Parish: E of Woodland, LA-432 1.5 km E of LA-67, S & Su 10113 Dg oliganthe (Chapm.) Small. 2” = 36. Florida. Franklin Co.: W of St. Teresa, -319 2.7 km § of Ochlockonee R., S & Jeff Semple 10340. Solidago altissima L. var. altissima. 2n = 36. Texas.” Grimes Co.: TX-105 1.5 m E of TX-Farm Rd-362, S & Su 10069. Harris Co.: N of Bayton, I-10 rest 242 Rhodora [Vol. 95 Table 1. Continued. area, S & Su 10076. Lamar Co.: S of Powderly, US-271 N of Loop Rd-286, S & Su 10002. 2n = 54. Mlinois. Hamilton Co.: IL-142 1 km SE of Dahl- green, S & Su 94/8. Kentucky. Christian Co.: KY-109 SE of Dawson Spring, N end of Pennyrite Forest State Park, S & Su 9434-1]. Harrison Co.: US-62 2 km SW of Clayville, 5 & Su 9592. Louisiana. Vernon Parish: N of Anacoco, US-171 3.4 km N of LA-111, S & Su 10046. Missouri. Washington Co.: NE of Potosi, ni 21 just N of MO-8, S, Su & Ah 9395. North Carolina. Jones Co.: Chadwick, US-17 ca. 5 km NE of Maysville, S & Su 9753. Pitt Co.: of Bethel, US- 64 0.2 km W of county line, S & Su 9737. South fociais Colleton Co.: NW of Jacksonboro, NC-64 2.5 km from US-17, S. & Su 9800. Tennessee.2 Montgomery Co.: US-79 0.5 km S of Kentucky state line, ca. 12 km NE of Clarkville, S & Su 9437. S. arguta Ait. ssp. arguta. 2n = & Pennsylvania. Juniata Co.: 5 km NW of Ickesberg. ridge top, S & Su 9485. S. arguia Ait. ssp. caroliniana (A. Gray) G. Morton. 2n = 9,,. South Carolina. Berkeley Co.: SC-45 ca. 5 km E of St. Stephen, Francis Marion Nat’! For., S & Su 9792, 2n = 18. Mississippi.2 Amite Co.: NE of Liberty, Holmes Rd. 6.7 km SE of MS-570, S & Su 10104. Missouri.2 Taney Co.: MO-76 11.8 km W of county line, S & Su 9935. 2n = 36. Kentucky. Adair Co.: NW of Toria, Cumberland Parkway (MP38.1), S & Su 9444. Laurel Co.: Daniel Boone St. For. junction of For. Rd.-119 and For. Rd-56, S : Fe 9613. S. bicolor L. 2n = 9,,. Maine. Sagadahoc Co.: S of B shige alae ine woods above The Basin tidal basin, S 10377. sn talon Perry 743 km NW of Ickesberg, S & Su 9487. S. caesia L. 2” = 18. Indiana. Putnam Co.: SE of Cloverdale, IN-42 4.1 km W of US-231. S & Su 9403. Kentucky. Powell Co.: Boone Nat’l For. Tunnel ia Rd... S & Su 9462. North Carolina. pe rg Co.: N of Durham, Pen- s Bend Nature Reserve. Eno R., S & Su fo eas L. var. canadensis. 2n = 9,. peas inthe, Plymouth Co.: E of Raynham, MA-104 just S of Lake oo S$ 9181. §. delicatula Small. 2n = 18. Texas. Anderson Co.: Co. Rd.-494 0.5 km W of US-287, 3.5 km W of Cayuga, S & Su 10014 S. erecta Pursh. 2n = 9,,. New Jersey. Atlantic Co.: US-40 5.6 km W of Mays Landing, S & Su 9501. North Carolina. Franklin Co.: NC-56 4.2 km W of county line, S & Su 9735. 2n = 18. Kentucky. Estill Co.: E of Ravenna, KY- 52 by Ravenna-Pryse Rd., S & Su 9454. Pulaski Co.: KY-1392 ca. 10 km S of Livingston, bluff of Rockcastle R., S & Su 9605. Mississippi.? Itawamba Co.: MS-23 10.1 km N of Tremont, S & Su 10175. S. fistulosa Mill. 2n = 9,,. North Carolina. Martin Co.: US-64 just W of Par- mele, S & Su 9739. 2n = 18. Alabama.? Mobile Co.: N of Theodore, Co. Rd- 59 ca. 2 km S of I-10, S & Su 10128. Mississippi.? Harrison Co.: N o D'Iberville, MS-67 N of I-10, S & Su 10125. New Jersey. Atlantic Co.: swampy woods by NJ-50 just S of US-40, S & Su 9506. North Carolina. Jones Co.: Chadwick, US-17 ca. 5 km NE of Maysville, S & Su 9751. South Carolina. Clarendon Co.: US-521 1.1 km SE of Wilson, S & Su 9785. S. flexicaulis L. 2n = 36. Kentucky. Estill Co.: KY-52 E of Ravenna on Raven- na-Pryse Rd., S & Su 9453. Menifee Co.: Boone, N.F., KY-715, river bank, 1993] Semple et al.—Chromosomes 243 Table 1. Continued. S & Su 9464. Pulaski Co.: KY-1392 ca. 10 km S of Livingston, bluff of Rockcastle R., S & Su 9606. S. aff. flexicaulis L. 2n = 90.' Kentucky. Leslie Co.: S of Stinnett, US-421 just S of where highway crosses river, S & Su 9619. [The plant was tall for the spe- cies and the heads were very large.] S. gigantea Ait. 2n = 18. Kentucky.” Leslie Co.: US-421 S of Stinnett, S & Su 9620. Mississippi. Lowndes Co.: Columbus, N end of tow, US-45 S of Chap- Rd., S & Su 10165. North Carolina. Franklin Co.: NC-56 4.2 km W of Nash County line, S & Su 9732. Washington Co.: E of Skinnerville, US-64 just W of NC-32, S & Su 9748. South Carolina.2 McCormick Co.: 4 km N of McCormick, along side rd off SC-28, S & Su 9819. 2n = 36. Illinois. Kakan- kee Co.: NW of Bradley, Kakankee R. State Park, S & Su 9871. White Co.: § of Norris City, junction of IL-1 and US-45, S & Su 9424. Missouri.? Douglas Co.: MO-14 at Twin Bridges (MO-181), woods by N. Fork White R., S & Su 9928. Hickory Co.: E of Wheatland, US-54 ca. 0.9 km E of Hicomo Picnic Area, S & Su 9945. 2n = 54. Nebraska.? Hooker Co.: N of Mullen, NE-97 at Middle Loup R., S, Su & Ah 9200. S. glomerata Michx. 2n = 108. North Carolina. Avery Co.: W of Linville, Grandfather Mt., cliff face at top of mt. SW of swing bridge parking lot, S & Su 9670; ca. 5000 ft. el., S & Su 9676. Mitchell Co.: top of Roan Mt., Roan High Bluff, 6250 ft. el., S & Su 966]. Yancey Co.: Mt. Mitchell, near top, ca. 6000 ft. el., S & Su 9685; midway along rd. to park, mid el., S & Su 9683 S. gracillima Torr. & Gray. 2n = 9,,. South Carolina.’ a Co.: NW of Barnwell, US-278 just N of Salkehatchie R., S & Su 9 S. hispida Muhl. 2n = 18. Ontario. Kenora Dist.: cotsane Clements s.n. S. juncea Ait. 2n = 18. Illinois.” Jefferson Co.: S of Mt. Vernon, I-64, S & Su 9414. Effingham Co.: Shelbyville, 2.5 km S of Sigel, S & Su 9874, Massachu- setts. Bristol Co.: vicinity of Horseneck Beach Park, S 9/87. Plymouth Co.: E of Raynham, MA-104 just S of Lake Nippenicket, S 9/82. S. fis oie Torr. & Gray. 2n = 9,,. Ontario. Bruce Co.: Bruce Peninsula, N er Bay, alvar below Cabot Head, S 10363. S. Janeane Mill. 2n = ca. 18,,.' South Carolina. Chesterfield Co.: US-1 just US-52, S & Su 9778. 2n = 36.' Massachusetts. Plymouth Co.: Center Hill Rd. 2.6 km N of MA-3A, S & Brouillet 3553. 2n = 27,,.' Rhode Island. Washington Co.: E of Charleston, W of Matunuck, near W end of Cards Pond Rd., S & Su 9544. 2n = 54.' Massachusetts. Plymouth Co.: W of Man- omet, Beaver Dam Rd. 0.9 km S of MA-3A, S & Su 9568 S. leavenworthii Torr. & Gray. 2n = 36. South Carolina. Allendale Co.: Yen- ome, US-278 S of Barnwell, S & Su 9807. S. lepida DC. 2n = 18. Ontario.2 Kenora Dist.: Peawanuck (S of old Winisk), Clements s.n.-1, Clements s.n.-2. S. ludoviciana (A. Gray) Small. 2n = 18. Texas.’ Fort Bend Co.: TX-36 5 km S of county line, just N of Farm Rd.-1489, S & Su 10074. 2n = 36. Texas. Grimes Co.: TX-105 2.6 km W of Plantersville, S & Su 10066. Polk Co.: US-190 E of Livingstone, 1.8 km E of Soda (16.4 km E of TX-146), S & Su 8 S. missouriensis Nutt. 2n = 9,,. South Dakota. Minnehaha Co.: NW of Hum- 244 Rhodora [Vol. 95 Table 1. Continued. boldt, SD-38 3.5 km W of SD-19, S & Xg 10182. 2n = 18. Nebraska. Lin- coln Co.: NE-97 24.2 km SE of Tyron, S, Su & Ah 9195, S. mollis Bartl. var. mollis. 2n = 18,,. South Dakota. Minnehaha Co.: NW of Humboldt, SD-38 3.5 km W of SD-19, S & Xg 10181. S. nemoralis Ait. ssp. decemflora (DC.) Brammall in Semple. 2n = 18,. South Dakota. Minnehaha Co.: NW of Humboldt, SD-38 3.5 km W of SD-19, S & Xg pee 81. S. nemoralis Ait. ssp. nemoralis. 2n = 9,,. North Carolina. Chowan Co.: NC-37 just N of Albermarle Sound Causeway, SE of Edenton, S & Su 9745. Jones Co.: US-17 ca. 5 km NE of Maysville, S & Su 9750. South Carolina. Berke- ley Co.: SC-45 ca. 5 km E of St. Stephen, S & Su 9790. 2n = 18. North Carolina. Guilford Co.: E of Greensborough, service rd. W of Rock Creek Dairy Rd. exit off I-40, S & Su 9713. 2n = 36. Kentucky. Pulaski Co.: KY- 1392 ca. 10 km S of Livingston, ara of Rockcastle R., S & Su 9609. S. nitida Torr. & Gray. 2n = 18. Louisiana. Vernon Parte N of Anacoco, US- 171 3.4 km N of LA-111, S & Su 10045. S. odora Ait. var. odora. 2n = 9,,. Louisiana.2 Webster Parish. LA-159 8.4 km S of Leton (LA-2), S & Su 10034. Massachusetts. Bristol Co.: vicinity of Horseneck Beach Park, S 9/86. New Jersey.? Salem Co.: W of Malaga, NJ- 55, S & Su 9499, South Carolina.? Colleton Co.: NW of Jacksonboro, NC-64 ca. 2.5 km from US-17, S & Su 9798. 2n = 18. Kentucky.’ Powell Co.: Boone Nat’! For., end of Tunnel Ridge Rd., S & Su 9460. Texas.? Newton Co.: NE of Burkeville, TX-63 2.5 km W of Sabine R., S & Su 10050. Roan Mt., S & Su 9655. Yancey Co.: Mt. Mitchell, midway along rd. to park, S & Su 9686. S. patula Muh. ex Willd. var. strictula Torr. & Gray.' 2n = 9,,. Mississippi. Harrison Co.: US-46 12 km N of Lyman, S & Su 10123. S. puberula Nutt. var. puberula. 2n = 9,,. Maine. Sagadahoc Co.: S of Bath, Meadowbrook Rd., woods above The Basin tidal basin, S 10376. York Co.: NW of Kennebunk, ME-35 1.1 km S of ME-111, Lyman, S & Su 9586. S. puberula Nutt. var. pulverulenta (Nutt.) Chapm. 2n = 9,,. North Carolina.’ Bladden Co.: 0.5 km SW of Carvers, S & Su 9771. South Carolina.? Barnwell W of Barnwell, US-278 just N of Salkehatchie R., S & Su 9813 S. pulchra Small.' 2n = 18,,. North Carolina. Brunswick Co.: NC-133 S of Or- ton Plantation, S & Su 9765. Onslow Co.: US-17 1.7 km N of Dixon (NC- 210), S & Su 9755. Pender Co.: Hampstead, NC-210 ca. 4.1 km W of US- 17, S & Su 9760 S. radula Nutt. 2n = 18. Louisiana. Natchitoches Parish: E of Ajax, LA-174 1.5 km SE of co. line, S & Su 10038. Missouri.2 Washington Co.: MO-8 6.6 km W of Potosi, S, Su & Ah 9388. S. rigida L. var. glabrata (Braun) Heard & Semple. 2n = 9,,. Texas. Dallas Co.: N of Cedar Hill, Cedar Hill Rd., opposite lot 1110, RR right-of-way, S & Su 7 S. rigida L. var. humilis (Porter) Heard & Semple. 2” = 18. Illinois. St. Clair 1993] Semple et al.—Chromosomes 245 Table 1. Continued. Co.: Fairview Heights, Community Park, S & Su 9413. Nebraska. Hooker Co.: N of Mullen, NE-97 by Middle Loup R., S, Su & Ah 9198. S. rupestris Raf. 2n = 36. Kentucky.” Robertson Co.: KY-617 0.5 km W of Piqua, banks of Blue Lick R., S & Su 9594. S. sempervirens L. var. mexicana (L.) hi 2n = 9,,. North Carolina. Beaufort Co.: S of Washington, just S of Pamlico R., S & Su 9749. 2n = 18. Florida Walton Co.: Freeport, US-331 just N BL: 20, S & Su 10133. S. sempervirens L. var. sempervirens. 2n = 18. Massachusetts. Bristol Co.: Somerset, N of town along Segreganset Estuary, S 9184. S. shortii Torr. & Gray. 2n = 18. Kentucky. Flemming Co.: S of town of Blue- licks, S banks of Bluelicks R., S & Su 9598. S. mae Nutt. var. speciosa. 2n = 18,,. South Carolina.’ Edgefield Co.: SC- 230 17 km N of county line, S & Su 9817 S. sphacelata Raf. 2n = 18. North eae Mitchell Co.: W of Bakersville, NC-226 1.5 km W of NC-80, S & Su 9649. S. stricta Ait. 2n = 9,,. North Carolina,’ — Co.: NC-133 W of Wil- mington, S & Su 9763. 2n = 18. South Carolina. Clarendon Co.: US-521 1.1 km SE of Wilson, S & Su 9786. Hampton Co.: US-601 22.4 km S of Hamp- ton, S & Su 9803. 2n = 36. Louisiana.” St. Tammany Parish: E of LaCombe, US-191 2.4 km E of LA-434, S & Su 10118. South Carolina.’ Colleton Co.: NW of Jacksonboro, NC-64 8.7 km from US-17, S & Su 9801. Texas.? Aus- tin Co.: TX-36 4.2 km N of Sealy (Farm Rd.-1094), S & Su 10071. Harrison Co.: N of oe vicinity of N. Maine St. exit off I-10, S & Su 10080. Orange Co.: NE of Vidor, TX-12 ca. 3 km NE of I-10, S & Su 10079. 2n = 54. Florida. — Co.: E of Ebro, FL-20 3 km E of FL-79, S & Su 10138. Walton Co.: FL-20 3.2 km E of Freeport (US-331N), S & Su 10134. North Carolina.2 NNW of Supply, NC-211 13.4 km SSE of Columbus Co. line, S & Su 9766. South Carolina. Berkeley Co.: NC-17 9.2 km SE of Jamestown (US-17alt.), S & Su 9796. Jasper Co.: NW of Ridgeland, SC-652 14.9 km SE of Pineland, S & Su 9805 S. tortifolia Ell. 2n = 18. Louisiana.” Sabine Parish: N of Many, LA-175 10.7 km N of US-171, S & Su 10040. South Carolina. Berkeley Co.: SC-45 ca. 5 km E of St. Stephen, Francis Marion Nat'l For., S & Su 9789. S. uliginosa Nutt. 2n = 18,,. Maine.? York Co.: Wells Beach, W of beach front, S & Su 9590. Massachusetts. Essex Co.: N of Newbury, 1.5 km W of Plum Is. R., opposite airport, S & Su 9576. New Jersey.’ Ocean Co.: Tuckerton, coastal cedar marsh E of town centre, S & Su 9524. Rhode Island.’ Washing- S of county line, S & Su 9964. Washington Co.: S of Fayetteville, US- 7 just S of AR-170, S & Su 9957. Minois. Kakankee Co.: NW of Bradley, Kakan- kee R. State Park, by Rocky R., S & Su 9870. Union Co.: Bald Knob Rd. N of Bald Knob, S & Su 9880, S & Su 9881. White Co.: S of Norris City, IL-1 60, S & Su 9906. Douglas Co.: MO-14 at Twin Bridges, woods by N. Fork of 246 Rhodora [Vol. 95 Table 1. Continued. White R., near MO-181, S & Su 9930. Hickory Co.: MO-64B W of Pitts- burg, S & Su 9940. Oregon Co.: N of Greer, MO-19 16.6 km S of US-60, vicinity of Mark Twain Nat’! For. Rd.-3174, S & Su 9918, S & Su 9919, S & Su 9920. Phelps Co.: § of Jerome, Co. Rd.-D at I-44 exit, S, Su & Ah 9385. Taney Co.: MO-76 11.8 km W of county line, S & Su 9934. Washington Co.: MO-8 6.6 km W of Potosi, S, Su & Ah 9389. Wayne Co.: Co. Rd.-F 0.1 km E of US-67 7.8 km S of Greenville, S & Su 9893; NE of Wappapello, MO-2 a few km NE of MO-D, S & Su 9886; E of Williamsville, MO-49 2.5 km E of Co. Rd.-A, S & Su 9900. gee Sse — Reveal & Keener var. ericoides. 2n = 10. Illinois. St. lai : Fairview Heights, Community Park, bluff of Mississippi R., S & Su 0408. Soe Grimes Co.: E of Navasota, TX-105 1.5 km E of Farm Rd.- 362, S & Su 10067. 2n = 10. Texas. Dallas Co.: N of Cedar Hill, area oppo- - 1110 Cedar Hill Rd., S & Su 10006. 2n = 10 + 0-3 supernumeraries. . Lamar Co.: S of Powderly, US-271 N of Paris, S & Su /0001. Vz perce (Nutt.) Reveal & Keener. 2n = 10. Nebraska. Hooker Co.: N of Mullen, bluff of Middle Loop R., just E of NE-97, S, Su & Ah 9197. 2n = Logan Co.: Magazine Mt., top of Cameron Bluff, S & Su 9970. Missouri. Oregon Co.: N of Greer, MO-19 just N of For. Rd.-3188 and Elev- en Point R., limestone bluff glade, S & Su 992]. Washington Co.: MO-21 NE of Potosi, just N of MO-8, S, Su & Ah 9398; MO-8 6.6 km W of Potosi, S, Su & Ah 9394 V. patens (Soland. in Ait.) Reveal & Keener var. gracilis (Hook.) Reveal & n= 5,,. Texas. Harrison Co.: NE of Marshall, TX-43 7.2 km NE of US-59, S & Su 10025. V. patens (Soland. in Ait.) Reveal & Keener var. patens. 2n = 20. Texas. Fort N of Orchard, TX-36 5 km S of county line, N of Farm Rd.-1489, S & Su 10073 V. patens (Soland. in Ait.) Reveal & Keener var. patentissimus (Torr. & Gray) veal & Keener. 2m = 20. Missouri. Bollinger Co.: 1.2 km W of Arab, S & 85. V. seats aso in Ait.) Reveal & Keener var. phlogifolius (Muhl.) Reveal & = 20. Ilinois.2 Hamilton Co.: 7.5 km E of McLeansboro at Co. Rd.- "L400E. S & Su 9421. Kentucky. Pulaski Co.: KY-461 7 km NE of Shop- ville, S & Su 945] sachusetts; S. patula Muhl. ex. Willd. var. strictula Torr. & Gray, 2n = 9,, (one population) from Mississippi; and S. pulchra Small, = 18,,(3 populations) from North Carolina. Many of the counts (99) represent the first report(s) for a taxon, or ploidy level of a taxon, for a particular province or state, based on data compiled on nearly 7000 chromosome number reports of North American asters and goldenrods (Semple, 1992): e.g., Aster puniceus L. (2n = 16) from Georgia, Heterotheca camporum (Greene) Shinners 1993] Semple et al.—Chromosomes 247 Table 2. Summary of chromosome number reports by genus by area. The numbers of counts representing first report(s) of a taxon, or ploidy level within a taxon, for each region are given as subscripts. Virgu- Eutha- Other Region Total Aster lus mia Solidago Genera Canada Ontario 9 5 —_ _ 3; 1 United States Alabama 6 4, - _ 1, 1, Arkansas 11 Ty 1 =- 2 1 Connecticut 1 1, - _ _ = rida 5 | _ _ 3 l Georgia 2 2 _ = - on Tilinois 25 11 Bp - 10, Indiana 3 a _ _ l - Kentucky 37 20, 1 _ 16, — Louisiana 14 6; = ~ 6; Ds aine 8 a _ _ 4, _ Massachusetts i) 6, — _ 8, 1 Minnesota 2 2 — - — a Mississippi 2 | 6, a - 5, - Missouri 35 b3, 4 _ LT, A Nebraska 4 ~ 1 ~ 3 =_ New Jersey 11 F - a 4, 2 New York - — — 1 _ North Carolina 49 ee — _ 26, 7 Oklahoma 6 6; = = _ —_ Pennsylvania 3 fr — 2; - Rhode Island 2 - — — 2. — South Carolina 29 3; = ] Le 2 South Dakota 5 = — — 3 —_ Tennessee 7 6, — aot 1, a Texas 36 24 4 5 12, 4 Wisconsin i _ _ es _ TOTALS 333 142,, 14, 5 147,, 25; var. glandulissimum Semple (2n = 36) from Alabama, Solidago lepida DC. (2n = 18) from Ontario, S. flexicaulis L. (2n = 90) from Kentucky, S. fistulosa Mill. (2n = 18) from Alabama, S. odora Ait. var. odora (2n = 94) from New Jersey and S. puberula Nutt. var. pulverulenta (Nutt.) Chapm. (22 = 9,) from South Carolina. The number of reports and the number of first reports for each province and state are summarized in Table 2 and each case is identified in Table 1. First reports are listed for the fol- 248 Rhodora [Vol. 95 lowing political regions: Alabama, Arkansas, Connecticut, Geor- gia, Illinois, Kentucky, Louisiana, Maine, Massachusetts, Mis- sissippi, Missouri, Nebraska, New Jersey, North Carolina, Oklahoma, Ontario, Pennsylvania, Rhode Island, South Carolina, Tennessee and Texas. ASTER PILOSUS COMPLEX The Aster pilosus Willd. complex includes all of the eastern North American members of Aster sect. Dumosi Torr. & Gray subsect. Porteriani (Rydb.) Semple: A. depauperatus (Porter) Fern., A. parviceps (Burgess) Mackenzie & Bush, A. pilosus var. pilosus, A, pilosus var. pringlei (A. Gray) Blake, and A. priceae Britt. Included in Table 1 are 14 counts representing all eastern mem- bers of the complex, whose cytogeography was examined in some detail earlier by Semple and Chmielewski (1985). Special note should be taken of the counts for four collections from Missouri hesitantly treated in Table 1 as A. aff. parviceps and A. aff. pilosus var. pringlei. One specimen (Semple et al. 9398) was tetraploid (2n = 32) and was from a cedar glade habitat; three collections were hexaploid (2n = 48) and also were from cedar glades. The specimens were glabrous to very sparsely pubescent, which is atypical for A. parviceps. Some of the glabrous hexaploid indi- viduals were very similar in growth form and general appearance to the octoploid A. priceae (Semple et al., 1989), which is endemic to a small region in southern Kentucky (Table 1, Semple & Suripto 9438 & 9440-1), central Tennessee and northern Alabama and Georgia, but the Missouri plants had white rays rather than the blue usual for A. priceae along with smaller heads and floral parts. Other individuals from the same cedar glades were slightly pu- bescent. The hexaploid level and glabrous condition is typical of A. pilosus var. pringlei, which is not previously reported for the Ozark region. Tetraploids are unknown in var. pringlei. Thus, the four collections from Missouri do not fit well into any of the three taxa to which they might be assigned. They cannot be hybrids between A. pilosus and another species in a different section or subsection because they have phyllaries and lower stem and ro- sette leaves typical of the subsection; these traits are modified in hybrids. Following Steyermark (1963), the collections could be assigned to A. pilosus var. demotus Blake, but the type of this 1993] Semple et al.