Corticolous Microlichens in Northeastern Brazil: Habitat Differentiation Between Coastal Mata Atlântica, Caatinga and Brejos de Altitude Author(s): Marcela E. S. Cáceres, Robert Lücking, Gerhard Rambold Source: The Bryologist, 111(1):98-117. 2008. Published By: The American Bryological and Lichenological Society, Inc. DOI: http://dx.doi.org/10.1639/0007-2745(2008)111[98:CMINBH]2.0.CO;2 URL: http://www.bioone.org/doi/ full/10.1639/0007-2745%282008%29111%5B98%3ACMINBH%5D2.0.CO %3B2 BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/ terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder. BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. Corticolous microlichens in northeastern Brazil: habitat differentiation between coastal Mata Atlântica, Caatinga and Brejos de Altitude MARCELA E. S. CÁCERES Universität Bayreuth, Lehrstuhl für Pflanzensystematik NWI, Abteilung Mykologie und Lichenologie, Universitätsstraße 30, 95440, Bayreuth, Germany. Current address: Departamento de Micologia, Centro de Ciências Biológicas, Universidade Federal de Pernambuco, 50670-420 Recife, PE, Brazil e-mail: marcela.caceres@uni-bayreuth.de ROBERT LÜCKING Department of Botany, The Field Museum, 1400 South Lake Shore Drive, Chicago, IL 60605-2496, U.S.A. e-mail: rlucking@fieldmuseum.org GERHARD RAMBOLD Corresponding author: Universität Bayreuth, Lehrstuhl für Pflanzensystematik NWI, Abt. Mykologie und Lichenologie, Universitätsstraße 30, 95440, Bayreuth, Germany e-mail: gerhard.rambold@uni-bayreuth.de ABSTRACT. Based on a study of 22 sites in northeastern Brazil, including the three main vegetation types, coastal Mata Atlântica (Zona da Mata), Caatinga and Brejos de Altitude (rain forest enclaves in Caatinga areas), we studied the distribution and habitat preferences of 456 crustose and microfoliose lichen species. Alpha-diversity ranged between three and 99 species per site, with Zona da Mata and Brejos de Altitude showing higher numbers than Caatinga sites. Beta-diversity (dissimilarity) was highest between Zona da Mata sites and, as a whole, the Zona da Mata showed the highest gamma-diversity, with a total of 334 species. Site ordination by non-metric multidimensional scaling (NMS), as well as cluster analysis, both using Sørensen’s coefficient of dissimilarity, show that Zona da Mata and Caatinga sites have distinctive lichen species compositions, with the isolated Brejos de Altitude being more similar to coastal Zona da Mata than to Caatinga sites. Exposed Zona da Mata sites have certain species in common with Caatinga sites but overall cluster with the Zona da Mata sites. The transitional Agreste vegetation (one study site) also appears transitional between Zona da Mata and Caatinga in its lichen species composition. Indicator species analysis for each vegetation type was performed by applying a MonteCarlo test. Other than ten ubiquitous taxa (found in all three vegetation types), 59 taxa were shared between Zona da Mata and Brejos de Altitude, 20 between Zona da Mata and Caatinga, and none between Brejos de Altitude and Caatinga. Dissimilarity values of Zona The Bryologist 111(1), pp. 98–117 Copyright E2008 by The American Bryological and Lichenological Society, Inc. 0007-2745/08/$2.15/0 Cáceres et al.: Microlichens in NE Brazil 99 da Mata versus Brejos de Altitude sites were high (0.77 or 23% shared species on average), as were those of Zona da Mata versus Caatinga sites (average of 0.92 or 8% shared species). Zona da Mata lichens had a higher proportion of Arthoniomycetidae (Arthoniales: Arthoniaceae, Roccellaceae) and Chaetothyriomycetidae (Pyrenulales: Pyrenulaceae), as well as Porinaceae and Thelotremataceae; frequently trentepohlioid photobionts, predominantly transversely septate and/or narrow ascospores, and lack of lichen substances. Brejos de Altitude lichens showed a higher proportion of Dothideomycetiae (Trypetheliaceae) and Ostropomycetidae (Ostropales: Gomphillaceae and Graphidaceae), as well as Pilocarpaceae; ascospores were predominantly thick-walled or muriform and hyaline. Caatinga sites were dominated by Lecanoromycetidae (Lecanorales: Lecanoraceae; Teloschistales: Physciaceae) and Pertusariales (Pertusariaceae); taxa were chiefly associated with chlorococcoid photobionts, ascospores were megalosporous, non-septate and/or brown, and showed a predominance of certain cortical substances (atranorin, lichexanthone other xanthones, pulvinic acid derivates), as well as norstictic acid as medullary substance. KEYWORDS. Atlantic rain forest, Brazil, crustose and microfoliose lichens, Alagoas, Paraı́ba, Pernambuco, Rio Grande do Norte, Sergipe. ¤ The Atlantic rain forest (Mata Atlântica) is one of the three major rain forest blocks in the Neotropics (Galindo-Leal & Câmara 2003; Morellato & Haddad 2000; Myers et al. 2000; Tabarelli et al. 2005; Whitmore 1990) and it is among the 25 world biodiversity hotspots (Costa et al. 2000; Myers et al. 2000). More than 95% of the original vegetation cover of the Mata Atlântica in northeastern Brazil has been eliminated or is strongly affected by human activities, mainly agriculture (sugar cane plantations), logging and the extension of urban areas like Recife and Salvador (Myers et al. 2000; Paciencia & Prado 2005; Silva Filho et al. 1998; Tabarelli et al. 2005; Whitmore 1990). One of the consequences of the extensive land use change since the beginning of the Portuguese colonization in Brazil five centuries ago is the increasing drought, which subsequently affects the already reduced and over-stressed rain forests remnants (Cáceres et al. 2000; Galindo-Leal & Câmara 2003; Paciencia & Prado 2005; Silva Filho et al. 1998; Tabanez & Viana 2000). The region of northeastern Brazil includes the states of Maranhão, Piauı́, Ceará, Rio Grande do Norte, Paraı́ba, Pernambuco, Alagoas, Sergipe and ¤ ¤ Bahia. Of these, Rio Grande do Norte, Paraı́ba, Pernambuco, Alagoas and Sergipe form the small northeastern coastal states focused upon in this study. The region has three main vegetation types: (1) coastal Atlantic rain forest (Zona da Mata), (2) Caatinga (Sertão) and (3) isolated montane Atlantic rain forest remnants within the Caatinga and the transitional Agreste vegetation, the so-called Brejos de Altitude (Andrade-Lima 1954, 1961; Galindo-Leal & Câmara 2003; Marcelli 1998; Rizzini 1977; Silva Filho et al. 1998; Whitmore 1990). The Mata Atlântica covers a narrow strip along the coast, extending from Rio Grande do Norte in northeastern Brazil to Rio Grande do Sul in southern Brazil, becoming broader and covering the Serra da Mantiqueira, Serra do Mar and Serra do Espinhaço at the latitudes of Minas Gerais and São Paulo states. In its northeastern part, it is characterized as a perennial forest with a pronounced dry season, while its southern part is more humid (Galindo-Leal & Câmara 2003; Morellato & Haddad 2000; Tabarelli et al. 2005; Whitmore 1990). The Caatinga is a dry thorn-bush, in some parts desert-like vegetation, while the Brejos de Altitude are Pleistocene rain forest remnants, isolated from the coastal vegetation 100 THE BRYOLOGIST 111(1): 2008 and located within the Caatinga (Rodal 1998), in areas of higher altitude (500–1100 m). As much as 95% of the original cover of the Atlantic rain forest has been deforested in northeastern Brazil, and many tree species have locally disappeared (Cardoso Silva & Tabarelli 2000; FIDEM 1987; Galindo-Leal & Câmara 2003; Myers et al. 2000; Ranta et al. 1998; Tabanez & Viana 2000; Tabarelli et al. 2005; Whitmore 1990). Hitherto, no comprehensive lichen inventory has been undertaken in northeastern Brazil, although