Dodota JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS | RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON ' Associate Editors VOLUME 63 1961 The New England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge 38, Mass. Dodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Assoçiate Editors Vol. 63 January, 1961 No. 745 CONTENTS: Notes on American Rorippa (Cruciferae). MECC: DE arrg LT Y- eee E E E CORR A Re-evaluation of the Generic Status of Ascyrum and Crookea (Guttiferae). William P. Adams and Nor- man TECHS IL OOS ON. etes eret TL aa, Chromosome Numbers of Some Brazilian Leguminosae. HL DAT GNE E. S. ITW Led ee erteilen COO A Decade of Botanizing in Illinois. Robert H. Mohlenbrock .................. e eeeeeeeeeee esee eee nee New Combinations in Grasses. Julian A. Steyermark Daa O L. FOUL COP eee aeaeaeyeesen (EERIE REV SNC Rhododendron maximum in Hopkinton and Harrisville, New Hampshire. A. R. Hodgdon and Radcliffe Pike .......... Campanular Persistence. Arthur Stanley Pease .................... 26 29 The Nem England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume, Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able. All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass. Printed by THE LEXINGTON PRESS, INC. Lexington, Mass. Rbodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 January, 1961 No. 745 NOTES ON AMERICAN RORIPPA (CRUCIFERAE) REED C. ROLLINS One of the most distinctive American species of Rorippa is R. sinuata (Nutt.) Hitchcock. This is because of the per- ennial habit and the fact that the trichomes are unusual, being somewhat vesicular and hemispherical in shape rather than elongated and pointed. On dried specimens, the tri- chomes appear to be flat and scale-like because they are collapsed. These are present rather sparsely along the mid- ribs of the under sides of the leaves and on the stems on specimens from the plains states, but an increase in the abundance of trichomes and a more extensive coverage of the plants occur on material from extreme western Texas to Arizona and northward through the Rocky Mountains and to the westward in the Columbia River valley (with certain exceptions mentioned below) and in the Great Basin region. Evidently these hemispherical-shaped trichomes are not wholly persistent, for they may be seen on the young foliage and pedicels of some specimens where they are completely absent from the mature leaves and pedicels. Ordinarily, the siliques are glabrous but plants with the greatest density of trichomes have them extending to the fruits, where they may be restricted to the valve edges next to the replum or they may extend to cover the entire valve surface. A speci- men from southwestern Colorado collected by T. S. Bran- degee apparently impressed Gray (1876) because of the roughness produced by the dense covering of trichomes on the siliques. This specimen became the type of Nasturtium trachycarpum Gray. According to Kearney and Peebles (1951) only the pubescent-fruited form occurs in Arizona. 1 2 Rhodora [Vol. 63 However, neither the extent nor the abundance of these peculiar vesicular trichomes is a safe basis for any tax- onomic separations and, accordingly, N. trachycarpum should be treated as a straight synonym of Rorippa sinuata. Rorippa sinuata occurs in patches, due to the spreading of the plants by underground roots. The species appears to be adapted to disturbed soils and finds roadsides and rail- road banks suitable for vigorous growth. It is not certain that it is actively increasing its geographic range but it could easily be doing so because such excellent pathways as transcontinental highways are open to it. The geographic range of R. sinuata appears to be from Saskatchewan and Minnesota south to Arkansas and Texas, west to Arizona and north to Washington. Munz and Keck (1959) cited it from Little Lake, Inyo County, and Modoc County, Cali- fornia, and these are likely areas in which to find plants of more easterly and northerly distribution. However, I have not personally seen California specimens of R. sinuata. THE TYPE OF RORIPPA SINUATA Nasturtium sinuatum was published by Torrey and Gray (1838) from Nuttall’s manuscript, where “banks of the Oregon and its tributaries; also in Arkansas" are given as source localities. On the sheet bearing the type in the Brit- ish Museum, two collections are present. One of these is marked “Sisymbrium sinuatum Arkansas", the other, “Na- sturtium sinuatum Columbia River & Arkansas." The speci- men marked “Arkansas” and the sterile shoot on the right [one of three fragments] of the second designation resemble very closely modern material from Arkansas, Oklahoma and Texas that is regularly referred to R. sinuata. The other two fragments are similar to modern specimens from the Co- lumbia River valley but, in contrast to most of the material of R. sinuata, they are nearly glabrous. Unfortunately, I used only a hand-lens in examining these specimens while I was at the British Museum in 1950 and I cannot now say with certainty that there are absolutely none of the charac- teristic vesicular trichomes present on them. In other re- spects, the Arkansas and Columbia River specimens appear to be similar enough to represent but a single species. How- 1961] Rollins — American Rorippa 3 ever, it was disturbing to find that the Nuttall specimen in the British Museum marked “Nasturtium curvisiliqua Co- lumbia Shores" does have a conspicuous covering of vesicu- lar trichomes and it certainly represents the species we think of as Rorippa sinuata. A Nuttall collection in the Gray Herbarium marked “Nasturtium curvisiliqua. Sisymbrium Hooker. Oregon River", in Nuttall's handwriting, has two fragments; one is Rorippa sinuata, the other not determin- able with certainty but definitely not R. sinuata. These latter must be the specimens mentioned by Torrey and Gray (l.c.) as differing from the description of Sisymbrium curvisili- quum Hooker. It appears to me that Nuttall either made mixed collections of Rorippa along the Columbia River or his collections were subsequently mixed. We know from Nuttall’s own collection and from Suksdorf 2430 near Bin- gen, with vesicular trichomes, and Suksdorf 2103, W. Klicki- tat County (glabrous or nearly so) that both types occur along the lower Columbia in Washington. I am satisfied that the two Suksdorf collections should be accommodated in the same species and it appears that no real difficulties will arise from accepting the Columbia River specimens on the type sheet at the British Museum as the holotype of R. sinuata. However, it does mean that the holotype is somewhat atypi- cal of the species as a whole. One source of some confusion, with regard to the typifica- tion of Rorippa curvisiliqua, is Torrey and Gray’s (l.c.) treatment of Nasturtium curvisiliqua in such a way that their intent was not made clear. Nuttall was merely given credit for the transfer of Sisymbrium curvisiliqua Hooker to the genus Nasturtium. Thus, Howell (1897) assumed that Nasturtium curvisiliqua of Nuttall was the basionym of Rorippa curvisiliqua rather than the rightful Sisymbrium curvisiliquum of Hooker. Actually Nuttall did not intend to publish Nasturtium curvisiliqua as a new species and this was not done for Nuttall by Torrey and Gray. OTHER SPECIES WITH VESICULAR TRICHOMES One reason for a careful review of the variation and dis- tribution of Rorippa sinuata, as given above, was to provide the basis for a proper assessment of specimens from Mexico 1 Rhodora [Vol. 63 that share many technical characteristics with it, including the possession of vesicular trichomes. A critical comparison of the Mexican material with R. sinuata shows that the speci- mens do not belong to it but represent a closely related un- described species. Rorippa ramosa Rollins, sp. nov. Prostrate dense perennial, up to 10 dm. in diameter; stems numerous, highly branched, sparsely pubescent with vesicular trichomes, 3-6 dm. long, branches present in the axil of nearly all leaves from base to apex of each stem; more generally pubescent than principal stems; leaves numerous, sessile, auriculate, thick, greyish-green, oblong to broadly lanceolate, pinnately lobed, 3-5 em. long, 5-12 mm. wide, lobes con- fluent toward base; midrib prominent on lower surface of leaf, pubes- cent with vesicular trichomes; inflorescences short, mostly less than 5 em. long; sepals oblong, glabrous or with a few trichomes present on the dorsal surface, hyaline-margined, non-saccate, 2-2.5 mm. long; petals pale yellow, spatulate, not differentiated into blade and claw, 2.5-3 mm. long, 0.75-1 mm. wide; pedicels widely spreading to ascend- ing, straight to slightly curved outward, sparsely covered with tri- chomes, 3-5 mm. long, expanded at summit; siliques divaricately spreading to erect, slightly curved inward, oblong to lanceolate, plump, obtuse below, tapering above, 6-10 mm. long, valves densely covered with vesicular trichomes along their margins; styles glabrous, 1.5- 2.5 mm. long; ovules numerous, funiculi slender; seeds plump, cordi- form, ca. 1.5 mm. in diameter, seed coat colliculate (cf. Murley, 1951) and lustrous; cotyledons accumbent. Fig. A-C. Herba perennis procumbens, caulibus numerosis ramosis 3-6 dm. longis; foliis crassis sessilibus aurieulatis oblongis vel late lanceolatis pinnatilobatis costatis 2-3 cm. longis 4-12 mm. latis sparse pubescenti- bus; sepalis non-saccatis oblongis 2-2.5 mm. longis; petalis flavis spathulatis 2.5-3 mm. longis; pedicellis divaricatis vel adscendentibus sparse pubescentibus 3-5 mm. longis; siliquis oblongis vel lanceolatis ad basi obtusis 6-10 mm. longis sparse pubescentibus; stylis glabris 1.5- 2.5 mm. longis; ovulis numerosis; seminibus cordiformibus colliculatis; cotyledonibus accumbentibus. Type in the Gray Herbarium, collected in a dry arroyo, 3 miles northwest of Ceballos, Durango, Mexico, May 4, 1959, D. S. Correll and I. M. Johnston 21449. Isotype in the Lundell Herbarium of the Texas Research Foundation. An additional collection of Rorippa ramosa is: San Lorenzo de La- guna, 70 miles south of Parras, Coahuila, May, 1880, E. Palmer 34 (GH, US). The general habit of growth of Rorippa ramosa is that of a densely leafy, highly ramified, flat, nearly circular plant. Rollins — American Rorippa 1961] Fic. B. silique X 2. Fic. C. Rorippa ramosa. Fic. A. habit sketch, X M4. Drawings by C. S. Tsao. A-C. replum showing numerous funieuli, Fic. 2 x 6 Rhodora [Vol. 63 It is possible that there are basal leaves on the young plants but none are evident on the mature specimens I have studied. It is more probable that a truly basal rosette of leaves is not a characteristic of the species. The illustration, fig. 1A, even though it shows only a portion of two main stems, gives an approximate idea of the intricate branching present. In habit alone, R. ramosa differs strikingly from R. sinu- ata. I have collected R. sinuata four times in Kansas, Colo- rado and Wyoming and in each instance colonies were found with the individual plants interconnected underground. Usu- ally a single stem, or at most three or four, emerges at a given locus. In contrast to this, R. ramosa has numerous stems arising at the summit of what appears on the speci- mens to be a tap-root. An important difference between the two species is in the size of the flowers. Those of R. sinuata are more than twice the size of the flowers of R. ramosa, the petals being over twice the width. The outer sepals of R. sinuata are saccate while those of R. ramosa are non- saccate. Other differences include the shorter, more erect and stouter pedicels, more lanceolate-shaped fruits, shorter styles, less angular seeds and more prominently colliculate seed-coats of R. ramosa as compared with R. sinuata. The leafiness of the branches, relatively short infructescences and repeated branching give R. ramosa a distinctive overall appearance. Rorippa ramosa is unquestionably related to R. sinuata from which it is geographically isolated. I have not seen any other material from Mexico that falls into this alliance with- in the genus. Rorippa Walteri (Ell. Mohr, Bull. Torr. Bot. Club 24: 23, 1897. A photograph of the type of Sisymbrium Walteri Elliott in the Gray Herbarium leaves no doubt as to the application of this name to plants that have at times been placed under Nasturtium tanacetifolium H. and A. (cf. Schulz, 1933). The type of S. Walteri is in the Charleston Museum of Charleston, South Carolina. There seems little doubt that Elliott’s renaming of the plants tentatively identified as 1961] Rollins — American Rorippa T Sisymbrium tanacetifolium by Walter (1788) provides the first available name clearly applied to this species. The only justification for taking up the specific name tanacetifolium would be to consider it to have been newly proposed by Hooker and Arnott, as has been done quite often, as a re- naming in another genus of the Walter described species. Hooker and Arnott (1834) did cite Walter and it is clear that they had his description in mind. However, if the name were to date from their work, it would even then not take priority over Elliott's Walteri. Walter did not cite Linnaeus as the author of Sisymbrium tanacetifolium in his book but he did use a question mark after tanacetifolium. It is incon- ceivable that he would have questioned the application of a name he was himself proposing. Rorippa Walteri occurs from South Carolina southward to Florida and westward, largely on the coastal plain, through Texas; on the eastern and western lowlands of Mexico and at least in Nicaragua in Central America. For the present study, I have not tried to determine the total geographic range of the species. It is possible that KR. Walteri has been carried by man somewhat outside of its natural range in Mexico and Central America because it is used as a salad plant, and it may be seen in the local mar- kets of western Mexico. As in most species of Rorippa, there is considerable variation in the leaf-pattern and habit of growth. These features are strongly influenced by the conditions of moisture and light under which the plants grow. Vesicular trichomes, somewhat longer and more con- stricted toward the base than those in Rorippa sinuata, are found on plants of R. Walteri, especially on the lower por- tions of the stems. The abundance decreases upward, often resulting in glabrousness on the upper parts of the plant. This distribution of trichomes is characteristic of specimens from the Atlantie slope of Mexico to South Carolina but most of the material from the Pacific slope of Mexico shows a different trichome distribution. Specimens from Nayarit to Sonora, and the single specimen I have seen from Nicar- agua, have glabrous stems but vesicular trichomes are pres- 8 Rhodora [Vol. 63 ent on the siliques. Specimens from Colima and Oaxaca are similar to those of the eastern range of the species. There appears to be a trend toward shorter pedicels and more un- evenly divided leaves in the western Mexican material. Fur- thermore, it seems that drier habitats are the rule in the western as compared to the eastern part of the range. All of these correlated characteristics, coupled with geographic segregation, suggest a divergent trend that may eventually result in separable taxa. At the present time, evidences of continuity throughout the species are found in a variety of structures and it seems wise to view the whole as a single species, although the addition of new knowledge might well require a revision in the direction of a recognition of a larg- er number of entities. Rorippa Walteri is most easily recognized by the pinnate- ly compound lower leaves with the leaflets at least dentate and often deeply lobed; the short (1-3 mm.) divaricate pedi- cels; minute flowers; terete, divaricately ascending siliques ; prominent styles and the distinctive, somewhat clavate vesicular trichomes either on the stems or the siliques or occasionally on both. A selection of specimens is cited below to aid in interpreting this species. UNITED STATES. South Carolina: Beaufort, J. R. Churchill 432 (GH) ; St. Johns, Berkeley, H. W. Ravenel s. n. (GH). Florida: 4 miles n. Crawfordville, Wakulla Co., Godfrey and Almodovar 52975 (GH); dried bottom of Lake Jackson, Leon Co., Hunnewell 13048 (GH); near Jacksonville, Duval Co., A. H. Curtiss 4589 (GH). Mississippi: 3 miles from Laurel, Jones Co., Cooley and Pease 3104 (GH); near Natchez, Sullivant s. n. (GH). Louisiana: New Orleans, T. Drummond 18 (GH). Oklahoma: Sapulpa, B. F. Bush 1233 (GH). Texas: Conquista Crossing, between Falls City and Deweesville, Karnes Co., Johnston 1462 (GH) ; Corpus Christi, Nueces Co., Heller 1487 (GH); Bastrop, Bastrop Co., E. J. Palmer 33381 (GH); Santa Elena Canyon, Brewster Co., Goodman and Waterfall 4653 (GH). Mexico. Tamaulipas: near Matamoros, April, 1831, Berlandier 879 (GH). Vera Cruz: Jalapa, Pringle 8087 (GH, US); Cordoba, Orcutt 3134 (GH, US). Sonora: vicinity of Alamos, Rose, Standley and Russell 13012 (GH, US) ; vicinity of Culiacán, Rose, Standley and Russell 14974 (GH, Us) ; 12 miles west of Culiacán, Gentry 7004 (GH). Nayarit: vicinity of Acaponeta, Rose, Standley and Russell 14246 (us). Colima: Manzanillo, Palmer 1344, (GH). NICARAGUA. Zelaya: Rio Grande, Antonio Molina R. 2172 (Us). 1961] Rollins — American Rorippa 9 Rorippa portoricensis (Sprengel) Stehlé, Rev. Bot. Ap- pliq. 26: 108. 1946. This name is based on Nasturtium portoricensis Sprengel (1825) which appears to have been described from the same Bertero collection cited from Puerto Rico as that used by De Candolle (1821) as the basis for Nasturtium palustre var. brevipes. De Candolle’s var. brevipes was first raised to specific rank by Grisebach in 1860 and the name Nasturtium brevipes (DC.) Griseb. was used by Schulz in various papers on the Cruciferae of the West Indies. However, the oldest and the correct specific name appears to be portoricensis. The relationship of R. portoricensis to R. Walteri is a very close one and there is some question as to whether an inter- pretation of the existing evidence as indicating a single spe- cies, probably with the recognition of a West Indian variety, is not more realistic than the acceptance of two species. Watson, in the Synoptical Flora of North America (1895), did treat the West Indian material as Nasturtium tanaceti- folium var. insularum. At that time, N. tanacetifolium was in common use for the species now known as Rorippa Wal- teri. However, a broader and more detailed study of the genus Rorippa than is at present possible is much needed, and should be looked to fór the settling of many such difficult questions of taxonomic interpretation as that posed by R. portoricensis. Assuming for the present that R. portoricensis is a good species, the material I have seen shows it to be present in Cuba, the Dominican Republic and Puerto Rico. The prin- cipal differences from R. Walteri are shown by the siliques, which are shorter with a nearly sessile stigma instead of a definite style, and the shorter, more ascending pedicels. Ves- icular trichomes similar to those of R. Walteri are present on the lower stems and usually the valve-margins of the Cuban specimens and at least along the valve-margins of the siliques (only occasionally on the lower stems) of the material from the Dominican Republic and Puerto Rico. These are the same two trichome distribution patterns found in R. Walteri. The following specimens have been deter- mined as R. portoricensis in the present study. 10 Rhodora [Vol. 63 Cusa. F. Rugel 235 (Gn) ; Pinar del Rio: Galafre, Britton and Cowell 9826 (GH); Las Guaaimas, O'Donovan 4687 (GH); Santa Catarine, C. Wright, Feb. 1860 (GH) ; “in Cuba Orientali", C. Wright 1562 (GH). DOMINICAN REPUBLIC: Pontezuela, Jiménez 2565 (US); Constanza, Türckheim 3031 (GH, US); La Estancia, Ekman 12124 (A, Us); La Cumbre, Raunkier 1084 (US). PUERTO Rico: Cabo-Rojo, Sintenis 699 (Gu); Adjuntas, Sintenis 4023 (US); Afasco, Sintenis 5625 (GH). — GRAY HERBARIUM OF HARVARD UNIVERSITY. LITERATURE CITED GRAY, ASA. 1876. Nasturtium trachycarpum, in Brandegee, Bull. U. S. Geol. and Geog. Survey 2: 233. HowELL, THoMas. 1897. A Flora of Northwest America. Portland, Oregon. KEARNEY, THOMAS H. and Ropert H. PEEBLES. 1951. Arizona Flora. University of California Press, Berkeley and Los Angeles. MUNZ, PHILIP A. and Davip D. Keck. 1959. A Flora of California. University of California Press, Berkeley and Los Angeles. MURLEY, MARGARET R. 1951. Seeds of the Cruciferae of Northeastern North America. Amer, Midl. Nat. 46: 1-81. ROoLLINS, R. C. 1960. The American Cruciferae of Sessé and Mociño. RHODORA 62: 11-20. ScHULZ, O. E. 1933. Beitráge zur Kenntnis der Gattung Nasturtium R. Br. II, Nasturtium tanacetifolium Hook. et Arn. und ver- wandte Arten. Repert. Spec. Nov. 34: 131-136. WALTER, TH. 1788. Flora Caroliniana. London. WATSON, S. 1895. Synoptical Flora of North America, p. 146-149. Cambridge, Mass. A RE-EVALUATION OF THE GENERIC STATUS OF ASCYRUM AND CROOKEA (GUTTIFERAE) WILLIAM P. ADAMS AND NORMAN K. B. ROBSON Recent intensive studies of the floral anatomy and taxon- omy of Hypericum and the segregates Ascyrum and Crookea have led to a re-evaluation of the generic status of these groups. A general review and study of the floral anatomy of Hypericum and related genera by Robson (1956) indi- cates that the species belonging to Ascyrum, Crookea and the sections Myriandra and Brathydium of Hypericum are closely related to each other. Recent studies by Adams (1959) in the taxonomy of these species appear to confirm Robson's idea that they form a very natural group. In the following discussion we will present evidence supporting the 1961] Adams and Robson — Ascyrum and Crookea 11 reduction of the genera Ascyrum and Crookea to the more inclusive genus Hypericum. Financial support of the senior author's studies in the taxonomy of Hypericum was provided by the following sources: the Fernald Fund for Study in Systematic Botany at Harvard University, established by Mr. F. W. Hunnewell of Wellesley, Massachusetts; a National Institute of Health grant-in-aid through the Department of Biology, Harvard University; and a research grant (RG-6305) to Dr. R. K. Godfrey of Florida State Uni- versity from the Division of General Medical Studies, Public Health Service. Ascyrum L., Gen. Pl. ed. 5. 342. 1754. Since it was first described by Linnaeus the genus Ascyrum has included those species having a tetramerous calyx and corolla, with two unequal pairs of sepals. The large genus Hypericum has long included those species with pentamerous corolla and calyx. Apparently the first author to challenge this classification was Crantz (1766) who transferred the Ascyrum species to the larger genus. Later authors, however, maintained As- cyrum as a distinct genus. During the past sixty years several taxonomists have questioned this segregation but, with the exception of Keller (1895), no one has attempted to revise the traditional classification. Coulter, in his treat- ment of the Hypericaceae for the Synoptical Flora of North America (1897), stated: “The propriety of a generic sep- aration from Hypericum is very doubtful.” In 1895 Keller treated Ascyrum as a section of Hypericum. Thirty years later, however, Keller (1925) reconsidered the problem and gave Ascyrum generic status but apparently with some mis- givings for he remarked: “Die Gattung ist jedenfalls nur künstlich von Hypericum zu trennen." Recently, in a re- vision of the Ascyrum species by Adams (1957), the question was briefly discussed but it was decided to treat the species as members of Ascyrum until the closely related species of Crookea and Hypericum § Myriandra could be studied. EVIDENCE FOR A MERGER OF ASCYRUM WITH HYPERICUM THE FLOWER. The tetramerous calyx and corolla, charac- ters which have been traditionally used to distinguish As- cyrum as a genus, occur not infrequently in many Hypericum species. In § Myriandra, 4-parted flowers have been ob- served in H. ellipticum Hook., H. myrtifolium Lam. and H. galioides Lam. and doubtless occur occasionally in other species. According to Milne-Redhead (1953), 4 sepals and 12 Rhodora [Vol. 63 petals may sometimes be found in the African H. kiboense Oliv. (S Humifusoideum). Tetramery appears to be the normal condition in H. filicaule Dyer ex Hook. f. a spe- cles of Hypericum from the Sikkim Himalaya which Dyer considered to belong to Ascyrum. Robson (1956), however, showed it to be closely related to other Himalayan species in that section. Flowers with pentamerous corollas, a characteristic which has been long used to separate Hypericum from Ascyrum, occur in Ascyrum pumilum Michx. Pentasepalous flowers have not yet been observed in Ascyrum although they occur not infrequently in Crookea (see below). The genus Ascyrum has been further characterized by the presence of two unequal pairs of sepals. This condition is not uncommon in many species of Hypericum in which the sepals are unequal because of their quincuncial development. The first (or exterior) two are almost opposite and more or less equal, the third (with one margin exterior and the other interior) is smaller, and the fourth and fifth sepals (or in- terior ones) are smaller still and nearly equal in size and shape. Unequal sepals are characteristic of H. macrosepalum Rehder, H. humifusum L., H. androsaemum L., H. filicaule, and several species in Hypericum § Myriandra. Floral characteristics common to Ascyrum and certain species of Hypericum include: persistent sepals and sta- mens; short to long styles; minute stigmas; lack of sepal articulation; petals which are yellow, convolute in the bud and usually quickly deciduous; numerous afascicular sta- mens; versatile anthers which dehisce laterally by longitu- dinal slits; 3 or 2 carpels; parietal placentation; dry, septi- cidal capsule; numerous small seeds; and an inflorescence which is obviously reduced to a single flower from a simple, 3-parted dichasium. ANATOMY OF THE FLOWER. Except for tetramery, the a- natomy of the flower in Ascyrum is very similar to that found in various species of Hypericum § Myriandra. The inner (smaller) sepals have unilacunar (not trilacunar) traces, but this is a common effect of reduction in the width of in- sertion of a foliage member (Robson, 1956). 1961] Adams and Robson — Ascyrum and Crookea 13 VEGETATIVE BODY. In growth habit the species of Ascyrum are not unlike various members of Hypericum § Myriandra. The low, bushy, suffruticose form of A. pumilum Michx. and A. multicaule Michx. is paralleled by that of H. buckleyi S. Wats. The erect, shrubby nature of A. stans Michx. and A. tetrapetalum (Lam.) Vail is very similar to that of H. cisti- folium Lam. and H. myrtifolium Lam. Winged stems are present in Ascyrum species and occur in practically all sections of the genus Hypericum. In Ascyrum, as well as in $ Myriandra of Hypericum, the secretory structures in the leaves, sepals and stems are composed of translucent or pellucid-punctate glands which take the form of dots or vittae (elongate tube-like sacs). The black punctate glands which are typical of such her- baceous species as H. punctatum Lam. and H. perforatum L. are absent from these groups, however. The leaves (as well as the sepals) of Ascyrum do not have a basal groove or articulation, a feature lacking in its closest relatives among the species of Hypericum § Myrian- dra as well. The leaf margin in Ascyrum is narrowed abrupt- ly into a thin hyaline zone which is easily seen in living plants but becomes obscure in dried material. This characteristic is also present in several species of Hypericum § Myriandra. ANATOMY OF THE STEM. A comparative study of the stem anatomy by Vestal (1937) showed that Ascyrum species dif- fer very little, if any, from the woody members of the genus Hypericum. Furthermore, Vestal found “a very constant homogeneity" with “no segregation of anatomical groups possible" among the one hundred or more species of Hyper?- cum which he studied. CHROMOSOMES. The haploid number of nine is present in five of the six species of Ascyrum for which counts have been made (Adams, 1959). The same haploid number is present in at least eighteen members of Hypericum § Myriandra (Hoar and Haertl, 1932; Adams, 1959). It has been found also in several species of sections Hypericum, and Triad- enioidea. In size and shape the meiotic chromosomes of Ascyrum and of S Myriandra are remarkably similar. No detailed karyotype analysis has been attempted on these 14 Rhodora [Vol. 63 species, however. GEOGRAPHIC DISTRIBUTION. Ascyrum and its closest rela- tives in Hypericum § Myriandra are native only in eastern North America, especially the southeastern portion. More- over, the other members of the section occur only in the same region. Crookea Small, Fl. Southeastern U.S. 786, 1335. 1903. The single species comprising the genus Crookea was first described by Torrey and Gray (1838) as Ascyrum micro- sepalum. That it occupied an anomalous position in this genus was suggested by their remark: “This species differs from all the others of the genus in the somewhat equal and very small sepals, as well as in the long style: it has the habit of Hypericum.” Many years later Sereno Watson (1878) transferred it to Hypericum. Coulter (1897) agreed with Watson, noting that this species showed very close affinity with Hypericum in all characteristics except, of course, the tetramerous flowers. In order better to accom- modate this rather anomalous species, Small (1903) dis- tinguished the monotypic genus Crookea. Later Keller (1925) treated it again as an Ascyrum. Recent studies (see below) suggest that both Crookea and Ascyrum are very closely related to $ Myriandra of Hypericum. EVIDENCE FOR A MERGER OF CROOKEA WITH HYPERICUM THE FLOWER. The flowers of Crookea are like Ascyrum in being tetramerous, but the nearly equal size and shape of its two pairs of sepals are definitely suggestive of Hyperi- cum. Sepal and petal number in Crookea are very variable even in flowers on a single plant. Many individual plants may have only 4-parted flowers. Not infrequently, however, plants are found which have typical hypericaceous penta- merous flowers. The same plants may possess flowers which are “intermediate” in sepal and petal number, size, and shape. As regards the total number of perianth parts, flow- ers with 4 sepals and 5 petals, or 5 sepals and 5 petals have been observed. The sepals and petals of a single flower may differ in size and shape and one or more may be much smaller than the others; not infrequently, two petals or two sepals 1961] Adams and Robson — Ascyrum and Crookea 15 may be partially fused, presenting a doubled appearance. A detailed analysis of the flower variation in Crookea will be presented by the senior author in a forthcoming taxonomic study of its single species and its relatives in Hypericum § Myriandra. All the other floral and vegetative morphological charac- teristics of Crookea occur as well in various Hypericum species. These include its low, bushy growth habit, winged stems, translucent secretory glands, numerous small seeds, absence of leaf and sepal articulation, rounded leaf margins, parietal placentation, and a haploid chromosome number of nine. SUMMARY To recognize Crookea, Ascyrum and Hypericum as distinct genera appears to us to require the use of rather arbitrary criteria as a means of delimiting these genera. An analysis of the constellation of morph- ological characteristics common to the species of Crookea, Ascyrum and certain species of Hypericum which suggest close genetic affinity indicates that they should be included in a single genus of which Crookea and Ascyrum represent extreme evolutionary developments. The reduction of the genera Crookea and Ascyrum to Hypericum makes necessary the following nomenclatural changes: Hypericum edisonianum (Small) Adams and Robson, comb. nov. Based on Ascyrum edisonianum Small, Man. Southeastern Fl. 868. 1933 [as Edisonianum]. Hypericum hypericoides (L.) Crantz, Institut. rei herbariae. 2:520. 1766. Ascyrum Hypericoides L., Sp. Pl. 2:788. 1753. Hypericum microsepalum (T. & G.) Gray ex S. Wats., Biblio. index to N. Am. botany. 1:456. 1878. Ascyrum microsepalum T. & G., Fl. N. Am. 1:157. Crookea microsepala (T. & G.) Small, Fl. Southeastern U. S. 786, 1335. 1903. Hypericum suffructicosum Adams and Robson, nom. nov. Based on Ascyrum pumilum Michx., Fl. Bor.-Am. 2:77. 1808. Non Hypericum pumilum Sesse & Moc., Fl. Mexic. ed. 2:177. 1894 [as pumillum ]. Hypericum stans (Michx.) Adams and Robson, comb. nov. Based on Ascyrum stans Michx., Fl. Bor.-Am. 2:77. 1803. Hypericum stragulum Adams and Robson, nom. nov. Based on As- cyrum multicaule Michx., Fl. Bor.-Am. 2:77. 1803. Non Hypericum multicaule Lam., Encyc. 4:178. 1797. Ascyrum spathulatum Spach, Hist. Nat. Vég. 5:462. 1836. Non Hypericum spathulatum (Spach) Steud., Nomencl. ed 2, 1:789. 1840, which was based on Myriandra spathulata. Hypericum tetrapetalum Lam., Encyc. 4:153. 1797. Ascyrum tetra- petalum (Lam.) Vail in Small, Fl. Southeastern U. S. 785. 1902. 16 Rhodora [Vol. 63 ACKNOWLEDGEMENTS The senior author wishes to express his appreciation to Dr. Reed C. Rollins, Dr. R. K. Godfrey, Dr. Carroll E. Wood, Jr. and Mr. George Argus for their interest and helpful criticisms. — DEPARTMENT OF BI- OLOGICAL SCIENCES, FLORIDA STATE UNIVERSITY, TALLAHASSEE, FLORIDA, AND THE ROYAL BOTANIC GARDENS, KEW, ENGLAND. LITERATURE CITED ADAMS, W. P. 1957. A revision of the genus Ascyrum (Hyperi- caceae). Rhodora 59: 73-95. 1959. The taxonomy of Hypericum section Myriandra (Hypericaceae). Unpub. Ph. D. Thesis, Harvard University. COULTER, J. M. 1897. Hypericaceae, in Gray, Synoptical Flora of North America. Vol. 1, part 1, 282-284. CRANTZ, H. J. N. 1766. Institutiones rei herbariae. Vienna. Hoar, C. S. AND E. J. HAERTL. 1932. Meiosis in the genus Hypericum. Bot. Gaz. 93: 197-204, pl. 1. KELLER, R. 1895. Hypericum, in Engler and Prantl, Die Natürlichen Pflanzenfamilien. Teil 3, Abt. 6. 194-242. . 1925. Ascyrum, in Engler and Prantl, Die Natürlichen Pflanzenfamilien. 2 Aufl. 21:154-237. MILNE-REDHEAD, E. 1953. Hypericaceae, in Turrill and Milne-Red- head, Flora of Tropical East Africa. London. RoBSON, N. K. B. 1956. Studies in the genus Hypericum. Unpub. Ph. D. Thesis, Edinburgh University. TORREY, J. AND ASA GRAY. 1838. Flora of North America. 1:156-157. VESTAL, P. A. 1937. The significance of comparative anatomy in es- tablishing the relationship of the Hypericaceae to the Guttiferae and their allies. Philippine Jour. Sci. 64:199-256. WATSON, S. 1878. Bibliographical index to North American botany. 1:456. CHROMOSOME NUMBERS OF SOME BRAZILIAN LEGUMINOSAE! B. L. TURNER AND H. S. IRWIN The junior author of this paper spent 5 months during 1958-59 in south-central Brazil collecting Cassia material in connection with a doctoral thesis problem. Since he was routinely collecting bud material of various species of this genus and shipping these air mail to the senior author for meiotic examination, he was able to include, as time and opportunity permitted, occasional bud collections of other This study was supported in part from funds provided by The University Research Institute of The University of Texas, We are indebted to Dr. Richard S. Cowan of the Smithsonian Institution for the identification of the specimens. 1961] Turner and Irwin — Brazilian Leguminosae 17 taxa of the family Leguminosae. The present contribution summarizes the results of a study of this latter material. Chromosome counts were made by the squash technique essentially as outlined by Turner (1956). Voucher speci- mens are deposited at The University of Texas Herbarium, the United States National Herbarium and elsewhere. 3 4 FIGURES 1-4. Camera lucida drawings of meiotic figures. Fic. 1. Camptosema tomen- tosum (n = 11). Fig. 2. Centrosema coriaceum (n = 11). Fic. 3. Galactia martii n = 10). Fic. 4. Periandra mediterranea var. mucronata (n = 11). All figures x ca. 2000. 18 Rhodora [Vol. 63 TABLE 1. SPECIES OF BRAZILIAN LEGUMINOSAE EXAMINED FOR CHROMOSOME NUMBERS. Voucher collection Species CAESALPINIOIDEAE n number Bauhinia aff. mollis (Bong.) Goras: Irwin 2584. n = 14 Walp. Bauhinia rufa Steud. MINAS GERAIS: Irwin 2395. n — 14 Caesalpinia ferrea Mart. MINAS GERAIS: Irwin 2368. n — 12 Caesalpinia spinosa (Molina) MINAS GERAIS: Irwin 2329. n= 12 Ktze. Caesalpinia sp. MINAS GERAIS: Irwin 2331. — n — 12 Copaifera langsdorffii Desf. MINAS GERAIS: Irwin 2394. n= 12 PAPILIONOIDEAE Aeschynomene elegans S.& C. MINAS GERAIS: Irwin 2081. n = 10 Camptosema tomentosum MINAS GERAIS: Irwin 2481. n= 11 Benth. (Fig. 1) Centrosema coriaceum Benth. MINAS GERAIS: Irwin 2503. m — 11 Centrosema coriaceum Benth. MINAS GERAIS: Irwin 2366. n=11 (Fig. 2) Crotalaria stipularia Desv. MINAS GERAIS: Irwin 2006. n= 16 Crotalaria striata Schrank MINAS GERAIS: Irwin 2018. 2n = 16 Crotalaria sp. MINAS GERAIS: Irwin 2468. n=8 Galactia martii DC. MINAS GERAIS: Irwin 2406. n= 10 (Fig. 3) Galactia martii DC. MINAS GERAIS: Irwin 2506. .. n =10 Indigofera cf. truxillensis MINAS GERAIS: Irwin 2176. — 8 H.B.K. Periandra mediterranea MINAS GERAIS: Irwin 2504. n= 11 (Vell.) Taub. Periandra mediterranea var. MINAS GERAIS: Irwin 2393. n= 11 mucronata (Benth.) Burk. (Fig. 4) CAESALPINIOIDEAE — Chromosome counts for species in the genus Bauhinia (n = 14) and Caesalpinia (n = 12) are consistent with reports for other species in these taxa (Dar- lington and Wylie, 1956). Including the present (Table 1), only 3 species of Copaifera have counts reported for them, 2 from South America and one from Africa (Mangenot and Mangenot, 1957). All were diploid with n = 12. PAPILIONOIDEAE — Chromosome counts for species of Aeschynomene (n — 10), Crotalaria (n — 8), Galactia (n = 10) and Indigofera (n = 8) are consistent with the basic numbers already established for these genera. The chromo- some number of Centrosema coriaceum (n — 11; fig. 2) dif- 1961] Mohlenbrock — Botanizing in Illinois 19 fers from that of the other three species of the genus reported. All of the latter are diploid with n = 10 (Frahm- Leliveld, 1957). Chromosome counts for taxa of Camptosema (n = 11) and Periandra (n = 11) are first reports for these genera. SUMMARY Chromosome counts for 17 taxa of Brazilian Leguminosae are reported, these include first reports for 15 species and two genera (Periandra, x = 11, and Camptosema, x = 11). Centrosema coriaceum (n = 11) was found to have a differ- ent basic number than has been previously reported for the genus. — BOTANY DEPARTMENT AND THE PLANT RESEARCH INSTITUTE, UNIVERSITY OF TEXAS, AUSTIN. LITERATURE CITED DARLINGTON, C. D. and A. P. WYLIE 1956. Chromosome atlas of flowering plants. Macmillan Company. New York. FRAHM-LELIVELD, J. A. 1957. Observations cytologiques sur quel- ques Légumineuses tropicales et subtropicales. Rev. Cytol. Biol. Vég. 18: 273-293. MANGENOT, S. AND G. MANGENOT. 1957. Nombres chromosomiques nouveaux chez diverses dicotylédones et monocotylédones d'Afrique occidentale. Bull. Jard. Bot. (Bruxelles) 27: 639-654. TURNER, B. L. 1956. Chromosome numbers in the Leguminosae. I. Amer. Jour. Bot. 43: 577-581. A DECADE OF BOTANIZING IN ILLINOIS ROBERT H. MOHLENBROCK The publication of Flora of Illinois, Second Edition (Jones, 1950) brought up to date the vascular plants known to occur in the state and at the same time fortunately paved the way for intensive botanizing throughout the state. As documentation for this renewed research on the Illinois flora, no less than eighty publications have appeared since 1950, many of them recording species previously unreported from Illinois. Efforts have been concentrated throughout the state — in the Chicago region by Steyermark, Swink, and Thieret, in the north western section by E. W. Fell, in the east central area by Jones and Ahles, in the west central section by V. Chase, Dobbs, Rexroat, and Winterringer, and in southern Illinois by Voigt and the writer. In addition, Evers has col- lected extensively throughout the state. 20 Rhodora [Vol. 63 Flora of Illinois (l.c.) enumerated 730 genera and 2202 species. In the subsequent decade, 71 genera and 350 spe- cies have been added, representing a thirteen per cent in- crease in the species total. Of these, 64 are species which are considered native to Illinois. It is these latter that are considered the most significant new finds to Illinois, since they denote natural range extensions. The area of greatest concentration of these newly found Illinois plants is across the extreme southern portion of Illinois, in or near the vi- cinity of the Shawnee Hills. Fifty-three of the 64 are known from 10 extreme southern counties. Several of the species fill in previously existing gaps in the distribution, while others mark extensions in the ranges. These are enumerated below according to their distribution. In addition, two endemics have been described from Illi- nois since 1950. These are Aster chasei G. N. Jones from Marshall, Peoria, and Tazewell counties and Cyperus grayi- oides Mohlenbrock from Mason and Whiteside counties. NATIVE SPECIES FILLING IN GAPS IN DISTRIBUTION Asplenium bradleyi Carex debilis Vitis lincecumii Glyceria pallida Lemna valdiviana Penstemon alluviorum Carex austrina Juncus diffusissimus Dicliptera brachiata Carex atherodes Juncus secundus Ruellia caroliniensis Carex caroliniana Polygonatum biflorum Eupatorium fistulosum Carex digitalis Smilax herbacea Rudbeckia bicolor Carex swanii Pilea fontana Solidago buckleyi Carex torta Draba cuneifolia Solidago rugosa Carex texensis Rubus alumnus NATIVE SPECIES EXTENDING RANGE TO THE NORTH Scirpus koilolepis Aristolochia nashii Heliotropium tenellum Carex oxylepis Crataegus collina Gerardia fasciculata Carex physorhyncha Prunus mexicana Galium virgatum Allium mutabile Hypericum lobocarpum NATIVE SPECIES EXTENDING RANGE TO THE NORTHWEST Carex decomposita Arisaema pusilla Tipularia discolor Carex striatula Wolffiella floridana Gaura filipes Carex styloflexa Trillium cuneatum Solidago boottii NATIVE SPECIES EXTENDING RANGE TO THE NORTHEAST Isoetes butleri Jussiaea leptocarpa Solidago strigosa Ranunculus harveyi Penstemon arkansanus 1961] Mohlenbrock — Botanizing in Illinois 21 NATIVE SPECIES EXTENDING RANGE TO THE EAST Talinum calycinum Viola viarum Vernonia crinita NATIVE SPECIES EXTENDING RANGE TO THE SOUTHEAST Pilea opaca NATIVE SPECIES EXTENDING RANGE TO THE SOUTHWEST Lycopodium flabelliforme Trillium erectum Carex emmonsti Solidago uliginosa NATIVE SPECIES EXTENDING RANGE TO THE WEST Rubus enslenii NATIVE SPECIES SHOWING DISJUNCT DISTRIBUTIONS Bromus nottowayanus Scleria reticularis Lipocarpha maculata Cimicifuga cordifolia PUBLICATIONS CONCERNING ILLINOIS VASCULAR PLANTS SINCE 1950 BarLEY, W. M. & J. R. SWAYNE. 1951. New Illinois Plant Records. Am. Midl. Nat. 46: 256. 1951. Some Southern Illinois Plant Records. Trans. Ill. Acad. 44: 40-41. BELL, R. 1956. Aquatic and Marginal Vegetation of Strip Mine Waters in Southern Illinois. Trans. Ill. Acad. 48: 85-91. BREWER, R. & E. D. TRINER. 1956. Vegetational Features of Some Strip-mined Land in Perry County, Illinois. Trans. Ill. Acad. 48: 73-84. BRIGGS, L. 1951. Study of the Compositae of Champaign County, Illinois. Trans. Ill. Acad. 44: 42-44. CALEF, R. T. 1953. Flora and Ecological Analysis of the Vegeta- tion of the Funk Forest Natural Area, McLean County, Illinois. Trans. Ill. Acad. 46: 41-55. DoBBS, R. J. 1953. New Plant Records for Illinois. RHODORA 54: 307. DRAPALIK, D. J. & R. H. MoHLENBROCK. 1960. The Taxonomic Status of Eleocharis elliptica Kunth. Am. Midl. Nat. 63: 143-148. Evers, R. A. 1950. Andropogon elliottii Chapm. in Illinois. RHODORA 52: 45-46. . 1951. Notes on the Illinois Flora. Am. Midl. Nat. 44: 617-621. . 1951. Four Plants New to Illinois Flora. RHODORA 53: 111-113. 1955. Hill Prairies of Illinois. Bull. Ill. Nat. Hist. Surv. 26(5) : 363-446. 1956. Two Plants New to the Illinois Flora. RHODORA 58: 49-50. 1958. Further Notes on the Illinois Flora. RHODORA 60: 142-144. 22 Rhodora [Vol. 63 1959. Illinois Flora: Notes on Eriochloa and Jussiaea. RHODORA 61: 307-308. FELL, E. W. 1956. Notes on a New Hybrid Carex. RHODORA 58: 318-320. 1957. Plants of a Northern Illinois Sand Deposit. Am. Midl. Nat. 58: 441-451. 1958. New Illinois Carex Records. RHODORA 60: 115- 116. FELL, E. W., et. al. 1955. Checklist of Vascular Plants of Boone County, Illinois. Trans. Ill. Acad. 47: 44-54. 1955. Flora of Winnebago County, Illinois. Nature Conservancy, Washington, D. C. 208 pp. Harpy, J. W. 1952. Rudbeckia amplexicaulis Vahl in Southern Illi- nois. RHODORA 54: 119. Jones, G. N. 1957. Vernonia crinita in Illinois. RHODORA 59: 119. 1958. Buchloé dactyloides in Illinois. RHODORA 60: 259-260, JONES, G. N., et al. 1951. Additional Records of Some Illinois Vas- cular Plants. Am. Midl. Nat. 45: 500-503. 1955. Vascular Plants of Illinois. Univ. Ill. Press, Urbana. 594 pp. KILBURN, P. D. 1959. The Forest- Prairie Ecotone in Northeastern Illinois. Am. Midl. Nat. 62: 206-217. MOHLENBROCK, R. H. 1954. Some Notes on the Flora of Southern Illinois. RHODORA 56: 227-228. 1954. Flowering Plants and Ferns of Giant City State Park. Dept. of Conservation and the Ill. State Museum. 24 pp. 1955. Contributions to the Flora of Southern Illinois. RHODORA 57: 319-322. 1955. The Pteridophytes of Jackson County, Illinois. I. Am. Fern Jour. 45: 143-150. 1955. The Pteridophytes of Jackson County, Illinois. II. Am. Fern Jour. 46: 15-22. 1956. Giant City State Park in the Sprine. Bull. Mo. Bot. Gard. 44: 49-56. . 1956. An Unusual Form of Asplenium pin- natifidum. Amer. Fern Jour. 46: 91-93. 1957. Missouri Ozark Plants in Illinois. Bull. Mo. Bot. Gard. 45: 68-72. 1957. Dryopteris clintoniana in Illinois. Amer. Fern Jour. 48: 122-123. 1958. Field and Herbarium Studies. RHODORA 60: 292-297. 1959. An Interesting Place for Orchids in Il- linois. The Living Museum 21: 389-390. 1961 | Mohlenbrock — Botanizing in Illinois 23 1959. A Floristic Study of a Southern Illinois Swampy Area. Ohio Jour. Sci. 59: 89-100. 1959. Plant Communities in Jackson County, Illinois. Bull. Torr. Bot. Club 86: 109-119. - . 1959. A New Species of Cyperus from the Illinois Sand Prairies. Brittonia 11: 255-256. 1960. 'The Cyperaceae of Illinois. I. Cyperus. Amer. Midl. Nat. 63: 270-306. —— . 1960. Isoetes melanopoda in Illinois. Amer. Fern Jour. 51: 181-184. MoHLENBROCK, R. H. & D. J. DRAPALIK. 1960. Eleocharis, Subseries Palustres, in Illinois. Am. Midl. Nat. 64: 224-228. MoHLENBROCK, R. H. & J. W. VoiGT. 1957. Contributions to the Flora of Southern Illinois. RuoporRA 59: 125-128. 1959. A Study of the Filmy Fern, Trichomanes boschianum. Am, Fern Jour. 49: 76-85. 1959. Tipularia discolor in Ilinois. Tax. Index 22: xxxix-xl. 1959. A Flora of Southern Ilinois. Southern Illinois University Press. 420 pp. 1960. New Plant Records from Illinois. RHODORA 62: 239-241. MoHLENBROCK, R. H. & W. R. WEBER. 1959. An Unusual Form of Asplenium bradleyi. Am. Fern Jour. 49: 159-161. Myers, R. M. 1950. A New Station for Marsilea quadrifolia in Illi- nois. Am. Fern Jour. 40: 256. NEILL, J. 1950. Isoetes melanopoda still grows in Illinois. Am. Midl. Nat. 44: 251. STEYERMARK, J. A. 1953. Elymus riparius in Illinois. RHODORA 55: 156. 1955. Epipactis helleborine in Illinois. KRHo- DORA 57: 72. STEYERMARK, J. A. & F. A. SWINK. 1952. Plants New to Illinois and to the Chicago Region. RHODORA 54: 208-213. 1955. Plants New to Illinois and to the Chicago Region. RHODORA 57: 265-268. 1959. Plants New to Illinois and to the Chicago Region. RHODORA 61: 24-27. STEYERMARK, J. A. , F. A. SWINK, & J. W. THIERET. 1957. Plants New to Illinois and the Chicago Region. RHODORA 59: 31-53. SWAYNE, J. R. & W. M. BAILEY. 1953. New Southern Illinois Plant Records. Am. Midl. Nat. 50: 509. Swink, F. A. 1952. A Phenological Study of the Flora of the Chi- cago Region. Am. Midl. Nat. 48: 758-768. THIERET, J. W. 1953. Dipsacus laciniatus in Illinois. RHODORA 55: 268. 24 Rhodora [Vol. 63 1957. Plants New to Illinois and to the Chicago Region. RHODORA 59: 289. THIERET, J. W. & R. A. Evers. 1957. Notes on Illinois Grasses. RHODORA 59: 123-124. THIERET, J. W. & S. F. GLASSMAN. 1958. Grasses New to Illinois and the Chicago Region. RHODORA 60: 264. Vorct, J. W. 1951. Additional Collections of Andropogon elliottii in Southern Illinois. RHODORA 53: 128-130. 1953. Plants Recently Found in Illinois. RHODORA 55: 290-291. 1955. Southern Illinois Flora: Recent Additions. RHODORA 57: 159-160. Voiet, J. W. & R. H. MoHLENBROCK. 1957. An Ozark Odyssey. Southern Illinois University Printing Service. 16 pp. Voret, J. W. & J. R. Swayne. 1955. French's Shooting Star in Southern Illinois. RHODORA 57: 325-332. WINTERRINGER, G. S. 1950. A New Orchid for Illinois. Am. Midl. Nat. 43: 763. 1952. Flowering Arundinaria gigantea in Il- linois. RHODORA 54: 82-83. 1953. Additional Notes on Arundinaria gi- gantea. RHODORA 55: 60. 1954. Breweria pickeringii in Illinois. RHO- DORA 56: 274-275. 1956. Notes on Collinsia violacea. RHODORA 58: 308-309, 1957. Arctium in Illinois. RHODORA 59: 44. 1958. Plant Notes from Illinois. RHODORA 60: 41-43. 1959. Notes on Cyperaceae from Illinois. RHODORA 61: 290-292. WINTERRINGER, G. S. & A. G. VESTAL. 1956. Rock-ledge Vegetation in Southern Illinois. Ecol. Monogr. 26: 105-130. — SOUTHERN ILLI- NOIS UNIVERSITY, CARBONDALE. NEW COMBINATIONS IN GRASSES! JULIAN A. STEYERMARK AND C. L. KUCERA During the course of study of the grass flora of Missouri, the authors have found it necessary to change the categories of several taxa, resulting in the following new combinations: Glyceria septentrionalis Hitche. var. arkansana (Fern.) Steyerm. & Kucera, comb. nov., based on Glyceria arkansana Fern., Rhodora 31: 49. 1929. "Work on this paper was completed during the period when the senior author re- ceived grants-in-aid (G 5623, 7117) from the National Science Foundation. 1961] Steyermark and Kucera — Grasses 25 Muhlenhergia Schreberi Gmel. var. curtisetosa (Scribn.) Steyerm. & Kucera, comb. nov., based on M. Schreberi cur- tisetosa Scribn., Rhodora 9: 17. 1907 (as subspecies) ; M. curtisetosa (Scribn.) Bush, Am. Midl. Nat. 6:35. 1919. As suggested by Gleason (New Ill. Fl. 1: 174. 1952), M. curtisetosa appears to be of doubtful taxonomic status as a species, and seems better regarded as a variety of M. Schreberi, which it closely resembles in general appearance. Sporobolus clandestinus (Bieler) Hitche. var. canovirens (Nash) Steyerm. & Kucera, comb. nov., based on Sporobolus canovirens Nash, in Britton, Man. 1042. 1901; S. asper var. canovirens (Nash) Shinners, Rhodora 56: 30. 1954. There is intergradation in spikelet length, relative length of palea and lemma, and degree of prolongation of the palea between S. clandestinus and S. canovirens. In the extremes of their variation, the two taxa appear quite distinct, but the frequent occurrence of transitional specimens which are difficult to place would indicate the reduction to varietal rank. Since both S. clandestinus var. clandestinus and var. canovirens possess pubescent lemmas, it is believed that this character warrants their being treated as varieties of S. clandestinus, rather than merged, as Shinners has done, as varieties of S. asper, which possesses glabrous lemmas. Sporobolus neglectus Nash var. ozarkanus (Fern.) Stey- erm. & Kucera, comb. nov., based on Sporobolus ozarkanus Fern., Rhodora 35: 109. 1933; S. vaginiflorus var. ozarkan- us (Fern.) Shinners, Rhodora 56: 29. 1954. The glabrous and short, pointed lemmas, together with the relatively less elongated apex of the palea apparently relate S. ozarkanus more closely to S. neglectus than to S. vaginiflorus. The relatively longer spikelets of S. ozarkanus, together with the strongly ciliate orifices of the leaf-sheaths, are points of resemblance between S. ozarkanus and S. vagi- niflorus, but occasional specimens of S. neglectus var. ne- glectus also exhibit ciliate orifices. The strongly ciliate orifices of the leaf-sheaths, believed by Fernald to be char- acteristic of S. ozarkanus, cannot be considered a distin- guishing feature of that taxon. 26 Rhodora [Vol. 63 Leptochloa filiformis (Lam.) Beauv. var. attenuata (Nutt.) Steyerm. & Kucera, comb. nov. based on Oxydenia attenuata Nutt. Gen. Pl. 1: 76. 1818; Leptochloa attenuata. (Nutt.) Steud., Syn. Pl. Glum. 1: 209. 1854. In their extremes, Leptochloa filiformis and L. attenuata appear to be distinct. However, many intergradations are found among specimens in Missouri with both types some- times appearing together. In general, L. filiformis var. filiformis is taller, attaining 1.2 m. in height, and the in- florescence is often larger with 20-100 stiff spikes, while L. filiformis var. attenuata is usually of shorter stature, and the inflorescence is usually smaller with only 10-30 flexuous spikes. Unfortunately, tall-growing plants, characteristic of L. filiformis var. filiformis, are found with the aristate glumes and smaller lemmas characteristic of L. filiformis var. attenuata, while low-growing plants, characteristic of L. filiformis var. attenuata, occur with the acute glumes and larger lemmas characteristic of L. filiformis var. filiformis. The same lack of correlation is noted occasionally between the greater or lesser length of the glumes with respect to the upper floret and the height of the plant. Deam (Grasses of Indiana, p. 198. 1929) also had difficulty in determining whether a specimen placed by him in Leptochloa attenuata should warrant specific or varietal status. — INSTITUTO BO- TANICO DEL MINISTERIO DE AGRICULTURA Y CRIA, CARACAS, VENEZUELA, AND UNIVERSITY OF MISSOURI, COLUMBIA, MIS- SOURI. RHODODENDRON MAXIMUM IN HOPKINTON AND HARRISVILLE, NEW HAMPSHIRE A. R. HODGDON AND RADCLIFFE PIKE! It becomes apparent that some of the many early reports by non-botanists of Rhododendron colonies in New Hamp- shire may be accurate, the occurrence in Hopkinton being a case in point. In 1874, C. S. Hitchcock stated that Rho- dodendron maximum grew in that township. On page 543 of volume I of his “Geology of New Hampshire" he made 1Published with the approval of the Director of the New Hampshire Agricultural Experiment Station as Scientific Contribution No. 249. 1961] Hodgdon and Pike — Rhododendron Maximum 27 the following comment as part of a general discussion of the few and scattered colonies of the species in the State, ““— I have got traces of it in Hopkinton and Hooksett." It is cer- tain from Hitchcock’s other comments that he knew the plant well. But it is by no means clear that he had more than verbal assurance that it grew wild in these two town- ships; nor does he state who his informants were. We have been inclined to rule out of consideration all such reports because in our own experience we have found that most of the general public in Maine and New Hampshire do not know R. maximum. In this instance however, the report of a station in Hopkinton proves to be authentic. Early in March 1959, Mr. Henry Mock, a senior at the University of New Hampshire and a resident of Contoocook brought us a specimen from a small wild colony which he stated grew on the farm of a Mr. Frank Kimball in Hopkin- ton. On June 2 we were shown the colony by Mr. Mock and Mr. Kimball. The stand is particularly vigorous and luxuriant with rather uniform stems some of the tallest of which were 12 or more feet high. There were few flower- buds for the current season and there were no seedlings nor small plants in the area, in this respect differing markedly from most of the other stands in both Maine and New Hampshire where seedlings and young plants are often numerous. The colony is rectangular in shape and is about 50 feet wide by 150 feet long. The regular shape of the colony, the uniform growth and the absence of young plants made it seem planted rather than wild. However, Mr. Kim- ball convinced us that the colony was quite natural. He recalled his father Herbert Kimball, who was born about 1862, stating that in his youth, the colony was vigorous, but that somewhat later (about 65 years ago according to Frank Kimball), the bigger protecting trees were cut off for lum- ber, after which the Rhododendrons declined seriously. In recent years with the growth of suitable species of shading and protecting trees in the vicinity of the stand it has made a remarkable recovery. The Harrisville Rhododendrons to our knowledge have not been reported previously. Mr. Tudor Richards of Dublin 28 Rhodora [Vol. 63 first learned of this colony from local residents a few months ago and made arrangement with Mr. Merle Jones of Han- cock who guided Mr. Richards and the senior author to the station on June 9, 1960. This part of Harrisville and ad- jacent Hancock is heavily wooded with considerable swamp- land and intervening rocky upland. The colony is very close to the Hancock line and is about three quarters of a mile east of Skatutakee Lake. Rhododendron plants are found over a total area of about one half acre. A dense growth of middle-sized to large shrubs occupies the wetter places while an equal quarter acre of drier footing on the eastern side has some isolated large plants as well as some scattered small individuals which must have started as seedlings in recent years. While a few plants are close to 10 feet in height, most of them fall short of this. It is evident that the plants comprising this colony have not yet attained their full growth: at least in other colonies that we have studied the biggest plants have nearly always been considerably taller than those in Harrisville. Here the older plants of earlier times presumably have been replaced by seedlings or rejuvenated sprout growth. This is a colony that undoubtedly will be improving during the next few years. This makes a total of 11 townships in which we have ob- served wild stands of Rhododendron in New Hampshire. These are Albany, Pittsfield, Barnstead, Hopkinton, Gran- tham, Manchester, Mason, Wilton, Fitzwilliam, Harrisville and Richmond. Are there still other stations in New Hamp- shire? In “The History of Weare” by William Little pub- lished in 1888 there is mention of the occurence of both Mountain laurel and Rhododendron in the township. Lean- der W. Cogswell in 1880 in his “History of the Town of Henniker" states that “rhododendron or river laurel adorns banks of Contoocook” which might refer to Kalmia latifolia. We have been told of a colony near the eastern end of Squam Lake probably in Sandwich. Thus there may be other sta- tions but it seems to us that we have now a fairly complete list of Rhododendron colonies in New Hampshire. Several years of diligent sleuthing on our part have resulted in dis- 1961] Pease — Campanular Persistence 29 closing only one New Hampshire station (Harrisville) that had not been reported in some published work. And this stand was well enough known locally to be a topic of con- versation at a party. — DEPARTMENT OF BOTANY AND DE- PARTMENT OF HORTICULTURE, UNIVERSITY OF NEW HAMP- SHIRE, DURHAM, NEW HAMPSHIRE. CAMPANULAR PERSISTENCE. — While walking on the rail- road in Randolph, N. H., near the former station of Appa- lachia, in the summer of 1920, I observed, on a gravelly embankment, one good-sized clump, about six inches in diameter and the same in height, of a many-stemmed Cam- panula, with small pale blue flowers on naked flexuous ped- uncles. Leaving most of the plant undisturbed, I placed a portion in the herbarium of the New England Botanical Club (Pease 18093), and by analysis and comparison with speci- mens in the Gray Herbarium identified the plant as Cam- panula divaricata Michx., which is now described in the eighth edition of Gray’s Manual as growing “in dry woods and rocky slopes, w. Md.,W.Va. and Ky.,s. to. Ga. and Ala." In my Vascular Flora of Coós County, N. H. (1924), p. 345, I have reported the plant as rarely adventive and persistent in 1923. Over the years from 1923 to the present I have watched the fate of this little pilgrim, and several years ago, when the railroad track was heavily reballasted with unpromising gravel, found its site deeply buried. For several years I con- sidered it as gone beyond recovery, but then it rose again from the gravel, and my annual visits recommenced. Then came another calamity ; some four or five years ago the track was again reballasted, this time with even more unpromising cinders, and I had again to mourn the loss of the Campanula. This summer (1960) it occurred to me to look again, and lo! there again it was at its accustomed place, rising through cinders as it had previously through gravel. Forty years, then, at least — for I do not know how long before 1920 it was first established here — this delicate little plant has survived an austere diet and violent attacks upon its security. It shows no disposition to increase, but whether 30 Rhodora [Vol. 63 this may be due to a lack of the insects needed to fertilize it or to other causes I know not. Mr. Walter Deane reported to the Botanical Club (RHo- DORA 4:243-244; 10:203-204) on the persistence of Cepha- lanthus occidentalis L. for 43 years in a pig-sty at Shelburne, N. H.; equally or perhaps even more notable is the experi- ence of this delicate little wild-flower, about eight feet from a heavily ballasted railroad track and at least five hundred miles from its natural home. — ARTHUR STANLEY PEASE, RANDOLPH, N. H. Volume 62, No. 744, including pages 325-364 was issued January 25, 1961. DUPLICATE BOOKS FOR SALE These books have library book plates and are used copies. some worn, some in need of binding. BUILLIARD, P. Dictionnaire élémentaire de botanique ou Exposition par ordre alphabétique, . . . Paris, 1783. vii, 242-- [8] p. 10 plates, 9 colored. folio ........................ $25.00 KERNER VON MARILAUN, A. The natural history of plants... From the German by F. W. Oliver . . . New York, 1895. Two volumes. Q ...........:.4 erret nnne nennen nennen enne 10,00 Moss, C. E. The Cambridge British Flora. volume 2 only plus the volume of plates. Cambridge, 1914. folio .................... 20.00 REICHENBACH, H. G. L. Icones florae Germanicae . . . In- complete. Volumes 1 through 17 bound in 9 and vol. 18 incomplete, pages 1-72, pl. 1202-1331. Volume 1 is the 2nd ed. rev. of 1850 with 192 plates. Text in Latin, the plates are uncolored. Leipzig, 1850, 1837-56. folio ............ 200.00 SMITH, JAMES EDWARD. Flora brittanica. London. 1800-04. Three volumes 8? .....cccccecccccccccccesseecceecesssssssseeeeccesseeeaeeeneeeeeeeeees 25.00 SMITH, JAMES EDWARD. A selection of the correspondence of Linnaeus and other naturalists from the original manu- scripts. London, 1821. 2 volumes. 8? ................2....... 25.00 ADDRESS THE LIBRARIAN GRAY HERBARIUM OF HARVARD UNIVERSITY 22 Divinity Avenue, Cambridge 38, Mass. ALAW Rerencnce Lisaaay MAR 31 1961 Dodota JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN Associate Editers IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Vol. 63 February, 1961 No. 746 CONTENTS: Notes on Lesser Antillean Ferns. George R. Proctor ............ 31 A New Name for the Puberulent Sesssile-leaved Uvularia. Robert L. Wubur te e eei TUE IHE 36 New Records for North Dakota. O. A. Stevens ........... eese 39 A Fern New to Rhode Island, Richard L. Champlin .............. 46 Fourteenth Report of the Committee on Plant Distribution. R. C. Bean, A. F. Hill and R. J. Eaton ................... 47 A New Manual for Ohio Vascular Plants (Review). Robert W. Long E E a a EE aE NG 55 An Indispensable Manual of Tropical Marine Botany (Review). I. Mackenzie Lamb .................. E eerte 57 The Nem England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume. Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able, All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass. Printed by THE LEXINGTON PRESS, INC. Lexington, Mass. Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 February, 1961 No. 746 NOTES ON LESSER ANTILLEAN FERNS GEORGE R. PROCTOR During the current preparation of a volume on the Pteri- dophytes for the forthcoming “Flora of the Lesser Antilles”, at east five new species have come to light, and a number of new combinations have become necessary. The purpose of the present paper is to describe three of these new species,’ and to make a total of 17 new combinations. HYMENOPHYLLACEAE Trichomanes trigonum Desv., var. fimbriatum (Backh.) Proctor, comb. nov., based on Trichomanes fimbriatum Backh., Cat. 12. 1861; Gard. Chron. 1862:44. Syn. T. su- perbum. v.d.B., Ned. Kr. Arch. 5::203. 1863 (not Backh., 1862). CYATHEACEAE Cyathea hodgeana Proctor, sp. nov. Caudex 5 m. altus, 7.5 em. diametro, gemma apicali squamis brun- nescentibus nitidis eroso-ciliatis dense obtecta. Stipites atrobrunnei minutissime et sparse puberuli, inermes, basi paleis deciduis eis gemmae apicalis similibus obtecti. Laminae ovatae, bipinnato-pinnati- fidae, ca. 1 m. longae. Rhachides primariae et secundariae brunneae et dense pallido-brunneo-furfuraceae; rhachidibus secundariis et costis YThis research supported by Grant No. G-4441 from the National Science Foundation, Division of Biological and Medical Sciences, for work on the flora of the Lesser Antilles in cooperation with Dr. Richard A. Howard of the Arnold Arboretum, Harvard University. ?Special thanks are due to Dr. L. M. Perry of the Arnold Arboretum, Harvard Uni- versity for the preparation of the three Latin diagnoses. 31 32 Rhodora [Vol. 63 supra dense brunneo-pubescentibus, subtus costis brunneas planas paleas numerosas ferentibus, et costulis minutas bullatas paleas pallido-brunneas ferentibus. Pinnae articulatae, anguste oblongae, usque ad 30 cm. longae et 11 cm. latae acuminatae, longe petiolulatae (petiolulis usque ad 1.5 em. longis). Pinnulae lineari-oblongae, pinna- tifidae, usque ad 5.5 em. longae et 1.1 cm. latae, acutae vel sub- acuminatae, petiolulatae (petiolulis 1-2 mm. longis). Segmenta ob- longa, usque ad 4 mm. longa et 2.5 mm. lata, apice rotundata vel obtusa, crenulata. Venae 4-5 ad latera, simplices vel uni-furcatae. Sori supramediani, lamina supra eis scrobiculata. Indusium nullum. Receptaculum parvum, capitatum, paraphyses capillares ferens. TYPE from moist forests bordering the Pegoua River in vicinity of Deux Branches, Dominica, West Indies, W. H. & Barbara T. Hodge 3420, collected May 6 & 7, 1940 (Holotype in the Gray Herbarium, consisting of 3 sheets). Cyathea imrayana Hook., var. caribaea (Jenm.) Proctor, comb. nov., based on Cyathea caribaea Jenm., Ferns B.W.I. & Guiana 57. 1898. Cnemidaria grandifolia (Willd.) Proctor, comb. nov., based on Cyathea grandifolia Willd., Sp. Pl. 5:490. 1810. POLYPODIACEAE Nephrolepis faleata (Cav.)C.Chr., forma furcans (Moore in Nicholson) Proctor, comb. nov., based on N. davallioides [var. ?] furcans Moore in Nicholson, Dict. Gard. 2:445, fig. 682. 1885; Schneider, Book of Choice Ferns 2: 590-592, fig. 144. 1893. ?Syn. N. exaltata var. monstri osa v. A.v.R., Ma- layan Ferns 162. 1908. This entity is usually filed in herbaria as a form of N. biserrata, but the present writer considers this identification to be incorrect. It is now associated with N. falcata on the basis of comparison with New Guinea speci- mens identified as that species in the herbarium of the Ar- nold Arboretum. However, it should be noted that the type of N. falcata came from the Philippines, and differs from the New Guinea material in having a scaly rachis. A mono- graphic study of these plants will probably demonstrate that the New Guinea and Philippines populations are at least varietally distinct. Forma furcans is definitely allied with the New Guinea taxon. According to Schneider, forma fur- cans was first introduced into European cultivation from Australia; the West Indian plants probably originated from 1961] Notes on Lesser Antillean Ferns 33 descendants of these Australian plants cultivated in Eng- land. Thelypteris muscicola Proctor, sp. nov. Rhizoma erectum apice paleaceum, paleis ligulato-attenuatis, usque ad 8 mm. longis, 1-1.5 mm. latis, brunneis, sparse pubescentibus. Stipes plerumque 9-15 cm. longus, glaber. Lamina pinnato-pinnatifida, ob- longo-lanceolata vel lanceolata, 45-80 cm. longa, 15-22 cm. lata, apice acuminata, basi abrupte angustata, cum pinnis perreductis 2-4-jugis. Rhachis et costae supra strigillosae, subtus fere glabrae; pilis omnibus acicularibus, non hamatis. Pinnae lineari-lanceolatae usque ad 2 cm. latae, sessiles, alte pinnatifidae. Segmenta oblonga, rotunda vel apice subacuta, plana, 3-4 mm. lata. Venae 6-10-jugae, simplices, segmentis basalibus exceptis. Sori supramediales ovales. Indusium ciliatum erec- tum margini longo cum vena conjunctum. Sporangia glabra. TYPE from mossy woodland on upper west slope of Nevis Peak, Nevis, West Indies, elev. 2500-3000 ft., Proctor 19354, collected March 5, 1959 (Holotype in the Arnold Arboretum Herbarium; isotypes at the Insti- tute of Jamaica and the State University of Utrecht). This species differs from T. germaniana by its glabrous stipes and tissue beneath, by the lack of hamate hairs, and by the elongate, laterally-attached indusium. It differs from T. hydrophila by its very much greater size; by the longer, darker, less hairy rhizome-scales ; by the rachis being nearly glabrous beneath, and lacking aerophores at the bases of the pinnae; and by the smaller indusium which lacks minute capitate glands. To no other West Indian species does it bear any near resemblance. Thelypteris antillana Proctor, sp. nov. Rhizoma breve erectum dense paleaceum; paleis anguste deltoideo- lanceolatis, usque ad 8 mm. longis, spadiceis, pubescentibus aliquando dentatis. Stipes plerumque 5-22 cm. longus, basi paleaceo, undique puberulus et minute capitato-glandulosus. Lamina pinnato-pinnatifida, lanceolata usque ad 50 cm. longa, 17 cm. lata, apice acuminata, basi abrupte angustata, cum pinnis perreductis 1-3-jugis. Rhachis minute capitato-glandulosa; partibus omnibus undique minute pubescentibus, subtus pilis in parte hamatis. Pinnae lineari-lanceolatae, usque ad 1.5 cm. latae acuminatae, sessiles, alte pinnatifidae. Segmenta numerosa anguste oblonga acuta, margine revoluto 2-3 mm. lato. Venae 7-10- jugae, integrae. Sori rotundi, mediales vel supramediales. Indusium persistens, dense longi-ciliatum. Sporangia glabra. TYPE from stunted elfin woodland on upper southwest spur of Verchild's Mountain, below Dodans (Dos D'Ans) Pond, St. Kitts, West Indies, elev. 2500-2700 ft., Proctor 19587, collected March 19, 1959 34 Rhodora [Vol. 63 (Holotype in the Arnold Arboretum Herbarium; isotvpe at the Insti- tute of Jamaica). Additional material: st. KITTS, Britton & Cowell 529, from summit of Mt. Misery, collected in 1901 (NY, US); this num- ber was reported with doubt as Dryopteris oligocarpa by C. Christen- sen, Smiths. Misc. Coll. 52: 371. 1909. A duplicate at the U. S. National Herbarium (perhaps the same sheet seen by Christensen) has been identified by C. V. Morton as D. hydrophila, but this material differs from the latter species in a number of significant details. DOMINICA, W. H. & Barbara T. Hodge 1857, from between Laudat and Freshwater Lake (GH). This species is very closely related to T. oligocarpa, as shown especially by the nature of the hairs on the underside of the blade. It differs from that variable species, however, by its larger, much more chaffy rhizome-scales ; by its medial or slightly supramedial (instead of nearly marginal) sori; and by its persistent, long-ciliate indusia. Comparison with various South American relatives of T. oligocarpa has failed to disclose any material that could be considered conspecific. T. antillana averages about twice the size of T. hydrophila, and further differs from the latter in its puberulent and minutely capitate-glandular stipes, by lacking aerophores along the rachis at base of pinnae, by having more numerous veins per segment (usually 7 - 10 pairs instead of 5 - 6), and by the larger indusium on which none of the hairs are gland- ular. Thelypteris invisa (Swartz)Proctor, corrected citation (based on Aspidium invisum Swartz, Jour. Bot. Schrad. 18002:34. 1801. Syn. Nephrodium invisum (Swartz) Desv., Mém. Soc. Linn. Paris 6:257. 1827); incorrectly given as *(Desv.) Proctor" in Rhodora 61:306. 1959(1960). I am indebted to Mr. C. V. Morton for pointing out this necessary correction. Ctenitis protensa (Afz.) Copel., var. funesta (Kunze) Proc- tor, comb. nov., based on Aspidium funestum Kunze, Linnaea 9:96. 1834. Alston (Kew Bull. 1932:309) raised this entity to specific rank (as Thelypteris), but the present writer pre- fers to follow Christensen in associating it at the varietal level with the typically African C. protensa. Ctenitis excelsa (Desv.) Proctor, comb. nov., based on Poly- podium excelsum Desv., Mém. Soc. Linn. Paris 6:249. 1827. 1961] Notes on Lesser Antillean Ferns 35 Syn. Dryopteris excelsa. (Desv.) C.Chr., Ind. Fil. 264. 1905 (excl. syn.). Dicranoglossum desvauxii (Klotzsch) Proctor, comb. nov., based on Taenitis desvauxii Klotzsch, Linnaea 20:431. 1847. Syn. Eschatogramme desvauxii (Klotzsch) C.Chr., Dansk Bot. Ark. 6(3) :37. 1929. The generic name Eschatogramme was published as a nomen nudum and must be rejected in favor of Dicranoglossum J.Sm., Bot. Voy. Herald 232. 1854. This has been pointed out previously by Pichi-Sermolli, Web- bia 9:365. 1953. Grammitis serricula (Fée)Proctor, comb. nov., based on Polypodium serricula Fée, Gen. Fil. 238. 1852. Grammitis knowltoniorum (Hodge)Proctor, comb. nov., based on Polypodium knowltoniorum Hodge, Amer. Fern Jour. 31 (3) :105, pl.1, figs. 4-6. 1941. Grammitis anfractuosa (Kunze ex Klotzsch) Proctor, comb. nov., based on Polypodium anfractuosum Kunze ex Klotzsch, Linnaea 20:375. 1847. Syn. Polypodium induens Maxon, Bull. Torr. Bot. Club 32:75. 1905. Grammitis jubaeformis (Kaulf.) Proctor, comb. nov., based on Polypodium jubaeforme Kaulf., Flora 6:364. 1823. Grammitis tenuicula (Fée) Proctor, comb. nov., based on Polypodium tenuiculum Fée, Gen. Fil. 239. 1852. Grammitis taxifolia (L.)Proctor, comb. nov., based on Polypodium taxifolium L., Sp. Pl. 1086. 1753. Grammitis pendula (Swartz) Proctor, comb. nov., based on Polypodium pendulum Swartz, Prodr. Veg. Ind. Occ. 131. 1788. Grammitis sericeolanata (Hooker)Proctor, comb. nov., based on Polypodium sericeolanatum Hooker, Sp. Fil. 4:221. 1862. Grammitis cultrata (Willd.) Proctor, comb. nov., based on Polypodium. cultratum Willd., Sp. Pl. 5:187. 1810. Grammitis mollissima (Fée) Proctor, comb. nov., based on Polypodium mollissimum Fée, Mém. Foug. 11:47, pl.12, fig.2. 1866. — INSTITUTE OF JAMAICA, KINGSTON, JAMAICA, WEST INDIES. 36 Rhodora [Vol. 63 A NEW NAME FOR THE PUBERULENT SESSILE-LEAVED UVULARIA ROBERT L. WILBUR! For more than three-quarters of a century the puberulent sessile-leaved bellwort, which has a range extending from New Jersey and Pennsylvania south to Georgia and as far west as West Virginia and Tennessee, was known scientific- ally as Uvularia puberula, the name given to it by Michaux in 1803. Asa Gray in 1839, as reported by Fernald (Rho- dora 41:537. 1939) identified the specimen in Walter’s her- barium, upon which “Anonymos pudica” was based, as Uvularia puberula. He made no new combination since the convention generally followed by Gray adopted the first epi- thet given to a species in the accepted genus as its proper binary name. In segregating the sessile-leaved species as the genus Oakesia, Sereno Watson in 1879 provided the new combina- tion O. puberula. A new complexity was introduced in 1889 by N. L. Britton who described what has proven to be the glabrous variant of this species as “Oakesia sessilifolia var. (?) nitida.” In 1893 Morong provided the combination “Uv- ularia sessilifolia nitida” and by 1908 Mackenzie had con- cluded that the plant from the New Jersey pine-barrens was more closely related to U. puberula. However, thinking its differences to be of specific rank, he made the combination U. nitida. Prior to Mackenzie’s publication, however, Small, agreeing that the sessile-leaved bellworts were generically distinct from Uvularia, published the name Oakesiella. This was necessary since Watson’s Oakesia (1879) was a later homonym of Oakesia Tuckerm. (1842), a later synonym of Corema D. Don (c.1826), a genus in the Empetraceae. Oakesia, it might be noted in passing, was a perfectly per- missible generic name in the Liliaceae under the Internation- al Code until about 1935 but such names were always taboo Grateful acknowledgment is made to the National Science Foundation for a grant of research funds to Duke University (NSF-Grant 5636) which make the present study possible. 1961] Puberulent Sessile-leaved Uvularia 37 under the provisions of the Code under which Small pub- lished (see Canon 16 b, Bull. Torrey Club 31:257. 1904). Small’s new generic name in his Flora of 1903 required among others the accompanying new combination of Oake- siella puberula. In 1933 Small maintained the genus Oake- siella and, overlooking the combination O. nitida made by Heller in 1910, in the appendix of the Manual again made the combination O. nitida although attributing it to Macken- zie. Fernald (Rhodora 37:407-409. 1935) discussed the thinner-leaved, glabrous variant and provided the combina- tion U. puberula var. nitida. Four years later Fernald (Rho- dora 41:536-538. 1939), investigating the ‘‘Anonymos pudic." of Walter, concluded that there was no reason to doubt Asa Gray's determination. He therefore made two new combinations, U. pudica and U. pudica var. nitida, even though such names based upon Anonymos of Walter “were subject to ridicule by some of the younger English botanists." In the nomenclatural sessions of the 1950 Stockholm Con- gress however these “younger English botanists” including and led by the then septuagenarian Dr. T. A. Sprague pre- vailed upon the majority of the delegates to declare definitely illegitimate “binary combinations of a specific epithet with the word Anonymos." (Article 23). The price of this in- sistence upon consistency will be at least nineteen changes in citation which have been largely ignored by most Ameri- can taxonomists. Unfortunately a new combination is re- quired for the above discussed bellwort. Almost all of these necessary changes in citation will result merely in the appearance of the relatively unfamiliar J. F. Gmelin, the author of the 13th edition of Linnaeus' Systema Naturae, in place of the long-familiar Thomas Wal- ter. Gmelin's Systema appeared in 1791, only three years after Walter's Flora, and for the most part Walter's species were included there. Most of the ‘““Anonymos” genera were either assigned new generic names or their species included in previously established genera. Gmelin however some- times made no reference whatever to Walter's species and once provided an entirely new epithet. 38 Rhodora [Vol. 63 Turning then to Gmelin (Syst. Nat. ed. 13. 2:546. 1791), we find the following: 414. ERYTHRONIUM. Cor. 6 petala, campanulata. Nectario tuberculis 2 petalorum alternorum basi adnatis. Dens canis. 1. E. foliis oppositis. Jacq. fl. austr. 5. app. t. 9. carolinia- 2. E. foliis alternis. Walt. flor. carol. p. 122. num. Index Kewensis wrongly attributed Erythronium caroli- nianum to Walter's Flora (p. 122), (an obvious error since the genus Erythronium is not included as such) and equated the name to the synonymy of Erythronium americanum over which, if true, it would have priority. For "Anonymos pudica", however, which of course is the same plant, Index Kewensis equates it to Uvularia perfoliata. This determina- tion is in accord with the disposition of this name by Mi- chaux (Fl. Bor.-Am. 199. 1803). Dietz (Ann. Mo. Bot. Gard. 39:225. 1952) argues that “the name Uvularia pudica should be discarded and Michaux's U. puberula restored." His stated reason was that Walter described the plant as having “foliis amplexicaulibus" which Dietz felt not to be true. The term *amplexicaule" is not too far from the mark for the sessile leaves, even in fruiting specimens, are often so broadly rounded at base that they do “clasp the stem." Even Michaux, whose name Dietz accepts, described the leaves of this species as *"subamplexicaulibus". Fernald has quoted notes, (Rhodora 41:537. 1939) made in 1839, in which Gray, while examining the contents of Walter’s herbarium, wrote that 'Anonymos (Erythronio aff.) pudica! = Uvularia puberula.” Certainly Asa Gray would never have mistaken U. puberula for one of the per- foliate species; as Fernald wrote 'there is no reason to doubt Gray's identification ; there is every reason to accept it." Apparently the first author to recognize this species fol- lowing Walter was J. F. Gmelin in 1791. This work in re- gard to Walter's species at least, was merely a compilation. The new species described under Anonymos by Walter were 1961] New Records for North Dakota 39 either provided with new generic names, utilizing no more information than provided by Walter and in most cases providing far less, or assigned to an established genus. Following Walter's hint (*Erythronio affinis?") as to the affinities of this species, Gmelin made it the second species of Erythronium but substituted the epithet carolinianum. Such a substitution by Gmelin is legitimate by our present Code since Walter’s epithets with “Anonymos” are no long- er "taken into consideration for purposes of priority". Hence the puberulent sessile-leaved bellwort again must sub- mit to a change in name. In the formal listing of synonymy below I include those names intended to apply to the glabrous or nearly glabrous representative which I do not believe merits formal recognition. Uvularia caroliniana (J. F. Gmelin) Wilbur, comb. nov. Anonymos pudic. Walt., Fl. Car. 123. 1788. nom. illeg. Art 23. Erythronium carolinianum J. F. Gmel., Syst. Nat. ed. 13. 2: 546. 1791. Uvularia puberula Michx., Fl. Bor.-Am.1: 199. 1803. O«kesia pube- rula (Michx.) S. Wats., Proc. Am. Acad. 14: 269. 1879. Oakesia ses- silifolia var. (?) nitida Britt., N. Y. Acad. Sci. 9: 13. 1889. Uvularia sessilifolia nitida (Britt.) Morong, Mem. Torrey Club 5: 111. 1893. Oakesiella puberula (Michx.) Small, Fl. SE. U. S. 271. 1903. Uvularia nitida (Britt.) Mackenzie, Torreya 8:14. 1908. Oakesiella nitida (Britt.) Heller, Muhlenbergia 6: 83. 1910. Uvularia puberula var. nitida (Britt.) Fern., Rhodora 37: 407. 1935. Uvularia pudica (Walt.) Fern., Rhodora 41:536. 1939. Uvularia pudica var. nitida (Britt.) Fern., Rhodora 41: 536. 1929.— DEPARTMENT OF BOTANY, DUKE UNI- VERSITY, DURHAM, N. C2 NEW RECORDS FOR NORTH DAKOTA' O. A. STEVENS Since the publication of my Handbook (Stevens, 1950), some 60 species have been added to the State list. Some are recently introduced weeds, some have been found for the first time, while others had been incorrectly identified. A few records for species previously reported are included in the following list where they are of special interest in ex- tending the known range. Moore (1951) noted some addi- tions and changes in status. Some of these are included here 1Contribution from the North Dakota Institute for Regional Studies. Journal Paper No. 12. 40 Rhodora [Vol. 63 and some are still problematical. The nomenclature follows Gray's Manual, 8th ed., for the most part. Specimens of most species have been deposited in the U. S. National Her- barium, in herbaria of the University of California, Univer- sity of Minnesota, Canadian National Museum and Science Service of Canada. Equisetum variegatum Schleich. Some poor material, Stevens 1218, collected at Valley City in 1950 was determined by C. V. Morton as var. nelsonii A. A. Eaton. Equisetum sylvaticum L. Tongue River, close to Pembina-Cavalier County line in 1958, Stevens and Moir 1767, Botrychium lunaria (L.) Sw. Low, sandy prairie, McHenry County, Stevens 1530. Cystopteris dickieana Sim. West of Grassy Butte, McKenzie County, Stevens and Moir 2243. Determined by C. V. Morton who notes that its status as a species is uncertain. Woodsia oregana D. C. Eaton. Material formerly reported as W. obtusa has been identified by R. T. Clausen as W. oregana, var cath- cartiana (Robins.) Morton. Pteretis pensylvanica (Willd.) D. C. Eaton. This was found in the Turtle Mts., Bottineau County, by Duane Green in 1956 and later col- lected by D. R. Moir. It occurs also in one ravine near Huff, Morton County. Onoclea sensibilis L. A second record is from Pembina County, Stevens in 1954. Juniperus virginiana L. Near Fargo in 1959, J. R. Nelson. I had wondered why this had not become established from seeds scattered by birds. It is not native within probably 200 miles. Typha angustifolia L. This was recognized as frequent in the Fargo area in 1958, It apparently has been spreading rather rapidly. Sparganium chlorocarpum Rydb. Lower Souris Wildlife Refuge, Hotchkiss and Hammond in 1955. Sparganium multipedunculatum (Morong) Rydb. N. E. Hotchkiss writes that there is a specimen in the U. S. National Herbarium col- lected at Riverside Marsh, south of Mandan, Metcalf 337. Ruppia occidentalis S. Wats. Iverson Lake, Burke County, Hotch- kiss and Knowlton 4656. Potamogeton richardsonii (A. Benn.) Rydb. should replace P. per- foliatus. Potamogeton friesii Rupr. Hotchkiss writes that he collected this in Burke County. Elodea nuttallii (Planch.) St. John. McHenry and Burke Counties, Hotchkiss. Bromus marginatus Nees. Black (H. T.) Butte, Slope County, Stevens and Moir 2297, det. Swallen. 1961] New Records for North Dakota 41 Bromus brizaeformis Fisch. and Mey. Sent by Bert Miller from Tuttle, Stutsman County, in 1956. In a grass field, probably not permanently established. Puccinellia cusickii Weatherby. Sheridan County, Stevens 2095, identified by J. R. Swallen. This was from a small area without glacial till that had also Artemisia cana, Opuntia polyacantha and Oenothera caespitosa, our easternmost records of the last two. Specimens col- lected in 1960 from Ward, Burleigh and Slope Counties are also re- ferred to it. Poa fendleriana (Steud.) Vasey. Bottineau County, Stevens 2209. Eragrostis diffusa Buck. A specimen from Minot by Bolley in 1891 has been determined by J. R. Swallen. Eragrostis perplexa L. H. Harvey. Reported (Harvey, 1954), from Mandan. Hordeum montanense Scribn. Late season material from Minot, Curtis Benton in 1953, seems to be this form. Hordeum pusillum Nutt. Medora in 1953, Stevens 1445, and Rhame in 1954, V. Facey. Danthonia spicata (L.) Beauv. Killdeer Mountains in 1960, Stevens and Moir 2254. Quite common along a woodland trail. Sporobolus airoides Torr. A flowering stalk was grown in the green- house from a piece of sod collected in Bowman County by Clayton Quinnild in 1957. Eleocharis parvula (R. & S.) Link. Burleigh County, Hotchkiss in 1940, and Grand Forks County, F. C. Seymour in 1951. Eriophorum viridicarinatum (Engelm.) Fern. One specimen from Walhalla, Pembina County, Stevens in 1938, seems this species, all others E. angustifolium Honck. Carex foenea Willd. (C. siccata Dew.) Turtle Mts., Bottineau County, Stevens 1610, det. Hermann. Carex prairea Dewey. Larimore, Stevens 1722; Turtle Mts., Stevens & Moir 1776; McHenry County, Stevens and Moir 1779. Carex aenea Fern. Towner County, Stevens 1531; Tongue River, Cavalier County, Stevens and Moir 1759, det Hermann; Williams County, Stevens in 1956. Carex molesta Mack. Richland County, Stevens 1347, det. Hermann. Carex peckii Howe. Richland, Pembina, Grand Forks, Bottineau, Dunn and Ward Counties. Carex richardsonii R. Br. McHenry and Cass Counties, Stevens 1534 and 1715. Carex garberi Fern. Burke County, Hotchkiss 6714. Carex haleana Olney. Richland County, Stevens 1257, det Hermann; Benson County by Lunell as C. shriveri. Carex parryana Dewey. Burke County, Hotchkiss 6715. Carex hallii Olney. Eddy County, Stevens 1209; reported by Mac- kenzie for Benson County and an old specimen, Bell 61, is from Rich- land County. 42 Rhodora [Vol. 63 Carex substricta (Küken.) Mack. McLean County, Stevens 1070, det. Hermann. Carex torreyi Tuckerm. Additional records are, Killdeer Mts. and Grassy Butte in 1960 and Moorhead, Minn. in 1959. Calla palustris L. Pembina County, Stevens 1338; Rolette County, Stevens and Moir 1926. Abundant in Tongue River west of Cavalier, Pembina County. Juncus gerardi Loisel. Fargo, Stevens 1497; Richland County, Bell 674; both determined by Hermann. Calochortus pudica (Pursh) Spreng. A specimen overlooked was col- lected at Medora, Billings County, by E. C. Moran who wrote that he saw it also in Bowman County. Uvularia sessilifolia L. Tongue River, Cavalier County, Stevens and Moir 1754. Populus acuminata Rydb. One tree was shown me near Medora, Billings County (south side of Dedication Butte, Stevens 1286), by Virgil Weiser; Dr. L. D. Potter found a small grove near Vim in Slope County in 1958. X Betula sandbergii Britton. One tree in poor condition at Carpenter Lake, Rolette County, Stevens and Moir 1775. Betula pumila L., var. glandulifera Regel. Near Larimore, Grand Forks County, Stevens 1431; Carpenter Lake, Rolette County, Stevens and Moir 1772; noted by Hotchkiss in McHenry County. Pilea fontana (Lunell) Rydb. Three of our specimens, all from the Richland County area, are this species, Ulmus pumila L. Cass, Burleigh, Stark and Mountrail Counties. Volunteer seedlings are frequent. Rumex stenophyllus Ledeb. First collected in this area by J. F. Brenckle in 1951 (No. 5106) in Spink County, South Dakota, and de- termined by Paul Aellen. In 1954 it was observed in quantity at Fargo (Stevens 1492) and later collected in McLean, Burleigh, Morton, Oliver and McKenzie Counties, Lóve and Bernard (1950) have given other records and a description of the plant. Rumex domesticus Hartm. This had been noted as a puzzling form and is well established; 21 counties, all parts of the State. Some old specimens are this species: Richland County, Bell in 1908; Pembina County, Bergman in 1912; Cass County, Stevens in 1920. Folygonum douglasii Greene. McKenzie and Slope Counties in 1960, Stevens and Moir 2240. Frequent in small water channels on north- facing slopes. Atriplex glabriuscula Edmonst. A considerable colony at Fargo in 1954, Stevens 1507. Salsola collina Pall. In 1949, one of my former students, Lars Rei- tan, called attention to some “different” Russian thistles in Barnes County and a specimen was identified by Paul Aellen as S. collina. We noted it with increasing frequency and in recent years it has sec emed 1961] New Records for North Dakota 43 the more common form. The bracts are appressed, the calyx wings very short and erect. Mirabilis albida (Walt.) Heimerl. One specimen from Oliver County, Stevens in 1938, and one from Valley City, Stevens in 1949, are per- haps this species. Both approach M. hirsuta. Mirabilis linearis (Pursh) Heimerl. Most specimens previously re- ferred to M. albida belong here. Cerastium scopulorum Greene? Black (H. T.) Butte, Slope County, Stevens and Moir 2292, seems to be this species rather than C. arvense. It was in a shady glade at the foot of a north-facing slope, July 14, 1960. Only one duplicate (us). Cerastium vulgatum L. Numerous complaints of this in lawns in Fargo began about 1950. Apparently it was generally established but gradually subsided. In an area on our campus, seeded in 1949, it was vigorous in 1955, weak in 1956 and not found the following year. Delphinium ajacis L. Rocket Larkspur. Streets at Belfield, Stark Co., Stevens and Moir 2238. Apparently well established. Myosurus aristatus Benth. Kenmare, Ward Co., Hotchkiss 6701. Chorispora tenella (Pallas) DC. Williston in 1960 by D. G. Hotch- kiss, county extension agent. Polanisia trachysperma T. & G. Previously reported as P. graveo- lens Raf. Ribes hirtellum Michx. Rolette County, Bergman 1503, was over- looked. To this have been added: Richland County, Stevens 1341; Cav- alier, Stevens in 1956; Pembina, Stevens 1327; Griggs, Stevens in 1952 and 1953; Bottineau, Stevens in 1957. Potentilla palustris L. N. E. Hotchkiss writes that he saw it in McHenry County but no specimens are available. Geum rivale L. Tongue River west of Cavalier, Pembina County, Stevens 1232. Agrimonia gryposepala Wallr. Killdeer Mts., Dunn County, Stevens & Kluender in 1935; Richland County, Stevens 1752. Caragana arborescens Lam. One bush at Valley City, Barnes County, seemed an escape but might have been planted. It is strange that volunteer plants are not found. The seeds germinate freely in bare soil by planted material. One at Cavalier, Pembina County in 1959 was a few rods from plantings. Althaea officinalis L. This volunteer indefinitely from plantings. Viola nephrophyllaGreene. Many of our specimens formerly labeled V. papilionacea have been changed to this. It is often abundant in boggy meadows. Viola bernardi Greene. Four old specimens are referred to this: Fargo, Bolley and M. Field in 1891 and 1892; Ransom County, Bell 653; Hankinson, A. D. Stoesz in 1934. Viola sarmentosa Nutt. A sterile specimen from Killdeer Mts., Dunn County, Stevens and Kluender in 1935, has been determined as this by 44 Rhodora [Vol. 63 Norman H. Russell. Further collections are needed to verify this report. Viola incognita Brainerd. Tongue River west of Cavalier, Pembina County, Stevens 1332, 1395, det. Russell. Elaeagnus angustifolia L. I had wondered why we found no volun- teer plants of this but since 1950 I find them in many places, especially in low areas, in fence rows and under telephone wires. Oenothera rhombifolia Nutt. The specimen so reported seems to be O. strigosa (Rydb.) Mack. & Bush. Myriophyllum verticillatum Michx. One sheet from Grant County, Bell 729, Sanicula gregaria Bickn. Richland County, Stevens 1344, det. Con- stance. Menyanthes trifoliata L. N. E. Hotchkiss reports seeing this 5 miles north of Towner, McHenry County, in 1932. Apocynum medium Greene. Some material from Fargo, Stevens 1251, seems to be this species. It was in an area where both A. andros- aemifolium and A. sibericum grew (now destroyed). Asclepias sullivantii Engelm. Specimens previously referred to A. purpurascens L. Asclepias lanuginosa Nutt. Dickinson, Stark County in 1960, J. R. Nelson, is a second record. I had been looking for it for 40 years and the Dickinson area has received more than usual attention. Cuscuta glomerata Choisy. This was found near Lisbon, Ransom County in 1950 by Snorri Thorfinnson; Stevens 1374. Navarretia propinqua Suksd. is our species according to Mason's treatment. Hackelia virginiana (L.) I. M. Johnston, At Fargo, Stevens 1495, a quantity of this was noted in woods along the Red River. It seemed unaffected by a leaf spot which was abundant on H. americana. An older specimen, Stevens in 1934, is H. virginiana and one from Valley City in 1957. Lappula redowskii, var. cupulata (A. Gray) M. E. Jones. At Medora in 1954 occasional plants, Stevens 1441, with cupulate nutlets, among abundant normal var. occidentalis, seemed pathological. These had not been seen since the previous report. Cryptantha macounii (Eastw.) Payson. A specimen from White Earth, Mountrail County, T. A. Haigh in 1898, was referred to this by A. J. Breitung. Amsinckia retrorsa Suks. The Fargo specimen, Stevens 629, should be called this according to Constance. Lycopsis arvensis L. Dickinson, Stark County in 1955, Stevens 1205. One plant in a field. Determined by F. Chisaki. Solanum interius Rydb. McClusky, Sheridan County, Stevens 1645, determined by Constance. We have been getting an occasional speci- men, none in quantity. 