Bulletin of the British Museum (Natural History) The lichen genus Usnea subgenus Neuropogon F. Joy Walker Botany series Vol 13 No 1 28 March 1985 The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in four scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology, and an Historical series. Papers in the Bulletin are primarily the results of research carried out on the unique and ever-growing collections of the Museum, both by the scientific staff of the Museum and by specialists from elsewhere who make use of the Museum's resources. Many of the papers are works of reference that will remain indispensable for years to come. Parts are published at irregular intervals as they become ready, each is complete in itself, available separately, and individually priced. Volumes contain about 300 pages and several volumes may appear within a calendar year. Subscriptions may be placed for one or more of the series on either an Annual or Per Volume basis. Prices vary according to the contents of the individual parts. Orders and enquiries should be sent to: Publications Sales, British Museum (Natural History), Cromwell Road, London SW75BD, England. World List abbreviation: Bull. Br. Mus. nat. Hist. (Bot.) Trustees of the British Museum (Natural History), 1985 The Botany series is edited in the Museum's Department of Botany Keeper of Botany: Mr J. F. M. Cannon Editor of Bulletin : Mr J . R . Laundon Assistant Editor: Dr A. J. Harrington ISBN 565 08004 ISSN 0068-2292 Botany series Vol 13 No 1 pp 1-130 British Museum (Natural History) Cromwell Road London SW7 5BD Issued 28 March 1985 The lichen genus Usnea subgenus Neuropogon t F. Joy Walker Department of Botany, British Museum (Natural History), Cromwell Road, London SW7 5BD Contents Synopsis 1 Introduction .. Previous collections and research Taxonomic review Materials and methods Results .. Morphology . Chemistry ... Distribution . BRITISH MUSEUM JNAtUNAL HISTORY) " PRESENTED .GENERAL LIBRARY Ecology Phytosociology 27 Biogeography 31 Discussion 36 Circumscription of the species 36 Generic concept and infrageneric classification 41 The subgenus Neuropogon 43 Key to the species 44 The species 47 1. U.acanthella 47 2. U. acromelana 48 3. U. antarctica 55 4. U. aurantiaco-atra 62 5. U.ciliata 74 6. U.durietzii 78 7. U. neuropogonoides 80 8. U. patagonica 82 9. U.perpusilla 85 10. U. pseudocapillaris 89 11. U. sphacelata 92 12. U. subantarctica 99 13. U. subcapillaris 104 14. U.taylorii 108 15. U. trachycarpa 110 Appendix I. Allied taxa 115 Appendix II. Excluded taxa 120 Acknowledgements 122 References 122 Index 129 Synopsis A world-wide taxonomic revision is presented for the lichen genus Usnea subgenus Neuropogon. In summary, this subgenus is here circumscribed to include polar-alpine species with a yellow to yellow-green thallus with varying black or violaceous black pigmentation; a black, or rarely brown, matt apothecial disc, often with excipular rays; and almost always a saxicolous habitat. Fifteen species are recognised, of which three are new: Usnea patagonica, U. pseudocapillaris, and U. subantarctica. Two described species, U. durietzii and U. neuropogonoides, are included in the subgenus for the first time; one name change, U. Bull. Br. Mus. not. Hist. (Bot.) 13 (1): 1-130 Issued 28 March 1985 2 F. J. WALKER sulphured to U. sphacelata, has had to be made resulting from nomenclatural considerations; three new combinations are made: U. acanthella (Lamb) F. J. Walker, U. perpusilla (Lamb) F. J. Walker, and U. subcapillaris (D. Galloway) F. J. Walker. Several taxa are reduced to synonymy, including U. melaxantha with U. aurantiaco-atra and U. rohmederi with U. perpusilla. The taxonomic section is preceeded by an outline of previous research and accounts are given of morphology, chemistry, distribution, ecology, phytosociology, and biogeography. Reasons for accepting subgeneric status are given. In delimiting species much consideration has been given to the polymorphic nature of Usnea species and hence a wide range of infraspecific variation is often tolerated. A detailed account is provided for each species with particular emphasis attached to variation, chemistry, and species concepts. Three allied species belonging to Usnea subgenus Usnea which may be pigmented are included in the key to the species and are discussed in Appendix I. Introduction Previous collections and research Species of Usnea subgenus Neuropogon form one of the most conspicuous groups of bipolar- alpine lichens, often being the dominant fruticose lichens found in such inhospitable localities. Apart from a distinctive variegated black and yellow-green colouration (yellow to ochraceous in herbaria) and comparatively large size, members of the group are more easily collected than other associated saxicolous lichens which are frequently crustose. Consequently the group has often formed the major part of collections made by botanists and non-botanists alike on expeditions to the antarctic and arctic from the late eighteenth century to the present day. Despite their prominence there are still certain areas, in particular the northern Andes and even Central and North America, where the subgenus is extremely rare and difficult to interpret. Frequently the only known collections from these areas have been made by explorers and mountaineers, for example Whymper; further field work is needed to fill in the gaps that still exist in distributional patterns and to provide the essential ecological information which is particularly needed for the study of this group. The first arctic Neuropogon specimens to be brought back to Europe by explorers and described as new taxa by contemporary lichenologists were collected during the late eighteenth and early nineteenth centuries. Early literature was restricted to the description of individual species and the first group of these, Lichen sulphureus (Konig, 1772) and L. pallidus (Retzius, 1779), were based on material collected in Iceland, whilst Usnea sphacelata (Brown, 1823) was described from arctic North America. During the same era expeditions to southernmost areas of South America also brought back Neuropogon material. The earliest southern hemisphere species to be described, Lichen aurantiaco-ater (Jacquin, 1781), i.e. Usnea aurantiaco-atra, was based on collections made by Commerson in 1767 from the Magellan Straits area. Other notable collectors of new taxa during the early nineteenth century include Cavanilles (Acharius, 1803) and Gaudichaud (Persoon, 1828), who both collected on the Falkland Islands, as well as Poeppig (Nees & Flotow, 1835) from southern Chile. Material was also collected by Darwin from Tierra del Fuego in 1833 and in his journal (Darwin, 1979) he likens the abundance of the extensive swards of Neuropogon around Cape Horn to a variety of grass when viewed from a distance. By the early nineteenth century explorers were beginning to survey the antarctic continent. The first Neuropogon species, Usnea fasciata (= U. aurantiaco-atra}, to be described as new from Antarctica was brought back to America from the South Shetland Islands (Torrey, 1823), probably by a whaling party. Hooker's account of the botany of the voyage of the ships 'Erebus' and Terror' (Hooker, 1844-47) and 'Lichenes antarctici' (Hooker & Taylor, 1844) provided the earliest accounts of antarctic lichens as a group and also recorded the presence of Neuropogon species on lies Kerguelen, one of the isolated subantarctic islands in the southern Indian Ocean. These islands were revisited by other expeditions during the latter part of the nineteenth century, with Neuropogon taxa being described from collections made by the 'Venus Transit' Expedition (Crombie, 18760) and the 'Challenger' Expedition (Stirton, 1881). This was not long after the description (Nylander, 1866) of the first Australasian taxon, Neuropogon melaxanthus var. ciliatus (= Usnea ciliata). USNEA SUBGENUS NEUROPOGON 3 Major expeditions to Antarctica on which notable lichen collections were made have been adequately listed by Dodge (1948) for the period 1839-1935 and Lamb (1964) for 1901-1958, together with details of relevant publications and herbaria. Arctic collections have mainly been the province of Scandinavian expeditions and relevant historical collections have been ade- quately cited by Lynge in various publications (for example: Lynge, 1932, 1941). Late nineteenth and early twentieth century lichenologists who contributed to the knowledge of the group included Miiller (1888, 1895), Du Rietz (1926), Rasanen (1932), and Zahlbruckner (1903, 1917), although many of the taxa described by them were subsequently reduced to synonymy. Howe (1915) produced a useful compilation of type data of South American taxa, whilst Du Rietz (1926) provided the first comparative account of the southern hemisphere species although he excluded relevant chemical information. The three twentieth century lichenologists that feature most prominently in the study of Neuropogon are Motyka, Lamb, and Dodge. These authors produced detailed taxonomic studies that were mainly based on collections made by contemporary antarctic expeditions and surveys. Important collections included those made by members of the British Antarctic Survey, formerly 'Operation Tabarin' prior to 1946, and the Falkland Islands Dependencies Survey (F.I.D.S.) 1946-1961.* Others include the British, Australasian, New Zealand Antarctic Research Expedition 1929-31 (B.A.N.Z.A.R.E.), and the United States Antarctic Service Expedition 1940-41 (U.S. A.S.). Lamb (19480) also studied some of the rich collections made in Patagonia by Santesson from 1939-41 (S, UPS) which have also been invaluable to the present author during this reappraisal of the group. Motyka (1936-38) included 12 species in the subgenus Neuropogon and also described an additional two species in the subgenus Euusnea nom. illeg. (Article 21.3), i.e. Usnea, that are accepted here as belonging to the group. Motyka was the first author to include any indication of medullary chemistry in his account, which was limited to spot tests with potassium hydroxide solution (K). The most detailed monographic account was produced by Lamb (19390) who revised Motyka's work and recognised 13 species and their varieties and forms. Lamb was the first author to make a detailed study of medullary chemistry of the group, using thallus spot tests and microchemical crystal tests, and to use this information on a taxonomic basis. Lamb later (1964) had access to some thin-layer chromatographic (TLC) data provided by Hale. This was a major step forward, although sometimes too much weight was attached to minor chemical differences which would today be regarded as individual races, rather than distinct taxa, by most taxonom- ists. Lamb subsequently (1948a, 1964) emended and added to his earlier account, eventually recognising a total of 15 species of Usnea subgenus Neuropogon. He also realised that many of his individual forms were simply chemical 'phases' of a particular species that were only significant at a distributional and ecological level. There have been few reports of medullary chemistry of Neuropogon besides those of Lamb. These include Hawksworth & Moore (1969), Golubkova & Schapiro (1970), Kashiwadani (1970), Filson (1974), and Ghogomu & Bodo (1982), which frequently report the lack, rather than the presence, of depsidones. Lamb's work provided a sound basis for the identification of Neuropogon species world-wide with particular emphasis on antarctic and some South American taxa. Meanwhile Dodge and associates (Dodge & Baker, 1938; Dodge, 1948, 19656, 1973) had much narrower species concepts and described a series of additional species. Many existing infraspecific taxa were raised to specific rank, often without the study of extant type specimens. Dodge's work was confined to Antarctica and the subantarctic islands of the Indian Ocean in which he recognised 21 species (Dodge, 1973) and six species (Dodge, 1948, 19656) respectively, accepting a grand total of 30 species in Neuropogon, including one arctic and two Australasian taxa. Some of the taxa described by Dodge have been examined by Lamb (1964) and tentatively reduced to synonymy. Lamb's interpretations of Dodge's material have, for the most part, been followed * The collections made by 'Operation Tabarin' were referred to by Lamb as F.I.D.S. both on herbarium labels and in subsequent publications (Lamb, 1964); both names are used here following label details. 4 F. J. WALKER here, particularly where type specimens have not been made available for study; many of the species distinguished by Dodge were separated on minute morphological differences with little or no reference to chemical data. More recent work has primarily concentrated on ecological observations and several plant associations have been described based on Neuropogon-containing communities with other lichens and bryophytes. Examples include Smith & Corner (1973), Follmann (1965a, 1967), Lamb (1970), and Gimingham & Smith (1970). Taxonomic work has been limited to the preparation of local antarctic and subantarctic floras following Lamb's treatment and includes floras of South Georgia (Lindsay, 1975), the South Orkney Islands (Smith, 1973), the South Sandwich Islands (Longton & Holdgate, 1979), Marion Island (Lindsay, 1977b), Macquarie Island (Filson, 1981), Mac. Robertson Land (Filson, 1966, 1975), Wilkes Land (Filson, 1974), and Bouvet0y (J0rgensen, pers. comm.). These floras have been produced in conjunction with recent expeditions or the work of permanent scientific bases (for example, the British Antarctic Survey) that have been established around the antarctic continent by various nations. The aim of this present study is to give a comprehensive account of the species belonging to Usnea subgenus Neuropogon. Special emphasis has been placed on the circumscription of the species in the light of the wide range of infraspecific variation, new chemical data, and relationships with closely allied taxa. There has been no critical work on the subgenus since that of Lamb (19390, 19480, 1964). Taxonomic review Neuropogon was published as a genus by Nees & Flotow (1835) within the Usneaceae (= Parmeliaceae, Henssen & Jahns, 1973) and was based on two species N. poeppigii and N. antennarius. N. poeppigii was subsequently transferred from the genus by later authors, details of which are given by Lamb (1939a, 1964), commencing with the transfer to Chlorea by Nylander (1860) and, more recently, to the subgenus (Motyka, 1936-38) or genus Protousnea (Krog, 1976). Nylander (1860) is consequently considered to have emended the original concept of the genus since the original description of Nees & Flotow circumscribed both Protousnea and Neuropogon, based on common features of thallus colour, anatomy, apothecial shape, and disc colour. However the citation 'Neuropogon (Nees & Flotow) Nyl.' by Lamb (19390) is erroneous as no new combination was actually made by Nylander. Nylander (1860: 272) is also considered to have, in effect, selected N. antennarius as the lectotype of Neuropogon, consequently Motyka's (1936-38) citation of N. melaxanthus as the type species is superfluous. The genus Neuropogon was relegated to the rank of section or subsection of the genus Usnea Hill ex Browne (Laundon, 1984) by Montagne, in Gay (1852), and to a subgenus by Jatta (1900), although he (Jatta, 1909) later accepted Neuropogon as a distinct genus. Jatta's combination (Jatta, 1900) Usnea subgenus Neuropogon (Nees & Flotow) Jatta was based on two species, Usnea arboricola and U. soleirolii, that were not subsequently regarded as belonging to the subgenus and were finally included in Lethariella subgenus Lethariella (Krog, 1976). For this reason Jatta's combination was wrongly regarded as invalid by Motyka (1936-38) and omitted by Lamb (1964). However, in accordance with the Code (Article 63.2) the combination must be accepted, as the type of Neuropogon is, by implication, included in the combination. Conse- quently the full citation at subgeneric level is Usnea subgenus Neuropogon (Nees & Flotow, emend. Nyl.) Jatta, and the citation 'Usnea subgenus Neuropogon (Nees & Flotow) Motyka' (Motyka, 1936) is incorrect (Culberson, 1966). Materials and methods The following account is based primarily on collections in institutional herbaria. I have also had access to material collected in recent years by the following botanists: James (Patagonia, New Zealand), Henssen & Vobis (Patagonia), Follmann (South America), Santesson (Patagonia), Hertel (Marion and Prince Edward Islands), Engelskj0n (Bouvet0y), Galloway (New Zea- land), Bratt (Tasmania), S0mme, Angard (Dronning Maud Land), Seppelt (Knox Coast, McDonald and Macquarie Islands), Halls (Bolivia), and, in addition, extensive collections made USNEA SUBGENUS NEUROPOGON 5 by various members of the British Antarctic Survey (Antarctic peninsula and islands). The material referred to in this study has been subjected to thin-layer chromatography (TLC) by means of standard methods (Culberson, 1972; Culberson & Johnson, 1976; Culberson et al., 1981; Walker & James, 1980). Type specimens have been examined and tested by TLC unless otherwise stated. In the text denotes sporadic or low concentration of a given substance. Details of morphology and anatomy were also studied using the scanning electron microscope (SEM) (Cambridge Stereoscan ISI 60A). Species descriptions are based on either type material or characteristic specimens in rare instances where the type is very atypical. Details of variation, species concept, chemistry and distribution are given for each species, together with notes on typification and nomenclature when applicable. Allied taxa, not considered to belong to Usnea subgenus Neuropogon, are given in Appendix I, and excluded taxa in Appendix II. Only selected records are cited for the commoner and more widespread species according to chemical race. Full lists are only given in instances where new records or critical or new taxa are involved. Author abbreviations follow guidelines laid down by Laundon (1979). Results Morphology For convenience various morphological features are discussed under subheadings. Emphasis is placed on those features that were found to be of value for species delimitation within the subgenus. Species may be divided initially into two groups, namely those which produce apothecia and lack vegetative propagules, and those which produce soredia, pseudoisidia, or isidia and only rarely produce apothecia. Habit and mode of branching, surface ornamentation, and branch anatomy are important diagnostic characters, whilst others, including pigmentation and faveolation were found to be less reliable. The relative importance of these characters is discussed below. It is important to emphasise that within any given species specimens may exhibit considerable variation due to modification by ecological factors (Filson, 1982; Hawks- worth, 1973) and consequently certain features may then be atypical of the species. Scanning electron microscopy was used in certain instances to clarify interpretation of some features, for example vegetative propagules, and to show similarities to Usnea subgenus Usnea. Habit and mode of branching: Most species within the subgenus can be distinguished by their saxicolous habitat with erect, cartilagineous main branches arising from either a delimited or proliferating holdfast. In some species the form and blackening of the holdfast, together with the initial branching pattern, are characteristic, whilst in other species this is a much more variable feature. For example, Usnea ciliata arises monopodially from a proliferating, often blackened, holdfast; U. durietzii branches a short distance above a solitary holdfast to give a tree-like habit; whilst in U. acromelana and U. sphacelata the form is much more variable, ranging from a delimited holdfast to a spreading colonial form. Occasionally thalli are subpendulous or subdecumbent. This is a feature of Usnea subcapillaris and, to a lesser extent, U. pseudocapillaris . Many antarctic species may atypically become subdecumbent when growing in very exposed habitats. Similarly, thalli, for example in U. aurantiaco-atra, may rarely become detached and develop a scrambling habit resembling that of U. neuropogonoides . The extent of branching in any given species is often very variable although the overall form is often characteristic. Some species remain virtually monopodial or more or less subdichotomous towards the apices, for example Usnea ciliata and U. taylorii. Others, for example U. antarctica and U. aurantiaco-atra, may be richly branched from a delimited holdfast. An angular, divergent, branching pattern is characteristic of U. subcapillaris and U. pseudocapillaris which both branch repeatedly from a confined holdfast to form a loosely interwoven network of fine branches. These two species are further characterised by the very friable nature of the secondary branches, the brittleness is accentuated in herbarium material. 6 F. J. WALKER Pigmentation: The extent of black or violaceous black pigmentation of the thallus is very variable within the subgenus and frequently appears to reflect particular ecological parameters (see p. 25). Pigmentation may be extensive or confined to apices or papillae. Pigmentation of the thallus base near the holdfast was rarely of value, owing to the lack of correlation with other features, although it was occasionally used. Although pigmentation is one of the main features of the subgenus it is not unique within Usnea s. lat. and its relative importance is discussed elsewhere under 'Generic concept' (p. 42). Black pigmentation of the apothecial disc is characteristic of most species, although there are two exceptions, Usnea trachycarpa and the rarely fertile U. subantarctica , which have a brown to rufous brown disc. Pigmentation may also be a feature of vegetative propagules and is only found when soredia or other structures, such as isidia and pseudoisidia, are partially corticate. Some specimens of the norstictic-salazinic race of Usnea aurantiaco-atra assume a pinkish colouration of the medulla in the herbarium. This has no taxonomic value and is not compatible with the pigment found in some Usnea species, as for example in U. roseola, as reported by Swinscow&Krog(1979). Surface ornamentation: Surface features may either be uniform or rather variable within the individual species. A smooth, waxy, rarely subfaveolate, thallus with blackened annulations is a constant feature of all four species belonging to the Usnea ciliata complex (U. acromelana, U. ciliata, U. pseudocapillaris and U. subcapillaris} . Such annulations may sometimes occur in other species, for example, U. patagonica, or may atypically, be the result of weathering effects or necrosis in species where they are usually absent. A smooth, waxy surface is also characteris- tic of U. taylorii which, in addition, sometimes has scattered, slightly raised, pale maculae, formed by protrusion of the axis towards the surface through the cortex. In some species, for example U. perpusilla and U. sphacelata, the surface is more variable, and may either be smooth and waxy or become more or less scabrid to subpapillate with minute, frequently pigmented, papillae. On occasions this variation may be observed in different parts of the same specimen. In those species which have a very lax medulla, the primary branches may become notably inflated and attenuated at the point of attachment. This is a particular feature of Usnea durietzii and abnormal forms of U. sphacelata. In contrast to Usnea s. str. , pseudocyphellae, as described by Swinscow & Krog (1979), do not occur in the subgenus, although they may be recognised when associated with certain forms of asexual propagule formation (see below); the use of the term 'soralium' in the broad sense is preferred in this context. Gaps in the cortex left on the branch after the breaking away of fibrils or erosion of pseudoisidia may sometimes resemble pseudocyphellae. Papillae: These are recognised as small hemispherical or conical protuberances composed mainly of cortex (Swinscow & Krog, 1979). Prominent papillae are a feature of several species, in particular Usnea antarctica and U. aurantiaco-atra and, to a lesser extent, U. subantarctica and U. trachycarpa. In these species the thallus usually lacks the waxy lustre that is a characteristic feature of epapillate species. The relative size and presence of pigmentation of papillae may sometimes be used as an additional distinguishing feature. For example, papillae are small and normally pigmented in U. sphacelata and usually coarser and generally lack pigment in U. antarctica and U. aurantiaco-atra. U. aurantiaco-atra has a very varied range of morphology which varies from almost smooth or faveolate to papillate or verrucose-rugose; variation of this diversity may rarely be observed in a single thallus of this species. Fibrils: Fibrils are regarded (Swinscow & Krog, 1979) as laterally developed appendages containing an axis as well as a medulla. Numerous stout, elongate fibrils on the main branches, derived from papillae, are a feature of Usnea trachycarpa imparting a bottle brush-type of appearance; in some individuals the surface in this species may, less typically, be more or less papillate or faveolate. Short, capillaceous, spreading fibrils are often a feature of U. subantarc- tica, whilst in U. patagonica the fibrils are usually replaced by extended, thin, rarely branching, lateral branches that bear soralia. USNEA SUBGENUS NEUROPOGON I Internal structure: Relative widths of cortex, medulla, and axis are important taxonomic features and are frequently expressed as a ratio. However, a wide range of variation may be exhibited in individual species and, in some instances (Swinscow & Krog, 1979), may be of little taxonomic significance. It is essential to examine well-developed main branches since the medulla may not be fully expanded in finer, secondary branches. Within Neuropogon the basic delimitation of the cortex, medulla, and central axis is uniform in all species except Usnea taylorii, where a broad axis is deeply invaded by strands of medullary tissue containing algal cells, often resulting in the unique formation of several separate axial strands in this species. In some species the presence of a lax, arachnoid medulla accompanied by a thin axis, often occupying less than half the diameter in main branches, is diagnostic and occurs in, for example, Usnea acanthella, U. durietzii, U. perpusilla, and U. sphacelata. In the latter two species there is considerable variation, and the medulla may only be slightly lax and the axis consequently occupying a greater portion of the main branch diameter. By contrast, a compact medulla with a broad axis is a characteristic feature of U. antarctica and U. aurantiaco-atra. In other species, for example U. trachycarpa, although the medulla may sometimes be relatively broad, it is often sublax, and even species which normally have a compact medulla, for example in the U. ciliata complex, there may occasionally be some degree of laxness, especially towards the axis. Examples of various transverse sections are given in Fig. 1. SEM work has revealed that some species with a lax medulla, for example Usnea acanthella, U. perpusilla, and U. trachycarpa, have medullary hyphae ornamented with small nodular outgrowths. This was not found to be of taxonomic value and has previously been reported in Usnea s. str. (Lopez-Figueras & Palacios-Prii, 1981) and also within the Parmeliaceae in Alectorias. lat. (Brodo & Hawksworth, 1977). Apothecia: Within the subgenus six species: Usnea aurantiaco-atra, U. ciliata, U . perpusilla, U. subcapillaris, U. taylorii, and U. trachycarpa, produce abundant apothecia and lack vegetative propagules and, to date, four asexual species: U. acromelana, U. antarctica, U.pseudocapillaris, and U. subantarctica are occasionally fertile. Apothecia have not been observed in the remaining five species: U. acanthella, U. durietzii, U. neuropogonoides, U. patagonica, and U. sphacelata. The position as well as the form of the apothecium may be a useful diagnostic feature in some instances. For example, in Usnea subcapillaris apothecia are lateral, whilst in U. taylorii, U. trachycarpa, U. aurantiaco-atra, and U. ciliata they are almost invariably subterminal, some- times with a short, geniculate appendage; in U. perpusilla they are often produced laterally in series along a branch, each with a broad area of attachment (Fig. 2). Exceptions to this arrangement frequently occur in all these species. Excipular rays always occur in Usnea trachycarpa, U. ciliata, and U. subcapillaris but are only rarely present in U. perpusilla and U. aurantiaco-atra; they are absent in U. taylorii. The undersurface ornamentation of the excipulum may be diagnostic and usually reflects that of the subtending branch. However, in U. perpusilla and, occasionally, in U. ciliata, the surface may become individually faveolate. The dark disc colour is also a characteristic of the subgenus, although sometimes pigmentation may not be fully developed in abnormal or immature apothecia in those species which normally have a black disc resulting in a greenish grey colouration. U. trachycarpa and U. subantarctica are unique in the subgenus in having a rufous brown disc. The spores are similar to those in Usnea s. str. and are simple, ellipsoid, hyaline and usually fall within the range 7-10(-12) x 5-7 (-8) /tm. Variation in size has not been studied statistically, but there appears to be little variation between species and no taxonomic importance has consequently been attached to them here. The structure of the ascus apex is identical to that of the subgenus Usnea and corresponds to the 'Leowora-type' described by Honegger (1978). Vegetative propagules: Three types of vegetative propagules are found within the subgenus: soredia, pseudoisidia, and isidia. Soredia are defined as clusters of fungal hyphae and algal cells without cortex, whilst pseudoisidia, which originate in the same way, are outgrowths from F. J. WALKER Fig. 1 Examples of transverse sections of main branches. A & B - U. taylorii, C - U. ciliata, D - U. antarctica, E & F - U. aurantiaco-atra, G-U. sphacelata, H-U. patagonica, l-U. perpusilla. USNEA SUBGENUS NEUROPOGON Fig. 2 Disposition of apothecia. A - U. ciliata (Bartlett 25962, BM) x2, B - U. perpusilla (Lamb 6046, CANL) x4, C- U. aurantiaco-atra (Lamb 1085, CANL) x2. soredia or soralia-like areas that become partially secondarily corticate in contrast to true isidia, which are corticate from conception. A detailed study of propagules and their formation has been provided by Beltman (1978) using the SEM; similarities with her findings occur in Neuropogon species. However, in some instances the clear distinction between soredia, pseudoisidia, and true isidia may be rather difficult to resolve since intergradation and regeneration often occurs. Such intergradation may occur within a species or an individual thallus and is the result of breakdown of the primary propagule, for example pseudoisidia, with the subsequent formation of the second, for example soredia. However, the primary type of propagule formed is specific for the species and consequently used as the diagnostic character. Examination of a range of structures using the SEM has shown that soredia are not corticate, whilst pseudoisidia often have a thin, incomplete, secondarily developed outer cortex, and true isidia have a primary, structured cortex. Differ- ences between vegetative propagules in selected species are given in Fig. 3. (a) Soredia: Shape and formation of soralia in Usnea, together with the type of soredia produced, can be diagnostic, although considerable variation may occur within a given species. According to previous investigations (Krog et al., 1980; Swinscow & Krog, 1975, 1979) soralia formation in Usnea is either primary or secondary. Primary soralia in, for example, U. glabratula, develop directly from the cortex by local breakdown and are usually plane or concave initially, sometimes becoming protuberant on development. Secondary soralia arise from pseudocyphellae subsequent to the breakdown of isidia, for example in U. subfloridana, and are often protuberant. However, in this species it is apparent that such soralia may become 10 F. J. WALKER B Fig. 3 Vegetative propagules. A isidia- U. acanthdla (holotype, BM), B soredia- U. antarctica (F.I.D.S. B1119, BM), C pseudoisidia - U. patagonica (Henssen & Vobis 244991, MB), D pseudoisidia - U. durietzii (Gibby & Barrett, July 1979, BM). Based on SEM studies; blackened areas represent cortex. Scales A = 1 mm, B-D = 50 /xm. corticate, thus producing pseudoisidia. This is in contrast to true isidia that are produced directly from the cortex. According to the above definition, soralia initiation in the majority of asexual species of Neuropogon may be classified as primary, although this distinction is often difficult to interpret in species that produce pseudoisidia. In most species such primary soralia arise directly from the thallus, for example in Usnea acromelana, U. pseudocapillaris, U. sphacelata, and U. subantarc- tica, and are initially concave, though they may become convex to globose. Soredia may become secondarily corticate, often very irregularly or thinly so, thus forming minute, more or less spherical pseudoisidia. This is a feature of species which appear to have pigmented soredia; for example in certain forms of U. acromelana and U. sphacelata, where the pigment corresponds to overlying fragments of cortex. Such structures can be distinguished by size from the cylindrical, more or less elongate pseudoisidia that are a feature of U. durietzii and U. patagonica. Soralia initiation in Usnea antarctica is unusual and may be classified as secondary in a different sense since they are produced on papillae and, unlike species in which soralia develop USNEA SUBGENUS NEUROPOGON 11 from the cortex, are not confined to apices or secondary branches. These soralia often have a distinct crateriform margin and rarely also produce small, dark, pigmented pseudoisidia. Soredia development in Usnea patagonica and U. durietzii may also be regarded as secondary since they are produced from the breakdown of pseudoisidia. This process appears to be cyclic since both structures may occur in an individual 'soralium'. In U. patagonica such soralia often arise from small papillae but do not have a distinct margin characteristic of U. antarctica. (b) Pseudoisidia: Pseudoisidia were defined by Dahl & Krog (1972) as isidia-like structures lacking a true original cortex, such as occur in Evernia prunastri. Beltman (1978) found the delimitation impractical in that species, although she observed intermediate structures amongst soralia and lobules that presumably corresponded to pseudoisidia. Other terms have been used to describe similar structures: 'soredial isidia' was used by Du Reitz (1924) and Maas Geester- anus (1947) to describe isidia-like structures formed in soralia to distinguish them from 'isidial soralia' or 'sorediose isidia' (Beltman, 1978) which are produced by breakdown of apices of isidia, as in Parmelia subaurifera. Similar small, corticate structures resembling isidia, particu- larly in Alectoria, have been referred to as 'soredialasten' (Henssen & Jahns, 1973); 'isidioid spinules' (Brodo & Hawksworth, 1977) and 'isidial soralia' (Jahns, 1980; Krog et al., 1980). All these terms appear to correspond to a structure that is characteristic of two species of Neuropogon, namely Usnea durietzii and U. patagonica. The term pseudoisidia is here preferred to describe the small, partially secondarily corticate, pigmented structures that are produced in soralia-like clusters and have the same origin as soralia. The cortex is often ill-defined unlike that found in true isidia. Pseudoisidia are either produced in delimited, soralia-like structures, or may be of secondary origin, formed by regeneration after the breakdown of true isidia as, for example, in U. torulosa and U. amblyodada (see Appendix I) where they may also erode to produce soralia. (c) Isidia: True isidia only occur in one species, Usnea acanthella, where they may be up to 1 mm in length. They arise as small clusters from tubercules on the surface of the thallus. Such isidia lack a central axis but have a true primary cortex, often with a minute fracture at the constricted base assumed to assist in dispersal, and fracture leaving a scar (Beltman, 1978; Du Rietz, 1924; Maas Geesteranus, 1947). Sometimes fibrillae are also produced in this and other species and may be distinguished by the presence of a central axis; they correspond to structures occasionally observed in Protousnea dusenii (Krog, 1976). Pycnidia: Pycnidia are immersed in the cortex and form irregular, hemispherical swellings, ranging from 100-200 /am in diameter, towards the apices of ultimate branches in pigmented or unpigmented areas. Individual loculi are separated by thallus tissue although often superficially appearing to be compound with several ostioles. Lindsay (1859) provided a detailed account of the pycnidia of Usnea taylorii and U. aurantiaco-atra (as Neuropogon melaxanthus) . Lamb (19390) also observed pycnidia in U. aurantiaco-atra, and described the conidia as 'staff-shaped, sometimes with a slight eccentric swelling', and subsequently (Lamb, 19480) described those of U. perpusilla (as U. rohmederi) as 'broadly fusiform'. Pycnidia are rare and difficult to observe, but have been examined in the following fertile species: Usnea aurantiaco-atra, U. perpusilla, U. taylorii, and U. trachycarpa. Examination of the type of U. trachycarpa f. elatior revealed the pycnidial wall to be hyaline whilst in U. aurantiaco-atra this was found to be pigmented throughout. More specimens should be examined before any conclusions may be reached, although this might indicate an additional taxonomic difference between species with rufous brown and black apothecial discs. Conidia were found to be of a similar size in all species, in the range 9-ll(-14) x l-l-7(-2) //in. Their shape conforms to sublageniform as described by Krog (1982) or are more or less narrowly fusiform but are slightly swollen at the proximal end. However, insufficient specimens have been examined to indicate the full range of infraspecific variation. There is apparently no conidial difference between the two races of Usnea aurantiaco-atra, at least from examination of the lectotypes of Neuropogon antennarius and Lichen aurantiaco-ater , both from subantarctic South America. Krog & Swinscow (1981) described conidial formation in the Parmeliaceae as endobasidial on 12 F. J. WALKER conidiophores of the bayonet type. The terms 'endobasidial' and 'exobasidial' of Steiner (1901) have often been misinterpreted or confused by previous authors (Lamb, 19390, 19480; Rogers, 1981) when applied to the subgenus Neuropogon. These terms have since been rejected by mycologists (Henssen & Jahns, 1973; Vobis & Hawksworth, 1981) and replaced by 'terminal' and 'lateral' as defined by Vobis (1980), each comprising a range of types of conidiophores. The conidiophores appear to correspond to type VI of Vobis & Hawksworth (1981) in which conidiogenous cells arise in branched chains with the conidia arising laterally. However, intermixed with these, some more or less conidiogenous cells are seen to arise directly from the wall tissue (type II) ; such cells are occasionally seen with one or two percurrent proliferations. In addition, a few conidiophores approximating to type V have been observed in which conidia are produced terminally. Chemistry Usnic acid is present in varying concentrations in the cortex of all species. A limited range of jS-orcinol depsidones (connorstictic, fumarprotocetraric, norstictic, protocetraric, psoromic, 2'-O-demethylpsoromic, and salazinic acids), rarely /3-orcinol depsides (squamatic and hypothamnolic acids), or fatty acids (murolic acid complex) occur in the subgenus. As in Usnea subgenus Usnea, frequently a particular species may exhibit more than one chemical race with either different, but often biosynthetically related, substances present. Many species also have an acid-deficient phase which may be dominant, for example, in U. taylorii and U. sphacelata; one species, U. acanthella, has no demonstrable chemistry. The substances, with the species in which they occur, are given in Table 1. Only those of a diagnostic value are given, whilst substances which may occur in conjunction with one of the main compounds, such as connorstictic acid, cph-1 (yellow accessory with fumarprotocetraric acid, Culberson et al., 1981) and unknown accessory substances, are omitted from the table. In some species the overall distribution and abundance of certain chemical races may vary but in areas where there is an overlap of two races thalli are encountered which have a composite chemical complement. This phenomenon is known to occur, for example, in Usnea aurantiaco- atra and U. subcapillaris and often produces evidence against accepting species based solely on chemical differences. Such mixed strains are omitted from Table 1 and are discussed further under 'Circumscription of the species' (p. 39) and U. aurantiaco-atra (p. 71). 1 . {$-orcinol depsides: Squamatic and hypothamnolic acids are the only two j8-orcinol depsides known within the subgenus Neuropogon, occurring with, or rarely replacing, /3-orcinol depsi- dones in a single species, Usnea subcapillaris. This is in contrast to the subgenus Usnea where a wide range of depsides is found. Consequently, in specific instances, the detection of jS-orcinol depsides can be a useful factor in separating certain species of the subgenus from closely allied taxa. For example, in the Usnea ciliata complex, squamatic acid is only known in a rare race of U. subcapillaris whilst it is frequently found in U. torulosa; a species belonging to the subgenus Usnea which is superficially similar to some ecotypes of U. acromelana. Conversely barbatic acid is sometimes found in U. torulosa but has not yet been found in any Neuropogon species. Further, in South American collections, the presence of the orcinol depside divaricatic acid in species of Protousnea is an additional character useful for separating members of that genus from Usnea neuropogonoides or decumbent forms of U. aurantiaco-atra. 2. ^-orcinol depsidones: The range of /3-orcinol depsidones found in the subgenus Neuropogon and related taxa were identified by TLC using solvent systems HEF and TDA following the method of Walker & James (1980). Solvent system G (toluene/ethylacetate/formic acid) of Culberson et al. (1981) was used to confirm the presence of salazinic acid and connorstictic acid in norstictic-salazinic acid chemotypes, using Parmelia sulcata and P. perforata as controls. The presence of connorstictic acid was only demonstrated when the concentration of norstictic acid was high. Stictic acid has not been found in the subgenus. An additional unknown yellow spot, Rf class TDA 1, HEF 2-3, accessory to norstictic acid, was sometimes found in Usnea trachycarpa. USNEA SUBGENUS NEUROPOGON 13 Table 1 Chemical properties of Usnea subgenus Neuropogon. nor sal pc fpc pso sqm hth fat frequency U. acanthella U. acromelana 1 U. acromelana 2 U. acromelana 3 U. antarctica 1 U. antarctica 2 U. antarctica 3 U. aurantiaco-atra 1 U. aurantiaco-atra 2 U. aurantiaco-atra 3 U. ciliata U. durietzii 1 U. durietzii 2 U. neuropogonoides 1 U. neuropogonoides 2 U. patagonica U. perpusilla 1 U. perpusilla 2 U. pseudocapillaris x x X X X X X X X X X X X c R L C R C C L C C R C C R C U. sphacelata 1 .... x ... R U. sphacelata 2 C U. subantarctica 1 x C U. subantarctica 2 C U. subcapillaris 1 x x ..... C U. subcapillaris 2 x R U. subcapillaris 3 .... x ... R U. taylorii 1 ... .... R U. taylorii 2 C U. trachycarpa 1 x x . . . . + C U. trachycarpa 2 x . . + R U. trachycarpa 3 + C Only substances of diagnostic importance are included, nor = norstictic acid, sal = salazinic acid, pc = protocetraric acid, fpc = fumarprotocetraric acid, pso = psoromic acid (including 2'-O-demethylpsoromic acid), sqm = squamatic acid, hth = hypothamnolic acid, fat = fatty acids. Symbols: x = constant; usually present as an accessory substance or occurring in low concentrations; 4- = present in most specimens. C = Common, widespread throughout range. R = Rare throughout range or from a single locality. L = Locally abundant, restricted distribution within range. Solvent system G was also particularly useful for identification of substances belonging to the fumarprotocetraric acid complex in Usnea aurantiaco-atra and U. antarctica. The substance cph-1 (Culberson et al., 1981) was frequently found as an accessory substance to fumarpro- tocetraric acid, producing a yellow spot on developed chromatograms. Its presence was confirmed using Cetraria straminea as a control and further by two-dimensional chromatography employing first solvent system G and then HEF, with suitable controls and following fig. 2 of Culberson et al. (1981). This method was also used to confirm the occasional presence of traces of salazinic acid and distinguish it from cph-2 (Culberson et al., 1981), in specimens containing 14 F. J. WALKER the fumarprotocetraric acid complex. For example, traces of salazinic acid were confirmed in a collection of Usnea antarctica from South Georgia (Lindsay 4327, A AS) and in U. aurantiaco- atra from Tierra del Fuego (Henssen & Vobis 24417a, MB), both containing fumarprotocetraric acid as the primary constituent. In the subgenus Usnea Swinscow & Krog (1979) found that although protocetraric acid nearly always excluded the production of salazinic acid it neverthe- less occurred in some thalli that contained both substances. In Neuropogon protocetraric and salazinic acid may occur together when norstictic acid is present, as for example in the Usnea ciliata complex. The same method of two-dimensional chromatography was used to check the identity of depsidones in a chemically mixed specimen of the two depsidone-containing races of Usnea aurantiaco-atra from the Falkland Islands (R. I. L. Smith 2572, AAS). It was also used to establish the identity of salazinic acid and protocetraric acid in Race 2 of the same species from Isla de Los Estudos (Staten Island) which lack norstictic acid. With two-dimensional chromato- grams it was sometimes found useful to run an additional control plate containing a mixed extract of the test specimen with known substances as an alternative, or in addition to, running one-way controls in each solvent on the same plate. The distribution of psoromic acid cannot reliably be used as a diagnostic feature since it occurs randomly throughout the subgenus as it tends to do in quite a few taxa of Usnea subgenus Usnea. Its presence is apparently of a spasmodic nature and it is frequently only known from a handful of collections in a single species with a different chemistry. It is also of interest to note that in two instances, in Usnea perpusilla and U. sphacelata, psoromic acid occurs in species which otherwise characteristically lack diagnostic medullary substances. A previous report of psor- omic acid in antarctic material of, for example, U. antarctica (Golubkova & Schapiro, 1970) is incorrect and refers to an unknown substance (see p. 15). 3. Fatty acids: A series of fatty acids which may be of diagnostic importance are found primarily in Usnea trachycarpa. These are related to lichesterinic acid and are here referred to as the murolic acid complex. Their chemical structure has already been elucidated by previous authors (Bodo & Molho, 1980; Ghogomu & Bodo, 1982) who found that two of the acids isolated corresponded to murolic and muronic acids, previously known from Lecanora muralis (Huneck et al., 1979). Ghogomu & Bodo (1982), using material from lies Kerguelen, identified the remaining four acids as 13-acetoxyprotolichesterinic acid, 13-acetoxylichesterinic acid, isomur- onic acid, and neuropogolic acid. The six acids were found to have Rf values ranging from 0-30 to 0-46 in TA. The murolic acid complex has here been demonstrated in all three chemical races of Usnea trachycarpa throughout its range. These were found to correspond to Rf classes 3 to 5 in TA, using norstictic, psoromic, and stictic acids as markers. Improved separation was obtained when plates were run twice over a distance of 11 cm. Additional acetone extracts were also run in solvent system G (Culberson et al., 1981), designed to separate depsidones with low Rf values. This system also gave improved separation of spots lying approximately between norstictic and stictic acids. Traces of additional substances were sometimes found which made it difficult to identify individual fatty acids. This may indicate that the complex also contains traces of other unknown fatty acids. A series of fatty acids, apparently belonging to the same complex, has also been found in Usnea patagonica. Frequently only two to four fatty acids were detected with the remaining acids absent or only present in trace amounts. The number of fatty acids detected in a single specimen on the TLC plate from this species varied according to the solvent system used or the number of times the plate was run. This may be due to variation in the concentration of the extract spotted on each plate, or, since U. trachycarpa uniformly resolved into six spots, may indicate that some different acids are involved which do not separate so readily in the solvent systems used. At least one fatty acid found in U. trachycarpa belonging to Rf class 5 in TA did not occur in U. patagonica. At least two fatty acids belonging to Rf classes 2-3 in TA occur in Usnea neuropogonoides and USNEA SUBGENUS NEUROPOGON 15 are present in both chemical races. These appear to be identical to those of lower Rf classes found in U. patagonica. Very rarely traces of similar, unidentified fatty acids occur in other species in the subgenus, for example Usnea acanthella, U. antarctica*, and U. subantarctica, but their presence is not constant enough to warrant taxonomic significance. 4. Accessory substances: Throughout the subgenus three undetermined substances were found which produced characteristic coloured fluorescence under long-wave UV light (see Table 2) on chromatograms before charring. On developed plates the spots are either colourless or have slight pinkish or yellow-grey colouration. These substances are here referred to as 'UV + unknowns' and appear at random in most species in the subgenus, with or without medullary substances, and are consequently of no taxonomic value. They are probably unstable or locally concentrated as it was often not possible to repeat the original results from the same thallus. In most species UV + unknowns are only present in trace amounts but in Usnea perpusilla they often occur in high concentrations. \ Table 2 TLC properties of UV 4- unknown substances in Usnea subgenus Neuropogon. Rf classes UV fluorescence colour of spot Substance TDA HEF TA before charring after charring () A 3 5 3 pale violet quench or purple pale pink B 2 5 2 pale yellow greenish grey pale yellow-grey C 2 3 1 pale violet quench or purple pale pink D 1-2 3 1 white quench or purple colourless The TLC properties of the UV + unknowns A, B, and C are given in Table 2 along with an additional unknown, D, which fluoresces white under long wave UV light and sometimes occurs in trace amounts with unknown C. Substances C and D may, in trace amounts, be mistaken for fumarprotocetraric or protocetraric acids which have similar Rf classes; however, the character- istic fluorescence of the unknown substances should be a distinguishing feature. Rf classes obtained for unknowns A, B, and C agree with those found by J0rgensen (pers. comm.), working on material from Bouvet0y, in Usnea antarctica and U. aurantiaco-atra. The occurrence of these unknown substances in antarctic Neuropogon specimens has previously been published by Golubkova & Schapiro (1970). One substance was incorrectly identified by them as psoromic acid and probably corresponds to unknown A. These substances have occasionally been detected in Usnea s. str. Distribution Figs 4-8 The subgenus Neuropogon has a bipolar distribution and largely replaces the subgenus Usnea in arctic, antarctic, and subantarctic regions. Only one species, Usnea sphacelata, is bipolar, with a circumpolar distribution in the northern boreal regions as well as in Antarctica, and with outliers in the subantarctic and North, Central, and South America. One species, U. acanthella, is apparently confined to the northern Andes, whilst the remaining species have antarctic or subantarctic-alpine distributions. Further details are to be found under each species. The distributions discussed here are arranged according to various geographical areas as a back- ground to biogeographical considerations and to substantiate, or comment upon, some records to be found in the literature. Areas where difficulties may arise in identification of species, due to variation or overlap of distributions, are also indicated. 1 . Northern Andean chain At present three species, Usnea acanthella, U. durietzii, and U. sphacelata, are known to occur in * Identified as murolic and neodihydromurolic acids by Huneck et al. in J. Hattori hot. lab. 56: 461-480 (1984). 16 F. J. WALKER Fig. 4 Antarctic and subantarctic distribution patterns. USNEA SUBGENUS NEUROPOGON 17 0-- ..*. m oo o u. sphacelata ^^ U. subantarctica . . U. trachycarpa Fig. 5 Antarctic and subantarctic distribution patterns. 18 F. J. WALKER 0-- O U. acromelana U. ciliata U. pseudocapillaris U. subcapillaris m oo o Fig. 6 Antarctic and subantarctic distribution patterns. USNEA SUBGENUS NEUROPOGON 19 Fig. 7 South American distribution patterns. 20 F. J. WALKER U. durietzii U. neuropogonoid.es U. patagonica Fig. 8 South American distribution patterns. USNEA SUBGENUS NEUROPOGON 21 this area. The possible occurrence of a fourth species, U. acromelana, remains uncertain or doubtful, whilst U. patagonica reaches its northern limit in Bolivia. The few collections examined have been referred, where possible, to one of these species, sometimes as atypical forms. Frequently specimens are depauperate, moribund, eroded, or extensively pigmented, and in such instances it is not possible to identify material conclusively. 2. Subantarctic South America This area includes Patagonia, Tierra del Fuego, and the Falkland Islands. The greatest concentration of species occurs in this region with three fertile, one sterile, and six asexually reproducing species known. All three fertile species Usnea aurantiaco-atra, U. perpusilla, and U. trachycarpa are known as far north as about 40S. Two other fertile species, U. ciliata and U. taylorii, have been incorrectly reported from this area, for example by Lamb (1948a) and Rasanen (1932), and misidentifications refer to either atypical, smooth forms of U. aurantiaco- atra, or weathered forms of U. perpusilla. The sterile U. neuropogonoides is only known from a small area on the Chilean-Argentinian border. U. aurantiaco-atra and U. trachycarpa also occur on the Falkland Islands. The six asexual species found in this area are Usnea acromelana, U. antarctica, U. durietzii, U. patagonica, U. sphacelata, and U. sub antarctica. Some of these species are reaching the limits of their distribution and consequently atypical morphotypes may occur which may lead to misidentification. For example, U. antarctica and U. durietzii here reach their northern and southern limits respectively. Atypical forms of these two species might be mistaken for U. patagonica but fortunately medullary chemistry can be used as an additional guideline since, unlike U. patagonica, the two species are usually represented by the depsidone-containing race. Morphological differences between all three species are summarized in Table 3. U. antarctica is the only asexual species known from the Falkland Islands and is represented there by a single specimen. Table 3 Differences between U. durietzii, U. patagonica, and U. antarctica. U. durietzii U. patagonica U. antarctica Habit/branching irregular to tufted regular with regular with with short laterals extended laterals extended laterals Pigmentation apices, base, apices, base, apices, soralia, pseudoisidia pseudoisidia, cortex cortex Medulla very lax lax or sublax compact Papillae rare common common Soralia often becoming remaining delimited with confluent delimited, ulcerose crateriform margin Pseudoisidia large, c. 100 /mi Small, c. 50/rni absent Chemistry norstictic/salazinic deficient/fatty fumarprotocetraric acids acids acid 3. Australasia (a) Australia: Two species Usnea acromelana and U. subcapillaris or possibly three (U. ciliata) occur in Australia. The subgenus is uncommon in Tasmania and is extremely rare at very high altitudes in the mountains of Victoria (see below). The majority of specimens examined so far from the mainland are referable to either Usnea torulosa or U. inermis of the subgenus Usnea. Rogers (1981) reported five species (unlisted) from Australia based on catalogues produced by Wetmore (1963) and Weber & Wetmore (1972). These were U. acromelana, U. antarctica, U. ciliata, U. aurantiaco-atra (as U. melaxantha), and U. sphacelata. The epithet 'melaxanthus' has frequently been misapplied to sorediate material resulting from Nylander's (1855) misconcep- 22 F. J. WALKER tion of the species. Examples include Darbishire (1912), Crombie (18796), and Wilson (1888, 1890, 1893) and again specimens are referable to Usnea subgenus Usnea or rarely, when in Tasmania, to U. acromelana. The record of U. sphacelata from Tasmania (Lawrence, 1834) may also refer to U. acromelana. U. antarctica does not occur in Australia (cf. Filson, 1982). (i) Victoria: Weber & Wetmore (1972) reiterated reports by Wilson (1888, 1890) based on material from Mt. Macedon (3727'S: 14434'E) and Mt. Hotham (3658'S: 14711'E). Unfortu- nately Wilson's main herbarium was lost in transit (Filson, 1976). Some duplicate material of Mt. Macedon collections has been traced (NSW!) and specimens belong to the barbatic acid race of U. torulosa. To date the only collections of Usnea acromelana examined are from the Bogong High Plains (3645'S: 14721'E) (MEL 18755!, MEL 1018193!). (ii) Tasmania: Usnea acromelana and U. subcapillaris are known from several alpine localities in Tasmania and were previously reported by Lamb (19390) and Bratt & Cashin (1976) respec- tively. Some specimens of U. acromelana examined have close affinities with U. pseudocapillaris but are somewhat coarser and are retained in the former species for the present, since a wider range of variation is found than amongst New Zealand populations of U. pseudocapillaris. U. ciliata may also occur in Tasmania, but unfortunately, I have not seen any definitive, fertile, material. The few examples seen are damaged and lack apothecia and soralia (for example MEL 1029344! and CHR 342744!), and could be either U. acromelana or U. ciliata. Dodge (1948) cited a specimen, possibly of U. ciliata, in herb. Stirton (GLAM) from Mt. Wellington, but this has not been traced. U. antarctica was reported from Tasmania by Du Rietz (1929). This was a misidentification of U. acromelana, a species which he does not include in his taxonomic treatment (Du Rietz, 1926) since he referred all northern hemisphere material to U. sulphurea and all southern hemisphere to U. antarctica (Du Rietz, 1929). Wilson's (1893) report of U. melaxantha from Table Mountain was based on Robert Brown's collections which mainly consist of U. acromelana. (b) New Zealand and subantarctic islands Five species have recently been recorded from the South Island of New Zealand (Galloway, 1985): Usnea acromelana, U. antarctica, U. ciliata, U. sphacelata, and U. subcapillaris; a further species, U. pseudocapillaris, is described here. Previous reports of U. aurantiaco-atra and U. trachycarpa (Mark & Bliss, 1970; Martin & Child, 1972) are erroneous and probably refer to U. ciliata. The subgenus is commonly represented in alpine areas east of the Main Divide, particularly in central Otago but is rare elsewhere, with U. antarctica and U. sphacelata confined to a few exposed, predominately high altitude localities. U. acromelana and U. ciliata are both known from Stewart Island although the subgenus is rare there and is confined to Mt. Anglem (CHR 343356! and CHR 342749! respectively). The subgenus is very rare in the North Island and, apart from a few recent collections, records are often based on historical data. U. acromelana, U. ciliata, and U. subcapillaris are all known to occur. One collection from Mt. Ruahine (Colenso 1164, WELT!), previously determined as U. ciliata, is Alectoria nigricans, although another specimen (Colenso C1776, BM!, WELT!) from the same locality 'prope Napier' (Miiller, 1896) is U. acromelana. Both U. acromelana and U. ciliata are known from Tongariro National Park and the subgenus has recently been refound by Bartlett as far north as Gisbourne, from Mt. Hikurangi in the Raukumara Range (Bartlett 25965, BM!) and from the Kaweka Range in Hawkes Bay (Bartlett 25961, BM! 25962, BM!), thus verifying the record in Martin & Child (1972). Only one collection is known from Chatham Island, that collected by Travers (BM!), which is an unusual, decumbent form of Usnea acromelana, growing at a much lower altitude than elsewhere in New Zealand. The subgenus does not occur on the Auckland Islands or on Campbell Island, where it is probably replaced by U. xanthopoga. U. antarctica is the only species examined from Macquarie Island where it is confined to a few localities at the north of the island. Sometimes specimens are somewhat subdecumbent and this has led to misidentifica- tions as that of U. laxissima (= U. sphacelata) reported by Filson (1981). USNEA SUBGENUS NEUROPOGON 23 4. Subantarctic islands of the Indian Ocean Three species are known, two fertile and one asexual. Of the two fertile species, Usnea taylorii and U. trachycarpa, the former is only known from lies Kerguelen and Crozet. U. antarctica is known from both island groups as well as from Marion Island in the Prince Edward Island group, thus completing the subantarctic circumpolar distribution of this species eastwards from Bouvet0y . Many sorediate taxa have been described from these islands and all are here regarded as synonymous with U. antarctica; the species is represented by a range of rather robust, depsidone-containing variants in this area. Specimens of U. aurantiaco-atra, said to have been collected from lies Kerguelen by Hooker (M!) are unique, and presumably originated from Cabo de Hornos (Patagonia) having been subsequently mislabelled. 