the major vegetation types are assumed to have high lichen diversity, possibly near 1000 species within the study region. Barros and Xavier-Filho (1972) published a catalogue of lichens housed in the herbarium of the Federal University of Pernambuco in Recife (URM), but a large part of the lichen samples cited in this compilation are from areas outside northeastern Brazil, including Europe, and came to the herbarium by exchange. Batista and collaborators made extensive collections of foliicolous lichens in the area (Silva & Minter 1995), but most of their identifications were subsequently found to be incorrect (Lücking et al. 1998, 1999). The most recent lichen inventory for this area, including a thorough revision of collections made by Batista and co-workers, covered foliicolous lichens only, reporting 191 species for Pernambuco state (Cáceres 1999; Cáceres & Lücking 2000; Cáceres et al. 2000; Lücking et al. 1999). Except for a few large-scale monographic treatments that mention a few collections from the area (Ahti 2000; Brako 1991; Frisch et al. 2006; Harris 1986, 1989; Kalb et al. 2000, 2004; Kashiwadani & Kalb 1993; Sparrius 2004; Staiger 2002; Staiger & Kalb 1995, 1999; Tehler 1993; Tibell 1996), and collections made by the German lichenologist Klaus Kalb mostly in Bahia state (Kalb 1981, 1982a–d, 1983, 1984, 1987, 2001, 2004; Kalb & Elix 1995), little is known about the corticolous lichens of northeastern Brazil, in particular the crustose and microfoliose species. Apart from the lack of knowledge of the lichen biota, much less is known about the distribution and ecology of lichens within the area. Such knowledge is indispensable to assess, for example, the feasibility of using lichens as bioindicators of the impact of deforestation and land use change. Foliicolous lichens have demonstrated potential as bioindicators of the impact of fragmentation on biodiversity in Atlantic rain forest remnants in Pernambuco (Cáceres et al. 2000), and since corticolous microlichens are not restricted to evergreen rain forests, their potential applications as bioindicators are more extensive. While knowledge of tropical lichen ecology is rudimentary, a few quantitative studies have been published, including montane rain forests in Colombia and Ecuador and lowland rain forest and savannas in Venezuela, Guyana and French Guiana (Cornelissen & Ter Steege 1989; Komposch & Hafellner 1999, 2000, 2002, 2003; Marcelli 1992; Martins 2006; Montfoort & Ek 1990; Nöske 2004; Nöske & Sipman 2004; Wolf 1993). Microlichens make up a significant part of the diversity of lowland forests, but were identified to species level only in a few studies (Komposch & Hafellner 1999, 2000, 2002, 2003; Marcelli 1992; Martins 2006). In northeastern Brazil, a recent study dealing with the diversity of corticolous lichens in three Atlantic rain forest remnants (Pereira et al. 2005a–c) was the first to assess microlichen diversity in individual forest fragments. Expanding from that study, the present paper is the first to examine the distribution and habitat preferences of corticolous microlichens across the three principal vegetation zones in northeastern Brazil, based on the identification of 456 species (Cáceres 2007). MATERIAL AND METHODS For the comparative analysis of the crustose and microfoliose corticolous lichen biota in the three vegetation types in northeastern Brazil, 22 localities (Fig. 1) were selected, representing the coastal Mata Atlântica or Zona da Mata (13 sites), transitional Agreste (1 site), Caatinga (5 sites) and Brejos de Altitude (3 sites). A detailed description of the study area and collecting sites is given elsewhere (Cáceres 2007; Cáceres et al. 2007a, b). Non-quantitative opportunistic sampling was applied to each site as suggested by Sipman (1996). At each site, lichens were collected from tree bark along the main trail through the site. Trees were inspected within a 20 m broad strip along both sides of the corresponding trail, with a distance of about Cáceres et al.: Microlichens in NE Brazil 101 Figure 1. Map of Brazil showing the five major geographic regions (A) and the location of the study area and sites (B). Three of the 17 dots each indicate two close localities, and one indicates three localities, for a total of 22. 10–20 m between each tree. About 50–100 trees were sampled at each site. Each lichen thallus recognized as potentially distinct in the field and featuring identifiable structures (ascomata, conidiomata, soralia, isidia, etc.) was collected, amounting to a total of 1–5(–10) thalli per tree and a total of 100–200 (–300) specimens per locality (see also Cáceres et al. 2007a, b). One site, the RPPN Fazenda São Pedro, was sampled three times repetitively, for a separate analysis comparing different sampling methods (Cáceres et al. 2007b). Methods and literature used for the identification of the lichen material are given with detail in Cáceres (2007). Most lichen samples were duplicated and sets were deposited in URM, B and F. For the statistical analysis, each species was assigned an abundance score for each site, based on the number of collections made: 0 5 absent, 1 5 rare (1–3 collections), 2 5 intermediate (4–10 collections), and 3 5 abundant (.10 collections). Alpha-diversity was calculated as the number of species per site, while gamma-diversity was calculated as the total number of species per vegetation type. Beta-diversity, that is dissimilarity between sites, was computed using the relative Sørensen coefficient of dissimilarity (McCune & Mefford 1999; McCune et al. 2002). Lichen species composition at each site was used to ordinate and classify sites by applying non-metric multidimensional scaling (NMS) as ordination method and cluster analysis based on the relative Sørensen coefficient of dissimilarity as classification method (McCune et al. 2002; McCune & Mefford 1999). Flexible beta 5 20.25 was used as the 102 THE BRYOLOGIST 111(1): 2008 clustering algorithm; this method resulted in tight clusters similar to those obtained from Ward’s method but contrary to the latter, flexible beta is compatible with a distance matrix derived from Sørensen’s coefficient of dissimilarity (McCune et al. 2002). Indicator species analysis was performed to detect species that can be classified as characteristic of a given vegetation type. For that purpose, a MonteCarlo test was performed on the original frequency (number of sites where species was present) and abundance data (categorized number of collections per site), that is the data were mixed randomly with 1000 repetitions, and it was tested whether the observed data distribution deviated significantly from the random distribution derived from the Monte-Carlo test, i.e., whether a given species was significantly more abundant and frequent within a given vegetation type than expected by random (McCune et al. 2002). Lichen species unique to each of the three major vegetation types (Atlantic rain forest, Brejos de Altitude and Caatinga) were used to test whether the observed frequency of selected character states (systematic affinity, morphology, anatomy, chemistry) among vegetation types differed significantly from the expected frequency based on the overall frequency of the character state assuming random distribution. Observed versus expected frequencies were compared within each vegetation type across all states of a given character, and a ChiSquare test was used to determine statistical significance of the observed differences. Twelve characters were used with different sets of character states (see Results for details). The