New Zealand Journal of Botany ISSN: 0028-825X (Print) 1175-8643 (Online) Journal homepage: http://www.tandfonline.com/loi/tnzb20 Distribution, habitat, and relation to climatic factors of the lichen genus Ramalina in New Zealand Peter Bannister , Jennifer M. Bannister & Daniel J. Blanchon To cite this article: Peter Bannister , Jennifer M. Bannister & Daniel J. Blanchon (2004) Distribution, habitat, and relation to climatic factors of the lichen genus Ramalina in New Zealand, New Zealand Journal of Botany, 42:1, 121-138, DOI: 10.1080/0028825X.2004.9512894 To link to this article: http://dx.doi.org/10.1080/0028825X.2004.9512894 Published online: 17 Mar 2010. Submit your article to this journal Article views: 204 View related articles Citing articles: 6 View citing articles Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tnzb20 Download by: [122.57.53.94] Date: 04 July 2017, At: 01:42 New Zealand Journal of Botany, 2004, Vol. 42: 121-138 0028-825X/04/4201-0121 © The Royal Society of New Zealand 2004 121 Distribution, habitat, and relation to climatic factors of the lichen genus Ramalina in New Zealand PETER BANNISTER JENNIFER M. BANNISTER Department of Botany University of Otago P.O. Box 56 Dunedin, New Zealand DANIEL J. BLANCHON Resource Management Research Group School of Landscape and Plant Science UNITEC Private Bag 92025 Auckland, New Zealand Abstract We present maps, based on herbarium records extensively supplemented by our own records, showing the distribution of the 14 species of Ramalina found in New Zealand (excluding the Kermadec, Chatham, and sub-antarctic islands). Species diversity is highest in two areas, eastern Northland (seven species) and coastal Otago (eight species). Only four species (R. canariensis, R. celastri, R. glaucescens, and R. unilateralis) are common to both main islands. Distribution is related to climate and altitude along a latitudinal gradient on the eastern side of New Zealand, from cooler, drier sites and higher altitudes in the south to warmer, moister sites and lower altitudes in the north. Species of Ramalina tend to be absent from sites with various combinations of high rainfall, high altitude, high water deficits, and extremes of temperature. Species reproducing asexually and sexually show similar relationships of frequency to spread, but with sexually reproducing species showing greater frequency at a given degree of spread. Keywords Ramalina; Ramalina arabum; Ramalina australiensis; Ramalina canariensis; Ramalina celastri; Ramalina erumpens; Ramalina B02091; Online publication date 30 March 2004 Received 24 December 2002; accepted 23 June 2003 fimbriata; Ramalina exiguella; Ramalina geniculata; Ramalina glaucescens; Ramalina inflexa; Ramalina meridionalis; Ramalina pacifica; Ramalina peruviana; Ramalina riparia; Ramalina unilateralis; distribution; habitat; climate; New Zealand; lichens INTRODUCTION Ramalina Ach. is a cosmopolitan genus of fruticose lichens found in locations from beyond the Arctic Circle to the Antarctic Peninsula and encompassing a range of climates from polar to tropical. Kirk et al. (2001) listed c. 200 species of Ramalina world-wide. Fourteen species of Ramalina occur in New Zealand (Blanchon et al. 1996). Species of Ramalina occurring on the outlying islands of the New Zealand geographic area are not included in this paper but are documented by Bannister & Blanchon (2003). R. arabum (Dill. ex. Ach.) Meyen & Flotow is no longer considered to occur in New Zealand and specimens described under this name constitute a new species, R. meridionalis D.Blanchon & J.Bannister (Blanchon & Bannister 2002; see also Müller Argoviensis 1894; Hellborn 1896; Zahlbruckner 1941; Martin 1966, 1968; Galloway 1985; Blanchon et al. 1996). The New Zealand species of Ramalina can be divided into three groups: those most prevalent in the warmer temperate regions of northern New Zealand (latitude <38°S), those most prevalent in the cooler temperate regions of southern New Zealand (latitude > 44°S), and those occurring in both North and South Islands. The northern group comprises R. meridionalis, R. australiensis Nyl., R. pacifica Asahina, R. peruviana Ach., R. exiguella Stirt., and R. geniculata Hook.f. & Taylor. The southern group consists ofR. erumpens D.Blanchon, J.Braggins & A.Stewart, R. fimbriata Krog & Swinscow, R. inflexa D.Blanchon, J.Braggins & A.Stewart, andR. riparia D.Blanchon, J.Braggins & A.Stewart. Four species occur in both North Island and South Island: R. canariensis J.Steiner andR. celastri (Spreng.) Krog & Swinscow are more frequent in North Island while 122 R. glaucescens Kremp. andR. unilateralis F.Wilson are more frequent in South Island. Six of the 14 species (R. celastri, R. exiguella, R.glaucescens, R. inflexa, R. geniculata, and R. exiguella) are consistently apothecial. R. meridionalis andR. australiensis are largely vegetative and probably spread via thallus fragmentation (Bannister & Blanchon 2003), but occasionally produce apothecia. The remaining species produce soredia; but apothecia occur occasionally on R. peruviana and rarely