DOI: 10.13102/sociobiology.v62i1.70-75Sociobiology 62(1): 70-75 (March, 2015) Open access journal: http://periodicos.uefs.br/ojs/index.php/sociobiology ISSN: 0361-6525 Soil-sampled termites in Two Contrasting Ecosystems within the Semiarid Domain in Northeastern Brazil: Abundance, Biomass, and Seasonal Influences Introduction Termites are among the most abundant arthropods in tropical ecosystems, and their densities can exceed 8000 individuals.m-² (Martius, 1994; Bignell & Eggleton, 2000). Termites have very important functions in most terrestrial ecosystems due to their roles in decomposition processes and in energy and nutrient fluxes (Matsumoto, 1976; Bignell & Eggleton, 2000; Bandeira & Vasconcellos, 2002). These organisms also influence the soil by redistributing material both horizontally and vertically and by increasing soil porosity (Lee & Wood, 1971; Wood & Sands, 1978; Jones et al., 1994). It is estimated that up to 50% of the decomposition of Abstract Termites are important in tropical terrestrial ecosystems due to their role in fragmenting organic material and thus in energy and nutrient flows. The present study compared soil- sampled termites in two contrasting ecosystems in the Brazilian semiarid domain: humid highland forests (“Brejo de Altitude”) and seasonally dry tropical forests (“Caatinga”). Thirty soil core samples (12 L each), divided into three soil levels (A: 0-10; B:10-20; C: 20-30 cm), were analyzed in each area, 15 during the rainy season and 15 during the dry season. Twenty- two termite species were encountered in the soil core samples, belonging to the Termitidae (20 species) and Rhinotermitidae (2 species). The abundance and biomass of termites in the humid highland forests were significantly greater during the rainy season (6080 individuals.m-² and 18.53 g.m-² fresh weight) than in the dry season (2215 individuals.m-² and 5.45 g.m-² fresh weight). The greatest abundance and biomass was encountered at the 0 and 20 cm depths. Abundance and biomass did not differ significantly among the different seasons in the seasonally dry tropical forests (rainy season: 862 individuals.m-² and 1.4 g m-² fresh weight; dry season: 250 individuals.m-² and 0.45 g m-² fresh weight) nor among the different soil levels. The observed increase in termite numbers and biomass during the rainy season in the humid highland forest indicates that their decomposition activities are asymmetrical during the year. The quantitative similarity of termite numbers and fresh weights during the year in the seasonally dry tropical forest area suggests that other factors beyond climatic differences influence the abundance and biomass of these termites, at least at a local scale. Sociobiology An international journal on social insects VFP Araújo, MP Silva, A Vasconcellos Article History Edited by Reginaldo Constantino, UnB, Brazil Received 20 April 2014 Initial acceptance 21 June 2014 Final acceptance 19 August 2014 Keywords Isoptera, vertical stratification, seasonally dry tropical forests, humid highland forests Corresponding author Alexandre Vasconcellos Laboratório de Termitologia Departamento de Sistemática e Ecologia Centro de Ciências Exatas e da Natureza Universidade Federal da Paraíba 58051-900, João Pessoa, Paraíba, Brazil E-mail: avasconcellos@dse.ufpb.br plant organic detritus in tropical forests can be attributed to termites (Bignell & Eggleton, 2000). More water is retained in the soil in arid and semiarid ecosystems due to the activities of termites, which is believed to be directly reflected in vegetation structures and local primary productivity (Holt & Conventry, 1990; Whitford, 1991). Though few, quantitative studies at different times of the year in the same ecosystem have revealed seasonal variations in abundance and assemblage composition that may impact the temporal contributions of termites to processes such as organic decomposition (Matsumoto, 1976; Matsumoto & Abe, 1979; Collins, 1989; Dibog et al., 1998; Martius, 1998; Vasconcellos, 2010). RESEARCH ARTICLE - TERMITES Universidade Federal da