Lichenologist 35(2): 137–146 (2003) doi:10.1016/S0024-2829(03)00017-3 The lichen genus Ramalina Ach. (Ramalinaceae) on the outlying islands of the New Zealand geographic area J. M. BANNISTER and D. J. BLANCHON Abstract: The diversity of species of Ramalina occurring on the outlying islands of the New Zealand geographic area is linked to their methods of dispersal and the origin and principal climatic features of the islands themselves. It appears that species of Ramalina have reached these islands by transoceanic, wind-borne dispersal of ascospores and soredia, not necessarily in the direction of the prevailing winds. Species have become established on the islands only if both climatic and habitat requirements have been met.  2003 The British Lichen Society. Published by Elsevier Science Ltd. All rights reserved. Key words: species of Ramalina, New Zealand geographic area, emergent oceanic islands, wind-borne dispersal. Introduction The genus Ramalina in Australia was revised by Stevens (1987), but the occurrence of species on Lord Howe, Norfolk and Macquarie Islands was not documented. Subsequently the species occurring on Norfolk Island were documented by Elix et al. (1992). Blanchon et al. (1996) revised the genus in New Zealand with distribution being based on herbarium collections. Recently the distribution of species of Ramalina on the main islands of New Zealand has been recorded and mapped more intensively (P. Bannister, J. M. Bannister & D. J. Blanchon, unpublished). The study reported here includes the species of Ramalina occurring on the outlying islands of New Zealand’s geographic area (Fig. 1): the Kermadec Islands, the Chatham Islands, the Bounty Islands, the Snares Islands, the Antipodes and Auckland J. M. Bannister: Department of Botany, University of Otago, P.O. Box 56, Dunedin, New Zealand. D. J. Blanchon: Resource Management Research Group, School of Landscape and Plant Science, UNITEC, Private Bag 92025, Auckland, New Zealand. 0024-2829/03/020137+10 $30.00/0 Islands and Campbell Island. This study also includes Norfolk Island, Lord Howe Island and Macquarie Island. Although they are Australian territory, biogeographically they have very close links to New Zealand (Wardle, 1978; Orchard 1994). In terms of climate, these islands range from subtropical to subantarctic. The Snares and Bounty Islands consist of basement rock (granite) and are part of the original New Zealand continental landmass (Turnbull 1999), while the other islands have arisen in isolation, by volcanic or tectonic activity. The presence of various species of Ramalina on the emergent islands implies both successful dispersal from other islands, or from mainland sites in New Zealand or Australia, and climatic conditions that allow establishment of the thallus and its subsequent growth and reproduction. The distribution of species on these geologically recent islands provides an interesting case study of long-distance dispersal. Material and Methods Synoptic climatic and other information, summarized in Table 1, was collated from a variety of sources (De Lisle 1965; Fineran 1969; Elix & Streimann 1989;  2003 The British Lichen Society. Published by Elsevier Science Ltd. All rights reserved. 138 THE LICHENOLOGIST Vol. 35 30°N Norfolk Id. Kermadec Is. Three Kings Is. Lord Howe Id. Australia North Island Sydney Auckland Tasman Sea 40° Wellington South Island Christchurch Chatham Is. Dunedin Stewart Island Tasmania Snares Is. Bounty Is. Antipodes Is. 50° Auckland Is. Campbell Id. Macquarie Id. Pacific Ocean 180°E F. 1. The New Zealand archipelago and surrounding islands. Selkirk et al. 1990; Wardle 1991; Kantvilas & Seppelt 1992; Archer & Elix 1994; Atkinson 1996; O’Connor 1999; Turnbull 1999; West & Rance 1999; Sykes et al. 2000; and others) and the records of the National Institute of Water and Atmospheric Research (NIWA). Literature reports of species of Ramalina have been checked and only those that have been confirmed from herbarium specimens are included in this study. Collections were borrowed from the following herbaria: ADT, AK, AKU, BM, CANB, CANU, CHR, FH, HO, MEL, MSC, O, OTA, WELT and the private collection of P. N. Johnson (Dunedin). Nomenclature is based on the taxonomic revisions of the genus in Australia (Stevens 1987) and New Zealand (Blanchon et al. 1996). Ramalina subfraxinea var. leoidea (Nyl.) G. N. Stevens and R. nervulosa var. luciae (Molho et al.) G. N. Stevens are currently re- ferred to as R. leoidea (Nyl.) Nyl. and R. luciae Molho et al. in the Australian checklist (McCarthy 2002). Two new species have been described since these revisions, R. stevensiae Elix (Elix et al. 1991) and R. meridionalis Blanchon & Bannister (Blanchon & Bannister, 2002). The accepted names of Ramalina species in the study area are as follows: R. australiensis Nyl., R. canariensis J. Steiner, R. celastri (Spreng.) Krog & Swinscow, R. erumpens Blanchon, Braggins & A. Stewart bis, R. exiguella Stirt., R. geniculata Hook. f. & Taylor, R. inflata (Hook. f. & Taylor) Hook. f. & Taylor, R. leoidea (Nyl.) Nyl. (syn.: R. subfraxinea var. leoidea (Nyl.) G. N. Stevens), R. luciae Molho, Bodo, W. L. Culb. & C. F. Culb. [syn.: R. nervulosa var. luciae (Molho et al.) G. N. Stevens], R. meridionalis Blanchon & Bannister, R. cf. microspora Kremp., R. pacifica Asah., R. peruviana Ach., R. stevensiae Elix and R. unilateralis F. Wilson. 