'JsA : Ref. HAROLD ROBINSON SERIES PUBLICATIONS OF THE SMITHSONIAN INSTITUTION Emphasis upon publication as a means of “diffusing knowledge” was expressed by the first Secretary of the Smithsonian. In his formal plan for the Institution, Joseph Henry outlined a program that included the following statement: “It is proposed to publish a series of reports, giving an account of the new discoveries in science, and of the changes made from year to year in all branches of knowledge.” This theme of basic research has been adhered to through the years by thousands of titles issued in series publications under the Smithsonian imprint, commencing with Smithsonian Contributions to Knowledge in 1848 and continuing with the following active series: Smithsonian Contributions to Anthropology Smithsonian Contributions to Astrophysics Smithsonian Contributions to Botany Smithsonian Contributions to the Earth Sciences Smithsonian Contributions to the Marine Sciences Smithsonian Contributions to Paleobiology Smithsonian Contributions to Zoology Smithsonian Folklife Studies Smithsonian Studies in Air and Space Smithsonian Studies in History and Technology In these series, the Institution publishes small papers and full-scale monographs that report the research and collections of its various museums and bureaux or of professional colleagues in the world of science and scholarship. The publications .are distributed by mailing lists to libraries, universities, and similar institutions throughout the world. Papers or monographs submitted for series publication are received by the Smithsonian Institution Press, subject to its own review for format and style.. only through departments of the various Smithsonian museums or bureaux, where the martuSCripti are given substantive review. Press requirements for manuscript and art preparation are outlined on the inside back cover. Robert McC. Adams Secretary Smithsonian Institution SMITHSONIAN CONTRIBUTIONS TO BOTANY • NUMBER 64 A Palynological Study of the Liabeae (Asteraceae). Harold Robinson and Clodomiro Marticorena SMITHSONIAN PUBLICATIONS ' *7” •' 7 ‘ > "O -tf s'lis h *• •» '< 1 SMITHSONIAN INSTITUTION PRESS City of Washington 1986 ABSTRACT Robinson, Harold, and Clodomiro Marticorena. A Palynological Study of the Liabeae (Asteraceae). Smithsonian Contributions to Botany, number 64, 50 pages, 208 figures, 1986. — Surface features and internal structure of the spines in the pollen of the Liabeae are illustrated by scanning electron microscopy and drawings made from oil immersion light microscopy. Obser- vations are presented for all 16 genera of the tribe, Paranephelius and Pseudonoseris of the Paranephelinae, Astroliabum, Bishopanthus, Cacosmia, Chionopappus, Ferreyranthus, Liabellum, Liabum, Microliabum, Oligactis, and Sinclairia of the Liabinae, and Chrysactinium, Erato, Munnozia and Philoglossa of the Munnoziinae. No examples of echinolophate pollen are found in the tribe, but irregular groupings of spines are found to be characteristic of all but Munnozia and Chrysactinium in the Munnoziinae and Paranephelius and Pseudonoseris in the Paranepheliinae. The distribution of caveate forms in the Liabeae is clarified. The presence of a large cavus-like space interrupted by thin basal columuellae is confirmed in the genus Paranephelius, correcting early observations showing the genus as an exception to the basic uniformity of exine structure in the tribe. However, subcaveate forms have been discov- ered in the tribe in the genus Munnozia. SEM study of broken tecta in the genus Munnozia show previously unsuspected great potential for variation between closely related species and species groups. The structure of the exine supports views derived from floral structures regarding relationships with other tribes. No support is seen for a phyletic position near the Senecioneae, and closest similarity is seen to the Vernonieae. A basic difference is discov- ered in the arrangement of the basal columellae between the Liabeae and the Vernonieae, being clustered or forming a hollow cylinder under each spine in the former but forming a single solid basal columella under each spine in the most nearly comparable members of the latter. Genera of the Vernonieae shown for comparison include the non-lophate Blanchetia, Pseu- dostifftia, and Vernonia, and the lophate Bishopalea, Heterocypsela, and Phyllo- cephalum. Official publication date is handstamped in a limited number of initial copies and is recorded in the Institution’s annual report, Smithsonian Year. Series cover design: Leaf clearing from the katsura tree Cercidiphyllum japonicum Siebold and Zuccarini. Library of Congress Cataloging in Publication Data Robinson, Harold Ernest, 1932- A palynological study of the Liabeae (Asteraceae). (Smithsonian contributions to botany ; no. 64) Supt. of Docs. no. SI 1.29:64 Bibliography: p. 1. Compositae — Pollen — Morphology — Atlases. 2. Compositae — Classification. 3. Palyno- taxonomy. I. Marticorena, Clodomiro. II. Title. III. Title: Liabeae (Asteraceae) IV. Series. QK1.S2747 no.64 [QK495.C74] 581s [583'. 55] 86-600032 Contents Introduction Materials and Methods Acknowledgments Pollen Structure of the Liabeae Paranephf.liinae Liabinae Munnoziinae Key to Types of Liabeae Pollen Palynological Evidence on the Relationships of the Liabeae in the Asteraceae Appendix Literature Cited Figures Page 1 2 3 3 4 5 6 9 10 14 21 23 A Palynological Study of the Liabeae (Asteraceae) Harold Robinson and Clodomiro Marticorena Introduction A recent series of studies (Robinson and Bret- tell, 1973a, 1974; Robinson, 1978, 1983a) has explored the taxonomy and evolution of the geographically restricted neotropical tribe Lia- beae, using mostly macroscopic and some micro- scopic structural features. Most recently the chromosome numbers of the tribe have been reviewed (Robinson et al., 1985). The present study of the pollen continues the effort to im- prove understanding of this previously most poorly understood tribe of the Asteraceae whose members were neither placed together nor rec- ognized at the tribal level in the traditional treat- ments of the family. A summary of the pollen characters of the tribe is given based on exami- nation of over half the species by at least oil immersion light microscopy. Generic patterns are noted, and detailed taxonomically significant variations in exine structure in closely related species groups are documented. The views pre- sented in the recent studies by the senior author regarding relationships and distinctions between the Liabeae and other tribes such as the Seneci- oneae and Vernonieae are extended by knowl- edge of the pollen structure. Harold Robinson, Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Clodomiro Marticorena, Departamento de Botdnica, In- stitute de Biologia, Universidad de Concepcion, Chile. The Liabeae was not recognized at the tribal level or properly placed together in a related group until this century (Rydberg, 1927; Blake, 1935; Sandwith, 1956; Cabrera, 1954; Robinson and Brettell, 1973a, 1974; Robinson, 1978, 1983a). In the earliest classifications the mem- bers of the tribe with a capillary pappus were placed in the Vernonieae (Cassini, 1828). Later these were placed in the Senecioneae (Bentham and Hooker, 1873) while genera with reduced or plumose pappus forms were described in var- ious other tribes such as the Heliantheae, Helen- ieae, and Mutisieae. The history of the tribe is dealt with more completely by Robinson ( 1 983a). The restricted neotropical distribution of the Liabeae and the failure to treat the group as taxonomically distinct resulted in their omission from the earliest palynological studies of the Asteraceae. It was in the study by Stix (1960) that the group was first included with observa- tions on 14 species that are presently considered members of the genera Liabum, Austroliabum, Liabellum, Munnozia, Sinclairia, and Paranephel- ius. The study by Stix cited some variation in internal structure of the tectum in the pollen of the tribe. In 1966, the pollen of Cacosmia was described and compared with that of Liabum by Skvarla and Turner. Light microscope oil im- mersion observations of thick sections of Chion- opappus pollen were used by Marticorena and Parra (1974) to confirm the relationship of that 1 2 genus to Liabum. Regularity of spine distribution and some variation in size of grains were men- tioned by Robinson and Brettell (1974). The results of transmission electron microscope stud- ies were included in the review of the family by Skvarla et al. (1977) where apparent variations between non-caveate forms with basal columellae and forms without basal columellae were noted in the tribe. The apparently anomalous Parane- phelius was shown to possess thin basal columellae by Feuer and Dillon in an oral presentation (1982). It remains for the present study to broaden the survey of the tribe to include all 16 genera, introduce results of scanning electron microscope observations of whole and broken pollen grains, and to compare the results with conclusions from the simultaneous phyletic stud- ies of the tribe (Robinson, 1983a). Materials and Methods A variety of techniques have been used for the study of pollen of the Liabeae. Many initial ob- servations have been made of material mounted in Hoyer’s solution (50 cmJ distilled water; 30 grams Gum Arabic, U.S.P. Flake; 200 grams Chloral hydrate; 20 cm'1 Glycerin). The medium is regarded as temporary, eventually altering the index of refraction and making the exine essen- tially invisible. Whole grains have also been mounted in glycerin jelly. Paraffin sections have been prepared by Marticorena. All the above material was viewed with the light microscope with X40 or oil immersion objectives. Results were comparable in detail to those by Stix ( 1 960). The sections, which are often fragmentary and folded, have been used in the present study pri- marily for reference and only one photograph of a section of Chionopappus prepared during the study of Marticorena and Parra (1974), is pre- sented herein. The slides of whole grains were used by Marticorena to produce ink drawings representing the internal spine structure in ver- tical view of over 100 species. These were grouped to show a range of columellar variation under different spines in grains from single col- SMITHSONIAN CONTRIBUTIONS TO BOTANY lections. Representatives (Figures 1-40), re- drawn from the Marticorena camera lucida draw- ings, are included herein to provide the reader with a more convenient survey of the variation of such structure in the tribe than is available in the SEM micrographs. No observations have been made in this study with the transmission electron microscope, but data from previous TEM studies by Skvarla et al. (1966, 1977) and Tomb (1975) are cited. In spite of some artifacts produced by breakage, the use of broken grains has proven much superior to TEM preparations for obtaining an understand- ing of the three-dimensional columellar struc- ture. A primary effort in the present study has in- volved use of the scanning electron microscope (SEM). Pollen samples have been removed from herbarium material or in one case from a living plant. In a few cases the pollen was not acetolized, but most samples were acetolized according to the procedure outlined in Erdtman (1966). Ace- tolyzed grains were fractured by crushing with the edge of