Paraíba, João Pessoa, PB, Brazil Sociobiology 62(1): 70-75 (March, 2015) 71 Seasonally dry tropical forests are characteristic of tropical environments with marked water restrictions during the dry season (Dirzo et al., 2011). These dry forests occur widely in the semiarid region of northeastern Brazil, which itself covers an area of approximately 735,000 km2 extending from 2° 54’ W to 17° 21’ S (Andrade-Lima, 1981; Prado, 2003). This region is characterized by high temperatures and high evapotranspiration potentials that compound the effects of the low and irregular rainfall rates – resulting in low water availability in the soil for 7 to 11 months (Andrade- Lima, 1981). However, the regional landscape is actually a mosaic of different phytophysiognomies such as seasonally dry tropical arboreal, shrub, and spiny forests that are adapted to these very dry conditions (Coimbra-Filho & Câmara, 1996). Additionally, there are “islands” of highland humid forest within the semiarid region of northeastern Brazil that are associated with landscapes that rise above 600 m, locally known as “Brejos de Altitude” (Andrade-Lima, 1982; Tabarelli & Santos, 2004). These highlands and plateaus offer more amenable environmental conditions, with greater soil and atmospheric humidity and denser vegetation cover (Andrade- Lima, 1966; Tabarelli & Santos, 2004; Rodal & Sales, 2008). As such, these unique landscapes function as biodiversity refuges, especially in the dry season, for organisms having a certain degree of mobility (Andrade-Lima, 1981). The present study sought to quantify the abundance, biomass, and vertical stratification of soil termites in two areas within the Brazilian semiarid domain – one with seasonally dry tropical forest vegetation (Caatinga), and the other with highland humid forest vegetation (Brejos de Altitude) – to determine the influences of the dry and rainy seasons on these arthropods. Materials and Methods The study sites were located in the Fazenda Almas Private Nature Reserve (seasonally dry tropical forest) and in the Mata do Pau Ferro State Ecological Reserve (highland humid forest) in Paraiba State, Brazil. The Fazenda Almas is a private protected area located in the municipality of São José dos Cordeiros (07°28’ S 36°52’ W) that encompasses 3,505 ha of seasonally dry tropical forest vegetation at altitudes ranging from 600 to 720 meters a.s.l. The mean annual precipitation, temperature, and humidity there are 560 mm ± 230 (sd), 25 oC, and 65% respectively (Governo do Estado da Paraíba, 1985), with rainfall concentrated between February and April. The Mata do Pau Ferro ecological reserve is located in the municipality of Areia (6° 58’ S, 35° 42’ W) and encompasses 607 ha of ombrophilous forest communities at altitudes between 400 and 600 m a.s.l. (Oliveira et al., 2006). It is situated within a micro-region of montane forests with mean annual rainfall, temperatures, and humidity of 1400 mm, 22 °C, and 85% respectively (Embrapa, 1972). Termites were collected in the Mata do Pau Ferro site during December 1998 and January 1999 during the dry season, and in June and July 1999 during the rainy season. Collections at the Fazenda Almas site were undertaken in May 2007 and November 2008 during the rainy and dry seasons respectively. The difference of the sampling period between the areas was about 8 years. However, for over 30 years, the areas have been free of human disturbance. Fifteen collection points were established at minimum distances of 10 m from each other, spaced along a transect, and termites were collected using the methodology proposed by the Tropical Soil Biology and Fertility Program (TSBF) (Anderson & Ingram, 1989), as modified by Silva and Bandeira (1999). Blocks of soil (20 cm x 20 cm x 30 cm: 12 liters) were extracted at each collection point and divided into three strata: A (0-10 cm), B (10-20 cm), and C (20-30 cm), resulting in 45 samples each season (90 for each forest type). The three different soil strata were separated and broken up in plastic trays, and the termites encountered