2003 Ramalina on o#shore islands of NZ—Bannister & Blanchon 139 T 1. The outlying islands of New Zealand showing latitude, area, age, maximum elevation (ME), mean annual temperature (MAT), mean annual rainfall (MAR) and total number of vascular plant taxa (VAS) Island Latitude Area (km2) Age (Ma) ME (m) MAT ((C) MAR (mm) VAS Norfolk Kermadecs Lord Howe Chathams Bounty Snares Antipodes Auckland Campbell Macquarie 28(58#S 29(30#S 31(33#S 44(00#S 47(05#S 48(01#S 49(41#S 50(30#S 52(35#S 54(30#S 35 33 17 950 1·3 2·6 21 625 113 128 3 1·6 7 2·5 – – 1·05 10 6 1·7 316 518 875 287 70 130 366 668 569 434 18·5 18·9 19 11 – 9·8* – 8·5 6 4·5 1313 1500 1675 800 – 1482* – 1520–2160 1400 895 445 265 459 472 0 21 72 233 212 45 (–) No information available. *Based on limited information. Study area Norfolk Island Norfolk Island is situated on the submarine Norfolk Ridge which extends from New Zealand to New Caledonia. It lies 1610 km NE of Sydney (Australia), 1100 km NNW of New Zealand and 670 km S of New Caledonia. Volcanic in origin, the island has been emergent since the late Pliocene. The climate is subtropical, rainfall occurs throughout the year but can be variable and short droughts may occur. Humidity is moderate, RH 72–81%, and the prevailing winds are from the west to south-west in winter, and from the north-east to south-east in the summer. The vascular flora is similar to that of Lord Howe Island and is more closely related to New Zealand than Australia (Orchard 1994). The vegetation consists of remnant rainforest, open Araucaria heterophylla (Salisb.) Franco woodland, pasture with isolated trees or small stands of Araucaria and weedy forest of Psidium littorale Raddi and Olea africana Mill. (Elix & Streimann 1989). Kermadec Islands The Kermadec Islands, 976 km NE of New Zealand, rise from a volcanically active submarine ridge, the Kermadec Ridge, that extends NE from the Taupo Volcanic Zone in the North Island of New Zealand. This is part of the subduction zone between the Australian and Pacific tectonic plates and two islands, Raoul (Sunday Island) and Curtis, remain actively volcanic. The climate is subtropical, with a mean average temperature of 18·9(C. In winter the winds and ocean currents are from the south-west, while in summer the prevailing winds and currents are from the north-east (Sykes et al. 2000). There are dry forests at low altitude (below 244 m) and wet forests at higher altitudes and in more moist habitats. The dominant tree in both forest types is Metrosideros kermadecensis W. R. B. Oliver. There are also areas of coastal scrub or low forest. The vascular flora is more closely related to that in New Zealand than it is to Norfolk Island and Lord Howe Island (Wardle 1991; Sykes et al. 2000). Lord Howe Island Lord Howe Island is situated on the Lord Howe Rise, an undersea volcanic plateau extending from New Zealand to a point between Queensland and New Caledonia. The island is the emergent part of a sea mount which is a remnant of a large shield volcano. Lord Howe Island has been above sea level since the late Miocene. It lies 700 km NE of Sydney and 1250 km NW of New Zealand. The climate is subtropical and humidity is moderate (RH 70–78%); prevailing winds in winter are westerlies, while in summer they come from the east. The vascular flora is similar to that of Norfolk Island and, like that flora, is more closely related to New Zealand than Australia (Orchard 1994). Three types of forest are found on the island: lowland exposed areas with a dry Drypetes/Cryptocarya forest with thickets of Howea spp.; lowland sheltered forests with Drimys howeana F. Muell., Cleistocalyx fullageri (F. Muell.) Merrill & Perry, Atractocarpus stipularis (F. Muell.) Puttock ex P. S. Green and Chionanthus quadristaminea F. Muell.; and, at higher altitudes, a cloud forest of Hedyscepe canterburyana H. Wendl. & Drude, Lepidorrachis mooreana (H. Wendl. & Drude) Burret, Dracophyllum fitzgeraldii F. Muell. and Negria rhabdothamnoides F. Muell. occurs (Archer & Elix 1994; Orchard 1994). Chatham Islands The Chatham Islands are situated on the Chatham Rise, part of the Campbell Plateau, a submerged portion of the New Zealand subcontinental landmass. The islands, 870 km E of Christchurch, have been formed by volcanic eruptions over a long period of time, accompanied by uplift and changes in sea level. The present landmass has been emergent for at least 2·5 million years. Summer droughts can occur, but there 140 THE LICHENOLOGIST Vol. 35 are frequent rain showers and