Interactions between carpenter bees and orchid bees (Hymenoptera: Apidae) in flowers of Bertholletia excelsa Bonpl.

(Lecythidaceae)
Charles Fernando dos SANTOS1 & Maria Lcia ABSY2
ABSTRACT

Competition between two species of bees for the same type of floral resource may generate antagonistic behavior between them, especially in cultivated areas where food resources are limited, seasonally and locally. In this study, was tested the hypothesis of antagonism between two solitary bee species of the family Apidae, Eulaema mocsaryi (Euglossini) and Xylocopa frontalis (Xylocopini), visiting the Brazil nut flowers (Bertholletia excelsa: Lecythidaceae) in a central Amazonia agricultural area. The visitation time was analyzed to detect the possible temporal overlap in the foraging of these bees. Furthermore, was analyzed their interspecific interactions for manipulating flower species visited by an opponent species, as well as attempts to attack this opponent. The individuals of Xylocopa frontalis visited the Brazil nut flowers before Eulaema mocsaryi, although the peak visitation of both did not presented significant differences. Neither of the species manipulated flowers recently visited by opponent species, and there were practically no antagonistic interactions between them. Thus, X. frontalis and E. mocsaryi shared the same food source in the flowers of B. excelsa due to differences in their time of visits and non-aggressive way of interacting with the opponent. This result has important implications for pollinating the Brazil nut, and a possible management of X. frontalis and E. mocsaryi, since these two were the most abundant pollinators in the studied locality. KEYWORDS: antagonism, behaviors, facilitation, resource partitioning, pollinators.

Interaes entre abelhas carpinteiras e abelhas das orqudeas (Hymenoptera: Apidae) em flores de Bertholletia excelsa Bonpl. (Lecythidaceae)
RESUMO

A competio entre duas espcies de abelhas por um mesmo tipo de recurso floral pode gerar comportamentos antagnicos entre elas, principalmente, dentro de reas cultivadas, onde o recurso alimentar limitado sazonalmente e localmente. No presente trabalho, foi testada a hiptese de antagonismo entre duas espcies de abelhas solitrias da famlia Apidae, Eulaema mocsaryi (Euglossini) e Xylocopa frontalis (Xylocopini) em flores da castanheira do Brasil (Bertholletia excelsa: Lecythidaceae) em uma rea agrcola da Amaznia Central. O horrio de visitao foi analisado a fim de se constatar a possvel sobreposio temporal no forrageamento dessas abelhas. Alm disso, suas interaes interespecficas de manipular flores visitadas pela espcie oponente ou tentar agredir essa espcie foram analisadas. Os indivduos de Xylocopa frontalis visitaram as flores da castanheira antes que E. mocsaryi, embora o pico de visitao de ambas no tenha apresentado diferenas significativas. Nenhuma das duas espcies de abelhas manipulou flores recm-visitadas pela espcie oponente e praticamente no houve interaes antagnicas entre elas. Desse modo, X. frontalis e E. mocsaryi compartilham a mesma fonte alimentar nas flores de B. excelsa devido s diferenas em seus horrios de visitas e ao modo no agressivo de interagir com a espcie oponente. Esse resultado tem implicaes importantes para a polinizao da castanheira e um possvel manejo de X. frontalis e E. mocsaryi, uma vez que essas duas espcies de abelhas foram os polinizadores mais abundantes na localidade estudada. PALAVRAS-CHAVE: antagonismo, comportamentos, diviso de recursos, facilitao, polinizadores.

1 2

Curso de Ps-Graduao em Entomologia Lab de Palinologia - Coordenao de Pesquisas em Botnica, Instituto Nacional de Pesquisas da Amaznia, Caixa Postal 478, Av. Andr Arajo 2936, Bairro Petrpolis, 69011-970 Manaus, AM, Brasil; [email protected]

89

vol. 42(1) 2012: 89 - 94

Interactions between carpenter bees and orchid bees (Hymenoptera: apidae) in flowers of Bertholletia excelsa bonpl. (lecythidaceae)

