Flora Novo-Galiciana A Descriptive Account of the Vascular Plants of Western Mexico Rogers McVaugh General Editor: William R. Anderson VOLUME 12 CompositaeZinnia violaces. drawn By ome of the units with the Royal Botanical Expedition 19 New Spain, 1787-1803, The plate, probably based om a plant feom Guerrero, represenis No. 332 of the Leones Florae Mexicanae of Sessé de Mocifi, an is repreduecd hereby permission of the Huat lnstsute for Botanical Documentation, Pxtsburgh. Fist ‘culated in Madrid in about 1790, wraler te name of Zima elegans, this species was soon widely dispersed 40 ther European botanical centers, and many of the most popular garden rinnias have been derived from it, We are sreatly indebted to Edna S. Newnan and the Horsee H, Rackham Seto! of Graduite Studies, University of ‘Michigan, for making this reproduction possibleFlora Novo-Galiciana A Descriptive Account of the Vascular Plants of Western Mexico Rogers McVaugh General Editor: William R. Anderson VOLUME 12 Compositae ANN ARBOR — THE UNIVERSITY OF MICHIGAN PRESS.To the memory of Sidney Fay Blake 1392-1959 Friend and Teacher Issued 17 September 1984 Copyright © by The University of Michigan 1984 All rights reserved Published in the United States of America by The University of Michigan Press and simultaneously in Rexdale, Canada, by John Wilev & Sons Canada, Li Manufactured in the United States of America Volume 12 of the Flora Novo-Galiciana is published with the support of National Science Foundation Grant No. DEB- 8118738. The Foundation provides awards for research and edu- cation in the sciences. The awardee is wholly responsible for the conduct of such research and preparation of the results for publi- cation, The Foundation, therefore, does not assume responsibility for such findings or their interpretation, Any opinions, findings, conclusions, or recomendations expressed in this publication are those of the author and do not necessarily reflect the views of the National Science Foundation. Library of Congress Cataloging in Publication Data (Revised for volume 12) McVaugh, Rogers, 1909- Flora Novo-Galiciana. Includes bibliographical references and index. Content 12. Compositae — — v. 14. Gramineae. 1. Botany—Mexico—Jalisco—Classification, 2. Botany —Mexico—Classification. 1. Anderson, William R. If, Title. TIT. Title: Novo-Galiciana, III. Title: Western Mexico. QK211.M38_ 1983 581.972 82-13537 ISBN 0-472-04814-7 (v, 14) ISBN 0-472-04812-0 (v. 12)Contents Compositae ......-....- - 1 Acknowledgments 0.0.00... 2 Morphology and Terminology 3 Division into Tribes. -....e eee 8 Delimitation of Genera vie Alphabetical Arrangement of Genera ....... 10 Tribes of Compositae in this Flora, with the Included Genera .......-.-. UL Keys to the Tribes... ....22200eeceee 12 Doubtful and Excluded Genera ....... 1125 Index to Names of Compositae .... 2. 1129 Map of Nueva Galicia ...... following IndexCOMPOSITAE! References (general): Bentham, George, & J. D. Hooker. Genera Planta- rum 2, pt. 1: 163-533. 1873. Hoffmann, O., in Engler & Prantl. Natiirl. Pflan- zenfam. 4, pt. 5: 87-387. 1894. Rydberg, P. A. Carduales, Ambrosiaceae, Cat- duaceae [key to tribes}. N. Amer. Flora 33: 1-46. 1922; Carduaceae, Helenicae (part). N. Amer. Flora 34: 1-80. 1914; Carduaceae, Helenieae (part, with certain genera by Harvey Monroe Hall), Tageteae (part). N. Amer. Flora 34: 81-180. 1915; Carduaceae, Tageteae (part), Anthemideac. N. Amer. Flora 34: 181-288. 1916; Carduaceae, Liabeae, Neurolzeneae, Senecioneae (part). N. Amer. Flora 34: 289-360. 1927. Gleason, H. A. Carduaceae, Vernonieae. N. Amer. Flora 33: 47-110. 1922, Heywood, V. H., J. B. Harborne, & B. L. Turner (eds.). ‘The biology and chemistry of the Compositae. Vol. 1: xiv, 1-619; vol. 2: xiv, 621-1189. Academic Press, 1977. References (pertinent to Mexican floristics): Blake, $. F. Asteraceae (1 certain genera by B. L. Robinson and J. M. Greenman), in Trees and Shrubs of Mexico by Paul C. Standley. Contr. U.S. Nat. Herb, 23: 1401-1641, 1681. 1926; Asteraceae described from Mexico and the southwestern United States by M. E. Jones, 1908-1935, Contr. U.S. Nat, Herb, 29: 117-137. 1945. Correll, D. S., & M. C, Johnston. Compositae, in Man. Vasc. Plants Texas, pp. 1523-1736. Texas Research Foundation, Renner, 1970. D’Arcy, W. G., et al. Compositae, in Flora of Panama. Ann. Missouri Bot, Gard. 62: 835-1322. 1975. Nash, D. L., L. O, Williams, et al. Compositae, in Flora of Guatemala. Fieldiana Bot. 24, pt. 12: i-ix, 1-603. 1976. Rzedowski, J. Claves para la identificacién de los géneras de la familia Compositae en México. Acta Cient. Potosina 7: 1-145. 1978. Annual or perennial herbs or [our species] often shrubs or small trees, diverse in habit, foliage and inflorescence, the individual flowers small and sessile in a close head on a common receptacle, sometimes individually subtended by a small bract (pale), and nearly always collectively surrounded by an involucre of few to many bracts (phyllaries), the flower-head as a whole often simulating a single flower; individual flowers perfect or unisexual, gamopetalous, regular or irregu- lar, commonly S-merous (not rarely 4-merous); calyx none, or represented by pappus of hairs, scales, or awns, or a ring or crown, or combinations of these: stamens as many as the coroila-lobes and alternate with them, epipetalous, the anthers elongate, united into a tube surrounding the style (rarely free), and in- trorsely dehiscent; ovary inferior, 2-carpellate, unilocular with one ovule, nor- mally becoming an achene; style usually 2-cleft; achenes very diverse in form and ornamentation. One of the largest families of flowering plants, the number of species esti- mated at 13,000-15,000 or more, or more than 10% of all flowering plants. They are cosmopolitan in distribution, and are represented by many spccies in almost TAn alternative name for the family, sanctioned by the International Code of Botanicat Nomen: clatuce, is Asteraceae. The type-genus of the family is Aster L.2 FLORA NOVO-GALICIANA all habitats except in tropical rain-forests and in strictly aquatic situations, where they are relatively uncommon. Many Compositae are cultivated for ornament, and a few (considering the total size of the family) are economically important in other ways (e.g., the common sunflower, cultivated for the oil of its edible seeds). In Nueva Galicia the family is represented by approximately 750 species in 144 genera. Of these almost all are native species; only about five or six species are known to represent recent introductions from outside the area. This is in marked contrast to such areas as California where more than 100 species, or approximately one-eighth of the total, are introduced, and occur as weeds or escapes from cultivation. The Compositae are relatively numerous in Nueva Galicia, as may be seen by comparing the numbers of native species occurring in various regions for which modern lists are available: Size of Region, Compared Region Native Species Native Genera with Nueva Galicia Arizona 923 135 cax2 California 693 43 cax3 Baja California 390 120 ca x 115 Sonoran Desert 416 um ca x25 Guatemala 593 135, caxi Venezuela 607 133, caxT Nueva Galicia ca 750 14g About haif of the species known from Nueva Galicia are characteristic of the Pacific slope of Mexico, ranging chiefly from Oaxaca or Guerrero to Sinaloa, Durango, Sonora, or southern Arizona. About three-quarters of the species are confined to continental Mexico west of the Isthmus of Tehuantepec. Very few species are strictly endemic, but a considerable number of gencra in various tribes have the major parts of their ranges on the Pacific slope (e.g., Bolanosa, Jaliscoa, Olivaea, Pericalia, Microspermum, Oxypappus, lostephane). About one-fourth of all the species are true shrubs or trees, and only about five-eighths of the species are strictly herbaceous. Woody species are well distributed among the several tribes, including about one-half of the Vernonieae (ten spp), three-tenths of the Heliantheae (about 80 spp.) and the Eupatoricae (about 50 spp.), and smaller fractions of the other tribes. Acknowledgments Among the names of Mexican Compositae, a disproportionately large number are based on type-specimens deposited in but few herbaria