1961] New Records for North Dakota 45 Solanum dulcamara L. Mrs. P. F. Debertin, Parshall, Mountail County, sent a specimen in 1952 that she found growing by her house. A plant at Fargo I am told has been so growing for several years. Linaria dalmatica (L.) Mill Roadside, Steele, Kidder County, Stevens 1194, and Dickinson, Stark County, Stevens in 1955. Linaria canadensis (L.) Dumort. One small plant on a sheet of Veronica peregrina, from Grant County, Bell 359, was recognized by F. W. Pennell. Mimulus guttatus DC. In a stream from a flowing spring near Ink- ster, Grand Forks County. This was first brought to our attention in 1956 by Dr. Vera Facey of the University of North Dakota, which has the area as a study preserve. Utricularia intermedia Hayne. McHenry County, Hotchkiss 4502 (US). Linnaea borealis L. Killdeer Mountains in 1960, Stevens and Moir. Previously from Turtle Mountains, E. E. Hotchkiss. Sherardia orientalis Boiss. & Hohen. Waste ground by railroad, Fargo in 1945, L. R. Waldron. Determined by Bernice G. Schubert. Galium labradoricum Wieg. McHenry County, Hotchkiss 4502 (US). Sambucus canadensis L. This is not native but seems more or less established. A colony on the Agricultural College grounds at Fargo (now destroyed) was said not to have been planted though many years before it had been planted in at least two other places (long since de- stroyed). It was received from Park River, Walsh County, where it was said to have been present for some years and was found at Shey- enne, Eddy County (Stevens and Moir 1661) in a native grove; these last two both in 1956. Galinsoga parviflora Cav. Fargo in 1958, Stevens 2029. These plants are becoming frequent about yards. Bellis perennis L. A few plants in bloom in a lawn at Fargo in 1955. Arctium lappa L. Kulm, LaMoure County, Brenckle in 1912; Arvilla, Grand Forks County, D. G. Walp in 1936. Arctium tomentosum Mill. Fargo in 1956, Stevens 1634. Petasites palmatus (Ait.) Gray. Turtle Mts., Bottineau County, Stevens and Moir 1790. Centaurea maculosa Lam. Two or three plants by railroad at Fargo, Stevens 1633; destroyed the next year. Stephanomeria tenuifolia (Torr.) Hall. Little Bad Lands, Stark County, V. Facey in 1952, Stevens in 1955. Tragopogon pratensis L. Fargo, Lee in 1891, and Lisbon, Bell 686, are this species but we have been unable to find it in recent years. Taraxacum kok-saghyz Rodin. Fargo, Stevens 1703. 'This was grown in a field plot in 1941 and 1942 but has not been observed in the area since. Some seed was cleaned at our building where a few plants con- 46 Rhodora [Vol. 63 tinue to thrive near the west wall but are not aggressive. — NORTH DAKOTA STATE UNIVERSITY, FARGO, LITERATURE CITED Harvey, L. H. 1954. New entities in North and Middle American Eragrostis. Bull. Torrey Club 81: 405-410. LOvE, DORIS AND FRÈRE JEAN-PAUL BERNARD (1958). Rumex steno- phyllus in North America. Rhodora 60: 54-57. Moore, JOHN W. 1951. Review. Ecology 51: 359-360. STEVENS, O. A. 1950. Handbook of North Dakota Plants. North Dakota Inst. Reg. Studies. Fargo. A FERN NEW To RHODE ISLAND. — Asplenium montanum, not previously recorded as growing in Rhode Island, has been found in a well-established colony on sandstone cliffs at West Glocester, Providence County. The fern grows with Polypodium virginianum on a west-facing ledge (about 150 feet long) where it receives mixed sun and shade. On Au- gust 18, 1960 when John Hudson and I collected specimens, there appeared to be no surplus moisture on the cliffs, yet the plants were luxuriant in crevices or hung from the seams of stony recesses. Records, mostly from the early 1900’s place Asplenium montanum in these nearby Connecticut towns: North Ston- ington, Franklin, Scotland, Salisbury, ete. Whether it has only recently invaded Rhode Island, or has simply eluded detection for many years is a moot question. John Hudson who led me to its location was in turn shown it by Lewis Carpenter of Hope, R. I. in 1957. Not until Hudson and I visited it, however, were any specimens collected. These have been deposited with the University of Rhode Island Herbarium and The New England Botanical Club. — RICH- ARD L. CHAMPLIN, JAMESTOWN, R. I. 1961] Committee on Plant Distribution 47 FOURTEENTH REPORT OF THE COMMITTEE ON PLANT DISTRIBUTION The thirteenth report included the Dicotyledoneae from Portulacaceae through Lauraceae. The present report deals with the families from Papaveraceae through Platanaceae, taken in the order of the eighth edition of Gray's Manual. The data for these reports have been compiled from the material found in the herbarium of the New England Botan- ical Club and in the Gray Herbarium. PRELIMINARY LISTS OF NEW ENGLAND PLANTS — XXXIX The sign + indicates that an herbarium specimen has been seen, the sign — that a reliable printed record has been found and the sign * is used for those plants which are not native in the New England area. Me. N.H. Vt. Mass. R.I. Conn. PAPAVERACEAE Adlumia fungosa (Ait.) Greene + + *Argemone alba Lestib. f. *Argemone mexicana L. *Chelidonium majus L. + + Corydalis aurea Willd. Corydalis flavula (Raf.) DC. Corydalis sempervirens (L.) Pers. + Dicentra canadensis (Goldie) Walp. — Dicentra Cucullaria (L.) Bernh. + *Dicentra eximia (Ker) Torr. *Dicentra formosa (Andr.) Walp. *Dicentra spectabilis (L.) Lem. + + + +++ ++++ ++] + + ++t+ ++++ +++++ *Eschscholtzia californica Cham. *Fumaria officinalis L. T + T *Glaucium flavum Crantz *Macleaya cordata (Willd.) R. Br. + — *Papaver dubium L. *Papaver Rhoeas L. + *Papaver somniferum L. + Sanguinaria canadensis L. + CAPPARIDACEAE *Cleome serrulata Pursh +++ +++ +++ | Er ++ $4+t4++4++ d + *Cleome spinosa Jacq. Polanisia graveolens Raf. -H CRUCIFERAE *Alliaria officinalis Andrz. —— * Alyssum Alyssoides L. + * Alyssum saxatile L. Arabis canadensis L. + Arabis divaricarpa Nels. Arabis Drummondii Gray + + ++ +++ tra 4 + + ++ 48 Rhodora [Vol. 63 Me. N.H. Vt. Mass. R.I. Conn. Arabis glabra (L.) Bernh. + + +o — + Arabis hirsuta (L.) Scop. var. pyenocarpa (M. Hopkins) Rollins Arabis laevigata (Muhl.) Poir. Arabis lyrata L. Arabis missouriensis Greene *Arabis procurrens Waldst. & Kit. *Arabidopsis Thaliana (L.) Heynh. Armoracia aquatica (Eat.) Wieg. *Armoracia lapathifolia Gilib. Barbarea orthoceras Ledeb. *Barbarea verna (Mill.) Aschers. *Barbarea vulgaris R. Br. *Barbarea vulgaris var. arcuata (Opiz) Fries *Barbarea vulgaris var. brachycarpa Rouy & Foucaud *Berteroa incana (L.) DC. *Berteroa mutabilis ( Vent.) DC. *Brassica hirta Moench *Brassica juncea (L.) Coss. *Brassica juncea var. crispifolia Bailey *Brassica Kaber (DC.) L. C. Wheeler var. pinnatifida (Stokes) L. C. Wheeler *Brassica Kaber var. Sehkuhriana (Reichenb.) L. C. Wheeler *Brassica Napus L. *Brassica nigra (L.) Koch *Brassica oleracea L. *Brassica Rapa L. Braya humilis (C. A. Mey.) Robins, var. leiocarpa (Trautv.) Fern. Cakile edentula (Bigel.) Hook. *Camelina microcarpa Andrz. *Camelina sativa (L.) Crantz *Capsella Bursa-pastoris (L.) Medic. *Capsella Bursa-pastoris var. bifida Crépin *Capsella rubella Reut. Cardamine bellidifolia L. Cardamine bulbosa (Schreb.) BSP. -- Cardamine Douglassii (Torr.) Britt. Cardamine flexuosa With. — *Cardamine hirsuta L. *Cardamine impatiens L. Cardamine Longii Fern. Cardamine parviflora L. var. arenicola (Britt.) O. E. Schulz Cardamine pensylvanica Muhl. Cardamine pensylvaniea var. Brittoniana Farw. *Cardamine pratensis L. *Cardamine pratensis f. plena G. Beck Cardamine pratensis var. palustris Wimm. & Grab. *Cardaria Draba (L.) Desv. *Chorispora tenella (Willd.) DC. *Coronopus didymus (L.) Sm. + *Coronopus procumbens Gilib. *Conringia orientalis (L.) Dumort. 4 + 4- + + + ++ + ++++ + n ++ +4] + E +4+4+4+4+4+ ++ + + + +++ l- + +++ ++ ++H+++++ +i + + +++ ++ +++ +++++ + 4+4+4+4++4++4+ +++++ + + ctl + ++] ++ + +I+]I+ + ++i +++4++ ++++ trc ++ ++++ + ++ | + ++ +4+++4++4+4 + ++ +++ ++ + + + 4 + + + + ++++4++4+4+ + + | + + ++ +++ + 1961] Committee on Plant Distribution Me. N.H. Vt. Dentaria anomala Eames Dentaria diphylla Michx. + + E Dentaria incisifolia Eames Dentaria laciniata Muhl. + + Dentaria maxima Nutt. -ł}- + Descurainia pinnata (Walt.) Britt. var. brachycarpa (Richards.) Fern. + -} Descurainia Richardsonii (Sweet) O. E. Schulz *Descurainia Sophia (L.) Webb ++ *Diplotaxis muralis (L.) DC. *Diplotaxis tenuifolia (L.) DC. Draba Allenii Fern. Draba arabisans Michx. Draba glabella Pursh Draba lanceolata Royle + + ae +++ Draba reptans (Lam.) Fern. *Draba verna L. *Draba verna var. Boerhaavii Van Hall *Eruca sativa Mill. *Erucastrum gallicum (Willd.) O. E. Schulz *Erysimum cheiranthoides L. Erysimum ineonspicuum (S. Wats.) MacM. +++ +++ | *Erysimum pannonicum Crantz *Erysimum repandum L. *Hesperis matronalis L. *Iberis amara L. *Iberis umbellata L. * Lepidium campestre (L.) R. Br. * Lepidium densiflorum Schrad. * Lepidium latifolium L. * Lepidium perfoliatum L. *Lepidium ruderale L. *Lepidium sativum L. Lepidium virginicum L. *Lobularia maritima (L.) Desv. *Lunaria annua L. *Malcolmia maritima R. Br. *Nasturtium officinale R. Br. *Nasturtium officinale var. microphyllum (Boenn.) Thell. ats + +++ + + |I+++++ ++ ++ ++ +++ + + +++] | *Nasturtium officinale var. siifolium (Ileichenb.) Koch *Neslia paniculata (L.) Desv. *Raphanus Raphanistrum L. *Raphanus sativus L. +++ ++ | *Rapistrum rugosum (L.) All. *Rorippa amphibia (L.) Bess. Rorippa islandica (Oeder) Borbas Rorippa islandica var. Fernaldiana Butt. & Abbe Rorippa islandica var. hispida (Desv.) Butt. & Abbe *Rorippa sessiliflora (Nutt.) Hitche. *Rorippa sylvestris (L.) Bess. *Sisymbrium altissimum L. *Sisymbrium Loeselii L. *Sisymbrium officinale ( L.) Scop. + ++ + + 4 + + ++ + + 49 Mass. R.I. Conn, ++++ +++ ++++++ + dE dBRRGGBGAB + + +++++++++ +++ TRES +++++ + I++ i ELI + + 50 Rhodora [Vol. 63 N.H. Vt. Mass. R.I. Conn, *Sisymbrium officinale var. leiocarpum DC. + + + *Sisymbrium orientale L. Subularia aquatica L. *Teesdalia nudicaulis (L.) R. Br. *Thlaspi arvense L. RESEDACEAE *Reseda alba L. *Reseda lutea L. *Reseda Luteola L. *Reseda odorata L. SARRACENIACEAE Sarracenia purpurea L. DROSERACEAE Drosera filiformis Raf. Drosera intermedia Hayne Drosera linearis Goldie Drosera rotundifolia L. Drosera rotundifolia var. comosa Fern. PODOSTEMACEAE Podostemum ceratophyllum Michx. CRASSULACEAE *Sedum acre L. *Sedum alboroseum Baker *Sedum anopetalum DC, *Sedum purpureum (L.) Link Sedum Rosea (L.) Scop. *Sedum rupestre L. *Sedum sarmentosum Bunge *Sedum spurium Bieb. *Sedum Telephium L. Sedum ternatum Michx. E *Sempervivum tectorum L. -+ -+ Tillaea aquatica L. + SAXIFRAGACEAE * Astilbe japonica (Morren & Dene.) Gray Chrysosplenium americanum Schwein. + + + + *Deutzia scabra Thunb. *Heuchera americana L. *Hydrangea paniculata Sieb. *Hydrangea quercifolia Bartr. *Hydrangea radiata Walt. Mitella diphylla L. Mitella prostrata Michx. Mitella nuda L. + Parnassia glauca Raf. + + + ++ + + +8 + + + + + + ++ + + E u- + + — + ++ + +++ ++ ++ E E + I+++ + + + ++ + + + + + TERR + $444 + + | ++ ++ E LEE ++ + ++ ++ +++ + +++ 4 Penthorum sedoides L. *Philadelphus coronarius L. *Philadelphus inodorus L. *Philadelphus pubescens Loisel. Ribes americanum Mill. -+ Ribes eynosbati L. +- Ribes glandulosum Grauer + *Ribes Grossularia L. + Ribes hirtellum Michx. Ribes hirtellum var. calcicola Fern. EE Ribes hirtellum var. saxosum (Hook.) Fern. +++ + ++ E + +$+t4+4¢)+4+4+444+4+4+4+4+44+ +++++ LEG TEE 1961] Committee on Plant Distribution 51 Me. N.H. Vt. Mass. R.I. Conn. Ribes laeustre (Pers.) Poir. -4 -- + + + Ribes missouriense Nutt. + *Ribes nigrum L. + + + -4 + *Ribes odoratum Wendland f. + + + + Ribes rotundifolium Michx. + + *Ribes sativum Syme + + -= + -+ 4. Ribes triste Pall. + + + + + Saxifraga aizoides L. + Saxifraga Aizoön Jacq. var. neogaea Butters + + Saxifraga oppositifolia L. -+ Saxifraga pensylvanica L. + + + + + + Saxifraga rivularis L. + Saxifraga stellaris L. var. comosa Poir. + Saxifraga virginiensis Michx. + ae + 4 + + Tiarella cordifolia L. + + + F + HAMAMELIDACEAE Hamamelis virginiana L. E + + + + + Hamamelis virginiana var. parvifolia Nutt. — + Liquidamber Styraciflua L. + PLATANACEAE Platanus occidentalis L. — + -+ -L + + The Saxifragaceae, Hamamelidaceae and Platanaceae were treated by C. H. Knowlton in 1916 (Rhodora XVIII- 245-248). The Capparidaceae, Resedaceae, Sarraceniaceae, Droseraceae, Podostemaceae and Crassulaceae were also treated by C. H. Knowlton in 1917 (Rhodora XIX-217-219). These groups are also included in the present report. As noted in previous reports the introduced plants have in general tended to spread in the intervening years. The native species show the same distribution as in these earlier reports with very few exceptions. Podostemum ceratophyl- lum has now been collected in both New Hampshire and Vermont. Parnassia glauca and Ribes rotundifolium are now represented by specimens from New Hampshire. Saxifraga Aizoön before known from northern Vermont only has now been collected on Mt. Katahdin in Maine. The geographical areas are in general the same as used previously. Again a large number of the plants included are not native to New England and the percentage of such plants is greater than in the two previous reports, fifty-six percent. If the Cruciferae alone were considered, the per- centage of introduced plants would be sixty-six. I. GENERALLY DISTRIBUTED. — Cardamine pensylvanica, Lepidium virginicum, Rorippa islandica var. Fernaldiana, Sarracenia purpurea, 52 Rhodora [Vol. 63 Drosera intermedia, D. rotundifolia, Chrysosplenium americanum, Ribes hirtellum. The number of species and varieties considered as generally distributed, eight, is small for the large number of the plants treated, Ia. GENERAL, EXCEPT CAPE COD. — Corydalis sempervirens, Rorippa islandica var. hispida, Ribes americanum, Tiarella cordifolia. Tiarella cordifolia also is absent from Rhode Island and from Bristol and Plymouth Counties in Massachusetts. Ib. GENERAL, EXCEPT CAPE COD AND THE MAINE COAST EAST OF THE KENNEBEC RIVER. — Sanguinaria canadensis, Arabis glabra, Dentar'a diphylla. Ic. GENERAL, EXCEPT CAPE COD AND WASHINGTON COUNTY, MAINE, — Arabis Drummondii. IIa. NORTHERN-NOT OR NOT MUCH SOUTH OF 43^. — Barbarea orth- oceras, Subularia aquatica. Barbarea orthoceras is distinetly northern and is represented in New England by only three stations, Fort Kent and St. Francis, Maine and Mt. Washington, New Hampshire. Subu- laria aquatica is also northern in its general distribution but in New England it has not been reported from Northern Maine; in fact there are only five collections in all from Maine. IIb. NORTHERN-NUMEROUS STATIONS SOUTH OF 43°.— Mitella nuda, Ribes glandulosum, R. lacustre, R. triste. III. ALPINE-ARCTIC. — Cardamine bellidifolia, Sedum Rosea, Saxi- fraga aizoides, S. Aizoón, S. oppositifolia, S. rivularis, S. stellaris var. comosa. Sedum Rosea in New England follows along the rocky coast of eastern Maine and has two stations in Vermont. Saxifraga aizoides, S. Aizoon, and S. oppositifolia are alpine but they are also calcicolous, as their occurrence only in northern Vermont indicates. S. Aizoon, al- though it prefers a calcareous soil, is not restricted to it and is found on Mt. Katahdin. IVa. SOUTHERN-GENERAL SOUTH OF 45^.— Cardamine parviflora var. arenicola, C. pensylvanica var. Brittoniana, Hamamelis virgini- ana. Although Cardamine pensylvanica var. Brittoniana is represent- ed by very few stations, it seems to belong to this category. IVb. SOUTHERN-GENERAL SOUTH OF 45^ BUT NOT ON MAINE COAST EAST OF THE KENNEBEC RIVER. — Saxifraga pensylvanica. IVc. SOUTHERN-GENERAL SOUTH OF 45^ BUT NEITHER CAPE COD NOR WASHINGTON COUNTY. — Adlumia fungosa, Dicentra Cucullaria, Ara- bis hirsuta var. pycnocarpa, Penthorum sedoides, Saxifraga virginien- sis. Va. CHIEFLY THE THREE SOUTHERN STATES, — Platanus occidentalis. Vb. CHIEFLY THE THREE SOUTHERN STATES BUT NOT ON CAPE COD, — Arabis canadensis, 1961] Committee on Plant Distribution 53 Vc. CHIEFLY THE THREE SOUTHERN STATES BUT NEITHER CAPE COD NOR WESTERN MASSACHUSETTS. — Arabis missouriensis, Draba reptans. VI. SOUTHWESTERN NEW ENGLAND CHIEFLY. — Cardamine Doug- lassii, Heuchera americana, Liquidambar Styraciflua. Cardamine Doug- lassii is also a calciphile. VIII. WESTERN NEW ENGLAND CHIEFLY. — Dicentra canadensis, Arabis lyrata, Dentaria incisifolia, D. laciniata, D. maxima, Sedum ternatum, Mitella diphylla, Arabis lyruta occurs in southwest Connec- ticut and Massachusetts and has two stations in southwest Vermont. Dentaria incisifolia occurs only in Sharon, Connecticut. VIII. COASTAL PLAIN. — Drosera filiformis, Ribes hirtellum var. calcicola. Ribes hirtellum var. calcicola is placed in this category with some hesitation. Outside of New England it is in general a calciphile. With- in our area it simulates a coastal plain type of distribution as it oc- curs in Rhode Island, on Cape Cod, Martha's Vineyard and Nantucket and has several stations between the Kennebec River and Mount Desert. This variety is not found on the coastal plain south of New England. IXa. CALCICOLOUS-CHIEFLY WEST OF THE CONNECTICUT RIVER IN THE SOUTH, IF IN THE EAST MOSTLY NORTH OF 45^. — Corydalis aurea, Ara- bis divaricarpa, Braya humilis var. leiocarpa, Cardamine flexuosa, Draba arabisans, D. glabella, D. lanceolata. Corydalis aurea is con- fined to western Vermont with the exception of Norwich in the eastern part of the state where it is represented by a collection made in 1877. Cardamine flexuosa has been collected at Wallingford and Smuggler’s Notch in Vermont. Braya humilis var. leiocarpa is confined to the Willoughby region of Vermont. Draba arabisans has been collected in Maine and Vermont only. Draba glabella has one station at Willough- by, Vermont. IXb. CALCICOLOUS-ALSO ON BASIC AND NEUTRAL SOILS IN EASTERN MASSACHUSETTS. — Arabis laevigata, Cardamine pratensis var. palus- tris, C. bulbosa, Parnassia glauca. X. MARITIME-IN VICINITY OF COAST, NO INLAND STATIONS. — Cakile edentula, Tillaea aquatica. XI. ESTUARINE. — Cardamine Longii, found only at Bowdoinham, Maine. A collection made at Newton Upper Falls, Massachusetts rep- resents an authenticated introduction. XII. MISCELLANEOUS. — Corydalis flavula, Polanisia graveolens, Armoracia aquatica, Dentaria anomala, Descurainia pinnata var. bra- chycarpa, D. Richardsonii, Erysimum inconspicuum, Rorippa islandica, Drosera linearis, D. rotundifolia var. comosa, Podostemum ceratophyl- lum, Mitella prostrata, Ribes Cynosbati, R. hirtellum var. saxosum, R. rotundifolium. Corydalis flavula is known only from Meriden and Middletown, Connecticut. Polanisia graveolens is confined to the Lake Champlain area of Vermont. Armoracia aquatica has two stations in n 54 Rhodora [Vol. 63 the Lake Champlain region at Ferrisburg and Shelburne, Vermont and a questionable one at East Boothbay, Maine. Dentaria anomala occurs at Plainville and Orange, Connecticut. Descurainia pinnata var, brachycarpa is occasional in the White Mountains of New Hamp- shire and the Lake Champlain area of Vermont. Descurainia Richard- sonii is apparently native at Moscow, Maine. Specimens from South Berwick, Maine, Acton, Massachusetts and Waterbury, Connecticut are presumably adventive. Erysimum inconspicuum is a western species adventive along railroad tracks at Canton and Falmouth, Maine and Gorham and Berlin, New Hampshire. Rorippa islandica has a very spotty distribution: two stations in New Hampshire, six in Massachusetts, one at Block Island, and two in Connecticut. Dro- sera linearis has been collected only in the marly bog at Crystal, Maine. Drosera rotundifolia var. comosa has been found at Mt. Desert, Maine and Waterville, New Hampshire. Podostemum ceratophyllum is occa- sional in some swift streams and rivers south of 45^. Mitella prostrata is known only at Gaylordsville, Connecticut. Ribes Cynosbati does not conform to any of the categories that have been used. It seems to occur chiefly in western New England but has four stations in West- ern Maine and appears on both Martha's Vineyard and Nantucket. Ribes hirtellum var. saxosum is represented by one collection from Mt. Desert, Maine. Ribes rotundifolium has been found at Crawford Notch, New Hampshire and at Danbury and Meriden, Connecticut, XIII. INTRODUCED SPECIES-GENERAL. — Armoracia lapathifolia, Bar- barea vulgaris, Brassica juncea, B. nigra, B. Rapa, Camelina sativa, Capsella Bursa-pastoris, Erysimum cheiranthoides, Lepidium densiflor- um, Sisymbrium altissimum, S. officinale var. leiocarpum, Thlaspi arvense, Ribes sativum. XIIIa. INTRODUCED SPECIES-NEITHER CAPE COD NOR NORTHERN MAINE. — Hesperis matronalis, Rorippa sylvestris, Sisymbrium offici- nale, XIIIb. INTRODUCED SPECIES WITH SOUTHERN TENDENCIES-CHIEFLY SOUTH OF 43^?.— Chelidonium majus, Fumaria officinalis, Alyssum alyssoides, Arabidopsis Thaliana, Barbarea vulgaris var. arcuata, Berteroa incana, Brassica hirta, B. Kaber var. pinnatifida, B. Kaber var. Schkuhriana, Camelina microcarpa, Cardamine pratensis, Con- ringia orientalis, Descurainia Sophia, Draba verna, D. verna var. Boerhaavii, Lepidium campestre, L. ruderale, L. sativum, Lobularia maritima, Nasturtium officinale, N. officinale var. microphyllum, Nes- lia paniculata, Raphanus Raphanistrum, R. sativum, Sedum. acre, S. purpureum, Sempervivum tectorum, Deutzia scabra, Philadelphus in- odorus, Ribes Grossularia, R. odoratum. XIIIc. INTRODUCED SPECIES-SPORADIC. — Papaver Rhoeas, P. somni- ferum, Barbarea verna, Brassica Napus, B. oleracea, Sisymbrium Loeselii, Reseda alba, R. lutea, Hydrangea paniculata (apparently well established at Lincoln, Massachusetts), Ribes nigrum. 1961] A New Manual for Ohio Vascular Plants 55 XIIId. INTRODUCED SPECIES-LOCAL. — Argemone alba, A. mexicana, Dicentra eximia, Eschscholtzia californica, Glaucium flavum, Macleaya cordata, Papaver dubium, Cleome spinosa, Alliaria officinalis, Alyssum saxatile, Barbarca vulgaris var. brachycarpa, Berteroa mutabilis, Brassica juncea var. crispifolia, Capsella Bursa-pastoris var. bifida, C. rubella, Cardamine hirsuta, Cardaria Draba, Chorispora tenella, Coronopus didymus, C. procumbens, Diplotax^s muralis, D. tenuifolia, Eruca sativa, Erucastrum gallicum, Erysimum repandum, Iberis ama- ra, Lepidium latifolium, L. perfoliatum, Lunaria annua, Nasturtium officinale var. siifolium, Rapistrum rugosum, Rorippa amphibia, Tees- dalia nudicaulis, Reseda Luteola, R. odorata, Sedum alboroseum, S. anopetalum, S. rupestre, S. sarmentosum, S. spurium, S. Telephium, Ph'ladelphus coronarius, P. pubescens. The following local introduced plants are represented by only one station in New England and it is doubtful if they should be considered as a real part of our flora: Dicentra formosa (Danvers, Massachu- setts); D. spectabilis (Westport, Connecticut); Cleome serrulata (Lawrence, Massachusetts); Arabis procurrens (Wakefield, Massa- chusetts); Cardamine impatiens (Peterborough, New Hampshire); Erysimum pannonicum (Westfield, Massachusetts); Iberis umbellata (Randolph, New Hampshire); Malcolmia maritima (Shelburne, New Hampshire); Rorippa sessilifolia (Salem, Massachusetts) ; Sisymbri- um orientale (Milton, Massachusetts); Astilbe japonica (Providence, Rhode Island); Hydrangea quercifolia (Norwalk, Connecticut); H. radiata (Fairfield, Connecticut). — R. C. BEAN, A. F. HILL, AND R. J. EATON. A NEW MANUAL FOR OHIO VASCULAR PLANTS.' — The ap- pearance of a new manual for the identification of Ohio vas- cular plants is worthy of special notice. For many years students have had to rely on J. H. Schaffner's manual? which has long since become out-dated and out-of-print. Dr. Weis- haupt, who is curator of the Ohio State Herbarium, has prepared a book that meets this pressing need for an up-to- date accounting of Ohio plants. Essentially, her manual is a series of dichotomous keys. Vascular plants are categorized as “Pteridophyta” or “Sper- matophyta". Within these groups artificial keys to families lVascular Plants of Ohio a manual for use in field and laboratory, by Clara G. Weishaupt. 309 pp. 84⁄4 x 11, 1960. Harold L. Hedrick, Columbus, Ohio, publisher. $5.50. ?Field manual of the flora of Ohio and adjacent territory. Columbus, Ohio. 1928. 56 Rhodora [Vol. 63 are given with page references to generic keys following the family name. A brief, general characterization of the family is presented followed by the key to genera. A further refer- ence to a genus name and number leads one to the key for species. This key is headed by a brief genus description. Other than those that appear in the key, there are no descrip- tions or notes for the species, although common names are given. In the back of the manual two special keys are in- cluded, a key to woody plants in leaf and one to woody plants in winter condition. A glossary and index are at the end of the book. The various diagnostic keys are the most praiseworthy feature of the manual ; for the most part they are clear, con- cise, and highly usable. Obviously, much careful work went into their construction. There is a minimum of technical language which should facilitate their use by students. The keys to grasses and to sedges are excellent. The key to cru- cifers, depending on both flowering and fruiting material being available, is less successful. Some keys are prefaced with a “suggestion” to aid in their use. The general approach of the author to species is conserva- tive rather than modern. Seldom are varieties or forms described. Very little attention is given to natural hybrids. One is included for Populus, but none for Viola or Quercus. No distinction is made between native and introduced plants. Thus, Ginkgo and Rheum are included without reference to their origin or distribution. On the other hand, Picea has been left out. Perhaps the most serious fault in the treat- ment of species has been the omission of any ecological, geographical, or economic comments. An additional refer- ence work will always be necessary for one who wishes to know as much about a species as he would ordinarily want to know. Species descriptions and comments would have added immeasurably to the value of the manual for students. It would be unfair to criticize this manual, which has been designed primarily as a diagnostic key, entirely on grounds of what it could have been. Specialists may well be disap- pointed in the treatment of species, and others may wish 1961] Manual of Tropical Marine Botany 51 for more information, such as maps, geologic history, draw- ings, and chromosome numbers, but as a clear and refined set of keys this book has few equals. One can hope that a revised edition will include not only these keys but also the usual supplementary information. — RoBERT W. LONG, OHIO WESLEYAN UNIVERSITY. AN INDISPENSABLE MANUAL OF TROPICAL MARINE BOTANY. — Biologists interested in the marine algal flora of the tropi- eal and subtropical coasts of the United States and the Caribbean area have long awaited the appearance of a com- prehensive taxonomic manual to facilitate the determination of their collections. Taking into consideration the spectacu- lar and highly diversified nature of the marine algal flora of this region, one is struck by the relative paucity of system- atic treatments relating to it; descriptions of tropical Amer- ican species have been widely scattered through general taxonomic works, such as Agardh’s Species Algarum (1820- 28), and the first important attempt to deal with them critically on a regional basis was that of W. H. Harvey in his Nereis Boreali-Americana (1852-58). For practical purposes of recent years, those wishing to identify marine algae of Florida and the Caribbean area have had recourse mainly to two manuals, namely Borgesen's Marine Algae of the Danish West Indies (1913-20) and W. R. Taylor's Ma- rine Algae of Florida, with special reference to the Dry Tor- tugas (1928). Both of these, although critical and valuable treatments, cover the marine algae flora of restricted areas only and make no claims to exhaustive coverage. The appearance of a marine algal flora for the whole of the Eastern American tropical and subtropical seaboard,’ from Bermuda and North Carolina to Southern Brazil, there- fore satisfies an acute and very long-felt need. Dr. Taylor’s book of 870 pages, with numerous plates of illustrations, contains descriptions of, and keys to, all the WILLIAM RANDOLPH TAYLOR: Marine Algae of the Eastern Tropical and Sub- tropical Coasts of the Americas. UNIVERSITY OF MICHIGAN PRESS, Ann Arbor, Mich., 1960. 8°, vii-ix +, 870 pp., 14 text-figs., 80 plates. $19.50. 58 Rhodora [Vol. 63 genera and species of marine algae known from this area, the total number being 272 genera, 760 species, and 140 infraspecific taxa. The treatment follows the same lines as the same author’s well known Marine Algae of the North- eastern Coast of North America (1937, 1957), in which, from Virginia to the Eastern Arctic, 401 species were re- corded; the approximate doubling of this number in the present manual emphasizes in a very striking way the rich- ness and diversity of the tropical and subtropical element along the Eastern American coasts. Like the earlier work mentioned above, Dr. Taylor’s new manual makes no claim to present a critically monographic treatment of all groups, which is of course a task yet to be accomplished, piecemeal, by generations of future workers. Nevertheless, it embodies the results of well over thirty years of practical acquaintance with the flora in shore and herbarium studies, and the spe- cies descriptions for the most part incorporate original and independent observations on the part of the author. Those species known only from the descriptions have been evalu- ated insofar as possible and, unless completely dubious, in- serted in the keys. Original references, except when unique, are not given for all species, but at the end of each descrip- tion there follows a bibliographic selection including all the more important geographic, and most of the morphological, data published on the species in the past. The distribution of each species is given in terms of the various islands and coastal segments comprising the area, together with infor- mation on the special ecology and mode of occurrence. The descriptions of genera and species, with the accom- panying keys, form the greater part of the book, the “Des- criptive Catalogue”, from page 44 onwards. The preceding pages contain a general introduction divided into the follow- ing sections: “Historical survey”, “Geographical distribu- tion”, “Algal habitats”, and “Collection and preservation”. For the non-specialist, the section on “Algal habitats” con- tains much of considerable general interest, illustrated by 14 full page reproductions of photographs taken by the author in Bermuda and Jamaica of various littoral and sub- littoral associations. 1961] Manual of Tropical Marine Botany 59 Latin descriptions of the new taxa (1 new family, Wur- demanniaceae in the Florideae, 8 new species and 4 new varieties) are assembled together as an appendix at the end of the descriptive section. The extensive bibliography contains references to all works in which the geographical distribution and essential morphology of the marine algae of the region has been dealt with. As with Dr. Taylor's treatment of the Northeastern Coast flora, considerable care has been taken to present a series of illustrations showing the general appearance and habitus of many of the commoner species, as well as of the details of their microscopic organization. In the present book the amount of illustration may be termed lavish, to the extent of 74 plates of line drawings and 6 of photographs. The drawings, apart from those executed by the author himself, were prepared by a team of several artists, and a certain disparity in treatment and technique is obvious, but all are extremely good and naturalistic representations, with the exception of a very few in which the artist, like his predeces- sors in ancient Egypt, seems to have had some difficulty in producing on a flat surface the illusion of a third dimension. The index is arranged by names of species and higher taxa; varleties are indented under the species to which they belong. One small error in alphabetical placing was noted on page 866, the duplication of entries for TITANOPHORA (J. Ag.) Feldm. For those working on aspects of tropical and subtropical marine biology in which an understanding of the marine plant life is essential, this book is indispensable for the routine identification of collections; while for the student with less specialized approach but with a lively interest in the remarkable world of offshore plant life in tropical and subtropical waters, its fascinatingly written introduction will stimulate, instruct, and probably lure not a few, whose interest has previously been held in check by lack of readily accessible information, to delve into the taxonomic profundi- ties of the descriptive section. 60 Rhodora [Vol. 63 Dr. Taylor is to be congratulated on the production in such eminently adequate format of this fundamental and extensively documented marine flora. — I. MACKENZIE LAMB. Volume 63, No. 745, including pages 1-30, was issued January 26, 1961. p^ D | REFERENCE LIBRARY Rhe Dodota JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Associate Editors Vol. 63 March, 1961 No. 747 CONTENTS: An Ecological Interpretation of Rhododendron Colonies in Maine and New Hampshire. A. R. Hodgdon and Radcliffe Pike ...........22 a 61 Taxonomic Fern Notes. I. Rolla Tryon ............. 70 A Variegated Foliage Form of Commelina. inani GV OKA UAAR ©. con tr eskacserisassesessous suze lene tr eerte de e thas 88 The Nem England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume. Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able. All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass. Printed by THE LEXINGTON PRESS, INC. Lexington, Mass. Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 March, 1961 No. 747 AN ECOLOGICAL INTERPRETATION OF RHODODENDRON COLONIES IN MAINE AND NEW HAMPSHIRE A. R. HODGDON AND RADCLIFFE PIKE Over a large part of its natural range Rhododendron max- imum is a local or rare species occupying certain unique habitats. Thus its disjunct distribution in the peripheral part of its range can be explained in part by the discontinu- ity of possible habitats. But after making extended observa- tions of rhododendrons in central and northern New Eng- land we have seen that there are many more apparently suitable places for the species to grow than there are colo- nies. One explanation of this interesting fact might be that wind dispersal results in the establishment of new colonies only here and there as the minute seeds are fortuitously blown into new areas. Yet wind dispersal for great dis- tances does not seem to warrant much consideration due first to the very protected and nearly windless sites in which the Maine and New Hampshire colonies are found and our failure to find seedling reproduction beyond one to two hun- dred feet away from mature plants. We might more reason- ably account for the more disjunct colonies on the basis of their being relics of an earlier more continuous distribution of the species, the assumption here being that various distur- bances have occurred to destroy them in many of the swampy areas that now seem to be entirely suitable for them. The recent fluctuations in size of a number of rhododendron col- Published with the approval of the Director of the New Hampshire Agricultural Experiment Station as Scientific Contribution No. 263. 61 62 Rhodora [Vol. 63 onies in Maine and New Hampshire as reported by us re- cently! give some support to the contention that a great decline in both number and size of colonies could indeed have resulted from the climatic changes that are known to have occurred many times in the post glacial period. A re- cent review? points up the nature of some of these climatic oscillations. The data presented in this paper make it seem likely that the limits of tolerance of rhododendron for certain climatic factors have been exceeded locally during some of the more severe climatic minima. At two periods in post glacial times climatic conditions were apparently such as to make possible the northward migration of warmer floras. These moderate periods were followed by decidedly colder more rigorous climatic conditions. For rhododendron to have persisted as a relic, one must postulate not only a once wider and more general distribution which could conceivably have occurred during these two warmer periods but also a continuity of suitable habitats from the time when the spe- cies was more or less continuously distributed. There have been several papers on Rhododendron maxi- mum emphasizing its distributional peculiarities and something of its ecology. Recently Iltis* has discussed in- terestingly and in some detail an outlying colony in the Coastal Plain of Virginia below Fredericksburg. He found that the majority of common species of associated trees and shrubs there were also among the dominant species listed by Spencer* in his study of 36 New Jersey Colonies and by Griggs® for the Sugar Grove Region of Ohio. The Virginian 1Hodgdon, A. R. and Pike, R. “Recent Changes in Some Rhododendron Colonies in Maine and New Hampshire". Rhodora 62: 87-93, April 1960. 2Dorf, Erlig. Climatic Changes of the Past and Present. American Scientist 48: 341- 364, September 19690. 3Titis, Hugh H. Studies in Virginia Plants II. Rhododendron maximum in the Virginia Coastal Plain and its Distribution in North America. Castanea 21: 114-24, September 1956. 4Spencer, Ernest L. ‘‘Natural distribution of Rhododendron maximum in New Jer- sey". Bull. Torr. Bot. Club 59: 401-*4, 1932. 5Griggs, R. F. A Botanical survey of the Sugar Grove Region. Ohio State University Bull. 18(25) : 273-75, 328, 1914. 1961] TABLE I. Colonies of Rhododendron maximum in Maine and New Hampshire. Hodgdon and Pike — Rhododendron 63 (The first 7 stations listed are in Maine, the others are in New Hampshire). Location “Township” 1. Lexington 2. Standish (1) 3. Standish (2) 4. Standish (3) 5. Standish (4) 6. Acton 7. Sanford 8. Albany 9. Pittsfield- Barnstead 10. Grantham 1l. Manchester (1) 12. Manchester (2) 13. Manchester (3) 14. Hopkinton 15. Mason 16. Wilton 17. Fitzwilliam 18. Richmond 19. Harrisville Characteristics of site Swamp and south facing slope mostly swamp (seedlings) Swamp and south facing slope mostly swamp (seedlings) Swampy woods (seedlings) Well drained ridge adjacent to swamp Swampy woods Gently south facing and well drained slope North facing slope of about 20% in steepest part (seedlings) Steep north facing mossy and wet ledges (seedlings) Swampy wooded pond north-facing shore (seedlings) Swampy woods (seedlings) Slope at eastern edge of swamp and in swamp: reported to have covered for- of acres in swamp merly scores (seedlings) Slope facing northeast Swampy woods and bed of stream Moist but scarcely swampy woods Swamp and adjacent west facing slope mostly swamp (seedlings) Northerly (seedlings ?) slope and bank of stream Mostly swampy woods but also adjoining slopes and ridges (seedlings) Swampy woods said to have covered 7-8 acres formerly (seedlings) Swampy woods (seedlings) Size (estimated) ló acre 315 acres ló acre 1/12 acre 1/20 acre 10x 6 ft. 5 acres 1/3 acre (concen- trated); few plants scattered over acre. 1/3 aere 34 acre 2 acres on slope, acreage in swamp uncertain but plants decidedly scattered there. 14 acre 1% acre 1/16 acre - acres 1/3 acre 15 acres 1% acres % acre Colony resembled the New Jersey stands also in having a strongly acid soil, pronounced shade and abundant water. Iltis agreed with Spencer's conclusion for New Jersey that the distribution of R. maximum in Virginia is governed by topography and not by climate. It should further be noted that the Virginia colony, all of those in New Jersey reported by Spencer and those in Ohio mentioned by Griggs (see references) are on north-facing slopes or, in the case of a few in New Jersey, in swamps. 64 Rhodora [Vol. 63 During recent field work we have recorded for each of our 19 Maine and New Hampshire colonies, such facts as the associated woody species, the absence or presence of seedling rhododendrons, the nature of the habitat — whether swamp or slope and, if so, the direction of slope as well as some other features. We are now in a position therefore to evaluate the ecological requirements of R. maximum by com- paring our colonies with those discussed by Iltis, Spencer and Griggs. Since we had no reason to doubt that the soils in all colonies were definitely acid and moreover because of the mass of scientific evidence that shows the oxylophytic character of the species, we decided at the outset not to include pH determinations as part of the record. We did attempt to determine the area of each stand, however, to provide a basis for evaluating the changes that may occur in the future. TABLE I. Table I shows 11 of our colonies to be chiefly in swamps, one other (Manchester (1)) divided between a swamp and an adjoining east-facing slope, 4 colonies on north-facing slopes, 2 very small colonies in non-swampy woodland and 1 small stand on a south-facing slope. But it should be re- marked that of the 11 swamp-colonies, 2 have excellent rhododendrons on their adjacent south-facing slopes and another has prolific plants on the west-facing adjacent slope. In contrast to the situation in New Jersey, the Coastal Plain of Virginia and in Ohio, swamps generally provide better habitats for rhododendrons in Maine and New Hampshire than do slopes. Also with us the direction of slope does not seem to be critical. We have pointed out elsewhere" that juxtaposition of swamps and adjacent slopes as alternative habitats for rhododendrons in many of the finest Maine and New Hampshire colonies may provide the species with a means of surviving environmental changes. Seedlings were noted in all of the larger colonies in Maine and New Hampshire and in some of the smaller stands as *Hodgdon, A. R. and Pike, R. “Recent Changes in Some Rhododendron Colonies in Maine and New Hampshire". Rhodora 62: 87-93, April 1960. 1961] Hodgdon and Pike — Rhododendron 65 well. Thirteen colonies (more than 68%) had seedlings in contrast to the situation in New Jersey where Spencer re- ported seedlings in only 5 of his 36 stations (less than 14%). Since Spencer's observations were made about 30 years ago we are led to speculate on the present condition of the New Jersey stands as a result of the tendency for temperatures in Northeastern United States to rise appreciably between TABLE II. List of tree-species associated with R. shire Stations: X indicates presence* maximum in Maine and New Hamp- "a $ n ~ m~ “~ a o are A u hh T od RE eva Besse RR z = 65 See UO TSO eS eee SesGusesee sz HnnnnecACHSOSSS DESPRE Acer rubrum AGONISTS AN Ke KEIO A AANA OX Tsuga canadensis Sex cp ah SRM Ip iho EN ey eo og b dib dro Betula lutea noe MAD do ebd doo» dedi Betula papyrifera X XX XXX X X XXX x Fagus grandifolia XN OX X X X X X XX x Pinus strobus xx x AKA ARANA Quercus rubra x x x X x x x x x X Picea rubens XXX x x x ENX Nyssa sylvatica CX X x x X X x Betula lenta x X X X x Fraxinus nigra XX x nc x Fraxinus americana x x x x Abies balsamea x x x x Quercus alba x x x Acer saccharum x? x Castanea dentata x x Betula populifolia x x Populus grandidentata X x Pinus resinosa x Thuja occidentalis x x Chamaecyparis thyoides x Quercus velutina X Q. coccinea 2c Q. prinus » n Carpinus caroliniana x Ostrya virginiana x Ulmus americana x Populus tremuloides ad Prunus pennsylvanica X Prunus serotina Tilia glabra 3c * The nomenclature in this paper follows that of Gray's Manual of Botany, 8th edition, 1950. 66 Rhodora [Vol. 63 the years 1930 and 1955.* Such temperature rise might easily further reduce areas suitable for seedling reproduction as our observations indicate that R. maximum seedling repro- duction takes place only on mossy generally moist sites. Tables IV and V, which compare the associated species of woody plants growing along with Rhododendron maximum in the 4 areas under consideration, point up certain similari- ties, notably the almost universal presence of Acer rubrum and the rather high frequency of T'suga canadensis. Kalmia latifolia, Cornus florida and Quercus alba are the three other species that are present in more than half of the New Jersey TABLE III. List of shrub-species associated with R. maximum in Maine and New Hampshire Stations. Pittsfield-Barnstead Grantham Manchester (3) Manchester (2) Hopkinton Lexington Standish (1) Standish (2) Standish (3) Standish (4) Acton Sanford Albany Manchester (1) Mason Wilton Fitzwilliam Richmond Harrisville Hamamelis virginiana Viburnum cassinoides Viburnum alnifolium x Nemopanthus mucronata x Acer pensylvanicum x x Vaccinium corymbosum x x Ilex verticillata x x Kalmia latifolia Acer spicatum : x x x Kalmia angustifolia x x x Epigaea repens var. glabrifolia x x x Linnaea borealis var. americana x Alnus rugosa x x Lyonia ligustrina x x Vaccinium angustifolium x x Salix Bebbiana x Lindera Benzoin x Sassafras albidum x Pyrus floribunda x Amelanchier laevis x Rhus radicans x Vaccinium myrtilloides x Lonicera canadensis x Viburnum recognitum x Sambucus pubens x x »* PET x "ET x A »* x A z PET nw ^ nun p A A »5 A AAA Pd bd "Braun, E. L. Deciduous Forests of Eastern North America, Blakiston, 1950. 1961] Hodgdon and Pike — Rhododendron 67 stands and might therefore be regarded as being character- istic associates of rhododendron in that area. However C. florida is absent from all Maine and New Hampshire colo- nies, occupying instead only certain well drained and warm exposures often with Quercus velutina — very different sites indeed from those of rhododendron. Mountain Laurel is sim- ilarly a disjunct species over most of its Maine and New Hampshire range; again it is adapted there to better drained TABLE IV. Presence of Common New Jersey woody species in four Rhododendron areas: the figure given is the percent of stands (to the nearest full number) in which a species was observed; X — present New Jersey Maine and Ohio Virginia Species Spencer 1932 New Hampshire Griggs 1914 Iltis 1956 Acer rubrum 94 100 X X Tsuga canadensis 72 95 X — Kalmia latifolia 69 21 »« bd Cornus florida 61 — xX xX Quercus alba 58 16 XC pi Quercus velutina 42 5 D< — Q. Prinus 42 5 X — Betula lenta 39 26 D< — B. lutea 39 79 — — Fagus grandifolia 28 58 x x Liriodendron tulipifera 25 — X x Chamaecyparis thyoides 5.5 5 — = habitats though occupying the same general range. Quercus alba on the other hand is a common forest tree in much of southern New Hampshire and southwestern Maine; yet it is met with in only 3 of the nineteen colonies ; it too grows more commonly in drier places. Of the remaining so-called “Common Species” Betula lutea is associated with rhododendron in 15 of our 19 stands though it is present in only 14 of the 36 New Jersey colonies and it is not mentioned by Griggs in Ohio nor is it present in the Virginian Colony studied by Iltis. Fagus grandifolia seems to do a bit better being present in all 4 areas but oc- cupying only 10 stands out of 36 in New Jersey and 11 of our 19. Of the other prevalent species in the Ohio, New Jersey and Virginia stands, Liriodendron tulipifera does not 68 Rhodora [Vol. 63 extend north of southern New England while Quercus Pri- nus barely gets into southern New Hampshire and Maine. Turning to the commonest Maine and New Hampshire associates, aside from those already mentioned, we find sev- eral species that would be found very rarely if at all in the TABLE V. Presence of Common New Hampshire Woody species in four Rhododendron areas Maine and Species New Hampshire New Jersey Ohio Virginia Acer rubrum 100 94 x x Tsuga canadensis 95 72 x — Betula lutea 79 39 — — Betula papyrifera 63 — -— -— Fagus grandifolia 58 28 x Pinus strobus 63 — — — Quercus rubra 53 — — X Hamamelis virginiana 47 sometimes present x x Picea rubens 47 — -— — Viburnum cassinoides 47 -— —- -— Nyssa sylvatica 42 — x -— Viburnum alnifolium 32 — — = Betula lenta 26 39 x — Fraxinus nigra 32 — -— — Nemopanthus mucronata 32 -— — — Acer pensylvanicum 26 — — — | Vaccinium corymbosum 26 — other 3 areas. Such include Betula papyrifera, Pinus stro- bus, Viburnum alnifolium, Picea rubens and Viburnum cas- sinoides. In Braun’s “Deciduous Forests of Eastern America" fre- quent mention is made of the types of forest communities in which Rhododendron maximum occurs, Tsuga canadensis being usually a dominant tree where rhododendron is found and Acer rubrum occurring commonly. The presence of other species of trees depends on the peculiarities of the habitat and the geographical area in which the rhododen- drons are found. These also are the only two species that appear repeatedly in a majority of colonies under considera- tion in this paper. Of the 15 other commonest associated species of trees in Maine and New Hampshire 11 are absent 1961] Hodgdon and Pike — Rhododendron 69 in New Jersey, 11 in Ohio and 13 in the Virginian Coastal Plain, the remaining species having no high order of coin- cidence in these places either. For local parts of the range of Rhododendron maximum there seems to be some usefulness in recognizing a charac- teristic rhododendron association, because the same group of associated species of plants is met with over and over again. But over the whole range the associates may change markedly; the nearly ubiquitous Tsuga may give way to Picea rubens at one extreme of climatic tolerance of rhodo- dendron or to Chamaecyparis thyoides at another. Good* has stated that each species has its particular range of tolerance. As a corollary it may be assumed that no two species, unless they are mutually dependent, have precisely the same range of tolerance of environmental conditions. A consideration of the associated woody species growing with Rhododendron maximum in various parts of its range lends support to this contention. Obviously none of the associated species has precisely the same tolerance of environmental factors as R. maximum. Those that are most frequently as- sociated with it over its entire range are those that pre- sumably most closely approach it in tolerance. But many of the common associated species in any one part of the range have very different amplitudes of environmental tol- erance from that of rhododendron and therefore will be absent from climatically different parts of the range. It seems preferable to interpret the facts of the Rhododendron maximum “association” in this way rather than to try to contrive a definite rhododendron association to embrace any considerable portion of the area that it occupies. On the other hand, it is to be expected that whenever environmental conditions are somewhat similar there will be essentially the same associates unless indeed these associates have had a different history and followed different migrational paths. It can be seen then that Tsuga canadensis and Acer rubrum have ranges of tolerance somewhat similar to that of R. maximum. But Betula lutea and other northern associated 5Good, R. A Theory of Plant Geography. New Phytologist 30: 155, 1931. 70 Rhodora [Vol. 63 species have ranges of tolerance overlapping that of R. max- imum only in the northern States. We must conclude from this comparison of Rhododendron maximum colonies in 4 outlying parts of its range that 1. There is no single physiographic situation to which the spe- cies is confined so long as an abundance of moisture is avail- able. 2. There is no such thing as a predictable association of species with which Rhododendron maximum is constantly to be found. 3. Within any particular climatic zone the rho- dodendron association is usually composed of essentially the same dominant species and often occupies similar physio- graphic situations. 4. The suggestion is made that Good’s concept of tolerance suggests a reasonable explanation of the observed diverse character of the rhododendron association. — DEPARTMENT OF BOTANY AND DEPARTMENT OF HORTICUL- TURE, UNIVERSITY OF NEW HAMPSHIRE, DURHAM, NEW HAMP- SHIRE. TAXONOMIC FERN NOTES. I ROLLA TRYON 1. Adiantum humile Kze. The name Adiantum humile Kze., based on a Poeppig collection from Peru, has seldom appeared in the literature since it was first published and to my knowledge has never been treated in a definitive manner. Mettenius identified Lechler 2319 and 2319a (B!) from Peru as A. humile but he did not publish these identifications in Filices Lechleri- anae; the specimens are Adiantum terminatum or a variant of it. An authentic specimen of Adiantum humile is at Vienna and a photograph of this specimen and fragments from it were obtained for the British Museum (Natural History) by the late A. H. G. Alston. This specimen has a valid claim to represent the name since the holotype was presumably destroyed with the Herbarium at Leipzig and since it is perhaps the only authentic material now extant (I saw no type material at B, BM, K, L, LE, P, S-PA, or U). I studied this authentic material and it unquestionably represents the spe- 1961] Tryon — Taxonomic Fern Notes 71 cies described by Maxon and Weatherby as Adiantum Kil- lipii. Such specimens as Guppy 6192, Tutin 366 and Lepri- eur 145 (all BM) are substantially identical to the Poeppig collection. The description of this species by Maxon and Weatherby (as A. Killipii) is an excellent one; it may be amplified by some comments about the variation of the indument. Nar- row scales (these two or three cells broad at their base), as well as trichomes, often occur on the under surface of the segments. The scales or trichomes occur especially toward the base of the ultimate segments and as Kramer' has noted, they are more persistent on the fertile segments than on the sterile ones, which indeed, may be glabrate. In some speci- mens only trichomes are present on the under surface of the segments, in others there are trichomes and also a few scales and in yet others the scales are more abundant than the trichomes. Adiantum humile occurs from British Honduras to Pana- ma, east to French Guiana and Trinidad, south to the state of Amazonas, Brazil and to Peru. The only recent collection that I have seen from Peru is: Quimiri Bridge, La Merced, Dept. Junin, Killip & Smith 24003 (NY, US). Adiantum humile Kze. Linnaea 9: 80. 1834. Holotype: “In sylvis fl. Huallagae superior. Peruv. ad Mission Tocache, Jun. 1830", Poeppig, Herb. Kunze, LZ, destroyed. Authentic specimen: “Maynas, alto fl. silvis primaeva, Toache [Toca- che], Maj. 1830", Poeppig, W, photograph and fragments BM! Adiantum Killipii Maxon & Weath. Amer. Journ. Bot. 19: 166. 1932. Holotype: Ancón Hill, Canal Zone, Panama, Killip 2752, US! (15 paratypes are also cited). 2. Adiantum lobatum Pres! (Plate 1255, fig. 1). Two recent collections by D. S. Correll and E. E. Smith from the Department of Lambayeque, Peru, apparently rep- resent this species which has been little, if at all, understood since it was published. These collections are rather similar to A. brasiliense and A. curvatum in the leaf-architecture 1Acta Bot. Neerland. 3: 482. 1954. 72 Rhodora [Vol. 63 and in the minute puberulence only on the upper surface of the rachis and lesser axes. However, in those two species the sori are oblong or oblong-lunate and the segments are strongly dimidiate. In the Peru material the sori are orbic- ular to suborbicular and although some segments are di- midiate (but not strongly so), others are subdimidiate or flabellate. A photograph of the holotype of A. lobatum, a single rather small leaf, agrees closely with the Correll & Smith collections in characters of the ultimate segments, and also, allowing for the difference in the size of the leaves, with its characters of leaf-architecture. The lack of a rhizome in the holotype introduces an element of uncertainty in its identity. Also Presl described the leaves as glabrous while in the Peru material they are minutely puberulent on the upper side of the rachis and other axes of the lamina. Presl, however, might well have overlooked this small character. While it is by no means certain, then, that the present speci- mens represent Presl’s species, they do not seem to belong to any other, and I think it is better to use Presl’s name for them, albeit tentatively, rather than to describe them as new. Adiantum lobatum Presl, Rel. Haenk. 1: 62, t. 10, f. 4. 1825. Holotype: Guayaquil, 1790, Haenke, PR, photograph GH! (Presl’s illustration is a very accurate copy of the speci- men; Pres] gave the locality as Mexico but the label accom- panying the photograph gives Guayaquil). Adiantum loba- tum “(Poir.)”’ Steud. Nomencl. Bot. 2: 275. 1824, in synony- my of Lindsaea lobata Poir. is invalid. Dept. Lambayeque, Peru: 27 km. from Olmos on road to Jaen, 1250 m., March 23, 1960, D. S. Correll & E. E. Smith P801 (GH,LL,US) ; 31 km. from Olmos on road to Jaen, 1400 m., March 23, 1960, D. S. Correll & E. E. Smith P808 (GH, LL). The following description will serve to orient this species, as I interpret it, within the genus. Rhizome rather slender, creeping, the petioles spaced but not distant, scales ca. 1-2 mm. long, long-triangular, acuminate to lanceolate- acuminate, wholly sclerotic, brown to atropurpureous; leaves to 60 cm. long, lamina deltoid to long-triangular, bipinnate to tripinnate at 1961] Tryon — Taxonomic Fern Notes 73 2. Fig. Adiantum lobatum, X 1/3, Correll & Smith P801 (GH). 55. Fig. 1. 00. Plate 12 Cheilanthes Orbignyana, X 1/4, Sagástegui 2937 (GH). 74 Rhodora [Vol. 63 the base, bipinnate below the large conform terminal pinna, rachis subflexuous, it and the other axes faintly glaucous, minutely puberu- lent on the upper, darker colored surface, glabrous and lighter colored beneath; ultimate segments toward the apex of the pinna oblong to trapeziform, subsessile to short-stalked, those toward the base of the pinna subflabellate or flabellate, with longer stalks, the terminal ones flabellate-cuneate, all entire to usually sparingly and moderately in- cised (sometimes strongly so), non-articulate, the dark color of the stalk entering the base of the segment; sterile vein tips end between the indistinct marginal crenulations; sori orbicular to suborbicular, borne on all margins of the fertile segment except the lower one. 3. Cheilanthes fractifera Tryon, Rhodora 62: 7. January, 1960 Cheilanthes Saundersii Alston, Lilloa 30: 110, t. 6. August, 1960. Holotype: Dept. Lima, prov. Huarochiri, dist. Surco, S. G. E. Saunders 350, BM; paratype: (from the same dis- trict) Saunders 219, BM! GH'! This recently recognized species has been independently described by myself and the late A. H. G. Alston. Although I did not see Saunders 350 at the British Museum, Alston's photograph (t. 6), his description and the citation of Saun- ders 219 (also a paratype of C. fractifera), all confirm the identity of his species with mine. Dr. Alston's paper was certainly prepared before I had seen any material of this species and it is unfortunate that its publication was delayed. In addition to the collections cited by myself (Correll & Smith P169, Saunders 219) and by Alston, two specimens collected by Rauh & Hirsch in Peru were seen at Berlin: Rimae Valley, Dept. Lima, P143 and Cerros de Caldera des- ert, Dept. Arequipa, P570. 4. Cheilanthes Orbignyana Mett. ex Kuhn (Plate 1255, fig. 2) Cheilanthes Orbignyana is evidently a valid species al- though a very rare one; there is no previous material of it at the Gray Herbarium or the United States National Her- barium. An isotype at Paris (La Laguna, Bolivia, D'Orbigny 388) is represented in the Gray Herbarium by descriptive notes by Mr. C. A. Weatherby and a photograph taken by Mrs. Weatherby. 1961] Tryon — Taxonomic Fern Notes 75 The following specimen, the first from Peru, compares very closely with these materials and I believe that its iden- tification with them is reasonably certain: La Pampa, Guz- mango, prov. Contumaza, Dept. Cajamarca, Peru, 2730 m., May 30, 1959, Sagástegui 2937 (GH,US). This may be dis- tinguished from the other Peruvian species that lack hairs or scales on the lamina by the following key. Pinnules, or most of them, sessile; rhizome scales with a dark sclerotic central portion and pale, thinner margins. C. Poeppigiana. Pinnules petiolulate or, at least most of them, on short broad stalks; rhizome scales entirely, but not always heavily, sclerotic. Indusia extending along the segment stalks and pinna-rachises. C. marginata. Indusia confined to the segments. Rhizome short, erect, the petioles clustered; petiole, especially in the apical half, flat to convex between small lateral ridges. C. Orbignyana. Rhizome short, creeping, the petioles rather spaced; petiole, es- pecially in the apical half, sulcate between the prominent lateral ridges. C. rufopunctata. 5. Costaricia Christ (Plate 1256, figs. 3, 4) The genus Costaricia, described by Christ in 1909 on the basis of a sterile fern collected by Wercklé in Costa Rica, has never been understood and Christensen listed it in Index Filicum Suppl. 1 and again in Suppl. 3 as a genus valde du- bium. The type material at Paris consists of Wercklé no. 238 mounted on two sheets; a photograph of one of them is given in fig. 3. This material is wholly sterile but while studying it some of its characters reminded me of two fertile collec- tions from Costa Rica I had examined while working on Dennstaedtia. A later comparison of these with my notes and photographs of the Wercklé collection indicated that they were portions of fertile leaves of Costaricia Werckleana and that the original material represented a juvenile plant. On the basis of the fertile material (Fig. 4) it is now possible to draw a more satisfactory description of the genus (or species). Rhizome (of juvenile plant) long-creeping, slender, pub- escent with more or less moniliform trichomes, their short cells with clear whitish side and brown end walls, the leaves 76 Rhodora [Vol. 63 borne singly at intervals; leaves probably to 1 m. or more tall, to tripinnate, acroscopic, herbaceous, bearing trichomes similar to those of the juvenile rhizome, especially on the upper surface in the grooves of the major axes and on each side of the costa of the penultimate segments ; veins free, the sterile tips ending well back of the margin and not enlarged ; pinnae and pinnules subarticulate; sorus terminal on a lobe, the receptacle, a continuation of the single vein, elongate within the rather cuneate, slightly bilabiate indusium which is formed of a more modified inner true indusium and a less modified outer opposed lobe of leaf tissue, these being almost wholly joined to the tissue of the lobe which extends well up on each side; sporangia developing in a basipetal sequence, paraphyses absent (although persistent stalks of previously maturing sporangia may simulate them), annulus vertical, or more or less displaced by crowding in the sorus, of 10-12 indurated cells, these extending from near the apex of the capsule down to the apex of the stalk. The characters strongly suggest that Costaricia Werckle- ana is a species of Dennstaedtia, although if this is true, then it is not at all closely related to any other of the American ones. It was for this reason (as well as the absence of in- formation about the rhizome) that I did not account for the fertile material in my paper on American Dennstaedtia. The most distinctive characters of C. Werckleana in comparison with American Dennstaedtias are the subarticulate pinnae and pinnules and the leaf tissue that extends well up on each side of the sorus. These characters, however out of place among the New World species, are known in some of those of southeast Asia. For example, D. ampla (Bak.) Bedd., D. glabrata (Ces.) C. Chr. and D. Elmeri Copel. have articu- late segments and D. scabra (Wall. ex Hook.) Moore some- times has a substantially identical sorus. It is possible then, or perhaps probable, that C. Werckleana is a species of Dennstaedtia. However, the rhizome of the adult plant, and especially its indument, must be known before it can be placed in Denn- staedtia with certainty. The adult rhizome may have a dif- 1961] Tryon — Taxonomic Fern Notes i. 9n p P T LOS nM nS US oo en 77 238 (P). Wercklé Fig. 3. Holotype, ca. X 1/3, (NY). Costaricia Werckleana. Plate 1256. Brade 336 2 Os Two pinnae, X 1/: 4. Fig. 78 Rhodora [Vol. 63 ferent type of indument than the juvenile one as in Spheno- meris in which the adult rhizome may bear scales while the juvenile one may bear only trichomes. Until more adequate material of Costaricia Werckleana is available, the possibili- ty that it represents an endemic genus can not be eliminated. For the present, then, I prefer to retain Costaricia as a dubious genus rather than to reduce it to Dennstaedtia and make a new combination that may be taxonomically incor- rect. Costaricia Werckleana Christ, Bull. Soc. Bot. Genève II, 1: 229, fig. on 230. 1909. Holotype: Costa Rica, 1905, Wer- cklé 238, P! Other specimens examined: (all from Costa Rica) La Palma, 1400 m., 1909, Brade 336 (NY, US ex Ro- senst., US ex C) ; La Hondura, 1200 m. 1933, Valerio 1815 (US). 6. Synonyms in Dennstaedtia The examination of certain type or authentic specimens has enabled me to place some of the dubious names listed* in my recent paper on American Dennstaedtia. Dennstaedtia divaricata (Sod.) C. Chr. = D. arborescens (Willd.) Ekman ex Maxon. Authentic specimen: Niebly, Ecuador, 1883, Sodiro, P! Dennstaedtia erosa (Kze.) Moore — D. obtusifolia (Willd.) Moore. A probable isotype: “Pampayaco, Peru, July, 1829, Poeppig 169" (det. Kunze), B! is taken to repre- sent this name since the holotype was presumably destroyed with the Herbarium at Leipzig. The data on the Berlin specimen are the same as those given by Kunze except that he cited “Diar. 1127" rather than “no. 169”. A fragment (one pinnule) at K! (ex Kunze, det. Kunze) is also D. obtusifolia as are probable isotypes at LE! and P! Dennstaedtia grandifrons Christ = D. distenta (Kze.) Moore. The type was not seen at Paris but an authentic specimen is D. distenta: Miinch 154 (Herb. Christ, det. Christ), P! Dennstaedtia Munchii Christ — Hypolepis sp. The holo- type, Münch 137, was not seen at Paris; a specimen at US ?Contrib. Gray Herb. 187: 52. 1960. Tryon — Taxonomic Fern Notes 79 1961] “(HD) 996 inDojspnODS P zodo'] ‘adAyoloy '9 ‘BI ' (Hd) ZEGT ut wuno4g dsd ‘a o IX 'njsn6Oun 'e/ TI X 'sisuoDupnguno DUIDIOYION DUIDIOYION °C "ng AmB EDS OL 9 SOUS HITE Ie mV pt KUO PONE Mq) "S E Yee RAVER LET PMI LY QANONNSISG “LYST Pld "aou toads ‘Gods Weetsue wuatl MODEYSSAR AVEO 80 Rhodora [Vol. 63 (ex Stanford) with this name but with number 76 is prob- ably an isotype incorrectly numbered. Dennstaedtia Orbignyana Mett. ex Kuhn — D. obtusifolia (Willd.) Moore. Holotype: D'Orbigny 278, (Herb. Mett.) B!;isotype, P! Dennstaedtia vagans (Bak.) Diels = D. arborescens (Willd.) Ekman ex Maxon. The holotype, Andes of Quito, Ecuador, Sodiro, K! is evidently a juvenile leaf-form. 7. Notholaena angusta, spec. nov. (Plate 1257, fig. 5) Rhizoma breve paleis rigidis obscuris scleroticis nitidis pectinato- ciliatis. Folia 5-20 cm. alta, petiolus quam lamina brevior teres rub- elle brunneus vel nigrescens fasciculo vasculari uno ceraceo-glandu- losus glande plerumque longe cauliculatis trichomatibus paucis magnis brunneis modice paleaceus paleis eis rhizomatis similibus nisi brevi- oribus ad 1 mm. longis. Lamina linearis modice pinnato-pinnatifida vel pinnato-pinnatisecta, rhachis modice paleacea teres vel subteres tri- chomatibus multis circa 0.5 mm. longis eis petiolorum similibus. Pin- nae ad 30-jugae oblongae obtusae plus minusve aequilaterales seg- mentis obtusis 2-3-jugis coriaceae pagina superiore modice albo-ceracea pagina inferiore dense albo-ceracea costa trichomatibus paucis magnis brunneis. Venulae 1-2-furcatae sporangia 64-sporis in apicibus geren- tes, margem immutatus leviter revolutus, Specimens examined (all from Hidalgo, Mexico) : Holo- typus: ca. 1 km. south of Tasquillo Bridge, Jan. 23, 1941, T.C. & E. M. Frye 3145 (US). Paratypi: Near Barranca de Veneado, 1800 m., Sept. 23, 1951, E. Matuda & D. B. Gold (Matuda Herb. no. 23571) (US); Near Tasquillo, 8000 ft., July 17, 1940, C. L. Hitchcock & L. R. Stanford 7249 (US) ; Río Tula, Puente Tasquillo, arid rocky slopes, March 8, 1937, M. Broun (PH). This new species will key out with Notholaena Schaffneri in my revision of the American species of Notholaena’. It may be separated by the following key which modifies and amplifies my published one. 29. Lower surface of the lamina and the rachis with large, brown trichomes; rhizome scales dark sclerotic, rigid, strongly pectinate- ciliate. 29a. 29a. Lamina elliptic-lanceolate, bipinnate-pinnatifid; the larger scales of the petiole ca. 2-3 mm. long, the trichomes of the rachis ca. 1 mm. long. . . . 29. N, Schaffneri. ?Contrib. Gray Herb, 179: 11. 1956, 1961] Tryon — Taxonomic Fern Notes 81 29a. Lamina linear, pinnate-pinnatifid to pinnate-pinnatisect; the large scales of the petiole ca. 1 mm. long, the trichomes of the rachis ca. 0.5 mm. long. . . . 29A. N. angusta. Notholaena angusta is related to N. Schaffneri in charac- ters of the indument: the long-stalked ceraceous glands gen- erally distributed on the petiole and rachis, the scales on the rachis, and the large brown trichomes on the lower surface of the pinnae. It is related to N. Ekmanii, N. cubensis and N. affinis in its linear, pinnate-pinnatifid lamina. These three species differ from N. angusta in lacking scales on the rachis and trichomes on the pinnae. They also either lack long-stalked glands on the primary axis, or, if these are sometimes present (in N. Ekmanii and N. affinis), they are confined to the upper surface. N. Ekmanii also differs from N. angusta in having a sulcate rather than terete or flattened petiole and rachis. In my treatment of Notholaena, I placed the Broun collec- tion cited above in N. affinis, and later identified the Hitch- cock & Stanford collection as the same species. It is now evident that all of this material from Hidalgo is a distinct species and that N. affinis, which occurs principally in Guate- mala and Honduras, is known to occur in Mexico only in Oaxaca. 8. Notholaena cantangensis, spec. nov. (Plate 1257, fig. 6) Rhizoma breve paleis medio nigrescentibus scleroticis marginibus fulvis pectinato-serrulatis. Folia 10-15 cm. alta caespitosa, petiolus quam lamina brevior teres brunneus vel rubelle brunneus fasciculo vasculari uno decidue paleaceus paleis magnis fulvis concoloribus pectinato-serrulatis nisi eis rhizomatis similibus ad basem. Lamina lanceolato-elliptica bipinnato-pinnatifida, rhachis teres vel subteres, paleacea paleis eis petiolorum similibus. Pinnae ad 15-jugae oblongae vel leviter latiores ad basem obtusae aequilaterales pinnulis 5-8-jugis obtusis coriaceae pagina superiore parce pubescens trichomatibus brevibus subcrassis cellulis brevibus pagina inferiore paleis imbricatis anguste ovato-lanceolatis acuminatis vel ligulato-acuminatis. Venulae 2-3-furcatae sporangia 64-sporis in apicibus gerentes, margem leviter mutatus planus vel leviter revolutus, sporae rugosae. Holotypus: Entre piedras, Cantange, ruta Celendín-Río Marañon, prov. Celendín, Dept. Cajamarca, Peru, 1450 m., 4 Junio 1960, A. López & A. Sagástegui 3366 (GH). 82 Rhodora [Vol. 63 This new species, and the following one, belong to a small group (previously of two species) that have the lamina with both hairs and scales but lacking wax. N. cantangensis is closely related to N. Hassleri of Paraguay which also has similar rather sparse hairs only on the upper surface of the lamina. Following the description of the next species, a modification of the appropriate portion of my key to the American species is presented in which both new ones are incorporated. This will serve to compare the four species of this group and to bring out their most distinctive characters. 9. Notholaena solitaria, spec. nov. (Plate 1258, fig. 7) Rhizoma ignotum. Folia circa 18-32 cm. alta, petiolus quam lamina brevior vel longior teres obscurus rubelle brunneus vel nigrescens fasciculo vasculari uno pubescens paleaceusve paleis angustis fulvis concoloribus nisi medio nigrescentibus scleroticis marginibus fulvis ad basem. Lamina anguste lanceolata vel anguste elliptica modice bipinnato-pinnatifida vel bipinnato-pinnatisecta, rhachis teres decidue paleacea paleis eis pinnarum similibus. Pinnae circa 10-15-jugae ob- tusae aequilaterales pinnulis anguste oblongis 6-9-jugis subcoriaceae pagina superiore tomentosa trichomatibus longis subcrassis cellulis longis pagina inferiore dense tomentosa sub paleis imbricatis anguste auriculatis vel cordatis marginibus vel tantum base longe ciliatis. Venulae 1-2 furcatae sporangia 64-sporis in apicibus gerentes, margem modice mutatus revolutus, sporae rugosae. Holotypus: Surinam, Hostmann 199 (Herb. Hook.-K) Mr. C. A. Weatherby annotated this specimen in 1938 as probably a new species. It is closely related to N. brachypus (Mexico to Costa Rica) but is amply distinct in the charac- ters of the scales at the base of the petiole, the complexity of the lamina and the scales on the under surface of the lamina. N. solitaria is geographically isolated from other species of the genus. There are six species in Colombia and of these only two widely distributed ones, N. aurea and N. sinuata, extend eastward to (north-central) Venezuela. One would have expected that a Notholaena from Surinam would be one of these two species rather than an endemic with its affinities in Central America. This species and the previous one form with N. brachypus and N. Hassleri, a small group that lacks ceraceous indument but has both hairs and scales 1961] Tryon — Taxonomic Fern Notes 83 on the lamina. These are distinguished in the following key which modifies and expands the one in my Revision of the American species of Notholaena'. 3. Indument of the lamina of both hairs and scales; spores rugose. 8. 8. Under surface of the pinnae densely tomentose beneath the im- bricate scales, the upper surface tomentose with long hairs these with long cells. 8a. 8a. Rhizome scales and those of the petiole base concolorous or nearly so, or sometimes with a brownish sclerotic center; lamina pinnate-pinnatifid, hairs of the upper surface of the pinnae very slender, scales of the lower surface with smooth or irregularly dentate margins, the larger ones strongly auriculate; sporangi- um with 32 spores. Mexico to Costa Rica. . . . 6. N. brachypus. 8a. Rhizome scales not known, those of the petiole base with a blackish sclerotic central area and pale margins; lamina bipin- nate-pinnatifid to bipinnate-pinnatisect, hairs of the upper surface of the pinnae rather stout, scales of the lower surface with long-ciliate margins (or the cilia confined to the base), usually narrowly auriculate or cordate; sporangium with 64 spores. Surinam. . . . 6A. N. solitaria. 8. Under surface of the pinnae paleate with imbricate scales, the upper surface sparsely pubescent with short thickish hairs, these with short cells, sporangium with 64 spores. 8b. 8b. Rhizome creeping, the petioles approximate to moderately spaced; pinnae inequilateral although not strongly so; scales of the under surface of the pinnae (especially of the pinna- rachis) predominantly ovate-acuminate to deltoid-acuminate; margins of the segments somewhat revolute, of broadly rounded lobes which have a hyaline border. Paraguay. . . . T. N. Hass- leri. 8b. Rhizome short-creeping, the petioles clustered; pinnae equi- lateral; scales of the under surface of the pinnae predominantly narrowly ovate-lanceolate and acuminate to ligulate-acuminate; margins of the segments flat to slightly revolute, very slightly modified. Peru. . . . TA. N. cantangensis. 10. Notholaena Stuebeliana (Hieron.) Tryon, comb. & stat. nov. (Plate 1258, fig. 8) Pellaea dealbata var. Stuebeliana Hieron. Hedwigia 48: 225, t. 12, fig. 15, 1909. Holotype: Sunibamba, valle Río Ut- cubamba, [Dept. Amazonas], Peru, 1800 m., Stübel 1048 (part), B! Pellaea Stuebeliana Hieron. l.c., in synonymy. *Op. Cit. 9. 84 buen 2 m NP WEM Rhodora [Vol. 63 e. d^. y'on lt p h't'b'p hu pr hl. LA Hostmann 199 (Xx). holotype, Notholaena Stuebeliana, X 4/10, Correll & Smith P841 (GH). b] " Notholaena solitaria, ca. X 1/3, T. Fig. od. 9 Plate 1 Fig. 8. 1961] Tryon — Taxonomic Fern Notes 85 Hieronymus described Pellaea dealbata var. Stuebeliana on the basis of a single leaf presumably collected in Peru along with Notholaena nivea (Stübel 1048 proper). In my Revision of American Notholaena? the name was cited as a synonym of Notholaena dealbata on the assumption that the specimen and the label had been erroneously associated with each other. Although Hieronymus did emphasize, in his des- cription, the significant rounded ultimate segments, these were not brought out well in the illustration. Three recent collections from the same general region now verify the occurrence of this plant in Peru and the adequate material allows an assessment of it: Cajamarca: Celendín, May 22, 1960, López & Sagástegui 3104 (GH) ; about 40 km. from Cajamarca on road to Chilete, March 24, 1960, D. S. Correll & E. E. Smith P841 (GH,LL, US). La Libertad: Camino de las Quishuas, Bolivar, May 31, 1960, López & Sagástegui 3288 (GH). Notholaena Stuebeliana grows in rocky places in northern Peru at elevations of 2250 to 3000 m. It is most closely re- lated to N. limitanea of Mexico and the adjacent southwest- ern United States; it is also closely related to N. dealbata of the central United States. The following key will serve to bring out the important characters of these three species, and to distinguish them ; it amplifies the key to species in my Revision of American Notholaena’. 57. Petiole moderately stout (rarely slender), it and the rachis dark reddish-brown; ultimate segments coriaceous, the veins not visible; sporangium with 32 spores. 57a. 57a. Rhizome scales somewhat thickened, their cells evident, rather straight or occasionally somewhat sinuous, dry in appearance; terminal segments of the pinnae and pinnules oval to usually ob- long. Southwestern United States and Mexico. . . . 51. N. limi- tanea. 57a. Rhizome scales thickened, their cells obscure, mostly strongly sinuous, oily in appearance (but not evidently viscid) (they may appear dry with age); termina! segments of the pinnae and pin nules (or their terminal lobe) orbicular, suborbicular, orb'eular- deltoid or infrequently oval. Northern Peru. . . . 51A. N. Stue- beliana. šOp. Cit. 87. Op. Cit. 14. 86 Rhodora [Vol. 63 57. Petiole slender, it and the rachis bright chestnut-brown; ultimate segments subherbaceous, the veins usually visible; sporangium with 64 spores. Central to south-central United States. . . . 52. N. dealbata. 11. Pteris Lechleri Mett. The species currently called Pteris Killipii Maxon has two earlier names. Mettenius described Pteris Lechleri ( Lechler 2533) in 1859, a name that was seldom, if ever, used after its publication. In 1867 Baker described Pteris vestita (Spruce 4063) but the specimen was not annotated by him and later Maxon cited it as a paratype of P. Killipii. An examination of the holotypes of these three names shows that they all represent the same species and accordingly Mettenius' name must be reinstated for it. Pteris Lechleri grows on the moist eastern slopes of the Andes in Peru and Bolivia. Pteris Lechleri Mett. Fil. Lechler. 2: 13, 1859. Holotype: Tatanara, [Valle Río San Gaban, prov. Carabaya, Dept. Puno], Peru, Lechler 2533, Herb. Mett.-B!, photo GH! Pteris vestita Baker, Syn. Fil. 169. 1867. Holotype: (not annotated by Baker), near Tarapoto, Peru, Spruce 4063, K!, fragment ex K, US!, photo US. Pteris Killipii Maxon, Amer. Fern Journ. 23: 107. 1933. Holotype: San Ramón, Dept. Junín, Peru, Killip & Smith 24697, US! ; paratypes: Killip & Smith 23919, Us!, Killip & Smith 23962, US!, Spruce 4063, K! 12. Recent fern collections from Peru In the past few years several collectors have obtained material from Peru that is of unusual interest and it seems worthwhile to comment on some of the principal features of these collections. This will serve to emphasize the regions in which new discoveries of special importance are being made and at the same time will acknowledge the efforts of these collectors who have added to our knowledge. The ferns mentioned at this time, with one exception, are all members of the Gymnogrammeae and all are from the An- dean region of Peru. The collections are those of P. Coro- nado in 1953 and 1955; D. S. Correll and E. E. Smith in 1958 and 1960; P. Hutchison in 1957; W. Rauh and G. Hirsch 1961] Tryon — Taxonomic Fern Notes 87 in 1954, W. Rauh in 1956’; A. Sagástegui (sometimes with A. López) in 1953-1960; and S. G. E. Saunders in 1954. Five of the new or rare species from these collections have been discussed in the previous portions of this paper and these are omitted from the following list. Adiantum alarconianum Gaud. — The first collections in Peru were made by Coronado (218, 228) in Tumbes in 1955. Adiantum concinnum Willd. — The first collections in Peru were made by Coronado (222, 229) in Tumbes in 1955; it was collected again in Lambayeque, somewhat to the south, by Correll & Smith (P791, P794) in 1960. Adiantum Henslovianum Hook. f. — The second collection in Peru is that of Correll & Smith (P807) from Lambayeque in 1960; the original collection by Matthews was probably made in northern Peru about 1835. Adiantum Poiretii var. hirsutum (Hook.) Tryon — First collected in Peru on lomas in Arequipa by Coronado (33, 42) in 1958. Cheilanthes notholaenoides (Desv.) Weath. — Known from Huánuco and Junin by two earlier collections, it was collected by Rauh & Hirsch (P1935) in Piura in 1954 and by Correll & Smith (P838) in Cajamarca in 1960. Notholaena obducta (Mett. ex Kuhn) Baker — First col- lected in Peru by P. Hutchison (1425) in Cajamarca in 1957. The principal range of the species is from Paraguay to Bo- livia; it is also known from several Colombian collections. The Peru locality is about half-way between these two areas. Notholaena peruviana Desv. — This species has been known to occur in the Rimac Valley (Lima) and southward on coastal lomas. Three recent collections extend its range considerably northward: Sagástegui (2220) in Cajamarca in 1953, Sagástegui (2665) in La Libertad in 1958, and Cor- rell & Smith (P822) in Lambayeque in 1960. Saffordia induta Maxon — This monotypic genus has been the rarest of American fern genera being known only from the original collection made by Safford in 1892. However, it "These collections were reported on by D. E. Meyer, Willdenowia 1: 642-653. 1956; 1: 704-708. 1957; 2: 23-26. 1958. I studied some of them at Berlin in 1960. 88 Rhodora [Vol. 63 has now perhaps lost this distinction since four collections of it have been made in 1954-1959, and its range has been considerably extended beyond the type locality in the Rimac Valley. In 1954 it was collected by Rauh & Hirsch (P186) at Matucana (Lima) which is probably the type locality (Safford gave only “along the Arroya [La Oroya] Railway in the mountains back of Lima"). In the same year Saun- ders (218) collected it near Surco, also in the Rimac Valley, somewhat below Matucana. This collection was sent to the British Museum (Natural History) and spores from it were germinated at Kew where a number of mature plants are now growing. In 1958 Sagástegui (205) collected it in La Libertad (prov. Otuzco) and in 1959 (2936) in Cajamarca (prov. Contumazá) somewhat to the northwest. All of the collections are from the rather dry western slopes of the Andes at elevations between 1900 and 2900 m.; and they were all made in the months of February to May which cor- responds to the “rainy” season in that zone. Stylites andicola Amstutz emend. Rauh and S. gemmifera rauh — This new genus of /soétaceae, from the borders of high lakes in central Peru, was first collected by Rauh & Hirsch (P271b) in 1954. It was collected again by Rauh (P186/56, P384/56) in 1956, by Amstutz (2000) in 1956 and by von Appen in 1958. The careful studies of W. Rauh and H. Falk? and D. E. Meyer’ have presented an exception- ally detailed and complete account of this unusual plant. Trachypteris pinnata (Hook. f.) C. Chr. — This rare fern was collected by Rauh & Hirsch (P2160) in Cajamarca (prov. Jaen) in 1954 and in the same province by Hutchison (1424) in 1957. — GRAY HERBARIUM, HARVARD UNIVERSITY. A VARIEGATED FOLIAGE FORM OF COMMELINA. — During the summer of 1958, Mr. Joseph Monachino of our staff col- lected a specimen of Commelina communis L. var. ludens (Miquel) C. B. Clarke, which he found growing along the shore walk of the Palisades near the George Washington Bridge in New Jersey. "Sitz. Heidelberg. Akad. Wissen. Math-naturwissen. Klasse 1959: 1-160, "Willdenowia 2: 32-40. 1958. 1961] A Variegated Foliage Form of Commelina 89 The interesting feature of the plant was its variegated foliage, the leaves being a light lemon color with narrow lines of light green running the entire length of the leaf. The greater portion of some of the leaves was either entire- ly a light lemon color, or slightly diffused with very thin, broken lines of a very light green. They resemble greatly the leaves of Tradescantia fluminensis Vell., the Wandering Jew. Having been careful to keep the roots moist, Mr. Mona- chino presented the plant to me, to grow in my wildflower garden. There it thrived until killed by the frost. Early in the Spring of 1959, while cleaning my garden I discovered that the crown was alive and that it was showing a few curled-up leaves of light cream color. I was rather dumbfounded as I had always considered the Dayflower to be an annual in this area; especially as our temperature had dropped, occasionally, to five degrees below zero. Later on in the Spring, it was discovered that because of someone’s faulty cultivation the plant had been destroyed. However, it was noted that the area was covered with a fine stand of young seedlings possessing the variegated foliage of the parent plant. These seedlings were permitted to fend for themselves and, by the end of the Summer, had estab- lished a fine colony of beautiful and interesting plants. These variegated plants were taller and more robust than the usual type of the plant, two patches of which were permitted to grow on either side of the variegated form. However, the flowers seemed somewhat smaller and lighter blue in color. A few of the variegated plants have been pulled and are being prepared for mounting and for exchange with other institutions. The balance will be left to reseed themselves, for my friends and I shall look forward to a renewed ac- quaintance with this pretty plant next vear. Therefore, I should like this variegated form to bear the name of — Commelina communis L. var. ludens (Miquel) C. B. Clarke forma aureostriata MacKeever, f. nov. A varie- tate differt foliis luteo-striatis. 90 Rhodora [Vol. 63 This form is similar to the variety except for its leaves, which are striped with yellow. The type specimen was collected by Frank C. MacKeever in Mount Vernon, N. Y., on October 1, 1959, and is deposited in the New York Botanical Garden. Isotypes are deposited at Chicago Natural History Mu- seum, Chicago, Ill.; Gray Herbarium, Cambridge, Mass.; Missouri Botanical Garden, St. Louis, Mo. ; Smithsonian In- stitution, Washington, D. C.; L. H. Bailey Hortorium, Ithaca, N. Y. — FRANK C. MACKEEVER, NEW YORK BOTANICAL GAR- DEN. Volume 63, No. 746, including pages 31-60, was issued March 27, 1961. DUPLICATE BOOKS FOR SALE These books have library book plates and are used copies, some worn, some in need of binding. BUILLIARD, P. Dictionnaire élémentaire de botanique ou Exposition par ordre alphabétique, . . . Paris, 1783. vii, 242-- [8] p. 10 plates, 9 colored. folio ......................... $25.00 KERNER VON MARILAUN, A. The natural history of plants... From the German by F. W. Oliver . . . New York, 1895. Two volumes. Q M— — —— 10.00 Moss, C. E. The Cambridge British Flora. volume 2 only plus the volume of plates. Cambridge, 1914. folio .................... 20.00 REICHENBACH, H. G. L. Icones florae Germanicae . . . In- complete. Volumes 1 through 17 bound in 9 and vol. 18 incomplete, pages 1-72, pl. 1202-1331. Volume 1 is the 2nd ed. rev. of 1850 with 192 plates. Text in Latin, the plates are uncolored. Leipzig, 1850, 1837-56. folio ............ 200.00 SMITH, JAMES EDWARD. Flora brittanica. London. 1800-04. Three volumes 8? ............. eese eee 25.00 SMITH, JAMES EDWARD. A selection of the correspondence of Linnaeus and other naturalists from the original manu- scripts. London, 1821. 2 volumes. 8° ................ 25.00 ADDRESS THE LIBRARIAN GRAY HERBARIUM OF HARVARD UNIVERSITY 22 Divinity Avenue, Cambridge 38, Mass. REFERENCK LIBRAM! APR 2 8 1961 Dodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS | RA PUHICARTETON BEAN Associate Editors IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER | ROLLA MILTON TRYON Vol. 63 April, 1961 No. CONTENTS: Some New Aspects of the Fern Platyzoma microphyllum. PECIA EIUS A E E E E E E a TEOR The Genus Lechea (Cistaceae) in the Southeastern United States. Robert L. Wilbur and Hazim S. Daoud .................. Mentezelia albescens and Lonicera Xylosteum in Missouri. Ernest J. PAUNO PRECES A New Variety of Rudbeckia fulgida. Robert E. Perdue, Jv... eiie eere tette isana The Nem England Botanical Club, Inc. » c 2 n ES i Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume. Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able. All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass. E ? Printed by THE LEXINGTON PRESS, INC. Lexington, Mass. Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 April, 1961 No. 748 SOME NEW ASPECTS OF THE FERN PLATYZOMA MICROPHYLLUM ALICE F. TRYON Since this north Australian fern was described by Robert Brown in 1810 and placed in the Gleicheniaceae, it has been the subject of several papers. Because of its unusual stelar structure and absence of leaf gaps it was considered particu- larly in theories relating to the origin of the stele. In the most recent of these morphological studies by J. M. Thomp- son (1919) the literature is reviewed. F. O. Bower (1926) in his survey of the primitive ferns cites Thompson's work and figures the reconstruction of the stele and the irreg- ularities of the sporangium. In connection with his studies on the Gleicheniaceae, R. E. Holttum (1956) has discussed the relationships of Platyzoma. He excludes it from that family on the basis of the filiform leaves, the occurrence of waxy indument and the terminal position and characters of the sporangium. He tentatively refers it to a place among the gymnogrammoid ferns on evidence drawn from a com- parison with Jamesonia. While there are good reasons for excluding Platyzoma from the Gleicheniaceae, there are some objections to its close alliance with Jamesonia. A study of Platyzoma was undertaken in search of new evidence which might help to establish its relations more certainly. Observations have been made on the following specimens in the Gray Herbarium and the United States National Her- barium’. II am most grateful for the loan of specimens from the United States National Her- barium, and for the opportunity to study the collections and the use of facilities at the Gray Herbarium. 