The orchid-bee fauna ( Hymenoptera : Apidae ) of ‘ Reserva Biológica de Una ’ , a hotspot in the Atlantic Forest of southern Bahia , eastern Brazil

The orchid-bee fauna of ‘Reserva Biológica de Una’ (REBIO Una), one of the largest Atlantic Forest remnants in southern Bahia, eastern Brazil, was surveyed for the first time. Baits with sixteen different scents were used to attract males of orchid bees. Eight hundred and fifty-nine males belonging to 26 species were actively collected with insect nets during 60 hours in January and February, 2009, and January, 2010. Euglossa avicula Dressler, 1982 and Euglossa milenae Bembé, 2007 have been recorded for the first time in the state of Bahia. It was found that REBIO Una has one of the most diverse and rich orchid-bee faunas of the entire Atlantic Forest domain and holds some rare species, such as Euglossa cyanochlora Moure, 1996.


Introduction
The Atlantic Forest has been considered one of the earth's biological "hotspots" (Myers, 1988;Mittermeier et al., 1999;Galindo-Leal and Câmara, 2003) and the region of the southern state of Bahia is probably the hottest section of this hotspot (see Laurance, 2009), with the highest levels of endemism for many taxonomic groups (e.g., Dean, 1995;Pacheco et al., 1996;Thomas et al., 1997Thomas et al., , 1998;;Sambuichi et al., 2008).Nevertheless, most of the original forest cover in southern Bahia was wiped out and only scattered fragments remain (Fundação SOS Mata Atlântica, 1993;Ribeiro et al., 2009), the largest of them not exceeding 25,000 ha (Ribeiro et al., 2009).Species endemic in such fragmented biomes are more prone to population declines and even to extinction.Concerning the Neotropical forest-dependent orchid bees (Hymenoptera: Apidae: Euglossina), Nemésio (2009Nemésio ( , 2010aNemésio ( , 2011b) ) and Nemésio et al. (2012) have warned that some orchid-bee species with restricted geographical distributions within the Atlantic Forest domain may be severely affected by their habitat loss.It is impossible, however, to make any conclusions about the conservation status of any animal group if even the faunistic composition and actual geographic distribution of its species are little or not at all known.
The 'Reserva Biológica de Una' is one of the largest preserves in the Atlantic Forest of southern Bahia and is reputed to holding astonishingly high diversities and levels of endemism for many taxonomic groups (reviewed by Martini, 2002).There have been only two other surveys for orchid bees in southern Bahia (Melo, 2005;Nemésio, 2011a), but these few collections have revealed a high diversity of new records for the region (Nemésio, 2011a, e;Nemésio et al., 2012) and even new species (e.g., Moure, BIOLOGY 1996;Nemésio 2011bNemésio , c, d, 2012)).The main goal of this study is to expand distributional and taxonomic knowledge on the orchid bees of the Brazilian Atlantic Forest, with special emphasis to the region of southern Bahia, by providing relevant data on the orchid-bee species found in a previously unsampled area of one of the most threatened biomes in the world.

Study sites
This study was conducted at 'Reserva Biológica de Una' (REBIO Una), a 18,000-ha preserve situated in the municipality of Una, in the state of Bahia, eastern Brazil in January and February 2009, and the summer of January 2010, when orchid bees are most actively foraging (e.g., Martins and Souza, 2005).The vegetation of REBIO Una is essentially dense Atlantic Rain Forest [Central Lowland Forest according to Thomas and Carvalho (2003)].Most of the preserve consists of secondary forests in various successional stages, since pastures, other crops and mainly cocoa plantations happened to take place in the area before the establishment of the preserve in 1979, but pristine forested areas can also be found at REBIO Una (see Martini, 2002).Precipitation in the area is ca.1,300 mm/year (Thomas and Carvalho, 2003).