a glass slide, or more often while still in the centrifuge tube by using a needle. All samples were pipetted on to cover slips and al- lowed to dry. Tbe samples on coverslips were thinly coated using an EFFA carbon coater, in- volving evaporation by the carbon-arc technique. They were then coated with gold-paladium using a DC sputtering unit. Specimens were examined and photographed with Cambridge Stereoscan MK 1 1 A, Cambridge Stereoscan S4-1 0, or Coates and Welter field emission microscopes. The terminology for pollen structure used in the present paper follows mostly that of Skvarla et al. (1977). Sizes of pollen grains are given over a broad range to account for variations due to method of preparation. Grains mounted in Hoy- er’s solution tend to swell most, those in glycerine jelly are about the same size or slightly smaller than those in Hoyer’s, the grains in water are distinctly smaller, and air- or vacuum-dried ma- terial, such as that examined by SEM, is com- pletely unexpanded and may be as small as % the Hoyer’s Solution diameter. Actual overlap in NUMBER 64 3 sizes in various genera should be less in any one medium than indicated by the general figures given. Measurements cited for the Liabeae pollen are of the equatorial diameters, but none of the grains of the tribe are sufficiently oblong or oblate in shape to produce a markedly different polar diameters. Since only the general sizes are mentioned for the Vernonieae the shorter polar diameters of the larger oblate grains of that tribe are not noted. Measurements in the study ex- clude the projecting parts of the spines. Speci- mens used in the study are listed in the appendix. Acknowledgments. — The SEM micro- graphs were prepared by the SEM Laboratory (supervised by Walter Brown) at the National Museum of Natural History (NMNH), Smithson- ian Institution. The microscopes were operated by Mary Jacque Mann and Susann Braden. Dr. Joan Nowicke is to be thanked for use of the facilities of the Palynological Laboratory in the NMNH Department of Botany, and for sugges- tions and help in the traditional methods of breaking pollen grains. Janice Bittner of the Pa- lynology Laboratory and Andrea Sessions helped by processing the pollen. Brian Kahn has helped in preparation of the plates. Pollen Structure of the Liabeae Stix (1960) in the initial study of pollen of the Liabeae, recognized three variations in the tribe. The first or andromachioides-type has large col- umellae restricted to the area under the spines, and most closely agrees with the type termed “Liaboid” by Skvarla et al. (1977). The species listed with this pollen type by Stix include five species now placed in Sinclairia, two species now placed in Liabellum, two species of Austroliabum, and Munnozia rusbyi. The second pollen type of Stix (1960), the umbellata-type, has smaller columellae under each spine and some basal columellae in the intervening parts of the exine. The three species cited by Stix include one now known as Munnozia hastifolia and the only two true Liabum species in her study. The pollen type is treated herein as a comparatively minor variation of the first type. The different examples seem to occur in species that are closely related to others of the andro- machioides-type with larger columellae under the spines. The remaining pollen type of Stix (1960), the ovatum-type, has the columellae partially to com- pletely separate from the footlayer. The colu- mellae have the form seen in the upper layer of the first and second types. This pollen type is cited by Stix from Liabum ovatum, which is a member of the genus Paranephelius. The type was examined by Skvarla et al. (1977) on the basis of a specimen presently determined as P. uniflora, and the suggestion of a cavus and the more senecoid rather than anthemoid nature was emphasized. The apparent anomally of a sene- coid pollen type in the Liabeae was partially resolved in the study by Eeuer and Dillon (1982), which discovered slender columellae under the spines that connect with the foot layer. The present understanding of the pollen struc- ture in the Liabeae is reviewed herein according to the revised concept of the tribe with 1 6 genera distributed in three subtribes, Paranepheliinae, Liabinae, and Munnoziinae (Robinson, 1983a,b). The pollen of the Liabeae (Figures 1-180) is generally characterized as follows: Tricolporate; 25-50 ^m in diameter, spherical to slightly oblate-spheroid or in some cases where colpi are closed appearing oblong; ecto-apertures meridionally elongate (Figures 41, 50, 69, 74, 104, 124) and endo-apertures transversely short- oval (Figure 121); echinate with spines often irregularly distributed and variously confluent with each other at the base (Figures 60, 62, 64, 67, 94), sometimes regularly dispersed on the intercolpar area (Figures 49, 50, 140, 141, 147, 148, 150). Internal structure basically with two levels (Figures 68, 88, 132, 179, 180) of colu- mellae. Usually with few (Figures 1, 2, 46) to many (Figures 156, 160) large basal columellae clustered under spine bases, rarely with addi- tional basal columellae dispersed in inter-spinal areas (Figure 118), columellae under spines sometimes coalesced into hollow cylinders (Fig- 4 ures 175, 176); basal columellae sometimes re- duced or separated from foot layer (Figures 162- 167) with partial to complete development of cavus; upper layer of columellae usually regular, short, dense, with upper ends forming a perfo- rate tectum, and covered below by unsupported or partially supported internal tecta forming a somewhat undulating or papillose surface in in- terspinal areas. Pollen size and spine distribution show varia- tion in the Liabeae. The generic pair of Liabum and Oligactis has the smallest average pollen size in the tribe at 25-35 /im. A number of other members of the subtribe Liabinae, such as Cacos- mia and Ferreyranthus, can have pollen as small, 27-30 /im and 29-35 respectively, but one species of the latter ranges up to 43 /im. Still, in other Liabinae, such as Austroliabum, 33-40 nm, Chionopappus, 37-40 /xm, and especially Sin- clairia, 35-55 fim and Liabellum, 47-52 /x m , the pollen is characteristically larger. The members of the other two subtribes are more consistent in size. Both genera of the Paranepheliinae have pollen mostly 35-45 /xm in diameter. The four genera of the Munnoziinae, with rare exceptions, range from 30-40 /xm in diameter. The primary exception is the typical element of Munnozia that ranges between 37-45 /um in diameter. For a sampling of size measurements see the appendix. Many genera of the Liabeae have a type of uneven distribution of the spines on the pollen that seems unique in the family. In these forms the spines seem to be in groups, sometimes in pairs or in short crests of three with confluent bases. This is especially evident in SEM views of some unacetolized grains (Figures 62, 63). Such grains seem to be best characterized by the irreg- ular placement of the spines along the lateral margins of the colpi (Figures 72, 74). This pollen form is found throughout the subtribe Liabinae and in Erato and Philoglossa of the Munnoziinae, but it is lacking in Munnozia and Chrysactinium of the Munnoziinae and in the Paranepheliinae. Fhe distribution in the tribe suggests that is the plesiomorphic condition in the tribe. It is corre- lated rather closely in its distribution in the tribe SMITHSONIAN CONTRIBUTIONS TO BOTANY with the apparent plesiomorphic condition of the basal columellae under the spines. The spines are more evenly disposed on the surfaces of the pollen in two pairs of genera in two separate, comparatively derived groups, Munnozia and Chrysactinium in the Munnoziinae, and Paranephelius and Pseudonoseris in the Para- nepheliinae. This arrangement is reflected in the regularly aligned spines along the lateral margins of the colpi (Figures 41, 50). The pattern is more evident in a preparation of unacetolized grains of Paranephelius (Figure 49), where the depres- sions between the bases of each of the spines are equal in spite of some displacement seen in the spines. The character can be observed under the light microscope. The separate and derived na- ture of the regular spine pattern in the two groups is indicated by the retention of the irreg- ular pattern in two of the genera of the Munno- ziinae and by the slight evidence of irregularity that survives along the margin of the colpus in Paranephelius. There is greater variation in the Liabeae in the internal structures of the pollen than in ex- ternal features. The variations between many species of Munnozia are perhaps the greatest thus far known for internal exine structure of pollen in any such closely related group. Variations include tendencies toward the caveate condition seen in the Paranepheliinae, Philoglossa, and Munnozia, the fusion of columellae into hollow cylinders in Chrysactinium and Munnozia, and the production of an anomalous inner series of col- umellae seen in Munnozia subgenus Kastnera. These and other details of pollen structure in the Liabeae are best reviewed in each subtribe indi- vidually. The subtribes are treated herein in the following order: Paranepheliinae, Liabinae, and Munnoziinae. Paranepheliinae. — The subtribe contains two genera that share long style branches, pale anthers, elongate raphids in the achene wall, narrow disk corollas, and leaves mostly in basal rosettes. The inflorescence has few heads and in the type-genus the heads are sessile in the rosette of leaves. The two genera, Paranephelius (Figures NUMBER 64 5 1, 2, 41-49) and Pseudonoseris (Figures 50-53), have pollen structure that supports their status as a related pair. Both have grains that appear caveate, with spines rather evenly distributed over the intercolpar regions. The illustrations of Paranephelius (as Liabum ovatum) by both Stix (1960, Abb. 1 1) and Skvarla et al. (1977, pi. 15: fig. F) are equivocal regarding complete separa- tion of the outer exine from the footlayer, and their preparations seem to have a columellate layer of the exine tightly contracted against the footlayer. A point in the TEM of Skvarla et al. (1977, pi. 15: fig. F) seems to show many colu- mellae closely appressed against the foot layer and at least one clear example of a basal attach- ment. However, closer examination of the most obvious basally attached element shows that it is not directly attached to the overlying tectum. Preparations in Hoyer’s Solution, thick sections viewed with the light microscope, and SEMs of broken grains indicate that the columellate exine is normally less close to the foot-layer, and that a cavus-like area is present in the intercolpar parts of the grain. Studies of Paranephelius with oil immersion (Figures 1, 2) and SEM (Feuer and Dillon, 1982; Figures 45-48) show that the grains are not strictly caveate. They have a few narrow basal columellae under the spines that reach the foot layer. The SEM also shows occa- sional short pegs on the foot layer, but it is most likely that the basally attached structure seen in the TEM photo of Skvarla et al. (1977) is part of one of the slender columellae that has been bent and cut. Broken pollen grains of Pseudonoseris have not been examined with the SEM, and oil immersion study does not show columellae of the type found in Paranephelius. If basal columellae are present in Pseudonoseris, they are more reduced than