were manually collected and preserved in 75% alcohol. The specimens were removed to the laboratory to be counted and weighed (fresh weight in g.m-²). The specimens were subsequently deposited in the permanent collection of the Termite Laboratory at the Federal University of Paraiba. The abundance and biomass data were log (x+1) transformed to guarantee the normality and homogeneity of variance. The t-test was performed to compare abundance and biomass of the dry tropical forest and humid highland forest sites. Analysis of variance (Factorial ANOVA), with the a posteriori Tukey test, was performed to determine any significant differences between the seasons and between the different strata within each study area. The statistical tests were performed using STATISTIC 5.0 software (Statsoft, 1995). Results Twenty-two termite species were encountered overall, belonging to 13 genera comprising 20 species of the family Termitidae and distributed among the subfamilies Apicotermitinae (7 species), Nasutitermitinae (6), Syntermitinae (3), and Termitinae (3), and two species of the family Rhinotermitidae. Only Diversitermes diversimilis (Silvestri) and Amitermes amifer Silvestri were encountered in both forest types (Table 1). Abundance (t test = - 6,7; df = 178; P < 0.001 ) and biomass (t test = - 5,7; df = 178; P < 0.001 ) were significantly higher in the humid highland forest. The Fazenda Almas site (seasonally dry tropical forest) yielded 10 termite species distributed among seven genera, of which eight species were encountered during the rainy season and seven during the dry season. The species Anoplotermes sp. 3, Grigiotermes sp. 2, Ruptitermes silvestrii (Emerson), and one unidentified species (of the family Termitidae), were only encountered in the soil during the rainy season. Constrictotermes cyphergaster (Silvestri) and Heterotermes sulcatus Mathews, on the other hand, were only encountered during the dry season. Amitermes amifer was predominant during both seasons in terms of both its abundance and biomass VFP ArAújo, MP SilVA, A VASconcelloS – Soil Termites in the Brazilian Semiarid Domain72 (Table 1). During the rainy season, the mean abundance and biomass of the termites was 862 ± 532 individuals.m-² and 1.40 ± 1.02 g.m-² (fresh weight), respectively, while during the dry season these values were 250 ± 195 individuals.m-² and 0.45 ± 0.38 g.m-² (fresh weight) respectively. Nonetheless, termite abundance and biomass did not differ significantly between the two seasons in this vegetation type (F1.84=1.91; P=0.17 and F1.84=0.93; P=0.33 respectively), nor between the different soil levels (F2.84=1.06; P=0.35 and F2.84=0.65; P=0.52 respectively). The same results were encountered when considering the interactive effects of the seasons and the soil strata on the termite fauna (F2.84=0.87; P=0.41 and F2.84=0.07; P=0.92 respectively) (Fig 1). The Mata do Pau Ferro site (humid highland forests) yielded 14 termite species distributed among 11 genera, of which nine were encountered during the rainy season and 12 in the dry season. Nasutitermes jaraguae (Holmgren) and Neocapritermes sp. were encountered only during the rainy season, while five species were only observed during the dry season (Diversitermes diversimiles, Embiratermes neotenicus (Holmgren), Ibitermes inflatus Vasconcellos, Subulitermes microsoma (Silvestri), and Amitermes amifer). The termite species Grigiotermes sp. 1 and Embiratermes parvirostris Constantino were predominant during both seasons, in terms of both their abundance and biomass (Table 1). A total of 6080 ± 1477 individuals.m-² and 18.53 ± 6.18 g.m-² (fresh Taxa Mata do Pau Ferro Fazenda Almas Abundance Biomass Layer Abundance Biomass Layer Feeding groupsRainy Dry Rainy Dry Rainy Dry Rainy Dry TERMITIDAE Apicotermitinae Anoplotermes sp 1 234 189 0.44 0.36 A,B,C S Anoplotermes sp.2 331 243 0.62 0.46 A,B,C S Anoplotermes sp.3 6 0.01 B S Grigiotermes sp. 1 1044 250 4.18 1.00 A,B,C S Grigiotermes sp. 2 5 0.01 A S Ruptitermes reconditus (Silvestri) 332 38 1.23 0.14 B,C L Ruptitermes