91 92 Rhodora [Vol. 63 AUSTRALIA. QUEENSLAND: near Hughenden, Brass & White 64 (frag- ment GH ex K); northern Queensland, Hann (Us); Thursday Island, N. J. Hey, in 1917 (us); Dimbulah, Cook District, Hubbard & Winders 6556 (US); Hughenden, Burke District, Hubbard & Winders 7592 (US); Rockingham Bay, F. Müller (GH); between Inglewood and Mill- merran, Darling Downs, C. T. White 9705 (GH) ; between Cecil Plains and Millmerran, Darling Downs, White & Webb 1177 (GH). NORTHERN TERRITORY: east of Borroloola, R. A. Perry 1851 (Us); east of Carlton Station, R. A. Perry 3005 (us); in Gulf of Carpentaria, Groote Ey- landt, R. L. Specht 197 (us); Port Bradshaw, Arnhem Land, R. L. Specht 716 (us); Oenpelli, Arnhem Land, R. L. Specht 1252 (us). WESTERN AUSTRALIA: Cambridge Gulf, F. Müller, (GH); east of Kim- berly Research Station, R. A. Perry 2566 (US). OBSERVATIONS ON PLATYZOMA MICROPHYLLUM Roots. — The roots are long, coarse and of a diameter nearly equal to that of the petioles. They arise from the lower surface of the rhizome. The outer tissue of the roots appears spongy and consists of large, lustrous, light brown, parenchyma cells. Long persistent root hairs arise from the surface of these cells forming dense mats in which parti- cles of white sand may be enmeshed. Rhizome. — The rhizome is dorsiventral, ca. 0.4 em. in diameter with a tomentum of rigid, concolorous, lustrous, rust colored, multicellular trichomes. The trichomes are relatively long and consist of about 30-60 cells (Fig. 1). These cells are as long as or up to four times longer than broad, and are not arranged in any regular sequence of size. The base of the trichome may be catenate (Fig. 2); the cells being flattened and twisted. The terminal cell is short and bulbous. The leaves are close but irregularly placed on the rhizome or fasciculate. Many arise from the upper and lateral surfaces but some of the pinnate leaves and most of the filiform ones arise from the lower surface of the rhi- zome. The longest rhizome (with attached leaves) was 4 cm. long and there were 4 circinate leaves at the apex, 34 pinnate leaves and about 50 filiform leaves. The filiform leaves in some specimens are fascicled while in others they are quite uniformly distributed along the entire rhizome. Details of the structure of the rhizome are given by Thomp- son (1916). It is described as a medullated protostele with 1961] Tryon — Platyzoma microphylium 93 a discontinuous inner endodermis and lacking leaf gaps. Pinnate leaves. — Fully developed leaves have an expanded apex with a terminal pinna although many leaves are not entirely developed. In the smallest complete leaf the petiole was 2.5 em. long, the lamina 11.0 cm. long and 0.3 cm. wide. The lamina of the longest leaf, with a broken apex, was 32.0 cm. long, 0.4 cm. wide and the petiole was 8.0 cm. long. Most leaves are about 24.0 cm. long. PETIOLE: The petiole is terete or oval near the rhizome, atro- purpureous and with sparse, multicellular, capitate tri- chomes similar to those of the rhizome. At the apex of the petiole the adaxial half is flattened in three planes and the abaxial half is convex. There are no stomata on the petiole but small mounds of sclerotic tissue rarely occur. RACHIS: The adaxial surface of the rachis is flat, slightly lighter in color than the petiole or sometimes greenish. The lateral surfaces, on which the pinnae are borne, are flat or slightly grooved. There are abundant, glandular, multicel- lular trichomes on the lateral and adaxial surfaces of the rachis (Fig. 3). The apical portion of the rachis is attenuate and heavily indumented. PINNAE: The lamina is once pinnate with about 500 small, simple pinnae which are easily detached from the rachis in dried specimens. Articulation occurs in the stalk without specialized tissue. Most leaves are wholly sterile. Fertile pinnae occur in zones usually in the terminal half or quar- ter of the lamina and on one or both sides of the rachis. One leaf consisted of 433 pinnae of which 11, borne on one side of the rachis, were fertile. The pinnae are somewhat blad- der-shaped with a central slit. The margins are more or less enrolled and nearly meet at the center of the pinna. The pinna stalk is usually brown, 0.1-0.25 mm. long, 0.2 mm. broad with a cushion of parenchyma cells at the apex which is attached to the lower epidermis and is very glandular. The cells of the lower epidermis are thin walled and about 4 to 6 times longer than broad (Fig. 4). In surface view the cell walls are undulate with rounded lobes. Stomata occur only on this surface and are abundant adjacent to the 94 Rhodora [ Vol. 63 costa and more diffuse towards the margins. Two- or three- celled, capitate glands (Fig. 5) which secrete copious quan- tities of yellow wax-like substance, are especially abundant along the veins. The mesophyll is thick and with many large lacunae. The cells of the upper epidermis are thick walled and have a shape and a pattern distinct from those of the lower epidermis (Fig. 6). The pattern, which can be ob- served in dried specimens under 15 X magnifications, con- sists of a band of elongate cells about 6 to 12 times longer than broad with undulating walls. This band extends through the center of the pinna from the stalk to the apex where it becomes flabellate. Adjacent to it and along the curved sides of the pinna the cells are broader and deeply dissected. The vascular system of the pinna is a short sym- podium with 6-12 lateral branches (Fig. 7). The central vein is slightly flexuous and terminates in an unequal di- chotomy. The lateral veins are unbranched and become broader toward their distal end. They terminate well back of the pinna margin and the ends are acute or slightly en- larged. The veins are situated in the spongy mesophyll somewhat closer to the upper than the lower epidermis. They consist of a cord of short, thick-walled tracheids so tightly joined that the entire sympodium can be withdrawn from a pinna softened in sodium hydroxide. STERILE PINNAE: These are rigid herbaceous, 0.5 - 4.0 mm. long, 0.5 - 2.0 mm. wide, ovate or somewhat orbicular with enrolled margins (Fig. 7). A central slit extends from the stalk to short of the apex and the pinna margins adjoining this are strongly papillate. There are some unusual pinnae on the leaves of one collection (Hubbard & Winders 6856). The lower pinnae on these are 2 or 3 times as long as the normal ones and clavate or boat-shaped (Fig. 8). The base is narrowed and decurrent on the broad and greenish ad- axial surface or on the light brown colored lateral surface of the rachis. If a pinna stalk can be distinguished from the decurrent tissue, it is longer and broader than those of the normal pinnae and only partially brown. The margin is irregular with protruding, elongate cells. The pinna is 1961] Tryon — Platyzoma microphyllum 95 thickened on its abaxial surface by a ridge of tissue extend- ing from the stalk to the apex. Such a ridge also occurs in otherwise normal pinnae of the lamina and there seem to be transitional forms between these elongate ones and the normal pinnae. FERTILE PINNAE: These are slightly larger than the sterile pinnae, orbicular and the margins are not or only slightly enrolled. The papillae from opposite margins may be some- what enmeshed and close the bladder-shaped pinnae. The fertile pinnae are often a darker brown color and more wrinkled than the sterile pinnae. Filiform leaves. — These are simple, linear structures with attenuate apices. The apical portion is straw colored or light brown, the basal portion is atropurpureous. They are 3.0 - 8.0 em. long, 0.05 em. wide. There are two prominent grooves, presumably on the adaxial side, and two or more shallow ones on the other side extending nearly the length of the leaves. Stomata may be more or less in rows in the shallow grooves (Fig. 9). Sporangia. — Sporangia are attached along the lateral veins usually nearer to the costa than to the distal end of the vein (Fig. 10). They have been observed on the distal portion of the vein but not in a terminal position. There is consid- erable variation in the size of the sporangia for among mature ones within the same pinna some may be twice the size of others. There is also variation in the shape, spore content and the position and structure of the annulus. Such irregularities are discussed and illustrated by Thompson (1917, p. 160) and Bower (1926, p. 209). There is however a rather uniform sporangial type which will be described here. Sporangia in several stages of development are found along a single vein and up to 13 have been observed on a single pinna. The sporangium stalk is short, usually of two or three tiers of cells. Three basal cells have been observed in immature sporangia. At the capsular end of the stalk there are either 3 or 4 cells which differ from other stalk cells in having striated thickenings. The capsule of the sporangium is spherical, the annulus is broad, slightly raised [Vol. 63 Rhodora 96 Platyzoma microphyllum 1961] Tryon — Platyzoma microphyllum 97 and usually more or less oblique with the capsule faces cor- respondingly unequal in size (Fig. 11). The annulus (Fig. 12) is composed of a single row of cells which may vary in size and shape, but its width is roughly one third the diam- eter of the capsule. There are about 40 indurated cells which extend from the base of the capsule around and below its apex. There are about 4 cells with thin and undulating walls which interrupt this annular series. These are followed by 7-14 indurated cells similar to those of the annulus. The sporangium opens between these indurated cells at a point usually above their center. There are a few, usually 2 or 4, cells with thin and undulating walls between these indurated ones and the sporangium stalk. The annulus is interrupted by the stalk and the rupture of the sporangium occurs be- tween several indurated cells which resemble those of the longer annular series. A unique characteristic of the spo- rangium is the cells of the capsule faces (Fig. 11). In surface view these are strongly undulating and somewhat resemble the cells of the lower epidermis. Spores. — The number of spores in a sporangium and their size seem to correlate with the size of the sporangium. The larger sporangia usually contain 16 large spores and small sporangia contain 32 small spores. There are also sporangia intermediate in size and in one of these, which was open, I found a single large spore nearly filling the capsule and 4 small spores. In six collections examined both 16- and 32- spored sporangia were found on the same pinna. Most Tracings from cleared material of Platyzoma microphyllum — Fic. 1. Rhizome tri- chome, X 10, Perry 2566. Fic. 2. Catenate base from rhizome trichome, X 15, Perry 2566. Fic. 3. Glandular trichomes from rachis, X 35, White 9705. Fic. 4. Portion of lower epidermis of pinna, X 45, Specht 1252. Fic. 5. Glands from abaxial surface of pinna, X 60, Specht 197. Fic. 6. Portion of the upper epidermis of the pinna. Elon- gated cells at the left are from the central band, X 45, Specht 1252. FiG. 7. Sterile pinnae on a portion of the rachis. The venation darkened in the upper; the lower showing the outline of the margins, the veins in broken lines, X 15, White 9705. Fic. 8. An elongate pinna, decurrent on a portion of the rachis, the apex torn by flattening, the vascular tissue shaded, X 15, Hubbard & Winders 8656. Fig. 9. Portion of the epidermis of a filiform leaf with stomata, X 45, Specht 1252. Fia. 10. Vascular system of a pinna with attached immature sporangia, X 15, Brass & White 64. Fic. 11. A sporangium with the aperture shaded. With indurated cells above and below the gap, the three celled stalk in the lower part, X 70, Brass & White 64. Fic. 12. An opened annulus of a sporangium, with a three celled stalk in the lower part, X 70, Brass & White 64. 98 Rhodora [Vol. 63 spores are tetrahedral with a trilete proximal face. In a few sporangia there were also bilateral spores some of which were trilete with ridges of unequal length and others were monolete. The spores are prominently sculptured. The proximal face has long, broad ridges which are somewhat parallel to the commissural ridges and are connected by shorter ones to form a reticulum. The distal face is also prominently sculptured but the rugae are mostly short and form circular or irregular loops. In the equatorial region there may be one or a few parallel ridges nearly continuous around the spore. The spores are often retained in open sporangia. They are also found free within the pinna, some- times covered with the yellow indument from the glands. DISCUSSION some of these observations are at variance with those previously reported by Thompson and Holttum. The first is the position of the sporangia which Thompson (1916) re- ports as terminal on the vein ends. Although sporangia may occur on the distal portion of the veins, I have not observed them to be exactly terminal in cleared pinnae. They occur on the lateral veins adjacent to the costa and along the veins to near the distal end. The position is best observed when the sporangia are immature. I interpret them as lateral rather than terminal on the veins. The apical growth of the leaves has been described as indefinite. Some leaves have undeveloped apices which may expand slowly ; however they do terminate in a completely developed apex before the lower pinnae have deteriorated. The leaves are determinate and the lamina is imparipinnate. The spores are mostly tetra- hedral and trilete as reported but a few spores have been observed which are bilateral and monolete. From this survey of Platyzoma it was noted that there are certain similarities with the Schizaeaceae. A medullated pro- tostele with sclerosed pith and inner endodermis are reported by Bower (1926) in species of Anemia and Sehizaea. The rhizome in these genera has multicellular trichomes. Re- duced filiform leaves with a central vascular trace and stomata are found in several species of Schizaea — S. fistu- 1961] Tryon — Platyzoma microphyllum 99 losa Labill., S. papuana Brause, and they are especially abundant in the North American S. pusilla Pursh. In the filiform leaves of Platyzoma the guard cells may be larger relative to the adjacent cells than in Schizaea, but the form of the intra-stomatal cells is similar. In the elongate pinnae which were found in one collection of Platyzoma there is a resemblance to the form of the lobes of the lamina in Schizaea. Since these pinnae have been found in only one specimen of Platyzoma they might be considered as ab- normal. In the company of other similarities with Schizaea, I am inclined to regard them as something more than an anomalous condition. On the leaves of many species of Anemia and Schizaea there are 2- or 3-celled capitate glands. These are similar to the glands on the pinnae of Platyzoma although the basal cell may be larger and there is no yellow- ish indument. The sporangium of Platyzoma is most re- markable in its spherical shape, short stalk, its broad, more or less oblique annulus which is scarcely raised from the surface of the capsule, and especially in the undulating cell walls of the capsule faces. In Moria and some species of Anemia the sporangia are spherical and short stalked or sessile. The cell walls of the capsule in Schizaea melanesica Sell. and S. confusa Sell. (Selling 1944b, 1947) are undu- lating and similar to those of Platyzoma. There is also similarity in the reticulate sculpture of the spores between S. confusa and Platyzoma. In his studies on the spores of Schizaea O. H. Selling (1944a) reports variations in size and seulpturing. In S. fistulosa there seem to be two sizes which show some correlation with the geographic distribu- tion. It would be of interest to determine whether such variation in spore size in Schizaea and Platyzoma might be related to the apogamous condition as it is in species of Pellaea (Tryon & Britton, 1958). In addition to an evaluation of similarities between Platyzoma and Schizaea, the distinctive character of the Schizaeaceae — the apical annulus — must be considered, and also the possibility of parallel reduction in these two genera. Resemblances in the sporangia are remarkable and 100 Rhodora [Vol. 63 there is little possibility of parallel reduction in this struc- ture. In Platyzoma the annulus is usually oblique — with unequal capsule faces, it is scarcely raised from the surface and is about one third the diameter of the capsule. There is a series of indurated cells in the stomial region and epidermal-like cells in the capsule faces. In these features the resemblance seems to be with Schizaea although there are similarities with species of Anemia and Mohria which have spherical sporangia, and in which there are several non-indurated cells in the apical portion of the sporangia. While the sporangia of Platyzoma are not identical to those of the Schizaeaceae, in this structure and in some aspects of the leaves there are resemblances closer to that group than to any other family. There is also similarity to the Polypodiaceae particularly in the interruption of the annulus by the sporangium stall. Platyzoma can be placed here in the subfamily Gymnogram- moideae and in the tribe Gymnogrammeae on the super- fieial sporangia following the course of the veins. In this tribe it would be allied to Eriosorus and Jamesonia by its monomorphic leaves and pubescent rhizome. It was tonta- tively placed here by Holttum (1956). in the course of the preparation of a revision of Jamesonia, i have studied its species in some detail and wish to remark upon a number of characters of Jamesonia which differ from Platyzoma. The species of Jamesonia grow mainly on the South American paramos at altitudes frcm 2350 to 5000 meters. The species center in Colombia and Venezuela and occur northward into southern Mexico and southward to Bolivia and eastern Brazil. In the paramos Jamesonia is usually found growing in the shelter of rocks with grass or moss. The rhizomes are horizontal, creeping, dichotom- ously branched, with spirally arranged leaves, and are sparsely to densely covered with simple, multicellular tri- chomes. In some species the trichomes have more massive bases and become rather scale-like. The leaves in some but not all species are indeterminate and there are no reduced 1961] Tryon — Platyzoma microphyllum 101 or filiform leaves such as in Platyzoma. Young leaves are produced behind the apex on older parts of the rhizome and it is possible that these or a cluster of old petioles may have been the basis of the report of reduced leaves. A dense pu- bescence or glutinous exudate is characteristic of the leaves of Jamesonia but ceraceous indument is rare. It is indeed of such rare occurrence on the abaxial surface of the pinnae that it has been observed there in only one collection — the type of J. ceracea Maxon. It is perhaps in the form of the pinnae that Jamesonia and Platyzoma differ most markedly. The pinnae in Jamesonia are generally flat with the margins more or less enrolled. The margin is usually membranous or ciliate. The pinnae are relatively long stalked, rarely subsessile and in two species adnate. The venation of the pinnae is dichotomous with the ramifications branching and terminating usually in numerous ultimate veins which extend to or near the pinna margin. The form of the an- nulus may be somewhat irregular but there are generally about 20 raised and indurated cells. There is a distinct stomial region with 2-4 indurated lip cells which are smaller than those of the annulus. The sporangial stalk may be short — one quarter or less than the length of the capsule or equal to it in length. The spores of Jamesonia are entirely different from those of Platyzoma in the type of sculpture. They are trilete with 3 broad smooth or verrucate planes on the proximal face; there is a prominent equatorial ridge or wing and 3 broad ridges forming a triangle on the distal face. Jamesonia is a specialized group, closely related and with species transitional to Eriosorus. Although there is some resemblance between Jamesonia and Platyzoma the relationship is not a close one. Platyzoma is properly excluded from the Gleicheniaceae and can, in the present classification, be placed in the Poly- podiaceae. This disposition however, adds to the problems of the definition of that family. There is general agreement that the Polypodiaceae, as treated by Christensen, is poly- phyletie but there is no clear understanding of the natural groups. 102 Rhodora [Vol. 63 It is my belief that the Schizaeaceae is a source from which some of those groups have been independently de- rived. Sinopteris and some species of Cheilanthes have similarities with Mohria; Eriosorus, Jamesonia and Ptero- zonium comprise a group having certain resemblances to Anemia. While in Platyzoma there are similarities with Schizaea in the shape of the pinnae and lobes of the lamina, the filiform leaves, the spores and particularly in the struc- ture of the sporangium. LITERATURE CITED Bower, F. O. 1926. The Ferns. Vol. II. The Eusporangiate and Other Relatively Primitive Ferns. 344 pp. University Press. Cambridge, England. BROWN, ROBERT. 1810. Prodromus Florae Novae-Hollandiae et Insulae Van Diemen. J. Johnson. London, England. CHRISTENSEN, CARL. 1938. Filicinae. In Verdoorn's Manual of Pteri- dology. The Hague, Netherlands. HorTTUM, E. R. 1956. On the nature and possible relationships of the fern genus Platyzoma R. Br. Kew Bull. No. 8: 251-258. SELLING, O. H. 1944a. Studies in the recent and fossil species of Schizaea, with particular reference to their spore characters. Med- del. Góteborgs Bot. Trad. 16: 1-112. 1944b. A new species of Schizaea from Melanesia and some connected problems. Svensk Bot. Tidskr. 38: 207-225. 1947. Further studies in Schizaea. Svensk Bot. Tidskr. 41: 431-450. THOMPSON, J. M. 1916. The anatomy and affinity of Platyzoma micro- phyllum R. Br. Trans. Roy. Soc. Edin, 51: 631-656. 1917. A further contribution to the knowledge of Platyzoma microphyllum R. Br. Trans. Roy. Soc. Edinb, 52: 157- 165. 1919. The morphology of the stele of Platyzoma microphyllum, R. Br. Trans. Roy. Soc. Edinb. 52: 571-595. TRYON, A. F. AND D. M. BRITTON. 1958. Cytotaxonomic studies on the fern genus Pellaea. Evolution 12: 137-145. 1961] Wilbur and Daoud — Lechea 103 THE GENUS LECHEA (CISTACEAE) IN THE SOUTHEASTERN UNITED STATES ROBERT L. WILBUR! AND HAZIM S. DAOUD The genus Lechea is a small group of about seventeen species found only in North and Central America and in the West Indies. Actually all of the species occur in the United States east of the Rocky Mountains except for one little known and seldom collected endemic from western Cuba. L. tripetala, which is found in Oklahoma, occurs in the United States only as one small population widely sepa- rated from the principal area of the species in Mexico and Guatemala. The genus is particularly well-represented in the southeastern United States (here defined as the area east of Texas and Oklahoma and south of Missouri, Ken- tucky and Virginia) where ten species occur and four of them are endemic. L. Torreyi cannot rank as an endemic due to its puzzling reappearance in the pine forests of Brit- ish Honduras. The variety of L. Leggettii found in the Southeast barely ranges beyond its borders but does occur in southeastern Virginia. No other region possesses so many species of the genus nor can claim so rich and varied an endemic element. Lechea is most strongly and clearly distinguished from all other genera within the family by its trimerous, persistent corolla ; sessile, fimbriate stigmas and the presence of three incomplete partititions within the capsule. Hence for all practical purposes there has never been any real confusion over generic limits. Lechea has usually been considered a diffieult genus due to the rather minute technical characters upon which the species distinctions rest. The admirable re- vision of Hodgdon (1938) has provided a firm foundation for an understanding of the genus by clarifying species limits and evaluating morphological variation. The key pro- vided, however, relies to a considerable degree upon char- acters not present during the season when most specimens 1Grateful acknowledgement is made to the National Science Foundation for a grant of research funds to Duke University (NSF-G5636) which were used in part for this study. 104 Rhodora [Vol. 63 are collected, and therefore is not as usable as might be desired. Even plants collected with mature fruit and seeds often have not yet developed the basal shoots often called for in the key. These shoots are commonly not present until very late in the growing season. Although it does not now seem possible to prepare a key for sterile or even young flowering material, it is thought that the following key will aid in the determination of fruiting specimens of those species found in the southeastern United States. Rafinesque (1836) in his monograph of the genus placed considerable emphasis upon the comparative lengths of the inner and outer sepals. The three subgenera recognized by him were actually largely characterized by the compara- tive lengths of the two whorls of the calyx. Since then most workers have placed considerable reliance upon this char- acter. We too have found it a useful character and hence have employed it; but having seen considerable variation, especially in L. racemulosa, have felt it necessary to place this species under both leads of the key. Small's key (1933) is rendered almost unusable by too complete a reliance upon the constancy of this character. For example one would be in a quandary in attempting to determine a specimen of L. villosa with Small’s key since in that species the outer sepals typically equal the inner and his key treats all species as having the outer lobes either longer or shorter than the inner sepals. It was hoped when the present study was undertaken that it might result in some suggestions as to the relationships of the species and perhaps even a grouping of them into formal series or sections. Rafinesque recognized three sub- genera based largely upon comparative lengths of the caly- cine whorls and also stamen and seed number. Spach in 1837 segregated the Texan species now known as Lechea san-sabeana as the monotypic genus Lechidium based largely upon the extreme development of the incomplete partitions found within the capsule. This taxon has since been treated as of subgeneric or sectional rank. Hodgdon accepted it as a section although admitting its close affinity to L. tenuifolia 1961] Wilbur and Daoud — Lechea 105 and L. tripetala. Small (1933) in his treatment of the genus, covering almost the same area as that in this paper, segre- gated the species into five named groups whose rank, as is eustomary in Small's Manual, is undesignated. Hodgdon declined to provide formal subgeneric categories except for the section Lechidium. Although the genus seems admirably suited for the construction of a phylogenetic tree based upon the rather obvious reduction series, our data are not suffi- ciently complete for us to present a phyletic scheme. We are accepting the nomenclature of Hodgdon's revision and are ignoring the perplexing monograph presented by Rafinesque in 1836. It seems probable, however, that some of the twenty new species published by Rafinesque will eventually prove identifiable and will have priority over some of the names now current. In the course of this study more than 4000 herbarium specimens were examined and annotated. For having made their collect’ons available to us in whole or in part, we are indebted to the curators of the herbaria indicated below by the abbreviations: BUS, CU, F, FLAS, FSU, GA, IA, KY, MICH, MISSA, NCSC, NCU, NO, NY, OS, PUL, SMU, TENN, TEX, UARK, USF, VDB, U. of So. Carolina. KEY TO THE SOUTHEASTERN SPECIES OF LECHEA 1. Pubescence of aerial stems for the most part strongly divergent and spreading. 2. Internal sepals conspicuously V- or U-shaped in cross-section with thin searious margins and a roughened, often sparsely pilose keel (the sepal otherwise glabrous); leaves, or at least many of them, over 1:5 em long; capsule thin-walled, readily splitting into 3 valves at maturity, subglobose, about equaling the calyx in length; widespread throughout much of the east- ern- United -States os cccesssescesesecovscrdeeeceatnedecisenstessdeveeocstes 1. L. villosa. 2. Internal sepals but slightly bowed in cross-section with texture appearing uniform and pubescence scattered across surface; leaves all less than 1 em long; capsule thick-walled, indehiscent ; ellipsoid or somewhat rounded- barrel-shaped, exserted from the calyx for at least 1/3-1/2 its length; restricted to peninsular Isloridapu Nec UMEN ITEM 2. L. divaricata. 1. Pubescence of aerial stems mostly closely appressed or lacking. 3. External sepals equaling or exceeding the internal sepals in length. 4. Calyx of fruit and of older flowers strikingly differentiated into an obconic to somewhat cylindrical base about 0.4-0.6 mm 106 Rhodora [Vol. 63 long, shiny and yellowish in color and scarious in texture contrasting considerably with the texture, color and appear- ance of the inner sepals; pedicels (at least after anthesis) averaging 2 mm or more in length.............. 7. L. racemulosa. Calyx-base of fruit and of older flowers not conspicuously differentiated in texture, color or appearance from the inner sepals; pedicels averaging less than 2 mm long. 5. Capsule completely enclosed by sepals, stigma not exposed after anthesis; fruit and inner sepals together sub- globose; cauline and rameal leaves averaging 10 or more times as long as broad................. eee 3. L. tenuifolia. Capsule slightly to conspicuously exserted, the sepals not enclosing the top of the fruit; stigma exposed on develop- ing and maturing fruit; fruit and inner sepals together cylindrical to pyriform (at least higher than broad); cauline and rameal leaves mosily less than 9 times as long as broad. 6. External sepals at least one-fifth longer than the inner sepals and usually equaling or exceeding the capsule in length; capsule equaling or exceeding the inner sepals by not more than one-fifth its length; cauline leaves usually elliptic to elliptic-oblong, usually less than 5 times as long as wide, those below the inflorescence commonly appearing whorled and often more than 2 mm widoe.....................s 4. L. minor. 6. External sepals about equaling the inner sepals and never equaling the capsule in length; capsule usually exceeding the inner sepals by about 1/3-1/2 its length; cauline leaves narrowly oblong to linear, usually 6 times (or more) as long as wide, those below the in- florescence alternate and less than 2 mm wide............ MM 5. L. patula or 3. External sepals shorter than the internal sepals. [d Leaves pubescent on both upper and lower surfaces (at least those of the basal shoots conspicuously pilose above and below while the cauline and rameal leaves are usually incon- spicuously pubescent over entire surface); flowers or fruits mostly clustered in 2's or 3's: capsule wall thickened and indurate; plants of peninsular Florida............ 10. L. cernua. Leaves variously pubescent below but glabrous on upper sur- face; flowers or fruits not appearing fascicled (attached separately); capsule wall thin or at least not conspicuously indurate. 8. Aerial stems perennial, suffruticose, clearly woody at base, with wiry woody branches; capsule exserted from the often spreading calyx by 1/3-1/2 its length; calyx 1961] Wilbur and Daoud — Lechea 107 sparingly short-pubescent to glabrous; known only from southern Georgia and Florida...................... 6. L. Deckertit. 8. Aerial stems annual, herbaceous, dying to the base each year; capsule equaling the calyx or exserted not more than 1/5 its length from the closely enveloping sepals; calyx moderately to densely pilose. 9. Capsule and calyx together narrowly obovoid, about twice as long as broad; capsule narrowly ellipsoid or ovoid, 1.5 times or more as long as broad; pedicel filiform, about 0.1 mm in diameter; basal leaves, when present, less than 5 times as long as wide; distribution primarily in the Piedmont and Mountains but also occurring in the Coastal Plain............ 7. L. racemulosa. 9. Capsule and calyx together broadly pyriform to obo- void, usually less than 1.5 times as long as broad; capsule ovoid to subglobose to broadly ellipsoid, about 1-1.3 times as long as broad; pedicel stout, at least 0.2 mm in diameter; basal leaves, when present, 6 or more times as long as wide; distribution typically of the Coastal Plain and the outer Piedmont. 10. Leaves abruptly tapering at apex into a hardened, shiny, conical callosity about 0.25 mm long; inner sepal clearly 3-nerved (often best demonstrated by moistening); pedicels averaging over 1.5 mm long; capsule exceeding the sepals by about 1/5 its length; seeds mostly 2................ 8. L. Leggettii. 10. Leaves pointed but not differentiated into a cal- lositv; inner sepal 1-nerved; pedicels averaging less than 1.5 mm long; capsule almost completely enveloped by the sepals; seeds mostly 3-6............ MOTEL EUM PARE re Bae ar Code 9. L. Torreyi. 1. L. villosa Ell. Hodgdon recognized three varieties of this species. The one found throughout much of eastern North America is var. villosa (—var. typica Hodgd.). Ap- parently overlapping the range of this eastern variety along the extreme western fringe of the species distribution is a relatively little collected plant of problematic distinctness called var. macrotheca Hodgd. Another variant of which there is too little material available for adequate appraisal is var. Schaffneri Hodgd. This is known only from a few stations in northeastern Mexico (Tamaulipas and San Luis Potosí). The conspicuously carinate inner sepals of L. vil- losa are diagnostic and together with several other distinc- 108 Rhodora [Vol. 63 tive features make this species easy to recognize and perhaps indicate a relatively isolated position within the genus. As is to be expected in so widespread a species, considerabie Figures 1-10. Sepals and capsules of Lechea (pubescence not shown.) Fig. 1. Lechea villosa; fig. 2. L. divaricata; fig. 3. L. tenuifolia; fig. 4. L. minor; fig. 5. L. patula; fig. 6. L. Deckertii; fig. T. L. racemvlosa; fie. 8. L. Leggettii; fig. 9. L. Torreui: fig. 10. L. cernua. 1961] Wilbur and Daoud — Lechea 109 variation is encountered which is striking even within the range of var. villosa. In the course of our studies two collections were examined that seemed definitely to be hybrids between L. villosa and L. racemulosa. As both of these species are extremely wide- ranging and in part share broadly overlapping areas, it is no doubt indicative of the effectiveness of whatever isolating mechanisms that they possess that so little evidence of hy- bridization was encountered. The putative hybrids were both collected from the sandy inner margin of the Coastal Plain at the outer margin of the range of L. racemulosa. One (Radford 26436 NCU) was from Edgefield Co., S. Carolina and the other (Wiegand & Manning 2072 CU) was collected in Moore Co., N. Carolina. The spreading stem pubescence of these plants resembles that of L. villosa and the sepals are keeled, but not so strongly, as they are in L. villosa. The leaves are shaped more like those of L. racemu- losa and the shape of the inflorescence also is more like that species. RANGE: The typical variant occurs beyond our area as far north as southern New Hampshire and as far west as eastern Texas and into Illinois. Specimens examined from the following counties: NORTH CAROLINA: Anson, Bladen, Brunswick, Carteret, Cherokee, Cleveland, Craven, Cumberland, Currituek, Dare, Duplin, Greene, Halifax, Harnett, Hyde, Iredell, Johnston, Jones, Lenoir, Montgomery, Moore, New Hanover, Northampton, Onslow, Pender, Richmond, Robeson, Scotland, Wake, Wayne, Wilson. SOUTH CAROLINA: Aiken, Allendale, Barnwell, Beau- fort, Calhoun, Charleston, Chesterfield, Clarendon, Colleton, Darling- ton, Dillon, Dorchester, Edgefield, Fairfield, Florence, Georgetown, Hampton, Horry, Lancaster, Lee, Lexington, Marion, Marlboro, Mc- Cormick, Newberry, Orangeburg, Richland, Saluda, Sumter, Williams- burg. GEORGIA: Atkinson, Baker, Brantley, Charlton, Clay, Decatur, Dougherty, Echols, Glynn, Jones, McDuffie, McIntosh, Pierce, Rich- mond. FLORIDA: Alachua, Citrus, Clay, Columbia, Dixie, Duval, Escam- bia, Franklin, Gilchrist, Hamilton, Hernando, Lake, Liberty, Leon, Levy, Madison, Manatee, Nassau, Okaloosa, Putnam. TENNESSEE: Bledsoe, Chester, Coffee, Dickson, Fayette, Franklin, Gibson, Hender- son, Hickman, Lawrence, Lewis, McNairy, Montgomery, Rutherford, Sumner. ALABAMA: Jefferson, Lee, Mobile, Montgomery, Sumter, Tal- lapoosa. MISSISSIPPI: Clarke, Hancock, Harrison, Hinds, Jackson, Jef- 110 Rhodora [Vol. 63 ferson, Lafayette, Lincoln, Monroe, Oktibbeha, Pike. LOUISIANA: Allen, Caddo, Claiborne, Grant, Jackson, Lincoln, Rapides, Sabine, Tangipa- hoa, Union, Vernon, West Feliciana. ARKANSAS: Baxter, Benton, Cal- houn, Clark, Cleveland, Columbia, Conway, Dallas, Drew, Faulkner, Garland, Hempstead, Independence, Izard, Jefferson, Johnson, Logan, Marion, Miller, Phillips, Pike, Prairie, Pulaski, Saline, Sevier, Stone, Union, Washington. 2. L. divaricata Shuttlew. ex Britt. A most distinctive species restricted in range to southern peninsular Florida. In our key, L. divaricata keys out next to L. villosa with which it was long confused (or at best treated as a Floridian variant). Hodgdon comments upon their relationship as follows: “It would be difficult to find any Lechea differing more from L. villosa in everything except quality of pubes- cence and the general shape of the leaves, both of which are vegetative characters and not in themselves necessarily indicative of close relationship." Hodgdon placed more con- fidence in the characters of the fruit which he felt indicated the relationships of this species to be closest to L. Deckertii and L. patula. As is usual when facts are few, speculation is unfettered; obviously too little is known at present to be at all dogmatic about relationships. Although L. divaricata is an isolated species, its relationships would seem to us somewhat closer to L. villosa than to L. Deckertii or L. patula. RANGE: Endemic to the southern half of peninsular Florida but apparently absent from the Keys. Specimens examined from the following counties of Florida: Brevard, Broward, Dade, Hernando, Highlands, Hillsborough, Lee, Manatee, Martin, Pinellas, Seminole. 3. L. tenuifolia Michaux. A widespread species easilv recognized by its very narrowly linear leaves (usually 1.5 mm or less wide), globose capsules and enveloping calyx, and its external sepals that are equal to or more commonly ex- ceed the inner. According to Hodgdon's treatment our repre- sentatives of this species belong to var. tenuifolia which is distinguished by its smaller stature, fruit and seeds from var. occidentalis Hodgd. whose range it somewhat overlaps only in northwest Illinois. In our area it can be confused in the young flowering stages with L. patula and in such 1961] Wilbur and Daoud — Lechea 111 cases the greater degree of pubescence and coarseness of the trichomes of L. tenuifolia are the best guide to its identity. RANGE: Hodgdon's typical variety extends beyond our area as far north as southern Maine and as far west as Min- nesota and eastern Texas. » M LÀ kd Ste eN n^» \ e 6 E "S 4 3 *. s Maps 1-6. Distribution of Lechea in the Southeastern United States. Map 1. Lechea villosa; map 2. L. divaricata; map 3. L. tenuifolia; map 4. L. minor; map 5. L. patula; map 6. L. Deckertii. 112 Rhodora [Vol. 63 Specimens examined from the following counties: NORTH CAROLINA: Cabarrus, Caswell, Davie, Edgecombe, Durham, Forsyth, Franklin, Granville, Halifax, Harnett, Lee, Northampton, Orange, Person, Row- an, Stokes, Union, Vance, Wake, Warren. SOUTH CAROLINA: Abbeville, Anderson, Cherokee, Chester, Chesterfield, Darlington, Fairfield, Greenwood, Laurens, Lee, Lexington, McCormick, Newberry, Richland, York. GEORGIA: Cherokee, McDuffie, Richmond. TENNESSEE: Benton, Carroll, Chester, Davidson, Dickson, Hardeman, Hickman, Humphreys, McNairy, Rutherford. ALABAMA: Colbert. MISSISSIPPI: Benton, Han- cock, Harrison, Lowndes, Oktibbeha, Webster. LOUISIANA: Bossier, Caddo, Caldwell, DeSoto, Grant, Natchitoches, Ouachita, Rapides, Sabine, St. Tammany, Winn. ARKANSAS:Baxter, Benton, Clark, Colum- bia, Conway, Crawford, Craighead, Dallas, Drew, Faulkner, Franklin, Garland, Greene, Hempstead, Hot Springs, Independence, Izard, Logan, Marion, Miller, Newton, Ouachita, Poinsett, Prairie, Pulaski, St. Fran- cis, Saline, Sebastian, Sevier, Stone, Washington, White, Yell. 4. L. minor L. The type-species of the genus. The only other named by Linnaeus, L. major, has been shown to be a synonym of Helianthemum canadense (L.) Michx. As was pointed out by Britton (1894), Linnaeus had com- pounded under L. minor at least two other species, L. villosa and L. maritima, but Britton's typification of the species, although at variance with previous interpretations, has been universally accepted. The species presents few problems in identification as it is easily distinguished by its long ex- ternal sepals and the comparatively broad and often ap- parently whorled leaves beneath the inflorescence branches. RANGE: This species extends north of our area to southern New Hampshire and from there west across southern On- tario to the area of Illinois about Chicago. Specimens examined from the following counties: NORTH CAROLINA: Bertie, Buncombe, Carteret, Cumberland, Duplin, Edgecombe, For- syth, Gates, Greene, Harnett, Henderson, Hertford, Hoke, Robeson, Scotland, Swain, Transylvania, Wake, Washington. SOUTH CAROLINA: Aiken, Allendale, Anderson, Beaufort, Berkeley, Charleston, Clarendon, Colleton, Dillon, Dorchester, Georgetown, Hampton, Horry, Jasper, Kershaw, Lee, Lexington, Sumter, Williamsburg. GEORGIA: Brantley, Calhoun, Decatur, Early, McDuffie, Seminole, Richmond. FLORIDA: Alachua, Bay, Bradford, Dix:e, Duval, Gulf, Hernando, Jackson, Jef- ferson, Leon, Madison, Marion, St. Johns, Wakulla. TENNESSEE: Coffee, Hamblen, Knox, Warren. ALABAMA: Baldwin, Mobile, Washington. MISSISSIPPI: Hancock, Harrison, Jackson. LOUISIANA: Calcasieu, Washington. 1961] Wilbur and Daoud — Lechea 113 5. L. patula Legg. A species of such variability that Small (1933) recognized three species. Hodgdon, after a careful appraisal of the suggested differences, could accept but one. The three species were placed by Small in two dif- ferent informally named groups. The group *Longisepalae" contained L. patula and L. prismatica Small while his L. exserta was assigned to the group ''Brevisepalae". Char- acteristic features of L. patula are its narrowly linear cau- line leaves, its sepals that usually either nearly equal one another or the outer ones somewhat the longer, its conspicu- ously exserted capsule capped by the striking reddish-brown fimbriate stigmas and very loosely enveloped about its basal half (to two-thirds) by the calyx. RANGE: Endemic to the southeastern Coastal Plain. Specimens examined from the following counties: NORTH CAROLINA: Bladen, Brunswick, Cumberland, Harnett, Hoke, Moore, Pender, Rich- mond, Scotland. SOUTH CAROLINA: Aiken, Allendale, Bamberg, Cal- houn, Chesterfield, Darlington, Jasper, Kershaw, Marion, Marlboro, Orangeburg, Richland, Williamsburg. GEORGIA: Baker, Ben Hill, Brantley, Clay, Decatur, Dougherty, Lee, McDuffie, Miller, Randolph, Richmond, Sumter. FLORIDA: Alachua, Bay, Bradford, Brevard, Brow- ard, Calhoun, Collier, Columbia, Dade, Dixie, DeSoto, Duval, Escambia, Flagler, Franklin, Gadsden, Gilchrist, Hernando, Highlands, Jeffer- son, Lake, Liberty, Lee, Leon, Levy, Madison, Manatee, Marion, Orange, Pasco, Pinellas, Polk, Putnam, Sarasota, Seminole, Suwannee, Sumter, Volusia, Walton. ALABAMA: Mobile. MISSISSIPPI: Harrison, Jackson. 6. L. Deckertii Small (including L. myriophylla Small). An unmistakable species distinguished by its suffruticose habit, more or less globose fruit with thin papery walls tardily dehiscent at maturity, and its exceedingly short external sepals and often strongly divergent inner sepals. Small (1933) distinguished two species by means of the following key to his group *Myriophyllae". Leaf-blades subulate to elliptic-subulate: capsule depressed-globose. cutem Ma M MM tends piaaiassc tangas p 13. L. Deckertii. Leaf-blades elliptic, sometimes narrowly so: capsules globose......... Re eRe ee Seer OE och cna ROREM 14. L. myriophylla. Hodgdon concluded that these were merely growth-dif- ferences. He found that there did seem to be a difference in calyx pubescence between the two populations described 114 Rhodora [Vol. 63 as separate species by Small with L. myriophylla having a nearly glabrous calyx. He concluded however that these largely vegetative variants could be included within the bounds of one species. We have seen considerably more material than was available to Hodgdon and our conclusion is no different than his. RANGE: Known only from the Coastal Plain of southern Georgia and throughout much of Florida. Specimens examined from the following counties: GEORGIA: Ben Hill, Coffee, Pierce. FLORIDA: Broward, Clay, Collier, Dade, DeSoto, Frank- lin, Highlands, Levy, Marion, Orange, Palm Beach, Putnam, Wakulla. 7. L. racemulosa Michx. The Appalachian distribution pattern of this species has been modified by a migration onto the Coastal Plain from upper North Carolina to Long Island (but scarcely at all further south) and by its sporadic Mars 7-10. Distribution of Lechea in the southeastern United States. Map 7. Lechea racemulosa; map 8. L. Leggettii; map 9. L. Torreyi; map 10. L. cernua. 1961] Wilbur and Daoud — Lechea 115 occurrence in parts of the glaciated Midwest. It is distin- guished by its relatively long, slender pedicels and its con- ical, hardened shiny calyx-base. It apparently rarely hybridizes with L. villosa under which name the inter- mediates are discussed. RANGE: Extending north of our area to Long Island and southern Ohio and Indiana. Reappearing in Indiana on the shore of Lake Michigan and also in eastern Missouri. Specimens examined from the following counties: NORTH CAROLINA: Alamance, Alexander, Alleghany, Anson, Bertie, Buncombe, Burke, Cabarrus, Caldwell, Caswell, Catawba, Chatham, Cherokee, Chowan, Clay, Cumberland, Davidson, Davie, Durham, Edgecombe, Forsyth, Franklin, Gates, Graham, Granville, Greene, Guilford, Halifax, Har- nett, Haywood, Henderson, Hertford, Iredell, Jackson, Johnston, Lee, Lincoln, Macon, Martin, McDowell, Mecklenburg, Mitchell, Montgom- ery, Moore, Nash, Northampton, Orange, Person, Pitt, Polk, Randolph, Richmond, Rockingham, Rowan, Rutherford, Sampson, Stanly, Stokes, Surry, Swain, Transylvania, Union, Wake, Warren, Watauga, Wilson, Yadkin, Yancey. SOUTH CAROLINA: Anderson, Cherokee, Darlington, Edgefield, Greenville, Greenwood, Laurens, McCormick, Oconee, Pick- ens, Saluda, Spartanburg, York. GEORGIA: Clarke, DeKalb, Fanin, Habersham, Paulding, Rabun, Union. TENNESSEE: Bledsoe, Blount, Cocke, Cumberland, Fentress, Franklin, Grainger, Hamilton, Hawkins, Knox, Monroe, Polk, Rhea, Roane, Scott, Sevier, Van Buren, Wayne, White. ALABAMA: Cullman, De Kalb. 8. L. Leggettii Britt. & Holl. Hodgdon treats the variant occurring in the South as var. ramosissima Hodgd. His two other varieties are found in the Northeast and Midwest. For the most part L. Leggettii is restricted to the Coastal Plain but surprisingly one collection from central Tennessee seemed to belong here and adds to the list of Coastal Plain plants found as disjuncts in Tennessee or Kentucky. The most striking vegetative characteristic of this species is the hardened, conical apex of the leaf. The short external sepals, the 3-nerved inner sepals and the subglobose to pyri- form fruiting calyx are a combination of characters which render this species rather easily identified. RANGE: var. ramossissima Hodgd., which alone occurs in the Southeast ranges beyond that region only into south- eastern Virginia. 116 Rhodora [Vol. 63 Specimens examined from the following counties: NORTH CAROLINA: Bertie, Beaufort, Bladen, Brunswick, Carteret, Chatham, Chowan, Columbus, Craven, Cumberland, Duplin, Edgecombe, Gates, Greene, Harnett, Hoke, Jones, Lee, Martin, Nash, Northampton, Onslow, Pam- lico, Pender, Pitt, Robeson, Sampson, Scotland, Washington, Wilson. SOUTH CAROLINA: Allendale, Beaufort, Chesterfield, Darlington, Dillon, Dorchester, Florence, Georgetown, Hampton, Jasper, Lee, Lexington, Orangeburg, Sumter, Williamsburg. GEORGIA: Brantley, Chatham, Dougherty, Liberty, McDuffie, McIntosh, Worth. FLORIDA: Baker, Bay, Bradford, Clay, Columbia, Dixie, Hamilton, Suwanee, Union, Wakulla. TENNESSEE: Coffee. ALABAMA: Baldwin. MISSISSIPPI: Hancock, Har- rison, Jackson, Wayne. LOUISIANA: Beauregard, Jefferson Davis, St. Tammany. 9. L. Torreyi Legg. ex Britt. It seems probable that both Britton (1894) and Small (1903, 1933), in attributing to L. maritima Legg. ex Britt. a range as far south as Georgia, actually mistook some specimens of L. Torreyi for it. L. maritima is not known south of Virginia. Leggett (18783) pointed out that there are two distinguishable elements within the populations now covered by L. Torreyi. The dif- ferences were largely overlooked until Hodgdon (1938) treated the two as varieties distinguished by him more or less as follows: Var. Torreyi: Seeds 4-6; calyx dark brown to slightly ferruginous and densely pilose with cinereous pubescence; fruit somewhat loosely arranged or scattered. Var. congesta: Seeds 3; calyx usually markedly ferruginous and pubescence not conspicuously cinereous; fruit densely clus- tered. The typical element was thought by Hodgdon to be re- stricted to Florida where var. congesta also was abundantly represented but which in addition ranged as far north as southeastern North Carolina and west into Mississippi. There is neither geographic segregation of the two types within peninsular Florida nor from the scanty information available is there any ecologic separation. The morphologi- cal differences are certainly not striking or clear-cut and, with the exception of the number of seeds in each capsule, are rather subjective in nature. Our study has not been exhaustive but we chose not to recognize the two varieties. More material and habitat information are required before 1961] Wilbur and Daoud — Lechea ALI a proper evaluation of them can be made. We would like to reemphasize that Hodgdon reports the reappearance of this species (the 3-seed variant) about 800 miles across the Gulf of Mexico in the pine forests of British Honduras. A far more conspicuous variant has come to our attention and is represented by two sheets in the Buswell Herbarium from the “Sandy scrub above Fort Lauderdale." It is com- pletely glabrous (leaves, stems and calyx) and of a dark reddish brown. There is but one vein in the inner sepal and that is strongly elevated — so much so as to form almost a keel. This vein extends to the very apex of the sepal. We have seen nothing remotely approaching these specimens but feel it best not to recognize them formally until more is known about the population to which they belong. RANGE: The range of this species is shown by the accom- panying map except for the reappearance of the species in British Honduras. Specimens examined from the following counties: NORTH CAROLINA: Brunswick, Pender. SOUTH CAROLINA: Beaufort, Georgetown. GEORGIA: Charlton, Coffee, Long. FLORIDA: Alachua, Baker, Bradford, Brevard, Broward, Charlotte, Citrus, Collier, Dade, Dixie, Duval, Franklin, Her- nando, Highlands, Lake, Lee, Manatee, Martin, Orange, Palm Beach, Pinellas, Polk, Putnam, St. Johns, Sarasota, Seminole, Suwannee, Vol- usia, Wakulla. ALABAMA: Baldwin. MISSISSIPPI: Jackson. 10. L. cernua Small. The species is so unlike any other within the genus that it is strange that it was overlooked until Small described it in 1924. The flowers and fruits are borne in clusters of 2-4 on stiff, often reflexed pedicels and the leaves are at least minutely pubescent upon both sur- faces. The fruits are 1-2-seeded with thickened valves. The species is as distinctive as any within the genus including L. san-sabeana (Buckl.) Hodgd., which alone has been pro- vided with a separate formal category by students of the genus. RANGE: Endemic to peninsular Florida. Specimens examined from the following counties in Florida: Brow- ard, Collier, Hernando, Highlands, Hillsborough, Indian River, Martin, Pinellas, Polk. — DEPARTMENT OF BOTANY, DUKE UNIVERSITY. 118 Rhodora [Vol. 63 LITERATURE CITED Britton, N. L. 1894. A Revision of the Genus Lechea. Bull. Torrey Club 21: 244-253. HopaGpoN, A. R. 1938. A Taxonomic Study of Lechea. Rhodora 40: 29-69, 87-131. (Reprinted as Contrib. Gray Herb. 121.) LEGGETT, W. H. 1878. Rafinesque's Lechea. Bull. Torrey Club 6: 246-252. RAFINESQUE, C. S. 1836. Monograph of Lechea. New Flora and Bot. of N. Am. p. 89-98. SMALL, J. K. 1903. Lechea i» Flora Southeastern United States. pp. 798 + 799. 1933. Lechea in Manual of the Southeastern Flora. pp. 881-884. MENTZELIA ALBESCENS AND LONICERA XYLOSTEUM IN MIS- SOURI. — While botanizing near Joplin, Missouri, recently I came upon a plant, obviously belonging to the family Loasa- ceae, which I did not recognize; and I was unable to find it in either the eighth edition of Gray's Manual or in Britton and Brown's Illustrated Flora. It proved to be Mentzelia albescens (Gill.) Griseb., pre- viously known only from much farther west, ranging from New Mexico and Texas through Mexico and to Argentina and Chile. The plant was described by Gray as Mentzelia Wrightii in Plantae Fendlerianae Novi-Mexicanae, Mem. Amer. Acad. 4: 48. 1849; and there are several other syno- nyms. The determination was confirmed by Dr. Robert E. Woodson, curator of the herbarium, Missouri Botanical Gar- den. Where first found, the plant was growing on mine dumps, perhaps at least fifty or sixty years old, on the north side of Turkey Creek about 215 miles n. w. of Joplin, Missouri. The dumps consist of crumbling limestone and dolomite with fragments and boulders of chert and limestone; and prac- tically no other vegetation was growing on them. The plants here were rather abundant, 3-5 dm. high, nearly simple or a little branched above. My collection data is No. 69221, Aug. 27, 1960. A few days later I again found the plant growing on chat piles of old mines about two miles up Turkey Creek and about a half mile n. w. of Joplin. The record here is No. 69227, along Loan Elm Road, Aug. 30, 1960. 1961] Palmer — Mentzelia and Lonicera 119 Then on Sept. 9th, 1960, while collecting near Carl Junc- tion, Jasper County, Missouri, about ten miles n. w. of the Joplin localities, I again found the plant growing on chat piles of abandoned mines, about a mile and a half from the town. The plants here were more abundant and averaged larger in size than at the Joplin localities ; some of them were up to 7-8 dm. high and more widely branched above the middle. The collection No. is 69263. The chat piles, on which Mentzelia albescens was growing, consist of crushed chert with a smaller admixture of lime- stone or dolomite or both from which the lead and zinc ore has been extracted. Because of its loose dry nature and the presence of alkalies and sulphides, the surface, if undis- turbed, remains sterile for many years, and no plant life can find lodgment on it. However, it seems that this south- western desert plant has found it to be a congenial habitat, for it is becoming abundant and is well established at all of the localities given above. Specimens of all the collections mentioned are deposited in the Ernest J. Palmer private herbarium, Webb City, Missouri. Duplicates of some of them will be sent to the Gray Herbarium, the herbarium of the Missouri Botanical Garden, and to other herbaria. Lonicera Xylosteum L. is another recent collection that appears to be new to the Missouri flora, though not to that of the Manual Range. A large plant was found growing in open upland woods, along a bank of an old electric railway grade, about half a mile north of Joplin, Missouri. It was collected under my No. 69188, ^ug. 6, 1960. The range given for this introduced species in the eighth edition of Gray's Manual is N. E. to Mich., s to N. J., Penn. and O. ERNEST J. PALMER, WEBB CITY, MISSOURI. A NEW VARIETY OF RUDBECKIA FULG:DA. — Rudbeckia fulgida Ait. var. auriculata var. nov. Folia radicalia elliptica, lamina 15-25 cm. longa, 5-8 cm. lata, acuta, basi attenuata; folia caulina acuta, integra vel grosse serrata; folia caulina inferiora sessilia, pandurata vel spatulata, basi angustata, auriculata; folia caulina mediocria sessilia pandurata, basi lata, auriculata; folia caulina superiora sessilia ovata vel ovato-lanceolata, auriculata vel truncata. 120 Rhodora [Vol. 63 Basal leaves elliptical, blade 15 to 25 cm. long, 5-8 cm. wide, acute, the base attenuate into a petiole that is 12 as long to as long as the blade; cauline leaves acute, coarsely and irregularly serrate; lower cauline leaves sessile, pan- durate with a narrow base or spatulate, strongly auriculate, those nearest the base (nodes 1-3) occasionally elliptic- spatulate but sessile and auriculate; middle cauline leaves sessile, pandurate with a broad base, strongly auriculate ; upper cauline leaves and bracts subtending the branches sessile, ovate to ovate-lanceolate, auriculate or truncate. TYPE: Moist soil along Alabama Highway 55, 11 miles south of McKenzie (2 miles north of Red Level), Covington Co., Alabama, R. E. Perdue, Jr. 2177, July 24, 1958. Type in the Gray Herbarium, isotype in the U. S. National Herbar- ium. This variety is very distinct from the others of R. fulgida. In my key to the varieties of this species', the plant de- scribed here keys out to either R. fulgida var. fulgida or R. fulgida var. spathulata (Michx.) Perdue. Varicty auri- culata is readily distinguished from each of these as well as from its other relatives by the typically auriculate leaves of the lower and middle parts of the stem. I first collected the new variety in very immature condi- tion during the summer of 1952. Not until 1958 was I able to return to the original locality and obtain additional speci- mens to provide a basis for the description of this new variety. My original collection was taken from a small colony extending about 25 feet along a ditehbank and con- sisting of about 25 or 30 individual plants. When I made the second collection, 6 years later, the colony had expanded to several hundred plants extending along the ditchbank for some 300 feet. On two separate occasions I looked extensively elsewhere in the vicinity of the type locality without finding other plants of this variety. — ROBERT E. PERDUE, JR., U. S. DEPARTMENT OF AGRICULTURE, AGRICULTURAL RESEARCH SERV- ICE, CROPS RESEARCH DIVISION, BELTSVILLE, MARYLAND. !Synopsis of Rudbeckia subgenus Rudbeckia. RHODORA 59 (708): 293-299 (1957). Volume 63, No. 747, including pages 61-90, was issued March 29, 1961. ! RKFERENCE LIBRARY JUN 2 6 1961 Dodota JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by i REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Associate Editors Vol. 63 May, 1961 No. 749 CONTENTS: Chromosome Numbers in the Compositae. V. Mexican and Guatemalan Species. B. L. Turner, J. H. Beaman and Tae ye PL O CIC A E AEE AOA AOOO OE E A E ET 121 Notes on the Grass Flora of the Chicago Region. Il. NI FEMGlassmaon 5s E UU IE NI esie ede e tS TR S 130 Some Notes on Myriophylium spicatum. Áskell Lóve ............ 139 Bidens connata var. gracilipes Fern. in Western Michigan. Bart Raward Sherf Pa.: Else eo NN TEES NM erre dee 145 New Plant Records from Kansas. Ronald L. McGregor, Vernon L. Harms and Jerold D. Poindexter ...................... 146 Eragrostis curvula from Illinois. Glen S. Winterringer .......... 148 Notes on the Flora of Penikese Island, Massachusetts. JO YEA ES UTE ende, CR CURIE One Sco US ERN GONE 149 Che Nem England Botanical Club, Inc. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume. Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able. All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass. Printed by THE LEXINGTON PRESS, INC. Lexington, Mass. Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 May, 1961 No. 749 CHROMOSOME NUMBERS IN THE COMPOSITAE. V. MEXICAN AND GUATEMALAN SPECIES! B. L. TURNER, J. H. BEAMAN AND H. F. L. ROCK The present study is based on bud material collected by Beaman and Rock during the summer of 1959 incidental to their separate studies on certain elements of the flora of Mexico and Guatemala. Chromosome counts were made by Turner from pollen mother cell squashes as outlined by Turner and Ellison (1960). The voucher specimens collected by Beaman (Table 1) are deposited in the Michigan State University Herbar- ium ; those collected by Rock are deposited in the University of Texas and Vanderbilt University Herbaria. The taxo- nomic identifications are those of the authors, except where otherwise noted. Table 1 should be consulted for a complete tabulation of the species studied, as some are not included in the discussion. DISCUSSION EUPATORIEAE — Eupatorium glabratum (n=17), E. sco- rodonioides (n=17). These counts are consistent with those previously reported for taxa of the section Eximbricata (Turner, Ellison and King, 1961). Eupatorium pazcuarense (n—25). Both apomictie and sexual species are known for this genus; the present meiotic figures were normal with 25 bivalents. Chromosomally the species is related to those taxa on a base of x=10. E. prunel- laefolium (n=50 univalents) is apparently apomictic, the meiotic chromosomes showing complete absence of pairing. 1This study was supported "by National Science Foundation Grants G-9025 and G-9045. 121 122 Rhodora [Vol. 63 Ageratum corymbosum (n=20). This species was also reported as a tetraploid by Turner, Ellison & King (1961). Stevia (n—34 univalents) . The count agrees with those obtained for several other Mexican species (Turner, unpub- lished), except that the latter have been completely regular at meiosis, showing 17 bivalents. ASTEREAE — Astranthium guatemalense (n=9). The present count was also determined by Beaman and Stouta- mire (unpublished) from the same bud collections. A. gua- temalense is very restricted, occurring, so far as known, only in the Sierra de los Cuchumatanes. Astranthium xanthocomoides (n—8). As indicated below, some species of Astranthium show a polyploid series on a base of z—4. This series apparently extends from the diploid to the hexaploid level. In this connection it is interesting to note that A. guatemalense (see above) is diploid with n=9; Stoutamire and Beaman (1960) and the present authors have reported n—18 for A. mexicanum. Further study, both morphological and cytological, is needed to determine if the apparent chromosomal base of x—9 for the latter two spe- cies has any phyletic significance. This information would seem particularly significant in view of the controversy con- cerning the probable ancestral basic number for the tribe Astereae (Turner, Ellison and King, 1961). Astranthium sp. (n—12). This material probably repre- sents an undescribed species. It is apparently hexaploid on a base of x—4, since the lowest diploid number reported for the genus is n=4 for A. integrifolium (Baldwin, 1941; Bea- man, unpublished). Erigeron (x—9). The chromosome numbers for the 4 spe- cies listed in Table 1 are consistent with the previous base numbers reported for other species in the genus. E. pubes- cens (n—36 univalents) is apparently apomictic. E. scapo- sus (n—18) is tetraploid showing 18 bivalents at meiosis I. Figs. 1-25. Camera lucida drawings of meiotic chromosomes, all approximately ; 1300. Fig. 1, Eupatorium glabratum (n=17). Fig. 2. E, prunellaefolium | (n—50 univalents). Fig. 3. Stevia sp. (»—34 univalents). Fig. 4. Astranthium wanthoco- moides (n—8). Fig 5. Astranthium sp. (n—12). Fig. 6. Erigeron pubescens (n= 36 univalents). Fig. 7. Erigeron sp. (n—36). Fig. 8. Grindelia oxylepis (n=6). Fig. 9. Machaeranthera tanacetifolia (n=4). Fig. 10. Bidens angustissima (n—10). 1961] Turner, Beaman and Rock—Chromosome Numbers 123 Cl 5 . e eU gà. e «0 N "2 ru o Shee. Carol A q AP QV CEN p ry € o? XS « 10 PEDI oy 0% " o4 Se AAR we 8 Nua t ^ 4 " NN | Y ae 12 l4 T Se G2 a ^s c) X [ON X 3 us” Ay av 4 v A, f Oy XA are ON. o>,” 2 e a «mà il WF I ev as . x $ AS "La ace Fig. 11. B. anthemoides (n—24). Fig. 12. Dugesia mexicana (n—18). Fig. 13. Het- erospermum pinnatum (n—25). Fig. 14. Verbesina hypomalaca (n—16). Fig. 15. V. serrata (n—17). Fig. 16. Zaluzania coulteri (n—16). Fig. 17. Zinnia angustifolia (n—11). Fig. 18. Bahia xylopoda (n—11). Fig. 19. Dyssodia pinnata (n—13). Fig. 20. Hymenoxys insignis (n—15). Fig. 21. H. odorata (n=15). Fig. 22. Nicolletia edwardsii (n=10). Fig. 23. Porophyllum amplexicaule (n=12). Fig. 24. Senecio toluccanus (n—20). Fig. 25. Senecio cf. cyclophyllus (n—23). 124 Rhodora [Vol. 63 Erigeron sp. (n—9). This collection, Beaman 2698, is apparently an undescribed species. Erigeron sp. (Beaman 2693, n=36) is likewise an undescribed species and is octo- ploid, there being 36 bivalents at Meiosis I. Chromosome numbers for the genera Gutierrezia (n=4) and Grindelia (n=6) agree with the basic numbers estab- lished for these genera by other workers. Determination of a voucher specimen of Gutierrezia glutinosa was made by Dr. O. Solbrig. The count for Chaetopappa is a first report. Machaeranthera gymnocephala (n—4). Jackson (1959) reported this species to be diploid with n—5; however, his New Mexican collection is apparently referable to M. ble- phariphylla (Gray) Shinners since Cronquist and Keck (1957) would recognize the latter as a valid taxon. It is a distinct perennial of the southwestern United States and adjacent Mexico. M. gymnocephala is a biennial (or an- nual?) or weak perennial of more southern distribution. If Jackson’s count applies to M. blephariphylla (n=5) the pres- ent chromosome count for M. gymnocephala (n=4) lends support to the recognition of it as a separate species. HELIANTHEAE — Bidens angustissima var. linifolia (n= 10). This is the lowest chromosome number reported for the genus to date. Previous counts have all been on a base of x—12 or 11 (Turner, Ellison and King, 1961). Verbesina hypomalaca (n—16). Turner, Ellison and King (1961) have reported counts of n=17 and 18 for this genus. Apparently Verbesina (sens. lat.) is multibasic with x=18, 17, 16. Zaluzania coulteri (n—16). The species identification is tentative. Previous chromosome counts for the genus have been on a base of z—18 (Turner and Johnston, 1961). Chromosome counts for Cosmos, Perymenium, Sanvitalia, Viguiera and Zinnia are consistent with basic numbers al- ready established for these genera (Turner, Ellison and King, 1961). Chromosome counts for the genera Dugesia (n—18) and Heterospermum (n==25) have not been previously reported. HELENIEAE — Dyssodia pinnata (n=13). This count agrees with a number of unpublished counts for the species 1961] Turner, Beaman and Rock—Chromosome Numbers 125 (Johnston and Turner, unpublished). Hymenoxys odorata (n—15). Chromosome counts of n —]11 have been reported by previous workers for H. odorata (Speece and Baldwin, 1952; Raven, unpublished) ; identifi- cation of the material from which the present count was made was verified by Dr. K. Parker. Nicolletia edwardsii (n—10). Raven & Kyhos (unpub- lished) have also found counts of n=10 for the genus. Chromosome counts for the genera Bahia, Baileya, Poro- phyllum, Psilostrophe and Tagetes are consistent with the basic numbers already established for these genera (Dar- lington and Wylie, 1956; Towner, 1958; Turner, Ellison & King, 1961). ANTHEMIDEAE — Achillea lanulosa (n—18). The species, as represented in Mexico, is similar morphologically to other collections from North America and has the same chromo- some number. SENECIONEAE — The chromosome counts for Senecio to- luccanus (n—20) and S. sanguisorbae (n—ca. 30) are con- sistent with counts reported for other species of the genus. Senecio cf. cyclophyllus (n—23). Darlington and Wylie (1956) list one other species, S. resedifolius, from Siberia on a base of z—23. Both species belong to the section Aurei (Greenman, 1903, 1907). SUMMARY Chromosome counts are reported for 47 taxa of Mexican Compositae. These include first reports for 34 species, some of which belong to previously unreported genera (Chaetopappa, m-—9; Dugesia, n—18; Heterospermum, n=25; Nicolletia, n—10; and Stevia, n=17). Astranthium, as presently understood, has been found to have spe- cies with n—4, 8, 9, 12 and 18. Machaeranthera gymnocephala was found to be diploid with ».—4, instead of » —5 as reported by a differ- ent investigator. Exceptional counts for the genera Bidens (m-—10), Melampodium (n-—11) and Verbesina (n-—16) are also recorded. Hymenoxys odorata, reported as n» —11 or 2n—22 by previous work- ers, was found to have Mexican populations with n=15. A count of 7.—23 for a Mexican species of Senecio was also obtained; the only other counts of n=23 for this genus have been from a Siberian species. — PLANT RESEARCH INSTITUTE, THE UNIVERSITY OF TEXAS, AUSTIN; BOTANY DEPARTMENT, MICHIGAN STATE UNIVERSITY, EAST LANSING; BIOLOGY DEPARTMENT, VANDERBILT UNIVERSITY, NASHVILLE. 126 Rhodora TABLE 1. Summary of collections studied Species EUPATORIEAE Ageratum corymbosum Zuee. Eupatorium glabratum H. B. K. Eupatorium pazcuarense H. B. K. Eupetorium prunellaefolium H. B. K. Eupatorium scorodonioides Gray Stevia sp. ASTEREAE Astranthium guatemalense Blake Astranthium mexicanum (Gray) Larsen Astranthium purpurascens (Rob.) Larsen Astranthium xanthocomoides (Less.) Larsen Astranthium xanthocomoides (Less.) Larsen [Vol. 63 n Locality chromosome number SAN LUIS POTOSI: La Capilla. 20 (S. W. of San Luis Potosi) Rock 456. HIDALGO: ca. 1 km. north of 17 Real del Monte, ca. 2770 m. (fig. 1) alt. Beaman 2370. GUATEMALA: Dept. of Huehte- 25 tenango; Sierra de los Cuchu- matanes, between kms. 324 and 325 on Ruta Nacicnal 9 N. (between Chemal and San Juan Ixeoy); ca. 3140 m. ale. Beaman 3033. STATE OF MEXICO: at Puerto del 50 Aire on Mexico-Puebla high- all way, 3196 m. alt. Beaman univalents 2903. (fig. 2 SAN LUIS POTOSI: La Capilla. 17 (S. W. of San Luis Potosi) Rock 457. STATE OF MEXICO: Llano Gran- 34 de. Rock 352. all univalents (fig. 3) GUATEMALA: Dept. of Huehue- 9 tenango; Sierra de los Cuchu- matanes, between kms. 324 and 325 on Ruta Nacional 9 N. (between Chemal and San Juan Ixcoy), ca. 3140 m. alt. Beaman 3027. TLAXCALA: Llano Grande. 18 Rock 353. HIDALGO: 6.7. mi. south of 8 Jacala. Rock 310. NUEVO LEON: ca. 26 mi. north- 8 east of Dr. Arroyo on west side of mtn. known locally as Picacho Onofre, ca. 3230 m. alt. Beaman 2697. HIDALGO: ridge ca. 2 kms. 8 south of Real del Monte, ca. (fig. 4) 2880 m. alt. Beaman 2737. 1961] Turner, Beaman and Rock—Chromosome Numbers 127 Astranthium sp. Chaetopappa bellioides Gray Erigeron pubescens H. B. K. Erigeron scaposus DC. Erigeron sp. Erigeron sp. Grindelia oxylepis var. eligulata Steyermark Gutierrezia glutinosa (Schauer) Sch. Bip. Machaeranthera tanacetifolia (H. B. K.) Nees (DC.) Machaeranthera gymnocephala Shinners HELIANTHEAE Bidens angustissima var. linifolia (Sch. Bip. ex Klatt) Sherff (DC.) Sherff Bidens anthemoides Cosmos diversifolius Otto in Knowles & Weste. Dugesia mexicana Gray Heterospermum pinnatum Cav. NUEVO LEON: top of Cerro 12 Potosi, ca. 3650 m. alt. Bea- (fig. 5) man 2661. NUEVO LEON: near Monterrey. 9 Rock 495. HIDALGO: ridge ca. 2 kms. 36 south of Real del Monte, ca. univalents 2880 m. alt. Beaman 2733. (rarely a few bivalents). (fig. 6) HIDALGO: ridge ca. 5 kms. 18 northeast of Pachuca; ca. 2640 m. alt. Beaman 2758. NUEVO LEON: ca, 26 mi. north- 9 east of Dr. Arroyo on west side of mtn. known locally as Picacho Onofre, ca. 3300 m. alt. Beaman 2698. NUEVO LEON: ca. 26 mi, north- 36 east of Dr. Arroyo on west (fig. 7) side of mtn. known locally as Picacho Onofre, ca. 2700 m. alt. Beaman 2693, NUEVO LEON: 41.2 mi. south of 6 Saltillo. Rock 271. (fig. 8) NUEVO LEON: ca. 8 mi. east of 4 Galeana on road to Linares, ca. 1850 m. alt. Beaman 2679. NUEVO LEON: 41.2 mi. south of 4 Saltillo. Rock 263. (fig. 9) SAN LUIS POTOSI: La Capilla. 4 (S. W. of San Luis Potosi). Hock 451. HIDALGO: ridge ca. 5 kms. 10 northeast of Pachuca, ca. 2640 (fig. 10) m. alt. Beaman 2763. VERACRUZ: Cofre de Perote. 24 Rock 391. (fig. 11) HIDALGO: ridge ca. 2 kms. 12 south of Real del Monte, ca 2880 m. alt. Beaman 2741. HIDALGO: ridge ca. 2 kms. 18 south of Real del Monte, ca. (fig. 12) 2770 m. alt. Beaman 2756, SAN LUIS POTOSI: La Capilla 25 (S. W. of San Luis Potosi). (fig. 13) Rock 463. 128 Melampodium montanum Benth. Perymenium mendezii DC. Sanvitalia ocymoides DC. y Sanvitalia procumbens L. Verbesina hypomalaca Rob. & Greenm. Verbesina serrata Cav. Viguiera stenoloba (Gray) Blake Zaluzania coulteri Hemsl. Zinnia angustifolia H. B. K. HELENIEAE Bahia absinthifolia Benth. Bahia xylopoda Greenm. Baileya pleniradiata Harv. & Gray Dyssodia pinnata Rob. Hymenoxys insignis (Gray) Cockerell Hymenoxys odorata DC. Nicolletia edwardsii Gray Porophyllum amplexicaule Engelm, Psilostrophe gnaphaloides DC. Rhodora [Vol. 63 GUATEMALA: Dept. of Huehve- 11 tenango; Sierra de los Cuchu- matanes, between kms. 324 and 325 on Ruta Nacional 9 N. (between Chemal and San Juan Ixeoy), ca. 3140 m. alt. Beaman 3043. HIDALGO: ridge ca. 5 kms. 15 northeast of Pachuca, ca. 2640 m. alt. Beaman 2764. QUERETARO: 4.7 mi. north of 16 Queretaro. Rock 435. HIDALGO: 6.7 mi. south of Jaca- 8 la. Rock 305, HIDALGO: ridge ca. 2 kms. 16 south of Real del Monte, ca. (fig. 14) 2880 m. alt. Beaman 2742. QUEKETARO: 4.7 mi. north of 17 Queretaro. Rock 431. (fig. 15) COAHUILA: Canon de Tvle, ca. 34 Saltillo. Rock 239. HIDALGO: ridge ca. 2 kms. 16 south of Real del Monte, ca. (ñg. 16) 2880 m. alt. Beaman 274: SAN LUIS POTOSI; La Capilla 11 (S. W. of San Luis Potosi). (fig. 17) Rock 449. DURANGO: 43.6 mi. northeast of 12 Durango. Rock 475. HIDALGO: ridge ca. 5 kms. 11 north of Pachvea, ca. 2640 m. alt. Beaman 2762. Paila. (fig. 18) COAHUILA: Rock 482. 16 NUEVO LEON: ca. 8 mi. east of 13 Galeana on road to Linares, (fig. 19) ca. 1850 m. alt. Beaman 2680, NUEVO LEON: top of Cerro 15 Potosi, ca. 3650 m. alt. Bea- (fig. 20) man 2649. NUEVO LEON: 41.2 mi. south of 15 Saltillo. Rock 264. (fig. 21) COAHUILA: Paila. Rock 481, 10 (fig. 22) COAHUILA: Saltillo. Rock 251. 12 (fig. 23) COAHUILA: Saltillo. Rock 252. 16 1961] Turner, Beaman and Rock—Chromosome Numbers 129 Tagetes lucida Cav. HIDALGO: ridge ca. 5 kms. 11 northeast of Pachuca, ca. 2640 m. alt. Beaman 2761. ANTHEMIDEAE Achillea lanulosa Nutt. STATE OF MEXICO: at Puerto del 18 Aire on Mexico-Puebla high- way, 3196 m. alt. Beaman 2901. SENECIONEAE Senecio sanguisorbae DC. NUEVO LEON: top of Cerro ca. 30 Potosi, ca. 3650 m. alt. Bea- man 2638. Senecio toluccanus DC. NUEVO LEON: ca. 26 mi. north- 20 east of Dr. Arroyo on west (fig. 24) side of mtn. known locally as Picacho Onofre, ca. 2700 m. alt. Beaman 2688. Senecio ef. cyclophyllus Greenm. TLAXCALA: Llano Grande. 23 Rock 354. (fig. 25) LITERATURE CITED BALDWIN, J. T., JR. 1941. Cytophyletic analysis of Astranthium in- tegrifolium. Bull. Torrey Club 68: 615-617. CRONQUIST, A. C. & D. D. KECK. 1957. A reconstruction of the genus Machaeranthera. Brittonia 9: 231-239. DARLINGTON, C. D. & A. P. WYLIE. 1956. Chromosome atlas of flower- ing plants. p. 519. Macmillan Co., New York. GREENMAN, J. M. 1903. Monographie der nord- und centralamerika- nischen Arten der Gattung Senecio. Engler Bot. Jahrb. 32: 1-33. .1907. New or noteworthy spermatophytes from Mexico, Central America, and the West Indies. Field Mus. Pub. Bot. 2: 247-287. JACKSON, R. C. 1959. In Documented Chromosome numbers of plants. Madrono 15: 49-52. SPEESE, B. & J. T. BALDWIN, JR. 1952. Chromosomes of Hymenoxys. Amer. Jour. Bot. 39: 685-688. STOUTAMIRE, W. P. & J. H. BEAMAN. 1960. Chromosome studies of Mexican alpine plants. Brittonia 12: 226-230. TOWNER, J. W. 1958. Amphiploidy in the cultivated marigolds (Tagetes). Proc. X Int. Cong. Gen. 2: 296 (Abstract). TURNER, B. L. & W. L. ELLISON. 1960. Chromosome numbers in the Compositae. I. Meiotic chromosome counts for 25 species of Texas Compositae including 6 new generic reports. Texas Jour. Sci. 12: 146-151. . & M. C. JOHNSTON. 1961. Chromosome numbers in the Compositae. III. Certain Mexican species. Brittonia 13: 64-69. , W. L. ELLISON & R. M. KING. 1961. Chromosome num- bers in the Compositae. IV. Mexican species, with phyletic inter- pretations. Amer. Jour. Bot. 48: 216-223. 130 Rhodora [Vol. 63 NOTES ON THE GRASS FLORA OF THE CHICAGO REGION. Il. S. F. GLASSMAN Since the publication of my last article on grasses of the Chicago region (1957), I have uncovered a substantial amount of additional information. This includes new county and regional records, notes on misidentified specimens or unverified records, a list of excluded species, and other per- tinent information. Specimens cited were examined by me at the following herbaria: Chicago Natural History Museum (F), Univer- sity of Illinois, Navy Pier (CHI), University of Illinois, Ur- bana (ILL), Illinois Natural History Survey (ILLS), Butler University (B), Indiana University (1), University of Notre Dame (ND), Purdue University (PUR), and University of Wisccnsin (WIS). I want to thank the curators of these her- baria for the privilege of studying their specimens. Aegilops cylindrica Host. LAKE (Indiana): Hammond, Glassman 4215 (CHI). A native of Europe which has been collected only in railroad yards in the region. First reported from the region (Cook county) by Thieret and Evers (1957). Agropyron smithii Rydb. var. smithii. COOK: Chicago, Thieret 2350, 2351, 2352 (F), A. Johnson 3 (F); Morton Grove, Glassman 3717, 4139, 4319 (CHI); Palos Park, Um- bach 34,68 (WIS). LAKE: (Illinois) : Along Milwaukee R. R., east of Round Lake Beach, Glassman 4117, 4119 (CHI). Not previously recorded for Cook and Lake counties. This spe- cies is native west of the Mississippi River, but is adventive in the Chicago region where it occurs principally along rail- road tracks. The report of A. smithii by Deam et al. (1942) from La Porte county is based on a glaucous specimen (Deam 52396, I) which appears to be A. repens (L.) Beauv. The glaucous character was seen in many specimens of repens as well as smithii. A number of specimens examined are rather atypical or abnormal for A. smithii. In Hill 87/1897 and Bebb 621, 980 1961] Glassman — Grass Flora of Chicago Region 131 from Lake county (Indiana) and Glassman 4302, 4304 from Will county, some spikes have two spikelets at a node instead of one and some of these spikelets are unusually narrow. Otherwise, they fit this species. Some other specimens ap- pear to be hybrids between A. repens and A. smithii. Before discussing them, a key distinguishing the two species is given below : Glumes broadest near the middle, usually distinctly nerved, symmetri- cal, with acute tips or awn-pointed; lemmas awnless, awn-pointed or awned; upper blade surface shallowly grooved, blades 3-12 mm. wide, flat or involute, with soft or sharp pointed tips. ...... A. repens Glumes broadest near the base, usually obscurely nerved, asymmetrical, with acuminate tips which are often awn-pointed; lemmas acuminate, mucronate or short awned; upper blade surface deeply and unevenly grooved, blades 2-5 mm. wide, usually involute, with sharp pointed Jiu cde geo, a iter MS. cer A. smithii. Umbach 3468 from Cook county, Glassman 4227 and Um- bach (no number) from Lake county (Indiana), and Moffatt 142 from Du Page county, have glumes which are mostly strongly nerved. In Umbach (no number), however, they are with acute or mucronate tips, whereas in the others the glumes are broadest near the base and asymmetrical with acuminate tips. In one locality east of Round Lake Beach, a number of specimens were collected which apparently represent a hybrid swarm. Glassman 4118, 4120 and 4121 have most of the characters of A. repens, but many of the glumes are shaped like A. smithii except that they are mostly strongly nerved; and 4119 is characteristic of smithii but most of the glumes are strongly nerved. Glassman 4117 seems to be intermediate between the two species. The glumes are shaped like smithii, but almost all are distinctly nerved; and the blades are flat and up to 9 mm. wide with sharp tips, but some of the surfaces are deeply grooved like smithii while others are shallowly grooved like repens. Agropyron smithii Rydb. var. molle (Scribn. & Smith) Jones. WILL: Gougars, Glassman 4312a, 4313 (CHI). First reported from the region (Cook and Du Page counties) by Thieret and Evers (1957) ; these are new records for Will county. Both specimens are unusual in that some of the spikes have 2-4 spikelets per node instead of one. The re- 132 Rhodora [Vol. 63 ports of A. dasystachyum (Hook.) Vasey from the Chicago region by Mosher (1918), Pepoon (1927), Hitchcock (1935), Fernald (1950), Chase (1951), and Jones and Fuller (1955) are based on Moffat 231 and Umbach (no number) from Du Page county. These specimens belong to A. smithii var. molle. Many of the reports of A. dasystachyum for Illinois state the habitat as "along the shores of Lake Michigan". This species is found in the above habitat in Wisconsin and Michigan, but not in Illinois or Indiana. A. smithii var. molle, on the other hand, occurs only in the vicinity of rail- road tracks in the Chicago region. In the Lake Michigan region, the two taxa can be distinguished as follows: Glumes usually asymmetrical, acuminate tipped, broadest below the middle, obscurely or distinctly nerved; lemmas and usually glumes scabrous to short pilose ...........eeeeee A. smithii var. molle. Glumes usually symmetrical, acute, broadest above the middle, dis- tinetly nerved; lemmas and often glumes more or less densely villous Sissevsussseuesssusseussaesuusussusscususisaucsssusessasscevensasseresssenisesessayeyersene A. dasystachyum. Andropogon scoparius Michx. DU PAGE: Wheaton, Moffatt 3315 (WIS). First record from Du Page county. Andropogon virginicus L. Apparently, a recent arrival in the Chicago region where it is found in sandy soil along roadsides. The first authentic reports of this species from Porter and La Porte counties were by Deam et al. (1951) and (1953). The Cook county reports of Higley and Raddin (1891), Mosher (1918) and Pepoon (1927) are based on a lost Shipman specimen which is probably A. scoparius. Arrhenatherum elatius (L.) Presl. DU PAGE: Wheaton, Glassman 4069 (CHI) ; WILL: Romeo, Umbach (CH1) ; Braid- wood, Vytanovych 4170 (CHI). First records for Du Page and Will counties. Avena fatua L. LAKE (Illinois): Highland Park, Glass- man 4087 (CHI); LAKE (Indiana): Hammond, Glassman 4214 (CHI). A European introduction found mainly along railroad tracks in the region. Both cited specimens are new records for those counties. Brachyelytrum erectum (Schreb.) Beauv. LAKE (Illinois) : Biltmore estates, Steyermark 65948 (F). Also known from Cook and Porter counties. 1961] Glassman — Grass Flora of Chicago Region 133 Bromus commutatus Schrad. cook: Chicago, Glassman 3630, 3637, 4252; Northlake, Glassman 4089; Lincolnwood, Glassman 4093, 4094 (CHI); LAKE (Illinois): Grayslake, Glassman 4110, 4112; Libertyville, Glassman 4116 (CHI) ; DU PAGE: Glen Ellyn, Glassman 4050, 4093 (CHI) ; Wheaton, Moffatt 3356 (Wis); West Chicago, Glassman 4071, 4072 (CHI) ; WILL: Gougars, Glassman 4297, 4307, 4310; Romeo, Umbach 1712 (CHI) ; LA PORTE: Trail Creek, Potzger 10281 (B). Not listed from the Illinois portion of the Chicago region by Jones and Fuller (1955), therefore, all four county records are new. In Indiana, only specimens from Lake county have been seen, hence the La Porte county record is also new. This species resembles B. racemosus L. rather closely and both are often found growing together along railroad tracks. A key differentiating the two taxa in the Chicago region follows: Plants about 50-70 cm. tall, panicles open, 10-20 em. long, main branch- es stiff and ascending, up to 10 em. long .................... B. commutatus. Plants usually 20-30 cm. tall, panicles somewhat contracted, 5-9 cm. long; main branchessup:to-4 cm. long 9-5 m n B. racemosus. Bromus inermis Leyss. DU PAGE: Glen Ellyn, Glassman 4046 (CHI); WILL: Gougars, Glassman 4811 (CHI). First records for Du Page and Will counties. Bromus japonicus Thunb. Cook: Saganashkee Slough, Glassman 4287, 4289 (CHI). Not previously reported from the Illinois part of the Chicago region. This species closely resembles B. commutatus, but can be distinguished from that taxon by the very wavy panicle branches which are spreading or drooping. Bromus latiglumis (Shear) Hitch. LA PORTE: Along Kan- kakee River, C. Ek (B). First record for that county. Bromus racemosus L. COOK : Chicago, Drouet 12736a; Lin- colnwood, Glassman 4093 (CHI) ; LAKE (Illinois) : South of Grayslake, Glassman 4110 (CHI); DU PAGE: Villa Park, Glassman 4082 (CHI). First specimens I have seen for the Chicago region. No specimens were found to verify the re- ports of Pepoon (1927) or Jones and Fuller (1955) for Cook County. Bromus secalinus L. wILL: Four miles east of Steger, 134 Rhodora [Vol. 63 Drouet and Louderback 13524 (CHI). Not previously re- ported from Will county. Bromus squarrosus L. PORTER: Dunes State Park, Drouet 12715 (CHI). First reported from the region (Lake county, Indiana) by Steyermark and Swink (1955). Diplachne fascicularis (Lam.) Beauv. Jones (1950) and Jones and Fuller (1955) list this taxon as Leptochloa fasci- cularis (Lam.) Gray; however, Fernald (1950) splits off Diplachne from Leptochloa and assigns each genus to sep- arate tribes. Only known in ihe Chicago region from three specimens (Moffatt 485, 554 and Moffatt, no number) from Chicago. One of these specimens, Moffatt (no number), is the basis for the report of Triplasis purpurea (Walt.) Chapm. by Pepoon (1927). Echinochloa walteri (Pursh) Heller. No specimens have been seen for the Lake county (Illinois) report of Jones and Fuller (1955). A specimen collected from Algonquin by Nason (ILL) in 1879, however, was mistakenly placed in Lake county instead of McHenry county where it belongs. Eleusine indica (L.) Gaertn. WILL: Braidwood, Evers 34800 (ILLS) ; LAKE (Indiana): Gary, Thieret 1417 (CHI). First records for these two counties. Elymus villosus Muhl. f. villosus. LAKE (Illinois): Bilt- more Estates, Steyermark 65947 (F); WILL: Wheatland, Umbach 4016, 4032, 7926 (WIS). Not previously reported from Lake and Will counties. Otherwise known from Cook and Porter counties in the region. Elymus villosus f. arkansanus (Scribn. and Ball) Fern. PORTER: Five miles west of Michigan City, Kriebel 10310 (PUR). First reported from the region (Du Page county) by Glassman (1957). Eragrostis capillaris (L.) Nees. LAKE (Indiana) : Gary, Thieret 1509 (CHI). The first specimen I have seen for the Indiana part of the Chicago region. Listed by Peattie (1930) from the Indiana Dunes but no specimen was cited. The Du Page county report of Pepoon (1927) from Wheaton, based on Moffatt 24,13 (CHI), is actually E. frankii C. A. Meyer. Eragrostis poaeoides Beauv. COOK: Chicago, Glassman 1961] Glassman — Grass Flora of Chicago Region 135 4240; Burnham, Glassman 4194; Northlake, Glassman 4086 (CHI) ; DU PAGE: West Chicago, Glassman 4076 (CHI) ; WILL: Gougars, Glassman 3837 (CHI); LAKE (Indiana): Gary, Glassman 3922; Hammond, Glassman 4108 (CHI). First reported from the Indiana part of the region by Glassman (1957) from Porter county; the Lake county records are new. No specimens have been found to support the Cook and Du Page county listings of Jones and Fuller (1955) ; these are the first plants seen by me for the Illinois portion of the region. Festuca rubra L. cooK: Chicago, Glassman 3637a; Skokie, Glassman 5300 (CHI) ; LAKE (Illinois) : Grayslake, Glassman 4111; Wauconda, Glassman 4128 (CHI). Probably intro- duced into the region as a pasture or lawn grass. At present, found along roadsides and in vacant lots. Originally reported from the region by Deam et al. (1947) from Porter county ; the Cook and Lake county records are new. Glyceria grandis Wats. ex Gray. LAKE (Illinois) : Four miles N.E. of Volo, Glassman 4129 (CHI) ; PORTER: Beverly Shores, Glassman 3875 (CHI). First authentic specimens seen by me for the Chicago region. No specimens were found to verify the listing of Pepoon (1927). Leptoloma cognatum (Schult.) Chase. WILL: Custer Park, Steyermark 64831, Swink 2489 (F). First record for Will county. Muhlenbergia mexicana (L.) Trin. f. ambigua (Torr.) Fern. LAKE (Illinois): Lake Villa, Evers 41719 (ILLS). First record for the Chicago region. Differs from the species in having long awned (4-10 mm.) lemmas. Panicum clandestinum L. WILL: Custer Park, Swink 2433 (F). First record for Will county. Otherwise known from Cook and Porter counties. Panicum dichotomiflorum Michx. var. puritanorum Sven- son. COOK: Morton Grove, Glassman 5591 (CHI). This is the first authentic record for the Chicago region. Chase (1951) listed this variety from Indiana which was probably based on a Jasper county specimen cited by Deam (1940). The following key differentiates the species from its variety: 136 Rhodora [Vol. 63 Culms rather robust, 50-100 em. long; spikelets acute, averaging 2.5 mm. long; plants of dry soil ................... sees P. dichotomiflorum. Clums rather slender, usually less than 50 cm. long; spikelets less pointed, averaging 2.0 mm. long; plants of damp or wet soil var. puritanorum. Pe eee eee ee eee ee Peer Creer re cee rere rere reer rere ere terre errr eter ss] Poa palustris L. LA PORTE: Eight miles S.W. of La Porte, Kriebel 8248 (PUR). Not previously reported from this county. Setaria glauca (L.) Beauv. WILL: Monee, Evers 21548 (ILLS) ; Plainfield, D. Chapp 1 (CHI). First records for Will county. Now reported from all counties in the Chicago re- gion. Deam (1929), (1940), Deam et al. (1946), Chase (1951), and Jones and Fuller (1955) call this plant S. lutes- cens (Weigel) Hubb., but Fernald (1950), Reeder (1951) and Gleason (1952) maintain that the valid name is S. glauca. Sporobolus asper (Michx.) Kunth. DU PAGE: Glen Ellyn. Swink (F). The first specimen I have seen for this county. Pepoon (1927) cites a Moffatt specimen from Glen Ellyn, but it is S. neglectus Nash. Triplasis purpurea (Walt.) Chapm. COOK: Along Belt R.R., Chicago, Thieret 1019 (F). Jones and Fuller (1955 list this species for Cook and Lake counties. The Cook coun- ty report is based on Moffatt, Chicago, Center Ave. and 47th St. (ILL), but the specimen is actually Diplachne fascicu- laris. 1 have not seen a Lake county, Illinois, specimen; Thieret 1019 may be the first authentic record for the Illi- nois part of the Chicago region. Otherwise known from all three counties in Indiana. EXCLUDED SPECIES Cinna latifolia (Trev.) Griseb. No specimens have been seen to support the reports of Mosher (1918) and Pepoon (1927) from Lake county, Illinois (based on Gates, Beach, in 1909) ; Jones (1945), (1950) and Jones and Fuller (1955) from Cook and Lake counties ; and Peattie (1930) from Lake county, Indiana. These entries probably should be referred to C. arundinacea L., a similar species. Deam (1940) lists C. latifolia under excluded species for Indiana. 1961] Glassman — Grass Flora of Chicago Region 137 Eragrostis reptans (Michx.) Nees. The listings of Higley and Raddin (1891) and Jones and Fuller (1955) for Cook, Lake and Du Page counties are based on specimens of E. hypnoides (Lam.) B.S.P., a closely related species. Leersia lenticularis Michx. No specimens have been seen to verify the reports of Pepoon (1927) and Fernald (1937). The listings of Jones and Fuller (1955) and Mosher (1918), as Homalocenchrus lenticularis Michx., for Cook county (based on Hill 76.1892, ILL, from Wolf Lake) and Will coun- ty (based on Hill 17.1911, ILL, from Homer) are actually Leersia virginica Willd. Muhlenbergia sobolifera (Muhl.) Trin. No specimens have been found to substantiate the reports of Higley and Raddin (1891) and Pepoon (1927), based on Babcock in 1885, Stony Island, Cook county; or Peattie (1930) from Clarke, Lake county, Indiana. Not listed by Deam (1929), (1940) or Jones and Fuller (1955) from the Chicago region. These reports most likely refer to M. tenuiflora (Willd.) B.S.P., a similar species. Oryzopsis racemosa (Sm.) Ricker. The report of Mosher (1918) from Cook county (based on Johnson, Glencoe) has not been verified. Deam (1929) cited this species for Porter county (based on Lyon, ND). In 1940, he omitted O. race- mosa from the Chicago region and transferred the Lyon specimen, which I have seen, to O. asperifolia Michx. Sporobolus brevifolius (Nutt.) Scribn. This entry is based on Hill 185.1906 (F, ILL) from Joliet, Will county (Hill, 1915). The specimen is Muhlenbergia cuspidata (Torr.) Rydb. Sporobolus clandestinus (Bieler) Hitch. The specimen is Sporobolus vaginiflorus (Torr.) Wood. Gates 1488, (F), from Leyden, Cook county, is the basis of this entry by Mosher (1918), Pepoon (1927) and Jones and Fuller (1955). Sporobolus virginicus L. This listing by Pepoon (1927) is based on Umbach 2290 (wis) from Lake county, Illinois; but the specimen is actually Agrostis palustris Huds. 138 Rhodora [Vol. 63 LITERATURE CITED CHASE, A. 1951. Hitchcock's Manual of Grasses of the United States, ed. 2, 1051 pp. Misc. Publ. 200, U. S. D. A. DEAM, C. 1929. Grasses of Indiana, 365 pp. Indianapolis. 1940. Flora of Indiana, 1236 pp. Indianapolis. DEAM, C. et al. 1942. Indiana Plant Distribution Records. II. 1941. Proc. Ind. Acad. Sci. 51: 120-129, 1944. Ibid. IV. 1943. l.c. 53: 105-115. 1946. Ibid. VI. 1945. l.c. 55: 50-64. 1947. Ibid. VII. 1946. l.c. 56: 106-112. 1951. Ibid. XI. 1950. l.c. 60: 82-90. 1953. Ibid. XIII. 1952. l.c. 62: 108-113. FERNALD, M. L. 1937. Local plants of the Inner Coastal Plain of Virginia. Rhodora 39: 321-366; 465-491. 1950. Gray's Manual of Botany, ed. 8, 1632 pp. American Book Co., New York. GLASSMAN, S. F. 1957. Notes on the Grass Flora of the Chicago Re- gion. Rhodora 59: 230-235. GLEASON, H. A. 1952. The New Britton & Brown Illustrated Flora of the Northeastern United States and Adjacent Canada. vol. 1, 482 pp. New York Botanical Garden. HIGLEY, W. K. AND RappIN, C. S. 1891. Flora of Cook County, Illi- nois and a part of Lake County, Indiana. Bull. Chicago Acad. Sci. 2: 138-147. HILL, E. J. 1915. Notes on the flora of the Chicago Region. Torreya 15: 21-26. HITCHCOCK, A. S. 1935. Manual of the Grasses of the United States. 1040 pp. Misc. Publ. 200. U. S. D. A. JONES, G. N. 1945. Flora of Illinois. Amer. Midl. Nat. Monogr. 2: 1-317. University of Notre Dame, South Bend. 1950. Flora of Illinois. 2nd ed. l.c. 1-368. JONES, G. N. AND FULLER, G. D. 1955. Vascular Plants of Illinois. 593 pp. University of Illinois Press, Urbana. MosHER, E. 1918. Grasses of Illinois. Univ. Ill. Agr. Exp. Sta. Bull. 205: 257-425. PEATTIE, D. C. 1930. Flora of the Indiana Dunes, 432 pp. Field Mu- seum of Natural History, Chicago. PEPOON, H. S. 1927. An Annotated Flora of the Chicago Area. Bull. Chicago Acad. Sci. 8: 1-554. * REEDER, J. 1951. Setaria lutescens, an Untenable Name. Rhodora 53: 27-30. STEYERMARK, J. A. AND SWINK, F. 1955. Plants new to Illinois or to the Chicago Area in Illinois. Rhodora 57: 266. THIERET, J. AND Evers, R. 1957. Notes on Illinois Grasses, Rhodora 59: 123-124. 1961] Love — Notes on Myriophyllum 139 SOME NOTES ON MYRIOPHYLLUM SPICATUM ÁSKELL LÓVE In the distant past it was generally assumed that the majority of species of higher plants of the boreal zone had a circumpolar distribution, perhaps somewhat interrupted in climatically adverse regions. Many species originally de- scribed from northern Europe were found to be represented also in America and eastern Asia. Although some students did observe a few differences between these populations, others minimized this, and the general hypothesis of circum- polar distribution of most of these plants was regarded as an indisputable fact despite the limited evidence. When larger collections were accumulated from different parts of the globe and critical botanists started to make detailed studies of the circumpolar species, several of these were found to be composed of morphologically distinct taxa each of which was characterized by its more or less defined area. In case such taxa are only races of the same species, classifiable at the subspecific level or lower, their combined area is that of a truly cireumpolar species. However, many are doubtlessly better regarded as species, and it is then a fallacy to classify them only as a part of a circumpolar species. Such a grouping may be very misleading from all geobotanical points of view because it indicates identity of units that are evolutionarily dissimilar, or could at least be expected to be so. In many cases such taxa are known to differ not only in morphology and distribution but also in chromosome number, so they cannot logically be interfertile races but must be regarded as evolutionarily distinct species on basis of their strong reproductive isolation. This was pointed out recently by Lóve (1954a, b, 1955) in connection with studies of some such taxa from Eurasia and North America. Collectively speaking, these taxa appear circum- polar, but from the point of view of modern evolutionary taxonomy and geobotany they are only quasi-circumpolar, because each such collective taxon is composed of two or more false vicariants that are true species of a more limited range and often of a different origin. 