on R. unilateralis and R. fimbriata, but have not been observed on R. canariensis, R. erumpens, or R. pacifica in New Zealand. Lichens that spread by asexual means are generally considered to be more frequent and widespread than those that produce only sexual spores (e.g., Bowler & Rundel 1975; Louwhoff 2001). The first maps based on herbarium records (Blanchon et al. 1996) have been expanded by our own records from the field and additional records from herbarium specimens not included in earlier maps. Since September 1997, we have visited localities all over New Zealand targeting under-recorded areas and species. Our main objective was to produce more complete maps of the distribution of the New Zealand species of Ramalina and use these to relate their distribution to climate using the climate response surfaces developed by Leathwick & Stephens (1998). A secondary objective is to use the information on frequency and location to examine whether those species reproducing asexually are more frequent and more widespread than those producing sexual spores. MATERIALS AND METHODS Blanchon et al. (1996) produced maps that relied largely on herbarium and other records collated by Blanchon (1994). Examination of these maps showed that there were areas of the country where species of Ramalina were either absent or unrecorded. Ongoing investigations began in September 1997 in the Dunedin area and Otago and were followed by specific visits to more distant areas in subsequent years. These were Stewart Island and western Fiordland, Jan 1998; Bay of Islands, Jun 1998; Taranaki-Manawatu, Nov 1998; Marlborough-Nelson and West Coast, Dec 1998; Southland, Apr 1999; Coromandel, Aug 1999; Wellington to East Cape, Nov-Dec 1999; Te Anau, Apr 2000; western and central Northland, Jun-Jul 2000; central North Island, Dec 2000; Southland, New Zealand Journal of Botany, 2004, Vol. 42 Apr 2001; Banks Peninsula and Mid Canterbury, July 2001; North Canterbury, Aug 2001; Marlborough, Apr 2002; Great Barrier Island, Dec 2002. Additional records were obtained by examining specimens from herbaria at Auckland (AK), Wellington (WELT), Lincoln (CHR), Canterbury University (CANU), recent additions at Otago (OTA), and the private collection of P. N. Johnson. Additional specimens collected by individuals (see Acknowledgments) were verified and included in the mapping. In total, 1535 separate sites were mapped, 609 from herbarium records and 926 from our own records (Fig. 1). Grid references (New Zealand Map Grid, NZMG) and altitude were determined for each site where our records were made, and from herbarium records that gave specific map references and altitudes. Landcare Research (Hamilton, New Zealand) provided estimates of climatic data for 1039 sites where species of Ramalina were found and 34 sites where no species of Ramalina were detected. Climatic variables were estimated from geographic location and elevation of each site using thin-plate splines fitted to average monthly climate data from meteorological stations (Leathwick & Stephens 1998; Leathwick 2001). Means and standard deviations for climatic variables were calculated for the sites in which a particular species of Ramalina was present, and species means provided the data for a principal components analysis using a centred correlation matrix (Kovach 1993). Ranges of climatic variables were subdivided into groups that were scaled either logarithmically (altitude) or linearly (temperature, precipitation, precipitation/evaporation ratios, solar radiation). Groups at the extremes of each range, where there were few records, were amalgamated. Percentage species frequency within each group was calculated from the ratio of the number of sites containing the species in question to the total number of sites within a particular group. Histograms showing species frequency with respect to annual means of minimum temperature and precipitation/evaporation (p/e) ratios, and altitude are included with graphs showing the geographic distribution of species. Species richness was determined by allocating records to 10-km squares on the New Zealand Map Grid (NZMG) and recording the number of species per square. Overall frequencies for a particular species were determined as the percentage of the recorded sites containing that species. Spread of species was measured as the distance (in km) across the longest axis of each species distribution, calculated Bannister et al.—Ramalina in NZ Fig. 1 Sites visited by the authors and associated collectors (see Acknowledgments) (O), and sites of herbarium specimens (Ø) with verified identifications of species of Ramalina (Blanchon et al. 1996, and subsequently). 