in the United States and Europe. For the privilege of studying types and other authentic specimens, I am indebted to many herbaria, but especially to the authorities of the British Museum (Natural History) (BM); the Conservatoite Botanique, Geneve (G, G- DC); the Gray Herbarium, Harvard University (GH); the Royal Botanic Gar- dens, Kew (K); the Muséum National d'Histoire Naturetle, Paris (P); and the United States National Museum (US). For comments, suggestions, and aid in many ways during the preparation of the text of this work, I am grateful to manyCOMPOSITAE 3 friends who took time to share their knowledge with me, though I did not always take their advice. ‘The drawings in the main are the work of Karin Douthit, to whom I am very grateful for her patience and attention to endless detail. Her talent speaks for itself. It is also a pleasure to acknowledge the assistance of Humberto Sénchez Cordova, who drew all or part of 24 figures (nos. 10, 11, 20, 29, 34, 42, 43, 54, 66, 69, 76, 7, 81, 99, 100, 115, 125, 128, 140, 144, 154, 155, 157, 179); Cathie Isaacs, who drew fig. 28, a~c, and fig. 153, e-g; Jeanne C. Koelling, who drew fig. 141, and Chester W. Laskowski, who drew fig. 159, a-f. Financiat support for most of the fieldwork in Nueva Galicia, for study of types and other specimens in herbaria in Europe, Mexico, and the United States, and for preparation of this Flora, has been generously provided by the National Science Foundation, Washington, D.C., through the following grants: G3305, 10665, GBS218X, GB31133, GB40366X (=BMS 7300738), DEB-7680638, and DEB-8004178. Other funds were provided by the Horace H. Rackham Schoo! of Graduate Studies and by the Office of Research Administration, University of Michigan, During the final preparation of the manuscript, working space and other facilities were placed at my disposal by the Department of Botany, Univer- sity of North Carolina at Chapel Hill. Morphology and Terminology In the keys and descriptions that follow, much of the terminology applies uniquely to the specialized structures that make the Compositae so distinctive among plant-families, and the reader will do well to make himself familiar with the morphology of the composite head, and with the terminology, before using the Flora. The flowers in a head may be all perfect (hermaphrodite, bisexual) and fer- tile, but there are many deviations from this condition. Very commonly there is a central group of disk-flowers, and a marginal row or rows of ray-flowers; a head composed wholly of disk-flowers is said to be discoid. The disk-fiowers are perfect and fertile, or sterile but functionally staminate, and the corolla is regular, tubular or trumpet-shaped to narrowly funnelform, with typically five short terminal lobes. Often there is an evident distinction between the proper mbe of the corolla (the narrow part extending upward from the attachment to the achene approxi- mately to where the filaments become free from the inside of the corolla), and the throat, which is often enlarged or inflated below the lobes. The ray-flowers are pistillate or newtral (without a style), and the corolla is tubular at the base only; above this it is flat, commonly bent to one side and often exhibiting traces of two or three of the lobes as terminal teeth. The flattened portion of the flower is known as a ray, or figule; in the head it may simulate a petal of a single flower and may be misinterpreted as such (as in the song, “picking petals off a daisy”). Heads having both rays and disk-flowers are said to be radiate, In some species the ligule of the marginal pistillate flowers does not develop, so the corolla is tubular, sometimes with terminal teeth; a head is said to be disciform when the central flowers are perfect, with tubular corollas, and the marginal flowers are pistillate or neutral but without ligules, their corollas tubular or wanting. Another4 FLORA NOVO-GALICIANA type of flower, superficially resembling the ray-flower of a radiate head, but differing in being perfect and in having usually five terminal teeth, is characteristic of the tribe Lactuceae, in which all the flowers are of this type. In the tribe Mutisicae, and in a few genera in other tribes, the outer lobes of the disk-flowers may be enlarged and spreading from the head, thus simulating three-lobed ligules. The anthers are often colored like the corolla, sometimes black or green; the connectives are sometimes pale, and usually prolonged distally into stiff or even rigid, colored or colorless, spoon-shaped oF flat, mostly ovate appendages much shorter than the body of the anther. The anther-sacs are often prolonged at base (the anthers then said to be sagittate or auriculate, or caudate, the auricles pro- tonged into a tail. The anthers open inward, and the pollen is pushed out through the anther-tube by the growth of the style. The styles of the ray-flowers are mostly similar in all groups, and those of sterile disk-flowers are often reduced and undivided. In the fertile disk-flowers the style-branches commonly diverge above the anther-tube, have distinctive forms and textures varying from tribe to tribe and from genus to genus, and tend to be stigmatic only on definite parts of their surface, sometimes bearing a sterile appendage at the tip. For descriptions of styles, see page 8. The achenes may be alike throughout the head, or those of the disk-flowers may differ markedly from those of the rays; sometimes even in discoid heads the central achenes differ from those of the Howers on the periphery. Achenes may be narrowly linear, or (commonly) slightly enlarged from base toward apex (cla- vate), ot variously ellipsoid, obovoid or subglobose. They may be round in cross- section (terete) or variously angled, ribbed or furrowed; in many groups they are compressed (i.e., more or less flattened and biconvex). Usually the disk-achenes are radially compressed (with one edge facing the center of the head and the other facing the periphery), and the ray-achenes tend to be tangentially com- pressed with the outer (abaxial) surface flat or convex and the inner surface convex or angled. In some groups the achenes are strongly flattened and have their margins prolonged into membranous or cartilaginous wings. At the base of the achene, where it is attached to the receptacle, there may be an enlarged and thickened cartilaginous margin or foot (carpopodium). The receptacle (the enlarged end of the axis upon which the flowers are borne) is said to be naked when the individual flowers are without subtending bracts (receptacular bracts, or pales, Latin paleae), or chaily (paleaceous) when such bracts are present. Its surface at maturity usually bears the scars showing the positions from which the achenes have fallen, and may be characteristically modi- fed. The achene-scars may appear as small pits in the surface, or sometimes be raised above it. The pits may be separated by thin chaffy partitions, as in a honeycomb (an alveolate receptacle), or the partitions or the entire receptacle may be more or less hairy or bristly Care should be taken to distinguish the pales, which subtend the disk-flowers, from the bracts that subtend the ray-flowers or, in discoid heads, sometimes the outer disk-flowers. These outer bracts are usually considered to represent an inner row or rows of phyllarics. They are usually readily distinguished from the pales, but sometimes are morphologically intermediate between the pales and the outer phyllaries (i.c., the sterile phyllaries which do not directly subtend a flower).COMPOSITAE 5 ‘The pappus, located at the summit of the achene, is very diverse in structure. The hairs, scales, or other elements of which it is composed may be arranged radially (as in many groups with terete or columnar achenes) or bilaterally (as in many groups with compressed achenes). When bilaterally arranged the elements tend to become reduced in number. When a pappus is composed of hairlike elements, these may vary in thickness from the most slender (capillary) to stiff, bristle-form elements that may be called awns rather than hairs. The awns may be flattened or triangular in cross-section, or widened toward the base. Awns and hairs may be smooth, or variously plumose (i.e., with