5. Antarctica For the purposes of this study Antarctica is here divided into two areas based on geological, climatic, and ecological considerations. The Antarctic peninsula and the antarctic continent as far west as the Ross Ice Shelf are included with the subantarctic islands of the South Orkneys, South Shetlands, South Sandwich islands, and South Georgia and Bouvet0y, and is considered separately from eastern Antarctica, which lies from Dronning Maud Land eastwards to Victoria Land. These approximately correspond to maritime and continental antarctic zones as defined by ecologists, for example Lindsay (1977 'a), although the maritime antarctic usually only includes the west coast of the Antarctic peninsula as far south as Marguerite Bay (Holdgate, 1970). (See 'Phytosociology', p. 28.) (i) Antarctic peninsula and islands: The three most common asexual species are Usnea antarctica, U. sphacelata, and U. subantarctica. U. antarctica replaces U. sphacelata and U. subantarctica in the island groups and is often the only species of the subgenus to be found in some localities, for example, the South Sandwich Islands. U. acromelana is also found in this area, but is rare or overlooked, and is confined to the northern part of the peninsula and some of the islands. Occasionally all four species may be found in a given locality and may exhibit a bewildering range of variation, besides, with the exception of U. sphacelata, sometimes producing apothecia. Usnea aurantiaco-atra is found on the west side of the Antarctic peninsula, the South Orkneys, the South Shetlands, and on South Georgia. U. antarctica and U. aurantiaco-atra both occur on Bouvet0y but are rare (Holdgate et al, 1968). The occurrence of U. trachycarpa is doubtful, based on a single sterile specimen (R. I. L. Smith 3453, AAS!) from the Antarctic peninsula; if supported by further fertile records this would represent a significant extension of its subantarc- tic distribution. (ii) Eastern Antarctica: Only two species, Usnea antarctica and U. sphacelata, have been identified from material from eastern Antarctica. Both species are circumpolar, although U. sphacelata is the commoner, usually occupying more exposed habitats (pp. 26-27). Where both species do rarely occur together one of them is often depauperate, abnormal, or even somewhat intermediate, and particular thalli may be difficult to identify. Unusual, weathered forms of U. sphacelata with a less lax medulla have led some authors (Dalenius & Wilson, 1958; Filson, 1974, 1975; Golubkova & Schapiro, 1970) to confuse this species with U. acromelana, a species that is confined to subantarctic regions and only extends its range as far south as the tip of the Antarctic peninsula. The occurrence of unusual forms has also resulted in the descrip- tion of a spectrum of microspecies which are here reduced to synonymy. Further specimens need to be examined before antarctic distributions can be assessed in detail. A report of U. aurantiaco-atra from Dronning Maud Land (Dalenius & Wilson, 1958) is a misidentification of U. sphacelata (UPS!). Further problems encountered in this area are discussed under 'Bio- geography' (p. 33). 24 F. J. WALKER 6. South Africa Two collections comprising saxicolous Usnea species collected at 1830 m from Cedarberg (Schelpe 1961 BOL!, 1966 p.p. BOL!) have been examined and have proved to be very interesting. Schelpe 1966 is a mixed gathering of an Alectoria species with an unknown, possibly undescribed, Usnea species. Thalli have scattered, pigmented, true isidia that are reminiscent of the type found in Usnea inermis and, like that species, contain psoromic acid. Part of the second collection (Schelpe 1961) was examined by Lamb (19480) and identified as possibly Neuropogon acromelanus (CANL 16944!) and Neuropogon sp. (CANL 17284!). Material identified by Lamb as Neuropogon acromelanus and part of the original collection (Schelpe 1961!) very closely conforms to Usnea patagonica and is tentatively referred to that species; for differences between the two see p. 84. Other thalli in this collection and the specimen referred to Neuropogon sp. by Lamb do not resemble Usnea patagonica quite so closely and remain unidentified. Further material is required, along with detailed study of South African montane species of Usnea, including types, before the taxonomic position of the Cedarberg specimens can be finally resolved. Taxa related to Usnea pulvinata aggregate, for example U. capensis, may also have to be considered. Since some specimens are tentatively included in Usnea patagonica it is remotely possible that the subgenus Neuropogon s. lat. is extremely rare at high altitudes in South Africa. Another species with a similar pattern of distribution is Pseudocyphellaria gilva which was described from South Africa and is also known from South America. According to Schelpe (pers. comm.) the Cedarberg locality is one of the coldest in the south-west Cape Mountains and is the richest locality for Umbilicariaceae in southern Africa, and consequently is the most likely refugia for any Neuropogon species. Ecology There has only been a brief opportunity for a first-hand study of the ecological requirements of Neuropogon species during the preparation of this account. Consequently much of the informa- tion presented here has been drawn from available literature and discussion with lichenologists who have made detailed field observations. As a result many gaps still exist, in particular for South America, and for species which have a restricted distribution or are only known from a few collections. One of the earliest ecological observations was made by Dumont d'Urville (1826) in his Flore des Malouines who noted that (translation) 'Usnea melaxantha (U. aurantiaco-atra) grew by preference on bare rocks exposed to the south-west winds, forming an unusual type of sward on smooth rock faces. These rocks were always arranged in fairly regular strata inclined at an angle of 40 to 50 degrees and running from east to west'. This observation reflects the general ecological requirements of the subgenus. These are a more or less acidic, saxicolous substrate and an exposed, predominately arctic-alpine environ- ment which is unfavourable to, and usually excludes, the subgenus Usnea. Species are able to flourish in harsh microclimates where temperature variation, radiation, and drought from freezing or high winds, are all major controlling factors. Such conditions are often termed ecological cold-deserts (Lindsay, 1917 a), since water availability is often the most important factor, and are not necessarily confined to polar regions. Consequently a remarkable altitudinal amplitude is exhibited by the subgenus, ranging from sea-level upwards in polar regions, to c. 5000 m in the Andes of tropical America. In areas where the climate is warmer and the rainfall higher, the subgenus is rapidly replaced by the subgenus Usnea. Here Neuropogon is confined to exposed, drier, upland areas, as for example, on Macquarie Island and the New Zealand shelf islands. Neuropogon species are rarely able to compete with the subgenus Usnea; an exception is Usnea acromelana which occupies similar habitats to U. torulosa. Species of Neuropogon are adapted ecologically, morphologically and physiologically (Ahmadjian, 1970) to these cold-desert conditions which impose stresses that are rarely encountered elsewhere (Lindsay, 19770). The presence of a thick cortex is probably an USNEA SUBGENUS NEUROPOGON 25 adaptation against water loss and, conversely, water uptake (Ahmadjian, 1970), although the latter can still occur via the soralia. In addition much mechanical strength, as found in the tough axial strand, is required to withstand wind-blast. It has often been observed that pigmented lichens are more frequent in Antarctica, and Lindsay (I977a) noted that thalli of Usnea sphacelata are more heavily pigmented in continental than in maritime Antarctica. This pigmentation may provide protection for the underlying algal cells against UV radiation (Ahmadjian, 1970), but more importantly absorbs heat which may assist in melting snow (Lindsay, \971a) and increase carbon assimilation (Kershaw, 1983). Neuropogon species have to withstand a great range of temperatures (Lindsay, 19770) and their metabolism shows a remarkably effective physiological adaptation to low temperatures, particularly when dry, and hence may be described as facultatively or obligatively psychrophilic (Lamb, 1970). Field and laboratory measurements of respiration and photosynthetic rates (Ahmadjian, 1970) have shown that thalli of Usnea sphacelata are able to maintain a positive metabolic balance of temperatures as low as 18-5C, whilst a negative balance occurs above +20C. Consequently an extended period of high temperatures would be damaging. Distribution of the subgenus is controlled by climatic, edaphic, and biotic factors (i.e. bird rookeries) (Lamb, 1970). Consequently there are few habitats in the polar regions where species of Neuropogon do not occur. Despite being able to withstand severe physiological drought, it is evident that some moisture is essential, either as mist or melt-water, although this requirement varies considerably according to the species. For example, in New Zealand species of the Usnea ciliata complex are absent from areas where there is prolonged snow cover, including the bases of exposed tors on mountain plateaux and in fellfields (Mark & Bliss, 1970). Such intolerance towards prolonged snow cover is similar to that exhibited by Parmelia olivacea in Scandinavia (Ahti, 1966). This is in contrast to some other macrolichens, for example Umbilicaria cylindrica (Mark & Bliss, 1970) and Cladonia bellidiflora (Lynge, 1937), which appear to require prolonged winter snow protection against wind-blast. Similarly in New Zealand, the Usnea ciliata complex is rarely found in the nival zone above the permanent snow-line, or in areas of permanent glaciation. In contrast U. sphacelata, a species that has its main distribution in the polar regions, is best adapted to withstand much harsher climatic conditions and, for this reason, mainly occurs in New Zealand (and Patagonia), in very exposed, snow-free, alpine areas and nunataks at higher altitudes. The distribution of Neuropogon species on the Antarctic continent, excluding the Antarctic peninsula, appears to be very local, thus reflecting the availability of suitable habitats which are limited to ice-free coastal areas and exposed nunataks (Ahmadjian, 1970). For example, the subgenus was not found on Mawson Rock, Mac. Robertson Land (Seppelt & Ashton, 1978) although species olAlectoria, Umbilicaria, and Xanthoria occurred, but has been recorded from many other localities to the west (Filson, 1966, 1975). There are also many localities where, as a result of under-collection, the subgenus has not been previously reported, for example at Hallett Station, Victoria Land (Rudolph, 1963, 1967). On the antarctic continent inland areas often receive more sunshine than coastal areas (Lamb, 1970) and consequently melt-water streams sometimes flow from inland nunataks to coastal outcrops allowing subsequent colonisation (Filson, 1982). The subgenus appears to have a southern distributional limit, since it is often absent from localities that might be considered favourable that support Umbilicaria species (Lindsay & Brook, 1971), although duration of snow-cover may again be a critical factor. Microlichens extend further south than macrolichens (Lamb, 1970; Siple, 1938). This may be due to the narrowing of the boundary layer of warmer air through which fruticose lichens extend (Kershaw, 1983), coupled with an increase in the effect of ice-blast, the relative harshness of which increases with decrease in temperature (Lindsay, 1977a). This is enhanced by the constant katabatic winds (Lamb, 1970; Greene & Longton, 1970) which, combined with the freeze-thaw action may modify thallus morphology (Filson, 1982). Edaphic factors are important since most species are obligate saxicoles, although a few may rarely secondarily occupy more diverse substrates. For example, Usnea antarctica may occa- 26 F. J. WALKER sionally grow on bryophytes, discarded timber, eroded peat, or soil (Lindsay, 1973, 1978; Smith & Corner, 1973), although the range of habitats it is able to colonise decreases towards its southern limit of distribution. Similarly, the South American species, U. acanthella and U. neuropogonoides , may become detached or fragmented and assume a terricolous habitat. The subgenus almost always occurs on acidic rocks and is regarded as calciphobous; it is only able to colonise stable ground (Smith, 1973). For example, in New Zealand, the species of the Usnea ciliata complex are virtually confined to chlorite schists and, to a lesser extent, greywacke or, rarely, rocks of volcanic origin. In contrast, in the subantarctic regions, U. antarctica is often found on volcanic rocks and lavas, for example on Macquarie Island (Huntley, 1971) and on Marion and Prince Edward Islands (Lindsay, 19776), although the species is unable to grow in the vicinity of active fumaroles that occur, for example, on the South Sandwich Islands (Longton & Holdgate, 1979). Similarly, in Patagonia, Neuropogon species are found on a wide range of substrates from lavas to basaltic rocks and sandstones. Observations in continental Antarctica (Siple, 1938) have shown that whilst Usnea antarctica and U. sphacelata both occurred on granitic and metamorphic sediments, such as schists and sandstones, U. antarctica was more widely distributed on the former. This species is also calciphobous since studies of schist-marble boundaries (Gimingham & Smith, 1970; Smith, 1973) show an abrupt demarkation at the interface. Similarly in the arctic, Lynge (1941) observed that U. sphacelata was common on basaltic or siliceous deposits and did not occur on purely calcareous rocks, although recent collections indicate that this species may rarely occur on slightly calcareous rocks (E. Hansen, pers. comm.) and sandstones. It may be possible that the hardness of the rock in addition to its composition may also influence colonisation. For example, on the Kar Plateau, South Victoria Land, Usnea antarctica was found to occur on dolerite boulders covered with a wind-blown deposit of sandstone grit (Schofield, 1972), thus providing a more suitable substrate. Similarly, in New Zealand, strongly foliated schists are easily colonised (Mark & Bliss, 1970). In addition to substrate composition aspect is also important in relation to microclimate. Despite being able to tolerate very exposed conditions, it is often noticeable that some degree of shelter is necessary against wind or, in maritime environments, against salt-spray. Field observations in New Zealand have shown that the growth of species of the Usnea ciliata complex is more luxuriant on the underhangs of sheltered vertical sides of tors and exposed ridges than on horizontal faces. U. subcapillaris , and presumably U. pseudocapillaris , tend to favour more sheltered habitats and crevices. Similarly in Antarctica, Usnea antarctica occurs in more sheltered habitats than U. sphacelata and consequently the species is usually found at lower altitudes, often growing more luxuriantly some distance inland than at sea-level (Lamb, 1964) due to mist-formation caused by coastal temperature inversions (Lindsay, 1971 a). Where both species occur together in continental Antarctica, U. antarctica is generally rare (Dodge, 1962; 0vstedal, 1978) and poorly developed, as it is just surviving at the southern limit of its distribution. Thalli are often small, atypical, and as such may be difficult to identify conclusively. Consequently communities dominated by U. antarctica, that are frequent in the subantarctic regions and on the Antarctic peninsula, are rare in continental Antarctica, being restricted to sheltered habitats (Longton, 1973). In contrast, U. antarctica cannot tolerate excessive moisture. For example on the South Shetland Islands, this species is found in a community with bryophytes and. Umbilicaria antarctica where melt-water collects, but is only prominent on dry surfaces and south-facing aspects (Lindsay, 19710). Within the subgenus species exhibit varying ecological requirements and tolerances, with asexual species often having wider distributions and occupying a wider range of habitats than their fertile counterparts. The absence of any sexually reproducing species from continental Antarctica may primarily be controlled by environmental conditions rather than distributional factors. Hawksworth (1973), in discussing ecological differences between primary and secon- dary species, observed that greater soralia production occurs in humid situations and that sorediate species are more able to utilise atmospheric moisture. For example, Usnea antarctica has much broader requirements than U. aurantiaco-atra, being able to colonise more varied habitats and occupy a wider range of altitudes (Lamb, 1964; USNEA SUBGENUS NEUROPOGON 27 Lindsay, 1971a). U. aurantiaco-atra is regarded as a montane, maritime species (Lindsay, 1971a; 1977) which tends to favour slightly more sheltered aspects (Smith & Corner, 1973) and is less tolerant of strong winds (Smith, 1973). This species is unable to produce mature spores towards the southern limit of its distribution (Lindsay, 19710). U. antarctica, in contrast to U. aurantiaco- atra, is slightly tolerant of nutrient-enriched melt-water (Lindsay, 1969, 1971a). This species is usually regarded as indifferent rather than nitrophilous (Lamb, 1964, 1970), since it may sometimes grow in nitrogenous habitats around bird rookeries with, for example Caloplaca regalis, Xanthoria elegans, and Lecanora aspidophora (Lindsay, 1971a). Both species of Usnea are regarded as halophobous (Lindsay, 19710), being replaced in coastal communities by Ramalina terebrata (Follmann, 19656), although U. antarctica may be able to tolerate some salt spray. From general observations on the distribution and ecology of Usnea antarctica and U. sphacelata throughout their respective ranges, it is apparent that the latter is characteristically a species of exposed habitats, usually not in close proximity to the sea, and is consequently the more common species is continental Antarctica. Consequently in the arctic, this species is only found where the climate is sufficiently continental (Bliss, 1981; Greene & Longton, 1970; K. Hansen, 1962; Lindsay, 1977a, 1978). The somewhat restricted distribution in the arctic may therefore be due to climatic as well as biogeographical factors, although its absence from large continental areas may be due to under-collecting rather than ecological reasons (E. Hansen, 1982). Lamb (1964), from studies in the Antarctic peninsula, indicated that although the distribu- tions of the two species overlap, U. sphacelata is often found at higher altitudes than U. antarctica and has a somewhat southerly and easterly distribution in that area. On the Antarctic peninsula U. sphacelata is often replaced by U. subantarctica at low altitudes, a species that appears to have ecological requirements that are intermediate between the other two species. In the subantarctic regions species occur that have different ecological requirements from the antarctic-arctic species. The requirements of the Usnea ciliata complex have already been discussed and these species from an alpine-southern temperate element which, in Patagonia, is represented by U. perpusilla and U. acromelana. In addition in this area a specialised group of species is prominent which may be referred to as a transitional arid-montane element. This comprises U. durietzii, U. neuropogonoides, U. patagonica, and U. trachycarpa, which are characteristic of areas of low rainfall between the main high Andean Cordillera and the Patagonian plains, although often extending into both areas. Such species occupy a variety of habitats and, although often occurring with arctic-alpine or alpine southern-temperate species, are primarily adapted to open, exposed, dry but misty, rather than necessarily polar-alpine, habitats. Phytosociology Antarctica and the surrounding subantarctic regions have been divided into various ecological zones, primarily based on phanerogamic communities, by numerous workers (see Holdgate, 1970), and those defined by Longton (1966) have been most widely accepted. Three main zones are recognised: the southern cold-temperate, the subantarctic, and the antarctic, based on the position of the subtropical and Antarctic Convergences (Holdgate, 1970; Skottsberg, 1960). The subantarctic regions have been variously defined (Bliss, 1979; Godley, 1960; Greene, 1964a; Skottsberg, 1960; van Zinderen Bakker, 1971; Wace, 1960) and various islands are included that lie in the vicinity of the Antarctic Convergence, either side of which marked climatic, and hence vegetational differences occur. For example, although South Georgia and Isla de Los Estados (Staten Island) lie on approximately the same latitude, the snow-line on the former is lower than the tree-line on the latter (Deacon, 1960). Consequently South Georgia has been included in the subantarctic zone for floristic and climatic affinities (Greene, 1964a, 19646; Lindsay, 1975), although possessing affinities with both the antarctic and subantarctic regions (Bliss, 1979). Similarly various classifications of arctic vegetation have been proposed, which have been 28 F- J- WALKER summarised by Bliss (1981). Bliss (1979, 1981) has also recently described a new classification for the vegetation of polar and alpine regions based on integrated ecological information. Broadly, various biomes (arctic, subarctic; alpine, subalpine; antarctic, subantarctic) were defined and each divided into low and high zones according to altitude or latitude. Within this regime each high zone may be further subdivided into desert and semi-desert units, thus reflecting the concept of ecological cold-deserts of Lindsay (1977a). This classification is much more flexible than earlier proposals since it is not based primarily on strict geographical or climatic boundaries. Subdivisions of each biome are based on the distribution of key genera which, in the polar desert subdivisions includes bryophytes and lichens. Consequently a particular island or region may support different vegetation zones and be placed in the relevant divisions, thus supporting evidence that gradients exist between zones (Longton, 1967). The divisions maritime (or oceanic) and continental antarctic (Holdgate, 1970) are retained here for convenience, since they reflect the distribution and ecology of particular Neuropogon species, although both are included in the polar desert subdivisions of the high antarctic biome by Bliss (1979). The maritime antarctic includes Bouvet0y, the South Sandwich Islands, the South Orkneys, South Shetlands, the Palmer Archipelago, and the western coast of the Antarctic peninsula as far south as Marguerite Bay. Continental Antarctica has been further subdivided into three zones (Holdgate, 1970; Schofield, 1972), but these are not discussed here. As a result of climatic differences, lichen communities in the maritime antarctic are more diverse, having some affinities with those of the boreal-arctic zones, than those found in continental Antarctica which are local, less diverse, and show affinities with the high arctic (Lindsay, 1978). Antarctic A range of lichen and bryophyte phytosociological communities have been described from the antarctic and subantarctic regions which reflect the general ecological differences that occur between Neuropogon species. In contrast, little ecological work has been undertaken in continental Antarctica and frequently communities described do not contain Neuropogon species. The most detailed research has been undertaken in the maritime antarctic (Allison & Smith, 1973; Gimingham & Smith, 1970; Lindsay, 1969, 19710; Longton & Holdgate, 1979; Redon, 1969, 1973; Smith, 1973; Smith & Corner, 1973) of which three groups are relevant: those in which Neuropogon species either form a major component, or are less important, or are absent. Fruticose and foliose lichens predominate (Lindsay, 19770) and three major community types dominated by Neuropogon may be recognised. The Usnea antarctica sociation is characteristic of windswept, gravel-covered cols, and windgaps; whilst that of U. aurantiaco-atra (as U. fasciatd) is found on boulder-fields; and that of U. sphacelata (as U. sulphured) is more characteristic of the east coast of the Antarctic peninsula and continental Antarctica. The first two community- types only are discussed in more detail below. Habitats on the west coast of the peninsula become drier with increasing latitude (Longton, 1967) and consequently vegetation is much sparser and communities less complex, when, for example, those of the Argentine Islands are compared with the South Orkneys (Smith & Corner, 1973). Usnea antarctica is the most widespread Neuropogon species in the maritime antarctic, and many communities have been based on this species, for example the alliance Neuropogonion antarcticum (Follmann, 1967). U. antarctica is the dominant species in the Andreaea- Usnea sociation on the South Orkneys (Smith, 1973) and forming the Usnea antarctica sociation on the South Shetlands (Smith, 1973, 1984) and South Orkneys (Allison & Smith, 1973), which is rare in continental Antarctica (Longton, 1973). The species is also found in the Lecideetum sciatraphae (Follmann, 1967). It is often a primary coloniser (Lindsay, 1978) of boulders in moraines and there is often a succession of U. antarctica followed by crustose species, including Buellia anisomera and B. russa (Lamb, 1970), although the reverse has been observed (Lindsay, 1978). Moss banks may be colonised by a range of lichen species (Smith, 1973) including USNEA SUBGENUS NEUROPOGON 29 Ochrolechiafrigida, Cladonia rangiferina, Sphaerophorus globosus , and U. antarctica, when the growing tips of the mosses are not covered by snow. U. antarctica is the major component of the Sphaerophoretum teneri (Redon, 1969) and is occasionally found in more halophobous com- munities with Himantormia lugubris and Andreaea species (Allison & Smith, 1973). In contrast Usnea aurantiaco-atra occurs in much more halophobous and nitrophobous communities than U. antarctica. For example, on the South Orkneys this species forms a distinct community on raised beaches with Himantormia lugubris over a range of a latitude with species of Andreaea, Rhizocarpon, and Lecidea. U. aurantiaco-atra is also found in communities with halophobous species, including Sphaerophorous and Stereocaulon. Lindsay (1971a) has de- scribed a series of communities along a gradient away from a slightly nitrophilic habitat dominated by U. antarctica that is finally replaced by the U. aurantiaco-atra-H. lugubris- Andreaea sociation. Occasionally communities may contain both Usnea antarctica and U. aurantiaco-atra (Smith, 1973). U. antarctica is characteristic of dry deposits of fine gravel and sandy soil and is often replaced on exposed, windswept ridges by U. aurantiaco-atra and Himantormia lugubris. However, U. aurantiaco-atra is also found in an Usnea-Umbilicaria-H. lugubris sociation on dry, exposed rock faces at higher altitudes, which are dominated by U. antarctica and species of Buellia, Lecidea and Rhizocarpon. Similarly, in the Andreaea-Grimmia-Usnea-Umbilicaria sociation (Smith & Corner, 1973) U. antarctica is the most prominent species on windswept outcrops whilst U. aurantiaco-atra is only locally abundant in slightly more sheltered situations. Subantarctic The subantarctic islands are all characterised by an extremely cool, oceanic or maritime climate with little annual or diurnal temperature variation (Du Rietz, 1960), high rainfall, constant high humidity, and strong westerly winds (Eaton, 1879; Greene, 1964a; Wace, 1960). Lichen communities are not so prominent, being confined to very exposed rocky situations on, for example, Marion Island (Lindsay, 1977ft) and Macquarie Island (Filson, 1981). Consequently most ecological and floristic work has been concentrated on phanerogamic communities (Greene, 19640, 1964ft; Hooker, 18790, 1879ft; Taylor, 1955; Wace, 1960). Two species, Usnea antarctica and U. aurantiaco-atra, that are characteristic of the maritime antarctic, occur in this region, with the latter being replaced by U. taylorii beyond the eastern limit of its distribution in similar habitats (Dodge & Rudolph, 1955). U. trachycarpa also occurs on lies Kerguelen but is characteristic of exposed, dry, windswept, rather than alpine, habitats. Consequently the species of Neuropogon occurring in the subantarctic are a mixture of antarctic and cold- temperate elements, bearing strongest affinities with the maritime antarctic. The lichen flora of South Georgia has affinities with the other subantarctic islands, Tierra del Fuego, the Falkland Islands, and the Scotia Arc. Communities dominated by fruticose and crustose lichens (Lindsay, 1975) were found to be similar to those of the South Orkneys. For example, an Usnea aurantiaco-atra (as U. /osa'ataO-crustose lichen community was widespread on moderately exposed, dry, boulders and cliff faces in which U. aurantiaco-atra often gave 80% cover with an understorey of mosses, Lecidea, Lecanora, Pertusaria, and Rhizocarpon species. In contrast at lower altitudes an U. antarctica-cmstose lichen community, tolerant of slightly nitrogenous melt-water, had less cover of Neuropogon and a greater variation of crustose lichen understorey. U. aurantiaco-atra also occurs in a markedly nitrophibous community, which has a restricted distribution in very dry situations at high altitudes, in which Pseudephebe pubescens and Alectoria miniuscula were often co-dominant, forming an understorey which virtually excluded crustose lichens and mosses. Lichen communities on Marion and Prince Edward Islands have been briefly outlined by Lindsay (1911 b). Communities on Marion Island were found to be local on exposed ridges and dominated by Usnea, Alectoria, Himantormia, and Umbilicaria species. U. antarctica (as U. insularis) was an associated species in a community dominated by Lecidea species on moderately exposed boulders and rock-faces. Little revision has been made of the lichen flora of lies Kerguelen since that made by Crombie (18760, 1877, 1879a) of Hooker's work (Hooker & 30 F. J. WALKER Taylor, 1844; Hooker, 1847) apart from Dodge's contributions (Dodge, 1948, 1966). Crombie (18790) reported U. taylorii from high altitudes on lies Kerguelen and his other records refer to U. antarctica and U. tr achy carp a which may sometimes occupy the same habitats. Southern cold temperate The southern cold-temperate zone has been variously defined (Darlington, 1965; Godley, 1960; Holdgate, 1970; Skottsberg, 1960) and very broadly encompasses regions of southern South America (including the Falkland Islands), New Zealand and the associated shelf islands, and even the south-eastern corner of Australia with Tasmania. In these regions most Neuropogon species reach their northern distributional limit and are primarily represented by a group of species that may be termed alpine-southern temperate, rarely with isolated occurrences of antarctic or subantarctic species. In Australasia this element is well-represented by the Usnea ciliata aggregate. Little ecological work has been undertaken in this region apart from studies in Central Otago, New Zealand by Mark & Bliss (1970). Similarly, little is known of the alpine lichen flora of Tasmania, where Neuropogon species are rare and confined to relatively few high altitude localities (c. 850-1250 m). The dominant alpine phanerogamic vegetation is different from that of New Zealand (Kirkpatrick, 1980) which indicates that climatic and vegetational differences exist at corres- ponding latitudes which exclude Neuropogon. Recent observations on the Rock and Pillar Range, Otago, have shown that a characteristic community, dominated by Neuropogon species, occurs on exposed, isolated rock tors in Celmisia-Poa herbfield at the boundary (c. 1200 m) of the low and high alpine zones (Mark & Bliss, 1970) or biomes (Bliss, 1981). Such communities were dominated by Usnea acromelana and U. ciliata with occasional small thalli of U. subcapillaris . Associated species included Alectoria nigricans, Coccocarpia palmicola, Hypogymnia lugubris, Lecanora polytropa aggr. , Lecidea spp., Menegazzia aeneofusca, M. castanea, M. lucens, M. nothofagi, Parmelia pet- riseda, Pertusaria dactylina, P. super ba, Umbilicaria spp., and Usnea torulosa. At lower altitudes Usnea ciliata is much rarer, for example on Mt. Maungatua, Otago (c. 880 m) and only U. acromelana was found growing with U. torulosa on scattered tors in tussock grassland. At even lower altitudes, c. 370 m, on an exposed plateau at the foot of the Rock and Pillar Range, only U. acromelana was found. This species was only represented by a few deformed thalli on schist boulders in a community dominated by Parmelia and Xanthoparmelia species, including P. mougeotina, P. petriseda, P. pseudosorediosa, P. reticulata, P. signifera, P. subrudecta, P. sulcata, X. mexicana, and X. tasmanica. Associated species included Buellia macularis, Lecanora blanda, Lecidea irrubens, Rhizocarpon geographicum aggr. , and Siphula coriacea. The climate of southern South America is more oceanic than at corresponding latitudes in the northern hemisphere and, in some respects, is more like the montane climate of the equatorial Andes (Troll, 1960). The mountainous Magellanic archipelago provides few habitats that are sufficiently dry (Holdgate, 1960) for Neuropogon species. A southerly decrease in the level of the permanent snow-line and a sharp climatic gradient from west to east (Auer, 1960) provides a wide range of habitats and hence communities. The mountainous region of the Argentine-Chile border is subject to almost constant winds and frequent storms (Shipton, 1959) which impose rigorous conditions. There is an abrupt change to the east from Nothofagus forest on the slopes of the higher mountains to the grasslands of the lower hills and undulating plateaux (Shipton, 1959) which extend eastwards to form the dry, semi-desert plains (mesetas) of Patagonia (Darlington, 1965). Ecological studies have previously primarily been directed towards the classiciation of phanerogamic vegetation (Godley, 1960; Holdgate, 1960; H. Weber, 1969). Previous lichenolo- gical investigations have mainly been taxonomic, adding to distributional records (Crombie, 18766; Grassi, 1952; Lamb, 1948a, 1955; Rasanen, 1932, 1939; Redon, 1972; Santesson, 1942) with little ecological investigation (Lamb, 1959; Mattick, 1951). Lamb (1959) made brief reference to a saxicolous alpine community composed of species of Neuropogon associated with USNEA SUBGENUS NEUROPOGON 31 Lecidea species, Pseudephebe pubescens, Rhizocarpon adarense, and Umbilicaria decussata, which was found to be a mixture of antarctic and subantarctic elements and was similar to those described from central Chile (Follmann, 1965a; Redon, 1973) probably based on U. patagonica (as U. acromeland). Redon (1972, 1974) described various alpine communities from Chile and found them to be similar to those occurring on the Argentinian side of the Cordillera. U. aurantiaco-atra was found in a community with Rhizocarpon geographicum, R. crystalligenum, Umbilicaria cylindrica, and U. nylanderiana. Altogether ten species of Neuropogon are found in southern South America which may be grouped into various floristic elements that form distinct communities or intergrade. Usnea sphacelata represents a continental antarctic element that is confined to high alpine areas, particularly those associated with areas of glaciation. U. antarctica, U. aurantiaco-atra, and U. subantarctica represent the northern limit of a maritime antarctic element. Alpine southern- temperate species are represented by U. acromelana, which is restricted to glacial valleys and moraines, and the endemic U. perpusilla which tends to replace U. aurantiaco-atra towards northern Patagonia. Perhaps the most interesting group of species are those which form a transitional arid-montane element, represented by U. durietzii, U. neuropogonoides, U. patagonica, and U. trachycarpa, which occur within a wide range of altitudes and are capable of extending into the mesetas as well as the lower alpine slopes of the Cordillera. Consequently many unusual communities occur which contain several of these floristic elements. For example, on the lower mountains of Tierra del Fuego, at c. 800 m, collec- tions have been examined consisting of U. antarctica and U. cf. subantarctica, together with un- usually fibrillate forms of U. aurantiaco-atra, U. perpusilla, and U. trachycarpa. The lichen flora is less diverse on the Falkland Islands since low altitude, maritime communi- ties containing Usnea antarctica are absent, possibly due to univestigated ecological factors or extinction due to changes in land-use. Only remnants of a unique lichenicolous heath commun- ity (Standring, 1983) now remain which support Protousnea species (Krog, 1976), and which is probably closely related to the Magellanic moorlands of Chile. On these islands strong winds, rather than low temperatures, favour the growth of U. trachycarpa whilst the maritime antarctic element is represented solely by U. aurantiaco-atra. Tropical-alpine The climate and the flora of the tropical high mountain zone of Central and South America have been compared to that of the subantarctic (Darlington, 1965; Troll, 1960; Wace, 1960) as for example, seen by a comparison of tussock grassland and the paramos. By contrast, these vegetation types are very different from those of the cold-temperate zone of the northern hemisphere. For example, in Ecuador, at Quito (2850 m) there is great diurnal but little annual temperature range (Troll, 1960) whilst at higher altitudes, close to the permanent snow-line (e.g. Mt. Chimborazo, c. 4750 m) there is little annual or diurnal temperature variation. Usnea durietzii is found at the lower altitudes (c. 3000-4000 m) and is characteristic of dry, exposed, rocky paramos in communities associated with U. acanthella, U. bogotensis, and Stereocaulon species, whilst U. sphacelata is confined to higher, alpine habitats. Biogeography Many authors (Ahmadjian, 1970; Dodge, 19650; Filson, 1982; J0rgensen, pers. comm.; Lamb, 1970) regard the antarctic lichen flora as a mixture of relict endemic elements, which survived Pleistocene glaciations, as well as immigrant elements that recolonised the continent from adjacent areas, for example Fuegia. Lindsay (19770) considered the nunatak cryptogamic flora of Antarctica was, in contrast to the arctic, somewhat impoverished as a result of polar isolation, more drastic climatic changes during the late Cenozoic, and slower rate of recolonisation. However, despite physiographical, climatic, and vegetational differences between the polar regions, taken in its entirety, the present-day lichen flora of Antarctica is as rich as that of the arctic, thus indicating a varied origin and opportunities for speciation. 32 F. J. WALKER The present day distribution of the subgenus Neuropogon has often been used as an example by biogeographers when discussing origins of bipolar disjunctions. Together with other genera which have a predominately southern hemisphere distribution, for example Menegazzia, Placopsis, and Pseudocyphellaria, it has been generally accepted (Du Rietz, 1929, 1940; Galloway, 1979; Henssen & Jahns, 1973; James, 1960; Lindsay, 1977a; Lynge, 1938, 1941) that the subgenus Neuropogon possibly had a monophyletic southern origin. However, many features are shared with the subgenus Usnea, for example pigmentation of pseudoisidia, and this could indicate either common ancestry or subsequent parallel development. This contrasts with other genera, for example Bryoria and Evernia, which are thought to have evolved in the northern hemisphere, since they are today poorly represented or absent in the southern hemisphere (Galloway, 1979). J0rgensen (1983) postulates that the present-day tricentric antarctic distribution of fertile species, as exemplified by Usnea aurantiaco-atra in southern South America-Antarctic Peninsu- la, U. ciliata in New Zealand, and U. taylorii in the subantarctic Indian Ocean, indicates ancient antarctic origins. This is despite difficulties in associating a cold-tolerant group with Gondwana- land when glaciation only occurred comparatively recently. A possible centre for subsequent speciation may have been in South America-west Antarctica, but J0rgensen proposes that some elements must have been present in Gondwanaland to account for present-day distributions. Krog (1982) considered that disjunct distributions of several genera in the Parmeliaceae could indicate origins in the Cretaceous or Late Jurassic. Present-day distributions of fertile Neuropo- gon species indicate that the subgenus may have been established in Gondwanaland prior to the opening up of the south Atlantic at the start of the Cretaceous (Raven & Axelrod, 1974). The asexual species Usnea patagonica, which occurs in Patagonia and possibly South Africa, may represent the remnants of a once much richer flora in South Africa that was subsequently wiped out. However, distributions based on asexual species are likely to be speculative in view of the likelihood of subsequent long-distance dispersal. Krog (1982) also suggested that the distribution of Neuropogon is an indication that the subgenus is more primitive than the more widespread and diverse subgenus Usnea. Today Usnea s. str. is mainly a tropical genus with only a few, all asexual, species sharing the same habitats with Neuropogon. This distribution indicates that Usnea may be a much younger group and any connection between the two subgenera must be very ancient. Certainly some of the South American species included in the subgenus, for example, Usnea durietzii and U. patagonica, have close affinities with the subgenus Usnea which are discussed under 'Generic concept', (p. 42). These species might either suggest evolution from Neuropogon to Usnea or convergent evolution, or represent remnants of an ancestral group from which Neuropogon s. str. adapted to tolerate polar environments and Usnea spread to occupy a wider range of habitats and less severe ecological conditions. Similarly, one species, U. neuropogonoides , has some affinities with Protousnea although the two groups differ in other respects, including chemistry. Protous- nea appears to be a further modification of the basic Usnea-type, and may be regarded (Krog, 1976) as a rather primitive group with possible affinities with hypothetical predecessors. The unique thallus anatomy of Usnea taylorii might be interpreted as an indication of a primitive species, but is more likely to have been secondarily derived from a species with a thick axis, for example U. aurantiaco-atra. This species is confined to the lies Kerguelen, Heard Island, and lies Crozet. There is geological evidence (Brundin, 1970; Harrington, 1965) that these islands are partly ancient and partly of recent volcanic origin; thus indicating a Gondwana- land connection with an ancient endemic flora (Dodge, 1948). Biogeographically these islands, with Marion and Prince Edward Islands, form the Kerguelen Province (van Zinderen Bakker, 1971). There is evidence that the so-called 'Grande Terre' of the lies Kerguelen, which straddles the Antarctic Convergence, had an aberrant type of glaciation with certain sheltered areas probably serving as refugia for flora and fauna during the glaciations (van Zinderen Bakker, 1970) ; A parallel may be drawn from studies on an endemic genus of chironimid midges found on lies Crozet (Brundin, 1970) which are thought to be the survivor of a group whose ancestors existed during the Jurassic-Cretaceous transition. This would indicate that such islands must have been USNEA SUBGENUS NEUROPOGON 33 isolated before the separation of Antarctica, New Zealand, and Australia, and like lies Kerguelen and Heard Island are thought to be of continental origin (Shields, 1976). The bicentric distribution of Usnea trachycarpa in Patagonia with the Falkland Islands and lies Kerguelen suggests that this could be a very old species with origins in Gondwanaland, since it is unlikely to have subsequently spread by long distance dispersal. The Falkland Islands are thought to have Precambrian origins (Shields, 1976). Even as early as 1879 Hooker (Hooker, 18790) recognised affinities between the vascular and cryptogamic floras of lies Kerguelen and South America. This is in contrast to the much younger subantarctic islands that were of recent volcanic origin, for example Marion and Prince Edward Islands, which were also subsequently glaciated (van Zinderen Bakker, 1970; Verwoerd, 1971). Despite the previous acceptance of an endemic sorediate species oiNeuropogon, Usnea insularis (= U. antarcticd) (Dodge, 1948; Lamb, 19390; Lindsay, 19776) it is consequently unlikely that such a species could have had sufficient time to evolve. It is more plausible that U. antarctica recolonised these islands post-glacially by long distance dispersal via the Scotia Arc, which is probably of continental origin, or with islands dating from Precambrian to Cretaceous times (Shields, 1976). Lindsay (19776) indicates that the lichen flora of the Prince Edward Island group has affinities with that of the lies Kerguelen, as exemplified by Orceolina kerguelensis , which has recently been found on lies Crozet (Tilman, BM). However, these islands floras also share a common bipolar element. Bouvet0y and the other islands of the Scotia Arc, which extends eastwards from Tierra del Fuego through South Georgia to the South Sandwich Islands then westwards through the South Orkneys towards the Antarctic peninsula (Greene, 1964), were similarity glaciated and con- sequently have a low incidence of endemism. Two main species, Usnea antarctica and U. aurantiaco-atra have recolonised these islands from Fuegia, and Bouvet0y represents the easterly limit of the latter species. Some ice-free nunataks may have existed along the Scotia Arc, although there is, as yet, no geological evidence, where U. antarctica and U. aurantiaco- atra may have survived glaciations (Lindsay, 1975). The South Sandwich Islands are of recent origin, formed from a crust c. 8 million years old, and have no exposed rocks older than c. 3 million years and are still actively volcanic. Consequently the flora must have arrived by transoceanic migration (Longton & Holdgate, 1979). Biogeographers attach considerable importance to the extent of endemism, particularly at species level, in a given area. Results are often conflicting and may, in some instances, be more relevant for phanerogams. With lichens it is essential for the flora of a particular region to be well documented and compared to that of adjacent areas. The flora of the antarctic regions is a test case where failure to do this can lead to recording of spurious endemism, as, for example, in the case of the taxa described by Dodge (19650, 1970). Many groups still require critical systematic revision, which may show that many endemic taxa are superfluous, merely representing environmental modifications rather than genetic differences (Filson, 1982). Geological evidence (Harrington, 1965) indicates that east and west Antarctica had different origins, with the western, younger part having close affinities with South America. The islands of the northern Antarctic peninsula are thought to have Middle Paleozoic origins (Shields, 1976). Dodge (1948, 19656, 1973) recognised endemic species of Neuropogon in east Antarctica derived from Usnea antarctica. Although this area is older, with Gondwanaland origins, such elements are now regarded as ecotypes (Ahmadjian, 1970). Their precursors may have survived Pleistocene glaciations on inland nunataks or ice-free refugia (Lamb, 1970) and subsequently spread along melt-water channels to colonise coastal areas (Filson, 1982). However there would not have been sufficient time for speciation to occur, since the rate of evolution in lichens is thought to be very slow (Lindsay, 19770; J0rgensen, 1983) and then growth is even slower in polar regions (Lamb, 1970). This is in contrast to the proposed relict indigenous elements which may have survived at high altitudes above the tree-line during more temperate eras and form the truly endemic portion of the flora. The Antarctic peninsula is considered (Dodge, 1973; Lindsay, 19770) to have a high incidence of specific, but low generic, endemism. This is exemplified by the occurrence of lichen groups which have developed unusual characteristics such as the possibly more highly evolved stipate 34 F. J. WALKER forms of many crustose genera (Lamb, 1970). Many of these species also occur in southernmost South America, having a distribution pattern similar to that of Usnea aurantiaco-atra and U. subantarctica. Such species probably evolved during the rapid phase of speciation that occurred in Patagonia after the initial break up of Gondwanaland, and spread postglacially into the Antarctic peninsula via the Scotia Arc. Limited dispersal of fertile species must have occurred by this route although the Arc was formed after the separation of the Antarctic peninsula and South America. The timing of the separation of the Antarctic peninsula is uncertain but an estimate of 29-3 million years B. P. is preferred by Crook (1981). Dodge (1973) recognised a greater number of Neuropogon species in the Antarctic peninsula than are accepted here. At present five, or possibly six, species are known, of which none are endemic. In contrast ten species occur in Fuegia, of which one fertile species, U. perpusilla, is endemic, and one, U. aurantiaco-atra, is common to the Antarctic peninsula and islands. It is clear from present-day distributions that there has at some stage been explosive speciation of Neuropogon in southernmost South America and subsequent spread, possibly as a response to drastic climatic changes (van Zinderen Bakker, 1970) and opening up of ecological possibili- ties. This does not necessarily prove that the area coincides with a centre of origin and could conversely be regarded as an area where many species occur as the result of continual replacement or accumulation of relicts (Darlington, 1965). There is evidence that the greater part of Tierra del Fuego was glaciated although refugia may still have existed for old species, such as U. trachycarpa. Other species, such as U. antarctica, may have evolved from a fertile primary species, i.e. U. aurantiaco-atra, and subsequently spread postglacially by long distance dispersal. South America has long been of biogeographical interest as a centre of speciation and a migration route to and from the northern hemisphere. With respect to Neuropogon, -it is apparent that at some stage taxa were isolated in the northern Andes. Such proximity to the tropics and varied climates has led to the evolution of a series of closely related, asexual species, such as Usnea acanthella and U. durietzii, which together with U. sphacelata are often difficult to identify and may indicate either a centre of great diversification or parallel evolution. Only one of these asexual, South American taxa managed to spread into North America. Usnea sphacelata probably had its origins in Patagonia since it may be the sorediate counterpart of U. perpusilla. Various theories as to how and when this migration occurred have been proposed or discussed by previous authors (Du Rietz, 1929, 1940; Lynge, 1941). It seems probable that migration north may have occurred prior to the recent joining of North and South America; possibly along a series of volcanic islands during the Late Cretaceous (Raven & Axelrod, 1974; J0rgensen, pers. comm.). The absence of a possible fertile counterpart in the northern hemisphere and the uniformity of the population in contrast to the greater variation found amongst southern hemisphere populations indicate recent migration to the northern hemisphere. A similar situation is seen in Placopsis and Menegazzia. However, it might be argued that less variation is found because climatic conditions have been more stable than in Antarctica during the late Cenozoic glaciations and the arctic was not isolated geographically (Lindsay, 19770). Spread in the northern hemisphere was apparently eastward towards Green- land, following the prevailing winds, rather than across the Bering Straits. The distribution of U. sphacelata is not truly circumpolar and is mainly in the Canadian eastern arctic, being rare between Novaya Zemlya and western arctic Canada, with a single record from the New Siberian Islands (Lynge, 1941). Thomson (1972) considered that the present day disjunct distributions of many arctic lichens may reflect the availability of suitable habitats rather than the spread from a centre of origin. Since U. sphacelata is more or less confined to island groups in the northern hemisphere rather than the continents, this might be a relevant proposition. Post-Pleistocene colonisation was probably rapid from North America by means of wind-blown propagules and thallus fragments. Biogeographic treatments of New Zealand have been the subject of much debate and were recently summarised by Craw (1978). Galloway (1979) comments that the cryptogamic and phanerogamic floras of New Zealand and southern South America have many common genera but few shared species. For example, in Usnea subgenus Usnea closely related species occupy USNEA SUBGENUS NEUROPOGON 35 similar habitats. U. xanthopoga and U. contexta are replaced by U. nobilis and U. pallida in South America (Galloway, 1979) and also U. torulosa and U. inermis by U. igniaria and U. nidulifera. Similarly different fertile species of Neuropogon also occur. This is also true for Menegazzia where few species are common to South America and New Zealand (P. W. James, pers. comm.). In the New Zealand alpine flora there is a considerable extent of endemism in phanerogams but this is not as high in the lichen flora. New Zealand was separated from Gondwanaland, and hence from Australia and Tasmania, approximately 80 million years B.P. (Raven & Axelrod, 1974), whilst separation of Australia and Antarctica did not occur until 60-53 million years B.P. (Crook, 1981). The subgenus Neuropogon is restricted to schists east of the main divide in the South Island and is rare in the North Island. This substrate specificity could indicate that a norstictic acid-containing precursor of the Usnea ciliata aggregate or even U. ciliata itself, was isolated at that time and subsequently underwent speciation with the resulting four taxa. This isolation occurred well before the more or less direct migration route between South America and Australia via Antarctica was broken, and hence explains why one would not expect U. ciliata to occur in South America, despite erroneous reports (Lamb, 1948a). In South America U. ciliata is replaced by U. aurantiaco-atra and U. perpusilla. The lichen flora of Australia has diverse origins (Rogers & Stevens, 1981) but how and when Neuropogon reached Tasmania and Victoria remains conjectural. As far as vascular plant dispersal is concerned there have been conflicting views as to whether migration was possible from east to west (Wardle, 1978) as well as from Tasmania to New Zealand with the prevailing winds (Raven, 1973). The former migration would had to have occurred if Neuropogon spread to Tasmania after separation of New Zealand. Evidence that this direction of migration has occurred in some groups of organisms is also given by Fleming (1976). An alternative explanation is that ancestors of Usnea ciliata were much more widespread in Gondwanaland and consequently already present in Tasmania and New Zealand prior to continental movements; this certainly seems to be true for species of Menegazzia (P. W. James, pers. comm.). Additional evidence for spread from New Zealand comes from the fact that U. ciliata has not reliably been recorded from Tasmania and the two species present, U. acromelana and U. subcapillaris could spread by soredia or thallus fragmentation. Slight differences between New Zealand and some Australian populations may either be due to ecological conditions or conversely indicative of a longer isolation. Long distance dispersal may have also been by other agents, including birds (Bailey & James, 1979). There is evidence (Burton, 1968) that gulls and skuas use lichens to line their nests. In addition, Wardle (1978) suggested that dispersal of sorediate taxa may have still occurred from Antarctica to New Zealand prior to glaciation. This may account for the presence of an antarctic element, Usnea antarctica and U. sphacelata, in the flora. It seems more probable that U. sphacelata, from its restricted distribution on very old outcrops, is a relict species. Further evidence also comes from the recent discovery of this species in north-west Nelson (herb. Bartlett 25810!, 25811!), together with other bipolar species including Solorina spongiosa (Galloway, pers. comm.). In contrast U. antarctica may have spread more recently from Fuegia or Antarctica. The sorediate Usnea acromelana is common in Australasia and infrequent in southernmost South America and the Antarctic peninsula area. It may be argued that this species subsequently spread from Australasia, where its fertile counterpart, U. ciliata, occurs, to Patagonia via Antarctica before the continents finally separated. However, this present-day distribution does not necessarily reflect earlier patterns and migration might have occurred in the opposite direction, although U. ciliata is today confined to Australasia. U. acromelana is absent from east Antarctica and associated subantarctic islands which may be a result of ecological factors, with habitats more suited to U. antarctica and U. sphacelata. It is also worth emphasising that some populations from the Antarctic peninsula and Patagonia are only tentatively included in U. acromelana and may eventually prove to be examples of parallel evolution. Ecological factors may also explain the limited distribution of U. subantarctica. Fleming (1976) considered that most of the outlying subantarctic islands of New Zealand were not connected to the mainland during the Tertiary, being partly formed during that era by 36 F. J. WALKER volcanism although with older, continental elements (Shields, 1976). He suggested that they were subsequently glaciated or had very cold climates with the result that many inhabitants are more recent colonists. However, Wardle (1978) drew attention to the high level of endemism in the vascular plant flora on the Antipodes, Stewart, Campbell, and Auckland Islands, which suggests they survived glaciations; on the other hand there is evidence that the Macquarie Island flora was to some extent impoverished by glaciations (van Zinderen Bakker, 1970). Dodge (1973) believed that there was no relationship between the antarctic lichen flora and that of lies Kerguelen and the subantarctic islands of New Zealand. However, it has now been shown that the New Zealand flora has a substantial austral element (Galloway, 1979) which has affinities with those of its subantarctic islands (Dodge, 1970) and even with lies Kerguelen, as illustrated by the distribution of the genus Steinera (Henssen & James, 1982) which is probably a relict of a once more widely dispersed flora lost from New Zealand and South America during glaciations (Galloway, 1979). The oceanic climate of Auckland and Campbell Islands is not suitable for Neuropogon species. It is probable that Usnea acromelana spread to Chatham Island from New Zealand, and U. antarctica to Macquarie Island, by later, long distance dispersal. The distribution and evolution of chemical races in Neuropogon has not been considered from a biogeographical standpoint. However, it does seem likely that the precursor of the subgenus might have contained norstictic acid, since this substance is the most widely distributed, with subsequent development of other races. Alternatively, many species, including U. taylorii, often lack medullary substances; a state that is the commoner phase of many antarctic species. Discussion Circumscription of the species The genus Usnea s. lat. is notorious when it comes to delimiting species since there is inherent a great plasticity of form. The subgenus Neuropogon is no exception. Complexes exist within which it may be difficult to delimit individual species because rare linking intermediates are encountered. Such groups of individuals are regarded as noda along an axis of variation and may either be considered as distinct species, if sufficient criteria exist, or else as variants of a single species. Often there are indications that convergent evolution has occurred and this may lead to misidentifications. Immature, damaged, or thalli from sites of ecological stress may prove anomalous or even impossible to identify. In his world monograph Lamb (1939a) attached importance to both morphological and chemical characters when delimiting taxa. The morphological features found to be useful criteria are cortical features, relative widths of axis and medulla and corresponding laxness, and colour of the apothecial disc, and presence of rays. In asexual species the type and formation of propagules is important and may sometimes be species specific. For example, soralia may be plane, crateriform, ulcerose, or globose, and soredia may or may not be pigmented. In some instances the habit and branching of the thallus as well as the form of blackening of the holdfast may also be of value besides distributional and ecological considerations. Frequently, during this study, it has been found that a combination of characters must be taken into account when identifying a particular species. This accounts for the wide range of infraspecific variation and any anomalies that may consequently occur. Microscopical characters, for example spores and conidia, have not been studied in depth and hence no great taxonomic value is attached to them in this instance, although they have belatedly been found to be significant in other instances within the Parmeliaceae (Krog, 1982). Lamb (1939fl) suggested that the structure or formation of conidia may provide useful evidence for separating genera. This theory has been proved by recent authors (Culberson & Culberson, 1980; Krog, 1982; Hawksworth, 1981) who have found them useful in separating genera or species. Krog (1976) used pycnidial morphology to separate Neuropogon from Protousnea based on absence of pigment in the ostiole rather than details of conidia structure or formation. Pycnidia are rather rare in Neuropogon, although further work on conidial shape and formation may provide useful information in confirming conspecificity of taxa such as Usnea aurantiaco- atra and U. melaxantha (see 'Morphology' p. 11). USNEA SUBGENUS NEUROPOGON 37 As mentioned earlier (p. 3) virtually all Lamb's taxonomic work on Neuropogon was carried out before the advent of TLC, relying on thallus spot tests and microchemical crystal tests. Consequently too much weight was sometimes placed on slight differences in concentra- tion of medullary substances. The use of TLC in delimiting species is discussed below. In