123 Sites visited Herbarium records from the grid co-ordinates of the two most distant sites. The linear relationship between the logarithm of frequency and spread was used to compare the responses of sexually and asexually reproducing species. North and South Island records of R. canariensis are widely separated and are treated as two distinct groups for the purposes of analysis. RESULTS AND DISCUSSION Recording and mapping Maps of species distribution must combine accurate species identification and location with adequate geographic coverage. The identity of herbarium specimens must be confirmed by inspection as not all herbarium specimens are correctly named. The revision of New Zealand species of Ramalina by Blanchon et al. (1996) enabled accurate identification of all the species of Ramalina found in New Zealand: previous herbarium records were checked and renamed where necessary (Blanchon 1994), as were herbarium specimens that came to light subsequently. There were obvious gaps in the maps of Ramalina distribution produced from herbarium and other records collated by Blanchon et al. (1996), and there are still evident gaps in our more recent maps (Fig. 1). Gaps may be due to absence of a species or lack of collection. Species may be overlooked or not collected because they are too common, rare, or small, or because they grow in rarely visited or inaccessible areas. By targeting these gaps, the range of many species has been extended. Similar gaps in the distribution of vascular plants (e.g., western Taranaki, East Cape, inland Canterbury) were noted by Raven & Engelhorn (1971). In order to cover New Zealand, our recording was necessarily restricted to areas that could be readily accessed by roads (or, in western Fiordland and southern Stewart Island, by boat). Consequently, some mountainous or remote areas (e.g., the Kaikoura Ranges, regions to the west of the Southern Alps) remain under-recorded, although areas at higher altitudes (>1000 m) or with high precipitation/evaporation ratios rarely support species of Ramalina. The mode of recording also affects distribution maps. Individual herbarium records usually consist of single species collected from a particular site, whereas we searched for all species of Ramalina at any site that we visited. When records were grouped by site, only 27% of herbarium records recorded more than one spe- New Zealand Journal of Botany, 2004, Vol. 42 124 "a * .9 L, 2• - 5 1| o r-M3 h H ^ in ID ID 00 ^H rs en en ^H vo o\ ID •* rs r s o c ^ ^ o t — o\ ^H o\ H H rH en t— t— rs O rS en ^H O O "Ñf rs en r~- r—< t~^ r~- Q ^ "Ñf ^O 00 0 0 <^ r~- "Ñt" ID ^ D ^ D r—< ID ^í" ^í" ^í" 'H" ID *sf 'sf l/^ 'sf 'sf *sf 'sf 'sf l/^ l/^ '/"} l/^ l/^ l/^ l/^ o CO S' • s ID S ID ID 47 47 en 46 49 43 o o o o o o o o o o o o o o en rs rs rs rs rs rs rs rs rs rs rs rs rs rs •3 H 1272 (390; (568; (382; (159; (187; (185; (186; (394; (228; (193; (i8i; (103; (217; (28Œ j II m cN oo ^o in r- cN i ^o r- in -Ñf co -Ñf ^o o ID en en en CoGîî ! ^o i il en vq ^-i ts rs_ rs_ Oí p o\ p rs_ VD I D en 3\ ö o\ oô r-' r-' oô o\ r-' r-' oô r-' r-' r-' rS oí M S •a g lu 1• | Í 2-S 2 .S-S S Ö -'S ^H O » es ' 0 O\ <~~ ta lu o! o! d oo cK I rS I• s s §11 o cies ofRamalina per site, whereas 49% of our records recorded more than one species. Species distribution and climate Annual means of climatic variables for each species of Ramalina and for sites where no species ofRamalina were found are presented in Table 1 and form the basis of a principal components analysis (PCA) that summarises the predominant trends of species' distribution with climate (Fig. 2). The first (horizontal) axis accounts for 51% of the variation in means, and the second (vertical) axis accounts for a further 27% of the overall variation. Latitude was deliberately excluded as a factor in the PCA, as it is correlated with most of the other factors, but it is also significantly associated with the first axis of the PCA (r = 0.69, P < 0.01). Consequently, the first axis has a latitudinal component, from low to high latitudes (i.e., from north to south), and is associated with increasing altitude and temperature range, decreasing solar radiation and water deficit, and cooler temperatures. As a result, species that are widespread throughout New Zealand would be expected to have average values for climatic factors and be located near the centre of the PCA (e.g., R. celastri). However, the relationship with latitude is weaker for the southern species (R. inflexa, R. riparia, R. erumpens, and R. fimbriata) and for the other species found in both islands (R. glaucescens, R. unilateralis) but with their main centres of distribution in South Island. These species are more strongly influenced by altitude, temperature range, and water deficit. •S S o 'S01 S g ^ > "l'a-a ill r-oorsoooo^or-rs^^orsiDr-rrS^OiD^oiDrSOOrSeniD^OiD g 'C « « g « .« -Si ^ "S -S ^ 8 - 2 lll'i Ml ¡iliSlî 8. § c¿ c¿ ce ce ce ce ce ce ce ce ce ce ce ce The climatic means for sites without Ramalina (Table 1) are distinctly different and strongly related to the second axis of the ordination. They are associated with high precipitation and p/e ratios, lack of water deficits, extremes of temperature, and higher altitudes (Fig. 2). However, climatic data were obtained for only a few sites (34) where species ofRamalina were apparently absent and coverage was by no means comprehensive, and did not include dry areas in the east of North and South Bannister et al.