hairlike branches) or barbed. There is no clear distinction between barbed hairs or awas (those with large sharp batbs, as in many species of Bidens, of on a fishhook), and barbellate hairs (those with small lateral projections, such as occur in most Compositae with capillary pappus). Bristles and awns are retrorsely barbed or barbellate if the projections point backward (toward the base) or antrorsely if they point toward the apex. Pappus elements that are relatively broad and fiat are commonly called scales; when these occur on the same achene with larger awns, or larger scales of a different series, the smaller elements may be called squamellae (diminutive of Latin squama, scale). ‘The phyllaries are commonly imbricated (j.e., overlapping like shingles in an essentially spiral arrangement), and often graduated (i.e., the inner series longest, and outer series successively shorter) The number of phyllaries making up the involucre, the number of disk- flowers, and the number of ray-flowers in a head, are variable within more or less Gefinable limits in different genera and species. The number of phyllaries may be constant throughout a genus (¢.g., always five in Stevia), or may vary even among the plants of a single species. The ray-flowers if present are relatively few, seldom more than about 35 and most often about five, eight, or 13. The disk-flowers may be constant in number throughout a genus (e-g., four in Mikania) or in a species (e.g., three in Piqueria triflora), or may vary within species and from group to group, sometimes becoming very numerous (€.g., 300-500 or even about 1000 in a single head). Both flowers and phyllaries tend to be spirally arranged, although this may not be apparent when their numbers are small. The actual number of phyllaries characteristic of a species, like the number of disk-flowers, or the number of rays, tends to cluster about one or more numbers of the Fibonacci series (1-2-3-5-8-13-2]-34--35, etc.), in which any term of the series represents the sum of the two preceding terms. Thus in a species in which the disk-flowers are known to vary in number from about 11 to 24, experience suggests that in most of the heads they will vary from about 11 to 15 or from about 20 to 22. The “inflorescence” in the Compositae, although often so-called, is not strictly analogous to that of other flowering plants, because in fact each head represents a flowering branch, not a single flower. Commonly the terminal branch (head) of the principal axis flowers first, followed by the heads terminating the lateral axes (if any) below this. Growth of any one axis is determinate (that is, it is stopped by the development of the flower-head), so that in old inflorescences the oldest (terminal) head is often found in mature or over-mature condition, much overtopped by the Jater-developing lateral branches. In some Compositae the flowering branches are opposite throughout, but very commonly the branches of6 FLORA NOVO-GALICIANA the inflorescence, at least the smaller ones, are alternate. Because the terminal heads usually flower first, true racemes and panicles are essentially unknown in the Compositae, although the terms are sometimes rather loosely used when the heads are arranged in so-called racemiform or paniculiform inflorescences. Individual heads may be sessile, or pedunculate at the ends of more or less elongated, leafy or naked branches. Often there is no sharp distinction between the reduced leaves along the flowering axis below the head and the specialized accessory bracts that may form a calyculus immediately subtending the involucre Sometimes the upper cauline bracts merge gradually with the outer phyllaries, with no perceptible gap between. The naked portion of the stem immediately below the head is termed the peduncle (also called pedicel by some authors on Compositae, by analogy with the flower-stalks in other plant-families). Strictly speaking, the length of the peduncle is the distance from the head to the first bract below it, but long, unbranched, and nearly naked flowering axes, even if bearing one or several smalll bracts, are often loosely designated in their entirety as “peduncles,” and their bracts as “peduncular bracts.” Morphological terms used in the descriptions of species, if not self-explana- tory, may be found in any good botanical glossary. Terms referring to various types of hairiness, to leaf-shape and toothing of leaf-margins, are sometimes subjective and difficult to define precisely. Plant-hairs and other superficial processes Plant hairs are generally outgrowths from epidermal cells, often superficially resembling animal hairs, and presumably named by analogy with them. Other types of outgrowths of similar origin are less hairlike; such srichomes may be several- or many-celled, greatly enlarged at the base, variously thickened and hardened, curved or hooked (uncinate), or tipped with small secretory cells (capitate glands,” gland-tipped hairs, or glandular hairs). Presence and absence of hairs are described by series of useful but not very precise terms. A surface without any hairs or other trichomes is said to be glabrous; (the word smooth is sometimes used erroneously as a synonym for glabrous, but strictly speaking smooth refers to a lack of roughness or irregularity, not a lack of hairs). The terms glabrate and glabrescent mean “made nearly glabrous” or “becoming gla- brous,” and refer chiefly to surfaces that are pubescent when young but tose all or @ part of their hairs in age. The most generally used adjectives denoting the Presence of hairs are pubescent and pilose, In common botanical usage, and in this Flora, pubescent refers particularly to surfaces that are downy with short soft hairs, but may also be used as a general term, i.e., a opposed to glabrous. Puberulent means minutely pubescent. Pilose (from Latin pilosus, hairy or shaggy) surfaces bear hairs that are longer and straighter; villous surfaces are those with long but somewhat softer hairs (for all practical purposes pilose and villous are interchangeable). More specialized terms refer to hairs and often superficial processes in differ- ent positions or those that are modified in shape or in other ways. These include arachnoid (cobwebby); ciliate (with hairs on the margin); floccose (with soft densely packed hairs falling away in little tufts); hirsute (with only moderatelyCOMPOSITAE, 7 course and stiff and mostly erect straight hairs); hispid (with coarse rigid erect hairs or bristles); lanate or lanose (woolly; tomentose, with fine matted and kinky hairs); papillose (pimpled; with minute rather soft tubercles); scabrous (rough; covered with short hard rigid points); sericeous (silky; i.e., with long straight glossy, soft, closely appressed hairs); sérigose (with stiff, rather short, closely appressed bristle-like hairs); tomentose (covered, often thickly, with curled or curved matted hairs). Leaves and ather plane surfaces Leaves and their divisions vary from filiform (literally, threadlike) or linear (narrow, with the two opposite margins parallel) to oblong (broader, the sides nearly parallel, the ends more or less obtuse), elliptic (longer than wide, with curved sides, broadest near the middle), ovate (broadest near the base, as in the longitudinal section of an egg); lanceolate (comparatively narrower than ovate, usually tapering to the tip, broadest below the middle), obovate or oblanceolate (as in ovate and lanceolate but reversed, the broadest part above the middle), or orbicular (circular). Perfect typical shapes are rare, and compound terms like elliptic-lanceolate or oblong-obovate are often used. Special terms often used are cuneate (wedge-shaped), rhomboid (more or less elliptic, but a little angular in the middle), deltoid (with a broadly triangular outline like the Greek A), cordate (having two round lobes at the base, like the heart in a pack of cards), hastate (with two acute diverging lobes at base, like the head of a halbert); pandurate or panduriform (fiddle-shaped; obovate, with a deep sinus on each side), ‘The tips of leaves and their divisions may be rounded (like an are of a circle), obtuse (terminating gradually in a rounded end), truncate (ending abruptly, as if cut off