Revisionary Study Research Articles

North American-East Asian Similarities in the Genus Ptychomitrium (Bryopsida)

A common species in eastern Asia, Ptychomitrium polyphylloides (C. Miill.) Jaeg. is synonymous with P. gardneri Lesq. from western North America. Ptychomitrium leibergii Best, from North America, is synonymous with P. sinense (Mitt.) Jaeg., which is widespread in eastern Asia. Both P. sinense and P. gardneri have disjunct East Asian-North American distributions. These two species are nearly sympatric in eastern Asia, but one occurs in dry forests of western North America while the other is confined to dry areas of south-central North America. During a revisionary study of Chinese Ptychomi- trium, we discovered that two common species of the genus in East Asia-P. sinense (Mitt.) Jaeg. and P. polyphylloides (C. Miill.) Jaeg.-are identical to P. leibergii Best and P. gardneri Lesq., respectively, both described from western North America. Pty- chomitrium leibergii, described by Best in 1906, is synonymous with P. sinense (based on Glyphomi- trium sinense Mitten 1865), while P. polyphylloides (based on Brachysteleum polyphylloides C. Miill.) is synonymous with P. gardneri, described by Les- quereux. These synonyms have important nomen- clatural consequences and also markedly expand the distributional range of these two species.

New Combinations in the Tribe Poraneae (Convolvulaceae) for the Flora of China

New combinations are made for 10 taxa formerly included in the genus Porana s.l. in the account of the Convolvulaceae for the Flora Reipublicae Popularis Sinicae, which are now considered to belong in the genera Dinetus, Poranopsis, and Tridynamia, as follows: Dinetus decorus, D. dinetoides, D. duclouxii, D. grandiflorus, D. truncatus, Poranopsis discifera, P. sinensis, Tridynamia megalantha, T. sinensis var. sinensis, and T. sinensis var. delavayi. A key to the genera is provided, and 11 additional taxa formerly accepted in Porana are reduced to synonymy. The genus Porana N. L. Burman comprises 57 nominate taxa, the majority distributed in mainland tropical Asia. Twenty-eight (14 species, 14 varieties) were recognized in the account of the Convolvulaceae for the Flora Reipublicae Popularis Sinicae (Fang & Huang, 1979). My revisionary study of the genus Porana and other taxa of the tribe Poraneae H. Hallier (Staples, 1987) indicates that there has been considerable overdescription in the genus Porana, and the number of species is actually much lower than the number of named species. Although a comprehensive taxonomic revision of Porana s.l. has been submitted to Systematic Botany Monographs (Staples, submitted manuscript), several new combinations, and reduction of additional names to synonymy, are needed for the account of the Convolvulaceae for the Flora of China, which is scheduled for publication before the revision will appear. This paper makes 10 new combinations needed for use in the Flora, and reduces to synonymy 11 other taxa of Porana recognized by Fang & Huang (1979). To summarize the major conclusions of the revision, Porana s.l. proved to be a polyphyletic assemblage comprising four distinct groups of species; a preliminary analysis indicated that these four species groups are not a monophyletic lineage inter se (Staples, 1987). I recognize them at generic rank, emending the generic concept of Porana to accommodate one group comprising three species, and take up the previously established generic names Tridynamia Gagnepain (1950), Poranopsis Roberty NOVON 3: 198-201. 1993. (1952), and Dinetus Sweet (1825) for the other three groups of species. Furthermore, the genus Cordisepalum Verdcourt is maintained as a genus distinct from and closely related to Poranopsis. These five genera may be distinguished by the following key. Full descriptions of all taxa, distribution data, ecological and phenological summaries, and detailed nomenclatural discussion are presented in the revision (Staples, submitted manuscript) so they are omitted here. KEY TO GENERA OF ASIATIC PORANEAE la. Leaf venation pinnate, or with a single basal pair of secondary veins; calyx about half as long as the corolla, sepals all + equal in length .. P orana s.s. lb. Leaf venation palmate; calyx minute, usually < 1/o as long as the corolla, if larger then sepals markedly unequal. 2a. Plants entirely herbaceous; flower buds with a minute apical tuft of hairs, otherwise glabrous externally Dinetus 2b. Plants woody (often with herbaceous stem tips); flower buds hairy externally. 3a. Sepals unequal in flower; fruiting sepals with 3-9 parallel longitudinal veins Tridynam ia 3b. Sepals (sub)equal in flower; fruiting sepals with 1 evident midvein. 4a. Corolla funnelform, white; fruit sessile on torus ........ Poranopsis 4b. Corolla rotate, yellowish; fruit borne on a slender stalk ...... Cordisepalum As construed here, Porana s.s. and Cordisepalum do not occur within the geographic range of the Flora of China. Some species of Poranopsis, Dinetus, and Tridynamia occur within the political boundaries of the People's Republic of China, and the following new combinations are proposed for them.