140 Rhodora [Vol. 63 One of the species that is still regarded as circumpolar by many is the aquatic Myriophyllum spicatum s. lat., originally described by Linnaeus (1753) from quiet waters in Europe, with references to localities from Lappland in the north to Montpellier in the south. A North American plant was iden- tified as the same species by Pursh (1814) and these two were regarded as being completely conspecific until Fernald (1919) pointed out that they differ in several characters so essential that he did not hesitate to give the American plant a new name, M. exalbescens. Although Fernald (l. c.) clear- ly showed that each of the two taxa has a wide and distinct gecgraphical range, in addition to their morphological differ- ences, later authors have often either ignored M. exalbescens (Rydberg, 1932; Tidestrom & Kittel, 1941) or agreed with Hultén (1947) in regarding it as only an American sub- species exalbescens, of the circumpolar species M. spicatum. Only a few chromosome numbers have so far been report- ed for species of the genus Myriophyllum. The first report, 2n— 14, was made by Scheerer (1939) on German popula- tions of M. alterniflorum L., and this has later been confirmed for Icelandic and American material by Lóve & Lóve (1956, 1958) and for Greenland plants by Jórgensen, Sóren- sen & Westergaard (1958). Scheerer (1940) also reported the number 2n--28 for German plants of M. verticillatum L., whereas Japanese plants so named have 2n—42, according to Harada (1952). M. tenellum Bigel. is diploid with 2n — 14 chromosomes, as determined by the present writer on mater- ial from Lac Ouareau in the Laurentian Mountains of Que- bee. Reports of the chromosome number of M. spicatum L. s. str. first were made by Love & Lóve (1948), who then had counted 2n—c. 36 in an inferior fixation of Icelandic materi- al, whereas later Lóve (1954a, b) corrected this to 2n— 28 also on the basis of faulty preparations from plants of two Icelandic localities. Recently better fixed material from Mývatn in northern Iceland has been found to have 2n— 42 chromosomes so the plant is apparently a hexaploid. In connection with the corrected report of 2n— 28 chromo- somes for M. spicatum, Love (l. c.) reported that M. exalbe- 1961] Love — Notes on Myriophyllum 141 scens is a diploid with 2n—14 chromosomes, as had been counted on material from Lake Manitoba. Consequently, since this indicated that these two related taxa were not only morphologically and geographically distinct as shown by Fernald (l. c.) but also separated by a reproductive barrier as indicated by the difference in chromosome number, it was concluded that this added considerable strength to the sepa- ration of these taxa at the specific level. Unfortunately, however, this report has later been found to be erroneous due to a mixup of notes; the material in question belonged to M. alterniflorum L. Specimens of true M. exalbescens, from Lake Manitoba and its marshes at Delta on its southern shores (cf. Love & Lóve, 1954), were found to have 2n— 42 chromosomes. That number has later been confirmed on plants from several places in the Rocky Mountains of British Columbia and Alberta, several lakes in different parts of Manitoba where material was fixed by the present writer or by Dr. J. C. Ritchie, and from some localities in western Ontario and in the Laurentian Mountains of Quebec. There is no doubt, therefcre, that all over the area of M. exalbescens it is characterized by the hexaploid number 2n— 42 chromo- somes. Since the chromosome number is the same in both taxa, observational cytotaxonomy cannot contribute a more certain solution to their problem of classification than could morphology and chorology ; in fact the occurrence of some reproductive barrier between M. spicatum s. str. and M. exalbescens can only be inferred as long as biosystematic experiments have not been performed including various populations of both. However, the present writer is of the opinion that these taxa may be biologically more effectively isolated than, e. g., Populus tremula and P. tremuloides and several other species pairs without differences in chromo- some number, and thus favors their being retained at the species level unti! experiments prove this to be wrong. The morphological differences between M. spicatum and M. exalbescens include many characteristics, but the most distinctive ones seem to be the following: (1) The number of pairs of leaf-divisions on each primary rachis is usually 142 Rhodora [Vol. 63 14-24 in M. spicatum, but only 4-14 in M. exalbescens. (2) Winter buds are unknown from M. spicatum, whereas they are typical of M. exalbescens. (3) The floral bracts are rhombic to elongate in M. spicatum but typically spathulate- ovate or even oblong-cochleiform in M. exalbescens. (4) The female bracts in M. spicatum are longer than the fruits, but rarely equal the fruits in length in M. exalbescens. (5) The bracteoles are reniform or suborbicular in M. spicatum, but always ovate in M. exalbescens. (6) The bracteoles are broader than long in M. spicatum, whereas they are longer than broad or of equal dimensions in length and breadth in M. exalbescens. (7) The stem of the dried plant is usually fulvous or olivaceous in colour but rarely somewhat whitened in M. spicatum, whereas it always is distinctly whitened in M. exalbescens. The species also differ in several quantative characters, but since these seem to overlap, as such charac- ters usually do, they are useful only when many measure- ments can be made and compared statistically. However, the differences between these two species are so obvious that even without knowledge of the occurrence of reproduc- tive isolation between them, they can confidently be consid- ered to be separate species. The differences are of a magnitude greater than that separating many good species delimited by classical taxonomists, and hundreds of species pairs with similar distribution but based on much weaker morpholcgical characteristics could easily be listed. In a paper reporting the results of a detailed study of populations in a New Jersey lake, Patten (1954) concluded that not only M. exalbescens but also M. spicatum must be met with in New Jersey. He was of the opinion that these spe- cies intergrade and should be regarded as races only of the same species. Unfortunately, no cytological studies have so far been performed on the material reported by Patten (l. c.) and the pollen fertility of the putative hybrids was not ob- served, nor was the fertility of the seeds, as far as the report goes. Based on the present knowledge of the differences be- tween the parents Patten (l. c.) presumed were involved in the hybridization, some introgression could perhaps be ex- pected if they happened to grow together at the same place. 1961] Love — Notes on Myriophyllum 143 One may perhaps venture to suppose that the results report- ed by Patten (l. c.) in fact do not support his assumption of the occurrence and introgression of M. spicatum in New Jersey, but that his presumed hybrids may rather have been formed between M. exalbescens and another more certainly North American species. Such a hybrid may or may not show some degree of fertility so that some introgression could be possible, although experimental studies are needed before it can be regarded as an acceptable solution of the problem pondered by Patten (l. c.). Such an assumption is, however, far more plausible and much less far fetched than is the hypothesis of the occurrence and introgression of M. spicatum, à species of another continent which is nowhere else even indicated from North America east of south western Alaska. When Fernald (1919) distinguished M. exalbescens and excluded M. spicatum from North America, he also described another new species, M. magdalense, later (Fernald, 1924) corrected to M. magdalenense. This latter taxon is closely related to M. exalbescens, differing mainly in more or less dubious fruit characters. It is met with only on the Magdalene Islands and certainly belongs to the group of endemic races, from the regions adjacent to the Gulf of St. Lawrence, which have been unduly classified at a much too high level. The present writer is of the opinion that this taxon is quite comparable to the few local variations known in Europe from M. spicatum and so proposes that it be reduced to the level of variety : M. exalbescens var. magdalen- ense (Fern.) Lóve, stat. nov. (based on M. magdalense Fer- nald, in Rhodora 21, 1919, p. 122, and M. magdalenense Fernald, in Rhodora 26, 1924, p. 198). Variations at the same level are expected to be discovered when detailed stud- ies are made of other somewhat isolated populations else- where on the continent, as indicated already by Fernald (1919) in mentioning the occurrence in Colorado of speci- mens with elongate bracts, and, thus, in this character reminding one of M. spicatum. The distribution area of M. exalbescens, according to Hultén (1947), includes the North American continent from 144 Rhodora [Vol. 63 westernmost Alaska to Greenland, south to California, Ari- zona, New Mexico, Kansas, northern Indiana, and Connecti- cut. It seems to be replaced by M. spicatum in south western Alaska and the Aleutian Islands. It is likely that the species also occurs in easternmost Asia, though it is ignored by the authors of the Flora USSR and other recent Asiatic floras. In Greenland, the species is met with only in the west-cen- tral parts of the country, between Sóndre Strómfjord and Umanak (cf. Bócher, Holmen & Jakobsen, 1957). There it belongs to a group of American plants of limited distribu- tion the origin of which has not yet been properly settled (cf. Iversen, 1953; Bócher, 1954). — LABORATORY OF BIO- SYSTEMATICS, BOTANICAL INSTITUTE, UNIVERSITY OF MON- TREAL, LITERATURE CITED BócHER, T. W. 1954. Oceanic and continental vegetational complexes in southwest Greenland. Medd. om Grónl. 148,1: 1-336. BócuER, T. W., HOLMEN, K. & JAKOBSEN, K. 1957. Grönlands flora. — Köbenhavn. FERNALD, M. L. 1919. Two new Myriophyllums and a species new to the United States. Rhodora 21: 120-124. 1924. Myriophyllum magdalenense, a correction. Rho- dora 26: 198. HARADA, I. 1952. Chromosome studies of some dicotyledonous water plants. Jap. Journ. Genet. 27: 117-120. HULTEN, E. 1947. Flora of Alaska and Yukon. VII. Lunds Univ. Arsskr. N. F. Avd. 2, Bd. 43, Nr. 1: 1067-1200. IVERSEN, J. 1953. Origin of the flora of western Greenland in the light of pollen analysis. Oikos 4: 85-103. JORGENSEN, C. A., SORENSEN, T. & WESTERGAARD, M. 1958. The flower- ing plants of Greenland. A taxonomical and cytological survey. Biol. Skr. Dan. Vid. Selsk. 9, 4:1-172. LINNAEUS, C. 1753. Species plantarum. — Holmiae. Love, A. 1954a. Cytotaxonomical remarks on some American species of cireumpolar taxa. Svensk Bot. Tidskr. 48: 211-232. 1954b. Cytotaxonomical evaluation of corresponding taxa. Vegetatio 5-6: 212-224. 1955. Biosystematic remarks on vicariism. Acta Soc. Fauna et Flora Fenn. 72, 15: 1-14. LóvE, A. & LÓvE, D. 1954. Vegetation of a prairie marsh. Bull. Torrey Bot. Club 81: 16-34. .1956. Cytotaxonomical conspectus of the Ice- landic flora. Acta Horti Gotob. 20: 65-291. 1961] Sherff — Bidens 145 .1958. The American element in the flora of the British Isles. Bot. Notiser 111: 376-388. PATTEN, B. C. 1954. The status of some American species of Myrio- phyllum as revealed by the discovery of intergrade material be- tween M. exalbescens Fern. and M. spicatum L. in New Jersey. Rhodora 56: 213-225. PunsH, F. 1814. Flora Americae Septentrionalis. I-II. London. RYDBERG, P. A. 1932. Flora of the prairies and plains of central North America. New York. SCHEERER, H. 1939. Chromosomenzahlen aus der schleswig-holstein- ischen Flora. I. Planta 29: 636-642. 1940. Chromosomenzahlen aus der schleswig-holstein- ischen Flora. II. Planta 30: 716-725. TIDESTROM, I. & KITTEL, T. 1941. A flora of Arizona and New Mexico. Washington. BIDENS CONNATA VAR. GRACILIPES FERN. IN WESTERN MICHI- GAN. — In a former article (Brittonia 11: 190. 1959), I reported the occurrence of this interesting variety as grow- ing on the east shore of Eagle Lake, Kalamazoo County, of southwestern Michigan. This location marked the farthest point west at which the variety was known to occur. It was originally described from Massachusetts, and subse- quently cited by me (The Genus Bidens, Bot. Ser. Field Mus. Nat. Hist. 16: 261. 1937) for "Maine and southward to Connecticut". More recently I found a specimen of this same variety among additional materials sent me for examination from the Herbarium of the University of Michigan at Ann Arbor, C. W. Bazuin 4282, wet, sandy shore of Wolf Lake, 3-4 miles east of Muskegon, Michigan, Oct. 5, 1941. Wolf Lake is roughly seventy miles farther north and twenty miles far- ther west than the previously known habitat, Eagle Lake: in Kalamazoo County. To learn if the variety was still pres- ent and perhaps well established at Wolf Lake, I made a journey there on October 7th, 1960, visiting the southwest shore. Many specimens were observed growing in the wet sand almost immediately at the water's edge. Twenty or more plants were gathered for making dried herbarium specimens. These will be sent to various herbaria for record purposes in the course of time. — EARL EDWARD SHERFF. 146 Rhodora [Vol. 63 NEW PLANT RECORDS FROM KANSAS RONALD L. MCGREGOR, VERNON L. HARMS AND JEROLD D. POINDEXTER Continued botanical exploration centered in southeastern Kansas has resulted in several additions to the state flora. The area in which all but one of the new records were found is one of a few square miles in the extreme southeastern corner of the state where Mississippian rocks, mainly lime- stone, are exposed. This area has a strong Ozarkian floristic element and has more species of vascular plants than any of the other much larger physiographic areas of Kansas. A few species known in Kansas from this area only are: [soé- tes butleri, Erythronium americanum, Saururus cernuus, Arabis laevigata, Vaccinium stamineum, Gillenia stipulacea, Crataegus spp., Cornus florida, Physocarpus opulifolius var. intermedius, Aster turbinellus, and Hieracium gronovii. All specimens listed are in the herbarium of The Univer- sity of Kansas, Lawrence, and were collected by the senior author in company with the others. Scirpus koilolepis (Steud.) Gleason. CHEROKEE CO.: low abandoned field, 1 mile northwest of Baxter Springs, April 30, 1960, R. L. McGregor 15343, This small rush occurred abundantly with Myosurus mini- mus, Sagina decumbens, and Lepidium densiflorum. Its range is extended west from southern Missouri and north from Oklahoma. Carex stipata Muhl. var. oklahomensis (Mackenzie) Glea- son. CHEROKEE CO.: open wooded creek valley, 1 mile north- west of Galena, June 18, 1960, R. L. McGregor 15728. A large colony of this sedge was found. The leaves aver- aged 3-5 mm. wide and perigynia averaging 4.2 mm. long and nearly as wide as high. It was known previously from southwest Missouri to eastern Texas. Carex squarrosa L. CHEROKEE CO.: moist ravine in woods, 5 miles east and 1 mile south of Crestline, July 16, 1960, R. L. McGregor 15822. 1961] McGregor, Harms and Poindexter — Records 147 We have seen no specimen to validate the previous report of this species from Kansas. Its range is extended west from Missouri. Sisyrinchium varians Dickn. CHEROKEE CO.: prairie at edge of oak-hickory woods, 5 miles east of Baxter Springs, May 15, 1960, R. L. McGregor 15588. Intergrades so completely with S. pruinosum Bickn.. that it should, perhaps, be included with that more southern spe- cies. Our collection extends the range to the north from Oklahoma. Mentzelia altescens (Gill.) Griseb. CHEROKEE CO.: cherty waste land in old lead and zinc mine area, 4 miles east and 1 mile south of Crestline, July 16, 1960, R. L. MeGregor 15823. Darlington (Ann. Mo. Bot. Gard. 21: 160. 1934) gave the range of this species as Oklahoma south ward through Mexi- co, Chile, and Argentina. The Oklahoma part of the range was based on a collection in Comanche County which is some 280 miles southwest of our station. It was a pleasant surprise, then, to find a colony of several hundred plants flourishing in southeastern Kansas. The colony extended eastward a few hundred yards into Missouri to a point 3 miles southwest of Carl in Jasper County. Gonolobus gonocarpos (Walt.) Perry. CHEROKEE CO.: rocky wooded ravine, 5 miles north of Galena, July 16, 1960, R. L. MeGregor 15809. A few plants in the area extend the range a little west from Missouri and north from Oklahoma. Erigeron tenuis T. & G. CHEROKEE CO. : sandy prairie bank, 1 mile northwest of Baxter Springs, May 14, 1960, R. L. McGregor 15577. This slender annual was abundant in the area and its known range is extended north and west from Oklahoma and Missouri. Lactuca salizna L. ANDERSON CO.: edge of low woods and rcadside, 2 miles north of Garnett, August 13, 1960, R. L. MeGregor 15972. Also collected in Douglas and Franklin Counties. This naturalized weedy plant was scattered in the area. 148 Rhodora [Vol. 63 Two plants had lanceolate to linear and entire leaves and are referable to the forma ruppiana (Wallr.) G. Beck. Our collection extends the known range westward from Missouri. Rudbeckia grandiflora (D. Don) DC. CHEROKEE CO.: blue- stem prairie hay meadow, 8 miles west of Baxter Springs, June 18, 1960. R. L. McGregor 15752. The range of this species is extended west from Missouri and north from Oklahoma. Though previously reported from Kansas by Stevens (Kans. Wild Flowers, Univ. of Kans. Press, 1947) his determination was based on a dwarf form of R. subtomentosa. — DEPARTMENT OF BOTANY, THE UNIVERSITY OF KANSAS, LAWRENCE. ERAGROSTIS CURVULA FROM ILLINOIS. — The spontaneous appearance and subsequent collection of an introduced plant is not likely to prompt comment unless it is useful or an unwelcome arrival. However, the presence of Eragrostis curvula (Schrad.) Nees in an Illinois locality may be of some interest to conservationists. This perennial South African bunchgrass was first introduced in the south western U. S. in 1934. Since it is easily established from seed there is little reason to doubt its spread from areas of previous introduction. This writer has no reports of E. curvula from other areas in Illinois or from adjacent areas. The collector of the Illinois specimens is R. T. Rexroat who stated in habitat notes that the grass was “growing in sand on the east side of woods and in a tight clump up to a foot across." It seems from Rexroat's observations that the plants had been there for some time and have persisted. According to information from the collector there were no "conservation area" plantings in the vicinity. The grass was not likely to have been deliberately seeded in that par- tieular area. Identity was verified by Dr. Jason Swallen. Collection data: Morgan County, Illinois, east of Meredosia. Sept. 21, 1960, R. T. Rexroat 7214, 7214A. Specimens are deposited in herbaria of the Illinois State Museum and the U. S. National Museum. — GLEN S. WINTERRINGER, ILLINOIS STATE MUSEUM. SPRINGFIELD. 1961] Moul — Flora of Penikese 149 NOTES ON THE FLORA OF PENIKESE ISLAND, MASSACHUSETTS EDWIN T. MouL Through the courtesy of the staff of the Botany course of the Marine Biological Laboratory at Woods Hole, the author had the opportunity to revisit Penikese Island on July 6, 1960. In the short time available some significant changes in the vegetation were noted, since my observation of 1948, which should be recorded. This island is famous as the site of the Anderson School of Natural History, con- ducted by Louis Agassiz in 1873. The flora of the island was first recorded by David Starr Jordan (1874). Later surveys were made by Lewis (1924), Fogg (1930) and Moul (1948). The island has remained essentially grass covered, but two species of plants that formerly were very common every- where have become extremely rare. These are the wild car- rot (Daucus carota L.) and the daisy (Chrysanthemum leu- canthemum L. var. pinnatifidum Lecoq. & LaMotte). In 1947 when these plants were common, the nesting bird pop- ulation consisted almost exclusively of the Common and Roseate Terns. Today the population is dominantly Herring Gulls with a few Black Backed Gulls. It is possible that the great population of larger birds has been responsible for the eradication of these plants. The two ferns, Dennstaedtia punctilobula (Michx.) Moore and Dryopteris thelypteris (L.) Gray var. pubescens (Law- son) Nakai, are no longer growing at their former sites. Datura stramonium L. was formerly very common along the strand line and around the ruins of the buildings, but only one plant was located this year. Raphanus raphanistrum L. which was confined to the vicinity of South Pond in 1947 has become established in large pure stands at the edge of the morainal cliffs, along the south shore. A number of the shrubs are spreading. Rubus laciniatus Willd., which formed a single thicket in a hollow near the south shore in 1947, has grown vigorously and spread, form- 150 Rhodora [Vol. 63 ing three large distinct clumps. Rhus copallina L. grew only along the shore near the wharf in 1947, now it completely ccvers the crest of a grassy hillside to the west of the old cottage, growing to two feet in height. The thickets of Sambucus canadensis L. have spread, forming large thickets ; this is particularly noticeable on the north side of the island near the Leper Cemetery. The tree population has decreased. All of the Pinus syl- vestris L. planted some years ago is dead. The maples ( Acer pseudoplatanus L.) maintain a constant height, level with the top of the protective morainal hill. They have recovered from the severe hurricane damage, which was evident in 1947. The single specimen of Populus deltoides Marsh., east of the wharf, represented by sucker only in 1947, still exists, but is shrub-like and about 6 feet tall. Dead twigs indicate killing back by wind laden with salt spray. Populus alba L. still forms extensive thickets on the slope above the cottage, but none of these trees is more than 5 feet tall. In 1948, I expressed the belief that the original tree cover, mentioned by Gosnold's naturalists in 1602, might return, but today the evidence indicates that a grass "subclimax" may persist into the future. — RUTGERS UNIVERSITY, NEW BRUNSWICK, N. J. LITERATURE CITED 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. JorpAN, D. S. 1874. The Flora of Penikese Island. Amer. Nat. VIII (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. Volume 63, No. 748, including pages 91-120, was issued April 21. 1961 QM REFERENCE LIGRAS 73. RET JUL 2 0 1961 Dodota JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Associate Editors Vol. 63 June, 1961 No. 750 CONTENTS: Synopsis of the Genus Xanthocephalum (Compositae). (DLUDRASOIDE QN EU SK OPP BERBER E ED S OI 151 Roots and the Taxonomic Differences between Botrychium oneidense and B. dissectum. W. H. Wagner, Jr. .............. 164 Two New Orchid Records for Ontario. H. Andrews .............. 175 A Range Extension for Parapholis incurva. FMI [989 DA ETT Ceres on sidissaeescessentaceuistascstesveces DE d coco AAE o 176 A New Species of Haplopappus from Mexico. Marshall C. Jolnston eee hehe l7 Contemporary Notes on Powell's Expeditions in the West (Review). Joseph EWAN eese 179. The New England Potanical Club, Inc. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume, Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able. All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass. Printed by THE LEXINGTON PRESS, INC. Lexington, Mass, Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 June, 1961 No. 750 SYNOPSIS OF THE GENUS XANTHOCEPHALUM (COMPOSITAE) OTTO T. SOLBRIG! Xanthocephalum is a small genus of homochromous Compositae-Astereae. It is composed of stout annuals or short-lived perennials which are characterized by a reduced pappus, glutinous campanulate or hemispheric involucre, and conspicuous yellow-rayed heads with a rela- tively large number of flowers, both ligulate and tubular. The genus is found in central and northern Mexico, from the state of Puebla to the United States border and in southern Arizona, New Mexico and Texas. The present investigation was prompted by the lack of any taxonomic work covering the species in their entirety. Its aim is therefore only to delimit somewhat precisely the taxa and to present a key to the species, and is not intend- ed as a monograph. In a previous paper (Solbrig, 1960), the characters of Xanthocephalum and related genera were discussed. The present investigation supplements that work by correcting information presented there and by adding some new data on generic characters. Material from the Gray Herbarium (GH) of Harvard University, the U. S. National Herbarium (US), and the herbaria of the University of California at Berkeley (UC) and the University of Michigan (MICH) was examined. In addition, type material from these herbaria and the Royal Botanical Gardens, Kew (K) was examined. To the direc- II am very grateful to Dr. Carroll Wood for reading the manuscript and making valuable suggestions. 151 152 Rhodora [Vol. 63 tors and curators of all these institutions Il am very grate- ful for making the material in their custody available. HISTORY OF THE GENUS Xanthocephalum was described by Willdenow in 1807, apparently based on material collected and described by Humboldt. No specifie name was given in the original description. Kunth (Humboldt, Bonpland and Kunth, 1820) referred the species X. centauroides (without a descrip- tion) to Willdenow's description, stating that he had not seen any material of it. This is, to my knowledge, the first specific epithet ascribed to Xanthocephalum, and may there- fore be considered the type species. In the same work, Kunth described the genus Xanthocoma with the species X. humile. He also redescribed Xanthocephalum centauroides as Pyrethrum Bonplandianum. Lessing (1832) realized the true identity of Pyrethrum Bonplandianum Kunth, but not that of Xanthocoma humile which he maintained. De Can- dolle (1836, 1837) also accepted both genera, describing a new species of Xanthocephalum, X. suffruticosum. He failed to realize the true identity of Pyrethrum Bonplandianum, with the result that he coined the name Xanthocephalum Bonplandianum, and redescribed under X. suffruticosum, what is in reality, X. centauroides. Another new species of Xanthocephalum is described by him as Keerlia linearifolia. In 1852 Asa Gray transferred this last to the genus Gutier- rezia in which, the epithet linearifolia being preoccupied, he coined the combination Gutierrezia Alamani. A year later, Gray described two more species of Xanthocephalum under Gutierrezia: G. gymnospermoides and G. Wrightii. In 1857 Regel, realizing that Gutierrezia gymnospermoides is not a true Gutierrezia, but apparently unaware of the existence of Xanthocephalum, described Guenthera viscosa, based on Gutierrezia gymnospermoides Gray. The following year, still another name, Grindeliopsis (invalid according to the present rules) was created for the same plant by Schultz Bipontinus. George Bentham (Bentham and Hooker, 1873) recognized the synonymy of all these generic names. In 1880, Asa Gray accepted Bentham’s concept and described still another species, Xanthocephalum sericocarpum. Hems- lev (1881-82) in Biology of Central America attempted the first and only revision of the genus. He also described in 1961] Solbrig — Xanthocephalum 153 4 ARIZONA x * NEW MEXICO p e A * t [3 0 o. e SONORA S COAHUILA TAMAULIPAS BENTHAMIANUS CENTAUROIDES GYMNOSPERMOIDES HUMILE LINEARIFOLIU M MEGALOCEPHALUM SERICOCARPUM *(üO*99»0 WRIGTHII GUERRERO PUEBLA Fic. 1. Distribution of the species of Xanthocephalum. Each symbol represents a locality, regardless the number of collections. Above, for Benthamianus, read Bentham- janum. 154 Rhodora [Vol. 63 that work two new species. Since then new species have been described by Fernald (1901) and Robinson (1893). Thus in all, four generic names and over fifteen species have been ascribed to Xanthocephalum. MORPHOLOGICAL AND CYTOLOGICAL CHARACTERS A detailed comparative account of the principal morpho- logical features has been presented elsewhere (Solbrig, 1960). It should be added that, contrary to what was stated then, not all species of Xanthocephalum are annuals, a few being short-lived perennials. Nevertheless, none are globose shrubs like species of Gutierrezia and Amphipappus. Like- wise, the blooming period is not strictly fall (that is the period between September 21 and December 21), but late summer and fall. This applies to all the genera indicated in that work as fall bloomers. Nevertheless no species blooms normally in the spring, as is the case with Amphi- pappus. It may be added that most species of Xanthoce- phalum seem to be restricted to the mountainous areas of central and northwestern Mexico and border areas of the United States, usually in dry pine forests. Two species are adapted to marshy habitats in central Mexico (fig. 1). Xanthocephalum | gymnospermoides, with six pairs of chromosomes, was the only species of the genus known eytologically (Raven, Solbrig, Kyhos and Snow, 1960). Since then I have had the opportunity to get a count on material of Xanthocephalum Wrightii from near Alpine, Arizona (Solbrig 3218, GH) which had only four pairs of chromosomes. In addition Dr. B. L. Turner? has also found n—4 for Xanthocephalum linearifolium, from La Cima Station, Distrito Federal, México (Beaman 3653, MSC). This eliminates one of the generic differences between Xanthocephalum and Gutierrezia (which has x=4). It is interesting to note that Xanthocephalum Wrightii is the species morphologically closest to Gutierrezia. Nevertheless, these results do not require a reassessment of the status of Xanthocephalum as a genus as outlined elsewhere (Solbrig, 1969). 2I am grateful to Dr. Turner for permission to use this hitherto unpublished count. 1961] Solbrig — Xanthocephalum 155 TAXONOMIC TREATMENT Xanthocephalum Willd. Ges. Naturf. Fr. Berlin. Mag. 1: 140. 1807. Xanthocoma HBK. Nov. Gen. et Sp. Plant. 4: 310. 1820. (Type: X. humile = Xanthocephalum humile). Guenthera Regel, Gartenflora 7: 44. 1858. (Type: G. vis- cosa — Xanthocephalum gymnospermoides). Grindeliopsis Sch. Bip. Bonplandia 6: 356. 1858 (Nomen nudum). Stout annuals or short-lived perennials, erect, glabrous to tomentose, often glandular-pubescent or resinous. Leaves petiolate, the cauline sometimes sessile, alternate, thin to subcoriaceous, linear-lanceolate to lanceolate, simple, entire, or divided. Heads solitary at the end of the branchlets, usually eymosely arranged, peduncles of variable length. Heads campanulate or hemispheric, involucral bracts in two to many rows, imbricate, yellowish, reddish or green, with or without green midribs and tips. Flowers pale to golden yellow, ligules usually more than twice the length of the involucre, numerous; tubular flowers campanulate, with a short tube and usually an expanded throat and five small triangular lobes; stamens five, tailless; styles of tubular flowers with triangular or elongate tips with collecting hairs occupying varying degrees of the outer surface of the stylar branches and with the stigmatic papillae always occupying the margins of the lower halves of the stylar branches below the collecting hairs; styles of ligulate flowers only papillate; pappus usually reduced to a low crown less than 0.5 mm. long, infrequently formed by short, irregular squamellae; achenes turbinate, swollen or flat, glabrous or tomentose. TYPE SPECIES: Xanthocephalum centauroides Willd. KEY TO THE SPECIES A. Achenes white-hairy at maturity. Receptacle conical....................--- oc AC RM CM TERR ME cn 6. X, sericocarpum. A. Achenes glabrous or sparsely pubescent. Receptacle flat or convex, not conical. B. Plants 5-10 em. high with entire leaves with entire margins......... SHIT p ree rH RELIER E DUE 2. X. humile. B. Plants more than 20 cm. high, or if less, then with serrate- dentate leaves. C. Leaves usually pinnatifid, sometimes serrulate, if serrulate then not glandular dotted................sss 1. X. centauroides. C. Leaves entire or dentate, occasionally serrated. D. Stems longitudinally grooved, without conspicuous glan- dular hairs. E. Heads 1.5 em. or more in diameter, ligules 15 mm. or more in length, leaves broad (5-10 mm.)........................ A rebos oo S ee 8. X. megalocephalum. 156 Rhodora [Vol. 63 E. Heads less than 1.5 cm. in diameter, ligules not more than 15 mm. in length, leaves narrow or broad. F. Leaves linear, 1-2 mm. broad, ligules 1-2 mm. wide. Plants from an annual rootstock, usually profusely branched.................... 5. X. Wrighti. F. Leaves lanceolate more than 2 mm. broad, ligules 3 mm. wide. Plants from a perennial rootstock, usually little branched................ 3. X. linearifolium. D. Stems round, with conspicuous glandular hairs. G. Involueres glabrous-resinous, leaves usually entire, but sometimes with dentate or serrated margins........ easpavsanqeasneeannsassesucessessssiuchbsnesosebuasey 4. X. gymnospermoides. G. Involucres glandular-pubescent, leaves always den- tate-serrate margined................ 7. X. Benthamianum. 1. Xanthocephalum centauroides Willd. ex HBK. Nov. Gen. et Sp. 4: 312. 1820. Pyrethrum Bonplandianum HBK. Nov. Gen. et Sp. 4: 300. 1820; Xanthocephalum Bonplandianum (HBK.) DC. Prodr. 6: 44. 1837. Based upon Humboldt & Bonpland “in Mexici locis altis prope Valladolid de Mechoacan". Xanthocephalum suffruticosum DC. Prodr. 6: 44. 1837. Based upon Ber- landier 538, “circa urbem Mexici" (G!). Grindelia coronopi- folia Lehm. Linnaea 5: 376. 1830. Based upon cultivated material (presumably living) of the Botanical Garden in Hamburg. Xanthocoma dentata Schauer, Linnaea 19: 725. 1847; Xanthocephalum dentatum (Schauer) Hemsley, Biol. Centr.-Am. Bot. 2: 111. 1881. Based upon Aschenborn 10 “Mexico”. Annual or short-lived perennial, 15-50 cm. high. Shoots usually several from base, racemose branched, grayish-green or green, gla- brous. Leaves lanceolate, 3-8 cm. long, 1-5 mm. wide, glabrous, borders usually serrate or toothed, sometimes almost entire, the teeth distinct and spaced. Heads cymosely arranged at the end of branchlets, in- volucres campanulate or hemispheric, 4-11 mm. wide, 3-6 mm. high; involucral bracts numerous, lanceolate, appressed, glabrous, arranged in 2-5 loose series, with green upper portions. Ligulate flowers 20-40, ligules 4-6 mm. long, 1-3 mm. wide, tube ca. 2 mm. long; tubular flowers numerous 40-60, with a very narrow tube expanding into an upper throat at least three times as wide; stylar branches with tri- angular tips covered with collecting hairs, and a long stigmatic portion. Pappus minute, achenes glabrous, terete or slightly appressed. 1961] Solbrig — Xanthocephalum 157 TYPE LOCALITY: Not known. (see under history of the genus). DISTRIBUTION: Central México, from México City to San Luis Potosí and Durango. MATERIAL STUDIED: México. DISTRITO FEDERAL. México City, Orcutt 4112 (GH), Pringle 6441 (GH, MICH, UC, US) ; prés México, Bourgeau 369 (GH, US); Valle de México, Schmitz s. n. (GH), Schaffner 207 (GH) ; no loc., Ghiesbreght 118 (GH). DURANGO. City of Durango and Vicinity, E. Palmer 265 (GH, MICH, UC, US), 501 (GH, UC, US). SAN LUIS POTOSÍ. San Luis Potosí Schaffner 312 (GH). VERA CRUZ. Mt. Orizaba, Maltrata, Seaton 10 (GH, US). 2. Xanthocephalum humile (HBK.) Sch. Bip. ex Hems- ley, Biol. Centr.-Am. Bot. 2: 111. 1881. Xanthocoma humile HBK. Nov. Gen. et Sp. 4: 311. 1820; Chrysanthemum humile (HBK.) Spreng., Linn. Syst. ed. 16, 3: 584. 1826. Annual or short-lived perennial herb 10-15 cm. high. Shoots few, unbranched from the base, brownish gray, glabrous and slightly fistu- lose. Leaves lanceolate, 0.5-5.0 em. long, 0.5-3.0 mm. wide, the lower petiolate and large, forming a loose basal rosette, the upper sessile and short, glabrous, acute, entire. Heads solitary at the end of the branches, involucre hemispheric, 5-10 mm. wide, 3-6 mm. high; involu- cral bracts lanceolate, glabrous, arranged in two loose series, with midrib and tips green. Ligulate flowers 20-30, ligule 4-6 mm. long, tube 1-2 mm. long; tubular flowers 30-50, not more than twice the number of ligulate flowers, the corolla with a narrow tube app. 2 mm. in length provided with a few trichomes, broadening into an upper throat of same length and app. twice as broad; stigmatic branches of styles of tubular flowers with a triangular tip covered with collect- ing hairs and a lower papillate region; pappus a very short tube; achenes terete or prismatic, glabrous. ILLUSTRATION: HBK. Nov. Gen. et Sp. 4: t. 412, 1820. TYPE LOCALITY: ''Crescit locis humidis Regni Mexicani, inter pagum Carpio et locum Sancti Christophori, Alt. 1180 hex. . ." (Humboldt). (Not seen). DISTRIBUTION: Central México. Known from a few localities in northern Puebla, México City and San Luis Potosí. MATERIAL STUDIED: México. DISTRITO FEDERAL. Valle de México, Schmitz 43 (GH), Pringle 3202 (GH, UC, vs), Pringle 7422 (vs); México, Bourgeau 163 (GH, US). PUEBLA. Near Lago Salado, 253 km. e. of México City, Weaver 865 (GH, US). SAN LUIS POTOSI. San Luis Potosí, C. Parry and E. Palmer 525 (GH, US). 3. Xanthocephalum linearifolium (DC.) Greenman, Publ. Field Mus. Nat. Hist. Bot. 2: 345. 1912. 158 Rhodora [Vol. 63 Keerlia linearifolia DC. Prodr. 5: 310. 1836; Gutierrezia Alamani Gray, Pl. Wright. 1: 91. 1852; Xanthocephalum Alamani (Gray) Benth. & Hook. ex Hemsley, Biol. Centr.- Am. Bot. 2: 109. 1881. Rhizomatous perennial 10-35 em. high, forming a dense mat. Shoots sparsely branched, greenish, somewhat fistulous, glabrous. Leaves 2-8 em. long, 2-8 mm. wide, lanceolate-spathulate, acute, glabrous, the margins entire, sessile or with a poorly defined petiole. Heads solitary at the end of the branches, involucre hemispheric, 6-15 mm. wide, 4-6 mm. high; involucral bracts broadly lance-triangular, glabrous, glu- tinous, appressed in 2 or 3 loose series, with green tips. Ray flowers few (10-20), conspicuous, ligule 5-10 mm. long, 2-4 mm, wide, tube 2-3 mm. long; tubular flowers numerous (60-80), about 5 times more than ligulate flowers, corollas broadly campanulate, 3-4 mm. high; appen- dages of the style covered almost entirely by collecting hairs, the papillate region short; pappus variable, usually a short crown or formed by minute paleae; achenes turbinate, 1-2 mm. lone, glabrous or pubescent, but never densely silky-pubescent. TYPE LOCALITY: “... in Mexico...” (Alaman). (Not seen). DISTRIBUTION: Restricted to the valleys of México City and Toluca and neighboring areas. MATERIAL STUDIED: México. DISTRITO FEDERAL. Cima Station, Pringle 11613 (GH, MICH, US), Moore 3439 (GH, US), Orcutt 3784 (GH, US), Lyonnet 841 (US), Harshberger 120 (GH); 38 km. s. of México, Hiteh- cock & Stanford 7039 (UC, US); 40 km. s. of México City, Schoonberger 8562 (MICH); Valley of México, Reiche 1 (US), s. n. (US). STATE OF MÉXICO. 12 mi. s. of Tlalpan, W. E. & Margaret Manning 531040 (GH); calcareous bluffs, valley of Toluca, Pringle 4195 (GH, MICH, UC, US); Las Cruces, Hinton 1036 (Us); 30 mi. w. of Toluca, Hitchcock & Stan- ford 7217 (US). 4. Xanthocephalum gymnospermoides (Gray) Benth. & Hook. ex Rothrock, in Wheeler, U. S. Geogr. Survey W. of 100th meridian 6: 140. 1878. (Incorrectly ascribed to Benth. & Hook. Gen. Pl. 2: 249. 1873). Gutierrezia gymnospermoides Gray, Pl. Wright. 2: 78. 1853. Guenthera viscosa Regel, Gartenflora 7: 44. 1858. (based on Gutierrezia gymnospermoides). Grindeliopsis gymnospermoides Sch. Bip. Bonplandia 6: 356. 1858 (nomen nudum). Stout annual up to two meters tall. Stem with brownish-green bark, little branched at the base, somewhat more towards the top. Leaves lanceolate, of variable size, about 0.5-3.0 em. broad and 3-15 em. long, 1961] Solbrig — Xanthocephalum 159 glabrous, the margins entire, toothed or slightly serrated, acute, the upper surface shiny and somewhat glutinous, the midrib prominent on lower surface, sometimes somewhat reddish in color. Branchlets, flowering shoots and pedicels often reddish, with prominent capitate glands in varying density. Heads numerous, cymosely arranged in crowded groups at the end of the branchlets. Involucre campanulate, glutinous, 3-7 mm. in diameter and 3-6 mm. in height; involucral bracts numerous, arranged in two loose series, glabrous, glutinous, tightly appressed, usually with green midrib and tips, the ends usually spreading. Receptacle flat or slightly convex, alveolate. Ligulate flowers 50-70, yellow, ligules 2-4 mm. in length, 1-1.5 mm. wide, about the same length as the filiform tube; tubular flowers 150-200, two to three times as many as ligulate flowers, corolla about 4 mm. long, with a narrow tube and expanded throat; style with collecting hairs re- stricted to the short triangular tips. Pappus variable, sometimes a low erown, sometimes a few irregular awns up to half as long as the tubular corolla. Achenes terete, glabrous or slightly pubescent, 1-2 mm. long. TYPE LOCALITY: “Low banks of the San Pedro, Sonora (now Ari- zona)" Wright (GH!). DISTRIBUTION: Along the mountains, northwestern México from Durango to the Arizona border; area of the Huachuca Mts. in Arizona, and Davis Mts. in Texas. MATERIAL STUDIED: México. CHIHUAHUA. Prezén Ortega, Shreve 8894 (us), I. M. Johnston 7935 (GH, US); Majalea, LeSueur, Mex-242 (GH), White 2378 (GH, MICH); Rio Sta. Maria, Thurber 747 (GH); between Casas Grandes and Sabinal, E. W. Nelson 6381 (US), 6379 (GH, US); near Colonia Juárez, E. W. Nelson 6321 (GH, US); 40 mi. from Guadalupe y Calvo, E. W. Nelson 4815 (GH, US) ; Memelichi, Rio Mayo, Gentry 2736 (GH, UC, US); Loreto, Rio Mayo, Gentry 2573 (GH, UC, US); s.w. Chihuahua, E. Palmer 439 (GH, US) ; near Parral, Gold- man 127 (GH, US). DURANGO, 5 mi. s.w. of Guadalupe Victoria, Gentry 8412 (GH, MICH, UC); El Salto to Pueblo Nuevo, Maysilles 7752 (MICH) ; Otinapa, E. Palmer 455 (US). SONORA. Cananea, Ponnelly 7 (uc). United States. ARIZONA. Cochise Co.: San Pedro River, Smart 404 (US); W. of Huachuca Mts., Shreve 7728 (MICH); Ft. Huachuca, Patzky 38 (us); Ramsey Canyon, Huachuca Mts., Hood 181 (US); Huachuca Mts., M. E. Jones s. n. (US), Lemmon 2738 (GH, UC, US); near Ft. Huachuca, Wilcox 38 (US). Pima Co.: Silver Lake, near Tucson, Toumey 608 (US); Sacaton, G. J. Harrison 6016 (GH). Santa Cruz Co.: Alkaline plains, Pringle s. n. (GH, US); Patagonia Mts., Kearney & Peebles 14832 (US); near Nogales, Peebles & Harrison 4726 (us); Nogales, Hood 174 (US); South Arizona, Rothrock 667 (GH, US). TEXAS. Davis Co.: Ft. Davis, Reed 866 (US) ; Davis Mts., Young 8. n. (UC). 160 Rhodora [Vol. 63 NOTE: Some of the specimens from Chihuahua (Gentry 2736, 2573) were devoid of ligulate flowers, although undistinguishable from normal ligulate plants in all other respects. This rayless form which apparently is geographically isolated might deserve varietal status. Nevertheless, since in some Compositae raylessness has proved to be determined by a single gene, it is felt that before giving formal recog- nition to this form, more work, especially of an experimental nature, is needed, 5. Xanthocephalum Wrightii (Gray) Gray, Proc. Am. Acad. 8: 632. 1873. Gutierrezia Wrightii Gray, Pl. Wright. 2: 78. 1852. Stout annual not more than 75-100 em. high. Shoots striate, often branching from the base. Leaves linear-lanceolate, usually 4-6 cm. in length, occasionally up to 10 em. long, not more than 3-5 mm. wide, surface glabrous, the margins entire, sometimes short-ciliate. A loose basal rosette of leaves is present in young plants, but absent in more mature ones, Heads disposed in loose, cymose arrangements. Involu- cre hemispheric, 4-10 mm, wide, 4-6 mm. high; involucral bracts imbri- cated, appressed, glutinous, with a green tip, sometimes slightly spreading. Ray flowers few (14-20), with a long and showy ligule 5-7 mm. long, 2-3 mm. wide and a short tube, 2-3 mm, long; tubular corollas about four times more numerous than the ligulate ones (50- 60), with a broad throat and short tube; style of tubular flowers with long subulate appendages covered with collecting hairs and with only a short lower papillate portion; pappus a low crown; achenes terete, glabrous, somewhat striate, 1-2 mm, long. TYPE LOCALITY: “between Barbacomori and Santa Cruz, Sonora" Wright 1177 (GH!). DISTRIBUTION: Northwestern México (Chihuahua, Durango, and Sonora) and southwestern United States (Arizona and New Mexico), in moist places at mid-altitudes in the mountains. MATERIAL STUDIED: México. CHIHUAHUA. Cañon de San Diego, Lumholtz 765 (GH, US); near Colonia Garcia, E. W. Nelson 6203 a (GH, US), Townsend & Barber 304 (GH, UC, US); 5 mi. S. of Garcia, Leopold 234 (vC); near Chichupa, Townsend & Barber 421 (GH, UC, US); lake near Chichupa, LeSueur 1370 (GH); Chuhuichupa, LeSueur 992 (GH); Sierra Madre, E. W. Nelson 6307 (GH, US), Pringle 1631 (uc); foothills, base of Sierra Madre, Pringle 1280 (GH, MICH, US), 1629 (UC); los Cascarones, Rio Mayo, Gentry 2681 (GH, UC, US); Canelo, Rio Mayo. Gentry 2008 (GH, UC, US); Majalca, Le Seuer, Mex- 30 (GH, UC); 38.2 mi. w. of viejas Casas Grandes, Tucker 2489 (UC, US). DURANGO. City of Durango, E. Palmer 823 (UC, US), E. W. Nel- son 4623 (US); 34 mi. w. of Ciudad Durango, Maysilles 7547 (US). SONORA. El Rancho del Roble, White 4261 (GH, MICH), 4247 (MICH); 1961] Solbrig — Xanthocephalum 161 between Las Tierritas and El Tigre, Phillips 684 (GH, MICH), White 3451 (GH, MICH); Las Tierritas del Temblor, White 3410 (GH, MICH) ; Barbacomori to Sta. Cruz, Thurber 1007 (GH); El Billito, White 4834 (MICH). United States. ARIZONA, Apache Co.: 8 mi. n. of Hannagan Meadow, White Mts., Kearney & Peebles 12428 (GH); White Mts. Gould & Robinson 4978 (UC); Me Nary, Whitehead 1633 (MICH); Buffalo Junction, Solbrig 3218 (GH), Parker & McClintock 7643 (uc, us). Cochise Co.: Mule Mts., Harrison & Kearney 6236 (GH); Rucker Canyon, Chiricahua Mts., Gould & Haskell 4603 (GH); Chiricahua Mts., Lemmon s.n. (UC); Carr Peak, Huachuca Mts., Goodding 869(GH, US); upper Miller Canyon, Huachuca Mts., Goodding 420 (GH); Reef Mine, Huachuca Mts., Darrow, Phillips, Gould & Pultz 1418 (GH); Ramsey Canyon, Huachuca Mts., Shreve 5038 (MICH), M. E. Jones s. n. (GH, UC) ; Huachuca Mts., Peebles, Harrison & Kearney 3387 (US), Harrison & Kearney 5777 (us), Holzner 2025 (US), Shreve 5074 (uc), Gould, Darrow, Phillips & Pultz 2486 (uc), Gould & Haskell 3375 (Uc) : Santa Cruz Co.: Wrighton Peak, Santa Rita Mts., Clark 12310 (GH); Santa Rita Mts., Griffiths & Thornber 169 (US); near Washington, Patagonia Mts., Kearney & Peebles 10107 (MICH, US) ; Roberts Ranch, Wooton s. n. (US). NEW MEXICO. Catron Co.: Mogollon Mts., Wooton s. n. (US), Saunders s. n. (UC), Metcalfe 458 (Us); Grant Co.: Pinos Altos Mts., Greene 200 (GH); Black Range, Pilsbury s. n. (US); Black Mts., Archer 358 (MICH); Silver City, Holzinger s. n. (US); Head of Cow Creek, Eggleston 16030 (GH) ; Iron Creek, Metcalfe 1481 (GH, UC, US). 6. Xanthocephalum sericocarpum Gray, Proc. Am. Acad. 15: 31. 1880. Xanthocephalum conoideum Hemsley, Biol. Centr.-Am. Bot. 2: 110. 1881. Based upon Coulter 299, “Real del Monte to Zacatecas" [Mexico] (K) (Isotype GH !). Stout annual 30-50 cm. tall, usually branching profusely from the base. Shoots greenish, fistulous, less than 5 mm. in diameter at the base and not more than 1-2 mm. at the upper ends, glabrous, often with minute bracteoles. Leaves linear, 3 to 5 cm. long, 1-3 mm. wide, glabrous, acute, border entire, petiole very short or absent. Heads solitary at the end of branchlets, sometimes forming a loose cyme. Involucres hemispheric, 6-12 mm. wide, 5-7 mm. high, involucral bracts broadly lanceolate, acute, tightly appressed in two loose rows, with green tips and midribs; receptacle conical, grooved; ligulate flowers 25-40, ligules 6-12 mm. long, 2-3 mm. wide, tube 2-3 mm. long; tubular flowers 100-150, about four times as many as ligulate flowers, the tube broadly campanulate, 2-3 mm. long; styles of tubular flowers with the distal halves of the stigmatic portions covered with collecting hairs, the lower half papillate; pappus a relatively well 162 Rhodora [Vol. 63 developed erown, 0.5 to 1.5 mm. long; achenes terete, silky pubescent at maturity. TYPE LOCALITY: “Near San Luis Potosi...” Parry and Palmer 369 (GH!). DISTRIBUTION: Northern México, from Jalisco and Querétaro north- ward. MATERIAL STUDIED: México. CHIHUAHUA, Pine plains, base of Sierra Madre, Pringle 1278 (GH, MICH, us); Namaquipa, LeSueur, Mex-337 (GH); near Balleza, Goldman 141 (GH, US) ; 12 mi. w. of San Antonio, C. Muller 3380 (GH, MICH, US); southwestern Chihuahua, E. Palmer 412 (GH, US) ; Mapula Mts., Pringle 1150 (GH) ; 34 mi. s. of Parral, Waterfall 12528 (MICH). DURANGO, City of Durango and vicinity, E. Palmer 823 (GH), 143 (GH, MICH, UC, US) ; Sandia Station, Pringle 13580 (GH, MICH, US); 34 mi. w. of Durango, Maysilles 7547 (MICH), 7689 (MICH, US). JALISCO. Road between Huajnilla and Mesquite, Rose 2560 (US). QUERÉTARO. Cazadero Station, Pringle 10066 (GH, UC, US); 15 mi. se. of San Juan del Rio, Waterfall 13987 (US). SAN LUIS porosf. San Luis Potosí, Schaffner 744 (US). 7. Xanthocephalum Benthamianum Hemsley, Biol. Centr.- Am. Bot. 2: 110. 1881. Stout annual 20-50 em. tall with little or no branching from the base, and only slightly branching above. Stems greenish, greenish brown or reddish, slightly fistulous at times, 1-5 mm. in diameter, usually covered with capitate glands, especially in the upper portions. Leaves broadly lanceolate, 2-8 cm. long, 5-10 mm. wide, the lower shortly petiolate, the upper ones sessile, glabrous or slightly pubescent, the margins coarsely serrate. Heads solitary at the end of branchlets, sometimes arranged in loose cymes. Involucres campanulate, 8-15 mm. broad, 4-7 mm. high; involucral bracts numerous, variously covered with capitate glands, loosely arranged in several series, the upper portions green, the tip usually extended. Ligulate flowers 30-40, ligules 6-10 mm. long, tube app. 2 mm. long; tubular flowers numerous (100- 200), four to six times as many as the ligulate, the very narrow tube followed by a wide throat; tips of stylar branches triangular, covered with collecting hairs, and with a large papillate region. Pappus a low crown; achenes terete, glabrous, 1-2 mm. high. TYPE LOCALITY: “South México, Aguas Calientes” Hartweg 110 (Isotype GH!). DISTRIBUTION: Central México, from México City to Durango. MATERIAL STUDIED: México. AGUASCALIENTES. Hartwegg 110 (GH). DURANGO. W. of Durango, Forber 17 (UC) ; 5 mi. n. of Coyotes, May- silles 8171 (MICH); Coyotes Hacienda, Maysilles 7853 (MICH) ; 60 mi. sw. of Durango, Maysilles 7781 (MICH); Rio Chico, Maysilles 7656 (MICH). STATE OF MÉXICO. Del Rio, Pringle 5336 (GH). MICHOACAN. 1961] Solbrig — Xanthocephalum 165 Patzcuaro, Kenoyer 1690 (GH). SAN LUIS POTOSI. Chiefly in the region of San Luis Potosí, C. Parry & E. Palmer 526 (GH, UC). ZACATECAS. Near Plateado, Rose 2720 (Us). NOTE: Xanthocephalum Benthamianum is closely allied to X. gymno- spermoides and might possibly be only a southern form of this last species. The serrated leaves and the glandular dotted involucres, the inereased general pubescence and the more southern distribution (with a considerable overlap) of X. Benthamianum seem nevertheless suffi- cient specific distinction from a typologieal point of view. Only field studies and crossing experiments which are lacking at present can determine the true status of X. Benthamianum. 8. Xanthocephalum megalocephalum Fernald, Proc. Am. Acad. 36: 505. 1900. Rhizomatous perennial 30-70 cm. high. Stems glabrous, greenish, slightly fistulous, little or not branched at the base, moderately so above. Basal rosette leaves spatulate-lanceolate when present, 10-15 em. long, 5-10 mm. broad, petiolate, acute, glabrous, Cauline leaves sessile, lanceolate, 4-6 cm. long, 5-15 mm. broad, glabrous. Heads borne singly or in pairs at the end of branchlets; involucre hemispheric, 10-20 mm. broad, resinous, with a green tip; ligulate flowers 15-30, ligules large, 12-20 mm. long, the tube 3 mm. long; tubular flowers 80- 100, about four times as many as ligulate flowers, campanulate, with a short tube and broad throat; stigmatic branches subulate, almost completely covered with collecting hairs, the stigmatic papillae re- stricted to the lower portion; pappus a short crown not more than half a millimeter in length. Achenes terete, 1-2 mm. long, glabrous. TYPE LOCALITY: “Chihuahua, Mt. Mohinora" E. W. Nelson 4890 (GH!). DISTRIBUTION: Uncommon in northern Durango and western Chihua- hua, México. MATERIAL STUDIED: México. CHIHUAHUA. Near Guachichi, Goldman 174 (GH, US) ; along road to Parral, near San Julián, E. W. Nelson 4932 (GH, US); Sierra Madre, Pringle 1630 (UC), 1279, (GH, UC, US) ; mountains north of Chihuahua, LeSueur 1377 (GH). DURANGO. State of Durango, Ibarra Garcia 412 (US). NOTE: This rare species shows close similarities to XY. Wrightii and to X. linearifolium. Nevertheless the large heads and long ligules were sufficiently distinct to permit an easy identification of X. megalocepha- lum in the material studied. 3RAY HERBARIUM, HARVARD UNIVERSITY. LITERATURE CITED BENTHAM, G., AND J. D. HOOKER, 1873. Genera plantarum. Vol. 2(1), 553 pp; London. DE CANDOLLE, A. 1836-37. Prodromus systematis naturalis. Vols. 5 and 6. Paris. 164 Rhodora [Vol. 63 FERNALD, M. L. 1901. Some new spermatophytes from México and Central America. Proc. Am. Acad. 36: 491-506, 1901. GRAY, A. 1852. Plantae Wrightianae Texano- Neo-Mexicanae. 146 pp. Washington. . 1873. Notes on Compositae and characters of certain genera and species, ete, Proc. Am. Acad. 8: 631-661. HEMSLEY, W. 1881-82. Biol. Centr.-Am. Bot. 2: 620 pp. London. HUMBOLDT, A., A. BoNPLAND AND C. KUNTH. 1820. Nova genera et species plantarum. 4: 312 pp. Paris. LESSING, C. E. 1832. Synopsis genera compositarum. 473 pp. Berlin. RAVEN, P., O. T. SorBRIG, D. KyHos AND D. SNow. 1960. Chromosome numbers in Compositae. I. Astereae. Am. Jour. Bot. 47: 124-152. REGEL, E. 1858. Im Botanischen Garten zu Petersburg geprüfte neuere and ältere Pflanzen. Gartenflora 7: 43-52. RoBINSON, B. L. 1893. Descriptions of new plants collected in México by C. G. Pringle in 1890 and 1891, with notes upon a few other species. Proc. Am. Acad. 27: 165-185. SCHULTZ BIPONTINUS, K. 1858. Neue Zierpflanzen. Bonplandia 6: 356. SoLBRIG, O. T. 1960. The status of the genera Amphipappus, Amphia- chyris, Greenella, Gutierrezia, Gymnosperma and Xanthocephalum (Compositae). Rhodora 62: 43-54. WILLDENOW, C. L. 1807. Einige Bemerkungen über die Pflanzen der Klasse Syngenesia. Ges. Naturf. Fr. Berlin, Mag. 1: 132-141. ROOTS AND THE TAXONOMIC DIFFERENCES BETWEEN BOTRYCHIUM ONEIDENSE AND B. DISSECTUM W. H. WAGNER, JR.' Since first distinguished nearly sixty years ago, the plant here treated as Botrychium oneidense (Gilbert) House has been a continuous source of taxonomic disagreement. It has been interpreted as a variety of B. multifidum (Gmel.) Rupr. or as a variety of B. dissectum Spreng. (Wagner, 1960a). It was originally described as a variety of a third species, B. ternatum (Thunb.) Sw. Only House (1921) seems to have recognized B. oneidense as a distinct species; but he changed his mind just three years later, and made it a vari- This study was supported in part by a grant from the Horace H. Rackham School of the University of Michigan and in part by National Science Foundation Grant G-10846, I am indebted to Miss Virginia M. Morzenti for her careful assistance. 1961] Wagner — Botrychium 165 ety of B. obliquum Muhl. (1924). In general, then, we can find agreement on only one point, viz., that — whatever it is— taxon oneidense must be a variety of some other species. No author, at least for any length of time, seems to have thought that B. oneidense is truly a distinct species, or that the difficulty of placing it with one or another of the several species of which it has been made a variety might be due to that fact. After personal observations of over sixty localities where the plants grow together in Ohio, Ontario, and Michigan, as well as fresh, living materials sent by others from southern Indiana and Maryland — a total of over 4,500 specimens of B. dissectum and 2,000 of B. oneidense — I feel convinced that these are distinct species (Wagner, 1955, 1960a). Botrychium dissectum is a highly variable plant, the most common and “normal” form of which is f. obliquum (Muhl.) Fern. The type form, dissectum, is often so lacerated that botanists have frequently treated it as a variety, or even as a species, distinct from taxon obliquum. However, I can- not find any differences between f. dissectum and f. obliquum other than those of marginal cutting, and there are numer- ous intermediates. In spite of the rather spectacular con- trast in their appearance, I cannot treat even the extremes as more than minor forms of the same species. On the con- trary, B. multifidum, B. oneidense, and B. ternatum each have a series of differences from one another and from B. dissectum that justify their interpretation as distinct species. This conclusion is substantially bolstered by the fact that all four of these species co-exist side-by-side in the same habitats in any combination, over a tremendous area of the northeastern United States (Wagner, 1960a, 1960b). The most significant point is perhaps that, in spite of over- lap in practically every character that differentiates these plants, there are correlated series of central tendencies for each taxon that in totality are clearly different. These en- sembles of average differences maintain themselves, where- ever these plants grow together, with monotonous regu- larity. I do not, in fact, believe that all of the differences between them have yet been found. The present report will 166 Rhodora [Vol. 63 PLATE 1259, Habitat forms of Botrychium oneidense: Top two rows, deep shade, Saginaw Co., woods along M-83, Sect. 22, R.6E, T.10N, Sept. 20, 1959, 9110. Bottom two rows, edges of cleared old fields, St. Clair Co., along Belle River, Sect, 28, T.5N, R.15E, May 12, 1957, 8393. 1961] Wagner — Botrychium 167 PLATE 1260. Habitat forms of Botrychium dissectum: Top two rows (9109) and bottom two rows (8394) from same localities as Plate 1259. 168 Rhodora [Vol. 63 describe a newly uncovered contrast between two of the species that was quite unexpected and apparently completely overlooked, not only by previous workers but by the present author as well It involves the root differences between botrychium oneidense and B. dissectum. In making studies of critical characters among these species, the populations must be compared in the same habitats (e.g., a shaded swamp; a low, wet, brushy field; or a second-growth wood). Differing localities and differ- ing habitats produce strong modifications as shown in Plates 1259 and 1260, which will be diseussed below. To ensure as precise a comparison as possible the rule followed in this study was to collect only paired. plants of the two species. To obtain each pair, a spot was sought where a plant of B. oneidense grew close to a comparable (i.e., of approximately equal size) plant of B. dissectum. The arbitrary limit of per- missible distance between them was five feet. Then the plant of B oneidense was dug up, along with the neighboring one of B. dissectum. Thus the average distance between the members of the pairs was between two and three feet. (The same rule was followed to compare other species, to be mentioned below.) The measurements of the diameters of the two largest roots of each living specimen were made 1 cm. from the stem to the nearest 0.1 mm., using calipers. RESULTS. The habitat variation of roots in B. oneidense is probably fairly well represented by the data of this study, but the same cannot necessarily be said for B. dissectum. The reason for this is that the morphological differences found between various localities of B. oneidense are consid- erably less pronounced than those in the other species of evergreen grapeferns (B. dissectum, B. multifidum, and B. ternatum). Botrychium oneidense seems to be the least variable of the species. The reason for its lesser environ- mental variability seems very likely to be due to its nar- rower definition of habitat: it is confined almost exclusively to low, wet, acid secondary woods and swamps. The most extreme leaf forms that have been found in this species are illustrated in Plate 1259. The lower two rows are from the most exposed habitat we have yet discovered — exposed 1961] Wagner — Botrychium 169 mossy areas along the edge of a dry field bordering a woods. The upper two rows are from a very deeply shaded habitat along the edge of a swamp. Corresponding specimens of B. dissectum from the same habitats are shown in Pl. 1260. Much more extreme specimens are known of the latter species. The five habitats we selected in which to compare the roots of these two species are what might be called ap- proximately average for B. oneidense — neither the most exposed nor the most deeply shaded. Differences between the means in root diameter of the two taxa in the different localities turned out to vary from 0.5 to 1.0 mm., as shown in Table 1. The roots of B. dissec- tum were always larger. This difference is readily evident to the naked eye when examining the root systems of a series of freshly collected specimens that have been washed. The greater thickness of the roots of B. dissectum is ac- centuated by the fact that they tend to be dark gray-brown in the region 1-5 cm. from the stem especially, but those of B. oneidense are paler, more delicate, and dominantly ivory- gray. Also the roots of the latter tend to have much less TABLE 1. Root diameters of mature plants, the two largest roots of each plant measured 1 em. from the stem. All (except *) paired specimens in same habitat, B. dissectum B. oneidense 1. 9218a (5 plants) 9219a (5 plants) 2.87 2.10 EE 9218b (9) 9219b (10) 3.08 2.15 3. 9263 (14) 9264 (14) 2.90 2.20 4. 0265272619) 9266 (19) 2.92 2.20 5. 9289 (28) 9290 (28) 3.05 2.53 6. 9307 (18) 9309 (18) 3.06 2.48 developed circular ridges on the roots. This is shown in the samples in Plate 1261. An anatomical examination was made of the histological nature of these differences. Specimens fixed in formalin- 170 Rhodora [Vol. 63 acetic-aleohol solution sectioned nicely on the freezing mi- crotome, and were examined under the compound micro- scope. The roots of both species are very fleshy, composed of a large cortex of starch-filled parenchyma cells and tra- versed by a narrow stele, the latter only 10-15 per cent the diameter of the whole root. The root steles of both are most commonly triarch, although the tetrarch condition is frequent near the rhizome, and small roots may be diarch. The pri- mary variation of the roots at the anatomical level seems to involve the relative development of the cortex and the modi- fication of the superficial layers. A thin-walled corky tissue PLATE 1261. Root systems of Botrychium: Three plants on left, 9264, B. oneidense; three on right, 9263, B. dissectum. (Photographed on frosted glass plate, with light from above and below). 1961] Wagner — Botrychium 171 ROOT DIAMETERS: lcm, from stem. BOTRYCHIUM ONEIDENSE EE Herbarium Collections t BOTRYCHIUM DISSECTUM t t b d 4 + A j 4 4 ——— TT v T t 4 | Herbarium Collections FIGURE 1. Root diameters of Botrychium. Curve numbers correspond to Table 1. Vertical scale represents number of individuals. 172 Rhodora [Vol. 63 forms in a varying number of layers from a cambium in the outer cortex, and this tissue, comprising dead, brown-walled cells, is responsible for the darker color of the roots of Botrychium dissectum. The same development is found in B. oneidense, but its extent is less, and thus the roots appear much paler on the average and have smoother surfaces. The development of the root periderm is associated with the formation of circular ridges, these more strongly developed in B. dissectum but only weakly developed in all but the most robust roots of B. oneidense. The differences between B. dissectum and the other spe- cies of evergreen grapeferns, B. multifidum and B. terna- twm, proved to be negligible or inconsistent. In drier field habitats, B. multifidum exceeded the associated B. dissectum in root diameter, but in very damp habitats (two localities, one at the edge of a bog, the other in low swampy woods) it tended to have narrower roots than B. dissectum. Botry- chium ternatum in three localities had roots the same thick- ness or somewhat more slender than those of B. dissectum growing with it. In general, the roots of B. multifidum and B. ternatum are more like those of B. dissectum; only B. oneidense showed a constant average difference, so far as our studies have been able to show. In order to test whether the difference in root diameter between B. dissectum and B. oneidense would also be shown in random herbarium collections, 18 specimens each of the two species were measured from the United States National Herbarium. The results showed an average of 2.6 mm. in diameter in the former, and 2.0 mm. in the latter, as meas- ured 1 cm. from the stem. A remeasurement of dried mate- rials that have been compared in the living state first shows that there is almost exactly 20 per cent reduction in thick- ness. The original differences between species will thus re- main in the dry state but on a somewhat smaller scale. According to this contraction of dried material, the National Herbarium samples would have averaged in the living state 2.1 mm. in diameter for B. dissectum. and 2.4 for B. onei- dense. These values compare favorably with those in Table l. 1961] Wagner — Botrychium 173 DISCUSSION. Although this report is based upon a sample of 187 plants from only 5 habitats, the methods used seem to be sufficiently reliable to say that the average difference we found is a valid one. The fact that a random collection of herbarium sheets gave essentially the same results is an indication that the root differences between B. dissectum and B. oneidense are probably characteristic of the two spe- cies over their range. One of the reasons that special care has been taken to assure that the differences between the roots are not en- vironmentally induced ones is that there is some reason to believe that several of the previously accepted “differences” TABLE 2. Summary of average differences of B. oneidense and B. dissectum. Habitat: Range: Leaf Blade: Pigmentation: Periodicity: Roots: B. oneidense Mainly uniform: low, wet, acid secondary woods and swamps; local and rare in most of its area. Narrow: New Brunswick and Wise. S. to Indiana and Mary- land; and to North Carolina and Tennessee in the mountains. Segments few, large, rounded, with nearly entire to shallowly denticulate Mature surfaces duller, smoother (less margins. *"veiny"). Unfolding leaves in spring lime green; late summer mature blade axes with av. 3-15 per cent pigmentation below; mid- winter laminae green where well exposed. mostly Vernation beginning in May, the new leaf averaging twice as tall during vernation stages; maturation of sporangia in last half of September. Root diameter 1 em. from stem av. 2.5 mm. alive, 2.0 dried; cireular ridges sparse in proxi- mal 1-5 em.; color dominantly ivory-gray to tan, only occasion- al large roots completely dark in basal 1-5 em. B. dissectum Diverse: open fields, meadows, dry woods, wet woods, ubiquitous and com- brushy swamps; mon in much of its area. Wide: Nova Scotia and Minne- sota S. to Mexico and Jamaica. Segments numerous, smaller, pointed, with variable margins denticulate to very deeply and coarsely lacerate. Mature sur- faces shiny, "veiny." Unfolding leaves in spring red- dish; late summer mature blade axes with av. 10-40 per cent pig- ment beneath; midwinter lam- inae bronze where well exposed. Vernation beginning in early June, the new leaf averaging one-half as tall during vernation stages; maturation of sporangia in October. Root diameter 1 cm. from stem av. 3.0 mm. alive, 2.5 mm. dried; circular ridges well developed in proximal 1-5 em.; color dom- inantly dark gray-brown in proximal 1-5 cm. 174 Rhodora [Vol. 63 between these species are actually based upon habitat modi- fications. For example, we have found no evidence that B. oneidense has a distinctively “thin texture of the blade" when growing with B. dissectum (Wagner, 1960b, table 4, p. 318). Likewise, we have found no evidence of a “marked tendency toward lower fertility": indeed, when growing sympatrically in the habitats, the two species are very much alike (Wagner, 1961). We examined the spores to find dif- ferences but they too are similar. However, there are a number of real average differences between these plants. It is, of course, possible that one or a few important genes could somehow control all of the differ- ences, but this seems unlikely because the characters in- volved are so diverse. The differences lie in six categories (listed in Table 2). It is probable that many of the details of contrasts within several of the categories are inter-related and are aspects of the same thing: for example, the three contrasts under “roots” are probably all correlated. Those under periodicity, under pigmentation, and under leaf struc- ture, may be also — but this is questionable. For example, the size and the shape of segments must surely be under different genetic control, judging from these characters in the other species. Therefore, the best statement of the dif- ferences between these two species, P. oneidense and B. dis- sectum, that we can make (including here the new root characters, as well as those studied previously) is as follows: they are sympatric species in northeastern North America, the range of one (B. dissectum) completely overlapping the other (B. oneidense), and they differ in the central tenden- cies of differences in six known categories: (1) Habitat; (2) Range; (8) Leaf blade structure; (4) Pigmentation; (5) Periodicity; and (6) Root size and development. VOUCHER SPECIMENS: (All Michigan field collections). Wayne Co., junct. Flat Rock Rd. and Expressway, R.9E, T.98, Sect. 19, May 20, 1960, 9218a (Dissectum), 9219a (Oneidense), 9218b and 9219b unpaired in same habitat. Monroe Co., woods W. side of Secor Rd., 0.4 mi. s. of Todd Rd., R.7E, T.8S, Sect. 30, August 4, 1960, 9263 (D), 9264 (O). Monroe Co., S.W. corner of Nolan and Exeter Rd., 1961] Andrews — Orchid Records 175 R.8E, T.5S, Sect. 24, August 4, 1960, 9265 (D), 9266 (0). Saginaw Co., Fordney R., N. of Brady Rd., Sect. 11, R.2E, T.9N, Aug. 13, 1960, 9289 (D), 9290 (O). St. Clair Co.: N. side of Rt. 21, 0.4 mi. W. of Beach Rd., Sect. 6, R.17E, T.6N, August 23, 1960, 9307 (D), 9309 (O). — UNIVERSITY OF MICHIGAN, ANN ARBOR. LITERATURE CITED CLAUSEN, ROBERT T. 1944. On the status of Botrychium dissectum var. oneidense. Amer. Fern Jour. 34: 55-60. House, HOMER D. 1921. Nomenclatorial notes on certain American plants. Amer. Midl. Nat. 7: 126-135, . 1924. Annotated list of the ferns and flowering plants of New York State. New York State Mus. Bull. no. 254. WAGNER, W. H., JR. 1955. Cytotaxonomic observations on North American ferns. Rhodora 57: 219-240. . 1960a. Evergreen grapeferns and the meanings of infraspecific categories as used in North American Peeridoph- ytes. Amer. Fern Jour. 50: 32-45. . 1960b. Periodicity and pigmentation in Botry- chium subg. Sceptridium in the northeastern United States. Torrey Club Bull. 87: 303-325. . 1961. On the relative development of fertile segments in Botrychium dissectum and B. oneidense. Amer, Fern Journ. (in press). Two NEW ORCHID RECORDS FOR ONTARIO. — While orchid huntin- last year, I met Mrs. J. C. Hiegins of Komoka, Ont., who told me that she had Liparis lilifolia growing in her woodlot. I was skeptical of her identification, since this species had not been reported previously in Ontario, and since the province is not included in the range given for the species in Gray's Manual. I was invited to visit the site this year, and on June 18th, 2ccompanied by several friends, I went to Komoka and was delighted to see Liparis lilifolia (L.) Richard for the first time. It was growing in a pre- dominantly beech-maple woods on the southwest facing slope cf the Oxbow Creek, emong herbaceous species usually found in this type of woods. Mrs. Higgins told us that this year she had counted 83 plants and she first noticed the orchid in this vicinity about 10 to 15 years ago. Photographs were taken, and Prof. Montgomery took one specimen for the herbarium of the Ontario Agricultural College, Guelph. 176 Rhodora [Vol. 63 On July 10th, on my return trip from observing orchids in the vicinity of Moosonee, James Bay, I visited Timmins, Ont., and was shown Orchis purpurella 'T. & T. A. Stephen- son by Mr. Fred Cowell, local naturalist and nature photog- rapher. About 20 plants were found growing along the mar- gin of Gillies Lake in the town of Timmins. Prof. Montgom- ery and I revisited the stand on July 16th and observed that the lake shore at this point had been filled in with rock, sand and gravel to make rcom for 2 roadway. The local residents occasionally discarded waste along this roadway, and the remains of packing material was still evident among the grass and weeds which made up most of the plant cover. It was probably the source of this orchid, native to Europe and Asia. Specimens have been deposited in the herbarium of the Ontario Agricultural College, Guelph, and of the National Museum of Canada, Ottawa. — H. ANDREWS, ROYAL ONTARIO MUSEUM, TORONTO, ONT. A RANGE EXTENSION FOR PARAPHOLIS INCURVA. — On May 4, 1960, specimens of Parapholis incurva (L.) C. E. Hubb. were collected (R. J. Lemaire 650) on Freemason Island, St. Bernard Parish, Louisiana. The various manuals report this grass, introduced from Europe, as occurring in the United States along the Atlantic coast from New Jersey and Pennsylvania to Virginia, on the Pacific coast in California and Oregon (Portland), and in Michigan. Since it appeared that a significant range extension for the species may be represented by this collection, a search was made of the limited amount of pertinent literature available to me, concerned with the states bordering the Gulf of Mexico. No published record of the species for Louisana or the gulf coast was found. Subsequent correspondence with Dr. Jason R. Swallen of the U. S. National Herbarium established that two collec- tions had previously been made on the gulf coast, both in Refugio County, Texas. Swallen collected this grass between Aransas Pass and Rockport (10276), and Whitehouse col- lected it at Bayside (21209). The collection site on Free- 1961] Johnston — Haplopappus 177 mason Island is about 485 miles east of the Texas records and appears to be a first record for Louisiana. Freemason Island is a small, elongated, sand and shell island in the Gulf of Mexico about 40 miles south of Biloxi, Mississippi and about 6 miles west of the main north-south arc of the Chandeleur Islands. Several small colonies of this grass were observed on the higher (three to four feet above mean gulf level) parts of the island. Some of the associated plants were Lepidium vir- ginicum L., Oenothera humifusa Nutt., Sonchus oleraceus L., Solidago sempervirens var. mexicana (L.) Fern., and Chloris petraea Swartz. Specimens have been deposited in the herbaria of Tulane University, New Orleans, Louisiana; Southwestern Louisi- ana University, Lafayette; Florida State University, Talla- hassee; and the U. S. National Herbarium, Washington, D. C. — R. J. LEMAIRE, U. S. FISH AND WILDLIFE SERVICE, GRAND ISLAND, NEBRASKA. A NEW SPECIES OF HAPLOPAPPUS FROM MExIco. — Haplo- pappus rhizomatus sp. nov. Plantae perennes, fasciculos laxos ad 20 em. alt., ad 100 cm. lat. formantes ; rhizomata 2-8 mm. crass. longa ramosa; internodi 1-2 mm. long. Folia lineari- lanceolata, 3-4 cm. long., 2 mm., lat., spinuloso-mucronata, marginibus integris, fimbrias villosas habentibus. Pedunculi breviores quam 1 em. Capitula singula terminalia radiata, ca. 1 em. diam. Receptaculum fimbrillatum. Phyllaria macu- lis apicalibus herbaceis perspicuis subviridibus praedita. Achaenia copiose sericeo-pubescentia. Pappus e setis multis inaequis grossis rigidis antrorsaliter barbellatis luteo-albidis constans. Subcespitose, very leafy, pubescent, perennial herbs form- ing loose clumps 15-20 cm. tall and 50-100 cm. across; rhi- zomes 2-3 mm. thick, rather elongated and branching. Stems 1-2 mm. thick, arcuate-erect, the branches many, diverging at low angles (30-40°) ; internodes only 1-2 mm. long or to 7 mm. just beneath the heads, sordid-cinereous with rather closely-set antrorse villous hairs. Leaves alternate, sessile, linear-lanceolate, near midstem 3-4 cm. long but reduced to 178 Rhodora [Vol. 63 1-2 em. just beneath the heads, ca. 2 mm. broad, apically acuminate and spinulose-mucronate, basally subamplexicaul, marginally entire, tending to roll upwards on drying and with a dense fringe of villous hairs, especially toward the base, superficially sordid olive-green and sparsely appressed pubescent with antrorse villous hairs, with an inconspicuous midvein and a close inconspicuous network of dark lateral veins. Peduncles mostly shorter than 1 em. Heads solitary and terminal on each branch, ca. 1 em. in diameter, radiate. veceptacle slightly convex, 4-5 mm. in diameter, alveolate and fimbrillate with irregularly subulate-lobed, persistent, chaffy structures ca. 0.7 mm. long surrounding the points of attachment of the florets. Involucre broadly campanulate, a. 5 mm. high (entire structure reflexed after deciduation of the achenes) ; phyllaries in ca. 3 ranks, lanceolate, the inner phyllaries longest, whitish chartaceous but with prom- inent greenish herbaceous tip-spots which are produced downward into narrow green mid-lines, the outer phyllaries short and mostly green-herbaceous. Ray-florets ca. 15-20 per head, pistillate and fertile, the corolla tube narrow, ca. 2-9 mm. long, the ligule yellow, oblong to oblanceolate, 6-8 mm. long, apically shortly acute. Disk-florets 40-60 per head, perfect, the corolla yellow, narrowly tubular, ca. 4.5 mm. long, 5-toothed. Achenes of the ray-florets shorter than those of the disk but otherwise similar, 1-1.5 mm. long, trigonally or tetragonally prismatic, copiously pubescent with long silky-white stiffly ascending hairs. Pappus of the disk long- er than that of the ray-florets but otherwise similar, of 70- 100 coarse stiff antrorsely-barbellated, buffy-white bristles very unequal in length, the longest of the disk ca. 4 mm. long, and of the ray-florets ca. 3 mm. long. Gametic chromo- some number as determined by B. L. Turner from pollen mother cells, »1— 4. TYPE: Nuevo Leon, Saltillo-Matehuala highway 5 miles north of the junction of the side road to Galeana, elev. ca. 6,000 ft., J. Graham and M. C. Johnston 4203, October 8, 1959. Holotype in the herbarium of the University of Texas, isotypes in the herbaria of the Universidad Nacional de México and the University of Kansas. 1961] Ewan — Powell's Expeditions 179 The species is known only from the type collection. Only a few of these plants were seen. They grew on the road em- bankment of a highway which was scarcely five years old, indicating a relatively short age and perhaps a weedy ten- dency. The surrounding country is a high, intermontane desert plain of interior, or in places karst, drainage, with fine caleareous subsaline or alkaline soil. The plain is a vast prairie-dog town. It lies in the rain shadow of the high Cerro Potosí (ca. 12,500 ft.) and adjacent mountains, to the east. The closest relationship of this species seems to be to one of the “subspecies” of Haplopappus lanceolatus (Hook.) T. & G., as treated by H. M. Hall (The Genus Haplopappus ..., Carn. Inst. Publ. 389, pp. 114-120, 1928) showing reduction of the "inflorescence." But that our plants are substantially distinct is evident from a study of herbarium specimens and of Hall's work, and is attested to by the specialist in the genus, Dr. Ray C. Jackson of the University of Kansas, who kindly examined isotype material. The type locality and surrounding area were searched in the fall of 1960, but no further plants of this species were found. It seems desirable to place the species on record even though it is known only from the one collection. This collec- tion was made during field study in northeastern Mexico sup- ported by the National Science Foundation through NSF- G9254 at the Plant Research Institute, The University of Texas, and carried out under the direction of Dr. Calvin MeMillan. The Latin diagnosis is the work of Hannah Croas- dale. — MARSHALL C. JOHNSTON, THE PLANT RESEARCH IN- STITUTE, THE UNIVERSITY OF TEXAS, AUSTIN. CONTEMPORARY NOTES ON POWELL’S EXPEDITIONS IN THE WEsT'. — From the book's foreword we learn that this is a supplement to a chapter [ten] of Prof. Watson's Illinois Wesleyan Story: 1850-1950 (Ill. Wesleyan Univ. Press, 1950) prepared by his widow, Julia S. Watson. Had not this journalism instructor retrieved these reports of the per- 'Professor Goes West: Illinois Wesleyan University — Reports of Major John Wesley Powell's Expeditions: 1867-1874. Edited by Elmo Scott Watson. Illinois Wesleyan Univ. Press, Bloomington, 1951. 138 pp. Sold exclusively by Frank Glenn, bookseller, 627 E. 46 St., Kansas City, Mo. $3.00. 180 Rhodora [Vol. 63 sonnel, routes, and general results of the various expeditions from contemporary newspaper accounts, they would rest unnoticed in the files of the Chicago Tribune, Rocky Moun- tain News, Western Mountaineer, etc., and the forgotten issues of the Illinois Wesleyan Alumni Journal. Our botan- ical interest in this book centers around ascertaining more exact localities for the plant collections made on the various surveys of Major Powell and his associates, than those re- corded in the literature (e.g. Amaranthus pow ellii S. Wats.). The reader will be only partially successful on this score but here are recorded now in more accessible form the names of the persons who accompanied Powell and some information on their itineraries. For example, the precise identity of Mrs. Almon Harris Thompson (née Ellen Pow- ell), the sister of Major Powell, is established as the object of commemoration in the name Astragalus thompsoniae S. Wats.; the early Illinois years of George Vasey are alluded to; random notes m woven into the reports of Francis Marion Bishop (1843-1933), a student of Powell who later became professor of natural science at the University of Deseret then politician, judge, and bailiff. Incidentally, the answer to my query on the captainey of Bishop (Rocky Mountain Naturalists, 165) is answered by Ralph V. Cham- berlain, Life Sciences at the University of Utah: background and history (Salt Lake City, 1950), a useful collateral ref- erence book for the Professor Goes West and in general for the history of biological exploration in the Far West. An index is sadly missed and Prof. Watson made no ex- haustive search in the Illinois Wesleyan University records to learn the full identity of all the persons mentioned in the narrative, However, for an unhurried story of Major Powell and his parties of students in the Old West here is good reading. Certainly these expeditions were among the first field laboratory attempts in this country which set off a succession of. college-sponsored excursions. — JOSEPH EWAN, TULANE UNIVERSITY, NEW ORLEANS, LA. Volume 63, No. 749, including pages 121-150, was issued June 15, 1961. ILOW REFERENCE LIBRAR: SEP 1 1961 Dodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Associate Editors Vol. 63 July, 1961 No. 751 CONTENTS: Spore Studies in the Genus Cystopteris. I. The Distribution of Cystopteris with Non-spiny Spores in North America. BE S KOMKA E E a 181 The Genus Pteris of Costa Rica. Edith Scamman „seee. 194 Omissions in Key to Xyris in Florida. R. Kral ......................