Sampling
Twenty hours of active sampling with insect nets were performed in each of the three selected sites in the preserve, totaling 60 hours, following the methodology proposed by Nemésio (2010aNemésio ( , b, 2011a, b), b): site-1 (15° 10' 07" S and 39° 03' 54" W, ca.60 m above sea level) was sampled on the 29 th and 30 th of January, 2009, from 08:00h to 15:00h and on the 20 th of January, 2010, from 08:00h to 14:00h; site-2 (15° 10' 26" S and 39° 03' 42" W, ca.60 m above sea level) was sampled on the 28 th (from 10:00h to 15:00h) and the 31 st (from 07:00 to 16:00h) of January, 2009, and on the 19 th of January, 2010 from 08:00h to 14:00h); site-3 (15°11'00" S and 39° 04' 01" W, ca.50 m above sea level) was sampled on the 1 st and 2 nd of February, 2009, from 07:00h to 14:00h and on the 21 st of January, 2010, from 07:00h to 13:00h.Sites 1 and 3 were located in areas with pristine, well developed forests.Site 2, for comparison, was located in an area in early to intermediate successional stage.At each site, 17 scent baits were placed ca.2.0 metres apart from each other at about 1.5 m above the ground.These baits were made of cotton wadding soaked with one of the following substances, known or believed to be attractive to orchid bees: benzyl acetate, benzyl alcohol, r-carvone, 1,8-cineole, p-cresol acetate, dimethoxybenzene, eugenol, β-ionone, methyl benzoate, methyl trans-cinnamate, heneicosane, methyl salicylate, skatole, tricosane, p-tolyl acetate, vanillin, and a mixture (1:1) of methyl trans-cinnamate and p-tolyl acetate.Baits with cineole, the most volatile compound, were recharged every hour.Bees arriving on the baits during the sampling period were collected with insect nets and killed with ethyl acetate.They were labelled as belonging to the project "Euglossina da Hileia Baiana" and were deposited at the Entomological Collection of 'Universidade Federal de Minas Gerais' (UFMG), where they were numbered from 17627-50352 to 17718-50681 (specimens collected in 2010) and from 18964-54409 to 19156-54953 (specimens collected in 2009).

Data analysis
Diversity was estimated through Shannon-Wiener diversity index (H'), as H' = -Σ p i ln (p i ), where p i is the proportion of the total number of species made up of the ith species (Pielou, 1975).Evenness (E) was estimated through the formula E = H'/ ln (S), where S is the species richness.The similarity in faunistic composition among all sites was estimated by the percent similarity index of Renkonen, recommended by Wolda (1981) for small samples.Based on those similarities, the areas were grouped using UPGMA (Sneath and Sokal, 1973).

Results
Eight hundred and fifty-nine orchid-bee males belonging to 26 species in four genera were collected at REBIO Una (Table 1).Euglossa carolina Nemésio, 2009 andEg. imperialis Cockerell, 1922 were the most abundant species and these two species together represented almost one third of all collected orchid bees (Table 1).Ten species were represented by three or less individuals; consequently, they were not recorded at all three sites (except Eg. liopoda Dressler, 1982 -Table 1).Eleven species were represented by 35 or more specimens, and only one of them, Eg. pleosticta Dressler, 1982, was not present in all three sites (Table 1).The same sampling effort was carried out in all three sites, allowing a direct comparison among their faunistic features.Site 2 presented the lowest abundance (ca.six specimens/hour) and number of species (17) (Table 1), as well as the lowest diversity (H' = 2.42).Twenty-one species were collected in both sites 1 and 3, but abundance was higher in site 3 (22 specimens/hour) than in site 1 (15 specimens/hour).Diversity was almost the same in these two sites (Table 1) as well as evenness, which showed a similar high value in all three sites (Table 1).
The ordination of the sites according to their faunas showed a great overall similarity among the sites.Sites 1 and 3, situated in the best preserved forest patches, grouped together with ca.70% similarity and then grouped to site 2, with ca.66% similarity.