Ezra Pound and the symbolist inheritance

In this revisionary study of Ezra Pound's poetics, Scott Hamilton exposes the extent of the modernist poet's debt to the French romantic and symbolist traditions. Whereas previous critics have focused on a single influence, Hamilton explores a broad spectrum of French poets, including Theophile Gautier, Tristan Corbiere, Jules Laforgue, Remy de Gourmont, Henri de Regnier, Jules Romains, Laurent Tailhade, Paul Verlaine and Stephane Mallarme. This exploration of Pound's canon demonstrates his logic in borrowing from the French tradition as well as a paradoxical circularity to his poetic development.

Realignments in American Croton (Euphorbiaceae)

Revisional studies in Croton to achieve a more nearly phylogenetic classification necessitate a number of nomenclatural changes. The genera Crotonopsis and Eremocarpus are reduced to sec- tions of Croton, and the new names Croton mi- chauxii and C. willdenowii are proposed for the two species of Croton sect. Crotonopsis. For Croton sect. Julocroton, the 25 species accepted are enu- merated, including 19 new names or new combi- nations. In the course of an unpublished review of the sections of Croton L., I have found it necessary to make a number of changes in classification. Con- sequently, in order to be able to cite representative species for the 38 sections of Croton, a number of nomenclatural changes must be validated. The generic circumscription of Croton that has been almost universally accepted by systematists for more than a century is that proposed by Jean Muell- er (1865, 1866, 1873). In contrast to the earlier classification of Baillon (1858), Mueller recognized only four genera in the Croton alliance (subtribe

Revision of Disepalum (Annonaceae)

The relationship between the Asian generaDisepalum Hook. f. andEnicosanthellum Ban (Annonaceae) is examined in a revisionary study. Despite the disparate perianth morphologies of the two genera, other characters support inclusion of all species in a single genus with nine species. The monocarp stalks in all species are found to be derived from the receptacle rather than the ovary, and are thus not homologous with those found in most other Annonanceae genera. Cladistic analysis indicates that the species on the mainland of Asia and of montane habitats are more primitive than the species of the lowlands and of Sumatra and Borneo. Two new combinations,Disepalum petelotii andD. plagioneurum, are necessary as a result of merging the two genera, and one species,D. aciculare, is described as new.