statistical analyses were carried out using STATISTICATM 6.0 and PC-ORD 4.0 (McCune & Mefford 1999). RESULTS Patterns of alpha-, beta- and gamma-diversity. A total of 456 species of corticolous crustose and microfoliose lichens were found in the three vegetation types in northeastern Brazil. A complete checklist and taxonomic treatment for the reported taxa is published elsewhere (Cáceres 2007). The number of species per site varied from three to 99 (Figs. 2, 3). The site with the highest number of species (99) was the repetitively sampled RPPN Fazenda São Pedro in the state of Alagoas; here, the first trip resulted in 53 species and the two subsequent trips produced an increase of 87%. The highest species numbers for sites sampled one time opportunistically were found for Brejos de Altitude, which contributed 84 (Brejo dos Cavalos) and 73 species (Parque Municipal de Bonito), respectively. Apart from RPPN Fazenda São Pedro, sites within the Zona da Mata had slightly lower species numbers, the richest being the Refúgio Ecológico Charles Darwin (71) and the Estação Ecológica de Gurjaú (60) in Pernambuco state. The three sites of exposed secondary vegetation within the Zona da Mata, RPPN Rosa do Sol, UFPE Campus, and the exposed secondary vegetation at Gurjaú, showed little variation in the number of species (16–19), while only eight taxa were found at the exposed secondary vegetation at Brejo dos Cavalos. One site, Estação Ecológica de Tapacurá, was located in the transitional Agreste region; it had 22 species. Within the Caatinga, the number of species varied between 23 and 54 per site, with the largest number reported for the most conserved Caatinga vegetation at IPA in Caruaru. The two sites representing exposed secondary Caatinga area, Garanhuns and the exposed secondary vegetation at IPA, had 3–12 species. Combining all sites within each of the three main vegetation types, the Zona da Mata yielded a total of 281 species, being the most diverse of the three vegetation zones (Fig. 3). For the exposed secondary Zona da Mata sites and the higher altitude Zona da Mata forests, 43 and 25 taxa were found in total, and the combined number for all Zona da Mata sites, including exposed secondary sites, was 334. The Brejos de Altitude localities represented the second most diverse region of the study area, with a total of 136 species. The three sites representing Caatinga vegetation comprised a total of 79 species, and the two exposed secondary Caatinga sites yielded 15 taxa, totaling 84 species for all five Caatinga sites. Because of the different number of sample sites per vegetation, total species numbers across vegetation are not directly comparable; logarithmic transformation of these numbers [S9 5 S/100 3 log(N)], where S9 5 logarithmically transformed Cáceres et al.: Microlichens in NE Brazil 103 Figure 2. Alpha-diversity (number of species) per site. Sites arranged according to major vegetation types and subtypes (anthropogenic variations), from high to low values for each type. Figure 3. Variation in alpha-diversity across sites and gammadiversity (total number of species) per vegetation type. Box and whiskers show mean, standard deviation (box) and min-max (whisker); bold numbers indicate gamma-diversity per vegetation subtype (differentiating between closed forest and exposed secondary vegetation), while thinner numbers indicate total gamma-diversity for main vegetation types (specifically Zona da Mata and Caatinga). diversity index, S 5 original species number per vegetation type, and N 5 number of sites per vegetation type, results in S9 5 2.9 for Zona da Mata, 4.5 for Brejos de Altitude, and 1.2 for Caatinga, indicating that the highest overall diversity is to be expected in Brejos de Altitude vegetation (which is supported by the higher average species numbers for the two study sites). Beta-diversity (Sørensen dissimilarity) in lichen species composition between any two sites was relatively high, with values ranging between 0.41 and 1.00 and a mean of 0.79 (Table 1). In other words, any two sites shared between 0% and 59% of their species, with a mean of 21%. When comparing Zona da Mata sites only, the smallest dissimilarity value between any two sites was 0.45, which means that a maximum of 55% of the species was shared between any two sites. The two sites representing Brejos de Altitude had a dissimilarity of 0.56, meaning that they shared 44% of the species. With respect to the 104 THE BRYOLOGIST 111(1): 2008 Table 1. Sørensen dissimilarity and percentage of similarity between sites (ranges and mean), arranged according to the three main vegetation types and their anthropogenic variations. Vegetation type Dissimilarity (range) Dissimilarity (mean) Similarity (range) Similarity (mean) Zona da Mata (0–100 m) Zona da Mata (0–100 m exposed) Zona da Mata (300–500 m montane) Brejos de Altitude Caatinga Caatinga (exposed) 0.45–1.00 0.67–0.85 — — 0.41–0.83 — 0.78 0.75 1.00 0.56 0.64 0.72 0–55% 15–33% — — 17–59% — 22% 25% 0% 44% 36% 28% All sites 0.41–1.00 0.79 0–59% 21% three undisturbed Caatinga sites, the smallest dissimilarity was 41% and the highest 83%, which means 17–59% of the species were shared between sites. Although mean similarity was highest between Brejos de Altitude sites and lowest between Zona da Mata sites, the differences were not statistically significant (Kruskal-Wallis ANOVA). Comparing sites across the three main vegetation types, dissimilarity values of Zona da Mata versus Brejos de Altitude sites were found to be relatively high, with an average value of 0.77 or 23% shared species (Table 2). Differences in species composition of Zona da Mata versus Caatinga were even more pronounced, with dissimilarity values averaging 0.92 (8% shared species). Similarly high dissimilarity values were found between sites representing Brejos de Altitude and Caatinga. While the average similarity of Zona da Mata versus Brejos de Altitude is not different from the average between Zona da Mata sites (22%) and the average for all sites (21%), similarity of Zona da Mata and Brejos de Altitude versus Caatinga sites is significantly lower (8% vs. 22%, 44% and 36%, respectively). This indicates that Caatinga lichen communities are more distinct from both Zona da Mata and Brejos de Altitude communities than Brejos de Altitude from Zona da Mata communities. Ordination and classification of sites. NMS (non-metric multidimensional scaling) ordination, based on lichen species composition at each site, revealed a distinct pattern reflecting the three main vegetation types and their anthropogenic disturbance variations (Fig. 4). Sites representing closed Zona da Mata vegetation in the upper and left portion of the diagram are clearly differentiated against the Caatinga sites in the lower right portion. The two Brejos de Altitude sites are found at the upper left extreme of this polarization axis, but fairly close to the Zona da Mata sites. The exposed secondary sites of the Zona da Mata (upper right portion of the diagram) show affinities with both Zona da Mata and Caatinga sites, but eventually cluster with Zona da Mata sites (Fig. 5). The two montane Zona da Mata sites at higher elevations do not cluster together and also do not cluster with the two Brejos de Altitude sites, suggesting distinct differences in their lichen species composition compared to the latter. The Agreste site (Tapacurá) falls intermediate between the Zona da Mata and the Caatinga sites, reflecting its transitional character. Contrary to the Zona da Mata Table 2. Sørensen dissimilarity and percentage similarity between