those of Paranephelius, which can by seen by light microscopy. The two genera, nevertheless, both show a tendency for great reduction in the basal columellae in pollen grains of comparatively large size. In its vegetative or floral characters, Pseudo- noseris seems less specialized than Paranephelius, but the pollen seems more specialized. It is Par- anephelius rather than Pseudonoseris that retains some irregularity of spine position and has more obvious remnants of basal columellae under the spines. The presence of slender basal columellae un- der the spines of Paranephelius pollen conforms to the anthemoid pattern of typical Liabeae and eliminates doubts about the senecioid pollen pat- terns or Senecionean relationships of the tribe. Nevertheless, the extremely slender form of the basal columellae represents a trend toward cavus development that is interpreted herein as apo- morphic within the tribe. The trend toward ca- vus development here and in some Munnoziinae indicates at least two separate instances of reduc- tion in basal columellae in the Liabeae. Similarities between the pollen of the Parane- pheliinae and Munnoziinae might indicate close relationship between the two, but this is not supported by other evidence. In fact, on the basis of such characters as style form, anther color, endothecial cells, and raphids of the achene wall, the two subtribes are at opposite extremes in the tribe. The surviving slight irregularity of spine distribution along the margin of the colpus that is evident in Paranephelius (Figure 41) supports the idea that the regularly disposed spine pattern of the Paranepheliinae arose from members of the tribe with more irregularly disposed spines on the pollen within the immediate relationship, and separately from that of the Munnoziinae. Eiabinae. — Members of the subtribe share pale anthers with fringed bases, narrow bases on the disk corolla throats, and leaves rarely or never in rosettes. The heads are often on short peduncles in crowded inflorescences. The style branches and the raphids in the achene wall may be long or short. The subtribe consists of ten genera, Austroliabum (Figures 3, 69-73), Bishop- anthus (Figures 54-57), Cacosmia (Figures 4, 58- 63), Chionopappus (Figures 5, 64-68), Ferreyr- anthus (Figures 6, 7, 78-82), Liabellum (Figures 11, 94-98), Liabum (Figures 12-20, 108-123), Microliabum (Figures 74-77), Oligactis (Figures 21-23, 99-107), and Sinclairia (Figures 8-10, 6 83-93), ranging north to central Mexico and the Greater Antilles and south to northern Argen- tina. In all these genera the spines are unevenly distributed in small subgroups on the pollen sur- face (Figures 60, 62-65, 67, 69, 70). As far as known, this type of arrangement is unique to the Liabeae but apparently plesiomorphic in the tribe. The pattern is reflected in cross-sections (Figure 68; Skvarla et ah, 1977) by the irregular thickness of the intercolpar exine and by the confluence between the bases of the adjacent spines. Size of the pollen grains, in contrast, is highly variable in the subtribe (Appendix). Size of pollen grains seems to be reflected in the subtribe in the size of the columellae, smallest in Liabum and Oligactis (Figures 12-23), and largest in Sinclairia and Liabellum (Figures 8, 10, 11). The two closely related genera with the smallest pollen grains, Liabum and Oligactis (25-35 jim), are also distinctive in the shorter stouter tips of the spines. This is evident in both the Andean (Figures 99-107) and West Indian (Figures 108- 111) material. The SEMs of broken pollen in Austroliabum, Ferreyranthus, Liabum, and Sinclairia, and sec- tioned grains of Chionopappus (Marticorena and Parra, 1974), all indicate a distinct group of large basal columellae under the spines. Nevertheless, oil immersion studies by Marticorena (Figures 3, 7, 11) show variations toward a fused hollow cylinder in each of these genera. The fusion might be the result of poorer resolution or of the vertical direction of the observation in the oil immersion studies. Still, the pollen in one SEM preparation of Microliabum (Figure 77), which is closely related to Austroliabum, may be inter- preted as an example of a fused hollow cylinder. There is no other SEM evidence of the trait in the subtribe. Fhe pollen form called the umbellatum-type by Stix (1960) has been examined in Liabum (Figures 12, 13, 18, 20, 118, 123) using SEM preparations. The pollen type differs from the more common form in the tribe by the basal columellae being somewhat smaller and more numerous, and extending into the areas between SMITHSONIAN CONTRIBUTIONS TO BOTANY the spines. Although named after Liabum umbel- latum (L.) Schultz-Bipontinus of Jamaica, the character does not occur in that species. The character does occur in species in Hispaniola that have sometimes been included in a broader con- cept of L. umbel latum. Two species from Hispan- iola have been examined. In L. subacaule Ryd- berg the columellae under the spines become narrower distally while those between the spines are often constricted at the base (Figure 1 1 8). There are also a few projections from the under surface of the internal tectum among the inter- spinal columellae. In L. barahonense Urban the supposed columellae between the spines are seen to be mostly projections from the under surface of the internal tectum and they are not connected to the foot-layer (f igures 122, 123). From the evidence, the umbellatum-type pollen in West Indian Liabum has columellae of basically differ- ent structure in the interspinal areas from those under the spines. This particular design of colu- mellae is probably restricted to the genus Liabum. The umbellatum-type pollen was also credited to Munnozia hastifolia by Stix (1960, as Liabum), but the latter seems to be different in the details of its structure. Most of the remaining genera of the subtribe have massive columellae clustered under the spines, but the survey of material under oil im- mersion by Marticorena has revealed a variant in Sinclairia deamii (B.L. Robinson and Bartlett) Rydberg having only small columellae (Figure 9). Examination with the SEM shows these columel- lae to be further distinguished by their common failure to reach the foot-layer (Figures 91, 92). The variant seems to represent an extreme re- duction of a type that is rare in the subtribe Liabinae, though it does occur commonly in the Munnoziinae. As interpreted herein on the basis of the exposed unbroken basal columellar sur- face, the SEM photograph of Microliabum (Figure 77) represents another example in the subtribe of basal columellae that are incompletely fused to the foot-layer. Munnoziinae. — The subtribe contains four genera, Erato, Philoglossa, Chrysactinium, and NUMBER 64 7 Munnozia that share short style branches, black anthers, quadrate raphids in the achene wall, somewhat expanded bases in the throats of the disk corollas, and a tendency for long-peduncu- late heads. Nevertheless, for purposes of discus- sion of the pollen, the subtribe falls into two distinct subgroups consisting of two genera apiece. On the basis of pollen alone, the two subgroups would not be considered closely re- lated. Erato (Figures 24, 124-133) and Philoglossa (Figures 25, 26, 134-139) both have the irregu- larly clustered pattern of spines on the pollen surface (Figures 124-127, 134-1 36) that is char- acteristic of the larger subtribe Liabinae. Thick sections and whole grains examined with oil im- mersion, and broken grains viewed with the SEM, show that Erato further resembles the Lia- binae by the large clustered basal columellae under each spine (Figures 24, 129, 131, 132, 133), all firmly attached to the foot layer. The presence of such pollen in Erato in the subtribe Munnoziinae strongly indicates that the form is plesiomorphic in the tribe. In Philoglossa the pollen is externally the same as Erato; however, oil immersion and SEM stud- ies show that the basal columellae under the spines are reduced in size and number. Such reduction is independent of the size of the grain that is the same size as in Erato. The oil immer- sion and SEM studies show that basal columellae are present (Figures 25, 26, 138, 139), a fact demonstrated also by Feuer and Dillon (1982). Still, the SEM preparations indicate that the col- umellae are weakly attached to the foot-layer and that a near-caveate condition exists (Figures 1 38, 1 39). The broken grains show a large expanse of papillose foot-layer with little evidence of colu- mellar scars. Philoglossa is the only member of the tribe characteristically combining such re- duction of the basal columellae with an irregular disposition of spines on the surface. The remaining two genera of the subtribe, Chrysactinium (Figures 27, 28, 140-146) and Munnozia (Figures 29-40, 147-180) have spines regularly distributed on the surface of the pollen with no merging of spine bases. Internally the columellae show a strong tendency for reduction, although a number of species (Figures 151-160) including the type of the genus, M. lanceolata, show the massive type seen in Erato and the Liabinae. Nevertheless, partial or complete col- umellae are clearly present in all members of the two genera that have been examined by light microscope or SEM. In this respect the pollen differs from that found in the Paranepheliinae. As seen in SEMs, however, there is a capacity in Munnozia for the most complete separation of the columellae from the foot-layer and thus the most technically complete cavus development in the tribe. In the Chrysactinium-Munnozia complex, the detailed structure of the columellae and tecta shows great variation, and the variations seem to correlate well with taxonomic groups within a genus. As such, the variations are of more tax- onomic value than variations in other parts of the tribe. In Chrysactinium, according to oil im- mersion studies (f igures 27, 28), the basal colu- mellae vary from separate strands to a single hollow cylinder, but they are rather small and regularly arranged. An SEM view shows that the columellae connect with the foot-layer (Figures 144, 145). The examples seen in the SEM prep- aration are all fused, with the hollow center evident in one view (Figure 146). The form of the columellae in the genus is not precisely matched in size, symmetry, attachment to the foot-layer, or variable dissection in any member of Munnozia. Munnozia itself has many different subgroups that can be distinguished by the internal struc- ture of the pollen walls. The type-species, Mun- nozia lanceolata Ruiz and Pavon (Robinson and Brettell, 1974), and a series of close relatives, M. cardenasii, M. foliosa, M. olearioides, M. peruensis, andM. rusbyi, all having rather coriaceous leaves, appear to have a characteristic cluster of large, separate, basal columellae in a ring under each spine (Figures 29, 33, 151-155). No members of the group have the columellae completely fused into a hollow cylinder. In this typical group, the 8 columellae, as observed in thick section and SEM, are attached to the foot-layer. A singularly dis- tinctive species, M. perfoliata, also has columellae grouped and unfused under the spines (Figure 156), but these columellae are less massive and are sometimes mixed with smaller columellae inside the circle. The even smaller circle of col- umellae in M. lyrata, as seen with oil immersion and SEM, seems to be unique to that