INTRODUCTION
Bees have different foraging strategies that allow them to divide floral resources in the same species of plants with other flower visitors. This division of resources may occur spatially depending on the distribution of flowers on the top of a tree, or between trees located in different parts of a given region (Gumbert and Kunze 1999; Goulson et al. 2001; Reader et al. 2005; Goulson et al. 2008; Ishii et al. 2008). Furthermore, there may also be temporal differences between the various floral visitors during the exploration of a particular food resource (Morse 1981; Nagamitsu and Inoue 1997). Although the intensity of competition among visitors may be high, the division of resources (spatial and temporal) may have a facilitating effect, so that they need not use antagonistic or aggressive strategies on other individuals that are also using the same food source. However, some species of eusocial bees can be aggressive, thus permitting them to have a greater access to the resource exploited (Johnson and Hubbell 1975; Nagamitsu and Inoue 1997; Nieh et al. 2005). However, antagonism does not necessarily regulate all interactions between such species during their encounters in the flowers (Wilms et al. 1996). Although there are several studies involving the division of floral resources between the social bees, mentioned above, there is no work, at least for the Central Amazonia region that examines specifically the interactions between species of non eusocial bees during foraging activities. Thus it was that we analyzed two species of pollinating bees of Bertholletia excelsa Bonpl. (Lecythidaceae), namely, the orchid bee which occurs exclusively in the Amazon Basin (Moure 2000; Oliveira 2006), Eulaema mocsaryi (Friese, 1899) (Apidae: Euglossini), and a carpenter bee with wide occurrence in Brazil (Silveira et al. 2002), Xylocopa frontalis (Olivier, 1789) (Apidae: Xylocopini). These two species belong to two of five genera of the family Apidae that pollinate B. excelsa flowers, and the other three genera are Bombus (Bombini), Centris and Epicharis (Centridini) (Mori et al. 1978; Mller et al. 1980; Nelson et al. 1985; Mori and Prance 1990; Maus 2002). Eulaema mocsaryi and X. frontalis were chosen over other visitor bees of B. excelsa, because of their apparent predominance in relation to other species, and taking into account the works that have been carried out at the same locality as our study, and which point out the importance of these visitors as abundant in other crops (Renner 1986/ 1987; Santiago 1994). With the aim of testing the hypothesis that E. mocsaryi and X. frontalis would have an antagonistic behavior in relation to the other, which might reflect on their patterns of activity in flowers of B. excelsa, was analyzed the interactions between them, and the visiting time of each to the B. excelsa flowers.

MATERIAL AND METHODS

The study was conducted within a culture of B. excelsa on the Aruan farm, in the Municpality of Itacoatiara / Amazonas Brazil. This farm has a wide coverage of Amazon forest primary vegetation. Collections and observations were made from November 2006 to January 2007, with temperature and relative humidity fairly constant, averaging 27 C and 82% respectively. However, average rainfall differed greatly: 98 mm (November 2006), 161 mm (December 2006) and 494 mm (January 2007) (Inmet 2007). The Bertholletia excelsa Bonpl. is a native Amazon rainforest tree that occurs in terra firme forest, and has great social and economic importance in the Amazon region due the nutritional value of its nuts (Mori and Prance 1987; 1990). Bee activity in B. excelsa occurs during the flowering period of this species (Cymerys et al. 2005), namely, from October to February, in the morning, from 5:30 a.m. to 11:00 a.m., at which time it offers pollen and nectar to visitors who can manipulate its ligule (Moritz 1984). Due to the Brazil nuts physiological system, after 11.00 a.m. the androecium and flower petals fall, and only on the next day do new flowers bud in the open inflorescence (Mller et al. 1980; Nelson et al. 1985; Maus 2002). From each of the three collecting months, were chosed three individuals of B. excelsa by drawing lots, and had scaffolding installed beside their canopy. The observations and collections of bees in the flowers occurred during three consecutive days in each of the nine trees sampled. The visiting hours of X. frontalis and E. mocsaryi were recorded for 10 minutes every half hour from 5:00 a.m. to 11:00 a.m. In the remaining 20 minutes, the floral visitors were collected with an entomological net and stored in deadly vials for later identification. Those bee species, whose individuals permitted a reliable visual identification were only registered, and were no longer collected. The variation in the number of X. frontalis and E. mocsaryi individuals sampled during this period was analyzed by the Variance analysis test (ANOVA, double-factor, without repetition). The differences between the visitation times and visitation peaks (acrophase) of the two bee species were analyzed, comparatively, by the Rayleigh test. These test results were used to compare the month by month intraspecific and interspecific visitations, by the Watson-Williams test. We recorded two different behaviors of the two species 60 times for each bee species, namely, the floral manipulation of a flower recently visited by the opposite species (1. Yes, 0. No), and the antagonistic encounters in the B. excelsa flowers between the two species (1. Yes, 0. No). We used the chi-square (X2) to examine differences in the frequencies of both behaviors of the two species. The significance level used