sites (ranges and mean), comparing sites across the three main vegetation types. Vegetation type Dissimilarity (range) Dissimilarity (mean) Similarity (range) Similarity (mean) Zona da Mata vs. Brejos de Altitude Zona da Mata vs. Caatinga Caatinga vs. Brejos de Altitude 0.57–1.00 0.69–1.00 0.87–0.98 0.77 0.92 0.92 0–43% 0–31% 2–13% 23% 8% 8% All sites 0.41–1.00 0.79 0–59% 21% Cáceres et al.: Microlichens in NE Brazil 105 Figure 4. NMS ordination of the 22 sites. Arrowed lines indicate correlations of the main axes with humidity (precipitation regime) and exposure (light level). M 5 Zona da Mata sites, S 5 montane Zona da Mata sites, B 5 Brejos de Altitude sites, A 5 Agreste site, C 5 Caatinga sites; black and gray dots indicate closed vegetation; white dots indicate exposed secondary vegetation. sites, the exposed secondary Caatinga sites do not cluster separately from the closed Caatinga sites, and the exposed Brejo de Altitude site clusters together with the Caatinga sites, far from the closed Brejos sites, indicating that exposed vegetation near Brejos is dominated by Caatinga lichens. The cluster analysis of the 22 sites chiefly supports the pattern observed in the NMS diagram, with the formation of two major clusters (Fig. 5). The first cluster unites all Zona da Mata sites, as well as Brejos de Altitude and Agreste sites, while the second cluster is comprised of the Caatinga and the exposed Brejos site. One exception is the Caatinga site Itabi, which in the ordination diagram (Fig. 4) is the Caatinga site closest to the Zona da Mata sites and in the cluster analysis clusters with the latter; this site is situated very close to Zona da Mata vegetation and therefore its lichen species composition appears to include a portion of Zona da Mata lichens. Within the large Zona da Mata cluster, there are two subgroups: one includes the two Brejos de Altitude sites and the high diversity Zona da Mata sites (Charles Darwin, Gurjaú, RPPN São Pedro and Serra Itabaiana), while the second includes the three 106 THE BRYOLOGIST 111(1): 2008 Figure 5. Cluster analysis of the 22 sites. M 5 Zona da Mata sites, S 5 montane Zona da Mata sites, B 5 Brejos de Altitude sites, A 5 Agreste site, C 5 Caatinga sites; black and gray dots indicate closed vegetation; white dots indicate exposed secondary vegetation. Cáceres et al.: Microlichens in NE Brazil Figure 6. Number of lichen species unique for, and shared between, each of the three main vegetation types. Numbers in boldface in white circles indicate number of shared species between vegetation types, and numbers in parentheses below indicate total number of species for combined vegetation types. exposed secondary sites, the Agreste and the Caatinga site, and all low diversity Zona da Mata sites. This suggests that low diversity, transitional and exposed sites all share a number of possibly common, widespread species with broader ecological amplitude (otherwise they would not cluster together), while the high diversity sites share more ecologically restricted and rare taxa. Still, the structure of the cluster diagram shows overall high dissimilarity between individual sites, even within smaller clusters. Vegetation types and their indicator species. Among the 456 corticolous crustose and microfoliose lichen species found in the present study, only ten were shared among all three main vegetation types: Dyplolabia afzelii, Lecanora helva, Malcolmiella fuscella, M. gyalectoides, M. leptoloma, M. vinosa, Phaeographis brasiliensis, Trypethelium ochroleucum, T. subeluteriae and T. tropicum. Fifty-nine taxa were shared between Zona da Mata and Brejos de Altitude, 20 between Zona da Mata and Caatinga and none between Brejos de Altitude and Caatinga, supporting the affinity of Brejos de Altitude with Zona da Mata rather than Caatinga in terms of lichen species composition. A total of 366 species were unique to one zone, either Zona da Mata (245), Brejos de Altitude (67) or Caatinga (54) (Fig. 6). Genera containing more than one species and unique to the Zona da Mata or particularly well-represented there are Arthonia, Arthothelium, Bactrospora, Bapalmuia, Bacidiopsora, Bathelium, Carbacanthographis, 107 Coccocarpia, Cresponea, Crocynia, Cryptothecia, Enterographa, Fellhanera, Herpothallon, Letrouitia, Myriotrema, Opegrapha and Sarcographa. Unique to Brejos de Altitude or particularly well represented there are Astrothelium, Calopadia, Echinoplaca, Eugeniella, Hemithecium, Lopezaria, Sagenidium, Trichothelium and Vainionora. Genera restricted to Caatinga or most commonly found there include Baculifera, Chrysothrix, Cratiria, Dirinaria, Hafellia, Lecanographa, Lecanora, Maronina, Ochrolechia, Pertusaria, Physcia, Pyxine, Rinodina and Stigmatochroma. Although 245 lichen species are unique to the Zona da Mata, none is significantly indicative for this type of vegetation (Table 3). This is due to the fact that the Zona da Mata sites included in the study (13) outnumbered the Brejos de Altitude (3) and Caatinga sites (5), so the deviations observed from expected indicator values (IV) are too small to produce significant p-levels. For example, a species present at all 13 Zona da Mata sites but absent from Brejos de Altitude and Caatinga would have an expected frequency within the Zona da Mata of 132/ 22 5 7.7 if randomly distributed, and the observed frequency of 13 would be within the 95% confidence interval of that value. On the other hand, a species present at all five Caatinga sites would only have an expected random Caatinga frequency of 52/22 5 1.14; in that case, the observed frequency lies outside the 95% confidence interval and the difference becomes significant at the 5% level. Accordingly, the Brejos de Altitude sites have 24 statistically significant indicator species and the Caatinga sites have eight (Table 3). The aspect of characteristic Zona da Mata, Brejos de Altitude and Caatinga lichen species is very different from each other (Figs. 7–9). Among the lichen species unique to the Zona da Mata, there is a higher proportion of subclasses Arthoniomycetidae (Arthoniales: Arthoniaceae, Roccellaceae) and Chaetothyriomycetidae (Pyrenulales: Pyrenulaceae), as well as the families Porinaceae and Thelotremataceae (Table 4). Because of the aforementioned explanation, this difference is not significant (Chi-Square test). Brejos de Altitude have a significantly higher proportion of Dothideomycetiae (Trypetheliaceae) and 108 THE BRYOLOGIST 111(1): 2008 Table 3. Indicator lichen species analysis for the three main vegetation types in northeastern Brazil. Columns 2–4 give the combined abundance/frequency score for each species within each vegetation type; IV indicates observed and expected indicator values (combined abundance/frequency scores) for the vegetation type in which the species is relatively most common. IV values for Zona da Mata are not statistically significant because of the high proportion of study sites belonging to this vegetation type (see text). Zona da Mata Brejos Caatinga Obs. IV Exp. IV p-level Zona da Mata Malcolmiella psychotrioides Pyrenula mamillana Malcolmiella badimioides Letrouitia domingensis Graphis glaucescens Letrouitia vulpina Pyrenula nitidula Cryptothecia striata Malcolmiella granifera Coenogonium subdentatum Arthonia bessalis Phaeographis crispata 53 40 53 40 33 33 33 27 27 27 27 27 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 53.3 40.0 53.3 40.0 33.3 33.3 33.3 26.7 26.7 26.7 26.7 26.7 30.7 27.2 31.0 28.1 26.6 26.9 27.1 23.8 24.7 25.3 25.1 25.2 0.100 0.103 0.107 0.127 0.250 0.292 0.299 0.417 0.445 0.489 0.492 0.500 Brejos