Gouania axilliflora (Rhamnaceae), a New Species from Peru

The new species, Gouania axilliflora, is described from the province of Cajamarca, Peru. The resemblance to the Brazilian Alvimiantha tricamerata Grey-Wilson is discussed. The genus Gouania L., comprising a few dozen species of tendriliferous, tropical vines, is more in need of revision than any other genus of Rhamnaceae. After 20 years of desultory study, I estimate that of the more than 60 specific epithets proposed for American species, fewer than 25 will survive a critical revisionary study. And I predict that even after critical revision, the distinctions between the survivors will continue to be weakened by the presence of intermediate specimens only arbitrarily referable to one or another recognized species. That is to say that most species of the genus are poorly differentiated. It is therefore remarkable to find, among recent Peruvian collections, a species so distinctive that one could make a case for the erection of a new genus. I have no hesitation in proposing this new taxon even in the absence of information on the base of the plant or the mature fruits. Gouania axilliflora M. C. Johnst., sp. nov. (fig. 1.-TYPE and only known collection: Peru, Department of Cajamarca, Province of Cajabamba, Condebamba valley, CajabambaCajamarca road, 78?9'W, 7?33'S, 2600-2100 m, 15 Feb 1983, D. N. Smith & R. Vdsquez M. 3396 (holotype: TEX!; isotypes: AMAZ, CPUN, HUT, MO!, USM). Plantae scandentes cirrhiferi, caulibus costatis sulcatis hispidis; petioli 2-4 mm longi; laminae foliorum anguste elliptico-ovatae 15-25 mm longae 7-11 mm latae, mucronatae, supra fere glabrae venatione impressa, subtus hispidae venis lateralibus 2 costaque prominentibus valde campto-acrodromis; stipulae 4-5 mm longae; axillae distales 2-florae; pedunculi pedicellique sub anthesi nulli; sepala ? 1 mm longa; petala 0.7-0.9 mm longa cuneata exunguiculata spathulata incisurata; discus glaber tenuis planus integer; styli rami ?0.6 mm longi; fructus (non satis maturus) ? 6 mm longus ? 5 mm diametro exalatus. Vine [fide collectors]; stems slender, with 8 relatively strong nearly hairless ribs alternating with 8 weaker ribs, the 16 grooves bearing dense, longitudinally oriented, crisped white, translucent hairs 0.4-1 mm long, each hair more or less appressed basally and weakly spreading distally; tendrils slender, 3-5 cm long, hooklike to tightly coiled. Leaves: petioles 2-4 mm long; blades narrowly elliptic-ovate, 15-25 mm long, 7-11 mm wide, adaxial surface more or less hairless and dull olive green and with impressed venation, abaxial surface between major veins with sparse hairs 0.4-0.9 mm long and with prominent venation including midrib and 2 arcuate lateral primaries departing midvein at extreme base of blade and conspicuously acrodrome to mucronate tip of blade; stipules linear, 4-5 mm long, deciduous. Inflorescence: 2-flowered glomerules in distal axils; flowers green, essentially sessile at anthesis; cupular structure at base of flower (interpreted as an expanded receptacle) hemispherical, about 1.5 mm in diameter at anthesis, densely hairy externally, adnate to side of ovary; sepals 5, triangular, about 1 mm long and wide, hairy abaxially, glabrous adaxially, falling sooner or later after anthesis or rarely persisting on younger fruit; petals 5, yellow-green(?), 0.7-0.9 mm long, cuneate but not clawed, distally broadened and spatulate, concavo-convex and hooding anthers, hairless, tip-notched about 0.1-0.2 the total length; filaments 5, about 0.3 mm long, anthers about 0.3 mm long; disk hairless, thin, flat, covering top of ovary and extending as an extremely narrow thin unlobed flange at the periphery of the top of the ovary; style-branches 3, clavellate, about 0.6 mm long, excurved. Fruit (not perfectly mature) about 6 mm long, about 5 mm in diameter, 3-angled, wingless (at this stage of maturity, at least); fruiting pedicels up to 4-5 mm long; ma-

A revisionary study of the genus Mobula Rafinesque, 1810 (Chondrichthyes: Mobulidae) with the description of a new species

On the basis of morphometries, external morphology, colouration, and tooth and denticle morphology, the genus Mobula Rafinesque (1810) consists of nine living species, one of which (M. munkiana sp. nov.) is described here as new. Although the species can be grouped into six divisions based on morphological and morphometric similarities, no formal subdivision of the genus is here proposed, because of the lack of sufficient information on some of the species involved: the study presented here is the first systematic approach towards an overall revision of this genus. Neotypcs are designated for M. eregoodootenkee (Cuvier, 1829) and M. tarapacana (Philippi, 1892). Mobula coilloti Cadenat & Rancurel (1960) and M. formosana Teng (1962) are considered to be synonyms of M. tarapacana (Philippi, 1892), M. rancureli Cadenat (1959) is a synonym of M. japanica (Muller & Henle, 1841), M. lucasana Beebe & Tee-Van (1938) is a synonym of M. thurstoni (Lloyd, 1908), M. draco (Gunther, 1872) is a synonym of M. kuhlii (Valenciennes in Muller & Henle, 1841), and Ceratobatis robertsii Boulenger (1897) is a synonym of M. hypostoma (Bancroft, 1831). The relationships between M. japanica and M. mobular are not clear, and must await more detailed information on M. mobular. For each species principal synonyms are given and discussed, and diagnosis, description, morphometric data and geographic range are presented. A key is included to aid in species identification. All Mobula species are marine tropical-subtropical, and three are probably worldwide; M. japanica and M. mobular range into warm temperate waters. Of the nine species accepted here, five are found in the Indo-West Pacific, four in the eastern Pacific, two in the western Atlantic, four in the eastern Atlantic, and one in the Mediterranean.