distinctive species, with the enlarged proximal ends that do not reach the foot-layer (Figures 34, 161-165). The foot-layer characteristically bears a small peg under each spine below the hollow center. The dissected pattern of columellae in Mun- nozia lyrata is in sharp contrast to the fused cylinder, which is mostly separated from the foot- layer, and seems to be characteristic of the subgenus Kastnera (Figures 39, 40, 172-180), M. senecionidis (Figure 35), and the latter’s relatives, M. affinis, M. angusta, M. convencionensis, M. cor- ymbosa, M. glandulosa, M. longifolia, M. oxyphylla, M. pinnulosa, M. silphioides (Figure 36), and M. venosissima (Figure 38). These fused columellae seem usually eccentric, and as seen by SEM, the eccentric form in M. senecionidis may be associ- ated with partial eccentric attachment to the foot- layer (Figure 166, 167). It is evident, on the basis of pollen, that the M. senecionidis group is more closely related to the subgenus Kastnera (Figures 39, 40) than to typical Munnozia (Figures 29, 33). According to Stix (1960), and as seen in this study in thick sections examined under the light microscope and in some broken grains under SEM, Munnozia hastifolia has columellae in the interspinal areas of the pollen wall. For this rea- son, the species was included in the umbellatum- type pollen by Stix (1960). Such a design seems anomalous in the group, because the two most closely related species, M. maronii and M. subvir- idis, have columellae of the Kastnera-M. senecion- idis- type according to oil immersion studies (Fig- ure 37). One additional distinctive pollen type is found in Munnozia jussieui. As observed under oil im- SMITHSQNIAN CONTRIBUTIONS TO BOTANY mersion by Marticorena (Figures 30-32) the col- umellae under the spines are of two forms, with numerous smaller ones surrounding a large cen- tral mass. The central mass seems to be solid, but on the basis of the outline of many samples, the central mass seems to consist of several large basal columellae that may be slightly separated. It seems unlikely that the species would be the only member of the tribe lacking a hollow or open space under the center of the spine. SEMs of broken grains (Figures 157-160) show that all the basal columellae are completely fused at the bases to the foot-layer. In Munnozia tenera of the subgenus Kastnera, there is, as usual, an outer layer of short uniform columellae, but there also is a second inner layer of similar short uniform columellae between the spines and adjacent to the foot-layer (Figures 172-180). Close examination shows that the proximal ends of the short columellae in both series are narrow and not, or only weakly, at- tached to the structure below them. Thus, the inner layer is not, or only weakly, attached to the foot-layer, and more surprisingly the outer layer is not, or only weakly, attached to the “internal tectum” that caps the inner layer of short colu- mellae. This differs from the structure of the “internal tectum” seen in other members of the tribe and described by Skvarla et al. (1977:156), which is attached to, or gives rise to, the outer layer of columellae. The inner layer of columel- lae is at a level equivalent to that occuppied under the spines in most Liabeae by the larger basal columellae attached to the foot-layer. In M. te- nera, it is notable that the cylindrical structure formed by the basal columellae under the spines is like the smaller columellae in not being at- tached to the foot-layer. The character may be more widely distributed in the subgenus Kast- nera, but it is not found in M. senecionidis, which is the closest relative outside of the subgenus that has been examined under SEM in this study (Figures 166, 167). NUMBER 64 9 Key to Types of Liabeae Pollen 1 . Pollen with spines arranged in irregular groups on the surface, not evenly aligned along margin of colpus, with bases of spines variably confluent 2 2. Exine nearly or truly caveate, with only slender basal columellae present under the spines Philoglossa 2. Exine not caveate, with stout or numerous basal columellae present under the spines .... 3 3. Pollen 25-35 /im in diameter; spines short with stout tips 4 4. With small numerous basal columellae under both the spines and areas between the spines (umbellata-type of Stix, 1960) . . . Liabum barahonense and some other Liabum species from Hispaniola 4. With large columellae restricted to area under the spines Oligactis and most Liabum; 3. Pollen mostly 30-50 /mi in diameter; spines with narrow tips 5 5. Pollen less than 40 jim in diameter Austroliahum, Cacosmia, Chionopappus, Erato, Ferreyranthus, and Microliabum 5. Pollen mostly 40-50 utm in diameter Liabellum, Sinclairia, and one Ferreyranthus species 1. Pollen with spines regularly arranged on surface of intercolpar area and mostly aligned along sides of colpus, each spine separated to equal depth on all sides 6 6. Pollen 35-45 /mi in diameter, with no or only very slender basal columellae under the spines 7 7. Spines regularly aligned along sides of colpi; basal columellae not visible under the spines using light microscopy Pseudonoseris 7. Spines slightly unaligned along sides of the colpi; a few slender basal columellae present under the spines Paranephelius 6. Pollen 27-45 /mi in diameter, with basal columellae forming clusters, rings, or fused cylinders under the spines 8 8. Basal columellae under the spines completely or almost completely separated from foot- layer at the base 9 9. Pollen 40-42 /mi in diameter; basal columellae in a circular unfused cluster under each spine Munnozia lyrata 9. Pollen 27-37 /mi in diameter; basal columellae under each spine usually fused into a single unit 10 10. With two layers of short, dense columellae in interspinal areas of tectum Munnozia subgenus Kastnera 10. With only one layer of short dense columellae in interspinal area of tectum Munnozia senecionidis 8. Basal columellae under the spines attached at bases to foot-layer 11 11. Basal columellae under each spine in small single symmetrical series about 2 /mi in diameter Chrysactinium 1 1. With at least some massive basal columellae under each spine, the clusters mostly 3-4 /mi in diameter 12 10 SMITHSONIAN CONTRIBUTIONS TO BOTANY 12. Basal columellae under the spines of two types, many small peripheral colu- mellae surrounding a central, massive, fused column Munnozia jussieui 12. A simple cluster of mostly large basal columellae under each spine 13 13. Pollen 28-30 ftrn in diameter; with basal columellae in both interspinal and spinal areas of tectum Munnozia hastifolia 13. Pollen 37-45 /im in diameter; with basal columellae only under the spines 14 14. Cluster under each spine with individual basal columellae often 2-3 iim in diameter typical Munnozia, M. lanceolata and its relatives 14. Cluster under each spine with individual columellae not more than 1.5 /x m in diameter Munnozia perfoliata Palynological Evidence on the Relationships of the Liabeae in the Asteraceae The pollen structure in the Liabeae reinforces recent conclusions regarding the two conflicting placements of the group in the family by various authors. Cassini (1828), Robinson and Brettell (1973b), Nordenstam (1977), Jansen and Stuessy (1980), and Robinson (1983a) have placed the genera of the Liabeae in or near the Vernonieae. In contrast, Bentham and Hooker (1873) fol- lowed by Hoffmann (1894) and Cronquist (1955) have placed the broadly interpreted genus Lia- bum in the Senecioneae. The two tribes involved have the distinction of belonging to two different subfamilies of the Asteraceae as interpreted by Carlquist (1976) and Robinson (1977). The Ver- nonieae share most characters with the Arcto- teae, Eremothamneae, Echinopsideae, Cynareae, Mutisieae, and Lactuceae in a group best recog- nized as the subfamily Cichorioideae, while the Senecioneae belong with the Eupatorieae, He- liantheae, Inuleae, Calenduleae, Astereae, and Anthemideae in a group best recognized as the Asteroideae. For comparisons of characters see Robinson and Brettell (1973b), Wagenitz (1976), Carlquist (1976), and Robinson (1977). As such the two tribes with which the Liabeae have been placed are not closely related, and there are many structural features by which the closest relation- ships of the tribe can be tested. Results of evi- dence from floral anatomy have most recently indicated that the Liabeae are close to the Ver- nonieae and other members of the Cichorioideae (Robinson and Brettell, 1973b; Nordenstam, 1977; Robinson, 1983a). Pollen evidence should reflect a similar relationship. A general trend has been noticed by Robinson and Brettell (1973b) where all pollen surfaces that depart from an evenly distributed echinate pattern seem to belong to the subfamily Cicho- rioideae. These variations include some distinc- tive lophate and oblong psilate types, but also include most members of the Vernonieae having less obvious sublophate forms in which the spines are only arranged around depressed areoles rather than restricted to well-defined ridges. The pollen in the Liabeae is echinate without obvious lophate or psilate variations, but the spines of most Liabeae are not evenly distributed on the surface of the grains. In all genera of the Liabeae except Chrysactinium, Munnozia, Para- nephelius, and Pseudonoseris, the spines tend to be grouped in small series or crests with some- what confluent bases. The gaps within the series are slightly but distinctly smaller than those be- tween the series. The grouping of spines in these Liabeae is unlike the variations seen in any other tribe of the Asteraceae, but by its irregularity it would seem to be excluded from the subfamily Asterioideae, and would fall into the Cichorioi- deae. Internal structure of the pollen wall shows more characters that can indicate relationship in the Asteraceae. Skvarla et al. (1966, 1977) have described features they termed a “cavus,” “inter- nal foramina,” and “internal tecta,” and on the basis of these structures they have recognized four basic pollen types in the family: helianthoid, senecioid, arctotoid, and anthemoid. Each of NUMBER 64 these types shows some broad systematic corre- lations in the family and furnish a useful basis for comparison with the Liabeae. The Liabeae do not have “internal foramina” in the pollen wall and therefore do not fall into the helianthoid classifiaction of Skvarla et al. (1977) that is defined by the presence of internal foramina with a cavus. The character is almost restricted to the subfamily Asteroideae, occur- ring especially in such tribes as the Eupatorieae and Heliantheae, which have never been consid- ered close relatives of the Liabeae. Internal for- amina are notably undeveloped in a few of the most highly evolved members of the Asteroideae, such as most true Senecioneae, the Anthemideae, and most Ambrosiinae of the Heliantheae. The fact that there are no internal foramina in the Liabeae conforms with placement in the Cicho- rioideae but does not in itself preclude relation to some Senecioneae. The Liabeae seem to lack a “cavus” in the pollen wall in their plesiomorphic form. The lack of the cavus provides a further basis for exclud- ing the Liabeae from the helianthoid grouping of Skvarla et al. (1977). The cavus is also char- acteristic of two of the other major pollen wall patterns described by Skvarla et al. (1966, 1977), the senecoid, and