de Altitude Chapsa platycarpella Herpothallon rubrocinctum Byssoloma chlorinum Byssoloma leucoblepharum Byssoloma aff. meadii Calopadia pruinosa Diorygma reniforme Echinoplaca leucotrichoides Malcolmiella hypomela Phaeographis kalbii Trichothelium horridulum Coenogonium geralense Chapsa dilatata Malcolmiella gyalectoides Malcolmiella leptoloma Porina nucula Malcolmiella atlantica Coenogonium strigosum Dyplolabia afzelii Phaeographis haematites Trypethelium tropicum Laurera megasperma Malcolmiella fuscella Coenogonium pyrophthalmum 5 0 0 0 0 0 0 0 0 0 0 0 11 4 2 3 2 6 8 10 1 0 0 0 84 95 100 100 96 100 100 100 100 100 100 94 73 78 82 85 86 76 71 71 70 47 48 47 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 1 0 8 0 0 0 84.3 94.9 100 100 95.7 100 100 100 100 100 100 93.7 73.2 77.6 82.2 84.9 85.7 76.3 70.6 71.4 69.8 46.9 47.9 46.9 28.9 21.4 18.3 18.6 21.1 18.6 18.6 18.6 18.5 18.3 18.2 21.4 29.2 28.9 27.9 26.3 24 28.4 30.6 28.4 28.5 17.8 17.4 18.1 0.003 0.004 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.007 0.008 0.010 0.010 0.010 0.011 0.019 0.023 0.027 0.036 0.049 0.049 0.050 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 83 67 50 50 67 50 43 42 83.3 66.7 50 50 66.7 50 43.1 41.7 27.6 24.8 22.4 21.8 25.1 22.1 24.5 25.1 0.006 0.022 0.024 0.024 0.026 0.026 0.051 0.060 Caatinga Pyrrhospora haematites Baculifera pseudomicromera Pertusaria flavens Maronina multifera Haematomma persoonii Lecanora hypocrocina Dirinaria leopoldii Pertusaria quassiae Cáceres et al.: Microlichens in NE Brazil 109 Figure 7. Selected lichen species characteristic of the coastal Mata Atlântica (Zona da Mata) vegetation in northeastern Brazil: A. Arhonia bessalis. B. Coenogonium subdentatum. C. Cryptothecia striata. D. Graphis glaucescens. E. Letrouitia domingensis. F. Malcolmiella badimioides. G. Phaeographis crispata. H. Pyrenula nitidula. 110 THE BRYOLOGIST 111(1): 2008 Figure 8. Selected lichen species characteristic of the Brejos de Altitude vegetation in northeatsern Brazil: A. Chapsa platycarpella. B. Diorygma reniforme. C. Dyplolabia afzelii. D. Laurera megasperma. E. Lopezaria versicolor. F. Phaeographis kalbii. G. Sagenidiopsis undulata. H. Trichothelium horridulum. Cáceres et al.: Microlichens in NE Brazil 111 Figure 9. Selected lichen species characteristic of the Caatinga vegetation in northeastern Brazil: A. Baculifera pseudomicromera. B. Cratiria lauricassiae. C. Dirinaria leopoldii. D. Haematomma persoonii. E. Maronina multifera. F. Ochrolechia africana. G. Pertusaria quassiae. H. Pyrrhospora haematites. 112 THE BRYOLOGIST 111(1): 2008 Table 4. Differences in the relative proportion of lichen species belonging to different higher taxa and showing different morphological, anatomical, and chemical features, between the three main vegetation types (Chi-Square test). Predominant taxa and features are indicated in boldface. TTT 5 highly significant (p , 0.001), T 5 significant (p , 0.05), (T) 5 tendential (p , 0.1), and (–) 5 not significant. Subclass Order/Suborder Family Thallus type Photobiont Ascomatal type Vegetative dispersal Ascospore type Ascospore septa Ascospore shape Ascospore color Chemistry Zona da Mata p-level Brejos de Altitude p-level Caatinga p-level Arthoniomycetidae Chaetothyriomycetiae Arthoniales Ostropales Pyrenulales Arthoniaceae Porinaceae Pyrenulaceae Roccellaceae Thelotremataceae squamulose trentepohlioid perithecia (–) Dothideomycetidae Ostropomycetidae Ostropales T Lecanoromycetidae T T TTT (–) Gomphillaceae Graphidaceae Pilocarpaceae Trypetheliaceae T Lecanorales Pertusariales Teloschistales Lecanoraceae Pertusariaceae Physciaceae (–) T (–) byssoid [none] lirellae (–) (–) (–) (T) TTT (–) isidia [none] transverse narrow [none] nil psoromic acid (–) (–) (–) (–) (–) (–) [none] thick-walled muriform broad hyaline nil (–) (–) T (–) (–) (–) microfoliose chlorococcoid apothecia stromata soredia megalosporous nonseptate broad brown atranorin lichexanthone norstictic acid pulvinic acids xanthones (–) Ostropomycetidae (Ostropales: Gomphillaceae and Graphidaceae), as well as Pilocarpaceae (p , 0.05), while Lecanoromycetidae (Lecanorales: Lecanoraceae; Teloschistales: Physciaceae) and Pertusariales (Pertusariaceae) are the predominant subclasses, orders and families found in Caatinga (p , 0.001). The predominant thallus type is squamulose for Zona da Mata, byssoid for Brejos de Altitude and microfoliose for Caatinga, but the observed differences are significant for the Caatinga only (Table 4). Lichens in the Zona da Mata frequently have trentepohlioid photobionts (p , 0.05), while those in Caatinga vegetation are associated with chlorococcoid photobionts (p , 0.001). Vegetative dispersal by isidia is more common within the Zona da Mata, while Caatinga lichens more frequently disperse by soredia, but the difference is not significant. The predominant ascomatal types are perithecia for Zona da Mata, lirellae for Brejos de TTT (T) T T (–) T TTT Altitude and apothecia and stromata for Caatinga, but the differences are not significant. Ascospores are predominantly transversely septate and/or narrow in lichens of the Zona da Mata (not significant), thickwalled or muriform (p , 0.05) and hyaline in those of Brejos de Altitude, while megalosporous (largespored) non-septate and/or brown (all p , 0.05) in Caatinga species. Both Zona da Mata and Brejos de Altitude lichens have no predominant secondary substances, except for psoromic acid in the first, but Caatinga lichens show a highly significant predominance of atranorin, lichexanthone and other xanthones, as well as pulvinic acid derivatives, as cortical substances, and norstictic acid as a medullary substance. DISCUSSION Few studies are available in which lichen species diversity, specifically microlichens, has been assessed for tropical vegetation. Also, the available studies all Cáceres et al.: Microlichens in NE Brazil used somewhat different sampling techniques, included different forest strata and identified the lichens to various taxonomic levels depending on growth form. While several studies included canopy lichens collected with tree-climbing techniques, often only macrolichens were identified to species level. Comparisons of the numbers found in this study with other tropical sites are therefore difficult, especially as a separate analysis of our data (Cáceres et al. 2007b) showed that opportunistic, repetitive and quantitative sampling can yield different species numbers, with opportunistic sampling underestimating lichen diversity by as much as 50%. Defining ‘‘site’’ as a single, continuous study area of about 1–5 ha in size, montane rain forests in Costa Rica, Colombia and Ecuador had 32–51 macrolichens and an undetermined number of microlichen species per site (Holz & Gradstein 2005; Nöske 2004; Nöske & Sipman 2004; Wolf 1993); in these cases, microlichens were left mostly undetermined or determined to genus level only. Montane forests are quite different from the relatively dry and hot northeastern Atlantic rain forest and Caatinga and generally have higher lichen biomass but fewer species, especially microlichens, than lowland forests. Lowland sites were investigated in Venezuela by Komposch and Hafellner (1999, 2000, 2002, 2003), in Guyana by Cornelissen and Ter Steege (1989), in French Guiana by Montfoort and Ek (1990), and in Brazil by Aptroot (2002), Marcelli (1992) and Martins (2006). These workers reported mostly macrolichens and left most of the microlichens unidentified, except for Aptroot (2002), Komposch and Hafellner (1999, 2000, 2002, 2003), Marcelli (1992) and Martins (2006), who counted 161–268 