contrast Dodge (1948, 1973) virtually ignored chemical data and attached considerable weight to small differences in morphology and thallus anatomy. His very narrow species concepts led to the description of a large number of taxa which cannot be accepted today. Reasons for reducing many of these taxa to synonymy are also discussed under the relevant species. From studying a wide range of material some indication has been obtained of the morpholo- gical variation of each species. Concepts may have to vary according to the species concerned, and a wider or narrower range of variation of a particular character accepted. Frequently a broader view must be taken of species which have a wider distribution. These are species which usually only spread by means of vegetative propagules. Such species may occupy a wider range of habitats or have a greater ecological tolerance and hence exhibit more variation than species with a limited range (Krog & Swinscow, 1981). Swinscow & Krog (1978) proposed the concept that widespread and variable Usnea species may be represented by different morphotypes or chemical races towards their distributional limits. At extremes such thalli may look very different and have often been interpreted as distinct species. Consequently several Neuropogon taxa accepted by previous authors are here reduced to synonymy (Table 5). An example is Usnea sphacelata where there are some differences between certain antarctic and arctic propulations, although this variation is not considered to be sufficient to warrant the acceptance of distinct taxa. In addition there may be considerable variation within populations resulting from ecological factors (Filson, 1982, Hawksworth, 1973) which may be misinterpreted. It is possible that the diverse forms met with whilst evaluating the variation of Usnea sphacelata, particularly in South American and antarctic populations, are the result of the response to differing ecological conditions which some authors would recognise as distinct subspecies. Hawksworth (1972) discussed the effect of ecological factors on species delimitation in lichens. He concluded that variations may be caused by phenotypic plasticity but in addition these may have a genetic origin and require taxonomic recognition. Hence, ideally, the need for careful field analysis before new taxa are recognised. It is very likely that montane and arctic-alpine conditions influence variation in thallus morphology of fruticose lichens. Previous authors have come to different conclusions. For example, Kristinsson (1969) studied Cetraria islandica and C. ericetorum in Iceland and found no consistent correlation between morphotypes and chemotypes, finding a range of forms intermediate between the two taxa. More recent world-wide studies by Karnefelt (1979) have revealed that the two Cetraria species may be further subdivided with geographical subspecies, using taxonomic and ecological criteria. Both species have a bipolar disjunction coinciding with racial differentiation and the recognition of a southern hemisphere subspecies for each. However, there are many examples where species, which have their main distribution in the northern hemisphere, and also occur in the southern hemisphere, do not develop a southern race, for example Cladonia mitis andAlectoria nigricans (Karnefelt, 1979). Usnea sphacelata could be regarded in a similar fashion as a series of subspecies. However, in many instances, as with U. acromelana, it seems more plausible that environmental factors are the primary cause of variation. Altitude may be important since thalli from high-altitude 'nunatak' situations are usually distinct, although these intergrade with lower altitude forms. Similarly, although the majority of southern hemisphere material of U. sphacelata examined is slightly different from that of the northern hemisphere, occasionally thalli 'true to type' do occur in populations of varied morphology. Depsidone-containing specimens belong to U. subantarc- tica. Thalli in shaded conditions are greener than those in exposed situations due to lower concentrations of usnic acid. Parallels exist in the pigmentation of antarctic Teloschistaceae (Filson, 1969). In more exposed habitats thalli are often smaller, may be more richly branched, or become subdecumbent, or may be more extensively pigmented. The development of a 38 F. J. WALKER violaceous black pigment in the outer layers of the cortex is thought to be a shielding mechanism against UV radiation. Lindsay (19770) observed that in the continental antarctic, thalli of Usnea sphacelata are heavily or even almost entirely pigmented, whilst in the maritime antarctic (Antarctic peninsula) thalli are only lightly variegated at branch apices. Ecological factors may also effect the development of the lax medulla in some species, for example U. perpusilla and U. sphacelata. Such environmentally induced variation may account for some abnormalities that are occasionally encountered. These include the rare occurrence of minute papillae in species which usually have a smooth surface, for example, U. acromelana and U. taylorii, or the uncharacteristic presence or absence of excipular rays in fertile species. Variation and development of atypical features in eastern antarctic populations of U. antarctica and U. sphacelata has led to the description and acceptance of many distinct taxa from this area by Dodge and his associates, who applied very narrow species concepts and recognised pockets and endemism around the eastern antarctic (Dodge, 1948, 19656, 1973). Some of Dodge's taxa are difficult to assign to either U. antarctica or U. sphacelata, particularly where they are based on poorly developed thalli or when holotype material is not available for study. In addition named material examined of a single species often included both U. antarctica and U. sphacelata. For example, U. subfoveolata and U. subpapillata were both described from the same locality in Queen Mary Land. Type specimens were examined by Lamb (1964) and photographed (Lamb, unpublished notes, A AS!) and are probably referable to U. antarctica, although I have examined material from the type locality, also cited by Dodge, which belongs to U. sphacelata. Both species are present in collections identified as U. subpapillata in US. Similarly Dodge described a further two species from an adjacent locality in Queen Mary Land. Usnea laxissima is a decumbent form of U. sphacelata, whilst U. pustulata is synonymous with U. antarctica. Occasionally, specimens from eastern Antarctica have been misidentified as Usnea acromela- na, (for example, Golubkova & Schapiro 1970; Filson, 1975), although in Antarctica the species is confined to the Antarctic peninsula. Lamb (1939a) described two forms of U. acromelana from Antarctica which are now recognised as belonging to U. sphacelata. Sometimes specimens of U. sphacelata have a very smooth, slightly waxy, unpigmented surface and a somewhat compact medulla. These features, combined with the production of black nodular soralia, suggest some affinities with forms of U. acromelana, although specimens lack the blackened annulations and compact medulla that are features of that species. The form of vegetative propagules is often species specific, but in rare instances there may be an intergradation between granular soredia, pseudoisidia, and true isidia, often with erosion and secondary regeneration. True isidia are only known in one species, U. acanthella, in Neuropo- gon. The formation of various propagules is discussed under 'Morphology' (p. 9). Examples include corticate outgrowths in soralia of Usnea acromelana and U. antarctica which may not be readily distinguished from the pseudoisidia of U. durietzii and U. patagonica. U. patagonica may be regarded as an intermediate species between U. antarctica and U. durietzii (Table 3). Some taxa are now regarded as conspecific which were originally distinguished on the form of the soralium; for example in U. acromelana and var. decipiens (Lamb, 19390). If such distinctions are maintained this should only be at infraspecific level, although sometimes the range of variation can be observed within a single thallus. In some instances it is difficult to distinguish asexually reproducing species where several species overlap in their distribution. This is generally the case in subantarctic South America (Patagonia) where U. antarctica, U. acromelana, U. durietzii, U. patagonica, U. sphacelata, and U. subantarctica are all known to occur. In addition the area is poorly collected. Frequently thalli examined are small or somewhat moribund and consequently have not developed sufficient features characteristic of a particular species. Many of the species are in the depsidone-deficient phase and so chemistry cannot be used as an additional guideline. Where distributions overlap it is also likely that forms intermediate between species may occur. In addition species are reaching their distributional limits and as a result often produce atypical forms. In some instances altitude may have the same effect on the form of the species as latitude. Such a phenomenon is often observed in New Zealand populations of U. acromelana USNEA SUBGENUS NEUROPOGON 39 where thalli from lower altitudes may become discoloured and produce contorted or grotesque forms, often with proliferation of soralia to produce corticate outgrowths resembling pseudo- isidia. In such instances the only species liable to cause confusion is U. torulosa. Recent advances in microchemical techniques resulting in the identification of lichen sub- stances, in particular the development of rapid methods of TLC, have led to differing opinions on the taxonomic status of chemical variants of a species. Some workers regard chemical variants as distinct species, although there are no morphological differences; others prefer to regard them as races of the single species concerned. A well-known example is the range of opinion concerning the Ramalina siliquosa complex which is summarised by Krog & James (1977). Whilst delimiting species within Neuropogon I have attempted, as with morphological variation, to evaluate each case of chemical variation individually, based on the study of a large amount of material. No taxonomic value is attached to accessory substances, for example UV + unknowns, which vary from abundance to absence within a given species. Similarly the occurrence of psoromic acid is rarely considered to be species specific, although it may rarely occur as a common denominator that unites morphologically uniform chemotypes, for example, in Parmelia direagens (Krog & Swinscow, 1981). The chemistry of Neuropogon species is not as complex as in some other genera in the Parmeliaceae since most of the comparatively few depsidones and depsides that are known are closely related chemically. Chemical data have supported morphological evidence that Usnea aurantiaco-atra and U. melaxantha are conspecific. Thalli are occasionally found with a mixed chemistry in areas where the two depsidone-containing races overlap and cannot be disting- uished morphologically or from depsidone deficient material. This is discussed more fully under U. aurantiaco-atra (p. 71) and is a further example of an instance of where extremes in variation were previously thought to represent distinct taxa. Various epithets have been applied to races of U. aurantiaco-atra at different times by previous authors and these are summarised in Table 4. Chemical data, in some instances, may prove useful in the recognition of species groups and species pairs. The concept of a species pair: a fertile species and a derived asexual species, has been accepted or rejected by various authors, usually the former. Du Rietz (1924) proposed two types of clones based on extant or extinct primary species and this concept has been followed up by Hale (1965), Poelt (1970), Swinscow & Krog (1978), and Krog & Swinscow (1981). Poelt (1970) referred to a 'species pair' when both the fertile and asexual counterparts were known and to 'secondary species' where the primary species is unknown, as is frequently the case in some fruticose groups. In contrast, Tehler (1982, 1983) considers that secondary asexual taxa are best expressed as formae rather than as distinct species. In Neuropogon, in the few instances where obvious species pairs occur, I prefer to regard the taxa as distinct species. Otherwise it would become increasingly difficult to delimit taxa at species level besides making allowances for the wide range of infraspecific variation and Table 4 Epithets applied to chemical races of Usnea aurantiaco-atra (Neuropogon aurantiaco-ater) by major authors. Author norstictic/salazinic acid fumarprotocetraric () Motyka (1936-38) melaxantha aurantiaco-atra aurantiaca fasciata strigulosa Lamb (19390) melaxantha aurantiaco-atra aurantiaca strigulosa Lamb (1948a) melaxantha aurantiaco-atra Lamb (1964) aurantiaco-atra fasciata Dodge (1973) aurantiaca aurantiaco-atra melaxantha fasciata siplei (?) strigulosa Lindsay (1975) aurantiaco-atra fasciata 40 F. J. WALKER chemical variation that already occurs. The chemical races of Usnea aurantiaco-atra have already been mentioned above in the context of species delimitation. U. antarctica, from similarities in morphology and anatomy, is thought to be the asexual counterpart. In addition U. antarctica is sometimes found fruiting and also the same two depsidone-containing races as well as a depsidone-deficient race are known. The norstictic-salazinic acid race is extremely rare and is only known from a single collection from the centre of distribution of the corresponding race of U. aurantiaco-atra. Similarly members of other species pairs may have more than one chemotype with no obvious morphological differences. Such chemotypes are regarded as races rather than distinct taxa. Either the primary or derived species may exhibit more than one chemical race. Examples exist in the Usnea ciliata complex which consists of four species dominated by norstictic acid and salazinic acid containing thalli. U. acromelana is without doubt the sorediate counterpart of U. ciliata, sharing a range of morphological features including a waxy, annulate surface, a compact medulla, and, when fertile, apothecia with rays and a blackened disc. In Australasia the two species overlap and only one chemical race, with norstictic and salazinic acids, is usually encountered; depsidone deficient material of U. acromelana is confined to a single locality in Victoria. U. acromelana has a more widespread distribution and is a rare species in south- ernmost South America, the Antarctic peninsula and islands. Within this South American- antarctic distribution three chemical races are known: (1) norstictic acid and salazinic acid, (2) psoromic acid, (3) depsidone deficient, indicating greater diversity at distributional limits. Two other Australasian species, U. subcapillaris and U. pseudocapillaris , belong to the complex and are derived from U. ciliata and U. acromelana respectively. Conversely only a norstictic- salazinic acid race is known in U. pseudocapillaris whilst three morphologically uniform chemical races occur in U. subcapillaris: (1) norstictic acid and salazinic acid, (2) squamatic acid hypothamnolic acid, (3) psoromic acid. Races 2 and 3 have a more limited distribution' and occasional thalli with mixed chemistry are known. The origin of race 2 is particularly intriguing since these substances do not occur in any other species in Neuropogon. Amongst the remaining species it is not so easy to pinpoint close relationships. There is often not sufficient evidence to indicate species pairs and also more asexual species exist than possible fertile, primary counterparts. Usnea sphacelata and U. perpusilla possibly form a species pair, based on the range of morphological and anatomical variation found in both species. Both species lack a depsidone-containing race, apart from the freak occurrence of psoromic acid. Specimens previously included in U. sphacelata, containing norstictic acid and/or salazinic acid, are here included in U. subantarctica; a species also separated on morphological and distribu- tional features. The distinction of U. subantarctica as a species from U. sphacelata might be considered to be rather tenuous, particularly in the depsidone-deficient phase. However, the rare occurrence of fertile material indicates other affinities which do not fall into the possible concept of an U. perpusilla-U. sphacelata species pair. The discovery of fertile U. sphacelata with the 'perpusilla' type of apothecia would finally settle this dilemma. The existence of a sorediate counterpart to Usnea trachycarpa remains uncertain, although two possible candidates exist: U. patagonica is often found in the same habitat in South America but has fewer fibrils, more or less delimited, blackened holdfast, is not known to contain depsidones (i.e. norstictic acid), but does contain the same or similar fatty acids as often occur in U. trachycarpa. The other, more likely, possibility is U. subantarctica which again is less fibrillate, more extensively pigmented, has ' trachycarpa' type of apothecia, often contains norstictic acid and/or salazinic acid, but lacks the fatty acids. It is possible that the sterile U. neuropogonoides may be derived from U. trachycarpa, but they do not represent a species pair. Lamb (1939) described the sorediate Usnea insularis from Marion Island and indicated that the species may have closest affinities to U. taylorii. However, U. insularis is here reduced to synonymy with U. antarctica on the basis of morphology, anatomy, and the presence of fumarprotocetraric acid, a depsidone that is extremely rare in U. taylorii. Further collections may reveal an asexual counterpart to U. taylorii. USNEA SUBGENUS NEUROPOGON 41 Generic concept and infrageneric classification The genus Usnea s. lat. has been divided into various genera or subgenera of which three, Neuropogon, Protousnea, and Lethariella, are frequently accepted as distinct genera. Other genera, for example Parmelia, have been subjected to a similar chequered existance and their delimitation may eventually rest on critical study of, for example, pycnidia, which are being found to be of value in delimitation of some orders, genera, and species (Krog, 1982; Hawksworth, pers. comm.). Meanwhile distinctions between neighbouring genera may remain somewhat obscure. This is particularly true of Neuropogon where some authors have accepted the group as a distinct genus, others as a subgenus of Usnea; two authors (Jatta, 1900, 1909; Lamb, 19390, 1964) changed their opinion. The main arguments have centered around whether thallus colour and pigmentation, disc colour, saxicolous habitat, and restricted distribution are sufficient criteria for separation at generic or subgeneric level. Previous authors have interpreted Neuropogon in various ways. Jatta's concept (1900) was based on two species that are now excluded from the subgenus. Du Rietz (1926) separated the subgenus from Usnea solely on the basis of apothecial disc colour and consequently had a much broader concept of Neuropogon than subsequent taxonomists. Eight out of the 13 species he included in his subgenus are now included in Lethariella and Protousnea (Krog, 1976). Motyka (1947) accepted Neuropogon as one of six subgenera of Usnea on the basis that there were not sufficient distinguishing characters to warrant generic status. Motyka (1936-38) further subdivided the subgenus into three sections: Sulphureae, Melaxanthae and Trachycarpae , grouping the species according to shared characteristics. All Motyka's species are still included within the subgenus today, although six taxa (five species and one variety) have subsequently been reduced to synonymy. Two species, Usnea durietzii and U. neuropogonoides , were described by Motyka (1936-38) and placed in sections Glabratae and Foveatae of the subgenus Euusnea respectively. These species, as pointed out by Motyka, have many Neuropogon-type features and are here included in the subgenus for the first time. Both Lamb (19390, 19480) and Motyka (1936-38) may be regarded as having a more conservative view of Neuropogon which will here be termed Neuropogon s. str. Dodge (1948) based his infrageneric classification on that of Motyka and accepted Neuropogon as a subgenus. In addition Dodge (1948, 1973) excluded five taxa from the subgenus and instead placed them in section Laevigatae, subsection Roccellinae of the subgenus Euusnea (i.e. Usnea s. str.) based on differences in the phycobiont. These taxa, all from the subantarctic islands of the southern Indian Ocean, were Usnea taylorii and four species that are here regarded as synonymous with U. antarctica. However, Dodge did consider these taxa to be somewhat intermediate between the two subdivisions and included them in the keys to both groups. A summary of the major taxonomic treatments is given in Table 5. Recently there has been a tendency by many lichenologists to accept at generic level subdivisions of established genera often based on detailed anatomical studies. For example, Brodo & Hawksworth (1977), using SEM showed that differences exist in the cortex structure in Alectoria s. lat. and used this as additional evidence to segregate genera. The SEM was used here to compare Usnea s. str. with Neuropogon and no anatomical differences were found. Results also compared favourably with previously published SEM work on Usnea s. str. (Lopez-Figueras & Palacios-Prii, 1981). Some examples of instances where Neuropogon has been given generic status include Krog (1976, 1982), Galloway (1983, 1984), and Rogers (1981). Krog (1976) compared Neuropogon with Protousnea and listed a series of characters that could be used to distinguish the two genera. This may be an acceptable separation since there are many differences, including surface ornamentation, medullary chemistry, and habitat. Krog (1982) accepted six usneoid genera: Neuropogon, Usnea, Protousnea, Evernia, Letharia, and Lethariella, using colour and pruinos- ity of the apothecial disc as distinguishing criteria. She also proposed a hypothetical relationship in which Protousnea is more closely related to Evernia than to Usnea and Neuropogon. Following Krog's classification the main difference between Usnea and Neuropogon lies in apothecial pigmentation and pruinosity. This is a much more acceptable distinction since, as will 42 F. J. WALKER Table 5 Examples of generic and infrageneric classification of Neuropogon . Source Classification A B Nees & Flotow (1835) Genus Neuropogon 2 1 Jatta (1900) Subgenus Neuropogon 2 Du Rietz (1926) Subgenus Neuropogon 13 5 Motyka (1936-38) Subgenus Neuropogon sect. Sulphur eae 2 2 sect. Melaxanthae 9 4 sect. Trachycarpae 1 1 Subgenus Euusnea sect. Foveatae 1 sect. Glabratae subsect. Pycnocladae 1 Lamb (1939a & 1948a) Subgenus Neuropogon 11 9* Lamb (1964) Subgenus Neuropogon 15 9 Dodge (1948 & 1973) Subgenus Neuropogon 25 5 Subgenus Euusnea sect. Laevigatae subsect. Roccellinae 5 2 Walker (present study) Subgenus Neuropogon 15 15 Column A represents the number of species included in Neuropogon by the given author. Column B represents the number of species accepted here; * = one variety raised to specific rank. Where dates are combined, numbers of species are totalled. become clear from this study, other criteria such as thallus colour and pigmentation (see below) and habitat may not be so reliable. However, even apothecial characters may not fit generic divisions. Krog (1976, 1982) gives examples of species included in Usneas. sir. which have flesh coloured to brown epruinose apothecial discs and therefore have close affinities to Protousnea, possibly representing an intermediate or primitive group. Consequently I prefer to follow the classification established by Henssen & Jahns (1973) and regard Neuropogon as a subgenus of Usnea. It has become apparent from this study that Neuropogon has often been loosely defined and frequently used as a term of convenience when applied, in particular, to antarctic taxa. If only species confined to the arctic and antarctic were considered (cf . Lamb, 1964; Dodge, 1973) there would probably be sufficient character differences to favour generic status - namely thallus colour and pigmentation, a saxicolous habitat, and limited distribution - thus conforming to Neuropogon s. str. However, when subantarctic and, particularly, South American taxa and allied species are studied it is clear that the delimitation between Neuropogon and Usnea is often rather obscure. Several South American taxa are included here in the subgenus because they share many characteristic features. Since apothecia in these are unknown at present, and hence disc colour and pruinosity, their true systematic position cannot be ascertained. The species that may be regarded as belonging to Neuropogon s. lat. are Usnea acanthella, U. durietzii, U. neuropogonoides, and U. patagonica. The high Andean species U. acanthella was included in Neuropogon by Lamb (19390) and described as a form of U. sphacelata (as U. sulphured). The variable extent of pigmentation and affinities to U. durietzii, combined with a wider substrate range, may indicate that U. acanthella is somewhat intermediate between the subgenera Usnea and Neuropogon. It is possible that the development of pigmentation may have arisen independently in different groups in response to harsh, exposed environments. A parallel may be found in the development of pigment in the genus Ramalina. R. tigrina is a terricolous species from the mist or fog zone of the Atacama desert in Chile (Follmann, 1966) and is characterised by a yellow-green thallus with prominent bands of black pigment. USNEA SUBGENUS NEUROPOGON 43 Development of pigment is also variable in the genus Alectoria; for example apices ofAlectoria sarmentosa ssp. sarmentosa may be occasionally striately blackened (Hawksworth, 1972). Usnea durietzii occurs throughout the Andean chain, occasionally occupying the same habitats as Neuropogon s. str. Thallus colour is the same green-yellow as U. acanthella and similarly thalli do not become as straminous or fuscous in herbaria as Neuropogon s. str. U. neuropogonoides is only known from a few aged collections and is a scrambling, sometimes terricolous, species from Patagonia. Thalli are sterile, and may have affinities with U. trachy car- pa. Similarly U. patagonica has limited pigmentation and shares common characteristics with U. antarctica and U. durietzii, the former species being accepted in Neuropogon s. str. It could also be argued that the two species, U. subantarctica and U. trachy carpa, which have red-brown to brown rather than black epruinose apothecial discs, might form a separate grouping or lie somewhere between Neuropogon s. str. and s. lat. U. trachy carpa is often scantily pigmented and has a subantarctic distribution whilst U. patagonica, a sorediate species that is rarely fertile, is more characteristic of Neuropogon s. str. and primarily has an antarctic distribution. If thallus pigmentation and saxicolous habitat were used as the main criteria for segregating Neuropogon many other species belonging to Usnea s. str. which exhibit some pigmentation would have to be included. However, a distinction has to be drawn at some point otherwise the delimitation of the group would become much more obscure. These species are excluded from Neuropogon s. lat. and, although occasionally sharing the same habitat, are either rarely pigmented or have pigmentation confined to apices or isidia. Three examples are given in Appendix I. Finally, thallus anatomy is sometimes used as a character for separating usneoid subgenera. The only Neuropogon species with a unique structure is Usnea taylorii in which the broad axis is invaded by medullary tissue, frequently resulting in sub-division or the formation of separate axial strands. In other species, for example U. antarctica and U. aurantiaco-atra, occasionally a small central cavity occurs in the axis without any significant penetration of medullary tissue or resultant segmentation. Dodge (1948) reported sub-divided axes in three Neuropogon species within the Roccellinae, although this only appears to be a constant feature of U. taylorii. Krog (1976) suggested that the axis of Neuropogon is most closely allied to that of Usnea subgenus Eumitria in which the axis is either tubular with a central cavity or is sometimes solid with a few longitudinal fissures (Swinscow & Krog, 1974). Krog (1976) also indicated similarities between the axes of Neuropogon and Lethariella subgenus Lethariella. The subgenus Neuropogon Usnea subgenus Neuropogon (Nees & Flotow, emend Nyl.) Jatta Sylloge lichenum italicorum: 54 (1900). - Neuropogon Nees & Flotow in Linnaea 9: 496 (1835). - Neuropogon Nees & Flotow, emend Nyl., Synopsis methodica lichenum 1: 272 (1860). Lectotype: Neuropogon antennarius Nees & Flotow (see Taxonomic review - p. 4). Description: Thallus fruticose, saxicolous, exceptionally terricolous or lignicolous, arising either from a small, delimited, or a broadly proliferating, basal holdfast, erect or subdecumbent, monopodial to subdichotomous or irregular, richly branched above, with or without short, lateral fibrils or numerous, attenuated secondary branches. Branches greenish yellow, becom- ing yellow or orange-brown on storage, variegated with violaceous black or black pigmenta- tion, especially towards the apices. Branches terete, or rarely angular, in transverse section. Surface smooth or sometimes becoming faveolate, shining-waxy or matt, pigmented-annulate or minutely scabrid, sometimes papillate to verrucose-rugose; rarely maculate. Cyphellae and pseudocyphellae absent. In section of three distinct zones; cortex, medulla, and axis. Cortex variable in thickness, (50)-100(-200) /am, of indistinct, strongly gelatinised pseudoparanchyma, sometimes forming a pallisade-like layer towards the exterior. Medulla compact, lax, or sublax, of interwoven unorientated hyphae. Photobiont: coccoid Chlorophyceae, forming a con- tinuous layer underlying the cortex. Axis chondroid, rigid, occupying 0-2-0-7(-0-9) of the branch diameter, cylindrical or slightly irregular, entire, very rarely sub-divided, of strongly 44 F. J. WALKER gelatinised longitudinal, fastigiate, paraplectenchymatous hyphae. Isidia rare; pseudoisidia and soredia often present. Apothecia lecanorine, subterminal or lateral, subsessile to geniculate, with or without a subtending spur. Disc black, rarely rufous brown, matt or subnitid, epruinose. Thalline excipulum concolorous with the thallus, smooth, faveolate, scabrid or verrucose- papillate, with or without numerous, marginal, pigmented rays. Thecium 60-75 //,m tall; epithecium aeruginose black or rarely brown, hymenium colourless below, blue-green or rarely brown above; hypothecium colourless to pale yellow. Asci clavate, c. 45 x 16 /xm, tholus amyloid (I + blue); ascospores 8 per ascus, simple, hyaline, ellipsoid, 7-10(-12) x 5-7 /u,m, thick-walled (c. 1 /Am). Paraphyses capitate, simple or branched, conglutinate. Pycnidia rare, pigmented, immersed in terminal branches; conidia sublageniform to filiform, 9-11 (-14) x l-l-7(-2) /am. Chemistry: depsides, depsidones, fatty acids, UV + unknowns, usnic acid. Distribution: polar alpine, predominately southern hemisphere, one species also northern hemisphere. Key to the species la Soredia or isidia absent, apothecia usually present 2 Ib Soredia or isidia present, apothecia rare 13 2a (la) Apothecia absent, thallus straggling or apices capillaceous, or decumbent. (For erect, sterile , juvenile thalli see 2b) 3 2b Apothecia present, thallus usually erect, tufted 5 3a (2a) Thallus prostrate, scrambling. Main branches ornamented, rarely entirely smooth, not waxy, without black-edged annulations. Pigmentation scant. Medullary chemistry various , fumarprotocetraric acid sometimes present . Patagonia , Antarctic islands 4 3b Thallus subdecumbent with numerous extended, capillaceous, pigmented secondary branches. Main branches smooth, waxy, with black-edged annulations. Medullary chemistry various, never containing fumarprotocetraric acid. Australasia 13. U. subcapillaris (p. 104) 4a (3a) Axis broad, greater than half branch diameter. Surface strongly verrucose or papillate. Medulla K + sordid brown, PD + red (fumarprotocetraric acid aggr.); K+ red, PD orange (norstictic and salazinic acids), or K , PD (no medullary substances). Antarctic islands, Falkland Is (abnormal form) 4. U. aurantiaco-atra (p. 62) 4b Axis thin, less than half branch diameter. Surface smooth, faveolate or minutely papillate. Medulla K-, PD+ yellow (psoromic acid); or K-, PD- ( fatty acids). Patagonia 7. U. neuropogonoides (p. 80) 5a (2b) Medulla lax, at least towards axis. Axis less than half branch diameter 6 5b Medulla compact. Axis half or more than half branch diameter 7 6a (5a) Apothecia lateral, rarely subterminal, subsessile, often in series; mature disc black, marginal excipular rays sparse (10-20) or absent. Surface smooth, waxy, rarely minutely papillate (x 10 lens); rarely with fibrils. Medulla K-, PD- (no medullary substances) ; or very rarely K - , PD + yellow (psoromic acid) . Patagonia 9. U. perpusilla (p. 85) 6b Apothecia subterminal; mature disc rufous brown; marginal excipular rays numerous (20+). Surface matt, faveolate to grossly papillate; with numerous fibrils. Medulla K+ red, PD orange (norstictic and salazinic acids); K-, PD+ yellow (psoromic acid) ; or K- , PD - ( fatty acids) . Kerguelia, Patagonia, Falkland Is 15. U. trachycarpa (p. 110) 7a (5b) Axis thick, partially sub-divided, often protruding through the cortex as pale maculae. Medulla reduced, penetrating axial cavities. K , PD (no medullary substances), very rarely K-,PD+ red (fumarprotocetraric acid). Kerguelia 14. U. taylorii(p. 108) 7b Axis entire, rarely with sinuose outline, not penetrated by the medulla. Medullary chemistry various 8 8a (7b) Surface smooth, subfaveolate, fibrils absent 9 8b Surface papillate or verrucose ; often with fibrils 12 USNEA SUBGENUS NEUROPOGON 45 9a (8a) Black-edged annulations frequent on main branches. Excipulum smooth with conspic- uous marginal rays. Australasia 10 9b Black-edged annulations normally absent. Excipulum faveolate or rarely papillate, marginal rays sparse or absent. Patagonia, Antarctic peninsula and islands 11 lOa (9a) Thallus entirely erect. Apothecia subterminal, rarely lateral. Medulla K+ red, PD orange (norstictic and salazinic acids) 5. U. ciliata (p. 74) lOb Thallus subdecumbent with numerous extended, capillaceous, secondary branches. Apothecia lateral. Medulla K+ red, PD orange (norstictic and salazinic acids, protocetraric acid); UV + -I- blue, K purple (squamatic, hypothamnolic acids); or K- , PD -I- yellow (psoromic acid) 13. U. subcapillaris (p. 104) lla (9b) Apothecia lateral, often in series; rarely subterminal, excipulum faveolate. Medulla K-, PD- (no medullary substances); rarely K-, PD + yellow (psoromic acid). Patagonia (abnormal form) 9. U. perpusilla (p. 85) lib Apothecia subterminal, rarely lateral, excipulum smooth to minutely papillate or verrucose. Medulla K -I- red, PD orange (norstictic and salazinic acids); K + sordid brown, PD + red (fumarprotocetraric acid aggr.); or K , PD (no medullary substances). Patagonia, Falkland Is., Antarctic peninsula and islands, including Bouvet0y (abnormal form) 4. U. aurantiaco-atra (p. 62) 12a (8b) Mature apothecial disc rufous brown, marginal excipular rays numerous (20+). Fibrils numerous on main branches. Medulla K -I- red, PD orange (norstictic and salazinic acids); K-, PD + yellow (psoromic acid); or K-, PD- ( fatty acids). Kerguelia, Patagonia, Falkland Is (abnormal form) 15. U. trachycarpa (p. 110) 12b Mature apothecial disc black (rarely brown when immature), marginal excipular rays sparse (10-20) or absent. Fibrils rare on main branches. Medulla K+ red, PD- orange (norstictic and salazinic acids); K+ sordid brown, PD+ red (fumarpro- tocetraric acid aggr.) or K , PD (no medullary substances). Patagonia, Falkland Is., Antarctic peninsula and islands, including Bouvet0y 4. U. aurantiaco-atra (p. 62) 13a (Ib) Thallus with true isidia 14 13b Thallus with soredia and/or pseudoisidia 17 14a(13a) Thallus yellow or yellow green; pigmented; rarely with fibrils. Pseudocyphellae absent or inconspicuous. Norstictic and salazinic acids absent 15 14b Thallus grey-green, with numerous fibrils. Pigmented isidia arising from prominent pseudocyphellae. Norstictic or salazinic acids usually present. South America [U. amblyoclada(p. 115)] 15a(14a) Medulla lax throughout. Axis less than half branch diameter. Surface smooth to subpapillate; faveolate to inflated 16 15b Medulla compact, rarely sublax. Axis more than half branch diameter. Surface waxy, epapillate, not inflated. Isidia delimited, frequently pigmented and regenerating to form pseudoisidia which may erode. Australasia [U. torulosa (p. 117)] 16a(15a) Thallus corticolous, rarely saxicolous, unpigmented, rarely subpapillate. Isidia scat- tered , often pigmented . Medulla K - , PD + yellow , U V - (psoromic acid) ; K - , PD - , UV+ + blue (squamatic acid); rarely K- , PD- , UV- ( fatty acid). Australasia [U. inermis(p. 117)] 16b Thallus saxicolous, rarely terricolous, often pigmented, subpapillate. Isidia unpig- mented, arising from distinct tubercules. Medulla K-, PD- (no medullary sub- stances). Peru, Ecuador, Bolivia (high altitudes) 1. U. acanthella (p. 47) 17a(13a) Medulla lax in main branches. Axis half or less than half branch diameter 18 17b Medulla compact in main branches . Axis more than half branch diameter 23 18a(17a) Main branches extensively inflated or articulate 18b Main branches only faveolate or slightly expanded 20 19a(18a) Thallus truly sorediate, extensively pigmented or variegated towards apices. Medulla K , PD (no medullary substances). Peru, Ecuador, Bolivia. High altitudes (abnor- malform) 11. U. sphacelata (p. 92) 19b Thallus pseudoisidiate, pigmentation confined to pseudoisidia, thallus base and branch 46 F. J. WALKER apices. Medulla K+ red, PD orange (norstictic and salazinic acids); rarely K , PD (no medullary substances) . South America 6. U. durietzii (p . 78) 20a(18b) Thallus with fibrils or grossly papillate, arising from a proliferating holdfast. Soredia present, pseudoisidia rare. Medulla K+ red, PD orange (norstictic and salazinic acids) ; or K - , PD - (no medullary substances) . Patagonia , Antarctic peninsula 12. U. subantarctica (p. 99) 20b Thallus smooth to subpapillate ( x 10 lens) , arising from a delimited, or rarely proliferat- ing holdfast. Soredia or pseudoisidia present. Medulla K , PD (no medullary substances or fatty acids); rarely K , PD+ yellow (psoromic acid); or K+ red, PD orange (norstictic and salazinic acids) 21 21a(20b) Soralia delimited, often globose and pigmented. Thallus extensively pigmented. Medullary chemistry various 22 21b Soralia eroded, effuse to ulcerose, with small blackened pseudoisidia. Thallus pigment confined to base and apices. Medulla K , PD (fatty acids). Patagonia, ?South Africa 8. U. patagonica (p. 82) 22a(21a) Medulla containing norstictic acid (K + red, PD orange) Patagonia (abnormal form) 12. U. subantarctica (p. 99) 22b Medulla depsidone deficient (K- , PD - ) or rarely containing psoromic acid (K- , PD + yellow). Bipolar 11. U. sphacelata (p. 92) 23a(17b) Thallus epapillate, waxy, black-edged annulations present or absent 24 23b Thallus papillate , subpapillate or with fibrils ; rarely waxy or annulate 26 24a(23a) Thallus bright yellow, torulose, lacking annulations. Pseudoisidia pigmented; regener- ating from true isidia. Medullary chemistry various but never containing norstictic and salazinic acids. Australasia [U. torulosa (p. 117)] 24b Thallus yellow-green (straminous in herbaria), monopodial to richly branched, usually at least with black-edged annulations. Medullary chemistry various, often containing norstictic and salazinic acids 25 25a(24b) Thallus subdecumbent, richly branched from a delimited holdfast, with numerous, fragile, capillaceous, divergent ultimate branches. Soralia pale, punctiform. Medulla K+ red, PD orange (norstictic and salazinic acids) . New Zealand 10. U. pseudocapillaris (p. 89) 25b Thallus erect, monopodial or moderately branched, often from a proliferating holdfast. Soralia plane, pale, rarely nodular, corticate then pigmented; pseudoisidia rare. Medulla K+ red, PD+ orange (norstictic and salazinic acids); K-, PD- (no medullary substances) (rare); or K , PD + yellow (psoromic acid) (rare). Australa- sia, Patagonia; Antarctic peninsula and islands (rare). [If lacking annulations and from Antarctica see also U. sphacelata (p. 92) 2. U. acromelana (p. 48) 26a(23b) Thallus monopodial or moderately branched arising from a proliferating holdfast; extensively pigmented 27 26b Thallus subdichotomously to richly branched arising from a delimited holdfast; extent of pigmentation variable 28 27a(26a) Surface matt, with fibrils or grossly papillate. Medulla K+ red, PD+ orange (norstictic and salazinic acids), or K- PD- (no medullary substances). Patagonia, Antarctic peninsula 12. U. subantarctica (p. 99) 27b Surface waxy, subpapillate. Medulla K , PD , (no medullary substances). Antarctica (abnormal form) 11. U. sphacelata (p. 92) 28a(26b) Soralia plane, crateriform; minute pseudoisidia rare. Thallus grossly papillate; usually matt; annulations absent. Medulla compact; K , PD (no medullary substances); or K+ sordid brown, PD + red (fumarprotocetraric acid); rarely K-l- red, PD orange (norstictic and salazinic acids - Falkland Is.). Antarctic continent, subantarctic islands , New Zealand (rare) , Patagonia (rare) 3. U. antarctica (p . 55) 28b Soralia plane, eroded, ulcerose; minute pigmented pseudoisidia frequent. Thallus smooth, waxy to subpapillate; occasionally annulate. Medulla sometimes lax in part; K - , PD - (fatty acids) Patagonia , ? South Africa 8. U . patagonica (p . 82) USNEA SUBGENUS NEUROPOGON 47 The species 1. Usnea acanthella (Lamb) F. J. Walker, comb. nov. Fig. 9 Neuropogon sulphureus f. acanthella Lamb in/. Linn. Soc. (Bot.) 52: 210 (1939). Type: Peru, [Puno], Carabaya, above Limbani, 4410 m, 18 November 1937, Dora Stafford 1 108 (BM! _ holotype; BM! , FH! - isotypes). [TLC: no medullary substances, traces of unidentified fatty acids, usnic acid.] Description: Thallus 2-4 cm, arising from a delimited, blackened, elongated, stalked holdfast, erect or rarely subdecumbent, subdichotomous to richly, divergently branched above, without fibrils, with loosely interwoven, capillaceous, spinulose secondary branches. Branches terete, greenish yellow, with broad bands of black pigment, inflated, weakly articulated. Cortex thin. Surface faveolate, matt, subpapillate. Medulla extremely lax, axis thin, occupying 0-2-0-3 of the branch diameter. Isidia present, fine, spinulose, unpigmented, c. 0-5 mm long, frequently eroding, arising in delimited clusters from partially corticate tubercules. Pseudoisidia and soredia absent. Apothecia and pycnidia not seen. TLC: no medullary substances, traces of unidentified fatty acids, usnic acid. Fig. 9 Usnea acanthella. Holotype of Usnea suiphurea f. acanthella Lamb (BM). Top. Whole thallus (xl-5). Bottom. Detail of isidia (xlO). 48 F. J. WALKER Distinguishing features: Usnea acanthella is characterised by its erect, spreading habit from a basal stalk and a richly branched, spinulose thallus, an inflated, subpapillate surface, a lax medulla lacking depsides and depsidones, a thin axis, and clusters of fine, unpigmented, true isidia. Distribution: Usnea acanthella is apparently restricted to saxicolous, or rarely musicolous- terricolous habitats, in open paramo at high altitudes in the northern part of the Andean chain. The species has been recorded from between 3500-4500 m in Peru, Ecuador, and Bolivia. Fig. 7. Chemistry: Only known in a depsidone deficient phase, with the occasional occurrence of traces of unidentified fatty acids. Variation: From the small number of collections studied, Usnea acanthella appears to be a constant species with little variation. The isidia often erode or appear nodular, but even when they are abraded or underdeveloped, the thallus is characterised by the prominent, pale tubercules from which they arise. Occasionally small fibrils may develop secondarily from such tubercules. There is some variation in the habit of the species, ranging from thalli with prominent, inflated main branches and relatively few ultimate branches, to richly branched, less inflated forms with extensively interwoven branches. Although the type material is uniform in its variegation, frequently with broad bands of pigmentation, other gatherings indicate that this may not necessarily be as constant a feature of this species. In recent collections from Ecuador some parts of the thalli are unpigmented, whilst other areas are totally blackened. Such presence or absence is reminiscent of the kind of blackening found in some corticolous species belonging to the subgenus Usnea, for example U. inermis (p. 117), where blackening only occurs in extremely exposed situations or when the thallus is moribund. However, the presence of some distinctly variegated thalli indicates that pigmentation is probably the norm for the species, although a wider range of material is required to verify its extent. Species concept: Lamb (1939a) described this taxon as a form of Usnea sphacelata (as U. sulphured), interpreting the pale isidia as outgrowths from healed-over soralia rather than recognising their true nature. The development of true isidia in this species widens the range of asexual propagules found in the subgenus and their development is a sufficiently distinct feature, combined with the characteristic habit of the thallus, to warrant separation at species level. The species superficially is very similar to Usnea durietzii and U. sphacelata, sharing the common features of a lax medulla and normally saxicolous habitat. The development of a basal stalk below and branching are very reminiscent of U. durietzii, although the branches in U. acanthella are generally finer and more fragile. U. acanthella may be distinguished from all other asexual species of the subgenus by the presence of true isidia. Specimens examined ECUADOR. Chimborazo: Cerro Payacorral, near El Altar, 3450 m, 2 August 1980, Cambridge Exped. 2 (BM); Mt. Achipungo, near Osogochi, 4500 m, 7 September 1980, Cambridge Exped. 42 (BM); near El Altar, near Riobamba, 4150 m, 3 August 1980, Cambridge Exped. 5 p.p. (BM). PERU. Cuzco: Huaya Pass, above Cuzco, 4500 m, 14 June 1973, B. Mullins 29 (BM). Puno: prope Azangaro, June 1854, W. Lechler PL Peruvianae 1758 p.p. (BM); Cordillera de Carabaya, Limbani (type locality). BOLIVIA. La Paz: Larecaja, Sorata, 4500 m, April 1897, G. Mandon 1737 p.p. (PC). 2. Usnea acromelana Stirton Figs 10-12 in Trans. Proc. N.Z. Inst. 30: 388 (1898). -Neuropogon acromelanus (Stirton) Lamb in/. Linn. Soc. (Bot.) 52: 218 (1939). Type: New Zealand, Selwyn Gorge, May 1894, T. W. N. BeckettLll, 'on trees' (thalli saxicolous) (BM! - holotype; CANL 16943! , CHR 343153! - isotypes; GLAM NHB 1927-8-562! - ? isotype). [TLC: norstictic acid, salazinic acid, usnic acid.] Neuropogon acromelanus var. decipiens Lamb in/. Linn. Soc. (Bot.) 52: 219 (1939). - Usnea acromelana var. decipiens (Lamb) Lamb in Br. Antarct. Surv. Sci. Rep. 38: 5 (1964). Type: Tasmania, summit of Table Mountain [= Mt. Wellington (Wilson, 1893)], R. Brown her Australiense, 1802-05, 523 (BM! - holotype; BM! - isotype). [TLC: norstictic acid, salazinic acid, usnic acid.] USNEA SUBGENUS NEUROPOGON 49 Fig. 10 Usnea acromelana. A Convex-globose soralia xlO (holotype of Neuropogon acromelanus var. decipiens Lamb, BM). B, C Fertile thalli. B Antarctic peninsula, Joinville 1., Smith 2680 (AAS) xl-5. C New Zealand, Otago, Old Man Range, James 1597 (BM) x 10. Description: Thallus (1 -5-2-4(-6) cm, arising from a broadly proliferating or rarely delimited, pigmented holdfast, usually erect, monopodial or subdichotomous, moderately branched above, usually lacking fibrils. Branches terete, yellow-green, continuously pigmented violaceous black towards the apices. Cortex thick. Surface smooth or rarely subfaveolate, waxy, 50 F. J. WALKER epapillate, with conspicuous black-edged annulations on main branches. Medulla compact, axis thick, occupying c. 0-5 of the branch diameter. Soralia numerous on primary and secondary branches, plane and discrete, sometimes becoming confluent, rarely convex-globose and pulverulent. Soredia granular, unpigmented, or partially corticate then pigmented. Pseudoisi- dia rare, isidia absent. Apothecia rare, subterminal, as in U. ciliata. Pycnidia not seen. TLC: (1) norstictic acid, salazinic acid, protocetraric acid, usnic acid; (2) psoromic acid, 2'-O- demethylpsoromic acid, usnic acid; (3) no medullary substances, usnic acid. Distinguishing features: Usnea acromelana is characterised by its erect, proliferating habit, and a monopodial to moderately branched thallus, often with violaceous black pigmentation, a smooth, waxy, black-annulate surface, a compact medulla usually containing norstictic and salazinic acids, a thick axis, and numerous emarginate soralia. Distribution: In Australasia Usnea acromelana occurs in the alpine regions of New Zealand and is mainly confined to the Southern Alps apart from isolated records from the North Island, Stewart Island, and Chatham Island; it is rather rare in Tasmania and is known only from a single locality in Victoria. The species is rarely found below c. 400 m, usually occurring between 900-1800 m, and is rare at higher altitudes, c. 2500 m, occasionally being replaced by U. sphacelata. Usnea acromelana is probably less frequent in southern South America (Patagonia) and the Antarctic peninsula than previously indicated (e.g. Follmann, 1965a; Lamb, 19480). For example, specimens from central Chile are referable to U. patagonica, although the northern limit of U. acromelana is still uncertain. Lamb (19390) reported a single gathering of U. acromelana from Peru. This was based on a mixed collection (BM!) of U. acanthella and a norstictic acid-containing thallus that appears to be U. durietzii. However, Dennis (I960) reported affinities between the fungus flora of South America and Australasia, finding the same species in Tasmania and in the paramos of western Venezuela. Consequently the possible occurrence of U. acromelana in the Andes north of Patagonia cannot be completely dismissed. Usnea acromelana is rare on the Antarctic peninsula and is probably confined to the northern tip and the western coast, where material has been examined from Adelaide, Brabant, Joinville and Wiencke Islands, frequently mixed in gatherings with Usnea antarctica, U. aurantiaco-atra, and U. subantarctica. The species is also tentatively identified from a few depauperate specimens from the South Orkney Islands, where U. antarctica is the dominant species. The species is not known from continental Antarctica and previous reports (Filson, 1975; Rudolph, 1963, 1966) refer to U. sphacelata, including the infraspecific taxa described by Lamb (19390). Fig. 6. Chemistry: Thalli containing norstictic acid and salazinic acid (Race 1) are by far the most frequent of the three chemical races recorded. In this race a trace amount of protocetraric acid is often present, but probably not always in sufficient quantity to enable detection by TLC. This is the most frequent race found in Australasia and is also the most widespread in the South American sector, being previously reported from Patagonia by Asahina (1967), and extends south to the Antarctic peninsula. The occurrence of psoromic acid (Race 2) is rare and unpredictable, and is known from isolated locations in Argentina and Chile, the former collection also including depsidone deficient thalli (Race 3). It is possible that there is a correlation between medullary substances and distribution, as Race 3 is chiefly confined to more southern latitudes; the only exception being the isolated record from Victoria, Australia. A parallel case occurs in Usnea antarctica where depsidone- containing thalli are only frequent towards the northern part of the distribution of that species. Variation: The above description is characteristic of the majority of specimens examined and is typical of much New Zealand material and, to a lesser extent, Tasmanian and South American- antarctic populations where a greater range of thallus morphology is often encountered. The greatest range of variation is found in branching, extent of pigmentation and form of the soralia. The extent of branching ranges from monopodial, with numerous primary branches arising from a proliferating holdfast, to richly branched, rarely from a solitary point of attachment. USNEA SUBGENUS NEUROPOGON 51 Fig. 11 Holotype of Usnea acromelana Stirton (BM). Top. xl. Bottom. Detail of cortical annulations and plane soralia x 10. 52 F. J. WALKER Study of a large number of collections, particularly from NewZealand (CHR), indicates that altitude may influence thallus morphology, since less characteristic thalli often occur at lower altitudes. For example, the type material of Usnea acromelana, from c. 600 m, includes two of the extreme forms encountered. Plants forming the holotype (BM) (Fig. 11) and one isotype (CANL) are infrequently branched, with only traces of pigmentation at the apices, lacking the waxy, violaceous black lustre, having inconspicuous annulations, and large (0-4-0-5 mm), plane soralia that are rarely pigmented-corticate. In contrast, isotype material in CHR (Fig. 12) is less robust, more richly branched, with scattered fibrils, fine, attenuate, pigmented secondary branches, prominent black-edged annulations, smaller (0-1-0-3 mm), unpigmented soralia, and a less densely interwoven medulla. Scantily pigmented forms are rare at higher altitudes where thalli tend to be smaller, more richly branched and more extensively pigmented, possibly as a response to greater exposure. In extreme instances extensive cracking, resulting from marked annulation formation, may give a slightly faveolate appearance. In typical specimens the soralia are pale, plane, or slightly excavate, sometimes with a tendency to become convex or slightly confluent, characteristically with a small amount of blackening inside resulting from overlying cortical fragments which, in very extreme instances, may form minute pseudoisidia, or, when damaged, produce spinules. Less typically, soralia may be small and plane, resembling those of Usnea pseudocapillaris , or can be large, extensively blackened, convex and globular to nodulose, (Fig. 10) as in the variety decipiens (Lamb, 19390). A range of soralia forms may be found on a single thallus. There appears to be a much wider range of variation in Tasmanian populations of Usnea acromelana than in New Zealand, particularly in soralia type and branching pattern; some specimens may lack annulations. Such unusual forms are occasionally found in other parts of the range of this species. In Tasmanian material two main types may be distinguished, but neither is here considered worthy of taxonomic status. The first conforms to variety decipiens and is small (2-4 cm), erect, richly branched, extensively pigmented, with large (0-5-1-5 mm), nodular, blackened soralia with compacted soredia. The second type is erect to subdecumbent, more divergently branched, and could be regarded as an intermediate between U. acromelana and U. pseudocapillaris. However this is considered to be a form of U. acromelana as secondary branches are notably coarser than those of U. pseudocapillaris. Ultimate branches are shortly attenuate and capillaceous, giving the thallus a slightly tasselled and interwoven appearance; the soralia are small, plane to punctiform, rarely pigmented, and confined to secondary branches. The thalli lack the segmented appearance, the extended, entangled secondary branches, and laxer medulla, characters which distinguish U. pseudocapillaris. In some respects South American and antarctic populations of Usnea acromelantfappear to be slightly different from those in Australasia. Thalli may frequently be more robust or more richly branched; short incipient fibrils are rare and are unrelated to papillae. In addition such thalli are usually extensively pigmented, including some banded or mottled variegation which is not generally a feature of this species. As in Australasian populations the extent of black-edged annulations is variable. Often folding and puckering of the surface may be more pronounced, becoming faveolate-ridged rather than simply annulate, which might lead to misidentification. Thalli still retain a violaceous tinge to the pigmentation and a waxy lustre, whilst soralia exhibit the same range of forms encountered in Tasmanian populations. Apothecia are rarely produced (Fig. 10) and are only known from a few localities, for example, from the Old Man Range, Otago, New Zealand (P. W. James 1597, BM!) and Joinville Island, Antarctic peninsula (R. I. L. 5m/r/z3680p.p.,AAS!). The New Zealand specimens bear subterminal or rarely lateral apothecia with a rayed excipulum, which frequently bear soralia, and a buff-grey to greenish black disc. Although sometimes lacking pigment, the disc colour is distinct from that of Usnea trachycarpa which is rufous brown. The Antarctic peninsula specimen belongs to Race 1 and bears a single apothecium, c. 2 mm diameter, with a black disc but lacks excipular rays. Lamb (19480) described a fertile specimen of this species from Wiencke Island, Antarctic peninsula. This has been examined (Lamb 1790 p.p. BM!) and is similar to the Joinville Island specimen. USNEA SUBGENUS NEUROPOGON 53 DTANY DIVIS Fig. 12 Isotype of Lfrnea acromelana Stirton (CHR) x 1 . Species concept: The diverse range of forms of Usnea acromelana that occur might appear to be distinct taxa if it were not for the existence of a range of intermediates. Initially, study of the varied type material of U. acromelana caused considerable speculation as to the delimitation of this species, however, these extremes are here considered to fall within the variation of a single species. This is particularly true of Tasmanian populations, and it is possible that the species is diverging into a range of entities, some of which already differ from New Zealand populations. Some of these forms may appear quite distinct when observed individually but en masse form a continuum. The diversity of chemical races in South American populations may indicate possible affinities with other species occurring in that sector. Such specimens are here referred to Usnea acromelana, despite some morphological ambiguities which are not sufficient to warrant recognition of separate taxa. Further collections and ecological data are essential to resolve this problem. Usnea acromelana is most closely related to U. pseudocapillaris and both belong to the U. ciliata complex. As frequently occurs within the subgenus, there are a few specific instances where it is difficult to conclusively separate the two species. However, study of populations indicates that there are sufficient distinguishing and constant characters. U. pseudocapillaris can normally easily be identified (p. 90) on morphological and chemical characters. On rare occasions in Australasia Usnea acromelana may be confused with Usnea torulosa 54 F. J. WALKER (Appendix I, p. 117). The two species may be separated by the twisted, entwined habit of U. torulosa, its brighter yellow colouration in the field (although this may be less obvious in stored material), lack of surface pigmentation, the presence of pigmented, true isidia, and by differences in depsidone content. The type of Neuropogon acromelanus var. inactivus (Lamb, 1939a) from Tasmania is U. torulosa. Vsnea acromelana can be distinguished from U. antarctica by the waxy surface and absence of papillae, and margin to the soralia; from U. durietzii and U. patagonica by differences in habit and absence of pseudoisidia; and from U. sphacelata and U. subantarctica by a compact medulla and smooth surface. Selected specimens examined Racel CHILE. Magallanes: Seno Skyring, Estancia Maria, near sea-shore, 28 April 1940, R. Santesson 7058 (S, UPS); Tierra del Fuego, [5432'S: 7004'W], 730 m, 24 February 1978, /. R. Peart s.n. (BM); Tierra del Fuego , Sierra Sorondo , N . slope , above Las Cotorras (c. 20 km ENE . of Ushuaia) , 800 m , 6 February 1 940 , R. Santesson 641d (S); Tierre del Fuego, Sierra Alvear, S. slope, above Las Cotorras (c. 20 km ENE. of Ushuaia), 700 m, 9 February 1940, R. Santesson 637d (S); Isla Navarino, Puerto Navarino, 10 m, 28 February 1940, R. Santesson 1232b (S); Canal Beagle, Yendegaia, by front of glacier, 50-100 m, 4 March 1940, R. Santesson 1373 (S), 1373 p.p. (UPS). ARGENTINA. Santa Cruz: Lago Argentine, near front of Ventrisquero Mayo, 105 m, March 1959, P. W. James 5053 (BM); Lago Argentino, Cerro Mayo, c. 1500 m, 25 February 1959, P. W. James 3605 (BM, S, UPS, US); Lago Argentino, Peninsula Magallanes, opposite Spegazzini Glacier, Ventrisquero Spegazzini, January 1959, P. W. James s.n. (BM); Calafate, 1959, P. W. James s.n. (BM); Lago San Martin, 1600 m, 2 February 1933, A. Donatlp.p. (H). Tierra del Fuego: Parque National Tierra del Fuego, Ushuaia, 'Weg zum' Glaciar Martial, c. 500 m, 7 December 1973, A. Henssen & G. Vobis 24417i p.p. (MB). SOUTH ORKNEY IS. Coronation I.: Wave Peak Buttress, 225 m, 9 September 1973, T. N. Hooker 84 (AAS), T. N. Hooker 87 (AAS), T. N. Hooker 116 (AAS); Sunshine Glacier, 180 m, 12 March 1972, T. N. Hooker 66 (AAS). ANTARCTIC PENINSULA. Graham Land: Joinville I. [6315'S: 5545'W], Mt. Alexander, 200-500 m, 27 March 1981, R. I. L. Smith 2680 p.p. (fertile) (AAS); Mt. Cardinall [6327'S: 5710'W], 225 m, 14 December 1946, /. D. Andrew F.I.D.S. D115-125 (BM); Gerlache Channel, Brabant I. [6415'S: 6220'W], Exped. Antarct. Beige 249 p.p. (BM ex herb. Vainio 354, 355); Palmer Archipelago, Wiencke I. [6448'S: 6325'W], N. buttress of Noble Peak, 225 m, 20 November 1944, /. M. Lamb 1790 p.p. (fertile) (BM); Loubet Coast, Adelaide I. , Rothera Point [6735'S: 6800'W], 10 m, 29 January 1976, /. Fenton 23 (AAS). AUSTRALIA. Tasmania: Central Southern Tasmania, Mt. Mawson, 1220 m, 4 December 1965, G. C. Bratt 2931a (CHR 343400); Mt. Mawson, southern peak, 1220 m, 11 May 1968, G. C. Bratt 68/348 [Lich. Exs. COLO 247] (BM, CHR 343460); Mt. Mawson, 19 February 1968, R. Filson 10580 (MEL 1029344); Lake Augusta, outcrop c. 0-5 km E. of dam, 840 m, 2 March 1970, G. C. Bratt & M. H. Bratt Bratt 70/243 (HO 35238), Bratt 70/229 (HO 35237); Mt. Wellington Plateau, 11 January 1969, G. C. Bratt el al. Bratt 69/7 (HO 35193), summit area, 1270 m, 20 December 1964, G. C. Bratt 1872b (HO 35188), 1850, 5. Mossman 796 (BM, E); Table Mtn, 1095 m, 18 June 1972, G. C. Bratt &J. A. Cashin Bratt 72/393 p.p. (HO 35179); Mt. Penny, 4 April 1969, G. C. Bratt & K. M. Mackay Bratt 69/162a&b (HO 35240). NEW ZEALAND. North Island. Wellington: Tongariro National Park, Tama Lakes, 1220 m, D. Scott 504 (OTA): unrealised: Colenso C1776 (BM, WELT). South Island. Nelson: Cobb Valley, Cobb Ridge, 850 m, 19 December 1982, /. K. Bartlett 26252 (herb. Bartlett, BM); Mt. Technical, above Lewis Pass, 14 January 1979, D. J. Galloway s.n. (CHR 343160 p.p.); Mt. Arthur, March 1879, A. Mckay s.n. (WELT L1505). Maryborough: Inland Kaikoura Range, Mt. Tapuaenuku, 20 March 1934,7. 