arctotoid. A cavus extends through most members of the subfamily Aster- oideae including the Senecioneae and is the best character after that of spine distribution for ex- cluding the Liabeae from the Senecioneae and the Asteroideae. The “lack” of a cavus is, in fact, the same as the “presence” of basal columellae connecting the outer exine to the foot layer in the intercolpar area, and this is the principal characteristic of the fourth or anthemoid pollen wall pattern of Skvarla et al. (1977). The distribution of the non-caveate or anthe- moid pollen of the type found in the Liabeae is of considerable interest, especially since it is the type found in most members of the subfamily Cichorioideae excluding the Arctoteae. Of the various pollen types, it is the one that most nearly aligns with the presently accepted subfamily clas- sification of the Asteraceae. There are important exceptions in the distribution of the anthemoid pollen type, however, including its presence in the Anthemideae of the Asteroideae and its ab- 1 1 sence in various Cichorioideae such as the Arc- toteae. In one way, it is unfortunate that the non- caveate pollen type has been called anthemoid by Skvarla et al. (1977), because the type is much more characteristic of members of the subfamily Cichorioideae. The Anthemideae is the one ma- jor group in the Asteroideae in which the anthe- moid pattern occurs. It should be emphasized further that the Anthemideae is not primitive within the subfamily Asteroideae and is not close to the Cichorioideae (Robinson, 1981). Caveate pollen, the type predominant in the Asteroideae, is found even in what seems to be a primitive element of the Anthemideae, in Ursinia (Robin- son and Brettell, 1973b; Skvarla et al., 1977). One could conclude from foregoing evidence that the non-caveate pollen of the Anthemideae represents a development of basal columellae in the intercolpar area from ancestral caveate types that did not have such columellae. The anthe- moid pollen pattern thus would have at least two separate origins in the Asteraceae. In the subfamily Cichorioideae, where the an- themoid pollen pattern predominates, there are some caveate forms such as those within the arctotoid pattern, some Lactuceae having an in- terrupted cavus, and a few Mutisieae such as Dasyphylium (which is caveate along the colpar margins). Bolick (1978) cites caveate pollen in Ethulia, Hoplophyllum, and Corymbium of the Ver- nonieae and Centaurea in the Cynareae, and in- dicates that the Arctoteae is approximately equally divided between caveate and non-caveate types. To these examples may be added those now known from the Liabeae. In these cases, the caveate pollen seems to be derived from non- caveate ancestors, and it has evidently been de- rived more than once. Apparently, the two subfamilies of the Asteraceae have the two dif- ferent types of pollen as basic, but each type, caveate and non-caveate, seems able to give rise to the other. The fact that the anthemoid pollen wall type occurs in the one specialized tribe of the Asteroideae, in the absence of any other supporting characters, does not particularly fa- vor the idea of asterioid relationships for the Liabeae. In addition to the cavus, the presence of an 12 “internal tectum” is used as a basis for differen- tiating the arctotoid pollen pattern within the subfamily Cichorioideae (Skvarla et al., 1977). This involves the most characteristic form of internal tecta that almost equally divides the re- gion of solid columellae. The arctotoid pollen is also claimed in Eremothamnus (Liens, 1970) of the tribe Eremothamneae (Robinson and Bret- tell, 1973c), but this should be checked. Illustra- tions of Skvarla et al. (1977, pi. 24: fig. F, H) suggest an arctotoid pattern in some Lactuceae such as Rafinesquia and Lapsana. Internal tecta of this type occur in pollen of other Cichorioi- deae, seeming to be most highly developed in the tribes Cynareae and Mutiseae. The latter two tribes show relationship to the Arctoteae in the form of the style, but they differ and are ex- cluded from the arctotoid type by Skvarla et al. (1977) because of the lack of a “cavus.” The Liabeae and Vernonieae lack the elaborate inter- nal tecta of the arctotoid type or the Cynareae and Mutisieae type, but enlargements at the proximal ends of the short outer columellae form a layer at the lower limit of the outer columellae that is also referred to as an internal tectum by Skvarla et al. (1977: 1 56). The latter layer is often prominent in the areas between the spines where there may be no basal columellae, and the struc- ture thus appears very different from the struc- ture referred to by the same name in the Arcto- teae and is structurally equivalent to the layer below the internal tectum of the Arctoteae. Nevertheless, the two structures are formed in the same manner and no other better term seems to have been provided as yet. Within the anthemoid pollen pattern Skvarla et al. (1977) have described a modified pattern they termed “Liaboid.” It is distinguished by the basal columellae being much broader (Figures 68, 89, 90, 113, 114, 120) than those noted for the typical anthemoid pollen pattern. The pat- tern is cited from ” Liabum Cacosmia, and the Vernonieae and is cited elsewhere by Skvarla et al. only in Scolymus of the Cichorioid Lactuceae on the basis of TEM studies by Tomb (1975). Observation during the present study show that Gundelia, of the prospective Cichorioid tribe SMITHSONIAN CONTRIBUTIONS TO BOTANY Gundelieae, has similar massive columellar struc- ture under the spines, but these do not reach the foot-layer. The limitation of the liaboid pollen pattern to members of the subfamily Cichorioi- deae seems to provide confirmation of the place- ment of the Liabeae in that subfamily. The dis- tribution of the liaboid pattern within the subfamily seems to parallel evidence of stylar form in suggesting closest relationship of the Liabeae to the Vernonieae and Lactuceae, espe- cially the former. Of the tribes that appear to be most closely related to the Liabeae, the pollen of the Lactu- ceae can be distinguished by various characters such as lophate surfaces or more complex inter- nal tecta. In the cases where Tomb (1975) and Skvarla et al. (1977) show internal structure more like that of the Vernonieae and Liabeae the structure is under the ridges of specialized echinolophate grains. Among the non-lophate members of the Lactuceae, there seem to be cavus-like areas periodically interrupted by col- umellae reaching the foot layer, a form seen in comparatively few derived members of the Lia- beae. The Vernonieae is regarded here as the prob- able closest relatives of the Liabeae, and the two tribes have been combined by some recent au- thors (Nash, 1976; Turner and Powell, 1977; Jansen and Stuessy, 1980). The two groups were placed adjacent to each other in the pollen study by Stix (1960) because of their similar wall struc- ture. It is of interest, therefore, that the pollen of the Vernonieae and Liabeae is in fact distin- guishable, and almost without exception the tribes show totally different trends in the details of their anthemoid structure. Surface features alone could be used to distin- guish most Vernonieae from the Liabeae. Only the recently described Pseudostiff tia (Robinson, 1979) in the Vernonieae appears to have no trace of a lophate surface pattern (Figures 181, 182). The remainder of the Vernonieae have various degrees of areole formation (Keeley and Jones, 1979). Most members of the tribe have a form termed the Lychnophora-type by Stix (1960), where the tectum is continuous between the NUMBER 64 13 colpi, but where slight areolae are present (Fig- ures 190-192). The polygonal areolae are more obvious under the light microscope where the columellae are visible to accentuate the pattern. The ridges in some genera, such as Bishopalea, are more obvious with the tectum still continuous across the areolae (Figures 200-202). More strongly lophate pollen forms in the tribe, such as Heterocypsela, have regularly arranged ridges with a restricted tectum (Figures 203, 204). The extreme form of lophate grain in the tribe has little or no tectum and has the columellae visible from a lateral view of the ridges. This is seen in FAephantopus and its relatives, as well as in some Paleo-tropical elements of the subtribe Vernoni- inae, such as Cyanthillium and Phyllocephalum (Figures 205-208; Skvarla et al., 1977; Keeley and Jones, 1979; Jones, 1981; Kirkman, 1981). The trend toward a lophate pollen surface seems basic to the Vernonieae, but is absent from the Liabeae. Size of pollen grains alone offers some distinc- tion between the tribes. Pollen of the Liabeae is mostly 25-40 jmi in diameter with some up to 50 /zm. The pollen of the Vernonieae is usually 40 /zm in diameter or more, and it is often over 50 jLzm. Based on a limited sample, Skvarla et al. (1977) indicated some differences in the detailed inter- nal structure of the liaboid pattern in the Ver- nonieae and Liabeae. On page 156 they stated the following. In Liabum the lateral branches of the distal ends of the columellae (sic basal columellae) form a uniform internal tectum. From this internal tectum a short, uniform set of columellae arise which in turn are capped by a perfo- rate tectum. Commonly, the tectum areas between the large columellae show gentle undulations. In the Ver- nonieae the major distinction from Liabum is that the lateral branches are frequently (but not always) a thick network of solid rods or tubules with complex anasto- mosing patterns. The distinction is evident to some degree in the examples of broken pollen grains seen in the present study. Many Vernonieae, including Blan- chetia (Figures 197-199) and Pseudostiff tia (Fig- ures 184-189), show a branching pattern that completely Fills the inter-spinal areas, while Ver- nonia fuertesii shows that even an unsupported internal tectum in the tribe does not bear a layer of regular short columellae as in the Liabeae (Figures 193-195). The most important difference in the internal structure of the pollen wall was first noted by Stix (1960) and has been confirmed in the pre- sent study by the light microscope and SEM views of broken grains. In the Vernonieae with well- developed spines, each spine seems to have a single solid basal columella centered under the spine, which is continuous with the apex of the spine (Figures 193-195, 197-199). The only exception seems to be Pseudostiff tia with basal columellae distributed more randomly through- out the spinal and interspinal areas with branches of more than one basal columella contributing to each spine (Figures 184-189). Though unlike any other Vernonieae, P seudo stiff tia is not like any Liabeae. In the Liabeae, the basal columellae are either grouped under the spines (Figures 1 2- 16, 29, 46, 89, 163-165) or are coalesced into a single hollow cylinder (Figures 40, 145, 146, 166, 167, 174-176). The coalesced form, which seems continuous with the apex of the spine, resembles the single columella of the Vernonieae except for the hollow center. Single solid colu- mellae centered under the spines do not seem to occur in the Liabeae except possibly in a special- ized form seen in Munnozia jussieui. In the latter case the oil immersion studies (Figures 30-32) suggest a characteristically higher degree of co- alescence than any other