silvestrii (Emerson) 5 0.02 A L Nasutitermitinae Constrictotermes cyphergaster (Silvestri) 9 0.02 A,B W Diversitermes diversimiles (Silvestri) 50 0.10 A 4 4 0.00 0.00 A,C W/L Nasutitermes callimorphus Mathews 226 23 0.58 0.06 A W Nasutitermes jaraguae (Snyder) 105 0.30 A,B W Subulitermes microsoma (Silvestri) 28 0.04 A S Velocitermes sp. 306 2 0.38 0.00 A L Syntermitinae Embiratermes neotenicus (Holmgren) 10 0.04 A,B W/S Embiratermes parvirostris Constantino 989 242 1.56 0.38 A,B,C S Ibitermes inflatus Vasconcellos 224 1.36 A,B S Termitinae Amitermes amifer Silvestri 30 0.09 A,B 314 118 0.61 0.23 A,B,C W/S Amitermes nordestinus Melo and Fontes 23 2 0.02 0.00 B,C W/S Neocapritermes sp. 81 0.34 A,B S Termitidae sp. 7 0.01 A RHINOTERMITIDAE Heterotermes longiceps (Snyder) 68 2 0.07 0.00 A,B W Heterotermes sulcatus Mathews 15 0.02 C W Total 3648 1329 11.12 2.54 432 150 0.75 0.27 Feeding groups: W= wood-feeders; S= soil-feeders; L= leaf-feeders; W/S= wood-soil interface feeders; W/L= wood-leaf interface feeder Table 1 - Abundances (individuals.m-²) and biomasses (g fresh weight.m-²) of termites in two forest types within the Brazilian semiarid domain: humid highland forest (“Brejo de Altitude” - Mata do Pau Ferro) and seasonally dry tropical forest (“Caatinga” - Fazenda Almas). Sociobiology 62(1): 70-75 (March, 2015) 73 weight) were encountered during the rainy season, while 2215 ± 1099 individuals.m-² and 5.45 ± 3.04 g.m-² (fresh weight) were found during the dry season. Termite abundance and biomass were both significantly greater during the rainy season (F1,84=7.21; P<0.01 and F1.84= 8.35; P<0.01 respectively), and vertical stratification was also identified (F2.84= 7.04; P<0.01 and F2.84= 3.13; P<0.05 respectively) (Fig 1). However, when considering the interactive effects of the season and the soil strata on the termite fauna, it was observed that their abundances and biomasses did not vary significantly (F2.84=0.73; P=0.41 and F2.84=0.07; P=0.92 respectively). Fig 1. Biomasses (biomass density, g.m-2) and abundances (numerical density, individuals.m-2) of termites in different soil strata (A: 0-10 cm; B: 10-20 cm; C: 20-30 cm), in two forest types in the semiarid domain of northeastern Brazil: seasonally dry tropical forest (“Caatinga” - Fazenda Almas) and thumid highland forest (“Brejo de Altitude” - Mata do Pau Ferro). Discussion Although they are both located within the Brazilian semiarid domain, the dry tropical forest and humid highland forest sites demonstrate distinct climatic characteristics and consequently very little similarity between their termite species (two species), abundances, and biomasses. Arthropods occupying the leaf litter/soil complex are generally influenced by the climatic characteristics of the host ecosystems (Adis et al., 1989; Harada & Bandeira, 1994; Dibog et al., 1998; Pinheiro et al., 2002; Doblas-Miranda et al., 2007). The assemblage structure and foraging of termites in the Neotropical region have been observed to be associated with climatic elements, especially temperature and rainfall regime, both in humid forests (Torres & Bandeira, 1985; Cancello et al., 2014) and in semiarid ecosystems (Moura et al., 2006; Araújo et al., 2010), yet the humid forest sites show higher rates of biomass and abundance of termites. Termite abundance in the soil at the dry tropical forests site was 862 individuals.m-2 during the rainy season and 250 individuals.m-2 during the dry season, while their biomass was 1.40 g.m-2 and 0.45 g.m-2 respectively. These values were larger than those reported by Melo and Bandeira (2004) for a number of microhabitats (soil, leaf litter, nests and wood) in another area of seasonally dry tropical forest experiencing anthropogenic pressure. In a study undertaken at the dry tropical forests site that considered only nest populations, Vasconcellos et al. (2007) recorded abundance and biomass values of 278.2 individuals.m-2 and 0.9 g.m-2, respectively, for C. cyphergaster. By combining these values, the abundances and biomasses of termites in this seasonally dry