spray-laden gales. Conditions are often overcast with high humidity, and frosts are infrequent. The prevailing winds are from the south-west. The original broad-leaved forest has been extensively fragmented by agriculture. The most abundant tree is Corynocarpus laevigata J. R. Forst. and G. Forst., but many other broad-leaved trees are present. Tree-heath, dominated by Dracophyllum arboreum Cockayne, occurs on drier peats, and scrub and fern-dominated communities occurs in coastal areas, upland sites and on peat domes. The islands lie near the subtropical convergence and although the climatic zone is cool-temperate, some of the vascular plants show a$nities with warmtemperate species (Wardle 1991; Atkinson 1996). on Campbell Island, these volcanoes were eroded by glaciation occurring from 1 million to 15 000 years ago. The temperatures are cool and equable, and while levels of sunshine are low, there are more hours of bright sunshine than on Campbell Island (estimated 960 h). Humidity is high with frequent rain (over 300 days a year). The prevailing strong winds are westerly and north-westerly (De Lisle 1965). The vegetation shows a distinct altitudinal zonation, with tussock grassland, forests of Olearia lyallii, dwarf forest of Metrosideros umbellata Cav., subalpine shrubland (Cassinia and Dracophyllum spp.) with herb moor and an alpine region dominated by mosses. The plant species are closely related to those in New Zealand (Wardle 1991; West & Rance 1999). Bounty Islands Campbell Island Situated on the Campbell Plateau, 624 km E of the southernmost point on the South Island of New Zealand, these small granite islands are part of the original New Zealand continental landmass. They are drenched by sea spray and support dense populations of seals and sea birds. There are no available climatic data, but cloudy, overcast conditions can be surmised with prevailing winds from the south-west. The Snares These small, granite islands are part of the New Zealand continental landmass, situated 209 km S of New Zealand. Little climatic information is available, but it appears that the climate is comparable to that of South West Cape on Stewart Island, with a high rainfall, over 200 rain days per year, and predominantly westerly winds (NIWA 2002, unpublished data). Tree-heath, comprising three woody species, Olearia lyallii Hook. f., Brachyglottis stewartiae (J. B Armstr.) B. Nordenstam and Hebe elliptica Pennell, covers most of Main Island, with tussock grassland in more exposed areas (Wardle 1991; West & Rance 1999). Antipodes Islands The Antipodes Islands are situated on the Campbell Plateau 872 km SE of New Zealand. They are volcanic in origin, and while detailed climatic information is not available, it has been suggested that the rainfall is probably only half that of the Auckland Islands (O’Connor 1999). The islands are peat-covered and there is no forest, the vegetation consisting of tussock grasslands with patches of shrubs and fern gullies (West & Rance 1999). Auckland Islands The Auckland Islands are situated on the Campbell Plateau 465 km S of New Zealand and are the eroded remnants of two shield volcanoes. Volcanic activity took place from about 20 to 10 million years ago, and as Campbell Island is situated on the Campbell Plateau, 700 km S of New Zealand. The Auckland Islands are 300 km to the north-west and Macquarie Island is 700 km to the south-west. Campbell Island is a glacially eroded remnant of a shield volcano that was active 11 to 6 million years ago. The climate is cloudy with uniformly cool temperatures and a narrow daily temperature range. The number of sunshine hours average 660 per year, and the island is moist, with over 300 rain days a year, and windy, with strong, persistent westerly and north-westerly winds (De Lisle 1965). The vegetation is similar to that on Auckland Island, but the forest is even more dwarfed. Metrosideros umbellata is absent and the dominant and tallest tree is Dracophyllum longifolium (J. R. Forst. and G. Forst.) R. Br. with a subcanopy of Myrsine divaricata A. Cunn. There are tussock grasslands, subalpine shrubs and a higher alpine area of rushes and cushion-forming plants (Wardle 1991; West & Rance 1999). Macquarie Island Macquarie Island is an emergent part of the Macquarie Ridge that extends 900 km SSE from Fiordland, New Zealand. The ridge formed as a result of sea floor spreading rather than volcanic activity c. 27 million years ago, and it forms part of the boundary between the Australian and Pacific plates. Islands may have emerged and foundered repeatedly along the crest of the ridge. One view is that the high points of the island were raised between 200 000 and 90 000 years ago (Hnatiuk 1993). However, sea level fluctuations and downward movements over the last 2 million years may mean that the island has been submerged for part of that time. The island may not have been glaciated, and its oceanic position and the fact that it