Remijia chelomaphylla (Rubiaceae), a New Species from Peru

Remijia chelomaphylla is known from one locality in eastern Peru. Although also distinct from other species of the genus because of vestiture and size and shape of the capsule, the type material is unique because it has lobed or sinuate leaf margins, rarely seen in the Rubiaceae. The wall of the foliar trichomes is sculptured with cuticular micropapillae. This sculpturing is reported to be found in other members of the Cinchoneae. Remijia DC. is a member of the tribe Cincho- neae (Rubiaceae). The genus was first described from Brazil in 1829, but species are now known from Venezuela, Colombia, Peru, and Bolivia. No species is known to be widespread throughout the entire range of the genus; in fact, most are restricted to relatively limited ranges. Remijia chelomaphylla is known from the val- ley of the Rio Monz6n, a tributary of the Rio Huallaga of eastern Peru. The type was col- lected in 1963, prior to extensive disturbance of this locality. Rapid deforestation along the Rio Huallaga and its tributaries has occurred within the past fifteen years because of a large increase in population (Gentry and Lopez-Pa- rodi 1980). The area is also a source of coca, Erythroxylum coca Lam. var. coca (Bohm et al. 1982), and much of the land has been cleared for its cultivation. Efforts to relocate the plant have been unsuccessful. Jose Schunke Vigo has collected extensively in the vicinity, but this species was not included in his collections. I was unable to locate the species on a recent field trip, and it is probable that the plant is rare or possibly even extinct. Although currently under revisionary study, the genus is estimated to contain about 25 species. Remijia chelomaphylla is unlike other species examined, especially in the apparent lobation of the leaves and the characteristic leaf vestiture.

Two New Species in the Genus Leochilus (Orchidaceae)

The following new species of Leochilus (Orchidaceae) are described: Leochilus puertoricensis from Puerto Rico, and Leochilus hagsateri from southwestern Mexico. Both species have sister species known from nearby locations that have prevented their recognition as distinct species, even though collections of both have beeen known for many years. Although authors have made over thirty combinations in the genus Leochilus, two species collected a number of times and known for 50 years or more have gone undescribed. Both species have closely related sister species, and specimens of the two unnamed species, one each from Puerto Rico and Mexico, have heretofore been identified as L. labiatus (Sw.) Kuntze and L. carinatus (Knowles & Westc.) Lindley, respectively. A thorough revisionary study of Leochilus (Chase unpubl.) revealed their distinctness from the other seven species in the

The Infrageneric Classification of Biscutella (Cruciferae)

A brief historical taxonomic sketch ofBiscutella is given. Revisionary study of the genus throughout its geographic areas has led to an infrageneric classification. Two subgenera are recognized, one with two sections.

A New Species of Litoria (Anura: Hylidae) from New South Wales, Australia

adjacent Great Dividing Range have a rich frog fauna which was the subject of a revisionary study by Moore (1961). There is no doubt that Moore's monograph stimulated work on the fauna throughout the continent. Included in the species described subsequently from eastern New South Wales are five species of the hylid genus Litoria Tschudi: L. brevipalmata Tyler, Watson and Martin; L. olongburensis Liem and Ingram; L. revelata Ingram, Corben and Hosmer; L. subglandulosa Tyler and Anstis (replacement name for L. glandulosa Tyler and Anstis, 1975, preoccupied); and L. tyleri Martin, Watson, Gartside, Littlejohn and Loftus-Hills. Here we describe an additional species of Litoria from sites upon the Great Dividing Range centered upon Armidale and Glen Innes.