—Ramalina in NZ Fig. 2 Principal component analysis of species means for maximum (mxt), average (avt), and minimum (mnt) temperature, temperature range (tra), altitude (alt), precipitation/evaporation ratio (p/e), annual water deficit (def), precipitation (ppt), and solar radiation (slr) as listed in Table 1. Vectors (arrows) indicate the relation of these variables to the species means. Species are listed in latitudinal order. Open circles indicate species of Ramalina found only in North Island (R. exiguella (exi), R. meridionalis (mer), R. peruviana (per), R. australiensis (aus), R. geniculata (gen), R. pacifica (pac)); crossed circles, species (and sites without Ramalina) occurring in both North and South Islands (R. canariensis (can), R. celastri (cel), no Ramalina (nil), R. glaucescens (gla), R. unilateralis (uni)); solid circles indicate species found only in South Island (R. inflexa (inf), R. fimbriata (fim), R. erumpens (eru), R. riparia (rip)). 125 PPt I p/e © nil slr def Islands (see Fig. 11). Our experience in the field suggested that species of Ramalina tended to be absent not only in areas with high precipitation or p/e ratios, at high altitudes, or with extremes of temperature (as indicated in Fig. 2), but also within closed forest and dry pastoral or agricultural areas. Northern species Ramalina australiensis, R. exiguella, R. geniculata, R. meridionalis, R. pacifica, and R. peruviana are found in North Island and adjacent islands (Fig. 3 5). These species are typically found at low altitudes and are often coastal in their distribution. Consequently they experience warm and equable temperatures, relatively high rainfall, and potentially high evaporation rates (Table 1). The distribution ofR. australiensis overlaps with that of R. meridionalis (Fig. 3) but it is more widespread in coastal and mainland locations. We found new locations on the Coromandel Peninsula and in eastern Bay of Plenty, on coastal rocks from Awanui to Lottin Point. The disjunct record in Wellington Harbour may be associated with the presence of hard coastal rocks. Herbarium records from inland sites at Mangatea Stream and Feilding (Merry Hill) appear to be associated with forest remnants. We visited the latter site but were unable to relocate the species (Bannister 2001). Ramalina meridionalis is restricted to offshore islands and mainland points and peninsulas in the north-east of North Island (Fig. 3). We have not added any new records. Herbarium records indicated that R. exiguella (Fig. 4) has a similar but more restricted distribution than R. meridionalis, but we found new locations on coastal mangrove and kauri on the mainland and made the first record for this species on Great Barrier Island (on mangroves in Whangaparapara Bay). Ramalina geniculata, R. pacifica, and R. peruviana are relatively common in the north of North Island but are also found further south, particularly in forest remnants. We found new sites for R. geniculata (Fig. 4) on the Coromandel Peninsula and in the Hunua Range to the west of Firth of Thames and extended its range to forested sites in eastern Taranaki and Wanganui-Manawatu. R. pacifica (Fig. 5) is less frequently found than R. geniculata but shows a similar distribution. We made the first R. meridionalis R. australiensis Mean minimum temperature (<3to>11°C ) Mean minimum temperature (<3to>11°C ) 3 - 2; 81 o 81 Altitude (log scale <4 to >1000 m) Altitude (log scale <4 to >1000 m) Fig. 3 Distribution of site records for Ramalina australiensis and R. meridionalis. Circles indicate sites visited by the authors and associated collectors, squares indicate herbarium records. Histograms relate species frequency to mean minimum temperature, precipitation/evaporation ratios and altitude. The scale for minimum temperature ranges from <3°C to >11°C in degree increments, that for precipitation/evaporation ratio ranges from <1 to 4 in increments of 0.5, with the last two bars indicating p/e ratios of 4-6 and >6, while the scale for altitude is logarithmic (<4 m, 4—10 m, 11-30 m, 31-100 m, 101-310 m, 311-1000 m and >1000 m). W o Bannister et al.—Ramalina in NZ record for this species on Great Barrier Island and new records in the Central North Island with records from forested sites in eastern Taranaki, WanganuiManawatu, and the Volcanic Plateau, thus linking northern sites with the otherwise isolated southernmost site on Mokopuna Island in Wellington Harbour. R. peruviana (Fig. 5) is most frequent in the northern part of North Island. Our records extend its known distribution in the northern parts of North Island and in the southern parts of its range (Hamilton area, Coromandel Peninsula, and the Hunua Ranges), and made the first records for this species on Great Barrier Island. Herbarium specimens record R. peruviana on islands in the Marlborough Sounds and the Chatham Islands (Bannister 1998) indicating that this species extends into cooler and drier areas than R. geniculata and R. pacifica. A specimen of R. peruviana at CHR, ostensibly collected by J. S. Thomson from a montane site in the Silverpeaks, Otago, is not