- 205 A Second Station for Stewartia ovata on the Coastal Plain. WOOD POTION LO emt UNE d ere rene E ate: 206 Che Nem England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume, Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able, All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass. Printed by THE LEXINGTON PRESS, INC. Lexington, Mass, Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 July, 1961 No. 751 SPORE STUDIES IN THE GENUS CYSTOPTERIS. I. THE DISTRIBUTION OF CYSTOPTERIS WITH NON-SPINY SPORES IN NORTH AMERICA DALE J. HAGENAH In an article entitled “An Overlooked North American Fern" the late A. H. G. Alston (1951), of the British Muse- um, called attention to a number of western collections of Cystopteris with rugose-verrucose spores (Fig. 1, C, B and D) rather than the spiny spores (Fig. 1, A) found in normal Cystopteris fragilis. Such plants had been known from Eurasia for many years and were frequently treated as a distinct species, C. dickieana Sim or C. Baenitzii Dórfl., by European botanists. In the past ten years Irene Manton (1950) has dealt with the history and cytology of such plants from Europe and Greenland; Ira L. Wiggins (1954) has compared the morphology of such plants from Alaska with that of Woodsia glabella; while D. Lóve and N. J. Freedman (1956) have published a review of the literature in regard to the nomenclature and distribution of these plants and reported a number of new localities. My own interest in the spores of this genus was the result of the finding of rugose-verrucose spores in two puzzling collections from Northern Michigan. After the publication of the Alston article I made a survey of all collections of Cystopteris from Michigan then in the herbaria of the Uni- versity of Michigan and the Cranbrook Institute of Science. Three more collections with such spores were found and re- ported (Hagenah, 1955). In the meantime non-spiny spores had been reported in material from Ontario and Minnesota by C. V. Morton (1952). In discussing Cystopteris fragilis as a subject for intensive research he wrote as follows (l.c.), 181 182 Rhodora [Vol. 63 “The whole problem has been complicated recently by Mr. Alston's report of another species, Cystopteris Dickieana, from the United States. What is this plant, indistinguish- able from fragilis morphologically (or is it?), but with spores similar to those of a Woodsia? Can a really valid FIGURE 1. A. Typical spiny spore, Lake Michigamme, Marquette County, Michigan, Hagenah 2580 (BLH). B. C. and D. Non-spiny spores: B. Steamboat Springs, Routt County, Colorado, Goodding 1625 (GH); C. Diana Bay, Hudson Strait, Quebec, Gardner 39570 (GH); D. Wiseman, Alaska, Scamman 2179 (GH). Camera lucida drawings by W. H. Wagner, Jr. species have the improbable range Scotland, Scandinavia, Siberia, Spain, Algeria, Turkey, Persia, Alaska, Alberta, California and Mexico? It seems as though “Dickieana” occurs sporadically throughout the range of fragilis. The study of this question will involve field work, as well as the examination of the spores of hundreds of specimens." That 1961] Hagenah — Spore Studies in Cystopteris 183 same year, and again in 1953, I revisited the location at Mt. Bohemia, Keweenaw County, Michigan, where I had made my first collection of such plants. Both years the plants in that colony had non-spiny spores, showing that this was the regular condition for that colony. A plant transferred from this station to the University of Michigan Botanic Gardens was studied cytologically by W. H. Wagner, Jr. (1955), and found to have the same chromosome number (n—84) as that reported by Manton for plants from Scotland, Norway and Greenland. Stimulated by Morton's comments (l. c.) and by the find- ings just described in Michigan collections, I decided to map the distribution of the non-spiny (rugose-verrucose) type of spore in North America. This work was undertaken as part of a comparative study of some American members of the genus. Some results of these studies have been reported elsewhere (Wagner and Hagenah, 1956a and 1956b). A comprehensive survey was made possible through the loan of the North American collections of Cystopteris by the Gray Herbarium. I wish to thank the staff of the Gray Herbarium for the opportunity of examining this fine series of specimens. This collection, containing over 900 sheets of specimens and including material from nearly every state and all of the Canadian provinces, provided an exceptional cross-section of the genus as it occurs in North America. In addition to the wide coverage it provided, a survey of this material was desirable because the collections from the northeastern United States and Canada had been studied and named by C. A. Weatherby (1935) during his investi- gation of the C. fragilis complex in that region. I am grate- ful to the Department of Botany of the University of Michigan for providing laboratory facilities during the preparation of the slides and especially to Dr. Warren H. Wagner, Jr., of that department, for his many suggestions and invaluable assistance, including preparation of the spore drawings. In the spore survey, preparations were made from all collections on which there seemed to be a chance that the spores were sufficiently mature for study. In some cases this resulted in slides with no mature spores although a 184 Rhodora [Vol. 63 likely pinnule had been selected by examination under mag- nification. In many cases where there were two or more well-developed plants on the same sheet, preparations were made from each, especially when there seemed to be some variation in their appearance. To make the preparations a drop of Euparal was placed on a slide, a drop or two of alco- hol placed on the spore-bearing pinnule selected, the spo- rangia and spores picked up with a needle and placed in the Euparal. Between each preparation the needle was flamed over an aleohol burner to prevent mixing. After stirring to distribute the spores in the medium, a cover-slip was added and pressure applied. While this method resulted in a fairly thick slide due to the presence of numerous sporangia, it was felt that the presence of sporangia, especially some with the spores still inside, was desirable. In a few cases some con- tamination on the herbarium sheet was found through the presence of more than one type of spore. New preparations were made in such cases. The spores of Cystopteris may be described briefly as bilateral, monolete, convex on one side, and either flat or concave on the other. The latter condition results in a *bean- shaped" profile. The outer layer, called the “sculptine” by Harris (1955) in his study of the spores of New Zealand ferns, has been found by Robert F. Blasdell (1959) to have three basic patterns of which only two, the echinate or spiny type (Fig. 1, A), and the rugose-verrucose type (Fig. 1, B, C and D), occur in North America. There is considerable variation in sculpturing within these basic types. In this study I have not as yet separated the rugose-verrucose spores into sculpturing sub-types and will refer to any of the variants of this type as “non-spiny.” When prepared in the manner just described it was found that the outer layer was generally more darkly pigmented in the non-spiny spores than in the spiny members of the fragilis complex. The outer layer seemed to be more brittle in the non-spiny spores and in a few cases cracked and flaked off under pressure. Both spiny and non-spiny spores showed a considerable tendency for the spores to fail to fill out to normal size or shape. However, the two basic types could still be distinguished for the outer layer tended to 1961] Hagenah — Spore Studies in Cystopteris 185 assume normal sculpturing in such aborted spores. Even in small, completely aborted spores from plants with spiny spores some definite spines were evident. The sculptine pattern could be determined under 100x magnification. For more detailed examination and for measurements, 430x was used. The genus Cystopteris in North America consists of C. bulbifera, C. montana, and the C. fragilis complex, the last being the most widely distributed and the most variable. In the collections which were sufficiently mature for the spore sculptine type to be determined, only spiny spores were Spiny and non-spiny spore types in Cystopteris fragilis as represented in the Gray Herbarium collections. Total Percent Percent Area Colls. Non-spiny Spiny I No. Quebec, Labrador, E. Arctic 48 52.1% 47.9% incl. Greenland II So. Quebec, New Brunswick, Nova 91 3.3 96.7 Scotia & Newfoundland III Eastern United States except Mich., 125 0.8 99.2 Wisc., & Minn. IV Great Lakes Region: Ontario, 48 16.7 83.3 Mich., Wisc., & Minn. V U. S. from Mississippi R. to the 37 5.4 94.6 Rockies VI Rocky Mt. & Pacific Coast States 162 60.5 39.5 & Western Canada VII Alaska 27 18.5 81.5 Total 538 26.4% 73.6% found in C. bulbifera (117 collections) and C. montana (17 collections). However, in the C. fragilis complex non-spiny (rugose-verruccse) spores were found in slightly more than 26 per cent of the 538 collections in which the sculptine pattern was identified. The percentage varied greatly from one geographic area to another, as shown in the table. The non-spiny spore type predominated in collections from the Rocky Mountain and Pacific Coast States and from Western Canada. Other areas of relative abundance of this type were around the Upper Great Lakes and from the Gulf of St. Lawrence north into the Arctic. In most of the states east of the Rockies the C. fragilis complex is represented for the 186 Rhodora [Vol. 63 most part by the taxa which have been designated by the varietal names mackayii, protrusa, simulans, tennesseensis, and laurentiana. Only spiny spores were found in the col- lections which had been identified as these varieties. Thus, at least in North America, the non-spiny type of spore is confined to plants which, by the characters used in the cur- rent manuals, would be identified as C. f. var fragilis. [= DENOYER-GEPPERT WALL OUTLINE SERIES NORTH AMERICA ; 7 E — | $. ~ g en 4 i FIGURE 2. Map of distribution of Cystopteris with non-spiny spores in North Ameri- ca, all localities except those in Michigan based on specimens in the Gray Herbarium. Newfoundland, the Maritime Provinces, and Gaspé were well represented in the Gray Herbarium material with a large proportion of the collections being C. f. var. fragilis. However, only three collections had non-spiny spores. All were from the vicinity of Bic, Rimouski County, Quebec, on 1961] Hagenah — Spore Studies in Cystopteris 187 the south shore of the St. Lawrence. Although not as plenti- ful as the other varieties, typical fragilis does occur in the Northeastern States. Only one collection had non-spiny spores. The locality data on the label is *Shores of Lake Champlain, N. Y." Four more stations have been found in Michigan since my prior report, but again the percentage of plants with non-spiny spores was small in proportion to the number of specimens examined. Since interpretation of the plants with non-spiny spores as a species or sub-species has been based on supposed Arctic affinities it is surprising to find that this type of spore was more abundant in collections from California (83%) than in those from either Alaska (18%) or Green- land (83%). The distribution of Cystopteris with non-spiny spores as found during this study is shown on the map (Fig. 2). Only collections examined by me have been mapped. Addi- tional records, mostly for Canada and the Arctic, may be found in the literature cited. Before the distinctive spores had been discovered the original C. dickieana was based on a characteristic frond pattern involving what has been described as “congested” pinnae and which still persists in cultivated plants descend- ed from the original stock, according to Manton. However, both Manton and Lóve note that non-spiny spores are found in plants with a diversity of leaf form. This was found true in the specimens in this study. Attempts to predict the spore type of herbarium specimens from their leaf architec- ture were incorrect more times than they were right. As described, two or more preparations were made from the same collection number where there were two or more plants. In eight such cases, plants with non-spiny and with spiny spores were found to have been collected and dis- tributed under the same collection number. In another case, a California collection (New York Falls, Amador County, Hansen 646) cited by Alston as having non-spiny spores in the specimen in the British Museum was found to have spiny spores in the Gray Herbarium specimen. This seems to indicate that the field differences between plants with the two spore types are not suflicient to prevent experienced 188 Rhodora [Vol. 63 field botanists from collecting the two types as one where they grow together. Although some collections do simulate small Woodsias, as noted by both Wiggins and Love, the tendency toward confusion between these two genera, as shown by the original specimen labels, is not confined to the plants with non-spiny spores. From blade texture of the specimens as well as the specimen data it was apparent that non-spiny spores occur in both sun and shade plants. The same is true for spiny spores. This bears out my own experience with the two types in Michigan. Considerable variation in spore size was noted early in the survey. A correlation between spore size and chromo- some number has been found to exist in the spiny-spored members of the genus in both Europe and North America so the scope of the survey was enlarged to include measure- ments of spore length. Random samples of ten spores from each slide were measured. In a sampling of more than 1,400 ncn-spiny spores the length, excluding the sculptured layer, was found to vary from 27 mu to 55.5 mu. This spread is nearly as great as that found for the three-leveled polyploid series in the spiny-spored types in the eastern United States. That series includes diploids (n—42), tetraploids (n—84), and hexaploids (n—126), of which, in Michigan material, the varieties protrusa, mackayii, and laurentiana are ex- amples of the three levels. In the non-spiny spores the aver- age size for the majority of collections falls within the sizes found for the tetraploid varieties of spiny-spored fragilis. The spores of the Mt. Bohemia, Michigan, plants which were investigated cytologically fall in this size class and the chromosome counts showed the plants to be tetraploid. This suggests that there may well be a three-level series in the non-spiny types (Fig. 1, B, C and D). With the exception of one collection from the Mistassini region of Quebec, all of the specimens with spores small enough to indicate a pos- sible diploid condition were from the western United States and Alaska, but not concentrated in any one area. On the other hand, plants with spores large enough to indicate a possible hexaploid condition were mostly from Canada and Alaska, nearly all outside the range of currently known hexaploids in the spiny-spored species. More study is needed 1961] Hagenah — Spore Studies in Cystopteris 189 on the average size and variation in the non-spiny spores of plants for which the chromosome number is known. This will be handled best by growing the plants either from living rhizomes or from recent collections of spores, preferably the former. Cultivation of plants from various localities is desirable also for another phase of the problem, the investigation of the various types of sculpturing. Some collections seem some- what intermediate between spiny and non-spiny spores, and Larsen (1952) has reported intermediates from Greenland. More cytological investigation and possibly even experi- mental hybridization may be necessary before we can deter- mine the relation of the non-spiny spored plants to the fragilis complex. Research on the spore size and sculpturing pattern prob- lems is limited by the fact that plants with non-spiny spores are not readily obtainable because they cannot be distin- guished except by microscopic examination of the spores, a test not easily applied under even the best of field condi- tions. However, in June, 1960, I was able to obtain about twenty such plants from two Michigan localities. The living plants were obtained by random sampling along transects in stations for which I was fortunate in having very pre- cise locality data. At one of these stations the sampling yielded about one-third plants with non-spiny spores while the other had a small but apparently pure stand of such plants, although plants with spiny spores were found only a few yards away. I would be glad to receive either living plants or collections with mature spores from other parts of the range. SPECIMENS WITH NON-SPINY SPORES EXAMINED DURING THIS STUDY All specimens cited are in the Gray Herbarium (GH) with the ex- ception of those from Michigan which are in the Herbarium of the University of Michigan (MICH) or the herbarium of Cranbrook Insti- tute of Science (BLH). NEW YORK: Shores of Lake Champlain, F. H. Horsford, June 1882. MICHIGAN: Alpena County: near Bolton, Hagenah 4506 (BLH); Chippew County: Near Drummond, Drummond Island, Mc- Vaugh 11360 (MICH, BLH) ; Huron County: Port Austin, C. A. Davis (MICH); Keweenaw County: Cliff Mine, near Phoenix, Hagenah 190 Rhodora [Vol. 63 3046; Mt. Bohemia, Hagenah 2001, 2003, 3021, Hagenah & Hall 777 (BLH); Marquette County: Huron Mountains, Hagenah 4014 (BLH); Partridge Island, Lake Superior, A. Dachnowski (MICH) ; Ontonagon Councy; Porcupine Mountains, Hagenah 1166 (BLH). WISCONSIN: Ashland County: Vogt Knob, Fassett 9220. MINNE- SOTA: Cook County: Grand Portage, Pease & Bean 26364. SOUTH DAKOTA: Mead County: Near Tilford, Palmer 37331. NEBRASKA: Thomas County: Plummer Ford, Dismal River, Rydberg 1452 (in part). MONTANA: Little Belt Mountains, Scribner 445; White Sulfur Springs, Scribner 443; Big Fork, Flathead Lake, Mrs. Jos. Clemens, Aug. 5, 1908; Gallatin County: Cottonwood Creek, Suksdorf 552 (in part). IDAHO: Blaine County; Near Martin, Macbride & Payson 3052; Elmore County: Upper Trinity Lake, Hitchcock & Muh- lick 10368; Latah County: Kendrick, Henderson 4791; Nez Perces County: Valley of Peter Creek, Sandberg, MacDougall & Heller 119; Owyhee County: Hot Hole, East Bruneau, Nelson & Macbride 1905. WYOMING: Yellowstone Falls, Rydberg & Bessey 3506 (in part); Laramie Hills, Nelson 9035; Lineoln Gulch, Nelson 2606; Owens Creek, Bighorn Mountains, J. G. Jack (in part); Fremont County: Sweetwater River at Farson-Lander Road, Porter 4980; Lincoln County: East of Afton, Payson & Armstrong 4980; Sweetwater County: Leucite Hills, Merrill & Wilcox 474. COLORADO: Rocky Mountains, Lat. 40-41, Dr. Geo. Vasey, Powell's Colorado Exploring Expedition; no locality, Addison Brown (type of C. fragilis var. laciniata Davenport) ; Crystal Creek, Gunnison Watershed, Baker 261; Tabeguache Basin, Payson 179; Horsetooth Mountain, Crandall 3976; Castillo County: Wagon Creek, Charlotte Horner (in part); Montrose County: Paradox Creek, Walker 224 (in part); Ute, Payson & Payson 3911; Park County: South Park, Miss E. L. Hughes; Routt County: Steamboat Springs, Goodding 1625; San Miguel County: Near Trout Lake, Payson & Payson 4120. UTAH: American Fork Canon, Wat- son 1367 (in part); Beaver County: Delano Ranger Station, Beaver Canyon, Maguire 19865; Box Elder County: Drum Canyon, Raft River Range, Maguire & Holmgren 22216; Cache County: Between Tony Grove Lake and Naomi Peak, Holmgren, Walker & Drummond 3576; Grand County: LaSal Mountains, Payson & Payson 4027; Juab County: Granite Canyon, Deep Creek Mountains, Maguire & Becroft 2465; Salt Lake County: Twin Lake outlet, near Brighton, Maguire 18656. NEVADA: Washoe Mountains, Watson 1367 (in part); Elko County: Cooper Mountain, Jarbridge Mountains, Maguire & Holmgren 22386. ARIZONA: Grand Canyon of the Colorado, MacDougal 196. CALIFORNIA: Kina River, Rothrock 364; High Mountain near Donner Pass, Torrey 596; Glen Alpine, Tahoe, Smiley 200; Alpine County: Pigeon Flat, Hoover 5355; Butte County: Mrs. R. M. Austin, June 1879; Butte Creek, Jonesville, Copeland, U. of C. Plants of Calif. 602; Eldorado County: Angora Lake, Smiley 10; Inyo County: Third 1961] Hagenah — Spore Studies in Cystopteris 191 Lake, Cottonwood Lakes, Alexander & Kellogg 3335; Onion Valley, west of Independence, Alexander & Kellogg 3162; Lone Pine Canyon east of Mt. Muir, Sharsmith 3298; Los Angeles County: Bear Creek below Bear Valley Dam, San Bernardino Mountains, Ewan 4880; Mariposa County: Merced River Canyon, Ware 536; Yosemite Valley, Abrams 4459; Mono County: Conness Cirque near Saddlebag Lake, Tioga Pass Region, Mason 11439; Nevada County: Ridge south of Donner Pass, Heller 7179; Placer County: Mt. Lincoln south of Summit Valley, Heller 12931; Plumas County: Mrs. R. M. Austin, Aug. 1882; American Valley, Mrs. R. M. Austin, July 1887; Riverside County: Strawberry Valley, San Jacinto Mountains, Grant 464; San Bernardino County: Bear Valley, San Bernardino Mountains, Abrams 4873; San Diego County: Spencer Valley, near Julian, Abrams 3798; Santa Cruz County: Santa Cruz, Dr. Anderson; Tulare County: Lower Kern River Canyon, Bacigalupi & Ferris 2451; Crabtree Meadow, Culbertson, C. F. Baker Dist. 4352; South Fork Kaweah River, Culbertson, C. F. Baker Dist. 4515; Tuolumne County; Dardanelle, Alexander & Kellogg 3744; Dana Fork of Tuolumne River, Tuolumne Meadows, Sharsmith 324; Siskiyou County: Panther Creek Meadows, Mt. Shasta, Cooke 13999. OREGON: Baker County: Alder Springs, Wallowa Mountains, Jones 6612; Grant County: Dixie Mountain, Blue Mountains, John Day Val- ley, Henderson 5587; Hood River County: Henderson 762; Wasco County: Dalles of the Columbia, Major Bullies. WASHINGTON: Douglas County: Egbert Spring, Sandberg & Leiberg 351; Okanogon County: Muchamuch Lookout, Thompson 6992; Chesaw, St. John, Courtney & Parker 5064; Pend Oreille County: Z Canyon, St. John 6469; Spokane County: Bank of Spokane River, opposite Fort Wright, Jennings & Jennings 8132; Cheney, Mrs. Susan Tucker; Newman Lake, Jennings & Jennings 8519; Walla Walla County: Waitsburg, R. M. Horner, May 1897. ALASKA: Rapids Lodge, Richardson Highway, Scamman 4; Eagle Summit, Steese Highway, Scamman 1970B; Wise- man, Scamman 2179; Nome, Anvil Creek and Dexter Creek, Seward Peninsula, Porsild & Porsild 1301; Camp Eilson, Mt. McKinley National Park, Nelson & Nelson 4100 (in part). GREENLAND: Uniiorfik Fjord, Vestside, Niaqornaq, M. P. Por- sild, Sept. 1934; Agpatsiait, 71? 5' N., M. P. Porsild, July 1935; Gothaab, Wetherill 31. ELLESMERELAND: Fram Harbour, H. G. Simmonds, July 1889; Harbour Fjord, Simmonds 2553. BAFFIN LAND: Lake Harbour, Malte 463; Cape Dorset, Malte 532. LABRADOR: Razorback Har- bor, Torngat Region, Abbe 9; Valley of the Bryant Lakes, Kangalak- siorvik, Torngat Region, Abbe 8; Flint Island near Manvers, Bryant 1. QUEBEC: Rimouski County: Bic, Fernald & Collins 804, 808 and 809; Anticosti Island: Riviere de la Chute, Victorin & Rolland 27 037; Riviere Des Caps, Victorin & Rolland 27 051; Mingan Islands: Ile au Fantome, Victorin & Rolland 18090; Grande Ile, Victorin & Rolland 192 Rhodora [Vol. 63 18086; Mistassini District: Ile Andre-Michaux, Rousseau & Rouleau 201; Ile Manitounouk, Rousseau & Rouleau 9; Baie de la Chute-Cachee, Peninsule du Dauphin, Rousseau & Rouleau 1084; Pointe de Basalte, Peninsule du Dauphin, Rousseau & Rouleau 1050 and 1051; Lac Wacha- gami, Rousseau & Rouleau 1306; Opitchouane, Peninsule D'Orvel, Rousseau & Rouleau 1157; Ungava District: Boat Opening, Manitou- nok Islands, Dutilly & Lepage 12990; Cape Jones, James Bay, Gardner 391237 ; Diana Bay, Hudson Strait, Gardner 39570; Port Burwell, Hud- son Strait, Malte 121048 and 121057. ONTARIO: Manitoulin Island: Gore Bay, Pease & Ogden 25014; West Bay, Pease & Ogden 25034; Algoma District: Garden River, Fassett 13312; Thunder Bay Dis- trict: Jackfish, Pease & Bean 23713 and 23717 (in part); Sibley Town- ship, Taylor, Losie & Bannan 22. SASKATCHEWAN: Cornwall Bay, Lake Athabaska, Raup 6573. ALBERTA: Peace Point, Wood Buffalo Park, Raup 1454; Edmonton, Moss 2701a; Nordagg, Mt. Coli- seum, Malte & Watson 1527 and 1554; Bertha Lake, Waterton Lakes National Park, Malte & Watson 2705; Jasper National Park: Pyramid and Patricia Lakes, Scamman 2789; Miette Hot Springs, Scamman 2400; Medicine Lake, Scamman 2485; Maligne Lake, Scamman 2576; Athabaska Glacier, Columbia Ice Field, Scamman 2726; Angel Glacier, Mt. Edith Cavell, Scamman 3401; Jasper, Scamman 3379. BRITISH COLUMBIA: Selkirk Mountains, Shaw 1095; Asulkan Glacier Trail, Selkirk Mountains, F. C. Prince, Aug. 1900; Carbonate Draw, Selkirk Mountains, Hacock, C. H. Shaw Dist. 285; Gorge, Carbonate Draw, Selkirk Mountains, Shaw 271; North bank of Peace River, below Wicked River, Raup & Abbe 4008; Mt. Selwyn, Raup & Abbe 3936; Hudson Hope, Peace River Valley, Raup & Abbe 3956; Alberni Region, Vancouver Island, Rosendahl 2054. SUMMARY: Plants distinguished only by the non-spiny (rugose-verrucose) sculpturing of the outer layer of their spores have been shown to be widespread and not uncommon in North America within most of the range of Cystopteris fragilis var. fragilis. This spore type was not found in plants identified as any of the Eastern North American members of the C. fragilis complex, i.e., the varieties mackayü, protrusa, simulans, tennesseensis, and laurenti- ana; in the endemic American species C. bulbifera; or in American collections of the cireumpolar species C. montana. The presence of two entirely different spore sculpturing types in plants which cannot be distinguished by any cur- rently known field characters or ensemble of characters seems most remarkable. However, the recognition of species on the grounds of spore sculpturing alone does not seem 1961] Hagenah — Spore Studies in Cystopteris 193 justifiable at this time. The significance of spore pattern as a taxonomic character in this genus and the relationships between plants of the two spore types are problems which are likely to be resolved only by such techniques as experi- mental hybridization and the cytological study of the re- sulting progeny. — CRANBROOK INSTITUTE OF SCIENCE, BLOOMF.ELD HILLS, MICHIGAN. LITERATURE CITED ALSTON, A. H. G. 1951. An overlooked North American fern. Am. Fern Jour. 41: 76-78. BLASDELL, R. F. 1959. A monographie study of the fern genus Cystopteris. Doctoral thesis, Department of Botany, University of Michigan, and available on microfilm from University Micro- films, Inc., Ann Arbor, Mich. HAGENAH, D. J. 1955. Notes on Michigan Pteridophytes, I. New county records in Osmundaceae and Polypodiaceae. Am. Fern Jour. 45: 65-80. Harris, W. F. 1955. A manual of the spores of New Zealand Pteri- dophyta. New Zeal. Dept. Sci. and Indus. Res. Bul. 116. LARSEN, K. 1952. Udbredelsen i Grónland af Cystopteris fragilis coll. med piggede og vortede sporer. Bot. Tidsskr. 49: 39-44. LóvE D., AND FREEDMAN, N. J. 1956. A plant collection from Southwest Yukon. Bot. Not. 109: 153-211. MANTON, I. 1950. Problems of cytology and evolution in the Pteri- dophyta. Cambridge Univ. Press. MoRTON, C. V. 1952. A suggestion for a cooperative study by mem- bers of the American Fern Society. Am. Fern Jour. 42: 31-35, WAGNER, W. H., JR. 1955. Cytotaxonomic observations on North American ferns. Rhodora, 57: 219-240. WAGNER, W. H. JR., AND HAGENAH, D. J. 1956a. A diploid variety in the Cystopteris fragilis complex. Rhodora, 58: 79-87. 1956b. Observations on some bulblet-producing populations of the Cystopteris fragilis complex. Am. Fern Jour. 46: 137-146. WEATHERBY, C. A. 1935. A new variety of Cystopteris fragilis and some old ones. Rhodora, 37: 373-378. WiGGINS, I. L. 1954. Cystopteris dickieana and Woodsia glabella in Arctic Alaska. Am. Fern Jour. 44: 97-108. THE GENUS PTERIS OF COSTA RICA EDITH SCAMMAN This paper on Pteris is the second of my studies of a genus of Costa Rica ferns. The first on Adiantum was pub- lished in Contributions from the Gray Herbarium 187: 3-22. 1960. This small country is truly a fern lover's paradise, and the days I spent collecting there during February and March, in 1951, 1953, 1955, and 1956, proved to be rewarding and worthwhile. ! Due to the large size and the multiple and complex branch- ing of some species of Pteris it is often difficult to obtain satisfactory specimens, and frequently only the extreme upper portion of a frond or a single pinna is found on herbar- ium sheets. The specimens that have been seen and cited here are from the U. S. National Herbarium, the Gray Herbarium, and a few from the New York Botanical Garden. They are arranged geographically and listed under the seven provinces into which Costa Rica is divided — Limon on the Atlantic, Guanacaste and Puntarenas on the Pacific, and Heredia, Alajuela, San José and Cartago in the interior of the country. The habitat and altitude given for each species refer only to the specimens seen from Costa Rica. PTERIS L. A large genus of mostly coarse ferns, herbaceous to cori- aceous, with erect or ascending fronds. Blades 1-4-pinnate, often only the basal parts decompound; veins all free, or only the basal pair joined, or joined in several rows of angular areoles. Sori linear and continuous, but not usually reach- ing the apices and sinuses of the segments, sporangia borne 1I was greatly aided during my experiences in the field by Dr. Leslie R. Holdridge of the staff of the Interamerican Institute of Agricultural Sciences at Turrialba, who was most generous in sharing with me his time and his knowledge of the country, and in providing means of transportation for procuring desired specimens. To Dr. Rolla M. Tryon of Harvard University I am deeply indebted for his con- tinued interest in this project and for his stimulating suggestions and help in many ways. The drawings were prepared by Mrs. Joyce Todd, most of them for Dr. Tryon's Ferns of Peru. 194 1961] Scamman — Pteris of Costa Rica 195 on a marginal connecting-vein, protected by the membran- ous, reflexed margin, which serves as an indusium. KEY TO SPECIES a. Pinnae (at least above the basal ones) entire, narrowly linear, not lobed or pinnatifid. b. b. Fronds large, usually over 1 m. long; veins freely areolate toward VITERTHCL EDU ES in gsnicasnchspsntawestiuva pk tara aA xe E BUS CERE 1. P. grandifolia. b. Fronds small, rarely up to 1 m. long; veins all free..................... aub de MR IE E 2. P. cretica. a. Pinnae (at least above the basal ones) deeply pinnatifid or more deeply divided. c. c. Veins all free. d. d. Basal pinnae deeply pinnatifid beyond the basal pinnules. e. e. One or two veins between costules arising from the costa of the pinna (at least toward base of pinna). Fig. 4. f. f. Sinuses mostly asymmetrical, pinna axils smooth to Slehtlvamuricate eee ie os 3. P. paucinervata. f. Sinuses uniformly symmetrical, pinna axils strongly muricabtes e o EMI i ee CUI 4. P. pungens. e. Veins all arising from the costule (of the pinnule). Fig. 5b. X ad ML KR ME EE 5. P. quadriaurita. d. Basal pinnae regularly pinnate-pinnatifid beyond the basal pinnules. g. g. Leaf-tissue coriaceous; segments acute and mucronate; short, firm awns on the upper surface of the costae; rachises and costae muricate beneath................... eese 6. P. muricata. g. Leaf-tissue soft, herbaceous; segments obtuse, crenulate at the apex; long, soft, whitish awns on the upper surface of the costae; rachises and costae smooth beneath................ penc EM NEM KNEE S E 7. P. muricella. c. Veins joined (at least along the costae; fig. 8b). h. h. Basal pinnae pinnatifid beyond the basal pinnules; basal veins joined in a narrow costal arc, the others free.............. 8. P. biaurita. h. Basal pinnae pinnate-pinnatifid (or more divided) beyond the basal pinnules; veins joined in several rows of angular areoles. i. i. Basal veins monoarcuate, one large areole along the costa be- tween the costules. j. j. Pinnatifid pinnae with the herbaceous tissue decurrent onto the rachis, especially in the apical ones............ 9. P. propinqua. j. Pinnatifid pinnae with the herbaceous tissue not decurrent onto the rachis. k. k. Apex of ultimate segments crenulate. 1. 1. Ultimate segments usually 5-10 cm. long (rarely 3 em.), pinnatifid pinnae or pinnules definitely petiolu- late, usually once pinnate at the base ........ 10. P. livida. 196 Rhodora [Vol. 63 1. Ultimate segments usually about 1 em. long (rarely to 2 cm.), pinnatifid pinnae or pinnules subsessile or some shortly petiolulate, usually pinnatifid to the base. RR 13. P. tripartita. k. Apex of ultimate segments sharply serrate. Segments numerous, close, oval-falcate, leaf-tissue coriaceous........ EE 12. P. podophylla. i. Two or more long areoles along the costa between costules. Segments lance-attenuate, oblique, sterile tips sharply serrate, leaf-tissue firm-herbaceous................. see 11. P. altissima. 1. PTERIS GRANDIFOLIA L. Sp. Pl. 2: 1073. 1753 A large fern, simply pinnate, with long narrow pinnae, mostly oblique, with entire thinly cartilaginous margins and sori continuous nearly to the attenuate tips. The veins are close and parallel, free near the costa, but anastomosing toward the margin. The rhizome is stout and creeping; the texture of the pinnae is membrano-herbaceous and trans- lucent. Mexico to Panama, to Peru; West Indies. Specimens seen: ALAJUELA: Cebadilla, Valerio 278 (us); Gorges of Machuca River near San Mateo, Biolley 2019 (vs), Rio Machuca, Biolley 17389 (GH, NY, US). 2. PTERIS CRETICA L. Mant. 130. 1767 This well-known fern of wide distribution is characterized by the lower pair or pairs of pinnae that are forked nearly to the base into long, narrow attenuate segments. The mar- gins of the sterile pinnae and tips of the fertile ones are usually spinulose-serrated. Tropical and subtropical regions in many parts of the world, either native or naturalized. Rocky woods and forested slopes from 1600 to 1900 m. Specimens seen: CARTAGO: La Banderilla, R. Torres 243 (US); Reventado, Lankester 721 (us). 3. PTERIS PAUCINERVATA Fée, Mém. Fam. Foug. 8: 73. 1857 The stipe of this seemingly rare fern is reddish brown, erect from a short ascending rhizome, which has long con- colorous brown scales. The terminal and lateral pinnae, usually 5-7 pairs, end in a long caudate tip; the linear- 1961] Scamman — Pteris of Costa Rica 197 oblong segments are decurrent at the base, the sterile ones with crenulate margins, the fertile ones slightly falcate. The lowest pinnae are irregularly forked; the texture of the oe ~ OO - [vec aC CC <¥ t LM Fic. 1-4. Fig. 1. P. grandifolia: la, a fertile pinna, X 1/4: 1b, portion of fertile pinna, X 1; le, portion of sterile pinna, X 1. Fig. 2. P. cretica: a fertile frond, x 1/2. Fig. 3. P. paucinervata: 3a, a fertile pinna, X 1/2; 3b, a fertile segment, X 1. Fig. 4. P. pungens: base of sterile pinna, X 3,4. blade is subeoriaceous. Veins are few (hence the name) and conspicuous, well-spaced at the base; some of them arise from the costa. This fern is variable and rare; perhaps it is a hybrid of P. pungens and P. muricella (or others). Mexico to Panama. Ravines in forests and on slopes of volcanoes from 1300 to 1800 m. Specimens seen: HEREDIA: Vara Blanca de Sarapiquí, between Poás 198 Rhodora [Vol. 63 and Braba volcanoes, Skutch 3636 (US). SAN JOSE: Tablazo, Valerio 265 (US). 4. PTERIS PUNGENS Willd. Sp. Pl. 5: 387. 1810 The erect stipe, which is castaneous at the base, is slightly muricate, as is the rachis especially at the axils of the pin- nae. The pinnae are from 2-5 pairs, with only the lowest 2-partite. The linear-lanceolate obtuse segments are close, parallel, and have a regular appearance as compared with those of P. paucinervata. As the rounded or pointed tubercles on the stipe and rachis can be observed in some other species, the distinguish- ing character of this Pteris is found in the short lowermost vein or veins which arise from the costa. The linear sori extend from the sinuses of the segments almost to its ser- rate tip. Mexico to Panama, to Peru and Bolivia; West Indies. In humid forests from 200 to 1100 m. Specimens seen: LIMON: Tsaki, Talamanca, Tonduz 9471 (Us); Finca Montecristo, on the Río Reventazón below Cairo, Standley & Valerio 48626 (Us); Los Diamantes, Rubber Plant Station, Scamman 5908, 7036 (GH); Hacienda Parismine Banana Co., Jimenez 1060 (NY, US). HEREDIA: Santa Clara Las Delicias, Biolley 10683 (US); Finca La Selva, Río Puerto Viejo, Scamman & Holdridge 7443, 7915 (GH). ALAJUELA: Llanuras de San Carlos, Brade 314 (Us); Surubres near San Mateo, Feb. 1906 Biolley (NY, US). SAN JOSE: San José, 1906 Biolley (US); Vicinity of El General, Skutch 2200 (GH, NY, US). CARTAGO: Turrialba, Maxon 182 (NY, US). 5. PTERIS QUADRIAURITA Retz. Obs. Bot. 6: 38. 1791 This common, widely distributed fern varies greatly in size from 15 em. to 1 m. The leaf has 7-15 pairs of pinnae, and the basal pair are 2-partite; it is suberect from a woody rhizome clothed with small, acicular dark scales with lighter margins. The segments are oblong to linear, rounded- obtuse, thin to firm-herbaceous and translucent. Mexico to Panama, to Brazil; West Indies. Tropics of Old World. On shaded stream banks and in deep ravines from 1000 to 1800 m. Representative specimens: 1901-1905 Wercklé (NY, US). LIMON: 1961] Scamman — Pteris of Costa Rica 199 Talamanca, Tonduz 8579 (US). HEREDIA: Confluence of Rio Puerto Viejo and Sarapiquí, Pittier 7475 (US); Barba, Scamman 7038 (GH). ALAJUELA: Alajuela, Alfaro 6045 (GH, US); La Verbena prés Alajue- lita, Tonduz 8791 (US); Near Zapote, Scamman 7622 (GH); San Ramon, Tonduz 17581 (NY, US). SAN JOSE: Aserri, Hunnewell 16535 (GH); San Sebastian near San José, Standley 49294 (US); Rio Torres, Alfaro 12, 18 (US); Forêts du Copey, Toxduz 11684 (Us); Santa Ana, Scamman 5906 (GH); Finca Ortuna, Desamparados, Scamman & Hold- ridge 7916 (GH). CARTAGO: Navarro, R. Torres 58 (us); Juan Viñas, Cook & Doyle 187, 200, 241 (Us); Cerro de La Carpintera, Standley 34270 (US); Turrialba, Scamman 5904, 5907 (GH) ; Tapanti, in Valley of Río Reventazón, Scamman & Holdridge 7917 (GH). PUNTARENAS: Cours superieur du Diquis, Pittier 10571 (Us). 6. PTERIS MURICATA Hook. Sp. Fil. 2: 193. 1858 Commonly called P. coriacea Desv. (a South American species). The stipe which is dark chestnut at the base arises from a thick, woody rhizome with dark scales with light dentate margins. Both stipe and rachis are flexuous with the coria- ceous pinnae stalked at the base, the lower pair bi- to tri- pinnate. Segments are linear-oblong, subfalcate, mucronate at the tip. This species is characterized by the hard spinous points on the rachises and costae. Costa Rica to Panama, to Peru and Bolivia. In wooded ravines and moist forests at high altitudes from 1300 to 3000 m. Specimens seen: Costa Rica 1901-1905 Wereklé (us), Vicinity of Coliblaneo, Maxon 314 (NY, US). HEREDIA: Volcán Barba, R. Torres 229 (US). ALAJUELA: Candelaria, Hoffmann 889 (US). SAN JOSE: Foréts du Copey, Tonduz 11898 (Us); La Hondura, Standley 37721 (GH, US), Along the road to La Hondura, Scamman & Holdridge 7912 (GH). CARTAGO: Cerro de La Carpintera, Standley 34489 (us); Robert’s on the road to Volcán Irazü, Scamman 5903 (GH). 7. PTERIS MURICELLA Fée, Mém. Fam. Foug. 8: 73. 1857 Picris mollis Christ, Bull. Herb. Boiss. 4: 658. 1896. Costa Rica: Foréts de San Marcos, Tonduz 7565; isotype US! This fern is unusual and distinctive among Costa Rican Pteris because of the soft and delicate tissue. The smooth, reddish-brown stipes rise erect from a thick rhizome with concolorous brown scales ; segments are oblong, obtuse, cren- ulated at the apex, with the sori occupying only the middle 200 Rhodora [Vol. 63 of the lobes. Long soft awns are common on the costae on their upper surface. Mexico to Panama. In moist dense forests from 1000 to 1800 m. Specimens seen: 1901-1905 Wercklé (US). ALAJUELA: La Palma, near San Ramón, Brenes 5364 (GH, US) ; Santiago, near San Ramón, Tonduz 17582 (us); Zarcero, Jan, 19 & 30, 1948, Austin Smith (US). SAN JOSE: Forêts de San Marcos, Tonduz 7565 (US); Vicinity of Santa María de Dota, Standley 41863 (GH, US) ; Cultures du Copey, Tonduz 11705 (US); Vicinity of El General, Skutch 2240 (GH, NY, US). CARTAGO: Valle del Río Navarro, Wercklé 16771 (US). 8. PTERIS BIAURITA L. Sp. Pl. 2: 1076. 1753 The stipe and rachis are light-colored, rising from a sub- erect, woody rhizome; the blade may have from 5-15 pairs of opposite pinnae, the basal pair 2-partite ; segments oblong to linear, obtuse, with entire margins, the fertile with rounded sinuses, the sterile acute. In this species only the basal veins are joined in a narrow costal are. Mexico to Panama to Brazil; West Indies. Tropics of Old World. In light woods at edge of fcrest; wet thickets, from 100 to 1000 m. Specimens seen: LIMON: La Colombiana Farm of United Fruit Co., Standley 36715 (us); Port Limón, June 15, 1874, Kuntze (NY). ALAJUELA: Surubres prés de San Mateo, Pittier 7009 (US), Surubres near San Mateo, Feb. 1906 Biolley (NY, Us); La Palma de San Ramón, Brenes 6396 (NY) ; San Pedro de San Ramón, Brenes 15085 (NY). SAN JOSE: El General, Skuteh 2203, 2224, 3931 (GH, NY, US). GUANACASTE: Nicoya, Cook & Doyle 678 (US). 9. PTERIS PROPINQUA Ag. Rec. Spec. Gen. Pterid. 65. 1839 Pteris costaricensis Rosenst. Fedde Rep. Spec. Nov. 22: 7.1925. Costa Rica, Brade 461; isotype NY! Stipe and rachis smooth, light-colored to the base; scales on the rhizome with dark centers; frond bipinnate below with especially the upper pinnae decurrent to the rachis; segments lanceolate, subfalcate, rather obtuse, mucronate and sharply serrated. Basal veins monoarcuate. Mexico to Panama, to Brazil; West Indies. In other Central American countries it has been found in open places and swamps from sea level to 300 m. 1961] Scamman — Pteris of Costa Rica 201 Specimen seen: ALAJUELA: Llanuras de San Carlos, Brade 461 (NY). 10. PTERIS LIVIDA Mett. Ann. Sc. Nat. V. 2: 222. 1864 Blades tripartite, the pinnatifid pinnae petiolulate, the long segments usually cut to the rachis at the base of their pinna; sori not reaching the inciso-crenate tips. The vivid green color of the leaf-tissue is distinctive. Costa Rica to Peru and Bolivia. MIAT | 9 / f fi ^ f) Fia. 5-7. Fig. 5. P. quadriaurita: 5a, a fertile basal pinna, X 1/2; 5b, portion of fertile pinna, X 1 1/2. Fig. 6. P. muricata: 6a, a fertile pinna, X 1/2; 6b, portion of a pinna with muricated points on costa beneath, X 2. Fig. 7. P. muricella: Ta, a fertile pinna, X 3/4; Tb, portion cf a pinna with awns on costa above, X 1. 202 Rhodora [Vol. 63 In ravines and damp forests in mountains from 900 to 2100 m. Specimens seen: 1901-1905 Wercklé (US). Vara Blanca between Poás and Barba, Scamman 7035 (GH); Along cart-road from Vara Blanca to La Concordia, Maxon & Harvey 8482 (US). SAN JOSE: El Copey, H. E. Stork 1549 (US); La Palma, Brade 26 (Us); San Jerónimo, Wercklé 580 (US). CARTAGO: Santa Clara de Cartago, Lan- kester 711 (GH, US); Foréts du Turrialba, Pittier 849 (US); Foréts du Roble, Massif de l'Irazü, Pittier 4179 (US). 11. PTERIS ALTISSIMA Poir. in Lam. Encycl. 5: 722. 1804 Pteris Kunzeana Ag. Rec. Spec. Gen. Pterid. 62. 1839. This large fern, sometimes reaching a height of 2 m., is the most common and widely distributed Pteris in Costa Rica. The blade is deltoid-ovate, broad, to tripinnate at the base, the basal pinnae are much the largest. The straw- colored stipe rises from a short, erect rhizome with shiny castaneous scales with a lighter dentate margin. The leaf- tissue is firm-herbaceous to subcoriaceous. The pinnae and pinnules vary greatly in shape and gen- eral aspect in different fronds, but the character which dis- tinguishes this from other similar species of Pteris is the 2-3-arcuate basal veins. Mexico to Panama, to Brazil and Bolivia; West Indies. In ravines and on hillsides in wet forests from 250 to 2200 m., from all the provinces of Costa Rica. Representative specimens: LIMON: Foréts de Tsaki, Talamanca, Tonduz 9440 (US); Los Diamantes, Holm & Iltis 369 (NY, US), Los Diamantes, Scamman 7034 (GH). HEREDIA: Yerba Buena, northeast of San Isidro, Stanley & Valerio 49235 (GH, US); La Concepción, Llanu- ras de Santa Clara, J. D. Smith 6870 (GH, US) ; Vara Blanca de Sarapi- qui, Skutch 3578 (GH, NY, US), Vara Blanca between Pods and Barba, Maxon & Harvey 8337 (US) ; Cinchona, Scamman 7619 (GH); La Paz- Waterfall, Scamman & Holdridge 7910 (Gn). ALAJUELA: Region of Zarcero, Austin Smith 383 (GH); Surubres prés San Mateo, Biolley 6 (US). SAN JOSE: Las Nubes, Scamman & Holdridge 7908 (GH); La Palma, Scamman 7618 (GH), Maxon & Harvey 7999 (US); Cerro Turrubares, Orotina, Jimenez 600 (US); Finca Ortuna, Desamparados, Scamman & Holdridge 7909 (GH). CARTAGO: Estrella, Cooper 6044 (GH, NY, US); Navarro Valley, H. E. Stork 1405 (GH); San Juan del Norte, Scamman 7620 (GH); Foréts de Juan Vinas, Jan. 25, 1890, Pittier (US), Juan Viñas, Cook & Doyle 212 (US). GUANACASTE: Cafetales at Hacienda Granadilla, Dodge & Thomas 6421 (GH); Upper 1961] Scamman — Pteris of Costa Rica 203 INS IU NS j N 171 SN uut. NNN NN Fic. 8-13. Fig. 8. P. biaurita: 8a, a fertile pinna, X 1/2; 8b, portion of a fertile pinna, X 1 1/2. Fig. 9. P. propinqua: a fertile pinna, X 3/4. Fig. 10. P. livida: a fertile pinna, X 1/4. Fig. 11. P. altissima: a fertile pinna, X 1/2. Fig. 12. P. podophylla: a fertile pinna, X 1/2. Fig. 13. P. tripartita: a fertile pinna, X 1/2. 204 Rhodora [Vol. 63 slopes of Cerro San José de Libano, Dodge, Hanckel & Thomas 7878 (GH). PUNTARENAS: Between Golfo Dulce and Río Terruba, Skutch 5412 (US). 12. PTERIS PODOPHYLLA Sw. Schrad. Journ. 1800.? 67. 1801 A tall conspicuous fern often growing at the edge of a forest, in the mountains. The wide frond is ternately di- vided, the lateral divisions 3-4-partite, the central one pin- nate-pinnatifid. The stipe is stout, thick and tawny, about 1 m. high, and somewhat muricated at the base. The linear- oblong falcate segments are regularly spaced on the costae with rounded sinuses, the sterile ones are finely spinulose- serrate as are the fertile near their tip. Texture is thick, coriaceous, and the basal veins are monoarcuate. Mexico to Panama, to Peru and Bolivia; West Indies. In partial shade at edges of forests and in clearings usually from 1000 to 2400 m. Specimens seen: 1901-1905 Wercklé (US). HEREDIA: Confluence of Río Puerto Viejo and Sarapiquí, Pittier 7491 (US); Cart road from Vara Blanca to La Concordia, Maxon & Harvey 8475 (US). | SAN JOSE: Santa María de Dota, Standley 43335 (GH, US) ; Foréts du Copey, Tonduz 11718 (US), El Copey, Jimenez 1107 (GH, US); Dans les bois humides à la Palma, Pittier 710 (US), Vicinity of La Pa!ma, Maxon & Harvey 8083 (Us); Highway near La Chonta, Scamman 5905 (GH); Rancho Redondo, on slope of Iraza, Scamman & Holdridge 7621 (GH). CARTAGO: Forêts del Roble, Massif de l'Irazü, Tonduz (Pittier) 4189, 4190 (NY, US) ; Foréts entre la Turrialba à la Río Birrio, Pittier 845 (US) ; Finca Navarro, Maxon 629 (NY, US). 19. PTERIS TRIPARTITA Sw. Schrad. Journ. 1800.2 67. 1801 This fern has often been described as "gigantic" or giant bracken. The leaves reach 2 m. or more in height from a very stout rootstalk. Fronds are tripartito, the pinnatifid pinnae more or less sessile. The short ultimate segments are linear-oblong, falcate, obtuse or acute, crenulate only at the tip, and regularly spaced and evenly cut by obtuse sinuses usually at some distance from the costa. A widely distributed fern, native to the Tropics of the Old World — Asia, Africa, Australia, Polynesia, etc. — but found as an escape occasionally in the New World, including cypress swamps and wet hammocks of Florida, and rarely in Central and South America. 1961] Scamman — Pteris of Costa Rica 205 Specimens seen: CARTAGO: Tapanti, in Valley of Río Reventazón, Scamman & Holdridge 7913 (GH); Orosi, Scamman & Holdridge 7914 (GH). Other species of Pteris described from Costa Rica are the following. They are probably synonyms of some of the species mentioned here, but the tynes have not been seen. P. longicaudadel Christ, in Pitt. Prim. Fl. Costar. 3: 21. 1901. Costa Rica, Foréts de Santo Domingo de Osa, Mars. 1896, Tonduz 10071. Cited as a synonym of P. pungens by Maxon. Sei. Sur. Porto Rico and V. I. 6: 434. P. macrodictya Christ, Bull. Herb. Boiss. II, 7: 267. 1907. Costa Rica, 1904, Wercklé. Probably = P. grandifolia. P. navarrensis Christ, Bull. Soc. Bot. Geneve II, 1: 227. 1909. Costa Rica, Valle del Río Navarro, 1400 m., Wercklé 16761. Probably = P. podophylla. P. prolifera J. E. Bomm., Bull. Soc. Bot. Belg. 35: 189. 1896, in synonymy. Costa Rica, Juan Vinas, Pittier 1841. It is a proliferous phase of P. quadriaurita or perhaps of a species of another genus. — GRAY HERBARIUM, HARVARD UNIVERSITY. OMISSIONS IN KEY TO XYRIS IN FLORIDA. — In Rhodora, Vol. 62, No. 743, 1960, two portions of the key to Florida Xyris were omitted from the manuscript, through an over- sight. On page 300, there should be a second 3 just above the first 4 which should read “3. Plants tuberculate-rough- ened only on scape ridges and/or leaf margins." On page 301, the second “15” should end with 14. X. smalliana. — R. KRAL. 206 Rhodora [Vol. 63 A SECOND STATION FOR STEWARTIA OVATA ON THE COASTAL PLAIN. — Grimes (1922) reported an extensive population of Stewartia ovata (Cav.) Weath. in James City County, Virginia, south of Williamsburg: Grimes 3818 (NY), June 24, 1921. Baldwin 14951 (GH) is from that same pop- ulation, June 29, 1954 ; the label has the following notations: "Small trees to eighteen feet. Petals white. Filaments pur- ple, rose, and yellow in various different plants." Coe (1959) mentions the variability of the plants at this station. The geographic range of S. ovata was mapped by Braun (1937) : the Williamsburg station seems to have been the only one known for the Coastal Plain (Kobuski, 1951). I recently found S. ovata in Lancaster County, Virginia, Harley 2265 (GH), June 18, 1960, a small population on the east bank of John Creek, a short distance above its conflu- ence with the Corrotoman River at Merry Point. Shrubs grew up to six feet in height. All the flowers observed were white with yellow stamens. A fruiting specimen was taken on Aug. 25th: Harley 2272 (GH). Associated with Stew- artia at this station are Galax, Kalmia, Asarum virginicum, ete. Several small colonies of Epigaea are near-by. — WINI- FRED J. HARLEY, MERRY POINT, VIRGINIA. LITERATURE CITED BRAUN, L. Lucy. 1937. Some relationships of the flora of the Cum- berland plateau and Cumberland mountains in Kentucky. Rhodora 39: 193-208. CoE, FREDERICK W. 1959. Stewartia malacodendron versus Stew- artia ovata grandiflora. National Horticultural Magazine 38: 175- 176. GRIMES, E. J. 1922. Some interesting plants of the Virginia Coastal Plain. Rhodora 24: 148-152. KoBuskKI, C. E. 1951. Studies in the Theaceae, XXI. The species of Theaceae indigenous to the United States. Journal of the Arnold Arboretum 32: 123-138. Volume 63. No. 750, including pages 151-180, was issued July 13, 1961 W REFERENCE LIBRARY SEP 13 1961 JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Associate Editors Vol. 63 August, 1961 No. 752 CONTENTS: Variance in Herbarium Specimen Identification and other Considerations Based Upon the Preparation of a Local Flora. Robert A. Davidson and Phyllis E. Davidson. .......... 207 A New Zephyranthes from Southern Texas. FIMogWERHOnes er E IURI cede 214 Merger of the North American Houstonia and Oldenlandia under Hedyotis. Walter H. Lewis ............ eene 217 Draba aprica in Oklahoma. Reed C. Rollins ............................. 223 Gisekia pharmacioides, a New Weed. R. K. Godfrey .............. 226 Heterostyly and Homostyly in Lithospermum canescens (Boraginaceae). H. G. Baker ................... ene 229 A New Species of Panicum from New Jersey. BEEN UM L..............e thoro eros E on BOO The New England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume, Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able, All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver’s blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass, Printed by THE LEXINGTON PRESS, INC. Lexington, Mass, Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 August, 1961 No. 752 VARIANCE IN HERBARIUM SPECIMEN IDENTIFICATION AND OTHER CONSIDERATIONS BASED UPON THE PREPARATION OF A LOCAL FLORA ROBERT A. DAVIDSON AND PHYLLIS E. DAVIDSON “A universal, to Aristotle, is any common noun, any name capable of universal application to the members of a class: so animal, man, book, tree, are universals. But these univer- sals are subjective notions, not tangibly objective realities ; they are nomina (names), not res (things) ; all that exists outside us is a world of individual and specific objects, not of generic and universal things; men exist, and trees, and ani- mals; but man-in-general, or the universal man, does not exist, except in thought; he is a handy mental abstraction, not an external presence or re-ality" (Durant, 1953 Some 2200 years after Aristotle, Gleason (1933) reflected : “It is now generally recognized that a species is an abstract mental concept. . . . To this concept is given a binominal name. . . . The assignment of the individual to a certain concept constitutes identification. . . . The name which ap- pears on a herbarium sheet represents the opinion of some person." Gleason also has pointed out that specific concepts vary through accumulation of study material and through changed mental attitudes and emphasizes that, '' . the probability that a specimen is correctly identified, that is, that it correctiy illustrates a certain specific concept, depends largely on the person making the identification and on the date when it was made." The foregoing comments are intended to serve as an initial warning that the data to follow, though objective in appear- 207 208 Rhodora [Vol. 63 ance, are at least partially quite subjective. Yet, since few would disagree that the boundaries of most present-day- recognized species are “. . . real, objective phenomena" (Stebbins, 1950), the data presented here may be of some interest. In 1952 work was begun which culminated in a survey of the vascular flora of 16 counties of southeastern Iowa (Dav- idson, 1957). The field work done in this connection en- compassed at least parts of four growing seasons during which some 15,000 miles were traveled, innumerable notes taken, and some 4,400 collections, totaling an estimated 9,000 individual specimens, made. In identifying these speci- mens! every effort was made to check as carefully as possible identifications of all specimens collected previously from southeast Iowa and deposited in the Herbarium of the State University of Iowa (1A). Eventually similar consideration was given to ali southeast lowa species included neither in the new collections not at LA but represented in the herbaria of Iowa State University, Parsons College, and Iowa Wesley- an College. A few critical specimens were examined at the Barnes’ Herbarium of the Davenport Public Museum. For the most part, however, the data cited bear upon representa- tives of the flora of southeast Iowa on deposit at IA. This herbarium houses some 125,000 vascular plants. Curated in the past by D. S. Shimek, L. M. Cavanaugh, W. A. Anderson, and currently by R. F. Thorne, the permanent collection at the time of the study was in good condition; nevertheless, special care was taken during the investiga- tion to discard or ignore fragmentary material which per se would be subject to non-reliable identification. Student collections, other than those associated with formal graduate- level research, have not been incorporated into the porma- nent collection and almost without exception specimens and their identifications reflect the field work and opinions of experienced taxonomists. 'Recent regional manuals (Fernald, 1950; Gleason, 1952), monographs and revisions, and other appropriate literature were consulted in identifying materials, in the appli- cation of names, and in the compilation of synonymy. In a few cases deviations from the literature were necessary to express other taxonomic opinions. 1961] Davidson — Specimen Identification 209 Reporting the data presented here is possible due to an earlier belief that differences in opinion regarding the cor- rect identity of herbarium specimens might relate to certain evolutionary considerations. When idealized this reasoning would run something like this : 1. In a given area, species which are clearly defined and not asso- ciated with close relatives nor subject to introgression, etc., are subject to greater unanimity of opinion regarding their identity than are species for which there are close relatives or between which introgression, hybridization, intergradation, etc., occur. 2. "Misidentification"" thus is more closely associated with “critical” taxonomic groups than with “non-critical” groups. 3. Critical groups are more often closely associated with recent and current evolutionary flux than with old evolutionary stability. 4. Misidentification is thus related to evolution and perhaps to evolu- tionary stage. The idealized “logical” conclusion follows that, e. g., if we detected 20% misidentifications within the family Composi- tae amongst 50 genera including 200 species while we found only 5% misidentifications within the family Leguminosae amongst 50 genera and 200 species, we might assume that the Compositae is (at least as represented in a given area) in a stage of greater evolutionary flux than is the Legumino- sae. It seems apparent, however, that such a scheme would be feasible only if numerous qualifications were made and if large floristic segments were studied within the framework of a taxonomy much more refined than that with which we now work. In any event, rather vague thinking along the preceding lines caused accurate records to be kept on each herbarium specimen considered to be misidentified. "These records may be summarized as follows: A total of 1252 species, comprising 507 genera and 124 families, currently are known to occur in southeast Iowa. Of these, 1148 were already represented in the herbaria consulted by the time the senior author's field work was initi- ated. Thus this more recent field work, resulting in the ?"Misidentification" is used for simplicity throughout this paper to mean: ‘‘difference in opinion regarding the identity of. . ." Thus a misidentification was scored each time one of us (R.A.D.) disagreed with the label identification, or the latest annota- tion, of a herbarium specimen. ` 210 Rhodora [Vol. 63 collection of 4435 numbers, increased the number of species comprising the flora of southeast Iowa previously uncollected in the area by 8.3%. One unrecorded species was collected for each 42.3 collection numbers. Said in another way, 2.3% of the new collections represented species previously un- known to the area. Of the 3997 herbarium specimens examined, 184 (or 4.6%) were considered misidentified (this number does not include approximately 35 specimens considered to be puta- tive hybrids). Upon correction these 184 specimens were included in the 507 specimens which represented the 159 species for which misidentified specimens were found. By relating the latter figure to the 1148 species represented in the herbaria we can calculate that 12.1% of all species was falsely represented by at least one herbarium specimen. This figure is of some importance inasmuch as the speed and clarity with which one arrives to a given species concept depends in large part upon the examination of a series of correctly identified specimens. Before re-identification, the 184 misidentified specimens allegedly represented 60 species (synonymy taken into ac- count here as elsewhere) not currently known to be present in southeast Iowa. Thus, for the segment studied, the her- barium was 4.8% richer in species than it should have been. It is suspected that this trend may reflect the taxonomists’ “unconscious” desire to find rarities. On the other hand, the 184 misidentified specimens when re-identified accounted for 21 species which are currently represented by one or more formerly misidentified specimens only. In other words, 21 species now considered validly present in the flora of southeast Iowa were neither represent- ed in the herbarium before the present study nor collected during it. Thus some 1.7% of the total flora was hidden in the herbarium through misidentification. Table 1 lists all families represented by 20 or more her- barium specimens, giving the included number of genera, species, specimens, misidentified specimens, and the percent- age of misidentification. Few, if any, generalizations on the 1961] Davidson — Specimen Identification 211 TABLE 1. DATA ON PLANT FAMILIES REPRESENTED BY TWENTY OR MORE HERBARIUM SPECIMENS Number Number Total "Misiden- Percent Family Genera Species Speci- tified” “Mis- mens identified” Equisetaceae 1 4 24 0 0 Aspidiaceae 8 15 68 0 0 Potamogetonaceae 1 11 31 2 6.4 Gramineae 51 118 441 36 8.1 Cyperaceae 9 72 178 10 5.6 Araceae 3 4 22 0 0 Commelinaceae 2 4 23 0 0 Juncaceae 2 11 23 6 26.0 Liliaceae 10 22 62 0 0 Salicaceae 2 16 64 1 1.5 Betulaceae 4 4 36 1 2.1 Fagaceae 1 9 73 0 0 Polygonaceae 4 27 84 4 4.7 Chenopodiaceae 3 12 24 4 16.6 Amaranthaceae 3 10 36 1 Zul. Caryophyllaceae 8 16 45 0 0 Ranunculaceae 12 28 133 0 0 Papaveraceae 4 7 23 1 4.3 Cruciferae 16 32 62 8 12.9 Saxifragaceae 6 9 29 0 0 Rosaceae 4 41 143 15 10.4 Leguminosae 28 55 215 3 1.3 Rutaceae 2 2 22 0 0 Polygalaceae 1 5 24 0 0 Euphorbiaceae 3 19 o 0 0 Anacardiaceae 1 4 41 0 0 Vitaceae 3 6 25 1 4.0 Malvaceae 6 7 20 0 0 Guttiferae it 9 28 3 10.7 Violaceae 1 16 58 d 29.3 Onagraceae 5 n 39 0 0 Umbelliferae 17 23 49 6 12.2 Primulaceae 3 8 29 1 3.4 Asclepiadaceae 2 12 21 0 0 Convolvulaceae 3 14 36 4 Tot Boraginaceae 7 11 37 0 0 Verbenaceae 2 6 42 0 0 Labiatae 20 40 161 13 8.0 Solanaceae 5 12 40 2 5.0 Scrophulariaceae 19 39 162 0 0 Acanthaceae 2 3 21 0 0 212 Rhodora [Vol. 63 Plantaginaceae 1 7 22 0 0 Rubiaceae 3 8 42 2 4.7 Caprifoliaceae 6 12 40 1 2.5 Campanulaceae 3 7 39 1 2.5 Compositae 52 153 509 32 6.2 cause of misidentification are discernable with any degree of certainty. The quality of misidentifications ranged from genera mistaken for other genera (e. g., Anthemis for Mat- ricaria, Crepis for Pyrrhopappus, Eragrostis for Leptoloma, Pontederia for Heteranthera, Thaspium for Pastinaca, etc.) to "closely related species" being confused (e. g., Mentha spp., Viola spp., Muhlenbergia spp., etc.). The relative tech- nicality of the taxonomy of a given group may or may not be important (at any rate, any attempt to isolate this as a single factor probably would be futile). The Euphorbiaceae, a rather technical group with few critical species (Euphorbia heterophylla L. and E. dentata Michx. being notable excep- tions), was represented by 57 specimens all correctly identi- fied. On the other hand approximately 12% of the 49 specimens representing the Umbelliferae, also technical but with few critical species noted, was misidentified. The Cyperaceae, at once quite technical and apparently including many taxonomic problems, was represented by 178 speci- mens of which only 5.6% was misidentified (the figure quite close to the 4.6% “Grand ‘Misidentification’ Average" for the entire herbarium segment studied). In the Gramineae, similarly technical and also with several critical species com- plexes, misidentifications were found in 8.1% of the 441 representative specimens. The Juncaceae, technical, but with few taxonomic problems encountered, was represented by 23 specimens of which 26% was misidentified. The taxonomy of the Chenopodiaceae and Amaranthaceae might be considered more or less equally technical, but of the two families species of the Amaranthaceae seem generally less well defined; yet of 24 chenopodiaceous specimens 16.6% was misidentified while of 36 amaranthaceous specimens only 2.7% was misidentified. Less technical, but with sever- al outstandingly difficult genera, the Rosaceae bore a misi- dentification figure of 10.4%. The large misidentification 1961] Davidson — Specimen Identification 213 percentage for the Violaceae (29.3€c of 58 specimens) prob- ably reflects the biological complexity (introgression, etc.) of Viola and ccncomitant difficulties in its taxonomic inter- pretation plus intensive recent work on the group. The surprising low misidentification percentages given for the Salicaceae and Fagaceae, both apparently containing biolog- ically complex entities, are probably best accounted for by the fact that many specimens considered uninterpretable were not included in the tally as neither were specimens considered putative hybrids. The number of families (some, e. g., the Scrophulariaceae and Ranunculaceae, represented by relatively large numbers of specimens) for which no misidentifications were detected seems surprisingly large. It should be re-emphasized that the “facts” and “figures” presented here are, actually, only quasi-facts and -figures. Without qualifications they are not strictly appropriate for mathematical manipulations ; with the qualifications that are indicated they seem even less so. These qualifications are of a compounding nature with each seriously affecting the others. In the first place, are the species under consideration actually real with objective, definable limits? Probably most are while some are not — what is the percentage of each in the total flora? How much error is the result of carelessness? For those species that are real, how adequate has been their perception, and how adequately have these perceptions been set to the language of keys, descriptions, ete? How uniformly do various taxonomists interpret this language (which as Gleason, op. cit., has pointed out changes with time) and how much more, or less, accurate are recent identifications than preceding ones? These are just a few of the questions that come to mind. Probably very little of significance can be deduced from this minor side-study. However, an attempt has been made to indicate some of the effects of recent field work on the known flora of a given sector and to quantitate differences of opinion regarding the composition of this flora. — DEPT. OF BOTANY, UNIVERSITY OF WISCONSIN, MADISON,WISCONSIN. 3Specimens were examined and annotated by Dr. Norman H. Russell. 214 Rhodora [Vol. 63 LITERATURE CITED DAVIDSON, R. A. 1957. The flora of southeastern Iowa. Ph.D. Thesis. State University of Iowa Libraries. DuRANT, W. 1953. The story of philosophy. Simon & Schuster. pp. 48-49. FERNALD, M. L. 1950. Gray’s manual of botany. 8th Ed. American Book Co. GLEASON, H. A. 1933. Annotations of herbarium sheets. Rhodora 35: 41-43. 1952. The new Britton and Brown illustrated flora of the northeastern United States and adjacent Canada. 3 Vols. New York Botanical Garden. Lancaster Press. STEBBINS, G. L., JR. 1950. Variation and evolution in plants. Colum- bia Univ. Press. p. 189. A NEW ZEPHYRANTHES FROM SOUTHERN TEXAS FRED B. JONES Zephyranthes refugiensis sp. nov. Bulbus subglobosus 2-2.7 cm. diam.; folia linearia; ad basim 2-3 mm. lat., usque ad 25 cm. long.; pedunculus 15-23 cm. alt.; spatha 2.2-2.8 cm, long. integra, fenestrata aut bifida, dimidio inferiore tubulari; pedicellus 8-16 mm. long.; peri- anthus erectus, tubo 1.5-2.4 em. long., viridi; segmenta perianthi ob- lanceolata ad lanceolata, flava; stylus erectus, antheras attingens; stigma album breviter trilobatum. Bulb subglobose, 2-2.7 cm. wide x 1.7-2.3 em. high, tunics dark brown; neck 4-5 em. long; leaves linear, 2-3 mm. wide at base, to 25 cm. long, grayish green, channelled on upper side, convex on lower side, apex subacute to rounded; peduncle 15-23 em, high, 3-4 mm. wide at base, 2-3 mm, wide at apex, round to slightly flattened, one-flowered; spathe membranous, 2.2-2.8 cm. long, entire, fenestrate or bifid, the lower half tubular, purplish; pedicel 8-16 mm. long; ovary 4-6 mm. long, 3-4 mm. wide; perianth erect, 3.4-4.5 em. long, the limb funnelform; perianth tube 1.5-2.4 cm. long, 2-3 mm. wide at base, 3-4 mm. wide at apex, yel- lowish green; perianth segments oblanceolate to lanceolate, vellow (Wilson, 2-3), greenish at base, often flushed with red on outside; peta- line segments 20-28 mm. long, 7-11 mm. wide; sepaline segments ap- proximately as long but usually 1 mm. wider; filaments inserted at the throat of the perianth tube, suberect, somewhat flattened, light green- ish yellow; sepaline filaments 7-10 mm. long, petaline filaments usually 1 mm. longer; anthers versatile, suberect, affixed much below the middle, 8-10 mm. long at anthesis, the pollen orange-yellow; style erect, greenish below, white in the upper part, reaching apexes of filaments or even of anthers; stigma shortly three-lobed, white; capsule deeply three-lobed; seeds D-shaped, 5-6 mm. long, black. 1961] Jones — A New Zephyranthes 215 FIGURE I. Photomicrograph (X1500) of the somatie chromosome complement of Zephyranthes refugiensis (2n 18) from a colchicine-treated root tip squashed in 2% acetic orcein. Arrows indicate two of the chromosome types readily distinguishing Z. refugiensis from Z. pulchella. (Courtesy of R. O. Flagg, The Blandy Experimental Farm, Boyce, Va.) 216 Rhodora [Vol. 63 TYPE: 14% miles east of Refugio, Refugio County, Texas, Fred B. Jones 4353, Oct. 26, 1960, Welder Wildlife Foundation Herbarium (iso- types to be distributed). The usual habitat of Z. refugiensis is an open swale, either in a brushy pasture or on prairie. The soil preference ap- pears to be a tight sandy loam. Flowering occurs five to ten days after a heavy shower, at which time water to a depth of several inches is likely to be standing over the bulbs. A flush of bloom follows each substantial rain from July to November. Other rain lilies which may be present in the swales and come into flower at the same time as Z. refugi- ensis are Z. pulchella, Cooperia Drummondii, C. Jonesii, C. Traubii and Habranthus texanus. Z. refugiensis is readily distinguished from Z. pulchella, to which it seems to have a close affinity, by the longer per- ianth tube. The lighter yellow perianth and decided frag- rance are also distinctive characteristics. Flagg (Fig. 1) reports that the plant differs cytologically from Z. pulchella. As presently known, the species is limited to northern and eastern Refugio County and a small adjoining area in Goliad County. ! — WELDER WILDLIFE FOUNDATION, SINTON, TEXAS. MERGER OF THE NORTH AMERICAN HOUSTONIA AND OLDENLANDIA UNDER HEDYOTIS* WALTER H. LEWIS Principally on the basis of seed, and to a lesser extent floral and fruit characteristics, the genera Houstonia L. and Oldenlandia L. are distinguished. A comparative study of these features for the North American species and, where possible, for the African Oldenlandia is presented using the following outline. ‘For making available the collections of rain lilies on deposit in their respective herbaria, I am grateful to the following: Dr. B. L. Turner, Herbarium of the Univer- sity of Texas; Dr. F. W. Gould, Tracy Herbarium of A. & M, College of Texas; Dr, Lloyd Shinners, Herbarium of Southern Methodist University. Dr. Hanna Croasdale kindly prepared the Latin diagnosis. "This study was supported by a grant from the National Science Foundation G-9800, 1961] Lewis — Houstonia and Oldenlandia 217 Subg. Houstonia’ — seeds few, large, dorsiventrally flattened, cra- ter-form to concavo-convex, with or occasionally without hilar ridge, en- dosperm corneous; capsules one-fourth to three-fourths inferior; flowers often dimorphic, corolla elongate. Subg. OLDENLANDIA — seeds numerous, minute, angled, rarely sub- globose, endosperm fleshy; capsules wholly inferior; flowers usually monomorphic, corolla short. These criteria are largely from the keys and descriptions of Gray (1860), Wernham (1916), Standley (1918), and Fernald (1950). SEED. A majority of the subg. Houstonia species have averages of 10-15 seeds per capsule. However, Rose (1890) reported ca. 40 seeds for H. brevipes (Rose) W. H. Lewis, Terrell (1959) recorded up to 35 seeds in the H. purpurea (L.) T. & G. complex, and to these are added averages of 30 seeds for the capsules of H. arenaria (Rose) W. H. Lewis and 50-60 for those of H. asperuloides Benth. Most subg. Oldenlandia species average over 50 seeds per capsule, ex- cepting H. microtheca (S. & C.) Steud. and H. watsonii W. H. Lewis from Mexico, with 20 and 30 seeds, respectively. The African Oldenlandia subg. Orophilum Brem. with 9 species and the questionable segregate Lelya Brem. are de- scribed as few-seeded (Bremekamp, 1952). These ex- amples illustrate a continuous and often overlapping range for seed frequency in the two subgenera. The typical seeds of 18 species of subg. Houstonia and subg. Oldenlandia are outlined in figures 1-18. The usual subg. Oldenlandia seed is small, e.g., H. corymbosa (L.) Lam. (fig. 12, 0.24 x 0.4 mm.), but this varies to the larger seeds of H. greenei (Gray) W.H. Lewis (fig. 17, 0.11 x 0.71 mm.) and H. microtheca (fig. 18, 0.65 x 0.90 mm.). These seeds are about the same size or even larger than those of H. (Hous- tonia) asperuloides, H. saxatilis W. H. Lewis, H. peninsul- aris (Brandg.) W. H. Lewis, H. arenaria, H. gracilenta (I. M. Johnst.) W. H. Lewis, H. umbratilis (Robins.) W. H. The North American Houstonia and Oldenlandia are discussed at the rank of sub- genus so that the accepted nomenclature for Hedyotis can be used. Only one of the Afro-Asian species used in evidence here has been transferred to Hedyotis. Because a detailed study of the more numerous Old World species was not attempted, a change in nomenclature at this time is premature and all species and the subg. Orophilum Brem. are retained under Oldenlandia. 218 Rhodora [Vol. 63 Lewis (fig. 6-11), and several additional species not illust- rated. Although the seeds of subg. Houstonia tend to be larger than those of subg. Oldenlandia, a discontinuity in size does not exist. The species of subg. Houstonia are described as having dorsiventrally flattened seeds with distinct craters varying to concavo-convex and with hilar ridges on the ventral face. Near median sections of H. crassifolia Raf. and H. wrightii (Gray) Fosb. seeds are typical representatives (fig. 1-2). Those of subg. Oldenlandia do not have seeds dorsiventrally flattened, craterformed, concavo-convex, nor ventrally ridged, but rather they are angular or more rarely sub- globose (fig. 12-14). These descriptions, however, are not always applicable. The seeds of H. (Houstonia) mullerae Fosb. (fig. 4), H. brevipes (fig. 5), and H. asperuloides (fig. 6) are not strongly dorsiventrally flattened, but are + angular; those of H. arenaria (fig. 9), H. gracilenta (fig. 10), and H. umbratilis (fig. 11) are angular. None are craterform and all lack hilar ridges. An intermediate seed type with only a shallow ventral groove may be noted for H. serpyllacea Schl. (fig. 3). On the other hand, the + dorsi- ventrally flattened seeds of H. (Oldenlandia) callitrichoides (Griseb.) W. H. Lewis (fig. 15) and the concavo-convex seeds of H. microtheca (fig. 18) with hilar ridges hardly match the characteristics of the subg. Oldenlandia. Such variability is not restricted to the American species, for the African O. divaricata Engl., O. benguillensis Hiern, O. luzu- loides Schum., and O. micrantha Chiov., all possess dorsi- ventrally flattened seeds. It is unfortunate that, in the Rubiaceae, there are a num- ber of characters in general use which, for practical taxo- nomic purposes, are difficult to utilize (Verdcourt, 1958). Among these I include the endosperm of small seeds. Most species of subg. Houstonia have corneous endosperms, but among the exceptions are H. arenaria, H. saxatilis, and H. brevipes, whose endosperms are decidedly fleshy. Those of H. (Houstonia) gracilenta and H. (Oldenlandia) callitri- choides are intermediate in texture, while the seeds of H. mi- 1961] Lewis — Houstonia and Oldenlandia 219 G Figs. 1-18. Outlines and near median sections of Hedyotis seeds, Figs. 1-11 subg. Houstonia, Figs. 12-18 subg. Oldenlandia. Fig. 1. H. crassifolia Raf.; fig. 2. H. wrightti (Gray) Fosb.; fig. 3. H. serpyllacea Schl.; fig. 4. H. mullerae Fosb.; fig. 5. H. brevipes (Rose) W. H. Lewis; fig. 6. H. asperuloides Benth.; fig. 7. H. saxatilis W. H. Lewis; fig. 8. H. peninsularis (Brandg.) W. H. Lewis; fig. 9. H. arenaria (Rose) W. H. Lewis; fig. 10. H. gracilenta (I. M. Johnst.) W. H. Lewis; fig. 11. H. umbratilis (Robins.) W. H. Lewis; fig. 12. H. corymbosa (L.) Lam.; fig. 13. H. pringlei (Rob- ins.) W. H. Lewis; fig. 14. H. herbacea L.; fig. 15. H. callitrichoides (Griseb.) W. H. Lewis; fig. 16. H. watsonii W. H. Lewis; fig. 17. H. greenei (Gray) W. H. Lewis; fig. 18. H. microtheca (Schl. & Cham.) Steud. Drawn with the aid of a camera lucida at X24 reduced by one-half in reproduction. 220 Rhodora [Vol. 63 crotheca, H. watsonii, and H. pringlei (Robins.) W. H. Lewis are corneous, contrary to the typification of subg. Oldenlan- dia. This feature is at best a generalization with fleshy, inter- mediate, and corneous endosperms known to each subgenus. FRUIT. The eastern North American species of subg. Houstonia often have capsules one-half or frequently less than one-half inferior. From these first described species, the taxon was sharply separated from its closest relatives and even removed from the Rubiaceae (to the Gentianaceae, DC. Prodr.4: 622, 1830). The study of the south western spe- cies, in particular H. arenaria, H. asperuloides, H. gracilenta, H. mucronata Benth., H. greenmanii Fosb., H. serpyllacea, and H. subviscosa (Wright ex Gray) Shinners, has shown that the subg. Houstonia capsules may be entirely or almost entirely inferior, i.e., not more than one-sixth free. Although the subg. Oldenlandia capsules are rarely free, the American H. microtheca and H. watsonii and the African O. juncoides Schum. and O. staelioides Schum. are not wholly inferior and are indistinguishable from those of the subg. Houstonia species mentioned. FLOWER. At least one American species, H. (Oldenlandia) pringlei, is dimorphic, as are the flowers of O. affinis (R. & S.) DC., O. umbellata L., O. verticillata Bullock ex Brem., O. somala Chiov., and O. eludens Brem. from Afro-Asia. Per- haps a greater per cent of species in the subg. Houstonia possess this characteristic than is known for subg. Olden- landia, but such a tendency has little taxonomic value. Many Mexican and eastern North American species under subg. Houstonia have rather elongate corollas, while the two widely distributed species of subg. Oldenlandia in North America, H. boscii DC. and H. uniflora (L.) Lam., have insignificant, minute corollas. The corollas of H. (Houstonia) australis Lewis & Moore, H. croftiae (Britt. & Rusby) Shin- ners, H. intricata Fosb., H. greenmanii Fosb., and H. sub- viscosa, however, rarely exceed 2 or 3 mm. in length and are about equal to those of the typical subg. Oldenlandia species. In contrast, H. (Oldenlandia) pringlei (5-8 mm.) and H. aestosperma (Robins. & Greenm.) W. H. Lewis (7-9 mm.) 1961] Lewis — Houstonia and Oldenlandia 221 from Mexico and O. cryptocarpa Chiov. (tube, 6 mm.), O. hymenophylla Brem. (tube, 7 mm.), and O. sepaneoides Schum. (tube, 7 mm.) from Africa have long corollas about equal to the longest in subg. Houstonia. No character currently in use separates the Houstonia from the pantropical Oldenlandia, and at best only tenden- cies are expressed. In his revision of the African species, Bremekamp (1952) has chosen, usually with considerable hesitation, to establish genera for those species varying by one or more “key” characters. Although fewer taxa are in- digenous to North America, at least 4 additional genera would be required to accommodate those species morphologi- cally between the (artificial) limits established for typical Houstonia and the North American Oldenlandia. Since the major characteristics of several species exhibit continuous ranges, even this procedure would not place all taxa in de- finite niches, and for this reason alone, such a generic re- alignment would ultimately collapse. On the basis of morphological and the supporting, though incomplete, cytological data (Lewis, 1959), the two genera are merged under Hedyotis following Torrey & Gray (1841), Fosberg (1941, 1954), and Shinners (1949). Most taxa have been transferred to Hedyotis, but a number of new combinations must be made to complete this transfer for the North American species. HEDYOTIS subgenus HOUSTONIA (L.) Gray HEDYOTIS ARENARIA (Rose) W, H. Lewis, comb. nov., based on Hous- tonia arenaria Rose, Contr. U. S. Nat. Herb. 1: 70, 1890. HEDYOTIS ASPERULOIDES Benth., f. BRANDEGEANA (Rose) W. H. Lewis, comb. nov., based on Houstonia brandegeana Rose, Contr. U. S. Nat. Herb. 1: 70, 1890. Distinguished from the typical form by its fine, delicate stems and slightly smaller (1-2 mm.) capsules which are + equal in width and length. HEDYOTIS BREVIPES (Rose) W. H. Lewis, comb. nov., based on Hous- tonia brevipes Rose, Contr. U. S. Nat. Herb. 1: 83, 1890. HEDYOTIS DRYMARIOIDES (Standl.) W. H. Lewis, comb. nov. based on Houstonia drymarioides Standl., Jour. Wash. Acad. Sci. 18: 162, 1928. HEDYOTIS EXIGULA W. H. Lewis, nom. nov., based on Houstonia gra- cilis T. S. Brandg., Zoe 5: 238, 1907. Not Hedyotis gracilis DC., Prodr. 4: 419, 1830. 222 Rhodora [Vol. 63 HEDYOTIS GRACILENTA (I. M. Johnst.) W. H. Lewis, comb. nov., based on Houstonia gracilenta I. M. Johnst., Proc. Calif. Acad. Sci. IV. 12: 1174, 1924. HEDYOTIS LONGIPES (S. Wats.) W. H. Lewis, comb. nov., based on Houstonia longipes S. Wats., Proc. Amer. Acad. 18: 97, 1883. HEDYOTIS NIGRICANS (Lam.) Fosberg, f. SALINA (Heller) W. H. Lewis, comb. nov., based on Houstonia salina Heller, Bot. Expl. S. Tex. (Contrib. Herb. Franklin & Marshall College 1) : 96, 1895. HEDYOTIS PALMERI (Gray) W. H. Lewis, comb. nov., based on Hous- tonia palmeri Gray, Proc. Amer. Acad. 17: 202, 1882. HEDYOTIS PENINSULARIS (T. S. Brandg.) W. H. Lewis, comb. nov., based on Houstonia peninsularis T. S. Brandg., Zoe 5: 160, 1903. HEDYOTIS SAXATILIS W. H. Lewis, nom. nov., based on Houstonia aus- tralis I. M. Johnst., Univ. Calif. Publ. Bot. 7: 446, 1922. Not Hedyotis australis Lewis & Moore, Southwest. Nat.3(1958) : 208, 1959. HEDYOTIS SINALOAE W. H. Lewis, nom. nov., based on Houstonia par- vula T. S. Brandg., Zoe 5: 221, 1905. Not Hedyotis parvula (Gray) Fosberg, Bull. Bishop Mus., Honolulu, no. 174: 54, 1943. HEDYOTIS UMBRATILIS (Robins.) W. H. Lewis, comb. nov., based on Houstonia umbratilis Robins., Proc. Amer, Acad. 45: 401, 1910. HEDYOTIS VEGRANDIS W. H. Lewis, nom. nov., based on Houstonia prostrata T. S. Brandg., Zoe 5: 105, 1901. Not Hedyotis prostrata Korth., Nederl. Kruidk. Arch. II. 2: 160, 1851. HEDYOTIS subgenus OLDENLANDIA (L.) Fosberg HEDYOTIS CALLITRICHOIDES (Griseb.) W. H. Lewis, comb. nov., based on Oldenlandia callitrichoides Griseb., Mem. Amer. Acad. II, 8: 506, 1863. HEDYOTIS CAPILLIPES (Griseb.) W. H. Lewis, comb. nov., based on Oldenlandia capillipes Griseb., Cat. Pl. Cub. 130, 1866. Not Hedyotis capillipes Schl. ex Hook. f., Fl. Brit. Ind. 3: 73, 1880, pro syn. HEDYOTIS GREENEI (Gray) W. H. Lewis, comb. nov., based on Olden- landia greenei Gray, Proc. Amer. Acad. 19: 77, 1883. HEDYOTIS PRINGLEI (Robins.) W. H. Lewis, comb. nov., based on Oldenlandia pringlei Robins., Proc. Amer. Acad, 27: 169, 1892. HEDYOTIS wATSONII W. H. Lewis, nom. nov., based on Oldenlandia ovata S. Wats., Proc. Amer. Acad. 18: 97, 1883. Not Hedyotis ovata Thunb. ex Maxim., Bull. Acad. Petersb. 29: 161, 1883. HEDYOTIS XESTOSPERMA (Robins, & Greenm.) W. H. Lewis, comb. nov. based on Oldenlandia xestosperma Robins. & Greenm., Proc. Amer. Acad. 32: 41, 1896. — DEPT. OF BIOLOGY, STEPHEN F. AUSTIN STATE COLLEGE, NACOGDOCHES, TEXAS. LITERATURE CITED BREMEKAMP, C. E. B. 1952. The African species of Oldenlandia L sensu Hiern et K. Schumann. Verh. Akad. Wet. Amst. 48: 1-297. 1961] Rollins — Draba Aprica 223 FERNALD, M. L. 1950. Gray’s Manual of Botany. 8th ed., 1623 pp. New York. FosBERG, F. R. 1941. Observations on Virginia plants, part I. Va. Jour. Sci. 2: 106-111. . 1954. Notes on plants of the eastern United States. Castanea 19: 25-37. GRAY, A. 1860. Hedyoteae. Proc. Amer. Acad. 4: 312-318. Lewis, W. H. 1959. Chromosomes of east Texas Hedyotis (Rubia- ceae). Southwest. Nat. 3 (1958): 204-207. Rose, J. N. 1890. Houstonia arenaria. Contr. U. S. Nat. Herb. 1: 70. SHINNERS, L. H. 1949. Transfer of Texas species of Houstonia to Hedyotis (Rubiaceae). Field & Lab. 17: 166-169. STANDLEY, P. C. 1918. Oldenlandieae. N. Am. Fl. 32(1) : 17-39. TERRELL, E. E.1959. A revision of the Houstonia purpurea group (Rubiaceae). Rhodora 61: 157-180, 188-207. TORREY, J., AND A. GRAY. 1841. Hedyotideae, Fl. N. Am. 2(1) : 37-43. VERDCOURT, B. 1958. Remarks on the classification of the Rubiaceae. Bull. Jard. Bot. Etat, Brux. 28: 209-281. WERNHAM, H. F. 1916. Tropical American Rubiaceae — VII. Jour. Bot. 54: 322-334. DRABA APRICA IN OKLAHOMA REED C. ROLLINS It is often said among botanists that rare plants receive an unwarranted amount of attention and by and large this appears to be true. Some of the stimulus for this is to be as- sociated with the attraction of novelty for its own sake but I am convinced that a real desire to make rarities more gen- erally known or to clear up doubtful information is often basic to the attention given. A case in point as to the latter has to do with Draba aprica Beadle. This little annual crucifer was first discovered by Thomas Nuttall in Arkansas in 1819 and later published for Nuttall by Torrey and Gray (1838) as Draba brachycarpa Nutt. B fastigiata Nutt. The same plant was found on Kenesaw Mt., near Marietta, Georgia, and described for Beadle by Small (1913) as D. aprica without reference to the earlier varietal name of Nuttall. When Fernald (1934) was working on Draba in the early 1930's, he stimulated L. M. Perry to search for this species on Kenesaw Mt., since it had not been recollected in the in- 224 Rhodora [Vol. 63 tervening years. The results were two collections by Perry and Myers in May, 1934. In his writeup of D. aprica, Fernald (l.c.) raised the question as to whether D. brachy- carpa. var. fastigiata and D. aprica were one and the same thing. The problem at that time also involved the question as to whether the material of Nuttall described as var. fasti- giata really came for Arkansas or was an inadvertent mix- ture, on the same sheet at the New York Botanical Garden, of Georgia material from other collections. Fernald assumed that the specimen at New York is the type (holotype) of D. brachycarpa and that the type (holotype) of var. fastigiata is on the same sheet. Just to put the record straight in one respect, the holotype of D. brachycarpa is at the British Museum. A photograph of it and an isotype are in the Gray Herbarium. The holotype of D. brachycarpa var. fastigiata should also be at the British Museum but I was unable to find it there on one of my visits. Perhaps the New York speci- mens are, in fact, the holotype in the latter case. Although he was wrong about the actual location of the holotype of D. brachycarpa, Fernald did call attention to the need of a search for D. aprica (D. brachycarpa var. fastigiata) in Arkansas to clear up the confusion as to the area of origin of Nuttall’s material. Steyermark (1940) came through with the discovery of Draba aprica in the Ozarks of southeastern Missouri, de- monstrating that the species still occurs in territory at least adjacent to that of Nuttall’s travels. If any doubt remains that the plant might occur where Nuttall went on his famous excursion up the Arkansas River, it should be completely dispelled by a collection made by Kenton Chambers and me in 1957. We had stopped a few miles north of Broken Bow in extreme southeastern Oklahoma to try for cytological material of the then recently described Streptanthus squami- formis Goodman (1956). In the same area with the Strep- tanthus on an open knoll sparsely covered with dwarfed oaks and pines, we found more than a dozen plants of Draba aprica. In the collection (Rollins and Chambers 5762, 3.7 miles north of Broken Bow, McCurtain Co., Oklahoma, 1961] Rollins — Draba Aprica 220 April 13, 1957; GH), we inadvertently included two piants of D. brachycarpa. Our collecting experience with these two species was similar to that of Nuttall, who one hundred and thirty-eight years earlier had mixed the two species in a collection that was primarily D. brachycarpa, with only a few plants of D. aprica included. Steyermark (l. c.) similar- ly found D. brachycarpa and D. aprica growing at the same site. Aside from showing that D. aprica is to be included in the Oklahoma flora, our collection is notable in that it is the only one since Nuttall's day from the area where he original- ly found it. Hitchcock (1941) has suggested that Draba aprica is indeed only a variety of D. brachycarpa, as originally pro- posed by Nuttall and published by Torrey and Gray. How- ever, I disagree with this conclusion. The larger and fewer seeds; the more elongate, densely pubescent instead of gla- brous siliques; the characteristic shortened branches and condensed infructescences; and the dentritic instead of eruciform trichomes of the leaf-surfaces are all distinctive features of D. aprica. A further suggestion that D. aprica might be a polyploid form because of the larger fruits and seeds seems unwarranted in view of the fact that the tric- homes are much smaller than in D. brachycarpa and might with the same justification be interpreted as evidence for the opposite viewpoint. — GRAY HERBARIUM OF HARVARD UNI- VURSITY. LITERATURE CITED FERNALD, M. L. 1934. Draba in Temperate Northeastern America. Rhodora 36: 361-865. Plate 319. GOODMAN, GEORGE J. 1956. A New Species of Streptanthus. Rhodora 58: 354-355. HiTCcHCOCK, C. Leo. 1941. A Revision of the Drabas of Western North America. Univ. Wash. Publ. Biol. 11: 118. SMALL, J. K. 1913. Flora of the Southeastern United States. Ed. 2, Appendix, p. 1336. STEYERMARK, JULIAN A. 1940. Draba aprica in the Ozarks of South- eastern Missouri. Rhodora 42: 32-33. TORREY, JOHN AND ASA GRAY. 1838. A Flora of North America. Vol. 1, pt. 1: 1-184, 226 Rhodora [Vol. 63 GISEKIA PHARNACIOIDES, A NEW WEED R. K. GODFREY! The loose recently tilled sandy soils of a number of orange groves in Orange County, central peninsular Florida, are presently inhabited by large numbers of plants of Gisekia pharnacioides L. (Phytolaccaceae). This is an Old World annual plant much resembling Mollugo verticillata L. in habit and certain superficial characteristics. It has not been previously reported as a naturalized introduction to this continent and I have no information as to when it first ap- peared or by what means it was introduced. I first saw the plant in a single orange grove in the summer of 1958. Later in that year Professor Erdman West, University of Florida, Gainesville, told me that a plant of this species had recently been sent to him for identification from an orange grove in the vicinity of Lake Alfred, Polk County, Florida. Travel- ling through central Florida in the summer of 1961, I saw it in numerous orange groves in the general vicinity of my first collection, between Oakland in Orange County and the Oceola County line, but did not see it elsewhere. Although this weed appears to flourish in cultivated, loose sands much as do Mollugo verticillata, Diodia teres Walt., and other annuals, it would not appear to be any more of a problem agriculturally than are other such ubiquitous weeds. There are several species of Gisekia which occur in differ- ent parts of the Old World. Specimens of our plants were sent to Dr. Carroll E. Wood, Jr., Harvard University, who very kindly compared them with specimens at the Gray Her- barium. He said that ours best matched specimens of Gisekia pharnacioides from India. Citations for my specimens are: Orange County, Florida: locally abundant in an orange grove, 5 miles south of Oak- land, Aug. 8, 1958, Godfrey 57373; abundant, loose sandy soil, orange grove, 3 miles north of jct. Fla. Rt. 530 and 545, south of Oakland, Godfrey and. Reinert 61029. This investigation was supported (in part) by a research grant, RG-6305, to the author from the Division of General Medical Sciences, Public Health Service. Illustra- tions were prepared by Mr. Grady W. Reinert. 1961] Godfrey — Gisekia pharnacioides 22 Gisekia pharnacioides is a somewhat succulent, glabrous annual with several subequal prostrate principal branches spreading radially from the base. The leaves are opposite, the branching is opposite except as one of the two laterals may fail to develop. Leaves short-petioled, narrowly lanceo- late, elliptic-lanceolate, or linear-oblong, entire; midvein n i Nd p 2 e DA NA ESE: x VS g n h c i — Fic. 1. Gisekia pharnacioides. a. Habit, all except one branch removed. b. Node with an axillary dichasium. e. Outline of cross-section of stem. d. Flower. e. Sepal. f. Stamen. g. Ovulary. h. Seed. prominent below, depressed above, lateral veins scarcely evident; upper leaf surface deep green, finely punctate, the lower pale green, somewhat roughly granular-punctate ; lowermost margins of the blades with translucent band-like edges which are decurrent on the petiole thence across the 228 Rhodora [Vol. 63 swollen nodes and finally decurrent along the stem as nar- row pebbly ribs to the next node. The stem body is essential- ly terete aside from the four ribs which are more or less obscured in drying; between the ribs lesser striae are some- times evident. The stems are reddish in color. Inflorescences are axillary dichasia and are produced more or less throughout the branches, the terminus of each major branch ending in a dichasium. In a given axil, the dichasium consists of a solitary short-stalked central flower with two lateral compact, compound, stalked dichasia, the stalks up to 1-2 em. long but frequently very unequal in length. Each of the laterals is subtended by a subulate bract. Individual flowers are small, about 2 mm. long. The calyx is comprised of five boat-like sepals, white-hyaline along their lower margins, and pink-hyaline at their obtuse tips. No corolla is present. Stamens five, alternating with the sepals and about equalling them in length, the filaments translucent, much dilated at the base and abruptly narrowed upward, the anthers white. Carpels five, free, flattened, oblique relative to positions of attachment and of the stigma, but nearly cireular in outline; carpel wall thin, translucent or almost transparent but densely clothed with white, thickish, spicule-like, but soft protuberances, somewhat erose-winged along the flattened margins. Style soft and flat, short, arising obliquely and incurving over the summit of the ovulary, stigmatic at the tip. Ovules one in each carpel. In fruit, the ovulary wall is thin and virtually trans- parent but the dried protuberances of its surface make it appear coarsely granular. The seed is oblique-lenticular, nearly circular in outline, very dark brown or black, its sur- face lustrous and finely punctate, so closely enveloped by the thin ovulary wall that it shows through. Individual plants vary greatly in size as is characteristic of many kinds of annuals. Where they grow very thickly, individuals are small, the branches tending to be ascending. Isolated plants are in general much more vigorous, have branches 6-8 dm. long, perhaps more, and are definitely pro- strate. — DEPARTMENT OF BIOLOGICAL SCIENCES, FLORIDA STATE UNIVERSITY, TALLAHASSEE, 1961] Baker — Heterostyly and Homostyly 229 HETEROSTYLY AND HOMOSTYLY IN LITHOSPERMUM CANESCENS (BORAGINACEAE) H. G. BAKER In the Boraginaceae, heterostyly of the distylous type was first investigated experimentally by Hildebrand (1864, 1865) and Darwin (1877) who both worked with species of Pulmonaria. Its occurrence in the family, however, had been noticed earlier by Torrey (1856), especially in Amsinckia, and Gray (1859) records it for Lithospermum. Much more recently, Johnston (1952), in his taxonomic survey of Litho- spermum, took care to distinguish heterostylous from homo- stylous species and used the distinction in his key for their identification. However, no experimental work has been done on this genus so that our knowledge of the population biology of its species is still at a primitive level. Therefore, it is of interest that two works published by nineteenth century botanists dealing with heterostyly in Lithospermum canescens (Michx.) Lehm., one of the showy North Ameri- can species, appear to have remained unknown to twentieth century writers on the genus even though they contain items of considerable moment. The first of these neglected works is by Erwin Smith (1879). This is of more than merely historical interest be- cause it also records the existence of homostylous plants of L. canescens amongst the plants which he collected in Michi- gan. In Johnston's recent treatment, L. canescens was considered to be exclusively heterostylous and, indeed, no other species of the genus has been recorded as containing indisputably heterostylous and homostylous plants. This need not be surprising for Baker (1960) has given rea- sons for believing that speciation is likely to follow quick- ly after a change in breeding system (and heterostyly promotes allogamy while derived homostyly is generally as- sociated with a greatly increased proportion of autogam- ous seed-production). Translating Smith's remarks on proportions into per- centages, it would seem that roughly 59% of the plants 230 Rhodora [Vol. 63 which he saw were short-styled (with stamens inserted at the mouth of the corolla-tube) while only 39% were long- styled. In the long-styled plants the length of the style varied from being included within the corolla-tube to be- ing conspicuously exserted, an appearance which could re- sult from the sampling of flowers of different ages should there be a differential elongation of corolla-tube and style during maturation of the flower. The remaining 2% of plants would be classified nowadays as “short-homostyles”’. Even these showed some variability, the anthers standing either at the level of the stigmas or just above them. Par- ticularly striking, however, was a tiny proportion of homo- stylous plants in which the stamens are inserted at two levels, two of them above the stigma and three on a level with it. The accuracy of such an observation is testified to by the subsequent description of an unequal (or zonal) inser- tion of stamens by Johnston (1952) in Lithospermum tournefortii Johnston (heterostylous) and L. decumbens Vent. (probably homostylous), as well as in several other genera of the Boraginaceae. However, it is in the pub- lished accounts of the floral structure of Amsinckia lun- aris Macbr. (Macbride, 1917; Suksdorf, 1931; Ray and Chisaki, 1957) that the closest parallel may be seen. Here, probably as in Lithospermum canescens, a heterostylous species has produced a homostylous variant in which the stamens are inserted at two levels, one of them correspond- ing with the height of the stigma and almost certainly leading to a high proportion of self-pollination. In Am- sinckia lunaris, according to Ray and Chisaki, the pollen grains from the two lower anthers are smaller than those from the upper trio (which may function in cross-pollin- ation). It would be most interesting to know if a similar situation obtains in the unusual plants of Lithospermum canescens. In view of this report of homostylous plants of L. cane- scens from Michigan, a survey was made of the floral structure of specimens from a variety of States in the her- baria at the University of California, Berkeley, and at 1961] Baker — Heterostyly and Homostyly 231 Stanford University. Also Dr. Paul Grun and Dr. R. B. Channell kindly made observations on material at Pennsyl- vania State and Vanderbilt Universities, respectively. Un- fortunately, no homostylous specimens were found. Overall, however, 39 short-styled plants were observed along with 24 long-styled. This preponderance of short-styled plants talles with Smith's (1879) observation from Michigan. If field-studies should show this to be a regular feature of naturally occurring material, it might be interpreted most reasonably as indicating that some self-pollination occurs (otherwise the two forms might be expected to occur in roughly equal numbers). A preponderance of short-styled plants could be produced if this type is self-pollinated con- siderably more frequently than is the case with the long- styled form, through pollen falling down the corolla-tube. Another possible cause of such a disparity in numbers, al- though of lesser likelihood, would be that both forms are selfed with roughly equal frequency but that the allele pro- ducing the long-styled condition is the genetical dominant. On selfing the heterozygous form produces both kinds in its progeny while the homozygous recessive form produces all recessives, creating an excess of this form. However, in almost all cases where the genetics of a distylous system is known it is the short-styled form which is genetically dominant (cf. Lewis, 1954). Only in the Plumbaginaceae is there good reason to believe that the opposite is true (Ba- ker, 1954 and unpub.), and preliminary results for the Boraginaceae from the genus Amsinckia (Ray and Chisa- ki, 1957), as well as a modern interpretation of Darwin's (1877) results with Pulmonaria, both suggest a contrary situation; that the short-styled condition is dominant. Another, apparently unique, feature of heterostyly in Lithospermum in a difference in shape between the pollen of long-styled and short styled plants. To Johnston (1952), his discovery of this shape-difference correlated with heter- ostyly demonstrated “a type of pollen dimorphism previously unreported.” However, in 1880, in the second of these ne- glected papers, C. E. Bessey described heterostyly in materi- al of L. canescens from central Iowa and included not only 232 Rhodora [Vol. 63 statistics on the sizes of the pollen grains of the two kinds of plants but also these comments on their shapes, “pollen grains ovoidal, slightly constricted in the middle" (on short- styled plants) and “pollen grains oblong, much constricted in the middle" (on the long-styled plants). The descriptions accord quite well with Johnston’s more detailed observations made three-quarters of a century later. This overlooked, earlier description of pollen shape dif- ferences associated with heterostyly recalls that dimor- phism in sculpturing of the grains was described as a part of flower dimorphism in the Plumbaginaceae by Macleod in 1887, but that this lay unused until Kulezynski (1932) ap- plied it to the identification of fossil pollen. Actually, neither Bessey's nor Macleod’s description of pollen di- morphism was the first report of such a phenomenon, for it had been seen in the Rubiaceae as early as 1868 by F. Miiller (1869), the significance of this demonstration re- maining unnoticed until recently (Baker, 1956). In fact, Bessey’s short paper was more concerned with a description of the variable relationship between stamen and style lengths in another species, Lithospermum longi- florum Pursh (which is now considered synonymous with L. incisum Lehm.). In this species, Bessey considered that heterostyly might be in process of development. However, in view of its production of cleistogamic flowers on a regu- lar seasonal basis, it seems more likely that his alternative suggestion, that this is a species which has moved toward inbreeding rather than away from it, is the correct one. However, had Bessey made his observations just a few years sooner, it is unlikely that they would have been rel- egated to the obscurity which has been their fate. On May 22, 1877, Asa Gray wrote to Charles Darwin saying, “I asked my good correspondent Prof. Bessey to see if Litho- spermum longiflorum (= angustifolium) being cleistoga- mous later, is, like its relatives, also dimorphous. Here is his first reply just in season to send you by this post. I for- got to ask him to examine pollen. I will do so" (unpub. let- 1961] Baker — Heterostyly and Homostyly 233 ter in Gray Herbarium, quoted from microfilm copies kind- ly lent by Dr. Hunter Dupree). Darwin replied on June 4, 1877, in a letter which, be- cause it does not mention the genus involved, has always been a mystery (cf. transcript in Holbrook, 1939). Now, we can see that it was to Lithospermum incisum that Dar- win was referring when he wrote, “Prof. Bessey’s case has come too late, as the sheets on this subject are printed ee The pollen grains and stigmas ought to be com- pared. The case seems to be well worth careful investiga- tion and I would have given my eyes for seeds formerly; but now I have done with the subject." He went on to give instructions for Bessey on methods of observation and ex- perimentation and point out that, "the case may be one merely of great variability or it may be one of incipient heterostylism." Bessey appears to have given some heed to the instructions because his paper did not appear till three years later. Meanwhile, Darwin (1877) had published the famous book to which he was making reference in his let- ter — “The Different Forms of Flowers on Plants of the Same Species." Surprisingly, Asa Gray, himself, never made a clear statement on dimorphism in L. canescens. In the “Synop- tieal Flora of North America" (Gray, 1878, and subsequent editions) his sole comment reads, *. . . in one form style about the length of the tube and stamens, inserted below its middle." Even with the deletion of the comma, one is merely left to infer the existence and appearance of a sec- ond form. Knuth (1899, page 120; in translation 1909, page 137) added to the camouflage of Bessey's discovery by giving an erroneous reference in the statement, “Darwin des- cribes the flowers of this species [L. canescens] as either heterostylous or very variable in regard to the length of the style." Actually, Darwin never described the heterosty- lism of any species of Lithospermum and, in all probabil- ity, the statement should have referred to Bessey and to L. incisum! Similarly, Knuth credits Darwin with a de- 234 Rhodora [Vol. 63 scription of cleistogamous flowers in L. incisum (Knuth, 1899, page 120, as L. longiflorum Pursh; in translation 1909, page 137, as L. angustifolium Michx.). Once again the credit should have been Bessey's. On another page, however, although rather inconspicuously, the same author does give Bessey credit for having seen heterostyly in L. canescens (Knuth, 1898, page 62; in translation 1906, page 50). Thus, despite a history of confusion and neglect lasting for the larger part of a century, the reproductive biology of Lithospermum canescens is commended to the attention of field-botanists and experimentalists living within its geo- graphical range in the eastern half of North America. This species promises to reward the student of natural popula- tions through variations in the proportions of the various flower-forms and the genetical studies which these may make feasible. — BOTANY DEPARTMENT, UNIVERSITY OF CALI- FORNIA, BERKELEY, CALIFORNIA. LITERATURE CITED BAKER, H. G. 1954. Dimorphism and incompatibility in the Plumbagi- naceae. Rapp. et Comm. 8ème Congr. Int. de Botanique, Paris, 1954, sect 10:133-134. ———————————.1956. Pollen dimorphism in the Rubiaceae. Evolution, 10: 23-31. ————————-.1960. Reproductive methods as factors in speciation in flowering plants. Cold Spring Harbor Symp. Quant. Biol., 24:177- 191. BEssEY, C. E. 1880. The supposed dimorphism of Lithospermum long- iflorum (L. angustifolium Michx. of Gray's Synoptical Flora). Amer, Nat. 14:417-421. DARWIN, C. R. 1877. The Different Forms of Flowers on Plants of the Same Species. London, England. GRAY, A. 1859. Manual of the Botany of the Northern United States. Revised edition. New York. GRAY, A. 1878. Synoptical Flora of North America. Vol. 2, Part 1. New York. HILDERBRAND, F. 1864. Uber den Dimorphismus von Pulmonaria offici- nalis. Verh. Nathist. Ver., Bonn, 21:56. . 