Chapman's "Caesar and Pompey" and the Uses of History

Previous articleNext article No AccessChapman's "Caesar and Pompey" and the Uses of HistoryRichard S. IdeRichard S. Ide Search for more articles by this author PDFPDF PLUS Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinkedInRedditEmail SectionsMoreDetailsFiguresReferencesCited by Modern Philology Volume 82, Number 3Feb., 1985 Article DOIhttps://doi.org/10.1086/391383 Views: 4Total views on this site Citations: 2Citations are reported from Crossref Copyright 1985 The University of ChicagoPDF download Crossref reports the following articles citing this article:Hediye ÖZKAN George Chapman’ın Ceasar and Pompey Trajedisi ve Arabulucu Cato, Selçuk Üniversitesi Edebiyat Fakültesi Dergisi , no.4444 (Dec 2020): 55–70.https://doi.org/10.21497/sefad.845326Toshinori IDE Mechanism of Cell Proliferation—Cell Cycle, Oncogenes, and Senescence, YAKUGAKU ZASSHI 126, no.1111 (Nov 2006): 1087–1115.https://doi.org/10.1248/yakushi.126.1087

The biology of false wireworms and their adults (soil-inhabiting Tenebrionidae) (Coleoptera): a review

AbstractSoil-dwelling tenebrionid larvae have developed in three evolutionary lines and are pests in all faunal regions. Description of and keys to most of the important species are available. However, revisional studies, on both adult and larval characters, are needed in many groups. The basic number of non-sex chromosomes appears to be 18, but the number and type of sex chromosomes are variable. Host ranges of both larvae and adults are usually extensive but usually only include plants. Development rates are related mainly to temperature, with life-cycles of one, two or three years. Larval development periods are rather long, and adults are long-lived. The number of instars is high, often above ten and frequently variable within a species. Sampling is best carried out by direct counts and pitfall trapping. In the species studied, mating follows a regular pattern and pheromones are involved. Species show definite temperature, moisture and soil-type preferences and these influence both spatial and temporal distribution. Recorded parasitoids include Diptera (mainly tachinids), Hymenoptera, mites, bacteria and fungi. Birds, carabids and a variety of other vertebrates are the main predators, but abdominal secretions and postural mechanisms provide some defence. Cultural control was formerly widely practised but has given way to chemical control.

A revisionary study of the taxon Cillenus Samouelle, 1819 and related forms (Coeloptera: Carabidae, Bembidiini)

AbstractThe 23 species of Bembidiini revised in this paper were earlier referred variously to subgenus (or genus) Cillenus Sam. or to closely allied taxa. It is shown that the five New Zealand species are taxonomically quite isolated and different from Bembidion; for these the new genus Zecillenus was erected. All remaining species are true members of genus Bembidion belonging to five different subgenera, all previously named. Subgenus Desarmatocillenus Netolitzky contains 13 species distributed from Tasmania to Japan; three other subgenera are restricted to China, Formosa and Japan, respectively. The single European species, B. laterale Samouelle, constitutes subgenus Cillenus s. str.—Six species, Zecillenus embersoni and 5 Bembidion, adelaideae, foochowense, hebridarum, kasaharai, seticorne, are described as new.—None of the subgenera shows clear affinities to any other subgenus of Bembidion. This was judged primarily on the structures of the internal sac of penis.

The Affinities of Costesia and Neosharpiella and Notes on the Gigaspermaceae (Musci)