5. Thomson 1523 p.p. (CHR 343806 p.p.), 2400 m, /. 5. Thomson 1506 p.p. (CHR 343791 p.p.); Branch River, Range above Gordon Stream, c. 900-1200 m, 23 January 1984, /. K. Bartlett 264185 p.p. (herb. Bartlett), 26419 p.p. (herb. Bartlett). Canterbury: Banks Peninsula, Mt. Sinclair, 3 February 1979, D. J. Galloway s.n. (CHR 343147); Ben Ohau Range, Glen Lyon Station, 1830 m, October 1958, /. Murray Mason 29 & 30 (BM, OTA, S, UPS); Kirkliston Range, 1520-1680 m, 25 March 1978, D. J. Galloway s.n. (CHR 343230); Four Peaks Range, Blue Mtn, c. 1640 m, 23 April 1979, D. J. Galloway s.n. (CHR 343224); Torlesse Range, Foggy Peak, 1680 m, November 1972, D. J. Galloway s.n. (CHR 343430). Otago: Humboldt Mtns, Mt. Minos, 2010 m, 1 January 1970, D. J. Galloway s.n. (CHR 342824); Rock and Pillar Range, 1160-1190 m, 18 September 1981, F. J. Walker s.n. (BM, UPS); Dunedin, Maungatua, 880 m, 17 September 1981, F. J. Walker s.n. (BM, UPS); Dunedin, Middlemarch-Mosgiel road, nr. Sutton, c. 370 m, 18 September 1981, F. J. Walker s.n. (BM); Old Man Range, 1220 m, 1 February 1963, P. W. James 1597 (BM), 1579 (BM); USNEA SUBGENUS NEUROPOGON 55 Matukituki Valley, Mt. Avalanche, 2560-2590 m, 15 February 1969, L. D. Kennedy s.n. (CHR 343382), Mt. Sir William, 2560 m, 27 December 1970, D. J. Galloway s.n. (CHR 342786 p.p.); Shepherd's Pass, 1980 m, January 1968, D. J. Galloway s.n. (fertile) (CHR 343411); Mt. Earnslaw, 2700 m, February 1972, A. A. Deans s.n. (CHR 342757); Remarkables, above Lake Alta, 1980 m, March 1967, L. D. Kennedy s.n. (fertile) (CHR 343440). Southland: Mid Dome, 1520 m, May 1970, G. Van Reenen s.n. p.p. (CHR 342837 p.p.); Mt. Barrier, Homer Tunnel, March 1959, M. A. Chapman Murray 4180 (OTA). Stewart I.: Mt. Anglem, summit, 984 m, February 1966, D. J. Galloway s.n. (CHR 343356, CHR 343417, CHR 343389). Chatham Is: Chatham I., Trovers s.n. (BM). Race 2 CHILE. Aisen: 'in valle superiore fluminis Aysen', c. 1400 m, 23 February 1897, P. Dusen s.n. (UPS). ARGENTINA. Santa Cruz: Lago Argentine, Cerro Norte, c. 800 m, 22 December 1958, P. W. James 404 p.p. (BM). Race 3 CHILE. Magallanes: Tierra del Fuego, Isla Navarino, 300 m, 1963, G. Follmann 14586 p.p. (M). ARGENTINA. Santa Cruz: Lago Argentine, Glaciar Moreno, 22 January 1967, P. R. San Martin s.n. (FH); Lago Argentine, Parque Nacional Glaciares, Peninsula Magallanes, 17 December 1973, A. Henssen & G. Vobis 24516a (MB); opposite Spegazzini Glacier, P. W. James s.n. p.p. [as Race 1] (BM); Lago Argentine, Bilbao, 2-3 February 1914, C. M. Hicken 16450 (BM); Cerro Norte, P. W. James s.n. p.p. [as Race 2] (BM). Unrealised: South Patagonia, 1900-01, comm. H. Prichard (BM). SOUTH ORKNEY IS: Coronation I., Wave Peak Buttress, 225 m, 9 September 1973, T. N. Hooker 85 (AAS), T. N. Hooker 86 (AAS), T. N. Hooker 115 (AAS), T. N. Hooker 111 (A AS), T. N. Hooker 118 (AAS). AUSTRALIA. Victoria: Basalt Hill, Bogong High Plains, [c. 2000 m], 29 January 1967, A. C. Beauglehole s.n. (MEL 18755), 22 January 1967, R. B. Filson 9504 (MEL 1018193). For further localities of Race 1 in Canterbury and Otago see lists held in BM and collections in BM, CHR, HO and OTA. 3. Usnea antarctica Du Rietz Figs 13-14 in Svensk hot. Tidskr. 20: 93 (1926). - Neuropogon antarcticus (Du Rietz) Lamb in/. Linn. Soc. (Bot.) 52: 210 (1939). Type: Regio Antarctica, South Victoria Land, Admiralty Range, 2000' s.m., 1900, C. R. Borchgrevink (UPS! - holotype; S!, O! -isotypes). [TLC: no medullary substances; UV+ unknowns, usnic acid.] Neuropogon melaxanthus f. sorediifer Crombie in/. Linn. Soc. (Bot.) 15: 182 (1876). - Usnea melaxantha var. sorediifera (Crombie) Mull. Arg. in /. Linn. Soc. (Bot.) 32: 200 (1896). - Usnea sulphurea var. sorediifera (Crombie) Vainio, Res. Voy. S. Y. Belgica, 1897-99, Botan., Lichens: 11 (1903). - Neuropogon antarcticus f. sorediifer (Crombie) Lamb in J. Linn. Soc. (Bot.) 52: 213 (1939). Type: Kerguelen Land, Royal Sound, Venus Transit Expedition, A. E. Eaton (BM! - lectotype, selected Lamb, 1939a; BM!, M!, UPS! - isolectotypes). [TLC: fumarprotocetraric acid aggr., usnic acid.] Usnea sulphurea var. granulifera Vainio, Res. Voy. S. Y. Belgica, 1897-99, Botan., Lichens: 11 (1903). - Usnea melaxantha var. granulifera (Vainio) Hue, Deux Exped. Antarct. Franf. Lichenes: 27 (1915). - Usnea granulifera (Vainio) Motyka, Lich. Gen. Usn. Stud. Monogr. 1: 35 (1936). Type: Detroit de Gerlache: dans les fentes d'une falaise rocheuse, Cap Anna Osterrieth, 6433' de latitude sud, Terre de Danco (9 e debarquement, 197 p.p.), M. Emile G. Racovitza (TUR 443 herb. Vainio 358! - lectotype, selected Dodge, 1973; herb. Dodge - isolectotype, not seen). [TLC: no medullary substances, usnic acid.] (see Note 1) Neuropogon insularis Lamb in/. Linn. Soc. (Bot.) 52: 215 (1939). - Usnea insularis (Lamb) Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 211 (1948). Type: Prince Edward Group: Marion L, 26 December 1873, Challenger Exped. (BM! - holotype; BM! - isotype). [TLC: fumarprotocetraric acid aggr., usnic acid.] Usnea crombiei ['Crombii'] Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 212 (1948); nom. inval. (Article 36.1.). Spec, orig.: Heard L, between Atlas Cove and Corinthian Bay, B.A.N.Z.A.R.E. B140-36 (herb. Dodge, not seen). [TLC: fumarprotocetraric acid, usnic acid (Lamb, 1964) AD! - loc. classicus.] Usnea crombiei var. sublaevis Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 213 (1948); nom. inval. (Article 36.1) Spec, orig.: Heard L, between Atlas Cove and Corinthian Bay, B.A.N.Z.A.R.E. B 140-40 (herb. Dodge, not seen). [TLC: fumarprotocetraric acid, usnic acid (AD! - loc. classicus).] Usnea floriformis Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 214 (1948). Type: Heard I., Atlas 56 F. J. WALKER Cove and Corinthian Bay, B.A.N.Z.A.R.E. B 140-41 (herb. Dodge - holotype, not seen). [TLC: fumarprotocetraric acid, usnic acid (AD! - loc. classicus).] Usnea propagulifera Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 213 (1948). Type: Heard I., Atlas Cove to Corinthian Bay, B.A.N.Z.A.R.E. B150-42 (herb. Dodge - holotype, not seen). [TLC: fumarprotocetraric acid aggr., usnic acid (AD! - loc. classicus).] Usnea pustulata Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 203 (1948). Type: Queen Mary Land, Possession Nunatak [6645'S: 9830'E], C. T. Harrisson A.A.E. 85-1 (herb. Dodge - holotype, not seen). [TLC: no medullary substances, (Lamb, 1948fl).] Wsnea subfoveolata Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 202 (1948). Type: Queen Mary Land, Hippo Nunatak [6626'S: 9806'E], C. T. Harrisson A.A.E. 82-2 (herb. Dodge - holotype, not seen). [Chemistry: no medullary substances (Lamb, unpublished notes, AAS!)] AD! - loc. classicus = U. sphacelata. (see Note 2) Wsnea subpapillata Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 204 (1948). Type: Queen Mary Land, Hippo Nunatak, C. T. Harrisson A.A.E. 82-3 (herb. Dodge -holotype, not seen). [Chemistry: no medullary substances (Lamb, unpublished notes, AAS!).] AD! - loc. classicus = U. sphacelata. Usnea crassa Zammuto in Dodge in Trans. Am. microsc. Soc. 84: 521 (1965). Type: Antarctica, Melchior Is., Eta (Bailey) I. [6419'S: 6255'W], 5 March 1941, P. A. Siple 349 (herb. Dodge -holotype, not seen; US! - isotype). [TLC: no medullary substances, usnic acid (isotype).] Usnea pseudofruticosa Zammuto in Dodge in Trans. Am. microsc. Soc. 84: 521 (1965). Type: Antarctica, Melchior Is., Gordon Lien I., 7 March 1941, P. A. Siple 352 (herb. Dodge - holotype, not seen; US! - isotype). [TLC: no medullary substances, usnic acid (isotype).] Note 1: Usnea sulphurea var. granulifera Vainio This taxon was based on four paratype collections, some of which were later further subdivided, in herb. Vainio (TUR). Although Lamb (19390) illustrated part of herb. Vainio 360 as the type it does not appear that the taxon was formally lectotypified. A small part of this collection in BM! was later annotated as isotype by Lamb. Motyka (1936) referred to material from lie Auguste as being from the locus classicus but did not indicate which Vainio specimen was the type. This locality is cited by Vainio under Racovitza 208 p.p., which in herb. Vainio is represented by numbers 359, 360, and 361. Consequently later formal lectotypification by Dodge (1973) must be recognised. It is clear that Dodge did not examine all the paratype collections since he selected herb. Vainio 358 (Cap Anna Osterrieth) on the grounds of this being the only fertile material, which is not the case. Herb. Vainio 358 is a mixture of U. aurantiaco-atra and fertile U. antarctica. It is assumed that Dodge based his lectotypification on U. antarctica and that a duplicate specimen in herb. Dodge also belongs to the species. Other paratypes examined belong to U. antarctica. Note 2: Usnea subfoveolata Dodge The systematic position of this taxon remains uncertain. Dodge (1948) regarded the species as being somewhat intermediate between U. frigida (= U. sphacelata) and U. antarctica, since it morphologically resembles the former species but its anatomy was closer to the latter. Type material, from Queen Mary Land, has not been made available but was examined by Lamb in MO (Lamb, 1964), who considered it to be nearer to U. antarctica. Material from the type locality in AD has been examined and is nearer to U. sphacelata, along with additional specimens in US, possibly determined by Dodge's student. Examination of a photograph of the type specimen (Lamb, unpublished notes, AAS!) shows some affinities with U. antarctica, although examination of the actual specimen is required for the systematic position to be finally ascertained. Description: Thallus (l-5)-2-5(-7-10) cm, arising from a delimited, rarely pigmented, holdfast, erect, dichotomous, richly branched above with numerous, attenuate branches, rarely with fibrils. Branches terete, yellow-green, variegated above with bands of black to violaceous black pigment, continuously pigmented towards the apices. Cortex variable in thickness. Surface matt, subpapillate to grossly papillate, papillae usually unpigmented, black- edged annulations absent. Medulla compact, axis thick, occupying 0-5 to 0-7 of the branch diameter. Soralia extensive throughout thallus, plane to excavate, rarely pulvinate, arising from papillae, delimited, often with a distinct crateriform margin. Soredia granular, unpig- mented, rarely partially corticate then pigmented. Pseudoisidia rare, isidia absent. Apothecia rare, subterminal, as in U. aurantiaco-atra. Pycnidia not seen. TLC: (1) fumarprotocetraric acid, protocetraric acid, cph-1, UV+ unknowns, usnic acid; (2) norstictic acid, salazinic acid, usnic acid; (3) no medullary substances, UV+ unknowns, usnic acid. USNEA SUBGENUS NEUROPOGON 57 ig. 13 Usnea antarctica. Top. Isotype of Usnea antqrctica Du Rietz (O) x 1. Bottom. Detail of soralia. South Shetland Is., Lindsay 471 (BM) xlO. 58 F. J. WALKER Distinguishing features: Usnea antarctica is characterised by its erect habit, a richly branched thallus arising from a delimited holdfast, pigmented towards the apices, a grossly papillate surface, a compact medulla, a thick axis, and numerous, plane, discrete, more or less marginate soralia. It is the only asexual species of the subgenus known to occasionally contain fumarpro- tocetraric acid. Distribution: Usnea antarctica is a circumpolar antarctic species which has its main centre of distribution in the region of the Antarctic peninsula and the associated islands of the Scotia Arc. It also occurs on the subantarctic islands, for example: Bouvet0y, Kerguelen, Macquarie, and Marion, and is rare in the Andes of southern South America (Lamb, 1964) as well as at high altitudes in New Zealand (Mark & Bliss, 1970; Martin & Child, 1972). The species is less frequent in continental Antarctica (Dodge, 1962; Dodge & Baker, 1938; Filson, 1966, 1974; Lamb, 1964; 0vstedal, 1983) whilst it is the only asexual species of the subgenus on the South Shetlands, South Sandwich Islands, Bouvet0y, lies Kerguelen, Heard Island, Me. Donald Islands, Marion Island, Macquarie Island, and South Georgia. Fig. 4. Chemistry: Thalli lacking medullary substances (Race 3) are most frequently encountered overall, whilst those containing fumarprotocetraric acid (Race 1) predominate, or are often the only race, in the subantarctic regions. Both Races 1 and 3 occur on the Antarctic peninsula, although Race 3 extends further south and is the only race found in continental Antarctica. Most South American specimens belong to Race 1 and any depsidone deficient material should be checked against Usnea patagonica and analysed by TLC for fatty acids. The presence of occasional traces of unidentified fatty acids in a few specimens of U. antarctica is not considered to be significant. Thalli containing high concentrations of fumarprotocetraric acid were initially regarded as a separate form, f. sorediifera, by Lamb (19390). Traces of salazinic acid were occasionally detected in Race 1 whilst the existence of a norstictic and salazinic acid-containing race (Race 2) occurs in the only specimen known from the Falkland Islands. UV+ unknowns, identical to those found in U. perpusilla, occur sporadically in Races 1 and 3. Psoromic acid has not yet been demonstrated; previous reports (Golubkova & Schapiro, 1970; Lamb, 19390) correspond respectively to an UV+ unknown and low concentrations of fumarprotocetraric acid that give a PD+ yellow reaction but are not detectable by TLC (BM!). Variation: Lamb (19390) considered Usnea antarctica to be one of the most variable species of the subgenus and discussed the problems involved in selecting suitable characters for delimiting the species. Frequently not all the distinguishing features are developed. However, the species can usually be separated from other asexual species solely on features of the cortex and soralia. Thallus size and extent of branching are variable, although the thallus almost always arises more or less dichotomously from a delimited holdfast. Secondary branching may be extensive with the production of numerous, relatively short, fine branches. Thalli may rarely become subdecumbent when growing in exposed situations or amongst bryophytes. The degree of violaceous black pigmentation is generally more extensive in richly branched specimens whilst in some instances pigment may be confined to branch apices or cortical fragments in the soralia. Pigment is usually lacking on primary branches, whilst the area above the holdfast is rarely pigmented. Primary branches are characterised by the presence of conspicuous unpigmented papillae. Papillation is sometimes very reduced particularly in smaller thalli, giving an almost faveolate or subfaveolate appearance. In such rare instances the thallus may have a slightly waxy lustre, rather than more characteristically dull or matt, thus resembling some forms of Usnea sphacelata. In these specimens the soralia tend to be very small and their margins become less prominent, or may even be lacking if the surface is exceptionally smooth. This modification has led to the description of several taxa which are here reduced to synonymy, although the interpretation of some of these has been difficult in the non-availability of holotype material. For example, U. pustulata represents an extensively sorediate form with a thin, compact, medulla, and an almost smooth, somewhat waxy, surface. From examination of material from near the type locality (AD ! ) and an unpublished photograph of the holotype taken USNEA SUBGENUS NEUROPOGON 59 by Lamb (A AS!), it is clear that this taxon agrees in all details with U. antarctica, although U. sphacelata also occurs in a mixed collection from Possession nunatak made by Borchgrevink (UPS!). Usnea subfoveolata and U. subpapillata were both described from the same locality. From examination of additional cited specimens (AD!) of both species from the type locality, it is evident that Dodge's original concepts (Dodge, 1948) of the taxa included some specimens of U. sphacelata. However, following Lamb's observations (Lamb, 1964) on the type specimens and examination of his unpublished photographs (AAS!), it is likely that the type material of both taxa is closer to U. antarctica. The pale patches in the cortex of the type of Usnea insularis appear to be the result of erosion combined with incipient soralia formation, rather than maculae formed as a result of extension of the^axis through the cortex as in U. taylorii. This type has been compared with U. antarctica from lies Kerguelen which has a similar eroded cortex and only a few, small papillae. Thallus anatomy is constant in as far as the axis usually occupies half or more of the diameter in main branches and the medulla remains densely interwoven, at least towards the cortex. A very wide range of variation occurs in the relative widths of each tissue, particularly amongst robust populations from lies Kerguelen and Marion and Prince Edward Islands. Such develop- ment has formed the basis for the description of a wide range of taxa that can no longer be regarded as distinct. For example, Usnea crassa is a very robust form of the species originally distinguished by an exceptionally thick axis, that is irregular in outline, and possesses a small central cavity, as well as a very thick cortex. The original description gives measurements of up to 260 fjim for the cortex, which presumably includes prominent papillae, since the maximum width for isotype material is c. 170 /xm, whilst c. 100 //,m is more usual in specimens from the Antarctic peninsula. Similarly U. pseudofruticosa was separated on slight differences in thickness of cortex, medulla, and axis. Robust material of Race 1 from Heard Island and lies Kerguelen was reported to have a partially sub-divided axis like Usnea taylorii and recognised as a distinct species, U. insularis (Dodge, 1948; Lamb, 19390; Lindsay, 1977ft). Occasionally thalli examined from these loca- tions, and also from Marion Island, have axes with a small central lumen and slightly irregular outline, but I have not found any thalli with a distinctly divided axis. U. crombiei was distinguished for similar reasons, although the original material indicates that the slight invagination of the axis is not particularly pronounced. Papillae give rise to soralia throughout the thallus which may extend down main branches, rather than remaining confined to apices as in Usnea sphacelata. In robust thalli, with large papillae, the soralia are very characteristic since papillae from which they are derived form a distinctive crateriform margin. Soralia are usually unpigmented, plane to slightly concave- excavate, rarely convex, and smaller than the branch diameter, although they may become confluent on ultimate branches, a feature that is sometimes characteristic of Race 1 from the subantarctic regions. In southern South America a wider range of variation of form of soralia occurs. Sometimes small pseudoisidia are produced in the soralium, but these are never as large or distinct as in U. durietzii or U. patagonica. Other variations include small, plane soralia in more extensively pigmented thalli with fibrils; globular, pigmented soralia are characteristic of a few thalli all belonging to Race 1 which also have a slightly lax medulla and were found in a mixed collection with U. cf. subantarctica (Santesson 641c, S!). When damaged, small, spinule- like projections, may be produced from soralia (Santesson 637 a, S!); this was regarded as a distinctive feature in U. propagulifera by Dodge (1948). Fertile material (Fig. 14) of Usnea antarctica is rare and apothecia are only produced in areas of optimum luxuriance, for example, the type of U.floriformisfrom Heard Island. Lamb (19390, 1948a, 1964) reported fertile material from the Antarctic peninsula, South Georgia, and Tierra del Fuego. In addition, fertile material belonging to Races 1 and 3 has now been found amongst collections from the South Orkneys, South Shetlands, South Sandwich Islands, and Bouvet0y; the only Falkland Island specimen (Race 2) also bears rudimentary apothecia. The apothecia are subterminal or lateral, or are sometimes produced in series along a single branch. They are usually small, cupular, or rarely expanded with an irregular to crenulate margin. The disc is 60 F. J. WALKER ,41. Fig. 14 Fertile thalli of Usnea antarctica. Antarctic peninsula, Anvers I., Lamb 8025 p.p. (FH) xl. black or rarely brown when immature. In fertile specimens soralia may be either abundant or scarce. One specimen (Race 1) from South Georgia (Lindsay 4327, AAS!) bears a superficial resemblance to U. ciliata in habit but has a verrucose surface and very scanty development of soralia. Species concept: The wide range of variation found in Usnea antarctica has led to the description and recognition of a number of species by various authors according to their particular species concepts (see synonymy), all of which here fall within the variation of the species. The holotype specimen of Usnea antarctica represents a less robust form and in some respects has some features in common with U. sphacelata. The thalli have a slightly lax medulla, slender, variegated secondary branches, are subpapillate, and have small, plane, eroded soralia, sometimes lacking the distinctive margin. It is surprising that Usnea antarctica was not recognised as a distinct species before Du Rietz' treatment (Du Rietz, 1926) of the subgenus; this taxon previously being included as a variety of U. sphacelata. Du Rietz (1926) based the species on two varieties previously assigned to U. sulphurea. Of these U. sulphurea var. granulifera is more typical of populations from the Antarctic peninsula and var. sorediifera representative of richly sorediate material of Race 1 form lies Kerguelen. Usnea antarctica is very probably the sorediate counterpart of U. aurantiaco-atra. Supporting evidence for this comes from the similarity of surface ornamentation and apothecia as well as from chemical data. The existance of three chemical races in the two species with similar distributions in South America and Antarctica lends support to this theory, in particular the validity of the single specimen of Race 3. This sorediate specimen is unique and was probably the result of a sporadic secondary formation from the norstictic-salazinic race of the fertile counterpart, on the Falkland Islands. It is unlikely that the specimen is misplaced since the remaining two specimens in the same herbarium packet belong to the norstictic-salazinic race of U. aurantiaco-atra which also has a restricted distribution (see p. 69). In areas where the distributions overlap, Usnea antarctica may be distinguished from U. sphacelata and U. subantarctica by the marginate soralia arising from large, unpigmented papillae, a compact medulla and broad axis; from U. durietzii and U. patagonica (Table 3) by USNEA SUBGENUS NEUROPOGON 61 differences in habit, basal pigmentation, a compact medulla, and absence of pseudoisidia; from U. neuropogonoides by habit and presence of soralia; and from U. acromelana and U. pseudocapillaris by the presence of papillae and lack of annulations, and sometimes also by chemical differences. Selected specimens examined Racel CHILE. Magellanes: Mt. Aymond, Straits of Magellan, 1872, Hassler Exped. s.n. (BM, FH, PC); Isla Navarino, Puerto Navarino, 10 m, 28 February 1940, R. Santesson 1234b (S); Tierra del Fuego, San Sebastian, 2 January 1896, P. Dusen 80 (BM); Canal Beagle, Yendegaia, by glacier, 50-100 m, 4 March 1940, R. Santesson 1373 p.p. (UPS). ARGENTINA. Santa Cruz: Cordillera Darwin, Cerro Mayo, above Seno Mayo, Lago Argentine, c. 1000 m, February 1959, P. W. James 73 (BM), near front of Ventrisquero Mayo, 90 m, March 1959, P. W. James 5052 (BM). Tierra del Fuego: Sierra Alvear, S. slope, above Las Cotorras, (c. 20 km ENE. of Ushuaia), 700 m, 9 February 1940, R. Santesson 637a (S, UPS), Sierra Sorondo, N. slope, above Las Cotorras, 800 m, 6 February 1940, R. Santesson 641c. p.p. (S). SOUTH GEORGIA. Cumberland East Bay, above Hope Point, 80 m, January 1973, D. C. Lindsay 4327 (fertile) (AAS); Whale Valley, 250 m, January 1973, D. C. Lindsay 3957 (fertile) (AAS); Royal Bay, 29 April 1902, C. Skottsberg s.n. (S); N. of Sandebugten, W. shore of Barff Peninsula, c. 15 m, 14 January 1961, S. W. Greene 939 (AAS, BM). SOUTH ORKNEY IS. Signy I.: Paal Harbour, 90 m, 2 April 1965, R. I. L. Smith 429 (AAS); Tern Cove, 15 m, 24 October 1973, T. N. Hooker 263 (fertile) (AAS). Coronation I.: Top of Wave Peak Buttress, 392 m, 29 September 1973, T. N. Hooker 194 (AAS). Laurie L: Cape Dundas, 60 m, 15 February 1971, M. McManmon 195 (AAS). SOUTH SHETLAND IS. King George I.: Admiralty Bay, Keller Peninsula, 24 December 1960, B. J. Taylor 324 (AAS), 50-60 m, 5 January 1953, B. Frodin 21a (UPS). ANTARCTIC PENINSULA. Graham Land: Joinville I. [6315'S: 5548'W], Mt. Alexander, 200-500 m, 27 March 1981, R. I. L. Smith 3680 p.p. (AAS); Trinity Peninsula, Hope Bay [6323'S: 5700'W], 105 m, 16 April 1945, F.I.D.S. D2375 p.p. (BM), Duse Bay [6332'S: 5715'W], 370 m, 6 January 1946, E. H. Back F.I.D.S. D2865 (BM); Palmer Archipelago, Wiencke I. [6448'S: 6325'W], N. buttress of Noble Peak, c. 210 m, 15 October 1944, /. M. Lamb Operation Tabarin 1307 (BM); Argentine Is., Galindez I. [6515'S: 6415'W], 27 December 1935, B.G.L.E. 1328-2 (BM); Fallieres Coast, Adelaide I. [c. 6760'S: 6820'W], nunatak c. 18 km N. of base hut [Rothera], 10 December 1962, F. Gibbs Killingbeck 207b (AAS, BM). MARION I. Tafelberg, E. slope, c. 320 m, 10 May 1982, H. Hertel 24 593 (M); Skua's Ridge, 80 m, 26 January 1972, A. de Villers 4-31 (BLFU); Johnny's Hill, 4 January 1951, R. W. Rand 3310 (BOL). PRINCE EDWARD I. vicinity of Kent Crater, c. 50 m, 1 May 1982, H. Hertel 24 343 (M). IS. KERGUELEN. Cliffs above Lake du Val Studer, 11 February 1963, R. B. Filson 4665 p.p. (BM); Royal Sound, 1897-98, R. Hall s.n. (MEL 9230). HEARD I. between Rogers Head, Corinthian Bay, West Beach, and the foot of the glacier from Big Ben (Kaiser Wilhelm Peak), 28 November-2 December 1929, B.A.N.Z.A.R.E. B140-43 (AD). MC.DONALD IS. 13 March 1980, /. Jenkin s.n. (herb. Seppelt 11123). MACQUARIE I. between North Mount and Handspike Point, 7 December 1968, D. McVean 6909 (BM); Perseverance Bluff, 100 m, 29 January 1982, R. D. Seppelt 12748 (herb. Seppelt). NEW ZEALAND. Canterbury: Kirkliston Range, 1680-1830 m, 25 March 1978, D. J. Galloway s.n. (BM, CHR 343299). Otago: Mt. Pisa, 1920 m, March 1968, D. J. Galloway s.n. (CHR 343771); Mt. Earnslaw, 2240 m, May 1961, R. G. Cunninghame s.n. p.p. (BM, OTA); Old Man Range, c. 1370 m, 1 February 1963, P. W. James 1159 p.p. (BM). Race 2 FALKLAND IS. (unrealised), 1824, Greville herb. (E). Race 3 SOUTH GEORGIA. Cumberland West Bay, Bore Valley, 105 m, 11 February 1961, 5. W. Greene 1979 (AAS, BM, FH); Cumberland East Bay, Dartmouth Point, 30 m, 5 December 1971, D. C. Lindsay 3458 (BM); 1921-22 Shackleton-Rowett Exped. 10 p.p. (BM). SOUTH SANDWICH IS. Candlemas I.: 180 m, 5 March 1964, R. E. Longton 535 (fertile) (BM). Bellinghausen I. : 200 m, 10 March 1964, M. W. Holdgate 825c (AAS, BM). Visokoi I. : Finger Point, 6 m, 9 January 1961, K. Archibald^ (AAS, BM, FH). SOUTH ORKNEY IS. Signy L: Borge Bay, Knife Point, 9 m, 8 October 1966, D. C. Lindsay 1255 (BM); Observation Bluff, 60 m, 21 January 1972, T. N. Hooker 1 (AAS). Laurie I. : Scotia Bay, 11 March 1953, A. 62 F. J. WALKER E. Hunziker 10238 (FH). Powell I.: between Cape Disappointment and Falkland Harbour, 30 m, 30 January 1965, R. I. L. Smith 263 (AAS, BM). SOUTH SHETLAND IS. King George I. : Admiralty Bay, mountain slope between Petrified Forest Creek and Ornithologist's Creek, 100m, 26 December 1979, fl. Ochyra 5000/79 (M). Deception I.: Kroner Lake, 30 January 1960, B. J. Taylor 29 (AAS, BM). Robert I.: 1963, G. Follmann 17441 (KASSEL). ANTARCTIC PENINSULA. Graham Land: Joinville I. [6315'S: 5545'W], Active Sound, 200 m, 27 March 1981, R. I. L. Smith 3735 (AAS); Palmer Archipelago, Anvers I. [6445'S: 6405'W], Arthur Harbour, Palmer Station, 10 m, 4 February 1965, /. M. Lamb Operation Tabarin 8025 p.p. (fertile) (FH); Marguerite Bay, Jenny I. [6744'S: 6825'W], 9 m, 17 October 1948, B. Stonehouse & V. E. Fuchs F.I.D.S. E466-la (BM). DRONNING MAUD LAND. [c. 15W]. Vestfjella, Basen, c. 300 m, 1968-69, A. Hjelle s.n. (O). EDWARD VII LAND. [c. 150W]. Rockefeller Mountains, Mt. Marguerite Wade, 17 December 1940, R. G. Frazier & F. A. Wade U.S. A.S. 320 (US). VICTORIA LAND. [c. 165E]. Geikie Land, 300 m, 1900, Southern Cross Exped. s.n. (BM); Hallett Peninsula, c. 300 m, 11 November 1962, E. D. Rudolph 62040 (LD). GEORGE V LAND. [c. 150E]. Dreadnought or Horn Bluff, 21 December 1912, A. L. McLean A.A.E. 46 (AD). KNOX COAST, [c. 110E]. Mitchell Peninsula, outcrop 1-5 km E. of Pidgeon I. , 21 December 1982, R. D. Seppelt 013357 (herb. Seppelt). QUEEN MARY LAND. [c. 95E]. Alligator Nunatak, 2 January 1913, C. T. Harrisson A.A.E. 72 (AD). KAISER WILHELM II LAND. [c. 90E]. Haswell I. , 20 December 1957, V. Rozov 101 (AAS, FH). MAC.ROBERTSON LAND. [c. 70E]. Tschuffert Peak, 9 February 1974, R. Filson 14900 Lich. Ant. Exs. 23 (AAS, BM, M, O); Hays Peak, Cape Bruce, 10 October 1962, R. Filson 4387 (AAS). MAC.ROBERTSON-ENDERBY LAND TRAVERSE, [c. 60E]. 1964-65, N. Leid Bratt 3272a (HO 35190). BOUVET0Y. Rustadkollen, 340 m, 7 March 1979, T. Engelskjon s.n. (BG, BM); S. of Kapp Circon- cision, c. 15 m, 31 March 1964, M. W. Holdgate 868a (AAS, BM, FH). MARION I. 26 December 1873, Challenger Exped. s.n. (type of U. insularis Lamb) (BM). HEARD I. Dreadnought (Horn) Bluff, 21 December 1912, A. L. McLean A.A.E. 46 (AD). MACQUARIE I. Wireless Hill, 90 m, 8 December 1968, D. McVean 6907 (BM); near Lake Scoble, 7 December 1968, D. McVean 6935 (BM); Green Gorge, 220 m, 26 October 1983, R. D. Seppelt 14297 (herb. Seppelt). The following may be consulted for further localities: South Georgia (Lindsay, 1974), South Sandwich Is. (Longton & Holdgate, 1979), South Orkney Is. (Smith, 1973), South Shetland Is. (Lindsay, 19710), Antarctic Peninsula (Lamb, 1964); together with lists held in BM and collections in AAS and BM. 4. Usneaaurantiaco-atra(Jacq.)Bory Figs 15-19 in Mem. Soc. Linn. Paris 4: 596 (1826). - Lichen aurantiaco-aterJacq. , Miscell. Austriac. 2: 369 (1781). - Neuropogon aurantiaco-ater (Jacq.) Lamb in/. Linn. Soc. (Bot.) 52: 221 (1939). Type: [Chile] Magellan Straits, 1767, Commerson, comm. D. Jussieu (PC! - lectotype, selected here; BM!; FI!, LINN (herb. Smith 1715.9)!, UPS (herb. Thunberg 26355)! - isolectotypes). [TLC: fumarprotocetraric acid aggr., usnic acid.] (see Note 1) Usnea melaxantha Ach., Method Lich.: 307 (1803). - Parmelia melaxantha (Ach.) Sprengel, Syst. Veg. 4 (1): 277 (1827). - Neuropogon melaxanthus (Ach.) Nyl. in Mem. Soc. Imp. Sci. Nat. Cherbourg 3: 170 (1855). Type: America [West Falkland Is.], Port Egmont, Cavanilles, Ach. 1887A (H-ACH! - lectotype; BM! - isolectotype). [TLC: norstictic acid, salazinic acid, connorstictic acid, protocetraric acid (trace), usnic acid.] (see note 2) Usnea fasciata Torrey in Am. J. Sci. 6: 106 (1823). - Usnea melaxantha var. yfasciata J. D. Hook., Flora Antarctica 2: 520 (1847). Type: New South Shetland, comm. Dr. Mitchill (NY! - lectotype, selected here). [TLC: fumarprotocetraric acid, protocetraric acid (trace), usnic acid.] (see Note 3) Corniculariaflavicans Pers. in Gaudichaud in Freycinet, Voyage autour du Monde, Botanique: 210 (1828). Type: lies Malouines [Falkland Is.], 1820, C. Gaudichaud 130 (PC! - holotype; G-DEL! - isotype; BM! - ?isotype). [TLC: norstictic acid, salazinic acid, protocetraric acid (trace), usnic acid.] Neuropogon antennarius Nees & Flotow in Linnaea 9: 497 (1835). - Usnea antennaria (Nees & Flotow) Massal., Memor. Lichenogr.: 73 (1853). Type: ad rupes Chil. austr. in summo cacumine Pico de Pilque prope Antuco [3720'S: 7141'W], legit Poeppig, ex herb. Kunze (VER (herb. Massalongo)! -lectotype, selected here). [TLC: norstictic acid, salazinic acid, usnic acid.] (see Note 4) USNEA SUBGENUS NEUROPOGON 63 Usnea melaxantha var. aAchariiJ. D. Hook., Flora Antarctica 2: 520 (1847); nom. illeg. (Article 26.1). Usnea melaxantha var. fiJacquinii J. D. Hook., Flora Antarctica 2: 520 (1847). Type: Lichen aurantiaco- ater Jacq. Usnea melaxantha var. subciliata Zahlbr. in Annls mycol. 1: 360 (1903). - Usnea sulphurea var. subciliata (Zahlbr.) Zahlbr., Cat. Lich. Univ. 6: 603 (1930). - Neuropogon strigulosus f. subciliatus (Zahlbr.) Lamb in/. Linn. Soc. (Bot.) 52: 231 (1939). Type: Patagonia, P. Neumann (W7177! -holotype). [TLC: fumarprotocetraric acid (trace), usnic acid.] Usnea melaxantha var. subciliata f. strigulosa Zahlbr. in Annls mycol. 1: 360 (1903). - Usnea melaxantha f . strigulosa (Zahlbr.) R. H. Howe in Bryologist 18: 61 (1915). - Usnea sulphurea var. subciliata f. strigulosa (Zahlbr.) Zahlbr., Cat. Lich. Univ. 6: 603 (1930). - Usnea strigulosa (Zahlbr.) Motyka, Lich. Gen. Usn. Stud. Monogr. 1: 30 (1936). - Neuropogon strigulosus (Zahlbr.) Lamb in/. Linn. Soc. (Bot.) 52: 229 (1939). - Neuropogon aurantiaco-ater f. strigulosus (Zahlbr.) Lamb in Lilloa 14: 152 (1948). - Usnea aurantiaco-atra f. strigulosa (Zahlbr.) Lamb in An. Parq. nac. B. Aires 7: 156 (1959) ['1958']. - Usnea fasciata f. strigulosa (Zahlbr.) Lamb in Br. Antarct. Surv. Sci. Rep. 38: 14 (1964). Type: Patagonia, P. Neumann (W7116!- holotype). [TLC: fumarprotocetraric acid, protocetraricacid, UV+ unknowns, usnic acid.] Usnea sulphurea var. normalis Vainio, Res. Voy. S. Y. Belgica, 1897-99, Botan., Lichens: 11 (1903) nom. inval. (Article 26.1). - Neuropogon aurantiaco-ater f. normalis Lamb in /. Linn. Soc. (Bot.) 52: 224 (1939) nom. inval. (Article 26.1.). Usnea sulphurea var. normalis f. activa Zahlbr. in K. svenska VetenskAkad. Handl. 57 (6): 44 (1917). - Usnea sulphurea f. activa (Zahlbr.) Zahlbr. Cat. Lich. Univ. 6: 603 (1930). Type: Falkland Inseln, Port Stanley, Sapper Hill, Schwedische antarktische Exped. 1901-03. (W 123 ! - holotype). [ TLC: norstictic acid, salazinic acid, usnic acid.] Usnea sulphurea var. spadicea Zahlbr. in K. svenska VetenskAkad. Handl. 57 (6): 45 (1917). - Usnea tayloriivar. subspadicea (Zahlbr.) Motyka in Rasanen in Suomal. eldin-ja kasvit. Seur. van. kasvit. Julk. 2(1): 11 (1932) nom. illeg. (Article 63.1). - Usnea melaxantha var. spadicea (Zahlbr.) Motyka, Lich. Gen. Usn. Stud. Mongr. 1: 35 (1936). - Neuropogon melaxanthus f. spadiceus (Zahlbr.) Lamb in/. Linn. Soc. (Bot.) 52: 228 (1939). Type: Falkland Inseln, Mt. Adams, Schwedische antarktische Exped. 1901-03, C. Skottsberg. (W121! - holotype). [TLC: norstictic acid, salazinic acid, usnic acid.] (see Note 5) Usnea melaxantha var. nigropallida Cengia-Sambo in Boll. Soc. hot. ital. 1: 91 (1926). - Neuropogon aurantiaco-ater f. nigropallida (Cengia-Sambo) Lamb in /. Linn. Soc. (Bot.) 52: 224 (1939). Type: Tierra del Fuoco, 1913, G. B. de Gasperi (FI! - holotype; FI! - isotype). [TLC: fumarprotocetraric acid, protocetraric acid, usnic acid.] Usnea taylorii var. kranckii Rasanen in Suomal. elain-ja kasvit. Seur. van. Julk. 2(1): 10 (1932). Type: [Chile] Fuegia occ. Fjordo Martinez, Seno Pliischow, c. 500 m.s.m., reg. alp., Expeditio Fennica 1928-29, 21 February 1929, E. H. Kranck (H! - lectotype, selected here). [TLC: fumarprotocetraric acid, protocetraric acid, usnic acid.] (see Note 6) Usnea trachycarpa var. eciliata Rasanen in Suomal. eldin-ja kasvit. Seur. van kasvit Julk. 2(1): 10 (1932). Type: [Chile] Fuegia occ. Puerto Yartou, Pico Nariz, 750 m.s.m., reg. alp. Expeditio Fennica 1928-29, H. Roivainen (H! - holotype; S - isotype, not seen). [TLC: norstictic acid, salazinic acid, protocetraric acid (trace), usnic acid.] Usnea aurantiaca Motyka, Lich. Gen. Usn. Stud. Monogr. 1: 29 (1936). - Neuropogon aurantiacus (Motyka) Lamb in/. Linn. Soc. (Bot.) 52: 229 (1939). Type: Falkland Is., D. Coleman, A. Zahlbr. Lich. Rar. Exs. 20 (W5133! - holotype; UPS! - isotype). [TLC: norstictic acid, salazinic acid, protocetraric acid (trace), usnic acid.] Neuropogon aurantiaco-ater f. egentissimus Lamb in /. Linn. Soc. (Bot.) 52: 225 (1939). - Usnea aurantiaco-ater f. egentissima (Lamb) Dodge, Lich. FI. Antarct. Cont.: 227 (1973). Type: Antarctica, Graham Land, Argentine Is., summit of Galindez I., 2 January 1936, Br. Graham Land ('Penola') Exped. 1934-37, 1340J (BM! - holotype). [TLC: no medullary substances, usnic acid.] Neuropogon melaxanthus f.fibrillifer Lamb in Lilloa 14: 154 (1948). Type: Falkland Is. , peat moors above Port Stanley, c. 60 m, 27 January 1946, 1. M. Lamb, 'Operation Tabarin' 2873 (BM! - holotype). [TLC: norstictic acid, salazinic acid, protocetraric acid (trace), usnic acid.] lUsnea siplei Zammuto in Dodge in Trans. Am. microsc. Soc. 84: 522 (1965). Type: Antarctica [Graham Land], Melchior Is., Omega I. [= Lystad I.], 6420'S: 6256'W, 14 March 1941, P. A. Siple 367 (herb. Dodge - holotype, not seen; US! - isotype). [TLC: fumarprotocetraric acid aggr., UV+ unknowns, usnic acid.] (see Note 7) 64 F. J. WALKER Note 1: Lichen aurantiaco-ater Jacq. (i) The hyphen should be retained in accordance with Article 73.1. (ii) There has been considerable confusion concerning the correct identity of this taxon. The holotype specimen has not been traced and Jacquin's personal lichen herbarium in Vienna is regarded as lost (Motyka, 1936; Lamb, 1939a). Consequently assumptions have been made as to the medullary chemistry of the type specimen . This has resulted in subsequent changes in use of various specific epithets , all of which are here regarded as synonyms (Table 4). Jacquin (1781) based the name Lichen aurantiaco-ater on material collected by Commerson, probably from the Magellan Straits. Examples of Commerson's collections have been traced in several herbaria and, since they presumably originated from PC, are regarded as isotypes. Two collections have been traced in PC of which one (herb. Jussieu !) very closely conforms with Jacquin's original illustration (Tab XI, fig. 2) and description. This specimen is accordingly designated here as the lectotype (Fig. 15), with isolectotypes selected in other herbaria. The specimen contains fumarprotocetraric acid which follows Lamb's original concept (Lamb, 1939a) of the type. Motyka annotated this specimen as Usnea aurantiaco-atra although wrongly listing it (Motyka, 1936) under U. aurantiaca, with the erroneous date of 1820. ' ^l^f-v^^t A .^ , 4 Fig. 15 Lectotype of Lichen aurantiaco-ater Jacq. (PC) x 1 . Most of Commerson's collections contain fumarprotocetraric acid, apart from the collection in BM which also includes norstictic and salazinic acid-containing thalli. In this particular case only the fumarprotocetraric acid-containing thalli are selected as isolectotype material. This lectotypification reflects the predominance of Race 1 (see p. 69) in the Magellan Straits area. Note 2: Usnea melaxantha Ach. The type collection of Usnea melaxantha Ach. (Ach. 1887 A + B) consists of two plants from different localities, both containing norstictic and salazinic acids. One gathering, collected by Menzies, is from USNEA SUBGENUS NEUROPOGON 65 7^: "*lTt7 " Fig. 16 Usnea aurantiaco-atra. Top. Isolectotype of Usnea melaxantha Ach. (BM) x 1. Bottom. Detail of apothecia. Graham Land, Galindez I., B.G.L.E. 1340 p.p. (BM) xlO. 'Staaten Land' and the other, from the Falkland Islands (Port Egmont), was collected by Cavanilles. Motyka (1936) regarded the Port Egrnont specimen as the nomenclatural type from the classic locality. His selection is therefore considered to effect the lectotypification of this taxon. This lectotype is illustrated by Lamb (1939a, pi. 11, fig. 28). A duplicate collection in BM (Fig. 16) is recognised as an isolectotype although the two plants are unlocalised. Note 3: Usnea fasciata Torrey Torrey probably described this taxon on material from two sources; that collected by Captain Napier of the U.S. Sealing Expedition of 1820-21 (Dodge, 1973) and an unknown collector, both of which were sent 66 F- J- WALKER to him by Dr S. L. Mitchill (Torrey, 1823). Consequently his description is likely to have been based on a mixed gathering. Lamb (1939a) illustrated a fertile specimen of Usnea aurantiaco-atra (NY!) as the holotype, but he later (Lamb, 1964) retracted this because Torrey, in his original description, stated that specimens lacked apothecia, only possessing 'cephalodia' which, from study of his illustration, correspond to a lichen parasite. It is evident, from examination of the herbarium sheet, that Lamb was sent only part of one thallus of the NY collection in a separate capsule. The original collection (Fig. 17), which was not seen by him, bears an inscription in Torrey 's handwriting. Consequently, it may be assumed, despite the presence of fertile material, that this material is authentic, especially as the journal citation on the packet lacks specific details. This collection comprises the 'parent' thallus of the fertile specimen sent to Lamb, as well as a sterile thallus of the same species (both lacking parasites) and two small, distinctly variegated, parasitised thalli of U. antarctica. j$**ARMQ CO. Fig. 17 Ifrnea aurantiaco-atra. Lectotype (centre thallus) of Usnea fasciata Torrey (NY) X 1. USNEA SUBGENUS NEUROPOGON 67 Torrey's illustration (Torrey, 1823; pi. 9 figs 1-4) which represents a large, conspicuously variegated thallus, lacking apothecia but clearly parasitised, must be taken into consideration. None of the thalli in the NY collection exactly correspond to this illustration, although the closest are the parasitised thalli of U. antarctica, despite difference in size and apparent lack of soralia. There is a remote possibility that Torrey may have misinterpreted the apothecia as well-developed 'cephalodia', since he described them as 'scattered, sometimes crowded and irregular', rather than recognising them as the same fertile structures that occur in the subgenus Usnea, although he might have regarded the fertile specimen as an older state of the same species. It is evident from the protologue (Torrey, 1823: 105) that he was fully aware of the distinction between apothecia and 'cephalodia'-like structures in U.florida (=? U. strigosa). Consequently there are three options relevant to the typification of Usnea fasciata based on this NY collection. (i) to follow Lamb's final interpretation (Lamb, 1964) and reject the collection as being the type of the name. A neotype would then have to be selected. (ii) Assume only part of the collection to be the type of the name and lectotypify the parasitised, sorediate material based on comparison with Torrey's illustration. This would result in a change of nomenclature of U. antarctica. (iii) Assume the entire mixed collection to be the type of the name and lectotypify on one element (Article 9:2) so as to preserve current usage, following Recommendation 7B, since the name can no longer be rejected under Article 70. None of these solutions is entirely satisfactory, but it is important to stabilise the situation and to save further misapplication or nomenclatural changes. Thus Usnea fasciata is here lectotypified on the fertile specimen, reducing the taxon to synonymy with U. aurantiaco-atra and preventing a confusing name change for U. antarctica. Note 4: Neuropogon antennarius Nees & Flotow Motyka (1936) reported examining the fragmentary holotype specimen in herb. Flotow (B) which was subsequently destroyed during the Second World War (Krog, 1976). Material of various species from the classic locality, collected by Poeppig, was sent to Kunze at Leipzig (Sayre, 1975). Some of these collections were subsequently distributed by Poeppig and Kunze as part of an unpublished exsiccatum, Poeppig Coll. pi. Chil. N. antennarius does not form either of the two examples of the exsiccatum that have been traced (BM!), and authentic material has not been found in Poeppig's personal herbarium (W), although collections of U. perpusilla from herb. Poeppig and herb. Kunze have been traced in M and PC, annotated U. fasciata and U. melaxantha respectively. The only authentic material of M antennarius traced is in herb. Massalongo (VER!) and originates from herb. Kunze. This specimen is presumably an isotype, since, besides bearing the same data as the above exsiccatum, bears the inscription 'Neuropogon antennarius Nees & Flotow' in Massalongo's handwriting and a small ink stamp he used to denote type material. This collection (Fig. 18) is consequently selected as the lectotype of Neuropogon antennarius, following Krog's lectotypification of N. poeppigii (Krog, 1976) based on an isotype traced in S. Note 5: Usnea sulphurea var. spadicea Zahlbr. This variety was validly published by Zahlbruckner in 1917, although examination of the type specimen (W 121) indicated that he possibly intended the epithet ' subspadicea' to be used since this name appears in his handwriting on the label. Motyka in Rasanen (1932) published the combination 'Usnea taylori var. subspadicea (Zahlbr.) Motyka' based on this unpublished name as indicated on his determination label attached to herb. NYL 36372 (H!). As both names are of the same rank 'Usnea taylori var. subspadicea' must be rejected under Article 63.1 as a superfluous name for Usnea sulphurea var. spadicea Zahlbr. Note 6: Usnea tayloriivar. kranckii Rasanen Examination of the two collections cited under this name by Rasanen (1932) has revealed that they are different species. One gathering from Martinez is fertile material of U. aurantiaco-atra and the other, from Yartou, is an infertile, immature specimen of U. trachycarpa. It is evident, from his notes on the herbarium packet, that Rasanen intended the fertile collection from Martinez to be the type of the variety since the presence of apothecia and spore measurements correspond with the published description. Note 7: Usnea sip lei Zammuto This species was described from the Antarctic peninsula by Zammuto (in Dodge, 19656), and, according to the description, the holotype is small and lacks apothecia and soredia. However, Dodge (1973) subsequently gave a similar description of the species but in the key describes it as being sorediate. A specimen, taken to be an isotype, bearing the same collection number (USAS 367, US!), and an additional collection (USAS 369, US!) from the same locality, both annotated U. siplei, have been examined. Both 68 F. J. WALKER Fig. 18 Lfrnea aurantiaco-atra. Lectotype of Neuropogon antennarius Nees & Flotow (VER) x 1. are immature specimens of t/. aurantiaco-atra as previously stated by Lamb (19480). The isotype bears a single, small, immature apothecium. There is always the possibility that the original gathering may not have been homogeneous, and for this reason the species is only tentatively included here. Additional material from the Melchior Archipelago, collected by the same expedition, has been distributed as Usnea fasciata (Vezda: Lich. Sel. Exs. 675) and is moderately fertile. USNEA SUBGENUS NEUROPOGON 69 Description: Thallus (3)-5-8(-10-13) cm, arising from a delimited or rarely proliferating pigmented holdfast, erect, dichotomous, richly branched above with numerous attenuate branches, fibrils rare. Branches terete or rarely angular, yellow-green, variegated above with bands of black to violaceous black pigment, continuously pigmented towards the apices. Cortex variable in thickness. Surface matt, smooth at base, becoming verrucose-rugose to subfaveolate-papillate, grossly papillate or ridged-faveolate above. Medulla compact, variable in extent, axis thick, occupying 0-5-0-6(-0-8) of the branch diameter. Soredia, pseudoisidia and isidia absent. Apothecia frequent, subterminal or rarely lateral, cupular, expanding on maturity. Disc black, excipulum verrucose-papillate, margin prominent, rays rare. Pycnidia infrequent towards branch apices. TLC: (1) fumarprotocetraric acid, protocetraric acid, cph-1, UV+ unknowns, usnic acid; (2) norstictic acid, salazinic acid, protocetraric acid, usnic acid; (3) no medullary substances, UV+ unknowns, usnic acid. Distinguishing features: Usnea aurantiaco-atra is characterised by its erect habit, a richly branched thallus which is pigmented towards the apices, a verrucose to papillate surface, a compact medulla, a thick axis, and frequent subterminal apothecia with a black disc, usually lacking marginal excipular rays. It is the only fertile species of the subgenus which frequently contains fumarprotocetraric acid. (Fig. 16) Distribution: Usnea aurantiaco-atra has a more restricted distribution than U. antarctica and is more or less confined to the west coast of the Antarctic peninsula, the islands of the Scotia Arc, including Bouvet0y, but excluding the South Sandwich Islands, and extends into subantarctic South America, including the Falkland Islands. The species does not occur in continental Antarctica (cf. Dalenius & Wilson, 1958) or Australasia. Specimens cited from lies Kerguelen, collected by Hooker, are erroneously labelled, having been subsequently mixed with collections of U. taylorii. Fig. 4. Chemistry: Three chemical races occur in Usnea aurantiaco-atra. Of these, specimens contain- ing fumarprotocetraric acid (Race 1) and the depsidone deficient Race 3 are the most commonly encountered and have a much wider distribution than the norstictic-salazinic acid containing Race 2. Race 1 is apparently slightly more frequent than Race 2, with its centre of distribution in the South Orkneys and South Shetlands Islands; this race is also known from the Antarctic peninsula, South Georgia, Tierra del Fuego and Chile as far north as c. latitude 46S. Race 3 extends further north, to c. 37S, and further east to Bouvet0y, than Race 1. Race 2 has its centre of distribution in the Falkland Islands, where Race 1 is only known from a few, possibly misplaced, collections. Race 2 overlaps Race 1 in southern South America, but is less frequent. It is confined to Tierra del Fuego apart from a single, isolated, northerly occurrence which forms the type specimen of Neuropogon antennarius: it is absent from the Antarctic peninsula. There are reports of occasional outliers of Race 2 from South Georgia (Lindsay, 1975) and rather tentatively from the South Shetlands (Motyka, 1936). Of these Lindsay's records were based on K and PD reactions which may have been misinterpreted since a high concentration of fumarprotocetraric acid can give a K+ red-brown reaction. In spite of these records, so far I have only confirmed (by TLC) two collections of Race 2 from South Georgia. Thalli of one gathering (R. I. L. Smith 259, AAS!) are extensively blackened and bear a superficial resemblance to Usnea ciliata, possibly due to the extremely exposed habitat given as a south-facing scree overlooking a glacier. Very rarely thalli with a mixed chemical complement of Races 1 and 2 are encountered, particularly where the distributions of the two races overlap. For example, thalli may contain fumarprotocetraric, protocetraric, salazinic, and norstictic acids, or Race 1 may additionally contain salazinic acid, or Race 2 may either lack norstictic or salazinic acid and have a higher concentration of protocetraric acid. The implications of such thalli with mixed or intermediate chemistries are discussed under 'species concept' below. Variation: Usnea aurantiaco-atra is a very variable, but usually easily recognised species, even when sterile. It exhibits a wide range of growth form, branch anatomy, and degree of pigmentation. The account of U. antarctica (p. 58) should be referred to for discussion of much 70 F. J. WALKER of the variation in branching, morphology, and anatomy. Thalli may rarely become subdecum- bent or straggling and then may be sparsely pigmented, sterile, and infrequently to richly branched; such specimens superficially resemble U. neuropogonoides . In rare instances pigment may be confined to branch apices and the apothecial disc, which seem to be more prevalent in Race 2. In contrast to Usnea antarctica thalli are often larger, though rarely more than 10 cm. In U. aurantiaco-atra a wider range of variation is found in surface ornamentation and branch anatomy, which, although frequently correlated with medullary chemistry, is not sufficiently distinct to merit taxonomic separation. For example, the lectotype of Lichen aurantiaco-ater, typical of Races 1 and 3, is minutely verrucose-papillate and is extensively branched with continuously pigmented ultimate branches. The axis occupies more than half the branch diameter and in transverse section branches are more or less terete and have a narrow medulla, c. 200 fj.m. In contrast, the type of Usnea melaxantha, typical of Race 2, has branches which are somewhat angular-indented in section resulting from faveolation and deeper depressions. The axis of U. melaxantha is somewhat irregular in outline and occupies slightly less than half the diameter coupled with a correspondingly wider medulla of up to 300 /x,m. Ornamentation ranges from almost smooth to obscurely papillate to coarsely verrucose- papillate becoming distinctly faveolate-ridged. False annulations, resembling those of Usnea ciliata, only occur in exceptionally smooth, scantily papillate or weakly verrucose thalli that have been extensively weathered, for example, as in the types of U. trachycarpa var. eciliata and Neuropogon antennarius. On occasions a range of ornamentation may be exhibited by a single thallus. Fibrils are only rarely produced as extended papillae or verruculae and are never as extensively developed as those of U. trachycarpa. Fibrillate forms previously given taxonomic status include N. melaxanthus \ar.fibrillifer and U. melaxantha var. nigropallida; one taxon, U. melaxantha var. subciliata f. strigulosa, was even raised to specific rank (Lamb, 1939a; Motyka, 1936). The axis may be somewhat irregular in outline and can vary in thickness together with the width of the medulla, which tends to be broader in Race 2. Overall it is rare for the axis to occupy more than 0-7 of the branch diameter. Exceptions include the type of Usnea taylorii var. kranckii, which is here treated as a synonym, as well as specimens examined from Bouvet0y (BG!) ; in these the axis occupies 0-7 to 0-8 of the branch diameter and the medulla is narrowed to c. 100 fj,m in primary branches, suggesting a possible affinity with U. taylorii. Apothecia are normally subterminal, rarely with a geniculate appendage, or may be lateral. In some instances the margin may become excluded and the apothecium irregular and reflexed, or even slightly crenulate in well-developed thalli. Only rarely is the disc pigmentation not fully developed (for example, R. I. L. Smith 2573, A AS!). The excipulum is papillate or minutely faveolate; irregular rays are only rarely produced and then in thalli which develop fibrils on main branches. Species concept: To date various taxa comprising the Usnea aurantiaco-atra-U . melaxantha group have been considered to be distinct species by different authors (Table 4). Dodge (1973) recognised six species, Motyka (1936) five; Lamb initially (Lamb, 1939a) accepted four species but finally modified his concept to include only two (Lamb, 1964). Apart from the ensuing nomenclatural confusion the delimitation of each species was by no means clear and depended on small variations in the relative size of axis and medulla, surface ornamentation or presence of fibrils, thallus size or the extent of branching. Variation in chemical reactions was also considered to be significant; for example the depsidone-deficient Neuropogon aurantiaco-ater i. egentissimus as well as Usnea sulphurea var. normalis f. activa which gave an intense K+ red reaction. Motyka (1936) considered that the species could be separated primarily on morphological features, but Lamb later (Lamb, 1948a) considered chemistry to be the only reliable feature and used thallus spot tests to distinguish Usnea melaxantha from U. aurantiaco-atra (i.e. Races 1 and 3 from Race 2). Lindsay, (1975) used chemistry initially to distinguish between the two taxa, but included differences in surface ornamentation in his descriptions. Eventually Lamb (1964) USNEA SUBGENUS NEUROPOGON 71 concluded that there were morphological and distributional differences between the two species which were supported by chemistry. These differences, however, were only apparent when contrasting whole populations rather than considering individuals. This approach is, however, in my estimation impractical, as admitted by Lamb (1964), since many thalli with intermediate characters may occur, particularly where the distributions of chemical races overlap. This leaves chemistry as apparently the only reliable criterion for separating U. aurantiaco-atra from U. melaxantha. These two taxa are here considered to form chemical races of a single species particularly as all the substances involved have a close biosynthetic relationship (Huovinen & Ahti, 1982). During this investigation it has been found that even chemistry may not, on very rare occasions, be sufficiently reliable to separate 'melaxantha'' and 'aurantiaco-atra' type thalli. TLC studies indicate that chemical data are not always correlated with minute morphological differences. Morphological similarities between the two chemical races are particularly appa- rent in material from the Magellan Straits area, where Races 1 and 2 overlap at the western limit of Race 2 and both races are often collected together. In particular Menzies' collections of 1787 from Isla de los Estados (Staten Island) include a diverse range of chemistries in an apparently morphologically and anatomically uniform gathering. These collections were widely distributed throughout herbaria and represent paratype material of Vsnea melaxantha (BM, E, H (ACH), LINN, PC, UPS, US). Out of 32 thalli examined by TLC one plant was depsidone deficient, 20 belonged to Race 1 and 11 to Race 2 although often lacking either norstictic or salazinic acid, but sometimes with traces of protocetraric acid. A further specimen from the same area (Castellanos 1542, BM!) only contained salazinic acid. Other examples of specimens with a mixed chemistry include a collection from Ushuaia (Henssen & Vobis 24 417a, MB !) in which two thalli contained salazinic acid in addition to the substances of Race 1, and a specimen from the Falkland Islands (R. I. L. Smith 2572, AAS!) which contained norstictic, salazinic, fumarprotocetraric and protocetraric acids. Examination of transverse sections from main branches of thalli in one of the Menzies collections (E), Fig. 19, representing both chemical races, are identical, with the axis occupying approximately half the branch diameter and with cortex and medulla widths 70-75 pm and 170-200 fjim respectively. An additional specimen from Isla de los Estados (H36372!) has a somewhat wider medulla and is more angular in transverse section and is characteristic of the so-called 'melaxantha' form. However, this thallus contains fumarprotocetraric acid (Race 1) instead of norstictic acid (Race 2). Surface ornamentation is frequently not so distinct in some of these specimens; there is a tendency in both races for the thallus to be smoother with less pronounced ornamentation, particularly towards the base, to occasionally become subnitid, but only rarely rupture to form annulations. Typically, fumarprotocetraric acid containing speci- mens (Race 1) have been described as being verrucose-papillate, rarely becoming scrobiculate- corrugated, whilst norstictic and salazinic acid containing specimens are described as being sparsely papillate becoming confluent and scrobiculate-corrugated (Lindsay, 1975). However, all these states are frequently found on the same thallus and are probably associated with its age and the degree of exposure. Slight difference in thallus colour has also led to the recognition of separate taxa that are here regarded as synonyms; for example Corniculariaflavicans. Likewise Motyka (1936) distinguished Usnea melaxantha from other taxa on the width and pinkish colour of the medulla, a feature that was subsequently found to be a storage artefact involving the breakdown of salazinic acid. Usnea aurantiaco-atra has most frequently been confused with U. ciliata and U. perpusilla. Apart from differences in distribution the species may be distinguished from U. ciliata by the ornamented surface, which lacks a waxy lustre and the typical pigmented annulations, and the frequent absence of excipular rays. The medulla of U. aurantiaco-atra is more compact than is usually found in U. perpusilla and the apothecia are frequently subterminal (Fig. 2) rather than lateral in series. The species may be distinguished from U. trachycarpa by differences in disc colour, abundance of fibrils and width of the medulla; and from U. taylorii by the latter's unique anatomy. 72 F. J. WALKER Fig. 19 Usnea aurantiaco-atra. a = Race 1, b = Race 2. Staten Land, February 1787, Menzies (E) X 1. USNEA SUBGENUS NEUROPOGON 73 Selected specimens examined Racel CHILE. Aisen: Coyhaique, Cerros Divisaderos (Cordon de Bella Vista), 1300 m, 13 November 1940, R. Santesson 6843 (S), 1400 m, 13 November 1940, R. Santesson 7355 (CANL 16965, S). Magallanes: Punta Arenas, Cerros Mina Rica, c. 500 m, 22 December 1940, R. Santesson 5244 (S, UPS); Tierra del Fuego, Porvenir, Morro Piedra, 300 m, 30 December 1940, R. Santesson 5398 (S); Isla Navarino, Puerto Navarino, 10 m, 28 February 1940, R. Santesson 1233 a & b (S); Cabo de Hornos, Hermite I., Forster's Peak, 1842, R. McCormick s.n. (BM); Tierra del Fuego, Canal Whiteside, Nose Peak, R. Santesson 5965, Lich. austram. ex Herb. Regnelliano 422 (BM, C, H, S, UPS). ARGENTINA. Chubut: Comodoro Rivadavia, 1924 'Euero 1 22 (BM). Tierra del Fuego: Sierra Alvear, S. slope, above Las Cotorras (c. 20 km ENE. of Ushuaia), 800-900 m, 7 February 1940, R. Santesson 639a (S, UPS); Monies Martiales, above Ushuaia, 10 March 1965, /. M. Lamb 8146 (FH, M); Beagle Channel, February 1963, E. E. Shipton s.n. (BM); Isla de los Estados, February 1787, A. Menzies s.n. [+ Races 2 & 3] (BM, E, H (ACH), LINN, PC, UPS, US). SOUTH GEORGIA. Cumberland West Bay, 2 km N. of Mt. Hodges, 28 February 1972, D. C. Lindsay 4285 (AAS, BM) ; Cumberland West Bay, near head of Sphagnum Valley, c. 150 m, 1 February 1961 , 5. W. Greene 1617 (AAS); Grytviken, 25 December 1909, C. A. Larsen s.n. (O, UPS). SOUTH ORKNEY IS. Signy I. : Factory Bluff, NE. , 19 January 1965, J. Price 9 (BM) ; Observation Bluff, 107 m, 6 February 1966, D. C. Lindsay 867 (AAS). Coronation I.: Foot of Wave Peak Buttress, Marshall Piedmark, 210 m, 20 August 1950, W. J. L. Sladen F.I.D.S. H602/3 (BM); Olivine Point, 30-60 m, 17 January 1965, R. I. L. Smith 156 (AAS, BM, FH). Laurie I. : Cape Geddes, 15-60 m, 13 February 1971, M. McManmon 140 (AAS). SOUTH SHETLAND IS. King George L: Admiralty Bay, Keller Peninsula, Mt. Flagstaff, 265 m, 9 January 1980, R. Ochyra 416/80 (M); Fildes Peninsula, 168 m, 16 February 1966, B. S. John & D. E. Sugden 25 (AAS, BM). Livingston I. : Robbery Beaches, Byers Peninsula, 75 m, 30 December 1965, D. C. Lindsay 518 (AAS). Half Moon I.: 75 m, 5 January 1966, D. C. Lindsay 576 (AAS, BM, FH). Robert L: English Strait, Copper Mine Cove, c. 60 m, 31 December 1934, 'Discovery' Exped. 