tropical forest area appeared to oscillate between 428 and 610 individuals.m-2 and between 1.17 and 1.65 g.m-2 respectively. It is important to note that these values do not include other termite species that construct conspicuous nests in the area, or those that forage within tree trunks. The abundance and biomass of termites found only in the soil in the humid highland forest site were larger than those previously recorded in other ecosystems in the Neotropical region (Martius, 1994; Bandeira & Vasconcellos, 2002; Vasconcellos, 2010). Other quantitative methods, including mounds and arboreal nests, would be likely to increase abundance and biomass data. Specifically in terms of the soil, the largest values recorded to the present time were 4331.54 individuals.m-2 and 7.65 g.m-2 fresh weights in an Atlantic Forest fragment in northeastern Brazil (Vasconcellos, 2010). Even including estimated values for other microhabitats (such as nests and decomposing wood), termite populations in humid highland forest were still the highest so far reported for the Neotropical region (Bandeira & Torres, 1985; Martius, 1994; Bandeira & Harada, 1998; Silva & Bandeira, 1999; Bignell & Eggleton, 2000). The abundances and biomasses of termites at the dry tropical forests site did not demonstrate significant variations between the dry and rainy seasons. Other studies undertaken in seasonally dry tropical forests in the region, however, demonstrate an effect on vertical distribution, foraging and abundance of insects in general (Moura et al., 2006; Araújo et al., 2010; Vasconcellos et al., 2010a), including termites (Melo & Bandeira, 2004; Vasconcellos et al., 2010b). Apparently, other characteristics, such as food availability, soil moisture, and interspecific interactions, could minimize climatic effects on soil fauna in seasonally dry tropical forest ecosystems. The humid highland forest site, on VFP ArAújo, MP SilVA, A VASconcelloS – Soil Termites in the Brazilian Semiarid Domain74 the other hand, demonstrated significant variations in its termite population parameters between the dry and rainy seasons. Although there have been very few publications focusing on seasonal variations of the termite fauna in humid tropical forests, Eggleton and Bignell (1995) suggested that abundance and biomass values in these ecosystems were greater in the rainy season. Bandeira and Harada (1998), however, reported higher values for these same parameters among soil termites during the dry season in the Amazon forest. In relation to the depth of the soil samples, it was observed that the greatest abundances and biomasses were encountered in the top 20 cm profiles, independent of the forest type (Walwork, 1970). According to Peterson and Luxton (1982), most of the edaphic biota is concentrated between 0 and 10 cm below the soil surface. This tendency does vary considerably, however, in different localities and when different taxa are considered (Walwork, 1970; Adis et al., 1989; Dindal, 1990; Harada & Bandeira, 1994). A number of studies in the Neotropical region have found that a 20 cm soil depth is the most adequate for collecting termites, especially in terms of species richness (Vasconcellos, 2010; Bandeira & Vasconcellos, 2002; Silva & Bandeira, 1999). The observed increase in termite numbers and biomass during the rainy season in the humid highland forest indicates that their decomposition activities are asymmetrical during the year. The quantitative similarity of termite numbers and biomass during the year in the seasonally dry tropical forest area suggests that other factors, beyond climatic differences, influence the abundance and biomass of these termites, at least at a local scale. The quantitative differences observed between the two forest types reflect the effects of different environmental variables on ecosystems, in spite of the fact that they are both embedded in the semiarid Brazilian domain. Acknowledgements The authors are grateful to the Braz family, owners of the RPPN-Fazenda Almas, for logistical support during the