species at a single site, including all lichen groups and forest strata. The latter numbers are consistent with the study by Aptroot (1997, 2001) who identified 173 lichen species on a single tree in Papua New Guinea, and recent inventories in Costa Rica, with about 300 corticolous lichen species each found at a lowland and a lower montane rain forest site (Lizano et al., in prep.; Moncada et al., in prep.). Based on these figures and the discussed uncertainties, the number of species per site observed in this study (as many as 84 for opportunistic and 99 for repetitive sampling) appear moderate, especially 113 compared to the figures by Aptroot (1997, 2001, 2002), Komposch and Hafellner (1999, 2000, 2002, 2003) and Martins (2006), who studied comparable vegetation types. On the other hand, the figures are similar to those reported by Pereira et al. (2005a–c) for three Atlantic rain forest remnants (including one site also studied here: Gurjaú), with 37, 45 and 53 species, respectively; these authors used the same opportunistic sampling technique. Assuming that opportunisting sampling underestimates lichen diversity by as much as 50% (mostly because inconspicuous, cryptic and frequently sterile taxa are not sampled; Cáceres et al. 2007b), and assuming that sampling of the forest understory only recovers a part of the diversity (25–75% depending on forest structure), the actual species numbers are probably 2–3 times as high as observed, which would result in estimated numbers, for the richest sites, of 180–250 species, within the same range as those reported by the aforementioned papers. Indeed, in a separate study using a quantitative approach sampling 47 trees in the understory at one site, a total of 150 species was found (Cáceres et al. 2007b), while opportunistic one-time sampling of that site yielded 53 species and repetitive sampling 99 species. Another explanation for the moderate species numbers is the potential effect of forest fragmentation, as observed in other organisms such as pteridophytes, trees and birds (Machado & Fonseca 2000; Paciencia & Prado 2005; Tabanez & Viana 2000). In studies in Amazonia, it has been shown that rain forest fragmentation specifically kills mature and large trees (Laurance et al. 2000), and since certain microlichen groups, such as Thelotremataceae (Rivas Plata et al. 2007), are dependent on such trees, fragmentation will eventually eliminate such lichen species. The stochastic component of such processes would result in higher dissimilarity between fragments, which is supported by the high beta-diversity values between sites observed in our study. High beta-diversity was more pronounced between Zona da Mata sites, compared to Caatinga and Brejos de Altitude sites, which agrees with the notion that anthropogenic fragmentation is more pronounced in the Zona da Mata, which naturally constitutes a continuous rain forest area (Machado & Fonseca 2000; Silva et al. 114 THE BRYOLOGIST 111(1): 2008 1998), whereas Brejos de Altitude sites have been naturally isolated during geological times; the latter are considered ‘‘Pleistocene refugia,’’ where species richness and composition were not significantly disturbed by the unfavorable climate conditions affecting their surrounding areas (Rodal 1998; Whitmore 1990). Caatinga lichens are more adapted to exposed conditions, for example, by frequently producing secondary cortical substances as demonstrated here, and hence less affected by anthropogenic disturbances than closed rain forest lichens. It can therefore be concluded that specifically in the Zona da Mata, forest fragmentation stochastically reduces lichen diversity within individual fragments, depending on the impact level, and at the same time increases beta-diversity between fragments. This underlines the need for protection of many different forest fragments in order to conserve high biodiversity (Machado & Fonseca 2000; Paciencia & Prado 2005; Tabanez & Viana 2000). Our study confirms findings of other studies in both temperate and tropical regions that different vegetation types support different lichen communities (Burgaz et al. 1994; Johnson 1981; Komposch & Hafellner 2003; Wolf 1993). In our case, strong differences were found between Zona da Mata and Caatinga sites, not only at different taxonomic levels but also regarding certain morphological, anatomical and chemical characters and character states. Because such characters often correlate with certain taxonomic levels, their evaluation as potential adaptations is limited. On the other side, the presence of predominant characters and character states in a speciose clade restricted to a certain vegetation type suggests that such characters can be considered preadaptations. In the present study, we found a distinct shift from Arthoniomycetidae, Chaetothyriomycetidae, Dothideomycetidae and Ostropales, with predominantly trentepohlian photobiont, squamulose-byssoid, often isidiate thalli lacking secondary substances, perithecia or perithecioid ascomata, and septate, hyaline ascospores, dominating Zona da Mata communities, towards Lecanoromycetidae (in particular Lecanoraceae and Physciaceae) and Pertusariales, with chlorococcoid photobionts, often sorediate thalli producing a wide array of cortical substances (atranorin, lichexanthone and other xanthones, pulvinic acid derivatives), apothecia, and often non-septate and/or dark brown ascospores, characterizing Caatinga communities. Most of the latter characters, in particular cortical substances and pigmented ascospores, are of advantage in an environment characterized by high solar insolation and extended drought. Therefore, certain similarity exists between Zona da Mata lichens from exposed secondary or canopy microhabitats and Caatinga communities, although overall they cluster with closed Zona da Mata communities. An interesting case is the abundant and widespread Pyrrhospora russula sensu lato, which was found at exposed Brejos de Altitude sites and on fallen canopy branches, as well as abundantly in the Caatinga. Recent revision (Kalb, pers. comm. 2006) divides the taxon into two species, P. haematites with norstictic acid and P. russula with fumarprotocetraric acid (Cáceres 2007). While P. haematites included all Caatinga and exposed Brejos collections (which are dominated by Caatinga lichens), P. russula was only found on a fallen canopy branch in a Brejos de Altitude site, underlining the difference between the two vegetation types. The fact that Brejos de Altitude sites are much more similar to Zona da Mata sites, although they are located within Caatinga vegetation, supports their interpretation as isolated Pleistocene remnants of the Atlantic rain forest (Rodal 1998; Whitmore 1990). Only ten (2%) of the reported 456 taxa are shared between all three vegetation types, which means that each vegetation type contributes substantially to the overall lichen diversity of northeastern Brazil. The relatively low number of species shared between Zona da Mata and Brejos de Altitude sites suggests that, apart from the interpretation of the latter as Pleistocene refugia, major climatic differences exist between the two types, Brejos de Altitude being more similar to montane rain forests in their precipitation regime (Rodal 1998). This is true even if the overall altitude of the studied Brejos does not exceed 800 m, because of the so-called Massenerhebungseffekt (‘‘mass elevation effect’’), that compresses altitudinal zonation of vegetation belts in low mountains close to coastal areas (Grubb 1971). Although none of the Cáceres et al.: Microlichens in