transversely), or emarginate (having a notch at the end), or they may be variously pointed; the different shapes are expressed by special terms. Apices may be acute (terminating in a point, with two almost straight lines converging at an angie less than 90°); attenuate (long-tapering, diminishing gradually in breadth); acuminate (tapering gradually or abruptly from inwardly curved sides into a pro- jecting point); subulate (awl-shaped; tapering to a very fine point); aristae (awned; abruptly ending in a straight subulate point, either long or short); mucro- nate (with a hard short point); cuspidate (tapering into a rigid, usually sharp, point), Leaf-bases may be cordate or, like the tips, rounded, obtuse, acute or attenu- ate. Leaf-margins may end abruptly at the summit of the petiole, or may be decurrent (prolonged below the base of the blade into broad or narrow wings bordering the petiole or the stem below the leaf). In this Flora the length of the petiole usually is stated in terms of the unmargined part only, but in some leaves with much prolonged, attenuate bases the margins may become very narrow toward the base, and the narrowly margined part is often loosely interpreted as a petiole. Leaf-margins may be entire (without teeth or other division) or sinuate (alter- nately with concavities and convexities), or variously toothed, The most general terms are dentate (toothed, the usually sharp teeth pointing straight outward), or8 FLORA NOVO-GALICIANA. denticulate (having smal! or minute teeth), Other common types of margins are crenate (with rounded teeth), serrate (saw-toothed: having sharp mote or less straight-edged teeth pointing toward the apex), or crenate-serrate (low rounded teeth pointing toward the apex). Leaves may be pinnately nerved or veined (penninerved), ot often 3-nerved (ie., with three nearly equal veins from the base of the blade or the two lateral veins prominent but not so strong as the midvein), or ‘riplinerved (i.¢., with a pair of strong ascending Jateral veins arising from the midvein definitely above the base). Division into Tribes The Compositae have traditionally been divided into tribes, the exact number of these varying from about 12 to about 20 depending upon taxonomic opinion. The principal divisions have been made upon characters of the anthers (e.g., whether or not caudate at base), the receptacle (e.g., whether or not chaffy), the corollas, the pappus, and especially the styles. The style-branches in general are receptive to pollen only on their upper (inner) surfaces; the stigmatic areas may cover most of the inner surfaces of the branches, or may be more or less restricted toward the bases and margins of the branches. When the stigmatic areas are thus restricted, the sterile tips of the branches may be truncate or obtuse, with or without a terminal brush of hairs, or prolonged into a subulate, sagittate or clavate appendage. When without appendages, the branches may be truncate, and penicillate (bearing a tuft of hairs at the end) (the so-called Senecioid style); or elongate, gradually attenuate, and acute at the end, and hispidulous on the out- side along their whole length (Vernonioid style); or less elongate, rounded at the end and usually less or not at all hispidulous (Inuloid style). When appendages are present they may be clavate, i.c., obtuse at the apex and usually thicker than the stigmatic parts of the branches (Eupatorioid style); or acute and tapering from the base and then either hairy on both sides (Helianthoid style), or glabrous at Jeast within (Asteroid style), ‘The tribes as understood and described in this Flora almost all have nearly their traditional circumscriptions, i.e., chiefly those of Bentham (in Bentham & Hooker, 1873). The major exception to this is the tribe Helenieae, which in its traditional sense is represented in Nueva Galicia by about 70 species in 20 genera. Most workers now agree that the original Helenieae were a heterogeneous assem- blage constituted by Bentham in an effort to maintain other tribes, mainly the Senecioneae and Heliantheae, in a more homogeneous state. The result was a tribe whose members were alike in having the receptacles naked (without chaff, or pales), but otherwise made up of several very different groups. ‘There has been no general agreement as to the disposition of the different groups of genera assigned to the old Helenicae. With respect to the genera represented in our flora, there seems to be a consensus that Bahia, Baileya, Florestina, Galeana, Helenium, and Schkuhria have their affinities with the Heli- antheae, which includes almost half the species of Compositae in Nueva Galicia. The anomalous genus Microspermum has by common consent been accepted into the Eupatorieae. Six genera (Chrysactinia, Dyssodia, Hydropectis, Pectis, Poro-COMPOSITAE 9 Phyllum, and Tagetes) are members of the recently revived tribe Tageteae, char- acterized in part by the translucent glands in the foliage and the phyllaries. Another anomalous genus, Jaumea, with a single representative in our flora, has been assigned to a newly proposed tribe, Coreopsideae (Turner & Powell, in Heywood et al., 1977, p. 724), or to a group in subtribe Coreopsidinae, tribe Heliantheae (Stuessy, in Heywood et al., 1977, p. 637). Our species, incidentally, has minute translucent foliar glands superficially suggestive of those of the Tage. teae. Finally, four rather different genera (Eutetras, Flaveria, Oxypappus, and Perityle) have been pushed toward the Senecioneae by Turner and his co-workers (e-g., A. M. Powell & B. L. Turner, Amer. J. Bot. 61: 87-93. 1974; Turner & Powell, in Heywood er al., 1977, chapter 25), but have not been accepted into the Senecioneae by Nordenstam, who recently published a systematic review of that tribe (in Heywood, et al., 1977, chapter 29). The case for inclusion of these four genera among the other genera of Senecioneze is not very thoroughly docu- mented by Turner & Powell, who indeed admit (1977, p. 733) that if the four fit neither into the Senecioneae nor the Heliantheae, their assignment to tribe may be delayed until they have been more thoroughly studied. In the present Flora, the members of the former tribe Helenieae have been assigned to Heliantheae except for Microspermum and the genera of Tageteae, This should raise no hackles among specialists in Compositae, unless for the disposition of Ewtetras, Flaveria, Oxypappus, and Perityle. Our species of these genera have opposite leaves (at least at the lower nodes), and none has a capillary Pappus. At least in these respects they seem out of place among our other genera Of Senecioneae, and more like our genera of Heliantheae, which seems already sufficiently diverse to contain them. It seems likewise probable that Liabwn, with vernonioid styles, yellow flowers, opposite leaves, milky juice, and multiseriate involucte, is out of place in the Senecioneae to which it has usually been assigned. It seems equally out of place in the Vernonieae, except for the characters of its styles and stamens, but some thoughtful authors, including Turner & Powell (in Heywood et al., 1977, p. 712) are inclined to put it there. Rydberg (1927) created for Liabum and four other segregate genera the new tribe Liabeae, and in this some recent authors have concurred (¢.g., H. Robinson & R. D. Brettell, Phytologia 25: 404—407 1973; and B. Nordenstam, in Heywood et al., chapter 29, 1977. A general work like the Flora Novo-Galiciana is an unsuitable forum for debates on the niceties of evolutionary classification, and most genera have been assigned to tribes without discussion, often on traditional lines without any firm conviction on the part of the author. The genera are arranged in one alphabetical sequence without division into tribes. The name of the tribe is indicated in the text for each gemus. Keys to the genera in each tribe, and a separate enumeration of the genera in each, are provided in the introduction. Delimitation of Genera The most difficult decisions that had to be made in the preparation of the following treatment were those involving generic limits, and the recognition or non-recognition of taxa that had been proposed as genera. Since before 197010 FLORA NOVO-GALICIANA there has been a flood of publications resulting from research on American Com- positae. At least two-thirds of the genera in our area have been more or less thoroughly investigated from the taxonomic point of view; the conclusions as expected range from the conservative to the extremely liberal. In order to achieve a more or less uniform generic concept throughout this and other volumes of the Flora Novo-Galiciana I have tried to keep to a middle course. It is almost impos- sible to be consistent. Jf there is a bias in the Flora it is probably toward a more inclusive generic concept, and toward genera whose names are well established in the literature, The reader will not find many of the recently proposed or resusci- tated small genera that have been segregated from the larger complexes like Eupatorium, Brickellia, Senecio, or Viguiera; as B. L.. Turner has well said (Brit- tonia 30: 344. 