research, and to Nilton Leite de Sousa Júnior for their input. This research was supported by CNPq/PELD (520062/2006-0). References Adis, J., Ribeiro, E.F., Morais, J.W. & Cavalcante, E.T.S. (1989). Vertical distribuition and abundance of arthropods from white sand soil of a Neotropical Campinarana Forest during the dry season. Studies on Neotropical Fauna and Environment, 24: 201-211. Anderson, J.M. & Ingram, J.S.M. (1989). Topical Soil Biology and Fertility: a Handbook of Methods. UNESCO / IUBS, 77p. Andrade-Lima, D. (1966). Esboço fitoecológico de alguns “Brejos” de Pernambuco. Boletim Técnico. Instituto de Pesquisas Agronômicas de Pernambuco, 8: 3-9. Andrade-Lima, D. (1981). The caatingas dominium. Revista Brasileira de Biologia, 4: 149–153. Andrade-Lima, D. (1982). Present day forest refuges in Northeastern Brazil. In: G.T. Prance (Ed.), Biological Diversification in the Tropics (pp. 245-254). New York: Columbia University Press. Araújo, V.F.P., Bandeira, A.G. & Vasconcellos, A. (2010). Abundance and stratification of soil macroarthropods in a Caatinga Forest in Northeast Brazil. Brazilian Journal of Biology, 70: 337-346. doi:10.1590/S1519- 69842010000400006. Bandeira, A. G. & Harada, A. Y. (1998). Densidade e distribuição vertical de macroinvertebrados em solos argilosos e arenosos na Amazônia Central. Acta Amazonica, 28: 191-204. Bandeira, A.G. & Torres, M.F.P. (1985). Abundancia e distribuição de invertebrados do solo em ecossistemas Amazônicos. O papel ecológico dos cupins. Boletim do Museu Paraense Emílio Goeldi. Ser. Zoologia, 2: 13-38. Bandeira, A. G. & Vasconcellos, A. (2002). A quantitative survey of termites in a gradient of disturbed highland forest in Northeastern Brazil. Sociobiology, 39: 429-439. Bignell, D.E. & Eggleton, P. (2000). Termites in ecosystems. In: T. Abe, D.E. Bignell & M. Higashi (Eds.), Termites: evolution, society, symbioses, ecology (pp. 363–387). Dordrecht: Kluwer. Cancello, E.M.; Silva, R.R; Vasconcellos, A.; Reis, Y.T. & Oliveira, L.M. (2014). Latitudinal Variation in Termite Species Richness and Abundance along the Brazilian Atlantic Forest Hotspot. Biotropica, 46: 441-450. doi: 10.1111/btp.12120 Coimbra-Filho, A.F. & Câmara, I.G. (1996). Os limites originais do bioma Mata Atlântica na região Nordeste do Brasil. Rio de Janeiro: Fundação Brasileira para Conservação da Natureza, 86 p. Collins, N.M. (1989). Termites. In: H. Lieth & M.J.A. Werger (Eds.), Tropical rain forest ecosystems (pp. 455–471). Amsterdam: Elsevier. Dibog, L., Eggleton, P. & Forzi, F. (1998). Seasonality of soil termites in a humid tropical forest, Mbalmayo, southern Cameroon. Journal of Tropical Ecology, 14: 841-850. Dindal, D. L. (1990). Soil biology guide, 1348p. Dirzo, R., Young, H.S., Mooney, H.A. & Ceballos, G. (2011). Seasonally Dry Tropical Forests: Ecology and Conservation. Washington: Island Press, 1st Edition. 394 p. Doblas-Miranda, E., Sánchez-Piñero, F. & González-Megías, A. (2007). Soil macroinvertebrate fauna of a Mediterranean arid system:Composition and temporal changes in the assemblage. Soil Biology and Biochemistry, 39: 1916 -1925. Eggleton, P. & Bignell, D.E. (1995). Monitoring the response of tropical insects to change in the environment: troubles with termites. In: R. Harrington & N.E. Stork (Eds.), Insects in a Changing Environment (pp. 474-97). London: Academic Press. EMBRAPA-SNLCS. (1972). Levantamento exploratório - reconhecimento de solos do Estado da Paraíba. Rio de Sociobiology 62(1): 70-75 (March, 2015) 75 Janeiro, Boletim Técnico, 15. Brasil. SUDENE-DRN. (Série Pedologia, 8). Governo do Estado da Paraíba. (1985). Secretaria da Educação. Universidade Federal da Paraíba. Atlas geográfico do Estado da Paraíba. Grafset. João Pessoa, 100 p. Harada, A.Y. & Bandeira, A.G. (1994). Estratificação e densidade de invertebrados em solo arenoso sob floresta primária e plantios arbóreos na Amazônia Central durante a estação seca. Acta Amazônica, 24: 103-118. Holt, J.A. & Coventry, R.J. (1990). Nutrient cycling in Australian savannas. Journal of Biogeography, 17: 