member of the tribe, but the SEM seems to show narrow separation (Figures 157-160) that was overlooked at the lower magnification. The difference between the basal columellae in the two tribes is regarded herein as fundamental, and observation of the range of variation in the Liabeae indicates that the cluster of separate columellae is plesiom- orphic in the tribe. The form most resembling the Vernonieae is apparently an isolated deriva- tive. Appendix List of specimens examined palynologically in the study with figures in which they appear and with measurements (in /um) of pollen in glycerine mounts and Hoyer’s (H) mounts. Liabeae Austroliabum A. candidum (Grisebach) H. Robinson & Brettell O. Kuntze s/n, Argentina, Figures 69-72 A. eremophyllum (Cabrera) H. Robinson 8c Brettell Venturi 6580, Argentina A. polytnnioides (R.E. Fries) FI. Robinson & Brettell Pflanz 4102, Bolivia, Figure 3 Venturi 5280, Argentina Wall & Sparre 622, Argentina, Figure 73 Bishopanthus B. soliceps H. Robinson Ring 9280, Peru, Figures 54-57 Cacosmia C. rugosa Humboldt, Bonpland, 8c Kunth Barclay & Juajibioy 8305, Ecuador Barclay & Juajibioy 8339, Ecuador Wurdack 1016, Peru, Figures 4, 58-61 Chionopappus C. benthamii (Bentham) Blake Asplund 10996, Peru, Figures 5, 64-67 Ferreyra 7663, Peru Ferreyra 1 1078, Peru Macbride 8c Featherstone 162, Peru Pennell 14480, Peru Chrysactinium C. acaule (Humboldt, Bonpland, 8c Kunth) Weddell Camp E-2231, Ecuador, Figures 27, 144-146 Haught 3261, Ecuador, Figures 140-143 C. amphothrix (Blake) H. Robinson 8c Brettell Ferreyra 9481, Peru C. caulescens (Hieronymus) H. Robinson & Brettell Jameson s/n, Ecuador, type of Liabum bicolor Blake Wurdack 764, Peru C. hieracioides (Humboldt, Bonpland, 8c Kunth) H. Robinson & Brettell Wurdack 1113, Peru, Figure 28 C. longiradiatum (Hieronymus) H. Robinson 8c Brettell Fagerlind 8c Wibon 1563, Ecuador Haught 3261, Ecuador Holmgren 560, Ecuador Erato E. polymnioides DeCandolle Ferreyra 8168, Peru H 38-40 H 33-37 35-37 33-37 30(-35), H 35 H 27-30 H 27 H 40 H 35-37 30 H 40 30-32 30-34 H 35-37 H 35-37 14 NUMBER 64 15 King 6551, Ecuador, Figures 124-130 Lugo 241, Ecuador, Figure 24 E. Sodiroi (Hieronymus) H. Robinson Barclay &Juajibioy 8314, Ecuador, Figures 131-134 35-40, Rose & Rose 22413, Ecuador E. stenolepis (Blake) H. Robinson Hutchison 1 190, Peru E. vulcanica (Klatt) H. Robinson Cuatrecasas 256, Colombia Steyermark 55981, Venezuela, type Liabum insigne Badillo v. Sneidern A. 504, Colombia Ferreyranthus F. excelsus (Poeppig & Endlicher) H. Robinson & Brettell Hutchison & Wright 3854, Peru, Figure 6 F. rugosus (Ferreyra) H. Robinson & Brettell Wurdack 469, Peru, type, Figures 82, 83 29-35, F. vaginans (Muschler) H. Robinson & Brettell Vargas I 1050, Peru F. verbascifolius (Humboldt, Bonpland, & Kunth) H. Robinson & Brettell Camp E-3906, Ecuador Hutchison & Wright 5133, Peru, Figures 78-81 Lehmann 7958, Ecuador, type of Liabum salviifolium Hieronymus F. vernonioides (Muschler) H. Robinson & Brettell Hutchison & Wright 5176, Peru, Figure 7 Kunkel 975, Peru Liabellum L. angustissimum (A. Gray) Rydberg Pringle 2501, Mexico Rose 3406, Mexico L. cervinum (B.L. Robinson) Rydberg Pringle 4398, Mexico L. palmeri (A. Gray) Rydberg Barnes & Land s/n, Mexico, Figures 94-98 Pringle 2328, Mexico Liabum L. acuminatum Rusby Williams 1605, Bolivia Woytkowski 34392, Peru L. amplexicaule Poeppig & Endlicher Klug 3 1 83, Peru L. barahonense Urban Howard 12087, Dominican Republic, Figures 12, 121-123 Jimenez 3284, Dominican Republic, Figures 13, 108-1 1 1 L. barclayae H. Robinson Barclay &Juajibioy 8316, Ecuador, type L. bourgeaui Hieronymus Skutch 4052, Costa Rica, Figure 14 Skutch 4730, Costa Rica Standley & Valerio 51861, Costa Rica L. crispum Schultz-Bipontinus Ekman 15643, Cuba L. cubense Schultz-Bipontinus Ekman 1542, Cuba L. domingense Rydberg 35, 36 35-37 36-38 32-34 35-37 H 35 30-34 39-41 H 43 H 50 50-52 47-50 28-31 30-34 34, 35 34-36 34, 35 27-32 16 SMITHSONIAN CONTRIBUTIONS TO BOTANY Valeur 994, Dominican Republic L. eggersii Hieronymus Camp E-3I96, Ecuador Fagerlind Sc Wibon 290, Ecuador, Figure 15 Haught 3446, Peru L. falcatum Rusby Smith 2012, Colombia, type, Figure 16 L. ferreyrii H. Robinson Hutchison, Wright & Straw 5948, Peru L. floribundum Lessing Camp E-3196, Ecuador Hutchison & v. Bismark 6337, Peru L. grandiflorum (Humbolt, Bonpland, Sc Kunth) Lessing Camp E-3123, Ecuador Rose, Pachano Sc Rose 23232, Ecuador, type of L. amplexans Blake L. igniarium (Humboldt, Bonpland, & Kunth) Lessing Heilborn 534, Ecuador, Figures 112-117 Lehmann 4896, Ecuador, type of L. lehmannii Hieronymus L. kingii H. Robinson Sydow 484, Ecuador L. melastomoides (Humboldt, Bonpland, Sc Kunth) Lessing Dryander 39172, Colombia Humbert et al. 25795, Colombia L. nigro-pilosum Hieronymus Mexia 6686, Ecuador L. nudicaule H. Robinson Hutchison 1191, Peru L. oblanceolatum Urban Sc Ekman Loigier 15402, Dominican Republic L. ovatifolium Urban Ekman H-12578, Dominican Republic, type L. polycephalum Urban Ekman H-5346, Haiti L. poiteaui (Cassini) Urban Ekman H-3713, Haiti, Figure 17 L. sandemanii H. Robinson Sandeman 4420, Peru, type L. selleanuin Urban Ekman H-1548, Haiti, type. Figure 18 L. solidagineum (Humboldt, Bonpland, Sc Kunth) Lessing Humbert 31002, Peru Hutchison 1 129, Peru Macbride Sc Featherstone 1519, Peru L. subacaule Rydberg Ekman H-1870, Haiti, Figure 19 Valeur 56, Dominican Republic, Figures 20, 1 18-120 L. umbellatum (L.) Schultz-Bipontinus Maxon 277, Jamaica Maxon 8575, Jamaica L. vargasii H. Robinson Vargas 10182, Peru, type L. wrightii Grisebach Ekman 5061, Cuba Ekman 14745, Cuba L. wurdacku Ferreyra 27-30 H 30-32 H 27 24-46 32-35 27-30 H 30-32 25-27 33-35 33-35 30-33 H 35 32-35 H 25-27 32, 33 32-35 30-32 30-32 32-35 31-34 NUMBER 64 17 Hutchison & Wright 5850, Peru Microliabum M. humile (Cabrera) Cabrera Okada 5953, Argentina, Figures 74-77 Munnozia subgenus Munnozia M. affinis (Blake) H. Robinson & Brettell Macbride 4337, Peru, type M. angusta (Blake) H. Robinson & Brettell Mexia 8152, Peru M. canarensis (Cuatrecasas) H. Robinson & Brettell Camp E-2836, Ecuador, type M. cardenasii (Cabrera) H. Robinson & Brettell Herzog 2149, Bolivia M. convencionensis (Cuatrecasas) H. Robinson & Brettell Vargas 4446, Peru, type M. corymbosa Ruiz & Pavon ex herb Pavon, possible type M.foliosa Rusby Bang 1 195, Bolivia, type Bang 1581, Bolivia, type of M. chrysanthemoides Rusby, Figure 29 Vargas 2761, Peru, type of Liabum herrerae Cabrera Metcalf 30754, Peru M. gigantea (Rusby) Rusby Bang 2379, Bolivia, type M. glandulosa (O. Kuntze) Rusby Cardenas 683, Bolivia M. hastifolia (Poeppig & Endlicher) H. Robinson & Brettell Uribe 3149, Colombia Hutchison 1191, Peru M. jussieui (Cassini) H. Robinson & Brettell Lehmann 8450, Colombia, type of Liabum nonoense var. microcephalum H ieronymus. Figure 32 Triana 1 145, Colombia, Figures 30, 157-160 v. Sneidern 1974, Colombia, Figure 31 M. lanceolata Ruiz & Pavon Humbert 30952, Peru, Figure 33 Hutchison, Wright & Straw 5928, Peru, Figures 147-155 M. longifolia Rusby Buchtien 3079, Bolivia, type M. lyrata (A. Gray) H. Robinson & Brettell Ferreyra 6968, Peru, Figures 34, 161-165 M. maronii (Andre) H. Robinson Mandon 240, Bolivia M. olearioides (Muschler) H. Robinson & Brettell Weberbauer 4417, Peru, type frag. M. oxyphylla (Cuatrecasas) H. Robinson & Brettell Woytkowski 34165, Peru, type M. perfoliata (Blake) H. Robinson & Brettell King & Guevara 6184, Colombia, Figure 156 M. peruensis (Cuatrecasas) H. Robinson & Brettell Woytkowski 34275, Peru, type M. pinnulosa (O. Kuntze) H. Robinson & Brettell Holway 616, Bolivia M. rusbyi (Britton) Rusby Cardenas 6278, Bolivia 31, 32 H 35 H 37-40 33-35, H 35 42-45 H 37 H 42-45 H 43-45 28-30 35 37 H 40-42 30-35 H 40-44 H 38-40 H 45 18 SMITHSONIAN CONTRIBUTIONS TO BOTANY M. senecionidis Benth Garcia-Barriga 12518, Colombia Cardenas 3277, Bolivia Haught 3295, Ecuador, Figure 35 King & Almeda 7842, Ecuador, Figures 166, 167 Pennell 769, Colombia, type of M. attenuata Rusby Schneider 7, Colombia v. Sneidern 2778, Colombia M. silphioides (Poeppig 8c Endlicher) H. Robinson 8c Brettel! Swingle 183, Peru, Figure 36 M. subviride (Blake) H. Robinson 8c Brettell Cook & Gilbert 1365, Peru, type. Figure 37 M. venosissima Ruiz & Pavoti Macbride 4312, Peru, type of Liabum pulchrum Blake, Figure 38 Munnozia subgenus Kastnera M. acostae Chung Asplund 9586, Ecuador M. annua (Muschler) H. Robinson & Brettell Hutchison & Wright 5084, Peru, Figures 39, 168-171 M. liaboides (Lessing) H. Robinson Ferreyra 6968, Peru Haught 3294, Ecuador M. nivea (Hieronymus) H. Robinson 8c Brettell Killip 8c Varela 34590, Colombia, Figure 40 Weberbauer 6022 M. tenera (Schultz-Bipontinus) H. Robinson & Brettel! Barclay 8c Juajibioy 5760 Killip 9775, Colombia, Figures 172-180 Oligactis subgenus Oligactis O. latifolia (Hieronymus) H. Robinson & Brettell v. Sneidern 354, Colombia O. sessiliflora (Humboldt, Bonpland, 8c Kunth) H. Robinson 8c Brettell Allart 370, Venezuela, type of Liabum tovarense Badillo Ariste-Joseph A-257, Colombia O. volubilis (Humboldt, Bonpland, 8c Kunth) Cassini Cuatrecasas 12855, Colombia, Figures 99-103 Gehringer 427, Venezuela, type of Liabum meridense Badillo Perez-Arbelaez & Cuatrecasas 8164, Colombia, type of Liabum boyacense Cuatrecasas Smith 2013, Colombia, type of Liabum biattenuatum Rusby, Figure 21 Oligactis subgenus Andromachiopsis O. coriacea (Hieronymus) H. Robinson 8c Brettell Camp E-4190, Ecuador, Figure 22 Mexia 7476, Ecuador O. ecuadoriensis (Hieronymus) H. Robinson 8c Brettell Lehmann 4897, Ecuador, type O. ochracea (Cuatrecasas) H. Robinson 8c Brettell Townsend A- 193, Peru O. pastoensis (Cuatrecasas) H. Robinson 8c Brettell Cuatrecasas 1 1949, Colombia, type O. pichinchensis (Hieronymus) H. Robinson 8c Brettell Mexia 7697, Ecuador, Figures 104-107 Sydow 605, Ecuador, Figure 23 27-30 30-32 H 34-37 H 35-37 H 33-35 H 35-37 27-30 25-27 33-37 H 33-35 H 33-35 30-32, H 35 27-30 NUMBER 64 19 Paranephelius P. asperifolius (Muschler) H. Robinson & Brettell Fiebrig 3538, Bolivia, type 38-40 P. bullatus A. Gray ex Weddell Macbride & Featherstone 1656, Peru Macbride 8c Featherstone 2131, Peru P. jelskii (Hieronymus) H. Robinson & Brettell Wurdack 760, Peru, Figure 1 35-37 Wurdack 1240, Peru, Figure 48 P. ovatus Weddell Pennell 13363, Peru 47-50 P. uniflorus Poeppig & Endlicher Cerrate 1367, Peru, Figures 45-47 36-40 Ferreyra 5580, Peru, Figures 41-44 Herzog 1840, Bolivia, Figure 2 42-45 Philoglossa P. mimuloides (Hieronymus) H. Robinson & Cuatrecasas Dodson 8c Thien 1467, Ecuador H 35 Firmin 52, Ecuador King 6633, Ecuador P. mimuloides forma sapida (Bristol) H. Robinson 8c Cuatrecasas Bristol 348, Colombia, type. Figure 25 P. peruviana DeCandolle Asplund 13735, Peru, Figures 138, 139 H 35-37 Ferreyra 4022, Peru, Figures 26, 134-137 P. pterocarpa Sandwith Soukup 4219, Peru H 30 Pseudonoseris P. discolor (Muschler) H. Robinson 8c Brettell Metcalf 3053 1 , Peru, Figures 50-53 H 37-40 P. striata (Cuatrecasas) H. Robinson 8c Brettell Weberbauer 7107, Peru, type H 38-40 Sinclairia S. andrieuxii (DeCandolle) H. Robinson 8c Brettell Cronquist 8c Sousa 10457, Mexico H 50-52 S. andromachioides (Lessing) Rydberg Pringle 5905, Mexico H 37-42 Purpus 2942, Mexico 35 S. brachypus Rydberg Pittier 1886, Guatemala, type S. caducifolia (B.L. Robinson & Bartlett) Rydberg King 8c Soderstrom 4988, Mexico Palmer 245, Mexico S. deamii (B.L. Robinson 8c Bartlett) Rydberg Edwards 