—Chromosomes 249 taxon is just a robust var. pringlei from southeastern Virginia (Semple and Chmielewski, 1985). If the collections are deter- mined to be more correctly assigned to A. pilosus var. pringlei than to A. parviceps, then the range of the former must be ex- panded to include disjunct populations in the northern Ozarks, and one of these would be the only known tetraploid member of the variety. A more definitive way of distinguishing members of the complex is needed than the usual approach based on head and floret size and the degree of upper stem pubescence, because these break down in the case of these glabrous-glabrate Missouri plants. The circumscription of either A. parviceps, A. pilosus var. pringlei or A. priceae will have to be modified to accommodate these cedar glade individuals. A detailed multivariate morpho- metric analysis is being undertaken to resolve the problem. SOLIDAGO STRICTA COMPLEX The Solidago stricta complex in the southeastern United States includes S. austrina Small, S. flavovirens Chapm., S. gracillima Torr. & Gray, S. perlonga Fern., S. pulchra Small, and S. stricta Ait. Cronquist (1980) distinguished individual species on the basis of rhizome, capitulescence and capitulum traits. Solidago stricta was distinguished by its elongate, stoloniferous rhizomes, which are lacking in S. pulchra and S. gracillima (including S. austrina, S. flavovirens, and S. perlonga). S dS. gracillima were distinguished on the ‘basis of stem height, the number of heads, and the number of florets per head; the widely distributed S. gracillima was tall with few florets per head, and the narrowly tributed S. pulchra was short with a few heads with many florets per head. Radford et al. (1968) treated the complex differently, merging S. austrina and S. pulchra into S. stricta, and they noted that the rare, narrowly distributed S. gracillima with more widely spreading capitulescence branches might be conspecific with S. Stricta. Three diploid, two tetraploid and three hexaploid reports (2n = 18, 2n = 36, and 2n = 54, respectively) have been published previously for members of the complex (Beaudry, 1963; Semple et al., 1981, 1984). With the addition of 17 counts for members of the complex reported here in Table 1, a tentative geographic pattern can be recognized now in S. stricta. Solidago stricta was 250 Rhodora [Vol. 95 found to be diploid in North and South Carolina, tetraploid in South Carolina, Florida, Louisiana and Texas, and hexaploid in New Jersey, North and South Carolina, and Florida. Thus, the western portion of the range (in the United States at least) is apparently occupied only by tetraploids, but from the Florida Panhandle northward all three ploidy levels may be encountered, with the hexaploid level being most frequently reported to date. Only a few counts have been reported for other members of the complex. For Solidago austrina, there is one diploid report (North Carolina, Beaudry, 1963) and one hexaploid report pub- lished as “‘S. stricta (= S. austrina)’ (North Carolina, Semple et al., 1984). The latter is probably just S. stricta, not S. austrina. Two diploid reports are known for S. gracillima (Alabama, Beaudry, 1963; South Carolina, Table 1 here). Solidago pulchra (sensu Cronquist) is known from only a few coastal counties in North and South Carolina and was previously unknown cytologically. It was found to be tetraploid (2n = 18,) at three populations in separate counties in North Carolina (Table 1). Our plants were clearly different from any of the S. stricta collections, being much shorter, having heads with a campanu- late-hemispherical versus cylindrical shaped involucres and bright yellow versus yellow-green phyllaries, and more ray and disc flo- rets per head. Differences in habitat preference may also exist, but a number of our collections came from disturbed areas whose original character was unclear; additional data are needed to clar- ify this distinction. Because stem height in goldenrods is a very plastic trait and because ploidy level is known to influence involucre height and the numbers of florets per head (Beaudry and Chabot, 1957; Beaudry et al., 1958; Beaudry, 1960, 1970; Melville and Morton, 1982; Heard and Semple, 1988; Semple et al., 1990), separating taxa in the S. stricta complex on the basis of these traits may no lead to a useful classification. It is clear from the chromosome counts now reported that at least one of the taxa in the complex includes more than one ploidy level and that diploids in all the taxa occur on the coastal plain in the Carolinas. A detailed cy- togeographic and multi ate of the com- plex is needed to determine the distribution of cytotypes through- out the range of each taxon and to determine how many of these taxa warrant recognition. 1993] Semple et al.—Chromosomes 251 SOLIDAGO ULIGINOSA The cytogeography of Solidago uliginosa was examined by Chmielewski et al. (1987), who reviewed previous reports (Beau- dry et al., 1958; Beaudry and Chabot, 1959; Beaudry, 1969: Ka- poor, 1970, 1977; Morton, 1981; Semple, 1981; Semple et al., 1981, 1984; Léve and Léve, 1982; Chmielewski, 1985). In those studies, diploids were common across the range of the species in eastern Canada and the northeastern United States. In contrast, tetraploids were reported from several disjunct areas: limestone pavements around northern Lake Huron in Ontario (five sites) and northern Lake Michigan in Michigan (two sites), near James Bay in northwestern Québec (Beaudry, 1969; one site), west of Montréal in southern Québec (Beaudry, 1969; one site) and the northeastern United States (Beaudry and Chabot, 1959; one site each in eastern Massachusetts and southeastern Pennsylvania; Beaudry, 1963; one site each in central Massachusetts and north- ern West Virginia). Since 1988 two additional diploid counts have been reported (Semple et al., 1989). To these previous reports we add four tetraploid counts here (Table 1; one each from near the coast in New Jersey, Rhode Island, Massachusetts, and Maine). To date, all counts from the southeastern portion of the species’s range are tetraploid. Thus, tetraploids are the dominant cytotype in two disjunct areas (one east and one west of the Appalachian Mountains) and occur rarely in several other parts of the range. Diploids are unknown in the southeastern portion of the range of the species. ACKNOWLEDGMENTS This work was supported by Natural Sciences and Engineering Research Council of Canada Operating Grants to Semple. We thank the following people for their assistance in the field: Tufail Ahmed, Jerry Chmielewski, Jeff Semple and Bambang Agus Sur- ipto. Viable rootstocks provided by Blair Clements from near Hudson Bay in northern Ontario were very much appreciated. LITERATURE CITED Beaupry, J. R. 1960. Studies on Solidago L. V. The Solidago rugosa complex. Canad. J. Genet. Cytol. 2: 389-396 252 Rhodora [Vol. 95 1963. Studies on Solidago L. V1. Additional eee numbers of taxa of the genus Solidago. Canad. J. Genet. Cytol. 5: 174, . 1969. Etudes sur les Solidago IX. Une sare liste de nombres chromosomiques des taxons du genre Solidago et de certains genres voisins. Naturaliste Canad. 97: 431-445. 1970. Etudes sur les Solidago. X. Le Solidago gigantea Ait. dans lest de l’Amérique du Nord. Naturaliste Canad. 97: 35-4 AND D. L. CHasBot. 1957. Studies on Solidago L.—I. S. altissima L. and S. canadensis L. Contrib. Inst. Bot. Univ. Montreal. 70: 65-7 AND 959. Studies on Solidago IV. The cheamononee numbers of certain —_ of the genus Solidago. Canad. J. Bot. 37: 209-288. , A. ZINGER AND J. SAINT-PIERRE. 1958. Etudes sur la genre Solidago L. II. la mise en évidence de la polyploidie endotaxonomique chez S. purshii Porter et les taxa similaires. Canad. J. Bot. 36: 663-670 CHMIELEWSKI, J. G. 1985. Documented plant chromosome numbers 1985: 1. Miscellaneous counts from Ontario and Quebec. SIDA 1: 251-253. , R. S. Rinatus AND J. C. SeMpLe. 1987. The cytogeography of Solidago ee Nutt. onsale Astereae) in the Great Lakes region. Canad. J. Bot. 65: 1045-1046. Cronqutst, A. 1980. Vascular Flora of the Southeastern United States. I. As- teraceae. Univ. North Carolina en Chapel Hi Il, NC. HEARD, S. B. AND J.C. SEMPLE. 1788. S. 2 plex (Compositae: Astereae): a multivariate tri io is and cl b Canad. J. Bot. 66: 1800-1 1807. Kapoor, B. M. 1970. Jn IOPB chromosome number reports XX VII. Taxon 19: 437-442. . 1977. Further ot i tk phology of some Solidago species. Cytologia 42: 241-253. Ove, A. AND D. Léve. 1982. Jn IOPB chromosome number reports. LXXV. Taxon 31: 344-360. MELVILLE, M.R. AnD J. K. Morton. 1982. A biosystematic study of the Solidago canadensis (Compositae) complex. I. The Ontario populations. Canad. J. Bot. 60: 976-997. Morton, J. K. 1981. Chromosome numbers in Compositae from Canada and the U.S.A. Bot. J. Linn. Soc. 82: 357-368. RaprorD, A. E., H. E. AHLEs AND C. R. BELL. 1968. Manual of the Vascular Flora of the Carolinas. University of North Carolina Press, Chapel Hill, NC. aio J.C. 1981. Misc. chromosome counts. Jn Chromosome number reports XXII. Taxon 30: 703-704 ——.. 1985. Chromosome number determinations in Fam. Compositae, Tribe Astereae. Rhodora 87: 517-527. . 1992. A geographi f ck rts for North (Compositac: Astereae). hac Missouri Bot. Am : + “a yf | 3 Garden 79: 95-109. _ . CHMIELEWSKI. 1981. Chromosome numbers of goldenrods, nn and Solidago, (Compositae-Astereae). Canad. J. Bot. 59: 1167-1173. ——— AND J. G. Stay 1985. The cytogeography of Aster pilosus (Com- 1993] Semple et al.—Chromosomes 253 positae-Astereae). II. Survey of the range, with notes on A. depauperatus, A. Pparviceps and A. sake "ech 87: 367-379. AND 87; omosome numbers in Fam. Compositae, Tribe Astereae. II. por tiie oe Rhodora 89: 319-325. . CHMIELEWSKI AND R. A. B MALL. 1990. A multivariate study of Solidago nemoralis (Compositae: Astereae) and comparison with S. cal- ifornica and S. sparsiflora. Canad. J. Bot. 68: 2070-2082. AND CHUNSHENG XIANG. 1992. Chromosome numbers in Fam. Casiponitac, Tribe Astereae. IV. Additional reports and comments on the cytogeography and status of some species of Aster and Solidago. Rhodora 94: 48-62 1989. Chromosome numbers in Fam. Com- — Tribe Astereae. IIL. Additional counts and comments on generic limits and ancestral base numbers. Rhodora 91: 296-314 . RINGIus, C. LEEDE R AND G. Morton. 1984. Cieceniaceie numbers Gf goldentods AY " e). II. Additional counts with comments on cytogeography. Brittonia 36: 280-292. A. 1963. Flora of Missouri. Iowa State University Press, Ames, IA. = & Q9 = > FE Ko] 5 2] 2° *) > a f) A 4 2 » DEPARTMENT OF BIOLOGY UNIVERSITY OF WATERLOO WATERLOO, ONTARIO, CANADA N2L 3G1 RHODORA, Vol. 95, No. 883/884, pp. 254-260, 1993 INTRASPECIFIC TAXONOMY AND COMPARISONS OF NRDNA ITS-2 SEQUENCES OF ARISAEMA RINGENS (ARACEAE) SUNG CHUL Ko,! STEVE L. O’KANE, JR., AND BARBARA A. SCHAAL ABSTRACT The second internal transcribed spacer (ITS-2) of ribosomal DNA (rDNA) was sequenced for the three phenotypes of Arisaema ringens (Araceae). These se- quences were invariant, and are therefore not systematically informative. How- ever, when considered along with the anatomical, morphological, cytological, and palynological data summarized here, the recognition of intraspecific taxa in A. ringens is not supported. Key Words: Arisaema, nrDNA, ribosomal DNA, transcribed spacer, ITS-2 INTRODUCTION Arisaema ringens (Thunberg) Schott (section Pistillata; Ara- ceae), Japan, China and S. Korea, was divided by Engler (1879) into two varieties, a sieboldii and 8 praecox, based on cataphyll and petiole color. The former has green cataphylls and pale violet petioles, and the latter has red cataphylls and pale green petioles. Workers have had varying opinions of these taxa. Palibin (1901) and Chung (1957) agreed with Engler in maintaining two varieties. Other taxonomists have viewed these phenotypes as different species (e.g., De Vriese, 1839; Nakai, 1952) or, more recently, united them into a single species (e.g., Ohwi and Kitagawa, 1983; Ohashi and Murata, 1980; Wu and Li, 1979). Koyama (1984) made Engler’s varieties into forms based on spathe color rather than cataphyll and petiole color, i.e., A. ringens f. sieboldii with dark purple spathes and A. ringens f. praecox with greenish spathes. Ko and Kim (1985) agreed with Koyama. In the field three phenotypes of Arisaema ringens are found. Based on the current intraspecific taxonomy, these phenotypes may be separated as follows: individuals with purple spathe mar- gins and petioles [var. sieboldii (de Vriese) Engl.], individuals with purple spathe margins and green petioles [var. praecox (de Vriese) Current address: Dept. of Biology, Han-nam University, 133 O-jung dong Tae-jun, Korea. 254 1993] Ko et al.— Arisaema ringens 255 Engl.], and individuals with greenish spathe margins and green petioles (f. praecox T. Koyama). If, however, petiole color is rec- ognized as a good taxonomic character, then characters of var. sieboldii and var. praecox are found within f. sieboldii. Several recent studies have examined the intraspecific system- atics of Arisaema ringens. Morphology and geographical distri- bution are treated in Ko and Kim (1985). Cytology, anatomy and palynology have also been studied (Ko et al., 1990; Oh et al., 1990; Ko et al., 1987; Ko and Kim, 1985; Hotta, 1971; and Ito, 1942). In order to further clarify the intraspecific relationships in A. ringens we examined the sequence of one of the internal spacers of large subunit rDNA (ITS-2). DNA sequences have proved extremely useful in other taxonomic studies (Crawford, 1990). The coding regions of rDNA (18S, 5.8S, and 28S) are not useful at the intraspecific level because they evolve very slowly in se- quence and almost not at all in length. These sequences are par- ticularly useful at the level of family and above (Baverstock and Johnson, 1990; Schaal and Learn, 1988). The intergenic spacer (IGS) evolves rapidly both in sequence and length, making it difficult to use except at the population level (Learn and Schaal, 1987; Jorgensen and Cluster, 1988). The ITS regions (ITS-1 and ITS-2), however, have an intermediate level of variation (Bald- win, 1992a; Schaal and Learn, 1988; Hamby and Zimmer, 1992) that makes them ideal for use at or below the generic level. No studies have reported ITS sequences at the subspecific level; at the specific level, percent sequence divergence ranges from 0.4% to 12.9%, excluding the single value of 0% for variation within Madia bolanderi (Baldwin, 1992a). METHODS Plants were grown in the Washington University greenhouse from corms collected in the Jeon-La-Nam-Do Province of Korea. Plants with purple spathes and purple petioles and peduncles were from near Dae-Heung Temple, Hae-Nam Kum. Plants with green or purple spathes and green petioles and peduncles were from Kun-Oe Myun, Wan-Do (Island) Kun. DNA was extracted from fresh tissue using the CTAB procedure of Doyle and Doyle (1987) with the addition of a phenol extraction. Primers “ITS-3” (GCATCGATGAAGAACGTAGC) and “ITS-4” (TCCTCC- GCTTATTGATATGC) for PCR amplification of ITS-2 were 256 Rhodora [Vol. 95 5.8S gtgaattgca gaatcccgtg aaccatcgaa tctttgaacg acagttgcgce ccgaggcctc ar I1S2 taggtcgagg gcacgcctgc ctgggcGTCA CGCCCTACGT CGCTCCCTGA CCCCCCCATA GAGTGTGGGG GGTGTTGAGG GATGCGGAGA TTGGCCCACC GTGCACGTGC GCGCAGGTTG AAGAACTCGA CCCTCCTGCC GGGCGATTAA CGGCGAGTGG TGGACGATGC TCATCGTCGC CGTAGTGCAC GCCCGCGCGT AAGGATGGGT TGACTGTgag ggaacccaat catcggagag Tr 28S acgatcgta tcttaaagat agggtagctc tttgatcgcg accccaggtc aggcggggcc egee: 3" Figure 1. Nucleotide sequence of the rDNA ITS-2 region of the three phe- notypes of Arisaema ringens including the 28S rDNA flanking region and most of the 5.8S rDNA. Overlap sequence from primers 3 and 4 is indicated in capitals. constructed from the consensus sequences reported by White et al. (1990). The fourth base (C) from the 5’ end of primer “ITS- 3” was changed to T to match the sequences for flowering plants reported elsewhere (e.g., Yokata et al., 1989). Both strands of the double stranded PCR products were directly sequenced using Se- quenase (US Biochemicals). Sequence overlap was obtained for 52.5 percent of the total sequence and 75.4 percent of ITS-2 (Figure 1). Sequencing protocols are essentially those given by US Biochemicals except that labeling and extension reactions were performed in a single step, Sequenase was diluted 1:6 rather than 1:8, and termination reactions were carried out at 49°C. The entire sequence of ITS-2 and most of the 5.88 rDNA was sequenced for one individual of each phenotype. RESULTS _ The sequences obtained for the three phenotypes were identical, indicating to us that no further sequences need be sampled for Arisaema ringens, especially as the three samples are from two different locations. ITS-2 was found to be 244 bases long with a G+C content of 61.5 percent (Figure 1). These values are similar to other reported ITS-2 sequences (Yokata et al., 1989; Kavanagh and Timmis, 1988; Takaiwa et al., 1985). The location of the ends of the 5.8S and 26S sequences are based on comparisons 1993] Ko et al.—Arisaema ringens 257 Table 1. Summary of characters of Arisaema ringens var. praecox, var. sie- boldii, and f. praecox. Taxon praecox sieboldii praecox Morphology Peduncle green purple green Petiole green purple green purple purple green Cytology! Chromosome no. (2n) 28 28 _ Karyotype Relative size Very long 1 1 - Long 9 8 = Medium ] 2 — ma 3 3 _ Total chromosomal length (um) Ave. 6.57 a3) - Range 3.95-8.83 2.07-4.35 = ype Metacentric 2 3 - Submetacentric 12 10 = Subtelocentric 1 — Anatomy—Cell wall specialization index? Peduncle 2.10 1.50 Petiole SAS 2.90 _ 00 3.90 4.85 —_ Ave. 3.20 3.08 - ‘Ko, S.C. & ¥. §. Kim, 1985. ? Ko, S.C. et al., 1990. with published ITS-2 sequences (Rathgeber and Capesius, 1989; Yokata et al., 1989; Kavanagh and Timmis, 1988; Kiss et al., 1988; Takaiwa et al., 1985). DISCUSSION AND CONCLUSIONS The intraspecific taxonomy of Arisaema, as noted previously, is currently confused, especially so if we wish to recognize the three variable but non-overlapping phenotypes recognizable in the field. In addition the names of these phenotypes are non- heirarchical, i.e., var. praecox, var. sieboldii, and f. praecox. Table 1 summarizes the anatomy, cytology, and morphology of the 258 Rhodora [Vol. 95 intraspecific taxa of Arisaema ringens. Forma praecox has not been the subject of previous studies other than morphology. Cytologically, Arisaema ringens var. praecox and var. sieboldii are very similar, both having 2” = 28 chromosomes and nearly identical karyotypes (Table 1). Although they share a pair of chro- mosomes that is slightly different from one another in length and chromosome type, these differences are somewhat subjective, e.g., chromosome length is dependent on tissue condition and age of tissue as well as parameters of staining. Geographically these taxa occur in the same subtropical to warm temperate region. Although electron density of the pollen ektexine of var. sieboldii is somewhat greater than in var. praecox, there are no significant differences in pollen size, shape, and wall architecture. Cross-sections of roots, petioles, petiolules, peduncles, and leaves show the same tissue arrangement and level of tissue specialization and differentiation in both taxa. Diameter and cell wall thickness of tracheids, and diameter and length of fibers in both taxa are not significantly different, and these characters have been shown not be phylo- genetically useful in Arisaema (Ko et al., 1990). Based on the level of specialization of cell wall thickenings, var. sieboldii is slightly more advanced in its root anatomy and var. praecox is slightly more advanced in areal organs. Although the distribution of colors of peduncle, petiole, and spathe are not overlapping, they are quite variable within a given taxon. Many morphological characters of plants, some of them quite striking, have been shown to be under the control of alleles of one or a few genes (Coen and Meyerowitz, 1991; Gottlieb, 1984; Hilu, 1983). It is probable that colors thought to distinguish in- traspecific taxa of Arisaema ringens are similarly controlled and represent intraspecific polymorphism rather than synapomor- phies of evolutionary lineages. Sequences of ITS-2 show no dif- ferences among the intraspecific taxa of Arisaema ringens. These sequences are known to be good markers of evolutionary lineages at or near the species level (Baldwin, 1992b; Baldwin et al., 1992; Suh et al., 1992). Insignificant difference in morphology and anat- omy, and the lack of ITS-2 sequence variation, suggests that the intraspecific taxa of Arisaema ringens should be consolidated and a single polymorphic taxon should be recognized. Further work is required to assay levels of ITS-2, as well as ITS-1, variation within and among other species of Arisaema. 1993] Ko et al.—Arisaema ringens 259 LITERATURE CITED BaLpwin, B. G. 1992a. Phylogenetic utility of the internal transcribed spacers of nuclear ribosomal DNA in plants: an example from the Compositae. Mol. Phylog. Evol. 1: 3-16. 1992b. ene of ‘Calycadenia oa poe Madiinae) based on nuclear 18-26S r p ITS) sequences. Amer. J. Bot. beikeiaips | . RopicHaux, G. D. Carr AND D. W. Kyuos. 1992. Evolutionary and biogeographic patterns in the Hawaiian silversword alliance (Composi- tae-Madiinae). Amer. J. Bot. 79(Supp.): eg stract). BAVERSTOCK, P. R. AND A. M. JOHNSON. 1990. Ribosomal RNA nucleotide sequence: a comparison of new methods i for its determination, and its use in phylogenetic analysis. — Syst. Bot. 3: 101-110. Coen, E. S. AND E. M. MEYEROWwr 1991. The war of the whorls: genetic interactions controlling flower Pia SRS Nature 353: 31-37. Cuuna, T. H. 1957. Korean Flora II. Shinzisa, Seoul. Pp. 776-779. CRAwForD, D. J. 1990. Plant Molecular Systematics. Wiley and Sons, New York ork. De Vrigse, W. H. 1839. Hortus spaaren-bergensis. I. /n: A. Engler DC. Mono- graphiae Phanerogamarum 2: 533-560. Doyte, J. J. AnD J. L. Doyte. 1987. A rapid DNA are ‘Saeed for small quantities of fresh leaf _— Phytochem. Bull. 19: ENGLER, A. 1879. Arisaema. In: DC., Monographiae ‘saith Masson, P. 0 Gorttuies, L. D. 1984. Genetics and mophological evolution in plants. Amer. Naturalist 123: 681-709. Hampy, R. K. AND E. A. ZimMER. 1992. Ribosomal RNA a in plant systematics. Jn: P. S. Soltis, D. E. Soltis ph 1 J. Doyle, si Molecular Systematics of Baise: Chapwaan and Hall, se! York. Ep: 50-91 Hitu, K. W. 1983. gle-g g ones 7 Biol. 16: 97- Hotta, M. 1971. Study of the family Araceae, general remarks. Jap. J. Bot. 20: 269-310. Ito, T. 1942. Chromosomen und Sexualitét vonder Araceae, I. Somatische Chromosomen-Zahlen einiger Arten. Cytologia 12: 313-3 JORGENSEN, R. A. AND P. D. CLuster. 1988. Modes and tempos in = theevoludon of nuclear ribosomal DNA: new characters for evolutionary studies and new markers for genetic population studies. Ann. Missouri Bot. Gard. 75: 1238 1247 Kavknane T. A. AND J. N. Tiwmis. 1988. Structure of melon rDNA and nucleotide sequence of the 17-25S spacer region. Theor. Appl. Genet. 76: on Kiss, T., M. Kis, S. ABEL AND F. Sotymosy. 1988. Nucleotide sequence of the 17S-25S spacer region from tomato rDNA. Nuc. Acids Res. 16: 7179. Ko, S.C. AND Y. S. Kim. 1985. A taxonomic study of genus Arisaema in Korea. Korean ds cig Tax. 15: 67-109. B _S. LEE AND Y. S. Kim. 1990. A phylogenetic study of > 260 Rhodora [Vol. 95 Arisaema by anatomical ag palrnokigual characters. I. Anatomical char- acters. Korean J. Plant T: 35. «Ko Hy TAE; FSO: ee S. Kim. 1987. A cytotaxonomic study n some species of Arisaema Korean J. Plant Tax. 17: 189-205. KoyaMA, T. 1984. Arisaema, pp. 256-262. In: F. G. Meyer and E. H. Walker, Eds., J. Ohwi, Flora of Japan. Smithsonian Inst. Washington, D.C. Learn, G. H. AND B. A. SCHAAL. 