1865. Experimente zur Dichogamie und zum Dimor- phismus. Bot. Zeit., 23: 13-15. 1961] Swallen — New Species of Panicum 235 HOLBROOK, C. F. 1939. Calendar of the Letters of Charles Robert Darwin to Asa Gray. The Historical Records Survey, Boston, Massachusetts. JOHNSTON, I. M. 1952. Studies in the Boraginaceae, XXIII, A survey of the genus Lithospermum. Journ. Arnold Arb. 33:299-363. KNUTH, P. 1898. Handbuch der Blütenbiologie. vol. 1. Leipzig. .1899. Ibid. vol. 2 (2). .1906. Handbook of Flower Pollination (trans. by J. R. A. Davis), vol. 1. Oxford. . 1909. Ibid., vol. 3. KULCZYNSKI, S. 1932. Die altidiluvialen Dryasfloren der Gegend von Przemysl. Acta Soc. Bot. Polon. 9:236-299. Lewis, D. 1954. Comparative incompatibility in angiosperms and fungi. Adv. in Genetics, 6:235-285. MACBRIDE, J. F. 1917. A revision of the North American species of Amsinckia. Contr. Gray Herb. N.S. 49:1-16. MACLEOD, J. 1887. Untersuchungen über die Befruchtung der Blumen. Bot. Centralbl. 29:150-154. MÜLLER, F. 1869. Ueber eine dimorphe Faramea. Bot. Zeit., 27:606- 611. Rav, P. M. AND H. F. CHISAKI. 1957. Studies on Amsinckia. IL A synopsis of the genus, with a study of heterostyly in it. Amer. Jour. Bot. 44: 529-536. SMITH, E. F. 1879. Trimorphism in Lithospermum canescens, Lehm. Bot. Gaz. 4:168-169. SUKSDORF, W. N. 1931. Untersuchungen in der Gattung Amsinckia. Werdenda 1:47-113. TORREY, J. 1856. Report on the botany of the (Whipple) expedition. tept. Route near 35th Parallel 4:124. A NEW SPECIES OF PANICUM FROM NEW JERSEY JASON R. SWALLEN A specimen of Panicum was received recently for ident- ification from Bayard Long of the Academy of Natural Sciences of Philadelphia, which has proven to be new. It was collected by Frank Hirst, an active amateur botanist in southern New Jersey, who recognized it as an unusual plant, and in whose honor the species is named. “It occurs dominantly in a small woodland pond in the pine barrens — growing in the water, much as Panicum spretum often does. This is a most interesting pond, the Panicum being 236 Rhodora [Vol. 63 associated with Lobelia boykinii, Paspalum dissectum, Core- opsis rosea, etc.” Panicum hirstii Swallen, sp. nov. Culmi 55-80 cm. alti, ramosi, glabri; ligula brevis, ciliata; laminae acuminatae, adscendentes, usque ad 11 cm. longae, 3-5.5 mm. latae, glabrae; panicula angustissima, usque ad 9 cm. longa; spieulae 1.8-2.1 mm. longae, glabrae; gluma prima obtusa enervis. Culms 55-80 cm. tall, erect or ascending, glabrous; sheaths mostly shorter than the internodes, glabrous or sparsely pilose on the margins near the summit, the uppermost usually elongate; ligule a dense line of hairs, 0.5 mm. long; blades stiffly erect or narrowly ascending, 4.5-11 cm. long, 3-5.5 mm. wide, flat, acuminate, glabrous, often tinged with purple; panicles 4.5-9 cm. long, about 5 mm. wide, the branches as much as 2.5 em. long, appressed, smooth or scaberulous; spikelets 1.8-2.1 mm. long, glabrous, the pedicels appressed, usually a little shorter than the spikelets; first glume broadly obtuse, nerveless, 0.5-0.8 mm. long; sec- ond glume and sterile lemma with hyaline margins, the lemma cuccu- late, subacute, as long as the fruit, the glume broadly obtuse, a little shorter than the fruit; fruit subacute, obscurely roughened, pale or yellowish. Autumnal phase sparingly branching from most of the nodes, the leaves and panicles not much reduced. Type in the U. S. National Herbarium No. 2,306,491, collected in Leipzig Ave. Pond, Germania, Atlantic Co., New Jersey, June 7, 1959, by Frank Hirst (No. 47). Additional material examined: NEW JERSEY: In water of small wood- land pond, pine barrens, in Galloway Township, southeast of Egg Har- bor City (this is the same locality as the one where the type was collected, fide Bayard Long), August 11, 1959, Bayard Long 85320. GEORGIA: Margin of pine-barren pond, Sumter County, August 23, 1900, Roland M. Harper 458; Cypress swamp, 1 mile north of Leary, Calhoun County, May 31, 1947, Robert H. Thorne 4313. Harper 458 was doubtfully referred to Panicum roanok- ense Ashe by Hitchcock and Chase in their revision of Pan- icum. (Contr. U. S. Nat. Herb.15: 197. 1910). P. hirstii is readily distinguished from P. roanokense, however, by the very narrow panicles with relatively — short-pedicelled spikelets, the shorter second glume, and the sparingly branching culms, the autumnal blades and panicles scarce- ly reduced. — DEPARTMENT OF BOTANY, U. S. NATIONAL MU- SEUM, SMITHSONIAN INSTITUTION, WASHINGTON, D. C. Volume 63, No. 751, including pages 181-206, was issued July 27, 1961. JW REFERENCE LIBRARY Noy € (Sci. Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Associate Editors Vol. 63 September, 1961 No. 753 CONTENTS: Evidence for the Hybrid Origin of Scirpus peckii. A fred ke Solvent MI N ep 237 Environmental Variation in Heterotheca subaxillaris. CORN BUT eee t oat xq cu E 243 Observations on the Sagittaria subulata Complex. I nestorme dams ONG R. He, GOdLToM eR E e EDUC 247 Che Nem England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume. Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able, All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. ) Second Class Postage Paid at Boston, Mass. Printed by THE LEXINGTON PRESS, INC. Lexington, Mass. TRbooora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 September, 1961 No. 753 EVIDENCE FOR THE HYBRID ORIGIN OF SCIRPUS PECKII ALFRED E. SCHUYLER On the basis of current morphological, cytological, and ecological studies by the author, the sedge Scirpus peckii Britt. appears very likely to be a hybrid between Scirpus atrovirens Willd. of section Taphrogeton, and either Scir- pus atrocinctus Fern. or Scirpus pedicellatus Fern. of sec- tion Trichophorum. Although S. peckii has had a rather complicated history according to the taxonomic literature (Brainerd, 1901), there has been little question in the past about regarding it as a distinctive species. Fernald stated (1900) that, “It is a very handsome and unique plant, not closely related to any described species." It is paradoxical that he also described one of the probable parents of S. peckii in the same paper. Fernald (1950) considered S. peckii to belong to still another section than those already mentioned, section Androcoma, while admitting (1900) that it had some characteristics of section Trichophorum. MORPHOLOGICAL EVIDENCE In order to discuss these plants in detail, I prefer to use comparative descriptive terms rather than quantitative terms. It should be borne in mind that these comparative terms only apply to S. atrovirens, S. atrocinctus, S. pedicella- tus, and S. peckii. Scirpus atrovirens, one of the putative parents, produces culms from an underground rhizome with relatively long internodes (Plowman, 1906). The leaves, involucral bracts, and scales have a comparatively low length/width ratio. The values for the bracts are expressed as quotients 237 238 Rhodora [Vol. 63 in table I. The scales have a conspicuous mucronate tip and are arranged in a steep spiral on the spikelet axis. The lateral rays of the inflorescence branch at relatively di- vergent, sometimes obtuse angles from the main rays. Bulb- lets often occur where the rays branch. The spikelets are arranged in compact glomerules, and all the flowers of the spikelet mature at about the same time. After the flowers mature, the spikelet ceases to grow. Hence, flowering and spikelet growth occur during a relatively short span of time. The flowers in the lower part of the spikelet gener- ally have two stamens, although sometimes there are three. Toward the top of the spikelet, the flowers lack stamens. The perianth bristles are short and inconspicuous, and have downward protrusions called barbs. Scirpus atrocinctus and S. pedicellatus, the other prob- able parents of S. peckii, are members of the Scirpus cy- perinus (L.) Kunth complex. The plants of this complex, along with Scirpus longii Fern., are commonly called wool- grasses. Except for S. longii, I think it is better to regard these woolgrass types as conspecific and only varietally distinct. However I will treat them in the traditional man- ner for convenience in this paper. In New York and Ver- mont, S. atrocinctus usually occurs in the uplands, while S. pedicellatus usually occurs in stream and river valleys. Morphologically the two types are very similar; they dif- fer mainly in color and size, S. atrocinctus is a slender plant with black scales, and S. pedicellatus is a robust plant with brown scales. Both types flower approximately at the same time as S. atrovirens, while S. cyperinus (sensu stricto) flowers later in the summer. These two woolgrass types have underground rhizomes with short internodes (Plowman, 1906) and form large tussocks. The leaves, involucral bracts, and scales have a comparatively high length/width ratio. Unlike S. atrovi- rens, the scales do not have a conspicuous mucronate tip and they are arranged in a shallow spiral on the spikelet axis. The lateral rays branch at acute angles from the main rays and only rarely do bulblets occur where the rays branch. The spikelets are solitary and the flowers mature 1961] Schuyler — Evidence for Hybrid Scirpus 239 gradually from the base to the tip of the spikelet. The tip continues to grow over a relatively long span of time. The flowers in the lower part of the spikelet generally have 1 or 2 stamens, but higher in the spikelet, most flowers lack stamens. However, in contrast to S. atrovirens, there are a few bisexual flowers scattered throughout the up- per part of the spikelet. The perianth bristles are long, con- spicuous, and without barbs. The suggested hybrid, S. peckii, produces culms in tus- socks similar to S. atrocinctus and S. pedicellatus. The leaves, involucral bracts, and scales have a length/width ra- tio which is intermediate between S. atrovirens and the wool- grasses. The scales do not have a conspicuous mucronate tip, and are arranged in a spiral which is intermediate between TA CN EN S. ATROVIRENS S. PECKII S. ATROCINCTUS Fic, I. 1. Perianth bristles, x 47 2. Seale tips, x 12 3. Meiotic chromosomes, polar views, metaphase I, x 455 4. Fruits and inclosed seeds, x 34.5. the steep spiral of S. atrovirens and the shallow spiral of the woolgrasses. The lateral rays branch at angles which are intermediate between those of the probable parents. As in S. atrovirens, bulblets often occur where the rays branch. The spikelets may be solitary or may occur in small glomer- ules. At first, the spikelet appears to mature in the same manner as S. atrovirens, when about 25 flowers mature at about the same time. However, in contrast to S. atrovirens, the tip of the spikelet does not cease to grow, but continues 240 Rhodora [ Vol. 63 on to form another tier of flowers. The flowers in the lower part of each tier have two, or occasionally three stamens, and toward the top of each tier, the flowers lack stamens. In this respect, each tier shows a close resemblance to the entire spikelet of S. atrovirens. The perianth bristles are about twice as long as those of S. atrovirens. Like S. atro- virens, the bristles bear barbs, but they are not as long nor are they always protruding downward. Scirpus peckii also appears to be intermediate between S5. atrovirens and the woolgrasses on the basis of anther length, style length, and fruit length. However, the differences are minute, and my initial measurements are not complete enough to demonstrate this convincingly. ECOLOGICAL EVIDENCE From my observations of S. peckii in eleven localities in TABLE I. A comparison of some characteristics of Scirpus atrovirens, S. peckii, S. atrocinctus, and S. pedicellatus.! Character S. atrovirens S. peckii S. atrocinctus S. pedicellatus Growth form Cespitose, 'Tussocks 'Tussocks but not tussock- forming Color at base of Green Black or Black or lowest involueral brown brown bract Length divided by 16-30 32-53 43-87 the width of the lowest involucral bract Seale length Seale tip Spikelet arrangement Flower maturation throughout the spikelet Bristle length Bristle margin Stamen number/ flower* *arranged in order of their frequency of occurrence; parentheses 1.3-1.7 mm. Mucronate Glomerate Almost simul- taneously 0.7-1.1 mm. Strongly barbed 0, 2, (1), (3) number is rarely observed. YDhe measurements included in this table came from 10 plants of 1.3-1.7 mm. Mucronulate Solitary to sub-glomerate Gradually in tiers 1.6-2.0 mm. Moderately barbed 2, 0, 1, (3) plants of S. peckii, and 15 woolgrass plants. 1.5-2.0 mm. Mucronulate Solitary Gradually 4.5-6.0 mm. Not barbed 0, 1, 2 indicate that the S. atrovirens, 15 1961] Schuyler — Evidence for Hybrid Scirpus 241 Vermont and New York, the following pattern appeared: S. peckii is widely distributed in these states, but rarely forms large populations; often only 1 to 5 tussocks are found in a partieular locality. Further evidence pertaining to this dis- tribution pattern has been provided by Ezra Brainerd (1901). He stated in regard to S. peckii, “In a drive of over seventy miles I found single plants of two or three culms each in six stations miles apart." Such an occurrence is unique among the leafy species of Scirpus which usually form sizeable populations. In the localities where I observed S. peckii, it was always associated with S. atrocinctus or S. pedicellatus, and, in all but two localities, S. atrovirens was also present. Furthermore, there was a morphological cor- relation between S. peckii and the woolgrass type in each locality. In localities where S. atrocinctus was the only wool- grass type, plants of S. peckii were slender, while in locali- ties where S. pedicellatus was the woolgrass type, plants of S. peckii were more robust. In one locality, where both wool- grass types were present, both forms of S. peckii, the slender and the robust, were present. I think that the unique distri- bution pattern, and the morphological correlation of S. peckii with the associated members of the woolgrass group, give strong additional support for the hybrid origin of this species. INDICATORS OF HYBRID STERILITY Further evidence of the hybrid nature of S. peckii is pro- vided by its pollen grains, chromosomes, and fruits. Almost all the pollen grains of S. peckii which I have observed, have highly distorted walls and their contents do not stain with aceto-carmine. They are in striking contrast to those ob- served from woolgrasses, which usually have undistorted walls and their contents readily stain with aceto-carmine. I have only been able to make an approximate determination of the chromosome number of S. peckii. There appear to be about 60 chromosomes in pollen mother cells undergoing meiosis. This number is approximately the sum of the hy- pothetical parents. S. atrocinctus has a haploid number of 34 and S. atrovirens has a haploid number of 25-30 (Hicks, 1928). The chromosome number of S. pedicellatus has not yet been determined. Also the units at meiotic metaphase in 242 Rhodora [Vol. 63 S. peckii are smaller in size than those of S. atrocinctus. Be- cause of their higher number and smaller size, I am inclined to regard the chromosomes of S. peckii as univalents. This condition would not be unusual in a hybrid between two fairly unrelated parents. All the fruits of S. peckii which I have observed, have lacked mature seeds. These abortive seeds lack mature embryos and an endosperm. Taxonomists have indirectly made use of these seedless fruits as a taxo- nomic character. Fernald (1950) mentions that the achenes are soft and whitish, and Gleason (1952) mentions that they are very pale. These characteristics are due to the lack of the yellow-brown seed which colors and fills the fruits in other leafy species of Scirpus. CONCLUSIONS In view of the evidence presented here, I am certain that this widespread northeastern species is really not so “unique” as it first appears. On the contrary, it appears to have achieved its unusual morphological characteristies by combining the characteristics of two well-known and dis- tinctive species. It is very likely a sterile hybrid which arose, and probably is still arising, over the wide geographical area in which the common parents co-exist. I am somewhat surprised that it has not previously been suggested that S. peckii is a hybrid. I think that the reason for this is due to the technical difficulties that we encounter while identifying many species-groups in the Cyperaceae. The search for differentiating characters has to be a critical one in order to make proper species identifications. How- ever, a critical search for similarities has little value for identification purposes. Hence, we naturally tend to over- emphasize differences and tend to ignore similarities which may have taxonomic significance. — UNIVERSITY OF MICHI- GAN, ANN ARBOR. LITERATURE CITED BRAINERD, E. 1901. Scirpus atratus a synonym of Scirpus Peckii. Rhodora 3:31-33. FERNALD, M. L. 1900. Some northeastern species of Scirpus. Rhodora 2:15-21. FERNALD, M. L. 1950. Gray's Manual of Botany. 8th ed. American Book Co., New York. 1961] Burk — Variation in Heterotheca 243 GLEASON, H. A. 1952. The New Britton and Brown Illustrated Flora of the Northeastern United States and Adjacent Canada. The New York Botanical Garden, New York. Hicks, G. C. 1928. Chromosome studies in the Cyperaceae with special reference to Scirpus. Bot. Gaz. 86:295-317. PLOWMAN, A. B. 1906. The comparative anatomy and phylogeny of the Cyperaceae. Ann. Bot. 20:1-33. ENVIRONMENTAL VARIATION IN HETEROTHECA SUBAXILLARIS C. JouN BURK A weedy composite, Heterotheca subaxillaris, is a promi- nent member of the dune-grass community on Bogue Bar- rier, the most southern island of the North Carolina Outer Banks. The plant occurs not only on the open dunes, but also along roadsides and pathways through the shrub zone and into the pine-oak-hickory forest which covers those portions of the island which are most protected from salt spray. The plants on the dunes are low-growing, nearly prostrate forms. The plants in the pine-oak-hickory forest grow erect and to a height of a meter or slightly more. The plants occurring in the shrub zone avc intermediate in size between these two extremes. While most of these plants flower in late summer and early fall, one population (which occurred on both sides of the road adjacent to the bridge which connects the north end of the island with the mainland) was observed in flower all the year-round, some cighty plants being in full bloom on Janu- ary 19, 1960. Most of the other Heterotheca plants on the island were in winter-rosette form at that time. The winter- flowering specimens resembled a plant described by Benke (1928) as H. subaxillaris var. petiolaris. They were shorter than typical H. subaxillaris and more densely pubescent. While a few of the flowering heads were aborted and de- formed, most of the capitula were larger (more than a centi- meter in diameter) than those of typical H. subawillaris. They flowered on short branches sent up from a persistent rosette, and, as Benke had noted, were very similar in ap- pearance to plants of the closely allied genus Chrysopsis. 244 Rhodora [Vol. 63 Stem cuttings were made from portions of the basal ros- ettes of a number of these plants, and clonal pairs of cut- tings were placed, one on short day (9 AM to 5 PM) and the other on long day (dawn to 11 PM) in the greenhouse at the University of North Carolina. Winter rosettes of typical H. subaxillaris had been brought into the greenhouse and placed on long and short days on December 9, 1959. The winter rosette specimens flowered within two months on long day (2 plants) ; they did not flower at all on short day but remained in vegetative condition until early May of 1960 when the experiment was concluded. The clone-cut- tings of the winter-flowering plants which had been placed on long day rooted quickly and began to flower within a month. The ones on short day rooted rather tardily. Many of the latter plants died. One of them flowered on April 26, 1959, and on May 2, two more plants were in bud out of a total of eight surviving cuttings. Of the long-day members, eight had flowered out of fifteen which had survived. It would appear from these preliminary experiments that typical H. subaxillaris requires long days to flower whereas Benke's variety petiolaris will flower on long or short day- lengths. It might be noted that the portion of the Outer Banks from Bogue Barrier north past Cape Hatteras is an “island” of localized mild climate along a coastline which, northward and southward, is more rigorous. The ability to flower all the year-round might well be a selective advantage under these conditions. In March and April of 1960, while these experiments were in progress, a revision of Heterotheca, Section Heterotheca, was published (Wagenknecht, 1960). In this work H. sub- axillaris var. procumbens was described. The description of this variety coincided exactly with that of the plants in the dune-grass community on Bogue Barrier, and it was sus- pected that H. subaxillaris var. procumbens might be an ecotypie derivative of typical H. subaxillaris. Thirty established plants of the beach variety and 30 es- tablished plants from the populations of tall-growing H. sub- axillaris from the pine-oak-hickory forest were transplanted from Bogue Barrier to alternating rows in the Botanical Gardens of the University of North Carolina in early June 1961] Burk — Variation in Heterotheca 245 of 1960. At the time of transplanting, the plants from the pine-oak-hickory forest had begun to bolt whereas those from the dunes had not. The transplants were watered only at the time of transplanting. After that, they were exposed to normal meteorological conditions. They were grown on a poor clay soil to which no fertilizers had been added. Never- theless, both sets of plants grew very well in Chapel Hill and, under the virtually uniform conditions provided in the Bo- tanical Garden plots, developed into virtually identical sets of plants. The heights of the transplants and the length of the longest stem of each are given in Table 1. The ratio of the height to the length may be considered a general measure TABLE l. Height, length, and height/length ratio of H. subaxillaris from dune and pine-oak-hickory populations contrasted with height, length, and height/length ratio of plants from the same populations grown in a uniform environment. 20 plants from each population were measured in centimeters and the longest stem of each plant was used for length measurement. Dune-grass Forest Dune-grass Forest ecads tvpes transplants transplants height range som 55-100 104-148 94-157 average height 27 81 124 124 length range 21-48 68-103 106-154 94-160 average length 33 86 129 124 height/length .82 .94 .96 OF of the degree of procumbence. Included in Table 1 are the heights and lengths of 20 specimens taken from each of the two populations on the island from which the transplants had been removed in June. These two latter sets of plants were collected on October 13, 1960; the transplants were up- rooted and measured on October 18. It would appear from the data that the Bogue Barrier plants referable to H. subaxillaris var. procumbens reflect the response of a common genotype to more rigorous envi- ronmental conditions. The other characters separating the variety from the species (degree of serration of the leaves, prominence of lateral veins) are subjective at best and were quite useless in separating the two sets of plants grown in the uniform environment from one another. The recent revision placed H. subaxillaris var. petiolaris in synonymy with H. latifolia Buckley. While the probable 246 Rhodora [ Vol. 63 photoperiodic nature of the variations in Benke's plant was recognized (it had been collected in March), it was con- sidered to be a deviant of the western species H. latifolia on the basis of its heavy pubescence. Examination of the sites upon which the photoperiodic variant occurs revealed no specimens referable to H. latifolia, as described in the re- vision, although numerous plants of typical H. subaxillaris were encountered. It would seem more likely that the day- neutral plants on Bogue Barrier were variants of the species already abundant on the island than that they were plants of a separate introduced species. One of the principal characters used to distinguish H. lati- folia from H. subaxillaris is the tendency of H. latifolia to grow as much as two meters in height while H. subaxillaris does not exceed one meter. An examination of Table 1 shows that this criterion is not valid. It should be mentioned that a specimen of H. subaxillaris from Bogue Barrier was cited in the revision as a representative specimen. The heavier pubescence attributed to H. latifolia and encountered in the winter-flowering plants on Bogue Barrier likewise does not appear to be a valid criterion for distinguishing two taxa since clonal plants of typical H. subaxillaris also developed the heavy pubescence when grown in the greenhouse over winter, regardless of the daylength to which they were ex- posed, "losing" the heavy pubescence when grown in the Botanical Gardens over the summer. Thus it would seem preferable, for the time being, to re- tain the former concept of H. subaxillaris as a single, highly variable species until a more extensive study of the effects of different environments upon its forms is made. Specimens representing all environmental variants cited, both naturally occurring and experimentally induced, have been deposited in the herbarium of the University of North Carolina.— DEPT. OF BOTANY, SMITH COLLEGE, NORTHAMPTON, MASS, LITERATURE CITED BENKE, H. C. 1928. Two new varieties of early spring plants — a Ranunculus from Missouri and a Heterotheca from Texas. Rho- dora 30: 201. WAGENKNECHT, B. L. 1960. Revision of Heterotheca, section Hetero- theca (Compositae). Rhodora 62: 61-76, 97-107. 1961] Adams, Godfrey — Sagittaria subulata Complex 247 OBSERVATIONS ON THE SAGITTARIA SUBULATA COMPLEX? PRESTON ADAMS AND R. K. GODFREY The Sagittaria subulata complex has been variously inter- preted by recent authors. Small (1933) recognized four species, S. subulata (L.) Buch., S. stagnorum Small, S. lorata (Chapm.) Small, and S. filiformis J. G. Smith. Seven years later Fernald (1940) interpreted the group as a single spe- cies, S. subulata, with four "fairly marked varieties," var. natans (Michx.) J. G. Smith, var. gracillima (Wats.) J. G. Smith, var. lorata (Chapm.) Fernald, and var. typica. The following year Clausen (1941) concurred with Fernald that S. subulata, S. natans, and S. lorata were conspecific. Clau- sen, however, considered Fernald's varieties natans and lora- ta to be phases of a single subspecies, ssp. lorata (Chapm.) Clausen. The variety gracillima was thought by Clausen to be a deep water phase of the tidal mud flat plant, then known as S. subulata ssp. typica. In a recent revision of the genus Bogin (1955) recognized a single species with three varieties. He noted that this species, S. subulata, is "exceedingly plastic" in its vegetative body. His varieties were proposed "only hesitantly” since they “are all similar florally, and merge into one another vegetatively without sharp demarcation in structure or distribution." Bogin’s conception of var. subulata included Fernald’s varieties natans, lorata, and typica as well as Small's S. stagnorum and part of Smith's S. filiformis. In his circumscription of a second taxon, var. gracillima, Bogin included a portion of Small’s S. lorata and part of S. filiformis J. G. Smith. Bogin's third variety included only the Florida plants named as S. kurziana by Glück in 1927. It is apparent from the foregoing brief resume that the Sagittaria subulata complex is not well understood if one may presume that about the same kind of sample was avail- able to each of the investigators. Much of the variance in interpretation is undoubtedly due to the exceedingly great plasticity which these plants, as well as other species in the "This investigation was supported (in part) by a research grant, RG-6305, to the junior author from the Division of General Medical Sciences, Public Health Service. 248 Rhodora [Vol. 63 genus, show in their vegetative responses to different en- vironments and/or environmental fluctuations. The junior author commenced to observe the Sagittaria subulata, complex in the field several years ago while con- ducting general collecting in the northern Florida region. During the summer of 1959 numerous observations were made by Dr. Norlan C. Henderson? and the senior author. Since that time the present authors have continued to study these plants and all of our collective experience includes a not inconsiderable amount of observation in nature where forms that appear to represent the extremes occur, and where certain differing kinds of habitat fluctuations were evident. In addition it includes some greenhouse culture experiments. It is our intent here to report upon this work only insofar as it leads us to offer some hypotheses as to the taxonomy of the complex and to suggest procedures for further intensive investigation to test these ideas. Throughout most of these investigations the authors have entertained the notion that at least three biological entities were involved. One of these, the largest and most conspicu- ous, was thought to be referable to Sagittaria kurziana Glück. Plants of this type are very abundant in the St. Marks and Wakulla river channels (Wakulla County, south of Tallahassee, Florida) as well as in several of the large springs and spring-fed streams in central Florida. A second kind of Sagittaria, considerably smaller in size, was observed on the tidal mud flats and banks along the lower portions of these two rivers. This plant was considered to be the same as the S. subulata of the tidal flats around Chesa- peake Bay and northward. A third aspect with floating dilated leaf blades was also recognizable and referred to Small's S. stagnorum (= S. natans Michx., not Pallas). However, recent observations of the complex, especially at the Wakulla-St. Marks stations, have demonstrated that this idea must be revised. It will be helpful in the following discussion to consider first the kind and nature of the three principal phases ob- Then a graduate student and an N.S.F. Research Participant, now at McMurray College, Abilene, Texas. 1961] Adams, Godfrey — Sagittaria subulata Complex 249 servable within the Sagittaria subulata complex. We think that an analysis of the morphology of the recognizable ex- tremes and the habitats in which they grow will provide a framework around which our other observations can be or- ganized. THE SAGITTARIA KURZIANA PHASE Plants of this type were first brought to the attention of botanists by Hugo Glück (1927) who had been guided to the type locality by Dr. Herman Kurz, then Professor of Botany at Florida State College for Women (now Florida State University) at Tallahassee. Glück and Kurz observed and collected the plant in the Wakulla and St. Marks rivers in 1926. The tremendous size of the phyllodes and scapes greatly impressed Glück who remarked enthusiastically that, to the best of his knowledge, it is the "largest Sagittaria of the whole world!" He noted phyllodes as long as 3.5 meters but the present authors have not encountered any longer than about 2.5 meters.’ The “Sagittaria kurziana" plants are characterized by the great length of their strap-like phyllodial leaves, several of which issue from a very short stem (Fig. 1). Each phyl- lode tends to be broadest above the middle, tapering gradual- ly toward the apex, and is more or less flattened with the veins usually evident as several prominent ridges on the lower surface (Fig. 2). When viewed from underwater these leaves are seen to cover the spring or stream bottom, a scene which brings to mind Agnes Arber's vivid descrip- tion of the deep water phase of S. sagittifolia, a European species. Mrs. Arber wrote (1920, p. 12) that "the ribbon- leaves of Sagittaria . . . have a singular beauty when seen forming, as it were, a meadow beneath the surface of the water, moving in the current in a way that recalls a field of wheat swayed by the wind." The inflorescence scapes of the “S. kurziana" type of plant also become very long, often reaching 15-25 dm. This distinctive Sagittaria seems to be much at home in the clear running water of the Wakulla and St. Marks rivers (and a few of their small tributaries). It is very abundant — The notation on Griscom 21059a (GH) that the leaves reach a length of fifty feet needs verification before being accepted. 250 Rhodora [Vol. 63 in the Wakulla River all the way to its source at the large fresh-water Wakulla Spring. The plant also grows in great abundance in at least ten other comparable springs and spring-fed streams of northern and central Florida. These aquatic habitats are unique, having been produced by a combination of climatic and geological forces (Whitford, 1956). A tremendous volume of water issues forth from Fic. 1-5. Fig. 1. “S. kurziana," habit. Note young plantlet from rhizome. FIG. 2. Phyllode of “S. kurziana,” cross-section, lower surface in face view. Fic. 3. “S. kurzi- ana," plant uprooted and floating. Compare the much elongate original phyllodia with those formed while floating (see text). Fic. 4. “S. subulata,” habit. Note young plant- let from rhizome. Fic. 5. Phyllode of “S. subulata," cross-section, lower surface in face view. 1961] Adams, Godfrey — Sagittaria subulata Complex 251 these giant springs, many having a flow of over three cubic meters per second. Around the springheads are often large open pools, and the water forms deep runs or streams which may be several miles long. Very little surface water enters and the chemical and thermal conditions are relatively stable, the light being the only important environmental variable (Whitford, 1956). The waters are hard in most of these springs, with large amounts of calcium, sodium, potassium and chlorides often present. The temperature is remarkably constant, varying less than 1° C. annually in many of the spring pools (slightly more down stream, of course). Wa- kulla Spring has a temperature of about 24° C. while several of those further south are a few degrees warmer. In the St. Marks and Wakulla rivers the plants of the "Sagittaria kurziana" type grow in water from about one to at least three meters deep (completely submersed where the waters are affected by the tide) forming broad zones. In MeBride's Slough, a spring and its stream that flows into the Wakulla River, the plants often grow throughout the width of the stream, especially near the springhead. Besides the Sagittaria other species of vascular plants are often present, including Najas guadalupensis, Vallisneria sp., and Potamogeton illinoensis. Dense beds of Chara, a large green alga, are also common. The leaves of these plants are coated with dense growths of algae, especially diatoms, filamentous greens, and blue-greens (Whitford, 1956). This investigator found that the algae form recognizable communities, well- developed on the Sagittaria leaves, depending upon the min- eral content of the springs. The importance of these large springs and spring-fed streams in Florida as natural laboratories ready-made for ecological studies under controlled conditions has been shown by Odum (1957). From intensive investigations of com- munity metabolism and productivity in one such spring, the famous Silver Springs near Ocala, Marion County, Florida, Odum found that a remarkable steady state is present and appears to have existed for a very long time. Under these optimum conditions Odum discovered that the Sagittaria plants are responsible for one-third of the primary produc- 252 Rhodora [Vol. 63 tion of organic matter in the spring community. The Sagit- taria no doubt plays a similar important role in the ecology of the other springs where it occurs. THE SAGITTARIA SUBULATA PHASE This distinctive Sagittaria of tidal mud flats and tidal shores was first collected by John Clayton sometime during the early seventeenth century. Clayton lived for many years in Gloucester County, Virginia (Stearn, 1957). While he visited and collected in other parts of Virginia, Clayton could have gotten his specimens of “S. subulata” near his home, since Gloucester County is bounded on the east by Chesapeake Bay and on the west by the York River. The “S. subulata” phase is a small plant, commonly with short phyllodia (Fig. 4) not unlike those of Lilaeopsis chinenses (with which we have known it to be intermixed) and which by means of subterranean runners forms dense mats. In size, the shoots grade from the tiny Lilaeopsis-like ones on the upper tidal shores to a condition where there are rosettes of ascending wider leaves reaching a length of 1.5-3 dm. or more on lower shores (Fig. 10 a-i), submersed even at mean low tide. “Grassy” mats are formed on the bottom which very much resemble those formed by Thalassia in purely saline situations. The phyllodia tend to be more or less lenticular in cross-section, especially in the lower half, and, when fresh, as much as two-thirds to three-fourths as thick as they are wide (Fig. 5). Generally the veins are in- conspicuous, not forming the prominent ridges as in the “S. kwurziana" form. A tendency for the phyllodia to be broad- ened near their apices, producing a narrow but definite blade, is often present in plants of “S. subulata” from New York to North Carolina (Figs. 7, 8, 9). The best development of the “Sagittaria subulata” form occurs on the tidal flats and along river shores which come under the influence of the tides. A few plants have been seen growing completely out of the water but they are probably covered during times of particularly high tide. For several hours each day (during the retreat and advance of the tide) most of the plants are exposed to the drying action of the sun and the air with little, if any, adverse 19611 Adams, Godfrey — Sagittaria subulata Complex 253 effects. Near the junction of the Wakulla and St. Marks rivers the “S. subulata” form is very abundant on the mud flats, often being virtually the only plant in a zone as wide as 30 meters. Seaward from this point even wider expanses of f | f | | \ || | | 10 | || : || nasi | TECH | MEC A MEAN] J n | TA g h ! | l / | Xs / (> \ A D » ^f^ JN J J E í E E lk || | | HUU ' from Collins, Rhode Island (GH). Fic. 7. “S. subulata” from House 25070, Ulster Co., N. Y. (GH). Note widened lamina. Fic. 8. “S. subulata” from Steele Aug. 28, 1896, District of Columbia (us). Fic. 9. “S. subulata”? from Steele Sept. 1, 1900, District of Columbia (us). Fic. 10 a-i. “S. subulata” from Adams 728A, Wakulla Co., Fla. (Fsu). Fic. 11 a-r. “S. kurziana," population G, Wakulla Co., Fla. (See text). “S, subulata var gracillima’ Fic. 6-11. Fic. 6. 254 Rhodora [Vol. 63 the tidal flats are thickly carpeted with this plant. Upstream, further away from the tidal influence, the colonies diminish in extent, until only a few scattered individuals are present, as in the case at the U. S. Rt. 98 highway bridge over the Wakulla River (about seven miles from the Gulf of Mexico). THE SAGITTARIA STAGNORUM PHASE Plants of this taxon were first described by the elder Michaux in his “Flora Boreali-Americana" (1803) under the name of S. natans (an epithet unfortunately predated by S. natans Pallas, a European plant). Typically, many of the phyllodia are dilated near the apex, producing linear-ovate to ovate or even slightly sagittate blades which float upon the surface of the water (Figs. 12, 13). During the winter, the floating leaves disappear, presumably being killed by a thin film of ice which occasionally forms in this region. This medium-sized, very distinctive plant reaches its most typical development in the quiet, relatively shallow waters of sink- hole ponds and pools, lake margins, and swamps and road- side ditches in which water stands for several months. In the small sinkhole ponds which are a characteristic feature of the landscape in northern Florida the “S. stagnorum” phase often forms a zone or ring of considerable width around the margins. When the plants are in flower an un- usual aspect suggesting confetti on the water surface is presented. ADDITIONAL FIELD AND GREENHOUSE OBSERVATIONS Having described in a general way the three main phases detectable within the Sagittaria subulata complex, we now turn our attention to the variations from these principal forms. Lest we be accused of so-called “typological think- ing" may we hasten to assure the reader that this manner of organizing our observations is purely arbitrary. Due to the unusual nature of the problem this particular manner of presentation appears to us to be the best method. We are fully cognizant of the elementary biological principle that populations, subspecies, species — taxa if you please — vary in differing degrees. If the reader so wishes, he may regard each “phase” as representing the midpoint on a variation 1961] Adams, Godfrey — Sagittaria subulata Complex 255 curve; in a sense they can be called by the old-fashioned word “extremes.” Population" A: Marion Co., Fla., near Silver Springs, east of Ocala. In a small drainage ditch which flows into an artificial boat basin near the Silver River (the stream from Silver Springs) Sagittaria plants with a most variable array of phyllodial shapes occur (Henderson and. Adams 2016 July 8, 1959, FSU, 8 sheets). All intermediate forms from strap-like phyllodia to leaves with fairly definite petioles and blades were observed, even on the same plant. The dilated leaf tips extended nearly vertically above the water surface with little or no tendency to become floating. Amidst the Sagit- taria were numerous other kinds of vascular plants, result- ing in the ditch being fairly choked with vegetation. The water depth in the middle of the ditch was about 70 em. On the other side of a small footbridge, at the junction of the drainage ditch and the boat basin, were many Sagittaria plants with only long, strap-shaped phyllodia. Here the water was much deeper, the current fairly swift, and very few other plants were present. Since the Sagittaria in the nearby Silver River and Silver Springs had the “S. kurzi- ana" appearance, this population was interpreted as being a deviation from the typical form of “S. kurziana." More recently, study of similar populations in other areas has strongly suggested that a re-interpretation is necessary (see below). Population B. Taylor Co., Fla. A colony similar in several features to population A was discovered growing in a “black” water, swift-flowing stream out of a Nyssa swamp (8 miles south of Perry, Godfrey and Houk 60460, Oct. 21, 1960, FSU, 13 sheets). Many of the phyllodia were elongate and ribbonlike but all transitions to leaves with broadly dilated blades (Fig. 14 g-p) were pres- ent, even on the same individuals. The leaves on several "The presence of extensive vegetative reproduction by means of rhizomes and corras rendere the collection of true population samples most difficult. Therefore, we have had to be satisfied with taking specimens along transits or by merely walking (or rather wading) about over the stand and picking up plants at irregular intervals (see fuller discussion of this phenomenon below). 256 Rhodora [Vol. 63 plants were cut off about 8-12 cm. above the base and the rootstocks were planted in wet soil contained in aquaria. The tanks were filled with 12-15 cm. of tap water. Within a few days after transplanting, new leaves were being pro- duced (Godfrey 604,60A, March 1961, FSU, 8 sheets). These, as well as subsequent ones, tended to have a widened floating blade with slender “petiole” (Fig. 14 a-f). Such plants have the typical vegetative form characteristic of “S. stagnorum”. Fic. 12-13. “Sagittaria stagnorum,” habit. Note the sagittate leaves. Population C: Levy Co., Fla. A population comparable to the Taylor County area (see B above) was found growing amidst a thick mat of the aquatic grass Hydrochloa caroliniensis in a drainage ditch bordering a cypress swamp (ca. 5.5 miles west of Bronson, Godfrey 60559, March 2, 1961, FSU, 4 sheets). In nature, these plants possessed phyllodia which exhibited much the same gamut of variations in shape as the Taylor County population except that the “petiole” of the leaves with blades were very much shorter and stouter. Within less than two weeks after being transferred to aquaria in the greenhouse, broad, floating leaf blades on narrow petiolar structures were being produced in abundance (the original leaves were not removed in this sample; they soon rotted away in the aquaria). In this example, also, the aquarium grown plants assumed the aspect typical of the "S. stagnorum" phase (Godfrey 60559A, April 17, 1961, FSU, 3 sheets). 1961] Adams, Godfrey — Sagittaria subulata Complex 257 Population D: Leon Co., Fla. For more than a year the senior author has continually observed a colony of Sagittaria growing in the “black” water of a Nyssa-Taxodium swamp. During a period of heavy rains and consequent high water level the plants were in about 0.5 to 1.2 meters deep. Many of the phyllodia were elongate and strap-like but, especially in the shallower water, various forms of widened leaf tips were produced (15 miles north of Tallahassee, Adams 431, April 19, 1960, FSU, 12 TN g 30cm — 48cm 30 cm ALLE [D LU. Fic. 14. “S. stagnorum," phyllodia: population B, Taylor Co., Fla.; a-f, plants grown in aquaria; g-p, plants in nature. (See text). 258 Rhodora [Vol. 63 sheets). About five weeks later, the water level had dropped considerably and the plants presented a much changed ap- pearance (Adams 474, May 21, 1960, FSU, 33 sheets). In the mud along the swamp margin were plants completely ex- posed to the air. Their leaves were very short, semi-erect, and with wide blades. Underground runners from these ter- restial plants were traced out into the swamp bottom for distances of one-half to a meter or more. At intervals in the water new shoots were produced from these runners. The phyllodia were elongate and ribbon-like as well as with dilated, floating tips. Several rootstocks of the submerged individuals were transferred to aquaria and, within two weeks, most of the new leaves were of the floating type (Adams 752, March 30, 1961, FSU, 3 sheets). Population E: Leon Co., Fla. Tremendous variation in the shape of the floating leaf blades is present in many populations referable to the "Sagittaria stagnorum" phase. The plants of one such colony in and on the exposed shores of a small pond had leaves show- ing all imaginable transitions from the ribbon-like phyllode to one with a dilated floating blade having a truncate or even somewhat sagittate base (Fig. 15) (1 mile north of Woodville, Henderson 2003, June 18, 1959, FSU, 5 sheets). A collection from the same pond (Godfrey 57484, Aug. 15, 1958, FSU, 27 sheets) made the previous summer is reveal- ing since much more variability in leaf blade shape and size is present (Fig. 16). In general, the widened lamina were longer than in the Henderson collection while their bases tended to be more sagittate. Such annual changes in a population are of great interest and need to be investigated further. The tendency for production of sagittate leaves is evidently a widespread phenomenon, judging from the avail- able herbarium material. Population F: Wakulla Co., Fla. Valuable information concerning the apparent effects of the local environmental conditions upon the vegetative bodies of the “Sagittaria stagnorum” form has been obtained by transect studies. One colony growing in a wide ditch adja- cent to a Nyssa swamp was sampled by collecting the plants 1961] Adams, Godfrey — Sagittaria subulata Complex 259 at intervals along a transect reaching from the muddy bank out into the ever-deepening water (6 miles south of Craw- fordville, Adams 263, Aug. 15, 1959, FSU, 8 sheets). Plants on the wet soil were dwarfed, their phyllodia rather short and with semi-erect dilated tips (Fig. 17a). Further out in the water (about 5-10 cm. depth) the typical floating blades j | ^ (Us "n SARAN Fig. 15-16. “Leaf” variation of “S. stagnorum,” population E, Leon Co., Fla. Note transition from narrow straplike phyllodia to blades with sagittate bases. 260 Rhodora [ Vol. 63 (Fig. 17 b-e) were found along with some ribbon-like phyllo- dia. With increase in water depth the phyllodia become much more elongate (Fig. 17 f-i) while the floating leaves virtually ceased to be produced (at about 1 meter deep). The length of the scapes varied directly with the water depth, also. Similar observations have been recorded by the authors at numerous other stations. Population G: Wakulla Co., Fla. In McBride’s Slough, a medium-sized spring-fed stream flowing into the Wakulla River, a Sagittaria not unlike the "S. kurziana" phase is extremely abundant (Adams and Mitchell 754, April 21, 1961, FSU, 18 sheets). In the fairly swift current of the deeper portions of the stream these plants have elongate, ribbon-like phyllodia (Fig. 11a, b) and correspondingly long inflorescence scapes. However, in the shallower water near the banks many plants with much 9 h i E a | mn 1 - — meters 0 1 2 3 Fic. 17. Phyllode variation in “Sagittaria stagnorum.” Diagrammatic representation of a transect from the shore out into the water of a drainage ditch. Water depth indicated by the ordinate. Transect distance by the abscissa. Phyllode length not to scale. 1961] Adams, Godfrey — Sagittaria subulata Complex 261 shortened phyllodia exhibiting various degrees of widening at the tips occur (Fig. 11 c-i). At the spring itself, the US. kurziana" aspect is present in the deeper water but a gradual decrease in the size of the plants occurs toward the periphery of the springhead pool. On the soft mud at the edge of the pool, the leaves on individual plants show a gradual increase in dilation of the apex, not infrequently even producing a floating blade (Fig. 11 j-r). Population H: Wakulla Co., Fla. Approximately a mile north of the village of Newport is a sulfur spring which forms a small stream emptying into the St. Marks River. In the swift current of this rivulet there occurs a Sagittaria with elongate strap-like phyllodia hardly distinguishable (if at all) from the "S. kurziana" phase so abundant in the nearby river (see Fig. 1). Along the muddy banks and in quiet backwaters created by large masses of Cicuta (and other vegetation) plants whose phyl- lodia are variously dilated are frequently encountered. Oc- casionally some of these individuals produce phyllodia with widened, floating blades not unlike those characteristic of the *S. stagnorum" phase (Fig. 13). On the muddy banks along the lower portion of this stream (near its junction with the river) are plants which are very difficult to distin- guish from the “S. subulata” form (similar to Figs. 4, 10 a-1) so very abundant further downriver where tidal fluctuation occurs (Adams and Mitchell 755, April 21, 1961, FSU). OBSERVATION ON METHODS OF REPRODUCTION Extensive vegetative reproduction via underground run- ners and corm-like structures is present in the Sagittaria subulata complex. The runners extend from the base of a plant and, after becoming several centimeters long, produce small white swellings similar to a corm or tuber (Figs. 1, 4, 12). From each such structure a new plantlet will arise. Not long afterwards, a new runner may form from the base of the young plant, a process often resulting in several to as many as fifty plants linked together (Wakulla Co., Fla., St. Marks River, 2 miles north of St. Marks, Henderson 2021, July 21, 1959, FSU, 3 sheets). Such a phenomenon is re- sponsible for the formation of vast mats or beds containing 262 Rhodora [Vol. 63 practically nothing but Sagittaria. In the Wakulla and St. Marks rivers large clumps of the “S. subulata” form often become detached from the tidal mud, float about on the surface, and are carried along by the current. Frequently the plants in these floating mats remain alive long enough to flower and produce mature fruits. Individual plants of the "S. kurziana" aspect are occasionally torn loose from their anchorage in the river bottom. If they happen to be- come caught-up in debris along the shore, or in quiet back- waters, very short phyllodia with dilated tips may be pro- duced (Fig. 3). Flowers on greatly shortened inflorescence stalks are often formed, even remaining long enough to mature the achenes. Young plantlets may often form at the lowest node of the inflorescence. Several instances of this phenomenon have been seen on plants in McBride's Slough (Adams 754A, April 21, 1961, FSU). Vegetative reproduction appears to be of paramount im- portance in the maintenance and local spread of plants of the Sagittaria subulata complex. The massive beds such as those formed in the Wakulla and St. Marks rivers replenish their numbers simply by the new plants which form from the corms produced on the subterranean runners. It would ap- pear that, in these dense “carpets and meadows" of Sagit- taria, seedlings would have little chance of becoming estab- lished, presumably being hindered by competition with exist- ing plants which originated from the runners. Reproduction by seeds no doubt occurs but probably plays à secondary role in the life cycle of these plants, at least once a colony becomes established. Re-establishment of the "S. stagnorum” phase following severe drouth resulting in dras- tic and often lengthy desiccation of the habitats is probably due to seeds. The junior author has observed situations where, after a severe drouth cycle of several years, no Sagittaria plants could be found in the dried-up ponds, ditch- es, and swamps. The drought was broken by much autumn rain following which these habitats rapidly became filled with water — and in the following very early spring the "S. stagnorum” form was found to be extremely abundant in the very same places were none was present earlier. This 1961] Adams, Godfrey — Sagittaria subulata Complex 263 observation implies that such rapid re-colonization may have been due to seeds since the underground runners would most likely have been killed by the several years of drouth. Abun- dant seeds are matured by the plants of the S. subulata com- plex. The pedicels of the carpellate flowers become reflexed after pollination (Figs. 1, 3, 4, 12) and the fruits reach maturity underwater. The fruit, a distinctive achene with variously developed wings, becomes detached from the receptacle and floats to the water surface. We have not been able to follow the subsequent events leading to germination of the seeds. NOTES ON FLOWERING Plants of the “Sagittaria kurziana" phase flower through- out the year. The other phases of the S. subulata complex apparently cease flowering in the winter. Beginning in March and April in the Tallahassee region blooming resumes and continues until about November. In Massachusetts S. subulata var. gracillima (= S. stagnorum Small?, see below) begins to flower about the middle of June and continues until the end of July (note attached to C. E. Faxon's specimens in the Gray Herbarium). Usually the pedicillated flowers are held a centimeter or two above the water surface during anthesis, the pollina- tion appearing to be due to small flies. In the tidal flats where the “Sagittaria subulata" phase occurs the flowers open when the tide is low, thus permitting them to be pollin- ated. On the deep water side of the mud flats we have seen partially-opened flowers and these appear not ever to have become emersed. Although in all the phases the staminate flowers usually are emersed at anthesis, there is as yet considerable question as to whether many female flowers (which ultimately produce achenes) are ever emersed. OBSERVATIONS ON THE INFLORESCENCE BRACTS Each whorl of branches in the inflorescence of Sagittaria was described by Bogin (1955) as being “almost always subtended by a ring of 3 bracts." In the S. subulata complex these “bracts” were said to be “typically connate at the base, sheathing, . . . and occasionally spathe-like." Our observa- 264 Rhodora [ Vol. 63 tions on living plants show that these so-called bracts are actually the split and withering portions of a floral envelope which completely encloses the young inflorescence and flower buds. As these organs develop, the floral envelope is split down one side nearly to the base, producing a spathe-like appearance. Frequently, similar tears occur in the other sides, often producing a 3-parted “bract” structure like that illustrated by Bogin. Shortly afterwards, the envelope be- gins to wither and die, usually breaking away, leaving a narrow “ring” of chlorophyllous tissue at the node. The floral envelope may reach a length of 2-3 em. in plants of the “S. kurziana" phase, especially at the lowermost whorl of the inflorescence. A similar envelope covers the corms, and is split by the growth of the young shoot. CHROMOSOME COUNTS Several mitotic chromosome counts on plants identified as members of the Sagittaria subulata complex have been pub- lished. Brown (1946) reported the diploid number of 22 chromosomes for “S. stagnorum Small, “S. Kurziana Glück", and a putative hybrid between plants of these two taxa. A diploid count of 22 chromosomes in plants of the tidal "S. subulata" was published by Baldwin and Speese (1955). Brown (1946) also found 22 chromosomes in somatic tissue of plants which he identified as “var. gracillima." No mention of this report was made by Bogin (1955) who claimed to have discovered “ca. 44" chromo- somes in root tips of “var. gracillima.” REMARKS ON THE STATUS OF S, SUBULATA VAR. GRACILLIMA Since being described by Sereno Watson (1890), this Sagittaria with very narrow elongate phyllodia has been subjected to a variety of opinion concerning its biological distinctness and its taxonomic standing. Most authors have recognized its close affinity with the S. subulata complex but their interpretations of its alliance within the group have varied considerably. Watson considered the plant to be a variety of S. natans Michx. (S. stagnorum Small) while Smith (1895) treated it as a variety of S. subulata (L.) Buch. Many years later Fernald (1940) concurred with 1961] Adams, Godfrey — Sagittaria subulata Complex 265 Smith's opinion but the following year Clausen (1941) re- garded var. gracillima as a deep water phase of the tidal flat plant S. subulata. Recently, Bogin (1955) and Beal (1960) have maintained gracillima as a variety of S. subu lata. Exception to the prevailing opinion concerning the rela- tionship of var. gracillima was made by Pierce (in Gleason, 1952). This author concluded that the plant was more prop- erly interpreted as a *deep water form" of Sagittaria grami- nea Michx. Our observations of the Sagittaria subulata complex strongly suggest that var. gracillima may represent merely the deep water aspect of "S. stagnorum.” We have seen many plants of the "S. stagnorum" phase in Florida and Georgia, both living and on herbarium sheets (Figs. 14d, 17i), which are hardly distinguishable from New England specimens (Fig. 6) traditionally classified as this variety. As discussed above, plants of “S. stagnorum” growing in water about a meter or more deep possess very elongate narrow phyllodia (Figs. 14 A, 17 i). The production of dilated leaf blades by plants of var. gracillima from Massachusetts simi- lar to House 25070 (Fig. 7) further suggests its close affini- ty with the “S. stagnorum” phase of the S. subulata complex. CONCLUDING REMARKS Despite the not inconsiderable amount of observation de- tailed in the foregoing discussion, we still do not feel justified in assigning taxonomic status to the three relatively distin- guishable phases (Figs. 1, 4, 12, 13) in this complex. We believe that their biological discreteness, or lack of it, must be investigated more thoroughly, probably entailing the following: a more intensive and extensive analysis of popu- lations or stands in relation to variable and fluctuating environments ; reciprocal transplants ; experimental culture; more exact and more extensive cytological study ; and, if possible, crossing experiments. The universal occurrence of vegetative reproduction and the consequent impossibility of discerning individual plants in this complex certainly require that population analysis be handled differently from 266 Rhodora [Vol. 63 that in groups whose populations are comprised of manifest-- ly discrete individuals. Possibly plants can be grown satis- factorily from seed thus yielding information not otherwise available. — DEPARTMENT OF BOTANY AND DACTERIOLOGY, DE- PAUW UNIVERSITY AND DEPARTMENT OF BIOLOGICAL SCIENCES, FLORIDA STATE UNIVERSITY. LITERATURE CITED ARBER, AGNES. 1920. Water Plants. Cambridge. BALDWIN, J. T. JR. AND B. M. SPEESE. 1955. Chromosomes of taxa of the Alismataceae in the range of Gray's Manual. Am, Jour. Bot. 42: 406-411. BEAL, E. O. 1960. The Alismataceae of the Carolinas. Jour. Elisha Mitchell Soc. 76: 68-79. BoGIN, CLIFFORD. 1955. Revision of the genus Sagittaria ( Alismata- ceae). Mem. N. Y. Bot. Gard, 9: 179-233. Brown, W. V. 1946. Cytological Studies in the Alismaceae. Bot. Gaz. 108: 262-267. CLAUSEN, R. T. 1941. "The variations of Sagittaria subulata. Tor- reya 41: 161-162. FERNALD, M. L. 1940. A century of additions to the flora of Virgin- ia. Part II. Rhodora 42: 408-409. GLEASON, H. A. 1952. The New Britton & Brown Illustrated Flora 1: 88-92. GLUCK, HuGo. 1927. A new Sagittaria from Florida: Sagittaria Kurziana. Bull. Torrey Club 54: 257-261. ODUM, H. T. 1957. Trophic Structure and Productivity of Silver Springs, Florida. Ecol. Monog. 27: 55-112. SMALL, J. K. 1933. Manual of the southeastern Flora. 22-26. SMITH, J. G. 1895. A revision of the North American species of Sagittaria and Lophotocarpus. Mo. Bot. Gard. Sixth Ann. Report. 1-64. STEARN, W. T. 1957. An introduction to the Species Plantarum and cognate botanical works of Carl Linnaeus. London. WATSON, SERENO. 1890. In Gray, Man. Bot, ed. 6, p. 556. WHITFORD, L. A. 1956. The communities of algae in the springs and spring streams of Florida. Ecology 37: 433-442. Volume 63, No. 752, including pages 207-236, was issued September 11, 1961. BLOW REFERENCE LIBRARY NOV 2 0 1961 Dodota JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Associate Editors Vol. 63 October, 1961 No. 754 CONTENTS: The Chromosomes of Podophyllum peltatum. Gabriele N. Mühling and G. B. Wilson ........... eene 267 A List of Algae from Selected Areas in Massachusetts. Faodar Pee Webber cM eo Rn Se 275 Further Notes on Chamaecyparis thyoides in ; New Hampshire. Henry I. Baldwin c.cccccccccccsscsesessseseseeseeeees 281 Which Side is Up? A Look at the Leaves of Oryzopsis. EAU GCs C VOSS cate c TIR E M 285 A Check List of Walter Deane’s Seedling Collection. JAN) CNR Jod ORE ee AEE eA ien ee OE ore 287 The Bulbiferous Ranunculus ficaria. Peter S. Green. and Joab L. Thomas anenee eaaa at 289 A Useful Multilingual Botanical Dictionary. IDOI JUD ote NT RR EE ENEMIES 29] A Notable Assemblage of Plants in New Hampshire. AVE TANIERI ECOL (rs ate a tibt ere ee 4 The Specific Epithet of the Pecan. John W. Thieret ........ 296 Che New England Botanical Club, Ine. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume, Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able. All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. n — mene — — t deum lle. Jud. dno» E-r e o0 Second Class Postage Paid at Boston, Mass. Printed by THE LEXINGTON PRESS, INC. Lexington, Mass. Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 October, 1961 No. 754 THE CHROMOSOMES OF PODOPHYLLUM PELTATUM GABRIELE N. MÜHLING AND G. B. WILSON! Podophyllum peltatum L., commonly known as May Apple, Mandrake or Pomme de Mai, represents one of the best sources of cytological material for class purposes. It is found in eastern North America, east from a line drawn from Min- nesota to Texas, including southern Ontario and Quebec. Representatives of the genus also occur in eastern Asia (Fernald, 1950). Large amounts can be collected easily and the sporogenous material is normally found to be in meiosis for about a two week period. Good cytological preparations can be made by simple techniques. The early stages of meio- sis especially pachytene can be fixed and stained giving a clarity that equals maize. The chromosomes are large and few in number. The advantages of the material do not seem to be recog- nized very generally probably because the cytology has not been described in any detail since 1926 when Kaufmann published his general description of the karyotype and meio- sis. It therefore seemed worthwhile reporting and illustrat- ing the work from our own collections of this material. MATERIALS AND METHODS Flower buds of Podophyllum peltatum L. were collected in the Oakland and Ingham Counties of Michigan. Collections "Department of Botany, Montana State University, Missoula, Montana; and Depart- ment of Botany and Plant Pathology, Michigan State University, East Lansing, Michi- gan. The authors wish to thank Mr. P. G. Coleman for his preparations of the illustra- tions. 267 268 Rhodora [Vol. 63 of anthers from these areas have been made yearly for the past six years. The most extensive collections were made in 1954 and 1959 and the main part of the data for this paper were obtained from these. Buds from the former year were fixed in the field in 3 parts absolute ethyl alcohol and 1 part glacial acetic acid and from the latter year in a 6:3:2 mix- ture of absolute methanol, chloroform and propionic acid (Piennar, 1955) and then placed under vacuum as soon zs possible. Microspore mother cells of anthers were prepared for analysis mostly by staining by the Feulgen technique after a 9-10 minute hydrolysis in 1 N HCl at 60° C. Each anther was carefully squashed on a slide to separate the sporogenous tissue from the tapetum. Slide preparations were dehydrated in a 9:1 mixture of tertiary butyl alcohol and absolute ethyl alcohol overnight and then mounted in diaphane. Some material was stained with aceto-carmine. Fixed anthers were macerated in 1 N HCI at 609 C for 5 minutes prior to staining. Before dehydration, excess car- mine was removed from preparations by running a drop of 45% acetic acid under the cover slip. If this is not done the cytoplasm becomes dark, often obscuring the division figures. DISCUSSION 1. Karyotype: Our observations of a haploid number of six chromosomes in P. peltatum are in agreement with those earlier reported by Litardiére (1921) and Kaufmann (1926). The kinetochores are found in three positions — each type appearing twice in the karyotype. In two chromo- somes the kinetochores are in a near-median position ; in two, in a submedian position and in two, in a subterminal posi- tion. The relative lengths of the arms of the chromosomes measured at anaphase II in the above order are: approxi- mately 1:1; 1:2; and 1:18. For convenience the three types shall be called A, B, and C respectively. One chromosome of each of the three types has a satellite. It appears on the long arm in types B and C (Plate 1262, fig. 7, 8 and 9). It is not possible to determine which arm contains the satellite in type A because of their near identical length and the lack of any other morphologically distinguishing characteristic 1961] Muhling and Wilson — Podophyllum 269 (Plate 1262, fig. 9). In some details this description of the satellites does not agree with that of Kaufmann (1926) which may be because of the difficulty in detecting them or because a different population was sampled. A number of 3 s ? E 8 9 PLATE 1262. Fig. 1. Pachytene showing nucleolar attachment. Fig. 2. Zygotene with DNA blobs. Fig. 3. Diakinesis and faint outline of nucleolus. Fig. 4. Interlocking chromosomes. Fig. 5. Probably a metaphase I with numerous fragments. Fig. 6. First anaphase bridge and fragment. Fig. 7-9. Anaphase I. Arrows point to chromosomes with satellites. 270 Rhodora [Vol. 63 peculiarities are noted in the occurrence of these satellites. First of all we did not observe them in somatic anaphases when they generally do appear in material which has them. Furthermore frequently a satellite is only visible in one chromatid of a first division anaphase chromosome, its part- ner having no indication of this structure whatsoever (Plate 1262, fig. 9). The total length of the complement of six chromosomes of this species is about 60 microns. The lengths of each of the chromosomes expressed in terms of the total length of the complement is approximately as follows: the median at- tached chromosome, 20% ; submedian 17% ; and subtermi- nal, 13%. 2. General description of meiosis: The key stages are shown in the figures in Plate 1263. They are consistent with the generalized textbook descriptions. Pachytenes are par- ticularly good. In cells where they are sufficiently spread out, it is possible to count the number of chromomeres in a chromosome. The interkinesis between the two meiotic divisions is well marked, sometimes being so complete that nucleoli appear although this seems to be somewhat unusual. Cytokinesis does not occur until the end of second division so that diads are sometimes confused with binucleate tapetal cells since they are somewhat similar in size and shape. A number of special features of the material deserve some comment: a. Thenucleolus: In somatic tissue the number of nucleo- li appears to vary from one to three. So far we have failed to determine with which chromosomes they are associated. In meiosis there is consistently a single nucleolus, which is associated with one of the two subterminally attached chrom- osomes. Although the satellite can not be seen all of the time, it is assumed that the nucleolar association is with the chromosome possessing the secondary constriction. The nucleolar organizer region appears to be in the proximal third of the long arm and is not associated with any obvious secondary constriction. In any case, the three secondary constrictions noted do not seem to be concerned with nucleo- 1961] Muhling and Wilson — Podophyllum 271 lar formation. In some plants such as maize (McClintock, 1934) the nucleolus develops from a definite organizing region in the area of the secondary constriction. b. Some microsporocytes that appear to be between zygo- tene and pachytene frequently show a large number of Feul- gen positive bodies (Plate 1262, fig. 2). Similar observa- PLATE 1268. Stages of meiosis in Podophyllum peltatum. Fig. 1. Zygotene; 2. Pachy- tene; 3, Diplotene; 4, Diakinesis; 5, Metaphase I; 6, Anaphase I; 7, Telophase I; 8, Prophase II; 9, Metaphase II; 10, Anaphase IT; 11, Late anaphase II; 12, Telophase I 212 Rhodora [Vol. 63 tions have been made in Trillium (Sparrow & Hammond, 1947) and Lilium (Cooper, 1952 and Takats, 1959). Spar- row and Hammond (1947) suggest that this represents a transfer of nuclear DNA to the cytoplasm. Similarly Cooper suggests that this represents movement of DNA from the tapetum to the locules and ultimately into the nuclei of the microsporocytes. After an extensive examination of this phenomenon, Takats (1959) concludes that there is no evidence for transfer of DNA to the microsporocytes from the tapetum during this stage in meiosis. He suggests that the extrusion may be caused by such factors as injury to the anthers at time of harvest and type of fixative used. We are inclined to take the view that they represent an abnormality of some sort and doubt that such cells proceed through meio- sis. Occasionally observations such as the configuration shown in Plate 1262, fig. 5 in which there is very considera- ble fragmentation suggests the possibility of a relationship between the abnormal zygotenes and pachytenes and the cells showing the chromatin pieces. Similar extreme frag- mentation beginning at first metaphase is reported by Gentcheff and Gustafsson (1940) in an apomict, Hieracium robustum. c. Populations examined so far quite consistently have re- vealed a first anaphase bridge and fragment suggesting heterozygosity for an inversion in what appears to be the median attached chromosome (Plate 1262, fig. 6). d. Chromosome interlocking as illustrated in Plate 1262, fig. 4 also occurs occasionally. 3. Chiasma frequency: Counts of the chiasma frequency at diakinesis were made in samples of both the 1954 and 1959 material. The average of the former year is 10.9 + .83 and the latter year 11.6 = .86, the difference not being sig- nificant. Some terminalization takes place between early diakinesis and first metaphase but the number of chiasmata is not notably decreased till the beginning of first anaphase. 4. Tapetal cells: It is quite common for the tapetal cells in plants to show considerable aberration with reference both to mitosis and number of chromosomes (D'Amato, 1961] Muhling and Wilson — Podophyllum 213 1952). P. peltatum is particularly good material for illus- trating aberrations. Some of the more striking ones are shown in Plate 1264. The tapetal cells begin division con- current with the onset of meiosis and these first divisions are generally normal. By diplotene, tapetal divisions begin to show abnormalities in the form of scattered metaphase configurations and the frequency of divisions is on the in- crease. Between first metaphase and second prophase, the rate of mitosis in the tapetum seems to reach its peak and also the degree of aberration has increased considerably. Polyploidy is very common now, often to the octoploid level and in a few cases probably much higher. In some apparent- 4 5 6 PLATE 1264. Abnormal stages in mitosis of the tapetum. Fig. 1-2. Polyploid pro- phase. Fig. 3. Polyploid metaphase. Fig. 4. “Scattered” metaphase. Fig. 5. Metaphase. Fig. 6. Tri-polar anaphase. 274 Rhodora [Vol. 63 ly clear cases the number is not an exact multiple of the basic number but one or more chromosomes is missing. Other characteristics are multinucleate cells with varying numbers of chromosomes in the different nuclei, star metaphases, star anaphases with three or more centers of aggregation. By the time tetrads are formed there is a definite decrease in the tapetal divisions and an over-all multinucleate condition exists in the interphases of this tissue. Indeed almost all of the mitotie aberrations which have been associated with chemical treatments or described for cancer cells are charac- teristically found in the tapetal cells of this plant. The question may be asked whether the abnormalities of the tapetum arise as a result of its function. In her studies on Solanum tuberosum Avanzi (1950) suggests that the tapetum definitely has a nutritive function and that the abnormalities especially in chromosome number, occur at the time when the sporogenous tissue needs its greatest food supply. Taylor (1959) working on Lilium longiflorum came to the conclusion that the primary function of the tapetal layer is to secrete material for wall formation of the micro- spores, Further work needs to be done on this intriguing phenomenon in the tapetum to determine its cause. SUMMARY The above account describes microsporogenesis in an- thers of Podophyllum peltatum L. and the concurrent mitotic activity of the tapetum. Some peculiarities related to these events are also mentioned. AVANZI, M. G. 1950. Endomitosi e mitosi a diplocromosomi nello sviluppo delle cellule del tappeto di Solanum tuberosum L. Cary- ologia 2: 205-222, Cooper, D. C. 1952. The transfer of desoxyribonucleic acid from the tapetum to the microsporocytes at the onset of meiosis. Amer. Naturalist 86: 219-229. D’AMATO, F. 1952. Polyploidy in the differentiation and function of plant tissues and cells. Caryologia 4: 311-358. FERNALD, M. L. 1950. Gray's Manual of Botany, eighth edition. American Book Company. GENTCHEFF, G. AND A. GUSTAFSSON. 1940. The balance system of meiosis in Hieracium. Hereditas 26: 209-249. 1961] Webber — Algae From Massachusetts 219 KAUFMANN, B. P. 1926. Chromosome structure and its relation to the chromosome cycle. II. Podophyllum peltatum. Amer. Ji Bot: 13: 355-363. LITARDIÈRE, R. DE. 1921. Remarque au sujet de quelques processus chromosomiques dans les noyaux diploidiques du Podophyllum peltatum L. Compt. Rend. Acad. Sci. Paris 172: 1066-1068. MCCLINTOCK, B. 1934. The relation of a particular chromosomal element to the development of the nucleoli in Zea mays. Z. Zell- forsch. 21: 294-328. PIENNAR, R. DEV. 1955. Combinations and variations of techniques for improved chromosome studies in the Gramineae. Jour. South African Bot. 21: 1-8. SPARROW, A. H. AND M. R. HAMMOND. 1947. Cytological evidence for the transfer of desoxyribose nuclei acid from nucleus to cyto- plasm in certain plant cells. Amer. J. Bot. 34: 439-445. TAKATA, S. T. 1959. Chromatin extrusion and DNA transfer during microsporogenesis. Chromosoma 10: 430-453. TAYLOR, J. H. 1959. Autoradiographic studies of nucleic acids and proteins during meiosis in Lilium longiflorum. Amer. J. Bot. 46: 477-484. A LIST OF ALGAE FROM SELECTED AREAS IN MASSACHUSETTS: EDGAR E. WEBBER For the past thirty years, work on the flora of Worcester County has been centered upon vascular plants. Numerous references may be found elsewhere. Recently, attention has been directed to other forms of plant life (1, 3). However, little is known concerning the algae of the county. In 1899 Stone (6) included a list of algae found in Lake Quinsiga- mond as part of a floristic study of that area. Auyang (2) has recently completed a survey of the algae in Lake Quinsig- amond. These two works appear to be the only ones to date dealing with the algal flora of the county, but they both are restricted to only one location. The writer (7) has just completed an ecological study of the algal populations in eleven selected stations in Worcester 1The taxonomic list, slightly modified, from a thesis, “The Ecology of Some Attached Algae in Worcester County, Massachusetts,” presented to the Faculty of the Graduate School of Cornell University in partial fulfillment of the requirements for the degree of Master of Science. 276 Rhodora [Vol. 63 County. No attempt will be made here at ecologically charac- terizing these stations; a listing of them will suffice. STATION NAME AND LOCATION 1 Eames Pond, Oxford, Mass. 2 Eddy Pond, Auburn, Mass. 3 The Cataracts, Worcester, Mass. 4 Silver Spring Brook, Paxton, Mass. 5 The Quag, Sterling, Mass. 6 Chaffin Pond, Holden, Mass. 7 Brook near Reservoir no, 4., Paxton, Mass. 8 Stream near Reservoir no. 2., Leicester, Mass. 9 Indian Lake, Worcester, Mass. 10 A tire depression, temporary habitat, near sta- tion 9. 11 Sargent Pond, Leicester, Mass. Following is a list of algal species collected in Worcester County during the course of the ecological study mentioned above (7). The collecting station appears in parentheses after each species. The taxonomic scheme is that used by Prescott (4) ; the desmids and diatoms, since they are not included in Prescott (4), are classified following Smith (5). CHLOROPHYTA CHLAMYDOMONADACEAE: Chlamydomonas sp. (6). VOLVOCACEAE: Pandorina morum (Muell.) Bory (2-b, 5-a, 5-b) ; Goni- um sociale (Duj.) Warming (10). HAEMATOCOCCACEAE: Haematococcus sp. (5; isolated pockets along shore). ULOTRICHACEAE: Ulothrix subconstricta G. S. West (11-b) ; U. tener- rima Kuetz. (3, 7-a); U. variabilis Kuetz. (6, T-a, 11-b); U. sp. (? cylindrieum) Prescott (7-a); Binuclearia tatrana Wittr. (11-a). MICROSPORACEAE: Microspora tumidula Hazen (T-a); M. sp. (4). CYLINDROCAPSACEAE: Cylindrocapsa geminella var, minor Hansg. (5-a). CHAETOPHORACEAE: Stigeoglonium aestivale (Hazen) Collins (1-b); S. lubricum (Dillw.) Kuetz. (2-a, 7-b); S. tenue (Ag.) Kuetz. (8-c); S. sp. (? glomeratum (Hazen) Collins or subsecundum Kuetz.) (7-b). COLEOCHAETACEAE: Coleochaete scutata Breb. (5-a); Chaetosphaeri- dium globosum Klebahn (6); Draparnaldia glomerata (Vauch.) C.A. Ag. (8-a); Aphanochaete repens A. Br. (11-a). CLADOPHORACEAE: Rhizoclonium hieroglyphicum (Ag.) Kuetz. (2-a). OEDOGONIACEAE: Bulbochaete scrobiculata Tiff.) Tiffany (11-a); B. sp. (? elatior) Prings. (11-a) ; B. sp. (? mirabilis) Wittr. (11-a) ; B. sp. (? setigra (Roth) C.A. Ag. or insignis Prings.) (11-a) ; Oedogo- 1961] Webber — Algae From Massachusetts 277 nium boscii (LeCl.) Wittr. (3); O. gracilius (Wittr.) Tiff. (6); O. reinschii Roy (6); O. subsexangulare Tiff. (11-a); O. sp. (? hystrici- num Trans. & Tiff.) (11-b) ; O. sp. (1-a, 1-b, 2-a, 2-b, 5-b). CHLOROCOCCACEAE: Golenkinia paucipina West & West (2-b). CHARIACEAE: Characium ambiguum Herm. (1-a, 2-a, 7-a, 7-b); C. naegelii A. Br. (7-b, 8-b) ; C. stipitatum (Bachm.) Wille (11-a) ; C. sp. (? pringsheimii) A. Br. (6). HYDRODICTYACEAE: Pediastrum araneosum var. rugulosum (G. S. West) G. M. Smith (6); P. biradiatum Meyen (2-b); P. boryanum (Turp.) Meneg. (1-a, 2-a, 2-b, 5-a, 6, 11-a) ; P. boryanum var. longi- corne Raciborski (5-a); P. boryanum var. undulatum Wille (1-a); P. duplex Meyen (l-a, 6); P. duplex var. clathratum (A. Br.) Lag. (l-a, 2-a); P. duplex var. cohaerens Bohl. (6); P. duplex var. gracillimum West & West (11-b); P. obtusum Lucks (6, 11-b) ; P. sculptatum G. M. Smith (5-a); P. tetras (Ehr.) Ralfs (l-a, 2-a, 2-b, 5-a, 5-b, 6, 11-a); P. tetras var. tetraodon (Corda) Rab. (1-c, 5-a, 5-b, 6, 11-a, 11-b); Sorastrum americanum (Bohl.) Schmidle (6); S. americanum var. undulatum G. M. Smith (6, 11-b); S. spinulosum Naeg. (6). COELASTRACEAE: Coelastrum cambricum Arch. (2-a, 11-b); C. micro- porum Naeg. (2-a, 5-a, 6, 8-a, 11-a, 11-b); C. sphaericum Naeg. (1-c, 2-a, 2-b). OOCYSTACEAE: Dictyosphaerium pulchellum Wood (2-a); Trochiscia obtusa (Reins.) Hansg. (6); T. reticularis (Reins.) Hanse. (6, 8-b); T. sp. (1-b, 4, 5-b) ; Eremosphaeria viridis DBy. (6); Oocystis crassa Wittr. (6); O. elliptica W. West (6); O. solitaria Wittr. (6); Neph- rocytium agardhianum Naeg. (5-a, 6, 11-b); N. obesum West & West (6); Dimorphococcus lunatus A. Br. (6); Ankistrodesmus convolutus Corda (2-a); A. falcatus (Corda) Ralfs (1-c, 6, 11-a) ; A. falcatus var. acicularis (A. Br.) G. S. West (11-a); A. falcatus var. mirabilis (West & West); G. S. West (2-a); A. spiralis (Turner) Lem. (6, 8-a); Selen- astrum minutum (Naeg.) Collins (5-a, 11-a); S. sp. (6); Kirchneriella lunaris (Kirch.) Moebius (6); K. obesa var. major (Barnard) G. M. Smith (2-a, 6); K. subsolitaria Schmidle (2-a, 6); Quadrigula lacus- tris (Chod.) G. M. Smith (6, 11-b); Tetraedron caudatum (Corda) Hansg. (6, 11-b) ; T. minimum (A. Br.) Hansg. (2-a, 2-b, 11-a, 11-b) ; T. obesum (West & West) Wille (6); T. trigonum (Naeg.) Hansg. (6); T. tumidulum (Reins.) Hansg. (11-b). SCENEDESMACEAE: Scenedesmus abundans (Kirch.) Chod. (2-a, 2-b) ; S. acuminatus (Lag.) Chod. (8-a); S. acutiformis Schroeder (L-a, 2-a, 2-b, 6, 11-b); S. arcuatus Lem. (2-a, 6); S. arcuatus var. capitatus G. M. Smith (2-b); S. arcuatus var platydisca G. M. Smith (2-a 6, 11-a, 11-b); S. armatus (Chod.) G. M. Smith (6); S. bijuga (Turp.) Lag. (1-a, 2-a, 2-b, 5-a, 5-b, 6, 8-a, 8-b, 11-a) ; S. bijuga var. alternans (Reins.) Hansg. (6); S. brasiliensis Bohlin (1-c, 2-b, 8-b, 11-b); S. carinatus (Lom.) Chod. (1-a); S. caudatus Corda (1-c, 2-a); S. cau- datus var. typicus Kirch. (1-c) ; S. denticulatus Lag. (2-a, 2-b, 6, 8-a) ; 278 Rhodora [Vol. 63 S. dimorphus (Turp.) Kuetz. (1-a, 2-a, 2-b, 6, 8-b) ; S. incrassatulus Bohlin (1-a) ; S. longus Meyen (5-b) ; S. obliquus (Turp.) Kuetz. (2-a, 2-b, 5-a, 6, 8-a); S. opoliensis P. Richter (2-b); S. quadricaudata (Turp.) Breb. (l-a, 1-c, 2-a, 2-b, 5-a, 5-b, 6, 8-a, 11-b); S. quadricau- data var. parvus G.M. Smith (1-a); S. quadricaudata var. quadrispina (Chod.) G. M. Smith (5-b); S. quadricaudata var. westii G. M. Smith (2-a, 5-b); Actinastrum sp. (? hantzchii) Lag. (2-a); Tetradesmus wisconsinense G. M. Smith (6). ZYGNEMATACEAE: Mougeotia scalaris Hass. (8-b); M. sp. (l-a, 1-b, 1-c, 2-b, 5-a, 6, 7-a, 7-b, 8-a, 8-b, 8-c, 11-b); Spirogyra hyalina Cleve (6); S. rivularis Rab. (2-b); S. sp. (l-a, 1-b, 2-a, 2-b, 2-c, 5-a, 5-b, 6, 7-b, 8-b, 11-b) ; Zygnema sp. (l-a, 1-b, 5-a, 5-b, 6, 8-a, 8-b, 11-b) ; Zyg- nemopsis desmidioides (West & West) Transeau (5-a). MESOTENIACEAE: Gonatozygon aculeatum Hastings (6); G. brebis- sonii Meneg. (11-a); Cylindrocystis americana West (8-c); C. brebis- sonii Meneg. (11-a); Netrium digitus (Ehr.) Itz. & Rothe (1-c, 5-a, 5-b, 6, 8-c) ; N. digitus var. constrictum West & West (6); N. naegelii (Breb.) West & West (6, 8-b); N. sp. (? oblongum) (DBy) Luet. (8-b) ; Spirotaenia condensata Breb. (1-a). DESMIDIACEAE: Closterium abruptum West (6); C. acerosum (Schrank) Ehr. (6); C. acuminatum Ralfs (1-c, 3, 4) ; C. cornu Ehr. (11-a) ; C. ehrenbergii Meneg. (5-a); C. jenneri Ralfs (1-c, 2-a, 2-b, 4); C. juncidum Ralfs (6); C. lanceolatum Kuetz. (6) ; C. leibleinii Kuetz. (1-a, 5-b, 6, 7-a) ; C. libellula Focke (1-a, 5-b) ; C. libellula var. intermedium Roy & Bis. (6) ; C. moniliferum Ehr. (4, 6, 8-b, 8-c) ; C. parvulum Naeg. (1-e, 3, 5-a, 5-b, 6, 8-b, 8-c, 11-a) ; C. regulare Breb. (6); C. rostratum Ehr. (1-a, 4, 5-a) ; C. setaceum Ehr. (6, 11-b); C. strigosum Ehr. (6); C. subtruncatum West & West (6); C. venus Kuetz. (1-a, 3, 6, 8-b) ; Penium navicula Breb. (6); P. sp. (? closter- oides) Ralfs (1-b); P. sp. (6) ; Pleurotaenium coronatum (Breb.) Rab. (6); P. maximum (Reins.) Lund (11-a); P. trabecula (Ehr.) Naeg. (5-a, 5-b, 6); Triploceras gracile Bail. (6); Euastrum abruptum Nord. (1-c, 6); E. abruptum forma minus West & West (11-a, 11-b); E. bi- dentatum Naeg. (6); E. binale (Turp.) Ehr. (6); E. binale forma minor G. S. West (11-a); E. elegans Kuetz, (3, 5-a, 11-a) ; E. evolutum West & West (6); E. evolutum var. integrius West & West (6); E. gemmatum Breb. (6) ; E. insulare (Wittr.) Roy (3, 6, 11-b) ; E. pictum Borg. (1-a) ; E. pulehellum Breb. (6, 11-a, 11-b) ; E. sinuosum var. re- ductum West & West (11-a, 11-b); Cosmarium amoenum Breb. (5-b, 6); C. amoenum var. tumidum Wolle (6) ; C. bioculatum Breb. (11-a, 11-b); C. boeckii Wille (3, 6, 11-a, 11-b) ; C. botrytis Meneg. (2-a) ; C. broomei Thwaites (1-a); C. circulare Reins. (1-c, 6, 8-c) ; C. coelatum Ralfs (3); C. connatum Breb. (6) ; C. contractum Kirch. (2-a, 6); C. excavatum Nord. (6); C. galeritum Nord. (2-a); G. globosum Bulnh. (5-a, 6) ; C. impressulum Elfv. (6); C. intermedium Delp. (5-a, 5-b); C. margaritatum (Lund) Roy & Bis. (5-a, 6) ; C. moniliforme (Turp.) 1961] Webber — Algae From Massachusetts 219 Ralfs (1-a); C. meneghinii Breb. (5-a, 6) ; C. naeglianum Breb. (6) ; C. nitidulum DeNot. (5-a); C. orbiculare Ralfs (5-b, 6); C. ornatum Ralfs (2-a); C. ornatum var. protractum Wolle (1-a); C. pachyder- mum Lund (6); C. portianum Arch. (l-a, 6, 11-b); C. punctulatum Breb. (6) ; C. pyramidatum Breb. (6) ; C. quadrum Lund (6); C. reg- nelii Wille (11-b); C. regnesii Reins. (6, 11-b) ; C. renniforme (Ralfs) Arch. (11-a); C. repandum forma minor West & West (6) ; C. specio- sum Lund (2-a); C. sublobatum Arch. (6) ; C. tumidum Lund (6); C. undulatum Corda (1-a, 1-c, 2-a, 3, 5-a, 5-b, 6, 8-c, 11-a, 11-b) ; C. undula- tum var. minutum Wittr. (5-a); C. undulatum (? var. wollei) West (5-b, 11-a) ; C. viride (Corda) Josh. (5-a, 6) ; C. sp. (? gonatum) Breb. (2-a); C. sp. (? pseudoconnatum) Nord. (8-c); Micrasterias ameri- cana (Ehr.) Ralfs (1-c); M. laticeps Nord. (6); M. mahabuleswaren- sis Hobson (5-a) ; M. pinnatifida (Kuetz.) Ralfs (6) ; M. radiata Hass. (5-b, 6) ; M. sol (Ehr.). Kuetz. (6); M. truncata (Corda) Breb. (5-a, 5-b); Xanthidium antilopaeum (Breb.) Kuetz. (1-b, 6); X. antilo- paeum var. minneapoliense Wolle (6) ; X. antilopaeum var. polymazum Nord. (5-a, 6); Staurastrum alternans Breb. (3); S. arctison var. glabrum West & West (5-b); S. aspinosum Wolle (6); S. bienianum var. ellipticum Wolle (5-b); S. brebissonii Arch. (6); S. brevispinum Breb. (5-a, 11-a) ; S. erenulatum (Naeg.) Delp. (6) ; S. dejectum Breb. (5-a); S. dickiei Ralfs (5-b); S. dilatatum Ehr. (2-a, 5-a, 5-b); S. gracile Ralfs (2-a, 5-a, 5-b, 6); S. granulosum (Ehr.) Ralfs (11-b) ; S. leptocladum Nord. (1-a, 5-a, 5-b, 6, 11-b) ; S. leptocladum var. diver- gens Nord. (1-c, 6) ; S. margaritaceum (Ehr.) Meneg. (5-a); S. mega- canthum Lund (6, 11-a); S. odontatum Wolle (6); S. paradoxum Meyen (11-a) ; S. pilosum Arch. (3) ; S. tetracerum Ralfs (6) ; Arthro- desmus convergens Ehr, (6); A. incus (Breb.) Hass. (1-c, 11-a) ; A. incus var. extensus Anderson (11-b); A. octocornis Ehr. (11-a, 11-b) ; A. triangularis Lag. (11-a, 11-b) ; Spondylosium papillosum West & West (11-a, 11-b) ; S. planum (Wolle) West & West (6, 11-a, 11-b); S. pulchellum Arch. (11-a, 11-b) ; S. pulchrum (Bail.) Arch. (6); Hy- alotheca dissiliens (Smith) Breb. (5-a, 6, 8-c) ; H. undulata Nord. (6) ; Onychonema filiforme (Erh.) Roy & Bis. (6); Sphaerozosma excava- tum Ralfs (6); S. filiforme Rab. (6); S. granulatum Roy & Bis. (6, 11-a); Desmidium aptogonum Breb. (6); D. baileyi (Ralfs) Wolle (6) ; D. grevelii (Kuetz.) DBy. (6) ; D. longatum Wolle (6) ; D. swart- zii Ag. (6); Gymnozyga moniliformis Ehr. (6); G. moniliformis var. gracilescens Nord. (6). CHRYSOPHYTA CHLOROTHECIACEAE: Ophyiocytium capitatum Wolle (11-a, 11-b) ; O. parvulum (Perty) A.Br. (6, 11-a, 11-b) ; O. sp. (1-c, 2-a). TRIBONEMATACEAE: Tribonema bombycinum (C.A. Ag.) Derbes & Solier (6, 7-a, 8-a) ; T. minus (Wolle) Hazen (2-a, 5-a, 7-b, 8-b). VAUCHERIACEAE: Vaucheria sp. (1-b, 3, 6, 8-c). 