568, Honduras, Figures 9, 91-93 H 37-42 Standley 19695, El Salvador, Figures 10, 88-90 S. deppeana (Lessing) Rydberg Botteri 1 177, Mexico 37, H 42-45 S. dimidia (Blake) H. Robinson 8c Brettell Bartlett 12602, Guatemala, type S. discolor Hooker 8c Arnott Molina et al. 17514, Costa Rica Ton 780, Mexico, Figures 84-87 37 20 SMITHSONIAN CONTRIBUTIONS TO BOTANY Williams & Molina 13728, Honduras S. glabra (Hemsley) Rydberg Hartman 120, El Salvador Mexia 8797, Mexico S. hypochlora (Blake) Rydberg Skutch 2031, Guatemala S. klattii (B.L. Robinson 8c Greenmail) H. Robinson 8c Brettell Conzatti 1751, Mexico, Figure 8 Pringle 6059, Mexico 5. moorei (H. Robinson 8c Brettell) H. Robinson 8c Brettell Moore 5518, Mexico, type 5. pittien Rydberg Pittier 9093, Costa Rica, type 5. platylepis (Schultz-Bipontinus ex Klatt) Rydberg Schipp S-738, Belize ■S. polyantha (Rlatt) Rydberg Allen 2886, Panama Skutch 3488, Costa Rica Standiey 42513, Costa Rica S. pringlei (B.L. Robinson & Bartlett) H. Robinson 8c Brettell Pringle 6214, Mexico S. sericolepis (Hemsley) Rydberg Bourgeau 2177, Mexico, type 5. vagans (Blake) H. Robinson 8c Brettell Skutch 1913, Guatemala, type 35-37 42-45 48-50 H 53-55 33-37 38-40 H 37-40 42, H 50-52 Vernonieae Bishopalea B. erecta H. Robinson King & Bishop 8729, Brazil, Figures 200-202 Blanchetia B. heterotricha DeCandolle Schery 736a, Brazil, Figures 196-199 Heterocypsela H. andersorw H. Robinson Anderson et al. 9223, Brazil, type, Figures 203, 204 Phyllocephalum P. scabridum (DeCandolle) Kirkman Stocks 127, India, Figures 205-208 Pseudostifftia P. kingii H. Robinson King et al. 8145, Brazil, Figures 181-189 Vernoma V . fuertesii (Urban) H. Robinson Ekman H-7351 , Haiti, type of Vernonia barkeri Ekinan ex Urban, Figures 190-195 Literature Cited Bentham, G., and J.D. Hooker 1873. Ordo 88: Conipositae. In Genera Plantarum, 2(1 ): 163-533, 536, 537. Blake, S.F. 1935. The Genus Chionopappus of Bentham (Astera- ceae). Journal of the Washington Academy of Sciences, 25(1 1 ):488-493. Bolick, M.R. 1 978. Taxonomic, Evolutionary, and Functional Consid- erations of Compositae Pollen Ultrastructure and Sculpture. Plant Systematics and Evolution, 130:209-218. Cabrera, A.L. 1954. Compuestas sudamericanas nuevas o criticas, II. Notas del Museo, Universidad Nacional de Eva Pe- ron, 1 7(84): 7 1 -80. Carlquist, S. 1976. Tribal Interrelationships and Phylogeny of the Asteraceae. Ahso, 8(4):465-492. Cassini, H. 1828. Vernoniees. In G. Cuvier, editor, Dictionnaire des Sciences Naturelles, 57:338-347. Paris [Reprinted 1975 in R.M. King and H.W. Dawson, editors, Cassini on Compositae, 3:1801-1811. New York: Oriole Editions.] Cronquist, A. 1 955. Phylogeny and Taxonomy of the Compositae. The American Midland Naturalist , 53(2):478-51 1. Erdtman, G. 1966. Pollen Morphology and Plant Taxonomy: Angios- perms. xii + 553 pages. Stockholm: Almquist and Wiksell. Eeuer, S.M., and M.O. Dillon 1982. Pollen Morphology and Ultrastructure of the Lia- beae (Asteraceae). Botanical Society of America Mis- cellaneous Publication, 162:93. Hoffmann, O. 1894. Compositae. In A. Engler and K. Prantl, editors, Die natiirlichen Pflanzenfamihen, 4(5):87-387. Leipzig. ' Jansen, R.K., and T.F. Stuessy 1980. Chromosome Counts of Compositae from Latin America. American Journal of Botany, 67(4):585- 594. Jones, S B. 1981 . Synoptic Classification and Pollen Morphology of Vernonia (Compositae; Vernonieae) in the Old World. Rhodora, 83(833):59-75. Keeley, S.C., and S.B. Jones I 979. Distribution of Pollen Types in Vernonia (Vernon- ieae: Compositae). Systematic Botany, 4(3): 195- 202. Kirkman, L.K. 1981. Taxonomic Revision of Centratherum and Phyllo- cephalum (Compositae: Vernonieae). Rhodora, 83(833): 1-24. Liens, P. 1970. Die Pollenkorner und Verwandtschaftsbeziehun- gen der Gattung Eremothamnus (Asteraceae). Mit- teilungen der Botanischen Staatssammlung, Miin- chen, 7:369-376. Marticorena, C., and O. Parra 1974. Morfologia de los granos de polen y posicion sistematica de Anisochaeta DC., Chionopappus Benth., Fedea Urb. y Gochnatia glomeriflora Gray (Compositae). Boletin Sociedad Biologia de Concep- cion, 47:187-197. Nash, D.L. 1976. Tribe I: Vernonieae. In D.L. Nash and L.O. Wil- liams, Compositae. Flora of Guatemala, 24(I2):4- 32. Nordenstam, B. 1977. Senecioneae and Liabeae: Systematic Review. In Heywood, Harborne, and Turner, editors, The Biology and Chemistry of the Compositae, pages 799- 830. London and New York: Academic Press. Robinson, H. 1977. An Analysis of the Characters and Relationships of the Tribes Eupatorieae and Vernonieae (Aster- aceae). Systematic Botany, 2(3): 1 99-208. 1978. 190 (2), Compositae — Liabeae. Flora of Ecuador, 8:1-62. 1979. Two New Genera of Vernonieae (Asteraceae) from Brasil, Heterocypsela and Pseudostifftia. Phy- tologia, 44(7):442-450. 1981 A Revision of the Tribal and Subtribal Limits of the Heliantheae (Asteraceae). Smithsonian Contri- butions to Botany, 51: 102 pages. 1983a. A Generic Review of the Tribe Liabeae (Astera- ceae). Smithsonian Contributions to Botany, 54: 69 pages. 1983b. Studies in the Liabeae (Asteraceae), XVI: New Taxa from Peru. Phytologia, 54(l):62-65. Robinson, H., and R.D. Brettell 1973a. Tribal Revisions in the Asteraceae, III: A New Tribe, Liabeae. Phytologia, 25(6):404-407. 21 22 SMITHSONIAN CONTRIBUTIONS TO BOTANY 1973b. Tribal Revisions in the Asteraceae, VIII: A New Tribe, Ursinieae. Phytologia, 26(2):76-85. 1973c. Tribal Revisions in the Asteraceae, XI: A New Tribe, Eremothamneae. Phytologia, 26(3): 163- 166. 1974. Studies in the Liabeae (Asteraceae), II: Prelimi- nary Survey of the Genera. Phytologia, 28(1 ):43- 63. Robinson, H., A.M. Powell, R.M. King, andJ.F. Weedin 1985. Chromosome Numbers in Compositae, XV: Lia- beae. Annals of the Missouri Botanical Garden, 72:469-479. Rydberg, P.A. 1927. (Carduales) Carduaceae, Liabeae, Neurolaeneae, Senecioneae (pars). North American Flora, 34(4):289-360. Sandwith, N.Y. 1956. Contributions to the Flora of Tropical America, LXI: Notes on Philoglossa. Kew Bulletin, 1956:289-293. Skvarla, J.J., and B.L. Turner 1966. Systematic Implications from Electron Micro- scopic Studies of Compositae Pollen: A Review. Annals of the Missouri Botanical Garden, 53:200- 256. Skvarla, J.J., B.L. Turner, V.C. Patel, and A.S. Tomb 1 977. Pollen Morphology in the Compositae and in Mor- phologically Related Families. In Heywood, Har- borne, and Turner, editors, The Biology and Chem- istry of the Compositae, pages 141-248. London and New York: Academic Press. Stix, E. 1960. Pollenmorphologische untersuchungen an Com- positen. Grana Palynologica, 2(2):41-104, plates 1-21. Tomb, A.S. 1975. Poilen Morphology in Tribe Lactuceae (Compos- itae). Grana, 15(l/3):79-89. Turner, B.L., and A.M. Powell 1977. Heienieae: Systematic Review. In Heywood, Har- borne, and Turner, editors, The Biology and Chem- istry of the Compositae, pages 699-737. London and New York: Academic Press. Wagnitz, G. 1976. Systematics and Phylogeny of the Compositae. Plant Systematics and Evolution, 125:29-46. NUMBER 64 23 • . » • _ • *.’■ 1 •’ ' ' 2 :• s C •> • *v o o r. O ? # # .. O 0 j.*4 • * - * 5 • * m ft •• %* •!' • - e v %# •• *5 o y w f ^ * * + . * •• /• •• ’• •• r •« %* • V* % • 8 9 • •• •% .% ^ * v .. .> •’ • # ^ *• l.1 . ■ • %' V o * •H •*« ^ *•« 0 •••'12 13 i .... i 5 urn .* % 0 O ♦4 •r. 14 V c % i* * • •* » * •V •V • • •« :Y* «; • % •• 15 :*♦ 0 *.*16 *1 %• V 17 Figures 1- -17. — Basal columellar patterns of the tribe Liabeae in vertical view (for citation of specimens see appendix): 1, Paranephelius jelskii; 2, P. uniflorus; 3, Austroliabum polymnioides ; 4, Cacosmia rugosa; 5, Chionopappus benthamii; 6, Ferreyranthus excelsus; 7, F. vernonioides; 8, Sinclairia klattii; 9, 10, 5. deamii ; 1 1, Liabellum angustissimum; 12, 13, Liabum barahonense; 14, L. bourgeaui; 15, /.. eggersii; 16, L. falcatum; 17, L. poiteaui. • •• 24 SMITHSONIAN CONTRIBUTIONS TO BOTANY % 9 25 • 4' 0 0 o a O ,*• 1 0 z », • • % * • • 27 •* * O **. 0 28 26 *9S /if?- 31 l» C? C %y #/ 29 32 A . . *. $ w.. • T * m W: * %% i# r# %; %;• * 30 _J 1 I L_ 0 I ^ • % -• *.* xJ v; -*.•34 ^ C) w w *37 • (J^3®CCl39 •V 0% / V V* t«* *#«» 33 5 gm ^ • © ^ • C • V1 A* 35 # *36 « ^ ) C ^ « *> >38 , o * © C O40 Figures 18-40. — Basal columellar patterns of the tribe Liabeae in vertical view: 18, Liabum selleanum; 19, 20, L. subacaule; 21, Ohgactis volubilis; 22, O. (Andromachiopsis) coriacea; 23, O. (A.) pichinchensis; 24, Erato polymnioides ; 25, Philoglossa mimuloides f. sapida; 26, P. peruviana DC; 27, Chrysactinium acaule; 28, C. hieracioides; 29, Munnozia foliosa; 30-32, M. jussieui; 33, M. lanceolata; 34, M. lyrata; 35, M. senecionidis; 36, M. silphioides; 37, M. subviridis; 38, M. venosissima; 39, M. (Kastnera) annua; 40, M. (K.) nivea. NUMBER 64 25 Figures 41-46. — Scanning Electron Micrographs of tribe Liabeae, Paranephelius uniflorus pollen: 41, colpar view; 42, polar view; 43, colpus; 44, intercolpar area; 45, broken tectum; 46, broken tectum. (Scale lines 41 = 10 /mi; 42-46 = 5 /mi.) 26 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 47-53. — SEM of tribe Liabeae, Paranephelius and Pseudonoseris pollen: 47, Parane- phehus uniflorus; 48, P. jelskii; 49, P. uniflorus, unacetolized grain. Pseudonoseris discolor: 50, colpar view; 51, polar view; 52, colpus; 53, spines. (Scale lines 47, 48 = 2 fim; 49-51 = 10 iim; 52 = 5 /xm; 53 = 4 /im.) NUMBER 64 27 Figures 54-59. — SEM of tribe Liabeae, Bishopanthus and Cacosmia pollen. Bishopanthus soliceps: 54, equatorial view with colpus; 55, polar view; 56, intercolpar region; 57, broken tectum. Cacosmia rugosa: 58, colpar view; 59, polar view. (Scale lines 54-56, 58, 59 = 10 nm; 57 = 2 fi m.) 28 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 60-65. — SEM of tribe Liabeae, Cacosmia and Chionopappus pollen. Cacosmia rugosa: 60, colpus; 6 1 , spines; 62, 63, unacetolized grains. Chionopappus benthamii: 64, colpar view; 65, polar view. (Scale lines 60, 61=5 /im; 62-65 = 10 fin l.) NUMBER 64 29 Figures 66-71. — SEM of tribe Liabeae, Chionopappus and Austroliabum pollen. Chionopappus benthamii: 66, spines; 67, colpus; 68, microphotograph of microtome section. Austroliabum candidum: 69, colpar view; 70, polar view; 71, spines. (Scale lines 66 = 2 (im; 67 = 5 ^m; 68- 70 = 10 jim; 71=4 fim.) 30 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 72-77. — SEM of tribe Liabeae, Austroliabum and Microliabum pollen. Austroliabum candidum: 72, colpus; A. polymnioides: 73, broken grain. Microliabum humile: 74, colpar view; 75, polar view; 76, spines; 77, broken grain. (Scale lines 72, 76 = 5 /im; 73 = 2 /xm; 74, 75 I 0 /xm; 77 = I /xm.) NUMBER 64 31 Figures 78-83. — SEM of tribe Liabeae, Ferreyranthus and Sinclairia pollen. Ferreyranthus verbascifolius: 78, colpar view; 79, polar view; 80, colpus; 81, spines. F. rugosus: 82, broken grain. Sinclairia discolor: 83, colpar view. (Scale lines 78, 79, 83 = 10 (im; 80, 81=5 /im; 82 = 2 Min.) 32 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 84-89. — SEM of tribe Liabeae, Sindairia pollen. Sindairia discolor: 84, polar view; 85, colpus; 86, spines; 87, unacetolized grain. 