90

vol. 42(1) 2012: 89 - 94

santos & absy

Interactions between carpenter bees and orchid bees (Hymenoptera: apidae) in flowers of Bertholletia excelsa bonpl. (lecythidaceae)

was 5%, and tests and graphs were generated by the statistical program BioStat 5.0. Additionally, other species of floral bee visitors were identified and quantified, and, although they were not included in the analysis of this study, they may eventually provide a basis for future works to complement the data investigated here.

RESULTS
There was no statistically significant variation in the quantity of individuals of X. frontalis observed during the study period (ANOVA, F = 1305, df = 12, p> 0.05); however, for E. mocsaryi variation in the quantity of visitor individuals (ANOVA, F = 7881, df = 12, p <0.05) was found. Eulaema mocsaryi, inclusive, was less abundant in all months in relation to X. frontalis, and in the last month the difference in the quantity between the two species was higher, practically double of the first two months. The other bee species recorded in the Brazil nut flowers are presented in Table 1. The average time of visitation for X. frontalis and E. mocsaryi throughout the entire sampling period is shown in Figure 1, where it observed the two bee species to have different peaks of activity; they were very close, however, with X. frontalis tending to visit flowers before E. mocsaryi. On analyzing the time variation of each species (Table 2) month by month, it appears that X. frontalis maintained the visiting pattern only between November 2006 and December 2006, and in the remaining period, this pattern had significantly changed. On the other hand, E. mocsaryi showed a more constant visiting pattern throughout the study period (Table 2). By the compared analysis of the visiting frequency of the two bee species (Table 3), was founded that in November 2006 and December 2006, there was a temporal overlap between the
Table 1 - Number of individuals and species of bees (Hymenoptera: Apoidea) potential pollinators of Bertholletia excelsa (Lecythidaceae). Aruan Farm, Itacoatiara, Amazonas, Brazil. November 2006 to January 2007 Family Tribes Bombini Species Bombus transversalis (Olivier, 1789) Centris dimidiata (Olivier, 1789) Centridini Centris flavifrons (Fabricius, 1775) Centris flavilabris Mocsry, 1899 Epicharis zonata Smith, 1854 Apidae Euglossini Euglossa intersecta Latreille, 1938 Eulaema bombiformis (Packard, 1869) Eulaema cingulata (Fabricius, 1804) Eulaema mocsaryi (Friese, 1899) Xylocopini Xylocopa frontalis (Olivier, 1789) Xylocopa muscaria (Fabricius, 1775) Total 77 16 45 20 27 49 576 29 737 1,037 14
a

Figure 1 - Average frequency of visitation of Xylocopa frontalis and Eulaema mocsaryi in flowers Bertholletia excelsa (Lecythidaceae). Aruan Farm, Itacoatiara, Amazonas, Brazil. Standard deviation: X. frontalis, november 2006 (5.408); december 2006 (3.425); january 2007 (8.549); E. mocsaryi, november 2006 (4.235); december 2006 (2.871); january 2007 (3.376).

two bee species, while in January 2007 only, there was indeed a distancing in the foraging peak between the two species in the B. excelsa flowers. Notably, it was the rainiest month of the three examined; however, at the moment we do not have the data to make a detailed analysis for exploring these results. Practically no individuals of E. mocsaryi and X. frontalis manipulate flowers after they had been recently visited by opponent species (X2 = 1.31, df = 1, p> 0.05). However, about half an hour after receiving such a visit, many flowers were manipulated by one of the two species (personal observation). Throughout the study, neither species changed their pattern of floral manipulation behavior (X. frontalis, X2 = 5.34, p> 0.05; E. mocsaryi, X2 = 04.04, p> 0.05, Figure 2). Eulaema mocsaryi showed more antagonistic behavior than X. frontalis (X2 = 4.60, p <0.05). However, although the test found statistical significance for this behavior, biologically (Figure 2), these antagonistic interactions between them
Table 2 - Comparison of the monthly frequency of visits Xylocopa frontalis and Eulaema mocsaryi the flowers Bertholletia excelsa (Lecythidaceae) during peak flowering in Aruan farm, Itacoatiara / AM. Xylocopa frontalis Eulaema mocsaryi Raylegh Watson-Williansb Raylegh Watson-Williansb Month to month Apa F r Ap F R November 2006: 07.6; 0.927; 07.9; 0.929; 4.04 0.06 December 2006 08.0 0.922 07.9 0.923 November 2006: 07.6; 0.927; 07.9; 0.929; 11.6* 0.01 January 2007 07.0 0.930 07.9 0.948 December 2006: 08.0; 0.922; 07.9; 0.923; 24.9* 0.02 January 2007 07.0 0.930 07.9 0.948
Ap: Acrophase. Peak hours of visitation in decimal b corrected comparisons a posteriori with Bonferroni * Significant