280 Rhodora [Vol. 63 OCHROMONADACEAE: Dinobryon sp. (6). RHIZOCHRYSIDACEAE: Lagynion sp. (? ampullaceum) (Stokes) Pas- cher (6). COSINODISCACEAE: Melosira sculpta Kuetz, (8-b, 8-c); M. varians C. A. Ag. (2-b); Stephanodiscus sp. (8-a). TABELLARIACEAE: Tabellaria fenestrata (Lyngb.) Kuetz. (l-a, 1-c, 2-b, 3, T-a, 8-a, 11-a); T. fenestrata (? var. asterionelloides) Grun. (1-c) ; T. floceulosa (Rothe) Kuetz. (1-a, 1-c, 3, 4, 5-a, 7-a, 8-a, 11-a). MERIDIONACEAE: Meridion circulare (Grev.) Ag. (1-c, 2-a, 3, 4, T-a); M. intermedium H. L. Smith (1-c, 3, 7-a). DIATOMACEAE: Diatoma anceps (Ehr.) Kirch. (4). FRAGILARIACEAE: Fragilaria sp. (? capucina) Desmaz. (1-a, 1-c, 2-a, 3, 4, 5-a, T-a, 8-a, 11-a) ; Synedra acuta Ehr. (1-c, 3, 4, 8-a) ; S. radians Kuetz. (2-a, 8-e, 11-a) ; S. ulna (Nitzsch) Erh. (3) ; S. ulna var. biceps (Kuetz.) Schoen. (8-a); S. ulna var. danica (Kuetz.) Grun. (4, 8-a, 8-b, 8-c). EUNOTIACEAE: Eunotia robusta var, diadema (Ehr.) Ralfs (11-a); E. robusta var. tetrodon Ralfs (3). NAVICULACEAE: Navicula diaphala (Ehr.) W. Smith (5-a); N. sp. (l-a, 8, 4, 5-a, T-a, 7-b, 8-b, 8-c, 10) ; Stauroneis inflata Kuetz. (8-c) ; S. sp. (3). GOMPHONEMATACEAE: Gomphonema acuminatum var. coronatum (Ehr.) Ralfs (11-a, 11-b); G. constrictum Ehr. (2-a, 11-a); G. dichoto- mum Kuetz, (7-b, 8-b); G. geminatum (Lyngb.) C.A. Ag. (T-a, 11-a); G. sp. (l-a, 1-b, 8-b). CYMBELLACEAE: Cymbella sp. (3, 4, 5-a); Epithemia sp. (5-b). SURIRELLACEAE: Surirella sp. (4). EUGLENOPHYTA EUGLENACEAE: Euglena polymorpha Dangeard (6, 10); E. sp. (3, 5-a, 5-b, 6, 10); Phacus sp. (3, 5-a) ; Trachelomonas hispida (Perty) Stein (6). PYRRHOPHYTA PERIDINIACEAE: Peridinium cinctum (Muell.) Ehr. (6). CYANOPHYTA CHROOCOCCACEAE: Microcystis aeruginosa Kuetz, (2-a, 2-b, 2-c, 9, 11-a) ; M. incerta Lem. (9) ; Merismopedia glauca (Ehr.) Naeg. (5-a, 5-b); M. tenuissima Lem. (6); Aphanothece castagnei (Breb.) Rab. (6); Coelosphaerium collinsii Drouet & Daily (11-a); Marssoniella elegans Lem. (6) ; Glaucocystis nostochinearum (Itz.) Rab. (6, 11-a) ; G. oocystiformis Prescott (6). OSCILLATORIACEAE:- Spirulina laxa G. M. Smith (6); S. sp. (5-a); Trichodesmium lacustre Kleb. (2) ; Oscillatoria agardhii Gomont (11- b); O. amphibia C.A, Ag. (11-a) ; O. articulata Gardner (1-a, 11-b); O. curviceps C.A. Ag. (6); O. lacustris (Kleb.) Geitler (3) ; O. sancta (Kuetz.) Gomont (6); O. subtilissima Kuetz. (6) ; O. tenuis C.A. Ag. (1-e, 8-a, 8-c, 10) ; O. sp. (? angusta) Koppe (8-b) ; Phormidium inun- 1961] Baldwin — Chamaecyparis thyoides 281 datum Kuetz. (1-b, 2-a, 4, 11-b) ; P. retzii (C.A. Ag.) Gomont (4, 8-b, 11-b); Lyngbya aerugineo-coerulea (Kuetz.) Gomont (3). NOSTOCACEAE: Anabaena affinis Lem. (2-a, 2-b, 2-c, 5-a); A. flos- aquae (Lyngb.) Breb. (11-a); A. oscillarioides Bory (8-c); A. spi- roides Kleb. (9). SCYTONEMATACEAE: Hapalosiphon hibernicus West & West (6) ; Sti- gonema mamillosum (Lyngb.) C.A. Ag. (11-a). RIVULARIACEAE: Gloeotrichia echinulata (J.E. Smith) P. Richter (6) ; Calothrix epiphytica West & West (5-b). RHODOPHYTA CHANTRANSIACEAE: Audouinella sp. (4). BATRACHOSPERMACEAE: Batrachospermum vagum (Roth) C.A. Ag. (4). — DEPT. OF BOTANY, CORNELL UNIVERSITY, ITHACA, N. Y. LITERATURE CITED 1. AHMADJIAN, V. 1958. The Fruticose and Foliose Lichens of Wor- cester County, Massachusetts. Rhodora 60: 74-86. 2. AUYANG, S. C. (Mrs.) 1960. Survey and Taxonomy of the Algae of Lake Quinsigamond. M.A. Thesis. Clark Univ. 3. Davis, R. L. 1959. An Ecological Study of Mosses in Worcester County. M.A. Thesis. Clark Univ. 4. PRESCOTT, G. W. 1951. Algae of the Western Great Lakes Area. Cranbrook Inst. of Sei. Bull. No. 31. Cranbrook Press. . SMITH, G. M. 1950. Freshwater Algae of the United States. Sec- ond ed. MeGraw-Hill Book Co. New York. 6. STONE, G. E. 1899. Flora of Lake Quinsigamond. Mass. Agr. College. Amherst. 7. WEBBER, E. E. 1961. The Ecology of Some Attached Algae in Worcester County, Massachusetts. M.S. Thesis. Cornell Univ. ct FURTHER NOTES ON CHAMAECYPARIS THYOIDES IN NEW HAMPSHIRE HENRY I. BALDWIN H. K. Svenson (Rhodora 31: 96-98, 1929) describes his three visits to the towns of Bradford, Windsor and Wash- ington, N. H. when he located Chamaecyparis thyoides (L.) ESP., in Bradford near the Washington line, presumably along the highway from East Washington to Bradford Cen- ter, and at Black Pond and Bagley's Pond in Windsor. He failed to find any stations for this species in the town of Washington, although he and Prof. Fernald made a consid- erable search for it. Recently, in connection with explora- 282 Rhodora [Vol. 63 tions made for the purpose of acquiring an area where this species might be preserved, the following observations were made. As mentioned by Svenson, this tree is continually being cut for posts and poles, and is fast disappearing. Flooding by beaver dams as well as man-made dams has also taken its toll. 1. Bradford swamp adjoining Bradford Mineral Springs is one of the most extensive habitats. It extends from the highway between East Washington and Bradford Center, south to the Bradford-Hillsboro town line and farther along the brook into Hillsboro a considerable distance, where cedar is found today as widely separated single specimens. Along the western side of the swamp near the Bradford-Washing- ton line there still remain clumps of larger trees up to 10" in diameter, with scattered smaller trees on the fringes of areas clear-cut in 1958. No trees have been found in Wash- ington; it seems probable that collections labelled “East of East Washington Village" were actually made in Bradford. The eastern side of the swamp also contains at least one un- touched stand on the fringe of an open black spruce bog. Some cedar is to be found north of the highway near the northern margin of the swamp. 2. Mud Pond (also known as Ayers or Nichols Pond) ele- vation 900’ (area 4.6 acres of which 3 are in Hillsboro) through which the Bradford-Hillsboro town line passes, has long been known to the writer as a station for cedar. When visited over 20 years ago the impression was gained that an extensive stand of fairly large trees surrounded the pond. Possibly these have since been cut. When visited in March 1961 the following observations were made: A small clump of cedar is situated on the southwest border of the pond. The trees are small — mostly 1" to 4" in diameter. There is also one small tree on the southeast shore, and one 5" tree on the east shore. All these are in the town of Hillsboro. There are also four trees along the northeast shore lying in Brad- ford. Carter Pond at the same elevation, and ringed by black spruce bog, apparently an identical habitat, was searched but no cedar found. 1961] Baldwin — Chamaecyparis thyoides 283 3. Bagley’s Pond, mentioned by Svenson as a station, lies partly in Windsor and partly in Hillsboro at an elevation of 1,146’. It is 40 acres in extent, of which only three acres are in Hillsboro near the outlet. This was visited on March 26, 1961 and as in the case of the other areas, explored on skis on a hard crust. It was easy to cover all the shore line and the swampy areas at the inlet and outlet. No trace of cedar was found. The water level appeared to have been raised about two feet by a beaver dam at the outlet at some time in recent years and pine, spruce, hemlock and some hardwoods killed, especially in the swamp around the inlet at the west end. A thorough search of this area failed to disclose any dead cedar or stumps. Since Svenson mentioned finding cedar in Bagley’s Pond in Windsor, it was presumably in this area where the trees occurred. 4. Black Pond in Windsor, elevation 1058’ (area 39 acres) has two main inlets on the south and west bordered by ex- tensive swamps. The water level was raised many years ago for a mill at the outlet, the present site of Windsor Mt. Camps. The dam has been repaired and maintained by the camp. The shores are rather steep in the main part of the pond and no cedar is encountered until one passes the princi- pal island. Part of swamps around the inlets are occupied by cedar of small size, mostly 1” to 4” in diameter, growing in thick clumps, usually associated with old stumps of the original cedar that was cut many years ago. The living trees are not over 10 feet in height. Mingled with these are numerous dead cedars of larger size (5"-6" diameter and 20' high) that may have been left in the original cutting and killed when the dam was built. All the presently living trees are in water, but the level must drop sufficiently in summer to permit them to survive. Some of the older stumps are one to two feet in diameter. Sections cut from these showed that growth had been extremely slow. One large stump had grown 0.3 inches during the last 10 years ; 0.45 inches between 10 and 20 years, and 0.6 inches from the 20th to 30th year, counting back from the bark. Allowing for more rapid growth in youth this would still make it over 284 Rhodora [ Vol. 63 100 years old. A 6" standing dead cedar averaged about 20 rings per inch. Nearby White Pond in Windsor and Stoddard was explored but no cedar found. 5. Swamp NW of Loverens Mill on Route 9, Antrim, N. H., elevation 1,036’. This extends for one mile north to the Windsor town line, and is separated from the Black Pond bog by a low height-of-land. Cedar is abundant in the poorly drained portions, disappearing wherever water move- ment becomes pronounced. It grows in mixture with black spruce Picea mariana, red maple and occasional white pine. There is no evidence of cutting in the southern portion, the only part visited. The largest cedars are 6" to 8" d.b.h. and younger, smaller trees are scarce due to the dense stand. The extent of the cedar area was not determined. 6. Robb Reservoir in Stoddard, elevation 1,275'. Cedar may be seen along the highway from Route 123, about 1 mile south of South Stoddard. This is a swamp at the south end of the Reservoir. Again the cedar occurs in mixture with black spruce and shrubs. (No cedar was found at the outlet or along the shores of Rye Pond in Nelsona short distance to the SE.) Cedar at the Stoddard site was commonly infected with Gymnosporangium biseptatum. Apparently Chamaecyparis has not been previously re- ported from Antrim, Hillsboro or Stoddard, N. H. Hodgdon & Steele! do not list these towns. Specimens from these areas have been deposited in the herbaria of the N. E. Botanical Club and the University of New Hampshire. This tree was found only in wet bogs or swamps with standing water the year round, and with imperfect drainage. Where water movement was evident no cedar was found. Common associated species were black spruce, Picea mari- ana (Mill) BSP., tamarack, Larix laricina (Du Roi) K. Koch, red maple, Acer rubrum L., mountain holly, Nemopan- thus mucronata (L.) Trel., highbush blueberry, Vaccinium corymbosum L. and cassandra, Chamaedaphne calyculata (L.) Moench. all growing in sphagnum. While cedar was found under these conditions, and in association with these " !HODGDON, ALBION R. AND FREDERIC L., STEELE, 1958. The Woody Plants of New Hampshire, Bull. 447, N. H. Agr. Exp. Sta. 1961] Voss — Leaves of Oryzopsis 285 species, the converse was not true. There are numerous other sites with apparently identical habitat conditions where cedar has not been found. It may very likely occur in other places not yet examined. Certainly it is not possible to state that Chamaecyparis does not occur in a township without a very detailed survey. This brief survey of Chamaecyparis at its known north- western limit in New Hampshire serves to indicate the pre- carious status of plants that cannot maintain themselves when the environment is altered. Reproduction is scanty, growth is slow and is inadequate to restore stocking when cutting and flooding destroy the seed bearing trees. Unless some areas are placed under protection it will be only a matter of time before all Chamaecyparis disappears from this region, as it apparently already has from Bagley's Pond. — FOX RESEARCH FOREST, HILLSBORO, NEW HAMPSHIRE, WHICH SIDE IS UP? A LOOK AT THE LEAVES OF ORYZOPSIS EDWARD G. Voss It was no less noted a critic than M. L. Fernald who as- serted: “errors once born never die but, on the contrary, by others not situated to know the facts are continually mis- taken for the truth and consequently perpetuated." (Rho- dora 44: 246. 1942.) After examination of a long series of Michigan specimens of a grass common in dryish woods throughout the state, Oryzopsis asperifolia Michx., I was not a little surprised to read in Fernald's 8th edition of Gray's Manual (1950) that the leaves of this species have a “glaucous lower surface" (italics are the original emphasis). Professor Fernald is in good, if not accurate, company. The first edition of Gray's Manual (1848) described the leaves as “pale underneath," and they have been similarly described in all subsequent editions. Torrey’s great Flora of New York (1843) — a source in which I often find ac- curate bits of description omitted by other authors — con- siders the leaves “glaucous underneath." The official verdict 286 Rhodora [Vol. 63 of the Manual of Grasses (1935 & 1951 editions) is “glau- cous beneath." Gleason's New Illustrated Flora (1952) says "pale... beneath.” Harrington (Man. Pl. Colo., 1954) uses the phrase *rather glaucous below." The embarrassing fact is that it is the upper (adaxial) surface of the leaf blades (a conspicuous 4-10 mm wide) which is glaucous. The truth is readily determined by fol- lowing a blade as it extends from the sheath. In his original description, Michaux said nothing about glaucousness, and I do not know who started the error; possibly it was Torrey. Many authors, whether intentionally or not, have kept their records clean by declining to report on this point: Pursh, Bigelow, Eaton, Wood, Britton & Brown, etc. The typically careful Deam (Grasses Ind., 1929) does not mention glaucousness, but does describe the blades as involute at the base, thus revealing his correct observation of the orientation of the leaf. (Descriptions of the margins as “revolute” show the same lack of understanding of which side is up.) Jennings (Wild Fl. West. Pa. & Upper Ohio Basin, 1953) merely describes the leaves as “glaucous,” not committing himself as to surface. In a cursory examination of literature, I have found no one who actually describes the upper surface as glaucous. Species possessing leaves with revolute margins and a glaucous under surface are not unusual. Involute margins and a glaucous upper surface would ordinarily, to be sure, be unexpected, although the leaves of Oryzopsis pungens (Spreng.) Hitche. are almost universally described as very narrow and involute and one might therefore expect that in O. asperifolia they are simply wide and involute. Assuming that some, if not all, authors have based their mention of a glaucous surface on an examination of plants and not on second-hand information, apparently the examination was not extended to checking just which side of the leaf was being observed. This is one case in which herbarium speci- mens are more easily interpreted than plants growing in the field. For in the latter, the leaves (essentially basal) spread out loosely over the ground, with the glaucous upper surface 1961] Eaton — Deane's Seedling Collection 287 usually appearing falsely to be the lower one, as the tough evergreen blades turn on their very narrow, involute, nearly terete, and evidently weak bases. However, one may easily trace the adaxial surface with the naked eye from the inside of the sheath, past the tiny ligule, through the groove in the slender base of the blade, to the broad glaucous surface with usually involute margins. The leaves of Oryzopsis racemosa (Sm.) Hitche., being cauline rather than basal, and even more conspicuous, have fared a little better in manuals and the score is nearly even. The Manual of Grasses says “pubescent beneath,” and Gleason uses the identical phrase. Torrey (under the syno- nym O. melanocarpa) again disappoints us, with “pubescent underneath." Deam, however, says “pubescent above," as does Fernald in the 8th edition of Gray's Manual, and as did the 7th edition (earlier ones omitting the point). Jones (Fl. Ill., 1950) says “the upper surface pubescent.” On all specimens which I have examined, the leaf blades are charac- teristically short-pilose above, although there is sometimes a little pilosity below in addition. If a moral is to be drawn from these simple observations, it is that those who write local floras have no basis for an accurate product other than painstaking examination of “na- ture, not books." — HERBARIUM, UNIVERSITY OF MICHIGAN, ANN ARBOR. A CHECK LIST OF WALTER DEANE'S SEEDLING COLLECTION RICHARD J. EATON The extensive and beautifully prepared herbarium of Wal- ter Deane came to the New England Botanical Club by be- quest in 1931. It included a bundle which he had designated as his “Seedling Collection" made in 1895 or thereabouts. It consists of meticulously prepared and well-mounted speci- mens representing fifty-two species in forty-three genera. For each species there are from one to five or more sheets with the specimens arranged in sequence according to age 288 Rhodora [ Vol. 63 from cotyledon stage onwards. In many cases a fruiting specimen from the putative parent plant is exhibited as a voucher; in others, a specimen in similar condition collected from the immediate vicinity. In every instance mature iden- tifiable material accompanies the seedlings. Mr. Deane’s label data and supplementary field notes are convincing: one is persuaded to accept the identifications without skepticism. It has been decided to insert this collection, appropriately segregated in separate species covers, in the organized her- barium of the Club. Before doing this, the appended check list was prepared, the nomenclature being revised to con- form to that of Gray’s Manual, 8th Edition. Without such a list the collection would be effectively lost among the eleven thousand or more species covers which enfold the two hun- dred thirty thousand mounted sheets in the Club herbarium. Actually, I think that a portion of the original collection, or perhaps a supplement to it, has been “lost” in this manner, because I have encountered from time to time an occasional sheet of seedlings labeled in Mr. Deane’s well-known hand- writing. From now on a record of such encounters should be kept so as to build up as complete a check list of the Deane seedlings as possible. 1. Abies balsamea 20. Silene Cucubalus 2. Picea mariana 21. Adlumia fungosa 3. Pinus Strobus 22. Cakile edentula 4. Alisma triviale 23. Sisymbrium officinale 5. Betula lutea var. leiocarpum 6. Ulmus americana 24. Tiarella cordifolia 7. Polygonum aviculare 25. Hamamelis virginiana 8. P. Persicaria 26. Geum urbanum 9. P. orientale 27. Prunus serotina 10. P. Convolvulus 28. Trifolium repens 11. P. scandens 29. Lathyrus japonicus var. 12. Chenopodium album glaber 13. Atriplex patula var hastata 30. Oxalis montana 14. Salicornia europaea 31. O. corniculata 15. Salsola kali 32. Acalypha virginica 16. Amaranthus retroflexus 33. Euphorbia polygonifolia 17. Portulaca oleracea 34. Acer pensylvanicum 18. Spergularia marina 35. A. rubrum 19. Stellaria media 36. A. spicatum 1961] Green and Thomas — Ranunculus ficaria 289 37. Impatiens capensis 46. Galium triflorum 38. Malva rotundifolia 47. Solidago sempervirens 39. Circaea alpina 48. Ambrosia artemesiifolia 40. Fraxinus americana var. elatior 41. F. nigra 49. Bidens connata var. 42. Lamium amplexicaule gracilipes . B. frondosa . Arctium tomentosum . Taraxacum officinale 43. Lycopus americanus 44. Campsis radicans 45. Plantago rugelii or c wre © C1 THE BULBIFEROUS RANUNCULUS FICARIA. — Recent exam- ination of this European species in two localities in the Boston area has shown that in both cases it is represented only by the bulbiferous variety, Ranunculus ficaria L. var. bulbifera Marsden-Jones. Cytotaxonomic studies carried out in Britain have shown that Ranunculus ficaria exists in at least two cytodemes, the diploid R. ficaria var. ficaria (R. ficaria var. fertilis Clapham) (2n = 16) and the tetraploid R. ficaria var. bulbifera (2n = 32). They are separable on a number of minor morphological and ecological features (see Marsden-Jones in Jour. Linn. Soc. Lond. Bot. 50: 39. 1935 or D. E. Allen in Proc. Bot. Soc. Brit. Is. 3: 45. 1958, or even van Tieghem in Ann. Sci. Nat. sér. 5, 5:88. 1866 who was naturally unaware of the cytological significance of his observations.) but most significant and noticeable of all is the fact that the tetraploid bears bulbils in the leaf axils and has a very reduced seed fertility whereas the diploid is quite fertile and does not produce bulbils. These bulbils do not become apparent until after the plant has been in flower for a few days when they rapidly enlarge to about the size of a grain of wheat. The two populations examined this spring, one in the garden of 383 South Street, Jamaica Plain and the other in the Case Estates at Weston, both show the production of abundant bulbils. Cytological examination of both popula- tions was made using acetocarmine squashes of the develop- ing bulbils. This proved to be very favorable material, par- ticularly during early stages of development, at which time a mass of cells near the apex of the bulbil is dividing quite 290 Rhodora [Vol. 63 rapidly. The chromosomes are rather long, however, and better preparations were obtained when the material was pre-treated for three hours in colchicine to shorten the chromosomes, fixed in Carnoy's solution, and softened for 10-15 minutes in 10% HCl. In both populations examined the bulbils were found to be composed primarily of tetraploid cells (2n = 32). However, occasional triploid cells were observed in several bulbils from both populations. A few cells seemed to have a chromosome number intermediate between triploid and tetraploid, but in all cells that could be counted with certainty the somatic number was either 24 or 32. Possibly the cytological situa- tion here is comparable to that observed in tissue cultures in which there is considerable variation in chromosome num- ber. (See Torrey, J. G. in 7th Symposium of Society for the Study of Development and Growth, 189-222. 1959.) Examination of the pollen revealed that less than 20% of the grains were either abortive or failed to stain normally. The remaining grains stained densely with acetocarmine and appeared normal, except for the fact that the size variation was rather high. It seems unlikely that the plants could be triploid and produce pollen that is over 80% fertile. Tri- ploids have been reported in Ranunculus ficaria from areas in which the diploid and tetraploid varieties overlap, but these triploids are characterized by a high percentage of abortive pollen (Neves in Bol. Soc. Brot. ser. 2, 16: 169. 1942, see also 46th (1955), Ann. Rept. John Innes Hort. Inst. 20- 21. 1956). The plants in the above populations produced no fertile achenes this spring, but in view of the reasonably high pollen fertility observed it is possible that these plants are outbreeders and appear sterile due to the lack of pollen from a different individual. The populations examined were small and probably consist of a single vegetative clone. It is hoped that cross pollinations and a study of meiotic material next spring will clarify this point. The specimens of Ranunculus ficaria in the Gray Herbari- um and the New England Botanical Club Herbarium were next examined and it is significant that whereas several spec- 1961] Love — Botanical Dictionary 291 imens show distinct axillary bulbils, none of the others may be identified with certainty as var. ficaria. Those where bul- bils were not apparent were all collected early in the flower- ing season before the end of April when the bulbils would not be seen even on var. bulbifera. Every one of the specimens gathered after the beginning of May shows the presence of bulbils. The examination of other herbaria would no doubt reveal other records and a tour of New England reveal other popu- lations, but the main object of this note is to draw the atten- tion of botanists in N. E. America to the existence of this polyploid and bulbiferous variety. It would also be most interesting to know whether it is only this variety that has been introduced from Europe. The following are the herbarium records we have seen of var. bul- bifera, CANADA. QUEBEC: comté de Jacques-Cartier, ville-Lasalle, dans les bruissons le long d'une clóture, 23 May, 1932. Marie-Victorin & Rolland-Germain 46, 833 (GH). UNITED STATES. MASSACHUSETTS: Mid- dlesex Co., Cambridge, spreading in William Brewster’s garden, 8 May 1914, Walter Deane (NEBC); Norfolk Cc., Milton, wild weed of my garden, 23 June 1923, N. T. Kidder (NEBC); Plymouth Co., South Hingham, May 1891, H. W. Cushing (GH) and moist soil in garden, transplanted from original locale, 10 May 1947, C. H. Knowlton (NEBC); Worcester Co., Lancaster, wild garden of Mrs. N. Thayer, never seen there before, June 1924, Mrs. J. E. Thayer (NEBC). PENN- SYLVANIA: Philadelphia, *Nurseries", Fairmont Park, 7 May 1910, H. St. John 111 (GH). (Herbarium specimens of plants that were ex- amined cytologically in this investigation have been deposited in the Gray Herbarium). — PETER S. GREEN AND JOAB L. THOMAS, ARNOLD ARBORETUM, HARVARD UNIVERSITY. A USEFUL MULTILINGUAL BOTANICAL DICTIONARY.' — In recent years more and more scientists in the New World have taken up the study of Russian in order to benefit more directly from the flood of Russian scientific books and papers now available to them through various channels. For most of them this pays, instead of waiting an inordinately long IN.N. Davidov and F. Kh. Bakhteyeva, 1960: Botanical Dictionary, Russian-English- German-French-Latin. — Glavnaya Redaktsiya Inostrannykh Nauchno-Teknicheskikh Slovarey Fizmatgiza, Leninsky Prospekt 15, Moskow V-71. 1 ruble 16 kop. 292 Rhodora [Vol. 63 time for an official translation, the appearance of a reference journal where the desired information may be found, or such. But with only a slight knowledge of Russian and per- haps a not too good dictionary at hand, it can at times be quite hard to get the exact meaning of terms. However, at least for botanists, this dilemma is now solved with the ap- pearance of a multilingual botanical dictionary, compiled by N. N. Davidov and edited by F. Kh. Bakhteyev. The book is actually a list of botanical terms and plant names directly translated from Russian into English, Ger- man and French as well as Latin, and permits quick transla- tion to or from any of these five languages. Thus it is actually of a much wider use than for those reading Russian texts only, and is a help also for English-speaking botanists read- ing French, German or even Latin works. According to the foreword, it is the first time that such a botanical dictionary has been published in Russia, and the authors are to be con- gratulated on their achievement. The dictionary contains about 6000 botanical terms of which some 30% refer to plant names only. The choice of vernacular names in English, French and German has cer- tainly been a difficult task, and some rarely used names are met with here and there. But as the Latin names — in some cases the specific name, in others only the generic name, but always the family name — are included also, no real difficul- ty of identification of the common name is met with. Names of wild plants are of course referable mostly to Russian spe- cies, but common names for a goodly number of cultivated plants from the world over are also included. Regarding the about 4000 proper terms themselves, it is pleasant to note that they cover the botanical field very wide- ly and seem to reach out into neighboring fields such as cytol- ogy and genetics as well. It is quite evident, however, that the authors have done their utmost to make the book compact, and to give for each Russian word, if possible, only a single-word expression in the other languages. This has resulted in the choice of a number of the English terms seemingly rather unfamiliar to 1961] Love — Botanical Dictionary 293 the young botanists who do not have the knowledge of Latin and Greek, which was so useful to their older colleagues. Such words as grumous, hiant, irrorate, vittate and poecilo- phyllous, to mention a few random examples, might have been better translated with a more modern expression such as, respectively: lumpy (about roots), gaping, sprinkled with dew, with bands or stripes, with various-colored leaves. Fortunately most of these unfamiliar words are explained by a glance at the corresponding German or French expres- sions, or are relatively easy to find in the Oxford, Webster. or some other widely used English dictionaries. So far, I have been unable to find the English counterpart of the word "trullifolius" only, but as the Botanical Dictionary itself gives the word "trulliformis" as corresponding to “saucer- shaped", it requires only a small portion of imagination to understand that *''trullifolius" means “with saucershaped leaves". This phrase would, of course, have been better to use than the old-fashioned Latin derivative. There are ad- mittedly a number of such outmoded English expressions in the book, but not nearly enough to cause concern, and as demonstrated, usually possible to get at by some round-about way. In rapidly reading through all the English terms listed in the English index, I came across a few unfortunate mis- spellings (e.g. skiophyte for sciophyte, dicliny instead of dic- linism, induvia for induvium) which have eluded the authors themselves, who found about 25 words from all the lan- guages requiring correction on an inset leaf at the end of the book. In comparison to the total number of words in the book, these misprints are indeed very few and in no case really serious or misleading. The few negative remarks above should not be allowed to obscure the more excellent properties of the book. A very good feature is the accentuation of the Russian words and the indication of the gender of all nouns in Russian as well as in German and French, even when these words are pre- ceded by adjectives. This is actually more information than is given in some current and in other respects excellent Rus- 294 Rhodora [Vol. 63 sian dictionaries, and a feature that will be very helpful to students using this book. The unusual arrangement of numbering each Russian word in the actual dictionary part of the book, and using these numbers in the clear and easily read English, German, French and Latin indices for a quick reference to the location of the particular word in the dictionary is highly laudable. It is particularly this feature that makes the use of the book so universal and not only restricted to those who want to read a Russian text or translate into Russian. Now anybody, say English-speaking, who wants to read e.g. a French, Ger- man or even Latin text, can use this handbook to full advan- tage, even if he is completely ignorant of any Russian, its letters or order of alphabet. The following line may serve as an example of the system: 453 BÉTKA f || bough, branch, limb || Ast m, Zweig m || branche f || ramus. The Botanical Dictionary is clothbound, of a handy, com- paet size, well printed and very easy to read. It is thus a pleasure to recommend this excellent book to all colleagues in the wide field of Botany and not only to those directly in- terested in Russian botanical literature. It could be said about it that it is truly an aid to international understanding. — Doris LOVE, INSTITUT BOTANIQUE, UNIVERSITE DE MONTREAL, CANADA. A NOTABLE ASSEMBLAGE OF PLANTS IN NEW HAMPSHIRE. — This note concerns an area of hardwood forest several of whose plentiful species of ground-flora suggest conditions unusually rich for east-central New Hampshire. The area — rather stony and originally alluvial, though most of it may be inundated only once in a decade or two — lies in the extreme northeast corner of Sandwich in Carroll County, N. H. It is reached from Route 113A at Wonalancet in Tamworth by crossing Wonalancet River at the bridge by the post office and then proceeding 0.7 miles up a gravelled public road to Squirrel Bridge (over the same river) 30 feet down a private road. 1961] Lincoln — Plants in New Hampshire 295 The plot of chief interest starts on the north edge of the private road — a narrow, wooded, dirt one — by a telephone pole about 70 feet south and west of the bridge and extends westerly along the road 100 feet. In width, it extends north- erly toward the river a mere ten feet ; beyond that width, the dense herbaceous cover thins out greatly and the more un- usual species disappear. The most unexpected plant is Asarum canadense L. This and Viola pensylvanica Michx. var. leiocarpa (Fern. & Wieg.) Fern. are perhaps the two most abundant herbs. They occur little if any beyond the ten foot line back from the road, and elsewhere in the general area and within 100 feet of the river they seem to be found only along the edge of the same road and of smaller wood roads and paths. Two other rich-woods plants in the plot are Botrychium virginianum (L.) Sw. and Dicentra Cucullaria (L.) Bernh. Other more commonplace herbs readily identifiable in June include, more or less in decreasing order of abundance there, Uvularia sessilifolia L., Viola rotundifolia Michx., Dryopter- is Phegopteris (L.) Christens., Arisaema atrorubens (Ait.) Blume, Smilacina racemosa (L.) Desf., Polygonatum pubes- cens (Willd.) Pursh, Trillium erectum L., Streptopus roseus Michx. var. perspectus Fassett, and the two species of Actea PT The two commonest shrubs on the plot are Acer spicatum Lam. and Prunus virginiana L. The two commonest trees are Acer saccharum Marsh. and Fraxinus americana L. Jug- lans cinerea L. and Tilia americana L. overhang parts of the plot. Additional plants of interest occurring nearby along the river or within 100 yards downstream and somewhat unusu- al for the region are Sanguinaria canadensis L., Viola Selkir- kii Pursh, and V. renifolia Gray. Frederic L. Steele assisted me with identifications in the field. The plot on which the above plants are growing, as well as adjoining land, is threatened with development for house lots. — ALEXANDER LINCOLN, JR., MEREDITH, NEW HAMP- SHIRE. 296 Rhodora [ Vol. 63 THE SPECIFIC ÉPITHET OF THE PECAN. — According to Article 73 of the International Code of Botanical Nomencla- ture (Regnum Vegetabile 8: 46. 1956), “The original spell- ing of a name or epithet must be retained, except that typo- graphic or orthographic errors should be corrected." Fernald (Rhodora 49: 194-196. 1947) showed that Juglans illinoin- ensis of Wangenheim (Beytr. z. teutsch. holzg. Forstwiss. Nordam. Holz. 54. 1787) is the earliest valid name for the pecan. Koch (Dendrologie 1: 593. 1869) transferred Juglans illinoinensis to Carya, but altered the spelling of the specific epithet to illinoénsis. Koch's version of the epithet (with or without the diaeresis) is the one now universally used — in violation of Article 73 — in North American botany. Rehder (Journ. Arn. Arb. 22: 572. 1941) expressed the opinion that the spelling illinoinensis used by Wangenheim is possibly “a typographical error or . . . a slip of the pen." Such would . not seem to be the case because illinoinensis occurs at least twice in Wangenheim's work: in the description of the *'Illi- nois Wallnut Tree" on page 54, and in the running head on page 55. Illinoinensis was surely intentionally spelled so. Admittedly, Wangenheim's spelling is perhaps not the best for a Latin adjective derived from a French word. *'Illinoi- sensis” or even “illinoensis’” may be preferred, although there would appear to be no hard and fast rule for the Latin- ization of a word such as Illinois. All this is beside the point, however. The fact remains that Wangenheim used illinoin- ensis and that this spelling must, according to the Code, be retained. The scientific name of the pecan is Carya illinoinen- sis, not C. illinoensis. — JOHN W. THIERET, CHICAGO NATUR- AL HISTORY MUSEUM. Volume 63, No. 753, including pages 237-266, was issued November 1, 1961 BLOW REFERENCE LIBRAR. 73 R47 DEC 2.0.1961 Hodova JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Associate Editors Volume 63 November, 1961 No. 755 CONTENTS: Introgression of Alnus serrulata and Alnus rugosa, Erederie L- Steele e NN ssl ITI ie 297 Some New or Otherwise Interesting Reports of Liliaceae from the Southeastern States. Bawara T Browne, JT eU ONSE 304 Hybrid Aspens in the Lower Peninsula of Michigan. TER e OUR a Sire A Fe cask ee cakes tec e hate sk OR ecaee user caress 311 A Disjunct Community of Chestnut Oak in Mississippi. EUIS: CEE CES: Lucae perisse cbr PEN AE Important Notice to Members, Subscribers and Librarians .... 326 The New England Botanical Club, Jne. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume, Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able. All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass. Printed by THE LEXINGTON PRESS, INC. Lexington, Mass. Rhodora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Volume 63 November, 1961 No. 755 Dui Ot Coo OF ALNUS SERRULATA AND ALNUS RUGOSA FREDERIC L. STEELE In the eighth edition of Gray's Manual of Botany, two species of alders in the subgenus Alnus are indicated as occurring in New Hampshire, each of which has a variety and several forms. The two species are distinguished as follows: Alnus rugosa (Du Roi) Spreng. has leaves with a rounded base, ovate or elliptical in shape with a tendency to be broadest below the middle, doubly-toothed leaf margin, strong veins on the green lower surface, prominent lenticels on the stems, and nutlets and bracts with a characteristic shape. Its variety americana (Regel) Fern. differs in having leaves glaucous rather than green underneath. Alnus serru- lata (Ait.) Willd. has leaves with angled or cuneate base, obovate and broadest above the middle, a margin almost singly toothed, weak or vanishing veins on the green or glutinous lower surface, inconspicuous lenticels, and nutlets and bracts differing from A. rugosa in a way to be stated later. Its variety subelliptica Fern. has broader leaves, although still broadest above the middle, and a rounded leaf base. | : The present study was started when the writer became interested in the northern limits of the ranges of these two species and their varieties in New England. A. rugosa var. americana is a well-marked variety occurring througnout northern New England and extending north to Labrador. Fernald (1945) states that this variety is decidedly more northern in range than var. rugosa. An investigation of the range of the latter from herbarium sheets presented some 297 298 Rhodora [Vol. 63 difficulties. The question of whether leaves are glaucous or green can only be determined from mature leaves and in natural light. Under these conditions, if the observer has had some practice, the distinction is readily apparent. Many of the herbarium specimens were collected in April when there were no leaves, so that the distinction can not be made. In other cases, the specimens were mounted in such a way that it was possible to examine the lower surfaces of the leaves. Subsequent field observations indicate that the new leaves of var. americana, as they appear in June are green ; by early July they become glaucous. Alder typically grows in swamps or open low land ; when it grows in shady environ- ments the leaves are thin and pale green with the glaucous character barely discernible. The material on some sheets that were labeled var. rugosa was of uncertain identity. Altered notations on many sheets indicated confusion of A. serrulata with A. rugosa var. rugosa. In Tamworth, New Hampshire, A. rugosa var. americana is common in open moist habitats. There plants are quick to invade a poorly-drained hayfield and are also common along streams. A. serrulata is near its northern limit in Tamworth and is confined to shores of lakes and occasionally along edges of rocky streams. Some plants occur that can be referred to A. rugosa var. rugosa. These, however, although having green leaves which set them off from var. americana are not readily distinguishable from A. serrulata. As the scope of this investigation was broadened to include the whole state of New Hampshire, this situation was observed to be characteristic. It seemed probably that hybridization and introgression were going on between A. rugosa var. americana and A. serrulata producing a mixed assemblage of plants, some of which could be referred to var. rugosa. To investigate this situation, it was decided to study popu- lations and analyze them by pictorialized scatter diagrams of the type devised by Anderson (1949, 1952, 1953). Charac- ters to be used were based on leaves, lenticels, and fruit. As there is considerable variation in leaves on sterile plants, especially on young shoots, only leaves on fruiting branches were used. Preferably these were collected after the middle 1961] Steele — Introgression in Alnus 299 of July when the fruit was mature. An alder colony typically consists of a number of separate clumps of plants, which, judging from the variation that occurs, are clonally distinct, although it is possible that in some cases a new clump could arise from a rootstock. In a colony extending for a distance of two or three hundred feet, many plants may not fruit, so the sampling consisted of collecting branches from available fruiting plants that appeared to be separate. Leaves on fruiting branches are typically quite small. Where possible a branch was collected with at least one undamaged leaf six to eight centimeters in length. Lenticel type is significant, and, as collection of bark is difficult, the character of the lenticels was noted at the time. Notches were then cut on the branches in accordance with a standardized code; three notches indicating prominent lenticels, two indicating that some were indefinite, and one indicating the lenticels were indistinct. Seven characters were scored including most of those used by Fernald in the eighth edition of Gray’s Manual. The angle at the base of the leaf was measured on a 15° degree sector scale. Single as opposed to double serration was determined by counting the major teeth on one leaf margin using certain “typical” leaves from well marked samples of the two species as standards for comparison. The character of the lower surface of the leaves, ranging from glutinous to glaucous, was determined by examining the leaves in daylight and again comparing with standards. Usually the distinction between green and glaucous was obvious, but where the glaucous character could barely be discerned, the leaves were marked pale. Sometimes it ap- peared as if the leaves were glaucous because of disease. To measure whether the leaf was broadest above or below the middle, it was placed on graph paper and the broadest point determined. The distance from this point to the mid- point of the leaf was divided by half the length of the leaf to form the “midratio”. This was taken as positive if the broad- est point was above the middle and negative if it was below. Lenticels, especially on larger stems, are usually promi- nent in the case of A. rugosa var. americana and indistinct 300 Rhodora [Vol. 63 in the case of A. serrulata. The lenticel character was re- corded at the time of collection as indicated above. Veins on the lower surface were ranked as prominent, intermediate or vanishing. This character was somewhat variable, but well-marked trends were apparent. The leaves were compared with characteristic leaves of A. serrulata and of A. rugosa var. americana and rated according to which one of the two they most nearly approached. Well developed nutlets and bracts of the two taxons seemed to be markedly different. The bracts of A. serrulata have a lower portion with straight narrow sides with the upper portion rather abruptly flaring; the bracts of A. rugosa var. americana have the lower portion with diverging sides, with the upper portion less distinct. Nutlets of var. americana tend to be rhombic with the top portion triangular. Nutlets of A. serrulata are obovate with prominent shoulders at the top. These differences become apparent when the fruit is mature; about the middle of July in New Hampshire. On the diagrams the fruit is labeled broad if the nutlets and bracts resemble var. americana; narrow if they resemble A. serrulata. Some fifty populations were sampled from different areas of New Hampshire and northern Massachusetts, with an effort made to sample different environments. Then popula- tions believed to be typical of the situation are shown in Figures 1 - 10. Following Hardin (1957), intermediates are labeled “‘species A" X “species B", while apparent back crosses are labeled as "species A" (X "species B") indicat- ing that "species B" contributed the minority of the genes. Figure 1 from a swamp shows typical A. serrulata. Figure 2 from a lake shore shows A. serrulata near the northern limit of its range where much A. rugosa var. americana is present. It appears to be good A. serrulata but may not be genetically pure. Figure 3 from a swamp shows A. serrulata Fic, 1-10. Pictorialized scatter diagrams showing introgression between A. serrulata and A. rugosa var. americana. All collections by the author. 1. A. serrulata (Boxford, Mass.). 2. A. serrulata (Tamworth, N. H.). 3. A. serrulata (X var. americana) (Hampton, N. H.). 4. A. serrulata (X var. americana) (Freedom, N. H.). 5. A. serrulata X var. americana (Hampton, N. H.). 6. A. serrulata X var. americana (Freedom, N. H.). 7. A. rugosa var. americana (X A. serrulata) (Tamworth, N. H.). 8. A. rugosa var. americana (X A. serrulata) (Milton, N. H.). 9. A. rugosa var. americana (Sandwich, N. H.). 10. A. rugosa var. americana (Ossipee, N. H.). 1961] Steele — Introgression in Alnus 301 E EM [o g i ae | v $ $6 $ E e? e & L $9. Fe, | L L l eae ,94 A T ii l l l ee i 4 - d 2 " E » | d T L i it L 1 1 it Bo, É | we b L L EL L | e | [5 [ 6 F wy E W 4 X i "d - LY " e E - by L l L L l L L L f ARES | 1 L 1 J L 1 l l 1 1 ] [.3 8 PY Ly if zl E oy wy Y E Y Kf E - x g g " 1 ji l l ji | | | i i} 1 L L J EEES l pi L L l1 L] [ 9 y NG TE E / 5p E BUT NE E YYY = u 3r Wy vy i > "A wy V wk 9 ES ie y yy | l l l l L L L L I! l l l L L fi l L L l J 60 g0 100 120 140 160 EXPLANATION OF SYMBOLS ABSCISSA ANGLE AT BASE ORDINATE NUMBER OF TEETH MIDRATIO LEAF COLOR LENTICELS VEINS +.2 0 GLAUCOUS O PROMINENT C/ STRONG 4e PALE S WEAK O WEAK Ò Oo O GREEN 9 VANISHING O VANISHING O -.l e GLUTINOUS e FRUIT -.2 6 BROAD ` INTERMEDIATE D NARROW fe) 302 Rhodora [Vol. 63 modified by genes from var. americana and contains three plants in the upper right of the diagram that could be called var. rugosa. Figure 4 from a sandy lake shore also shows modification of A. serrulata by genes from var. americana. In this case some of the plants might be referred to A. serru- lata var. subelliptica. Figure 5 from a swamp and Figure 6 from a lake shore show populations about intermediate be- tween A. serrulata and var. americana. Figure 7 and Figure 8 both from swamps show var. americana modified some- what by A. serrulata and include plants that could be referred to var. rugosa. Figure 9 from a poorly drained field shows good var. americana unmodified by any A. serru- lata. Figure 10 from the edge of an acid peat bog shows var. americana apparently modified by the acid environment. The leaves are smaller and the major teeth less prominent. TABLE 1 TAMWORTH FREEDOM Plant No. Leaf angle Width/lengthPlant No. Leaf angle Width/length 1 75 .45 1 90 .65 2 75 .5 2 105 6 3 90 5 3 105 7 4 90 6 4 120 6 5 90 7 5 120 .75 6 105 7 6 120 8 T 120 6 7 135 7 8 120 M 8 150 8 These populations are apparently typical of the alder situ- ation, judging from field studies in northern New England and herbarium studies for the whole area in eastern North America. A. rugosa var. americana is a distinct taxon with constant characteristics, especially in the northern part of its range where no A. serrulata is present. A. serrulata is a distinct species, with a range more southern than that of var. americana, extending north to north-central New Hamp- shire and central Maine. When the two taxa occupy the same area, introgression occurs producing intermediates, and plants that approach one or the other of the supposed parent. These intermediates do not seem to demand any particular hybridized environment, although they normally 1961] Steele — Introgression in Alnus 303 occur in places that may be regarded as somewhat intermedi- ate, and that have almost invariably been disturbed by man. Plants close to var. americana but green underneath could be referred to var. rugosa; however, these usually are vari- able and resemble A. serrulata in some respects. The key characters that distinguish var. subelliptica from var. serrulata are leaves gradually rounded at base as op- posed to cuneate or subacute, and broad leaves with a width- length ratio of .6 to .9 as opposed to narrower leaves with a ratio of .35 to .65. Width to length ratios were measured for a number of populations. Table 1 shows some typical results. Var. subelliptica appears to be one extreme of the normal variation of A. serrulata populations, although in some cases these populations may be modified by gene flow from var. americana. There is no indication that var. subelliptica is a geographic variety. There are various named forms of both species and varieties, some of them depending on pubescence of lower surfaces of leaves. Many populations of both species had leaves varying from glabrous to varying degrees of pubes- cence. A. serrulata forma nanella is described as a dwarf shrub with small leaves and aments. Alders growing in adverse environments, such as peat bogs or silty shores of rivers are frequently stunted. A. serrulata forma emargi- nata has emarginate leaves. It is not infrequent to find an alder with some of the leaves emarginate. In conclusion it appears that in the subgenus Alnus, in the northeast, the two good native taxa are A. serrulata and A. rugosa var. americana. The latter can readily be distin- guished by the glaucous character of the lower surface of the mature leaves. Any plant with this trait, if not diseased, will possess the other characteristics of the taxon. Plants with deep green or glutinous leaves and cuneate leaf bases will possess the specific characters of A. serrulata. These two taxa hybridize and backcross frequently, the resulting gene exehange producing a variable assemblage of plants including some intermediates that would be referred to A. rugosa var. rugosa. 304 Rhodora [Vol. 63 ACKNOWLEDGEMENTS. During the course of this study the author has been greatly helped by the encouragement and criticism of Dr. A. R. Hodgdon, Durham, N. H. Extensive use was made of the Herbarium of the New England Botanical Club. Dr. R. C. Rollins kindly allowed me to use the Gray Herbarium at Harvard University. — ST. MARY'S -IN-THE-MOUNTAINS, LITTLETON, N. H. LITERATURE CITED ANDERSON, E. 1949. Introgressive Hybridization. John Wiley & Sons. . 1952. Plants, Life and Man. Little, Brown & Co. . 1953. Introgressive Hybridization. Biol. Review 28: 280-307. FERNALD, M. L. 1945. American Representatives of Alnus incana. Rhodora 47: 333-361. . 1950. Gray’s Manual of Botany 8th Edition. Ameri- can Book Co. HARDIN, J. W. 1957. Studies in the Hippocastanaceae. IV. Hybridi- zation in Aesculus. Rhodora 59: 188. SOME NEW OR OTHERWISE INTERESTING REPORTS OF LILIACEAE FROM THE SOUTHEASTERN STATES' EDWARD T. BROWNE, JR. Several interesting plants in the Liliaceae have been found as a consequence of field work in the southeast. As might be expected, this family is one of the best represented ones in our herbaria, but it is still incompletely known. One of these collections consisted of several specimens of Aletris collected by Dr. W. H. Duncan of the University of Georgia. Dr. Duncan found both white- and yellow-flowered Aletris plants growing in approximately equal numbers in the same population in Chatham Co., Georgia, and they ap- peared at first to be plants of A. farinosa L. and A. lutea Small, respectively. Since perianth color and shape are the two most important characteristics for specific separation in the North American species of Aletris, the identifications appeared to be obvious. However, upon closer examination, it was discovered that in neither of these types was there semi-epigyny, one of the This paper was supported in part by grants from the Faculty Research Fund, Graduate Sehool, University of Kentucky. 1961] Browne — Liliaceae of Southeastern States 305 most outstanding floral characteristics of A. farinosa. Micro- scopic and macroscopic examination of leaves of these plants revealed that both exhibited the narrow hyaline margin which is typical of the leaves of A. lutea. From this informa- tion it has been concluded that this population consists of typical A. lutea and white-flowered individuals of A. lutea. Thus, the latter are hereby designated as a new color form as follows: Aletris lutea Small, forma albiflora E. T. Browne, Jr., f. nov. Forma nova floribus albis. This new form differs from the typical form only in flower color. The perianth tube is cylindric with recurved tepals. This is the first reported color form of North American Aletris species. Because of the importance attached to flower color in the identification to species of plants in this genus, careful review of the specimens presently deposited in her- baria will be necessary for proper determination. It is significant that A. farinosa has apparently not been collected in the Lower Coastal Plain of Georgia, and although these specimens are obviously those of a population of A. lutea, this latter species has not turned up in the collections made in connection with the Flora of the Carolinas Project which includes a geographic area in part not very far removed from this Chatham County locality (H. E. Ahles, personal com- munication). Specimens cited: GEORGIA. Chatham Co. Open area, low ground at edge of cypress stand, 15.3 mi. W4°S of Savannah City Hall. Coastal Plain Province. W. H. Duncan, 20982. 13 June 1958. The holotype has been retained in the Herbarium of the University of Georgia. Dr. Duncan has presented an isotype and a specimen of the typical form (Duncan, 20980) to the Herbarium of the University of Kentucky. Another Aletris variant was observed in mixed popula- tions of A. lutea and A. obovata Nash. These individuals had cream-colored flowers instead of yellow and white flowers as in A. lutea and A. obovata, respectively. The perianth tips of A. lutea are recurved while those of A. obovata are not. This variant is characterized by a perianth of intermediate color "Appreciation is expressed to Dr. W. L. Carr, Department of Ancient Languages, University of Kentucky, for the Latin diagnoses in this paper. 306 Rhodora [Vol. 63 and shape but without recurved perianth tips. Since these plants were observed only in mixed populations of the two species previously referred to, or in the near vicinity of populations of both species, the variant individuals are strongly suggestive of probable interspecific hybridization between these two species. Therefore, this putative hybrid is designated in the following way : Aletris X Tottenii E. T. Browne, Jr., hybr. nov. Aletris X Tottenit, hybrida naturalis nova inter A. luteam et A. obovatam. Herbae mediae inter species parentales. Folia basalibus rosettis, hyalino margine, colore sulfureo-viridi, lanceolata, 52-96.5 mm. longa et 6.2-11.5 mm. lata. Calami et bracteati racemi 5.1-7.7 dm. alti cum multis floribus. Flores horizontales aut prope horizontales, colore aliquantum vario inter ochroleucum et ebureum (cream-colored). Ultimae partes peri- anthorum non recurvatae. Forma et magnitudo tubae perianthi ali- quantum variae inter species parentales. Plants intermediate between the parental species. Perianth segment tips not recurved (perhaps appearing so in pressed material). Specimens examined: GEORGIA — Berrien Co. Highway right of way, Ga. 135, 1.8 mi. S of Berrien-Atkinson Co. line. Flowers cream color. May 7, 1960. Elizabeth M. and E. T. Browne, Jr. 2516; Colquitt Co. Highway right of way, US 319, 0.1 mi. N of Colquitt-Thomas Co. line. Flowers varying in color from light to deep cream color in the three specimens collected at this locality. May 8, 1960. Elizabeth M. and E. T. Browne, Jr. 2524. (holotype, KY); Cook Co. Highway right of way, Ga. 76, 5.8 mi. N of Cook-Brooks Co. line. Flowers light cream color. May 7, 1960. Elizabeth M. and E. T. Browne, Jr. 2519. This putative hybrid was first reported in the literature by Harper (1905, 1906). He did not name the hybrid, but he did suggest its hybrid nature. Since in form of the perianth there is a somewhat closer resemblance to A. obovata than to A. lutea, some botanists might suggest that this variant constitutes only a color form of A. obovata. However, this suggestion is not tenable since these plants are found only in mixed populations of A. lutea and A. obovata or in the near vicinity of pure populations of both species. In addition, while these species usually occur in separate populations, especially along highways they have become mixed presumably as the result of highway maintenance operations, and this fact has undoubtedly contributed to their hybridization. Apparently, the parental species are not highly interfertile since there are very few of the putative 1961] Browne — Liliaceae of Southeastern States 307 hybrids to begin with, and there is no evidence to indicate that offspring are produced beyond the F, generation. The putative hybrid, Aletris X Tottenii, has been named for Dr. Henry Roland Totten, Professor of Botany in the University of North Carolina, Chapel Hill. A double-flowered form of Lilium superbum L. was found near Oakland, Garrett Co., Maryland, in 1954. Bulbs of this form were transplanted to Chapel Hill and Elon College, N. C., where they flowered the two following years. While this form is not as attractive as the typical form, it is worthy of mention since this is the first reported double-flowered variation of a native species of Lilium. As such, this new form is being designated as: Lilium superbum L., forma plenum E. T. Browne, Jr., f. nov. Forma nova cum tepalis plenis. In this double-flowered form some of the stamens have been changed entirely or in part to tepals. The pistil is unaffected, and plants of this form could, therefore, be fertile. This collection represents the only cccurrence of this variation among several thousand individuals of this species observed at this locality and elsewhere by the writer in Mary- land, Virginia, West Virginia, North Carolina and Kentucky. Specimens cited: MARYLAND. Garrett Co. Near Oakland. July 6, 1954. Edward T. Browne, Jr., s.n. (holotype, NCU; isotype, KY). A student in an undergraduate botany class has discovered a population of Trillium pusillum Michx. in Casey Co., Ken- tucky. This is the first report of this species from the state. The determination has been verified by Stanley J. Smith, New York State Museum, Albany. Mr. Smith writes that this species has been previously reported from only four other widely separated localities in the United States: the Chesapeake Bay region, the Carolinas, the Ozarks and the Sabine River region of Texas, and he feels that the Kentucky specimens are nearest those from the Ozarks, T. pusillum Michx., var. ozarkanum (Palmer and Steyerm.) Steyerm. (Stanley J. Smith, personal communication). Specimen cited. KENTUCKY. Casey Co. Pricetown, 4 mi. S of Liberty and E of US 127. Open field, rocky soil. April 28, 1961. Glenn W. Murphy, 92. (KY). In connection with a taxonomic monograph of the Lili- aceae of Kentucky which the writer is undertaking prelimi- nary to biosystematic studies in this family, several species 308 Rhodora [Vol. 63 have been found in herbaria or collected which have appar- ently not been reported for the state. Among these species is Smilacina stellata (L.) Desf. which would be expected to occur in Kentucky on the basis of its distribution in adjacent states, but, aecording to the distribution given by Small (1953) and Fernald (1950) and the smaller works of Mc- Farland (1942) and Braun (1943), the species does not occur in Kentucky. However, S. stellata is reported from Kentucky (as Vagnera stellata) in the second edition of Britton and Brown (1913), but it is not reported for Ken- tucky in the New Britton and Brown (Gleason, 1952). Al- though Galway (1945) indicates the presence of S. stellata in Kentucky on the distribution map of this species in her monograph of North American Smilacina, she does not cite any specimens from the state. Specimens cited: KENTUCKY. Jefferson Co. In sylvis. Louisville, Ky. C. Mohr, 2921. (Annotated by Desma H. Galway, 1939; Mo, 147191); Kentucky. (No county or other data given). 1840. C. Short, s.n. (GH). Xerophyllum might be expected to occur in Kentucky based on its distribution in adjacent states. However, this writer has not collected any specimens of this genus in Kentucky although a considerable effort has been made 1o locate plants. One collection apparently made in the state has come to the writer's attention nevertheless. If the valid- ity of this collection can be established, it will constitute a new state record since this genus is not given for Kentucky in any of the manuals, floras or papers of more limited ex- tent. On the sheet in question is an annotation label of K. J. Stone who has written: “X. tenax (Pursh) Nutt. Locality highly doubtful." X. tenax has a reported range from Wyo- ming to British Columbia and California (Abrams, 1940). On the other hand, X. asphodeloides (L.) Nutt. has a range from “Virginia to Georgia and Tennessee" (Fernald, 1950), and this fact would support the identification of this speci- men as X. asphodeloides. ln addition, in floral character- istics this specimen most closely approximates the latter species. Fernald (1950) points out that X. asphodeloides 1961] Browne — Liliaceae of Southeastern States 309 also occurs on “sandy pinelands, New Jersey, Delaware and North Carolina." Situations similar to these, though of not the same geological origin, occur in various areas of wesi- central and western Kentucky, and it is not unlikely, there- fore, that this specimen was collected in one of these places. It should be further pointed out that the place of collection, Bowling Green, is situated in the Barrens area of Kentucky. Specimen cited: KENTUCKY. (Warren Co.) Bowling Green. 1903. Sadie F. Price, s.n. (MO). Intensive field work has been under way in an attempt to relocate this species as well as Smilacina stellata in Ken- tucky. Atthis writing these efforts have so far proved futile. In regard to Xerophyllum, much acreage of the Barrens has been put into pasture and cultivation. It may be that both of these species, which in recent times have not been abun- dant, are now extinct or nearing this status in the state. Fernald (1950) indicates that Yucca Smalliana Fern. is probably only introduced and escaping north of North Caro- lina and Tennessee. This species occurs widely in Kentucky, and in some instances, it is freely reproducing and persisting around abandoned homesites. Braun (1943) writes that Y. filamentosa L. is “occasional along roadsides and railroad tracks". The writer has not seen any specimens of Y. fila- mentosa in a natural state anywhere in Kentucky, and pre- sumably it is actually Y. Smalliana to which this reference applies if these species are distinct. This latter species is not recorded by Fernald (1950), Gleason (1952), Small (1933), McFarland (1942) or Braun (1943) as occurring in Kentucky. While perhaps it may not be native to the state, Y. Smalliana should certainly be considered a member of the vascular flora of this area. Specimens cited: KENTUCKY. Barren Co. US 68, 0.8 mi. E of Barren- Warren Co. line. July 15, 1961. Elizabeth M. and E. T. Browne, Jr. 4570; Boone Co. US 25, 5.2 mi. N of Boone-Kenton Co. line. Open field between highway and Southern Rwy. tracks. Scattered plants. July 8, 1961. Elizabeth M. and E. T. Browne, Jr. 4469; Boyle Co. US 68, just S of Boyle-Mercer Co. line at bridge. July 15, 1961. Elizabeth M. and E. T. Browne, Jr. 4507; Carter Co. Dry roadside cut about % mi. E of bridge over Tygarts Creek, Ky. 182. July 1, 1961. Elizabeth M. and E. T. Browne, Jr. with J. C. Warden. 4421; Edmonson Co. Ky. 259, 6.9 310 Rhodora [Vol. 63 mi. E of junct. Ky. 185 and Ky. 70. Dry roadside bank. July 16, 1961. Elizabeth. M. and E. T. Browne, Jr. 4598; Fayette Co. Lexington. Persisting and spreading after cultivation on vacant tract of land across the Southern Rwy. tracks from the passenger station near S. Broadway. July 6, 1961. Elizabeth. M. and E. T. Browne, Jr. 4460; Grant Co. US 25, 6.8 mi. N of Grant-Scott Co. line. Open dry hillside. Plants of all ages. Very abundant. July 8, 1961. Elizabeth M. and E. T. Browne, Jr. 4466; Mercer Co. US 68, 11.3 mi, S of Mercer- Jessamine Co. line. Roadside. July 15, 1961. Elizabeth M. and E. T. Browne, Jr. 4504; Scott Co. US 25, 5.4 mi. N of Scott-Fayette Co. line. Open roadside. July 8, 1961. Elizabeth M. and E. T. Browne, Jr. 4462; Woodford Co. US 421, 3.6 mi. S of Woodford-Franklin Co, line. Steep road fill. Several plants in row as if planted. July 9, 1961. Elizabeth M. and E. T. Browne, Jr. 4496. Allium ampeloprasum L. has been collected in three coun- ties of west-central Kentucky. In each instance, these plants were found growing without cultivation apparently having originated from this source. As this species has not been previously reported from Kentucky, its mention here is justi- fied. Undoubtedly, there are other localities in the state, other than those cited below, where this species is natural- ized, but no other reports are available from elsewhere. As far as this writer is able to determine, this is apparently the first report of the occurrence of the typical variety of A. ampeloprasum in the United States as an escape (Small, 1933; Fernald, 1950 ; Gleason, 1952) since Fernald describes var. atroviolaceum (Boiss.) Regel as having “deep purple" flowers and Gleason, giving no varietal status to this, de- scribes the flowers as “purple-red”. The flowers of my col- lections were of several stages of development, and their color was light lavender becoming slightly darker upon drying. Specimens cited: KENTUCKY. Butler Co. Ky. 185, 6.2 mi. N of ferry over Green River. Wet road cut. July 16, 1961. Elizabeth M. and E. T. Browne, Jr. 4595; Edmonson Co. Ky. 259, 6.9 mi. E of junct. Ky. 185 and Ky. 70. Dry roadside bank. July 16, 1961. Elizabeth M. and E. T. Browne, Jr. 4599; Warren Co. Ky. 67, 3.2 mi. E of junct, Ky. 67 and Ky. 263. Open roadside. July 16, 1961. Elizabeth M. and E. T. Browne, Jr. 4591. The writer wishes to express his appreciation to the curators of the Gray Herbarium, the Herbarium of the Missouri Botanical Garden and the Herbarium of the University of North Carolina for permitting examination of specimens referred to herein. Dr. Herbert P. Riley has 1961] Barnes — Hybrid Aspens in Michigan 311 kindly read the manuscript. — UNIVERSITY OF KENTUCKY, LEXINGTON, KENTUCKY. LITERATURE CITED ABRAMS, L. 1940. Illustrated Flora of the Pacific States. v. 1. Stan- ford University Press, Stanford, Calif. BRAUN, E. Lucy. 1943. An Annotated Catalog of the Spermatophytes of Kentucky. Cincinnati, The Author. FERNALD, M. L. 1950. Gray's Manual of Botany, ed. 8. American Book Co., New York. GALWAY, DESMA H. 1945. The North American Species of Smilacina. Amer. Midl. Nat. 33: 644-666. GLEASON, H. A. 1952. The New Britton and Brown Illustrated Flora. N. Y. Bot. Gard., New York. HARPER, R. M. 1905. Phytogeographical Explorations in the Coastal Plain of Georgia in 1904. Bull. Torrey Bot. Club 32: 451-457. 1906. A Phytogeographical Sketch of the Altamaha Grit Region of the Coastal Plain of Georgia. Ann. N. Y. Acad. Sci. 17: 1-414. MCFARLAND, F. T. 1942. A Catalogue of the Vascular Plants of Kentucky. Castanea 7: 77-108. SMALL, J. K. 1933. Manual of the Southeastern Flora. The Author, New York. HYBRID ASPENS IN THE LOWER PENINSULA OF MICHIGAN' BURTON V. BARNES? INTRODUCTION One explanation for the polymorphism and diversity found in the aspens, Populus tremuloides Michaux and P. grand:- dentata Michx., is an exchange of genes between these spo- cies through hybridization and backcrossing. Anderson (1949) and others have demonstrated the importance of introgression as a vital force in evolution. In a study of the natural variation and clonal development of the aspens in the Lower Peninsula of Michigan I searched for hybrids on two research sites on forest land of the University of Michi- Information from the author's dissertation submitted in partial fulfillment of the requirements of the degree of Doctor of Philosophy at the University of Michigan, Ann Arbor. ?Research Forester, Intermountain Forest and Range Experiment Station, Forest Service, U. S. Dept. of Agriculture, stationed at Northern Idaho Forest Genetics Center, Moscow, Idaho. 312 Rhodora [Vol. 63 gan Biological Station, near Pellston, Michigan, and in other parts of Lower Michigan. In this paper I shall list and describe aspen hybrids and putative introgressants dis- covered in the Lower Peninsula of Michigan and shall discuss briefly the ubiquity of the P. grandidentata X P. tremuloides hybrid in southeastern Michigan as compared with its appar- ent rarity in the northernmost part of Lower Michigan. HYBRIDS AND INTROGRESSANTS POPULUS GRANDIDENTATA X TREMULOIDES A hybrid between P. grandidentata and P. tremuloides was first reported and described by Victorin (1930). Heim- burger (1936) produced hybrid plants by crossing the two species. He reported a scarcity of natural grandidentata- tremuloides hybrids and said this might be explained by the fact that “P. tremuloides flowers about a week to ten days before P. grandidentata in nature." Pauley (1956) reported that P. tremuloides flowered from 10 to 14 days before P. grandidentata and stated that the hybrids were not infre- quent in central and eastern Massachusetts. He had observed scattered hybrid individuals and hybrid swarms as well as what were presumed to be F, plants or backcrosses to one of the parents. In April 1956 I discovered one hybrid clone on an aban- doned field in Section 17, T. 10 N., R. 6 E., Saginaw County, Michigan. This hybrid was tentatively identified on the basis of the scattered hairs that were found on the terminal and lateral bud seales. Fallen leaves collected at the base of several trees in the group closely resembled the hybrid leaves illustrated by Pauley (1956). A second hybrid was discovered in October 1957 on the forest land of the University of Michigan Biological Station (NW1/2, Section 25, T. 37 N., R. 4 W., Emmet County). The 13 living ramets of this clone were severely damaged by hypoxylon canker and were apparently unable to compete successfully with vigorous neighboring clones of quaking and bigtooth aspen. In August 1958 Dr. Warren H. Wagner, 3SBarnes, Burton V. 1959. Natural variation and clonal development of Populus tremuloides and P. grandidentata in Northern Lower Michigan. 334 pp. Unpublished. 1961] Barnes — Hybrid Aspens in Michigan 313 Jr., discovered a hybrid swarm in Monroe County. Conse- quently, in the fall of 1958 a more intensive search for hy- brids was directed in southeastern Michigan. Thirty hybrid clones and putative introgressants were discovered in eight 3- to 8-hour trips. Many more hybrid clones could be located by systematic search in southeastern Michigan. Dr. Wagner located five additional clones in 1960. Hybrid clones have now been located in 10 counties of Michigan's lower penin- sula. Location of these hybrid clones and putative introgres- sants is shown in the Appendix. Specimens of these will be deposited in the University of Michigan Herbarium. Hybrids are typical in many localities in southeastern Michigan where the parent species occur. The number of hybrid clones is estimated at from 1 to 5 percent of those present in the localities observed. On the University of Michigan Biological Station land, however, the frequency is much lower, and this seems to be true for the entire Douglas Lakes area of the northern tip of the Lower Peninsula. The apparent abundance of the hybrid in Lower Michigan may be due to several factors. The difference in flowering time between bigtooth and quaking aspen may not be as great as one might suspect. I can offer no proof, however, that this barrier is stronger in northern Michigan than in southeastern Michigan. I did observe numerous receptive female flowers on several P. tremuloides clones near Pellston, Michigan, at the time when P. grandidentata pollen was being discharged. These flowers were usually located at the base or tip of a catkin and represent an intraclonal lag in flowering. Einspahr and Joranson (1960) reported a some- what similar intraclonal lag in flowering for seven aspen clones when their flower buds were forced to develop in the greenhouse. This intraclonal lag was apparently verified on only one tree under natural conditions. This developmental phenomenon may be closely related to the early-late leaf situation described by Critchfield (1960). Late flowering may be largely restricted to greenhouse conditions because not all flower buds may be in the same developmental stage when branches are collected 4 to 6 weeks before normal flowering time. 314 Rhodora [Vol. 63 While intraclonal variation might explain the source of receptive female P. tremuloides flowers,there are at least two other explanations. Pauley (1956) pointed out that in areas of temperature inversion the flowering of female P. tremu- loides clones may be retarded until it corresponds with the flowering time of neighboring P. grandidentata clones. Rec- ognizing the wide range in the variability of both species, interclonal differences in flowering time might also account for pollination of P. tremuloides flowers by P. grandidentata. Establishment of aspen seedlings is extremely difficult (ibid.). This may be the reason why the hybrid is appar- ently much more abundant in southeastern Michigan than in the northern tip of the Lower Peninsula. In southeastern Michigan many hybrid clones were discovered in abandoned fields or along the edge of cultivated fields where they bor- dered a woodlot. There appears to be more repeated distur- bance, primarily by cultivation, in southeastern Michigan than on University of Michigan Biological Station property and surrounding areas. More disturbance, coupled with more fertile soil and a somewhat more moderate climate in southeastern Michigan, probably means that the establish- ment of seedlings is easier. Most ramets of the hybrid clones were not more than 2 or 3 inches in diameter at breast height and were probably younger than 20 years in age. Leaf samples from 25 of the hybrid clones were measured and examined in detail. In general, the size of the hybrid leaves is intermediate between that of the parent species. The leaf shape of the hybrid clones varies as much as that within either of the parent species. To illustrate part of the variation among hybrid clones, two leaves typical of eight hybrid clones and typical leaves of the parent species are pictured in Fig. 1. INTROGRESSION Introgression, the flow of genes from one species into an- other through repeated backcrossing to either parent, has *At least 50 leaves from each hybrid clone were measured to obtain the mean blade width, blade length, and petiole length. The number of teeth along both margins of at least 30 leaves of each clone was counted. 1961] Barnes — Hybrid Aspens in Michigan 315 COMPARISON OF LEAF CHARACTERISTICS POPULUS TREMULOIDES 20 06 Of [i n i N CLONE 50 CLONE 99 ^ CLONE 6 POPULUS GRANDIDENTATA X POPULUS TREMULOIDES 00 0O 00 o0 CLONE 317 CLONE 313 CLONE 304 CLONE 320 /l DO 6 CLONE 33! CLONE 310 CLONE 321 CLONE 327 POPULUS GRANDIDENTATA e > CLONE 106 CLONE ! Figure | Comparison of leaf characteristics of P. tremuloides, P. grandidentata, and hybrids. 316 Rhodora [Vol. 63 been described in Acer (Desmarais, 1952), Quercus (Cooper- rider, 1957), and in several other plant genera (Anderson, 1949, 1953; Stebbins, 1950). A few hybrids collected in southeastern Michigan closely resemble P. tremuloides, yet have several characteristics of P. grandidentata. A series of frequency polygons illustrates leaf differences between the hybrids and the parent species (Figs. 2, 3, 4,5). Since leaf specimens were not collected from the parents in southeast- ern Michigan, data from the aspen clones studied intensively in the forest of the University of Michigan Biological Station were used (Barnes, 1959). The hybrid clones are more nearly like P. tremuloides in respect to blade width, blade length, and petiole length. The number of teeth per leaf ' side, however, corresponds more closely to P. grandidentata. Of the four characters illustrating the intermediacy of the hybrid clones, the number of teeth per leaf side and the blade length are the most explicit quantitative characters sepa- rating hybrids from the parent species. Hairiness of the terminal and lateral bud is one of the most striking differences between P. tremuloides and P. grandidentata. 'This character, however, lends itself less easily to quantification than the four leaf attributes. Ander- son (1957), with a semigraphical method, used four of the more distinct features to characterize the differences among clones of the hybrids and parent species (Fig. 6). The hybrids are intermediate between the two parents, but they resemble P. tremuloides more closely than P. grandidentata. Three of the putative hybrid clones are suspected as being introgressants or backcrosses (clones 310, 317, and 333). Without additional study and analysis of the parent and hybrid populations in southeastern Michigan, one can report only that the leaf characteristics of several hybrid clones closely resemble P. tremuloides. Usually, F, hybrids between distinet species are uniform in their characteristies, while segregates appear in the second and following generations (Anderson, 1949). However, the variation within both parents is so diverse and as yet only so summarily described, that F, plants seeming to resemble P. tremuloides in one, two, or more characteristics might be expected in the F, generation. 1961] Barnes — Hybrid Aspens in Michigan 317 Sympatric introgression may occur between two distinct species that are ecologically or physiologically separated. co P. grandidentata —— P. tremuloides : ——— P. grandidentata X EN 2. n P. tremuloides ik j! a lO T is ! i uJ l ós | ee EHE pO + | \ wù 6 3E l \ o | \ a T ! i w 4 f H \ a / = + A \ 224 FON + / ee E i i 0 pt —À——————————--— [6] l 2 3 4 5 61 8 3 10 BLADE LENGTH -Centimeters e c» a Mt P. grandidentata bagel E — —— P. tremuloides + ——-P. grandidentata X n IOS P. tremuloides zx: A 9 8 1! e zT: || u T p O16 l | a BE \ o4 | 2 c7] 2 EX N mo "i LM HR / x [6] 4 4 4 n L 4 n n L | O | 2 3 4 5 6 7 8 9 10 BLADE WIDTH - Centimeters Fic. 2. Frequeney polygons illustrating differences in blade length for clones of the hybrid and the parent species. Fic. 3. Frequency polygons illustrating differences in blade width for clones of the hybrid and the parent species. Anderson (1948, 1949) pointed out that where the species are separated by habitat requirements, only when the “‘habi- tat is hybridized” can the F, hybrids find an ecological con- dition suitable for establishment and survival. In succeeding 318 Rhodora [Vol. 63 generations, according to Anderson (1949), the hybrids segregate but the habitat usually does not. Thus, hybrid plants possessing about the same habitat requirements of either of the parent species, i.e. backcrosses to either parent, are favored over F, or successive generation segregates. P. tremuloides and P. grandidentata, however, are not isolated ecologically by their habitat requirements. The habitat of P. tremuloides overlaps that of P. grandidentata, and both species are frequently growing together on the same site. Therefore, disturbance by man or natural agents, is prob- ably more important as a prerequisite to seedling establish- ment than “hybridizing the habitat." Thus, it is probable that in southeastern Michigan, F, and F, hybrids, as well as introgressants, would be about equally fitted for a given habitat that is suitable for both parents. The variability of the parental species, P. tremuloides and P. grandidentata, may be due to multiple alleles for genes in the gene pool, mutation, drift, and introgression. Without further analysis it is difficult to say what combination of these factors is responsible for the polymorphism exhibited by the two species. POPULUS ALBA X TREMULOIDES Two individuals of the P. alba X tremuloides hybrid were discovered in the Lower Peninsula. One clone was located at the edge of an abandoned field near the juncture of Woods Road and Highway M-18 (SE1/4, SE 1/4, Section 23, T. 23 N., R. 2 W., Gladwin County. The form of this tree resembles that of P. tremuloides rather than P. alba trees in Lower Michigan. A second P. alba X tremuloides hybrid was discovered along the Deckerville Road, 81/2, Section 30, T. 13 N., R. 10 E. Tuscola County. This hybrid was recognized by its extremely shiny green leaves and its silvery bark. P. alba X P. tremuloides hybrids were reported in Canada by Heimburger (1936) and Peto (1938). According to Peto, the P. alba X tremuloides hybrid was not as frequent as the alba X grandidentata hybrid in the vicinity of Ottawa, Canada. The leaves of the P. alba X tremuloides hybrid are inter- 1961] Barnes — Hybrid Aspens in Michigan 319 mediate in shape and serration between the parents. P. alba leaves are palmately lobed and tomentose on the undersur- face. The margin of P. alba leaves is irregularly and shallow- ly serrate. The leaves of the hybrid are not lobed but are irregularly toothed. The teeth of the hybrid leaves resemble . P. grandidentata ——— P. tremuloides ——— P. grandidentata X P. tremuloides Eb. $e 1 2 emu u + IN z 8] DE 9 IM T JN 6l gs u o "d H he E l . ui ] 1. +4 : ME zc \ ; z \ = + bæ- °- 0 Ct tt (0) | 2 3 4 5 6 T 8 PETIOLE LENGTH - Centimeters . P. grandidentata IO a 5 MEO MOM HMM d Co — — P. tremuloides a a s ——- P. grandidentata X m put P. tremuloides oren px i : S | B 2d /^ i 4] E TAN, N /N Pd a B : i / \ , u Lr : Vv 7: 5 > roe : AVAAN V 5 : z + É [6] fL fec jus THERETO Late CERT NN SOR MEE GNIS E o Eo cnl T T j T T T T T T T T T T T T T T T Oo 4 8 l2 1620" 24 2832F 36 NUMBER OF TEETH PER LEAF SIDE Fic. 4. Frequency polygons illustrating differences in petiole length for clones of the hybrid and the parent species. Fig. 5. Frequency polygons illustrating differences in number of teeth per leaf side for clones of the hybrid and the parent species. those of P. tremuloides more than those of P. alba. The numbers of teeth per leaf side for the Gladwin and Tuscola County hybrids were 14.6 and 15.1, respectively. The under- 320 Rhodora [Vol. 63 side of the immature leaves of the hybrid is moderately tomentose and becomes glabrous as the leaves mature. Buds of the hybrids were covered with a tomentum that was not as dense as the tomentum on buds of P. alba. The immature shoots were also covered with a moderate amount of tomentum that did not entirely disappear from the new shoots during their initial year. The bark of the hybrid had many diamond-shaped cracks — a typical characteristic of P. alba trees of the same size. The bark was rougher in texture and more silvery-white in color than bark of typical sapling or pole-size trees of P. tremuloides. Catkins collected from the hybrid in Gladwin County were predominantly male, but some female flowers were discov- BLADE LENGTH IN CM ^ o ~ 7 O ` ipe 8 " E 2 p" -Qm 89 a tom 39 Om Ei 20- ig S è 3 BQ" oni e 20.195 ig- EEG 1922] i Zz z ss igo 777 32 8 3 e al Wc 9 io RA gE n E | m^ ? E § 3 5 7 " | © š à $ it o 8 m ias iln LJ » " 7" > aia TEM o T N W^ io § » L E ; NU sf 2 * $7 To) o SẸ oo O zO DA S IS = E "D P v “O se) ri à 4 : H B ERES i a —. < aww Si wyeanu P009 dul Poppe A 4) nor - 7O8mAS 38033 avcn2w m TI HT t Fic. 6. Pictorialized scatter diagram, contrasting four characters in different clones of P. tremuloides, P. grandidentata, and their hybrids. 1961] Barnes — Hybrid Aspens in Michigan 321 ered on several eatkins. Peto (1938) reported that one of the three hybrids near Ottawa, Canada, was “monoecious.” An interesting phenomenon observed on one of the catkins from the predominately male hybrid was a row of female flowers proceeding from the base to the tip along one side of the catkin. The bracts of the aments in the hybrid are characterized by digits that are not as deeply cleft as in P. tremuloides, but not as shallow as in P. alba. The stigma color of the hybrid (pink) is intermediate between that of its parents. The receptive stigma of P. tremuloides is typically red or scarlet, whereas that of P. alba is “yellow-green to rose" (Amann, 1956). POPULUS ALBA X GRANDIDENTATA One clone of the alba X grandidentata hybrid was dis- covered in Section 17, T. 10 N., R. 6 E., Saginaw County. This clone consisted of about 10 to 15 ramets, the largest of which was about 8 inches in diameter at breast height. Victorin (1935) was probably the first to recognize that P. alba was hybridizing with the native aspens of Quebec. Heimburger (1936) stated that the P. alba X grandidentata hybrid was probably the one that Victorin had observed, since the parents flower approximately at the same time. Heimburger (1936) discovered several hybrid trees, and Peto (1938) listed what were apparently seven different clones of this hybrid from Ontario and Quebec. McComb and Hansen (1954) reported two hybrid clones from south- eastern Iowa, which were described by Little, Brinkman, and McComb (1957). Subsequently Gatherum (1960) has reported two additional clones from Iowa. SUMMARY Thirty-eight clones of the P. grandidentata X tremuloides hybrid are reported from 10 counties in the Lower Peninsula of Michigan. Although the majority of hybrids have leaf and bud characteristics intermediate between the two par- ents, three clones are suspected of being introgressants. The reasons for the abundance of hybrids and the greater fre- quency of hybrid clones in southeastern Michigan than in the northern part of the Lower Peninsula are discussed. 322 Rhodora [Vol. 63 Two clones of the P. alba X tremuloides hybrid and one clone of the P. alba X grandidentata hybrid are reported and described. APPENDIX Location of P. grandidentata X tremuloides clones in the Lower Peninsula of Michigan Clone Number of mM E number trees in clone Location 301 5 Section 17, T. 10 N., R. 6 E. , Saginaw Orau- ty 302 13 NW1/4, SE1/4, Section 25, T. 37 N, R. 4 W., Emmet County 303 3 SE1/4, SW1/4, Section 25, T. 37 N, R. 4 W., Emmet County 304 2 Section 6, T. 7 S., R. 7 E., Monroe County 305 2 Section 6, T. 7 S., R. 7 E., Monroe County 306 1 Section 6, T. 7 S., R. 7 E., Monroe County 307 2 Section 6, T. 7 S., R. 7 E., Monroe County 308 4 Section 6, T. 7 S., R. 7 E., Monroe County 309 3 Section 6, T. 7 S., R. 7 E., Monroe County 310 1 Section 6, T. 7 S., R. 7 E., Monroe County 311 1 Section 6, T. 7 S., R. 7 E., Monroe County 312 3 Section 6, T. 7 S., R. 7 E., Monroe County 313 8 NE1/4, Section 33, T. 10 N., R. 5 E., Sagi- naw County 314 6 S1/2, SE1/4, Section 23, T. 4 S., R. 7 E., Washtenaw County 315 6 S1/2, SE1/4, Section 23, T. 4 S., R. 7 E. Washtenaw County 316 ca. 50 N1/2, NW1/4, Section 28, T. 4 S., R. 8 E., Wayne County 317 ca. 20 N1/2, NW1/4, Section 28, T. 4 S., R. 8 E., Wayne County 318 3 N1/2, NE1/4, Section 30, T. 4 S., R. 8 E., Wayne County 319 1 N1/2, NE1/4, Section 30, T. 4 S., R. 8 E., Wayne County 320 ca. 30 Section 5, T. 8 N., R. 7 E., Genesee County 321 ca, 12 Section 5, T. 8 N., R. 7 E., Genesee County 322 2 SE1/4, Section 17, T. 9 N., R. 7 E., Genesee Countv 323 3 SE1/4, Section 17, T. 9 N., R. 7 E., Genesee County 324 1 SE1/4, NW1/4, Section 19, T. 7 S., R. 6 E., Monroe County 1961] Barnes — Hybrid Aspens in Michigan 323 Clone Number of ik UN page Er an c MN number trees in clone Location 325 1 SE1/4, NW1/4, Section 19, T. 7 S., R. 6 E., Monroe County 326 1 E1/2, SW1/4, Section 24, T. 7 S., R. 5 E., Lenawee County 327 1 E1/2, SW1/4, Section 24, T. 7 S, R. 5 E, Lenawee County 328 1 E1/2, SW1/4, Section 24, T. 7 S, R. 5 E, Lenawee County 329 ca. 50 E1/2, SW1/4, Section 24, T. 7 S, R. 5 E. Lenawee County 330 al NWI1/4, Section 30, T. 6 S., R. 7 E., Monroe Countv 331 2 SE1/4, NE1/4, Section 33, T. 3 S., R. 9 E., Wayne County 332 6 SE1/4, NE1/4, Section 33, T. 3 S., R. 9 E., Wayne County 333 1 SE1/4, NE1/4, Section 33, T. 3 S., R. 9 E., Wayne County 334 Not recorded Cedar Lake, Waterloo Recreation Area, near camp, Washtenaw County. (Wagner s.n.) 335 Not recorded East side Jefferson Rd., Section 15, T. 16 N., R. 2 E., Midland County. (Wagner 8993) 336 Not recorded Edge of Muskegon State Park, SW1/4, Sec- tion 16, T. 10 N., R. 17 W., Muskegon County. (Wagner 9250) 337 Not recorded North side Water-Munith Rd., Section 15, T. 1 S., R. 2 E., Jackson County. (Wag- ner 9259) 338 Not recorded M-83, 4.7 mi. south of Frankenmuth, T. 10 N., R. 6 E., Section 22, Saginaw County. (Wagner 9285) LITERATURE CITED AMANN, GOTTFRIED. 1956. Bäume and Sträucher des Waldes. Verlag J. Neumann-Neudamm, Melsungen. 231 p. ANDERSON, EDGAR. 1948. Hybridization of the habitat. Evolution 2: 1-9. 1949. Introgressive hybridization. John Wiley & Sons, Inc., New York. 109 p. . 1953. Introgressive hybridization. Biological Re- view 28: 280-307. 1957. A semigraphical method for the analysis of complex problems. National Acad. Sci. Proc. 43: 923-927. 324 | Rhodora [Vol. 63 COOPERRIDER, MIQUAKO. 1957. Introgressive hybridization between Quercus marilandica and Q. velutina in Iowa. Am. Jour. Bot. 44: 804-810. CRITCHFIELD, WILLIAM B. 1960. Leaf dimorphism in Populus tricho- carpa. Am. Jour. Bot. 47: 699-711. DESMARAIS, Yves. 1952. Dynamics of leaf variation in the sugar maples. Brittonia 7: 347-388. EINSPAHR, DEAN W., AND PHILIP N. JoRANSON. 1960. Late flowering in aspen and its relation to naturally occurring hybrids. Forest Sci. 6: 221-224. GATHERUM, GoRDON E. 1960. Current tree improvement research in Iowa. Proceedings, 1st Central States Forest Tree Improvement Conf., pp. 16-19. HEIMBURGER, CARL C. 1936. Report on poplar hybridization. Forestry Chronicle 12: 285-290. LITTLE, ELBERT L., JR, KENNETH A. BRINKMAN, AND A. L. MCCOMB. 1957. Two natural Iowa hybrid poplars. Forest Sci. 3: 253-262. MCCOMB, A. L., AND NORMAN J. HANSEN. 1954. A naturally occurring aspen-poplar hybrid. J. Forestry 52: 528-529. PAULEY, Scorr S. 1956. Natural hybridization of the aspens. Univ. Minnesota Forestry Note 47, 2 p. PETO, F. H. 1938. Cytology of poplar species and natural hybrids. Canad. Jour. Res. 16: 445-455. STEBBINS, G. LEDYARD, JR. 1950. Variation and evolution in plants. Columbia Univ. Press, New York. 648 p. VICTORIN, FRÈRE MARIE. 1930. Les variations laurentiennes du Popu- lus tremuloides et du P. grandidentata. Contrib. Lab. Bot. Univ. Montreal 16. 16 p. 1935. Flore Laurentienne. Imprimerie de La Salle, Montreal. 917 p. A DISJUNCT COMMUNITY OF CHESTNUT OAK IN MISSISSIPPI EDWARD G. ROBERTS Coker and Totten (1944) and Harrar and Harrar (1946) report the range of chestnut oak (rock oak, rock chestnut oak), Quereus prinus L. as being south to Georgia and Ala- bama. The range map of Munns (1938) shows it in the northeastern tip of Mississippi — Tishomingo County. Mat- toon and Beal (1936) report “It is found in the extreme northeastern counties, where it is common on the sandstone bluffs.” Lowe (1913) writes of the vegetation of the North- Central Plateau of Mississippi that the typical upland for- 1961] Roberts — Chestnut Oak in Mississippi d20 ests of the more sandy eastern parts of the region contain, among other things, Quercus prinus. This geographic de- scription would include the western part of Calhoun County. But then Lowe goes on to report chestnut oak as being pres- ent in the bottoms of the same region and in the hills and bottoms of the deep loessal soils region to the west. There is unquestionably some confusion between Quercus prinus and Quercus michauxti. In a later bulletin, Lowe (1921) reports that chestnut oak is found in the northern part of the Pon- totoc Ridge and, elsewhere in the bulletin, in the high hills about the headwaters of the Hatchie River. These seem to be two descriptions for essentially the same area. Further on in the bulletin he gives the range of chestnut oak as Tish- omingo County and the high, rocky divides of Itawamba, Alcorn, and Tippah Counties. Itawamba is immediately to the south of Tishomingo. Alcorn and Tippah are immediate- ly to the west and are at the headwaters of the Hatchie. Recently, Newton Hanson of Bruce, Mississippi, forester for the E. L. Bruce Company, took the writer to see some chestnut oak in northwestern Calhoun County, Mississippi. One tree by the side of the road is over 16 inches d.b.h. It is in every respect a typical chestnut oak. Several others are within a few hundred feet of it. Mr. Hanson says that they are scattered over perhaps a thousand acres on a ridge which runs roughly north and south. Superficially, at any rate, there seems to be nothing to distinguish this ridge from countless others in north Mississippi. The area where the trees were seen is about three-quarters of a mile east of the Yalobusha County line and about three and one quarter miles south of the Lafayette County line in Section 22, Town- ship 11 South, Range 3 West from the Chickasaw Base Line and Chickasaw Meridian or at Latitude 34° 7’ N, Longitude 89° 30' W. This location is some eighty miles southwest of the range shown by Munns and perhaps sixty miles south and west of the locations reported by Lowe (1921). The day may not be far distant when man converts this oak-hickory ridge to loblolly pine with the elimination of this disjunct chestnut oak community. —- FORESTRY SCHOOL, MIS- SISSIPPI STATE UNIVERSITY. 326 Rhodora [Vol. 63 LITERATURE CITED CoKER, W. C. AND H. R. ToTTEN. 1944. Trees of the southeastern states. 3rd ed. The University of North Carolina Press. Harrar, E. S. AND J. G. HARRAR. 1946. Guide to southern trees. Mc- Graw-Hill. LowE, E. N. 1913. Notes on the Flora of Mississippi in Forest con- ditions of Mississippi. Miss. State Geol. Sur, Bull. No. 11. LowE, E. N. 1921. Plants of Mississippi. Miss. State Geol. Sur. Bull. No. 17. MATTOON, W. R. AND J. M. BEAL. 1936. Forest trees of Mississippi. Miss. Agr. Ext. Ser. Bull. No. 32. MUNNS, E. N. 1938. The distribution of important forest trees of the United States. U. S. D. A. Miscl. Pub. No. 287. IMPORTANT NOTICE to members, subscribers and librarians Beginning with Volume 64 to be published in 1962, Rhodora will become a quarterly journal. It is anticipated that Volume 64 and subsequent volumes will contain ap- proximately the same number of pages as previously pub- lished recent volumes. It will be the aim of the editors to have each quarterly number consist of 96 pages, including the cover. The above change does not imply any other changes in editorial policy. The Editors Volume 63, No. 754, including pages 267-296, was issued November 10, 1961. LOW Rrcreptpp- broretiy $4 Dodota JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Conducted and published for the Club, by REED CLARK ROLLINS, Editor-in-Chief ALBERT FREDERICK HILL STUART KIMBALL HARRIS RALPH CARLETON BEAN Assocddbs Editore IVAN MACKENZIE LAMB ROBERT CRICHTON FOSTER ROLLA MILTON TRYON Vol. 63 December, 1961 No. 756 CONTENTS: Viola Rafinesquii: Nomenclature and Native Status. LOU HE SEITE DS aee reee aired eon pase onec ose et e rd 327 The Genus Oleandra of Costa Rica. NEED SE LIGUE BSEC LS IARE E ae 335 aw Vascular Plants on the Cinder Cone of Paricutin Volcano we Be AW JORN H. RECOUNT Aaner E d Ha cic inris 340 A Weedy Crucifer Again Reaches North America. CME O: Kol uou. ae ee ee C m 845 An Albino Fruited Form of Gaultheria procumbens. TATE OES COUPE S eL sessiles d dug taescts re CULTE CERT 346 Corrections in Reports of the Committee on Plant Distribution. R. C. Bean, A. F. Hill and R. J. Eaton ........ 