in its original packet and the duplicate held in OTA contains only a specimen of a species of Usnea. Consequently, this unlikely record has been rejected. Species occurring in North and South Islands Ramalina canariensis, R. celastri, R. glaucescens, andR. unilateralis (Fig. 6,7) are found in both North and South Islands. R. canariensis andR. celastri are more frequent in North Island, whereas R. glaucescens and R. unilateralis are more common in South Island. The climate of North Island is generally warmer, wetter, and sunnier than that of South Island. Species with a distribution in both North and South Islands tend to occupy appropriate habitats. For example, R. glaucescens andR. unilateralis are found at higher altitudes that provide a cooler climate in North Island (median altitudes of 490 and 580 m, respectively, in contrast to 200 and 160 m in South Island). Conversely, R. celastri is less frequent in cooler upland areas and has a lower altitudinal limit (590 m) in southern South Island than in northern South Island (790 m) or North Island (950 m). Ramalina canariensis was previously described as subtropical and maritime in distribution (Blanchon et al. 1996); an inland specimen from Barryville (485 m) was considered to be aberrant. We have found R. canariensis at altitudes up to 260 m in altitude in inland forest sites from Northland to eastern Taranaki and WanganuiManawatu and also in coastal sites at low altitudes (<60 m) to the north and south of Otago Harbour (Fig. 6), where lack of severe frosts permits the 127 growth and naturalisation of plants from warmer climates, including the Canary Islands and Madeira. Ramalina celastri is a widely distributed species (Fig. 6). In South Island, it is absent or infrequent in dry and upland regions away from the east coast and in the wetter southern and western areas of South Island. We failed to find R. celastri in western Fiordland, from Puysegur Point to Milford Sound, and it is infrequent elsewhere on the west coast of South Island, where it is the only species of Ramalina recorded. As a common species, it has apparently often been ignored and not recorded or collected (see Raven & Engelhorn 1971). For example, there were no herbarium records for R. celastri in western Taranaki and inland areas of East Cape in North Island where we found it to be the only species of Ramalina present. It is most frequent at lower altitudes with warm temperatures and moderate to high rainfall, and less frequent in drier and colder areas, and in sites with extremely high rainfall or p/e ratios. Ramalina glaucescens (Fig. 7) is common and widespread in the eastern part of South Island, including upland and dry inland areas where R. celastri is rare or absent. It apparently does not occur west of the Main Divide of the Southern Alps. There were only a few herbarium records from North Island, mostly from the Manawatu region with an outpost in Gisborne (Blanchon et al. 1996). Our new records link these scattered herbarium records and show that R. glaucescens occurs throughout the central and eastern parts of North Island. Although R. glaucescens often co-occurs with R. celastri, it shows marked differences in its habitat preferences and is often the only species of Ramalina to be found at higher altitudes, in dry areas, and in sites with high seasonal variation in temperature. Ramalina unilateralis (Fig. 7) was previously considered to be widespread but rare, and most common on rocks in exposed upland and subalpine areas. We have found it to occur most frequently on trees and shrubs at lower elevations (particularly in shrubland or on forest edges), where it may have been overlooked and possibly confused with species of Usnea, with which it often co-occurs. It is relatively common in moister areas of Otago and Southland, in coastal regions, and on the eastern foothills of the Southern Alps and other mountain ranges, particularly where dry lowlands abut on moister uplands. We have extended its known distribution in the north of South Island, in the southern and central parts of North Island, and eastwards R. exiguella ü CD O" CD ü 12 10 - CD 8 - 6 4 2 0 Mean minimum temperature (<3to>11°C ) >, o CD =3 O" CD R. geniculata 30 -, 14 -i Mean minimum temperature (<3to>11°C ) 3 -i ü CD 2 - 10 8 - er CD 1 0 2; i & £ o Altitude (log scale <4 to >1000 m) Fig. 4 Distribution of site records for R. exiguella and 7?. geniculata. See Fig. 3 for further explanation. Altitude (log scale <4 to >1000 m) R. pacifica o CD er CD 30 25 20 15 10 5 0 -, - cr CD TTTÏTf1 >, 12 -i ë 10-1 CD cr cr CD CD P/E ratio Fig. 5 Distribution of site records for R. pacifica and R. peruviana. See Fig. 3 for further explanation. a a' 2; N B Altitude (log scale <4 to > 1000 m) I 1 Mean minimum temperature (>3to>11°C ) Mean minimum temperature (<3to>11°C ) o R. peruviana 30 -, 25 20 15 10 5 0 -I Altitude (log scale <4 to > 1000 m) R. canariensis R. celastri 1 o 12 -, 10 - g CD 8 - CD O" CD 6 4 2 0 CT CD ill Mean minimum temperature (<3to>11°C ) Mean minimum temperature (<3to>11°C ) 3 -i Ü Ü CD CD CT CD 2; i & £ o