1978), one needs “compelling reasons” for the elevation of “this or that ‘exceptional’ species” to generic rank. On the other hand, once a genus has existed unchallenged for a half-century or more, the mere fact that it has not been. challenged for so long gives it at least a specious right to recognition. Thus the Flora includes Bolanosa, Chromolepis, Decachaeta, lostephane, Mexianthus, Oli- vaea, Pericalia, and Pseudelephantopus, which may or may not have any better qualifications than Barkleyanthus, Neohintonia, Phanerostylis, Pittocaulon, Stes- sya, and Telanthophora, Even here some seeming, inconsistencies may be de- tected; Cymophora (1907), Leucactinia (1915), and Leucosyris (1897) are not admitted to the Flora as independent genera in spite of their venerability, but Digitacalia (1968) and Pseudoconyza (1961) are accepted. In these instances as in all others, taxonomic considerations as I see them have prevailed. ‘The following new names or new combinations are published in this volume of the Flora: Hieracium hintonii Beaman, Lasianthaea rosei (Greenm.) McVaugh, Senecio glinophyllus (H. Rob. & Brett.) McVaugh, Viguiera cordata var. websteri (Turner) McVaugh, Viguiera michoacana (Turner & Davies) McVaugh, and Zin- nia microglossa (DC.) McVaugh. Alphabetical Arrangement of Genera ‘The arrangement of the genera in a single alphabetical series is intended to facilitate access to any particular genus of Compositae among the approximately 140 others in this Flora. There are advantages in grouping the genera of Composi- tae by tribes, because then related (and presumably similar) genera and their species are near one another in the book and easier to find once the reader has placed the group in a tribe. In this Flora most of that seeming advantage is illusory because the one tribe Heliantheae includes almost half of all the genera and species in the family. If the genera of Heliantheae were to be alphabetically arranged, such groups as Aldama and Sclerocarpus would still be very far apart in the text, even though they are very closely related if not congeneric. One alterna- tive is to arrange the genera in some kind of systematic order within the tribes. This makes it difficult to locate any particular genus without consulting the index. In the second place there is no generally accepted order. in the printed pages of a book the order is necessarily linear, but in evolutionary terms the relationships are best expressed by three-dimensional diagrams, so that closely related genera appear together in groups, like twigs on one branch of the evolutionary tree. In two major modern revisions of the taxonomy of the Heliantheae, the authorsCOMPOSITAE ul agree upon the limits of certain infra-tribal taxa, but they disagree upon the limits of many others, they disagree upon the order in which the infra-tribal groups are to be presented, and they disagree upon the limits of the tribe itself. (Stuessy, Tod F., in Heywood et al., 1977, chapter 23; Harold Robinson, Smithson. Contr. Bot. 51: iv, 102 pp. 1981). For those wishing to familiarize themselves with tribal groupings as accepted in this Flora, lists of the included genera arranged by tribes are set forth below, and the treatment of each genus in the text includes the name of the tribe to which it is assigned. Tribes of Compositae in this Flora, with the Included Genera ANTHEMIDEAE (p. 18) Hleischmannia Achillea Gymnocoronis ‘Artemisia Hofmeisteria Chrysanthemum Jaliscoa Cotula Mexianthus Matricaria Microspermum Soliva Mikania Piptothrix ASTEREAE (p. 18) Piquerta Aphanostephus Stevia Archibaccharis Trichocoror ae Doubtful and Excluded Genera Astranthium Erythradenia Baccharis Isocarpha Chaevopappa Piqueriopsis Conyza Egletes HELIANTHEAE (p. 24) Erigeron Aldama Grindelia Ambrosia Gymnosperma Bahia faplopappus aiteya Heterouhecs Baltimora Olivaea Berlandiera Solidago. Bidens Townsendia Calea Xanthocephalum Calyptocarpus Chromolepis CARDUEAE (p. 20) Chosanthellum Centaurea G i. Gentaur ‘oreopsis Cosmos Dahlia EUPATORIEAE (p. 21) Deiilia Ageratella Desmanthodium Ageratum Eclipta ‘Alomia Encelia Barroetea Euphrosyne Brickellia Entetras Carminatia Flaveria Carphochaete Horestina Decachaeta Galeana Eupatorium Galinsoga2 FLORA NOVO-GALICIANA HELIANTHEAE (continued) Guardiola Helenium ‘Helianthus Heliopsis Heterosperma Hymenoclea ostephane Iva Jaegeria Jaumea Lagascea Lasianthaea Melampodium Melanthera Milleria Montanoa ‘Otopappus Oxypappus Parthentum Perityle Perymenium Podachaenium Polymnia Rumfordia Sclerocarpus Sigesbeckia Simsia Spilanthes Synedrella ‘ithonia Tridax Trigonospermum Verbesina Viguiera Wedelia Xanthium Zaluzania Zinnia Doubtful and Excluded Genera INULEAE (p. 35) Gnaphalium Pluchea Pseudoconyza LACTUCEAE (p. 35) Hieracium Lactuca Pinaropappus Pyrrhopappus Sonchus Taraxacum LIABEAE (p. 36) Liabum MUTISIEAE (p. 37) Chaptalia Jungia Onoseris Perezia Trixis SENECIONEAE (p. 37) Digitacalia Erechtites Odontotrichum Pericalia Pippenalia Psacatium Senecio TAGETEAE (p. 40) Chrysactinia Dyssodia Hydropectis Pectis Porophyllum Tagetes VERNONIEAE (p. 42) Eryngiophyllum Bolanosa Haplocalymma Elephantopus Sabazia Pseudelephantopus Thelesperma Vernonia Keys to the Tribes Key A. Introductory Key. 1, Plowers all igulate and perfect; ligules S-toothed at apex; herbaceous plants with Iatex. Lactuceae (p. 35). 1. Ligulate flowers, if present, pistillate or neutral, arranged around the periphery of the head; figules 2—4-toothed at apex; plants commonly without conspicuous Intex.COMPOSITAE, 2B 2, Ligolate flowers present, the ligules sometimes very small 3, Pappus none. Key B, 3. Pappus present, 4, Pappus (at least in part) of awns oF scales, oF a crown or cup. Key C. 4. Pappus (at least in part) of capillary bristles, Key D. 2. Ligulate flowers none, the heads discoid or disciform. 5. Pappus none Key E. 5. Pappus present, 6. Pappus (at Ieast in part) of awns or scales, or a crown or cup. Key F. 6. Pappus (at least in part) of capillary bristles Key G, Key B. Ligulate flowers present, pappus none. 1, Receptacle naked (without pales). 2. Leaves opposite, or the uppermost alternate Heliantheae (p. 24). 2. Leaves alternate or basal. 3. Ligules commonly more than 3 mmm wide, yellow. 4 Leaves all in a basal rosette, peltate, radiately S-lobed almost to the base; scapes T-headed. Pippenalia in Senecioneae, 4, Leaves cauline or partly basal, not peltate, entire or toothed or pinnately lobed; heads solitary or few. Heliantheae (p. 24) 3. Ligules rarely as much as 3. mm wide, usually 2 mm ot less, variously colored, Astereae (p. 18) 1. Receptacte chatty (with pales) 5, Style-branches truncate, penicillate; phyllaties with dry scarious margins and tips; heads ‘humerous in a dense terminal corymb; ray- and disk-flowers white; leaves altemate, dissected, Achillea in Anthemideae, 5. Style-branches mostly with acute to attenuate, hairy tips; combinations of characters not as. ‘above. Heliantheae (p. 24). Key C. Ligulate flowers present; pappus (at least in part) of awns or scales, or a crown or cup. 1, Receptacle naked (without pales. 2. Leaves, and usually phyllaries, with translucent, usually conspicuous, linear to ellipsoid or ‘globose oil glands, Tageteae (p. 40) 2. Leaves and phyllaries without oil glands, sometimes with impressed or superficial resinous drops of exudate 3. Pappus partly of awns or bristles, partly of scales or a short erown, 4. Pappus of many long capillary bristles and an outer series of much shorter scales; leaves ‘opposite. Liabum in Liabeae, 4. Pappus not of numerous long bristles and an outer series of scales. 