14851 (BM). ANTARCTIC PENINSULA. Graham Land: Joinville I. [6315'S: 5545'W], Mt. Alexander, 200-500 m, 27 March 1981, R. I. L. Smith 3680 p.p. (AAS); Palmer Archipelago, Melchior Is., Omega [Lystad] I. [6420'S: 6256'W], 28 January 1951, A. Martinez s.n. (FH), 6-9 m, 14 March 1941, P. A. Siple U.S.A.S. 369 (US), Anvers I. [6445'S: 6405'W], Arthur Harbour, 20 m, 6 February 1965, /. M. Lamb & M. Zimmerman 8091 (FH), Wiencke I. [6448'S: 6325'W], N. buttress of Noble Peak, 225 m, 20 November 1944, /. M. Lamb Operation Tabarin 1790 p.p. (BM), /. M. Lamb Operation Tabarin 1789 (BM, CANL 16961, FH); Argentine Is., Galindez I. [6515'S: 6415'W], Woozle Hill, 35-50 m, 5 March 1981, R. I. L. Smith 3304 (AAS), summit, 1 January 1936, B.G.L.E. 1340 p.p. (BM, FH), Uruguay I. [6514'S: 6414'W], 15m, 25 October 1964, R. W. M. Corner 603 (A AS); Graham Coast, near Cape Tuxen [6744'S: 6825'W], 26 January 1961, K. Archibald 33 (AAS). Race 2 CHILE. Bio Bio: Type of Neuropogon antennarius (VER). Magallanes: Tierra del Fuego, Canal Whiteside, Puerto Yartou, Nose Peak, 700 m, 5 February 1941, R. Santesson 6820 (S); Cabo de Hornos, Hermitel., 1839-43, /. D. Hookers.n. (BM, E); Magellan Straits, Cerros Yartou, 8 March 1928, herb. Th. Fries (UPS). ARGENTINA. Tierra del Fuego: Cerro Garibaldi, SE. of Lago Escondido, between Ushuaia and Rio Grande, 1961, R. M. Schuster 58346 p.p. (CHR 342819, FH); Isla Navarino, Puerto Navarino, 10 m, 28 February 1940, R. Santesson 1232c (S); Canal Beagle, W. of Lapataia Bay, 1898, M. E. G. Racovitza 188 p.p. (herb. Vain. 344-TUR458). FALKLAND IS. East Falkland I.: Stanley, Sapper Hill, 4 April 1965, /. Price s.n. (BM), Stanley, 30 January 1946, /. M. Lamb Vezda: Lich. Sel. Exs. 496 (BM, M, MB); Mt William, near Stanley, near sea-level, 30 January 1946, /. M. Lamb Operation Tabarin 2891 (CANL 17185), Operation Tabarin 2888 (BM); Mt. Longdon, 174 m, 2 April 1967, D. C. Lindsay 1659 (AAS, H); between Darwin and Goose Green, 6 m, 17 September 1963, R. W. M. Corner 53 (AAS, FH); N. of Wireless Hill, 100 m, 28 February 1977, R. I. L. Smith 2574 (AAS). West Falkland Is.: Roy Cove, 1909-11, E. Vallentin 88 (BM). SOUTH GEORGIA, c. 1 km SE. of Brocken, 180 m, 23 January 1972, D. C. Lindsay 4036 (AAS); E. of Swinhoe Peak, between Hamburg Lakes and Hamburg Glacier, 690-695 m, 6 November 1976, R. I. L. Smith 2519 (A AS). Race3 CHILE. Bio Bio: Antuco, Reynolds 141 (BM). Cautin: Andes de Villarrica, 1891, Neyer s.n. (M). Magallanes: Cape Spencer, /. D. Hooker s.n. (E); Cabo de Hornos, Hermite L, St. Martin's Cove, R. 74 F. J. WALKER McCormick s.n. (BM); Cabo de Hornos, Voyage of H.M.S. Adventure and Beagle 1826-30, King s.n. (BM). ARGENTINA. Neuquen: Parque Nacional Lanin, N. of Lago Lacar, Cerro Malo, c. 1900 m, 28 January 1968, /. H. de Haas 1290 (U 3136735). Rio Negro: Cerro Rigi, near Lago Frias, c. 1780 m, 15 February 1950, /. M. Lamb 6045 (CANL 16962, UPS), 1660 m, /. M. Lamb 6043 (CANL 16964, FH, H, US). Tierra del Fuego: Fuegia med., Cerro Milladeo, 850 m, 17 January 1929, H. Roivainen s.n. (H). SOUTH GEORGIA. Head of Sphagnum Valley, near Echo Pass, Cumberland West Bay, 15 January 1961, E. A. Coleman Greene 1071 (AAS,BM,FH); Wilson Harbour, 19 January 1927, 'Discovery' Exped. s.n. (BM). SOUTH ORKNEY IS. Signy I. : Hill between Factory Cove and Paal Harbour, 105 m, 20 December 1961 , M. W. Holdgate 225 (AAS). Coronation I.: Saunders Point, 0-9 m, 8 October 1950, W. J. L. Sladen F.I.D.S. H 612/21 (BM). SOUTH SHETLAND IS. King George L: Keller Range, 21 February 1947, D. Nicholson & A. Reece Vezda: Lich. Sel. Exs. 972 (BM). Livingston I. : New Plymouth, Byers Peninsula, 19 m, 14 December 1965, D. C. Lindsay 366 (AAS). Nelson L: Harmony Cove, 2 December 1954, O. Kuhnemann 35075 (FH). ANTARCTIC PENINSULA. Graham Land: Joinville I. [6315'S: 5545'W], Mt. Alexander, 250 m, 27 March 1981, R. I. L. Smith 3682 (AAS); Palmer Archipelago, Wiencke I. [6448'S: 6325'W], N. buttress of Noble Peak, 225 m, 20 November 1944, /. M. Lamb Operation Tabarin 1789 (BM); Argentine Is., Galindez I. [6515'S: 6415'W], above Stella Creek, 24 m, 21 April 1964, R. W. M. Corner 542 (BM), Corner I. [6516'S: 6414'S], 6 m, 18 August 1964, R. W. M. Corner 566 (AAS), Graham Coast, Loubat Point [6504'S: 6356'W], Lemaire Channel, 300 m, 19 March 1964, J. Clennell Kennett 56 (AAS, BM). BOUVET0Y. Rustadkollen SW., 323 m, 7 March 1979, T. Engelskj0n s.n. (BG, BM). Races 1+2 (see p. 71) ARGENTINA. Tierra del Fuego: Parque Nacional Tierra del Fuego, 'Weg zum' Glaciar Martial, c. 500 m, 7 December 1973, A. Henssen & G. Vobis 24 417a (MB). FALKLAND IS. East Falkland L: to N. of Wireless Hill, 28 February 1977, R. I. L. Smith 2572 (AAS). The following may be consulted for further localities: South America (Lamb, 1948a), South Georgia (Lindsay, 1975), South Orkney Is. (Smith, 1973), South Shetland Is. (Lindsay, 197 la), Antarctic Peninsula (Lamb, 1964); together with lists held in BM and collections in AAS and BM. 5. Usneaciliata(Nyl.)DuRietz Figs 20-21 inSvensk. hot. Tidskr. 20: 91 (1926).-Neuropogonmelaxanthusvsir. aV/atoNyl.in/. Linn. Soc. (Bot.) 9: 245 (1866). - Neuropogon ciliatus (Nyl.) Krempelh. in Ver. zool.-bot. Ges. Wien. 18: 313 (1868). - Usnea melaxantha var. ciliata (Nyl.) Mull. Arg. in Bull. Soc. r. Bot. Belg. 31(2): 26 (1892). Type: New Zealand, Tarndale, Nelson Mountains, January 1861, Dr A. Sinclair. (H-N YL 36365 ! -holotype; BM!, E! - isotypes). [TLC: norstictic acid, salazinic acid, protocetraric acid, usnic acid.] Note: The combination Usnea ciliata (Mull. Arg.) Vainio (Vainio, 1909) was invalidly published in synonymy (Article 34.1) and consequently does not predate the name U. ciliata (Nyl.) Du Rietz. Further, Vainio, in later publications (Vainio, 1913, 1915, 1923), made it clear that he intended the name to be published as U. trichoidea * U. ciliata, thus making a combination below specific rank. Description: Thallus (-3)-5-10(-12) cm, arising from a broadly proliferating, pigmented holdfast, erect, monopodial or subdichotomous, moderately branched above, fibrils absent, lacking extended secondary branches. Branches terete, yellow-green, continuously pig- mented violaceous black towards the apices. Cortex thick. Surface smooth, or rarely sub- faveolate, waxy, epapillate, with conspicuous black-edged annulations. Medulla compact, axis thick, occupying c. 0-5 of the branch diameter. Soredia, pseudoisidia and isidia absent. Apothecia frequent, subterminal, geniculate with a subtending spur, or additionally lateral; cupular or plane becoming reflexed on maturity. Disc black, excluded margin, excipulum smooth or subfaveolate, marginal rays limited in number, stout, branch-like, pigmented. Pycnidia not seen. TLC: norstictic acid, salazinic acid, protocetraric acid, usnic acid. Distinguishing features: Usnea ciliata is characterised by its erect habit, arising from a proliferating holdfast, and a monopodial to moderately branched thallus, often with violaceous black pigmentation, a smooth, waxy, black-annulate surface, a compact medulla containing USNEA SUBGENUS NEUROPOGON 75 Fig. 20 Usnea ciliata. Isotype of Usnea melaxantha var. ciliata Nyl. (BM) x 1. norstictic and salazinic acids, a thick axis, and subterminal, geniculate apothecia with a black disc and conspicuous excipular rays. Distribution: Usnea ciliata is confined to Australasia and is only known with certainty from New Zealand, where it has a similar, but slightly more restricted distribution, to U. acromelana, often absent at lower altitudes. Its presence in the North Island has now been confirmed by recent collections made by Bartlett (p. 22), although the species is undoubtedly rare and less frequently fertile there; occasionally Alectoria nigricans has been mistaken for such sterile thalli (for example, Mt. Ruahine, Colenso 11 64 A WELT!). The species is also rare on Stewart Island. Tasmanian records remain uncertain and are either based on damaged material lacking either apothecia or soredia (for example, Bratt, CHR342744!), or have not been traced (for example, Dodge, 1948) . However, it is possible that U. ciliata does rarely occur in Tasmania, although this cannot be verified in the absence of fertile material. Lamb (19390, 1948a, 1959) cited several collections of Usnea ciliata from South America. Many of these have been examined and are less typical, almost smooth, forms of either Usnea 76 F. J. WALKER Fig. 21 Usnea ciliata. New Zealand, Otago, Rock & Pillar Range, February 1967, Galloway (BM). Left. Detail of cortical annulations xlO. Right. Detail of apothecium xlO. aurantiaco-atra or, alternatively, U. perpusilla, in which the medulla is not characteristically lax. In both instances the apothecia bear some excipular rays and any annulations present are of a spurious nature, merely the result of weathering. The specimen figured by Lamb (19390 pi. 7 fig. 13, herb. Vainio 344 TUR 458!) is U. aurantiaco-atra (Race 2). Fig. 6. Chemistry: To date only one chemical race, that containing norstictic and salazinic acids, is known from the large number of collections examined. Variation: Usnea acromelana should be referred to (p. 50) for the account of variation in branching, pigmentation, morphology and anatomy, which may, to a similar extent, be influenced by altitude. However, U. ciliata does not exhibit such a wide range of variation as U. acromelana. The degree of violaceous black pigmentation is often extensive in more richly branched forms, but in extreme instances it may only be present in the apothecial disc, annulations and branch and ray apices. The thalli are often taller than those of Usnea acromelana and tend to show less variation in the proliferating nature of the holdfast. Small, tufted, densely branched forms, with fine, tapering secondary branches may be distinguished from U. subcapillaris by their erect habit and shorter, stouter branches (see p. 106). Apothecia are usually subterminal and conspicuously geniculate with a subtending spur that may be longer than the excipular rays. Additional, lateral, apothecia may sometimes be produced which tend to be subsessile. Their size is variable, as is the presence, number, length, and occasional branching, of excipular rays, which are not as extensively produced as in Usnea trachycarpa. Young apothecia are cupular, becoming expanded or sometimes convex at maturity, when they are irregular to crenulate. Only in rare instances is the black pigmentation of the disc more or less lacking or not fully developed. The excipulum is usually smooth but may sometimes be subfaveolate, thus reflecting the subtending branch morphology; the ma r " : - often excluded. USNEA SUBGENUS NEUROPOGON 77 Species concept: Usnea ciliata is a comparatively uniform species which is considered to be the primary species related to the sorediate U. acromelana. Although it is closely allied to U. subcapillaris , only in very rare instances are thalli encountered that appear to be intermediate between the two taxa. U. ciliata may be distinguished from U. subcapillaris by its erect rather than subdecumbent habit, the coarser, shorter secondary branches, and subterminal, rather than lateral, apothecia. Usnea ciliata may be separated from other fertile species by distribution, although it is readily distinguishable on morphological characters. Selected specimens examined NEW ZEALAND. North Island. Gisborne: Mt. Hikurangi, c. 1740 m, 14 November 1983, 7. K. Bartlett 25965 (BM). Hawke's Bay: Mt. Kaweka, c. 1840 m, 9 November 1983, 7. K. Bartlett 25960 (BM),7. K. Bartlett2596l (BM). Wellington; Mt. Ruapehu, Turoa Site, 1740 m, 6 July 1969, D. J. Galloway s.n. (CHR 342790, CHR 343281, CHR 343292, CHR 343339); near Wellington [? Tararua Range], 7. Buchanan s.n. (BM, GLAM NHB 1927-8-347); North Ruahine Range, c. 1000 m, 17 October 1983, J. K. Bartlett 26688 (herb. Bartlett). South Island. Nelson: Mt. Cobb, c. 1740 m, 19 December 1982,7. K. Bartlett 24124& (herb. Bartlett, BM); near Dunn Saddle, 1460 m, 12 January 1983, 7. K. Bartlett 25962 (BM); Kakapo Peak, c. 1520 m, 16 December 1982,7. K. Bartlett 26253 (herb. Bartlett, BM),7. K. Bartlett 25812 (herb. Bartlett, BM); Mt. Robert, Lake Rotoiti, 1430 m, 16 January 1960, D. Scott 426 (BM); St. Arnaud Range, 1680 m, 22 December 1967, A. F. Mark s.n. (CHR 343290 p.p., CHR 343321); Mt. Technical, above Lewis Pass, 14 January 1979, D. J. Galloway s.n. (CHR 343160 p.p.); Mt. Aorere,7. K. Bartlett s.n. (CHR 343235); Lake Sylvester, 1620 m, 18 December 1967, A. F. Mark s.n. (CHR 342797, CHR 342798). Marlborough: Mt. Black Birch, 1220 m, 1 January 1969, B. V. Sneddon s.n. (BM, CHR 342822); Blue Mtn, near head of Waihopai River, c. 1830 m, 1934, W. Martin s.n. (CHR 375942); Upper Awatere Valley, Shingle Peak, 1310m, 6 January 1970,A F. Marks.n. (OTA 27124); Awatere, Mt.Harkness (collector unknown) (CHR 160678); Inland Kaikoura Range, Mt. Tapuaenuku, 20 March 1934,7. 5. Thomson 1523 p.p. (CHR 343806 p.p.), 2700 m, August 1969, P. Lusk s.n. (CHR 343309), February 1961, B. C. Aston s.n. (WELT L193), 1859, Sinclair s.n. (BM); Inland Kaikoura Range, Mt. Mitre, 27 January 1954, R. Mason & D. R. McQueen s.n. (CHR 160686 p.p.); Upcot, February 1916, B. C. Aston s.n. (WELT L190). Canterbury: Torlesse Range, Foggy Peak, 1680 m, 18 December 1962, P. W. James 1918 p.p. (BM), 12 November 1972, G. C. Bratt 72/1880c (HO 35166); Mt. Torlesse, March 1934, 7. 5. Thomson 1618 (CHR 343800); Four Peaks Range, Blue Mountain, c. 1640 m, 23 April 1979, D. J. Galloway s.n. (CHR 343225); Mt. Peel, 1740 m, January 1972, D. J. Galloway s.n. (BM, CHR 343437), c. 1680 m, H. H. Allan 2 (UPS); Mt. Peel, Rangitata River, 1520 m, 6 May 1960, D. Scott 458 (OTA); Kirkliston Range, 1680-1830 m, 25 March 1978, D. J. Galloway s.n. (CHR 343254); Two Thumb Range, Mt. Richmond, November 1968, A. F. Mark s.n. (CHR 343302); Two Thumb Range, Mt. Dobson, 1830 m, 15 January 1959, D. Scott 302 (OTA); Craigieburn Range, 1520 m, October 1968, L. J. W. Strung s.n. (CHR 342783); Temple Basin, Arthur's Pass, 21 February 1943, V. D. Zotov s.n. (CHR 160683); Cass, Day Creek, 1070 m, 6 February 1936, M. Sutherland s.n. (WELTL194); Ben Ohau Range, Glen Lyon Station, 1830m, October 1958, Mason 161 & 162 (OTA); Porter's Pass, February 1874, 910 m, 5. Berggren s.n. (UPS); Banks Peninsula, Castle Rock, 460 m, October 1967, P. F. Johnson s.n. (CHR 342785). Otago: Rock and Pillar Range, 1160-1190 m, 18 September 1981, F. J. Walker s.n. (BM, UPS), c. 1020 m, 4 April 1970, G. Degelius NZ-346 (herb. Degelius), 2000 m, 1 December 1969, N. M. Adams s.n. (WELT L359); Old Man Range, 1370 m, 1 February 1963, P. W. James 1576 (BM), 1680 m, April 1968, D. J. Galloway s.n. (BM, CHR 343341); Mt. Sir William, 2560 m, 27 December 1970, D. J. Galloway s.n. (CHR 34278 p.p.); Humboldt Mtns, Mt. Nox, 1950 m, 31 December 1969, D. J. Galloway s.n. (CHR 342791 p.p.); Remarkables, 1220 m, February 1968, D. J. Galloway s.n. (CHR 343334); Matukituki Valley, Mt. Avalanche, 1680 m, 5 April 1969, L. D. Kennedy s.n. (CHR 342756); Mt. Maungatua, 910 m, March 1954, 7. Murray 554 (OTA); Lake Ohau, Mt. Sutton, 1070 m, May 1958, J. Murray 1783 (BM); Mt. Roy, 1 November 1972, G. C. Bratt 72/1507 (HO 35174) ;Mt. Pisa, 1860m, March 1968, D. 7. Galloway s.n. (CHR 342799) ; Silverpeaks, Gap Ridge, 610m, 19 March 1961, D. J. Galloway s.n. (CHR 342754). Southland: Mid Dome, 1520 m, May 1970, G. Van Reenen s.n. (CHR 342837 p.p.); West Dome, 1070 m, 11 January 1970, P. N. Holdsworth s.n. (CHR 342806); Ridge between Takahe Valley and Ettrick Burn, 1400 m, 16 February 1969, G. Van Reenen s.n. (CHR 342818); Thomson Mtns, 1370 m, October 1967, A. F. Mark s.n. (CHR 342813); Grave-Talbot Pass, Milford Track, 1830m, 25 January 1963, O. Fletcher s.n. (CHR 160684). Stewart I. : Mt. Anglem,992 m, February 1966, D. J. Galloway s.n. (CHR 342749). For further localities in Canterbury and Otago see lists held in BM and collections in CHR, HO and OTA. 78 6. Usnea durietzii Motyka F. J. WALKER Fig. 22 Lich. Gen. Usn. Stud. Monogr. 2: 503 (1937). Type: [Chile] W. Lechler: Plantae Magellanicum, Insula Elizabethae [5407'S: 7311'W] Oct. med. 1882 (UPS! - holotype; BM!, M!, PC! - isotypes). [TLC: norstictic acid, salazinic acid, usnic acid.] Description: Thallus 2-3(-4) cm, arising from a single, elongated, blackened holdfast, erect, ' richly branched, often 0-5-1-0 cm above the base, main branches clustered, fibrils absent, with short, recurved, capillaceous secondary branches. Branches terete, greenish yellow, articu- late, conspicuously inflated above a constricted base, black-pigmented only at the apices. Cortex thin. Surface matt, coriaceous, usually epapillate. Medulla broad, very lax in main branches, axis thin, occupying 0-2-0-4 of the branch diameter. Pseudoisidia numerous on all branches, in soralia-like clusters, small (x 10 lens), spinulose, black-pigmented, often becoming confluent or eroding. Isidia and soredia absent. Apothecia and pycnidia not known. TLC: (1) norstictic acid, salazinic acid, usnic acid; (2) no medullary substances, usnic acid. Fig. 22 Usnea durietzii. Top. Holotype of Usnea durietzii Motyka (UPS) xl. Bottom. Detail of pseudoisidia. Peru, Ancash, July 1979, Gibby & Barrett (BM) xlO. USNEA SUBGENUS NEUROPOGON 79 Distinguishing features: Usnea durietzii is characterised by its erect, often stalked, tufted habit, with short, irregular laterals and inflated, articulated main branches; a smooth, matt surface with numerous clustered, pigmented pseudoisidia which give a sooty appearance to the thallus; and a lax medulla, usually containing both norstictic and salazinic acids. Distribution: Usnea durietzii has a rather disjunct, predominately western distribution throughout the Andean Cordillera from Panama and northern Venezuela to the Magellan Straits and Tierra del Fuego, although the species is very rare in Chile and Argentina, virtually being replaced there by U. patagonica. The species is characteristic of dry, exposed, rather than necessarily alpine, habitats. It is usually confined to high altitudes, c. 3000-4000 m, rarely occurring at lower levels towards the southern limit of its distribution as , for example , in the type locality. It has previously been recorded from the east side of the Andes in Prov. Mendoza, Argentina (Rasanen, 1939), and an illustration of Usnea cf. condensata by Asahina (1967) from Patagonia is clearly this species. In southern Patagonia U. durietzii has been collected with U. patagonica and U. trachycarpa (Santesson, 525 S!, Santesson 1920, S!). Fig. 8. Chemistry: Usnea durietzii is characteristically found to contain norstictic and/or salazinic acids (Race 1); depsidone-deficient populations (Race 2) are rare and are apparently confined to northern part of the range. Variation: Usnea durietzii is a rather variable species with a wide range of growth forms resulting from the irregular, repeated branching which occurs at a short distance above the holdfast. Frequently there is a proliferation of ultimate branches producing an untidy, tufted, 'witches-broom'-like structure consisting of numerous recurved branchlets with pigmented apices. Sometimes the basal black, stalk-like structure is absent and the thallus proliferates directly with a cluster of laterals. Fibrils, such as occur in U. amblyoclada (Appendix I, p. 115) and U. trachycarpa, are absent. Pigmentation is confined to the thallus base and apices and to the pseudoisidia. The degree of development of pseudoisidia is variable and they may be incompletely developed or erode to form soredia. The pseudoisidia are always black-pigmented and tend to be produced in soralia-like clusters dispersed throughout the thallus. Such pseudoisidia are larger than those of Usnea patagonica and consequently can be distinguished from the flatter, partially corticate structures that are sometimes produced in other asexual species. Papillae are usually absent although they rarely occur on main branches in those specimens that are of uncertain taxonomic position, which seem to have affinities with Usnea amblyoclada or U. sphacelata. In these specimens the medulla is not so characteristically lax as in U. durietzii. The holotype specimen (Fig. 22) is a grotesque, knarled form of the species with proliferating branchlets and eroded pseudoisidia, although it is still characterised by a lax medulla, matt surface and tufted habit. Isotype collections are more typical. As the species often grows in exposed situations some features in consequence may not be fully developed as, for example, laxness of the medulla or development of pseudoisidia. Species concept: Usnea durietzii is here included in the subgenus by virtue of habitat, distribution, and pigmentation. Fertile 'material of the taxon, or the recognition of a fertile counterpart, is required in order to confirm this assumption. This is a well-defined species, which may have affinities with the subgenus Usnea, for example with U. amblyoclada and U. nigropapillosa. The distribution overlaps that of U. amblyoclada from which it may be distinguished by the lack of fibrils and pseudocyphellae and also by the presence of clustered pseudoisidia rather than scattered true isidia. It also occurs with U. bogotensis which is a much more robust, unpigmented species with conspicuous true soralia that are only rarely pigmented, a compact medulla, and prominent white annulations and reticulations. Usnea durietzii has some affinities with Usnea patagonica but may usually be distinguished by habit, surface features, propagule size, and medullary chemistry (Table 3). The presence of pigmented pseudoisidia, rather than pale true isidia, distinguish the species from U. acanthella. It may be separated from other asexual species within the subgenus by differences in soralia and surface characters, and often by habitat and distribution. 80 F. J. WALKER Specimens examined Race I PANAMA. Chiriqui: Chiriqui, volcano crater's edge, 3300-3450 m, 12 December 1948, P. F. Scholander 169534 (US, CANL 17249). VENEZUELA. Merida: Valle del Mufafi, paramo de Mucuchies, Sierra Nevada, 3650-3750 m, 9 April 1975, M. E. Hale & M. Lopez Figueiras 44650 (US); trail between Laguna Negra and Mucubaji, Sierra Nevada, 3500 m, 8 April 1975, M. E. Hale & M. Lopez Figueiras 44455 (US); Rangel, Sierra Nevada de Santo Domingo, paramo de Mucubaji, close to Laguna Grande (Laguna de Mucubaji), c. 3500 m, 11 October 1981, M. Lindstrom 526 (GB); [unrealised] , 5500 m, 1846, Voyage de Funck & Schlim 987 (PC). COLUMBIA. Santander: Paramo del Almorzadero, 3970 m, 17-19 November 1978, /. Aguirre 1090 (U 387342b). Boyaca: Paramo de La Rusia, Laguna Negra, 3745 m, 14 December 1912, A. M. Cleef7229 (BM); Sierra Nevada del Cocuy, Alto Valle Lagunillas, SE. of La Laguna Cuadrada, 4080 m, 26 November 1972, A. M. Cleef& P. A. Florschutz 5558 (U 390939), 5559 (U 390940). ECUADOR. Pinchincha: Volcan Sincholahua, plateau along Rio Pita, c. 3600 m, 30 May 1973, L. Holm-Nielsen et al. 6655 (GB); Cerro Iliniza, c. 4400 m, 7 March 1972, L. Arvidsson & D. Nilson 945 (BM, GB). Cotopaxi: Cotopaxi National Park, c. 60 km SE. of Quito, Pampa de Limpios, 3900 m, 13 June 1976, S. R. Gradstein, H. Sipman & T. de Vries 164 p.p. (BM, U). PERU. Ancash: Quebrada Honda, Cordillera Blanca, 4260 m, July 1979, M. Gibby&J. Barrett s.n. (BM). Junin: Huytapallana, above Huancayo, c. 4600 m, 19 May 1973, B. Mullins 25 (BM); Cerro de Pasco, Conception, 4260-4560 m, 1969, B. Mullins s.n. (BM). Cuzco: Huaya, Huaya Pass, above Cuzco, 4500m, 14 June 1973, B. Mullins 29 (BM); Pisac, near Cuzco, 3700 m, 14 April 1973, B. Mullins 24 (BM); above Cuzco, 3500-3600 m, 16 June 1973, B. Mullins 30 (BM). Ayacucho: near Ninabamba, 3900 m, 25 May 1973, B. Mullins 27 (BM) Apurimac: between Piccas and Huancaramay, 1932, M. M. Woods, n. (BM). Puno: prope Azangaro, June 1854, W. Lechler PI. Peruvianae 1758 (BM, M). BOLIVIA. La Paz: Copacabana, near Lake Titicaca, 3800 m, 10 July 1973, B. Mullins 31 (BM). CHILE. Malleco: c. 32 km SW. of Angol, 16 February 1958, W. J. Eyerdam 10794 (BM, FH). Magallanes: Is. Grafton, I. Elisabeth [5407'S: 73H'W] (type locality); prope Sandy Point, W. Lechler PI. Magellani- cae 108 (M); Rio de los Cruzeros, 60 km NNE. of Punta Arenas, 26 April 1940, R. Santesson 1920 p.p. (S). ARGENTINA. Mendoza: Las Heras, Quebrada del Portero Puerta, near Cacheuta, 1800 m, June 1939, R. Leal s.n. (LD). Tierra del Fuego: Ushuaia, small island in Beagle Channel off the town, 30 January 1940, R. Santesson 532 p.p. (S). Unlocalised: 'Jahina', Peruviae, W. Lechler PI. Magellanicae 139 (M). Race 2 VENEZUELA. Merida: [unrealised], 1842, Funck & Schlim 9 & 7 (BM, PC); Hale & Lopez Figueiras 44650 p.p. (as Race 1) (US); Rangel, Sierra Nevada de Santo Domingo, paramo de Mucubaji, close to Laguna Grande (Laguna de Mucubaji), c. 3500 m, 11 October 1981, M. Lindstrom 519 (GB); Paramo de Mucubaji, on highest point of road Merida-Caracas, track from Laguna de Mucubaji to Laguna Negra, 3500 m, 24 January 1979, H. Sipman & M. Lopez Figueiras 11289 (U). COLUMBIA. Cundinamarca: prope Bogota, /. Weir 12 (BM). BOLIVIA. Larecaja: near Sorata, Rampe de 1'Apacheta, 4500 m, 1857, G. Mandon PL Andinum Boliviensium 1737 (BM, PC). 7. Usnea neuropogonoides Motyka Fig. 23 Lich. Gen. Usn. Stud. Monogr. 1: 73 (1936). Type: Argentina: terr. Santa Cruz, Rio Fosiles, c. 1000 m, in rupibus, May 1905, P. Dusen. (UPS! - holotype). [TLC: no medullary substances, usnic acid.] Description: Thallus up to 13 cm long, suberect to scrambling-prostrate, often lacking a distinct holdfast, divergent, dichotomous, infrequently branched to form a lax, spreading habit, rarely with fibrils, with short, divaricate, capillaceous secondary branches with attenuate, subcornute apices. Branches terete, occasionally subinflated or slightly angular, yellow- green, minutely variegated or black-pigmented only at the apices. Cortex c. 100 /am. Surface matt, smooth to faveolate or papillate, often fracturing. Medulla broad, lax, or sublax, axis thin, occupying c. 0-25 of the branch diameter. Apothecia, pycnidia, soredia, pseudoisidia and isidia not known. TLC: (1) psoromic acid, 2'-O-demethylpsoromic acid, fatty acids (murolic acid complex), usnic acid; (2) depsidone deficient, fatty acids (murolic acid complex), usnic acid. Distinguishing features: Usnea neuropogonoides is characterised by its lax , suberect to scramb- ling habit, divergent branching, scant pigmentation, faveolate to papillate surface, lax medulla often containing psoromic acid, and lack of sexual or vegetative propagules. USNEA SUBGENUS NEUROPOGON 81 Fig. 23 Holotype of Usnea neuropogonoides Motyka (UPS). Top. Whole thallus x 1. Bottom. Detail of branching xlO. 82 F. J. WALKER Distribution: Usnea neuropogonoides is only known from a few collections from Patagonia, and is apparently characteristic of the windswept mesetas to the east of the Andean Cordillera. Fig. 8. Chemistry: From the few collections (six) examined it appears that thalli more frequently contain psoromic acid (Race 1) than lack depsidones. Traces of two or three fatty acids of the murolic acid complex were found in some thalli. Variation: It is difficult to assess the extent of variation in this species from the few extant collections. The suberect or sprawling habit and loosely interwoven branches with pigmented apices appear to be constant. Pigmentation is scarce on the main branches, being confined to fractures and, rarely, papillae, thus bearing resemblance to the rare, straggling forms of Usnea aurantiaco-atra. Although fibrils are infrequent, the surface faveolation and papillation, the sublax medulla, and the thin axis are reminiscent of some of the variation found in U. trachycarpa. The lack of reproductive structures suggests that this species is possibly disseminated by thallus fragmentation or by abrasion of fibrils or papillae. One specimen from Fuegia, 'San Isideo Point' (BM!), cited by Crombie (1876ft), possibly belongs to this species, but is somewhat moribund and has a partially eroded cortex which could indicate the incipient production of isidia or pseudoisidia. Occasionally, in other thalli, nodular, white, soralia-like outgrowths are formed on main branches as a result of cortical damage or thallus fracture; these are anomalous structures and do not appear to have any taxonomic significance. Species concept: Usnea neuropogonoides is included in Neuropogon since, although sterile, it has several characteristic features of the subgenus. Examination of further material may prove that this is an extreme form of another species, the most likely being U. trachycarpa, or that it even belongs to the subgenus Usnea, only producing pigmentation in very exposed habitats or when moribund as is seen in U. torulosa (p. 120). The presence of papillae and depsidones rather than depsides (i.e. divaricatic acid) and different habitat requirements, indicate that although superficially similar, this species should not be included in Protousnea as delimited by Krog (1976). In addition, the main branches remain virtually terete and do not become so markedly angular in section or inflated at branch points, as, for example, in P. scrobiculata. However, in spite of this, certain similarities do exist, notwithstanding the limited morphological characters, which also define Protousnea. Usnea neuropogonoides can be distinguished from decumbent forms of other Neuropogon species by the matt, faveoiate to papillate surface, thin axis, broad, sponge-like, more or less lax medulla, and lack of propagules. Specimens examined Race I ARGENTINA. Santa Cruz: Lago San Martin, 4 February 1903, P. Dusen s.n. (H); between Lago San Martin and Lago Viedma, 2 February 1903, Spegazzini s.n. (BM, H); Chilean border, Lago San Martin, Laguna Theuen Aiken, [48-49S], 4 February 1903, C. Hoberg 121 (BM); Estancia Librun, [c. 51S: 71E], Shipton Exped. 1958-59, P. W. James s.n. p.p. (BM). Race 2 ARGENTINA. Santa Cruz: Belgrano Pass, 22 December 1908, C. Skottsberg (UPS); Estancia Librun (as Race 1), P. W. James s.n. p.p. (BM). 8. Usnea patagonica F. J.Walker, sp. nov. Fig. 24 Diagnosis: Thallus fruticosus, pigmento nigro in cortice ad apices basimque suffusus. Rami extensi teretes saepe fibrillosi, papillis sparsis vel numerosis. Medulla sublaxa. Soralia pseudoisidiis minutis nigrescentibus excavata. Apothecia ignota. Acidum usnicum continens, et acidia aliphatica ad acidum murolicum pertinentia. Typus: Argentina, Santa Cruz, Patagonia, Lago Argentine [4980'-51S: 72- 7330'W], Cordillera Cristales, 2000 m, on boulder in valley, 26 December 1958, P. W. James 31 (Shipton Exped.) (BM! -holotype; UPS!, US! -isotypes). [TLC: 3 fatty acids of the murolic acid group, usnic acid.] Description: Thallus l-5-3(-4-6) cm, arising from a delimited, pigmented holdfast, erect, subdichotomous, richly branched a short distance above, with numerous, clustered, inter- USNEA SUBGENUS NEUROPOGON 83 HoLO Type Specimen Fig. 24 Holotype of Usnea patagonica F. J. Walker (BM). Top. Whole collection x 1. Bottom. Detail of eroded soralia with pseudoisidia x 10. woven, extended lateral branches, fibrils and capillaceous, flaccid ultimate branches. Branches terete, greenish yellow, black-pigmented only at or towards the apices. Cortex thin, 50-75 fjim. Surface subnitid, subfaveolate-papillate (xlO lens) then scabrid, occasionally rupturing. Medulla lax or sublax, at least towards axis, axis thick, occupying c. 0-5-0-6 of the branch diameter. Soralia extensive throughout thallus, plane, irregular or ulcerose, rarely becoming effuse to confluent and excavate, arising from small papillae, an inconspicuous margin, with numerous, minute (xlO lens), pigmented pseudoisidia. Soredia rare, true isidia absent. Apothecia and pycnidia not known. TLC: no medullary substances, 1-^ fatty acids (murolic acid complex), usnic acid. Distinguishing features: Usnea patagonica is characterised by its erect, richly branched, touselled habit with extended, lateral branches which rarely bear fibrils; a subnitid, subpapillate surface with numerous, minute, pigmented pseudoisidia, produced in ulcerose soralia, which give the thallus a dusted, sooty appearance, and a sublax medulla, frequently containing fatty acids. 84 F. J. WALKER Distribution: Usnea patagonica is mainly confined to the southern half of South America and is a species of the transitional arid-montane zone, frequently between the altitudes 500-1500 m, often occurring with U. trachycarpa and, less frequently, with U. durietzii or even U. perpusilla, having a slightly more eastern distribution. U. patagonica has been most frequently collected from the region of the Patagonian Lakes, for example Lago Argentino, but is rare further north, and is known from a single collection from Bolivia. Specimens cited by Follmann (19650) from central Chile also appear to belong to this species. A single collection from South Africa is tentatively assigned to this species. Fig. 8. Chemistry: All specimens examined were depsidone-deficient, containing usnic acid and usually one to four fatty acids, of Rf classes TDA 2-4, HEF 3-4, belonging to the murolic acid complex, which also occurs in Usnea trachycarpa. The presence of these fatty acids and lack of a depsidone-containing race may be regarded as additional characters for recognising U. patago- nica as a distinct species. Traces of the UV+ unknowns are occasionally present. Variation: The general form of Usnea patagonica shows little variation apart from the extent of production of secondary branches and extended laterals, which usually replace fibrils. In a single gathering thalli may either bear numerous, capillaceous, ultimate branches or may entirely lack laterals and instead have truncated apices. Thalli are characteristically repeatedly branched, sometimes a short distance above the blackened holdfast, giving a tufted, tasselled habit. The surface may be subnitid to waxy, and in smaller, immature thalli, is occasionally devoid of papillae which lack pigmentation. Rarely annulations are formed resembling those of Usnea acromelana, although in U. patagonica this is usually confined to the finer, laterals rather than main branches. The laxness of the medulla may vary, but even when not well developed, the hyphae are more loosely interwoven in the proximity of the axis. Such variation is similar to that found in Usnea trachycarpa; the primary branches do not become inflated as in U. durietzii. There is little variation in the form of the soralia; the production of small, pigmented pseudoisidia is a characteristic feature of the species. The pseudoisidia are rarely as well- developed as in Usnea durietzii and arise in more or less delimited, discrete soralia which rarely become confluent as in that species. Occasionally, soralia may resemble the less well-developed forms found in U. antarctica, but generally they are more eroded in U. patagonica and lack the distinctive crateriform margin. Table 3 summarises the differences between these three species and Fig. 3 illustrates their asexual propagules. Black pigmentation is normally confined to the holdfast, pseudoisidia, and branch apices, although the ultimate branches may rarely be variegated or extensively pigmented. High altitude material from Cedarberg, South Africa (Schelpe 1961, BOL!, CANL 16944!), originally determined by Lamb as Usnea acromelana, very closely conforms to U. patagonica, and is tentatively included in this species. Some thalli have a blackened base, are richly branched above with pigmented apices, and have excavate, ulcerose, soralia-like areas which contain small, pigmented pseudoisidia. Thalli also lack diagnostic medullary substances, containing traces of three to five fatty acids of the murolic acid complex. Other thalli in the same collection do not resemble U. patagonica so closely, and may belong to another taxon, since they are less extensively pigmented, have more fibrils, a markedly faveolate surface, and a laxer medulla lacking fatty acids. However, they do possess similar soralia-like areas that, less frequently, produce poorly developed pseudoisidia. Species concept: Many specimens of Usnea patagonica have been previously determined as U. acromelana or U. durietzii according to the extent of pigmentation, branching, or development of pseudoisidia. In many respects this species might be regarded as an intermediate entity between several asexual species that occur in South America, and is, for these reasons, included within the subgenus despite the lack of fertile material. Usnea patagonica appears to be most closely related to U. durietzii with which it shares many common features. The two species may be distinguished by differences in their habit and branching which, in U. durietzii, is much more irregular with the production of clusters of fibril-like, stunted branchlets rather than fine, extended laterals of U. patagonica. In U. USNEA SUBGENUS NEUROPOGON 85 patagonica branches are never inflated and pseudoisidia are less prominent. The holotype of U. durietzii is very close to, but distinct from, more robust forms of U. patagonica and, in addition, contains traces of norstictic acid. The presence of fatty acids rather than depsidones, preferred habitat and distribution may, conversely, indicate affinities between Usnea patagonica and Race 3 of U. trachycarpa with fibrils of the latter species being replaced by extended laterals bearing soralia. In contrast, the general habit of the new species is similar to depauperate forms of U. antarctica from the South Orkney and South Shetland Islands, although the thallus is less robust with somewhat flaccid ultimate branches. The extent of pigmentation is also similar and is frequently sparse; unlike the majority of specimens of U. antarctica the thallus always has a blackened basal portion above the holdfast. Usnea patagonica may be distinguished from U. subantarctica and U. sphacelata by the form of the soralium, presence of pseudoisidia, less extensive pigmentation, and different ecological parameters; from U. acromelana by the presence of papillae and pseudoisidia, a laxer medulla and by fewer, usually unpigmented, cortical annulations. Specimens examined BOLIVIA. La Paz: Prov. Murillo, Valle de Chuquiaguillo, Incachaca, 4200 m, 7 April 1921, E. Asplund56 (UPS). CHILE. Santiago: Cordillera Central, SW., 3000 m, 1964, G. Follmann 13446-L (LD), 13435-L (KAS- SEL), 13442-L(UPS), 13443-L (UPS). Cautin: Cordillera Lonquimay, 2050m, 4 November 1930, R. P. A. Hollermayer s.n. (H). Aisen: Estancia Nirehuao, Bano Nuevo, 23 September 1940, R. Santesson 5010 p.p. (S), 5021 (S); Estancia Nirehuao (25-30 km N. of Rio Coyhaique), 20 November 1940, R. Santesson 4821 (S), 25 November 1940, R. Santesson 4865 (S); Estancia Coyhaique Alto (near Cerro Coyhaique), c. 100 m, 18 November 1940, R. Santesson 4632 p.p. (S); Coyhaique Alto, 1000 m, 18 November 1940, R. Santesson Lien, austroam., ex herb. Regnelliano 423 p.p. (C, H, M, S); Coyhaique, Cerros Divisaderos (Cordon de Bella Vista), 1200 m, 13 November 1940, R. Santesson 4435a (S), 4435b (S, UPS). Magallanes: Rio de los Cruzeros (60 km NNE. of Punta Arenas), 26 April 1940, R. Santesson 1920 p.p. (S); Natales, Cerro Dorotea, 9 May 1940, R. Santesson 2137 p.p. (S), 8254 p. p. (S); Lago del Toro (Lago Maravilla), La Peninsula, 10 March 1941, R. Santesson 6325 (S); Tierra del Fuego, Isla Navarino, 300 m, 1963, G. Follmann 14591 (UPS), 13978 (KASSEL), 14587 (KASSEL), 14586 p.p. (M); 14589 (LD); Isla Navarino, Puerto Navarino, 10 m, 28 February 1940, R. Santesson 1234a (S, UPS); Paine National Park, Lake Pehoe, 129 km N. of Puerto Natales, 18 m, 11 March 1974, C. Neher 76 (LAM 202156). ARGENTINA. Rio Negro: Cerro Leones, near summit, W. bank of Lago Nahuel Huapi, 1 January 1974, A. Henssen & G. Vobis 24625 (MB); Parque Nacional Nahuel Huapi, Cerro Catedral, Liftstation, c. 2225 m, 6 January 1974, A. Henssen & G. Vobis 24674f p.p. (MB); Cerro Otto, c. 10 km W. of San Carlos de Bariloche, 1200 m, 29-30 December 1980, K. Kalb s.n. (herb. Kalb). Santa Cruz: Lago Viedma, 2 February 1903, P. Dusen s.n. p.p. (H); N. coast of Lago Viedma, 1200 m, 2 April 1903, Hoberg 123 (BM); Lago Argentine, Calafate, weg nach Punta Bandera, Campo Anita Fuss des Cerro Moyano, 300-400 m, 20 December 1973, A. Henssen & G. Vobis 24535d (MB); Calafate, Cuevas de Hualichu, 18 December 1973, A. Henssen & G. Vobis 24523a (MB); near Calafate, 1959, P. W. James 4/120b (BM); Parque Nacional Los Glaciares, Lago Roca, Jeronima, 16 December 1973, A. Henssen & G. Vobis 244991 (MB); Lago Roca, Cordon de los Cristales, c. 1000 m, 26 December 1958, P. W. James s.n. (BM); Cordillera Cristales (type locality); Lago Roca c. 1000 m, December 1958, P. W. James 16 (BM); 20 (BM); Lago Roca, Cerro del Fraile, January 1959, P. W. James s.n. (BM); Rio Fosiles, c. 1000 m, April 1905, P. Dusen s.n. p.p. (M, WU 2997); Uppsala Glacier, Estancia La Christina, 16 December 1958, P. W. James 5/49 p.p. (BM), 5/10 (BM). Tierra del Fuego: Ushuaia, La Peninsula, 3 January 1940, R. Santesson 559 p.p. (S); Sierra Sorondo, N. slope above Las Cotorras (c. 20 km ENE. of Ushuaia), 800 m, 6 February 1940, R. Santesson 641f p.p. (S); Sierra Alvear, S. slope, above Las Cotorras, 700 m, 9 February 1940, R. Santesson 637b (UPS). Uncertain determination SOUTH AFRICA. Cape Province: Cedarberg. Between Tafelberg (Table Mt.) and Spout, 1830 m, 18 January 1947, E. Schelpe 1961 p.p. (BOL, CANL 16944). 9. Usnea perpusilla (Lamb) F. J. Walker, comb nov. Figs 25-26 Neuropogonperpusillus Lamb in J. Linn. Soc. (Bot.) 52: 234 (1939). Type: Chile, Pico de Pilque, prope Antuco, Poeppig PI. Chil. Ill, 278(56) (BM! - holotype; B, G - ?isotypes (Motyka, 1936), not seen). [TLC: no medullary substances, usnic acid.] 86 F. J. WALKER Neuropogon rohmederi Lamb in Lilloa 14: 158 (1948). - Usnea rohmederi (Lamb) Lamb in An. Parq. nac. B. Aires 7: 157 (1959) ['1958']. Type: Argentina, Chubut, Lago Futalaufquen, 1800-2000 m, 1945, G. Rohmeder (LIL 2275 - holotype, not seen; BM!, CANL 17205!, H!, UPS!, US! - isotypes). [TLC: no medullary substances, UV+ unknowns, usnic acid.] Neuropogon rohmederi f. ushuaiensis Lamb in Lilloa 14: 160 (1948). Type: Argentina, Tierra del Fuego, Sierra Alvear, the southern slope, above Las Cotorras (c. 20 km ENE. of Oshuaia), 900-1000 m, 7 February 1940, R. Santesson 640c (640a p.p.) (S! - holotype; CANL 17209! , UPS! - isotypes). [TLC: no medullary substances, usnic acid.] Description: Thallus (2)-5-10(-14) cm, arising from a proliferating or rarely delimited, unpigmented, holdfast, erect, dichotomous, infrequently to richly branched above. Bran- ching often divergent with short, subcapillaceous, sometimes shortly attenuate then deflexed, laterals. Fibrils very rare. Branches terete, yellow-green, variegated with bands of violaceous black pigment, continuously pigmented towards the apices. Cortex variable in thickness. Surface smooth, subnitid to waxy, incomplete fractures, often becoming markedly faveolate, never verrucose, rarely scabrid with minute, pigmented papillae (xlO lens). Medulla lax, rarely sublax, axis thin occupying 0-25-0-5 of the branch diameter. Soredia, pseudoisidia and isidia absent. Apothecia frequent, l-8(-12) mm diameter, lateral, subsessile, often in series, rarely subterminal with a geniculate spur, reniform or irregular on maturity. Disc black, excipulum smooth to faveolate-reticulate, margin thin, often excluded, rays rare. Pycnidia rare. TLC: (1) psoromic acid, 2'-O-demethylpsoromic acid, 1 to 4 UV -I- unknowns, usnic acid; (2) no medullary substances, 1 to 4 UV+ unknowns, usnic acid. Distinguishing features: Usnea perpusilla is characterised by its erect, moderately branched thallus, often extensively pigmented towards the apices, a more or less smooth, waxy, subfaveolate surface, a lax medulla normally lacking medullary substances, a thin axis, and lateral or rarely subterminal apothecia with a black disc, with a faveolate-reticulate excipulum, usually lacking marginal rays. Distribution: Usnea perpusilla is confined to the Andean Cordillera in southern South America, occurring in Argentina and Chile from Tierra del Fuego northwards to about latitude 37S. It is characteristic alpine-southern temperate species, often found at high altitudes, between 1000 and 2000 m, in communities with such species as U. aurantiaco-atra and, occasionally, U. trachycarpa and U. patagonica. Fig. 7. Chemistry: The majority of specimens examined lack medullary substances although some may contain localised high concentrations of three or even four UV+ unknowns (p. 15). Specimens containing psoromic acid (Race 2) are only known from a single locality towards the northern limit of the distribution of the species. Variation: Usnea perpusilla is a moderately variable species exhibiting a range of growth form, branch anatomy, and apothecial form. The size of the thallus is very variable, ranging from small thalli, c. 2 cm, which bear small, cupular apothecia, as in the type specimen, to larger, c. 14 cm, erect forms that may become subdecumbent. The extent of branching is also variable, ranging from almost monopodial forms arising from a proliferating holdfast to lax, spreading forms, arising from a delimited holdfast, as in the type of U. rohmederi. Where branching is extensive the thallus may produce short, fibril-like, recurved, subcapillaceous, ultimate branches. Specimens recently collected from Prov. Rio Negro, Argentina (herb. Kalb!, BM!) illustrate well two features of the variation found in this species; namely, seemingly superficial morpholo- gical similarities to Usnea ciliata, and an intermediate range of thallus size between the types of U. perpusilla and U. rohmederi. The main branches are usually unpigmented near the base, with apices and secondary branches being either conspicuously variegated or continuously pigmented violaceous black towards the tips. The surface is characteristically smooth and waxy and may become faveolate or slightly inflated if the underlying medulla is very lax. In such instances the cortex may fracture forming incomplete, unpigmented annulations that are not a constant diagnostic feature as in Usnea USNEA SUBGENUS NEUROPOGON 87 Fig. 25 Usnea perpusilla. A-Isotype of Neuropogon rohmederi Lamb (CANL). Top. Whole thallus xl. Bottom. Detail of excipulum xlO. B(inset)- Holotype oiNeuropogonperpusillus Lamb (BM) x2. 88 F. J. WALKER Fig. 26 Usnea perpusilla. Isotype of Neuropogon rohmederi f . ushuaiensis Lamb (C ANL) x 1 5 . ciliata. The thallus is never strongly verrucose-papillate, as in U. aurantiaco-atra, and only rarely, in more robust forms where fibrils are present, do small, usually black-pigmented, papillae occur, giving the thallus a slightly scabrid texture, as in the type of U. rohmederi f. ushuaiensis (Fig. 26). Typically the axis is very thin in main branches, occupying about a third of the branch diameter. The degree of laxness of the medulla is very variable, and in some instances may only be looser and arachnoid in close proximity to the axis. Specimens in which the medulla is more compact and where the axis correspondingly occupies a greater proportion of the diameter, have previously been mistaken for Usnea ciliata or U. aurantiaco-atra; they may usually be disting- uished from these species by other well-defined characters. Positioning and size of apothecia may vary in Usnea perpusilla. In well-developed thalli the apothecia are often produced serially along a branch, a feature more rarely encountered in other fertile species, for example U. ciliata and U. aurantiaco-atra. Less frequently, in more richly branched thalli, the apothecia are subterminal, and then often have a geniculate subtending spur which may in turn produce additional apothecia, thus reflecting the characteristic habit of the species. The apothecia are subsessile, having a broad area of attachment; frequently various stages of development may be observed on a single branch, ranging from a small, rounded, cupular form to the more typical reniform shape with a deflexed, irregular outline and a virtually excluded thalline margin. Faveolation of the excipulum is usually a constant feature. Marginal excipular rays are infrequent and, when present, may vary in number, length and extent of pigmentation. Their formation may be correlated with the extent of branching since they are most likely to be present in richly branched specimens. Species concept: Examination of the type specimens of Usnea perpusilla and U. rohmederi, including f . ushuaiensis, and study of additional material has shown that these taxa fall within the variation of a single species. The holotype of U. perpusilla (BM!) is very fragmentary (Fig. 25), although additional collections (M!, PC!) that may be part of Poeppig's original gathering, are USNEA SUBGENUS NEUROPOGON 89 better developed. Examination of slides of sections of the holotype prepared by Lamb (BM!) show that mature asci are very rare, an indication that apothecia are in fact immature, and that the medullary tissue, although not expanded, shows signs of becoming arachnoid-lax in the proximity of the axis. Lamb (1948fl) separated Usnea rohmederi f. ushuaiensis on the presence of minute, pig- mented papillae, an almost smooth excipulum with marginal rays, and slight differences in spore size and thecium height, all of which are here accepted as infraspecific variation. Usnea perpusilla has often previously been mistaken for V. ciliata and may be distinguished by the lax medulla, absence of conspicuous, pigmented annulations, a frequently variegated, rather than continuously pigmented, thallus, and often by the serial position of the apothecia. (Fig. 2). The two species have a marked difference in distribution. In U. ciliata apothecia are normally terminal or sublateral, also with a geniculate spur, whilst in U. perpusilla they are more frequently produced laterally, often up to five in series, with a much broader area of attachment, sometimes producing an acute geniculation of the branch. Forms with fibrils may be distinguished from Usnea trachycarpa by the difference in disc colour; the species can be distinguished from U. aurantiaco-atra by the degree of ornamentation and differences in branch anatomy. Specimens examined Racel ARGENTINA. Rio Negro: Cerro Rigi, near Lago Frias, c. 1750 m, 15 February 1950, I. M. Lamb 6046 (CANL 17016, UPS). Race 2 CHILE. Bio Bio: 'in summ. And. cacum.'. Kunze s.n. (M); In cacum. mentis, Pico de Antuco, 1835, Kunze s.n. (PC); Pico de Antuco, ex Kunze, s.n. (PC); Poeppig PI. Chil. Ill [c. 37S] (type locality). Cautin: Cordillera Lonquimay, 2050 m, 4 November 1930, R. P. Hollermayers.n. (H). Unlocalised: 'Chil. bor. rupes marit.' Poeppig s.n. (PC). ARGENTINA. Neuquen: Cordillera Suangulo [38-41S], 2130 m, 16 January 1926, Kew Andes Exped. , H. F. Comber 470 (BM, E); Paso Pino Hachado - Lonquimay [Chilean Border], 1900 m, 10 January 1948, A. Pfister 8123 (CANL 17014, UPS); Parque Nacional Nahuel Huapi, Brazo Rincon, Cerro Dormilon, Perez Moreau 6793 p.p. (BM, H), 2 February 1940, Perez Moreau 4543 p.p. (H); Parque Nacional Lanin, Cerro Malo, N. of Lago Lacar, c. 1900 m, 28 January 1968, /. H. de Haas 1290-A (U 313666b). Rio Negro: Parque Nacional Nahuel Huapi, Cerro Rigi, Lago Frias, 23 January 1940, Perez Moreau, s.n. (S); /. M. Lamb 6046 (as Race 1) (BM, UPS); Parque Nacional Nahuel Huapi, Cerro Catedral, Liftstation, c. 2225 m, 6 January 1974, A. Henssen & G. Vobis 24674f (MB), 24677d (MB), 24677g (MB); Parque Nacional Nahuel Huapi, Cerro Catedral, c. 20 km SW. of San Carlos de Bariloche, 1850 m, 2 January 1981, K. Kalb s.n. (herb. Kalb); Parque Nacional Nahuel Huapi, Cerro Catedral, near Bariloche, 2000 m, 10 February 1950, /. M. Lamb 5949 (BM, CANL 17206, FH, H, UPS), 5951 (CANL 17207, UPS), 5956 (CANL 17208), 5962 (CANL 36562), 5964 (CANL 36563); Parque Nacional Nahuel Huapi, Cerro Otto, c. 10 km W. of San Carlos de Bariloche, 1250 m, 29-30 December 1980, K. Kalb s.n. (BM, herb. Kalb); Parque Nacional Nahuel Huapi, Capitan, 30 April 1933, E. & A. Ljungner 1367 (S); Cerro Goye, 1670 m, 25 January 1944, /. C. Montiel s.n. (CANL 17204); Lago Nahuel Huapi, Puerto Manzano, 19 January 1966, H. & F. Walter 170 (M). Chubut: Chilean frontier [4428'S: 7134'W], 1500 m, 13 February 1902, Hoberg s.n. (BM); Lago Mendez, Cerro Torrecillas, c. 1000 m, 6 December 1940, N. Kuhnemann 4792 (BM); Lago Futalaufquen, (type locality - N. rohmederi). Santa Cruz; Lago Viedma, t\ 1400 m, Shipton Exped. , January 1959, G. C. Bratts.n. (BM, CHR 343330, FH); Lago Argentine, Cerro Mayo, Seno Mayo, c. 1100 m, Shipton Exped., February 1959, P. W. James s.n. (BM). Tierra del Fuego: Cerro [unrealised] S. of Estancia 'La Marina', 500 m, 1921, Argentina Faculty of Science Exped. 564 p.p. (BM); Sierra Alvear (type locality - N. rohmederi f. ushuaiensis), 800-900 m, 7 February 1940, R. Santesson 639d (S); Sierra Sorondo, above Las Cotorras, c. 20 km ENE. of Ushuaia, 800 m, 6 February 1940, R. Santesson 641e (UPS, S); Monte Marcial, SE. slope, above Ushuaia, 700 m, 29 January 1940, R. Santesson 450c (UPS, S). For further localities see Lamb (1948a). 10. Usnea pseudocapillaris F. J.Walker, sp. nov. Fig- 27 Diagnosis: Usneae subcapillari affinis, sed thallo minore et soraliis parvis punctiformibus differt; apothecia rara. Typus: New Zealand, South Island, Otago, Humboldt Mountains, Mt. Nox, 1950 m, 31 90 F. J. WALKER December 1969, D. J. Galloway (CHR 343226! - holotype; BM! - isotype). [TLC: norstictic acid, salazinic acid, protocetraric acid, usnic acid.] Description: Thallus 2-3 cm, arising from a delimited, pigmented holdfast, erect, becoming subpendulous or subdecumbent. Brancing extensive, divergent, dichotomous, with numer- ous, delicate, capillaceous, spreading branches and short, attenuate, deflexed laterals; usually lacking fibrils. Branches terete, yellow-green, continously pigmented or variegated violaceous black towards the apices. Cortex thin. Surface smooth, waxy, epapillate, easily fracturing, forming regular, black-edged annulations. Medulla sublax towards the axis, axis occupying 0-3 - 0-5 of the branch diameter. Soralia small, punctiform, plane to concave or eroded. Soredia farinose, unpigmented. Pseudoisidia absent. Apothecia rare, immature, as in U. subcapillaris . Pycnidia not seen. TLC: norstictic acid, salazinic acid, protocetraric acid, usnic acid. Distinguishing features: Usnea pseudocapillaris is characterised by its spreading to subdecum- bent habit and a richly branched thallus with numerous, fragile, divergent, flexuose-capillaceous secondary branches bearing small, unpigmented soralia. This species also has a waxy, pig- mented-annulate surface, and a sublax medulla containing depsidones. Distribution: Usnea pseudocapillaris appears to be confined to the South Island of New Zealand, occurring in alpine habitats with other species of the U. ciliata complex. It is so far only known from a restricted area in Central Otago, lying within the distribution of U. subcapillaris, and a single locality in NW. Nelson. It is possible that this species may also occur in Tasmania, where both U. acromelana and U. subcapillaris are known. Some Tasmanian specimens examined are morphologically very similar to this species but, as they lack soredia and apothecia, they are therefore probably best regarded as immature thalli of U. subcapillaris (for example: Mt. Wellington, herb. Degelius A-414!) pending the discovery of more typical material. Fig. 6. Chemistry: Only one chemical race, that containing norstictic and salazinic acids, has been detected. However, it is possible that other races may be discovered since both the related Usnea subcapillaris and U. acromelana have more than one chemical race. Variation: The thalli of this new species are small, rarely exceeding 4-5 cm, are characteristi- cally richly, dichotomously branched, and arise from a delimited holdfast. The fine, capil- laceous, divergent secondary branches form a lax, subdecumbent, loosely-interwoven network, whose fragility is accentuated on storage in herbaria. The primary branches are frequently widely diverging and produce a characteristic angulose network of secondary branches which also occurs in Usnea subcapillaris; however, in U. pseudocapillaris these are not as extensive or extended. The diameter of secondary branches may vary; typically they are short, capillaceous, 0-1-0-3 mm diameter, resembling those of Usnea subcapillaris. Rarely secondary branches may be up to 0-5 mm in diameter and in such instances there is a more gradual transition in width from main branches. In all specimens the ultimate branches are short, divergent or often reflexed and shortly attenuate, giving a netted appearance to the thallus. This is also a feature of richly branched forms of U. acromelana and less well-developed forms of U. subcapillaris. Formation of a secondary holdfast and ensuing regeneration may occur in this species, and in U. subcapillaris, where decumbent branches are in contact with the substrate, as for example in the holotype of this new species. The smooth, waxy surface is characteristic of the Usnea ciliata complex, along with the rupturing of main branches to form black-edged annulations. Such annulations are more widely dispersed than in U. acromelana, although rarely absent; they tend to accentuate the character- istic divergent branching of the thallus. Occasionally the effect of the laxness of the underlying medulla in conjunction with cortical fracturing may give rise to partially inflated segments. The primary branches are usually unpigmented, apart from the annulations, whilst the extent of pigmentation in secondary branches ranges from complete blackening to narrow bands of variegation. USNEA SUBGENUS NEUROPOGON 91 NOtoType Specimen Fig. 27 Holotype of Usnea pseudocapillaris F. J. Walker (CHR). Top. Whole collection xl. Bottom. Detail of branching and soralia x 10. The soralia are usually minute and indistinct and only occasionally become convex, globose, confluent or pigmented in more robust specimens. The presence of soralia may account for the shorter and stouter nature of the secondary branches in this species when compared with Usnea subcapillaris . Apothecia are extremely rare and only immature examples are known on two thalli which form part of the type collection. Species concept: Usnea pseudocapillaris belongs to the U. ciliata complex and may be regarded as the sorediate counterpart to U. subcapillaris. Plants are generally smaller and slightly more erect than U. subcapillaris and have shorter, less trailing, secondary branches. Unlike U. ciliata 92 F. J. WALKER and U. subcapillaris the delimitation between this species and U. acromelana may not always be so easily discerned, since occasional thalli exhibit intermediate characteristics. However, this is a rare occurrence and, despite the seemingly close relationship, the persistence of a distinct, divergent branching pattern, capillaceous secondary branches, and a much more flaccid, fragile nature are here considered to merit specific rank. It is generally possible to separate coarser specimens of U. pseudocapillaris from the finer forms of U. acromelana using a combination of some of the characteristic features of the former, even though a single thallus may not exhibit all these features. Short, divergent primary branches are typical of U. pseudocapillaris; more richly branched forms of U. acromelana tend to be more compact with less divergent branches which lack segmentation, and have more numerous, closely arranged annulations. Usnea pseudocapillaris may easily be distinguished from typical forms of U. acromelana which are sparingly branched, erect, and arise from a proliferating holdfast, and from U. antarctica and U. sphacelata by differences in habit, branching, ornamentation, and medullary chemistry. Specimens examined NEW ZEALAND. South Island. Nelson: KakapoPeak,c. 1520m, 16 December 1982,7. K. Bartlett 25814 (herb. Bartlett, BM); Tasman Mountains, Douglas Range, c. 1400 m, 17 December 1983, J. K. Bartlett 36109d p.p. (herb. Bartlett, BM). Otago: Park Pass, 1200 m, 19 February 1968, D. J. Galloway s.n. (CHR 343384 p.p.), 1430 m, May 1968, D. J. Galloway s.n. (CHR 343380 p.p.); Humboldt Mountains, Mt. Amphion, above Park Pass, 1830 m, February 1971, D. J. Galloway s.n. (BM, CHR 343485); Humboldt Mountains, Mt. Nox (type locality); Sugar Loaf Saddle, 1070 m, May 1966, D. J. Galloway s.n. (CHR 343446 p.p.; BM, CHR 343413 p.p.). 1280 m, February 1968, D. J. Galloway s.n. (CHR 343408 p.p.); Forbes Mountains, western slopes, 1680 m, 2 December 1967, A. F. Mark s.n. (CHR 343462); Earnslaw Burn Valley, 1220 m, January 1968, D. J. Galloway s.n. (CHR 343362 p.p.); Wilkin Valley, Mt. Jumbo, 1830 m, January 1969, A. F. Mark s.n. (CHR 343385); Young Range, 1430 m, March 1968, D. J. Galloway s.n. (CHR 343436 p.p., CHR 343447 p.p.), 1520, March 1968, R. J. Nilsson s.n. (CHR 343396 p.p.); unnamed Virgin, N. of Pope's Nose, 2350 m, 11 February 1969, L. D. Kennedy s.n. (CHR 343366 p.p.); Mt. Roy, above Wanaka, 1 November 1972, G. C. Bratt 72/1506 (HO 35175 p.p.). Uncertain determination Otago: Mt. Ward, summit, 2590 m, January 1961, R. G. Cunninghame Murray 5629 (OTA). 11. Usnea sphacelata R.Br. Figs 28-29 Chloris melvilliana: 49 (1823 ['1821']). - Usnea melaxantha var. sphacelata (R. Br.) J. D. Hook., Bot. Ant. Voyage 'Erebus & Terror' 1: Flora Antarctica 2: 520 (1847). - Usnea sulphurea var. sphacelata (R. Br.) Vainio, Re. Voy. S. Y. Belgica, 1897-99, Botan., Lichens: 11 (1903). Type: Melville L, Mr James Ross 114 (BM! - holotype; BM! - isotypes). [TLC: no medullary substances, usnic acid.] (see Note 1) Lichen sulphureus J. Konig in Olafsen & Povelsen, Reise igien. Island Appendix: 16 (1772), non Lichen suphureus Retz. (1769) [Article 64.1]. - Usnea sulphurea Th. Fr. in K. svenska VetenskAcad. Handl. 7(2): 9 (1867) [Article 72, note l].-Neuropogon melaxanthus f. sulphurea (Th. Fr.) Hue in Nouv. Archs Mus. Hist. not. Paris Ser. 3. 2: 272 (1890). - Neuropogon sulphureus (Th. Fr.) Hellbom in Bih. K. svenska VetenskAkad. Handl. 21 (3/13): 21 (1896). Type: Iceland (not extant), (see Note 2) Lichen pallidus Retz., Fl. Scandin. Prodrom.: 234 (1779), non Lichen pallidus Schreber (1771). [Article 64.1]. Type: Lichen sulphureus J. Konig. (see Note 2) Usnea frigida Dodge & G. Baker in Ann. Mo. hot. Gdn 25: 603 (1938). Type: Marie Byrd Land, Edsel Ford Range, Mt. Rea-Cooper, P. Siple, F. A. Wade, S. Corey & O. D. StancliffRl (Herb. Dodge -holotype, not seen). Neuropogon acromelanus var. inactivus f. picatus Lamb in /. Linn. Soc. (Bot.) 52: 220 (1939). - Usnea picata (Lamb) Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. 7: 205 (1948). Type: [South Victoria Land] Cape Adare or Sastrugi, Br. Antarct. 'Terra Nova' Exped. 