Altitude (log scale <4 to >1000 m) Fig. 6 Distribution of site records for R. canariensis and R. celastri. See Fig. 3 for further explanation. Altitude (log scale <4 to >1000 m) ü R. unilateralis R. glaucescens 100 -i ü CD =3 O" CD CD O" CD -i - 15 10 5 -J 0 1 Mean minimum temperature (<3to>11°C ) Mean minimum temperature (<3to>11°C ) ü o CD CD er O" CD CD 35 30 25 20 35 -i 30 25 20 15 10 5 0 I - Altitude (log scale <4 to >1000 m) Fig. 7 Distribution of site records for R. glaucescens and R. unilateralis. See Fig. 3 for further explanation. Altitude (log scale <4 to >1000 m) a a' 2; N R. inflexa R. fimbriata CD Mean minimum temperature (<3to>11°C ) Mean minimum temperature (<3to>11°C ) Ü 2; i & £ o Altitude (log scale <4 to >310 m) Altitude (log scale <4 to >1000 m) o. Fig. 8 Distribution of site records for R. fimbriata and R. inflexa. See Fig. 3 for further explanation. Bannister et al.—Ramalina in NZ to the Gisborne region (where it overlaps with the range ofR. glaucescens). Southern species There are four species that occur only in South Island: Ramalina inflexa, R.fimbriata, R. erumpens, andR. riparia (Fig. 8, 9). Herbarium records of R. inflexa (Blanchon et al. 1996) indicated a discontinuous distribution along coastal regions on the east coast of South Island from Cheviot to Invercargill. Our records (Fig. 8) extend its northern limit beyond Cheviot (Hundalee) and show that its distribution is continuous with extension further inland in the southern part of its range (sea level to 560 m). It is common at moderate altitudes, and absent from the driest and warmest areas, but is found over a greater range of temperatures and p/e ratios than the more restricted R. erumpens and R. riparia. Ramalina fimbriata (Fig. 8) is a rare species confined to Otago. It occurs under overhangs on schist tors at the highest altitudes (Galloway 2002), and on stream-side cliffs at moderately high altitudes, but also on coastal rocks and cliffs (Bannister 1998). It was previously considered to be confined to high altitudes (Blanchon et al. 1996), and is most frequent at high altitudes (Fig. 8), although most specimens were collected from sites below 1000 m. R.fimbriata is absent from the driest and wettest sites and tends to be most frequent in cooler locations (Fig. 8). Ramalina erumpens (Fig. 9) is an uncommon species, confined to coastal regions of South Island from Banks Peninsula to western Southland. We have extended its distribution within this range and identified herbarium specimens of R. banzarensis (CANU) from Bird Island, Foveaux Strait (Bannister & Blanchon 2002), and a previously unidentified Ramalina specimen from Banks Peninsula (CHR) as R. erumpens, and added them to our maps. Ramalina riparia (Fig. 9) shares a coastal distribution with R. erumpens from just north of Dunedin southwards. Within this range, R. riparia is more frequently encountered than R. erumpens. Like R. erumpens, R. riparia is most frequent at lower altitudes with moderate temperatures and rainfall. Habitat and substratum New Zealand species of Ramalina are mostly corticolous, and only R. fimbriata and R. meridionalis are predominantly saxicolous (Table 2). There are differences between our records of substrata and those from herbarium records. Herbarium records show proportionately greater occurrence of 133 some species on rock (R. canariensis, R. celastri, R. peruviana, R. unilateralis) and proportionately greater (R. peruviana) or lesser (R. celastri, R. glaucescens, R. inflexa) occurrence of some species on native trees and shrubs than in our records. This almost certainly relates to collection and recording. Our records were largely from roadside and inland sites and targeted areas where few collections had been made. These were biased towards native shrubs as these often proved to support species of Ramalina, whereas herbarium records were biased by the substantive collections that had been made from Auckland northwards in coastal areas and on islands which presented an extensive rocky shoreline. R. inflexa was most frequently recorded on exotic trees and this reflects its distribution in agricultural areas along the eastern coast of South Island in which there are very few native trees or shrubs. Likewise, records from wood (usually gates and fence posts) relate to agricultural areas with few native trees and shrubs, particularly in the south of South Island. Spread, range, and frequency The species of Ramalina that are established in New Zealand have three possible methods of propagation: vegetative spread by thallus fragmentation, dispersal of soredia, and sexual spread by ascospores (Bannister & Blanchon 2003). The establishment of a lichen from an ascospore requires that a Table 2 Occurrence (% of records) of species of Ramalina on various substrata. Figures in bold indicate the type of substratum on which each species has been most frequently recorded. Woody plants Species Rock R.fimbriata R. meridionalis R. australiensis R. canariensis R. peruviana R. geniculata R. erumpens R. pacifica R. exigue lla R. unilateralis R. glaucescens R. riparia R. celastri R. inflexa 100 98 33 35 17 12 10 6 0 3 1 0 18 1 Native Exotic Wood 0 2 67 61 67 80 85 88 89 82 57 65 41 40 0 0 0 4 17 8 5 6 11 11 35 30 36 0 0 0 0 0 0 0 0 0 4 7 5 5 4 55 R. erumpens R. riparia o CD er CD 8 -i 7 6 5 4 3 2 1 0 - Mean minimum temperature (<3to>11°C ) o CD S" Mean minimum temperature (<3to>11°C ) 5 -i 4 3 2 1 0 P/E ratio (<1 to >6 2; o i & £ o Altitude (log scale <4 to >1000 m) Altitude (log scale <4 to >1000 m) O Fig. 9 Distribution of site records for R. riparia and R. erumpens. See Fig. 3 for further explanation. Bannister et al.