Discussion
The strategy of intensive sampling using the same methods as employed here, over a few days during the season when orchid bees are most actively foraging, has been demonstrated to be adequate for surveys focusing on this group of bees (Nemésio 2010b, 2011a, b, Nemésio et al., 2012).As pointed out by Nemésio (2010b), twenty hours of sampling during the season when orchid bees are most abundant provides similar results to one year of monthly samplings of three to four hours per sampling (e.g., Rebêlo andGarófalo, 1991, 1997;Santos and Sofia, 2002;Nemésio andSilveira, 2007, 2010).This is because year-round sampling occurs predominantly during less favourable seasons for orchid bees.The high abundance recorded by Nemésio (2010bNemésio ( , 2011a, b, this study) , b, this study) is suggestive of the efficiency of this method, when areas to be sampled are distant and costs of regular visits to each area are very high.Although seasonal fluctuations in the populations of orchid-bee species cannot be measured with this sampling strategy, the faunistic composition and relative abundances of each species that visit the baits can be estimated with reasonable confidence.This may be particularly important in areas that are ill-sampled or have not been sampled at all, and may improve our knowledge on the occurrence and distribution of many species.Neves and Viana (2003) listed 33 species for the Atlantic Forest of Bahia, but Nemésio (2009Nemésio ( , 2011a) ) showed that some of them are junior synonyms of other species found in the area and other species were misidentifications.The only other studies carried out in southern Bahia are those by Melo (2005), at 'RPPN Estação Veracel', a private preserve belonging to Veracel Celulose in the municipalities of Porto Seguro and Santa Cruz Cabrália, some 100 km south of REBIO Una, and Nemésio (2011a), at 'Parque Estadual da Serra do Conduru' (PESC), a state preserve in the municipalities of Ilhéus and Uruçuca, some 60 km north of REBIO Una (see Nemésio 2011a).Nemésio (2011a) compared the list of orchid-bee species of both 'Estação Veracel' and PESC and recognised 26 species occurring  avicula Dressler, 1982;Eg. milenae Bembé, 2007;and Eg. monnei Nemésio, 2012), however, were not collected by Melo (2005) or Nemésio (2011a), raising the number of known species in the same area (since REBIO Una lies exactly between both preserves) to 29.The two former species are here reported for the first time in the state of Bahia.Euglossa monnei had already been listed to Bahia when originally described (see Nemésio 2012).The bee occurring in Bahia listed by Neves and Viana (2003) and Nemésio (2009Nemésio ( , 2011a) ) as Euglossa analis Westwood, 1840 was later recognised as a different species, Euglossa marianae Nemésio, 2011b, and it is hereby treated as such.
Recent surveys of orchid bees have emphasized the occurrence of new species or species not previously known to a given region or area (e.g.Morato, 2001;Nemésio 2010bNemésio , 2011a), but equally relevant can be the absence of some species expected to occur in a given area.In this respect, there is a noticeable absence of Euglossa cognata Moure, 1970, a bee known to occur in the Atlantic Forest of Espírito Santo and Minas Gerais (reviewed by Nemésio, 2009) and also listed for the Atlantic Forest of Bahia by Neves and Viana (2003) -although these latter authors did not indicate the exact locality where the examined specimens came from -and Nemésio (2009), for Una and Porto Seguro.Euglossa cognata seems to be a forest-dependent species and Nemésio (2009) suggests it should be considered as vulnerable according to IUCN criteria.The absence of this species in three relatively large forest remnants (PESC, REBIO Una, and RPPN Estação Veracel) in southern Bahia suggests not only that they may be geographically restricted, but also that its population may be declining in areas where it had been previously recorded, a fact already observed for its close ally Eg. marianae (see Nemésio, 2011b, Nemésio, 2013).
Diversity and richness were very similar among the three sampled sites at REBIO Una (Table 1), which is also reflected in the overall similarity among sites.It is noticeable, however, that the similarity among all sites varied from 66 to 69%, a relatively low value when compared to the observed overall similarity among sites situated in different and isolated forest remnants of urban areas (Nemésio andSilveira, 2007, 2010) or among sites situated in a large forest preserve in southeastern Brazil (Nemésio and Silveira, 2006), when similarities of over 80% were observed.This suggests that either the sampling effort, in spite of resulting in a high number of collected bees, is still not effective enough to thoroughly sample each site or that within-habitat heterogeneity (see Armbruster, 1993) should be considered when sampling areas with higher diversity.Thus, more sampling sites should be selected for these areas to better sample the entire area.It is remarkable and suggestive of the second hypothesis, for example, that Eg. pleosticta was not recorded at site 1, but was relatively common at both sites 2 and 3 (Table 1), in spite of the physiognomic differences between these latter sites.
In spite of the absence of Euglossa cognata, REBIO Una presented the highest richness and diversity not only when compared to PESC and 'Estação Veracel', but also to all other forest preserves in the Atlantic Forest to date, with the possible exception of 'Reserva Natural Vale', in Linhares, state of Espírito Santo (Bonilla-Gómez, 1999).It should be considered, however, that samplings at REBIO Una were carried out in late January and early February, when some species of Eufriesea were not active anymore.Maybe for this reason, only one species of Eufriesea was recorded at REBIO Una in the present study (see Table 1), when it is known that at least two other species occur in the region (see Nemésio 2011a).REBIO Una should also be highlighted for holding some very sensitive forest-dependent species, such as Euglossa marianae and Eg.cyanochlora.It should also be remarked that it is in REBIO Una region that the rare Eg. cyanochlora seems to present the highest genetic variability (see Nemésio et al., 2012).
Similar to many other "protected" natural areas in eastern Brazil, the situation faced by this preserve is problematic (few personnel to patrol the area; presence of hunters and wood dealers; legal problems with former owners of the land; invasion by "alleged" Indians who claim the area).The presence of apparently sensitive orchid-bee species, such as Euglossa marianae (see Tonhasca Junior et al., 2002;Nemésio and Silveira, 2006;Nemésio, 2010bNemésio, , 2011b)), Euglossa roubiki Nemésio, 2009and Euglossa amazonica Dressler, 1982(see Nemésio, 2010b, 2011b), besides Euglossa cyanochlora, suggests that this area still holds important habitat for most, if not all, local orchid bees.Nevertheless, if the conservation problems listed above are not solved in the near future, there will be almost nothing to be preserved, due to the accelerated damages caused to the area and the very slow governmental response to this situation.

Table 1 .
Diversity, evenness, species richness and number of specimens of each orchid-bee species collected at sites 1 to 3 at 'Reserva Biológica de Una', state of Bahia, eastern Brazil, after 20 hours of sampling in each site.