5. Leaves alternate. Astereae (p. 18), 5. Leaves opposite at least below Heliartheae (p. 24). 3. Pappus of more of less uniform scales, or a short crown. 6. Leaves lobed, to deeply pinnately or bipinnately divided, 7. Pappus of distinct, conspicuous scales; at least the Jower leaves opposite. Heliantheae (p. 24). 7. Pappus a low, sometimes dissected or Ambriate crown; leaves alternate.4 FLORA NOVO-GALICIANA 8. Phyllaries with dry cearious or hyaline tips and margins: style-tips mostly truncate, penicillate ‘Anthemideae (p. 18) 8, Phyllaries not as above; style-tips acute to attenuate, hairy on the backs. ‘Astereae (p. 18), 6, Leaves entire or dentate 9. Plants strongly resinous-glutinous. 9, Plants not markedly glutinous. Astereae (p. 18) 10. Disk of the head globose: stems winged. Helenium in Heliantheoe. 10. Disk not globose; stems wingless. 11. Pappus of distinct, well-developed scales. Heliarsheae (p. 24). 11. Pappus a short crown, sometimes with 1 or more elongate bristles. 12. True ligules none, the marginal flowers perfect with enlarged outer lobes simulating ligues. . Microspermum in Eupatorieae, 12, Marginal flowers ligulate, pistllate Astereae (p. 18). 1, Receptacle chatty (with pales). Heliantheae (p. 24). Key D. Ligulate flowers present; pappus (at least in part) of capillary bristles. 1. Leaves, and usually phyllaries, with translucent, usually conspicuous, linear to ellipsoid or ‘globose oil glands. Tageteae (p. 40), 1. Leaves and phyllaries without oi} glands, sometimes with impressed or superficial resinous drops ‘of exudate. 2. Leaves opposite 3, Pappas of many long capillary bristles and an outer series of much shorter scales. Liabum in Liabeae. 3, Pappus not of numerous long bristles and an outer series of scales, Heliantheae (p. 24) 2. Leaves alternate or basal 4. Woolly scapose perennials, the scapes I-headed; anthers caudate at base. Chaptalia in Mutsieae. 4. Habit various, not as above; anthers not caudate at base (all flowers pistillate in Baccharis). 5. Style-branches truncate and penicillate; phyllaries commonly subequal, the involucre ‘usually subtended by a calyculus of smaller bracts. Senecio in Senecioneae. 5, Style-branches with acute to attenuate, hairy tips; phyflaries often conspicuously gradue ‘ated, the involucre without a well-marked ealyeulus at base. Astereae (p. 18), Key E, Ligulate flowers none; pappus none. 1. Heads unisexual, the staminate and pistillate heads on the same plant, the staminate above the pistillate; pistilate heads with few Bowers inclosed in a nutlike, burlike, or winged invoucre Heliantheae (p. 24). 1. Heads not unisexual, all alike; involuere not burlike nor winged 2. Involucre very flat, the few flowers more oF less concealed between two broad, unequal, suborbicular-obovate phyllaries. Delilia in Heliancheae. 2, lavolucre not strongly flattened. 3, Heads 1-flowered (seldom 2-flowered), aggregated into secondary capitula resembling ordi nary heads, Helianaheae (p. 24) 3. Heads with several or many flowers, not aggregated into secondary capitula. 4. Outer flowers pistillate and fertile; inner flowers perfect hut often sterile 5. Pistillate flowers with tubular corollas, 6. Leaves alternateCOMPOSITAE, 15 7. Leaves pinmnately di ided; plants Maceose or sitky-tomentose Artemisia in Anthemideae, 7. Leaves narrow, entire; plants gummy, essentially glabrous. Gymnosperma in Astereae, 6. Leaves opposite Heliantheae (p. 24). 5, Pistillate flowers without corellas. 8. Receptacle naked. Anthemidece (p. 18) 8. Receptacle chaffy Heliansheae (p. 24) 4. Outer flowers perfect. 9. Receptacle without pales (naked), 10, Outer flowers with showy bilabiste corollas simulating ligule. ‘Microspermum in Eupatorieae, 10. Outer flowers not enlarged and simulating ligules. 11, Receptacle narrowly conic, 3-6 mm high; small annuals with finely dissected alternate leaves. Matricaria in Anthemideae, 11. Receptacle flit or somewhat convex; leaves opposite or alternate, toothed or entire, never finely dissected. 12, Flowers yellow; style-branches 0.7 mm long, not or scarcely exserted, truncate and glandular-papillose at tips; leaves opposite Flaveria in Heliantheae. 12. Flowers never yellow; style-branches elongate, exserted, filiform-clavate, ob- tuse, smooth at tips; leaves opposite or alternate. Eupatorieae (p. 21) 9. Receptacte with pales (chaffy).. 13, Style-branches filiform-clavate, long-exserted, obtuse at tips, minutely papillose but ‘not hairy; woody branches hollow with elliptic openings. Jaliscoa in Euparorieae, 13, Style-branches acute (0 attenuate, not long-exserted, hairy. Heliartheae (p. 24). Key F. Ligulate flowers none; pappus (at least in part) of awns or scales, or a crown or cup. 1, Heads 1-flowered (seldom 2lowered), aggregated into secondary capituta resembling ordinary heads. 2. Phyllaries fused into a tube resembling a 3~6-loothed calyx, Lagascea in Heliantheae. 2. Phyllaries distinct. 3. Phyllaries 8, in 4 decussate pairs; pappus of 56 awned scales 4,5-6.5 mm long, Vernonieae (p. 42). 3. Phyllaries 5, unequal, the 2 inner ones largest: pappus a laciniate crown less than 1 mm ong, Mexianthus in Eupacorieue, 1, Heads with several or many flowers or, if 1- or 2-flowered, not aggregated into secondary capitula 4, Receptacle without pales (naked) 5. Leaves and usually phyllaries with transhucent, usually conspicuous, linear to ellipsoid or globose oil glands Tageteae (p. 40). 5, Leaves and phyllaries without oil glands, sometimes with impressed or superficial resinous drops of exudate. 6, Phyllaries and flowers 5 each. Stevia in Eupatoriee. 6, Phyllaries more numerous; flowers 1-many, seldom 5, 7. Marginal corollas of the head bilabiate, the enlarged outer lobes simulating a ligule Microspermum in Eupatorieae, 7. Marginal corollas essentially regular, the outer lobes not conspicuously enlarged, 8. Pappus, at least im part, of awns, bristles, or bristle-tipped scales.16 FLORA NOVO-GALICIANA 8. Pappus of 2 awns, and squamellae; achenes strongly fattened. Perityle in Heliantheae. 9. Pappus of 4-10 or more numerous awns, bristles, or briste-tipped scales; achenes prismatic, commonly 4-5-angled. 10. Pappus of 4 broad squamelae alternating with 4 barbellate awns. Eutetras in Heliantheae, 10. Pappus of 5-10 bristies or bristle-tipped awns, or of numerous bristles, 11, Pappus of 5-10 bristles or bristle-tipped awns 12, Heads aggregated into secondary sessile clusters or head-like capitula. Vernonieae (p. 42). 12. Heads variously aggregated in branching inflorescences, not in secondary ccapitula. Eupatorieae (p. 21). 11. Bristles aumerous. 13, Leaves alternate Vernonia in Vernonieae. 13. Leaves opposite Liabum in Liabeae. 8, Pappus of scales only (none of them bristles or bristle-tipped), or coroniform. 14. Principal leaves divided or dissected; annuals. 15. Leaves alternate; flowers more than 100 in a head; receptacle narrowly conic, 3-6 mm high. Matricaria in Anthemideae, 15, Leaves opposite below, altemate above; flowers 40 or fewer; receptacle Mat or convex, Heliantheae (p. 24). 14, Leaves entire to dentate. Eupatoriene (p. 21) 4, Receptacle with pales (chatty) 16. Pappus of awns or bristles, with or without additionat seales or squametiae 17. Leaves alternate; flowers purple; pappus double. the outer series of fa linear scales, the nner of longer, linear, flattened bristles. Bolanosa in Vernoniene. 19, Leaves, at least the lower ones, opposite; Rowers yellow or white, if purplish the awas 2-3 only, or subequal in 1 series. Heliantheae (p. 24). 16. Pappus of scales only, or coroniform, without awns or bristles. 18, Style-branches filiform-clavate, Jong-exserted, obtuse at tips, minutely papillose but not hairy; inner phyliaries linear of lineay-lanceolate, mostly narrowly acute Eupatorieae (p. 21). 18, Style-branches short, acute to attenuate, with hairy tips; inner phyllaries oblong to ‘ovate, usually with rounded or obtuse tips Heliantheae (p. 24). Key G. Ligulate flowers none; pappus (at least in part) of capillary bristles. §, Leaves, and usually phyllaries, with translucent, usually conspicuous, linear to elliptic or globose cil glands, Tageteae (p. 40). 