Dodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by ALBION REED HODGDON, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN ROBERT CRICHTON FOSTER ROLLA MILTON TRYON RADCLIFFE BARNES PIKE | LORIN IVES NEVLING, JR. ; | Associate Editors VOLUME 71 1969 Che Nem England Botanical Club, Jue. Botanical Museum, Oxford St., Cambridge 38, Mass. "F V "AeCncens ND Dovova JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by ALBION REED HODGDON, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN ROBERT CRICHTON FOSTER Associate Editors ROLLA MILTON TRYON RADCLIFFE BARNES PIKE LORIN IVES NEVLING, JR. J Vol. 71 January-March, 1969 No. 785 CONTENTS: The Phytogeography of Subalpine Black Spruce in New FRINGE. James A. Toeri .....................uu l 5 s eee i The Alsophila swartziana complex (Cyatheaceae). SINON ME L LU ri exis a bred sO R j 7 The Flora of Monomoy Island, Massachusetts. Zum T. Moul u u S S c LE 18 A Floristic and Ecological Study of Pitcher Plant Bogs in South Mississippi. L. N. Eleuterius amd S. B. Jones, Jr. 29 The Bromeliaceae of Bolivia. Lyman B. Smith ................... 35 Taxonomy of Perityle Section Pappothrix (Compositae- Peritylanae). A. Michael Powell ........................ 58 Disjunction and Endemism in Croton alabamensis. Joe A. Farmer and Joab L. Thomas .................................... 94 (Continued on Inside Cover) The Nef Hngland Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge, Mass, 02188 CONTENTS: — continued Book Review: A Definitive New Book on the Flora of Northwestern North America. Reed C. Rollins ................ 104 Book Review: A New Wildflower Guide. Frederic L. Steele ccccccccccccccsssssssssscsccccesecsscsesccsceceeeccesecessessreeeeees 107 Contributions to the Marine Algae of Newfoundland. Arthur C. Mathieson, Clinton J. Dawes and Harold J. HUMM .... eee eee eene nennen enne 110 Edith Scamman: An Appreciation. Nicholas Polunin ........ 160 Added Note on Edith Scamman. Albion R. Hodgdon ........ 165 Notes on the Flora of Coós County, New Hampshire. Stuart K. Harris eese eene eee entree eene 166 Additions to the Kansas Flora. Rudy G. Koch .................... 175 Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 71 January-March No. 785 THE PHYTOGEOGRAPHY OF SUBALPINE BLACK SPRUCE IN NEW ENGLAND JAMES A. TEERI! Prostrate plants of black spruce (Picea mariana (Mill.) BSP.) occur above the treeline on many of the higher mountains of northern New England. The dwarf spruces appear to belong to a subalpine race adapted to the cool, windy upper-montane environment. The present study re- sulted in the re-elevation of the short-needled prostrate plants to varietal status. Possible origins and phytogeo- graphic relations of the variety are discussed with respect to recent interpretations of late-glacial events in north- eastern North America. Picea mariana (Mill.) BSP. var. semiprostrata (Peck) comb. nov. Based on Picea brevifolia var. semiprostrata Peck, Rep. Bot. N.Y. State Mus. 283. 1897. Plants of var. semiprostrata lack a single upright apex, and may or may not possess a single horizontal main stem (fig. 1a, b). The needles average about 4 mm in length, with a range of 2.2 to 5.5 millimeters. The needles are thick, glaucous, and slightly adaxially curved. Glandular- tipped trichomes profusely cover the new growth during the first year. These plants have a prostrate growth form usually less than 4 dm in height. The average amount of annual apical elongation is about 3 centimeters. Plants of "Present address: Department of Botany, Duke University, Durham, North Carolina 27706. 2 Rhodora [Vol. 71 var. semiprostrata occur in the subalpine regions of the higher mountains of northeastern North America. Peck (1897) described Picea brevifolia var. semiprostrata as a “. . . small, half-prostrate, shrub-like spruce . . . on the exposed summits of the high peaks of the Adirondack mountains." He based the variety on short needles and glaucous foliage. Rehder (1907) later described Picea mariana var. brevifolia as a small tree usually found in "... mountain bogs...” of northeastern North America. The var. brevifolia was characterized by needles 4 to 12 mm in length. Rehder recognized that the variety he described did not include the most prostrate, short-needled black spruce of the mountains. He mentioned: “The low prostrate form of the exposed tops of high mountains de- scribed by Peck as P. brevifolia var. semiprostrata, will probably also be found in New England.” Blake (1913) recombined Picea brevifolia var. semiprostrata Peck to P. mariana f. semiprostrata (Peck) Blake. The plant thus described had “. . . a creeping alpine form . . ." and needles 3 to 6 mm in length. The present information in- dicates that Blake's concept of the f. semiprostrata is the same as P. mariana var. semiprostrata. Black spruce becomes dominant in the flora at treeline in many of the higher mountains of New England, The spruce plants below the treeline and at treeline are of the partially upright growth form (fig. 1c) or the normal tree form (fig. 1d). These two forms are about 214 to 3 meters in height, and have needles with an average length of 7.5 to 8.5 millimeters. The range in needle length is from 5.0 to about 9.0 mm in plants of these two forms. Above the treeline, black spruce occurs in growth forms varying from partially upright to prostrate (fig. 1b). The Fig. 1l. The growth forms of black spruce in the mountains of New England. a. Vegetative derivative of var. semiprostrata. b. A typical plant of var. semiprostrata above treeline on Mt. Washington. c. Partially-upright form of long-needled black spruce from below treeline. d. An upright tree of black spruce from several hundred feet below treeline. 1969] Black Spruce — Teeri 4 Rhodora [Vol. 71 needle length of a given plant can be correlated with its growth form. Each plant has a distinct range of needle size, The most prostrate plants have the shortest needle length, about 4 mm; it is these plants which constitute the var. semiprostrata. The cline of needle length does not appear to be related directly to altitude above treeline nor exposure of the site. Frequently, long-needle plants (aver- age needle length ca. 8 mm) occur near the upper altitu- dinal limit of the species. In several sites, long needled, partially upright plants grow immediately beside quite prostrate plants which have an average needle length less than half that of the larger plants. The colonies appear to be of approximately the same age. The short-needle plants possess considerable vigor and show little evidence of killing back or needle damage. Longer needled plants often show needle damage and winter injury. The complex series of plants of intermediate morphology and winter resistance suggests hybridization as the most likely ex- planation for the cline of types. The genetic extremes of this cline must be the long-needled upright trees and the short-needled prostrate plants. The great number of indi- viduals, each slightly distinct from the others in growth form and needle length suggests a polygenic mechanism of inheritance. Sexual reproduction of black spruce is not frequent in the present climatic regime in the subalpine regions of New England (Griggs 1942). Cones appear sporadically and in most years few or no cones are produced above tree- line. During the summer of 1967, some scattered black spruce plants in the subalpine zone produced a few cones, most of which were male cones. The same plants produced no cones during the summer of 1968. Seedlings are scarce in the subalpine region. However, it is apparent from the present wide distribution of black spruce types that in the past there has been a good breeding population of subalpine black spruce. Current geological evidence indicates that the deglacia- tion of New England took place in an erratic fluctuating 1969] Black Spruce — Teeri 5 manner (Borns 1963). Borns (1963) has suggested that in late-glacial time, the White Mountains of New Hampshire and adjacent Maine were the center for a late-persisting local ice cap. Late-persisting local ice has been inferred for Mt. Katahdin, Maine (Tarr 1899). If a local ice cap did persist in the White Mountains, then the repopulation of the mountains by many taxa of the present subalpine flora could have come from a variety of directions. As habitats became available, suitable plants probably followed the retreating ice closely. Present evidence (Teeri unpub.) suggests that the subalpine black spruce (var. semipro- strata) is suited in its ecologic and physiologic require- ments to such late-glacial habitats. The var. semiprostrata, with short needles, minimal biomass, and a prostrate growth form, is particularly suited to conditions of climate unsuitable for many other woody taxa. It may be that the evolution of var. semiprostrata occurred in the postglacial mountain habitats of its present range. In this case, the taxon may still be in the process of development. Alternatively, the variety may represent the remnants of a population adapted to the environmental conditions at the periphery of the retreating Wisconsin glacier. The late-glacial conditions may well have been similar, near the ice margins, to present environmental conditions in the higher New England mountains. Thus, in the second case, one would not have to postulate the separate evolution of the same taxon in the rather dis- junct sites of its present range. Instead the evolution of var. semiprostrata could have occurred in a more or less continuous manner, perhaps at the margin of the glacia- tion. The present range of the variety (i.e. the subalpine regions of northeastern North America) would then be a function of relatively continuous migration and subsequent isolation due to increasingly unfavorable postglacial con- ditions in the lowlands. As the less winter-resistant red spruce later occupied the mountains, the present mixture of hybrids between black and red spruce (Morgenstern and Farrar 1964) became possible in the middle montane 6 Rhodora [Vol. 71 region. Observations of the past two years have shown an intergradation of many of the characters of the two species just below the treeline on many New England mountains. The striking difference in needle length and growth form between plants of var. semiprostrata and adjacent more typical black spruce, suggests strongly a genetic distinction of the prostrate population. Plants of var. semiprostrata are morphologically distinct, possess con- siderable vigor, and occupy a distinct range. Thus, the category of varietas seems to best fit this taxon. ACKNOWLEDGEMENTS I thank the following persons for assistance and helpful discussion: A. R. Hodgdon, R. B. Pike, T. S. Teeri. Daniel Sanders prepared the illustrations in fig. 1. DEPARTMENT OF BOTANY UNIVERSITY OF NEW HAMPSHIRE DURHAM 03824 LITERATURE BLAKE, Š. F. 1913. Six weeks’ botanizing in Vermont. II. Addi- tional notes on plants near Burlington. Rhodora 13: 200-201. Borns, H. W. 1963. Preliminary report on the age and distribu- tion of the late Pleistocene ice in north central Maine. Am. Jour. Sci. 261: 738-740, GRIGGS, R. F. 1942. Indications as to climatic changes from the timberline of Mount Washington. Science 95: 515-519. MORGENSTERN, E. K., and J. L. FARRAR. 1964. Introgressive hy- bridization in red spruce and black spruce. Faculty of Forestry, Technical Report No. 4. Univ. of Toronto. PECK, C. H. 1897. Report of the state botanist, New York State Museum. Pp. 282-283. REHDER, A. 1907. Some new or little known forms of New England trees. Rhodora 9: 109-117. TARR, R. S. 1899. Glaciation of Mount Ktaadn, Maine. Geol. Soc. America, Bull. 11: 433-448, THE ALSOPHILA SWARTZIANA COMPLEX (CYATHEACEAE).* RAMÓN RIBA The genus Alsophila R. Brown comprises about 350 spe- cies, widely distributed in the tropical and subtropicals re- gions of the world, with about 150 species in America, This genus can be distinguished from the others of the family, in a traditional sense, by the stem and petioles with scales, and the sori without an indusium or with only a minute scale-ike indusium. In a more restricted sense it would include only such species of the traditional genus as have flabelloid petiole scales. The group of Alsophila, swartziana falls under Alsophila in either classification. I agree with Holttum, (1957), who considers that the morphology of the petiole scales is a. character to define natural groups in this family. Holttum recognized two basic types of scales: setiferous scales, with marginal denticula- tions bicolorous or concolorous with the entire body of the scale, and flabelloid scales, with a margin formed by very fragile cells, very different from the cells of the body of the scale. In the American species of the genera of tree-ferns, however, the structure presented by the petiole scales is more complicated and more diverse than a setiferous type and a flabelloid type. Maxon (1922) established an informal group of species related to Alsophila armata. (— A. swartziana) based prin- cipally on the complete absence of indusium, the degree and type of dissection of the leaf and the pubescence. The South American members of this group were not treated. In the *The present paper is a synopsis of “Revisión monográfica del complejo Alsophila swartziana Martius (Cyatheaceae)," to be pub- lished in An. Inst. Biol. Univ. Nac. Autón. México, 37. Ser. Bot. The research was made possible by a John Simon Guggenheim Fellowship, at the Gray Herbarium, Harvard University, during 1966. The sup- port of the Guggenheim Foundation is gratefully acknowledged. Dr. Rolla Tryon provided helpful guidance during my studies, and field work, in part, was supported by his grant (4184) from the National Science Foundation. 8 Rhodora [Vol. 71 present paper other characteristics of the petiole scales have been used to define the group, in addition to those employed by Maxon, and five species must be excluded from the group: A. microdonta, A. myosuroides, A. notabilis, A. mex- icana and A. acutidens. It includes the American species of Alsophila with subflabelloid scales in which the border is more or less differentiated and the margin of the petiole scales of mature leaves or croziers has dark denticulations (fig. 1). The group of Alsophila swartziana, then, includes three species of the West Indies, five of Mexico and Central America, four South American species, and a single one of both Central and South America. KEY TO THE SPECIES a. Veins of the fertile segments simple, or once forked near the border of the segment; sori borne below the branching of the vein; Jamaica. ....................... ee eee eene ne nnne sensns sss rns nani niii rnn nnn 1. A. estelae a. Veins of the fertile segments medially forked at least once; sori borne at the branching of the vein. b. b. Petiole abundantly pubescent, the trichomes sometimes deciduous and then leaving a hard elevated base and a scabrous surface (smooth in A. swartziana of the Greater Antilles). c. c. Paraphyses turgid, persistent, dark red and shorter than the sporangia; Cocos Island. ..................... seen 2. A. nesiotica c. Paraphyses slender, obtuse or acute at the tip, deciduous or persistent, longer than the sporangia. d. d. Upper surface of the segments with abundant trichomes on the eostules, veins and between the veins. e. e. Margin of petiole scales dark denticulate throughout; apex of pinnules long-attenuate, often curved; Costa Rica to Ecuador. ........................... nemen 3. A. trichiata e. Margin of petiole scales smooth or with a few dark teeth apically: apex of pinnules acuminate, nearly straight; Venezuela, Colombia to Bolivia. ............ 4. A. tryonorum d. Upper surface of the segments glabrous or with trichomes only on the costules and veins. f. f. Upper surface of veins with trichomes (when deciduous the bases evident); mature pinnules with few or no bul- late scales along the costa beneath; Mexico to Honduras. MEME 5. A. scabriuscula g. Sinuses of segments quadrangular in the basal portion of the pinnules; Mexico, Guatemala, Honduras. ............ 5a. var. scabriuscula Mhsethsssusstósasssssseosssssssssseseonceseeesenesoeeeovaseee 1969] Alsophila — Riba 9 g. Sinuses of segments acute to rounded in the basal portion of the pinnules; Guatemala. .............. mm ied MM UTE MAU D UE E bb. var. guatemalensis f. Upper surface of veins glabrous; pinnules beneath with abundant bullate scales along the costa; a flattish scale nearly always present between the costule and the fork of the vein; Jamaica, Haiti and Dominican Republic. 6. A. swartziana Pesocssssosssssssosesesososstsssesssoovoesosvoossosscscescssaseneeoooe b. Petioles glabrous or sparsely pubescent; the trichomes some- times deciduous and then leaving a prominent scar and a nearly smooth surface. h. h. Pubescence of lower surface of the costa and costules only of abundant, flexuous, catenate reddish ‘trichomes; eastern Cuba ec n n 5 S A n ata MEIN Cree sss 7. A. strigillosa h. Pubescence of lower surface of the costa and costules princi- pally of abundant, stiff, terete trichomes. i. i. Petiole scales with a white body, or with a dark area only at the base; southeastern Brazil. j. j. Rachis glabrous, prickly with sparse slender spines, slightly hirsute at the apex; segments lanceolate, acute; fertile and sterile veins once forked. ........ 8. A. hirsuta j. Rachis pubescent or subglabrous, unarmed; segments linear, obtuse or acute; fertile and sterile veins with one or two pinnate pairs of veinlets. ......... e 9. A. rufa i. Petiole scales, or most of them, with a narrow or broad, dark central stripe and a white border. k. k. Margin of the petiole scales smooth or with a few dark teeth apically or with a few marginal cells slightly darker than the others; veins impressed above; Colombia to Bolivia. ........................................................ 10. A. conjugata k. Margin of the petiole scales dark denticulate throughout; veins not impressed above. I. l. Rachis without spines, sometimes slightly muricate; lower surface of costules with flexuous trichomes; upper surface of costules and veins with thin and flexuous trichomes; Costa Rica and Panama. ....... Ti E 1 11. A. stipularis l. Rachis with scattered, small, sharp spines; lower sur- face of costules with stiff trichomes; upper surface of segments glabrous or glabrescent. m. m. Petiole scales with continuous dark teeth along the margin, with a broad dark central stripe; Guate- mala. |... eorr eee ero tst eso rt aoo oS E N 12. A. pansamalana m. Petiole scales with whitish cells between the dark teeth along the margin, with a narrow dark central 10 Rhodora [Vol. 71 stripe or sometimes with a dark area at the base; Mexico. ................ eeeeeeeeerrenes 13. A. bicrenata TAXONOMIC AND GEOGRAPHIC NOTES ON THE SPECIES. 1. Alsophila estelae Riba, Rhodora 69: 67, f. 5-7. 1967. Type: Mount Horeb, Blue Mountains, Jamaica, Riba 214 (MEXU) ; Isotype: (GH) Alsophila estelae is endemic to Jamaica and is one of the most graceful tree-ferns of that island. At first sight it can be easily confused with A. swartziana, a species that grows in the same places. However, the slender trunk (3-4 m tall and ca. 4-6 cm thick) and the crown of small leaves make it a distinctive species in the field. Alsophila estelae has been collected only three times in Portland and St. Thomas Parishes. 2. Alsophila nesiotica Maxon, Contrib. U. S. Nat. Herb 24: 43. 1922. Type: Cocos Island, Costa Rica, Pittier 16229 (us). The affinity of Alsophila nesiotica with the other species in this group is uncertain. The petiole scales do not have the margin dark denticulate, but the border is well differ- entiated, its cells are smaller than those of the center and are oriented in a fan-like manner, and a few marginal cells at the apex are slightly darker than the others, without any differentiation in teeth. Material of the croziers has not been available, but it is anticipated that they will have dark denticulate scales, as in A. conjugata and A. tryonorum. This species grows only on Cocos Island, Costa Rica, where it has been collected four times. 3. Alsophila trichiata Maxon, Contrib. U. S. Nat. Herb. 24: 44, t. 15. 1922. Type: Cana, Panama, R. S. Williams 928 (US). Alsophila trichiata (fig. 3) is a species which grows at low elevations, from almost sea level in Costa Rica and Pan- ama and northern Venezuela, up to near 1,000 m of altitude in Ecuador. This species is related to A. tryonorum, but that species grows at higher elevations (up to 2,800 m) and 1969] Alsophila — Riba 11 Cn Fig. 1. Border of petiole scale of Alsophila trichiata, > 25. Fig. 2. Winged trichomes from lower surface of costa of Alsophila pansamalana, X 40. Fig. 3. Distribution of Alsophila trichiata. Fig. 4. Distribution of Alsophila tryonorum. Fig. 5. Distribution of Alsophila conjugata. 12 Rhodora [Vol. 71 is easily distinguished by characters of the pubescence and the scales given in the key. 4. Alsophila tryonorum Riba, Rhodora 69: 66, f. 1-4. 1967. Type: Cuesta de Fusagasugá, Dept. Cundinamarca, Colombia, Cuatrecasas 8036 (US). Alsophila tryonorum shows an evident relation with A. trichiata because of the similarity of the distribution of the pubescence, However, in A. trichiata the rigid trichomes are longer and the catenate trichomes of the petiole are not as abundant as in A. tryonorum; the petiole scales in A. trichiata have contiguous dark denticulations along the mar- gin, while in A. tryonorum the margin has very few dark denticulations or none, except in crozier scales where the dark denticulations are always present. Another relation is with A. conjugata based on the similarity of the petiole scales together with the abundant winged trichomes (cf. fig. 2) and bullate scales on the lower surface of the seg- ments. However, the petiole in A. conjugata is glabrous, while in A. tryonorum it is strongly pubescent with rigid and catenate trichomes. Alsophila tryonorum (fig. 4) grows in the high mountain forests of northern Venezuela and in the Andean region of Colombia and northern Ecuador. 5. Alsophila scabriuscula Maxon, Proc. Biol. Soc. Wash. 32: 125. 1919. Type: Cubilquitz, Alta Verapaz, Guatemala, Tiirckheim (J. D, Smith Exsice. 7806) (US); Isotype (NY). This species grows in Mexico, Guatemala and Honduras, in the shade of humid tropical forests. Variety guatemalen- sis Riba, Rhodora 69: 68, f. 8, 1967 [Type: Sierra de los Cuchumatanes, Dept. Huehuetenango, Guatemala, Steyer- mark 49417 (F); Isotype: (US)] is confined to central Guatemala. The differences between the two varieties are in the sinuses and the size of the leaf. In var. guatemalensis the sinuses between the segments are acute to narrowly quad- rangular, rather than definitely quadrangular, and the leaf 1969] Alsophila — Riba 15 and its parts are smaller than in var. scabriuscula. Although the two varieties occur in the same general region they are distinctive in these characters. 6. Alsophila swartziana Martius, Icon. Pl. Crypt. Brasil. 73, t. 49. 1834. Type: based on Polypodium armatum Swartz, hence hav- ing the same type. Polypodium armatum Swartz, Prodr. Veg. Ind. Occ. 134. 1788. Type: Jamaica, Swartz (S-PA), photograph and frag- ment (US). Alsophila armata (Swartz) Presl, Tent. Pterid. 62. 18536, not A. armata Martius, 1854. Alsophila swartziana is confined to Jamaica, Haiti and Dominican Republie; specimens identified as this species, from Venezuela and Brazil prove to be A. tryonorum in Venezuela and A. hirsuta in Brazil. Alsophila swartziana is closely related to A. bicrenata, but that species has the upper surface of the segments glabrous, and the petiole smooth ; it also has a close relation to A. scabriuscula var. guate- malensis, but that variety does not have the characteristic scales between the receptacle of the basal sori and the costule. 7. Alsophila strigillosa Maxon, Contrib. U. S. Nat. Herb. Ba. ot. t £2, 1922, Type: Nima-nima, eastern Cuba, C. Wright 1062 (YU), fragment (US); Isotype: (GH). This species has been collected only twice in eastern Cuba. I have been able to examine several sheets of the type collection; besides this, Maxon cites one sheet, Wright 891 in the Eaton Herbarium (YU), as containing one pinna of Cyathea arborea and one pinna of A. strigillosa. Alsophila strigillosa is one of the best defined species of the Alsophila swartziana group, as brought out by the characters given in the key. It seems to be related to A. swartziana by the similarity of the scale border and margin, and the pubescence of the upper surface of the segments, with just a few stiff hairs on the costule. 14 Rhodora [Vol. 71 8. Alsophila hirsuta (Presl) Kunze, Linnaea 9: 98. 1834. Cyathea hirsuta Presl, Delie. Prag. 190. 1822. Type: Corcovado, Rio de Janeiro, Brazil, Pohl (PR or PRC), frag- ment (NY). Alsophila hirta Kaulfuss, Enum. Fil. 249. 1824, ex char. This species is known only from Brazil, in some areas of the states of Minas Geraes, Rio de Janeiro, Guanabara, and Sao Paulo. It is closely related to A. rufa, of the same region, differing from the latter by the characters given in the key. 9. Alsophila rufa Fée, Crypt. Vasc. Brasil. 1: 165, t. 39, f. 1. 1869. Type: Rio de Janeiro, Brazil, Glaziou 2291 (P). Alsophila rufa occurs in the same region of Brazil as A. hirsuta. The two have sometimes been confused, even con- sidered as the same species, but they are clearly distinct. 10. Alsophila conjugata Spruce ex Hooker & Baker, Syn. Fil. 37, 1866. Type: Chimborazo, Ecuador, Spruce 5745 (K), fragment (US). This species (fig. 5) is exclusively Andean in distribution, and is the most widely distributed species in the group; it grows from northern Colombia to Bolivia, always in damp forests, ravines, creeks, or along streams, with an alti- tudinal distribution which varies from 650 m in Peru to 2,800 m in Colombia. 11. Alsophila stipularis Christ, Bull. Herb. Boiss. II, 4: 958. 1904, Type: Costa Rica, Wercklé (P); Isotype: (us). Alsophila stipularis grows in the higher mountain regions of Costa Rica and Panama between 1000 and 2000 m. This species seem to be related to A. conjugata by the glabres- cence of the principal axes of the leaf (petiole, rachis and secondary rachises). The petiole scales, however, are differ- ent, having a dark denticulate margin with contiguous dark teeth while in A. conjugata the dark teeth are nearly lack- ing. 1969] Alsophila — Riba 15 12. Alsophila pansamalana Maxon, Contrib. U. S. Nat. Herb. 24: 40. 1922. Type: Pansamalá, Alta Verapaz, Guatemala, Tiirckheim (J. D. Smith Exsice. 1008) (US). This species is confined to Guatemala, between 1000 and 2000 m; there is no indication on the collections I have seen concerning the habitat. Its closest relative apparently is A. scabriuscula from which it is readily distinguished by its freely armed, glabrous (rather than the scabrous) rachis. The winged trichomes, developed in several species, are il- lustrated in fig. 2 from the lower surface of a segment. 13. Alsophila bicrenata (Liebmann) Fournier, Mex. Pl. Crypt. 134. 1872. Cyathea bicrenata Liebmann, Dansk. Vid. Selsk. Skrift. V, 1: 289. 1849. Type: between Trapiche de la Concepcion and Totontepec, Oaxaca, Mexico, Liebmann (C), photograph (US) ; Isotype: (US). This species is confined to Mexico, from Veracruz to Guer- rero, south to Chiapas. It grows in the shade of humid tropical forest, between 1000 and 2000 m. Alsophila bicrenata is closely related to A. scabriuscula var. scabriuscula; the petiole scales in both taxa have the same general structure and the habit is very similar. Al- sophila bicrenata has a glabrous or slightly pubescent petiole, the lower surface of the veins is without long tri- chomes, and the sinuses between segments are acute or rounded, it has the persistent bullate scales at the base of the costules beneath, these sometimes covering the basal half of the costule. DISCUSSION. The differences between the species included in this group are largely in microscopic characteristics, because their gross morphology does not seem to offer characters of spe- cific value. In the whole family, the habit and general morphology are very similar, with a few exceptions. In the 16 Rhodora [Vol. 71 species of the Alsophila, swartziana group the principal dif- ferences are in the type of pubescence and place of attach- ment of the trichomes. In one species, Alsophila nesiotica, the paraphyses are distinctive; in another, Alsophila estelae, the venation is distinctive; and in two South American spe- cies, Alsophila conjugata and A. tryonorum, the petiole scales lack the usual dark denticulations on the margin. The concept of species adopted in this paper is derived from closely related sympatric species in four areas: Al- sophila swartziana and A. estelae in the Greater Antilles, A. scabriuscula var. scabriuscula and A. bicrenata in south- ern Mexico, A. tryonorum and A. conjugata in the Andean region, and A. hirsuta and A. rufa in Brazil. There is no morphological continuity between these species, in each re- gion, and I consider each one of the pair to represent a good species. To define the distributional center of the group, it is necessary to look for the more primitive species, or at least the less evolved ones. The principal characteristics which are considered as primitive in this group are: a pubescent petiole, a less differentiated border on the petiole scales, and a petiole scale margin without dark denticulations. Accordingly, the most primitive species are the Andean Alsophila conjugata and A. tryonorum, the latter present- ing the most primitive characteristics. Alsophila trichiata, Central America to Ecuador, is at about the same evolu- tionary level. I suppose, in a purely speculative way, that the distribu- tional center of the species is in the Andean region, an area with many other species of tree-ferns. From this region the migration evidently proceeded north through Central Amer- ica to Mexico and east to the Antilles. The Antillean species probably were derived from the principal stem which gave rise to the Mexican species. The sympatric groups found in Brazil probably derived from the species of the Andean re- gion, which now has its austral limit in the Bolivian moun- tains. In regard to Alsophila nesiotica, the endemic species 1969] Alsophila — Riba 17 on Cocos Island in the Pacific Ocean, it could be derived from the group represented by Alsophila trichiata on the western slope of the Andean region. INSTITUTO DE BIOLOGIA UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO. REFERENCES, HoLTTUM, R. E. 1957. The scales of the Cyatheaceae. Kew Bull. 41-45. Maxon, W. R. 1922. The North American species of Alsophila grouped with A. armata. Contrib. U. S. Nat. Herb. 24: 33-46. FLORA OF MONOMOY ISLAND, MASSACHUSETTS EDWIN T. MOUL! The flora of Monomoy Island, Massachusetts, has been investigated during the summers of 1965, 1966, and 1967 in July and August as part of the Systematics-Ecology Pro- gram of the Marine Laboratory at Woods Hole. This is part of the survey of the plants of the marine fringe being made for the Cape Cod area under that program. Monomoy Island is a large mass of sand, one-quarter to one mile wide, built up by shore-drifting from the northern shores of the Cape. It extends into Nantucket Sound for 8 miles, south of Chatham. At present it is under the ad- ministration of the Bureau of Sport Fisheries and Wildlife and is reserved as a Wild Life Refuge. The island is built up of a number of curving sand ridges which have become connected by wind-formed dunes. Several natural fresh water ponds and marshes have developed in the hollow of some of the dunes. A number of artificial ponds have been dug by the Bureau of Wildlife to provide food and resting places for migratory wild fowl. The survey of the island was made by a group of associ- ates from the Systematics-Ecology Program from the labor- atory, including Mr. Peter Schwamb, curator of the Gray Museum. We were assisted by Mr. Dana Eldridge, a native of Orleans, Massachusetts. Three areas were covered as completely as possible by doubling back and forth across the island. Herbarium specimens were collected and deposited in the Herbarium of the Gray Museum at the Marine Labor- atory, with duplicates placed in the Chrysler Herbarium at Rutgers, The State University, New Jersey. The areas surveyed were as follows: Area #1 at the southern tip of the island, known as Monomoy Point, around Big Station Pond and Powder Hole. This is an area of dunes, brackish ponds, and fresh water marshes; Area *Systematics-Ecology Program Contribution #155, supported in part by a grant from the Ford Foundation to Systematics-Ecology Program, Marine Biological Laboratory. 18 1969] Monomoy Island — Moul 19 #2 was north of this around the Old Monomoy Light Tower and the abandoned Coast Guard Station; Area #3 was at Inward Point and around Hospital Pond. North of Inward Point the island is completely covered with typical dune vegetation. The flora of islands change over the years. Examination of the records of the flora of Penikese Island (Jordan 1874, Lewis 1924, Fogg 1930, Moul 1948, Moul 1961) will show some drastic changes over the years. So the list of species presented here should be considered the summer flora of Monomoy Island at the present time. A number of plants reported for Monomoy Island by S. G. Cross and C. E. Cross in 1954 (unpublished list provided by Mrs. Marcia G. Norman) were not found. Some of the plants included in this 1954 list are possible misidentifications, since a number of them do not occur in the range of Gray’s Manual. Arcto- staphylos uva-ursi, Kalmia angustifolia and Chrysanthe- mum leucanthemum listed as present by Cross and Cross cannot be conspicuous members of the flora, since they were not found even after extensive search in 1967. Moul noted (1961) that Chrysanthemum leucanthemum, former- ly common on Penikese Island has become extremely rare. The dominant vegetation type of the island is the dune grass community (Ammophila breviligulata) on the pri- mary dunes. With the grass are the usual associates Lathyrus japonicus, Solidago sempervirens, and Xanthium echinatum. This community occupies the whole northern end of the island and the east and west shores. The second- ary dunes have dune grass on them, but are principally covered with a Hudsonia-Cladonia association. Artemisia caudata, Chrysopsis falcata, and Lechea maritima are com- monly associated with Hudsonia. These secondarv dune communities are very extensive in the Inward Point area and the area around the Monomoy Light Tower. The inner dunes at Monomoy Point (area #1) are covered with a growth of Ammophila and Phragmites communis, Thickets typical of the main Cape, consisting of Myrica pensylvanica, Rosa rugosa, Prunus maritima, and Rhus radicans, occur in the protected hollows and slopes of the inner dunes. 20 Rhodora [Vol. 71 Beach vegetation on the Nantucket Sound or west coast of the island is typical of the Cape beaches. Cakile edentula, Xanthium echinatum, Lathyrus japonicus, and Atriplex arenaria are the principal plants of the association. This community is missing on the east or ocean shore. Extensive cat-tail (Typha) marshes occur north of Big Station Pond at Monomoy Point (area #1). Typha angusti- folia forms a dense phalanx, beneath which grows Galium palustre, Ptilimnium capillaceum, and Pluchea purpurascens in dense stands. The waterways, probably made by musk- rats, are filled with Ludwigia palustris. The wet depressions at Monomoy Point and the Light- house area have Festuca rubra and Spartina patens in them. The other species associated with the grasses are Polygonum spp., Rumex crispus, Lycopus americanus, Scutellaria epilo- biifolia, Teucrium canadense, and Iris versicolor. This type of vegetation is also found along the shores of the many artificial ponds. Cat-tails and Onoclea sensibilis also occur along the shores of some of these ponds. The sand removed in digging the ponds has been piled into artificial dunes, which are now covered with Rhus radicans, Myrica pensyl- vanica and Prunus maritima. In the center of the island at area #2 are a few small thickets made up of Salix discolor, Amelanchier laevis, Vac- cinium corymbosum, and Prunus serotina, with dense mar- gins of Myrica pensylvanica. A few Pinus rigida grow in the protection of these thickets. The shrubs and trees here show wind pruning due to salt spray. A number of bogs occur in depressions of the dunes in area #2. Vaccinium macrocarpon and Sphagnum palustre form the dense ground cover. Viola lanceolata, Lycopodium inundatum, Drosera rotundifolia, and Pogonia ophioglos- soides grow in these bogs. Polytrichum commune occurs around the drier margins of the bogs. The weedy species such as Daucus carota, Phytolacca americana, and Achillea millefolium are confined principally to the area of the lighthouse and abandoned Coast Guard station, showing the influence of man. The most mature vegetation occurs at Inward Point (area 1969] Monomoy Island — Moul 21 #3). This part of the island has been used for years by summer visitors and a number of old cottages are located here. What probably was formerly a fresh water pond, and still may be flooded during wet summers, is dotted with hummocks of Juncus effusus. Ptilimnium capillaceum, Hypericum virginicum, Osmunda cinnamomea, and Iris versicolor form a dense cover in the wet bottom. A few patches of open water persist. Dicranum scoparium grows in the drier portions of the marsh along the border. The thicket surrounding this marsh is quite dense and contains most of the woody species reported for the island. Vaccini- um corymbosum is very common, and in the summer of 1965 bore a prolific crop of berries, but had no fruit in 1966 or 1967. Pinus sylvestris has apparently been planted here in the thickets, and the trees have produced cones regularly. The largest typical New England type salt marsh occurs north of Inward Point. It is dissected by creeks, and there are a number of dry “pannes” in which scattered Salicornia spp. and Plantago oliganthos grow. 'This marsh extends almost completely aeross the island and merges with the dune grass of the eastern dunes. Nomenclature used in the list of higher plants is that of Fernald (1950). Nomenclature for the mosses is that of Grout (1940). The assistance of Dr. David Fairbrothers and Mary Hough in identification of a number of taxa is acknowledged. LICHENS Cladonia sylvatica (L.) Hoffm. Bare sandy areas on dunes, between vegetation. Areas £1, £2. Cladonia sp. Between Hudsonia plants on dunes near center of island. Area £3. BRYOPHYTES Dicranum scoparium Hedw. Path around fresh water marsh in thickets. Area #3. Polytrichum commune L. var. perigoniale (Mx.) Bry. Eur. Carpeting areas around edge of sphagnum-cranberry bogs. Area £2. Sphagnum palustre L. In the bogs. Area #2. 22 Rhodora [Vol. 71 VASCULAR PLANTS Lycopodium inundatum L. Fairly common in sphagnum bogs in the hollows of dunes in Area #2. Osmunda cinnamomea L. Fresh water marsh in Area #3. Inward Point. Dryopteris thelypteris (L.) Gray. var. pubescens (Lawson) Nakai. Commonest fern, Beneath cat-tails in Area £1. Also around artificial ponds and fresh water marshes in areas £2 and £3. Onoclea sensibilis L. Fairly common. In cat-tail marsh, around artificial ponds and bogs in Areas #1 and £2. Juniperus virginiana L. Small trees to 2 feet in height around abandoned lighthouse. Area #2. Pinus rigida Mill. A few low scrubby trees in the center of the island, Area #2. Frequently protected by willow and shad-bush thickets. Pinus sylvestris L. Planted around a fresh water marsh. In thickets. Cones present. Inward Point. Area #3. Typha angustifolia L. Very common in low marshy area between dunes. Monomoy Point. Area #1. Ruppia maritima L. In brackish water ponds near Mono- moy Point. Covered with blue-green algae. Area #1. Triglochin maritima L. In the Spartina patens zone of the salt marsh. Inward Point. Area #3. Agrostis alba L. var. palustris (Huds.) Pers. Drier parts of the hollows between dunes and ponds. Area #1, and at the edge of salt marsh and dunes at Inward Point, area #3. Agrostis scabra Willd. Shade of willow thicket. Center of island. Area #2. Ammophila breviligulata Fern. Common grass of all dune areas. Bromus tectorum L. Bare sandy spots in center of island near cottages. Inward Point. Area #2, Elymus virginicus L. var. halophilus (Bickn.) Wieg. Very common at edge of salt marsh and dunes. Upper beach next to dunes. Inward Point. Area #2. Festuca rubra L. Wet swale, area #1. Edge of cranberry bog, area £2. 1969] Monomoy Island — Moul 23 Panicum virgatum L. Border of fresh water marsh and dunes, area #1. Dunes and marsh border, east side of island, area #3. Phragmites communis Trin. Very common around brackish ponds, on dunes, and in swales. Monomoy Point, area #1. Spartina alterniflora Loisel. Brackish pond, area #1. Salt marsh in area #3. Spartina patens (Ait.) Muhl. Drier part of swale, area #1. Salt marsh in area #3. Carex comosa Boott. Wet area in center of island, area #2. Carex longii Mackenzie. Swale near brackish ponds. Fair- ly common at other places in the dunes, area #1. Cyperus grayii Torr. Bare sandy area between Hudsonia dunes, area #1. Dunes on east side of island, Inward Point, area #3. Eleocharis halophila Fern. & Brack. Swale on border of fresh water ponds, area #1. Eleocharis parvula (R. & S.) Link. Edge of brackish ponds. Very common, area #1. Eriophorum virginicum L. Wet area in hollow of dunes, area £2. Scirpus americanus Pers. Swale, border of brackish ponds, area #1. Dune-marsh transition zone, east side of island. Inward Point, area #3. Scirpus cyperinus (L.) Kunth. Marsh area north of brack- ish ponds, area #1. Wet thickets in middle of island, area £2. Juncus acuminatus Michx. Wet area near brackish ponds, area #1. Juncus effusus L. In fresh water marsh. Inward Point, area #3. Juncus gerardi Loisel. Salt marsh. Inward Point, area #3. Juncus greenei Oakes & Tuck. Wet area near brackish ponds, area #1. Sandy soil, hollow of dunes, area #2. Smilacina stellata (L.) Desf. var. crassa Vict. Small plants. In path cut through thickets around fresh water marsh, area #3. 24 Rhodora [Vol. 71 Smilax rotundifolia L. Tangle on shrubs of thickets, area #3. Iris versicolor L. Around artificial ponds in area #2. In fresh-water marsh at Inward Point, area #3. Sisyrinchium angustifolium Mill. Around fresh-water ponds, area #2. Pogonia ophioglossoides (L.) Ker. In sphagnum-cranberry bog between dunes in area #2. Fruiting. Salix discolor Muhl. Forming small thickets, center of island in area £2. Identified by David Fairbrothers. Myrica pensylvanica Loisel. Very common in thickets, area #1. Also present in areas #2 and #3. Polygonum pensylvanicum L. In swale, bordering brackish water ponds, area #1. Polygonum punctatum Ell. var. leptostachyum | (Meisn.) Small. Fairly common in swale bordering brackish ponds, area #1. Also bordering fresh-water ponds, center of island in area #2. Rumex acetosella L. Beach in area #1, Path in thickets in area #3. Rumex crispus L. Wet area back of dunes bordering brack- ish ponds in area #1. Rumex maritimus L. var. fueginus (Phil) Dusen. Small plants growing in belt around brackish pond, just above water level, area #1. Atriplex arenaria Nutt. Beach vegetation, areas #1 and #3. Flowers and seeds. Atriplex patula L. var. hastata (L.) Gray. Dominant in beach vegetation, area #1. Salicornia bigelovii Torr. Bare areas in salt marsh, area #3. Salicornia europaea L. Salt marsh. “Pannes” and in S. patens zone, area #3. Salicornia virginica L. Salt marsh. “Pannes” and in S. patens zone, area #3. Suaeda maritima (L.) Dumort. Edge of salt marsh, area #3. Salsola kali L. Dominant on beach, area #1. 1969] Monomoy Island — Moul 25 Phytolacca americana L. Around Monomoy Light Tower, area #2. Arenaria peploides L. var. robusta Fern. Beach on ocean side, area #3. Dianthus armeria L. In thickets of bayberry, between fresh- water marsh and dunes, area #1. Spergularia marina (L.) Grisch. In sand at edge of brack- ish ponds in area #1. Bare area in salt marsh, area #3. Nymphaea odorata Ait. In fresh-water ponds (artificial), area £2. Cakile edentula (Bigel. Hook. Beach, area £1. Lepidium virginicum L. Along edge of brackish ponds, higher ground at edge of bayberry thickets, area £1. Drosera rotundifolia L. Sphagnum-cranberry bogs between dunes. Area £2. Amelanchier laevis Wieg. Common in thickets in center of island. Areas £2 and £3. Potentilla egedei Wormsk. var. groenlandica (Tratt.) Polunin. Fairly common in Juncus gerardi zone of salt marsh, area #3. Prunus maritima Marsh. A common component of thickets in center of the island and around fresh-water and brackish water ponds. Areas #1, #2, and #3. Prunus persica (L.) Batsch. In a hedge, along side aban- doned house at lighthouse, area £2. Prunus serotina Ehrh. Part of dune thickets, center of island. Areas £2 and £3. Pyrus arbutifolia (L.) L.f. Thickets in center of island and around fresh-water marsh. Areas £2 and #3. Rosa carolina L. Thickets along road in dunes, area #3. Rosa nitida Willd. Roadside and dune, area £3. Rosa rugosa Thunb. Dunes and thickets. Areas £1 and #2. Rubus hispidus L. Trailing in path through thickets around fresh-water marsh. Rubus idaeus L. var. strigosus (Michx.) Maxim. Thickets along roadsides and dunes. With Rhus radicans. Fruits. Area #3. 26 Rhodora [Vol. 71 Spiraea tomentosa L. Edge of sphagnum-cranberry bog. Area #2. Lathyrus japonicus Willd. Part of beach vegetation and frontal dunes. Very common. Areas #1, #2, & #3. Euphorbia polygonifolia L. Plants on beach, smaller ones in dunes. Area #1. Rhus radicans L. Very common. In thickets, back of dunes. Areas #1, #2, and £3. Ilex verticillata (L.) Gray. Thickets, tops of dunes in cen- ter of island, area #3. Hypericum boreale (Britt.) Bickn. Dry area between fresh- water marsh and dunes, area £1. Hypericum virginicum L. Wet area between dunes, area £2. Fresh-water marsh at Inward Point, area £3. Hudsonia tomentosa Nutt. Very common on secondary dunes. All areas. Lechea maritima Leggett. Scattered over dunes in area £3. Viola lanceolata L. Sphagnum-cranberry bog in hollow of dunes. Area #2. Ludwigia palustris (L.) Ell. var. americana (DC) Fern. & Grisc. Very abundant in fresh-water marsh, area #1. Daucus carota L. Weedy area around old lighthouse station, area #2. Ptilimnium capillaceum (Michx.) Raf. Wet areas near brackish and fresh-water marshes and ponds, areas #1 and #2. Lyonia ligustrina (L.) DC. Part of thickets around fresh- water marsh, area #3. Vaccinium angustifolium Ait. var. nigrum (Wood) Dole. Thickets on dune top, center of island, area #3. Vaccinium corymbosum L. Thickets around fresh-water marsh. Fruit abundant 1965, none in 1966 and 1967, area £3. Vaccinium macrocarpon Ait. Small plant in wet swale in area #1. Abundant in several bogs in hollow of dunes, area #2. Trientalis borealis Raf. Sandy soil in path and under thickets around fresh-water marsh. Area #3. 1969] Monomoy Island — Moul 27 Limonium carolinianum (Walt.) Britton. Bare areas and in Spartina patens zone of salt marsh. Area #3. Lycopus americanus L. Wet area around brackish ponds. Area #1. Scutellaria epilobiifolia Hamilton. Common around edges of artificial fresh-water ponds. Area #2. Teucrium canadense L. Common around artificial fresh- water ponds. Area #2. Linaria canadensis (L.) Dumont. Sandy soil on dunes. Flowering over and seed formed. Area #3. Linaria vulgaris Hill. Depression in dunes near west shore. Not blooming. Area #3. Verbascum thapsus L. Edge of a depression near fresh- water ponds. Area £1. Plantago oliganthos R. & S. Bare area in salt marsh. Area #3. Galium palustre L. Wet area near brackish ponds. Very common. Area #1. Viburnum dentatum L. Thickets around fresh-water marsh. Area #3. Achillea millefolium L. Weedy area around old lighthouse, area #2. Artemisia caudata Michx. Common on dunes. Better devel- oped in center of island. Areas #1 and #2. Artemisia stelleriana Bess. Beach vegetation. Area #1. Aster sp. Near old lighthouse. Area #2. Baccharis halimifolia L. Low dunes and high part of salt marsh. Area #3. Chrysopsis falcata (Pursh.) Ell. Very common on dunes in all areas. In 1965 attacked and defoliated by cater- pillars. Cirsium vulgare (Savi) Tenore. Edge of bayberry thickets near brackish ponds, area £1. Erichtites hieraceifolia (L.) Raf. Weedy area around old lighthouse. Wet area. Area £2. Gnaphalium obtusifolium L. Between dunes and brackish ponds. Area #1. 28 Rhodora [Vol. 71 Pluchea purpurascens (Sw.) DC. var. succulenta Fern. Swale at border of brackish water ponds in Typha marsh, area #1. Solidago sempervirens L. Beach and dunes. Areas #1, #2, and #3. Solidago tenuifolia Pursh. Between dunes and near fresh- water ponds, area #2. Between dunes and salt marsh, area #3. Sonchus asper (L.) Hill. Edge of pond and thickets. Area #1. Xanthium echinatum Murr. Beach and dunes. Areas #2 and #3. LITERATURE CITED FERNALD, M. L. 1950. Gray’s Manual of Botany. Amer. Book Co., New York. Focc, JOHN M., JR. 1930. The Flora of the Elizabeth Islands, Mass. Rhodora 32: 119-132; 147-161; 167-180; 208-221; 226-258; 263-281. GROUT, A. J. 1940. List of mosses of North America north of Mexico. Bryologist 43: 117-131. JORDAN, D. S. 1874. The Flora of Penikese Island. Amer. Nat. VII (4): 193. LEwis, I. F. 1924. Flora of Penikese, Fifty Years After. Rhodora 26: 181, 211, 222. Mout, E. T. 1948. Flora of Penikese Island. Rhodora 50: 288-304. —— —— 1961. Notes on the Flora of Penikese Island, Mass., Rhodora 63: 149-150. DEPARTMENT OF BOTANY. RUTGERS — THE STATE UNIVERSITY NEW BRUNSWICK, NEW JERSEY. 08903 A FLORISTIC AND ECOLOGICAL STUDY OF PITCHER PLANT BOGS IN SOUTH MISSISSIPPI L. N. ELEUTERIUS and S. B. JONES, JR. An interest in the effects of fire on the vascular flora of pitcher plant bogs led to a floristic and ecological study of such bogs in south Mississippi. This included the intensive comparison of 2 peaty bog, which had been burned annually in the winter for the past seven years, with a similar bog which had not been burned for three years. The study area is located in Harrison County, Mississippi, northwest of Gulfport and approximately six miles from the Gulf Coast, and on the southern edge of the Southern Pine Hills. The bogs occupy a position between the hilltop pine forest and tne bottomland forest along Bayou Bernard. The two bogs that were studied are both open with a few scattered pine and cypress trees. The soil is an acid, Portmouth, sandy loam with a high organic matter content. The bogs in this area appear to have been only slightly disturbed by the activities of European man. The ecological aspects of this study are reported in detail in Eleuterius (1968); however, some of our conclusions, observations, and hypotheses are briefly summarized in this paper. We do not suggest that our observations have not been previously noted, but we feel that they should be mentioned here. Pitcher plant bogs have been discussed in a number of papers including Garren (1943), Harper (1906), Plummer (1963 and 1966), and Wells (1928). Since the earlier report by Harper (1906), there have been changes in the vascular flora of pitcher plant habitats in Georgia due to the disturbance of these ecosystems. These changes were described by Pullen and Plummer (1964). We believe that Sarracenia alata is dependent upon fire to maintain the open bog, as are a number of our native bog orchids. In the burned bog, Sarracenia alata was more vigorous, with both larger leaves and rhizomes; while in the absence of fire Sarracenia alata tends to de- crease in abundance. The addition of 6-12-12 fertilizer 29 30 Rhodora [Vol. 71 (1 1b/100 sq. ft.) and ammonium nitrate (2 1b/100 sq. ft.), divided into two applications each, caused a decrease in Sarracenia alata. The bogs are not deficient in N-P-K as some have suggested; however, the amount of N-P-K de- creased during the growing season. The nutrient levels in pounds per acre were: N, April 90, September 18; P, April 160, September 140; K, April 150, September 100. Plant succession on the bogs was toward a “sedge-woody species" dominated community. Fire, however, retards this suc- cession, and we believe that the pitcher plant community is a "fire type" vegetation. Clip-quadrat studies pointed out that fire increases plant productivity. The greatest number of species were found in the burned bog, while the unburned bog was dominated by sedges, with a reduction in the total number of species. Although plants are in flower from March to November, there are two major phenological peaks, one in June and one in August. The unique floral beauty and composition of the bogs make them of scientific and aesthetic importance. These ecosystems, however, are in danger of being destroyed by man's activities; such as, the expansion of suburban and industrial complexes northward from the coast, and the draining of the bogs and the planting of pine by public and private interests. A phylogenetic list of the plant species is presented below. The phylogenetic arrangement and nomenclature generally follows that of Radford, et al. (1964), although Small (1933) and Hitchcock and Chase (1950) were also used in identification. Species not considered characteristic of bogs were also collected because they were present within the bogs due to slight alteration of the ecosystem by either man, ant hills, crayfish, or slight elevations or depressions. They are indicated by an asterisk (*) in the phylogenetic listing. We would like to thank several individuals for their help with the project, including: Drs. Ben Channel, Robert Kral, and Sidney McDaniel, and Mr. F. H. Sargent. The 1969] Pitcher Plant Bogs — Eleuterius & Jones 31 project was partially supported with the aid of a N.S.F. Grant (GB-4635) to the Mississippi Flora Project. PHYLOGENETIC LIST OF THE TAXA; LYCOPODIACEAE — Lycopodium alopecuroides L., L. adpressum (Chapm.) Lloyd & Underw., L. carolinianum L., L. prostratum Harper. OSMUNDACEAE — Osmunda, cinnamomea L., O. regalis var. spectabilis (Willd.) Gray. PTERIDACEAE — Pteridium aquilinum (L.) Kuhn. BLECHNACEAE — Anchistea virginica (L.) Presl., Lorinseria areolata (L.) Presl. PINACEAE — Pinus elliottii Engelm., P. palustris Mill., P. taeda L. TAXODIACEAE — Taxodium ascendens Brongn. POACEAE — Andropogon glomeratus (Walt.) BSP., Anthaenantia vil- losa (Michx.) Beauv., Aristida affinis (Schult.) Kunth., A. pur- purascens Poir., A. stricta Michx., Arundinaria gigantea (Walt.) Muhl., Ctenium aromaticum (Walt.) Wood., Eragrostis specta- bilis (Pursh) Steud., Manisuris cylindrica (Michx.) Kuntz., Muhlenbergia expansa (DC.) Trin., Panicum aciculare Desv. ex Poir., P. commutatum Schult., P. lanuginosum Ell., P. leucothrix Nash., P. longiligulatum Nash., P. repens L., P. sphaerocarpon Ell., P. verrucosum Muhl., P. virgatum L., P. wrightianum Schribn., Paspalum difforme LeConte., P. floridanum Michx., * P. notatum Flugge., P. urvillei Steud., Sporobolus junceus (Michx.) Kunth., Tridens ambiguus (Ell) Schult., T. strictus (Nutt.) Nash. CYPERACEAE — Carex glaucescens Ell., C. lurida Wahlenb., C. stylo- flexa Buckl., Dichromena colorata (L.) Hitchc., D. latifolia Baldw., Eleocharis tuberculosa (Michx.) R. & S., Rhynchospora caduca ElL, R. cephalantha Gray., F. chalarocephala Fern. & Gale., R. chapmanii M. A. Curtis., R. corniculata (Lam.) Gray., R. filifolia Gray., R. glomerata, (L.) Vahl., R. gracilenta Gray., R. harveyi W. Booth., R. inexpansa (Michx.) Vahl., R. macra (Clarke) Small., R. microcephala Britt. ex. Small., R. oligantha Gray., R. plumosa E!l., R. rariflora (Michx.) Ell., R. tracyi Britt., Scir- pus americanus Pers., Scleria baldwinii (Torr.) Steud., S. geor- giana Core., S. oligantha Michx., S. reticularis Michx. XYRIDACEAE — Xyris ambigua Beyr., X. baldwiniana R. S., X. caro- liniana Walt., X. difformis Chapm., X. fimbriata Ell., X. flexuosa Muhl. ex Ell., X. iridifolia Chapm. ERIOCAULACEAE — Eriocaulon compressum Lam., E. decangulare L., E. lineare Small., Lachnocaulon anceps (Walt.) Morong., L. minus (Chapm.) Small. BROMELIACEAE — *Tillandsia usneoides L. 32 Rhodora [Vol. 71 JUNCACEAE — Juncus biflorus Ell., J. brachycarps Engelm., J. cana- densis J. Gay ex La Harpe., J. effusus L., J. elliottii Chapm., J. marginatus Rostk., J. scirpoides Lam., J. validus Cov. LILIACEAE — Aletris aurea Walt., A. farinosa L., A. lutea Small., Lilium catesbaei Walt., Melanthium virginicum L., Smilax lauri- folia L., S. rotundifolia L., S. smallii Morong., Tofieldia racemosa (Walt.) BSP., Zigadenus densus (Desr.) Fern., Z. glaberrimus Michx. AMARYLLIDACEAE — Hypoxis micrantha Pollard. HAEMODORACEAE — Lachnanthes caroliniana (Lam.) Dandy., Lophiola americana, (Pursh.) Wood. IRIDACEAE — Sisyrinchium albidum Raf. ORCHIDACEAE — Calopogon pallidus Chapm., C. pulchellus (Salisb.) R. Br., Cleistes divaricata (L). Ames., Habenaria blephariglottis (Willd.) Hook., H. ciliaris (L.) R. Br., H. cristata (Michx.) R. Br., H. nivea (Nutt.) Spreng., Pogonia ophioglossoides (L.) Ker-Gawl., Spiranthes praecox (Walt.) S. Wats., S. vernalis Engelm. & Gray. MYRICACEAE — Myrica cerifera L., M. inodora Bart., M. pensylvanica Loisel. PHYTOLACCACEAE — *Phytolacca americana L. MAGNOLIACEAE — Magnolia virginiana L. LAURACEAE — Persea borbonia (L.) Spreng. SARRACENIACEAE — Sarracenia alata Wood & Wood., S. psittacina Michx. DROSERACEAE — Drosera brevifolia Pursh., D. capillaris Poir., D. fili- formis Raf., D. intermedia Hayne. SAXIFRAGACEAE — Itea virginica L. HAMAMELIDACEAE — Hamamelis virginiana L., Liquidambar styra- ciflua L. ROSACEAE — *Rubus trivialis Michx., Sorbus arbutifolia (L.) Heynh. FABACEAE — *Clitoria mariana L., *Lespedeza capitata Michx., *Petalostemum pinnatum (Walt. ex J. F. Gmel.) Blake. LINACEAE — Linum striatum Walt., L. virginianum L. POLYGALACEAE — Polygala cruciata L., P. cymosa Walt., P. incarnata L., P. lutea L., P. mariana Mill. EUPHORBIACEAE — *Croton capitatus Michx., *Euphorbia chamaesyce L., *E. corollata L., *E. maculata L. ANACARDIACEAE — Rhus vernix L. CYRILLACEAE — Cliftonia monophylla (Lam.) Sarg., Cyrilla racemi- flora L. AQUIFOLIACEAE — Ilex coriacea (Pursh) Chapm., I. glabra (L.) Gray., I. vomitoria Ait. ACERACEAE — Acer rubrum L. RHAMNACEAE — Ceanothus americanus L. VITACEAE — *Vitis rotundifolia Michx. 1969] Pitcher Plant Bogs — Eleuterius & Jones 33 MALVACEAE — Hibiscus aculeatus Walt. HYPERICACEAE — Hypericum cistifolium Lam., H. denticulatum H.B.K., H. fasciculatum Lam., H. stans (Michx.) Adams & Robson. VIOLACEAE — Viola lanceolata L., V. primulifolia L. LYTHRACEAE — Cuphea carthagensis (Jacq.) Macbr. MELASTOMATACEAE — Rhexia alifanus Walt., R. cubensis Griseb. R. lutea Walt., R. mariana L., R. Nashii Small., R. virginica L. ONAGRACEAE — Ludwigia hirtella Raf. L. pilosa Walt., *Oenothera fruticosa L. HALORAGACEAE — Proserpinaca pectinata Lam. APIACEAE — Centella asiatica (L.) Urban., Eryngium integrifolium Walt., E. yuccifolium Michx. var. synchaetum Gray., Hydrocotyle umbellata L., Oxypolis filiformis (Walt.) Britt. NYSSACEAE — Nyssa sylvatica Marsh., N. sylvatica Marsh. var. biflora (Walt.) Sarg. CLETHRACEAE — Clethra alnifolia L. ERICACEAE — Gaylussacia dumosa (Andr.) T. & G., Lyonia ligustrina (L) DG: SYMPLOCACEAE — Symplocos tintoria (L.) L’Her. STYRACACEAE — Styrax americana Lam., S. grandifolia Ait. LOGANIACEAE — Cynoctonum sessilifolium J. F. Gmel., Gelsemium sempervirens (L.) Ait. f., *Polypremum procumbens L. GENTIANACEAE — Sabatia campanulata (L.) Torr., S. dodecandra (L.) BSP., S. macrophylla Hook. ASCLEPIADACEAE — Asclepias cinerea Walt., A. lanceolata Walt., A. longifolia Michx., A. michauxii Dene. CONVOLVULACEAE — Cuscuta gronovii Willd. POLEMONIACEAE — Phlox carolina L., P. pilosa L. VERBENACEAE — Lippia nodiflora (L.) Michx. LAMIACEAE — Hyptis alata (Raf.) Shinners., Lycopus rubellus Moench., Scutellaria integrifolia L., Trichostema dichotomum L. SCROPHULARIACEAE — Agalinis aphylla (Nutt.) Raf., A. purpurea (L.) Pennell., Buchnera floridana Gandoger. LENTIBULARIACEAE — Pinguicula caerulea Walt., P. lutea Walt., Utricularia cornuta Michx., U. fibrosa Walt., U. juncea Vahl., U. subulata L. ACANTHACEAE — Ruellia noctiflora (Nees) Gray. RUBIACEAE — Cephalanthus occidentalis L., *Diodia teres Walt., D. virginiana L. CAPRIFOLIACEAE — *Lonicera sempervirens L., Viburnum nudum L. CAMPANULACEAE — Lobelia floridana Chapm., L. glandulosa Walt., L. puberula Michx. ASTERACEAE — Aster dumosus L., A. lateriflorus (L.) Britt., A. linarii- folius L., A. paludosus Ait., A. pilosus Willd., A. reticulatus Pursh., A. umbellatus Mill., Balduina uniflora Nutt., Bidens frondosa L., Boltonia asteroides (L.) L’Her., B. diffusa Ell., 34 Rhodora [Vol. 71 Cacalia lanceolata Nutt., Carduus lanceolatus L., C. lecontei (T. & G.) Pollard., C. spinosissimus Walt., Carphephorus pseudo- liatris Cass. Chaptalia tomentosa Vent., Chondrophora nudata (Michx.) Britt., Coreopsis angustifolia Ait., *C. lanceolata L., C. major Walt., C. pubescens Ell, Elephantopus elatus Bertol., *Erechtites hieracifolia (L.) Raf., *Erigeron canadensis L., *E. philadelphicus L., *E. quercifolius Lam., *E. strigosus Muhl. ex Willd., E. vernus (L.) T. & G., Eupatorium album L., *E. capilli- folium (Lam.) Small, E. coelestinum L., E. hyssopifoliwm L., E. leucolepis (DC.) T. & G., E. rotundifolium L., *Helenium amarum (Raf.) H. Rock., H. vernale Walt., Helianthus angusti- folius L., *H. divaricatus L., H. heterophyllus Nutt., H. radula (Pursh) T. & G., Heterotheca graminifolia (Michx.) Shinners., H. mariana (L.) Shinners, H. nervosa (Willd.) Shinners., Liatris aspera Michx., L. elegans (Walt.) Michx., L. gracilis Pursh., L. spicata (L.) Willd. var. resinosa (Nutt.) Gaiser., Pluchea rosea Godfrey., Rudbeckia hirta L., Solidago odora Ait., S. petiolata Mill., Stokesia laevis (Hill) Greene., Trilisa odora- tissima (J. F. Gmel.) Cass., Vernonia altissima Nutt. LITERATURE CITED ELEUTERIUS, L. N. 1968. Floristics and ecology of coastal bogs in Mississippi. M.S. Thesis. Univ. of Southern Mississippi, Hatties- burg. GARREN, K. H. 1943. Effects of fire on vegetation of southeastern United States. Bot. Rev. 9: 617-654. HARPER, R. M. 1906. A phytogeographical sketch of the Altamaha Grit Region of the Coastal Plain of Georgia. Ann. N.Y. Acad. 7: 1-415. HITCHCOCK, A. S. and A. CHASE. 1950. Manual of the grasses of the United States. USDA Misc. Pub. 200. 1050 p. PLUMMER, G. L. 1963. Soils of pitcher plant habitats in the Georgia Coastal Plain. Ecology. 44: 727-734. 1966. Foliar absorption in carnivorous plants. Carolina Tips 29: 25-26; 29-30. PULLEN, T. M., JR. and G. L. PLUMMER. 1964. Floristie changes within pitcher plant habitats in Georgia. Rhodora 66: 375-381. RADFORD, A. E., H. E. AHLES, and C. R. BELL. 1964. Guide to the vascular flora of the Carolinas. The Book Exchange, Univ. of North Carolina, Chapel Hill. 375 p. SMALL, J. K. 1933. Manual of the Southeastern Flora. (Reprint) The University of North Carolina Press, Chapel Hill. 1552 p. WELLS, B. W. 1928. A southern upland grass-sedge bog: An eco- logical study. N. C. Agr. Exp. Sta. Tech. Bull. 32. DEPARTMENT OF BIOLOGY, UNIVERSITY OF SOUTHERN MISSISSIPPI, HATTIESBURG. DEPARTMENT OF BOTANY, UNIVERSITY OF GEORGIA, ATHENS THE BROMELIACEAE OF BOLIVIA LYMAN B. SMITH Originally this paper was prepared as an integral part of Robert C. Foster’s Flora of Bolivia and although it is no longer possible to so include it, I am continuing that for- mat in order to relate to his work as closely as possible. Reference to the “Catalogue” is to R. C. Foster, A catalogue of the ferns and flowering plants of Bolivia, Contrib. Gray Herb. 184, 1958. a. Seeds appendaged; ovary wholly or partly superior; fruit a cap- sule. b. Appendages of the seeds entire, not plumose; ovary often only partly superior; leaves mostly spinose-serrate; plants almost always terrestrial uu T S. Subfamily 1. Pitcairnioideae. c. Filaments not forming a tube; petals free. d. Petals twisted together after anthesis, mostly large (13-80 mm.); plants usually large and coarse; leaf-blades narrowly triangular, never narrowed toward base; ovary nearly or quite superior. ............................................ 1. Puya. d. Petals free after anthesis. e. Ovary wholly superior. f. Petals naked, small (3-9 mm.); plants small (25- 100 em.) and delicate; seeds narrow, caudate-append- aged at each pole. ........................................ 2. Fosterella. f. Petals each bearing a single large scale at base. g. Scape evident, bearing a cambium layer; inflorescence paniculate, perennial; plants shrubby. ............ HEC tU vn 3. Deuterocohnia. g. Scape lacking; inflorescences one-flowered, at the ends of the branches; plants low and cushion-forming. .... MEL or 4. Abromeitiella. e. Ovary partly inferior; flowers large and conspicuous, usually zygomorphie; petals often appendaged; seeds winged or caudate. ............. eene 5. Pitcairnia. c. Filaments forming a tube to which the petals are joined by their centers; petals with their margins free, yellow or orange; seeds winged. ............ mH 6. Dyckia. b. Appendages of the seeds plumose; ovary nearly or quite supe- rior; leaves always entire; plants chiefly epiphytic. „eseese Hcet vmi e riis EN MPH Subfamily 2. Tillandsioideae. h. Petals free, or if slightly joined then with the corolla-tube deeply included in the calyx; inflorescence of one or more 35 36 Rhodora [Vol. 71 distichous-flowered spikes or rarely simple and polystichous or even one-flowered. i. Petals naked. ............. sss 7. Tillandsia. i. Petals each bearing two scales on the inner surface near the base. MM 8. Vriesea. h. Petals closely agglutinated and simulating true fusion, the corolla-tube about equaling the calyx or longer; flowers al- ways polystichous. .............. eene 9. Guzmania. a. Seeds always naked; ovary wholly inferior; fruit always baccate, fleshy to coriaceous; leaves usually spinose-serrate; plants often epiphytic. ................. ra rra, Subfamily 3. Bromelioideae. j. Petals naked; inflorescence compound. k. Filaments free from each other; petals free, mot fleshy; flowers distichous. ............s sees 10. Streptocalyx. k. Filaments forming a tube to which the fleshy petals are joined along their centers but with their margins free; flowers polystichous. ................. nn ennt entes 11. Bromelia. j. Petals appendaged. l. Ovaries always remaining distinct; inflorescence compound or simple. m. Petals straight, mostly small; sepals usually mucronate. .... — ÉOÉ— ERR 12. Aechmea. m. Petals recurved in a tight coil at anthesis, large (5-9 cm.) ; Sepals unarmed. ............. eee 13. Billbergia. l. Ovaries fused with each other and with the fleshy bracts to form a syncarp; inflorescence with an apical coma, simple. n. Inflorescence with a small inconspicuous coma, never pro- ducing basal shoots; plant propgating by elongate rhizomes; petals bearing vertical folds. ...................... 14. Pseudananas. n. Inflorescence with a large conspicuous coma, often with basal shoots; rhizome lacking; petals bearing well developed scales. ——————— 'ÉRRRRRRRRRRRRRRRRRRRERREA 15. Ananas. 1. Puya Molina Perennial, often long-caulescent, simple or branching, coarse, often flowering several meters high. Leaves densely rosulate, the blades narrowly triangular, usually spinose-serrate with coarse teeth. In- florescence simple or compound, the branches fertile throughout or with the apical half sterile. Flowers showy, regular. Sepals free, shorter than the petals. Petals free, usually spreading at anthesis, twisted together afterwards, naked or rarely with paired vertical folds. Stamens usually included, the filaments free. Ovary superior or nearly so, glabrous, the style slender, elongate. Capsule primarily loculicidal. Seeds broad, alate. a. Inflorescence and its branches fertile throughout or nearly so. . RR Subgenus 1. Pitcairniopsis. 1969] Bromelliaceae — Smith 37 b. Ultimate axes of the inflorescence more or less visible; neither the inflorescence nor its branches strobiliform. c. Inflorescence compound. ............u1 ms ns s.s Subkey I. ca Inflorescence SINDEN uynuy sb Subkey II. b. Ultimate axes of the inflorescence completely covered; inflores- cence or its branches strobilate (sometimes the flowers or the bracts reflexed but the axis still covered). d. Inflorescence obviously compound (at least by the lobed out- UTC MEM NERA ptt e Subkey III. d. Inflorescence simple or pseudosimple with the short branches covered by the primary bracts. u sm ua. Subkey IV. a. Inflorescence compound with branches sterile in large part toward Ciro dup saran n puasa SS i au sh ee Subgenus 2. Puya. Subgenus 1. PITCAIRNIOPSIS Subkey I a. Branches of the inflorescence more than twice as long as the pri- mary bracts, divergent to spreading; inflorescence usually broad. b. Floral bracts coarsely pectinate-serrate. ................ 1. P. ultima. b. Floral bracts entire or nearly so. c. Sepals sparsely lepidote or glabrous. d. Branches at least three times as long as the primary bracts. e. Racemes lax throughout, the rhachis wholly visible. f. Sepals alate-carinate throughout, even. ........ 2. P. alata. f. Sepals carinate or alate only at base, strongly sulcate when dry in most cases. g. Pedicels soon glabrous. h. Floral bracts about equaling the pedicels; pedicels 15-18 mm. long. ............................. 3. P. Hofstenii. h. Floral bracts about half as long as the pedicels. i. Primary bracts about equaling the sterile bases of the branches; sepals lanceolate, 21 mm. long. .... MONI aa asses auri Dap ans 4. P. potosina. i. Primary bracts much shorter than the sterile bases of the branches; sepals oblong, 15 mm. Jong. creer menn ii Rat dir RR qasa 5. P. Trollüi. g. Pedicels persistently vestite. j. Pedicels 20-25 mm. long; floral bracts oblong-ovate. HELME EE e EAD MENTI 6. P. Rutyi. j. Pedicels 5-15 mm, long; floral bracts broadly ovate or elliptic. k. Sepals alate at base, 25-32 mm. long. .................... S s ass it RE 7T. P. cristata. k. Sepals not more than angled at base. l. Leaf-blades tomentose beneath with coarse linear spreading scales; sepals 17 mm. long. .... PR Sia spaces ne ENIRO 8. P. laxa. 38 Rhodora [Vol. 71 l. Leaf-blades with subspreading suborbicular scales beneath; sepals 20-28 mm. long, ............ uu a n L ''— 9. sanctae-crucis. e. Racemes subdense; rhachis only partially exposed at an- thesis. m. Sepals 17 mm. long, scarcely if at all exceeding the floral bracts; pedicels glabrous. ............ 10. P. Fiebrigii. m. Sepals 28-40 mm. long; pedicels tomentose-lepidote. n. Floral bracts lanceolate, nearly flat. .. 11. P. Meziana. n. Floral bracts ovate, inflated. o. Primary bracts much shorter than the long, sterile bases of the branches; floral bracts much exceeded by the sepals. .......... ee 12. P. secunda. o. Primary bracts far exceeding the short, sterile bases of the branches; floral bracts nearly equaling the sepals. .............. eene 13. P. stenothyrsa. d. Branches little more than twice as long as the primary bracts. p. Floral bracts lanceolate, attenuate. ........ 14, P. glareosa. p. Floral bracts elliptic, ample, rounded and apiculate. q. Pedicels to 25 mm. long. .....................-— 15. P. olivacea. q. Pedicels not over 10 mm. long. ........ 13. P. stenothyrsa. c. Sepals densely and persistently vestite. r. Indument ferruginous-stellate with stiff-rayed trichomes. .... vevuceceeecsscceeeececcessesnnseeeeeeceeeeeecesenseeaeeeseeeenteneeenes 16. P. ferruginea. r. Indument obscurely stellate with soft-rayed trichomes. s. Pedicels over 15 mm. long. t. Floral bracts drying dark; pedicels Stout. ............... MEME 17. P. riparia. t. Floral bracts not drying dark; pedicels slender. u. Primary bracts about one-fifth as long as the branches; sepals acute, 30 mm. long. ....... 18. P. alba. u. Primary bracts almost half as long as the branches; sepals obtuse and mucronulate, 35-40 mm. long. ........ MNMMMMRRRERRRRRERERENCREREMR 15. P. olivacea. s. Pedicels less than 15 mm. long; sepals narrow, acute or attenuate with a narrow blunt apex. v. Lateral branches over 30 em. long; floral bracts broadly elliptic. sessavaeurssservevessesvernesntsenerctorserenssyenioss® 19. P. Pearcei. v. Lateral branches not over 9 em long; floral bracts ovate. :NEMMNMNMMNMRNRRRMRRMMMMWMMMMMMMMMMEMMMMMMMMM 20. P. penduliflora. a. Branches of the inflorescence not more than twice as long as the primary bracts; inflorescence narrow. w. Sepals glabrous or with a fine appressed deciduous stellate or lepidote indument. x. Leaf-blades 12-30 mm. wide. 1969] Bromelliaceae — Smith 39 y. Branches spreading to slightly deflexed; inflorescence drying almost Diack pee un UU n T n 21. P. atra. y. Branches erect or ascending. z. Sepals 12 mm. long, ovate-elliptic. ........ 22. P. dyckioides. z. Sepals 26-35 mm. long, narrow. A. Floral bracts attenuate, lanceolate, slightly exceeding the pedicels sini ar oce ee s cesses 14. P. glareosa. A. Floral bracts acute, elliptic, about equaling the sepals. DEMON D n A n wal nn 2 EE 23. P. glabrescens. x. Leaf-blades 2-10 mm. wide, covered beneath with pale, nar- row, spreading: seales; lateral branches few-flowered. B. Pedicels 4-7 mm. long, longer or shorter than the floral bracts; leaf-blades 2-3.5 mm. wide. ............ 24. P. tuberosa. B. Pedicels 1 mm. long, the floral bracts 15 mm. long; leaf- blades -10 mm. wide oesie usss tese s. 25. P. reducta. w. Sepals densely lanate. ................................................ 26. P. mollis. Subkey II a. Sepals 12-60 mm. long; flowering shoot over 3 dm. high. b. Inflorescence densely and coarsely ferruginous-stellate. ................ DEI Ld anti Rice tn EEE E ED 16. P. ferruginea. b. Inflorescence white-tomentellous to glabrous. e. Floral bracts laciniate-spinose; sepals nearly 6 cm. long. ........ mone N PENE N EEEE TE SE = ER 27. P. mirabilis. e. Floral bracts subspinulose to entire; sepals 25-40 mm. long. d. Sepals 25 mm. long; floral bracts uniform, narrowly lance- E n i, ius O EE E E E 28. P. Kuntzeana. d. Sepals 30-40 mm. long; lower floral bracts much larger and ampler than the others. ................ nu... 11. P. Meziana. a. Sepals 7-11 mm. long; flowering shoot slender, 9-30 cm. high. e. Trichomes of the leaf-blade linear, spreading; flowering shoot DASE RD om hiph. o oe ee roo 24. P. tuberosa. e. Trichomes of the leaf-blade broad, subspreading; flowering shoot ure Bib. ias a Tu G dama a n s s 29. P. minima. Subkey III a. Sepals glabrous or with fine appressed deciduous stellate or lepidote indument. b. Lateral branches well developed, the flowers not fasciculate. c. Floral bracts much exceeded by the sepals, ovate, inflated. ........ "ERR 13. P. stenothyrsa. c. Floral bracts nearly equaling to exceeding the sepals. d. Sepals 26 mm. long; floral bracts lustrous black. ................ com D c Eius ec M 2I E atra. d. Sepals 17 mm. long; floral bracts dull, paler. ........................ 10. P. Fiebrigü. sssssssssssssssssssssessssssesssssesesssesssesssssssssssssssssssssssesssesee 40 Rhodora [Vol. 71 b. Lateral branches very short with the flowers fasciculate, but the inflorescence interrupted at base and obviously compound; Sepals 11 mm. long. ......... sse 30. P. paupera. a. Sepals densely lanate or tomentose, the indument mostly persistent after anthesis. e. Floral bracts much exceeded by the sepals. f. Leaf-blades densely appressed-lepidote beneath; floral bracts broadly ovate, ca. 35 mm. long... II a TL, 26. P. mollis. f. Leaf-blades soon wholly glabrous; floral braets elliptic-oblong, to 55 mm. long. seses, 31. P. valida. e. Floral bracts exceeding the sepals. g. Sepals 45 mm. long, narrowly lanceolate or linear; racemes secund. MM 32. P. lasiopoda. g. Sepals 20-30 mm. long; racemes not secund. h. Branches ellipsoid, to 17 cm. long; pedicels 15 mm. long; Sepals 30 mm. long. ............ sees 38. P. Ugentiana. h. Branches globose, 6 cm. long; pedicels 7 mm. long; sepals 20 mm. long. sesser 34. P. ctenorhyncha. Subkey IV a. Outer bracts (the primary bracts or in a simple inflorescence the floral bracts) conspicuously serrate. b. Apices of the thin outer bracts reflexed; inflorescence densely brown-lanate. sese 35. P. Fosteriana. b. Apices of the outer bracts erect; sepals soon glabrous and con- trasting with the densely vestite pedicels. .... 28. P. glabrescens. a. Outer bracts entire or very obscurely serrulate. c. Apical part of each outer bract reflexed. d. Inflorescence compound; flowers fasciculate. e. Sepals 17-18 mm. long. f. Inflorescence 8 cm. in diameter; outer bracts 7 cm. long. RN 36. P. tristis. f. Inflorescence 3 cm. in diameter; outer bracts 5 cm. long. 1 (——— 'ÁÉÁÉÁÉEN 37. P. leptostachya. e. Sepals 25-30 mm. long. £. Bracts all membranaceous and subhyaline, soon shattered; leaf-blades nearly glabrous. ............ 38. P. Cardenasii. g. Bracts firmer, persistent; leaf-blades densely and con- spicuously appressed-lepidote beneath. .... 39. P. Herzogii. d. Inflorescence simple. h. Floral bracts narrow, leaving much of the flower exposed (juvenile) ; sepals 25 mm. long. ................ 28. P. Kuntzeana. h. Floral bracts ample; sepals 14 mm. long. ........ eee (— RR 40. P. tunarensis. c. Apical part of each outer bract erect. i. Inflorescence compound with the flowers fasciculate. 1969] Bromelliaceae — Smith 41 j- Scape well developed; axis of the inflorescence elongate; sepalis 11 mm. long s. unn es 80. P. paupera. j. Scape very short, completely hidden; inflorescence capitiform with all the axes aborted; sepals 35-40 mm. long, ................ PORE Se tr NM Tact Sale TES rr TES 41. P. nana. i. Inflorescence simple. k. Sepals thinly vestite to glabrous, l. Sepals 12-13 mm. long. .................... s ns 42. P. humilis. l. Sepals 25 mm. long. ............. ct 28. P. Kuntzeana. k. Sepals densely lanate, 22 mm. long. ........ 43. P. Brittoniana. Subgenus 2. PuYA a. Caudex branching and continuing to live after the fruition and death of the flowering shoot; sepals 18 mm. long, exceeding the ÜOENISEERADC 400 LIO ies necant L AT SSES 44. P. Weddelliana. a. Caudex simple and columnar, dying after the fruition of the flower- ing shoot; sepals 40 mm. long, exceeded by the floral bracts. ........ PEE SD S a T L E. 45. P. Raimondi. 1. Puya ultima L. B. Smith in Contrib. U. S. Nat. Herb. 29: 540, fig. 94 (1954). To 1.5 m. high. Leaf-blades 6 cm. wide, covered be- neath with white, appressed scales, becoming wholly glabrous, the spines ascending, 10 mm. long, broad, black. Inflorescence bipinnate, ample, very lax, glabrous after anthesis. Primary bracts broadly ovate, acuminate, 5 cm. long, about equaling the sterile, naked bases of the branches, pectinate-serrate, subchartaceous. Branches spread- ing, slender, 5 dm. long, lax. Floral bracts ovate, acuminate, 2 cm. long, equaling or exceeding the slender pedicels, pectinate-serrate. Flowers ascending. Sepals narrowly triangular, strongly carinate, 40 mm. long. LA PAZ: SUR vUNGAS: La Paz to Irupana, 1500 m., M. B. Foster 2575 (US; type). 2. P. alata L. B. Smith in Phytologia, 7: 422, t. 2, fig. 3, 4 (1961). P. spathacea of Catalogue Leaf-blades over 7 dm. long, 3 cm. wide, glabrous above, densely vestite beneath and on the margins with pale, subappressed scales, the spines lax, flat, antrorse, 10 mm. long. Inflorescence compound. Primary bracts elliptic, apiculate, 2 em. long, shorter than the sterile base of the branch, entire, sparsely and min- utely lepidote with linear white trichomes. Branches slightly flexuous, 3 dm. long, 3 mm. in diameter, very lax, angled, soon glabrous; floral bracts like the primary bracts, erect, much exceeding the pedicels. Pedicels slender, to 10 mm. long, sparsely pale-lepidote. Flowers pendent-secund. Sepals lanceolate, rounded and apiculate, 26 mm. long, even, soon glabrous, alate-carinate. PoTosi: LINARES: Lagunil- las, 1000 m., Cárdenas 2851 (F; type). 3. P. Hofstenii Mez in Fedde, Rep. Nov. Spec. 3: 8 (1906). P. Hauthalii Mez in Fedde, Rep. Spec. Nov. 16: 65 (1919). Over 1 m. high. Leaf-blades 5 dm. long, 5 em. wide, densely cinereous-lepidote, 42 Rhodora [Vol. 71 especially beneath, the spines antrorse, 5 mm. long. Inflorescence laxly bipinnate, 3-7 dm. long, glabrous. Primary bracts ovate, acute, about equaling the sterile bases of the branches, inflated, entire, stramineous. Branches spreading to suberect, to 20 em. long, lax. Floral bracts ovate-elliptic, about equaling the slender, 15-18 mm. long pedicels. Flowers suberect to deflexed. Sepals lanceolate, rounded and apiculate, 26 mm. long. Petals 40 mm. long, naked or appendaged. POTOSI: NOR CHICHAS: Tala Chorolque near Guadalupe, Hauthal 101 (B; type of Puya Hauthalii Mez); Quechisla, 3500 m., Cárdenas 339. Argentina. 4. P. potosina L. B. Smith in Phytologia, 7: 423, t. 2, fig. 7, 8 (1961). Leaf-blades over 5 dm. long, 25 mm. wide, densely vestite with pale, appressed scales but becoming glabrous above, the spines antrorse, 3 mm. long, brown. Scape 8 mm. in diameter at apex, glabrous. Scape-bracts small and exposing most of the upper scape, the ovate, serrulate base 2 cm. long, the slender, linear blade over 5 em. long. Inflorescence laxly bipinnate, 7 dm. long, glabrous, somewhat glaucous. Primary bracts like the upper scape-bracts but with much shorter blades or merely apiculate, about equaling the sterile bases of the branches. Branches suberect, nearly straight, to 25 cm. long, 4 mm. wide at the flattened base, lax. Floral bracts ovate, apiculate, to 15 mm. long, entire, thin. Pedicels slender, en- larged upward, to 30 mm. long. Flowers more or less secund, often subverticillate. Sepals lanceolate, rounded and apiculate, 21 mm. long, carinate at base, sulcate when dry. Petals 4 cm. long, naked. PoTosí: NOR CHICHAS: Quechisla, 3400 m., Cárdenas 3738 (US; type). 5. P. Trollii L. B. Smith in Phytologia 16: 461, t. 1, fig. 10, 11 (1968). Leaves over 7 dm. long; blades ca. 2 cm. wide, covered with pale appressed scales beneath, laxly serrate with dark antrorse spines 2 mm. long. Scape unknown. Inflorescence amply bipinnate, lax, soon glabrous, the fine white stellate trichomes uniform and completely deciduous. Primary bracts ovate, acuminate, much shorter than the sterile bases of the branches. Branches subspreading, to 26 cm. long, laxly flowered. Floral bracts broadly ovate, apiculate, mostly about half as long as the pedicels, thin, nerved, entire. Pedicels slender but strongly enlarged upward, to 20 mm. long. Flowers spreading. Sepals oblong, rounded, 15 mm. long, carinate toward base, sulcate when dry. Petals ca. 3 em. long. Potosi: Montecanto, Troll 661 (M; type). 6. P. Rusbyi (Baker) Mez in DC. Monog. Phan. 9: 482 (1896). Flowering over 1 m. high. Leaf-blades over 35 cm. long, 25 mm. wide, covered with appressed white scales on both sides, becoming glabrous. Inflorescence amply bipinnate, covered with fine, white, stellate scales. Primary bracts many times shorter than the branches. Branches lax, 20-30 em. long. Pedicels subspreading, slender, 20-25 1969] Bromelliaceae — Smith 45 mm. long, persistently vestite. Sepals lance-oblong, apiculate, 20-25 mm. long, more or less alate at base. Petals 4 cm. long, green and deep purple. Stamens exserted. LA PAZ: MURILLO: La Paz, 3000 m., Rusby 2846 (NY; type); Balls 5900 (F, US); San Jorge, 3600 m., West 6402 (UC). 7. P. cristata L. B. Smith in Lilloa, 14: 95, fig. 7, 8 (1948). Flower- ing 1-1.5 m. high. Leaf-blades 7 dm. long, covered beneath with ap- pressed, white scales, the spines antrorse, 12 mm. long, brown. In- florescence compound, minutely white-flocculose. Primary bracts ovate, acuminate, about a third as long as the axillary branches, serrulate. Racemes lax, to 14 cm. long, the naked sterile base to 35 mm. long. Floral bracts acuminate from a broadly ovate base, exceeding the pedicels, subchartaceous when dry, very minutely serrulate. Flowers divergent, not secund. Pedicels slender, 10-15 mm. long, 3-angled at apex. Sepals ovate-lanceolate, acuminate, 25-32 mm. long, strongly crested at base, nerved when dry. Petals 4 cm. long, green. COCHA- BAMBA: AYOPAYA: Sailapata, 3000 m., Cárdenas 3311 (GH; type). 8. P. laxa L. B. Smith in Phytologia, 6: 195, t. 1, fig. 7, 8 (1958). Flowering 8 dm. high. Leaf-blades 27 cm. long, caudate-acuminate with a long entire apex, densely tomentose with coarse white scales especially beneath, the spines antrorse, slender, 5 mm. long, brown. Scape 4 mm. in diameter, soon glabrous. Scape-bracts ovate, thin, the lower with small foliaceous blades, the upper merely acuminate, entire, much shorter than the internodes. Inflorescence laxly bipin- nate, soon glabrous except for the white-puberulent pedicels. Primary bracts broadly ovate, apiculate, entire, to 25 mm. long, thin, mostly about half as long as the sterile bases of the branches. Branches spreading, straight, very slender, lax, the lateral to 23 cm. long. Floral bracts broadly ovate, apiculate, to 13 mm. long, slightly ex- ceeding the slender pedicels, entire, membranaceous. Flowers not secund. Sepals lance-oblong, apiculate, 17 mm. long, thin, sulcate, slightly carinate at base. Petals 3 cm. long, violet. SANTA CRUZ: VALLE GRANDE: Pulquina, 1500 m., Cárdenas 5092 (US; type). 9. P. sanctae-crucis (Baker) L. B. Smith in Proc. Amer. Acad. 70: 154 (1935). Piteairnia robusta Rusby in Bull. N. Y. Bot. Gard. 6: 488 (1910). Densely aggregated, flowering 1-2 m. high. Leaf-blades to 35 mm. wide, glabrous above, covered beneath with subspreading pale-cinereous scales, the spines mostly antrorse, 10 mm, long, broad, black. Scape stout, white-tomentose-lepidote, becoming glabrous. Scape-bracts erect, imbricate, subfoliaceous, Inflorescence amply bi- pinnate, lax, dark red except the petals, white-tomentose-lepidote, soon becoming glabrous. Primary bracts ovate, acuminate to apiculate, from about half as long as the naked sterile bases of the lower branches to slightly exceeding those of the upper ones, serrulate. Branches divergent, 14-30 cm. long, subdense. Floral bracts ovate, 44 Rhodora [Vol. 71 apiculate, 15-22 mm. long, thin-coriaceous, often obscurely dentic- ulate. Flowers suberect, more or less secund. Pedicels 5-10 mm. long, rather slender. Sepals lanceolate, acute, 20-28 mm. long, exceed- ing the floral bracts, sulcate when dry, the posterior ones obscurely carinate. Petals 4 cm. long, dark violet, BOLIVIA: Castelnau s. n. P; type); LA Paz: MURILLO: Songo, Bang 892 (B, F, G, GH, M, MO, NY, US); cAUPOLICÁN: Apolo, R. S. Williams 2655 (NY; type of Pitcairnia robusta Rusby). 10. P. Fiebrigii Mez in Fedde, Rep. Spec. Nov. 3: 9 (1906). Leaves unknown. Scape-bracts large, ovate, glabrous, serrate toward apex, the blades narrow, armed with stout, uncinate, castaneous, 4 mm. long spines, densely villous-lepidote beneath. Inflorescence bi- pinnate, glabrous. Lowest primary bracts like the scape-bracts, the others elliptie, entire, barely exceeding the short sterile bases of the branches. Branches to 23 cm. long, lax at base, dense toward apex. Floral bracts very broadly elliptic, 25 mm. long, nearly or quite equaling the sepals, inflated, entire. Flowers suberect. Pedicels to 15 mm. long, stout, thickened upward. Sepals ovate-elliptic, apiculate, 17 mm. long, subrigid. Petals 25 mm. long. Bolivia: Fiebrig 3210 (B; type); 3211. 11. P. Meziana Wittm. in Meded. Rijks Herbar. 29: 85 (1916). Pitcairnia grandiflora of Catalogue. Flowering to 2.4 m. high. Leaves 45-85 em. long, the blades 35 mm. wide, soon glabrous above, covered beneath with appressed, cinereous scales, the spines mostly antrorse, 10 mm. long. Inflorescence simple and pyramidal-cylindric or few-branched, finely white-tomentellous, all but the pedicels be- coming glabrous, the axis stout. Floral bracts lanceolate, acuminate, pungent, the lowest 9 cm. long, ample, exceeding the sepals, sub- spinulose, submembranaceous, strongly nerved, the upper much re- duced. Pedicels rather slender, to 15 mm. long. Sepals lanceolate, apiculate, 30-40 mm. long. LA PAZ: MURILLO: Miraflores to Obrajillo, 3600 m., Buchtien 8531 (US); La Paz, 3600 m., Foster 2560 (US); LoAIZA: Teneria, Quimzacruz Cordillera, 3000 m., Herzog 2487 (L; type). 12. P. secunda L. B. Smith in Phytologia 7: 424, t. 2, fig. 11, 12 (1961). Flowering 1-2 m. high. Leaf-blades 35 mm. wide, glabrous above, covered beneath with coarse, subappressed, whitish scales, the spines antrorse, flat, 10 mm. long. Scape 2 cm. in diameter, densely and finely pale-floccose. Scape-bracts densely imbricate, the lower foliaceous, the upper acuminate from a broadly ovate base, serrulate, red. Inflorescence laxly bipinnate, at first finely white-floccose. Pri- mary bracts like the upper scape-bracts, much shorter than the long sterile bases of the branches. Branches divergent, 15-30 cm. long, subdensely secund-flowered with the slender rhachis covered toward apex. Floral bracts ovate, acuminate, 20-25 mm. long, about equaling 1969] Bromelliaceae — Smith 45 the middle of the sepals, rather thin, red, obscurely serrulate, soon glabrous. Pedicels slender, 5-10 mm. long. Sepals linear-lanceolate, apiculate, 28 mm. long, ecarinate, soon glabrous. Petals dark violet. COCHABAMBA: Yungas, 3000 m., M. B. Foster 2550 (US); CHAPARE: Aguirre to Incachaca, 2000 m., Cárdenas 4381 (US; type). 13. P. stenothyrsa (Baker) Mez in DC. Monog. Phan. 9: 481 (1896). Piteairnia Bangii Baker in Mem. Torr. Bot. Club, 6: 124 (1896). Plant to 1.5 m, high. Leaf-blades 40 mm. wide, densely white-lepidote beneath, the spines antrorse, nearly straight, 6 mm. long. Inflores- cence amply bipinnate. Primary bracts very broadly ovate, laciniate- spinulose, the lower short-acuminate, the upper broadly acute. Branches to 20 cm. long, bearing 10-20 more or less secund flowers, tomentose-lepidote or glabrate. Floral bracts ovate, acute, 20 mm. long, exceeded by the sepals, entire or minutely serrulate, inflated, red, tomentellous toward base. Pedicels 10 mm. long, rather stout, ferruginous-tomentose. Sepals narrowly elliptic-triangular, rounded and mucronulate, 33 mm. long, 9 mm. wide, sparsely lepidote. Petals 5 em. long, violet. LA PAZ: LARECAJA: Sorata, 2650-3300 m., Mandon 1189 (K; type); Bang 1631 (BM; type of Pitcairnia Bangii Baker; GH, NY, US); 2700 m., Cardenas 1217 (GH). COCHABAMBA: TOTORA: Incallacta to Totora, 3800 m., Cárdenas 2373 (GH); CAPINOTA: Pan- duro, Río Caine, Cárdenas 3444 (GH); CERCADO: Taquina, 12 km. north of Cochabamba, 2700 m., Cárdenas & Cutler 1408 (GH). 14. P. glareosa L. B. Smith in Lilloa, 14: 96 (1948). Plant to 1 m. high. Leaf-blades 5 cm. wide, densely white-lepidote beneath, the spines antrorse, curved, brown. Inflorescence bipinnate, minutely white-flocculose. Primary bracts oblong-ovate, acuminate, to 7 cm. long, a half to a third as long as the axillary branches, subchartaceous when dry, minutely serrulate. Racemes subsessile, suberect to diver- gent, the rhachis 14-15 em. long, sulcate. Floral bracts lanceolate, acuminate, slightly exceeding the pedicels, minutely erose, mem- branaceous. Flowers divergent. Pedicels slender, to 17 mm. long. Sepals narrowly lanceolate, acuminate, 26 mm. long, slightly carinate at base, subcoriaceous. Petals 4 em. long. LA PAZ: PACAJES: Calacoto, 3600 m., Cardenas 3583 (GH; type). 15. P. olivacea Wittm. in Meded. Rijks Herbar. 29: 83 (1916). To 1.5 m. high. Leaf-blades 3-4 em. wide, pale-lepidote, becoming glab- rous, the spines 5 mm. long, dark, slender. Scape elongate, glabrous. Seape-bracts broadly ovate, soon disintegrating. Inflorescence amply paniculate, lax, olivaceous-glaucous, covered with fine white stellate trichomes when young. Primary bracts suborbicular, apiculate, 4 cm. long, exceeding the short, sterile, naked bases of the branches, serrate, subchartaceous, soon disintegrating. Branches divergent, simple, laxly few-flowered, 10-12 cm. long. Floral bracts elliptic, apiculate, 3 em. long, equaling or exceeding the slender pedicels, serrulate. Sepals 46 Rhodora [Vol. 71 lanceolate, obtuse and mucronulate, 35-40 mm. long, thickened at base but not carinate. Petals 6-7 cm. long, pale lilac. SANTA CRUZ: CORDIL- LERA: Quebrada de Charagua, 950 m., Herzog 1216 (L; type); Lagunillas, 1050 m., Cárdenas 4786 (US); VALLE GRANDE: Valle Grande, 2000 m., Cardenas 5300 (US). 16. P. ferruginea (R. & P.) L. B. Smith, Phytologia 16: 461 (1968). Pitcairnia ferruginea R. & P. Fl. Peruv. 3: 36 (1802). Pitcairnia consimilis Baker in Journ. Bot. 19: 266 (1881). Very variable, flower- ing to nearly 4 m. high. Stem stout, woody, covered with the remains of dead leaves. Leaf-blades narrowly triangular, attenuate, 5 cm. wide, coarsely spinose-serrate, white-lepidote beneath. Scape stout, rather woody, ferruginous-stellate, then glabrous. Scape-bracts ovate, at- tenuate, at least the lowest serrate. Inflorescence usually paniculate, laxly pyramidal, covered with ferruginous stellate scales. Primary bracts like the upper scape-bracts, much shorter than the branches and usually not equaling their sterile bases. Branches secundly many- flowered. Floral bracts ovate, acute, entire, usually shorter than the pedicels. Pedicels slender, 1-6 em. long. Sepals oblong, obtuse, 12- 45 mm. long, densely ferruginous-stellate. Petals to 14 em. long though often much smaller, greenish to various shades of purple, usually with 2 vertical calli near base but varying to completely naked in a single flower. Stamens included. Ovary 2/3 superior. Ovules nar- rowly winged. Capsule globose, 25 mm. long. LA PAZ: MURILLO: La Paz, 3600 m., Balls 5881 (US); 3000 m., Bang 152 (BM, GH, US); 3800 m., Buchtien s. n. (BM, GH); 810 (US); 3600 m., Foster 2561 (US); Rose 18854 (US); 3000 m., Rusby 2845 (NY); 3450 m., Wil- liams 2355 (NY); Cerros de Obrajes, 3600 m., Cárdenas 2403 (GH); La Paz to Hacienda Chaco, 1200 m., Foster 2577 (US); LARECAJA: Sorata, 2650-3700 m., Mandon 1173 (BM, GH, K, S; type no. of Pitcairnia consimilis Baker); YUNGAS: 1800 m., Rusby 2847 (F, NY, US); Cordillera Real, Nequejahuira, 2400 m., Tate 634 (NY). WITH- OUT LOCALITY: Kuntze s. n. (NY). Ecuador, Peru. 17. P. riparia L. B. Smith in Phytologia, 7: 424, t. 2, fig. 9, 10 (1961). Flowering about 1.5 m. high. Leaf-blades 2 cm. wide, glabrous above, covered beneath with cinereous scales, the spines lax, slender, pale, uncinate, mostly antrorse, 4-5 mm. long. Scape 2 cm. in diameter, soon glabrous except the nodes. Scape-bracts deciduous, unknown. Inflorescence laxly bipinnate, narrowly pyra- midal, 7 dm. long, drying black, densely white-lanate. Primary bracts ovate, acute, ample, much exceeding the short, flat, sterile bases of the branches, thin-coriaceous, becoming glabrous, the margins crisped when dry. Lateral branches few, basal, 10-15 cm. long, lax, their floral bracts shorter then the sepals, the terminal branch elongate with floral bracts the size of the primary bracts and much exceeding the sepals. Pedicels rather stout, 20 mm. long. Flowers more or less 1969] Bromelliaceae — Smith 47 secund. Sepals lance-elliptic, apiculate, 42 mm. long, dull and rather rugose on drying. Petals over 6 cm. long. LA PAZ: MURILLO [?]: Road to Hacienda Chaco, eastern Cordillera, M. B. Foster 2578 (US; type). 18. P. alba L. B. Smith in Phytologia, 5: 49, t. 7, fig. 8-5 (1954). Plant 2 m. high. Leaves unknown. Inflorescence laxly bipinnate, per- sistently white-flocculose. Primary bracts suborbicular, apiculate, 3 em. long, densely serrate. Branches divergent, 15 cm. long, sub- dense with pendent-secund flowers. Floral bracts elliptic, acute, about equaling the pedicels, entire, subcoriaceous. Pedicels slender, to 25 mm. long. Sepals narrowly lanceolate, acute, 30 mm. long, ecarinate. Petals over 5 cm. long, white, bearing 2 long vertical calli near the base. TARIJA: O'CONNOR: Palos Blancos, 800 m., Cdrdenas 4952-a (US; type). 19. P. Pearcei (Baker) Mez in DC. Monog. Phan. 9: 480 (1896). Over 1 m. high. Leaf-blades 40 mm. wide, glabrous or sparsely lepi- dote throughout, the spines antrorse, 10 mm. long. Inflorescence amply bipinnate. Primary bracts narrowly triangular, densely spinulose- serrate, much shorter than the sterile naked bases of the branches. Branches ascending from a suberect to spreading base, over 3 dm. long, flavous-tomentose especially toward apex. Floral bracts broadly elliptic, apiculate, 20 mm. long, entire, membranaceous, glabrous or puberulent. Flowers spreading, more or less secund. Pedicels to 10 mm. long, rather slender, densely short-tomentose. Sepals narrowly lanceolate, apiculate, 32 mm. long, much exceeding the floral bracts, finely appressed-tomentose. Without locality: Bang s.n. (GH, NY, type); Milluguaya, 1300 m., Buchtien 4287 (US); LARECAJA: Tipuani, US). LA PAZ: NOR YUNGAS: Coroico, 900-1200 m., Pearce s. n. (K; 1400 m., Buchtien 7179 (US). COCHABAMBA: Puente Villa, 1400 m., Cárdenas 3689 (GH, US); Pasorapa, Cardenas 5924 (US). 20. P. penduliflora L. B. Smith in Contrib. Gray Herb. 98: 12, t. 4, fig. 3, 4 (1932). Flowering 6 dm. or higher. Leaf-blades (on the basis of the lower scape-bracts) glabrous above, sparsely and minutely lepidote beneath, the spines antrorse, castaneous. Scape rather slender, soon glabrous. Scape-bracts ovate, serrulate, submembranaceous, red, the lower with a long blade. Inflorescence few-branched, 25 cm. long, very finely white-tomentose. Primary bracts ovate, acute, much shorter than the branches, serrulate, submembranaceous, red, nerved. Branches laxly few-flowered, briefly sterile at apex, complanate at base, the lateral not over 9 em. long. Floral bracts like the primary but entire, slightly exceeding the pedicels. Flowers pendulous. Pedi- cels to 13 mm. long. Sepals narrowly lanceolate, acuminate, 35 mm. long. Petals 45 mm. long, greenish violet when dry. LA PAZ: LARECAJA: Tipuani, 1400 m., Buchtien 7182 (US; type). 48 Rhodora [Vol. 71 21. P. atra L. B. Smith in Contrib. U. S. Nat. Herb. 29: 536, fig. 90 (1954). Caulescent with a caudex like that of Yucca aloifolia, the flowering shoot over 2 m. high. Leaf-blades 25 mm. wide, soon gla- brous, the spines mostly antrorse, 5 mm. long, dark. Scape 2 cm. in diameter, tomentose-lepidote, soon glabrous. Scape-bracts densely im- bricate, subfoliaceous, the blades subdensely lepidote above with fine, irregular, white scales. Inflorescence densely but distinctly bipinnate, covered with fine, stellate, ferruginous scales, soon glabrous. Pri- mary bracts broadly ovate, acute, 8 em. long, even or faintly striate, black, lustrous. Branches spreading to slightly deflexed, ellipsoid, 9-10 cm. long exclusive of the short stout sterile base, 5-6 em. in diameter. Floral bracts imbricate, very broadly elliptic, obtuse, 4 cm. long, exceeding the sepals, sparsely and very minutely serrulate, sub- chartaceous, black, lustrous. Pedicels slenderly cylindric, 8 mm. long. Sepals narrowly elliptic, obtuse, 26 mm. long, carinate, lustrous. COCHABAMBA: Above Yungas, 3150 m., Foster 2551 (US; type); CARRASCO: Jatum Pino, 3000 m., Cárdenas 5285 (US). 22. P. dyckioides (Baker) Mez in DC. Monog. Phan. 9: 486 (1896). Stemless, 4-7 dm. high. Leaf-blades 12-20 mm. wide, glabrous above, covered beneath with pale, appressed scales, The spines antrorse, 3 mm. long, curved. Scape 15 mm. in diameter, furfuraceous at first. Scape-bracts imbricate, broadly ovate, acuminate to a linear blade, densely serrulate. Inflorescence bipinnate, subdensely cylindric, 2-4 dm. long, 5-8 cm. in diameter, the axes furfuraceous when young. Primary bracts like the scape-bracts but more coarsely serrate and the upper bladeless, 45-60 mm. long. Branches erect or divergent, slightly if at all exceeding the primary bracts, subdensely few-flowered. Floral bracts ovate-elliptic, apiculate, nearly or quite equaling the sepals, entire or serrulate, thin, nerved. Pedicels 5-7 mm. long, slender. Sepals ovate-elliptic, 12 mm. long, glabrous, nerved. Petals 20 mm. long, green. CHUQUISACA: TOMINA: Weddell 3739 (P; type); Toma- bamba, Weddell 3881 (P). TARIJA: CERCADO: Tarija, West 8323 (UC). Argentina, 23. P. glabrescens L. B. Smith in Contrib. U. S. Nat. Herb. 29: 537, fig. 91 (1954). Flowering 6-8 dm. high. Leaf-blades 2-3 cm. wide, at first covered on both sides with gray appressed scales, the spines slender, curved, 10 mm. long. Scape erect, shorter than the leaves, stout. Inflorescence bipinnate, densely cylindric, densely and finely white-lanate with the bracts and sepals soon becoming glabrous. Primary bracts persistent, elliptic, ample, acute, to 13 em. long, equaling the branches, opaque, serrate. Floral bracts elliptic, acute, about equaling the sepals, serrulate. Pedicels slender, 16 mm. long. Sepals lance-ovate, acuminate, much thickened centrally but not winged, 35 mm. long, coriaceous. Petals 55 mm. long. COCHABAMBA: TARATA: Llallaguani to Anzaldo, 3000 m., Cardenas 4383 (US); 4384 1969] Bromelliaceae — Smith 49 (US; type); CERCADO: Taquina, 10 km. east of Cochabamba, 2900 m., Eyerdam 25311 (UC, US); Tunari, 2850 m., Foster 2538 (US); 2539 (US); AYOPAYA: Parangani, 2560 m., Cárdenas 4811 (US). 24. P. tuberosa Mez in DC. Monog. Phan. 9: 483 (1896). Flowering plant less than 3 dm. high including the very short erect rhizome. Leaf-sheaths forming a globose bulb 3 cm. in diameter, the blades linear, flexuous, filiform-acuminate, 15-18 cm. long, 2-3.5 mm. wide, glabrous above, covered with pale, narrow, spreading scales beneath, the spines spreading or ascending, fine, 1-2 mm. long, Scape slender, white-flocculose, becoming glabrous, Scape-bracts ovate with long, filiform blades, pale-lepidote. Inflorescence simple or with a few short branches at base, cylindric, 7-10 cm. long, white-flocculose at first. Primary bracts ovate, acuminate, 2 cm. long, longer or shorter than the laxly few-flowered branches, minutely serrulate. Floral bracts ovate, acute, thin, red. Pedicels slender, 4-7 mm. long, longer or shorter than the floral bracts. Sepals lance-oblong, acute, 7-9 mm. long. Petals 14 mm. long, blue-violet. Bolivia: “Peru” Haenke (M; type); SANTA CRUZ: VALLE GRANDE: Cuesta de Los Monos, 1000 m., Cárdenas 4647 (US); El Fuerte to Samaipata, 1800 m., Cárdenas 5198 (US). Note: Haenke was near the two Cardenas localities and most spe- cies of Puya have very restricted ranges, so his material probably came from this same general area. 25. P. reducta L. B. Smith in Phytologia, 10: 472, t. 1, fig. 1, 2 (1964). Leaf-blades ca. 4 dm. long, glabrous above, subdensely tomentose beneath with white linear trichomes, the spines slender, pale, antrorse, 4 mm. long. Scape erect, ca. 3 mm. in diameter, finely white-stellate. Scape-bracts erect, ovate with a caudate apex, red, the upper ones about equaling the internodes. Inflorescence laxly bipinnate, sparsely white-stellate, becoming glabrous. Primary bracts like the upper scape-bracts but not caudate, nearly the same size and form as the floral bracts, exceeding the axis of the very short few- flowered lateral branches. Branches divergent to spreading with a very short naked sterile base, subdense but with the axis largely ex- posed. Floral bracts suberect, broadly ovate, acute and apiculate, 15 mm. long, ecarinate, broadly convex, red with a narrow scarious margin, subchartaceous. Pedicels about 1 mm. long but slender and distinct. Sepals linear-lanceolate, narrowly obtuse and apiculate, 27 mm. long. Petals ca. 4 cm. long, naked. LA PAZ: CAUPOLICÁN: Altunkama, 2600 m., Cardenas 5687 (US; type). 26. P. mollis Baker ex Mez in DC. Monog. Phan. 9: 488 (1896). Leaf-blades 4 cm. wide, glabrous above, covered beneath with fine, appressed, cinereous scales, the spines antrorse, uncinate, pale brown, 5 mm. long. Scape unknown. Inflorescence compound, dense, stout- cylindric or ellipsoid, densely white-lanate. Primary bracts ovate, 50 Rhodora [Vol. 71 acuminate, some nearly equaling the branches, others about half as long, thick, coriaceous, slightly serrulate toward apex. Branches sub- erect, to 10 cm. long with a tuft of sterile bracts at apex. Floral bracts straight, broadly ovate, acute, 35 mm. long, much exceeded by the sepals, entire, thick, coriaceous. Pedicels stout, 15 mm. long. Sepals subacute, 28 mm. long, very thick. Petals 6 cm. long, greenish violet. Stamens included. LA PAZ: LARECAJA: Sorata, 2700-3000 m., Mandon 1190 (BR, GH, P); Bang 1629 (K; type; BM, GH, US). 27. P. mirabilis (Mez) L. B. Smith in Phytologia 16: 461 (1968). Piteairnia mirabilis Mez in Fedde, Repert. Spec. Nov. 3: 6 (1906). Flowering to 1.5 m. high. Leaves persistent, the sheaths broadly ovate, very thick, forming an epigaeous bulb, the blades linear, attenu- ate, not contracted at base, 10-15 mm. wide, coarsely spinose-serrate, soon glabrous. Scape stout, glabrous. Scape-bracts erect, exceeding the internodes but exposing the scape, broadly ovate, the lower with long spinose blades, the upper with short ones. Inflorescence laxly racemose, cylindric, 3 dm. long, the axis stout, straight. Floral bracts suberect to spreading, broadly ovate-elliptic, attenuate, 30 mm. long, exceeding the pedicels, laciniate-spinose. Pedicels suberect at anthesis, becoming decurved, 12 mm. long. Sepals subtriangular, slightly asym- metric and obscurely mucronulate at apex, nearly 6 cm. long, ecari- nate, thick, coriaceous. Petals 95 mm. long, yellow-green, naked. Stamens included. Ovary 3/4 superior. Ovules alate. COCHABAMBA: CERCADO: Cerro San Pedro, Cochabamba, 2560 m., Cárdenas 3835 (GH); Adolfo 177 (US); QUILLACOLLO: Cerro del Calvario, 2560 m., Cárdenas 3578 (GH). TARIJA: ARCE: Bermejo, 1400 m., Fiebrig 2320 (B; type); CONNOR: Chuquiaca, Fiebrig 2420a (B). Argentina. 28. P. Kuntzeana Mez in DC. Monog. Phan. 9: 490 (1896). Flower- ing over 1 m. high. Leaves unknown. Scape stout, densely white- tomentose when young, becoming glabrous. Scape-bracts (upper) re- flexed, lance-triangular, ca. 15 cm. long, sparsely laciniate-spinulose. Inflorescence simple, densely cylindric, subacute, 30 em. long, ca. 7 em. in diameter, densely tomentulose when young, becoming glabrous. Floral bracts narrowly lanceolate, equaling or exceeding the sepals, entire, submembranaceous, the lowest reflexed, the others erect or suberect. Flowers erect or suberect. Pedicels slender, to 12 mm. long. Sepals subelliptic, triangular-acute, 25 mm. long, 8 mm. wide. Petals obtuse, 46 mm. long, brownish. Stamens exserted. COCHABAMBA: CERCADO: Tunari, 1300 m., Kuntze s. n. (NY; type). 29. P. minima L. B. Smith in Phytologia 7: 423, t. 2, fig. 5, 6 (1961). Flowering only 10 em. high. Leaves forming a subbulbous rosette, to 12 em. long, the blades 6 mm. wide, covered on both sides with sub- orbicular, slightly spreading, cinereous scales, becoming glabrous above, the spines retrorse, 1.5 mm. long, yellow-brown. Scape slender, stellate-lepidote, mostly hidden by the leaves. Scape-bracts imbricate, 1969] Bromelliaceae — Smith 51 the upper ovate, acute, thin. Inflorescence simple, 4 cm. long, sub- dense, few-flowered, sparsely stellate-lepidote. Floral bracts like the upper scape-bracts, to 13 mm, long, much exceeding the pedicels, bright red. Pedicels slender, 8 mm. long. Sepals oblong, broadly acute and apiculate, 11 mm. long. Petals 15 mm. long. TARIJA: MEN- DEZ: Sama, 2500 m., West 8345 (UC; type). 30. P. paupera Mez in Fedde, Repert. Spec. Nov. 3: 14 (1906). Stem- less, flowering 22 cm. high. Leaf-blades 7 mm. wide, velutinous- lepidote especially beneath, the spines dense, slender, 3 mm. long. Scape slightly exceeding the leaves, densely velutinous. Scape-bracts erect, exceeding the internodes, sublanceolate, bladeless, subglabrous, minutely spinose. Inflorescence subsimple, rather densely cylindric, slightly or not interrupted at base, 10 em. long, 15 mm. in diameter. Branches basal with one normal and one abortive flower, shorter than the bracts. Floral bracts ovate, acuminate, 20 mm. long, denticulate, glabrous, nerved. Flowers erect. Pedicels 6 mm. long. Sepals oblong, obtuse, 11 mm. long, minutely araneose. Petals 15 mm. long, blue or violet. TARIJA: CERCADO: Tucumilla, 2600 m., Fiebrig 2005 (B; type). 31. P. valida L. B. Smith in Phytologia, 10: 474, t. 1, fig. 3, 4 (1964). Flowering 3 m. high. Leaf-blades 4 cm. wide, soon wholly glabrous, laxly serrate with brown stiffly antrorse brown spines 3 mm. long. Scape erect, over 4 cm. in diameter (! photo). Scape-bracts early deciduous, unknown, Inflorescence subdensely bipinnate, over 1 m. long, except for the petals and genitalia covered with a fine white dense appressed tomentum. Primary bracts small and incon- spicuous (! photo). Branches subspreading, strobilate, nearly 3 dm. long, 7 cm. in diameter. Floral bracts elliptic-oblong, acute and apicu- late, to 55 mm. long, distinctly exceeded by the sepals, coriaceous, rigid, very minutely serrate toward apex. Flowers suberect. Pedicels subeylindric, 2 cm. long, 5 mm. in diameter. Sepals lance-elliptic, rounded and apiculate, 35 mm. long, coriaceous when dry, ecarinate. Petals naked, twisted together after anthesis. CHUQUISACA: CAMPERO: Río Chico, 2500 m., Cárdenas 6012 (US; type). 32. P. lasiopoda L. B. Smith in Proc. Amer. Acad. 70: 153 (1935). Leaves and scape unknown. Inflorescence laxly bipinnate, tomentose- lepidote. Primary bracts ovate, exceeding the long, slender, sterile bases of the branches, minutely serrate, submembranaceous, red. Racemes secund, dense, ellipsoid, 10 em. long, 5 cm. in diameter. Floral bracts like the primary bracts but smaller and entire, about equaling the sepals. Flowers erect. Pedicels slenderly obconic, 7 mm. long. Sepals narrowly lanceolate, uncinate at apex, 45 mm. long, ecarinate. Petals acute, 6 cm. long, dark violet. Stamens about equal- ing the petals or slightly exserted. BENI: YACUMA: Reyes, 500 m., Rusby 2232 (NY; type). PUNO: CARABAYA: Ollachea to Quillabamba, Vargas 17515 (US). 52 Rhodora [Vol. 71 83. P. Ugentiana L. B. Smith in Phytologia, 13: 147, t. 7, fig. 1, 2 (1966). Leaf-blades 6 em. wide, obscurely lepidote between the nerves beneath, covered with appressed cinereous scales above, the spines antrorse, 3 mm. long. Scape 5 cm. in diameter, soon glabrous. Seape-bracts persistent, exceeding the internodes, the upper broadly triangular with short serrate blades, partly exposing the scape. In- florescence lax, bipinnate, over 1 m. long, finely and densely whitish- stellate including the petals. Primary bracts much shorter than the axillary branches. Spikes subspreading, strobilate, subsessile, the central ones to 17 cm. long. Floral bracts lanceolate, acute, nearly 5 em. long and about equaling the sepals, Pedicels slenderly sub- cylindric, 15 mm. long. Sepals lanceolate, narrowly obtuse, 30 mm. long. Petals linear. Capsules subglobose, 2 cm. long. CHUQUISACA: OROPEZA: Sucre to Aiqule, 1950 m., Ugent 4998 (US; type; WIS). 34. P. ctenorhyncha L. B. Smith in Phytologia, 5: 49, t. 7, fig. 6-9 (1954). Plant 1.2 m. high. Leaf-blades 35 mm. wide, glabrous above, covered beneath with appressed, gray scales, the spines lax, red-brown, 4 mm. long. Scape stout, white-lanate. Scape-bracts erect and exceed- ing the internodes but not wholly covering the scape, broadly ovate with small, subfoliaceous blades, very dark brown, subcoriaceous, per- sistent, serrate, floccose. Inflorescence densely bipinnate, fusiform, 3 dm. long, 8 em. in diameter, densely white-lanate. Primary bracts like the scape-bracts but their narrowly triangular apices notably pectinate and not at all foliaceous, 6 em. long, equaling or exceeding the globose, few-flowered spikes. Floral bracts elliptic, apiculate, 35 mm. long, exceeding the sepals. Pedicels rather slender, 7 mm. long. Sepals elliptic, broadly acute, 20 mm. long, coriaceous. Petals 4 cm. long, light yellow-green. LA PAZ: LARECAJA: Anilaya to Larecaja, 3500 m., Cardenas 4867 (US; type). 35. P. Fosteriana L. B. Smith in Journ. Wash. Acad. Sci. 40: 216, fir. 1 (1950). Plant 1 m. high. Leaf-blades 5 cm. wide, glabrous above, minutely pale-lepidote between the nerves beneath, the spines dark, flat, straight or uncinate, 9 mm. long. Scape very short, con- cealed by the leaves. Inflorescence very densely bipinnate, thick- cylindrie, rounded at apex, densely brown-lanate. Primary bracts membranaceous, pale green, the suborbicular base equaling the sepals, the long, narrowly triangular apex reflexed, coarsely serrate. Branches much abbreviated, few-flowered. Floral bracts broadly elliptie, long- caudate, exceeding the sepals, subentire, membranaceous, lightly green, drying stramineous. Pedicels rather slender, 1 cm. long. Sepals linear-lanceolate, acute, 45 mm. long, ecarinate, minutely and sparsely serrate, submembranaceous. Petals broadly elliptic, ultramarine, dark purple at base. L4 PAZ: MURILLO: La Rinconada, about 75 km. east of La Paz, 4400 m., M. B. Foster 2513 (US; type). 36. P. tristis L. B. Smith in Phytologia, 10: 475, t. 1, fig. 5-7 (1964). 1969] Bromelliaceae — Smith 53 Flowering 6 dm. high. Leaf-blades 15 mm. wide, glabrous above, cov- ered beneath with appressed white scales, laxly serrate with dark uncinate spines 4 mm. long. Scape exceeding the leaves, coarsely brown-tomentose. Scape-bracts subfoliaceous, strict, densely imbricate, brown-tomentose toward base. Inflorescence densely ovoid, broadly rounded, 10 cm. long, 8 cm. in diameter, obscurely bipinnate, densely and coarsely brown-tomentose except the petals. Primary bracts at- tenuate from a suborbicular base, to 7 cm. long, membranaceous, finely nerved, entire, the apical half reflexed at anthesis. Branches nearly aborted, 2-flowered. Floral bracts like the primary bracts but short-attenuate and only 35 mm. long. Pedicels short but slender and distinct. Sepals oblong-elliptic, rounded and apiculate, 18 mm. long, thin. Petals over 3 cm. long, blue. CoCHABAMBA: CHAPARE: km. 89, Cochabamba to Chapare, 3900 m., Cardenas 6067 (US; type). 37. P. leptostachya L. B. Smith in Lilloa, 14: 97, fig. 11, 12 (1948). Flowering 4-5 dm. high. Leaf-blades 12 mm. wide, glabrous above, covered beneath with subappressed, cinereous, black-centered scales, the spines lax, dark, uncinate, 4 mm. long. Scape short. Scape-bracts very densely imbricate, shortly foliaceous-laminate from a broadly elliptic, subchartaceous, lepidote sheath. Inflorescence slenderly cy- lindric, depauperate-paniculate, 3 cm. in diameter. Primary bracts acuminate from a broadly ovate base, 5 cm. long, entire, membrana- ceous, sparsely brown-lanate at base, strongly reflexed above the middle. Branches nearly aborted, 2-flowered, wholly hidden. Floral bracts elliptic, acuminate, 25-30 mm. long, much exceeding the sepals, entire, membranaceous, brown-lanate. Pedicels slenderly obconic, 4 mm. long. Sepals elliptic, broadly acute or obtuse, 17 mm. long, thin, ecarinate, strongly nerved, lanate. Petals 3 cm. long, blue. CocHA- BAMBA: CHAPARE: Cuesta de Llanta, Aduana near Colomi, 3400 m., Cardenas 3581 (GH, type). Peru. 38. P. Cardenasii L. B. Smith in Lilloa, 14: 94, fig. 6 (1948). Plant 1.5 m. high. Leaf-blades 3 cm. wide, nearly glabrous, the spines very lax, brown, straight or curved, 8 mm. long. Scape very short, hidden by the leaves. Inflorescence densely cylindric, appearing simple but actually bipinnate, 25 em. long, 12 em. in diameter, pale-lanate, gla- brous in fruit. Primary bracts ample, about equaling the branches, obscurely serrulate, membranaceous, reflexed, fragile. Branches dense- ly few-flowered. Floral bracts like the primary bracts, exceeding the sepals. Pedicels slenderly obconic, 10 mm. long. Sepals broadly lance- olate, acute, 25 mm. long, ecarinate, thin-coriaceous. Petals 4 cm. long, blue. COCHABAMBA: CERCADO: Tunari, 3800 m., Cardenas 4082 (US; type); 3900 m., M. B. Foster 2540 (US). 39. P. Herzogii Wittm. in Meded. Rijks Herbar. 29: 86 (1916). Flowering 2 m. high. Leaf-blades 2 cm. wide, glabrous above, densely cinereous-lepidote beneath, minutely spinose-serrate. Scape erect or 54 Rhodora [Vol. 71 slightly curved, elongate, stout, much exceeding the leaves. Scape- bracts refracted, soon deciduous. Inflorescence densely cylindric or ellipsoid, compound, 2-3 dm. long, covered with a yellowish-white wool. Primary bracts large, broadly ovate or obovate with a deltoid apiculus, fuscous when dry, suberose, slightly exceeding the few- flowered spikes, recurved toward apex; floral bracts like the scape- bracts, equaling or slightly exceeding the sepals. Pedicels 10 mm. long. Sepals ovate-oblong, subacute, 30 mm. long, covered on both sides with a dense easily detached wool. Petals 5 cm. long, flavous. Stamens included. COCHABAMBA: CHAPARE: Incacorral, Herzog 2269 (L; type); Colomi, 3300 m., Cardenas 2251 (GH); AYOPAYA: Saila- pata, 3700 m., Cárdenas 3306 (US); CERCADO: Tunari, Kuntze s. n. (NY); 3600 m., M. B. Foster 2541 (US); Cárdenas 6276 (US); TOTORA: Incallajta to Carrasco, Cárdenas 6275 (US). 40. P. tunarensis Mez in DC. Monog. Phan. 9: 498 (1896). Flower- ing 4-7 dm. high. Leaf-blades 3 dm. long, 15 mm. wide, woody, very stiff; densely cinereous-lepidote beneath, the spines dense, subspread- ing, 5 mm. long. Scape short, stout. Scape-bracts erect, imbricate, broadly ovate, acuminate. Inflorescence simple, densely cylindric, 20 cm. long, 3 cm. in diameter, many-flowered. Floral bracts elliptic or ovate-elliptic, apiculate, 20 mm. long, about equaling the petals but decurved at apex, entire, scarious, pulverulent-lepidote, flavous. Pedi- cels very short. Sepals ovate-elliptic, acuminate, 14 mm. long, carinate, coriaceous, flavous, subglabrous or sparsely lepidote. Petals 17 mm. long, dull red-brown. Stamens exserted. COCHABAMBA: CERCADO: Tunari, 1400 m., Kuntze s. n. (NY; type); Toralapa to Cochabamba 3700 m., Cárdenas 3580 (US); TOTORA: Quirusillani, 3800 m., Cár- denas 2372 (GH); TARATA: Anzaldo, 3600 m., Cárdenas, Cutler & Gandarillas 7639 (US); Cocha to Vilavila, 2850 m., M. B. Foster 2555 (US). 41. P. nana Wittm. in Mededeel. Rijks Herb. 29: 85. (1916). Pit- cairnia nana (Wittm.) L. B. Smith in Contrib. Gray Herb. 89: 7 (1930). Leaves many in a spreading rosette, the blades linear-tri- angular, over 6 dm. long, 25 mm. wide, glabrous above, densely cinereous-lepidote beneath, laxly serrate with dark antrorse spines 5 mm. long; scape very short and hidden; inflorescence semiglobose, dense, 15 cm. in diameter. Outer bracts ovate or ovate-oblong with a narrowly deltoid apex, entire, 45-50 mm. long, 15-20 mm. wide, sub- coriaceous, lustrous, roseate, drying to yellow-brown with a very dark base, subfurfuraceous to glabrous. Floral bracts like the outer bracts but narrower and more furfuraceous, about equaling the sepals. Pedicels broadly obconic, 15 mm. long, compressed. Sepals narrowly lanceolate, attenuate to a pungent apex, 35-40 mm. long, 10 mm. wide, carinate. Petals subacute, slightly exceeding the sepals, naked, blue- green. Stamens barely exserted. Ovary !4 inferior. Ovules winged. 1969] Bromelliaceae — Smith 55 SANTA CRUZ: VALLE GRANDE: Samaipata, 2100 m., Herzog 1856-a (L; type); same, alt. 1800 m., Cárdenas 5533 (US). 42. P. humilis Mez in DC. Monog. Phan. 9: 498 (1896). P. Wer- dermannii Harms of Catalogue. Cespitose with a rhizome 15 cm. in diameter, the flowering shoot 15-25 cm. high. Leaf-blades 7 mm. wide, cinereous-lepidote especially beneath, the spines lax, spreading, 3 mm. Scape very short, white-tomentellous, becoming glabrous, Scape- bracts densely imbricate, broadly ovate-elliptic with foliaceous blades. Inflorescense simple, strobilate, fusiform, 6-8 cm. long, 25 mm. in diameter, few-flowered, the axis densely white-tomentose. Floral bracts strict, broadly ovate, acuminate, 20-28 mm. long, exceeding the petals, entire, pungent, appressed-lepidote, becoming glabrous, nerved. Pedicels 2-3 mm. long. Sepals subtriangular, acutish, 12-13 mm. long, carinate, sparsely lepidote, becoming glabrous especially toward apex. Petals obtuse, 15-20 mm. long, about equaling the sta- mens, blue or violet. COCHABAMBA: CHAPARE: Sacaba, 3000 m., Kuntze s. n. (NY; type); Incachaca, 2800 m., Werdermann 2006 (B, type of P. Werdermannii Harms) ; TOTORA: Siberia, Comarapa to Pojo, Vogel 468 (US). Potosi: Porco [?]: Tagua Comufio, 4200 m., West 6370 (UC). CHUQUISACA: SUCRE: Guerralona, 3000 m., Cárdenas 4128 (US). 43. P. Brittoniana Baker, Handbk. Bromel. 124 (1889). Leaf-blades 15-18 mm. wide, glabrous above, sparsely white-lepidote beneath, the spines lax, spreading, uncinate, brown, 5 mm. long. Scape stout. Scape-bracts densely imbricate, the upper ovate with a foliaceous blade, lanate. Inflorescence simple, densely cylindric, over 15 cm. long, densely pale-ferruginous-lanate. Floral bracts ovate, ‘acuminate, 60 mm. long, 38 mm. wide, equaling or exceeding the flowers, entire, thin-coriaceous. Sepals lance-oblong, acute, 22 mm. long, coriaceous. Petals obtuse, 4 em. long, green. Stamens included. LA PAZ: LARECAJA: Ingenio del Oro between Sorata and Mapiri, 3000 m., Rusby 2849 (NY; type). 44. P. Weddelliana (Baker) Mez in DC. Monog. Phan. 9: 475 (1896). Caudex large, forking, stout, bearing rosettes of leaves at the apices of the branches. Leaf-blades 4 cm. wide, glabrous and glaucous above, pale-lepidote beneath, the spines straight, erect, slender, red, 5 mm. long. Scape 1.5 m. high, much exceeding the leaves. Inflorescence amply paniculate, very densely pyramidal, many-flowered. Branches many-flowered and white-tomentose toward base, sterile, bracteate, and glabrous toward apex. Floral bracts lanceolate, acute, 15 mm. long, exceeded by the sepals, entire. Flowers erect. Pedicels 10 mm. long. Sepals subobtuse, 18 mm. long, becoming glabrous. Petals obtuse, 35 mm. long, the blade bright blue. Stamens included. CHUQUISACA: CINTI: Culpina, Correll et al B-644 (US); TARIJA: 56 Rhodora [Vol. 71 PROVINCE [?]: Weddell 4001 (P; type); MENDEZ: San Lorenzo, 1800 m., Cárdenas 210 (GH). 45. P. Raimondii Harms in Notizbl. 10: 213 (1928). “Chuqui Kayara.” Plant columnar, 9.5 meters high, dying after the formation of fruit, the caudex simple, erect, 4 m. high, 6 dm. in diameter, sur- mounted by the dense leaf-rosette. Leaves very many, 1-1.25 m. long, the blades 9 cm. wide, the spines lax, stout, uncinate, dark brown, 10 mm. long. Scape very short and stout, hidden by the leaves. In- florescence densely bipinnate, slenderly cylindric, attenuate, 4.3 m. long, 6 dm. in diameter, white-lanate throughout. Primary bracts broadly ovate, apiculate, much shorter than the branches, entire, membranaceous, white. Branches spreading or reflexed, sterile toward apex. Floral bracts densely imbricate, ovate-lanceolate, acuminate, exceeding the sepals, membranaceous, subglabrescent. Pedicels cylin- dric, stout, 15 mm. long. Sepals lanceolate, acute, 4 cm. long. Petals 6-8 cm. long, greenish white, often purplish, the blade broad. La PAZ: INGARI: Comanche, 75 km. southwest of La Paz, Rose 18875 (NY, US) ; 4050 m., M. B. Foster 2566 (US) ; COCHABAMBA: CERCADO- CHAPARE: Huacanqui, 3000 m., Cárdenas 4380 (US) ; 2400 m., M. B. Foster 2546 (US); TARATA: Tarata, 3900 m., Miller (US); TOTORA: Totora, 3800 m., Cárdenas 2371 (GH); COCHABAMBA-SUCRE: Cuesta de Alalas, Furlong (GH). Peru. Fig. 1. Excluded Species: P. boliviensis Baker of Catalogue. Endemic to Chile. P. micrantha Mez of Catalogue. Endemic to Argentina. 1969] Bromelliaceae — Smith 57 a p- AX RESEO LELA : BORE Figure (1) A: Puya Raimondii (after Cardenas 4380), habit from photo; B: section of leaf X 1; C: branch of inflorescence X 1/2; D: flower TAXONOMY OF PERITYLE SECTION PAPPOTHRIX (COMPOSITAE — PERITYLANAE)' A. MICHAEL POWELL This small section of the genus Perityle is comprised of five species which are distributed throughout Trans-Pecos Texas and in the Organ Mountains of southern New Mexico. Asa Gray (1852) proposed the section to include a species of Laphamia (now synonymous with Perityle), L. rupestris, which exhibited a unique pappus of about 20 bristles. Rydberg (1914) elevated Gray's section to generic rank, apparently relying on the multi-bristled pappus as the pri- mary distinguishing feature. Other species of Perityle have a pappus of 0-2(3) bristles (often 4-6 in P. staurophylla) with or without squamellae. Everly (1947), in her revision of Perityle, retained Pappothrix as a separate genus because the group “can be clearly defined on the basis of a pappus of 15-30 awns." Pappothrix indeed is characterized by its pappus, but as Macbride (1918), Shinners (1959), and others have concluded, the distinction does not warrant generic status apart from Perityle where pappus variability is a hallmark. Considering most other morphological fea- tures Pappothrix and Perityle are quite similar, and this is substantiated by chromatographic evidence (Powell, 1968a). Sectional status is maintained for Pappothrix be- cause the species comprise a natural phyletic unit within Perityle. Other information concerning the revised tax- onomy of Perityle is presented elsewhere (Powell, 1968a). ECOLOGICAL CONSIDERATIONS Species of section Pappothrix, and in fact most all taxa of the genus Perityle, exist in rather bizarre habitats, oc- curring only in soil-filled crevices of otherwise solid rock. ‘Supported by NSF Grants GB-2400 and GB-5046. I am grateful to the curators of the various herbaria who loaned specimens utilized in this work. I also acknowledge Professor B. L. Turner who suggested the study of Perityle and contributed valuable counsel during the course of this investigation. 58 1969] Perityle — Powell 59 Plants may be distributed in perpendicular bluffs, large boulders, or any rock mass of considerable size and stabil- ity. Plants are never found growing in soil, even among smaller rocks at the base of cliffs. This would appear to represent very stringent conditions for the survival of these plants, especially in the arid southwestern United States. Examination of most populations reveals that a majority of the plants are relatively old, having developed strong, thick woody rootstocks. Dispersal of seeds seems to be effected by fortuitous deposition of achenes by wind and washing from summer thundershowers. No dispersal vec- tors such as birds or ants have been observed carrying achenes, but it is likely that such animals do spread seeds accidentally. Plant genera with which the species of Pappothrix are commonly associated in their rock habitats are Heterotheca spp. and Hedyotis spp. Less often associated are Selaginella spp., Galium spp., Crotalus lepidus, and Cirsium turneri. Greenhouse tests of all the section Pappothrix species have revealed that they are almost completely self-incom- patable. This has facilitated hybridization experiments to the point of determining percent of seed-set, but achene germination is very poor. Manipulation of conditions has shown that dormancy at normal refrigerator temperatures increases the percent of germination about 30%. At present it is difficult to determine if poor germination is due to natural or induced phenomena. The absence of self-compatability brings out an unre- solved question about pollen transfer under natural condi- tions. Plants in their natural habitat normally produce heads with a very high percent of well-developed achenes. Ants and various other small insects (such as beeflies) have been observed visiting flowers and crawling over plants, but there is no evidence that these animals effect pollination. There is no apparent regularity with which populations of section Pappothrir species are distributed in the hun- dreds of likely habitats available throughout the total dis- tribution area. The most successful populations are usually located on north and east-facing bluffs or in narrow canyons 60 Rhodora [Vol. 71 where the plants receive only partial sun, but in some areas which appear to offer ideal habitats, plants of these species have not been found. The surface geology of the Trans-Pecos region of Texas and adjacent areas provides important considerations for understanding the distribution of section Pappothrix species (Sellards, Adkins, and Plum- mer, 1958). In general, this region has been subjected to violent geologic events, including intrusive and extrusive Tertiary volcanic activity, cataclismic mountain-forming, faulting, and extensive erosion. Any wide-spread ancestral species of section Pappothrix (possibly P. cernua) would have become increasingly more isolated with progressive erosion especially because of the restricted habitats. Occu- pation of new habitats by expanding populations appears to be controlled by fortuitous dispersal, regardless of the desirability of a location. There is some indication that a positive correlation exists between geological events and biological evolution in Perityle, but it is not advisable to speculate extensively at this time, CHROMOSOMAL STUDIES Meiotic chromosome numbers have been compiled with the utilization of methods similar to those outlined by Turner and Johnston (1961). Voucher specimens are de- posited in the Sul Ross State College Herbarium. Chromosomes of all section Pappothrix species are hetero- morphic in size, with 2-3 bivalents being smaller and 2-3 bivalents being larger than the others. One or 2 of the smaller bivalents often divide precociously at anaphase I, and thus could lead an observer to believe that unpaired univalents, fragments, or extra chromosomes are occurring. However, careful observation has demonstrated that meio- sis is regular in diploid (n = 17) and polyploid (n = 34, 68, 85) plants. The chromosome numbers presented in Table I are the first reports for each species of section Pappothrix, All Species so far examined exhibit diploid chromosome num- bers with the exception of Perityle cinerea. Diploid and polyploid numbers have been recorded for P. rupestris. 1969] Perityle — Powell 61 Chromosome numbers have not been particularly helpful in determining the interspecific taxonomic relationships within section Pappothrix or the affinity of it with sects. Laphamia and Perityle. The basic number of Pappothrix seems clearly established as + — 17. Aneuploid numbers above and below n = 17 exist in several species of Perityle (Powell, 1968b), and it is possible that the ancestral basie number for the genus Perityle is n — 18 or 19, but the preponderance of species exhibit chromosome numbers of n = 17 or polyploids thereof. Populations of P. rupestris seem to possess a remarkable ability to establish polyploidy. Populations with diploid, tetraploid (n = 34), octaploid (m = 68), and decaploid (n = 85) chromosome numbers have been discovered (Table I). The explanation for this high polyploid toler- ance is not clear at this time. Perhaps a partial solution is correlated with the stringent ecological conditions under which the species live, and their perennial habit. Also, H. Table I. Species of Perityle section Pappothrix Examined for Chromo- some Number. Location and Voucher Species n Number Perityle cernua 17 + 1 frag.* NEW MEXICO: Doña Ana Co. Dripping Springs Can- yon. Sikes & Babcock 328. P. cinerea ca. 68 TEXAS: Pecos Co. 6 mi. e. of Bakersfield. MeKenzie 128. ca. 68 TEXAS: Pecos Co. 7-Mile Mesa, near Ft. Stockton. Powell 1274. P. vitreomontana 17 TEXAS: Brewster Co. n. side, Glass Mts. Warnock & Powell 19111. 17* TEXAS: Brewster Co. n. side, Glass Mts. Sikes 110. P. quinqueflora 17 TEXAS: Culberson Co. s. McKittrick Canyon, Guad. Mts. Powell 1246. 17 NEW MEXICO: Eddy Co. 6 mi. w. of Whites City. Powell 1319. 62 Species P. rupestris var. rupestris P. rupestris var. rupestris ca. ca. ca. ca. Rhodora n Number 17* 17 17 17 17 34 34 34 34 34 34 34 34 34 II + 1 frag. 68* [Vol. 71 Location and Voucher TEXAS: Jeff Davis Co. ca. 20 mi. w. of FM 2017, San Carlos Tunnel. Sikes & Babcock 326. TEXAS: Jeff Davis Co. 17 mi. n. of Alpine. Powell 1229. TEXAS: Jeff Davis Co. Boy Scout Camp, n. of Ft. Davis. Sikes 33. TEXAS: Jeff Davis Co. 10 mi. n. of Alpine, near Mitre Peak. Sikes 4a. TEXAS: Jeff Davis Co. Madera Canyon, n.w. of Ft. Davis. Sikes 18. TEXAS: Jeff Davis Co. Madera Canyon, n.w. of Ft. Davis. Sikes 17. TEXAS: Jeff Davis Co. Madera Canyon, n.w. of Ft. Davis. Sikes 19. TEXAS: Jeff Davis Co. 2 mi. n. of Ft. Davis. Powell 1228, TEXAS: Jeff Davis Co. Davis Mt. State Park. Sikes 15. TEXAS: Jeff Davis Co. 6 mi. n.w. of Ft. Davis. Sikes 16. TEXAS: Jeff Davis Co. 6 mi. n.w. of Ft. Davis. Sikes 15. TEXAS: Jeff Davis Co. Timber Mt. n. of Ft. Davis. Sikes 45, TEXAS: Jeff Davis Co. 16 mi. n. of Alpine. Powell 1316. TEXAS: Jeff Davis Co. 11 mi. n. of Marfa. Powell 1550, TEXAS: Jeff Davis Co. 13 mi. s. of Toyahvale. Sikes 91. Ca < 1969] Species P. rupestris var. albiflora ca. ca. ca. ca. *Indicates chromosome drawing. Perityle — Powell 63 n Number 17 17 17 17 17 17 17" Location and Voucher TEXAS: Brewster Co. 6 mi. n. of Alpine. Sikes 2. TEXAS: Brewster Co. Basin, Chisos Mts. Sikes 41. TEXAS: Brewster Co. n. side, Cathedral Mt. Sikes 29. TEXAS: Presidio Co. s. side, Goat Mt. Sikes 48. TEXAS: Jeff Davis Co. 10 mi. n. of Alpine, near Mitre Peak. Sikes 4b. TEXAS: Brewster Co. 12 mi. s. of Marathon. Sikes 20, 21, 22. TEXAS: Brewster Co. 17 mi. s. of Marathon. Powell 1408. TEXAS: Brewster Co. 3 mi. W. of Alpine. Powell 1320. TEXAS: Presidio Co. 9 mi. s. of Marfa. Powell 1237. TEXAS: Presidio Co. 10 mi. s. of Marfa. Powell 1238. TEXAS: Presidio Co. n.w. side, Goat Mt. Sikes 30. TEXAS: Brewster Co. Sun- ny Glen, 4 mi. w. of Al- pine. Sikes 1. TEXAS: Brewster Co. 12 mi. s.e. of Alpine, Doubt- ful Canyon. Sikes 93. Lewis (1967) has suggested “that autopolyploids may have a tremendous advantage over diploids at the time of the establishment of new colonies from a single normally out- crossing individual." The mechanism for the origin of poly- ploid populations is not apparent. The close morphic similarity of polyploids and diploids suggests autoploidy as the most logical mechanism, although, as mentioned above, meiosis even in high polyploids appears regular with no "m í [Vol. Rhodora 64 a + ts 9 “ e $ e =?» 4 7 S u ew Ry my x gue RE rs ea 2J eres eo a ^ et “AÒ y í x a. +y t 1969] Perityle — Powell 65 multivalent configurations observed. However, if all poly- ploid populations which have been examined are old and well established, it is likely that any meiotic irregularities would have been eliminated. W. Lewis (1967) and H. Lewis (1967) have revived the classic question concerning the potential of “autopolyploids” to evolve as successful populations. Perhaps the polyploids in Perityle section Pap- pothrix and Chamaesaracha (Solanaceae; Powell and Av- erett, 1967) could be added to the examples presented by the above authors. There is no evidence that interspecific hybridization has been responsible for the polyploids. In fact, all species of section Pappothrix are strictly allopatric and the topog- raphy throughout the total range of these species indicates that separation occurred long ago. Perityle rupestris var. albiflora is known to occur with P. parryi (sect. Perityle) in the Chisos Mts. of Texas, and fertile hybrids apparently are produced between the two taxa (Powell, 1968a). Peri- tyle cinerea cohabits with P. angustifolia (sect. Laphamia) on a Cretaceous caprock mesa in eastern Pecos Co. No hybrids between the latter two species have been found. CHEMICAL STUDIES Chemieal studies were carried out by the use of 2-dimen- sional paper chromatography following the techniques re- ported by Alston and Turner (1963). Distinctions in spots were made in accordance with their different positions on chromatograms and their color variation under ultraviolet light with and without the presence of ammonia vapor. Figures 1-7. Meiotic chromosomes of some Perityle section Pap- pothrix species. Fig. 1. P. cernua, n — 17, (Sikes & Babcock 328). Fig. 2. P. cernua, n = 17 II + 1 round fragment denoted by an arrow, (Sikes & Babcock 328). Fig. 3. P. cernua, n — 17 + 1 round frag- ment (arrow); late anaphase I showing characteristic behavior of fragment, (Sikes & Babcock 328). Fig. 4. P. vitreomontana, n = 17, (Sikes 110d). Fig. 5. P. quinqueflora, n = 17, (Sikes 326b). Fig. 6. P. rupestris var. rupestris, n = 68, (Sikes 91b). Fig. 7. P. rupestris var. albiflora, n = 1". (Powell 1408). All X ca. 1300 except for Fig. 3 which is X ca. 1700. 66 Rhodora [Vol. 71 Mere patterns of compounds (probably flavonoids) were employed in the comparison of species, but care was taken to ensure that pattern data were representative of whole populations and not individual plants or seasonal variants (Powell and Tsang, 1966). Each species can be recognized chromatographically by slight differences in patterns (Figs. 8-13). Specific differences are manifested by the consistent presence or absence of unique compounds (oblique lines in Figs, 8-11), and by small combinations of spots which are present in some species but absent in others. The interpretation of chemical data from Perityle rupes- tris is more difficult because of the occurrence of different patterns for some yellow and white-flowered diploid and polyploid populations of this species (Powell and Tsang, 1966). An understanding of the complex chromatographic variation in P. rupestris has not been realized, but after additional testing it is clear that one vexing phenomenon brought out by Powell and Tsang does not hold true in all instances; viz., that diploid white-flowered and polyploid yellow-flowered plants have the ability to produce yellow compounds in much greater abundance than diploid yellow- flowered and polyploid white-flowered plants. A detailed analysis of the chromatographic data obtained for the sec- tion Pappothrix species will not be possible until a thorough study of the P. rupestris complex has been completed. The main purposes of illustrating the chromatographic patterns (Fig. 8-13) are (1) to show the distribution of yellow com- ponents among the section Pappothrix species, (2) to em- phasize the overall similarity of patterns in this subgeneric Figures 8-13. Chromatographie patterns of the section Pappothrix species of Perityle. Fig. 8. P. cernua. Fig. 9. P. vitreomontana. Fig. 10. P. cinerea. Fig. 1l. P. quinqueflora. Fig. 12. P. rupestris var. rupestris (n = 17). Fig. 13. P. rupestris var. albiflora (n — 17). Darkened spots appear yellow under ultra violet light. Spots which appear to be species specific are denoted by oblique straight lines in association with the spots. No species specific or variety specific spots are designated for P. rupestris because the variation in patterns pro- duced by diploid-polyploid populations of this taxon is not well under- Stood. 1969] Perityle — Powell 67 68 Rhodora [Vol. 71 taxon, and (3) to point out the correlated chemical-mor- phological distinctiveness of P. cernua (Fig. 8). Subsequent discussions based largely on morphic evidence (see Phylo- genetic Considerations and Taxonomy) suggest a prominent phyletic distance between P. cernua and the other four species of section Pappothrix. The copious yellow components (darkened spots) have proven to represent the basis of the tentative chromato- graphic profile for the whole genus Perityle. Not all species of Perityle produce the yellow compounds, but those that do not are considered to be relatively removed from the more primitive species on other grounds. Our current knowledge of these compounds is insufficient to permit more than speculation, but this avenue of investigation appears especially promising. PHYLOGENETIC CONSIDERATIONS Section Pappothrix stands as a monophyletic assemblage of species within the genus Perityle. As discussed earlier the primary unifying character is the pappus of numerous bristles. A connection of section Pappothrix with section Laphamia is seen through P. vitreomontana. The pappus of the latter species is reduced to 2-3 main bristles and numer- ous vestigial nubs (occasionally several other main bristles are evident), This suggests strongly that section Pappothrix is ancestral to section Laphamia (at least the Texas spe- cies), a subject to be discussed further with the taxonomy of section Laphamia (Powell, 1969). No likely extant precursor to section Pappothrix has been discovered. Knowledge of such a taxon would be of theoret- ical importance because it is possible that a Pappothrix-like predecessor fostered the sections Pappothrix and Perityle. As suggested above, a forerunner of the genus Perityle pos- sibly had a chromosome number of n = 18 or 19, or maybe even n = 20, but at present it appears just as possible that the base number could have been n = 17. It is difficult to predict an ancestral species of section Pappothrix, but the progenitor of this phylad must be sought from among P. cernua, P. quinqueflora, or P. rupes- 1969] Perityle — Powell 69 tris. These three species are basically diploid, yellow-flow- ered, and produce a well-developed pappus. Of the remain- ing species, P. cinerea is a white-flowered polyploid and P. vitreomontana is a white-flowered diploid with a reduced pappus. The largest number of characters which could be con- strued as primitive in the section Pappothrix line are ex- hibited by P. cernua. These features mainly are solitary, yellow-flowered heads with ca. 150 florets each, large semi- succulent leaves which are glabrous, ca. 35 pappus bristles (compared with 20-30 > 10-14 in other species), and a diploid chromosome number. However, there are factors which might appear to vitiate such a status for P. cernua. The species is extremely limited in distribution with a lo- cation at the very northwestern periphery of the section Pappothrix range. Additionally, chromatographic studies of P. cernua have revealed a relatively small quantity of the yellow compounds which are so abundant in other diploid species of this section. Some factors support the possibility that Perityle quin- queflora was the preludial taxon of section Pappothrix. Even though the latter species is noted for its numerous reduced heads with 5-6 florets, the leaf morphology is sim- ilar to that of P. cernua which has what are considered to be the most primitive vegetative features. Perityle quinque- flora is a widely distributed diploid species occurring mostly in geologically old limestone areas. Additionally, chromato- graphic studies have revealed that this species produces yellow compounds in abundance. A few plants in all popu- lations throughout the range of P. quinqueflora exhibit heads with 6-8 florets (this is especially true in the popula- tion at San Carlos Tunnel in extreme western Jeff Davis Co.), suggesting that reduction in number of florets has been coincident with the evolution of the taxon. Since plants of Perityle rupestris produce heads with 8-12 florets, it is possible to consider this taxon as ancestral to P. quinqueflora and perhaps to the other species of sec- tion Pappothrix as well. However, P. rupestris is located mostly in geologically more recent areas than is P. quinque- 70 Rhodora [Vol. 71 flora, and the leaves are thinner and more pubescent than are those of the latter species. Even though glabrous con- ditions might be evolved from pubescent ones in many Com- posite taxa, this does not seem to have been the case in the derivation of species of section Pappothrix. At least in P. rupestris a more dense indument seems to be correlated with polyploidy, and the optimum in pubescence of section Pappothrix species is seen in the octaploid P. cinerea. When one evaluates the total evidence available, it is apparent that P. cernua must be regarded as the ancestral taxon of the section Pappothrix. In the first place it would be difficult to visualize how the few, numerous-flowered heads of P. cernua could have evolved from a species with numerous 5-12 flowered heads, especially since P. cernua is diploid. The determination of the chromosome number of this species was not unequivocal, but certainly the number is » — 17 or 18. Regrettably, only one head of buds from this rare species has been available for examination, and it was produced in the greenhouse. I have interpreted the chromosome number to be n = 17 + 1 fragment because late anaphase I stages show 17 chromosomes at each pole in many cells, and occasionally 18 at one pole and 17 at the other. A few cells revealed that one chromosome, or portion of a chromosome, was not attached to the spindle apparatus and was left suspended between the poles (Figs. 1-3). Such behavior of chromosome fragments (and univalents) is well known to be associated with reduction in chromosome num- ber. This corresponds with other cytological evidence since there appears to be an aneuploid reduction series in Perityle as manifested by several species (Powell, 1968b). Secondly, the marginal location of Perityle cernua within the total section Pappothrix range may be construed as en- hancing rather than negating a primal position for the taxon. Axelrod (1958) has emphasized that speciation with- in the general Madro-Tertiary Geoflora was stimulated by adaptation of taxa to drier habitats which came into exist- ence at the margins of deserts, Raven (1964) has discussed catastrophic selection with regard to taxa which become isolated at the peripheries of relatively mesic-xeric habitats. 