1987. Population subdivision for ribosomal DNA repeat variants in Clematis fremontii. Evol. 41: 433-438. Nakal, T. 1952. A synoptical sketch of Korean flora. Bull. Nat. Sci. Mus. 31: 1-152. On, B. U., S. C. Ko, W. P. HoNG AND Y. S. Kim. 1990. A phylogenetic consid- eration of Arisaema by anatomical and palynological characters. II. Paly- nological characters. Korean J. Plant Tax. 20: 37-42. Onasut, H. AND J. MuraTA. 1980. Taxonomy of the Japanese Arisaema. J. Fac. Sci. Univ. Tokyo, Sect. S, Bot. 12: 281-336. Oxnwi, J AND M. KrraGAwa. 1983. New Flora of Japan. Shibundo, Tokyo. Pp. PALIBIN, J. 1901. Conspectus Flora Korea III. Act. Hort. Petrop. 19: 101-151. RATHGEBER, J. ANDI. Capestus. 1989. Nucleotide sequence of the 18S-25S spacer region from mustard DNA. Nucl. Acids Res. 17: 7522. SCHAAL, B. A. AND G. H. LEARN. 1988. Ribosomal DNA variation within and among plant populations. Ann. Missouri Bot. Gard. 75: 1207-1216 Sun, Y, L. B. THEIN AND E. A. ZIMMER. 1992. Molecular evolution of the internal transcribed spacers of nuclear ribosomal DNA and a phylogeny of the Win- teraceae. Amer. J. Bot. 79(Supp.): shi aagroeemckam TAKAIWA, F. OHARU AND M. SucruraA. 1985. Nucleotid nce of the 178-258 pei region from rice rDNA. PI. Mol. Biol. 4: 355- 364. Wuirte, T. J., T. Bruns, S. LEE AND J. TAyLor. 1990. Amplification and direct es of fungal ribosomal RNA genes for phylogenetics, pp. 315-322. n: M. A. Innis, D. H. Gelfand, J. J. Sninsky and T. J. White, Eds., PCR ete Academic Press, New York. AND H. Li. 1979. Flora Reipublicae Popularis Sinicae. Institutum Bicteniaten Kummingense Academiae Sinicae. Tormus 13: 116-194. YoKATA, Y., T. KAwWATA, Y. ImtpA, A. KATO AND S. TANIFUJI. ae Nucleotide sequences of the 5.88 rRNA gene and internal transcribed spacer regions in carrot and broad bean ribosomal DNA. J. Mol. Evol. 29: 294-301. pinilammiccaiel OF BIOLOGY, CAMPUS BOX 1137 YN UNIVERSITY ST. LOUIS, MO 63130 RHODORA, Vol. 95, No. 883/884, pp. 261-276, 1993 ANTENNARIA PULVINATA GREENE: THE LEGITIMATE NAME FOR 4A. AROMATICA EVERT (ASTERACEAE: INULEAE) JERRY G. CHMIELEWSKI ABSTRACT Canonical variates analysis was used as an analytical technique to document morphological discontinuities among individuals of Antennaria media (n = 63), A, pulvinata (n = 103), A. rosea (n = 64) and A. umbrinella (n = 65). Evaluation of the defined classification criterion indicated that 94% of the specimens were classified correctly. The classification criterion was subsequently used to classify type collections of A. aromatica (n = 26) into one of the previously « defined — Results based on these analyses as well as § previously p onstrate that A. aromaticaand A. 1 li The analyses also indicate that ra pulvinata and . rosea exhibit morphological integrity relative to each other and would best be treated as distinct species. The four species, A. media, A. pulvinata, A. rosea and A. umbrinella, exhibit morpho- logical integrity and should be treated as distinct. A list of synonymy is provided or A. pulvinata Key Words: Ant 7 dia, A. pulvinata, A. rosea, A. umbrinella, A. aromatica INTRODUCTION Antennaria aromatica Evert was originally described as a sexual diploid Cordilleran species (Evert, 1984). Subsequent studies have demonstrated that pa pane iS ‘morphologically, cytologically tl ariable (Bayer and Stebbins, 1987; Bay- or; 1984, 1989a), as seit as more widely distributed (Chmielewski and Chinnappa, 1988a; Bayer, 1989a), than initially proposed. Additional interest in A. aromatica stems from the implication that the species is one of the progenitors of the A. rosea Greene polyploid complex, specifically, that entity described as A. rosea subsp. pulvinata (Greene) R. Bayer (Bayer, 1989b). Evert (1984) was able to differentiate Antennaria aromatica from both A. media Greene and A. umbrinella Rydberg, other members of the complex, by its copious and persistent glandu- losity, widely cuneate-spatulate basal leaves and distinctive cit- ronella-like odor. Bayer (1989b) concluded that his subsp. pul- vinata resembled A. aromatica, except in that it lacked glandular hairs. Although Greene (1898) described both “female” and “male” specimens of A. pulvinata, Bayer (1989b) concluded that his subsp. pu/vinata is entirely pistillate, apomictic and polyploid. 261 262 Rhodora [Vol. 95 Antennaria aromatica was initially considered endemic to southwestern Montana and adjacent parts of Idaho and Wyoming sit ts limestone talus and xeric habitats (Evert, 1984). The spec tly reported from Alberta, British Columbia and California (Chmielewski and Chinnappa, 1988a). Bayer (1991) rejected the latter range extension, citing specimen misidentifi- cation, as determined by either inspection or deduction, and in- stead supported the distribution for the species presented by Bayer (1989a). Although Antennaria aromatica was originally described as an endemic, morphologically uniform, sexual diploid species, five years later it was more appropriate to consider the species as wider ranging (Chmielewski and Chinnappa, 1988a; Bayer 1989a), reproductively variable (either sexual or apomictic) and cytolog- ically variable (diploid, tetraploid or hexaploid) (Bayer and Steb- bins, 1987; Bayer 1989a). Reproductive biology, hybridization, polyploidization and dioecism have all contributed to morpho- logical intergradation among species of Antennaria and thereby magnified the present state of confusion not only in this species but also the genus (Fernald, 1950; Cronquist, 1955; Anderson, 1959; Hultén, 1968; Welsh, 1974; Scoggan, 1979; Porsild and Cody, 1980; Moss, 1983). This investigation incorporates the results of a phenetic study of gross morphology with information published during the past decade to determine whether it is possible to distinguish consis- tently and confidently among Antennaria aromatica, A. media, A. pulvinata (including subsp. albescens E. Nelson), A. rosea and A. umbrinella. Having established those taxa which are distinct, some comments are offered concerning the presumed evolution- ary history of the group. MATERIALS AND METHODS Herbarium specimens, including type material, utilized in the phenetic study were borrowed from ALTA, ARIZ, CAN, COLO, DAO, F, LEA, MO, MONT, MOR, MT, NY, OS, PH, RM, SLRO, UAC, UBC, UC, US, WIN and ws (Holmgren et al., 1990). Specimens of each species were selected on the basis of completeness and comparability in terms of developmental stage. Only mature pre-dehiscent speci- mens were included. These specimens reflect a range in plant size, 1993] Chmielewski— Antennaria 263 Table 1. Morphological characters included in the phenetic study of Anten- naria aromatica, A. media, A. pulvinata, A. rosea and A. umbrinella. Structure Character Vegetative Basal leaf length (mm): BASALLL Basal leaf width (mm): BASALLW Vegetative-Reproductive Plant height (mm): HEIGHT umber of cauline leaves: NOCAULL Cauline leaf length (mm): CAULLL Cauline leaf width (mm): CAULLW Number of capitula: NOCAP Involucre height (mm): INVOHT Outer phyllary length (mm): OUTPHYL Outer phyllary width (mm): OUTPHYW Inner phyllary length (mm): INNPHYL Inner phyllary width (mm): INNPHYW Reproductive Pappus length (mm): PAPPUSL Corolla length (mm): COROLLAL involucre size, habit and provenance. No a priori restrictions were placed on the total number of specimens considered necessary to represent the range of morpho-geographic variation within each species. Herbarium specimens (” = 295) were identified prior to analysis as Antennaria aromatica (n = 26 type wre A. media (n = 63), A. pulvinata (n = 103), A. rosea (n = 64) and A. um- brinella (n = 65). Qualitative characters used to identify the spec- imens to species prior to analysis included habit, degree of pu- bescence, color, shape and texture of the phyllaries, exsertion of the style and the presence or absence of papillae on the achenes. These characters were not utilized in the phenetic study. Both pistillate and staminate plants were included in the phenetic study. Data were collected for 14 quantitative characters for each spec- imen (Table 1). These characters were selected primarily because they were uncorrelated (|r| < .70), described the general form of vegetative and reproductive structures, were not used directly in the classification of individuals to a priori groups, were not ips- ative and were previously shown to be useful in differentiating among species of Antennaria (Chmielewski and Chinnappa, 1988b, 1991; Chmielewski et al., 1990a, 1990b). Several of these char- acters, including plant height, leaf shape, number of capitula, size of capitula and floret size, are typically used to differentiate among species of Antennaria. The data matrix contained no missing values. 264 Rhodora [Vol. 95 Canonical variates analysis was facilitated with the SAS (SAS Institute Inc., 1989) DISCRIM ( lid d posterr options) and CANDISC procedures. To meet the assumptions of multi- variate normality (Gilbert, 1968), length measurements were transformed to their logarithms (base 10) and count data were transformed to their square roots prior to initiating multivariate analysis. Classificatory discriminant analysis was used to classify specimens of Antennaria media, A. pulvinata, A. rosea and A. umbrinella, defined a priori on one or more characters that were not included in the analysis, to species. Correct classification rates and Geisser assignment probabilities were used as indicators of separation among the groups. Tests for equality of group centroids were performed as part of the canonical analysis. The same data set was used to both define and evaluate the classification crite- rion. The classification criterion was subsequently used to classify the type collections of A. aromatica (n = 26) into one of the previously defined species. Antennaria aromatica was not treated as a separate a priori group in the discriminant analysis because the importance of glandularity in the separation of A. rosea subsp. pulvinata from A. aromatica was in my estimation overempha- sized. The copious glandularity that is evident on the holotype is less pronounced to lacking on some of the other type collections. Additionally, some specimens that superficially appear non-glan- dular do possess glands beneath the tomentum. Thus, A. aro- matica is not considered to be consistently glandular. Trivariate plots of the canonical variate scores were used to facilitate vi- sualization of the results of the multidimensional analysis (DeltaPoint Inc., 1992). Mean value + standard deviation was determined for each character for each species. RESULTS Canonical variates analysis (CVA) of 14 quantitative morpho- logical characters was used to document discontinuities among specimens of Antennaria media, A. pulvinata, A. rosea and A. umbrinella. Evaluation of the classification criterion indicated that 94% of the specimens were classified correctly (Table 2). Geisser assignment probabilities were generally greater than .98 for the 278 correctly assigned specimens. The low error count, in conjunction with the high Geisser assignment probabilities, in- 1993] Chmielewski— Antennaria 265 Table 2. Classification summary by number of observations and percent clas- sified into species for the discriminant analysis. Number of Observations and Percent Classified into Species From Species media pulvinata rosea umbrinella media 54 8 0 1 85.71 12.70 00 1.59 pulvinata 6 97 0 0 5.83 94.17 .00 .00 rosea 1 0 63 0 1.56 .00 98.44 .00 umbrinella 0 1 0 64 .00 1.54 .00 98.46 dicates that classification of additional specimens, that is, t collections of A. aromatica, through the use of the classification criterion, would yield acceptable results. The first and second discriminant functions accounted for 90.5% (73.1 and 17.4% respectively) of the total variation (eigenvalues of 4.263 and 1.015 respectively) among specimens. Ordination of the 295 specimens by canonical analysis is presented as a series of trivariate plots (Figure 1) in which mean canonical scores are indicated for each species. The Mahalanobis distance between groups and associated F-value indicate the group centroids were significantly different (P < .0001). The canonical correlation of the first discriminant function (.900), or that squared as the pro- portion of the variance in the function explained by the four groups (.810), indicates high to moderate correlation with the groups. A strong relationship between the specific linear combi- nation of variables, the canonical variables and the groups is indicated. Descriptive statistics (mean + standard deviation) are used to identify character discontinuities for the predefined groupings of specimens (Table 3). Except for STEMHT, NOCAULL, CAULLL, BASALLL, NOCAP and INNPHYW which exhibited moderate i no overiay among all or some a the groups, the remaining harac d mong the groups. These char- acters had the highest loadings on the first canonical axis. ype specimens of Antennaria aromatica were classified, through the use of the classification criterion, to either A. media (7%, n = 266 Rhodora [Vol. 95 Canonical Axis Z &exrwlrzi OP DP MN WD Canonical Axis 2 “eg. k) 6 . 5 ‘ Canonical Axis Z fans = + > - Canonical Axis 2 Figure 1. Ordination of a. Antennaria media (—.818, 1.277, —1.017); b. A. pulvinata (—2.238, —.302, .564); c. A. rosea (2.908, .754, .750); and d. A. um- brinella (1.476, —1.502, —.646). Group centroids are shown in parentheses (ca- nonical axes |, 2 and 3 respectively). 2) or A. pulvinata (92%, n = 24). Geisser assignment probabilities were typically greater than .95 for these specimens. DISCUSSION In describing Antennaria aromatica, Evert (1984) noted that the species was morphologically uniform throughout its range. Subsequent comparison of the holotype to average species values via a mean similarity matrix employing Pearson product-moment correlation coefficients (see Bayer, 1988a, Table 1) indicated that 267 Chmielewski— Antennaria 1993] €S°0 ¥ 087 ISO + Ore p90 F ESE LS0 ¥ LSE ISO ¥ PS (ww) TWTIOUNOO 7S°0 ¥ 99'€ 6S°0 + €b'r 7L'0 ¥ LSP €9°0 + €L'b 180 ¥ 98'¢ (ww) T§AddVd ¢7'0 + 79'°0 L1‘0 ¥ 80 70 + 060 1770 ¥ LL'0 €£°0 ¥ €0'l (ww) MAHdNNI 79°0 + SIP ¢9°0 + LOS 890 # SES 19°0 + 80's ¢L'0 + L8'b (ww) TAHdNNI 070 + SO'l €770 ¥ 60'1 S70 F 8E'l 7'0 + €7'l 170 ¥ 071 (ww) MAHA LAO vS'0 ¥ 87'€ v9'0 + I8'€ 89'°0 F P0'r p90 + 06'€ 6L°0 ¥ PLE (ww) TAHdLNO €L°0 ¥ 8L'P 1L'0 + 89°S S80 + €6'S 99°0 + gs°¢ 6L°0 + 7E'S (ww) THOANI LTE F078 6L°L + OT TI If l + pr'p 6b 1 + 00'S 660 * TPE dVOON €8°0 ¥ 6P'E 67'l + 86'€ Ol + Sip 96°0 * €€'€ OT I + 8@'F (ww) A TTVYSVd 8I'€ + IE Ol Ors + ELI 68°7 + 7L'8 79E F LOTI v0'7 + S78 (ww) TTTVSVa €$°0 ¥ 731 b8°0 + S77 8r'0 + O8'I 7S0 F ILI €9'°0 = L8'l (ww) MTINVO 167 + 9611 979 + 8I'IZ Sr F £66 Sle + EPI €77 F E78 (ww) TTIAVO 60°7 F 6F'8 lee F STTI 781 + 96°9 661 + $9°9 891 + Z's TINVION S6'€ + 9TOI co°8 + 18°61 197 + TI'S €9°7 F 86'S Or l + TIE (tu9) LHWALS Djjau1aquin Dasod pyouiajnd pipaut DINDULOAD JaoRvIey) uoxe | “Apnys oousyd oy} UI popnyour exe} dy} JO YOR JOJ UONeIADP psepur}s = onyea urow jo Arewuwing “¢ JQeL 268 Rhodora [Vol. 95 it is “atypical of the species, consisting of really small diploids that do not occur on limestone talus” (Bayer, pers. comm. 1989). Examination of available type material (n = 26) indicated that A. aromatica, like most A ki and Chinnappa, 1988b, 1991; Chmielewski et al., 1990a, 1990b), is indeed vari- able, exhibiting different degrees of plasticity, with respect to basal leaf length and width, plant height, cauline leaf length and width, degree of pubescence, glandularity and involucre height. Anten- naria aromatica is therefore a species which illustrates that the nomenclatural type is not necessarily the most typical or repre- sentative element of the taxon; it is merely that element with which the name is apagerrine| associated (Lanjouw et al., 1966). Initially described as a diploid sp (Evert, 1984), Antennaria aromatica was subsequently reported to include both tetraploid and hexaploid individuals (Bayer, 1984, 1989a; Bayer and Steb- bins, 1987). Apparently some confusion exists with respect to the identification of the tetraploid specimen designated CO-458 (Bay- er and Stebbins, 1987), as it was included with both A. aromatica (toward A. media) and A. rosea (toward A. aromatica) by the authors. This confusion notwithstanding, it is likely that the dip- loid and polyploid states possessed by individuals of the species have contributed, at least in part, via the gigas effects of poly- ploidization (Stebbins, 1971) to the observed patterns of plasticity (see also Bayer, 1989b) noted above. Because so few cytological determinations are available for the species, statistically valid conclusions pertaining to the relationship between chromosome number and morphology are not possible at this time. Results based on canonical variates analysis indicate that the degree of morphological overlap among the four taxa, Antennaria media, A. pulvinata, A. rosea and A. umbrinella, as defined in this study, is minimal. High assignment rates and Geisser assignment probabilities support this conclusion. The conclusion that A. me- dia, A. rosea and A. umbrinella are morphologically distinct is not a novel one as the species are generally treated as such (see Bayer, 1988a, 1988b, 1989c, 1990a; Chmielewski and Chinnappa, 1988b; Chmielewski et al., 1990a, 1990b for discussion). The question remains, how does Antennaria pulvinata fit in with these species? Direct visual comparison of type material indicates that frag- ments of the holotypes of Antennaria aromatica (RM346356) and 1993] Chmielewski—Antennaria 269 A. pulvinata (NDG058584) are indistinguishable, as are the para- type of A. aromatica (F) and the isotype (CAN 105666) and para- type (CAN105674) of A. pulvinata. Although two type collections of Antennaria aromatica were originally assigned to A. media, through the use of the classifi- cation criterion, the 92% success rate observed is in line with the capabilities of the classification criterion as it was defined through the analysis. The classification criterion obtained through canon- ical variates was initially evaluated at a 94% success rate. Thus, it is unrealistic to expect a higher classification success rate when it is used to classify collections, such as the types of A. aromatica, which were not used to define the classification criterion. The analysis demonstrates not only that A. aromatica and A. pulvinata are indistinguishable but also that morphological intergradation among species of Antennaria is a facet of reality. In attempting to clarify the confusion that surrounds the iden- tity of Antennaria pulvinata, Bayer (1989a) concluded that it re- sembles A. aromatica closely, except that it lacks glandular hairs. This character is not infallible however, as glandularity, even among type collections of A. aromatica, ranges from a dense cover to trace amounts or even lacking. Additionally, some specimens that superficially appear non-glandular do possess glands beneath the tomentum. Thus A. aromatica is not consistently copiously glandular, but rather variable in this regard. Inasmuch as Antennaria pulvinata and A. aromatica are iden- tical, the basis for inclusion of the former in A. rosea is elusive. If A. pulvinata should be included at some taxonomic rank in the A. rosea polyploid complex, as proposed by Bayer (1989b), then the results of canonical variates analysis should support this con- tention. Support would include lower assignment rates for both taxa, as well as lowered Geisser assignment probabilities. Neither of these scenarios is true however in the present study. In cir- cumscribing A. rosea, Bayer (1989b) concluded that subsp. pu/- vinata is entirely pistillate, apomictic and polyploid. This non sequitur is unacceptable however as Greene (1898) described both “female” and “male” specimens of A. pulvinata. The protologues of A. pulvinata and A. rosea, gross morphology of type material and the results of canonical variates analysis, all contribute to support the conclusion that the two taxa are morphologically distinct. The two taxa, A. pulvinata and A. rosea, do exhibit mor- 270 Rhodora [Vol. 95 phological integrity (see Table 3) and should therefore be treated as distinct. As such, it would be inappropriate to include the former species as an infraspecific subdivision of the latter. Although originally described as an endemic, morphologically uniform, sexual, diploid species, Antennaria aromatica is more appropriately described as wider ranging (Chmielewski and Chin- nappa, 1988a; Bayer, 1991), both sexual and apomictic (Bayer, 1989a), diploid and polyploid (Bayer, 1989a) and morphologi- cally variable. Gross morphology of type and non-type material, the Pearson product-moment correlation coefficients presented by Bayer (1988a), the protologues of A. aromatica and A. pulvi- nata, the fallibility of using glandularity as a diagnostic character and the results of canonical variates analysis all contribute, to varying degrees, to support the conclusion that A. aromatica and A, pulvinata are conspecific. The latter specific epithet is deemed the legitimate name for the plant and the former a synonym. A list of synonymy is provided for A. pulvinata. TAXONOMY The following key may be used to differentiate among the An- tennaria media, A. pulvinata, A. rosea and A. umbrinella. A. Plants generally tall, greater than 1 dm, typically 1.5-2.5 dm; cauline leaves long, 1.5+ cm, numerous; basal leaves typ- ically lanceolate, occasionally spatulate-lanceolate, 1.2+ cm long; immature involucres nodding ............ A. rosea A. Plants generally compact, less than 1 dm, typically 0.5-1.0 dm; cauline leaves shorter, <1.5 cm, fewer; basal leaves spatulate, obovate, spatulate-obovate, cuneate spatulate to occasionally oblanceolate, typically <1.2 cm; immature in- volucres not nodding B B. Stolons elongate, strongly upsurgent at tips and usually woody; basal leaves tufted about stem; tips of involucral bracts dirty white or light brown ...... A. umbrinella B. Stolons short, not upsurgent; basal leaves not tufted about stem; tips of involucral bracts conspicuously darker than base, brown, green, black C C. Basal leaves typically spatulate-lanceolate with per- manent woolly tomentum; inner bracts narrow, lin- ear, tips usually sharp pointed, acute ... A. media oS WS 8 CLS eh ee a ee Rae ee we Oe ee ee Oe Re ea ee S eet Bae aE Se ee we 1993] Chmielewski— Antennaria 271 C. Basal leaves widely cuneate-spatulate to occasionally oblanceolate; persistently white tomentose on both surfaces; inner bracts broader, frequently obtuse .. Paha ek ne vee es eee eae A, pulvinata LIST OF SYNONYMY a pulvinata Greene, Pittonia 3: 287, 1898. Type: Can- a: Alberta, Moose Mtn, Elbow River, Rocky Mountains, ie 29, 1897, J. Macoun 18491. (HOLOTYPE: NDG058584!; IsoTyPe: CAN105666!). J. Macoun 18493, 18495, 18498. (PARATYPES: CAN105675!, CAN105674!, NDG058583!, NDG058584!). J. Macoun 18492. (ToporyPe: CAN105676!). Antennaria rosea subsp. pulvinata (Greene) R. Bayer, Brittonia 41: 59, 1989. Antennaria pulvinata subsp. albescens E. Nelson, Proc. U.S. Natl. Mus. 23: 702, 1901. Antennaria rag (E. Nelson) Rydb. Rocky Mts. 917, 1067, 1917. Tyre: United States: Idaho, Texas District, Mount Hops, Salmon River Mountains, Aug i, "1895, L. F. Henderson 3870. (HoLoTyPE: US228762!). Ss aroma tica Evert, Madrono 31: 109, 1984. Type: United States: Mon- tana, Carbon Co., hens Range along Hwy 212, ca. .8 km N of Quad Cr., Aug 5, 1981, E. F. Evert 3406. (HOLOTYPE: RM346356!; IsoTYPEs: MO3229695!, MOR52039!, UC1509580!). F. W. Pennell, F. B. Cotner and R. L. Schaeffer 23835. (PARATYPES: F1576300!, PH853371!). R. J. Bayer and G. L. Stebbins 8092. (PARATYPE: OS153623!). E. F. Evert 1780, 3015, 3420. (PARATYPES: RM346360!, RM346359!, RM346357). P. A. Robertson 1110. (PARATYPE: RM330754!). G. L. Stebbins and R. J. Bayer 8113. (PARATYPE: UAC41448)). The original species description of Antennaria pulvinata (Greene, 1898), supplemented by the species description for A. aromatica (Evert, 1984), deemphasizing glandularity as a diagnostic char- acter for the latter, adequately reflects the morphology of the former as it was applied at the completion of this study. Species descriptions previously presented for A. média (including A. pul- chella Greene, Bayer, 1990b), A. rosea (Chmielewski et al., 1990a) and A. umbrinella (Chmielewski et al., 1990a) are deemed ac- curate and were applied accordingly in this study. The distributions presented for Antennaria media (Bayer, 1990b), A. pulvinata (as discussed for A. aromatica in Chmie- lewski and Chinnappa, 1988a), A. rosea (Chmielewski and Chin- nappa, 1988b) and A. umbrinella (Chmielewski et al., 1990a) are considered to accurately represent the ranges of these species as they are presently understood. 212 Rhodora [Vol. 95 EVOLUTIONARY RELATIONSHIPS Antennaria media sensu lato, considered by some to be one of the more important western species of Antennaria (see Bayer, 1987a), has been implicated in the hybridization events which resulted in A. alpina (L.) Gaertner, A. parvifolia Nuttall and A. rosea. Additionally, A. media is considered to possess a pivotal genome, being partially responsible for A. parlinii Fernald over- lapping morphologically with A. neodioica Greene; A. neodioica with A. parvifolia; A. parvifolia with A. rosea; and A. rosea with A. alpina (Bayer, 1987a). The importance of A. media in the evolution of all these taxa is in my estimation overemphasized as these apparent relationships represent speculation based on little or no data. A recent revision of Antennaria media sensu lato led Bayer (1990b) to conclude that the diploid cytotype (A. pulchella Greene) should be recognized as distinct from the polyploid cytotypes represented by A. media. This conclusion, founded on the results ofa phenetic study, warrants discussion. First, although polyploid A. media occurs throughout the alpine zone of many western North American mountain ranges, extending from the south- western United States to the western arctic, only specimens from the southern portion of its distribution were included in the anal- ysis. Second, the data set utilized was incomplete and therefore subject to produce unsupported results (see Chmielewski and Chinnappa, 1992). Third, except for A. scabra Greene, the taxa included in the list of synonymy were not utilized in the analysis. Fourth, many additional taxa which were previously included in the complex were excluded without comment from the phenetic study (A. atriceps Fernald ex Raup, A. compacta Malte, A. fusca E. Nelson, A. maculata Greene, A. modesta Greene, A. mucronata E. Nelson, A. pallida E. Nelson and A. stolonifera A. E. Porsild). Additional taxa which I believe should be considered in a revision of A. media sensu lato, but which were placed in synonymy with A. rosea (Bayer, 1989b), include A. chlorantha Greene, A. lanulosa Greene, A. isolepis Greene and A. tomentella E. Nelson. Fifth, although A. scabra was considered most similar to A. aromatica based on the results of an average similarity matrix (Bayer, 1988a), Bayer (1990b) failed to address this discrepancy, concluding that it was most similar to A. pulchella. Similar discrepancies based on the results of an average similarity matrix (Bayer, 1988a) exist 1993] Chmielewski— Antennaria 2738 for A. austromontana E. Nelson (most similar to A. aromatica), A. candida Greene (most similar to A. umbrinella) and A. modesta (most similar to A. aromatica). A phenetic study which includes specimens from throughout the range of A. media sensu lato and also incorporates all those taxa which may be part of the complex is necessary before a classification scheme may be legitimately proposed. For the time being it would be prudent to treat these dark green-black phyllaried taxa as members of A. media sensu lato. Bayer (1987b) postulated that Antennaria rosea is a polyploid compilospecies, the result of multiple hybridizations among sev- eral amphimictic species. Results based on phenetic analyses in- dicated that A. aromatica, A. corymbosa E. Nelson, A. media, A. microphylla Rydberg and A. umbrinella are the major sexual pro- genitors of the complex, whereas A. marginata Greene and A. rosulata Rydberg are minor contributors (Bayer, 1990a). Chmie- lewski et al. (1990a) questioned the likelihood of such a compli- cated pattern of hybridization events across sectional boundaries. The latter authors instead stated that phenetic analyses, based on a substantially larger sample of specimens, supported a hybrid origin among A. corymbosa, A. microphylla and A. umbrinella. Although phenetic analyses have led the aforementioned authors to support different positions relative to the origins of A. rosea, isozyme data can be used as independent evidence to support the hypothesis of Chmielewski et al. (1990a), that is, that only A. corymbosa, A. microphylla and A. umbrinella are among the sex- ual progenitors of the complex (Bayer, 1989d). Antennaria umbrinella, a member of section Alpinae (Chmie- lewski et al., 1990c) was recently referred to as a member of section Dioicae (Bayer, 1987b), as well as section Alpinae (Bayer, 1989c). A sexual species, A. umbrinella is most similar to A. pulvinata, the strongly upsurgent, typically woody stolon tips representative of the former species accounting for morphological divergence between the taxa. The role of this species as one of the progenitors of the A. rosea polyploid complex was adequately discussed pre- viously. ACKNOWLEDGMENTS The curators at ALTA, ARIZ, CAN, COLO, DAO, F, LEA, MO, MONT, MOR, MT, NY, OS, PH, RM, SLRO, UAC, UBC, UC, US, WIN and ws are 274 Rhodora [Vol. 95 thanked for the loan of type material and additional specimens. J. C. Semple is thanked for his helpful comments and suggestions in revising a pre-submission draft of the manuscript. The anon- ymous reviewers of the submission copy are also thanked for their comments and suggestions. LITERATURE CITED AnpberSON, J. P. 1959. Flora of Alaska and Adjacent Parts of Canada. lowa State ageing! Prt Ames, IA. BAYER, R.J. 1984 1 for North Amer- ican species of propia (Asteraceae: Inuleae). Syst. Bot. 9: 74-83. . 1987a. Evolution and phylogenetic enn of the Antennaria (As- teraceae: Inuleae) polyploid agamic complexes. Biol. Zent. bl. 106: 683-698. 1987b. Morphometric analysis of dete North American Antennaria (Asteraceae: Inuleae). I. Sexual species of sections Alpinae, Dioicae, and Plantaginifoliae. Canad. J. Bot. 65: 2389-2395. 1988a. Typification of western North American Antennaria Gaertner (Asterackae: Inuleae); sexual species of sections Alpinae, Dioicae, and Plan- taginifoliae. Ane 37: 292-298. ———. 1988b. Morphometric analysis of western North American Antennaria (Asteraveas: Inuleae). I. Sexual species of sections Alpinae, Dioicae, and ip ae Canad. ue Bot. 65: 2389- 2395. 1989a. rs M4 moatira and A. peers s (Asteraceae: Inuleae) i in : the western North America Cor- dillera. Madrofio 36: 248-259. A taxonomic revision of the ie rosea (Asteraceae: In- uleae Gnaphaliinae) 2 at complex. Brittonia 41: 53-60. 1989c. Patterns of isozyme variation in pias North American An- ieninatiz (Asteraceae: inl) II. Diploid and polyploid species of section Alpinae. Amer. J. Bot. 79-691. 1989d. Patterns Se isozyme variation in the Antennaria rosea (Astera- Inuleae) polyploid agamic complex. Syst. Bot. 14: 389-397. 1990a. Investigations into the evolutionary history of the Antennaria ng — Inuleae) polyploid complex. Pl. Syst. Evol. 169: 97-110. 990b. A systematic study of Antennaria media, A. pulchella and A. hor: apap Inuleae) of the Sierra Nevada and White Mountains. Madrofio 37: -183. = IS8E. ae ne on the geographic range of Antennaria aromatica Evert (Asteraceae: Inuleae). SIDA 14: 505-506. AND G. - STEBBINS. JE. Chromosome numbers, patterns of distri- bution, and ap : Inuleae). Syst. Bot. 12: 305- 319. CHMIELEWSKI, J.G. ANDC. C. CHINNAPPA. 1988a. Range extension of Antennaria aromatica Evert (Asteraceae: — SIDA 13: 256-258. 1988b. The ge Inuleae) in North 1993] Chmielewski— Antennaria a os America: multivariate analysis of variation patterns in A. rosea sensu lato. Canad. J. Bot. 66: 1583-1609. 1991. Gender dependent and independent morphological morphism in sexual and apomictic Antennaria monocephala DC. sensu os Canad. J. Bot. 69: 1433-1448. —. eee Comauennaey < on Bayer’ S use of incomplete data ma- f North Amer- ican sleet commentary. Canad. J. Bot. 70: 2313-2315 AND J. C. SEMPLE. 1990a. The genus Aphewmarla (Asteraceae: hateae) j in western North America: morphometric analysis of Antennaria alborosea, A. corymbosa, A. marginata, ay ee A. parvifolia, A. rosea and A. umbrinella. Pl. a fe Evol. 169: ‘Ob. eect - intraspecific variation in Anten- naria alborosea, A. co a mbosa, A. marginata, A. ig A. parvifolia and A. lilt Asteraceae Inuleae). Pl. Syst. Evol. 169: 150. ARNER. 1990c. Pollen morphology - aie Amer- fica species nae phates (Asieccens: Inuleae). Canad. J. Bot. 68: 1233- 1238. Cronaguist, A. 1955. Vascular Plants of the Pacific > elu Part 5. Com- positae. University of Washington Press, Seattle, De.taPornt Inc. 1992. DeltaGraph Professional, clita! 0.3. DeltaPoint Inc., Evert, E. F. 1984. A new species of A ia(A ) from Montana and Wyoming. Madrofio 31: 109-112 FERNALD, - L. 1950. Gray’s Manual of Botany, 8th ed. D. Van Nostrand Co., ork, NY. reagan a 1968. On ot using qualitative variables. Am. Stat. As- oc. J. 63: 1399-14 oe E. L. 1898s. Sate of the Compositae VII. 2. Some northern species of Antennaria. Pittonia 3: 273-298. HobLmaren, P. K., N. H. HOLMGREN AND L. C. BARNETT. 1990. Index Herba- ticeina. 8th ed., Part I. The Herbaria of the World. The New York Botanical Garden, Bronx, Huttén, E. 1968. Flora of Alaska and Neighbouring Territories. A Manual of Vascular ae Stanford University Press, Stanford, CA. LANjouw, J. S., 2 Mamay, R. McVauacu, W. Rosyns, R. C. Ro.uins, R. Ross, J. ae u, G. M. Scuutze, R. DE VILMORIN AND F. A. STAFLEU. 1966. Sosa Code of Botanical Nomenclature. The International Bureau for Plant Taxonomy and Nomenclature. 106 Lange Nieuwstraat, Utrecht, Netherlands. Moss, E.H. 1983. Flora of Alberta, 2nd ed. Revised by J. G. Packer. University of Toronto Press, Toronto, ON. Porsitp, A. E. AND W. J. Copy. 1980. Vascular Plants of Continental Northwest Territories, Canada. National Museums of Canada, Ottawa SAS Institute Inc. 1989. SAS/STAT a Guide, Version 6, "4th od, Vol. 1. SAS Institute Inc., Cary, NC. 943 p Scoccan, H. J. 1979. Flora of Fa Part 4. Dicotyledoneae (Loasaceae to Compositae). National Museum of Natural Sciences, Publications in Botany, No. 7(4). 276 Rhodora [Vol. 95 Stespsins, G. L. 1971. Chromosomal Evolution in Higher Plants. Edward Arnold (Publishers) Ltd., London WExLsu, S. L. 1974. Anderson’ s Flora of Alaska and Adjacent Parts of Canada. Braeis Young University Press, Provo, UT DEPARTMENT OF BIOLOGY SLIPPERY ROCK UNIVERSITY SLIPPERY ROCK, PA 16057 RHODORA, Vol. 95, No. 883/884, pp. 277~—284, 1993 THREE NEW SPECIES OF THE GENUS PELEXIA (ORCHIDACEAE, SPIRANTHINAE) FROM COLOMBIA Dariusz L. SZLACHETKO ABSTRACT The genus Pelexia Poit. ex Lindl. occurs in the tropics and subtropics of both Americas: from Argentina in the south to Mexico and Florida in the north. Ac- cording to Garay (1982) it contains 67 species. Pelexia is distinguishable from closely related Sarcoglottis Presl by an “externally observable line of adnation” between ovary and lateral sepals. ae studying] seh materials from COL, pands, I found some specimens y pl They are described here for the first time. Key Words: Pelexia, new species, Colombia PELEXIA CUNDINAMARCAE, sp. nov. (Figure 1) Involucrum sub angulo recto ad ovarium incurvatum, tubu- loso-convolutum. Calcar saccelliforme, apice angusto, discreto, ad ovario soluto. Auriculae labelli crassae, carnosae, ad unguiem adnatae, epichilio longitudinis aequantes. Hypochilium obova- tum, epichilium triangulare, minutum, valde incrassatum. Roots clustered, fleshy. Stem 210-450 mm tall, 5-6 mm in diameter at base, 2 mm in diameter below inflorescence, erect, rather stout, in the upper part glandular. Cauline bracts 4-5, her- baceous, thin, acute, closely adnate to the stem, longer than in- ternodes, the uppermost reaches the base of the inflorescence, glabrous except the uppermost, which is sparsely glandular in the basal portion. Leaves 3-5, basal; petiole long and narrow, up to 120 mm long; blade 55-95 mm long, 13~30 mm wide, lanceolate, acute, in pressed plants grayish-pale violet on the upper surface, dull red on the lower. Inflorescence 40-115 mm long, 8-20-flow- ered, lax, subsecund. Flowers medium size, perianth tube forms a right angle with the ovary, spur saccate with narrow projection on apex. Floral bracts 15-22 mm long, lanceolate, acute, longer than ovary, herbaceous, thin, delicate, densely glandular outside. Pedicel 1.5-5 mm long, twisted. Ovary 12-15 mm long, densely glandular. Dorsal sepal 11.5 mm long, 4 mm wide, ovate-lan- ceolate, acuminate, 3-nerved, glandular outside. Free portion of a a | 278 Rhodora [Vol. 95 SCHNEIDER Figure 1. Pelexia aleaaiiigebad Szlach.: sl rsaes and floral bract; b—dorsal sepal; c— petal; d—fre p; f—basal lip auricles, spread; g—rostellum remnant. _ lateral sepals 12.5 mm long, 3 mm wide, lanceolate, acute, slightly asymmetric, 3-nerved, glandular outside. Petals 11.5 mm long, 2.6 mm wide, linear-falcate, acute, single-nerved, slightly adnate to the dorsal sepal, ciliate on outer margins. Lip boat-formed, straight, shortly clawed; claw very thick in the center. Hypochile 1993] Szlachetko— Pelexia 279 12.5-13 mm long, 7.2 mm wide, obovate in general outline, thick- ened and papillate along midnerve and near the base, thin on margins: basal lip auricles 4-5 mm long, 4.7—5.5 mm wide when spread, connate with the claw with free apices about 1.3-1.5 mm long, fleshy, very thick. Epichile 3.5 mm long, 4-4.2 mm wide, obovate to cordate, obtuse to subacute, papillate, thicker than hypochile, fleshy. Gynostemium 9.5 mm long, erect; column foot adnate to the ovary on distance 6.5 mm long, with free apical part 2 mm long; rostellum 3 mm long, broad at base; viscidium 0.9 mm long; rostellum remnant 2.3 mm long; anther 3.6 mm long. Type: COLOMBIA, Deptos. Cundinamarca-Boyaca—entre Villa Pinzon y Ventaquemada, 2800-2900 m. Schneider 679/1, 30 March 1953 (HOLOTYPE: COL 68326). PARATYPE: COLOMBIA, Dept. Boyaca—bei Ventaquemada, 288-2900 m. Schneider 679, 30 March 1953 (s). ETYMOLOGY: eipthet of this new entity based on the name of the Department in Colombia where the species was found. Pelexia cundinamarcae is characterized by the position of the perianth versus ovary, shape of the spur and lip form, i.e., perianth segments set perpendicularly on the ovary apex; nectar spur is saccate with narrow, finger-like projection on apex; basal lip au- ricles are fleshy, thick, connate with thick claw and equal in length with the small, triangular and very fleshy epichile. Hypochile is obovate to cordate. Pelexia cundinamarcae is a terrestrial plant, found in dense, moist forest. Sepals and petals are green, lip is greenish. PELEXIA ASINUS, sp. nov. (Figure 2) Pelexiae novofriburgensi affinis, sed floribus strictis, apice plus minusve patentibus, pedicello longo, erecto, calcari coniformi, apice soluto, auriculis labellis magnis, carnosis, epichilio trans- verse ovali duplo longioribus et hypochilio ovato-rhomboidali recedit. Roots fleshy, clustered. Stem 520 mm tall, 6 mm in diameter at base, about 1 mm in diameter below inflorescence, erect, slight- ly pendulous in upper part, delicate, above the leaves sparsely glandular, densely glandular along inflorescence axis. Cauline bract single, narrow, acute, short, herbaceous, sparsely glandular at base. 280 Rhodora [Vol. 95 Figure 2. Pelexia asinus Szlach.: a—flower with floral bract; b—lip; c—dorsal sepal; d—petal; e—free portion of lateral sepal; f—rostellum remnant. Leaves 5, forming a basal rosette and 2 on the stem; petiole 120 mm long, narrow, delicate; blade 85-100 mm long, 25-28 mm wide, lanceolate, acute, asymmetric, thin, delicate, reddish tinge. Inflorescence 120 mm long, 10-flowered, lax, subsecund. Flowers erect, medium-sized, with narrow slightly falcate spur. Floral bracts 1993] Szlachetko— Pelexia 281 ANO, DIAZ 3193 Figure 3. Pelexia sinuosa Szlach.: a—flower and floral bract; b—dorsal sepal; c—petal; d—lateral sepal; e—lip. 282 Rhodora [Vol. 95 24 mm long, ovate-lanceolate, subacute, herbaceous, thin, glan- dular outside. Pedicel about 5-7 mm long, twisted. Ovary 16 mm long, narrow, densely glandular. Dorsal sepal 16.5 mm long, 5 mm wide, oblong, widest and concave in the middle, obtuse, rather fleshy, densely glandular outside in the basal half. Free part of lateral sepals 20 mm long, 4 mm wide, more or less erect, lanceolate, acute, fleshy, densely glandular outside in the basal half. Petals 16.5 mm long, 3.5 mm wide, falcate-oblanceolate, obtuse, single-nerved, glandular up to *%4 on outer margins. Lip straight with the apex bent forward, clawed; basal lip auricles 6 mm long, 1.6 mm wide, almost completely connate with the claw, free portion about | mm long, very thick, fleshy; hypochile 10 mm long, 8 mm wide, obovate, thickened along midnerve, mar- gins thin, agglutinate to the clinandrium; epichile 3 mm long, 4.8 mm wide, elliptic, obtuse, fleshy, papillate. Gynostemium 12 mm long, erect; column foot adnate to the ovary along 7 mm, with free 2.5 mm apical part; rostellum remnant 2 mm long, broad at base; anther 4 mm long. Type: COLOMBIA, Dept. Cundinamarca— Rio Sabaneta, mas abajo de El Dintel, 2200 m. Schneider 679/2, 13 June 1953 (HOLOTYPE: COL). ETYMOLOGY: asinus (Lat.)