1, Leaves and phyllaries without translucent oil glands, sometimes with impressed or superficial sesinous drops of exudate, 2. Leaves opposite or verticillate, or the uppermost, and those of the inflorescence, alternate, 3. Achenes strongly flattened. 4, Receptacle naked. 5, Pappus bristles 1-3. Perityle in Heliantheae. 5. Pappus bristles 25-30 Barroetea in Eupatorieae. 4, Receptacle chaify. Spllanthes in Heliantheae. 3. Achenes prismatic, or terete to plumply biconvex. 6, Outer flowers with bilabiate corollas, the outer “lip" simulating 2 ligule Microspermum in Eupatorieae.COMPOSITAE a 6, Outer flowers essentially regular, the outer corolia-lobe not enlarged. 7. Achenes with 8-10 or more ribs 8. Pappus double, with an outer series of short scales. Liabun in Liabeae. 8. Pappus of a single series of bristles. Brickelia in Eupatoricae. 7. Achenes with 4-6 ribs (the summber very rarely varying within wider limits), never fattened. 9. Receptacle chaffy: pappus-bristles 8-15, very fragile and mostly falling as the achenes mature. 16, Perennial herbs with 4-angled stems; heads solitary or few, on peduncles 3-10 em long; flowers $0-200; anthers black ‘Melanthera in Heliantheae. 10, Shrubs, the woody branches hollow and with elliptic openings to the outside; ‘heads numerous in compact terminal clusters: flowers 20 or fewer; anthers pale. Jaliscoa in Eupatorieae. 9. Receptacle naked or, if chatty, the pappus-bristles persistent and more numerous. Eupatorieae (p, 21). 2. Leaves alternate (in Hofmeisieria crowded at flowering nodes and appearing verticillate), or basal 11, Heads entirely unisexual, or the outer flowers pistillate only, or neutral (without pistils). 12, Plants dioecious o potygamo-dioecious, at least the “staminate” heads unisexual, the “pistillate” heads sometimes with 1 or more perfect (but sterile) flowers ‘Astereae (p. 18), 12. Plaots not dioecious nor polygamo-dioecious, the heads all alike; outer flowers pistillate ‘or neutral, the inner perfect. 13, Receptacle densely hairy, the hairs 1.5 cm long; heads very large, up to $-10 em wide, the 40-50 marginal flowers sterile, neutral, with enlarged, showy, pink corolla Centaurea in Cardueae. 13. Receptacle naked, glabrous; heads much smaller, seldom more than } em wide, the ‘marginal flowers not enlarged, pistillate, fertile 14, Anthers caudate at base; style-branches truncate or rounded at the tip Inudeae (p. 35). 14. Anthers obtuse or somewhat auriculate at base; style-branches with short sterile appendages at tips. 15. Involucre more or less cylindric, subtended by a calyculus of smaller bracts; achenes about 10-ribbed Erechtites in Senecioneae. 15, Involucre more or less campanulate, without a subtending ealyculus; achenes biconvex, seldom linear or fusiform, then terete or with few inconspicuous angles. Conyza in Astereae. 41, Flowers all alike and perfect, the heads neither unisexual nor with marginal pistllate flowers. 16. Corollas more or less bilabiate, irregular, with the 3 outer lobes (occasionally 2 or 4 lobes) larger; anthers caudate at base. ‘Mutsieae (p. 37) 16. Corollas not bilabiate, regular; anthers various. 17. Pappus-bristles plumose. 18. Receptacle naked, glabrous or sometimes short-pubescent; plants not spiny. Brickellia in Eupatorieae. 18, Receptacle without pales, but densely hairy, the hairs often 1 cm long; plants more ‘or less spiny. Cirsium in Carducae. 17. Pappus bristles barbellate, not plumose. 19. Receptacle chaffy, most of the flowers subtended by pales: flowers 75 or more in a head Bolanosa in Vernonieae. 19. Receptacle naked, sometimes hairy, if chafly the flowers fewer than 20 in a head.18 FLORA NOVO-GALICIANA 20. Style-branches slender, tapering to narrow acute tips, pubescent on the outer surface; phyllaries imbricated in several series, the inner progressively longer (graduated); leaves pinnately veined, chiefly or entirely eauline, anthers with sagittate base: pappus double, of an outer series of short scales or bristles, and ‘a mmuch longer inner series; flowers never yellow. Vernonia in Vernonieae, 20, Style-branches various, not as above; phyllaries graduated or subequal; pappus ‘usually of a single ceries of bristles; leaves cauline, or in basal rosettes. 21. Anthers long-caudate at base. Perezia in Mutisieae. 21. Anthers rounded to auriculate at base. 22, Style-branches long-exserted, obtuse, glabrous, filiform or with thickened clavate tips; plants various; flowers never yellow. Eupatorieae (p. 21). 22, Style-branches included or shorily exserted, more of fess hairy at the blunt or ‘acute tips; flowers yellow, eream-color or ceddish purple. 23, Small gummy shrubs with oblanceolate leaves 1-3 em tong; phyllaries strongly graduated in 3-4 series; flowers yellow. Maplopappus in Astereae. 23, Plants herbaceous, sometimes up to 2-3 m high, or with all leaves basal; eaves broad and large, commonly more than $ cm long and wide; phyllaries subequal. Senecioneae (p. 37). ‘Tribe ANTHEMIDEAE Shrubs or herbs, often aromatic; leaves mostly alternate, often dissected; phyliaries imbricated, wholly scarious, or with scarious apex or margins; recepta- cle flat or convex, naked or chaffy; pappus of small scales, or wanting; anthers (at Teast in ours) obtuse to bluntly auriculate at base; style-branches truncate, the tips naked or with brushlike margins, the stigmatic surfaces in two marginal stripes. ‘About 100 genera and 1400 species, about three-fourths of these in the North- ern Hemisphere. Very few (¢.g.. various species of Artemisia) are native in the New World. 1, Heads radiate, the ligules often small 2. Receptacle chatfy; heads numerous in a dense terminal corymb; flowers white; peremniat her's ‘with dissected leaves and a strong aromatic odor. Achillea 2, Receptacle naked. 3. Receptacle conic or hemispheric. : Matricaria. 3. Receptacle fiat or slightly convex. Chrysanthemum, 1. Heads discoid. 4, Marginal flowers of the head without corollas, pistillate and fertile: achenes strongly ‘compressed. 5, Heads pedunculate; achenes of the marginal flowers pedicellate. Conula 5. Heads sessile; achenes of the marginal flowers sessile. Sotiva 4. Marginal flowers with corollas; achenes not compressed. 6, Heads solitary or few; disk-flowers more than 100, all perfect and fertile Matricaria 6. Heads very mumerous in a spikelike or branched and often nodding panicle; central (per fect) flowers fewer than 20, the marginal pistllate flowers few or many. Aniemisia Tribe ASTEREAE Perennial or annual herbs, sometimes shrubs, rarely small trees; leaves usu- ally alternate; phyllaries mostly imbricated in several series; receptacle usually naked: heads usually hetcrogamous, the outer flowers ligulate and pistillate, theCOMPOSITAE 9 central flowers tubular and bisexual (heads unisexual and the plants dioecious in a few American genera); pappus of bristles or scales or a short crown or, especially in North American genera, wanting; anthers obtuse at base, with a more or less triangular terminal appendage; style-branches (in the disk-flowers) flattened on the inner surface, tipped with subulate to triangular appendages covered with collecting hairs, and margined with stigmatic lines at base; achenes various, the epidermis always a single layer of cells thickened on three sides About 135 genera and 2500 species (according to J, Grau; chapter 19, in Heywood er al., 1977), predominantly extra-tropical |. Plants dioecious, or polygamo-dioccious (i.e., with some heads functionally staminate, others with some marginal flowers pistillate and fertile); rays absent or very short. 2. Plants potygamo-dicecious, the “pistillate” heads with one or more central staminate flowers. Archibaccharis, 2, Plants dioecious, the heads strictly unisexual Baccharis. 1. Planis neither dioecious nor polygamo-dioecious, the heads all alike and fertile; rays frequently present and conspicuous. 3. Rays and disk-flowers yellow, 4. Pappus none, or coroniform, sometimes a crown with 2 or 3 small scales. 5. Ray-corotlas small, equalling or shorter than the disk-flowers. Gymnosperma. 5. Rey-coreltas conspicuous, much longer than the disk-flowers, Xanthocephalur. 