348 A Botanical Atlas of the Danish Flora. (Review) DONA SE TOON fence cea MM ru Me Er M 349 Md NM M Lu iocis dict Ut iuecilisteseseneno censi loeis 350 Important Notice to Members, Subscribers and Librarians .... back cover The Nem England Botanical Club, huc. Botanical Museum, Oxford St., Cambridge 38, Mass. RHODORA.—A monthly journal of botany, devoted primarily to the flora of North America and floristically related areas. Price, $6.00 per year, net, postpaid, in funds payable at par in United States currency in Boston; single copies (if available) 60 cents. Back vol- umes 1-58, with a few incomplete, can be supplied at $5.00 per volume, Volume 59— available at $6.00. Somewhat reduced rates for complete sets can be obtained upon application. Scientific papers and notes, relating directly or indirectly to the plants of North America, will be considered by the editorial com- mittee for publication. Articles concerned with systematic botany and cytotaxonomy in their broader implications are equally accept- able, All manuscripts should be double-spaced throughout, Please conform to the style of recent issues of the journal. Illustrations can be used only if the cost of engraver's blocks is met through the author or his institution. Forms may be closed five weeks in advance of publication. Extracted reprints, if ordered in advance, will be furnished at cost. Address manuscripts and proofs to Reed C. Rollins, Gray Herbarium, 22 Divinity Avenue, Cambridge 38, Mass. Subscriptions and orders for back issues (making all remittances payable to RHODORA) should be sent to Albert F. Hill, Botanical Museum, Oxford Street, Cambridge 38, Mass. Second Class Postage Paid at Boston, Mass. Manufaetured by THE LEXINGTON PRESS, INC. Lexington, Mass. iTRbooora JOURNAL OF THE NEW ENGLAND BOTANICAL CLUB Vol. 63 December, 1961 No. 756 VIOLA RAFINESQUII: NOMENCLATURE AND NATIVE STATUS LLOYD H. SHINNERS The name Viola bicolor Pursh, Fl. Am. Sept. 1: 175, 1814 (Dec. 1813), was long rejected as a later homonym of V. bicolor Gilibert, Fl. Lithuanica 2: 123, 1782. It has been pointed out, however, that Gilibert used his own peculiar system of descriptive species names consisting of varying numbers of words (Hylander, 1945; McVaugh, 1949). That is to say, he used abbreviated polynomials, many of them short enough to simulate Linnaean binomials, but shown by their association with indubitable polynomials to be really the latter. Hence V. bicolor Gilibert does not invalidate V. bicolor Pursh, an authentic binomial. For this reason I employed the latter in my Spring Flora of the Dallas-Fort Worth Area, Texas (1958), and distributed specimens so named. Unfortunately there does exist a validly published homonym earlier than that of Pursh, not listed in Index Kewensis. This is Viola bicolor Hoffmann, Deutschlands Flora (ed. 2) p. 170, 1804. Though only briefly described incidentally to the account of V. tricolor L., the name is both valid and legitimate. Hence V. Rafinesquii Greene, Pittonia 4: 9, 1899, remains the correct name for the American plant if considered a species distinct from the European V. Kitai- beliana R. & S. Fernald (1938) considered it so similar to the Old World species that he reduced it to varietal rank as V. Kitaibeliana var. Rafinesquu. But the morphological similarity is not the whole story. Breeding behavior (specifically, the occurrence of cleistogamy in V. Rafinesquii but not in V. Kitaibeliana) 327 328 Rhodora [Vol. 63 and chromosome number (so far as known, from rather few counts) are different. Further, the question of nativity !s intimately involved with taxonomic evaluation. For reasons to be detailed later, I believe that V. Rafinesquii is indigenous to North America, and that this gives additional support to its recognition as a species distinct from V. Kitaibeliana. Two sources of evidence have been taken to indicate an Old-World origin for our plant. The first, its weedy behavior in the Atlantic States, must be rejected, since a number of field weeds in the East are immigrants, not from the Old World, but from areas farther west in North America. ( This is more fully discussed below under item 2, Geography and Ecology.) The second, its close resemblance to members of the V. Kitaibeliana complex of southern and eastern Europe, is not at all as conclusive as Fernald (and Wilmott) found it. We have had a number of publications on the so-called amphigean or amphi-Atlantic members of the floras of north- eastern North America and northwestern Europe. But our knowledge of the flora of the Southern United States, and more particularly its relationships with the Mediterranean and other floras, is in a primitive state. For the present one suggestive example must suffice. The small umbelliferous genus Bifora comprises two species in the Mediterranean region, one in Indo-China, and one in the United States (Texas, Oklahoma, Arkansas). The Mediterranean and American species are extremely similar in general appear- ance, though differing more markedly in technical morpho- logical details than do Viola Rafinesquii and the V. Kitaibeli- ana complex. Bifora americana is annual and decidedly weedy. In North Central Texas it is a familiar late-spring wild flower, on “prairies, rocky slopes, and roadsides, lime- stone areas; very common, often abundant" (Shinners, 1958). In Dallas it is a common weed in vacant lots. It has also been found introduced as a roadside weed in Rusk Coun- ty, Texas, about 100 miles east of its main range. Neverthe- less its native status is beyond question. So close is the parallel with the Viola Rafinesquii — V. Kitaibeliana situa- tion that the morphological similarities between the latter cannot be taken as proof of common nativity. The sources of evidence in support of the nativity of V. 1961] Shinners — Viola Rafinesquii 329 Rafinesquii in North America are more numerous. While no single one can be taken as conclusive beyond all question (owing in part to the incompleteness of our knowledge, especially of the cytology, breeding behavior, and micro- forms of V. Kitaibeliana), there is remarkable agreement among them. Taken together, I believe that they demon- strate that Viola Rafinesquii is a native American species. The arguments may be grouped under six overlapping and closely interrelated headings. 1. EARLY RECORDS. Fernald observed that the earliest record of this species was in 1739 in Gronovius’s Flora Vir- ginica (actually 1743, since it appeared in the addenda on p. 182 in part 2, published in the latter year). The brief description fits quite well, but there is no information as to exact locality, habitat, or abundance. Pursh, in describing it as V. bicolor, said only that it grew “in fields of Pennsyl- vania and Virginia." This sounds weedy enough, but a field recently cleared from wilderness, and still surrounded by wilderness, is a different thing from fields in the 20th Cen- tury. The importance of this point will appear in the discus- sion of geography and ecology (item 2, below). Barton (1818), treating the area within 10 miles of Philadelphia, reports its occurrence “on the grassy borders of cultivated fields bordering Cooper's creek, Jersey, not far from the Market-street ferry and the Burlington road." Nuttall (also in 1818), without specifying any localities, said *apparently native." Other early reports give stronger evidence of the plant’s being native. Elliott (1817) says “Found near the Chatahouchie river, Creek nation, by Dr. Latham," far re- moved from cultivated fields. His predecessor in the area, Thomas Walter, remaining close to civilization, did not find it. Schweinitz, writing at Salem, North Carolina, in 1821, declared that “this interesting Viola grows with us, along the river bottoms and in retired mountain vallies in such a manner as to leave no doubt, that it is a true native." West of the Mississippi (where the plant is today and has long been far more abundant than to the east) there are likewise early testimonials in support of native status. Reverchon (1880) had no doubt of its being native in Dallas County, Texas. T. C. Porter (1880) quoted Reverchon and added 330 Rhodora [Vol. 63 evidence of its natural occurrence in Colorado: “Mr. Rever- chon, in the last number of the GAZETTE, reports it from Dallas county, Texas, and says, ‘I am satisfied it is native. I have met it in large patches in remote woods and prairies, sometimes very far from settlements.’ To this I may add the fact that it has also been collected in Colorado by Mr. Wm. A. Henry, who thus wrote me Aug. 29, 1876 — ‘I send you more of the violet. It grows on a warm, dry slope at the mouth of Boulder canon, in a rather inaccessible place. I have seen a few stalks farther up the canon. It blooms very early, along with Leucocrinum montanum, so that it has probably escaped the notice of other collectors. I gathered it three years before in the same place. It could have been introduced, but I greatly doubt that seeds of recent introduc- tion could have reached the spot where I found these plants. " It was collected in Oklahoma (‘Indian Territory") in 1875 (Butler, no other data; SMU). These early records do not support Fernald's belief that the plant was an intro- duction from Europe. 2. GEOGRAPHY AND ECOLOGY. Both the geographic area and weedy behavior of Viola Rafinesquii are strikingly like those of certain other species of unquestionably native status. Bifora americana, has already been cited. Mirabilis albida, first described in Walter's Flora Caroliniana, is rare and local east of the Mississippi River, but common and wide- spread west of it, especially in Texas and Oklahoma. There it is distinctly weedy, appearing on roadsides, pastures, and old fields. There is an endemic var. lata in eastern Texas (Shinners, 1951). Hedyotis nigricans, first described from Florida, is abundant and weedy on limestone outcrops, dis- turbed prairies, and eroding ground in central Texas and Oklahoma, but in eastern Texas and Louisiana and in states farther east it is much less common. Helenium amarum (H. tenuifolium) early in the 19th Century occurred as far east as Mississippi. Gray's Synoptical Flora (1886) says “Ar- kansas to Mississippi, Florida, and Texas; becoming a naturalized weed throughout Southern Atlantic States." Undoubtedly its indigenous range centered in Texas, where (in strikingly resemblance to Mirabilis albida) there is an endemic var. badium (perhaps better treated as a distinct 1961] Shinners — Viola Rafinesquii 331 species). I long ago reported instances of the same species occurring in the same general region in populations of two origins : one native, the other introduced. Sporobolus neglec- tus and S. vaginiflorus var. inaequalis are present in south- ern Wisconsin as very localized natives on rocky hillsides and, at the same time, as rapidly spreading railroad weeds largely introduced from elsewhere (Shinners, 1941). This I believe has been precisely the history of Viola Rafinesquii. Fernald's argument that because of its weedy behavior and habitats in the eastern states it must have been introduced from Europe cannot be accepted. The occurrence of the plant in Colorado is not out of order for a native of the Gulf Southwest. Thalictrum dasycarpum var. hypoglaucum, ranging as far west as British Columbia and Arizona, occurs east to Missouri, Arkansas, and Louisi- ana (Boivin, 1944). In Texas it occurs only in the eastern third of the state, so that there is a very wide gap in its range. Brickellia grandiflora, a Rocky Mountain and Pacific species, is known from a few stations in Arkansas and Missouri (Robinson, 1917), a distribution pattern resem- bling that of Viola Rafinesquii in reverse. Aster laevis, a common species of the northeastern United States, also occurs in Colorado, but is absent from a large area in between. (In the northern Rocky Mountain region var. laevis is replaced by var. Geyeri.) 3. VARIATION. According to Wilmott (quoted by Fernald, 1938, p. 446), one of the major differences between V. Ra- finesquii and V. Kitaibeliana was in the sepals: with strongly ciliate margin in the former, with little or no ciliation in the latter. In north-central Texas and eastern Oklahoma, V. Rafinesquii occasionally has completely glabrous sepals, and there are rare intermediates with sepals ciliate only in basal portion or very sparsely: Fernald evidently had insufficient material from this area, for he states that the ciliation is “essentially constant." I have seen specimens with entirely glabrous sepals from Logan, Love, Payne, and Pontotoc counties in Oklahoma, and from Wise County, Texas; with partially glabrous sepals from Cherokee and Delaware coun- ties in Oklahoma, and from Wise County, Texas. It might at first be thought that this weakens the case for taxonomic 332 Rhodora [Vol. 63 distinctness and native status of V. Rafinesquii. But there is a difference in sepal shape (see further under item 4, below), so that ciliation is not the only point of difference. The geographic distribution of the glabrous variants is, however, quite significant. If the plant were an alien which had first been introduced into the Atlantic states, one would not expect a minor variation to be thus restricted (though it is remotely possible). But if it is native, and in process of developing endemic races, that is precisely what one would expect. The occurrence of endemic varieties of Mirabilis albida and Helenium amarum within the Gulf Southwest, mentioned under item 2, are suggestive examples. I believe that, when taken in conjunction with the other items being cited as evidence of native status, the localized glabrous vari- ants of V. Rafinesquii strengthen the case. 4. ABSENCE FROM THE OLD WORLD. Fernald and Wilmott were unable to match V. Rafinesquii with any forms in the V. Kitaibeliana complex. In checking through European floras, I found in Pereira Coutinho’s Flora de Portugal (ed. 2, p. 500, 1939) V. Kitaibeliana var. Machadiana, described as "papiloso-aspera." Specimens of V. Rafinesquii were sent to A. R. Pinto de Silva of the National Agronomic Station at Sacavem, Portugal, with a request for an opinion as to whether they might be var. Machadiana. He very kindly took them to Lisbon (LISE) where, with the help of Dr. L. G. Sobrinho, they were compared with a paratype and other specimens determined by Coutinho. He reports (in letter of 25 December 1956) that “your V. Ranfinesquii is different from Machadiana and that its original country cannot be Portugal.” He found that the main differences were in the basal leaves (with ovate limb and the incisions of the crenate teeth more **opened" in Machadiana; with depressed-orbicu- lar limb, the incisions of the crenate teeth not “opened” in Rafinesquii) and in the sepals (lanceolate, broader near the base but narrower, and glabrous at the margin or scarcely, sparsely and hispidly ciliate in the upper part and near the base in Machadiana; sepals constricted as a shoe and regu- larly hispidly and strongly ciliate in Rafinesquii). He adds “T have observed also some specimens of V. Kitaibeliana (determined as so!) from South Spain (Sierra Nevada) 1961] Shinners — Viola Rafinesquii 333 (LISE). These have sepals with a little more cilia but they are still different from your Rafinesquit.” The opinion of Fernald and Wilmott that V. Rafinesquii must be some very rare and local European race remains without proof. 5. CLEISTOGAMY. Gershoy (1934, p. 13) states briefly that only V. Rafinesquii in the sec. Melanium produces cleisto- gamous flowers. I have observed apparent cleistogamy in the species in northeastern Texas, but rarely. Color varia- tions in roadside colonies suggest, however, that more careful observation might show it to be fairly common. Some colo- nies show considerable variation in flower color, while others — or certain patches within extensive stands — are per- fectly uniform. This is what one might expect if the plants reproduce by both cross- and self-pollination. I have no information about the European members of the section beyond Gershoy's statement. He remarks that he considers V. Rafinesquii to be “functionally biennial” ; as known to me, it is a winter annual. He considers the three forms of V. Kitaibeliana for which chromosome counts were made as annual. 6. CHROMOSOME NUMBER. For V. Kitaibeliana, Gershoy reports haploid numbers of 7, 18, and 24. For V. Rafinesquii he gives 17, but does not state the number of counts or source of material on which this figure was based. At my request, Dr. Walter H. Lewis of Stephen F. Austin State College, Nacogdoches, Texas, counted two plants growing wild on his campus and found the diploid number to be 34, thus confirm- ing Gershoy's count for this species. Additional counts from different localities are needed, both in America and in Europe, before we can be reasonably certain that the same number is not found on both sides of the Atlantic, but the data so far available certainly fall into the same pattern as other lines of evidence. It is concluded that Viola Rafinesquii is a native American plant with its major area in northeastern Texas, Oklahoma, and Arkansas, with outlying scattered stations farther east and in Colorado, that it has increased and spread as a weed since settlement, and that, although very similar to members of the V. Kitaibeliana complex of the Old World, it is prefer- ably treated as a distinct species, the weak morphological 334 Rhodora [Vol. 63 differences being supported by differences in breeding be- havior and chromosome number, as well as by geographic remoteness. I wish to express my very deep appreciation to A. R. Pinto de Silva for his great trouble in comparing specimens, making tracings and microfilms, and supplying transcripts of all the Portuguese literature dealing with the Viola Kitai- beliana group; to Dr. L. G. Sobrinho for his assistance in checking the material at Lisbon; to Dr. Walter H. Lewis for the chromosome counts of V. Rafinesquii; to Dr. U. T. Water- fall for the loan of specimens in the Herbarium of Oklahoma State University ; and to Dr. G. B. Van Schaack, Librarian, Missouri Botanical Garden, for a copy of the original de- scription of V. bicolor Hoffmann. — SOUTHERN METHODIST UNIVERSITY, DALLAS 22, TEXAS. LITERATURE CITED BARTON, WiLLIAM P. C. 1818. Compendium Florae Philadelphicae. Vol. 1. (Viola tricolor, pp. 124-125.) BorviN, BERNARD. 1944. American Thalictra and their Old World allies. 114. T. dasycarpum var. hypoglaucum. Rhodora 46: 482- 483. ELLIOTT, STEPHEN. 1817. A Sketch of the Botany of South-Carolina and Georgia. Vol. 1l. (Viola arvensis, pp. 302-303.) FERNALD, M. L. 1938. Noteworthy plants of southeastern Virginia. Viola Kitaibeliana var. Rafinesquii. Rhodora 40: 443-446. GERSHOY, A. 1934. Studies in North American Violets. III. Chromo- some numbers and species characters. Vermont Agr. Exp. Sta. Bull. 367. Gray, ASA. 1886. Synoptical Flora of North America: the Gamo- petalae. (Helenium tenuifolium, vol. 1 pt. 2 pp. 347-348.) HYLANDER, NILS. 1945. Nomenklatorische and systematische Studien ueber nordische Gefaesspflanzen. Uppsala Univ. Arsskr. 1945-7. (Comments on Gilibert, p. 16.) McVavGH, Rocers. 1949. Questionable validity of names published in Gilibert's floras of Lithuania. Gentes Herbarum 8 (1): 83-90. NUTTALL, THOMAS. 1818. The Genera of North American Plants. Vol. 1. (Viola bicolor, p. 170.) PORTER, T. C. 1880. Viola tricolor, L., var. arvensis, DC. Bot. Gaz. 5: 13-14. PuRsH, FREDERICK. 1813. (“1814”). Flora Americae Septentrionalis. Vol. 1. (Viola bicolor, p. 175.) REVERCHON, J. 1880. Notes on some introduced plants in Dallas county, Texas. Bot. Gaz. 5: 10. 1916] Scamman — Oleandra of Costa Rica 335 ROBINSON, BENJAMIN LINCOLN. 1917. A Monograph of the Genus Brickellia. Mem. Gray Herb. I. (B. grandiflora, pp. 111-113.) DE SCHWEINITZ, LEWIS D. 1822. Attempt of a monograph of the Linnaean genus Viola, comprising all the species hitherto observed in North-America. Amer. Journ. Sci 5: 48-81. (V. bicolor, p. T8.) SHINNERS, LLovp H. 1941. Notes on Wisconsin grasses — II. Muh- lenbergia and Sporobolus. Amer. Midl. Nat. 26: 69-73. (Sporobo- lus neglectus, pp. 70-71; S. vaginiflorus var. inaequalis, p. 71.) 1951. The North Texas species of Mirabilis (Nyctaginaceae). Field & Lab. 19: 173-182. 1958. Spring Flora of the Dallas-Fort Worth Area, Texas. (Viola bicolor, p. 271.) THE GENUS OLEANDRA OF COSTA RICA EDITH SCAMMAN This paper is the third of my studies of a genus of Costa Rican ferns, and follows the same plan as the previous ones on Adiantum! and Pteris.? Again I want to express my gratitude to Dr. Leslie R. Holdridge of San José for his great help during my seasons of collecting in Costa Rica, and to Dr. Rolla M. Tryon of Harvard University for his advice and assistance so gener- ously given in the preparation of these papers. The drawings have been prepared especially for this article by Miss Ruth Hsu. OLEANDRA Cavanilles Oleandra is an isolated genus and is considered to be an old one. It is restricted almost entirely to the Tropics with numerous species in Asia to Polynesia, Africa, West Indies and Central and South America. The fronds are simple and entire, usually lanceolate-elliptical and firm, often lustrous in texture resembling the leaves of the Oleander, whence the name. In some species the rhizome is stout (with appressed scales) and more or less erect, producing a shrubby growth habit. In others it is more slender (with spreading scales) and twines about tree trunks. The stipes are articulate (the portion below the joint being called a phyllopodium) and short or long, solitary or clustered. Veins are free (Fig. 3), 1Contrib. Gray Herb. 187: 3-22. 1960. "Rhodora 63: 194-205. 1961. 336 Rhodora [Vol. 63 and the round dorsal sori are borne irregularly in a row or rows, generally near the costa (Fig. 1), with orbicular or reniform indusia. The scales on the costa of some species are characteristic and serve as an aid to identification. The treatment and key have been adapted from Maxon, Contr. U. S. Nat. Herb. 17: 392-398. 1914. KEY TO THE SPECIES a. Rhizomes slender, the spreading scales not entirely concealing the rhizome (Fig. 2); phyllopodia usually long (0.5-3 em.), slender and naked (except at the base), like the upper portion of the stipe. b. b. Rhizomes brownish with a rather dense covering of persistent scales; leaf-tissue subcoriaceous. ............... eee 1. O. articulata. b. Rhizomes pruinose with fewer, more laxly spreading, deciduous scales; leaf-tissue membrano-papyraceous. ................ 2. O. Bradei. a. Rhizomes stout, pruinose, the appressed scales strongly imbricate, completely concealing the rhizome (Fig. 6); phyllopodia short (rarely to 0.5 cm. or a little more), stout and scaly at least at first, like the rhizome. c. c. Stipes 0.5 to 2 em. long above the articulation, the fronds variable in size and shape, narrowly or abruptly cuneate at base, minutely glandular-pubescent beneath, but soon glabrous; indusia not ciliate, nearly glabrous. ................ eee 3. O. costaricensis. c. Stipes wanting or nearly so above the articulation, never more than 0.5 em. long, the frond tapering downward to a long-attenu- ate, alate base, minutely pubescent; indusia long-ciliate and pilose. IERRRRTEEIREIIS TERRE EEERE EISE ERR ET ER ERIREREREREREREEESSERN ET PERERISREERRTTEERERERYATERE 4. O. decurrens. 1. Oleandra articulata (Swartz) Presl, Tent. Pterid. 78. 1836. FIGS. 1-3. Oleandra nodosa (Willd.) Presl, Tent. Pterid. 78. 1836. Of the four species of Oleandra collected in Costa Rica this is the only one of widespread distribution, occurring in the West Indies and Central and South America; two of the others seem to be endemic to Costa Rica. Climbing up the trunks of trees in the forest, the wide- creeping, brownish rhizomes are densely covered with spreading, ferruginous, linear-filiform scales with a dark- centered peltate base. The fronds, lance-linear to oblong- acuminate, caudate at the apex, are glabrous and shining. Fic. 1-3. Oleandra articulata (from Scamman 7600): Fig. 1. leaf and portion of rhizome, X 4. Fig. 2. portion of rhizome, X "4. Fig. 3. venation, X M. Fic, 4. O. Bradei (from Scamman & Holdridge 7886): lamina, X 4. Fic. 5-6. O, costaricensis 1916] Scamman — Oleandra of Costa Rica 337 (from Scamman 7604): Fig. 5. leaf and portion of rhizome, X %. Fig. 6. portion of rhizome, X 94. Fic. 7. O. decurrens (from Brenes 14240): portion of leaf and rhizome, X 34. Fic. 8. O. panamensis (from Pittier 5322): portion of leaf and rhizome, X 4. Fic. 9. O. guatemalensis (from Tuerckheim 983): portion of leaf and rhizome, X 4. 338 Rhodora [Vol. 63 The costae are elevated beneath, with cordate, deltoid to long triangular, castaneous scales. Guatemala to Panama, to Bolivia and Brazil ; West Indies. In moist forests from sea level to 1200 m. Specimens seen: Cocos Island (Pacific) : Waldo L. Schmitt 132 (US), A. Stewart 239 (GH). HEREDIA: Finca La Selva, Río Puerto Viejo, Scamman & Holdridge 7884 (GH). CARTAGO: San Juan del Norte, Scamman 7600 (GH). 2. Oleandra Bradei Christ, Bull. Soc. Bot. Genève II, 1:231. 1909. FIG. 4. Based on Oleandra nodosa, var. caudata, Christ, Bull. Herb. Boiss. II, 4:964. 1904. Type: Talamanca, Costa Rica, 1000 m., 1898, Pittier 12699. Oleandra Bradei has been collected often in Costa Rica, especially in the region of La Palma. The whitish rhizomes and their fewer, more deciduous, reddish-brown, lanceolate scales distinguish this from the former species. The fronds are generally wider and of thinner texture than those of O. articulata. The often ovate-elongate blade, with undulate margins is variable in size and shape. The costa in the speci- mens seen, has substantially no scales. Known only from Costa Rica. On mossy tree trunks at edge of forests from 300 to 2500 m. Specimens seen: LIMON: Los Diamantes, Rubber Plant Station, Scamman 7041 (GH); Hamburg Finca, on the Río Reventazón below Cairo, Standley & Valerio 48691 (US.). ALAJUELA: Zarcero, Austin Smith 48/194 (US); La Palma de San Ramón, M. Quiros 264 (GH), La Palma near San Ramón, A. Brenes 5113 (US). SAN JOSE: Las Nubes, Jan. 30, 1938, Wm. Knight (us), Scamman & Holdridge 7885 (GH); La Palma, Standley 33066 (GH, US), 38188 (US), Maxon 389,404 (us), H. E. Stork 420 (us), H. Berlolini 608 (us), Scamman 7601 (GH); On the Road to La Hondura, Maxon & Harvey 7904 (GH, US), Scamman & Holdridge 7886 (GH), Standley 37911 (US). CARTAGO: Navarrito, Lankester 750 (US). 3. Oleandra costaricensis Maxon, Contr. U.S. Nat. Herb. 17:397. 1914. FIGs. 5-6. Type: La Palma, Costa Rica, 1459 m., September 1898, A. Tonduz 12551 (US). This species, the most common in Costa Rica, belongs to 1916] Scamman — Oleandra of Costa Rica 339 the second group of Oleandras with the stout, woody rhizome (covered with closely appressed scales) growing more or less erect on forest trees or on bushy banks. The fronds are at first minutely glandular-pubescent beneath, lustrous in tex- ture and often iridescent. The sori are rather large with nearly glabrous indusia. The scales on the costa are charac- teristic, reddish brown, deltoid to ovate-lanceolate, and deep- ly lacerate. Costa Rica, also Colombia. On roadside banks or climbing high on trees in moist forests, the stems sometimes supported by other plants, from 600 to 1700 m. Specimens seen: 1901-1905, Wercklé (US); Nov. 1886, J. J. Cooper (GH, US). LIMON: On hills near Moravia, Williams 16181 (US). HEREDIA: Cinchona, Scamman 7604 (GH). SAN JOSE: La Palma, Ton- duz 12550 (us), Standley 38229 (US), Scamman 7602 (GH); La Hon- dura, Standley 36210 (US); Vicinity of El General, Skutch 2817, 2961 (GH, US). CARTAGO: Orosi, finca del Dr. Valverde, A. Brade 16836 (us), Vicinity of Orosi, Standley 39611 (US); Tapanti (Orosi), R. Torres 185 (US), Scamman & Holdridge 7887 (GĦ); Juan Viñas, Cook & Doyle 222 (us), Holdridge 7603 (GĦ); La Estrella, Standley 39259 (GH, US), 29393 (Us); East Turrialba, Aug. 9, 1924, A. Alfaro (GH, vs); El Muñeco, on Río Navarro, S. & R. Torres 50991 (Us); La Fortuna entre Cervantes et Pacayas, March 1906, Biolley (US); Pejivalle, Stork 2819 (US); San Isidro de Cartago, Stork 4529 (US). GUANACASTE: La Tajana, north of Tilarán, Standley & Valerio 45928 (us), El Silencio near Tilarán, Standley & Valerio 44702 (US). 4. Oleandra decurrens Maxon, Contr. U. S. Nat. Herb. 17:396. 1914. FIG. 7. Type: El General, Costa Rica, January 1897, Pittier 10649 (US). This rare species has the stipe almost lacking, never more than 0.5 em. long, the linear-oblanceolate, pubescent fronds tapering gradually downward to a slender, long-attenuate, alate base. The indusia are small and long-ciliate, and the many scales on the costa are linear-lanceolate, long-attenuate and bright brown. The leaf-tissue is rigidly herbaceous. Known only from Costa Rica. From 1100 to 1500 m. Specimens seen: ALAJUELA: Des collines de Piedades prés San Ramón, A. Brenes 14240 (GH, US). SAN JOSE: El General, Pittier 10649 (US). One variety described from Costa Rica has not been 340 Rhodora [Vol. 63 placed. Although described under O. trinitensis it is prob- ably O. costaricensis. O. trinitensis Maxon, var. subcostaricensis Suesseng. & Losch, Mitteil. Bot. Staats. München 1: 23. 1950. Type: Turrialba, Costa Rica, 1500 m., June 18, 1932, Kupper 1612, not seen. Two other species, Oleandra guatemalensis Maxon and Oleandra panamensis Maxon grow in Guatemala and in Panama to Colombia respectively. These might possibly occur in Costa Rica, although no specimens of them have been seen from there. O. panamensis (Fig. 8) differs from the related O. decur- rens in having the lamina base rather abrupt and the leaf distinctly stipitate above the articulation. In O. decurrens the lamina base is gradually reduced and the leaf is subses- sile or nearly so above the articulation. O. guatemalensis (Fig. 9) has long slender phyllopodia like the upper portion of the stipe and small distinctive heart-shaped, almost concolorous scales at the sides of the costa. It differs in these characters from O. costaricensis which has the phyllopodium like the rhizome and more elong- ate, deeply lacerate costal scales. — GRAY HERBARIUM, HAR- VARD UNIVERSITY. VASCULAR PLANTS ON THE CINDER CONE OF PARICUTIN VOLCANO IN 1960! JoHN H. BEAMAN Plants growing on the cinder cone of Parícutin Volcano in 1958 have been reported previously (Beaman, 1960). The present study is based on collections made there on Septem- ber 8, 1960, two years after the first visit and approximately eight and a half years after volcanic activity ceased. Voucher specimens are filed in the Herbarium of Michigan State Uni- versity. I am indebted to Dr. R. L. Hauke, Dr. R. MeVaugh, Dr. T. R. Soderstrom, and Dr. W. H. Wagner, Jr. for help in the determination of certain species as indicated below. Ascent and descent of the cone were made on the south- west side. The entire circumference of the rim and both ‘Supported by Grant G-9045 from the National Science Foundation. 1961] Beaman — Plants on Paricutin 341 crater vents were examined. The physical features of the cone were about the same in 1958 and 1960. A description and illustrations of the cone, based on a 1957 visit, have been published by Segerstrom (1960). EQUISETACEAE. EQUISETUM HYEMALE L. var. AFFINE A. A. Eat. (Beaman 4417). The determination of this species was made by Dr. Hauke. Several stems were present on the northwest rim. These were relatively small, and strobili had not been produced. POLYPODIACEAE. PELLAEA TERNIFOLIA (Cav.) Link var. TERNIFOLIA (Beaman 4414). Only one small plant, growing on the northeast wall of the crater, was found. This is the area where most of the ferns were collected during the 1958 visit, but not where P. ternifolia was previously located. The 1960 collection was sterile, but one of the plants obtained in 1958 had produced sporangia. PITYROGRAMMA TARTAREA (Cav.) Maxon (Beaman 4415, 4423, 4431). This was one of the most frequently observed species on the cone. The determination of collection 4415, which includes only very young plants, was suggested by Dr. Wagner. The previously reported un- determined juvenile fern (Beaman 2408D) may also be this species. Both of these collections are from the northeast wall of the crater. Collection 4423 from the northwest rim of the cone includes plants with sporangia. Collection 4431, from near a fumarolic vent on a small lava outerop at the southwest base of the cone, is of plants at about the same developmental stage as those in collection 4423. GRAMINEAE. DIGITARIA SANGUINALIS (L.) Scop. (Beaman 4412). Several plants were found in flower on the northeast saddle. In 1958 this area was devoid of plants except for one dicot seedling. CALAMAGROSTIS MCVAUGHII Sohns (Beaman 4426, 4429). The deter- mination of collection 4429 was made by Dr. Soderstrom. In 1958 this species (previously called C. pringlei) was found on a lava flow near the east base of the cone. Plants with young inflorescences were well established in 1960 on the northwest rim (4426) and at the southwest base near a fumarolic vent on a small lava outcrop (4429). SPOROBOLUS CONFUSUS (Fourn.) Vasey (Beaman 4410). Flowering and fruiting plants of this annual species were abundant near the east peak in a local area which was kept moist by a fumarolie vent. SALICACEAE. SALIX HARTWEGII Benth. (Beaman 4423). This collection was made at the southwest base of the cone near a fumarolic vent on a small lava outcrop. Several plants were also observed on the rim of the cone. Eggler (1959) noted that S. hartwegii was one of only two species present in 1950 in volcanic ash on the adjacent north slopes of Cerro Tancítaro. 342 Rhodora [Vol. 63 PHYTOLACCACEAE. PHYTOLACCA ICOSANDRA L. (Beaman 4424). Several fruiting speci- mens were found near the west peak. This collection makes it possible to determine the previously reported undetermined dicot seedling ( Beaman 2404) as the same species. ERICACEAE. GAULTHERIA ODORATA Willd. (Beaman 4408). A single plant, about 22 em. high and with one inflorescence, was found near the east peak. Presumably this shrub had attained its full development in two years, because it was not observed during the 1958 visit. HYDROPHYLLACEAE. WIGANDIA KUNTHII Choisy (Beaman 4420). The identity of this collection was suggested by Dr. McVaugh. Three seedlings were found cn the northwest rim of the cone. COMPOSITAE. ASTER SUBULATUS Michx. var. AUSTRALIS (A. Gray) Shinners (Bea- man 4411, 4438). Since 1958 this species (previously called A. exilis) has become well established in the area near the east peak. No flower- ing plants were found in 1958, but in 1960 many of the plants were in flower. One tiny seedling (collection 4428, probably of this species) was found on the southwest side of the cone about 40 m. below the rim. BACCHARIS GLUTINOSA Pers. (Beaman 4428). A fairly large, sterile plant was found near a fumarolic vent on a small lava outcrop at the southwest base of the cone. Large flowering specimens (Beaman 4392) were collected about three kilometers east of the cone, where the plants were growing in voleanic ash. CONYZA CANADENSIS L. (Beaman 4413). One small flowering speci- men was collected in the northeast saddle on the rim of the cone. CONYZA CORONOPIFOLIA H. B. K. (Beaman 4406). This species was obtained from near the east peak where it was also collected in 1958. The one plant was in flower and of about average size for the species. ERIGERON EXILIS A. Gray in S. Wats. (Beaman 4422). One depauper- ate, but flowering, specimen was found on the northwest rim of the cone, EUPATORIUM PAZCUARENSE H. B. K. ? (Beaman 4427, 4432, 443? ). This species was found at three places on the cone. Two plants were located on the northwest rim (4427); several plants were found at the scuthwest base on a small lava outerop (4432); and a small seedling was obtained on the southwest side about 40 m. below the rim (4437). GNAPHALIUM AMERICANUM Mill. (Beaman 4407, 4430). Two plants of collection 4407, both in flower, were obtained near the east peak. A larger population (4430) was found at the southwest base near a fumarolie vent on a small lava outcrop. GNAPHALIUM SEMIAMPLEXICAULE DC. (Beaman 4425, 4436). One small, sterile plant (4425) was located near the west peak. Two other plants (4436), in poor condition, were present on the southwest side of the cone about 40 m. below the rim. Specimens of this species collected in 1958 were more mature and in better condition that those found in 1960. 1961] Beaman — Plants on Paricutin 343 HETEROTHECA INULOIDES Cass. var. ROSEI Wagenknecht (Beaman 4416). One vigorous rosette was found on the northwest rim of the cone. A flowering specimen (Beaman 4404) was collected near San Juan Viejo (formerly San Juan Parangaricutiro) in soil with a high volcanic ash content. SONCHUS OLERACEUS L. (Beaman 4421). A very small seedling was found on the northwest rim of the cone, but a flowering specimen ( Beaman 4389) was collected on a volcanic ash deposit near the south- west base of the cone. UNDETERMINED MATERIAL. Two additional species (Beaman 4418 and 4419) were obtained on the northwest rim of the cone. Collection 4418 consists of one rosette, and 4419 is of two small, caulescent plants. Both may be species of the Compositae. DISCUSSION The vegetation on the cone of Parícutin Volcano was much denser and more species were present in 1960 than in 1958. In 1960 the cone flora was made up of 22 species in eight families and 20 genera, while 14 species in six families and 11 genera were present in 1958. Six species (Pityrogramma calomelanos, Pteridium aquilinum var. feei, Pinus monte- zumae ?, Aegopogon cenchroides, Buddleia cordata, and Gnaphalium attenuatum) reported from the 1958 visit were not relocated in 1960. The number of species of ferns de- creased from five to two. The Gramineae increased from one to three species, although the one species found in 1958 was not recollected in 1960. The Compositae, with 10 species in 1960, had doubled its representation in the two-year period. The two woody species (Pinus montezumae ? and Buddleia cordata) found in 1958 were not relocated in 1960, but they were replaced by three other woody species (Salix hartwegii, Gaultheria odorata, and Wigandia kunthii). In 1958 the moisture from fumarolic vents appeared to be very important in enabling plants to become established on the cone. At that time a majority of the species were con- centrated near such areas. In 1960 some of the most vigor- ous growth was near fumarolic vents around the east peak and at the southwest base. Many plants, however, were also found on the rim of the cone in areas away from fumarolic vapors. Coarse rubble on the outside of the cone and on the crater walls apparently is still not a habitat which vascular plants have been able to successfully invade. 344 Rhodora [Vol. 63 In view of the species present in 1960, it still appears that dispersal by wind is the most important method of transport of propagules to the cone. Nevertheless, animal agencies, possibly birds, were very likely involved in the transport of Phytolacca icosandra and Gaultheria odorata, species with fleshy fruits. Animals may also be responsible for the dis- persal of the grasses, particularly of Digitaria sanguinalis, which produces relatively large caryopses. The Equisetum, Polypodiaceae, Salix, Wigandia, and Compositae probably were all dispersed by wind. Nearly half of the species found on the cone are weedy. These are Digitaria sanguinalis, Phytolacca icosandra, Wigandia kunthii, Aster subulatus var. australis, Conyza canadensis, C. coronopifolia, Gnaphalium americanum, Heterotheca inuloides var. rosei, and Sonchus oleraceus. It is not surprising that species which are well adapted for colonizing other disturbed and open habitats should be among the first plants to become established on the cone. The species most vigorous in appearance and numerous in individuals were Pityrogramma tartarea, Calamagrostis mcvaughii, Digitaria sanguinalis, Sporobolus confusus, Salix hartwegii, Phytolacca icosandra, Aster subulatus var. aus- tralis, and Gnaphalium americanum. All of these species except Salix hartwegii had produced reproductive structures. Seven of the species which were present on the cone in 1958 were also found in 1960, thus demonstrating that these pio- neer species were able to maintain themselves on the cone during the two-year period. Four species, Pityrogramma tartarea, Phytolacca icosandra, Aster subulatus var. austra- lis, and Eupatorium pazcuarense ? have become more abun- dant since 1958. — DEPARTMENT OF BOTANY AND PLANT PATHOLOGY, MICHIGAN STATE UNIVERSITY, EAST LANSING. LITERATURE CITED BEAMAN, J. H. 1960. "Vascular plants on the cinder cone of Parícutin Volcano in 1958. Rhodora 62: 175-186. EGGLER, W. A. 1959. Manner of invasion of volcanic deposits by plants, with further evidence from Parícutin and Jorullo. Ecologi- cal Monographs 29: 267-284. SEGERSTROM, K. 1960. Erosion and related phenomena at Parícutin in 1957. Geol. Surv. Bull. 1104-A. 1-18. 1961] Rollins — A Weedy Crucifer 345 A WEEDY CRUCIFER AGAIN REACHES NORTH AMERICA. — The identification of a roadside weed from Yancey County, North Carolina was not difficult, but determining the correct name for the plant turned up several conflicts and proved to be a small “tour de force" through the literature. The plant, originally described as Brassica cheiranthos by Vil- lars,! is apparently native in western, central and southern Europe. If the species were to be retained in Brassica, as was done for many years, there would be no nomenclatural problem. The plant would be referred to as Brassica. cheir- anthos Vil. However, most of the recent works dealing with this species place it in another genus. O. E. Schulz? not only revived the invalidly published generic name Brassicella of Fourreau? but in addition appears to have misapplied the name Brassica erucastrum L. to the species here considered. At least, this is essentially the conclusion of Dandy*, who transferred Brassica cheiranthos Vill. to Rhyncosinapis to replace the name R. erucastrum (Vill. Dandy earlier pro- vided for the same taxon.^ I do not know whether Dandy went over the same literature and specimens that were examined by Pugsley* but he reached the same conclusion, that the names Brassica cheiranthos and B. erucastrum do not apply to the same species as was assumed to be the case by Schulz. Pugsley points out that "the sheet in the Linnean Herbarium labelled “7. erucastrum?” by Linnaeus, contains two small plants of Rhaphanus Raphanistrum L. that look much more like Erucastrum obtusangulum than Brassica cheiranthus." Assuming for the moment that the species is not to be retained in Brassica and is not properly referrable to Sinapis or any other of the long established generic segregates of Brassica, we come back to the two divergent treatments of Schulz and Dandy. It is clear that Brassicella, published merely as a nomen in Fourreau's work, has to be dated from Schulz and was illegitimate when proposed because Rhyn- 1Prosp. Pl. Delph. 40. 1779. ?Engler's Bot. Jahrb. 54: Beibl. No. 119. 52-3. 1916. 3Ann. Soc. Linn. Lyon 16: 330. 1868. *Watsonia 4: 41-42. 1957. 5Clapham, Tutin and Warburg, Fl. Brit. Isles 158. 1952. $Journ. Bot. 74: 326. 1936. 346 Rhodora [Vol. 63 cosinapis had been validly published by von Hayek’ in the meantime. Schulz was wrong both as to the generic and to the specific names. We agree then, with Dandy, that the species is to be referred to by the name Rhyncosinapis cheiranthos (Vill.) Dandy. The specimen that evoked the above was collected by Harry E. Ahles, No. 42855, with J. A. Duke, June 7, 1958, roadside, 1.8 miles southwest of Burnsville on N.C. 197, Yancey County, North Carolina. It was distributed as Dip- lotaxis muralis. This is not the first time Rhyncosinapis cheiranthos has reached the North American continent. Nearly one hundred years ago when Addison Brown was combing the ballast fillings around New York City for new introductions he found it in Hoboken, New Jersey.’ I have examined Brown’s specimens and they are correctly identi- fied. At first, he used the name Brassica monensis, but later made the correction to B. cheiranthos. This species appar- ently did not gain a continuing foothold in our flora at that time and we have seen no evidence of its presence from then until that provided by the North Carolina collection cited above. The fact that this came from a roadside habitat far inland from a coastal port indicates that the species may well be on its way to becoming a part of our weedy flora. — REED C. ROLLINS, GRAY HERBARIUM OF HARVARD UNIVERSITY. AN ALBINO FRUITED FORM OF GAULTHERIA PROCUMBENS On September 11, 1958, I stopped my car in a temporary parking place opposite the Naval Facilities Station at Tom Nevers Head, Nantucket Island, Mass. As I left the car, I noticed that the front wheels were on a luxuriant bed of Gaultheria procumbens L. Intrigued by the robust appear- ance of the plants, I made a closer inspection of them. To my surprise, I discovered that the fruits were a creamy white with many of them being almost pure white. Outside the perimeter of this bed of albino-fruited plants, there was a far greater area matted solidly with the usual red-fruited form. The fruits, flowers, and foliage of these plants were considerably smaller by comparison. Several TBeih. Bot. Centralbl. 27: 260. 1911. 5Bull. Torrey Bot. Club 7: 123. 1880. 1961] MacKeever — Albino Fruited Gaultheria 347 finger-like drifts of the more robust albino-fruited plants extended well into the area of the red-fruited plants showing that soil conditions played no major part in the dwarf condi- tion of the latter. Gaultheria procumbens L., forma leucocarpa MacKeever. Differt ab planta typica in fructibus albis. This form is similar to the species ex- cept for its albino fruits. The type specimen, N309, was collected by Frank C. MacKeever at Tom Nevers Head, Nantucket Island, Mass., August 12, 1959, and is deposited in the herbarium of the New York Botanical Garden. It may be of interest to write something of the history of this plant. My first encounter with it took place on my first visit to the island and I had then no intentions of doing a floristic study of the plant-life there. For that reason I collected only enough material to serve as voucher specimens, the two specimens being given the reassigned collection num- ber N1 and deposited in the herbarium of the New York Botanical Garden, and in the herbarium of the Museum of Natural Science, Nantucket, Mass. Upon realizing that this plant was a new form, I wished I had more sets of the original collection to present to other institutions. Therefore, on August 12, 1959, I recollected it in order that a set of this number, N309, could be presented as an isotype to Gray Herbarium, Cambridge, Mass.; De- partment of Agriculture, Plant Research Institute, Ottawa, Ont., Canada ; and Instituto de Botanico, Sao Paulo, Brazil. The sets of N309, now deposited in the above mentioned institutions, were collected from the same group of plants as those of N1. However, the fruits of N1 are somewhat more mature than those of N309, as N1 was collected later in the season. On September 21, 1960, I took the Nantucket Garden Club for “a nature hike". We visited the station of f. leucocarpa only to find great piles of dirt and debris spread over the colony. A few albino-fruited plants were found, but the fruits possessed somewhat of “a blush" on the more exposed surfaces. These “blushed” fruits were not as white as those of the original collections. However, it does seem certain that the albino fruited plants were not completely destroyed. — FRANK C. MACKEEVER, NEW YORK BOTANICAL GARDEN. 348 Rhodora [Vol. 63 CORRECTIONS IN REPORTS OF THE COMMITTEE ON PLANT DISTRIBUTION In the Fourteenth Report (RHODORA 63: 47-55) the fol- lowing corrections resulting from what we now consider to be misidentifications should be noted : — Arabis canadensis L. Delete the + sign in the column for Maine. The reported occurrence of this species in Maine was based on the misidentification of a depauperate speci- men of Hesperis matronalis L. in young fruit collected on July 3, 1901, by Kate Furbish at Skowhegan in Somerset County. The nearest and otherwise most northeasterly sta- tion for A. canadensis so far discovered is some one hundred fifty miles to the southwest in Rockingham County, New Hampshire. Ribes rotundifolium Michx. Delete the + sign in the New Hampshire column. A specimen from Crawford Notch in the White Mountains, well north of the normal range for the species, collected by H. H. York on July 18, 1919, labeled R. rotundifolium, proves to be Ribes hirtellum Michx. Ribes cynosbati L. (Page 53, Line 17 et seq.) Delete the second sentence and substitute: It occurs sparingly in west- ern Maine south of 45? and is frequent elsewhere in New England west of a line drawn from Strong in Franklin County, Maine, through Jaffrey in Cheshire County, New Hampshire, Hartland in Hartford County, Connecticut, to New Milford in Litchfield County, this line forming a well- marked boundary with the exception of a single station in southeastern New Hampshire and two stations in north-cen- tral Connecticut. Three errors in earlier Reports have been pointed out by Hodgdon and Steele'. Accordingly, the following plants should be excluded from the New Hampshire list: Juniperus horizontalis Moench. (Fifth Report, RHODORA 28: 43). Car- ya tomentosa Nutt. (Eleventh Report, RHODORA 58: 127). Betula papyrifera Marsh. var. commutata. (Regel.) Fern. (Eleventh Report, RHODORA 58: 128). — R. C. BEAN, A. F. HILL AND R. J. EATON, COMMITTEE ON PLANT DISTRIBUTION. 'HopnGDoN, ALBION R., AND FREDERICK L. STEELE — The Woody Plants of New Hampshire. S. D. 447. Agr. Exp. Sta. U. of N. H., Durham, N. H., 1958. 1961] Tryon — A Botanical Atlas 349 A BOTANICAL ATLAS OF THE DANISH FLORA! — This fine illustrated work contains figures, drawn from living materi- al, of most of the species of Bryophyta and vascular plants of the Danish flora. The first volume, with 515 plates, is devoted to the Angiosperms and the second one, with 240 plates, to the Bryophyta, Pteridophyta and Gymnosperms. One to several species are illustrated on each plate. The whole plant, or a portion of it, is presented and in addition some of its more significant details. The second volume has a parallel English text and especial- ly in this volume there is an introductory description and discussion of the larger families or groups. This text gives a valuable account of the life-history of the group and the seasonal development of the different structures. The ecolo- gy of the group is discussed and closely correlated with its life-history. The phyletie theory presented at the end of the second volume is somewhat novel in presenting the Angiosperms as triphyletic. Each of the three lines is considered to have evolved from separate groups of the Bryophyta through groups of the lower vascular plants. This phylogeny is developed chiefly through a concept of the conifer cone and the Equisetum and Selaginella strobilus as primitive kinds of "flowers". While this theory will be of interest, its mor- phological basis may not be acceptable to many. These volumes are not only useful as an iconography of the principal elements of the Danish flora, but also for in- formation about the smaller structures of the species (sta- mens, ovaries, seed, peristome, spores, etc.) and about their ecological life-history as well. — ROLLA TRYON, GRAY HER- BARIUM, HARVARD UNIVERSITY. 1Botanisk Atlas, Danmarks Daekfreede Planter, by Olaf Hagerup and Vagn Peters- son. vol. 1. pp. 1-550. 1956. vol. 2, pp. 1-299. 1960. 4°. Ejnar Munksgaard. Kobenhaven. 350 Rhodora [Vol. 63 INDEX TO VOLUME 63 New scientific names and combinations are printed in bold face type Abies balsamea 65, 288 Acalypha virginica 288 Acer pensylvanicum 66, 68, 288; pseudoplatanus 150; rubrum 65, 66, 67, 68, 69, 284, 288; saccharum 65, 295; spicatum 66, 288, 295 Achillea lanulosa 125, 129 Actinastrum (? hantzchii) 278 Adams, P. and Godfrey, R. K., Observations on the Sagittaria subulata Complex 247 Adams, W. P. and Robson, N. K. B. A Re-evaluation of the Generic Status of Ascyrum and Crookea (Guttiferae) 10 Adiantum alarconianum 87; bra- siliense 71; concinnum 87; curvatum 71; Henslovianum 87; humile 70, 71; Killipii 71; lobatum 71, 72, Plate 1255; Poiretii var. hirsutum 87; terminatum 70 Adlumia fungosa 47, 52, 288 Aegilops cylindrica 130 Aeschynomene elegans 18 Aegopogon cenchroides 343 Ageratum corymbosum 122, 126 Agrimonia gryposepala 43 Agropyron dasystachyum 132; repens 130, 131; smithii 130, 131; var. molle 131, 132; var. smithii 130; palustris 137 Albino Fruited Form of Gaul- theria procumbens, An 346 Aletris farinosa 304. 305; lutea 304, 305, 206; f. albiflora 305; obovata 305, 306; Aletris x Tottenii 306, 307 Alisma triviale 288 Alliaria officinalis 47, 55 Allium ampeloprasum 310; var. atroviolaceum 310; mutabile 20 Alnus rugosa 66, 297; var. amer- icana 297, 298, 299, 300, 302, 303; var. rugosa 297, 298, 302, 303; serrulata 297, 298, 300, 302, 303; f. emarginata 303; f. nanella 303; var. subelliptica 297, 302, 303 Althaea officinalis 43 Alyssum Alyssoides 47, 54; sax- atile 47, 55 Amaranthus powellii 180; retro- flexus 288 Ambrosia artemisiifolia var. ela- tior 289 Amelanchier laevis 66 Amsinckia lunaris 230; retrorsa 44 Anabaena 281 Andrews, H., Two New Orchid Records for Ontario 175 Andropogon scoparius 132; vir- ginicus 132 Ankistrodesmus 277 Anonymos pudica 36, 37, 38; var. nitida 37 Aphanochaete repens 276 Aphanothece castagnei 280 Apocynum androsaemifolium 44; medium 44; sibericum 44 Arabidopsis Thaliana 48, 54 Arabis canadensis 47, 52, 348; divaricarpa 47, 53; Drummon- dii 47, 52, glabra 48, 52; hir- suta var. pyenocarpa 48, 52; laevigata 48, 53, 146; lyrata 48, 53; missouriensis 48, 53; procurrens 48, 55 Arctium lappa 45; tomentosum 45, 289 Argemone alba 47, 55; mexicana 47, 55 Arisaema atrorubens 295; pusilla 20 Aristolochia nashii 20 1961] Armoracia aquatica 48, 58; lapathifolia 48, 54 Arrhenatherum elatius 132 Artemisia cana 41 Arthrodesmus 279 Asarum canadense 295; virgini- cum 206 Asclepias lanuginosa 44; sulli- vantii 44 Ascyrum and Crookea (Gutti- ferae), A Re-evaluation of the Generic Status of 10 Ascyrum edisonianum 15; Hy- pericoides 15; microsepalum 14, 15; multicaule 18, 15; pumilum 12, 13, 15; spathula- tum 15; stans 13, 15; tetrape- talum 13, 15 Aspidium funestum 34; invisum 34 Asplenium bradleyi 20; monta- num 46 Aster Chasei 20; exilis 342; lae- vis 331; var laevis 331; var. Geyeri 331; subulatus var. aus- tralis 342, 344; turbinellus 146 Astilbe japonica 50, 55 Astragalus thompsoniae 180 Astranthium guatemalense 122, 126; integrifolium 122; mexi- canum 122, 126; purpurascens 126; xanthocomoides 122, 126 Atriplex glabriuscula 42; patula var. hastata 288 Avena fatua 132 Baccharis glutinosa 342 Bahia absinthifolia 128; xylopo- da 123, 128 Baileya pleniradiata 128 Baker, H. G., Heterostyly and Homostyly in Lithospermum canescens (Boraginaceae) 229 Baldwin, H. I., Further Notes on Chamaecyparis in New Hamp- shire 281 Barbarea orthoceras 48, 52; verna 48, 54; vulgaris 48, 54; Index to Volume 63 351 var. arcuata 48, 54; var. bra- chycarpa 48, 55 Barnes, B. V., Hybrid Aspens in the Lower Peninsula of Michi- gan 311 Batrachospermum vagum 281 Bauhinia rufa 18 Beaman, J. H., Vascular Plants on the Cinder Cone of Paricu- tin Volcano in 1960 340 Beaman, J. H., Rock, H. F. L., and Turner, B. L., Chromo- some Numbers in the Compo- sitae. V. Mexican and Guate- malan Species 121 Bean, R. C. Hil, A F, and Eaton, R. J., Corrections in Reports of the Committee on Plant Distribution 348; Four- teenth Report of the Commit- tee on Plant Distribution 47 Bellis perennis 45 Berteroa incana 48, 54; mutabilis 48, 55 Betula lenta 65, 67, 68; lutea 65, 67, 68, 69, 288; papyrifera 65, 68; var. commutata 348; popu- lifolia 65; pumila var. glandu- lifera 42; X Betula sandbergii 42 Bidens angustissima 122; var. linifolia 124, 127; anthemoides 123, 127; connata var. gracili- pes 145; 289; frondosa 289 Bidens connata var. gracilipes Fern. in Western Michigan 145 Bifora americana 328, 330 Binuclearia tatrana 276 Botanical Atlas of the Danish Flora, A (Review) 349 Botrychium dissectum 164, 165, Plate 1260, 168, 169, Plate 1261, 171, 499. 178, 174: f. dissectum 165; f. obliquum 165; lunaria 40; multifidum 164, 165, 168, 172; obliquum 165; oneidense 164, 165, Plate 352 Rhodora 1259, 168, 169, Plate 1261, 171, 172, 173, 174; ternatum 164, 165, 168, 172; virginianum 295 Brachyelytrum erectum 132 Brassica cheiranthos 345, 346; erucastrum 345; hirta 48, 54; juncea 48, 54; var. crispifolia 48, b5; Kaber var. pinnatifida 48, 54; var. Schkuhriana 48, 54; monensis 346; Napus 48, 54; nigra 48, 54; oleracea 48, 54; Rapa 48, 54 Braya humilis var. leiocarpa 48, 53 Brazilian Leguminosae, Chromo- some Numbers of Some 16 Brickellia grandiflora 331 Bromus brizaeformis 41; com- mutatus 133; inermis 133; japonicus 133; latiglumis 133; marginatus 40; nottowayanus 21; racemosus 133; secalinus 133; squarrosus 134 Browne, E. T. Jr., Some New or Otherwise Interesting Re- ports of Liliaceae from the Southeastern States 304 Buddleia cordata 343 Bulbochaete scrobiculata 276 Burk, C. J., Environmental Vari- ation in Heterotheca subaxil- laris 243 Caesalpinia ferrea 18; spinosa 18 Cakile edentula 48, 53, 288 Calamagrostis mcvaughii 341, 344; pringlei 341 Calla palustris 42 Calochortus pudica 42 Calothrix epiphytica 281 Camelina microcarpa 48, 54; sativa 48, 54 Campanular Persistence 29 Campanula divaricata 29 Campsis radicans 289 Camptosema tomentosum 17, 18 Capsella Bursa-pastoris 48, 54; [Vol. 63 var. bifida 48, 55; rubella 48, 55 Carragana arborescens 43 Cardamine bellidifolia 48, 52; bulbosa 48, 53; Douglassii 48, 53; flexuosa 48, 53; hirsuta 48, 55; impatiens 48, 55; Longii 48, 53; parviflora var arenicola 48, 52; pensylvanica 48, 51; var. Brittoniana 48, 52; pra- tensis 48, 54; f. plena 48; var. palustris 48, 53 Cardaria Draba 48, 55 Carex aenea 41; atherodes 20; austrina 20; caroliniana 20; debilis 20; decomposita 20; digitalis 20; emmonsii 21; foenea 41; garberi 41; halea- na 41; hallii 41; molesta 41; oxylepis 20; parryana 41; peckii 41; physorhyncha 20; prairea 41; richardsonii 41; squarrosa 146; stipata var. oklahomensis 146; striatula 20; styloflexa 20; substricta 42; swanii 20; texensis 20; torreyi 42; torta 20 Carpinus caroliniana 65 Carya illinoensis 296; illinoinen- sis 296; tomentosa 348 Castanea dentata 65 Centaurea maculosa 45 Centrosema coriaceum 17, 18, 19 Cephalanthus occidentalis 30 Cerastium arvense 43; scopulor- um 43; vulgatum 43 Chaetopappa bellioides 127 Chaetosphaeridium globosum 276 Chamaecyparis thyoides 65, 67, 69, 281 Chamaedaphne calyculata 284 Champlin, R. L., A Fern New to Rhode Island 46 Characium ambiguum 277; nae- gelii 277; stipitatum 277 Check List of Walter Deane’s Seedling Collection, A 287 1961] Cheilanthes fractifera 74; mar- ginata 75; notholaenoides 87; Orbignyana Plate 1255, 74, 75; Poeppigiana 75; rufopunctata 75; Saundersii 74 Chelidonium majus 47, 54 Chenopodium album 288 Chicago Region. II. Notes on the Grass Flora of 130 Chloris petraea 177 Chorispora tenella 48, 48, 55 Chromosome Numbers in the Compositae. V. Mexican and Guatemalan Species 121 Chromosome Numbers of Some Brazilian Leguminosae 16 Chromosomes of Podophyllum peltatum, The 267 Chrysanthemum humile 157; leucanthemum var. pinnatifi- dum 149 Chrysosplenium americanum 50, 52 Cimicifuga cordifolia 21 Cinna arundinacea 136; latifolia 136 Circaea alpina 289 Cleome serrulata 47, 55; spinosa 47, 55 Closterium 278 Cnemidaria grandifolia 32 Coelastrum cambricum 277; mi- croporum 277; sphaericum 277 Coleochaete scutata 276 Coleosphaerium collinsii 280 Commelina, A Variegated Foli- age Form of 88 Commelina communis var. lu- dens 88; f. aureostriata 89 Conringia orientalis 48, 54 Contemporary Notes on Powell’s Expeditions in the West (Re- view) 179 Conyza canadensis 342, 344; cor- onopifolia 342, 344 Cooperia Drummondii 216; Jonesii 216; Traubii 216 Copaifera langsdorfii 18 Index to Volume 63 353 Coreopsis rosea 236 Cornus florida 66, 67, 146 Coronopus didymus 48, 55; pro- cumbens 48, 55 Corrections in Reports of the Committee on Plant Distribu- tion 348 Corydalis aurea 47, 53; flavula 47, 53; sempervirens 47, 52 Cosmarium 278, 279 Cosmos diversifolius 127 Costa Rica, The Genus Oleandra of 335; The Genus Pteris of 194 Costaricia Werckleana 75, 76, Plate 1256, 78 Crataegus collina 20 Crookea (Guttiferae), A Re- evaluation of the Generic Status of Ascyrum and 10 Crookea microsepala 15 Crotalaria stipularia 18; striata 18 Cryptantha macounii 44 Ctenitis excelsa 34; protensa 34; var. funesta 34 Cuscuta glomerata 44 Cyathea caribaea 32; grandifolia 32; hodgeana 31; imrayana var. caribaea 32 Cylindrocapsa geminella var. minor 276 Cylindrocystis americana 278; brebissonii 278 Cyperus grayioides 20 Cystopteris Baenitzii 181; bulbi- fera 185, 192; dickieana 40, 181, 187; Dickieana 182; fragi- lis 181, 182, 183, 185, 189; var. fragilis 186, 192; var. lauren- tiana 186, 188, 192; var. mac- kayii 186, 188, 192; var. protrusa 186, 188, 192; var. simulans 186, 192; var, tennes- seensis 186, 192; montana 185, 192 Danish Flora, A Botanical Atlas 354 of the (Review) 349 Danthonia spicata 41 Daoud, H. S. and Wilbur, R. L., The Genus Lechea (Cistaceae) in the Southeastern United States 103 Datura stramonium 149 Daucus carota 149 Davidson, P. E. and Davidson, R. A., Variance in Herbarium Specimen Identification and Other Considerations Based Upon the Preparation of a Local Flora 207 Deane's Seedling Collection, A Check List of Walter 287 Decade of Botanizing in Illinois, A 19 Delphinium ajacis 43 Dennstaedtia ampla 76; arbores- cens 78, 80; distenta 78; di- varicata 78; Elmeri 76; erosa 18; glabrata 76; grandifrons 18; Munchii 78; obtusifolia 78, 80; Orbignyana 80; punctilo- bula 149; scabra 76; vagans 80 Dentaria anomala 49, 53, 54; di- phylla 49, 52; incisifolia 49, 53; laciniata 49, 53; maxima 49, 53 Descurainia pinnata var. bra- chycarpa 49, 53, 54; Richard- sonii 49, 53, 54; Sophia 49, 54 Desmidium 279 Deutzia scabra 50, 54 Diatoma anceps 280 Dicentra canadensis 47, 53; Cu- cullaria 47, 52, 295; eximia 47, 55; formosa 47, 55; spectabilis 47, 55 Dicliptera brachiata 20 Dicranoglossum desvauxii 35 Dictyosphaerium pulchellum 277 Digitaria sanguinalis 341, 344 Dimorphococcus lunatus 277 Diodia teres 226 Diplachne fascicularis 134, 136 Rhodora [Vol. 63 Diplotaxis muralis 49, 55, 346; tenuifolia 49, 55 Disjunct Community of Chestnut Oak in Mississippi, A 324 Draba Allenii 49; aprica 223, 224, 225; arabisans 49, 53; brachycarpa 224, 225; var. fastigiata 224; f fastigiata 223; cuneifolia 20; glabella 49, 53; lanceolata 49, 53; reptans 49, 53; verna 49, 54; var. Boerhaavii 49, 54 Draba aprica in Oklahoma 223 Draparnaldia glomerata 276 Drosera filiformis 50, 53; inter- media 50, 52; linearis 50, 53, 54; rotundifolia 50, 52; var. comosa 50, 53, 54 Dryopteris excelsa 35; Phegop- teris 295; thelypteris var. pubescens 149 Dugesia mexicana 123, 127 Dyssodia pinnata 123, 124, 128 Eaton, R. J., A Check List of Walter Deane's Seedling Col- lection 287 Eaton, R. J., Bean, R. C. and Hill, A. F., Corrections in Re- ports of the Committee on Plant Distribution 348; Four- teenth Report of the Commit- tee on Plant Distribution 47 Echinochloa walteri 134 Ecological Interpretation of Rhododendron Colonies in Maine and New Hampshire, An 61 Elaeagnus angustifolia 44 Eleocharis parvula 41 Eleusine indica 134 Elodea nuttallii 40 Elymus villosus f. arkansanus 134; f. villosus 134 Environmental Variation in Het- erotheca subaxillaris 243 Epigaea repens var, glabrifolia 66 1961] Equisetum hyemale var. affine 341; sylvaticum 40; variega- tum 40; var. nelsonii 40 Eragrostis capillaris 134; curvu- la 148; diffusa 41; perplexa 41; hypnoides 137; poaeoides 134; reptans 137 Eragrostis curvula from Illinois 148 Eremosphaeria viridis 277 Erigeron exilis 342; pubescens 122, 127; scaposus 122, 127; tenuis 147 Eriophorum angustifolium 41; viridicarinatum 41 Eruca sativa 49, 55 Erucastrum gallicum 49, 55; obtusangulum 345 Erysimum cheiranthoides 49, 54; inconspicuum 49, 53, 54; pan- nonicum 49, 55; repandum 49, 55 Erythronium americanum 38, 146; carolianum 38, 39 Eschatogramme desvauxii 35 Eschscholtzia californica 47, 55 Euastrum. 278 Euglena polymorpha 280 Eunotia robusta var. diadema 280; var. tetrodon 280 Eupatorium fistulosum 20; gla- bratum 121, 122, 126; pazcua- rense 121, 126, 342, 344; prunellaefolium 121, 122, 126; scorodonioides 121, 126 Euphorbia dentata 212; hetero- phylla 212; polygonifolia 288 Evidence for the Hybrid Origin of Scirpus peckii 237 Ewan, J., Contemporary Notes on Powell’s Expeditions in the West (Review) 179 Fagus grandifolia 65, 67, 68 Fern New to Rhode Island, A 46 Festuca rubra 135 Florida, Omissions in Key to Xyris in 205 Index to Volume 63 355 Fourteenth Report of the Com- mittee on Plant Distribution 47 Fragilaria (? capucina) 280 Fraxinus americana 65, 288, 295; nigra 65, 68, 289 Fumaria officinalis 47, 54 Further Notes on Chamaecypar- is thyoides in New Hampshire 281 Galactia martii 17, 18 Galinsoga parviflora 45 Galium labradoricum 45; tri- florum 289; virgatum 20 Gaultheria odorata 342, 343, 344; procumbens 346; f. leucocarpa 347 Gaura filipes 20 Genus Lechea (Cistaceae) in the Southeastern United States, The 103 Genus Oleandra of Costa Rica, The 335 Genus Pteris of Costa Rica, The 194 Gerardia fasciculata 20 Geum rivale 43; urbanum 288 Gillenia stipulacea 146 Gisekia pharnacioides, A New Weed 226 Glassman, S. F., Notes on the Grass Flora of the Chicago Region. II 130 Glaucium flavum 47, 55 Glaucocystis nostochinearum 280; oocystiformis 280 Gloeotrichia echinulata 281 Glyceria arkansana 24; grandis 135; pallida 20; septentrionalis var. arkansana 24 Gnaphalium americanum 342, 344; attenuatum 343; semi- amplexicaule 342 Godfrey, R. K. and Adams, P., Observations on the Sagittaria subulata Complex 247 Godfrey, R. K., Gisekia pharna- cioides, A New Weed 226 356 Golenkinia paucipina 277 Gomphonema acuminatum var. coronatum 280; constrictum 280; dichotomum 280; gemina- tum 280 Gonatozygon aculeatum 278; brebissonii 278 Gonium sociale 276 Gonolobus gonocarpos 147 Grammitis anfractuosa 35; cultrata 35; jubaeformis 35; knowltoniorum 35; mollissima 35; pendula 35; sericeolanata 35; serricula 35; taxifolia 35; tenuicola 35 Grindelia coronopifolia 156; oxylepis 122; var. eligulata 127; gymnospermoides 158 Guatemalan Species, Chromo- some Numbers in the Com- positae. V. Mexican and 121 Guenthera viscosa 152, 155, 158 Gutierrezia Alamani 152, 158; glutinosa 124, 127; gymno- spermoides 152, 158; Wrightii 152, 160 Gymnosporangium 284 Gymnozyga moniliformis 279; var. gracilescens 279 biseptatum Habranthus texanus 216 Hackelia americana 44; virgini- ana 44 Hagenah, D. J., Spore Studies in the Genus Cystopteris. I. The Distribution of Cystopteris with Non-spiny Spores in North America 181 Hammamelis virginiana 51, 52, 66, 68, 288; var. parviflora 51 Hapalosiphon hibernicus 281 Haplopappus lanceolatus 179; rhizomatus 177 Harley, W. J., A Second Station for Stewartia ovata on the Coastal Plain 206 Harms, V. L., Poindexter, J. D. Rhodora [Vol. 63 and McGregor, R. L., New Plant Records from Kansas 146 Hedyotis arenaria 221; asperu- loides f. brandegeana 221; australis 222; brevipes 221; callitrichoides 222; capillipes 222; drymarioides 221; exigula 221; gracilenta 222; gracilis 221; greenei 222; longipes 222; nigricans 330; f. salina 222; ovata 222; palmeri 222; parvu- la 222; peninsularis 222; prin- glei 222; prostrata 222; sax- atilis 222; sinaloae 222; umbratilis 222; vegrandis 222; watsonii 222; xestosperma 222 Helenium amarum 330, 332; var. badium 330; tenuifolium 330 Helianthum canadense 112 Heliotropium tenellum 20 Hesperis matronalis 49, 54, 348 Heterospermum pinnatum 123, 127 Heterostyly and Homostyly in Lithospermum canescens (Boraginaceae) 229 Heterotheca inuloides var. rosei 343, 344; latifolia 245, 246; subaxillaris 243, 244, 245, 246; var. petiolaris 243, 244, 245; var. procumbens 244, 245 Heuchera americana 50, 53 Hieracium gronovii 146; robus- tum 272 Hill, A. F., Eaton, R. J. and Bean, R. C., Corrections in Reports of the Committee on Plant Distribution 348; Four- teenth Report of the Commit- tee on Plant Distribution 47 Hodgdon, A. R. and Pike, R., An Ecological Interpretation of Rhododendron Colonies in Maine and New Hampshire 61; Rhododendron maximum in Hopkinton and Harrisville, New Hampshire 26 1961] Homalocenchrus lenticularis 137 Hordeum pusillum 41; montan- ense 41 Houstonia arenaria 217, 218, 219, 220, 221; asperuloides 217, 218, 219, 220; australis 220, 222; boscii 220; brandegeana 221; brevipes 217, 218, 219, 221; callitrichoides 218, 219; co- rymbosa 217, 219; crassifolia 218, 219; croftiae 220; dryma- rioides 221; gracilenta 217, 218, 219, 220, 222; gracilis 221; greenei 217, 219; greenmanii 220; herbacea 219; intricata 220; longipes 222; microtheca 217, 218, 219, 220; mucronata 220; mullerae 218, 219; pal- meri 222; parvula 222; penin- sularis 217, 219, 222; pringlei 219, 220; prostrata 222; pur- purea 217; salina 222; saxatil- is 217, 218, 219; serpyllacea 218, 219, 220; subviscosa 220; umbratilis 217, 218, 219, 222; uniflora 220; watsonii 217, 219, 220; wrightii 218, 219; xestos- perma 220 Houstonia and Oldenlandia un- der Hedyotis, Merger of the North American 217 Hyalotheca dissiliens 279; un- dulata 279 Hybrid Aspens in the Lower Peninsula of Michigan 311 Hydrangea paniculata 50, 54; quercifolia 50, 55; radiata 50, 55 Hydrochloa caroliniensis 256 Hymenoxys insignis 123, 128; odorata 123, 125, 128 Hypericum androsaemum 12; buckleyi 13; cistifolium 13; edisonianum 15; ellipticum 11; filicaule 12; galioides 11; humifusum 12; hypericoides 15; kiboense 12; lobocarpum 20; macrosepalum 12; micro- Index to Volume 63 357 sepalum 15; myrtifolium 11, 13; perforatum 13; punctatum 13; stans 15; stragulum 15; suffructicosum 15; tetrapeta- lum 15 Iberis amara 49, 55; umbellata 49, 55 Ilex verticillata 66 Illinois, A Decade of Botanizing in 19; Eragrostis curvula from 148 Impatiens capensis 289 Important Notice to Members, Subscribers and Librarians, back cover Index to Volume 63, 350 Indigofera cf. truxillensis 18 Indispensable Manual of Tropi- cal Marine Botany, An (Re- view) 57 Introgression of Alnus serrulata and Alnus rugosa 297 Irwin, H. S. and Turner, B. L., Chromosome | Numbers of Some Brazilian Leguminosae 16 Isoetes butleri 20, 146 Johnston, M. C., A New Species of Haplopappus from Mexico 177 Jones, F. B., A New Zephyran- thes from Southern Texas 214 Juglans cinerea 295; illinoinensis 296 Juncus diffusissimus 20; gerardi 42;secundus 20 Juniperus horizontalis 348; vir- giniana 40 Jussiaea leptocarpa 20 Kalmia angustifolia 66; latifolia 28, 66, 67 Kansas, New Plant Records from 146 Keerlia linearifolia 152, 158 Kirchneriella lunaris 277; obesa 358 var. major 277; subsolitaria 277 Kral, R., Omissions in Key to Xyris in Florida 205 Kucera, C. L. and Steyermark, A. J., New Combinations in Grasses 24 Lactuca saligna 147; f. ruppiana 148 Lamb, I. M., An Indispensable Manual of Tropical Marine Botany (Review ) 57 Lagynion (? ampullaceum) 280 Lamium amplexicaule 289 Lappula redowskii var. cupulata 44; var. occidentalis 44 Larix laricina 284 Lathyrus japonicus var. glaber 288 Lechea cernua 1061 108, 114, 117; Deckertii 107, 108, 110, 111, 113; divaricata 105, 108, 110, 111; exserta 13; Legget- tii 103, 107, 108, 114, 115; var. ramosissima 115; major 112; maritima 112, 116; minor 106, 108, 111, 112; myriophylla 113, 114; patula 106, 108, 110, 111, 113; prismatica 113; racemu- losa 104, 106, 107, 108, 109, 114; san-sabeana 104, 117; tenuifolia 104, 106, 108, 110, 111; var. occidentalis 110; var. tenuifolia 110; Torreyi 103, 107, 108, 114, 116; var. con- gesta 116; var. Torreyi 116; tripetala 103, 105; villosa 104, 105, 107, 108, 109, 110, 111, 112, 115; var. macrotheca 107; var. Schaffneri 107; var. typica 107; var. villosa 107, 109 Leersia lenticularis 137; virgini- ca 137 LeMaire, R. J., A Range Exten- sion for Parapholis incurva 176 Lemna valdiviana 20 Rhodora [ Vol. 63 Lepidium campestre 49, 54; densiflorum 49, 54, 146; lati- folium 49, 55; perfoliatum 49, 55; ruderale 49, 54; sativum 49, 54; virginicum 49, 51, 177 Leptochloa attenuata 26; fascic- ularis 134; filiformis 26; var. attenuata 26; var. filiformis 26 Leptoloma cognatum 135 Lesser Antillean Ferns, Notes on 31 Lewis, W. H., Merger of the North American Houstonia and Oldenlandia under Hedyo- tis 217 Lilaeopsis chinenses 252 Lilium longiflorum 274; super- bum 307; f. plenum 307 Linaria canadensis 45; dalmatica 45 Lincoln, A. Jr, A Notable Assemblage of Plants in New Hampshire 294 Lindera Benzoin 66 Lindsaea lobata 72 Linnaea borealis 45; var. ameri- cana 66 Liparis lilifolia 175 Lipocarpha maculata 21 Liquidamber Styraciflua 51, 53 Liriodendron tulipifera 67 Lithospermum angustifolium 234; canescens 229, 230, 231, 233, 234; decumbens 230; in- cisum 232, 233, 234; longi- florum 232, 234; tournefortii 230 Lithospermum canescens (Bor- aginaceae), Heterostyly and Homostyly in 229 Lobelia boykinii 236 Lobularia maritima 49, 54 Long, R. W., A New Manual for Ohio Vascular Plants (Re- view) 55 Lonicera canadensis 66; Xylos- teum 119 1961] Lóve, A., Some Notes on Myrio- phyllum spicatum 139 Lóve, D., A Useful Multilingual Botanical Dictionary ( Review) 291 Lunaria annua 49, 55 Lycopus americanus 289 Lycopodium flabelliforme 21 Lycopsis arvensis 44 Lyngbya aerugineo-coerulea 281 Lyonia ligustrina 66 Machaeranthera blephariphylla 124; gymnocephala 124, 125, 127; tanacetifolia 122, 127 MacKeever, F. C., An Albino Fruited Form of Gaultheria procumbens 346; A Variegated Foliage Form of Commelina 88 Macleaya cordata 47, 55 Maine and New Hampshire, An Ecological Interpretation of Rhododendron Colonies in 61 Malcolmia maritima 49, 55 Malva rotundifolia 289 Marssoniella elegans 280 Massachusetts, A List of Algae from Selected Areas in 275; Notes on the Flora of Peni- kese Island 149 McGregor, R. L., Harms, V. L., and Poindexter, J. D., New Plant Records from Kansas 146 Melampodium montanum 128 Melosira sculpta 280; varians 280 Mentzelia albescens 118, 119, 147; Wrightii 118 Mentzelia albescens and Loni- cera Xylosteum in Missouri 118 Menyanthes trifoliata 44 Merger of the North American Houstonia and Oldenlandia under Hedyotis 217 Meridion circulare 280; inter- Index to Volume 63 359 medium 280 Merismopedia glauca 280; tenu- issima 280 Mexican and Guatemalan Spe- cies, Chromosome Numbers in the Compositae. V. 121 Mexico, A New Species of Hap- lopappus from 177 Michigan, Bidens connata var. gracilipes Fern, in Western 145; Hybrid Aspens in the Lower Peninsula of 311 Micrasterias americana 279; laticeps 279; mahabuleswaren- sis 279; pinnatifida 279; radi- ata 279; sol 279; truncata 279 Microcystis aeruginosa 280; in- certa 280 Microspora tumidula 276 Mimulus guttatus 45 Mirabilis albida 43, 330, 332; var. lata 330; hirsuta 43; linearis 43 Mississippi, A Disjunct Com- munity of Chestnut Oak in 324 Missouri, Mentzelia albescens and Lonicera Xylosteum in 118 Mitella diphylla 50, 53; nuda 50, 52; prostrata 50, 53, 54 Mohlenbrock, R. H., A Decade of Botanizing in Illinois 19 Mollugo verticillata 226 Mougeotia scalaris 278 Moul, E. T., Notes on the Flora of Penikese Island, Massachu- setts 149 Muhlenbergia curtisetosa 25; cuspidata 137; mexicana f. ambigua 135; Schreberi 25; Schreberi curtisetosa 25; var. curtisetosa 25; sobolifera 137; tenuiflora 137 Mühling, G. N. and Wilson, G. B., The Chromosomes of Podo- phyllum peltatum 267 Myosurus aristatus 43; minimus 146 260 Myriophyllum alterniflorum 140, 141; exalbescens 140, 141, 142, 143; var. magdalenense 143; magdalenense 143; spicatum 140, 141, 142, 143, 144; tenel- lum 140; verticillatum 44, 140 Myriophyllum spicatum, Some Notes on 139 Najas guadalupensis 251 Nasturtium brevipes 9; curvisi- liqua 3; officinale 49, 54; var. microphyllum 49, 54; var. siifolium 49, 55; palustre var. brevipes 9; portoricensis 9; sinuatum 2; tanacetifolium 6, 9; var. insularum 9; trachy- carpum 1, 2 Navarretia propinqua 44 Navicula diaphala 280 Nemopanthus mucronata 66, 68, 284 Nephrocytium agardhianum 277; obesum 277 Nephrodium invisum 34 Nephrolepis biserrata 32; daval- lioides (var. ?) furcans 32; exaltata var. monstruosa 32; faleata 32; f. furcans 32 Neslia paniculata 49, 54 Netrium (? oblongum) 278; digitus 278; var. constrictum 278 New Combinations in Grasses 24 New Hampshire, An Ecological Interpretation of Rhododen- dron Colonies in Maine and 61; Further Notes on Chamae- cyparis thyoides in 281; A Notable Assemblage of Plants in 294; Rhododendron maxi- mum in Hopkinton and Har- risville, 26 New Jersey, A New Species of Panicum from 235 New Manual for Ohio Vascular Plants, A (Review) 55 New Name for the Puberulent Rhodora [ Vol. 63 Sessile-leaved Uvularia, A 36 New Plant Records from Kansas 146 New Records from North Da- kota 39 New Species of Haplopappus from Mexico, A 177 New Species of Panicum from New Jersey, A 235 New Variety of Rudbeckia ful- gida, A 119 New Zephyranthes from South- ern Texas, A 214 Nicolletia edwardsii 123, 125, 128 North America, A Weedy Cruci- fer Again Reaches 345 North American Houstonia and Oldenlandia under Hedyotis, Merger of the 217 North Dakota, New Records for 39 Notable Assemblage of Plants in New Hampshire, A 294 Notholaena affinis 81; angusta Plate 1257, 80, 81; aurea 82; brachypus 82, 83; cantangen- sis Plate 1257, 81, 82, 83; cubensis 81; dealbata 85, 86; Ekmanii 81; Hassleri 82, 83; limitanea 85; nivea 85; obduc- ta 87; peruviana 87; Schaff- neri 80, 81; sinuata 82; soli- taria 82, 83, Plate 1258; Stuebeliana 83, Plate 1258, 85 Nyssa sylvatica 65, 68 Oakesia puberula 86, 39; sessili- folia var. (?) nitida 36, 39 Oakesiella nitida 37, 39; puber- ula 37, 39 Observations on the Sagittaria subulata Complex 247 Oedogonium boscii 277; gracilius 277; reinschii 277; subsex- angulare 277 Oenothera caespitosa 41; humi- fusa 177; rhombifolia 44; stri- gosa 44 1961] Ohio Vascular Plants, A New Manual for (Review) 55 Oklahoma, Draba aprica in 223 Oldenlandia affinis 220; benguil- lensis 218; callitrichoides 222; capillipes 222; cryptocarpa 221; divaricata 218; eludens 220; greenei 222; hymenophyl- la 221; juncoides 220; luzu- loides 218; micrantha 218; ovata 222; pringlei 222; se- paneoides 221; somala 220; staelioides 220; umbellata 220; verticillata 220; xestosperma 222 Oldenlandia under Hedyotis, Merger of the North Ameri- can Houstonia and 217 Oleandra articulata 336, 338; Bradei 336, 338; costaricensis 336, 338, 340; decurrens 336, 337, 339, 340; guatemalensis 837, 340; nodosa 336; var. caudata 338; panamensis 337, 340; trinitensis 340; var. sub- costaricensis 340 Onoclea sensibilis 40 Ontario, Two New Orchid Rec- ords for 175 Onychonema filiforme 279 Oocystis crassa 277; elliptica 277; solitaria 277 Ophyiocytium capitatum 279; parvulum 279 Opuntia polycantha 41 Orchis purpurella 176 Oryzopsis asperifolia 137, 285, 286; melanocarpa 287; pun- gens 286; racemosa 137, 287 Oryzopsis, Which Side is Up? A Look at the Leaves of 285 Oscillatoria 280 Ostrya virginiana 65 Oxalis corniculata 288; montana 288 Oxydenia attenuata 26 Palmer, E. J., Mentzelia albes- Index to Volume 63 361 cens and Lonicera Xylosteum in Missouri 118 Pandorina morum 276 Panicum clandestinuia 125; di- chotomiflorum 136; var. puri- tanorum 135, 136; hirstii 236; roanokense 236; spretum 235 Papaver dubium 47, 55; Rhoeas 47, 54; somniferum 47, 54 Parapholis incurva 176 Paricutin Volcano in 1960, Vas- cular Plants on the Cinder Cone of 340 Parnassia glauca 50, 51, 53 Paspalum dissectum 236 Pease, A. S., Campanular Per- sistence 29 Pecan, The Specific Epithet of the 296 Pediastrum 277 Pellaea dealbata var. Stuebeli- ana 83, 85; Stuebeliana 83; ternifolia 341; var. ternifolia 341 Penium (? closteroides) 278; navicula 278 Penstemon alluviorum 20; ar- kansanus 20 Penthorum sedoides 50, 52 Perdue, R me Jr. A New Variety of Rudbeckia fulgida 119 Periandra mediterranea 18; var. mucronata 17, 18 Peridinium cinctum 280 Perymenium mendezii 128 Petasites palmatus 45 Philadelphus coronarius 50, 55; inodorus 50, 54; pubescens 50, 55 Phormidium retzii 281 Physocarpus opulifolius var. intermedius 146 Phytolacca icosandra 342, 344 Picea mariana 284, 288; rubens 65, 68, 69 inundatum 281; 362 Pike, R. and Hodgdon, A. R., An Ecological Interpretation of Rhododendron Colonies in Maine and New Hampshire 61; Rhododendron maximum in Hopkinton and Harrisville, New Hampshire 26 Pilea opaca 21; fontana 20, 42 Pinus montezumae ? 343; resin- osa 65; strobus 65, 68; Stro- bus 288; sylvestris 150 Pityogramma calomelanos 343; tartarea 341, 344 Plantago rugelii 289 Platanus occidentalis 51, 52 Platyzoma microphyllum 92, 96 Pleurotaenium coronatum 278; maximum 278; trabecula 278 Poa fendleriana 41; palustris 136 Podophyllum peltatum 267, 268, 271, 273, 274 Podostemum ceratophyllum 50, 51, 53, 54 Poindexter, J. D., McGregor, R. L., and Harms, V. L., New Plant Records from Kansas 146 Polanisia graveolens 43, 47, 53; trachysperma 43 Polygonatum biflorum 20; pubes- cens 295 Polygonum aviculare 288; Con- volvulus 288; douglasii 42; orientale 288; Persicaria 288; scandens 288 Polypodium anfractuosum 35; cultratum 35; excelsum 34; induens 35; jubaeforme 35; knowltoniorum 35; mollissi- mum 35; pendulum 35; serice- olanatum 35; serricula 35; taxifolium 35; tenuiculum 35; virginianum 46 Populus acuminata 42; alba 150, 318, 319, 320, 321; P. alba x grandidentata 318, 321, 322; x tremuloides 318, 322; x P. Rhodora [ Vol. 63 tremuloides 318; deltoides 150; grandidentata 65, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320; P. grandidentata x tremuloides 312, 321, 322; x P. tremuloides 312, 315, 317, 319; tremula 141; tremuloides 65, 141, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320, 321 Porophyllum amplexicaule 123, 128 Portulaca oleracea 288 Potamogeton friesii 40; illinoen- sis 251; perfoliatus 40; rich- ardsonii 40 Potentilla palustris 43 Proctor, G. R., Notes on Lesser Antillean Ferns 31 Prunus mexicana 20; pennsyl- vanica 65; serotina 65, 288; virginiana 295 Psilostrophe gnaphaloides 128 Pteretis pensylvanica 40 Pteridium aquilinum var. feei 343 Pteris altissima 196, 202, 203; biaurita 195, 200, 203; coriacea 199; costaricensis 200; cretica 195, 196, 197; grandifolia 195, 196, 197, 205; Killipii 86; Kun- zeana 202; Lechleri 86; livida 195, 201, 203; longicaudadel 205; macrodictya 205; mollis 199; muricata 195, 199, 201; muricella 195, 197, 199, 201; navarensis 205; paucinervata 195, 196, 197, 198, podophylla 196, 203, 204, 205; prolifera 205; propinqua 195, 200, 203; pungens 195, 197, 198, 205; quadriaurita 195, 198, 201, 205; tripartita 196, 203, 204; vestita 86 Puccinellia cusickii 41 Pyrethrum Bonplandianum 152, 156 Pyrus floribunda 66 1961] Quadrigula lacustris 277 Quercus alba 65, 66, 67; coccinea 65; michauxii 325; Prinus 65, 67, 68, 324, 325; rubra 65, 68; velutina 65, 67 Range Extension for Parapholis incurva, A 176 Ranunculus ficaria 289, 290; var. bulbifera 289, 291; var. fer- tilis 289; var. ficaria 289, 291; harveyi 20 Raphanus Raphanistrum 49, 54, 149, 345; sativus 49, 54 Rapistrum rugosum 49, 55 Re-evaluation of the Generic Status of Ascyrum and Crookea (Guttiferae), A 10 Reseda alba 50, 54; lutea 50, 54; Luteola 50, 55; odorata 50, 55 Rhizoclonium hieroglyphicum 276 Rhode Island, A Fern New to 46 Rhododendron maximum 26, 27, 61, 62, 63, 64, 65, 66, 68, 69, 70 Rhododendron maximum in Hop- kinton and Harrisville, New Hampshire 26 Rhus copallina 150; radicans 66 Rhyncosinapis cheiranthos 346; erucastrum 345 Ribes americanum 50, 52; cynos- bati 50, 58, 54, 348; glandu- losum 50, 52; Grossularia 50, 54; hirtellum 48, 50, 52, 348; var. calcicola 50, 53; var. sax- osum 50, 53, 54; lacustre 51, 52; missouriense 51; nigrum 51, 54; odoratum 651, 54; rotundifolium 51, 53, 54, 348; sativum 51, 54; triste 51, 52 Roberts, E. G., A Disjunct Com- munity of Chestnut Oak in Mississippi 324 Robson, N. K. B. and Adams, W. P., A Re-evaluation of the Generic Status of Ascyrum Index to Volume 63 363 and Crookea (Guttiferae) 10 Rock, H. F. L., Turner B. L., and Beaman, J. H., Chromo- some Numbers in the Com- positae. V. Mexican and Guatemalan Species 121 Rollins, R. C., Draba aprica in Oklahoma 223; Notes on American Rorippa (Cruci- ferae) 1; A Weedy Crucifer Again Reaches North America 345 Roots and the Taxonomic Dif- ferences between Botrychium oneidense and B. dissectum 164 Rorippa amphibia 49, 55; curvi- siliqua 3; islandica 49, 53, 54; var. Fernaldiana 49, 51; var. hispida 49, 52; portoricensis 9; ramosa 4, 5, 6; sessiliflora 49, 55; sinuata 1, 2, 8, 4, 6, 7; sylvestris 49, 54; Walteri 6, 7, 8, 9 Rorippa (Cruciferae), Notes on American 1 Rubus alumnus 20; enslenii 21; laciniatus 149 Rudbeckia bicolor 20; fulgida var. auriculata 119; var. ful- gida 120; var. spathulata 120; grandiflora 148; subtomentosa 148 Ruellia caroliniensis 20 Rumex domesticus 42; steno- phyllus 42 Ruppia occidentalis 40 Saffordia induta 87 Sagina decumbens 146 Sagittaria filiformis 247; gra- minea 265; kurziana 247, 248, 249, 250, 251, 252, 2583, 255. 260, 261, 262, 263, 264; lorata 237; natans 247, 248, 254, 264; sagittifolia 249; stagnorum 247, 248, 254, 256, 257, 258, 259, 260, 261, 262, 263, 264, 364 Rhodora 265; subulata 247, 248, 250, 252, 258, 261, 262, 263, 264, 265; var. gracillima 247, 253, 268, 264, 265; subulata var. lorata 247; var. natans 247; var. typica 247 Sagittaria subulata Complex, Observations on the 247 Salicornia europaea 288 Salix Bebbiana 66; hartwegii 341, 343, 344 Salsola collina 42; kali 288 Sambucus canadensis 45, 150; pubens 66 Sanguinaria canadensis 47, 52, 295 Sanicula gregaria 44 Sanvitalia ocymoides 128; pro- cumbens 128 Sarracenia purpurea 50, 51 Sassafras albidum 66 Saururus cernuus 146 Saxifraga aizoides 51, 52; Aizoon 51, 52; var. neogaea 51; oppositifolia 51, 52; pen- sylvanica 51, 52; rivularis 51, 52; stellaris var. comosa 51, 52; virginiensis 51, 52 Scamman, E., The Genus Olean- dra of Costa Rica 335; The Genus Pteris of Costa Rica 194 Scenedesmus 277-278 Schizaea confusa 99; fistulosa 99; melanesica 99; papuana 99; pusilla 99 Schuyler, A. E., Evidence for the Hybrid Origin of Scirpus peckii 237 Scirpus atrocinctus 237, 238, 239, 240, 241, 242; atrovirens 237, 238, 239, 240, 241; cyperinus 238; koilolepis 20, 146; longii 238; peckii 237, 238, 239, 240, 241, 242; pedicellatus 237, 238, 239, 240, 241 Scleria reticularis 21 Second Station for Stewartia [Vol. 63 ovata on the Coastal Plain, A 206 Sedum acre 50, 54; alboroseum 50, 55; anopetalum 50, 55; purpureum 50, 54; Rosea 50, 52; rupestre 50, 55; sarmen- tosum 50, 55; spurium 50, 55; Telephium 50, 55; ternatum 50, 53 Seedling Collection, A Check List of Walter Deane's 287 Selenastrum minutum 277 Sempervivum tectorum 50, 54 Senecio resedifolius 125; san- guisorbae 125, 129; toluccanus 123, 125, 129; cf. cyclophyllus 123, 125, 129 Setaria glauca 136; lutescens 136 Sherardia orientalis 45 Sherff E. E., Bidens connata var. gracilipes Fern. in Western Michigan 145 Shinners, L. H., Viola Rafines- quii: Nomenclature and Native Status 327 Silene Cucubalus 288 Sisymbrium altissimum 49, 54; curvisiliqua 3; curvisiliquum 3; Loeselii 49, 54; officinale 49, 54; var. leiocarpum 50, 54, 288; orientale 50, 55; tanaceti- folium 7; Walteri 6, 7 Sisyrinchium pruinosum 147; varians 147 Smilacina racemosa 295; stellata 308, 309 Smilax herbacea 20 Solanum dulcamara 45; interius 44; tuberosum 274 Solbrig, O. T., Synopsis of the Genus Xanthocephalum (Com- positae) 151 Solidago boottii 20; buckleyi 20; rugosa 20; sempervirens 289; var. mexicana 177; strigosa 20; uliginosa 21 Sonchus oleraceus 177, 343, 344 1961] Sorastrum americanum 277; var. undulatum | 277; spinulosum 277 Sparganium chlorocarpum 40; multipedunculatum 40 Specific Epithet of the Pecan, The 296 Spergularia marina 288 Sphaerozosma excavatum 279; filiforme 279; granulatum 279 Spirogyra hyalina 278; rivularis 278 Spirulina laxa 280 Spirotaenia condensata 278 Spondylosium papillosum 279; planum 279; pulchellum 279; pulchrum 279 Spore Studies in the Genus Cys- topteris. I, The Distribution of Cystopteris with Non- spiny Spores in North America 181 Sporobolus airoides 41; asper 25, 136; var. canovirens 25; brevi- folius 137; canovirens 25; clandestinus 25, 187; var. canovirens 25; var. clandes- tinus 25; confusus 341, 344; neglectus 25, 331; var. neglec- tus 25; var. ozarkanus 25; vaginiflorus 25, 137; var. inaequalis 331; var. ozarkanus 25; virginicus 137 Staurastrum 279 Stauroneis inflata 280 Steele, F. L., Introgression of Alnus serrulata and Alnus rugosa 297 Stellaria media 288 Stephanomeria tenuifolia 45 Stevens, O. A., New Records for North Dakota 39 Stewartia ovata 206 Steyermark, J. A. and Kucera, C. L., New Combinations in Grasses 24 Stigeoglonium aestivale 276; (glomeratum or subsecundum) Index to Volume 63 365 276; lubricum 276; tenue 276 Stigonema mamillosum 281 Streptanthus squamiformis 224 Streptopus roseus var. perspec- tus 295 Stylites andicola 88; gemmifera 88 Subularia aquatica 50, 52 Swallen, J. R., A New Species of Panicum from New Jersey 235 Synedra acuta 280; radians 280; ulna 280; var. biceps 280; var. danica 280 Synopsis of the Genus Xantho- cephalum (Compositae) 151 Tabellaria fenestrata 280; (? var. asterionelloides) 280; flocculosa 280 Taenitis desvauxii 35 Tagetes lucida 129 Talinum calycinum 21 Taraxacum kok-saghyz 45; offi- cinale 289 Taxonomie Fern Notes. I. 70 Teesdalia nudicaulis 50, 55 Tetraedron caudatum 277; mini- mum 277; obesum 277; tri- gonum 277; tumidulum 277 Tetradesmus wisconsinense 278 Texas, A New Zephyranthes from Southern 214 Thalictrum | dasycarpum var. hypoglaucum 331 Thelypteris antillana 33, 34; ger- maniana 33; hydrophila 33, 34; invisa 34; muscicola 33; oligocarpa 34 Thlaspi arvense 50, 54 Thieret, J. W., The Specific Epi- thet of the Pecan 296 Thuja occidentalis 65 Tiarella cordifolia 51, 52, 288 Tilia americana 295; glabra 65 Tillaea aquatica 50, 53 Tipularia discolor 20 Trachelomonas hispida 280 366 Trachypteris pinnata 88 Tradescantia fluminensis 89 Tragopogon pratensis 45 Tribonema bombycinum 279; minus 279 Trichodesmium lacustre 280 Trichomanes fimbriatum 381; superbum 31; trigonum var. fimbriatum 31 Trifolium repens 288 Trillium cuneatum 20; erectum 21, 295; pusillum 307; var. ozarkanum 307 Triplasis purpurea 134, 136 Triploceras gracile 278 Trochiscia obtusa 277; reticu- laris 277 Tryon, A. F., Some New Aspects of the Fern Platyzoma micro- phyllum 91 Tryon, R., A Botanical Atlas of the Danish Flora (Review) 349; Taxonomic Fern Notes. I. 70 Tsuga canadensis 65, 66, 67, 68, 69 Turner, B. L., Beaman, J. H. and Rock, H. F. L., Chromosome Numbers in the Compositae. V. Mexican and Guatemalan Species 121 Turner, B. L. and Irwin, H. S., Chromosome Numbers of some Brazilian Leguminosae 16 Two New Orchid Records for Ontario 175 Typha angustifolia 40 Ulmus americana 65, 288; pu- mila 42 Ulothrix subconstricta 276; tenerrima 276; variabilis 276 Useful Multilingual Botanical Dictionary, A (Review) 291 Utricularia intermedia 45 Uvularia caroliniana 39; nitida 36, 39; perfoliata 38; puberula Rhodora [Vol. 63 36, 38, 39; var. nitida 37, 39; pudica 37, 38, 39; var. nitida 37, 39; sessilifolia 42, 295; sessifolia nitida 36, 39 Vaccinium angustifolium 66; corymbosum 66, 68, 284; myrtilloides 66; stamineum 146 Vagnera stellata 308 Variance in Herbarium Speci- men Identification and Other Considerations Based Upon the Preparation of a Local Flora 207 Variegated Foliage Commelina, A 88 Vascular Plants on the Cinder Cone of Paricutin Volcano in 1960 340 Verbesina hypomalaca 123, 124, 128; serrata 123, 128 Vernonia crinita 21 Veronica peregrina 45 Viburnum alnifolium 66, 68; cas- sinoides 66, 68; recognitum 66 Viguiera stenoloba 128 Viola bicolor 327, 329, 334; in- cognita 44; Kitaibeliana 327, 328, 329, 331, 332, 333, 334; var. Machadiana 332; var. Rafinesquii 327; nephrophylla 43; papilionacea 43; pensyl- vanica var. leiocarpa 295; Rafinesquii 327, 328, 329, 330, 331, 332, 333, 334; renifolia 295; rotundifolia 295; sarmen- tosa 43; Selkirkii 295; tricolor 327; x Viola bernardi 43; viarum 21 Viola Rafinesquii: Nomenclature and Native Status 327 Vitis lincecumii 20 Voss, E. G., Which Side is Up? A Look at the Leaves of Oryzopsis 285 Form of Wagner, W. H. Jr., Roots and TT 1961] the Taxonomic Differences be- tween Botrychium oneidense and B. dissectum 164 Webber, E. E., A List of Algae from Selected Areas in Massa- chusetts 275 Weedy Crucifer Again Reaches North America, A 345 Which Side is Up? A Look at the Leaves of Oryzopsis 285 Wigandia kunthii 342, 343, 344 Wilbur, R. L., A New Name for the Puberulent Sessile-leaved Uvularia 36 Wilbur, R. L. and Daoud, H. S., The Genus Lechea (Cistaceae) in the Southeastern United States 103 Wilson, G. B. and Miihling, G. N., The Chromosomes of Podophyllym peltatum 267 Winterringer, G. S., Eragrostis curvula from Illinois 148 Wolffiella floridana 20 Woodsia glabella 181; obtusa 40; oregana 40; var, cathcartiana 40 Xanthidium antilopaeum 279; var. minneapoliense 279; var. Index to Volume 63 367 polymazum 279 Xanthocephalum Alamani 158; Benthamianum 153, 156, 162, 163; Bonplandianum 152, 156; centauroides 152, 153, 155, 156; conoideum 161; dentatum 156; gymnospermoides 153, 154, 155, 156, 158, 163; humile 153, 155, 157; linearifolium 153, 154, 156, 157, 163; mega- locephalum 153, 155, 163; seriocarpum 152, 153, 155, 161; suffruticosum 152, 156; Wrightii 153, 154, 156, 160, 163 Xanthocoma dentata 156; humile 152, 155, 157 Xerophyllum asphodeloides 308; tenax 308 Xyris smalliana 205 Yucca filamentosa 309; Smalli- ana 309 Zaluzania coulteri 123, 124, 128 Zephyranthes pulchella 215, 216; refugiensis 214, 215, 216 Zinnia angustifolia 123, 128 Zygnemopsis desmidioides 278