A contribution to the knowledge of the orchid bee fauna of the Los Amigos Biological Station , Madre de Dios , Peru ( Hymenoptera : Apidae : Euglossini )

sister organization, Asociación para la Conservación de le Cuenca Amazónica (ACCA). The station occupies 453 hectares on a terrace between the two rivers and ranges in elevation from 225–296 meters above sea level. Habitats include a mix of primary tropical humid forests, mature secondary forests that were disturbed >50 years ago, bamboo and palm forests, swampy forests around oxbow lakes, and some maintained cut areas. Two areas about 600 m apart were chosen for surveys of orchid bees. Site 1: “Forest” (12.55236° S; 70.10994° W). This site was located along a trail within a secondary, mature closed-canopy forest punctuated by occasional tree-falls, and containing a fairly sparse understory. Disturbance in this area was minimal though logging operations were occurring several km away. Site 2: “Airfield” (12.55631° S; 70.10751° W). This disturbed area consisted of the north end of an abandoned airfield (~1000 x 50 m). It was fully open, devoid of large trees, and primarily occupied by shrubs, grasses, bamboo, and vines. This site was surrounded on all sides by mature forest equivalent to that of Site 1.


Introduction
Orchid bees, tribe Euglossini, are a conspicuous component of the Neotropical bee fauna, known for their striking metallic body coloration and for the remarkable behavior of the males, which collect aromatic compounds principally by visiting (and in turn pollinating) flowers of Orchidaceae (Dressler 1982).The compounds are manipulated and stored via a series of morphological modifications of the legs, and exposed during mating (Eltz et al. 2007).Once the diverse chemicals involved in the attraction of the euglossine males to orchid flowers were discovered (Dodson et al. 1969) they were used as bait to sample and collect orchid bees, which comprise five genera and a little over 200 species (Nemésio and Ramussen 2011).Orchid bees have been inventoried at several sites throughout the Amazon Basin (e.g., Morato 1994;Rasmussen 2009).Still, it is thought that only about 1/3 of the species have been described (Roubik and Hanson 2004).New species continue to be discovered (e.g., Hinojosa-Díaz and Engel 2007;2011a, b;Nemésio 2007;2011a, b;Roubik 2004).Some species are also known pollinators of Brazil nuts (Myers et al. 2000;Mori and Prance 1987), and some authors suggest that they can be used as bio-indicators of habitat quality (Brown and Albrecht 2001;Hedström et al. 2006).
Herein, we present results of a survey of orchid bee diversity at the Los Amigos Biological Station, Madre de Dios, Peru.This is the first survey of the station and the second census after Dressler (1985) and Pearson and Dressler (1985) from this part of the Amazon basin.

Study Area
Los Amigos Biological Station (LABS) is located 100 km west of Puerto Maldonado in Madre de Dios, Peru near the confluence of the Madre de Dios and Los Amigos rivers.It is administered by the U.S.-based Amazon Conservation Association (ACA) and its South American A contribution to the knowledge of the orchid bee fauna of the Los Amigos Biological Station, Madre de Dios, Peru (Hymenoptera: Apidae: Euglossini) minutes of collecting time using chemical baits were spent in the airfield and forest sites, respectively.In a separate study by CSC and DJB, flight intercept traps (FIT) placed near the forest transect captured five bees during 7-13 of June, 2011; these specimen data are included because the FIT sampled the single specimen of Eulaema (Apeulaema) cingulata Fabricius.

Specimen Processing
Specimens were temporarily stored in glassine envelopes, and dried using silica gel.They were pinned, labeled, and databased at the Snow Entomological Museum Collection (SEMC) at the University of Kansas, Lawrence, Kansas, and were identified by IHD.Taxonomic names follow Nemésio and Rasmussen (2011).The study was conducted under Peru permit 056-2011-AG-DGFFS-DGEFFS to CSC and DJB and specimens are vouchered in the SEMC and in the collection of the Museum of Natural History, San Marcos University, Lima, Peru (MUSM).

Results and Discussion
Table 1 summarizes the orchid bees captured at the airfield and forest sites as well as their attraction to the methyl salicylate and eucalyptus oil baits.No bees were attracted to Listerine® or vanilla extract and will not be discussed further.In total, 93 individuals, three genera, and 15 species were captured.Over twice as many individuals and about 1.5 times more species were taken at the forest site than at the airfield site.A total of 88 individuals and 13 species were collected with the chemical baits; five additional individuals of four species (two additional) were collected with the flight intercept traps.About 1/3 of the total bees were caught between stations as a result of attraction to the insect net, which attained incidental residue of chemical lures from contact with the cotton cords and the bees.These bees are referred to in the table as hand collected, and should be regarded as being attracted to a mixture of methyl salicylate and eucalyptus oil.
A total of 67 individuals of 13 species in three genera were collected in the forest site with 62 being captured using chemical baits, and five with the use of a flight intercept trap (FIT).For the baits used, 16 individuals across seven species were collected with the methyl salicylate bait; 26 individuals across five species were collected with the eucalyptus oil bait; and 22 individuals across nine species were hand collected in Site 1.In the disturbed area (Site 2), a total of 26 individuals belonging to eight species and three genera were collected.Of the two baits which had any success, four individuals belonging to three species were collected using the eucalyptus oil bait; 14 individuals belonging to five species were collected using the methyl salicylate bait; and eight individuals belonging to five species were hand collected.Dressler (1985) and Dressler and Pearson (1985) reported four genera and 39 species of orchid bees resulting from an intensive two-year survey of the orchid bees of the Tambopata Reserved Zone.This area is 90 km southwest of the LABS and shares similar habitats and climate.Though there are no previously published studies of the orchid bees of LABS, the two areas undoubtedly share a highly similar orchid bee fauna.In fact, all 15 species captured in this study were previously known from Tambopata.

Taxonomic issues
Euglossa occidentalis and Eu.orellana were described in the last decade (Roubik 2004), consequently both names are absent from the two 1985 studies.However, both species can be added to the Tambopata list, since Roubik described both from specimens identified as Euglossa chalibeata Friese, some of them part of the material collected for the Tambopata studies (Dressler 1985;Pearson and Dressler 1985).
Pooling results from both sites, and ignoring the hand collections, the lures attracted similar numbers of species (seven for eucalyptus oil, eight for methyl salicylate) and individuals (28 for eucalyptus oil, 30 for methyl salicylate).However, the composition of species collected by each lure was distinctive, with no overlap in the airfield site and a 33% (three of nine species) overlap in the forest site.This demonstrates the utility of this combination of chemical baits; additional baits (e.g.eugenol, skatole, cineole, methyl cinnamate, benzyl acetate, etc.) would undoubtedly augment this list of species.
Nearly twice as many individuals and 1.5 times as many species were collected in the forest site than the airfield site, but this may reflect the greater collecting effort in the forest site.There may be differences in number of individuals and species composition between the two sites.Collecting with additional baits, for longer periods, and in more than one season might tease apart these variables, sample more species, discover new species, and further complement knowledge of the orchid bees of LABS.