4. Pappus well developed (at least in the disk-flowers), not a short crown, 6. Pappus of flat narrow scales. [Sec Gutierrezia under Xanthocephalum.) 6. Pappus of hairs, bristles, or awas, 7. Pappus of the ray-flowers none, or a minute irregular ring 0.1~0.2 mm high; pappus of the disk-flowers of long bristles, Heterotheca, 7. Pappus of bristles or awns, essentially alike in ray- and disk-flowers 8. Pappus of 2 t0 12 fragile bristles or awns that soon fall from the achene. 9. Pappus of antrorsely barbellate soft bristles: phyllaries in about 3 series, united at the base; stipitate-glandular aquatic annuals, Otivaca, 9. Pappus (in our species) of entire (not barbeliate), bright white, stiff awns; phylla- Fies in 4 to 8 series, distinct ta the base; gummy biennials or perennials of upland habitats Crindelia 8. Pappus of more numerous and persistent bristles or hairs 10. Heads small and numerous, om one-sided branches in well-defined, usually pyra- ‘midal panicles; plants herbaceous Solidago, 30. Heads few and relatively large, if small and panicled then the plants shrubby, Haplopeppus. 3. Ray-flowers none, or not yellow, 11. Raysflowers none. 12, Pappus none, or a very short minutely toothed crown. Gymnosperma, 12. Pappus of bristles 13, Flowers all hermaphrodite. Heplopappus. 15: Outer flowers with filiform corollas, pistllate: plants herbaceous, mostly weedy; (see also “pistillate” plants of Archibaccharis, which are mostly shrubs). Conyza, 11. Ray-flowers present 14. Pappus none, or minute, én both ray- and disk-flowers. 45, Liguies ca 2 man tong; weedy, glandular-pubescent, lowland plants Egletes. 15, Ligules (3.5-) 6-10 (~15) mm long; plants of high plateaus and mountains20 FLORA NOVO-GALICIANA. 16. Achenes cohimnar, subterete to 4-angulate; receptacle depressed-hemispherie to conical Aphanustephuss. 16, Achenes laterally compressed. 17. Receptacle flat or slightly conver; rays 40 oF mor Erigeron 17. Receptacle conical; rays 10 10 35. Astranthium 14, Pappus well developed, at least in the disk-flowers: 18. Pappus in the ray-flowers none, or greatly reduced, and much shorter than that of the disk-flowers, 19, Pappus of the ray-flowers none; plants annual, glandular-puberuleat, Aster. 19, Pappus of the ray-lowers 0.5-1 mm tong, that of the disk-flowers 1.5-2.5 mm; ‘small narrow-leaved strigose perennials, to be expected in the extreme northeast ‘em part of Nueva Galicia. Townsendia. 18. Pappus present and essentially alike in ray- and disk-flowers. 20, Pappus of S (4-9) awns and as many scales. Chaetopappa. 20, Pappus of capillary bristles or (rarely) a short crown with bristle-tipped lobes. 21. Rays inconspicuous, the ligule shorter than the corofla-tube and scarcely if at all exceeding the pappus. 22. Marginal pistillate Mowers very many (often 100 or more) and several times more numerous than the perfect flowers at the center of the head; plants mostly woolly or glandular-pubescent. Conyze, 22, Pistillate flowers (rays) 30-40 in 1 to 2 series, hardly more numerous than the flowers; @ narrow-leaved, glabrous plant. Aster, 21, Rays conspicuous, the ligule longer than the tube and exceeding the pappus. 23. Involuere evidently graduated, the phyllarics in several series, the outer shorter; achenes not or scarcely compressed, commonly about S-ribbed: Iigules 15~40 (~50); style-appendages relatively long, acute, deltoid to sagit- tale or subulate. ‘Aster. 23. Involuere not evidently graduated, the phyllarics commonly subequal; achenes ‘stially compressed, with 2 (rarely 4) prominent nerves; ligules (40-) 50- 200; style-appendages relatively short, obtuse or deltoid and subacute Enigeron ‘Tribe CARDUEAE Perennial or annual herbs; cauline leaves alternate, often pinnatifid, incised, or prickly; basal leaves often present and conspicuous; heads usually discoid, phyllaries in several series, imbricated, mostly ending in a spine or appendage; receptacle commonly bristly or hairy; pappus mostly of multiseriate bristles; anthers appendaged at apex, auriculate to caudate at base; style-branches flat, elliptical, often united and diverging only at tips, velvety outside, with a promi- nent circular and often hairy thickening below them. ‘About 65 genera and 2000 or more species (according to M. Dittrich, chapter 36 in Heywood et al,, 1977). Most of the genera and species are found in the Mediterranean region and in southwestern Asia. Only Cirsium and Centaurea are represented by indigenous species in North America. 1, Flowers all perfect, the outer ones not conspicuously different; achenes basifixed; leaves prickly ‘margined; phyllaries with prickly tips and often spine-tipped lobes; pappus plumose. Cirsium 1. Marginal flowers neutral but showy, their corollas about twice as long as those of the central flowers; achenes obliquely of laterally attached: leaves not prickly-margined; phyllaries with dark, peetinately lobed tips: pappus barbellate Centaurea.COMPOSITAE, a Tribe EUPATORIEAE References: Robinson, B. L. A generic key to the Compositae-Eupator- jeae, Proc. Amer. Acad. 49: 429-437. 1913. King, R. M., & H. Robin- son. Studies in the Eupatorieae (Asteraceae). C [100]. A key to the genera of Nueva Galicia, Mexico. Phytologia 24: 267-280. 1972. Mostly herbs or shrubs, sometimes small trees or vines; leaves mostly oppo- site or in some genera the uppermost or all alternate; blades pinnately or palmately veined, entire to lobed or strongly dissected; involucre uniseriate to multiseriate, commonly imbricated; phyllaries herbaceous to coriaceous, usually distinct; receptacle flat to hemispherical or conical, usually naked (without chaff, or pales); heads discoid, the flowers perfect and fertile; corollas 5 (~4)-lobed, rarely irregular, white, pink, blue, purple, or red, never yellow: pappus of scales. awns, or bristles, or coroniform or wanting; anthers 5 (~4), the appendages vesti- gial to elongate, the bases rounded or hastate, never caudate: style-branches well-developed, the stigmatic surface of two distinct lateral lines toward the base, the appendages commonly exserted, filiform or clavate, terete of flattened, often colored like the corollas; achenes prismatic or flattened, usually 5- or 10-ribbed, commonly with a distinctive sterile base (carpopodium). A diverse tribe of more than 2000 species, mostly of the New World (H. Robinson & R. M. King, in Heywood e¢ ai., 1977, chapter 15). Well-known and Gistinctive genera of more than 100 species each, such as Brickellia, Mikania, and Stevia, are represented in Mexico by numerous species. What was traditionally understood to be the largest genus of the tribe, by far, namely Eupatorium, with more than 1000 species, has been subjected in recent years to an intensive and frenetic scrutiny. This has led to the production of new and revived genera, new species, and new names, to an extent unmatched in modern times. In the opinion of R. M. King & H. Robinson, as expressed in more than 200 publications between 1967 and 1981, the name Eupatoriu is to be restricted to a group of 37 38 species in the North Temperate Zone, and the very numerous tropical Ameri- can species formerly referred to Eupatorium are all to be transferred to other genera, The names applied to our species by King & H. Robinson are listed below where appropriate, with comments as necessary. A fuller discussion of the work of these authors appeared in 1982 (Contr. Univ. Michigan Herb. 15: 181— 190). 1 am indebted to them (King & H. Robinson, 1972) for suggestions as to the best method of contrasting the genera in the key that follows, 1. Individual heads with 1 (rarely 2) Gowers, aggregated into spherical glomerules resembling ‘many-fowered heads, 2. Pappus a crown of 5-7 firm white opaque scales 0.5 mun long: achenes evidently narrowed Deneath the pappus Mexianthus. 2. Pappus of about 30 coarse bristles; achenes nat strongly narrowed distally, Eupatorium monanthwm. 41. Individual heads with 3 or more flowers, solitary or variously grouped or clustered, 3. Marginal flowers of the disk with enlarged 3-parted outer lip simulating a ligule, and a much smaller bilobed inner Kp, Micraspermum, 3. Disk-flowers regular, without enlarged outer lip. 4. Achenes without pappus, of the pappus a short, scarious, entire of lobulate crown, if of bristles these only 0.1-0.2 mm long,