NE Brazil taxa reported for the Zona da Mata is significantly indicative for that vegetation, because of the high number of sites compared to other vegetation types, many are representative for this biome. Most Arthonia, Coenogonium, Malcolmiella and Porina species are restricted to the Zona da Mata, and other exclusive genera and species include Letroutia domingensis, L. vulpina, Phaeographis crispatula, Pyrenula mamillana and P. nitidula, among others. Brejos de Altitude sites are characterized by a high number of indicative species, including Lecanactis epileuca, Sagenidiopsis undulata and Lopezaria versicolor, the latter being a typical montane taxon. Baculifera spp., Cratiria spp., Dirinaria spp., Haematomma persoonii, Lecanora spp., Maronina multifera, Pertusaria spp. and Pyrrhospora haematites, are significant indicator species for Caatinga sites. Although indicative for these vegetation types, most of these species are widespread in the tropics and found in similar habitats, such as the Amazon lowland rain forest (Zona da Mata lichens), the lower Andes (many Brejos de Altitude lichens) and dry forests and savannas in Central and South America and the Cerrado in central Brazil (most Caatinga lichens). Among the 456 species found in the study area, only 18 (4%) are new and potentially endemic, almost all found in Zona da Mata and Brejos de Altitude sites (Cáceres 2007). Thus, while each site and vegetation type substantially contributes to the overall lichen diversity in the study area, the taxonomic contribution of that area to neotropical lichen diversity is relatively low. ACKNOWLEDGMENTS The work was supported by a Ph.D. grant to the first author from the Deutscher Akademischer Austauschdienst (DAAD). The first author thanks Leonor Costa Maia and her staff at the Departamento de Micologia, Centro de Ciências Biológicas, Universidade Federal de Pernambuco, for logistic and organizational support in order to access the localities sampled. Franciso Quintella and his wife, owners of the RPPN Fazenda São Pedro, are warmly acknowledged for their help during the repeated visits to the site. Tatiana Gibertoni provided welcome company and transportation during visits to several localities. Drs. André Aptroot, Klaus Kalb and Harrie Sipman assisted in the identification or confirmation of certain lichen groups. Dr. Luciana Zedda is thanked for revising the manuscript and giving some useful suggestions concerning the statistical analyses. 115 LITERATURE CITED Ahti, T. 2000. Cladoniaceae. Flora Neotropica Monograph 78: i–iv, 1–363. Andrade-Lima, D. 1954. Contribution to the study of the flora of Pernambuco, Brazil. Universidade Rural de Pernambuco Monografia 1: 1–154. ———. 1961. Tipos de floresta de Pernambuco. Anais da Associação de Geógrafos do Brasil 12: 69–85. Aptroot, A. 1997. Lichen biodiversity in Papua New Guinea, with the report of 173 species on one tree. Bibliotheca Lichenologica 68: 203–213. ———. 2001. Lichenized and saprobic fungal biodiversity of a single Elaeocarpus tree in Papua New Guinea, with the report of 200 species of ascomycetes associated with one tree. Fungal Diversity 6: 1–11. ———. 2002. New and interesting lichens and lichenicolous fungi in Brazil. Fungal Diversity 9: 15–45. Barros, L. M. & L. Xavier-Filho. 1972. Catálogo dos liquens do herbário do Departamento de Botânica da Universidade Federal de Pernambuco. Pages 45–55. In XXIII Congreso Nacional de Botânica, Garanhuns, PE, Brazil. Brako, L. 1991. Phyllopsora (Bacidiaceae). Flora Neotropica Monograph 55: 1–66. Burgaz, A. R., E. Fuertes & A. Escudero. 1994. Ecology of cryptogamic epiphytes and their communities in deciduous forests in Mediterranean Spain. Vegetatio 112: 73–86. Cáceres, M. E. S. 1999. A new foliicolous Byssoloma (lichenized Ascomycetes: Pilocarpaceae) from the Atlantic rainforest in Pernambuco, Brazil. Mycotaxon 71: 383–386. ———. 2007. Corticolous crustose and microfoliose lichens of northeastern Brazil. Libri Botanici 22: 1–168. ——— & R. Lücking. 2000. Three new species and one new combination of foliicolous lichens and lichenicolous fungi from the Atlantic rainforest in Pernambuco state, Brazil. Nova Hedwigia 70: 217–226. ———, ——— & G. Rambold. 2007a. Phorophyte specificity and environmental parameters versus stochasticity as determinants for species composition of corticolous crustose lichen communities in the Atlantic rainforest of northeastern Brazil. Mycological Progress (in press). ———, ——— & ———. 2007b. Efficiency of sampling methods for accurate estimation of species richness: Corticolous microlichens in the Atlantic rain forest of northeastern Brazil. Ecotropica (in press). ———, L. C. Maia & R. Lücking. 2000. Foliicolous lichens and their lichenicolous fungi in the Atlantic rain forest of Brazil: diversity, ecogeography and conservation. Bibliotheca Lichenologica 75: 47–70. Cardoso Silva, J. M. & M. Tabarelli. 2000. Tree species impoverishment and the future flora of the Atlantic forest of northeast Brazil. Nature 403: 72–74. Cornelissen, J. H. C. & H. Ter Steege. 1989. Distribution and ecology of epiphytic bryophytes and lichens in dry 116 THE BRYOLOGIST 111(1): 2008 evergreen forest of Guyana. Journal of Tropical Ecology 5: 131–150. Costa, L. P., Y. L. R. Leite, G. A. B. Fonseca & M. T. Fonseca. 2000. Biogeography of South American forest mammals: endemism and diversity in the Atlantic forest. Biotropica 32: 872–881. FIDEM. 1987. Região Metropolitana do Recife: Reservas Ecológicas. Recife, Governo do Estado de Pernambuco, Secretaria de Planejamento do Estado de Pernambuco. Fundação de Desenvolvimento da Região Metropolitana do Recife. Frisch, A., K. Kalb & M. Grube. 2006. Contributions towards a new systematics of the lichen family Thelotremataceae. Bibliotheca Lichenologica 92: 1–556. Galindo-Leal, C. & I. G. Câmara (eds.). 2003. The Atlantic Forest of South America: Biodiversity Status, Threats, and Outlook. Center for Applied Biodiversity Science and Island Press, Washington, D.C. Grubb, P. J. 1971. Interpretation of the ‘Massenerhebung’ effect on tropical mountains. Nature 229: 44–45. Harris, R. C. 1986. The family Trypetheliaceae (Loculoascomycetes: lichenized Melanommatales) in Amazonian Brazil. Acta Amazonica 14(1–2, Suplemento): 55–80. ———. 1989. A sketch of the family Pyrenulaceae (Melanommatales) in eastern North America. Memoirs of The New York Botanical Garden 49: 74–107. Holz, I. & S. R. Gradstein, S. R. 2005. Cryptogamic epiphytes in primary and recovering upper montane oak forests of Costa Rica—species richness, community composition and ecology. Plant Ecology 178: 89–109. Johnson, E. A. 1981. Vegetation organization and dynamics of lichen woodland communities in the Northwest Territories, Canada. Ecology 62: 200–215. Kalb, K. 1981. Lichenes Neotropici ausgegeben von Klaus Kalb. Fascikel I (No. 1–40). Neumarkt, 18 Sep 1981. ———. 1982a. Lichenes Neotropici ausgegeben von Klaus Kalb. Fascikel II (No. 41–80). Neumarkt/Opf., 22 Feb 1982. ———. 1982b. Lichenes Neotropici ausgegeben von Klaus Kalb. Fascikel III (No. 81–120). Neumarkt/Opf., 2 Apr 1982. ———. 1982c. Lichenes Neotropici ausgegeben von Klaus Kalb. Fascikel IV (No. 121–160). Neumarkt/Opf., 16 Aug 1982. ———. 1982d. Lichenes Neotropici ausgegeben von Klaus Kalb. Fascikel V (No. 161–200). Newmarkt/Opf. 19 Nov 1982. ———. 1983. Lichenes Neotropici ausgegeben von Klaus Kalb. Fascikel VI (No. 201–250). Neumarkt/OPf., 7 Apr 1983. ———. 1984. Lichenes Neotropici ausgegeben von Klaus Kalb. Fascikle VIII (No. 301–350). Neumarkt/OPf. ———. 1987. Brasilianische Flechten. 1. Die Gattung Pyxine. Bibliotheca Lichenologica 24: 1–89. ———. 