—donkey; an al- lusion to the general outline of the lip, which suggests the head of this animal. At first glance Pelexia asinus seems to be similar to P. novofri- burgensis (Reichenb. f.) Garay, but it is easy to distinguish by the erect flowers, set on the long and erect pedicel, narrow conical- falcate spur apex, very large and fleshy basal lip auricles, which are about twice the size of the elliptic epichile. Hypochile is ob- ovate. PELEXIA SINUOSA, sp. nov. (Figure 3) Habitu Pelexiae funckianae similis, sed vaginis caulinis multis, floribus majoribus et inflorescentia densa differt. Bracteae longae, intus glandulosae. Labellum basi angulato-flexuosum, plico cras- SO, Carnoso inter auriculis praeditum. Roots fleshy, clustered. Stem 370 mm tall, 5-7 mm in diameter at base, 2-3 mm in diameter below inflorescence, erect, rather stout, glandular in upper half, completely covered by cauline bracts. Cauline bracts 6-12, longer than internodes, reach the inflores- 1993] Szlachetko— Pelexia 283 cence, lanceolate, spacious, acute, herbaceous, uppermost glan- dular at base. Leaves 7, forming a basal rosette; petiole 55-90 mm long, narrow; blade 45-90 mm long, 18-40 mm wide, broad- lanceolate, acute, seems to be fleshy. Inflorescence 130-180 mm long, 15-25-flowered, dense. Flowers very large, suberect, with saccate spur. Floral bracts 34-43 mm long, lanceolate to oblong- lanceolate, acute, thin, herbaceous, glandular outside. Pedicel 5 mm long, twisted. Ovary 18-23 mm long, narrow, almost erect, glandular. Dorsal sepal 28-30 mm long, 5.5-6.5 mm wide, ob- long-lanceolate, subacute, thin, delicate, concave at apex, densely glandular outside. Lateral sepals 34-35 mm long, 44.5 mm wide, falcate-lanceolate, acute, thin, glandular outside. Petals 28-30 mm long, 4.8-5.5 mm wide, linear in lower part, falcate-lanceolate in upper, acute to obtuse, thin, glandular on outer margins. Lip set on long and narrow claw, basal lip auricles partially free from claw (2.3-2.8 mm long), 4.7-5 mm wide when spread, thickened at apices, ciliate outside; between auricles fleshy, papillate ridge, lip strongly S-incurved above the ridge; hypochile 28-30 mm long, 8.5—9.8 mm wide, thin, delicate, more or less linear in basal part and ovate in apical, sometimes with two small knobs at apex, fused with the column; epichile 4-5 mm long, 6-6.5 mm wide, cordate, obtuse, papillate, thicker than hypochile, bent forward in natural position. Gynostemium 21-25 mm long, slender, erect; column foot adnates to the ovary along 8.5-10 mm, with free apex 4 mm long; rostellum 3-4 mm long, membraneous, delicate; viscidium 1-1.5 mm long; rostellum remnant 3 mm long; anther 6 mm long; pollinarium 7.2-8 mm long. Type: COLOMBIA, Dept. del Choco—kilometro 55 de la carre- tera Ansermanuevo-San Jose del Palmar, 1700-1950 m. Terres- tre. Flores verdes, labelo verde amarilliento. Lozano & Diaz 3193, 19 March 1980 (HOLOTYPE: COL 200472). PARATYPE: COLOMBIA, Dept. de Bolivar—requon du rio San Jorge. Claes 32, sine dat. (P). ETYMOLOGY: sinuosa (Lat.)—sinous; in reference to the S-curvature of the lip base. This new species is superficially similar to P. funckiana (A. Rich. & Gal.) Schlecht., but the flowers are almost twice as large as in the last species. They are arranged in multiflowered and dense inflorescence. Cauline bracts are numerous and longer than internodes covering the whole stem completely. Very long floral 284 Rhodora [Vol. 95 bracts are glandular outside. But the most characteristic feature of P. sinuosa is S-formed lip bending in the anterior portion with fleshy, thick ridge between basal lip auricles. ACKNOWLEDGMENTS I am grateful to Prof. dr hab. Ryszard Ochyra for latinization of the diagnoses, and to the Curators of cot and s for the loan of herbarium materials and Curator of p for the hospitality during my personal visit. LITERATURE CITED Garay, L. A. 1982. A generic revision of the Sprianthinae. Bot. Mus. Leafl. Harv. Univ. 28(4): 278-425. DEPARTMENT OF PLANT ECOLOGY AND NATURE PROTECTION UNIVERSITY OF GDANSK AL. LEGIONOW 9 PL 80-441 GDANSK, POLAND RHODORA, Vol. 95, No. 883/884, pp. 285-324, 1993 ALGAL VEGETATION OF THE YORK RIVER ESTUARY AND THE ADJACENT OPEN COAST OF SOUTHERN MAINE! ARTHUR C. MATHIESON, EDWARD J. HEHRE, AND MARTIN COSTA ABSTRACT A total of 72 seaweed taxa were recorded from the York River Estuary, Maine, psi ¥ Point) and 12 estuarine locations within New Hampshire and northern Massa- chusetts. In contrast to the Great Bay Estuarine System (New Hampshire/Maine) aati sieedoas succes occur upstream to approximately 8.5 miles, a relatively within the York River after 2-3 miles, presumably due to hydrographic variability, the dominance of saltmarsh habitats and reduced availability of rocky substrata. Overall, species composition within the York River Bay Estuary as well as saltmarsh habitats of the Hampton-Seabrook Estuary Sys- tem (New Hampshire). Relatively few of the “southerly” taxa common to the from the York River Estuary, and the occurrence of Porphyra amplissima and Codium fragile ssp. tomentosoides from southern Maine. Key Words: seaweeds, estuarine, coastal, York River, Maine, New Hampshire, Massachusetts INTRODUCTION In contrast to the southeastern United States, where saltmarsh (Spartina spp.) and eelgrass communities (Zostera marina) are ominant primary producers of organic carbon within estuarine habitats, seaweeds play a major role within north temperate es- tuaries of the Gulf of Maine (Josselyn and Mathieson, 1980). Estuarine seaweeds within the latter geography provide valuable habitats for a myriad of organisms, generate significant amounts of organic carbon via primary productivity, and contribute to detrital cycles and an ‘“‘outwelling” of biological productivity to ' Scientific Contribution Number 1830 from the New Hampshire Agricultural Experiment Station; also issued as Contribution Number 276 from the Jackson Estuarine Laboratory and the Center for Marine Biology. 285 286 Rhodora [Vol. 95 nearshore coastal waters (Josselyn and Mathieson, 1978, 1980; Mann, 1972, 1982). Although these fundamental roles are be- coming better known, very few detailed assessments of estuarine seaweed resources within the Gulf of Maine exist, except from New Hampshire (Mathieson and Fralick, 1972; Mathieson and Hehre, 1982, 1983, 1986; Mathieson and Penniman, 1986b, 1991; Mathieson et al., 1981) and Massachusetts (Mathieson and Fra- lick, 1973; Webber and Wilce, 1971). Thus, the only compre- hensive seasonal collections of estuarine seaweeds in Maine have been made within the Piscataqua and Salmon Falls Rivers of the Great Bay Estuary System (cf. Mathieson and Hehre, 1986; Math- ieson and Penniman, 1991), while only selected collections have been made at other northernmost sites (Bell and MacFarlane, 1933; Eaton, 1873; Stone et al., 1970). The present study was initiated to enhance our knowledge of Maine estuarine seaweed populations. Specifically, we have sum- marized distributional and floristic patterns of seaweed popula- tions within the York River Estuary (Maine), comparing them with previous synopses of the Hampton-Seabrook (New Hamp- shire), Merrimack River (Massachusetts) and Great Bay Estuarine Systems (New Hampshire and Maine), including the various sub- sets of the latter ecosystem (Figures 1 and 2). The proximity of the four estuarine systems within the southern Gulf of Maine (Figure 1), plus their varying sizes, habitats, industrial develop- ment and eutrophication, provides several meaningful compari- sons. METHODS AND MATERIALS Seasonal collections and observations were made at twenty sites within the York River Estuary, ranging from just outside York Harbor to a point about 1.5 miles below the river’s tidal headwater (Figure 2, Table 1). A detailed description of each study site, including its location, habitat(s) and substrata is given in Appen- dix 1. Collections were made during 1978, 1985 and periodically thereafter. Overall, methods of collection and identification were similar to those outlined in Mathieson and Penniman (1991). Representative samples of all conspicuous seaweeds at each site were collected from the littoral (on foot) and sublittoral zones (by SCUBA). The nomenclature of South and Tittley (1986) was ap- plied for most taxa. The author(s) of each binomial and many 1993] Mathieson et al.— York River Algae 287 3 ft a 71°40’ 71° ‘30! FS Cape Elizabeth iy Kencabiiny, ey (7 miles }) x SS" Cape Arundel 43 10'_ 2 3G “Great Bay Estuary System =" Portsmouth : £3 Isles of Shoals in . fe Gulf of zi? NH Open Maine J Coast (Nearshore) Seabrook: .) YU Hampton-Seabrook Estuary System ° ! 42 50 a Ee 1 mile aA acy A : Ann \ Figure 1. The northern New England coastline between southern Maine 7 northern Massachusetts showing the locations of the York River, Great Ba Hampton- ‘agape and Merrimack River Estuaries, plus Cape Neddick and Sea Point, Main pertinent nomenclatural changes since Taylor (1957) are included in Table 2. Approximately 500 voucher specimens are deposited in the Albion R. Hodgdon Herbarium (NHA) at the University of New Hampshire in order to document the river’s flora. A com- [Vol. 95 Rhodora 288 YHOSHYVIT HUOA ONOd IW S.N3uuVvA LINIOd NY3ALSV3 ANIOd YOOU ONIVOY 39aIve TWM3aS S31dWVS HALYM = AYA LLY STW ¢/t x 433Y9 TNH Y3aI0 | ——> MHOA q9aINaaNVUCOs 9% ONOd H3LINOE Lona OOS 390i SLLND OOUS SHADOU YAAIY HHOA OOH LTAWS a MYOA aaa 1993] Mathieson et al.— York River Algae 289 parison of two nearshore open coastal sites in southern Maine (Cape Neddick and Sea Point) plus twelve other estuarine habitats between Maine and Massachusetts is given (Figure 1, Table 1), utilizing several published records (Mathieson, 1979; Mathieson and Fralick, 1972, 1973; Mathieson and Hehre, 1986; Mathieson et al., 1981; Mathieson and Penniman, 1986b, 1991). Surface water temperature, salinity, pH and total suspended solids were recorded (winter/spring, 1978) at eight sites, ranging from the mouth to the tidal headwaters. Temperature and salinity were recorded (in situ) with a hand-held thermometer and re- fractometer (.1°C & .5%o accuracy), while pH was measured with an Orian pH meter. Total suspended solids (mg/liter) were de- termined according to Strickland and Parsons (1968). Seasonal assessments (approx. monthly) of the same parameters were doc- umented near the mouth of York Harbor between September, 1991 and January, 1992 (i.e., near station 4 in Appendix 1) in conjunction with a comparative investigation of the York and Piscataqua Rivers (R. Langan, unpubl. data). Records of in situ temperature and salinity were recorded with a YSI conductivity meter (model CST), while ionic concentrations were measured with a Fisher Alumet pH meter. Total suspended solids were determined as outlined above. HABITAT DESCRIPTION The York River Estuary, located in southern York County, Maine, between Cape Neddick (York) and Sea Point (Kittery), extends approximately 7 miles inland from its mouth near West- ern Point to its tidal headwaters in East Eliot (Figures | and 2). Most of the estuary is within York, except for a small tidal marsh in East Eliot. The source of the river is York Pond, which is in the northeast corner of Eliot township. Three major brooks drain into the river near its headwaters (Smelt Brook in York and Cutts Ridge and Rogers Brooks in Eliot), while three large tributaries (Boulter Pond Brook, Cider Hill Creek and Dolly Gordon Brook) are within .5 mile of the Maine Turnpike (I-95). Boulter and Figure 2. Twenty study sites within the York River Estuary (Maine) and the adjacent open coast, including the location of water sampling sites. 290 Synopsis of collecting sites between southern Maine and northern Rhodora le 1. Massachusetts, including the York River Estuary Indi- vid- Distance from Geographical ual Open Coast eas ites (miles) References Nearshore open 2 =nearshore Mathieson (1979) coastline of open coast Mathieson & Hehre (1986) southern Maine: Cape Neddick (York) and Sea Point (Kittery) York River Estu- 20 0-7 present study ary, _ (York and Eliot) Great bi Estuary 147 0Q-22.7 Doty and Newhouse (1954) yst Hehre and Mathieson (1970) New rele Mathieson and Hehre (1986) Maine Mathieson et al. (1981, 1983) Mathieson and Penniman (1986b, 1991) Norall et al. (1982) Reynolds and Mathieson (1975) Piscataqua River, 59 0-12.1 Mathieson and Hehre (1986) N.H./Maine Mathieson et al. (1977) Little Bay, New 21 8.6-12.4 Reynolds and Mathieson (1975) Hampsh Great Bay, New 16 12.9=15.7 Mathieson and Hehre (1986) Hampshire Mathieson et al. (1982) Bellamy River, 10 10.4-14.3 Mathieson and Hehre (1986) New Hampshire Cocheco River, 17 12.6-15.8 Mathieson and Hehre (1986) New Hampshire Lamprey River, 9 15.8-17.5 Mathieson and Hehre (1986) New Hampshire Oyster River, New 14 11.5-14.2 Mathieson and Hehre (1986) Hampshire Salmon Falls Riv- 16 12.5-16.5 Mathieson and Hehre (1986) er, Hew Hamp- shire/Maine Sqamscott River, 16 16.2-22.7 Mathieson and Hehre (1986) New Hampshire Winnicut River, 4 15.8-17.3 Mathieson and Hehre (1986) New Hampshire Hampton-—Sea- 49 0-5.2 Hehre and Mathieson (1970) [Vol. 95 1993] Mathieson et al.— York River Algae 291 Table 1. Continued. Indi- vid- Distance from Geographical ual Open Coast Areas Sites (miles) References brook Estuary, Mathieson and Fralick (1972) New Hampshire Mathieson and Hehre (1986) Merrimack River Mathieson and Fralick (1973) Estuary, Massa- chusetts Middle Ponds, which lie to the north, are sources of the first two tributaries, while Dolly Gordon Brook originates from a saltmarsh to the southeast. As outlined i in | Appendix l, the prone available for sea- rs, cobbles, pebbles and rock outcrops primarily occur in fee energy areas (i. e., due to waves) near the mouth of the river, while York Harbor is mostly sandy (cf. Figure 2). Several artificial structures (bridge abutments, pilings and boat moorings) are located immediately upstream from York Harbor, while saltmarshes dominate most shorelines between Route 1A (Sewall Bridge) and tidal headwaters (Figure 2). The main channel of the river is relatively straight, with only one large meander at Ramshead Point, as opposed to its tribu- taries that have sharply developed meanders indicative of youth- ful streams. About one mile upstream from Scotland Bridge the river branches into two major channels associated with the con- fluence of Smelt Brook and the York River (Figure 2). A major oxbow occurs downstream from this confluence, while several smaller ones are found upstream. Each inner channel has a steep- sided shoreline that cuts across broad expanses of saltmarsh with peat and clay soils. The saltmarshes are dominated by Spartina alterniflora, S. patens and Juncus gerardii, along with several other vascular plants. In some instances (e.g., near house construction), the marshes are very narrow and only contain S. alterniflora. Overall, the most expansive mudflat habitats are found near York Harbor where saltmarshes are either reduced or absent. Figure 3 illustrates “typical” winter/spring hydrographic con- ditions within the estuary. Surface water temperatures ranged from approximately 3.0°C (0 miles) to 0°C (4.4 miles), while sa- linities were approximately 31%c on the open coast and varied from 0-16%o at station 8 (4.4 miles inland). Thus, surface water temperatures during late winter and spring are relatively low and 292 Rhodora [Vol. 95 Table 2. Systematic list of algal species found within the York River Estuary and the adjacent open coast of Maine, with synonyms in parentheses referring to names in Taylor (1957) CHLOROPHYCEAE MICROSPORALES Microspora pinnae aden (Wille) Lagerheim ULOTRICHALES lotrichaceae Ulothrix flacca (Dillwyn) Thuret in Le Jolis Ulothrix speciosa (Carmichael ex Harvey in Hooker) Kiitzing ULVALES Percursariaceae Percursaria percursa (C. Agardh) Bory Ulva so sailor fulvescens (C. Agardh) ai et Gardner Enteromorpha clathrata (Roth) Grev Enteromorpha flexuosa (Wulfen ex ae J. Agardh ssp. paradoxa (Dill- ng Enteromorpha intestinalis aero Link Enteromorpha linza (Linnaeus) J. Agardh Enteromorpha nay (OXF. Miter) J. Agardh Ulva lactuca Linnaeu Ulvaria obscura exeizing} Gayral [= Monostroma fuscum (Postels et Ru- precht) Wittrock Ulvaria oxysperma (Kiitzing) Bliding [= Monostroma oxyspermum Kiitz- ing} Monostromata ee minima Seca ex gene Kylin {includes Enteromorpha ococca, E. m oo stirs (Thuret) ile Monostroma pulchrum Farlow PRASIOLALES Prasiolaceae Prasiola stipitata Suhr in Jessen ACROSIPHONIALES Acrosiphoniaceae “Codiolum pusillum” (Lyngbye) Kjellman [= sporophytic stage in the life history of sae eg wormskioldii and probably conspecific with Co- diolum gre. Urospora as (Roth) Areschoug rospora wormskioldii (Mertens in Hornemann) Rosenvinge [includes U. collabens (C. Agardh) Holmes et Batters] Spongomorpha arcta (Dillwyn) Kiitzing Spongomorpha spinescens Kiitzing 1993] Mathieson et al.— York River Algae 293 Table 2. Continued. CLADOPHORALES Cladophorac Cha weet brachygona Harvey [including C. cannabina (Areschoug) Kjellman] Chaetomorpha linum (O. F. Miller) Kiitzing Chaetomorpha melagonium (Weber et Mohr) Kiitzing Chaetomorpha picquotiana Montagne ex Kiitzing [= C. atrovirens Taylor] Cladophora sericea ieee: eaNe: [includes C. glaucescens, C. rudol- phiana, C. osa and C. gracilis] ea riparium (Roth) | Kiting ex Harvey Rhizoclonium tortuosum (Dillwyn) Kiitzing BRYOPSIDALES Bryopsidac aN ene (Hudson) C. Agardh Cod cota Jragile (Suringar) Hariot ssp. tomentosoides (van Goor) Silva PHAEOPHYCEAE ECTOCARPALES rpac : Ba, eat Harv Ectocarpus siliculosus (Dillwyn) Lyngbye [includes E. dasycarpus, E. con- fervoides and E. penicilla ger Giffordia granulosa (J. E. S matin srse Karan auak Kylin [= Ectocarpus tomento- soides M hyrae Kuckuck Pilayella littoralis (Linnaeus) Kjellman Spongonema tomentosum (Hudson) Kiitzing [= Ectocarpus tomentosus (Hudson) Lyngbye] Ralfsiaceae Petroderma maculiforme (Wollny) Kuckuck Pseudolithoderma extensum (P. Crouan et H. Crouan) S. Lund [= Lithoder- sum (Crouan) Hamel] Ralfsia verrucosa (Areschoug) J. Agardh Sorapion sane (Wille) Rosenvinge Myrionematace: wane Pelee Taylor Elachista Elachista} Vekack (Velley) Areschoug [includes Elachista lubrica Ru- precht] CHORDARIALES Leathesiaceae (= Corynophlaeaceae) Leathesia difformis (Linnaeus) Areschoug 294 Rhodora [Vol. 95 Table 2. Continued. Chordari ee ae oe (O. F. Miller) C. Agard Eudesme virescens (Carmichael ex Harvey in aia J. Agardh SPHA' Sphacelariaceae Sphacelaria arctica oe [= Sphacelaria racemosa Greville var. arctica ey) Rei pe aead te a. (Roth) C. Agardh DESMARESTIALES arestiaceae Desmarestia aculeata (Linnaeus) Lamouroux Desmarestia viridis (O. F. Miiller) Lamouroux DICTYOSIPHONALES Punctariaceae Punctaria plantaginea (Roth) Greville ictyosiphonaceae Dictyosiphon pelea (Hudson) Greville Scytosiphonacea Petalonia fascia es. F. Miiller) Kuntze “*Ralfsia bornetii’” Kuckuck [sporophytic stage in the life history of Peta- lonia fascia] Scytosiphon lomentaria (Lyngbye) Link var. /omentaria LAMINARIALES Chordac Chorda poe (Linnaeus) nappa aia tomentosa Lyngby Laminari porn aes (Mertens) Bory Laminaria digitata (Hudson) Lamourou Laminaria saccharina (Linnaeus) ane Saccorhiza dermatodea (de la Pylaie) J. Agardh Alariaceae Alaria esculenta (Linnaeus) Greville FUCALES Ascophyllum nodosum (Linnaeus) Le Jolis Ascophyllum nodosum ecad scorpioides (Reinke) Hauck [= Ascophyllum mackaii (Turner) Holmes et Batters] Fucus distichus — emend. Powell ssp. distichus Powell [= Fucus filiformis Gmel Fucus distichus Saviiek emend. Powell ssp. edentatus (de al Pylaie) Powell [= Fucus edentatus de la Pylaie} 1993] Mathieson et al.— York River Algae 295 Table 2. Continued. Fucus distichus Linnaeus emend. Powell ssp. evanescens (C. Agardh) Powell [= Fucus evanescens C. Agardh] Fucus spiralis Linnaeus Fucus vesiculosus Linnaeus RHODOPHYCEAE: BANGIOPHYCIDAE PORPHYRIDIALES Goniotrichaceae Stylonema alsidii (Zanardini) Drew [= Goniotrichum alsidii (Zanardini) Howe] BANGIALES Erythropeltidaceae Eoshrarich carnea (Dillwyn) C. Agardh Bangiac poten AEST (Roth) C. Agardh Porphyra amplissima (Kjellman) Setchell et Hus in Hus Porphyra miniata (C. Agardh) C. Agardh Porphyra umbilicalis (Linnaeus) J. Agardh RHODOPHYCEAE: FLORIDEOPHYCIDAE NEMALIALES Acrochaetiaceae Audouinella alariae (H. Jonsson) Woelkerling [= Kylinia alariae (H. Jons- son) Kylin] Audouinella purpurea (Lightfoot) Woelkerling [= Rhodochorton purpu- reum (Lightfoot) Rosenvinge Audouinella secundata (Lyngbye) Dixon [includes Kylinia secundata and K. virgatula] Bonnemaisoniaceae Bonnemaisonia hamifera Hariot [= Asparagopsis hamifera (Hariot) Oka- ura] PALMARIALES Palmariaceae Devaleraea ramentacea (Linnaeus) Guiry [= Halosaccion ramentaceum (Linnaeus) J. Agardh Palmaria palmata (Linnaeus) Kuntze [= Rhodymenia palmata (Linnaeus) eville] CRYPTONEMIALES Dumontiaceae r Dumontia contorta (Gmelin) Ruprecht [= D. incrassata (O. F. Miller) Lamouroux] Choreocolacaceae Harveyella mirabilis (Reinsch) Schmitz et Reinke 296 Rhodora [Vol. 95 Table 2. Continued. Kallymeniaceae Callocolax neglectus Schmitz ex Batte Callophyllis cristata (C. Agardh) Kaitzine [= Euthora cristata (Linnaeus ex Turner) J Doge h] Gloiosiphoniac aiaotisloote tee (Hudson) Carmichael ex Berkeley HILDENBRANDIALES Hildenbrandiacea Hildenbrandia eq (Sommerfelt) Meneghini [= H. prototypus Nardo] CORALLINALES Corallinaceae Clathromorphum sa (Strémfelt) Foslie [= Phymatolithon evanescens (Foslie) Foslie] Corallina officinalis an us Dermatolithon pustulatum (Lamouroux) Foslie [= Lithophyllum pustula- tum (Lamouroux) Foslie te laeve (Strémfelt) Adey [= Lithothamnium laeve (Strémfelt) era corallinae (Crouan frat.) Heydrich Lithothamnion glaciale Kjellman Phymatolithon laevigatum (Foslie) F Phymatolithon lenormandii rechone in J. Agardh) Adey [= Lithotham- nion lenormandii (Areschoug lie] Pneophyllum fragile Kiitzing [= pane lejolisii (Rosanoff) Howe] GIGARTINALES Phyllophoraceae Ahnfeltia plicata (Hudson) Fries Ceratocolax hartzii Rosenvinge ‘ymnogo S crenulatus (Turner) J. Agardh [= G. norvegicus (Gunne- rus) J. Agardh] Pauionbore pseudoceranoides (Gmelin) Newroth et A. Taylor [= P. mem- branifolia (Goodenough et Woodward) J. A: Phyllophora truncata (Pallas) Zinova [includes P. brodiaei and P. inter- rupta] Petrocelidaceae Mastrocarpus stellatus (Stackhouse in Withering) Guiry [= Gigartina stel- lata (Stackhouse) Batters] Gigartinaceae Chondrus crispus Stackhouse Polyideaceae ar rotundus (Hudson) Greville [= P. caprinus (Gunnerus) Papen- uss] 1993] Mathieson et al.