91

vol. 42(1) 2012: 89 - 94

santos & absy

Interactions between carpenter bees and orchid bees (Hymenoptera: apidae) in flowers of Bertholletia excelsa bonpl. (lecythidaceae)

Figure 2 - Frequency of behaviors performed* by Xylocopa frontalis and Eulaema mocsaryi in flowers Bertholletia excelsa (Lecythidaceae). Floral manipulation of a species in a new flower species visited by the opposite; Encounter antagonistic, attempted expulsion / aggression directed to the opposite species. Aruan Farm, Itacoatiara, Amazonas, Brazil. November 2006 to January 2007. * 60 records per species for each of the two behaviors.

Table 3 - Comparison of the frequency of visitation between Xylocopa frontalis and Eulaema mocsaryi in the flowers of Bertholletia excelsa (Lecythidaceae) during peak flowering in Aruan farm, Itacoatiara / AM. X. frontalis vs. E. mocsaryi Raylegh Apa November 2006 December 2006 January 2007
a

Watson-Williansb F r

07.6; 07.9 08.0; 07.9 07.0; 07.9

2.11 0.1 17.62*

0.927; 0.929 0.922; 0,923 0.930; 0.948

Ap: Acrophase. Peak hours of visitation in decimal b corrected comparisons a posteriori with Bonferroni * Significant

were rare and, therefore, it does not seem to modulate the interactions between them. During the study, there was no such change for either X. frontalis (X2 = 2.03, p> 0.05) or E. mocsaryi (X2 = 2.46, p> 0.05, Figure 2 - AE).

DISCUSSION
Our hypothesis that there would be a temporal differentiation in the peak activities of X. frontalis and E. mocsaryi was partially confirmed for two reasons. First, because this phenomenon occurred clearly only in one (January 2007) of the three months studied; second, because such temporal differentiation does not appear to have been very well regulated by our prediction tha the antagonistic encounters between the two bee species would reflect in different patterns of visitation. The non-evident overlap at the peaks of activity between the bee species, which was important only in January 2007,

occurred where the quantity of X. frontalis was practically double that of E. mocsaryi, and when there was a much higher rainfall than in previous months. Therefore, other factors may be contributing to the temporal division of resources between these bees, among them, environmental factors, and possibly the type of phenology of each of the bee. Other studies have shown, as recorded here, that Xylocopa species seems, preferentially, to visit flowers of Lecythidaceae early in the morning, unlike the species of Euglossini that visit them a little later in the same morning (Maus 2002; Aguiar and Gaglianone 2008). But regardless of which species visits the flowers most often before that of its competitor, neither manipulates flowers that have been manipulated by the counterpart, as examined here. To be able to recognize a flower already visited by other species may provide an advantage to the organism that is able to discriminate and make choices, because it maximizes foraging efficiency, as the individual would not lose time and energy trying to access a flower whose resources have been exhausted (Goulson et al. 1998; Goulson et al. 2001; Reader et al. 2005). The competition between floral visitors may be reduced by dividing resources between them, which implies a structuring of the community (Wilms et al. 1996). In this way, non-temporal difference in foraging of X. frontalis and E. mocsaryi, during November and December of 2006, shows that, at least during this period, the competition was not very strong, and what could be happening, therefore, is a facilitation between them. The records about the antagonistic encounters reinforce this line of thought, because of the several possible encounters between the two species in the B. excelsa flowers, the attempts of expulsion/aggression between them were rare. This finding is interesting because even bee species that are recorded visiting food resources aggressively, also may present some kind of division in the use of resources with other floral visitors (Nagamitsu and Inoue 1997; Kajobe and Echazarreta 2005; Kaminski and Absy 2006; Tan 2008). Therefore, the daily activities of X. frontalis and E. mocsaryi in B. excelsa flowers and how these bees interact with one another affect their foraging strategies. This demonstrates that they can share the same floral resources without showing prominent antagonism between them, which could have important implications for pollinating B. excelsa. Additional studies could analyze the other bee species, identified here, and the role that they represent in pollination of B. excelsa.

