New record and distribution map of Parmotrema rubifaciens ( Parmeliaceae , Ascomycota ) in the Neotropics

Parmotrema rubifaciens is recorded in this paper for the first time for Paraguay, expanding its known geographic distribution approximately 430 km southeast from its nearest locality. The species is characterized, illustrated, and compared with most related taxa. In addition, a map of its geographic distribution in the Neotropics is presented.


Introduction
In the humid tropics, due to deforestation of primary forests, secondary forests and plantations are becoming increasingly widespread land-use systems in human dominated areas (Barlow et al. 2007).Despite their quick expansion and potential importance, the biodiversity conservation values of secondary and plantation forests remain poorly understood (Hartley 2002;Dunn 2004), especially in relation to butterfly diversity.Along with the availability of larval and adult food plants, habitat quality appeared to be one of the most important parameters that is used to determine butterfly community structure (Barlow et al. 2007).However, several studies (Bowman et al. 1990;Lawton et al. 1998;Ramos 2000) have discussed the potential of butterfly diversity in secondary forests, but diversity and species richness of butterflies across different secondary vegetation gradients were poorly understood.
Among insects‚ butterflies are ideal subject for ecological studies of landscapes (Thomas and Malorie 1985), and their value as indicators of biotope quality is being increasingly recognized because of their sensitivity to minor changes in micro-habitat, in particular, light levels (Kremen 1992).To a large extent, butterflies (being a pollinating agent) contribute to the growth, maintenance and expansion of flora in the tropical regions where these insects show high abundance and species diversity (Bonebrake et al. 2010).
The northeastern region of India is home to a rich diversity of butterflies and other insects, due to vegetative richness (Alfred et al. 2002;Majumder et al. 2011), and it is also globally recognized as one of 25 biodiversity hotspots (Myers et al. 2000).Nonetheless, its biodiversity is under threat due to deforestation and habitat modification.A perusal of the literature suggests that 76 species of butterflies were previously recorded from the Variation in butterfly diversity and unique species richness along different habitats in Trishna Wildlife Sanctuary, Tripura, northeast India from November through February.Annual average rainfall is 255.89 mm.Temperatures vary from 6.8°C in January to 37.7°C in June.The sanctuary is well-known for its Asian population of Bison (Bos garus Smith) and migratory water birds in winter among many other wild animals.The present study is funded by Japan International Cooperation Agency (JICA).

Sampling regime
Field surveying of butterfly fauna was conducted from May 2010 to October 2010, following Modified Pollard Walk Method (MPWM) (Pollard 1977) in four distinct habitats of TWS.Four Permanent Line Transects (PLTs) (approx. 1 km long and 5 m wide) were laid based on floral composition in the four habitat types, namely: mature secondary mixed moist deciduous forest (TWS I); regenerated secondary mixed moist deciduous forest (TWS II); secondary mixed moist deciduous forest with bamboo patches (TWS III) and a grassland of exotic species (Pennisetum purpureum Schumach) (TWS IV) (Table 1).The degree of anthropogenic pressure is distinct among the four habitats in the study area.Butterflies were counted during sunny days at a constant speed in each transect from 8 am to 12 am local time for four consecutive days.This was repeated at 30 day intervals, maintaining the same spatial scale in each of the four sampling sites.Collected butterflies were identified using field guides (Haribal 1992;Kunte 2000;Kehimkar 2008), and followed the classification given by Ackery (1984).Vouchers of collected specimens are maintained in the Department of Zoology, Tripura University.

Data Analysis
The Shannon diversity index was applied to estimate butterfly species diversity along the habitats (Shannon and Wiener 1949).This index was calculated by the equation H s = -∑p i In p i .Where, p i is the proportion of individuals found in the ith species and 'In' denotes the natural logarithm.Species dominance across habitats was estimated by Simpson's dominance index (Simpson 1949).This index was used to determine the proportion of more common species in a community or an area by the following formula D s = ∑ s i=1 [n i (n i -1)]/[N (N-1)] where, n i is the population density of the ith species, and N is the total population density of all component species in each site.Comparisons of butterfly species composition between different forest habitats was estimated using single linkage cluster analysis based on Bray-Curtis similarity (McAleece 1998).Biodiversity Pro version 2 (Lambshead et al. 1997)  was used for data analysis.