lies north of the Antarctic convergence (the sharp boundary between the cold Antarctic waters and warmer subantarctic waters) ameliorates the temperature slightly. The climate is uniformly cool, cloudy (with only 850 h sunshine per year) and windy (the prevailing winds are westerly and north-westerly). Rain falls on more than 300 days per year and the relative humidity is 89%. Macquarie Island lacks trees and shrubs, the vegetation consisting of tall tussock grassland, short grassland 2003 Ramalina on o#shore islands of NZ—Bannister & Blanchon 141 T 2. Occurrence of species of Ramalina on the outlying islands of New Zealand and the presence of these species in Australia and New Zealand. Species R. R. R. R. R. R. R. R. R. R. R. R. R. R. R. Nor How Ker Cha Ant Auk Cam Mac + australiensis canariensis celastri erumpens exiguella geniculata inflata leoidea luciae meridionalis cf. microspora pacifica peruviana stevensiae unilateralis + + + + + + + + + + + + + + + + + + + Total + + + + + + NZ + + + + AU + + + + + + + + + 6 4 8 + 4 + + 1 1 1 + + + + + 2 10 10 Key to locations: Norfolk Island (Nor), Lord Howe Island (How), Kermadec Islands (Ker), Chatham Islands (Cha), Antipodes Islands (Ant), Campbell Island (Cam), Auckland Islands (Auk), Macquarie Island (Mac), Australia (AU), New Zealand (NZ). with herbs on upper slopes and fellfield covering about 45% of the plateau uplands (Selkirk et al. 1990). A greater proportion of the vascular flora is shared with New Zealand than Australia (Wardle 1978). Lichen communities are found on cli#s and rocks in the coastal area, in herbfield and on vegetation, peat and rock in the fellfield (Selkirk et al. 1990; Kantvilas and Seppelt 1992). Results Species of Ramalina Norfolk Island The most important lichen habitats are rainforest, open Araucaria woodland, exposed rock surfaces inland and on the foreshore, and fenceposts (Elix et al. 1992). The most recent Australian checklist (McCarthy 2002) lists six species of Ramalina: R. arabum, R. exiguella, R. leiodea, R. pacifica, R. peruviana and R. stevensiae. While earlier lists had also included R. australiensis, R. canariensis, R. celastri and R glaucescens, Elix et al. (1992) were unable to examine specimens and verify these names. We have examined specimens of those taxa listed in the Australian checklist in AK, CANB, CHR, HO and MEL (Table 2). Blanchon & Bannister (2002) have referred specimens previously named R. arabum to the newly described and mainly corticolous R. meridionalis. Kermadec Islands Ramalina australiensis, R. celastri, R. exiguella, R. geniculata, R. luciae, R. pacifica and R. peruviana have been identified from specimens in AK, AKU, CHR and WELT (Table 2). Specimens with inflated thalli, previously labelled R. microspora, are R. geniculata, while a few in AK, CHR and WELT, originally labelled R. geniculata, are tentatively identified as R. cf. microspora. Ramalina microspora Kremp. s. str. appears to be restricted to coastal habitats in the Hawaiian Islands. Elix & McCarthy (1998) listed two other species, R. farinacea (L.) Ach. and R. fastigiata (Pers.) Ach. We identified as R. pacifica a specimen collected by W. R. B. Oliver in 1908, sent to Kew and determined as R. farinacea (AK 20267). Another specimen collected by Oliver and labelled R. fastigiata, was found to contain divaricatic acid and may belong to R. cf. microspora. Ramalina fastigiata contains evernic acid and does not occur in Australia or New Zealand. 142 THE LICHENOLOGIST Lord Howe Island The main lichen habitats are found in the three types of forest, on exposed rock surfaces and on isolated trees in pasture. While Elix & McCarthy (1998) listed R. complanata (Sw.) Ach., R. leiodea and R. peruviana from Lord Howe Island, Archer & Elix (1994) corrected R. complanata to R. subfraxinea, without confirming its identity from specimens. It is probable that this name refers to R. leoidea (syn.: R. subfraxinea subsp. leiodea Nyl.). Specimens in CANB have been determined by us as R. leiodea, R. meridionalis, R. pacifica and R. peruviana (Table 2; Blanchon & Bannister 2002). Ramalina leiodea and R. peruviana were collected from basalt and shrubs, R. meridionalis was found only on rock and R. pacifica only on shrubs. Chatham Islands There are only a few specimens of Ramalina in CHR, HO, OTA, WELT and hb P. Johnson, with few habitat details provided. The species are R. canariensis, R. celastri, R. erumpens and R. peruviana (Table 2). Bounty Islands There are no vascular plants on these islands, the vegetation consisting of algae and crustose lichens (Wardle 1991). No species of Ramalina have been recorded (Table 2). The Snares Although lichens have been collected on the Snares as well as on the Mutton Bird Islands and Codfish Island (o# the coast of Stewart Island), once again Ramalina appears to be absent (Fineran 1969) (Table 2). Antipodes Islands Specimens in CHR, HO, MSC and O indicate that R. erumpens is the only species