427-432. Jones, C.G., Lawton, J.H. & Shachak, M. (1994). Organisms as ecosystem engineers. Oikos, 69: 373-386. Lee, K.E. & Wood, T.G. (1971). Termites and soils. London and New York: Academic Press, 251 p. Martius, C. (1994). Diversity and ecology of termites in Amazonian forests. Pedobiologia, 38: 407–428. Martius, C. (1998). Occurrence, body mass and biomass of Syntermes spp. (Isoptera, Termitidae) in Reserva Ducke, Central Amazonia. Acta Amazonica, 28: 319-324. Matsumoto, T. (1976). The role of termites in an equatorial rain forest ecosystem of West Malaysia. I. Population density, biomass, carbon, nitrogen and calorific content and respiration rate. Oecologia, 22: 153–178. Matsumoto, T. & Abe, T. (1979). The role of termites in an equatorial rain forest ecosystem of West Malaysia. II. Leaf litter consumption of the forest floor. Oecologia, 38: 261–274. Mélo, A. C. S. & Bandeira, A. G. (2004). A qualitative and quantitative survey of termites (Isoptera) in an open shrubby caatinga in Northeast Brazil. Sociobiology, 43: 707-716. Moura, F.M.S.; Vasconcellos, A.; Araujo, V. F. P. & Bandeira, A. G. (2006). Seasonality in foraging behavior of Constrictotermes cyphergaster (Termitidae, Nasutitermitinae) in the Caatinga of Northeastern Brazil. Insectes Sociaux, 53: 472-479. Oliveira, F.X., Andrade, L.A. & Felix, L. P. (2006). Comparações florísticas e estruturais entre comunidades de Floresta Ombrófila Aberta com diferentes idades, no Município de Areia, PB, Brasil. Acta Botanica Brasilica, 20: 861-873. Peterson, H. & M. Luxton. (1982). A comparative analysis of soil fauna populations and their role in decomposition processes. Oikos, 39: 287-388. Pinheiro, F.; Diniz, I. R.; Coelho, D. & Bandeira, M. P. S. (2002). Seasonal pattern of insect abundance in the Brazilian Cerrado. Austral Ecology, 27: 132-136. Prado, D. (2003). As caatingas da América do Sul. In: I. R. Leal, M. Tabarelli. & J.M.C. Silva (Eds), Ecologia e conservação da Caatinga (pp. 3–73). Recife: Editora Universitária, Universidade Federal de Pernambuco. Rodal, M.J.N. & Sales, M.F. (2008). Panorama of the Montane Forests of Pernambuco, Brazil. In: W.W. Thomas (Ed.), The Atlantic Coastal Forest of Northeastern Brazil (pp. 541-559). New York: The New York Botanical Garden, Bronx. Silva, E.G. & Bandeira, A.G. (1999). Abundância e distribuição de cupins (Insecta: Isoptera) em solo de Mata Atlântica, João Pessoa, Paraíba, Brasil. Revista Nordestina de Biologia, 13: 13- 36. STATSOFT. STATISTIC 5.0. (1995). Inc. Tulsa. EUA. Tabarelli, M. & Santos, A.M.M. (2004). Uma Breve Descrição Sobre a História Natural dos Brejos Nordestinos. In: K.C. Porto, J.J.P. Cabral & M. Tabarelli (Orgs.), Brejos de altitude em Pernambuco e Paraíba: história natural, ecologia e conservação (pp. 17-24). Ministério do Meio Ambiente, Brasília. Vasconcellos, A. (2010). Biomass and abundance of termites in three remnant areas of Atlantic Forest in northeastern Brazil. Revista Brasileira de Entomologia, 54: 455-461. Vasconcellos, A. P. S. & Bandeira, A. G. (2007). Biomass and populational structute of Constrictotermes cyphergaster (Silvestri) (Isoptera, Termitidae) in the dry forest of Caatinga, Northeastern Brazil. Neotropical Entomology, 36: 693-698. Vasconcellos, A.; Andreazze, R.; M.P.; Oliveira, E. S. & Oliveira, U. (2010a). Seasonality of insects in the semi-arid Caatinga of northeastern Brazil. Revista Brasileira de Entomologia, 54: 471- 476. Vasconcellos, A.; Bandeira, A.G.; Moura, F.M.S.; Araújo, V.F.P.; Gusmão, M.A.B. & Constantino, R. (2010b). Termite assemblages in three habitats under different disturbance regimes in the semi- arid Caatinga of NE Brazil. Journal of Arid Environments, 74: 298-302. Wallwork, J.A. (1970). Ecology of soil animals. England. McGraw - Hill Publishing Company Ltd, 283p. Whitford, W.G. (1991). Subterranean termites and long-term productivity of desert rangelands. Sociobiology, 19: 235-243. Wood, T.G. & Sands, W.A. (1978). The role of termites in ecosystems. In: M.V. Brian (Ed.), Production ecology of ants and termites (pp. 245–292). Cambridge: Cambridge University Press.