Results and Discussion
During the systematic survey, a total of 1005 individuals of 59 species of butterfly belonging to 48 genera and five families were recorded from the four habitat types at TWS (Table 2, 4).The 59 species recorded in the present study compared suitably with the 72 species recorded in another study of northeast India (Ali and Basistha 2000) that showed more or less similarity in habitats and climatic conditions.Among the families, Nymphalidae was dominant with 23 species followed by the Lycaenidae (13 species), Papilionidae (8 species), Hesperiidae (8 species) and Pieridae (7 species) (Table 2).Members of the Nymphalidae were always dominant in the tropical region because most of the species are polyphagous in nature, consequently helping them to live in all the habitats.Additionally, many species of this family are strong, active fliers that might help them in searching for resources in large areas (Eswaran and Pramod 2005;Krishna Kumar et al. 2007;Raut and Pendharkar 2010;Padhye et al. 2006).
A high proportion of nymphalid species indicates high host plant richness in the Trishna sanctuary area.Notwithstanding, all the recorded species are widely distributed in India (Wynter-Blyth 1957) and only Troides helena L. is listed as an endangered species in CITES Appendix II (Collins and Morris 1985).However, 9 species: viz.Lethe europa (F.), Cepora nerissa (F.), Castalius rosimon (F.), Narathura selta (de Niceville), Pantoporia hordonia (stoll), Megisba Malaya (Horsfield), Arhopala fulla (Hewitson), Elymnias malelas (Hewitson) and Baoris farri (Moore) are listed in the Indian Wildlife Protection Act of 1972 as under Schedule I and II (Anonymous 2006).An aphidophagous butterfly, Spalgis epius Westwood (Kehimkar 2008), belonging to the family Lycaenidae is recorded from the sanctuary.Out of 59 species, 21 species (35.59%) have been recorded from only a single habitat type, and hence, referred to as unique species (Table 4).TWLS IV contain one of these [Spalgis epius (Westwood)], whereas TWLS I, TWLS II and TWLS III contain seven, five, and eight unique species, respectively.Results suggested that the structural complexity and vegetation diversity of each habitat type might facilitate a definite set of microhabitats suitable for a particular species.Thirty one species (52.54%) were recorded for the first time in the state (Table 4).
High species richness of butterflies in secondary forest habitats was reported by earlier workers (Bowman et al. 1990;Lawton et al. 1998;Ramos 2000), and several studies revealed that habitat specificity is directly linked to the availability of host plants for larvae and adults (Grossmueller and Lederhouse 1987;Thomas 1995).In general, all four habitats showed high species richness and diversity index values.Particularly, among the study sites, TWS I ranked highest based on species richness and diversity indices (D s; H s ) followed by TWS II, TWS III and TWS IV (Table 3, Table 4).Species richness and diversity of butterfly in the regenerated secondary forest (TWS II) is also high because of rapid vegetative succession that provides suitable foliage and nectar for larval and adult stages of butterflies.Moreover, some butterflies (Papilio demoleus Linnaeus, Graphium agamemnon Linnaeus, Castalius rosimon Fruhstofer; Eurema hecabe (Moore) were observed in the core area of TWS II, mudpuddling in the wild boar (Sus scrofa Linnaeus) inhabited sites.Pits created by wild boars serve as collection sites of mineralrich feces and urine perfect for mudpuddler butterflies, especially females that require minerals for reproduction (Ramos 1996).TWS III showed moderate species richness and diversity due to half of the sampling area being rich in secondary vegetation, and the remaining half being dominated by bamboo patches.Most of the species recorded in TWS III are found in the secondary vegetation and only a few species Lethe europa (Fabricius), Lethe mekera (Moore) and Mycalesis perseus (Fabricius) were confined to bamboo patches because they prefer bamboo as a host plant (Kehimkar 2008).Comparatively, TWS IV (grassland of exotic species) showed poor species richness and diversity of butterflies.A similar impact of exotic species was found in the case