1969] Perityle — Powell 71 General principles advanced by both authors appear to fit the proposed evolution of section Pappothrix species. Third- ly, it is not yet possible to assess sophisticated conclusions from chemical data. Such data rest ultimately upon the identification of compounds and information regarding their biosynthesis. However, since the obviously more advanced species of section Pappothrix (e.g., Perityle cinerea and P. vitreomontana) produce large amounts of the basic yellow compounds, it is likely that an increased capacity to make the substances is derived. The compounds in question do appear in P. cernua but in relatively smaller quantity. A comparison of present day topography with presumed Late Tertiary geology gives some insight into how the evolution of section Pappothrix species could have come about. According to Sellards et al. (1958), a Cretaceous limestone cap extended virtually intact across the Trans- Pecos into New Mexico. With the activity of Tertiary vul- canism and subsequent erosion, many of the rock-bluff potential habitats of section Pappothrix species progressive- ly must have become more disjunct. Assuming that the distribution of an ancestral species with vegetative and floral morphology approaching present day P. cernua be- came widespread across an ecologically uniform area before the mountain chains were well separated, the stage would have been set for speciation through geographical isolation. Later erosion could have accounted for more rapid and complete isolation of existing populations such as we see today in the Trans-Pecos. The restriction of section Pap- pothrix species habitats to mountain regions and other prominent land features (such as remnants of limestone caprocks) seems to support such an evolution of taxa. The broken mountain systems which extend from south- ern New Mexico through the Trans-Pecos into Mexico ap- parently were once continuous enough to allow the estab- lishment of a mesophytie montane forest in the higher altitudes, although Wells (1966) has suggested that many of the mesophytic species have never existed in continuum between some of the mountains. All species of section Pap- pothrix occur at various altitudes in these mountains, ex- 71 [Vol. Rhodora 72 'OL9X AOATYB[eL 9i? SAIL 194470 IV '(oS3u*€X 740j19]D ‘iea Siuqsadnus *q JO Əut4)xə uzəul)nos) 'S}W SosrqO pue ‘(asuet sL4sədnia “ABA sisodn4 “gq (817099) ‘sW seq '(93ue4 n4oyonbumb *q Jo oureijxo uieujyiou) “SI ən[epen5 ‘(vnULa2 `J) `s1IÑN u€34() ay} ur pojeoo[ oi? soroeds osouj Jo syeyqey orsour ‘suryeeds AppAn*[esd ^3ueuudo[pAep səroəds Jo uorjoe4rp o[qQeqoid ƏY} MOYS SMOIIY ‘atay poyetassexe A[jeais oie nunQuow -094]14 `d PUB DNULAD `J JO SLAE [euonnqirdsip IJL '(NIO) DILIM ‘`J PUL '(LIA) DuDjuowoo4j2 ‘q ‘(IATV) DLOYig]D “IBA $142052dm4 `J poioMopg-93IQ^ JO sasuei oy} ojeudIsop svere po[ddrg ‘seul, [euoseip Aq pezeorpur sI (ana) SISIN “IBA s14252dn4 `q pue ‘(ONIND) naoyonbuimb *q *(wa0) onmu4oo `q VX patamoy-MoT[ad ay}, Jo uon -nqiyqsip Aep 3uesedq ‘satoeds xwyjoddyg jo uomnmoəs ajhjiwag JO eouenbes Aaeuorynjoas pesoddng ‘pT IIA ITVIS $31IW 00! os ; I8'IV ONINO O 1969] Perityle — Powell 73 cept Perityle cinerea, and it is probable that conditions at least were conducive to the migration of a precursoral taxon. Assuming first of all that section Pappothrix represents a northeastern fringe element of the Madro-Tertiary Geo- flora, Figure 14 shows how the various section Pappothrix species could have arisen after distribution along the moun- tains in which they are now located from northern into more southern areas. Presumably Perityle cernua represents a relictual mesophytic population of ancestral section Pap- pothrix which migrated into the southerly mountains when relatively moist environments prevailed. Since the floral features of Perityle cernua (large, solitary heads) differ so markedly from those of P. quinqueflora and P. rupestris, it is conceivable that a prototype of the latter taxa developed the characteristic smaller, numerous heads long before their divergence as separate species. On the basis of the above postulate, it would follow that P. cernua has been isolated much longer than have P. quinqueflora and P. rupestris. Possibly the latter species gradually became isolated through erosion of their elevated habitats and subsequently becoming adapted to the more xeric habitats in which they now occur. For example, the northernmost population of P. quinqueflora in the Guadalupe Mts. resembles P. cernua in vegetative habit, and a southernmost population at San Carlos Tunnel resembles P. rupestris in the same way. Probably, the San Carlos population of P. quinqueflora was most recently isolated from P. rupestris. The northern- most populations of P. rupestris are more like P. quinque- flora vegetatively than the more xerophytic southern populations of this species to the south. It is reasonably certain that yellow-flowered P. rupestris preceded the origin of white-flowered P. rupestris var. albiflora. White-flowered populations are distributed entirely south of the yellow- flowered colonies in somewhat drier habitats (Figure 14). The white-flowered Perityle cinerea and P. vitreomontana apparently evolved from offshoots of the white-flowered line arising out of P. rupestris. That an ancestral taxon might have given rise to two separate species independently is suggestive of a similar circumstance reported in Chaenac- 74 Rhodora [Vol. 71 tis by Kyhos (1965). With these section Pappothriz species, however, geographic separation is proposed as the mech- anism of initial isolation rather than the aneuploid reduc- tion of chromosome number and structural rearrangements of chromosomes which occurred in Chaenactis. Perityle cinerea could have originated from a population which mi- grated into the present Cretaceous caprock area at a time when a more mesic and continuous environment existed throughout the region. Wells (1966) has discovered that pinyon pine was present across the same general territory as recent as Pleistocene. Both P. cinerea and P. vitreomon- tana seem to be manifestations of differential adaptation to more xeric habitats. P. cinerea has become established as a octaploid species, while P. vitreomontana developed as a semi-depauperate diploid species. If speculations about the ramification of Perityle cinerea and P. vitreomontana are accurate, an interesting question remains about why their older phylogenetic antecedent did not diverge to a similar degree. Current distribution of the three taxa would suggest that P. rupestris var. albiflora, was never disassociated from P. rupestris var. rupestris, and their relatively uniform environments, as were P. cinerea and P. vitreomontana. Morphological evidences of the relationships projected above can be gleaned from the taxonomic discussions which follow. TAXONOMY Perityle sect. Pappothrix A. Gray, Pl. Wright. 1: 100. 1852. Pappothrix (A. Gray) Rydb., N. Amer. Fl. 34: 26. 1914. TYPE: Laphamia rupestris A. Gray. Plants small suffrutescent perennials, or the stems herbaceous and brittle, growing in crevices of limestone or igneous rock bluffs, gla- brous, puberulent, hirsute, pilose, or tomentose-canescent; leaves alter- nate or opposite, 1.0-4.0 cm long, 0.6-3.0 em wide, ovate, reniform, or cordate, entire to serrate or serrate-lobed; inflorescence of solitary and erect or nodding heads, or several erect heads, partially obscured by leaves; heads broadly to narrowly campanulate or subcylindric, 0.4-1.4 em wide; involucre of 2 equal or subequal series; ray flowers absent; disc flowers 5-150; disc corollas yellow or white; style branches linear, tapering to a minutely pubescent point; achenes black, 1.8- 1969] Perityle — Powell 75 3.0 mm long, oblong to narrowly obconical, partially flattened to sub- cylindric, usually rounded or angled on one or both surfaces, with rather inconspicuous calloused margins, the margins, angles, and surfaces short-pubescent; pappus of (2-7)10-35 slender, unequal bristles, 1.0-3.5 mm long, minutely antrorse-ciliate; base chromosome number, x — 17. Since the taxonomy of the other two sections of Perityle will be published separately, a key to all three sections is presented here to facilitate the identification of each sub- generic group. KEY TO THE SECTIONS Pappus of (2-7-)10-35 bristles; achenes subcylindrical to partially flattened, margins short-pubescent. ........................ sect. Pappothriz. Pappus of 0-2(-3-6) bristles; achenes flattened, margins short-pubes- cent or profusely ciliate. Pappus of bristles only (rarely none) or with inconspicuous, ves- tigial squamellae; achene margins sparsely short-pubescent. .... Lud uiui di sa Wa E sect. Laphamia. Pappus of bristles (rarely none) and a crown of squamellae; achene margins profusely ciliate. ............................ sect. Perityle. KEY TO THE SPECIES OF PERITYLE SECTION PAPPO- THRIX Heads large, florets ca. 150; corollas yellow; Organ Mts., New Mex- ICO 02 E x E 1. P. cernua. Heads small, florets 5-16(20) ; corollas yellow or white. Florets 5-6(7-8); leaves somewhat lustrous, dark olive-green, subsucculent to coriaceous, typically reniform to subreniform, essentially glabrous to puberulent, rarely hirtellous. ................ E ddnde ro ee 2. P. quinqueflora. Florets (7)8-16(20); leaves lighter green, rather thin, typically ovate, subdeltoid, to subcordate, notably pubescent. Corollas yellow or white; pappus of 20-30 well-developed bristles; leaves pilose to hirsute; widely distributed in igneous or limestone rocks in Davis Mts., Texas. Corollas yellow. ................ 3a. P. rupestris var. rupestris. Corollas white. ................ 3b. P. rupestris var. albiflora. Corollas white; pappus reduced or well-developed; leaves pilose to tomentose-canescent. Pappus reduced to 2-3(6) main bristles, less often 10-20, including vestigial nubs; leaves pilose; Glass Mts., Pappus of 14-20 well-developed bristles; leaves tomen- ORR ole 4. P. vitreomontana. tose-canescent; Cretaceous limestone, Pecos, and Up- ton Counties. ‘Texas. |... 00902... eha 5 P. cinerea. 76 Rhodora [Vol. 71 Figure 15. Habit sketch of Perityle cernua (Sikes 65); X 1. 1. Perityle cernua (Greene) Shinners, Southw. Nat, 4: 204-209. 1959. Fig. 15. Laphamia cernua Greene, Bull. Torr. Bot. Club 25: 122, 1898. TYPE: NEW MEXICO: DoÑA ANA Co: Organ Mts., at 6500 ft., 4 Sept. 1897, E. O. Wooton 476? (Holotype, ND-G! Isotypes US!). Pappothrix cernua (Greene) Rydb. N. Amer. Fl. 34: 27. 1914. Plants 3-6 em high, closely appressed to rocks, essentially glabrous, dark, lustrous green; leaves alternate or opposite on very short stems, the leaves 2.3-4.0 cm long, 2.0-3.0 cm wide, ovate-reniform to ovate- cordate, the margins unevenly serrate-dentate, essentially glabrous; petioles 0.7-1.5 cm long; inflorescence of solitary and erect or nodding heads often partially obscured by leaves, the peduncles 1.0-1.9 cm long; heads broad-campanulate, 1.0-1.2 cm long, 1.2-1.4 cm wide; receptacle convex, rounded, 3.0-4.0 mm wide, ca. 2.0 mm high; bracts ca. 28, 8-9 mm long, 1.3-1.5 mm wide, linear-lanceolate; disc flowers ca. 150; dise corollas golden yellow, 5.5-6.0 mm long, the tube 1.7-2.0 mm long, minutely glandular-pubescent, the throat narrowly funnel- form, 2.8-3.8 mm long, sparsely glandular-pubescent, the lobes nar- row, acute, 0.7-0.8 mm long; achenes 2.1-2.5 mm long, partially flattened; pappus of ca. 35 slender, unequal bristles, 1.2-1.5(2.0) mm long, minutely antrorse-ciliate; anthers 2.0-2.5 mm long; chromosome number, » — 17 II + 1 frag. An extremely rare endemic species, known only from the type locality at Dripping Springs Canyon, growing in lime- 1969] Perityle — Powell 77 stone rock walls, w. side, Organ Mountains, Dona Ana County, New Mexico. Spring-Fall. (Fig. 17). Representative Specimens: NEW MEXICO: DoNa ANA Co: Organ Mts., 29 Aug. 1894, E. O. Wooton s. n. (US); n. face of Dripping Springs Canyon, 12 mi. e. of Las Cruces, Organ Mts., S. Sikes 65 (SRSC) : Dripping Springs Can- yon, Organ Mts., 10 May 1967, S. Sikes and C. Babcock 328 (SRSC). The most anomalous species of section Pappothria is Perityle cernua. Easily recognized by its small habit, large, solitary heads, and relatively large leaves, P. cernua is not easy to relate to other species. This is true mainly because there are no extant section Pappothrix species with such large heads. It is probable that P. cernua represents the ancestral species of section Pappothrix, exhibiting such primitive characters as yellow flowers, few, large heads with numerous florets, and large, glabrous, thick leaves. Vegetatively, notwithstanding the large heads, P. cernua appears closest to P. quinqueflora. A.CANO "OT Figure 16. Habit sketch of Perityle quinqueflora (Powell 1319); x 1. 78 Rhodora [Vol. 71 2. Perityle quinqueflora (Steyermark) Shinners, Southw. Nat. 4: 205. 1959. Fig. 16. Laphamia quinqueflora Steyermark, Ann. Mo. Bot. Garden 19: 392, 393. 1932. TYPE: TEXAS: CULBERSON Co: niches in exposed limestone cliffs, lower MeKittrick Canyon, Guadalupe Mts., alt. 1900 m., 20 July 1931, Moore and Steyermark 3547 (Holotype, GH! Isotypes, cas! ps! uc! US!). Pappothrix quinqueflora (Steyermark) Everly, Contrib. Dudley Herb. 3: 375-396. 1947. Plants 7-30 em high, densely leafy, essentially glabrous to puberu- lent; leaves opposite, often becoming alternate above, 1.5-2.5(3.8) cm long, 0.8-2.0(3.3) em wide, subsucculent to coriaceous, somewhat lus- trous, dark olive-green, typically reniform to subreniform, sometimes short-cordate, the margins subentire, dentate-toothed or shallow-lobed, the teeth lobes typically obtuse or rounded but rarely acute, apex obtuse, rounded, or rarely subacute, subcuneate at the base, essen- tially glabrous to puberulent, rarely hirtellous; petioles 5-8(12) mm long; inflorescence typically of several to many heads in a cluster, borne on rather slender peduncles 2-10(15) mm long; heads cylindrical to very narrowly campanulate, 7-8(9) mm long, 2-3 mm wide, often partially obscured by leaves; receptacle flattened to slightly convex, ca. 1.0 mm wide; involucre of 5-6 bracts in essentially one series, 5-6(9) mm long, (1.0)1.3-1.7 mm wide, strongly keeled; disc florets 5-6(8) ; dise corollas yellow, 3.8-4.3 (rarely 5.0) mm long, the tube 1.0-1.3 mm long, minutely glandular-pubescent, the throat tubular, 2.0-2.8 (rarely 2.8) mm long, sparsely glandular-pubescent, the lobes acute, reflexed 0.5-1.0 mm long; achenes 1.9-2.0 (rarely 2.5-2.9) mm long, flattened; pappus of 25-30 slender bristles, 1.9-2.5(3.5) mm long, the bristles unequal in length, minutely antrorse-ciliate; anthers 1.5-2.0 mm long; chromosome number, n — 17. Growing extensively in crevices of limestone bluffs (rare- ly igneous in the Sierra Vieja and Eagle Mts.), in high canyons and caprock. Spring-Fall. (Fig. 17 ). Representative Specimens: NEW MEXICO: Eppy Co: Guadalupe Mts., 12 July 1939, A. L. Hershey 619 (CAS) ; crevices of limestone rock in canyon to Carlsbad Caverns, 30 May 1965, A. M. Powell 1319 (SRSC); s.w. portion of Dark Canyon, Guadalupe Mts., 1 Sept. 1965, Powell & Sikes 1379 (SRSC). CULBERSON Co: Pine Canyon, Guadalupe Mts., 20 June 1964, D. S. Correll & C. Hanson 29791 (LL); South McKittrick Canyon, 2 July 1958, Correll & I. M. Johnston 19177 (LL); Beach Mts., 8 mi. n. of Van Horn, 3 June 1938, H. C. Cutler 1978 (CAS); South McKittrick Canyon, 25 Aug. 1950, W. V. Fischer s.n. (uc); South McKittrick Canyon, 11 July 1948, L. C. Hinckley 4467 (sRSC); upper McKittrick Canyon, 21 June 1947, F. G. Meyer & L. E. Meyer 2178 (UC) ; between 79 Perityle — Powell 1969] [911 peso[p-j[eu e Aq pəlouəp st uorje[ndod otrjeduiÁsS € Jo uorjeoo[ BY} !sS14qsədmna "q JO SatjelIVA ay} jo Áymunuoosrp ay} səzıseydwə əu pios V '"(e[gueri peso[p) 2242242 «gq :(o|duerij uədo) DUPJUOWMOILJUA `d “(SƏP peso[p) n4oyfiq]o "IBA SIJSIƏANAL ‘q É (SƏP uədo) sL4sədma “rea Siqsədmuia ‘q ‘(sərenbs uədo) vsoyənbunb 'q * CW 'N “SUA ue34O ‘atenbs peso[p) DNULII `J :SƏət99ds xiwyjoddyg UON AÂJ jo uorjnquijsu[ “LT IMELA ITVIS x — —ñ $31IW 00! yy — 0$ i ° ° s A i ed e i i s o0 Se o oo 80 Rhodora [Vol. 71 South & North McKittrick Canyons, 15 July 1945, R. McVaugh 7398 (LL, SMU); Smith Canyon, Guadalupe Mts., 17 July 1945, R. McVaugh 7418 (LL, SMU); South McKittrick Canyon, 4 July 1964, A. M. Powell, B. Calvert, & P. Tsang 1246 (SRSC) ; Victoria Canyon, Sierra Diablo, 10 Aug. 1946, L. H. Shinners 8865 (SMU, TEX); Pine Springs Canyon, Guadalupe Mts., 15 Aug. 1946, Shinners 9062 (SMU); McKittrick Canyon, 15-17 Aug. 1924, P. C. Standley 40562 (us); upper Pine Springs Canyon, 15 Sept. 1948, B. L. Turner & B. H. Warnock 123 (LL, SMU, SRSC) ; Smith Canyon, 10 July 1949, Turner and Warnock 1265 (SMU, SRSC); McKittrick Canyon, 30 Aug. 1950, Warnock 9456 (LL, SRSC); Victoria Canyon, Diablo Mts, 11 July 1956, Warnock 14091 (LL, SRSC) ; South McKittrick Canyon, 11 Sept. 1962, Warnock 204,07 (SRSC); South McKittrick Canyon, 18 May 1958, Warnock & M. C. Johnston 16529 (SRSC); lower Pipe Line Canyon, 5 July 1958, Warnock & Johnston 16589 (sRsc); North McKittrick Canyon, 4 Aug. 1962, Warnock & T. Rogers 18864 (SRSC); Pine Top Mt., 28-29 July 1952, G. L. Webster 4578 (SRSC) ; Guadalupe Mts., 15 Aug. 1916, M. S. Young s.n. (TEX) ; Guadalupe Mts., 2 Sept. 1916, M. S. Young s.n. (TEX, SRSC). HUDSPETH Co: Quitman Canyon, Quitman Mts., 13 Oct. 1952, D. S. Correll 15065 (LL); s. Sierra Diablo Plateau, 9 Sept. 1961, Correll & Johnston 24362 (LL); upper limestone slopes, Victoria Canyon, 10 July 1945, R. McVaugh 7359 (LL, SMU); Van Horn, 4 Sept. 1925, B. C. Tharp 3818 (TEX) ; head of Victoria Canyon, 18 Aug. 1953, Warnock 11495 (SRSC); s.e. of Old Love Ranch, Eagle Mts., 22 Aug. 1946, U. T. Waterfall 6684 (NY). JEFF Davis Co: San Carlos Tunnel, 22 July 1967, Powell & Sikes 1536 (SRSC); ca. 20 mi. w. of FM 2017, San Carlos Tunnel, 11 May 1967, Sikes & Babcock 326 (SRSC). PRESIDIO Co: Bracks Canyon, Sierra Tierra Vieja, 11 June 1941, L. C. Hinckley 1675 (SMU, TEX, US) ; near rimrock, Sierra Tierra Vieja, 4 Oct. 1941, Hinckley 2185 (LL, NY); Vieja Pass, Vieja Mts., 3 Aug. 1960, Warnock 19600 (SRSC). It is clear that Perityle quinqueflora is most closely re- lated to P. rupestris. In fact, the several collections from Sierra Tierra Vieja (Jeff Davis Co. and Presidio Co.) are diffieult to distinguish from P. rupestris. Normally P. quin- queflora is readily distinguished from P. rupestris by 5(6) flowered heads, 5(6) involucral bracts, rather fleshy, dark green, subglabrous leaves with margins varying from sub- entire to serrate-lobed, and habitat mainly in limestone rock. The Sierra Vieja collections at San Carlos Tunnel approach P. rupestris by exhibiting 5-7(8) florets and bracts per head, and thinner, lighter green, hirtellous leaves with deep serrations of the leaf margins. Even though the Sierra Vieja population closes the morphological gap between these 1969] Perityle — Powell 81 two taxa, there is no evidence of intergradation. Geograph- ically P. quinqueflora and P. rupestris remain well separated (Fig. 17). It is interesting to speculate about the phylogeny of these related entities. Perityle quinqueflora features several re- duced characters, especially the 5-membered heads and essentially glabrous habit, and could thus be viewed as de- rived from P. rupestris. However, the former species is distributed in the geologically older areas, for the most part, and is known only as a diploid. Perityle rupestris on the other hand, at least the yellow-flowered variety, occupies the geologically more recent tertiary igneous intrusive and extrusive formations in Jeff Davis Co., and occurs in diploid and polyploid forms. The white-flowered variety of P. rupestris inhabits both limestone and igneous formations. As discussed above, most evidence concerning the re- lationship of P. quinqueflora and P. rupestris seems to in- dicate an evolution of both species from a widespread ancestor much like P. cernua. Hypothetically, two large populations could have been isolated geographically in re- cent geological history when their relatively continuous rock-bluff habitats were eroded away. Once they were sep- arated the populations would have been free to follow their independent evolutionary paths. It is possible that a fluc- tuating Pleistocene environment could have allowed sec- ondary sympatry after initial isolation if erosion of intermediate habitats was not complete. There is no evi- dence for the latter unless one considers the somewhat morphologically intermediate colony of P. quinqueflora at San Carlos Tunnel to have resulted from hybridization with P. rupestris. At this time it appears feasible to interpret the San Carlos population as one which has been most re- cently segregated from P. rupestris. da. Perityle rupestris (Gray) Shinners var. rupestris, Southw. Nat. 4: 204-209. 1959. Fig. 18. Laphamia rupestris Gray, Pl. Wright. 1: 100. 1852. TYPE: TEXAS: JEFF Davis Co: crevices of rocks, on mountains, in the Pass of the Limpia, Aug. 1849, Charles Wright 300 82 Rhodora [Vol. 71 (Holotype, GH! Isotypes, GH! NY! US! Photograph, ps!). Pappothrix rupestris (Gray) Rydb. N. Amer. Fl. 34: 26. 1914. Plants 6-20(35) em high, puberulent, pilose, or hirsute, profusely branched and leafy; leaves opposite or alternate, usually alternate, 1.0-2.5(3.5) cm long, 0.7-1.7(3.0) em wide, ovate to broadly so, sub- deltoid, subeordate, or rarely subreniform, the margins irregularly serrate or serrate-lobed, the serrations or lobes acute (rarely obtuse), pilose, hirsute, or rarely puberulent; petioles 0.3-1.3 em long; in- florescence of solitary heads or small, loose clusters, heads borne on rather short peduncles, often partially obscured by leaves; receptacles flattened to slightly convex, 1.0-1.5 mm wide; heads subcylindric, sub- funnelform, or narrowly companulate, 6-8 mm long, 3-5 mm wide; bracts 8-10, 4-7 mm long, 1-2 mm wide, typically oblong to oblong- lanceolate, keeled; disc flowers (7) 8-12; disc corollas yellow, 4.0-5.2 mm long, the tube 1.0-1.3 mm long, minutely glandular-pubescent, the throat tubular to subfunnelform, 2.2-3.0 mm long, the lobes acute, 1.0-1.3 mm long; achenes (1.8) 2.0-2.5 mm long, subtruncate or taper- ing toward the base, flattened (rarely nearly cylindric); pappus of 20-30 slender, unequal, bristles, 2.0-3.0(3.5) mm long; anthers 2.0-2.5 mm long; chromosome numbers, n — 17, 34, 68. An extremely variable taxon, common in crevices of igneous boulders and bluffs, from ca. 10 mi. north of Alpine Figure 18. Habit sketch of Perityle rupestris var. rupestris (Pow- ell 1228); X 1. 1969] Perityle — Powell 88 throughout Jeff Davis County and probably north Presidio County. Spring-Fall. (Fig. 17). No mention of flower color was made in Gray’s original description of Laphamaa rupestris, but the morphology of the type specimens and Wright’s collection site indicate that this is a member of the yellow-flowered variety. Representative Specimens: TEXAS: JEFF Davis Co: e. slope, Haystack Mt., 6050 ft., 27 May 1959, T. J. Allen 353 (SRSC); 10 mi. s.e. of Ft. Davis, 28 April 1947, V. L. Cory 53568 (NY, SMU); Fort Davis, 19 Sept. 1920, W. W. Eggleston 17425 (NY, US); Fort Davis, 20 Sept. 1920, Eggleston 17474 (NY, US); rocky hillside back of old Ft. Davis, 9-12 July 1921, R. S. Ferris & C. D. Duncan 2683 (DS); 12 mi. n. of Alpine, 18 Oct. 1933, H. T. Fletcher 1482 (sRSC); Goat Canyon, Mt. Livermore, 7 Sept. 1935, L. C. Hinckley s.n. (TEX); n. side of Timber Mt., near Madera Springs, 24 July 1949, L. C. Hinckley & L. Hinckley 1923 (SMU, SRSC); Limpia Canyon at Wild Rose Pass, 11 Sept. 1949, Hinckley & Hinckley 338 (sRSC); ridge s. of Mt. Livermore, 21 Aur, 1935, Hinckley 356 (LL); Mt. Livermore, Aug., 1935, Hinckley 50* (sRSC); Limpia Canyon near Ft. Davis, 8 Oct. 1944, Hinckley 3381 (SRSC) ; igneous rock above old Ft. Davis, 6 Nov. 1958, M. C. Johnston 3578 (SRSC); igneous rock, 18 mi. n. of Alpine, 6 Nov. 1958, Johnston 3589 (SRSC); 15 mi. n. e. of Ft. Davis, canyon e. of Limpia Creek, 24 May 1949, R. McVaugh. 10626 (SRSC, US); Little Aguja Canyon, 12 June 1931, J. A. Moore & J. A. Steyermark 3041 (CAS, DS, NY, UC, US); clefts and ledges of porphyritic rocks, Davis Mts., 7 Oct. 1926, E. J. Palmer 32029 (TEX); crevices in basaltric rocks, on the Limpia, C. C. Parry s.n. (NY) ; ca. 2 mi. n. of Ft. Davis along Limpia Creek, 6 June 1964, A.M. Powell & B. Calvert 1228 (SRSC); ca. 17 mi. n. of Alpine along Musquiz Creek, 6 June 1964, Powell & Calvert 1229 (SRSC); large igneous boulders, 16 mi. n. of Alpine, 14 May 1965, A. M. Powell & P. R. Powell 1316 (sRSC) ; 11 mi. n. of Marfa, 19 Aug. 1967, Powell 1550 (sRSC); 10 mi. n. of Alpine, near Mitre Peak, 5 June 1965, S. Sikes 4a (SRSC) ; Davis Mt. State Park, 3 Sept. 1965, Sikes 15 (SRSC); 6 mi. n. w. of Ft. Davis, 3 Sept. 1965, Sikes 15, 16 (SRSC); Madera Canyon, n. w. of Ft. Davis, 3 Sept. 1965, Sikes 17, 18, 19 (sRSC); Boy Scout Camp n. of Ft. Davis, 30 Sept. 1965, Sikes 33 (SRSC); Timber Mt., n. of Ft. Davis, 5 June 1966, Sikes 45 (SRSC) ; 13 mi. s. of Toyahvale, 28 Aug. 1966, Sikes 91 (SRSC); rock crevices, Fern Canyon, 20 Oct. 1945, O. Sperry T1322 (us); Fern Canyon, Aug. 1932, T. L. Steiger 954 (NY); lower Limpia Canyon, 9 May 1937, B. H. Warnock T80 (TEX, US) ; Musquiz Canyon, 16 mi. n. of Alpine, 19 June 1947, Warnock 6158 (SRSC); n. slopes of Timber Mt. above Madera Springs, 20 July 1947, Warnock 6493 84 Rhodora [Vol. 71 (SRSC); upper Madera Canyon, Mt. Livermore, 11 Sept. 1947, War- nock & Hinckley 7458 (SRSC); upper Madera Canyon, 25 Oct. 1947, Warnock 7619 (sRSC); Wild Rose Pass, Limpia Canyon, 24 May 1949, Warnock & McVaugh 8649 (SRSC) ; n. side Timber Mt., 14 May 1950, Warnock & N. Tallant 9058 (SRSC); 12 mi. n. of Alpine, 15 Sept. 1963, Warnock & Powell 19289 (SRSC); 12 mi. n. of Alpine, 11 Sept. 1962, Warnock 20404 (SRSC); 15 mi. n. of Alpine, upper Rose Canyon, 1 July 1940, Warnock 20909 (TEX); 20 mi. s. s. e. of Kent, 31 July 1943, U. T. Waterfall 5406 (cas, NY); Davis Mt. Park, 23 June 1941, E. Whitehouse 12129 (SMU); Mt. Livermore, 14 Aug. 1914, M. S. Young s. n. (TEX). 3b. Perityle rupestris (Gray) Shinners var. albiflora Powell, Sida 3: 180. 1967. TvPE: TEXAS: BREWSTER Co: in crevices of igneous bluffs, ca. 3 mi. w. of Alpine in Tronto Pass, 3 June 1965, A. M. Powell & S. Sikes 1320 (Holotype, SRSC; isotypes, GH, SMU, TEX, US). Disc corollas white; disc florets 8-18; achenes 1.8-2.5 mm. long; pappus bristles 18-25, 1.0-2.0 mm. long; chromosome numbers, n = 17, 34, 85. A variable taxon, common in crevices of igneous and limestone bluffs and boulders, from ca. 10 mi. n. of Alpine, south throughout Brewster County, extreme south Jeff Davis County, and eastern Presidio County. This variety is distinguished by white disc corollas, smaller leaves in most populations, tendency for more florets per head, pappus bristles mostly as long as or shorter than the achenes, and geographic distribution (Fig. 17). Representative Specimens: TEXAS: BREWSTER Co: ca. 15 mi. s. e. of Marathon, 13 Sept. 1961, D. S. Correll & M. C. Johnston 24577 (LL); Pena Blanca Mts., 6 mi. e. of Hwy. 385, 30 Sept. 1962, Correll & H. B. Correll 26109 (LL); Pine Canyon, Chisos Mts., 17 June 1963, Correll & D. C. Wass- hausen 27867 (LL); vicinity of Alpine, 7 June 1926, D. Demaree & E. J. Palmer 104 (SMU); n. slope, Cathedral Mt., 21 July 1960, L. Doyles & B. Jobes 71 (SRSC); Lizzard Mt., 2 mi. w. of Alpine, 21 Sept. 1934, H. T. Fletcher 144 (SRSC); Lizzard Mt., 10 Sept. 1929, Fletcher 1384 (SRSC); s. end of Del Norte Mts., 6 Sept. 1947, L. C. Hinckley & F. M. Churchill 4112 (smu, us); Alpine, 3 Oct. 1950, R. Hughes 12 (NY); upper slopes Cathedral Mts., 26 June 1959, J. Jackson 130 (SRSC); ca. 16 mi. s. e. of Marathon, 13 July 1964, A. M. Powell & B. Calvert 1252 (sRSC); Tronto Pass, ca. 9 mi. w. of Alpine, 3 June 1965, Powell & S. Sikes 1320 (SRSC) ; 17 mi. s. of Marathon, 31 May 1966, Powell, T. Watson, & Sikes 1408 (SRSC); Hidden Valley near Alpine, 20 June 1941, R. Rose-Innes & B. Moon 1969] Perityle — Powell 85 1261 (TEX); Sunny Glen, 4 mi. w. of Alpine, 4 June 1965, Sikes 1 (SRSC); 6 mi. n. of Alpine, 5 June 1965, Sikes 2 (SRSC); 12 mi. s. of Marathon, 4 Sept. 1965, Sikes 20, 21, 22 (sRSC); n. side of Cathe- dral Mt., 23 Sept. 1965, Sikes 29 (SRSC) ; Basin, Chisos Mts., 26 Sept. 1965, Sikes 31 (sRSC); Doubtful Canyon, 12 mi. s. e. of Alpine, 2 Sept. 1966, Sikes 93 (SRSC); s. slopes, Cathedral Mt., 21 July 1960, M. Snowden & R. Rice 374 (SRSC); Paradise Canyon, 22 Sept. 1935, O. E. Sperry T325 (LL, SRSC, US); Alpine, May, 1932, T. L. Steiger 1043 (NY); Jackson Ranch, near Alpine, 13 Aug. 1938, B. H. Warnock T513 (vs); Bluff Cave, 6 mi. s. w. of Alpine, 24 June 1940, Warnock W907 (TEX); 6 mi. n. w. of Alpine, 17 May 1947, Warnock 5580 (LL, SMU, SRSC, TEX) ; 16 mi. s. of Marathon, 29 June 1947, Warnock 6127 (SRSC, TEX); Doubtful Canyon, 25 mi. s. of Alpine, 18 Sept. 1947, Warnock & Hinckley 7527 (sRsc); Doubtful Canyon, 18 Sept. 1947, Warnock & Hinckley 7535 (SMU, SRSC); Doubtful Canyon, 11 July 1948, Warnock 7983 (LL, SMU, SRSC); upper Pine Canyon, Chisos Mts. 1 Aug. 1950, Warnock 9179 (SRSC); base of Bailey Peak, Chisos Mts., 6 Sept. 1950, Warnock 9593 (SRSC); 18 mi. s. of Marathon, 17 June 1964, Warnock 20419 (SRSC); 3 mi. w. of Alpine, 15 Sept. 1964, Warnock 20491 (SRSC); 20 mi. s. e. of Marathon, 29 Sept. 1964, Warnock 20501 (srsc); Pena Blanco Springs, 9 Oct. 1964, Warnock 20548 (SRSC). JEFF Davis Co: 10 mi. n. of Alpine, near Mitre Peak, 5 June 1965, Sikes 4b (SRSC). PRESIDIO Co: e. slope, Goat Mt., 1 Sept. 1959, L. Campsey 315 (SRSC) ; e. slope, Goat Mt., 22 Aug. 1959, Campsey 1060 (SRSC); below San Estaban Lake, 12 mi. s. of Marfa, 18 Aug. 1940, Hinckley 1296 (LL, NY); Alamito Creek, below San Estaban Lake, 26 June 1964, Powell 1237 (SRSC); bluffs, s. w. edge of Marfa Flats, ca. 13 mi. s. of Marfa, 26 June 1964, Powell 1238 (SRSC) ; n. w. side of Goat Mt., 23 Sept. 1965, Sikes 30 (SRSC); s. side of Goat Mt., 11 June 1966, Sikes 48 (SRSC); San Estaban Canyon, 28 May 1957, Warnock & W. D. McBride 14487 (LL, SRSC); below San Estaban Lake, 24 June 1964, Warnock 20411 (SRSC); Goat Mt., 5 Aug. 1941, Warnock 21338 (SMU, TEX). Perityle rupestris shares its closest relationship with P. quinqueflora. This subject is discussed further in connec- tion with the latter species. The primary distinguishing features of P. rupestris var. rupestris include (7) 8-12 flow- ered heads, and serrate-lobed leaves which are pilose to hirsute. Perityle rupestris var. albiflora is readily delimited by its white flowers. As a relatively wide-ranging species P. rupestris is one of the most variable in the genus, being comprised of many populational and cytological entities. In distribution the species extends from the north edge of the Davis Mts. south 86 Rhodora [Vol. 71 to the Chisos Mts. in Brewster Co., and eastern Presidio Co. Practically every disjunct mountain segment in this general area (those with rock bluffs) serves as the habitat for a population of P. rupestris which is different either in ploidy level or a few morphic features. However, geographically and morphologically only two varieties seem worthy of recognition. No mixed populations of yellow and white- flowered plants have been found on either side of the imagi- nary line seen in Figure 19. One small sympatric colony of predominately white-flowered plants was discovered in an intermediate locality which is denoted in Figure 19 along the separatory imaginary line. At least three other populations of Perityle rupestris are reasonably distinct morphologically, cytologically, and geo- graphically, but at the present, it seems that the differences are not sufficient to allow for formal recognition. Locations of the three populations are noted in Figure 19, and ex- amples of each are as follows: 1) Sikes 45, Jeff Davis Co., n. slopes of Timber Mt. above Madera Springs; flowers yellow, large heads, large, subreniform leaves, chromosome number, n — ca. 34. 2) Powell 1408, Brewster Co., 17 mi. s. of Marathon; flowers white, restricted to Caballos Novaculite expo- sures in the Marathon Basin, chromosome number, n = 17. 3) Powell 1237, Presidio Co., 9 mi. s. of Marfa, below San Estaban Dam; flowers white, leaves large, chromosome number, n = 34. Several other polyploid populations of both yellow and white-flowered varieties have been discovered (Table I), but these do not deviate notably from the typical morphol- ogy of their respective varieties. Many individual plants of most populations have been checked for their chromo- some number, and the ploidy level has been shown to be consistent in each population, or at least in each cluster of plants examined. However, two qualified exceptions to the latter condition have been detected. In Madera Canyon of Jeff Davis Co., tetraploids (n = 34) and diploids (n = 17) 1969] Perityle — Powell 87 Figure 19. Distribution of diploid and polyploid populations of Perityle rupestris: Open circles (n = 17); closed circles (m = 34); triangle (n = 68); and square (n = 85). The broken line desig- nates a uniform disjunction of yellow-flowered populations (P. rupes- tris var. rupestris) to the north and white-flowered populations (P. rupestris var. albiflora) to the south. An arrow points to the inter- mediate location of a sympatric colony of predominately white- flowered plants. Locations of three populations with cytological and/or slight morphological differences from typical P. rupestris are denoted by voucher collection numbers and arrows. were collected about 20 feet apart. On Goat Mountain in Presidio Co., tetraploids were collected on the northwest side and diploids on the south side, and it is likely that plants with both ploidy levels are mixed at snme point along the west side of the mountain. There seems to be no pattern for the evolution of polyploid populations of P. rupestris (Fig. 19). It is clear, however, 88 Rhodora [Vol. 71 that once polyploids become established they are successful, even to the exclusion of diploids or other ploidy levels. No direct evidence has been accrued to elucidate the origin of polypoids in P. rupestris. The similar morphol- ogy of diploids and polyploids seems to justify speculation that all the polyploids in this species have arisen through autoploidy or perhaps segmental alloploidy in the sense of Stebbins (1950). There is no indication that other species of section Pappothrix have contributed to an alloploid origin of any polyploid populations of P. rupestris even thougi preliminary interspecific hybridization experiments (Powell, 1968a) with diploid species have shown that there is no stringent reproductive isolation between species. Reasonable conjecture would lead one to believe that populational isolation of a wide-spread P. rupestris began with the erosion of semi-continuous mountain systems in the Trans-Pecos region of Texas. A cursory examination of the present discontinuous topography and ecology (mainly habitat in crevices of rocks) in the Trans-Pecos indicates that interspecific and even large-scale interpopu- lational outcrossing of P. rupestris populations was prob- ably precluded. This is substantiated by both the allopatric distribution of the extant species of section Pappothriz, and in particular by the allopatry of the various different populations of P. rupestris. The anomalous populations of P. rupestris are being studied chromatographically, with some interesting results (Powell and Tsang, 1966). A comprehensive systematic study of this diploid and polyploid complex is in progress. 4. Perityle vitreomontana Warnock, Southw. Nat. 12 (4): 475-476. 1967. TYPE: TEXAS: BREWSTER Co: limestone bluffs, w. slopes of Old Blue, Glass Mountains, ca. 4500 ft., 22 Sept. 1963, B. H. Warnock and A. M. Powell 19111 (Holotype, SRSC! Isotype, TEX!). Plants 3-10 cm high, pilose, densely leafy; leaves opposite, be- coming alternate above, 0.7-1.0 (1.3) em long, 0.5-0.8 (1.2) em wide, ovate to ovate-deltoid, the margins obtuse-serrate to incised-lobed, rarely laciniate, typically obtuse, rarely subacute at the apex, sub- 1969] Perityle — Powell 89 truncate to subacute at the base, pilose, strongly veined; petioles 2-4 mm long; inflorescence typically of loosely clustered heads, rarely solitary, borne on short peduncles, (0.1) 0.2-0.3 (0.8) em long; heads narrowly campanulate, 6-7 mm long, 3-4 mm wide, often mostly obscured by leaves; receptacle essentially flattened to slightly convex, 1-2 mm wide; bracts (9) 10-12, 4-5 mm long, 0.8-1.0 mm wide, obscurely keeled; disc flowers 11-16; disc corollas white, 4.2-4.5 mm long, the tube 0.8-1.0 mm long, minutely glandular-pubescent, the throat tubular, ca 2.5 mm long, sparsely glandular-pubescent, the lobes narrow, acute, reflexed, 0.8-1.0 mm long; achenes 1.9-2.0 mm long, flattened, typically truncate at base and apex; pappus typically of 2-(3) very slender bristles, 1.2-2.0 mm long, often with 3-6 bristles, most shorter than the main 2, rarely with 10-20 short to vestigial bristles and nubs, the bristles minutely antrorse-ciliate; anthers ca. 2.0 mm long; chromosome number, » — 17. A rare edemic found only on limestone bluffs in the Glass Mountains, north and north-west side, eastern Brew- ster County (Fig. 17). Representative Specimens: TEXAS: BREWSTER Co: limestone bluffs, n. side, Glass Mts., S. Sikes 110 (SRSC); on Baldy Peak, Glass Mts., 13 July 1940, B. H. Warnock W247 (SRSC, TEX, UC); w. slopes of Gilliland Peak, Glass Mts., 5400 ft., 29 Aug. 1947, B. H. Warnock 6953 (SRSC); w. slopes of Old Blue, Glass Mts., 4700 ft., 27 July 1957, Warnock and W. D. McBride 15663 (LL, SRSC); near top of Old Blue, Glass Mts., 6500 ft., 15 Aug. 1962, Warnock 20405 (SRSC) ; Old Blue, Glass Mts., 5000 ft., 18 June 1964, Warnock 20412 (SRSC). This remarkable species exhibits a superficial relationship with Perityle bisetosa. The small habit, white-flowered discoid heads, and 2 main pappus bristles of P. vitreo- montana are features which resemble those of P. bisetosa. However, leaf and flower morphology indicate a close affinity of P. vitreomontana with P. rupestris var. albiflora, and a close study of the pappus of P. vitreomontana shows a structure which is reduced and derived from the section Pappothrix type. Characteristically there are 2 or 3 prominent bristles of the pappus, but in each case vestigial nubs are present between the main ones. By examination of several individuals, and often in different florets of the same head, one may observe a larger number of prominent bristles, 4-10, and rarely even 10-20, Vestigial nubs are almost always present when less than 10-15 rather well- 90 Rhodora [Vol. 71 formed bristles predominate. Perityle vitreomontana is regarded, therefore, as belonging with the section Pappo- thrix, a disposition also supported by general morphology and chromatographic data (Fig. 9; Powell & Tsang, 1966). 5. Perityle cinerea (A. Gray) Powell, sida 3: 278. 1968. Fig. 20. Laphamia cinerea A. Gray in Torr., Bot. Mex. Bound. 82. 1859. TYPE: TEXAS: on rocks near Escondido Creek, Sept., Bigelow s. m. (Holotype, GH! Isotypes, NY! Us! Photograph, ps!). Pappothrix cinerea (A. Gray) Rydb., N. Amer. Fl. 34: 27. 1914. Plants 8-25 (45) em high, tomentose-canescent, densely leafy; leaves opposite below, alternate above, 1.2-2.5 em long, 0.9-1.8 em wide, broadly ovate to subdeltoid, rarely subreniform, the margins deeply serrate with lobelets obtuse to acute, obtuse to acute at the apex, subtruncate to subacute at the base, tomentose-canescent, but veins still conspicuous; petioles 0.6-1.0 em long; inflorescence of solitary heads or clusters of heads borne on rather stout peduncles, 0.3-1.0 em long; heads narrowly campanulate, 0.8-1.0 cm long, 4-7 mm wide, sometimes partially obscured by subtending leaves; receptacle flat- tened, 1.5-2.0 mm wide; bracts 8-10, 5-7 mm long, 1.1-2.0 mm wide, keeled to obscurely so, tomentose, inner series essentially like the outer except more membranous along the margins and less pubescent; disc florets 10-15; disc corollas white, 5.3-6.0 mm long, the tube 1.3-1.6 mm long, minutely glandular-pubescent, the throat tubular, 2.8-3.2 mm long, sparsely glandular-pubescent, the lobes narrow, acute, re- flexed, 0.8-1.2 mm long; achenes 2.6-3.0 (3.5) mm long, partially flattened; pappus of 14-20 slender bristles, the bristles unequal in length, but most are subequal, minutely antrorse-ciliate; anthers (2.3) 2.5-3.0 mm long; chromosome number, n = ca. 68. Rare in crevices of Cretaceous limestone caprock of mesas, Spring-Fall (Fig. 17). Representative Specimens: TEXAS: Pecos Co: on rocks near Escondido Creek, Sept., Bigelow s. n. (GH, NY, US); 6 mi. e. of Bakersfield, April, M. Mc- Kenzie 128 (SRSC); 7-Mile Mesa, near Ft. Stockton, 15 Sept. 1964, A. M. Powell 1310 (SRSC); cracks in caprock, 17 July 1943, B. C. Tharp 43-965 (TEX, US) ; limestone caprock and on top of mesa, 1 July 1943, Tharp 43-966 (ND, POM, RM, SRSC, TEX, US) ; 7-Mile Mesa e. of Ft. Stockton, 16 Aug. 1962, Warnock 20403 (sRsc); Ft. Stockton, 14 June 1907, H. H. York 291 (Ds, SRSC, TEX). UPTON Co: ca. 10 mi. s. of Rankin, Noltke Hill, 14 Sept. 1961, Correll & Johnston 24607 (LL). Plants of this species are strikingly cinereous. Individuals of a single population may be suffruticose and rather tall 1969] Perityle — Powell 91 Figure 20. Habit sketch of Perityle cinerea (Powell 1310); X 1. or very short and appressed to rocks, at least in some in- stances depending on whether or not they are in reach of sheep which browse the plants. Perityle cinerea appears to be most closely related to P. vitreomontana or the white- flowered variety of P. rupestris, notwithstanding the to- mentose-canescent pubescence which is not approached by any other species of the genus except the Guadalupe Island endemic, P. incana. The type locality of Escondido Creek, now called Tunis Creek, is about 30 miles east of Fort Stockton. No plants of this species could be found in the small rock bluffs along 92 Rhodora [Vol. 71 the creek, and this is not a typical habitat for P. cinerea. Therefore, it appears that the original collection was made somewhat near there but hieher up on the mesas which abound in the area along the route taken by Bigelow’s party. The distribution of Perityle cinerea overlaps that of P. angustifolia (sect. Laphamia) along the Cretaceous cap- rock exposures in eastern Pecos Co. Plants of these two species have been observed growing side-by-side (but not extensively intermixed) about 6 miles east of Bakersfield. No interspecific hybridization has been detected, but fertile progeny would be unlikely since the chromosome number of P. cinerea is n = ca, 68 as compared to n = 17 for P. angustifolia, DEPARTMENT OF BIOLOGY SUL ROSS STATE COLLEGE ALPINE, TEXAS 79830 LITERATURE CITED ALSTON, R. E. and B. L. TURNER. 1963. Natural hybridization among four species of Baptisia (Leguminosae). Amer. J. Bot. 50: 159-173. AXELROD, D. I. 1958. Evolution of the Madro-Tertiary Geoflora. Bot. Rev. 24: 433-509. EvERLY, M. L. 1947. A taxonomic study of the genus Perityle and allied genera. Contrib. Dudley Herb. 3: 375-396. GRAY, A. 1852. Plant. Wright. I. 99-100. Kvnos, D. W. 1965. The independent aneuploid origin of two species of Chaenactis (Compositae) from a common ancestor. Evol. 19: 26-43. LEWIS, H. 1967. The taxonomic significance of autopolyploidy. Taxon 16: 267-271. Lewis, W. H. 1967. Cytocatalytie evolution in plants. Bot. Rev. 33: 105-115. MAaCBRIDE, J. F. 1918. Reclassified or new Compositae, chiefly North American Helenieae. Contrib. Gray Herb. n.s. 56: 37-39. POWELL, A. M. 1968a. Additional discussions pertaining to the con- generic status of Perityle and Laphamia (Compositae). Sida 3: 270-278. 1968b. Chromosome numbers in Perityle and re- lated genera (Peritylanae-Compositae). Amer. J. Bot. 55: 820- 828. 1969] Perityle — Powell 98 1969. Taxonomy of Perityle Benth. (Compositae- Peritylanae). II. Section Laphamia A. Gray. (in manuscript). , and J. AVERETT. 1967. Chromosome numbers of Chamaesaracha (Solanaceae) in Trans-Pecos Texas and adjacent regions. Sida 3: 156-162. , and H. S. TsANG. 1966. Preliminary chromato- graphic studies of Laphamia and related genera of Compositae. Southw. Nat. 11: 190-195. RAvEN, P. H. 1964. Catastrophic selection and edaphic endemism. Evol. 18: 336-358. RYDBERG, P. A. 1914. Carduaceae-Helenieae-Peritylanae. N. Amer. Fl. 34: 11-27. SELLARDS, E. H., W. S. ADKINS, and F. B. PLUMMER. 1958. The geology of Texas — Vol. I— Stratigraphy — The University of Texas Bulletin No. 3232, The Univ. of Tex. Press. SHINNERS, L. H. 1959. Species of Laphamia transferred to Perityle (Compositae-Helenieae). Southw. Nat. 4: 204-209. STEBBINS, G. L. 1950. Variation and evolution in plants. Columbia Univ. Press, New York, 643 pp. TURNER, B. L. and M. C. JOHNSTON. 1961. Chromosome numbers in the Compositae. III. Certain Mexican species. Brittonia 13: 64-69. WELLS, P. V. 1966. Late Pleistocene vegetation and degree of pluvial climatic change in the Chihuahuan Desert. Science 153: 970-975. DISJUNCTION AND ENDEMISM IN CROTON ALABAMENSIS JOE A. FARMER AND JOAB L. THOMAS Croton alabamensis E. A. Smith ex Chapman is one of the rarest shrubs in the United States. It is known from two river valleys in central Alabama and is very narrowly distributed within this area (Map I). In the Herbarium of the University of North Carolina there is one sheet of Cro- ton alabamensis which is reported from Tennessee. The label data on this sheet is as follows: Tullahoma, Coffee County, Tennessee, T. G. Harbison 725, Aug. 10, 1899. There is no other known record of this species outside of Alabama, and due to the very distinctive appearance of the plant at all seasons of the year it is unlikely that this species occurs in Tennessee at the present time. However, anyone doing field work in the vicinity of Coffee County should certainly be on the lookout for this rare plant. The species was first discovered in 1877 by Eugene A. Smith along the Cahaba River in Bibb County, Alabama. In 1905 Dr. Roland Harper found it along the Warrior River in Tuscaloosa County. Since that time the known range has been extended very little (Farmer, 1962). The species is now known from ten sites occupying something less than 100 acres irregularly distributed over about 30 square miles; about 25 square miles in the Cahaba River drainage, and about 5 square miles in the Warrior River drainage. Within these limited areas the species forms dense thickets giving the superficial appearance of one or more large clones, but no evidence of asexual reproduction of any kind has been found in this spe- cies, The distance between Croton sites on the Warrior and the nearest site on the Cahaba is approximately 25 air miles. This distance is apparently sufficient to prevent gene ex- change between the populations in the two areas even in this wind pollinated species. Thus Croton alabamensis is par- ticularly unusual in that it combines narrow endemism with 94 1969] Croton — Farmer and Thomas 95 Q ALABAMA | Z Cumberland “ani Plateau E wis | "m “TUSCALOOSA "j s". “ae? LEGEND: © WARRIOR POPULATION O CAHABA POPULATION x TYPE LOCALITY ! Map I. The distribution of Croton alabamensis. The stippled area represents Coastal Plain. Note that the distribution of this species is near, but always above, the Fall Line shown in dashes. a disjunct distribution. In addition there is an interesting variation pattern both within and between the disjunct pop- ulations. The Warrior population differs from the Cahaba population in several morphological characteristics, and there is some evidence that they differ physiologically as well. 96 Rhodora [Vol. 71 ON LP? TYPICAL LEAF OF WARRIOR POPULATION 1. Elliptical form 2. Emarginate apex 3. Average size: 37*9.5 (50) mm by 8019.7 (50) mm B. TYPICAL LEAF OF CAHARA POPULATION 1. Ovate form 2. Acute apex 3. Average size: 25+7.5 (50) mm by 6618.9 (50) mm C. EÚ TYPICAL LEAF OF PLANTS OP THE CAHABA POPULATION CHARACTERIZED BY UNUSUALLY SMALL LEAVES Average size: 10.47 2.47 (14) by 24.67.18 (1l) mm r 4. . . Fig. I. Leaf variation in Croton alabamensis. 1969] Croton — Farmer and Thomas 97 VARIATION BETWEEN WARRIOR AND CAHABA POPULATIONS Mature plants of Croton alabamensis vary in height from about 1/2 to 3 1/2 m averaging about 11/2 m. In the War- rior population the mature shrubs average about 20% taller than those in the Cahaba population. Several differences are apparent in the size and shape of the leaves (Fig. I). The average leaf blade size for Warrior plants is 37 + 9.5 (50) mm. The largest single leaf found in this population measured 75 by 170 mm. For the Cahaba population the plants can be divided readily into two categories based on leaf size as indicated in Fig. I. Most of the plants are like those indicated by B in Fig. I, where the average size was 25 + 7.5 (50) mm by 66 + 18.9 (50) mm; whereas for the fewer, extremely small-leaved plants, (Fig. I C) the aver- age leaf size was found to be only 10.4 + 2.47 (14) mm by 24.6 + 7.18 (14) mm. Leaf shape in the Warrior plants also differs consistently from that of Cahaba plants. Leaves of Warrior plants are elliptical with an apex which varies from slightly to distinctly emarginate; those from Cahaba plants are more ovate with an acute apex which is not emarginate. Reproductive structures also show a consistent size dif- ference between plants from the two river valleys. Flowers, fruits and seeds from Warrior plants are approximately 20% larger than those from Cahaba plants. It was also observed that in the Cahaba population relatively more branches bear predominantly pistillate inflorescences, as compared with approximately equal numbers of staminate and pistillate inflorescences found on Warrior plants, Nor- mal time of first flowering for the species is within the first two weeks of March, but Cahaba flowering usually pre- cedes Warrior flowering by about a week, Erratic occur- rence of a few flowers at almost any time of year is not uncommon but was observed to occur with greater fre- quency among Cahaba plants. CYTOLOGY Previously it was thought that the plants from these two river valleys differed in chromosome number, the Warrior 98 Rhodora [Vol. 71 Fig. II. Metaphase II configurations of the two populations of Croton alabamensis. Above, Cahaba population; below, Warrior pop- ulation. 1969] Croton — Farmer and Thomas 99 material having n = 32, the Cahaba material n = 16. (Far- mer 1962, Webster 1967). This correlated very nicely with the size differences noted above. Recent studies have shown that the situation is not that simple. During February and March 1968, ehromosome counts were made on 15 different individuals from the Warrior population, and 18 different individuals from three different sites in the Cahaba drain- age. The latter included individuals representing most of the morphological variation described above. All of the in- dividuals examined from both populations were found to have the number n — 32. This is not to imply that the original counts were in error, although no 16-chromosome plants were found in this survey of 33 individuals from four different sites. The authors are convinced that at least the Cahaba populations contain individuals with the num- ber » = 16 as well as individuals with n = 32, although the latter number apparently occurs with much greater frequency. The lower chromosome number previously re- ported for Cahaba material was seen in both mitotic and meiotic divisions. The mitotic counts were made on radicles of germinating seeds, the meiotic counts on pollen mother cells. The counts made during the spring of 1968 were primarily on pollen mother cells, although one count from each river system was made on root tips of softwood cut- tings. Voucher specimens of three individuals from each population counted are deposited in the herbarium of the University of Alabama. For the Cahaba population the voucher specimens are Joab L. Thomas #1615, 1616, 1617; for the Warrior population, Joab L. Thomas #1628, 1629, 1630. Meiosis appeared to be normal in all of the material ex- amined, Diakinesis was rarely seen in this material, but when this stage was encountered bivalent pairing was evi- dent and no clear multivalent formations were seen. The chromosomes of Croton alabamensis are very small, and there is considerable size variation within the chromosome complement. Correlated with a difference in flower size noted between Warrior and Cahaba material, bud size and 100 Rhodora [Vol. 71 anther size at the time of meiosis is approximately 20% larger in Warrior material than in Cahaba material. For this reason, and perhaps for others, Warrior material was much easier to work with, and it was always easier to find division figures in this material. An interesting relationship was noted between division synchrony and weather conditions at the time the material was collected. Generally, in material collected during an unusually cold period a high degree of asynchrony was seen, whereas material collected a day or two following a short warm period would invariably show more synchrony. HABITAT A brief description of the habitat of Croton alabamensis is particularly interesting and relevent because an equally rare species, Neviusia alabamensis has recently been found growing along side the Croton in one small area along the Warrior River. The habitat consists of moderately-to-steep- ly-sloping terrain with shallow soil derived from and un- derlain by sandstone, shale, or limestone, The area is also characterized by high soil and air temperature and intense drought during the summer. During late summer, in par- ticular, extreme drought is a common feature of the rock- outcrop habitat. The relief of the Warrior and Cahaba areas is similar. Elevations range mostly from 300 to 400 feet above sea level. The Croton thickets in the Cahaba drainage are mostly on gentle slopes of approximately 20 degrees and are usually facing south or southwest. Those of the Warrior area are found primarily on north-facing bluffs where the slopes are estimated at greater than 45 degrees (Fig. III). The difference in exposure is thought to be primarily responsible for the generally higher soil and air temperatures recorded in the Cahaba habitat. Also, in spite of the steep slope, the Warrior habitat usually has more soil moisture due to seepage through the shale. In both areas the thickets are usually found about midway up the slopes, seldom extending either to the top or to the base of the slopes. 1969] Croton — Farmer and Thomas 101 Fig. III. Habitats of Croton alabamensis. Above, Cahaba area; below, Warrior area. 102 Rhodora [Vol. 71 Prominent outcrops of bare rock are visible throughout the area of both the Warrior and the Cahaba populations, but the rocks are of different geological ages and mineral composition in the two areas. The rocks of the Warrior area are primarily Pottsville shale and sandstone of the Pennsylvanian, whereas those of the Cahaba are mostly Ordovician limestone. Likewise soil pH shows some varia- tion between the two areas, that of the Cahaba ranging from 6.5 to 7.8, that of the Warrior from 6.0 to 6.8. Morphological and physiological characteristics of Croton alabamensis which apparently contribute to its survival in this habitat include the following. Flower buds develop and become well differentiated in late spring and are dormant before the beginning of summer extremes of drought and high temperature. The following spring the pre-formed buds flower and develop fruits rapidly, seed dehiscence usually being complete by the end of May. Heavy seed crops are produced each year, and the seeds are released by a catapult mechanism with adequate range to seed most of the available substratum. This is thought to be an im- portant factor in maintaining the locally dense populations of Croton, Because of dormancy the seeds do not germinate until early the following spring. Young seedlings rapidly become established and enter dormancy early in the sum- mer. In this manner the seeds and the seedlings survive the drought and high temperatures of late summer. Adap- tations favoring moisture conservation within the plant during these extremes include: (1) stomata on the lower leaf surface only; with the stomata, both leaf surfaces, and the stems densely covered by scale-like, peltate trichomes ; (2) drooping or wilted leaves during most of the daylight hours; (3) abscission of many of the leaves during late summer; and (4) general plant dormancy from mid-sum- mer to early spring. The shallow rooting habit is correlated with shallow-habitat-soil and facilitates rapid water ab- sorption following rains. Thus the habitat factors which are apparently most critical in determining the distribution of this species are 1969] Croton — Farmer and Thomas 103 the slope, the shallow soil over rock, and the high soil and air temperatures associated with middle and late summer drought. Croton alabamensis seems to be particularly well adapted to this apparently unfavorable habitat. AIR FORCE ARMAMENT LABORATORY EGLIN AFB, FLORIDA 32542 DEPARTMENT OF BIOLOGY, UNIVERSITY OF ALABAMA, UNIVERSITY 35486 LITERATURE CITED FARMER, JOE A. 1962. An Ecological Life History of Croton alabamen- sis E. A. Smith ex Chapman. Unpublished Ph.D. Thesis, Uni- versity of Alabama. WEBSTER, GRADY L. 1967. The Genera of Euphorbiaceae in the South- eastern United States. Jour. Arn. Arb. 48: 303-430. A DEFINITIVE NEW BOOK ON THE FLORA OF NORTHWESTERN NORTH AMERICA! From the rich experience of many weeks and months in the field working closely with the plants he has written about, coupled with thousands of hours of study in her- baria and libraries in many parts of the world, Dr. Eric Hultén has drawn the material presented in a Flora of Alaska and Neighboring Territories. This is a superb book in every respect. Without question, it will be the standard and authoritative reference concerning plants of the region it covers for many years to come. That region includes, besides Alaska, the Chukchi (Chukotka) Peninsula in north- eastern Siberia, the Yukon, extreme northwestern British Columbia and the Mackenzie District of the Northwest Territories, Canada. A distinctive feature of this book is the fact that each species or infraspecific taxon is individually illustrated and maps of its range within the area covered and of its total geographical range are given. The line drawings of each ultimate taxon were prepared by the well-known botanical artist Mrs. Dagny Tande-Lid. These are botani- cally accurate and artistically pleasing. There are eight full pages of color including forty-nine photographs, all of excellent quality. These are in one section near the middle of the book. They show dramatically some of the varied environments available to plants in the area and some of the more spectacular and interesting plants of the flora, Long descriptions of the individual taxa are not a feature of Hultén's book and we view this as a step in the right direction. The short descriptions, illustrations and maps are adequate to fix the identity of each taxon, Keys to the families, genera, species and infraspecific taxa are com- plete. Some 1,974 taxa belonging to 1,559 species are treated in the work. "Flora of Alaska and Neighboring Territories: A Manual of the Vascular Plants, by Eric Hultén. Stanford University Press, Stan- ford, Calif. xxii -- 1008 pp. illus. $35. 104 1969] Book Review — Rollins 105 Some very involved taxonomic problems are present in the flora dealt with by Hultén. The species of some genera are largely apomictic and in other genera there is an un- usual amount of hybridization. In other instances, such as the genus Carex with one hundred and twenty taxa, it is the complexity and great reticulate divergence that makes a satisfactory taxonomic treatment difficult. Thirty-four species of Draba are treated and here is one case where this is the only satisfactory taxonomic treatment of the genus for the area. One is impressed again with the enormous number of herbaceous species compared to the woody plants of the flora and with the relatively high pro- portion of monocotyledons vs. dicotyledons. This, of course, is expected in an arctic and subarctic region. But there are large numbers of species in some genera that are rather unexpected. For example, twenty-two taxa of Pedicularis seem somewhat unreal when considered from the viewpoint of representation in the rest of North America. Eric Hultén has studied the plants of the North circum- polar region assiduously and with perseverance for nearly fifty years. He began with an expedition to the Kamchatka Peninsula in 1920. Later he made numerous collections in the Aleutian Islands and the islands of the Bering Straits. His field work included extended periods along coastal Alaska, in the interior of that State and throughout north- western Canada as well as along the Arctic shores of both Canada and Alaska. In fact, his field research has con- tinued down through the last few years right to the present, in spite of his retirement status. I remember and still appreciate the wonderful stories Hultén told of his experiences in the unexplored part of Kamchatka, not the least of which were those concerning his unintentional involvement in the dying phases of the Communist Revolution then in progress in southern Kam- chatka. With other graduate students in the 1930’s, we were privileged to hear accounts of his botanical exploits in the Aleutian Islands and in Northwest America. At the time, he was at the Gray Herbarium studying the collections he had brought back from these remote places. 106 Rhodora [Vol. 71 In building toward the book under review, which in a way is a culminating presentation of a major aspect of the basic findings of a lifetime. Dr. Hultén has proceeded in a very deliberate and resourceful way. His early pub- lication on the Flora of Kamchatka was followed by a book on the Flora of the Aleutian Islands. Then followed an “Outline of the History of Arctic and Boreal Biota during the Quaternary Period". A major contribution was his “Flora of Alaska and Yukon” published between 1941 and 1950. The Amphi-Atlantic Plants and their Phytogeogra- phical Connections, published in 1958, treated many arctic or subarctic taxa pertinent to the present work and the relevance of the more recent “Circumpolar Plants” is at once apparent. All of these were major contributions and show in a very positive way the full depth of the author’s background for writing the book at hand. The volume is excellently produced with a good quality of paper and a pleasing format. Although the cost may make it prohibitively expensive for many individuals, libraries and institutions having any interest at all in the plant cover of a large segment of North America cannot afford to be without this highly significant book. REED C. ROLLINS GRAY HERBARIUM OF HARVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS 02138 A NEW WILDFLOWER GUIDE Field Guide to Wildflowers of Northeastern and North Central North America. Roger Tory Peterson and Margaret McKenny. Hough- ton Mifflin, 1968. $4.95 In the past decade there has been an increasing interest by the public in the identification of wild flowers, with con- sequent demand for a useful popular guide. Several success- ful bird guides have been published and it has seemed that the format of these could be imitated to produce a useful flower guide. The problem has turned out to be difficult. A standard bird guide lists all species east of the Rockies in about 300 pages. For a guide to be successful, financially, the publishers demand that as large a territory as possible be covered. The standard now seems to be the United States west to the Dakotas and south to North Carolina. This guide in addition includes parts of Ontario and Quebec. Even excluding grasses, sedges and other such plants, there are very many species in a territory of this size and most guides omit many, including some quite common ones. Thus Wherry includes 500 species with 428 illustrations. Peterson includes almost 1300 species with 1344 illustra- tions, which mean that anyone trying to identify a flower, especially in the northeastern part of the country, prob- ably will find it listed in the guide. To achieve this admir- able coverage in a pocket size guide, the descriptions are necessarily brief. Included are range, habitat, size and one or two key points. Those wishing a more complete descrip- tion should refer to a more detailed guide. As stated in the introduction, the book is designed for those people who are unable to use keys, but belong to the picture-matching school, which is the average amateur's approach to plant identification, and, except for checking the range, is about as far as most will go. The system is based on (1) color, (2) general shape or structure, (3) distinctions between similar species. For this last purpose only a leaf of the plant may be shown as in the case of some of the blue violets. Color is the basic approach, and to adhere to it the traditional procedure of arrangements 107 108 Rhodora [Vol. 71 by families is abandoned. Thus the first part of the book consists of white flowers. Here for example are found all white orchids in one section, the wide-leaved orchids on one page, the spurred orchids on the next and the showy on the third. Cypripedium acaule appears here and also in the pink section of the book. In various other cases where there may be color confusion, flowers are listed twice. The color sections after white, are yellow, pink to red, violet to blue, orange, green and brown. Clearly in some cases a user would have to check 2 different color sections to locate a plant. All but a few of the plants are illustrated but many of the illustrations are line drawings with the color indicated at the top of the page and in the text. No doubt many users would prefer color for all illus- trations, which, however, would presumably make the book much more expensive. The procedure of having one family spread through several sections of the book seems strange at first sight. In practice, however, it is quite workable. A user, aware he had a Legume, would refer to the front of the book where there is a brief description of the Pea Family and page listings for the various colors which are readily checked. A user with Desmodium canadense should locate it without difficulty on the appropriate page and would probably correctly identify it. If he happened to have one of the other 9 Desmodiums listed he might well have one difficulty in making a determination although an effort is made to indicate key points. Of necessity a number of the Desmodiums are omitted but the selection seems to be rea- sonable. An effort is made to make the user aware of the significance of families. For each species, the family is indicated with a pictogram designed to call attention to a characteristic of the family. For the Legumes a generalized pea-like flower is used. The symbolism is varyingly suc- cessful, but the perceptive user is likely to acquire some appreciation of family. Amateurs demand common names for plants. Many have become standard but in other cases it has been necessary 1969] Book Review — Steele 109 to supply them. The author has in some cases translated the scientific name and in others has endeavored to devise an appropriate one. It is to be hoped that widely recog- nized common names will come into general use for flowers as they have for birds and most trees and shrubs. Some users will not find the color key always satisfactory with plants of intermediate hue being listed under a less familiar shade. The distinctions between closely related species, as in the Violets, are not always workable. These are minor criticisms. The book is excellent and sets new standards for popular guides. It also would be of consider- able use to the professional botanist who might be working with unfamiliar groups in new territory. FREDERIC L. STEELE ST. MARY’S-IN-THE-MTS. LITTLETON, N. H. 03561 CONTRIBUTIONS TO THE MARINE ALGAE OF NEWFOUNDLAND:. ° By ARTHUR C. MATHIESON, CLINTON J. DAWES and HAROLD J. HUMM INTRODUCTION Relatively little is known of the benthonic marine algae of Newfoundland. De la Pylaie (1824, 1829) listed several collections from the Island, but as suggested by Taylor (1954) his catalogue is too old to be currently useful. The only recent account of marine algae from Newfoundland is that given by Wilce ( 1959) who collected at four locations on the northwest and central west coasts. His collections were made while enroute to the Labrador Peninsula. There are so few other published records (see Taylor, 1957 for listings) that the composition of the Newfoundland marine flora has been inferred from collections of the nearby islands of St. Pierre and Miquelon (Le Gallo, 1947, 1949). The present paper summarizes the species found during two trips (1948 and 1967) to Newfoundland. Collections were made from a variety of coastal environments. During the month of August, 1948, a survey of the dis- tribution and abundance of algae of potential economic value was made by Humm for the Newfoundland Industrial De- velopment Board. A general collection was made and field notes were kept. In July and August, 1967, Dawes and Mathieson made extensive collections and field observations along the coast of N ewfoundland. The second trip was con- ducted in conjunction with a survey of algae of economic importance, and it was sponsored by Marine Colloids Inc. ‘Published with the approval of the Director of the New Hampshire Agricultural Experiment Station as scientific contribution number 462. "Contribution No. 37 from the Department of Botany and Bacteriol- ogy, University of South Florida. 110 1969] Marine Algae — Mathieson, Dawes & Humm 111 PROCEDURES 1948: Since only the month of August was available for field work the areas visited were limited to the more popu- lated and accessible localities of the eastern coast, partic- ularly the Avalon and Burin Peninsulas, and to the portion of the north coast from Cape Freels to Gander Bay — in- cluding Fogo Island and New World Island (Fig. 1). The portions of the eastern coastline with an accessible road were covered by automobile. Stops were made at each cove, village or coastal area where there was accessible shallow water and where a small boat could be obtained. The coastal areas included the lower part of the Burin Peninsula, the eastern side of Placentia Bay from Argentia to St. Bride’s, the eastern side of St. Mary’s Bay, Trepassey Bay, the east- tern side of the Avalon Peninsula, Conception Bay and the eastern side of Trinity Bay (Fig. 1). The 38-foot motor vessel Duckhawk was used to visit sta- tions in upper Placentia Bay and the western side of St. Mary’s Bay, including many islands in both and the coast- line and islands along the eastern portion of the north coast between Cape Freels and Notre Dame Bay. Most of these areas were inaccessible except by boat. Stations were ex- amined and representative collections were made from a dory which was launched from the larger vessel. Station descriptions are given for each location where collections were made. Herbarium voucher specimens from the sum- mer collections of 1948 are in the personal herbarium of H. J. Humm and the Herbarium of Duke University. 1967: The locations visited during 1967 were selected on the basis of previous observations (i.e. the 1948 field notes of H. J. H.) and (or) after examination of hydrographic charts. Most of the locations were shore sites, but a number of offshore islands were studied in Conception Bay by means of a fishing vessel (Fig. 1). A skiff and outboard motor were available for use at all stations. Collections were made at most locations in the littoral zones (on foot) and sublit- toral zones (by diving with a snorkel or hooka apparatus). A record of surface water salinity and temperature was 112 Rhodora 1 NEWFOUNDLAND STATIONS T T T T I miles miles o 40 o 300 | CANADA DON d Ne etoundiend Quebec MALLA 4) CA K u Su, 48 o "d S ` ` o* ` S e Here bey i Y Q i 7 Atlentic Ocean "Pl i w^ : D z Netre Deme Md à Bey 1 » < N WI s tv. Beneviste Bey J Ch: ba UT `> Se : FOUNDLAND V ' : NEW é : TH wj LJ f ^ A HL we Y N fe 1 en k . ». 27^ f i ' et T wait 27, if ) H P owe Ë i ^ 2." H s E pr T * eit “ et 3 T. . S M bey fes . 1948 o 1948 & 1967 . 1967 F ATLANTIC OCIAN " L se A ñ L °°. + i d Map of Newfoundland showing sites of collections. Fig. 1. 1969] Marine Algae — Mathieson, Dawes & Humm 113 made at most stations with a thermometer and a set of hy- drometers. All of the field data were recorded at the site with a portable tape recorder. Representative collections were made of all the conspicuous species at each station. The collections were preserved (4% formalin) for later identification and preparation of herbarium specimens (about 900 herbarium voucher specimens). All of the 1967 collections from the Southern Shore, Trepassey Bay, St. Mary’s Bay, Placentia Bay, and Fortune Bay are deposited in the Herbarium of the University of South Florida, Tampa (see Description of Stations for specific details of locations, dates, etc.) ; all other 1967 collections are deposited in the Herbarium of the University of New Hampshire, Durham. A variety of references was consulted for identification of species and determination of presently known distributions along the Atlantic Coast of North America (Adey, 1964, 1965, 1966; Blomquist and Humm, 1946; Bornet and Flah- ault, 1886-88; Cardinal, 1964, 1965, 1966, 1967a, b, c, d; Collins, 1909; Edelstein and McLachlan, 1966, 1967a, b; Edelstein, McLachlan and Craigie, 1967; Fritsch, 1935, 1945; Gomont, 1892; Humm, in press; MacFarlane and Milligan, 1965; Scagel, 1966; Taylor, 1957; Widdowson, 1964; Wilce, 1959, 1965; Wulff et al. 1968; Zaneveld, 1966). The nomenclature of the recent British Checklist (Parke and Dixon, 1964) has been applied in most cases. The ver- tical distribution of the major algal components described in this article is according to the biological classification of Lewis (1964). DESCRIPTION OF THE AREA Newfoundland, an island of 42,734 square miles and a province of Canada, is located off the East Coast of North America, northeast of the Gulf of St. Lawrence between 47° and 52° north latitude and 52° and 59° west longitude.* The coast is rocky (often precipitous) and usually lacks sandy beaches. In general, there is not much shallow water. The *The province is properly designated as Newfoundland and Labra- dor; the Island as Newfoundland. 114 Rhodora [Vol. 785 coastal waters are remarkably clear with visibility usually better than 50 feet. Although Newfoundland is farther south than the British Isles, its climate and surrounding waters are cooler. The Labrador Current flows southward from Baffin Bay along the coast of Labrador and along the eastern shore of New- foundland to the Grand Banks southeast of the Island where it converges with the Gulf stream which flows east-north- east. A small branch of the Labrador current leaves the main current north of the Island and enters the Strait of Belle Isle, moves into the Gulf of St. Lawrence and then southwestward through Cabot Strait between the south- western tip of Newfoundland and the eastern end of Nova Scotia, These currents keep the surface waters surrounding the Island generally a little below 0? C during much of the winter season. During summer, the water temperatures vary between 14 and 18° C. The highest temperatures (20 to 229 C— in August) are found in small bays and estuaries. The surface water salinities on the open coast are lower than typical open ocean waters (i.e. they range from 32 to 33 0/00) — primarily because of melting ice. Lower salinities occur in bays, harbours and estuaries which are fed by fresh-water streams. In many cases, the fresh water forms a surface layer with low salinity and high temperature. Several types of marine algal habitats can be distinguish- ed in Newfoundland depending upon the degree of wave action: exposed open coasts; semi-exposed beaches or coves; sheltered harbours, bays, or estuaries, The open coastal areas are subjected to heavy wave action and exhibit a pro- nounced zonation of plants and animals. A variety of red, brown and green algae is present in the littoral zone; vari- ous kelps (particularly Alaria esculenta) and red algae dom- inate the sublittoral zone. The substrate in such areas usually consist of large boulders or rocky ledges. Numerous coves interrupt the open coast line. They are formed by small streams and are bordered by steep rocky cliffs. Wave action is reduced, but the scouring effects of ice action are often conspicuous. A variety of fucoids can be found in the lit- 1969] Marine Algae — Mathieson, Dawes & Humm 115 toral zone, as well as various green and brown algae. Alaria esculenta is not a dominant component of the sublittoral zone. The substrate at the headlands of the coves consists of coarse sand or pebbles with scattered boulders. Sheltered harbours, bays or estuaries form the upper arms of larger bays — e.g. Conception or Trinity Bay. The vegetation in the littoral zone is dominated by Ascophyllum nodosum, Fucus vesiculosus, filamentous brown and many green al- gae. Dictyosiphon foeniculaceus, Chordaria flagelliformis, Chorda filum, Ahnfeltia plicata, Chondrus crispus, Poly- siphonia spp. and Zostera marina dominate the sublittoral zone. Their relative abundance varies with the type of sub- strate (small boulders, pebbles, sand and silt) and the de- gree of fresh water runoff. Mudflats form the upper ends of these sheltered locations, and relatively few species are found here except fucoids and green algae. RESULTS A total of 155 species of marine algae was collected dur- ing the two trips to Newfoundland — including 10 Cyano- phyta, 36 Chlorophyta, 47 Phaeophyta and 62 Rhodophyta. The following annotated list of species includes some syn- onymy, the collection sites, habitat notes, noteworthy taxo- nomic features and records of presently known geographical distribution along the Atlantic Coast of North America. The distribution of each species is designated numerically for its occurrence in one of the following geographical areas: #1 Newfoundland northward #2 Northern New England to Newfoundland or north- ward #3 Southern New England to Newfoundland or north- ward #4 New Jersey — Maryland to Newfoundland or north- ward #5 North Carolina to Newfoundland or northward #6 South Carolina to Newfoundland #7 Tropics (at least Florida) to Newfoundland or north- ward 116 Rhodora [Vol. 785 A Description of Stations follows the List of Species. LIST OF SPECIES Cyanophyta CYANOPHYCEAE CHROOCOCCALES CHROOCOCCACEAE *Anacystis dimidiata (Kiitzing) Drouet et Daily! Occasional as an epiphyte on Rhizoclonium riparium which was growing on woodwork in the littoral zone at Forest Field. #7 ^*Not previously recorded from Newfoundland. Gomphosphaeria aponina, Kützing Mixed with Rhizoclonium riparium on woodwork in the littoral zone at Forest Field. #7 ENTOPHYSALIDACEAE Chlorogloea conferta (Kützing) Setchell et Gardner ex Gardner ( = Entophysalis conferta. (Kützing) Drouet et Daily) Found as an epiphyte on Ceramium areschougii at Ship Harbour and on Rhodymenia palmata and other algae at Marystown. #7 *Entophysalis granulosa Kützing ( = Entophysalis deusta (Meneghini) Drouet et Daily) On the intergenicula of Corallina officinalis at Dog Cove. #7 NOSTOCALES OSCILLATORIACEAE *Lyngbya confervoides (C. Agardh) Gomont Found once as an epiphyte on Enteromorpha intestinalis at O’Donnell’s Beach. #7 * Microcoleus tenerrimus Gomont Mixed with Rhizoclonium riparium on woodwork in the littoral at Forest Field. #7 * Schizothrix calcicola (C. Agardh) Gomont Found as an epiphyte on Ceramium areschougii at Ship Harbour, and on pier pilings at Big Barachois. #7 1969] Marine Algae — Mathieson, Dawes & Humm 117 * Spirulina subsalsa (Oersted) Gomont Mixed with mats of Rhizoclonium spp. at Forest Field. #7 RIVULARIACEAE *Calothrix confervicola ( (Roth) C. Agardh) Bonet et Flah- ault Found twice; once as an epiphyte on Ceramium areschougii at Ship Harbour and once as an epiphyte on Polysiphonia flexicaulis at Boswarlos (a). Probably of world wide dis- tribution, but apparently not recorded previously from Newfoundland. #7 *Calothrix crustacea (Thuret) Bornet et Flahault Found on rocks, pier pilings and on various plants in the littoral zone at Harbour Grace (c), Sweet Bay, Embree, Laurenceton and Margaree. #7 Chlorophyta CHLOROPHYCEAE CHLOROCOCCALES CHLOROCOCCACEAE Codiolum pusillum (Lyngbye) Kjellman in Foslie f. longipes (Foslie) Collins Forming blackish-green patches on rocks in the upper lit- toral zone at Renews, Ship Cove and Carbonear Island. #2 TETRASPORALES PALMELLACEAE Gloeocystis scopulorum Hansgirg Forming a blackish-green layer on rocks in the littoral zone (mixed with Ulothrix flacca) at Great Paradise. #2 ULOTRICHALES CHAETOPHORACEAE *Entocladia wittrockii Wille Found within the cell walls of Elachista fucicola, which in turn was growing on Ascophyllum nodosum in the eulit- toral zone at Great Paradise. #3 *Pilinia lunatiae Collins On a shell of Lunatia heros in the eulittoral zone at Big Barachois. #2 Pringsheimiella scutata (Reinke) Marchew Found as an epiphyte on Polysiphonia lanosa at Point Crewe and on Ceramium rubrum at Dog Cove. #7 118 Rhodora [Vol. 785 *Pseudendoclonium marinum (Reinke) Aleem et Schulz ( = Protoderma marinum Reinke in Taylor, 1957) Uncommon; found on rocks in the mid and lower eulittoral at Placentia Sound, Sweet Bay, mouth of Gander River and Gander Bay, Embree, Davidsville (b), and Glenburnie (a). #7 ULOTRICHACEAE *Stichococcus marinum (Wille) Hazen Free filaments entangled in Elachista fucicola, which in turn was epiphytic on Ascophyllum nodosum, at Great Paradise. #3 Ulothrix flacca (Dillwyn) Thuret in Le Jolis Found as an epiphyte on Fucus vesiculosus and Ascophyl- lum nodosum, and on rocks in the littoral zone at Great Paradise. #4 ULVACEAE *Capsosiphon fulvescens (C. Agardh) Setchell et Gardner Occasional on mud covered rocks in the littoral zone of pro- tected coves and bays — particularly in brackish water areas. Collected at Marystown, Milton, Lethbridge (a), Sweet Bay, mouth of Gander River and Gander Bay, Big Cove, Brown’s Arm, Laurenceton, all stations in Humber Arm, York Harbour, Boswarlos (a), Winter House, and Woody Point. #4 * Enteromorpha erecta (Lyngbye) J. Agardh Found on scattered rocks in the littoral zone of mud flats at Brown’s Arm and the mouth of the Gander River and Gander Bay. #3 Enteromorpha groenlandica (J. Agardh) Setchell et Gard- ner Found as an epiphyte on Cladophora flexuosa at Lethbridge (a) and on pier pilings at Big Barachois. #2 Enteromorpha intestinalis (L.) Link Common on rocks in the upper-middle eulittoral zone — particularly in areas of fresh water run-off. Collected at O’Donnell’s Beach, Salmonier Arm, Port au Bras, Lawn, Brigus Bay, Hant’s Harbour, Adeyton, Lethbridge (a), Sweet Bay, mouth of Gander River and Gander Bay, 1969] Marine Algae — Mathieson, Dawes & Humm 119 Davidsville (b), Noggin Cove, Embree, Brown’s Arm, Humber Arm (a,b), Benoit’s Cove, Stephenville Cross- ing, Boswarlos (a, b), Winter House and Glenburnie. #7 * Enteromorpha linza (L.) J. Agardh Occasional on rocks in the eulittoral and upper sublittoral at Trepassey Harbour (b), Point Verde, Cuslett, Garnish, Harbour Grace (c), Sweet Bay and Frenchman’s Cove. #7 Enteromorpha micrococca Kützing Found on rocks and pier pilings in the littoral fringe and upper eulittoral at Renews, Harbour Grace (c), Sweet Bay and Glenburnie. £2 Enteromorpha minima Nageli ( = Blidingia minima (Nägeli ex Kützing) Kylin) Found once on boulders in the upper eulittoral-littoral fringe at Long Beach. #7 * Enteromorpha plumosa Kützing Found entangled amongst other algae in the eulittoral zone at Marystown, and in a high, marshy tide pool at the mouth of the Gander River and Gander Bay. #7 *Enteromorpha prolifera (O. F. Müller) J. Agardh Found on rocks in the eulittoral zone at Marystown. #7 Monostroma fuscum (Postels et Ruprecht) Wittrock f. blyttii (Areschoug) Collins Present on rocks and as an epiphyte in the lower eulittoral and sublittoral at Trepassey Harbour (a), Lawn, and Hant’s Harbour. #3 Monostroma grevillei (Thuret) Wittrock On rocks in the eulittoral zone at Big Barachois. #3 Monostroma oxyspermum (Kiitzing) Doty On rocks in the eulittoral zone at Marystown. #7 Monostroma pulchrum Farlow Collected once as an epiphyte on Chondrus crispus in the lower eulittoral at Margaree. #3 *Percursaria percursa (C. Agardh) Rosenvinge Found on scattered rocks and pier pilings in the upper eulittoral zone of muddy bays — particularly brackish water areas. Collected at Noggin Cove, Marystown and Embree where it was mixed with Enteromorpha and Rhizo- clonium spp. #4 120 Rhodora [Vol. 785 CLADOPHORALES CLADOPHORACEAE Chaetomorpha atrovirens Taylor Found as entangled mats amongst various plants in the sublittoral and lower eulittoral zones at Riverhead, Sal- monier Arm, Tickles, Dog Cove, St. Mary’s, Ship Harbour and Bloomfield. Specimens listed here are distinguished according to the characteristics given by Taylor (1957). However, we are not convinced that C. atrovirens and C. linum are distinct, for there is considerable variation of color and size in natural populations of this complex. #3 “Chaetomorpha linum (O. F. Miiller) Kiitzing Found mixed with C. atrovirens in the sublittoral zone at Salmonier Arm, and entangled amongst other algae in Placentia Sound. #7 *Chaetomorpha melagonium (Weber et Mohr) Kiitzing Present on boulders and vertical ledges in the lower eulit- toral and sublittoral zones at Placentia Sound, Branch, Harbour Mille and Margaree. #4 Cladophora flexuosa (O. F. Müller) Harvey ( = C. sericea (Hudson) Kiitzing sensu van den Hoek) Present on rocks in high tide pools or on scattered rocks in the littoral zone at Riverhead, Salmonier Arm, St. Mary’s Harbour, Marystown, Lethbridge (a) and Brown’s Arm. Often mixed with C. gracilis. #7 Cladophora gracilis (Griffiths ex Harvey in Mackay) Kützing (— C.sericea (Hudson) Kützing sensu van den Hoek) In high tide pools or on scattered rocks in the littoral zone at Salmonier Arm, Bloomfield, Hant's Harbour, Lethbridge (d), mouth of Gander River and Gander Bay, Benoit's Cove, Bloomfield, and Margaree. £4 Cladophora ruprestris (L.) Kützing Entangled amongst other algae along the town front at Marystown. £3 * Rhizoclonium kerneri Stockmayer Entangled amongst other algae in the mud flat area at Marystown. #7 1969] Marine Algae — Mathieson, Dawes & Humm 121 *Rhizoclonium riparium (Roth) Harvey v. implexum (Dillwyn) Rosenvinge Found as free-floating masses or entangled mats amongst various plants throughout the eulittoral zone; however, it is often most conspicuous in high, marshy tide pools. Col- lected at Trepassey Harbour, O’Donnell’s Beach, Forest Field, Hant’s Harbour, Milton, Gull Cove, Lethbridge (a), Sweet Bay, Bloomfield, mouth of the Gander River and Gander Bay, Embree and Margaree. #7 *Rhizoclonium tortuosum Kiitzing Found as an epiphyte and on rocks in the lower eulittoral at Riverhead, St. Mary’s Harbour and Point Crewe. #4 Spongomorpha arcta (Dillwyn) Kützing Relatively common on rocks in the lower eulittoral and upper sublittoral at Bay Bulls, Ferryland, Trepassey Harbour (b), Salmonier Arm, Point Verde, Branch, Goose- berry Cove, Ship Cove, Red Harbour, Port au Bras, Gar- nish, Carbonear Island, Harbour Grace (c), Brigus Bay, Hant’s Harbour, Long Beach, Lethbridge (a,c), Bloom- field and Margaree. #4 Spongomorpha hystrix Stromfelt Found on rocks in shallow water at Ferryland and the east coast of the Avalon Peninsula. #2 Spongomorpha lanosa (Roth) Kiitzing Found once on rocks in the lowest eulittoral at Laurence- ton. #3 *Spongomorpha spinescens Kiitzing Collected twice on rocks in the lowest eulittoral at Trepas- sey Harbour and Brigus Bay. #2 Urospora collabens (C. Agardh) Holmes et Batters On woodwork in the littoral zone at Dog Cove, #2 Urospora penicilliformis (Roth) Areschoug Found once on a pier piling at Harbour Grace (c) ; mixed with various blue green algae in the littoral fringe and up- per eulittoral. #4 122 Rhodora [Vol. 785 Phaeophyta PHAEOPHYCEAE ECTOCARPALES ECTOCARPACEAE Ectocarpus confervoides (Roth) Le Jolis ( = E. arctus Kützing in Parke and Dixon, 1964) Common as an epiphyte on various plants in the eulittoral and upper sublittoral — less common on rocks; collected at Ferryland, Trepassey Harbour (a,b), St. Mary's, Sal- monier Arm, Point Verde, Red Harbour, Port au Bras, Garnish, Spread Eagle, Kelly's Island (b), Bell Island, Carbonear Island, Harbour Grace (c), Brigus Bay, Hant's Harbour, Long Beach, Shoal Harbour, Lethbridge (c), mouth of Gander River and Gander Bay, Laurenceton, Humber Arm (b), Frenchman's Cove stations and Woody Point. #7 *Ectocarpus fasciculatus Harvey Found as an epiphyte on various plants in the lower eulit- toral-upper sublittoral at St. Mary's, Point Verde, Goose- berry Cove and Margaree. £4 Ectocarpus siliculosus (Dillwyn) Lyngbye Found occasionally as an epiphyte on Chorda filum and Chordaria flagelliformis in the lower eulittoral-upper sub- littoral at St. Mary's, Salmonier Arm, Lawn, Harbour Grace (a), Kings Beach, Sweet Bay, Humber Arm (b) and Frenchman's Cove (b). £7 * Ectocarpus tomentosus (Hudson) Lyngbye Found as an epiphyte and entangled amongst other algae in the eulittoral and sublittoral at Trepassey Harbour (b), Salmonier Arm, Placentia Sound, Spread Eagle and York Harbour. #4 *Giffordia granulosa (J. E. Smith) Hamel On rocks in the sublittoral zone at Dog Cove. #4 *Giffordia ovata (Kjellman) Kylin On Chordaria flagelliformis from the mainland west of Chamber’s Island. #3 *Giffordia sandriana (Zanardini) Hamel 1969] Marine Algae — Mathieson, Dawes & Humm 123 Found once on rocks in the eulittoral zone at Trepassey Harbour (b). #3 Pilayella littoralis (L.) Kjellman A common epiphyte on Ascophyllum nodosum and Fucus spp. in the mid-lower eulittoral ; occasionally on rocks. Col- lected at Bay Bulls, Ferryland, Trepassey Harbour (a, b), St. Mary's, O’Donnell’s Beach, Salmonier Arm, Point Verde, Perch Cove, Branch, Gooseberry Cove, Port au Bras, Grand Bank, Foxtrap, Harbour Grace (c), Brigus Bay, Hant's Harbour, Long Beach, Adeyton, Lethbridge (a), Sweet Bay, Bloomfield, Big Cove, Laurenceton, Hum- ber Arm (b), Stephenville Crossing, Frenchman's Cove (b) and Margaree. £4 RALFSIACEAE * Ralfsia clavata (Harvey in Hooker) Crouan frat. Found once on the intergenicula of Corallina officinalis, which was growing in the upper sublittoral at Kelley's Island (b). £4 Ralfsia fungiformis (Gunner) Setchell et Gardner Found once on rocks in the lower eulittoral at Lethbridge (b) mixed with Ralfsia verrucosa. #2 Ralfsia verrucosa (Areschoug) J. Agardh Common on rocks and shells (often in tide pools) through- out the eulittoral zone at Trepassey Harbour (a), Sal- monier Arm, Tickles, Little St. Lawrence Harbour, Dildo, Spread Fagle, Hant's Harbour, Adeyton, Lethbridge (a-c), Sweet Bay, Bloomfield, mouth of Gander River and Gander Bay, Embree, Big Cove, Laurenceton and Humber Arm (a). #4 SPHACELARIALES SPHACELARIACEAE Sphacelaria cirrosa (Roth) C. Agardh Found on pier pilings in the mid-low eulittoral zone at Laurenceton and as an epiphyte on Polysiphonia nigres- cens in the sublittoral at Frenchman’s Cove (b). #3 Sphacelaria racemosa Greville v. arctica (Harvey) Reinke Found once on mud covered rocks in the eulittoral at Hum- ber Arm (a). #1 124 Rhodora [Vol. 785 Sphacelaria plumosa, Lyngbye ( = Chaetopteris plumosa (Lyngbye) Kiitzing in Taylor, 1957) Found once as an epiphyte on Dictyosiphon foeniculaceus in the lower eulittoral at Laurenceton. #2 STYPOCAULACEAE Halopteris scoparia (L.) Sauvageau Found on scattered rocks (often sand covered) in the sub- littoral zone at Frenchman's Cove (b) and Boswarlos (a). #2 TILOPTERIDALES TILOPTERIDACEAE *Haplospora globosa Kjellman A single gametophytic specimen was collected from a rock in the upper sublittoral at Gooseberry Cove; mixed with Porphyra miniata, Ceramium rubrum, and Rhodomela con- fervoides. #3 CHORDARIALES CHORDARIACEAE Chordaria flagelliformis (O. F. Müller) C. Agardh Common on rocks (often mud covered) in the lower eulit- toral-upper sublittoral zone; occasionally as an epiphyte in the same zone. Often mixed with Dictyosiphon foenicu- laceus and difficult to distinguish from it in the field. Col- lected at Bay Bulls, Ferryland, Trepassey Harbour (a, b), St. Mary’s, O’Donnell’s Beach, Salmonier Arm, Tickles, Point Verde, Ship Harbour, Cuslett, Gooseberry Cove, Ship Cove, Red Harbour, Port au Bras, Lamaline, Grand Bank, Garnish, North Harbour, Foxtrap, Bell Island, Car- bonear Island, Harbour Grace (c), Brigus Bay, Hant’s Harbour, Long Beach, Shoal Harbour, Lethbridge (a, b), Sweet Bay, Embree, Big Cove, Humber Arm (c), French- man’s Cove (b), Margaree, Bottle Cove and Boswarlos (a,b). #4 Eudesme virescens (Carmichael ex Harvey in Hooker) J. Agardh Found once growing on rocks in the lower eulittoral at Trepassey Harbour (a). #3 1969] Marine Algae — Mathieson, Dawes & Humm 125 *Hudesme zosterae (J. Agardh) Kylin Epiphytic on Zostera marina in the sublittoral zone at Sal- monier Arm and Spread Eagle; also on stones along the town front at Marystown. #7 Sphaerotrichia divaricata (C. Agardh) Kylin On rocks (often mud covered) and epiphytic on Fucus vesiculosus in the eulittoral and sublittoral zones of pro- tected bays at Lethbridge (a, c, d), Embree and Big Cove. #4 CORYNOPHLAEACEAE Leathesia difformis (L.) Areschoug Epiphytic on Corallina officinalis and other plants in the lower eulittoral-upper sublittoral zone at Lamaline and Margaree. #5 ELACHISTACEAE Elachista fucicola (Velley) Areschoug A common epiphyte on Ascophyllum nodosum and Fucus spp. from the mid eulittoral to the upper sublittoral at Bay Bulls, Ferryland, Trepassey Harbour (a,b), O’Donnell’s Beach, Point Verde, Parker’s Cove, Port au Bras, Harbour Mille, Spread Eagle, Kelley’s Island (b), Brigus Bay, Hant’s Harbour, Long Beach, Adeyton, Lethbridge (c), Sweet Bay, mouth of Gander River and Gander Bay, Nog- gin Cove, Big Cove, Brown’s Arm, Frenchman’s Cove (b), Margaree and Boswarlos (b). #4 Elachista lubrica Ruprecht Growing on fucoids in the sublittoral zone (on a pier piling) at Marystown. #2 MYRIONEMATACEAE: * Ascocyclus distromaticus Taylor *Found once as an epiphyte on Zostera marina in the sub- littoral at Salmonier Arm. £3 * Myrionema strangulans Greville Found as an epiphyte (on Enteromorpha erecta, Chaeto- morpha atrovirens and kelps) at the mouth of the Gander River and Gander Bay, Placentia Sound, Gooseberry Cove, Harbour Mille, Kelly's Island (b), Bay Bulls and Trepas- sey Harbour (b). #5 126 Rhodora [Vol. 785 DESMARESTIALES DESMARESTIACEAE Desmarestia aculeata (L.) Lamouroux On rocks in the sublittoral zone at Trepassey Harbour (b), Salmonier Arm, Dog Cove, Ship Cove, Garnish, Bell Island, Carbonear Island, Brigus Bay and Shoal Harbour. #4 Desmarestia viridis (O. F. Müller) Lamouroux On rocks in the sublittoral zone at Ferryland, Trepassey Harbour (a), St. Mary’s, Salmonier Arm, Dog Cove, Gooseberry Cove, Grand Bank, Harbour Mille, Spread Eagle, Foxtrap, Kelly’s Island (b), Bell Island, Carbonear Island, Hant's Harbour and Margaree, #4 DICTYOSIPHONALES DICTYOSIPHONACEAE *Dictyosiphon ekmanii Areschoug On rocks in shallow water at Big Barachois. #2 Dictyosiphon foeniculaceus (Hudson) Greville Very common on mud covered rocks in the lower eulittoral- upper sublittoral — particularly in sheltered bays. Often mixed with Chordaria flagelliformis which is difficult to distinguish in the field. Collected at Bay Bulls, Ferryland, Trepassey Harbour (a, b), St. Mary’s, O'Donnell's Beach (a), Salmonier Arm, Tickles, Point Verde, Patrick’s Cove, Placentia Sound, Gooseberry Cove, Ship Cove, Red Har- bour, Port au Bras, Little St. Lawrence Harbour, Lamaline, North Harbour, Dildo, Foxtrap, Kelly’s Island (b), Bell Island, Harbour Grace (c), Brigus Bay, Hant's Harbour, Long Beach, Adeyton, Shoal Harbour, Lethbridge (a, c), Sweet Bay, Bloomfield, mouth of Gander River and Gan- der Bay, Embree, Big Cove, Laurenceton, Humber Arm (b), Frenchman's Cove (b), Margaree, Boswarlos (a, b) and Woody Point. #4 PUNCTARIACEAE Asperococcus echinatus (Mertens) Greville Found once on rocks in a mid-eulittoral tide pool at Hant's Harbour. £3 *Delamarea attenuata (Kjellman) Rosenvinge 199] Marine Algae — Mathieson, Dawes & Humm 127 On rocks in the eulittoral zone at Bay Bulls, Point Verde, Ship Cove, Long Beach, Shoal Harbour and Margaree. #3 *Myriotrichia filiformis Harvey Found as an epiphyte on Cystoclonium purpureum from the mainland west of Chamber’s Island, and on Dictyo- siphon foeniculaceus from Placentia Sound. #4 *Punctaria latifolia Greville Found on scattered rocks in the eulittoral zone at Port au Bras, North Harbour and Dildo. #4 *Punctaria plantaginea (Roth) Greville Uncommon, on rocks in the eulittoral zone at Long Har- bour, Lamaline and Frenchman’s Cove (b). #4 SCYTOSIPHONACEAE Petalonia fascia (O. F. Miller) Kuntze On rocks (often in tide pools) and epiphytic on various algae in the eulittoral zone at Trepassey Harbour (b), Cuslett, Gooseberry Cove, Port au Bras, Lawn, Garnish, Bacon Cove; Carbonear Island, Hant’s Harbour, Long Beach and Sweet Bay. The variety zosterifolia was found once at Renews. #7 *Scytosiphon lomentarius (Lyngbye) Link On rocks (often in tide pools) and occasionally epiphytic on various plants in the eulittoral zone at St. Mary’s, Ship Cove, Grand Bank, Garnish, Foxtrap, Kelly’s Island (b) and Carbonear Island. #5 STRIARIACEAE Stictyosiphon tortilis (Ruprecht) Reinke Present on rocks and pier pilings in the eulittoral and upper sublittoral zones of the protected bays at Brown’s Arm, and Laurenceton. Mixed with Enteromorpha spp, Chordaria flagelliformis and Dictyosiphon foeniculaceus. #2 LAMINARIALES ALARIACEAE Alaria esculenta (L.) Greville (including A. musaefolia (De la Pylaie) J. Agardh sensu Widdowson, 1964). Relatively common in the sublittoral zone (usually below 3 feet) of rocky exposed coastal areas. 128 Rhodora [Vol. 785 Collected at Bay Bulls, Trepassey Harbour (b), St. Mary’s, Point Verde, Branch, Cuslett, Gooseberry Cove, Ship Cove, Red Harbour, Grand Bank, Carbonear Island, Brigus Bay, Hant’s Harbour, Long Beach, Dog Cove and Margaree. #3 CHORDACEAE Chorda filum (L.) Stackhouse Common on small rocks and shells in the sublittoral zone of sheltered bays and harbours. Collected at Trepassey Harbour stations, Riverhead, O’Donnell’s Beach (a), Sal- monier Arm, Dog Cove (adrift), Point Verde, Red Har- bour, Port au Bras, Little St. Lawrence Harbour, Lawn, Garnish, Harbour Mille, Dildo, Spread Eagle, Foxtrap, Bell Island, Harbour Grace (a), Brigus Bay, Hant’s Har- bour, Adeyton, Lethbridge (a-d), Sweet Bay, mouth of Gander River and Gander Bay, Embree, Big Cove, Brown’s Arm, Laurenceton, Humber Arm (b), Frenchman’s Cove (a,b), York Harbour, Bottle Cove, Boswarlos (b) and Woody Point. #4 LAMINARIACEAE Agarum cribrosum (Mertens) Bory Common in the sublittoral zone of exposed and semi- exposed areas below 5 feet. It doesn't appear to be eaten by urchins which were usually found in large numbers in the same areas. Collected at Trepassey Harbour (b), St. Mary's, Point Verde, Lamaline, Point Crewe, Lories, Burin Peninsula, Foxtrap, Kelly's Island (b), Carbonear Island, Harbour Grace (a), Brigus Bay, Hant's Harbour, and Margaree. £2 *Laminaria, digitata (Hudson) Lamouroux Present in the sublittoral zone of rocky exposed coastal areas. Collected at Branch, Cuslett, Gooseberry Cove, Grand Bank, Harbour Mille and Brigus Bay. Frond seg- mentation varies considerably on different specimens, but each has a consistent + — anatomy (i.e. mucilage ducts are present in the blades and absent from the stipes, Wilce, 1965). #3 1969] Marine Algae — Mathieson, Dawes & Humm 129 *Laminaria nigripes J. Agardh Found in the same exposed rocky habitat as L. digitata, but it was much less common. Collected at Margaree and Red Rock Point. It has a ++ anatomy (Wilce, 1965). #1 Laminaria saccharina (L.) Lamouroux sensu Wilce The plants designated here are of three ecotypes (Wilce, 1965), which are distinguished according to their anatomy (i.e. ++, +— and — —). They are found on rocks in the sublittoral zone of exposed or semi-exposed coastal areas. The + — ecotype (L. saccharina (L.) Lamouroux, in Taylor, 1957) was the least common of the three and it it was only found at Little St. Lawrence Harbour. The +-+ ecotype (L. groenlandica Rosenvinge in Taylor, 1957) was found at North Harbour, Carbonear Island and Margaree. The — — ecotype (L. agardhii Kjellman, in Taylor, 1957) was the most common of the three and was found at all the Trepassey Harbour stations, St. Mary’s, Riverhead, Salmonier Arm, Tickles, Ship Harbour, Lawn, Lamaline, Spread Eagle, Brigus Bay and Bloomfield. The — — and + + ecotypes (as they are designated by Taylor, 1957) have not previously been recorded in Newfound- land. #4 Saccorhiza dermatodea (De la Pylaie) J. Agardh On rocks in the sublittoral zone of exposed coastal areas. Collected at Ferryland, O’Donnell’s Beach, Point Verde, Branch, Gooseberry Cove, Ship Cove, Lamaline, Garnish, Harbour Mille, Bacon Cove, Bell Island, Carbonear Island, Brigus Bay, Frenchman’s Cove (b), Margaree and Bottle Cove. #2 FUCALES FUCACEAE Ascophyllum nodosum (L.) Le Jolis Very common on rocks from the mid eulittoral to the upper sublittoral at semi-exposed coastal and sheltered locations. Collected at Bay Bulls, Trepassey Harbour stations, River- head, O'Donnell's Beach, Salmonier Arm, Parker's Cove, Port au Bras, Spread Eagle, Foxtrap, Hant's Harbour, Adeyton, Lethbridge (a-c), Sweet Bay, Bloomfield, mouth 150 Rhodora [Vol. 785 of the Gander River and Gander Bay, Noggin Cove, Em- bree, Big Cove, Laurenceton, Humber Arm (b), Benoit’s Cove, Frenchman's Cove (b), Margaree, Boswarlos (a), Winter House, Glenburnie and Woody Point. #5 Fucus distichus (L.) emend, Powell ssp. distichus Powell On rocks in high eulittoral tide pools at Big Barachois, Garnish, Carbonear Island, Hant’s Harbour, Big Cove and Margaree. #2 Fucus distichus (L.) emend. Powell ssp. edentatus (De la Pylaie) Powell Found twice on rocks in the lower eulittoral — upper sub- littoral at Hant’s Harbour and Margaree. #4 Fucus distichus (L.) emend. Powell ssp. evanescens (C. Agardh) Powell On a vertical rock wall in the lower eulittoral at Bacon Cove. £3 Fucus vesiculosus L. Common on rocks from the mid eulittoral to the upper sublittoral at semi-exposed coastal and sheltered areas. Collected at Bay Bulls, Ferryland, Trepassey Harbour sta- tions, Riverhead, O’Donnell’s Beach, Salmonier Arm, Tickles, Point Verde, Ship Harbour, Cuslett, Parker's Cove, Port au Bras, Lamaline, Grand Bank, Garnish, Har- bour Mille, Dildo, Spread Eagle, Foxtrap, Kelly's Island (b), Carbonear Island, Harbour Grace (c), Brigus Bay, Hant's Harbour, Long Beach, Adeyton, Shoal Harbour, Lethbridge (a-d), Sweet Bay, Bloomfield, mouth of Gander River and Gander Bay, Noggin Cove, Mason's Cove, Em- bree, Big Cove, Brown's Arm, Laurenceton, Humber Arm (a,b), Benoit's Cove, Frenchman's Cove, Margaree, Stephenville Crossing, Boswarlos (a,b), Winter House, Glenburnie and Woody Point. £5 Rhodophyta RHODOPHYCEAE BANGIOPHYCIDAE GONIOTRICHALES GONIOTRICHACEAE *Asterocystis ramosa (Thwaites in Harvey) Gobi ex Schmitz. 1969] Marine Algae — Mathieson, Dawes & Humm 131 Found once epiphytic on Ceramium areschougii at Ship Harbour. #7 *Goniotrichum alsidii (Zanardini) Howe Epiphytic on various plants in the eulittoral and sublittoral zones of sheltered locations. Collected at Salmonier Arm, Great Paradise, Dog Cove, Placentia Sound, Point Verde, Lethbridge (a, c), Bloomfield, Brown's Arm, Humber Arm (b), Frenchman’s Cove (b) and Boswarlos (a,b). #7 BANGIALES BANGIACEAE *Bangia ciliaris Carmichael Found as an epiphyte on larger algae at Marystown, Top- sail Beach, Hant’s Harbour and Bloomfield. #4 Porphyra miniata (C. Agardh) C. Agardh Growing on rocks or epiphytic on selected plants (e.g. Chondrus crispus, Gigartina stellata) in the lower eulit- toral — upper sublittoral at all Trepassey Harbour sta- tions, Point Verde, Branch, Gooseberry Cove, Ship Cove, Red Harbour and Lawn. #2 Porphyra umbilicalis (L.) J. Agardh Found on rocks and epiphytic on various plants in the eulittoral zone at St. Mary’s, Branch, Green Point, Goose- berry Cove, Ship Cove, Grand Bank, Point Verde, Port au Bras, Garnish, Harbour Mille, Dildo, Harbour Grace (c), Brigus Bay, Adeyton, Sweet Bay, Humber Arm (b), mouth of Gander River and Gander Bay, York Harbour, Winter House and Woody Point. The forma epiphytica Collins was also found on Fucus vesiculosus at Sweet Bay. #5 ERYTHROPELTIDACEAE *Erythrotrichia carnea (Dillwyn) J. Agardh Found epiphytic on Ceramium areschougii at Ship Har- bour and on various algae in front of the town at Marys- town. #7 FLORIDEOPHYCIDAE NEMALIONALES ACROCHAETIACEAE * Acrochaetium amphiroae (Drew) Papenfuss 132 Rhodora [Vol. 785 Found epiphytic on Rhodymenia palmata at Point Verde and Harbour Mille. #3 * Acrochaetium attenuatum (Rosenvinge) Hamel, f. Found epiphytic on Chaetomorpha atrovirens at Dog Cove. #3 * Acrochaetium flexuosum Vickers Found epiphytic on Chaetomorpha atrovirens from Pla- centia Sound. #7 *Kylinia alariae (Jonsson) Kylin Epiphytic on Alaria esculenta at Ferryland and at Port au Bras. #2 *Kylinia compacta (Jao) Papenfuss Found once epiphytic on Cladophora gracilis at Lethbridge (a). x3 *Kylinia hallandica (Kylin) Kylin Found once epiphytic on Petalonia fascia at Trepassey Harbour (b). #1 *Kylinia moniliformis (Rosenvinge) Kylin Epiphytie on Giffordia ovata in shallow waters along the mainland shore west of Chamber's Island. #3 *Kylinia secundata (Lyngbye) Papenfuss Growing as an epiphyte on Sertularia, which in turn was on Fucus vesiculosus, at St. Mary's. #3 *Kylinia virgatula (Harvey) Papenfuss Found once epiphytic on Cladophora gracilis at Lethbridge (a). #7 * Rhodochorton purpureum (Lightfoot) Rosenvinge In shaded rock crevices at Witless Bay and Gull Cove. #3 BONNEMAISONIACEAE Trailliella intricata (J. Agardh) Batters Epiphytic on various plants in the sublittoral zone of pro- tected harbours or bays. Collected at Salmonier Arm, Point Verde, Ship Harbour, Cuslett, North Harbour, Frenchman’s Cove (b), York Harbour and Stephenville Crossing. According to Harder (1948) Trailliella intricata is the tetrasporophyte of Asparagopsis hamifera. How- ever, the latter plant is only recorded from Long Island and southern Massachusetts (Taylor, 1957). Chihara 1969] Marine Algae — Mathieson, Dawes & Humm 133 (1961) has recently given an extensive review of the life history of the group. #3 CRYPTONEMIALES CHOREOCOLACEAE *Choreocolax polysiphoniae Reinsch Parasitic on Polysiphonia lanosa, which in turn was epi- phytic on Ascophyllum nodosum. Collected at O’Donnell’s Beach, Parker’s Cove, Garnish and Margaree. #3 CORALLINACEAE Clathromorphum circumscriptum (Stromfelt) Foslie (As Phymatolithon compactum (Kjellman) Foslie in Tay- lor, 1957) On rocks and shells in the lower eulittoral and sublittoral at St. Mary’s, Salmonier Arm, Tickles, Branch, Hant’s Harbour and Margaree. #2 Corallina officinalis L. On rocks in eulittoral tide pools and in the sublittoral zone at Trepassey Harbour (b), Salmonier Arm, Tickles, Dog Cove, Point Verde, Cuslett, Gooseberry Cove, Ship Cove, Lamaline, Foxtrap, Kelly’s Island (b), Carbonear Island, Hant’s Harbour and Margaree. #2 Leptophytum laeve (Stromfelt) Adey (= Lithothamnium laeve (Strémfelt) Foslie, in Taylor, 1957) On stones in the upper sublittoral on the west side of Sound Island. #3 Lithothamnium glaciale Kjellman On rock surfaces and shells in the lowest eulittoral and sublittoral at Riverhead, O’Donnell’s Beach, Salmonier Arm, Lamaline, Foxtrap, Kelly’s Island (b), Harbour Grace (a), Hant’s Harbour and Sweet Bay. This plant forms a conspicuous residual cover in areas of heavy sea urchin grazing. #2 *Lithothamnium tophiforme Unger Found loose in shallow water at Ship Harbour, The validi- ty of this species is subject to question (Adey, 1966). #2 *Phymatolithon rugulosum Adey Found once on rocks in the sublittoral zone at Salmonier Arm. #2 134 Rhodora [Vol. 785 GLOIOSIPHONIACEAE Gloiosiphonia capillaris (Hudson) Carmichael ex Berkeley Found on rocks in the lower eulittoral — upper sublittoral at Branch and Gooseberry Cove; also found epiphytic on Chordaria flagelliformis from the mainland west shore of Chamber’s Island. #3 HILDENBRANDIACEAE Hildenbrandia prototypus Nardo On rock surfaces and shells in the eulittoral and upper sub- littoral zones at Trepassey Harbour (a), Salmonier Arm, Placentia Sound, Parker’s Cove, Spread Eagle, Lethbridge (a-c), Sweet Bay, Bloomfield, mouth of Gander River and Gander Bay, Embree, Big Cove and Humber Arm (a). #7 POLYIDEACEAE *Polyides rotundus (Hudson) Greville (= Polyides caprinus (Gunnerus) Papenfuss in Taylor, 1957) On stones in the sublittoral zone at Perch Cove. #3 SQUA MARIACEAE *Rhodophysema elegans (Crouan frat ex J. Agardh) Dixon (including Rhododermis parasitica Batters sensu Dixon, 1964) Present on rocks in the lower eulittoral zone at Sweet Bay, epiphytic on Spongomorpha arcta in the eulittoral zone at Salmonier Arm and Lethbridge (c) and at the base of Ascophyllum nodosum and on a vertical rock wall (eulit- toral) at Placentia Sound. #2 *Rhodophysema georgii Batters (= Rhododermis georgii (Batters) Collins in Taylor, 1957) Found once epiphytic on Zostera marina leaves in the sub- littoral zone at Lethbridge (d). #3 GIGARTINALES GIGARTINACEAE Chondrus crispus Stackhouse Relatively common on rocks in the eulittoral (usually the 1969] Marine Algae — Mathieson, Dawes & Humm 135 lowest part) and sublittoral zones at all Trepassey Har- bour stations, St. Mary’s, Riverhead, O’Donnell’s Beach, Salmonier Arm, Tickles, Point Verde, Ship Harbour, Cus- lett, Parker’s Cove, Red Harbour, Little St. Lawrence Harbour, Harbour Mille, North Harbour, Hant’s Harbour, Shoal Harbour, all Lethbridge stations, Bloomfield, Em- bree, Humber Arm (b), Benoit’s Cove, Margaree, Stephen- ville Crossing and Boswarlos (a). #4 Gigartina stellata (Stackhouse) Batters On rock surfaces in the lower eulittoral and upper sub- littoral at Trepassey Harbour (a) and Branch. #3 NEMASTOMATACEAE *Platoma bairdii (Farlow) Kuckuck On stones in the lower eulittoral zone at Perch Cove. £3 PHYLLOPHORACEAE Ahnfeltia plicata (Hudson) Fries Present on rocks (often sand and mud covered) in the lower eulittoral and sublittoral at Salmonier Arm, Dog Cove, Branch, Placentia Sound, Cuslett, Gooseberry Cove, Shoal Harbour, Lethbridge (a,b), mouth of the Gander River and Gander Bay, Embree, Humber Arm (b) and Boswarlos (b). #4 Phyllophora brodiaei (Turner) Endl. Found on rock surfaces in the sublittoral zone at Salmonier Arm; mixed with P. membranifolia. #4 Phyllophora membranifolia (Goodenough et Woodward) J. Agardh Found in the sublittoral zone at Salmonier Arm. #4 RHODOPHYLLIDACEAE Cystoclonium purpureum (Hudson) Batters Growing as an epiphyte and on rocks in the sublittoral zone at O'Donnell's Beach, along mainland west of Cham- ber’s Island, Point Verde, Ship Harbour, Branch, Harbour Mille and North Harbour. #4 RHODYMENIALES RHODYMENIACEAE Halosaccion ramentaceum (L.) J. Agardh Relatively common on rock surfaces in the lower eulittoral 156 Rhodora [Vol. 785 and upper sublittoral at Bay Bulls, Trepassey Harbour stations, Point Verde, Branch, Taylor Bay, Cuslett, Goose- berry Cove, Ship Cove, Red Harbour, Lawn, Harbour Mille, Bell Island, Carbonear Island, Brigus Bay, Hant's Harbour, Long Beach and Margaree. #2 Rhodymenia palmata (L.) Greville On rocks and epiphytic on kelps in the sublittoral zone at Bay Bulls, Trepassey Harbour stations, Riverhead, O'Don- nell’s Beach, Salmonier Arm, Point Verde, Branch, Cuslett, Gooseberry Cove, Ship Cove, Red Harbour, Port au Bras, Little St. Lawrence Harbour, Harbour Mille, North Har- bour, Bell Island, Carbonear, Harbour Grace (c), Brigus Bay and Margaree. £4 CERAMIALES CERAMIACEAE * Antithamnion americanum (Harvey) Farlow in Kjellman On woodwork in the eulittoral zone at Marystown. £4 * Antithamnion boreale (Gobi) Kjellman Found as an epiphyte on Desmarestia aculeata in the sub- littoral zone at Brigus Bay. £2 * Antithamnion cruciatum (C. Agardh) Nageli Epiphytic on larger algae along the town front at Marys- town. #7 Antithamnion pylaisaei (Montagne) Kjellman Epiphytic on other algae along the town front of Marys- town and at Perch Cove. #3 “Ceramium areschougii Kylin Found once in the sublittoral zone at Ship Harbour. #3 “Ceramium elegans (Ducluzeau) C. Agardh Found as an epiphyte on large algae at Marystown and Humber Arm (b). #2 *Ceramium rubriforme Kylin On Chondrus crispus along the mainland west of Cham- ber’s Island; also at Dog Cove. #3 Ceramium rubrum (Hudson) J. Agardh v. pedicellatum Duby Collected on rocks in the lower eulittoral and sublittoral at Point Verde, Branch, and Gooseberry Cove. #7 1969] Marine Algae — Mathieson, Dawes & Humm 137 *Plumaria elegans (Bonnemaison) Schmitz Mixed with Membranoptera alata around the bases of Agarum cribrosum in the lower eulittoral — upper sub- littoral zones at several locations on the Burin Peninsula. #4 * Ptilota plumosa (Hudson) C. Agardh On rocks in the sublittoral zone at Point Crewes. #2 Ptilota serrata Kützing On rocks and epiphytic on various plants (particularly the stipes and holdfasts of kelps) in the sublittoral zone at Point Verde, Branch, Brigus Bay, Hant’s Harbour, York Harbour, Stephenville Crossing and the Burin Peninsula. #4 DELESSERIACEAE * Membranoptera alata (Hudson) Stackhouse (including Membranoptera denticulata (Montagne) Kylin in Taylor, 1957). Found as an epiphyte on Desmarestia aculeata in the sublittoral zone at Brigus Bay and mixed with Plumaria elegans around the bases of Agarum cri- brosum at many locations on the Burin Peninsula. £2 Phycodrys rubens (Hudson) Batters Found once on rocks in the sublittoral zone at Brigus Bay. #4 RHODOMELACEAE * Polysiphonia arctica J. Agardh Found once on rocks in the sublittoral zone at Ship Har- bour. #2 *Polysiphonia elongata (Hudson) Greville ez Harvey in Hooker On rocks in the upper sublittoral along the mainland shore west of Chamber’s Island, and at Marystown. Fine branches were present throughout the summer. #3 * Polysiphonia fibrillosa (Dillwyn) Harvey in Hooker On rocks and epiphytic on various algae in the lower eulit- toral and sublittoral at Point Verde, Kelly’s Island (b), Sweet Bay, mouth of the Gander River and Gander Bay, Big Cove, and Humber Arm (b). #4 *Polysiphonia flexicaulis (Harvey) Collins On rocks and epiphytic on various algae in the lower 138 Rhodora [Vol. 785 eulittoral and sublittoral at Salmonier Arm, Marystown, Garnish, Hant's Harbour, Lethbridge (c), Laurenceton and Boswarlos (a,b). #3 *Polysiphonia harveyi Bailey Found once epiphytic on Chordaria flagelliformis in the upper sublittoral zone at Hant’s Harbour. #6 Polysiphonia lanosa (L.) Tandy Epiphytic on Ascophyllum nodosum at O’Donnell’s Beach, Parker’s Cove, Garnish and Margaree. #4 *Polysiphonia nigra (Hudson) Batters Found once on rocks in the sublittoral zone at Salmonier Arm. #4 *Polysiphonia nigrescens (Hudson) Greville On rocks in the sublittoral zone of semi-exposed coasts (Hant’s Harbour and Bell Island) and sheltered estuarine regions (Salmonier Arm, Brown’s Arm, Humber Arm (b), Frenchman’s Cove (b) and Stephenville Crossing). It is most common in the latter regions where it is mixed with Chorda filum, Chordaria flagelliformis, Dictyosiphon foeniculaceus etc. #6 * Polysiphonia novae-angliae Taylor Found once epiphytic on Chordaria flagelliformis in the sublittoral zone at Sweet Bay. #4 Polysiphonia urceolata (Lightfoot ex Dillwyn) Greville var patens (Dillwyn) Harvey On rocks in the lowest eulittoral and sublittoral at Marys- town, Point Verde, Perch Cove, Kelly’s Island (b), Bell Island, Carbonear Island, Brigus Bay, Hant’s Harbour and Sweet Bay. #4 Rhodomela confervoides (Hudson) Silva On rocks in the lower eulittoral (often in tide pools) and in the sublittoral at Trepassey Harbour stations, Salmonier Arm, Point Verde, Branch, Cuslett, Gooseberry Cove, Ship Cove, Lawn, Harbour Mille, Kelly’s Island (b), Car- bonear Island, Brigus Bay, Hant’s Harbour, Margaree and Stephenville Crossing. #4 1969] Marine Algae — Mathieson, Dawes & Humm 139 DESCRIPTION OF STATIONS Southern Shore Bay Bulls: A semi-exposed bay approximately 2 miles long and 3⁄4 mile wide with steep sides. The substrate consisted of large outcroppings. Sea urchins grazing and the scour- ing effects of ice were visible in the sublittoral zone. Visited August 2, 1967. (Approximate position: Lat. 47° 18’ N, Long. 52° 49’ W). Witless Bay: A semi-exposed bay on the east coast of the Avalon Peninsula. Visited August 22, 1948. (Approximate position: Lat. 47° 18.5’, Long. 52° 51’ W). Ferryland: A semi-exposed bay formed from a ridge of off- shore islands. The substrate consisted of large boulders, which showed conspicuous ice scouring. Numerous sea urchins were evident. Visited August 22, 1948 and August 4, 1967. (Approximate position: Lat. 47° 01’ N, Long. 52° 53’ W). Fermeuse: A sheltered bay (about 4 miles long and 14 mile wide) with a small creek flowing into it. Substrate of scat- tered boulders grading into mud (at the head). Visited August 4, 1967. (Approximate position: Lat. 46° 58.3’ N, Long. 52° 36.5’ W). Renews: A sheltered cove with a large creek flowing into it. Substrate of scattered boulders grading into mud (at the head). Visited August 22, 1948 and August 4, 1967. (Approximate position: Lat. 46° 56’ N, Long. 52° 59’ W). Cappahayden: An exposed open coastal shore. Substrate of large boulders grading into small stones. Visited August 22, 1948 and August 4, 1967. (Approximate position: Lat. 46° 52’ N, Long. 52° 56.6’ W). Trepassey Bay Trepassey Harbour (a): A semi-exposed area on Powles Peninsula (head) just south of the Observatory Station — outside of the berm. Substrate ranging from large boulders to small pebbles. In August 1967 the surface water salinity was 33.2 0/oo, and the temperature 15°C. Visited August 22, 1948 and August 4 & 5, 1967. (Approximate position: Lat. 46° 42.5’ N, Long. 53° 23.5’ W). 140 Rhodora [Vol. 785 Trepassey Harbour (b): A sheltered location in the north- west arm of Trepassey Bay — near the Fishery. An area of steep cliffs. Substrate ranging from small boulders to large rocks. Visited August 22, 1948 and August 5, 1967. (Approximate position: Lat. 46° 45' N, Long. 53° 26' W). Biscay Bay: A sheltered bay due east of Trepassey Har- bour. Visited August 22, 1948 and August 4, 1967. (Ap- proximate position: Lat. 46° 47 N, Long. 53° 18’ W). St. Mary's Bay St. Mary's: A semi-exposed beach with shingle-pebble and occasional rock outcrops — in St. Mary's Harbour. Ice scouring evident on rocks, Visited August 11, 1948 and August 6, 1967. (Approximate position: Lat. 46° 56’ N, Long. 53° 34’ W). Coots Pound: A sheltered berm area enclosing a river out- let in St. Mary's Harbour. Substrate of shingle-pebbles and with occasional outcrops of boulders. Surface water salinity 32 0/00, temperature 20°C, Visited August 6, 1967. (Ap- proximate position: Lat. 46° 56.8’ N, Long. 53° 30.5’ W). Riverhead: A sheltered area at the head of St. Mary’s Harbour. Substrate of shingles-pebbles and with occasional boulders. Visited August 11, 1948 and August 6, 1967. (Approximate position: Lat. 46° 58’ N, Long. 53° 32’ W). O’Donnell’s Beach (a): A sheltered beach with pebbles and boulders. Surface water salinity 34 o/oo, temperature 20°C, Visited August 6, 1967. (Approximate position: Lat. 47° 04.8’ N, 58° 33.5’ W). O’Donnell’s Beach (b): A sheltered area 31% miles south of O’Donnell’s Beach; substrate of flat limestone ledges projecting into a sandy beach. Visited August 6, 1967. (Approximate position: Lat. 47° 05’ N, Long. 53° 37’ W). Admiral’s Beach: A sheltered beach with a shingle sub- strate. Surface water temperature 18°C. Visited August 6, 1967. (Approximate position: Lat. 47° 01’ N, Long. 53° 37’ W). Salmonier Arm: A number of locations were examined in the sheltered bay. The substrate ranged from large boul- ders to pebbles, sand and silt from the mouth of the arm 1969] Marine Algae — Mathieson, Dawes & Humm 141 to the river itself. On August 6, 1967 the surface water salinity ranged from 15 o/oo (at the head) to 35 o/oo (at the mouth), while the surface water temperatures ranged from 23°C (at the head) to 20°C (at the mouth). The following stations were occupied : (a) 11% miles south of Mt. Carmel — west side of the arm. Visited August 6, 1967. (Approximate posi- tion: Lat. 479 07.8' N, Long. 53? 31' W). (b) Mt. Carmel — west side of the arm. Collections were made at the pier near the church. Visited August 6, 1967. (Approximate position: Lat. 47° 08' N, Long. 53? 30' W). (c) 11% miles north of Mt. Carmel — west side of the arm. Visited August 6, 1967. (Approximate posi- tion: Lat. 47° 09.5’ N, Long. 53? 27 W). (d) 3 miles south of Salmonier River — west side of the arm. Visited August 6, 26 & 27, 1967. (Approxi- mate position: Lat. 47° 10.3’ N, Long. 53? 26.5’ W). (e) Forest Field — near St. Catherines on the east side of the arm. Visited August 4, 1968. (Approximate position: Lat. 47° 09.3’ N, Long. 53° 25.5’ W). (f) 3 miles south of the Salmonier River — east side of the arm. Visited August 6, 1967. (Approximate position: Lat. 47° 07.5’ N, Long. 53° 26.5’ W). (g) 114 miles north of New Bridge — east side of the arm. Visited August 6, 1967. (Approximate posi- tion: Lat. 47° 08.5’ N, Long. 53? 28.5’ W). (h) 11% miles north of St. Joseph's Church — east side of the arm. Visited August 6, 1967. (Approximate position: Lat. 47? 07.5' N, Long. 53? 29.5’ W). Tickles: A sheltered berm area consisting of shingles-peb- bles. Only a few large boulders were evident. Surface water salinity 33.3 0/oo, temperature 20°C. Visited August 7, 1967. (Approximate position: Lat. 47° 10.5’ N, Long. 53° 33.5’ W). Dog Cove: A semi-exposed cove west of Cape Dog. Visited August 11, 1948. (Approximate position: Lat. 47° 02’ N, Long. 53? 44’ W). Branch: A semi-exposed cove of about 1 mile in width. 142 Rhodora [Vol. 785 Substrate ranged from rock ledges, pebbles to sand. In August, 1967, the water was of two layers: an upper silty, brackish and warmer layer (25 o/oo and 20°C) and a lower layer which was clear, more saline and colder (35 0/oo and 10°C). Visited August 12, 1948 and August 9, 1967. (Ap- proximate position: Lat. 46° 53’ N, Long. 53° 59’ W). Placentia Bay and the Extreme Southern Tip of the Burin Peninsula Perch Cove: A semi-exposed cove on the eastern side of Placentia Bay — about 1.3 miles south of Cuslett. Visited August 12, 1948. (Approximate position: Lat. 46° 53’ N, Long. 54° 12’ W). Cuslett: A small semi-exposed cove with steep rocky sides; approximately 14 mile long and wide. Substrate at the head of the cove consisted of pebbles. Visited August 12, 1948 and August 9, 1967. (Approximate position: Lat. 46° 54.5’ N, Long. 54° 10.5’ W). Patrick’s Cove: A semi-exposed cove approximately 6 miles north of Cuslett. Visited August 12, 1948. (Approximate position: Lat. 46° 59’ N, Long. 54° 07.5’ W). Gooseberry Cove: A small semi-exposed cove with steep rocky sides; approximately 14 mile wide and 3⁄4 mile long. Substrate of scattered boulders below the ledges. Visited August 9, 1967. Approximate position: Lat. 47° 03.9’ N, Long. 54° 06’ W). Ship Cove: A small semi-exposed cove (approximately 9/4 mile wide and 1 mile long) with the same topography and substrate as Gooseberry Cove except that the headland sub- strate was coarse sand, Visited August 12, 1948 and Aug- ust 9, 1967. (Approximate position: Lat. 47° 06’ N, Long. 54° 03’ W). Big Barachois: A sheltered beach at the mouth of the Big Barachois River. Visited August 12, 1948. (Approximate position: Lat. 47° 06’ N, Long. 54° 05’ W). Green Point: A semi-exposed beach approximately 2 miles south of Placentia Harbour. Visited August 12, 1948. (Ap- proximate position: Lat. 47° 11’ N, Long. 54° 03 W). Point Verde: An exposed open coastal site. Substrate con- 1969] Marine Algae — Mathieson, Dawes & Humm 143 sisting of large boulders. Surface water salinity 35 0/00, temperature 17°C. Visited August 8, 1967. (Approximate position: Lat. 47° 14’ N, Long. 54° 00.5’ W). Ferndale: A sheltered cove at the entrance to Placentia Harbour; with steep sides. Substrate of boulders and stones. Visited August 9, 1967. (Approximate position: Lat. 47° 13’ N, Long. 53° 54’ W). Fox Harbour: A sheltered cove (approximately 15 feet deep) in Placentia Sound; polluted waters. Substrate of scattered boulders and mud. Visited August 9, 1967. (Ap- proximate position: Lat. 47° 19.2’ N, Long. 53° 54.8’ W). Ship Harbour: A semi-exposed harbour in Placentia Sound; with steep vertical sides; substrate of large boulders. Visit- ed August 12, 1948 and August 9, 1967. In late August 1967 the surface water temperature was 15°C. (Approxi- mate position: Lat, 47° 22’ N, Long. 53° 55’ W). The Placentia Sound area was also visited by Humm on August 13 and 15, 1948. Long Harbour: A sheltered harbour on the northwest coast of the Avalon Peninsula. Visited August 13, 1948. (Approximate position: Lat. 47° 27’ N, Long. 53° 48’ W). Arnold’s Cove: A sheltered cove with long beaches and small cliffs. Substrate of scattered boulders. Surface water salinity 33.6 0/00, temperature 16°C. Visited August 12, 1967. Approximate position: Lat. 47° 45.8’ N, Long. 53° 59.4’ W). North Harbour: A sheltered harbour with a flat sloping bottom. Substrate of silt-covered pebbles and limestone outcrops. Surface water salinity 32 0/00, temperature 18°C. Visited August 12, 1967. (Approximate position: Lat. 47° 50.8’ N, Long. 54° 06’ W). Garden Cove: A sheltered area with a strong tidal current. Substrate of large boulders. Surface waters are turbid, brackish (18 0/00) and warm (20°C). Visited August 12, 1967. (Approximate position: Lat. 47° 50.8’ N, Long. 54° 50.8’ W). Swift Current: A sheltered site west of Garden Cove; sub- strate of scattered boulders. Surface water temperature 144 Rhodora [Vol. 785 20°C, salinity 30 0/00. Visited August 10, 1967. (Approxi- mate position: Lat. 47° 53’ N, Long. 54° 15’ W). Sound Island: Collections were made on the west side of the Island. Visited August 12, 1948. (Approximate posi- tion: Lat. 47° 47’ N, Long. 54° 13.5’ W). Chamber’s Island (west of): Collections were made along the mainland west of the Island. Visited August 15, 1948. (Approximate position: Lat. 47° 35’ N, Long. 54° 22’ W). Haystack: A sheltered cove on the northeastern port of Long Island, Visited August 15, 1948. (Approximate posi- tion: Lat. 47° 37’ N, Long. 54° 07’ W). Merasheen: A semi-exposed cove on the south-west corner of Merasheen Island. Visited August 15, 1948. (Approxi- mate position: Lat. 47° 22’ N, Long. 54° 23’ W). Long Island: An island about 1 mile wide and 4.7 miles long — due south of Petit Forte on the western side of Placentia Bay. Visited August 15, 1948. (Approximate position: Lat. 47° 19’ N, Long. 54° 42’ W). Petit Forte: A sheltered coastal area due north of Long Island, and on the western side of Placentia Bay. Visited August 15, 1948. (Approximate position: Lat. 47° 20 N, Long. 54° 42’ W). Little Harbour: A sheltered harbour due west of Baine Harbour. Visited August 15, 1967. ( Approximate position: Lat. 47° 18' N, Long. 54° 48' W). Great Paradise: A semi-exposed site on the northwest coast of the Burin Peninsula. Visited August 15, 1948. ( Approx- imate position: Lat. 47° 18’ N, Long. 54° 37’ W). Parker’s Cove: A sheltered cove (about 1 mile long). Col- lections were made from the pier and on scattered boulders. Surface water temperature 20° C, salinity 30 o/oo. Visited on August 10, 1967. (Approximate position: Lat. 47° 25’ N, Long 54° 50’ W). Baine Harbour: A protected harbour with poor algal growth, Visited August 11, 1967. (Approximate position: Lat. 47° 22’ N, Long. 54° 55’ W). Red Harbour: A semi-exposed rocky site with large rock outcrops and boulders. Surface water temperature 17° C, 1969] Marine Algae — Mathieson, Dawes & Humm 145 salinity 25 o/oo. Visited on August 10, 1967. (Approxi- mate position: Lat. 47? 17.5' N, Long. 54? 59' W). Marystown: A sheltered site along the town front at Marys- town; within Mortier Bay. Visited August 17, 1948. (Ap- proximate position: Lat. 47? 10' N, Long. 55? 10' W). Port aw Bras: A semi-exposed cove with steep cliffs. Lo- cated about 14 mile south of the town. Visited on August 17, 1948 and August 11, 1967. In August 1967 the surface water salinity was 29 0/oo and the temperature 15° C. (Approximate position: Lat. 47° 14.3’ N, Long. 55° 08’ W). Bull Cove: A rocky coastal area with polluted water and poor visibility. Visited on August 11, 1967. (Approximate position: Lat. 479 03' N, Long. 55? 09' W). Little St. Lawrence Harbour: A protected harbour. Col- lections were made in the north side of the harbour. The substrate (boulders) was covered with fine sediment. The water was of two layers: an upper silty, brackish and warmer layer (15 o/oo and 18? C) and a lower layer which was clear, more saline and colder (32.8 0/oo and 109 C). Visited on August 11, 1967. (Approximate posi- tion: Lat. 46° 56’ N, Long. 55° 21’ W). Lawn: A semi-exposed rocky shore line with large boulders. The upper surface layers was warmer and less saline (15 o/oo and 18° C) than the lower layer (32.8 o/oo and 10° C). Visited on August 11, 1967. (Approximate position: Lat. 46° 56’ N, Long. 55° 32’ W). Taylor Bay: A semi-exposed bay on the southern tip of the Burin Peninsula — just north of Point au Gaul. Visited August 17, 1948. (Approximate position: Lat. 46° 50’ N, Long. 55° 43’ W). Lamoline: A semi-exposed shallow bay with scattered boul- ders. Surface water salinity 33.3 0/oo, temperature 17? C. Visited on August 11, 1967. (Approximate position: Lat. 46° 52’ N, Long. 55° 48’ W). Lories: An exposed location on the southern tip of the Burin Peninsula; near Point May. Visited August 17, 1948. (Approximate position: Lat. 46° 51’ N, Long. 55° 55’ W). 146 Rhodora [Vol. 785 Point Crewe: An exposed location approximately 3 miles west of Lories. Visited August 17, 1948. (Approximate position: Lat. 46° 52.5’ N, Long. 55° 58’ W). Fortune Bay Grand Bank: An exposed coastal area with substrate of large boulders and pebbles. Surface water salinity 33.2 o/oo, temperature 18° C. Visited August 11, 1967. (Ap- proximate position: Lat. 47° 06’ N, Long. 55° 48’ W). Garnish: An exposed coastal area with substrate grading from large outcrops to sand. Collection was made 1% mile south of the town. Visited August 11, 1948 and August 11, 1967. In August, 1967, the surface water salinity was 32 o/oo and the temperature 16° C. (Approximate posi- tion: Lat. 47° 14’ N, Long. 55° 22.2’ W). Harbour Mille: A semi-exposed coastal area with steep cliffs. Bottom irregular and of large boulders. Surface water salinity 33 o/00, temperature 19° C. Visited August 12, 1967. (Approximate position: Lat. 47° 35’ N, Long. 54° 56.5’ W). Little Harbour East: A “V” shaped protected cove with a river flowing into it; approximately 14 mile south of Little Harbour East proper. A delta extends about 14 mile into the cove and it consists mostly of large pebbles. No large stones were present except along the sides. The surface water salinity at the head was brackish (18.2 0/oo) and it increased towards the mouth; temperature 20° C. Visited August 12, 1967. (Approximate position: Lat. 47° 35.5’ N, Long. 54° 50.8’ W). Trinity Bay Dildo: A sheltered beach located approximately l5 mile south of Dildo. Substrate of scattered boulders and pebbles which were covered with sediments. Visited August 15, 1967. (Approximate position: Lat. 47° 34’ N, Long. 52° 32’ W). Spread Eagle: A sheltered beach located approximately 1 mile south of Spread Eagle. Substrate and topography sim- ilar to that at Dildo. Visited August 15, 1967. (Approxi- mate position: Lat. 47° 32.5’ N, Long. 52° 35.5’ W). 1969] Marine Algae — Mathieson, Dawes & Humm 147 Gull Cove: A semi-exposed cove on the east side of Trinity. Visited August 25, 1948. (Approximate position: Lat. 47° 48’ N, Long. 53° 31’ W). Hant’s Harbour: Collections were made on the exposed coastal side and the sheltered inner harbour side. The sub- strate on the exposed side ranged from large outcrons of rocks to scattered boulders, while scattered boulders and mud were present on the harbour side. The sublittoral area showed extensive grazing by sea urchins and also ice scour- ing. Surface water salinity 33.1 0/oo, temperature 12° C. Visited August 17, 1967. (Approximate location: Lat. 489 01’ N. Long. 53? 16.5’ W). Long Beach: Located on the South Arm of Random Sound. A semi-exposed location with scattered boulders in the lit- toral and sublittoral zones. Surface water salinity 33.8 0/00, temperature 11? C. Visited August 18, 1967. (Approximate position: Lat. 48° 01' N, Long. 53° 50.5’ W). Adeyton: Located on the North West Arm of Random Sound. A sheltered beach with scattered boulders in the littoral and sublittoral zones. Visited August 18, 1967. (Ap- proximate position: Lat. 48° 05' N, Long. 53° 55’ W). Shoal Harbour: A sheltered harbour with few scattered boulders in the littoral zone and a sand-mud bottom in the sublittoral. Surface water salinity 35.5 0/00, temperature 13° C. Visited August 18, 1967. (Approximate position: Lat. 48° 12’ N, Long. 54? 00’ W). Milton: Collection was made at a bridge which crosses the extreme northwest portion of Smith Sound. The area is a protected, shallow bay with occasional boulders on a mud flat. Visited August 18, 1967. (Approximate position: Lat. 48° 13’ N, Long. 53° 49.5’ W). Conception Bay Kelly’s Island (a): An exposed coastal area on the south west corner of the Island. Substrate of large outcrops and scattered boulders. Visited August 16, 1967. (Approximate position: Lat. 47° 33’ N, Long. 53° 01’ W). Kelly’s Island (b): An exposed coastal area on the north- west side of the Island. Substrate and topography as at 148 Rhodora [Vol. 785 station “a”. The sublittoral area showed extensive grazing of sea urchins, and ice scouring was also evident. Much of the vegetation was restricted to cracks and crevices. Visited August 16, 1967. (Approximate position: Lat. 47° 33.3’ N, Long. 53° 00’ W). Foxtrap: A semi-exposed coastal area with scattered boul- ders. Much of the substrate was denuded by ice action. Surface water salinity 31.6 0/00, temperature 14.4° C. Visited August 16, 1967. (Approximate position: Lat. 47° 31’ N, Long. 52° 58’ W). Bell Island: Collections were made on the southwest corner of the Island in a semi-protected cove area which had steep cliffs. The substrate ranged from large outcrops, boulders, pebbles to sand. Surface water temperature 15° C. Visited on August 16, 1967. (Approximate position: Lat. 47° 37’ N, Long. 539 00.5’ W). Carbonear Island: Collections were made on the east side of the Island in an exposed rocky beach. Substrate of large boulders. Surface water salinity 31.0 o/oo, temperature 8° C. Visited August 16, 1967. (Approximate position: Lat. 47° 41.5’ N, Long. 58° 12.5’ W). Harbour Grace (a): A semi-exposed portion at the mouth of the harbour. Substrate of large rock outcrops and scat- tered boulders. The vegetation was heavily grazed by sea urchins. Visited August 16, 1967. (Approximate position: Lat. 47° 41.5’ N, Long. 53° 12.5’ W). Harbour Grace (b) — Kings Beach: A sheltered beach southwest of station “a”. Substrate of scattered boulders which were heavily covered with sediments, The water was very dirty in appearance. Visited August 16, 1967. (Ap- proximate position: Lat. 47° 40.5’ N, Long, 53? 14' W). Harbour Grace (e): A sheltered beach in the town of Har- bour Grace. Collections were made from the pier pilings. Surface water salinity 21.4 0/oo, temperature 6? C. Visited August 16, 1967. (Approximate position: Lat. 47° 41’ N, Long. 539 13' W). Brigus Bay: An exposed coastal area with steep cliffs. Sub- strate of large massive outerops of rocks grading into boul- 1969] Marine Algae — Mathieson, Dawes & Humm 149 ders in the sublittoral zone. Surface water salinity 31.4 o/oo, temperature 12° C. Visited August 16, 1967. ( Approximate position: Lat. 47° 31’ N, Long. 53° 13’ W). Topsail Beach: A semi-exposed location east of Kelly’s Island. Collections were made by Mr. Frank Martin on May 5, 1953. (Approximate position: Lat. 47° 32’ N, Long. 52° 56’ W). Bacon Cove: A semi-exposed site south of Colliers Point. Visited on August 8, 1948. (Approximate position: Lat. 47° 25’ N, Long. 53° 13’ W). Bonavista Bay Lethbridge (a): A sheltered beach in Goose Bay. Substrate of scattered boulders in the littoral zone grading into a muddy sublittoral area. Surface water salinity 33.2 0/00, temperature 20° C. Visited August 18, 1967. (Approximate position: Lat. 48° 22’ N, Long. 53° 49.5’ W). Lethbridge (b): A sheltered beach 1 mile past Lethbridge on the road towards Brooklyn. Substrate and topography the same as station “a”. Surface water salinity 30.8 0/00, temperature 23° C. Visited August 18, 1967. (Approximate position: Lat. 48° 22’ N, Long. 53° 50.5’ W). Lethbridge (e): A sheltered beach 2 miles past Lethbridge (north) on the road towards Brooklyn. Substrate, topog- raphy and vegetation as at stations “a” and “b”. Surface water salinity 32.8 o/oo, temperature 20° C. Visited Au- gust 18, 1967. (Approximate position: Lat. 48° 23.7’ N, 538° 51’ W). Lethbridge (d): A sheltered beach 3 miles past Lethbridge on the road towards Brooklyn. Substrate of fine gravel grading into scattered rocks and mud in the sublittoral zone. Visited August 18, 1967. (Approximate position: Lat. 48° 24.7’ N, Long, 53° 49.5’ W). Sweet Bay: A sheltered location with large outcrops of rocks and scattered boulders. The vegetation was localized in cracks and crevices because of extreme ice action. Visited August 19, 1967. (Approximate position: Lat. 48° 27.2’ N, Long. 53° 38.1’ W). Bloomfield: A sheltered beach in Goose Bay —1 mile past 150 Rhodora [Vol. 785 the post office. Substrate of scattered boulders grading into mud in the sublittoral zone. Surface water salinity 33.8 0/00, temperature 15° C, Visited August 19, 1967. (Approximate position: Lat. 48° 23’ N, Long. 53° 54.9’ W). Sir Charles Hamilton Sound Gander Bay (a) — Mouth of Gander River and Gander Bay: A mud flat with abundant fresh-water runoff. Substrate of scattered boulders grading into mud in the sublittoral. Sur- face water salinity 22.7 o/oo, temperature 18° C. Visited August 20, 1967. (Approximate position: Lat. 49° 15.7’ N, Long, 54° 29.2’ W). Gander Bay (b): A mud flat area 1.2 miles north of station “a”. Substrate and topography the same as station “a”. Surface water salinity 30.2 0/oo, temperature 20° C. Visited August 20, 1967. (Approximate position: Lat. 49? 15.8' N, Long. 54? 28' W). Gander Bay (c): A mud flat area 4.1 miles north of station “a”. Substrate and topography the same as at stations “a” and “b” Visited August 20, 1967. (Approximate position : Lat. 49° 20.4’ N, Long. 54° 21’ W). Davidsville (a): A mud flat area with scattered boulders and a few pebbles. Opposite the post office in Davidsville. Collection was made from a pier piling at the location. Sur- face water salinity 23.5 o/oo, temperature 18° C. Visited August 20, 1967. (Approximate position: Lat. 49° 21’ N, 54° 21.5’ W). Davidsville (b): A mud flat area with scattered rocks — 1.5 miles north of Davidsville. Visited August 20, 1967. (Approximate position: Lat. 49° 22.5’ N, Long. 54° 21’W). Carmanville: A mud flat area with scattered rocks (oppo- site a general store), 14 mile before the ferry dock to Fogo Island. Visited August 20, 1967. (Approximate position: Lat, 49° 23’ N, Long. 54° 19’ W). Noggin Cove: A sheltered cove with little vegetation pres- ent. Most of the substrate was mud, but occasional rocks were present. Surface water salinity 34.0 o/oo, temperature 21.5° C. Visited August 20, 1967. (Approximate position: Lat. 49/ 23’ N, Long. 54° 22’ W). 1969] Marine Algae — Mathieson, Dawes & Humm 151 Notre Dame Bay Lewisporte: A sheltered beach in Burnt Bay ; with brackish water. Substrate of pebbles. Visited August 21, 1967. (Ap- proximate position: Lat. 49° 15’ N, Long. 55° 04’ W). Embree: A sheltered beach in Little Burnt Bay. Substrate of large boulders. A collection was made from a pier piling at the location and from the boulders. Surface water salinity 31.2 0/00, temperature 18? C. Visited August 21, 1967. (Approximate position: Lat. 49° 18.5’ N, Long. 55° 01’ W). Mason's Cove (a): A sheltered beach in Little Burnt Bay with large scattered boulders. Visited August 21, 1907. (Approximate position: Lat, 49° 19.5’ N, Long. 55° 01.2’ W). Mason’s Cove (b): A sheltered cove (approximately 14, mile wide) three miles north of Mason’s Cove. No name applied to it on the charts. Substrate of scattered rocks and mud. Visited August 21, 1967. (Approximate position: Lat. 49° 20.5’ N, Long. 55° 03’ W). Big Cove: A sheltered cove at the end of the road in Little Burnt Bay. Substrate of scattered boulders. Visited August 21, 1967. (Approximate position: Lat, 49° 21’ N, Long. 55° 04’ W). Brown’s Arm: A mud flat area with brackish water. Sub- strate of scattered boulders and mud. Visited August 21, 1967. (Approximate position: Lat. 49° 16.2’ N, Long. 55° 08.5’ W). Laurenceton: A sheltered beach in the Bay of Exploits. Substrate of small rocks. The area was comparatively bar- ren of vegetation because of ice jams, Surface water salinity 34.8 o/oo, temperature 19° C. Visited August 21, 1967. (Ap- proximate position: Lat. 49° 16.2’ N, Long. 55° 17’ W). Bay of Islands Humber Arm (a): A sheltered bay five miles out of Corner Brook. Substrate of small rocks grading into sand-mud in the sublittoral. Surface water salinity 25.1 o/oo, tempera- ture 19° C. Visited August 23, 1967. (Approximate posi- tion: Lat, 48° 57.5’ N, Long. 57° 58’ W). Humber Arm (b): A sheltered beach eight miles out of 152 Rhodora [Vol. 785 Corner Brook. The substrate and topography are the same as at station “a”. Visited August 23, 1967. (Approximate position: Lat. 48° 58.8’ N, Long. 57° 55.5’ W). Benoit’s Cove: A sheltered cove in Humber Arm. Substrate and topography the same as Humber Arm stations “a” and “b” A freshwater stream emptied into the cove. Visited August 23, 1967. (Approximate position: Lat, 49° 01’ N, Long. 58° 08’ W). Frenchman’s Cove (a): A sheltered cove with gravel sub- strate. Visited August 23, 1967. (Approximate position: Lat. 49° 03’ N, Long. 58° 12’ W). Frenchman’s Cove (b): A sheltered location approximately 8 miles west of Frenchman’s Cove. Substrate of scattered rocks and pebbles. Visited August 23, 1967. (Approximate position: Lat. 49° 03’ N, Long. 58° 18’ W). Cabot Strait Margaree: An exposed rocky shore west of Port aux Basques. Substrate of large rock outcrop and scattered boulders. Visited August 24, 1967. (Approximate position: Lat. 47° 34.5’ N, 59° 04’ W). Gulf of St. Lawrence Bottle Cove: An exposed coastal beach. Substrate of sand and a few scattered boulders. Visited August 24, 1967. (Ap- proximate position: Lat. 47° 42’ N, Long. 59° 15.5’ W). Red Rock Point: An exposed coastal area with scattered boulders on a sandy beach. Visited August 24, 1967. (Ap- proximate position: Lat. 47° 41.2’ N, Long. 59° 15’ W). St. George’s Bay Stephenville Crossing: A sheltered beach of scattered boul- ders grading into sand-mud. Visited August 24, 1967. (Ap- proximate position: Lat, 48° 31’ N, Long. 58° 28’ W). Port au Port Bay Boswarlos (a): A semi-exposed sandy beach with occa- sional rock outcrops. Located 1 mile before Boswarlos. Visited August 24, 1967. (Approximate position: Lat. 48? 35.3' N, Long. 58? 49' W). 1969] Marine Algae — Mathieson, Dawes & Humm 153 Boswarlos (b): A semi-exposed sandy beach with the same topography and substratum as station “a”. Visited August 24, 1967. (Approximate position: Lat. 48° 36’ N, Long. 58° 50’ W). Winter House: An exposed open coastal beach south of the “Bar” on Port au Port Bay. Substrate of large scattered boulders and sand. Visited August 24, 1967. (Approximate position: Lat. 48° 42’ N, Long. 58° 48’ W). Bonne Bay Glenburnie: A sheltered location 1.1 miles past Glenburnie in the South Arm of Bonne Bay. Substrate of scattered boulders and mud. The transparency of the water was very good. Surface water salinity 32.8 0/oo, temperature 22? C. Visited August 25, 1967. (Approximate position: Lat. 49° 27’ N, Long. 57° 54’ W). Woody Point: A sheltered location at the ferry dock at Woody Point — South Arm of Bonne Bay. Visited August 25, 1967. (Approximate position: Lat. 49° 30.5’ N, 57° 54.8’ W). DISCUSSION! Of the 155 species of marine algae listed by us from a variety of coastal environments around Newfoundland 81 species appear to be new records for the Island. In addition Collins (1905) records 2 species which we have not seen 'A recent paper by Lee (Lee, R. K. S. 1968, “A collection of marine algae from Newfoundland I. Introduction and Phaeophyta". Nat. Canad. 95: 957-978) has been published since our manuscript was submitted for publication. He records 26 species of brown algae from Newfoundland and the French islands of St. Pierre and Miquelon as well as 22 red and 16 green algae. Five of the brown algae were designated as new records (Acrothrix novae-angliae, Desmarestia media, Punctaria plantaginea, Laminaria agardhti and L. groenland- ica). We have not seen the first two species, but record the last three. Although no distributional information is given on the red and green algae, one of the green (Urospora mirabilis) and four of the red algae (Clathromorphum compactum, Euthora cristata, Ptilota pecti- nata (?) and Pantoneura baerii) have not been recorded by us. Ac- cording to our information a total of 185 species of marine algae is now known from Newfoundland. 