of Ramalina found on these islands (Table 2). It has been found growing on rock and shrubs of Coprosma spp. Auckland Islands The Auckland Islands are the type locality for R. inflata, and this is the only species of Vol. 35 Ramalina found there (Table 2). It grows epiphytically on shrubs including Dracophyllum and Myrsine spp. and on the small trees of Metrosideros umbellata. There are specimens in BM, MSC, OTA and WELT. Campbell Island Ramalina inflata (on Dracophyllum and Myrsine spp.) is the only species in MSC and OTA (Table 2). Macquarie Island The Australian checklist (McCarthy 2002) includes four species: R. banzarensis, R. farinacea, R. inflata and R. unilateralis. Specimens in ADT, BM, CANB, FH and MEL were determined as R. erumpens and R. unilateralis (Table 2). Specimens of R. erumpens had been previously determined as R. banzarensis C. W. Dodge or R. farinacea. Bannister & Blanchon (2002) have placed R. banzarensis incertae sedis, suggesting that it was impossible to determine whether the name referred to R. erumpens, R. unilateralis or possibly R. inflata, although no specimens of R. inflata have been found on the island. One packet from FH labelled R. inflata var. gracilis was found to be a specimen of Leifidium tenerum (Laurer) Wedin. Indeed it is highly unlikely that R. inflata is present on Macquarie Island as there are no suitable woody plants for this corticolous species. The specimens of R. erumpens and R. unilateralis have been collected mainly from rock, but also from timber, peat on rock and one specimen of each has been collected from the base of small perennial plants. Discussion The biota of emergent, oceanic islands becomes established by long-distance, transoceanic dispersal. New arrivals are successful only if they reach suitable substrata and microclimates, and colonization takes place in spite of the fact that the chances of successful dispersal and survival for an organism are low. The outlying, emergent islands around New Zealand (except the Bounty Islands and the Snares) are all geologically comparatively young, and the 2003 Ramalina on o#shore islands of NZ—Bannister & Blanchon T 3. Types of propagule produced by species of Ramalina found on the islands Species australiensis canariensis celastri erumpens exiguella geniculata inflata leoidea luciae meridionalis cf. microspora pacifica peruviana stevensiae unilateralis Ascospores Soredia ++ – +++ – +++ +++ +++ +++ + ++ +++ + ++ +++ + – +++ – +++ – – – – +++ – – +++ +++ – +++ –, propagule not known; +, propagule found only rarely; ++, some thalli do not produce the propagule; +++, typically all specimens produce large numbers of propagules. species of Ramalina found on the islands have arrived presumably from Australia, Tasmania, New Zealand or other older islands by transoceanic dispersal. Lichens can be dispersed by thallus fragments, and some species of Ramalina produce short, fragile branches or long, fine branches that may break from the parent thallus and be dispersed. Büdel & Scheidegger (1996) considered that long, Usnea-like Ramalina thalli are torn and dispersed by strong winds. Species of this type include R. australiensis and R. meridionalis, but it is not known whether these fragments could be dispersed over long distances. Most species of Ramalina produce either ascospores or soredia, although both propagules may be present in some species (Table 3 ). Ascospores are ellipsoidal, straight or slightly curved, one-septate, colourless and thin-walled (Blanchon et al. 1996). Lichen ascospores are liberated when the contents are ejected from a mature ascus which bursts apically after a build-up of hydrostatic pressure (Ingold 1965). In Ramalina the contents are discharged as packets of six to eight ascospores, for a distance of 4–10 mm (personal observation). Ascospores are shot into turbulent 143 air above the 1–2 mm deep boundary layer of still air that covers the surface of the apothecia, thus allowing dispersal by air currents. Soredia are liberated passively from soralia by water, abrasion by small invertebrates or birds, but wind provides more successful long-distance dispersal by lifting the soredia from dry soralia into turbulent air (Bailey 1976). Successful dispersal can occur only if propagules remain viable during transportation, and ascospores are considered to be resistant to long-distance aerial transport (Galloway & Aptroot 1995). Desiccation and sunlight are thought to harm some ascospores, although thick-walled and pigmented ascospores are thought to have a better chance of survival (Pedgley 1982). Ramalina ascospores are both thin-walled and hyaline (Blanchon et al. 1996), but as the ascospores are liberated only when the apothecia are wet, they may be dispersed in cloudy conditions or rainy weather, and this may allow for a greater rate of survival. Nine species producing only ascospores occur on the islands (Table 3). Following dispersal, an ascospore must germinate