of many native herbivores, particularly those species for which native plants serve as a potential food plant (Nagy et al. 1998).The cultivation of exotic grass species will largely create a problem for the host-specific butterfly species of the sanctuary that chiefly rely on locally available plant species for their survival.Among the four habitats, TWS I showed low dominance value (0.038) followed by TWS II (0.041), TWS III (0.045) and TWS IV (0.054), respectively (Table 3).The low  Cluster analysis based on the Bray-Curtis single linkage similarity value revealed the percent similarity between butterfly species composition across the four habitat types.The open habitat of long exotic grass (TWS IV) stood out clearly from the other three habitats and showed linkage at 50.86% (which represent lowest) similarity.Degraded forest habitat (TWS III) was linked at 58.16% similarity to the cluster of habitats of primary forest (TWS I) and regenerated forest (TWS II) which showed highest similarity in species composition (70.55%) (Figure 2).Results showed that more than 50% of the butterfly species recorded in all the sampling sites were the same despite differences in habitat characteristics.Study conducted by Novotny et al. (2007) in tropical forests showed low beta diversity of herbivorous insects.The high similarity value of butterfly fauna between different habitats of TWS is an indicator of low beta diversity in this small forest area which in turn indicates availability of more or less similar niches in each of the studied habitats.This fact is also supported by Proctor (1986), who opined that tropical forests contain more microhabitats per unit ground area than their temperate counterparts.However, Klopfer and MacArthur (1961) suggested that in tropical forests species may reside not in the number of niches available, but in an increase in the similarity of coexisting species.The extent to which all these informal explanations apply is a matter of further study at micro-habitat level.
The present study confirmed the existence of a wide diversity of butterflies in the selected habitat types of Trishna Wildlife Sanctuary and indicates specifically that mature secondary and regenerated forests had the highest butterfly diversity and species richness, while exotic grasslands have negative influences on local butterfly community structure.However, to understand the influence of different landscape elements on butterfly community structure, long-term butterfly censusing and monitoring (both temporally and spatially) is required in Trishna Wildlife Sanctuary.Moreover, the presence of 9 threatened species and 21 unique butterfly species in the sanctuary makes it an important butterfly habitat in the state for future conservation and management programs.Acknowledgments: Authors are thankful to Dr. Durgadas Ghosh, Department of Zoology, Tripura University for his encouragement during the study.Thanks are also due to Koushik Majumdar, Department of Botany, Tripura University for identification of exotic grass species and his valuable suggestions at the time of manuscript preparation.We thank the wildlife warden for the permission and accommodation in the Trishna wildlife sanctuary.Thanks to two anonymous reviewer for their valuable comments on the earlier draft of the manuscript.Special thanks are extended to Dipankar Kishore Sinha for his help during the field survey.

Figure 2 .
Figure 2. Single linkage cluster analysis between habitats based on Bray-Curtis similarity.

Table 1 .
Characteristics of the four habitat types at Trishna Wildlife Sanctuary.

Table 2 .
Family wise composition of butterflies showing number of genera, species and individuals recorded from Trishna Wildlife Sanctuary.

Table 3 .
Diversity indices of butterfly communities in four habitats of Trishna Wildlife Sanctuary.
dominance value of TWS I indicate that butterfly species are more or less evenly distributed in terms of abundance compared to TWS II, TWS III and TWS IV.The present results indicate an availability of host plants for each butterfly species and efficient resource sharing by each species recorded in the TWS I.

Table 4 .
List of butterfly species and their abundance recorded in Trishna Wildlife Sanctuary.