— York River Algae 297 Table 2. Continued. Cystocloniaceae Cystoclonium purpureum (Hudson) Batters Fimbrifolium dichotomum (Lepeschkin) Hansen [= Rhodophyllis dichoto- ma (Lepeschkin) Gobi] RHODYMENIALES Champiaceae Lomentaria baileyana (Harvey) Farlow Lomentaria clavellosa (Turner) Gaillon Lomentaria orcadensis (Harvey) Collins ex Taylor CERAMIALES Ceramiaceae seisloveribgo gee floccosa (O. F. Miiller) Whittick [= Antithamnion floc- um (O. F. Miiller) Kleen Gis byssoides Arnott ex Harvey in Hooke Callithamnion tetragonum (Withering) S. F. Gray [= - baileyi Harvey] Ceramium rubrum (Hudson) C. Agardh Ceramium strictum Harvey Plumaria plumosa (Hudson) Kuntze [= P. elegans (Bonnemaison) hmitz Ptilota serrata Kitzin Scagelia pylaisei (Montagne) Wynne [= Antithamnion pylaisaei (Mon- base otic n Delesseria pictorial (Hudson) Stackhouse [includes M. denticulata (Montagne) Kyl Phycodrys rubens ee Batters aceae Dasya baillouviana (Gmelin) Montagne [= D. pedicellata (C. Agardh) C. sd h] Rhodomelaceae Panes denudata (Dillwyn) a ex Harvey Polysiphonia flexicaulis (Harvey) Colli Polysiphonia harveyi Bailey Polysiphonia lanosa (Linnaeus) Tandy Polysiphonia nigra (Hudson) Batters Polysiphonia nigrescens (Hudson) Greville Polysiphonia subtilissima Montagne Polysiphonia urceolata (Lightfoot ex Dillwyn) Greville Rhodomela confervoides (Hudson) Silva uniform, while a conspicuous reduction in salinities occurs within inner estuarine habitats (3.3—4.4 miles), particularly during spring runoff (late March-April). The pattern of total suspended solids showed peak values near the open coast (approx. 177 mg/liter) 298 Rhodora [Vol. 95 and a general decrease upstream (47-25 m/liter), except fora slight peak near the I-95 bridge (57 mg/liter); pH varied from approx- imately 8.2—8.0 between 0-2.4 miles, with lower values occurring upstream (7.7—7.4). Seasonal variability of hydrographic conditions near York Har- bor are illustrated in Figure 4. Surface water temperatures ranged from approximately 16.0°C (September, 1991) to 2.0°C (Febru- ary), while salinities varied from approximately 32.0% (Septem- ber, 1991) to 21.0% (April), pH from 8.4 (November, 1991) to 7.6 (December, 1992), and total suspended solids from 22.9 (Feb- ruary) to 5.6 mg/liter (December, 1992). Each parameter shows a relatively broad range throughout the year, with spring runoff occurring in late March-April. Greater hydrographic amplitudes were no doubt evident within inner estuarine stations (cf. Figure 3), although they weren’t documented. Monthly values for total suspended solids during 1991-1992 were never as high as those recorded after a major “northeaster” during February of 1978 (approx. 177 mg/liter; Figure 3). Overall, water quality within the York River appears to be very good, as the region is sparsely populated (particularly upriver of Route 1) and totally devoid of industry. The only signs of pol- lution are evident near Rice’s Bridge (i.e., stations 13 and 14) where two discharge pipes occur. A localized enhancement of siltation and turbidity was evident during repair work on the I-95 Bridge (Figure 3). SPECIES COMPOSITION AND DISTRIBUTIONAL PATTERNS A total of 131 taxa were recorded from the York River Estuary and two nearby open coastal sites in southern Maine (Cape Ned- dick and Sea Point), including 61 Rhodophyceae, 39 Phaeophy- ceae and 31 Chlorophyceae (Figure 5, Table 3). Seventy-two taxa were found in the York River (29 Rhodophyceae, 19 Phaeophy- ceae and 24 Chlorophyceae), while 59 were restricted to the open coast. The seaweeds within this open coastal-estuarine gradient exhibited three major distributional patterns (Figure 5, Table 3): (1) Coastal—restricted to the nearshore open coast (59 taxa or 0 (2) Cosmopolitan—present in both estuarine and open coastal environments (45 taxa or 34%). 1993] Mathieson et al.— York River Algae 299 SALINITY (0/00) 8 TEMPERATURE (°C) np Oo a 1.0 05 0.0 s 384 200 g 8.2 ra 160 = ie 8.0 fm < < = Th 120 © i 2 g 7.6 80 a . a= 5 7.4 \ mg/l ue Ww e--LL wae ao a re) Wome ---- ay ---- Fv me 40 on S 72 ~~ = o) aa 7.0 -0 0.0 6.4 O82 2:4 3.4 4.0 4.4 DISTANCE (MILES) FROM OPEN COAST Figure 3. Surface water hydrographic conditions (i.e., temperature, salinity, PH and total suspended see within the York River Estuary during February and March (salinity only) of 1978 (3) Estuarine—restricted to the York River Estuary (27 taxa or 21%). As outlined in Table 3 approximately 47% of the coastal taxa occurred at both Cape Neddick and Sea Point, while the remain- Go =) =) Rhodora < ck Ne) a = r=.) & E z g — < °o pe al 5 Qa i ud Oo a z ra ° a ‘ 3 oO = n 9 o =. o a 2 = ° = ~~ < c Po uw a ai ma s —- fu ae | 2 < w 4 See ms oS ~~ S aS Ss @ oS 2. 2s ok US Peea Seas eae eee 1. o Q on s < > ear SN, = Saar} Figure 4. Monthly variability of surface water hydrographic conditions (.¢., temperature, salinity, pH and total suspended solids) at the mouth of the Yurk River Estuary between September, 1991—December, 1992. der (53%) were restricted to a single station (i.e., Codiolum pusil- lum, Codium fragile ssp. tomentosoides, Enteromorpha linza, Pra- siola stipitata, Ascocyclus distromaticus, Chorda filum, C. tomentosa, Dictyosiphon foeniculaceus, Eudesme virescens, Lam- inariocolax tomentosoides, Leathesia difformis, Mikrosyphar porphyrae, Punctaria plantaginea, Saccorhiza dermatodea, Spha- celaria arctica, - cirrosa, Spongonema tomentosum, Antitham- nionella floccos alariae, A. purpurea, Callithamnion tetragonum, beiaiarn ramentacea, Gloiosiphonia capillaris, Gymnogongrus crenulatus, Harveyella mirabilis, Lomentaria or- 1993] Mathieson et al.— York River Algae 301 #green #brown # red Perennial Cosmopolitan Estuarine NUMBER OF TAXA a ao o Total Aseasonal Annual or Pseudoperennial Perennial Estuarine NUMBER OF TAXA Aseasonal Annual or Pseudoperennial Figure 5. Patterns of species richness, local distribution and longevity of sea- weed taxa within the York River Estuary and at two nearshore open coastal sites in southern Maine (Cape Neddick and Sea Point). cadensis, Plumaria plumosa, Pneophyllum fragile, Porphyra am- plissima, P. linearis and Scagelia pylaisei). Cosmopolitan taxa exhibited varying estuarine distributional patterns (cf. Table 3). The fucoid brown algae Ascophyllum no- dosum and Fucus vesiculosus plus the red alga Dumontia contorta occurred at 86-96% of coastal and estuarine stations, while twelve Others occurred at >60% of these sites (e.g., Blidingia minima, Enteromorpha intestinalis, Monostroma grevillei, Urospora pen- icilliformis, Ascophyllum nodosum ecad scorpioides, Petalonia fas- cia, Pilayella littoralis, Scytosiphon lomentaria var. lomentaria, Chondrus crispus, Hildenbrandia rubra, Polysiphonia harveyi and Porphyra umbilicalis). Some taxa exhibited restricted (<20%) and 302 Rhodora [Vol. 95 ¢ q Table 3. Summar study si sites within the York River Estuary, plus the atiaciat open coast at Cape Neddick and Sea Point Maine. an a ee oe ee eae Chlorophyta mae minima eee x x plum sane sy cunie Chaetomorpha brachygona x Chaetomorpha lin Ke. poe x x x Chaetomorpha melagoniu x x Chaetomorpha picquotiana x x me oe OX Cladophora sericea Wx “Codiolum pusillum” x Codium fragile ssp. tomentosoides x Enteromorpha clathrata Enteromorpha flexuosa ssp. paradoxa x Enteromorpha intestinalis ae x oo Enteromorpha linza x Enteromorpha prolifera te ey x Microspora pachyderma Monostroma grevillei Me oe x ox Monostroma pulchrum % Percursaria percursa Prasiola stipitata x hizoclonium riparium x x Rhizoclonium tortuosum x x a,’ Spongomorpha arcta ees a Ulva lactuca x xX x x Ulvaria obscura iL x a vk Vos x x AdAnn. 50.0 (50) AAnn. 18.2 (10) yoke ox Ann 22.7 (15) Ann 9.1 (0) es, © Oe | OX > ae x x Ann 59.1 (55) eee ke x Xx Ann. .5 (60) x oo. s- AS x AAnn.or 50.0 (45) c PPer x x RO x Ann 41.0 (35) > xy “% x * x x %* ADA 36.4 (40) SR CR + a We a) ote ee) x x Ann 86.4 (85) x XxX x xX % = =u X x Ann 59.1 (55) 1 7 2-2 WU Hew Ae Per. 9.1 (0) Per. 9.1 (0) Ann 4.5 (0) ee ae a or a ee ee oe ee ee ee 91.0 (90) 304 Rhodora [Vol. 95 Table 3. Continued. NE SP -1.-2 4 6 7 Ascophyllum nodosum ecad scorpioides x x Chorda filum x oment x Chordaria flagelliformis x x Desmarestia acu > a: esmarestia viridis ex Dictyosiphon foeniculaeus x Ectocarpus fasciculatus X x Ectocarpus siliculosus x x X Elachista fucicola tates Pe x x Eudesme virescens x Fucus distichus ssp. distichus Kok Fucus distichus ssp. edentatus aa Fucus distichus ssp. evanescens er ee Fucus spiralis x s x Fucus ve. Rae a oe x x ox Giffordia granulosa Laminaria digitata ae Laminaria sacchari OX x Laminariocolax tomentosoides x Leathesia difformis x Mikrosyphar porphyrae x Petalonia fasci Se oR x x Xx Petroderma maculiforme x Pilayella littoralis toe Xx xX Pseudolithoderma extensum Punctaria plantaginea x ‘Ralfsia bornetii x Ralfsia verrucosa x = x X X Sacchorhiza dermatodea x Scytosiphon lomentaria var. lomentaria : a a ie X x X Sorapion kjellmanii x Sphacelaria arctica X Sphacelaria cirrosa x Spongonema tomentosum x Total Phaeophyta taxa 32 18 10 5 11 e439 Mean (+SD) Phaeophyta taxa (OE 8) Rhodophyta Ahnfeltia plica x x peer emai, floccosa % Audouinella alariae 1993] Mathieson et al.— York River Algae Table 3. Extended, continued. 8 9 10 11 12 13 14 15 16 17 18 19 20 Longevity %* Me Ee oe eS ee Pere 68.2 (75) Ann 4.5 (0) Ann 4.5 (0) Ann 9.1 (0) Per. 9.1 (0) Ann 9.1 (0) Ann 4.5 (0) Ann 9.1 (0) Ann 13.6 (10) x Me x x Per. 54.5 (50) Ann 4.5 (0) Per: 9.1 (0) 5.4 Per 18.2 (0) x Per 18.2 (10) Xx Per. 18.2 (10) i OK GE ER Ge CR i Ra Le Pee Be UR er. 95.5 (95) x Ann 4.5 (5) Per. 9.1(0 Per. 13.6 (5) Ann 4.5 (0) Ann 4.5 (0) Ann 4.5 (0) x x Koes x Se ee Ann 81.8 (80) x x Per. 13.6 (15) x Be! Re pe Raw ON oe x «x Ann TCT) x Xx X Per. 13.6 (15) An 5 (0) SX Per.? 13.6 (15) x x x xk be Per. 50.0 (45) Ann 4.5 (0) x 5 n> oi, x x Ann 72.7 (70) x Per, 9.1 (0) Per. 4.5 (0) Per 4.5 (0) Per.? 4.5 (0) 7 a Oe ie i a a a: i ae a Per: 9.1 (0) AAnn 4.5 (0) Ann. 4.5 (0) 306 Table 3. Rhodora Continued. [Vol. 95 NL SP 1 2 3 Audouinella purpurea Audouinella secundata Callocolax neglectus oe cristata Ceramium rubru pier strictum — hartzii Chondrus crispus Clathromorphum circumscriptum Leptophytum laeve Polysiphonia harveyi i i i al i i i de a i i i al ~ OKO ~ i i i ol i i i i i oe a 1993] Mathieson et al.— York River Algae 307 Table 3. Extended, continued. 8 9 10 11 12 13 14 15 16 17 18 19 20 Longevity %* Per. 4.5 (0) AAnn 9.1 (5) Ann 22.7 (15) Per 9.1 (0) 34 x ok ot Ann 22-125) Per 4.5 (0) Per, 9.1 (0) x Per 13.6 (5) x Per: 22.7 (15) x x Ann. 9.1 (10) Per. 9.1 (0) x Xx MAX oe Ey x Per. TE.3 (75) Per. 18.2 (10) Per 22.7 (15) Per. 9.1 (0) x x Ann. 9.1 (0) Per. 9.1 (0) Per 4.5 (0) SR Bi Xe Sk Ann 86.4 (85) x : meee § Ann 18.2 (20) Per. 4.5 (0) Ann 4.5 (0) Per 4.5 (0) Per. 4.5 (0) KX XS oe ow x. x Per. 773-75) Per. 4.5 (5) Fer 9.1 (5) Per 18.2 (10) x Ann 4.5 (5) x Per.? 9.1 (10) Per 4.5 (0) Per 27.3420) Per 9.1 (0) Per 9.1 (0) Per 9.1 (0) Per 9.1 (0) Per. 9.1 (0) Per 9.1 (0) Per. 9.1 (0) Per. 4.5 (0) Per. 4.5 (0) Per. 9.1 (0) x Ann 4.5 (5) Per 9.1 (0) x Ree es ee ee Ann. 63.6 (60) 308 Rhodora (Vol. 95 Table 3. Continued. MISE Sle. (2. eid 4 Si GA Polysiphonia lanosa x x xx Polysiphonia ni Polysiphonia nigrescens x x x x Porphya am i X Porphyra linearis > Porphyra a Porphyra umbilicalis < x x x. x x Ptilota serrata x Rhodomela confervoides x2 Scagelia pylaisei 4 Stylonema alsidii Total Rhodophyta taxa ao.3s & 8 1D 7 8 2 38 Mean (+SD) Rhodophyta taxa (6:5 2:3) Grand total seaweed taxa 97 72 22 16 B23 DW Zl Mean (+SD) total seaweed taxa (21.4 & 3.2) * The two % values represent calculations based upon a total of 22 combined and 20 estuarine stations, respectively. patchy estuarine distributions (Chaetomorpha brachygona, Cla- dophora sericea, Enteromorpha flexuosa ssp. paradoxa, Fucus distichus ssp. edentatus, F. distichus ssp. evanescens, F. spiralis, Laminaria saccharina, Audouinella secundata, Callophyllis cris- tata and Polysiphonia urceolata), while others were restricted (<20%) to the outer estuary (e.g., Rhizoclonium tortuosum, Ec- tocarpus siliculosus, Clathromorphum circumscriptum, Litho- phyllum corallinae and Lithothamnion glaciale). Estuarine taxa also showed various distributional patterns (cf. Table 3). Ulvaria oxysperma and Ascophyllum nodosum ecad scor- pioides occurred at 36% and 68% of the sites, respectively. Most taxa either exhibited a restricted (<20%) inner distribution (i.e., Capsosiphon fulvescens, Microspora pachyderma, Giffordia gran- ulosa, Polysiphonia denudata and P. nigra) or a patchy/restricted (<20%) distribution (e.g., Bryopsis plumosa, Enteromorpha clath- rata, Percursaria percursa, Ectocarpus fasciculatus, Giffordia granulosa, Petroderma maculiforme, Pseudolithoderma extensum, Ralfsia bornetii, Sorapion kjellmanii, Ceramium strictum, Dasya baillouviana, Erythrotrichia carnea, Leptophytum laevae, Lo- 1993] Mathieson et al.— York River Algae 309 Table 3. Extended, continued. 8 9 10 11 12 13 14 15 16 17 18 19 20 Longevity %* x x X. & ke ox Bee 50 (45) x =x Pér,? 9.1 (10) Xe Per. 22.7 (20) x x Per. 9.1 (10) x Per 13.6 (5) Ann 4.5 (1) Ann. 4.5 (1) Ann. 9.1 (0) x x x Me i ee x x Ann. 68.2 (65) Per. 9.1 (0) Per: 9.1 (0) AAnn. 4.5 (0) x Ann. 4.5 (5) 28 21 19° 262) 21: 27g ae a ee mentaria baileyana, L. clavellosa, Phymatolithon lenormandii and Polysiphonia subtilissima). FLORISTIC COMPARISONS Figure 6 compares species richness patterns in the York River Estuary with two nearshore open coastal sites (Cape Neddick and Sea Point) and twelve other estuarine habitats in southern Maine, New Hampshire and Massachusetts. Ninety-five taxa were re- corded at Cape Neddick, while Sea Point and the York River Estuary both had 72 taxa, albeit their species compositions were very different. A comparison of the York, Great Bay, Hampton- Seabrook and Merrimack River Estuaries suggests an interme- diate pattern of species richness (Figure 6). A “typical” estuarine reduction pattern (cf. Mathieson and Penniman, 1991) is evident within the ten primary subareas of the Great Bay Estuary System (Table 1), ranging from the Piscataqua River (143 taxa), Little Bay (130 taxa), Great Bay (90 taxa), and inner riverine sites like the Salmon Falls (16 taxa) and Winnicut Rivers (4 taxa). The York River has a more diverse flora than the Hampton-Seabrook Estuary (63 taxa) or the tidal waters of the Merrimack River (25 310 Rhodora [Vol. 95 taxa). Thus, the York River has its greatest affinities to several mid-estuarine habitats within the Hampton-Seabrook and Great Bay Estuaries (Table 1). A comparison of the number and per- centage of York River taxa within these same open coastal and estuarine sites confirms many of the above described patterns (cf. Figure 6). The greatest floristic affinities were between the York and Piscataqua Rivers (i.e., 68 taxa and 94%), followed by Little Bay (63 and 88%), Great Bay (52 and 72%) and the Hampton- Seabrook Estuary (41 and 57%). By contrast, the York River showed reduced affinities with the Merrimack River (20 taxa and 28%), as well as a variety of inner estuarine sites (e.g., Winnicut River, 4 taxa and 5.5%). The York River only shares a modest number of taxa with the nearby coastal sites at Cape Neddick (42 and 58%) and Sea Point (41 taxa and 57%). The numbers of seaweed taxa at each site within the York River Estuary, plus those at Cape Neddick and Sea Point, are illustrated in Figure 7 and Table 3. Overall, patterns of species richness within the estuary were irregular and reduced in numbers. A total of 16-28 taxa per site were recorded within the York River (mean = 21.4 + 3.2 taxa), while Cape Neddick and Sea Point had 95 and 72 taxa, respectively. A comparison of the mean number of taxa per site within the York River and twelve other estuarine habitats is given in Figure 8. The York River is comparable to the Piscataqua River (mean = 25.3 + 24.9) and Great Bay (mean = 22.5 + 18.0), while it is lower than Little Bay (mean = 34.1 + 29.0). By contrast, the York River has a greater number of taxa per site than several mid- and inner-estuarine areas like the Oyster (mean = 12.6 + 7.9) and Winnicut Rivers (mean = 1.3 + 1.6). It is also more diverse than the Hampton-Seabrook (mean = 10.5 + 5.5) and Merrimack River Estuaries (mean = 3.5 + 5.2). Thus, the York River is more spatially consistent (mean = 21.4 + 3.2) than the Piscataqua River, Little Bay and Great Bay, and is analogous to several inner-estuarine sites like the Cocheco (+4.3) and Salmon Falls Rivers (+3.4)—albeit the species rich- ness within the latter rivers is much lower than in the York River (Table 3). LONGEVITY PATTERNS Of the 131 taxa recorded from the York River, Cape Neddick and Sea Point, 62 were annuals (47%), 67 (51%) perennials and TOTAL TAXA * < x < ~ c Ww —— > a =< ee ec 2 fe) : 3s = ae Oe Geo « i OC Rm fo ee ee ge ee oe ane 6. CU cy > ee | z < He Figure 8. Mean (+SD) number of green, brown and red algal taxa/site within the York River Estuary and fourteen other geographical areas. See Table 1 for a synopsis of number of collection sites/area. son, 1978; Mathieson and Hehre, 1986). Aside from such hydro- graphic variability, a conspicuous reduction in the amount of rocky substrata is present upstream within the York River Es- tuary. A combination of both physical factors probably causes gross fluctuations in numbers of taxa per site, a general reduction in the number of species upstream and the ultimate dominance of riverine habitats by ephemeral Chlorophyceae (Coutinho and Seeliger, 1984; Josselyn and West, 1985; Ketchum, 1983; Mathie- son and Penniman, 1986b; Wilkinson, 1980). Several specific examples of limited rocky substrata can be cited. Foremost, the upper reaches of the York River are com- posed of steep channels and dissected saltmarshes, which have a limited flora dominated by Blidingia minima, Enteromorpha spp.., Ulva lactuca, Urospora spp., Ulothrix spp., Ulvaria spp., Rhizo- clonium riparium, Fucus vesiculosus, Ascophyllum nodosum and A. nodosum ecad scorpioides. Many of the latter taxa occur as unattached or entangled masses (Norton and Mathieson, 1983). For example, the marsh ecad scorpioides of A. nodosum grows abundantly at diverse sites (Table 3), forming entangled masses amongst Spartina alterniflora. By contrast, only a single detached 314 Rhodora [Vol. 95 marsh ecad of F. vesiculosus was found at station 16 where it grew as a partially embedded fragment (Appendix 1, Table 3). The morphology and proliferous habitat of the latter plant are similar to F. vesiculosus megecad limicola (Baker and Bohling, 1916). Although marsh ecads of F. vesiculosus are common in Europe (cf. Norton and Mathieson, 1983), they are uncommon and spo- radically distributed in New England (Chapman, 1939). Distributional patterns of several epiphytic algae (e.g., Elachista fucicola, Pilayella littoralis, Callithamnion byssoides, Ceramium rubrum and Polysiphonia lanosa) paralleled the occurrence of the two dominant canopy fucoids, Ascophyllum nodosum and Fucus vesiculosus (cf. Table 3), presumably because of reduced avail- ability of rocky substratum. The occurrence of epiphytic F. vesicu- losus on A. nodosum ecad scorpioides at two sites (i.e., 7 and 14) should also be noted, as we know of no other records of its oc- currence on the ecad. Epiphytic Fucus may occasionally grow on attached A. nodosum, particularly on isolated rock outcrops with- in estuarine mudflats such as in Great Bay, New Hampshire (A. Mathieson, unpubl. data). Patchy distributional patterns of sev- eral attached taxa on scattered rock outcrops should also be noted, including Ectocarpus spp., Laminaria saccharina, Bangia atro- purpurea, Corallina officinallis, Clathromorphum circumscrip- tum, Leptophytum laeve, Lithophyllum corallinae, Lithothamnion glaciale, Phymatolithon lenormandi and Mastocarpus stellatus (Table 3). Lastly, the restricted estuarine occurrence of Callo- phyllis cristata on Littorina littorea at station 8 (1.6 miles inland) may have been due to its host’s migration from the open coast. A comparison of the York, Piscataqua and Merrimack Rivers is instructive as all three are outer estuarine habitats with con- trasting floras (cf. Mathieson and Fralick, 1972, 1973; Mathieson and Penniman, 1986b; Mathieson et al., 1981). Open coastal species drop out after approximately 1.0 mile on the Merrimack versus 2-3 miles on the York (cf. Table 3) and approximately 8.5 miles on the Piscataqua River (Reynolds and Mathieson, 1975; Mathieson et al., 1983). The latter site on the Piscataqua River (i.e., Dover Point) has strong tidal currents (approx. 5 knots) and diverse rocky substrata. A comparison of temperature and salinity conditions within the York River (cf. Figures 3 and 4) shows that they are comparable to those found at Dover Point (i.e., —2.0 to 24.1°C and .9 to 30.3%), while hydrographic variability (partic- ularly salinity and siltation) is maximal on the Merrimack (cf. 1993] Mathieson et al.