1910 (BM! - holotype). [TLC: no medullary substances, usnic acid.] Neuropogon acromelanus var. inactivus f. scabridulus Lamb in/. Linn. Soc. (Bot.) 52: 220 (1939). - Usnea scabridula (Lamb) Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B 7: 204 (1948). Type: Antarctica [unlocalised] , Campbell D. Mackellar (BM! - holotype). [TLC: no medullary substances, usnic acid.] Neuropogon ciliatus var. subpolaris Lamb in /. Linn. Soc. (Bot.) 52: 217 (1939). - Usnea subpolaris (Lamb) Dodge, Lich. Fl. Antarctic Cont. 237 (1973). Type: South Victoria Land, Cape Sastrugi, Evans Cove, Br. Antarct. 'Terra Nova' Exped. 1910 (BM! - holotype). [TLC: no medullary substances, usnic acid.] Usnea laxissima Dodge in Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 198 (1948). - Neuropogon antarcticus USNEA SUBGENUS NEUROPOGON 93 var. laxissima (Dodge) J. Murray in Trans. R. Soc. N.Z. (Hot.) 2: 71 (1963). Type: Queen Mary Land, Possession Nunatak, 15 December 1912, C. T. Harrisson, AAE 84 (herb. Dodge - holotype, not seen; MEL 1012296! - isotype, CHR! - isotype). [TLC: no medullary substances, usnic acid (isotype).] (see Note 3) Neuropogon lambii Imshaugh in Rhodora 56: 154 (1954). Type: U.S.A., Washington, Mt. Rainier National Park, Yakima Park, 1830 m, 19 August 1942, P. F. Scholander (FH! - holotype). [TLC: no medullary substances, UV+ unknowns, usnic acid.] lUsneastriata Zammuto in Dodge in Trans. Am. micros. Soc. 84: 522 (1965). Type: Edward VII Peninsula, Rockefeller Mountains, Mt. Breckinridge, 7803'S: 15528'W, R. G. Frazier & F. A. Wade 315 (herb. Dodge - holotype, not seen), (see Note 4) Note 1: Usnea sphacelata R. Br. Type material is present in the BM. Material incorporated into collections bearing the species number and Brown's handwriting is recognised as holotype material (Fig. 28). This conforms with current research into Brown's bryophyte collections in BM (Harrington & Ellis, unpublished). Several isotypes are known, including boxed material of lichens and phanerogams forming 'Flora Antarctica' of Captain Parry's First Voyage. The name Usnea sphacelata was originally published by Brown (1823) in Chloris Melvilliana which was a preprint of his account in the Appendix to Capt Parry's Voyage (Brown, 1824) with independent pagination (Stafleu & Cowan, 1976); the 1824 reference has always been cited by previous authors (Zahlbruckner, 1930; Motyka, 1936; Lamb, 1939a). Note 2: Lichen sulphureus J. Konig. This name is a later homonym of Lichen sulphureus Retz. (Retzius, 1769) and is consequently invalid (Article 64.1). Retzius (1779) gave the taxon a new name, Lichen pallidus , but this is also invalid, being a later homonym of Lichen pallidus Schreber. The correct name for this taxon is consequently Usnea sphacelata R. Br. Note 3: Usnea laxissima Dodge Isotype material (Fig. 29) represents a weathered, decumbent form of U. sphacelata. Murray (1963) wrongly regarded this taxon to be a variety of U. antarctica, although a record from Macquarie Island (Filson, 1981) is probably referable to that species. Note 4: Usnea striata Zammuto This taxon is only known from the type collection from Edward VII Peninsula (Dodge, 1973) which has not been made available for study. The original description indicates that the taxon is synonymous with U. sphacelata. Description: Thallus 1-5-5 cm, arising from a delimited or proliferating, rarely pigmented holdfast, erect, sparsely to richly branched above with capillaceous, attenuate branches. Fibrils usually absent. Branches terete, yellow-green, conspicuously variegated above with bands of black pigment or continuously pigmented violaceous black towards the apices. Cortex variable in thickness. Surface subnitid or matt, smooth to faveolate, rarely inflated, scabrid with minute, often pigmented papillae (xlO lens). Medulla lax or rarely sublax, axis thin, occupying 0-2-0-4 of the branch diameter. Soralia numerous, confined to ultimate branches, plane, emarginate, becoming convex-pulvinate to globose on maturity. Soredia granular, frequently partially corticate then pigmented. Pseudoisidia rare, isidia absent. Apothecia and pycnidia not seen. TLC: (1) psoromic acid, 2'-O-demethylpsoromic acid, usnic acid; (2) no medullary substances, 2-3 UV+ unknowns, usnic acid. Distinguishing features: Usnea sphacelata is characterised by its erect, usually richly branched habit, a subnitid to subpapillate surface with prominent bands of pigment, a lax medulla, usually lacking depsidones, a thin axis, and plane to nodular, often pigmented, soralia more or less confined to secondary branches. Distribution: Usnea sphacelata is the only known bipolar species of the subgenus. In the arctic its distribution is almost circumpolar including the New Siberian Islands in the east to Melville Island in the west. The species is known from Greenland, Iceland, Svalbard, Franz Joseph Land, Novaya Zemlya, Jan Mayen, and a few islands of arctic Canada, with its main distribution in the Canadian eastern arctic; it is rare in western arctic Canada. Its distribution has been adequately mapped by Lynge (1941) and Thompson (1972) and there are few recent additions. Figs 5 & 7. 94 F. J. WALKER HOLOType Specimen Fig. 28 Holotype of Lfrnea sphacelata R. Br. (BM) xl. Inset. Detail of soralia x 10. In Greenland Usnea sphacelata has a northerly distribution, although it is absent from the north coast. The species is frequent north of latitude c. 75N on the east coast, rarely extending to latitude 68-70N, and on the west coast as far south as Disko (Lynge, 1941; Lamb, 1939ft). The species has only recently been discovered in southern Greenland (E. Hansen, 1982) at Nakalaq (C!) slightly south of the area investigated by K. Hansen (1971) who failed to find the species. In the arctic Usnea sphacelata is characteristically, but not obligately, an alpine species USNEA SUBGENUS NEUROPOGON 95 occurring, for example in Greenland, on the upper levels of rock falls and precipices (Lynge, 1932) and only occasionally at lower altitudes (K. Hansen, 1962). Towards the southern range of distribution in the arctic the species is confined to higher altitudes, and rarely occurs below 500-600 m in Iceland (Lynge, 1941) and western Svalbard (Lynge, 1941; Weber, Lich. Exs. 591); in the northern part of its range it occurs down to sea-level, for example in Franz Josef Land. The species is very rare in the U.S.A. , being only known from the type locality of Usnea lambii in Washington State and an additional record nearby (W. Weber, 1973) . A single gathering from c. 4400 m in Mexico (Mt. Orizaba, Metzger, FH!, UPS!) forms a link between the North American distribution and that of the northern Andes. In South America, Usnea sphacelata occurs in Ecuador at very high altitudes, and is very rare in Venezuela and Peru as far south as latitude c. 15S. South of this latitude, apart from isolated, mainly high-alpine, subantarctic localities in New Zealand and Patagonia, there appears to be a significant gap in the distribution before the circumpolar antarctic populations are encountered. Many previous records from South America (as U. sulphured} are erroneous and are referable to U. patagonica, U. durietzii, and U. subantarctica. The species is absent, and is replaced by Usnea antarctica, from the Falkland Islands, South Orkney, and South Shetland Islands, as well as the South Sandwich Islands. On the antarctic continent U. sphacelata is circumpolar and is more frequent than U. antarctica, although occasionally misidentified as that species (for example, Follmann Lich. Exs. 399, BM!). As in the arctic it is frequently a species of high altitude, exposed locations, although also occurring at lower altitudes (see Bowra et al, 1966; Kashiwadani, 1970; Lindsay, 1972; 0vstedal, 1983). On the Antarctic peninsula the species occurs at a range of altitudes (Lamb, 1964) on the west coast, mainly south of the Antarctic Circle, and throughout the east coast, although it is often replaced at lower altitudes by U. subantarctica in the north-east and south-west. Chemistry: Race 1 , containing psoromic acid , is extremely rare and is only known from a single collection from the Andes in Peru tentatively assigned to this species. All other specimens examined were found to be depsidone-deficient; occasionally containing traces of UV+ unknowns. Specimens from Patagonia and the Antarctic peninsula containing norstictic and/or salazinic acids, previously referred to -this species (Lamb, 19480, 1964), belong to Usnea subantarctica, and, from the northern Andes, to U. durietzii. Variation: Arctic populations of Usnea sphacelata are much more uniform than those from the southern hemisphere and are characterised, as found in the type specimen, by a richly branched, strikingly variegated thallus, with pigmented, nodular soralia. The thallus may arise from a delimited holdfast, although this is more frequently proliferating and spreading. Branching is more or less dichotomous and regular with the production of extended, variegated laterals which in turn may give rise to capillaceous, extensively pigmented, ultimate branches. The thallus may be smooth to subfaveolate and is usually subnitid to waxy or rarely matt; it lacks pigmented annulations. Small, often pigmented, papillae are usually present; only rarely in a few arctic specimens are the main branches waxy, epapillate and unpigmented, as, for example, the type of Usnea lambii; the medulla is usually lax and arachnoid in main branches, but in some instances may not be fully expanded and then the axis occupies a greater proportion of the branch diameter. Soralia are characteristically large, pulvinate, emarginate, sometimes geniculate; they tend to occur at intervals towards the apices of secondary branches, and are often broader than the subtending branch. Soredia often become partially corticate and pigmented but do not produce distinct pseudoisidia; they only lack black pigment when poorly developed as small punctiform soralia, or if the thallus is weathered or moribund. In the northern Andes variation within populations is more diverse but is difficult to assess since the species appears to be rare. The 20 specimens examined are often extensively pigmented or lack several distinctive features of the species. Some thalli reflect the variation found in arctic populations. Others differ in having a delimited holdfast, are richly branches with extended laterals bearing minute, unpigmented, eroded soralia on a subnitid, subfaveolate to 96 F. J. WALKER subpapillate thallus. Other variants resemble Usnea acromelana and are virtually monopodial, arise from a proliferating holdfast, and are infrequently branched with large, pale, excavate soralia; however, they may be distinguished by a lax or sublax medulla and absence of annulations. Rarely the medulla may become markedly inflated or small pseudoisidia are produced, thus superficially resembling U. durietzii. Occasionally thalli, tentatively referred to this species, are found in collections of U. durietzii; these are extensively pigmented and papillate, and have plane soralia and a compact medulla (for example, Arvidsson & Nilson 945 p.p.,GB!). Material conforming to arctic populations is still present, but less frequent, in the antarctic regions and appears to be confined to lower altitudes where it is often replaced by Usnea subantarctica in areas where their distributions overlap. In such instances the two species may be separated by the way in which secondary branches are produced and sometimes by their medullary chemistry. Material described as fertile U. sulphurea (Lindsay, 1969) is referable to U. subantarctica (p. 102). Thalli of Usnea sphacelata from Antarctica, particularly from high altitudes, are erect or rarely subdecumbent and are often more or less monopodial arising from a proliferating holdfast, occasionally branching extensively towards the apices. Rarely in subdecumbent thalli, are the secondary branches extensive, divergent and capillaceous with flexuose apices; produc- ing a fragile, interwoven network which resembles, but is not as extensive as, that of Usnea subcapillaris and U. pseudocapillaris (for example, DronningMaud Land, H. U. Sverdrupfjella, Angard, TROM!). Pigmentation is confined to apices and tends to be continuous rather than variegated and is violaceous black. The cortex is often thicker and the surface subnitid, sometimes epapillate, rarely faveolate and fracturing, whilst the medulla may only be sublax. Consequently such thalli have sometimes been mistaken for species of the U. ciliata complex (Lamb, 19390, 1948a; Dodge, 1973; Murray, 1963) and described as distinct taxa; Neuropogon ciliatus var. subpolaris is a sparsely sorediate, decumbent form of U. sphacelata whilst in N. acromelanus var. inactivus f . picatus and f . scabridulus the medulla is only lax towards the axis and weathering of the thallus has given rise to development of untypical annulations. Specimens from New Zealand and Patagonia resemble those antarctic populations that bear a superficial resemblance to Usnea acromelana. In these the medulla may only be slightly lax, the surface epapillate, and the soralia small, plane and unpigmented. In New Zealand the two species may easily be distinguished by differences in medullary chemistry, whilst in Patagonia and the Antarctic peninsula more critical examination of morphological features is required. Species concept: Although the range of variation of antarctic populations of Usnea sphacelata is much greater than that in the arctic, I do not consider this sufficient for the recognition of separate taxa. In addition thalli resembling the type from the arctic are also present in Antarctica, as well as a range of intermediate forms. The only distinction that has been made here is the recognition of a new species, U. subantarctica (p. 99), which is characterised by a more open branched habit with dispersed fibrils, often forming tassel-like apices; in contrast U. sphacelata has fewer, broader, extended laterals, usually lacking fibrils, and lacking a norstictic acid-containing race. Dodge & Baker (1938) only recognised Usnea sphacelata from Patagonia northwards, whilst Dodge later (1973) referred to this as solely an arctic species, describing most continental antarctic material as a distinct species, U. frigida, in addition to his other taxa that are here also regarded as synonyms. From the original description and illustrations (Dodge & Baker, 1948), and from Lamb's subsequent report (Lamb, 1964) on a prepared slide of the holotype of U. frigida and his examination of material from the type locality ex MO, (Lamb, unpublished notes, A AS!) it is clear that this taxon is a synonym of U. sphacelata. In addition specimens examined from a nearby locality (KASSEL 24772, ex herb. Dodge!) and material distributed by OS as U. frigida (BM! , CANL! , KASSEL!) though depauperate, are also clearly referable to this species. Some thalli of Usnea sphacelata bear a superficial resemblance to those of Usnea acromelana, particularly from New Zealand and Patagonia. This might suggest either a parallel evolution of characters from a common ancestry or, remotely, some indication of hybridisation between the USNEA SUBGENUS NEUROPOGON 97 Fig. 29 Usnea sphacelata. Variation in southern hemisphere populations. Top. New Zealand, Otago, Mt. Aspiring, Kennedy (CHR 343354) x 1. Bottom. Isotype of Usnea laxissima Dodge (MEL 1012296) x 1-5. two species. However, on evidence, it is more likely that U. sphacelata has closest affinities with U. perpusilla. Features common to U. sphacelata and U. perpusilla include a diverse range of habit and similar branching, pigmentation, surface ornamentation, variation in medulla width, and lack of chemistry. 98 F. J. WALKER Occasionally thalli are encountered that are somewhat intermediate between Usnea sphacela- ta and U. antarctica (see synonymy in both species) , but these similarities are superficial and are most likely to occur in immature specimens. Close examination reveals the characteristic anatomical, surface, and soralia features of the two species. The soralia of U. sphacelata tend to be more globose than in U. antarctica and lack the distinctive crateriform margin, and are less widely distributed throughout the thallus. The species may be distinguished from U. subantarc- tica by differences in branching and papillation; from U. acromelana by a laxer arachnoid medulla and lack of pigmented annulations; from U. durietzii and U. patagonica by the absence of pseudoisidia; from U. acanthella by the absence of true isidia. Selected specimens examined Racel PERU. Puno: San Antonio de Esquilache [1608'S: 7022'W], 4750 m, 20 May 1937, D. Staff ordlGl (BM, FH). Race 2 JAN MA YEN. [7100'N: 900'W], Eskkrateret, 22 July 1930,7. Lid s.n. (BM). ICELAND. Sydri Bjargh611 [c. 6540'N: 1645'W], 8 km NE. of Reykjahlid, Sudur Pingeyjarsysla, c. 550 m, 11 August 1982, C. D. & D. H. Dalby s.n. (BM); Nordur-Mulasysla, Austurfjallgardur, Modrudal- sorafi [6520'N: 1542'W], 680 m, 10 July 1979, H. Hertel 21586 (M); Ytri Baegisa, Akureyki [6541'N: 1804'W], 610 m, August 1963, B. A. RowlandLll (BM); Tungnafjellsjokull [6445'N: 1755'W], 1300 m, 8 August 1967, H. Kristinsson 23095 Vezda: Lich. Sel. Exs. 886 (BM); Kjallfell, 26 July 1895, Stefansson s.n. (C). GREENLAND. Disko I.: Lyngmarksfjeldt [6916'N: 5335'W], 200 m, 16 July 1974, P. Mellergard & E. S. Hansen s.n. Lich. Groen. Exs. C 102 (BM, C, M), head of Nordfjord [6757'N: 5419'W], 26 June 1952, P. Gelling Lich. Groen. Exs. C 152 (BM, C, M); Svartenhuk, Simiutaq [7153'N: 5225'W], 1200 m, August 1949, N. Berg s.n. (C); Angmagssalik district, S. of Pourquoi-Pass glacier [6639'N: 3540'W], 1500 m, 8 August 1966, M. Zumbuhls.n. (C); Nakalaq [6059'N: 4555'W], 1400 m, 20 August 1975, D. Olsen L-1405 (C). CANADA. North West Territories: Ellesmere Land, North Kent I. [7630'N: 9000'W], 13 July 1901, Simmons s.n. (BM); Franklin district, Bathurst I., E. side of May Inlet, N. of Purcell Bay [7623'N: 10047'W], c. 25 m, 9 July 1963, Weston Blake 19a (C, UPS); Parry Is., Melville I [c. 75N: 112W] (type locality U. sphacelata)}. Ross s.n. (BM). NOVAYA ZEMLYA [7440'S; c. 55E]. see Lynge (1941). NEW SIBERIAN IS. [c. 76N: 140E]. see Lynge (1941). FRANZ JOSEF LAND. Hooker I. [81S: 53E], Ssedov 'promitory', 1930, V. P. Savicz Savicz: Lich. Rossica (5) 43 (M), [Northbrook I.] Cape Flora, 1901, Palibin s.n. Elenkin: Lich. Flor. Rossiae 113a (BM, FH, US). SVALBARD. Spitzbergen: Isfjord area, near Longyearbyen [7812'N: 1540'E], Mt. Nordenskjold, 100 m, 3 August 1975, H. Hertel & H. Ullrich s.n. Lich. Exs. COLO 591 (BM, C, M); Svalhardhytta, c. 200 m, A. Hoeg & J. Lid s.n. Krypt. Exs. Vindobon. 3070 (BM, C, M, WU 2886); Wahlenbergsbay (Hinlopen Strait), 1861, A. J. Malmgrens.n. Fries. Lich. Scand. Exs. 51 (BM, M); Brandewijnsbay, 150m, O. Torrell s.n. Zahlbr.-Redinger, Lich. Rar. Exs. 384 (BM, WU3085). UNITED STATES. Washington: (type locality olNeuropogon lambit). MEXICO. Veracruz: SE. of Mt. Orizaba [1851'N: 9708'W], Citlaltepetl, 4330-4540 m, 17 August 1966, U. V. Metzgers.n. (FH, UPS). VENEZUELA. Merida: paramo de Campanario, 4000 m, 19 September 1938, /. Hanbury-Tracy 168 (BM). Barinas: Pico de las Piedras Blancas, 4762 m, 25 May 1952, V. Vareschi 1259 p.p. (US). COLUMBIA. Caldas: Nevado del Ruiz, NW. side, c. 1500 m, 3 February 1979, H. Sipman & H. Valencia 10423 (BM, U). ECUADOR. Pichincha: Quito, Pinchincha, crater margin, 30 June 1939, E. Asplunds.n. Lich. austroam. ex herb. Regnelliano 360 (BM, C, H, LD); E. slopes Cerro Iliniza, c. 4400 m, 7 March 1972, L. Arvidsson & D. Nilson 945 p.p. (GB); summit of Corazon, 4861 m, E. Whymper 1323 (10) (BM). Napo: Antisana, 4900 m, E. Whymper 1329 (2) (BM). Bolivar: W. side of Chimborazo, 4780 m, 1967, Cambridge Lichen Exped. s.n. (BM). Chimborazo: NW. face of Chimborazo, 3700-4300 m, January 1859, Wagner s.n. (M). ARGENTINA. Santa Cruz: Patagonia, Lago Viedma, Volcan nunatak, c. 13 km from nearest land in Hielo Continental, January 1959, G. Bratts.n. (BM). ANTARCTIC PENINSULA. Graham Land: Joinville I. [6318'S: 5756'W], N. end of Active Sound, 200 m, 27 March 1981, R. I. L. Smith 3736 p.p. (AAS), Crown Prince Gustav Channel, Alectoria I. [6359'S: USNEA SUBGENUS NEUROPOGON 99 5837'W], 30-60 m, 18 August 1945, /. M. Lamb Operation Tabarin 2465 (BM), Nordenskjold coast [c. 6430'S: 5900'W], nunatak 6-4 km NW. of head of Larsen Inlet, c. 910 m, 23 September 1960, C. G. Brading 105a (AAS), Oscar II coast, N. side of Churchill Peninsula [c. 6630'S: 6250'W], 460 m, November 1963, R. Tindall 18 (AAS), Foyn coast, Cape Robinson [c. 67S: 64W], E. of Mt. Hayes, 6 October 1964, G. Stubbs B. A.S. Misc. 6 (AAS), Fallieres coast, Adelaide I. [c. 6715'S: 6830'W], 'Reptile Ridge', Yellow Bluff, 350 m, 9 March 1981, R. I. L. Smith 3453 p.p. (AAS). Palmer Land: Wilkins coast [6920'S: 6341'W], 1070 m, 23 October 1965, M. J. Cousins 22 (AAS), Alexander I., Ablation Valley [7048'S: 6830'W], c. 5 m, January 1979, D. England2692 (AAS), Alexander I., Two-Step Cliffs [7151'S: 6817'W], c. 600 m, 4 February 1979, D. England 2688 (AAS), Black coast, Cape Bryant area [7219'S: 6040'W], 45 m, 23 January 1948, K. S. P. Butler & B. Stonehouse F.I.D.S. E1228 (BM), George VI Sound, Eklund Is. [7320'S: 7200'W], 20 November 1949, V. E. Fuchs et al. F.I.D.S. E604 (BM, FH). DRONNING MAUD LAND. [c. 15W]. H. U. Sverdrupfjella, Jutulr0ra [c. 20W], 1500 m, 11 December 1970, /. Angard s.n. (BG), Vestfjella [c. 13-16W], Miiren, 21 January 1977, L. S0mme 204(102) (BG, BM). ELLSWORTH LAND. [c. 72-74W]. Johnson nunatak, survey point, 7 January 1962, J. T. R. Molholm 6 (BM, CANL 21331, KASSEL 26669), Barnes nunatak, c. 1560 m, 9 January 1962, /. T. R. Molholm 10 (AAS). MARIE BYRD LAND. Edsel Ford mountains [c. 145W], peak 1207, 9 December 1940, H. P. Gilmour U.S. A.S. 234 (US) (type locality of Usneafrigida Dodge & G. Baker), Mt. Grace McKinley [c. 148W], 1934, Second Byrd Exped. (KASSEL 24772). EDWARD VII LAND. [c. 150W]. Rockefeller Mountains, Mt. Paterson, 18 December 1940, R. G. Fitzsimmons U.S. A.S. 278 (US), Mt. Breckenridge, 14 December 1940, R. G. Frazier & F. A. Wade U.S. A.S. 312 (US). VICTORIA LAND. [c. 165E]. Yule Bay, Birthday Ridge, 17 December 1981, L. Kappen 3001 (KIEL), Follmann: Lich. Exs. 399 (BM), Hallett Peninsula, 10 January 1963,7. Cranfieldl (BM). KNOX COAST, (c. 110E). Bailey Peninsula, near Casey Station transmitter building, 35 m, 13 December 1982, R. D. Seppelt 13103 (herb. Seppelt). QUEEN MARY LAND [c. 98E]. Hippo nunatak, 29 December 1912, C. T. Harrisson s.n. A.A.E. 71-2 (AD),A.A.E.71-3(AD). MAC.ROBERTSON LAND. [c. 70E]. Falla Bluff, 9 February 1974, R. Filson 14910 Lich. Ant. Exs. 22 (AAS,BM,M,O). ENDERBY LAND. [c. 50E]. 3 March 1961, E. Korotkevish 70 (AAS), 1961, G. Konovalov 72 (FH). PRINCE OLAV COAST, [c. 44E]. Molodyozhnaya, c. 40 m, 8 February 1968, H. Kashiwadani LA.R.E. 4118 (FH,H). SOYA COAST, [c. 39E]. c. 20 m, 12 January 1968, H. Kashiwadani J.A.R.E. 3823 (H), J.A.R.E. 3773 (H). NEW ZEALAND. Nelson: Cobb Valley, c. 1490 m, 18 December 1982, /. K. Bartlett 25810 (herb. Bartlett, BM). Otago: Mt. Aspiring, NE. to Coxcomb Ridge, 2790-2850 m, 13 February 1969, L. D. Kennedy s.n. (CHR 343354), 2290 m, January 1957, B. J. Wilkins s.n. Murray 1040 (OTA), NW. Ridge, 2590 m, 27 January 1969, D. J. Galloway s.n. (CHR 342745); Pope's Nose, NE. Ridge, 2650 m, 8 February 1969, L. D. Kennedy s.n. (CHR 343319); Mt. Sir William, 2560 m, 27 December 1970, D. J. Galloway s.n. p.p. (CHR 342781 p.p.); Matukituki Valley, Mt. Avalanche, 2560-2590 m, 15 February 1969, L. D. Kennedy s.n. p.p. (CHR 343358 p.p.). The following may be consulted for further localities: arctic regions (Lynge, 1941; Thomson, 1972), Antarctic peninsula (Lamb, 1964) (excluding specimens here referred to Usnea subantarctica), continental Antarctica (Bowra et al., 1966; Dodge, 1973; Kashiwadani, 1970; Lindsay, 1972; 0vestedal, 1978); lists held in BM and collections in AAS and BM. 12. Usnea subantarctica F. J. Walker, sp. nov. Figs 30-31 Diagnosis: Thallus fruticosus, pigmento nigro in cortice ad apices basimque conspicue coloratus. Rami irregulares teretes fibrillosi, papillis numerosis nigriscentibus. Medulla laxa vel sublaxa. Soralia convexa vel nodulosa nigrescentia. Apothecia rara disco rufescenti, apotheciis Usneae trachycarpae similia. Acidum usnicum et acidum norstictum continens. Typus: Antarctic peninsula: Western shore of bay east of old base, on west, coast of Horseshoe I. , Marguerite Bay, Fallieres Coast, 7-5 m, 23 February 1965, R. E. Longton 1275 (AAS! - holotype; BM! - isotype). [TLC: norstictic acid, UV+ unknowns, usmc acid.J Description: Thallus 1-5-3-5 cm, arising from a proliferating, rarely pigmented, holdfast, erect or rarely subdecumbent, moderately branched above with short, divergent, flexuose, 100 F. J. WALKER attenuate secondary branches. Fibrils extensive, irregularly dispersed, to give an open, spinulose to tassel-like habit. Branches terete, yellow-green, continuously black pigmented towards the apices or variegated with bands of pigment. Cortex thin. Surface matt, conspicuous- ly scabrid with small, pigmented papillae. Medulla lax or sublax, axis thin, occupying 0-3-0-5 of the branch diameter. Soralia numerous, confined to ultimate branches, plane becoming convex to pulvinate or nodular, rarely marginate. Soredia granular, often partially corticate then pigmented. Pseudoisidia rare, isidia absent. Apothecia rare, as in U. trachycarpa. Pycnidia not seen. TLC: (1) norstictic acid, salazinic acid, usnic, acid; (2) no medullary substances, usnic acid. Distinguishing features: Usnea subantarctica is characterised by its erect, spreading, richly branched, irregular habit, a matt, papillate, a more or less extensively pigmented surface, a lax or sublax medulla, often containing norstictic and/or salazinic acids, and nodular, often pigmented, soralia more or less confined to secondary branches. Distribution: Usnea subantarctica has a distribution similar to that of U. aurantiaco-atra, although it is largely restricted to the Antarctic peninsula; it is rare in southern South America. The species is absent from continental Antarctica and the islands of the Scotia Arc, but may eventually be found elsewhere in the subantarctic regions. On the Antarctic peninsula the distribution is similar to that of Usnea sphacelata, being more frequent on the north-east coast and at lower latitudes on the west coast. The species has been found in association with U. aurantiaco-atra, U. acromelana, and U. antarctica (R. I. L. Smith 3680, AAS!) and also with U. antarctica and U. sphacelata (R. I. L. Smith 3736, AAS!). In this area the species is usually confined to lower altitudes, frequently replacing U. sphacelata, and is rarely found above c. 500 m where U. sphacelata predominates. Its ecological requirements appear to be somewhat intermediate between those of U. antarctica and U. sphacelata. 'The majority of specimens on which Lamb (1964) based his distribution map of depsidone- containing material of U. sphacelata (as U. sulphured) have been examined and are referable to this species as well as some depsidone-deficient material. Usnea subantarctica is less frequent in Patagonia, often being replaced by other asexual species, including U. patagonica, and is there confined to higher altitudes in the alpine region of the Andes, mainly between 1000-2000 m, where it has been collected with U. antarctica and U. trachycarpa. Fig. 5. Chemistry: Two chemical races have been detected in Usnea subantarctica . The dominant race contains norstictic acid (Race 1) with occasional traces of salazinic or connorstictic acids or rarely just salazinic acid, whilst the other (Race 2) is depsidone-deficient. UV+ unknowns may occur, but are rare and only present in low concentrations. Similarly traces of unidentified fatty acids have also been detected, but these are not a constant feature as in U. trachycarpa or U. patagonica and hence do not have taxonomic significance. Both races have similar distributions and are of approximately the same frequency; however, the presence of norstictic acid does appear to be of salient importance in circumscribing the species. Variation: Usnea subantarctica is usually characterised by an irregular habit that is reminiscent of an 'untidy' specimen of U. sphacelata, but usually has more fibrils, especially towards the apices giving a tassel-like appearance. Indeed, immature, small thalli of the two species, particularly from the Antarctic peninsula, may be indistinguishable on morphological charac- ters, although chemical and ecological data may assist in identification. Thalli may be sparingly branched, with extensive, small, blackened papillae, or may be richly branched and possess fibrils. Secondary branches are produced irregularly along the main axis, giving an open, spreading, divergent, habit; only infrequently may main branches remain in a close cluster or are richly branched from the base, as found in some forms of Usnea sphacelata. Rarely thalli may resemble northern hemisphere forms of U. sphacelata but are usually more irregularly branched with fibrils. Variation in surface ornamentation and branch anatomy is similar to that of Usnea trachycar- pa, and much of the discussion of that species is referable here (p. 113). However, some differences do occur: pigmentation is often more extensive, occasionally resulting in prominent USNEA SUBGENUS NEUROPOGON 101 Fig. 30 Holotype of Usnea subantarctica F. J. Walker (AAS) . Top. Whole thallus x 1 . JBottom. Detail of apothecium and soralia x 10. variegation, even of the fibrils; papillae on primary branches may less frequently give rise to fibrils; the surface is rarely faveolate or subnitid, although may occasionally fracture due to collapse of the underlying lax medulla. Soralia are more or less confined to branch apices or secondary branches and are usually convex to globose, often pigmented, well spaced and wider than the subtending branch, and are similar to some forms of Usnea sphacelata, particularly from the arctic. Soralia production tends to foreshorten the ultimate branches (cf . U. patagonicd) often resulting in flexuose geniculation of the branch. Soralia appear to arise directly from the cortex, by localised breakdown, and are 102 F. J. WALKER not usually derived from papillae, although occasionally a surrounding margin, similar to that of U. patagonica or U. antarctica, is produced in instances where cortical papillae are particularly numerous and prominent. Apothecia are rare and are identical to those of Usnea trachycarpa. They are only known from the holotype where they are immature, and an additional collection (Fig. 31) in which they are well-developed (R. I. L. Smith 829, AAS!). The fruits of the latter specimen were described by Lindsay (1969) and ascribed to Usnea sulphured. Fig. 31 Fertile thalli of Usnea subantarctica. Antarctic peninsula, Horseshoe I. , Smith 829 (AAS) x 1-5. Some specimens from Patagonia are more variable and, because of the range and variation of species found, may require more critical examination and are only tentatively included under this species. Others tend to have more conspicuous, stouter, spreading fibrils. Specimens tentatively referred to this species include one cited by Lamb (1948), as Usnea sulphurea, from Sierra Alvear, Tierra del Fuego (Santesson 640e, S!), containing norstictic acid, and a depsi- done-deficient thallus from a nearby locality (Santesson 641c, S!) mixed with U. antarctica. These show some similarities to papillate forms of U. perpusilla, which might indicate affinities with U. sphacelata, since they have fewer fibrils and have less prominent papillae. Species concept: Although morphologically very similar to Usnea sphacelata, U. subantarctica is regarded as a distinct species, by virtue of differences in habit and branching, the presence of prominent fibrils and papillae, and apothecia that are identical to those of U. trachycarpa. Differences between the two species have been discussed above and also under U. sphacelata (p. 96). It might be argued that some specimens of U. subantarctica are very similar to some northern hemisphere forms of U. sphacelata. However, arctic forms of the latter species have extended laterals rather than short, undivided fibrils and lack the short, spreading, attenuate fibrils on primary branches. Despite the lack of fertile material it is tentatively proposed (p. 40) that Usnea sphacelata is the sorediate counterpart of U. perpusilla and consequently would be expected to have a black rather than rufous brown apothecial disc. Consequently it appears that U. subantarctica has a closer relationship to U. trachycarpa than to U. sphacelata-U. perpusilla. This supposition is enhanced by the presence of norstictic acid in a significant proportion (c. 60%) of the specimens. However, a number of depsidone deficient gatherings are only tentatively referred to this species. If there were excluded the incidence of Race 1 would be 70% . Although traces of fatty acids have been detected in U. subantarctica, it is not certain that these have affinities with the murolic acid complex of U. trachycarpa. For this reason there must remain some doubt as to USNEA SUBGENUS NEUROPOGON 103 whether these taxa represent a species pair. However, specimens are often very similar to high-alpine forms of U. trachycarpa, originally described as f. elatior by Lamb (19480), from Patagonia which are extensively pigmented and have scattered, short, attenuate fibrils and frequently lack medullary substances and fatty acids. The absence of fatty acids might be the consequence of more exposed habitats and this would explain their absence from U. subantarc- tica, particularly in the Antarctic peninsula. It is evident from study of the synonymy of Usnea sphacelata that there is no existing name for the new species. Furthermore, existing names under U. sphacelata refer to material from continental Antarctica, outside the distributional range of U. subantarctica. The epithet 'subantarctica' is selected to refer to the distribution of the species, which although limited, does extend into the subantarctic area. It is likely that the species may eventually be found to be more widespread. A possible relationship to alpine forms of Usnea trachycarpa is also indicated, since that species also has a characteristic, albeit wider, distribu- tion in the subantarctic. The distribution may eventually be compared to that of generally accepted species pairs, for example U. ciliata-U. acromelana and U. aurantiaco-atra-U . antarctica, in which the asexual species extends further south than the fertile counterpart. Usnea subantarctica may be separated from other asexual species of Neuropogon by its untidy to tassel-like appearance and close resemblance to some forms of U. trachycarpa. The form of the soralium may resemble some variants of U. sphacelata and U. acromelana or may sometimes be marginate, although this is as non distinct as the plane, crateriform soralium of U. antarctica. The lack of distinct pseudoisidia separate the species from U. durietzii and U. patagonica whilst the lax medulla and the matt, surace with papillae and fibrils distinguish the species from U. acromelana. Specimens examined Racel CHILE. Aisen: Lago San Martin [49S], 1600 m, 2 February 1933, A. Donat 3 p.p. (H). ARGENTINA. Santa Cruz: Peninsula Magallanes, near Puerto Handera, c. 1000 m, December 1958, P. W. James 59 p.p. (BM); Rio Fosiles, c. 1000 m, April 1905, P. Dusen s.n. p.p. (H). ANTARCTIC PENINSULA. Graham Land: Joinville I. [6318'S: 5548'W], Mt. Alexander, 200-500 m, 27 March 1981, R. I. L. Smith 3680 p.p. (AAS); Trinity Peninsula, Duse Bay [6348'S: 5717'W], NE. of unnamed nunatak, [6336'S: 5701'W], 385 m, 27 March 1946,7. D. Andrew F.I.D.S. D109-6 p.p. (BM), Bald Head, 60-150 m, 18 November 1945, /. M. Lamb Operation Tabarin 2801 p.p. (BM), Duse Bay, [6333'S: 5722'W], 13 November 1945, G. Davies F.I.D.S. D2632 (BM, FH), Duse Bay, View Point [6336'S: 5701'W], c. 15 m, 10 November 1945, /. M. Lamb F.I.D.S. D2622 (BM), Crown Prince Gustav Channel, Beak I. [6337'S: 5720'W], c. 9m, 28 December 1945, /. M. Lamb F.I.D.S. D2712 (BM), Egg I. [6341'S: 5742'W], 120 m, 11 November 1945, /. M. Lamb F.I.D.S. D2756 (+ Race 2) (BM), Pitt Point [c. 63S: 58W], c. 30 m, 17 August 1945, /. M. Lamb Operation Tabarin 2448 (BM), Operation Tabarin 2452 p.p. (BM), Operation Tabarin 2454 p.p. (BM), Operation Tabarin 2456 p.p. (BM), Church Point, [6341'S: 5754'W], 75 m, 10 December 1946, J. D. Andrew F.I.D.S. D114-2 (BM), F.I.D.S. D114-3 (BM), foot of East Russell Glacier, [6344'S: 5817'W], 550 m, 14 December 1946, A. Reece F.I.D.S. D369-8a (BM), 'Mt. Fitzroy' [6348'S: 5829'W], 240m, 19 December 1946, A. Reece F.I.D.S. D37 (BM), Vega I. [6350'S: 5725'W], False Island Point, 6-18 m, 5 December 1945, /. M. Lamb Operation Tabarin 2724 (BM), D2722 (BM), Tabarin I. [6352'S: 5700'W], 150 m, 6 January 1946, E. H. Back Operation Tabarin 2852 (BM); James Ross I. [6347'S: 5747'W], Cape Lachman, c. 30 m, 21 November 1945, /. M. Lamb F.I.D.S. D2677 (BM p.p. , CANL 17256), F.I.D.S. D2778 (BM), James Ross I., the Naze, 6-22-5 m, 26 November 1945, /. M. Lamb Operation Tabarin 2789 p.p. (+ Race 2) (BM); Palmer Archipelago, Wiencke I. [6449'S: 6322'W], E. of Wall Range, 370 m, 8 October 1944, /. M. Lamb F.I.D.S. A1323 (BM); Fallieres Coast, Adelaide I. [6715'S: 6830'W], 'Stork' nunatak, 550 m, 9 March 1981, R. 1. L. Smith 3416 (AAS); Fallieres Coast, Marguerite Bay, Bourgeois Fjord [6740'S: 6705'W], 15 December 1936, B.G.L.E. 1518-3 (BM), Ridge I. [6742'S: 6706'W], 4 August 1936, B.G.L.E. 1484-3 (BM), Jenny I. [6744'S: 6825'W], E. coast, 15 m, 30 January 1961, B. J. Taylor 444b (AAS), Horseshoe I. [6751'S: 6712'W], (type locality), 7-5 m, 23 February 1965, R. E. Longton 1275 (BM), 24 February 1967, R. I. L. Smith 829 (fertile) (AAS), 6-10 m, 22 February 1977, R. I. L. Smith 2165 (AAS), between Lystad Bay and former B. A. S. station, 20-30 m, 22 February 1977, R.I.L. Smith 2168 (AAS, BM), Camp Point, (between Square Bay and Calmette Bay), [6758'S: 6719'W], 27 October 1949, B. Stonehouse F.I.D.S. E627a (BM), Debenham Is. [6808'S: 6707'W], 39 m, 7 December 1947, B. Stonehouse F.I.D.S. E1116 (BM), 104 F. J. WALKER Neny Fjord, Red Rock Ridge [6818'S: 6705'W], 120 m, 19 January 1948, A. R. C. Butson F.I.D.S. E1158 p.p. (BM), c. 12 m, 19 January 1948, B. Stonehouse F.I.D.S. E1176 (BM), Refuge I. [6821'S: 6710'W], 21-36 m, 29 November 1940, H. M. Bryant U.S.A.S. 32 (US); unrealised, Terra Firma Land', 21 June 1936,B.G.L.E. 1482 (BM). Uncertain determination ARGENTINA. Tierra del Fuego: Sierra Alvear, S. slope, above Las Cotorras (c. 20 km ENE. of Ushuaia), 900-1000 m, 7 February 1940, R. Santesson 640e (640a p.p.) (S). Race 2 CHILE. Aisen: Lago San Martin, Glaciares, c. 1500 m, 13 February 1933, A. Donats.n. (H). ARGENTINA. Rio Negro: Parque Nacional Nahuel Huapi, Cerro Catedral, c. 2225 m, 6 January 1974, A. Henssen & G. Vobis 24677f (MB). Santa Cruz: Rio Fosiles, c. 1000 m, April 1905, P. Dusen s.n. p.p. (FI). ANTARCTIC PENINSULA. Graham Land: Joinville I. , N. end Active Sound [6318'S: 5756'W], 200 m, March 1981, R. I. L. Smith 3736 p.p. (AAS); Trinity Peninsula, Cape Longing [6433'S: 5850'W], 15-30 m, March 1961, J. Killingbeck 84 (AAS, BM, FH), 85 (AAS), 86 (AAS, FH), Duse Bay (as Race 1) F.I.D.S. D109-6 p.p. (BM), Crown Prince Gustav Channel, Egg I. [6341'S: 5742'W], 120 m, (as Race 1) F.I.D.S. D2756 (FH), Alectoria I. [6359'S: 5837'W], 30-60 m, 18 August 1945, /. M. Lamb F.I.D.S. D2467 (BM), foot of East Russell Glacier [6344'S: 5817'W], 14 December 1946, A. Reece F.I.D.S. D369-12a (FH), Vega I. [6350'S: 5725'W], False Island Point, 6-18 m, 5 December 1945, I. M. Lamb F.I.D.S. D2718 (BM), James Ross I. [6355'S: 57'40W], Herbert Sound, c. 90 m, 23 November 1945, V. RussellF.I.D.S. D2734 (BM), James Ross I., Cape Gage [6412'S: 5816'W], c. 6 m, 29 November 1945, /. M. Lamb F.I.D.S. D2843 (BM); Fallieres Coast, Marguerite Bay, Neny Fjord, Roman Four Promontory [6813'S: 6658'W], 27 m, 8 December 1947, B. Stonehouse F.I.D.S. E1072a (BM); George VI Sound, Alexander I. [7211'S: 6905'W], Stephenson nunatak, c. 520 m, 5 December 1949, V. E. Fuchs & R. J. Adie F.I.D.S. E616-7 (BM), Alexander I., SE. corner, 385 m, 4 December 1949, V. E. Fuchs et al. F.I.D.S. E612(BM). Uncertain determination ARGENTINA. Tierra del Fuego: Sierra Sorondo, N. slope above Las Cotorras (c. 20 km ENE. of Ushuaia), 800 m, 6 February 1940, R. Santesson 641c p.p. (S). For details of specimens of uncertain determination (Race 2) from the Antarctic peninsula see specimens in AAS, BM and FH and list held in BM. 13. Usnea subcapillaris (D. Galloway) F. J. Walker, comb. nov. Fig. 32 Usnea ciliata var. subcapillaris D. Galloway in N.Z. Jl Bot. 6: 470 (1968). - Neuropogon subcapillaris (D. Galloway) D. Galloway in N.Z. Jl Bot. 21: 195 (1983). Type: New Zealand, South Island, Otago, Remarkables Range, 2000 m, on rock, February 1968, D. J. Galloway s.n. (CHR 343756 [OTA 68-222]! - holotype; CHR! - isotypes (2), BM! - isotype). [TLC: norstictic acid, salazinic acid, protocetraric acid, usnic acid.] Description: Thallus (2)-5(-15) cm, arising from a delimited, pigmented holdfast, pendu- lous, rarely subdecumbent and spreading. Branching extensive, divergent, dichotomous, with numerous extended, delicate, capillaceous secondary branches predominating, and short attenuate, deflexed laterals; usually lacking fibrils. Branches terete, yellow-green, con- tinuously pigmented or variegated violaceous black towards the apices. Cortex thin. Surface smooth, waxy, epapillate, easily fracturing, forming regular black-edged annulations. Medulla sublax towards the axis; axis occupying 0-3-0-5 of the branch diameter. Soredia, pseudoisidia and isidia absent. Apothecia rare, lateral, as in U. ciliata. Pycnidia not seen. TLC: (1) norstictic acid, salazinic acid, protocetraric acid, usnic acid; (2) squamatic acid, hypothamnolic acid, usnic acid; (3) psoromic acid, 2'-O-demethylpsoromic acid, usnic acid. Distinguishing features: Usnea subcapillaris is characterised by its subdecumbent to pendulous habit, and a richly branched thallus with numerous, extended, fragile, divergent, flexuose- capillaceous secondary branches. It has a waxy, pigmented-annulate surface, a sublax medulla, containing depsides or depsidones, and occasionally apothecia of the 'ciliata' -type. Distribution: Usnea subcapillaris occurs mainly in the South Island of New Zealand; it is widely distributed in alpine localities, generally between 1000 and 2000 m, often alongside U. ciliata, although tending to favour slightly more sheltered aspects. This species is only known in USNEA SUBGENUS NEUROPOGON 105 Fig. 32 Usnea subcapillaris. Holotype of Neuropogon ciliatus var. subcapillaris D. Galloway (CHR). Top. Whole thallus xl. Bottom. Detail of apothecia and cortical annulations xlO. 106 F. J. WALKER the North Island from an unlocalised gathering (Colenso 2086, BM!, s.n., PC!) which was probably collected on the Ruahine Range. It is rare in Tasmania, being confined to a few high-altitude locations and the species was first identified from there by Bratt & Cashin (1976). Fig. 6. Chemistry: Three chemical races have now been identified in Usnea subcapillaris (Galloway, 1984). Race 1 predominates whilst the other two races are very rarely encountered and are restricted to a few localities in mixed populations with Race 1 . Protocetraric acid is only present in high concentrations in Race 1 and such thalli subsequently lack, or only contain traces of, norstictic and/or salazinic acids (for example, CHR 343473, herb. Bartlett 24724c). Interesting- ly, such chemical diversity does not occur in U. ciliata and U, pseudocapillaris or in the majority of Australasian populations of U. acromelana. The presence of squamatic and hypothamnolic acids (Race 2) in the medulla of this species is unique within the subgenus and may easily be demonstrated by a brilliant white medullary fluorescence under UV light and a K+ purple reaction. Race 3, containing psoromic acid, is only known from a few localities, one of which, Mt. Hutt (CHR 343463, CHR 343483!), is unusual since some thalli have a mixed chemistry of two, or even all three, races. Variation: The accounts of Usnea ciliata (p. 76), U. acromelana (p. 50), and U. pseudocapil- laris (p. 90) may be referred to for much of the variation in branching, pigmentation, morphology, and anatomy found in U. subcapillaris. This species most closely resembles U. ciliata but differs in the pendulous habit, a more or less confined, rarely proliferating holdfast, repeated, divergent branching, and extensive, capillaceous, interwoven secondary branches. The medulla may also be slightly laxer, occupying a larger proportion of the branch diameter. The surface has more regularly spaced, conspicuous annulations which sometimes result in thallus fragmentation. When frequent on main branches, such annulations produce a character- istic segmented appearance, sometimes resulting in slight subsequent constrictions. Secondary branches are more extensive than in Usnea pseudocapillaris , tending to form the most distinctive feature of the thallus, and are often up to two or three times the length of primary branches. The fragile and fragmentary nature of the secondary branches becomes more pronounced on storage. Adjacent thalli of richly branched specimens, with diverging ultimate branches, often become mutually entangled. U. pseudocapillaris and U. subcapillaris are only likely to be confused when they are immature and lack any reproductive structures. In less typical forms the secondary branches are somewhat coarser, although thalli are still moderately to richly branched and retain the characteristic divergent habit with short, attenuate ultimate branches. This is a particular feature of some Tasmanian collections in which there is a gradual transition between the coarse primary branches and the capillaceous secondary branches which are shorter, broader and more frequently divided than in the majority of New Zealand specimens examined. Apothecia are rare in New Zealand whilst in Tasmania they are only known from a single collection (HO 35239!). They are usually lateral on secondary branches and have a subtending, extended apical branch or spur which frequently divides. They are usually smaller, c. 5 mm, than in Usnea ciliata and pigmentation of the disc may rarely be undeveloped. Marginal excipular rays vary in number but are usually longer, finer and may divide, thus resembling secondary branches. Their scarcity may be a reflection of the ease at which the thallus fragments and regenerates, thus providing an alternative method of reproduction. This might explain the wider distribution of this species than U. ciliata in Australasia. Species concept: Usnea subcapillaris is now considered (Galloway, 1983) to be a distinct species from U. ciliata and a parallel can now be found between the related species, U. pseudocapillaris and U. acromelana. U. subcapillaris may also be regarded as the fertile counterpart to U. pseudocapillaris within the U. ciliata complex. It may be distinguished from U. ciliata by the pendulous or subdecumbent habit resulting from repeated, more or less divergent, dichotomous, division of extensive, capillaceous, secondary branches which give a fragile, lax, combed or frequently entangled appearance. Other characteristic features include the shorter, USNEA SUBGENUS NEUROPOGON 107 often widely divaricately branched main branches and sometimes the slightly laxer nature of the medulla and deep fissured annulations. It is extremely rare for thalli to be encountered that are difficult to assign to U. subcapillaris or U. ciliata. Coarser, suberect, or subdecumbent forms from Tasmania are included within the accepted variation. There does not appear to be any correlation between chemical race and morphological variation. Usnea subcapillaris may be distinguished from the other fertile species of Neuropogon by its restricted distribution besides morphological characters. Selected specimens examined Racel AUSTRALIA. Tasmania: Mt. Wellington, summit, 1250 m, 15 March 1970, G. Degelius A-414 (herb. Degelius), 1250 m, 20 December 1964, G. C. Bratt 1872a (CHR 343280), Table Mountain' [Mt. Wellington], (part of type gathering of U. acromelana var. decipiens Lamb (BM)); Central Plateau, Mt. Penny, near Arthur's Lake, 1134 m, 4 April 1969, G. C. Bratt & K. M. Mackay 69/161 (CHR 343323, HO 35239 p.p.); Table Mountain, 1095 m, 18 June 1972, G. C. Bratt & J. A. Cashin Bratt 72/393 p.p. (HO 35178), Bratt 394 (HO 35179); Sandbank Tier, 5 April 1969, G. C. Bratt 69/212 (HO 35195); Seager's Lookout, c. 10 km NNW. of Maydena, 1240 m, 30 May 1970, G. C. Bratt & F. N. Larkin Bratt 70/680b p.p. (HO 35236 p.p.). NEW ZEALAND. North Island, unrealised [Ruahine Range?], Colenso 2086 (BM), s.n. (PC). South Island. Nelson: saddle between Mt. Aorere and Mt. Cobb, c. 1580 m, 19 December 1982, /. K. Bartlett241246p.p. (BM); Mt. Robert, Lake Rotoiti, 1430m, 16 January 1960, D. Scott s.n. (BM, OTA); St. Arnaud Range, Lake Rotoiti, 1520 m, February 1959, Mason 647 (BM, OTA), St. Arnaud Range, 1680 m, 22 December 1967, A. F. Mark s.n. (CHR 343290 p.p., CHR 343476); Mt. Technical, above Lewis Pass, 14 January 1979, D. J. Galloway s.n. (CHR 343275); Lake Sylvester, 1620 m, 18 December 1967, A. F. Mark s.n. (CHR 343488); Lake Peel, 1430 m, /. K. Bartlett s.n. (CHR 343223). Marlborough: Black Birch Range, nr. Altimarlock trig, 10 January 1971, L. B. Moore s.n. (CHR 162621); Inland Kaikoura Range, Mt. Mitre, 2 January 1954, R. Mason & D. R. McQueen 2760 (CHR 160686). Canterbury: Mackenzie Country, Lower Godley River Valley, 1830 m, 21 December 1958, D. Scott 99 (BM, OTA); Torlesse Range, Foggy Peak, summit rocks, 1680 m, 18 December 1962, P. W. James 1918 (BM), 1680 m, November 1972, D. J. Galloway s.n. (CHR 343491), 12 November 1972, G. C. flratf 72/1880b (HO 35170); Kirkliston Range, 1680-1830 m, 25 March 1978, D. J. Galloway s.n. (CHR 343178, CHR 343252 p.p.), 1830 m, December 1964, C. J. Burrows s.n. (CHR 343477); Mt. Peel, 1740 m, January 1972, D. J. Galloway s.n. (CHR 343437 p.p., CHR 343499); Mt. Cook National Park, Mt. Annan, 2740 m, R. B. Filson s.n. (MEL 1026826); Arthur's Pass, Mt. Bealey, 1680 m, 19 August 1968, L. D. Kennedy s.n. (CHR 343452); Two Thumb Range, Mt. Richmond, November 1968, A F. Marks.n. (CHR 343489 p. p.). Otago: Rock and Pillar Range, 1160-1190 m, 18 September 1981, F. J. Walker s.n. (BM), S. end, 980 m, June 1967, T. S. Choate s.n. (CHR 343347), 1480 m, February 1967, D. J. Galloway s.n. (CHR 343490), 980 m, June 1967, D. J. Galloway s.n. (CHR 342766); Young Range, 1430 m, March 1968, D. J. Galloway s.n. (CHR 343447 p.p., CHR 343472); Old Man Range, 1490 m, November 1962, P. W. James 443b (BM), 1130 m, 1 February 1963, P. W. James 1579 (BM), 1220 m, 1 February 1963, P. W. James 1597 (BM), 1680 m, April 1968, D. J. Galloway s.n. (CHR 343467, CHR 343478, CHR 343497), 1968, D. J. Galloway s.n. (CHR 343492), 1490m, 30 August 1968, L. D. Kennedy s.n. (CHR 343479), 1520m, A. F. Marks.n. (CHR 343470); Mt. Pisa, 1920 m, March 1968, D. J. Galloway s.n. (CHR 343468, CHR 343475, CHR 343480, CHR 343494), 1860 m, March 1968, D. J. Galloway s.n. (CHR 343473, CHR 343486), March 1968, D. J. Galloway s.n. (CHR 343448); Rees Valley Slopes, Leary Peak, 1680 m, January 1968, D. J. Galloway s.n. (CHR 343474, CHR 343287); Coronet Peak, 1520 m, February 1968, D. J. Galloway s.n. (CHR 343471); Humboldt Mtns, Mt. Minos, 2010 m, 1 January 1970, D. J. Galloway s.n. (CHR 343338, CHR 343488, CHR 343499); Mt. Erebus, 1980 m, 14 January 1970, D. J. Galloway s.n. (CHR 343754); Bedford Valley, 1680 m, 26 December 1970, D. J. Galloway s.n. (CHR 343482); Remarkables, 1980 m, March 1966, D. J. Galloway s.n. (CHR 343487); Coronet Peak, 1520 m, February 1968, D. J. Galloway s.n. (CHR 343755); Shepherd's Pass, 1980 m, January 1968, D. J. Galloway s.n. (CHR 343450); Mt. Brewster, 2130 m, March 1968, D. J. Galloway s.n. (CHR 343493); Mt. Roy, 1 November 1972, G. C. Bratt 72/1506 (HO 35175); Unnamed Virgin, N. of Pope's Nose, 2350 m, 11 February 1969, L. D. Kennedy s.n. (CHR 342787 p.p.); Mt. Pisgah, c. 1600 m,/. 5. Thomson 1983 (CHR 343797). Race 2 NEW ZEALAND. South Island, Nelson: above Cobb Lake, 1070 m, December 1967, A. F. Mark s.n. (CHR 343498). Canterbury: Four Peaks Range, Blue Mountain, c. 1640 m, 23 April 1979, D. J. Galloway s.n. (CHR 343228), 23 April 1979, D. J. Galloway s.n. (CHR 343179); Mt. Peel, 1740 m, January 1972, D. 108 F. J. WALKER J. Galloway s.n. (CHR 343500); Mt. Hutt, 1520 m, C. J. Burrows s.n. (CHR 343495 p.p.), 1830 m, C. J. Burrows s.n. (CHR 343483 p.p.); Two Thumb Range, Mt. Richmond, November 1968, A. F. Mark s.n. (CHR 343489 p.p.). Otago: Sugarloaf Saddle, 1070m, May 1966, D.J. Galloway s.n. (CHR 343446 p. p.), 1280 m, February 1968, D. J. Galloway s.n. (CHR 343408 p.p.); Young Range [as Race 1] (CHR 343477 p.p.), 1520 m, March 1968, R. Nilsson s.n. (CHR 343481); Humboldt Mtns, Mt. Nox, 1950 m, 31 December 1969, D. J. Galloway s.n. (CHR 342791, CHR 343469). Race 3 NEW ZEALAND. South Island, Canterbury: Mt. Hutt, C. J. Burrows s.n. [as Race 2] (CHR 343463 p. p., CHR 343483 p.p.); Four Peaks Range, Tripps Peak, 24 April 1979, D. J. Galloway s.n. (CHR 343337 p.p.). Otago: Rock and Pillar Range, 1280 m, December 1964, D. J. Galloway s.n. (CHR 343286, CHR 343451), 1400 m, July 1968, L. D. Kennedy s.n. (CHR 343496), 1400 m, February 1967, D. J. Galloway s.n. (BM). For further localities see Galloway (1968) and lists held in BM. 14. Usnea lay loriiJ.D. Hook. Figs33-34 in Hooker & Taylor in Lond. J. Bot. 3: 657 (1844). - Alectoria taylorii (J. D. Hook.) Nyl. in Mem. Soc. Imp. Sci. Nat. Cherbourg 5: 98 (1857). - Neuropogon taylorii (J. D. Hook.) Nyl., Synops. Lich. 1: 273 (1860). Type: Kerguelen's Land, Antarctic Expedition of H. M.S. Discovery ships 'Erebus' and 'Terror' 1839-1843, /. D. Hooker (BM! - lectotype, selected here; BM!, E!, FH, not seen, M! - isolectotypes). [TLC: no medullary substances, usnic acid.] Note: The epithet 'taylori' is corrected to 'taylorii' under Article 73.10 following Recommendation 73C.1. Many original collections were widely distributed by Hooker and the species is best lectotypified on the collection illustrated by Lamb (1939o) in BM in preference to the herb. Taylor collection (FH) indicated by Dodge (1948). Description: Thallus (4)-5-7(-10) cm, arising from a proliferating, rarely pigmented, holdfast, erect, monopodial to dichotomous, infrequently branched above with flexuose, tapering to subcornute apices. Branches terete, corneous, yellow-green, continuously or irregularly pig- mented with black pigment towards the apices. Cortex thin. Surface subnitid, epapillate, smooth but mottled with slightly raised, unpigmented maculae. Medulla compact, reduced, invading axial cavities. Axis occupying c. 0-9 of the branch diameter, partially sub-divided, forming several strands, frequently abutting the cortex and protruding to form pale maculae. Soredia, pseudoisidia and isidia absent. Apothecia frequent, subterminal with a prominent, geniculate spur or rarely lateral and subsessile; cupular, plane or sinuose-deflexed on maturity. Disc black, excipulum smooth or minutely faveolate, margin excluded, rays absent. Pycnidia rare towards apices. TLC: (1) fumarprotocetraric acid (trace), usnic acid; (2) no medullary substances, usnic acid. Distinguishing features: Usnea taylorii is characterised by its erect, proliferating habit and a monopodial to infrequently branched, often extensively pigmented, corneous thallus; a subnitid maculate surface, a much reduced medulla, usually lacking medullary substances, a broad, sub-divided axis and subterminal, geniculate apothecia with a black disc lacking excipular rays. Distribution: Usnea taylorii is endemic to Kerguelia. Material has been examined from lies Kerguelen and lies Crozet and Dodge & Rudolph (1955) have recorded the species from Heard Island (A.N.A.R.E. 250, MEL, not seen). Specimens from South America misidentified as U. taylorii by Rasanen (H!) are referable to U. perpusilla. Specimens collected by Hooker, apparently from South America, are erroneously labelled (see p. 69). U. taylorii replaces U. aurantiaco-atra in Kerguelia and is often found in communities there with U. antarctica and U. trachycarpa. Fig. 4. Chemistry: All collections examined from lies Kerguelen lack medullary substances, including fatty acids. Race 1, containing low concentrations of fumarprotocetraric acid, is only recorded from the single collection from lies Crozet. Variation: This is probably the most distinctive species of the subgenus. It has a unique axis structure throughout the thallus giving rise to the formation of separate axial strands that are USNEA SUBGENUS NEUROPOGON 109 Lecto-Fype Specimen Fig. 33 Lectotype of Usnea taylorii J. D. Hook. (BM) x 1. invaded by medullary tissue, resembling a coaxial cable; as illustrated by Rienke (1895). The extrusion of the axis through the narrow medulla and the otherwise cortex produces the characteristic pale, raised maculae that resemble the pseudocyphellae olAlectoria ochroleuca on an otherwise smooth surface. This often results in a mottled, rather than variegated, pattern on the extensively pigmented ultimate branches. Very rarely these maculae may form papilla- like structures which remain ecorticate. Collections are mostly uniform, the only variation being in the extent of thallus branching and pigmentation which range from scantily pigmented unbranched thalli to more richly branched forms that are extensively pigmented towards the apices. Main branches are usually thick and corneous, and are often flexuose with tapering, subcornute apices, giving the thallus a sinuose habit. Apothecia are characteristically subterminal and are particularly conspicuous on sparingly branched thalli. In contrast, in more richly branched thalli apothecia tend to be smaller, slightly cupular, subsessile, and lateral. Pycnidia are rare, and may be confused with a range of 110 F. J. WALKER Fig. 34 Usnea taylorii. Left. Detail of maculae. Challenger Exped. 1874, Molseley (BM) xlO. Right. Detail of apothecia. Is. Kerguelen, January 1960, Tilman (BM) xlO. parasites, including Lecidea alectoriae which was originally described (W. Lindsay, 1859) from this species. Species concept: Usnea taylorii appears to be a highly evolved species which may be closely related to U. aurantiaco-atra. U. taylorii replaces U. aurantiaco-atra in the Kerguelen region and shares several morphological characteristics besides the existence of a race containing fumarpro- tocetraric acid. U. aurantiaco-atra does not occur east of Bouvet0y where populations lack depsidones, have a very thick axis, and a reduction in the extent of papillation. A sorediate counterpart is not known, although some populations of Usnea antarctica, previously described as distinct taxa, for example U. crombiei (Dodge, 1948) and U. insularis (Lamb, 1939a), have slightly irregular but undivided axes. Usnea taylorii may be distinguished from all other fertile species within the subgenus by the unique branch anatomy with the production of maculae. Specimens examined Racel Is. CROZET. (unrealised) c. 610 m, 1959-60, W. H. Tilman s.n. (BM). Race 2 Is. KERGUELEN. Cliffs above Lake du Val Studer, 11 February 1963, R. B. Filson 4665 p.p. (BM); Swain's Bay and Observatory Bay, Venus Transit Exped. 1874-75, A. E. Eaton s.n. (BM, E, M), A. Balfour s.n. (E); Royal Sound, A. E. Eaton s.n. (UPS), Royal Sound, nr. Port Jeanne d'Arc, c. 460 m, 11 February 1930, B.A.N.Z.A.R.E. B176 (AD, FH); Crater Hill, Christmas Harbour and Cumberland Bay, Ross' Antarctic Exped. 1839-43, R. McCormick s.n. (BM); Baie des Baleiniers, Petit Mt. Ballon, 300 m, January 1960, W. H. Tilman s.n. (BM, FH); unrealised - Challenger Exped., January 1874, H. N. Moseley s.n. (BM); Challenger Exped. (GLAM NHB 1927-8-348, GLAM NHB 1927-8-349); O. Ring s.n. (O); March 1931, A de la Rue s.n. (UPS); December 1898, W. Schimpers.n. (UPS). HEARD I. see Dodge & Rudolph (1955). 15. Usnea trachycarpa (Stirton) Mull. Arg. Fig. 35 in Nuovo G. hot. ital. 21: 37 (1889). - Neuropogon trachycarpus Stirton in Scott. Nat. 6: 105 (1881). Type: Kerguelen's Land (So. Antarctic), [January, 1875], Challenger Exped. 6 (BM! -holotype; BM! - isotype). [TLC: norstictic acid, fatty acids, usnic acid.] (see Note) USNEA SUBGENUS NEUROPOGON 111 Usnea naumannii Miill. Arg. in Bot. Jb. 4: 54 (1883). Type: Kerguelen, Betsys Cove, 1875, Dr Naumann (G! - holotype). [TLC: norstictic acid, fatty acids, usnic acid.] Usnea melaxantha var. angulosa Miill. Arg. in Flora, Jena 71: 528 (1888). - Usnea sulphurea var. angulosa (Mull. Arg.) Zahlbr., Cat. Lich. Univers. 