ACKNOWLEDGEMENTS
The authors are very grateful to Marcos Vinicius Bastos Garcia (Empresa Brasileira de Pesquisa Agropecuria,

92

vol. 42(1) 2012: 89 - 94

santos & absy

Interactions between carpenter bees and orchid bees (Hymenoptera: apidae) in flowers of Bertholletia excelsa bonpl. (lecythidaceae)

EMBRAPA Amaznia Ocidental) co-advisor of the first author in the Masters Degree program in Tropical Biology and Natural Resources / Instituto Nacional de Pesquisas da Amaznia (INPA), Mrcio Luis Oliveira (INPA) for the identification of bees Euglossini; Gabriel Melo (Universidade Federal do Paran) for identifying the bees Bombini, Centridini and Xylocopini, Henrique Manuel Salgado (Universidade do Estadual Paulista) and Srgio Dias Hilrio (Universidade de So Paulo) for their collaboration in statistical analysis; George Tokuwo Nakamura (INPA) by revising the text to english; Conselho Nacional de Desenvolvimento Cientfico e Tecnolgico (CNPq) for financial support (process n. 472563/2006-2008 and 480944/2004).

Maus, M.M. 2002. Reprodutive phenology and pollination of the Brazil nut tree (Bertholletia excelsa Humb. and Bonpl. Lecythidaceae) in eastern Amazonia, p. 245-254. In: Kevan, P.; Imperatriz-Fonseca, V.L. (Eds.). Pollination bees: The conservation link between agriculture and nature. Ministry of environment, Braslia, Distrito Federal. 313 pp. Morse, D.H. 1981. Interactions among syrphid flies and bumblebees on flowers. Ecology, 62: 81-88. Mori, S.A.; Prance, G.T. 1990. Taxonomy, ecology and economic botany of the Brazil nut (Bertholletia excelsa, Humb and Bonpl: Lecythidaceae). Advances in Economic Botany, 8: 130-150. Mori, S.A.; Prance, G.T. 1987. A guide to collecting Lecythidaceae. Annual Report Missouri Botanical Garden, 74: 321-330. Mori, S.A.; Prance, G.T.; Bolten, A.R. 1978. Additional notes on the floral biology of neotropical Lecythidaceae. Brittonia, 30: 113-130. Moritz, A. 1984. Biological studies of flowering and fruiting of the Brazil nut tree (Bertholletia excelsa H.B.K), Embrapa - Cpatu, (Documentos Embrapa, 29), Belm, Par, BR. 82 pp (in Portuguese). Moure, J.S. 2000. The species of the genus Eulaema Lepeletier, 1841 (Hymenoptera, Apidae, Euglossinae). Acta Biologica Paranaense, 29: 1-70 (in Portuguese). Mller, C.H.; Rodrigues, I.A.; Mller, A.A.; Mller, N.R.M. 1980. Brazil nut trees: results of research, Embrapa - Cpatu (Miscelnea, 2), Belm, Par, BR. 25 pp (in Portuguese). Nagamitsu, T.; Inoue, T. 1997. Aggressive foraging of social bees as a mechanism of floral resource partitioning in an Asian tropical rainforest. Oecologia, 110: 432-439. Nelson, B.W.; Absy. M.L.; Barbosa, E.M.; Prance, G.T. 1985. Observations on flower visitors to Bertholletia excelsa H. and B. K. and Couratari tenuicarpa A. S. SM (Lecythidaceae). Acta Amazonica, 15: 225-234. Nieh, J.C.; Kruizinga, K.; Barreto, L.S.; Contrera, F.A.L.; ImperatrizFonseca, V.L. 2005. Effect of group size on the aggression strategy of an extirpating stingless bee, Trigona spinipes. Insectes Sociaux, 52: 147-154. Oliveira, M.L. 2006. New hypothesis of phylogenetic relationships for the genera of Euglossini, and for the species of Eulaema Lepeletier, 1841 (Hymenoptera: Apidae: Euglossini) Acta Amazonica, 36: 373-286 (in Portuguese). Reader, T.; MacLeod, I.; Elliott, P.T.; Robinson, O.J.; Manica, A. 2005. Inter-order interactions between flower visiting insects: foraging bees avoid flowers previously visited by hoverflies. Journal of Insect Behavior, 18: 51-57. Renner, S.S. 1986/ 1987. Reproductive biology of Bellucia (Melastomataceae). Acta Amazonica, 16: 197-208. Santiago, F.F. 1994. Aspects of the reproductive biology of Bixa orellana L. (Urucum) in the Central Amazon. Dissertao de Mestrado, Instituto Nacional de Pesquisas da Amaznia/ Fundao Universidade do Amazonas, Manaus, Amazonas. 95 pp (in Portuguese).