on a suitable substratum where it must resynthesize the lichen thallus with an appropriate photobiont. In Ramalina, the photobiont is a species of Trebouxia (Purvis & James 1992), and although it is rarely found outside lichens and is thought to be a poor competitor (Honegger 2001), there is evidence that Trebouxia exists in the free-living state (Galun 1988; Sanders & Lücking 2002). Sanders & Lücking (2002) present evidence to show early stages of lichenization in foliicolous lichens where ascospores have reassociated with algal symbionts after germination. The abundance of species lacking vegetative diaspores indicates that lichenization must be a common process (Galun 1988). Murtagh et al. (2000) found evidence of self-fertilization in lichens, indicating that a population could develop from a single dispersed ascospore. Following dispersal, soredia do not have the problem of lichenization as the photobiont is already enclosed in the fungal hyphae, but they still have to establish a thallus on a 144 THE LICHENOLOGIST suitable substratum. If dispersal by thallus fragments occurs, the fragments must develop a holdfast for attachment prior to the lichen thallus becoming re-established. It is not known whether any species of Ramalina are able to re-attach a fragment to a substratum. However, we have observed that thalline branches of the Hawaiian species R. microspora and the Australasian species R. australiensis and R. inflata can form new holdfasts. Various methods of transoceanic dispersal of lichen propagules have been suggested, including animals, flotation or rafting and air currents. Bailey (1976) considered that wind was the most likely agent for long-distance dispersal, and Louwho# (2001) has suggested that winddispersal is more influential than the combined e#ects of bird-dispersal and rafting for Pacific Parmeliaceae. There is little direct evidence for transoceanic dispersal by animals. Small invertebrates are often associated with lichens, and some feed on ascospores, fungal hyphae or algal cells. The possibility of long-distance dispersal of propagules arises when the invertebrates are themselves dispersed. Tardigrades, in an anabiotic condition, have been reported as being wind-dispersed together with small lichen propagules (Gerson & Seaward 1977). Many mites feed exclusively or occasionally on lichens, and they occur on subantarctic islands, having previously been dispersed from island to island or from the southern tips of landmasses (Gressitt 1967). Several species of mites found on Macquarie Island also occur in New Zealand (Sømme 1985). It has often been suggested that birds contribute to long-distance dispersal by carrying lichen propagules on their feathers or feet. One, often quoted, example is of a Royal Albatross on the Auckland Islands, that had soredia and thallus fragments washed from its feet after it was forced to run through scrub (Bailey & James 1979). This information presumably led Purvis (2000) to suggest that birds such as the Royal Albatross may transport lichen propagules over very long distances. However, Vol. 35 dispersal in this way is highly unlikely as albatrosses spend the time between breeding seasons in the air or on the sea surface; they do not touch land until they return to the same island breeding colony (McClelland 1999). Smith (1995), in a study of Hawaiian lichens, also considered it unlikely that birds were responsible for the dispersal of lichens to the islands. Marine currents are unlikely to disperse lichen fragments or thalli of species of Ramalina. Experimental work on immersion of coastal lichen thalli in sea water showed that while certain thalli could withstand immersion for several days, thalli of R. siliquosa (Huds.) A. L. Sm., a xeric-supralittoral species, do not survive immersion (Fletcher 1976). Both R. australiensis and R. meridionalis can grow in the splash zone and might be transported by seawater if their thalli could survive immersion. In fact, the occurrence of R. meridionalis on rocky islands and peninsulas in northern New Zealand (Blanchon & Bannister 2002; P. Bannister, J. M. Bannister & D. J. Blanchon, unpublished.) suggests this might occur. Although ascospores of marine ascomycetes survive in sea water, there appear to be no published records indicating that ascospores of lichenized ascomycetes can survive immersion. Most lichen propagules are probably transported by wind as ascospores and soredia are small and light and are thus wellsuited for long-distance dispersal by air currents (Bailey 1976; Wedin 1995). The length of Ramalina ascospores varies between 10 and 20 µm (Blanchon et al. 1996) and soredia are roughly spherical and 25– 50 µm in diameter (personal observation). These dimensions are similar to those of small (10–25 µm) and medium-sized pollen grains (25–50 µm) (Moar 1993). Air sampling traps have recorded cryptogam spores, pollen grains, mites and wingless insects over land and oceans