154 Rhodora [Vol. 785 (Porphyra linearis and Calothriz scopulorum), while Tay- lor (1957) lists 14 (Bolbocoleon piliferum, Cladophora ex- pansa, Ulva lactuca, Alaria pylaii, Chorda tomentosa, Fucus miclonensis, F. spiralis, Laminaria longicruris, Bangia fuscopurpurea, Euthora cristata, Furcellaria fastigiata, Phymatolithon laevigatum, Rhodomela lycopodioides, and Rhodophyllis dichotoma), and Wilce (1959) lists 7 others (Scytonema sp., Prasiola crispa, Laminariocolax tomento- soides, Leptonema fasiculatum, Dumontia incrassata, Peys- sonnelia rosenvingii, and Phymatolithon lenormandi). At present a total of 178 species of benthonic marine algae are recorded from Newfoundland. Detailed seasonal investiga- tions will no doubt add substantially to this number, for most collections have been restricted to the summer months and nothing is known of the winter-early spring vegetation. A comparison of the marine flora of the nearby Islands of St. Pierre and Miquelon (Delamare, et al 1888; De la Pylaie, 1829; Hariot, 1889; Le Gallo, 1947, 1949; Taylor, 1957) indicates that seven other species (Lyngbya aestuarii, Pleurocapsa kerneri, Cladophora albida, Antithamnion floccosum, Gracilaria verrucosa, Sorocarpus micromorus, and Omphalophyllum ulvaceum) as well as the forma scorpi- oides of Ascophyllum nodosum are recorded for the vicinity of Newfoundland. Some of these will probably also be found in Newfoundland. Of the red, brown and green algae reported by us 47% were recorded from the Labrador Peninsula (Wilce, 1959), 89% from the Canadian Maritime Provinces (Edelstein and McLachlan, 1966, 1967a,b; Edelstein, McLachlan and Craigie, 1967; MacFarlane and Milligan, 1965) and 50% from Jaffrey Point, New Hampshire (Mathieson, et al, in press). It is evident that there is a high degree of similarity between the floras in all of the areas. Of the 178 species recorded from Newfoundland only 3 do not extend south of the Island, 37 extend only to northern New England, 45 more stop in southern New England, while 46 reach New Jersey-Maryland, 6 North Carolina, 2 South Carolina and 39 reach to the tropics. Thus the marina flora of Newfound- 199] Marine Algae — Mathieson, Dawes & Humm 155 land is primarily subarctic in character, but it has a strong component of boreal and cosmopolitan species. In contrast, the marine flora of Labrador is primarily composed of arctic and subarctic components (Wilce, 1959) and it has a larger arctic element than Newfoundland. Several of the species recorded from Labrador are not known from Newfoundland or the Maritime Provinces. However, additional collections from the northern coasts of Newfoundland may show a somewhat larger arctic component. A comparison of the number of species in the Maritime Provinces (Edelstein and McLachlan, 1966, 1967a, b; Edel- stein, McLachlan and Craigie, 1967 ; MacFarlane and Milli- gan, 1965), Newfoundland and Labrador (Wilce, 1959) indicates that there is a reduction in the number of species farther north — i.e. as the winter conditions become more extreme. In fact, according to Wilce (1959), the arctic- subarctic floras of the north are primarily characterized by a lack of species and color diversity. Similarly fewer spe- cies are recorded from the northern than from the southern shores of Newfoundland. Although ice scouring is more rigorous in northern than southern Newfoundland, many of the southern locations show pronounced scouring. Ex- tensive grazing bv sea urchins (Strongylocentrotus dro- bachiensis) is evident in the sublittoral zones (from 10 to at least 30 feet) of many exposed and semi-exposed open coasts. The combined effects of sea urchin grazing and ice scouring eliminate the bulk of the mid-sublittoral vegetation on the latter coasts. The residual flora in such areas con- sists mostly of crustose corallines, Agarum cribrosum, Des- marestia spp. and occasional small plants of other species in crevices. ACKNOWLEDGEMENTS With reference to the 1948 collections, grateful acknowl- edgement is made to Mr. Arthur H. Monroe, President of Fishery Products Ltd., for arranging transportation ( car), use of his motor vessel Duckhawk, and assistance with vari- ous other details; to the Newfoundland Industrial Develop- ment Board for the invitation to conduct the survey and for 156 Rhodora [Vol. 785 facilities provided; to Mr. Dennis Monroe for his valuable service as guide and field assistant; to Dr. W. Templemen, Director of the Fishery Research Laboratory at St. John, for making available equipment and laboratory facilities; and to Mr. William Abbot for use of his automobile along the Burin Peninsula. With reference to the 1967 collections, the writers wish to thank Marine Colloids Inc. of Springfield, New Jersey for use of their equipment and their support. Particular ap- preciation is extended to Mr. Thomas L. Chapman, of Marine Colloids Inc., who organized, managed and provided excellent facilities for the trip, and to Mr. William Ander- son, of the same firm, who participated in the collecting dives, DEPARTMENT OF BOTANY, UNIVERSITY OF NEW HAMPSHIRE DURHAM 03824 DEPARTMENT OF BOTANY AND BACTERIOLOGY UNIVERSITY OF SOUTH FLORIDA, TAMPA 33620 MARINE SCIENCE INSTITUTE UNIVERSITY OF SOUTH FLORIDA BAY CAMPUS, ST. PETERSBURG 33701 REFERENCES ADEY, W. H. 1964. The genus Phymatolithon in the Gulf of Maine. Hydrobiologia, 24 (1-3): 377-420. 1965. The genus Clathromorphum (Corallinaceae) in the Gulf of Maine. ibid., 26 (3-4): 539-573. 1966. The genera Lithothamnium, Leptophyllum (nov. gen.) and Phymatolithon in the Gulf of Maine. ibid., 28 (3-4): 321-370. BLoMQUIST, H. L. and H. J. HUMM. 1946. Some marine algae new to Beaufort, North Carolina. J. Elisha Mitchell Sci. Soc., 62: 1-8. BORNET, E. and C. FLAHAULT. 1886-1888. Revision des Nostocacées. Heterocystées. Ann. Sci. Nat. Bot., Ser 7, 3 (1886): 323-381; 4 (1886): 343-373; 5 (1887): 51-128; 7 (1888): 171-262. CARDINAL, A. 1964. Etude sur les Ectocarpacées de la Manche. Beihefte zur Nova Hedwigia. 15, 86 p. - 1965. Liste préliminaire des algues benthiques de la baie des Chaleurs. Rapp. Ann. 1964, Sta. Biol mar. Grande- Riviere: 41-51. 1969] Marine Algae — Mathieson, Dawes & Humm 157 1966. Additions à la liste des algues benthiques de la baie des Chaleurs. Rapp. Ann. 1965, Sta. Biol. mar. Grande Riviére: 35-43. 1967a. Inventaire des algues marines benthiques de la baie des Chaleurs et de la baie de Gaspé (Quebec). I. Phaéo- phycées. Naturaliste can. 94: 233-271. 1967b. ibid II. Chlorophyeées. Naturaliste can, 94: 447-469. 1967c. ibid III. Rhodophycées. Naturaliste can. 94: 735-760. 1967d. Liste preliminaire des algues marines ben- thiques des Isles-De-La-Madeleine. Rapp. Ann. 1966, Sta. Biol. mar. Grande-Riviere: 29-33. CHIHARA, M. 1961. Life cycles of the Bonnemaisoniaceous algae of Japan I. Sci. Reports Tokyo Kyoiku Daigaku, Sec. B, 10 (153): 121-154. CoLLINS, F. S. (Ed.). 1905. Phycological notes of the late Isaac Holden II. Rhodora 7 (82): 222-243. 1909. The green algae of North America. Tufts College Studies (Sci. Ser.), 2(3): 79-480. DELAMARE, E., F. RENAULT, J. CARDOT. 1888. Flora miquelnensis, florule de l'ile Miquelon, 78 pp. Paris (not seen). DE LA PYLAIE, A. J. M. B. 1824. Quelques observations sur les productions d ' île de Terre-Neuve, et sur quelques Algues de la cóte de France appartenant au genre Laminaire. Ann. Sci. Nat., 4: 174-184. 1829. Flore de L ' Ile de Terre-Neuve et les Iles Saint Pierre et Miclon. 128 pp. Paris. Dixon, P. S. 1964. Taxonomic and nomenclatural notes on the Florideae IV. Bot. Notiser, 117 (1): 56-78. EDELSTEIN, T. and J. MCLACHLAN. 1966. Investigations of the marine algae of Nova Scotia I. Winter flora of the Atlantic Coast Can. J. Bot. 44(8) : 1035-1055. 1967a. ibid III. Species of Phaeophyceae new or rare to Nova Scotia. Can. J. Bot. 45 (2): 203-210. 1967b. ibid IV. Species of Chlorophyceae new or rare to Nova Scotia. Can. J. Bot. 45 (2): 211-214. EDELSTEIN, T., J. MCLACHLAN and J. S. CRAIGIE. 1967. Investiga- tions of the marine algae of Nova Scotia II. Species of Rhodo- phyceae new or rare to Nova Scotia. Can. J. Bot. 45 (2): 193- 202. FRITSCH, F. E. 1935. The structure and reproduction of the algae. I. xviii + 791 p. Cambridge Univ. Press. 158 Rhodora [Vol. 785 1945, ibid II. xiv + 939 p. Cambridge University Press. GOMONT, M. 1892. Monographie des Oscillariées. Ann. Sci. Nat., Bot. Sér. 7 t. 15: 263-368; t. 16: 91-264. HARDER, R. 1948. Einordnung von Trailliella intricata in den Gen- erationswechsel der Bonnemaisoniaceae. Nachr. Acad. Wiss. Gottingen, Math.-Phys. K1., Biol. — Physiol. — Chem, Abt. Pp. 24-27. HARIOT, P. 1889. Liste des algues recuillies a L'île Miquelon par M. le docteur Delamarre. Journ de Bot, 3 (9-11): 154-157, 181-183. HuMM, H. J. Distribution of marine algae along the Atlantic Coast of North America. Phycologia (in press). LAMB, M. and M. H. ZIMMERMANN. 1964. Marine vegetation of Cape Ann, Essex County, Massachusetts. Rhodora 66 (767): 217-254. LE GALLO, C. 1947. Algues Marines des Iles Saint-Pierre et Mique- lon. Le Nat. Canad., 74 (11, 12): 293-318. 1949. Esquisse Générale de la Flore Vasculaire des Iles Saint-Pierre et Miquelon, suivie d'un supplément sur les algues marines. Contr. Inst. Bot. Montreal, 65: 1-84. Lewis, J. R. 1964. The ecology of rocky shores. xii + 323 p. Eng- lish Univ. Press Ltd, London. MACFARLANE, C. I. and G. M. MILLIGAN. 1965. Marine algae of the Maritime Provinces of Canada, a preliminary check list. Nova Scotia Research Foundation, Halifax, 24 pp. Mimeographed. MATHIESON, A., E. HEHRE and N. REYNOLDS. Investisations of New England marine algae. I: A floristic and descriptive ecological study of the marine algae at Jaffrey Point, New Hampshire. Nova Hedwigia (in press). PARKE, M. and P. S, Drxon. 1964. A revised check-list of British marine algae. J. mar. biol. Assoc. U. K. 44 (2) : 499-542. SCAGEL, R. F. 1966. Marine algae of British Columbia and northern Washington, Part I: Chlorophyceae (green algae). Nat. Mus. Canada, Biol. Ser, No. 74, Bull. No. 207; viii + 257 p. TAYLOR, W. R. 1954. Cryptogamic flora of the arctic II. Algae: Non-Planktonic. Bot. Review, 20 (6, 7): 363-399. 1957. Marine algae of the northeastern coast of North America. viii + 509 p. Univ. Michigan Press, Ann Arbor. Wippowson, T. B. 1964. A study of variation in the genus Alaria Greville. Ph.D. thesis, unpublished. University of British Colum- bia. ix + 229 p. WILCE, R. T. 1959. The marine algae of the Labrador Peninsula and northwest Newfoundland (ecology and distribution). Nat. Mus. Canada, Biol. Ser. No. 56, Bull. No. 158; iv + 108 p. 1969] Marine Algae — Mathieson, Dawes & Humm 159 1965. Studies in the genus Laminaria III. A revision of the North Atlantic species of the Simplices Section of Lami- naria. Botanica Gothoburgensia III, pp. 247-256. Wutrr, B. L., E. M. T. Wurrr, B. H. RoBISON, J. K. LOWRY and H. J. HUMM 1968. Summer marine algae of the jetty at Ocean City, Maryland. Chesapeake Science 9 (1): 56-60. ZANEVELD, J. Š. 1966. The marine algae of the American Coast between Cape May, N. J. and Cape Hatteras, N. C. I. The Cyanophyta. Bot. Mar. 9 (34): 101-128. EDITH SCAMMAN: AN APPRECIATION A squawky ‘Maineac’ voice, ‘thistles’ in the old straw hat, and always radiating from her warmest of hearts the kind- ness and generosity of human nature at its very best: that is how a loving but candid biographer might characterize Edith Scamman — straight from the heart in turn as she would have wished. For years we used to laugh affectionate- ly together about these first two items — for instance how, when she was with us in England in the summer of 1939 at the declaration of war on Hitler’s Reich, her rendering of ‘Keep the Home Fires Burning’ involved a sudden ascent to a high-pitched shriek each time for the ‘Burning’ and ‘Yearning’; and how, the year before that, the already old straw hat had been stood upon by a pony in Iceland which would not get off it for all my urging and belaboring until several of the ‘thistles’ had been scattered to the wind. It was Edith herself who was on that stolid pony, and our thoughtful Icelandic hosts later explained that they had chosen it ‘for the old lady’ as the only one in the region which would not only never bolt but would not even budge when belabored! And there, prolonging our happy botanical wanderings and investigations of the plant life of the sites of the Viking farmsteads whence had come the early colo- nists of Greenland and the American mainland, we ran so short of cash that we could not even afford our accustomed weekly luxury of a copy of The Saturday Evening Post. Yet there was also somewhere in Edith Scamman’s make- up an indomitable spirit of enterprise and even adventure — which sent her many times to the wilds of Alaska and remote parts of Costa Rica and elsewhere, even in later life, in search of new plants for which she had a keen eye. Though one did not know, and did not want to know, her age, she must have been well on in her seventies before she ceased her often exacting botanical collecting trips alone in distant lands, and to the end she did not give up her work in the Gray Herbarium, though it was always difficult to persuade her to publish the results of it. Most aspects of her botanical labors have been well described by others (for 160 1969] . Edith Scamman — Polunin 161 Plate 1406 Edith Scamman, in the Sierra de Talamanca of Costa Rica, April 1951. example the Tryons in Am. Fern J. 58, pp. 1-4, 1968, who also give a useful ‘Bibliography of Edith Scamman’). Yet to the taxa of vascular plants that they mention as named for her should be added at least Cerastium scammaniae Polunin (in Rhodora 53, p. 227, 1951), which she photo- graphed and studied at my request in two of its few known stations overlooking Bering Straits. For although her chief interest everywhere was in the ferns, Miss Scamman’s col- lecting in the North was far more general. Backing her steadfast enterprise, in spite of her relatively frail form and 162 Rhodora [Vol. 785 eternal womanliness, were an almost tough persistence and the spirit of the frontiersman in which she evidently took after her pioneering, west-wandering father (who was once wounded in a real western hold-up, and participated as a banker in the California gold-rush in the early 1850s). Nevertheless, Edith delighted to relate how a forebear, a Colonel Scamman of nearby Biddeford, Maine, had been court-martialled (and acquitted) for going with his men to Bunker Hill, Maine, when ordered to the Battle of Bunker Hill! Edith’s father being many years older than her mother, she was left to care for the latter on his death and did so to the preclusion of the happiness of marriage which one would have wished for her. Perhaps in unconscious partial com- pensation, she lavished affection on her friends of all ages and situations and seemed always to revel in their news and correspondence. Yet the brave pioneer spirit had evidently been inherited from her venturesome father, so that after her mother’s death, when already approaching her sixties, she began her serious study of plant life and to that end often adventurous travelling, ‘tramping the countryside with a joy in living and with none of the fears that most of us who were her friends would have known’ (according to the eulogy delivered at the Appleton Chapel, Harvard Yard, by the Reverend Dr. Frederick M. Meek on 4 December 1967). As a result of her inquisitiveness of mind and warmth of spirit she was beloved by children (including my own), and had several Alaskan and Indian babies named Edith after her; moreover to crown her pioneering there were friends who were anxious to see her admitted as the first woman member of that venerable bachelor establish- ment the New England Botanical Club — which now, only a little too late, has acceded to the inevitable march of time. Botanically, Miss Scamman’s observation and collecting were critical, painstaking, and often valuable — inter alia because of the out-of-the-way places which she was able to reach and, often, sojourn in. Notable among these were Eagle Summit in central Alaska and the interior of Costa 1969] Edith Scamman — Polunin 163 Rica where her headquarters in a dive named ‘Bar Ken- tucky’ might have been characterized by having ‘cockroaches as big as rats and rats as big as cats’. As a result of her personal field activities, Harvard has benefitted particularly by some 5,000 sheets of vascular plants from various parts of Alaska and about 1,400 from Costa Rica representing some 500 species of ferns— almost all of her own collect- ing. Her botanical publications were few but reliable, and sometimes displayed a Wordsworthian quality that was, however, consistent with scientitic accuracy. Though well educated at Thornton Academy, Saco, and then at Wellesley and Radcliffe Colleges (at which latter she obtained an M.A. in English Literature in 1909), Edith Scamman was at heart a perpetual student, attending courses ranging from Systematic Botany and Ecology to Chinese at Harvard and elsewhere until almost her latest years; and she had an infectious quality of passing on her enthusiasms to a wide range of friends. It was in this course-taking connection that I first got to know her in the middle 1930s in the old Gray Herbarium situated in the erstwhile Harvard Botanical Garden which occupied the corner of Garden and Linnaean Streets, Cambridge. That was a happy place indeed to work — often practically through the night! There we were both devoted (if not wholly devout!) disciples of the late Professor Merritt L. Fernald, another Maine stalwart of undying memories, and I well recall how on one early occasion we made a field trip with his class for some days into northern New Hampshire and Vermont to observe the spring awakening of plant life in that delightful region. The trip was memorable in many ways, not the least of which was the faithfully chugging participation of Miss Scamman's already ancient brown Dodge car — which was still in the barn of her home in Saco, Maine, and even occasionally used, when we occupied her fine family house there some twenty years later. Her frugality was entirely for herself and by her nature: she had considerable property in California as well as in Maine, and to others was generosity personified. This is thought- 164 Rhodora [Vol. 785 fully expressed in a wide range of legacies to friends and institutions. Intensely religious in her early life, and active for many years on missionary boards, Edith Scamman become more and more convinced of the ultimate values of scientific truth and reasoning as a satisfactory basis for human life as her serious study of plants progressed through its third decade and into its fourth. The last time I went to Cambridge, Mas- sachusets, to spend some days with her, in May of 1967, this philosophy seemed to prevail, though we did not discuss it specifically. She was already enfeebled, suddenly looking incredibly aged (she had seemed scarcely to change at all in the preceding thirty and more years), but still stubbornly refusing to move from the modest room in which she lived to more secure and comfortable quarters just across the street — affectionately arranged for by friends. And in spite of the setting up of a calling-system by which I would be brought over from Europe if I could be of any help, or at least in time to see her once again, this failed to function before she died in Cambridge on 4 November 1967, less than a month prior to her 85th birthday. It can truly be said, and I know I am reflecting the thoughts of numerous friends in many corners of the earth in saying so with deep feeling, that the world is not the same without Edith Scamman, hav- ing lost something rare and uniquely precious with her passing. NICHOLAS POLUNIN 1249 AVUSY, GENEVA, SWITZERLAND ADDED NOTE ON EDITH SCAMMAN When Edith Scamman first appeared at the old Gray Herbarium on Garden Street in Cambridge it was to take Professor Fernald’s course. That was in the fall of 1935. I recall that he was somewhat disturbed that she was en- rolled in the Radcliffe section. I was his teaching assistant and he muttered something about being on my guard about someone from Saco. Very soon he reversed his opinion when her unusual qualities of character, intelligence and gen- erosity asserted themselves. It must have taken me longer, for in later years she often reminded me that I had, on one occasion, criticized her drawings in the laboratory, remark- ing something to the effect that “if you are interested it seems to me you should be able to do better work." I went along on that memorable spring field trip men- tioned above in Dr. Polunin's account. Unlike most of Pro- fessor Fernald's instructional excursions I recall that one more for the gathering of personalities than plants. That same spring Miss Scamman was planning her first botanical trip. In late May, before she left Cambridge, Dr. Lyman Smith and I, probably in my Model A Ford, drove out with her past Concord to some reasonably good roadside col- lecting area and gave her a lesson in the rudiments of scien- tific plant collecting. In one other role I was privileged to play a part in Edith Scamman's career, This had to do with New England Bot- any. Under Professor Charles Weatherby's direction she had done a study of the ferns of New Hampshire. Both Mr. Weatherby and Miss Scamman had approached me about some possible source of publication. It was a bit of good fortune that being at that time Secretary of the New Hampshire Academy of Science, I could wield some influence in getting her “Ferns and Fern Allies of New Hampshire" published by that organization. A. R. HODGDON DEPARTMENT OF BOTANY UNIVERSITY OF NEW HAMPSHIRE DURHAM, NEW HAMPSHIRE 03824 165 NOTES ON THE FLORA OF CO6S COUNTY, NEW HAMPSHIRE, III. STUART K. HARRIS Two collecting seasons have passed since the data covered in Note II (Rhodora 69: 29-34 (1967) were accumulated and the County continues to produce interesting material. Professor Pease cited 1689 taxa from the County in his Flora of Northern New Hampshire. There are now addi- tional distributional data on over 750 of these taxa and more than fifty new taxa have been added. Most of the additions are adventive weeds, garden escapes or forms and hybrids. While one expects to find introduced plants on dumps, some of them have turned up in rather remote places. During World War II the fire tower on the summit of Signal Mountain in Millsfield was manned by a woman and she established a flower garden at the base of the tower ; some of her plants still persist. When the Appalachian Mountain Club built their new MizpahHut on Mount Clinton they made the highest cultivated lawn in New Hampshire around the hut and a number of plants came in with the grass seed. But what is Lysimachia punctata doing in a gravel bank on Crystal Mountain at the end of a logging road up Four Mile Brook in Dix Grant nearly fifteen miles back in the woods? In Note II I discussed the dilemma of the boundaries of Cutts Grant and hazarded the guess that the collections cited by Pease for this township and actually collected there could be counted on the fingers of one hand. It would only take three fingers; I have examined all the cited specimens in the herbarium of the New England Botanical Club, finding 101 of the 105 Pease citations. Chimaphila umbellata, var. cisat- lantica, Solidago squarrosa and S. Randii were collected in the Grant; the localities given on the labels of all the other cited specimens indicate that they were collected in Sargent Purchase. During the summers of 1966 and 1967 I have made three trips into Cutts Grant up the Dry River Trail 166 1969] Coós County — Harris 167 and the Grant now has an authentic flora of some 140 taxa, but none of them are strictly mountain species. During the past two summers I have also tried to improve the representation from the unorganized townships in the County. The known flora of Dix Grant has increased from five to 205 taxa and the Atkinson and Gilmanton Academy Grant from 17 to well over 200. Odell and Kilkenny are much better represented than formerly and Chandler Pur- chase has risen from zero to about 80. There are now not more than five townships represented by less than 100 taxa; all are in the southern half of the County. After another collecting season I hope there will be none. The year of collection is given in parentheses following the specimen citation. Equisetum arvense L., var. boreale (Bong.) Ledeb. — Errol, Harris 30844 (67); Cambridge, Harris 30808 (67); Dummer, Pease 375538 (54); Kilkenny, Harris 31152 (67). Unreported in the Flora but the Dummer specimen in the herbarium of the Nev England Botanical Club had been so identified by Pease. Probably other County speci- mens identified as the species are this variety. Osmunda cinnamomea L., forma frondosa (T. & G.) Britt. — Kil- kenny, moist ground in Willard Notch, ‘Harris and W. Tiffney, Jr. 30935 (67); Bean Grant, between Mizpah Hut and Mt. Jackson, Harris 29721 (66). Pease cites one collection from Randolph. Pinus sylvestris L. — Milan, east side of the river, Harris 31380 (67). Spontaneous seedlings in a roadside plantation, The only other County collection was made by Deane in Shelburne. Juniperus communis L., var. depressa Pursh. — Stark, roadbed of Grand Trunk Railway, Crystal, Harris 30487 (66); Berlin, roadside on east side of river, Harris 31387 (67). This seems to be a scarce plant in the County. Pease cites it from eight townships but the only other station I have seen is the cited specimen from Dummer which I found in an abandoned pasture on Dummer Hill in 1948; this colony was still doing well in 1967. Sparganium androcladum (Engelm.) Morong — Milan, Long Pond, West Milan, Harris 30580 (66). Previously reported only from the Second (Dartmouth) College Grant. Elodea Nuttallii (Planch.) St. John — Cambridge, Bog Brook, Harris 29864 (66). Now known from six townships; this species is probably more common than collections indicate. Lolium perenne L. — Errol, roadside at Seven Islands, Harris 30997 (67); Bean Grant, lawn of Mizpah Hut, 3800 feet, Harris 29725 (66). Previously reported only from Randolph. 168 Rhodora [Vol. 785 Lolium multiflorum Lam., var. diminutum Mutel — Pittsburg, town dump, Harris 31512 (67). Second record for the County; the first, from Millsfield, was reported in Note II. Avena fatua L. — Pittsburg, about old stable near Moose Pond, Harris and Scott 30416 (66). New to the County. A few plants mixed in an abundant stand of A. sativa which was new to Pittsburg; Harris and Scott 31417 (66). Sporobolus neglectus Nash — Dummer, Grand Trunk Railway tracks, Harris 30729 (66); Milan, Grand Trunk tracks near Copper- ville, Harris 30614 (66). The only previous record is from the ad- joining town of Stark and also on the railway. Calamagrostis Pickeringii Gray — Chandler Purchase, summit of Mount Pleasant, Harris 31766 (67). Previous collections are all from Sargent Purchase. Alopecurus pratensis L. — Atkinson and Gilmanton Academy Grant, clearing at Hell Gate, Harris 30923 (67). The only other collections are from Lancaster and Shelburne. Phalaris canariensis L.— Pittsburg, town dump, Harris 31513 (67); Errol, town dump, Harris 30132 (66); Milan, dump, West Milan, Harris 31282 (67). Two previous collections, also from waste ground. Panicum dichotomiflorum Michx., var geniculatum (Wood) Fern. — Northumberland, waste ground, North Road, Groveton, Harris 30126 (66). The third record for the County. Panicum miliaceum L. — Colebrook, town dump, Harris 30244 (66) ; Errol, town dump, Harris 30131 (66); Stratford, foot of Sugarloaf Trail, Harris 29738 (66); Milan, dump West Milan, Harris 29662 (66); Gorham, town dump Cascade, Harris 31463 (67). Three previ- ous reports. The lookout on the Sugarloaf fire tower feeds birds. Cyperus strigosus L. — Gorham, town dump Cascade, Harris 31468 (67). Pease cites the species from two towns on the Connecticut River where it may well be native but it is certainly an introduction on the Gorham dump. Arisaema Stewardsonii Britt. — Atkinson and Gilmanton Academy Grant, low ground at Hell Gate, Harris 30880 (67). Reported from Colebrook as new to the County in Note II. Lemna minor L. — Colebrook, roadside pool, Harris 30239 (66); Stark, farm pond, Harris 28796 (66). This brings the County repre- sentation to five townships; Lemna seems to be a difficult genus to find in northern New Hampshire. Allium tricoccum Ait. — Milan, rich moist hillside, West Milan, Harris 28681 (66). Known from three other towns. Cypripedium acaule Ait., forma acaule — Pittsburg, Indian Stream, Chapman, Nugent, Nugent and Tiffney (64); Stratford, road to Trio Ponds, Harris 28798 (66). See Note II for a discussion of the distribution of the pink-flowered form. 1969] Coós County — Harris 169 Epipactis Helleborine (L.) Crantz — Dixville, damp roadside near the Balsams, Harris 30223 (66). Previously known only from Mt. Prospect in Lancaster where it was first collected in 1943. Salix discolor >< humilis — Pittsburg, Indian Stream, Chapman, Nugent, Nugent and Tiffney (64). First report for the County. Salix gracilis, var. textoris X S. rigida — Pittsburg, Indian Stream, Chapman, Nugent, Nugent and Tiffney (64). First report for the County. Populus alba L. — Stark, spontaneous seedling in dooryard, Harris 31376 (67); Milan, roadside on east side of river, Harris 31381 (67). Previously reported from two townships. Ulmus rubra Muhl. — Dummer, road up Dummer Hill, Harris 29098 (66). I had been told at least ten years ago that in the 1800's U. rubra had been grown on Dummer Hill and the bark sold in Port- land, Maine. At the time I had searched for remaining trees without success. While walking out from camp one morning in 1966 I noticed some samaras on the road which looked different from those of U. americana. They had fallen from two trees of U. rubra, which over- hang the road and which are at least eighteen inches in diameter at breast height. Both trees show evidence of Dutch Elm Disease but they again produced fruit in 1967. I can find no evidence of seedling trees in the vicinity. Ulmus americana L., forma pendula (Ait.) Fern. — Cambridge, roadside, Harris 30815 (67). The second township for the County. Ulmus americana, forma laevior Fern, — Pittsburg, Hall Stream, Chapman, Nugent, Nugent and Tiffney (64). The second report for the County. Ulmus americana, forma intercedens Fern, — Atkinson and Gil- manton Academy Grant, vicinity of Hell Gate, Harris 30878 (67). New to the County. All four forms are now known from the County; of these forma alba (Ait.) Fern. is by far the most common. Polygonum aviculare L., var. vegetum Ledeb. — Dummer, waste ground Paris, Harris 29905 (66); Milan, Emery farm West Milan, Harris 27394 (65). Previously known from the railroad in Gorham. Polygonum Persicaria L., forma albiflorum Millsp. — Atkinson and Gilmanton Academy Grant, roadside, Harris 30002 (66). First report for the County. Polygonum Persicaria, var. ruderale (Salisb.) Meisn. — Gorham, town dump Cascade, Harris 31470 (67). New to the County. Chenopodium urbicum L.— Pittsburg, roadside Indian Stream, Harris and Scott 30418 (66). The third record for the County. Lychnis Flos-cuculi L. — Pittsburg, logging road east of Parma- cheene Bog, Harris and Scott 29150 (66). First reported for the County in Note II. In June 1966 a lady in Berlin told me she had found Ragged Robin “on the right side of the road somewhere between Dixville Notch and Pittsburg.” I promptly searched that stretch of 170 Rhodora [Vol. 785 road, a matter of some thirty miles, without success. Knowing that she had found it on her way to go birding with Fred Scott, I looked him up to see if she had shown him the plant and told him where she found it. He had never heard of Ragged Robin and I showed him the picture in his old Reed Flower Guide. In less than a week he sent me a fragment of the species which he had discovered in the locality cited above. Thus there are now two definite stations for the species in the County plus a third yet to be relocated. Rancunculus trichophyllus Chaix, var. trichophyllus — Cambridge, Bog Brook, Harris 29845 (66). The sixth township for the County. Ranunculus repens L. — Pittsburg, logging road east of Parma- cheene Bog, Harris and Scott 29125 (66); Randolph, roadside Keenan Brook, Harris and W. Tiffney, Jr. 30925 (67). This introduced species is now known from four townships. With the exception of the Pease collection made on the Ravine House lawn in Randolph, all collections have been made on logging roads well off the beaten path. Aconitum Napellus L. — Dummer, cellar hole Dummer Hill, Harris 27580 (65). The old farm where this still persists was abandoned at least fifty years ago. Sanguinaria canadensis L. — Atkinson and Gilmanton Academy Grant, low ground at Hell Gate, Harris 30874 (67). The other four townships from which this species is known are all on the Connecticut River; the Academy Grant is in the Androscoggin watershed. Fumaria officinalis L. — Milan, Emery farm West Milan, Harris 29541 (66). The second report for the County. This species has ap- peared as a weed in a vegetable garden for at least twenty years; I did not find it again in 1967. Raphanus Raphanistrum L., forma albus (Scheubler & Martens) Hayek — Berlin, site of old dump Jerico, Harris 27870 (65) ; Gorham, town dump Cascade, Harris 31466 (67). One other report from the County. Sedum spurium Bieb. — Milan, dump West Milan, Harris 29678 (66). New to the County. Saxifraga cernua L. Saxifraga Aizoon Jacq. — Both of these Saxifrages are new to New Hampshire. They were found on Mount Washington in 1939 by Dr. John A. Churchill; the station was relocated in 1967. For full details see Rhodora 69: 483-486 (1967). These are the most exciting finds made in the County for several decades. Potentilla canadensis L.— Pittsburg, Indian Stream, Chapman, Nugent, Nugent and Tiffney (64). The first collection for the County. Lupinus polyphyllus Lindl. — Millsfield, summit of Signal Moun- tain, Harris 31356 (67). Persistent in old garden at base of fire tower. Third collection for the County. Trifolium repens L., forma phyllanthum (Sm.) Fieri & Beguinet — Milan, roadside West Milan, Harris 29171 (66). New to the County. 1969] Coos County — Harris 171 Lotus corniculatus L. — Stark, roadside, Harris 30971 (67) ; North- umberland, roadside Groveton, Harris 30972 (67). I reported this as new to the County in 1965, Rhodora 67: 196; it is now known from six townships. Rhus typhina L. — Errol, roadside, Harris 30837 (67); Milan, field near Cedar Pond, Harris 31166 (67). The Errol station extends the known range of the species in the Androscoggin valley north by some thirty miles. Vitis vinifera L. — Milan, dump West Milan, Harris 31271 (67). New to the County. I was interested to find that seedlings of a table grape which I assume to be this species occur on dumps in northern New Hampshire just as they do on dumps in Essex County, Massachu- setts. They apparently are not hardy enough to withstand New England winters. Malva moschata L., forma moschata — Pittsburg, town dump, Harris 28979 (66); Second College Grant, site of old farm, Harris 31338 (67). Both collections seem to be the typical form which is unreported for the County. There are numerous collections of forma laciniata and some of forma heterophylla, however. Althaea rosea Cav. — Milan, dump West Milan, Harris 29681 (66). This is another fairly regular inhabitant of dumps. Not reported previously from the County. Abutilon Theophrasti Medic. — Pittsburg, site of Depot Camp, Indian Stream, Harris and Scott 30422 (66). Another first collection for the County. Viola arvensis Murr. — Stark, dooryard Crystal, Harris 31139 (67). The third township for the County. Lythrum Salicaria L. — Dummer, waste ground Paris, Harris 29894 (66); Whitefield, roadside ditch, Harris 29173 (66). As yet this is not a common weed in the County there being only two previous collections. It would not be surprising if Pittsburg should be the next town where it is reported since I have collected it in Quebec not more than a hundred yards over the border above Third Lake. Epilobium coloratum Biehler Second College Grant, along Swift Diamond, Harris 30050 (66); Stratford, roadside along Nash Stream, Harris 30099 (66). The only other collection was made in Whitefield in 1897. Panax trifolium L. — Pittsburg, Hall Stream, Chapman, Nugent, Nugent and Tiffney (64). There are only three other towns repre- sented. Possibly this species is more abundant than collections indi- cate because the flowering period is over in early June before most botanists start collecting in the County. Moneses uniflora (L.) Gray — Dixville, shore of Mud Pond, Harris 30205 (66); Cambridge, shore of Bog Brook, Harris 29823 (66); Cutts Grant, between Shelter #2 and Mizpah Hut, C. C. Harris 29720 (66). Pease cites three towns for this species; this brings the total to six. 172 Rhodora [Vol. 785 Gaultheria, procumbens L., forma suborbiculata Fern, — Stratford, summit North Percy Peak, C. C. Harris 30483 (66). The second col- lection for the County. Lysimachia punctata L. — Dix Grant, gravel bank at end of road up Four Mile Brook, Harris 30445 (66), 31061 (67). New to the County. An odd place to find a garden escape since there is no evi- dence that there ever had been a garden within miles of the spot. Convolvulus arvensis L. — Milan, tracks of Grand Trunk Railway near Copperville, Harris 30604 (66). First report for the County. Myosotis sylvatica Hoffm. — Pittsburgh, Hall Stream, Chapman, Nugent, Nugent and Tiffney (64). The second report for the County. Dracocephalum parviflorum Nutt. — Bean Grant, edge of lawn at Mizpah Hut, Harris 29730 (66). New to the County. The station was still extant in 1967. Lamium maculatum L., forma lacteum (Wallr.) G. Beck — Mills- field, summit of Signal Mountain, Harris 31355 (67). New to the County. Monarda media Willd. — Dummer, beside old cellar hole Dummer Hill, Harris 29693 (66). New to the County. This long abandoned farm once had an herb garden. Professor Pease collected Asarum canadense here in 1930 and it is now abundant for a hundred yards along the road; I am sure the Asarum came from the herb garden rather than being native to Dummer Hill. Solanum. sarachoides Sendtner — Gorham, town dump Cascade, Harris 31465 (67). New to the County. This weed is still not com- mon in New England but seems to be spreading in the area. Euphrasia canadensis Townsend — Dix Grant, road up Four Mile Brook, Harris 30461 (66); Wentworth Location, road to Dixie Dam, Harris 30441 (66); Dixville, roadside near the Balsams, Harris 30211 (66) ; Colebrook, roadside, Harris 30240 (66) ; Millsfield, logging road, Harris 31366 (67); Dummer, road to Dummer Pond, Harris 30352 (66). Pease cites five towns and describes the species as local. In the northern half of the County it appears to be fairly common and has distinetly weedy characteristics. Mr. Leo Charette has recently pointed out that the two collections of Forbes cited from Green Grant are E. americana rather than E. canadensis and I am sure that other collections cited by Pease and by me could well be referred to E. americana or some other related species. Specific lines in this con- fusing genus are not clearly drawn and until the genus has had a thorough revision it seems to me that it is best to treat collections as a single highly variable species. Rhinanthus Crista-galli L. — Atkinson and Gilmanton Academy Grant, roadside, Harris 30248 (66); Clarksville, road to Clarksville Pond, Harris 29014 (66); Second College Grant, road along the Swift Diamond, Harris 30081 (66); Wentworth Location, field, Harris 28915 (66); Errol, roadside, Harris 28953 (66). Pease knew this 1969] Coos County — Harris 175 species only from Pittsburg where he first found it in 1938. In Note II I reported it from the northern edge of Randolph. Apparently in the northern half of the County it is now a rapidly spreading weed. Galium erectum Huds. — New to the County. The Pease collection #30134 (48) cited from Whitefield under G. Mollugo is a perfectly good specimen of this rather dubious species. Cephalanthus occidentalis L. — Milan, Long Pond, Harris 30549 (66), North Branch, Harris 30588 (66). It seems strange that But- tonbush is a rare plant in Coos County. Pease collected it in North- umberland in 1909 and Deane found a single bush in a hen yard in Shelburne and traced its history over a number of years in Rhodora. In August 1966 a lady in West Milan showed me a flowering head which was in a bouquet her grandson had brought her from a pickerel fishing trip. His mother told me that he had fished in Long Pond, so I put my boat in there and found a few sterile bushes. A few days later the boy told me he had fished the North Branch and again by boat I found abundant flowering material there. While searching in Long Pond I also found Xyris montana which is cited in Pease only from Pike Pond in Stark and Goose Pond in Milan. Goose Pond appears as a locality in Milan on a number of Pease's collections. This poses a problem. I have been unable to find Goose Pond, Milan, on any map and I have asked numerous inhabi- tants of Milan who have lived there all their lives and none have ever heard of it. Since Pease also cites specimens from Long Pond, Goose Pond is not apparently a synonym for that. Eupatorium rugosum Houtt. forma villicaule Fern. — Pinkham Grant, Glen Road, Floyd (03); cited in Pease as E. rugosum. New to the County. Aster puniceus L., forma albiflorus (Farw.) Shinners — Dixville, logging road near Mud Pond, Harris 30181 (66); Odell, between Little Bog Pond and Trio Ponds, Harris 30515 (66). This doubles the representation of this taxon in the County. Aster junciformis Rydb. — Dalton, Pease 39316 (58). The second collection from the County. Aster nemoralis Ait., forma albiflorus Fern. — Milan, Long Pond, Harris 30564 (66). Long Pond is on the outlet of South Pond in Stark where Pease made the only other collection of this taxon in the County. Gnaphalium sylvaticum L. — Dix Grant, along the Swift Diamond, Harris 29559 (66); Dixville, logging road near Mud Pond, Harris 30142 (66); Erving Location, logging road, Harris 28181 (65); Columbia, roadside Cilley Hill, Harris 28309 (65); Odell, Trio Pond road, Harris 30508 (66); Dummer, road to Dummer Ponds, Harris 30354 (66). First collected in Success in 1908 and not found again until 1942. The species is frequent now in the northern half of the County. 174 Rhodora [Vol. 785 Ambrosia trifida L. — Berlin, waste ground Mechanic Street, Harris 27959 (65). There are only two other collections from the County. I first saw the plant in Berlin in the early 1950’s and told Pease but at the time it was not convenient to collect a specimen, On visiting the same site in 1965 I was surprised to discover that the species had persisted more than ten years and pleased to be able to validate the report in Pease with a specimen. Helianthus laetiflorus Pers. — Dummer, Grand Trunk Railway, Harris 30733 (66). The third town for the County. Helianthus tuberosus L. — Errol, town dump, Harris 31505 (67). Known from three other towns. Helenium nudiflorum Nutt. — Milan, schoolyard West Milan, Harris 30438 (66). The fourth town for the County. Calendula officinalis L. — Milan, dump West Milan, Harris 29685 (66). The second collection for the County. Centaurea montana L. — Dummer, lawn at our camp on Dummer Hill, Harris 27407 (65). Two other collections from the County. This Species was planted in our flower garden nearly twenty years ago. There are now dozens of clumps of it on the lawn and the surrounding field. Crepis capillaris (L.) Wallr. — Dixville, logging road near Mud Pond, Harris 30185 (66). The third town for the County. I collected it again in Dixville in 1967 on a logging road several miles from the cited location. BOSTON UNIVERSITY BOSTON, MASSACHUSETTS ADDITIONS TO THE KANSAS FLORA Field studies undertaken during a floristic survey of Cowley County, Kansas, have resulted in the collection of two taxa previously unreported for the state. They are listed below. The specimens cited in this note may be found in the Herbarium of Oklahoma State University. Special thanks are due to Dr. U. T. Waterfall for his help and encourage- ment during the study. Ipomoea Shumardiana (Torr.) Shinners. Shinners (1961) made the above combination which replaced I. Carel- toni Holz. This taxon, originally described as a Texas-Okla- homa endemic, can be separated from I. longifolia Holz. by its ovate-lanceolate leaf-blades widest near the base, and its smaller calyces. It differs from I. pandurata (L.) G. Meyer in leaf characters. The collections (Koch 1707 and 1530) are from an abandoned field and adjacent railroad right-of- way one half mile north of Winfield and extend the known range of the species eighty miles north from Payne County, Oklahoma. Melothria pendula L. The first collection (Koch 1989) was made from the shade of a stand of Populus deltoides Marsh. along the Arkansas River northeast of Arkansas City. The following summer it was taken from the margin of a small stream near Gueda Springs (Koch 4077) where it was partially shaded by the dense growth of Ambrosia trifida L. These collections extend its known range some one hundred miles north and west of previously reported sites in Oklahoma (Waterfall, 1966) and Missouri (Steyer- mark, 1963). RUDY G. KOCH DEPARTMENT OF BOTANY AND NEBRASKA STATE MUSEUM, UNIVERSITY OF NEBRASKA, LINCOLN, NEBRASKA 68508 175 176 Rhodora [Vol. 785 LITERATURE CITED SHINNERS, L. H. 1961. Ipomoea Shumardiana (Convolvulaceae), a neglected Oklahoma-North Texas endemic. Southwest Nat. 6: 101. STEYERMARK, J. A. 1963. Flora of Missouri. Ames: Iowa State Uni- versity Press. WATERFALL, U. T. 1966. Keys to the flora of Oklahoma, ed. 3. Still- water, Oklahoma: Privately published. Volume 71, No. 785 including pages 1-176, was issued April 7, 1969. FARLOW REFERENCE LIBRARY “73 1198 RU? Hovdova JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by ALBION REED HODGDON, Editor-in-Chief ALBERT FREDERICK HILL D STUART KIMBALL HARRIS RALPH CARLETON BEAN ROBERT CRICHTON FOSTER » Associate Editors ROLLA MILTON TRYON RADCLIFFE BARNES PIKE LORIN IVES NEVLING, JR. Vol. 71 April-June, 1969 No. 786 CONTENTS: Contributions to the Fungus Flora of Northeastern North America V. Howard E. Bigelow and Margaret E. Barr eee 177 A Review of the Varieties of Polymnia uvedalia. James R- Wels: insiren as bu ME 204 Iris verna L.: Lectotype and Variety Description. Mauriee E. Edward |... iana a m 212 The Vascular Flora of Oklahoma — Additions and Com- ments. R. John Taylor and Constance Taylor .................... 215 The Bromeliaceae of Bolivia. (concluded) Lyman B. Shah: uu u R a A n a s: 220 (Continued on Inside Cover) Che Nef England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge, Mass. 02138 CONTENTS: — continued Chromosome Counts of Kalmia Species and Revaluation of K. polifolia var. microphylla. Richard A. JAYNES ................................ rs... 280 Cytotaxonomic Notes on the Species of Cirsium Native to Southeastern United States. Gerald B. Ownbey and Willis A. Olson ............... eee 285 Continued Botanizing on the Wolf Islands, New Brunswick. A. R. Hodgdon and R. B. Pike ........ eese 297 Isotria medeoloides on the Piedmont of Virginia. A. M. Harvill, Jv. Leere nnne nnne 303 A Rhode Island Station for Luminous Moss. Richard L. Champlin ........2............. a eterne 305 More Additions to the Flora of Grand Manan, New Brunswick. A. R. Hodgdon, R. B. Pike and Harry R. Tyler, Jr. ............ 306 Revision of the Genus Dahlia (Compositae, Heliantheae — Coreopsidinae). (to be continued) Paul D. Sorensen ........ 309 QTRbooora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 71 April-June No. 786 CONTRIBUTION TO THE FUNGUS FLORA OF NORTHEASTERN NORTH AMERICA. V. Howarp E. BIGELOW AND MARGARET E. BARR?’ Since our last report on the higher fungi of the northeast, continued field work and the laboratory study of specimens have resulted in additional records of fruiting as well as the discovery of more new species. Some interesting variations of well known species have been found also. The following constitute a miscellaneous assemblage which does not be- long to monographic work currently in progress. As previously, the data on agarics are from the studies of the senior author, while the junior author has contributed the information on ascomycetes. The colors mentioned within quotation marks are from Ridgway, R. 1912. Color Standards and Color Nomenclature. Washington, D.C. Unless stated otherwise, the collections cited are deposited in the Herbarium of the University of Massachusetts, Amherst (MASS). BASIDIOMYCETES Armillaria decorosa (Pk.) Smith & Walters, Mycologia 39: 622. 1947. Material examined. Vermont: Bigelow 13940, Bingham Falls, near Stowe, Lamoille Co., 9 Sept 1964. Solitary or scattered, on hard- wood log. 1T. Rhodora 62: 186-198. 1960. II. Rhodora 64: 126-137. 1962. IH. Rhodora 65: 289-309. 1963. IV. Rhodora 68: 175-191. 1966. Contribution from the Department of Botany, University of Massa- chusetts, Amherst. *Mrs. Howard E. Bigelow. 177 178 Rhodora [Vol. 71 This rare species is described and illustrated by Smith and Walters (1947), and by Farlow and Burt (1929, as a Tricholoma). The following provides additional informa- tion on microscopic characters. Spores 5-6 X 3-3.5 z, elliptic, smooth, amyloid. Basidia 15.5-21 X 5.5-6 u, 4-spored. Cystidia not differentiated. Pileus cutis brown in KOH, pigment in cell contents and in thickened smooth walls, hyphae cylindric or composed of chains of + sausage-shaped cells, 5.5-19 u in diameter, 31-55 u long, trama pale brown in KOH, hyphae cylindric to inflated, 4-11.5 u in diameter, walls thin or slightly thickened, hyaline or faintly brown. Hymenophoral trama regular to subparallel, hyaline in KOH, hyphae cylindric or slightly inflated, 4-9 u in diameter. Clamp connections present. As Smith and Walters have indicated, A. decorosa seems closely related to the lignicolous species of Cystoderma, I did not find the sphaerocysts typical of Cystoderma in the Vermont carpophores, but certainly the sausage-shaped cells of the cutis hyphae suggest a possible precursor of the cuticular sphaerocysts, Armillaria viscidipes Peck, New York State Mus. Rep. 44: 16. 1891, Plate 1406. Pileus 6.5-13.5 em broad, broadly convex to plane, broadly subumbonate at times, margin incurved at first, finally upturned in age, even, surface dry when collected but probably viscid in wet weather, somewhat shining, innately fibrillose in places or diffracted scaly, in age cracked or pitted at times, color a sordid buff; context thick, white, tough, odor pungent, taste mild. Lamellae adnexed at first becoming rounded or slightly sinuate, narrow to moderately broad (up to 1 cm), close or crowded, forked at times, whitish becoming cream-colored, dingy in age, edges even, slightly undulate. Stipe 5-11 cm long, apex 1.5-3.4 em thick, equal or base ventricose (up to 4 em thick), abruptly tapered in substrate, solid (white within and continuous with pileus flesh), apex white, fibrillose to somewhat scabrous (about !4 of stipe length), partial veil thin and soon collapsed, central portion and base viscid, brownish stained when gluten dried, rather shiny and somewhat zoned or appressed scaly. Spores 6-8 X 4-5.5 z, elliptic to broadly elliptic, smooth, not amyloid, spore deposit white. Basidia 31-44 X 6-8 4, 4-spored. Cystidia not differentiated. Pileus surface with very thin gelatinous pellicle, cutis thin, hyphae mostly cylindric, (2-)4-12.5 u in diameter; tramal hyphae cylindric to inflated, 7.5-18.5 (-21.5) u in diameter, walls often some- what thickened. Hymenophoral trama of subparallel hyphae, slightly 1969] Fungus Flora — Bigelow and Barr 179 Plate 1406 Plate 1406. Armillaria viscidipes Peck. X 1. diverging near subhymenium, hyphae cylindric, 2.5-5(-8.5) 4&4 in diameter. Clamp connections absent. Refractive hyphae present but rare. Gregarious, on mossy hummock near stream through open hemlock stand. Material examined. Massachusetts: Bigelow 14999, Conway, Frank- lin Co., 31 Oct 1966; 15590, same locality, 16 Oct 1968. This Armillaria also appears to be rare according to the records in the literature. Besides the orginal description based on specimens found in New York, the only other 180 Rhodora [Vol. 71 records from the east appear to be those of Hesler (1945). He collected A. viscidipes in Kentucky and Tennessee. Kauffman (1921, 1925) has reported fruitings in Colorado and Oregon. Cystoderma amianthinum var. sublongisporum Singer, Pap. Mich. Acad. Sci., Arts, Lett. 30: 112. 1945. Material examined. Massachusetts: Bigelow 7883, Mt. Toby, Sun- derland, Franklin Co., 4 Oct 1958; 7965, same locality, 2 Nov 1958; 10035, Womens Club State Forest, New Salem, Franklin Co., 7 Nov 1961. New Hampshire: 12519, Oliverian Brook trail, White Moun- tains Nat. Forest, 7 Sept 1963. Vermont: 13673, top of Mt. Mans- field, 29 Aug 1964; 13812, 13818, Ranch Brook trail, near Stowe, Lamoille Co., 1 Sept 1964; 13941, Bingham Falls, near Stowe, 9 Sept 1964; 14011, same locality, 11 Sept 1964; 14360, same locality, 3 Oct 1964, Quebec: 4783, 4784, 4785, Provincial Nursery, near Berthier- ville, 9 Oct 1956; 4861, near Farnham, 28 Oct 1956; 4871, Provincial Nursery, 30 Oct 1956; 6124, Lac Cascapédia, Gaspé Pare, 21 Aug 1957. Usually scattered or gregarious on mossy conifer logs, occasionally solitary on ground under conifers. Although not quite as common in the northeast as var. amianthinum, var, sublongisporum is widespread and often encountered. The spore size of specimens must be deter- mined to separate the two varieties unless var. amianthinum f. rugosoreticulum is involved, The latter can be distin- guished in the field by the rugose cap and often a green corn odor when the flesh is crushed. Cystoderma pulveraceum (Pk.) Smith & Singer, Pap. Mich. Acad. Sci., Arts, Lett. 30: 104. 1945. Material examined. Vermont: Bigelow 9819, near Whitingham, Windham Co., 10 Sept 1961. Solitary, on moss covered very decayed log under conifers, Hygrophorus conicus (Fr.) Fr. Epicrisis, p. 331. 1888. form. Plate 1407. Pileus 1.8-5.5 em broad, conic at first then the margin spreading, umbo prominent and blunt, surface moist, radiate-fibrillose at first, often becoming squamulose, fibrils and scales black, ground color deep red at first, paler in age to reddish-orange then finally orange, ground color of expanded pilei obscured by black of scales except on the margin; context thin, firm but brittle, watery dingy citrine fading to whitish, finally black, no odor and taste. 1969] Fungus Flora — Bigelow and Barr 181 Plate 1407 Plate 1407. Hygrophorus conicus (Fr.) Fr. form. >< 1. Lamellae adnexed to free, elose to subdistant, broad (up to 9 mm), ventricose or rounded near pileus margin, narrowest at the stipe, faces pale yellow at first, becoming sordid and darker yellow, finally reddish and with a bloom near the pileus margin, edges whitish at first, red marginate at times, blackening, uneven. Stipe 3.5-8 em long, 5-9 mm thick, equal or the base enlarged (up to 14 mm), hollow (interior blackening when cut), surface fibrillose striate, fibrils black over a greenish yellow ground color, often with a reddish blush in part or over the whole length. Spores 10-12 X 5.5-7 n, elliptic to obovate or oblong, slightly bent at times, smooth, olive black at times in KOH, not amyloid. Basidia 182 Rhodora [Vol. 71 31.5-46 X 7-8.5 u, 4-spored, often olive-blackish in KOH, blackness in contents, granular or diffuse. Cheilocystidia present on some gills, clavate to rostrate or mucronate, 30-60 u long, 9-13 u in diameter, agglutinated and blackened at times. Pleurocystidia not found. Pileus cutis dark olive black in KOH, pigment in contents — spotted or homogeneous, wall thickened at times, hyphae cylindric or some- what inflated, 1.5-11 u in diameter, trama olivaceous-yellow in KOH, appearing cellular at times, hyphae inflated, 15-19 &4 in diameter, walls somewhat thickened. Hymenophoral trama parallel, hyphae cylindric to somewhat inflated, 4-17 u in diameter. Clamp connec- tions present, Refractive hyphae present. Gregarious, on humus under hardwoods. Material examined. Vermont: Bigelow 12998, Bingham Falls, near Stowe, Lamoille Co., 22 July 1964; 13032, Stowe, 23 July 1964; 13243, Bingham Falls, 3 Aug 1964. With the black squamules over brieht pileus colors, the red gills, and red blush on the stipe, this interesting varia- tion of H. conicus is striking in the field. Typical specimens were found in the same general locality about the same time and weather conditions did not appear to be respon- sible for the different appearance. Comparison has been made with H. conicoides Orton, which also has red pig- ments, but the spores of the Vermont collections are identi- cal with those of typical H. conicus and not elongate as in H. conicoides. As far as I have been able to determine, cheilocystidia have not been reported for H. conicus before, These are difficult to identify in young lamellae, but are quite abun- dant in expanded ones. The red color leaches out of sections readily in KOH, but is obscured by other pigments if blackening has occurred. Hygrophorus cossus Fr., Epicrisis p. 321. 1838. Material examined. Massachusetts: Bigelow 15373, Conway State Forest, Franklin Co., 10 Oct 1967. Gregarious, under hardwoods. Hygrophorus fibrillosus Bigelow, sp. nov. Plate 1408. Pileus 1.5-2.3 cm latus, convexus, margine decurvatus et incurva- tus, siccus, opacus, fibrillosus, pallido-bubolinus; caro albida, crassa, solida, odore et sapore farinaceo. Lamellae decurrentes, confertae, angustae, incrassatae, subvinoso-griseae, Stipes 2.5-4.2 cm longus, 6-9 mm crassus, aequalis vel deorsum attenuatus, solidus, siccus, 1969] Fungus Flora — Bigelow and Barr 185 Plate 1408 Plate 1408. Hygrophorus fibrillosus Bigelow, sp. nov. PX A opacus, velutinus, pileus concolorus. Sporae (6-)7.5-9(-10) X (3-) 3.5-4.5 mu, ellipticae vel oblongae vel subcylindricae, levi, amyloideis. Cheilocystidia 22-30 X 3-6 u. Hyphis fibulatis. Typum legit H. E. Bigeloy, n. 13331, Monroe State Park, Wash- ington Co., Vermont, 6 Aug 1964, in Herb. Univ. Mass. conservatum. Pileus 1.5-2.3 em broad, convex with a strongly incurved and in- rolled margin, surface dry and dull, heavily matted fibrillose under a lens, dingy pale buff colored (dingy "light buff"); context thick, dingy whitish, firm, odor and taste farinaceous; KOH — reddish brown. Lamellae adnate, close, narrow, thickened, forked at times, gray (*pale drab gray"), edges even. 184 Rhodora [Vol. 71 Stipe 2.5-4.2 cm long, 6-9 mm thick at apex, equal or tapering downward, solid (dingy whitish within), slightly curved at times, central, surface dry and dull, velvety under a lens, concolorous with pileus. Spores (6-)7.5-9(-10) X (3-)3.5-4.5 4, elliptic to oblong or rarely subcylindric, smooth, amyloid. Basidia 32-45 X 6-8 u, 2- and 4-spored. Cheilocystidia cylindric or nearly so, 22-30 u long, 3-6 u thick, smooth, hyaline, flexuous at times, forked or irregular at times. Pleurocystidia not found. Pileus cutis an interwoven mass of cystidioid end cells, cylindric to subclavate, sometimes broadly fusoid or irregu- lar, 4.5-11.5 # in diameter, cells 24-70 u long, brownish in KOH, pig- ment dilute in ce!l contents and in smooth thin walls, trama of similar hyphae only compact. Hymenophoral trama subregular but slightly divergent near the subhymenium, brownish in KOH, hyphae cylindric, 3-6 u in diameter. Clamp connections present. Gregarious, on soil and humus under birch and maple. August. Material examined. Vermont: Bigelow 13331 (type), Monroe State Park, Washington Co., 6 Aug 1964. All parts of the fresh carpophore show a color change to reddish or reddish brown when touched with drops of 3% KOH, but this is a curious reaction as the color is only at the edges of the drop. When revived sections of the cap and gills are mounted in KOH for microscopic study, the gills turn reddish promptly but soon fade out to pallid or yellowish. In addition, the cap and stipe darken overall to a light gray in drying. By the amyloid spores and KOH reaction on the fresh carpophore Hygrophorus fibrillosus would seem close to H. angelesianus Sm. & Hesl., but the latter is distinctive by broader spores (4-5.5 x), no cystidia, a viscid pileus, and dark sordid gray to grayish brown colors in pileus and stipe. The species of Dermoloma with amyloid spores as well as Tricholoma griseum Dennis and Tricholoma hygro- phorus Joss. nom. nud. were also compared to H. fibrillosus, but none of these have the same diagnostic characters. Tricholoma hygrophorus is perhaps the nearest to H. fibril- losus but differs principally by the broad and distant lamellae and spores 5.9-6.5 (-7) X 4.5-5.2 y. Hygrophorus marginatus var. concolor Smith, Pap. Mich. Acad. Sci., Arts, Lett. 38: 59. 1953. Material examined. Maine: Bigelow 11258, Cedar Lake, T3R9, 1969] Fungus Flora — Bigelow and Barr 185 Penobscot Co. 24 Aug 1962. Massachusetts: 9714, Hawley Bog, Franklin Co., 10 Aug 1961. New Hampshire: 12257, Pinkham Notch, White Mountains Nat. Forest, 15 Aug 1963. Rhode Island: 14406, Beach Pond State Park, Exeter Co., 22 July 1965. Vermont: 13084, Ranch Brook, near Stowe, Lamoille Co., 27 July 1964; 13223, Bing- ham Falls, near Stowe, Lamoille Co., 3 Aug 1964. Hygrophorus marginatus var. olivaceus Smith & Hesler, Lloydia 5: 40. 1942. Material examined. Maine: Bigelow 10997, Abol Field near Baxter State Park, Piscataquis Co., 14 Aug 1962; 11331, Upper Togue Pond, Piscataquis Co., 27 Aug 1962; New Hampshire: 11870, Sawyer Rock, White Mountains Nat. Forest, 20 July 1963. Vermont: 12922, Lake Mansfield, Lamoille Co., 20 July 1964; 13408, Morristown, Lamoille Co., 12 Aug 1964. Hygrophorus recurvatus Peck, New York St. Mus. Bull 157: 28. 1912. Plate 1409 Plate 1409. Hygrophorus schulzeri Bresadola. X 2. 186 Rhodora [Vol. 71 Plate 1410. Mycena citrinomarginata Gillet. X 2 a. Material examined. Massachusetts: Bigelow 14985, Conway, Franklin Co., 24 Oct 1966. Gregarious or subcespitose, on lichen-covered very rotten conifer stump in old pasture. Hygrophorus schulzeri Bresadola, Fungi Trident. 1: 57. 1884. Plate 1409. Material examined. Connecticut: Bigelow 15493, Audubon Center, Sharon, Litchfield Co., 27 July 1968. Massachusetts: 15225, Mt. Toby, Sunderland, Franklin Co., 10 Aug 1967. Gregarious or subcespitose, under hardwoods or mixed woods. Hygrophorus tephroleucus (Fr.) Fr. var. tephroleucus, Epicrisis p. 328. 1838. 1969] Fungus Flora — Bigelow and Barr 187 Material examined. Vermont: Bigelow 14366, Worcester, Wash- ington Co., 5 Oct 1964. Gregarious, on sphagnum patch in spruce woods. Mycena citrinomarginata Gillet, Les Hyménomycétes, p. 266. 1874. Plate 1410. Pileus 4-20 mm broad, subcylindrie at first becoming conic then conic-campanulate, subviscid but soon merely moist, opaque, sulcate striate, edge crenate or ragged in age, glabrous, fuscous umber fresh, hygrophanous and slowly fading to sordid yellow; context thin, not fragile, concolorous and fading with pileus, odor strong, earthy or raphanoid when flesh is crushed, taste similar. Lamellae adnate to subdecurrent, close to subdistant, narrow to moderately broad, grayish, edges fimbriate and brown marginate. Stipe 2-5 em long, up to 2 mm thick at apex, equal or the base slightly enlarged, base tomentose, surface above with a thin fibrillose coating, glabrescent, central, hollow, often curved, sordid yellow. Spores (7-)8-10(-12.5) X 4.5-6 u, elliptic, smooth, amyloid. Basidia 23.35 X 6-8 u, 4-spored. Cheilocystidia 22-55 u long, fusoid ventricose, apex 1.5-3 4 broad, 6-8 u broad at enlarged portion, branched and flexuous at times, hyaline. Pleurocystidia not found. Pileus cutis vinaceous in Melzer’s, brown in KOH at first but fading to yellowish, pigment in cell contents, surface with thin gelatinous pellicle, cutis of flexuous and branched hyphae, 1-2 u in diameter, with numerous digitate processes, some hyphae originating in hypoderm, cellular hypoderm conspicious, cells 22-50 & broad, tramal hyphae cylindric to inflated, 2-16 » in diameter. Hymenophoral trama subcellular, hyphae or cells 3.5-24 u in diameter, dextrinoid in Melzer's. Clamp connections present. Gregarious or cespitose, on lawn in the open. Material examined. Massachusetts: Bigelow 14911, Baptist Hill, Conway, Franklin Co., 2 Oct 1966; 15613, same locality, 20 Oct 1968. The collection cited appears to be the dark form of M. citrinomarginata discussed by Smith (1947). Besides pileus color, this form differs in other respects from the pale form as described by Smith. He gives the odor and taste as ‘not distinctive", the pellicle as “non-gelatinous”, and no mention is made of digitate processes as found on the cutis hyphae of no. 14911. Whether these differences are sufficient to separate the dark form with at least varietal status is uncertain in view of the extreme variability found in this species by Smith. 188 Rhodora [Vol. 71 Plate 1411 Plate 1411. Tricholomopsis sulphurea (Pk.) Bigelow. X 1%. 1969] Fungus Flora — Bigelow and Barr 189 Tricholomopsis sulphurea (Pk.) Bigelow, comb. nov. Plate 1411. Clitocybe sulphurea Peck, New York State Mus. Rep. 41: 62. 1888. Pileus 2.5-6(-9.4) em broad, convex at first with an incurved mar- gin, becoming broadly convex or plane with a decurved margin, subumbonate to somewhat umbilicate or shallowly depressed, margin even, surface glabrous, often lubricous when moist, at times slightly fibrillose at the margin under a lens, dise sometimes slightly velvety, subhygrophanous, a watery dull yellow moist (near “honey yellow”), paler yellow faded (near “eolonial buff” or “deep colonial buff"), streaked when fading; context thin, pliant, pale yellow, odor and taste mild. Orange to reddish in KOH then brown. Lamellae adnate or somewhat sinuate, close to subdistant, broad (up to 10 mm), abruptly tapered near pileus margin, not forked or intervenose, light yellow (“maize yellow” to “naples yellow"), dark- ening in age (near “chamois”) but not changing color when bruised, edges crenate. Stipe 2-4(-5) em long, 5-9(-11) mm thick at apex, equal or the base swollen, usually curved, stuffed then hollow, surface innately fibrillose striate, concolorous with pileus or lamellae but fading more slowly, eccentric at times. Spores (5.5-)6-8 X (4-)4.5-5.5 (-6) 4, broadly elliptie, smooth, not amyloid, white in deposit. Basidia 23-37(-41.5) X (4-)6-7.5 4, 4-spored, hymenium yellow to golden in KOH. Cheilocystidia abun- dant, clavate to bulbous, pedicellate, 25-58 & long, (8-)11-23(-27) u in diameter, hyaline or contents yellowish to orangish, walls slightly thickened, smooth, agglutinated in age. Pleurocystidia scattered or rare, cylindric to somewhat fusoid or subclavate, (25-)46-70 u long, 4.5-8.5 » in diameter, hyaline and shiny, smooth. Pileus section red- dish at first in KOH but soon fading, cutis dingy orangish or golden- yellow, pigment as fine encrustrations or in slightly thickened walls, cutis hyphae mostly cylindric, 4-8.5 u in diameter, trama paler than cutis, hyphae cylindric to inflated, 4-15.5 u in diameter, walls thin or slightly thickened. Hymenophoral trama of parallel or subparallel hyphae, hyaline or pale yellowish, hyphae cylindric or somewhat inflated, 4-13 & in diameter, walls thin or slightly thickened. Clamp connections present. Colored refractive hyphae present. Seattered or gregarious, on conifer logs. September and October. Material examined: Massachusetts: Bigelow 9395, Mt. Toby, Sunderland, Franklin Co., 2 Oct 1960; 11569, Shutesbury, Franklin Co., 9 Oct 1962; 14658, Conway, Franklin Co., 19 Oct 1965; 15309, Conway, Franklin Co., 10 Oct 1967. New York: C. H. Peck, Witten- burg Mountain, Sept (type) (NYS). Vermont: Bigelow 14270, Ranch Brook trail, near Stowe, Lamoille Co., 28 Sept 1964. The discovery that this species of Peck’s is a Tricholo- mopsis necessitated a comparison with three other North 190 Rhodora [Vol. 71 American species of this genus. The types of Agaricus (Pleurotus) sulphureoides Peck, Agaricus (Tricholoma ) flavescens Peck, and Tricholomopsis bella Smith, were re- examined with special attention to the details of pigmenta- tion in the pileus cutis. On the basis of microscopic char- acters, the species can be separated into two groups. A sulphureoides and A. flavescens had no encrusted hyphae when studied with a phase microscope and 100X oil im- mersion objective, but encrusted hyphae were found on the cuticular hyphae of the types Clitocybe sulphurea Peck and T. bella. In the first two, the coloration appears to be dilute in the cell contents and in slightly thickened smooth walls. In addition to the differences in pigment, C. sulphurea and T. bella had larger cheilocysts (viz. 35-60 x 15-23 uq, and 23-75 X 10-25 A, respectively) than A. sulphureoides and A. flavescens (33-50 X 7.5-13 mand 31-44 X 9-17.5 u respectively). Further distinction between the species is by means of macroscopic features. With the types of A. sulphureoides and A. flavescens (NYS) are watercolor sketches and these present different appearing agarics. A sulphureoides has numerous small dark brown scales over a dull sulphur yellow ground color and a small brownish umbo. In com- parison with the sketch of A. flavescens, A. sulphureoides is smaller, has thinner flesh, and is a generally more in- tensely colored species. A. flavescens is light yellow (almost citrine yellow), glabrous, and with a rather robust Tri- cholomoid stature. In the original descriptions Peck stated that A. sulphureoides can be smooth, and that A. flavescens can crack into minute scales on the disc, but these condi- tions are not evident in the Sketches. The microscopic characters of the two types are nearly identical. T. bella is separated from T. sulphurea also by field characters. The pileus surface of T. bella is coarsely fibril- lose with squamules on the umbo and margin. The color is darker — dull ochraceous beneath darker ochraceous fibrils in contrast to the yellow shades of T. sulphurea. The lamellae of T. bella stain dull rusty brown when bruished, 1969] Fungus Flora — Bigelow and Barr 191 a state not observed in T. sulphurea. Pleurocystidia are rare and difficult to find in sections of T. sulphurea, but are abundant in T. bella. The latter has been found in New England and is described below to add information to the type description. By the scaly cap, T. ormata (Fr.) Sing. appears to be closest to T. sulphureoides, but this European species has spores which are elliptic-cylindric, 5.5-8 X 3.3-4 u, accord- ing to Favre (1952). The cheilocysts of T. ornata often have appendages, which I did not find in the type of A. sulphureoides. There is no information given by Favre about the details of pigmentation nor about pleurocystidia. In Smith’s (1960) treatment of Tricholomopsis a fibril- lose veil was emphasized in distinguishing T. sulphureoides from T. flavescens (as T. thompsoniana) and T. bella. The relation of C. sulphurea was not known at this time and thus not compared. Of the four types, none are known definitely to have a veil by the author’s descriptions. The sketches with the types of A. sulphureoides and A. flaves- cens (NYS) do not show veils unfortunately, and the specimens of all four type collections are too expanded to show any distinct veils even if present when young. All were examined under the dissecting microscope, but the very few fibrils visible on the pilei margins or stipe apices at times were inconclusive evidence of a veil. However, veils do exist in this group as Smith reported for T. sul- phureoides, and as known now in T. bella, Bigelow no. 15325. The presence or absence of a veil is a difñcult characteristic to use in diagmosis of a species in this group though, as rarely does one collect carpophores small enough to dis- tinctly show, or rule out, this structure. Of possible use in further assisting the separation of the four species discussed above is the reaction of fresh car- pophores to 3% KOH. All of the species have not been tested yet, but two collections of T. sulphurea showed a distinct orange to red color when drops of KOH were placed on any part. Slowly, the color changed to dull brown. 192 Rhodora [Vol. 71 Plate 1412 Plate 1412. Tricholomopsis bella Smith. X 1%. 1969] Fungus Flora — Bigelow and Barr 193 Tricholomopsis bella Smith, Brittonia 12: 64. 1960. Plate 1412. Pileus (0.8-)2.5-6.5 cm broad, convex with a strongly incurved margin at first, expanding to broadly convex, with a low broad umbo at times, margin edge tomentose in buttons then appressed squamu- lose, appressed fibrillose-scaly or coarsely radiate fibrillose to the rather velvety disc, subhygrophanous at times, ochraceous (nearest “antimony yellow"), disc slightly darker in expanded pilei, paler and more yellowish in age and loss of moisture; context thin except on disc, firm, ochraceous to yellow, no odor and taste. Lamellae adnexed usually but adnate or short decurrent in large specimens, close, moderately broad, concolorous with the pileus or more yellowish, not fading, edges crenate, becoming brown in age and bruising. Stipe 2.5-4.5 cm long, 5-10 mm thick at apex, equal or base en- larged, hollow, surface fibrillose, heavily fibrillose at apex in buttons and fibrils attached to pileus edge, soon ruptured, concolorous with rnoist pileus but base becoming watery brownish when bruised. Spores 5.5-7 X (3.5-)4.5-5.5 u, broadly elliptic or elliptic, smooth, not amyloid, white to faintly cream in deposit. Basidia 36-40 X 5.5-7 u, 4-spored. Cheilocystidia clavate to clavate-bulbous, pedicellate, at times basidioid or irregular-cylindrie, hyaline or golden yellow, 23-65 u long, (4.5-)10-22.5(-26) „u broad, usually abundant. Pleurocystidia basidioid to cylindric in shape but refractive, pale yellowish, em- bedded, 31-61 » long, 5-6 u broad, scattered at times. Pileus cutis brownish golden in KOH, pigment finely encrusted or in slightly thickened walls, hypae mostly cylindric, 5.5-11.5 4 in diameter, trama pale yellow, hyphae cylindric or somewhat inflated, 4.5-15.5 u in diameter. Hymenophoral trama subparallel, yellowish in KOH (paler than hymenium), hyphae mostly slightly inflated, (6-)8.5-19 u in diameter, walls often slightly thickened. Clamp connections present. Yellow refractive hyphae present. Scattered on conifer log. September. Material examined. Massachusetts: Bigelow 15325, Conway, Frank- lin Co., 28 Sept 1967. Michigan: Smith 42508 (type), Wilderness Point, Emmet Co., 7 Sept 1953. (MICH) New Hampshire: Bigelow 12429, South Conway, Carroll Co., 4 Sept 1963. ASCOMYCETES Berlesiella nigerrima (Blox.) Sacc. Rev. Mycol. 10: 7. 1888. Figs. 1-5. Sphaeria nigerrima Bloxam ex Currey, Linn. Soc. London Trans. 22: 272. 1858. Pleospora nigerrima (Blox.) Sacc. Syll. Fung. 2: 276. 1883. Dothidea episphaeria Peck, New York St. Mus. Rept. 30: 64. 1878. Phyllachora episphaeria (Pk.) Sacc. Syll. Fung. 2: 608. 1883. 194 Rhodora [Vol. 71 Bertia parasitica H. Fabre, Ann. Sci. Nat. Bot. 6 sér. 9: 95. 1878. Homostegia parasitica (Fabre) Rehm, Hedwigia 26: 96. 1887. Berlesiella parasitica ( Fabre) v. Hóhnel, Sitzungsber. Kais. Akad. Wiss. Wien, Math.-nat. Kl. Abt. 1, 124: 60. 1915. Pachythyrium parasitica (Fabre) Arnaud, Soe. Myc. Fr. Bull. 69: 302. 1953. Nom. non spec. Trichothyrina parasitica (Fabre) v. Arx in Müller & v. Arx, Beitr. Kryptogamenfl. Schweiz 11(2): 559. 1962. Nom. non spec. Valsa pulviniceps Peck, New York St. Mus. Rept. 32: 50. 1879. Cryptospora pulviniceps (Pk.) Sacc. Syll. Fung. 9: 940. 1891. Anthostoma pulviniceps Peck in Ellis & Ev. North Amer. Pyreno. p. 578. 1892. Cucurbitaria setosa Ell. & Ev. Phila. Acad. Nat. Sci. Proc. 42: 241. 1890, non C. setosa Winter. Berlesiella setosa (Ell. & Ev.) Sacc. Syll. Fung. 9: 915. 1891. Gibberidea setosa (Ell. & Ev.) Kuntze, Rev. Gen. Pl. 3(3): 481. 1898. Cucurbitaria echinata Ell. & Ev. North Amer. Pyreno. p. 240. 1892, nom. nov. for C. setosa Ell. & Ev. Berlesiella echinata (Ell. & Ev.) Berl. Icon. Pyren. 2: 112. 1897. Pleosphaeria echinata (Ell. & Ev.) Miller, Mycologia 33: 78. 1941. Stromata velvety black, superficial on ostioles or stromata of members of Diatrypaceae, pulvinate, 1-2 mm diameter or larger by confluence, 120-460 4 high, surface regular or raised into rounded protuberances by erumpent upper portions of locules, roughened by hyphal tips, short cells, or setae up to 48 u long, dark reddish brown to black; locules entirely immersed in stroma or partially erumpent, globose or laterally compressed and conie, 50-120 mu in diameter, 85-200 u high, locule wall reddish brown, composed of several layers of polygonal cells and narrow hyaline inner layers, stromatic tissues of polygonal brown or reddish brown cells, blackened toward surface, often in upright rows in interior, opening by lysigenous canal and pore, stuffed with + gelatinizing hyphae and appearing periphysate. Asci 36-67 (7.5-)9.5-16(-24) u, oblong to saccate, bitunicate, apex rounded and thickened, narrowed to foot-like base, in gelatinizing remnants of locule tissue. Ascospores 12-20(-24) X 4-6(-7.5) 4, light dull brown or olivaceous brown, obovate-fusoid, straight to inequilateral, (3-4-)5-6-(7-) septate, one vertical septum in one or all of middle cells but not extending through end cells, slightly constricted at the septa and cells protruding out to form a rather irregular outline, one or two globules in each cell, wall smooth, over- lapping biseriate or crowded in the ascus. 1969] Fungus Flora — Bigelow and Barr 195 Superficial, hypersaprophytic on old stromata of members of Dia- trypaceae. Material examined. Massachusetts: Barr 5248, Mohawk Trail State Forest, Franklin Co., 4 Oct 1968; New York: Peck, Maryland, Otsego Co., Sept (type of Dothidea episphaeria, NYS); Richmondville, Scho- harie Co., Sept 1878 (type and isotype of Valsa pulviniceps and An- thostoma, pulviniceps, NYS). Ontario: Dearness, London, July 1894 (as Cucurbitaria echinata, Ell. & Ev. N.A.F. 3109). France: Rehm, Ascom. exs. 2175 as Bertia parasitica (MICH); Flageolet, Rigny- sur Arroux, Saone-et-Loise, 1918 as Berlesiella nigerrima (NYS). The collections listed above vary in appearance of the stroma. Completely immersed locules with only scattered hyphal tips roughening the surface occur in Dothidea episphaeria (fig. 3). Slightly protruding apices of locules, covered by numerous short black cellular contorted setae are found in Bertia parasitica (fig. 1). Mostly erumpent locules whose stromatic surface bears more elongate pointed setae in Cucurbitaria echinata and Valsa pulviniceps (fig. 2) complete the series. The locules, asci, and ascospores are all similar within the size range described. Only one rather variable species seems to be involved. Chenantais (1921) has discussed the synonymy of nigerrima — setosa — parasitica. He was unable to study Peck's Dothidea episphaeria to determine its disposition. Peck misinter- preted the fungus called Valsa pulviniceps, for his de- scription included that of perithecia of the Eutypa as well as the locules with asci and ascopores of the Berlesiella. Fabre's Bertia parasitica has been involved in the synonymy of Trichothyrium epimyces (Sace., Bomm. & Rouss.) Theissen. This latter species is also hypersaprophytic on other pyrenomycetes, but has small flattened ascostromata with radiating wall structure and hyaline one-septate ascospores. Fitzpatrick (1923), Arnaud (1953), and Miiller and von Arx (1962) included Bertia parasitica with Microthyrium epimyces and Nitschkea flageoletina Sace. However, Fabre’s original description and figures of Bertia parasitica indicate that his fungus was indeed Berlesiella rather than Trichothyrium, The genus Berlesiella is presently the only stromatic member of the small family Herpotrichiellaceae Munk. 196 Rhodora [Vol. 71 f^ E VAL SDV GS ey GN OH Wy < D E < V" 9 V | V7. 10 = Figures 1-11 Figs. 1-5. Berlesiella nigerrima: 1-3, sections through stromata of varying types, 4, ascus, 5, ascospores. Figs. 6-8. Seynesiella exigua: 6, ascostroma in section, 7, ascus, 8, ascospores. Fig. 9. Leptosphaeria macrospora, ascospores. Fig. 10. Pleospora vitalbae, ascospores. Fig. 11. Thyridaria rubro-notata, ascospore. Earlier (in Bigelow and Barr, 1963) I followed Müller and von Arx (1962) in placing members of this family among the Pleosporaceae, but now consider that the Herpotri- chiellaceae must be upheld. Despite the few species pres- ently recognized and the lack of developmental studies, the mature locule represents more the Dothidea than the Pleospora type o£ development. The family Herpotrichiel- laceae is tentatively considered most closely allied to the Capnodiaceae in the Dothideales. Seynesiella exigua Barr, sp. nov. Figs. 6-8. Ascostromata 200-265 u diametro, 100 & alto, poro 10.5-12 u pertuso, peridio lateralis brunneis, radiatis, basalis tenuissima. Asci 36-60 X 7.5-9 s, bitunicatae, oblongis vel saccatis. Ascosporae 10.5-15 X 3-4.5 mu, pallidae viridae, ellipticae, infra medium septatae. Specimen typieum in foliis emortuis Juniperi communi L., prope 1969] Fungus Flora — Bigelow and Barr 197 “Lac Diable, Mont Albert, Gaspé Provincial Park, Quebec, 19 Aug 1957”, a M. E. Barr n. 2202 lectus; in herb. Univ. Mass. depositum. Ascostromata 200-265 u in diameter, about 100 u high, pore 10.5-12 u wide, basal wall thin and light brown, side walls blackish brown, of radiating rows of cells. Asci 26-60 X 7.5-9 p, bitunicate, apex thickened, oblong to somewhat saccate, sessile, pseudoparaphysate. Ascospores 10.5-15 X 3-4.5 u, greenish hyaline, elliptic, straight to inequilateral, septate below middle, constricted at septum, wall smooth, contents with two or three globules in each cell, crowded in the ascus. On dead leaves of Juniperus communis L. Material examined. Maine: Barr 4789, Yellowhead Point, near New Harbor, Lincoln Co., 30 July 1965. New Hampshire: 3880, South Conway, Carroll Co., 11 July 1963. Quebec: 2194, 2202 (type), Lac Diable, Mont Albert, Gaspé Prov. Park, 19 Aug 1957. Vermont: 4278, Stowe Pinnacle trail, near Stowe, Lamoille Co., 11 July 1964. This fungus is similar in appearance to Seynesiella juniperi, the type species of the genus. It differs from the previously described species by its much smaller ascospores. Stigmatea sylvatica Sacc. was described from leaves of Juniperus, but has ascostromata 110-120 » in diameter and hyaline ascospores 10 X 3 y. Seynesiella, with applanate ascostromata, radiating wall structure, and thin basal wall, seems best placed in the Microthyriaceae of the Hemi- sphaeriales. Three species belonging to the Pleosporaceae are re- ported from North America for the first time. These are: Leptosphaeria macrospora (Fckl.) Thümen, Myc. univ. n. 1359. 1879. Fig. 9. Material examined. Maine: Barr 3339, on Rumex obtusifolius, Katahdin Lake trail, Baxter State Park, Piscataquis Co. 7 July 1962. Vermont: 4482, on Eupatorium fistulosum, Munroe State Park, North Duxbury, Washington Co., 6 Aug 1964. The ascospores of these two collections are in the low end of the size range given by Holm (1957) — (20-)28-33 X 4-5.5 u for the North American collections, 29-40 X 4.5-6 u for European collections. They approach L. senecionis (Fckl.) Winter in size (24-30 X 6.5-7.5 u according to Holm), but differ in shape. Those of L. macrospora taper to a point especially at the basal end, while in L. senecionis the ends are obtuse. L. macrospora is an entirely different fungus from Pyrenophora macrospora (Schroet.) Wehm., although both were originally described as Pleospora. Schroeter’s name is the later 198 Rhodora [Vol. 71 and a new name must be provided for the species now called Pyrenophora macrospora. Pleospora vitalbae (de Not.) Berl. Nuov. Giorn. Bot. Ital. 20: 70. 1888. Fig. 10. Material examined. Maine: Barr 3741, on Clematis virginiana, Katahdin Lake trail, Baxter State Park, Piscataquis Co., 1 Sept 1962. The ascospores of P. vitalbae have a single vertical septum in one or two of the middle cells, and the species was described in Leptosphaeria. This fungus is known from Clematis in Europe and also from Ilex from Argen- tina, but has not been reported from North America to my knowledge. Thyridaria rubro-notata (Berk. & Br.) Sacc. Syll. Fung. 2: 141. 1883. Fig. 11. Material examined. Vermont: Barr 4306, on Acer bark, Stowe Hollow, Lamoille Co., 14 July 1964, The grouped ascostromata of T. rubro-notata have a vinaceous tinge to the hyphae of the papillate apices. Wehmeyer (1941) de- Scribed the species from European Specimens. Enchnoa subcorticalis (Peck) Barr comb. nov. Figs. 12-14. Sphaeria subcorticalis Peck, New York St. Mus. Rept. 28: 77. 1876. Trichosphaeria subcorticalis (Peck) Sacc. Syll. Fung. 1: 454. 1882. Perithecia scattered to grouped beneath bark and often adhering to bark when the latter is removed, 540-1000 z in diameter, depressed globose and collapsing cupulate, surrounded by fringing hyphal tomentum, hyphae dark brown, thick-walled, distantly septate, 4.5-6.5 u wide; perithecial wall fragile but thick, externally 50-66 u wide, composed of layers of polygonal brown cells, internally 80-115 uw wide, composed of layers of light brown to hyaline compressed cells; apical pore minute. Asci numerous, unitunicate, arising from inner wall layers among delicate, hyaline paraphyses, of two sizes: 1) clavate, long-stipitate (55-100 X 5-6 u), p.sp. 44-55 X 10-16.5 4, with rounded or conical apex; 2) elliptical, short-stipitate, 27.5-30 >< 5.5-6.5 4, neither ascus type bearing apical apparatus or staining differentially, irregular opening visible at apex of empty asci, 8-spored. Ascospores similar in both types of asci, 7.5-11(-14) X 1.5-2.5 (-3) u, greenish hyaline, yellowish, to light olive- or gray-brown, cylindric to allantoid, ends rounded, rarely narrowed, 1-celled or at times with a pseudo- septum, contents minutely guttulate or as single globule, wall thin and smooth, clustered near apex of large asci, overlapping biseriate in small asci. On old branches of Carpinus caroliniana, among remnants of stromatic pyrenomycetes. 1969] Fungus Flora — Bigelow and Barr 199 Figures 12-20 Figs. 12-14. Enchnoa subcorticalis: 12, perithecia from above and in section, 13, two asci illustrating size differences, 14, ascospores. Figs. 15-18. Pseudomassaria inversa: 15, perithecium, 16, ascus, 17, ascus apex with amyloid annulus, 18, ascospores. Figs. 19-20. Pseudomassaria thistletonia: 19, ascus, 20, ascospores. Material examined. Massachusetts: Barr 5131, Mt. Toby, Sunder- land, Franklin Co., 14 May 1968. New York: Peck, North Greenbush, Rensselaer Co., June 1874 (type, NYS). For making disposition of Peck's fungus, the character- istics of large depressed perithecia surrounded by hyphal tomentum and allantoid ascospores agree with those of other species of Enchnoa. The formation of two sizes of asci is a distinctive feature of this species, and indeed is most uncommon in the pyrenomycetes. The genus Enchnoa is not well known and has been relegated to several families in the past. Von Hohnel’s (1909, 1918) suggestion of close relationship between Enchnoa and Calosphaeria has much merit. These two genera, with a few others, comprise the family Calosphae- riaceae in which the perithecia are scattered to grouped, 200 Rhodora [Vol. 71 with a thiek but fragile wall, containing asci in broad paraphyses, with ascospores typically allantoid, rarely elliptical or short cylindrical. In the Xylariales, the Calo- sphaeriaceae seem most closely related to the Sphaeriaceae. Pseudomassaria inversa Barr, sp. nov. Figs. 15-18. Perithecia 270 & diametro, 350 u alta, conica, erumpente vel super- ficialia, peridium usque ad 23 „u crassum, brunneum vel hyalinum introrsum, pilis septatis, 65-350 # X 4-5.5 u vestitis. Asci 65-88 X 13-17.5 n, elliptici vel oblongi, unitunicati, ad apicem cum poro amy- loideo, paraphysati. Ascosporae 17.5-21(-25.5) X 5-6.5 u, hyalinae vel viridae, ellipticae vel ovatae, apicis versus uniseptatis, cellula apicalis 6-7.5 u longa. Specimen typicum in foliis Rhododendri lapponici, prope “Alpine Gardens, Mount Washington, New Hampshire, 19 Aug 1963", a M. E. Barr n. 4104a lectus; in Herb. Univ. Mass. depositum. Perithecia erumpent to superficial, scattered on lower leaf surface, 270 u in diameter, 350 & high, conic with papillate apex, wall of large blackish brown cells, up to 23 & wide, the outermost layers polygonal and brown, inner layers compressed and hyaline, scattered setae over much of wall and grouped below papillate apex, setae septate, 65-350 mw long, 4-5.5 4 wide, not tapered, tips rounded and pallid, dark brown below. Asci 65-88 X 13-17.5 4, elliptic or oblong, unitunicate, apex rounded-truncate, base short stipitate, apical annulus refractive in water, amyloid in Melzer’s, in deli- quescent paraphyses. Ascospores 17.5-21(-25.5) X 5-6.5 gu, greenish hyaline, elliptic to ovate, straight, ends rounded, 1-septate toward upper end, not constricted, upper cell 6-7.5 u long, contents guttulate, wall smooth, overlapping biseriate in the ascus. On dead fallen leaves of Rhododendron lapponicum. Material examined. New Hampshire: Barr 3859c, Alpine Gardens, Mt. Washington, White Mountains Nat. Forest, 9 July 1963; 4104a (type), 4109b, same locality, 19 Aug 1963. This species deviates from all others in Pseudomassaria by the septation of ascospores in the upper rather than lower third. In all other respects, P. inversa is typical of the genus and is closest to the species such as P. erumpens and P. oxydendri. P. inversa cannot be confused with the other species occurring on Rhododendron leaves, P. thistle- tonia or Chaetapiospora rhododendri, since they are smaller, immersed in the host tissues or only partially erumpent. This and the following species are members of the family Amphisphaeriaceae in the Xylariales. 1969] Funzus Flora — Bigelow and Barr 201 Pseudomassaria thistletonia (Cke.) von Arx, Ber. Schweiz. Bot. Gesell. 62: 355. 1952. Figs. 19-20. Physalospora thistletonia Cooke, Grevillea 18: 74. 1890. Pseudapiospora thistletonia (Cke.) Petrak & Sydow, Ann. Mycol. 27: 98. 1929. Perithecia immersed, few in gray or pallid areas in leaf spots, 140-170 & in diameter, depressed globose, wall thin, 15 » wide, layers of light brown, compressed cells, blackened around apieal pore, pore periphysate, apex glabrous or bearing one to two setae. Asci 42-75 >x 11-15 u, elliptical, unitunicate, apex and base rounded, apical annulus amyloid, no pulvillus visible, in thin-walled deliquescent para- physes. Ascospores 15.5-20 X 5.5-7.5 u, greenish or yellowish hyaline, elliptic to obovate, tapered to rounded ends or base pointed at times, straight to inequilateral, l-septate near base, not constricted, basal cell 2-3.5(-4) u long, contents guttulate, wall thin and smooth, in fresh material surrounded by gelatinous coating 4 4 wide, obliquely biseriate in the ascus. Epiphyllous in leaf spots formed by Guignardia rhodorae (Cke.) B. H. Davis, on Rhododendron maximum. Material examined. Maryland: Cory, New Germany, 23 Aug 1946 (BPI, as Venturia rhododendri). New Hampshire: Barr 5152, Rho- dodendron State Park, Fitzwilliam, Cheshire Co., 25 June 1968. P. thistletonia was described from England and later discussions of the species apparently refer only to the type collection. The specimens cited agree with the descrip- tions. Chaetapiospora rhododendri (Tengwall) von Arx differs in conie, partially erumpent perithecia bearing numerous elongate setae, and narrow ascospores 12-18 X 3-4.5 u, P. thistletonia is closely related to P. leucothoés (Cke.) Petrak & Sydow, and may be distinguished from the latter species by the few setae and by the host plant. Records of three members of the Pezizales are included. These species appear to be collected infrequently. Plicaria trachycarpa (Currey) Boud. Soc. Myc. Fr. Bull. 1: 102. 1885. Material examined. Maine: Barr 3732, Bear Brook, T4 R9, Pisca- taquis Co., 31 Aug 1962. Pseudombrophila deerrata (Karst.) Seaver, North American Cup- Fungi (Operculates), p. 141. 1928. Material examined. Massachusetts: Barr 5141, Conway, Franklin Co., 15 June 1968. Gregarious to cespitose on overwintered cabbage stalks. 202 Rhodora [Vol. 71 This discomycete was reported from North Dakota and New York by Seaver. The collection cited contained several dozen small pinkish to brown apothecia. The micro- scopic characters agree with Seaver's (1942) and Dennis’s (1968) descriptions, although I found the paraphyses to be somewhat enlarged at the apex and encrusted with brown pigment. Trichophaea bullata Kanouse, Mycologia 50: 131. 1958. Material examined. Connecticut: Barr 5200, Sharon Audubon Center, Litchfield Co., 27 July 1968. Massachusetts: 5058, Conway State Forest, Franklin Co., 29 Aug. 1967. Apparently T. bullata is not uncommon in eastern North America, judging from these recent collections. Like several other species of Trichophaea, it has the aspect of a miniature Humaria hemispherica (Fr.) Fckl, with whitish hymenium and brown exterior and marginal hairs. The enlarged bases of the hairs provide a ready means of identification of the species. Acknowledgments We would like to thank Dr, Alexander H. Smith, Director, University of Michigan Herbarium, Dr. Robert L. Shaffer, Curator of Fungi, University of Michigan Herbarium, Stanley J. Smith, Senior Curator of Botany, New York State Museum, Dr. Chester R. Benjamin, Principal My- cologist, Mycology Investigations, Beltsville, Maryland, for the opportunities of examining type and additional material. Most of the research reported was made possible through the support of National Science Foundation Grant G 19534. DEPARTMENT OF BOTANY UNIVERSITY OF MASSACHUSETTS AMHERST 01002 LITERATURE CITED ARNAUD, G. 1953. Mycologie concrète: Genera II (suite et fin). Soc. Myc. Fr. Bull. 69: 265-306. BIGELOW, H. E. and M. E. Barr. 1963. Contribution to the fungus flora of northeastern North America III. Rhodora 65: 289-309. 1969] Fungus Flora — Bigelow and Barr 208 CHENANTAIS, J. E. 1921. Notules mycologiques. Soc. Myc. Fr. Bull. 37: 61-67. DENNIS, R. W. G. 1968. British Ascomycetes. Cramer, Lehre. FARLOW, W. G. and E. A. Burt. 1929. Icones Farlowianae. Cam- bridge. FAVRE, J. 1952. Bribes mycologiques. Ber. Schweiz. Botan. Gesell. 62: 402-411. FITZPATRICK, H. M. 1923. Monograph of the Nitschkieae. Myco- logia 15: 23-67. HESLER, L. R. 1945. Notes on southern Appalachian Fungi. VI. J. Tenn. Acad. Science 20: 233-238. VON HOHNEL, F. 1909. Fragmente zur Mykologie. 442. Über Enchnoa chaetomioides Penz. & Sace. Sitzungsber. kais. Akad. Wiss. Wien, Mathn.-nat. Kl. 118: 1505-1508. 1918. Mykologische Fragmente. CCLXII. Über die allantoidsporigen Sphaeriaceen. Ann. Myc. 16: 127-132. HoLM, L. 1957. Études taxonomiques sur les Pleosporacées. Symb. Bot. Upsalienses 14(3): 1-188. KAUFFMAN, C. H. 1921. The mycological flora of the higher Rockies of Colorado. Pap. Mich. Acad. Sci, Arts & Lett. 1: 101-150. 1925. The fungus of Mt. Hood, with some new species. Pap. Mich. Acad. Sci, Arts & Lett. 5: 115-148. MÜLLER, E. and A. J. von AnRx. 1962. Die Gattungen der didymo- sporen Pyrenomyceten. Beitr. Kryptogamenfl. Schweiz 11(2): 1-922. SEAVER, F. J. 1942. The North American Cup-fungi (Operculates). Suppl. Ed. New York. SMITH, A. H. 1942. The North American species of Mycena. Ann Arbor. 1960. Tricholomopsis (Agaricales) in the western hemisphere. Brittonia 12: 41-70. and M. B. Walters. 1947. Additional notes on the genus Armillaria. Mycologia 39: 622-625. WEHMEYER, L. E. 1941. The genus Thyridaria (Pyrenomycetes). Lloydia 4: 241-261. A REVIEW OF THE VARIETIES OF POLYMNIA UVEDALIA' JAMES R. WELLS More than fifty years ago three varieties of Polymnia uvedalia (L.) L. were described (Blake 1917). The varieties are: uvedalia — pedicels and inflorescence branches densely stipitate-glandular, sessile glands present; hispid-pilose hairs none or few; floridana — pedicels and inflorescence branches with sessile or nearly sessile glands; hispid-pilose hairs infrequent; densipilis — glands inconspicuous; hispid- pilose hairs frequent. Although authors of recent manuals and floras usually have failed to treat these varieties, they were listed in a recent review of the genus (Wells 1965). Some additional comment upon their significance is perhaps justified as part of an overall effort to achieve a better understanding of this Western Hemisphere genus. The distribution of Polymnia uvedalia by varieties (Fig. 1 plus Bermuda) is based upon examination of types as well as other herbarium material from BLH, F, GH, ILLS, IND, LSU, MEXU, MICH, MO, MSU, ND, NY, NCU, OS, OU, PH, SIU, SMU, TAES, TENN, TEX, US, and WIs. See Lanjouw and Stafleu (1964) for key to these abbreviations. Varieties are un- determined at sites indicated by “+”, and the e. central Ill. record comes from R. P. Wunderlin (pers. comm.). In com- menting upon varietal distributions, Blake (1917) noted that in the northern and eastern portion of the species range var. uvedalia predominates while in the west var. densipilis is most common. In the southern and southeastern part of the U. S. var. floridana is most frequently encountered. When the three varieties are mapped separately (Fig. 1) Blake’s statement is substantiated but it should be noted that there is considerable overlap in ranges. Thus, varietal ‘This research was supported by grants from the National Science Foundation GB-6413 and from Cranbrook Institute of Science. 204 1969] Polymnia uvedalia — Wells 205 ` ; 4 — e UVEDALIA m FLORIDANA A DENSIPILIS Fig. 1. Distribution of P. uvedalia (plus Bermuda). Varieties un- determined at sites marked “+”. distribution patterns represent only tendencies. In the field I have observed one variety growing within a few feet of another variety. The apparent absence of P. uvedalia from the northern portions of Ill., Ind., and Ohio is likely a result of prairie peninsular extension into those regions (Dan- sereau, pers, comm.). Otherwise, it seems to conform to most of the Gray's Manual range of coverage. Although Bermuda is the only location of Polymnia wve- dalia known to me outside the U. S., facets do not now support Blake's data (1917) on the occurrence of only var. densipilis on that island. Examination of Bermuda material 206 Rhodora [Vol. 71 included specimens of all three varieties and in 1966 I col- lected living material of var. floridana there. Results of attempts at hybridization among the varieties are listed below (Table 1). These results are in agreement Table 1. SUMMARY OF HYBRIDIZATION ATTEMPTS Crosses Pollen Parent | Ovule Parent) Attempted Fruits Formed P. uvedalia var. | No. Percent floridana densipilis | 68 61* 90 densipilis floridana | 59 37 63 uvedalia foridana | 53 | .10* | 19 . floridana uvedalia 50 38 76 densipilis uvedalia | 57 42 | 74 uvedalia densipilis 95 54 57 floridanaB uvedalia | 58 29° 5 uvedalia floridanaB 54 23 43 floridanaB floridana | 42 | 15* | 36 ` _ floridana floridanaB — | 51 29 57 floridanaB densipilis 65 40 62 densipilis floridanaB 54 27 50 TOTAL 706 405 57 * X° test results indicate significance at the 5% level in this reciprocal cross. B Indicates Bermuda origin of var floridana plant. with earlier reports (Wells 1966). The percentages of fruits formed are probably somewhat reduced due to accidental breakages of vascular connections between pistillate flowers and receptacle during emasculation. (Emasculation was carried out prior to anthesis in every cross.) The higher ercentages of fruits formed in this study than formerly (Wells 1966) may reflect my increased facility in handling the material. With the exception of var. floridana from Ber- muda the same cultures were used. Without emasculation, higher percentages of fruits are formed but self-fertilization and pseudogamy cannot be completely ruled out (Wells 1966). 1969] Polymnia uvedalia — Wells 207 In all of the 706 crosses only 4 plants were involved. Data involving var. floridana from Bermuda correspond to data obtained for the other two varieties from the U. S. Judging from comparative crossabilities between all varieties with var. floridana from Bermuda there is no apparent evidence for a distinct genetic race on this island. An unexpected reduction in fertility is indicated when var. floridana of U. S. origin is employed as the ovule parent. Significant differences involving this individual may arise due to chance occurrence of a genotypically unique floridana plant from the United States. Such events may occasionally be observed when minimal numbers of plants are involved in an experi- mental treatment. Parent plants as well as their progeny had gametic counts of n=16 and meiotic pairing appeared normal. Only one cytological anomaly was detected and this tetraploid re- sulted from the cross of vars. floridana and uvedalia. Pollen stainability using analine blue in lactophenol was at least 90% in all cases, The fruits of Polymnia uvedalia have thick, woody coats. Removal of at least 50% of the fruit coat prior to their placement on moist filter paper resulted in almost 100% germination. The seedlings grew rapidly under greenhouse conditions and the F, plants began to flower in about 70 days. Fruits matured approximately 21 days later. Vestiture differences between vars. densipilis and flori- dana represent, to a great extent, a matter of degree rather than kind; therefore, phenotypic data from F,, F., or back- crosses would represent at best a mere guess. Thus, Tables 2, 3, and 4 lack data showing crosses between these two varieties. It is clear from Table 1 that vars. densipilis and floridana can be crossed reciprocally. Seeds from these reciprocal crosses are viable and a total of 23 plants flow- ered. Peduncle vestiture appeared to be “intermediate” al- though this represents only a crude estimate. Backcross data involving var. floridana as the ovule par- 208 Rhodora [Vol. 71 Table 2. SUMMARY OF BACKCROSSES AND CROSSES AMONG F, PLANTS Pollen Parent| Ovule Parent Crosses Fruits Formed (Variety) (Variety) Attempted No. Percent flor x uve uve 39 6 15 flor x uve flor 100 35 35 flor x uve florB | 84 64 76 = Total 223 105 47 dens x uve wee 7 3 43 dens x uve dens 59 25 42 uve x dens dens 25 15 60 Total 91 43 47 Total 314 148 47 dens x uve dens x uve 22 10 45 uve x dens dens x uve 7 3 43 dens x uve uve x dens 28 5 18 florB x uve florB x uve Petal ot = in florB x uve uve x florB 25 11 44 uve x florB uve x florB 41 28 68 uve x florB florB x uve 8 T oS 39 26 67 Total 113 72 64 Total 170 90 53 ent (Table 2) show substantially higher percentages of fruits formed when floridana from Bermuda was the ovule parent than was obtained by employing its United States counterpart as the ovule parent. This is in agreement with data from the corresponding crosses in Table 1, again point- ing up the unique genotype of the floridana plant from the United States. The percentages of fruits formed from crosses involving F, plants (Teble 2) are sufficiently high that only slight reduction in fertility, if any, is indicated for the second generation. The F. seeds germinated readily but more than half of them did not flower. A suggestion of genetic inccm- patability due to excessive homogeneity is obtained via 6 albino seedlings which appeared in the F. generation. These 6 seedlings resulted from crosses involving all varieties. What then is the mode of inheritance of the varietal char- 1969] Polymnia uvedalia — Wells 209 Pl. 1413. Peduncle vestiture variation from the same plant. A = var. uvedalia while B lies between vars. floridana and densipilis. acters? An examination of the data in Table 3 shows that offspring may resemble either parent or may be “inter- mediate." It is well to keep in mind that more than one category of intermediacy may obtain but a detailed pubes- cence measurement per unit area not only would be imprac- tical but meaningless because peduncle vestiture may vary altitudinally on the same plant. For purposes of this study I arbitrarily used the uppermost 1 cm of peduncle vestiture to characterize each plant. Pilose hairs may be longer or shorter in offspring than in parents. Glandular hairs vary likewise but independently. Although plants in culture for 6 years have maintained their varietal integrity, the photo- graphs below are of two “good” varieties from different peduncles of the same plant (Pl. 1413). The Fig. A would key out to var. wvedalia, while B lies between var. densipilis and floridana. Other similar cases were observed. Thus, although some genetic basis for varietal traits is adequately demonstrated (Table 3) there is likewise evi- dence for environmental influences upon phenotypic expres- sion. One might also hypothesize highly mutable gene(s) 210 Rhodora [Vol. 71 Table 3. PHENOTYPES OF F;, Fz, AND BACKCROSS PLANTS. Pollen Parent Ovule Parent Peduncle Vestiture Type Total (Variety) (Variety) Like Intermediate Like Pollen Ovule Parent Parent — florB uve 0 7 1 8 uve florB 3 2 1 6 flor uve 3 3 2 8 uve flor 0 1 2 3 uve dens 1 2 0 3 dens uve 0 2 11 13 uve x florB uve x florB 0 1 0 1 dens x uve dens x uve 0 1 0 1 uve x florB florB xuve 2likevar.uve 6 0 8 florB x uve florB x uve 0 1 0 1 florB x uve uve x florB 1 1 0 2 florB x wve florB 0 13 0 13 causing such a phenotypic aggregation. With respect to P. wvedalia Blake (1917) points out that “On the Bermuda Islands . . . it is certainly introduced and seems to be rapidly naturalizing itself. The plant is not noticed in Reade's list of Bermuda plants (1883), and as it is unlikely that so conspicuous a plant could have escaped his notice, it seems very probable that it is a recent introduction." Owing to the tendency for individual varietal distributions to occur in different parts of the United States it is unlikely that more than one variety was originally introduced there although all three are recorded from there now. The var. floridana is a likely candidate for the species' first establishment since its range of distribution includes portions of the United States closest to Bermuda. (North Carolina is some 568 miles distant.) Moreover, this variety is approximately 20% self-fertile, which is the highest incidence reported (Wells 1966) among the three varieties, Outbreeding from a single var. floridana introduction could have produced the present range of phenotypes. Future comments upon peduncle vestiture in P. wvedalia must take into account the following facts: 1969] Polymnia uvedalia — Wells 211 1) Two types of pubescence obtain; namely, stipitate glandular hairs and pilose hairs. 2) The two hair types appear to vary independently in size and in frequency. 3) Results from some crosses suggest a quantitative in- heritance of peduncle vestiture although other crosses indicate a simple dominant-recessive relationship. 4) Individual peduncle vestiture often varies from the proximal to distal regions. 5) Environmentally induced variation is possible as in- dicated by the different vestiture types “varieties” ob- served on peduncles of the same plant. In view of all the facts, little usefulness is realized in continued recognition of the three varieties of P. uvedalia according to Blake (1917). What is true for P. uvedalia likewise probably holds for the closely related P. maculata from Mexico and Central America which Blake treated sim- ilarly. CRANBROOK INSTITUTE OF SCIENCE BLOOMFIELD HILLS, MICHIGAN 48013 REFERENCES CITED BLAKE, S. F. 1917. Polymnia uvedalia and its varieties. Rhodora. 19: 46-48. LANJOUW, J., and F. A. STAFLEU. 1964. Index Herbariorum. Inter- national Bureau for Plant Taxonomy and Nomenclature of the International Association for Plant Taxonomy. Utrecht, Nether- lands. WELLS, J. R. 1965. A taxonomic study of Polymnia (Compositae). Brittonia. 17: 144-159. 1966. Some experimental taxonomic studies of Polym- nia (Compositae). Va. Jour. Sci. 17: 120-127. IRIS VERNA L.: LECTOTYPE AND VARIETY DESCRIPTION' MAURICE E. EDWARDS? Linnaeus (1753) in his description of Iris verna did not specify a plant specimen. Instead, he cited two documentary references, (1) Gronovius Flora Virginica (1739), which cited a single specimen collected by John Clayton in Vir- ginia, and (2) Plukenet's Almagestum Botanicum (1696), which recognized material collected by John Banister and also referred to an earlier work of Plukenet, Phytographia (1691), containing the first known reference to I. verna. Consequently a holotype is not absolutely discernible from the original Linnaean publication. Since Linnaeus is known to have assisted Gronovius in the preparation of Flora Virginica, many Clayton specimens cited in that work are today recognized as Linnaean types (Fernald, 1940). Adding to this the idea that Linnaeus utilized Flora Virginica in the preparation of Species Plan- tarum, it is only reasonable to favor the Clayton specimen when selecting the type for the species. However, while examining specimens on loan from 31 herbaria in the United States, an instance was noticed where photographs of Linnaean herbarium specimens, including one specimen probably misidentified as J. verna and with duplicate photographs deposited in two herbaria, had been mistakenly labelled as types. The circulated photographs do not resemble an J. verna. Moreover, Dykes (1913) in his monograph of Iris had stated that the Linnaean herbarium specimen of this species was incorrectly identified and was "This article results from literature and herbarium research for a Master of Arts thesis directed by Dr. R. B. Channell (Chairman, De- partment of General Biology, Vanderbilt University, Nashville, Ten- nessee), Whom the author wishes to acknowledge for his suggestion of the problem and assistance throughout the study. "Present address: Department of Bacteriology & Botany, Syracuse University, Syracuse, New York 13210. 212 1969] Iris verna — Edwards 218 probably an I. prismatiea Pursh, which the circulated photo- graphs resemble. The Clayton specimen, as first suspected from photo- graphs and later confirmed by written communication from Dr. William T. Stearn of the British Museum, does not possess rhizomes, which are critical to the identification of the two currently recognized taxa — var. verna (coastal element) and var. Smalliana (montane element). John Clayton is known to have collected extensively in the Coastal Plain of Virginia where the coastal element of I. verna grows. He is also known to have collected in other physio- graphic provinces of the state where the montane element grows, including the Blue Ridge and Shenandoah Valley (Fernald, 1940). Therefore, the actual source in Virginia of the Clayton specimen is to some degree questionable. In the absence of compelling evidence to the contrary it seems appropriate to associate the Clayton specimen with the coastal form of variation, as initially was done by Small (1929) and presumably also by Fernald (1947). To avoid possible future confusion Clayton’s specimen 253 of Iris verna is hereby designated the lectotype of the species. It is preserved in Herbarium Gronovius, British Museum (Na- tural History). Iris verna, distributed between Pennsylvania and the Gulf Coast, consists of at least two geographical varieties, de- scribed by Small (1929, 1931) as mountain and coastal forms. Later the montane element was proposed by Fernald (1947) as a new taxonomic variety, but Fernald inadver- tently failed to provide a Latin diagnosis when he proposed I. verna var. Smalliana, and thus technically the varietal name of the taxon is considered invalid. Therefore, to ful- fill the requirements of Article 36 of the International Code of Botanical Nomenclature (Lanjouw, 1966) the following Latin description is presented. Iris verna L. var. Smalliana Fern. var. nov. Iris vernalis pumila; corolla imberbis; odor nullus aut minor quam in typica; internodia brevia et radices multae secundum rhi- zoma; folia fere longiora et latiora quam in typica. Iris 214 Rhodora [Vol. 71 verna L. var. Smalliana Fern. in Rhodora 49: 214-215 (1947), nom. nud. and I. verna, (Mountain Form) Small in Addisonia 14: 15, plate 456 (1929). (GH — type, locally abundant in the oak barrens, east of Crossville, Cumberland County, Tennessee, Svenson 7635, 2 May 1936; Isotypes — BKL, TENN.) LITERATURE CITED Dykes, W. R. 1913. The Genus Iris. University Press, Cambridge. FERNALD, M. L. 1940. Chron. Bot. 6: 27-28. . 1947. Rhodora 49: 214-215, LANJoUW, J. 1966. International Code of Botanical Nomenclature. Utrecht. LINNAEUS, C. 1753. Species Plantarum 1: 39-40. SMALL, J. K. 1929. Addisonia 14: 15-16. ——. 1931. Addisonia 16: 87-88. DEPARTMENT OF GENERAL BIOLOGY VANDERBILT UNIVERSITY, NASHVILLE, TENNESSEE 37208 "The author wishes to acknowledge Dr. F. L. Newton (Classical Studies Department, Vanderbilt University) for his correction of the Latin diagnosis. THE VASCULAR FLORA OF OKLAHOMA — ADDITIONS AND COMMENTS B. JoHN TAYLOR AND CONSTANCE TAYLOR By the last of February, a number of plants may be found in flower or fruit in the United States south of the 36° north latitude, especially in the central and eastern portions. Few botanists, however, care to go afield when they must go encumbered by sufficient clothing to ward off the sharp wind and cold, wet ground of late winter or early spring. We began field work this spring (1968) the last two weeks in February. This proved very rewarding and thoroughly enjoyable, especially after a particularly wet, dreary fall and winter. Starting collections in the early part of the growing season has at least two advantages: (1) Very small plants may be seen at this time before the major plant cover has developed. Some small annual plants bloom and produce fruit only in the early part of the growing season and then are gone; others continue to flower into summer, but being covered by larger plants are seldom found and are sometimes missed entirely. (2) One may also gain a new and broader perspective of the flora. Some early flowering plants have one growth form at this time; but, later flowering specimens may look entirely different. It may be that, in some cases, different seasonal ecotypes com- prise populations that continually flower throughout much of a growing season. Occasionally, specimens of the early flowering forms are encountered in a folder composed mainly of specimens collected later in the growing season. They are usually ignored as atypical and not representative, when indeed they are quite typical of one flowering form of the species. This report is based mainly on the results of our field work in 1968. Five of the following species are reported as new to the Oklahoma flora and are marked with an asterisk. All specimens cited were collected by John and Connie 215 216 Rhodora [Vol. 71 Taylor and are deposited at the Bebb Herbarium, Norman, Oklahoma, with duplicates of most deposited at SMU, Dallas, Texas. Ophioglossum crotalophoroides Walt. This diminutive member of the Adder’s-tongue ferns was recently reported as new to the Oklahoma flora from Bryan County (Taylor and Taylor 1967). From herbarium material and literature available, it might be concluded that this species is rare. Our observations this spring indicate this is not so. It was found in many sandy, overgrazed pastures throughout the northwestern portion of the Old Gulf Coastal Plain. Proba- bly, because of its very small size and early fruiting, it is often overlooked. Fruiting specimens were found in Bryan County as early as February 10 and no later than May 1. We now have additional specimens from ‘the following sites in OKLAHOMA: ATOKA Co., 4811, 5 mi. ne of COLEMAN; CHOCTAW Co., 4795, 2 mi. w of BOSWELL; JOHNSTON Co., 4814, 3 mi. e, 1 mi. s of FILLMORE; MARSHALL Co., 4855, 2 mi. w, 4.5 mi. s of KINGSTON; MC- CURTAIN CO., 4881, 8 mi. w of IDABEL and 4864, 1.5 mi. w of VALLIANT. Specimens were also collected from the following locations in TEXAS: BowiE Co., 4842, 1.25 mi. w of DEKALB; FANNIN Co., 4832, 2 mi. ne of IVANHOE; LAMAR Co., 4833, 4 mi. ne of DIRECT; RED RIVER Co., 4838, 1 mi. e of WOODLAND. *Ophioglossum vulgatum L. This larger member of the Adder's-tongue ferns has been reported from a number of Texas counties including Denton, from Hempstead Co., Arkansas, and from Crawford Co., Kansas (Clausen 1938, Correll 1956). An Adder's-tongue fern was collected from a number of swampy areas and wet sandy bottomlands in Oklahoma and Texas this spring. These specimens proved to be O. vulgatum. This species usually has a much wider sterile blade and its rounded tip always lacks the apiculate tip of O. engelmanni, which it somewhat resembles, The veins of O. engelmanni form large areoles inclosing smaller ones; O. vulgatum lacks this characteristic. Likewise O. engelmanni is almost always found in limestone or tight clay soils in this area, while O. vulgatum occurs in sandy, wet bottoms. Its fruiting fronds can be found as much as three weeks earlier. 1969] Oklahoma — Taylor and Taylor 217 We have the following collections from OKLAHOMA: BRYAN Co., 4829, 4.5 mi. e of BENNINGTON, and 4846, 2 mi. w, 2.5 mi. n of OBER- LIN; CHOCTAW Co., 4875, 4 mi. n of SAWYER; McCunTAIN Co., 4897, 2 mi. n of VALLIANT; PUSHMATAHA Co., 4877, 1.5 mi. s of OLETA. TEXAS locations are: LAMAR Co., 4837, 1.5 mi. w of FAULKNER; RED RIVER Co., 4841, 1.5 mi. e of WOODLAND. *Lepuropetalon spathulatum (Muhl.) Ell. This very small member of the Saxifragaceae has been reported as close to Oklahoma as Kaufman Co., Texas (Shinners, 1958). Flow- ering specimens of this species were found in Bryan Co., Oklahoma in mid-February. By the last of April, it could not be located in any of the areas from which it had been previously collected. It is usually common, sometimes abundant, where it is found — in sandy overgrazed pas- tures, old fields, and along roadsides. It is probably quite abundant and wide spread throughout its range, but is seldom collected because of its early flowering date and minute size. All of our specimens had only five stamens instead of the reported ten. All collections listed below were collected from soils derived from sandstone formations. OKLAHOMA locations are as follows: ATOKA Co., 4810, 5 mi. ne of COLEMAN; BRYAN Co., 4803, overgrazed pasture on northwest edge of DURANT; CHOCTAW Co., 4863, 2 mi. w. of BOSWELL; JOHNSTON CoO., 4813, 3 mi. e, 1 mi. s of FILLMORE; MARSHALL Co., 4856, 2 mi. w, 4.5 mi. s of KINGSTON; McCurTAIN Co., 4886, 2 mi. n of the Little River; PUSHMATAHA Co., 4880, 2.5 mi. e of oLETA. We also have specimens from the following TEXAS locations: BOWIE Co., 4843, 1.25 mi. w of DEKALB; FANNIN Co,. 4831, 2 mi. ne of IVANHOE; GRAYSON Co., 4835, 2.5 m. e of DENISON; LAMAR Co., 4834, 4 mi. ne of DIRECT; RED RIVER Co., 4839, 1.5 mi. e of WOODLAND. *Schrankia occidentalis (W & S) Standl. Turner (1959) lists the range of this species of sensitive briar as the lower panhandle of Texas, occurring as far north as Crosby Coun- ty. Our specimen 3865 was collected in the spring of 1967 from along U.S. 287, 3 miles north of the Oklahoma-Texas boundary in Cimarron County. Leaflets of our material are smooth and lack the pronounced veination of S. uncinata, the other species known from this area. Epilobium glandulosum Lehm. var. adenocaulon (Haussk.) Fern. was reported as new to the Oklahoma flora 218 Rhodora [Vol. 71 by Taylor and Taylor (1967). The specimens on which the report was based came from an isolated population in a branch of Tessequite Canyon in extreme western Cimarron County. Abundant material of this species has been subse- quently collected from a canyon 16 miles N and 2 miles W of Boise City in northcentral Cimarron County. *Verbena brasiliensis Vell. Specimens of this South American adventive were collected from a roadside ditch about three-fourths of a mile NE of the small Johnston Co. community of Troy, Oklahoma. Gleason (1963) gives its range in this area as Arkansas and Louisiana and Gould (1962) lists it from Texas. It differs from a similar species, V. bonariensis, known to the Oklahoma flora, by absence of sessile amplexicaulate leaves. Our material has very narrow petioles, which sometimes may have a very thin margin along the basal portions. Spikes tend to be less compact, narrower, and a little longer than those of V. bonariensis. Hedyotis rosea Raf. was reported to the Oklahoma flora from northeast and north central Oklahoma by Waterfall (1953). His Keys to the Flora of Oklahoma (1966) lists it from a number of scattered locations in the same general area. We have seen Bebb Herbarium specimens from Cleve- land County, and collected it in Johnston County during the spring of 1963. Collections this spring added Specimens from Bryan, Marshall, MeCurtain, and Pushmataha Coun- ties, indicating this plant grows throughout the eastern half of Oklahoma. It is often found with the ubiquitous H. erass- ifolia, but no intermediates were observed. Although H. crassifolia is rather variable in this area in flower color, flower size, as well as other characteristics, H. rosea is easily distinguished where the two occur together. From our ob- servations, it seems that H. rosea is never as abundant, begins to flower slightly later, completes flowering sooner, and tends to favor habitats with slightly different edaphic conditions than H. crassifolia where they occur in the same area. It may be that a mechanism similar to that described by Lewis and Moore (1959) for H. australis is also involved in H. rosea. 1969] Oklahoma — Taylor and Taylor 219 * Aster brachyactis Blake. Specimens of this western aster were collected 7 miles east of Kenton on Willow Creek, in western Cimarron Co., on Oct. 4, 1964. The material had been grazed by stock, thus was short and somewhat atypical in growth form. Dr. G. J. Goodman was kind enough to identify our 2552. It does not appear to have previously been reported for our Oklahema flora. SOUTHEASTERN STATE COLLEGE DURANT, OKLAHOMA 74701 LITERATURE CITED CLAUSEN, R. T. 1938. A monograph of the Ophioglossaceae. Mem. Torr. Bot. Club 19: 125. CORRELL, D. S. 1956. Ferns and Fern Allies of Texas, Renner, Texas: Texas Research Foundation. GLEASON, H. A. 1963. Illustrated Flora of the Northeastern United States and Adjacent Canada. New York City, N. Y.; Hafner Publishing Co., 3 vols., Ed. 3. GouLp, F. W. 1962. Texas Plants — A checklist and ecological sum- mary. Texas Agr. Exp. Sta. MP-585. Lewis, W. H. and D. M. Moore. 1958. Hedyotis australis (Rubiaceae), a new species from the south central United States. Southwest- ern Natur. 3: 208-11. SHINNERS, L. H. 1958. Spring Fiora of the Dallas-Ft. Worth area, Texas. Publ. by Lloyd H. Shinnners, Dallas. TAYLOR, R. JoHN, and C. TAYLOR. 1967. Additions to the vascular flora of Oklahoma — II. Rhodora 69: 481-483. TURNER, B. L. 1959. The Legumes of Texas. Austin: Univ. Texas Press. WATERFALL, U. T. 1953. The identity of Hedyotis rosea Raf. Rhodora 55: 201-203. . 1966. Keys to the Flora of Oklahoma, ed. 3, Res. Found., Okla. State Univ., Stillwater. THE BROMELIACEAE OF BOLIVIA LYMAN B. SMITH Continued from page 57 2. Fosterella L. B. Smith Terrestrial, ombrophilous, stemless. Leaves rosulate, entire or ser- rulate, the blades more or less narrowed toward base. Scape erect, elongate. Inflorescence usually laxly paniculate, often tripinnate. Flowers small, homomorphic, perfect. Sepals free. Petals free, naked much exceeding the sepals, usually white. Stamens of the inner series very short-adnate to the base of the petals, the anthers linear, basi- fixed, the pollen dry, wind-borne. Ovary wholly superior, glabrous, the placentae elongate, the style very slender. Capsule septicidal or loculicidal. Seeds bicaudate. a. Flowers neither secund nor nutant; inflorescence amply tripinnate. b. Stamens longer than the petals; leaf-blades 8 mm. wide. ............ ——— RERRRRRRRERRRRRRRERRRRRRRRREERRRN l. F. Pearcei. b. Stamens shorter than the petals; leaf-blades 20-30 mm. wide, serrulate toward base. ............ ees 2. F. albicans. a. Flowers secund or nutant, usually both, c. Leaf-blades serrulate, at least toward base. d. Leaf-blades serrulate for most of their length, pungent; scape- bracts serrulate; flowers 5 mm. long. ........ 3. F. Weddelliana. d. Leaf-blades serrulate only toward base, filiform-acuminate; scape-bracts entire; flowers 3 mm. long. e. Scape-bracts bladeless; leaf-blades flat, uniform. ............... ———— HR 4. F. Rusbyi. e. Scape-bracts with large foliaceous blades; leaf-blades chan- neled. MM 5. F. graminea. e. Leaf-blades entire. f. Inflorescence densely villous; floral bracts more than twice as long as the pedicels, acuminate. ......... sss 6. F. villosula. f. Inflorescence glabrous; floral bracts merely acute, mostly Shorter than the pedicels. g. Flowers 5 mm. long; sepals subtriangular. ........ ee RR 7. F. schidosperma. g. Flowers 8-9 mm. long. h. Sepals obtuse; leaves to 25 cm. long. .... 8. F. penduliflora. h. Sepals acute; leaves about 15 em. long. ........ 9. F. gracilis. 1. Fosterella Pearcei (Baker) L. B. Smith in Phytologia, 7: 172 (1960). Lindmania Pearcei (Baker) Mez of Catalogue. Flowering to 5 dm. high. Leaf-blades linear, narrowed toward base, caudate- 220 1969] Bromeliaceae — Smith 221 attenuate, 8 mm. wide, channeled, covered beneath with pale appressed scales, glabrous above. Scape erect. Scape-bracts strict, densely im- bricate, subfoliaceous. Inflorescence laxly tripinnate, subthyrsoid, attenuate, 4 dm. long, arachnoid. Primary bracts linear-triangular, attenuate, shorter than the sterile bases of the branches. Branches ascending, lax. Floral bracts triangular-ovate, exceeding the pedicels, the upper 3.5 mm. long. Flowers erect or suberect, not notably secund, 3.5 mm. long. Sepals semiorbicular, broadly rounded, 2 mm. long. Petals slightly exceeding the sepals. Stamens exserted, the anthers 2 mm. long. DEPARTMENT UNKNOWN: Butuco, 1865, Pearce s. n. (BM; type). LA PAZ: SUR YUNGAS: Basin of Rio Bopi, San Barto- lome near Calisaya, 750-900 m., Krukoff 10328 (GH, NY). 2. F. albicans (Griseb.) L. B. Smith in Phytologia 7: 171 (1960). Flowering 8-10 dm. high. Leaf-blades linear, soft-acuminate, nar- rowed and serrulate toward base, 20-30 mm. wide, cinereous-lepidote especially beneath. Scape erect, 7 mm. in diameter. Scape-bracts sub- foliaceous, much exceeding the internodes, entire. Inflorescence tripin- nate, thyrsoid, to 4 dm. long, pale-lepidote. Primary bracts narrowly triangular, shorter than the branches. Branches divergent, subdense- ly 10-20-flowered. Floral bracts broadly ovate, acuminate, 4-5 mm. long, slightly exceeded by the sepals. Pedicels 1.5 mm. long. Flowers spreading, not secund. Sepals ovate, subobtuse, 4 mm. long. Petals 5.5 mm. long. COCHABAMBA: CHAPARE: Sacaba, Steinbach, 5699 (LIL). Argentina. 3. F. Weddelliana (Brongn. ex. Baker) L. B. Smith in Phytologia, 7: 172 (1960). Lindmania Weddelliana (Brongn.) Mez of Catalogue. Flowering 5 dm high. Leaf-blades ligulate, narrowed to a pungent apex, somewhat constricted at base, ca. 16 cm. long, 20 mm. wide, serrate at least at base with spines 0.5 mm. long, brown-lepidote especially beneath. Scape slender, glabrous. Scape-bracts imbricate, serrulate. Inflorescence laxly tripinnate, 20 cm. long, glabrous. Pri- mary bracts narrowly triangular, caudate-attenuate, much shorter than the branches. Branches erect, very slender, sublaxly 15-30- flowered. Floral bracts suberect, broadly elliptic, short-acuminate, hyaline, the upper 2 mm. long. Pedicels slender, exceeding the floral bracts. Flowers secund, nutant, 5 mm. long. Sepals ovate-elliptic, broadly rounded, 2 mm. long. Petals white, exceeding the stamens. LA Paz: Yungas, Weddell 4233 (P; type). 4. F. Rusbyi (Mez) L. B. Smith in Phytologia, 7: 172 (1960). Lindmania Rusbyi Mez of Catalogue. Flowering 5 dm. high. Leaf- blades linear, attenuate to a soft filiform apex, slightly narrowed toward base, 2 dm. long, somewhat dimorphic with broad and narrow forms, to 25 mm. wide, chartaceous-membranaceous, sparsely serru- late at base, pale green and glabrate above, covered beneath with a 222 Rhodora [Vol. 71 membrane of whitish scales. Scape erect, slender, glabrous. Scape- bracts lanceolate, filiform-attenuate, longer or shorter than the inter- nodes, entire. Inflorescence laxly 2-3-pinnate, 3 dm. long, the axes glabrous. Primary bracts like the scape-bracts, much shorter than the branches. Branches curved-ascending, elongate very slender, laxlv 8-15-flowered. Floral bracts broadly ovate, acute, minute, membrana- ceous. Flowers nutant, secund 3 mm. long. Sepals ovate-elliptic, rounded, 1.5 mm. long. Petals greenish. Stamens included. La PAZ: Yungas, Bang 2571 (B; type; BM, F, G, GH, K, NY, US); Nor YUNGAS: Polo-Polo, Coroico, 1100 m., Buehtien 3674 (NY, US); Mil- luguaya, 1300 m., Buchtien 4285 (US). Peru. 5. F. graminea (L. B. Smith) L. B. Smith in Phytologia, 7: 171 (1960). Lindmania graminea L. B. Smith of Catalogue. Flowering over 1 m. high. Leaf-blades linear, filiform-attenuate, contracted at base but not really petiolate, 6 dm. long, 12 mm. wide, strongly channeled, subdensely serrate toward base with spines 1 mm. long, glabrous above, covered beneath with a membrane of white scales. Scape erect, elongate, slightly flexuous, 5 mm. in diameter, glabrous. Scape-bracts very large but spreading and not concealing the scape, entire, the lowest foliaceous with small sheaths, the highest filiform- attenuate from a very narrowly triangular base. Inflorescence laxly and very amply tripinnate, over 6 dm. long, glabrous. Lowest pri- mary bracts like the top scape-bracts, large but shorter than the branches, the highest greatly reduced. Branches curved upward, laxly secund-flowered. Floral bracts broadly elliptic, less than half as long as the pedicels. Pedicels slender, 2 mm. long. Sepals elliptic, obtuse, 1.5 mm. long. Petals 3 mm. long, equaling the stamens, white. LA PAZ: LARECAJA: Mapiri, San Carlos, 700 m., Buchtien 417 (US; type). 6. F. villosula (Harms) L. B. Smith in Phytologia, 7: 172 (1960). Lindmania villosula Harms of Catalogue. Propagating by basal off- shoots, flowering over 6 dm, high. Leaves few, rosulate, 2-3 dm. long, entire, membranaceous, the blades linear-lanceolate, attenuate, nar- rowed toward base, 25-40 mm. wide, sparsely flocculose beneath, glabrous above. Scape erect, flocculose. Seape-bracts densely imbri- cate, linear-lanceolate, attenuate, 3-5 cm. long, thin. Inflorescence laxly subtripinnate, subcylindric, 25 cm. long, many-flowered, floccu- lose-villous. Primary bracts like the scape-bracts, 2-4 cm. long. Branches curved-ascending, subdense. Floral bracts lance-ovate, at- tenuate, nearly 7 mm. long. Pedicels very short. Flowers nutant, secund. Sepals oblong-ovate, subacute, 4-5 mm. long. Petals broadly oblanceolate, subobtuse, 7 mm. long. Stamens included. L4 PAZ: SUR YUNGAS [?]: Beni River Valley, 1000 m., Rusby 393 (NY); Cocna- BAMBA: CHAPARE: Incachaca to San Antonio, 1600 m., Werdermann 2120 (B; type). 1969] Bromeliaceae — Smith 223 7. F. schidosperma (Baker) L. B. Smith in Phytologia, 8: 500 (1963). Cottendorfia Rusbyi Baker in Bull. Torr. Bot. Club, 29: 697 (1902). Lindmania Weberbaueri Mez of Catalogue. Flowering 9 dm. high. Leaf-blades broadly lanceolate, attenuate, narrowed toward base, 25 cm. long, 4 cm. wide, entire, membranaceous, glabrous. Scape erect, slender, glabrous. Scape-bracts ovate, attenuate, shorter than the internodes, entire. Inflorescence laxly bipinnate, 3 dm. long, 75 mm. wide, many-flowered, glabrous. Primary bracts like the scape-bracts, many times shorter than the branches. Branches curved- ascending, 8 cm. long, lax. Floral bracts ovate, acute, 1.5 mm. long, shorter than the pedicels, erect. Flowers nutant, rather obscurely secund, 5 mm. long. Sepals subtriangular, obtuse, 2.5 mm. long. Petals subelliptic, acute, erect at anthesis. Stamens included. LA Paz: Yungas, 2000 m., Rusby 2541 (NY; type of Cottendorfia Rusbyi Baker; US); Sur Yungas: Rio Bopi Basin, San Bartolome, Calisaya, 750-900 m., Krukoff 10482 (GH, US). Peru. 8. F. penduliflora (C. H. Wright) L. B. Smith in Phytologia, 7: 172 (1960). Lindmania penduliflora, (C. H. Wright) Stapf of Cata- logue. Flowering 25-60 cm. high, propagating by suckers. Leaf- blades linear-lanceolate with a soft narrow apex, 35 em. long, 35 mm. wide, thin, glabrous above, sparingly scurfy beneath. Scape slender, glabrous. Scape-bracts lanceolate, attenuate, entire, the upper remote. Inflorescence laxly paniculate, 2 dm. long, glabrous. Primary bracts like the upper scape-bracts, 5-10 mm. long, hyaline. Branches simple or the lowest divided and 2 dm. long, laxly many-flowered. Floral bracts ovate, acuminate, 1-1.5 mm. long, thin, green. Pedicels slender, 2, rarely 3, mm. long. Sepals lanceolate, obtuse, 3.5 mm. long, green with white margins. Petals lance-oblong, 8-9 mm. long, over 2 mm. wide, white. Stamens included. SANTA CRUZ: CORDILLERA: Lagunillas, Cordillera of Cara, 1200 m., Cárdenas 2856 (GH); CHI- qurros: Robore, 416 m., Cárdenas 2994 (GH). Peru, Argentina. 9. F. gracilis (Rusby) L. B. Smith in Phytologia, 7: 171 (1960). Catopsis gracilis Rusby in Bull. N. Y. Bot. Gard. 6: 489 (1910). Lindmania gracilis (Rusby) L. B. Smith of Catalogue. Flowering 30-45 em. high. Leaves few, 15 cm. long, entire, the outer ones white-tomentose-lepidote, the blades linear-lanceolate, attenuate, nar- rowed toward base, 11 mm. wide. Scape erect, slender. Scape-bracts narrowly triangular, attenuate, much shorter than the internodes. In- florescence laxly bipinnate or subtripinnate. Primary bracts like the seape-bracts, shorter than the sterile bases of the branches. Branches curved-ascending, very slender, sublaxly many-flowered. Floral bracts ovate, acute, shorter than the pedicels. Pedicels slender, 3 mm. long. Flowers nutant, secund. Sepals ovate, acute, 2 mm. long. Petals lanceolate, 5-6 mm. long, equaling the stamens. LA PAZ: SUR YUNGAS: 224 Rhodora [Vol. 71 Guanai, 500 m., R. S. Williams 738 (NY; type); Santa Ana, Bopi, 500 m., Whate 1086 (NY). 3. Deuterocohnia Mez Spreading shrubs. Leaves many, densely rosulate, the blades nar- rowly triangular, usually spinose-serrate. Scape woody with a distinct cambium layer. Inflorescence perennial, putting forth new branches from buds on the axis every year, 2-4-pinnate. Branches polystichous- flowered, surrounded by buds at the base. Flowers sessile or very short-pedicellate. Sepals free. Petals free, actinomorphic, erect, much exceeding the sepals, bearing a single large incised scale at the base. Stamens free, the anthers linear, dorsifixed, the pollen suleate. Ovary superior, glabrous, the placentae linear. Ovules numerous, bearing a narrow dorsal wing that is more or less produced at the apex. Capsule septicidal to the base. Seeds winged. Branches of the inflorescence laxly flowered, the rhachis almost wholly exposed. ....cccccsssccsscsescescssssssscssesssscsccseseeseee 1. D. longipetala. Branches of the inflorescence strobilate, the rhachis completely covered. D ———— s sss 2. D. strobilifera. 1. Deuterocohnia longipetala (Baker) Mez in Mart. Fl. Bras. 3(3): 506 (1894). Flowering 8-10 dm. high. Leaf-blades 3-4 dm. long, attenuate to a slender caudate entire apex, covered with white transparent scales and beneath with cinereous opaque ones, densely serrate with spines 3-4 mm. long. Scape 5-6 mm. in diameter, glabrous. Scape-bracts linear-triangular, serrate, the upper shorter than the internodes. Inflorescence laxly paniculate, many-flowered, the axes subferruginous-tomentellous or glabrous. Primary bracts minute, about equaling the buds. Branches spreading, slender, simple or compound, 10-20 cm. long. Floral bracts very broadly ovate, apicu- late, 4 mm. long. Flowers erect, sessile, 17-25 mm. long. Sepals strongly asymmetric, ovate, obliquely truncate, 8-10 mm. long, few- nerved. Petals yellow with a green spot at apex, equaling or exceeding the stamens. Anthers 4-6 mm. long, green. SANTA CRUZ: CERCADO [?]: Sierra de Santa Cruz, 2000 m., Kuntze s. n. (NY); Comarapa to Vallegrande, 2100 m., Cárdenas 4007 (US); CORDILLERA: Tareira, Lagunillas, 1000 m., Cárdenas 2850 (GH). Peru, Argentina. 2. D. strobilifera Mez in Fedde, Repert. Spec. Nov. 3: 15 (1906). Forming loose polsters, the flowering shoot 15-20 cm. high. Leaf- blades 10-20 em. long, 18 mm. wide, covered with a whitish or silvery coat of scales, laxly serrate in the typical variety with curved sub- spreading spines 5 mm. long. Scape 6-15 cm. long. Scape-bracts foliaceous, exceeding the internodes. Inflorescence laxly bipinnate from a few branches. Spikes sessile, ellipsoid, obtuse, strobilate, 3-5 cm. long, 18 mm. in diameter. Floral bracts ovate, acute, 7 mm. long, about equaling the sepals. Flowers sessile or subsessile, very 1969] Bromeliaceae — Smith 225 variable in size, glabrous. Sepals broadly elliptic, somewhat asym- metric, 6-9 mm. long, mucronulate. Petals elliptic-oblong, 15 mm. long, yellow, the basal scale more or less bifurcate. Stamens equaling or exceeding the petals. CHUQUISACA: SUR CINTI: Camataqui, 2500 m., Fiebrig 2933 (B; type). Argentina. 2a. D. strobilifera var. inermis L. B. Smith in Contrib. U. S. Nat. Herb. 29: 535 (1954). Leaf-blades entire. PoTOSÍ: SUR CHICHAS: Oploca to Atocha, 3000 m., Cárdenas 3741 (GH, US). CHUQUISACA: CINTI: Sivingamayu, 3400 m., Cárdenas 4094 (US; type). TARIJA: CERCADO: Villazón to Tarija, 3400 m., West 8379 (GH). Argentina. 4. Abromeitiella Mez Abromeitiella brevifolia (Griseb.) Castellanos in Anal. Mus. Nac. Hist. Nat. Buenos Aires, 36: 371, tt. 2, 3, 6 (1931). Navia brevifolia Griseb. Symb. Argent. in Goett. Abh. 24: 332 (1879). "Terrestrial polster-forming herbs with much branched dichotomous stems. Leaves densely rosulate at the ends of the branches, 22 mm. long, 7 mm. wide, 2.5 mm. thick, the blades narrowly triangular, fleshy, serrate, lepidote. Scape none. Inflorescence 1-3-flowered in the center of the rosette. Flowers sessile, 31 mm. long, green. Sepals free, more or less asymmetric, mucronulate, 10-18 mm. long, 6 mm. wide. Petals free, distinctly unguiculate, 30 mm. long, 7 mm. wide, bearing a single serrate 5 mm. scale at base. Stamens of the second series short-adnate to the petals, the pollen longitudinally sulcate. Style fili- form. Ovary pyramidal, 5 mm. long, 3 mm. in diameter, the placentae extending the whole height of the locule. Capsule globose, 9 mm. in diameter, septicidal. Seeds fusiform, caudate at both ends. TARIJA: ARCE [?]: Valle del Tambo, Lorentz & Hieronymus 947 (GOET; type). Argentina. 5. Pitcairnia L'Hérit., nomen conservandum Plants mostly terrestrial, rarely epiphytic, mostly stemless, but a few species definitely caulescent. Leaves fasciculate or imbricate and many-ranked along a stem, entire or spinose-serrate, the sheaths usually small, often bulbous-thickened, the blades linear to broadly elliptic with a definite petiole or narrowly triangular, sometimes dimorphic with some blades reduced to horny spinose-serrate spines, deciduous in some species. Scape usually developed. Inflorescence simple cr compound. Flowers perfect, pedicellate to sessile. Sepals free, attenuate to obtuse. Petals free, slightly zygomorphic in most species, naked or appendaged on the inside near the base. Stamens from slightly exserted to shorter than the petals, anthers linear. Style filiform. Ovules many, caudate in all the Bolivian species. Fruit mostly a septicidal capsule. Seeds caudate at both ends. 226 Rhodora [Vol. 71 a. Inflorescence branched (uncertain in P. platystemon). b. Petals naked; sepals 20-22 mm. long. c. Petals long-attenuate; pedicels 23 mm. long. ................... aa... ————————— 1. P. platystemon. c. Petals acute or obtuse, not attenuate; pedicels 9-15 mm. long. d. Leaf-blades narrowed at base but not petiolate. .................... "—— S 2. P. multiramosa. d. Leaf-blades distinctly petiolate. e. Leaf-blades white-furfuraceous beneath; spines of petiole 3 mm. long. er 3. P. odontopoda. e. Leaf-blades not conspicuously white beneath; spines of petiole 1.5 mm. long. .......... eene 4. P. divaricata. b. Petals appendaged; sepals 11-18 mm. long. f. Sepals attenuate; branches elongate. ............ 5. P. paniculata. f. Sepals obtuse; branches abbreviated. ................ 6. P. inermis. a. Inflorescence simple. g. Floral bracts inconspicuous; inflorescence lax. h. Petals naked; sepals 8-22 mm. long. i. Pedicels 23 mm. long, exceeding the floral bracts; petals long-attenuate. ......... essere 1. P. platystemon. i. Pedicels 6-15 mm. long. j. Sepals 8 mm. long; leaf-blades persistent, 30-40 mm. wide. IAS a 7. P. Trianae. j. Sepals 18 mm. long; leaf-blades deciduous, 2 mm. wide. — —— UU eenen 8. P. Cardenas. h. Petals appendaged; sepals 20-42 mm. long. k. Sepals 42 mm. long, attenuate; flowers spreading to reflexed. IS 9. P. erassa. k. Sepals 20-30 mm, long, obtuse; flowers erect to subspreading. MM 10. P. lanuginosa, g. Floral bracts conspicuous, broad and more or less imbricate; inflorescence dense, ............eeceeeeeeeeeens 11. P. Brittoniana. l. Pitcairnia platystemon Mez in DC. Monog. Phan. 9: 421 (1896). Leaves and scape unknown. Inflorescence possibly compound but known only from fragments 31-38 cm. long, glabrous. Axis more or less glaucous, sulcate, especially toward apex. Floral bracts broadly ovate, acuminate, recurved-spreading, 15-20 mm. long. Flowers erect. Pedicels very slender, to 23 mm. long. Sepals attenuate, uncinate- recurved, 22 mm. long, 4.5 mm. wide, ecarinate. Petals attenuate, 34 mm. long, yellow or green when dry, naked. Stamens included. Ovary over 1/2 superior. Ovules caudate. SANTA CRUZ: CHIQUITOS: Santiago, d’Orbigny 937 (P; type). 2. P. multiramosa Mez in DC. Monog. Phan. 9: 419 (1896). Flower- ing 1 m. or higher. Leaves dimorphic, some reduced to narrowly ovate acute sheaths, covered with a thin quickly deciduous membrane of 1969] Bromeliaceae — Smith 221 white scales, others foliaceous, often over 1 m. long, the blades narrowly lanceolate, narrowed toward base but not petiolate, 35 mm. wide, submembranaceous when dry, minutely furfuraceous or glabrate beneath, serrate toward apex with green antrorse spines 1 mm. long. Scape erect, very stout, glabrous. Scape-bracts erect, lanceolate, attenuate, the upper shorter than the internodes. Inflorescence amply paniculate, 5 dm. long, many-flowered, subglabrous. Primary bracts narrowly triangular, shorter than the sterile basal 1⁄3 of the branches. Branches strict, to 25 em. long. Floral bracts ovate-triangular, 5 mm. long. Flowers suberect. Pedicels very slender, 15 mm. long. Sepals attenuate, 22 mm. long, 4 mm. wide, slightly carinate at base, the margin pale and subscarious. Petals acute, 45 mm. long, red, naked. Stamens included. Ovary 3/4 superior. Ovules caudate, CHUQUISACA: AZERO: Weddell 3637 (P; type); Rio Azero to Monteagudo, 1900 m., Cardenas 5732 (US). 3. P. odontopoda Baker, Handbk. Bromel. 93 (1889), Flowering about 1 m. high. Leaves all persistent, dimorphic, the outer with very broadly elliptic dark sheaths covered with a membrane of brown scales and blades reduced to spinose-serrate spines, the inner foli- aceous with elliptic sheaths, the petioles 30 cm. long, deeply chan- neled, serrate with dark antrorse spines 3 mm. long, the blades linear-lanceolate, attenuate, 22-26 mm. wide, entire, reddish beneath, glabrous at maturity. Scape about equaling the leaves, slender. Scape-bracts exceeding the internodes, narrowly triangular, filiform- attenuate, entire. Inflorescence paniculate, pyramidal, the axes white- tomentulose. Primary bracts narrowly triangular, many times shorter than the branches. Branches subspreading, straight, lax, the lateral 6-10 em. long. Floral bracts ovate-elliptic, acuminate, 5 mm, long. Pedicels slender, 9-12 mm. long. Sepals subtriangular, acute, 20 mm. long, white-puberulent. Petals acute, 4 cm. long, red, naked. Ovary over 3/4 superior. Ovules caudate. LA PAZ: NOR YUNGAS: Coroico, 1500-2000 m., Pearce s. n. (K; type); SUR YUNGAS: Chirca, 2500 m., Eyerdam 25357 (F, UC). 4. P. divaricata Wittm. in Meded. Rijks Herbar. 29: 81 (1916). Flowering over 8 dm. high. Leaves dimorphic or trimorphic, the median over 1 m. long with linear blades 2-3 cm. wide, the sheaths bulbous-imbricate, ovate, covered outside with a membrane of gray- fuscous scales, the petiole distinct, 20-25 cm. long, spinose-serrate, the blades 8-9 dm. long, broadly white-striate along the median nerve when dry, the innermost leaves reduced to broad sheaths with a spinose apex. Scape erect, stout, glabrous. Scape-bracts erect, the highest narrowly lanceolate, filiform-attenuate, shorter than the inter- nodes, entire. Inflorescence amply paniculate, to 3 dm. long. Primary bracts narrowly triangular, much shorter than the sterile bases of the branches. Branches divaricate, to 13 cm. long, 4-9-flowered, slightly 228 Rhodora [Vol. 71 glaucous-pruinose. Floral bracts ovate-elliptic, acuminate, about half as long as the pedicels. Pedicels slender, 10 mm. long. Sepals lanceo- late, acuminate and recurved at apex, 20 mm. long, nearly or quite ecarinate, glabrous. Petals linear, subobtuse, to 5 cm. long, red, naked. Stamens included. Ovary about % inferior. Ovules caudate. SANTA CRUZ: CORDILLERA: Quebrada de Charagua, 900 m., Herzog 1212 (L; type; B). 5. P. paniculata (R. & P.) R. & P. Fl. Peruv. 3: 36, t. 260 (1802). P. biattenuata. Rusby in Bull N. Y. Bot. Gard. 4: 457 (1907). Flowering over 2 m. high. Stem erect, stout, ringed with leaf-scars. Leaves densely fasciculate, dimorphic or trimorphic, the outer reduced, scale-like or some with setiform spinose blades, the inner elongate, contracted above the sheath but not always petiolate, attenuate, to 1.5 m. long, 35 mm. broad, remotely spinulose-serrate, glabrous above, sparsely furfuraceous beneath. Scape white-furfuraceous. Scape- bracts narrowly triangular, filiform-caudate, the uppermost about equaling the internodes. Inflorescence laxly paniculate, pyramidal, 5 dm. or longer, the axis furfuraceous. Primary bracts narrowly triangular, mostly shorter than the sterile bases of the branches. Branches suberect to spreading, secundly many-flowered. Floral bracts lance-ovate, acute, much shorter than the 13 mm. long pedicels. Sepals narrowly triangular, attenuate, 15-18 mm. long. Petals 45 mm. long, red, appendaged. Stamens included. Ovary almost wholly superior. Ovules caudate. LA Paz: MURILLO: La Paz to Hacienda Chaco, 1350 m., M. B. Foster 2576 (US); vuNGAS: Puente Villa, 1600 m., Cárdenas 3582 (GH); LARECAJA: Guanai to Tipuani, Bang 1352 (BM, GH, US); Tipuani, 1400 m., Buchtien 5349 (GH); NOR YUNGAS: Coripati, Bang 2155 (BM, G, GH, MICH, US). Peru. 6. P. inermis (Meyer in Presl) Meyer ex Schult, f. in R. & S. Syst. Veg. 7 (2): 1238 (1830). Stemless, to 5 dm. high. Leaves all alike, deciduous above the serrulate base, not at all contracted above the broadly ovate dark sheath; blade entire, linear, attenuate, 4-9 dm. long, 7-12 mm. wide, white-lepidote beneath. Scape stout, soon glabrous. Scape-bracts much exceeding the internodes, lance-ovate, entire, filiform-caudate. Inflorescence thyrsoid, its axis stout, soon glabrous. Primary bracts almost or quite equaling the short branches. Floral bracts scale-like, shorter than the slender 7 mm. long pedicels. Sepals obtuse, 11-15 mm. long, ecarinate. Petals 30-35 mm, long, scarlet, rapidly fading to white, appendaged. Stamens included. Ovary more than 3/4 superior, DEPARTMENT AND PROVINCE UN- KNOWN: Coromillo, Pearce 521 (fide Mez, Pflanzenreich, iv. Fam. 32: 248, 244). 7. P. Trianae André, Enum. Bromél. 4 (Dec. 18, 1888); in Rev. Hortic. 60: 564 (Dec. 16, 1888). Flowering to 2 m. high. Leaves 1969] Bromeliaceae — Smith 229 numerous, persistent, 12 dm. long, the blades linear, attenuate at both ends but not petiolate, 3-4 cm. wide, soon glabrous, spinose toward base. Scape very stout, at first white-arachnoid. Scape-bracts strict, very narrowly triangular, the lower spinose-serrate, the upper entire, shorter than the internodes. Inflorescence simple, many-flowered, very dense except near base, 3-5 dm. long, the axis straight, stout, arachnoid. Floral bracts triangular-ovate, mostly shorter than the pedicels. Flowers spreading to reflexed. Pedicels very slender, 15 mm. long. Sepals oblong, obtuse, 8 mm. long, ecarinate. Petals obtuse, to 45 mm. long, equaling the stamens, white or pale yellow, naked. Ovary at least 3/4 superior. Ovules short-caudate. LA PAZ: MURILLO [?]: Eastern slope of Eastern Cordillera, 1650 m. M. B. Foster 2574 (US). Ecuador, Colombia. 