6: 603 (1930). Type: Argentina, Patagonia [unlocalised], Claraz (G! -holotype). [TLC: norstictic acid, usnic acid.] Usnea trachycarpa var. sublaevis Miill. Arg. in Hedwigia 34: 139 (1895). Type: Lich. Exot. 141, [Argentina] Patagonia, Moreno, 1894, Otto Kuntze 713 (G! - holotype). [TLC: no medullary subst- ances, usnic acid.] Usnea trachycarpa var. trachycarpoides Vainio, Res. Voy. S. Y. Belgica, 1897-99, Botan., Lichens: 12 (1903). - Neuropogon trachycarpus f. trachycarpoides (Vainio) Lamb in /. Linn. Soc. (Bot.) 52: 233 (1939). - Usnea trachycarpoides (Vainio) Dodge, Lich. Fl. Antarct. Cont.: 238 (1973). Type: Argentine, He des Etats, Golfe Saint Jean, lat. 5524'S, sur les parois verticales, 1898 M. Emile G. Racovitza 182 (TUR! - holotype). [TLC: norstictic acid, salazinic acid, protocetraric acid, usnic acid.] Usnea hyyppae Rasanen in Suomal. elain-ja kasvit. Seur. van. kasvit. Julk. 2(1): 11 (1932). Type: [Argentina] Fuegia media, Cerro Milladeo, reg. alp. 800 m.s.m., Expeditio Fennica 1928-29, H. Roivainen (H! - holotype; H! - isotype). [TLC: no medullary substances, usnic acid.] Usnea taylorii var. subciliata Rasanen in Suomal. elain-ja kasvit. Seur. van. kasvit. Julk. 2(1): 11 (1932). Type: [Chile] Fugia media, Cerro Chico, reg. alp. 550 m.s.m. , Expeditio Fennica 1928-29, H. Roivainen (H! -holotype). [TLC: no medullary substances, usnic acid.] Neuropogon substrigulosus Lamb in /. Linn. Soc. (Bot.) 52: 231 (1939). - Neuropogon trachycarpus f. substrigulosus (Lamb) Lamb in Lilloa 14: 156 (1948). - Usnea substrigulosa (Lamb) Dodge. Lich. Fl. Antarct. Cont: 237 (1973). Type: Falkland Islands, (no collector given) (BM! - holotype). [TLC: norstictic acid, salazinic acid, protocetraric acid, fatty acids, usnic acid.] Neuropogon trachycarpus f. elatior Lamb in Lilloa 14: 157 (1948). Type: [Argentina] Tierra del Fuego, Sierra Alvear, the Southern slope, above Las Cotorras (c. 20 km ENE. of Ushuaia), rocks in the alpine region, 900-1000 m, 1940, R. Santesson 640b, (S! - holotype; S!, CANL 17258! - isotypes). [TLC: no medullary substances, usnic acid.] Note: The specimens figured by Lamb (1939a) are only part of the holotype material of Neuropogon trachycarpus and were selected from Stirton's herbarium by Miss A. L. Smith and are here regarded as an isotype. The original capsule bearing Stirton's handwriting was mislaid and not traced by Lamb or Dodge (1948). This capsule has recently been traced in the BM and bears the inscription 'Kerguelen's Land (So. Antarctic) Challenger Expedition (6) - melaxanthus trachycarpus (Strn.) axis rather solid, compact. Med. fibres arachnoid. K-; I-. Type.' This material forms the holotype and consists of two slightly larger thalli than the isotype, one of which bears three small apothecia. Description: Thallus (2-3)-4(-7-9) cm, arising from a proliferating, unpigmented, holdfast, erect, monopodial or dichotomous, infrequently to moderately branched above. Fibrils numerous on all branches, short, c. 5 mm, spreading, variegated or continuously black- pigmented. Branches terete or rarely slightly angular, yellow-green, continuously pigmented towards the apices. Cortex variable in thickness. Surface matt, smooth at the base becoming markedly faveolate to richly papillate or scabrid with numerous conspicuous fibrils above. Medulla lax or sublax, axis occupying 0-3-0-5 of the branch diameter. Soredia, pseudoisidia and isidia absent. Apothecia frequent, subterminal, cupular, expanding on maturity. Disc rufous brown, excipulum faveolate to verrucose-papillate with numerous marginal, attenuate, pigmented rays. Pycnidia infrequent towards apices. TLC: (1) norstictic acid, salazinic acid, protocetraric acid, 2-6 fatty acids (murolic acid complex), usnic acid; (2) psoromic acid, 2'-O-demethylpsoromic acid, 2-6 fatty acids (murolic acid complex), usnic acid; (3) no medullary substances, 2-6 fatty acids (murolic acid complex), usnic acid. Distinguishing features: Usnea trachycarpa is characterised by its erect, proliferating habit, an infrequently branched thallus with numerous fibrils, a faveolate-papillate surface, a lax to sublax, arachnoid, medulla, often containing depsidones and fatty acids, a thin axis, and subterminal apothecia with a rufous brown disc with numerous marginal and submarginal excipular rays. Distribution: Usnea trachycarpa is known from southern South America (Patagonia and Tierra del Fuego) as far north as latitude c. 42S, the Falkland Islands and lies Kerguelen, and has also been recorded by Dodge & Rudolph (1955) from Heard Island (Gilchrist 2, MEL, 112 F. J. WALKER Fig. 35 Usnea trachycarpa. Top. Holotype oiNeuropogon trachycarpus Stirton (BM) x 1 . Bottom. Detail of apothecium. Patagonia, James 28 (BM) x 10. USNEA SUBGENUS NEUROPOGON 113 not seen). A single, sterile specimen from the Antarctic peninsula is tentatively referred to this species (R. I. L. Smith 3453 p.p. , AAS!), although it is not known elsewhere in Antarctica or the islands of the Scotia Arc. A thallus fragment amongst the type of U. acromelana var. decipiens (BM!) from Tasmania appears to belong to this species, but is probably misplaced since the species is not otherwise known in Australasia. Usnea trachycarpa is a characteristic species of dry, exposed conditions and is frequently found at relatively low altitudes, for example near sea-level to c. 200 m on lies Kerguelen and the Falkland Islands. It has a broader altitudinal range in South America, ranging from 30 to 1000 m in Tierra del Fuego and Santa Cruz up to c. 2250 m further north in Rio Negro. In the Falkland Islands this species is often replaced by Usnea aurantiaco-atra whilst in southern South America it is sometimes found in communities with U. aurantiaco-atra and U. perpusilla and a range of asexual species, particularly U. patagonica. Fig. 5. Chemistry: The most common race throughout the range of Usnea trachycarpa is Race 1, containing norstictic acid, usually with salazinic acid, and frequently with detectable (by TLC) concentrations of connorstictic acid and/or protocetraric acid. An additional unknown subst- ance that was yellow after charring (Rf class TD A 1 , HEF 2-3) was found in some thalli from lies Kerguelen, but did not appear to be of taxonomic significance. Race 2, containing psoromic acid, is very rare and is only known from a few collections from a single locality in Argentina. Up to six fatty acids of the murolic acid complex (see p. 14) occur in Usnea trachycarpa and have been demonstrated by TLC in all three chemical races, although their presence and number may vary within a single collection. They are particularly persistant in Race 1 from lies Kerguelen, whilst from elsewhere their distribution appears to be more spasmodic, for instance only occurring in about half a selection of 30 South American specimens tested. The UV+ unknowns were only rarely detected in this species. Variation: Usnea trachycarpa is, especially when fertile, an easily recognised and uniform species throughout its range. The extent of branching is reminiscent of U. ciliata (p. 76). Thalli range from virtually monopodial forms, often scantily pigmented at the apices of branches and fibrils, to more richly branched, extensively pigmented forms that are characteristic of higher altitudes and more exposed localities. Papillae are only pigmented in the more extensively pigmented, high-altitude forms. Some populations are more robust, up to c. 9 cm tall, and have previously been recognised as distinct taxa, for example Neuropogon substrigulosus and N. trachycarpus f . elatior. The number and extent of fibrils is variable although they are nearly always present; forming a characteristic feature of the species. The surface is smooth, matt and free from fibrils near the holdfast, but becomes papillate-faveolate above with the production of small, uniform fibrils of c. 5 mm in length. Only very rarely is the thallus smooth to minutely papillate throughout, with fewer fibrils. Various taxa have been described based on the extent of fibril production and range from specimens virtually free of fibrils, including Usnea trachycarpa var. sublaevis and U. taylorii var. subciliata from Argentina, to extensively fibrillar forms, including Neuropogon substrigulo- sus from the Falkland Islands^ The type specimen from lies Kerguelen, although small, is fairly typical of the species. However, the medulla is more compact than is frequently found in more robust specimens, as seen in the type of Usnea naumannii from the same island group, although it still possesses an arachnoid texture, and slight inflation and characteristic faveolation of main branches. The degree of laxness and hence relative widths of medulla and axis, are the most variable feature of this species. Many gatherings from Patagonia are typical of a laxer form, for example the types of U. melaxantha var. angulosa and U. hyyppae, whilst in Neuropogon trachycarpus f . elatior the medulla is extensive and lax with the axis only occupying about a third of the branch diameter. Apothecia are uniform and are normally subterminal. When young they are cupular but are partially expanded on maturity, sometimes becoming irregular. The disc is always rufous brown in freshly collected specimens and quite distinct from the colour of the thallus. The disc only rarely becomes partly blackened when moribund. The extent and length of excipular rays may 114 F. J. WALKER vary. These are usually fine, short, unbranched, black-pigmented or variegated, and are numerous around the margin or the excipulum and less frequent over the rest of the surface. Variation in number, thickness and length of rays has led to the description of distinct taxa, including Usnea taylorii var. subciliata and Neuropogon substrigulosus . Species concept: Variation in thallus size, ornamentation, morphology, pigmentation, and anatomy in Usnea trachycarpa is not considered sufficient to recognise additional or infraspecific taxa; a parallel can be drawn within U. aurantiaco-atra. Reduction in the extent of black pigmentation, including the presence of a rufous brown pigment in the apothecial disc, may indicate some affinities with species of Usnea subgenus Usnea (see p. 43), for example U. hieronymii (var. adustd) and U. densirostra, which are saxicolous, have fibrils, a pale buff disc, and exhibit very limited pigmentation. These species have a more northern distribution and are not characteristic of montane areas. Usnea trachycarpa is possibly the primary species of U. subantarctica and may also be closely related to U. patagonica and U. neuropogonoides (see p. 42). It may easily be distinguished from all other fertile species of the subgenus by the rufous brown apothecial disc with numerous short rays and, even when sterile, by the numerous fibrils on all branches. Selected specimens examined Racel CHILE. Magallanes: Magellan Straits, Punta Arenas, Steinmann s.n. (BM, H, M, PC); Natales, Cerro Dorotea, 9 May 1940, R. Santesson 2135 (S), R. Santesson 2136 (S); Tierra del Fuego, Isla Navarino, Puerto Navarino, 10 m, 28 February 1940, R. Santesson 1232a (UPS, S); Cabo de Hornos, Cape Spencer,/. D. Hooker s.n. (BM). ARGENTINA. Santa Cruz: Lago Argentine, Cerro del Fraile, near Estancia Lago Roca, 270 m, 1958-59, P. W. James 764 (BM); Lago Argentine, Cordillera Cristales, Cerro del Fraile, c. 1000 m, 26 December 1958, P. W. James 29 (BM); hills above Calafate, 1235 m, 1958-59, P. W. James 420a (BM); Calafate, 'weg nach' Punta Bandera, Campo Anita Fuss des Co. Moyano, c. 500 m, December 1973, A. Henssen & G. Vobis 245405 (MB); Lago Viedma, 1200 m, 2 April 1903, P. Dusen s.n. (H); Rio Fosiles, c. 1000 m, 1975. P. Dusen s.n. (FH, H, M p.p., WU 2997 p.p.); Depto. Guar Aiken, 60 km SW. of Rio Gallegos, 30 m, 4 January 1939, W. J. Eyerdam et al. 24093 (BM p.p., FH). Tierra del Fuego: Parque Nacional Tierra del Fuego, Ushuaia, 'weg zum' Glacier Martial, c. 750 m, 7 December 1973, A. Henssen & G. Vobis 24420J (MB); Ushuaia, small island in Beagle Channel, 30 January 1940, R. Santesson 525 p.p. (S); Sierra Alvear, S. slope, above Las Cotorras, c. 20 km ENE. of Ushuaia, 650 m, 9 February 1940, R. Santesson 636b (S, UPS). FALKLAND IS. East Falkland Is.: Goose Green, Fish Creek, 9 m, 22 September 1963, R. W. M. Corner 67 (AAS); Mt. Longdon, near Port Stanley, 174 m, 2 April 1967, D. C. Lindsay 1657 (AAS), D. C. Lindsay 1660 (AAS), D. C. Lindsay 1661 (AAS); Fairy Cove, near Port Stanley, near sea-level, 30 January 1946, /. M. Lamb F.I.D.S. 2892 (BM, CANL 17257); N. of Wireless Hill, 100 m, 28 February 1977, R. I. L. Smith 2571 (AAS); Port Stanley, W. Lechler, pi. ins. Maclovian. 69c (BM, M); Murray Heights and Lookout Rocks, Stanley, 30 March 1965, /. Price s.n. (BM). West Falkland Is. : Crooked Inlet, Roy Cove, 31 August 910 (collector unknown) (BM); Cooke Hill and Roy Cove, 1909-11, E. Vallentin 89 (BM). Is. KERGUELEN. Royal Sound, Venus Transit Exped. 1874-75, A. E. Eaton s.n. (BM, E, M, UPS); Swain's Bay, Venus Transit Exped. 1874-75, A. E. Eaton s.n. (BM, E, M); Observatory Bay, Venus Transit Exped. 1874-75, A. E. Eaton s.n. (BM, E, M); Molloy Point, Venus Transit Exped. 1874-75, /. P. Kidder s.n. (FH); Baie Islandaise (Baie des Baleiniers), 210 m, January 1960, W. H. Tilman s.n. (BM, FH); above Lake du Val Studer, 11 February 1963, R. Filson 4665 p.p. (BM). Race 2 ARGENTINA. Santa Cruz: Depto. Guar Aiken [ as Race 1], W. S. Eyerdam et al. 24093 (BM), Guar Aiken, 18 January 1940, R. Santesson 300a (S, UPS), R. Santesson 300b (S). Race 3 CHILE. Aisen: Coyhaique Alto, 1000 m, 18 November 1940, R. Santesson 4600 (S), R. Santesson 4632 p.p. (S), Lich. Austro-amer. ex Herb. Regnelliano 423 (BM, M, UPS, S). ARGENTINA. Neuquen: Paso do Pino Hachado, 13 November 1941, Perez Moreau 5982 (H). Rio Negro: Parque Nacional Nahuel Huapi, Cerro Catedral, c. 2225 m, 6 January 1974, A. Henssen & G. Vobis 24677e (MB); Cerro Leones, 1 January 1974, A. Henssen & G. Vobis 24625 p.p. (MB). Chubut: Rio Corcovado, February 1903, Illin 553 (BM); Pampa Chica, 13 November 1908, C. Skottsberg s.n. (UPS). Santa Cruz; Patagonia, Lago Argentine, C. Megluno s.n. (BM); Rio Gallegos, Estancia Stag River, R. USNEA SUBGENUS NEUROPOGON 115 Tweedie 71 (BM); Rio Fosiles, c. 1000 m, April 1905, P. Dusen s.n. (FI, H, M, UPS). Tierra del Fuego: Parque National Tierra del Fuego, Ushuaia, 'Weg zum' Glacier Martial, c. 750 m, 7 December 1973, A. Henssen & G. Vobis 2443 la (MB); Sierra Alvear (type locality of Neuropogon tr achy carpus f. elatior Lamb) 800-900 m, 7 February 1940, R. Santesson 639b (UPS, S), 650 m, 9 February 1940, R. Santesson 636b p.p. (UPS); Sierra Sorondo, N. slope, above Las Cotorras, c. 20 km ENE. of Ushuaia, 800 m, 6 February 1940, R. Santesson 641b (S, UPS); Monte Martial, above Ushuaia, 700 m, 29 January 1940, R. Santesson 450b (UPS, S). is. KERGUELEN. part of type collection (BM); December 1898, W. Schimper (UPS). Uncertain determination ANTARCTIC PENINSULA. Graham Land: Adelaide I. [c. 6715'S: 6830'W], 'Reptile Ridge', Yellow Bluff, 350 m, 9 March 1981, R. I. L. Smith 3453 p.p. (AAS). For further localities see lists held in BM. HEARD IS. see Dodge & Rudolph (1955). Appendix I. Allied taxa Descriptions and brief notes on the following three taxa, Usnea amblyoclada, U. inermis, and U. torulosa, are appended to assist in the identification of some of the species of Usnea subgenus Usnea that can occur in communities with Neuropogon species and may rarely be somewhat darkly pigmented. In general, this pigmentation is scant in all three and lacks the violaceous lustre which is a characteristic of many species of Neuropogon; more extensive pigmentation only seems to occur in damaged or moribund specimens. In Australasia only two species of the subgenus Usnea, U. torulosa and, to a lesser extent, U. inermis, occupy similar habitats to Neuropogon, both appear to virtually replace the latter subgenus in alpine habitats. They may be distinguished, particularly from species of the U. ciliata complex, by the develop- ment of true isidia. A larger number of saxicolous Usnea species have been described from South America including a well-defined group of related species which Motyka (1938) assigned to the subsection Roccellinae of the Laevigatae. In general they are robust species, often with large amounts of usnic acid, and are rarely pigmented. Examples include U. bogotensis, a yellow-green species with more or less globose soralia, and the non-sorediate, fertile, U. roccellina, which probably form a species pair. Motyka also included the fertile species U. amaliae and U. hieronymii (and its var. adusta) in this subsection. However, U. hieronymii appears to be closely allied to, or possibly the same as, U. densirostra, a species which Motyka (1937) assigned to a completely different subsection, the Densirostrae within the section Setulosae, U. densirostra is the only species in this subsection which could be mistaken for a species of Neuropogon. This species resembles, in particular, U. trachycarpa in habit, but is uniformly grey-green and has a pale, not a rufous brown, apothecial disc; it also has a more northern distribution. Motyka also included U. capensis (South Africa), U. amblyoclada (South America) and U. glomerata (Australasia) in the same subsection. Of these species U. amblyoclada is possibly the sorediate counterpart of U. hieronymii. Several taxa, for example Usnea igniaria (UPS - holotype, not seen) and U. nidulifera (UPS! - holotype) which are usually corticolous and only rarely saxicolous, also occasionally occur in similar, mainly upland, habitats in southern South America. These two species usually lack pigmentation and superficially resemble, but replace, U. inermis; both sometimes contain psoromic acid. U. nidulifera may sometimes be extensively pigmented (cf. U. inermis), but when fertile, has a pale disc (for example, Argentina, Rio Negro, Cerro Otto, 1980, herb. Kalb!). 1. Usnea amblyoclada (Mull. Arg.)Zahlbr. Fig. 36 Cat. Lich. Univ. 6: 534 (1930). - Usnea barbata var. amblyoclada Mull. Arg. in Flora, Jena 72: 509 (1889). Type: Serra Ventana, in Argentinia merid., Lorentz (G! - holotype; BM! - isotype). [TLC: norstictic acid, salazinic acid, usnic acid.] Note: The combination 'Usnea amblyoclada (Mull. Arg.) Motyka' has not formally been made (Motyka, 1937) and would be superfluous. Description: Thallus saxicolous, rarely lignicolous, 2-3 cm, arising from a delimited, pigmented, holdfast, erect, richly branched above with clustered laterals and subcornute or slightly deflexed apices. Fibrils numerous, slightly articulate, often eroding. Branches terete, grey-green, rarely black-pigmented at the apices. Cortex thin, cartilagineous. Surface matt, papillate, rarely slightly inflated, then fracturing; pseudocyphellae numerous. Medulla compact or sublax, axis thick, occupying 0-5 of the branch diameter. Isidia numerous on ultimate branches and fibrils, arising from pseudocyphellae, apices black-pigmented, 116 F. J. WALKER often eroding. Pseudoisidia and soredia absent. Apothecia and pycnidia not known. TLC: (1) norstictic acid, salazinic acid, galbinic acid, usnic acid; (2) fumaprotocetraric acid, protocetraric acid, salazinic acid, usnic acid; (3) no medullary substances, usnic acid (rare - lignicolous). Distribution: Usnea amblyoclada is apparently confined to the northern part of the South America, where it is either rare or under-collected; it is recorded from Brazil, Argentina, Uruguay, and Peru (Motyka, 1937; Osorio, 1980; Swinscow & Krog, 1976). Previous reports of U. pulvinata from South America are assumed to refer to this taxon. Recent collections from Bolivia, Ecuador, and Peru (BM!) contain material referable to this species. Chemistry: The primary medullary substances are norstictic acid usually with salazinic acid. In addition to type material only six collections have been examined; in these galbinic acid was found as an accessory substance in two samples. A further specimen contained salazinic, protocetraric and fumarprotocetraric acids. Lignicolous material, from Bolivia, lacked medullary substances, although norstictic acid occurred in saxicolous thalli. It is likely that other chemical races may occur and in this diversity the species would resemble U. pulvinata (Swinscow & Krog, 1976). Fig. 36 Usnea amblyoclada. Holotype of Usnea barbata var. amblyoclada Mull. Arg. (G). Top. Whole thallus xl. Bottom. Detail of isidia and pseudocyphellae xlO. USNEA SUBGENUS NEUROPOGON 117 Discussion: The taxonomic position of this taxon remains uncertain. Swinscow & Krog (1976) initially accepted Usnea amblyoclada to be conspecific with Usnea pulvinata, although subsequently they (Swins- cow & Krog, 1979) expressed some doubt about their original interpretation. I have examined the holotypes of both taxa (in G and L respectively) and, pending a more detailed study of South American and African material, prefer to regard the two taxa as distinct species, possibly in different aggregates. U. pulvinata belongs to the U. bornmuelleri aggregate which is characterised by the presence of blackened, true isidia, a cartilagineous cortex, and usually saxicolous habitat. U. nigropapillosa, a saxicolous endemic from Tristan da Cunha (J0rgensen, 1977), belongs to the same aggregate, and is characterised by the presence of fibrils, blackened papillae and branch apices, but lacks true isidia (O! - holotype). In contrast to U. pulvinata, U. amblyoclada appears to have closer affinities with U. densirostra, a saxicolous, fertile South American species. Usnea amblyoclada is included here since it could occasionally be confused with U. durietzii. However, the former is grey-green, rather than yellow-green, has true isidia, not pseudoisidia, is more copiously fibrillate, and lacks the inflated main branches and a stalked holdfast, the features which characterise U. durietzii. Isidia in U. amblyoclada are produced extensively throughout on secondary branches and fibrils, and are not in delimited clusters as in U. acanthella. 2. Usnea inermis Motyka Fig. 37 Lich. Gen. Usn. Stud. Monogr. 1: 109 (1936). Type: Australia, Victoria, ad flumen Murray, in silvis, Wawra (W! - holotype). [TLC: psoromic acid, 2'-O-demethylpsoromic acid, usnic acid.] Description: Thallus corticolous, rarely saxicolous, l-2(-5) cm, arising from a delimited, usually unpigmented, holdfast, erect to subpendulous, dichotomous to irregular, repeatedly branched above with numerous, subarticulate laterals and reflexed, subcornute apices. Short fibril-like branches sometimes present. Branches terete or irregular, green to yellow-green, sparsely black-pigmented at the apices. Cortex thin. Surface waxy, smooth to faveolate, rarely subpapillate, unpigmented annulations frequent. Medulla lax, rarely sublax, axis thin, occupying less than 0-5 of the branch diameter. Isidia frequent throughout thallus, dispersed, arising from small pseudocyphellae, pigmented apices, becom- ing extensive then confluent, often eroding. Pseudoisidia and soredia absent. Apothecia rare. TLC: (1) squamatic acid, usnic acid; (2) psoromic acid, 2'-O-demethylpsoromic acid, usnic acid; (3) no medullary substances, usnic acid. Distribution: Usnea.inermis is apparently confined to Australasia. In New Zealand it is characteristically a corticolous or lignicolous species, chiefly occurring at low altitudes and rarely above 600 m, although it is often as important element of the lichen flora of subalpine scrub on twigs of, for example, Discaria and Leptospermum. Similarly, in Tasmania it is primarily a corticolous or lignicolous species of dry forests from the coast up to the subalpine zone. However, in Victoria and New South Wales, U. inermis frequently shares saxicolous habitats with U. torulosa at higher altitudes, between 1500 and 2200 m. Chemistry: Traces of an unidentified fatty acid were occasionally found in all three races. The psoromic acid race was found to be less frequent than the squamatic acid-containing race. Discussion: Usnea inermis should rarely be mistaken for members of the subgenus Neuropogon since it has many distinguishing features, besides different ecological parameters. For example, the thallus is much greener than the majority of Neuropogon species, and may easily be separated from Australasian taxa by the characteristic lax medulla and randomly scattered true isidia. Thallus blackening often appears to be more prevalent in thalli growing in very exposed situations and is often the result of tissue necrosis rather than pigmentation, although a tendency to limited pigmentation at the branch apices appears to be characteristic of the species. It may be distinguished from U. torulosa by the laxer medulla and the dispersed, rather than delimited, isidia. 3. Usnea torulosa (Mull. Arg.)Zahlbr. Fig. 38 Cat. Lich. Univ. 6: 594 (1930). - Usnea dasypogoides f . torulosa Mull. Arg. in Flora, Jena 66: 19 (1883). Type: Mt. Koscuisko, New South Wales, Findlay 554 (G! - holotype). [TLC: squamatic acid, usnic acid.] Usnea glomerata Motyka, Lich. Gen. Usn. Stud. Monogr. 2: 315 (1937). Type: Nova Hollandia, saxicola, 1876, Archer (O - holotype, not seen). Usnea aurescens Motyka, Lich. Gen. Usn. Stud. Monogr. 2: 561 (1938). Type: New Zealand, Cave Hill, Dunedin, (J. S. Thomson T245) M23 (herb. Motyka - holotype, not seen, BM!, CHR 343970! - isotypes). [TLC: barbatic acid, 4-O-demethylbarbatic acid, squamatic acid, usnic acid.] 118 F. J. WALKER Fig. 37 Usnea inermis. Top. Holotype of Usnea inermis Motyka (W) xO-5. Bottom. Detail of isidia. A- holotype x 10. B - New South Wales, Filson (MEL 1018181) x 10. USNEA SUBGENUS NEUROPOGON 119 _ HER81M UUU *> B Fig. 38 Usnea torulosa. A - holotype of Usnea dasypogoides f . torulosa Mull. Arg. (G) x 1 . B - Isotype of Usnea aurescens Motyka (CHR) xl.C- Detail of isidia. New Zealand, Otago, Old Man Range, James 1579(BM)xlO. 120 F. J- WALKER Neuropogon acromelanus var. inactivus Lamb in /. Linn. Soc. (Hot.) 52: 220 (1939). Type: Tasmania, Mt. Wellington, 1876, W. Campbell, ex herb. Stirton (BM! -holotype; BM! -isotype). [TLC: squamatic acid, usnic acid.] Description: Thallus saxicolous, (3)^t-6(-8) cm, arising from a proliferating or rarely delimited, pigmented, holdfast, erect to subdecumbent, dichotomous, richly branched above with numerous, short, clustered, entwined laterals with capillaceous apices. Fibrils rare. Branches terete, pale yellow, unpigmented or rarely blackened at the apices. Cortex thick, rigid. Surface waxy, rarely subfaveolate or annulate, papillae absent. Medulla compact, rarely sublax, axis thick, occupying 0-5-0-6 of the branch diameter. Isidia forming delimited clusters, arising from minute pseudocyphellae, pigmented apices, often eroding to from soralia-like areas. Pseudoisidia infrequent. Soredia absent. Apothecia rare. TLC: (1) squamatic acid, usnic acid; (2) psoromic acid, 2'-O-demethylpsoromic acid, squamatic acid, usnic acid; (3) barbatic acid, 4-O-demethylbarbatic acid, squamatic acid, usnic acid; (4) no medullary substances, usnic acid. Distribution: Usnea torulosa is widely known from saxicolous alpine or subalpine habitats in New Zealand and Australia, occurring with or sometimes replacing Neuropogon species. The species has a much wider ecological amplitude, and is found at lower altitudes; it is also more catholic in its choice of substrates. It seems more able to withstand competition from large foliose lichens and to tolerate some degree of nutrient enrichment, as often occurs in bird-perch communities with Parmelia signifera. U. torulosa also has some preference for habitats near standing water, often colonising depressions in boulders or rocks by lake sides. It also extends into areas with wetter climates, for example on the west coast of South Island, New Zealand, areas where Neuropogon is absent. Chemistry: The squamatic acid containing race (Race 1) was found to be the most common, with the barbatic acid race (Race 3) being less frequent and the psoromic acid race (Race 2) rare. Discussion: The taxonomic position of Usnea aurescens remains somewhat uncertain, although it is here included within the wide range of variation of U. torulosa. The isotype (Fig. 38) is a much less robust entity with fine, extended secondary branches with small eroded soralia-like areas that only produce a few true isidia. It is possible that some of the barbatic acid-containing specimens may eventually prove to belong to a distinct taxon. These are characterised by the presence of pronounced globular soralia-like structures which only rarely produce small pseudoisidia and appear to lack true isidia, although these entities are included within the variation of this species for the present. This probably represents another phase in the erosion-regeneration cycle of isidia production, where isidia regenerate to form pseudoisidia but ultimate- ly erode to produce large, usually excavate, soralia-like structures. Usnea torulosa is included in this appendix since it is frequently found associated with U. acromelana and might, in instances where the characteristic habit or the isidia are not well-developed, be mistaken for scantily pigmented forms of that species. In the field the two species are easily distinguished since U. torulosa is a much brighter yellow, in contrast to the yellow-green colour of U. acromelana. Pigmentation is scarce, being confined to the holdfast, isidia, and rarely the apices of secondary branches. Medullary chemistry is particularly useful in cases where identity is uncertain, since norstictic and salazinic acids, which are a feature of Australasian populations of U. acromelana, do not occur in U. torulosa. Usnea torulosa may be distinguished from U. inermis by the differences in habit, branch anatomy, and, frequently, habitat. Only rarely do the isidiate areas in U. torulosa become confluent towards the branch apices and then resemble the random, dispersed distribution found in U. inermis. Appendix II. Excluded taxa Letharia wandelensis Hue, Exped. Antarc. Franc.. 1903-1905, Botan., Lichens: 6 (1908). Type: Graham Land, Booth- Wandel I., Exped. Antarc. Franc,. 277, 299 (PC! - lectotype, selected here, CANL 16959! - isolectotype). From the original description and examination of the type material it is obviously that this species was based on a mixture of two taxa; namely a moribund, weathered specimen of Usnea antarctica overgrown by, amonst other lichens, a species of Caloplaca. Zahlbruckner (1926) subsequently created a new genus, Lethariopsis , to accommodate this 'species' based on the Caloplaca-type spores. Lamb (19486) discussed the identity of Letharia wandelensis and reported that, after careful study of the apothecia, it was impossible to determine the Caloplaca (or Teloschistes) species concerned. Since Hue (1908) mentioned the occurrence of Polycaulonia (Caloplaca) regalis and Polycaulonia coralligera (i.e. Xanthora candelaria - Lamb, 19486) as associated species, it is possible that the apothecia may belong to one of these. Lamb (19486) considered the former species to be the most likely, although he also found USNEA SUBGENUS NEUROPOGON 121 fragments of other lichens, including a species ofPhyscia, overgrowing the thallus. Dodge (1973) retained the genus Lethariopsis but suggested that Lamb's interpretation was referable to Usnea pseudofruticosa (U. antarctica) parasitized by Caloplaca clnericola. Examination of the type material (Fig. 39) shows that there is not sufficient material present to ascertain the identity of the Caloplaca species. Some clue to its identity may come from the parasite found in the apothecia described by Hariot (in Hue, 1908) as Endococcus wandelensis. This is a synonym of Polycoccum rugulosarium (Hawksworth in Pegler et al. , 1980) which is reported from the apothecia of Caloplaca regalis and Caloplaca rugulosa. There is apparently no material of the parasite still associated with the type specimen (D. Hawksworth, pers. comm.). HERB. MUS. PARIS. Fig. 39 Type of Letharia wandelensis Hue (PC) x 1 -5. Under the Code (Article 9.2) the original description could adequately be used to lectotypify either the Usnea or the Caloplaca species concerned, since the taxon can no longer be rejected under Article 70. If the Usnea part was selected as the lectotype of Letharia wandelensis the name would predate Usnea antarctica. Consequently Letharia wandelensis Hue is here lectotypified on the fertile part, as an unidentified species of Caloplaca. This follows Recommendation 7B of the Code in selecting a lectotype so as to preserve the current usage of the epithet 'antarctica'. Ramalina scopulorum var. E J. D. Hook., Flora Antarctica 2: 522 (1847):*Spec. orig.: dry granite rocks, Cape Horn, J. D. Hooker (BM!); Kerguelen's Land, Anderson (not traced). Material of this unnamed variety from lies Kerguelen was thought to be a species of Neuropogon (Crombie, 1879a; Dodge, 1948), possibly Usnea trachycarpa. However, examination of original material from Hermite Island, Cape Horn, /. D. Hooker 66 (BM!) and 76 (BM!), shows it to be Ramalina terebrata. It is possible that the Kerguelen specimen may have been mislabelled or misplaced, since according to Crombie (1879a) the genus Ramalina does not occur there. Usnea barbata var. (3 sulphurea Taylor & J. D. Hook., Flora Antarctica 1: 194 (1845). This taxon was described from the Auckland and Campbell Islands and, although the type has not been traced, is probably referable to Usnea xanthopoga. Zahlbruckner (1930: 602) gives this variety as a synonym of U. sulphurea (i.e. U. sphacelata). Usnea cornicularia Ach., Lich. Univ.: 619 (1810). Type: New Zealand, Forster (UPS - herb. Thunberg 26348!). Zahlbruckner (1930: 601/2) cited this taxon as a synonym of Usnea sulphurea (i.e. U. sphacelata) whilst Motyka (1936) excluded it from the subgenus. The type specimen is a corticolous Usnea-like species of Ramalina (Galloway, 1985). Usnea falklandica Motyka, Lich. Gen. Usn. Stud. Monogr. 2: 472 (1937). Type: Falkland Islands, unlocalised (W- not traced). From the brief description of this species it could either represent a species of Protousnea or Usnea s. lat., although not cited in the former by Krog (1976). Consequently this taxon may belong to the subgenus Usnea or even represent a scrambling form of U. aurantiaco-atra with scantily pigmented apices. 122 F. J. WALKER Usnea lutescens Stirton in Trans. Proc. N.Z. Inst. 30: 388 (1898). The unlocalised type, presumably from Australasia, has not been traced in BM or GLAM. If corticolous it might be referable to U. ciliifera (Motyka, 1937: 544), but if saxicolous it could be U. ciliata, and would then become the valid name for that species. Acknowledgements My sincerest thanks go to Mr P. W. James for his guidance, encouragement, and patience throughout the preparation of this work. I thank Mr J. K. Bartlett, Professor G. Degelius, Dr K. Kalb, and Dr R. D. Seppelt for loans of specimens and the directors and curators of the herbaria AAS, AD, BG, BLFU, BOL, C, CANL, CHR, E, FH, FI, G, GB, GLAM, H, HO, KASSEL, KIEL, L, LAM, LD, LINN, M, MB, MEL, NSW, NY, O, PC, S, STU, TRH, TROM, TUR, VER, W, WU, U, UPS, and US for the loan of material in their care. I thank staff as AAS, CHR, OTA, and WELT for all their help whilst working on their collections. I am also indebted to Dr R. I. L. Smith of the British Antarctic Survey for giving me access to the manuscript notes of Dr I. M. Lamb. My thanks go to Professor G. T. S. Baylis and Dr C. D. Meurk for assistance with field excursions in New Zealand. I also thank Dr D. J. Galloway, Mr E. S. Hansen, Mr P. W. James, Professor P. M. J0rgensen, Dr H. Krog, Dr D. O. 0vstedal, and Dr R. I. L. Smith, amongst others, for informative discussion on distribution, ecology, and biogeography based on their own observations. Recognition for advice on various nomenclatural matters goes primarily to Mr J. R. Laundon, and also to Dr O. Almborn, Mr A. O. Chater, Professor P. M. J0rgensen, and Dr N. K. Robson. I thank Dr B. J. Coppins and Professor D. L. Hawksworth for assistance with interpretation of fine structures. I am also indebted to Miss K. P. Kavanagh of the British Museum (Natural History) for help in preparation of Latin diagnoses and to Miss A. M. Burnet and Dr A. Melderis for their language expertise. The photographs were taken by Mr P. R. Hurst of the Photographic Unit, BM, and thanks also go to Miss L. G. M. Hosking for typing the manuscript. References Acharius, E. 1803. Methodus Lichenes 2. Stockholm. Ahmadjian, V. 1970. Adaptations of antarctic terrestrial plants. In M. W. Holdgate (Ed.), Antarctic ecology 2: 801-811. London. Ahti, T. 1966. Parmelia olivacea and the allied non-isidiate and non-sorediate corticolous lichens in the northern hemisphere. Acta hot. fenn. 70: 1-68. Allison, J. S. & Smith, R. I. L. 1973. The vegetation of Elephant Island, South Shetland Islands. Bull. Br. Antarct. Surv. 33/34: 185-212. Asahina, Y. 1967. Lichens collected in Tierra del Fuego by Yoshida, member of the scientific expedition to Patagonia, Hokkaido University 1966. Miscnea bryol. lichen. Nichinan 4 (6): 89-94. Auer, V. 1960. The Quaternary history of Fuego-Patagonia. Proc. R. Soc. B, 152: 507-516. Bailey, R. H. & James, P. W. 1979. Birds and the dispersal of lichen propagules. Lichenologist 11: 105. Beltman, H. A. 1978. Vegetative strukturen der Parmeliaceae und ihre Entwicklung. [Bibltheca lich. 11]. Vaduz. Bliss, L. C. 1979. Vascular plant vegetation of the southern circumpolar region in relation to antarctic, alpine and arctic vegetation. Can. J. Bot. 57: 2167-2178. 1981 . North American and Scandinavian tundras and polar deserts. In L. C. Bliss, O. W. Heal & J. J. Moore (Eds), Tundra ecosystems: a comparative analysis: 8-24. Cambridge. Bodo, B. & Molho, D. 1980. Structure des acides isomuronique et neuropogolique nouveaux acides aliphatiques du lichen Neuropogon trachy carpus. Phytochemistry 19: 1117-1120. Bowra, G. T., Holdgate, M. W. & Tilbrook, P. J. 1966. Biological investigations in Tottanfjella and central Heimefrontfjella. Bull. Br. Antarct. Surv. 9: 63-70. Bratt, G. C. & Cashin, J. A. 1976. Additions to the lichen flora of Tasmania 2. Pap. Proc. R. Soc. Tasm. 110: 139-148. Brodo, I. M. & Hawksworth, D. L. 1977. Alectoria and allied genera in North America. Opera Bot. 42: 1-164. Brown, R. 1823 ['1821']. Chloris melvilliana. London. 1824. A list of plants collected in Melville Island. In W. E. Parry, Journal of a voyage for the discovery of a north-west passage. Supplement to the Appendix. XI. Botany. London. Brundin, L. 1970. Antarctic land faunas and their history. In M. W. Holdgate (Ed.), Antarctic ecology 1: 41-53. London. USNEA SUBGENUS NEUROPOGON 123 Burton, R. W. 1968. Breeding biology of the brown skua, Catharacta skua lonnbergi (Mathews), at Signy Island, South Orkney Islands. Bull. Br. Antarct. Surv. 15: 9-28. Craw, R. C. 1978. Two biogeographical frameworks: implications for the biogeography of New Zealand. A review. Tuatara 23: 81-114. Crombie, J. M. 1876a. Lichenes Terrae Kergueleni: - an enumeration of the lichens collected in Kerguelen Land by the Rev. A. E. Eaton during the Venus-Transit Expedition in 1874-75. J. Linn. Soc. (Bot.) 15: 180-193. 18766. On the lichens collected by Professor R. O. Cunningham in the Falkland Islands, Fuegia, Patagonia, and the Island of Chiloe during the voyage of H. M.S. 'Nassau', 1867-9. /. Linn. Soc. (Bot.) 15: 222-234. - 1877. Revision of the Kerguelen lichens collected by Dr. Hooker. /. Bot. Lond. 15: 101-107. 1879a. Transit of Venus Expedition 1874/1975. Botany of Kerguelen Island. Enumeration of the plants hitherto collected in Kerguelen Island by the 'Antarctic', 'Challenger', and 'British Transit of Venus' Expeditions. IV. Lichens. Phil. Trans. R. Soc. B (extra vol.) 168: 46-52. 18796. Enumeration of Australian lichens in herb. Robert Brown (Brit. Mus.) with description of new species. J. Linn. Soc. (Bot.) 17: 390-401. Crook, K. A. W. 1981. The break up of the australian-antarctic segment of Gondwanaland. Monographiae biol. 41: 1-14. Culberson, C. F. 1972. Improved conditions and new data for the identification of lichen products by a standard thin-layer chromatographic method./. Chromat. 72: 113-125. , Culberson, W. L. & Johnson, A. 1981. A standardised TLC analysis of j3-orcinol depsidones. Bryologist 84: 16-29. & Johnson, A. 1976. A standardised two-dimensional thin-layer chromatographic method for lichen products. J. Chromat. 128: 253-259. Culberson, W. L. 1966. Jatta's subgeneric classification of the lichens. Bryologist 69: 214-231. & Culberson, C. F. 1980. Microconidial dimorphism in the lichen genus Parmelia. Mycologia 72: 127-135. Dahl, H. & Krog, H. 1972. Macrolichens of Denmark, Finland, Norway and Sweden. Oslo. Dalenius, P. & Wilson, O. 1958. On the soil fauna of the antarctic and of the subantarctic islands. The Oribatidae (Acari). Ark. zool. II, 11: 393-425. Darbishire, O. V. 1912. The lichens of the Swedish antarctic expedition. In N. O. G. Nordenskjold, Wissenschaftliche Ergebnisse der Schwedischen siidpolar-expedition 1901-1903. 4(11): 1-74. Stockholm. Darlington, P. J. 1965. Biography of the southern end of the world. Cambridge, Mass. Darwin, C. 1979. Letter to Prof. Henslow, written from Valparaiso July 24th 1834. In R. D. Keynes (Ed.), The Beagle Record. Selections from the original pictorial records and written accounts of the voyage of H.M.S. Beagle: 227. Cambridge. Deacon, G. E. R. 1960. The southern cold temperate zone. Proc. R. Soc. B, 152: 441-441. Dennis, R. W. G. 1960. Cryptogamic botany (Discussion) on the botany of the southern cold temperate zone. Proc. R. Soc. B, 152: 539-540. Dodge, C. W. 1948. Lichens and lichen parasites. Rep. B.A.N.Z. Antarctic Res. Exp. B, 7: 1-126. 1962. Expedition Antarctic Beige. Lichens. Bull. Jard. hot. EtatBrux. 32: 301-308. 1965a. Lichens. Monographiae biol. 15: 194-200. 19656. Lichenological notes on the flora of the antarctic continent and the subantarctic islands: VI. New taxa from the antarctic continent and adjacent islands. Trans. Am. microsc. Soc. 84: 507-529. 1966. Lichens from Kerguelen, collected by E. Aubert de la Rue. CNFRA 15: 1-8. 1970. Lichenological notes on the flora of the antarctic continent and the subantarctic islands IX-XI. Nova Hedwigia 19: 439-502. 1973. Lichen flora of the Antarctic continent and adjacent islands. Canaan, New Hampshire. & Baker, G. E. 1938. Botany of the second Byrd antarctic expedition. II. Lichens and lichen parasites. Ann. Mo. hot. Gdn 25: 515-718. & Rudolph, E. D. 1955. Lichenological notes on the flora of the antarctic continent and the subantarctic islands. I-IV. Ann. Mo. hot. Gdn 42: 131-149. Dumont d'Urville, J. S. C. 1826. Flore des Malouines. Mem. Soc. linn. Paris 4: 537-621. Du Rietz, G. E. 1924. Die Soredien und Isidien der Flechten. Svensk hot. Tidskr. 18: 371-396. 1926. Om slaktena Evernia Ach., Letharia (Th. Fr.) Zahlbr. emend DR. och Usnea Ach. subgenus Neuropogon (Nees et Plot.) Jatta. Svensk hot. Tidskr. 20: 89-93. 1929. The discovery of an arctic element in the lichen flora of New Zealand and its plantgeographical consequences. Rep. Austalas. Ass. Advmt. Sci. 19: 628-635. 1940. Problems of bipolar plant distribution. Acta phytogeogr. suec. 13: 215-282. 124 F. J. WALKER 1960. Remarks of the botany of the southern cold temperate zone. Proc. R. Soc. B, 152: 500-507. Eaton, A. E. 1879. Transit of Venus Expedition 1874/1875. The collections from Kerguelen Island. Introductory notes. Phil. Trans. R. Soc. B (extra vol.), 168: 1-8. Filson, R. B. 1966. The lichens and mosses of Mac. Robertson Land. A.N.A.R.E. Rep. B(2), 82: 1-169. - 1969. A review of the genera Teloschistes and Xanthoria in the lichen family Teloschistaceae in Australia. Muelleria2: 65-115. 1974. Studies in antarctic lichens II. Lichens from the Windmill Islands, Wilkes Land. Muelleria 3: 9-36. 1975. Studies in antarctic lichens V. Lichenes Antarctic! Exsiccati, Fascicle I, with additional notes on the taxonomy of each species. Muelleria 3: 146-158. 1976. Australian lichenology: a brief history. Muelleria 3: 183-190. 1981. Studies on Macquarie Island lichens 1: General. Muelleria 4: 305-316. 1982. Lichens of continental Antarctica. J. Hattoribot. Lab. 53: 357-360. Fleming, C. A. 1976. New Zealand as a minor source of terrestrial plants and animals in the Pacific. Tuatara 22: 30-37. Follmann, G. 1965a. Eine epipetrische Flechtengesellschaft der zentralchileschen Hochkordillere mil Kennzeichnendem Neuropogon acromelanus (Stirt.) Lamb. Ber. dt. hot. Ges. 78: 247-260. 1965b. Una associacon nitrofila de liquenes epipetricos de la Antarctica occidental con Ramalina terebrata Tayl. et Hook, como especie caracterizante. Publnes Inst. antart. chil. 4: 1-18. 1966. Eine neue Ramalina- Art aus der Ceruchis-Gruppe. Willdenowia 4: 227-233. 1967. Vegetationsanalytische Untersuchungen an Flechtengesellschaften zwischen Atacamawiiste und Grahamland. Ber. dt. bot.Ges. 80: 199-205. Galloway, D. J. 1968. A new variety of Usnea (Neuropogon) ciliata (Nyl.) DR. , in southern New Zealand. N.Z.JIBot. 6:470-472. 1979. Biogeographical elements in the New Zealand lichen flora. In D. Bramwell (Ed.), Plants and islands: 201-224. London. - 1983. New taxa in the New Zealand lichen flora. N.Z. Jl Bot. 6: 191-200. 1985. Flora of New Zealand: Lichens. Wellington. Gay, C. 1852. Historia fisica y politico de Chile. Botdnica 8 (1). Paris. Ghogomu, R. T. & Bodo, B. 1982. Structural elucidation of 13-acetoxylichesterinic and 13- acetoxyprotolichesterinic acids, two aliphatic lichen metabolites from Neuropogon trachy carpus. Phy- tochemistry 21: 2355-2358. Gimingham, C. H. & Smith, R. I. L. 1970. Bryophyte communities in the maritime antarctic. In M. W. Holdgate (Ed.), Antarctic ecology 2: 801-811. London. Godley, E. J. 1960. The botany of southern Chile in relation to New Zealand and the subantarctic. Proc. R. Soc. B, 152: 457-475. Golubkova, N. S. & Schapiro, I. A. 1970. De chemotaxonomia specierum antarcticum nonnullarum generis Neuropogon Nees et Fw. Nov. Sist. Nizsh. Rast. 7: 277-282. Grassi, M. M. 1952 ['1950']. Contribution al catalogo de liquenes Argentines. I. Lilloa 24: 5-294. Greene, S. W. 1964a. Plants of the land. In R. E. Priestley, R. J. Adie & G. de Q. Robin (Eds), Antarctic research: 240-253. London. 1964ft. The vascular flora of South Georgia. Sclent. Rep. Br. Antarct. Surv. 45: 1-58. & Longton, R. E. 1970. The effects of climate on antarctic plants. In M. W. Holdgate (Ed), Antarctic ecology 2: 786-800. London. Hale, M. E. 1965. A monograph of Parmeliasubgenus A mphigymnia. Contr. U.S. natn. Herb. 36: 193-358. Hansen, E. S. 1982. Lichens from central east Greenland. Meddr Gr0nland, Biosci. 9: 1-33. Hansen, K. 1962. Macrolichens from central west Greenland collected on the botanical expedition in 1958. Meddr Gr0nland 163 (6): 1-64. 1971. Lichens in south Greenland. Distribution and ecology. Meddr Gr0nland 178 (6): 1-84. Harrington, H. J. 1965. Geology and morphology of Antarctica. Monographiae biol. 15: 1-71. Hawksworth, D. L. 1972. Regional studies in Alectoria (lichens) II. The British species. Lichenologist 5: 181-261. 1973. Ecological factors and species delimitation in the lichens. In V. H. Heywood (Ed.), Taxonomy and ecology: 31-69. London. -1981. The lichenicolous coelomycetes. Bull. Br. Mus. nat. Hist. (Bot.) 9: 1-98. & Moore, D. M. 1969. Some lichens from Tierra del Fuego with notes on their chemical constituents. Bryologist 72: 247-251. Henssen, A. & Jahns, H. M. 1973 ['1974']. Lichenes. Stuttgart. & James, P. W. 1982. The lichen genus Steinera. Bull. Br. Mus. nat. Hist. (Bot.) 10: 227-256. USNEA SUBGENUS NEUROPOGON 125 Holdgate, M. W. 1960. The Royal Society expedition to southern Chile. Proc. R. Soc. B, 152: 434-441. - 1970. Introduction. Part XII - Vegetation. In M. W. Holdgate (Ed.), Antarctic ecology 2: 729-732. London. -, Tilbrook, P. J. & Vaughan, R. W. 1968. The biology of Bouvet0ya. Bull. Br. Antarct. Surv. 15: 1-7. Honegger, R. 1978. The ascus apex in lichenized fungi I. The Lecanora-, Peltigera- and Teloschistes- types. Lichenologist 10: 47-67. Hooker, J. D. 1879a. Transit of Venus expedition 1874/1875. Observations on the botany of Kerguelen Island. Phil. Trans. R. Soc. 168: 9-16. 18796. Enumeration of the plants hitherto collected in Kerguelen Island by the 'Antarctic', 'Challenger', and 'British Transit of Venus' Expeditions. I. Flowering plants, ferns, Lycopodiaceae and Characeae. Phil. Trans. R. Soc. 168: 17-23. 1844-47. The botany of the antarctic voyage . . . Flora antarctica: 1 & 2. London. & Taylor, T. 1844. Lichenes antarctici . . . /. Bot. Lond. 3: 634-659. Howe, R. H. 1915. The Usneas of the world, 1752-1914 with citations, type localities, original descriptions, and keys. Part II - South America. Bryologist 18: 38-43; 52-62. Hue, M. A. 1908. Expedition antarctique Franc,aise (1903-1905). Botanique: Lichens. Paris. Huneck, S., Schreiber, K., Hofle, G. & Snatzke, G. 1979. Neodihydromurol- und Murolsaure, zwei neue y-lactoncarbonsauren aus Lecanora muralis. J. Hattori hot. Lab. 45: 1-23. Huntley, B. J. 1971. Vegetation. In E. M. van Zinderen Bakker, J. M. Winterbottom & R. A. Dyer (Eds), Marion and Prince Edward Islands: 98-160. Cape Town. Huovinen, K. & Ahti, T. 1982. Biosequential patterns for the formation of depsides, depsidones and dibenzofurans in the genus Cladonia (lichen-forming ascomycetes). Annls hot. fenn. 19: 225-234. Jacquin, N. J. 1781. Miscellanea austriaca. ... 2. Vindobonae [Wien]. Jahns, H. M. 1980. Fame -Moose, Flecten, Mittel- Nord- und Westeuropas. Munchen. James, P. W. 1960. Cryptogamic botany (Discussion) on the botany of the southern cold temperate zone. Proc. R. Soc. B, 152: 540. Jatta, A. 1900. Sylloge Lichenum Italicorum. Trani. 1909. Flora italica cryptogama. 3 (3). Rocca S. Casciano. J0rgensen, P. M. 1977. Foliose and fruticose lichens from Tristan da Cunha. Skr. norske Vidensk-Akad. mat.-nat. Kl. 36: 1^0. 1983. Distribution patterns of lichens in the Pacific region. Aust. J. Bot. Suppl. Ser. 10: 43-66. Karnefelt, 1. 1979. The Brown fruticose species of Cetraria. Opera Bot. 46: 1-150. Kashiwadani, H. 1970. Lichens from the Prince Olav Coast, Antarctica. Scient. Rep. Jap. Antarct. res. Exped. E, 30: 1-21. Kershaw, K. A. 1983. The thermal operating-environment of a lichen. Lichenologist 15: 191-207. Kirkpatrick, J. B. 1980. Tasmanian high mountain vegetation I - A reconnaissance survey of the Eastern Arthur Range and Mount Picton. Pap. Proc. R. Soc. Tasm. 114: 1-20. Konig, J. G. 1772. Flora Islandica. In E. Olafsen & B. Povelsen, Reise igiennem Island 2: 3-20. Gor0e. Kristinsson, H. 1969. Chemical and morphological variation in the Cetraria islandica complex in Iceland. Bryologist 72: 344-357. Krog, H. 1976. Lethariella and Protousnea, two new lichen genera in Parmeliaceae. Norw. J. Bot. 23: 83-106. 1982. Evolutionary trends in the foliose and fruticose lichens of the Parmeliaceae. /. Hattori hot. Lab. 52:303-311. & James, P. W. 1977. The genus Ramalina in Fennoscandia and the British Isles. Norw. J. Bot. 24: 15-43. , 0sthagen, H. & Tonsberg, T. 1980. Lavflora Norske busk-og bladlav. Oslo. & Swinscow, T. D. V. 1981. Parmelia subgenus Amphigymnia (lichens) in East Africa. Bull. Br. Mus. nat.Hist. (Bot.) 9: 143-231. Lamb, I. M. 1939a. A review of the genus Neuropogon (Nees & Plot.) Nyl., with special reference to the antarctic species. J. Linn. Soc. (Bot.) 52: 199-237. - 19396 ['1940']. Lichens from East Greenland, collected by the Wagner expedition, 1935-36. Nytt Mag. Naturvid. 80: 263-286. 1948a. Further data on the genus Neuropogon. Lilloa 14: 139-168. 19486. New, rare or interesting lichens from the southern hemisphere. Lilloa 14: 201-251. 1955. New lichens from northern Patagonia, with notes on some related species. Farlowia 4: 423-471. 1959 ['1958']. La vegetacion liquenica de los parques nacionales Patagonicos. An. Parq. nac. B. Aires 7: 3-188. 126 F. J. WALKER - 1964. Antarctic lichens 1. The genera Usnea, Ramalina, Himantormia, Alectoria, Cornicularia. Sclent. Rep. Br. Antarct. Surv. 38: 1-34. 1970. Antarctic terrestrial plants and their ecology. In M. W. Holdgate (Ed.), Antarctic ecology 2: 733-751. Laundon, J. R. 1979. Deceased lichenologists: their abbreviations and herbaria. Lichenologist 11: 1-26. 1984. Proposals to conserve or reject. (730) Proposal to emend Cladonia Hill ex Browne, 1756, nom. cons., and delete Cladona Adanson, 1763, nom. rej. (Ascomycetes: Lecanorales). Taxon 33: 110-113. Lawrence, R. 1834. Notes on an excursion up the Western Mountains of Van Dieman's Land. J. Bot. Land. 1: 235-241. Lindsay, D. C. 1969. Further data on antarctic Usneaceae. Bull. Br. Antarct. Surv. 20: 33-40. - 19710. Vegetation of the South Shetland Islands. Bull. Br. Antarct. Surv. 25: 59-83. - 19716. Notes on antarctic lichens: IV. Lichens from Tottanfjella, Dronning Maud Land. Bull. Br. Antarct. Surv. 25: 99-100. 1972. Lichens from Vestfjella, Dronning Maud Land. Meddrnorsk Polarinst. 101: 1-21. 1973. Estimates of lichen growth rates in the maritime antarctic. Arct. alp. Res. 5: 341-346. - 1975 ['1974']. The macrolichens of South Georgia. Sclent. Rep. Br. Antarct. Surv. 89: 1-91. - 1977a. Lichens of cold deserts. In M. R. D. Seaward (Ed.), Lichen ecology: 183-209. London. 19776. The lichens of Marion and Prince Edward Islands Southern Indian Ocean. Nova Hedwigia 28: 667-689. 1978. The role of lichens in antarctic ecosystems. Bryologist 81: 268-276. & Brook, D. 1971. Notes on antarctic lichens: IV. Lichens from Tottanfjella, Dronning Maud Land. Bull. Br. Antarct. Surv. 25: 95-98. Lindsay, W. L. 1859. Memoir on the spermagones and pycnides of filamentous, fruticulose, and foliose lichens. Trans. R. Soc. Edinb. 22: 101-301. Longton, R. E. 1966. Botanical studies in the antarctic during 1963-64 and 1964-65 seasons. Bull. Br. Antarct. Surv. 10: 85-95. - 1967. Vegetation in the maritime antarctic. Phil. Trans. R. Soc. B, 252: 213-235. 1973. A classification of terrestrial vegetation near McMurdo Sound, continental Antarctica. Can. J. Bot. 51:2339-2346. - & Holdgate, M. W. 1979. The South Sandwich Islands: IV. Botany. Sclent. Rep. Br. Antarct. Surv. 94: 1-53. Lopez-Figueras, M. & Palacious-Prii, E. L. 1981. Organizacion morfologica del genero Usnea. Cuad. Centra. Miscrosc. electron. 2: 7-46. Lynge, B. 1932. Om utbredelelsen av endel arktiske laver. Svensk hot. Tidskr. 26: 401-430. - 1937. Lichens from west Greenland collected chiefly by Th. M. Fries. Meddr Gr0nland 118 (8): 1-225. - 1938. Lichens from the west and north coasts of Spitzbergen and the north-east land collected by numerous expeditions. I. The macrolichens. Skr. norske Vidensk-Akad. mat.-nat. Kl. 1938 (6): 1-136. 1941. On Neuropogon sulphureus (Konig) Elenk., a bipolar lichen. Skr. norske Vidensk-Akad. mat.-nat. Kl. 1940 (10): 1-35. Maas Geesteranus, R. A. 1947. Revision of the lichens of the Netherlands I. Parmeliaceae. 2. Glossary of technical terms. Blumea 6: 41-60. Mark, A. F. & Bliss, L. C. 1970. The high-alpine vegetation of Central Otago, New Zealand. N.Z. Jl Bot. 8: 381-451. Martin, W. & Child, J. 1972. Lichens of New Zealand. Wellington. Mattick, F. 1951. Wuchs- und Lebenformen, Bestands- und Gesellschaftsbildung der Flechten. Bot. Jb. 75: 378-424. Motyka, J. 1936-38. Lichenum generis Usnea studium monographicum, pars systematica. 1 & 2. Leopoli. - 1947. Lichenum generis Usnea studium monographicum, pars generalis. Annls Univ. Mariae Curie-Sktodowska C, 1: 277^76. Miiller, J. 1888. Lichenologische Beitrage XXX. Flora, Jena 71: 528-552. - 1895. Lichenes exotici. IV. Hedwigia 34: 139-145. - 1896. Lichenes Colensoani . . . /. Linn. Soc. (Bot.) 32: 197-208. Murray, J. 1963. Lichens from Cape Hallet area, Antarctica. Trans. R. Soc. N.Z. (Bot.) 2: 59-72. Nees von Esenbeck, C. G. & Flotow, V. 1835. Einige neue Flechtenarten. Linnaea 9: 495-502. Nylander, W. 1855. Essai d'une nouvelle classification des lichens (second memoire). Mem. Soc. Sci. nat. Cherbourg 3: 161-202. 1860. Synopsis methodica lichenum 1. Paris. 1866. Lichenes Novae Zelandiae, quos ibi legit anno 1861 Dr Lauder Lindsay. /. Linn. Soc. (Bot.) 9: 244-259. USNEA SUBGENUS NEUROPOGON 127 Osorio, H. S. 1980. Contribution to the lichen flora of Uruguay XIII. Lichens from Sierra Mahoma, San Jose department. Phytologia 45: 215-220. 0vstedal, D. O. 1983. Some lichens from Vestfjella, Dronning Mauds Land, Antarctis. Cryptogamie Bryol. Lichenol. 4: 217-226. Pegler, D. N., Spooner, B. M. & Smith, R. I. L. 1980. Higher fungi of Antarctica, the subantarctic zone and the Falkland Islands. Kew Bull. 35: 499-562. Persoon, C. H. 1827-28. Fungi. In M. C. Gaudichaud, Voyage autour du Monde . . . Bot. : 165-215. Paris. Poelt, J. 1970. Das Konzept der Artenpaare bei den Flechten. Vortr. GesGeb. Bot. 4: 187-198. Rasanen, V. 1932. Zur Kenntnis der Flechtenflora Feuerlands sowie der Prov. de Magallanes, Prov. de Chiloe und Prov. de Nuble in Chile. Suomal. eldin-ja kasvit. Seur. van. kasvitJulk. 2 (1): 1-65. 1939. Contribution a la flora liquenologica sudamericana. Liquenes recogidos por el Dr A. Donat en el Parque Nacional de Nahuel Huapi y por Adrian Ruiz Leal, en Mendoza. An. Soc. dent, argent. 128: 133-147. Raven, P. H. 1973. Evolution of subalpine and alpine plant groups in New Zealand. N.Z. Jl Bot. 11: 177-200. & Axelrod, D. I. 1974. Angiosperm biogeography and past continental movements. Ann. Mo. hot. Gdn. 61: 539-673. Redon, J. 1969. Nueva asociacion de liquenes muscicolas de la Antarctica occidental, con Sphaerophorus tenerLam., como especie caracterizante. Bolnlnst. antarct. chil. 4: 5-11. 1972. Liquenes del Parque Nacional 'Vicente Perez Resales', Provincia Llanquihue, Chile. An. Mus. Hist. nat. Valparaiso 5: 117-126. 1973. Beobachtungen zur Geographic und Okologie der chilenschen Flechtenflora. /. Hattori hot. Lab. 37: 153-167. 1974. Observaciones sistematicas y ecologicas en liquenes del Parque Nacional 'Vicente Perez Resales. Mn. Mus. Hist. nat. Valparaiso!: 169-225. Retzuis, A. J. 1769. Anmarkningar vid skanes ort-historie. K. svenska VetenskAkad. Handl. 30: 243-255. 1779. Florae Scandinaviae Prodromus. Stockholm. Rienke, J. 1895. Abhandlung iiber Flechten IV. Skizzen zu einer vergleichended Morphologic des Flechtenthallus. Jb. wiss. Bot. 28: 358-486. Rogers, R. W. 1981. The genera of Australian lichens. St. Lucia, Queensland. - & Stevens, G. N. 1981. Lichens. Monographiae biol. 41: 593-603. Rudolph, E. D. 1963. Vegetation of Hallett Station area, Victoria Land, Antarctica. Ecology, Brooklyn 44: 585-586. 1967. Lichen ecology and microclimate studies at Cape Hallett, Antarctica. In S. W. Tromp & W. H. Weihe (Eds), Biometerology 2: 900-910. Oxford. Santesson, R. 1942. Lichens from the Nahuel Huapi National Park in the Argentine. Ark. Bot. 30 (6): 1-12. Sayre, G. 1975. Cryptogamae Exsiccatae. An annotated bibliography of exsiccatae of algae, lichenes, hepaticae, and musci. V. Unpublished exsiccatae. I. Collectors. Mem. N. Y. hot. Gdn 19: 277-423. Schofield, E. 1972. Preserving the scientific value of cold desert ecosystems: past and present practices and a rationale for the future. In B. C. Parker (Ed.), Proceedings of the colloquium on conservation problems in Antarctica: 193-227. Blacksburg. Seppelt, R. D. & Ashton, D. H. 1978. Studies on the ecology of the vegetation at Mawson station, Antarctica. Aust. J. Ecol. 3: 373-388. Shields, O. 1976. A summary of the oldest ages for the world's islands. Pap. Proc. R. Soc. Tasm. 110: 35-61. Shipton, E. 1959. Explorations in Patagonia. Geogrl. J. 125: 312-325. Siple, P. A. 1938. The second Byrd antarctic expedition. Botany, 1. Ecology and geographical distribution. Ann. Mo. hot. Gdn 25: 467-514. Skottsberg, C. 1960. Remarks on the plant geography of the southern cold-temperate zone. Proc. R. Soc. B, 152: 447^57. Smith, R. I. L. 1973 ['1972']. Vegetation of the South Orkney Islands with particular reference to Signy Island. Scient. Rep. Br. Antarct. Surv. 68: 1-124. 1984. Colonization and recovery by cryptogams following recent volcanic activity on Deception Island, South Shetland Islands. Bull. Br. Antarct. Surv. 62: 25-51. & Corner, R. W. M. 1973. Vegetation of the Arthur Harbour- Argentine Islands region of the Antarctic peninsula. Bull. Br. Antarct. Surv. 33/34: 89-122. Stafleu, F. A. & Cowan, R. S. 1976. Taxonomic literature. A selective guide to botanical publications and collections with dates, commentaries and types 1. 2nd ed. [Regnum veg. 94]. Utrecht. Standring, K. 1983. Falklands wildlife at the cross-roads. Nat. world 7: 26-28. 128 F. J. WALKER Steiner, J. 1901. Uber die Function und den systematischen Wert der Pycnoconidien der Flechten. In 'Festschrift zur feier des zweihundert-jahrigen des K.K. Staats-gymnasiums im VIII. Bezirke': 119-154. Wien. Stirton, J. 1881. On the genus Usnea, and another (Eumitrid) allied to it. Scott. Nat. 6: 99-109. Swinscow, T. D. V. & Krog, H. 1974. Usnea subgenus Eumitria in East Africa. Norw. J. Bot. 21: 165-185. 1975. The Usnea undulata aggregate in East Africa. Lichenologistl: 121-138. 1976. The Usnea bornmuelleri aggregate in East Africa. Norw. J. Bot. 23: 23-31. - 1978. Pendulous species of Usnea in East Africa. Norw. J. Bot. 25: 221-241. - 1979. The fruticose species of Usnea subgenus Usnea in East Africa. Lichenologist 11: 207-252. Taylor, B. W. 1955. The flora, vegetation and soils of Macquarie Island. A.N.A.R.E. Rep. B, 2 (2): 1-192. Tehler, A. 1982. The species pair concept in lichenology. Taxon 31: 708-714. 1983. The genera Dirina and Roccellina (Roccellaceae). Opera Bot. 70: 1-86. Thomson, J. W. 1972. Distribution patterns of American arctic lichens. Can. J. Bot. 50: 1135-1156. Torrey, J. 1823. Art. V. Description of a new species of Usnea, from New South Shetland. Am. J. Sci. 6: 104-106. Troll, C. 1960. The relationship between the climates, ecology and plant geography of the southern cold temperate zone and of the tropical high mountains. Proc. R. Soc. B, 152: 529-532. Vainio, E. A. 1909. Lichenes Insularum Philippinarum, 1. Philipp. J. Sci. C, 4: 651-662. - 1913. Lichenes Insularum Philippinarum, 2. Philipp. J. Sci. C, 8: 99-137. 1915. Additamenta ad lichenographicam antillarum illustrandam. Suomal. Tiedeakat. Toim. A, 6 (7): 1-226. 1923. Lichenes Insularum Philippinarum, 4. Suomal. Tiedeakat. Toim. A, 19 (15): 1-84. Van Zinderen Bakker, E. M. 1970. Quaternary climates and antarctic biogeography. In M. W. Holdgate (Ed.), Antarctic ecology 1: 31-40. London. 1971. Introduction. In E. M. van Zinderen Bakker, J. M. Winterbottom & R. A. Dyer (Eds), Marion and Prince Edward Islands: 1-15. Cape Town. Verwoerd, W. J. 1971. Geology. In E. M. van Zinderen Bakker, J. M. Winterbottom & R. A. Dyer (Eds), Marion and Prince Edward Islands: 40-62. Cape Town. Vobis, G. 1980. Bau und Entwicklung der Flechten-Pycnidien und ihrer Conidien. [Bibltheca lich. 14]. Lehre. & Hawksworth, D. L. 1981. Conidial lichen-forming fungi. In G. T. Cole & B. Kendrick (Eds), Biology of conidial fungi 1: 245-273. London. Wace, N. M. 1960. The botany of the southern oceanic islands. Proc. R. Soc. B, 152: 475-490. Walker, F. J. & James, P. W. 1980. A revised guide to microchemical techniques for the identification of lichen products. Bull. Brit. Lichen Soc. 46 (suppl.): 13-29. Wardle, P. 1978. Origin of the New Zealand mountain flora with special reference to trans-Tasman relationships. N.Z. Jl Bot. 16: 535-550. Weber, H. 1969. Zur naturlichen Vegetationsgliederung von Siidamerica. Monographiae biol. 19: 475-518. Weber, W. A. 1973. Rediscovery ofNeuropogon lambii. Bryologist76: 305. & Wetmore, C. W. 1972. Catalogue of the lichens of Australia exclusive of Tasmania. Beih. nov. Hedwigia 41: 1-137. Wetmore, C. M. 1963. Catalogue of the lichens of Tasmania. Revue bryol. lichen. 32: 223-264. Wilson, F. R. M. 1888. Descriptions of two new lichens and a list of additional lichens new to Victoria. Victorian Nat. 5: 29-31. - 1890. Lichens from the Victorian alps. Victorian Nat. 6: 178-180. - 1893. Tasmanian lichens 1. Pap. Proc. R. Soc. Tasm. 1892: 133-178. Zahlbruckner, A. 1903. Neue Flechten. Annls mycol. 1: 354-361. 1917. Botanische Ergebnisse der Schwedischen Expedition nach Patagonien und dem Feuerland 1907-1909. K. svenska VetenskAkad. Handl. 57 (6): 1-62. 1926. Lichens B. Spezieller Teil. In H. G. A. Engler & K. A. E. Prantl (Eds), Die naturlichen Pflanzenfamilien. 8: 61-270. 2nd ed. Leipzig. 1930. Catalogus lichenum universalis 6: 530-606. Leipzig. USNEA SUBGENUS NEUROPOGON Index 129 Accepted names are in roman and synonyms in italic; new names and principal references are in bold. An asterisk (*) denotes a figure or table. acanthella 2, 7, 10*,11, 12, 13*, 15, 19*, 26, 31, 34, 38, 42, 43, 45, 47*, 48, 50, 79, 98, 117 Acharii 63 acromelana 5, 6, 7, 10, 12, 13*, 18*, 21, 22, 23, 24, 27, 30, 31, 35, 36, 37, 38, 40, 46, 48, 49*, 50, 51*, 52, 53*, 54, 55, 61,75, 76, 77, 84, 85, 90, 92, 96, 98, 100, 103, 106, 120 activa 63, 70 adarense (Rhizocarpon) 31 adusta 114, 115 aeneofusca (Menegazzia) 30 Alectoria 7, 11, 24, 25, 29, 41, 43 alectoriae (Lecidea) 110 amaliae 115 amblyoclada (Miill.Arg.) Motyka 115 amblyoclada (Miill.Arg.) Zahlbr. 