—Ramalina in NZ 135 Fig. 10 Frequency (%) of fertile (open circles) and vegetative (solid circles) species of Ramalina in New Zealand in relation to their spread (km). 100 q O CD 10 - O D Q 1 I 0 I I I I I I I 200 400 600 800 1000 1200 1400 1600 Spread (km) germinating spore forms a symbiotic relationship with a compatible alga (a species of Trebouxia in the case of Ramalina) whereas soredia are asexual reproductive structures consisting of algae surrounded by fungal hyphae. Consequently it is often considered that lichens producing soredia will be more successful in colonising new sites than those spread by ascospores alone. For example, Bowler & Rundel (1975) stated that "Propagation through vegetative diaspores appears to allow a higher survival rate, so that the frequency and abundance of plants using the asexual reproductive pathway is greater than that of species depending upon the germination of ascospores and a resynthesis of the symbiosis". Similar statements are often repeated in the literature, but seldom, if at all, tested quantitatively. Our distributional data gave us the opportunity to make a simple test. When log frequency is plotted against spread (Fig. 10), there is an overall linear relationship which is highly significant (F1,13 = 87.5, P < 0.001). The individual regressions for sexual and asexual species are also significant (P < 0.01). Analysis of covariance showed that the two regressions had significantly different elevations (P = 0.014) with the sexual species showing higher frequencies than asexual species at any particular level of spread. The two species with the highest frequencies and spread are R. celastri and R. glaucescens, both sexual species. R. unilateralis, a sorediate species, is the third most frequent and widespread species. This is counter to the generally held view that sorediate species are more common and widespread than apothecial species. The British Isles have the same number ( 14) of species of Ramalina as New Zealand and distribution maps of British Lichen Society's database at Bradford University (M. R. D. Seaward pers. comm. 2002) show that, in contrast to New Zealand where many species have a discrete geographic range, the more common British species show wide latitudinal spread (often from Shetland to the Channel Islands). The most frequent species is Ramalina farinacea (sorediate), but the three next most frequent species (R. fastigiata, R. siliquosa, R.fraxinea) are apothecial, while the fifth and sixth most frequent species (R. subfarinacea, R. canariensis) are sorediate. As in the New Zealand examples, there is little evidence to suggest that British and Irish species of Ramalina with soredia are any more widespread or frequent than those producing ascospores. Species diversity and frequency High numbers of species of Ramalina are found in 10-km squares in the north-east and south-east of New Zealand, but R. celastri is the only frequent species in common. There are, however, areas where only R. celastri is found. Squares lacking any species of Ramalina tend to be associated with areas of high rainfall and/or high altitude or extreme temperatures (Fig. 2, 11). 136 New Zealand Journal of Botany, 2004, Vol. 42 Fig. 11 Numbers of species ofRamalina for each 10-km square investigated in this study. Bannister et al.—Ramalina in NZ Diversity of species of Ramalina is high in Northland and coastal South Otago. These areas are in parts of the country where there is a high degree of endemism in the vascular flora (Wardle 1991; McGlone et al. 2001), but the high degree of endemism found in north-west Nelson is not associated with high species diversity of species of Ramalina. Eight of the 14 species of Ramalina found in New Zealand are found in the coastal strip from Otago Peninsula southwards, and the Otago Peninsula has also been observed to have a high diversity of vascular plant species (Rogers & O verton 2000). Species diversity is inevitably related to habitat. Species of Ramalina are relatively light-demanding and are often found in relatively open habitats (e.g., in scrub, forest edges, and upper canopy). The most common species (e.g., R. celastri and R. glaucescens) are found in a wide range of habitats and on a variety of substrata and are often considered to be "weedy" species associated with anthropogenic disturbance. Other species are particularly associated with native forest remnants (R. erumpens, R. riparia, R. pacifica, R. geniculata), mangrove and coastal forest (R. exiguella), or native shrubs (R. unilateralis). Conservation of species diversity within Ramalina will depend on conservation of both native and exotic habitats (e.g., R. inflexa is most frequently found on exotic trees in an otherwise agricultural landscape on the east coast of South Island). CONCLUSION Previous maps showing the distribution of species of New Zealand lichen genera include those for Pseudocyphellaria (Galloway 1988), Ramalina (Blanchon etal. 1996), and Sticta (Galloway 1997). The maps presented in this paper differ principally in their breadth and intensity of coverage. Accurate maps and locations have enabled us to make generalisations about the biodiversity, conservation status, and distribution of species of Ramalina in relation to geographic, habitat, and climatic factors. We hope that the results of this study will encourage others to follow suit, as accurate inventory not only tells us what is currently here but also provides a baseline for assessing future changes of species distribution. 137 ACKNOWLEDGMENTS We thank Barbara Anderson, Catherine Beard, Robyn Bridges, David Galloway, Jennifer Ide, Allison Knight, Kelvin Lloyd, Janice Lord, Keith Thompson, Isolde Ullmann, and Hugh Wilson for providing specimens from places that we had not visited, and M. R. D. Seaward for allowing us to use his unpublished distribution maps of British species of Ramalina. REFERENCES Bannister, J. M. 1998: Some observations on the genus Ramalina in Otago and Southland, NZ. Australasian Lichenology 43: 18-19. Bannister, J. M. 2001: Searching for Merry Hill. Botanical Society of Otago Newsletter 27: 9-10. Bannister, J.; Blanchon, D. J: 2002: The typification and status of the name Ramalina banzarensis Dodge. Australasian Lichenology 51: 14-16. Bannister, J. M.; Blanchon, D. J. 2003: The lichen genus Ramalina Ach. (Ramalinaceae) on the outlying islands of the New Zealand geographic area. The Lichenologist 35: 137-146. Blanchon, D. J. 1994: The genus Ramalina Ach. in New Zealand. Unpublished MSc thesis, University of Auckland, Auckland, New Zealand. 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. Bowler, P. A.; Rundel, P. W. 1975: Reproductive strategies in lichens. Botanical Journal of the Linnean Society 70: 325-340. Galloway, D. J. 1985: Flora of New Zealand lichens. Wellington, Government Printer. Galloway, D. J. 1988: Studies in Pseudocyphellaria (lichens). I. The New Zealand species. Bulletin of the British Museum (Natural History), Botany Series 17: 1-267. Galloway, D. J. 1997: Studies on the lichen genus Sticta (Schreber) Ach. IV. New Zealand species. The Lichenologist 29: 105-168. Galloway, D. J. 2002: Notes on the high-alpine species of Lecanora from schist underhangs in southern New Zealand, and a new name for L. parmelinoides. Australasian Lichenology 51: 20-32. 138 Hellborn, P. J. 1896: Lichenaea neo-zeelandica seu lichenes Novae Zeelandiae a Sv. Berggren annis 1874-75 collecti, additis ceteris speciebus indidem huc usque cognitis breviter commemoratis. Bihang till Konglica Svenska Vetenkaps-Akademiens Handlingar 21(3/13): 1-150. Kirk, P. M.; Cannon, P. F.; David, J. C.; Ansell, A. E. 2001: Ainsworth & Bisby's Dictionary of the Fungi. 9th ed. Wallingford, CABI Bioscience. Kovach, W. L. 1993: A multivariate statistical package for IBM-PC's. ver. 2.1. Kovach Computing Services, Pentraeth, Wales, UK. Leathwick, J. R. 2001: New Zealand's potential forest pattern as predicted from current species-environment relationships. New Zealand Journal of Botany 39: 447-464. Leathwick, J. R.; Stephens, R. T. T. 1998: Climate surfaces for New Zealand. Unpublished Landcare Research Contract Report LC9798/126, held at Landcare Research, Hamilton, New Zealand. Louwhoff, S. H. J. J. 2001: Biogeography of Hypotrachyna, Parmotrema, and allied genera (Parmeliaceae) in the Pacific Islands. Bibliotheca Lichenologica 78: 223-246. Martin, W. 1966: Census catalogue of the lichen flora of New Zealand. Transactions of the Royal Society of New Zealand (Botany) 3: 139-159. New Zealand Journal of Botany, 2004, Vol. 42 Martin, W. 1968: Supplement to census catalogue of New Zealand lichens. Transactions of the Royal Society of New Zealand (Botany) 3: 203-208. McGlone, M. S.; Duncan, R. P.; Heenan, P. B. 2001: Endemism, species selection and the origin and distribution of the vascular plant flora of New Zealand. Journal of Biogeography 28: 199-216. Müller Argoviensis, J. 1894: Conspectus systematicus lichenum Novae Zelandiae. Bulletin de l'Herbier Boissier 2, Appendix 1: 1-114. Raven, P. H.; Engelhorn, T. 1971: A plea for the collection of common plants. New Zealand Journal of Botany 9: 217-222. Rogers, G.; Overton, J. 2000: Regional patterns of plant species richness in southern New Zealand. New Zealand Journal of Botany 38: 609-627. Wardle, P. 1991: Vegetation of New Zealand. Cambridge, Cambridge University Press. Zahlbruckner, A. 1941: Lichenes Novae Zealandiae a cl. H. H. Allan eisque collaboratoribus lecti. Denkschriften der Akadamie der Wissenschaften in Wien: mathematisch-naturwissenschaftliche Klasse 104: 249-380.