including those at polar latitudes and up to an altitude of 3–6 km (Pedgley 1982). Spores of rust fungi and aphids are recorded as having crossed the Tasman Sea in air currents from Australia, and these organisms have become 2003 Ramalina on o#shore islands of NZ—Bannister & Blanchon established in New Zealand (Close et al. 1978). Evidence of wind-dispersal from Australia and New Zealand comes from modern pollen rain studies on Campbell Island (McGlone & Meurk 2000), the Antipodes Islands (Moar 1969) and the Chatham Islands (Dodson 1976). Aphids on Campbell Island are thought to have dispersed from New Zealand (Close et al. 1978). For successful dispersal to the islands, propagules need be airborne for only a few days as some winds can carry airborne organisms 1000 km or more in a day (Pedgley 1982). Thus, trans-Tasman crossings might take only 2–3 days (Close et al. 1978). It is not surprising that current distributions of Ramalina species on these islands appear to confirm that dispersal of airborne organisms mainly takes place in the direction of the prevailing winds. Norfolk Island and Lord Howe Island have winds from the direction of Australia in winter, and from New Zealand in summer, and both have mixtures of New Zealand and Australian Ramalina species. The Kermadec Islands have winds from the direction of New Zealand in the winter and from the Pacific (the north-east) in summer, with six species of Ramalina in common with New Zealand and two additional Pacific species. The Chatham Islands receive winds from southern New Zealand and the Subantarctic, and have species in common with both. The Auckland Islands and Campbell Island receive westerly and north-westerly winds and have one species, which is also found in Australia, to the north-west. However, the direction of plant dispersal to the subantarctic islands only partly conforms to the prevailing winds. For example, the flora of Macquarie Island is derived largely from that of New Zealand despite the infrequency of favourable winds (Wardle 1978). Moreover R. erumpens occurs on Macquarie Island, but it does not occur in Tasmania or mainland Australia. It presumably has dispersed from New Zealand where it is confined to the south-eastern part of the South Island (P. Bannister, J. M. Bannister & D. J. Blanchon, unpublished). This 145 species has also reached the Chatham Islands and the Antipodes Islands. In conclusion, we suggest that the species of Ramalina occurring on the outlying islands of the New Zealand geographic area have arrived by transoceanic transport, winddispersal of either ascospores or soredia, not necessarily in the direction of the prevailing winds, and that species have only become established on the islands if both climatic and habitat conditions have been met. We thank Peter Johnson and the curators of the herbaria listed for loans of specimens and for providing other information, Daphne Lee for discussing the geological ages with J.M.B., Peter Bannister, David Galloway, Mel Galbraith and Mark Large for constructive criticism of the draft manuscript, NIWA for climatic information, Matt McGlone for use of the map in Fig. 1 and two anonymous referees for helpful comments. R Archer, A. W. & Elix, J. A. (1994) The lichens of Lord Howe Island. 1. Introduction and the genus Pertusaria (Pertusariaceae). Telopea 6: 9–30. Atkinson, I. (1966) Major habitats. In The Chatham Islands (I. Atkinson et al., eds): 49–61. Christchurch: Canterbury University Press. Bailey, R. H. (1976) Ecological aspects of dispersal and establishment in lichens. In Lichenology: Progress and Problems (D. H. Brown, D. L. Hawksworth & R. H. Bailey, eds): 215–247. London: Academic Press. Bailey, R. H. & James, P. W. (1979) Birds and the dispersal of lichen propagules. Lichenologist 11: 105. Bannister, J. & Blanchon, D. J. (2002) The typification and status of the name of Ramalina banzarensis Dodge. Australasian Lichenology 51: 14–16. Blanchon, D. J. & Bannister, J. (2002) Ramalina meridionalis, a new species from New Zealand, Norfolk Island and Lord Howe Island. Australasian Lichenology 51: 17–19. Blanchon, D. J., Braggins, J. E. & Stewart, A. (1996) The lichen genus Ramalina in New Zealand. Journal of the Hattori Botanical Laboratory 79: 43–98. Büdel & Scheidegger, C. (1996) Thallus morphology and anatomy. In Lichen Biology (T. H. Nash III, ed.): 37–64. Cambridge: Cambridge University Press. Close, R. C., Moar, N. T., Tomlinson, A. I. & Lowe, A. D. (1978) Aerial dispersal of biological material from Australia to New Zealand. International Journal of Biometeorology 22: 1–19. De Lisle, J. F. (1965) The climate of the Auckland Islands, Campbell Island and Macquarie Island. Proceedings of the New Zealand Ecological Society 12: 37–44. 