8. P. cardenasii L. B. Smith in Phytologia, 8: 505, t. 3, fig. 5-7 (1963). P. rigida of Catalogue, non Mez (1904). Reproducing vegeta- tively by scaly rhizomes, flowering 3-6 dm. high. Leaves forming bulbous rosettes that soon appear lateral by the elongation of the rhizome, entire, vestite with linear white subspreading scales, the sheaths broadly ovate, 2 cm. long, strongly nerved, castaneous toward base, the blades dimorphic, some persistent, reduced, broadly tri- angular, thin, stramineous, others deciduous along a straight trans- verse line, linear, attenuate, 10 cm. long (young), 2 mm. wide, fascicu- late. Scape very slender, white-flocculose especially toward base. Scape-bracts erect, the lower imbricate, subfoliaceous and with decid- uous blades, the upper narrowly triangular, shorter than the inter- nodes. Inflorescence secundly racemose, red, sparsely white-flocculose. Floral bracts narrowly triangular, exceeding the slender 6-8 mm. long pedicels. Flowers suberect. Sepals oblong, acute, 18 mm. long, carinate at base. Petals rounded-apiculate, 45 mm. long, naked. Stamens included. Ovary 4/5 superior. Seeds caudate. SANTA CRUz: VALLE GRANDE: Cuesta de Monos, 1000 m., Cárdenas 4648 (US; type). 9. P. crassa L. B. Smith in Lilloa, 14: 94 (1948). Leaves and scape unknown. Inflorescence simple, laxly many-flowered, glabrous with age, the axis 1 cm. in diameter, lightly sulcate. Floral bracts re- flexed, broadly lanceolate, acute, 35 mm. long, exceeding the pedicels. Flowers spreading or reflexed. Pedicels 25 mm. long, 2 mm. in diam- eter, angled. Sepals narrowly triangular-lanceolate, 42 mm. long, carinate, nerved. Petals appendaged. Ovary 2/3 superior. Seeds cau- date. LA PAZ: Yungas: Puente Villa, 1400 m., Cárdenas 3690 (GH; type). 10. P. lanuginosa R. & P. Fl. Peru. 3: 35, t. 258 (1802). P. sub- petiolata Baker in Journ. Bot. 19: 267 (1881). P. sessiliflora Rusby in Bull. N. Y. Bot. Gard. 4: 457 (1907). Stemless, flowering 6-12 dm. high. Leaves persistent, dimorphic, the outer with broadly ovate 230 Rhodora [Vol. 71 dark castaneous sheaths and reduced subulate spinose-serrate blades, the inner to over 1 m. long with the blades ensiform, fiilform-at- tenuate, slightly constricted toward base, 16-22 mm. wide, glabrous above, densely white-lepidote beneath, serrate at base with spines 2 mm. long. Scape furfuraceous or glabrescent. Scape-bracts nar- rowly triangular, tomentose-lepidote, the upper much shorter than the internodes. Inflorescence laxly racemose, to 4 dm. long, sparsely furfuraceous or glabrous. Floral bracts ovate, attenuate or apiculate, from shorter than the pedicels to slightly exceeding them. Flowers erect to subspreading, fragrant like pineapple. Pedicels 15-25 mm. long. Sepals narrowly subtriangular, obtuse, 20-30 mm. long. Petals to 9 cm. long, red or green with purple lines, bearing a large oblong scale, Stamens included. Ovary 5/6 superior. Seeds long-caudate. BOLIVIA: Bang s. n. (NY; type of P. sessiliflora Rusby); LA PAZ: NOR YUNGAS: Polo-Polo, Coroico, 1100 m., Buchtien 3675 (NY, US). 11. P. Brittoniana Mez in DC. Monog. Phan. 9: 451 (1896). Ter- restrial or epiphytic, caulescent, flowering 1-1.5 m. high. Leaves over 1 m. long, the sheaths inconspicuous, the petioles 15-40 cm. long, coarsely spinose-serrate, the blades broadly oblanceolate, acute, 95 mm. wide, entire, soon glabrcus. Inflorescence simple, 3-6 dm. long, dense, secund-flowered, the axis sparsely and finely brown-lepidote. Floral bracts ovate, acute, much shorter than the sepals, thin and soon disintegrating. Pedicels to 8 mm. long. Sepals lanceolate, attenuate with an involute uncinate apex, more or less asymmetric, 30-85 mm. long, ecarinate, thick, dark red, sparsely tomentose-lepidote toward apex. Petals broadly obtuse, 4 cm. long, orange, bearing 2 narrow obliquely attached scales near the base. Stamens included. Ovary more than % superior. Ovules long-caudate. LA PAZ: YUNGAS: Bang 592 (US; type; NY); LARECAJA: Tipuani, 1400 m., Buchtien 5348 (GH, US). 6. Dyckia Schult. f. Small to large herbs, Rhizome stout, often repent. Leaves densely rosulate, the sheaths ample, rather fleshy, the blades long and narrow, not contracted at base, spinose-serrate, rigid. Scape evident, erect, usually lateral. Inflorescence amply paniculate to simple, its bracts inconspicuous. Flowers relatively small, yellow to red. Sepals free, shorter than the petals. Petals with their bases fused into a tube with the bases of the stamens. Stamens included or exserted, the filaments free or connate above the common tube. Ovary wholly su- perior. Style distinct to none, Ovules many, alate. a. Filaments free above the common tube formed by the petals and stamens together. b. Stamens exserted; inflorescence glabrous. c. Styles distinct; inflorescence simple. ........ 1. D. leptostachya. c. Styles none; inflorescence compound, ............ 2. D. pulquinensis. 1969] Bromeliaceae — Smith 231 b. Stamens included; inflorescence cinereous-tomentose, .................... Sa TPE TTR RE rns gS bod etc SEO AT 3. D. ferox. a. Filaments connate above the common tube formed by the petals and stamens together; inflorescence simple, minutely furfuraceous; SEVERA GRIN Tee eS n n s hee eed 4. D. gracilis. 1. Dyckia leptostachya Baker in Gard. Chron. 1884(2): 198 (1884). D. boliviensis Mez in DC. Monog. Phan. 9: 524 (1896). Leaves un- known. Scape slender, glabrous. Scape-bracts remote, small, very broadly ovate. Inflorescence simple, 15 cm. long, subdense, glabrous. Floral bracts reflexed, suborbicular, 4 mm. long, much exceeded by the sepals. Flowers subsessile, erect, 17 mm. long, red. Sepals broadly elliptic, emarginate, 9 mm. long. Petal-blades erect, very broadly obovate, subtruncate, obscurely if at all carinate. Stamens barely exserted, free above the common tube with the petals. Style half as long as the ovary. SANTA CRUZ: CHIQUITOS: Santiago, d’Orbigny 858 (P; type of D. boliviensis Mez). 2. D. pulquinensis Wittm. in Meded. Rijks Herbar. 29: 88 (1916). Leaves 14 cm. long, the blades 6 mm. wide, very laxly serrate with stout retrorse spines 5 mm. long. Scape very slender. Scape-bracts remote, triangular-ovate, acute, the upper ones 8 mm. long. In- florescence few-branched. Primary bracts like the upper scape-bracts, much shorter than the sterile bracteate bases of the branches. Branches flexuous, very slender, laxly and subsecundly flowered. Floral bracts broadly ovate, apiculate, 3 mm. long. Flowers subsessile, 15 mm. long, glabrous. Sepals broadly ovate-elliptic, apiculate, 8 mm. long, fuscous when dry. Petals broadly obovate, rounded at apex, 14 mm. long, golden yellow, forming a tube with the stamens 1.5 mm. high. Stamens slightly exserted, free above the common tube. Ovary slenderly ellipsoid, 9 mm. long, the stigmas sessile. SANTA CRUZ: VALLE GRANDE: Pulquina, Herzog 1849 (L; type); Cárdenas 6075 (US); San Isidro, Comarapa, Vogel 477 (US). 3. D. ferox Mez in DC. Monog. Phan. 9: 511 (1896). D. Meziana O. Kuntze of Catalogue. Flowering 6 dm. high. Leaf-blades 8-20 mm. wide, covered beneath with pale closely appressed scales, laxly serrate with acicular antrorse spines 4-6 mm. long. Scape slender, soon glabrous. Scape-bracts remote, very broadly ovate, abruptly acute, entire. Inflorescence simple or few-branched, cinereous-tomentose to glabrous. Primary bracts scarcely larger than the floral bracts. Floral bracts spreading or reflexed, suborbicular-ovate, obtuse or apiculate, much shorter than the sepals. Flowers subsessile. Sepals broadly elliptic, broadly rounded, ca. 5 mm. long, subglabrous. Petals to 12 mm. long, the blades suberect, subrhombic, scarcely undulate, ecarinate or nearly so. Stamens included, the filaments free above the common tube with the petals. Styles 3, very short. SANTA CRUZ: 232 Rhodora [Vol. 71 CORDILLERA: Cerro de Charagua, 1200 m., Cárdenas 2679 (GH). Para- guay, Argentina. 4. D. gracilis Mez in DC. Monog. Phan. 9: 516 (1896). Flowering 5 dm. high. Leaf-blades 3 dm. long, 20-25 mm. wide, covered beneath with closely appressed pale cinereous scales, serrate with spines 4 mm. long. Scape very slender, glabrous. Scape-bracts remote, broadly ovate with a short acuminate apex, entire. Inflorescence simple, lax, minutely furfuraceous. Floral bracts reflexed, suborbicular, apiculate, ca. 4 mm. long. Flowers suberect, subsessile, slender. Sepals broadly elliptic, emarginate, 7-9 mm. long. Petals 14-18 mm. long, equaling or slightly shorter than the stamens, the blade erect, obovate, nearly or quite ecarinate. Stamens connate above the common tube with the petals. Style 2-3 times shorter than the ovary. SANTA CRUZ: CHIQUITOS: San Rafael, d’Orbigny 1006 (P; type); ORIENTE: Roboré, 230 m., Cárdenas 2917 (GH); vELASCO: San Ignacio, J. S. Nelson s. n. (US); Cardenas 5624 (US). Key to the species of Tillandsioideae in fruiting condition (The numbers refer to the species in the respective genera). a. Spikes with the flowers distichous (sometimes becoming secund) or else the inflorescence reduced to a single flower. b. Sepals asymmetric, free, oblong or broadest near the apex, not over 9.5 mm. long ............ Tillandsia subgenus Pseudo-Catopsis. b. Sepals symmetric or, if slightly asymmetric, then ovate or lanceolate and broadest near the base. c. Leaf-blades ligulate or lanceolate with curving sides, mostly flat, thin, and obscurely lepidote ........................... Subkey 1. c. Leaf-blades very narrowly triangular or linear, frequently involute or densely and coarsely lepidote or both. d. Inflorescence compound ................... eem Subkey II. d. Inflorescence simple ................... eee Subkey III. a. Spikes with the flowers in more than 2 ranks ................ Subkey IV. Subkey I a. Inflorescence compound. b. Primary bracts much shorter than the branches. c. Flowers suberect, imbricate .............................. Tillandsia 1-7. c. Flowers spreading, very lax .......... a... Tillandsia 28. b. Primary bracts covering most or all of each lower spike ............ Tillandsia 12, 13. a. Inflorescence simple. d. Flowers imbricate. e. Inflorescences many, axillary .............. rr... Tillandsia 18. e. Inflorescence single, central. f. Floral bracts ecarinate, inflated, the keel curved ............ BEMMNMMMMWMWMMMMMMMMMMMRMMMMMEERRFRREFEEEEBBMMWNEIKIE!NIHENMEMMEIMMM Tillandsia 17. 1969] Bromeliaceae — Smith 283 f. Floral bracts carinate, laterally compressed, the keel sigmoid in outline .................... eere s Vriesea 3. d. Flowers divergent, lax ..................( nn s. Vriesea 4. Subkey II a. Primary bracts much shorter than the branches or spikes. b. Leaves spirally recurved and holding the plant to a branch ........ a NE iu i ces aspas sassa Tillandsia 29. b. Leaves not more than recurved, not spiral. c. Inflorescence amply tripinnate, elongate; sepals 10-13 mm. IT ARRA Tillandsia 8, 9. c. Inflorescence bipinnate or rarely a basal branch divided. d. Leaf-sheaths large, castaneous, strongly contrasting with the blades; floral bracts ecarinate. e. Floral bracts 25-35 mm. long ...................... Vriesea 1. e. Floral bracts 7-13 mm. long. f. Inflorescence subdigitate; spikes 9-16 cm. long ........ D TRA NE RE Tillandsia 10. f. Inflorescence elongate; spikes 4-7 cm. long above the sterile base ....................................... Vriesea 2. d. Leaf-sheaths not contrasting greatly with the blades. g. Floral bracts distinctly shorter than the 10-13 mm. long sepals; plants somewhat caulescent; leaf-blades covered with more or less spreading cinereous scales DIMICARE Som Tillandsia 29, 30. g. Floral bracts mostly equaling to much exceeding the sepals. h. Seales of the leaf-blades very closely appressed. i. Floral bracts 10 mm. long, even, glabrous; leaf- blades thick ....54552: E Tillandsia 25. i. Floral bracts 15-24 mm. long .... Tillandsia 53-55. h. Scales of the leaf-blades more or less spreading. j. Floral bracts 30 mm. long, cinereous-lepidote .... MI s rra oos Tillandsia 11. j. Floral bracts 11-25 mm. long, mostly glabrous sansasunesseeararscsannanoanarnnnsitegesetsbantenedis Tillandsia 22, 24. a. Primary bracts covering most or all of each lower spike. k. Leaf-blades linear-subulate ........................................ Tillandsia 52. k. Leaf-blades narrowly but definitely triangular .. Tillandsia 14-16. Subkey III a. Plants stemless; scales of the leaf-blades very closely appressed ; leaf-sheaths conspicuous. b. Leaf-sheaths dark castaneous, contrasting with the blades ........ Ms kO ce AA SSS ss re CETTE e TETTE TT u u. Vriesea 1. b. Leaf-sheaths little if at all darker than the blades, .................... A GeO dcc no NUR mice Tillandsia 52-54. 234 Rhodora [Vol. 71 a. Plants caulescent; scales of the leaf-blades usually somewhat spreading; leaf-sheaths inconspicuous. c. Leaves in more than 2 ranks. d. Scape wholly covered by its bracts. e. Flowers imbricate. f. Floral bracts equaling or exceeding the sepals; sepals 15-20 mm. long. g. Floral bracts all densely and persistently lepidote AE (Tillandsia 11.) g. Floral bracts glabrous or nearly so. h. Leaves strict, covered with appressed incon- spicuous scales ............................ Tillandsia 23. h. Leaves lax or contorted, covered with more or less spreading: conspicuous scales. i. Floral bracts to 7 em. long, much exceeding the sepals seses Tillandsia 21. i. Floral bracts 10-27 mm. long. j. Floral bracts much exceeding the sepals, wholly covering the rhachis, 27 mm. long MEME Tillandsia 19. j. Floral bracts equaling or barely exceeding the sepals, exposing much of the rhachis, 10-25 mm. long. k. Plant long-caulescent, the leaves mostly shorter than the stem, the blades 4-6 mm. wide ........................ Tillandsia 32. k. Plant short-caulescent, the leaves much longer than the stem, the blades to 20 mm. wide ................ Tillandsia 22. f. Floral bracts distinctly shorter than the sepals; sepals 6.5-17 mm. long. l. Plant moss-like; sepals 6.5 mm. long .................... seeteneehhseeeeeee nennt nt reser eese tient aa ese eee ee nnns Tillandsia 36. l. Plant not moss-like; sepals 10-17 mm. long .......... a Tillandsia 29-32. e. Flowers lax, appressed to the geniculate rhachis ............ —————— HERR Tillandsia 35. d. Scape exposed for most of its length or else wanting; scape- bracts not more than 2 ...... Ja aaa... Tillandsia 37, 38. c. Leaves in exactly 2 ranks. m. Stem wholly concealed by the imbricate leaf-sheaths, short; scape more or less developed. n. Sepals free or equally short-connate; 1-2 scape-bracts often present. o. Leaf-blades narrowly triangular, mucronate, usually well over 2 mm. in diameter. 1969] Bromeliaceae — Smith 235 p. Leaves tomentose-lepidote ................ Tillandsia 33. p. Leaves appressed- or slightly pruinose-lepidote .... PAs de h. s E sss Tillandsia 38-40. o. Leaf-blades linear with a soft blunt apex, not over 2 mm. in diameter. q. Floral bracts glabrous .................. Tillandsia 34. q. Floral bracts densely lepidote ........ Tillandsia 41. n. Sepals much connate posteriorly; scape naked. ................ mot hoe oec Cis ai a a ee PP mere ape cee Tillandsia 42. m. Stem exposed between alternating bunches of leaves, several Meters HN length e u a ms s... Tillandsia 43. Subkey IV a. Inflorescence compound, the spikes lax or sublax and showing the YR eV i Pepin P abe bet foie u n susct rR Aye Alo TE e T Guzmania 1. a. Inflorescence simple, dense with the rhachis completely covered. b. Leaf-blades ligulate. c. Inflorescence much longer than broad, acute .. Guzmania 3, 4. c. Inflorescence about as broad as long. d. Outer bracts of the inflorescence forming an involucre .... DOM TENE E A E o ien Guzmania 2. d. Outer bracts of the inflorescence not forming an involucre. e. Leaf-blades broadly rounded and retuse .. Guzmania 5. e. Leat- blades acute? ... u D m mas a Tillandsia 13. b. Leaf-blades narrowly triangular or linear-subulate. f. Leaf-blades linear-subulate; floral bracts coriaceous or sub- coriaceous, densely lepidote ................................ Tillandsia 52. f. Leaf-blades narrowly but definitely triangular; floral bracts thin. g. Sepals glabrous. h. Sepals ovate, 19 mm. long ........................ Tillandsia 20. h. Sepals lanceolate, 10 mm. long .............. Tillandsia 26. g. Sepals lepidote, thick, coriaceous .............. Tillandsia 2. 7. Tillandsia L. Caulescent or acaulescent herbs of very variable habit. Leaves rosulate or fasciculate or distributed along a stem, polystichous or distichous, entire, ligulate or narrowly triangular or linear. Scape usually distinct. Inflorescence various, usually of distichous-flowered spikes or sometimes reduced to a single polystichous-flowered spike by the reduction of the spikes to single flowers or rarely the whole inflorescence reduced to a single flower. Sepals usually symmetric, free, or equally or posteriorly joined. Petals free, naked. Stamens of various lengths relative to the petals and pistil. Ovary superior, glabrous. Ovules usually many, caudate. Capsule septicidal. Seeds erect, narrowly cylindrie or fusiform, the plumose appendage white, straight, basal. 236 Rhodora [Vol. 71 a. Stamens equaling the petals or shorter. b. Sepals symmetric, or if slightly asymmetric then ovate or lanceolate and broadest below the middle. c. Stamens appearing in the throat of the corolla; style slender, much longer than the ovary. d. Filaments straight; flowers almost always distichous. e. Stamens only a little shorter than the narrow suberect entire petal-blades .................... 1. Subgenus Allardtia. e. Stamens barely exceeding the claws of the petals; petal-blades spreading, broad, crenate-serrate; in- florescence simple ...................... 2. Subgenus Aérobia. d. Filaments more or less transversely plicate; flowers often polystichous in a simple inflorescence. ........... ee S A 3. Subgenus Anoplophytum. c. Stamens deeply included; style short and stout. f. Petal-blades broad, conspicuous; inflorescence compound or simple ....................... Y 4. Subgenus Phytarrhiza. f. Petal-blades narrow, ineonspicuous; inflorescence simple. ——— 'G''— 5. Subgenus Diaphoranthema. b. Sepals asymmetric, nearly or quite free, broadest near the apex, not over 9.5 mm. long ................. 6. Subgenus Pseudo-Catopsis. a. Stamens longer than the petals, exserted .... 7. Subgenus Tillandsia. 1. Subgenus Allardtia a. Flowers distichous; inflorescence compound or simple. b. Inflorescence compound. c. Spikes much exceeding the primary bracts. d. Leaf-blades ligulate, the sides not evenly attenuate. e. Sepals equally free or subfree. f. Floral bracts ample and wholly concealing the rhachis; posterior sepals carinate, 25-45 mm. long. ................. a... —Ó——— RR 1. T. Fendleri. f. Floral bracts narrow and exposing much of the rhachis. g. Sepals 22-40 mm. long; inflorescence amply tripin- nate. RM 2. T. maxima. g. Sepals 12-15 mm. long; inflorescence bipinnate or the lowest branches sometimes divided. ........ 8. T. rubella. e. Sepals connate posteriorly more than elsewhere. h. Inflorescence amply tripinnate; sepals 14-19 mm. long. i. Spikes fascicled, linear, many-flowered, ........................ RR 4. T. boliviana. i. Spikes pinnate along 35 cm. long branches, narrowly lanceolate, 5-7 flowered. ................ 5. T. Krukoffiana. h. Inflorescence bipinnate; sepals 15-26 mm. long. j. Leaves densely lepidote on both sides; sepals 26 mm. long. ...................... sarasa 6. T. Kumtzeama. 1969] Bromeliaceae — Smith 237 j. Leaves glabrous above; sepals 15 mm. long. ................ SE EA Or E E A eT UB CO LLG HO A 7. T. confinis. d. Leaf-blades narrowly triangular with straight evenly at- tenuate sides. k. Inflorescence amply tripinnate, elongate; sepals 10-13 mm. long. l. Spikes dense; floral bracts exceeding the sepals. ............ Be E i, teeta s 8. T. longifolia. l. Spikes very lax; floral bracts shorter than the sepals. .... FeO EEE a eee oon rr 9. T. marantoidea. k. Inflorescence not more than bipinnate; sepals 8-21 mm. long. m. Sepals obtuse, 8 mm. long; inflorescence of several SPIKES ooo sh cu nt SMS sss 10. T. Rusbyi. m. Sepals acute, 21 mm. long; inflorescence of 2 spikes. .... Xue t E E E SI 11. T. boliviensis. c. Spikes slightly (T. calocephala) if at all exceeding the lower primary bracts. n. Leaf-blades ligulate, the sides not evenly attenuate. o. Inflorescence tripinnate; sepals ecarinate. .... 3. T. rubella. o. Inflorescence bipinnate. p. Spikes 1-4-flowered; leaf-sheaths green or purplish. q. Sepals about equaling the floral bracts, even; leaf- blades 4-6 cm. wide. ............................ 12. T. violascens. q. Sepals exceeding the floral bracts, nerved; leaf-blades ISOM SWide.................. t aeuo o DR arenes 13. T. biflora. p. Spikes 6-7-flowered; leaf-sheaths very dark castaneous, nearly black. .................................... 14. T. fusco-guttata. n. Leaf-blades narrowly triangular, the sides evenly attenuate; spikes to 3-flowered; sepals acute. r. Plant steraless; leaves 20-40 cm. long; inflorescence 5-6 Cm: OM po ovine caves em e reete rre iiid 15. T. sphaerocephala. r. Plant caulescent; leaves 6-10 cm. long; inflorescence 25 rra Di 2 PNEU e cec 16. T. calocephala. b. Inflorescence simple. s. Leaf-blades ligulate. t. Scape central, solitary; spike 20 cm. long, 40 mm. wide. .... axcogx e RETENIR: VS, oc Eon 17. T. Walteri. t. Scapes axillary, numerous; spike to 8 cm, long, 15-20 mm. wo RSEN m 18. T. complanata. s. Leaf-blades narrowly-triangular. u. Floral bracts all densely and persistently lepidote. ................ M (11. T. boliviensis probably has a simple inflorescence). u. Floral bracts glabrous or nearly so. ........ 19. T. Cardenasii. a. Flowers polystichous; inflorescence simple. 238 Rhodora [Vol. 71 v. Leaf-blades ligulate; petals blue, lavender, or purple. ................ deasscavsesesssdnensusuntssssisyedsntoensnsesssssssuvennssssatnaseésesssecessssiscaanvee 13. T. biflora. v. Leaf-blades narrowly triangular; petals yellow. .... 20. T. ixioides. 2. Subgenus Aérobia A single species. .........22............... r. 21. T. xiphioides. 3. Subgenus Anoplophytum a. Flowers distichous; inflorescence usually compound. b. Leaves covered with conspicuous spreading scales especially along the margins; floral bracts 20-25 mm. long, nearly or quite glabrous, ecarinate. .............. essere 22. T. Lorentziana. b. Leaves covered with appressed or subappressed scales; floral bracts 10-20 mm. long. c. Plant caulescent; inflorescence simple. ........ 23. T. caulescens. c. Plant stemless; inflorescence usually compound. d. Floral bracts strongly nerved, usually cinereous-lepidote, carinate. ....................... a... 24. T. didisticha. d. Floral bracts nearly or quite even, lepidote at apex, ecari- nate, ————— 25. T. vernicosa. a. Flowers polystichous; inflorescence simple. e. Sepals membranaceous, much longer than broad, glabrous. ........ m— MEME 26. T. tenuifolia. e. Sepals coriaceous, suborbicular, lepidote. ........ 27. T. Pohliana. 4. Subgenus Phytarrhiza a. Leaf-blades ligulate, flat; spikes very lax. ....... 28. T. laxissima. a. Leaf-blades very narrowly triangular, mostly involute; spikes dense. b. Scape-bracts imbricate or nearly so, covering most or all of the scape; leaves polystichous. c. Leaves strongly coiled, stout; inflorescence long, often tripin- hate, isssascuscnsoarececocssseeensvedsesceresessesersderseessssovevsibicsces 29. T. Duratii. c. Leaves merely curved, slender; inflorescence not more than bipinnate, digitate or simple. d. Leaf-blades covered with subappressed scales; plant caules- D. APR 30. T. Reichenbachi. d. Leaf-blades covered with narrow spreading scales. e. Leaves longer than the stem; inflorescence usually com- pound. ........................... ethernet 31. T. streptocarpa. e. Leaves shorter than the elongate stem; inflorescence simple. RR 32. T. paleacea. b. Scape-bracts remote, leaving most of the setiform scape exposed; leaves distichous. f. Floral bracts densely tomentose-lepidote, about equaling the sepals; leaf-blades 2-5 mm. in diameter; petals yellow. ........ MM 33. T. crocata. 1969] Bromeliaceae — Smith 239 f. Floral bracts glabrous or nearly so, mostly shorter than the sepals; leaf-blades 1-2 mm. in diameter; petals blue or violet. Ail | eh S a ua E uy atas can 34. T. bandensis. 5. Subgenus Diaphoranthema a. Scape-bracts numerous, about equaling the internodes; leaves polystichous. b. Spike lax, linear, the flowers less than twice as long as the IDLernodegspes «m eei aaa a E pan shaya ete 35. T. loliacea. b. Spike dense or 1-flowered, lanceolate, the flowers more than twice as long as the internodes, ........................ 36. T. tricholepis. a. Scape-bracts not more than 2, the scape when present largely naked. e. Leaves 4-9 mm. long, polystichous, the plant resembling a coarse OERE oeoa a a eae a n rer 37. T. bryoides. c. Leaves much larger, distichous (sometimes polystichous in T. funebris). d. Stem wholly concealed by the imbricate leaf-sheaths, short; scape more or less developed. e. Sepals free or equally short-connate; 1-2 scape-bracts often present. f. Leaf-blades narrowly triangular, mucronate, usually well over 2 mm. in diameter. g. Floral bracts glabrous. ............................ 38. T. funebris. g. Floral bracts densely lepidote. h. Leaf-blades laterally compressed; leaf-sheaths indis- tinet, but a little broader than the blades, enfolding the stem only by their extreme base. .... 39. T. Gilliesii. h. Leaf-blades terete; leaf-sheaths distinct, much broader than the blades, enfolding the stem for most of their length. .................................... 40. T. myosura. f. Leaf-blades linear with a soft blunt apex, not over 2 mm. in diameter; inflorescence dense or 1-flowered. .................... T ————'ÓÉetpe 41. T. recurvata. e. Sepals much connate posteriorly; scape naked. .................... RN 42. T. capillaris. d. Stem exposed between alternating bunches of leaves, several meters n length, -........ eene eric ua nan: 43. T. usneoides. 6. Subgenus Pseudo-Catopsis a. Inflorescence simple or few-branched; floral bracts suborbicular, exceeding the flowers, membranaceous. ............ 44. T. Seemannii. a. Inflorescence much branched. b. Floral bracts about equaling the sepals or distinctly longer. c. Floral bracts even or faintly few-nerved near the margin, without a midnerve. d. Floral bracts not incurved, almost exactly equaling the sepals; spikes to 9 cm. long. ........................ 45. T. spiculosa. 240 Rhodora [Vol. 71 d. Floral bracts incurved toward apex, definitely exceeding the sepals; spikes 4 cm. long. .................... 46. T. monticola. c. Floral bracts sulcate and with a definite midnerve, exceeding DIRECTE ERR 47. T. pendulispica. b. Floral bracts distinctly shorter than the sepals. e. Inflorescence tripinnate. f. Leaf-blades broadly ligulate; inflorescence sparsely whitish- lepidote. ....ceccsccceccssssecccesecesceeeeeeessseccesevseseesessteeeeses 48. T. pardina. f. Leaf-blades narrowly triangular; inflorescence ferruginous- lepidote. ................................. e eene ee nnnnnne nnns 49. T. Bakeri. e. Inflorescence bipinnate; leaf-blades narrowly triangular. g. Scape-bracts remote; primary bracts many times shorter than the branches. ................... aa... 50. T. parviflora. g. Scape-bracts mostly equaling or exceeding the internodes; primary bracts at least half as long as the branches. ........ eene eee enenene ih nethne tentent S 5l. T. adpressa. 7. Subgenus Tillandsia a. Leaf-sheaths nearly flat, not forming a pseudobulb. b. Leaf-blades linear-subulate; inflorescence digitate or simple and with polystichous flowers. ............. cesses 52. T. juncea. b. Leaf-blades narrowly but definitely triangular. c. Floral bracts coriaceous; about equaling the sepals. ............... "N en erann i eene eenaa eneee enaena 53. T. polystachia. c. Floral bracts submembranaceous, much exceeding the sepals. —— ERREUR 54. T. Valenzuelana. a. Leaf-sheaths inflated and forming a pseudobulb; scape-bracts vaginiform; sepals ecarinate. ................................ 55. T. paraénsis. l. Tillandsia Fendleri Griseb. in Goett. Nachr. 1864: 17. 1865. T. Bangii Baker in Mem. Torr. Bot. Club 6: 124 (1896). T. Deppiana var. latifolia (Griseb.) L. B. Smith of Catalogue. Stemless, to 2 m. high. Leaves 6-10 dm. long, the blades ligulate, triangular-attenuate, to 8 cm. wide, flat, nearly or quite glabrous. Scape erect, stout. Scape- bracts densely imbricate, foliaceous. Inflorescence mostly bipinnate. Primary bracts like the upper scape-bracts, much shorter than the branches but much larger than the floral bracts. Spikes acute, com- planate, densely 6-many-flowered, mostly spreading, 10-30 cm. long, 5 em. wide, mostly short-stipitate. Floral bracts obovate, carinate toward apex, ample and concealing the rhachis, coriaceous, rigid, glabrous, even, about equaling the sepals. Flowers short-stipitate. Sepals oblong-lanceolate, 25-45 mm. long, carinate, equally subfree. Petals slightly exceeding the stamens, blue. LA PAZ: SUR YUNGAS: Asunta, near Evenay, 690-750 m., Krukoff 10568 (GH, NY, US). COCHABAMBA: CERCADO: Cochabamba, Bang, 1283 (G, GH, US; type- number of T. Bangii Baker); CHAPARE: Antahuacana, Espíritu Santo, 750 m., Buchtien s. n., 2224 (US). West Indies to Venezuela and Bolivia. 1969] Bromeliaceae — Smith 241 2. T. maxima Lillo & Hauman in Anal. Mus. Nac. Buenos Aires, 29: 416 (1917). T. australis Mez of Catalogue. T. maxima var. densior L. B. Smith in Lilloa, 14: 97 (1948). Flowering 8-24 dm. high with the inflorescence extended. Leaves many in a spreading rosette, the blades ligulate, subrounded, apiculate, 6-12 cm. wide, subglabrous. Scape very stout, erect and then decurved, Scape-bracts densely im- bricate, foliaceous. Inflorescence amply tripinnate, very variable, gla- brous. Primary bracts like the scape-bracts, ligulate, mostly much shorter than the branches. Branches bearing 1-7 spikes, the rather large sterile base bracteate. Secondary bracts much reduced, ovate, acute. Spikes lanceolate to linear, 11-30 cm. long, sublax to subdense. Floral bracts suberect, narrow and exposing most of the rhachis, broadly elliptie, equaling or slightly exceeding the sepals, even (! Cas- tellanos) and apparently fleshy, becoming coarsely rugose when dry, ecarinate, not incurved. Sepals free, subobtuse, 22-40 mm. long, ecarinate toward apex. Petals blue or violet. Stamens included. LA PAZ: LARECAJA: Sorata, Cerro de S. Iminapi, 2650 m., Mandon 1187 e. p. (BM). CocHABAMBA: AYOPAYA: Sailapata, 2500 m., Cárdenas 3387 (GH); CAMPERO: Río Caine, 2800 m., Cárdenas 2113 (GH, US; type number of T. maxima var. densior L. B. Smith); MIZQUE [?]: Sevingania to Vilavila, 3900 m., M. B. Foster 2554 (US); TOTORA [?]: Km. 450, Cochabamba to Santa Cruz, 1200 m., Cárdenas 5207 (US). TARIJA: ARCE: La Merced, Bermejo, 1700 m., Fiebrig 2161 (GH, US). Argentina. 3. T. rubella Baker in Journ. Bot. 26: 44 (1888). Stemless, propa- gating by stolons 9-15 dm. long, flowering 5-13 dm. high. Leaves 3-6 dm. long, the blades ligulate, acute with a recurved apex, 3-5 cm. wide, subglabrous, sometimes with fine purple spots. Scape 9-14 mm. in diameter, glabrous. Scape-bracts densely imbricate, lance-elliptic, acute. Inflorescence 2-3-pinnate, ellipsoid to subthyrsoid, dense toward apex, interrupted below, glabrous or nearly so, red. Primary bracts spreading, ovate, the lowest attenuate and exceeding the branches, the upper acute, shorter than the branches. Branches short with spikes solitary or fasciculate. Secondary bracts scarcely larger than the floral bracts. Spikes lanceolate, acute, 4-6 cm. long, 16-20 mm. wide, slightly complanate, subdensely 8-12-flowered. Floral bracts imbricate but partially exposing the rhachis, broadly ovate, subobtuse, 14-20 mm. long, slightly exceeding the sepals, submembranaceous, strongly nerved. Sepals ovate, subacute, 12-15 mm. long, equally subfree, ecarinate, Petals erect, 23 mm. long, the blades narrowly elliptic. Stamens included. LA PAZ: LARECAJA: Sorata, 2650 m., Mandon 1187 (GH; type-number); COCHABAMBA: CERCADO: Tunari, 2400 m., Foster 2533 (US). According to M. B. Foster, the young shoots of this species and of T. maxima are edible and are sold in the markets under the name “Horka”. 242 Rhodora [Vol. 71 4. T. boliviana Mez in Bull. Herb. Boiss. (ser. 2) 4: 1130 (1904). Plant large, known only from fragments. Leaves in a funnelform rosette, 4 dm. long, the blades ligulate, subacute and apiculate, 35 mm. wide, densely appressed-lepidote on both sides. Seape unknown. Inflorescence amply tripinnate, glabrous. Branches with 6 spikes, at base 2 sterile bracts, the axis compressed. Secondary bracts incon- spicuous. Spikes subsessile, nutant, linear, acute, 17 cm. long, 12 mm. wide, strongly complanate, to 32-flowered, Rhachis geniculate, angled. Floral bracts ovate-elliptic, obtuse, 17 mm. long, slightly exceeding the sepals, about 3 times as long as the internodes but not covering the rhachis, coriaceous, strongly nerved, obscurely carinate toward apex. Flowers subsessile, 22 mm. long. Sepals elliptic, obtuse, 14 mm. long, chartaceous, prominently nerved, connate for 6 mm. posteriorly and 2 mm. anteriorly. Petals violet, the blades suberect, elliptic. Stamens included. WITHOUT DEPARTMENT: Bang 2202 (GH, type- number). 5. T. Krukoffiana L. B. Smith in Contrib. Gray Herb. 154: 36 (1945). Flowering at least 2 m. high. Leaves to 8 dm. long, the blades ligu- late, acute or attenuate, 6 cm. wide, glabrous. Scape unknown. In- florescence ample, very laxly 3-pinnate, the axes slender, glabrous. Primary bracts ovate, acute, 4 em, long, shorter than the long sterile bases of the branches. Branches spreading, to 35 em. long, bearing 1-5 sterile bracts at base. Secondary bracts like the primary but much smaller, half as long as the spikes. Spikes spreading or deflexed, narrowly lanceolate, acute, 45-65 mm. long, 10-15, mm. wide, com- planate, subdensely 5-7-flowered. Floral bracts elliptie, acute, to 25 mm. long, carinate toward apex, subchartaceous, nerved, sparsely and obscurely lepidote. Flowers subsessile. Sepals elliptic, 19 mm. long, nerved, glabrous, the posterior carinate, connate for 4 mm. LA Paz: SUR YUNGAS: Rio Bopi Basin, San Bartolome, Calisaya, 750-900 m., Krukoff 10503 (GH; type). 6. T. Kuntzeana Mez in DC. Monog. Phan. 9: 790 (1896), Stemless, flowering to over 1 m. high. Leaves many, the blades attenuate, 30 mm. wide, covered on both sides with pale appressed scales. Scape erect, stout, glabrous. Scape-bracts imbricate, elliptic, acute or at- tenuate, coriaceous, lepidcte. Inflorescence pinnate from 4-5 spikes, 2 dm. long. Primary bracts like the upper scape-bracts, much shorter than the spikes. Spikes erect, sessile, 10 em. long, slightly complanate, densely 8-flowered. Floral bracts elliptic, acute, mucronate, 32 mm. long, 17 mm. wide, exceeding the sepals, sharply carinate, incurved at apex, coriaceous, even, glabrous. Flowers subsessile. Sepals linear, subacute, 26 mm. long, submembranaceous, nerved, glabrous, the pos- terior 1/3 connate. COCHABAMBA: CERCADO: Tunari, Kuntze (NY; type). 1969] Bromeliaceae — Smith 245 7. T. confinis L. B. Smith, Phytologia, 4: 218, t. 2, figs. 5-7 (1953). T. subtropicalis L. B. Smith in Phytologia, 8: 505 (1963). Stemless, flowering 3 dm. high. Leaves 20-23 cm. long, the blades ligulate with a long attenuate apex, 15-20 mm. wide, covered beneath with fine appressed cinereous scales, glabrous above. Scape erect, slender. Scape-bracts densely imbricate, the lowest subfoliaceous, the others elliptic, glabrous and lustrous except for the small narrow blade. Inflorescence laxly bipinnate from 5-7 spikes, glabrous. Primary bracts ovate, about half as long as the spikes. Spikes subspreading, subsessile, lanceolate, acute, 45 mm. long, 12 mm. wide, strongly com- planate, dense. Floral bracts wholly concealing the rhachis, ovate, acute, 18 mm. long, sharply carinate with a straight apex, coriaceous, even, lustrous. Sepals lanceolate, acute, 15 mm. long, the posterior carinate, 2/3 connate. LA PAZ: LARECAJA: Hacienda Casana, Tipuani, 1400 m., Buchtien 7185 (US; type of T. subtropicalis L. B. Smith) ; same, 7187. Venezuela, Colombia, Ecuador, Peru. 8. T. longifolia Baker, Handbk. Bromel. 185 (1889). Flowering 2.3 m. high. Leaves 8-10 dm. long, the blades very narrowly triangu- lar, 3-4 cm. wide, subdensely appressed-lepidote on both sides. Scape erect, very stout. Scape-bracts imbricate, foliaceous. Inflorescence pyramidal, tripinnate. Primary bracts ovate, the lowest with long foliaceous blades exceeding the branches. Branches suberect, to 2 dm. long, the sterile bases laxly bracteate. Spikes short-stipitate, linear- lanceolate, acute, 4-12 em. long, complanate, densely 4-12-flowered. Floral bracts ovate, obtuse, to 15 mm. long, broadly convex with a thickened midnerve but scarcely carinate, coriaceous, nerved, gla- brous. Sepals obovate, obtuse, 10-12 mm. long, short-connate pos- teriorly. Stamens included. SANTA CRUZ: CERCADO: Yungas de San Mateo, Comarapa, 2800 m., Steinbach 8467 (GH). Colombia, Vene- zuela. 9, T. marantoidea Rusby, Descr. S. Amer, Pls. 4 (1920). Caulescent, flowering over 1 m. high with an erect stem 15-20 cm. long. Leaves to 6 dm. long, the blades linear, attenuate, 10-12 mm. wide, very obscurely lepidote beneath. Scape-bracts erect, 6-15 cm. long, at- tenuate. Inflorescence laxly tripinnate, 45 cm. long, glabrous, Pri- mary bracts ovate, acute, 3 cm. long, much exceeding the short naked sterile bases of the branches, subchartaceous, red. Branches spreading, subdigitately divided, the median to 10 cm. long, the lower shorter and more remote. Secondary bracts like the floral bracts. Spikes 5-8 em. long, very lax with the flowers subspreading and about 1 cm. apart. Rhachis nearly straight, very slender, angled. Floral bracts elliptic, obtuse, 13-15 mm. long, exceeded by the sepals, ecarinate, thin, nerved. Pedicels stout. Sepals lanceolate, obtuse, 13 mm. long, ecarinate, nerved. Petal-blades 7 mm. long. Stamens included, WITH- OUT DEPARTMENT: Bang 2570 (NY; type; US). 244 Rhodora [Vol. 71 10. T. Rusbyi Baker, Handbk. Bromel. 176 (1889). T. Buchtienü H. Winkl. in Fedde, Repert. Spec. Nov. 7: 107 (1909). T. Guentheri Harms in Notizbl. 10: 794 (1929). Leaves about 30 in a dense rosette, exceeding the inflorescence, the sheaths very large, covered with castaneous scales, the blades very narrowly triangular, 10 mm. wide, involute-subulate, covered with cinereous scales. Scape suberect, slender, half as long as the leaves. Scape-bracts densely imbricate, long-laminate. Inflorescence bipinnate, distichous-branched, 11 cm. long and wide, of about 14 spikes. Primary bracts erect, broadly ovate, attenuate, scarcely exceeding the sterile bracteate bases of the branches, lepidote. Spikes suberect to spreading, nearly uniform, linear-lanceolate, acute, 9-16 em. long, 6 mm. wide, complanate, about 20-flowered. Floral bracts laxly imbricate, subacute, 11-12 mm. long, exceeding the sepals, ecarinate, subchartaceous, nerved, lepidote. Flowers subsessile. Sepals narrowly elliptic, obtuse, 8 mm. long, equally subfree, subcoriaceous, glabrous. Petals white. Stamens in- cluded. La PAZ: Yungas, 1800 m., Rusby 2163 (NY; type) ; LARECAJA: San Carlos, Mapiri, 850 m., Buchtien 369 (B; type of T. Guentheri Harms); Hacienda Casana, Tipuani, 1400 m., Buchtien 7180 (NY); MURILLO: La Paz, Buchtien 1268 (US; type-number of T. Buchtienti H. Winkl.); OMASUYOS: Copacabana, Mapiri, 850-950 m., Krukoff 11181 (NY). WITHOUT LOCALITY: Bang (NY). Peru. 11. T. boliviensis Baker in Mem. Torr. Bot. Club, 4 (3): 267 (1895), emend. L. B. Smith in Contrib. Gray Herb. 98: 15 (1932). Flowering 2 dm. high. Leaves fasciculate-rosulate, about equaling the inflores- cence, the blades narrowly triangular, subulate-attenuate, covered with slightly spreading cinereous scales, coarsely fimbriate-lepidote along the margins. Scape almost hidden by the leaves. Scape-bracts imbricate, lanceolate, apiculate or short-caudate, densely cinereous- lepidote. Inflorescence of a terminal spike 8 cm. long and a lateral 4-5 em. long. Primary bract like the scape-bracts, half as long as the spike. Spikes oblong-lanceolate, 4-10-flowered. Floral bracts imbri- cate but exposing the rhachis, lanceolate, acute, 30 mm. long, ecarinate, cinereous-lepidote. Flowers subsessile. Sepals narrowly lanceolate, acute, 21 mm. long, free, glabrous, nerved, the posterior alate-carinate. LA PAZ: MURILLO: La Paz, 3300 m., Bang 159a in part (NY; type). 12. T. violascens Mez in DC. Monog. Phan. 9: 797 (1896). Caules- cent, flowering 4-7 dm. high. Stem decumbent, stout, covered with the remains of old leaves. Leaves 3-5 dm. long, the sheaths distinct, elliptic, to 18 cm. long, dark violet, the blades ligulate, attenuate, 4-6 em. wide, minutely and obscurely punctulate-lepidote. Scape erect, to 14 mm. in diameter, glabrous. Scape-bracts densely imbricate, foliaceous with subinflated sheaths. Inflorescence densely bipinnate, thyrsoid, 15-25 em. long, 10 cm. in diameter, red or green. Primary bracts suberect, the middle and lower subfoliaceous and exceeding 1969] Bromeliaceae — Smith 245 the spikes, the upper suborbicular, apiculate, about equaling the spikes. Spikes sessile or with a short naked stipe, elliptic, subdensely 4-flowered, 4 em. long. Floral bracts broadly ovate, rounded, about equaling the sepals, carinate, even, glabrous. Sepals elliptic, broadly rounded, 8-15 mm. long, chartaceous, even, glabrous. LA PAZ: LARE- CAJA: Sorata, 2700 m., Mandon 1174 e. p. (P; type); Unduavi, 3200 m., Buchtien 2599 (NY). COCHABAMBA: CHAPARE: Llanta Aduana, Colomi to Cochabamba, 3500 m., Cárdenas 2258 (GH); Incachaca, 2700 m., hb. Marnier-Lapostolle s. n. (US); AYOPAYA: Choro, 3200 m., Cárdenas 4382 (US); CERCADO [?]: Yungas, 3300 m., M. B. Foster 2547 (US); 3150 m., 2549 (US) ; CARRASCO: Sehuenka, Cárdenas 6166 (US). SANTA CRUZ: VALLE GRANDE: Comarapa, 2800 m., Steinbach 8561 (US). 13. T. biflora R. & P. Fl. Peruv. 3: 41, t. 268 (1802). Stemless, to 35 em. high. Leaves 2 dm. long, obscurely punctulate-lepidote, often purple-spotted, the blades ligulate, acute, 2 em. wide. Scape slender, usually curved. Scape-bracts densely imbricate, foliaceous. Inflores- cence densely bipinnate or rarely reduced to a single polystichous- flowered spike, ovoid, glabrous or obscurely punctulate-lepidote. Primary bracts broadly ovate, thin, inflated, exceeding the lower and middle spikes, the lower attenuate, the upper apiculate. Spikes laxly 1-3-flowered, short-stipitate. Floral bracts broadly ovate, much shorter than the sepals, carinate, nerved, thin. Sepals oblong, obtuse or broadly acute, 12-15 mm. long, subfree, carinate, nerved. Petals ligulate, 2 cm. long, blue to purple. Stamens included. WITHOUT DE- PARTMENT: Bang 1802 (F, GH, NY, US; distributed as “Guzmania tricolor”); LA PAZ: LARECAJA: Sorata, 2700 m., Mandon 1174 in part (S). Peru to Venezula. 14. T. fusco-guttata Mez in Bull. Herb. Boiss. (ser. 2) 5: 101 (1905). Stemless, flowering 4-6 dm. high. Leaves 2-4 dm. long, the blades ligulate, acute or attenuate, 15-40 mm. wide, obscurely brown- punctulate, spotted. Scape erect or curved. Scape-bracts imbricate, foliaceous. Inflorescence densely bipinnate or interrupted at base, subpyramidal, 6-12 cm. long, 4-5 cm. in diameter. Primary bracts ovate-elliptic, long-laminate with recurving apices, exceeding the middle and lower spikes, punctulate. Spikes ovate-elliptic, acute, 25- 30 mm. long, 12-15 mm. wide, complanate, densely 6-7-flowered. Floral bracts broadly ovate, equaling or slightly shorter than the sepals, carinate, glabrous, slightly nerved, subcoriaceous. Sepals obtuse, 10 mm. long, equally subfree, coriaceous. Petals 16 mm. long, blue. Stamens included. COCHABAMBA: AYOPAYA: Sailapata, 2700 m., Cár- denas 3030 (GH, US). Province [?]: Monte Punco, 3000 m., Cárdenas 5276 (US). CHAPARE: Chapare, 2000 m., Cárdenas 6055 (US). Peru. 15. T. sphaerocephala Baker in Journ. Bot. 26: 141 (1888). Stem- less, flowering to 2 dm. high. Leaves 2-4 dm. long, covered with ap- 246 Rhodora [Vol. 71 Figure (2) A: Tillandsia sphaerocephala (after Bang 159); habit X 1/2; B: spike X 1; C: anterior sepal X 1; D: posterior sepal X 1; E: petal and stamen X 1. 1969] Bromeliaceae — Smith 247 pressed lustrous scales, the blades narrowly triangular, 2 em. wide, involute-subulate toward apex, erect or decurved. Scape usually curved and ascending. Scape-bracts densely imbricate, foliaceous. Inflores- cence bipinnate, densely capitate, 5-6 cm. long. Outer primary bracts with a broad ovate base and long triangular apex that exceeds the inflorescence, the inner ovate, obtuse, apiculate, equaling the spikes. Spikes 1-2-flowered, strongly complanate, Floral bracts ovate, 23-30 mm. long, carinate, sparsely lepidote to glabrous, strongly nerved. Sepals ovate, acute, 20-25 mm. long, carinate, Posteriorly connate for 3 mm. Petals to 30 mm. long. Stamens included. LA PAZ: MURILLO: La Paz, 3000 m. Bang 159 (GH, NY, US); Rusby 2848 (NY); Obrajes, 3550 m., Buchtien 4540 (US); 3300 m., 6384 (US); 3600 m., Cárdenas 2419 (GH); San Jorge, 3550 m., Buchtien 8819 (US); LARECAJA: Sorata, 3100 m., Mandon 1188 (K; type) ; MUNECAS: Chara- zani, 2700 m., Cárdenas 3867 (GH); INQUISIVI: Pongo de Quime, 3000 m., White 182 (NY). COCHABAMBA: CERCADO: Tunari, 1950 m., M. B. Foster 2534 (US). Fig. 2. 16. T. calocephala Wittm. in Meded. Rijks Herbar. 29: 90 (1916). Caulescent, flowering 22 cm. long. Leaves 6-10 em. long, covered with subappressed cinereous scales, the blades narrowly triangular, sub- ulate-attenuate, 10-12 mm. wide, more or less secund-recurving. Scape very short, Scape-bracts foliaceous and only slightly reduced. In- florescence densely bipinnate, ovoid, 25 mm. long, 15-20 mm. in diam- eter. Primary bracts suborbicular, slightly shorter to slightly longer than the spikes, the lowest foliaceous-laminate. Spikes sessile, lanceo- late, 2-3-flowered, complanate, Floral bracts imbricate, lance-ovate, 10 mm. long, slightly shorter than the sepals, carinate, thin, nerved, lepidote. Sepals lanceolate, 12 mm. long, carinate, glabrous, short- connate posteriorly. Petals 15 mm. long, violet. Stamens included. LA Paz: MURILLO: La Paz, Shepard 234 (GH). COCHABAMBA: CHA- PARE[?]: Rio Montehuaiko, Herzog 2300 (L; type). Peru. 17. T. Walteri Mez in Fedde, Repert. Spec. Nov. 3: 43 (1906). Stem- less, flowering 5 dm. high. Leaves cinereous-lepidote, the sheaths ovate, longer than the blades, the blades lanceolate, attenuate, 2 dm. long, 4 cm. wide. Scape erect, stout. Scape-bracts imbricate, the lower ones caudate, Inflorescence simple, dense, distichously 12-16-flowered, lanceolate, 2 dm. long, 4 cm. wide, slightly complanate. Rhachis straight, alate, glabrous. Floral bracts imbricate and concealing the rhachis, broadly elliptic, obtuse, 5 cm. long, exceeding the sepals, in- flated, ecarinate, nerved, glabrous. Pedicels stout, 4 mm. long. Sepals equally subfree, obtuse, 40 mm. long, nerved, glabrous. Petals 60 mm. long, violet. Stamens included. COCHABAMBA: AYOPAYA: Sailapata, 2500 m., Cardenas 3312 (GH). Peru. 18. T. complanata Benth. Bot. Voy. Sulphur, 173 (1846). Leaves many in a dense rosette, 3-4 dm. long usually, mostly exceeding the 248 Rhodora [Vol. 71 inflorescences, spotted, streaked, or suffused with dark purple especial- ly toward base, 'the blades ligulate, abruptly acute or subobtuse, 3-5 em. wide, obscurely punctulate-lepidote. Scapes many and axillary, at length decurved with the leaves, 1-2 mm. in diameter, glabrous. Scape- bracts erect, usually imbricate, narrowly lanceolate, subchartaceous, glabrous or nearly so. Inflorescences simple, lanceolate or linear, acute, dense, to 8 em. long, 15-20 mm. wide, usually complanate, glabrous. Floral bracts elliptic, obtuse, 15-25 mm. long, exceeding the sepals, even or nerved, ecarinate. Flowers subsessile, Sepals lanceolate, acute, 10-15 mm. long, the posterior ones carinate, usually much connate. Petals ca. 2 em. long, rose, purple, or blue. Stamens included, LA PAZ: MURILLO: La Paz, 3000 m., Rusby 2855 (NY); LARECAJA: Mapiri, San Carlos, Sarampiuni, 600 m., Buchtien. 372 (NY); Haci- enda Casana, Tipuani, 1400 m., Buchtien 7186 (US). COCHABAMBA: CHAPARE: Antahuacana, Espíritu Santo, 750 m., Buchtien 2253 (US); Locotal, 1700 m., Steinbach 9596 (GH). Peru to Costa Rica and the West Indies. 19. T. Cardenasii L. B. Smith in Proc, Amer. Acad. 70: 154 (1935). Caulescent, flowering 20-25 cm. high. Stem to 1 dm. long, simple. Leaves to 2 dm. long, the blades linear-triangular, filiform-attenuate, 8 mm. wide, densely villous-lepidote with basally produced cinereous scales, involute. Scape erect, 1 mm. in diameter, glabrous. Scape- bracts lanceolate, exceeding the internodes, dark purple, appressed- lepidote, foliaceous-laminate. Inflorescence simple, lanceolate, 5 cm. long, 1 cm. wide, complanate, densely 4-7-flowered. Floral bracts like the scape-bracts but glabrous, 27 mm. long, ecarinate, submembrana- ceous, Flowers subsessile. Sepals lanceolate, attenuate, 20 mm. long, free, glabrous, nerved. Petals ligulate, lilac. Stamens included. CHU- QUISACA: PROVINCE [?]: Cerro Macho, 2730 m., Cárdenas 491 (GH; type). 20. T. ixioides Griseb. in Goett. Abh. 24: 333 (1879). More or less caulescent, flowering 1-3 dm. high. Leaves many, 10-18 cm. long, covered with appressed cinereous scales, the blades very narrowly triangular, 10 mm. wide, deeply channeled becoming involute toward apex. Scape erect or ascending, equaling or exceeding the leaves, slender, glabrous. Seape-bracts imbricate, ovate, subacute, thin, stramineous, lepidote. Inflorescence simple, of 3-7 polystichous flowers, sublax, 45 mm. long. Floral bracts suberect, broadly ovate or elliptic, subacute, ca. 20 mm. long, exceeded by the sepals, thin, stramineous, lepidote toward apex. Pedicels stout, to 5 mm. long, Sepals ovate, acute, to 19 mm. long, connate more highly posteriorly, carinate, nerved, glabrous. Petals to 34 mm. long, yellow, the blade elliptic. Stamens included, the filaments straight. CHUQUISACA: TOMINA: 2200 m., Cardenas 4093 (GH, US). Paraguay, Uruguay, Argentina. 1969] Bromeliaceae — Smith 249 21. T. xiphioides Ker-Gawl. in Bot. Register, 2: t. 105 (1816). T. Friesii of Catalogue, non Mez. Flowering 15-30 cm. or higher. Stem from very short to 15 cm. long, simple to few-branched. Leaves many, polystichous but sometimes almost distichous, curved or con- torted, to 25 cm. long, covered with cinereous or ferruginous sub- spreading scales, the blades narrowly triangular, subulate-attenuate, flat, to 2 em. wide. Scape from almost none to 12 cm. long. Scape- bracts densely imbricate, elliptic-oblong, thin, the lower ones caudate and lepidote, the upper apiculate and nearly or quite glabrous. In- florescence simple, of 2-10 distichous flowers, lance-oblong, acute, to 12 cm. long. Floral bracts densely imbricate, lance-oblong, acute, to 7 cm. long, much exceeding the sepals, submembranous, ecarinate, nearly or quite glabrous. Flowers to 10 cm. long. Sepals linear- lanceolate, attenuate, to 42 mm. long, free, glabrous, thin, nerved. Petals white, fragrant, the blade broadly elliptie, obtuse, spreading, 2 cm. wide, crenate-serrate. Stamens barely included or exserted from the throat of the corolla, the filaments straight. POTOSI: NOR CHICHAS: San Antonio, Cárdenas 93 (GH). Brazil, Uruguay, Ar- gentina. 22. T. Lorentziana Griseb. in Goett. Abh. 19: 217 (1874). T. Lor- entziana var. simplex O. Ktze. Rev. Gen. 3 (2): 304 (1898). Caules- cent, flowering 2-7 dm. high. Leaves 2-4 dm. long, the blades very narrowly triangular, to 20 mm. wide, covered with spreading cinereous or brownish scales, often carinate, the margins coarsely ciliate-lepi- dote. Scape erect, slender. Scape-bracts imbricate and usually cover- ing the scape, the lower foliaceous, the upper lanceolate, acute, densely lepidote. Inflorescence laxly bipinnate or sometimes simple, usually 12-15 em. long with 4-8 spikes. Primary bracts like the upper scape- bracts, much shorter than the spikes. Spikes divergent, straight, oblong-lanceolate, 6 cm. long, 15 mm. wide, strongly complanate, the bracts of the sterile base as large as the floral bracts. Rhachis flexuous, slender, angled, glabrous. Floral bracts suberect, imbricate but exposing the rhachis, acute, mostly 20-25 mm. long, slightly ex- ceeding the sepals, ecarinate, papyraceous, glabrous or nearly so. Sepals oblong, acute, 16-18 mm. long, subfree. Petals 35 mm. long, white. Stamens included, the filaments plicate. COCHABAMBA: CERCADO: Tunari, Kuntze s. n. (NY; type of T. Lorentziana var. simplex O. Ktze.) ; PROVINCE [?]: Abra de Sacaba, 2600 m., Cárdenas 2211 (GH); QUILLACOLLO: Cerro del Calvario, 2560 m., Cárdenas 3577 (GH). SANTA CRUZ: VALLE GRANDE: Pampa Grande, 1200 m., Steinbach 6034a (F, GH). Brazil, Paraguay, Argentina. 29. T. caulescens Brongn. ex Baker, Handbk. Bromel. 168 (1889). Plant to 45 cm. long. Stem much branched, to 30 em. long. Leaves very many, rigid, arching and secund or strict, 10-15 cm. long, densely appressed- or subpruinose-lepidote, the blades narrowly tri- 250 Rhodora [Vol. 71 angular, 5 mm. wide, involute, usually with a strong median ridge beneath. Scape distinct to almost none. Scape-bracts densely im- bricate, the lower ones foliaceous, the upper elliptic, acute or apiculate, nerved, roseate, more or less lepidote. Inflorescence simple, distichous- ly 14-flowered or less, linear or linear-lanceolate, acute, 5-7 cm. long, 10-12 mm. wide, strongly complanate. Floral bracts densely imbricate but partially exposing the rhachis, lance-ovate, acute, 15-20 mm. long, exceeding the sepals, ecarinate, subcoriaceous, glabrous, roseate. Sepals lanceolate, acute, 15 mm. long, equally subfree or connate posteriorly to 3 mm. Petals white, 25 mm. long. Stamens included, the filaments plicate. LA Paz: SUR YUNGAS [?]: Rio Bopi Basin, 1000 m., Rusby 670 (NY); San Bartolome, Calisaya, 750-900 m., Krukoff 10042 in part (MICH) ; White 536 (NY); alt. 500 m., White 1068 (K, NY). Peru. 24. T. didisticha Baker in Journ. Bot. 26: 16 (1888). T. crassifolia Baker, Handbk. Bromel. 174 (1889). Guzmania complanata Wittm. of Catalogue. Stemless, flowering 25-35 em. high. Leaves 1-3 dm. long, the blades very narrowly triangular, channeled, 6-20 mm. wide, covered with subappressed cinereous scales. Scape erect, slender. Scape-bracts covering the scape, elliptic, acute, densely lepidote. In- florescence bipinnate or at base subtripinnate, dense, fan-shaped with all the spikes in one plane, 6-14 cm. long. Primary bracts erect, nearly imbricate, like the scape-bracts, much shorter than the spikes. Spikes curved-spreading from an erect base, narrowly lanceolate, 3-8 em. long, strongly complanate, the sterile base with imbricate re- duced bracts, Floral bracts imbricate and covering the rhachis, ovate, acute, 11-15 mm. long, equaling the sepals, carinate, thin, strongly nerved, densely cinereous-lepidote to nearly glabrous. Sepals narrowly lanceolate, acuminate, slightly connate posteriorly. Petals 15-20 mm. long, white. Stamens included, the filaments plieate. BENI: CERCADO MOJOS: Trinidad to Missiones Guarayos, 250 m., Werdermann 2357 (MO). LA PAZ: CAUPOLICAN: Apolo, 1440 m., Williams 1484 (K, NY); NOR YUNGAS: Milluguaya, 900 m., Buchtien 4286 (US) ; SUR YUNGAS: San Bartolome, Calisaya, Río Bopi Basin, 750-900 m., Krukoff 10045 e. p. (F). COCHABAMBA: AYOPAYA: Santa Rosa, 2000 m., Cárdenas 4876 (US). CHUQUISACA: AZERO: Camatinga, Machareti, Herzog 1201 (L; type of Guzmania complanata Wittm.). SANTA CRUZ: CERCADO: Santa Cruz, 1000 m., Kuntze (NY); CHIQUITOS: Castelnau 2 (P; type of T. crassifolia Baker); Roboré de Chiquitos, 416 m., Cárdenas 2958 (GH); Santiago de Chiquitos 700 m., Cutler 7027 (GH); SARA: Buena Vista, 450 m., Steinbach 6664 (S). Brazil, Paraguay, Argen- tina. 25. T. vernicosa Baker in Journ. Bot. 25: 241 (1887). T. polyphylla Baker of Catalogue. Stemless, flowering to 3 dm. high. Leaves to 18 em. long, covered with tightly appressed fine scales, the blades 1969] Bromeliaceae — Smith 251 very narrowly triangular, rigid, 20 mm. wide, deeply channeled. Scape erect, equaling or shorter than the leaves. Scape-bracts densely imbricate, broadly elliptic, lepidote, the lower short-laminate, the upper rounded. Inflorescence digitate from 3-5 spikes, to 7 cm. long. Primary bracts like the upper scape-bracts, much shorter than the spikes. Spikes erect, sessile, narrowly lanceolate, 10 em. long, 9 mm. wide, complanate, densely 20-flowered. Floral bracts broadly elliptic, obtuse, 10 mm. long, about equaling the sepals, ecarinate, nearly or quite even, coriaceous, lepidote at apex, rose when dry. Sepals el- liptic, obtuse, free, rigid, even, glabrous. Petals 20 mm. long, white. Stamens included, the filaments plicate. LA PAZ: LARECAJA: Sorata, San Pedro, 2550 m., Mandon 1183 (P; type of T. polyphylla Baker). CHUQUISACA: ORIENTE: Charagua, 800 m., Cárdenas 2689 (F, GH). SANTA CRUZ: CERCADO: Sierra de Santa Cruz, Kuntze (NY) ; VELASCO: 200 m., Kuntze (NY); San José de Chiquitos, 255 m., Cutler 7084 e. p. Paraguay, Argentina, 26. T. tenuifolia L. Sp. Pl. 286 (1753). T. pulchella Hook. of Cata- logue. T. pulchella var. rosea of Catalogue, non T. rosea Lindl. More or less caulescent with stem to 25 cm. long, often branching and as- sociated in great mats, rather variable. Leaves 5-10 em. long, typical- ly exceeding the inflorescence, densely and minutely appressed-lepidote throughout, the blades narrowly triangular, subulate-attenuate, 2-7 mm. wide, channeled. Scape erect or ascending, slender, short. Scape- bracts imbricate, elliptic, caudate, membranaceous, rose. Inflorescence simple, polystichous-flowered, ovoid, densely 4-10-flowered. Floral bracts like the scape-bracts but suborbicular and less caudate, much exceeding the sepals. Sepals lanceolate, acute, 10 mm. long, much connate posteriorly, membranaceous, glabrous. Petals 20 mm. long, blue, white, or rose. Stamens included, the filaments plicate. LA PAZ: LARECAJA: Sorata, 2650 m., Mandon 1184 (S); Bang 1628 e. p. (G, GH); NOR YUNGAS: Polo-polo, Coroico, 1100 m., Buchtien 3672 (US); INQUISIVI: Quimes, 2400 m., White 218 (NY). COCHABAMBA: CERCADO: Tunari, Kuntze (NY); CHUQUISACA: CHARCAS: Santiago, 950 m., I. S. Nelson 58-455 (US); TOMINA [?]: Bartolo, Monteagudo to Sucre, 1350 m., Brooke 5684 (NY). SANTA CRUZ: VALLE GRANDE: Samaipata, 1800 m., Cárdenas 3521; 1500 m., Cárdenas 4604 (US). West Indies to Argentina. 27. T. Pohliana Mez in Mart. Fl. Bras. 3 (3): 597, t. 111 (1894). Stemless or short-caulescent, flowering 15-36 em. high. Leaves few or many, rosulate, to 28 cm. long, covered with spreading cinereous scales; sheaths small, narrow and indistinct; blades very narrowly triangular, 1-2 em. wide, strongly channeled. Scape erect, more or less curved, slender. Scape-bracts imbricate and covering the scape, the lower foliaceous, the upper elliptic with narrow blades. Inflores- cence simple with a few polystichous flowers, dense or subdense, 35- 252 Rhodora [Vol. 71 95 mm. long, densely cinereous-lepidote except the petals. Floral bracts suberect to spreading, broadly elliptic, inflated, exceeding the flowers, the lower caudate. Flowers subsessile. Sepals free, very broadly elliptic or suborbicular, apiculate, broadly convex, 11-15 mm. long, thick, coriaceous. Petals obtuse, 18-22 mm. long, white. Stamens included. SANTA CRUZ: VELASCO: San Ignacio, I. S. Nelson 58-G-14 (US). Brazil, Paraguay, Argentina. 28. T. laxissima Mez in Bull. Herb. Boiss. (ser. 2) 5: 108 (1905). Leaves many in a dense rosette, 3 dm. long, the blades ligulate or slightly contracted at base, acute, 3 cm. wide, chartaceous, minutely appressed-lepidote. Scape unknown. Inflorescence diffusely tripinnate, the axes slender, glabrous. Primary bracts spreading, lance-ovate, shorter than the sterile bases of the branches. Spikes linear, 9 cm. long, very laxly 8-16-flowered. Floral bracts elliptic, obtuse, 7 mm. long, ecarinate, subcoriaceous, nerved, glabrous. Flowers spreading, sessile. Sepals subsymmetric, narrowly elliptic, obtuse, 13 mm. long, subequally short-connate, coriaceous, nerved, glabrous. Capsule slenderly cylindric, abruptly acute, 35 mm. long. WITHOUT DEPART- MENT: Bang 2301 in part (GH, NY, US; type-number). 29. T. Duratii Vis. in Nuov. Sagg. Padov. 5: 271 (1840). Caulescent, flowering, 2-6 dm. long. Stem stout, curved, simple, to over 1 m. long. Leaves to 45 em. long, covered with coarse subappressed cinereous scales, the blades very narrowly triangular, spirally recurved and grasping branches of trees, 20 mm. wide, involute-subulate, rigid, thick. Scape erect, stout. Scape-bracts imbricate, elliptic, densely cinereous-lepidote, the lower linear-laminate, the upper apiculate. In- florescence 2-3-pinnate or rarely simple, 60 cm. long. Primary bracts like the upper scape-bracts, erect and sometimes exceeding the lower spikes. Spikes strict or curved outward at apex, lanceolate or linear- lanceolate, 3-8 cm. long, 1 cm. wide, complanate, subdense. Floral bracts erect, enfolding the base of the flower and exposing the rhachis, elliptie, exceeded by the sepals, ecarinate, coriaceous, densely lepidote in the typical variety. Sepals elliptic, 14 mm. long, unequally short-connate, coriaceous, even, glabrous. Petals 30 mm. long, the blades rhombic-orbicular, 16 mm. long, blue. Stamens deeply included, exceeding the pistil. TARIJA: CHACO: Villamontes, Yaquacuna, Troll 387 (B); PROVINCE ?: Weddell 4035 in part (P). Paraguay, Argen- tina, Uruguay. 29a. T. Duratii var. saxatilis (Hassler) L. B. Smith in Phytologia, 16: 78 (1968). T. decomposita Baker, Handbk. Bromel. 168 (1889). T. Weddellii Baker of Catalogue. T. confusa Hassler var. saxatilis Has- sler in Ann. Conserv. & Jard. Bot. Genéve, 20: 329 (1919). Floral bracts nearly or quite glabrous, lustrous. LA PAZ: LOAIZA: Teneria, 3000 m., Herzog 2488 (L). COCHABAMBA: AYOPAYA: Santa Rosa, 2000 m., Cárdenas 4375 (US); CERCADO: Tunari, Kuntze (NY, US). CHU- 1969] Bromeliaceae — Smith 253 QUISACA: ORIENTE: Charagua, 800 m., Cardenas 2792 (F, GH). SANTA CRUZ: VELASCO: West Velasco, 200 m., Kuntze (NY). TARIJA: Prov- ince [?]: Weddell 4035 in part (P; type of T. Weddelli Baker). Para- guay, Argentina, Uruguay. 90. T. Reichenbachii Baker, Handbk. Bromel. 166 (1889). T. tucu- manensis Mez of Catalogue. T. Herzogii Wittm. of Catalogue. Caules- cent, flowering over 2 dm. high but usually much smaller. Leaves to 14 em. long, the blades recurving but not spiraling, very narrowly triangular, involute-subulate, 8 mm. wide, covered with subappressed cinereous scales. Scape erect, slender. Scape-bracts imbricate, sub- lanceolate, subacute, nerved, lepidote. Inflorescence compactly bipin- nate from 2-8 spikes or sometimes simple, to 70 mm. long, 45 mm. wide. Primary bracts like the scape-bracts but obtuse, shorter than the spikes. Spikes subsessile, erect to somewhat decurved, linear, acute, 30 mm. long, 9 mm. wide, densely 5-7-flowered, complanate. Floral braets enfolding the flowers and exposing the rhachis, elliptic, obtuse, shorter than the sepals, nerved, glabrous, sublustrous. Sepals elliptic, rounded, 12 mm. long, unequally short-connate, subcoriaceous, even, glabrous. Petal-blades suborbieular, 7 mm. long, blue-violet. Stamens deeply included, exceeding the pistil. SANTA CRUZ: CORDIL- LERA: Cumbarute, 800 m., Herzog 1151 (L; type of T. Herzogii Wittm.); ORIENTE: Charagua, 800 m., Cárdenas 2682 (GH). CHU- QUISACA: TOMINA: Montecanto, 2000 m., Troll 77-A (B). TARIJA: CHACO: Curuenda, 200 m., Cárdenas 2532 (F, GH). Argentina. 31. T. streptocarpa Baker in Journ. Bot. 25: 241 (1887). T. soraten- sis Baker of Catalogue. T. condensata Baker of Catalogue. T. apoloen- sis Rusby of Catalogue. Flowering 1-5 dm. high. Stem 1 dm. to almost lacking. Leaves 8-50 em. long, covered with slender spreading cinereous scales, the blades spreading or recurved, linear-triangular, long-attenuate, to 15 mm. wide at base, involute-subulate. Scape erect, slender, nearly glabrous. Scape-bracts imbricate, lanceolate, acute, densely lepidote, nerved, the lowest with slender foliaceous blades. Inflorescence bipinnate from 2-12 spikes or rarely simple. Primary bracts like the upper scape-bracts, erect, much shorter than the spikes. Spikes linear to elliptic, arched-spreading from an erect densely bracteate sterile base, complanate, 3-12-flowered. Floral bracts lanceolate, acute or subobtuse, slightly shorter than the sepals, nerved, lepidote or glabrous. Flowers erect. Sepals free or very short-con- nate, oblong, broadly acute or obtuse, 10-13 mm. long, glabrous. Petals ca. 25 mm. long, the blades spreading, very broadly obovate, blue or purple. Stamens deeply included, exceeding the pistil. LA PAZ: LARE- CAJA: Sorata, 2600 m., Mandon 1185 (K; type of T. soratensis Baker) ; INQUISIVI: Suri, 2400 m., White 229 (NY); CAUPOLICAN: Apolo, 1440 m., Williams 1483 (NY; type of T. apoloensis Rusby); CERCADO: [?]: Río La Paz, 1180 m., I. S. Nelson 387 (US). COCHABAMBA: AYOPAYA: 254 Rhodora [Vol. 71 Sailapata, 2500 m., Cárdenas 3305 (GH); CERCADO: Tunari, 1300 m., Kuntze s. n. (NY). SANTA CRUZ: SARA: Buenavista, 450 m., Stein- bach 6665 (S, Z); VELASCO: West Velasco, 200 m., Kuntze s. n. (NY); VALLE GRANDE: San Isidoro, Comarapa, 1700 m., Vogel 474 (US). TARIJA: CHACO: Villamontes, Yaquacuna, 500 m., Troll 387 in part (B). CHUQUISACA: AZERO: Weddell s. n. (P; type of T. condensata Baker). Brazil, Paraguay, Uruguay, Argentina, Peru. 32. T. paleacea Presl, Rel. Haenk. 1: 125 (1827). T. scalarifolia Baker of Catalogue. Flowering 1-7 dm. long. Stem at least to 35 cm. long, much branched, appearing stout because of the leaf-sheaths. Leaves more than distichous but relatively few-ranked and widely spaced so that the blade and the upper half of the sheath are clearly visible, covered with slender cinereous or finally fuscous scales, the blades abruptly spreading, usually contorted, to 12 em. long, 4-6 mm. wide. Scape erect, slender, from very short to over 15 em. long, glabrous or nearly so. Seape-bracts imbricate or nearly so, elliptic, apieulate, lepidote, the lower filiform-laminate. Inflorescence simple, lanceolate with 1-12 distichous flowers, acute, to 5 em. long, dense, complanate. Floral bracts ovate or elliptie, about equaling the sepals, ecarinate, sulcate, soon glabrous. Sepals lanceolate, 10-17 mm. long, free, glabrous. Petal-blade spreading, suborbicular, blue or violet. Stamens deeply included, exceeding the pistil. BOLIVIA: Pentland S. n. (K; type of T. scalarifolia Baker). LA PAZ: LOAIZA [?]: La Granja, 2600 m., Julio 158 (BM, US). Chile, Peru, Colombia. 33. T. crocata (E. Morr.) Baker in Journ. Bot. 25: 214 (1887). T. Mandonii Morr. ex Mez of Catalogue. Flowering 15-35 em. long. Stem simple or few-branched, to 2 dm. and longer. Leaves distichous, 1-3 dm. long, covered with fine narrow reflexed scales, the blades spreading or recurving, linear, long-attenuate, 2-5 mm. in diameter. Scape terminal, erect or nearly so, slender, 5-15 cm. long, retrorse- tomentose like the leaves, naked or with a single foliaceous bract. Inflorescence always simple and distichous-flowered, lanceolate or elliptic, acute, 1-4 em. long, dense, 2-6-flowered. Floral bracts ovate, acuminate, to 2 em. long, about equaling the sepals, densely tomentose- lepidote. Flowers fragrant. Sepals sublanceolate, broadly acute or obtuse, thin, nerved, densely appressed-lepidote except in extreme age. Petals to 2 em. long, the blade suborbicular, obtuse, 6-8 mm. wide, bright yellow. Stamens deeply included, exceeding the pistil. La PAZ: LARECAJA: San Pedro to Coaconi, 2650 m., Mandon 1180 (K; type of T. Mandonii Morr. ex Mez; BM, GH, S). Brazil, Uruguay, Argentina. 34. T. bandensis Baker in Journ. Bot. 25: 934 (1887). T. pilosa L. B. Smith in Phytologia 7: 175 (1960). Flowering 1-2 dm. long. Stem much branched, 4-6 dm. long. Leaves densely distichous, 5-7 cm. long, densely pruinose- or tomentose-lepidote, the blades suberect 1969] Bromeliaceae — Smith 255 to spreading, linear, involute, long-attenuate, 1-2 mm. in diameter. Seape terminal, erect or decurved, to 9 cm. long, ca. 0.5 mm. in diameter, lepidote at least toward base. Scape-bracts 1-2, remote, elliptic, the upper acute or apiculate, the lower usually filiform-lami- nate and appressed-lepidote. Inflorescence simple with 2-4 distichous flowers, linear-lanceolate, 2-3 em. long, 3-5 mm. wide. Floral braets not imbricate, ovate-elliptic, acute, 9-12 mm. long, the lower ones equaling or shorter than the sepals and sometimes sparsely lepidote. Flowers erect, appressed to the rhachis. Sepals subelliptie, acute, equally subfree, thin, nerved, glabrous. Petals 15-16 mm. long, the blade broadly elliptic or obovate, 5 mm. long, blue or violet. Stamens deeply ineluded, exceeding the pistil. SANTA CRUZ: VALLE GRANDE: Cerro de Alto-Mairana, 2000 m., Steinbach 6039; FLORIDA: Saipina, 1800 m., Cárdenas 5513 (US; type of T. pilosa L. B. Smith). 35. T. loliacea Mart. ex Schult. f. Syst. Veg. 7: 1204 (1830). T. quadriflora Baker of Catalogue, Stem simple or branched, rarely more than 4 em. long. Leaves densely polystichous, 2-3 (-4) cm. long, densely and coarsely pruinose-lepidote, the blades erect to suberect or arching-secund, very narrowly triangular, 3-5 mm. wide. Scape terminal, straight or curved, to 10 cm. long, less than 1 mm. in diameter, lepidote. Scape-bracts many, about equaling the internodes, elliptic, acute, nerved, densely lepidote. Inflorescence simple, linear with up to 16 distichous flowers, to 4 cm. long, lax. Rhachis strongly geniculate, excavated next the flowers, lepidote. Floral bracts not imbricate, closely enfolding the flowers, ovate, acute, equaling or shorter than the sepals, thin, nerved, densely lepidote. Flowers erect and appressed to the rhachis. Sepals lanceolate, acute, to 9 mm. long, glabrous, nerved, equally short-connate. Petals 10 mm. long, pale violet (! Mez) to yellow, the blade narrow, acute. Stamens deeply included, exceeding the pistil. BOLIVIA: Mandon 1176 (! Mez); COCHABAMBA: AYOPAYA: Weddell 4175 (P). SANTA CRUZ: CHIQUITOS: San José de Chiquitos, 255 m., Cutler 1083 (GH) ; ORIENTE: Charagua, 800 m., Cárdenas 2681 (GH); vELASCO: 200 m., Kuntze s. n. (NY). Brazil, Paraguay, Argentina, Peru. 36. T. tricholepis Baker in Journ. Bot. 26: 42 (1888). Plant moss- like. Stems many from a single point, densely massed, at first bearing flowers when short and simple but later elongate and much branched, to 22 em. long. Leaves densely polystichous, typically to 15 mm. long, the blades narrowly triangular, involute, appressed or slightly diver- gent, reflexed-tomentose with fine linear-lobate scales. Scapes at first terminal, soon becoming pseudo-axillary by the elongation of the stem, 1-4 cm. long, 0.3-0.5 mm. in diameter, suleate, soon glabrous. Scape-bracts evenly distributed, nearly or quite equaling the inter- nodes, lanceolate, acute, 5-8 mm. long, thin, lepidote. Inflorescence simple, of 1-5 distichous flowers, narrowly lanceolate, to 17 mm. long. 256 Rhodora [Vol. 71 Floral bracts like the scape-bracts but broadly ovate, enfolding the rhachis for about half their length, slightly shorter than the sepals. Flowers erect, imbricate. Sepals lanceolate, acute, 6.5 mm. long, equally short-connate, thin, glabrous or nearly so. Petals linear, yellow, the blade indistinct, obtuse, spreading. Stamens deeply in- cluded, exceeding the pistil. BENI: MoJos: Trinidad, 250 m., Werder- mann 2396 (S). LA PAZ: INQUISIVI: Inquisivi to Cañamina, 1700 m., White 272 (NY); CAUPOLICAN: Apolo, Williams 1486 (BM, NY, US); LARECAJA: Poquerani, San Pedro near Sorata, 2500-2650 m., Mandon 1179 in part (K; type; BM, NY, P, S). SANTA CRUZ: FLORIDA: Comarapa, 2000 m., Cárdenas 4386 (US); Vogel 473 (US); CERCADO: Sierra de Santa Cruz, Kuntze s. n. (NY); ORIENTE: Charagua, 800 m., Cárdenas 2013 (GH); Agua Caliente, Charagua, Cárdenas 2611 (F, GH); cuiQUrTOS: San José de Chiquitos, 255 m., Cutler 7084-B (GH). CHUQUISACA: AZERO: Camatindi, near Cumbarute and Woyuywe, Rio Parapiti, Herzog 1195 (L, Z); VALLE GRANDE: El Palo, 1500 m., Steinbach 3805 (GH). TARIJA: O'CONNOR: Villamontes, Tar- tagal, 500 m., Pflanz 2151 (US). WITHOUT LOCALITY: Pearce s. n. (BM); Weddell 3655 (P). Brazil, Paraguay, Argentina. 36a T. tricholepis var. macrophylla L. B. Smith in Lilloa, 14: 98 (1948). Short-caulescent. Leaves to 5 em. long, the blades very slender. LA PAZ: LARECAJA: Charopampa, Mapiri, 570 m., Buchtien 1270 (US; type); NOR YUNGAS: Polo-Polo, Coroico, 1100 m., Buchtien 4541 (US). 37. T. bryoides Griseb. ex Baker in Journ. Bot. 16: 236 (1878) ; in Goett. Abh. 24: 334 (1879). Plant small and with a habit like Lycopo- dium Selago. Stems many from a single point, densely massed, simple or few-branched, rarely more than 5 em. long. Leaves densely poly- stichous, 4-6 (-9) mm. long, the blades erect, subtriangular, stout, to 2 mm. wide, densely cinereous-lepidote. Scape often lacking so that the 1-flowered inflorescence is sunk among the terminal leaves, when present evidently elongating only after anthesis, erect, slender, gla- brous, naked or with a single lanceolate bract enfolding its extreme base, to 3 cm. long. Inflorescence usually terminal but sometimes becoming pseudoaxillary by the elongation of the stem, always 1- flowered. Floral bract triangular-ovate, 7 mm. long, hyaline with a single median excurrent nerve, glabrous or nearly so. Sepals narrowly elliptic, 5-9 mm. long, hyaline, 3-nerved, equally short-connate. Petals linear, fleshy, sulphur-yellow drying to orange-brown. Stamens deeply included, exceeding the pistil. LA PAZ: MURILLO [?]: Huaricana, 800 m., Buchtien 806 (NY, US); MURILLO: Cotaña Illimani, 2450 m., Buchtien 4144 (GH); LARECAJA: Sorata, 2500-2650 m., Mandon 1179 in part (BM); Sorata, Poqueraria San Pedro, 2400-2600 m., Mandon 1179 bis (G); LoAIZA [?]: Chivesi, 4000 m., Pentland 29 (P). CocHa- BAMBA: AYOPAYA: Sailapata, 2500 m., Cárdenas 3309 (GH); TARATA 1969] Bromeliaceae — Smith 257 [?]: Parotani, 2400 m., Kuntze s. n. (NY). Potosi: PROVINCE [?]: Rio Blanco, 3000 m., Cárdemas 318 (US). CHUQUISACA: SUR CINTI: Camataqui, 1800 m., Cárdenas 213 (GH). SANTA CRUZ: ORIENTE: Charagua to Izozog, 800 m., Cárdenas 2690 (GH); CORDILLERA: Cum- barute and Woyuywe, Rio Parapiti, Herzog (Herzog). WITHOUT LO- CALITY: Miers 7592-7592x (BM). Argentina, Peru. 38. T. funebris Castellanos in Anal. Mus. Nac. Hist. Nat. Buenos Aires, 37: 502 (1933). Rarely flowering over 10 cm. long. Stems sev- eral from a single point, simple or few-branched, 2-5 cm. long. Leaves densely distichous or polystichous, to 5 em. long, the blades divergent to reflexed, contorted, triangular-subulate, 2-3 mm. in diameter, dense- ly cinereous-lepidote. Scape distinct, to 5 em. long, slender, glabrous. Scape-bracts 2, remote, usually a somewhat lepidote one near the base of the scape and a glabrous one near its apex. Inflorescence 1-2- flowered, glabrous. Floral bracts like the upper scape-bract but more ovate, about equaling the sepals, even. Sepals lanceolate, acute, 10 mm. long, equally short-connate. Petals 13 mm. long, coffee-colored drying to dark orange-brown, the blade distinct, subrhombic, obtuse. Stamens deeply included, exceeding the pistil. SANTA CRUZ: ORIENTE: Charagua to Izozog, 800 m., Cárdenas 2688 (GH). Paraguay, Ar- gentina. 39. T. Gilliesii Baker in Journ. Bot. 16: 240 (1878). Flowering to over 2 dm. long. Stems many from a single point, forming a dense subglobose mass, simple or few-branched. Leaves densely distichous, 2-8 cm. long, covered with subappressed subsymmetrical cinereous to ferruginous scales, the blades arching-recurved, narrowly triangular, much compressed laterally, attenuate with a very narrow channel above. Scape almost lacking to 13 cm. long, appressed-lepidote, naked or bearing a single lanceolate bract. Inflorescence 1-3-flowered, densely cinereous-lepidote, the rhachis flexuous. Floral bracts erect, lance- ovate, acute, to 20 mm. long, about equaling the sepals, not more than twice as long as the internodes, ecarinate, nerved. Sepals oblong- lanceolate, to 16 mm. long, equally subfree. Petals narrow. Stamens deeply included, exceeding the pistil. COCHABAMBA: CERCADO: Arani, 2800 m., Cárdenas 2418 (GH). CHUQUISACA: SUR CINTI: Camataqui, 1800 m., Cárdenas 214 (GH); ciNTI: Puca Khasa near Tacaquira, Hammarlund 313 (S). Argentina, Peru. 40. T. myosura Griseb. ex Baker in Journ. Bot. 16: 240 (1878); in Goett. Abh. 24: 333 (1879). Flowering to over 3 dm. high. Stems many from a single point, forming a dense subglobose mass, simple or few-branched. Leaves densely distichous, 5-17 cm. long, the blades strongly recurved, triangular-subulate, 3-5 mm. in diameter, evenly rounded when fresh with a channel along most of the upper side, covered with cinereous subappressed to reflexed-pruinose scales. Scape always distinct, to 20 cm. long, lepidote or glabrous, naked or with 258 Rhodora [Vol. 71 a single lanceolate lepidote bract well removed from the inflorescence. Imflorescence simple, linear, to 8 em. long, laxly 1-8-flowered, densely cinereous-lepidote. Floral bracts remote, rarely more than twice as long as the internodes, ovate, attenuate, to 21 mm. long, longer to shorter than the sepals, nerved. Flowers strict. Sepals oblong-lanceo- late, 9-15 mm. long, equally subfree, thin. Petals linear, to 20 mm. long, yellowish white. Stamens deeply included, exceeding the pistil. COCHABAMBA: CERCADO: Cochabamba, 2530 m., Jiminez 22 (US). PoTosi: NOR CHICHAS: Cotagaita, 2600 m., Troll 3369 (B). Uruguay, Argentina. 41. T. recurvata L. Sp. Pl. (ed. 2) 1: 410 (1762). Rather variable, flowering 4-23 cm. high. Stems densely massed, simple or few- branched, 1-10 cm. long. Leaves densely distichous, 3-17 cm. long, densely pruinose-lepidote, cinereous or ferruginous, the blades usually recurved, linear, terete, 0.5-2 mm. in diameter, soft with a weak apex. Scape terminal, always distinct, to 13 cm. long, about 0.5 mm. in diameter. Scape-bracts linear-lanceolate, lepidote, 1 or rarely 2 im- mediately below the inflorescence, sometimes one next the inflorescence and one remote. Inflorescence 1-2(-5)-flowered, dense. Floral bracts like the scape-bracts but smaller, mostly equaling or exceeding the sepals, nerved, densely lepidote. Sepals lanceolate, 4-9 mm. long, free or equally short-connate, usually glabrous. Petals narrow, pale violet or white. Stamens deeply included, exceeding the pistil. BENI: CERCADO MOJOS: Trinidad, 250 m., Werdermann 2530 (S). LA PAZ: MURILLO: Cotana, Illimani, 2450 m., Buchtien 4143 (NY); LARECAJA: Sorata, Mandon. 1177 (GH, NY, S); Williams 2424 (BM, NY); In- genio del Oro, 3000 m., Rusby 2165 (NY, US); CAUPOLICÁN: Apolo, Williams 1485 (BM, NY, US). SANTA CRUZ: PROVINCE [?]: Achira, 1000 m., Steinbach 8224 (GH); SARA: Santa Cruz, 450 m., Steinbach 3807 in part (GH); 7452 (BM, F, MO, S); VALLE GRANDE: El Palo, 1500 m., Steinbach. 3807 in part (GH). Southeastern United States to northern Argentina and Chile. 42. T. capillaris R. & P. Fl. Peruv. 3: 42, t. 271, fig. e (1802). Very variable in both size and form, flowering to 16 cm. high, Stems many from a single point, densely massed, simple or branched. Leaves distichous, 1-4(-9) em. long, densely and finely pruinose-lepidote, cinereous or ferruginous, the blades erect to spreading, typically linear and less than 2 mm. in diameter, obtuse. Scape typically pseudo- axillary, elongate, slender, always naked, glabrous or nearly so. Inflorescence normally 1-flowered. Floral bracts ovate, acute or apicu- late, thin with 5 or more strong nerves, typically soon glabrous, usually equaling or exceeding the sepals. Sepals lanceolate, acute or obtuse, to 8 mm. long, connate posteriorly. Petals narrow with scarce- ly distinet blade, white, yellow, or brown. Stamens deeply included, exceeding the pistil. L4 PAZ: MURILLO: La Paz, Buchtien 381 in part 1969] Bromeliaceae — Smith 259 (S); Claude-Joseph 1109 (US); Weddell 4339 (P) ; Río Abajo, Huari- cana, 2700 m., Buchtien 807 (F); Cotana to Illimani, 2450 m., Buch- tien 4026 (GH); Obrajes, 3300 m., Buchtien 6383 (US); OMASUYOS: Achacache, 4000 m., Mandon 1181 (BM). COCHABAMBA: CERCADO: Cochabamba, 2700 m., Buchtien 2416 (NY, US); Cala-Cala, Ehrhorn (UC); Tunari, 1950 m., M. B. Foster 2536 (US); Sacaba, 2500 m., Cutler 7427 (GH); PROVINCE [?]: Recoleta, 2560 m., Cárdenas 4379 (US). SANTA CRUZ: CERCADO: Santa Cruz, 600 m., Kuntze (NY). TARIJA: CERCADO: Tarija, 2000 m., Cárdenas 211: 215 (GH); PROV- INCE [?]: Escayo, 3600 m., Fiebrig 3570 (BM); San Luis, Pearce s. n. (BM). Peru, Argentina. 42a. T. capillaris forma incana (Gill.) L. B. Smith in Proc. Amer. Acad. 70: 211 (1935). Leaves short and stout, mucronate, appressed and ascending so that they form an almost continuous plane. Scapes evident after anthesis and mostly exceeding the leaves, pseudo-axil- lary. Floral bracts with at least 5 strong nerves meeting near the apex, glabrous or subglabrous, LA PAZ: MURILLO: Obrajes, 3450 m., Buchtien 9364 (GH). COCHABAMBA: CERCADO [?]: Cerro San Pedro, 2600 m., Jiminez 23 (US). Potosi: NOR CHICHAS: Chorolque, 3900 m., Cárdenas 91 (GH); FRIAS: Potosí, 4000 m., Cárdenas 209 (GH). Argentina. 42b. T. capillaris forma cordobensis (Hieron.) L. B. Smith in Proc. Amer. Acad. 70: 211 (1935). Leaves widely spaced so that the greater part of each sheath is uncovered. Scapes evident after anthesis, usually terminal, mostly exceeding the leaves. Floral bracts with at least 5 strong nerves meeting near the apex, densely and persistently lepidote. LA PAZ: LARECAJA: Ingenio del Oro, 3300 m., Rusby 2165 (F, GH); Sorata, 2600 m., Mandon 1177 in part (NY); Williams 2424 in part (US); MURILLO: Cotana to Illimani, 2540 m., Buchtien 6380 (US). Argentina. 42c. T. capillaris forma virescens (R. & P.) L. B. Smith in Proc. Amer. Acad. 70: 212 (1935). T. Williamsii Rusby of Catalogue. Scapes never elongating much even after anthesis, mostly shorter than the leaves. Floral bracts with at least 5 strong nerves meeting near the apex. LA PAZ: MURILLO: La Paz, Bang 123 (G, NY, US); Buch- tien s. n. (US); 107 (GH); 155 in part (US); 381 in part (BM, MO, POM, S); 805 in part (MVM, NY, S, US); 2556 (NY); 9159 (GH); 9160 (GH); 9863 (GH); 9365 (GH); Jaffuel 585 (GH); d'Orbigny 298 (P); Pentland 27 (P); 3300 m., Rusby 2164 (F, NY, US); Wed- dell s. n. (P); Obrajes, 3550 m., Buchtien 9365 (GH); PACAJES: Ul- loma to Callapa, 400 m., Asplund 6382 (US); OMASUYOS: Achacache, 4000 m., Mandon 1181 (P); ANDES: Las Peñas, Mandon 1178 in part (P). ORURO: cERCADO: Cerro de Oruro, Herzog 2447 (L); PooPó: Pazña, 4200 m., Buchtien 1271 (US). COCHABAMBA: CERCADO: Cocha- bamba, 3000 m., Kuntze (NY). Potosi: FRIAS: Potosí, d’Orbigny 260 Rhodora [Vol. 71 1493 (P); Aroifilla, 3800 m., Cárdenas 4385 (US). TARIJA: CERCADO [?]: Escayache, Gross 3373 (B). Peru, Argentina, Chile. 42d. T. capillaris forma Hieronymi (Mez) L. B. Smith in Proc. Amer. Acad. 70: 213 (1935). Habit of the typical form. Floral bracts with only 3 strong nerves meeting near the apex, the other nerves if any short and with free apices. L4 PAZ: MURILLO: Palca to Illimani, 3600-4300 m., Hauthal s. n. (S). SANTA CRUZ: VALLE GRANDE: El Palo, 1500 m., Steinbach 3807 in part (GH). TARIJA: CERCADO: Tarija, 2000 m., Cárdenas 212 (GH). Argentina. 43. T. usneoides L. Sp. Pl. (ed. 2) 1: 411 (1762). Growing pendent from trees in slender branching strands up to 8 m. long. Roots usually absent. Stem less than 1 mm. thick, sympodial, the internodes 3-6 em. long with only the extreme base covered by the leaves, curved. Leaves distichous, 5 em. long, densely lepidote, the blades filiform, less than 1 mm, thick. Scape lacking. Inflorescence reduced to a single flower. Floral bracts ovate, shorter than the sepals, lepidote. Sepals narrowly ovate, acute, to 7 mm. long, thin, nerved, glabrous, equally short-connate. Petals narrow, 9-11 mm. long, pale green or blue. Stamens deeply included, exceeding the pistil. LA PAZ: MURILLO: La Paz, Buchtien 155 in part (GH); Jaffuel 583 (GH); 3000 m., Bang 107 (BM, F, GH, MO, NY); Illimani, Buchtien s. n. (BM); Cotaña to Illimani, 2600 m., Buchtien 6381 (US); Obrajes, 3300 m., Buchtien (F, MO, NY, POM, S) ; LARECAJA: Sorata, Mandon 1102 (NY); Brooke 6524 (NY); SUR YUNGAS: Sirupaya, Yanacachi, 2100 m., Buchtien 390 (US). COCHABAMBA: CERCADO: Morachata to Cochabamba, Cutler 7698 (GH); Tunari, 1300 m., Kuntze s. n. (NY). CHUQUISACA: CINTI: Huarinota, 3200 m., Hammarlund 375 (S). SANTA CRUZ: FLORIDA: Comarapa, 2000 m., Steinbach 8548 (AA, US). South- eastern United States to central Argentina and Chile. 44. T. Seemannii (Baker) Mez in DC. Monog. Phan. 9: 737 (1896). Stemless or short-caulescent. Leaves 10-15 cm. long, the sheaths broadly ovate, dark brown, the blades narrowly triangular, to 8 mm. wide. Scape erect, very slender, usually shorter than the leaves. Seape-bracts suberect, suborbicular, shorter than the internodes. Spikes 1 or 2, densely few-flowered. Floral bracts imbricate, sub- orbicular, acute or apiculate, 14 mm. long, exceeding the flowers, membranaceous, pale-lepidote. Sepals asymmetric, obtuse, 9.5 mm. long, thin, nerved, lepidote, Petals 12-15 mm. long. Stamens included. COCHABAMBA: CHAPARE: Llanta Aduana, 3500 m., Cárdenas 2252 (GH); CERCADO [?]: Yungas, 3300 m., Foster 2548 (US). Peru, Ecuador, Colombia. 45. T. spiculosa Griseb. in Goett. Nachr. 1864: 17 (1865). T. triticea Burchell ex Baker of Catalogue. T. micrantha Baker of Catalogue. Stemless, flowering to 8 dm. high. Leaves 3-4 dm. long, obscurely 1969] Bromeliaceae — Smith 261 punctulate-lepidote, often irregularly purple-spotted, the sheaths large, broadly ovate, dark brown, the blades rounded and apiculate or acute, 2-4 cm. wide. Scape erect, often much exceeding the leaves. Scape- bracts narrowly elliptic, about equaling the internodes or the upper ones slightly shorter. Inflorescence bipinnate or tripinnate, lax. Pri- mary bracts narrowly ovate, much shorter than the branches. Spikes linear, dense, 9 cm. long, 6-9 mm. wide, spreading, usually long-stipi- tate with sterile bracts at the base. Floral bracts broadly ovate, (5-) 6-9 mm. long, equaling or slightly exceeding the sepals, convex, ecari- nate, coriaceous, even, glabrous or nearly so. Sepals asymmetric, elliptic, even, glabrous. Petals orange or yellow. Stamens included. La Paz: Yungas 1800 m., Rusby 2852 (NY; type of T. micrantha Baker) ; SUR YUNGAS: San Bartolome, Calisaya, 750-900 m., Krukoff 10512 (GH, NY); LARECAJA: Copacabana, Mapiri, 850-950 m., Krukoff 11155 (NY). SANTA CRUZ: COMARAPA: Yungas de San Mateo, 2500 m., Steinbach 8416 (GH). WITHOUT LOCALITY: Bang 2301 in part (F, GH). Peru and Brazil to Colombia and Trinidad. 46. T. monticola Mez & Sodiro in Bull. Herb. Boiss. (ser. 2) 4: 1135 (1904). Stemless, flowering 7 dm. high. Leaves 3 dm. long, punctulate-lepidote, the sheaths forming a pseudobulb, the blades oblong, acute, 35 mm. wide, rigid when dry. Scape slender, much exceeding the leaves. Scape-bracts elliptic, slightly exceeding the internodes. Inflorescence erect, laxly tripinnate, subthyrsoid, 30 cm. long, 8 cm. wide, glabrous or nearly so. Primary bracts elliptic, ex- ceeding the sterile base of the branch. Branches subdistichously in- serted, usually bearing 3 spikes. Spikes oblong, acute, 40 mm. long, 13 mm. wide, laxly 14-flowered, complanate, the rhachis geniculate. Floral bracts not imbricate nor concealing the rhachis at maturity, broadly ovate, ecarinate, 8 mm. long, exceeding the sepals, coriaceous, even, especially the upper ones incurved at apex. Sepals strongly asymmetric, obovate, broadly rounded, 6 mm. long, sparsely lepidote. Petals 7 mm. long. Stamens included. LA PAZ: LARECAJA: San Carlos, Mapiri, 850 m., Buchtien 367 (NY); 750 m., Buchtien 1266 (US). Ecuador. 47. T. pendulispica Mez in DC. Monog. Phan. 9: 745 (1896). T. triangularis Rusby of Catalogue. Stemless, flowering about 5 dm. high. Leaves 3 dm. long, densely and finely appressed-lepidote, often purple-spotted, the sheaths very large, ovate, forming an ellipsoid pseudobulb, the blades linear or narrowly triangular, attenuate, 8-12 mm. wide. Scape erect, slender, glabrous. Scape-bracts lanceolate, acute, much shorter than the internodes. Inflorescence bipinnate or the lowest branches divided, of 16 spikes or often less, Primary bracts like the scape-bracts but obtuse, usually much shorter than the sterile base of the branch. Spikes narrowly oblong, 5 cem. long, 1 cm. wide, nearly uniform, rather densely 10-20-flowered, the lower 262 Rhodora [Vol. 71 ones long-stipitate and pendulous or arching-decurved, the rhachis geniculate. Floral bracts, suborbicular-ovate, 7 mm. long, exceeding the sepals, ecarinate, submembranaceous, nerved, soon glabrous. Sepals asymmetric, obovate, rounded-emarginate, 5 mm. long, even, soon gla- brous. Petal-blades 2 mm. long, yellow. Stamens included. CocHA- BAMBA: CERCADO: Cochabamba, Bang s. n. (NY; type of T. triangu- laris Rusby). Peru, Ecuador. 48. T. pardina L. B. Smith in Lilloa, 14: 98 (1948). Stemless, flowering to 57 em. high. Leaves to 24 cm. long, densely and min- utely pale-lepidote, the blades ligulate, very broadly acute or sub- rounded and apiculate, 5 cm. wide, flat, dark-spotted. Scape erect, slender. Scape-bracts elliptic, about equaling the internodes, densely pale-lepidote. Inflorescence laxly tripinnate, 33 cm. long, sparsely lepidote. Primary bracts like the scape-bracts, shorter than the sterile bracteate bases of the branches. Branches spreading. Spikes dense or interrupted toward base, 10-17-flowered, the rhachis geniculate. Floral bracts broadly ovate, shorter than the sepals, ecarinate, even. Sepals asymmetric, broadly obovate, 5 mm. long, even. LA PAZ: LARE- cAJA: Hacienda Casana, Tipuani, 1400 m., Buchtien 7183 (US; type). 49. T. Bakeri L. B. Smith in Contrib. Gray Herb. 95: 45 (1931). T. flexuosa (Baker) Mez of Catalogue. Stemless or short-caulescent, flowering to 11 dm. high. Leaves 3 dm. long, punctulate-lepidote, thin, the sheaths narrowly ovate-elliptic, dark brown, the blades nar- rowly triangular, caudate, pale-lepidote. Scape erect, slender. Scape- bracts lance-elliptie, attenuate, the upper ones somewhat shorter than the internodes. Inflorescence erect or pendulous, tripinnate, 40 cm. long, 14 em. wide, lax, the axes flexuous, slender. Lower branches with a long stipe that equals or exceeds the primary bracts, composed of 2-3 spikes. Spikes oblong, 4-5 cm. long, laxly 8-14-flowered, the rhachis straight or nearly so, densely ferruginous-lepidote. Floral bracts ovate, acute, sherter than the sepals, nerved, ferruginous-lepi- dote. Sepals asymmetric, elliptic, broadly rounded, 4-5 mm. long, ferruginous-lepidote. Petal-blades 1 mm. long, yellow. Stamens in- cluded. LA PAz: LARECAJA: Anansa, Cerro de Uacani, 2700 m., Man- don 1172 (K; type). Peru. 50. T. parviflora R. & P. Fl. Peruv. 3: 41, t. 259 (1802). Stemless, flowering less than 3 dm. high. Leaves subbulbous-rosulate, 15 cm. long, finely pale-lepidote, the sheaths large, broadly ovate to sub- orbieular, the blades linear, attenuate, 4-8 mm. wide. Inflorescence erect or pendulous, bipinnate, pale-lepidote, the axes slender, flexuous. Primary bracts lanceolate, not more than equaling the sterile base of the branch. Spikes linear, very laxly 6-20-flowered, usually long- stipitate, spreading. Floral bracts ovate, acute, about half as long as the sepals, nerved, Flowers suberect to spreading. Sepals asym- metrie, 3-4 mm. long. Petals slightly exceeding the sepals, yellow. 1969] Bromeliaceae — Smith 263 Stamens included. LA PAZ: SUR YUNGAS: Calisaya, Rio Bopi, 1800 m., Cárdenas 1216; 1376 (GH). SANTA CRUZ: PROVINCE [?]: Cerro Hosana, 1400-1900 m., Steinbach 3457 (GH); COMARAPA: Yungas de San Mateo, 2500 m., Steinbach 8404 (GH). Peru. 51. T. adpressa André, Enum. Bromél. 6 (Dec. 13, 1888); in Rev. Hortic. 60: 566 (Dec. 16, 1888). Stemless, 2-7 dm. high. Leaves usually forming an ovoid pseudo-bulb, 15-30 cm. long, the sheaths broadly elliptic-ovate, dark brown, the blades narrowly triangular, to 20 mm. wide, densely pale-lepidote. Scape erect or nearly so. Scape bracts from shorter than to exceeding the internodes, narrowly ovate-oblong, long-caudate, densely appressed-lepidote. Inflorescence bipinnate, nar- row, erect or arching, the axis straight to geniculate. Primary bracts attenuate to caudate, usually equaling or exceeding the spikes. Spikes 8-12-flowered, typically erect or suberect. Floral bracts shorter than the sepals, ovate, acute, densely tomentose-lepidote. Flowers spread- ing. Sepals strongly asymmetric, obovate, 5 mm. long. The typical variety not found in Bolivia. Ecuador to Costa Rica. 51a. T. adpressa var. Tonduziana (Mez) L. B. Smith in Contrib. Gray Herb. 89: 8 (1930). Spikes spreading or reflexed at maturity. LA PAZ: LARECAJA: Tipuani Valley, Hacienda Casana, 1400 m., Buch- tien 7188 (NY, US). COCHABAMBA: CARRASCO: Phaecha, 2700 m., Cárdenas 6287 (US); PROVINCE [?]: *Schuenka" Schuenca?, Cárdenas s. n. (US). Peru to Costa Rica. 52. T. juncea (R. & P.) Poir. in Lam. Encyc. Suppl. 5: 309 (1818). Flowering 2-4 dm. high, often propagating by scaly branching rhi- zomes. Leaves many in a fasciculate rosette, about equaling the in- florescence or shorter, the sheaths triangular-ovate, ferruginous, the blades linear-subulate, filiform-attenuate. Scape erect or ascending. Scape-braets densely imbricate, lance-elliptic, attenuate, filiform-lami- nate, densely pale-lepidote. Inflorescence very densely digitate from a few spikes or sometimes reduced to a single densely polystichous- flowered spike, ovoid, rarely over 7 em. long. Primary bracts like the upper scape-bracts, subinvolucrate below the inflorescence, their sheaths slightly shorter than the lower spikes. Spikes sessile, elliptic or lanceolate, acute, to 4 em. long, slightly complanate. Floral bracts imbricate, broadly ovate, broadly acute or obtuse, carinate in the distichous-flowered spikes, nearly or quite even, densely lepidote. Sepals lanceolate, acute, 15-20 mm. long, coriaceous, even, glabrous or nearly so, much connate posteriorly. Petals tubular-erect, to 4 em. long, violet. Stamens and pistil exserted. LA PAZ: SUR YUNGAS: San Bartolome, Calisaya, Rio Bopi Basin, 750-900 m., Krukoff 10042 (GH, NY). CocHABAMBA: CERCADO: Tunari, Kuntze s. n. (NY). Peru to the West Indies. 264 Rhodora [Vol. 71 53. T. polystachia L. Sp. Pl. (ed. 2) 1: 410 (1762). Flowering 20- 65 cm. high. Leaves many in a dense subspreading rosette, usually much exceeded by the inflorescence and arched-recurving, covered with fine pale appressed scales, often with a narrow purple margin, the sheaths broadly ovate or elliptic-oblong, large and conspicuous, pale ferruginous toward the base, the blades long-attenuate, 10-25 mm. wide, usually flat. Scape erect or ascending, slender. Scape-bracts foliaceous with erect densely imbricate sheaths and long spreading or recurved blades, green, red, or purple. Inflorescence pinnately or subdigitately compound or very rarely simple, usually subcylindric or slenderly fusiform, dense, to 3 dm. long. Primary bracts elliptic, acute, the lower ones short-laminate, much shorter than the spikes. Spikes erect, sessile, linear, acute, 3-10 cm. long, 1 cm. wide, com- planate. Floral bracts barely to densely imbricate, broadly ovate, 15-20 mm. long, about equaling the sepals, even and glabrous or nearly so, slightly carinate toward apex. Sepals elliptic, coriaceous, glabrous, short-connate posteriorly. Petals tubular-erect, 3 cm. long, violet. Stamens and pistil exserted. BENI: CERCADO MOJOS: Trinidad, Missiones Guarayos, 250 m., Werdermann 2299 (MO, S). LA PAZ: SUR YUNGAS: San Bartolome, Calisaya, 750-900 m., Krukoff 10516 in part (MO). SANTA CRUZ: CHIQUITOS: Puerto Suarez, Fiedrich 8675 (AS). Mexico and the West Indies to Brazil. 54. T. Valenzuelana A. Rich. Fl. Cub. Fanerog. 2: 267 (1853). Stem- less, flowering 2-6 dm. high. Leaves to 4 dm. long, finely cinereous- lepidote, the sheaths concolorous with the blades, large, the blades narrowly triangular, 10-25 mm. wide, flat. Scape erect or ascending, slender, glabrous. Seape-bracts imbricate, ovate, inflated, the lower ones laminate. Inflorescence simple or pinnately compound from a few spikes. Primary bracts like the upper scape-bracts, their sheaths much shorter than the spikes but their blades sometimes exceeding them. Spikes oblong, acute, 5-20 cm. long, 1-2 em. wide, complanate, often rather lax, 6-17-flowered, especially the terminal spike with sterile bracts at base. Floral bracts mostly 2-3 times as long as the internodes, usually exposing the rhachis, elliptic-oblong, obtuse or apiculate, 2 em. long, much exceeding the sepals, nerved, mostly ecari- nate, subglabrous. Sepals oblong, obtuse, connate posteriorly. Petals linear, 3 cm. long, lilac or violet. Stamens exserted. COCHABAMBA: AYOPAYA: Naranjito (Choco), 2200 m., Cardenas 4416 (US). WITH- OUT LOCALITY: Bang (NY). Florida and southern Mexico to Bolivia. 56. T. paraénsis Mez in Mart. Fl. Bras. 3(3): 586 (1894). T. bolivien- sis Baker in part but not as to type. Stemless, flowering 15-46 cm. high. Leaves to 35 em. long, minutely pale-lepidote, the outer ones reduced to small acute sheaths, the inner sheaths large and forming an ovoid pseudobulb, the blades narrowly triangular, abruptly acute, involute. Scape curved, suberect. Scape-bracts elliptic, equaling or 1969] Bromeliaceae — Smith 265 slightly exceeding the internodes, densely lepidote, the lower long- laminate, the upper apiculate. Inflorescence simple or of two spikes. Primary bract like the upper scape-bracts, much shorter than the axillary spike. Spikes linear, to 25 cm. long, complanate, 6-17-flowered. Floral bracts erect and imbricate but later convolute about the flowers and exposing the rhachis, broadly elliptic or suborbicular, 25-32 mm. long, ecarinate, nerved, densely lepidote. Sepals elliptic, acute, 24 mm. long, free, ecarinate, lepidote. Petals 33-70 mm. long, erect, red. Stamens exserted. LA PAZ: LARECAJA: Tuiri, Mapiri, 490-750 m., Krukoff 10908 (F, GH, MO, NY); SUR YUNGAS: Santa Ana, Rio Bopi, 400 m., White 1087 (MICH, NY); Rio Bopi Basin, San Bartolome, Calisaya, 750-900 m., Krukoff 10045; 10229 (NY). SANTA CRUZ: ICHILO: Rio Yapacani, 350 m., Steinbach 7596 bis (GH, NY). WITH- OUT LOCALITY: Bang 159-A in part (K, NY). Western Brazil, Peru, Colombia. 8. Vriesea Lindl. nomen conservandum See Phytologia 13: 84 (1966). Stemless herbs. Leaves densely rosulate, entire. Inflorescence simple or compound. Spikes usually distichous-flowered. Floral bracts conspicuous, enfolding or concealing the bases of the flowers. Sepals free. Petals nearly or quite free, bearing 2 scales on the inside at base. Ovary superior, the style elongate. Ovules many, caudate. Seeds fusiform, with a long straight basal coma. a. Leaf-blades narrowly triangular, densely cinereous-lepidote on both sides. b. Floral bracts convex, ecarinate, 25-35 mm. long. ................ een UU DM E RON PCENA T DUET E 1. V. incurva. b. Floral bracts carinate, 13 mm. long. .................... 2. V. heterandra. a. Leaf-blades ligulate; inflorescence simple. c. Floral bracts dense, much exceeding the sepals, chartaceous. .... oh di Ma NAAR RPSL OU TAI QUE: 3. V. heliconioides. c. Floral bracts lax, about equaling the sepals, subcoriaceous, ....... E eatin cj deiese sca saan s m. 4. V. Maxoniana. 1. Vriesea incurva (Griseb.) R. W. Read in Phytologia 16: 458. 1968. Tillandsia incurva Griseb. (“incurvata”) of Catalogue. Flower- ing 15-40 cm. long. Leaves many in a subglobose rosette, 15-35 cm. long, the sheaths large, distinct, broadly ovate to suborbicular, dark castaneous, the blades narrowly triangular, 2-3 cm. wide, cinereous. Scape slender, short. Scape-bracts barely imbricate, obovate or el- liptie, densely lepidote, the lower caudate. Inflorescence pendulous, simple or digitate with 2-5 spikes. Primary bracts scarcely larger than the floral bracts. Spikes strict, linear or lance-linear with sev- eral sterile bracts at base, acute, 10-24 em. long, strongly complanate, 7-16-flowered. Floral bracts two to three times as long as the inter- nodes but usually exposing most of the rhachis, elliptic, obtuse, 25-35 266 Rhodora [Vol. 71 mm. long, ecarinate at maturity, red, submembranaceous, nerved to almost even, lepidote to glabrous. Sepals free, elliptic, 15-20 mm. long, ecarinate, even, soon glabrous. Petals 35 mm. long, yellow. Stamens exserted. LA PAz: LARECAJA: San Carlos, Mapiri, 850 m., Buchtien 373 (US). Greater Antilles to Bolivia. 2. V. heterandra (E. André) L. B. Smith in Contrib. U. S. Nat. Herb. 29: 443 (1951). Flowering to 35 cm. high. Leaves 3-4 dm. long, covered with appressed cinereous brown-centered scales, the sheaths large, dark castaneous, the blades narrowly triangular, 15-20 mm. wide, Scape curved. Scape-bracts imbricate, broadly ovate with long subsetaceous blades, densely lepidote. Inflorescence bipinnate, 15-20 em. long. Primary bracts erect, like the scape-bracts but the upper ones merely apiculate, about equaling the sterile bases of the spikes or shorter. Spikes linear-lanceolate, acute, spreading and 4-7 cm. long above the sterile base, dense, complanate. Floral bracts ovate, acute, 13 mm. long, exceeding the sepals, lepidote. Sepals elliptic- oblong, acute, glabrous. Petals 18 mm. long, pale rose, the blades crenulate. Stamens included. LA PAZ: LARECAJA: Hacienda Casana, Tipuani, 1400 m., Buchtien 7184 (US). Peru to Colombia. 3. V. heliconioides (HBK.) Walp. Ann. Bot. 3: 623 (1852). V. disticha O. Ktze. Rev. Gen, 3(2) : 304 (1898), as to material cited not as to basionym. Guzmania obtusa Rusby of Catalogue. Rarely flower- ing over 4 dm. high. Leaves about 2 dm. long, the blades ligulate, acute or attenuate, 15-30 mm. wide, subglabrous, Scape erect, usually much shorter than the leaves. Scape-bracts densely imbricate, very broadly ovate, acute. Inflorescence simple, distichous- or rarely poly- stichous-flowered, oblong, dense, 20 em. long, 6 em. wide, 6-18-flowered. Floral bracts very broadly ovate and triangular-acute or subrhombic, to 45 mm. long, chartaceous, the base suberect, the middle spreading horizontally and the apex incurved and sharply carinate, bright red above the middle, greenish yellow at the apex. Flowers suberect. Sepals lanceolate, attenuate, 27 mm. long, thin. Petals ligulate, acute, white. Stamens included. BENI: YACUMA: Rurrenabaque, White 1811 (NY; type of Guzmania obtusa Rusby); SANTA CRUZ: ICHILO: Yapa- cani, Kuntze (NY). Southwestern Brazil to Colombia and Central America. 4. V. Maxoniana (L. B. Smith) L. B. Smith in Phytologia, 6: 441 (1959). Tillandsia Maxoniana L. B. Smith of Catalogue. V. icterica Castellanos of Catalogue. Flowering 2-5 dm. high. Leaves 35-55 cm. long, green and lustrous, the sheaths elliptic, large, the blades arch- ing, ligulate, acute, 20-25 mm. wide, flat. Scape erect, slender, Scape- bracts barely imbricate, elliptic, attenuate or apiculate. Inflorescence simple, laxly few-flowered, 11-20 cm. long, strongly complanate, gla- brous. Floral bracts divergent and not imbricate at anthesis, elliptic, acute, 45-55 mm. long, about equaling the sepals, carinate, incurved, 1969] Bromeliaceae — Smith 267 subcoriaceous, greenish yellow. Sepals narrowly elliptic, acute. Petals ligulate, flavous. Stamens exserted. LA PAZ: NOR YUNGAS: Polo-Polo, Coroico, 1100 m., Buchtien 4538 (US; type); SANTA CRUZ: FLORIDA: Samaipata, 1500 m., Steinbach 8221 (GH). Argentina. 9. Guzmania R. & P. Stemless herbs, mostly epiphytic. Leaves entire, the sheaths usually conspicuous. Inflorescence simple or compound, the spikes always polystichous-flowered. Sepals usually somewhat connate. Petals close- ly agglutinated but not truly connate, naked, yellow or white. Stamens included, the filaments more or less agglutinated to the petals. Ovules many. Capsule septicidal. Seeds with a long basal straight usually brownish coma. a. Inflorescence compound, its spikes lax to subdense but the rhachis always visible: dub t... U. U GS Pusta, 1. G. Roezli. a. Inflorescence simple, dense, the rhachis wholly covered. b. Outer bracts of the inflorescence forming a cyathiform involucre that exceeds and conceals the large flowers. ........ 2. G. lingulata. b. Outer bracts of the inflorescence never concealing more than the lowest flowers. c. Leaf-blades obtuse to attenuate, apiculate but never retuse. d. Floral bracts densely brown-lepidote; sepals attenuate. ........ nuc METRE DD Pta 3. G. calothyrsa. d. Floral bracts sparsely and obscurely lepidote; sepals broad- Iyi rounded sent n amo tetera 4. G. Melinonii. c. Leaf-blades broadly rounded and retuse. ............ 5. G. retusa. 1. Guzmania Roezlii (E. Morr.) Mez in DC. Monog. Phan. 9: 948 (1896). Flowering 6-10 dm. high. Leaves 4 dm. long, the sheaths broadly ovate, finely brown-lepidote, concolorous or purple-striped, the blades ligulate, rounded and apiculate, 3-4 cm. wide, glabrous above. Seape erect, slender, exceeding the leaves, glabrous. Scape-bracts ovate, shorter than the internodes. Inflorescence laxly few-branched, glabrous, Primary bracts like the scape-bracts, enfolding the sterile bases of the spikes. Spikes spreading, elongate with long sterile prophyllate bases, lax or subdense but the rhachis always visible. Floral bracts ovate, narrowly obtuse, shorter than the sepals, green. Flowers suberect to spreading, 23 mm. long. Sepals elliptic, obtuse, 16-18 mm. long, connate for 2 mm., coriaceous. Corolla white, its lobes recurved-spreading. LA PAZ: LARECAJA: San Carlos, Mapiri, 650 m., Buchtien 1272 (US); SUR YUNGAS: Rio Bopi Basin, San Barto- lome, Calisaya, 750-900 m., Krukoff 10516 (GH, MO, NY). Peru to Costa Rica and Guiana. 2. G. lingulata (L.) Mez in DC. Monog. Phan. 9: 899 (1896). Rather variable in habit, flowering about 3 dm. high. Leaves 30-45 em. long, the sheaths ovate, conspicuous, sometimes violet-striped, the blades 268 Rhodora [Vol. 71 ligulate, acute, obscurely punctulate-lepidote. Scape-bracts densely imbricate, the lower ones strict, foliaceous, the uppermost lanceolate, usually red, forming a more or less spreading involucre. Inflorescence to 7 em. wide, 10-50-flowered. Floral bracts linear, shorter than the flowers. Flowers erect, 45 mm. long. Sepals linear, obtuse, glabrous. Petals linear, cucullate, white at the apex. Stamens highly aggluti- nated with the petals. Coma reddish brown. LA PAZ: LARECAJA: San Carlos, Mapiri, 850 m., Buchtien 366 (GH, US); NoR YUNGAS: Polo- Polo, Coroico, 1100 m., Buchtien 3670 (US); 3673 (NY, US). South- western Brazil to Central America and the West Indies. 3. G. calothyrsa Mez in DC. Monog. Phan. 9: 910 (1896). Flowering to 1 m. high. Leaves 3-8 dm. long, densely punctulate-lepidote be- neath, the blades linear, attenuate, 2-4 em. broad. Scape stout, erect. Scape-braets densely imbricate, lanceolate, attenuate, ferruginous- lepidote. Inflorescence strobilate, fertile throughout, 6-18 cm. long. Floral bracts broadly elliptic, acute, membranaceous, densely brown- lepidote. Sepals narrowly elliptic, attenuate, 22 mm. long, connate for 6 mm., thin, lepidote becoming glabrous. Petals obtuse. LA PAZ: LARECAJA: San Carlos, Mapiri, 750 m., Buchtien 1265 (US); SUR YUNGAS: Rio Bopi Basin, San Bartolome, Calisaya, 750-900 m., Kru- koff 10488 (GH, NY). Peru, Colombia. 4. G. Melinonis Regel in Gartenfl. 34: 116 (1885). Flowering 20-23 em. high. Leaves 3-5 dm. long, the blades ligulate, broadly acute or obtuse and apiculate, 2-4 em. wide, soon glabrous. Scape straight, glabrous. Scape-bracts imbricate, strict, broadly ovate, acute, red. Inflorescence ellipsoid or subeylindric, 5-10 cm. long. Floral bracts about 4-ranked, broadly ovate, nearly equaling the flowers, sparsely and obscurely lepidote, nerved, bright red. Flowers to 33 mm. long. Sepals broadly elliptic, obtuse, 15 mm. long, short-connate, firm, even, glabrous. Petals obtuse, yellow or white. Coma white. LA PAZ: LARECAJA: Mapiri, 3000 m., Bang 1582 (GH). COCHABAMBA: CHAPARE: Antahuacana, Espiritu Santo, 750 m., Buchtien 4539 (US); Puerto Polonia, Rio Coni, San Antonio, 395 m., Cárdenas & Cutler 7362 (GH). Ecuador, Colombia, French Guiana. 5. G. retusa L. B. Smith in Fieldiana Bot. 28: 148 (1951). Leaves 5 dm, long, exceeding the inflorescence, minutely and obscurely lepi- dote, the blades ligulate, broadly rounded and retuse, 6 cm. wide. Scape erect, slender. Scape-bracts densely imbricate, elliptic, apicu- late. Inflorescence simple, densely ellipsoid, 4.5 cm. long, 2 cm. wide, fertile throughout. Floral bracts orbicular, 2 cm. long, exceeding the sepals, coriaceous, lustrous, punctulate-lepidote, nerved or nearly even. Sepals elliptic, 13 mm. long, equally connate for 5 mm., cari- nate, coriaceous, sparsely lepidote at apex. Petals 23 mm. long, the free lobes broadly elliptic. Stamens included. COCHABAMBA: CHAPARE: Locotal, 1600 m., Steinbach 9623 (GH, US). Colombia, Venezuela. 1969] Bromeliaceae — Smith 269 10. Streptocalyx Beer Stemless herbs. Leaves spinose-serrate. Scape elongate to almost none. Inflorescence central, bipinnate, dense. Flowers sessile. Sepals free or nearly so, asymmetric with a broad lateral wing. Petals free, narrow, naked. Stamens and pistil included. Pollen grains with 2 or 4 pores. Ovary farinose, the epigynous tube distinct. Fruit some- what fleshy. Seeds ellipsoid or ovoid, naked. Floral bracts ovate, about equaling the ovary; scape very short and hidden in the center of the rosette. ................ 1. S. Fuerstenbergit. Floral bracts reniform, much shorter than the ovary; scape evident. HERREN IPIE aet iis dt RENE 2. S. Poeppigiü. 1. Streptocalyx Fuerstenbergii (E. Morr. & Wittm.) E. Morr. in Belg. Hortic. 33: 16 (1883). Leaves 50-75 cm. long, the sheaths large, elliptic, covered with a brown membrane of fused scales, the blades ligulate, attenuate to a pungent apex, 2 cm. wide, laxly serrate with curved spines, sparsely lepidote. Scape short, stout. Scape-bracts densely imbricate, broadly ovate with narrowly triangular pungent blades, densely fimbriate-serrate, bright rose, pale-lepidote. Inflores- cence bipinnate, very dense, slenderly pyramidal, 3 dm. long, 1 dm. wide, white-farinose except the petals. Primary bracts like the scape- bracts but bladeless, apiculate, erect, slightly exceeding the spikes. Spikes 6 em. long, sublax, to 9-flowered. Floral bracts ovate, attenuate, about equaling the ovary, entire. Flowers distichous, divergent. Sepals free or nearly so, 15 mm. long exclusive of the terminal spine, 7-8 mm. wide, entire. Ovary ellipsoid, 8 mm. long, enlarged in fruit, the epigynous tube urceolate. BENI: YACUMA: Reyes, Rurrenabaque, Fleischmann 66 (S). Brazil. 2. S. Poeppigii Beer, Bromel. 141 (1857). Flowering 5-8 dm. high. Leaves to 16 dm. long, the sheaths elliptic, ample, dark castaneous, densely buff-lepidote, the blades linear, attenuate to a dark pungent cusp, 3-7 em. wide, laxly serrate with dark curved spines 1-4 mm. long, densely white-lepidote beneath, soon glabrous above. Scape curved, red. Scape-bracts imbricate, broadly ovate with a linear blade, fimbriate-serrate, bright rose. Inflorescence densely bipinnate, slenderly cylindric or pyramidal, 2-4 dm. long, 9-15 em. in diameter, white-farinose. Primary bracts like the scape-bracts but merely apicu- late, about equaling the spikes, divergent. Spikes 5-7 em. long, laxly 2-10-flowered. Floral bracts reniform, mucronulate, much shorter than the ovary. Sepals free, strongly asymmetric, 16-19 mm. long, mucronate. Petals bright purple. Ovary cylindric, the epigynous tube large, the placentae apical. LA PAZ: LARECAJA: Sarampiuni, San Carlos, Mapiri, 500 m., Buchtien 370 (US). Brazil, Peru, Colom- bia, Surinam. 11. Bromelia L. Coarse terrestrial herbs, spreading by rhizomes or offsets. Leaves rosulate, bearing large curved spines along the margins. Inflorescence 270 Rhodora [Vol. 71 scapose, always compound. Sepals free, unarmed. Petals without a definite claw, united into a tube by the filaments but their margins free, without appendages, fleshy. Stamens included, their filaments forming a tube of varying lengths according to the species. Berry succulent, relatively large. Seeds flattened, naked. a. Inflorescence very laxly paniculate, the branches to 18 cm. long. .... beccuucecececuueceecuuececuuccecuueeeesceueeeecccseeeeeseeeeeceesuueeeeesueeeea 1. B. Hieronymi. a. Inflorescence dense, the branches less than 5 em. long. b. Inflorescence cylindric, to 22 em. long; sepals becoming glabrous. :NEMMNRNNRNNRNRNNRNRMRWMWMMWMWMMWMWMWMWMKNKKNMNMHyY,MHAÀNEMMMMMMMMMMMM 2. B. Balansae. b. Inflorescence globose, 6 cm. in diameter; sepals densely and persistently lepidote, ................ eee 3. B. serra. 1. Bromelia Hieronymii Mez in Mart. Fl. Bras. 3(3): 199 (1891). Flowering 1 m. high, propagating by very stout rhizomes. Leaves 7 dm. long, the sheaths small, thick and forming a subglobose bulb, covered with a membrane of fused scales, the blades 30 mm. wide, mostly glabrous above, very closely pale-lepidote beneath. Scape rather slender. Scape-bracts subfoliaceous, exceeding the internodes. Inflorescence very laxly paniculate, glabrous, glaucous. Primary bracts subfoliaceous but small and much shorter than the branches. Branches suberect to spreading, 12-18 em. long, 7-13-flowered. Floral bracts triangular, 5-6 mm, long. Pedicels rather slender, 5-11 mm. long. Sepals free, ovate, obtuse, entire, 7 mm. long. Petals 15 mm. long, connate for 7 mm., lilac. Ovary ellipsoid, 20 mm. long. SANTA CRUZ: CORDILLERA: Oriente, Izogzog, 700 m., Cardenas 2755 (GH). Para- guay, Argentina. 2. B. Balansae Mez in Mart. Fl. Bras. 3(3): 191 (1891). Flowering to over 1 m. high, propagating by stolons. Leaves over 1 m. long, the blades attenuate, 25 mm. wide, glabrous above, closely pale-lepidote beneath, laxly serrate with antrorse spines. Scape erect, stout. Scape- bracts like the inner leaves, imbricate. Inflorescence densely cylindric, 18-22 cm. long, 7-9 em. in diameter, the axes densely white-lanuginose. Primary bracts large covering most of the flowers. Branches to 10- flowered. Floral bracts narrowly elliptic, 30 mm. long, equaling the centers of the sepals, subglabrous. Flowers subsessile, to 45 mm. long. Sepals elliptic, obtuse, to 20 mm. long, coriaceous, allutaceous, soon glabrous. Petals linear, obtuse, 25 mm. long, connate for 7-8 mm., deep violet with broad white margins, glabrous. Filaments more than 1/2 connate. Ovary clavate, trigonous. Fruit to 47 mm. long, fleshy, edible. Seeds many, flat, 5 mm. wide. BENI: PROVINCE [?]: Rio Guaporé, Santo Antonio de Manoelito, Black & Cordeiro 52-15327 (IAN). Brazil, Paraguay, Argentina. 3. B. serra Griseb. in Goett. Abh. 24: 328 (1879). Flowering to 4 dm. high, propagating by long scaly stolons. Leaves 1.5 m. long, the 1969] Bromeliaceae — Smith 271 blades attenuate, 4 cm. wide, laxly serrate with antrorse uncinate spines 5 mm. long, lepidote, green above, cinereous beneath. Scape short, stout, densely lepidote. Scape-bracts foliaceous, densely imbri- cate, bright red. Inflorescence densely globose, 6 cm. in diameter. Pri- mary bracts broadly ovate, 4 cm. long, covering most of the fasciculate flowers. Branches short, 5-9-flowered. Floral bracts ligulate, obtuse, cucullate, 30 mm. long, carinate, densely pale-lepidote. Pedicels 5-10 mm. long in fruit. Sepals oblong, cucullate, 15 mm. long, strongly carinate, white, lepidote. Petals elliptic, obtuse, connate for 5 mm. blue-purple with white base and margins. Filament-tube 3 mm. high. Ovary subcylindric, angled, 2 cm. long. Fruit ovoid, 40 mm. long. BENI: CERCADO MOJOS: Trinidad, Missiones Guarayos, 250 m., Werder- mann 2300 (S). SANTA CRUZ: VELASCO: 200 m., Kuntze (NY); COR- DILLERA: Oriente, Izogzog, 700 m., Cardenas 2756 (GH). Paraguay, Argentina. 12. Aechmea R. & P. Large or medium-sized herbs. Leaves rosulate, Scape conspicuous. Inflorescence simple or compound, very variable. Flowers sessile. Sepals mostly asymmetric with a large wing on the right. Petals free, bearing 2 scales. Second series of stamens more or less joined to the petals, the pollen grains with 2 or 4 pores. Ovules caudate or obtuse. Berry usually dry. Seeds small, rugose, dark. a. Inflorescence compound. b. Floral bracts neither decurrent nor forming pouches about the flowers. e. Floral bracts forming a sheath that completely encloses the base of the ovary; inflorescence densely ferruginous-lepidote. s T la a a aa ETE sa 1. Ae. Kuntzeana. e. Floral bracts not completely enclosing the base of the ovary, their margins failing to meet. d. Floral bracts equaling or exceeding the ovaries, navicular. e. Spike-rhachis excavated, ........................ 2. Ae. angustifolia. e. Spike-rhachis merely angled. ................ 3. Ae. brachyclada. d. Floral bracts distinctly shorter than the ovaries. f. Upper scape-bracts reflexed, massed beneath the inflores- UE, Serratë..................... u a. K ON 4. Ae. Castelnavii. f. Upper scape-bracts suberect, entire. .... 5. Ae. tocantina. b. Floral bracts decurrent and forming pouches around the flowers. 2000 SHORE dide RECTUS E E A uo i nes 6. Ae. distichantha. a. Inflorescence simple, strobilate, densely white-lanate; sepals un- (0. 00 csi APERIRE E 00 7. Ae. bromeliifolia. 1. Aechmea Kuntzeana Mez in DC. Monog. Phan. 9: 208 (1896). Leaf-blades ligulate, subrounded and terminating in a stout dark spine, nearly 9 cm. wide, serrate with black antrorse spines, densely lepidote beneath. Scape erect, stout, ferruginous-tomentose. Upper 272 Rhodora [Vol. 71 scape-bracts reflexed, elliptic, serrate with dark spines. Inflorescence bipinnate, subdensely cylindric, 20 cm. long, 5 em. wide, ferruginous- lepidote. Branches suberect to spreading, 2-flowered, very short, all fertile but especially the highest with the terminal bracts changed to elongate spines with pulvinate bases. Floral bracts strongly asym- metric, much wider than long, bearing a dark suberect spine 21 mm. long, enfolding the ovary, entire, the margins free from the rhachis. Flowers 25 mm. long without the petals. Sepals strongly asymmetric, free, minutely mucronate. Ovary 6 mm. long. BENI: CERCADO MOJOS: Trinidad, Missiones Guarayos, 250 m., Werdermann 2420 (MO); YACUMA: Rurrenabaque, 300 m., Rusby (Mulford Exped.) 1252 (NY). SANTA CRUZ: ICHILO: Yapacani, 400 m., Kuntze (NY; type). 2. Ae. angustifolia Poepp. & Endl. Nov. Gen. & Spec. 2: 43, t. 159 (1838). Ae. boliviana Rusby of Catalogue. Ae. inconspicua Harms of Catalogue. Flowering to 75 cm. high. Leaves 5-7 dm. long, the sheaths 1 dm. long, forming an ellipsoid tank, densely and finely brown-lepidote, the blades ligulate, 3-6 cm. wide, cinereous-lepidote, spinose-serrate, concolorous. Scape erect, becoming glabrous. Scape- bracts elliptic, acute, serrate, red. Inflorescence bipinnate, cylindric, dense or lax, fertile throughout, 15-36 cm. long. Lowest primary bracts like the scape-bracts, exceeding the spikes, the others abruptly re- duced to about the size of the floral bracts. Spikes spreading or re- flexed, distichously 10-flowered, 20-55 mm. long, the rhachis undulate, excavated. Floral bracts spreading, cymbiform, broadly ovate with a straight mucro, 5 mm. long, barely exceeding the ovary. Sepals asymmetric, mucronate, 4-5 mm. long, free. Petals linear, mucronate, 8-10 mm. long. Ovary ellipsoid, the placentae apical, the ovules cau- date. LA PAZ: LARECAJA: Mapiri, San Carlos, 850 m., Buchtien 368 (B; type of Ae. inconspicua Harms; NY); 374 (US); 750 m., Buch- tien 1264 (US); NOR YUNGAS: Polo-Polo, Coroico, 1100 m., Buchtien 4537 (US); SUR YUNGAS: Río Bopi Basin, San Bartolome, Calisaya, 750-900 m., Krukoff 10510 (NY); Krukoff 10511 (GH, NY); LARE- CAJA: Copacabana, Mapiri, 850-950 m., Krukoff 11368 (NY); CAUPOLI- CAN: Tumupasa, 540 m., Williams 525 (NY). WITHOUT LOCALITY: Bang s. n. (NY; type of Ae. boliviana Rusby). 3. Ae. brachyclada Baker, Handbk. Bromel. 37 (1889). Flowering 6 dm. high. Leaves 4 dm. long, the sheaths ovate-elliptic, forming a large ventricose tank, the blades erect, ligulate, attenuate, 30 mm. wide, densely serrate with spreading or recurving spines 5 mm. long, densely pale-lepidote. Scape erect, slender, about equaling the leaves, densely white-tomentellous. Scape-bracts lanceolate, acute, entire, erect, especially the lower ones shorter than the internodes. Inflores- cence bipinnate throughout, wholly fertile, cylindric, 20 cm. long, densely tomentellous. Upper primary bracts greatly reduced. Branches spreading, less than 2 em. long. Floral bracts spreading, spinose- 1969] Bromeliaceae — Smith 273 attenuate from a broadly ovate base, navicular, to 6 mm. long, scarcely if at all adnate to the rhachis. Flowers to 16 mm. long. Sepals strongly asymmetric, mucronate, to 4 mm. long, subfree. Petals acute. Ovary ellipsoid, to 4 mm. long, the epigynous tube distinct, urceolate, the placentae apical, the ovules long-caudate. SANTA CRUZ: Chiquitos: Weddell s. n. (P; type). 4. Ae. Castelnavii Baker, Handbk. Bromel. 39 (1889). Ae. Sprucei Mez in DC. Monog. Phan. 9: 226 (1896). Leaves unknown. Scape straight, stout, ferruginous-furfuraceous toward apex. Scape-bracts reflexed, narrowly ovate, attenuate to a pungent spine, to 20 cm. long, 45 mm. wide, densely serrate with spreading teeth, bright pur- ple, membranaceous. Inflorescence tripinnate, densely subcylindric, 40 cm. long, 75 mm. wide, densely furfuraceous. Lower primary bracts like the scape-bracts, the others greatly reduced. Branches spreading, wholly fertile, divided to apex, about 5-flowered. Floral bracts chiefly spinose from an elliptic base, to 5 mm. long. Flowers to 17 mm. long. Sepals strongly asymmetric, 6 mm. long with a 3 mm. terminal spine, nearly free. Petals rounded, reddish when dry. Stamens included. Ovary slenderly ellipsoid, 8 mm. long, the epigynous tube very short, the placentae apical, the ovules caudate. SANTA CRUZ: PROVINCE [?]: Castelnau s. n. (P; type). Brazil, Venezuela, Colombia, Costa Rica. 5. Ae. tocantina Baker, Handbk. Bromel. 39 (1889). Flowering 1-2 m. high. Leaves to over 1 m. long, the sheaths broadly elliptic, large, the blades ligulate, subacute and attenuate to a stout spine, 4-10 cm. wide, repand-serrate with straight or slightly curved flat brown spines, glabrous above, covered beneath with minute pale-ferruginous scales. Scape erect, ferruginous-furfuraceous. Scape-bracts imbri- cate, lance-elliptie, acute, pungent, entire, especially the lower fer- ruginous-lepidote. Inflorescence subdensely cylindrie, bipinnate or at base tripinnate, 17-33 cm. long, 7-8 em. wide, ferruginous-tomentulose. Lower primary bracts like the scape-bracts, exceeding the branches, the others much smaller, the highest subfiliform, very short. Spikes spreading, laxly 4-8-flowered, to 4 cm. long, wholly fertile, the rhachis nearly straight. Floral bracts distichous, spreading, ovate, mucronate, entire, 5 mm. long, usually shorter than the ovary, nerved, the margins free from the rhachis. Flowers 15 mm. long. Sepals strongly asym- metric, short-mucronate, 5 mm. long, short-connate. Petal-scales fim- briate. Ovary ellipsoid, the epigynous tube short but distinct, the placentae apical, the ovules caudate. BENI: CERCADO MOJOS: Trinidad, Missiones Guarayos, 250 m., Werdermann 2464 (MO, S). SANTA CRUZ: CERCADO: El Palmar, 500 m., Cardenas 4649 (US); VALLEGRANDE: Los Monos, I. S. Nelson G-228 (US). Brazil, Venezuela, Surinam. 6. Ae. distichantha Lem. in Jard. Fleur. 3: t. 269 (1853). Flowering 3-7 dm. high. Leaves 3-10 (-15) dm. long, covered on both sides with a membrane of pale fused scales, the sheaths elliptic or oblong, to 274 Rhodora [Vol. 71 3 dm. long, usually much wider than the blades, the blades ligulate, typically with a broad apex, pungent, 25-80 mm. wide, finely to coarse- ly serrate. Scape erect, white-flocculose, Scape-bracts densely im- bricate, elliptic, attenuate, rose, mostly entire. Inflorescence bipinnate, typically lax or sublax and broadly pyramidal, rose and white-floc- culose except the petals. Primary bracts broadly ovate, apiculate, often shorter than the spikes. Spikes subsessile, typically spreading and dense, the lateral ones with 2-12 distichous flowers, the terminal with more and polystichous flowers. Floral bracts pouch-shaped, entire, truncate, mucronulate, about equaling the ovary, thin, nerved. Flowers suberect, 15-29 mm. long. Sepals asymmetric, oblong or subquadrate, mucronulate, 5-13 mm. long, free or short-connate. Petals obtuse, purple or blue, the scales serrate. Stamens included. Ovary short-cylindrie or obconic, 4-6 mm. long, the epigynous tube large, crateriform, the placentae subcentral, the ovules acute or ob- tuse. Not reported from Bolivia. Brazil, Argentina, Paraguay. 6a. Ae. distichantha var. Schlumbergeri E. Morr. ex Mez in Mart. Fl. Bras. 3(3) : 343 (1892). Ae. involucrata Rusby of Catalogue. Ae. involucrifera Mez of Catalogue. Leaves usually attenuate. Inflores- cence dense, elongate, slenderly cylindric or fusiform. Spikes erect, few-flowered. LA PAZ: SUR YUNGAS: Irupana, 1800 m., Cárdenas 4378 (US). COCHABAMBA: Province [?]: Monte Punco, Cárdenas 5588 (US). SANTA CRUZ: VELASCO: 200 m., Kuntze s. n. (NY); VALLE GRANDE: Samaipata, 1800 m., Cárdenas 3520 (GH); CHIQUITOS: Wed- dell 3512 (P); Santiago de Chiquitos, 700 m., Cutler 7026 (GH). WITHOUT LOCALITY: Bang s. n. (NY; type of Ae. involucrata Rusby and Ae, ?nvolucrifera Mez). Brazil, Paraguay, Argentina. 7. Ae. bromeliifolia (Rudge) Baker ex Benth. & Hook. f. Gen. Pl. 3: 664 (1883). Ae. ellipsoidea Rusby of Catalogue. Flowering 7-9 dm. high. Leaves 12-20 in a tubular rosette, 6-12 dm. long, covered on both sides with a membrane of white fused scales, the sheaths ovate to elliptic-oblong, 1-3 dm. long, the blades ligulate, very variable, at- tenuate to rounded and apiculate, 4-9 cm. wide, concolorous, laxly serrate with spines to 10 mm. long. Scape erect, stout, densely white- lanate. Scape-bracts lance-ovate, acute, entire, thin, the upper ones densely imbricate. Inflorescence densely spicate, ellipsoid or cylindric, fertile throughout, to 15 em. long, 3-4 em. wide, densely white-lanate with only the petals exposed at first. Floral bracts broader than long, truncate, thick, coriaceous, bicarinate, enfolding the ovary, much shorter than the sepals. Sepals suborbicular, emarginate, 7 mm. long, equally short-connate. Petals oblong, emarginate, 15 mm. long, green- ish yellow, soon turning black, the scales fimbriate. Ovary tomentose, the placentae apical, the ovules long-caudate. LA PAZ: SUR YUNGAS: basin of Rio Bopi, San Bartolomé, Calisaya, 750-900 m., Krukoff 1969] Bromeliaceae — Smith 275 10248 (GH, NY); Asunto, 840 m., White 633 (NY; type of Ae. el- lipsoidea Rusby). SANTA CRUZ: CHIQUITOS: Santiago, 640 m., Cutler 7035 (GH). Argentina to Central America. 13. Billbergia Thunb. Stemless herbs. Leaves rosulate or fasciculate, the sheaths large, the blades ligulate. Scape erect or arching. Scape-bracts ample, thin. Inflorescence simple, spicate. Sepals unarmed. Petals free, spirally recurved at anthesis (subgenus Helicodea), bearing 2 scales at base, the claw elongate, the blade narrow. Stamens exserted at anthesis, free from the petals or nearly so, the pollen grains with longitudinal folds when dry but no pores. Epigynous tube evident, the ovules many, a. Sepals narrow, 15-24 mm. long. b. Floral bracts shorter than the ovary; sepals 20-24 mm. long. .... dandi d d MR TRE 1. B. microlepis. b. Floral bracts exceeding the ovary and the lowest exceeding the sepals; sepals 15-20 mm. long. .................................... 2. B. Meyeri. a. Sepals broad, 10-12 mm. long; floral bracts shorter than the ovary. c. Ovary slightly or not at all sulcate, wholly white-farinose; floral bracts minute or wanting; petals green. .... 3. B. decora. c. Ovary sulcate with the ridges usually glabrous and dark; floral bracts evident. d. Petals clear yellow throughout, ................ 4. B. Cardenasii. I Peas bIUA Lus rennen enm eo pna Edna 5. B. velascana. l. Billbergia microlepis L. B. Smith in Phytologia 8: 504, t. 3, figs. 3, 4 (1963). Leaves 50-75 cm. long, densely appressed-lepidote on both sides, white-banded beneath; the blades ligulate, subacute, apiculate, 5-7 em. wide, laxly serrate with flat black spreading teeth 1.5-5 mm. long. Scape slender, decurved, finely pale-lepidote. Scape- bracts elliptic, acute, 13 cm. long. Inflorescence lax, 15 em. long, the rhachis densely pale-lepidote. Floral bracts broadly ovate, apiculate, the lowest almost equaling the ovary, the others much shorter. Flowers sessile. Sepals subequal, lance-oblong, attenuate, 20-24 mm. long, socn glabrous except at base. Petals over 14 cm. long, greenish when dry. Ovary ellipsoid, sulcate with dark lines against the white in- dument, 9 mm. high exclusive of the 3 mm. short-cylindric epigynous tube. LA PAZ: LARECAJA: Copacabana, Mapiri, 850-950 m., Krukoff 11158 (GH; type). 2. B. Meyeri Mez in Engler, Bot. Jahrb. 30: 148 (1902). Leaves few in a long tubular rosette, 40-75 cm. long, pale-lepidote and white- banded beneath, the sheaths very large, narrowly elliptic, the blades acute, 35 mm. wide, sublaxly serrate with antrorse curved spines 1-2 mm. long. Scape decurved, slender, white-farinose, Scape-bracts large, lanceolate, attenuate, the lower erect and about equaling the 276 Rhodora [Vol. 71 Figure 3. A: Billbergia Meyeri (after I. S. Nelson G-35), habit from photo; B: inflorescence X 1; C: sepal X 1; D: section of ovary X 1. 1969] Bromeliaceae — Smith Zit internodes, the upper divergent and massed beneath the inflorescence. Inflorescence sublax, 7-11 cm. long, densely white-farinose except for the petals, the rhachis straight. Floral bracts large, at least exceed- ing the ovary and the lowest ample and exceeding the sepals, ovate, acute, violet. Flowers divergent. Sepals free, equal, triangular, at- tenuate, 15-20 mm. long. Petals 45-50 mm. long, green or greenish white, the scales coarsely crenate. Ovary ellipsoid, the epigynous tube 4-5 mm. deep. SANTA CRUZ: CERCADO: Hacienda Sauce, Santa Cruz, I. S. Nelson G-35 (US). Brazil. Fig. 3. 3. B. decora Poepp. & Endl. Nov. Gen. & Spec. 2: 42, t. 157 (1838). B. boliviensis Baker, Handbk. Bromel. 81 (1889). Leaves few in a tubular or narrowly funnelform rosette, to 7 dm. long, white-banded beneath and sometimes yellow-spotted as well. Blades broadly acute or subrounded and apiculate, to 6 cm. wide, sublaxly serrate with stout antrorse spines. Scape decurved, more or less white-farinose. Scape-bracts lanceolate or elliptic, acute, entire, bright rose, the upper ones massed beneath the inflorescence. Inflorescence sublax, 10-15 em. long, densely white-farinose except the petals. Floral bracts minute or lacking. Flowers 9-10 em. long. Sepals suboblong or broadly ovate, broadly acute and apiculate, to 12 mm. long. Petals wholly green. Ovary ellipsoid, scarcely or not at all sulcate, without spots or dark glabrous lines, the epigynous tube short. LA PAZ: NOR YUN- GAS: Unduavi, 2400 m., Rusby 2853 (NY; type of B. boliviensis Baker). Brazil, Peru. 4. B. Cardenasii L. B. Smith in Phytologia, 4(6): 382 (1953). Leaf- blades broadly acute and apiculate, 4 dm. long, 35 mm. wide, covered on both sides with coarse cinereous appressed scales, laxly serrate with dark curved teeth 2 mm. long. Scape slender, densely and very finely white-flocculose. Scape-bracts lanceolate, acute, 13 cm. long, rose. Inflorescence densely many-flowered, the axis white-flocculose. Floral bracts ovate, acute, shorter than the ovary, membranaceous, glabrous except the base. Sepals equal, ovate, broadly acute, 10 mm. long, thin, nerved, orange, nearly glabrous. Petals 5 cm. long, yellow, the scales denticulate. Ovary ellipsoid, strongly trigonous, 10 mm. long, sulcate, white-flocculose, the epigynous tube small and narrow. COCHABAMBA: CHAPARE: cultivated, probably from Yungas del Cha- pare, Cárdenas 4907 (US; type). 5. B. velascana Cárdenas in Bromel. Soc. Bull. 7: 35 (1957). Flower- ing 6 dm. high. Leaf-blades abruptly acute, 6 dm. long, 2-5 em. wide, spinulose, gray-green, white-banded. Scape 4 dm. long, 8 mm. thick, densely flocculose. Scape-bracts lanceolate, 16-19 cm. long, rose. In- florescence many-flowered, the rhachis straight, white-flocculose. Floral bracts ovate, acute, 8 mm. long, membranaceous, rose. Flowers 6-8 cm. long. Sepals lance-ovate, acute, 12 mm. long, green toward base, 278 Rhodora [Vol. 71 violet toward apex, flocculose. Petals 5 cm. long, 6 mm. wide, blue. Ovary ovoid-conical, 12 mm. long, trigonous, sulcate, white-flocculose. SANTA CRUZ: VELASCO: San Ignacio, 500 m., Hammerschmidt 5486 in hb. Cárdenas (type). 14. Pseudananas Hassler ex Harms Pseudananas sagenarius (Arruda) Camargo in Rev. Agric. Piraci- caba, 14 (7, 8): reprint p. 4 (1939). Caulescent, flowering 1 m. high, propagating by long basal stolons. Stem leafy, straight, 30 em. high, 5 em. in diameter. Leaves 30-40, arching, to 12 dm. long, the blades linear, 6-7 em. wide, deeply channeled, coriaceous, reddish green and lustrous above, densely lepidote beneath, laxly serrate with curved antrorse and retrorse spines 10 mm. long. Scape erect, stout, 2-3 dm. high, furfuraceous. Scape-bracts foliaceous, reduced, reddish. In- florescence simple, strobilate, ovoid with the apical coma inconspicuous or lacking, 17 cm. long, 9 em. in diameter, 150-200-flowered. Floral bracts densely imbricate, lanceolate, attenuate, nearly equaling the petals, serrulate, red, cinereous-lepidote, persistent. Flowers connate by their ovaries and forming a syncarp. Sepals asymmetric, ovate, apiculate, 10-13 mm. long, conduplicate or merely channeled, thick with membranous margins, red orange, sparsely lepidote. Petals 4-5 cm. long, the blade lanceolate, purple, the claw bearing lateral folds that overlap the filaments, Stamens included. Syncarp 20 cm. long, succulent. SANTA CRUZ: CHIQUITOS: Puerto Sucre, near Brazilian frontier, Baker & Collins s. n. (GH). Brazil, Paraguay. f 15. Ananas Mill. Not producing stolons. Leaves densely rosulate, scarcely enlarged at base. Scape evident, erect. Inflorescence densely strobilate, crowned with a tuft of sterile foliaceous bracts, often producing slips at the base. Flowers sessile. Sepals free, obtuse, slightly asymmetric. Petals free, erect, violet or red, each bearing 2 slenderly funnelform scales. Stamens included. Ovaries coalescing with each other and with the bracts and axis to form a fleshy syncarp, the epigynous tube short, the placentae apical, the ovules caudate. Syncarp 15 em, long or usually much less, with scant unpalatable flesh at maturity; scape elongate, slender; seeds numerous. ........ ("RM 1. A. ananassoides. Syncarp well over 15 em. long at maturity with copious palatable flesh; scape stout and usually short; seeds lacking or very rare. beeseeeesussseeeescececssreeresesreoresassssasssereeeerssssesssreeeeessssroeooerreeoesseee 2. A. comosus. 1. Ananas ananassoides (Baker) L. B. Smith in Bot. Mus. Leafl. Harvard Univ. 7: 79 (1939). Flowering over 1 m. high. Leaves many, more than 16 dm. long, the sheaths densely ferruginous-lepidote, the blades linear, long-attenuate, rarely over 35 mm. wide, subdensely serrate with antrorse spines, green, minutely white-lepidote. Scape 1969] Bromeliaceae — Smith 279 slender, soon glabrous. Scape-bracts large, subfoliaceous. Inflores- cence subcylindric, the apical coma small at anthesis but becoming greatly enlarged. Floral bracts broadly elliptic, attenuate, 15-20 mm. long, exceeding the sepals, pungent, serrulate. Sepals broadly elliptic, 7 mm. long. Petals 18-17 mm. long. Fruit slightly enlarged after anthesis, scarcely edible, readily breaking from the scape. Seeds numerous. SANTA CRUZ: CHIQUITOS: Robore, Oriente, 250 m., Cár- denas 2929 (GH). Paraguay, Brazil, Venezuela, Colombia. 2. A. comosus (L.) Merrill, Interpr. Rumph. Amboin. 133 (1917). Leaves coarsely and laxly spinose-serrate, Scape short, stout. Scape- bracts foliaceous, serrate. Inflorescence large, many-flowered. Floral bracts soon exposing the apices of the ovaries, relatively inconspic- uous, weakly serrulate or entire. Syncarp well over 15 cm. long at maturity with copious palatable flesh. Seeds lacking or very rare. Tropical lowlands in cultivation but no herbarium specimens. See Encyclopedia Britannica, ed. 14, 3: 815 (1929). SMITHSONIAN INSTITUTION WASHINGTON, DISTRICT OF COLUMBIA 20560 CHROMOSOME COUNTS OF KALMIA SPECIES AND REVALUATION OF K. POLIFOLIA VAR. MICROPHYLLA RICHARD A. JAYNES Kalmia L. is a small North American genus of woody shrubs belonging to the Ericaceae. There are 7-10 species in the genus depending on the authority. Two taxa, K. poli- folia var. polifolia Wangenh.' and K. p. var. microphylla (Hook.) Rehd. are morphologically very similar and thus have often been placed in the same species. Although these two taxa can be readily crossed (Jaynes 1968), other evi- dence obtained in the course of genetic studies suggests that var. microphylla should be treated as a separate species (K. microphylla) as originally done by Heller (1898). In this paper their taxonomic status is considered in the light of cytological and other evidence. Chromosome counts of four other species of Kalmia are also reported for the first time. Previously, Hagerup (1928) reported n = 12 for Kalmia latifolia and n = 24 for K. polifolia (as K. glauca), counts which are verified by this report. Callan (1941), however, reported 2n — 44 for K. polifolia, which appears to be in- correct. The counts given here were made during meiosis, usually at metaphase I, on pollen mother cells using an aceto-car- mine smear technique. Buds were fixed in ethyl alcohol (95%) and glacial acetic acid (3:1) for at least 24 hours, then rinsed and stored in 70% ethyl alcohol at 4C. Stained preparations were photographed under phase contrast. The list of species with geographic sources and the counts obtained are presented in Table 1. One or two plants per source were used and a minimum of four figures were ex- ‘Synonym K. p. var. rosmarinifolia (Pursh) Rehd. There is ap- parently no evidence to substantiate the existence of a var. rosmarini- folia as well as a var. polifolia (Wood 1961) as is suggested by the use of the name K. p. var. rosmarinifolia. 280 1969] Kalmia — Jaynes 281 Plate 1414. Figs. 1-5. Chromosomes of Kalmia species, MI, sources in paren- thesis. Fig. 1. K. cuneata (North Carolina) n = 12. Fig. 2. K. hirsuta (Alabama) n = 12. Fig. 3. K. microphylla (Idaho) n = 12. Fig. 4. K. polifolia (Connecticut) n = 24. Fig. 5. K. polifolia X K. microphylla (Nova Scotia X Washington) 14 II, 8 I: m congruent to 36. Magnif. X 2,600. 282 Rhodora [Vol. 71 amined to verify the chromosome number. The haploid chromosome number of all the species is 12, with the excep- tion of Kalmia polifolia (n = 24). Photomicrographs (Fig- ures 1-5) verify the counts reported and illustrate the small size of the chromosomes. A summary of the evidence suggesting that Kalmia micro- phylla should be treated as a species distinct from K. poli- folia follows: 1) The two taxa are geographically separated, K. microphylla occurring in western North America at high elevations and K. polifolia occurring to the east of the Rocky Mountains and at lower elevations, 2) The plants are dif- ferent physiologically: K. microphylla is a low growing plant adapted to a very short growing season as might be ex- pected of an alpine species, When cultivated under Connecti- cut conditions it flowers about May 1 and the pollen mother cells undergo meiosis about mid-June, approximately two weeks before K. polifolia. 3) The two taxa have completely different crossing relationships with other species in the genus (Jaynes 1968). Kalmia microphylla forms hybrids only with K. polifolia, whereas the latter taxon can, with difficulty, form hybrids with three other species: K. angusti- folia, K. hirsuta, K. latifolia. 4) The chromosome numbers of the two taxa are different (Table 1). Metaphase I chromosome counts of F, hybrids of K. polifolia-K. micro- phylla using plants of the following sources: Oregon X Sas- katchewan, Oregon X Maine, Nova Scotia X Washington, and New Hampshire X Washington indicate that the hy- brids are triploids with 2x = 36, but the exact number was difficult to verify because of univalents, bivalents, and sec- ondary associations, Figure 5 is atypically clear in this regard, but even here there is some uncertainity in dis- tinguishing bivalents from univalents. As would be ex- pected, all the hybrids between the two taxa are highly sterile. Pollen fertility as determined by acetocarmine stain- ing was 0-15% for the hybrids as compared to 90% or higher for the parental taxa. On the basis of all the pre- ceding information it is concluded that K. microphylla is a valid species and not a variety of K. polifolia. 1969] Kalmia — Jaynes 283 Table 1. Chromosome numbers and geographic source of Kalmia species. Species Geographic source n K. angustifolia L. North Carolina 12 K. cuneata Michx. North Carolina 12 K. hirsuta Walt. Alabama 12 K. latifolia L. Connecticut 12 K. polifolia Wangenh. Connecticut 24 " Minnesota 24 n New Hampshire 24 K. microphylla (Hook.) Heller Idaho 12 di Oregon 12 él Washington 12 Voucher specimens of the material examined have been placed in the herbarium of the Connecticut Agricultural Ex- periment Station. Attempts to produce a tetraploid Kalmia microphylla with colchicine for the purposes of comparing it to K. poli- folia have been unsuccessful. But it seems highly unlikely that a doubling of the chromosome number of K. micro- phylla would produce a plant with the longer internodes, rangy habit, and other characteristics of K. polifolia. The presence of normal bivalents during meiosis in K. polifolia and the apparent lack of trivalents or secondary associa- tions are additional reasons to believe that this species is not an autotetraploid of K. microphylla. It also appears unlikely that K. polifolia is an allotetraploid of K. micro- phylla and one of the other presently known Kalmia species since K. microphylla is unable to cross with the other spe- cies. However, K. polifolia may have originated as an allo- tetraploid from a hybrid between K. microphylla and an ancestral, presently non-existent, species of Kalmia. 284 Rhodora [Vol. 71 SUMMARY Kalmia, polifolia and K. microphylla have commonly been treated as varieties of K. polifolia but from cytological, physiological, geographic, and genetic evidence presented in this report it is concluded that they are valid species. The chromosome number of K. polifolia is verified as n = 24. K. microphylla, K. angustifolia, K. cuneata, and K. hirsuta are reported for the first time, » — 12, and K. latifolia is verified as n — 12. GENETICS DEPARTMENT CONNECTICUT AGRICULTURAL EXPERIMENT STATION NEW HAVEN 06504 REFERENCES CALLAN, H. G. 1941. The cytology of Gaulthettya wisleyensis (Mar- chant) Rehder. A new mode of species formation. Annals of Botany, N.S. 5: 579-585. HAGERUP, O. 1928. Morphological and cytological studies of bicornes VI. Dansk Bol. Ark. 5: 1-26. HELLER, A. A. 1898. New and interesting plants from western North America. III Bull. Torrey Botan. Club 25: 580-582. JAYNES, R. A. 1968. Interspecific crosses of Kalmia. Amer. J. Bot. 55: 1120-1125. Woop, C. E., JR., 1961. The genera of Ericaceae in the southeastern United States. J. Arnold Arb. 41: 10-80. CYTOTAXONOMIC NOTES ON THE SPECIES OF CIRSIUM NATIVE TO THE SOUTHEASTERN UNITED STATES GERALD B. OWNBEY AND WILLIS A. OLSON’ All of the southern thistles are classified under the sub- genus Eucirsium, section Onotrophe of Petrak (1917). Fol- lowing a modified version of Petrak’s classification, the chromosomal relationships of the southern species may be summarized as follows: Subgenus EUCIRSIUM, Section ONOTROPHE Subsect. ODORATA: 2n = 30, 32, 34, (33, 35). C. pum- ilum, C. repandum, C. lecontei, C. horridulum, C. vittatum. Subsect. CAMPANULATA: 2n = 24, (26). C. nuttallii Subsect. ACANTHOPHYLLA : Series ALTISSIMA (including series Mutica) : 2n — 18, 20, (21, 22, 23, 30, 31). C. altissimum, C. terrae-nigrae, C. discolor, C. muticum, C. carolinia- num. Series PANICULATA: 2n — 22, (23, 24). C. texanum. Series UNDULATA: 2n = 26, (22-34). C. undulatum. Series VIRGINIANA: 2n = 28. C. virginianum. We know of no prior published records of chromosome counts for the following species: C. repandum Michx., C. lecontei T. & G., C. vittatum Small, C. nuttallii DC., C. ter- rae-nigrae Shinners, C. carolinianum | (Walt.) Fern. & Schub., C. texanum Buckl. and C. virginianum (L.) Michx. Much of the material used in these studies was obtained from plants brought in from the wild and subsequently grown in the greenhouse over a period of several months. The remainder came from seedlings obtained from seeds, also from the wild, some of which were grown to maturity. "This study was supported by National Science Foundation grants G9071 and GB2727 made to the senior author. We wish to thank Dr. Lloyd H. Shinners, Dr. John W. Thieret, Mr. Fred B. Jones and Mr. Gary R. Baker for assistance in collecting seeds and living plants. We are indebted to Dr. Yu-tseng Hsi and Mrs. Siu-tsun Hsi for many excellent cytological preparations. 285 286 Rhodora [Vol. 71 Root tips were obtained and subsequently handled as de- scribed by Ownbey & Hsi (1963) and by Ownbey (1968). Preparations for observations of meiotic stages were made according to procedures described by Ownbey, l.c. Voucher specimens for nearly all of the chromosomal records have been deposited in the herbarium of the University of Min- nesota. Sect. ONOTROPHE, Subsect. ODORATA The Odorata as circumscribed here includes three species groups of equivalent taxonomic rank, The first group is formed of C. pumilum and C. repandum, the second of C. lecontei and the third of C. horridulum and C. vittatum. We have transferred C. repandum and C. lecontei from Petrak's subsect. Campanulata of the sect. Onotrophe as we believe this reflects their actual genetic relationships. 1. Cirsium pumilum (Nutt.) Spreng. subsp. pumilum. 