— York River Algae ai3 Mathieson and Fralick, 1973; Miller et al., 1971). The Merrimack River has the most depauperate flora (only 25 taxa), the York is intermediate (72 taxa) and the Piscataqua the most diverse (144 taxa). Several adverse physical factors (cf. above), as well as do- mestic and industrial pollution, are probably responsible for the reduced diversity of the Merrimack River. Besides being one of the most polluted rivers in New England (cf. Anonymous, 1984, 1987; Jerome et al., 1965; Lyons et al., 1982; Mathieson and Fralick, 1973; Miller et al., 1971), the Merrimack is also one of the largest sources of freshwater discharge into the Gulf of Maine (Apollonio, 1979). As noted by several investigators (Clokie and Boney, 1980; Daly and Mathieson, 1977; Edwards, 1972; Littler, 1980; North et al., 1964; Patrick, 1963, 1964, 1973; Round, 1981; Wilkinson, 1980), patterns of low species diversity are typical responses to stress, often allowing only a few tolerant species to dominate in both numbers and biomass. For example, the abun- dance of many ulotrichalean green algae, such as Ulva lactuca, Enteromorpha and Ulvaria (Monostroma) spp., typifies many eu- trophied estuarine habitats (Cotton, 1910; Fritsch, 1935; Sawyer, 1965). The latter species are not only tolerant of extremes of pollution but of gross fluctuation in hydrographic conditions. While industrial development and eutrophication are limited on the York River (cf. above), they are intermediate on the Piscataqua and highest on the Merrimack (Mathieson and Penniman, 1991). Aside from these contrasting patterns, the Piscataqua River also has the greatest diversity of habitats (e.g., tidal rapids, saltmarsh- es, boulders, cobbles, shingle, etc.) and stable substrata. Josselyn and West (1985) compare species richness within sev- eral estuaries, including San Francisco Bay and Great Bay. Beyond the mouth of the former site a relatively low number of taxa occurs upstream (i.e., 21-61 taxa/site). Similar patterns of low species richness are known from several other estuaries with high sedimentation and limited solid substrata, including the York, various subsets of the Great Bay and Hampton-Seabrook Estu- aries (Figure 8), the Chesapeake Bay (Mathieson and Fuller, 1969; Orris 1980), several British and Dutch estuaries (Hartog, 1967; Nienhuis, 1975; Wilkinson, 1980), and several Icelandic fjords (Munda, 1969, 1978). Within such turbid habitats there is a rel- atively small pool of common estuarine species (Munda, 1969, 1972; Orris, 1980; Wilkinson, 1980). For example, Orris (1980) lists 62 taxa from the upper reaches of the Chesapeake Bay in 316 Rhodora [Vol. 95 Maryland, with Cladophora, Enteromorpha, Ulva, Ceramium and Polysiphonia being most frequent. A comparison of the species richness within the York River and the Great Bay Estuary System (i.e., all ten subareas in Figure 6) also showed pronounced differences, with 72 and 168 taxa occurring at the two sites, respectively. As noted above, such contrasting patterns are probably due to a variety of factors, in- cluding habitat diversity, hydrographic stability, pollution, avail- ability of rocky substrata, etc. (Murray and Littler, 1989; Murray et al., 1980). With the exception of the crustose brown alga So- rapion kjellmanii, all of the taxa recorded from the York River are known from the Great Bay Estuary (Mathieson and Hehre, 1986). By contrast, a greater number of freshwater, coastal and estuarine taxa are known from the Great Bay Estuary than the York River. Several of these estuarine taxa represent disjunct “southerly” plants (Bird et al., 1976; Mathieson and Hehre, 1986; Novaczek et al., 1987) that are either absent (e.g., Chondria bail- eyana, Gracilaria tikvahiae, Polysiphonia elongata) or rare within the York River (e.g., Bryopsis plumosa, Ceramium strictum, Das- ya baillouviana, Lomentaria baileyana, Polysiphonia denudata and P. subtilissima). A similar pattern exists within the Hampton- Seabrook Estuary: the absence of many “southerly” taxa and a dominance of saltmarshes (Mathieson and Fralick, 1972). Pre- sumably these warm water plants are either unable to grow and/ or compete with other saltmarsh-inhabiting taxa. The vast sub- tidal mudflats of the Great Bay Estuary may be some of the most suitable habitats for such warm water taxa in northern New En- gland; these habitats are essentially absent from the York and Hampton-Seabrook Estuaries, except near outer estuarine sites where water temperatures are too low for optimal growth and/or survival (Figure 4; see also Fralick and Mathieson, 1975; Guo and Mathieson, 1992; Novaczek et al., 1987). Although this study has primarily dealt with the algal flora of the York River Estuary, our coastal collections have documented two new records from southern Maine and New Hampshire. Fore- most, the “arctic” taxon Porphyra amplissima, as recently defined by Bird and McLachlan, 1992, has been recorded from Cape Neddick (Table 3). Previously it was included under P. miniata (Taylor, 1957). An evaluation of herbarium specimens at Harvard University (FH) and the University of New Hampshire (NHA) has 1993] Mathieson et al.— York River Algae 317 documented its presence from several open coastal sites in Maine, as well as nearshore and estuarine sites within New Hampshire (Hehre and Mathieson, 1993). Some of these samples date to ca. 1880. The presence of P. amplissima near the mouth of the York River suggests it may be found within this habitat as well. Recent expansion of the “alien” species Codium fragile ssp. tomento- soides (Van Patten, 1992) should also be noted (Mathieson and Hehre, 1986). It was initially found at Cape Neddick in November of 1989 and is now abundant at the Isles of Shoals (particularly on Appledore and Smuttynose Islands, Maine). One attached specimen was recently found on the nearshore open coast of New Hampshire at Jaffrey Point, Newcastle (September, 1992), while drift specimens were collected during the same period at nearby Rye Ledge (Rye, New Hampshire) and within the inner reaches of the Great Bay Estuary (Weeks Point). Further expansion of its distributional range may provide a variety of problems (Van Pat- ten, 1992) ACKNOWLEDGMENTS We would like to express our sincere gratitude to the following people: ‘‘Ned”’ McIntosh, former captain of the R/V Jere A. Chase, who assisted with several of the seaweed collections and hydro- graphic studies. Richard Langan allowed us to use his unpublished hydrographic data from the York River, the acquisition of which was supported by the U.S. Department of Navy. Several indi- viduals, including Ellen Crane, Chris Emerich Penniman, and Morgan Hardwick-Witman, are acknowledged for their help with collecting many of the samples from the York River Estuary. Lastly, the first author would like to acknowledge the encourage- ment, support and help of his wife, Myla Mathieson. LITERATURE CITED ANONYMous. 1984. National Status and Trends Program for Marine Environ- mental Quality. NOAA, Washington, D ——.. 1987. Strategic Assessment of Near Coastal Waters, Northeast Case Study—EPA/NOAA Team on Near Coastal Waters, Washington APOLLONIO, S. 1979. The Gulf of Maine. Courier of Maine Books. Rockiand, E. BAKER, S. M. AND M. H. BouLING. 1916. On the brown seaweeds of the salt 318 Rhodora [Vol. 95 marsh. II. Their systematics, relationships, morphology and ecology. J. Linn. Soc. London, Bot. 43: 325-380. Bett, H. P. AND C. MACFARLANE. 1933. The marine algae of the Maritime Provinces of Canada. I. List of species with their distribution and prevalence. Can. J. Res. 9: 265-279. Birp, C. J., T. EDELSTEIN AND J. MCLACHLAN. 1976. Investigations of oi marine algae of Nova Scotia. XII. The flora of Pomquet Harbour. Can. J. B 2726-2737. AND . McLACHLAN. 1992. Seaweed Flora of the Maritimes 1. Rho- dophyta— The Red Algae. Biopress Ltd., Bristol, England CHAPMAN, V. J. 1939. Some algal complexities. A. Rhizoclonium tortuosum (Dillw.) Kiitz. and Chaetomorpha turtuosa Kiitz. B. Marsh forms of Fucus spiralis and F. vesiculosus. Rhodora 41: 19-28. CioxrE, J. J. P. AND A. D. Boney. 1980. The assessment of changes in intertidal ecosystems following major reclamation work: framework for interpretation of algal-dominated biota and the use and misuse of data. Jn: J. H. Price, D E. G. Irvine and W. F. Farnham, Eds., The Shore ne adn Vol: 2: Ecosystems. Academic Press, New York. Cotton, A. D. 1910. On the growth of Ulva latissima in water polluted by sewage. Bull. Misc. Inform. Roy. Bot. Gard. Kew. pp. 15-19 CouTINuo, R. AND U. SEELIGER. 1984. The horizontal distribution of the benthic algal flora in the Patos Lagoon Estuary, Brazil, in relation to salinity, sub- stratum and wave exposure. J. Exp. Mar. Biol. Ecol. 80: 247-257. intertidal seaweeds and selected invertebrates at Bound Rock, New Hamp- shire, U. S. A. Mar. Biol. 43: 45-55. Dory, M. S. AND J. NEwHouse. 1954. The distribution of marine algae into estuarine waters. Am. J. Bot. 41: 508-515. Eaton, D. C. 1873. List of marine algae collected near Eastport, Maine, in August and September, 1873, in connection with the work of the U. S. Fisheries Commission, under Prof. S. F. Baird. Trans. Conn. Acad. Arts and Sci. 2: 343-350. Epwarps, P. 1972. Benthic algae in polluted estuaries. Mar. Poll. Bull. 3: 55-60. FRALICK, R. A. AND A. C. MATHIESON. 1975. Physiological ecology of four Polysiphonia species (Rhodophyta, Ceramiales). Mar. Biol. 29: 29-36. E; 35. The Structure and a of the Algae, Vol. I. Cambridge Univ. Press, Cambridge, England. Guo, Z. AND A.C. MATHIESON. 1992. Physiological ecology of four ulvoid green Heure, E. J. anp A.C. MATHIESON. 1970. Investigations of New England marine algae. III. Composition, seasonal occurrence, and reproductive periodicity of the marine casi in New ‘Hampshire. Rhodora 72: 194-239 1993. ll et Hus: new records of an Agrctic” seaweed in ‘southern Maine, New Hampshire and northern Massachusetts. Rhodora 95: 184-187. JEROME, W. C., A. P. CHEsMorE, C. O. ANDERSON, JR. AND F. Grice. 1965. A study of the Marine Resources of the Merrimack River Estuary. Monograph ayn h wry } ie ai a - eee 1993] Mathieson et al.—York River Algae 319 Ser. No. 1, Div. of Mar. Fish., Dept. of Natl. Res., Commonwealth of Mas- tts. JOsSELYN, M. N. A. C. MATHIESON. 1978. Contribution of receptacles from the fucoid Ascophyllum nodosum to the detrital pool of a north temperate estuary. Estuaries 1: 258-261 AND . 1980. Seasonal influx and d neosies yr litter in a north temperate estuary. i see 71: 197-208. AND 1985. The distribution and temporal dynamics of the estuarine ions community of San Francisco Bay. Hydrobiologia 129: 152. £ +1} Ketcuum, B. H. 1983. Estuarine characteristics. Jn: B. H. Ketchum, Ed., systems of the World, Vol. 26: Estuaries and Enclosed Seas. Elsevier Se Publ. Co., Amsterdam Kinng, O. 1970. Tempexature, In: O. Kinne, Ed., Marine Ecology, Vol. 1, Part 1. Wiley-Interscience, New Yor . 1971. Salinity. Jn: O. Kinne, Ed., Marine Ecology, Vol. 1, Part 2. Wiley- Interscience, New York. Litter, M. M. 1980. Southern Cali community structure and variability. Jn: J. H. Price,.D:E, G. Irvine and W. F. Farnham, Eds., The Shore Environment, Vol. 2: Ecosystems. Academic ee 1 4 Lyons, W. B., P. B. ARMSTRONG, R. P. O’Nem, AND H. E. GAupeTTeE. 1982. Trace metal discharge into Great Bay Estuary, New Hampshire. Univ. New Hampshire Sea Grant Technical Rept., UNH-SG-176. Mann, K. H. 1972. Ecological energetics of the seaweed zone in a marine bay on the Atlantic coast of Canada. II. Productivity of the seaweeds. Mar. Biol. 14: 199-209. . 1982. Ecology of Coastal Waters: A Systems Approach. Univ. California Press, Berkeley. Matuieson, A. C. 1978. The distribution, composition and phenology of New England estuarine seaweeds. J. Phycol. 14 (suppl.): 37. 1979. Vertical distribution and longevity of subtidal seaweeds in north- ern New England, U.S.A. Bot. Mar. 20: 511-520. AND R. A. FRALICK. 1972. Investigations of New England | marine algae V. The algal vegetation of the Hampton-S near psi New Hampshire. Rhodora 74: 406-435. 1973. Benthic algae and vaseine plants of the lower Mer- nae River and adjacent shoreline. Rhodora 7 AND S. W. Futter. 1969. A preliminary faniatueictioen of the benthonic marine algae of the Chesapeake Bay region. Rhodora 71: 524-534 ——— AnD E..J. HEHRE. sisi The composition, a occurrence x and re- ti New Hampshire. Rho- dora 84: 41 1-437. AND ———. 1983. The composition and seasonal periodicity of the marine Chlorophyceae in New Hampshire. Rhodora 85: 275-299. ———— AND ———. 1986. A synopsis of New Hampshire seaweeds. Rhodora 88: 1-139. ———, C. Neerus anp C. EmeEri . 1983. Benthic ecology in an CH PENNIMAN estuarine tidal rapid. Bot. Mar. 26: 213-230. 320 Rhodora [Vol. 95 . PENNIMAN. 1986a. A phytogeographic interpretation of the marine fora from the Isles of Shoals, is S. A. Bot. Mar. ann 13-434. AND 1986b. S of N w England seaweeds along an open coastal-estuarine Scorient. i. Mar. aie 161- we AND ——-. 1991. not England estuarine and open coastal seaweed populations. Nova Hedwigia 52: 453-485. ——,P.K SSE AND E. TVETER-GALLAGHER. 1982. The effects of ice on Ascophyllum nodosum within the Great Bay Estuary System of New Hampshire-Maine. J. ae col. 18: 331-336. ——., N. B. REYNoLDs AND E. J. HEHRE. 1981. Investigations of New England marine algae. II. The species composition, distribution, and zonation of seaweeds in the Great Bay Estuary System and the adjacent open coast of New Hampshire. Bot. Mar. 24: 533-545. ———,, E. Tverer, M. DALy AND J. Howarpb. 1977. Marine algal ecology in a New Hampshive tidal rapid. Bot. Mar. 20: 277-290. Miter, B., D. NORMANDEAU, G. PIEHLER, P. HALL, A. MATHIESON, R. FRALI D. TuRGEON, P. MAHONEY AND W. Owen. 1971. Ecological study of Wier. rimack River Aan AORN: presented to U. S. Army, Corps of Engineers by Normandeau Associates, Inc. and Viast Inc MunpaA, I. 1969. Differences in the algal vegetation of tio Icelandic fjords, ie cia ap Iceland) and Reydafjérdur (East Iceland). In: R. Margalef, ., Proceedings of the Sixth International Seaweed Symposium, Santiago ‘. asain Spain, sons aae de la Marine Mercante, Made id. 1972. General features of the Icelandic Coast. Acta nat. Islandica 21: 1-34. . 1978. Salinity dependent distribution of benthic algae in estuarine areas of oe pasar Bot. Mar. 21: 451-468. Murray, M. M. Littter. 1989. Seaweeds and seagrasses of southern ap ene t lists for twenty-one rocky intertidal sites. Bull. So. ere Acad. Sci. 79. , M. M. Litrter ANDI. A. ABpotr. 1980. Biogeography of the California rine algae with emphasis on the southern California islands. Jn: D. M. nen Ed., The California Islands: Proceedings ofa ‘soar sicoetaas Sym- Harv. (Chlorophyceae: Cladophorales) in the estuarine areas of the River Rhine, Meuse and Scheldt. Thesis, University of Groningen, Bronder Offest NorAL.L, T. L., A. C. MATHIESON AND C. EMERICH PENNIMAN. 1982. Nutrient and hydrographic data for the Great Bay Estuarine System, New Hampshire- Maine, Part 1, September, Recta 1975. Jackson Estuarine Lab- oratory Contrib, No. 150, UNH D/TR Nortn, W. J., K. A. CLENDENNING, L. G. Jom, J. B. Lackey, D. L. LeiGHTton, HUL, M. C. SARGEANT AND H. L. Scorren. 1964. Investigation of the Effects of Discharged Wastes on Kelp. State Water Quality Control Bd. (California), Publ. 26. Norton, T. A. ANDA.C. MATHIESON. 1983. The biology of unattached seaweeds. 1993] Mathieson et al.— York River Algae 321 n: F. E. Round and D. J. Chapman, Eds., Progress in Phycological Research, Vol. 2. Elsevier, Amsterdam. NovaczeK, I., C. J. Brrp AND J. MCLACHLAN. 1987. Phenology and temperature tolerance of the red algae Chondria baileyana, Lomentaria baileyana, 2 nar globulifera, and Dasya baillouviana in Nova Scotia. Can. J. Bot 57-62. Orris, P. K. 1980. A revised species list and commentary on the macroalgae of the Chesapeake Bayt in Maryland. Estuaries 3: 200-206. Patrick,R. 1963. Thest ying ecological oo Ann. New York Acad. Sci. 108: 359~364. 1964. A discussion of natural and abnormal diatom communities. Jn: D. E. Jackson, Ed. — and Man. ‘Plenum Press, New York . 1973. Use of algae, In: J. Cairns and K. “s Dickson, Eds., Biological Methods for the Asean of Water Quality. ASTM, STP 528. PATTEN, VAN, P. 1992. Aliens among us. Nor’Easter (fall/winter), pp. 8-13. , C. A. AND A. hee MATHIESON. 1985. Photosynthesis of Gracilaria Bi eee Mc Lachl ay Estuary, New Hampshire. Bot. Mar. 28: 427-435. REYNOLDs, N. B. AND A. C. MATHIESON. 1975. Seasonal occurrence and ecology of marine algae in a New Hampshire tidal rapid. Rhodora 77: 512-533. Rounpb, F. E. 1981. The Ecology of Algae. Cambridge Univ. Press, Cambridge. SAwyver, C. M. 1965. The sea lettuce problem in Boston Harbor. J. Water Poll. Cont. Fed. 37: 1122-1133. Soutn, G. R. AND I. Titttey. 1986. A Checklist and Distributional Index of the Benthic Algae of the North rg Ocean. Spec. Publ. Huntsman Mar. Lab. and British Museum (Natural His Stone, R. A., E. J. HeHRE, J. M. ssaaegany AND A. C. MATHIESON. 1970. A preliminary checklist of the marine algae of Campobello Island, New Bruns- wick, Canada. Rhodora 72: 313-338. STRICKLAND, J. D. H. AND T. R. Parsons. 1968. A pene Handbook of Seawater Analysis. Bull. Fish. Res. Bd. Canada 167, Taytor, W. R. 1957. The Marine Algae of the se iareta got of North America. Univ. Michigan Press, Ann Arbor Wesper, E. E. anp R. T. Witce. 1971. Benthic salt marsh algae at Ipswich, Massachusetts. Rhodora 73: 262~291. Witkinson, M. 1980. Estuarine benthic algae and their environments: a review. In: J. H. Price, D. E. G. Irvine and W. F. Farnham, rs The Shore Envi- ronment, Vol. 2: Ecosystems. Academic Press, New al Yor SON. 1976. ti id of attached algae in estuaries. Mar. Poll. Bull. 7: 183-184. DEPARTMENT OF PLANT BIOLOGY JACKSON ESTUARINE LABORATORY UNIVERSITY OF NEW HAMPSHIRE DURHAM, NH 03824 ane Appendix 1. York River Rhodora [Vol. 95 d adjacent ope tal study sites (Cape Neddick and Sea Point) in southern Maine. ta- from tion coast Longitude Latitude Description Open Coastal NL 0.0 SP 0.0 York River 1 0.0 ~) 2 ‘pb Le) = - rs S oo oo — On 70°35'25”W 70°30'32”W 70°37'36”W 70°38'12”"W 70°38'25"W 70°38'45"W 70°38'32”W 70°39'00"W 70°39'07"W 70°39'45"W 43°09'56’"N 43°05'26"N 43°08'06’N 43°07'18’N 43°07'26°N 43°07'31’"N 43°07'44’N 43°07'49"N 43°07'42’"N 43°07'56’"N Nubble Light, Cape Neddick, Maine; an exposed site with substrata pri- marily consisting of massive rock outcrops Sea Point, Kittery, Maine; semi-ex- posed site with substrata consisting of boulders (cobbles), rock out- crops and sand Cove between Roaring Rock Point and Cow Beach Point; mostly rocky but some sand and gravel habitats also present South bank of river, across from Stage Neck; a rocky habitat with limited mudflats North bank of river by Stage Neck Condominiums; a rocky habitat with boulders and rock outcrops South bank of river, just north of the new Rte. 103 bridge; a variety of habitats including mudflats, salt- marsh and some rocks North bank of river at new Rte. 103 bridge; habitats include mudflats, Bridge and Rte. 103 bridge; sub- stratum varied, including mudflats, sand, cobbles and boulders; fast- flowing water draining into York River from Barrells Millpond North bank of river at Sewall Bridge; substrata consisting of mudflats and saltmars North bank of river at Sewall Bridge, upstream from #7; substrata main- ly consisting of sand, gravel, and some limited saltmarsh with scat- tered small rocks 1993] Mathieson et al.— Appendix 1. York River Algae 323 Continued. Sta- tion Miles rom coast Longitude Latitude Description i) Oo _ N ve) ay ps eal On — ~ — oo — \o 2.0 70°40'32”"W 2.4 70°40'25"W 2.4 70°40'15”"W 2.6 70°40'45"W 2.8 70°41'10"W 3.4 70°43'30"W 3.4 70°41’25"W 3.6 70°41'45"W 4.0 70°42'05"W 4.2. 