Aguiar, W.M.; Gaglianone, M.C. 2008. Behavior of flower visiting bees of Lecythis lurida in the north of Rio de Janeiro state. Revista Brasileira de Entomologia, 52: 277-282 (in Portuguese). Cymerys, M.; Wadt, L.; Kainer, K.; Aryolo, V. 2005. Brazil nut trees, p. 61-73. In: Shanley, P.; Medina, G. (Eds.). Fruitful and useful plants for life Amazon. CIFOR and Imazon, Belm, Par (in Portuguese). Goulson, D.; Lye, G.C.; Darvill, B. 2008. Diet breadth, coexistence and rarity in bumblebees. Biodiversity and Conservation, 17: 3269-3288. Goulson, G.; Chapman, J.W.; Hughes, W.H.O. 2001. Discrimination of unrewarding flowers by bees; direct detection of rewards and use of repellent scent marks. Journal of Insect Behavior, 14: 669-678. Goulson, D.; Hawson, S.A.; Scout, J.C. 1998. Foraging bumblebees avoid flowers already visited by conspecifics or by other bumblebees species. Animal Behavior, 55: 199-206. Gumbert, A.; Kunze, J. 1999. Inflorescence height affects visitation behavior of bees a case study of an aquatic plant community in Bolivia. Biotropica, 31: 466-477. Ishii, H.S.; Kadoya, T.; Kikuchi, R.; Suda, S.I.; Washitani, I. 2008. Habitat and flower resource partitioning by an exotic and three native bumble bees in central Hokkaido, Japan. Biological Conservation, 141: 2597-2607. Johnson, L.K.; Hubbell, S.P. 1975. Contrasting foraging strategies and coexistence of two species on a single resource. Ecology, 56: 1398-1406. Kajobe, R.; Echazarreta, C.M. 2005. Temporal resource partitioning and climatological influences on colony flight and foraging of stingless bees (Apidae; Meliponini) in Ugandan tropical forests. African Journal of Ecology, 43: 267-275. Kaminski, A.C.; Absy, M.L. 2006. Bees visitors of three species of Clusia (Clusiaceae) flowers in Central Amazonia. Acta Amazonica, 36: 259-264. INMET, 2007. Instituto Nacional de Metereologia (www.inmet.gov. br/html/observacoes.php). Accesso em 13/03/2007.

LITERATURE CITED

93

vol. 42(1) 2012: 89 - 94

santos & absy

Interactions between carpenter bees and orchid bees (Hymenoptera: apidae) in flowers of Bertholletia excelsa bonpl. (lecythidaceae)

Silveira, F.A.; Melo, G.A.R.; Almeida, E.A.B. 2002. Brazilian bees: sistematic and identification. MMA e Fundao Araucria, Curitiba, PR, BR. 253 pp (in Portuguese). Tan, N.Q. 2008. Pollinatin ecology of Melaleuca cajuputi, Nypa fruticans and their flower visitors. Journal of Apicultural Research, 47: 10-16.

Wilms, W.; Imperatriz-Fonseca, V.L.; Engels, W. 1996. Resource partitioning between highly eusocial bees and possible impacto f the introduced africanizad Honey bee on native stingless bees in the Brazilian atlantic rainforest. Studies of Neotropical Fauna and Environment, 31: 137-151. Recebido em 11/10/2010 Aceito em 01/03/2011

94

vol. 42(1) 2012: 89 - 94

santos & absy