2n = 30 NORTH CAROLINA. CHATHAM CO.: Route 64, 4 miles east of the Chatham County courthouse, Pittsboro, Olson & Ownbey 836, (2 plants). Chromosomal information for this species is consistent with that from earlier reports by Ownbey & Hsi (1963) and Frankton & Moore (1966) based upon specimens from Connecticut, Massachusetts and New York. Our specimens Figs. 1-18. Chromosomes of Cirsium drawn with camera lucida, all X 1000. Fig. 1. C. altissimum, O. & O. 802, n — 9. Fig. 2. C. altis- simwm, O. & O. 802, 2n — 18. Fig. 3. C. carolinianum, O. & O. 191, n = 10. Fig. 4. C. carolinianum, O. & O. 797, 2n — 20. Fig. 5. C. discolor, O. & O. 837, 2n — 20. Fig. 6. C. horridulum, O., O. & Thieret 798, n — 16. Fig. 7. C. horridulum, O., O. & Thieret 798, 2n — 32. Fig. 8. C. lecontei, O. & O. 807, 2n — 32. Fig. 9. C. muticum, O., O. & Thieret 799, pl. 1, 2n — 20. Fig. 10. C. muticum, O., O. & Thieret 799, pl. 6, » — 10 II + 1 I. Fig. 11. C. muticum, O., O. & Thieret 799, pl. 6, 2n = 21. Fig. 12. C. muticum, O., O. & Thieret 799, pl. 6, 2n — 22. Fig. 13. C. muticum, O., O. & Thieret 799, pl. 3, 2n — 23. Fig. 14. C. nuttallii, O. & O. 810, 2n — 24. Fig. 15. C. nuttallii, O. & O. 825, 2n — 26. Fig. 16. C. pumilum, O. & O. 836, 2n — 30. Fig. 17. C. repandum, O. & O. 828, 2n — 30. Fig. 18. C. terrae- nigrae, O. & O. 791, 2n — 18. 1969] Cirsium — Ownbey and Olson D $ wee ç LJ [] . NET umv & . ° f 9 `. e J ee ^? ` a 9 r] ` eo” ^" 10 b ° a ` 9 dd N 14 "e 7. ⁄ ` pee ge `` hp 2 1` . Pd . `. 13 4 ‘ .. 4 .... °. ` k - * 4 ae ` is . = - 2 S . ~ itle . - - Nut Š 16 * 287 288 Rhodora [Vol. 71 from North Carolina indicate that the sporophytic number of 30 is constant for the species throughout its range. One or two additional chromosomes may be present in hybrids of C. pumilum with plants of C. horridulum having a sporo- phytic number of 32, as demonstrated by Moore & Frankton, le. Hybrids may appear spontaneously where the two spe- cies occur together. 2. Cirsium repandum Michx. 2n = 30 NORTH CAROLINA. WAYNE CO.: 7 miles southeast of Goldsboro turnoff, Route 70, Olson & Ownbey 835, (2 plants). SOUTH CAROLINA. BERKELEY CO.: 7.3 miles northeast of Summer- ville, Route 17A, Olson & Ownbey 828, (1 plant); 7.8 miles northeast of McClellanville Post Office, Route 17, Olson & Ownbey 831, (1 plant). Although our data for this species are minimal, the sporophytic chromosome number is probably 30 throughout its range. It is closely related to C. pumilum, having the same chromosome number and similar morphological fea- tures. 3. Cirsium lecontei T. & G. 2n = 32 FLORIDA. WALTON CO.: 1.0 mile west of Santa Rosa Beach, Route 98, Olson & Ownbey 807, (1 plant). MISSISSIPPI. HANCOCK CO.: 7.2 miles west of Waveland, Route 43 (1 mile west of junction of routes 90 and 43), Olson & Ownbey 804, (1 plant). Our limited chromosomal data indicate that this species belongs in the subsect. Odorata. It is a distinctive species of wetland habitats and in this regard differs from other members of the subsection, 4. Cirsium horridulum Michx. 2» — 32 (33, 34) FLORIDA. FRANKLIN CO.: 1 mile west of Apalachicola, Olson & Own- bey 808, 2n — 32 (1 plant), 33, 34 (1 plant). GEORGIA. GLYNN CO.: just north of Brunswick along Route 17, Olson & Ownbey 824, (1 plant). LOUISIANA. BATON ROUGE PARISH: just south of Port Allen, Ownbey 3621, (3 plants). LAFAYETTE PARISH: northern environs of Lafayette, Olson, Ownbey & Thieret 198, n — 16, (1 plant), 2n — 32 (3 plants). MISSISSIPPI, FORREST CO.: 3.2 miles north of Maxie, Route 49, Olson & Ownbey 806, (1 plant). NORTH CAROLINA. ONSLOW CO.: 7.6 miles north of Folkstone, Olson & Ownbey 834, (3 plants). 1969] Cirsium — Ownbey and Olson 289 TEXAS. HARDIN CO.: just north of Silsbee, Ownbey 3623, (2 plants). PANOLA-RUSK CO, line: 8.4 miles northeast of Mt. Enterprise (9.9 miles southwest of Clayton), Ownbey 3625, (2 plants). This species had a chromosome number of 2n = 32 in all collections examined during this study with a single excep- tion from Franklin Co., Florida where the number was 33 or 34 in one plant. Lewis, Stripling & Ross (1962) reported counts of 2n = 34 for two collections, one from Franklin Co., and one from Leon Co., Florida. Moore & Frankton (1966) report 2n — 34 in one collection of this species from Massachusetts. They also report hybridization of this spe- cies with C. pumilum in Massachusetts. Cirsium horridulum is very closely related to C. vittatum and it is often difficult to distinguish them, especially as to herbarium specimens. In the field, C. horridulum appears to be a larger and coarser plant with predominantly ochro- leucous flowers which appear earlier in the spring. Popula- tions which we classified as C. vittatum consisted of more slender, branched plants with lavender flowers appearing later in the spring. C. horridulum is found from eastern Texas to New England in the coastal and adjacent states; C. vittatum is of more restricted distribution in Florida and the coastal areas of the southeastern states. Natural hybridization between C. horridulum and C. vit- tatum is to be expected and, in fact, the difficulties of separating the two may be due to this factor. Extensive analysis of this species complex is much needed. 5. Cirsium vittatum Small. 2n — 32, 34 (33, 35) FLORIDA. CITRUS CO.: 1.6 miles south of Crystal River, Route 19, Olson & Ownbey 812, 2n. — 32 (2 plants), 33 (1 plant). CLAY Co.: ca. 1.2 miles south of the junction of routes 218 and 21 (on 218) near Middleburg, Olson & Ownbey 823, n = 16 + 1 (1 plant), 2n = 33 (1 plant), 35 (1 plant). COLLIER CO.: just west of Paolita Sta., Route 41, Olson & Ownbey 814, 2n — 32 (1 plant), 34 (2 plants), 35 (1 plant). DADE CO.: just south of Florida City, Olson & Ownbey 815, 2n — 32 (3 plants), 34 (2 plants); about 1 mile southeast of Kendall, east of Route 1, Olson & Ownbey 816, 2n — 32 (1 plant). VOLUSIA CO.: 7.6 miles southeast of Pierson (2 miles southeast of Barberville), Olson & Ownbey 819, 2n — 32 (1 plant), 33 (2 plants), 34 (1 plant), 290 Rhodora [Vol. 71 34, 35 (1 plant). WAKULLA CO.: 1 mile west of Newport, Route 98, Olson & Ownbey 809, 2n = 32 (3 plants). SOUTH CAROLINA. GEORGETOWN CO.: southwestern outskirts of And- rews, Route 41, Olson & Ownbey 829, 2n = 32 (4 plants). DORCHESTER co.: northern environs of Summerville, Olson & Ownbey 827, n — 16 (1 plant), 2n — 32 (2 plants). This species is characterized by variation in the sporo- phytic chromosome number even within a single population. Normally, the number appears to be 32 or 34, but accessory chromosomes which are morphologically indistinguishable from those of the regular complement are sometimes pres- ent. In one instance the chromosome number in cells of a single root tip was not constant, being either 34 or 35. The differences between C. vittatum and C. smallii Britt. are inconsequential and we do not attempt to distinguish them. When combined, the older name, C. vittatum, must be adopted for the resulting taxon. For comments regard- ing the distinctions between C. vittatum and C. horridulum, see the latter species. Subsect. CAMPANULATA 6. Cirsium nuttallii DC. 2n = 24 (26) FLORIDA. HERNANDO CO.: 12.8 miles south of Homosassa Springs, Route 19, Olson & Ownbey 813, (4 plants). LEVY CO.: 1 mile south- east of Chiefland, Olson & Ownbey 811, (4 plants). OKEECHOBEE CO.: 1 mile south of Okeechobee, Route 441, Olson & Ownbey 817, (4 plants). PUTNAM CO.: 1.7 miles north of Palatka, Route 17, Olson & Ownbey 820, (4 plants). TAYLOR CO.: 1 mile southeast of Salem, Olson & Ownbey 810, (2 plants). GEORGIA. GLYNN CO.: ca. 10 miles north of Brunswick, Olson & Ownbey 825, 2n = 24 (2 plants), 26 (1 plant). SOUTH CAROLINA. HAMPTON CO.: 0.5 mile south of Hampton-Col- leton Co. line, Olson & Ownbey 826, (3 plants). With the exception of a single plant with 2» — 20, the chromosomal data for this species is very consistent. The species has no close relatives among the southern thistles. Subsect. ACANTHOPHYLLA, Series ALTISSIMA We believe that Petrak’s series Mutica of the Acantho- phylla should be combined with the series Altissima in order to more clearly portray the close chromosomal and morpho- logical relationships of this group of species. 1969] Cirsium — Ownbey and Olson 291 7. Cirsium altissimum (L.) Spreng. 2n — 18 MISSISSIPPI. ADAMS CO.: between Washington and Natchez, Olson & Ownbey 802, n = 9 (2 plants), 2n = 18 (3 plants). OKLAHOMA. LE FLORE CO.: 15.2 miles southeast of Hodgen (5.0 miles southeast of junction of routes 59 and 259), Ownbey 3627, (1 plant). TEXAS. CASS CO.: 1 mile south of Red Hill, Route 8, Ownbey 3626, (2 plants). The sporophytic chromosome number of 18 for the three collections reported here is consistent with earlier counts by Frankton & Moore (1963) and Ownbey & Hsi (1963). The only deviation from this figure was 2n = 20 reported in one Iowa collection by Ownbey & Hsi, Lc. All earlier counts came from plants from the northern part of the range of the species. The present counts are from the south- ern and southwestern periphery of the range. Plants from the Mississippi population were taller and with more deeply lobed cauline leaves than is usual in this species. 8. Cirsium terrae-nigrae Shinners (C. filipendulum En- gelm., not Lange). 2n = 18 TEXAS. ELLIS C0.: 2.0 miles southwest of Bardwell, Olson & Ownbey 793, (4 plants). KAUFMAN C0.: 1.4 miles northeast of Scurry, Route 34, Olson & Ownbey 791, (5 plants). This species is closely related to C. altissimum with which it shares the same chromosome number, 25 — 18, the lowest so far recorded for the genus Cirsium. Both species develop thickened, fusiform roots which are especially conspicuous during the rosette stage and tend to disappear during for- mation of the flowering stems. 9. Cirsium discolor (Muhl.) Spreng. 2n — 20 NORTH CAROLINA. ALLEGHENY CO.: along Blue Ridge Parkway west of Cherry Lane, alt. ca. 3600 ft., Olson & Ownbey 837, n = 10 (1 plant), 2n — 20 (1 plant). Chromosomal data for this species has been recorded by Ownbey (1951), Frankton & Moore (1963) and Ownbey & Hsi (1963). The sporophytic number of 20 has been found to be constant except for one seedling studied by Frankton & Moore, grown from Virginia seed, which had 21. The present collection was made in North Carolina at the south- eastern extremity of the species range. 292 Rhodora [Vol. 71 10. Cirsium muticum Michx. 2n — 20 (21, 22, 23) LOUISIANA. LAFAYETTE PARISH: southwestern environs of Lafayette, Olson, Ownbey & Thieret 199, n = 10 (1 plant), 10 + 1 (3 plants), 2n = 20 (2 plants), 21 (2 plants), 20, 21, 22 (1 plant), 23 (1 plant). Earlier reported counts for this species are as follows: Ownbey (1951) and Ownbey & Hsi (1963), 2n = 20; Frankton & Moore (1963), 2n — 20, 22. Frankton & Moore reported an apparent case of triploidy in Florida material. They studied five plants with 2n — 30 and one plant, which exhibited some morphological peculiarities, with 2n — 31. Preliminary studies of one plant indicated that 15 bodies, presumably bivalents, were regularly present at metaphase I. The additional chromosome in one plant was thought to be due to meiotic irregularity and not to be an accessory chromosome in the usual sense. Considerable instability in chromosome numbers was seen in plants of Olson & Ownbey 799. Pollen mother cells at metaphase showed 10 normally paired bivalents in all plants, but in three plants there was an additional smaller, un- paired chromosome. Root tip cells had 20-25 chromosomes in individual plants. In one plant the root tips differed in their chromosomal complements, having 20, 21 or 22 chro- mosomes in individual cases. Presumably this could result from mitotic misdivisions of the accessory chromosome. C. muticum from Louisiana differs in many small ways from the species as we are accustomed to see it in the northern wetlands. A more detailed experimental study of the south- ern races of this species would be valuable. 11. Cirsium carolinianum (Walt.) Fern. & Schub. 2n — 20 TEXAS. POLK CO.: 0.1 mile east of junction of Texas farm roads 942 and 2500, on Route 942 east of Leggett, Olson & Ownbey 797, n = 10 (2 plants), 2n — 20 (2 plants). More chromosomal data for this species is needed, but the data now available indicate that the species is correctly placed in the series Altissima. The plants studied came from the southwestern edge of the species range. 1969] Cirsium — Ownbey and Olson 293 ° ` v6 UNIV. -@ ¿ ' e 2 e ` P d á d ° ` oy z mr) > e ° ° .. M s^ A tos" e zi , "v. e D du ` 19 * 20 ~ , ° a - ee ° 22 « ^ a á = o um - Ld ` - *. ^. [| `. - - [| e °. ° ° . *? é ror ` (^ ec @ + ° = . > - ` 23 7 . ^ M e E b. - ad ih 24 br re L4 25 Te vu ME, - ee ats PLI ' ` ‘ -— E . 2 se ve * * ..... ° sus les ‘ ` v 7 te "a. E ; *. om ee,’ . e Ld - Su E e . e ` - ° 2 = 26 on. E. "ea 27 28 * ^ ` Figs. 19-28. Chromosomes of Cirsium drawn with camera lucida, all >x 1000. Fig. 19. C. texanum, O. & O. 792, n = 11 II + 1 I. Fig. 20. C. texanum, O. & O. 792, 2n = 23. Fig. 21. C. texanum, O. & O. 794, 2n = 22. Fig. 22. C. undulatum, O. & O. 795, 2n = 26. Fig. 23. C. virginianum, O. & O. 830, 2n — 28. Fig. 24. C. vittatum, O. & O. 827, n = 16. Fig. 25. C. vittatum, O. & O. 815, 2n = 32. Fig. 26. C. vittatum, O. & O. 812, 2n = 33. Fig. 27. C. vittatum, O. & O. 814, pl. 4, 2n — 35. Fig. 28. C. vittatum, O. & O. 814, pl. 2, 2n = 84. Series PANICULATA 12. Cirsium texanum Buckl. 2n = 22 (23, 24) TEXAS. BELL CO.: 3 miles south of Temple, Ownbey & Baker 2977, 2n — 22 (1 plant); 4.6 miles south of Temple city limits on the road to Academy, Ownbey & Baker 2978, 2n, = 22 (2 plants). DALLAS CO.: 294 Rhodora [Vol. 71 Route 114, 4 miles northwest of the crossing of the Elm Branch of the Trinity River, Ownbey & Baker 2972, 2n = 22 (2 plants). ELLIS co.: l.l miles southwest of Bardwell, Olson & Ownbey 792, n = 11 (2 plants), 11 + 1 (1 plant), 2n — 22 (1 plant), 23 (3 plants), 24 (1 plant). FALLS CO.: 2.0 miles east of Marlin, Olson & Ownbey 796, n — 11 (3 plants), 2» — 22 (2 plants). HILL CO.: Lake Whitney State Park, Olson & Ownbey 794, 2n — 22 (2 plants). MC CULLOCH CO.: 9 miles west of Brady city hall, Route 87, Ownbey & Baker 2988, n = 11 (2 plants), 2n = 22 (2 plants). PECOS CO.: Tunis Spring, 20 miles east of Ft. Stockton, Route 290, Ownbey & Baker 3001, 2n — 24 (1 plant). SAN PATRICIO CO.: ca. 2 miles southwest of Port- land, Jones s.n., n — 11 (2 plants), 2n — 22 (2 plants) ; 7 miles south of Taft, Jones s.n., n = 11 + 1 (2 plants), 2n = 23 (2 plants). TOM GREEN CO.: Concho River crossing, 2.4 miles northeast of Tankersly, Qwnbey & Baker 2996, 2n — 22 (6 plants), 24 (1 plant). Rather extensive sampling of this species indicates that the sporophytic chromosome number is normally 22, but sometimes is 23 or 24. No meiotic material of plants with 94 — 24 was studied so it is unknown whether 11 or 12 bivalents are formed at metaphase I. In plants with Ən = 22, 11 bivalents were seen at metaphase I; in plants with 2n — 23, 11 bivalents plus 1 univalent were ordinarily present. Jones s.n., from south of Taft, San Patricio Co., Texas, was seen to have 11 bivalents plus 1 univalent or 10 bivalents and 1 trivalent at metaphase I, indicating that the accessory chromosome in this collection is a homologue capable of undergoing normal pairing. Jones (personal communication) states that the two populations from San Patricio Co. were distinguishable in the field both on the basis of morphology and habitat. Possibly the extra chro- mosome in the one population produces a visible effect in this instance. C. texanum is, however, a variable species and there may actually be two distinguishable taxa lumped together under this binomial at the present time. The ques- tion is under study. Although a very common thistle in its natural area, C. texanum is not found outside Texas and adjacent Mexico. Ownbey & Baker 2988 was collected near the type locality of the species. Ownbey & Baker 3001 was collected near the type locality of Carduus austrinus Small which, judging from a study of the type specimens of the two, is a synonym 1969] Cirsium — Ownbey and Olson 295 of Cirsium texanum. Carduus austrinus is based upon Wright 1291 which, according to A. Gray (1853, p. 101), came from “Low and grassy flats around Escondido Springs, between the Pecos and the Limpio; June." Escondido Springs, according to Geiser (1935), is situated 19.5 miles east of Comanche Springs (now Fort Stockton). Petrak (1917, p. 419) erroneously lists Carduus austrinus Small as a synonym of Cirsium ochrocentrum Gray var. helleri (Small) Petrak. Petrak, (l.c., p. 447 et seq.) did not distinguish C. texa- num Buckl. from what later became C. terrae-nigrae Shin- ners (ie. Cirsium filipendulum Engelm. and synonyms), and the synonyms given by Petrak for C. texanum, with the exception of C. virginianum var. T. & G. and C. greenei Petrak, properly belong under C. terrae-nigrae. Series UNDULATA 13. Cirsium undulatum (Nutt.) Spreng. 2» — 26 TEXAS. HILL C0.: Lake Whitney State Park, Olson & Ownbey 1795, (4 plants). This collection comes from the southeastern extremity of the range of the species. The sporophytic chromosome num- ber is the same as reported on all previous occasions. Earlier reports for this species are as follows: Hsi (1960), Frank- ton & Moore (1961), Ownbey & Hsi (1963), Ownbey & Hsi (in press). Series VIRGINIANA 14. Cirsium virginianum (L.) Michx. 2» — 28 SOUTH CAROLINA. CHARLESTOWN CO.: 5.3 miles northwest of Mc- Clellanville Post Office, Olson & Ownbey 830, (2 plants). No chromosomal irregularities were observed in the very limited sample of this species. The species is not closely allied to others in the area and its relationships to other North American thistles are obscure. DEPARTMENT OF BOTANY UNIVERSITY OF MINNESOTA MINNEAPOLIS, 55455 MINNEHAHA ACADEMY MINNEAPOLIS, 55406 296 Rhodora [Vol. 71 LITERATURE CITED FRANKTON, C. and R. J. Moore. 1961. Cytotaxonomy, phylogeny, and Canadian distribution of Cirsium undulatum and Cirsium flod- manii. Canad. Jour. Bot. 39: 21-33. . 1963. Cytotaxonomy of Cirsium mu- ticum, Cirsium discolor, and Cirsium altissimum. Canad. Jour. Bot. 41: 73-84. GEISER, S. W. 1935. Charles Wright's 1849 botanical collecting-trip from San Antonio to El Paso; with localities of new species. Field and Laboratory 4: 23-32. GRAY, ASA. 1853. Plantae Wrightianae Texano-Neo-Mexicanae, Part 2. Smiths. Contr. Knowledge 5 (Art. 6): 1-119. Hsr, YU-TSENG. 1960. Taxonomy, distribution and relationships of the species of Cirsium belonging to the series Undulata. Ph.D. Thesis, University of Minnesota. University Microfilms, Ann Arbor, Michigan. Lewis, W. H., H. L. SmRIPLING and R. G. Ross. 1962. Chromosome numbers for some angiosperms of the southern United States and Mexico. Rhodora 64: 147-161. MoonE, R. J. and C. FRANKTON. 1966. An evaluation of the status of Cirsium pumilum and Cirsium hillii. Canad. Jour. Bot. 44: 581- 595. OwNBEY, G. B. 1951. Natural hybridization in the genus Cirsiwm — I. C. discolor (Muhl. ex Willd.) Spreng. X C. muticum Michx. Bull. Torrey Bot. Cl. 78: 233-253. . 1968. Cytotaxonomie notes on eleven species of Cir- sium native to Mexico. Brittonia 20: 336-342. . and Yu-rsENG HsiI. 1963. Chromosome numbers in some North American species of the genus Cirsium. Rhodora 65: 339-354. 1969, Chromosome numbers in some North American species of the genus Cirsium — II. Western United States. Madrono (in press) PETRAK, F. 1917. Die nordamerikanischen Arten der Gattung Cirsium. Beih. Bot. Centr. 35: 223-567. CONTINUED BOTANIZING ON THE WOLF ISLANDS, NEW BRUNSWICK BY A. R. HODGDON AND R. B. PIKE Several species of more than passing interest have turned up on the Wolf Islands since our last report (Rhodora 66: 413-417, 1964). To make our list complete we shall discuss these and any other new records of vascular plants for the Wolves. (1) Typha latifolia L. During the eighth year of our field work on the Wolf Islands we noticed a small colony of Cat- tail at the northern end of the barrier pond at Southwest Cove on East Wolf Island. We were unable to reach the plants to collect them on that visit because of very high water in the treacherous bog, but in the following year, 1968 we did make collections of vegetative plants and noted that the colony was spreading rapidly. We have visited this pond and adjacent bog many times, starting with nearly our first trip to the islands and even searched in particular for Typha, fully expecting to find it since some habitats seemed to us to be ideal for Cat-tails not only here but in other places on both East and South Wolf Islands. Indeed, in one of our papers (Rhodora 66: 141, 1964) Typha was singled out as a remarkable absentee. It is certainly of the greatest interest that, in the many thousands of years that the Wolf Islands have been available for plant colonization, we should happen to be witnesses to the arrival of a weedy and almost ubiquitous species which almost certainly was not introduced by man. (2) Poa subcaerulea Sm. Specimens of what we call P. subcaerulea were first collected by us in open turfy areas above ledges at Southwest Cove, East Wolf Island on July 1, 1964. On June 4, 1968, we again collected specimens in the same locality. When we made our first collections, we didn’t recognize the species but we believed we were getting some- thing quite different from any grass we had seen before. 297 298 Rhodora [Vol. 71 Later at the Gray Herbarium we attempted to match speci- mens with those in the collection in the genus Poa where ob- viously it belonged. Peter Green, working there at the time, looked over our shoulders at our specimens and suggested Poa subcaerulea without hesitation. In early 1968 we sent duplicate material to the herbarium at Kew and under cor- respondence of April 18, 1968, from Peter Green he quotes Dr. Charles E. Hubbard, the eminent grass specialist, who says of it, “The gatherings are both Poa subcaerulea Sm. recognized by the acutely acuminate glumes and few panicle branches.” Peter Green in the same letter went on to say that Dr. Hubbard was “very interested to see the specimens and confirm the presence of this species in North America from personal observation.” There is in the Gray Herbarium at the present time a folder of specimens of Poa subcaerulea at the end of the genus. We have notes on some nineteen of these collections, many of them made by Fernald and his co-workers during his years of active botanizing of areas about the Gulf of St. Lawrence. These specimens were collected all the way from Labrador to Prince Edward Island and Nova Scotia, The range given in Gray’s Manual ed. 8, 1950, p. 117 is “Lab.s to Nfld., M. I. and e. N.B.” Whatever the range, whether that given in Gray’s Manual or a somewhat wider one fol- lowing the data in the folder of specimens in the Gray Her- barium, the Wolf Island record is considerable of a range extension all the way from Eastern New Brunswick or from the head of the Bay of Fundy. There are several points of view about Poa subcaerulea that demand discussion. One relevant problem relates to the origin of the plant, whether native or introduced. Hultén, 1958 in “The Amphi-Atlantic Plants," p. 10 states of it, “It belongs to a very critical group, the area of which cannot be given for the present. The most reasonable is to regard it as a European plant introduced in Eastern America and elsewhere.” Roland and Smith in “The Flora of Nova Scotia,” part 1 of the Rev. Ed. 1966, pp. 84-85, treat Poa subcaerulea as part of the P. pratensis complex but make 1969] Wolf Islands — Hodedon and Pike 299 the following pertinent observation regarding its status as a native population. “Poa subcaerulea appears to be a native coastal type with bluish-tinged spikelets.” We agree with their disposition of it as native but consider it to have more distinctiveness than the color of spikelets. Polunin in his “Circumpolar Arctic Flora,” 67, 1959, commenting on P. pratensis sl. including P. subcaerulea, P. irrigata and P. alpigena states “fully cireumpolar in distribution." From the habitat and occurrence of our Wolf Island speci- mens we would be inclined, as some others have been, to call the plant native. It seems to us unlikely that Poa subcaerulea was brought in from Europe, especially after examining material collected by Fernald and his associates in Western Newfoundland where many of the stations are in wild and remote coastal areas. The second major problem about Poa subcaerulea is its status as a species. In complexes such as that of Poa praten- sis and its relatives, we ought to consider whether there are distinctive populations that are morphologically unique, are geographically isolated and that presumably have evolved independently. Assuming, as we do, that Poa subcaerulea is native along the coast of Eastern America and following Hubbard, Fernald and others, considering it to be mor- phologically distinct from typical Poa pratensis, we are pre- pared to accept Poa subcaerulea as a species that must have evolved out of the P. pratensis complex at some fairly remote time. This is further shown by its disjunct occurrence on both sides of the Atlantic. Often the specific lines separating P. pratensis from related species are blurred, perhaps large- ly because of interfertility between the related taxa and consequent introgression. Hybridization is so very common in many plant groups in nature that the great swarms of confusing hybrids, often wrongly interpreted, tend to ob- scure the evolutionary lines. It seems to us, at this time, that Poa subcaerulea merits treatment as a species closely related to Poa pratensis. (3) Carex deflexa Hornem. Reported erroneously as C. Emmonsii Dew. (Rhodora 65: 89, 1963) and discussed as to 300 Rhodora [Vol. 71 its occurrence on the Wolves (Rhodora 66: 144-145, 1964). Our first station, in turf over ledges bordering the sea was very limited, there being few plants. We were fortunate in 1967 to find a more extensive growth of the same sedge in a more inland situation on South Wolf Island. Careful study in the herbarium rules out C. Emmonsii for either station and make it evident that we have here the more northern C. deflexa. In our 1964 paper (loc cit) we had demonstrated the boreal character of the Wolf Island flora contrasted to that of Grand Manan, the presence of Carex Emmonsii being an anomalous and disturbing matter since it might, with more likelihood, have been on Grand Manan because of its more southern affinities rather than on the Wolves. Now that the plant is recognized as C. deflexa, which fits the bo- real pattern of the Wolf Island flora, our thesis is strength- ened. Incidentally, Carex deflexa has not been collected on Grand Manan. The earlier collection named C. Emmonsii by error should be changed to C. deflexa. It is our Pike & Hodgdon no. 17787. (4) Carex pallescens L. var. neogaea Fern. A widely dis- tributed sedge of not pronouncedly northern distribution. We have found this but once in an open area near the south- western end of the barrier beach at Southwest Cove on Au- gust 17, 1967. There are many records on Grand Manan and there is nothing startling about its presence on East Wolf Island, except its extreme scarcity, there being but one plant seen. (5) Rumex orbiculatus Gray. Edge of barrier pond at Southwest Cove, East Wolf Island, September 19, 1964. Even at this late time of year the fruits were still so immature that they proved to be nearly useless as aids to identification. However, the valves of the calyx lack teeth or bristles while other features of the plant including the very large leaves and the wet habitat preference belong to R. orbiculatus. (6) Heracleum maximum Bartr. A single large plant growing beside the stairway leading to the lighthouse on South Wolf, August 17, 1967. We had passed this way many times in earlier years and hadn’t seen it. This part of the 1969] Wolf Islands — Hodgdon and Pike 301 Wolf Islands had been occupied more extensively than any other part and probably this merely represents the reap- pearance of a species that had been there at some previous time and had been dormant at the time of our early visits. (7) Solidago puberula Nutt. A widely distributed golden- rod fairly common on Grand Manan. Only a few specimens were seen on East Wolf Island at the edge of the bog at Paul’s Cove on September 6, 1965, the only station we have found on the Wolves. The specimens were small but other- wise quite typical. In 1964 (Rhodora 66: 413-416, 1964) we provided some statistics on the flora of the Wolf Islands. These we will now revise, incorporating all additions and changes since then. Table I Total taxa on the Wolf Islands — 338 East South Fat Flat Gull Wolf Wolf Pot Wolf Rock Comparative areas of islands 42 22 6 3 1 Taxa recorded 302 169 107 .96 35 Ratio of taxa to area 7.24 145 195 32 38 Some changes are also required in the statistics on boreal taxa of the Wolf Islands and Grand Manan. The initial comparison of these two groups of islands showed that the Wolf Islands had 43 boreal taxa of which 33, confined to the Wolves, could be regarded as boreal disjuncts while Grand Manan had a total of only 22 boreal taxa. The boreal Betula papyrifera var cordifolia, which is very common on the Wolves and which we had reported from Grand Manan (Rhodora 64: 100-101, 1962) should have been added at that time to the Grand Manan list making 23 boreal taxa, of which 11 were shared with the Wolves. With the further addition of 3 boreal taxa to the Wolf Islands, one of which was shared with Grand Manan (Rho- dora 64: 413-416), the total of Wolf Island boreal taxa then would have become 46 rather than the reported 45 and the taxa shared with Grand Manan, 12 rather than 11. 302 Rhodora [Vol. 71 In this paper we are adding 2 more boreal species con- fined to the Wolf Islands, Poa subcaerulea and Carex de- flexa, bringing the total of boreal taxa on the Wolf Islands to 48, while in the current paper on Grand Manan (Rhodora 71: 306, 1969) one additional boreal species Salix pellita is reported from that group of islands and the taxon Ana- phalis margaritacea var subalpina is also newly reported from Grand Manan though found previously on the Wolf Islands. A tabular up-to-date summary of the boreal taxa on the two island groups is given in table 2. Table 2 Boreal taxa on Wolf Islands and Grand Manan Total On On boreal taxa Wolf Islands Grand Manan Shared 60 (100%) 48 (80%) 25 (41.7%) 13 (21.7%) It is evident that further collecting on the Wolf Islands and on Grand Manan continues to support the thesis that the Wolf Islands have a remarkably prominent boreal ele- ment in their flora contrasted to that of Grand Manan. DEPARTMENTS OF BOTANY AND PLANT SCIENCE UNIVERSITY OF NEW HAMPSHIRE DURHAM, NEW HAMPSHIRE 03824 ISOTRIA MEDEOLOIDES ON THE PIEDMONT OF VIRGINIA Plant distribution on the Piedmont of Virginia is more poorly known than for any physiographic province of the state. Aside from a northern portion of the Piedmont where Washington botanists frequently collect, and Amelia County where Mr. J. B. Lewis worked many years ago, collections from this part of Virginia are scanty and come from scat- tered localities. Travel in the area quickly demonstrates the reasons for a paucity of collections. The Piedmont is botanically uninteresting compared with other regions of Virginia, and it is difficult to find good collecting grounds. Oak-hickory and pine forests form much of the dominant vegetation and few habitats are available for rare and local species. Dr. G. W. Ramsey with his coworkers, and the author with much help from many interested persons, are making concerted efforts to close gaps in our knowledge of plant distribution and ecology of the region. A result of this work is the discovery of several relict areas which have an im- portant bearing on the ecology and vegetational history of Virginia. Such an enclave with many relict populations occurs in southern Buckingham County. Here soils of lower wooded slopes have very high calcium, magnesium, and pH values, and in these soils grow many plants which are rare or un- known elsewhere in the Piedmont. Among localized species are Carex jamesii Schwein., Tradescantia virginiana L., Trillium sessile L., Cypripedium calceolus L., Ostrya vir- giniana (Mill.) K. Koch, Ulmus rubra Muhl., Draba ramosis- sima Desv., Rhus aromatica Ait., Euonymus atropurpureus Jacq., Aesculus octandra Marsh., Hybanthus concolor (Frost.) Spreng., Panax quinquefolius L., Phlox divaricata L., Conopholis americana L., Campanula americana L., C. divaricata Michx., and Polymnia canadensis L. Many popu- 303 304 Rhodora [Vol. 71 lations within the general area are localized colonies of few individuals. Many of these species have one or more outliers on the coastal plain where they are edaphic relicts of more wide- spread distributional patterns of the past. Since these lo- calized plants have their main population centers in the mountains to the west, we were greatly surprised to find a single individual of the rare Isotria medeoloides ( Pursh) Raf. formerly known also as Pogonia affinis, in the Buck- ingham County area. Barbara J. Harvill found the orchid growing on a wooded north-facing slope. Dr. Donovan Correll kindly confirmed its identity. This pogonia is ex- tremely rare in Virginia and was heretofore known only on the coastal plain (Baldwin, 1968; Grimes, 1921; Taylor, 1934). The discovery adds another species to those with remarkable disjunct populations in Virginia. A. M. HARVILL, JR. LONGWOOD COLLEGE FARMVILLE, VIRGINIA REFERENCES BALDWIN, J. T., JR. 1968. Obituary for an orchid. Virginia Wildlife 24: 22. Grimes, E. J. 1921. A new station for Pogonia affinis. Rhodora 23. 195-197. TAYLOR, RAYMOND L. 1934. The present status of Pogonia affinis. Claytonia 1: 5. A RHODE ISLAND STATION FOR LUMINOUS MOSS: Owing to the apparent scarcity of Schistostega pennata, it seems worthwhile to record a station for it in Rhode Island. The moss occurs in a graphite mine in the town of South Kingstown, Washington County. The Indians reportedly first worked this vein of graphite and called the area Cajoot. As late as the 1890’s graphite was mined there, and the Schistostega has come in since then. In 1956 when David Nichols was surveying the Nar- ragansett Pier, R.I. quadrangle for a surface geology map, his father, Reginald Nichols, who accompanied him, noted the glowing moss and took color photographs of it. The moss grows in a north-facing tunnel located part way down the twenty-foot mine pit. It does not grow at the very mouth of this tunnel but begins at about six feet in and extends further another yard. At no time between May and September when I observed did the sun’s direct ray fall on the moss habitat; hence probably it does not at any other time. Schistostega grows only on the floor of this cavern, not ascending the walls, and it seems more partial to flat, moist chunks of graphite than to the crumbled earth between them, Its glow at this particular site is best viewed at a forty-five degree angle with some variation either way. Despite traffic in the form of youngsters exporing the abandoned mine, the moss survives although in less quan- tity than when first seen in 1956. A specimen has been deposited at the University of Rhode Island Herbarium by Dr. Richard Hauke of the Botany Department. RICHARD L. CHAMPLIN JAMESTOWN RHODE ISLAND 02835 MORE ADDITIONS TO THE FLORA OF GRAND MANAN, NEW BRUNSWICK In 1962 (Rhodora 64: 98-102) six new taxa were reported for Grand Manan Island to be added to the Weatherby and Adams Flora (A list of the Vascular Plants of Grand Manan, Contr. Gray Herbarium LVIII, 1945). During the summer of 1965, one of the present authors (Harry Tyler) then an undergraduate student at the University of New Hampshire, acted as a summer instructor at the Gerrish House Society which is located at Grand Harbour. He made particular use of his time in searching for additions to the list of Grand Manan plants. The following taxa reported in this paper were all collected by Tyler and, like those newly reported in 1962, are deposited in the Herbarium of the University of Hampshire (NHA). Salix pellita Anderss. In roadside gravel at North Head. The specimen lacked catkins but had leaves with silky-vel- vety lower surfaces and essentially entire or obscurely serrate margins, satisfactorily matching representative herbarium material of S. pellita. Polygonum aviculare L. var. vegetum Ledeb. White Head Island. This suberect and broad-leaved variety of the heteromorphous P. aviculare often occurs in the Bay of Fundy area. Although Weatherby and Adams did not distinguish varieties in this complex it might be well to do so now considering the great difference be- tween this plant and the more common typical weedy form. Lynchnis alba Mill. In dry soil at Castalia. Both male and female plants were collected and the female flowers had 5 styles which distinguishes it from the often confused Silene noctiflora L. Its white flowers and long atten- uate sepals characterize L. alba. Sorbaria sorbifolia (L.) A. Br. Thicket by road at Whale Cove. This seems to have been an escaped colony, but 306 1969] Grand Manan — Hodgdon, Pike and Tyler 307 may represent only persistent plants from earlier cul- tivation. In any event it should probably be counted as part of the natural flora of the islands. Pyrus floribunda Lindl. Ingall’s Head. No flowers or fruits were present but the densely pubescent young twigs and lower leaf-surfaces eliminate the more common P. melanocarpa (Michx.) Willd. which was reported by Weatherby and Adams, Pyrola elliptica Nutt. Deciduous woods at Castalia. The rel- atively long and thin blades and the many-flowered long-cylindric racemes serve to characterize this spe- cies. The numerous specimens collected are all typical. Lysimachia quadrifolia L. In gravel along a back road at Grand Harbour. The perfectly typical specimens leave no doubt about the identification. The presence of the species in Washington County, Maine would certainly lead one to expect the plant in New Brunswick and per- haps more particularly on Grand Manan since Weather- by and Adams comment on certain affinities of the Grand Manan Flora to that of southern Maine. We have seen specimens from Machias and Steuben in Washington Co. but, so far as we can determine, this is the first record from the Maritime Provinces. Mentha arvensis L. var. arvensis. Weed at Grand Harbour. The two small specimens collected in flower, have the leaves definitely broadest below the middle and rounded abruptly at the base to the short petioles. This variety, probably introduced, is widely distributed and to be expected. However, Weatherby and Adams stated in particular that the Mentha arvensis that they had seen from Grand Manan was the var. villosa (Benth.) S. R. Stewart. Anaphalis margaritacea (L.) C. B. Clarke var. subalpina Gray. Gravelly back road at Grand Harbour, The varieties of Anaphalis margaritacea are not always clearly defined but within a certain range of variation 308 Rhodora [Vol. 71 the numbers of leaves and distribution of pubescence on them seem to be reasonably good characters. The two specimens at hand at early maturity have about 36-38 leaves, the upper ones, just below the inflores- cence, of good size and all leaves somewhat floculent- tomentose above. Weatherby and Adams had identified their Grand Manan material as var. intercedens Hara. It is interesting that careful search on Grand Manan has not yet disclosed the conspicuous var. margaritacea with its bright green upper leaf-surfaces which is com- mon on the nearby Wolf Islands. A. R. HODGDON, R. B. PIKE & HARRY R. TYLER, JR. DEPARTMENT OF BOTANY AND DEPARTMENT OF PLANT SCIENCE UNIVERSITY OF NEW HAMPSHIRE, DURHAM 03824, AND DEPARTMENT OF OCEANOGRAPHY OREGON STATE UNIVERSITY, CORVALLIS 97331 REVISION OF THE GENUS DAHLIA (COMPOSITAE, HELIANTHEAE — COREOPSIDINAE) PAUL D. SORENSEN The genus Dahlia (Compositae, Heliantheae — Coreopsi- dinae) famed because of its numerous ornamental forms of gardens, was named by Cavanilles in 1791 honoring An- dreas Dahl, a Swedish botanist and pupil of Linnaeus. It is eomposed of four sections, twenty-seven species, and four infraspecific taxa and is largely restricted to the highlands (1500-4300 meters elevation) of México and Central Amer- ica. Two species, Dahlia coccinea and D. imperialis, are also found in some countries of South America where they are believed to have been introduced. The genus encompasses some extremely diverse life forms ranging from dwarfed perennial herbs scarcely over four decimeters tall (D. scapi- gera) to huge arborescent plants which, in some instances, ascend to eight or nine meters (D. imperialis), and to scrambling epiphytic vines which sprawl among the tree- tops in rain forest (D. macdougallüi). The great majority of the species have extremely restricted ranges, but one (D. coccinea) may be considered a roadside weed and is very wide-ranging throughout México and Guatemala. Most historical accounts of the genus Dahlia have dealt with its rather important role in horticulture. Little or no attention has been given the fact that the genus is composed first of all of a group of wild species some of which have served as the progenitors of our garden subjects. Similarly workers who have written about the wild species have sel- dom taken into account the aspects of the genus in horti- culture. As is discussed in detail later in this work, there appear to be just two of the naturally occurring species of Dahlia which have been involved, through processes of hy- bridization and selection, in the production of the numerous garden forms now known in cultivation the world over. In the past many hybrid variants were singled out and er- roneously given formal taxonomic rank and provided with 309 310 Rhodora [Vol. 71 Latin names, often also with Latin and vernacular diag- noses. Such treatment of these cultivars has given them the semblance of being botanical species. The purpose of the present study is to direct attention primarily to the syste- matics of the wild species where they occur naturally, while borrowing from the horticultural literature such names and information as are appropriate in clearing up misunder- standings in nomenclature. Elsewhere I shall present a full exploration of the history of Dahlia from the viewpoint of its cultivation dealing at length with its role in horticulture. The formal history of Dahlia begins in the late 18th Cen- tury in Spain when Antonio José Cavanilles described and illustrated Dahlia pinnata in volume one of his Icones et Descriptiones Plantarum, 1791. Cavanilles, then a senior member of the staff of the Royal Botanic Garden in Madrid (not the director as has so often been repeated in articles on Dahlia, see Colmeiro, 1875), had received seeds of Mexi- can plants from Vicente Cervantes, a botanist at the Mexican Botanic Gardens in the city of México, Plants grown from these seeds were cultivated in Madrid and they formed the basis of the material used by Cavanilles to prepare the six volumes of his Icones issued during the years 1791-1801. Among these seeds were those of the original Dahlia pinnata. as well as some of two additional species, D. rosea and D. coccinea which Cavanilles subsequently described in volume three of the Icones, 1796. Thus from the beginning Dahlias have been plants of gardens, During the 17th and 18th centuries Europeans had a lively interest in the plants being brought back from the New World and botanical institutions freely exchanged seeds of noteworthy introductions. In this atmosphere Dahl- ias were, within a matter of a few years, found in the major botanie gardens of France, Italy, Germany and England. Presumably the earliest source of these plants was Madrid whence the gardens of France and England received their seeds, but meanwhile other plant collectors had also intro- duced Dahlia seeds into Europe from México. In any case, it is clear by the large numbers of variants which arose 1969] Dahlia — Sorensen 311 among the plants in their gardens that Dahlias were creat- ing a mild sensation among European horticulturists of the time. Willdenow was the first to present a generic treatment of Dahlia. In his ed. 4 of Linneaus’ Species Plantarum, 1803, he included the three species described by Cavanilles but he adopted in place of Dahlia, the name Georgina believ- ing that Dahlia Cav. was preoccupied by Dahlia Thunberg (1792) of the Hamamelidaceae for which Persoon sub- sequently substituted the name Trichocladus. Although Georgina eventually came to be widely used, and is today a vernacular name for the garden Dahlia in Scandinavia and the countries east of the Rhine, botanists of the time mostly recognized Willdenow’s error in dates and continued to publish notices, using the name Dahlia, In 1804, Thouin, working from material of the genus under cultivation in the gardens of the Paris Natural History Museum, pub- lished what must be taken as the first treatise on the culti- vation of Dahlia. At the same time he offered a brief taxonomic review of the species accompanied by the world’s first published colored portrait of them. This portrait must have caused considerable interest because almost at once it was reproduced in several gardening journals of the period and undoubtedly helped to create further enthusiasm for these new garden plants. Willdenow, in 1809, again re- vised his classification of Georgina recognizing two species. One of his species was given the aptly chosen name G. vari- abilis under which he placed all of the cultivated forms with ligules of purple, lilac, or rosaceous colors. For these he gave the varietal names of purpurea, lilacina, and pallida. Similarly, he placed all of the scarlet, orange, and yellow- rayed forms under G. coccinea and provided the correspond- ing varieties coccinea, crocea, and flava. One can appreciate Willdenow’s problem as the number of Dahlias developed in gardens created an ever-increasing and bewildering array of new and different forms. The name variabilis seemed to fit the garden Dahlia so well that it was immediately adopted almost to the exclusion of all others. 312 Rhodora [Vol. 71 The following year, 1810, de Candolle, seemingly in deference to the lead established by Willdenow, published a revision of Dahlia under the name Georgina, This paper appeared in the Annals of the Paris Natural History Muse- um whose editors added a disclaimer stating they were publishing the paper just as de Candolle requested despite their feeling that the name Dahlia should have been used. The generic treatment offered by de Candolle is very sim- ilar to that of Willdenow on the infraspecific level but for G. variabilis Willd. he substituted G. superflua, and for G. coccinea (Cav.) Willd., G. frustranea. These new names had a brief acceptance but de Candolle himself even sub- merged them in the synonymy of his treatment of Dahlia for the Prodromus (1836). In 1829, Desfontaines offered in his ed. 3 of the Catalogus Plantarum Horti Regii Parisien- sis, the combination DaAlia variabilis (Willd.) Desf. in the synonymy of which he placed Willdenow's Georgina varia- bilis and de Candolle's G. superflua. Since that time, Dahlia variabilis has become widely used and persists even in cur- rent literature as the name for the cultivated garden Dahlia. During the remainder of the 19th and in the early years of the 20th centuries, several more species of Dahlia were described. A good summary of the taxonomy and nomencla- ture of the genus was offered by Hemsley in 1879 and again in 1881. Between the time of Hemsley and the present the greatest contributions to our understanding of the wild Dahlia species came from the work of E. E. Sherff who, at the time of his death, April 1966, was regarded as a leading authority on this and other genera of the Compositae. In 1930, in connection with his monographic work on the Com- positae, Subtribe Coreopsidinae, Sherff described D. linearis and thereafter regularly published notes including descrip- tions of several more new taxa until, in 1955, for an in- stallment of the North American Flora, he contributed the most recent systematic treatment of the genus. Here he recognized 29 taxa, including 18 species, 8 varieties, and 3 forms. In the present treatment, one section, six species, and two varieties are proposed as new. Of the 29 taxa recog- 1969] Dahlia — Sorensen 313 nized by Sherff 12 are maintained here, six are reduced to synonymy and ten are recombined at different taxonomic levels. Table 1. Conspectus of the species and summary of chro- mosome numbers of the Genus Dahlia. Section 1. Pseudodendron Sherff 1. D. excelsa Benth. 2. D. imperialis Roezl ex Ortgies (n — 16) 3. D. tenuicaulis Sorensen, sp. nov. (n — 16) Section 2. Epiphytum Sherff 4. D. macdougallii Sherff Section 3. Entemophyllon Sorensen, sect. nov. 5. D. scapigeroides Sherff 6. D. foeniculifolia Sherff 7. D. linearis Sherff (n = 17) 8. D. rupicola Sorensen, sp. nov. (n = 17) 9. D. dissecta S. Wats 9a. var. dissecta (n 17) 9b. var. sublignosa Sorensen, var. nov. Section 4. Dahlia 10. D. merckii Lehm. (n = 18) 11. D. apiculata (Sherff) Sorensen, comb. & stat. nov. 12. D. cardiophylla Blake & Sherff (n = 16) 13. D. purpusii Brandg. 14. D. pinnata Cav. (n = 32) 15. D. pteropoda Sherff 16. D. brevis Sorensen, nom. nov. (n = 16) 17. D. rudis Sorensen, sp. nov. 18. D. moorei Sherff 19. D. hintonii Sherff 20. D. mollis Sorensen, sp. nov. (mo 16) 21. D. atropurpurea Sorensen, sp. nov. (n = 32) 29. D. australis (Sherff) Sorensen, comb. & stat. nov. 22a. var. australis (n — 16 & 32) 29b. var. liebmannii (Sherff) Sorensen, comb, nov. 22c. var. chiapensis Sorensen, var. nov. 29d. var. serratior (Sherff) Sorensen, comb. & stat. nov. 314 Rhodora [Vol. 71 23. D. sherffii Sorensen, sp. nov. (n = 32) 24. D. scapigera (A. Dietr.) Knowles & Westc. (n — 16) 25. D. barkerae Knowles & Westc. (n — 32) 26. D. tenuis Robinson & Greenman (n — 16) 27. D. coccinea Cav. (n — 16 & 32) Evidence supporting the taxonomic and nomenclatural changes herein proposed for the genus DaAlia has been gathered from three main avenues of research: field studies of the intra- and interpopulational variability, morphologic studies using both living materials and herbarium specimens, and chromosomal studies. Pertinent data are discussed where appropriate in connection with the taxonomic treat- ment of the species to which they apply. However, it is useful at this point to offer a brief explanation of the mode of morphologic variability found among the species and to call attention to certain features of the foliage and the termi- nology used. Also, for the purposes of the present treat- ment, the principal details of the chromosomal studies are outlined below. A full report on this phase of the research will be presented elsewhere. Chromosomal Studies Chromosome numbers are now known for 17 taxa of the genus Dahlia as follows (see also Table 1): n = 16 7 spp. n = 17 3 spp. n = 18 1 sp. n — 16 & 32 2 spp. Ishikawa (1911) was the first to publish a report on Dahlia chromosomes. He found the numbers of D. variabilis (i.e., D. pinnata) and D. coronata (i.e., D. coccinea) to be 2n = 64 and 32 respectively. Dahlia imperialis wag re- ported to have 32 (n = 16) chromosomes by Belling (1924) and in 1929 Lawrence added a report on D. merckii (n = 18). Thus, at the outset of this study, three of the four numbers found in the genus were known. The fourth, - 1969] Dahlia — Sorensen 315 n = 17 (D. dissecta, D. linearis, and D. rupicola), is re- ported here for the first time. Several interesting remarks may be made concerning the distribution of chromosome numbers within Dahlia. In some instances the chromosome numbers correlate remark- ably with morphologic data. This is demonstrated particu- larly well among the taxa of the newly described section Entemophyllon which on morphologic grounds are clearly allied. All are diploids and all have the same chromosome number, n = 17. Moreover, the aneuploid D. merckii, n = 18, comes morphologically closest to the members of this section. Similarly, the two species of "tree-dahlias" of section Pseudodendron are diploids and share the same chromosome number, n = 16. The only tetraploids (n = 32) known in the genus occur within the section Dahlia where they number about as many as the diploids. Two of the species, D. coccinea and D. australis, are composed of both diploid and tetraploid races. A majority of the species have rather restricted geographic ranges and it is significant that most of these are diploids. Notable exceptions are D. merckii and D coccinea which are rather wide ranging though the tetraploid races of D. coccinea extend over a much greater area than do the diploid. Morphological Considerations The size, shape, and segmentation of the compound leaves of some Dahlia species are extremely variable. To promote clarity in the use and understanding of the keys and de- scriptions a brief explanation is provided below of the type of materials, the technical characters, and the terminology employed in the systematic treatment. It is hoped that this will also serve as a guide to the type of materials one should gather when making future collections of Dahlia. a) All descriptions of the species are based upon data derived from mature plants collected in the field, not greenhouse-grown individuals. The latter often 1A previous report (Turner et.al., 1962) of the chromosome number of D. dissecta as n = 18 was based on a specimen (Rock M-351 [TEx]) of D. merckii. 316 Rhodora [Vol. 71 exhibit altered characters such as smaller leaf segments and longer petioles. Plants in their first year of growth from seed usually produce Juvenile foliage unlike that on the same individual grown from tubers in later years. b) Measurements of stems and leaves are based on median leaves and adjacent portions of the stems. c) Length of leaves includes the petiole and terminal leaflet. d) Width of leaves is measured at the basal pair of leaflets or lobes. e) Length of leaflets is based solely on the basal pinna and is measured from the primary rachis to the tip of the terminal pinnule. In the same manner, the length of pinnules is measured from the rachilla to the tip of the basal pinnule on that axis. The necessity of designating, in this way, precisely which portion of the leaf is being measured is seen from the selected types of leaves portrayed in Fig. 1. Note that the basal pinna and pinnules are always the largest and that they diminish greatly in size the more distal they are. f) Shapes of leaflets are of the blades of ultimate seg- ments as, for example, in the case of bipinnate leaves, the basal pinnule of the basal pinna. Ex- ceptions to this occur in some descriptions when Specific reference is made to another segment such as the terminal leaflet or terminal pinnule. g) The petiole is measured from the stem to the junction of the basal pair of leaflets. In the case of pinna- Fig. 1. Selected leaf types XX !4 and their corresponding petiolar cross-sections X 2 of Dahlia spp. A. a compound leaf with alternate pinnules, Dahlia rupicola; B. a compound leaf with alternate pinnules, Dahlia merckii; C. a compound leaf with opposite pinnules, from an artificial hybrid, Dahlia australis var. australis X coccinea; D. a pinnatissect leaf, Dahlia pinnata ; a. stipel, note the compound stipels on the leaf of A; b. rachis internode; c. rachis node. 317 Dahlia — Sorensen 1969] 318 Rhodora [Vol. 71 tifid leaves (Fig.1-D) when the limits of the petiole are not clear, measurements are from the stem to the junction of the first (i.e. basal or proximal) pair of main lateral veins. Petiolules and pseudopetiolules are measured in the same way. h) Petiolar cross-sections are taken from a point mid- way between the stem and basal pair of leaflets or main lateral veins. Types encountered are shown in Fig. 1. If working with dried material it is possible to see clearly all the diagnostic features of these cross-sections following hydration in warm water. i) The junction of the primary rachis with the secondary rachis or rachilla is termed a “rachis node". Con- sequently, a “rachis internode” denotes the portion of the axis between the junctions of successive pairs of leaflets (see Fig. 1). j) The leaves of various species within DaAlia may be sorted into four main classes: simple leaves, com- pound leaves with alternate pinnules (Fig. 1-A, & -B), compound leaves with opposite pinnules (Fig. 1-C), and pinnatifid or pinnatisect leaves (Fig. 1-D). The degree of dissection of compound leaves, whether once-pinnate, bi-, or tri-pinnate, has been based solely on the basal pinnae. k) The occurrence of stipels, as seen in the leaves of Fig. 1-A, -B, & -C) is diagnostic with some species and not so with others. Attention is being called to them so the user may be aware of the structure to which the term refers. l) The principal diagnostic features of the heads are the rays (for their color) and the outer whorl of involucral bracts (for their position, size, shape, and ornamentation). In all cases, the character- 1969] Dahlia — Sorensen 319 istics of these structures are based on heads in anthesis. The nature of the materials needed for the identification of most Dahlia species is implicit in the foregoing notes. However, the species of the section Pseudodendron, the so- called “tree-dahlias”, require special mention. Most leaves of these species are too large to collect intact. A single leaf may measure 80 to 90 centimeters long. Collectors should select a single basal pinna of a median leaf and supplement the specimen with notes on total length of the leaves and the number of primary leaflets. Notes should also be made concerning the height and habit of the plants. Collections of the suffrutescent species should include a portion of the two-year-old stem. Various authors and collectors have, at times, confused species of Dahlia and Cosmos. This has been particularly true of Cosmos diversifolius which is easily mistaken for Dahlia scapigera, and D. brevis which has been identified as Cosmos concolor. At anthesis Cosmos species are dis- tinguished by a cylindrical brush of hairs on each of the five staminal filaments, while in species of Dahlia the fila- ments are smooth. In fruit the achenes of Dahlia are obcompressed and mostly 2-sided or slightly 3-angled in cross-section, those of Cosmos, quadrangular. The fleshy outer involucral bracts of Dahlia, especially as seen in fresh material, are quite distinct from the thin bracts of Cosmos. Species of the closely allied and deceptively simi- lar Hidalgoa, often called climbing-dahlia, are distinguished by their sterile disc florets, and their petioles which twist and cling in the manner of Clematis. I cannot conclude this introduction without acknowledging with sincere thanks the help of many persons — Dr. Thomas E. Melchert, University of Iowa, for suggesting the study and guiding me from its beginning; Dr. George W. Martin, Professor Emeritus, for his advice and encouragement over the years; Daniel J. Crawford, for his companionship in the field in México and assistance with other phases of 320 Rhodora [Vol. 71 the research; Dr. Jerzy Rzedowski, Curator of the Her- barium, Instituto Politécnico Nacional, México, D. F., for his help with locating many obscure place names in México and his advice on highway travel in that country. Many thanks are due the staff of the Arnold Arboretum, Dr. Richard A. Howard, Director, Dr. Carroll E. Wood, Jr., Dr. Bernice Schubert, and Dr. Gordon P. DeWolf, Jr., who critically read the manuscript and offered many helpful suggestions in nomenclatural and editorial matters. During two field trips through México and Central America I was able to collect 18 of the 31 taxa recognized here. In addition to recollecting many poorly known species and making extensive observations of natural populations, I was able to gather propagating materials of species which had never before been brought under cultivation. Being able to observe living materials of the Species pass through their life cycle greatly added to my present understanding of the genus. The study is otherwise based on herbarium specimens borrowed from the following institutions (the standard abbreviations of Index Herbariorum are used consistently throughout this work): A Arnold Arboretum, Harvard University BH Bailey Hortorium, Cornell University CAS California Academy of Sciences DS Dudley Herbarium, Stanford University F Field Museum of Natural History GH Gray Herbarium, Harvard University IA The University of Iowa ISC Iowa State University MEXU Herbario Nacional, Instituto de Biología, Universidad Nacional de México MICH University of Michigan MSC Michigan State University MO Missouri Botanical Garden NY New York Botanical Garden POM Pomona College RSA Rancho Santa Ana Botanic Garden TEX University of Texas 1969] Dahlia — Sorensen 321 UC University of California US U. S. National Herbarium, Smithsonian Institution WIS University of Wisconsin I am grateful to the curators of these herbaria for the loan of specimens. I should like also to acknowledge the assistance of the National Science Foundation whose grant (No. GB-3851) to Dr. Thomas E. Melchert helped to defray the expenses of the field work. Specimens cited in the lists of exsiccatae are arranged in a geographic sequence by country from east to west. The states of México are in order more or less by tier from north to south and within each tier from east to west. A complete first set of my own collections from México and Guatemala is on deposit at IA. SYSTEMATIC TREATMENT Dahlia Cav. Icones et Descr. Pl. 1: 56. 1791. Type species: Dahlia pinnata Cav. (see Sherff, 1955.) Georgina Willd., Linn. Sp. Pl. ed. 4. 3: 2124. 1803. Type species: Georgina purpurea Willd. Georgia Sprengel, Anleitung zur Kenntniss der Gewacshe, ed. 2. 2: 567. 1818. Type species: Georgia variabilis Willd. Scrambling epiphytes, herbaceous or suffrutescent perennials, de- veloping from fascicled tubers or with the rootstock tuberously thick- ened, in one epiphytic species the roots aérial (D. macdougallii). Stems erect, rarely vine-like, arising singly (sect. Dahlia), 2-4 (sect. Ps^udcdendron), or many in a cluster (sect. Entemophyllon), the indi- vidual stems mostly unbranched except in the flowering portion or branching freely from the base (D. merckii); internodes hollow or solid, sometimes the pith chambered (sect. Entemophyllon); nodes septate or the septa sometimes incomplete. Leaves opposite or whorled (8 per node); simple to 3-pinnately compound, often stipellate; mar- gins ciliolate or eciliolate. Heads radiate, on long, slender, naked peduncles; involucral bracts distinctly bi-seriate, the outer (4-)5(-6-7) erect, spreading or reflexed at anthesis, fleshy, dorsally several-lined, green, narrowly linear to broadly ovate or obovate, the inner (7-) 8(-9), membranous, many-lined, variously tan, red, reddish-brown or brown, often the tips colored purple or red, margins scarious or white- hyaline, ovate, subacute, obtuse, or emarginate, enlarging in fruit 322 Rhodora [Vol. 71 to about twice their size at anthesis; chaff scarious, light tan suffused with red or purple, resembling the inner bracts but with fewer lines, subtending each disc floret, ovate, obtuse or subacute; ray florets neutral or pistillate sterile, rays white or whitish-lavender to deep purple or yellow to various shades of orange to deep blackish-scarlet, narrowly or broadly ovate, acute, often minutely denticulate, some- times the veins dorsally minutely pubescent; dise florets yellow, yellow with red or purple tips or the limb purple throughout, 17-172 per head, hermaphrodite, fertile, tubular (all or some replaced by sterile or fertile ligulate florets in many horticultural forms), corolla expanded into a 5-toothed, cylindrical limb, the teeth erect, spreading, recurved or strongly revolute; filaments glabrous; style branches linear-lanceolate, bearded dorsally for 34 their length, ventrally near the tips only, spreading or slightly reflexed at anthesis. Achenes grayish-black to black at maturity, more or less linear to linear- oblanceolate, rarely broadly spatulate, obcompressed, obscurely sulcate, sometimes puberulent (minutely ‘tuberculate in sect. Entemophyllon), often slightly constricted just beneath summit and capped by a shallow tan disc (or the disc swollen and black in D. moorei); pappus obsolete or consisting of 2 minute rudiments, these sometimes elongated to 1 mm., rarely greatly elongated into weak, flexuous filiform, often caducous, threads. Chromosome num- bers, n — 16, 17, 18, and 32. KEY TO THE SECTIONS A. Plants liana-like, producing aérial roots to 22 m. long; stems scrambling for great distances in the tree tops of rain forest west of Tehuantepec. ........................................ Sect. 2. Epiphytum. A. Plants growing erect, not vine-like; roots tuberous and anchored in the soil, not aérial. B. Plants suffrutescent, 21-90 dm. tall; ligneous stems of the previous year 1.5-10 cm. diam.; pinnules opposite on the ra- chilla. .......................................... Y Sect. 1. Pseudodendron. B. Plants herbaceous, 4-30 dm. tall or subshrubs 9-15 dm. tall, if the latter, then the ligneous stems of the previous year 0.3-2 cm. diam.; pinnules opposite or alternate on the rachilla. C. Pinnules or pinnular segments invariably alternate on the rachilla; petioles solid; leaves bipinnatisect to bipinnate- pinnatisect. ...................................... Sect. 3. Entemophyllon. C. Pinnules or pinnular segments opposite on the rachilla or if alternate then the petiole hollow or the leaves merely simple. ERR Sect. 4. Dahlia. KEY TO THE SPECIES Section 1. Pseudodendron A. Principal leaves 50-90 cm. long 1969] Dahlia — Sorensen 323 B. Primary pinnae 3-7; outer involucral bracts linear. ................ ec ome dec cnm S E Ro 1. D. excelsa. B. Primary pinnae 9-15; outer involucral bracts obovate or RPE ate V T EET 2. D. imperialis. A. Principal leaves 20-43 cm. long. ........................ 3. D. tenuicaulis. Section 2. Epiphytum ONERE PECES CR PN NORTE HIE ATE: 4. D. macdougallii. Section 3. Entemophyllon A. Outer involucral bracts 5-9 mm. long; petioles solid. B. Peduncles 1.5-10 cm. long; outer bracts acute; current year's growth developing from woody stems, these usually with cham- IBered DIU a nn TS... Q S S pss 7 D. linearis. B. Peduncles 12-37 cm. long; outer bracts obtuse or subacute; plants herbaceous or with short (5-15 cm.) lengths of the previous year’s stems becoming slightly woody, sometimes with chambered pith. C. Plants herbaceous; heads 3.5-6 cm. diam. ........ AUN TUM cel Ir PM EC 9a. D. dissecta var. dissecta. C. Plants subherbaceous, short (5-15 cm.) lengths of the pre- vious year's stems becoming slightly woody (or with some stems herbaceous, having developed directly from the tu- berous rootstock); heads 5.5-8.5 cm. diam. ............................ E ISIS bare ietesssceosundeccbes 9b. D. dissecta var. sublignosa, A. Outer involucral bracts 9-18 mm. long; petioles solid or hollow. D. Stems living more than one season and branching to produce the current season’s growth; petioles solid. E. Ultimate segments or their lobes (whichever are narrower) of median leaves 0.5-3.2 mm. wide. ............ 8. D. rupicola. E. Ultimate segments of median leaves 5-16 mm. wide. ............ besescceeeceesesteestettttstttessetssatesstssttttttsseee De D. seapigeroides. D. Stems living only one season, the current season’s growth arising from a crown or tuberous rootstock beneath the soil; petioles solid or hollow. F. Petioles solid. ................. esee 6. D. foeniculifolia. [F. Petioles hollow. ........ 10. D. merckii (of Sect. 4. Dahlia)] Section 4. Dahlia A. Outer involucral bracts reflexed at anthesis (sometimes merely spreading but not ascending). B. Median leaves compound or if simple then the petiole more than 3 mm. long. C. Rays white, whitish-lavender, light purple to dark blackish purple. D. Current year's growth branching from ligneous stems of the previous year. ............................ 11. D. apiculata. 324 Rhodora [Vol. 71 D. Current year's growth from subterranean tubers or rootstock. E. Petioles broadly auriculate-winged at the base, 1-4.2 em. wide, upper surface of leaves glabrous or es- sentially so, lower surface pilose or merely pubescent; stem 6-furrowed,. ................ a... 15. D. pteropoda. E. Petioles without wings or if winged then not form- ing broad auricles at the base. F. Leaves and stems densely soft pubescent, pilose on lower surfaces of leaves (hairs diminishing on branches of the inflorescence and on parts of the heads). ................. r... 20. D. mollis. F. Leaves and stems not densely pubescent nor pilose. G. Outer involucral bracts flagelliform with long- attenuate tips, 10-17 mm. long, broadest at or near the base, 1.5-4 mm. wide; achenes 6-8 mm. long, constricted slightly below the sum- mit to form a black disc less than 1 mm. wide and barely 0.4 mm. high. ........ 18. D. moorei. Outer involucral bracts not long flagelliform, sometimes acuminate, but the tips not long attenuate, broadest at or beyond the middle. H. Outer involucral bracts 14-24 mm. long, dorsally glabrous. I. Rachis, petiole and _ petiolule usually winged; the terminal leaflet or segment with (11-)12-15 teeth per side; blades flexible when dry, surfaces not strongly bi-colored; outer bracts (4-)5-10 mm. wide. ...2.2.2. eee 17. D. rudis. . Rachis, petiole and petiolule not con- spicuously winged; the terminal leaflet or segment with 3-9(-11) teeth per side; leaflets brittle and stiff when dry, sur- faces bi-colored, the lower usually lighter green; outer bracts 2-6 mm. wide. .... 22d. D. australis var. serratior. H. Outer involucral bracts 2-12(-14) mm. long, if longer then dorsally pilose. J. Petioles solid, more than 1.2 cm. long. K. Outer involucral bracts glabrous on the dorsal suface or with a few scattered hairs. L. Rays drying to dark blackish purple; outer bracts ventrally appressed pubescent; leaflets (ex- 1969] Dahlia — Sorensen 325 cept sometimes the terminal leaf- let or segment) rounded at the base; hairs along the midrib be- neath widely spaced, 12 or fewer per cm. ..... 21. D. atropurpurea. L. Rays drying to rose or light pur- ple. M. Leaves coarse textured, often scabrous, simple or more com- monly pinnatisect to pinnate- pinnatisect; rachis, petiole and petiolule winged, often broadly so; margins coarsely toothed, the teeth protruding beyond the outline of the blades; basal pinna or lobe longer than the petiole. ........ NC er 14. D. pinnata. M. Leaves upon drying becoming stiff but not coarse textured nor scabrous, 1- or 2- pinnate. N. Upper surface of leaves rugose, glabrous or a few scattered hairs near the apex, lower surface con- spicuously lighter green or silvery green; margins revolute or obscurely so; native in Oaxaca, Chia- pas and Guatemala. ........ PENES 22. D. australis incl. vars. chiapensis & serratior. N. Upper surface of leaves smooth, not rugose, with scattered or broadly spaced antrorsely curved hairs and lines of very minute hairs following the main veins; lower surface not or but slightly lighter green; margins flat; native in Chihuahua, Durango and northern ZACHRUBC REUS eL en 326 Rhodora [Vol. 71 K. Outer involucral bracts copiously hairy or pilose on the dorsal surface. MII sa 19. D. hintonii. J. Petioles hollow or if solid then 0.8-1.2 em. long. O. Leaves (including petiole) 14-38 cm. long; petiole hollow, 4-16 cm. long. 1. nn 10. D. mercku. O. Leaves (including petiole) 5.5-8 cm. long; petiole solid, 0.8-1.2 cm. long. "m 22b. D. australis var. liebmanii. C. Rays yellow, orange, orange-scarlet or deep blackish scarlet. P. Leaves (including petiole), 6-11 cm. long, pinnate-pin- natifid or pinnate-pinnatisect; plants 2.5-6 dm. tall; rays yellow. ...................... eene 26. D. tenuis. P. Leaves (including petiole), 12-35 cm. long, simple to 3- pinnate; plants 4.5-30 dm. tall; rays yellow, orange, orange-scarlet or deep blackish scarlet. .................. IAS AAA a 27. D. coccinea. B. Median leaves simple, not lobed nor divided, sessile or the petioles 1-3 mm. long. Q. Base of blade clasping the stem, the leaves sessile; margins with 5-8 teeth per side. ........................ 12. D. cardiophylla. Q. Base of blade narrowed or rounded, the leaves sessile or short stalked, the petiole 1-3 mm. long; margins with 10-18 teeth per side. ............ een 13. D. purpusii. A. Outer involucral bracts erect or at least ascending at anthesis. R. Leaves without stipels; outer bracts 4-7 mm. wide. ................-. deassnesessansenccacessesseresaresereevesecsisussksneuassanseassesseansoovessensennns= 16. D. brevis. R. Leaves usually with stipels present at the 1st and/or 2nd rachis node; outer bracts 1.8-3 mm. wide. S. Outer bracts 6-6.5 mm. long; leaves (esp. on the under- surface and stems conspicuously hairy, the hairs shiny white, 2-5 mm. long, clearly multi-cellular; leaflets glossy dark green and rugose above. .............— 25. D. barkerae. S. Outer bracts 7-11 mm. long; leaves and stems glabrous or sparsely pubescent at the nodes on the stem and rachis, sometimes also dorsally along veins of leaflets; leaflets pale green and rugose above. ........ee A 24. D. scapigera. Section 1. Pseudodendron Sherff, Am. Jour. Bot. 38: 69. 1951. Type species: Dahlia lehmannii Hieron. (— Dahlia, imperialis Roezl ex Ortgies) Suffrutescent perennials 2-6 (-9) m. tall, stems becoming woody and often lasting more than one season, the older ones (1.5-)2-10 cm. 1969] Dahlia — Sorensen 327 diam., internodes hollow, nodes strongly septate, leaves 1-, 2- or 3-pinnate, the principal ones 20-90 cm. long, apices of ultimate seg- ments long attenuate, midveins tan colored beneath, heads obliquely erect or pendulous, numerous, rays white or whitish-lavender to light or rosaceous purple. (Species 1-3.) 1. Dahlia excelsa Benth. in Maund, Botanist. 2: pl. 88. 1838. TYPE: Mexico: Valley of México (the exact locality not stated), 7 Oct. 1834, William Bates s.n. (Holotype: K; photo. of type: F!). Dahlia excelsa var. anemonaeflora Benth. Ibid. Type: that of D. excelsa Benth. Dahlia, variabilis var. excelsa (Benth.) Walpers, Suppl. Prim. ad Rep. Bot, Syst. 2: 602. 1843. Type: that of D. excelsa Benth. *Roots fascicled, some perennial, cylindrical, and fibrous, others swelling into tubers. STEM perennial, very thick, becoming woody, growing to the height of twenty feet and upwards, less branched, and assuming more the aspect of a tree than any other species, hollow inside, smooth and glaucous on the surface, marked with horizontal rings formed by the broad stem-clasping base of the petioles, and sometimes emitting near the base a great quantity of fibrous roots. LEAVES opposite, attaining the length of two feet and a half, by about two feet in breadth, doubly pinnatipartite, the general petioles broadly connate round the stem, the segments borne on sharp partial footstalks, those of the lower leaves ovate and heart-shaped at the base, those of the upper leaves, especially the end ones, often contracted at the base, acuminate, toothed, nearly smooth or with a few short scattered hairs, of a pale glaucous green under- neath. FLOWER HEADS on long opposite monocephalous peduncles, collected five to eight together in a sort of corymb at the end of the branches, with occasionally a few axillary solitary ones along the stem. OUTER INVOLUCRE consisting of five or six linear spreading foliaceous squamae, the inner one of about twelve, oblong, obtuse, erect, membranous ones. FLORETS in the specimen we have seen all altered from their natural state, so as to give to the head of flowers the form known in other Dahlias by the name of anemone-flowered. 328 Rhodora [Vol. 71 The florets of the ray appear to be nearly in their ordinary state, and to be naturally neutral and sterile, those of the disk are shorter but all converted into irregularly formed semi-ligulate sterile florets having lost their original colour to assume that of the ray. All traces of the organs of fructification are obliterated in all the florets.” The foregoing is quoted directly from Bentham (1838) who based his description of Dahlia excelsa on living speci- mens. For the present study the only material I have been able to examine are two photographs (housed at F) of the type specimens. Sherff (1951a) writes of having collected Dahlia excelsa in two places in Guatemala, September, 1949, under his numbers 5061, “Chichicastenango . . . garden at the south side of Santo Thomas church," and 5062, “road- side thicket above Sololá." In an effort to obtain additional specimens for study I revisited these localities, in the com- pany of Dr, T. E. Melchert and Mr. D. J. Crawford, during late summer, 1966. Our search for this species was unsuc- cessful. At the garden of the Santo Tomás church we ob- served a large plant of Dahlia imperialis. It resembled the typical wild plants of that species we had seen in great numbers along roadsides in the general vicinity of Sololá earlier that day. Later an effort was made to locate Sherff's collections of Dahlia excelsa referred to above. It was expected these would be deposited in the herbarium of the Field Museum in Chicago. Other Dahlia specimens were found there which Sherff had collected in Guatemala during the same trip, September, 1949, but his numbers 5061 and 5062 were not found. Nevertheless, I have assumed that Sherff did, in fact, have the benefit of additional material of Dahlia excelsa on which to base his amplified description which he offers of the species in North American Flora (Sherff, 1955). From his description the following salient features may be added to Bentham's original description given above: “|. the larger leaves 6-7.5 dm. long including the glabrate petiole (this more or less 2 dm. long), 4-6 dm. 1969] Dahlia Sorensen 329 wide, their primary divisions (pinnae) each with but 1-5 leaflets (the terminal one often irregularly 2-3-parted), . margins coarsely serrate (with up to 21 teeth on a side), .. . ultimate segments to 1.7 dm. long and about 7 cm. wide; flowering heads . . . about 9 cm. across, ... outer involeral bracts 5 or 6, linear . . . ligules lilac, oblong, about 3.5 cm. long, about 1.5 cm wide, . achenes not seen." Valley of México (?) southeastward to southwestern Guatemala above Sololá. In addition to the problem of the identity of Dahlia ex- celsa, there is the question of the place of origin of the type. Bentham (1838) writes, “The Dahlias are all native in the mountainous districts of México, where the present species Dahlia excelsa had probably been discovered some years ago, as it appears to have long been cultivated in the Botanic Garden of the City of México." As seen in its photograph, the type in the Hooker Herbarium (K) bears a handwritten label stating the plant was collected in the Valley of México, but there is no mention on the label of a botanic garden. One wonders, therefore, how Bentham could have known about its having “long been cultivated in the Botanic Garden of the City of México," except had not William Bates, the collector, so informed him. Although it is not specifically stated, the indication is that the original material was collected by Bates in a Botanie Garden and he, perhaps assuming the plant had been introduced from nearby moun- tains, merely noted on the label the “valley of México." Certainly this region of México is botanically the best- known in that country, yet no specimens from this area, of Dahlia excelsa or any other “tree-dahlia,” have been seen, except two specimens of Dahlia imperialis which have come from contemporary gardens. On the characteristics displayed in the photographs of the type, Dahlia excelsa may be distinquished from Dahlia imperialis by its 3-7 (not 9-15) primary leaflets and from the new Dahlia tenuicaulis by the overall length of its leaves, “6-7.5 dm. long including the . . . petiole” (not 20-43 330 Rhodora [Vol. 71 cm. long). Despite these distinctions and despite the un- fortunate lack of material by which to judge, I strongly feel that Dahlia excelsa and Dahlia imperialis are variants of the same species. Nevertheless, I have chosen to retain Dahlia excelsa in the present treatment on a provisional basis. 2. Dahlia imperialis Roezl ex Ortgies in Regel, Gartenflora 12: 243. 18683. TYPE: exact place and date of col- lection not known; the plants collected in México by Roezl, brought to Zurich where cultivated, then to Berlin, the Royal Botanic Gardens, where cultivated and described, Roezl s.n. (Holotype: B[?]). Dahlia arborea Hort. ex Regel, Cat. Pl. Horti Aksakoviano 1860. nomen nudum. Dahlia, maximiliana Hort. ex Hooker f. Bot. Mag. pl. 7655. 1899. This name also used in Gard. Chron. II. 11: 216 & 12: 525. 1879, but without adequate description. Lectotype: pl. 7655, Hooker f. loc. cit. Dahlia lehmannii Hieron. Bot. Jahrb. 19: 55. 1894. Type: Leh- mann. 4894, Colombia (Holotype: B[?]; Isotype: F!). Dahlia dumicola Klatt, Bot. Beibl. zur Leopoldina 6. 1895 & Bull. Soc. Bot. Belg. 35: 289. 1896. Lectotype: Pittier & Tondue 8469, Costa Rica (F!). Dahlia maxonii Safford, Jour. Washington Acad. 9: 371. 1919. Type: Maxon & Hay 3295, Alta Verapaz, Guatemala (Us!). Dahlia lehmannii var. leucantha Sherff, Am. Jour. Bot. 38: 70. 1951. Type: Austin Smith A-620, San Luis de Zarcero, Costa Rica (F!). Herbaceous or suffrutescent perennial, 2-6(-9) m. tall. Stems of the previous year becoming woody, 7-10 cm. diam.; stems of the current year essentially unbranched except in the flowering portion, 1-4 cm. diam., hollow but septate at the nodes, smooth, obscurely 6-angled, glaucous, green or reddish-green when fresh, when dry becoming minutely many-striate with conspicuous lenticels, surface of older portions rough and tan or brown with enlarged lenticels. Leaves 50- 90 cm. long, 2- or 3-pinnate; primary leaflets 11-15, basal pinnae of principal leaves 20-38 cm. long; simple basal pinnules of lowest pinnae 5.5-14 em. long, 3.5-6 em. wide, ovate to oblong elliptical, acuminate or rarely merely acute at tips, mostly tapered or rounded at base, sometimes truncate or subcordate, sessile or stalked, the stalks rarely exceeding 5 mm.; the upper surface green, with hairs of two kinds, those on the nerves minute and appressed, those between the nerves longer, erect or slightly bent, the lower surface lighter green 1969] Dahlia — Sorensen 331 or gray-green, reticulate veiny, with hairs long, flexuous, conspicu- ously multi-cellular, mostly following the nerves; margins obscurely or conspicuously ciliolate, regularly and more or less evenly serrate with (6-)10-18 teeth per side; stipels occurring irregularly, when present then usually at the fourth rachis node but sometimes at more proximal nodes as well, infrequently also at the first node on the basal rachilla; petiole 12-25 cm. long, adaxially flattened or longi- tudinally grooved, abaxially rounded, glabrous, solid or sometimes hollow, strongly dilated and fleshy at the base, connate with the petiole base of the opposite leaf to form a conspicuous cupule at the node; rachis glabrous except for a cluster of hairs at the junctions with the primary pinnae, adaxially grooved, becoming narrowly winged esp. in the distal portions. Heads numerous, sometimes 150- 300 per plant, obliquely erect to slightly nodding (pendulous in some horticultural clones), 9-15 em. diam., slightly campanulate at anthesis; outer involucral bracts reflexed at anthesis, 6-14 mm. long, 3-9 mm. wide, obovate or oblanceolate, obtuse or acute, rarely subacuminate, dorsally many-striate, glabrous, ventrally appressed pubescent, the hairs stramineous or brown; inner bracts 15-25 mm. long, 7-13 mm. wide, ovate, obtuse, many-lined, brown with hyaline margins; rays white or whitish-lavender to rose-purple, 3.5-6 cm. long, 1.5-3 cm. wide, elliptical or ovate, acute or subacute, denticulate, tube of the ligulate florets pubescent, the hairs sometimes gland-tipped; disc florets 128-172, yellow or yellow with reddish tips, 9-11 mm. long, corolla lobes reflexed and inwardly curled; style branches about 3 mm. long, less than 1 mm. wide, lanceolate-attenuate, sometimes abruptly rhomboid-broadened near the middle; chaff (in fruit) about 2 cm. long, brown, sometimes with reddish or purple tips; achenes 13-17 mm. long, 1.8-2.5 mm. wide, linear or slightly oblanceolate. Chromo- some number, n = 16. Rocky slopes and fields, 2500-9000 ft., central Chiapas southeast in Guatemala, San Salvador, Costa Rica, and Colombia. Flowering Sept.-Feb. Map 1. MEXICO. cHtrAPAS: from 3 to 10 mi. E. of San Cristóbal de las Casas along route 190, Breedlove 7300 (ps, F, MICH), Carlson 1613 (F), Melchert et al. 6462 (IA), 6464 (IA), Miranda 9188 (MEXU), Moore & Cetto 8189 (BH, MEXU), Nelson 3173 (GH, Us [2]); vicinity of Tenejapa, Breedlove 6814 (ps, F), 7233 (ps, F[2]); 5 mi. S. of Jitotél, Breedlove 8878 (ps, F); Cerro Ovando, Matuda 2233 (GH, MEXU, MICH, US); Cerro Male, Matuda 4628 (Mo, NY); near Moto- zintla, Matuda 15452 (MEXU). COSTA RICA. without specific lo- cality, Stork 2992 (F). ALAJUELA: San Luis de Zarcero, Austin Smith A620 (F, MO), H236 (F, MO), NY1369 (F, GH, NY). SAN JOSE: Las Nubes, Coronado, Echevarria 160 (F, UC), Stork 4722 (GH, MICH, NY, UC); Santa Marfa de Dota, Standley 41857 (GH, Us). EL SAL- 332 Rhodora [Vol. 71 a f f N l a PP " AR. X wow e `. ° 15N 4 °. ' , 10N ee N 6S VD a ssw a -+ SN . * 5e Map 1. Distribution of Dahlia imperialis Roezl ex Ortgies. VADOR. AHUACHAPAN: Apaneca, Calderon 2422 (F, US). SANTA ANA: Volcán de Santa Ana, Tucker 1297 (uc). GUATEMALA. ALTA VERAPAZ: vicinity of Cobán, Standley 69469 (F), 90887 (F), 91114 (F); Socoyocté, Maxon & Hay 3295 (Type of D. maxonii— US) ; above Santa Cruz, Standley 71026 (F). CHIMALTENANGO: Chichavac, Skutch 65 (US), 711 (GH); above Las Calderas, Standley 80260 (F). CHIQUIMULA: vicinity of El Barriol, Steyermark 30774 (F). GUATE- MALA: NE. of Guatemala City, White 5142 (MICH). HUEHUETE- NANGO: vicinity of San Mateo Ixtatán, Steyermark 49887 (F); Breedlove 8761 (DS, F); above Concepción, Goodman & Melhus 3691 (F, ISC); near Huehuetenango, Melchert et al. 6452 (1A); vicinity of Tunima, Steyermark 48418 (F). JALAPA: vicinity of Jalapa, Steyer- mark 32110 (F), 32406 (F), 32547 (F), Standley 77180 (F), 77228 (F). JUTIAPA: Volcán Suchitán, Steyermark 31903 (F). QUETZALTENANGO: vicinity of Santa Maria de Jesús, Standley 67162 (F); 84874 (F); slopes of Volcán Santa Maria, Steyermark 34414 (F), 34415 (F); slopes of 1969] Dahlia — Sorensen 333 Volcán Zunil, Standley 67468 (F), Steyermark 34676 (F); SE. of San Martin Chile Verde, Standley 83790 (F), SE. of Palestrina, Standley 84205 (F); along route 1, E. Quetzaltenango, Melchert et al. 6434 (IA). SACATEPEQUEZ: slopes of Voleán de Agua, Bell & Duke 16972 (MICH, UC), Hunnewell 14869 (GH), Standley 59425 (F); near Antigua, Kellerman 7096 (F, MICH, NY, US); San Rafael, J. D. Smith 2857 (F, US); above Las Calderas, Standley 59972 (F) barranca above Duenas, Standley 63214 (F). SAN MARCOS: along road be- tween San Marcos and San Rafael Pié de la Cuesta, Standley 85320 (F), 86289 (F, GH); Cenaguilla, Heyde & Lux 4239 (F, GH, MO, NY, US). SOLOLA: E. of Sololá along road above Lake Atitlan, Melchert et al. 6439 (IA), 6443 (IA). TOTONICAPAN: 5 mi. E. of Quetzalten- ango, Melchert et al. 6434 (14). COLOMBIA. carpias: Río Quindío, above Armenia, Pennell et al. 8677 (GH, NY, US). CAUCA: Vicinity of Popayán, Cuatrecasas 23407 (F), Fosberg 20290 (NY, US), Killip & Lehmann V 38056 (GH, UC, vs), Lehmann 4894 (Holotype of D. lehmannii — B [?], Isotype — F); Sneidern 5677 (US); El Tambo, Sneidern 1035 (F, GH, NY). CUNDINAMARCA: vicinity of Bogotá, Ping 157 (F, MO), Soejarto 281 (GH); between El Salto and El Colegio, Cuatrecasas 8185 (F, US) ; Taguate Valley, Grant 9652 (US); Subatá, Haught 6114 (US); near Charquito, Langenheim 3359 (uc); barranca del Rio Bogotá, Schultes et al. 4045 (US). HUILA: Rio Balsallas, Rusby & Pennell 929 (GH, NY, US). SANTANDER: E. slope of Paramo del Hatico, Killip & A. C. Smith 20558 (A, GH, NY, US); valley above Suratá, Killip & A. C. Smith 16530 (GH, NY, US). The following specimens from cultivation: GUATEMALA. SACATEPE- QUEZ: near Antigua, Sherff 5054 I, II, III, IV, V (F). MEXICO. DISTRITO FEDERAL: Martinez s'n. (MEXU), in Cetto Garden, Jardines Pedregal near Tizapán, Moore s.m. (BH). UNITED STATES. CALI- FORNIA: Golden Gate Park, San Francisco, Walther s.n. (CAS). IOWA: University of Iowa Experimental Gardens, Iowa City, Sorensen 6249 (1A). After having observed the so-called “tree-dahlias” in their native environment as well as having grown living plants in the gardens and the greenhouses at the Univer- sity of Iowa, I can find no sound biological basis for main- taining the several taxa into which they have been segre- gated previously. Dahlia, lehmannii var. leucantha was established on the basis of its whitish rather than lavender ligules and its geographic restriction to parts of Costa Rica, Reports from the field (Burger, personal communication, observations of Standley, 1938, and specimen label of Stork 4722 [GH, MICH, 334 Rhodora [Vol. 71 NY, UC]) indicate that ligule color in wild populations ranges from white to lavender or light purple. This sug- gests that genetic control of ligule color is most likely multi-factorial, the color-range due to random segregation of quantitative and/or qualitative factors. This same color range has also been observed in populations of D. merckii, dissecta, and scapigera. Sherff’s (1951 & 1955) recognition of Dahlia lehmannii as a separate species from D. imperialis stems from the fact that plants on which he based the latter name were from an unusual and widely cultivated horticultural clone.’ Individuals of this particular clone are, according to Sherff (1955), characterized by excessively drooping heads and smaller ultimate leaf segments (4-7 cm. not 5.5-14 cm. long). I was able to obtain cuttings from what I believe is the same clone mentioned by Sherff (footnote 10, 1951a). During the summer of 1966 these plants were grown along- side wild plants raised from seeds collected in the moun- tains of Chiapas (Breedlove 7424, voucher at DS). By mid-October all of the plants had attained a height of about 5.5 ft. and bore fully developed leaves. At this stage indi- viduals of the two groups were, on the basis of their total morphologic appearance, essentially indistinguishable from each other. Only detailed comparison of them revealed minor differences. For example, the unfolding leaves of the Chiapas plants produced a slight reddish pigmentation which was not seen in the cloned individuals. It was also observed that their mature leaflets were apically more acuminate and less pubescent on the upper surface. Unfor- tunately these plants were cut down by frost before reaching the flowering stage. A plant from the cultivated clone which had been planted in the earthen floor (not in a pot) of the *'tree-house" at the University of Iowa grew to maturity and flowered. Its heads were seen to be no more pendulous than those on "Its clonal nature is suggested by the information (Turlington, per- sonal communication,) that “D. imperialis has never set seed; it has always been propagated by cuttings or root division." 1969] Dahlia — Sorensen 335 plants observed in wild populations in Chiapas and Guate- mala. As each head on this plant neared the end of anthesis, an apparent abscission layer formed on the peduncle just beneath the head causing the heads to fall. This “‘flower- drop” is perhaps the result of the heads not having been suitably pollinated considering that this plant is repro- ductively self-incompatible. Such an hypothesis is, as yet, untested. Sherff noted that Dahlia imperialis (i.e., the cultivated clone) bore leaflets much smaller than the wild “tree- dahlias” (ie. D. lehmannii). I have observed that when these plants are grown in pots in the greenhouse their leaves tend to be smaller than when field grown or when grown in the earthen floor of a tree-house. It is quite likely that Sherff used a greenhouse-grown specimen for his de- scription. He writes (1951b) of having “raised noble specimens of this species [i.e.. D. imperialis sensu Sherff (1955)] for several years in the greenhouses of the Uni- versity of Chicago . . . and of the Chicago Teachers Col- lege." In my judgment the cultivated form of Dahlia imperialis does not differ sufficiently from the wild plants to warrant treating it as a distinct species. The minor differences displayed in the cultivated Individuals, as compared to those collected in the field, are of no greater magnitude than one finds from plant to plant within a reasonably large wild population in its native environment. Dahlia imperialis may be distinguished from the other "tree-dahlias" by its 9-15 (not 3-7) primary leaflets. It should perhaps be emphasized that median leaves, or at least a portion of them, are essential in determinations of these Species. 3. Dahlia tenuicaulis Sorensen sp. nov. TYPE: MEXICO: OAXACA: 13 mi. NE. of village of Ixtlán de Juárez along route 175, Oaxaca-Tuxtepec, in cloud forest, about 9000 ft. 25 Aug. 1965, Melchert & Sorensen 6194A (Holotype: IA!, 125,512 & 125,513). 336 Rhodora [Vol. 71 [ 1255312 | I nios s RATE UNIVERSITY OF KWA "d ce idi —fa ai i MEXICO Shull [ STATE OF qz i Fig. 2. Dahlia tenuicaulis Sorensen. Photograph of type (14), x 1/3. 1969] Dahlia — Sorensen S01 Suffrutex, 21-37 dm. altus. Caules lignosi anni praecedentis 1.5-3 cm. diametro. Folia pinnata vel bipinnata, 27-43 cm. longa; foliolis 3-7, pinnis basilaribus 10-18 cm. longis et petiolulis 2-4 cm. longis, marginibus regulatim serratis cum dentibus in quoque latere 8-14. Capitula 12 vel plura in quoque ramo principali, squammis exteriori- bus involucri reflexis sub anthesi, 8-12 mm. longis, 2-4 mm. latis. Flores ligulati lilacini, 2.8-4 cm. longi, 1-2 cm. lati; flores tubulosi 70-80. Achenia lineari-oblanceolata, 11-15 mm. longa, 1-2 mm. lata. Chromosomatum numerus: n = 16. Fig. 2. Cool cloud forests of Pinus, Abies and Arbutus, deep loamy soil, 8500-10,000 ft., from the type locality in Oaxaca westward on tall peaks to near Santa Monica (about 1049 30' w. long, 19? 15' n. lat.), Jalisco. Flowering Aug.-Oct. Map 2. Map 2. Distribution of Dahlia tenuicaulis Sorensen. MEXICO. MICHOACAN: Cerro Torricillas, ‘Hinton 12364 (GH, MICH, NY, US). JALISCO: slopes of Nevado de Colima, McVaugh 11745 (MICH), 12926 (MICH), Melchert et al. 6396 A-C (IA); Cerro de Manantlan, 15-20 mi. SE. of Autlan, McVaugh 13873 (MICH); near Santa Monica, McVaugh 14110 (MICH). OAXACA: slopes of Cerro San Felipe, Conzatti & Gonzalez 552 (GH, MEXU, US), Conzatti 2259 (F), Pringle 5875 (GH), C. L. Smith 295 (F, IA, MO, NY, US); 358 Rhodora [Vol. 71 from 10 to 13 mi. NE. of Ixtlan de Juárez, Melchert & Sorensen 6193 (IA); 9 mi. above La Soledad along road to Oaxaca, Moore 8247 (BH, MEXU) ; near K-152 below Suchixtepec, Moore 8250 (BH, MEXU) ; 18 mi. SW. of Oaxaca, Nelson 1355 (us); Cerro de Pluma, vicinity of Suchixtepec, Reko 3716 (US). GUERRERO: vicinity of Cerro Teo- tepec, Hinton 11083 (GH), 11137 (GH, MICH), Rzedowski & McVaugh 163 (MICH); 40 k. W. of Chilpancingo, Feddema 2746 (MICH); summit of Sierra Madre near Chilpancingo, Nelson 2193 (us); river gorge near Omiltemi, Sharp 441534 (F). Among the "tree-dahlias" Dahlia tenuicaulis is the only species endemic in México. It is most easily distinguished from D. excelsa and D. imperialis by the overall smaller size of its leaves (27-43 em. not 50-90 em. long). From D. imperialis it is further distinguished by its 1- or 2-pin- nate (not 3-pinnate) leaves, by its fewer number of disc florets (70-80 not 120-170) and by its rather slender ligne- ous stems which barely exceed 2.5 em. diam. (not 4-10 cm. diam.). There is complete overlap between the species in the size and long-attenuate apices of their leaflets or ulti- mate segments. In the field it was observed that some individuals of D. imperialis produced their current season's growth directly from tuberous roots not exclusively from the previous year's stems as does D. tenuicaulis. Unlike D. imperialis, which is usually found in open disturbed areas along roadsides and in fields, D. tenuicaulis appears to be restricted to dense Pinus, Abies and Arbutus cloud forests at elevations of 8,500 to 10,000 ft. Section 2. Epiphytum Sherff, Bot. Leafl. 4: 21. 1951. Type species: Dahlia macdougallii Sherff. Plants epiphytic on tall trees, the stems clongate, solid, ligneous, the roots elongate, aérial, reaching down into the earth. (Species 4.) 4. Dahlia macdougallii Sherff, Bot. Leafl. 2: 5. 1950, & Chicago Nat. Hist. Mus. Bull. 22(6) : 5-7, illus. 1951. TYPE: MEXICO: OAXACA: about 50 mi. W. of Tehu- antepec, 7000-7500 ft., in rain forest above town of Santo Tomás Teipan, Nov. 1949, MacDougall s.n. (Holotype: F, sheet number 1,449,390; Isotype: F!). 1969] Dahlia — Sorensen 339 Perennial, epiphytic on tall trees. Stems solid and woody, about 5 em. diam., reaching about 10 m. to the tree tops where the flowering branches sprawl for considerable distances. Aérial roots about 5 cm. thick, up to 22 m. long. Uppermost leaves (median or principal leaves not seen) pinnate, 18-27 cm. long; leaflets 3-5, basal pinnae 13-15 em. long, 4-5 cm. wide, ovate or elliptic-ovate, long-acuminate, the base of principal pinnae at least broadly rounded or almost cor- date, sometimes asymmetric bearing a sessile lobe on the proximal side at the base of the main blade; surfaces bicolored, dark green above, gray-green and conspicuously veiny beneath, the upper pubes- cent along the principal veins and near margins, the lower almost glabrous but with a cluster of brown hairs at the junctions of the main veins; margins ciliolate, the cilia soft and flexuous, antrorsely curved, sharply and regularly serrate-dentate with 15-23 teeth per side; petiole 4.5-10 em. long, flattened above and rounded below, semi- circular in cross-section, essentially glabrous; rachis glabrous; stipels present at median rachis nodes and sometimes at the base of the ter- minal leaflet when the latter is not decurrent-winged. Heads numerous, at least as many as 14 per main flowering branch, erect or obliquely erect; outer involucral bracts spreading or reflexed at anthesis, 10-14 mm. long, 2-3 mm. wide, lanceolate or rarely oblanceolate, acute or acuminate, dorsally many-lined, glabrous, ventrally appressed pubes- cent; inner bracts 12-15 mm. long, 3-6 mm. wide; rays white or at least whitish, 4.2-5 em. long, 1.2-2.2 em. wide, elliptical or oblong lanceolate, mostly acute at the apex; disc florets yellow; achenes not seen. I have seen specimens from the type locality only, but Sherff (1951b) states that Thomas MacDougall has ob- served the species on Cerro Chivato in the territory of the village of San José Quianite which is in the same general area as the type locality. Flowering Nov. MEXICO. OAXACA: above town of Santo Tomas Teipan, MacDougall 21984 (Topotype: NY). Though not as large, the leaves of Dahlia macdougallii bear a strong resemblance to those of the 'tree-dahlias." The manner of arrangement of the heads in corymbose clusters at the end of flowering branches is also a feature this species shares with D. imperialis and D. tenuicaulis. Its unique habit, however, distinguishes it from all other known species. Sherff (1962) states that some European scientists have been at work on the cytology of D. macdougallii but in a 340 Rhodora [Vol. 71 personal communication with him and with the directors of the botanic gardens at Stockholm and Goteborg, Sweden, there is no indication that this work has ever been carried forward. Section 3. Entemophyllon Sorensen, sect. nov. Type species: Dahlia dissecta S. Wats. Herba perennis vel suffrutex, 6-15 dm. alta, Caules numerosis e radicibus tuberosis, plus minusve perennis, lignescentes, 4-20 mm. diametro, caules novi 1.5-12 mm. diametro, saepe cum medulla dis- sepimenta vel cum dissepimentis incompletis. Folia subtiliter dissecta, pinnati-pinnatisecta vel tripinnati-pinnatisecta cum foliolis oppositis et pinnulis alternis. Squammae exteriores involucri erectae vel ascen- dentes sub anthesi. Flores ligulati albi, albido-lavanduli vel lilacini, inferne saepe cum macula lutea vel luteo-viridi. Achenia minute tuberculata. Chromosomatum numerus: m — 17. (Species 5-9) (see Map 3). Species 5-9 which comprise the new section Entemo- phyllon form a coherent evolutionary series within the genus Dahlia. Members of this section share a number of characteristics either rare or absent among the other species of the genus. Cytological information has been gathered for three of the five taxa of this section: Dahlia dissecta var. dissecta, D. foeniculifolia, and the new D. rupicola. All have a meiotic chromosome number of n = 17, a number not re- ported previously for the genus (cf. footnote 1). Pairing appears to be completely regular, forming 17 perfect biva- lents at diakinesis. In habit of growth all of the species produce many stems arising from a common tuberously thickened rootstock. This is in marked contrast to the usual single-stemmed (or in sec- tion Pseudodendron, 3-4-stemmed) habit of all other species of the genus. The stems of several species in section Entemo- phyllon live for more than one season becoming quite woody and giving full grown plants a shrubby aspect. Their wood, however, remains rather weak and brittle and the stems rarely exceed 1.5 cm. diam. Pith in the stems of the cur- rent season’s growth (of all taxa except D. foeniculifolia and D. dissecta var. dissecta) is solid in the early stages 1969] Dahlia — Sorensen 341 becoming chambered in age or sometimes the diaphragms (about 3 mm. apart) themselves disintegrate leaving be- hind annular rudiments on the wall of the pith cavity. When crushed or bruised, the herbage of these species emits a particularly strong carroty odor. The segmentation of their leaves is probably the most remarkable characteristic of these species. They all are rather highly dissected (somewhat less so in D. scapigeroides) and are at least pinnate-pinnatisect or more often bipinnate to bipinnate- pinnatisect. The primary leaflets are arranged oppositely on the rachis, but the pinnules of the secondary and lesser ranks are arranged alternately on their rachillae. This characteristic is found with the same consistency in just one other species, D. merckiü. A rather interesting occurrence attending the circum- scription of section Entemophyllon took place while I was preparing for a chromatographic analysis of the secondary biochemical compounds in the leaves of these species. As a first step in this technique, dried leaves were crushed and placed in a flask with acidified methyl alcohol (0.5% or 196 conc. HC1, V:V). After about 12 hours on an orbital shaker the liquid turned milky-green. Upon filtration it was seen that the milky color had been caused by the forma- tion of a water-white, crystalline compound. This precipi- tate apparently is limited within the genus to the six taxa of section Entemophyllon. The precise nature of the com- pound remains to be determined, but efforts toward its elucidation are presently in progress. 5. Dahlia scapigeroides Sherff, Am. Jour. Bot. 34: 145. 1947. TYPE: MEXICO: HIDALGO: near Ixmiquilpan, Aug. 1905, Rose, Painter & Rose 9157 (Holotype: GH!; photo. F!; Isotype: US!). Dahlia scapigeroides var. typica Sherff, Am. Jour. Bot. 34: 145. 1947. Type: as for Dahlia scapigeroides Sherff. Suffruticose perennial reaching 24.5 dm. tall. Stems of the previous season 6-8 mm. diam.; pith chambered or the diaphragms incomplete with only annular rudiments of the septa on the inside wall of the 542 Rhodora [Vol. 71 pith cavity; exterior surface glabrous, pruinose, longitudinally striate, the conspicuous lenticels orbicular or vertically elongate; stems of the current year 5-7 mm. diam., longitudinally striate, glabrous, pruinose; internodes 2-3 em. long. Leaves pinnate-pinnatisect to bi- pinnate, 10-16 em. long, leaflets 5-7, opposite on the rachis, basal pinnae 5-9.5 cm. long, sessile or stalked, petiolules 2-11 mm. long, ultimate segments alternate on the rachilla, 1-2.5 em. long, 5-16 mm. wide, ovate-lanceolate, acute; surfaces more or less glabrous, occa- sionally with a few very fine scattered hairs below; margins eciliolate, irregularly and coarsely serrate, 1-4 teeth per side; stipels 0.6-4.5 em. long, 2-15 mm. wide, ovate, acute, occurring irregularly, when present then only at the junctions of the petiolulate pinnae; petiole 5-7 em. long, crescent-shaped in cross-section, grooved above; rachis groved above, glabrous except for a cluster of hairs at the junctions with the pinnae. Heads obliquely erect at anthesis, 1-2 per branch on peduncles 9-14 cm. long; outer involucra] bracts erect at anthesis, 14-17 mm. long, 5-11 mm. wide, broadly ovate, obtuse, 4-8-lined, the nerves blackish, sometimes faintly reticulate veiny, ventrally puberulent, dorsally glabrous; inner bracts 15-18 mm. long, 6-10 mm. wide; chaff (in fruit) 18-20 mm. long, 5-8 mm. wide; rays lavender or rose-purple, 2.5-4 em. long, 14-18 mm. wide, ovate, acute or denti- culate; dise florets yellow; style branches 2.8-3 mm. long, less than 0.5 mm. wide; achenes 12-15 mm. long, 1.7-2 mm. wide, linear, slightly wider above the middle, longitudinally suleate, surfaces beset with numerous minute tuberculae; pappus absent or consisting of 2 minute, squarrose rudiments usually less than 1 mm. long. In dry, rocky places among thorny scrub, 4700-6500 ft., western Hidalgo. Flowering June-Sep. Map 3, open circles. MEXICO. HIDALGO: canons, vicinity of Ixmiquilpan, Purpus 1341 (F, MO, NY, UC), 5577 (Uc);Barranea Grande de Metztitlán, Moore 3097 (BH, GH); barranca between Jacala and Tepetates, Moore 5036 (BH); near highway SE. of Jacala, Moore & Wood 3932 (BH, GH, NY, uc): along road from Zimapán to Mina Loma del Toro, Moore & Wood 4407 (BH, GH). When Sherff (1947) established Dahlia scapigeroides and its variety apiculata (i.e., D. apiculata) he referred to their similarity in growth habit with D. seapigera. He noted especially the shortened internodes with leaves crowded near the base of the stem. D. scapigera, however, is herbaceous (not suffrutescent) and its current year’s growth is from underground tubers, not from ligneous stems of the previous year as in D. scapigeroides. In addition to its woody stems, several other character- 1969] Dahlia — Sorensen 343 © DAHLIA SCAPIGEROIDES @ DAHLIA POENICULIFOLIA A DAHLIA LINEARI @ DAHLIA RUPICOLA @ DAHLIA DISSECTA VAR. SSECTA Hr DAHLIA DISSECTA VAR. SUBLIGNOSA Map 3. Distribution of the species comprising Section 3, Ente- mophyllon: Dahlia scapigeroides Sherff, open circles; Dahlia foeni- culifolia Sherff, solid diamonds; Dahlia linearis Sherff, solid triangle; Dahlia rupicola Sorensen, solid squares; Dahlia dissecta var. dis- secta, closed circles; Dahlia dissecta var. sublignosa Sorensen, stars. istics of D. scapigeroides (e.g. chambered pith, alternate pinnules, a crystalline precipitate in an acidified methanolic extract of its leaves) clearly place it in the section Ente- mophyllon. It may be distinguished from other members of this section on the following characteristics: from D. dissecta vars. and D. linearis on the size of its outer in- volucral bracts (5-11 mm. not 1.8-5 mm. wide) ; and from D. foeniculifolia and D. rupicola on the width of its ulti- mate segments (5-16 mm. not 0.5-2 mm. wide). 6. Dahlia foeniculifolia Sherff, Am. Jour. Bot. 38: 70. 1951. TYPE: MEXICO: NUEVO LEÓN : Dulces, Nombres, open rocky slope, 2000 m., 8 Aug. 1948, Meyer & Rogers 2881 (Holotype: Us!; Isotypes: F!, GH!, MICH !, MO}, US!). Herbaceous perennial, producing several canes 14-19 cm. tall from a tuberously thickened rootstock. Stems 5-12 mm. diam., glabrous 344 Rhodora [Vol. 71 except for a cluster of hairs at the nodes, light reddish-brown or purple; internodes 8.5-13.5 em. long, many-striate. Leaves bipinnate- pinnatisect, 8-33 cm. long (including petiole)*; primary leaflets 7-17, the basal pinnae 15-19 cm. long, their pinnules alternate on the rachilla, the ultimate segments narrowly linear 0.5-2 mm. wide, tips long-attenuate, one-nerved, pubescent in the young foli- age, mature leaves minutely pubescent or glabrate, green on both sides; margins smooth; petiole from less than 0.5 cm. to 3.5 cm. long; rachis essentially glabrous except for a cluster of hairs at the junction with the pinnae. Heads few, 1-4 per stem, obliquely erect, elevated on peduncles 8-15 cm. long; outer involucral bracts erect at anthesis, ovate to oblong-ovate, 14-18 mm. long, 8-11 mm. wide, conspicuously several to many-purple-lined, the lines running almost to the obtuse or acute tips, dorsally sparingly pubescent or glabrous, ventrally puberulent; inner bracts 15-18 mm. long, 6-9 mm. wide; rays lavender or light reddish-purple, 3.5-6 cm. long, 1.6-2.1 em. wide, oblong-ovate, acute or denticulate; dise florets yellow, the anthers yellow and long exserted; achenes not seen. Rocky slopes and ledges, about 6500 ft., Nuevo León and Tamaulipas. Flowering June-Aug. Map 3, solid diamonds. MEXICO. TAMAULIPAS: 3-5 km. S. of Huitsachél, 6000 ft., Stan- ford et al. 2130 (F, NY, TEX, Us). Dahlia foeniculifolia shares with D. dissecta var. dissecta its herbaceous habit and the absence of chambered pith. The two are easily distinguished on the dimensions of their outer involucral bracts (14-18 mm. long in the former, 5-7 mm. long in the latter, including also var. sublignosa). The outer bracts of D. foeniculifolia compare in size to those of D. scapigeroides from which it is separated by the much narrower ultimate segments of its leaves (0.5-2 mm. not 5-16 mm. wide) and its herbaceous habit. The highly dis- sected leaves with very narrow ultimate segments of D. foeniculifolia resemble strongly those of the allied D. rupi- cola and D. linearis. These species, though apparently both quite closely related to D. foeniculifolia, have a suffrutescent growth habit. 7. Dahlia linearis Sherff, Bot, Gaz. 89: 364. 1930. TYPE: MEXICO: QUERÉTARO: Querétaro (the exact locality "Only one specimen (Meyer & Rogers 2881 at F) has complete median leaves measuring 29-33 cm.; other specimens I have examined possess only leaves from the flowering portion of the stems, these measure 8-15 cm. long. 1969] Dahlia — Sorensen 345 not given), 1850 m., 1910-1913, Agnie/ 10286 (Holo- type: Mo!, photo. F!; Isotype: US!). Suffruticose perennial 8-13 dm. tall, freely branching from tuber- ously-thickened roots and ligneous stems of the year before. Stems of the previous year 4-8 mm. diam., longitudinally striate, some parts covered with a scaly gray bloom; lenticels conspicuous, vertically oblong, the larger ones 1-2 mm. long; pith chambered; stems of the current year 1.5-5 mm. diam., glabrous, many sulcate, markedly pruinose esp. in the middle portions, the pith solid in younger parts, in older parts becoming chambered; median internodes 3.5-11 cm. long. Principal leaves pinnate-pinnatisect to bipinnate-pinnatisect, 12-18 em. long; primary leaflets 7-9, their pinnules alternate on the rachilla, the basal pinnae 3.5-9 cm, long; ultimate segments 0.9-2.3 em. long, their lobes 1-5.5 mm. wide, linear-lanceolate, one-nerved, tips long attenuate-mucronate; surfaces glabrous or with a line of fine hairs along the veins above; stipels usually present on median leaves, attached at the basal and second rachis nodes, 1-5 cm. long; margins smooth; petiole 4-6.7 cm. long, glabrous or with two lines of fine hairs above in the proximal portion; rachis grooved above, minutely hispidulous or glabrous but usually with a cluster of hairs at the junction with the pinnae. Heads obliquely erect at anthesis, usually 2-6 on a stem, on peduncles 1.5-10 cm. long; outer involucral bracts 5-9 mm. long, 2-3.5 mm. wide, oblong-lanceolate, tips acute or subacuminate, dorsally glabrous, green, 4-7 purple-lined, ventrally sparsely pubescent; inner bracts 11-15 mm. long, 4-8 mm. wide, usually glabrous but sometimes with fine hairs near the base; chaff (in fruit) 1012.5 mm. long, 4-6 mm. wide, stramineous with a smudge of yellow near tip; rays purple with a yellow spot at base, 1-2.1 cm. long, 0.9-1.3 cm. wide, ovate, acute or denticulate; disc florets 43-48, yellow, anthers conspicuous; achenes 8-11 mm. long, 1.5-2 mm. wide, linear-oblanceolate, blackish, minutely puberulent and tuberculate, obscurely many-suleate; nappus obsclete or consisting of 2 erect or squarrose rudiments. Chromosome number, n — 17. Of uncertain extent (the only exact locality being that of Melchert & Sorensen 6240 below), on dry ledges and cliffs 5500-6500 ft. in the vicinity of Querétaro City, Querétaro, México. Flowering Aug. Map 3, closed triangle. MEXICO. QUERETARO: Arsène 10083 (us); 5 mi. N. of Ciudad Querétaro along route 57, 150 yds. E. of highway, Melchert & Soren- sen 6240 (IA). On the shape and segmentation of its leaves and leaflets Dahlia linearis seems closely allied to D. rupicola and D. foeniculifolia. Its suffrutescent habit distinguishes it from the herbaceous D. foeniculifolia, From D. rupicola, appar- 346 Rhodora [Vol. 71 ently its closest ally, it separates on such technical features as the size of its outer involucral bracts (5-9 mm. not 10-14 mm. long and 2-3.5 mm. not 4-7 mm. wide), the smaller diam. of its heads (3-4.2 em. not 5-6.5 cm., including the rays) and the overall dimensions of its median leaves (12-18 em. not 20-35 em. long). 8. Dahlia rupicola Sorensen, sp. nov. TYPE: Mexico: DURANGO: 11 mi. W. of La Ciudad along route 40, about 6000 ft, 5 Sep. 1966, Melchert, Sorensen & Crawford 6306 (Holotype: IA!). Suffrutex, 9-15 dm. altus. Caules anni praecedentis lignosi 1.5- 2 cm. diametro, lenticellis prominentibus verticaliter elongatis; caules novi 6-11 mm. diametro, glabri (et) pruinosi (que). Folia bipinnati- pinnatisecta, 20-35 em, longa, foliolis primariis 9-11, tribus paribus inferioribus manifeste petiolulatis, pinnulis alternis, pinnis basilaribus 12-18 cm. longis, libis segmentorum ultimorum 0.6-2 mm. latis, lineari- attenuatis, uninervatis, apice flexuoso-acicularibus; stipellis 8-12 em. longis, plerumque (nonnisi) in nodis basilaribus et secundis rachidi affixis. Capitula 2-4; pedunculis 6-11 cm. longis; Squamis ex- terioribus involucri late ovatis, acutis vel acuminatis, erectis vel leviter potentibus sub anthesi, in dorso 5-8-lineatis, marginibus an- guste albo-hyalinis, 10-14 mm. longis, 4-7 mm. latis. Flores ligulati lilacini, 2.2-3 em. longi, 7-11 mm. lati; flores tubulosi 85-109, lobis corollae reflexis sub anthesi. Achenia linearia, dilatata supra medium, 10-12 mm. longa, 1.5-2 mm. lata. Chromosomatum numerus: n = 17. Fig. 3. Abundant along cliffs and rock ledges for a distance of about 8 miles from the type locality westward along route 40, Durango, México. Flowering Sep. Map 3, squares. MEXICO. DURANGO: 16 mi. W. of La Ciudad, about 15 mi. E. of El Palmito, Sinaloa, 5800-6000 ft., 5 Sep. 1966, Melchert et al. 6312 (1A). Dahlia rupicola is extremely abundant along the main east-west highway through Durango. It is not surprising that it has remained uncollected so long as most of the plants were growing in crevices of exposed cliffs 50-150 ft. above the highway and largely inaccessible either from above or below. When we first encountered this large popu- lation of the species we despaired of ever reaching any of the plants, but it was seen that some could be reached from a small ledge midway up the cliff. A cooperative effort with 1969] Dahlia — Sorensen 347 MEXICO: stare OR DURANG rapicota X in Thomas E. Malchert, Pavi D. Sorensen, Daniel J, Crawinré Ma 6 11) Taxonomic Studies in the Corsepsidiese © Sepi 1065 Herbarium of The University nf lowe á Fig. 3. Dahlia rupicola Sorensen. Photograph of type (IA), x 1/3. 348 Rhodora [Vol. 71 one person doing the climbing and the others with binocu- lars in hand calling out directions from below finally suc- ceeded. At a station farther along the highway we found a fortuitous landslide had carried down several undamaged plants which we also collected. With one or more of the species of section Entemophyllon Dahlia rupicola shares a shrubby habit, pinnules arranged alternately on the rachilla, finely dissected leaves, a meiotic chromosome number of n = 17, and the presence of a white crystalline precipitate in an acidified methanolic ex- tract of its leaves. Dahlia rupicola most resembles D. linearis and D. foeni- culifolia. These three species all have very finely dissected leaves with rather narrow and long-attenuate ultimate segments. Dahlia rupicola is readily distinguished from D. foeniculifolia by its woody habit and from D. linearis, its closest ally, on such technical features as the dimensions of its outer involucral bracts (10-14 mm. not 5-9 mm. long and 4-7 mm. not 2-3.5 mm. wide), the greater diameter of its heads (5-6.5 cm. not 3-4.2 em. diam. including the rays) and its larger leaves (20-35 cm. not 12-18 cm. long). 9. Dahlia dissecta S. Wats. Proc. Am. Acad. 26: 141. 1891. TYPE: MEXICO: SAN LUIS POTOSÍ: limestone ledges, San José Pass, July 1890, Pringle 3167 (Holo- type: US!; Isotypes: F!, GH!, ISC!, MO!, MSC!, NY!, RSA!, UC!, US!). Freely branching perennial herb, 3-9 dm, tall, a cluster of many herbaceous stems arising from a large, horizontally-elongate ligneous tuber or rootstock, or the current season's stems branching from short (5-15 cm.) woody portions of the previous year's stems, Stems slender, terete, 1.5-4 mm. diam., glabrous, green or reddish-brown when fresh, drying to pale green or tan; internodes 1.5-16 cm. long, the upper solid, the lower hollow, the pith sometimes chambered. Principal leaves pinnate-pinnatisect to bipinnate-pinnatisect, 10-19 cm. long; primary pinnae 5-9, opposite on the rachis, the basal pinnae 2.5-10 em. long, the pinnules alternate on the rachilla, the ultimate segments 1.0-2.5 cm. long, their lobes 0.5-9 mm. wide, broadest at the base or at the middle and abruptly tapered into a mucronate tip; margins smooth or irregularly short ciliolate, surfaces green on both sides or obscurely lighter green beneath, upper surface glossy when 1969] Dahlia — Sorensen 349 fresh, glabrous or with very minute hairs along the main veins above; stipels rare or absent; petiole 2-9 cm. long, slightly flattened above; rachis glabrous or with a cluster of hairs at the rachis nodes, usually grooved above, the decurrent bases of the pinnae sometimes forming wings along the margins esp. in the terminal portions of the leaf. Heads in 2’s or 3’s, obliquely erect (nodding in fruit), peduncles 12-35 em. long; outer involucral bracts erect or very slightly spread- ing, 5-7 mm. long, 1.8-5 mm. wide, oblong or ovate, obtuse; inner bracts 9-15 mm. long, 4-7 mm. wide; rays pale lavender to light purple, fading when dry to light or dark tan, 2-4 cm. long, 0.9-1.7 (-2) em. wide, ovate, acute or denticulate; disc florets about 44, 6-6.5 mm. long, corolla lobes strongly recurved, yellow or yellow with purple tips, sometimes with 5 distinct dark veins; style branches about 3 mm. long, 0.5 mm, or less wide. In the present treatment two infraspecific taxa of Dahlia dissecta are recognized as follows: 9a. Dahlia dissecta var. dissecta. TYPE: MEXICO: SAN LUIS PoTOsÍ: limestone ledges, San José Pass, July 1890, Pringle 3167 (Holotype: US!; Isotypes: F!, GH !, ISC!, Mo!, MSC!, NY!, RSA!, UC!, US!). Perennial herb, 3-7.6 dm. tall. Stems 1.5-3 mm. diam.; internodes 1.5-15 em. long, pith rarely chambered. Leaves 10-15 cm. long; basal pinnae 2.5-5.2 cm. long, ultimate segments 1.0-1.8 cm. long, their lobes 2-9 mm. wide; petiole 2-7 cm. long. Peduncles 12-27 cm. long; outer involucral bracts 5.5-7 mm. long, 1.8-4 mm. wide, ovate, obtuse; inner bracts 9-15 mm. long, 4-6.5 mm. wide; rays 2-3.2 cm. long, 0.9-1.2 em. wide; achenes narrowly obovate, 9-11 mm. long, 1.5-2 mm. wide, blackish, minutely tuberculate, many-sulcate; pappus absent or consisting of 2 minute rudiments. Chromosome number: n = 17. Rocky slopes and ledges, 6500-8000 ft., Hidalgo and San Luis Potosí. Flowering July-Aug. Map 3, closed circles. MEXICO. HIDALGO: Distr. of Zimapán, Barranca de San Vincente, Moore & Wood 4425 (NY), Melchert & Sorensen 6205 A-C (IA), 6213 A-D (IA). SAN LUIS POTOSI: San Juan del Llanos, Halsted s.n. (NY); arroya near San Ruizo, Whiting 983 (MICH). The label on an early specimen collected by Halsted (NY) gives the locality as “San Juan del Llanos." Sherff (1955) understood this to be a reference to the town of “Llanos” in northern Puebla, shown on the American Geographic Society's millionth map. I think the small village of San Juan del Llanos in San Luis Potosi (NE. of 22° N. & 100° 350 Rhodora [Vol. 71 Fig. 4. Dahlia dissecta, var. sublignosa Sorensen. Photograph of type (NY), X 1/3. 1969] Dahlia — Sorensen 351 W.) is the more likely, as this locality would fall well within the known range of Dahlia dissecta (see Map 3). 9b. Dahlia dissecta S. Wats. var. sublignosa Sorensen, var. nov. TYPE: MEXICO: TAMAULIPAS: 3 mi. N. of Miquihuana in pine forest (99°47 N. Lat., 23936' W. Long.), about 7000 ft., 14 July 1949, Stanford, Ruther- ford & Northcroft 2436 (Holotype: NY!; Isotypes: MICH!, TEX!, UC!, US!). Herba perennis vel subherbacea 7-9 dm. alta. Caules hic annus ramificibus brevi caules (5-15 cm.) longitudinibus anni praecedentis. Folia simulans Dahlia dissecta var. dissecta autem plus grandis (11-19 cm. longa) et segmentis ultimis plus angustis (0.5-5.5 mm. latis). Pedunculis 12-35 cm. longus; involucri squamis exterioribus erectis vel effusis leviter sub anthesi, 5-7 mm. longis, 2.5-5.5 mm. latis. Flores ligulati diluti-lavanduli ad lilacini, 2.8-4 cm. longi, 1.2- 1.7(-2) em. lati; flores tubulosi flavis, 5-atronervati. Fig. 4. Known only from the type locality and from an area 0°10’ of long. W. of the type locality, 7000-8000 ft., Tamaulipas, México. Flowering July. Map 3, stars. MEXICO. TAMAULIPAS: between Marcella and Hermosa, Stanford et al. 2436 A (NY, TEX, US). On morphologic grounds the two infraspecific taxa of Dahlia dissecta may be distinguished by the following key: + Stems of the current year arising from a crown of tuberous roots or rootstock; leaves 10-15 cm. long, ultimate segments 2-9 mm. wide; outer involucral bracts 1.8-4 mm. wide. ................................ x OPERA asa sas EEE 9a. D. d'ssecta var. dissecta. +Stems of the current year branching from short (5-15 cm.) ligneous portions of the previous year’s ‘growth; leaves 11-19 em. long, ultimate segments 0.5-5.5 mm. wide; outer involucral bracts 2.5-5.5 mm. wide. ............ 9b. D. dissecta var. sublignosa. In addition to the above distinctions it should be pointed out that var. sublignosa tends to be generally more robust attaining heights of 7-9 dm. as compared to the height of var. dissecta (3-7.6 dm.). There is also a considerable geo- graphic isolation between the two taxa. Dahlia dissecta var. sublignosa is isolated from the nearest part of the range of var. dissecta (at San José Pass in San Luis Potosí) by about 85 miles. 352 Rhodora [Vol. 71 Section 4. Dahlia. Type species: Dahlia pinmata Cav. Sect. Neocaulon Sherff, Am. Jour. Bot. 38: 71. 1951. Type species: Dahlia coccinea Cav. Stems herbaceous (lignescent in D. apiculata) mostly with hollow internodes and septate or imperfectly septate nodes; leaves simple to 3-pinnate, pinnules opposite or sub-opposite on the rachilla (alter- nate on the rachilla in D. merckii); petioles solid (hollow in D. merckii); outer involucral bracts erect, spreading, or reflexed. Chromosome numbers, n = 16, 18, & 32. (Species 10-27.) 10. Dahlia merckii Lehm. Del. Sem. Hort. Hamb. 1839; Schlectendal, Linnaea 14(Lit.-Ber.): 130. 1840. TYPE: the original material grown in the Hamburg Bot. Gard. from seeds collected in México, the exact locality not known, H. J. Merck s.n. (Lectotype: GH!). Dahlia glabrata Lindl. in Edwards Bot. Reg, 26: pl. 29. 1840. Type: based on plants cultivated in England from seed collected in México, exact locality not known, G. F. Dickson, Esq. s.n. Lectotype: pl. 29, Lindley, loc. cit. Dahlia cosmaeflora Jacques, Ann. de Flore et Pom. (Jour. des Jard.) 1840: 345, pl. s.n. 1840. Lectotype: pl. facing p. 945, Jacques, loc. cit, Dahlia decaiseana Verlot, Rev. Hort. 1864: 31, pl. s.m. 1864. Lecto- type: pl. facing p. 31, Verlot, loc. cit. Dahlia seapigera var. scapigera f. merckii (Lehm.) Sherff, Am. Jour. Bot. 34: 141. 1947. Type: that of Dahlia merckii Lehm. Herb, 4.5-18.5 dm. tall, freely branching near the base, the leafy portions averaging 1/3 to 1/2 the height of mature flowering plants. Stems 3-9 mm. diam., terete or 6-angled, sometimes strongly 6-ridged (best seen in fresh material), glabrous or sometimes a few hairs gathered at the nodes, usually red when fresh, drying to brownish- red or tan, hollow except for septa at the nodes; internodes 1-27.5 cm. long, the lower often the shortest and usually increasing in length toward the topmost and longest. Leaves once-pinnate to bi- pinnate-pinnatisect, 14-38 cm. long; primary leaflets 5-7, opposite on the rachis (leaflets of the secondary and tertiary ranks alternate on their rachillae), the basal pinnae 6.5-16 cm. long, sessile or stalked, the petiolules or pseudopetiolules 9-40 mm. long; ultimate segments broadly ovate to slightly obovate, 2-5.2 cm. long, 0.9-3.2 em. wide (the length often about twice the width), the termina] seg- ment usually somewhat larger than its companion segments; apices acute to acuminate; surfaces bicolored, lighter green beneath, the upper glabrous or sparsely pubescent, minutely appressed pubescent along the principal veins, the lower glabrous or sparingly pubescent; 1969] Dahlia — Sorensen 353 margins ciliolate, the cilia now long, soft and flexuous, now short and rather stiff, irregularly serrate to dentate-crenate, 1-5 teeth per side; stipels usually present, 8-46 mm. long, 3-32 mm. wide, ovate, tips acute or acuminate, attached at each rachis node, except plants from Nuevo León and Tamaulipas on which stipels occur only at the basal rachis node or not at all (stipels sometimes also attached at the basal rachilla node); petiole glabrous, 4-16 cm. long, strongly dilated at the base, when fresh forming a conspicuous fleshy ridge around the stem, medianly terete or hemispherical, hollow; rachis glabrous except for a cluster of hairs at the junction with the pri- mary pinnae, often U-shaped with two parallel ridges above. Heads often numerous, 10-90 per plant (fewer in material from northwest México), obliquely erect or subcernuous, elevated well above the leafy portion of stem on several to many branches, these again branched and terminated by 2-19 heads on peduncles 1-30 cm. long; outer involucral bracts spreading or reflexed at anthesis, linear to slightly spatulate, apically obtuse, 9.5-11.5 mm. long, 1.8-2.2 mm. wide; inner bracts 12-15 mm. long, 5-7 mm. wide; chaff at anthesis hyaline to yellow, sometimes suffused with purple, 4.5-9.5 mm. long, 3-6 mm. wide, enlarging in fruit to 14-18 mm. long, 6-10 mm. wide; rays whitish (said to be white on many labels) to pale lavender to purple, 2-2.9 cm. long, 1.2-2 cm. wide, broadly elliptical, acute or denticulate; disc florets 47-58, corolla yellow, yellow with purple tips or purple, 6-9 mm. long; style branches linear-lanceolate, 2.5-3 mm. long, less than 1 mm. wide; achenes more or less linear, broad- est above the middle, abruptly narrowed at the top and capped by a shallow disc, 5.5-10 mm. long, 1.1-2.3 mm. wide, ventrally flattened and dorsally rounded cr V-shaped in cross-section, longitudinally suleate, dark brown to black (tan with dark speckling when im- mature); pappus absent or consisting of 2 minute rudiments. Chrom- osome number, n = 18. Rocky slopes, ledges and lava fields (“Pedregales”), 6000-10,500 ft, Nuevo León and Tamaulipas in northern México, south to northern Guerrero, westward from Vera- cruz to western México state, frequent on cool mountain slopes of Distrito Federal, México state and Morelos. Flowering June-Oct. (Dec.). Map 4. MEXICO. TAMAULIPAS: E. side of Cerro Linadero, Meyers & Rogers 2894 (F. MICH, MO, US); Cañon 4 km. W. of Miquihuana, Stanford et al. 767 (F, GH, Mo, NY); E. & S. slopes of Cerro Peña Nevada, Stanford et al, 2537 (NY, TEX, UC, US) ; NUEVO LEON: vicinity of Galeana & Cerro Infernillo, Mueller & Mueller 837 (F, GH, MICH, TEX), Taulor 183 (r, MO, TEX) ; vicinity of Dulces Nombres, Meyer & Rogers 2894 (F, MICH, MO, US), 3009 (F, MICH, MO, US) ; Cerro Potosí, McGregor et al. 296 (MISC); along trail from Trinidad to Cerro de 354 Rhodora [Vol. 71 Map 4. Distribution of Dahlia merckii Lehm. la Cebolla, Muller 2879 (GH, MICH, UC); Cerro Grande, Straw & Forman 1889 (MICH). SAN LUIS POTOSI: without specific locality, Parry & Palmer 477 (Gu, 180, NY, US); slopes of Cerro Alvarez, MeVaugh 12272 (MICH) Palmer 211 (F, GH, MEXU, NY, US), Pennell 17831 (MICH, NY); Cerro Aguilla, Lundell 5374 (MICH, Us); 22 km. W. of Santa Catarina, Roe & Roe 2173 (A, WIS). HIDALGO: vicinity of Real del Monte (i.e, Mineral del Monte), Ehrenberg 368 (F, GH), Mexia 2766 (CAS, MICH, MO, NY, UC), Martinez s.m. (F); vicinity of Cerro de Pachuca, Rose & Hay 5594 (us), Rose et al. 8877 (US); Guerrero near Omitlán, Martinez 15048 (MO); along road from Pachuca to Parque Nacional el Chico, Moore & Wood 4105 (GH, NY); Cerro Santa Ana, NE. of Apán, West B-18 (WIS). VERACRUZ: Loma Grande on Cerro Orizaba, Balls 5349 (uc, us); Maltrata at Punta de Zecatál, Matuda S-115 (MICH), 1363 (DS, GH, MEXU, MICH, NY, US); 21 mi. W. of Orizaba along route 150, Melchert & Sorensen 6200 A-H (1A); Cerro Citlaltepetl, Seaton 277 (GH), Cambre de Acaltzingo above Orizaba, L. O. Williams 13496 (F). PUEBLA: La Cima near Jajalpa, Kuntze 23746 (NY), 23747 (NY); Cerro Tepoxuchil, vicinity of Puebla, Arséne 1396 (us); Los Chinos, San Manuel de la Sierra, Balls, B5306 (vc, vs); Puente del Emperador La Venta, Sharp 44542 (F); 14 mi. W. of Texmelucan, King 3564 (MICH, TEX, UC); above Coxcatlán between Apala and summit of Cerro Chichiltepec, C. E. Smith et al. 3817 (F, US) ; along Tehuacán-Orizaba highway below Puerto del Aire, C. E. Smith et al. 3924 (F, MEXU, NY, US). 1969] Dahlia — Sorensen 355 MEXICO: along route 190 between México City and Puebla, mostly in the vicinity of Río Frío, Iltis et al. 1093 (WIS), Matuda 21453 (MEXU, NY), Melchert & Sorensen 6149 A-G (1A), Rock M-351 (TEx), Sharp 44100 (F), Sharp & Gilly 46 (MICH); vicinity of Amecameca along road to summit of Voleán Popocatepetl, Balls 5157 (us), Bea- man 2040 (MEXU, MISC), 2113 (GH, MISC, UC, US), Purpus 1553 (POM, UC), Urbina s.n. (MEXU); vicinity of Huehuetoca, Gold & Eheberle 21770 (MEXU, US), Matuda 29070 (MEXU, NY) ; Mtns. near Guadalupe, Bourgeau 802 (or 862) (GH, vs); along route 57, México City to San Juan del Río at k-167, Hawkes et al. 1395 (F); Cerro de Trane- pantla, Matuda 21685 (NY); Cerro de Jocotitlán, Matuda 31003 (MEXU) ; 2.5 mi. S. of Tenango along route 55, Melchert & Sorensen 6121A & 6121-POP (IA); 26 mi. W. of México City along route 15, Melchert & Sorensen 6238 A-E (IA); 55 km. SE. of México City, Weaver 743 (GH, US). DISTRITO FEDERAL: on pedregal (lava fields) along route 95 between México City and Cuernavaca, Clark 7277 (MO, NY), Manning & Manning 531050A (GH), Miranda et al. 2258 (F, GH, MICH, TEX), Melchert & Sorensen 6104 (IA), Ortenburger et al. 16M7382 (F), Powell & Edmondson 743 (MICH, TEX), Rusby 377 (NY, US) ; lava fields on the slopes of Cerro Ajusco, Harshberger 118 (BH), Iltis et al. 143 (wis), Matuda 19169 (MEXU, US), Rose et al. 9258 (GH); slopes of Volcán Xitle, Matuda 19589 (MEXU, NY). MORELOS: precipitous slopes near Parque Nacional de Zempoala, Melehert & Sorensen 6106 (1A), Ugent & Flores 1484 (wis); lava fields above Cuernavaca, Matuda 21476 (MEXU, NY), Pringle 13039 (CAS, F, GH, MICH, US). GUERRERO; Cuernavaca to Taxco, Chute M-90 (MICH). Sherff (1947, 1955 & 1959) reduced Dahlia merckii to the rank of “forma” under D. scapigera var. scapigera, ex- plaining that this taxon was “best recognized as a some- what localized color-form [i.e., with purple disc corollas] of D. scapigera" (see Table 2). This interpretation was based on his otherwise broad concept of D. scapigera (q.v.) proper for which the majority of the specimens he cited possessed yellow disc corollas. I have observed that the purple color of the disc florets mentioned by Lehmann (ex Schlechtendal, 1840) is a localized color-form but one which varies in its intensity and in the amount of the corolla tube so colored from plant to plant within a population. The same sort of variation has been seen in other Dahlia species in three of the four sections, notably D. australis, D. scapigera, D. imperialis, and D. dissecta. Aside from its unique chromosome number, n = 18, Dahlia merckii combines several characteristics which dis- 356 Rhodora [Vol. 71 tinctly set it apart from the remaining species. These are the alternate pinnules, hollow petioles, compact and greatly branched growth of the vegetative portion of the plant, red stems, smooth texture of its leaves, conspicuous stipels, and its large number of flowering heads on compound peduncles. The occurrence of alternate pinnules on the leaves of D. merckii could cause confusion with the species of section Entemophyllon, the leaves of which all possess this character. D. merckii may be distinguished from them by its hollow petioles and by its spreading or reflexed (not ascending or erect) outer involucral bracts. 11. Dahlia apiculata (Sherff) Sorensen, comb. & stat. nov. TYPE: MEXICO: PUEBLA: Tlanislotepec, 7000-8000 ft., July 1909, Purpus 3850 (Holotype: uc!). Dahlia scapigeroides var. apiculata Sherff, Am. Jour. Bot. 34: 145. 1947. TYPE: that of Dahlia apiculata. Suffruticose perennial. Stems of the previous year solid or hollow, 3-6 mm. diam., purplish-gray, pruinose, glabrous, longitudinally wrin- kled, lenticels conspicuous; stems of the current year 2.5-3 mm. diam., solid, longitudinally striate, glabrous except for a cluster of brown hairs at the nodes; median internodes 1-5 cm. long. Leaves once- pinnate, 7.5-14 cm. long; leaflets 5-9, basal pinnae 2.5-4.5 cm. long, 1-2 em. wide, lanceolate, acute; surfaces glabrous except for a line of minute hairs along the main vein above; margins lacking cilia, sharply serrate with 1-4(-5) apiculate teeth per side; stipels absent; petiole 2-4 cm. long, crescent-shaped in cross-section, puberulent above in the longitudinal groove; rachis glabrous except for a cluster of hairs at the junctions with the pinnae. Heads obliquely erect at anthesis, mostly solitary, sometimes in 2’s and 3’s, on peduncles 8.5- 17 cm. long; outer involucral bracts reflexed at anthesis, 12-17 mm. long, 3-5 mm. wide, oblanceolate, acute, 5-8-lined, usually glabrous, sometimes minutely puberulent on ventral surface; inner bracts 13- 20 mm. long, 4-8 mm. wide, ovate-lanceolate, obtuse to subacute; chaff (in fruit) tan, about 20 mm. long, 6-8 mm. wide, ovate, obtuse; rays purple (fide Purpus 3112, MO), 3.5-5 cm. long, 2-2.7 cm. wide, ovate, acute or denticulate; disc florets yellow; achenes (immature) 9-10 mm. long, 2-2.2 mm. wide, oblanceolate or spatulate, pappus obsolete or consisting of 2 minute rudiments. Rocky slopes at 7000-8000 ft. in the vicinity of its type locality near the border of Oaxaca and at about 5000 ft. in the Cafion del Tomellin and Amaloyán, Oaxaca. Flowering May-Dec. Map 5, stars. 1969] Dahlia — Sorensen O T MEXICO. PUEBLA: Coscomates and Los Naranjos, vicinity of San Luis Tultitlanapa near border of Oaxaca, Purpus 3112 (Ds, F, GH, MO, NY, UC, US). OAXACA: Almaloyas [Amaloyán] at Santa Caterina, 1000 m. Conzatti 1647 (F); Tomellin Canon, Pringle 5660 (GH, MICH, US); Almaloyán, Seler 1395 (GH, MO). Sherff (1947) first described Dahlia apiculata as a variety of D. scapigeroides. He presumably based this alliance on such similarities as: (1) their woody habit, (2) their manner of producing short, leafy shoots with the leaves crowded near the base, and (3) their scapose flower- ing heads. Despite such similarities, however, these taxa are apparently not closely related. In the present treat- ment, D. scapigeroides is allied with the members of the newly created section Entemophyllon. This change is based on a number of characteristics which it shares with the members of that group. With this realignment D, apiculata remains the only lignescent member in the section Dahlia. It may be observed from the list of specimens cited above that Dahlia apiculata is known from just two locali- ties. The first is that of the type in Puebla and the other from a location 30 mi. to the south in Oaxaca. Plants col- lected in each of these geographic areas are seen to be somewhat distinct morphologically. The specimens collected in Oaxaca generally display fewer (3 not 5) and slightly larger (5-8 cm. not 2.5-4.5 cm. long) leaflets than those from the type locality. Both groups do, however, share the suffrutescent habit noted above. Furthermore, the leaves of the two groups are alike in their smooth texture, absence of stipels and eciliolate margins, Additional material of this species with more extensive notes on height, habit, and chromosome number is much to be desired. 12. Dahlia cardiophylla Blake & Sherff in Sherff, Field Mus. Publ. Bot. 11: 453. 1936 (a new name for Coreopsis cordifolia Sessé & Moc). TYPE: MEXICO: GUERRERO: mts. of Tixtla, Aug. 1787-1804 (exact date not known), Sessé, Mocino, Castillo & Maldonado 4027 (Holotype: MA, photo. F!). Coreopsis cordifolia Sessé & Moc. Pl. Nov. Hispan. 147. 1890. & ed. 2. 137. 1893. TYPE: as for Dahlia cardiophylla Blake & Sherff 358 Rhodora [Vol. 71 above non Coreopsis cordifolia (Schtz.-Bip.) Drake del Cast. Ill. Fl. Ins. Mar. Pacif. 208. 1890; Fl. Polyn. Franc. 109. 1892. which is based on Bidens cordifolia Schtz.-Bip. Flora 39: 361. 1856. Herb, 7-12 dm. tall, unbranched except in the flowering portions. Stems 3-9 mm. diam., terete, many-sulcate, glabrous, stramineous cr green; internodes 5-6.5 cm. long. Leaves simple, sessile and cordate-clasping, their bases slightly connate, 4-8.5 em. long, 4-8.5 cm. wide; surfaces glabrous throughout, conspicuously bicolored, the upper dark green, the lower glaucous green, the principal veins noticeably stramineous; margins eciliolate, coarsely serrate or dentate with 5-8 teeth per side. Heads solitary or in 2’s or 3’s, erect or obliquely erect at anthesis, on peduncles 3-6 cm. long; outer in- volueral bracts reflexed at anthesis, 5-8 mm. long, 2.5-5 mm. wide, broadly obovate or spatulate, obtuse, green, glabrous, dorsally one- several-nerved; inner bracts 1-1.5 em. long, 4.5-6 mm. wide, greenish- yellow; chaff yellow; rays spreading, 1-1.4 cm. long, 6.5-8 mm. wide, ovate, acute or denticulate, deep purplish-scarlet when fresh, drying to blackish-purple; disc florets yellow; achenes not seen. Chromosome number, n = 16. Precipitous, open or lightly-wooded, rocky slopes, 5500- 6000 ft., mts., vieinity of Tixtla, Guerrero, México. Flower- ing Aug. MEXICO. GUERRERO: 8 mi. E. of Tixtla along new road to Chilapa at point where road crosses Barranca de Honda, Melchert & Sorensen 6142 A-K (IA). Dahlia cardiophylla is a rather rare and interesting species which, until recently, has apparently not been col- lected, since Sessé et al. botanized in México (1787-1804). Its leaves (simple, sessile and cordate-clasping) are quite distinct from any other known species. In only one other species, D. purpusii, are the leaves believed to be exclusively simple, but in this species the leaves are rounded or nar- rowed (not cordate-clasping) at their base, and they are sometimes short petiolate. 13. Dahlia purpusii Brandg. Univ. Calif. Publ. Bot. 6: 76. 1914. TYPE: MEXICO: CHIAPAS: Cerro del Boqueron (now probably Cerro Ovando), Sep. 1915, Purpus 6680 (Holotype: UC!; Isotypes: GH!, MO!, NY). Herb (probably perennial) at least 4 dm. tall. Stems 4-6 mm. diam. many-striate, stramineous, glabrous; internodes 3.5-11 cm. long. Leaves simple, elliptical or almost ovate, short acuminate, base tapered and slightly rounded, 8-12 cm. long, 4-7 em. wide, sessile or very short petiolate; petiole, if evident, 1-3 mm. long; surfaces glab- 1969] Dahlia — Sorensen 359 rous, slightly bicolored, salient veiny and lighter green beneath, margins eciliolate, serrate-crenate, 10-18 teeth per side. Heads soli- tary or in 2’s and 3’s, erect or obliquely erect on peduncles 5-8.5 em. long; outer involucral bracts reflexed at anthesis, 8-12 mm. long, 3- 4.5 mm. wide, ovate-lanceolate to obovate, acute or obtuse, glabrous, dorsally many-lined and sometimes reticulate veiny; inner bracts 14-15 mm. long, 5-8 mm. wide; rays probably purple, about 3 cm. long, 10-13 mm. wide, ovate, acute or denticulate; disc florets probably yellow; mature achenes not seen. Known only from the type locality. Flowering Septem- ber. Specimens of the type collection of Dahlia purpusii, ap- parently the only ones in existence, are somewhat inade- quate. None of the specimens is complete and the labels lack information on the habit of the plants. For these reasons the accompanying description must be considered provisional. Despite the scant material available the validity of this species is not questioned. Its simple leaves are quite unlike those of any other Dahlia. The only other simple-leaved species is D. cardiophylla, but the leaves of that species are sessile and cordate-clasping (not rounded or narrowed at the base). 14. Dahlia pinnata Cav. Icones et Descr. Pl. 1: 57, t. 80. 1791. TYPE: exact place of collection and date not known; originally described by Cavanilles from plants cultivated in the Royal Botanic Gardens, Madrid, from seed collected in México and sent to Spain by Vin- cente Cervantes, ca. 1789 (Lectotype: t. 80, Cava- nilles, loc. cit.*). Georgina purpurea Willd. Linn. Sp. Pl. ed. 4. 3: 2124. 1803. Type: based on Dahlia pinnata Cav. Dahlia sambucifolia Salisb. Parad. Lond. t. 16. 1805. Lectotype: t. 16 in Salisb. loc. cit. Dahlia pinnata var. nana Geo. Jackson in Andrews Bot. Repos. 7: pl. 483. 1807. Lectotype: pl. 483 in Andrews, loc. cit. “In a personal communication, Dr. J. H. Beaman, Michigan State University, informs me that he made a deliberate search for the type of D. pinnata when he visited the herbarium of the Royal Botanic Gardens, Madrid, Spain, in 1963, but was unable to locate it. Types of other species described by Cavanilles were there indicating that had a type been preserved it would have been kept at MA. 560 Rhodora [Vol. 71 Georgina variabilis [var.] lilacina Willd. Hort. Berol. 2: pl. 94. 1809. Lectotype: pl. 94, Willd. loc. cit. Georgina variabilis [var.] pallida Willd. Ibid, pl. 95. Lectotype: pl. 95, Willd. loc. cit. Georgina variabilis [var.] purpurea Willd. Ibid. pl. 93. Lectotype: pl. 93, Willd. loc. cit. Georgina superflua a rubra DC. Ann. Mus. Hist. Nat. Paris. 15: 310. 1810. Type: based on Georgina variabilis [var.] purpurea Willd. (G-bC, microfiche libraries A-GH). Georgina superflua B purpurea DC. Ibid. Type: based on Dahlia pinnata Cav. (G-DC, microfiche libraries A-GH). Georgina superflua Y lilacina (Willd.) DC. Ibid. Type: based on Georgina variabilis [var.] lilacina Willd. (G-bC, microfiche libraries A-GH). Georgina superflua è pallida (Willd.) DC. Ibid. Type: based on Georgina variabilis [var.] pallida Willd. (G-bC, microfiche libraries A-GH ). Georgina superflua e flavescens DC. Ibid. 311-312. Lectotype: based on plants obtained from seed of Georgina superflua Y lilacina (Willd.) DC. (G-pc, microfiche libraries A-GH). Dahlia purpurea Poir. Lam. Encycl. Suppl. 2: 444. 1812. Type: based on Georgina purpurea Willd. Dahlia superflua. (DC.) Ait. f. Hort. Kew. ed. 2. 5: 87-88. 1813. Type: based on Georgina superflua [vars.] DC. Coreopsis georgina [var.] nuda Cass. Dict. Sci. Nat. 18: 442. 1820. Type: based on Georgina superflua DC. and Georgina variabilis [vars.] Willd. Dahlia variabilis (Willd.) Desf. Cat. Pl. Horti Paris. ed. 3. 182. 1829. Type: based on Dahlia pinnata Cav. and Georgina variabilis [vars.] Willd. Dahlia pinnata var. variabilis (Willd.) Voss in Vilmorin's Blumen- gart. 1: 489. 1894. Type: based on Dahlia pinnata Cav. and Geor- bina variabilis [vars.] Willd. Perennial herb, 10-18 dm. tall, usually unbranched except in the flowering portion. Stems 8-12 mm. diam., scabrous, often rather red or maroon; internodes 9-16 cm. long, hollow. Leaves simple or pinnatisect to bipinnate, coarse-textured, 13-25 cm. long; leaflets 3-5, the basal pinnae (or basal segments) 5.5-13 em. long, 4-8 cm. wide, ovate elliptical, acute; surfaces sparingly pubescent or strigillose, ventrally the hairs short and stiff, esp. along the veins, dorsally the hairs softer, flexuous, following the principal veins; margins ciliolate, the cilia very stiff, coarsely serrate or more commonly dentate, some- times lacerate; stipels rare, tending to occur with greater regularity on bipinnate leaves; petiole 3.5-12 cm. long, or leaves with a broadly- winged midrib, then pseudopetiolate, often pubescent on the back, broadly crescent-shaped in cross-section; rachis of compound leaves often winged with the decurrent bases of the distal primary pinnae. 1969] Dahlia — Sorensen 561 Heads few, 2-8 per plant, occurring in 2's ad 3’s, erect or obliquely erect at anthesis, on peduncles 5.5-11 cm. long; outer involucral bracts reflexed at anthesis, 10-14 mm. long, 4.5-8 mm. wide, obovate, acute, glabrous, dorsally several-lined; inner bracts 15-21 mm. long, 6-10 mm. wide; rays light-purple or lavender-purple often with a yellowish or rosy spot at the base, spreading, 3.8-4.2 cm. long, 1.7-2.6 em. wide, ovate, acute or denticulate; chaff resembling the inner bracts, pale yellowish drying to tan; disc florets 96-144, mostly yel- low, sometimes with light-purple veins, corolla lobes spreading at anthesis, the tube 9-11 mm. long, style branches 4-5 mm. long, less than 1 mm. wide, linear-lanceolate, long-attenuate; achenes 11-13 mm. long, 2-8 mm. wide, linear-oblanceolate, slightly appressed pubescent near base. Chromosome number, » — 32. Rocky slopes, 7000-10,000 ft., southern Hidalgo south to northern Guerrero, west from the México-Puebla border at Volcán Popocatepetl to western México state, frequent in the voleanic mountains just west and southwest of México City. Flowering May-Nov. Map 5, solid dots. MEXICO. SAN LUIS POTOSI: chiefly in the region of San Luis Potosí, Parry & Palmer 479 (GH, ISC, MO). HIDALGO: vicinity of Cerro de Pachuca, Miranda 4454 (MEXU), Rose et al. 8784 (US); Cerro de los Pitos, Matuda 21542 (MEXU). MEXICO: slopes of Río We DaHLIA APICULATA @ DAHLIA PINNATA Q DAHLIA MOLLIS Map 5. Distribution of Dahlia apiculata (Sherff) Sorensen, stars; Dahlia pinnata Cav., Closed circles; and Dahlia mollis Sorensen, open circles. 