5. deamii: 88, 89, Standley 19695, broken grains. (Scale lines 84, 87 = 10 ^m; 85, 86 = 5 ^m; 88, 89 = 2 /um.) NUMBER 64 33 Figures 90-95. — SEM of tribe Liabeae, Sinclairia and Liabellum pollen. Sinclairia deamii: 90, St and ley 19695 ; 91-93, Edwards 568. Liabellum palmeri: 94, coipar view; 95, polar view. (Scale lines 90 = 2 /im; 91-93 = 1 ^m; 94, 95 = 10 um.) 34 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 96-101 . — SEM of tribe Liabeae, Liabellum and Oligactis pollen. Liabellum palmeri: 96, polar view; 97, colpus; 98, spines. Oligactis volubilis: 99, colpar view; 100, 101, polar views. (Scale lines 96, 97, 99-101 = 10 /um; 98 = 5 /mi.) NUMBER 64 35 Figures 102-107. — SEM of tribe Liabeae, Oligactis pollen. O. volubilis: 102, colpus; 103, spines. Oligactis (Andromachiopsis) pichinchensis: 104, colpar view; 105, polar view; 106, colpus; 107, spines. (Scale lines 102, 106 = 5 nm; 103, 107 = 2 jim; 104, 105 = 10 jam.) 36 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 108-113. — SEM of tribe Liabeae, Liabum pollen. L. umbellatum: 108, colpar view; 109, polar view; 1 10, colpus; 111, spines. L. igniarium: 112, 113, broken grains. (Scale lines 108, 111, 112, 113 = 5 nm ; 109, 1 10 = 10 /mi.) NUMBER 64 37 Figures 114-1 19. — SEM of tribe Liabeae, Liabum pollen, broken grains. 1 14-117, L. ignia- rium. 118, 1 19, L. subacaule. (Scale lines 114 = 5 fim; 115-119 = 2 fim.) 38 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 120-125. — SEM of tribe Liabeae, Liabum and Erato pollen. L. subacaule: 120, broken grains. L. barahonense: 1 21-123, broken grains. Erato polymnioides: 1 24, colpar view; 125, polar view. (Scale lines 120 = 2 m'11; 121-123 = 1 /xm; 124, 125 =10 jim.) NUMBER 64 39 Figures 126-133. — SEM of tribe Liabeae, Erato pollen. Erato polymnioides: 126, polar view; 127, colpus; 128, spines; 129, 130, broken grains. E. sodiroi: 131-133, broken grains. (Scale lines 126 = 10 fim; 127 = 5 Mm; 128-133 = 2 /mi.) 40 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 134-139. — SEM of tribe Liabeae, Philoglossa peruviana pollen: 134, near polar view; 135, colpar view; 136, colpus; 137, spines; 138, 139, broken grains. (Scale lines 134, 135 = 1 0 /im; 1 36 = 5 /rm; 1 37, 138 = 2 ixm; 1 39 = 1 pi.) NUMBER 64 41 Figures 140-146. — SEM of tribe Liabeae, Chrysactinium acaule pollen: 140, colpar view; 141, polar view; 142, colpus; 143, spines; 144-146, broken grains. (Scale lines 140, 141 = 10 /tm; 142 = 5 jim; 143-146 = 2 /mi.) 42 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 147-153. — SEM of tribe Liabeae, Munnozia lanceolata pollen: 147, colpar view; 148, polar view; 149, colpus; 150, spines; 151-153, broken grains. (Scale lines 147, 148 = 10 /nn; 149, 150 = 5 fiirr, 151-153 = 2 /um.) NUMBER 64 43 Figures 154-160. — SEM of tribe Liabeae, Munnozia pollen, broken grains: 154, 1 bb,Munnozia lanceolata. 156, M. perfoliata. 157-160, M. jussieui. (Scale lines 154-156, 158 = 2 u-n: 157, 159, 160 = 1 Mm.) 44 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 161-167. — SEM of tribe Liabeae, Munnozia pollen, broken grains: 161-165, M. lyrata. 166, 167, M. senecionidis. (Scale lines 161, 164, 165 = 1 ^m; 162, 163, 166, 167 = 2 NUMBER 64 45 Figures 168-173. — SEM of tribe Liabeae, Munnozia pollen (subgenus Kastnera). M. (Kastnera) annua: 168, colpar view; 169, polar view; 170, colpus; 171, spines. M. (Kastnera) tenera: 172, 173, broken grains. (Scale lines 168, 169 = 10 pm; 170, 171=2 jmi; 172, 173 = 1 /im.) 46 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 174-180. — SEM of tribe Liabeae, Munnozia ( Kastnera ) tenera, pollen, broken grains (179, arrows showing two layers of small columellae). (Scale lines = 1 nm.) NUMBER 64 47 Figures 181-189. — SEM of tribe Vernonieae, Pseudo stifjtia kingii, pollen: 181, colpar view; 3 82, polar view; 183, spines; 184-189, broken grains. (Scale lines 181, 182, 1 84 = 10 183, 185, 187 = 5 fim; 186, 188, 189 = 2 /an.) 48 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 190-195. — SEM of tribe Vernonieae, Vernonia fuertesii pollen: 190, colpar view; 191, polar view; 192, spines; 193-195, broken grains. (Scale lines 190, 191 = 10 /tm; 192, 194, 195 = 2 urn; 193 = 0.5 /mi.) NUMBER 64 49 Figures 196-202. — SEM of tribe Vernonieae, Blanchetia and Bishopalea pollen. Blanchetia heterotricha: 196, surface with spines; 197-199, broken grains. Bishopalea erecta: 200, whole grains; 201, polar view; 202, surface. (Scale lines 196, 197, 202 = 5 /im; 198, 199 = 2 /jiii; 200, 201 = 10 /an.) 50 SMITHSONIAN CONTRIBUTIONS TO BOTANY Figures 203-208. — SEM of tribe Vernonieae, Heterocypsela and Phyllocephalum pollen. Heter- ocypsela andersonii: 203, equatorial view; 204, broken tectum. Phyllocephalum scabridum: 205- 207, grains; 208, reticulum, showing two strata. (Scale lines 203, 205-208 = 10 /um; 204 = 2 uni.) REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION Manuscripts intended for series publication receive substantive review (conducted by their originating Smithsonian museums or offices) and are submitted to the Smithsonian Institution Press with Form SI-36, which must show the approval of the appropriate authority designated by the sponsoring organizational unit. Re- quests for special treatment — use of color, foldouts, case-bound covers, etc. — require, on the same form, the added approval of the sponsoring authority. Review of manuscripts and art by the Press for requirements of series format and style, completeness and clarity of copy, and arrangement of all material, as outlined below, will govern, within the judgment of the Press, acceptance or rejection of manuscripts and art. Copy must be prepared on typewriter or word processor, double-spaced, on one side of standard white bond paper (not erasable), with 1 Va" margins, submitted as ribbon copy (not carbon or xerox), in loose sheets (not stapled or bound), and accompanied by original art. Minimum acceptable length is 30 pages. Front matter (preceding the text) should include: title page with only title and author and no other information, abstract page with author, title, series, etc., following the established format; table of contents with indents reflecting the hierarchy of heads in the paper; also, foreword and/or preface, if appropriate. First page of text should carry the title and author at the top of the page; second page should have only the author’s name and professional mailing address, to be used as an unnumbered footnote on the first page of printed text. Center heads of whatever level should be typed with initial caps of major words, with extra space above and below the head, but no other preparation (such as all caps or underline, except for the underline necessary for generic and specific epithets). Run-in paragraph heads should use period/dashes or colons as neces- sary. Tabulations within text (lists of data, often in parallel columns) can be typed on the text page where they occur, but they should not contain rules or numbered table captions. Formal tables (numbered, with captions, boxheads, stubs, rules) should be submitted as carefully typed, double-spaced copy separate from the text; they will be typeset unless otherwise requested. If camera-copy use is anticipated, do not draw rules on manuscript copy. Taxonomic keys in natural history papers should use the aligned-couplet form for zoology and may use the multi-level indent form for botany. If cross referencing is required between key and text, do not include page references within the key, but number the keyed-out taxa, using the same numbers with their corresponding heads in the text. Synonymy in zoology must use the short form (taxon, author, year: page), with full reference at the end of the paper under “Literature Cited.” For botany, the long form (taxon, author, abbreviated journal or book title, volume, page, year, with no reference in "Literature Cited”) is optional. Text-reference system (author, year: page used within the text, with full citation in “Literature Cited" at the end of the text) must be used in place of bibliographic footnotes in all Contributions Series and is strongly recommended in the Studies Series: “(Jones, 1910:122)" or “...Jones (1910:122).” If bibliographic footnotes are required, use the short form (author, brief title, page) with the full citation in the bibliography. Footnotes, when few in number, whether annotative or biblio- graphic, should be typed on separate sheets and inserted imme- diately after the text pages on which the references occur. Exten- sive notes must be gathered together and placed at the end of the text in a notes section. Bibliography, depending upon use, is termed “Literature Cited," “References," or “Bibliography." Spell out titles of books, articles, journals, and monographic series. For book and article titles use sentence-style capitalization according to the rules of the lan- guage employed (exception: capitalize all major words in English). For journal and series titles, capitalize the initial word and all subsequent words except articles, conjunctions, and prepositions. Transliterate languages that use a non-Roman alphabet according to the Library of Congress system. Underline (for italics) titles of journals and series and titles of books that are not part of a series. Use the parentheses/colon system for volume (number): pagination: “10(2):5-9." For alignment and arrangement of ele- ments, follow the format of recent publications in the series for which the manuscript is intended. Guidelines for preparing bibli- ography may be secured from Series Section, SI Press. Legends for illustrations must be submitted at the end of the manuscript, with as many legends typed, double-spaced, to a page as convenient. illustrations must be submitted as original art (not copies) accompanying, but separate from, the manuscript. Guidelines for preparing art may be secured from Series Section, SI Press. All types of illustrations (photographs, line drawings, maps, etc.) may be intermixed throughout the printed text. They should be termed Figures and should be numbered consecutively as they will appear in the monograph, if several illustrations are treated as components of a single composite figure, they should be desig- nated by lowercase italic letters on the illustration; also, in the legend and in text references the italic letters (underlined in copy) should be used: “Figure 9b." Illustrations that are intended to follow the printed text may be termed Plates, and any components should be similarly lettered and referenced: “Plate 9b.” Keys to any symbols within an illustration should appear on the art rather than in the legend. Some points of style: Do not use periods after such abbrevi- ations as “mm, ft, USNM, NNE." Spell out numbers “one" through “nine” in expository text, but use digits in all other cases if possible. Use of the metric system of measurement is preferable; where use of the English system is unavoidable, supply metric equiva- lents in parentheses. Use the decimal system for precise meas- urements and relationships, common fractions for approximations. Use day/month/year sequence for dates: “9 April 1976.” For months in tabular listings or data sections, use three-letter abbre- viations with no periods: “Jan, Mar, Jun,” etc. Omit space between initials of a personal name: “J.B. Jones.” Arrange and paginate sequentially every sheet of manu- script in the following order: (1) title page, (2) abstract, (3) con- tents, (4) foreword and/or preface, (5) text, (6) appendixes, (7) notes section, (8) glossary, (9) bibliography, (10) legends, (11) tables. Index copy may be submitted at page proof stage, but plans for an index should be indicated when manuscript is sub- mitted.