70°42'30"W 4.4 70°43'15"W 43°07'52"N 43°07'37’N 43°07'50"N 43°08'04"N 43°07'56"N 43°07'51"N 43°08'13"N 43°08'18’"N 43°08'32’N 43°08'50"N 43°09’01"N North bank of river, midway be- tween Sewall Bridge and Rams- head Point; extensive mudflats with rock outcrops and a narrow band of saltmarsh South bank of river on River Mead- ow Lane; just south of Ramshead Point; extensive mudflats North bank of river approaching Ramshead Point; substrata consist- ing of rock outcrops, large boul- ders, saltmarshes and mudflats North bank of river, just opposite Ramshead Point; near a private dock with boulders and cobbles, as well as mudflats and saltmarshes North bank of river, just down- stream from Rice’s Bridge on Rte. 1; some rocky areas but mostly mudflats with some saltmarsh South bank of river, at Rice’s Bridge on Rte. 1; substrata primarily con- sists of mudflats and saltmarsh, with some bes rocks; two outfall pipes pre: North bank of river, ae I-95 bridge; extensive mudflats and very limit- ed saltmarsh; rocks and cobbles occur at bridge pilings Mouth of Cider Hill Creek (north bank) just above the I-95 bridge; extensive mudflats with scattered rocks and some saltmars Mouth of Boulter Pond Creek (north bank) between I-95 and Scotland Bridge; substrata primarily consist- h substrata primarily consisting of 324 Rhodora [Vol. 95 Appendix 1. Continued. ta- from tion coast Longitude Latitude Description mudflats, saltmarsh and bridge pil- ings 20 4.7 70°43'30"W 43°09'12”N___ North bank of river, near a large farm—upstream from Scotland Bridge on Rte. 91; substrata pri- marily consisting of mudflats and some saltmarsh RHODORA, Vol. 95, No. 883/884, pp. 325-341, 1993 reciept STUDIES IN THE MICONIEAE (MELASTOMATACEAE). V. MICONIA : ae RYS, CIRCUMSCRIPTION AND RELATIONSHIPS WALTER S. JUDD AND ELIZA KARPOOK ABSTRACT Miconia stenobotrys, a member of section Chaenopleura, is considered to con- stitute a species from a phenetic as well as a phylogenetic standpoint. It occurs widely in the Cordillera Central/Massif du Nord of Hispaniola, growing mainly in forests dominated by Pinus soap The species is circumscribed broadly, including M. azuensis, M. artibotinensis, M. buchii, and M. leptoneura, as well as populations with very seulcttie 4-flanged stems from the Parque Nacional J ose del: Carmen Ramee! Loma la Vieja region | of the Dominican Republic. A is presented. Miconia stenobotrys is distinguished from its presumed closest rel- atives, i.e., M. Krugii, M. samanensis, and M. zanonii, by its falcate leaves that are slightly to strongly V-folded along the midvein and distinctive elongated inflorescences. It is quite variable in several morphological features, especially in extent of development of U-shaped stem flanges, number of multicellular, long- stalked, gland-headed hairs on the stems and leaves, degree of sche ty of the leaf margin, placement of secondary veins of the lamina, and flower size. However, no phenetic discontinuities were discerned among analyzed en ae and the above listed characters often vary within populations. Key Words: Miconia stenobotrys, Miconia sect. Chaenopleura, Melastomataceae, phylogenetic species concept INTRODUCTION In the course of field work in connection with a taxonomic revision of Miconia section Chaenopleura Benth. & Hook. and a floristic treatment of Melastomataceae for the Flora of Hispaniola a diverse array of populations representing a species (or several related species), i.e., the Miconia stenobotrys complex, was col- lected. These plants, easily recognized by their leaves that are falcate and slightly to strongly V-folded along midvein and their distinctive elongated cymose inflorescences, are referable to Mi- conia sect. Chaenopleura (see Cogniaux, 1891; Judd and Skean, 1991). The West Indian members of section Chaenopleura form a distinctive, diverse, an which may be diagnosed by the possession of an actinomorphic an- droecium, that is, the stamens form a radially symmetrical pattern 25 326 Rhodora [Vol. 95 around style (instead of being deflexed to one side of flower). The stamens are white and glabrous, and the obovate anthers each open by two longitudinal slit-like pores. The group is also char- acterized by globose fruits that turn from red to blue (or blue- white) at maturity and angular-obovoid seeds with a more or less smooth testa (see Judd and Skean, 1991, figure 10B). Fifty-six Antillean species of Miconia sect. Chaenopleura have been (or are in the process of being) described. As discussed herein, some of these appear to be synonymous with other species. A few un- described species, however, recently have been discovered (Judd, unpubl. data), and the group likely is represented by ca. 50 species in the Greater Antilles (see Judd and Beaman, 1988; Judd and Skean, 1987; Judd et al., 1988; Judd and Skean, submitted a and b; Liogier and Martorell, 1982; Le6én and Alain, 1957; Moscoso, 1943; Proctor in Adams, 1972). Specific limits among the populations comprising the Miconia stenobotrys complex have been especially troublesome, with some botanists recognizing several species, i.e., M. stenobotrys, M. azuensis, M. buchii, M. artibotinensis, and M. leptneura (Cognaux in Urban, 1908; Urban and Ekman in Urban, 1929; Moscoso, 1943) and others (in herbarium identifications) applying the names M. azuensis or M. stenobotrys more broadly. Miconia stenobotrys is here re-circumscribed broadly and a detailed description and an illustration are provided. Its infraspecific pattern of variation is assessed through traditional taxonomic procedures, i.e., a sur- vey of numerous populations based on field-studies and herbar- ium material, and a Principal Components Analysis (PCA). The species is characterized eco-geographically and a list of specimen citations is presented. Miconia stenobotrys, as re-circumscribed, is compared with several closely related species, and its phylo- genetic relationships are assessed briefly. MEASUREMENTS/HERBARIUM CITATIONS All measurements (except for plant height, flower and fruit color, which were taken from information given on specimen labels or observed in the field, or floral and fruit measurements, which were taken from liquid-preserved or rehydrated material) included in the species descriptions come directly from dried herbarium material. The inflorescence length was measured from terminal flower to point at which first branch-pair emerges from 1993} Judd and Karpook— Miconia a27 axis, and the peduncle is defined as the internode separating cyme from uppermost leaf pair. The two or more prominent veins running in convergent arches toward the leaf apex are termed secondary veins since they clearly branch from the midvein. The detailed description has a format similar to that of other descrip- tions published as part of a series of taxonomic studies in the Miconieae, which allows for easy comparison; see Judd and Bea- man (1988), Judd and Skean (1987), Judd et al. (1988), and Judd and Skean (submitted a and b). In the citation of specimens, abbreviations of institutions follow the eighth edition of Index Herbariorum (Holmgren et al., 1990). PHENETIC STUDIES Miconia stenobotrys, as traditionally delimited, consistently has been confused with several phenetically similar “species,” i.e., M. azuensis, M. buchii, M. artibotinensis, and M. leptoneura. In addition, a population of plants clearly belonging within the Mi- conia stenobotrys complex, but with 4-flanged stems, i.e., two flanges extending below each point of petiole attachment and joining to form a conspicuous, U-shaped flange ca. 1-2 mm broad at the adjacent lower node, recently has been discovered in the Dominican Republic, prov. San Juan, near the “Forestry House” above Los Frios, in the Parque Nacional José del Carmen Ra- mirez; see Garcia 1245, Judd 6693, 6694, 6698, Skean 3269. Somewhat similar plants were collected by Eric Ekman on the nearby Loma la Vieja, i.e., Ekman H13410 and H13441. Typi- cally, the twigs of M. stenobotrys (or the other species listed above) are square to rectangular in cross-section, but more or less lack flanges. The phenetic separation between the above listed species of the Miconia stenobotrys complex, along with the plants with conspicuously 4-flanged stems, was evaluated by a Principal Com- ponents Analysis of 114 flowering or fruiting shoots representing the full range of observed variation; see discussion of this tech- nique in Sneath and Sokal (1973) and Wiley (1981). This analysis employed the CLUSTAN mainframe computer program, version 3.2 (Wishart, 1987). Variation in 12 continuously varying or “count” characters of leaf, stem, and inflorescence morphology (Table 1) was scored for each specimen and the taxa ordinated along the first two principal components (Figure 1). These char- acters show variation among populations of the Miconia steno- 328 Rhodora [Vol. 95 able 1. Characters used in PCA of Miconia stenobotrys complex. (All node and leaf measurements taken at third node below cyme , Petiole length hWNe 3 i =] p _ . Distance passes base of lamina and point at which major secondary veins join with midve Distance seid major secondary veins and leaf margin (as measured at lamina midpoint . Number of teeth per 2 cm (centered at lamina midpoint) . Length of largest tooth (located at any point along leaf margin) . Flange width (measured at widest point) Number of multicellular, long-stalked, gland-headed hairs per internode . Inflorescence length 11. Pedicel length 12. Basal inflorescence branch length at OC OND botrys complex, and several have been used in specific delimi- tations within the group (see Cogniaux in Urban, 1908; Urban and Ekman in Urban, 1929). The resulting diagram (Figure 1), describing 60% of the variation contained in the original data set, indicates that there are no phenetic discontinuities among ana- lyzed specimens, although some divergence is evident between specimens from the Jarabacoa—Constanza/El Rubio regions, the Lagunas de Cenobi-La Cidra—La Leonor/Hinche regions, the Parque Nacional José del Carmen Ramirez-—El Frio region, an the Morne Belance region. Plants of the Rio Magua region (as represented by Valeur 736), the Restauracién—Rio Artibonito re- gion and populations to the west in Haiti (e.g., Ekman H3396, Ekman H6275), and Loma la Vieja region (as represented by Ekman H13440, Ekman H13441) are more or less intermediate between the phenetically divergent populations of the Constanza— Jarabacoa and the Lagunas de Cenobi—La Cidra—La Leonor regions (see Figures 1 and 2). The strongly stem-flanged plants of the Parque Nacional José del Carmen Ramirez—El Frio region, al- though distinctive, show an overlapping pattern of phenetic vari- ation with plants of the Lagunas de Cenobi—La Cidra—La Leonor region, the Hinche region, and the Loma la Vieja region (Figure 1). Considerable variability is present, but no correlated character states were found that could be used to delimit the members of this complex into two or more phenetically delimited species. Variation is present in all of the characters included in Table 1 1993] Judd and Karpook— Miconia 329 Figure 1. Plot of the first two principal components resulting from phenetic analysis of Miconia stenobotrys complex of Hispaniola. PC I describes 45% of the variation; PC II describes 15% of the variation. Circles = specimens collected in the Parque Nacional José del Carmen Ramirez-—El Frio region, i.e., Judd 6693, 6694, 6698, & Garcia 1245; triangles = specimens collected by Judd in the Con- stanza—Jarabacoa region, i.e., 2955, 5125, 5143, 6626, 6627, 6629, 6653, 6659; dots = other specimens. Circled dots Spite variation expressed by plants of a particular geographical region: | = Con oe region; 2 = Lagunas de Cenobi-La Cidra~La Leonor—Hinche ict 3 = Parque icionat José del Car- men Ramirez-E] Frio region; 4 = Morne Belance ee 5 = Rio Magua region; 6 = Loma la Vieja region; and 7 = Restauracién-Rio Artibonito region. (see description), but is especially remarkable in extent of flange development (stems lacking flanges to conspicuously 4-flanged), marginal condition (leaves entire to clearly serrate, and teeth with or without long-stalked gland-headed hairs), indumentum (stems and leaves with multicellular, long-stalked, gland-headed hairs, or such hairs lacking), and placement/number of secondary leaf veins (2 or 4, placed ca. .7-6 mm in from margin, and + basa to strongly suprabasal). Although not included in the phenetic analysis, flower size is also quite variable, especially the length of the petals and internal calyx lobes. (As with the vegetative and inflorescence characters, floral features show more or less contin- uous variation among populations.) Certain character states are 330 Rhodora [Vol. 95 20° tty oa gure 2. Geographical distribution of certain morphological features in Mi- conia stenobotrys. Dots = leaves serrate/serrulate nearly entire length, circles = leaves + entire, half-closed circles = intermediates; tail up = stems + flanged, no tail up = stems lacking flanges; tail down = marginal teeth associated with mul- ticellular long-stalked hairs, no tail down = marginal teeth not associated with long-stalked hairs (or teeth lacking). quite oat distributed within the species’ geographical range, although som mmon ina particular geographic region than another; see discussion below (and Figure 2). It is clear that different character states show independent geographical trends. In contrast, M. stenobotrys is easily distinguished from presum- ably related species (see Phylogenetic studies section). Thus the members of the Miconia stenobotrys complex are all referable to a single species, under either the phenetic or traditional-taxonom- ic species concepts. That is, Miconia stenobotrys is a morpholog- ical entity that shows no major internal discontinuities in its pattern of variation, and is separated from other such entities by a consistent morphological gap (see Judd, 1981, 1986; Judd and Beaman, 1988). In contrast to many members of Miconia sect. Chaenopleura, M. stenobotrys has been well collected; populations from the Con- stanza—Jarabacoa region are especially well represented in her- baria (see Figure 1). It is clear, however, that recent collecting has - 1993] Judd and Karpook— Miconia 33] expanded our understand of this species’ pattern of variation, e.g., specimens collected by Judd in the Parque Nacional José del Carmen Ramirez—El Frio region (Figure 1, circles) and in the Constanza-Jarabacoa region (Figure 1, triangles). Additional col- lections are needed from the Rio Magua and Loma la Vieja regions (Figure 1, groups 5 and 6). PHYLOGENETIC STUDIES Miconia stenobotrys has more or less glabrescent leaves, stems, inflorescence axes, and hypanthia, i.e., they do not have a con- spicuous indumentum of ferrugineous stellate-branched hairs found in most Antillean members of Miconia sect. Chaenopleura. Other glabrescent species include M. barkeri Urban & E. Ekman, M. calycina Cogn., M. krugii Cogn., M. leptantha Urban & E. Ekman, M. samanensis Urban, and M. zanonii Judd, Skean, & Beaman (all of Hispaniola), and these species may constitute a monophyletic group if the glabrescent condition is taken as syn- apomorphic. (The phylogenetic relationships of these species will be considered in more detail in a planned cladistic analysis of the Antillean species of section Chaenopleura.) Miconia stenobotrys is distinguished from the above listed glabrescent species (and from any other Antillean member of the section) by its V-folded and usually strongly falcate leaf blades and distinctive elongated inflorescences, i.e., 4.5-31.5 cm long, 2-5 cm wide, with 3 to 13 major branch-pairs (each of which is a + raceme-like cyme). Among the populations of M. stenobotrys these similarities are certainly synapomorphic, and the species is, therefore, tentatively considered to be monophyletic, i.e., a cladospecies (Donoghue, 1985; Mishler, 1985; Mishler and Brandon, 1987; de Queiroz and Donoghue, 1988). No population or group of populations within the Miconia stenobotrys complex can be segregated as distinct species because none are diagnosable, i.e., no population possesses a fixed char- acter that distinguishes it from all individuals belonging to other populations. All populations of M. stenobotrys, in contrast, are clearly diagnosable from populations of related species, e.g., M. krugii, M. samanensis, M. zanonii, by fixed character differences. Thus, Miconia stenobotrys also constitutes a phylogenetic species in the sense of Cracraft (1989), Nixon and Wheeler (1990, p. 218), and Davis and Nixon (1992, p. 427): “the smallest aggregation Ee Rhodora [Vol. 95 of populations . . . diagnosable by a unique combination of char- acter states in comparable individuals.” The geographical pattern of variation among populations of Miconia stenobotrys is outlined below. The populations in the northwestern portion of the Cordillera Central (Dominican Re- public), especially in those in the vicinity of Jarabacoa, Constanza, and El Rubio, show a high frequency of individuals with strongly toothed leaves (Figure 2). Some individuals have a very slight development of stem-flanges, although most lack such structures or have them restricted to the primary inflorescence axis. Plants with leaf blades having strongly suprabasal secondary veins are more frequent in these populations than in those further east in the Cordillera Central or in the mountains of Haiti. Populations of the Lagunas de Cenobi-La Cidra-La Leonor region (Cordillera Central, Dominican Republic) and the area northeast of Hinche (Massif du Nord, Haiti) are distinctive be- cause of their entire margined leaves with the secondary veins placed very near the lamina margin (Figure 2). Occasional plants with flanged stems occur in these regions. Populations in the vicinity of Loma la Vega and the “Forestry House” (near El Frio) at the Parque Nacional José del Carmen Ramirez (Cordillera Central, Dominican Republic) are very similar to those of the La Cedra, La Leonor, and Hinche populations, but show a much stronger and consistent development of stem- flanges esas pi and 3). Interestingly, the nearby and similar pop ulation near Las Lagunas (to the south ‘of Loma la Vieja) Eats stem flanges. Plants of the Restauracién—Rio Artibonito region (Cordillera Central, Dominican Republic) and populations to the west in Haiti (Figure 2) are slightly serrulate and show a variable devel- opment of stem flanges. Plants of the Rio Magua region (Cordil- lera Central, Dominican Republic) show variation in degree of marginal toothing and inflorescence length. These plants are thus somewhat intermediate between the three groups of populations discussed above (Figures | and 2). ay Figure 3. onli stenobotrys: a, habit; b, c, leaves; d, leaf margin; e, flower; petal; g, berries; h, stem, showing U-shaped flanges; i, stem, lacking flanges; j, stem, X-Section; i multicellular, long-stalked, gland-headed hairs, on stem. G : wgeawe Ha A LH . Ni 334 Rhodora [Vol. 95 The plants of the Morne Belance region (Massif du Nord, Haiti) are distinguished in that their leaves are strongly serrate with a long-stalked, gland-headed hair at the apex of each tooth. Plants of other populations usually lack such hairs on their teeth. Ju- venile leaves or leaves of rapidly growing shoots, however, usually possess such hairs, even in populations in which reproductive- shoot leaves are more or less entire. A similar pattern of variation, i.e., long-stalked, gland-headed hairs on teeth of juvenile leaves, is seen in several other species of section Chaenopleura, e.g., M. adenocalyx Urban & E. Ekman, M. ferruginea (Desr.) DC., M. krugii, M. samanensis, M. santanana Judd & Skean (ined.). This condition, therefore, is not used as the basis for segregating the Morne Belance plants (as M. buchii). Long-stalked, gland-headed hairs are also moderately scattered on stems and leaves of these plants. These hairs, however, are also found on stems (and less commonly leaves) of some plants in the remaining populations of M. stenobotrys. In fact, the density of gland-headed hairs is often quite variable, even within a single population. It is note- worthy that in the strongly flanged plants of the Loma La Vieja— El Frio region these hairs are usually lacking. In summary, although certain plants of particular populations are quite different in appearance from some plants of other pop- ulations, when the total pattern of variation is considered, no diagnosable units within Miconia stenobotrys are discernable (Fig- ures | and 2 TAXONOMIC TREATMENT Miconia stenobotrys (L. C. Rich.) Naud., Ann. Sc. Nat., 3 ser., 16: 240. 1851. Melastoma stenobotrys L. C. Rich. in Bonpl. Mélast. 66, t. 30. 1816. Chaenopleura stenobotrys (L. C. Rich.) DC., Prodr. 3: 197. 1828. Miconia stenobotrys (L. C. Rich.) Naud., Ann. Sc. Nat. 3 sér. Ps ries 1851. Type: ““montibus insulae Hispaniolae,” L. C. Richard s.n., p, Miconia buchii Cogn. in Urban, Symb. stil = 448. 1908. Type: HartTI. Dept. de L’Art pay Massif du Nord, Gros-Morne, Morne Belance, 1100 m, 26 Sept. 25, E. L. Ekman H4914 (Hovorype: s!; isorTyPe: us!). Miconia aribonitenss Urban & E. Ekman, Arkiv Bot. 22A(17): 43. 1929. Type: c. Prov. Monte Criste [Dajabon], Cordillera Central, near Las Rosas by Rio tedite 500 m, 6 June 1926, E. L. Ekman H6275 (HoLotyPre: s!; isoTyPes: y!, Ny!, s!, us!). Miconia leptoneura Urban & E. Ekman, Arkiv Bot. 22A(17): 49. 1929. Type: 1993] Judd and Karpook— Miconia 335 HAITI. Dept. de L’Artibonite, Massif du Nord, near Hinche, between Cerca- Carvajal and Bois-Charles, 700 m, 12 May 1926, E. L. Ekman H6075 (HoLorypeE: s!; IsOTYPES: NY!, s!, us!). Miconia azuensis Urban & E. Ekman, Arkiv Bot. 22A(17): 50. 1929. TyPE: DoMI- (CAN REPUBLIC. Prov. Azua, Cordillera Central, near Las Lagunas, 750 m, 13 June 1926, E. L. Ekman H6374 (Hovotype: s!). Shrub or small tree to ca. 4 m tall. Indumentum of multicellular, minute, globular to matted and + irregularly branched hairs, these usually ephemeral and all parts of plant glabrescent, sometimes long-stalked, gland-headed hairs to 1.9 mm long (rarely elongate, stout-stalked, shortly branched hairs). Young twigs rectangular to square in cross-section, 2-5 mm wide, non-ridged with a nodal line, or with 4 flanges .2-2.3 mm broad, 2 extending below each point of petiole attachment and joining to form a very obscure to conspicuous U-shaped flange at the adjacent lower node, these sometimes restricted to inflorescence axis, glabrous, or with very sparse, minute, globular to matted and + irregularly branched hairs, these ephemeral, sometimes also with very sparse to sparse, long-stalked, gland-headed hairs, especially on portion of inter- node just above node (rarely with sparse, elongate, stout-stalked, shortly branched hairs); internodes .5-4.5(-9) cm long. Leaves with petiole 2-24 mm long, the indumentum sparse, minute, globular to matted and branched hairs, quickly glabrescent, oc- casionally with very sparse, long-stalked, gland-headed hairs (or rarely elongate, stout-stalked, shortly branched hairs); blade ovate to oblong (elliptic), 2.5-15.7