362 Rhodora [Vol. 71 Hondo Canyon, Matuda 26182 (MExU, NY), Pringle 3165 (F, GH, LCU, MEXU, MICH, MISC, MO, NY, RSA, UC); Amecameca, Fisher 197 (F, MO, US); near Huisquilnango, Matuda 21092 (MEXU); Cerro Jilote- pec, Matuda 29121 (MEXU, NY); San Geronimo, Villa del Carbon, Matuda 29229 (MEXU, NY, US); Cerro de Azompan, Tequezquinahuac, Matuda 31170 (us); 7 mi. S. of Tenancingo along route 55, Melchert & Sorensen 6125 A-D (IA); about 10 mi, W. of Atlaleomulco near small R.R. station of Bassoco, Melchert & Sorensen 6233 A-E (IA); about 12 mi. SE. of San Juan del Rio along route 57, Melchart & Sorensen 6239 A-B (IA). DISTRITO FEDERAL: slopes of Cerro de Ajusco, Pringle 13546 (CAS, GH, MICH, US), Pringle 21556 (us), Rose & Hay 5534 (GH, NY, Us); Santa Fé, Bourgeau 736 (GH, US); at Tizapan (near Villa Obregon), Bourgeau 746 (NY); Lomas de Mix- coac, Lyonnet 838 (vs); Tlalpám, rocky hills, Russell & Souviron 32 (US). Dahlia pinnata is the type species of the genus Dahlia. The name was based by Cavanilles on plants under cultiva- tion in Madrid. These as seen by the plate accompanying the original description (Cavanilles, 1791), already had undergone artificia] selection for heads producing multi- seriate ligules. This fact has led to widespread usage of the name D. pinnata, or some nomenclatural equivalent such as Georgina variabilis and Dahlia variabilis, for any ‘‘dou- ble-flowered" or otherwise unusual form originating in gardens. As was the practice in the period, Cavanilles gathered seed from the Dahlia plants in his care (then numbering three species including D. rosea and D. coccinea) and sent them to various botanists and botanical institutions of Europe with which the Madrid gardens corresponded. Among those receiving seeds were the botanical garden at Montpelier, France, the Museum of Natural History, Paris, Holland House, Kensington, England, and the Royal Botanic Garden, Berlin. Following this distribution, which occurred in the years 1798-1804, Dahlias quickly entered into the horticultural trade throughout Europe. The number of double-flowered and other forms rapidly increased. Con- committant with this was an increase in the number of new names appearing in botanical and horticultural litera- ture. By 1814, fully double-flowered forms had been de- veloped in Louvain, Belgium (Hibberd, 1891), and in 1817, 1969] Dahlia — Sorensen 363 Breiter listed more than 75 “varieties” in cultivation at Leipzig. The number of garden varieties or cultivars has never ceased to grow. In 1934, Howe (1936) states there were more than 14,000 named cultivars which had been produced in the past or were currently in the trade. All of these numerous horticultural aspects have been consist- ently equated by authors, from the early 1800’s to the present, with the original Dahlia pinnata. There is perhaps little doubt that a trace of the original Dahlia pinnata could have been found, were this possible at the time, in the genetic make-up of the many double- flowered Dahlias which arose in Europe during the early 19th Century. In my judgment, however, it is a mistake to name all of these D. pinnata. Nor should any single species be regarded as the basic one from which these numerous forms were selected. In 1929, Lawrence dem- onstrated on the basis of biochemical analysis, corro- borated by my own findings (cf. discussion following systematic treatment of Dahlia coccinea), that the so- called garden Dahlia (designating collectively all Dahlia cultivars), a fertile tetraploid, n = 32, was of hybrid origin initially having come from a cross between D. coc- cinea with red or yellow ligules and one of the double- purple-flowered species. Of the two purple-rayed species of the genus known to produced double-flowered forms, Dahlia pinnata is the more likely to have been involved in this cross, inasmuch as the other is D. imperialis, a diploid, n = 16, and one of the ‘‘tree-Dahlias” (cf. Pope- noe, 1920). If one recognizes the hybrid nature of the garden Dahlia, the next question to consider is when did the original hybridization occur. Was the D. pinnata of Cavanilles al- ready a hybrid at the time of its description, or did the hybridization which led to the subsequent development of the earliest “garden Dahlias” take place after seeds of the original material had been distributed to various botanical institutions in Europe? I am inclined to think the latter was the case. Had the seeds Cavanilles received from 364 Rhodora [Vol. 71 México come from an interspecific cross or from a plant which had been derived previously from such a cross, in- volving in either case D. coccinea as one of the parents, they would have resulted in plants with ligule colors potentially ranging, in different individuals, from white through light and dark purple to yellows, oranges, and scarlets.^ Yet, the only color Cavanilles (1791) gives for the ligules is purple, “Corolla . . . feminis magna, coeru- leo-rubens". (The terms “feminis magna" allude to the pistillate, but sterile, ligulate florets. Only the disc florets are hermaphroditic and fertile.) The explanation given above for discontinuing the use of Dahlia pinnata as the correct name for the garden Dahlia does not solve the problem of naming the many forms now found in cultivation. It would be incorrect to designate the cultivated Dahlias as D. coccinea X pinnata as has been done often for some taxa known to be the first generation of an inter-specific cross. The development of our modern cultivars has involved repeated hybridization between existing hybrids as well as between hybrids and wild species. Because of this it is best to utilize cultivar names for all those garden forms which are not otherwise clearly selections from among the naturally occurring variants within a wild species population. Unfortunately, at the present time no key to Dahlia cultivars exists. The best means one has to identify cultivars is by making comparisons of his material with portraits and descrip- tions. To date the most complete list of Dahlia cultivars with descriptions relating to the characteristics of the heads is the booklet entitled 7968 Classification of Dahlias, com- piled and published jointly by the American and Central States Dahlia Societies. In this booklet the names of culti- vars are alphabetically arranged within groups based on the diameter of the heads. After each name are symbols *One usually expects the F, to be fairly uniform. Such is not the case in Dahlia. The great variability in Dahlia plants is due to the prevailing self-incompatibility which maintains in all progeny a rather high degree of heterozygosity. 1969] Dahlia — Sorensen 365 which denote the color of the rays and the horticultural category (i.e. whether Formal, Decorative, Pompon, Colla- rette, etc.). The purpose of this listing is to be able to determine the application of a name by checking the de- scription ; however, I have found it practicable to use it for the identification of an unknown by reading through the descriptions within a given group and narrowing the num- ber of possibilities to three or four. Thereafter it becomes necessary to search out portraits for comparison in trade magazines and catalogues. Dahlia pinnata sensu stricto, is characterized by its rather coarse, often scabrous leaves which vary in their degree of segmentation from simple and unlobed to bi- pinnate. The petioles, petiolules and rachises are conspicu- ously winged by the decurrent bases of leaflets and pinnules. The variable margins are often rather coarsely toothed, sometimes lacerate, the teeth usually protruding beyond the outlines of the blades. The variable lobing and segmentation of its leaves are in part due to environ- mental influence. Plants grown in the greenhouse from wild tubers collected in México (Melchert & Sorensen 6125 & 6233) produced simple to pinnate-pinnatifid leaves. Simple leaves developed during winter months (without supplementary illumination) and the pinnate-pinnatifid leaves developed during spring and summer months or during winter months with supplementary illumination. The geographic range of Dahlia pinnata is sympatric with D. rudis and D. atropurpurea. Dahlia rudis, to which D. pinnata is closely related, is distinguished by its smooth- textured, flexible (when dry) leaves and by its much larger outer involucral bracts (15-25 mm. not 10-14 mm. long). Dahlia atropurpurea is distinguished from D. pinnata by the color of its rays which dry to a dark, almost blackish, purple and by its non-winged petiolules and rachises. To be continued Volume 71, No. 786 including pages 177-365, was issued June 30, 1969. FARLow REF ERENc; RY / Hodora | JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by ALBION REED HODGDON , Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN ROBERT CRICHTON FOSTER Associate Editors ROLLA MILTON TRYON RADCLIFFE BARNES PIKE LORIN IVES NEVLING, JR. Vol. 71 July-September, 1969 No. 787 CONTENTS: Revision of the Genus Dahlia (Compositae, Heliantheae — Coreopsidinae), (concluded) Paul D. Sorensem ............ 367 Winterbud Production and Function in Brasenia Schreberi. Pankin S. Adama Lesser ia. ne 417 Cyperus fuscus in Nebraska and South Dakota. Ronald R. Weedon and H. A. Stevens oe eb 433 A Seventeenth Century Record for Stewartia. u P iss m 434 Parasitism in Ximenia (Olacaceae). k a sss s E 439 (Continued on Inside Cover) The New England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge, Mass, 02138 CONTENTS: — continued Phytoplankton Flora of Newfound and Winnisquam Lakes, New Hampshire. Gerhard K. Gruendling and Arthur C. Mathieson ............ 444 Magnolia tripetala L. and Aralia spinosa L. in St. Louis County, Missouri. Neil A. Harriman |... eem 478 Eragrostis domingensis (Pers.) Steud. New to the United States. Olga Lakela ............ ' eee 479 An Incorrect Correction. Frank S. Crosswhite .......... eee eene nnne 480 A Re-evaluation of Cuphea, Sect. Leptocalyx, Series Bustamantae. Shirley A. Graham. uss 481 Stuart Kimball Harris .......................... n... 493 Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 71 July-September, 1969 No. 787 REVISION OF THE GENUS DAHLIA (COMPOSITAE, HELIANTHEAE — COREOPSIDINAE) PAUL D. SORENSEN (Continued from page 365) 15. Dahlia pteropoda Sherff, Am. Jour. Bot. 34: 147. 1947. TYPE: MEXICO: PUEBLA: Cerro del Gavilán, 8000- 9000 ft., Aug. 1909, Purpus 3851 (Holotype: Uc!). Perennial herb at least 7 dm. tall. Stems about 5 mm. diam., deeply 6-furrowed, more or less pilose, densely so at the nodes, internodes 3.5-7 cm. long. Leaves once-pinnate to bipinnatisect, 13-25 cm. long; leaflets 3-9, elliptical, acute (sometimes abruptly narrowed into the short acuminate tip); basal pinnae 3-8 cm. long, 3-5 cm. wide; sur- faces slightly bicolored, ventrally glabrous or essentially so, dorsally pilose; margins densely ciliolate, the cilia rigid and sharp, regularly serrate to dentate-crenate, the undissected pinnae with 8-16 teeth per side; stipels sometimes present at the basal and second rachis nodes, broadly ovate, 15-22 mm. long, 14-18 mm. wide; petiole 4-11 cm. long, 1-4.2 em. wide, broadly auriculate winged, the wings usually coarsely- toothed, more or less glabrous above, pilose beneath; rachis broadly winged 7-12 cm. wide, glabrous above except at the junction with the pinnae where densely pubescent, pilose beneath. Heads in 2’s, sometimes 3’s, obliquely erect at anthesis; rays light purple, 22-28 mm. long, 8-14 mm. wide, elliptic-ovate, acute or denticulate; outer involucral bracts reflexed at anthesis, obovate, acute, 8-10 mm. long, 1-2.5 mm. wide, glabrous, dorsally reticulate veiny, the 4-8 main veins parallel; inner bracts 11-14 mm. long, 4-6 mm. wide; disc florets yellow; the immature achenes black spotted, slightly spatulate- linear, almost 9 mm. long, 2 mm. or less wide; pappus obsolete or persisting as 2 minute rudiments. Known only from southern Puebla, 7000-9000 ft. Flower- ing August. 367 368 Rhodora [Vol. 71 MEXICO. PUEBLA: vicinity of San Luis Tultitlanapa, Purpus 2536 (GH, UC); Cerro del Gavilan, Purpus 3852 (Topotype: Uc). The leaves of Dahlia pteropoda with their huge auriculate wings at the base of the petioles are so distinctive there can be no confusion with it and any other known species of the genus. The texture of its leaves suggests a close relationship with D. pinnata, an opinion also expressed by Sherff (1947). 16. Dahlia brevis Sorensen, nomen nov. TYPE: MEXICO: MEXICO: on limestone bluffs at Flor de Maria, Aug. 1890, Pringle 3164 (Holotype: Us!; Isotypes: F!, GH!, LCU!, MO!, MSC!, NY!, RSA!, UC!). Dahlia pubescens S. Wats. Proc. Am. Acad. 26: 142. 1892. Type: based on Pringle 3164 as above, non Dahlia pubescens A. Brongn. ex Neumann, Rev. Hort. II. 4: 305. 1845 (see excluded species). Dahlia barkerae sensu Sherff, Am. Jour. Bot. 33: 508. 1946. Herb 4-7 dm. tall, sparingly branched, growing from very shallow tubers. Stems 2-5 mm. diam., many-striate, pubescent, increasingly hairy at the nodes; median internodes 4-8 cm. long. Leaves sometimes simple but more commonly once-pinnate or once-pinnatisect, rarely bipinnatisect, 3.5-14 cm. long; leaflets or divisions 5-1 (-9), opposite, rarely sub-opposite or alternate; the basal pinnae 12-40 mm. long, 4-12 mm. wide, oblong-lanceolate, acute or subacute to almost obtuse; the terminal leaflet or segment smaller than its companions; surfaces light-green or obscurely bicolored, ventrally glabrous or sparingly pubescent, dorsally conspicuously pubescent, especially along the prin- cipal veins; margins ciliolate, the cilia very short, stiff and almost cartilaginous, antrorsely curved, entire or irregularly serrate with lor 2 teeth per side; stipels absent; petiole 7-22 mm. long, 0.5-4 mm. wide, usually winged from the decurrent bases of its lateral divisions, base of petiole often beset with long, broad-based, multicellular hairs; rachis winged, pubescent, especially at the junction with the leaflets. Heads 1-3, rarely 4-6, per plant, obliquely erect at anthesis, on peduncles 10-24 cm. long; outer involucral bracts erect at anthesis, 6-9 mm. long, 2.5-3 mm. wide, broadly obovate, obtuse, conspicuously purple striate with 5-9 parallel veins; inner bracts 10-14 mm. long, 4-7 mm. wide, conspicuously many purple-lined; rays light purple, spreading at anthesis, 10-25 mm. long, 6-14 mm. wide, elliptical, acute or denticulate; disc florets yellow, corolla with purple veins, lobes purple margined or entirely purple; chaff (in fruit) 12-13 mm. long, 7 mm. wide; achenes 6-7 mm. long, 2-3 mm. wide, broadly spatulate, minutely puberulent; pappus obsolete or consisting of 2 minute tubercles. Chromosome number, n — 16. 1969] Dahlia — Sorensen 369 Moist rocky slopes and rocky ledges, 8500-10,000 ft., northwest México state. Flowering Aug.-Sep. MEXICO. MExico: At K-167 near Calpulalpan on route 57, México City to San Juan del Rio, 3100 m., Hawkes et al. 1396 (F); about 10 mi. W. of Atlacomulco at Estacion Bassoco along gravel road to El Oro, 9000 ft., Melchert & Sorensen 6235 A-E (IA). The present species was first described by Watson (1891) who chose for it the name Dahlia pubescens. It is now clear that the name D. pubescens is a later homonym and for this reason I have given the species the new name D. brevis alluding to its diminutive habit. As is evident in the list of synonyms above, the name Dahlia barkerae sensu Sherff (1946, 1955, 1962, 1966) was used with reference to this species. It is clear from the materials at hand and from my observations of living plants both in the field and in cultivation that such usage was incorrect. The name D. barkerae Knowles & Westcott is correctly referable to the taxon which Sherff (1946) had described as the new variety D. scapigera var. arsenei (see table 2). On the texture, color and shape of its leaves, Dahlia brevis strongly resembles a dwarfed D. pinnata sensu stricto. D. pinnata may be distinguished by its greater height, outer involucral bracts reflexed at anthesis, larger leaves, larger heads and chromosome number of n = 32. Several specimens of D. brevis I have examined were pre- viously identified as species of Cosmos and vice versa. Cosmos species are, however, easily distinguished by the tufts of hairs on their filaments and their 4-angled (not obcompressed) achenes. 17. Dahlia rudis Sorensen, sp. nov. TYPE: MEXICO: DISTRITO FEDERAL: Salazar (a railroad station SW. of Mexico City), 2800 m., Sep. 1941, Lyonnet 3372 Holotype: US!). Herba parennis, 9-30 dm. alta. Folia plerumque pinnata, interdum bipinnata, 16-25 cm. longa; foliolis sessilibus vel petiolulatis 9-25 mm. longa; pinnis basilaribus 11-15 cm. longis, 2.5-5 em. latis, foliolis ter- 370 Rhodora [Vol. 71 Ky Z A. Plato 1419 Fig. 5. Dahlia rudis Sorensen. Photograph of type (US), X 1/3. 1969] Dahlia — Sorensen 371 minalibus ovato-lanceolatis, longo-acuminatis vel attenuatis; nervo medio dorsaliter stramineo; marginibus regulatim serratis cum dentibus in quoque latere 10-16; stipellis in rachidis nodis basilaribus et interdum secundis affixis; petiolo valde lunato in sectione trans- versale, alato saepe pinnis basilaribus decurrentibus, basi valde am- plexicaule praesertim ad apicem plantae. Capitula solitaria vel in binis vel trinis, pedunculis 7-16 cm. longis; involucri squamis exterioribus reflexis sub anthesi, 15-25 mm. longis, 4-10 mm. latis, dorsaliter 6-10-nervis, interdum intervenium reticulatis. Flores ligulati lilacini, 2-3 em. longi, 1-1.5 em. lati. Fig. 5. In high mountains, 8600-10,000 ft., from near Amecameca in eastern México state westward in northern Morelos, Dis- trito Federal, and to the vicinity of Temascaltepec, south- western México state. Flowering Aug.-Oct. MEXICO. HIDALGO: 10 km. NNE of Apam, West H-5 (wis). MEXICO: Distr. of Temascaltepec, Hinton 1885 (Mo, vs), 8332 (GH, us); Amecameca, Kuntz 23660 (NY); Criadero, Matuda 19411 (MEXU, US) ; W. of Transfiguración, Russell & Souviron 215 (Us); Cerro del Pueblo de Hoacalco de Cuautitlán, Salasar s.n. (MEXU). MORELOS: Lago Zempoala, Matuda 25575 (MEXU, NY); from cultiva- tion: UNITED STATES. CALIFORNIA: grown from Mexican seed at Golden Gate Park, Walther s.n. (CAS). Among the specimens of Dahlia rudis cited above only two have labels providing information on the height of the plants (3 ft. & 3 m.), only one gives its habit (herb) and none provide what appear to be median leaves. The accompanying description must, therefore, be considered provisional. Certain characteristics of Dahlia rudis suggest a rela- tionship to the 'tree-dahlias," the leaflets with long attenuate apices and tan colored midveins on the undersur- face and the large, often reticulately-veined, outer involucral bracts. Several of the specimens have, in fact, been identi- fied with names of "tree-dahlias" as D. maximiliana, and D. excelsa. Dahlia rudis is distinguished by its conspicuous sheathing petioles (of the upper leaves) and its very large outer bracts. Most specimens show relatively few heads per branch (2-4) but one, at least, shows several more buds rather close to anthesis. 372 Rhodora [Vol. 71 18. Dahlia moorei Sherff, Bot. Leafl. 5: 22. 1951. TYPE: MEXICO: HIDALGO: near K-238 along route 85, limestone ledges and streamside thickets, Barranca de San Vicente, 1800-2000 m., 24 Sep. 1941, H. E. Moore 5068 (Holotype: F!; Isotypes: BH!, GH!, UC!). Herb 18-25 dm. tall. Stem 5-8 mm. diam, drying to reddish-brown or tan, many-striate, glabrous; internodes 12-21 cm. long, hollow. Leaves bipinnate, 20-30 cm. long; leaflets 5-9, the basal pinnae 14-17 cm. long, sessile or stalked, the petiolules 12-30 mm. long; ultimate segments opposite or alternate on the rachilla, 3.5-7 em. long, 1.5-2.3 cm. wide, ovate-lanceolate, apices long-attenuate or flagelliform; surfaces slightly bicolored, lighter green beneath, ventrally pubescent, the hairs occurring between the veins, dorsally puberulent, the hairs following the veins; margins ciliolate, the cilia slender and flexuous, coarsely and regularly serrate, the terminal leaflet with 10-18 teeth per side; stipels present at the basal and second nodes on the rachis, 1.5-4 cm. long, 0.8-2 cm. wide; petiole 2-12 cm. long, glabrous, crescent-shaped in cross-section, longitudinally grooved above; rachis glabrous except at the junction with the pinnae. Heads subnumerous, 6-12 per main branch, obliquely erect, on peduncles 4.5-16 cm. long; outer involucral bracts reflexed at anthesis, 10-17 mm. long, :1.5-4 mm. wide (near the base), linear-lanceolate, becoming long-attenuate or flagelliform, glabrous, dorsally 5-8-lined, ventrally sometimes strongly cross-wrinkled; inner bracts 13-15 mm. long, 4-7 mm. wide; rays deep rose-purple, 2-2.6 cm. long, 10-14 mm. wide, elliptical, acute or denticulate; disc florets yellow below, purple above; style branches purple; anthers yellow with purple veins; achenes linear- spatulate, glabrous or puberulous, many-suleate, 6.5-7 mm. long, 1.8-1.6 mm. wide, constricted slightly below the summit to form a black disc less than 1 mm. wide and barely 0.4 mm. high, pappus obsolete or consisting of 2 minute rudiments. Known only from the type locality in the Barranca de San Vicente, Hidalgo. Flowering Aug. MEXICO. HIDALGO: Near K-238 along route 85, Bar- ranca de San Vincente between Zimapán & Jacala, Moore & Wood 4441 (Topotypes: A, BH, NY, UC, WIS). Certain floral characteristics found in Dahlia moorei such as the purple style branches and the thickened, black apical achenial disc are unique among the dahlias. Its long- attenuate, almost flagelliform outer involucral bracts are like those of D. hintonii but in that species the bracts are dorsally pilose, not glabrous. The leaves of D. moorei, with 1969] Dahlia — Sorensen 878 ovate-lanceolate leaflets, their apices long acuminate and margins evenly serrate, resemble those of D. rudis from which it is distinguished by its flagelliform (not broadly obovate or oblanceolate) outer bracts and its large number of heads (6-12 not 1-4) per branch. 19. Dahlia hintonii Sherff, Am. Jour. Bot. 34: 138. 1947. TYPE: MEXICO: GUERRERO: Distrito de Minas, Toro Muerto, on boulder in pine forest, 2150 m., 27 July 1939, Hinton et al. 14499 (Holotype: GH!; Isotypes: MICH!, NY!, UC!, US!). Herb 10-15 dm. tall. Stems unusually slender, 2 mm. or less diam., tan, glabrous or a few soft hairs at nodes. Sub-median* leaves petiolate, once-pinnate, 11-17 cm. long; leaflets 3-5, basal pinnae 5-8.5 cm. long, 1-2 cm. wide, lanceolate, becoming long-acuminate or attenuate; surfaces slightly bicolored, dorsally lighter green, the upper glabrous or essentially so, the lower densely pilose; margins finely ciliolate, the cilia antrorsely curved, regularly or irregularly serrate, 3-7 teeth per side; stipels absent; petiole 1-3 em. long, puberulent, crescent-shaped in cross section, grooved above; rachis pubescent, increasingly so at the junction with the primary pinnae. Heads 4-8 per branch, erect or obliquely erect; outer involucral bracts linear or slightly linear-spatulate, reflexed at anthesis, 11-20 mm. long, 1-2 mm. wide, dorsally pilose, ventrally pubescent, becoming almost glabrous near tips; inner bracts 13-16 mm. long, 4.5-7 mm. wide, pubescent in the middle, glabrous toward margins; rays ovate, acute or denticulate, 13-18 mm. long, 8-10 mm. wide, said to be purple; disc florets yellow with lobes of the corolla dark; achenes not seen. Known only from the type locality in south-central Guerrero, just west of Cerro Teotepec. Flowering July. Certain distinctive characteristics of Dahlia hintonii are unique in the genus. Its once-pinnate leaves are composed of unusually narrow, lanceolate leaflets 3-5 times longer than wide and its rather long and slender outer involucral bracts are minutely but densely pubescent. 20. Dahlia mollis Sorensen, sp. nov. TYPE: MEXICO: HIDALGO: 12 mi. S. of Jacala, 1 m. S. of Minas Viejas, along route 85, about 7800 ft., 30 Aug. 1965, Melchert & Sorensen 6209 (Holotype: IA!). *No median leaves present on Hinton 14499. 374 | 125130 I | TE vsvensery "pesono ` Fig. 6. Rhodora [Vol. 71 i Holotype MEXICO SHEET | + £ STATE or HIDALGO Dahlia mollis Soren. About 1 f Pla aw at ah. Plate 1420 Dahlia mollis Sorensen. Photograph of type (IA), X 1/3. 1969] Dahlia — Sorensen Sid Herba perennis, 10-16 dm. alta. Caules omnino dense pubescentibus, nodis copiose pilos. Folia pinnati, 13-29 cm. longi, foliolis 3-5, pinni basilaribus 4-11 cm. longis, 3.5-5 cm. latis, late ovatis, apice acutis, basi contractis vel rotundatis (foliolis terminalis interdum basi truncatis vel subcordatis), ventraliter pubescentibus, pili brevis, dorsaliter subtomentosis; marginibus dense et conspicue ciliatis, ser- rato-crenatis, dentibus, in quoque latere 7-12; stipellis absentibus; rhachi et petiolo pubescentibus, nodis rhacheos floccosis. Capitula in binis vel trinis, pedunculis glabris, 3.5-14 cm. longis; involucri squamis exterioribus effusis vel reflexis sub anthesi, 7-11.5 mm. longis, 3-3.6 mm. latis, dorsaliter 5-9 nervis, glabris. Flores ligulati lilacini, 2.7- 3.5 em. longi, 12-18 mm. lati. Chromosomatum numerus: n — 16. Fig. 6. Rocky slopes and ledges, 7800-8400 ft., western Hidalgo, México. Flowering Aug.-Sep. Map 5, open circles. MEXICO. HIDALGO: 13 to 14 mi. S. of Jacala, 2 mi. S. of Minas Viejas along route 85, 8000 ft., Melchert & Sorensen 6210 A-C (IA), Melchert et al. 6504 (IA); trail from Zimapán to mines of El Monte, 7800-8400 ft., Moore & Wood 4469 (BH). The most distinctive feature of Dahlia mollis is the dense pubescence of the dorsal leaf surface. Only one other species in the genus, D. hintonii, has leaves that are as velvety and soft to the touch. The shape and size of the leaflets of the two are quite distinct, those of D. hintonii being narrowly lanceolate and smaller. The shape and size of the leaves of Dahlia mollis resemble those of D. barkerae, but in that species the upper surface is dark, glossy green and rugose and the lower surface is not nearly so tomentose. Furthermore, plants of D. barke- rae are much smaller (3-7 dm. tall) and have a higher chromosome number (n = 382). 21. Dahlia atropurpurea Sorensen, sp. nov. TYPE: MEXICO: MÉXICO: 4 mi. SE. of Sultepec on Cerro Cualtepec, about 8500 ft., 27 Sep. 1966, Melchert, Sorensen & Crawford 6498 (Holotype: IA!). Herba perennis, 15-22 dm. alta. Folia pinnata ad bipinnata, 16-25 em. longa; foliolis primariis 5-7, pinnis basilaribus ovato-lanceolatis, 6-13 cm. longis, 2.3-3.8 cm. latis, petiolulatis 1.3-3 em. longis, apice longo-acuminatis, basi rotundatis; paginis bicoloribus, dorsaliter pallido-viridibus vel cano-viridibus, parce pubescentibus, praesertim ad nervos, ventraliter glabris, rugosis; marginibus serrato-crenatis, 376 Rhodora [Vol. 71 dentibus in quoque latere 6-10; petiolis 2.4-5.5 cm. longis; stipellis nodis in basilibus et medianibus rhachidibus affixis. Pedunculis 6.5-11 cm. longis; exterioribus involueri squamis reflexis sub anthesi, 5-9.5 mm. longis, 3-5 mm. latis, dorsaliter et nervis purpureis 5-9, glabris, ventraliter parce adpresso-pubescentibus vel glabris. Flores ligulati atropurpurei vel atropurpureo-nigri, 2.2-3.5 cm. longi; flores tubulosi lutei vel lutei cum apicibus purpureis. Chromosomatum numerus: n — 32. Fig. 7. Known only from the type locality, about 8500 ft., on Cerro Cualtepec near Sultepec, México state and from near Chiriagua, Distr. of Minas, Guerrero. Flowering Sep. MEXICO. GUERRERO: Chiriagua, distr. of Minas, 2140 m., Hinton et al. 10665 (GH). Several characteristics of Dahlia atropurpurea distin- guish it from other species or species complexes in Dahlia. Its rays, which dry to a dark, almost blackish purple, are not found in any other of the purple-rayed species except D. cardiophylla from which it is easily separated by its compound (not simple) leaves. Dahlia atropurpurea is sympatric with D. pinnata and D. rudis, two species with which it might possibly be con- fused. Each of these species is distinguished by its leaf characteristics. The leaves of D. atropurpurea are charac- terized by: (1) drying to a firm and smooth (not coarse nor scabrous) texture, (2) the absence of wings on the rachis, petiole, and petiolules, (3) the 6-10 teeth per margin and (4) the rounded (not attenuate-narrowed) bases of the petiolulate leaflets. The rather coarse-textured, some- times scabrous leaves of D. pinnata are winged, often strongly so, on the rachis, petiole and petiolules and their leaflet bases are attenuate-narrowed. The leaves of D. rudis dry to a smooth texture but are rather flexible and thin and their leaflets are attenuate-narrowed at the base and have 10-16 teeth per margin. D. rudis is further dis- tinguished from D. atropurpurea by its larger outer in- volucral bracts (15-25 mm. not 5-9.5 mm. long). On leaf dimensions and segmentation there is consider- able overlap between D. atropurpurea and the D. australis- sherffii species complex. In addition to their geographic 1969] Dahlia — Sorensen 377 425135 HERBARUM STATE 7 ; oF oMa SY Fo mema 2 Ç TOT Ct Be ae s ae A [ARNOLD ARBORETUM war di, aoe S w ¿ = . , MEXICO: wareon mexico SHEET 2 í Z, Dahlia atropurpurea Soren. Holot pe Near summit ef Cerro de Cunltepec aboot 4 km. * $8. of Sultepec. Plants growing on rocky cliffs in n pine sone P : along road leading to summit of mountain, About 8500 feet, Cytological Voucher: n = 32 (30 UI & 1 IV) Themes B. Melchert, Posi D. kerana, Doniol 3, Crowd Ma. £403 Hodis in the s Herbarium of The Unioorsity of losa nn 1968. Plate 1421 Fig. 7. Dahlia atropurpurea Sorensen, Photograph of type (IA), x 1/3. 378 Rhodora [Vol. 71 distinctions the latter species are best distinguished on the nature of their attentuate-narrowed (not rounded) leaf bases. 22. Dahlia australis (Sherff) Sorensen, comb. & stat. nov. TYPE: MEXICO: OAXACA: Cerro de San Felipe, 2500 m., 1 Sep. 1897. Conzatti & Gonzales 543 (Holo- type: GH!). Herbaceous perennial, 4.6-12.3 dm. tall, branching freely or un- branched except in the flowering portion. Stem 2-6 mm. diam., essentially glabrous to conspicuously pubescent, sometimes scabrous on the lower portions, hairs becoming dense at the nodes; internodes 3.5-24 (-28.5) cm. long, hollow or sometimes solid. Leaves usually once-pinnate but ranging from merely pinnatifid to bipinnate, 3-21 cm. long; primary leaflets 3-5(-7), basal pinnae 1.5-9.5 cm. long, ultimate segments oblong or rhombic-ovate to ovate-lanceolate, apex acute or acuminate, base attenuate-narrowed, sessile or stalked, petiolule 1-13 mm. long, terminal leaflet often 3-lobed, sessile or pseudopetiolulate; surfaces usually bicolored, green or dark green above, gray or silvery-green and salient veiny beneath, the upper rugose, glabrous or with a few scattered hairs near the margins and apex, often with a line of fine hairs along the principal veins, the lower usually pubescent esp. along the veins; margins ciliolate, regu- larly or irregularly serrate to crenate-dentate with 2-10 teeth per side, slightly revolute; stipels usually present, attached at the basal (rarely also at the second or median) rachis node; petiole 0.9-10 cm. long, sometimes very slightly winged in the distal parts, grooved above and crescent-shaped in cross-section, almost glabrous to notice- ably pubescent; rachis usually narrowly winged, glabrous or pubes- cent, the hairs becoming dense at the junction with the primary pinnae. Heads solitary or in 2’s and 3’s, erect or obliquely erect; outer involucral bracts reflexed at anthesis, 7-15 mm. long, 1.5-6 mm. wide, linear-oblanceolate to oblong-spatulate or obovate, dorsally glabrous, 4-8-lined, ventrally sometimes minutely appressed pubescent; inner bracts 10-18 mm. long, 3-9 mm. wide, brownish or sometimes suffused with purple at the tips, rarely purple throughout; rays purple; chaff resembling inner bracts; dise florets 39-67, 6-9 mm. long, the limb ranging from yellow throughout to purple throughout with many intermediates, corolla lobes often reflexed or revolute at anthesis; style branches long attenuate, 2.5-3.2 mm. long, less than 0.8 mm. wide; achenes (derived from greenhouse plants) 7-11.4 mm. long, 2-3 mm. wide, linear-oblanceolate to obovate, blackish or speckled gray; pappus obsolete or consisting of 2 minute rudiments. Chromo- some numbers, n = 16 & 32. 1969] Dahlia — Sorensen 879 As treated here Dahlia australis (Sherff) Sorensen is composed of four infraspecific taxa as outlined and de- scribed below. These are var. australis, var. chiapensis, var. serratior, and var. liebmannii (see Table 2). Each of these is geographically isolated (see Map 6) and, to a limited extent, morphologically distinct. Excepting var. liebmannii, a little-known taxon marked by its rather small leaves and short petioles, none of the varieties is easily distinguished on any single morphologic feature. Their distinguishing characteristics are primarily of a quantita- tive nature and are discussed in turn following the taxo- nomic treatments of each variety. For the convenience of the user, the following key to the infraspecific taxa of Dahlia australis is provided. A. Petiole 0.9-1.3 em. long; leaves 3-8 CM. LONG. eerste ttetseteens rr rco ID ON ONERE 22b. D. australis var. liebmannii. A. Petiole (2-)3.8-10 em. long; leaves 7.5-21 cm. long. B. Plants mostly of a compact growth habit with median inter- nodes (1-)3-15 em. long, stems with leaves often inferiorly clustered; outer involucral bracts 7-12 mm. long. C. Terminal leaflet often 3-lobed, sessile with strongly de- current margins forming a pseudopetiolule; leaflets (of l-pinnate leaves) oblong-ovate, markedly bicolored, the lower surface much lighter or silvery green. .......... * ME MEME 22a. D. australis var. australis. C. Terminal leaflet usually not lobed nor divided and without strongly decurrent margins forming a pseudopetiolule; leaflets (esp. the terminal one) often narrowly rhombic- ovate, not strongly bicolored, the lower surface rarely markedly lighter or silvery-green. ....... nm tp n du TI 29c. D. australis var. chiapensis. B. Plants mostly taller growing with longer internodes (7-)13-19 (-28.5) cm. long, stems leafy; outer involueral bracts (9-) 25 t5 mm. Ong. "eeseseeeessseeszse 29d. D. australis var. serratior. Dahlia australis is apparently quite closely related to D. sherffii of Durango and Chihuahua of northwestern México (see Map 6). Indeed, attempts to derive artificial hybrids between them have been highly successful. The hybrid plants are very robust, flower the first season of growth, possess no visible deleterious irregularities at ana- 380 Rhodora [Vol. 71 phase and produce virtually 100% stainable pollen (using cotton-blue dissolved in lactophenol). Although geographically isolated by about 900 miles, there is a considerable amount of morphological overlap between the taxa comprising the Dahlia australis-sher ffi species complex. Fox example, a few individuals of D. sherffii display the markedly bicolored leaves which charac- terize D. australis. Similarly, the leaves on a few specimens of D. australis have upper leaf surfaces just as pubescent as those typically found in D. sherffii. Also, the dimensions and, to a limited extent, the segmentation of leaves and leaflets of both species overlap completely, and both pro- duce leaflets rather attenuate-narrowed (not rounded) at the base. Map 6. Distribution of the Dahlia australis-sherffii species com- plex: Dahlia australis var. australis (Sherff) Sorsensen, open circles; Dahlia australis var. liebmannii (Sherff) Sorensen, diamond; Dahlia australis var. chiapensis Sorensen, triangles; Dahlia australis var. serratior (Sherff) Sorensen, stars; Dahlia sherffii Sorensen, closed circles. 1969] Dahlia — Sorensen 381 Dahlia australis is distinguished from D. sherffii (1) by its markedly bicolored leaves which are much lighter, almost silvery-green beneath, (2) by the mostly glabrous (not pubescent) and rugose (not smooth) upper surface of its leaves, and (3) by its sometimes slightly revolute (not flattened) leaflet margins. 224. Dahlia australis (Sherff) Sorensen var. australis, TYPE: MEXICO: oaxacA: Cerro de San Felipe, 2500 m., 1 Sep. 1897, Conzatti & Gonzalez 543 (Holo- type: GH!). Dahlia scapigera var. australis et f. australis Sherff, Am. Jour. Bot. 34: 143. 1947. Type: that of Dahlia australis var. australis. Dahlia scapigera var. australis f. purpurea Sherff, Ibid. p. 145. Type: Conzatti & Gonzalez 402, Oaxaca, México (Holotype: GH!). Herbaceous perennial 7-12 dm. tall. Stem 2-5 mm. diam., glabrous or pubescent; internodes 3.5-12 cm. long. Leaves usually once-pinnate but ranging from pinnatifid to bipinnate, 7.5-21 cm. long; basal pinnae 2-9.5 cm. long, sessile or rarely stalked, petiolule 10-13 mm. long, terminal leaflet often 3-lobed, sessile or pseudopetiolulate; sur- faces conspicuously bicolored, green or dark green above, gray or silvery-green and salient veiny beneath, the upper rugose, glabrous or with a few scattered hairs near the margins and apex, the lower pubescent, esp. along the veins; margins irregularly serrate to crenate-dentate, 2-10 teeth per side, slightly revolute; petiole 2-8 (-10) em. long. Heads with outer involucral bracts 7-12 mm. long, 1.5-3 mm. wide, narrowly oblanceolate; rays purple, 2.8-3.5 cm. long, about 1.5 em. wide, ovate acute; disc florets 39-67, 6-9 mm. long, the limb from yellow throughout to purple throughout with many inter- mediates, corolla lobes often reflexed or revolute; achenes 8-9 mm. long, 2.8-3 mm. wide, linear-oblanceolate. Chromosome numbers, n = 16 & 32. Known only from ledges and rocky slopes at altitudes of 7000 to 11,000 ft. in the vicinity of Cerro de San Felipe and Cerro Zempoalteptl, north-northeast and northeast of Oaxaca City, Oaxaca, México. Flowering June-Oct. Map 6, open circles. MEXICO. oaxaca: summit and slopes of Cerro de San Felipe near Ciudad Oaxaca, Comzatti 55 (GH, US), 4078 (MEXU), 544 (GH, us), Nelson 1161 (vs), Pringle 5620 (GH, MEXU), C. L. Smith 294 (14) ; various localities from about 12 to 48 mi. N. of Ciudad Oaxaca along route 175 to Valle Nacional, Cronquist 9655 (MEXU, MICH, NY, TEX), King 2076 (MICH), Krueger & Gillespie 35 (CAS, GH, MO), 382 Rhodora [Vol. 71 Melchert & Sorensen 6173 A-C (1A), 6174 A-G (1A), 6184 (1A), 6185 A-D (IA), 6192 A-E (IA), Roe & Roe 2004 (A, wis); slopes of Cerro Zempoalteptl, Hallberg 896 (MICH), Nelson 640 (vs). Sherff (1947, 1955) established Dahlia australis var. australis as var. australis of D. scapigera sensu lato (see Table 2). Within his var. australis he described two sub- varietal taxa, f. australis with yellow disc florets and f. purpurea with purple disc florets. In the present treat- ment Sherff’s var. australis has been separated from D. scapigera sensu stricto (see Table 2) and elevated to the rank of species. Moreover, neither of the two color forms recognized by him are treated as separate taxonomic units inasmuch as the intensity and the amount of the disc corollas so colored varies from plant to plant within a Single population. Dahlia australis var. australis is the only taxon within this species for which cytological information is available. In this regard, it is a rather interesting taxon because among the purple-rayed dahlias it is the only one repre- sented by both diploid and tetraploid races. During August 1965, Dr. Thomas Melchert and I sampled 5 populations (representing 12 individual chromosome counts) along a 41-mile stretch of highway (route 175) leading northeast from Oaxaca City to Valle Nacional: at mile 12 we counted ^ = 32 (6173); at mile 15, n = 32 (6174); at mile 17, n = 16 (6184) ; at mile 21, n = 32 (6185) ; and near mile 41, n = 16 again (6192). To date I have been unable to detect morphologie markers which would distinguish the diploid and tetraploid populations or individuals. Morphologically, the Oaxacan Dahlia australis war. australis is distinguished from its allied varieties by several characters of its leaves: (1) surfaces markedly bicolored, much lighter green, almost silvery-green beneath, the upper noticeably rugose, glabrous, or with a few scattered hairs near the margins and apex, the lower often strongly reticu- late veiny, pubescent, esp. along the principal veins; (2) ‘Similar color variation is met with in several other Dahlia SPP., D. merckii, D. sherffit, D. brevis, D. imperialis, and D. dissecta. 1969] Dahlia — Sorensen 383 the terminal leaflet often 3-lobed, sessile or pseudopetiolu- late; (3) the basal leaflet of 1-pinnate leaves oblong-obvate ; (4) margins crenate-dentate, the teeth often protruding slightly beyond the outline of the blade, 22b. Dahlia australis var. liebmannii (Sherff) Sorensen, comb. nov. TYPE: MEXICO: HIDALGO (?): be- tween San Andres and San Miguel, Oct. 1842, Lieb- mann 688 (—9782) (Holotype: F!; Isotype; US!). Dahlia scapigera var. liebmannii Sherff, Am. Jour. Bot. 143. 1947. Type: that of Dahlia australis var. lieb- mannii. Herb, at least 4.6 dm. tall. Stems 2-3.5 mm. diam., glabrous or sparsely pubescent with a few scattered soft hairs; internodes 14-24 em. long. Leaves once-pinnate, 3-8 cm. long; the basal pinnae 1.5-4 em. long, 0.5-1.3 em. wide, ovate or ovate-lanceolate, sessile or short- stalked, the petiolule 1-3 mm. long; surfaces bicolored, conspicuously lighter green beneath, rugose above, the upper glabrous or puberulent, the lower pubescent, esp. along the veins; margins evenly serrate, 5-8 teeth per side, slightly revolute; petioles 9-13 mm. long. Heads with outer involucral bracts 7-11 mm. long, 1.5-2 mm. wide, linear- oblanceolate; rays light purple, 1.7-2 cm. long, 8-10 mm. wide; disc florets purple-tipped; achenes not seen. Known only from the type locality. Map 6, diamond. Very little can be said of this rather obscure taxon. A number of characters preserved in the type material (strongly bicolored leaves, slightly revolute margins, rugose texture of the upper surface of leaflets, slender stems) indi- cate that assigning it varietal rank under Dahlia australis is the most reasonable course to follow. This action is taken despite its apparent long geographic disjunction from the center of distribution of D. australis var. australis. Collections of other Dahlias species by Frederick Lieb- mann have been definitely located in northeastern Puebla end adjacent Veracruz. This suggests that the type locality “between San Andres and San Miguel” could be a reference to two small mountain villages located east-northeast of Ixmiquilpan in central Hidalgo. Their location is not far distant from areas known to have been visited by this early collector. 384 Rhodora [Vol. 71 Dahlia australis var. liebmannii may be distinguished from other D. australis vars. by its smaller leaves (3-8 cm. long) and its rather short petioles (0.5-1.3 cm. long). 22c. Dahlia australis var. chiapensis Sorensen, var. nov. TYPE: MEXICO: Chiapas: about 9 mi. SE. of San Cristóbal de las Casas along route 190, among boulders and on rocky ledges in zone of pine and oak, about 7500 ft., 28 Sep. 1966, Melchert, Sorensen, & Craw- ford 6458 (Holotype: IA!). Herba perennis 4.5-7.5(-9.6) dm. alta, foliis aggregatis in caulibus base dense foliatis et internodiis 1-3 cm. longis, vel foliis plus dis- tantibus in caulibus elongatis et internodis 5-15 cm. longis. Caules superne glabri, inferne praesertim ad nodis plus pubescentes. Folia pinnatifida ad pinnata, rarenter bipinnata, 9-18 cm. longa; foliolis 3-5, angusto-ovatis ad rhombico-ovatis, pinnis basilaribus 3-7 cm. longis, 0.8-3 cm. latis, saepe asymetricis cum lobo singulari sessili prope basin in margine proximali; paginis bicoloribus, dorsaliter pallido-viridibus vel cano-viridibus, praesertim ad nervos pubescenti- bus, ventraliter vere glabris; marginibus serratis ad dentato-crenatis dentibus in quoque latere 3-6; stipellis, ubi praesentia, ovatis ad rhombicis; petiolis 3.8-9.5 cm. longis. Capitula (1-)2-3 in quoque ramo, erecta ver subcernua; involucri squamis exterioribus effusis vel reflexis sub anthesi, 7-11 mm. longis, 1.8-3 mm. latis. Flores ligulati lilacini, + 2.4 cm. longi, 1.5 em. lati; flores tubulosi 44-52, luteis, interdum apicibus lilacini, Fig. 8. Rocky slopes and ledges, 7000-8500 ft., in highlands of Central Chiapas. Flowering Aug.-Nov. Map 6, triangles. MEXICO: CHIAPAS: from about 9 mi. SE. to about 6 mi. W. of San Cristóbal de las Casas along route 190, Breedlove 7095 (ps, F, MICH), 7307 (DS, F, MICH). 13762 (DS, MEXU), Breedlove & Raven 13412 (ps), Ghiesbreght 154 (GH, NY), 558 (GH, MO, NY), MacDougall s.n. (F), 365 (us). The geographic range of Dahlia australis var. chiapensis is separated from that of var. australis in Oaxaca by rough- ly 250 miles. As with most taxa in the genus Dahlia, these varieties are found only at rather lofty elevations. It is important, therefore, to note that the elevations of the area between the distribution centers of vars. australis and chiapensis are much below those of their known localities. This is easily determined from topographic maps of the area. It is unlikely that this intervening area offers suitable 1969] Dahlia — Sorensen 885 ARNOLD ARBORETUM | MEXICO: STATOR CHIAPAS Dahlia australis var. chiapensis Sore About 9 mi SK of San Cristobal de las Casas Along route 190. T E Thomas B. Malchert, Foui D. Seemen, Daniel 3. Crestor Me DOTE Tessossóv Modes in the Cocoquaidinam. $ HOLOTYPE Metsan e! TR. te A ga Plate 1422 Fig. 8. Dahlia australis var. chiapensis Sorensen, Photograph of type (IA), X 1/3. 386 Rhodora [Vol. 71 habitats where other populations of D. australis would be found. For these reasons, it is believed that these varieties are effectively isolated from one another by the 250-mile disjunction between their distribution centers. Dahlia australis var. serratior of Guatemala, a close ally of var. chiapensis, is similarly isolated by about 130 miles and the absence of suitable habitats in the area between their dis- tribution centers. As mentioned in the discussion of Dahlia australis the features which mark the varieties of this species are mostly quantitative in nature. Thus, while closely resembling var. australis and var. serratior, var. chiapensis may be distin- guished by the following characteristics: (1) mostly com- pact growth habit with leaves crowded near base; (2) undersurface of leaflets lighter green, rarely markedly lighter or silvery-green; (3) terminal leaflet usually un- lobed or divided and without strongly decurrent margins forming a pseudopetiolule; (4) leaflets, esp. the terminal one, often narrowly rhombic-ovate; (5) basal pinnae often asymmetrical with a single sessile lobe near the base of the blade on the proximal edge. Without careful scrutiny these lobes tend to resemble stipels. 22d. Dahlia australis var. serratior (Sherff) Sorensen, comb. & stat. nov. TYPE: GUATEMALA: Dept. HUEHUETENANGO: rocky bushy bank, San Juan Atitan, 8200 ft., 9 Sep. 1934, Skutch 1180 (Holotype: GH!). Dahlia scapigera var. scapigera f. serratior Sherff, Am. Jour. Bot. 34: 142. 1947. Type: that of Dahlia australis var. serratior. Perennial herb, 6.4-12.3 dm. tall. Stems 5-6 mm. diam., glabrous or sparsely pubescent throughout, sometimes slightly scabrous on lower portions; internodes (7-)13-18.5(-28.5) cm. long. Leaves once- pinnate to pinnate-pinnatifid, 12-20 cm. long; leaflets oblong or broadly rhombic-ovate, the basal pinnae 4-8 cm. long 1.3-3.7 em. wide, sessile or pseudopetiolulate, rarely stalked, the terminal leaflet some- what larger than its companion leaflets; surfaces bicolored, lighter green beneath, the upper rugose, usually glabrous, sometimes slightly pubescent near margins and apices, the lower glabrous to distinctly pubescent, conspicuously veiny; margins regularly serrate to dentate- 1969] Dahlia — Sorensen 387 crenate, 5-9 teeth per side; petiole (3-)6-10 em. long. Heads with outer involucral bracts (9-)12-15 mm. long, 2-6 mm. wide, oblong spatulate or more often obovate, strongly tapered toward the base; rays light or rosaceous purple, ovate, 3.4-4 cm. long, 1.4-1.9 cm. wide, acute and denticulate; disc florets 50-53, corolla yellow, tube about 8 mm. long; achenes linear-oblanceolate, 10-11.2 mm, long, 2-2.2 mm. wide. Rocky banks, ledges and cliffs, 8200-12,100 ft., high- lands of southwestern Guatemala. Flowering Aug.-Oct. Map 6, stars. GUATEMALA: HUEHUETENANGO: in the Sierra de los Cuchumatanes, esp. in the vicinity of San Juan Atitan, J. R. Johnston 1975 (F), Skutch 1244 (GH), Steyermark 50318 (F), 51974 (F). SOLOLA: Los Encuentros, J. R. Johnston 986 (F). TOTONICAPAN: woods between Totoni- capán and crest of mountains on road to Los Encuentros, Moore & Cetto 8193 (BH). From Dahlia australis vars. australis and chiapensis, var. serratior may be distinguished by several characters as follows: (1) plants producing leafy stems (not compact with leaves clustered near the base) ; (2) midrib of leaflets often tan-colored beneath; (3) terminal leaflets not 3-lobed nor divided, broadly rhombic-ovate; (4) basal pinnae mostly pseudopetiolulate, rarely stalked; (5) outer in- volucral bracts typically rather large, (9-)12-15 mm. long. 23. Dahlia sherffii Sorensen, sp. nov. TYPE: MEXICO: DURANGO: 24 mi. W. of Durango, 2 mi. 8. of El Soldado along route 40, Durango-Mazatlan, among boulders on a steep slope above highway in a plateau region of pine and oak, about 7500 ft., 4 Sep. 1966, Melchert, Sorensen & Crawford 6288 (Holotype: IA!). Herba perennis, 6-13 dm. alta. Caules plerumque glabri, interdum pilis fasciculibus in nodis. Folia pinnata ad bipinnata, 11-27 cm. longa; foliolis primariis 3-7; pinnis basilaribus 6-13 cm. longis, seg- mentis ultimis ovato-lanceolatis, acutis, basi attenuatis, interdum obliquis; paginis leviter bicoloribus, abaxialiter pallido-viridibus vel cano-viridibus ventraliter leviter pubescentibus, dorsaliter pubes- centibus, praesertim in nervis; marginibus dentatis vel grosse serratis, dentibus 2-8 in quoque latere; stipellis plerumque in nodis basilaribus, 388 Rhodora [Vol. 71 interdum in secundis etiam. Pedunculis 5-22 cm. longis; involucri squamis exterioribus effusis vel reflexis sub anthesi, 9-12 mm. longis, 3-4.2 mm. latis, ventraliter, glabris vel adpresso-pubescentibus dorsa- liter 5-12-lineatis. Flores ligulati lilacini inferne saepe luteo-maculati, 2.8-4.2 cm. longi, 1.8-2.1 cm. lati; flores tubulosi 56-68, lobis corollae effusis nec reflexis nec revolutis; acheniis 7.1-11.5 mm. longis, 2.2-3 mm. latis. Chromosomatum numerus: n — 32. Nomen celebrare Earl Edward Sherff (1886-1966), hominem longe studiosum Core- opsidinae. Fig. 9. Open or lightly-wooded rocky slopes, ledges, fields, and roadsides, 6000-8500 ft., Sierra Madre Occidental of Chi- huahua and Durango. Flowering Aug.-Sep. Map 6, closed circles. MEXICO. CHIHUAHUA: vicinity of Pilares de Majalca about 40 mi. NW. of Chihuahua City, Hawkes et al. 1235 (F), Le Seur 24 (CAS, F, GH, MO, TEX, UC), White 2383 (GH); 10 to 16 road mi. S. of Madera on road to Temósachie, Melchert et al. 6274 A-C (IA), 6272 pop (1A), Muller 3466 (GH, UC); vicinity of Cerro Mohinora, Gentry et al. 17993 (US) Straw & Forman (MICH); vicinity of Majorachic (Maguarichic), Knobloch 1188 (MICH, MO), 5292 (F, MSC). DURANGO: from 34 to 63 road mi. N. of Coyotes RR., Maysilles 7512 (F, MEXU, MICH, NY), 7928 (MICH), 8374 (MICH), 8401-A (MICH); 20 km. S. of El Salto at Arroyo del Infierno, Gordon 74 (MICH); 24 mi. SW. of Durango along route 40, King 3743 (TEX); 36 mi. W. of Durango along route 40, Melchert & Sorensen 6031 (IA), 6032 (IA); 10 mi. E. of La Ciudad along route 40, Melchert et al. 6304 (IA); 21.7 mi. NE. of El Paraiso along road to El Salto, Ownbey & Ownbey 1981 (F, UC, US). Dahlia sherffüi is a new species which was discovered among those taxa formerly treated by Sherff (1955) as part of D. scapigera sensu lato (see Table 2). Although geographically separated by about 900 miles, D. sherffii seems quite closely related to D. australis and its varieties of Oaxaca, Chiapas, and Guatemala as discussed above. From the pattern of distribution of this species (Map 6, closed circles) one is given the impression that D. sherffii is distributed only in two geographic areas isolated from each other by about 250 miles. It is believed this distribu- tion pattern reflects more the distribution of places where collections have been made than the real distribution of the species. This belief is based on first-hand knowledge of the type of habitat of the species and on a survey of 1969] 125138 | MERRARIUM STATE UNIVERSITY OF KWA POGS0ONIED Fig. Dahlia — Sorensen 389 MEXICO: STATE OP SURANGO Thomas Ë. Melchert, Peut D. Sorensen. Daniel J. Crawford — Nw. Taxanemic Studio in the Coropsidinse. oe ia d sao p " Plate 1423 9. Dahlia sherffii Sorensen. Photograph of type (IA), X 1/3. 390 Rhodora [Vol. 71 topographical maps of the area. It is seen from these maps that the region between the distribution centers of this species is continuously mountainous and would be expected to provide suitable habitats the entire distance. Moreover, this range of mountains continues for a considerable dis- tance south and north of these distribution centers. It is reasonable, therefore, to expect the present range extremi- ties of D. sherffii to be altered as the area becomes more thoroughly investigated. A certain amount of morphological diversity is displayed in the collections of D. sherffii available to me. This varia- tion prmarily involves the amount of segmentation of the leaves (from 1-pinnate to bipinnate-pinnatifid), and it correlates partly with the north-south distribution of the species. For example, the leaves on several specimens col- lected in Chihuahua display a greater amount of segmenta- tion than those collected in Durango. In some cases, however, almost an equivalent amount may be observed between populations within the respective geographic areas. Despite such variation the species is held together by a number of technical characters of its leaves. These are listed below along with the contrasting characteristics which distinguish the closely related D. australis, men- tioned above, from D. sherffii: (1) the upper surface of leaves pubescent (not glabrous), the hairs broadly but evenly spaced between the main veins throughout; (2) the lower surface slightly lighter green (not markedly lighter or silvery green) ; (3) the texture of leaflets smooth (not rugose); and (4) margins of leaflets flat (not revolute). 24. Dahlia scapigera (A. Dietr. Knowles & Westc. FI. Cab. 3: 113, pl. 118. 1839. TYPE. MEXICO: MEXICO or MICHOACAN: between San Joaquin and Angangueo (presumed to be the Angangueo, Michoa- can just W. of the México state boundary), Sep. 1830, Schiede s.n. (Holotype: B[?]; Isotype: US!). Georgina scapigera A. Dietr., in Otto & Dietr. Allg. 1969] Dahlia — Sorensen 391 Garten. 1: 197. 1833. Type: that of Dahlia scapi- gera. Dahlia scapigera var. typica f. typica Sherff, Am. Jour. Bot. 34: 139. 1947. Type: that of Dahlia scapigera. Herb, 2.8-4.5 (-6) dm. tall, the principal leaves crowded, rosette-like, near the base of the stem. Stem 2-3 mm. diam., glabrous, sometimes with a few hairs at the nodes; internodes 0-4 mm. (in shade forms the internodes may elongate to 2 cm.). Leaves pinnate to pinnate- pinnatifid, rarely almost bipinnate, 6-20 cm. long; leaflets 3-5(-7), sessile, rarely very short stalked, broadly ovate or ovate-lanceolate, 13-35 mm. long, 4-19 mm. wide; surfaces bicolored, lighter green beneath, the upper rugose, glabrous, the lower salient veiny, glabrous or sometimes a few scattered hairs along the principal veins; margins irregularly ciliolate, the cilia short and stiff, serrate with (1-)3-4(-5) teeth per side; stipels often present and then at the second node from the base; rachis glabrous except for a few soft hairs at the junction with the leaflets. Heads few, seldom more than one per plant, campanulate, subcernuous; outer involucral bracts erect or ascending (not strongly reflexed) at anthesis, 7-11 mm. long, 1.8-3 mm. wide, linear-oblanceolate or spatulate, glabrous, dorsally 3-5- lined, inner bracts 12-15 mm. long, 4-6 mm. wide; chaff yellowish, drying to tan; rays whitish-lavender to light-purple, sometimes with a yellow spot at the base, 1.8-3.5 cm. long, 0.8-1.5 cm. wide, ovate, acute or denticulate; disc florets 48-62, yellow, often tinged with purple at tips, 7-8 mm. long, corolla lobes spreading at anthesis, style branches 3-4.2 mm. long, about 0.8 mm. wide, linear lanceolate- attenuate; achenes (from greenhouse plants) 5-7 mm. long, 1.2-2 mm. wide, spatulate or at least linear-oblanceolate, blackish, minutely pubescent, many-sulcate pappus obsolete or consisting of 2 minute rudiments. Chromosome number, n = 16. Cool, open or lightly-wooded, rocky volcanic slopes, in deep loamy soil, 8600-12,600 ft., from Distrito Federal, just southwest of México City, west to Pico de Tancítaro, Michoacán, and on several high volcanic peaks in between. Flowering July-Sep. Map 7, closed circles. MEXICO. MEXICO: slopes of Volcán de Toluca, Balls 5047 (uc, US), Barkley et al. 79 (F, TEX, US), Melchert & Sorensen 6119 A-T (IA), 6217 A-B (1A), Roe et al. 303 (A, wis), Ugent et al. 1211 (WIS); Distr. of Temascaltepec, Las Cruces, Hinton 1039 (GH, MO, US); 9 mi. NE. of San Francisco Cheje along route 130, King 3570 (DS, MICH, NY, RSA, TEX, UC, US) ; woods near Aculco, Matuda 28883 (MEXU, NY); Cerro de Jilotepec, Matuda 29117 (NY). DISTRITO 392 Rhodora [Vol. 71 25N + %°%w © DAHLIA BARKERAE @ DAHLIA SCAPIGERA Map 7. Distribution of Dahlia barkerae Knowles & Westc., open circles; and Dahlia scapigera (A. Dietr.) Knowles & Westc., closed circles. FEDERAL: Peña de las Charras, Russell & Souviron 142 (CAS, US). MICHOACAN: summit and slopes of Cerro Tancitaro, Leavenworth 273 (F), 688 (F), 1103 (F, GH, MO); Cerro Pelón near Zitacuaro, Hinton 13239 (GH, MICH, NY, UC, US); mountains near Pátzcuaro, Pringle 4162 (F, GH, ISC, MEXU, MICH, MO, NY, UC, US); Cerro de San Miguel (state not given), Reiche s.n. (MEXU). In a series of articles Sherff (1946, 1947, 1955) gradually enlarged his concept of Dahlia scapigera (see Table 2) to include a wide assortment of taxa. To accommodate these he established a number of taxonomic varieties and forms which were brought together in their final version in his revision of Dahlia (1955). I have had the opportunity to observe all but two of Sherff’s intraspecific taxa of D. scapigera in the field. From these observations and after examining numerous herbarium specimens matched with Sherff’s published exsiccatae it was apparent that five rather distinct, and in some cases, unrelated species were included within D. scapigera. Briefly, these five species are characterized as 1969] Dahlia — Sorensen 393 SENSU SHERFF (1955) PRESENT TREATMENT D. SCAPIGERA (N = 16) F, SCAPIGERA D. SHERFFII (N = 32) F, MERCKII VAR, SCAPIGERA— D. MERCKII (N= 18) D. AUSTRALIS VAR, SERRATIOR F. xn SEI D. AUSTRALIS VAR. CHIAPENSIS F, AUSTRALIS VAR, AUSTRALIS — m D. AUSTRALIS VAR, AUSTRALIS F. PURPUREA (N = ]6 & 22) VAR, LIEBMANNII D. AUSTRALIS VAR, LIEBMANNII VAR, ARSENEI D. BARKERAE (N = 32) Table 2. Disposition of Dahlia scapigera sensu Sherff (1955) follows: (1) D. scapigera sensu stricto, plants of compact growth habit found only at very high elevations on volcanic peaks in south-central México and hardly over 4 dm. tall with mostly 1-pinnate, glabrous leaves (chromosome num- ber, n = 16); (2) D. barkerae, plants of compact growth habit, resembling D. scapigera but with larger and very hairy leaves and leaflets, stems remarkably hairy, ridges and cliffs, vicinity of Morelia, Michoacán (chromosome number, n = 32); (3) D. sherffii sp. nov., plants 6-13 dm. tall, leafy stems, 1-3-pinnate, pubescent leaves, on rocky ledges and along roadsides in northwestern México (chromosome number, n — 32); (4) D. merckii, plants 4.5-18.5 dm. tall, leafy portion 1/3 to 1/2 the height, leaves glabrous, mostly bipinnate with pinnules alternate on the rachilla, petioles hollow and fleshy, ledges and open lava fields in south-central México (chromosome number, » — 18) ; and (5) D. australis incl. 3 vars., leaves strongly bi- colored, the lower surface much lighter or silvery green, pubescent, upper surface rugose, glabrous, centra] Oaxaca eastward into western Guatemala (chromosome number, n = 16 & 32). These segregates are discussed in turn fol- 394 Rhodora [Vol. 71 lowing their individual taxonomic treatment. Table 2 con- trasts and summarizes the present treatment and that of Sherff (1955). Dahlia scapigera sensu stricto (chromosome number, n = 16) is a rather specialized species characterized by its extremely dwarfed, almost rosette-like, growth habit and by its relatively few campanulate flowering heads (usually only 1 or 2 per plant). It seems most closely related to, and possibly was involved in the derivation of, D. barkerae (chromosome number, n — 32). The ranges of these two species are almost sympatric and morpho- logically the species have in common a rugose leaf texture, mostly once-pinnate leaves with 3-5 sessile leaflets, a com- pact growth habit and erect or ascending outer involucral bracts. In addition to its chromosome number, D. scapigera is easily distinguished from D. barkerae by its fewer heads and by its more nearly glabrous (not conspicuously pubescent) herbage. When seen in the fresh condition, the leaves of D. barkerae are a much darker glossy green above. 25. Dahlia barkerae Knowles & Westc. Fl. Cab. 2: 28. 1858; & Fl. Cab. 3: 147, pl. 127. 1839. Type: Cultivated in Springfield, near Birmingham, England, 1837, from material collected in the vicinity of Valladolid (now called Morelia), Michoacán, México, 1836-1837, Barker s.n. (Holotype: BM [?]). Dahlia scapigera var. arsenei Sherff, Am. Jour. Bot. 33: 508. 1946. Type: Arséne 5520 Cerro Azul, vicinity of Morelia, June 1910 (Holotype: GH!; Isotypes: F!, MO!, NY!, US!). Herb, leaves crowded near base, 3-4.5 dm. tall, or with elongate internodes, then 6-7 dm. tall. Stem 3-4 mm. diam., densely pubescent, the hairs 2-4 mm. long, conspicuously multicellular, glistening white or silvery, flexuous; internodes of compact plants 0.0-1(-6.5) cm., of elongate plants 6-14 em. (the latter growing in shaded habitats, the former in more open situations). Leaves once-pinnate, 16-29 cm. long; leafllets (3-)5-7, 1-3.5 em. long, 0.7-2.8 em. wide, the terminal leaflet much larger than its companion leaflets, broadly ovate to suborbicular, apex acute or subacute, rounded, truncate or even cor- 1969] Dahlia — Sorensen 895 date at base, sessile or stalked, the petiolules less than 5 mm. long; surfaces bi-colored, the upper dark, shiny green, noticeably rugose (esp. when fresh), glabrous or minutely pubescent along the deeply impressed veins, the lower lighter green or gray-green, densely pubescent esp. along the conspicuous veins; margins ciliolate and coarsely serrate to dentate with 1-5 teeth per side; stipels sometimes present, attached at either the first or second rachis nodes; petiole 2.5-6 em. long, slightly winged esp. in proximal portion, pubescent; rachis pubescent, the hairs becoming floculent at the junctions with the leaflets. Heads several on a plant, in 2’s and 3’s, obliquely erect or slightly nodding, on peduncles 5.5-13 cm. long; outer involucral bracts erect at anthesis, dorsally glabrous, 3-5-lined, 6-6.5 mm. long, 2.5-3 mm. wide, elliptical or obovate, apex subacute or obtuse; inner bracts about 14 mm. long, 4.5-7 mm. wide, purple-lined; rays light purple, 26-32 mm. long, 14-17 mm. wide, ovate, acute or denticulate; chaff membranous, resembling the inner bracts; disc florets 48-60, yellow, the corolla about 8 mm. long, (excluding ovary), its lobes spreading or reflexed at anthesis; style branches 2.7-3 mm. long; achenes not seen. Chromosome number, » — 32. Rocky slopes and ledges, about 7000 ft., within a 20 mile radius of Morelia, Michoacán. Flowering June-Sept. Map 7, open circles. MEXICO. MICHOACAN: Cerro Azul, vicinity of Morelia, Arsène 18 (F), 5520 (GH); Campanario, Arsène 6787 (F [photo]) ; 17 mi. E. of Morelia along route 15, cañon of the Río del Salto, about 8500 ft., Melchert et al. 6414 A-C (IA). Dahlia barkerae was described in 1838 from material cultivated in England following its introduction from “the forests of Valladolid" (the Valladolid now called Morelia in Michoacán) in México. In 1946, Sherff applied the name D. barkerae to a group of specimens (Pringle 3164) which had been named D. pubescens by S. Watson (1891), but now called D. brevis. In the same paper (1946) Sherff de- scribed the new variety D. scapigera var. arsenei, The description was drawn from plants collected on Cerro Azul near Morelia. In September 1966 the type locality of D. scapigera var. arsenei was revisited and plants were found (Melchert et al. 6414 A-C) which seemed to be Sherff's variety. Subsequently, however, the original de- scription and accompanying plate of D. barkerae were seen and from these it became apparent that Sherff had mis- applied the name D. barkerae to the species which had 396 Rhodora [Vol. 71 been described as D. pubescens by Watson (here referred to as D. brevis). Dahlia barkerae seems closely related to the more widely distributed D. scapigera. With the latter it shares a com- pact habit, mostly once-pinnate leaves with leaflets that have a rugose texture on the upper surface, erect (not greatly spreading or reflexed) outer involucral bracts and slightly pendulous heads. D. barkerae may be distinguished from D. scapigera by its more branched habit with heads developing mostly in 2’s and 3’s, by its generally larger leaflets which are dark shiny green above and light green beneath, by its densely pubescent stem and lower leaf sur- faces, and by its chromosome number of n = 32. 26. Dahlia tenuis Robins. & Greenm. Proc. Am. Acad. 32: 48. 1896. TYPE: MEXICO: OAXACA: 18 mi. SW. of Oaxaca, 7500-9500 ft., 10 Sep. 1894, Nelson 1364 (Holotype: US!; Isotypes: F!, GH!). Herb 2.5-6 dm. tall. Stem slender, 1.5-4 mm. diam., obscurely many-striate, conspicuously pubescent, the hairs mostly retrorsely curved, more dense at the nodes; internodes 2.5-9.5 cm, long. Leaves pinnate-pinnatifid or pinnate-pinnatisect, 6-11 cm. long; leaflets or primary divisions 5-7, 2-3.5 em. long, lanceolate, irregularly lobed, pinnatisect or merely toothed with 1-3 teeth per side; surfaces bi- colored, dorsally lighter green and conspicuously veiny, the upper surface pubescent along the main veins, the lower pubescent, esp. along the principal veins; margins obscurely ciliolate, the cilia blunt, irregularly spaced; stipels sometimes present at the basal rachis node; petiole 0.8-4.5 cm. long, longitudinally grooved above, rounded beneath, pubescent, sometimes narrowly winged with the decurrent bases of the leaflets; rachis pubescent, densely so at the junctions with the pinnae, narrowly winged. Heads 1-4 per plant or on each main branch, erect or obliquely erect, slightly campanulate at anthe- sis; outer involucral bracts reflexed at anthesis, 5-8 mm. long, usually less than 1.0 mm. wide, oblanceolate or spatulate, acute or sometimes obtuse, glabrous, dorsally purple striate with 4-7 parallel lines; inner bracts 12-14 mm. long, 3-8 mm. wide; chaff (in fruit) 11-13 mm. long, 4.5 mm. wide; rays yellow, 1-2.5 cm. long, 6-9 mm. wide, ovate, acute or denticulate; disc florets 17-35, yellow, 3.5-4.5 mm. long, corolla lobes reflexed or revolute; achenes 6-9 mm. long, 1.5-1.8 mm. wide, linear-spatulate, dark gray or almost black, obscurely sulcate, pappus absent or consisting of 2 minute aristae less than 1 mm. long. Chromosome number n — 16. 1969] Dahlia — Sorensen 897 Steep rocky wooded slopes in zone of pine, 5000-8500 ft., from about 40 miles northwest of, to 18 miles southwest of Oaxaca, Oaxaca. Flowering Aug.-Sep. MEXICO. oAXACA: from about 30 to 39 mi. N. of Oaxaca along route 190, litis et al. 1290 (MICH, TEX, WIS), Jackson 7192 (IA), Melchert & Sorensen 6167 A-O (IA), 6179 A-M (IA), 6181 (14); Clavallinas Mts., Pringle 5807 (GH, MEXU, UC); mts. near Telixtla- huaca, L. C. Smith 481 (GH). The relationship of Dahlia tenuis to any other known Dahlia species is, at present, obscure. Its yellow rays sug- gest an affinity with D. coccinea, but its dwarfed habit, small heads and low number of disc florets are quite dif- ferent from those of any other species. 27. Dahlia coccinea Cav. Icones et Descr. Pl. 3: 33, t. 266. 1796. TYPE: exact place and date of collection not known; originally described by Cavanilles from plants cultivated at the Royal Botanic Gardens, Madrid, which had been grown from seed collected in México and sent to Spain by Vicente Cervantes, Director of the Mexican Botanic Garden, ca. 1789. (Holotype: MA, photo. F!). Georgina coccinea (Cav.) Willd. Linn. Sp. Pl. ed. 4. 3: 2124-2125. 1803. Type: based on Dahlia coccinea Cav. Dahlia crocata Sessé, Elenchus Hort. Reg. Matriti an. 1805, nom. nud., ef. Dahlia crocata Sessé ex Lagasca below. Dahlia bidentifolia Salisb. in Wm. Hooker, Parad. Lond. t. 19. 1805. Type: based on Dahlia coccinea Cav. Georgina coccinea a coccinea (Cav.) Willd. Hort. Berol. 2: pl. 96. 1809. Lectotype: pl. 96, Willd. loc. cit. 1809. Georgina coccinea B crocea Willd. Ibid. Lectotype: pl. 96. Willd. loc. cit. 1809. Georgina coccinea Y flava Willd. Ibid. Lectotype: pl. 96. Willd. loc. cit. 1809. Georgina frustranea a coccinea (Willd.) DC. Ann. Mus. Hist. Nat. Paris. Type: based on Georgina coccinea a coccinea (Cav.) Willd. (G-DC, microfiche libraries A-GH). Georgina frustranea B crocea (Willd.) DC. Ibid. Type: based on Georgina coccinea B crocea Willd. (G-DC, microfiche libraries A-GH). Georgina frustranea Y flava (Willd.) DC. Ibid. Type: based on Georgina frustranea Y flava Willd. (G-DC, microfiche libraries A-GH). 398 Rhodora [Vol. 71 Dahlia crocea (Willd.) Poir. Lam. Encycl. Suppl. 2: 445. 1812. Type: based on Georgina coccinea B crocea Willd. Dahlia frustranea (DC.) Ait. f. Hort. Kew. ed. 2, 5: 88. 1813. Type: based on Georgina frustranea [vars.] DC. Dahlia crocata Sessé ex Lagasca, Elenchus Pl. Hort. Reg. Matriti an. 1815 et Gen. Sp. Pl. Nov. aut Minus Cognitarum, 30. 1816. Lectotype: description p. 30, Lagasca loc. cit. 1816. Coreopsis georgina [var.] pruinosa Cass. Dict. Sci. Nat. 18: 443. 1820. Type: based on Georgina coccinea [vars.] Willd. and Georgina frustranea [vars.] DC. Georgina crocata (Sessé ex Lagasca) Sweet, Brit. Fl. Gard. 3: t. 282. 1829. Type: based on Dahlia crocata Sessé ex Lagasca. Georgina cervantesii Lagasca ex Sweet, op. cit. II. 1: t. 22. 1829. Lectotype: t. 22, Sweet loc. cit. Dahlia cervantesii (Lagasca ex Sweet) DC. Prodr. 5: 494. 1836. Type: based on Georgina cervantesii Lagasca ex Sweet (G-DC, microfiche libraries A-GH). Coreopsis coronata Sessé & Mocino, Pl. Nov. Hisp. 147. 1890, non Coreopsis coronata L. (1793). Lectotype: description p. 147, Sessé & Mocino loc. cit. 1890. Dahlia pinnata var. cervantesii (Lagasca ex Sweet) Voss in Vil- morin's Blumengart. 1: 489. 1894, Type: based on Georgina cer- vantesii Lagasca ex Sweet. Dahlia pinnata var. coccinea (Cav.) Voss Ibid. Type: based on Dahlia coccinea Cav. Dahlia pinnata var. gracilis (Ortgies) Voss Ibid. Type: based on Dahlia gracilis Ortgies. Coreopsis crassifolia Sessé & Mocifo, Fl. Mex. ed. 2: 194. 1894. Type: Sessé, Mocino, Castillo & Maldonado 2926 (Holotype: MA; Isotype: F!). Dahlia chisholmii Rose, Proc. U.S. Nat. Mus. 29: 439. 1905. Type: Chisholm s.n. near Arcelia, Guerrero, México (Holotype: US!; Isotypes: US!). Dahlia popenovii Safford, Jour. Washington Acad. 9: 369, fig. 3. 1919. Type: Popenoe 682, Dept. Zacatepequez, Guatemala (Holotype: US!). Dahlia coronata Hort. ex Sprague, Bull. Misc. Info. Kew (Kew Bull.) 1929: 53. 1929. Type: Sprague cited no specimens in his treat- ment of this species, however, Sherff (1947) states there are at K four sheets of the Sprague study material, from plants cultivated at the John Innes Institution, Merton, England (Lectotype: K). Dahlia gentryi Sherff, Am. Jour. Bot. 29: 332. 1942. Type: Gentry 6275, Sinaloa, México (Holotype: F!; Isotypes: ps!, GH!, MICH!, MO!, NY!). 1969] Dahlia — Sorensen 399 Dahlia coccinea var. steyermarkii Sherff, Ibid. 31: 280. 1944; op. cit. 33: 508. 1946. Type: Steyermark 50341, Dept. Huehuetenango, Guatemala (Holotype: F!). Dahlia coccinea var. palmeri Sherff, Ibid. 33: 508. 1946. Type: Edw. Palmer 494, vicinity of Durango, Durango, México (Holo- type: US [2 sheets]!; Isotypes: F!, GH!, Mo!, NY!). Dahlia coccinea var. coccinea (Willd.) Sherff, Ibid. 34: 149. 1947. Type: based on Dahlia coccinea Cav. and Georgina coccinea [vars.] Willd. Dahlia coccinea var. gentryi Sherff, Ibid. 34: 152. 1947. Type: based on Dahlia gentryi Sherff. Herb 4.5-30 dm. tall. Stems usually unbranched in the flowering portion, 2-25 mm. diam., glabrous to densely pubescent esp. at the nodes, sometimes conspicuously glaucous, variously colored green to purple when fresh; nodes usually septate or incompletely so; inter- nodes hollow, rarely solid, 2.5-30 cm. long. Leaves opposite or if whorled then with 3 per node, extremely variable from simple and unlobed to tripinnate, 12-35 cm. long including petiole; primary leaf- lets 3-11, sessile or petiolulate; basal pinnae 4.5-16 cm. long, its basal pinnule (of 2- or 3-pinnate leaves) 1.8-7 em. long, terminal segment of leaves, pinnae and pinnule commonly slightly larger than its com- panion leaflets; ultimate segments broadly ovate to ovate-elliptical or narrowly ovate-lanceolate, rarely slightly obovate, apices usually acute or slightly acuminate bases usually tapered, sometimes rounded, truncate, and even slightly cordate; surfaces usually bicolored, lighter green or gay-green beneath, the upper glabrous or pubescent, some- times coarsely scabrous, the lower rarely almost glabrous, usually conspicuously hairy along the veins; margins ciliolate, the cilia usually short and very stiff, infrequently longer and flexuous, vari- ously crenate-dentate to coarsely or finely serrate with (1-)2-14 teeth per side; petiole grooved above, crescent-shaped in cross-section, sometimes very narrowly winged, glabrous or pubescent, 1-11 cm. long; primary rachis glabrous or pubescent, mostly with a conspicuous cluster of hairs at the junctions with the primary pinnae, often grooved above, sometimes winged, especially in the distal portions; stipels occurring irregularly, when present then usually at the second rachis node, simple to compound, sessile or stalked. Heads usually in 2's and 3's, erect or obliquely erect, on peduncles 2-30 cm. long; outer involucral bracts spreading or reflexed at anthesis, variously ovate to obovate or spatulate, usually obtuse but sometimes long acuminate-attenuate, 6-15 mm. long, 3-7 mm. wide, dorsally several to many-lined, occasionally reticulate veiny, glabrous, ventrally gla- brous or puberulent, often cross-wrinkled; inner bracts brownish or scarlet throughout or just the tips scarlet, 11-17 mm. long, 3-8 mm. wide; rays 1.6-4 em. long, 10-18 mm. wide, ovate-elliptical or merely 400 Rhodora [Vol. 71 ovate, acute or denticulate, rarely lacerate at apex, lemon yellow to orange, orange scarlet, or deep blackish scarlet, sometimes variegated yellow and orange in a variety of combinations, dorsally puberulent along the veins, the hairs often gland-tipped, esp. those on the tube of the ligulate florets; chaff at anthesis 10-14 mm. long, 2-7 mm. wide, obtuse, pale greenish-yellow, scarlet or scarlet merely at the tips, dise florets 71-157, yellow or sometimes scarlet tipped, 8-10 mm. long, corolla lobes spreading or erect; style branches linear- lanceolate, flexuous, 4-4.5 mm. long, less than 0.8 mm, wide, achenes linear-oblanceolate to obovate or spatulate, 8-13 mm, long, 1.8-5.5 mim. wide, gray to black, sometimes black-speckled, usually minutely tuberculate, sometimes obscurely many-striate, of varied dimensions and texture sometimes within the same head; pappus obsolete or con- sisting of 2 minute rudiments, rarely greatly elongated into fine filiform threads, these very delicate and flexuous, usually deciduous. Chromosome number, n = 16 & 832. Cliffs, ledges, steep rocky slopes, fields and roadsides, from about 1500 to 11,000 ft. Sinaloa, Chihuahua, Coa- huila, Nuevo Leon, and Tamaulipas in Northern México, south in moutains and plateau regions to southern and southeastern México and Guatemala; reported (H. H. Iltis, personal communication) naturalized in mountainous re- gions of Peru, should be sought also in other Andean countries of South America. Flowering June-Nov. Map 8. MEXICO. TAMAULIPAS: E. side of Cerro Linadero, Meyer & Rogers 2747 (F, GH, MICH, MO, US), 2900 (BH, F, MO, US); between Marcella and Hermosa, Stanford et al. 2624 (NY, TEX, UC, US). NUEVO LEON: mts. in the vicinity of Monterrey and Cerro de la Cebolla, Muller 2877 (GH, UC), Muller & Muller 238 (F, GH, MEXU, TEX), R. F. Smith M612 (TEX). COAHUILA: Buenos Aires, Kenoyer & Crum 2833 (GH, MICH); La Casita, Kenoyer & Crum 4130 (GH); ledges and banks, Carneros Pass, Pringle 3166 (CAS, F, GH, MEXU, MO, MSC, NY, RSA, UC, US). CHIHUAHUA: Sierra Canelo, Rio Mayo, Gentry 2493 (ps, F, GH, MEXU, MO, UC, US); Cascada de Basascachic, Hewitt 149 (GH); Maguarichic (Mojarachic) SW. of San Juanito, Knobloch 1270 (MICH, MSC), 5994 (MSC, US). SAN LUIS POTOSI: 6 mi. W. of Ahuacatlán, Graber 177 (Us), 198 (us); Charcas, Lundell 5525 (MICH, US); 25 to 30 mi. E. of San Luis Postosí along route 86, McGregor et al. 574 (MSC), 646 (Msc), Powell 1144 (TEX). ZACATECAS: Along road to Huejuquilla el Alto, Jalisco, MeVaugh 17687 (MiCH); 13 mi. W. of Sombrerete, Soderstrom 747 (MICH, US); 18 km. W. of Concepción del Oro, Stanford et al. 601 (DS, GH, MO, NY, UC). DURANGO: 5 to 20 mi. W. of Durango City along route 40, Melchert et al. 6280 A-B & pop (IA), 6282 (IA), Waterfall 12551 (GH, MICH, US); 7-8 m. E. of 1969] Dahlia — Sorensen 401 s & a | f : A RV J N h: BE N . t mur - ç E now | . ° I ll 2 A - ER * E t Pii l se of » A L (e 2 e A . y x > P D DX je ae N 5 le/ Q9 e N P1 à ° x I i f rw S 5 e s ves s JE o ee (ON F E 3 xw / + ° napa mA q p 2 š : x o~— de. `A ° zi IN S we ¢ X or” m a & ( ° y i " i 1. e /*. E e Map 8. Distribution of Dahlia coccinea Cav. El Palmito along route 40, Melchert et al. 6316 (1A), 6318 A-B (14) ; Cafion of the Río Jaral about 15 mi. NW. of Coyotes Station, Cron- quist 9576 (MEXU, MICH, NY, US, TEX) ; fields along route 40, 29 mi. NE. of Durango City, King 3760 (MICH, NY, TEX, UC, US); at Coyotes Hacienda, 63 road miles WSW. of Durango City, Maysilles 7634 (F, MEXU, MICH, US); Ramos to Indé, Nelson 4714 (US); 16 mi. W. of El Ciudad along route 40, Melchert et al. 6313 (1A); Durango City and vicinity, Palmer 494 (F, GH, MO, NY, UC, us); mts. S. of La Purisma, Shreve 9182 (GH). SINALOA: Ocurahui, Sierra Suratato, Gentry 6302 (DS, GH, MICH, NY, US); 1-2 m. SW. of Sinaloa-Durango border along route 40, King 3722 (DS, MEXU, MICH, NY, TEX, UC, US); Yxtagua, Ortega 124 (MEXU) ; near Colomos, Rose 1779 (vs); Cerro del Perico, San Ignacio, Salazar 534 (US). HIDALGO: 15-20 mi. N. of Zimapán in the Barranca de San Vicente, along route 85, Fearing & Thompson 62 (TEX), Melchert & Sorensen 6206 (14), Moore 3376 (BH, GH), Straw & Gregory 1145 (GH, MEXU, MICH, RSA); 4-5 mi. N. of Jacala along route 85, Moore 1790 (BH), Weber & Charette 11911 (vs); 12 mi. W. of Pachuca along route 85, Melchert & Soren- sen 6203 A-F (IA); 7 mi. S. of Jacala, Melchert & Sorensen 6207 & 6207B (1A); 12 mi. S. of Jacala, Melchert & Sorensen 6208 (IA); between Zacualtipan and Olotlá, Moore 3292 (GH); along route 105 by Rio Panotlan, Moore 5330 (BH, GH, UC); near Ixmiquilpan, Rose et al. 9158 (GH); K-257, N. of Zimapán, Schnooberger 8002 (MICH). QUERETARO: 15 mi. SE. of San Juan del Rio, Waterfall & Wallis 402 Rhodora [Vol. 71 13966 (IA). GUANAJUATO: along road between Guanajuato and Dolores Hidalgo, Johnston & Ettlinger 2634A (TEX), Kenoyer 2111 (GH), Moore & Wood 4787 (BH). JALISCO: slopes and ledges in the Barranca del Rio Verde (locally known also as Rio Blanco), Detling 8512 (MICH), McVaugh 17342 (MICH), Melchert & Sorensen 6063 A-L (IA), 6066 (IA); slopes of Nevado de Colima (Volcán Fuego), MeVaugh 13495 (MICH), 13555 (MICH), Melchert et al. 6393 (IA), Urbina s.n. (MEXU); 5-8 mi. W. of Los Volcanes along road to Puerto Vallarta, Melchert et al. 6382 A-F (IA), 6383 (IA); near Guadalajara, Chisholm s.n. (MEXU); cliffs of Cerro de Talcozagua, Iltis et al. 707 (MEXU, MICH, NY, TEX, WIS); Cerro del Tigre, 3 mi. S. of Mazamitla, McVaugh 12967 (w'CcH), 12972 (MICH); SW. of Ojuelos along road to Aguascalientes, McVaugh 16762 (MICH); 15 mi. WSW. of Ojuelos, McVaugh 17015 (MicH); E. slopes of Cerro de los Gallos, MeVaugh 17087 (MICH); SE. slopes of Cerro Gordo above San Ignacio, McVaugh 17504 (MICH); barranca 5 mi. SW. of Tecalitlán, McVaugh 18129 (MICH); 2 mi. NW. of Tequila, McVaugh 18624 (MICH); Cerro del Halo along road to San Isidro, McVaugh. & Koelz 1250 (MiCH); 20 mi. E. of Ixtlán del Río, Nayarit, along route 15, Melchert et al. 63835 (1A); near summit of Cerro Etzatlán, Melchert et al. 6357 A-I (14); 8 mi. N. of Ciudad Guzman along road from Sayula, Melchert et al. 6391 (14); 15 mi. S. of Mazamitla along route 110, Melchert et al. 6399 (14); between Mesquitec and Monte Escobedo, Rose 2622 (GH, US); between Bolanos and Guada- lajara, Rose 3018 (vs), 3085 (vs); 15 mi. SE. of Autlán, Wilbur & Wilbur 1967 (MICH, US), 2016 (MICH, US). AGUASCALIENTES: Cerro del Laurel about 10 mi. SE. of Calvillo, McVaugh 18423 (MiCH). NAYARIT: 9-16 m. W. of Tepíc along route 15, McVaugh 18927 (MICH), 18938 (MICH), Melchert & Sorensen 6327 (IA); along road S. of Tepic to Compostela, McVaugh 16422 (MICH), Melchert et al. 6328 (IA); 3.5 mi. NW. of Ahuacatlán, Feddema 423 (MICH); slopes above lake NE. of Santa María del Oro, Feddema 699 (MicH), 730 (MICH); SW. slopes of Cerro Sangangüey, Feddema 914 (MICH); 20 mi. W. of Ahuacatlán along route 15, Melchert et al. 6330 A-C & pop (IA); along road from Tepíc to Jalcojotán, Mexia 605 (CAS, F, GH, MO, NY, UC, US); between Dolores and Santa Gertrudis, Rose 2036 (Us). VERACRUZ: San Miguel to La Joya, Balls B5491 (vc); Tlalcolulán, near Piletas, Dodds 104 (MICH); about one mile from Jalapa, Mac- Daniels 401 (F); 19 mi. W. of Orizaba along route 150 to Tehuacán, Melehert & Soresen 6198 A-E (14) ; Zacualpán, Purpus 1866 (F, MO, UC, US); Puente de San Miguel, Rangman 3425 (US); at W. end of Dam above Tuzpango, Sharp 441726 (F). PUEBLA: about 3 mi. N. of Atlixco along route 190, Melchert & Sorensen 6152 A-F (14); Chila-Zapotitlán, Miranda 2838 (MEXU); Amozoc, Pringle 9631 (GH, US); vicinity of San Luis Tultitlanapa, Purpus 3113 (F); near Tehuacán, Rose & Hay 10121 (GH, US); Esperanza, Seaton 357 (F, 1969] Dahlia — Sorensen 403 GH, NY, US); Puente del Emperador, Sharp 44525 (F); along Tehua- cán-Orizaba highway on slopes below Puerto del Aire, C. E. Smith et al. 3895 (F, MEXU, US). TLAXCALA: along route 140, 35 mi. E. of Texcoco, Montgomery & Root 9033A (MSC). MEXICO: along route 130 from Toluca to Temascaltepec, King 3579 (MICH, NY, TEX, UC, us), Melchert & Sorensen 6220 A-D (1A), 6221 (IA), 6224 A-B (IA); vicinity of Temascaltepec, Hinton 2028 (F, MEXU, MO, NY, us), 7965 (GH), 7999 (F, GH, MO, US), Matuda 27706 (NY); Valle de Bravo, Matuda 27195 (MEXU), Melchert et al. 6431 (IA), Moore & Cetto 5482 (BH, GH, UC); vicinity of Ixtapán de la Sal along route 55, Mick & Roe 350 (wis), Paray 2769 (MEXxU), Roe & Roe 1889 (A, WIS); vicinity of Villa Guerrero along route 55, Melchert et al. 6490 (IA), Roe & Roe 1840 (A, WIS); Santa Fé, Cuajimalpa, Matuda 21395 (MEXU, NY), Salasar s.n. (MEXU); Š. side of Guada- lupe dam, Corzo & Guzman 9 Aug. 1963 (MICH, TEX, WIS); near Huisquilnango, Matuda 21091 (MExU); Tepotzotlan, Matuda 21714 (NY); near Santo Tomas, Matuda 27124, (MEXU); Villa de Allende, Matuda 27646 (MEXU, NY); near Huehuetoca, Matuda 29079 (MEXU) ; Cerro Jilotepec, Matuda 29118 (NY); Real de Abajo near Sultepec, Matuda 29289 (NY); Cerro de los Capulines, Matuda 31343 (US) ; 9 mi. S. of Tenango, Melchert & Sorensen 6122 A-D (IA); 16 mi. S. of Tenancingo along route 55, Melchert & Sorensen 6129 A-C (1A) ; 7 mi. S. of Atlalcomulco along route 55, Melchert & Sorensen 6231 (IA); 1 km. SE. of Cahuacán, Montoya 51 (TEX); 8 km. SW. of Luvianos along road to Nanchititla, Rzedowski 20735 (MEXU). DISTRITO FEDERAL: Pedregal de San Angel, Barkely & Paxson 515 (F, MEXU, MICH, TEX), Martinez s.n. 14 Aug. 1959 (MEXU), Matuda 19484 (MEXU), Mexia 2730 (CAS, MICH, NY, UC); lava beds between México City & Cuernavaca along route 95, Degener & Degener 26267 (F), Matuda 21357 (NY), 21283 (NY, UC), Strother 498 (TEX) ; near Eslaba (Eslava), Pringle 11510 (CAS, F, GH, MICH, us), Rusby 378 (NY, US), 379 (Ny, US), 380 (NY, US), 381 (NY, US), 382 (NY); pedregal near Ajusco; Matuda 19168 (MEXU) ; Lomas de Chapultepec, Martinez sm. 1 Oct. 1954 (MEXU). MORELOS: near Cuernavaca, MacDaniels 281 (F), Matuda 21470 (NY); El Tezcal, 9 km. SW. of Tepoxtlan, Clausen 313 (MEXU, NY); Valle de Tepeite, Lyonnet 2416 (US), about 8 mi. SSW. of Tres Cumbres on road to Zempoala, Melchert & Sorensen 6107 (1A); 10 mi. N. of Cuernavaca, Melchert et al. 6487 (IA). MICHOACAN: numerous localities between Morelia and Ciudad Hidalgo along route 15, King 3613 (DS, MICH, NY, TEX, US), 3614 (DS, MEXU, MICH, NY, TEX, UC, US), 3619 (TEX), Melchert & Sorensen 6089 A-G (IA), 6091 (IA), 6095 A-K (IA), 6096 A-K (IA), 6097 (IA), Melchert et al. 6412 (IA), 6417 (1A), 6426 (IA), 6427 (1A), Ugent & Flores 1712 (wis); 1745 (wis); along route 15 E. of Zacapu, Iltis et al. 435 (TEX, WIS), Melchert et al. 6409 (IA), 6410 (IA); vicinity of Cerro Tancitaro and Apatzingán, Hinton 404 Rhodora [Vol. 71 15257 (F, GH, MO), Leavenworth 590 (F, NY), 640 (F, GH); district of Coalcomán, Hinton 12266 (GH, MICH, NY, UC, US), 15936 (NY), 15954 (NY); along route 37 S. of Uruapan, King & Soderstrom 4739 (MICH, NY, TEX, UC, US), Melchert & Sorensen 6081 (IA); 5 mi. N. of Patzcuaro, Barkley et al. 2702 (F, TEX); 5.5 mi. N. of San Pedro Jacuro on road to San Andres, Beaman 4260 A-C (GH, MSC, TEX) ; Cerro Santa Maria, 8-10 km. SW. of Jiquilpán, Feddema 200 (MICH), 205 (MICH); cañon 2 mi. N. of Zitácuaro, Hitchcock & Stanford 7206 (Ds, POM, UC); 12 mi. NW. of Zitácuaro, King 3604 (DS, MEXU, MICH, NY, TEX, UC, US); near Cojumatlan, Melchert & Sorensen 6073 (1A); 32 mi. E. of Zamora, Melchert & Sorensen 6075 A-D (1A); 15 mi. N. of Cheran along route 37, Melchert & Sorensen 6076A (1A); 3 mi. S. of Escalante along route 160, Melchert & Sorensen 6086 A-K (IA); 6 m. E. of Tuzpán, Melchert & Sorensen 6102 (1A); 5 mi. W. of Jacona, Melchert et al. 6400 A-B (IA); 6.4 km. NE. of Pátzcuaro on the road to Zurumütaro, Ugent & Flores 2115 (WIS). CHIAPAS: in the vicinity of San Cristóbal de las Casas, Breedlove 11342 (ps), 11905 (ps), Melchert et al. 6459 (1A); along trail from Tenejapa to Paraiso, Breedlove 6813 (Ds, F, MICH); 3 mi. W. of Huistan, Breedlove 12413 (ps); barrio of Tuk, paraje of Matsab, Breedlove 12510 (Ds, MEXU) ; along trail from Tenejapa to Colonia San Antonio, Breedlove 12798 (ps); Miramar, Matuda 785 (MEXU, MICH); Puerto Arriba, Miranda 7261 (MEXU) ; Cerro del Boqueron (Cerro Ovando), Purpus 6667 (F, MO, NY, UC). OAXACA: along route 175 N. of Oaxaca to Ixtlan de Juárez, King 3511 (MICH, TEX), Melchert & Sorensen 6170 (IA), 6172 (IA); vicinity of Cerro Zempoaltepetl, Hallberg 1061 (MicH); Monte Alban, Kenoyer 1522 (GH); about 24 mi. N. of Oaxaca along route 190, King 3520 (TEX); 10 mi. S. of Huajuapán, Melchert & Sorensen 6156 A-E (IA). GUERRERO: vicinity of Taxco, Abbott 147 (Gu), 396 (GH), Manning & Manning 531012 (GH); along road from Chil- pancingo E. to Tixtla, Melchert & Sorensen 6136 (1A), 6143 A-B (14), 6143 A-B (1A), 6146 (14); District of Minas, Hinton 9221 (GH, MICH, NY, UC, US), Hinton 14496 (GH, MO, NY, US); district of Montes de Oca, Hinton 11350 (GH, MICH, NY, UC, US); Cañon de Zopilote between Mexcala and Zumpango along route 95, Moore 5197 (BH); 3 km. beyond Acahuizotla along road to Acapulco, Moore & Wood 4676 (BH, UC); about 24 mi. S. of Chilpancingo along route 95, Melchert & Sorensen 6131 A-C (14); 6 mi. N. of Ocotito along route 95, Melchert & Sorensen 6133 A-E (14); Canon de la Mano near Iguala, Rose et al. 9330 (GH, US). GUATEMALA. BAJA VERA- PAZ: 14 mi. N. of Salama along route 5, King 3356 (TEX). CHIMA- TENANGO: near Tecpán, White 5264 (F, MICH). HUEHUETENANGO: about 11 mi. S. of Huehuetenango along route 9N, King 3390 (MICH, NY, TEX, UC, US); between Sololá and Santa Eulalia, Steyermark 1969] Dahlia — Sorensen 405 49971 (F); between Chemal and Calaveras, Steyermark 50341 (F, GH); Cerro Pixpix above San Ildefonso Ixtlahuacán, Steyermark 50607 (F, NY); 2/3 of way up Cerro Chiquihui above Carrizal, Steyermark 50789 (F); mtns. across river from San Juan Atitan, Steyermark 51976 (F). JALAPA: between Jalapa and Montana Miramundo, Steyermark 32864 (F). QUETZALTENANGO: fields along route 1 about 3 mi. E. of Quetzaltenango, King 3204 (MICH, TEX, UC). SACATEPEQUEZ: slopes of Volcán de Agua N. of Santa Maria de Jesús, Standley 59373 (F). SOLOLA: along route 1 between Sclolá and Panajachel, King 3231 (TEX), Melchert et al. 6438 (IA). TOTONICAPAN: in San Cristóbal Totonicapán, Breedlove 11465 (Ds). ZACAPA: ravine along Río Lima between Río Hondo and summit of Cerro Finca Alejandria, Steyermark 29613 (F). Dahlia coccinea is perhaps the most complex species of the genus. Its characteristics are nearly all superlative. It has the greatest geographic range (see Map 8), the greatest amount of variation in its foliar and floral features, perhaps the showiest “flowers,” and is the cn!y Dahlia which might be called a weed. The great range of variation in nearly all its characteristics, as portrayed in the foregoing de- scription, has not gone unnoticed by previous workers. Sherff (1955), attempting to make some taxonomic “sense” of the species, established the following four varieties: 1. var. steyermarkii, simple to irregularly 1-pinnate leaves, usually coarse-textured, stems subvillous or tomentulose. var. coccinea, leaves 1- or 2-pinnate, regularly di- vided, stems glabrous or nearly so. 3. var. palmeri, leaves 2- or 3-pinnate, stems robust, reddish-glaucous, ligules orange or scarlet. 4. var. gentryi, leaves 2- or 3-pinnate, stems delicate, straw-colored, ligules yellow. It is my belief that D. coccinea is a single, very wide- ranging, polymorphic species and that its “varieties” are often merely extreme variants which arise independently within geographically distinct (but not necessarily repro- ductively isolated) breeding populations. In the course of this study I have encountered these “varieties” in the field and have made certain observations about them and about the species as a whole. These are listed below with certain N 406 Rhodora [Vol. 71 facts obtained from subsequent laboratory studies of my collections. a) The “variety” to which an individual plant belongs, unless its features are extreme, is difficult, if not impossible, to determine. b) The geographic range of each variety is sympatric with one or more of the other varieties. c) In several populations two of these varieties (not always the same two) were growing side by side, some- times with intermediate forms between them (cf. Melchert et al. 6318 A & B, 6357 A-E, & 6409). d) All the varieties occupy more or less the same type of habitat. e) In most populations ligule color, leaf shape and size, chromosome number, height, and other features vary in- dependently. For example, a population of scarlet rayed plants may have leaves ranging from irregularly-pinnatifid (var. steyermarkii) to tri-pinnate (var. palmeri) (cf. Melchert et al. 6382 & 6383). f) Although diploid and tetraploid races are sympatric throughout most of the range of the species, the tetraploid races extend farther southeast into Guatemala, In several localities populations of the two races were found within a mile or two of one another but were normally not con- tiguous except as noted below. g)In one instance both diploids and tetraploids were found in the same population. These plants were morpho- logically similar and were found growing within a few feet of one another (cf. Melchert & Sorensen 6063 A-F). h) Many diploids and tetraploids throughout most of the range of the species are morphologically indistinguish- able. i) Several of the populations discovered in remote places and isolated from the main body of the species (cf. Mel- chert et al. 6280 & 6393) turned out to be diploids. All of the tetraploids were found in disturbed areas along road- sides and fields as were also some of the diploids. Dahlia coccinea, as with most Dahlia species, is repro- 1969] Dahlia — Sorensen 407 ductively self-incompatible. This has been true of five individuals, 3 of them diploids and 2 tetraploids, which I have grown in the greenhouse from tubers collected in México. In cases where this form of barrier to inbreeding prevails a high degree of heterozygosity is maintained in the population and all plants of each generation are, in a sense, “hybrids.” Each of these “hybrids” contains two, or the segments of two or more, genomes allowing for a tremendous amount of stored variability. Such reproductive self-incompatibility in combination with a high percentage of heterozygosity provides for a full exploration of the genotypic and phenotypic potential of any given breeding population in each generation. Obviously, not every geno- type can survive, and the habitat does not seem to be strongly selective in its choice of those which are allowed to survive (items c & d above). In small populations, therefore, there is an abrupt difference from plant to plant in those characteristics used to determine the varieties. Lacking intermediates their identification becomes a rela- tively simple matter. On the other hand, if this breeding unit or population is very large, the range of variability is more apt to be continuous. Obviously, in such cases it becomes biologically unsound and taxonomically artificial to designate a few of the specimens “varieties” gentryi, steyermarkii or palmeri and then name all the remaining variants var. coccinea. Therefore, my argument until now has been from the viewpoint that the extreme variants of D. coccinea should not be given formal taxonomic recogni- tion but should be regarded as segments of a broad morpho- logical spectrum. On the other hand, I do not wish to create the impression that every discrete population of D. coccinea is a mass of morphologically diverse individuals. I have observed cer- tain populations in which there is great uniformity. For example, five miles west of Durango, in the general vicinity of the type locality of D. coccinea var. palmeri, an isolated population was found at the head end of a box canyon. Of those collected (Melchert et al. 6280 A, B & pop.), each 408 Rhodora [Vol. 71 plant was morphologically similar with scarlet rays and 3-pinnate leaves, and they fit Sherff’s var. palmeri quite well. The single chromosome count obtained was n — 16. Again, about 12 miles northwest of Pachuca, Hidalgo, a population (Melchert & Sorensen 6203 A-F) of the diploid plants was found growing among thorny scrub and cactus plants in a region of semi-desert. All plants had glabrous, bipinnate leaves and unusually large, deep-orange or orange-scarlet ligules. This population was also quite iso- lated from other populations of D. coccinea. Both of the above examples by themselves possess all of the requirements for varietal status. They are geographic- ally limited, have mutually exclusive morphologic markers and are isolated, at least locally, from the main body of the species. If these populations were all that existed of D. coccinea they might even be regarded as distinct species. It is my opinion that such units with morphologic and geographic uniformity, as described in the above examples, do exist within D. coccinea, The problem has been that, until now, the overwhelming majority of the specimens in herbaria are from inhabited areas and from near railroads and along roadsides. What will be needed in order to search out these units is much more collecting from remote, undis- turbed and reproductively isolated populations. The many morphological variants of Dahlia coccinea are held together as a species primarily on the basis of their ligule color (item e, above). There is a very limited amount of biochemical evidence in support of this. Five individual plants of D. coccinea under greenhouse cultiva- tion from tubers collected in México by Melchert & Sorensen during 1965 and 1966 produced heads with ligule colors as follows: 6086, yellow-orange; 6203, deep-orange; 6220, bright scarlet with yellow at base; 6224, yellow-orange; 6318, deep-scarlet. Utilizing 2-dimensional paper chroma- tography, biochemical profiles of the flavonoid compounds in the ligules of these plants were obtained as follows: the pigments were extracted from freshly collected rays by placing them in a flask with acidified methanol (0.5% cone. 1969] Dahlia — Sorensen 409 HCl by volume). After 12-20 hours of agitation the extract was condensed and spotted on Whatman No. 3 filter paper. The chromatograms were developed via descending chroma- tographic techniques. The solvent for the first run con- sisted of tertiary butyl alcohol: glacial acetic acid: water (3:1:1 v/v) and for the second run in the second dimension, 15% by volume of glacial acetic acid and water. When the chromatograms were dried they were viewed in UV light both with and without ammonia fumes. An essentially identical basic profile involving 8 compounds (3 visible as yellow in daylight, 5 visible only in UV) were found in each sample. An additional compound, probably an anthocyanin (visible as magenta in daylight), was found in varying quantities on some and absent on others. An identical test was made of the ligules of three other species which produced light purple ligules in various shades. The profile of these was quite unlike that of 7, coccinea rays. On the basis of these exceedingly preliminary and quali- tative data it is concluded that the varying colors of intact rays of D. coccinea heads are due to differing concentra- tions of anthocyanin. Underlying the variable anthocyanin pigment, however, there is, uniformly present throughout the species, a basic group of 8 compounds. Several of these apparently contribute very little to the visible color of the rays but, interestingly, are strikingly different from the basic profile seen in ligules of the purple-rayed species. In summary, Dahlia coccinea is a wide-ranging, poly- morphic species. The species is held together by its ray colors which range from lemon yellow to deep blackish- scarlet and by the absence of clear morphologic discon- tinuities in the great range of variation displayed by its vegetative parts. For the present, it appears best not to recognize any formal subspecific taxa, but it is expected that eventually some could be established. This must await the accumulation of a much greater amount of information on the morphology, cytology, and geographic distribution of the species and of its various morpho-geograhic units. 410 Rhodora [Vol. 71 Doubtful, Excluded and Rejected Names Dahlia acutiflora DC. Prodr. 5: 494. 1836. This name is listed by de Candolle as a synonym for Dahlia cervantesii (Lagasca ex Sweet) DC. (i.e. Dahlia coccinea Cav.) how- ever, he remarks that the ligule colors range from purple to scarlet which suggests it was of hybrid origin. Dahlia astrantiaeflora (Sweet) G. Don in Sweet's Hortus Brit. ed. 3, 353. 1839. This should be regarded as a cultivar. Dahlia fulgens Hort. ex Sweet, Brit. Fl. Gard: 282. 1829. Sweet offers this name as a synonym for Georgina crocata (Sessé ex Lagasca) Sweet (ie. Dahlia coccinea Cav.) without reference to its earlier publication. As nearly as I can determine this is its first appearance in the literature. Dahlia gigantea Bull. in Loudon, Ladies’ Flower-Garden 2: 6. 1844. Nomen nudum. Dahlia juarezii Van der Berg, Gard, Chron. II. 12: 433, fig. 66. 1879. and Ibid. 594. 1879. This name has been widely used in reference to all Dahlia hybrids which produce revolute rays. As noted in the second reference of 1879 above, the original plant which served as the progenitor of these hybrids may have been referable to Dahlia coccinea Cav. Dahlia platylepis A. Brongn. ex Neumann, Rev. Hort. II. 4: 305. 1845. The description accompanying this name is confusing and cannot be related to any known taxon. The height of the plant is given as 3 m. suggesting that it might be referable to a species of section Pseudo- dendron but the petiole measurements given as 10-12 cm. would exclude those taxa. Dahlia pubescens A. Brongn. ex Neumann, Rev, Hort. II. 4: 305. 1845. Apparently Brongniart intended this name to refer to a segregate of Dahlia platylepis with the same description applying to both names. The added re- marks that Dahlia pubescens had more hairy leaf sur- faces is no help in relating it to a known taxon. Dahlia purpurea Cav. in Desfontaines, Tab. de lEcole 1969] Dahlia — Sorensen 411 Bot. Mus. Hist. Nat. Paris. 106. 1804. The epithet is that of Thouin (1804) but Desfontaines attributes it to Cavanilles without reference to its publication. Dahlia pusilla in Herb. Zucc. ex DC. Prodr. 5: 494. 1856. This name was apparently regarded by de Candolle as a synonym for a horticultural variant of Dahlia, variabilis (Willd.) Desf. It should be regarded as a cultivar. Dahlia repens Hartweg in litt. ex Benth. Trans. Hort. Soc. London II. 2: 395. 1840. Nomen nudum. Dahlia repens Hort. ex Verlot, Rev. Hort. 1864: 31. 1864. Nomen nudum. It is not known whether there is any connection at all between the Dahlia repens referred to by Verlot and the species given the same name by Hart- weg above. Dahlia rosea Cav. Icones et Descr. Pl. 3: 33, t. 265. 1796. This name has been a source of constant error and, in accordance with the provision of Art. 69, International Code of Botanical Nomenclature. 1966, should be re- jected. Many authors have considered it a synonym of Dahlia pinnata Cav. (or of some nomenclatural equiva- lent of Dahlia pinnata) while others have related it to the "tree-Dahlias". In the original description the phrase “pinnulis secondariis nunc alternis, nune oppo- sitis" suggests it may be an earlier name for what is treated here as Dahlia merckii Lehm. However, later authors such as Thouin and de Candolle who had access to plants grown from the seeds of the original study ma- terial of Cavanilles do not mention this trait. Dahlia royleana Knowles & Weste. Fl. Cab. 3: 186. 1840. Description inadequate, Dahlia sphondyliifolia Salisb. Trans. Roy. Hort. Soc. E591. 1820. This is a superfluous name apparently offered by Salisbury as a more descriptive epithet for Dahlia rosea Cav. noted above. Dahlia zimapani Roezl in litt. ex Ortgies, Gartenflora 1861: 406, t. 247. 1861. — Cosmos diversifolius Otto. Georgina, astrantiaeflora Sweet, Hortus Brit. ed. 2. p. 310. 1830. This should be regarded as a cultivar. 412 Rhodora [Vol. 71 Georgina bipinnata (Cav.) Sprengel, Syst. Veg. 3: 611. 1826. = Cosmos bipinnatus Cav. Georgina petroselinifolia Liebmann in litt. Bot. Zeit. 2: 687 & 734. 1844. Nomen nudum. PAUL D. SORENSEN THE ARNOLD ARBORETUM HARVARD UNIVERSITY JAMAICA PLAIN, MASSACHUSETTS 02130 LITERATURE CITED AITON, f., W. T. 1813. Hort. Kew. ed. 2. 5: 87, 88. BELLING, JOHN. 1924. Homologous and similar chromosomes in diploid and triploid Hyacinths. Jour. Gen. 10: 60. BENTHAM, GEORGE. 1838. Dahlia exclesa in Maund, The Botanist 2: pl. 88, DE CANDOLLE, A. P. 1810. Note sur les Georgina. Ann. Mus. Hist. Nat. Paris. 15: 310. 1836. Prodr. 5: 494. CAVANILLES, ANTONIO JOSE. 1791. Icones et Descr. Pl. 1: 57, t. 80. 1796. Op. cit. 3: 33, t. 265 & 266. COLMEIRO, MIGUEL. 1875. Bosquejo historico y estadistico del Jardin Botanico de Madrid. 105 pp., Madrid. DESFONTAINES, RENATO. 1829. Cat. Pl. Horti Reg. Paris. ed. 3. HEMSLEY, W. B. 1879. Dahlias. Gard. Chron. II. 12: 437, 524- 025, 557. illus. 1881. Biol. Centr. Am. 2: 196, 197. HowE, MARSHALL A. 1936. In New Taylor Gard. Dict. 196-198. ISHIKAWA, M. 1911. Cytologische Studien von Dahlien. Bot. Mag. Tokyo 25: 288. LAWRENCE, W. J. C. 1929. The genetics and cytology of Dahlia species. Jour. Gen. 21: 125-158, illus. PoPENOE, WILSON. 1920. The tree-Dahlia of Guatemala. Jour. Hered. 11: 264-268, illus. SAFFORD, W. E. 1919. Notes on the genus Dahlia with descriptions of two new species from Guatemala. Jour. Washington Acad. 9: 364-373, illus. SHERFF, EARL EDWARD. 1930. New or otherwise noteworthy Com- positae. IV. Bot. Gaz. 89: 364-365. 1946. Some new or otherwise noteworthy dicotyledonous plants. Am. Jour. Bot. 33: 508. 1947. New or otherwise noteworthy Com- positae. X. Am. Jour. Bot. 34: 138-158. 1951a. Miscellaneous notes on new or other- wise noteworthy dicotyledonous plants. Am. Jour. Bot. 38: 54-73. 1969] Dahlia — Sorensen 418 1951b. Dahlias — new facts about these garden favorites. Chicago Nat. Hist. Mus. Bull. 22(6): 5-7, illus. 1955. In North Am. Flora II. part 2: 45-59, 1959. Notes on the Coreopsidinae from Mexico and the United States. Brittonia 11: 189. 1962. The native dahlias of Mexico. Bull. Louisiana Soc. Hort. Res. 2: 62-76, illus. 1966. Notes on the occurrence of the rare Dahlia barkerae (Compositae). Sida 2: 267, 268. STANDLEY, PAUL C. 1938. Dahlia rosea in Flora of Costa Rica. Bot. Series: Field Mus. Nat. Hist. 18: 1448. THOUIN, ANDRE. 1804. Memoire sur la culture des Dahlia. Ann. Mus. Hist. Nat. Paris 3: 420-423, illus. THUNBERG, CARL P. 1792. Skr. Naturh.-Selsk. Kjóbenh. 2: 133, t. 44. TURNER, B. L., M. PowELL, and R. M. KING. 1962. Chromosome numbers in the Compositae. VI. Additional Mexican and Guate- malan species. Rhodora 64: 251-271. WATSON, SERENO. 1891. Proc. Am. Acad. 26: 142. WILLDENOW, C. L. 1803. Linn. Sp. Pl. ed. 4. 3: 2124, 2125. 1809a. Hortus Berol. 2: pl. 93-96. 1809b. Enum. Pl. Hort. Reg. Bot. Berol. 899, 900. Index to Exsiccatae Abbott 147, 396 (27). Agniel & Arséne 10286 (7). Arséne 18 (25); 49 (25); 1396 (10); 5520, 6787 (25); 10083 (7). Balls 5047 (24); 5157, B5306, 5349 (10); B5491 (27). Barkley & Paxson 515 (27). Barkley et al. 79 (24); 2702 (27). Bates s.n. (1). Beaman 2040, 2113 (10); 4260 A-C (27). Bell & Duke 16972 (2). Bourgeau s.n, Aug. 1866, 736, 746 (14); 802 (or 862) (10). Breedlove 6813 (27); 6814 (2); 7095 (22c); 7233, 7300 (2); 7307 (22c); 7424, 8761, 8878, (2); 11342, 11465, 11905, 12413, 12510, 12798 (27); 13762 (22c). Breedlove & Raven 13412 (22c). Calderon 2422 (2). Carlson 1613 (2). Chisholm s.». May 1905, s.n. Nov. 1904 (27). Chute M-90 (10). Clark 7277 (10). Clausen 313 (27). Conzatti 55 (22a); 1647 (11); 2259 (3); 4078 (22a). Con- zatti & Gonzalez 402, 543, 544 (22a); 552 (3). Corzo & Guzman s.n. 9 Aug. 1963 (27). Cronquist 9576 (27); 9655 (22a). Cuatre- casas 8185, 23407 (2). Degener & Degener 26267 (27). Detling 8512 (27). Dodds 104 (27). Echevarria 160 (2). Ehrenberg 368 (10). Fearing & Thompson 62 (27). Feddema 200, 205, 423, 699, 730, 914 (27) ; 2746 (3). Fisher 197 (14). Fosberg 20290 (2). 414 Rhodora [Vol. 71 Gentry 2493, 6275, 6302 (27); 17993 (23). Ghiesbreght 154, 558 (22c). Gold & Eheberle 21770 (10). Goodman & Melhus 3691 (2). Gordon 74 (23). Graber 177, 198 (27). Grant 9652 (2). Gregg 377 (10). Hallberg 896 (22a); 1061 (37). Halsted s.». (9a). Harshberger 118 (10). Haught 6114 (2). Hawkes, Hjerting, & Lester 1235 (23); 1395 (10); 1396 (16). Hewitt 149 (27). Heyde & Lux 4223, 4239 (2). Hinton 1039 (24); 1885 (17); 2028, 7965, 7999 (27); 8332 (17); 9221 (27); 11083, 11137 (3); 11350, 12266 (27); 12364 (3); 13239 (24); 14496 (27); 14499 (19); 15257, 15936, 15954 (27). Hitchcock & Stanford 7207 (27). Hunnewell 14869 (2). Iltis, Koeppen & Iltis 143 (10) ; 435, 707 (27); 1093 (10); 1290 (26). Jackson 7192 (26). Johnston, J. R., 986, 1975 (22d). Johnston & Ettlinger 2634A (27). Kellerman 7096 (2). Kenoyer 1522, 2111 (27). Kenoyer & Crum 28533, 4130 (27). Killip & Lehmann V 38506 (2). Killip & A. C. Smith 16530, 20558 (2). King, R. M. 2076 (22a); 3564 (10); 3570 (24); 3579, 3604, 3613, 3614, 3619, 3722 (27); 3743 (23); 3760, 3204, 3231, 3356, 3390, 3511, 3520 (27). King & Soderstrom 4739 (27). Knobloch 1188 (23); 1270 (27); 5292 (23); 5994 (27). Krueger & Gillespie 35 (22a). Kuntze 23660 (17); 23746, 238747 (10). Langenheim 3359 (2). Leavenworth 273 (24); 590, 640 (27); 1103 (24). Lehmann 4894 (2). LeSeur 24 (23). Liebmann 668 (—9782) (22b). Lundeli 5374 (10); 5525 (27). Lyonnet 838 (14); 2416 (27); 3372 (17). MacDaniels 401, 281 (27). MacDougall 365 (22c); 21984 (4). Mc- Gregor, Harms, Robinson, Rosario, & Segal 296 (10); 574, 646 (27). McVaugh 11745 (3); 12272 (10); 12926 (3); 12967, 12972, 13422, 13495, 13555 (27); 13874, 14110 (3); 16762, 17015, 17087, 17342, 17504, 17687, 18129, 18423, 18624, 18927, 18938 (27). Mc- Vaugh & Koelz 1250 (27). Manning & Manning 531012 (27); 531050A (10). Martinez 15048 (10); sn. 14 Aug. 1959, s.n. 1 Oct. 1954 (27). Matuda S-115 (10); 785 (27); 1363 (10); 2233, 4628, 15452 (2); 19168 (27); 19169 (10), 19411 (17); 19484 (27); 19589 (10); 21091 (27); 21092 (14); 21283, 21357, 21395 (27); 21453 (10); 21470 (27); 21476 (10); 21542 (14); 21685 (10); 21714 (27); 25575 (17); 26182 (14); 27124, 27195, 27646, 27706 (27); 28883 (24); 29070, 29079 (10); 29117 (24); 29118 (27) 29121, 29229 (14); 29289 (27); 31003 (10); 31170 (14); 31343 (27). Maxon & Hay 3295 (2). Maysilles 7512 (23); 7634 (27); 7928, 8374, 8401A (23). Melchert & Sorensen 6031, 6052 (23); 6063 A-L, 6066, 6073, 6075 A-D, 6076 A, 6081, 6086 A-K, 6089 A-G, 6091, 6095 A-K, 6096, 6097, 6102 A-C (27); 6104, 6106 (10) ; 6107 (27); 1969] Dahlia — Sorensen 415 6119 A-T (24); 6121 A, 6121 (10); 6122 A-D (27); 6125 A-D (14); 6129 A-C, 6131 A-C, 6133 A-E, 6136 (27); 6142 A-K (12); 6143 A-B, 6146 (27); 6149 A-G (10); 6152 A-F, 6154 A-B, 6156 A-E (27); 6167 A-O (26), 6170, 6172 (27); 6173 A-C, 6174 A-G (22a); 6179 A-M, 6181 (26); 6184, 6185 A-D (22a); 6187 (27); 6192 A-E (22a); 6193, 6194 A-E (3); 6198 A-E (27); 6200 A-H (10); 6203 A-F (27); 6205 A-C (9a); 6206, 6207, 6207 B, 6208 (27); 6209, 6210 A-C (20); 6213 A-D (9a); 6217 A-B (24); 6220 A-D, 6221, 6224 A-B, 6231 (27); 6233 A-E (14); 6235 A-E (16); 6238 A-E (10); 6239 A-B (14); 6240 A-B (7). Melchert, Sorensen, & Crawford 6272, 6274 (23); 6280 A-B, 6282 (27); 6288, 6304 (23); 6306, 6312 (8); 6313, 6316, 6318 A-B, 6327, 6328, 6330 A-C, 6335, 6357 A-I, 6382 A-F, 6383, 6386, 6391, 6393 (27); 6396 A-C (3); 6399, 6400 A-B, 6409, 6410, 6412 (27); 6414 A-C (25); 6417, 6426, 6427, 6431 (27); 6434 (2); 6438 (27); 6439, 6443, 6452 (2); 6458 (22c); 6459 (27); 6462, 6464 (2); 6487, 6490 (27); 6498 (21); 6504 (20). Meyer & Rogers (27); 2881 (6); 2894 (10); 2900 (27); 3009 (10). Mexia 605, 2730 (27); 2766 (10). Mick & Roe 350 (27). Miranda 2838 (27); 4454 (14); 7261 (27); 9688 (2). Miranda, Barkley & Webster 2258 (10). Montgomery & Root 9033 A (27). Montoya 51 (27). Moore H. E. 1790 (27); 3097 (5); 3292, 3376 (27); 5036 (5); 5068 (19); 5197, 5330 (27); 8247, 8250 (3). Moore & Cetto 5482 (27); 8189 (2); 8193 (22d). Moore & Wood 3943 (5); 4105 (10); 4407 (5); 4425 (9a); 4441 (18); 4469 (20); 4676, 4787 (27). Muller, C. H. 2877 (27); 2879 (10); 3466 (23). Muller & Muller 238 (27); 837 (10). Nelson 640, 1161 (22a); 1355 (3); 1364 (26); 2193 (3); 3173 (2); 4714 (27). Ortega 124 (27). Ortenburger et al. 16M732 (10). Ownbey & Ownbey 1981 (23). Palmer, Edw. 211 (10); 494 (27). Paray 2769 (27). Parry & Edw. Palmer 477 (10); 479 (14). Pennell 17831 (10). Pennell et al. 8377 (2). Popenoe 682 (27). Powell 1144 (27). Powell & Edmond- son 743 (10). Pring 157 (2). Pringle 3164 (16); 3165 (14); 3166 (27); 3167 (9a); 4162 (24); 5620 (22a); 5660 (11); 5807 (26); 5875 (3); 9631, 11510 (27); 13039 (10); 13546, 21556 (14). Purpus 1553 (10); 1341 (5); 1866 (27); 2536 (15); 3112 (11); 3113 (27); 3850 (11); 3851, 3852 (15); 5577 (5); 6667 (27); 6680 (13). Rangman 3425 (27). Reiche s.n. (24). Reko 3716 (3). Rock M-351 (10). Roe & Roe 1840, 1889 (27); 2004 (22a); 2173 (10). Roe et al. 808 (24). Rose 1779, 2036, 2622, 3018, 3085 (27). Rose & Hay 5534 (14); 5594 (10); 10121 (27). Rose, Painter & Rose 8784 (14); 8877 (10); 9157 (5); 9158 (27); 9258 (10); 9330 (27). Rusby 376 (11); 377 (10); 378, 379, 380, 381, 382 (27). Rusby & 416 Rhodora [Vol. 71 Pennell 929 (2). Russell & Souviron 32 (14); 142 (24); 215 (17). Rzedowski 20735 (27). Rzedowski & McVaugh 163 (3). Salasar s.n, 7 Sep. 1918 (27); s.m. 6 Nov. 1911 (17); 534 (27). Schnooberger 8002 (27). Schultes et al. 4045 (2). Seaton 277 (10); 357 (27). Seler 1395 (11). Sessé, Moçiño, Castillo, & Mal- donado 2926 (27); 4026 (12). Sharp 44100 (10); 44525 (21); 44542 (10) ; 441534 (3) ; 441726 (27). Sharp & Gilly 46 (10). Sherff 5054 I, II, III, IV, V (2). Shreve 368 (10); 9182 (27). Skutch 65, 711 (2); 1180, 1244 (22d). Smith, Austin A620, H236, NY1369 (2). Smith, C. E., Peterson, & Tejeda 3817 (10); 3895 (27); 3924 (10). Smith, C. L. 294 (22a); 295 (3). Smith, J. D. 2857 (2). Smith, L. C. 481 (26). Smith, R. F. M612 (27). Sneidern 1035, 5677 (2). Soderstrom 747 (27). Soejarto 281 (2). Sorensen 6249 (2), see also “Melchert & Sorensen” and “Melchert, Sorensen, & Crawford" above. Standley 41857 (2); 59373 (27); 59425, 59972, 63214, 67162, 67468, 69469, 71026, 77180, 77280, 80260, 83790, 84205, 84874, 85320, 86289, 90887, 91114 (2). Stanford, Rutherford, & Northeroft 601 (27); 767 (10); 2130 (6); 2436, 2436A (9b); 2537 (10); 2624 (27). Steyermark 29613 (27); 30774, 31903, 32406, 32547 (2); 32864 (27); 34414, 34415, 34676, 48418, 49887 (2); 49971 (27); 50318 (22d); 50341, 50607, 50789 (27); 51974 (22d); 51976 (27). Stork 2992, 4722 (2). Straw & Forman 1389 (10); 1929 (23). Stray & Gregory 1145 (27). Strother 498 (27). Taylor 183 (10). Tucker 1297 (2). Ugent & Flores 1484 (10); 1712, 1745, 2115 (27). Ugent, Uegcnt, & Flores 1211 (24). Urbina s.n. (10) ; s.n. Oct. 1886 (27). Walther s.m. Nov. 1939, 30 Sep. 1954 (17). Waterfall 12551 (27). Waterfall & Wallis 13966 (27). Weaver 743 (10). Weber & Charett 11911 (27). West B-18 (10); H-5 (17). White 2383 (23); 5142 (2); 5264 (27). Whiting 983 (9a). Wilbur & Wilbur 1967, 2016 (27). Williams, L. O. 13496 (10). Williams, L. O. et al. 25821 (2). WINTERBUD PRODUCTION AND FUNCTION IN BRASENIA SCHREBERI FRANKLIN S. ADAMS! Propagule-like reproductive structures are regularly found among many of the higher aquatic Angiosperms. When such structures are dormant overwinter and germi- nate in the spring to produce new plants, they are generally called winterbuds. (fig. 1.) In cases where the complete life cycles of these plants have been observed, it can be rea- sonably said that asexual vegetative reproduction through winterbud production and dissemination plays a significant role in the seasonal maintenance and distribution of these higher aquatic plants (Arber 1963, Sculthorpe 1967). One such plant receiving only limited attention in the literature is Brasenia Schreberi Gmel., which produces large highly developed winterbuds (Arber 1963, Chrysler 1938) or tu- rions (hibernacula) (Sculthorpe 1967). (fig. 3-a.) The genus Brasenia is a monotypic aquatic group in the family Nymphaeaceae belonging in the subfamily Cabom- boideae, the closest living relative is the genus Cabomba. Brasenia is distributed throughout the world in the tem- perate zones and the higher elevations of the subtropics with the exception of continental Europe. However, fossil evidence indicates that Brasenia grew widely throughout Europe during the Pleistocene period (Arber 1963, Engler and Prantl 1891, Fassett 1957, Ridley 1930). No comprehensive study of the life history or ecological treatment of the genus Brasenia exists. Early anatomical data are available however (Schrenk 1888). In addition, Chrysler in 1938 published a limited account of the function of winterbuds of Brasenia based upon field observations in "This study was conducted while the author was an NDEA Graduate Fellow at the University of New Hampshire. I should like to express my gratitude to Mr. Wendell Berry, who, as a graduate student at U.N.H., assisted immeasurably in the early stages of this study. Special thanks are also due to Professors Albion R. Hodgdon and Radcliffe B. Pike for their generous assistance and encouragement throughout. 417 [Vol. 71 Rhodora 418 L^ 199 " KU TUE UTI imd = 1969] Brasenia — Adams 419 southern Maine and laboratory experiments conducted in New Jersey with the material from Maine. Briefly, Chrysler suggests that Brasenia winterbuds abscise in the fall and by remaining dormant on the bottom of the pond overwinter and germinate in the spring thereby fulfilling an important reproductive function. In a review of the aquatic Angiosperms Arber in 1963, based at least in part upon the work of Chrysler (1938), states that winterbuds of Brasenia are specialized struc- tures of reproduction. In addition, Yeo in 1966, in a utili- tarian study of yields of winterbuds found in the com- mercial water of the western Rockies and California, has reached essentially the same conclusions based upon yield data and literature references. Most recently, Sculthorpe (1967), also citing Chrysler, has provided a most convincing description of Brasenia winterbud (turion) production and function. Conversely, preliminary observations conducted by the author during the 1965 season in east-central New Hamp- shire suggested that Brasenia winterbuds are not effective reproduction structures, at least at this latitude and longi- tude. Thus it is this question of functional vegetative propagation via winterbud production and dissemination casually observed and reported by Chrysler that the present paper attempts to resolve. With the foregoing objective clearly defined, the task of designing an adequate research program was simplified. In order to satisfactorily understand the function of Brasenia winterbuds, all events that constitute the so-called “normal” life cycle of the plant would have to be routinely observed and recorded. Two observational sites were selected on the bases of accessibility and differing ecological characteristics. One, Plate 1424 Figure 1. Functional winterbuds of several aquatic vascular plants. a. Ceratophyllum demersum L. b. Elodea Nuttalli (Planch) St. John. c. Utricularia vulgaris L. d. Utricularia pupurea Walt. e. Vallisneria, americana Michx. $43012YJ 1Y1N3WNOSIAN3 OL NOILV13¥ NI HLMOUD VINISWYS “AON O "des ‘Boy Ajor eun[ iow sdy Jow ê L 1 1 i i 1 1 1 Tipuose: [Vol. 71 Rhodora 420 wañoq u! yiðuej Aog *%%4- s. t42-$s - mio sn 9*ab9| — 618) pias —-, $ *inipiedue,; 1969] Brasenia — Adams 421 Wheelwright Pond in Lee, New Hampshire: of moderate size, 100+ acres; of varying depth, one to thirty feet; of pH 6.8 to 7.4 depending upon the time of year, was selected as the primary site, Four observation stations were es- tablished which provided the bulk of the quantitative data reported herein. The second site in Nottingham, New Hampshire on the first branch of the Little River is a small beaver dam and pond: of approximately four acres in size; of relatively uniform depth, two to eight feet; surrounded by a sphag- num swamp at the margins; of relatively acid composition, approaching pH 5.0 at times of low water, was used pri- marily as a confirmation station. In addition, it was possible to follow the course of winterbud distribution for several miles downstream and, most importantly, to record their fate. Typical life history growth phenomena were observed and recorded with time throughout the 1966 growing season which included: initiation of growth; growth rates; flower- ing; winterbud initiation and productivity; winterbud distribution and postseasonal development including ulti- mate fate. Equated to temperature and photoperiod, the combined data for Wheelwright Pond are summarized in figure 2 and explained below. A total of 10 Brasenia plants, selected at random within site #1, were individually identified with a tagged stake thrust securely into the bottom muck adjacent to each plant. The individual plants possessed from one to three attached winterbuds consisting of from one to three segments. (see fig. 3 and 4) Daily observations were made during the period April 11 to April 30. Initiation of growth as “germi- nation" was recorded as having occurred when the first Figure 2. Summary graphical representation in time of various growth phenomena associated with the life cycle of Brasenia Schre- beri Gmel. at Wheelwright Pond, Lee, New Hampshire, during the 1966 growing season. The phenomena of growth have been equated with temperature and photoperiod (day length) at this latitude, 43* 08’ North and Longitude, 71° 00’ West. Detailed explanation in text. 422 Rhodora [Vol. 71 observable meristem of a winterbud was measured to have elongated one inch. Of the ten plants observed, only one failed to grow, the remaining nine plants germinated with- in a two day period, April 24 and 25. The process of germi- nation was confirmed at the three remaining Wheelwright stations and Nottingham during the same period. The initial attempt to determine growth rate as a func- tion of meristem elongation had to be abandoned in the second week. The profusion of leaves being produced throughout the colony made it virtually impossible to re- locate tagged structures from one day to the next. The remaining data were obtained as the average daily number of new leaves arriving at the surface within two sample plots, each measuring 40 X 50 feet. At best, the composite curve “growth rate-leaves” can only be presented as an approximation of real growth rates. Despite these inade- quacies, the observed data are in general agreement with the literature values for hydrophytes. (Arber 1963, Scul- thorpe 1967) No attempt was made to quantitate flowering. Flower- bud initiation (a), day of flowering (b), and period of flowering are presented as they relate in time to all other observed growth phenomena. (See fig. 2.) Since all observed active shoot apicies possess a winter- bud-like morphology (containing only one segment) at all times during the growing season, it was difficult to deter- mine precisely when initiation of true winterbuds had occurred. When a total of ten such structures were observed to have developed multiple segments, winterbud initiation was assumed to have occurred. Subsequent observations confirmed these decisions, The curve *winterbud development" corresponds to the total number of winterbuds produced by twenty plants selected at random and observed weekly during the period July 23 to September 28. In excess of 6,000 winterbuds were counted, examined and collected during the 1966 and 1967 seasons. In the interest of clarity, these data are presented twice. First, 1969] Brasenia — Adams 423 as the events occurred in time (see fig. 2, secondary germi- nation and decay) ; second, as the events occurred quanti- tatively (see fig. 5, summary of winterbud activity). The significance of these data is presented in the results and discussion section below. A '"Friez" (Bendix Aviation Corporation) model 201-W dual-recording thermograph provided an accurate account- ing of water and air temperatures on a continuous basis at site #1, station #1, Wheelwright Pond. The instrument was housed on a roofed but open-sided platform supported by a fixed tripod and elevated three feet above the surface of the water, twenty-five feet from shore. The air tem- perature probe was shaded from sun and rain but otherwise exposed to the atmosphere at all times, The remote water temperature probe was located on the shaded side of the north oriented leg of the tripod at a depth of 55 cm., the average depth of attached winterbuds observed at the site at the time of installation. The temperature data were averaged, means calculated (mean low, mean, mean high), and plotted on the basis of four day intervals for each parameter, water and air. Official U. S. Weather Bureau Station S-79, Durham, New Hampshire, data were averaged and the four day means plotted to indicate the range of differences between at- mospheric temperatures over land and water. For con- venience, all temperatures are expressed in Fahrenheit. Photoperiodicity, expressed as “day length in hours,” was calculated from official U. S. Coast and Geodetic Survey Sunrise and Sunset Tables (U. S. Department of Commerce 1966) for Boston, Massachusetts, and corrected to Latitude 43° 08’ North and Longitude 71° 00’ West, the geographical location of Wheelwright Pond. An experiment designed to test the winterbud’s ability to germinate in relation to water depth was conducted at the Wheelwright Pond Site #1. A long upright stake was driven securely into the bottom muck in nine feet of water. At two foot intervals from the surface downward were attached four 6” X 8” perforated cans with clear polyethy- 424 Rhodora [Vol. 71 jene covers containine three winterbuds each; the cans were arranged vertically along the south-facing side of the stake. The entire apparatus was installed in place on December 2, 1965 and retained throughout the winter and spring seasons. Surface observations were made weekly throughout and the winterbuds collected on June 11, 1966. (See fig. 4.) An experiment designed to test the effectiveness of the winterbud as a reproductive propagule was conducted at the Wheelwright Pond Site #1, during the 1966 season and early 1967, First, 20 ungerminated winterbuds were so- cured to separate six foot segments of light cotton string and tethered individually to stakes scattered throughout the area in water depths ranging from 1 to 5 feet. Second, several hundred detached ungerminated winterbuds were liberated in the same areas and their progress followed daily for several weeks and weekly thereafter. Attempts to attach identifying markers restricted their free move- ment and were abandoned. Consequently, it was possible to observe accurately fewer than twenty-five of the total number released. Nevertheless, the results proved uniform throughout. In total, over 400 individual observations were made on a daily or weekly basis throughout the period of this study. A4 X 6 X 4 foot aerated aquatics tank was established in the University greenhouse for the purpose of observing the phenomenon of winterbud germination under controlled conditions particularly in response to freezing. Despite certain difficulties in controlling the growth of undesirable algae, useful data were obtained; winterbuds subjected to freezing failed to germinate. DISCUSSION All terminal growth, including lateral growth of the rhizome exclusive of the roots, is mediated by complex structures resembling winterbuds at all times during the year. The only obvious morphological difference between actively growing meristems and dormant winterbud struc- 1969] Brasenia — Adams 425 tures involves size and complexity. Terminal growing points are less than 3 cm, in length and possess one lateral leaf, one shoot tip and one primordia of each meristem at any point in time. The dormant structures are essentially multiples of these, thickened transversely and containing up to ten distinct subunits arranged sequentially and/or laterally and varying in length from 3 cm. to a maximum of greater than 20 cm. Furthermore, the overwintering rhizome is tipped by a winterbud which is larger than those found during the growing season and smaller than most found at the tips of shoots in the fall. Significant anatomical and cytological differences between these unique structures are suspected based upon preliminary examinations. Although Brasenia colonies are perennial, no single in- dividual will be more than one season old; no segment of a rhizome will be more than two seasons old. Attempts to isolate the decay organisms which assist this process were inconclusive. Winterbuds can and do germinate. However, there are limitations imposed upon such germinations by the environ- ment. The ability of winterbuds to germinate successfully is affected by water depth. The compensation depth, beyond which germination is incomplete, lies between six and seven feet within the conditions encountered in this study. (See fig. 4) The extent to which leaf, shoot, and root develop- ment is affected by depth is pronounced, however, the physiological significance of such differences has not been investigated. Winterbuds can not withstand freezing. In every case, the slightest degree of freezing involving meristematic tis- sues precluded germination. The first winterbuds that were clearly destined to become overwintering structures were collected on July 30, 1966. The ultimate fate of such structures varied widely and is summarized in figure 5. For convenience (according to their fate within the conditions encountered by this study) it is possible to divide the production of winterbuds into two categories: (1) winterbuds maturing prior to 1 Octo- 426 Rhodora [Vol. 71 ber; and (2) winterbuds maturing after 1 October. The number of winterbuds counted and observed in each cate- gory exceeds 2,500. Winterbuds, despite the presence of an abscission-like layer at the base (Schrenk 1888, confirmed by Chrysler 1938), have not been observed to absise in the way the presence of such a structure would suggest. Abscission, other than in response to external forces, occurs naturally in response to decay of the upright stem portion near the rhizome, The largest and most complex winterbuds were collected as free-floating plant portions prior to October 1 (60% of the total production). Conversely, all winterbuds maturing after 1 October, remained attached to the parent plant unless forcibly removed (40% of the total production). With few exceptions, leaf abscission appears to function normally in both the attached and free-floating conditions. The differences between normal leaf abscission and abnor- mal winterbud abscission suggest the presence of a special winterbud abscission mechanism. Fate of winterbuds associated with free-floating plant parts: All at first float. All floating winterbuds sank when their mucilage covering was reduced by approximately one half. As the temperature of the water decreased during this period, the time required for the mucilage covering to be reduced increased and the winterbuds floated longer. The sunken winterbuds either germinated and then decayed or decayed directly without visible signs of germination having occurred (fig. 2, secondary germination and fig. 3, b-c). It is probably significant that secondary (precocious) germination occurred in the fall at the same time, day- length, when germination occurred in the spring, Small groups of these partially decayed winterbuds were observed Plate 1425 Figure 3. Winterbud structures of Brasenia Schreberi Gmel., illus- trating the phenomena of precocious germination and decay. a. typical attached winterbud, b. precocious germination, phase 1., petiole de- velopment and elongation of apical meristems, c. precocious germina- tion, phase 2., root development from nodal region, d. early stages of decay, e. advanced stages of decay. 1969] Brasenia — Adams 427 428 Rhodora [Vol. 71 at shoreline at the time of icing-in. Winter ice conditions would preclude their survival even were the process of decay incomplete, therefore, the natural fate of winter- buds released prior to 1 October is death. Fate of winterbuds remaining attached to parent plants after 1 October: Forty percent of the tota] production re- mained attached as of 1 October. Thirty-five percent of the total production overwintered and in the attached condition provided the source of the next year’s growth. (None of the attached winterbuds were observed to germinate pre- cociously in the fall.) Of the 5% accidentally or intention- ally released, 4% of the total production were observed to float immediately and remained floating at the margins of the pond at the time of icing-in. All died. One percent, because of entanglement in other submersed vegetation were protected, at least from freezing. It is calculated that fewer than .25% of the total production possessed the potential for survival for the following reasons. (fig. 5.) Of the many thousands of winterbuds produced in 1965 and observed to germinate in 1966, only eight produced mature plants. These were at the Nottingham site and two were collected and preserved during the summer of 1966. Of the six remaining survivors, only two survived the sum- mer, and neither one was found again in the spring of 1967. Exactly what happened to them is not known. It is con- ceivable that with the spring thaw the entire plants were torn free and carried away due to their exposed position. Furthermore, the author has never observed well estab- lished Brasenia plants growing in swift water, not even in occasionally swift water areas. Additional evidence supports the view that free-floating winterbuds are not effective vegetative reproductive agents. Plate 1426 Figure 4. Results of detached winterbud germination at various depths, illustrating observable morphological differences and the com- pensation level below which germination was incomplete. Groups a-d, arranged according to size: < 3 em, 3-6 em, > 6 cm, and depth, a. 1 foot, b. 3 feet, c. 5 feet, d. 7 feet. 429 i Adams "'asenia — Br 1969] 450 Rhodora [Vol. 71 In the case of the tethered winterbuds (see page 424) at the Wheelwright Pond site, all germinated coterminously with the initiation of growth among the attached plants. They did so while sinking and refloating from one to several times, a factor, which if not for their tethered condition would have resulted in their floating into very shallow waters. At best, growth was limited as none of these struc- tures were able to establish roots in the bottom muck due primarily to entanglement with established vegetation. At the end of the 1966 growing season, only one winterbud re- mained alive; all others had died and decayed. Similarly, in the case of the released winterbuds, (see page 424) germination occurred as the winterbuds alter- nately floated and sank. Prevailing westerly winds carried most into shallow water which were then dessicated as the high waters of spring receded. The few winterbuds that sur- vived for longer periods did so through entanglement and ultimately suffered the same fate as the tethered winter- buds. The Nottingham beaver dam provided an unusual op- portunity to witness the phenomena of winterbud germina- tion and ultimate demise en masse. Beaver dam construc- tion facilitates the passage of water through the structure, only in cases of very high water are beaver dams subjected to extensive overflow. Thousands of winterbuds, maturing after 1 October and surviving the winter via entanglement in existing vegetation below the ice, upon release in the spring, floated to the dam and formed an extensive mat approximately one foot thick, thirty feet wide and along the entire eighty foot length of the dam. By mid-July, the waters immediately above the dam were literally obscured by the massive growth of Brasenia winterbuds. By Septem- ber, none remained, all had died and were decaying. The causes underlying this strange behavior have not been in- vestigated. Among several suspected causes are, excessive crowding, competition for nutrients, lack of contact with a favorable substrate and conditions favoring decay. Sev- eral trips below the dam for a distance of two miles revealed 1969] Brasenia — Adams 431 no evidence of winterbud survival save the several plants which were previously discussed. One additional aspect of winterbud behavior needs to be discussed concerning the presence or absence of roots at germination. Free floating winterbuds upon germination always developed roots which emerged from the nodal re- gion from which leaves also develop. Conversely, germina- ting winterbuds when attached to the parent plant were never observed to form roots. In addition, winterbud-like structures found on the rhizome during the growing season (see page 425) formed roots only following the initiation of upright stems bearing leaves. Cautious speculation sug- gests the existence of growth substance mediation, probably in the form of an inhibition response. A SUMMARY OF WINTER BUD ACTIVITY aoe Total Production Freed Before Oct-1 After Oct-1 60% | 40% ——- 35% remain attached all die | . — 5» freed _ .4* winter kill survival 1* improbable probable 15% 25% Figure 5. Summary explanation of the fate of winterbuds freed before 1 October, 1966 and those remaining attached to the parent plant or freed following 1 October, 1966. Detailed explanation in text. 432 Rhodora [Vol. 71 SUMMARY 1. In the areas studied, the abscised winterbud struc- ture of Brasenia Schreberi Gmel. is, in fact, an ineffective agent of vegetative reproduction. 2. In the areas studied, the attached overwintering winterbud structure is, in fact, absolutely essential for the annual survival of the species. 3. In the areas studied, winterbuds do not: a. sink immediately upon abscission. b. overwinter intact on the bottom of the pond. c. germinate and survive, except rarely when freed. 4. If it can be demonstrated at a future time, in another climate that the winterbud does, in fact, constitute an effective agent for vegetative reproduction, it would be reasonable to conclude that the Brasenia winterbud has a positive survival value. Otherwise, the Brasenia winterbud must remain an evolutionary enigma — at best an example of Gailbraithian economics, an economy of waste. UNIVERSITY OF NEW HAMPSHIRE, DURHAM PRESENT ADDRESS 203 BUCKHOUT LABORATORY THE PENNSYLVANIA STATE UNIVERSITY, UNIVERSITY PARK 16802 LITERATURE CITED ARBER, AGNES. 1963. Water Plants, A study of aquatic Angio- sperms. Hafner, New York. CHRYSLER, MINTIN A. 1938. The winter buds of Brasenia. Bull. Torr. Bot. Club. 65: 277-283. ENGLER, A. & K. PRANTL. 1891. Die Natürlichen Pflanzenfamilien, III Teil, 2. Abteilung. Verlag von Vilhelm Engelmann, Leipzig. FASSETT, NORMAN C. 1957. A Manual of Aquatic Plants. Univ. of Wisconsin Press, Madison. RIDLEY, HENRY N. 1930. The Dispersal of Plants Throughout the World. L. Reeve & Co. Ltd., London. SCHRENK, JOSEPH. 1888. On the histology of the vegetative organs of Brasenia peltata Pursh. Bull. Torr. Bot. Club. 15: 29-47. SCULTHORPE, C. D. 1967. The Biology of Aquatic Vascular Plants. St. Martin's Press, New York. 1969] Brasenia — Adams 433 U. S. DEPARTMENT OF COMMERCE. 1966. Tide Tables, Table #4, Local Meantime of sunrise and sunset. U. S. Govt. Printing Office, Washington. Yeo, R. R. 1966. Yields of certain aquatic plants. Weeds. 14: 110-113. CYPERUS FUSCUS IN NEBRASKA AND SOUTH DAKOTA: Seattered specimens of Cyperus fuscus L. were found growing near colonies of Bidens cernua L. and B. comosa (Gray) Wieg. on the wet open sandy flats of the South Platte River 0.8 mile south of Sutherland, Lincoln County, Nebraska on 24 September 1968 (Weedon 4733, KANU). Cyperus fuseus L. was also found on the wet sandy clay soil along the bank of the Keya Paha River 17 miles south of Colome, Tripp County, South Dakota on 16 September 1968 (Stephens 29293, KANU). Cyperus fuscus L. is widespread in Eurasia, especially in rice fields, but appears to be rarely fcund in the United States. This Um- brella-sedge (galingale) has previously been reported at scattered stations from Massachusetts to western New York and Virginia. Our collections represent extensions west- ward from the previously known range of the species in the eastern states to areas in the Midwest. Duplicate specimens were verified as to their identification by Dr. R. L. Mc- Gregor (KANU) and by Dr. Tetsuo Koyama (NY). RONALD R. WEEDON AND H. A. STEPHENS DEPARTMENT OF BOTANY UNIVERSITY OF KANSAS, LAWRENCE 66044 A SEVENTEENTH CENTURY RECORD FOR STEWARTIA J. T. BALDWIN, JR. A population of Stewartia ovata (Cav). Weath.' is on both sides of the Colonial National Historical Parkway south of Williamsburg, Viginia, from the site of William Parks’ paper mill to College (Archers Hope) Creek’, a distance of about two miles. The plant most likely occurs also to the west of College Creek. As early as 1637 the whole area was being patented and cleared; it now sup- ports a relatively young woods of mixed composition of which in various places Stewartia is a striking member. Some individual trees of the genus there reach a height of fifteen feet. Flowers are white, with purple, rose, or yellow filaments (Baldwin 14951, GH). Because the stamens are numerous, differences in flower colors are marked. This population of Stewartia ovata is well-known (Grimes, 1922; Kobuski, 1951; Coe, 1959; Harley, 1961; Massey, 1961; Harvill, 1965), but for a second Coastal Plain station in Virginia, in Lancaster County, there is a single report (Harley, 1961). Otherwise, records are for the moun- tains, hence the name Mountain-Camellia. We now learn that The Reverend Mr. John Clayton (1657-1725) found the Archers Hope plant in the seven- ‘Weatherby (1939) explains that the long-accepted Stewartia pen- tagyna L'Hér. is invalid and that the combination given is necessary. "The mill, dating from 1744, not only supplied the needs of William Parks in Williamsburg but also those of other printers including Benjamin Franklin in Philadelphia. "The colonists landed at the mouth of this creek on May 12, 1607, and proceeded upriver the next day to establish Jamestown. Captain Gabriel Archer advocated settlement at the creek but was outvoted. For more than a century thereafter the creek was known as Archers Hope and was then named for the College of William and Mary. With transfer of the Capital to Williamsburg in 1699 Archer's judg- ment was vindicated, and just as his creek afforded an approach from the James River to Williamsburg so did Queen's Creek from the York. For a time consideration was given to the construction of a canal to join the two creeks. 434 1969] Stewartia — Baldwin 435 teenth century. He arrived in Virginia in May of 1684 to be rector of James City Parish; in May of 1686 he sailed for England, “taking with him some drawings of plants, probably some collections, and certainly a vast fund of information” (Berkeley and Berkeley, 1965). Thereafter he pursued his ministerial work in England and Ireland and followed his natural bent for things scientific. In 1687, in a long letter to Robert Boyle, Clayton wrote an “Account of Virginia’, which is a delight to read (Berkeley and Berkeley, 1968) ; therein, after a picturesque and accurate description of the flowering dogwood, he stated: “I remem- ber near the place caled Coll. Bacons Landing, over agst Archers hope [See Figure 1.], I found a Shrub had a flower like the false flour of this dogg tree only it was a pentape- wt SEAR TRI y z regati & TRECE : r o fir W o uA Washington Q appo. ug JI Piankarank Tüver E 4 P we a "n V M ne. TN, Cherry V y < LP 4 o r [ro à Ç O t. ` üs wigs" i " dy ae SN Lv" d Cc “STERG 5 wf, Milford Haven E "o Ju" mes far Ç O tr hu Vhverkmorten eg = Ë | € > N A A dk o a me YTY aw AS B N AO Q S 2 P A M e 9 x a PT = "m A z Pp EU De elt ‘ b TM T 3 Ç M , n "i New Point Comfort Tap 1 € qw M . ous Cherto Oy . [^ t gam, Qara S : E ^ Y| M "< b^ s ë C ° Ñ Sorority p. y D » 4 > ^ f nar Qu DaS E 'ILLIAMS, 3. BURG e. A + - 3 m» ho Iland nue Ww Green es 8 y Ww fi "S "Bm GFF yt YorkO , `V Wand 2. JAN E RE S WARWICK S 4 Bm. ss CAPE ees t w. ON š 8 ENS a: m i 9.4 DW ads o S SS us , ees prre nnd KUHAR Em 3 Nei et pes To Z) MRE pcuibcrey, E CITY COUNT J aeo .'z pit P» p4. Wa. E 2 Jae new š ° ellos , 7A Y z AMA ^ M UA F ag E eost ete pto it = gy £ A coenam cH io e oT T Y A " fS ` x Yo Ber mor m d z Ñ iv f C Burwell “hat! sl oint Comfort ee A M ` ^ š £ | $ /£ m = < qu illowbys Pt E : < ul em n» A e. 2 ig M» A eot ge of — a Sia P ° O Cshirid m dj "ñ a: ts P f pes V» < ¿ — ut #@ RY P p di Norfolk i N s> «f Toen AE ` V J#EsSRRIN CESS EANAN w e9 a < gÙ "2 > (fs oS DU Iu : Y Mead j ^ ¿ >> jg 3 / o SAT. x x oW ee ruf Z ç ad Figure 1. Portion of the 1751 Fry and Jefferson map of Virginia, taken from the facsimile, 2nd ed., Charlottesville, University Press of Virginia, 1966. 436 Rhodora [Vol. 71 Figure 2. Stewartia ovata from the Archers Hope population as drawn by Frances Dayton, 1956. talon, the Dogg tree flower being as I s’d a tetrapetalon, afterwards comeing that way, I went to see wt fruit it bear, they were not nere ripe, but were about the bignesse of a Walnut, & had 5 ridges ran down the fruit, winged, How bigg they might grow I know not” 1969] Stewartia — Baldwin 437 The reverend gentleman was, of course, recording: obser- vations on the plant that exactly one hundred years later Cavanilles was to describe as Malachodendron ovatum. Compare Clayton’s description with the drawings made by Frances Dayton of specimens taken in 1956 from the location visited by the rector in 1685: flowering, June 29; fruiting, July 29 (Figure 2). The letter quoted above was written fifty-four years before Linnaeus described Stew- artia, sixty-one years before John Mitchell published Malachodendron, sixty-six years before Linnaeus (in Species Plantarum, 1753) made the combination Stewartia Malachodendron (type species), two hundred thirty-four years before Grimes collected specimens at approximately the same place (Grimes 3818, NY: in flower, June 24, 1921; on dry wooded slope along tributary of College Creek south of Williamsburg). Clayton associated in his memory the plants that came to be named Cornus florida and Stewartia ovata, Inter- esting it is, therefore, that the country people in the en- virons of Williamsburg sometimes call Stewartia as they find it in the woods Summer Dogwood (so reported to me by J. B. Brouwers, former Landscape Superintendent of Colonial Williamsburg who transplatend specimens from the wild into town). Though only two species of Stewartia occur in the United States — about six in Asia, the American representatives have a complex taxonomic history (Fernald, 1936) which, I suspect, needs further clarification. John Clayton (1686- 1773), the botanist, sent dried specimens of Stewartia to Gronovius in Holland and at least one living plant to Marc Catesby in England. Clayton collected in areas where we now know only S. ovata and not S. Malachodendron, though in the years that have intervened since Clayton arrived in Virginia in 1705 many Coastal Plain populations of either or both species might have been completely destroyed. But we do know from the Archers Hope stand that Stewartia does have survival persistence even in an agricultural area. Fernald seemed to be under the impression that Clayton 438 Rhodora [Vol. 71 sent back specimens only of S. Malachodendron: 1 question that. John Mitchell wrote a letter to Marc Cateby (Natural History of Carolina, etc.) with the important statement: "There are five styles, as I showed you in a specimen I have." Mitchell was at the time discussing his genus Mala- chodendron, whence Linnaeus took the specific epithet. Nowadays, S. ovata is considered to be unique in having five styles. COLLEGE OF WILLIAM AND MARY WILLIAMSBURG, VIRGINIA 23185 LITERATURE CITED BERKELEY, EDMUND and DOROTHY S. BERKELEY. 