The genera Costesia Ther. and Neosharpiella Robins. & Delg. are- transferred from the Funariaceae to the Gigaspermaceae. Neosharpiella turgida (Mitt.) Robins. & Delg. is reported for the first time from Chile. The morphology of the six genera currently placed in the Gigaspermaceae is compared and generic inter-relationships discussed. SEM micrographs of the spores of four genera are given. In the revised sense presented, the Gigaspermaceae have a widely scattered, predominantly Southern Hemispheric distribution. Nomenclatural notes on several taxa are appended and several lectotypes are proposed. In the course of revisionary study on the Funariaceae, I have had occasion to examine material of the Central and South American genus Neosharpiella Robins. & Delg. and the South American genus Costesia Ther., as well as type material of most of the species currently placed in the Gigaspermaceae. This paper presents evidence that Neosharpiella and Costesia are improperly placed within the Funariaceae and that they are better placed in the Gigaspermaceae sensu Brotherus. The Funariaceae and the Gigaspermaceae were both placed in the Funariales by Broth- erus (1924). The Funariaceae may be characterized by the following set of characters: 1. annual or biennial habit; 2. erect, little-branched stems (very rarely prostrate and more or less rhizomatous); 3. leaf cells thin-walled and without thickened corners; 4. costae well developed, very rarely absent; 5. usually autoicous sexual condition; 6. perigonial pa- raphyses with globose terminal cells; 7. capsules various, with stomata enclosed by a single guard cell and a slit-like pore; and 8. spores relatively large, though usually less than 50 gm in diameter. The Gigaspermaceae, on the other hand, may be characterized by: 1. apparently perennial habit; 2. prostrate, rhizomatous stems giving rise to erect branches; 3. leaf cells thin- or firm-walled, often undulate in outline and usually with thickened corners; 4. costae variously developed; 5. usually synoicous or paroicous sexual condition; 6. perigonial paraphyses filiform, without inflated terminal cells; 7. capsules gymnostomous, with stomata enclosed by two guard cells (rarely imperfectly formed); and 8. spores large, usually in excess of 50 gm diam. Collectively, Costesia and Neosharpiella are known from South America and Mexico where they grow as weeds on soil. Although appearing acrocarpous, the plants exhibit an unusual growth habit similar to that of the unrelated genus Drummondia and to which neither the term acrocarpous nor pleurocarpous can be applied without qualification. They proliferate by fragile, nearly hyaline rhizomes that give rise to the more visible erect branches. The apices of the leafy shoots give rise to sporophytes with exserted, broadly obovoid capsules resembling those of some Physcomitrium species, but which have a characteristic strongly wrinkled appearance when dry. Costesia was described by Theriot (1917) on the basis of two Chilean specimens col- lected by N. Costes in 1915. To my knowledge, the sole representative of the genus, C. 007-2745/80/466--476$1.35/0

CLETHRA OCCIDENTALIS (CLETHRACEAE): A RE‐EVALUATION

SummaryA statistical analysis of variation in Clethra occidentalis sensu lato was undertaken to reexamine its limits. Methods used are t‐tests and principal component analysis. From the evidence presented, C. occidentalis is redefined to include only Jamaican populations. It is suggested that C. hondurensis be recognized and that the disposition of other Mexican and Central American populations cannot be resolved without further revisionary study of the entire C. occidentalis complex.

New Taxa and New Combinations in the Orthotrichaceae of Mexico

New Taxa and New Combinations in the Orthotrichaceae of Mexico

Phyletic and Cladistic Relationships in Lipochaeta (Compositae)

A recent computer technique, Character Compatibility, was used to examine cladistic relationships among the 27 taxa (two sections, Lipochcaeta and Aphanopappus) of the genus Lipochaeta. The results were compared with a previous, intuitively derived, concept of relationships within the genus. The computer analysis produced four possible evolutionary trees. One tree was constructed primarily from reproductive characters which were previously shown to be conservative evolution- arily. The latter was in closer agreement with the intuitive analysis than were the other three. The computer study was helpful in assessing the previously obscure relationships of two taxa, Lipochaeta bryanii and L. degeneri. The intuitive phylogram and the computer cladograms were in close agreement regarding the relationships of the two sections to each other. Character Compatibility promises to be quite useful in plant systematics, but it should be used only after the plant group is well under- stood, i.e., it should be used as a final step rather than in the initial phases of a revisionary study.