11,45,79, 115, 116*, 117 angulosa 111, 113 anisomera (Buellia) 28 antarctica 5, 6, 7, 8*, 10*, 11, 13*, 14, 15, 16*, 21*, 22, 23, 25, 26, 27,28,29,30,31,33,34,35,36, 38, 40, 41, 43, 46, 50, 54, 55, 56, 57*, 58, 59, 60*, 61, 62, 66, 67, 69, 70, 84, 85, 92, 93, 95, 98, 100, 102, 103, 108, 110, 120, 121 antarctica (Umbilicaria) 26 antennarius (Neuropogori) 4, 11, 43, 62, 67, 68*, 69, 70 arboricola 4 aspidophora (Lecanora) 27 aurantiaca 39*, 63, 64 aurantiaco-atra 2, 5, 6, 7, 8*, 9*, 11, 12,13*, 14, 15, 16*, 21, 22, 23, 24, 26, 27, 28, 29, 31, 32, 33, 34, 35, 36, 39*, 40, 43, 44, 45, 50, 56, 60, 62, 63, 64*, 65*, 66*, 67, 68*, 69, 70, 71, 72*, 73, 74, 76, 82, 86, 88, 89, 100, 103, 108, 110, 113, 114, 121 aurescens 117, 119*, 120 barbata 121 bellidiflora (Cladonia) 25 blanda (Lecanora) 30 bogotensis 31, 79, 115 bornmuelleri 117 Bryoria 32 Buellia 29 Caloplaca 120, 121 candelaria (Xanthoria) 120 capensis 24, 115 castanea (Menegazzia) 30 Cetraria 37 Chlorea 4 ciliata (Miill.Arg.) Vainio 74 ciliata (Nyl.) Du Rietz 2, 5, 6, 7, 8*9*, 12, 13*, 14, 18*, 21, 22, 25,26,27,30,32,35,40,45,50, 53, 60, 69, 70, 71, 74, 75*, 76*, 77,86,88,89,90,91,96,103, 104, 106, 107, 113, 115, 122 ciliifera 122 cinericola (Caloplaca) 121 condensata 79 contexta 35 coralligera (Polycaulonia) 120 coriacea (Siphula) 30 cornicularia 121 crassa 56, 59 crombiei55, 59, 110 crystalligenum (Rhizocarpon) 31 cylindrica (Umbilicaria) 25, 31 dactylina (Pertusaria) 30 dedpiens 38, 48, 49*, 52, 113 decussata (Umbilicaria) 31 densirostra 114, 115, 117 Densirostrae 115 direagens (Parmelia) 39 durietzii 1, 5, 6, 7, 10*, 11, 13*, 15, 20*, 21*, 27, 31,32, 34, 38, 41, 42, 43, 46, 48, 50, 54, 59, 60, 78*, 79, 80, 84, 85, 95, 96, 98, 103, 117 dusenii (Protousnea) 11 eciliata 63, 70 egentissima 63, 70 elatior (Neuropogon) 11, 103, 111, 113 elegans (Xanthoria) 27 ericetorum (Cetraria) 37 Eumitria 43 Euusnea 3, 41, 42* Evernia 32, 41 falklandica 121 fasdata 2, 28, 29, 39*, 62, 65, 66*, 67,68 fibrillifer (Neuropogon) 63, 70 flavicans (Cornicularia} 62, 71 florida 67 floriformis 55, 59 Foveatae 41, 42* frigida (Ochrolechia) 29 frigida (Usnea) 56, 92, 96 geographicum (Rhizocarpon) 30, 31 gilva (Pseudocyphellaria) 24 Glabratae41,42* glabratula 9 globosus (Sphaerophorus) 29 glomerata 115, 117 granulifera 55, 56, 60 hieronymii 114, 115 Himantormia 29 hyyppae 111, 113 igniaria 35, 115 inadivus (Neuropogon) 54, 120 inermis 21, 24, 35, 45, 48, 115, 117, 118*, 120 insularis 29, 33, 40, 55, 59, 110 irrubens (Lecidea) 30 islandica (Cetraria) 37 Jacquinii 63 kerguelensis (Orceolina) 33 kranckii 63, 67, 70 Laevigatae 41, 42*, 115 lambii (Neuropogon) 93, 95 laxissima 22, 38, 92, 93, 97* Lecanora 7, 29 Lecidea 29, 30, 31 Letharia41,121 Lethariella 4, 41,43 Lethariopsis 120, 121 lucens (Menegazzia) 30 lugubris (Himantormia) 29 lugubris (Hypogymnia) 30 lutescens 122 macularis (Buellia) 30 melaxantha 2, 4, 11, 21, 22, 24, 36, 39*, 62, 64, 65*, 67, 70, 71 Melaxanthae 41, 42* Menegazzia 32, 34, 35 mexicana (Xanthoparmelia) 30 miniuscula (Alectoria) 29 mitis (Cladonia) 37 mougeotina (Parmelia) 30 muralis (Lecanora) 14 naumanii 111, 113 Neuropogon 4, 41, 42*, 43, 44 neuropogonoides, 1, 5, 7, 12, 13*, 14, 20*, 21, 26, 27, 31,32, 40, 41, 42, 43, 44, 61,70, 80, 81*, 82, 114 nidulifera 35, 115 nigricans (Alectoria) 22, 30, 37, 75 nigropallida 63, 70 nigropapillosa 79, 117 nobilis 35 normalis 63 nothofagi (Menegazzia) 30 nylanderiana (Umbilicaria) 31 ochroleuca (Alectoria) 109 olivacea (Parmelia) 25 130 F. J. WALKER pallida Motyka (Usnea) 35 pallidus Retz. (Lichen) 2, 92, 93 pallidus Schreber (Lichen) 92, 93 palmicola (Coccocarpia) 30 Parmelia 30, 41 Parmeliaceae 4,1, 11 patagonica 1, 6, 7, 8*, 10*, 11, 13*, 14, 15, 20*, 21*, 24, 27, 31,32, 38, 40, 42, 43, 46, 50, 54, 58, 59, 60, 79, 82, 83*, 84, 85, 86, 95, 98,100,101,102,103,113,114 perforata (Parmelia) 12 perpusilla 2, 6, 7, 8*, 9*, 11, 13*, 14, 15, 19*, 21, 27, 31,34, 35, 38,40,44,45,58,67,71,76,84, 85, 86, 87*, 88*, 89, 97, 102, 108, 113 Pertusaria 29 petriseda (Parmelia) 30 Physcia 121 picata 92, 96 Placopsis 32, 34 poeppigii (Neuropogon) 4, 67 Polycaulonia 120 polytropa (Lecanora) 30 propagulifera 56, 59 Protousnea 4, 12, 31, 32, 36, 41, 42, 82, 121 prunastri (Evernia) 11 pseudocapillaris 1, 5, 6, 7, 10, 13*, 18*, 22, 26, 40, 46, 52, 53, 61, 89, 90, 91*, 92, 96, 106 Pseudocyphellaria 32 pseudofruticosa 56, 59, 121 pseudosorediosa (Parmelia) 30 pubescens (Pseudephebe) 29, 31 pulvinata 24, 116, 117 pustulata 38, 56, 58 Pycnocladae 42* Ramalina 42, 121 rangiferina (Cladonia) 29 regalis (Caloplaca) 27, 120, 121 regalis (Polycaulonia) 120 reticulata (Parmelia) 30 Rhizocarpon 29 roccellina 115 Roccellinae41,42*,43, 115 rohmederi2, 11,86, 87*, 88 roseola 6 rugulosa (Caloplaca) 121 rugulosarium (Polycoccum) 121 russa (Buellia) 28 sarmentosa (Alectoria) 43 scabridula 92, 96 scopulorum (Ramalina) 121 scrobiculata (Protousnea) 82 Setulosae 115 signifera (Parmelia) 30, 120 siliquosa (Ramalina) 39 soleirolii 4 sorediifera 55, 58, 60 spadicea 63, 67 sphacelata2, 5,6, 7, 8*, 10,12, 13*, 14, 15, 17*, 19*, 21, 22, 23, 25,26,27,28,31,34,35,37,38, 40, 42, 45, 46, 48, 50, 54, 56, 58, 59, 60, 79, 85, 92, 93, 94*, 95, 96, 97*, 98, 99, 100,101, 102, 103, 121 Sphaerophorus 29 spongiosa (Solorina) 35 Steinera 36 Stereocaulon 29, 31 straminea (Cetraria) 13 striata 93 strigosa 67 strigulosa39*,63,7Q subantarctica 1, 6, 7, 10, 13*, 15, 17*, 21, 23, 27, 31,34, 35, 37, 38, 40, 43, 46, 50, 54, 59, 60, 85, 95,96,98,99, 100, 101*, 102*, 103, 104, 114 subaurifera (Parmelia) 11 subcapillaris 2, 5, 6, 7, 12, 13*, 18*, 21, 22, 26, 30, 35, 40, 44, 45, 76, 77, 90, 91, 92, 96, 104, 105*, 106, 107, 108 subciliata Rasanen 111, 113, 114 subciliata Zahlbr. 63 subfloridana 9 subfoveolata 38, 56, 59 sublaevis Dodge 55 sublaevis Miill.Arg. Ill, 113 subpapillata 38, 56, 59 subpolaris 92, 96 subrudecta (Parmelia) 30 subspadicea 67 substrigulosa 111, 113, 114 sulcata (Parmelia) 12, 30 sulphurea Taylor & J.D. Hook. 121 sulphurea Th.Fr. 2, 22, 28, 42, 48, 60, 92, 95, 96, 100, 102, 121 Sulphureae41,42* sulphureus J. Konig (Lichen) 2, 92, 93 sulphureus Retz. (Lichen) 92, 93 superba (Pertusaria) 30 tasmanica (Xanthoparmelia) 30 taylorii 5, 6, 7, 8*, 11, 12, 13*, 16*, 21,23,29,30,32,36,38,40, 41, 43, 44, 59, 67, 69, 70, 71, 108, 109*, 110* Teloschistes 120 terebrata (Ramalina) 27, 121 tigrina (Ramalina) 42 torulosa 11, 12, 21, 22, 24, 30, 35, 39, 45, 46, 53, 54, 82, 115, 117, 119*, 120 trachycarpa 6, 7, 11, 12, 13*, 14, 17*, 21, 22, 23, 27, 29, 30, 31,33, 34,40,43,44,45,52,67,70,71, 76, 79, 82, 84, 85, 86, 89, 100, 102, 103, 108, 110, 111, 112*, 113, 114, 115, 121 Trachycarpae 41, 42* trachycarpoides 111 trichoidea 74 Umbilicaria 25, 29, 30 ushuaiensis (Neuropogon) 86, 88*, 89 Usnea 4, 41, 42, 43 wandelensis (Endococcus) 121 wandelensis (Letharia) 120, 121* Xanthoparmelia 30 xanthopoga 22, 35, 121 Xanthoria 25 British Museum (Natural History) Ferns of Jamaica A guide to the Pteridophytes G. R. Proctor This flora records and describes the 579 species and 30 varieties of ferns occurring in Jamaica. The succinct species descriptions include relevant synonymy and incorpo- rate distributional data both within and outside Jamaica. Special emphasis is given to the subtle distinctions between closely related species and all genera are illustrated. Keys to the genera and species facilitate a wider use of the flora in the West Indies and northern South America. The author, one time Senior Botanist in charge of the Herbarium of the Science Museum, Kingston, Jamaica, is an outstanding field botanist and his expertise is reflected in the practicality of the flora and especially in the habitat and ecological information. This volume represents an important addition to our knowledge of the flora of the West Indies. 1985, 650pp (approx), 135 line illustrations, 22 maps. Hardback. 565 00895 1 50.00 Titles to be published in Volume 13 The lichen genus Usnea subgenus Neuropogon. By F. Joy Walker Cytotaxonomic studies of the ferns of Trinidad (3 papers). By A. C. Jermy & T. G. Walker Some genera of the Biddulphiaceae (diatoms) with interlocking linking spines. By Robert Ross & Patricia A. Sims Photoset by Rowland Phototypesetting Ltd, Bury St Edmunds, Suffolk Printed in Great Britain by Henry Ling Ltd, Dorchester UM i Bulletin of the I British Museum (Natural History) Cytotaxonomic studies of the ferns of Trinidad A. C. Jermy & T. G. Walker Botany series Vol 13 No 2 30 May 1985 The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in four scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology, and an Historical series. Papers in the Bulletin are primarily the results of research carried out on the unique and ever-growing collections of the Museum, both by the scientific staff of the Museum and by specialists from elsewhere who make use of the Museum's resources. Many of the papers are works of reference that will remain indispensable for years to come. Parts are published at irregular intervals as they become ready, each is complete in itself, available separately, and individually priced. Volumes contain about 300 pages and several volumes may appear within a calendar year. Subscriptions may be placed for one or more of the series on either an Annual or Per Volume basis. Prices vary according to the contents of the individual parts. Orders and enquiries should be sent to: Sales, tish Museum (Natural History), Cromwell Road, London SW75BD, . Br. Mus. nat. Hist. (Bot.) Trustees of the British Museum (Natural History), 1985 The Botany series is edited in the Museum's Department of Botany Keeper of Botany: Mr J. F. M. Cannon Editor of Bulletin : Mr J . R . Laundon Assistant Editor: Dr A. J. Harrington Editor's Assistant : Miss M . J . Short ISBN 565 08005 9 ISSN 0068-2292 British Museum (Natural History) Cromwell Road London SW75BD Botany series Voll3No2ppl31-276 Issued 30 May 1985 Cytotaxonomic studies of the ferns of Trinidad Contents 1 . The climate , geology , and vegetation of Trinidad with particular reference to the ecology of ferns. By A. C. Jermy 2 . The cytology and taxonomic implications . By T. G . Walker BRITISH 3. Descriptions of new species and hybrids and a new combination. By A. C. Jermy & T.G.Walker JL " (SENgRA- Page 133 149 251 Cytotaxonomic studies of the ferns of Trinidad 1. The climate, geology, and vegetation of Trinidad with particular reference to the ecology of ferns Anthony Clive Jermy Department of Botany, British Museum (Natural History), Cromwell Road, London SW7 5BD Contents Synopsis 133 Introduction 133 Climate 134 Rainfall 134 Temperature 136 Humidity 136 Winds 136 Geology and soils 138 Solid geology 138 Soils 138 Vegetation 140 Seasonal (lowland) rain-forest formations 140 Dry evergreen forest 141 Montane formations 141 Swamp communities 144 Savanna 144 Man-made habitats 145 Acknowledgements 146 References 146 Synopsis The topography, climate, geology, and vegetation formations found on the island of Trinidad are briefly summarised and each discussed in relation to the ecology of some of the more dominant or otherwise interesting fern species. Introduction Trinidad is an island in the tropical belt situated between 102' and 1050' N. latitude and 6055' and 6150' W. longitude. The political state includes the smaller island, Tobago, 30-5 km (19 miles) NE. of Trinidad and 192 km (120 miles) SW. of Barbados. Trinidad lies close to the South American continent and Corozal Point in the north-west, together with a chain of islands, of which Chacachacare, Huevos, Monos, and Caspar Grande are the most important, is only 11 km (7 miles) from Venezuela. In general topography two mountain systems stretch across almost the entire island, the Northern and Southern Ranges, whilst a third, the Central Range, lies somewhat diagonally across its middle portion. The Northern Range is formed by up-folding of Upper Jurassic and Cretaceous rocks and, as such, forms the eastern extremity of the eastern branch of the Andean chain which stems from Colombia and passes through northern Venezuela. The highest points are Aripo, 925 m (3085 ft) and El Tucuche, 922 m (3072 ft). The southern uplands are relatively Bull. Br. Mus. nat. Hist. (Bot.) 13 (2): 133-147 Issued 30 May 1985 134 FERNS OF TRINIDAD low hills, the highest being Peak 404 (304 m; 1013 ft) in the Trinite Hills; these are formed from a southern anticlinal trend or series of folds separated by a complicated fault system. It is an area that has been, at least in its western part, intensively explored for oil and very much cut over in putting down wells and pipelines, most of which are now dry and defunct. Emission of natural gases in this area gives rise to a phenomenon known as 'mud volcanoes'. Another feature in this area, at La Brea, is the Pitch Lake, covering 46 hectares (115 acres) of natural deposit of asphalt. The Central Range is another broad anticlinal uplift of Cretaceous and Eocene rocks flanked by Miocene formations, which include the prominent limestone hills of Tamana, Biche, and Brigand Hill. Between these hills are broad synclinical depressions that in the north are bounded by a large east to west fault at the base of the Northern Range; the resulting submerged areas are now filled with Pliocene and Pleistocene deposits. In the west of this area is the extensive Caroni delta and surrounding mangrove and swamp. Any expedition or field project that sets out with the intention of collecting a comprehensive range of a group of plants in a given area will of necessity collect ecological data, which, although often condensed on an herbarium label may not otherwise be published. The four excursions made by the author and T. G. Walker between 1963 and 1973, during which extensive collections of ferns were made and formed the basis of research reported in paper 2 (Walker, 1985), are no exception, and these short notes will hopefully give a background to the environment and provide data around which the biology of the ferns of Trinidad may be discussed. Climatic recording in Trinidad and Tobago is not centrally organized and a variety of institutes and organisations take recordings for their own use. The data presented here are substantially taken from Berridge (1981). The major climatic controls of these islands are their latitude, their landmass size, and relative position in the ocean and to the Bermuda/ Azores anticyclone, the effect of air-mass migrations (mainly the NE. and SE. Trades), and more locally the effect of topography. The seasons display monsoonal characteristics, the dry season coinciding for the most part with the northern hemisphere winter and the wet with summer. Collecting for this project was done for the most part in that wet season July-October, although one trip was made by M. G. and T. G. Walker in April. It is interesting to note that whilst the ferns of open habitats, e.g. road banks, and epiphytes in the more exposed crowns of trees responded quickly to the increased summer rainfall in flushing and quickly sporing, those species of the forest floor took some while before the effect of the increased rainfall took effect. Even those species in the deeper, more or less constantly moist ravines appeared to exhibit seasonality in spore production. Climate Rainfall Trinidad's annual rainfall totals vary from over 3048 mm (120 in) in the NE. to approximately 1524 mm (60 in) in the NW. and SW. peninsulas (see Fig. 1) and are related to the topography and upwind location in the local precipitation regime. The difference in amounts between the wet and dry season is considerable, the wet season maps showing a distribution from 2032 mm (80 in) to 635 mm (25 in), and the dry season from 1016 mm (40 in) to 254 mm (10 in). Tobago exhibits the same patterns differing only in amounts. Table 1 gives the mean monthly rainfall and Table 2 the mean number of days in each month when less than 1 mm of rain fell. Diurnal variation is more marked than might be expected in the small landmass. There is a tendency for midday and early afternoon showers inland. On the east coast convergence of air masses from land and sea regularly give showers in the evening and night. The intensity of rain is greater in the wet season. Topography will influence local convection currents, and a few areas of the higher ground will get almost daily rain throughout the year. CLIMATE, GEOLOGY & VEGETATION Table 1 Mean monthly rainfall (in mm). Based on Piarco records after Berridge (1981). 135 Jan. 80-9 Feb. 50-6 Mar. 33-6 Apr. 50-9 May 118-7 June 262-3 July 254-1 Aug. 247-0 Sept. 180-4 Oct. 160-8 Nov. 206-6 Dec. 150-0 Table 2 Mean number of days in each month with rainfall of less than 1 millimetre. Based on Piarco records after Berridge (1981). J 20 F 19 M 25 A 23 M 19 J 10 J 11 A 12 S 15 O 15 N 13 D 16 6130' km 2250 / 11750 2000 2000 1030' 61 30' 6100 Fig. 1 Isohyetal map of mean annual rainfall recorded over 30 years (1939-1968). (Adapted from Berridge, 1981.) 136 FERNS OF TRINIDAD Table 3 Monthly mean (dry bulb) and means of maximum and minimum temperatures (in C). Based on Piarco records after Berridge (1981). J FMAMJ JASOND Monthly mean Mean minimum 24-5 20-4 24-7 20-4 25-3 20-9 26-2 21-9 26-6 22-9 26-1 23-1 26-0 22-7 26-1 22-6 26-2 22-6 26-1 22-5 25-5 22-0 24-9 21-2 Mean maximum 30-6 30-3 31-1 31-8 31-6 30-6 30-7 31-0 31-4 31-3 30-7 30-0 Temperature Such is the effect of the Caribbean Sea and the Atlantic Ocean on the temperatures of Trinidad that the seasonal variation is only about 2-3C in the mean (see Table 3). Diurnally the variations are greater - approximately 10-1 5C between night and day. The mountain ridges, in the Northern Ranges at least, are high enough to show some decrease in temperature with altitude which in turn affects the vegetation and the plants which grow in this situation. Humidity For most of the island humidities are high throughout the year, and the relative humidity (i.e. the percentage rates of actual water vapour in the atmosphere for a given temperature against its total saturated capacity) lies in the region of 80%, with extremes from almost 100% to 50%. Even the air close to roadside or river banks in the open may be quite humid due to evaporation from seepage water or vegetation and as such forms excellent conditions for prothalli and young sporophyte growth. Winds Wind is a highly variable parameter, influenced by local effects of topography, land heating, etc. Berridge (1981) stresses that extra local records might give a more accurate picture of the variation of wind direction, which is biologically of greater significance than the traditional impression of dominant westerlies in the dry season and strong easterlies during the wet. Berridge has frequently recorded surface winds from the west in both seasons, and also from the south, south-west, north, and north-west in the Gulf of Paria. Such winds could be effective in dispersing spores and establishing species from both the South American mainland and the Lesser Antilles. CLIMATE, GEOLOGY & VEGETATION 137 61 30 61 00 1030 61 30 61*00' Fig. 2 Geological map of Trinidad. (Modified from Kugler , 1959.) Jurassic/quartz-mica schists, quartzites, and phyllites Cretaceous andesite phyllitic shales grits with volcanic tuffs Eocene and Paleocene chalky marls Oligocene marls and clays , ,. glauconitic and calcareous sands \\/iir\c*f*r\f* and limestones Pliocene clays, silts, and sands Pleistocene silty terraces 138 FERNS OF TRINIDAD Geology and soils Solid geology Trinidad is geologically part of the South American plate which in Jurassic/Cretaceous times (c. 130 million years ago) was moving away from Africa upon the splitting up of Gondwanaland. Orogenic movements that uplifted the sandstone hills of Roraima, the eastern Venezuelan mountains, and the Trinidadian Northern Ranges in turn lowered the coast areas and Orinoco basin, which became filled with later deposits, mainly calcareous clays and silts and, to a lesser extent, sands. The geology consists of a series of metamorphic rocks (slatey phyllites with subordinate limestones, quartzite, and grits) of Upper Jurassic origin, with Cretaceous shales and siltstones (Fig. 2). The limestone valleys and sinkholes of the Guanapo Heights are particularly rich in ferns, and large cave systems are found at Gaspara, Aripo, and Oropuche. In the north-east near Sans Souci is the only intrusion of volcanic rocks seen in the island, an intrusion that partially metamorphosed the adjacent rocks. A broad anticlinal uplift was pushed up in late Miocene times with strong folding from Pointe-a-Pierre through Tabaquite to Mount Harris, Tamana, and Brigand Hill where often rubbly limestones are exposed. The southern shoulder of this uplift, around Naparima Hill exposes, amidst faults and folds, the Cretaceous core of hard siliceous siltstone. Thus, here, as in the Northern Ranges, juxtaposition of acid and basic rocks exists. The southern uplands, low altitude hills stretching from Palo Seco to Guayaquayare form the southern anticlinal trend. Its highest peak lies in the Trinite Hills and again contains many fractures and faults giving interbedded sands, silts, and clays as mentioned above, with occasional chalky marls and limestones. An interesting feature of this district is the occurrence of mud volcanoes where silty mud flows or periodically 'erupts' through conical vents due to pressures exerted by natural gas escaping along fault fractures. They are in no way connected with normal volcanic action, but are ecologically significant in that they form areas of fine silt ready to be colonised by seedlings and in damper areas by prothalli, in what would otherwise be a closed grassland or scrub community. The intervening lowland between these folds was filled with sands, gravels, and alluvium in Pleistocene times. Superimposed on this geology is an erosion pattern of drainage systems involving six major rivers: Caroni, Oropuche (northern), Guaracara, Nariva, Oropuche (southern), and Ortoire. Many smaller streams drain the northern flanks of the Northern Range and the southern edge of the southern hills. Within the Caroni and northern Oropuche systems, are highly productive agricultural areas interspersed with river terraces, one of which is ecologically important, the Aripo Savanna. The estuaries of these rivers, and especially the Nariva, contain swamps and tidal mangroves. Soils The soils of Trinidad have been discussed by Hardy (1981), and the reader is referred to that paper for details. Seventy-eight soil types are described for Trinidad by Chenery (1952) and other soil surveyors. Six major sub-divisions are used in mapping and a simplified map is given here (Fig. 3) from Hardy (1981). The Highland Upland soils all occur on sloping land and are all free draining sands, loams and loamy clays usually developed on fine-grained metamorphic sediments, limestone, and sub-basic igneous rock. They are all restricted to the Northern Range where the drainage is almost entirely by few, south-flowing rivers partly filled with alluvium. CLIMATE, GEOLOGY & VEGETATION 139 1030 km 10 20 61*30' 61*30' 61*00' 1030- Beach sand Deep hydromorphic soils Deep alluvial soils Terrace soils Intermediate upland soils High upland soils 6100 Fig. 3 Simplified soils map of Trinidad. (From Hardy, 1981 , with permission of Edward Arnold.) The Intermediate Upland soils are those seen in the hills of the centre and the rolling country in the south. Mainly they are clays with occasional sandy clay-loams and loamy fine sands. Terrace soils range from gravelly sandy clay-loam, sands such as at Aripo Savanna and fine sandy loams to clays. The deep alluvial soils are again clays predominantly, and sandy clay-loams. Whilst much of this area (28% of the country) is under agriculture it is along river banks and on such nutrient-rich soils that much of the fern flora may be found. This may not be due solely to the soil type but also to the availability of water and light where watercourses make a break in forest canopy. The Deep Hydromorphic soils are to be found in the Caroni and Nariva delta regions where the drainage is impeded and the fern flora of such areas restricted to swamp species (see p. 144). 140 FERNS OF TRINIDAD Vegetation Beard (1946) was the first to describe in any detail the natural vegetation of Trinidad and, two years before, of Tobago (Beard, 1944). The main factor controlling vegetation formations in the tropics is humidity/water availability, although tree species composition may be largely in- fluenced by the chemical or physical nature of the soil or substrate. The same may be said for ferns and other pteridophytes, the majority of which have adapted themselves to low light intensities by having a low photosynthetic compensation point, as high humidity is most frequently associated with shady gullies and under dense canopy. A number of species in all groups are, however, adapted for other more exposed situations (mountain tops, landslips, road banks, savanna, and open scrub). The vegetation of Trinidad and Tobago consists of 45% (by area) of forest formations of various kinds either in state or private ownership. Of this area 25-6% of the total land area (129,968 ha; 324,920 acres) are legally proclaimed Forest Reserves, with the intention of them remaining under forest in perpetuity (Chalmers, 1981), the rest being other kinds of state-owned or private forests. The majority of the Forest Reserves are in the east and south-east seasonal lowland areas with very small areas in the ever- wet montane region of the Northern Range. Trinidad still lacks a comprehensive system of nature reserves to conserve representative examples of all natural forest formations. Of the remaining 55% of land surface, some 50% is, or has been, put over to agricultural use since the European colonisation of 1532. In the lowlands smallholdings or larger estates and plantations are still maintained, but many areas in lower montane forest once planted with cocoa have reverted to secondary forests or shifting cultiva- tion. They can be an interesting habitat for ferns. The ecological characteristics of Beard's six major categories of vegetation formation are discussed in relation to the biology of some of the ferns found there. Seasonal (lowland) rain-forest formations Beard (1946) divided these into three associations: (a) evergreen seasonal forest; (b) semi- evergreen seasonal forest; and (c) deciduous seasonal forest. Seasonal formations compose almost the whole of the remaining forest (excluding plantations of teak and pine) in the lowland centre and south of the island. The distribution of rainfall is thus important, those areas with a higher rainfall producing evergreen forest and the lower rainfall deciduous. Other factors such as the degree of slope , porosity , or water retention capabilities of the soil/country rock must also be considered. Thus the forest up to 300 m on the steeper Central Range mountain slopes on shallow skeletal soils over limestone gives not an evergreen but a semi-deciduous forest type. The evergreen seasonal forest is in areas with over 1778 mm (70 in) annual rainfall, with an average of three months each with under 102 mm (4 in) rain but over 51 mm (2 in). Beard calls this forest the Carapa guianensis - Eschweilera subglandulosa association and described five facies of which the Mora forest is the most distinctive, Mora excelsa Benth. being often 60% of the standing crop. Ferns are conspicuous in such untouched or dense forest. On the ground Adiantum obliquum Willd. ,A. latifolium Lam. , A. pulverulentum L. , Asplenium abcissum Willd. , A. salicifolium L. , Cyclopeltis semicordata (Swartz) J.Sm., Danaea nodose (L.) Sm., Diplazium grandifolium (Swartz) Swartz, Pteris tripartita Swartz, Rumohra adiantiformis (Forst.) Ching may be fre- quent. Above this herb layer at 12-27 m is an almost continuous canopy with occasional emergents rising to 30 m. Epiphytes are few on these emergent branches. Asplenium serratum L., Dicranoglossum desveauxii (Klotzsch) Proctor, Oleandra articulata (Swartz) C. Presl, Polypodium ciliatum Willd., P. polypodioides (L.) Watt, Vittaria lineata (L.) Sm. are recorded on fallen trees or on lower trees near watercourses. On the boles of trees are Campy loneurum spp., Lomariopsis marginata Kuhn, and Polybotrya caudata Kunze. CLIMATE, GEOLOGY & VEGETATION 141 In the dry period many of these epiphytes will show signs of stress and obviously wilt, only to be revived on wetting. Some species show special adaptation, such as Polypodium polypo- dioides, in which the fronds are densely covered with overlapping scales and in which the frond itself curls up in drought conditions, thus conserving water. Much of the seasonal forest has been cut over, and in the more open glades amongst stands of Curculigo scorzonerifolia (Lam.) Baker, Cyathea microdonta (Desv.) Domin, and Trimezia martinicensis (Jacq.) Kunth, are thelypterids like Christella dentata (Forssk.) Brownsey & Jermy, Goniopteris megalodus C. Presl, and G. paucijuga (Klotzsch) Pichi-Serm. G. poiteana (Bory) Ching, G. tetragona (Swartz) C. Presl, and Menisciwn nesioticum (Maxon & Morton) Jermy & T. Walker occur especially along trackways and by streams, with Fimbristylis spp., Torulinium odoratum (L.) Hooper, Pityrogramma calomelanos (L.) Link, Selaginella broad- wayii Hieron., and Tectaria incisa agg., including T. x bulbifera Jermy & T. Walker and T. vivipara Jermy & T. Walker. Epiphytes in the more open forest compete with bromeliads, but Ananthacorus angustifolius (Swartz) L. Underw. & Maxon, Polypodium ciliatwn, P. crassifolium L., P. phyllitidis L., and Polytaenium feei (Schaffner ex Fee) Maxon may be found on the smaller branches especially near river courses where light increases. Didymoglosswn angustifrons Fee, D. krausii (Hook. & Grev.) C. Presl, D. punctatum (Poiret) Desv., D. pusillum Swartz, and Microgonium kappler- ianum (Sturm) Pichi-Serm. are occasional on larger boles, but rarely together on the same tree. Semi-deciduous forest areas have the equivalent of five months each with under 1000 mm (40 in) of rain but over 25 mm (1 in). The annual precipitation is 1250-1750 mm. Beard gives two associations depending on rainfall and soil type. They are to be found along the north coast and down the east of the island, becoming the dominant forest type on steeper slopes of the southern hills. They were only briefly visited by the present author. Lianes appeared to be very conspicuous and bromeliads common on the larger trees both on limbs and trunks; other epiphytes, especially ferns and mosses, were almost absent. One facies of the Trichilia- Brosimum association, that containing a substantial amount of Ficus tobagensis Urb. and Cedrela odorata Roemer, is seen on the well drained skeletal soils over limestone at Tamana, Brigand Hill, and Cumaca. Here the fern flora increases in the upper parts of these mountains wiihAdiantum spp. , Asplenium spp. , Danaea nodosum (L.) Sm. , Diplazium cristatum agg. , and Tectaria spp. (especially T. purdiei (Jenman) Maxon). Deciduous seasonal forest lies in areas with 750-1250 mm (30-50 in) of rain per year and with the equivalent of five months with less than 100 mm (4 in) rain each of which two months have under 25 mm (1 in). The formation has a highly discontinuous emergent layer at 12-18 m and an almost continuous layer at 3-10 m. The forest has suffered considerably from human activities. It is to be found on the islands of the NW. peninsula and on the peninsula itself on foothills of the western end of the Northern Range. Epiphytes are virtually absent, although Polypodium aureum L. may be common especially in littoral situations. Dry evergreen forest Coccoloba scrub and Roystonea palm associations are generally termed littoral woodland. These are limited to near the sea on the east coast and have not been studied floristically as far as ferns are concerned. It may be noted here however that Cydosorus interrupta (Willd.) H. Ito is common in the short herb layer underneath coastal coconut palm estates. Normally this is a swamp or ditch species and it must be able to tolerate considerable salt in the strong winds; it has a tough leathery leaf tissue. Montane formations The terminology of the altitudinal zones (formations) of this rain forest follows Beard (1946), although in comparison with S.E. Asia for instance, Beard's Lower Montane Rain Forest, at an altitudinal range of 250-750 m (Fig. 4), would be included in a lowland forest zone by Whitmore (1975). 142 FERNS OF TRINIDAD Lower montane forest is restricted to the Northern Range in an area receiving 1875-3750 mm (75-150 in) of rain per year, evenly distributed. For the most part these forests are over schists, although where they are on skeletal soils over limestone Beard (I.e.) calls them seasonal montane forest, claiming subsoil drainage is so free that it is depressed to the point of seasonal drought. The structure, life-form, and flora are intermediate between lower montane forest and the seasonal evergreen forest described above. Beard gives the following as generally diagnostic of this type of rain-forest: Byrsonima spicata (Cav.) Rich., Licania biglandulosa Griseb., L. ternatensis Hook, f., and Sterculia caribaea R.Br. The boles of the trees are never excessively large, being, on average, around 2 m in girth. They are unbranched for 12 m before opening to a narrow crown, which forms a closed canopy between 20 and 30 m. On the larger boles pyrenocarp lichens cover large areas. Small Hymenophyllaceae, e.g. Didymoglossum and Microgonium species are commonly adpressed to the smooth bark. Epiphytes are relatively few, being on the larger branches; bromeliads such as Tillandsia and Vriesia, and aroids are dominant, and ferns like Asplenium serratum L., Campy loneurum ssp., Polybotrya spp., and Salpichlaena volubilis (Kaulf.) J.Sm. are frequent. The ground flora of Marantaceae and Zingiberaceae is otherwise predominantly of ferns, frequently many individuals of the same species covering many square metres. This dominance is seen in Cydopeltis semicordata (Swartz) J.Sm., Tectaria purdiei, and several Adiantum spp., e.g. A. tetraphyllum Humb. & Bonpl. ex Willd. and A. villosum L. , especially on deeper loamy soils in the lower foothills. Furthermore, species assemblages often change drastically in the next hollow of the hill, suggesting that spore dispersal does not reach very far. At or just above ground level in these denser forests the air movement must be nearly nil. In the deeper forest the herb layer includes Adiantum obliquum Willd. , A. pulverulentum L. , and A. serratodentatum Willd., Dennstaedtia adiantoides (Humb. & Bonpl. ex Willd.) T. Moore, D. ordinata (Kaulf.) T. Moore, Diplazium cristatum (Desv.) Alston, D. grandifolium (Swartz) Swartz, D. striatum (L.) C. Presl, Metaxya rostrata (Kunth) C. Presl, Pteris arborea L. , P. inaequalis (Fee) Jenman, and Tectaria incisa Cav. The paths (traces) through this forest often result in moist clay vertical banks in shade which make an excellent nursery for prothalli and sporelings of many genera, e.g. Blechnum (especially B. fraxineum Willd. and B. occidentale L.), Cyathea, Metaxya, Salpichlaena and Selaginella. Gleichenia may also germinate but lack of light prohibits any substantial growth. Montane rain-forest is defined by Beard (I.e.) as those limited areas on the Northern Range above 750 m (see Fig. 4), mostly on the Aripo Massif. The rainfall is estimated as around 5000 mm (200 in) well distributed throughout the year, and for much of the latter part of the day and during the night cloud or mist covers the area. The average temperatures lie between 15-5 and 21C. There is a general transition zone to the lower montane, stretching down especially along the ridges to 540 m, where Richeria and Eschweilera trinitensis A. C. Smith & J. S. Beard, such characteristic species of the montane forest, are joined in equal dominance by Licania biglandulosa. The canopy is closed at 15-19 m and few trees exceed 1-5 m girth. There is a lower storey of small trees with a number of palms and tree ferns, many of which grow where a wind-break through the canopy occurs. Both lianes and epiphytes are luxuriant, with Carludovica, aroids, bromeliads, and orchids, and with many ferns, especially Hymenophyllaceae (Mecodiumpoly- anthos (Swartz) Copel., Sphaerocionium hirsutum (L.) C. Presl, Trichomanes arbuscula Desv., Vandenboschia hymenophylloides (Bosch) Copel.) and Asplenium serra Langsd. & Fischer, Cochlidium linearifolia (Desv.) Maxon, Dicranoglossum desveauxii, D. polypodioides (Hook.) Lellinger, Elaphoglossum lingua (Raddi) Brackenr., E. longifolium (Jacquin) J.Sm., E. undulatum (Willd.) T. Moore, Lycopodium aqualupianum Spring, L. linifolium L., Polypo- dium astrolepis Liebm., P. ciliatum, P. nematorhizon Eaton, P. pectinatum L., Polytaenium feei, and Xiphopteris serrulata (Swartz) Kaulf. Climbing into the canopy especially on the ridges are Oleandra articulata, O. trinitense Maxon, Polybotrya caudata, and P. cervina (L.) Kaulf. CLIMATE, GEOLOGY & VEGETATION 143 On the ground and in denser vegetation, gullies, and similar niches the following are locally abundant, especially on Aripo and El Tucuche: Asplenium macilentum Kunze ex Klotzsch, Ctenitis aripensis (C. Chr. & Maxon) Lellinger, C. tucuchensis Jermy & T. Walker, Didymoch- laena truncatula (Swartz) J.Sm., Diplazium caracasanum (Willd.) Kunze, D. centripetale (Baker) Maxon, Goniophlebium nephrodioides (Klotzsch) Proctor, Hemidictyum marginatum (L.) C. Presl, Hypoderris brownii Hook., Lindsaea lancea (L.) Bedd., Pteris inaequalis (Fee) Jenman, Saccoloma inaequale (Kunze) Mett., and Tectaria incisa. On banks in the shade are Davalliopsis elegans (L. C. Rich.) Copel., Feea osmundoides (DC.) Copel., Lecanolepis membranaceum (C. Presl) Pichi-Serm., and Selenodesmium rigidum (Swartz) Copel., with Selaginella cladorrhizans A.Braun, 5. porelloides (Lam.) Spring, and Lycopodiella cernua (L.) Pichi-Serm. in more open areas. In the upper reaches of montane forest, tree ferns often increase and Cyathea aspera (L.) Swartz, C. sagittifolia (Hook.) Domin, C. spectabilis (Kunze) Domin, and C. surinamensis (Miq.) Domin are common. On the summit of El Tucuche Lophosoria quadripinnata (Gmel.) C. Chr. is rare but well established in association with Gleichenia brittonii (Maxon) C. Chr. Elfin woodland is a thicket formation of tree ferns, palms, and Clusia intertexta Britt. , with a few other small tree species, reported by Beard from El Aripo, from 840 m to its summit of 925 m. Signs of similar vegetation were also seen by the author on the west ridge of El Tucuche. Trees attain heights of 6-7-5 m, c. 30 cm girth, and emerge from the dense layer of tree ferns (Cyathea caribaea Jenman and C. tenera (Hook.) T. Moore) and palms (Euterpe broadwayana Becc. and Prestoea pubigera Nicholson). The ground layer is virtually bare. The tree stems, branches, and twigs are covered with mosses and lichens, and occasional Hymenophyllaceae (e.g. Sphaero- cionium hirsutum) and Grammitidaceae. Entrada Point Morne Mai d'estomac 648m Heights of Oropouche 658m Gatera Point Elfin Woodland Montane Rain Forest Semi-montane Lower Montane Rain Forest Lowland type forests Seasonal Rain Forest Fig. 4 Diagrammatic section E-W along the Northern Ranges, Trinidad, indicating the altitudinal distribution of vegetation formations. (Adapted from Beard, 1946.) There is a considerable intergrading of lower montane and montane forest, here called 'semi-montane'. 144 FERNS OF TRINIDAD Swamp communities Several swamp communities were described by Beard; there are, however, few ferns associated with them. The palm swamps near Omega Island in the Nariva delta in the east are characterised by Roystonea oleracea (Jacquin) Cook on the seaward edge, and Mauritia flexuosa L.fil. on the landward side. Acrostichum aureum L. may be associated with the former, especially where this abuts on to mangrove swamp. Mangrove swamps (woodland), mainly Avicennia-Laguncularia-Rhizophora, are scattered around the coast and occasionally contain plants of Acrostichum aureum. On the inland side, and on or around alluvial lagoons, herbaceous swamps occur, especially at Caroni, Nariva, Roussillac, and Icacos. A main component is Cyperus giganteus Vahl, forming dense stands up to 3-6 m high, with Montrichardia aborescens Schott and floating Leersia hexandra frequent; Acrostichum species, usually A danaeifolium Langsd. & Fischer, can form very large stools over 1-5 m across. A. danaeifolium is much more tolerant of edaphic conditions than is A aureum and Adams & Tomlinson (1979) note that in Florida the former species extends inland from saline mangrove conditions, just as it does on Aripo Savanna in Trinidad, to fresh-water swamps, often in sink-holes in hammocks and in disturbed marl sites; in contrast A. aureum appears to be confined to coastal mangrove swamps. Proctor (1977) also comments that A danaeifolium in the Lesser Antilles sometimes occurs near freshwater streams not near the sea, and Stolze (1981) records it up to an elevation of 1200 m in Guatemala. Blechnum serrulatum L. C. Rich, is another dominant fern in these swamps and in more inland marshes, e.g. pools in savannas. In keeping with its swamp habitat, the species is abundantly supplied with aerenchyma in the rhizome and stipe, although this is lacking in the root and rachis. A feature which, as far as is known is unique (David Cutler, pers. comm.) is the presence of abundant microscopic hairs on the outsides of the cell walls which line the intercellular spaces. These may best be seen in the aerenchyma but are not wholly confined there, also being prominent in the intercellular spaces of the mid-rachis which lacks aerenchy- ma. The individual hairs are 3-6 /-im long and form a dense coating (Walker, in prep.); their function is unknown. Growth in this species is seasonal, taking place at the onset of the wet season. The Aripo Savanna populations inhabit many of the 'islands' in the savanna and these are often subjected to burning. A plant was gathered which was on burnt-over ground and was producing new fronds, despite many of the growing points of the branching rhizome having been burnt and parts of the rhizome being extensively charred. It may be surmised that B. serrulatum requires the stimulus of a dry period to produce fertile fronds, since at both Newcastle and Kew where plants have been in cultivation for about 20 years, they produce sporangia only very rarely under uniform watering conditions, in contrast to the situation in the field where whole stands may be fertile (T. G. Walker, pers. comm.). Ceratopteris thalictroides (L.) Brongn., and C. lockhartii (Hook. & Grev.) Kunze may also be found in inner coastal swamps with Marsilea sp. and Cyclosorus interrupta. Savanna The term savanna is used to mean a short grass-sedge community with occasional shrubs, usually on firm white (leached) sands and not in swamps. Occasionally where hollows do form and give rise to marsh forest, the margin of such communities is characteristically ringed by so-called palm marsh, an association of Bactris sp., Mauritia, and Chrysobalanus, with Blechnum serrulatum mentioned above. Epiphytes on the palms include Lycopodium linifolium, Nephrolepis rivularis (Vahl) Mett., Oleandra articulata, Poly podium brasiliense Poiret, and Trichomanes arbuscula Desv. The open savanna is dominated by sedges (Rhynchospora holoschoenioides (L. C. Rich.) Herter, R. globosa Roemer & Schultes, etc.) and with Paspalum pulchellum Hochst., Lageno- carpus tremulus Nees, Xyris grisebachii Malme, and the orchids Epistephium parviflorum Lindl., Pogonia rosea (Lindl.) Reichb. f., Otostylis brachystalix (Reichb. f.) Schltr., and CLIMATE, GEOLOGY & VEGETATION 145 Cyrtopodium sp. (Richardson, 1963). Lycopodiella caroliniana (L.) Pichi-Serm. is found here in communities that are sporadically burnt. Kornas (1979) found similar resistance to fire in this species in Zambia. A scrub of Byrsonima-Chrysobalanus is found where Lindsaea portoricensis Desv., L. stricta (Swartz) Dryander var. parvula (Fee) Kramer and, most characteristically, Schizaea pennula (Swartz) Hook, were common. Large plants of Pityrogramma calomelanos were seen here with fronds up to 200 cm and turning reddish before dying; they proved to be hybrids (Walker, 1985) . Man-made habitats Fern species that can withstand open conditions can be incorrigible weeds, and any new road opened up in the clay soils of the tropics, especially when these cut into the side of a mountain and present steep clean-cut banks, will soon have a dense fern tangle. Gleichenia is dominant on such banks and G. bifida (Willd.) Sprengel, G. interjecta Jermy & T. Walker, and G. remota (Kaulf.) Sprengel are common right to the mountain tops, frequently with their hybrids; Dicranopteris pectinata L. Underw. is less abundant. Haufler & Adams (1982) draw attention to the three dimensional gametophyte of G. bifida and suggests it would confer an advantage to the species during the daily fluctuation in solar radiation, over a flat, plate-like prothallus which would lose more water. Stokey (1950) indicates that gametophytes of this species reach sexual maturity more rapidly than any other member of the Gleicheniaceae she had studied. This also would be beneficial in such exposed habitats. The following species formed a typical species assemblage on the roadbanks of the Northern Range: Adiantopsis radiata (L.) Fee, Adiantum obliquum, A. petiolatum Desv., A. terminatum Kunze, A. tetraphyllum, Anemia pastinacaria Prantl, A. phyllitidis (L.) Swartz, Blechnum occidentals, B.fraxineum, Cyathea microdonta, C. spectabilis, HemionitispalmataL., Lindsaea portoricensis, Lycopodiella cernua, Nephrolepis biserrata (Swartz) Schott, N. exaltata (L.) Schott, Selaginella cladorrhizans, S. diffusa (C. Presl) Spring, 5. mnioides A.Braun, S. platy basis Baker, 5. trifurcata Baker, and Trichomanes pinnatum Hedwig. In wet runnels along roadsides Cyclosorus interrupta is frequent and Trichomanes pinnatum Hedwig is sometimes found, often in full sun. It may be noted (Walker, pers. comm.) that meiosis in filmy ferns virtually ceases in pronounced dry seasons, whereas in Gleichenia, for example, it continues throughout. In Britain, in cultivation, most tropical ferns have two major spore-producing flushes, spring and autumn, when day and nights are more or less equal as in the tropics. The following points may be made on individual species. Anemia pastinacaria has a very marked seasonal growth pattern, and during the wet season the fronds were all green and producing spores abundantly. In the dry season T. G. Walker (pers. comm.) reports that plants from the same locality were in a state of dormancy with the fronds brown and crisp. In patches the bank had been burnt over and the Anemia plants survived this treatment. On sending the dormant rhizomes back to Newcastle, the dormancy was quickly broken by regular watering and new fronds were rapidly produced. This behaviour corresponds to that shown in the Type III 'summer green' pattern noted by Kornas (1977) for a number of ferns growing in the seasonal climate of Zambia. In this Anemia the dead fronds persist, attached to the rhizome for some time. The plants did not survive long enough in cultivation to observe if this seasonal pattern of events was genetically determined and would persist under a regular uniform regime of watering, as has been noted in Adiantopsis radiata (see below). One of the prominent features of the genus Adiantopsis is the articulation of the pinnules to the stipicel, and, in the case of A. radiata at least, in the dry season the pinnules are frequently shed, leaving the black stipe and its branches bare. Eventually the stipe itself drops. This behaviour is evidently inherent, as, according to Walker (pers. comm.), in cultivation under a more or less uniform regime of watering, the plant continues to behave in the same way, losing first its pinnules and then its stipe. Normally there tends to be a short rest before the production of new fronds. It may be noted in passing that, unlike a number of ferns whose pinnules are 146 FERNS OF TRINIDAD deciduous, e.g. Nephrolepis spp. , those of A. radiata do not shed when herbarium specimens are being prepared and thus behave like Adiantum deltoideum Swartz, in which herbarium plants show both bare stipes and fully furnished fronds as in nature. Hemionitis palmata has accessory reproduction by bulbils borne on the fronds. Under damp conditions these bulbils grow and develop into new plants, whilst under dry conditions the buds remain inactive. When in a uniform environment of the greenhouse (Walker, pers. comm.) plants tend to die back irregularly, i.e. apparently not at any fixed period, although several plants in a pan tend to behave as a unit, all dying back together. The die-back does not appear to be related to dry conditions in the field, as the species in Jamaica tends to live in rather dry situations and is fertile at most times of the year. In areas such as Trinidad dry-season fires are ecologically important. Kornas (1979) discusses this factor in detail for Zambia. He mentions no information being available for the fire resistance or susceptibility of Pteris vittata L. Walker records (pers. comm.) collecting a specimen on an open rocky grassed area near Port of Spain where fire had previously raged. Other species there included Adiantum lucidum (Cav.) Swartz and A. villosum. Similarly the only stand of Pteridium aquilinum var. arachnoideum (Kaulf.) Brade in Trinidad is subject to seasonal fires. Cocoa plantations are shady and have a characteristic ground flora which includes several Adiantum species (e.g. A. macrophyllum Swartz, A. tetraphyllum), Asplenium laetum Swartz, Goniopteris poiteana, Selaginella diffusa, and on the trees themselves Poly podium phyllitidis and Polytaenium brazilianum (Desv.) Bened. Ananthacorus angustifolius, Lycopodium spp., Lygodium micans Sturm, and L. venustum Swartz are common at the edges of such plantations, in gardens, and along the borders of secondary scrub. Acknowledgements I thank Dr T. G. Walker for numerous comments and observations as a result of his fieldwork in Trinidad, and Dr C. D. Adams for commenting on the MS. I also thank Miss Alison Paul for the care with which she has drawn the maps and Edward Arnold for permission to reprint Fig. 3. References Adams, D. C. & Tomlinson, P. B. 1979. Acrostichwn in Florida. Am. FernJ. 69: 42-46. Beard, J. S. 1944. The natural vegetation of the Island of Tobago, British West Indies. Ecological Monogr. 14(2): 136-163. 1946. The natural vegetation of Trinidad. Oxford Forestry Mem. 20: 1-152. Berridge, C. E. 1981. Climate. In St G. C. Cooper & P. R. Bacon (Eds), The natural resources of Trinidad and Tobago: 2-12. London. Chalmers, W. S. 1981. In St G. C. Cooper & P. R. Bacon (Eds), The natural resources of Trinidad and Tobago: 78-105. London. Chenery, E. M. 1952. The soils of central Trinidad. Port of Spain. Hardy, F. 1981. Soils. In St G. C. Cooper & P. R. Bacon (Eds), The natural resources of Trinidad and Tobago: 23-42. London. Haufler, C. H. & Adams, W. W. 1982. Early gametophyte ontogeny of Gleichenia bifida (Willd.) Spreng.; phylogenetic and ecological implications. Am. J. Bot. 69: 1560-1565. Kornas, J. 1977. Life-forms and seasonal patterns on the pteridophytes in Zambia. Acta Soc. Bot. Pol. 46: 669-690. 1979. Distribution and ecology of the pteridophytes in Zambia. Warsaw/Krakow. Kugler, H. G. 1959. Geological map of Trinidad. Petroleum Association of Trinidad. CLIMATE, GEOLOGY & VEGETATION 147 Proctor, G. R. 1977. In R. A. Howard (Ed.), Flora of the Lesser Antilles, 2. Pteridophyta. Harvard. Richardson, W. D. 1963. Observations on the vegetation and ecology of the Aripo savannas, Trinidad. /. Ecol. 51: 295-313. Stokey, A. G. 1950. The gametophyte of the Gleicheniaceae. Bull. Torrey Bot. Club 77: 323-339. Stolze, R. G. 1981. Ferns and fern allies of Guatemala II. Polypodiaceae. Fieldiana Bot. II, 6: 1-522. Chicago. Walker, T. G. 1985. Cytotaxonomic studies of the ferns of Trinidad 2. The cytology and taxonomic implications. Bull. Br. Mus. not. Hist (Bot.) 13: 149-249. Whitmore, T. C. 1975. Tropical rainforests of the Far East. Oxford. Cytotaxonomic studies of the ferns of Trinidad 2. The cytology and taxonomic implications Trevor George Walker Department of Plant Biology, Ridley Building, The University of Newcastle uponTyne, Newcastle upon TyneNEl 7RU Contents Synopsis 149 Introduction 151 Materials and methods 151 Chromosome fixing and karyotyping 151 Spore and stomata measurements 153 Notes on individual taxa 154 I. Marattiaceae 161 II. Schizaeaceae 162 III. Adiantaceae 168 IV. Hymenophyllaceae 178 V. Gleicheniaceae 185 VI. Polypodiaceae 191 VII. Grammitidaceae 202 VIII. Cyatheaceae 205 IX. Dennstaedtiaceae 206 X. Thelypteridaceae 208 XI. Aspleniaceae 213 XII. Blechnaceae 229 General discussion 242 Acknowledgements 244 References .. 245 Synopsis Several visits to Trinidad and Tobago provided material for the cytotaxonomic study on ferns which is described here. Cytological information is available for 155 taxa or cytotypes, and the notes on individual taxa amplify some of this information both in regard to taxonomic problems and to interrelationships on the specific or generic level. An important new development in technique is the production of standardised karyotypes of a number of taxa and some of the uses to which analyses of them may be put are discussed. Taxonomic implications resulting from these cytological studies are discussed and some morphological features of the taxa and hybrids concerned are compared. The fern flora of Trinidad contrasts with that of Jamaica in having a significantly lower incidence of polyploidy and the possible reasons for this are discussed. However, similarities are shown with Jamaica in the commonplace occurrence of hybridity and the rarity of agamospory. Bull. Br. Mus. nat. Hist. (Bot.) 13 (2): 149-249 Issued 30 May 1985 150 FERNS OF TRINIDAD 12 15a 16 14a 61 30 61 00 Fig. 1 Map of Trinidad showing collecting localities. 1 Scotland Bay. 2 Morne Catherine. 3 Lady Chancellor Road, St Anns. 4 Port of Spain. 5 Loanga. 6 El Tucuche. 7 Maracas Falls. 8 Mt Tabor. 9 Naranja. 10 St John's Valley, Tunapuna. 11 St Augustine. 12 Caura Valley. 13 Blanchisseuse. 14a Aripo Heights. 14b Aripo Valley. 15a Morne Bleu - Blanchisseuse Road - La Laja. 15b Las Lapas Trace. 16 Brasso Seco. 17 Long Stretch Reserve. 18 Aripo Road. 19 Guanapo Valley. 20 Grand Fond Road. 21 Caroni Swamp. 22 Princess Margaret Highway. 23 Churchill-Roosevelt Highway. 24 Valencia. 25 Sangre Grande. 26 Arima. 27 Guiaco- Valencia Forest Reserve. 28 Arena Forest Reserve. 29 Melajo Forest Reserve. 30 Aripo Savanna. 31 Near Guiaco, confluence of rivers. 32 Corosal Hill. 33 Tabaquite. 34 Mt Harris. 35 Brasso Venado. 36 Mt Tamana. 37 Brickfield. 38 Pitch Lake. 39 Point Fortin Experimental Farm. 40 Irois. 41 Grass swamp near Fullarton. 42 Icacos Point swamp. 43 Quarry Forest. 44 Texaco Village, Trinity Hills. 45 Biche. CYTOLOGY 151 Introduction Whilst carrying out a cytotaxonomic investigation of the fern flora of Jamaica (Walker, 1966a, 19736), it was felt to be desirable that another part of the Caribbean region should be examined for comparative purposes and to test overall conclusions. Trinidad was considered to be ideal for such an investigation, being at the other end of the chain of West Indian islands from Jamaica and lying so close to the South American mainland that floristically it can be regarded as part of the latter. An account of the general environment of Trinidad is given in paper 1 of this trilogy (Jermy, 1985). The present survey of the cytology of the fern flora of Trinidad and Tobago is the outcome of three major periods of field work carried out from 1963 to 1974, and the subsequent cultivation of plants at the University of Newcastle upon Tyne and the Royal Botanic Gardens, Kew. These were integrated with laboratory studies at Newcastle by T. G. Walker and herbarium studies at the British Museum (Natural History) by A. C. Jermy. The first and most comprehensive collecting expedition was made in July to September 1963 by the author and A. C. Jermy with the financial support of the Nuffield Foundation, the Godman Exploration Fund, and of the respective institutions. A second visit in April 1966 by T. G. and M. G. Walker was also supported by the Nuffield Foundation. A further visit by A. C. Jermy, funded by the British Museum (Natural History) and accompanied by J. Simmons (R.B .G. , Kew) in August to October 1974 allowed collecting to extend to Tobago. The locations of the collecting sites are shown in Fig. 1. One advantage of the visit occurring in April was that this was during the marked dry season and observations were made as to how some of the ferns reacted to this rainless period. In addition to the above, a small but very valuable amount of material has also been made available by other collectors, in particular Dr Alice Fay. During the survey it became clear that five previously unrecognised taxa were present and these, together with nine of the 14 different hybrid combinations found, have been described and validated in paper 3 (Jermy & Walker, 1985). In addition a new combination has been made there, namely in Meniscium. In the notes the arrangement and concept of families and genera follow those of Crabbe, Jermy & Mickel (1975), with the exception of Polypodium s.l. where the segregate genera are simply indicated informally. Materials and methods Chromosome fixing and karyotyping Cytological fixations were made in the field in 1963 and 1966 and have been heavily sup- plemented by those made on the living collections at Newcastle and Kew. The methods involved in obtaining meiotic and root tip squashes are standard, whilst the supplementary use of young croziers as a source of dividing somatic cells in those epiphytes in which roots either developed only very sparsely or which cannot be cropped without great disturbance to the plant, have been described previously (Walker, 19736). Voucher specimens of all cytologically authenticated material are deposited in BM, with duplicates, where available, in TRIN and herb. T. G. Walker. Karyotyping has been carried out for a number of species. This has raised problems of how to show the essential facts in the most economical manner and in a form which can be standardised for comparative work in the future. Many attempts were made before evolving a method which appears to achieve these aims. One of the principal difficulties encountered in fern karyotypes as opposed to those of other organisms is a direct consequence of the numbers involved. Whilst different forms of chromosomes may be perfectly obvious in a plant with say 2n = 10 (Fig. 2) the same forms may not be so immediately obvious when part of a chromosome complement of 2n = 74 for example (Fig. 28B). Furthermore, there is no standard way of setting out the data and 152 FERNS OF TRINIDAD i MII ii B Fig. 2 Complements of 2n = 10 extracted from karyograms of: A, Poly podium latum (2n = 74). B, tetraploid P. aureum (2n = 148). Compare with the complete karyograms in Figs 29 and 25C respectively. almost any issue of a journal in which a high proportion of its articles are devoted to karyotypic studies (e.g. Cytologia, 1978) will show a multiplicity of methods of presentation. Whilst it is true that a number of papers concerned with fern karyotypes have been published by Japanese workers, viz. Kawakami (1970, 1971), Takei (1969), Tatuno & Kawakami (1969), Tatuno & Okada (1970), Tatuno & Takei (1969a & b), Tatuno & Yoshida (1966, 1967), these have been undertaken predominantly with a particular end in view, namely the demonstration of alleged ancestral basic chromosome numbers (b). Their methods have not proved to be particularly suitable for the present purposes and hence an independent system has had to be devised. It has been found in a much wider survey of fern karyotypes (Walker, unpub.) that a study of chromosome length can give very useful information. In order to get the most informative results experiments were carried out to determine the most satisfactory means of measurement that would be both accurate and at the same time produce suitable groupings or classes of chromosome length which could be illustrated graphically and from which valid inferences could be made. The method adopted here appears to meet these criteria and involves photographing well-spread somatic metaphase nuclei and enlarging to a magnification of x 3000. The individual chromosomes are measured on the enlargements to the nearest mm, each 1 mm representing a 'unit of length'. Such a unit is equal to 0-33 /um actual length. Thus a chromosome of say 19 units length is actually 6-33 /u,m long. However, for most purposes it is more convenient to use the units of length as they stand rather than to convert them. This method of measurement has proved to give a very satisfactory grouping of results. If the measurement is made to an allegedly greater degree of accuracy, the accuracy becomes more apparent than real - see Bentzer et al. (1971) whilst if smaller units of length are employed the overall pattern tends to be obscured. Conversely, if these units are larger this produces too little separation of the length classes in those species which show only a very small range in actual lengths of the chromosomes such as in diploid Polypodium aureum (from 2-0 /x,m to 4-0 /u,m). Thus measurements taken from enlargements of x 4000 and x 2000 fail to meet the requirements set out above. The karyograms are stylised whereby the chromosomes are presented as solid columns of uniform width in which the length and centromere positions, together with satellites if present, are depicted accurately in standard form. All chromosomes of the same length are grouped together, starting in each group with those in which the centromere is most nearly median and finishing with those in which it is most nearly terminal. The length of the chromosomes is given below each group in units of length and, as has been noted above, these may be converted to actual measurements by dividing this figure by 3-0. The total number of chromosomes in a complement with their centromere positions is presented in the form of Tables, the nomenclature used being that of Levan, Fredya & Sandberg (1965). In their system two fixed points of a chromosome are recognised, the median point (M) which is at the exact mid-point of the chromosome and the terminal point (T) at the extreme end. The chromosome arm between these two points is divided into four equal regions these being, from the centre to the end, the median region (m), the submedian region (sm), the subterminal region (st) and the terminal region (t). Levan et al. (1965) gives several methods of calculating the centromere position, the one used here involving the ratio of the length of the long arm to the CYTOLOGY Table 1 L/S ratios and centromere positions. Centromere position Ratio as in Levanetal. (1965) Ratio used in present survey Median point (M) Median region (m) Submedian region (sm) Subterminal region (st) Terminal region (t) Terminal point (T) 1-0 1-0-1-7 1-7-3-0 3-0-7-0 7-0-00 00 1-0 >l-0-l-7 >l-7-3-0 >3-0-7-0 >7-0- J10844:! J11001 T6027,T6028 > T6036, T6038 T6040, T6041 T6043,T6044 ; Fay 417 J2045 J10999:3 T6354 T6088 T6444 Tl 1066, Tl 1072 T7397.T11067 \ T11068,T1107l| J10996,J10978:3 T6667 J 11047 T6064 T10523 J 10945 J11332 T10632 T10633 T10634 T6614 30 12 8 12 12 10 30 12 15a 3 10 18 16 12 12 16 Tobago 7 7 7 37 n = 134 2n = 116 2n = 58 'n' = 114 Us and 38 Is + 38 Us n = c. 76; 2n - c. 152 2n = 60 2n = 60 c. 76 Is + c. 37 Us, meiosis irregular n = 30 n = 60; 2n = 120 n = 30; 2n = 60 2n = 60 2n = 60 n = 30 n = 30 n = 30 2n = 60 n = c. 30 2n = 60 n = c. 30 2n = 60 n = c. 30; 2n = 60 n = 122 4x 2x 3x agamosporous 4x sexual 2x 2x 5x sterile hybrid 2x 4x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x 2x hyper-8x CYTOLOGY 155 Table 2 - cont. Taxon Voucher collection No. Locality reference Chromosome number Ploidy A. terminatum Kunze T6111 12 n = 60