1965. The Rever- end John Clayton. University Press of Virginia, Charlottesville. Eds. 1968. Another *Ac- count of Virginia": By the Reverend John Clayton, The Virginia Magazine 76: 415-436. COE, FREDERICK W. 1959. Stewartia malachodendron versus Stew- artia ovata grandiflora. Nat. Hort. Mag. 38: 175-176. FERNALD, M. L. 1936. Plants from the outer Coastal Plain of Vir- ginia. Rhodora 38: 376-404, Pl. 440-4406. GRIMES, E. J. 1922. Some interesting plants of the Virginia Coastal Plain. Rhodora 24: 148-152. HARLEY, WINIFRED J. 1961. A second station for Stewartia ovata on the Coastal Plain. Rhodora 63: 206. HARVILL, A. M., JR. 1965. The mountain element in the flora of the Peninsula of Virginia. Rhodora 67: 393-398. KoBUSKI, C. E. 1951. Studies in the Theaceae, XXI. The species of Theaceae indigenous to the United States. Jour. Arnold Arboretum 32: 123-138. MassEY, A. B. 1961. Virginia Flora. Tech. Bull. 155, Va. Agric. Experiment Station, Blacksburg. WEATHERBY, C. A. 1939. Two new combinations in Stewartia. Rhodora 41: 198. PARASITISM IN XIMENIA (OLACACEAE) R. DEFILIPPS Ximenia is a small genus of tropical shrubs and trees in the dicotyledonous family Olacaceae. A trend of parasitism occurs in the order Santalales, which includes Olacaceae, Loranthaceae, Viscaceae, Santalaceae, and Misodendraceae, as circumscribed by Kuijt (1968). In the Olacaceae, which is thought to be the plexus from which all other Santalalean families take their origin, parasitism has been documented in three of the approximately twenty genera: Ximenia, Olax, and Ptychopetalum. Root-hemiparasitism in the Olacaceae was first reported by Heckel (1900), who observed haustoria of Ximenia americana L. var. americana attached to Piper chaba Hunter. He planted Ximenia seeds collected near Libreville, Gabon, in individual pots containing Tamarindus indica L. (Leguminosae), Erythroxylon coca Lam. ( Erythro- xylaceae), Hura crepitans L. (Euphorbiaceae), Ficus lauri- folia Hort. (Moraceae) and Piper chaba Hunter (as Chavica officinarum Mig., Piperaceae). Of these intended host species, only Piper was attacked after two years. Barber (1907) observed haustoria of Ximenia americana var. americana in Madras, India, attached to roots of a dicotyledonous plant. Some haustoria attained the large size of one square inch in surface area. The haustorial and host tissues were illustrated, but the host species was not named. The present study was made to confirm parasitism in Ximenia. In November, 1965, shrubs of X. americana var. americana were received from a natural habitat near Immokalee, Florida, and specimens of Pelargonium hor- torum Bailey (Geraniaceae) were later placed in a pot with one of them. When harvested in June 1968, numerous haustoria were found attached to the host roots (Fig. 1). Several of the large Ximenia shrubs are now growing in the absence of host plants, which may indicate that these organisms are facultative, rather than obligative, parasites. 459 440 Rhodora [Vol. 71 CSS T a Am A2 ome Jo a Ë j $ C)S x p i Ur P Fig. 1. Haustoria of Ximenia americana L. var. americana (pale) attacking roots of Pelargonium hortorum Bailey (dark) and con- comitantly parasitizing itself (arrows), X 1. 1969] Ximenia — DeFilipps 441 Seeds of Ximenia americana var. americana were ob- tained from Immokalee in September 1966 and grown in individual pots. In May, 1967, these intended host species were selected arbitrarily and introduced into the pots: Bryophyllum pinnatum (Lam.) Oken (Crassulaceae), Rhoeo spathacea (Swartz) Stearn (Commelinaceae), Zamia flori- dana DC. (Cycadaceae). All of the host plants, harvested in October, 1967, and January, and June, 1968, had haus- toria attached to their roots. This caused neither a visible lack of vigor in the hosts nor increase in vigor in the para- sites. Heckel (1899, 1900) noted that Ximenia americana var. americana, when growing in the absence of a host, will produce haustoria which attack its own roots, the subter- ranean portion of the stem, and the endosperm of the seed from which the plant grows. In the present study, the shrub which parasitized Pelargonium concomitantly produced haustoria which attached to its own roots (Fig. 1). A specimen of X. caffra Sonder var. natalensis Sonder, grown alone from seed obtained from Salisbury, Rhodesia, also formed a number of haustoria which parasitized its own roots. Other examples of this phenomenon in the Santalales are cited by Fineran (1965). The attachment of Ximenia haustoria to non-living ob- jects has not been previously reported; prior reports exist for Olax (Olacaceae) and genera in the Santalaceae, Loran- thaceae and Scrophulariaceae. In this study, haustoria on roots of X. americana var. americana shrubs were observed attached to a small stone, a piece of charcoal, bits of plastic intentionally sown in potting soil, and to the cellophane-like coating on the interior of a metal can in which one speci- men was growing (Fig. 2). Therefore, the presence of living host roots is not always necessary for the initiation of Ximenia haustoria. A similar conclusion was made by Piehl (1962) for Melampyrum and Fineran (1965) for Exocarpus. The attachments to non-living objects are FAA-preserved in the Southern Illinois University Herbarium. Histological 442 Rhodora [Vol. 71 Fig. 2. Haustoria of Ximenia americana L. var. americana at- tached to non-living objects. Upper left: Stone, X 10. Lower left: Cellophane-like coating on metal, X 2. Upper right: Plastic, X 3.5. Center right: Charcoal, X 4. Lower right: Cellophane-like coating removed from metal, X 2. 1969] Ximenia — DeFilipps 443 studies of the parasite-host attachments and attachments between Ximenia roots on the same plant are in progress in the laboratory of Dr. B. Fineran, Botany Department, University of Canterbury, Christchurch, New Zealand. Plants were grown in the Production Greenhouse, Southern Illinois University, Carbondale. Mr. R. Wadlow, of Immokalee, Florida, supplied seeds and shrubs of Ximenia americana. Mr. T. Miller and Dr. H. Wild, Chief Botanist, Federal Herbarium, Salisbury, Rhodesia supplied seeds of X. caffra. The figures are by Mr. C. Seliger. DEPARTMENT OF BOTANY SMITHSONIAN INSTITUTION WASHINGTON, D. C. 20560 LITERATURE CITED BARBER, C. 1907. Parasitic trees in southern India. Proc. Cam- bridge Philos. Soc. 14: 246-256. FINERAN, B. 1965. Studies on the root parasitism of Exocarpus bidwillii Hook.f. VI. Haustorial attachment to non-living ob- jects and the phenomenon of self-parasitism, Phytomorphology 15: 387-399. HECKEL, E. 1899. Sur le processus germinatif dans la graine de Ximenia americana L. et sur la nature des écailles radiciformes propres à cette espéce. Rev. Gen. Bot. 2: 401-408. 1900. Sur le parasitism du Ximenia americana L. Compt. Rend. Acad. Paris 131: 764-765. Kut, J. 1968. Mutual affinities of Santalalean families. Brittonia 20: 136-147. PreHL, M. 1962. The parasitic behavior of Melampyrum lineare and a note on its seed color. Rhodora 64: 15-23. PHYTOPLANKTON FLORA OF NEWFOUND AND WINNISQUAM LAKES, NEW HAMPSHIRE GERHARD K. GRUENDLING ARTHUR C. MATHIESON Few comprehensive seasonal studies have been made of New England phytoplankton species (particularly in large lakes), and most of these investigations have been restricted to Massachusetts (Auyang, 1962; Croasdale, 1935; Spencer, 1950; Stone, 1900; Wright, 1964) and Connecticut (Hutch- inson, 1944; Hylander, 1928; Patrick, 1943; Riley, 1940). Investigations of the freshwater algal flora of New Hamp- shire have dealt primarily with its rich desmid populations (Cushman, 1930b, c, 1905a, b, c, 1906, 1907a, b, 1908a, b; Hastings, 1892; Whelden, 1942; Wood, 1869). The New Hampshire Fish and Game Department has conducted limited investigations of phytoplankton in conjunction with its biological surveys, but only generic designations were given (Edmundson, 1938; Edmundson and Fuller, 1937 ; Cole, 1939). As far as we are aware, only two other pub- lished accounts describe phytoplankton species from New Hampshire (Collins, 1909, 1912, 1918b; Gustafson, 1942). The present systematic account of phytoplankton is a portion of a detailed investigation of the composition, periodicity, and abundance of phytoplankton species in rela- tion to the trophic status of Newfound and Winnisquam Lakes, New Hampshire (Gruendling and Mathieson, in press). Newfound Lake is located in Grafton County (Figure 1), while Winnisquam Lake is in Belknap County (Figure 2). Both lakes are preglacial land forms which were slightly modified (mainly deepened) by Pleistocene glaciation (Billings, 1956). The lakes are similar in size, maximum depth, altitude, underlying geology, and certain chemical and physical characteristics, The chief difference between the two lakes is that Newfound Lake is oligo- trophic, while Winnisquam Lake is receiving treated sewage 444 1969] Phytoplankton — Gruendling and Mathieson T T T 43° 42 NEWFOUND LAKE — —.-> —— F r 1Mi . d um. z I 43° 40' mM A eet ` r ` . . s e E +" »* - E — 43^ 38' own p o' sS P ed 1 wot - \ vitae V 1 445 7*aa' 1 Fig. 1. Map of Newfound Lake. 446 Rhodora [Vol. 71 a T T 7 T — —71T = . 3 PII 24 r Laconi’ ! WINNISQUAM LAKE SCALE © EY —————— ° e" PETI ^ ral} 3 LACONIA - m | "oo ND NLIS Yawac Ming, "NI. T2 - 4 SEWAGE TREATMENT tant N 4 H 2 E w `. 4 ~ “co VON = “ Um “o, ` s , E 4 * 30! 2 p EET p - ^ ^ z - LN rd - wot ut vd» oo P dd ow - ^T aot "d ` - - - Ld ' z - ntt \ nto i L 1 L Fig. 2. Map of Winnisquam Lake. 1969] Phytoplankton — Gruendling and Mathieson 447 effluent and is in the early stages of eutrophication. The following paper describes the floristic differences between the two lakes and gives specific information regarding the seasonal and vertical distribution of many phytoplankton species. New records for the state of New Hampshire and New England are also designated Regular collections (approximately fortnightly) were made at Newfound and Winnisquam Lakes from June 25, 1966 to November 10, 1967. Three stations were occupied at each lake and samples were taken from various depths (1, 3, 5, 10, 15, and 20 meters) with a 4-liter Van Dorn water sampler as well as from vertical hauls (20 meters) with a #25 plankton net. Living samples, which were con- centrated through a .45 ,» millipore filter or a plankton net, were used whenever possible for the identification of organisms. Fixed samples (1% Acid Lugol's Solution) were only used occasionally for identification of species. The generic designations listed by Smith (1950) were used whenever possible. Specific identifications were determined from a variety of sources: Ahlstrom (1937) ; Eddy (1930) ; Huber-Pestalozzi (1938-1961) ; Patrick and Reimer (1966) ; Prescott (1962); Smith (1920, 1924); Teiling (1967); West and West (1904-1912); West, West and Carter (1923). Voucher specimens have been deposited in the University of New Hampshire Herbarium. Synopsis of Species' CYANOPHYCEAE *Anabaena circinalis Rabenhorst (Figure 7). Common at Winnisquam Lake during the summer; mixed with Anabaena flos-aquae, but not as abundant. Akinetes were present throughout the growing season (June to October). *Anabaena flos-aquae (Lyngbye) De Brébisson (Figure 10). A major constituent of water blooms at Winnisquam Lake in July and August, often occurring in large tufts. 1 *presumed to be a newly published record for New Hampshire. **presumed to be a newly published record for New England. 448 Rhodora [Vol. 71 Rarely found at Newfound Lake. Akinetes were present throughout the growing season (June to November). ** Anabaena scheremetievi Elenkin (Figure 6). Found in small quantities at Winnisquam Lake from August to November. Akinetes were present in all collections. Aphanizomenon flos-aquae (L.) Ralfs (Figure 11). Pres- ent throughout the year at Winnisquam Lake. The largest quantities were found during the summer and fall. Akinetes were only observed from September to November, 1967. During the winter and spring the tri- chomes were “rafted” while throughout the rest of the year they occurred as single trichomes. Cells were slightly smaller (3.9-5.0 » in diameter) than those listed by Prescott (1962). * Aphanothece clathrata G. S. West. Abundant in the sur- face waters at Newfound Lake from June to November. *Chroococcus limneticus Lemmermann. Common at New- found Lake from June to November. *Chroococcus minimum (Keissl.) Lemmermann. Rare; found at Newfound Lake from June to November. *Gloeothece linearis Naegeli. Common throughout the water column at Newfound Lake from June to Novem- ber. *Gloeotrichia echinulata (J. E. Smith) Richter (Figure 5). Present near the surface of Winnisquam Lake from July to September. Akinetes were present throughout the Fig. 3. Dinobryon cylindricum Imhof with statospores, 500. Fig. 4. Dinobryon bavaricum Imhof with statospores, X500. Fig. b. Gloeotrichia echinulata (J. E. Smith) P. Richter with akinetes, 500. Fig. 6. Anabaena scheremetievi Elenkin with akinete, 500. Fig. 7. Anabaena circinalis Rabenhorst with akinetes, 500. Fig. 8. Melosira italica var. tenuissima, (Grun.) Mull. with auxo- spore, X500. Fig. 9. Germinating auxospore of Melosira italica var. tenuissima, Fig. 10. Anabaena flos-aquae (Lyngbye) De Brébisson with aki- netes, 500. Fig. 11. Aphanizomenon flos-aquae (L.) Ralfs with akinete, Phytoplankton — Gruendling and Mathieson 449 1969] 450 Rhodora [Vol. 71 growing season. Although it never formed water blooms, on occasion it was dense enough to be seen with the un- aided eye. *Gomphosphaeria aponina var. delicatula Virieux. Present throughout the year at both lakes, but never in great abundance. G. aponina has been recorded from Massachu- setts (Croasdale, 1935) and from Connecticut (Hylander, 1928), but the var. delicatula has not previously been recorded from New England. Gomphosphaeria lacustris Chodat. A dominant organism throughout the year at Newfound Lake. Its maximum abundance was recorded during July and August. **Gomphosphaeria lacustris var. compacta Lemmermann. Found in small quantities throughout the year at New- found Lake; only evident from August to November at Winnisquam Lake. *Merismopedia punctata Meyen. Common at Newfound Lake from June to November. *Merismopedia tenuissima Lemmermann. Abundant at Newfound Lake from June to November; a maximum peak was recorded in July. *Oscillatoria granulata Gardner. Most common in shallow areas of Winnisquam Lake after heavy rainfalls. Prob- ably not a true planktonic species. *Oscillatoria limnetica Lemmermann. Found throughout the year at Winnisquam Lake; the dominant organism in April, May, and November and December. Its populations showed a marked seasonal difference vertically within the water column. During the winter, large quantities were present in the upper three meters but by the end of March the bulk of the population had settled to 10-15 meters — even though it was still increasing in numbers. During the remainder of the year it again sank to lower depths. **Spirulina laxa G. M. Smith. Found at Winnisquam Lake from July to September. Only evident at one location (10 to 15 meters) ; rare, *Tolypothrix lanata Wartmann. Found in small quantities 1969] Phytoplankton — Gruendling and Mathieson 451 from July to November at Winnisquam Lake. Only col- lected in shallow portions of the lake, and probably not a true planktonic species. Akinetes were never observed. CHLOROPHYCEAE * Ankistrodesmus faleatus (Corda) Ralfs. Relatively com- mon at Winnisquam Lake from June to October but never reaching large numbers. * Ankistrodesmus falcatus var. acicularis (A. Braun) G. S. West. Present at both lakes from June to October, but not as common as the typical plant. * Ankistrodesmus falcatus var. mirabilis (W. and G. S. West) G. S. West. Very common throughout the year at Winnisquam Lake. One of the dominant organisms dur- ing the spring overturn (April and May). *Botryococcus braunii Kuetzing. Common throughout the year at both lakes; particularly during the winter at Newfound Lake. The color of the mucilage sheath varied from dark green to dark orange-brown. **Botryococcus protuberans var. minor G. M. Smith (Fig- ure 20). Present throughout the year at Newfound Lake, but not as abundant as B. braunii. Only found at Winni- squam Lake from February to early June. *Characium curvatum G. M. Smith. Found during the summer at Winnisquam Lake; growing epiphytically on fragments of Oedogonium. *Characium gracilipes Lambert. Found as an epiphyte on Oedogonium during August and September at Winni- squam Lake. **Characium limneticum Lemmermann (Figure 15). Found at Winnisquam Lake from July to September; rare. **Chlamydomonas dinobryonis G. M. Smith (Figure 12). Cells growing within the lorica of Dinobryon cylindricum. Observed a few times during June and July at Newfound Lake. **Chlorangium stentorinum (Ehrenberg) Stein. Found occasionally during July and August at Winnisquam Lake; growing attached to copepods. 452 Rhodora [Vol. 71 **Closteriopsis longissima Lemmermann (Figure 19). Found throughout the year at Winnisquam Lake but never in great abundance. It was always confined to deep water samples (15 meters and below) where light was barely detectable. Closterium acerosum (Shrank) Ehrenberg. A few speci- mens were observed in September and November, 1966 at Winnisquam Lake; rare. Closterium dianae Ehrenberg. Seen once in June at Winnisquam Lake. Closterium kuetzingii De Brébisson. The most common species of Closterium found at both lakes; present throughout the year but never in any great quantity. Closterium moniliferum (Bory) Ehrenberg. Seen once in November at Winnisquam Lake. *Coelastrum microporum Naegeli. Common from June to November at Winnisquam Lake. Only sporadically pres- ent throughout the summer at Newfound Lake. *Cosmarium bioculatum De Brébisson. Common through- out the year at Newfound Lake. It was usually found at depths below 10 meters. Cosmarium contractum var. papillatum W. and G. S. West. The most abundant species of Cosmarium found at both lakes. It was most common at Newfound Lake, where it occurred throughout the year. At Winnisquam Lake it was only recorded from June to November. Fig. 12. Chlamydomonas dinobryonis G. M. Smith, 500. Fig. 18. Uroglenopsis americana (Calkins) Lemmermann, X500. Fig. 14. Schroederia judayii G. M. Smith, X750. Fig. 15. Characium limneticum Lemmermann, 500. Fig. 16. Colacium arbuscula Stein, X500. Fig. 17. Colacium vesiculosum Ehrenberg, 500. Fig. 18. Elakatothrix viridis (Snow) Printz, X500. Fig. 19. Closteriopsis longissima Lemmermann, 500. Fig. 20. Botryococcus protuberans var. minor G. M. Smith, X500. Fig. 21. Kirchneriella contorta (Schmidle) Bohlin, X750. Fig. 22. Nephrocytium limneticum (G. M. Smith) G. M. Smith X500. Fig. 23. Nephrocytium lunatum W. West, X500. Phytoplankton — Gruendling and Mathieson 453 1969] 454 Rhodora [Vol. 71 *Cosmarium punctulatum De Brébisson. Common from July to November at both lakes. *Crucigenia rectangularis (A. Braun) Gay. One of the dominant Chlorophyceae found at Newfound Lake; evi- dent in all collections except in April and May. Its popu- lation size was relatively stable throughout the growing season. The alga was only observed a few times from July to November at Winnisquam Lake. Desmidium baileyi (Ralfs) Nordstedt. Found from July to October at Winnisquam Lake. Present as very long chains; uncommon. Desmidium swartzii C. A. Agardh. Sporadically present throughout the summer and early fall at both lakes. *Dictyosphaerium ehrenbergianum Naegeli. Occurring in small quantities from September to November at Winni- squam Lake; uncommon. *Dictyosphaerium pulchellum Wood. Very abundant throughout the year at Winnisquam Lake. It was the dominant Chlorophyceae during the summer and early fall. Less common at Newfound Lake, but still occurring throughout the year. *Dimorphococcus lunatus A. Braun. Common from Sep- tember to November at Winnisquam Lake. *Elaktothrix gelatinosa Wille. Common from June to No- vember at Winnisquam Lake. in very small quantities throughout most of the year (except April and May) at both lakes. Euastrum didelta (Turpin) Ralfs. Only a few specimens of this species were found during October and November, 1967 at Newfound Lake. *Euastrum pulchellum De Brébisson. Common from June to November at Newfound Lake. *Eudorina elegans Ehrenberg. Very abundant throughout the year at Winnisquam Lake; reaching peak numbers during July and August. *Gloeocystis ampla (Kuetzing) Lagerheim. Found occa- sionally from July to October at Winnisquam Lake. 1969] Phytoplankton — Gruendling and Mathieson 455 *Gloeocystis gigas (Kuetzing) Lagerheim. Common at both lakes from June to November. However, it was most abundant at Newfound Lake where it was a domi- nant member of the Chlorophyceae population during the summer and fall. Gloeocystis vesiculosa Naegeli. One of the dominant green algae present at Winnisquam Lake from June to Novem- ber. Its maximum development occurred during July and August. Gonatozygon aculeatum Hastings. Abundant on two occasions (August and September, 1966) in the shallow portions of Winnisquam Lake. Gonatozygon monotaenium De Bary. Found once at New- found Lake during August and September; rare. *Gonatozygon pilosum Wolle. Abundant during August and September, 1966 at Newfound Lake; never seen again after September, 1966. Gymnozyga moniliformis Ehrenberg. Observed once in September, 1967 at Newfound Lake. Hyalotheca dissiliens (J. E. Smith) De Brébisson. Found sporadically throughout the summer and fall at both lakes. **Kirchneriella contorta (Schmidle) Bohlin (Figure 21). A few specimens were found in two collections made in November, 1967 at Winnisquam Lake; rare. *Kirchneriella lunaris (Kirchner) Moebius. The common- est species of Kirchneriella found at Winnisquam Lake; recorded from July to November. *Kirchneriella lunaris var. dianae Bohlin. Found several times during August at Winnisquam Lake; mixed with the typical plant. **Kirchneriella obesa var. aperta (Teil. Brunnthaler. Un- common, but found in most collections throughout the year at Winnisquam Lake. Only the typical plant has been recorded in New England — i.e. from Connecticut (Conn and Webster, 1908; Hylander, 1928) and Massa- chusetts (Croasdale, 1935). 456 Rhodora [Vol. 71 Micrasterias radiata Hassall. One specimen was seen in September, 1966 at Winnisquam Lake. Micrasterias radiosa (Lyngbye) Agardh. Observed once at Winnisquam Lake in August, 1967. Mougeotia sp. Vegetative material was found in the plankton during the spring and summer at Winnisquam Lake. **Nephrocytium ecdysiscepanum W. West. A few speci- mens were observed during July and August at Winni- squam Lake. **Nephrocytium limneticum (G. M. Smith) G. M. Smith (Figure 22). Very common at Winnisquam Lake from June to November. Only a few specimens were found at Newfound Lake during August to October. **Nephrocytium lunatum W. West (Figure 23). Common from July to October at Winnisquam Lake. Oedogonium spp. A number of different vegetative fila- ments were observed in the plankton from May to October at Winnisquam Lake. It was often heavily epiphytized. *Oocystis borgei Snow. The most common species of Oocytis found at both lakes. It was evident throughout the year, but it was never in any great abundance. *Oocystis elliptica W. West. Only a few specimens were found during August and September, 1966 at Winni- squam Lake; rare. *Oocystis lacustris Chodat. A few specimens were found during October and November, 1966 at Winnisquam Lake; rare. *Oocystis parva W. and G. S. West, A common species of Oocystis at Winnisquam Lake from July to September. *Pediastrum araneosum (Raciborski) G. M. Smith. Found only twice in September at Winnisquam Lake; rare. *Pediastrum boryanum var. longicorne Raciborski. Found once at Newfound Lake during October, 1966. The var. longicorne is differentiated from the typical species be- cause of its swollen apical lobes. *Pediastrum duplex Meyen. The most common species of Pediastrum found at Winnisquam Lake. Evident 1969] Phytoplankton — Gruendling and Mathieson 457 throughout the year, but never in very large numbers. *Pediastrum duplex var. clathratum (A. Braun) Lager- heim. Common throughout the year at Winnisquam Lake. *Pediastrum duplex f. gracilimum W. and G. S. West. Com- mon at Winnisquam Lake from July to November. Ac- cording to Prescott (1962) it is a growth form of the typical plant. *Pediastrum simplex var. duodenarium (Bailey) Raben- horst. Found once in June, 1967 at Newfound Lake. *Pediastrum tetras (Ehrenberg) Ralfs. A few specimens were found in July and August at Winnisquam Lake; uncommon, *Pediastrum tetras var. tetraedon (Corda) Rabenhorst. More abundant at Winnisquam Lake than the typical plant; present from July to November. Pleurotaenium trabecula (Ehrenberg) Naegeli. Present sporadically throughout the summer and fall at Winni- squam Lake. *Quadrigula closterioides (Bohlin) Printz. Common throughout the year at Newfound Lake. Less common at Winnisquam Lake, and only occurring from June to November. *Quadrigula lacustris (Chodat) G. M. Smith. Common at Winnisquam Lake from July to October. *Scenedesmus abundans (Kirchner) Chodat. Only found a few times during July and August at Winnisquam Lake; rare. *Scenedesmus acutiformis Schroeder. Observed at Winnis- quam Lake from June to November. *Scenedesmus arcuatus var. platydisca G. M. Smith. Com- mon from June to November at Winnisquam Lake. *Seenedesmus bijuga (Turp.) Lagerheim. Found occa- sionally during July and August at Winnisquam Lake. *Scenedesmus brasiliensis Bohlin. Common at Winnisquam Lake from June to November. *Scenedesmus dimorphus (Turp.) Kuetzing. Common at Winninsquam Lake from June to November. *Scenedesmus opoliensis P. Richter. Observed once in 458 Rhodora [Vol. 71 October and once in November, 1967 at Winnisquam Lake. Scenedesmus quadricauda (Turp.) De Brébisson. The dominant species of Scenedesmus in Winnisquam Lake; present throughout the year. *Scenedesmus quadricauda var. parvus G. M. Smith. Only found a few times during August and September at Winnisquam Lake. *Scenedesmus quadricauda var. westii G. M. Smith. Found a few times from August to October at Winnisquam Lake. *Scenedesmus serratus (Corda) Bohlin. The only species of Scenedesmus found at Newfound Lake. It was found from June to November, but never in any great abund- ance. **Schroederia judayi G. M. Smith (Figure 14). Very abundant from April to November at Winnisquam Lake. Maximum populations were evident in the spring and early fall, and generally in the upper five meters of the water column. The cells were either straight or arcuate. *Schroederia setigera (Schroeder) Lemmermann. Present throughout the year at Winnisquam Lake, but not as abundant as S. judayi. *Sorastrum spinulosum Naegeli. Only found a few times from August to October at Winnisquam Lake; rare. *Sphaerocystis schroeteri Chodat. Very common from April to October at Winnisquam Lake. Sparse at New- found Lake, and only occurring in July and August. The Fig. 24. Staurastrum pentacerum (Wolle) G. M. Smith (top view) X500. Fig. 25. Staurastrum pentacerum (end view), X500. Fig. 26. Staurastrum limneticum var. cornutum G. M. Smith (top view), X 500. Fig. 27. Staurastrum limneticum var. cornutum (end view), x 500. Fig. 28. Staurastrum pseudopelagicum W. & G. S. West (top view), X500. Fig. 29. Staurastrum manfeldtii Delp. (Top view), 500. Fig. 30. Staurastrum pingue Teiling (top view), 500. 1969] Phytoplankton — Gruendling and Mathieson 459 28 ^T PS S58 PISS »» YA MK 29 460 Rhodora [Vol. 71 plant was typically found as a many celled colony, but occasionally as a single cell. Spirogyra spp. A number of different vegetative fila- ments (i.e., some with single and others with double chloroplasts) were found in the plankton at Winnisquam Lake from April to November. *Spondylosium planum (Wolle) W. and G. S. West. A common desmid from June to November at Winnisquam Lake. Only found occasionally in the fall at Newfound Lake. Staurastrum anatinum var. longibrachiatum W. and G. S. West. A few specimens were found in each collec- tion made throughout the year at Winnisquam Lake. The typical plant was previously recorded from Massachu- setts (Cushman, 1930a; Stone, 1900), Maine (West, 1888), and New Hampshire (Collins, in PHYC. BOR-AM. XL, #1984). *Staurastrum ankyroides Wolle. Found once in Newfound Lake in September, 1966. Staurastrum arctiscon (Ehrenberg) Lund. Found once from each lake; rare. **Staurastrum cingulum (W. and G. S. West) G. M. Smith. Present sporadically throughout the year at Winnisquam Lake. Staurastrum johnsonii W. and G. S. West. Common throughout the year, except April and May, at Winni- squam Lake. Typically the plant is biradiate, but one specimen was found in which one semicell was triradiate and the other biradiate. **Staurastrum limneticum var. cornutum G. M. Smith (Figures 26 & 27). Common at both lakes from August to November. Found with 4, 5, and 6 processes. **Staurastrum manfeldtii Delp. (Figure 29). Abundant from August to November, 1966 at Winnisquam Lake. Not seen during 1967. **Staurastrum pentacerum (Wolle) G. M. Smith (Figures 24 and 25). The most common species of Stawrastrum found at Newfound Lake. Abundant throughout the year. 1969] Phytoplankton — Gruendling and Mathieson 461 The number of processes ranged from 3-5; the “janus” forms with 3+4 or 3+5 were the most common. *Staurastrum pilosum Archer. Present at Winnisquam Lake in small numbers during October and November, 1966, but abundant from June to October, 1967. **Staurastrum pingue Teiling (Figure 30). The dominant desmid at Winnisquam Lake; maximum populations oc- curred during October and November. It was generally most abundant at approximately 10 meters. *Staurastrum pseudopelagicum W. and G. S. West (Figure 28). Common at Winnisquam Lake from June to Novem- ber. *Staurastrum vestitum Ralfs. Found a few times in August and September at Winnisquam Lake; rare. **Staurodesmus bulnheimii (Raciborski) Brook (Equals Arthrodesmus bulnheimii in Smith, 1924). Found once in September, 1967 at Newfound Lake. Staurodesmus cuspidatus (De Brébisson) Teiling. (Equals Staurastrum cuspidatum in Smith, 1924). The most abundant species of Staurodesmus found at New- found Lake; present throughout the year. *Staurodesmus dickiei (Ralfs) Lillieroth. (Equals Stau- rastrum dickiei in Iréné-Marie, 1939). Only found a few times during May and June at Winnisquam Lake. *Staurodesmus extensus var. joshuae (Gutwinski) Teiling. (Equals Arthrodesmus incus var. extensus in Smith, 1924). Present sporadically throughout the year at both lakes. However, it was more abundant at Newfound Lake. *Staurodesmus incus var. ralfsii (W. West) Teiling. (Equals Arthrodesmus ralfsii in Smith, 1924). Common from February to July at both lakes, but most abundant when the water temperatures are low. *Staurodesmus mamillatus var. maximum (W. West) Teiling. (Equals Staurastrum cuspidatum var. canadense in Smith, 1924). Very common throughout the year at Newfound Lake; less abundant at Winnisquam Lake and only occurring from July to November. 462 Rhodora [Vol. 71 Staurodesmus megacanthus (Lund) Thunmark. (Equals Staurastrum megacanthum in Smith, 1924). Very com- mon throughout the year at Newfound Lake; less abun- dant at Winnisquam Lake and only occurring from April to November. **Staurodesmus subtriangularis (Borge) Teiling. (Equals Arthrodesmus triangularis var. subtriangularis in Smith 1924). Present sporadically throughout the year at both lakes. *Stylosphaeridium stipitatum (Bachmann) Geitler and Gimesi. Found commonly as an epiphyte on Gompho- sphaeria at Newfound Lake; collected from July to October. *Tetraedron limneticum Borge. Observed once in August, 1967 at Newfound Lake. *Ulothrix subconstricta G. S. West. Very common from June to November at Winnisquam Lake. *Ulothrix tenerrima Kuetzing. Common from April to May at Winnisquam Lake. *Ulothrix variabilis Kuetzing. Found occasionally from June to August at Winnisquam Lake. Xanthidium antilopaeum (De Brébisson) Kuetzing. Found a number of times during September to November at Winnisquam Lake. Xanthidium antilopaeum var. polymazum Nordstedt. Found once in October, 1966 at Newfound Lake. EUGLENOPHYCEAE **Colacium arbuscula Stein (Figure 16). Found attached to various zooplankton. Most common at Newfound Lake, where it was seen every month except February and March. It was seen from June to August at Winnisquam Lake. **Colacium vesiculosum Ehrenberg (Figure 17). Found throughout the year at both lakes. Epizooic on various Cladocera; more common than C. arbuscula. *Euglena spirogyra Ehrenberg. Found once (April) in 1969] Phytoplankton — Gruendling and Mathieson 465 shallow water at Winnisquam Lake — after a rainfall. Probably not a true planktonic species. XANTHOPHYCEAE **Gloeobotrys limneticus (G. M. Smith) Pascher (Figure 42). Common at Newfound Lake from August to November. Usually evident below ten meters. Less abundant at Winnisquam Lake where it was found from July to October. *Qphiocytium capitatum Wolle. Found only rarely at New- found Lake from September to November. *Stipitococcus capense Prescott. Found at a shallow sta- tion at Winnisquam Lake. Epiphytic on various fila- mentous algae, especially Oedogonium, and once on the lorica of Dinobryon divergens. Found from July to Sep- tember; rare. CHRYSOPHYCEAE *Chrysosphaerella longispina Lauterborn. Common at Newfound Lake from July to November; reaching maxi- mum numbers in July and August. Generally evident below 10 meters. Rarely found at Winnisquam Lake (September to November). **Diceras phaseolus Fott. Found sporadically throughout the summer at Newfound Lake; rare. *Dinobryon bavaricum Imhof (Figure 4). Very common throughout the year at both lakes. Maximum populations were evident in June and July at Newfound Lake. Stato- spores were also evident at the same time. The maximum quantities were usually found at 10 meters. The size of the lorica in the upper portion of the colony was usually longer (90-1004) than that in the lower portion (65-754). **Dinobryon crenulatum W. and G. S. West (Figures 35 & 36). Found sporadically throughout the year at New- found Lake; rare. *Dinobryon cylindricum Imhof (Figure 3). Common throughout the year at both lakes. At Newfound Lake the maximum numbers were found during June, July 1969] Phytoplankton — Gruendling and Mathieson 465 and November, while at Winnisquam Lake they were most abundant from April-June. The production of statospores coincided with the population peaks. *Dinobryon divergens Imhof. Less common than D. ba- varicum or D. cylindrum, Present throughout the year at Newfound Lake, but not seen from February to May at Winnisquam Lake. The lorica of D. divergens was often difficult to distinguish from that of D. cylindricum, Dinobryon sertularia Ehrenberg. Common throughout the year at both lakes; forming peak numbers in the fall. **Dinobryon suecicum Lemmermann (Figure 38). Found throughout the year at Newfound Lake. Often difficult to detect because of its very small size; rare. **Dinobryon vanhoeffenii (Krieger) Bachmann (Figure 37). Found from June to November at Newfound Lake. Its populations showed very rapid pulsations; maximum numbers were recorded in July and August. **Epipyxis tabellariae (Lemmermann) Smith (Figure 40). Found as an epiphyte on Tabellaria and Asterionella during the fall diatom pulse at Newfound Lake (October and November) and during the spring diatom bloom at Winnisquam Lake; uncommon. **Epipyxis utriculus Ehrenberg. A common epiphyte on Tabellaria at Newfound Lake from August to November. Mallomonas acaroides Perty. Only found during October and November at Newfound Lake; rare. **Mallomonas elegans Lemmermann. Abundant during the spring phytoplankton pulse (March to May) at Winni- Fig. 31. Peridinium willei Huitfeld-Kaas, X500. Fig. 32. Peridinium inconspicuum Lemmermann, 750. Fig. 33. Peridinium limbatum (Stokes) Lemmermann, > 500. Fig. 34. Glenodiniwm aciculiferum (Lemmermann) Lindem, X500. Fig. 35 & 36. Dinobryon crenulatum W. & G. S. West, x 500. Fig. 37. Dinobryon vanhoeffenii (Krieger) Bachmann, 500. Fig. 38. Dinobryon suecicum Lemmermann, 1500. Fig. 39. Rhodomanas lacustris Pascher & Ruttner, > 1000. Fig. 40. Epipixis tabellariae (Lemmermann) Smith, X500. Fig. 41. Synura adamsii G. M. Smith, X500. Fig. 42. Gloeobotrys limneticus (G. M. Smith) Pascher, 500. 466 Rhodora [Vol. 71 squam Lake, disappearing quickly after water tempera- tures increased. **Mallomonas fastigata Zacharias. The most common species of Mallomanas found at both lakes, Evident from June to November. **Mallomonas producta (Zacharias) Iwanoff. Relatively common at both lakes from June to November; found occasionally during the winter at Winnisquam Lake. **Mallomanas pseudocoronata Prescott. Common through- out the year at Newfound Lake. **Mallomonas tonsurata Teiling var. alpina (Pascher and Ruttner) Kuetzing. Common throughout the year at Newfound Lake. During the fall and the period of ice cover it occurred in maximum numbers near the surface, while during the summer it was most abundant at 10 and 15 meters. **Synura adamsii G. M. Smith (Figure 41). Found at Winnisquam Lake from May to November; rare. Synura uvella Ehrenberg. Very common throughout the year at both lakes. However, it was more abundant at Winnisquam Lake where it formed maximum numbers during the summer and fall. It was generally found below 5 meters. **Uroglenopsis americana (Calkins) Lemmermann (Figure 13). Common from June to November at Newfound Lake. Large pulses of populations developed suddenly in August, 1966 and July, 1967 and disappeared just as quickly. It was much less common at Winnisquam Lake and was only found in small quantities from May to August, 1967. BACILLARIOPHYCEAE Asterionella formosa Hass. Very common at both lakes. Large populations were evident in the spring (April) and fall (November) at Winnisquam Lake. Such a periodic cycle was less evident at Newfound Lake because the diatom was not as abundant. Large populations were recorded under ice cover. * Asterionella formosa var. gracillima (Hantz) Grun. Much 1969] Phytoplankton — Gruendling and Mathieson 467 less common than the typical species; only evident from February to early May. The cells were often in chain-like colonies, as illustrated by Huber-Pestalozzi (1942). *Cyclotella bodanica Eulenst. The most common species of Cyclotella found at Newfound Lake. Evident through- out the year and reaching maximum numbers during the summer. The maximum quantities were found at 10 meters and below. *Cyclotella compa (Ehrenberg) Kuetzing. Common throughout the year at both lakes; maximum numbers were evident during the spring and late fall between 5 and 10 meters. *Cyclotella glomerata Bachmann. Found at Newfound Lake when the water temperatures were quite low. Most abundant from February to early May, but even then it was uncommon. * Cyclotella stelligera Cleve and Grein. Found at Winni- squam Lake from September to November; rare. Cymbella ventricosa Kuetzing. Found at Winnisquam Lake during the period of the spring overturn (April and May). It is probably not a true planktonic species. *Fragilaria capucina Desm. Found sporadically at both lakes; uncommon. *Fragilaria crotonensis Kitton. Very common throughout the year at both lakes. However, it was more abundant at Winnisquam Lake where it formed maximum numbers during November, 1966 and July to August, 1967. The largest concentration of populations occurred between 3 and 5 meters. *Melosira ambigua (Grun.) Mueller. Abundant through- out the year at both lakes. The maximum populations were evident during the spring and fall at Newfound Lake, while during the fall at Winnisquam Lake. *Melosira distans (Ehrenberg) Kuetzing. Found through- out the year at Newfound Lake; always in very small quantities. *Melosira italica (Ehrenberg) Kuetzing. Very abundant throughout the year at Winnisquam Lake. Maximum 468 Rhodora [Vol. 71 numbers were found from February to May with the largest peak occurring at the time of the spring overturn. Lesser peaks occurred during the fall overturn (October and November). During the spring and fall overturn populations were evenly distributed throughout the water column, while throughout the rest of the year they steadily settled out of the photic zone. Such observa- tions are similar to those made by Lund (1954, 1955) upon Melosira italica (Ehrenberg) Kuetzing subsp. sub- artica O. F. Müller. *Melosira italica var. tenuissima (Grun.) Mueller (Figures 8 & 9). Very common at Winnisquam Lake throughout the year. The development of this variety was similar to the typical plant, with maximum development occur- ring in the spring. Auxospores were observed during April, and they were present until early June, when many appeared to be germinating. *Meridion circulare var. constrictum (Ralfs) V. H. Present in small numbers during the spring overturn at Winnisquam Lake. Probably not a true planktonic species. *Rhizosolenia eriensis H. L. Smith. Very common at New- found Lake during July and August. Its maximum abundance was evident below 10 meters. It was sporadic- ally present at Winnisquam Lake; being most conspicuous during the winter and spring. The morphological varia- tion of the plant is extremely great. *Synedra ulna (Nitz.) Ehrenberg. Common throughout the year at both lakes. However, it was more abundant at Winnisquam Lake, where it reached its maximum peaks during May and June. "Tabellaria fenestrata (Lyngbye) Kuetzing. Common throughout the year at both lakes; but most common at Winnisquam Lake. Tending to reach maximum peaks after all other diatoms in the spring, and before all others in the fall. *Tabellaria flocculosa (Koth) Kuetzing. Common through- 1969] Phytoplankton — Gruendling and Mathieson 469 out the year at Newfound Lake. Only occasionally observed at Winnisquam Lake from April to June. DINOPHYCEAE Ceratium hirundinella (O. F. Mueller) Dujardin. Common at both lakes from April to November. All specimens were of the form robustum as described by Huber- Pestalozzi (1950). **@lendodinium aciculiferum (Lemmermann) Linden. (Figure 34). A few specimens were observed in each col- lection made at Winnisquam Lake from March to early June; rare. The alga has the characteristic three spines on the hypovalve. *Gonyaulax palustris Lemmermann. Common at New- found Lake from February to early June. *Gymnodinium fuscum (Ehrenberg) Stein. Present spo- radically throughout the year at Newfound Lake. *Gymnodinium palustre Schilling. Found at Newfound Lake in February, March and April; rare. *Peridinium cinctum (O. F. Mueller) Ehrenberg. Found sporadically throughout the year at both lakes; never in any great abundance. **Peridinium inconspicuum Lemmermann (Figure 32). Common at Newfound Lake from June to October; maxi- mum numbers were evident during August. Most of the populations were restricted to the 10 meter collections. **Peridinium limbatum (Stokes) Lemmermann (Figure 33). The dominant dinoflagellate at Newfound Lake. Found throughout the year, but reaching maximum num- bers in the fall. It was found in very small quantities throughout the year at Winnisquam Lake, and was evenly distributed throughout the water column. **Peridinium willei Huitfeld-Kaas (Figure 31). Common throughout the year at Newfound Lake. Maximum num- bers were recorded during the winter. *Peridinium wisconsinense Eddy. Common from July to October at Newfound Lake. Maximum numbers were evident during September and October. 470 Rhodora [Vol. 71 CRYPTOPHYCEAE *Cryptomonas erosa Ehrenberg. Common throughout the year at Newfound Lake. *Cryptomonas erosa var. reflexa Marsson. A few speci- mens of this variety were found from February to early June at Newfound Lake. *Cryptomonas marssonii Skuja. Found sporadically from July to November at both lakes. Cryptomonas ovata Ehrenberg. Very common throughout the year at both lakes, but more abundant at Winnisquam Lake. The maximum populations were reached in the spring and fall. **Rhodomonas lacustris Pascher and Ruttner (Figure 39). Abundant throughout the year at both lakes; with peaks in the spring and fall. Very large numbers were found at Winnisquam Lake in April and May. SUMMARY A total of 185 taxa of fresh algae (primarily phytoplank- ton) were identified at Winnisquam and Newfound Lakes, New Hampshire. Of these, 149 are presumed to be newly published records for New Hampshire and 45 are presumed to be new records for New England. Of the 142 taxa recorded at Winnisquam Lake, 85 were found exclusively at this location. The green algae con- tributed the greatest number of species, but the bulk of the standing crop was composed of blue-green algae. During the summer the phytoplankton flora was primarily composed of members of the Chlorophyceae and Cyanophyceae. Mem- bers of the Bacillariophyceae, Chrysophyceae and Chloro- phyceae (mainly desmids) contributed the largest number of species during the spring and fall Diatoms were the major component of the winter flora. A total of 100 taxa were recorded at Newfound Lake, and 43 of these were found exclusively at Newfound Lake. The largest number of species were green algae. During the summer and fall the phytoplankton was primarily com- posed of members of the Cyanophyceae, Chlorophyceae 1969] Phytoplankton — Gruendling and Mathieson 471 (especially desmids) and Chrysophyceae. The diatoms and the golden-brown algae contributed the largest number of species during the winter and spring. ACKNOWLEDGEMENTS The authors wish to express their gratitude to the fol- lowing people: Dr. H. T. Croasdale of Dartmouth College for aid in the identification of desmids and review of the manuscript; Dr. Ruth Patrick for the aid in the identifica- tion of Melosira species; and Mr. G. L. Byers, Director of the New Hampshire Water Resources Research Center. The work upon which this publication is based was sup- ported in part by funds provided by the United States Department of the Interior Office of Water Resources Research, as authorized under the Water Resources Act of 1964, Public Act 88-379, through the Water Resource Research Center of the University of New Hampshire. DEPARTMENT OF BOTANY, UNIVERSITY OF NEW HAMPSHIRE, DURHAM. Present address: DEPARTMENT OF BIOLOGICAL SCIENCES STATE UNIVERSITY COLLEGE PLATTSBURGH, NEW YORK DEPARTMENT OF BOTANY, UNIVERSITY OF NEW HAMPSHIRE, DURHAM 03824 REFERENCES TO NEW ENGLAND FRESHWATER ALGAE AHLSTROM, E. H. 1937. Studies on variability in the genus Dino- bryon (Mastigophora). Trans. Amer. Microsc. Soc. 56: 139-159. AUYANG, T. S. 1962. A survey of the algae of Lake Quinsigamond. Rhodora 64: 49-59. BAILEY, J. W. 1847-1848. Notes on the algae of the United States. Amer. Jour. Sci. and Arts. Ser. 2. 3: 80-85, 399-408; 6: 37-42. BENNETT, J. L. 1888. Plants of Rhode Island. Proc. Providence Franklin Soe. Providence Press Co., Providence, R. I. 128 pp. BENOIT, R. J. and J. J. Curry. 1961. Algae blooms in Lake Zoar, 472 Rhodora [Vol. 71 Connecticut. Algae and Metropolitan Wastes, U. S. Public Health Service, SEC TR W61-3: 18-22. BILLINGS, M. P. 1956. The geology of New Hampshire Part II — Bedrock Geology. New Hampshire State Planning and Develop- ment Comm., Concord, N. H. 203 pp. CALKINS, G. N. 1892. On Uroglena, a genus of colony-building in- fusoria observed in certain water supplies in Massachusetts. Ann. Rept. Massachusetts State Bd. of Health 23: 647-657. 1901. Seasonal distribution of microscopic organ- isms in surface water. Ann. Rept. Massachusetts State Bd. of Health. CHASE, E. S. 1947. Experience with Anabaena in Kenoza Lake, Haverhill, Massachusetts. Jour. New Engl. Water Wks Assn. 61: 128-134 CoLE, R. M. 1939. Studies on the plankton of the Connecticut watershed. Im: Biological Survey of the Conn. Watershed. N. H. Fish and Game Biol. Surv. Rep. 4: 190-194. CoLLINS, F. S. 1904. Algae of the Flume. Rhodora 6: 229-231. 1905. Phycological notes of the late Isaac Holden. Rhodora 7: 168-172. 1909. The green algae of North America. Tufts College Studies, Sci. Ser. 2: 79-480 (Reprinted and repaged by G. E. Stechert, New York, 1928). 1912. The green Algae of North America. Supple- mentary Paper. Ibid. 3: 69-109 (Reprinted and repaged by G. E. Stechert, New York, 1928). 1916. Notes from the Woods Hole Laboratory. 1915. Rhodora 18: 90-92. 1918a. Notes from the Woods Hole Laboratory. 1917. Rhodora 20: 141-145. 1918b. The green algae of North America. 2nd Supplementary Paper. Tufts College Studies, Sci. Ser. 4: 1-106 (Reprinted and repaged by G. E. Stechert, New York, 1928.) Conn, H. W. 1905. A preliminary report of the protoza of the fresh water of Connecticut. Conn. Geol. and Nat. Hist. Surv, Bull. No. 2, 69 pp. and L. W. WEBSTER. 1908. A preliminary report of the algae of the freshwaters of Connecticut. Conn. Geol and Nat. Hist. Surv. Bull, No. 10, 78 pp. CooPER, G. P. 1939. A biological survey of thirty-one lakes and ponds of the Upper Saco River and Sebago Lake drainage sys- tems in Maine. Maine Dept. of Inland Fisheries and Game. Fish Surv. Report, No. 2. and J. L. Fuller. 1945. A biological survey of I 1969] Phytoplankton — Gruendling and Mathieson 473 Moosehead Lake and Haymock Lake, Maine. Maine Dept. of In- land Fisheries and Game. Fish Surv. Rept. No. 6, 160 pp. CowLES, R. P. and C. E. BRAMBEL. 1936. A study of the environ- mental conditions in a bog pond with special reference to the diurnal vertical distribution of Gonyostomum semen. Biol. Bull. 71: 286-298. CROASDALE, H. T. 1935. The freshwater algae of Woods Hole, Massachusetts. Philadelphia. 134 pp. 1948. The flora of Penikese, seventy-four years after. II. Fresh and brackish water algae of Penikese Island, Rhodora 50: 269-279. CUSHMAN, J. A. 1908a. List of desmids found in Carver’s Pond, Bridgewater, Massachusetts. Rhodora 5: 79-81. 1903b. Notes on New England desmids I. Rho- dora 5: 221-225. 1903c. Notes on New England desmids II. Rho- dora 5: 252-255. 1904a. Notes on Micrasterias from southeastern Massachusetts. Bull. Torrey Bot. Club 31: 393-397. 1904b. Division in desmids under pathological conditions. Rhodora 6: 234. 1905a. A contribution to the desmid flora of New Hampshire. Rhodora 7: 111-119. 1905b. A contribution to the desmid flora of New Hampshire. Rhodora 7: 251-266. 1905c. Notes on the zygospores of certain New England desmids with description of a few new forms. Bull. Torrey Bot. Club. 32: 223-229. 1905d. The desmid flora of Nantucket. Bull. Torrey Bot. Club 32: 549-563. 1906. New England desmids of the subfamily Saccodermae. Bull. Torrey Bot. Club. 33: 343-351. 1907a. A synopsis of the New England species of Pleurotaeniwm. Rhodora 9: 101-106. 1907b. A synopsis of the New England species of Tetmemorus. Bull. Torrey Bot. Club. 34: 599-601. 1908a. New England species of Closterium. Bull. Torrey Bot. Club 35: 109-133. 1908b. A synopsis of the New England species of Micrasterias. Rhodora 10: 97-111. DRouET, F. 1937. Three American Oscillatoriaceae. Rhodora 39: 277-280. 1988. Some Myxophyceae from Nantucket Island, Massachusetts. Rhodora 40: 74-76. 1938. The Oscillatoriaceae of southern Massachusetts. Rhodora 40: 221-241; 255-273. 474 Rhodora [Vol. 71 and A. COHEN. 1935. The morphology of Gonyostomum semen from Woods Hole, Massachusetts. Biol. Bull. 68: 422-439. Eppy, S. 1930. The fresh-water armored or thecate dinoflagellates. Trans. Amer. Microse. Soc. 49: 277-321. EDMUNDSON, W. T. 1938. Notes on the plankton of some lakes in the Merrimack watershed. /»: Biological survey of the Merri- mack watershed. N.H. Fish and Game Biol. Surv. Rept. 3: 207-210. and J. L. FULLER. 1937. Food conditions in some New Hampshire lakes. In: Biological Survey of the Androscoggin, Saco, and Coastal watersheds. N. H. Fish and Game Biol. Surv. Rept. 2: 95-99. FARLOw, W. G. 1877. Remarks on some algae found in the water supplies of the city of Boston. Bull. Bussey Institution 2: 75-80. FLINT, L. H. 1917. The algae of Vermont. Vermont Bot. and Bird Club, Joint Bull. 3: 12-18. FoRBES, F. F. 1890. Study of algae growth in reservoirs and ponds. Jour. New Eng. Water Wks. Assoc. 4: 196-210. GoopwIn, H. A. 1943. Plankton of the lakes and ponds. In: A biological survey of lakes and ponds of the central coastal area of Maine. Maine Dept. of Inland Fisheries and Game. Fish Surv. Rept. 5: 56-66. GRUENDLING, G. K. and A. C. MATHIESON. Phytoplankton popula- tions in relation to trophic levels of lakes in New Hampshire, U.S.A. Nova Hedwigia (in press). GUSTAFSON, A. H. 1942. Notes on some freshwater algae from New England. Rhodora 44: 64-69. 1952. The role of algae in some Massachusetts lakes. Fisheries Report for lakes of northeastern Massachusetts — 1949. Div. Fish and Game. Bur. Wildlife Res. and Manage- ment, Commonwealth of Massachusetts. 3-13. HARVEY, F. L. 1888. The freshwater algae of Maine I. Bull. Torrey Bot. Club. 15: 155-161. 1889. The freshwater algae of Maine II. Bull. Torrey Bot. Club. 16: 181-188. 1892. The freshwater algae of Maine III. Bull. Torrey Bot. Club. 19: 118-125. HasTiNGS, W. N. 1892. New desmids from New Hampshire. Amer. Monthly Microsc. Jour. 13: 153-155. Horus, F. S. 1901. On the distribution of growths in surface water supplies and on the method of collecting samples for examinations. Trans. Amer. Microsc. Soc. 22: 49-62. HUBER-PESTALOZZI, G. 1938-1961. Das Phytoplankton des Süss- wassers. Vol. 16 Pt. 1-5. In: Thienemann, A. (ed.) Die Bin- nengewasser. E. Schweizerbart'sche Verlagsbuchhandlung, Stutt- gart. 1969] Phytoplankton — Gruendling and Mathieson 475 HuTCHINSON, G. E. 1944. Limnological studies in Connecticut VII. A crictical examination of the supposed relationship be- tween phytoplankton periodicity and chemical changes in lake water. Ecology 25: 3-26. HYLANDER, C. J. 1922. Preliminary report of the desmids of Con- necticut. Rhodora 24: 213-224. 1924. Supplementary report on the desmids of Connecticut. Rhodora 26: 203-210. 1928. The algae of Connecticut. Conn. Geol. and Nat. Hist. Surv. Bull. 42. Hartford. 245 pp. IRENE-MarIE. 1938. Flore Desmidiale de la region de Montreal. Laprairie, Canada. 547 pp. Jao, C. 1934. Oedogonium in the vicinity of Woods Hole, Massa- chusetts. Rhodora 36: 197-214. 1936. Notes on Oedogonium and Bulbochaete in the vicinity of Woods Hole, Massachusetts. Rhodora 38: 67-73. KENNEDY, A. S. 1948. The algal environment in relation to fish. Jour. New Eng. Water Wks. Assn. 62: 196-201. LAMBERT, F. D. 1909. Two new species of Characium. Rhodora 11: 65-74. Lewis, I. F. 1924. The flora of Penikese, fifty years after. Rhodora 26: 181-195; 211-220. and W. R. TAYLOR. 1921. Notes from the Woods Hole Laboratory, 1921. Rhodora 23: 249-256. 1929. Notes from the Woods Hole Laboratory, 1928. Rhodora 30: 193-198. LUND, J. W. G. 1954. The seasonal cycle of the plankton diatom Melosira italica (Ehr.) Kutz., subsp. subarctica O. Mull. Jour. Ecol. 42: 151-179. 1955. Further observations on the seasonal cycle of Melosira italica (Ehr.) Kutz., subsp. subarctica O. Mull. Jour. Ecol. 43: 91-102. MooRE, G. T. 1899. The pollution of water supplies by algae. Rhodora 1: 100-102. PALMER, C. M. 1942. Additional records for algae, including some of the less common forms. Butler Univ. Bot. Stud. 5: 224-234. 1958. Algae and other interference organisms in New England Water Supplies. Jour. New Eng. Water Wks. Assn. 72: 27-46. PARKER, H. N. 1911. Note on a growth of Synura in Lake Co- chituate, Mass. Trans. Amer. Microsc. Soc. 30: 141-144. PATRICK, R. 1943. The diatoms of Linsley Pond, Connecticut. Proc. Acad. Nat. Sci. Phila. 95: 53-110. and C. W. REIMER. 1966. The diatoms of the United States exclusive of Alaska and Hawaii. Vol. 1. Monograph No. 13, Acad. Nat. Sci. Phila., Philadephia. 688 pp. 476 Rhodora [Vol. 71 PoULToN, E. M. 1930. Further studies on the Heterokontae: Some Heteronkontae of New England, U.S.A. New Phytologist 29: 1-26. PRESCoTT, G. W. 1935. Notes on the desmid flora of New England I. The genus Euastrum in Massachusetts. Rhodora 37: 22-29. 1935. Notes on the desmid flora of New England II. Desmids from Cape Cod and the Elizabeth Islands. Rhodora 37: 113-121. 1962. Algae of the western great lakes areas. Rev. Ed. Wm. C. Brown., Publishers, Dubuque. 977 pp. and H. T. CROASDALE. 1937. New or noteworthy freshwater algae of Massachusetts. Trans. Amer. Microsc. Soc. 56: 269-282. 1942. The algae of New England II. Additions to the freshwater algal flora of Massa- chusetts. Amer. Midl. Nat. 27: 662-676. RILEY, G. A. 1940. Limnological studies in Connecticut III. The plankton of Linsley Pond. Ecol. Monogr. 10: 279-305. SETCHELL, W. A. 1895. Notes on some Cyanophyceae of New England. Bull. Torrey Bot, Club 22: 424-431. SMITH, G. M. 1920. Phytoplankton of the inland lakes of Wisconsin. Part I. Wisconsin Geol. Nat. Hist. Surv. Bull. No. 57. Madison. 243 pp. 1924. Phytoplankton of the inland lakes of Wiscon- sin. Part II. Wisconsin Geol. Nat. Hist. Surv, Bull No. 57. Madison. 227 pp. 1950. The fresh-water algae of the United States. 2nd ed. McGraw-Hill Book Co., Inc, New York. 719 pp. SPENCER, J. L. 1950. The net phytoplankton of Quabbin Reservoir, Massachusetts, in relation to certain environmental factors. Ecology 31: 405-425. STONE, G. E. 1900. Flora of Lake Quinsigamond. F. S. Blan- chard and Co., Printers, Worcester, Mass. 12 pp. TAYLOR, W. R. 1921. Additions to the flora of Mt. Desert, Maine. Rhodora 23: 65-68. TEILING, E. 1967. The desmid genus Staurodesmus: A taxonomic study. Arkiv för Botanik 6: 467-629. TERRY, W. A. 1907. A partial list of Connecticut diatoms with some account of their distribution in certain parts of the state. Rhodora 9: 125-140. UNGER, W. B. 1941. A preliminary survey of the protozoa of Beaver Lake near Salsbury Cove, Maine. Bull. Mt. Desert Island Biol. Lab. 17-18. WEBBER, E. E. 1961. A list of algae from selected areas in Massa- chusetts. Rhodora 63: 275-281. 1969] Phytoplankton — Gruendling and Mathieson 477 1964. A rare blue-green alga from Massachusetts. Rhodora 66: 163-164. WEsT, G. S. and W. West. 1904-1912. A monograph of the British Desmidiaceae. Vol. 1-4. The Ray Soc., London. West, G. S., W. WEST, and N. CARTER. 1923. A monograph of the British Desmidiaceae. Vol. 5. The Ray Soc., London. West, W. 1888. The desmids of Maine. Jour. Bot. 26: 339-340. 1889. List of desmids from Massachusetts. Jour. Roy. Microse. Soc. 9: 16-22. 1889. The freshwater algae of Maine. I. Jour. Bot. 27: 205-307. 1891. The freshwater algae of Maine II. Jour. Bot. 29: 353-357. and G. S, West. 1895. New American algae. Jour. Bot. 83: 52. WHELDEN, R. M. 1939. Notes on New England Algae I. Cyclo- nexis and Actidesmium. Rhodora 41: 133-137. 1942. Notes on New England Algae II. Some interesting New Hampshire algae. Rhodora 44: 175-187. 1943. Notes on New England Algae III. Some interesting algae from Maine. Farlowia 1: 9-23. WHIPPLE, G. C. 1894. Synura. Amer. Monthly Microsc. Jour. 15: 257-262. 1897. Biological studies in Massachusetts. No. 2. Amer. Nat. 31: 576-581. and H. N. PARKER. 1899. Note on the vertical dis- tribution of Mallomonas. Amer. Nat. 33: 485-491. WILLE, N. 1899. New forms of green algae. Rhodora 1: 149-150. WOLLE, F. 1877. Freshwater Algae Il. Bull. Torrey Bot. Club 6: 137-141. 1877. Freshwater Algae III. Bull. Torrey Bot. Club 6: 181-189. 1880. Freshwater Algae IV. Bull. Torrey Bot. Club 7: 43-48. Woop, H. C. 1869. Desmids from the White Mountains. Proc. Acad. Nat. Sci. Phila. 1869: 15-19. 1872. A contribution to the history of the fresh- water algae of North America. Smithsonian Institute Contribu- tion to Knowledge 19: 1-262. WRIGHT, R. T. 1964. Dynamics of a phytoplankton community in an ice-covered lake. Limnol. Oceangr. 9: 163-178. 478 Rhodora [Vol. 71 MAGNOLIA TRIPETALA L. AND ARALIA SPINOSA L. IN ST. LOUIS COUNTY, MISSOURI: Steyermark, in his monumental Flora of Missouri (lowa State University Press, Ames, 1963), discusses at length the possibility of the natural occurrence of Magnolia tripetala L. in Missouri, and presents convincing arguments that previous ascrip- tions of this species to Missouri are based on mislabelled specimens (p. 670). On 7 June 1968, while collecting in a second-growth deciduous woods of about 100 acres in Creve Coeur, St. Louis County, Missouri, I found a single tree with multiple large trunks from the base which is un- doubtedly M. tripetala. The tree is growing deep in the woods, and gives every appearance of being native — I have located no cultivated specimens in the adjacent city area, and the trunks are 20 to 30 feet high, indicating that the tree could scarcely have been transplanted to such a site; moreover, houses have only recently been built in this area, and the woods remain somewhat isolated and remote from major disturbance. The stand of trees, known locally as the Tegethoff property, is bounded grossly by I-244 on the east, Mason Road on the west, Halsgame Lane on the north, and Hubler Road on the south. Three specimens of leaves, “cones”, and twigs are deposited as vouchers in the her- barium of Wisconsin State University-Oshkosh (Harriman 3071). In the same area, about 100 yards south of the Magnolia tree, there occurs a large grove of Aralia spinosa L. Steyer- mark reports that this tree occurs generally through south- eastern Missouri (p. 1112), but admits the plant to the native flora of St. Louis County only doubtfully. Again, I was unable to find any cultivated specimens of this tree in the vicinity, and the area in the woods where the tree occurs is also quite remote from roads, paths, and houses; it therefore appears definitely to be native to St. Louis County, its northernmost station in Missouri. A specimen of one large leaf, mounted on three sheets, is deposited 1969] Eragrostis — Lakela 479 in the herbarium, Wisconsin State University-Oshkosh (Harriman 3085). NEIL A. HARRIMAN BIOLOGY DEPARTMENT, WISCONSIN STATE UNIVERSITY-OSH KOSH OSHKOSH 54901 ERAGROSTIS DOMINGENSIS (PERS.) STEUD. NEW TO THE UNITED STATES: This grass was encountered during a collecting trip to the upper Florida Keys in the late summer of 1968. The discovery was timely for its inclusion in the “Subtropical Flora of Florida.” Perennial by short rhizomes. Culms 0.6-1.5 m tall, erect, rigid, relatively stout, 3 mm in diameter at lower internodes. Basal leaves few per culm; sheaths of previous years 5-8 mm wide at base, persistent, silvery on adaxial surface. Nodes purplish-brown. Culm leaves 3 or more with overlapping, pale green prominently veined sheaths, pilose at summit; ligule a fringe of hairs less than 0.5 mm long; the base of blade banded within, with silky puberulence; blades 2-7 mm wide, flat, linear, long-attenuate, involute at tips; midrib white, veins on adaxial surface scaberulous. Panicle 3-5.5 dm long, strict; the lower 1 or 2 racemes solitary, distant; the upper, fas- cicled; rachis and raceme branches wiry, glabrous. Spikelets short- pedicelled, 7-12(18)-flowered, glabrous; glumes 1 mm long, acute; lemmas 1.5-1.8 mm long, ovate-acute, with hyaline margins, deciduous with the fruit; paleas persistent, nearly as long as the lemmas, scaberulous on the veins. Fruit amber, 0.6-0.8 mm long, oblong in outline, slightly contracted to apex, truncate-oblique at base; the body patently longitudinally striate. The grass was discovered on Plantation Key, Monroe County, Florida. Collection L31625 was made August 27, 1968, from a small colony in a transition zone of dune sand and grassy berm along the old road, east side of US £1. Its striet habit, bony hardness of the culm, and narrow, elongate panicle are outstanding differentiating characters. (Contribution no. 43 from the Botanical Laboratories, University of South Florida, Tampa. 480 Rhodora [Vol. 71 The author is indebted to Dr. LeRoy H. Harvey who, during a recent visit to the University of South Florida Herbarium, named the plant by sight in a stack of Era- grostis collections. Known to occur on beaches, rocky soil near the coast, Cuba, Jamaica, Hispaniola, Leeward and Windward Is- lands, and, according to Dr. Harvey, Veracruz, Mexico. OLGA LAKELA UNIVERSITY OF SOUTH FLORIDA TAMPA 33620 AN INCORRECT CORRECTION, In a recent Rhodora article the use of connecting vowels in epithets of Latin origin was examined. Penstemon tubaeflorus was corrected to Penstemon tubiflorus. Recommendation 73G in the Inter- national Code of Botanical Nomenclature, however, clearly states that irregular forms can be used to reveal etymolog- ical distinctions and that when such irregularities occur in the original spelling of existing compounds, they should be retained. Thomas Nuttall intentionally used the ae spelling to indicate that the flowers were shaped like a horn (tuba) rather than like a waterpipe (tubis). It is truly a happy occasion when one discovers that his botanical predecessors were not altogether as illiterate as he has sometimes sup- posed. FRANK S, CROSSWHITE DEPARTMENT OF BOTANY UNIVERSITY OF WISCONSIN, MADISON 53706 A RE-EVALUATION OF CUPHEA, SECT. LEPTOCALYX, SERIES BUSTAMANTAE SHIRLEY A. GRAHAM In the course of studying the collections of Cuphea from Mexico deposited in the Natural History Museum in Paris, type material was found representing four of the five species of Sect. Leptocalyx, the third series. With the loan of additional types from Kew and Halle herbaria it was possible to compare authentic material of all species in this poorly known group. The species involved were C. busta- manta, C. calaminthifolia, C. corniculata, C. debilis, and C. imberbis. The latter was regarded by its author as a member of Sect. Heterodon, but as will be evident from the morphology, it clearly belongs to Sect. Leptocalyx. The first named species, Cuphea bustamanta, is fre- quently collected in pine-oak woods in Mexico and is easily recognized. It is fully described here for the first time and the position of an intraspecific taxon is re-evaluated. The remaining species are known only from type material and a few additional collections. Distinctions between them as based on the protologues are minor and examination of the types confirms the suspicion that C. calaminthifolia, C. corniculata, C. debilis, and C. imberbis represent a single taxon. They are united under the earliest epithet, cala- minthifolia. The series, established by Koehne (Fl. Bras. 13(2) : 236. 1877) was numbered but unnamed. Hence, the name Busta- mantae is proposed, based on the type species of the series. The series is a particularly distinctive one within the genus, immediately recognizable by its slender trailing habit, transversely oriented (at right angles to the stem), 2-armed hairs, and 9-staminate flowers. Because Koehne’s description of the series is very brief and fails to mention a number of prominent characters which the species share, including the characteristic transverse malphigaceous-like hairs, an expanded diagnosis is given below. 481 482 Rhodora [Vol. 71 Cuphea, Sect. Leptocalyx, Series Bustamantae S. A. Gra- ham, ser. nov. Type: C. bustamanta Lex. Series 3 Koehne, Fl. Bras. 13(2) : 236. 1877. Herbae graciles repentes, verisimiliter perennes, ubique secus caules prostratos radicantes. Caules, petioli, et pedi- celli pilis transversalibus malpighiaceis praeditis. Folia membranacea, ovato-elliptica apice acuto ad acuminato. Inflorescentia racemosa, foliata, flores paucis, Calyx gra- cilis, intus glaber, 13-30 mm longus, calcare descendente munitus ; appendices lobos aequantes ad superantes multum. Petala 6, unequales, generaliter purpurea. Stamina 9. Ovula (8-)12-18(-23). Discus deflexus in calcare, saepe teretiusculus. KEY TO THE SPECIES Calyx bright red with a greenish white area encircling the mouth; lobes of the calyx small, + equal in size; appendages 2-4 times the length of the lobes; dorsal petals 3-5 mm long, purple-black (rarely yellowish green); ventral petals 1-1.5 mm long, yellowish green, early deciduous. Michoacán to Oaxaca. ............ ess 1. C. bustamanta Calyx straw-colored to partly or wholly pale reddish purple; dorsal lobe of the calyx distinctly larger than the other five; appendages slightly less than to 3 times the length of the ventral lobes of the calyx; dorsal petals 5-12 mm long, purple; ventral petals 1.5-4 mm long, purple or white. Southern San Luis Potosi to Veracruz, Jalisco, and Michoacán. ........................ a... 2. C. calaminthifolia 1. Cuphea bustamanta Lex. in LaLlave & Lex., Nov. Veg. Descr. 1: 21. 1824. (Fig. la, b.) Type: MICHOACAN: Habitat prope Vallisoletum [Morelia], LaLlave & Lexarza s.n. (Holotype, there is no specimen which can be considered an authentic type at G and ap- parently all original material is lost; Neotype, 29 mi E of Morelia on Mex. 15, 2585 m, 6 Aug 1962, S. A. Graham 154 (MICH!) This specimen compares favor- ably with the protologue and was collected in the same general region as the missing holotype.) Cuphea reipublicae Rob. & Seat. Proc. Am. Acad. 28: 106. 1893. Type: MICHOACAN: Near Patzcuaro, 28 Jul 1892, C. G. Pringle 4142 (Holotype, GH; iso- types, P! US! VT). The type number was incorrectly cited in the protologue as Oct 1892, Pringle 4112. 1969] Cuphea — Graham 485 fa 1 Nf VISUM US L£ oN eee aa WwW 7 t imme = ASO S) x — . J / FIG. Ib = k <= | J | NN = | FG. la A / SV | —V^ i ATE Í AN J < ELEA] £ > t-Y V A 7 y Ld NU RNG E ae J m A ~~ "2 G ^ p £ | | C. BUSTAMANTA À C. CALAMINTHIFOLIA @ FIG. 3 1b, flower X114, (Graham 154). Fig. 2. C. ealaminthifolia Schlectd., flower X114, (Anderson & Laskowski 3819). Fig. 3. Distribution of species in Mexico, each symbol representing one or more collections at that locality. 484 Rhodora [Vol. 71 According; to Davis (H. B. Davis, Life and Works of Cyrus Guernsey Pringle, p. 100) and the Pringle Herbarium (pers. comm.), Pringle 4112 is Euphorbia torrida DC., collected 17 Jun 1892, at Rascon, San Luis Potosi and Pringle 4142 is C. reipublicae Rob. & Seat., collected 28 Jul 1892 by brooks near Patz- cuaro. All type material including the specimen pre- sumably studied by Robinson and Seaton at GH bears the same printed label numbered 4142. Although Pringle also was in Patzcuaro in October, his collec- tion numbers there were mainly 4200’s and 4300’s and no mention is made in his diary of recollecting no. 4142, nor are any specimens known to bear the data cited in the protologue. Cuphea bustamanta forma reipublicae (Rob. & Seat.) Koehne, Bot. Jahrb. 23 (Beibl. 57): 29. 1897. Parsonia bustamanta (Lex.) Standl. Contr. U.S. Natl. Herb, 23: 1021. 1924. Herbs, slender, trailing, probably perennial, with fibrous, adventitious roots occurring along the length of procum- bent stems up to 1 m long; stems bearing erect to decum- bent branches up to 40 cm tall, covered with dense, minute, two-armed, colorless, transversely oriented hairs and a row of longer, non-branched hairs. Leaves ovate to elliptic, the blades 20-50 mm long, 10-28 mm wide, the apex acumi- nate, the base obtuse to generally attenuate; petioles 5-15 mm long, bearing transverse two-armed hairs; blades thin, dark green above, pale beneath, both surfaces bearing minute, stiff hairs mixed with longer, weak hairs, the long hairs on the lower surface sometimes confined to the mar- gins of the leaf; leaves scarcely diminishing in size toward the apex of the stem. Inflorescence a leafy, few-flowered raceme; flowers 1 per node, interaxillary on bibracteolate pedicels 4-13 mm long; bracteoles linear, green, 1-3 mm long. Calyx slender, nearly straight, 22-30 mm long including a downward curving, basally rounded spur 2-3 mm long, bright red except for a greenish white area encircling the mouth of the calyx; externally bearing short, stiff, colorless, 1969] Cuphea — Graham 485 distally oriented hairs mixed with scattered, longer, color- less, weak hairs, the latter most concentrated at the proxi- mal end of the calyx; internally glabrous, neither bialate nor vesiculate; mouth of calyx slightly flaring, oblique with the dorsal side appearing shorter; neck of the calyx not contracted in fruit. Lobes of the calyx small, exceeded by linear, green appendages, 2-4 times the length of the lobes. Petals 6, the two dorsal ones obovate, generally purple-black, 3-5 mm long including a claw 1 mm long, (rarely yellowish green, then 1-2 mm long), subtended at the base by a flattened, yellow scale, free along one edge and included within the calyx; ventral petals yellowish green, obovate, to oblong, ca. 1 mm long, early deciduous. Stamens 9, mostly exserted, glabrous, the filaments and anthers deep purple. Pollen oblate, distinctly triangular in polar view; tricolporate, the pores equatorially arranged, equidistant, not protruding; syncolpate; exine tectate, finely and uni- formly striate over the entire surface of grain, striae ex- tending to the poles; diameter 37 u. Gynoecium with punc- tate to capitate sigma and exserted, glabrous style; ovules 12-15 (-23 fide Koehne) ; seeds lenticular, orbicular in out- line, 2 mm in diameter. Disc horizontal in spur, 1.5 mm long, thick, nearly terete, slightly tapering to a blunt apex. n = 12. Growing generally on steep, moist banks and along streams in pine-fir-oak forests at elevations from 1800-3600 m. Known from Michoacan to Oaxaca (Fig. 3). According to LaLlave and Lexarza, flowering throughout the year. The species may be recognized by the long, slender, red calyx with linear, green appendages which markedly exceed the calyx lobes, its 9-staminate condition and the procum- bent stems and branches bearing transverse two-branched hairs. Although Koehne, in his monograph of Cuphea (1903, p. 171) cites C. reipublicae Rob. & Seat. as C. bustamanta var. reipublicae (Rob. & Seat.) Koehne, he actually made the combination under the rank forma. The type of the form differs from typical C. bustamanta only in its yel- 486 Rhodora [Vol. 71 lowish green, rather than purple dorsal petals. Other character differences cited by Koehne were the 1 m height of the plant and the “somewhat broader" leaves lacking long (hispid) hairs in the middle of the ventral leaf surface. Examination of two isotypes reveals that the “height” of 1 m was actually the length of the procumbent, adventiously rooted main stem, the height of the erect branches being 15-40 em, which is typical of the species. The width of the leaves falls within that of typical C. bustamanta, and the arrangement of long hairs on the leaf is so variable on specimens examined that this is eliminated as a distin- guishing character of the form. In keeping with my taxo- nomic treatment of variability in other species of Cuphea, I choose not to recognize, with formal taxonomic rank, geographically isolated collections such as these which demonstrate but a single, minor character difference. Additional specimens examined: D.F.: Km. 61, road from Mexico City to Cuernavaca, 4600 m [incorrect?], 11 Aug 1929, Y. Mexia 2714 (MICH, MO). GUERRERO: Pie de la Cuesta, Toro Muerto, 2900 m, 17 Dee 1937, G. B. Hinton 11084 (US); Galeana, Teotepee, 3000 m, 25 Dec 1937, Hinton 11129 (Mo, Us); Montes de Oca, San Antonio-Buenos Aires, 22 Apr 1938, Hinton 14041 (MICH, MO, US); Galeana, Teotepec, 2600 m, 18 Mai 1939, Hinton 14275 (Mo, US); Mina, Petlacala, Barranca del Ranchito, 1920 m, 1 Jan 1938, Mexia 9064 (us); 1960 m, 8 Jan 1938, Mexia 9093 (Mo, US); Omiltemi, 20 km W de Chilpancingo, 2250 m, 1 Sep 1962, J. Rzedowski 15925 (MICH); 5 km W de Camotla, Munie. Chichihualco de Leonardo Bravo, 2600 m, 8 Apr 1963, Rzedowski 16403 (MICH); El Asoleadero, 15 km al oeste de Camotla, about 45 km W of Chilpancingo, 2650 m, 2 Dee 1963, Rzedowski 18073 (MICH), C. Feddema 2850 (MICH); + 2 km NE del Campamento El Gallo, approx. 17^ 28' N, 100° 13’ W, 2650 m, 27-29 Jan 1965, Rzedowski & R. McVaugh 135 (MICH). MEXICO: At state line between Michoacán and México on Mex. 15, 2600 m, 16 Jul 1961, S. A. Graham 53 (MICH); Temascaltepec, Comunidad, 2510 m, 6 Jul 1932, Hinton 964 (vs), Cajones, 9 Jun 1935, Hinton 7955 (MICH, MO), Pantoja, 9 Sep 1935, Hinton 8429 (US); ca. 6 mi S of Tenancingo and 8 mi N of Villa Guerrero on Route 55, 2600 m, 11 Aug 1960, J. T. Mickel 689 (MICH); Low hills along Hwy 15, 17 mi E of Zitácuaro, 14 Oct 1962, W. A. Weber & L. A. Charette 11879 (MICH). MICHOACAN: Cerro Azul, pres Morelia, 6 Apr 1910, F. Arsené s.n. (P); Temaxcal, 18 Mai 1939, T. C. & E. M. Frye 26067 (Mo, Us); on Mex. 15, 28 mi E of Morelia, 2150 m, 17 1969] Cuphea — Graham 487 Jul 1961, Graham 57 (MICH); Zitacuaro, 1800 m, 15 Mai 1938, Hinton 11853 (us); Tancítaro, 2050 m, 26 Nov 1940, «Hinton 15721 (US) (petals all yellow) ; 6 mi N of Tancítaro, 2615 m, 24 Jul 1940, W. C. Leavenworth 335 (MICH, MO); Mt. Tancítaro, 2300 m, 7 Mai 1941, Leavenworth & Hoogstraal 4018 (Mo), 2425 m, 9 Jul 1941, Leavenworth & Hoogstraal 4033 (Mo); 40 km east of Morelia, sobre la carretera a Zitácuaro (Km 269), 2600 m, 20 Jul 1964, Rzedowski & de la Sota 18361 (ENCB, MICH). MORELOS: Cerro Zempoala, cerca de las Lagunas de Zempoala, 3200-3400 m, 10 Nov 1957, L. Paray 2539 (ENCB); Valle del Tepeite, Aug 1932, E. Lyonnet 1531 (US); Lagunas de Zempoala, 17 Sep 1938, Lyonnet 2457 (US); barranca above Cuernavaca, 2000 m, 31 Jul 1896, Pringle 6382 (Mo, US), 11 Nov 1902, Pringle 9786 (Mo, vus), 21 Jul 1904, Pringle 11981 (MICH, US); near Cuernavaca, 27-30 Mai 1899, J. N. Rose & W. Hough 4409 (US). Oaxaca: 1842, M. Ghiesbreght 102 (P); vicinity of Cerro Zempoaltepetl, at Patio de Arena, about 5 km E of summit, 2800 m, 9 Aug 1950, Boone Hallberg 892 (MICH); Campamento Río de Molino, 4 km SW of San Miguel Suchistepec, 2250 m, 21 Sep 1965, Rzedowski 21061 (MICH). Locality unknown: Mexique?, 1826 (Photo, Field Mus. neg. no. 33463 from G at MICH). 2. Cuphea calaminthifolia Schlectd., Linnaea 12: 274. 1838. (Fig. 2.) Type: VERACRUZ: Cuesta grande de Chiconquiaco, Sep 1829, Schiede s.n. (Holotype, HAL!; isotypes, Photo of CGE isotype, MICH neg. no. 1732, 17381, P!). Cuphea corniculata Koehne, Fl. Bras. 13(2): 286. 1877. Type: HIDALGO?: between Tampico and Real del Monte near Pachuca, Berlandier 362 (Holotype, probably was at Berlin; isotype, P!). No specimen was cited in the protologue but Koehne later (Bot. Jahrb. 2: 415. 1882) gave the locality of the species as “Inter Tampico et Real del Monte!’’, then in his monograph (Pflanzenr. 216. 171. 1903) cited the single collection, Berlandier 362 from “Zwischen Tampico und Real del Monte bei Pachuca." Cuphea debilis Hemsl., Diag. Pl. Nov. Mex. 3: 51. 1880, (amplified in Hemsl., Biol. Centr. Am. 5: 439. 1880). Type: VERACRUZ: Jalapa, Dr. Coulter 156 (Holotype, K!). Cuphea, imberbis Rose in Koehne, Bot. Jahrb. 41: 94. 1907. Type: HIDALGO: wet banks, Trinidad Iron 488 Rhodora [Vol. 71 Works, 29 Jul 1904, C. G. Pringle 8979 (Holotype, US!; isotypes, MICH! Photo of Msc isotype, MICH neg. no. 1039, 1040! Mo! P!). Herbs, slender, trailing, probably perennial, with weak, decumbent stems 20-65 cm long bearing adventitious roots along their length, and covered with minute, two-branched, colorless, transversely oriented hairs and longer, weak, non-branched hairs, the hairs especially dense on younger portions. Leaves broadly ovate to narrowly elliptic, the blades 15-60 mm long, 8-30 mm wide, the apex acuminate, the base attenuate; petioles 4-10 mm long, bearing same hair types as the stem; blades thin, dark green above, pale gray-green below, both surfaces bearing minute stiff hairs mixed with sparse to dense, longer, weak hairs; leaves scarcely diminishing in size toward apex of the stem, In- florescence a leafy, few-flowered raceme; flowers 1 or 2 per node, interaxillary (or occasionally axillary) on bibrac- teolate pedicels 7-15 mm long; bracteoles linear, green or purple, less than 1 mm long. Calyx very slender, 12-27 mm long including a downward curving, basally rounded spur 1-2(-3) mm long, calyx at anthesis 1.5 mm wide, pale reddish purple to straw-colored; externally bearing dense, short, stiff, colorless, distally oriented hairs and scattered longer, weak, colorless or purple hairs, the latter often more concentrated at the proximal end of the calyx; in- ternally glabrous, neither bialate nor vesiculate; mouth of the calyx slightly flaring, not oblique; neck of the calyx scarcely contracted in fruit, Upper lobe of the calyx broadly deltoid, apparently deep purple within, 1.5 mm long, 2 mm wide, distinctly larger than the other five lobes which are less than 1 mm long and ca. 1 mm wide; appendages oblong to spathulate, slightly less than to three times the length of the small calyx lobes. Petals 6, greatly unequal, the two dorsal ones purple, obovate to oblong, (6-)8-12 mm long, tapering basally to a very short claw, subtended at the base by a thickened scale which is obliquely truncate to corniform along the free edge and included within the calyx; ventral petals purple to pale purple or white, 1-2 (-6) 1969] Cuphea — Graham 489 mm long, obovate. Stamens 9, 3 or 4(-6) exserted, glabrous. Pollen oblate, triangular to oval-triangular in polar view ; tricolporate, the pores equatorially arranged, equidistant, not protruding; syncolpate ; exine tectate, striate over entire surface of grain, striae extending to the poles; diameter 334. Gynoecium with capitate stigma and exserted, gla- brous style; ovules (8-)12-18; seeds dark brown, lenticular, nearly orbicular in outline, 2 mm long, 1.75 mm wide. Disc horizontal in spur, ca. 1 mm long, thick, nearly terete, slightly tapering to a blunt apex. An infrequently collected species, ranging from southern San Luis Potosí to Veracruz on fhe east and presently known from Jalisco and northern Michoacán in the western mountains (Fig. 3). Its habitat is identical to that of C. bustamanta, though the elevations at which it grows, 1450- 2500 m. are somewhat lower. It has been collected in bloom from May through December. Described by Schlectendal as not a very elegant plant, its distinguishing features are the weak trailing stems with transversely oriented malpighaceous-like hairs, slender, often colorless to pale reddish purple calyx with large dorsal calyx lobe, 9 stamens, and 6 unequal deep purple to pale or white petals. Vegetatively, it is very similar to C. bustamanta but the calyx of that species is red. lacks the large dorsal calyx lobe, and has smaller purple-black dorsal petals. In Koehne's kev (1903) the species are contrasted by axillary vs. interaxillary position of the flowers. Exami- nation of specimens has shown, however, that the position of the flower is generally interaxillary in both species, so that they cannot be distinguished by this character. Koehne, in initially describing Cuphea corniculata, sug- gested it might be merely a variety of C. calaminthifolia. He later reiterated this suggestion (Bot. Jahrb. 2: 414. 1882), but still later, in his monograph of the genus, omitted comment regarding such a relationship. The fea- tures of C. corniculata which should distinguish it from C. calaminthifolia are more a matter of semantics than actual character differences. Examination of type material 490 Rhodora [Vol. 71 shows that the obovate dorsal petals of C. calaminthifolia and the cuneate oblong petals of C. corniculata are essen- tially the same shape, those of the former varying from obovate to oblong, and the remaining petals on the isotype of C. corniculata being oblong. The only other character which separates the species is the shape of the scale. There is one flower on the isotype of C. corniculata and of the two scales present there, one terminates in a short horn-like process, while the other may be classified with the scales of C. calaminthifolia as obliquely truncate in shape. The other variable characters of the two species are identical, overlap, or are insignificantly different. Table I lists the characters taken from the type material representing this taxon and the others to be discussed in which variation is found. Characters not included on the table are considered identical on all the specimens examined. Color of the calyx and petals varies among the types but could not be deter- mined with certainty, due to the age and poor preservation of the specimens. Hemsley, in describing Cuphea debilis, suggested it was closely allied to C. calaminthifolia, but saw no specimens of the latter. The type of C. debilis consists of one whole and one partial stem, both much wilted at the time they were pressed, and each bearing a single flower. The general dimensions of the type are indeed smaller than the other taxa included within C. calaminthifolia, but the entire aspect of the type is a depauperate one. The differ- ences between C. debilis and C. calaminthifolia lie in the former’s generally smaller dimensions, fewer ovules, and the relatively long appendages of the calyx. The length of the appendages, as can be seen from the comparative table, falls within the range of variation of the species. The small dimensions and ovule number may be attributed to poor growing conditions of the plant, though even if con- sidered typical of the population are still within reasonable limits of size variation for the species. The species Cuphea imberbis, on the basis of its enlarged dorsal lobe, was considered by both Rose and Koehne to 491 Cuphea — Graham 1969] GI əyeouna) ÁA[enbr[qo unu gG-GT Suo unu GT-OI xuo[qo -918A0qO0 Seqo[ uey} Iioduo[| X-Z oje[nu3eds 0} Suo[qo unu Qc-LT unu QT-9 unu OT- uiu 0g-eI ww 09-47 wd 0F-08 siq4oqua 8 pepunod .O 9je2un.n K[onbt[qo uu Z-T Suo[ unu 9-G 91€A0q0 seqo[ ucun IeSuo| X €-Z a}e[nyyeds 0} Suo[qo unu GT-éT unu G[-¿I wut L -G unu QT-L uut 6¿-LI w 08 MdL ord d VI 9je2unij A[enbri[qo -ULLOJIUIO9 wu ¿-G'I suo, WU OI suo[qo Soqo[ uey} ieSuo[ Xe '3uo[qo wut LT-ST unu QT-8 unu Qa unu go-cT ww 09-0v wd $9-0G D20]Ó214409 8T-SI 9je2unaj AJenbi[qo Suo WU 9-g Suo[ WU G'8-9 37e A0qo səqo[ 3urpoe»xe ADWusus 0} [enbe ‘97840 suo, unu Gc-GTI Suo[ utu ppo suo] unu G f 9prA WU gT-8 Zuo; ur Ze-GT wd 09-06 010fv1uwun]no səna Q se[eog s[e3eq [equa A s[ejod [?940([ sodepueddy XÁ[9)) S[9?tped seporeq SOABI'T 7431H [eure3e]y ƏdAL WOI SIƏPLILYO AQLHLA JO uosmedwog 'I AQEL 492 Rhodora [Vol. 71 belong to Sect. Heterodon. However, the combination of trailing habit, long calyx, 9 stamens, transversely oriented malpighaceous-like hairs, and pollen morphology places it more naturally in Sect. Leptocalyx, series Bustamantae. The characters from the type of C. imberbis fall within the range of variation displayed by the other types studied and the species is unquestionably synonymous with C. calaminthifolia, The pollen morphology displayed by Cuphea calaminthi- folia deserves special comment. As recorded elsewhere (Graham & Graham, 1967) Cuphea is a distinctly eury- palynous genus with species frequently having a unique morphology. The section Leptocalyx palynologically is a natural one, characterized by triangular to oval-triangular, syncolpate, much striated pollen with non-protruding or slightly protruding pores. Section Heterodon, though more diverse in pollen characters, has oval-triangular to orbi- cular, non-syncolpate or syncolpate pollen with striae mostly concentrated around the highly protruding pores. The pollen of C. calaminthifolia is definitely of the type found only in Sect. Leptocalyx and substantiates the species placement there, rather than in Sect. Heterodon. Pollen morphology further suggests that the "key" character of Sect. Heterodon, the large dorsal calyx lobe, does not delimit a natural group of species but may well have arisen more than once in the genus in distantly related taxa (also cf. C. koehneana and C. paucipetala in Graham, 1968, p. 5). Thus taxa with this character need to be evaluated from the standpoint of their total morphology for a more natural taxonomic placement. Additional specimens examined: HIDALGO: Km 254, 14 Jul 1968, Irma Schnooberger 7999 (MICH); Km 310.5, between Zimapan and Tamazunchale, 15 Jul 1948, Schnooberger 8036A (MICH). JALISCO: Cerro del Muñeco or Sierra de Manantlán, 30-35 km SE of Autlán, 2100-2450 m, 29 Sep 1966, W. R. Amderson & C. W. Laskowski 3819 (MICH); Sierra de Manantlán, 15-20 mi SE of Autlán, about 2 mi from Aserradero San Miguel Uno, 2250-2400 m, 4-5 Nov 1952, Me- Vaugh 13867 (MICH, US); Near Santa Mónica, probable lat. ca, 20° N Long., ca. 104^, 30' W, 1950-2050 m, 12-13 Nov. 1952, McVaugh 14050 (MICH, US); 6-7 road-miles NW of San Miguel de la Sierra 1969] Cuphea — Graham 495 (40-50 km, airline, west of Ayutla), 1900-2000 m, 4 Nov 1962, Mc- Vaugh 22063 (MICH); mountains east of Mamantlán, about 15 mi S- SE of Autlán by way of Chante, ca 2500 m, 25 Jul 1949, R. L. & C. R. Wilbur 1832 (MICH), 27 Jul 1949, Wilbur & Wilbur 1878 (MICH). MICHOACAN: Dist. Coalcomán, Sierra Torricillas, 2360 m, 12 Oct 1938, G. B. Hinton 12354 (MICH, MO, US), Jul 1939, Hinton 15021 (Us), Barroloso, 2250 m, 21 Oct 1939, Hinton 15355 (MICH, MO, US). PUEBLA: Teziutlán, 1846 m, 20 Dec 1963, R. C. Koeppen s.n. (MICH); Chin- antla, Mai 1841, Liebmann 3795 (US). SAN LUIS POTOSÍ: 5 km NW de Ahuacatlán, Munic. de Xilitla, sobre el camino a El Lobo, 1450 m, 30 Aug 1957, J. Rzedowski 9256 (ENCB). VERACRUZ: San Miquel, Jalapa, 2000 m, 3 Jun 1938, E. K. Balls 4719 (Us); Coscomatepec, 10 Mai 1937, E. Matuda 1306 (MicH), Cerro Punta Coscomatepec, 10 Mai 1937, Matuda 1507 [1307?] (MICH); prope San Andres, Aug 1828, C. J. W. Schiede s.n. (HAL); prope Jalapam, Mai 1829, Schiede s.m. (HAL), Jun 1829, Schiede s.m. (HAL). The author gratefully acknowledges the aid of the cura- tors of the herbaria cited for the loan of valuable type material and information regarding special taxonomic prob- lems, The illustrations are by Susan Cielensky. DEPARTMENT OF BIOLOGICAL SCIENCES KENT STATE UNIVERSITY KENT, OHIO 44240 LITERATURE CITED GRAHAM, A. and S. A. 1967. Pollen morphology and taxonomy of Cuphea (Lythraceae). Rev. Paleobot. Palynol. 3: 155-162. GRAHAM, S. A. 1968. New species of Cuphea (Lythraceae) from Mexico. Brittonia 20: 1-10. KoEHNE, E. 1903. Lythraceae in Pflanzenreich IV. 216. 326 pp. Stuart Kimball Harris Notice has been received of the death of Dr. Stuart K. Harris on July 30, 1969. A long time active member of the New England Botanical Club and one of its editorial board for nearly as many years, Dr. Harris was President of the Club at the time of his death. A suitable memorial will appear in an early number of Rhodora. The Editor Volume 71, No. 787 including pages 367-193, was issued September 30, 1969. JAN Riodova JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by ALBION REED HODGDON, Editor-in-Chief ALBERT FREDERICK HILL x STUART KIMBALL HARRIS RALPH CARLETON BEAN ROBERT CRICHTON FOSTER \ Associate Editors ROLLA MILTON TRYON RADCLIFFE BARNES PIKE LORIN IVES NEVLING, JR. Vol. 71 October-December, 1969 No. 788 CONTENTS: Some Preliminary Statistics of the Flora of Southern Florida. Robert W. Long, Olga Lakela, and Rose Broome 495 Additions to the Flora of Saint Lawrence Island, Alaska Steven B. Young ........... erret SECUN Jurado HERCNTA 502 Triphora trianthophora (Sw.) Rydb. in Maine Lesley M. Eastman alla eee eene tenentes 509 Floristic Comparison of Three Bird Islands in the Gulf of Maine. A. R. Hodgdon and Radcliffe B. Pike .................. 510 A Preliminary Investigation of the Benthonic Marine Algae of Chesapeake Bay Region Arthur C. Mathieson and Stephen W. Fuller .................... 524 (Continued on Inside Cover) The New England Botanical Club, Jne- Botanical Museum, Oxford St., Cambridge, Mass. 02138 CONTENTS: — continued Etymology of Psychotria in View of a New Use of the Genus. Homer V. Pinkley essere 535 Glaucium flavum Crantz from Cape Cod Edward J. Hehre and Joan B. Conway ...... eee 540 Chromosome Studies in Habenaria Forrest C. Bent cicccccccccccccsccccecesscssscccessccccesscessesccessecesssecesaseesece 541 New Chromosome Counts in Ulmus and Platanus Frank S. Santamowr, Jv. asserentes 544 Chromosome Numbers of some Brazilian Angiosperms James R. Coleman and Lyman B. Smith eee 548 On a Weed Species of Rorippa. Reed C. Rollins ............... 552 Viburnum wurdackii, a New Peruvian Species T. R. Dudley coccecccccccccccssssscsssscsessesscssececscssssscacesssecaceacsecsesseeacarees 554 A Range Extension for Suaeda americana in New England Irene H. Stuckey sse cene 560 Notes on Recent Books. R. B. Pike accenn 561 Book Review: A Model Flora of Ohio — “The Monocoty- ledoneae" by E. Lucy Braun. Henry K. Svenson ............ 562 Validation of the Name Dahlia australis Paul D. Sorensen ......... 1 aaa eee 565 Studies in Eupatorium (Compositae) I. Revision of Eupatorium bellidifolium & Allied Species II. A New Species of Eupatorium (Section Hebeclinum) J. L. Grashoff and J. H. Beaman ooon 566 A New Weed Introduction in New England Atriplex sibirica C. V. Morton ...... l aasan LMY 580 Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 71 October-December, 1969 No. 788 SOME PRELIMINARY STATISTICS OF THE FLORA OF SOUTHERN FLORIDA: ; ROBERT W. LONG, OLGA LAKELA, and C. ROSE BROOME? We have been working towards the completion of a concise, descriptive manual of the vascular plants of south- ern Florida for approximately the past four years. Our interest in this work stems from the fact that this part of Florida, unique botanically in the continental United States because of its indigenous tropical flora, does not have an adequate, modern, taxonomic treatment for its plants. The area has not been covered in whole or in part by a manual since Small’s Flora of Miami (1913a), Flora of the Florida Keys (1913b), and Manual of the Flora of Southeastern U. S. (1933). It has been our aim to produce a new manual in the shortest possible time; publication cannot be delayed for an indefinite period to allow for detailed, lengthy re- search that is required to provide solutions for all the of South Florida. We are grateful for the aid furnished by grant no. B-16945 from the National Science Foundation and by the Cooley Botanical Research Fund, University of South Florida. We wish to thank also Dr. Richard A. Howard, Harvard University, for kindly agreeing to read the manuscript. "Research assistant. Present address, Department of Botany, Duke University, Durham, N. C. 495 496 Rhodora [Vol. 71 herbarium work. Since that time we have placed the revised names and records on “key sort” punch cards; this report is based largely on the statistics derived from an analysis of the flora using this method. There has been considerable interest in the plant ecology of southern Florida, and many of these studies included lists of plants for the area. Millspaugh's Flora of the Sand Keys (1907) is a detailed ecological and floristic survey of the small islands lying west of Key West. Harshberger (1914) described 16 “formations” of plants and included species’ lists in an ecological survey of southern Florida which he defined as the whole of Manatee, De Sota, Lee, Palm Beach, Dade, parts of Osceola, St. Lucie, and Monroe counties, Harper (1927) described southern Florida as that portion of the State extending from the Florida Keys north to Manatee Co. on the west coast, and to Indian River Co. on the east coast, an area of about 17,000 square miles. He listed 430 species for this area. In another ecological survey Davis (1943) limited southern Florida to Charlotte Co., the southern halves of Highland and Okeechobee Cos., St. Lucie Co. and the area south with a total of about 13,000 square miles. He classified the vegetation into 7 main types of communities: high pine forest and scrub, pine flatwoods rockland pine forest and saw palmetto, hammock forests, inland swamps cypress forests and bay tree forests, man- grove swamps and salt marshes, freshwater marshes and wet prairies, and coastal beach and dune vegetation. He identified 850 species of vascular plants including 90 species of trees, 125 species of small trees and shrubs, 130 species of grasses, and 90 species of sedges and rushes. Other note- worthy ecological studies of southern Florida that have species’ lists are Phillips (1940), Kurz (1942), Egler (1950), Laessle (1958), and Alexander (1955, 1959). We have limited our manual range to the southernmost counties in Florida, Collier, Dade, and Monroe, an area of approximately 5000 square miles, because of the general floristic correlations of this part of the State. DeWolf (1964) has estimated that the number of species ought to 1969] Florida Flora — Long, Lakela and Broome 497 be from 400-2000, mostly around 1000 in an area of this size and at this latitude. Our preliminary results show that the total number of species of vascular plants is 1,596. We expect this number to change slightly with revisionary study now in progress. Herbaceous species are 1,032 (about 65% of the total flora), woody species are 564 (about 35% ) of the total. From a taxonomic standpoint, the flora is predominately an herbaceous one. The general distribution of taxa is as follows: Families Genera Species Pteridophytes 11 36 74 Gymnosperms 4 4 7 Angiosperms 155 556 1515 monocots 29 165 538 dicots 126 391 977 totals 170 596 1596 The largest family in numbers of species and varieties is the Poaceae with 194. Next in size is Asteraceae with 153, Fabaceae with 134, Orchidaceae with 70, Cyperaceae with 62, Euphorbiaceae with 61, Rubiaceae with 40, and Malva- ceae with 32. The relative size and floristic diversity of the vegetation of southern Florida can be noted when one compares these results with the statistics of other floras. Small listed 146 families, 522 genera and 878 species of seed plants in his Flora of Miami (1913a) and 233 families, 1518 genera, and 5557 species of seed plants for the Southeastern States (1933). Comparisons with Small’s numbers are not too meaningful since he recognized a number of segregate species, genera, and families that are not generally ac- cepted today. Erdtman West, long-time student of the flora of Florida, estimated the total number of vascular species for the State to be about 3,000 (in litt.), an esti- mate considerably higher than that of Moldenke (1944). Fernald (1950) has 168 families, 1,133 genera, and 5,523 species of vascular plants for Central and Northeastern U. S. and Canada. Leon and Alain (1962) tabulated 181 families, 1,296 genera, and 5,785 species of seed plants for 498 Rhodora [Vol. 71 the Cuban flora with about 50% of the species endemic. D’Arcy (1967) listed 349 genera and 550 species of Dico- tyledons for the small island of Tortola, Virgin Islands. Dickson (1953) has 102 families and 415 species on 25 of the Lower Florida Keys, certainly a large number for so small an area. Ward’s checklist (1968) listed 318 genera and 1,183 species of lower vascular plants, Gymnosperms, and Monocots for Florida. ORIGIN OF THE FLORA The flora is composed of a tropical element and a non- tropical or temperate element. There are 970 species (about 61%) that are generally tropical in distribution, and 626 (about 39%) that are primarily nontropical in origin or distribution. The flora is predominately tropical, and the most significant contribution has come from the Antillean area of tropical America (Howard, 1954) ; 880 of the 970 species (about 91%) are also found in the West Indies. Of the 1,032 herbs 539 (about 52%) are West Indian or Caribbean in distribution. The woody flora has an even higher percentage of tropical Species with 77% of the trees, shrubs, and woody vines of southern Florida being tropical in distribution. The shrubs in particular are predominately Antillean in origin: 250 of 329 species are tropical, and 182 are West Indian species. Most of the 626 non-tropical species are also found in continental United States, and their distribution apparently extends into south- ern Florida. The identification and relationships of the endemic flora of Florida has been of concern to botanists for a number of years, and the problems are still largely unresolved (James, 1961). Howard (1954) accepted 385 species of Angiosperms as endemic to Florida. The total number of Species and varieties apparently endemic to peninsular Florida that occur in southern Florida is 160, or about 10% of the flora; 139 (87%) are dicots, and 60% of the dicots are herbaceous. We believe this number of endemics is much too high, and our revisionary study has already 1969] Florida Flora — Long, Lakela and Broome 499 reduced this number by demonstrating that some of Small’s “endemic” species are in reality West Indian taxa. It is probable that less than 5% of the flora is truly endemic. This ought not to be surprising when one considers that southern Florida is geologically very young even in com- parison with other areas in peninsular Florida such as the Lake Wales Ridge that has been continually available to plants since the Pliocene or earlier. PLANT ASSOCIATIONS We have been concerned with the plant ecology of the manual range. Particular attention has been given to the identification and the floristic composition of the important plant associations or communities. We have identified 13 principal associations chiefly by means of the local habitat. The numbers of species that may be found in these asso- ciations is as follows: scrub vegetation, 76; hardwood hammocks, 306; freshwater swamps, 188; dry pineland, 303; seasonally wet pineland, 361 (floristically the richest association) ; mangrove, 13; salt marsh, 23; wet prairies, 172; dry prairies, 69; coastal strand, 115; ruderal or dis- turbed areas, 250; aquatic and marshland, 191; and, marine waters, 5. Many of the species are ecotypically diverse and they can be found in more than one kind of habitat. Addi- tional, fairly well-defined plant communities can be segre- gated from the major ones listed here by refinements in the classification. One of the truly difficult problems we have had to deal with in the preparation of this flora is the identification of those species introduced into southern Florida mostly as ornamentals or useful plants that have become established outside of cultivation. If reasonable evidence is available that indicates the plant has escaped and is spreading ap- parently into new areas, we have included the species as part of the total flora along with the indigenous plants. The introduced and naturalized flora is approximately 250 species, or about 16% of the flora. We are including chromosome numbers with species’ 500 Rhodora [Vol. 71 descriptions where a count has been reported. Chromosome numbers are known for 229 species in the manual range (about 14% of the flora); 190 of these are herbaceous species. Most of the chromosome counts have not been obtained from materials collected in southern Florida, but rather are based on plants of other parts of the species” ranges. SUMMARY Good (1964, p 32) places southern Florida in the “Neo- tropical Kingdom”, Caribbean Region together with the West Indies, Bahamas, and Bermudas in his classification of the world’s floristic regions. Our preliminary statistics confirm that the flora is predominately a tropical one having close taxonomic relationships with the Antillean area. However, there are significantly more herbaceous species than woody ones. Ecologically, the richest plant associa- tions are the wet or dry pinelands and the tropical hard- wood hammocks. The former have high proportions of non-tropical or temperate species, the latter are predomi- nately West Indian in origin. One of the important by-products of the preparation of a flora treatment such as this has been the identification of a large number of taxonomic problems that merit bio- systematic investigation. It is our hope that a new manual of the flora will not only stimulate interest in plants of the area, but also will point out to botanists problems of all kinds for further research in southern Florida. DEPARTMENT OF BOTANY AND BACTERIOLOGY UNIVERSITY OF SOUTH FLORIDA TAMPA, FLORIDA 33620 LITERATURE CITED ALEXANDER, T. R. 1955. Observations on the ecology of the low hammocks of southern Florida. Quar. Jour. Fla. Acad. Sci. 18: 21-27. 1959. Ecology of the Pompano Beach Hammock. Quar. Jour. Fla. Acad. Sci. 21: 299-304. D’Arcy, W. G. 1967. Annotated checklist of the Dicotyledons of Tortola, Virgin Islands. Rhodora 69: 386-450. 1969] Florida Flora — Long, Lakela and Broome 501 Davis, J. H. 1943. The natural features of southern Florida, Fla. State Geol. Surv. Bull. 25, Dept. Conservation, Tallahassee. DEWoLr, G. P. JR. 1964. On the sizes of floras. Taxon 13: 149-153. Dickson, J. D., R. Woopsury, and T. R. ALEXANDER. 1953. Check- list of flora of Big Pine Key, Florida and surrounding keys. Quar. Jour. Fla. Acad. Sci. 16: 181-197. EcGLER, F. E. 1950. Southeast saline Everglades vegetation, Florida, and its management. Vegetatio Acta Geobotanica 3: 213-265. FERNALD, M. L. 1950. Gray’s Manual of Botany, 8th ed. American Book Co., N. Y. Goop, R. 1964. The Geography of the Flowering Plants, 3rd ed. Longmans Green and Co., Ltd. London. HARPER, R. M. 1927. Natural resources of southern Florida. Im, 18th annual report of the Fla. State Geol. Surv., Tallahassee, 18: 27-192. HARSHBERGER, J. W. 1914. The vegetation of South Florida. Trans. Wagner Free Institute Sci. Phil. 3: 51-189. Howarp, R. A. 1954. Contribution of the Caribbean flora to the southeastern coastal plain. (paper addressed to Systematic Section, Amer. Inst. Biol. Sci. meeting, Gainesville, Fla.) JAMES, C. W. 1961. Endemism in Florida. Brittonia 13: 225-244. Kunz, H. 1942. Florida dunes and scrub, vegetation and geology. Fla. State Geol. Surv. Bull. 23, Dept. Conservation, Tallahassee. LAESSLE, A. M. 1958. The origin and successional relationship of sandhil vegetation and sand-pine scrub. Ecol. Monog. 28: 361-386. LAKELA, O. and F. C. CRAIGHEAD. 1965. Annotated checklist of the vascular plants of Collier, Dade, and Monroe counties, Florida. Fairchild Trop. Gard. and Univ. Miami Press, Coral Gables. LEON, H. and H. ALAIN. 1962. Flora de Cuba. Tomo V. Uni- versidad de Puerto Rico, Rio Piedras. MiLLsPAUGH, C. F. 1907. Flora of the Sand Keys of Florida. Field Col. Mus. Publ. 118, Bot. Ser. 2: 191-245. MoLpENKE, H. N. 1944. A contribution of our knowledge of the wild and cultivated flora of Florida, I. Amer. Midl. Nat. 32: 529-590. PHILIPS, W. S. 1940. A tropical hammock on the Miami (Florida) Limestone. Ecol. 21: 166-175. SMALL, J. K. 1913a. Flora of Miami. New York, N. Y. 1913b. Flora of the Florida Keys. New York, N.Y. 1933. Manual of the Southeastern Flora. Univ. North Carolina Press, Chapel Hill. Warp, D. B. 1968. Checklist of the Vascular Flora of Florida, Part 1, Fla. Agr. Exp. Sta. Bull. 726, Gainesville. ADDITIONS TO THE FLORA OF SAINT LAWRENCE ISLAND, ALASKA STEVEN B. YouNG During the summers of 1966 and 1967, I had the oppor- tunity to spend several months traveling around Saint Lawrence Island by Eskimo skinboat and on foot.' During this time, botanical collections were made at all locations on the island known to have been visited by earlier botanists, and a number of hitherto unstudied areas were visited. A total of about 1,100 numbers, representing some 235 species of vascular plants, was collected. A complete set of speci- mens from these collections will be deposited at the Gray Herbarium. Duplicates will be distributed. The results of this study are being prepared for publica- tion in the form of a major paper. However, due to the present amount of interest in the Beringean region, it seems appropriate to publish a preliminary report on the new material from Saint Lawrence. Saint Lawrence Island lies in the northern portion of the Bering Sea, some 200 miles due south of the Bering Strait. The largest of the Bering Sea islands, Saint Law- rence has a land area of approximately 2,000 square miles, roughly equal to that of the State of Delaware. The island's terrain is quite varied. About one half of the land surface consists of poorly drained lowlands which usually contain great numbers of small tundra ponds. There are also large areas of rolling uplands, which include several small moun- tain ranges. The highest peak on the island rises to about 2,200 feet. Sandy barrier beaches and backshores make up a small but floristically rich part of the total land area of the island. Saint Lawrence's climate is polar maritime. Summer skies are nearly always overcast, and daytime temperatures seldom rise much above 50? F. During the winter months "This study was supported by grants from the Fernald Fund of the Gray Herbarium, The Institute of Evolutionary Biology at Har- vard University, and Sigma Xi. 502 1969] Flora of St. Lawrence Island — Young 505 the warming effect of the surrounding sea is largely abro- gated by the presence of pack ice, and mean temperatures are only slightly above 0° F. Storms with high winds are frequent throughout the year. Precipitation, mostly in the form of snow, is heavy. The high winds and heavy snow- fall combine to form deep drifts in sheltered areas, and these drifts may persist throughout the summer, even at sea level. In spite of the island’s location some 250 miles south of the Arctic Circle, Saint Lawrence supports a typical arctic tundra vegetation. There are no trees, and many of the shrubby species which are usually an important component of the tundra vegetation in low arctic areas are rare or absent on the island. Because of its location, Saint Lawrence is of considerable biogeographic interest. The island is located in the center of the area which is believed to have been emergent during the glacial maxima of the Pleistocene epoch, forming the Bering Land Bridge (c.f. Hopkins, 1967). There is some evidence that the most recent land connection between Eurasia and North America was by way of Saint Law- rence Island, rather than across the narrowest part of the Bering Strait (Hopkins, loc. cit.). The land bridge was subject to only local glaciation during the Pleistocene, and the entire Beringean Region undoubtedly served as a re- fugium for many elements of the arctic and boreal biota during the glacial maxima (Hultén, 1937). Hultén (1941-1950) lists specimens of some 190 species of vascular plants as having been collected on Saint Law- rence Island. In a later volume (Hultén, 1968) additional records bringing the total known flora of the island to 208 species are indicated by dots on distribution maps. The sources of these records and the locations of voucher speci- mens are not listed, however. The majority of the new records are undoubtedly based on specimens collected by Otto William Geist in the late 1930s, but some other col- lections may be included. Some of the new records also stem from new identification of older material and from 504 Rhodora [Vol. 71 new interpretations of the taxonomy of some of the species represented. Sixty-six species not known to Hultén (1941-1950) to occur on Saint Lawrence Island are represented in my collections. A list of these species is given in Table 1. Thirteen of these species are represented by Hultén (1968) as having been collected from the island, but the reports are not substantiated. These species are denoted by an asterisk in Table 1. About half of the species listed in Table 1 are confined to a few protected localities in the southern part of the island. This is particularly true of the aquatic and semi- aquatic species. About 15 species, otherwise unknown on Saint Lawrence, were found along about two miles of river bank in the south central portion of the island. Table 1. Additions to the known flora of Saint Lawrence Island Lycopodium clavatum L. ssp. monostachyon (Grev. and Hook.) Sel. Lycopodium alpinum L. Equisetum scirpoides Michx. Equisetum palustre L. Cystopteris fragilis (L.) Bernh. (subspecies indet.) Dryopteris fragrans (L.) Schott. Sparganium hyperboreum Laest. Potamogeton perfoliatus L. ssp. Richardsonii (Bennett) Hult. Potamogeton subsibiricus Hagstr. Potamogeton vaginatus Turcz. Triglochin palustris L. Calamagrostis neglecta (Ehrh.) Gaertn. Poa alpigena (E. Fries) Lindm. Festuca altaica Trin. Eriophorum Scheuchzeri Hoppe var. Scheuchzeri Eriophorum vaginatum L. ssp. vaginatum Kobresia simpliciuscula (Wahlenb.) Mack. 1969] Flora of St. Lawrence Island — Young 505 Carex Jacobi-peteri Hult. Carex pyrenaica Wahlenb. ssp. micropoda (C. A. Mey.) Hult. Carex stylosa C. A. Mey. Carex livida (Wahlenb.) Willd. Juncus sp. (aff. J. Fauriensis Buch.) Luzula multiflora (Retz.) Lej. ssp. multiflora Luzula arctica Blytt Veratrum album L. ssp. oxysepalum (Turcz.) Hult. Rumex graminifolius Lamb. Polygonum Bistorta L. ssp. plumosum (Small) Hult.* Claytonia arctica Adams Stellaria ciliatosepala Trautv. Minuartia arctica (Stev.) Aschers. & Graebn. Minuartia obtusiloba (Rydb.) House* Melandrium affine (J. Vall.) Hartm. Delphinium brachycentrum Ledeb.* Ranunculus aquatilis L. (subspecies indet.) Ranunculus Gmelini DC. ssp. Gmelini Ranunculus pedatifidus J.E. Sm. ssp. affinis (R. Br.) Hult. Ranunculus reptans L. Cardamine microphylla Adams* Draba macrocarpa Adams Draba borealis DC.* Draba nivalis Liljebl. Draba alpina L. Draba hirta L. Braga humilis (C. A. Mey.) Robins. (ssp. indet.) Saxifraga oppositifolia L. ssp. oppositifolia* Rubus arcticus L.* Potentilla Egedei Wormsk. var. Egedei* Potentilla elegans Cham. and Schlecht. Potentilla uniflora Ledeb.* Hedysarum alpinum L. ssp. americanum (Michx.) Fedtsch. Callitriche verna L. Epilobium anagallidifolium Lam. Epilobium angustifolium L. Epilobium palustre L. 506 Rhodora [Vol. 71 Viola biflora L.* Ligusticum mutellinoides (Crantz) Willar* Phyllodoce caerulea (L.) Bab.* Vaccinium uliginosum L. Primula nivalis Pall. Trientalis europaea L. ssp. arctica (Fisch.) Hult. Gentiana auriculata Pall. Evritrichium aretioides (Cham.) DC. Galium Brandegei Gray Campanula lasiocarpa Cham. ssp. lasiocarpa Campanula uniflora L. Aster sibiricus L. Species marked with an asterisk are treated as members of the Saint Lawrence Island Flora by Hultén (1968), but no substantiation is given. The following records are of particular interest. Carex Jacobi-Peteri was previously known only from the type location at Cape Prince of Wales (Hultén, 1968). Al- though not common on Saint Lawrence, this species was found at several locations. Carex livida from Saint Lawrence represents a range extension of over 500 miles, the nearest previously known station being in central Alaska (Hultén, 1968). The Saint Lawrence record is also apparently the first of this species from well within the true arctic. The specimens treated as Juncus sp. might be considered either as depauperate specimens of J. Fauriensis or as an undescribed species. Only a single small station was found. As far as I can discover, J. Fauriensis has not been re- corded from Alaska. Gentiana auriculata has apparently been recorded only once before in Alaska, from Attu Island. Two infraspecific taxa not included by Hultén (1968) in the flora of Saint Lawrence Island are represented in my collections. These are: Deschampsia caespitosa (L.) Beauv. var. glauca (Hartm.) Sam. and Pedicularis sudetica Willd. ssp. albolabiata Hult. Hultén (1941-1950) lists a number of species as being 1969] Flora of St. Lawrence Island — Young 507 of somewhat doubtful occurrence on Saint Lawrence Island. Im some cases the records are old and the speci- mens are no longer available, while in other cases the proveniance of the specimens is in some doubt. In the course of my work, I was able to substantiate a number of old or questionable records. Table 2 lists species which are represented in my collections from Saint Lawrence Island, and whose occurrence on the island might previously have been considered somewhat questionable. A few species which have been reported from Saint Lawrence should be excluded from the flora of the island until their occurrance there can be substantiated by voucher specimens. Some of these situations are the result of specimens being incorrectly labeled as to collection loca- tion, while others result from confusion of the taxonomy of such difficult groups as Carex, Draba, and Potentilla. In either case, one can hardly discredit the record without presenting a detailed discussion of each individual case, which is outside the scope of this paper. Let it suffice to say that some 20 to 30 species which have at one time or another been recorded as occurring on Saint Lawrence should be excluded from the flora of the island until their occurrence there can be substantiated. In most of these cases, I haye made intensive searches for the species in question at the location from which they were supposedly recorded, and have failed to find them, The species to be excluded from the Saint Lawrence flora will be discussed in a forthcoming paper. A similar situation exists with respect to some infra- specific taxa. Hultén (1968) claims that 16 of the species known to occur on Saint Lawrence are represented there by two infraspecific taxa, while two species, Rubus arcticus and Senecio atropurpureus (Ledeb.) Fedtsch. are sup- posedly represented by three infraspecific taxa apiece. My field work leads me to believe that there are very few cases in which two good taxa can be distinguished within a single species on Saint Lawrence Island. The dis- continuities which appear to exist are usually the result 508 Rhodora [Vol. 71 Table 2. Species collected on Saint Lawrence whose occurrence was previously substantiated only by old or questionable records. Hierochloe pauciflora R. Br. Eriophorum callitrix Cham. Carex subspathacea Wormsk. Carex Ramenskii Kom, Carex nesophila Holm Carex rariflora (Wahlenb.) J. E. Small Carex misandra R. Br. Betula nana L. ssp. exilis (Sukatsch.) Hult. Ranunculus Turneri Greene Saxifraga spicata D. Don Pyrola grandiflora Radius Andromeda polifolia L. Linnaea borealis L. Artemisia globularia Cham. ex Besser of insufficient collecting. For example, Rubus arcticus ap- parently reproduces mainly by vegetative means on Saint Lawrence, with the result that the plant usually occurs as discrete colonies, most of which can probably be considered to consist of a single individual. Nevertheless, the range of morphological variation within a single colony is often so great that three separate subspecies could be distin- guished using the key given by Hultén (1968). Under a fairly conservative taxonomic treatment, it appears that the known vascular flora of Saint Lawrence Island consists of about 240 species. Probably no more than 10 of these species are represented by two or more infraspecific taxa. INSTITUTE OF POLAR STUDIES THE OHIO STATE UNIVERSITY, COLUMBUS 43210 1969] Flora of St. Lawrence Island — Young 509 LITERATURE CITED HoPKINs, D. M. (ed.) 1967. The Bering Land Bridge. Stanford: 495 pp. HULTEN, E. 1937. Outline of the history of arctic and boreal biota during the Quaternary Period. Stockholm: 198 pp. 1941-1950. Flora of Alaska and the Yukon. Lund Univ. Arsskr., Bd. 37-46: pp. 1-1902. 1968. Flora of Alaska and neighboring territories. Stanford: 1008 pp. TRIPHORA TRIANTHOPHORA (SW.) RYDB. IN MAINE A new station for this orchid was found in Stow, Oxford County on August 24, 1969. At the present time this is the only known existing occurrence in Maine. Seymour (Flora of New England 1969) does not record it for the state, although it was reported by Jean Wallace (“The Orchids of Maine”, University of Maine Bulletin LIII No. 12: 1951) as growing in the Town of Raymond, Cumberland County in 1915. Two records of the species in Maine are given in Miss Wallace’s study, the earlier of which is an article entitled ““Pogonia pendula in Maine" by LeRoy Harvey (Rhodora II: 211-212, 1900). The station at Stow consists of many individuals growing in small groups on a wooded hillside. In lieu of specimens, photographs in color taken at the station have been deposited at the Herbaria of the Univer- sity of New Hampshire and the New England Botanical Club. LESLEY M. EASTMAN OLD ORCHARD BEACH, MAINE 04064 FLORISTIC COMPARISON OF THREE BIRD ISLANDS IN THE GULF OF MAINE* A. R. HODGDON AND RADCLIFFE B. PIKE The influence of birds on the flora of islands is of major interest to biogeographers. Birds are reported to carry propagules for great distances. Cruden (1966) discussing the findings of Darwin, Kerner, Guppy and Willis states, “on the basis of numerous examples which these observers give, there is little doubt that birds act as dispersal agents”. There is, thus, some reason to suppose that islands far from land may become populated to a considerable degree by plants whose reproductive structures were transported by birds. Certain islands also support enormous concentra- tions of sea birds which react with their limited environ- ment in many ways to facilitate or inhibit plant growth. By carefully documenting the flora of three bird islands in the Gulf of Maine, we hope to cast some light on the matter of dispersal of plants to bird islands and on the selective effects birds have on the flora. The three islands to be compared are Machias Seal Island off the coast of Eastern Maine, Matinicus Rock offshore of Matinicus Island in the Gulf of Maine, and Gull Rock, one of the Wolf Islands in New Brunswick in the Bay of Fundy. Some years ago we published an account of the vascular plants of Machias Seal Island (1962) emphasizing the in- stability of the flora as indicated by changes in the kinds and abundance of plants recorded at three different times, 1947-48 by Hawksley, 1960 by Pike, and again in 1962 by Pike and two experienced assistants. So far as we can determine the literature on the plants of bird islands in Northeastern America is sparse or else *This research was supported by a grant from the Central Uni- versity Research Fund of the Graduate School of the University of New Hampshire (CURF 273). Published with the approval of the Director of the New Hampshire Agricultural Experiment Station as Scientific Contribution No. 449. 510 1969] Bird Islands — Hodgdon and Pike 511 buried in the ornithological literature. One paper which relates somewhat to our work is by Smith and Schofield (1959). They observed and collected the vascular flora of Ciboux and Hertford Islands off the east coast of Cape Breton in Victoria County, Nova Scotia. Hertford Island of 27 acres lies about one and one-half miles east of the mainland of Cape Breton while Ciboux of 35 acres is out another half mile. These two islands are much closer to the mainland than the three bird islands discussed in this paper and presumably have become invaded by heavy bird popu- lations in recent time, indicated in part by the presence of several living woody species of trees and shrubs and by evidences of earlier forest conditions. The fact that woody species were diminishing on these islands is in line with our observations. Considerable work on bird islands has been carried out in Great Britain admirably summarized with many refer- ences by Darling and Boyd (1964). North Rona, an island of 300 acres reaching a height of 355 feet and lying some 45 miles from the mainland, resembles our bird islands in some ways. In several visits of botanists no woody species have been recorded and, at each visit, some new species appeared and others were absent. The first botanical record made gave 35 species; on the next visit 4 of these were missing but 12 new ones appeared. Again a few years later 8 of those recorded earlier could not be found but 4 new ones appeared. This fluctuating record is somewhat similar to our findings on Machias Seal Island (1962). A repeat visit to Machias Seal Island was planned to follow after an interval of several years to discover what further changes might have taken place. The earliest convenient time, when weather and other circumstances allowed, was on August 8, 1967, after a month or more of almost continuous fog. No striking changes were noted in 1967 compared with the 1962 observations and list. It should be kept in mind that the records made by Hawksley in the years 1947 and 1948 reflected, by his own account, the pressure of heavy grazing and much interference with 512 Rhodora [Vol. 71 nature. The establishment of Machias Seal Island as a bird sanctuary with consequent protection of its plant life must have helped to stabilize its vegetation. The few additional species that we found for the first time in 1967 are so indi- cated in the accompanying list. Some of these probably were overlooked in September 1960 and 1962 because of the exceedingly lush growth of overtopping asters in those years. At the time of our visit in early August 1967, the asters were still in an early stage of growth. The next logical step involved making a comparison of Machias Seal with other islands having somewhat similar environmental parameters. Matinicus rock, far out to sea south of Rockland, Maine, and several miles farther out than Matinicus Island itself, came immediately to mind. Both Machias Seal and Matinicus Rock which are of approx- imately the same size, ten to twelve acres, have had a long recorded history of occupancy by puffins, razor-billed auks and arctic terns, Both also have supported lighthouses but, being treacherous of access, have repelled human visitors except students of birds. In early 1966 we had received permission from Dr. & Mrs. Buckheister and the National Audubon Society to visit Matinicus Rock at a time of year when there would be no disturbance to nesting birds, but it was not until September 2, 1967, that we succeeded in getting there. Storms in 1966 and almost continuous fog in the summer of 1967 made it more difficult to reach Matinicus Rock than any of the many other islands we have visited during our eight years of botanizing in the Bay of Fundy and Gulf of Maine. Our visit in early September proved to be very difficult; strong, cold winds and rough water, as well as some fog and rain, permitted us only about one and one-half hours ashore to make collections, record observations, and make a photographic reconnaissance. Matinicus Rock is aptly named, the island is chiefly ex- posed granite and, unlike Machias Seal Island which has extensive “meadows” of asters, umbellifers, docks, grasses, sedges, sorrel, etc., the “Rock” has only continuous ledge with many interrupted limited patches of vegetation. Both 1969] Bird Islands — Hodgdon and Pike 513 islands have some pools, those on Matinicus Rock being much more numerous and thus accounting perhaps for the presence there of two species of Scirpus. On the other hand, Callitriche, an ideal pool inhabitant, was present on Machias Seal Island but absent from the “Rock” indicating that, after all, chance must play an important role in the migration of plants to these islands. Certainly such pools of varied extent, some of them toward the highest parts of Matinicus Rock, would be excellent habitats for many kinds of aquatic or paludal plants. During our brief stay, it was possible to visit all obvious habitats and nearly all of the separate patches of vegeta- tion. A total of 64 species was collected, this being a larger number than was taken at any one visit to Machias Seal Island. Considering the much larger area in vegetation on Machias Seal, we could accept this list as being reasonably complete for the year 1967 but, knowing the inadequacies of single collecting visits made under even the best of con- ditions, we must be frank to state that undoubtedly there are some species on Matinicus Rock yet to be discovered. Gull Rock is different in several respects from the other two bird islands. It lies closer to other bits of land, being less than a half mile from the nearest of the other Wolves; it is much smaller and rises more precipitately on all sides from the sea thus affording less opportunity for supralit- toral plants to grow and it supports gull populations instead of tern-auk-puffin societies of Machias Seal and Matinicus Rock. However, it shares certain characteristic features with the others. All three are far from the mainland, Machias Seal 11 miles, Matinicus Rock 19 miles, and Gull Rock 8 miles. There is, at the present time, little direct human interference with vegetation on any one of the three and no species of woody plant is found on any of them. While there is no close positive correlation in general of size of island and diversity of habitats, the islands under consideration do show such a correlation, Gull Rock, the smallest, having certainly fewer kinds of niches for vascu- 514 Rhodora [Vol. 71 lar plants. This presumably is reflected in the presence of only 34 species of vascular plants there. The influence of birds on these island floras is shown in many ways. All woody plants are excluded, embracing en- tire families or major taxonomic groupings, eg. Gymno- spermae, Salicaceae, Myricaceae, Corylaceae, Saxifragaceae, Empetraceae, Aquifoliaceae, Aceraceae, Ericaceae and Caprifoliaceae. Other major groups absent are Equiseta- ceae, Lycopodiaceae and Polypodiaceae, the only fern noted being the cinnamon fern, Osmunda cinnamomea on one island. No Orchidaceae were present while the following families were each represented sparsely on one island only by a single species: Liliaceae, Violaceae, Onagraccae and Labiatae. The above groups are all well represented on non-bird islands in the Gulf of Maine, sometimes in the immediate vicinity as in the Wolf Island group. It is evi- dently the environment created by the birds en masse that inhibits the growth of these species. The Wolf Islands other than Gull Rock are forested except for limited clearings and bird lawns although some forested portions in areas adjacent to bird lawns have been invaded by birds. Where concentrations of gulls are heavy, the conifers in particular are killed, but there seems to be stimulation of growth in many other groups. For example, in these situations we have noted an increase in growth in Rubus idaeus, Ribes lacustre, R. hirtellum, R. glandulosum, several species of Solidago and Aster, Achillea Millefolium, Iris versicolor, Ligusticum scothicum, Conioselinum chinense, Streptopus spp. Dryopteris spinulosa var. americana and some other species. It seemed to us that the mere release of these plants from the overstory of conifers would hardly have accounted for such lush development. Apparently up to a high level of fertility, many woody plants and various other kinds as well are stimulated but beyond this level toxicity results. It might also happen that direct contact of plants and birds in such cases serves to destroy the plants. Experimental work is needed to show in what way 1969] Bird Islands — Hodgdon and Pike 515 the environment inhibits the germination and or subse- quent development of many kinds of plants. The conclusion seems warranted that the environment provided by nesting birds on small islands in the Gulf of Maine serve initially to screen out and eliminate a consider- able percentage of potential migrant plants or of plants already present and to select the majority of its species from relatively few taxonomic groups. It follows that certain families of plants will be strongly represented; the Gramineae account for 16 out of the total of 114 species on the three islands. The Polygonaceae in- clude 8, Caryophyllaceae 8, Leguminosae 7 and the Com- positae 19. Thus, these 5 families comprise a bit over 50% of the total bird island flora. This contrasts markedly with the same families comprising 28.5% of the total Wolf Island vascular flora. Three categories may be made of the bird island plants, first the species that occur on all three islands, second those on two and finally those found only on one. Twenty-one species are common to all the islands. These are generally of common occurrence in the region with no particular adaptation for the bird islands as is shown by the presence of all of them in considerable abundance on the non bird island members of the Wolf Island group as well as their frequency of occurrence elsewhere in the coastal areas nearby. This group of species is well adapted to the cli- matic and edaphic conditions of the region and also posseses effective means of dispersal. The list of species that are present on any two of the three bird islands but are lacking from the third includes only 18 species. Thirteen species are shared by Machias Seal Island and Matinicus Rock and are absent from Gull Rock. Two of this group, Juncus bufonius and Plantago juncoides are of very general distribution in the area and their absence from Gull Rock or at least the fact that we didn't find them there is attributable to the fact that Gull Rock has a very abrupt rocky shore providing few or no suitable habitats for either species. The other 11 species 516 Rhodora [Vol. 71 of the group absent from Gull Rock are in general somewhat less widely distributed on the other Wolf Islands than those species of the first group and their absence perhaps may be accounted for also by the scarcity of suitable habi- tats. Four species are found on Machias Seal and Gul] Rock but are absent from Matinicus Rock while only one species, Lathyrus japonicus, occurs on Gull Rock and Matinicus Rock only. The remaining 75 species or nearly 66% of the bird island flora are present on only one of the three bird islands. Many of these casual species are weeds which in all probability were brought to the islands by man during his occupation of Machias Seal and Matinicus Rock where the majority of these are found. Many weeds have propagules which undobutedly may be carried far by air currents, water or birds. It is to be noted however that Ambrosia artemisii- folia which is absent from two of the bird islands, being found only on Matinieus Rock, is absent also from all of the Wolf Island group. Several years ago we paid a brief visit to another bird island known as White Horse Island, à Cormorant rookery and the nearest land to the Wolves. One of the dominant weeds there was Ambrosia artemisii- folia growing with huge nettles (Urtica sp.) making a most disagreeable tangle virtually covering the entire island. Both species are absent from Gull Rock though we have observed nettles on three of the other Wolf Islands. Human visitors on White Horse Island must be more frequent than on Gull Rock since the former is only a mile or two away from Campobello Island and the mainland of New Bruns- wick. Also it is much more likely that bird visitors from the settled islands and mainland could have introduced propagules more often on White Horse Island than on the Wolves because of the shorter distance to travel. Several species seem to be of chiefly fortuitous occurrence on the three bird islands such as Callitriche heterophylla, Arenaria lateriflora and Galium tinctorium on Machias Seal Island and Ranunculus Cymbalaria, Convolvulus sepium and Mertensia maritima on Matinicus tock. The 1969] Bird Islands — Hodgdon and Pike 517 habitats for these native species as well as for most of the introduced weeds would seem to be about equally favorable on both Machias Seal and Matinicus Rock. It is evident also that many of the records of plants found on only one island relate in some way to the proximity of that island to other land masses nearby, for example Gull Rock to the other Wolf Islands, Machias Seal to Grand Manan and the Coast of Maine and Matinicus Rock to Monhegan Island and Matinicus Island. The conclusion seems tenable that these bird islands, following the initial screening out or elimination of many kinds of plants including all woody plants, become populated to a great extent by introduced weeds or by those native species that occur abundantly on the nearest bodies of land. This is not to rule out the probability that part of the bird island flora is residual, having occupied these islands from early post glacial times or at least from some time previous to their domination by birds. It would appear that man has introduced many species of plants to suitable habitats on these islands. The evidence seems to show also that occasionally as with Callitriche on Machias Seal Island, birds may well be agents of wide dispersal. It should be kept in mind that puffins, razor-billed auks and arctic terns, which spend their time at sea except when nesting, are not likely candidates for effective transport of plant propagules. Herring gulls on the other hand are in part markedly ter- restrial and theoretically should be effective as agents of dispersal. The failure to find strong correlation of floras on the several islands and the frequent absence of poten- tially adapted species such as Ambrosia artemisiifolia which often are abundant a few miles or less away from an island, suggest to us that dispersal over miles of open ocean to bird islands is not frequently accomplished, Our study indicates that the screening effect of bird populations on bird island floras is more apparent than their transport of propagules. Herbarium specimens from the three bird islands dis- 518 Rhodora [Vol. 71 cussed are deposited in the Herbarium of the University of New Hampshire (NHA). DEPARTMENTS OF BOTANY AND OF PLANT SCIENCE UNIVERSITY OF NEW HAMPSHIRE DURHAM, NEW HAMPSHIRE 03824 LITERATURE CITED CRUDEN, ROBERT W. 1966. Birds as agents of long-distance dispersal for disjunct plant groups of the temperate Western Hemisphere. Evolution 20: 517-532. DARLING, F. FRAZER and J. M. Boyp. 1964. The Highlands and Islands, Chapter 10 and Bibliography; The New Naturalist Series, Collins, London. SMITH, E. C. and W. B. SCHOFIELD. 1959. Contributions to the Flora of Nova Scotia VI; Notes on the Vegetation of the Bird Islands. Canad, Field Nat. 73: 155-160. PIKE, R. B. and A. R. HopGpoN. 1962. Recent Changes in the Flora of Machias Seal Island. Rhodora 64: 340-3406. Bird Islands — Hodgdon and Pike 519 1969] Pd b FM" NH pd P4 4 A Pd PA >d s PA pd PS PS X M20 SODINLLVIN X (L961) X (1961) X X X X X aNV'ISI TVds SVIHOVIN SUBISIUDI “TBA “TT SUAISAUDI LIAN ‘UIIA $n22210]20 “IBA “STON $nsopnjnd. snd urag ‘UIIA 420249 “ABA TYLA snpyna sndriag "T wn4040po WUNYJUVLOYIJUYV "TT 2842204d, wne]y q ‘slag ('spnH) siqsnpod ‘IBA 0910 `V VIV 'A€^ I 0910 $2280406 V "nN (XI) sesuappuns s1804580140]707) UL ("I) vsonxay visdaupqoso(q "| $novu.aa snw "AOW $7$0]]14. “TVA “T $140w240 SW] ^T “T wnyoqnl wnop.o H ‘anvag (T) suadas wo4ftdo45 y "| susnjod ` q "[ $:18u2304d, *q "[ 0$8294dAw02 *q I DnUUD DOT "399A Y "uio, (Alay Wuqliog) DUMASI “IVA ‘YVBVIM Y UAT (WOH) vpnavadnvd ‘q "399A Y UIIA 02:012u24mm] DUIULIINA DAQNL “IVA J DAQNs DIN{SA J "[ Damowmpuurs DPUNWS O SANV'ISI AYIE AHUHL NO SVUOTA AO LSIT AAILVAYVANOD [Vol. 71 Rhodora 520 un MOOd TINO PS EO wA A aia "ET — 00 SAOINILYW a A v» uw ~ n rs PS X X X X (LOGIT) X (IN V'ISI "IV MS SVIHOVIN “‘{Spuouipy »juosni4qD]D 'V cep "I D]n)pd xojdi4]y `¿F "I UNQUY uwnipodouay) “TP "UI SHYNAQOAUOD 'q “OF UI D24D2942S49 4 `d cog TT annan wmnuoDfi]joq ‘8g "UD?»uesojoy "Y Lg VI 8ndsi42 ^W | 79g ARID) $njp]noiq40 "qp eg ‘WAVY $»2)1j529140p xawmngap pg ‘stg snpapod xowmnep ee '"qpA3 spana A ZS "ry SM82294 DINAN “TE VI 40JONSLIA SULT “OE AUDIT) 4010101 WNLYIUIASIS "6g Jseq (UT) 29948 vwuiovjuuS gg "uxo Y neuoqong snjrydoppy iea smuofnq *f "I $n240fnQ snounfg "1c 0314142 “ABA "Uer 0214142 *7) `9Z ‘QUAYLE M 72202]0d :O OZ UII sopo]pwA4owW `) "FS Áəu[O 20g D) “EZ IYNYJYIS n140do02$ :O ZZ ‘Yjueyny (urtexonp) viyanjsosanyds “ABA “LOT "(Si9q) suoosowwna4q 7D) "TZ Bird Islands — Hodgdon and Pike 521 1969] > > PA PA pd >< >< >< "T "dp o» bd | | | X201 SQOINILVIN e 9A Pe pA X < pA Fd "SF x (1961) X A CN V'ISI "IV US SVIHOVW "uxep (XY) $ns0514]8 *A€A I snappr snqnay urun[od (HEIL) DNPUMUIOLD “ABA “YSULLOM 29p3D7] `d "I Dwui428uD ` q "I DNDIALOU 0]]13u930d ‘doag (I) $esogp wnpog ‘SSOD (I) vaounl n2issnag "I wnagjsvunidpoy snunydvy `MOOH (Sig) »mjuopo apyng “IPE (CT) s240280d-0s4ngp vjjosdo2) [yn wnwnpbfijod wnaj2wvif L ‘I $2420 "qp 933joure'] $7:50]]tà "ABA “J Suado. "ap qusinq PUDUR snpnounuvy "I UMMM *2 ‘TT ?$u2à8.40D wnsp.o7) "I vauwveb ‘S O[[LIAQ (7T) vipaw 014001218 "I 94014220] DILDUAL V "I suaqunooud nuibog uo(T ('S1oq) sisuappuno `Ç JINDY (‘Qpury) Duwwadsora) “IBA 'qostin) (T) vunu 0140] Dod S UTIA YNY `S ulay (Slag) VUDNLIMUD nponng "GP VF [Vol. 71 Rhodora "uH xx "nn ~ FX X 300 SDOINLLVIN p > > x (¿961)uəəs x x x p PA 0606 m ANV'ISI IVdS SVIHOVIN 781312A 222142140 ` I PUeSUMOT, SISUIPVUDI “yz 'sutqoq 2pupa vispvsydny "I mab tg ‘TT DADUDIING wnunjog `Xxu9!IIN snoyrun sndoofiz ABID VS (7I) Dwn D4S10149 W wnidas ILA J wnidas $njna]oawo;) "T DUNAVU LNNIE) UIIA (I) wmnpion] wninajdoja09 I UNDNYIJOIS wnoijsnDv] I AADI WNLDD ‘ula ('MsSSne]H) uo)npoouəpp “ABA "unqorq wnsompun]£ wnigojid A Jexeg (Sxueg) $u2]]nd *dsqns pO 2:73/50]3/20 W MONA `q4Ə9Ə]N Sisuadna $uo1j0dau] ysing wpjfiydosajay 91[21421]10 7) "I $24480:20d. "T "[ snavuodnl SNALÄYIDT "I 022047) *A paeuoroq 020/12$n8uD MNA "p wnpiaqfü ` L "| Suadas *[ "[ ?$u2304d UNUO fru, '68 '88 "48 '98 "G8 ¥8 68 ‘68 "I8 08 ‘6L `8L “LL “OL “GL ‘tL ‘GL "eL “TL “OL 69 "89 “LO Bird Islands — Hodgdon and Pike 523 1969] X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X M00 300 (GNV'ISI TINS SQOINILVW "IV S SVIHOVW IH (I) 4odsp 'g ‘I $n22042]0 SNYJUO Ç 19q9A 9210102910 ` L "Zipuy WNULLAdSOLYRLI WNIDLVALD I, “I SUDUWUNINN UOPOJUOIT ‘doog (I) ISUIAALD wmis4i;) IIO (SSI) $2piot4021432714 014021420 AT "p malo fo] W `V ‘NN ?So)maum)] `V ‘BUOJ s2]D240Q 0p9])13/9 V "[ DSopuoL/ suapig I DYO{USIUWLAILD DISOLQUY "TT wnsourbyn wnyoydvur) TEW 9720joqwun `V ‘Ula sisuowunyot `V "[ 25120-1408 `V (xo[duro)) "T snaonqoj 431S V "I suasasadmas "S! "uio (usinq) s0]ptà “rea “yy 080m. obnpyog I UMALOJIU `D) "I 2w140d y UNN "Sw soyjuvbyo `d "uio (pnoeuieg) suardiap ‘rea WET sopioount *q "| Lolou obp)un] I "] 2218-038147) SNYJUDVUW A A PRELIMINARY INVESTIGATION OF THE BENTHONIC MARINE ALGAE OF THE CHESAPEAKE BAY REGION ARTHUR C. MATHIESON AND STEPHEN W. FULLER’ Few studies have been conducted upon the benthonic marine algae of Chesapeake Bay. Zaneveld (1966a) de- scribed the Cyanophyta flora of the area (i.e. between Cape May, New Jersey, and Cape Hatteras, North Carolina), but no systematic account of the other divisions is recorded. Zaneveld and Barnes (1965) described the reproductive periodicity of several species of seaweeds from the lower Chesapeake Bay. Wulff et al. (1968) have described the summer marine algae from a jetty on the open coast at Ocean City, Maryland. In the present paper we summarize the species found at 63 stations on the Chesapeake Bay and the Patuxent River (see Fig. 1 and Tables I-III for details of locations). All collections were made by the junior author in connection with a summer (1968) marine botany course given at the Chesapeake Biological Laboratory of the University of Maryland. The Bay is characterized by brackish waters and a lack of stable substrate. At most stations the bottom was sandy, but muddy areas were also evident. Seaweeds were attached to scattered boulders, pebbles, shells, pilings, cement blocks or other solid substrates. The surface water salinities ranged from 18.1 o/oo in the lower Bay to 3.4 0/oo in the upper Bay, and to 1.9 o/oo in the upper Patuxent River. The surface water temperatures at the same locations ranged from 24.8 to 30°C — most readings being near 27°C. The greatest temperature difference between the surface and bottom levels of water (35 feet in depth) was 2.5°C. Several shore collections were made at the mouth of the Patuxent River from June 26 to August 7, 1968, (Fig. 1 "Published with the approval of the Director of the New Hampshire Agriculture Experiment Station as scientific contribution number 457. 524 1969] Algae of Chesapeake Bay — Mathieson & Fuller 525 SCALE IN MILES | | | , \ CHESAPEAKE BAY ó 5 » 15 20 25 | E A = ü 37° | | Fig. 1. Chesapeake Bay and Patuxent River Stations. 526 Rhodora [Vol. 71 and Table I). Offshore collections were made in the Bay on July 1, 8, 19 and 25, 1968 (Table II) and the Patuxent River on August 7, 1968 (Table III). The offshore collec- tions were made aboard the motor vessel Orion or the Blue- fish. Samples were obtained with a steel frame trawl or with oyster tongs. Herbarium voucher specimens were made for all conspicuous species at each station. A complete set of specimens is deposited in the Herbarium of the Uni- versity of Maryland, while a partial set has been deposited in the Algal Herbarium of the University of New Hamp- shire. The nomenclature of the recent British Checklist (Parke and Dixon, 1964) has been applied in most cases. List of Species CHLOROPHYTA Cladophorales *Cladophora flexuosa (O. F. Müller) Harvey (Equals C. sericea (Hudson) Kützing sensu van den Hoek) Dredged in 5 feet of water at station F-5. *, fCladophora flexuosa (Dillwyn) Harvey f. densa Collins Found floating at station A-7. Previously recorded from Rhode Island (Collins, 1902). *, Cladophora gracilis (Griffiths ez Harvey in Mackay) Kützing Equals C. sericea (Hudson) Kützing sensu van den Hoek) Attached to rocks at stations A-9 and A-10. Dredged with Zostera marina at station B-5. Previously recorded from New Jersey to Newfoundland (see Taylor, 1957, for references). *Rhizoclonium riparium (Roth) Harvey Found once attached to pier pilings at station A-1. Ulotrichales *Enteromorpha clathrata (Roth) J. Agardh Entangled amongst Zostera marina at station A-3. *Indicates a new record for Maryland. fIndieates a southern extension of range on the Northeast Coast of North America. 