146 THE LICHENOLOGIST Dodson, J. R. (1976) Modern pollen spectra from Chatham Island, New Zealand. New Zealand Journal of Botany 14: 341–347. Elix, J. A., Din, L. B. & Samsudin, M. W. B. (1991) New species of Ramalina (lichenized Ascomycotina) from Australasia and Malaysia. Mycotaxon 40: 41–44. Elix, J. A. & McCarthy, P. M. (1998) Catalogue of the lichens of the smaller Pacific islands. Bibliotheca Lichenologica 70: 1–361. Elix, J. A. & Streimann, H. (1989) The lichens of Norfolk Island. 1: Introduction and the family Parmeliaceae. Proceedings of the Linnaean Society of New South Wales 111: 103–121. Elix, J. A., Streimann, H. & Archer, A. W. (1992) The lichens of Norfolk Island. 2: The genera Cladonia, Pertusaria, Pseudocyphellaria and Ramalina. Proceedings of the Linnaean Society of New South Wales 113: 57–76. Fineran, B. A. (1969) The flora of the Snares Islands, New Zealand. Transactions of the Royal Society of New Zealand, Botany 3: 237–270. Fletcher, A. (1976) Nutritional aspects of marine and maritime lichen ecology. In Lichenology: Progress and Problems (D. H. Brown, D. L. Hawksworth & R. H. Bailey, eds): 359–384. London: Academic Press. Galloway, D. J. & Aptroot, A. (1995) Bipolar lichens: a review. Cryptogamic Botany 5: 184–191. Galun, M. (1988) Lichenization. In Handbook of Lichenology, Vol. 3 (M. Galun, ed.): 153–169. Boca Raton: CRC Press. Gerson, U. & Seaward, M. R. D. (1977) Licheninvertebrate asociations. In Lichen Ecology (M. R. D. Seaward, ed.): 69–119. London: Academic Press. Gressitt, J. L. (1967) Introduction. In Entomology in Antarctica (J. L. Gressitt, ed.): 1–27. Washington: American Geophysical Union. Hnatiuk, R. J. (1993) Subantarctic Islands. Flora of Australia 50: 53–62. Honegger, R. (2001) The symbiotic phenotype of lichen-forming ascomycetes. In The Mycota IX, Fungal Associations (R. Hock, ed.): 165–188. Berlin: Springer Verlag. Ingold, C. T. (1965) Spore Liberation. Oxford: Clarendon Press. Kantvilas, G. & Seppelt, R. D. (1992) The lichen flora of Macquarie Island: introduction and an annotated checklist of species. ANARE Research Notes 87: 1–20. Louwho#, S. H. J. J. (2001) Biogeography of Hypotrachyna, Parmotrema and allied genera (Parmeliaceae) in the Pacific islands. Bibliotheca Lichenologica 78: 223–246. McCarthy, P. M. (2002) Checklist of Australian Lichens. Canberra: Australian Biological Resources Study, http://www.anbg.gov.au/abrs/lichenlist. McClelland, P. (1999) Birds. In New Zealand’s Subantarctic Islands (T. O’Connor, ed.): 58–72. Auckland: Reed Books. McGlone, M. S. & Meurk, C. D. (2000) Modern pollen rain, subantarctic Campbell Island, New Zealand. New Zealand Journal of Ecology 24: 181–194. Vol. 35 Moar, N. T. (1969) Pollen analysis of a surface sample from Antipodes Island. New Zealand Journal of Botany 7: 419–423. Moar, N. T. (1993) Pollen Grains of New Zealand Dicotyledonous Plants. Lincoln: Manaaki Whenua Press. Murtagh, G. J., Dyer, P. S. & Crittenden, P. D. (2000) Sex and the single lichen. Nature 404: 564. O’Connor, T. (ed.) (1999) New Zealand’s Subantarctic Islands. Auckland: Reed Books. Orchard, A. E. (ed.) (1994) Flora of Australia. Oceanic Islands 1 Vol. 49. Canberra: Australian Government Publishing Service. Pedgley, D. (1982) Windborne Pests and Diseases. Chichester: Ellis Horwood Ltd. Purvis, O. W. & James, P. W. (1992) Ramalina Ach. (1910). In The Lichen Flora of Great Britain and Ireland (O. W. Purvis, B. J. Coppins, D. L. Hawksworth, P. W. James & D. M. Moore, eds): 524–529. London: Natural History Museum Publications. Purvis, W. (2000) Lichens. London: The Natural History Museum. Sanders, W. B. & Lücking, R. (2002) Reproductive strategies, relichenization and thallus development observed in situ in leaf-dwelling lichen communities. New Phytologist 155: 425–435. Selkirk, P. M., Seppelt, R. D. & Selkirk, D. R. (1990) Subantarctic Macquarie Island: Environment and Biology. Cambridge: Cambridge University Press. Smith, C. W. (1995) Notes on long-distance dispersal in Hawaiian lichens: ascospore characters. Cryptogamic Botany 5: 209–213. Sømme, L. (1985) Terrestrial habitats—invertebrates. In Antarctica (W. N. Bonner & D. W. H. Walton, eds): 106–117. Oxford: Pergamon Press. Stevens, G. N. (1987) The lichen genus Ramalina in Australia. Bulletin of the British Museum (Natural History), Botany Series 16: 107–223. Sykes, W. R., West, C. J., Beever, J. E. & Fife, A. J. (2000) Kermadec Islands Flora, Special edn. Lincoln: Manaaki Whenua Press. Turnbull, I. (1999) Geology. In New Zealand’s Subantarctic Islands (T. O’Connor, ed.): 25–34. Auckland: Reed Books. Wardle, P. (1978) Origin of the New Zealand mountain flora, with special reference to trans-Tasman relationships. New Zealand Journal of Botany 16: 535–550. Wardle, P. (1991) Vegetation of New Zealand. Cambridge: Cambridge University Press. Wedin, M. (1995) The lichen family Sphaerophoraceae (Calicales, Ascomycotina) in temperate areas of the Southern Hemisphere. Symbolae Botanicae Upsalienses 31(1): 1–102. West, C. J. & Rance, B. (1999) Vegetation. In New Zealand’s Subantarctic Islands (T. O’Connor, ed.): 25–34. Auckland: Reed Books. Accepted for publication 12 February 2003