Biosystematics of Trichogramma and Trichogrammatoidea Species

erepresent a large group of minute parasitic wasps which attack eggs of various insects, many of which are of economic importance. The small size of trichogrammatids and the relatively uniform morphology of species within each genus have made taxonomic studies difficult, and this has resulted in many nomenclatorial problems. Descriptions of species have often been rather vague, frequently based on one sex, and, in the absence of good type specimens, verification of descriptions has not always been possible. The most recent and comprehensive revisionary study of the family has been made by Doutt & Viggiani (6) , who not only developed an excellent key for separation of the 70-odd genera and subgenera, but have also synonymized many of these groups. These authors have observed that the minuteness of these winged insects has apparently increased their chances of dispersion by wind, and as a result they are even found on remote isolated oceanic islands. Moreover, they are unique in that they occupy all kinds of habitats, from swampy marshlands to hot dry deserts, and occur on low-lying vegetation or in strictly arboreal habitats.

Evolution and Adaptive Radiation in Lipochaeta (Compositae) of the Hawaiian Islands

The Wedelia-like ancestors of Lipochaeta were probably dispersed to Hawaii by water, becoming established either on Kauai or Oahu. From the original colonizers, 25 species and four infraspecific taxa evolved. Polypoloidy played a major role in the formation of two independent evolutionary units. Among the diploids, the present distribution suggests that the species have evolved allopatrically on single islands through geographical isolation. The distribution pattern of the tetraploids suggests that they evolved allopatrically on separate islands rather than within different habitats on single islands as with the diploids. An analysis of leaf flavonoid chemistry has shown that the diploids synthesize few compounds, all flavonols. In contrast, the tetraploids elaborate many compounds, both flavones and flavonols. Although the diploids show marked morphological differences, their flavonoid systems have diverged very little. The broader array of compounds in the tetraploids suggests that their chemical systems are still under intense selection pressures. Adaptive radiation describes the process by which a group of organisms adapts to a broad variety of situations (Carlquist, 1970). One can argue that most groups have radiated to some extent, so to be more meaningful the term should be restricted to a small group of organisms that illustrates unexpected diversification. For adaptive radiation to become maximal, an array of habitats must be available to which few of the species initially are adapted. Perhaps the best natural laboratories for the study of this phenomenon are volcanic oceanic islands because: 1) an island mass of volcanic origin often has an environmental range from tidewater marsh through dry lowland to wet forest, or even to an alpine zone, providing a broad variety of habitats concentrated into a relatively small geographical area; and 2) plants that are dispersed to and become established on oceanic islands are isolated from mainland ancestral populations, and therefore, the initial gene pool is limited to very few individuals. If one assumes there are no later introductions of the same taxon, then all of the diversity that develops can be traced to the original colonizers. To further our understanding of adaptive radiation and evolution in an island-centered plant group, Lipochaeta De Candolle, endemic to the I This paper is a portion of a Ph.D. dissertation submitted to The Ohio State University, Columbus. I thank Tod F. Stuessy for his advice throughout the study and acknowledge support for the field work from National Science Foundation Grant GB41175 and from Sigma Xi Grants-in-Aid of research for chemical supplies. 2 Botany, The Ohio State University, Columbus, Ohio 43210. This content downloaded from 207.46.13.124 on Wed, 22 Jun 2016 06:11:00 UTC All use subject to http://about.jstor.org/terms 384 SYSTEMATIC BOTANY [Volume 1 Hawaiian Islands, was selected for study. In addition to morphological comparisons, many taxa were analyzed for chromosome number (Gardner, 1976) and leaf flavonoid chemistry. Data from the latter were utilized to determine whether flavonoids are correlated with patterns of adaptive radiation. Numerous studies have documented the utility of these compounds for indicating degrees of relationship among plant taxa (Turner &< Alston, 1959; Alston & Turner, 1962; Levin, 1967; Crawford, 1974; Giannasi, 1975), yet very few studies (Gupta c Gillett Sc Lim, 1970) have attempted to use chemical data to help interpret patterns of evolution among island-centered plant groups. Lipochaeta is listed (Carlquist, 1965) as an example of a genus that has undergone considerable adaptive radiation. Detailed analysis, however, including a revisionary study that recognizes 25 species (Gardner, 1976), has shown that although the genus represents an excellent system for determining evolutionary relationships, it illustrates only a moderate degree of adaptive radiation. The purpose of this paper is to discuss the origin of Lipochaeta and indicate evolutionary and adaptive trends within the genus.