1969] Algae of Chesapeake Bay — Mathieson & Fuller 527 Enteromorpha intestinalis (L.) Link Found unattached at stations A-3 to A-7, A-9, A-10, B-9, C-12, E-3, E-4, F-2, and F-5. Found attached (rocks, shells or logs) at stations A-3, A-6 and A-7. Enteromorpha minima Nageli (Equals Blidingia minima (Nageli ex Kiitzing) Kylin) On pier pilings at station A-4 and on rocks at station A-6. Enteromorpha prolifera (O. F. Miiller) J. Agardh Found once on pier pilings at station E-3. *Monostroma oxyspermum (Kützing) Doty Found once on a rock wall at station A-6. *Pseudendoclonium marinum (Reinke) Aleem et Schulz (Equals Protoderma marinum Reinke in Taylor, 1957) Found on rocks at stations A-7 and A-8. * *Ulothrix flacca (Dillwyn) Thuret in Le Jolis Found on pier pilings at stations A-1 and A-4 and on rocks at station A-7. Previously recorded from New Jer- sey to Baffin Island (see Taylor, 1957, for references). Ulva lactuca L. Found unattached and free-floating at stations A-3, A-4, A-6, A-8, A-10, B-1, B-5, B-9, C-1, C-4, C-5, C-6, C-8, C-12, E-1, E-2 and E-3. Attached to rocks or shells at stations A-3 and A-6. RHODOPHYTA Nemalionales * Acrochaetium flexuosum Vickers Found as an epiphyte on Zostera marina at station F-5. * *Trailliella intricata (J. Agardh) Batters Common as an epiphyte on Ceramium strictum, Poly- siphonia harveyi, Zostera marina and other species. Present at stations B-1, B-2, B-3, B-5, B-6, and B-10. According to Harder (1948) T. intricata is the tetra- sporophyte generation of Asparagopsis hamifera. T. intricata was previously recorded from Long Island to Newfoundland (see Taylor, 1957, for references). Rhodymeniales Champia parvula (C. Agardh) Harvey Dredged at stations B-3, B-4, B-5, B-6 and B-8. 528 Rhodora [Vol. 71 Gigartinales Agardhiella tenera (J. Ag.) Schmitz Dredged at station D-3. Gracilaria verrucosa (Hudson) Papenfuss Dredged or in drift at stations A- 8, A-10, B-1, B-2, B-3, B-4, B-6, B-8, B-11, C-4, C-7, C-8, C-12, D-1, D-3, E-1, E-2, and E-4, Ceramiales Ceramium rubrum (Hudson) C. Agardh Epiphyte on Gracilaria verrucosa, Zostera marina and other plants — only occasionally unattached, Present at stations A-3, A-6, A-10, B-1, B-2, B-3, B-4, B-5, B-8, B-9, B-11, C-4, C-7, C-8, C-12, D-1, E-3, E-4, and F-5. Ceramium strictum Har vey A common epiphyte on Gracilaria verrucosa, Zostera marina and Ceramium rubrum. Present at stations A-3, A-6, A-7, A-10, B-1, B-2, B-3, B-4, B-5, B-6, B-8, B-9, B-11, C-6, C-8, C-12, D-1, F-2, and F-5. Dasya pedicellata (C. Agardh) C. Agardh Found unattached at stations A- 10, C-3, C-4, C-5, C-6, E-1 and F-2. Polysiphonia harveyi Bailev Found unattached at stations A- 6, A-7, A-10, B-1, B-5, D-8, B-9, C-4, C-5, C-6, C-7, C-8, C-12, D-1, E-1, E-2 and E-3. Epiphytic on Zostera marina and other aquatic flowering plants at stations E-4, F-2 and F-5, Spyridia filamentosa (Wulfen) Har vey in Hooker Found unattached at stations B-2 to B- 7, B-10, B-11 and D- 1. Found as an epiphyte on Zostera marina at station B-1 and on Ceramium strictum at station D-1. PHAEOPHYTA Sphacelariales Sphacelaria fusca (Hudson) C. Agardh Found once epiphytic on Zostera marina at station F- 5; mixed with Acrochaetium flexuosum and Enteromorpha intestinalis. Of the 23 taxa of marine algae found in the Chesapeake Bay and Patuxent River, 11 are new records for the state 1969] Algae of Chesapeake Bay — Mathieson & Fuller 529 of Maryland and 4 are southern extensions of known dis- tributional ranges. Only one species of brown algae was recorded by us, although a few others are known for the lower Bay (Zaneveld and Barnes, 1965). The paucity of vegetation in the area is primarily related to low salinities and a lack of stable substrate. The biomass and diversity of species in the Chesapeake Bay is much less than in estua- rine areas of New England (e.g. Great Bay, New Hamp- shire, or Penobscot Bay, Maine), where fucoid and green algae dominate. More detailed seasonal investigations in the Chesapeake Bay will no doubt reveal additional species which are known for the vicinity (Wulff et al., 1968; Wulff and Webb, 1969; Zaneveld, 1965, 1966b). We would like to express our gratitude to the faculty, staff and students at the Chesapeake Biological Laboratory of the University of Maryland, who assisted the junior au- thor in collecting specimens and hydrographic data. Partic- ular appreciation is extended to Dr. R. W. Krauss, who directed the summer (1968) marine botany course, and to Miss Jean Snider who supplied temperature and salinity information for the Patuxent River stations. DEPARTMENT OF BOTANY AND JACKSON ESTUARINE LABORATORY UNIVERSITY OF NEW HAMPSHIRE DURHAM, NEW HAMPSHIRE 03824 REFERENCES CoLLINs, F. S. 1902. The marine Cladophoras of New England. Rhodora 4(42): 111-127. HARDER, R. 1948. Einordnung von Traillella intricata in den Gene- rations-wechsel der Bonnemaisoniaceae. Nachr. Acad. Wiss. Göttingen, Math.-Phys. K1., Biol.-Physiol.-Chem. Abt. p. 24-27. PARKE, M. and P. S. Dixon. 1968. Check-list of British marine algae — second revision. J. mar. biol. Assoc. U. K. 48(3): 783- 832. TAYLOR, W. R. 1957. Marine algae of the northeastern coast of North America. viii + 509 p. Univ. Michigan Press, Ann Arbor. Wutrr, B. L., E. M. T. WULFF, B. H. RoBINSON, J. W. Lowry and H. J. Humm. 1968. Summer marine algae of the jetty at Ocean City, Maryland. Chesapeake Sci. 9(1): 56-60, 530 Rhodora [Vol. 71 WULFF, B. L. and K. L. WEBB. 1969. Intertidal zonation of marine algae at Gloucester Point, Virginia. Chesapeake Sci. 10(1): 29-35. ZANEVELD, J. S. 1965. The benthic marine algae of Virginia. Vir- ginia J. Sci. 1965, p. 346. 1966a. The marine algae of the American Coast between Cape May, New Jersey, and Cape Hatteras, North Carolina. I. The Cyanophyta. Bot. Mar. 9(34): 101-128. 1966b. The benthic marine algae of Delaware. A preliminary checklist. Scientific Series, Publication No. 2, Insti- tute of Oceanography, Old Dominion College, Norfolk, Virginia, 35 p. ZANEVELD, J. S. and W. D. BARNES. 1965. Reproductive periodicities of some benthic algae in lower Chesapeake Bay. Chesapeake Sci. 6(1): 17-32. Table I. Shore Stations Station # Location June 26 to August 7, 1968 A-1 Pilings at Chesapeake Biological Laboratory boathouse A-2 Pilings at Chesapeake Biological Laboratory pier A-3 Beach on s. shore Patuxent R., just w. of entrance to Naval Station seaplane harbour A-4 Drum Point, mouth of Patuxent River A-5 #5 beacon, mouth of Patuxent River A-6 Rock retaining wall on Naval Base property, s. shore of Patuxent River A-7 Abandoned Cedar Point Light House A-8 Beach midway between Cedar Pt. and Point No Point — south of Patuxent River mouth A-9 Cove Point, just north of Patuxent mouth A-10 Beach north of Cove Point 9 19% VL c8 9'8 ‘sg upppog pue I# ueo ueo^jeq ‘Aeg ATTY qQ*4) — 47 = 9 39% 9c v8 L'8 ‘S urypog go 'spÁ 008T Ye) p `s[ uosieq Jo 'spA 006 ‘Avg £y qea Jo YMOWN 9-O x OT £'ec 0'9c c8 r8 puis] səuəweooO yo ‘a ‘spå 006 — €7O 2 OI 6'93 -0'97 6°, PL I# ues pu? Y SAO.LIEN ‘ST ie[doq useMjog +9 5 ZT e'96 g'az r6 "8 "T'spsdaeug yo ‘99# unu pot go *9 ‘SpA 00ST — $72 = 0% 6'vG 8' vc 8'8 r8 ‘YI ‘sy sdieug yo ‘99# unu pet Joos "IMT g- s3 ,88 L` 0:86 SFT 9°L ‘VJ ‘SI SAILS go OLT ‘99# unu pex zo m's seppu Get TO = g961 “g Aine A E 9 €'9c ELS LOT vor "q umaq Jo »- ‘a Kong uorjnijsqo iesu 7 3uexnjeq Jo YMON TrA E el £'86 7'83 Vel wal WS] ergg 4aədooH pue p# unu per ueo^jedg Ora | OT 6 LZ 8c eel ZST pue[s] u3roAspoo[g jo ysve sapu GT — 6H Ne 6 9'L6 uA EET 0'8T pue[s] uyaomsxspoo[g jo ysve '[eoqgg UT HYS — 8H 3 TLS 9'L6 GOT LOT 9# unu pat rou “eA ‘punog eyowmo0g L-A » ral 8'86 0°62 BIT Tor yool) peoig Yo ‘punog exourooqd 9-A i) 9 P82 g'9L (punog əywoutoood put ‘Y xəssəuu S -auuy ər $39euuoo) Y99019 peo:ig ur [euuteWopr]W G-q S KA! 9'LZ 9°82 Vor vor yaoarg ayeg Jo ujnour y ay xosseumuuy epar] va E 08 G°9% 1°93 eer eT "FI “SI 5 səwef Jo ‘os ‘spA 000g «eg 1943540 YY PAD — £&H X 88 0°98 LS TST (er presi) ° go "QI ‘ST səureí pue uoovoq 6# usemjoq ABMPI cea L OT L6 86 O L CTT uosveq ZH 1eeu “SI YSIEW “OS JO AIS pue[ oH Ta 50 896T ‘T AME < n Tuda "S œ uwuonoq ui0310q IS w0}}0g FS = © Jo Do 00/0 uorueoəo' uomse1s un yydeq o1nj39.roduyo J, AUNES suonejig Avg II eL [Vol. 71 Rhodora 532 BS UI01104T jo dəq {8S “8S 6 OL VOI 8'86 6'86 Or LOT 66 L'8c 6€ e 8I U 81 6°96 8 96 U6 06 ISS &'96 €6 E8 8'8 T6 wo07409 IS wo}jog IS Jo 00/0 91n3e.reduro T A3tut[e S VULIe SMOIIEN ÁƏəuld ye sud uo z Surjeopg puno sjue[d ‘SMOIIEN jJuoy uoov9q JO `U “SpA 008 'swo.reN sddeuy JO 'M uo9€2qQ Jo WU "Sp& 00G 'SMO.LIeN sddeuy jo *q 8961 “ce A[nf YIU OPTUMODT MA 1978919 JO 'spÁ 0081 'usiepy uo.oure([ JO Tq "1d PAA uoovoq TH JO `A 'spA OOPT 'usiepy uo.aure( yo asne ‘VT d IPUDUIA, Jo ` ‘SpA oorr ‘VI 3uroq Avasuryg əptsur 3snf uootoq [ix Jo atoysur ‘spA 00% “sI uuAMr) 94ous go 'spÁ 0061 “IT deip HOM Jo "^ 'sp& 0012 ysiey vournp go ‘spd 0082 qsiepy vourny go ‘spk 00ST ystey veurnp) Yo ‘spd 0001 :sSeip aou30 ¢ qsiepy vournr) yo SpA O0ZI ‘E# UBS Jo “MU ‘spA 009g ‘Aeg yolqon 896I '6T A[nf 9 9T “OT :SSVIp € *g 3t unu pax pue ‘sy sowe p UseMjeq [euuRYyD 'ST S4o[Áu L jo `u 3sní Sr yorya “sy səurep JO AUM “SPA oopp UOI€20'T oa ou ra «LD, FO a's LoyJINF A[oarssooons sFvip oeiu[ “g '8-2 # uonejg iG L8c 6'87 LOI 8'Or j[SrT [Boys sioxieq ‘Seg SUH — 6-M ,978 atoysyo ‘spd OOP] ISMH PAD — 8M 6-9 sppous urod sewoygy L-A OT LZ 08 Aeg AIS 9-4 ST-0T ve ‘sg uosqt go “spA 00F G-H jS e'gc 9'6c Ld GL "qd Iepa Jo 'spA QOOT-008 ‘teary 491sə9uO + QI uro11oq woog OS woog oS # Jo Ds 00/0 UOTyRIO'T uonejg 1969] Algae of Chesapeake Bay — Mathieson & Fuller 533 u1də(T əınzerodwə T ÁAqruneçs [Vol. 71 Rhodora 534 0g O1Oq[I€]y 1aMoT YO 06 86 F 986 oS 6'I ID uA014900 pue HHO pue[[oH useanjoq AeAprur so[quo aamod peəu4əA O ST Dl66c c 66 89 Vc juroq 94} go ‘sph oor poplooead atom S P L ƏL Yq ueureni[ Yo ‘spk 00T LI €'6c c 08 6'8 U8 08# unu pot Jo 'u *sp& 001 pəpaoəə. 9194 S 2 L UL Id JPY 'uonejg amog W sSur[rq "^4 3SursolleH ‘yae1g eue) Jo ‘spk 00g ,0€ L'8c cóc 0'OI v6 "d SunsolleH Yo ‘gZ# unu pat Jo 's *sp& 001 "d ueprteus Jo ‘u 'spÁ (cg ,S$ C66 Cog VII 901 ‘td ueptreus YO ZF uto Jo ‘a “spA COT ‘SJ Sowooig jo `w 'spÁ 00Z 06 L66 & 06 OTT CTT ‘ST Somootg YO gT# uooeəq Jo 's 'spÁ 00Z ,001 6€'96 8'6c 8'9I 9'el 9»u9neq 3uroq pue quroq 9uod JEH u92419q [euutQqoprur pop.ooei S P L "d ?uoq JEH JO “SpA 062 uoxej o3o« saeip OMI, ST 9'96 & 66 6'9r TET TƏAM Jo ujnour “Yq 30H JO 8961 '0c Isn3ny pəpaooə. Ayrur[eg pue soinjeteduiay, 5) Jsnsny poj»o[[oo so[dureg u0330gq ui0j)0q 2jg woy0g djs JO 00/0 UorjeoorT yydaq 91n39.roduro T, Aqyrur[es suon?js IAY Yuəxnieq III AQL # uontejs ETYMOLOGY OF PSYCHOTRIA IN VIEW OF A NEW USE OF THE GENUS HOMER V. PINKLEY During ethnobotanical research among the Kofan Indians of eastern Ecuador, Psychotria viridis R. & P. was ob- served and for the first time, identified as an admixture in the preparation of their hallucinogenic beverage known as yáje (6, 7,8). In the Kofán language, this plant is known as oprito, which, interestingly enough, is the name given to the human-like images which these Indians call *the heaven- ly people" and which they see while under the effect of the narcotic drink. They say that “. . . in order to increase their visions and to make them of longer duration . . ." the leaves and young shoots of the plant are added (1). Psychotria viridis is a glabrous shrub or small tree grow- ing usually in low land which may be flooded part of the year or quite frequently a short distance from streams or small rivers in the primary forest. The plant is character- ized by *. . . stipules large, acuminate, thin, brown, cadu- cous ; leaves short-petiolate, obovate or obovate-oblong, acute or short-acuminate at the apex, cuneately attenuate to the base; inflorescence terminal, pedunculate, spicate-panicu- late, the lower branches verticillate, the minute flowers ses- sile in distant glomerules; corolla greenish white; fruit red." (10) The plant has a wide distribution from Central America south to Bolivia. Because this is the first report of the use of a member of the Rubiaceae as an hallucinogen, I became interested in the etymology of the generic name. One could easily be- lieve that Psychotria is derived from the Greek wy (psyche), meaning mind, soul, or breath, but I do not feel this is the case. Linnaeus (4) derived Psychotria from Pat- rick Browne's name Psychotrophum (2). Linnaeus took Psycho- from Psychotrophum and added the nominative ending tria. Unfortunately, neither Linnaeus nor Browne state their reason for deriving these names. The following 535 536 Rhodora [Vol. 71 interpretation, which I believe to be wrong, became the accepted dictionary explanation of the origin of Psychotria. Wittstein (11) states, “Psychotria L. (Rubiaceae) Zus. aus wvyn (Seele, Leben) und toeqetv (ernárhen, erhalen) ; aus den Samen den Ps. herbacea bereitet man nach P. Browne auf Jamaica ein angenehmes coffeeáhnliches Getrank. Linné zog den ursprünglich von Browne gebildeten Namen Psy- chotrophum zusammen." Furthermore, in agreement with Wittstein, J. K. Small (9) states: “The generic name Psy- chotria is derived from a combination of two Greek words signifying soul-nourishment; the seeds of some species are used in the tropics as a substitute for coffee, hence the name wild coffee." But upon careful study of Browne's state- ment, “The seeds of all the species are pretty much like those of coffee," I surmised that he was not at all referring to a use of the seed but was rather describing and mor- phologically comparing the seed (2). After describing his "genera," Browne would often present a brief botanical summary giving an inclusive description of the group. This is what he did in the case of Psychotria which was later misinterpreted by Small, Wittstein, and perhaps, even by Linnaeus since the meaning of Browne's word is not fully expressed in the name Linnaeus coined. Before coming to this conclusion, I searched several early sources to see if any similar uses of Psychotria had been recorded, but I found none. Moreover, both Dr. Dennis Adams of the University of The West Indies and Mr. George R. Proctor of The In- stitute of Jamaica, through correspondence, reported that they had no records of Psychotria being used as a drink or a substitute for coffee in Jamaica. Emqually significant, Patrick Browne did not mention a use for this genus as he did with others when a use was known to him. What is probably the real etymology of Psychotria was discovered when I encountered the following statement con- cerning Psychotria in Hortus Jamaicensis by John Lunan: “This was so named from the Greek of an herb in Dio- scorides, so called from its delighting to grow in cool situ- ations.” (5). This name was then found in Book iv. i. of 1969] Psychotria — Pinkley 537 Dioscorides (3) under the name Kestron. He states, “Kes- tron that which is called Psychotrophon because it is found in the coldest places, which ye heighth of a cubit, or more, 4 cornered, ye leaves long, soft, like ye oak jagged round about, smelling well, but they are greater towards ye root, & on the top of ye stalks lies ye seed encased in an ear as of Thymbra.” Although this plant has been identified as Betonica Alopecurus, this is the first use of Psychotrophon trom which Browne made Psychotrophum and later Lin- naeus coined Psychotria. Hence, Psychotria is derived from the Greek word zo (psycho) meaning to breathe, to blow, to chill and the word tocqoz (trophos) meaning growing in, an adjectival form of the verb toeqew (trephein) meaning to grow, to nourish. Browne coined the word Psychotrophum which means grow- ing in a cool situation, probably because the species known to him were “. . . all natives of the woods, and grow best in a rich shady soil.” (2). Although the etymology of Psy- chotria does not turn out to be what one would expect, this does not, of course, detract from the significant role this species has in the hallucinogenic beverage of the Kofan In- dians and other aboriginal groups of the Amazon (figs. 1 & 2). Acknowledgements: To mention all who aided me in this paper, which is clearly interdisciplinary, would require a long list of names. Special thanks is given to my professor, Dr. Richard Evans Schultes, who has been a constant in- spiration and who turned my direction toward the Kofan ethnobotany. Also I wish to thank Mr. M. B. Borman, mis- sionary-linguist to the Kofáns, for his patience and kind assistance in translating the language. The assistance of Mr. Leonard Muellner in helping me with the Greek was greatly appreciated. I also wish to acknowledge the efforts of and give credit to Dr. John D. Dwyer of St. Louis Uni- versity, Department of Biology for studying my collections of Psychotria, #225 and #235 which as voucher specimens are now preserved in the Economic Herbarium of Oakes Ames in the Botanical Museum of Harvard University. May 538 Rhodora [Vol. 71 E hi A E opcm. ee r ien Fig. 1. Kofán Shaman administering the hallucinogenie drink yaje prepared from Psychotria viridis and species of Banisteriopsis. Plate 142 Fig. 2. Kofán Indians preparing to drink the hallucinogenic con- coction during an all night yáje ceremony. 1969] Psychotria — Pinkley 589 I express thanks to those whose names I did not include, especially those of the Gray Herbarium of Harvard Uni- versity. Dr. Ara Der Marderosian and Dr. Kenneth M. Kensinger have permitted me to publish the following per- sonal communication about their recent findings for which I wish to express my thanks. Current investigation by Dr. Ara Der Marderosian of the Philadelphia College of Pharmacy and Science has appar- ently shown the presence of Dimethyltryptamine, a known hallucinogenic chemical compound, in plant material col- lected by the anthropologist Kenneth M. Kensinger of Tem- ple University while carrying out field work among the Cashinahua Indians of Peru. This material was identified as Psychotria, possibly P. viridis by Homer V. Pinkley, and voucher specimens are now preserved in the Economic Her- barium of Oakes Ames in the Botanical Museum of Harvard University. BOTANICAL MUSEUM HARVARD UNIVERSITY CAMBRIDGE, MASSACHUSETTS LITERATURE CITED BoRMAN, M. B. 1967. Letter, November 15, 1967. 2 BROWNE, PATRICK. 1756. The Civil and Natural History of Jamaica. London. 3 DioscoRIDES (First century A.D.). The Greek Herbal. Translated by John Goodyer (1655); edited by Robert T. Gunther (1939). Hafner Publishing Company. 4 LINNE, CaROLI v. 1764. Genera Plantarum. Holmiae. 5 LUNAN, JOHN. 1814. Hortus Jamaicensis. St. Jago de la Vega Gazette, Jamaica. 6 SCHULTES, RICHARD EvANS. 1967. “The Place of Ethnobotany in the Ethnopharmacologic Search for Psychotomimetic Drugs.” in Ethnopharmacologie Search for Psychoactive Drugs edited by Daniel H. Efron. Washington, D. C. T . 1968. “The Impact of Spruce’s Amazon Explorations on Modern Phytochemical Research." Ciencia E Cultura, Vol. 20, No. 1, pp. 37-49. 8 . 1969. *De Plantis Toxicariis E Muno Novo Tropicale Commentations IV." Botanical Museum Leaflets, Harvard University, pp. 158-164. = 540 Rhodora [Vol. 71 9 SMALL, JOHN K. 1928. “Psychotria Sulzneri" Addisonia 13: 47-48. 10 STANDLEY, PAUL C. 1980. The Rubiaceae of Colombia. Field Museum of Natural History Botanical Series Vol. VIII, No. 1, Chicago. 11 WITTSTEIN, G. C. 1852. Etymologisch-botanisches Handwórter- buch. Ansbach. GLAUCIUM FLAVUM CRANTZ FROM CAPE COD In the summer of 1968, we discovered a single plant of Glaucium flavum in West Falmouth, Massachusetts (Barn- stable County). The plant was growing in sand two-three feet above the high tide level near the West Falmouth Jetty. We returned to this area in August, 1969 and observed three separate colonies of Glaucium (20 plants in all), growing with the two dominant components of the beach flora, Ammophila breviligulata and Artemisia Stelleriana. Most of the plants bore at least one fruiting capsule. The West Falmouth station is a slight range extension for Glaucium. It previously had been reported from Bristol County and Dukes County (the Elizabeth Islands), Massa- chusetts, and from Bristol, Newport and Prudence Island, Rhode Island (Seymour, 1969. Flora of New England). We were able to find one other New England record in the New England Botanical Club Herbarium, from Gay Head, Martha's Vineyard, Massachusetts (Dukes County). Voucher specimens were collected and deposited in the University of New Hampshire, New England Botanical Club and Southampton College Herbaria. EDWARD J. HEHRE DIVISION OF NATURAL SCIENCES SOUTHAMPTON COLLEGE OF LONG ISLAND UNIVERSITY SOUTHAMPTON, NEW YORK 11968 JOAN R. CONWAY DEPARTMENT OF BOTANY UNIVERSITY OF MASSACHUSETTS AMHERST, MASSACHUSETTS 01002 CHROMOSOME STUDIES IN HABENARIA FoRREST C. BENT The family Orchidaceae has received attention from numerous authors who have been interested in economic utilization of various taxa but less attention has been given to naturally occurring forms in temperate zone floras. Cor- rell (1950) considered the family and its distribution north of Mexico in North America, and recognized twenty-six species, seventeen varieties and two hybrid taxa in the genus Habenaria. When more local treatments are con- sidered it is noted that the number of taxa decreases and also that there is not complete unanimity concerning no- menclature. Thus, Fernald (1950) recognized twenty-six species, twelve varieties, four hybrids and four forms in the genus in north-eastern North America. Essentially the same taxonomic treatment was given by Donly (1963) and by Roland and Smith (1966), these latter describing ten species, five varieties, one form and one hybrid for the Atlantic provinces of eastern Canada. Cytologically, the genus Habenaria has received only limited attention. Earlier reports included the work of Humphrey (1932, 1933 and 1934) and Harmsen (1943). Darlington and Wylie (1955) reported on only two of the local species and followed the earlier nomenclatural system in which the genus was split into at least seven genera (Rydberg, 1932, reprinted 1965). Subsequently only oc- casional reports have been published (Ornduff, 1967 and Lóve, 1968). The present study was undertaken to deter- mine the cytological nature of Nova Scotian species of the genus Habenaria. The materials studied were collected chiefly in Queens County of Nova Scotia, but with scattered collections also from Kings, Antigonish and Inverness counties. Voucher specimens may be found in the herbarium of the Biology Department, Acadia University. Collections were made in the summers of 1964 and 1965, consisting of inflorescences, 541 542 Rhodora [Vol. 71 root-tips and young shoots according to the conditions of the plant at time of collection. Where possible the three types of material were collected from the same plant. Materials were fixed in Farmer’s fluid (three parts abso- lute alcohol to one part glacial acetic acid) for a maximum of twenty-four hours, after which they were transferred to seventy percent alcohol and stored under refrigeration. Root-tips and shoots were hydrolyzed in one normal hydro- chloric acid for three minutes and then squashed in aceto- carmine. Flower buds were dissected and the pollinia smeared in aceto-carmine. The chromosome numbers of twelve taxa of the genus Habenaria in Nova Scotia were determined (see Table I), with the numbers 42 and 84 being noted in somatic material. Of these, the numbers for H. fimbriata and H. flava have not been previously reported while the other counts confirm reports by Humphrey (1932, 1933, 1954) and Harmsen (1943). However, the count of 2n = 84 for H. hyperborea is noteworthy. Humphrey had reported 2n = 42 and Harmsen reported 2n = 84 for this taxon. The latter Table I. Chromosome numbers in the genus Habenaria Taxon 2n — previous record X H. Andrewsii White ex Niles 42 42 H. blephariglottis (Willd) Hook 42 42 H. clavellata (Michx.) Spreng. 42 42 H. dilatata (Pursh) Hook. 42 42 H. fimbriata (Ait.) R. Br. 42 H. flava (L.) R. Br. 42 H. Hookeri Torr. 42 42 H. hyperborea (L.) R. Br. 84 42, 84 var. huronensis (Nutt.) Farw. H. lacera (Michx.) Lodd. 42 42 H. orbiculata (Pursh.) Torr. 42 42 H. psycodes (L.) Spreng. 42 42 H. viridis (L.) R. Br. 42 42 var. bracteata (Muhl.) Gray 1969] Habenaria — Bent 548 author commented that the lower count is found in the more southerly part of the range of this species. Since Nova Seotia differs but little from Minnesota in latitude (where Humphrey obtained his material) and has milder winters, it might be questioned whether Harmsen’s con- clusion is valid. A study of material from a wider geo- graphic area should elucidate this problem. The present study indicates that the somatic number of 42 is widespread in this genus. BIOLOGY DEPARTMENT, ACADIA UNIVERSITY, WOLFVILLE, NOVA SCOTIA LITERATURE CITED CORRELL, D. S. 1950. Native Orchids of North America. Waltham, Massachusetts. DARLINGTON, C. D. & A. P. WYLIE. 1955. Chromosome Atlas of Flowering Plants. London, England. DowLy, J. F. 1963. The Orchids of Nova Scotia. Mill Village, Nova Scotia. (Privately printed) HARMSEN, L. 1948. Studies on the cytology of arctic plants. Med- delser om Gronland 131: 1-15. HuMPHREY, L. M. 1932. Somatic chromosomes of certain Minne- sota orchids. Amer. Nat. 66: 471-474. 1933. The somatic chromosomes of Cypripedium hirsutum and six species of Habenaria. Proc. Iowa Acad. Sci. 40: 75-78. 1934, The somatic chromosomes of Cypripedium hirsutum and six species of the genus Habenaria. Amer. Nat. 68: 184-187. Live, A. 1968. IOPB chromosome number reports. Taxon 17: 61. ORNDUFF, R. 1967. Index to Plant Chromosome Numbers for 1965. Utrecht, Netherlands. RoLAND, A. E. and E. C. SMITH. 1966. The Flora of Nova Scotia. Proc. N. S. Inst. Sci. 26: 3-238. RYDBERG, P. A. 1932. Flora of the prairies and plains of central North America. New York. NEW CHROMOSOME COUNTS IN ULMUS AND PLATANUS FRANK S. SANTAMOUR, JR. Ulmus and Platanus are among the genera that have been given high priority in the project on Selection and Breeding of Shade and Ornamental Trees, begun at the U.S. National Arboretum in 1967. As part of an over-all genetics research program, we have attempted to obtain cytological data on all trees used for breeding and observa- tion. Some of the plants are seedlings grown from seed obtained from native sources while others are older, estab- lished trees. This paper is primarily a report on the chromo- some numbers of elm and plane species and varieties which have not been previously noted in the literature. The meth- ods used in this study were standard aceto-carmine squash techniques with either root tips or pollen mother cells. Voucher specimens of Arboretum trees, parent trees from which seed was collected, and seedling progenies are on deposit in the herbarium of the National Arboretum. ULMUS The first report on elm chromosome numbers was made by Krause (1930), who determined the haploid number in U. montana With. (U. glabra Huds.) to be n = 14. Krause (1931) also found n — 14 in U. campestris L. (U. carpini- folia Gleditsch.) and 2n — 28 in U. americana Willd. Walker (1932) reported a different haploid number (n = 15) for U. pumila L. and U. fulva Michx. (U. rubra Michx.). She also determined the number of U. americana L. to ben — 28 or n — 30, and ascribed the difference to variation between trees. Later work, including that of Sax (1933a), has shown that 14 is indeed the normal haploid (and probably basic) chromosome number in elms and that most Species are diploid with 2» = 28. On the subject of U. americana L., Sax reported both the normal and pendulous forms as tetra- ploids with n — 28. 544 1969] Ulmus and Platanus — Santamour 545 In the most recent edition of the “Chromosome Atlas” (Darlington and Wylie, 1955), two chromosome numbers are listed for U. americana. The diploid count of 2n = 28 is ascribed to Krause (1930) — it should be Krause (1931) — while the reference for the tetraploid number is Sax (1933a). The question immediately arises as to whether the “U. americana” of the Krause study was the same spe- cies that Linnaeus described. Willdenow is given by Krause as the authority for the scientific name; and since Krause cbtained his material as seed or seedlings from a commercial firm, the problem of identification is more difficult. In view of past and present cytological studies on U. americana L., it appears likely that Krause’s plants were not true Amer- ican elm and that his diploid count may be eliminated from further consideration. Recent studies in American elm, along with other elm species, are reported below. U. americana L. n = 28 Five trees (ST-324, 331, 334, 336, 341) growing at the National Arboretum. Received as seedlings from U.S. Soil Conservation Service, Manhattan, Kansas in March, 1954. The author has previously determined native trees in Massachusetts and Pennsylvania to be tetraploids. U. americana var. floridana (Chapm.) Little 2n = 56 Seedlings raised from seedlot NST-44, collected by A. E. Squillace, U.S. Forest Service, in Fort White, Florida, in March, 1968. U. alata Michx. 2n = 28 Seedlings raised from seedlot NST-45, collected by A. E. Squillace, U.S. Forest Service, in O'Leno Park, Florida, in March, 1968. U. crassifolia Nutt. 2n = 28 Seedlings raised from seedlot NST-8, collected by L. Hin- ton, Jr. from native trees in Houston, Texas, in Novem- ber, 1967. U. macrocarpa Hance n — 14 Tree ST-370 growing at National Arboretum. One of five surviving trees raised from seedlots (P.I. 138008- 546 Rhodora [Vol. 71 158017) received in 1940 from garden of Summer Palace, Peiping, China. U. parvifolia Jacq. n = 14 Tree ST-300 of accession NA 25526. Received in 1929 as P.I. 82487 from Imperial Botanical Garden, Keijo, Japan. U. serotina Sarg. n = 14 Tree ST-305 of accession NA 18375. Source unknown. Tree now 35 feet tall. PLATANUS The earliest cytological study in Platanus was made by Winge (1917), who reported n = 21 for P. orientalis L. However, Pizzolongo (1958) has determined that Winge’s specimen was not oriental plane but P. X acerifolia (Ait.) Willd., the putative hybrid or hybrid derivative of the cross between P. orientalis and P. occidentalis L. Later inter- pretations of n = 8 (Brouwer, 1924) and n = 10-11 (Bretz- ler, 1924), for Platanus species have been shown to be incorrect: Sax (1933b) found n = 21 in both P. occiden- talis and P. X acerifolia and Pizzolongo (1958) reported that P. orientalis was also a diploid with n = 21, 2n = 42 chromosomes, The chromosome numbers of individuals of the above species were checked and, in addition, the following counts were made: P. mexicana Moric. 2n =42 Seedling plant (NA 29764) supplied by L. Lowrey, Hous- ton, Texas, in 1967. Grown from seed collected at Horse- tail Falls, south of Monterey, Mexico. P. occidentalis var. glabrata (Fern.) Sarg. 2n = 42 Seedling plant (NA 29763) supplied by L. Lowrey, Hous- ton, Texas, in 1967. Grown from seed collected along Blanco River in central Texas. P. racemosa Nutt. 2n = 42 Seedlings raised from seedlot NST-32, collected by J. Dourley, Rancho Santa Ana Botanic Garden, Claremont, California, in 1968. 1969] Ulmus and Platanus — Santamour 547 P. wrightii Wats. Iņ = 49 Seedlings raised from seedlot NST-106, collected by W. E. Earle in the New River Mountains, Arizona, in 1968. RESEARCH GENETICIST, U. S. NATIONAL ARBORETUM CROPS RESEARCH DIVISION, AGRICULTURAL RESEARCH SERVICE U. S. DEPARTMENT OF AGRICULTURE WASHINGTON, D. C. 20002 LITERATURE CITED BRETZLER, E. 1924. Beitrage zur Kenntnis der Gattung Platanus. Bot. Archiv. 7: 388-417. Brouwer, J. 1924. Studies in Platanaceae. Rec. des Trav. Bot. Neerl. 21: 382-396. DARLINGTON, C. D. and A. P. WYLIE. 1955. Chromosome Atlas of Flowering Plants. (ed. 2) George Allen and Unwin Ltd., 519 pp. KRAUSE, OTTO. 1930. Cytologische Studien bei den Urticales. Ber. Deutsch. Bot. Ges. 48: 9-13. 1931. Zytologische Studien bei den Urticales unter besonder Beruckichtigung der Gattung Dorstenia. Planta 13: 29-84. PizzoLONGO, PAoLo. 1958. Ricerche cario-tassonomiche su alcune Hamamelidales. Ann. Bot. (Roma) 26: 1-18. Sax, KARL, 1933a. Chromosome numbers in Ulmus and related genera. Jour. Arnold Arb. 14: 82-84. 1933b. Species hybrids in Platanus and Campsis. Jour. Arnold Arb. 14: 274-278. WALKER, RvTH I. 1932. Chromosome numbers in Ulmus. Science 75: 107. WINGE, O. 1917. The Chromosomes. Their numbers and general importance. Compt. Rend. Trav. Lab. Carlsberg. 13: 131-175. CHROMOSOME NUMBERS OF SOME BRAZILIAN ANGIOSPERMS JAMES R. COLEMAN AND LYMAN B. SMITH The material for this study was collected by the senior author while he was serving as research taxonomist at the Instituto de Botánica of Sáo Paulo, Brazil. The chromosome numbers were determined by the senior author and the material was identified by the junior author, with the ex- ception that Justicia leucophloea was determined by Mr. Dieter Wasshausen. The methods used in the cytological studies are identical to those described in a previous paper (Coleman, 1968). A complete set of voucher specimens has been deposited in the United States National Herbarium. Chromosome numbers are reported for 16 species (Table 1). The report of n = 8 for Eryngium ebracteatum confirms previous reports (Hamel, 1955; Bell and Con- stance, 1966) for that species as does the count of » — 12 confirm previous reports (Goodspeed, 1923; Lewis et al, 1962) for Nicotiana glauca, The remaining 14 reports are evidently the initial reports for the respective species. The report of n — 28 for Justicia (Rhytiglossa) leuco- phloea supports the position that Rhytiglossa is not distinct from Justicia (Wasshausen, personal communication). AI- though counts for species of Justicia range from n = 9 (Pal, 1964) to n = 17 (Ellis, 1962), the most frequently reported number is n = 14 (Grant, 1955). The report of n = 28 is the first report of polyploidy in the genus. The reports for Lafoensia pacari, n = c. 10, (Lythra- ceae), Deianira erubescens, n = 14, (Gentianaceae) and Prestonia acutifolia, n = 16-17, (Apocynaceae) are evi- cently the initial reports for these genera. The report of n = 12 for Loasa rupestris is the first report of that number in Loasa. The occurrence of n = 12 in Loasa is of some interest since Darlington and Ammal (1945) considered the genus dibasic with x — 7, 15 and at 548 1969] Brazilian Angiosperms — Coleman & Smith 549 least one species has been reported as n = 19 (Tschischow in Cave, 1963). It may therefore be hypothesized that the n = 19 species possibly had an amphidiploid origin from hybridizing n = 7 and n = 12 species. DEPARTMENT OF BOTANY, THE UNIVERSITY OF GEORGIA ATHENS, GEORGIA 30601 UNITED STATES NATIONAL HERBARIUM, SMITHSONIAN INSTITUTION WASHINGTON, D.C. 20560 LITERATURE CITED BELL, C. R. and L. Constance. 1966. Chromosome numbers in Umbelliferae. III. Amer. Jour. Bot. 53: 512-520. CAVE, M. S. (ed.). 1963. Index to plant chromosome numbers. Uni- versity of North Carolina Press, Chapel Hill. COLEMAN, J. R. Chromosome numbers in some Brazilian Compositae. Rhodora 70: 228-240. DARLINGTON, C. D. and E. K. J. AMMAL. 1945. Chromosome atlas of cultivated plants. Allen and Unwin Ltd., London. 397 pp. ELLIS, J. L. 1962. Chromosome numbers in some members of Acanthaceae. Sci. and Cult. 28: 191-192. GOODSPEED, T. H. 1923. A preliminary note on the cytology of Nico- tiana species and hybrids. Svensk. Bot. Tidsk. 17: 472-478. GRANT, W. F. 1955. A cytogenetic study in the Acanthaceae. Brit- tonia 8: 121-149. HAMEL, J. L. 1955. Etude caryologique de quelques Eryngium. Bull. Soc. Bot. France. 102: 488-502. Lewis, W. H., H. L. STRIPLING, and R. G. Ross. 1962. Chromosome numbers in some angiosperms of the southern United States and Mexico. Rhodora 64: 147-161. PAL, M. 1964. Chromosome numbers in some Indian angiosperms. 60 (B): 347-350. [Vol. 71 Rhodora 550 69% “BSOULLIOT pue əssoq uəƏA]əq “D%@40A[V :sUIOr) JO 931g C9G “ESOULIO, PUL 9ssoq UIIMJƏq “@pë4oA[V :SUIOr) Jo o3e3g €09 `sodureoO sop əsop OBS Jo yynos uy 6 :o[neq ogg Jo 9784S 88F ‘94I Jo ordrorun]y “(Əə|əuren : eryeg jo o1e3g GLP `9994] JO ordrorun]y *eare[ourer) :erqeg Jo oje3g a 98€ “BIURS Bose] IIU 'vejues vosey op emay :SI49D SUUIJW JO Ə161S P09 -'o[neq OVS ‘opeisy op onbreq :o[neq ogg Jo o3e3g GPS 'eurnuosLt0;) Jo ordrorun]y *oqurzoepa19A :erqeg Jo oje1g 0g0,. -"eurueiiop) :erueg Jo oje1g Age I PEL LI-9I VI OT? él pro 9I II yr? `umuəs “YH ('uquəg) D1]0f17no0 0140]82.4] AVAOV NAOQOdV ag»e[uos 2 “wey Su2282Qn49 D42096] AVAOVNVILNAD ‘We wnjv2ojon4qo wmniDufiAng IV SHAT'TTGEGEIN 1 HH 3$ unnd Disuoofv'T AVAOVYH LAT "uper 81/252dn.4 DSDO'T AVAOVSVOT neusyong `T 2200412010 wmnjoondo.[ IVO V'IOWVdOL ‘Od PLOYLAILQ 014013047) ‘Yue (HEW P SƏN) WUNIINLLOI Dwut2so3dwum) AYSONINWNÐAT "qesrir) CHEW) $730]0qns SNLOPOULY IT SVSIOVINSITV 551 Brazilian Angiosperms — Coleman & Smith 1969] $8€ oed1() sop viieg [euoroeN enbaieq :orrpuef ap on Jo 94¥4g 06P `Ə9Ə4] JO ordrorun]y ‘Lapur :erqeg JO 3484S 6LY '9goe1[ Jo otdporun]y 'eire[ourer) Jo ymos wy gg :Iueg Jo 9784S egg ‘sowo pue essod uoaAjeq 'epeiOA[V :SBION JO 93€38 9gG -'?urjuall0) :BIyeg Jo 9484S PZE 9?unualio) VILY Jo 9784S 68F ‘9I JO OIdPIUNW *exro[ourer) :erqeg Jo 9181S ‘alojoq A[pPI[VA epeur uəəq jou SEY 3r ‘STSUAM -ay Xepu[ ur pue snuioipoiq ayy ur 'AuiuouAs ur 92143 peusr[qnd uəəq seu uorjeurquioo əy} ysnouyVy “LEST ‘Che TIL ‘Apolg “Od Ul SƏƏN ?oopdoono] vssopbyhyy 'AOU ‘quod ‘uasneysse Mm (SƏƏN) mvoojudoono] Di01)8T f. x “LoyjNe IOIUasS Jo əsou1 SLaquInu uonoə[[0O + 86 cI él él rat eI x y UOSNBYSSB M (SƏƏN) va07Ydoona] v121?8n f aAVGOVHINVOV ‘yqueg Duoyig v140]2Dwy AVAOVIEVTAHd ONS ‘ds wnunjogy "uypues wnf{ns wnunjos WIA "o "I wna4odsp wnunjog "I wnjojnaiund wnwp]og urusa NINDI 00002 N AVGOVNV'TIOS ON A WEED SPECIES OF RORIPPA. — The area of origin and proper identity of widespread weeds are often difficult to establish. This is particularly true when differ- ent names are used in different parts of the world for the same species. A case in point is that of Rorippa indica (L.) Hiern. We have seen specimens of the apetalous variety of this species collected from a number of places in the West- ern Hemisphere, especially in Brazil; in the West Indies (Jamaica and Puerto Rico) and in the United States from Louisiana and from near Portland, Oregon. Besides R. in- dica, the names most commonly used for this species on the specimens and in the books are Rorippa heterophylla (Blume) Williams and Rorippa montana (Wall.) Small, but specimens were found misidentified in such remote gen- era as Sisymbrium and Cardamine. Rorippa indica is native in Eastern Asia and is present there in both petalous and apetalous infraspecific taxa. All of the evidence I have been able to assemble supports the position taken by Hochreutiner' that the plants of this weedy species can readily be accommodated in two varieties, var. indica and var. apetala. The original material of Lin- neaus possesses petals and the plants of var. indica are slightly taller with shorter siliques than is characteristic of var. apetala. 'The American material should be referred to F. indica (L.) Hiern, var. apetala (DC) Hochr. Hiern: definitely transferred the epithet indica to Rorippa, thus making superfluous the same transfer made by Hochreutiner (l.c.). The transfer by Hiern was made on an unnumbered page following p. XXV under a heading, “additions and corrections”. The place is sufficiently obscure to be easily overlooked, but the index does give Rorippa indica and refers to p. XXVI. R. heterophylla and R. montana appear to be straight synonyms of R. indica. The name Nasturtium montanum was given in Wallich’s list without description. I have not attempted to determine the first place a descrip- tion was supplied because that is irrelevant to the present 'Hochreutiner, B. P. H. Candollea 2: 370-371. 1925. "Hiern, W. P. Cat. Welw. Afr. Pl. pt. 1, following p. XXV. 1896. 552 1969] Rorippa — Rollins 553 consideration. Certainly Small’s use of the name in the ap- pendix to his Flora of Southeastern United States (Ed. 2) is not correct. It is interesting that the apetalous type of R. indica is the one that has become the weed rather than the variety with petals. It is a safe assumption that var. apetala does not require insect pollination. It is most probably both self compatible and predominately autogamous. Thus, it un- doubtedly possesses the appropriate breeding system to spread far outside of its native area. This conforms to expectations. REED C. ROLLINS, GRAY HERBARIUM HARVARD UNIVERSITY VIBURNUM WURDACKII, A NEW PERUVIAN SPECIES T. R. DUDLEY Most botanists will concur that an immense need exists for accelerated exploration, collectine, and systematic and ecological studies in Latin America. South America, Cen- tral America and Mexico have long been recognized as major centers of diversity and evolution within the genus Vibur- num (Caprifoliaceae). The studies on South American Viburnum (Killip and Smith 1929, 1930; Killip 1937), al- though adequate when published for herbarium material then extant, obviously cannot be expected now to be defini- tive. A modern, critical revision of the Andean taxa is much needed. This fact is greatly emphasized as the floras of remote, unexplored, and previously uncollected areas are gradually studied concurrently with the development of modern methods and concepts in systematics and ecology. Whereas the taxa of Viburnum from two other great cen- ters of heterogeneity of Eastern Asia and North America are rather thoroughly collected and taxonomically docu- mented, the Viburnum taxa of Latin America are essentially unknown in the broad sense of taxonomic relationships. Increased field work in Latin America and ne plus ultra analysis of all collections of Viburnum, as well as of other genera, will without doubt yield new taxa, define and solve outstanding problems of plant multiformity and evolution in the tropics, and interpret inter-taxa affinities. The Peruvian taxon of Viburnum described here is based on analysis of two collections discovered by the author in several herbaria, while surveying Viburnum for projected revisionary and monographic treatment. The line drawing was prepared by Mrs. Regina O. Hughes, formerly illus- trator for New Crops Division, U.S. Department of Agri- culture, Beltsville, Maryland. The photograph of the in- florescence was provided by Dr. J. J. Wurdack from his collection of field photographs. 554 1969] Viburnum wurdackii — Dudley 555 Viburnum wurdackii Dudley, sp. nov. Quae species affinis videtur V. weberbaueri Graebn. (= V. reticu- lato Ruiz & Pavon ex Oerst. — sensu lato) sed foliis sinuato-undulatis (nec valde dentatis), late ovato-ellipticis, maioribus, subtus toto densis fulvo-tomentosis implexis e pilis stellatis dimorphis compositi, inflorescentiis radiis pedunculo maioribus et robustiore, floribus multum congestioribus sessilibus vel interdum subsessilibus, corolla per anthesin majore 5-8 mm. diametro dense pubescentibus e pilis stellatis tecti, lobis margine valde et aequaliter papillosis apice ovato- spathulatis, calycis lobi rotundi margine ciliolati, ovario calyce corollaque toto densis pilis stellatis fulvis implexis tectis, fructibus immaturis pubescentibus e pilis stellatis densis compositis differt. Frutex vel arbor parva, 3-5 m. alta; ramis, ramulis, foliis subtus, petiolis, inflorescentiis, ovariis, calycinis lobisque dense crasseque luteolis ad fulvis indumento obsitis. Folia subcoriacea, estipulata, late ovato-elliptica vel raro parum obovata, (6-) 10-16 (-20) em. longa, 4-9 (-11) em. lata, apice acuta vel rotundata, mucronata, basi sub- truncata ad rotundata, margine sinuato-undulata, supra viridia sed semper cum pilis stellatis minutis aeque dispersis disjunctis tecta, subtus pallidiora densiore fulva dimorpha pubescentia e pilis stellatis appressis interspersis pilis stipitatis composita, nervis primariis 4, subtus prominente elevatis (sicut costa media), supra uterque levis- sime impressis, nervis secundariis transversis subtus distincte elevatis, supra uterque levissime impressis; petioli 1-2 em. longi, canaliculati, dense pubescenti. Cyma corymbiformis, 7-12 (-20) em. longo et lato, pilis luteolis vel fulvis crebris appressis, pedunculo 1-6 cm. longo, crasso, radii primarii 3-6, usque ad 6 cm. longi; bracteae bracteo- laeque plerumque minutissimae anguste obovatae cauducae pubes- centae; flores numerosi congesti, sessiles vel interdum subsessiles in radiis ordinis tertii siti basi bracteis bracteolisque oppositis suffulti; ovarium 1-2 mm. longum, campanulato-cylindricum densim pubescens; calycis lobi rotundo-orbicularis, 0.5-0.6 mm. longi, densim pubescentes, margine ad apicem ciliolati; corolla alba, rotata, 5-8 mm. diametro, 2 5-4 mm. longa, extus pubescens, lobis rotundato-spathulatis pariter aequaliter apillosis; stamina corolla superanta, corollae basi affixa, filamentis 3-4.5 mm. longis, antheris flavis, oblongo-ovatis, c. 1 mm. longis. Drupa immatura, viridia, 3-4 mm. longa, 2-3 mm. lata, ellip- tica, indumento denso e pilis stellatis tecta, apice praecipue 5-10 pilis fasciculatis vel simplicibus conspersa, calyce styloque persistentibus coronata, stylo cylindrico, c. 0.5-0.8 mm. longo, calycem paulo superante, stigmate capitato trilobulato. TYPE LOCALITY: Peru, Molinopampa-Diosan pass in Provin- cia de Chachapoyas, Departamento de Amazonas ; the type is J. J. Wurdack 1648. 556 Rhodora [Vol. 71 y rag aypana nia NT ue ys. © Wag hee Fig. Viburnum wurdackii Dudley. Holotype: J. J. Wurdack 1648. A. Habit X 1, second leaf from bottom, right side shows upper sur- face; second leaf from bottom, left side shows lower surface. B. Indumentum on leaf lower surface of intermeshing appressed stellate hairs and interspersed stipitate stellate hairs, XX 40. C. Bracteoles subtending secondary and tertiary rays of inflorescence, size is variable according; to position, X 4. D. Cluster of flowers on tertiary 1969] Viburnum wurdackii — Dudley 557 DISTRIBUTION: Presently known only from two Peruvian collections in Departamento Amazonas. It is recorded as a white-flowered shrub or tree 3-5 m. tall, growing in moist scrub forest at 1500 m. (F. Woytkowski) and 2700-3100 m. (Wurdack). PERU. PROVINCIA DE CHACHAPOYAS, Departamento de Ama- zonas: south side of Molinopampa-Diosan pass, 8 Aug. 1962, J. J. Wurdack 1648 (us, holotype; NY, isotype) ; De- partamento de Amazonas: Mendoza, 6 Aug. 1963, F. Woyt- kowski 8124 (GH, MO). The new species described above resembles Viburnum weberbaueri Graebn. (Bot. Jahrb. Syst. 37: 433. 1906) in some respects, but it is clearly distinguished as follows. The leaves of V. wurdackii Dudley are larger, broadly ovate to elliptie, with sinuate to undulate margins and with dimor- phic indumentum on the lower surface composed of stipitate stellate hairs interspersed with heavily matted, intermesh- ing and appressed stellate hairs. In Viburnum weberbaueri, the leaves are mostly obovate or oblong-ovate, and dentate or denticulate, at least above the middle. The stellate hairs on the lower leaf surface are of variable density and are consistently stipitate, and the indumentum on the upper leaf surface is much sparser than that of V. wurdackii. Accordingly, the upper leaf surface of V. weberbaueri often appears glabrous. The in- florescence of V. wurdackii with 3-6 primary rays and heavier, more robust rays and peduncle is larger than that of V. weberbaueri with 5-7 primary rays. A significant difference is noted in the bracts subtending the rays of the inflorescence. Those of V. weberbaueri are oblong-lance- olate, retained through anthesis and measure up to 2 cm. long (subtending the primary rays). The bracts of: V. wurdackii are early caducous and at most measure 0.8 mm. long. The flowers of V. wurdackii are usually sessile, very inflorescence rays showing disposition of bracteoles, X 5. E. Flower at anthesis, X 10. F. Style and top of ovary, X 20. G. Corolla spread out, X 10. H. Dense indumentum of stellate hairs from surface of ovary, X 40. I. Young ovary shortly after corolla has dropped, X 10. 558 Rhodora [Vol. 71 Plate 1428 Field photograph of inflorescence of V. wurdackti, + 3⁄5 normal size. (Holotype: J. J. Wurdack 1648). 1969] Viburnum wurdackii — Dudley 559 densely congested, and larger than the short pedicelled flowers of V. weberbaueri. The corollas of V. weberbaueri are glabrous, or at most with few (three or four) stellate hairs, lobes are smooth, or weakly and unevenly papillose. The corolla lobes of V. wurdackii are densely pubescent (of more than 20 intermeshing stellate hairs), and strongly and evenly papillose. The calyx lobes of V. weberbaueri are acute and almost glabrous (except of apical hairs) in con- trast with the rounded or obtuse, densely pubescent and ciliate calyx lobes of V. wurdackii. The ovaries of V. weber- baueri are glabrous or occasionally sparsely pubescent, with few readily displaced furcate and stellate hairs; those of V. wurdackii are densely pubescent with intermeshing stel- late hairs that persist on the fruit at least until half mature. Unfortunately, mature fruit was not present in the speci- mens examined. Only the F. Woytkowski specimen from MO has immature fruit. The duplicate (GH) did not have fruit, nor did the J. J. Wurdack specimens. The plant is named for Dr. J. J. Wurdack, the original collector of the new species who led the Smithsonian Institution Peruvian Expedition in 1962, and is internationally recognized for his investigations of the Melastomataceae. BIBLIOGRAPH Y KiLLIP, E. P. & A. C. SMITH. The genus Viburnum in northwestern South America. Bull. Torrey Bot. Club 56(5): 265-274. 1929. The South American species of Vibur- num. Bull. Torrey Bot. Club 57(4): 245-258. 1930. KILLIP, E. P. in J. F. MACBRIDE. Flora of Peru, part 6, No. 2, Field Museum Nat. Hist. 13: 282-287. 1937. UNITED STATES NATIONAL ARBORETUM WASHINGTON, D.C. 20002 A RANGE EXTENSION FOR SUAEDA AMERICANA IN NEW ENGLAND — Suaeda americana (Pers.) Fern. was found for the first time in Rhode Island on the salt marsh at Jerusalem in September, 1968. According to Fer- nald, this species has a range from Cóte N ord, Quebec, south to southern Maine. Gleason cites it as extending from Que- bec to Nova Scotia and Maine. Several hundred plants of Suaeda americana grew mixed with S. linearis (Ell.) Moq. on the upper levels of the marsh at the base of the dune between the marsh and Point Judith Pond, and extended into lower levels where S. linearis did not grow. Where the two species of Suaeda grew together, each species was distinct with no evidence of intergrading. The tidal marsh at Jerusalem lies between Potter Pond and Point Judith Pond and behind a barrier beach bordering Block Island Sound. When present conditions are compared with a map made in 1839, that part of the marsh where the Suaeda is now abundant was once open pond, the breachway was west of the present location, and the dune was made when the present breachway into Point Judith Pond was dredged. Loose sand still extends from the dune several hundred feet into the marsh. Suaedo. linearis is a common species found on most tidal marshes in Rhode Island. Suaeda americana was probably introduced when the breachway was dredged, possibly in debris clinging to the dredge from a previous work site. A few plants of Chenopodium leptophyllum Nutt., another species new to Rhode Island, were also found in the sandy areas extending from the dune into the marsh, but since the range of this species is from Maine to Virginia, finding it was not as unexpected as finding Suaeda americana. A colony of perennial saltwort, Salicornia virginica L. was also found here and may have been introduced at the same time as the other species named, but while it is rare in Rhode Island, it has been reported previously. Viburnum dentatum L., another relatively rare species, and one which is usually found near the sea shore in Rhode Island was 560 1969] Suaeda americana — Stuckey 561 found growing on an island in the marsh in close proximity to V. recognitum Fern. Voucher specimens dated September 23, 1968 have been deposited in the Herbarium of the University of Rhode Island. IRENE H. STUCKEY DEPARTMENT OF AGRONOMY AND MECHANIZED AGRICULTURE UNIVERSITY OF RHODE ISLAND KINGSTON 02881 NOTES ON RECENT BOOKS Nightshades — The Paradoxical Plants by Charles B. Heiser, Jr., W. H. Freeman and Company, San Francisco. 1969. This anecdotal yet authoritative little book on the Sola- naceae is intentionally aimed at the popular market. The treatment of the genera is rather uneven. Many of the “old wives tales” associated with some of the species are dis- cussed often without seeming resolution. The chapter on Burbank and the *Wonderberry" is unduly labored and prolonged. The discussions of the geographical and genetic origins of cultivated members of the family, along with a useful bibli- ography, are valuable parts of the book from the botanist’s viewpoint. Handbook of North American Nut Trees, Richard A. Jaynes, Editor. Humphrey Press, Geneva, N. Y. 1969. This is a horticultural book and without much doubt has the most authoritative sources of cultural information on nut trees that are now being grown or can be grown in this country. The contributors are specialists in their fields. There is much for the botanist, especially the discussions of natural and extended ranges of our native species and the records of spontaneous hybrids. What may be the final report on Castanea dentata is in- cluded in the chapter on chestnuts by the two greatest au- thorities on this almost extinct and once noble member of our flora. R. B. PIKE UNIVERSITY OF NEW HAMPSHIRE DURHAM, NEW HAMPSHIRE 03824 BOOK REVIEW A MODEL FLORA OF OHIO — “THE MONOCOTYLEDONEAE” by E. Lucy BRAUN. The Ohio State University Press. Columbus, Ohio, 1967. 464 pages. Price $10. Profusely illustrated. Here is another book by the indefatigable Lucy Braun. And, as usual, it is a good one. This imposing volume, the first of a series entitled “The Vascular Flora of Ohio”, ex- tends “from cat-tails to orchids”, ie. Typhaceae through Orchidaceae. The title of the book, and the Bromus illustra- tion on the cover are gilt-edged; so is the arrangement of printing, subject matter, and illustrations. We have as a beginning a “preface” instead of a “foreword” (cf. Fowler’s “Modern English Usage”), and this general good quality continues throughout the book. In this preface we read that “the usability is not only for the student or botanist, but also by the amateur and the fieldworker in any of the nat- ural sciences” and that “the illustrations have been drawn from fresh specimens whenever possible except for a large proportion of the Pondweeds, Rushes, Grasses, and Sedges". In the Introduction, which occupies nine pages, there is a series of maps with comment on them, and comment also on some of the troubles of nomenclature. These maps, as might be expected of Miss Braun’s wide interests, cover Ohio phenomena: January temperatures, length of average growing season, snowfall, geological maps of rock types and of glaciation, and a detailed landform map. Practically all of the native Ohio species are illustrated by fine line drawings. There is a general dearth in the liter- ature of illustrations made from living plants, and such illustrations may help to bridge the gap between Gray’s Manual and Gleason’s “Britton & Brown”. One could wish for more detailed drawings of spikelets in the grasses, espe- cially if drawn from living specimens, sinc» they are such a difficult group for anybody. Even in Carex the drawings are excellent, and Miss Braun in Carex nomenclature for- tunately follows Hermann, who knows Carez, The drawings are especially beautiful in the Liliaceae and Orchidaceae. 562 1969] Book Review — Svenson 563 For each native species there is a distribution map with the counties outlined, as in Deam’s “Flora of Indiana”. It will be interesting to compare these Indiana and Ohio plant maps, and also similar maps in some of the Southern states. This comparison would not do for New England, where the counties are very large. There will always be some differences of opinion as to nomenclature. I would like to see the name Arisaema tri- phyllum reestablished for our common northern species of jack-in-the-pulpit. Even Linnaeus, who was familiar with the Burser Herbarium, concluded that the “Brazilian” speci- men came from Canada, and the subject was long ago admirably treated by Juel, under the title “The French apothecary’s plants in Burser’s Herbarium”, in RHODORA, vol. 33 (1931). Also I should like to see the differences mentioned in the key to Lilium superbum and L. michi- ganense brought out more clearly in the illustrations. 1 am so obtuse that I cannot make out the distinction between the two species, although I have collected both of them. The name “dog’s-tooth lily" for Erythroniwm is a new one. It is far better than the saccharine “fawn lily" and *trout lily", both manufactured names; but lacks the fine historical background and long usage of the real common name “dog’s-tooth violet". The name “violet” was not at all confined to the genus Viola throughout the ages, but in- cluded many flowers with the “ion” or violet coloration of plants which belonged also to the Cruciferae and Li- liaceae, The European “erythro-ion” (i.e. Erythronium) has purplish flowers. The subject together with a discussion of many similar problems, is well taken up by E. A. Lopez in “Apuntes para un concepto del genero y la especie en la Historia de la Botanica” in Anales del Jardin Botanic» de Madrid, vol. 34: 315-355 (1944). One more point: the hazy Eleocharis palustris of Europe should be omitted. It has no real basis in America, but the native E. Smallii is instead founded on identifiable material. See RHODORA 49: 64-65 (1947) and North American Flora 18, pt. 9: 526 (1957). 564 Rhodora [Vol. 71 As evidence to show that he has read the text carefully, the reviewer is expected to come up with at least some tri- fling errors. These are hard to find, and consist of a few misspellings in the index, some of which are reflected in the text. The correct examples are Anthoxanthum pueli, Carex hormathodes, Carex tuckermani, Rhoeo, and Epipac- tis helleborine; also the name is Lumnitzer, and not “Lam- nitzer” under Carex varia, but the error comes from Gray’s Manual. But such trivialities occur in every publication, and they do not really detract any from this admirable piece of work. HENRY K. SVENSON OSTERVILLE, MASS, 02655 ERRATA Page 7, line 2 for subtropicals read subtropical Page 8, 4th line from bottom eliminate “g” Page 25, line 7 for Grisch read Griseb. Page 27, line 34 for Erichitites read Erechtites Page 27, line 34 for hieraceifolia read hieracifolia Page 32, line 1 for brachycarps read brachycarpus Odd pages 37-57, heading for Bromelliaceae read Bromeliaceae Page 123, -line 4 for Fucus read Fucus Page 183, line 5 for Bigeloy read Bigelow Page 419, Fig. 1 caption, line 3 for purpurea read purpurea Page 452, caption Fig. 18 for Elakatothriz read Elaktothrix Page 465, caption Fig. 34 for Glenodinium read Glendodinium Page 465, caption Fig. 39 for Rhodomanas read Rhodomonas VALIDATION OF THE NAME DAHLIA AUSTRALIS Article 33, International Code of Botanical Nomencla- ture, 1966, states, “A new combination . . . is not validly published unless its basionym . . . is clearly indicated and a full and direct reference given to its author and original publication. . . ." Compliance with this article has proved useful in the elimination of errors and confusion. On page 378 of Rhodora 71 (No. 787) I proposed the new combination Dahlia australis failing to make reference to its basionym until two pages hence where it is given under the heading of Dahlia australis var. australis. Mr. Robert M. King noted this and suggested to me such an arrangement might not be in compliance with the code. I agreed that, in the man- ner I had presented it, the basionym was not “clearly indi- cated." In an effort to correct this omission, thereby avoid- ing further confusion and perhaps ultimate invalidation of the name, I have repeated here the essential nomenclatural details of Dahlia australis. Dahlia australis (Sherff) Sorensen, comb. & stat. nov. TYPE: MEXICO: OAXACA: Cerro de San Felipe, 2500 m., 1 Sep. 1897. Conzatti & Gonzalez 543 (Holotype: GH!). Basionym: Dahlia scapigera var. australis et f. australis Sherff, Am. Jour. Bot. 34: 143. 1947. Type: that of Dahlia australis as indicated above. For purposes of possible future competition in considera- tion of the rule of priority (Article 11, 1.c.), the date of publication of the name Dahlia australis is the publication date of this correction. (For additional notes and descrip- tions of D. australis and its infraspecific taxa, see the full treatment in Rhodora 71: 378 ff.) PAUL D. SORENSEN THE ARNOLD ARBORETUM HARVARD UNIVERSITY JAMAICA PLAIN, MASSACHUSETTS 02130 565 STUDIES IN EUPATORIUM (COMPOSITAE) L REVISION OF EUPATORIUM BELLIDIFOLIUM AND ALLIED SPECIES J. L. GRASHOFF AND J. H. BEAMAN! Eupatorium bellidifolium Benth. is the earliest described member of a closely allied complex of herbaceous Mexican and Central American species of section Eximbricata. The complex also includes E. lasium Robins., concerning which Robinson (1923) noted that “This beautifully distinct spe- cies belongs in a small group striking by reason of their loose inflorescence.” Three other species which he included in the group (Robinson 1923; 1934) are E. muelleri Sch. Bip. ex Klatt, E. choricephalum Robins., and E. arsenei Robins. During an examination of various Mexican and Central American species of Eupatorium in the Gray Her- barium and the U. S. National Herbarium a collection from Honduras determined as E. muelleri came to our attention. It appeared to be misdetermined but nevertheless belonged with the complex. In connection with the study of this and several similar specimens, which are here described as a new species, we have found it desirable to review the group as defined by Robinson and to prepare a revision of these species. | Although Robinson’s conclusion that the species are close- ly related seems well justified, they do not appear to be as sharply delimited from other species of Eupatorium as his statement concerning the loose inflorescence might imply. Their relatedness is suggested also by the herbaceous habit, "The project was initiated with support of NSF Grant GE-4051 for Undergraduate Research Participation at Michigan State University and continued while Grashoff was a summer participant in a similar program at the Smithsonian Institution and Beaman held a National Academy of Sciences — National Research Associateship at the Smith- sonian. We appreciate the facilities for the study provided by the curators of the Gray Herbarium, Field Museum, University of Michi- gan Herbarium, and the U. S. National Herbarium. We are grateful to Dr. M. Kabalin for editing the Latin diagnosis. 566 1969] Eupatorium — Grashoff and Beaman 567 generally cordate leaves with dentate or crenate margins, corollas with ampliate throats, the shape, size and pubes- cence of the achenes, and brittle pappus. But these char- acters individually ‘or collectively do not afford a clear distinction of the group. Still other species which are not distantly related include E. pazcuarense H.B.K., E. roth- rockii A. Gray, E. mariarum Robins., E. aschenbornianum Schauer, and E. ciliatum Less. An investigation of all such species, however, is at present beyond practical limits. KEY TO THE SPECIES 1. Pedicels and outer phyllaries glabrous ...... Hem 2 1. Pedicels and/or outer phyllaries pilose or with glandular tri- MIU BER 3 2. Leaves cordate, crenulate, extending well up the stem into the than 7 mm wide, with ca 50 florets ................ 1. E. anchisteum 2. Leaves ovate to orbicular, acuminate, cuneate or truncate, seldom cordate, confined to the lower 1/4 of the plant, or if present above then much reduced; heads 6-9 mm wide, with ca 85-125 florets ............ eene enn 2. E. muelleri 3. Inflorescence branches, pedicels, and outer phyllaries with glandu- lar trichomes; leaves subrosulate; phyllaries linear; heads few on long pedicels .......... eene 3. E. bellidifolium 3. Inflorescence branches, pedicels, and outer phyllaries with non- glandular trichomes, phyllaries lanceolate to ovate; heads few to many, usually on short pedicels ........... mmm 4 4. Leaf blades with nerves branching from above the base, with large white moniliform trichomes on both surfaces; head ca 8 mm wide, with 60-87 florets ........ 4. E. lasium 4. Leaf blades with nerves branching at the very base, often appearing glabrous at least on the lower surface; heads 5-7 mm wide, usually with 30-60 florets .......... 5. E. choricephalum 1. Eupatorium anchisteum Grashoff & Beaman, sp. nov. Type: HONDURAS. MORAZAN: region of El Jicarito, above El Zamorano, 900-1000 m, Paul C. Standley 27471, Nov. 14, 1950 (us 2024848, holotype, MSC photo 9237; GH). Herba perennis erecta, 4-6 dm alta. Caulis gracilis, basi densissime cum sordidis trichomatibus duplo-pilosus, Folia opposita membranacea, superne gradatim parvescentia. Petioli 1.5-3 em longi, duplo-pilosi. Laminae 2.5-5 cm latae, 3.5-5 cm longae, orbiculari-ovatae, acuminatae, cordatae, crenatae ad crenulatae, utrinque sed praecipue inferne secus venas cum tricomatibus moniliformibus grandibus albidis, 3-7 568 Rhodora [Vol. 71 09454 E HONDURAS CP + (buago Natural Hutoarr Mokeum Enplan torig >a keri Sch. sip, Abpwe arw woo rost khàys 5 a is rill 448) WV. Uocky Sine oreat., “reet ' Ca Bon, Heslo witte x Š š Dept Mesue ë oie Ves EL Rascal WIED STATES NATIONAL MUSEUM ico put qe Plate 1429 Figure 1. Eupatorium anchisteum Grashoff & Beaman, holotype. Insert: head and floret from Steyermark 30650 (F). Many of the fragile pappus bristles have fallen away ( 2.2). 1969] Eupatorium — Grashoff and Beaman 569 nervis a brevi-acuminatae basi maxime diffusis. Inflorescentia in paniculam terminalem laxe disposita, ramis subglabratis, pedicellis 7-15 mm longis. Capitula 5 mm alta, cum ca 50 flosculis. Involucra campanulata. Phyllaria ca 20-22, obtusa ad acuta, subaequaliter herbacea, glabra, lanceolato-ovata ad oblanceolato-obovata, scarioso- marginata, ciliata. Corollae ca 2.7 mm longae, albae, ad lobulos pubescentes, lobuli 0.5 mm longi. Achaenia nigra, ca 1.5 mm longa, praecipue secus venas cum trichomatibus bifidis. Setas pappi multae, albae, barbellatae, fragiles, 3 mm longae. Erect perennial herb 4-6 dm tall; stem slender, terete, densely doubly pilose below with grayish trichomes; leaves opposite, mem- branaceous, becoming gradually smaller above; petioles 1.5-3 cm long, doubly pilose; blades 2.5-5 cm wide, 3.5-5 cm long, orbicular- ovate, acuminate, cordate, crenate to crenulate, with white moniliform trichomes on both surfaces and especially along the veins beneath, veins 3-7, branching from the short-acuminate base; inflorescence a loose terminal panicle, the branches subglabrous, pedicels 7-15 mm long; heads ca 5 mm high, 6-7 mm wide, with ca 50 florets; involucre campanulate, phyllaries 20-22, obtuse to acute, subequal, herbaceous, glabrous, lanceolate-ovate to oblanceolate-obovate, scarious margined, ciliate; corollas 2.7 mm long, white, pubescent on and near the lobes, lobes 0.5 mm long, achenes black, ca 1.5 mm long, with bifid trichomes especially along the ribs; pappus of numerous fragile white barbellate bristles 8 mm long. Figure 1. Additional specimens examined, GUATEMALA: CHIQUIMULA: along Rio Tacó between Chiquimula and Montaña Barriol, 3-15 mi NW of Chiquimula, 500-1200 m, Steyermark 30650 (F). JALAPA: mountains along the road between Jalapa and Paraiso, 1400-1700 m, Standley 77248 (F); vicinity of Jalapa, 1360 m, Standley 76538 (F); Cerro Alcoba, just east of Jalapa, 1800-1700 m, Steyermark 32577 (F). SACATEPEQUEZ: near Antigua, 1500-1600 m, Standley 58628 (F). CHIMALTENANGO: Alameda, Johnston 356 (F); along road from Chimaltenango to San Martin Jilotepeque, 1500-1900 m, Standley 57980 (F); near Finca La Alameda, near Chimaltenango, 1830 m, Standley 59096 (F). HONDURAS. MORAZÁN: San Antonio, Rodriguez 1458 (F); El Quebracho above El Zamorano, ca 950 m, Standley 3877 (F); along and near Rio Agua Amarilla above El Zamorano, 1000- 1200 m, Standley 13837 (F); along Rio Agua Amarilla above El Jicarito, 900-1100 m, Standley 27320 (F). EL PARAÍSO: near cumbre of Yuscarán Road, 1450 m, Standley 29294 (F). Irregularities in pollen production in Eupatorium anchis- teum suggest that it may be apomictic. Pollen was not ob- served in any specimens of this species. At anthesis the contents of the anthers resemble oil droplets of varying sizes which apparently lack an exine. Full fruit set never- 570 Rhodora [Vol. 71 theless occurred in many of the specimens. Two previous investigators (Holmgren 1919; Sparvoli 1958, 1960) have demonstrated apomixis in species of Eupatorium in which fruit was set despite abortive pollen. The following specimens of E. anchisteum have pre-an- thesis florets with the anthers devoid of recognizable pollen grains, the contents instead resembling oil droplets: JoAn- ston 356 (F) ; Rodriguez 1458 (F) ; Standley 377 (F), 27320 (F), 76538 (F), 77248 (F); Steyermark 30650 (F). The following specimens have no pre-anthesis florets but an ex- amination of post-anthesis florets revealed “oil-droplets” : Standley 27471 (Msc), 57980 (F), 58628 (F), 59096 (F). Specimens of E. anchisteum have been determined pre- viously as E. muelleri, and it is to this species that E. an- chisteum seems most closely related, The specific epithet anchisteum (Gr. next of kin) is meant to express this rela- tionship. There is no indication of intergradation between the two taxa, however, even though they overlap in range and appear mixed in a single collection. There are very clear morphological differences between them, which we believe justify recognition of the two taxa as species. Although E. anchisteum is most similar to E. muelleri, it also has features in common with E. aschenbornianum and E. choricephalum. From the former it may be distin- guished by the glabrous phyllaries as well as fewer heads on longer pedicels (hence a more lax inflorescence). The pubescence at the base of the stem of E. aschenbornianum is reddish whereas that in E. anchisteum is usually white; thus the stems of E. aschenbornianum appear tawny when dried while those of E. anchisteum have a grayish appear- ance. The densely pilose stem, the presence of large monili- form hairs on the leaves, and the glabrous outer phyllaries of E. anchisteum distinguish it from E. choricephalum. 2. Eupatorium muelleri Sch. Bip. ex Klatt Eupatorium muelleri Sch. Bip. ex Klatt, Leopoldina 20: 90. 1884. Type: MEXICO. OAXACA (?): "Tala." Liebmann 67 (C, holotype, MSC photo 4578; GH). 1969] Eupatorium — Grashoff and Beaman 571 Mallinoa corymbosa Coult. Bot. Gaz. 20: 47. 1895. Type: GUATEMALA. SANTA ROSA: Volcan J umaytepeque, 6000 ft, Heyde & Lux ex J. D. Smith 4255 (F, holotype, Msc photo 9149; US). Erect perennial herb 30-60 cm high; stems terete, striate, purple, rather densely doubly pilose below, becoming glabrous above; leaves opposite, confined to the lower 1/4 of the plant, sometimes subrosulate (upper leaves, when present, much reduced and short-petiolate to subsessile) ; blades 25-80 mm long, 10-60 mm wide, ovate to ovate- rhomboid (occasionally suborbicular), acuminate, truncate to cuneate- acuminate at the base, 3-7 nerved, sparsely pubescent on both surfaces with large moniliform trichomes especially along the nerves below; inflorescence an open, few-headed corymb; heads 4-6 mm high, 6-9 mm wide, with 85-125 florets, on long glabrous pedicels up to 7.5 cm long; involucre campanulate, phyllaries biseriate, subequal, glabrous, ovate, obtuse to acute, scarious margined, ciliate; corollas white, lobes pubescent; achenes 1.5 mm long, hispid especially along the ribs; pappus of brittle white barbellate bristles 3.5 mm long. Specimens examined. MEXICO. JALISCO: km 926-927 on Mex 80 from Guadalajara to Barra de Navidad, Grashoff 188 (Msc); steep moun- tainsides 3-5 road miles above (north of) La Cuesta, road to Talpa de Allende; elevation 950-1000 m, McVaugh 21246 (MICH); cultivated from seed of the preceding, Grashoff 277 (MSC). HIDALGO: Tulan- cingo, ravine near Apulco, 2200 m, Moore 2098 (GH). STATE OF MEX- 1co: Dist. of Temascaltepec, Cajones, 2480 m, Hinton 2874 (US); Tejupileo, 1340 m, Hinton 2710 (F, GH, US). MICHOACAN: Uruapan, 4500 ft, Pringle 10121 (F, GH, MSC, Us). GUERRERO: Mina, Toro Muerto, 2350 m, Hinton et al. 14764 (MICH, US). OAXACA: Sierra de San Felipe, 7500 ft, Pringle 6055 (GH, Msc, US); 16 km al NE de Oaxaca sobre la carratera a Ixtlan de Juarez, Rzedowski 19214a (msc); Cerro de San Felipe, 6000-7000 ft, C. L. Smith 305 (F). OAXACA (?): Tala [Talea ?], Liebmann 67 (c, holotype; GH). CHIAPAS: one mile N of Junction of Mex 195 and Mex 190 on Mex 195, Grashoff 236 (MSC). GUATEMALA. HUEHUETENANGO: along road E of San Sebastián, ca 1800 m, Standley 81454 (F); near crossing of Río San Juan Ixtlán E of San Rafael Pétzal, ca 1730 m, Standley 83060 (F). CHIQUIMULA: along Río Tacó between Chiquimula and Montana Bar- riol, 3-15 miles NW of Chiquimula, 500-1200 m, Steyermark 30641 (F). JALAPA: vicinity of Jalapa, 1360 m, Standley 76696 (F); Mon- taña Durazno, 2 mi E of San Pedro Pinula, 1400-1800 m, Steyermark 32976 (F). CHIMALTENANGO: Alameda, Johnston 493 (F); Barranco de La Sierra, SE of Patzám, ca 2100 m, Standley 61602. SANTA ROSA: Volcán Jumaytepeque, 6000 ft, Heyde & Lux ex J. D. Smith 4255 (F, holotype of Mallinoa corymbosa; US). HONDURAS. COMAYAGUA: vicin- ity of Siguatepeque, 1080-1400 m, Standley 56032 (F). MORAZÁN: 572 Rhodora [Vol. 71 along and near Rio Agua Amarilla, above El Zamorano, 1000-1200 m, Standley 13837. The relationships of E. anchisteum and E. muelleri are discussed under the former species. Like E. anchisteum, E. muelleri appears to be apomictic. The following specimens of E. muelleri have pre-anthesis florets with anthers devoid of recognizable pollen grains, the contents instead resem- bling oil droplets; Grashoff 188 (Msc), 236 (MSC), 277 (MSC) ; Hinton 2710 (F); Pringle 10121 (Msc) ; Standley 77626 (F); Steyermark 30641 (F). The following speci- mens have no pre-anthesis florets but an examination of post-anthesis florets revealed “oil droplets”: Johnston 483 (F) ; Standley 50632 (F), 61602 (F), 81454 (F). One speci- men (Standley 83060, F) contained only post-anthesis florets with anthers which appeared totally empty. When grown in the experimental garden and greenhouse, plants of E. muelleri exhibit a more robust habit than those in the native habitat. They achieve a higher stature (to one meter), have a greater degree of branching, and have leaves distributed farther up the stem. They maintain, however, the large heads in open inflorescences and the characteristic leaf shape of the species. Eupatorium muelleri is very similar to E. lasium, from which it is readily separated only by its glabrous pedicels. In other respects the two species appear to intergrade or to form two extremes of a continuum. Eupatorium lasium usually has a more robust habit with a much-branched in- florescence and a greater number of smaller heads. The specimens now referred to E. lasium may be local isolates of the more widespread E. muelleri, but such a conclusion would have to be reinforced by more extensive collections than are now available. 3. Eupatorium bellidifolium Benth. Eupatorium bellidifolium Benth. Pl. Hartweg. 43. 1840. Type: MEXICO. HIDALGO: mountains near Apuleo, Hartweg 330 (K, holotype, MsC photo 1644). Erect scapose perennial herb 15-35 em tall; stems terete, sparsely doubly pilose below, glandular and pilose above; leaves subrosulate, 1969] Eupatorium — Grashoff and Beaman 573 opposite, petiolate, reddish beneath; blades obovate-orbicular, 1-6 cm long, 0.8-4 em wide, crenate or serrate, base subcuneate-acuminate, with 3-7 nerves branching from a short distance above the base (occasionally appearing to branch directly from the base, particularly in depauperate specimens), pilose with large moniliform trichomes on both surfaces, especially along the veins beneath; inflorescence a diffuse, few-headed corymb, occasionally with but two or three heads; pedicels densely pubescent with both glandular and eglandular tri- chomes; heads with 40-100 florets, 8-10 mm wide, 6-8 mm high; involucre campanulate, phyllaries virescent, lanceolate to linear- lanceolate, acuminate, subequal, 2- (3-) ribbed, 3-nerved, the outer with glandular trichomes; corollas white, 4 mm long, glabrous except on the lobes; achenes dark brown, 1.5-2.0 mm long, hispid especially along the ribs; pappus of brittle white barbellate bristles 3 mm long, sometimes reddish. Specimens examined. MEXICO. SINALOA: Sierra Madre Occidental, in forest at Loberas summit, 6350 ft. Goodman 7420 (US) ; Sierra de la Durango, Cerro de las Cruces, San Ignacio, 2160 m, Montes & Salazar 105 (Us). JALISCO: Sierra del Halo, near lumber road leaving the Colima highway, 7 miles SSW of Tecalitlán and extending south- easterly to San Isidro, 1700-1800 m, McVaugh & Koelz 1285 (MICH). HIDALGO: mountains near Apulco, Hartweg 330 (K, holotype). STATE OF MEXICO: Temascaltepec, Cumbre-Cruz, Hinton et al. 8688 (F, GH, US). MICHOACAN: Zitacuaro, Cerro Aguila, 2750 m, Hinton et al. 13555 (GH, MICH, US). GUERRERO: Cruz de Ocote, ca 25 km WSW of Camolta, Mun. de Chichihualco ca 50-60 km W of Chilpancingo, Fed- dema 2819 (MICH) ; idem loc., Rzedowski 14945 (MICH); Cerro Alqui- trán cerca de Mazatlán, municipio de Chilpancingo, Rzedowski 23663, (msc); Mina, Campo Marado, 2420 m, Hinton et al. 11171 (GH, US); Cerro Azul, 2500 m, Hinton et al. 14945 (MICH, US). OAXACA: near first summit of road about 9 miles NE of Mitla on the road to San Lorenzo, 8000 ft, Ernst & Delgadillo 2228 (us); cultivated from seed of the preceding, Grashoff 276 (Msc); Ixtlan de Juarez, Kruger & Gillespie 44 (GH); Sierra de San Felipe, 8000 ft, Pringle 6056 (GH, MICH, MSC, US); Sierra de Clavellinas, 8000 ft, C. L. Smith 367 (F, us); 16 km NE of Oaxaca, sobre la carretera a Ixtlan de Juarez, Rzedowski 19214 (MICH). CHIAPAS: near Piedracitos 7 km E of San Cristobal las Casas, Mun. of Chamula, 7200 ft, Breedlove 8005 (MICH). GUATEMALA. ZACAPA: between Loma El Picacho and Cerro de Monos, 2000-2600 m, Steyermark 42760 (F). Three specimens were examined for anther content. Two had pre-anthesis florets with anthers completely devoid of any recognizable pollen grains, the contents instead resem- bling oil droplets. These were: Grashoff 276 (MSC), Rzedow- ski 23633 (Msc). Pringle 6065 (MSC) contained only post- 574 Rhodora [Vol. 71 anthesis florets but an examination of the anthers revealed few pollen grains and many “oil droplets". 4. Eupatorium lasium Robins. Eupatorium lasium Robins, Contr. Gray Herb. 68: 21. 1923. Type: MEXICO. MICHOACAN: vicinity of Morelia, vers la Loma de la Huerta, 1950 m, Arsène 5788 (GH, holotype, MSC photo 5980; US). Erect perennial herb 5-6 dm or more tall; roots of 6-12 gray-black fibrils ca 2 mm thick; stems terete, ca 5 mm thick at base, densely doubly pilose below, becoming sparsely and singly pilose above; leaves opposite, long-petiolate, ovate, broadly cordate, short-acuminate, coarsely crenate to obtusely dentate; blades white-pilose on both sur- faces with large moniliform trichomes, 3-7 cm long, 2.7-6 em wide, 3-T-nerved from just above the acuminate base; petioles sometimes winged; inflorescence a lax panicle, bracts foliaceous but much re- duced, pedicels 5-18 mm long, short pubescent; heads about 8 mm high, 8 mm wide, with 60-87 florets; involucre campanulate; phyllaries 23-30, subequal, lanceolate to oblanceolate, obtuse to acute, scarious margined, the apex long ciliate; corollas white, throat and/or lobes pubescent, throat enlarged, cylindrical, lobes ca 1.8 mm long, deltoid; achenes black, ca 1.8-2.2 mm long, hispid especially along the ribs; pappus of thin brittle white barbellate bristles ca 4 mm long. Specimens examined. MEXICO. MICHOACAN: vicinity of Morelia, vers la Huerta, 1950 m, Arséne 5440 (US); Arsóne 5788 (Gn, holotype; US); Loma Santa Maria, 1950 m, Arséne 5867 (US); Cerro San Miguel, 2100 m, Arséne 7272 (us); Cerro Azul, 2100 m, Arséne 7275 (US). 5. Eupatorium choricephalum Robins. Eupatorium choricephalum Robins. Proc. Amer. Acad. 54: 239. 1918. Type: MEXICO. JALISCO: near Guadalajara, Pringle 11824 (GH, holotype; F, MICH, MSC, US). Eupatorium arsenei Robins. Contr. Gray Herb. 104: 10. 1954. Type: MEXICO. MICHOACAN: vicinity of Morelia, Car- rindapaz, Arséne 6080 (GH, holotype, MSC photo 5928; us). Erect or slightly decumbent perennial herb about 1 m tall; stems terete, often purplish, doubly pilose below, singly pilose above; leaves opposite, petiolate, becoming subsessile and reduced above, broadly ovate, acuminate, broadly cordate to truncate at the base, crenate, the base of the blade acuminate at the insertion of the petiole, blade ca 3-7.5 em long, 2.5-6 em wide, thin, sparsely pilose on the veins with small trichomes, appearing glabrous without magnification, veins 3-8, 1969] Eupatorium — Grashoff and Beaman 575 separating at the very base of the blade; infloreseence usually an open panicle but clusters of heads sometimes remote; pedicels filiform, 3-18 (-25) mm long, pubescent; heads 5-6.5 mm long, 5-7 mm wide, with 30-60 (-84) florets; involucre campanulate, phyllaries virescent, subequal, oblong to linear-lanceolate or linear-oblong, obtuse, acute, or often acuminate, 2-ribbed, 3-nerved, scarious margined, ciliate, the outer dorsally pilose; corollas white, throat ampliate, lobes 0.3-0.9 mm long; achenes dark brown, 1.5-2 mm long, hispid especially along the ribs; pappus bristles white, short barbellate, brittle, 4 mm long. Specimens examined. MEXICO. CHIHUAHUA: “South-western Chi- huahua,” Palmer 276 in 1885 (GH, us), Mapula Mts., Pringle TAT (GH, MICH, MSC). SINALOA: San Ignacio, San Juan, Lasbreas, Ortega 6885 (F). DURANGO: city of Durango and vicinity, Palmer 857 in 1896 (F, GH, MSC, US). NAYARIT: mountains 10 miles SE of Ahuacatlan on the road to Barranca del Oro and Amatlan, 1100-1300 m, McVaugh & Koelz 782 (MICH). JALISCO: Barranca de San Juan de Dios ca 15 km E of Pihuamo, eastern slopes of Sierra de los Corales, Mun. de Tecalitlan, 1200-1300 m, Feddema 2201 (MICH) ; Sierra del Tigre, 3 miles S of Mazamitla, 2100-2200 m, McVaugh 13605 (MICH); MeVaugh & Koelz 418 (MicH) ; McVaugh & Koelz 419 (MICH) ; Sierra de la Campana along road to Mascota 7-8 miles NW of Los Volcanes, 1900-2000 m, McVaugh 13692 (MICH) ; steep mountainsides 4.5 miles NNE of Talpa de Allende N of the road, 1450-1500 m, McVaugh 20203 (MICH) ; moun- tains 12-15 miles SSE of Autlan on lumber road to Corralitos, 4-10 miles above (SE) Ahuacapan, 1500-2000 m, MeVaugh & Koelz 998 (MICH) ; in the barranca at Puente San Pedro, 5 miles SW of Tecalit- lan, 1200 m, McVaugh & Koelz 1310 (MICH); Chalapa, Palmer 112 in 1886 (GH, MICH, MSC, US); near Guadalajara, Pringle 11824 (GH, holo- type; F, MICH, MSC, US); Calabazas, Reko 4868 (us); Mata de Bule, Sierra de los Corales, Mun. de Tecalitlan, 1300 m, Rzedowski 17316 (MICH, MSC); Llano Verde, cerca de los Corales, Mun. de Tecalitlan, Sierra de los Corales, 1600 m, Rzedowski 17446 (MICH, MSC). VERA- CRUZ: Orizaba, Botteri 494 (GH). MORELOS: Cuernavaca, Bourgeau 1215 (F, GH); Huitzilac, Lyonnet 819 (us); near Cuernavaca, Pringle 11342 (GH, US). STATE OF MEXICO: Temascaltepec, Rincon, Hinton 129 (US). MICHOACAN: vicinity of Morelia, Cerro San Miguel, 2200 m, Arsène 5302 (us); vers la Huerta, 1950 m, Arsène 5439 (us); Arsène 6619 (us); Carrindapaz, 2200 m, Arsene 6080 (GH, holotype of E. arsenei; US); 5 miles N of Patzcuaro, 6300 ft, Cronquist 9726 (MICH) ; W-facing slopes of Cerro de Carboneras above Rio Cupatizio, ca 22 km S of Uruapan, King & Soderstrom 4809 (us); near Uruapan, 4800 ft, Pringle 10099 (F, GH, MICH, MSC, US). GUERRERO: Mina, Chilacayote- Manzito, 1850 m, Hinton et al. 14936 (us); 2 km N of Taxco, over road to Amacuzac, 1750 m, Rzedowski 21494 (MSC). Eupatorium choricephalum is a widely distributed, vari- 576 Rhodora [Vol. 71 able Mexican species from which E. arsenei cannot be satis- factorily separated. Robinson (1934) distinguished the two species on the basis of the narrower, more acute phyllaries and more diffuse panicle in E. choricephalum. From speci- mens now available, however, it appears that variation in these characters is continuous between the extremes. Notes and sketches preserved in the Gray Herbarium suggest that Robinson used Palmer 857 in 1896 in drawing up his de- scription of E. choricephalum, but he designated as type Pringle 11824 (misprinted 11924). These two specimens differ principally in the size of the crenations of the leaf margins. Eight specimens were examined for anther content. One (Pringle 11824, Msc) had pre-anthesis florets with anthers which contained few “oil droplets” and many pollen grains of irregular sizes and shapes, a high percentage of which were aborted. Five specimens with pre-anthesis florets had anthers which contained few if any pollen grains and many “oil droplets": Palmer 712 in 1886 (Msc); Pringle 747 (MSC); Rzedowski 17316 (Msc), 17446 (Msc), 21494 (Msc). Two specimens had only post-anthesis florets but an examination of the anthers revealed the presence of “oil droplets” indicative of pollen sterility: Palmer 857 in 1896 (MSC) ; Pringle 10099 (Msc). LITERATURE CITED HOLMGREN, J. 1919. Zytologische Studien iiber die Fortpflanzung bei den Gattungen Erigeron und Eupatorium. Kongl. Svenska Vetenskapsakad. Handl. 59: 1-72. ROBINSON, B. L. 1923. Records preliminary to a general treatment of the Eupatorieae. — III. Contr. Gray Herb. 68: 3-43. ROBINSON, B. L. 1934. Records preliminary to a general treatment of the Eupatorieae, — XI. Contr. Gray Herb. 104: 3-49. SPARVOLI, E. 1958. Osservazioni cito-embriologiche in Eupatorium riparium Reg. — Nota I: cariologia. Ann. Bot. (Rome) 26: 120- 127. SPARVOLI, E. 1960. Osservazioni cito-embriologiche in Eupatorium riparium Reg. — Nota II: Megasporogenesi e sviluppo del game- tofito femminile. Ann. Bot. (Rome) 26: 481-504. Il. A NEW SPECIES OF EUPATORIUM (SECTION HEBECLINUM)' J. L. GRASHOFF AND J. H. BEAMAN During an examination of undetermined specimens of Eupatorium at the U. S. National Herbarium, the Field Museum, and the University of Michigan Herbarium, two collections were found representing a previously unde- scribed species which is characterized as follows: Eupatorium ovandense Grashoff & Beaman, sp. nov. Type: MEXICO. CHIAPAS: Mt. Ovando,’ 1000 m, E. Matuda 3922, Nov. 14-18 1939 (us 2422509, holotype, MSC photo 9236; MICH). Herba grandis, suffrutescens. Rami quinquesuleati, fistulosi. Folia glabrata, ovata cum ca 5-11 lobis acutis. Petioli 2.5-7 em longi, glabri, saltem superne alati. Laminae 5-13 em latae, 6-17 cm longae, irreg- ulariter crenatae, utrinque glandibus sessilibus aureis. Inflorescentiae paniculatae, conicae ad subglobosae, ramis viscido-pilosis. Capitula multa. Involuera campanulata, ca 4-5 mm alta, ca 3-4 mm lata. Phyllaria ca 15, pilosa, sessili-glandulosa, 1-2 costata, exteriora lanceo- lata et acuta, 2 mm longa, interiora lineari-acuta, 4.5 mm longa, ciliata. Receptacula reticulato-ciliata. Flosculi ca 15. Corollae ca 4 mm longae, amethysteae, infundibuliformes sine tubis constrictis, superne raro sessili-glandulosae, basi duplo-bulbosae. Lobi breves vix 0.5 mm longi, obtusi. Rami styli 5 mm longi, subclavati. Achaenia columnaria, brunnea, 2-3 mm longa, 5-costata, precipue secus costas hispida. Setae pappi ca 24, albae, barbellatae, 3 mm longae. Large suffrutescent herb; stem 5-grooved, fistulose; leaves ovate with 5-11 acute lobes, glabrate; petioles 2.5-7 cm long, winged (at least above), glabrous; blades 5-13 cm wide, 6-17 cm long, with yellow sessile glands on both surfaces, irregularly crenate; inflores- cence a large conical to subglobose panicle, the branches viscid-pilose, heads numerous; involucre campanulate, ca 4.5 mm high, ca 3-4 mm wide; phyllaries ca 15, pilose and sessile-glandular, 1-2-ribbed, the ‘We appreciate the facilities provided for this study by the curators of the Field Museum, University of Michigan Herbarium, and the U. S. National Herbarium. Dr. M. Kabalin kindly edited the Latin diagnosis. "Mount Ovando is not indicated on the standard maps of Mexico. From the local populace its location was established as the first peak (over 2000 m high) to the NW of Esquintla, Chiapas, in the Sierra de Soconusco. 577 578 Rhodora [Vol. 71 FLORA MEXICANA Matuda Herbarium a on m aks Se Alt. 1000 a, diete STATES ROTOR, ud FEAR UN Sepa terse apandaanse Donsie Biome. Hab a Maiei HOLOTYPE NT "oaa. (ResHorr BED tee - gp. E Miik Plate 1430 Figure 1. Eupatorium ovandense Grashoff & Beaman, holotype. Insert: involucre; phyllaries showing change in size and shape from outer (left) to inner (right); floret with many of the fragile pappus bristles broken off (> 2). 1969] Eupatorium — Grashoff and Beaman 579 outer lanceolate, acute, 2 mm long, the inner linear-acute, 4.5 mm long, ciliate; receptacle reticulate-ciliate; florets ca 15, corollas ca 4 mm long, purple, funnelform without constricted tubes, rarely sessile-glandular above, doubly bulbous at the base; lobes short, scarcely 0.5 mm long, obtuse; style branches 5 mm long, subclavate; achenes columnar, brown, 2-3 mm long, 5-costate, hispid especially along the ribs; pappus of ca 24 white barbellate bristles 3 mm long. Figure 1. Additional specimens examined. Mexico. CHIAPAS: Mt. Ovando, Escuintla, Matuda 16258, November 14, 1945 (F, US). Eupatorium ovandense is evidently related to E. incomp- tum DC. (section Hebeclinum), similarly large, suffrutes- cent herb which occurs in central and southern Mexico and Guatemala. These species may be distinguished as follows: Phyllaries acute, becoming linear within; florets purple, rarely with yellow sessile glands on the lobes; leaves without a prominent veining pattern on the abaxial sur- face ne E. ovandense Phyllaries obtuse, of approximately equal width through- out; florets not purple but with many orange or brown sessile glands throughout the corollas; leaves with a con- spicuous reticulate veining pattern on the abaxial sur- face cccccccssasaccscecascesesosesescessssnsessosessesssnsseensosensnsanantnnansnsnnnnanananenneaia E. incomptum Postscript. While this manuscript was in press a publi- cation by R. M. King and H. Robinson (Brittonia 21: 275- 284. 1969) has proposed transfer of Eupatorium ovandense to the genus Decachaeta. They consider structure of the anther appendage to be the unifying character of Deca- chaeta. Anther appendages are variable in still other species of Eupatorium, however, and may relate to pollina- tion mechanisms (Grashoff and Beaman, Brittonia, in press). Furthermore, Decachaeta sensu King and Robinson would appear to include several discordant elements result- ing in too much diversity for a small segregate genus. We therefore consider E. ovandense and E. incomptum more appropriately left in Eupatorium. DEPARTMENT OF BOTANY AND PLANT PATHOLOGY MICHIGAN STATE UNIVERSITY EAST LANSING 48823 A NEW WEED INTRODUCTION IN NEW ENGLAND, ATRIPLEX SIBIRICA. — Some years ago Prof. Burton N. Gates sent me a plant for identification that was not in- cluded in the eighth edition of Gray’s Manual. It was an Old World type of Atriplex, and so I sent it to Prof. Paul Aellen, of Basel, Switzerland, the chief authority on the Chenopodiaceae. Aellen identified the specimen as Obione muricata Gaertn., a native of Siberia and Turkestan. The status of Obione Gaertn. as a genus distinct from Atriplex L. is still debatable and perhaps a matter of in- dividual preference. Ulbrich, in the second edition of Engler and Prantl’s Die Natiirlichen Pflanzenfamilien (vol. 16c, 1934) recognized it, and Aellen follows him, although with some reservations. The chief, and perhaps only, character is in the fruiting pistillate perianth segments being united to above the middle or throughout, rather than nearly free as in typical Atriplex species such as A. patula L., A. hor- tensis L., and A. rosea L. There are two “Obione” species that occur within the Gray’s Manual range, viz. A. argentea and A. arenaria Nutt. From these Atriplex sibirica L. (Obione muricata Gaertn.) differs in having the fruiting perianth segments united to the apex and strongly spiny on the back throughout. Since the perianth is incurved at the apex, the mature fruit is almost completely enclosed except for a minute apical tube. The following description is translated from Aellen’s description in his paper “Die orientalischen Obione-Arten" (Verhandl. Naturforsch. Gesell Basel 49: 131-137. 1937- 38). Atriplex sibirica L. Sp. Pl. ed. 2, 1493. 1763. Obione muricata Gaertn. Fruct. 2: 198, t. 126, f.5, 1791. Obione sibirica (L.) Fisch. Cat. Hort. Gorenk. 25. 1808. Annual, up to 60 em. high, many-branched, with elongate, glabrate branches; middle leaves mostly alternate, oval- deltoid, up to 7 cm. long (including the petiole 1 cm. long), and 3 em, wide, with simple or double lobe-like teeth below the middle, above the middle with sinuately curved teeth, the apex more or less acute, the base rounded, slightly 580 1969] Atriplex sibirica — Morton 581 decurrent into the petiole, beneath with prominent nerves, gray-pulverulent, above glabrescent; smaller leaves lance- olate, more or less entire; male flowers very few in balls at the end of branchlets; pistillate perianth up to 6 mm. long (including the pedicel), up to 4 mm. broad, stellate- scaly when young, the lower part round, connate upwardly to a small opening, woody in fruit, irregularly tuberculate and echinate, with a pedical 2-3 mm. long, ca. 1 mm. thick at the upper end, inconspicuously ridged from the decurrent nerves; pericarp membranous, more or less adherent; seed entirely enclosed within the woody perianth segments, ca. 2 mm. in diameter, red-brown, with upwardly turned rad- icle and 2 small separated styles. Specimen examined: Dump of Queensbury Combing Company, Auburn, Worcester County, Massachusetts, July 16, 1949, Burton N. Gates 31888 (us), Clark Univ., New England Bot. Club, (GH). It will be interesting to find out if this plant persists. Since it grows naturally in a cold region, it may very well find New England a congenial habitat. C. V. MORTON, SMITHSONIAN INSTITUTION WASHINGTON, D.C. 20560 INDEX TO VOLUME 71 New scientific names and combinations are printed in bold face type Abies 337, 338 Abromitiella 35, 225; 225 Abutilon Theophrasti 171 Acer rubrum 32 Aceraceae 514 Achillea borealis 523; lanulosa 523; Millefolium 20, 27, 514, 523 Aconitum Napellus 170 Acrochaetium | amphiroae attenuatum 132; 132, 527, 528 Acrothrix novae-angliae 153 Adams, Franklin S., Winterbud Production and Function in Brasenia Schreberi 417-432 Aechmea 36, 271; angustifolia 271, 272; boliviana 272; bra- chyclada 271, 272; bromelii- folia 271, 274; Castelnavii 271, 273; distichantha 271, 273, var. Schlumbergeri 274; ellipsoidea 274, 275; inconspicua 272; in- volucrata 274; Kuntzeana 271; tocantina 271, 273 Aesculus octandra 303 Agalinis aphylla 33; purpurea 33 Agardhiella tenera 528 Agaricus flavescens 190; sulphur- eoides 190, 191 Agarum cribrosum 125, 137, 155 Agropyron repens 519 Agrostis alba var. alba 519, var. palustris 22, 519; scabra 22 Ahnfeltia plicata 115, 135 Alaria esculenta 114, 115, 127, 132; musaefolia 127; pylaii 154 Alaska, Additions to the Flora of Saint Lawrence Island 502 Aletris aurea 32; farinosa 32; brevifolia 131; flexuosum lutea 32 Allium tricoccum 168 Alopecurus pratensis 168 Alsophilia 7, 8; acutidens 8; armata 7, 13; bicrenata 10, 13, 15, 16; conjugata 9-12, 14, 16; estelae 8, 10, 16; hirsuta 9, 13, 14, 16; hirta 14; mexicana 8; microdonta 8; myosuroides 8; nesiotica 8, 10, 16; notabilis 8; pansamalana 9, 11, 15; rufa 9, 14, 16; scabriuscula 8, 12, 15, var. guatemalensis 9, 13, var. scabriuscula 8, 12, 15, 16; stipularis 9, 14; strigillosa 9, 13; swartziana 7-10, 13, 16, The Complex 7; trichiata 8, 10- 12, 16, 17; tryonorum 8, 10-13, 16 Althaea rosea 171 Ambrosia artemisiifolia 516, 517, 523; trifida 174, 175 Amelanchier laevis 20, 25 Ammophila breviligulata 19, 22, 540 Amphisphaeriaceae 200 Anaboena circinalis 447-449; flos- aquae 447-449; scheremetievi 448, 449 Anacystis dimidiata 116 Ananas 36, 278; ananassoides 278; comosus 278, 279 Anaphalis margaritacea 307, var. intercedens 308, var. margari- tacea 308, var. subalpina 302, 307 Anchistea virginica 31 Andromeda polifolia 508 Andropogon glomeratus 31 Angelonia biflora 551 Ankistrodesmus falcatus 451, 585 586 var. acicularis 451, var. mira- bilis 451 Anthaenantia villosa 31 Anthostoma pulviniceps 194, 195 Anthoxanthum odoratum 519; puelii 564 Antithamnion americanum 136; boreale 136; cruciatum 136; floccosum 154; pylaisaei 136 Aphanizomenon flos-aquae 448, 449 Aphanothece clathrata 448 Aquifoliaceae 574 Aralia spinosa 478, in St. Louis County, Missouri 478 Arbutus 337, 338 Arctostaphylos uva-ursi 19 Arenaria lateriflora 516, 521; peploides var. robusta 25 Arisaema Stewardsonii 168; tri- phyllum 563 Aristida affinis 31; purpurascens 31; stricta 31 Armillaria decorosa 177, viscidipes 178-180 Artemisia caudata 19, 27; globu- laria 508; Stelleriana 27, 540 Arthrodesmus bulnheimii 461; incus var. extensus 461; ralfsii 461; triangularis var. subtri- angularis 462 Arundinaria gigantea 31 Asarum canadense 172 Asclepias cinerea 33; lanceolata 33; longifolia 33; michauxii 33 Ascocyclus distromaticus 125 Ascomycetes 193 Ascophyllum nodosum 115, 117, 118, 128, 125, 129, 138, 134, 138, f. scorpioides 154 Asparagopsis hamifera 182, 527 Asperococcus echinatus 126 Aster 514; brachyactis 219; du- mosus 33; foliaceus 523; johannensis 523; junciformis 178; Rhodora [Vol. 71 173; lateriflorus 33; linarii- folius 33; nemoralis f. albi- florus 173; novi-belgii 523; paludosus 33; pilosus 33; puni- ceus f. albiflorus 173; reticu- latus 33; sibiricus 506; sp. 27; umbellatus 33, 523 Asteraceae 497 Asterionella 465; formosa 466, var. gracillima 466 Asterocystis ramosa 130 Atriplex 580; arenaria 20, 24, 580; argentea 580; glabrius- cula 520; hortensis 580; patula 580, var. hastata 34, 520; rosea 580; sibirica 580, A New Weed Introduction in New England 580 Avena fatua 168; sativa 168 Baecharis halimifolia 27 Balduina uniflora 33 Baldwin, J. T. Jr., A Seventeenth Century Record for Stewartia 434-438 Bangia ciliaris 131; fuscopurpu- rea 154 Banisteriopsis 538 Barr, Margaret E. and Howard E. Bigelow, Contributions to the Fungus Flora of North- eastern North America V. 177- 203 Basidiomycetes 177 Beaman, J. H. and J. L. Gras- hoff, Studies in Eupatorium (Compositae) I. Revision of Eupatorium bellidifolium and Allied Species 566-576; II. A New Species of Eupatorium (Section Hebeclinum) 577-579 Bent, Forrest C., Chromosome Studies in Habenaria 541-543 Berlesiella 195; echinata 194; nigerrima 193, 195, 196; para- 1969] sitica 194; setosa 194 Bertia parasitica 194, 195 Betonica Alopecurus 537 Betula nana ssp. exilis 508; papyrifera var. cordifolia 301 Bidens cernua 433; comosa 433; cordifolia 358; frondosa 33, 523 Bigelow, Howard E. and Mar- garet E. Barr, Contributions to the Fungus Flora of North- eastern North America V. 177- 203 Billbergia 36, 275; boliviensis 277; cardenasii 275, 271; de- cora 275, 277; Meyeri 275, 276; microlepis 275; velascana 275, 277 Blidingia minima 119, 527 Bolbocoleon piliferum 154 Bolivia, The Bromeliaceae of 35, 220 Boltonia asteroides 33; diffusa 33 Botryoccoceus braunii 451; pro- tuberans var. minor 451-453 Brasenia 417, 419-421, 425, 428, 430, 432; Schreberi 417, 421, 426, 432, Winterbud Produc- tion and Function in 417 Brassica juncea 521 Braun, E. Lucy, A Model Flora of Ohio — “The Monocotyle- doneae" (Book Review) 562- 564 Braya humilis ssp? 505 Bromelia 36, 269; Balansae 270; Hieronymii 270; serra 270 Bromeliaceae 35, 220, of Bolivia 35, 220; subfam Bromelioideae 36; subfam. Pitcairnioideae 35; Tillandsioideae 35, 232 Bromus 562; tectorum 22 Broome, C. Rose, Robert W. Long and Olga Lakela, Some Pre- liminary Statistics of the Flora Index to Volume 71 587 of Southern Florida 495-501 Bryophyllum pinnatum 441 Buchnera floridana 33 Cabomba 417 Cacalia lanceolata 34 Cakile edentula 20, 25, 521 Calamagrostis canadensis 519; neglecta 504; Pickeringii 168 Calendula officinalis 174 Callitriche 513; heterophylla 516, 517, 522; verna 505 Calopogon pallidus 32; pulchellus 32 Calosphaeria 199 Calosphaeriaceae 199, 200 Calothrix confervicola 117; crus- tacea 117; scopulorum 154 Campanula americana 303; diva- ricata 303; lasiocarpa ssp. lasiocarpa 506 Camptosema coriaceum 550 Cape Cod, Glaucium flavum from 540 Capnodiaceae 196 Caprifoliaceae 514 Capsella Bursa-pastoris 521 Capsosiphon fulvescens 118 Cardamine 552; microphylla 505 Carduus austrinus 294, 295; lanceolatus 34; lecontei 34; spinosissimus 34 Carex 105, 507, 562; brunnescens var. sphaerostachya 520; ca- nescens var. canescens 519; comosa 28; crinita var. crinita 520; deflexa 299, 300, 302; Em- monsii 299, 300; glaucescens 31; hormathodes 520, 564; Jacobi-peteri 505, 506; jamesii 303; livida 505, 506; longii 23; lurida 31; misandra 508; neso- phila 508; paleacea 520; palles- cens var. neogaea 300; pyre- naica ssp. micropoda 505; Ramenskii 508; rariflora 508; 588 scoparia 520; silicea 520; sty- losa 505; styloflexa 31; tucker- mani 564; varia 564 Carphephorus pseudoliatris 34 Carpinus caroliniana 198 Carum Carvi 522 Caryophyllaceae 515 Castanea dentata 561 Catopsis gracilis 223 Ceanothus americanus 32 Centaurea montana 174 Centella asiatica 33 Cephalanthus occidentalis 33, 173 Ceramium areschougii 116, 117, 131, 136; elegans 136; rubri- forme 136; rubrum 117, 124, 528, var. pedicellatum 136; strictum 527, 528 Cerastrum arvense 521; vulga- tum 521 Ceratium hirundinella 469, f. robustum 469; seammaneae 161 Ceratophyllum demersum 419 Chaenactis 73, 74 Chaetapiospora rhododendri 200, 201 Chaetomorpha 125, 132; gonium 120 Chaetopteris plumosa 124 Chamaesaracha 65 Champia parvula 527 Champlin, Richard L., A Rhode Island Station for Luminous Moss 305 Chaptalia tomentosa 34 Characium curvatum 451; gra- cilipes 451; limneticum 451-453 Chavica officinarum 438 Chenopodium album 520; lepto- phyllum 560; urbicum 169 Chesapeake Bay Region, A Pre- linary Investigation of the Bethonic Marine Algae of the 524 atrovirens linum 120; 120, mela- Rhodora [Vol. 71 Chimaphila umbellata var. cis- atlantica 166 Chlamydomonas dinobryonis 451- 453 Chlorangium stentorinum 451 Chlorogloea conferta 116 Chondrophora nudata 34 Chondrus crispus 115, 119, 131, 134, 136 Chorda filum 115, 122, 128, 138; tomentosa 154 Chordaria flagelliformis 115, 122, 124, 126, 127, 134, 138 Choreocolax polysiphoniae 133 Chroococcus limneticus 448; mini- mum 448 Chrysanthemum leucanthemum 19 Chrysopsis falcata 19, 27 Chrysosphaerella longispina 463 Cirsium 285, 286, 291, 293; altis- simum 285, 286, 291; arvense 523; carolinianum 285, 286, 292; discolor 285, 286, 291; filipendulum 291, 295; Greenii 295; horridulum 285, 286, 288- 290; lecontei 285, 286, 288; muticum 285, 286, 292; nut- tallii 285, 286, 290; ochrocen- trum var. helleri 296; pumilum 285, 286, 288, 289, subsp. pumilum 286; repandum 285, 286, 288; ser. Altissima 285, 290, 292; ser. Mutica 290; ser. Paniculata 286; ser. Undulata 285; ser. Virginiana 285; sect. Onotrophe 285; smallii 290; subg. Eucirsium 285, 286; sub- sect. Acanthophylla 285, 290; subsect. Campanulata 285, 290; subsect. Odorata 285, 286, 288; terrae-nigrae 285, 286, 291, 295; texanum 285, 293-295; turneri 59; undulatum 285, 293, 295; virginianum 285, 293, 1969] 295; vittatum 285, 286, 289, 290, 293; vulgare 27 Cladonia 19; sp. 21; sylvatica 21 Cladophora albida 154; expansa 154; flexuosa 118, 120, 526, f. densa 526; gracilis 120, 132, 526; rupestris 120; sericea 120, 526 Clathromorphum circumscriptum 133; compactum 153 Claytonia arctica 505 Cleistes divaricata 32 Clematis 198, 319; virginiana 198 Clethra alnifolia 33 Cliftonia monophylla 32 Clitocybe sulphurea 189-191 Clitoria mariana 32 Closteriopsis longissima 452, 453 Closterium acerosum 452; dianae 452; kuetzingii 452, monili- ferum 452 Codiolum pusillum f. longipes 117 Coelastrum microporum 452 Coelopleurum lucidum 522 Colacium arbuscula 452, 453, 462; vesiculosum 452, 453, 462 Coleman, James R. and Lyman B. Smith, Chromosome Num- bers of Some Brazilian Angio- sperms 548-551 Compositae 515, 581; Melampodiinae 581 Conioselinum chinense 514 Conopholis americana 303 Convolvulus arvensis 172; sepium 516, var. sepium 522 Conway, Joan R. and Edward J. Hehre, Glaucium flavum Crantz from Cape Cod 540 Corallina officinalis 116, 123, 125, 133 Coreopsis angustifolia 34; cordi- folia 357, 358; coronata 398; crassifolia 398; georgina [var.] nuda 360, [var.] pruinosa 398; subtribe Index to Volume 71 589 lanceolata 34; major 34; pu- bescens 34 Cornus florida 437 Corylaceae 514 Cosmarium bioculatum 452; con- tractum var. papillatum 452; punctulatum 454 Cosmos 319, 369; bipinnatus 412; concolor 319; diversifolius 319, 411 Cottendorfia Rusbyi 223 Crepis capillaris 174 Crosswhite, Frank B., An Incor- rect Correction 480 Crotalaria breviflora 550 Crotalus lepidus 69 Croton alabamensis 94-103, Dis- junction and Endemism in 94; capitatus 32 Cruciferae 563 Crucigenia rectangularis 454 Cryptomonas erosa 470, var. re- flexa 470; marssonii 470; ovata 470 Cryptospora pulviniceps 194 Ctenium aromaticum 31 Cucurbitaria echinata 194, 195; setosa 194 Cuphea 485, 486, 492, A Re- evaluation of 481; bustamanta 481-483, 485, 486, 489, f. reipublicae 484, 485, var. rei- publicae 485; calaminthifolia 481-483, 487, 489, 490-492; carthagensis 33; corniculata 481, 487, 489-491; debilis 481, 487, 490, 491; imberbis 481, 487, 490-492; koehneana 492; paucipetala 492; reipublicae 482, 484, 485; sect. Heterodon 481, 492; sect. Leptocalyx 481, 492; ser. Bustamantae 481, 482, 492 Cuscuta gronovii 33 Cyathea arborea 13; bicrenata 590 Rhodora 15; hirsuta 14 Cyatheaceae, The Alsophila swartziana Complex 7 Cyclotella bodanica 467, compa 467; glomerata 467; stelligera 467 Cymbella ventricosa 467 Cynoctonum sessilifolium 33 Cyperaceae 62 Cyperus fuscus 433, in Nebraska and South Dakota 433; grayii 23; strigosus 168 Cypripedium acaule 108, f. acaule 168; calceolus 303 Cyrilla racemiflora 32 Cystoclonium purpureum 127, 135 Cystoderma 178; amianthinum var. amianthinum 180, f. rugo- soreticulum 180, var. sublongi- sporum 180; pulveraceum 180 Cystopteris fragilis ssp? 504 Dahlia 309-315, 318, 319, 321, 340, 359, 362, 364, 384, 392, 397, 405, 406; acutiflora 410; apiculata 313, 323, 342, 352, 356, 357, 361; arborea 330; astrantiaeflora 410; atropur- purea 313, 325, 365, 375-377; australis 318, 315, 325, 355, 376, 378-381, 383, 386, 388, 390, 393, 565, Validation of the Name 565, var. australis 313, 316, 379-384, 386, 387, 393, 565, var. chiapensis 313, 325, 379, 380, 384-387, 393, var. lieb- mannii 313, 326, 379, 380, 383, 384, 393, var. serratior 313, 324, 325, 379, 380, 386, 387, 393; barkerae 314, 326, 368, 369, 375, 392-396; bidentifolia 397; brevis 313, 319, 326, 368, 369, 382, 395; cardiophylla 313, 326, 857-359, 376; cervantesii 398; chisholmii 398; coccinea [Vol. 71 309, 310, 314-316, 326, 352, 362-364, 397-399, 401, 405-410, var. coccinea 399, 405, var. gentryi 399, 405, 407, var. palmeri 399, 405-408, var. steyermarkii 399, 405-407; coronata 314, 398; cosmaeflora 352; crocata 397, 398; crocea 398; decaiseana 352; dissecta 318, 315, 334, 340, 348, 348, 349, 351, 355, 382, var. dissecta 318, 328, 340, 343, 344, 349, 351, var. sublignosa 313, 323, 343, 344, 350, 351; dumicola 330; excelsa 313, 323, 327-330, 338, 371, var. anemonaeflora 327; foeniculifolia 313, 323, 340, 343-345, 348; frustranea 398; fulgens 410; gentryi 398, 399; gigantea 410; glabrata 352; gracilis 398; hintonii 313, 326, 372, 373, 375; imperialis 309, 313, 314, 323, 326, 328- 330, 332, 334, 335, 338, 339, 355, 363, 382; juarezii 410; lehmannii 326, 330, 334, 335, var. leucantha 330, 333; line- aris 312, 313, 315, 323, 343- 345, 348; macdougallii 309, 313, 321, 323, 338, 339; maxi- miliana 330, 371; maxonii 330; merckii 313-316, 321, 323, 326, 334, 341, 352, 354-356, 382, 393, 411; mollis 313, 324, 361, 373- 375; moorei 313, 322, 324, 372; pinnata 310, 313, 314, 316, 321, 325, 352, 359, 360-365, 368, 369, 376, 411, var. cervantesii 398, var. coccinea 398, var. gracilis 398, var. nana 359, var. varia- bilis 360; platylepis 410; pope- novii 398; pteropoda 313, 324, 367, 368; pubescens 368, 369, 395, 396, 410; purpurea 360, 410; purpusii 313, 326, 358, 1969] 359; pusilla 411; repens 411; Revision of the Genus 309, 367; rosea 310, 362, 411; royleana 411; rudis 318, 324, 365, 369, 370, 373, 376; rupicola 313, 315, 316, 323, 340, 343-348; sambucifolia 359; scapigera 309, 314, 319, 326, 334, 342, 355, 382, 388, 390-394, 396, var. arsenii 369, 393-395, var. australis 381, 382, 393, 565, f. australis 381, 382, 393, 565, f. purpurea 381, 382, 393, var. liebmannii 383, 393, var. scapi- gera 393, f. merckii 352, 355, 393, f. serratior 386, 393, f. scapigera 393, var. typica f. typica 391; scapigeroides 313, 323, 341-344, 357, var. apicu- lata 342, 356, 357, var. typica 341; sect. Dahlia 313, 322, 323, 352; sect. Entemophyllon 313, 322, 323, 340, 341, 348, 357; sect. Epiphytum 313, 322, 323, 338; sect. Pseudodendron 313, 322, 326, 410; sect. Neocaulon 352; sherfñi 314, 325, 376, 379-382, 387-390, 393; sphon- dyliifolia 411; superflua 360; tenuicaulis 313, 323, 329, 335- 339; tenuis 314, 326, 396; vari- abilis 312, 314, 360, 362, 411, var. excelsa 327; zimapana 411 Dasya pedicellata 528 Daucus carota 20, 26 Dawes, Clinton J., Arthur C. Mathieson and Harold J. Humm, Contributions to the Marine Algae of Newfound- land 110-159 Decachaeta 579 DeFilipps, R, Parasitism in Ximenia (Olacaceae) 439-443 Deianira erubescens 548, 550 Delamarea attenuata 126 Index to Volume 71 591 Delphinium brachycentrum 505 Dermoloma 184 Desmarestia aculeata 126, 136, 137; media 153; spp. 155; viridis 126 Desmidium baileyi 454; swartzii 454 Desmodium canadense 108 Deschampsia caespitosa var. glauca 506; flexuosa 519 Deuterochonia 35, 224; longi- petala 224; strobiifera 224, var. inermis 225 Dianthus armeria 25 Diceras phaseolus 463 Diehromena colorata 31; latifolia 31 Dieranum scoparium 21 Dietyosiphon ekmanii 126; foeni- culaceus 115, 124, 126, 127, 138 Dictyosphaerium ehrenbergianum 454; pulchellum 454 Dimorphococcus lunatus 454 Dinobryon bavaricum 448, 449, 463, 465; crenulatum 463-465; cylindricum 448, 449, 451, 463, 465; divergens 463, 465; sertu- laria 465; suecicum 464, 465; vanhoeffenii 464, 465 Diodia teres 33; virginiana 33 Dothidea episphaeria 193, 195 Dothideales 196 Draba 105, 507; alpina 505; borealis 505; hirta 505; macro- carpa 505; nivalis 505; ramo- sissima 303 Dracocephalum parviflorum 172 Drosera brevifolia 32; capillaris 32; filiformis 32; intermedia 32; rotundifolia 20, 25 Dryopteris fragrans 504; spinu- losa var. americana 514; Thelypteris var. pubescens 22 Dudley, T. R., Viburnum wur- dackii, a new Peruvian Species 592 554-559 Dumontia incrassata 154 Dyckia 35, 230; boliviensis 231; ferox 231; gracilis 231, 232; leptostachya 230, 231; Mezeana 231; pulquinensis 230, 231 Eastman, Lesley M., Triphora trianthophora (Sw.) Rydb. in Maine 509 Echinodorus subulatus 550 Ectocarpus arctus 122; confer- voides 122; fasciculatus 122; siliculosus 122; tomentosus 122 Edwards, Maurice E., Iris verna L.: Lectotype and Variety Description 212-214 Elachista fucicola 117, 118, 125; lubrica 125 Elaktothrix gelatinosa 454; viri- dis 452-454 Eleocharis halophila 23; palustris 563; parvula 23; Smallii 563; tuberculosa 31 Elephantopus elatus 34 Eleuterius, L. N. and S. B. Jones Jr., A Floristic and Ecological Study of Pitcher Plant Bogs in South Mississippi 29-34 Elodea Nuttallii 167, 419 Elymus arenarius var. villosus 519; virginicus 519, var. halo- philus 22 Empetraceae 514 Enchnoa 199; subcorticalis 198, 199 Enteromorpha clathrate 526; erecta 118, 125; groenlandica 118; intestinalis 116, 118, 527, 528; linza 119; micrococca 119; minima 119, 527; plumosa 119; prolifera 119, 527; spp. 119, 127 Entocladia wittrockii 117 Entophysalis conferta 116; deusta Rhodora [Vol. 71 116; granulosa 116 Epilobium anagallidifolium 505; angustifolium 505; coloratum 171; glandulosum var. adeno- caulon 217, 522; palustre 505 Epipactis Helleborine 169, 564 Epipyxis tabellariae 464, 465; utriculus 465 Equisetaceae 574 Equisetum arvense var. boreale 167; palustre 504; scirpoides 504 Eragrostis 480; domingensis 479, new to the United States 479; spectabilis 31 Erechtites hieracifolia 27, 34 Ericaceae 514 Erigeron canadensis 34; philadel- phicum 34; quercifolius 34; strigosus 34; vernus 34 Eriocaulon compressum 31; de- cangulare 31; lineare 31 Eriophorum callitrix 508; Scheu- chzeri var. Scheuchzeri 524; vaginatum ssp. vaginatum 504; virginicum 23 Eritrichium aretioides 506 Eryngium ebracteatum 548, 550; integrifolium 33; yuecifolium var. synchaetum 33 Erythronium 563 Erythrotrichia carnea 131 Erythroxylon coca 438 Euastrum didelta 454; pulchellum 454 Eudesme virescens 124; zosterae 125 Eudorina elegans 454 Euglena spirogyra 462 Euonymus atropurpureus 303 Eupatorium 566, 570, 577, 579; A New Species of 577; album 34; anchisteum 567-570, 572; arsenei 566, 574, 576; aschen- bornianum 567, 570; bellidifo- 1969] lium 566, 567, 572, Revision of 566; capillifolium 34; chorice- phalum 566, 567, 570, 574-576; ciliatum 567; coelestinum 34; fistulinum 197; hyssopifolium 34; incomptum 579; lasium 566, 567, 572, 574; leucolepis 34; mariarum 567; muelleri 566, 567, 570, 572; ovandense 577- 579; pazeuarense 567; roth- rockii 567; rotundifolium 34; rugosum 173, f. villicaule 173; sect. Eximbricata 566; sect. Hebeclinum 577, 579 Euphorbia chamaesyce 32; corol- lata 32; maculata 32; polygoni- folia 26; torrida 484 Euphorbiaceae 497 Euphrasia americana 172, 522; canadensis 172, 522; Randii 522 Euthora cristata 153, 154 Eutypa 195 Exocarpus 441 Fabaceae 497 Farmer, Joe A. and Joab L. Thomas, Disjunction and En- demism in Croton alabamensis 94-108 Festuca altaica 504; rubra 20, 22, var. rubra 519 Ficus laurifolia 438 Forida, Some Preliminary Statis- tics of the Flora of Southern 495 Fosterella 35, 220; albicans 220, 221; gracilis 220, 223; gram- inea 220, 222; Pearcei 220; penduliflora 220, 223; Rusbyi 220, 221; schidosperma 220, 223; villosula 220, 222; Wed- delliana 220, 221 Fragilaria capucina 467; croton- ensis 467 Fucus distichus ssp. distichus 130, ssp. edentatus 130, ssp. Index to Volume 71 593 evanescens 130; miclonensis 154; spiralis 154; spp. 123, 125; vesiculosus 115, 118, 125, 130- 132 Fuller, Stephen W. and Arthur C. Mathieson, A Preliminary Investigation of the Benthonic Algae of the Chesapeake Bay Region 524-534 Fumaria officinalis 170 Furcellaria fastigiata 154 Galium Aparine 523; Brandegei 506; erectum 173; Mollugo 173; palustre 20, 27; ssp. 59; tinc- toriura 523 Gaultheria procumbens f. subor- biculata 172 Gaylussacia dumosa 33 Gelsemium sempervirens 33 Gentiana auriculata 506 Georgia 321; variabilis 321 Georgina 311, 312, 321; astran- tiaeflcra 411; bipinnata 412; cervantesii 398; coccinea 311, 312, 397-399, a coccinea 311, 397, B crocea 311, 397, y flava 311, 397; crocata 398, 410; frustranea 312, a coccinea 397, B crocea 397, Y flava 397; petroselinifolia 412; purpurea 321, 359, 360; scapigera 390; superflua 312, 360, a rubra 360, B purpurea 360, Y lilacina 360, 8 pallida 360, e flavescens 360; variabilis 311, 312, 360, 362, [var.] lilacina 311, 360, [var.] pallida 811, 360, [var.] pur- purea 311, 360 Gibberidea setosa 194 Giffordia granulosa 122; 122, 132; sandriana 122 Gigartina stellata 131, 135 Glaucium flavum 540 Glaux maritima 522 Glendodinium aciculiferum 464, 465, 469 ovata 594 Gloeobotrys limneticus 463-465 Gloeocystis ampla 454; gigas 455; scopulorum 117; vesiculosa 455 Gloeothece linearis 448 Gloeotrichia echinulata 448, 449 Gloiosiphonia capillaris 134 Gnaphalium obtusifolium 27; syl- vaticum 173; uliginosum 523 Gomphosphaeria 462; aponina 116, 450, var. delicatula 450; lacustris 450, var. compacta 450 Gonatozygon aculeatum 455; monotaenum 455; pilosum 4 Goniotrichum alsidii 131 Gonyaulax palustris 469 Gracilaria verrucosa 154, 528 Graham, Shirley A., A Re-evalu- ation of Cuphea, sect. Lepto- calyx, Series Bustaraantae 481- 493 Gramineae 515 Grashoff, J. L. and J. H. Bea- man, Studies in Eupatorium (Compositae), I. Revision of Eupatorium bellidifolium and Allied Species 566-576, II. A New Species of Eupatorium (section Hebeclinum) 577-579 Gruendling, Gerhard K. and Ar- thur C. Mathieson, Phytoplank- ton Flora of Newfound and Winnisquam Lakes, New Hampshire 444-477 Guignardia rhodorae 201 Guzmania 36, 235, 267; calothyrsa 267, 268; complanata 250; lin- gulata 267; Melinonii 267, 268; obtusa 266; retusa 267, 268; Roezlii 267; tricolor 245 Gymnodinium fuscum 469; palus- tre 469 Gymnospermae 514 Gymnozyga moniliformis 455 Habenaria X Andrewsii 542; blephariglottis 32, 542; Chro- Rhodora [Vol. 71 mosome Studies in 541; ciliaris 32; clavellata 542; cristata 32; dilatata 542; fimbriata 542; flava 542; Hookeri 542; hyper- borea var. huronensis 542; lacera 542; mivea 32; orbicu- lata 542; psycodes 542; viridis var. bracteata 542 Halopteris scoparia 124 Halosaccion ramentaceum 135 Hamamelis virginiana 32 Haplospora globosa 124 Harriman, Neil A., Magnolia tripetala L. and Aralia spinosa L. in St. Louis County, Mis- souri 478-479 Harris, Stuart K. 493, Notes on the Flora of Coós County, New Hampshire 166-174 Harvill, A. M. Jr., Isotria medeo- loides on the Piedmont of Vir- ginia 303-304 Hedyotis australis 218; crassi- folia 218; rosea 218; spp. 59 Hedysarum alpinum ssp. ameri- canum 505 Hehre, Edward J. and Joan R. Conway, Glaucium flavum Crantz from Cape Cod 540 Heiser, Charles B. Jr., Night- shades — The Paradoxical Plants (Review) 561 Helenium amarum 34; nudiflorum 174; vernale 34 Helianthus angustifolius 34; di- varicatus 34; heterophyllus 34; lactiflorus 174; radula 34; tu- berosus 174 Helicodea 275 Hemisphaerales 197 Heracleum maximum 300 Herpotrichiellaceae 195, 196 Heterotheca graminifolia — 34; mariana 34; nervosa 34; spp. 59 Hibiseus aculeatus 33 1969] Hidalgoa 319 Hierochloe pauciflora 508 Hildenbrandia prototypus 134 Hodgdon, A. R., Added Note on Edith Scamman 165; and R. B. Pike, Continued Botanizing on the Wolf Islands, New Brunswick 297-302; and Rad- cliffe B. Pike, Floristic Com- parison of Three Bird Islands in the Gulf of Maine 510-523; R. B. Pike and Harry R. Tyler, Jr, More Additions to the Flora of Grand Manan, New Brunswick 306-308 Homostegia parasitica 194 Hordeum jubatum 519 Hudsonia 19, 21, 23; tomentosa 26 Hultén, Eric, Flora of Alaska and Neighboring Territories: A Manual of the Vascular Plants (Book Review) 104 Humaria hemisphaerica 202 Humm, Harld J., Arthur C. Mathieson and Clinton J Dawes. Contributions to the Marine Algae of Newfound- land 110-159 Hura crepitans 439 Hyalotheca dissiliens 435 Hybanthus concolor 303 Hydrocotyle umbellata 33 Hygrophorus angelesianus 184; conicoides 182; conicus 182, f. 180-182; cossus 182; fibrillosus 182-184; marginatus var. con- color 184, var. olivaceus 185; recurvatus 185; schulzeri 185, 186; tephroleucus var. tephro- leucus 186 Hypericum boreale 26; cistifolium 33; denticulatum 33; fascicu- latum 33; stans 33; virginicum 21, 26 Hypoxis micrantha 32 Imdex to Volume 71 595 Hyptis alata 33 Ilex 198; coriacea 32; glabra 32; verticillata 26; vomitoria 32 Impatiens capensis 522 Ipomoea Careltoni 175; longifolia 175; pandurata 175; Shumar- diana 175 Iris 212; prismatica 213; verna 212-214, Lectotype and Variety Description 212, var. Smalliana 213, 214, var. verna 213; versi- color 20, 21, 24, 514, 520 Isotria medeoloides 304 Itea virginica 92 Jaynes, Richard A., Chromosome Counts of Kalmia Species and Revaluation of K. polifolia var. microphylla 280-284; ed., Hand- book of North American Nut Trees (Review) 561 Jones, S. B. and L. N. Eleuterius, A Floristic and Ecological Study of Pitcher Plant Bogs in South Mississippi 29-34 Juncus acuminatus 23; biflorus 32; brachycarpus 32; bufonius 515, 520, var. halophilus 520; canadensis 32; effusus 21, 23, 32; elliottii 32; Fauriensis 505, 506; gerardi 23, 25; greenei 23; marginatus 32; scirpoides 22; sp. 505, 506; validus 32 Juniperus communis 197, var. de- pressa 167; virginiana 22 Justicia 548; leucophloea 548, 551 Kalmia 280, 283; angustifolia 19, 282-284; cuneata 281, 283, 284; glauca 280; hirsuta 281-284; latifolia 280, 282-284; micro- phylla 280-284; polifolia 280- 284, var. microphylla 280, 284, Revaluation of 280, var. poli- folia 280, 282, 284, var. ros- marinifolia 280, X microphylla 281, 282 596 Kansas, Additions to the Flora of 175 Kirchneriella contorta 452, 453, 455; lunaris 455, var. dianae 455; obesa var. aperta 455 Kobresia simpliciuscula 504 Koch, Rudy G., Additions to the Kansas Flora 175-176 Kylinia alariae 132; compacta 132; hallandiea 132; monili- formis 132; secundata 132; vir- gatula 132 Labiatae 514 Lachnanthes caroliniana 32 Lachnocaulon anceps 31; minus 31 Lafoensia pacari 548, 550 Lakela, Olga, Eragrostis domin- gensis (Pers.) Steud. new to the United States 479-480; Robert W. Long and C. Rose Broome, Some Preliminary Statistics of the Flora of Southern Florida 495-501 Laminaria agardhii 129, 153; digitata 128, 129; groenlandica 129, 153; longicruris 154; ni- gripes 129; saccharum 129 Laminariocolax tomentosoides 154 Lamium maculatum f. lacteum 172 Laphamia cernua 76; cinerea 90; quinquefolia 78; rupestris 58, 74, 81, 83 Lathyrus japonicus 19, 20, 26, 516, 522; palustris 522 Leathesia difformis 125 Lechea maritima 19, 26 Leguminosae 515 Lemna minor 168 Leontodon autumnalis 523 Lepidium virginicum 25 Leptonema fasciculatum 154 Leptophytum laeve 133 Rhodora [Vol. 71 Leptosphaeria 198; macrospora 196, 197; senecionis 197 Lepuropetalon spathulatum 217 Lespedeza capitata 32 Liatris aspera 34; elegans 34; gracilis 34; spicata var. res- inosa 34 Ligusticum | mutellinoides 506; scothicum 514, 522 Liliaceae 514, 562 Lilium catesbaei 32; michigan- ense 563; superbum 563 Limonium carolinianum 27 Linaria canadensis 27; vulgaris 27 Lindmania gracilis 223; graminea 222; Pearcei 220; penduliflora 223; Rusbyi 221; villosula 222; Weberbaueri 223; Weddelliana 221 Linnaea borealis 508 Linum striatum 32; virginianum 32 Lippia nodiflora 33 Liquidambar styraciflua 32 Lithothamnium glaciale laeve 133; tophiforme 133 Loasa 548; rupestris 548, 550 Lobelia floridana 33; glandulosa 33; puberula 33 Lolium multiflorum var. diminu- tum 168; perenne 167 Long, Robert W., Olga Lakela and C. Rose Broome, Some Pre- liminary Statistics of the Flora of Southern Florida 495-501 Lonicera sempervirens 33 Lophiola americana 32 Lorinseria areolata 31 Lotus corniculatus 171 Ludwigia palustris 20, var. amer- icana 26; hirtella 33; pilosa 33 Lunatia heros 117 Lupinus polyphyllus 170 Luzula arctica 505; multiflora Ssp. multiflora 505 133; 1969] Lychnis alba 306; Flos-cuculi 169 Lycopodiaceae 514 Lycopodium adpressuri 31; al- opecuroides 31; alpinum 504; carolinianum 31; clavatum ssp. monostachyon 504; inundatum 20, 22; prostratum 31; Selago 256 Lycopus americanus 20, 27; ru- bellus 33; uniflorus 522 Lyngbya aestuarii 154; confer- voides 116 Lyonia ligustrina 26, 33 Lysimachia punctata 166, 172; quadrifolia 307 Lythrum Salicaria 171 Magnolia tripetala 478, in St. Louis County, Missouri 478; virginiana 32 Maine, Floristic Comparison of Three Bird Islands in the Gulf of 510; Triphora triantho- phora in 509 Malachodendron 437; ovatum 437 Mallinoa corymbosa 571 Mallomonas acaroides 465; ele- gans 465; fastigiata 466; pro- ducta 466; pseudocoronata 466; tonsurata 466 Malva moschata f. heterophylla 171, f. laciniata 171, f. mos- chata 171 Malvaceae 497 Manisuris cylindrica 31 Massachusetts, The Flora of Monomoy Island 18 Mathieson, Arthur C. and Ste- phen W. Fuller, A Preliminary Investigation of the Benthonic Marine Algae of the Chesa- peake Bay Region 524-534; and Gerhard K. Gruendling, Phy- toplankton Flora of Newfound and Winnisquam Lakes, New Hampshire 444-477; Clinton J. Index to Volume 71 597 Davis and Harold J. Humm, Contributions to the Marine Algae of Newfoundland 110-159 Matricaria matricarioides 523 Melampyrum 441 Melandrium affine 505 Melanthium virginicum 32 Melosira 471; ambigua 467; dis- tans 467; italica 467, subsp. subarctica 468, var. tenuissima 448, 449, 468 Melothria pendula 175 Membranoptera alata 137; denti- culata 137 Mentha arvensis var, 307, var. villosa 307 Meridion circulare var. constric- tum 468 Merismopedia punctata 450; tenuissima 450 Mertensia maritima 516, 522 Micrasterias radiata 456; radiosa 456 Microcoleus tenerrimus 116 Microthyriaceae 197 Microthyrium epimyces 195 Minuartia arctica 505; obtusiloba 505 Mississippi, A Floristie and Eco- logical Study of Pitcher Plant Bogs in South 29 Missouri, Magnolia tripetala and Aralia spinosa in St. Louis County 478 Monarda media 172 Moneses uniflora 171 arvensis Monostroma fuscum f. blyttii 119; grevillei 119; oxyspermum 119, 527; pulchrum 119 Morton, C. V., A New Weed In- troduction in New England, Atriplex sibirica 580-581 Mougeotia sp. 456 Moul, Edwin T., Flora of Mono- 598 moy Island, Massachusetts 18- 28 Muhlenbergia expansa 31 Mycena citrinomarginata 187 Myosotis sylvatica 172 Myrica cerifera 32; inodora 32; pensylvanica 19, 20, 24, 32 Myricaceae 514 Myrionema strangularis 125 Myriotrichia filiformis 127 186, Nasturtium montanum 552 Nebraska, Cyperus fuscus in South Dakota and 433 Nephrocytium ecdysiscepanum 456; limneticum 452, 453, 456; lunatum 452, 458, 456 Neviusia alabamensis 100 New Brunswick, Continued Bot- anizing on the Wolf Islands 297; More Additions to the Flora of Grand Manan 306 New England, A New Weed In- troduction, Atriplex sibirica, in 580; The Phytogeography of Subalpine Black Spruce in 1 New Hampshire, Notes on the Flora of Coós County 166; Phytoplankton Flora of New- found and Winnisquam Lakes 444 Newfoundland, Contributions to the Marine Algae of 110 Nicotiana glauca 548, 551 Nitschkea flageoletina 195 North America, A Definitive New Book on the Flora of North- western 104; Contributions to the Fungus Flora of Northeast- ern 177 Nymphaea odorata 25 Nyssa sylvatica 33, var. biflora 33 Rhodora [Vol. 71 Obione 580; muricata 580; sibir- ica 580 Oedogonium 451, 463; spp. 456 Oenothera fruticosa 33 Oklahoma, The Vascular Flora of — Additions and Corrections 215 Olax 439, 441 Olson, Willis A. and Gerald B. Ownbey, Cytotaxonomie Notes on the Species of Cirsium Na- tive to the Southeastern United States 285-296 Omphalophyllum ulvaceus 154 Onagraceae 514 Onoclea sensibilis 20, 22 Oocystis borgei 456; elliptica 456; lacustris 456; parva 456 Ophiocytium capitatum 463 Ophioglossum crotalophoroides 216; engelmanni 216; vulgatum 216 Orchidaceae 497, 514, 562 Oscillatoria granulata 450; lim- netica 450 Osmunda cinnamomea 21, 22, 31, 514, 519, f. frondosa 167; re- galis var. spectabilis 31 Ostrya virginiana 303 Ownbey, Gerald B. and Willis A. Olson, Cytotaxonomie Notes on the Species of Cirsium Native to Southeastern United States 285-296 Oxypolis filiformis 33 Pachythyrium parasitica 194 Panax quinquefolius 303; trifo- lium 171 Panicum aciculare 31; commuta- tum 31; dichotomiflorum var. geniculatum 168; lanuginosum 31; leucothrix 31; longiligula- tum 31; miliaceum 168; repens 31; sphaerocarpon 31; ver- rucosum 31; virgatum 23. 31: 1969] wrightianum 31 Pantoneura baerii 153 Pappothrix 58, 59, 61, 74; cernua 76; cinerea 90; quinquefolia 78; rupestris 82 Parsonia bustamanta 484 Paspalum difforme 31; floridanum 31; notatum 31; urvillei 31 Pediastrum araneosum 456; bory- anum var. longicorne 456; du- plex 456, var. clathratum 457, f. gracilimum 457; simplex var. duodenarium 457; tetras 457, var. tetraedon 457 Pedicularis sudetica ssp. albola- biata 506 Pelargonium 441; hortorum 439, 440 Penstemon tubaeflorus 480; tubi- florus 480 Percursaria percursa 119 Peridinium cinctum 469; incon- spicuum 464, 465, 469; limba- tum 464, 465, 469; willei 464, 465, 469; wisconsinense 469 Perityle 58, 60, 61, 68, 75; an- custifolia 65, 92; bisetosa 89; cernua 60, 61, 65, 66, 68-73, 75- 77, 79, 81; cinerea 60, 61, 65, 66, 69-75, 79, 90-92; incana Ops quinquefolia 61, 65, 66, 68, 69, 72, 78, 75, 77-81, 85; Parryi 65; rupestris 60, 61, 66, 68-70, 73, 80, 81, 85-88, var. albiflora 63, 65, 66, 72-75, 79, 81, 84, 85, 87, 89, var. rupestris 62, 65, 66, 68, 72, 74, 75, 79, 81, 82, 85, 87, 91; sect. Laphamia 61, 65, 68, 75, 92; sect. Pappo- thrix 58-61, 64-66, 68-75, 77, 79, 88-90; sect. Perityle 61, 65, 75, staurophylla 58; vitreomon- tana 61, 65, 66, 68, 69, 71-75, 79, 88-91 Persea borbonia 32 Petalonia fascia 127, 132, var. Index to Volume 71 599 zosterifolia 127 Petalostemum pinnatum 32 Peterson, Roger Tory and Mar- garet McKenna, Field Guide to Wildflowers of Northeastern and North Central North America (Book Review) 107 Peyssonnelia rosenvingii 154 Pezizales 201 Phalaris canariensis 168 Phleum pratense 519 Phlox carolina 33; 303; pilosa 33 Phragmites communis 19, 23 Phycodrys rubens 137 Phyllachora episphaeria 193 Phyllodoce caerulea 506 Phyllophora brodiaei 135; mem- branifolia 135 Phymatolithon compactum 133; laevigatum 154; lenormandi 154; rugulosum 133 Physalospora thistletonia 201 Phytolacea americana 20, 25, 32 Picea brevifolia var. semipro- strata 1, 2; mariana 1, f. semi- prostrata 2, var. brevifolia 2 var. semiprostrata 1-6 Pike, R. B., Notes on Recent Books 561 Pike, Radcliffe B. and A. R. Hodgdon, Continued Botanizing divaricata on the Wolf Islands, New Brunswick 297-302; Floristic Comparison of Three Bird Islands in the Gulf of Maine 510-523; and Henry R. Tyler, Jr, More Additions to the Flora of Grand Manan, New Brunswick 306-308; Notes on Recent Books 561 Pilayella littoralis 123 Pilinia lunatiae 117 Pinguicula caerulea 33; lutea 33 Pinkley, Homer V., Etymology of Psychotria in View of a New 600 Use of the Genus 535-540 Pinus 337, 338; elliottii 31; pal- ustris 31; rigida 20, 22; sylves- tris 21, 22, 167; taeda 31 Piper chaba 439 Pitcairnia 35, 225; Bangi 45; biattenuata 228; Brittoniana 226, 230; Cardenasii 226, 229; consimilis 46; crassa 226, 229; divaricata 226, 227; ferruginea 46; grandiflora 44; inermis 226, 228; lanuginosa 226, 229; mira- bilis 50; multiramosa 226; nana 54; odontopoda 226, 227; panic- ulata 226, 228; platystemon 226; rigida 229; robusta 43, 44; sessiliflora 229, 230; subpetio- lata 229; Trianae 226, 228 Plantago juncoides 515, var. deci- piens 523; major 523; oligan- thos 21, 27, 523 Platanus 544, 546; X acerifolia 546; mexicana 546; New Chro- mosome Counts in Ulmus and 544; occidentalis 546, var. gla- brata 546; orientalis 546; race- mosa 546; wrightii 547 Platonia bairdii 135 Pleosphaeria echinata 194 Pleospora 197; nigerrima 193; vitalbae 196, 198 Pleosporaceae 196, 197 Pleurotaenium trabecula 457 Pleurotus 190 Plicaria trachycarpa 201 Pluchea purpurascens 20, var. succulenta 28; rosea 34 Pleurocapsa kerneri 154 Plumaria elegans 137 Poa alpigena 299, 504; annua 519; compressa 519; irrigata 299; palustris 519; pratensis 298, 299, 519; subeaerulea 297- 299, 302 Poaceae 497 Pogonia affinis 304; pendula 509; Rhodora [Vol. 71 ophioglossoides 20, 24, 32 Polunin, Nicholas, Edith Scam- man: An Appreciation 160-164 Polygala cruciata 32; eymosa 32; incarnata 32; lutea 32; mari- ana 32 Polygonaceae 515 Polygonum aviculare 306, 520, var. vegetum 169, 306; Bistorta ssp. plumosum 505; Convol- vulus 520; pensylvanicum 24; Persicaria 520, f. albiflorum 169, var. ruderale 169; puncta- tum var. leptostachyum 24; ssp. 20 Polyides caprinus 134; rotundus 134 Polymnia canadensis 303; macu- lata 211; uvedalia 204, 205, 207, 210, 211, var. densipilis 204- 210, var. floridana 204-210, var. uvedalia 204-210 Polypodiaceae 514 Polypodium armatum 13 Polypremum procumbens 33 Polysiphonia arctica 137; elong- ata 137; fibrillosa 137; flexi- caulis 117, 137; harveyi 138, 527, 528; lanosa 117, 133, 138; nigra 138; nigrescens 123, 138; novae-angliae 138; spp. 115; urceolata var. patens 138 Polytrichum commune 20, var. perigoniale 21 Populus alba 169; deltoides 175 Porphyra linearis 154; miniata 124, 131; umbilicalis 131, f. epiphytica 131 Potamogeton fperfoliatus ssp. Richardsonii 504; subsibiricus 504; vaginatus 504 Potentilla 507; anserina 521; can- adensis 170; Egedii var. Egedii 505, var. groenlandica 25, 521; elegans 505; norvegica 521; uniflora 505 1969] Powell, A. Michael, Taxonomy of Perityle, Section Pappothrix (Compositae-Peritylanae) 58- 93 Prasiola crispa 154 Prestonia acutifolia 548, 550 Primula nivalis 506 Pringsheimiella scutata 117 Proserpinaca pectinata 33 Protoderma marinum 118, 527 Prunus maritima 19, 20, 25; per- sica 25; serotina 20, 25 Pseudananas 36, 278; sagenarius 278 Pseudapiospora thistletonia 201 Pseudendoclonium marinum 118, 527 Pseudomassaria erumpens 200; inversa 199, 200; leucothoés 201; oxydendri 200; thistle- tonia 199-201 Pseudombrophila deerrata 201 Psychotria 535-537, Etymology of 585; viridis 535, 538, 539 Psychotrophum 535, 537 Pteridium aquilinum 31 Ptilimnium capillaceum 20, 21, 26 Ptilota pectinata 153; 137; serrata 137 Ptychopetalum 439 Puceinellia laurentiana 519; pau- pereula var. alaskana 519 Punctaria latifolia 127; planta- ginea 127, 153 Puya 35, 36, 49; alata 37, 41; alba 38, 47; atra 39, 48; boliv- iensis 56; Brittoniana 41, 55; Cardenasii 40, 53; cristata 37, 43; ctenorhyncha 40, 52; dyc- kioides 39, 48; ferruginea 38, 39, 46; Fiebrigii 38, 39, 44; Forsteriana 40, 52; glabrescens 39, 40, 48; glareosa 38, 39, 45; Hauthalii 41, 42; Herzogii 40, 53; Hofstenii 37, 41; humilis plumosa Index to Volume 71 601 41, 55; Kuntzeana 39-41, 50; lasiopoda 40, 51; laxa 37, 43; leptostachys 40, 53; Meziana 38, 39, 44; micrantha 56; min- ima 39, 50; mirabilis 39, 50; mollis 39, 40, 49; nana 41, 54; olivacea 38, 45; paupera 40, 41, 51; Pearcei 38, 47; penduliflora 38, 47; potosina 37, 42; Rai- mondii 41, 56, 57; reducta 39, 49; riparia 38, 46; Rusbyi 37, 42; sanctae-crucis 38, 43; sec- unda 38, 44; spathacea 41; stenothyrsa 38, 39, 45; subg. Piteairniopsis 36, 37; subg. Puya 37; tristis 40, 52; Trollii 37, 42; tuberosa 39, 49; tunar- ensis 40, 54; Ugentiana 40, 52; ultima 37, 41; valida 40, 51; Weddeliana 41, 55; Werder- mannii 55 Pyrenophora macrospora 197, 198 Pyrola elliptica 307; grandiflora 508 Pyrus arbutifolia 25; floribunda 307; melanocarpa 307 Quadrigula closterioides 457; la- custris 457 Ralfsia clavata 123; fungiformis 123; verrucosa 123 Ranunculus acris 521; aquatilis ssp? 505; Cymbalaria 516, 521; Gmelini 505; pedatifidus ssp. affinis 505; repens 170, var. vil- losus 521; reptans 505; tricho- phyllus var. trichophyllus 170; Turneri 508 Raphanus Raphanistrum 521, f. albus 170 Rhexia alifanus 33; cubensis 33; lutea 33; mariana 33; Nashii 33; virginica 33 Rhinanthus Crista-galli 172, 523 Phizoclonium kerneri 120; ripa- rium 116, 526, var. implexum 602 121; ssp. 117, 119; tortuosum 121 Rhizosolenia eriensis 468 Rhode Island, A Station for Luminous Moss 305 Rhodochorton purpureum 132 Rhododendron lapponicum 200; maximum 201 Rhododermis georgii 134; para- sitica 134 Rhodomela confervoides 124, 138; lycopodioides 154 Rhodomonas lacustris 464, 465, 470 Rhodophyllis dichotoma 154 Rhodophysema elegans 134; georgii 134 Rhodymenia palmata 116, 132, 136 Rhoeo 564; spathacea 441 Rhus aromatica 303; radicans 19, 20, 25, 26; typhina 171; vernix 32 Rhynchospora caduca 31; cepha- lantha 31; chalarocephala 31; chapmanii 31; corniculata 31; filifolia 31; glomerata 31; gra- cilenta 31; harveyi 31; inex- pansa 31; macra 31; micro- cephala 31; oligantha 31; plumosa 31; rariflora 31; tracyi 31 Rhytiglossa 548; leucophloea 548 Riba, Ramon, The Alsophila swartziana complex (Cyathea- ceae) 7-17 Ribes glandulosum 514; hirtellum 514; lacustre 514 Rollins, Reed C., A Definitive New Book on the Flora of Northwestern North America (Review) 104-106; On A Weed Species of Rorippa 552-553 Rorippa heterophylla 552; indica 552, var. apetala 552, 5538, var. indica 552; montana 552 Rhodora [Vol. 71 Rosa carolina 25; nitida 25; ru- gosa 19, 25 Rubiaceae 497, 535 Rubus arcticus 505, 507, 508; his- pidus 25; idaeus 514, var. strigosus 25, 521; trivialis 32 Rudbeckia hirta 34 Ruellia noctiflora 33 Rumex acetosella 24, 520; cris- pus 20, 24, 520; domesticus 520; graminifolius 505; maritimus var, fueginus 24; obtusifolius 197; orbiculatus 300, 520; pal- lidus 520 Ruppia maritima 22 Sabatia campanulata 33; dode- candra 33; macrophylla 33 Saccorhiza dermatodea 129 Sagina procumbens 521 Salicaceae 514 Salicornia bigelovii 24; europaea 24; spp. 21; virginica 24, 560 Salix discolor 20, 24, X humilis 169; gracilis var. tectoris X rigida 169; pellita 302, 306 Salsola Kali 24 Sanguinaria canadensis 170 Santamour, Frank S., New Chro- mosome Counts in Ulmus and Platanus 544-547 Sarracenia alata 29, 30, 32; psit- tacina 32 Saxifraga Aizoon 170; cernua 170; oppositifolia ssp. oppositi- folia 505; spicata 508 Saxifragaceae 514 Scamman, Edith, An Apprecia- tion 160; Added Note on 165 Scenedesmus abundans 457; acutiformis 457; arcuatus var. platydisca 457; bijuga 457; brasiliensis 457; dimorphus 457; opoliensis 457; quadri- cauda 458, var. parvus 458, var. Westii 458: serratus 458 1969] Schistostega pennata 305 Schizothrix calcicola 116 Schrankia occidentalis 217 Schroederia judayi 452, 453, 458; setigera 458 Scirpus 513; americanus 23 3i; cyperinus 23 Scleria georgiana 31; oligantha 31; reticularis 31 Scutellaria epilobiifolia 20, 27; in- tegrifolia 33 Scytonema sp. 154 Scytosiphon lomentarius 127 Sedum Rosea 521; spurium 170 Selaginella spp. 59 Senecio atropurpureus 507 Sertularia 132 Silene noctiflora 306 Sisymbrium 552 Sisyrinchium albidum 32; angus- tifolium 24; montanum 520 Smilacina stellata 520, var. crassa 23 Smilax laurifolia 32; rotundifolia 24, 32; smallii 32 Smith, Lyman B., The Bromelia- ceae of Bolivia 35-57, 220-279; and James R. Coleman, Chro- mosome Numbers in Some Bra- zilian Angiosperms 548-551 Solanaceae 561 Solanum asperum 551; Dul- camara 522; nigrum 522; panieulatum 551; rufum 551; sarachoides 172; sp. 551 Solidago 514; odora 34; petiolata 34; puberula 301; Randii 166; rugosa var. villosa 523; sem- pervirens 19, 28, 523; squar- rosa 166; tenuifolia 28 Sonchus asper 28, 523; oleraceus 523 Sorastrum spinulosum 458 Sorbaria sorbifolia 306 Sorbus arbutifolia 32 Sorenson, Paul D., Revision of the Index to Volume 71 603 Genus Dahlia (Compositae-Hel- iantheae-Coreopsidinae) 309- 416; Validation of the Name Dahlia australis 565 Sorocarpus micromomus 154 South Dakota, Cyperus fuscus in Nebraska and 433 Sparganium androcladum 167; hyperboreum 504 Spartina alterniflora 23; patens 20, 22-24, 27 Spergularia canadensis 521; ma- rina 25, var. leiosperma 521 Sphacelaria cirrosa 123; fusca 528; plumosa 124; racemosa var. arctica 128 Sphaeria nigerrima 193; subcor- ticalis 198 Sphaeriaceae 200 Sphaerocystis schroeteri 458 Sphaerotrichia divaricata 125 Sphagnum palustre 20, 21 Spiraea tomentosa 26 Spiranthes praecox 32; vernalis 32 Spirogyra spp. 460 Spirulina laxa 450; subsalsa 117 Spongomorpha arcta 191, 134; hystrix 121; lanosa 121; spin- escens 121 Sporobolus junceus 31; neglectus 168 Spondylosium planum 460 Spyridia filamentosa 528 Staurastrum anatinum var. longi- brachiatum 460; ankyroides 460; arctiscon 460; cingulum 460; cuspidatum 461, var. can- adense 461; dickiei 461; john- sonii 460; limneticum var. cornutum 458-460; manfeldtii 458-460; megacanthum 462; pentacerum 458-460; pilosum 461; pingue 458, 459, 461; pseudopelagicum 458, 459, 461; vestitum 461 604 Staurodesmus bulnheimii 461; cuspidatus 461; dickiei 461; ex- tensus var. joshuae 461; incus var. ralfsii 461; mamillatus var. maximum 461; megacan- thus 462; subtriangularis 462 Steele, Frederic L., A New Wild- flower Guide (Book Review) 107-109 Stellaria ciliatosepala 505; graminea 521; media 521 Stephens, H. A. and Ronald R. Weedon. Cyperus fuscus in Nebraska and South Dakota 433 Stewartia 434, 487, A Seventeenth Century Record for 434; Mal- achodendron 437, 438; ovata 434, 436-438; pentagyna 434 Stichococcus marinum 118 Stictyosiphon tortilis 127 Stigmatea sylvatica 197 Stipitococcus capense 463 Stokesia laevis 34 Streptocalyx 36, 269; Fuersten- bergii 269; Poeppigii 269 Streptopus spp. 514 Strongylocentrotus drobachiensis 155 Stuckey, Irene H., A Range Ex- tension for Suaeda americana in New England 560-561 Stylosphaeridium stipitatum 462 Styrax americana 33; grandi- folia 33 Suaeda americana 521, 560, A Range Extension in New Eng- land for 560; linearis 560; maritima 24; Richii 521 Svenson, Henry K., A Model Flora of Ohio — “The Mono- cotyledoneae" by E. Lucy Braun (Book Review) 562-564 Symplocos tinctoria 33 Synedra ulna 468 Synesiella exigua 196; juniperi Rhodora [Vol. 71 197 Synura adamsii 464-466; uvella 466 Tabellaria 465; fenestrata 468; flocculosa 468 Tamarindus indica 438 Taraxacum erythrospermum 523; officinale 523 Taxodium ascendens 31 Taylor, Constance and R. John Taylor, The Vascular Flora of Oklahoma — Additions and Corrections 215-219 Taylor, R. John and Constance Taylor, The Vascular Flora of Oklahoma — Additions and Corrections 215-219 Teeri, James A., The Phytogeog- raphy of Subalpine Black Spruce in New England 1-6 Tetraedon limneticum 462 Teucrium canadense 20, 27 Thalictrum polygamum 521 Thomas, Joab L. and Joe A. Far- mer, Disjunction and En- demism in Croton alabamensis 94-103 Thyridaria rubro-notata 196, 198 Tilandsia 36, 232-235; adpressa 240, 263, var. Tonduziana 263; apoloensis 253; australis 241; Bakeri 240, 262; bandensis 239, 254; Bangii 240; biflora 237, 238, 245; boliviana 236, 242; boliviensis 237, 244; bry- oides 239, 256; Buchtienii 244; calocephala 237, 247; capil- laris 239, 258, f. cordobensis 259, f. Hieronymi 260, f. in- cana 259, f. virescens 259: Car- denasii 237, 248: caulescens 238, 249; complanata 237, 247; condensata 253, 254; confinis 237, 243; confusa var. saxatilis 252; crassifolia 250; crocata 238, 254; decomposita 252; 1969] Deppiana var. latifolia 240; didisticha 238, 250; Duratii 288, 252, var. saxatilis 252; Fendleri 236, 240; flexuosa 262; Friesii 249; funebris 239, 257; fusco-guttata 237, 245; Gilliesii 239, 257; Guentheri 244; Her- zogii 253; incurva (incurvata) 265; ixioides 238, 248; juncea 240, 263; Krukoffiana 236, 242; Kuntzeana 236, 242; laxissima 238, 252; loliacea 239, 255; longifolia 237, 243; Lorentziana 238, 249, var. simplex 249; Mandonii 254; marantoides 237, 243; maxima 236, 241, var. densior 241; Maxoniana 266; monticola 240, 261; mierantha 260, 261; myosura 239, 257; paleacea 238, 254; paraénsis 240, 264; pardina 240, 262; parviflora 240, 262; penduli- spica 240, 264; pilosa 254, 255; Pohliana 238, 251; polystachia 240, 262; pulchella 251, var. rosea 251; quadrifolia 255; recurvata 239, 258; Reichen- bachii 238, 253; rosea 251; ru- bella 236, 237, 241; Rusbyi 237, 244; scalarifolia 254; See- mannii 239, 260; soratensis 253; sphaerocephala 237, 245, 246; spiculosa 239, 260; strep- tocarpa 238, 253; subg. Aéro- bia 236, 238; subg. Allardtia 236; subg. Anoplophytum 236, 238; subg. Diaphoranthema 236, 239; subg. Phytarrhiza 236, 238; subg. Pseudo-Catop- sis 232, 236, 239; subg. Til- landsia 236, 240; subtropicalis 243; tenuifolia 238, 251; tri- angularis 261, 262; tricholepis 239, 255, var. macrophylla 256; triticea 260; tucumanensis 253; usneoides 31, 239, 260; Valen- Index to Volume 71 605 zuelana 240, 264; vernicosa 238, 250; violascens 237, 244; Walteri 237, 247; Weddellii 252, 253; Williamsii 259; xiphi- oides 238, 249 Tofieldia racemosa 32 Tolypothrix lanata 450 Tradescantia virginiana 303 Trailliella intricata 132, 527 Trichocladus 311 Tricholoma 178, 190; griseum 184; hygrophorus 184 Tricholomopsis 189, 191; bella 190-193; flavescens 191; ornata 191; sulphurea 188-191; sul- phureoides 191; thompsoniana 191 Trichophaea bullata 202 Trichosphaeria subcorticalis 198 Trichostema dichotomum 33 Trichothyrina parasitica 194 Trichothyrium 195; epimyces 195 Tridens ambiguus 31; strictus 31 Trientalis borealis 26; europaea ssp. arctica 506 Trifolium hybridum 522; pra- tense 522; repens 522, f. phyl- lanthum 170 Triglochin maritima 22; palustris 504 Trilisa odoratissima 34 Trillium sessile 303 Triphora trianthophora 509 Tropaeolum glaziovii 550 Tyler, Harry R. Jr., A. R. Hodg- don and R. B. Pike, More Ad- ditions to the Flora of Grand Manan, New Brunswick 306- 308 Typha angustifolia 20, 22; lati- folia 297 Ulmus 544; alata 545; americana 169, 544, 545, f. alba 169, f. intercedens 169, f. laevior 169, In 0341 3934 606 Rhodora [Vol. 71 f. pendula 169, f. floridana 545; Vicia angustifolia 522; Cracca campestris 544; carpinifolia 544; crassifolia 545; fulva 544; glabra 544; macrocarpa 545; montana 544; New Chromo- some Counts in Platanus and 544; parvifolia 546; pumila 544; rubra 169, 308, 544; sero- tina 546 Ulothrix flacca 117, 118, 527; sub- constricta 462; tenerrima 462; variabilis 462 Ulva lactuca 154, 527 United States, Cytotaxonomic Notes on the Species of Cir- sium Native to the Southeast- ern 285; Eragrostis domingen- sis new to 479 Uroglenopsis americana 452, 453, 466 Urospora collabens 121; mirabilis 158; penicilliformis 121 Urtica gracilis 520, viridis 520; sp. 516 Utricularia cornuta 33; fibrosa 33; juncea 33; purpurea 419; subulata 33; vulgaris 419 Vaccinium angustifolium var. nigrum 26; corymbosum 20, 21, 26; macrocarpon 20, 26; uligi- nosum 506 Vallisneria americana 419 Valsa pulviniceps 194, 195 Venturia rhododendri, 201 Veratrum album ssp. oxysepalum 505 Verbascum thapsus 27 Verbena bonariensis 218; brasil- iensis 218 Vernonia altissima 34 Viburnum 554; dentatum 27, 560; nudum 33; recognitum 561; reticulatum 555; wurdackii 555- 559, A New Peruvian Species 554; weberbaueri 555, 557, 559 522 Viola 563; arvensis 171; biflora 506; lanceolata 20, 26, 33; Mackloskii ssp. pallens 522; primulifolia 33 Violaceae 514 Virginia, Isotria medeoloides on the Piedmont of 303 Vitis rotundifolia 32; vinifera 171 Vriesea 36, 233, 265; disticha 266; heliconioides 265, 266; heterandra 265, 266; incurva 265; Maxoniana 265, 266 Ronald R. and H. A. Cyperus fuscus in and South Dakota Weedon, Stephens, Nebraska 433 Wells, James R., A Review of the Varieties of Polymnia Uvedalia 204-211 Xanthidium antilopaeum 462, var. polymazum 462 Xanthium echinatum 19, 20, 28 Xylariales 200 Ximenia 439, 441, 443; americana 443, var. americana 439-442; caffra 443, var. natalense 441; Parasitism in 439 Xyris ambigua 31; baldwiniana 31; caroliniana 31; difformis 31; fimbriata 31; flexuosa 31; iridifolia 31; montana 173 Young, Steven B., Additions to the Flora of Saint Lawrence Island, Alaska 502-509 Yucea aloifolia 48 Zamia floridana 441 Zigadenus densus 32; rimus 32 Zostera marina 115, 125, 526-528 glaber- 134.