2001. New or otherwise interesting lichens. I. Bibliotheca Lichenologica 78: 141–167. ———. 2004. New or otherwise interesting lichens II. Bibliotheca Lichenologica 88: 301–329. ——— & J. A. Elix. 1995. The lichen genus Physcidia. Bibliotheca Lichenologica 57: 265–296. ———, R. Lücking & E. Sérusiaux. 2000. Studies in Bacidia sensu lato (lichenized Ascomycetes: Lecanorales). I. The genus Bapalmuia. Mycotaxon 75: 281–309. ———, B. Staiger & J. A. Elix. 2004. A monograph of the lichen genus Diorygma—a first attempt. Symbolae Botanicae Upsalienses 34(1): 133–181. Kashiwadani, H. & K. Kalb. 1993. The genus Ramalina in Brazil. Lichenologist 25: 1–31. Komposch, H. & J. Hafellner. 1999. List of lichenized fungi so far observed in the tropical lowland rain forest plot Surumoni (Venezuela, Estado Amazonas). Fritschiana 19: 1–10. ——— & ———. 2000. Diversity and vertical distribution of lichens in a Venezuelan tropical lowland rain forest. Selbyana 21: 11–24. ——— & ———. 2002. Life form diversity of lichenized fungi in an Amazon lowland rain forest. Bibliotheca Lichenologica 82: 311–326. ——— & ———. 2003. Species composition of lichen dominated corticolous communities: a lowland rain forest canopy compared to an adjacent shrubland in Venezuela. Bibliotheca Lichenologica 86: 351–367. Laurance, W. F., P. Delamônica, S. G. Laurance, H. L. Vasconcelos & T. E. Lovejoy. 2000. Conservation: rainforest fragmentation kills big trees. Nature 404: 836. Lücking, R., M. E. S. Cáceres & L. Costa Maia. 1999. Revisão nomenclatural e taxonômica de liquens foliı́colas e respectivos fungos liquenı́colas registrados para o Estado de Pernambuco, Brasil, por Batista e colaboradores. Acta Botanica Brasilica 13: 115–128. ———, E. Sérusiaux, L. C. Maia & C. G. Pereira. 1998. A revision of the names of foliicolous lichenized fungi published by Batista and co-workers between 1960 and 1975. Lichenologist 30: 121–191. Machado, R. B. & G. A. B. Fonseca. 2000. The avifauna of Rio Doce Valley, southeastern Brazil, a highly fragmented area. Biotropica 32: 914–924. Marcelli, M. P. 1992. Ecologia liquênica nos manguezais do sulsudeste brasileiro. Bibliotheca Lichenologica 47: i–vii, 1–288. ———. 1998. History and current knowledge of Brazilian lichenology. Pages 25–45. In M. P. Marcelli & M. R. D. Seaward (eds.), Lichenology in Latin America: History, Current Knowledge and Applications. Companhia de Tecnologia de Saneamento Ambiental, Estado de São Paulo, São Paulo. Cáceres et al.: Microlichens in NE Brazil Martins, S. M. A. 2006. Estudo da comunidade liquenizada epı́fita em Dodonaea viscosa L. na restinga do Parque Estadual de Itapuã, Viamão, RS. Ph.D. Thesis, Instituto de Botânica de São Paulo, Brazil. McCune, B., J. B. Grace & D. L. Urban. 2002. Analysis of Ecological Communities. MjM Software, Gleneden Beach, OR. ——— & M. J. Mefford. 1999. PC-ORD. Multivariate Analysis of Ecological Data, Version 4.0. MjM Software, Gleneden Beach, OR. Montfoort, D. & R. Ek. 1990. Vertical Distribution and Ecology of Epiphytic Bryophytes and Lichens in a Lowland Rainforest in French Guiana. Herbarium, Institute of Systematic Botany, Utrecht. Morellato, L. P. C. & C. F. B. Haddad. 2000. Introduction: the Brazilian Atlantic forest. Biotropica 32: 786–792. Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. Fonseca & J. Kent. 2000. Biodiversity hotspots for conservation priorities. Nature 403: 853–858. Nöske, N. M. 2004. Effekte anthropogener Störung auf die Diversität kryptogamischer Epiphyten (Flechten, Moose) in einem Bergregenwald in Südecuador. Ph.D. Thesis, Göttingen University, Germany. ——— & H. J. M. Sipman. 2004. Cryptogams of the Reserva Biológica San Francisco (Province Zamora-Chinchipe, southern Ecuador) II. Lichens. Cryptogamie, Mycologie 25: 91–100. Paciencia, M. L. B. & J. Prado. 2005. Effects of forest fragmentation on pteridophyte diversity in a tropical rain forest in Brazil. Plant Ecology 180: 87–104. Pereira, E. C., A. M. Silva, N. H. Silva & M. Marcelli. 2005a. Liquens corticı́colas de fragmentos de floresta Atlântica dos estados de Pernambuco e Alagoas. I: Reserva de Gurjaú. Page 43. In S. Eliasaro, C. G. Donha & M. Marcelli (eds.), VII Encontro de Grupo Latino Americano de Liquenólogos, Curitiba, Brasil, Programas e Resumos. Universidade Federal do Paraná. ———, ———, ——— & ———. 2005b. Liquens corticı́colas de fragmentos de floresta Atlântica dos estados de Pernambuco e Alagoas. II: Usina Serra Grande. Page 44. In S. Eliasaro, C. G. Donha & M. Marcelli (eds.), VII Encontro de Grupo Latino Americano de Liquenólogos, Curitiba, Brasil, Programas e Resumos. Universidade Federal do Paraná. ———, ———, ——— & ———. 2005c. Liquens corticı́colas de fragmentos de floresta Atlântica dos estados de Pernambuco e Alagoas. III: Usina Colônia. Page 43. In S. Eliasaro, C. G. Donha & M. Marcelli (eds.), VII Encontro de Grupo Latino Americano de Liquenólogos, Curitiba, Brasil, Programas e Resumos. Universidade Federal do Paraná. Ranta, P., T. Blom, J. Niemela, E. Joensuu & M. Siitonen. 1998. The fragmented Atlantic rain forest of Brazil: size, shape 117 and distribution of forest fragments. Biodiversity and Conservation 7: 385–403. Rivas Plata, E., R. Lücking & H. T. Lumbsch. 2007. When family matters: an analysis of Thelotremataceae (lichenized Ascomycota: Ostropales) as bioindicators of ecological continuity in tropical forests. Biodiversity and Conservation (in press). Rizzini, C. T. 1977. Tratado de Fitogeografia do Brasil. 2nd Ed. Âmbito Cultural Edições Ltda., Rio de Janeiro. Rodal, M. J. N. 1998. Florestas Serranas de Pernambuco: Localização e Conservação dos Remanescentes de Brejos de Altitude. Imprensa Universitária, UFPE, Recife. Silva, M. & D. W. Minter. 1995. Fungi from Brazil, Recorded by Batista and Co-workers. CAB International, Wallingford. Silva Filho, A. A., E. R. Toniolo, M. Gabı́nio & S. F. S. Oliveira. 1998. Mapeamento da cobertura florestal nativa lenhosa do estado de Pernambuco. Documento de Campo FAO 17: 1–32. Sipman, H. J. M. 1996. Corticolous lichens. In S. R. Gradstein, P. Hietz, R. Lücking, A. Lücking, H. J. M. Sipman, H. F. M. Vester, J. H. D. Wolf & E. Gardette (eds.), How to sample the epiphytic diversity of tropical rainforests. Ecotropica 2: 59–72. Sparrius, L. B. 2004. A monograph of Enterographa and Sclerophyton. Bibliotheca Lichenologica 89: 1–141. Staiger, B. 2002. Die Flechtenfamilie Graphidaceae. Studien in Richtung einer natürlicheren Gliederung. Bibliotheca Lichenologica 85: 1–526. ——— & K. Kalb. 1995. Haematomma-Studien. I. Die Flechtengattung Haematomma. Bibliotheca Lichenologica 59: 1–198. ——— & ———. 1999. Acanthothecis and other graphidioid lichens with warty periphysoids or paraphysis-tips. Mycotaxon 73: 69–134. Tabanez, A. A. J. & V. M. Viana. 2000. Patch structure within Brazilian Atlantic forest fragments and implications for conservation. Biotropica 32: 925–933. Tabarelli, M., L. P. Pinto, J. M. C. Silva, M. Hirota & L. Bedê. 2005. Challenges and opportunities for biodiversity conservation in the Brazilian Atlantic forest. Conservation Biology 19: 695–700. Tehler, A. 1993. Schismatomma and three new or reinstated genera, a reassessment of generic relationships in Arthoniales. Cryptogamic Botany 3: 139–151. Tibell, L. 1996. Caliciales. Flora Neotropica Monograph 69: 1–78. Whitmore, T. C. 1990. An Introduction to Tropical Rainforests. Clarendon Press, Oxford. Wolf, J. H. D. 1993. Diversity patterns and biomass of epiphytic bryophytes and lichens along an altitudinal gradient in the northern Andes. Annals of the Missouri Botanical Garden 80: 928–960. ms. received January 16, 2007; accepted June 6, 2007.