On the female of Metagonia taruma ( Araneae : Pholcidae ) , ecology of the pholcid spiders in the Urucu River Basin , Amazonas , Brazil and new records from Brazilian Amazonia

In this study we describe the unknown female of Metagonia taruma Huber, 2000, which was discovered after sampling in two forest gap types at Porto Urucu (Urucu River Basin, Coari, Amazonas, Brazil), and also provide information on the community ecology and natural history of the sampled species. The female of M. taruma is similar to that of M. samiria (Huber, 2000) by having an epigynum with a slightly projecting broad scape with a distal pocket; it differs by the larger pore plates. We collected twelve Pholcidae species at Porto Urucu and M. taruma was the most frequent and abundant. The populations of Carapoia ocaina Huber, 2000 and Mesabolivar aurantiacus (Mello-Leitão, 1930) present homogeneous sex ratios, while M. taruma and Mesabolivar sp. were female biased. Only two species (M. taruma and Mesabolivar sp. ) exhibited differences in abundance in each forest gap type, being higher at the poorly regenerated gaps. Thus, the use of Pholcidae species as ecological indicators is promising. We also present new records, throughout the Amazon Basin, for the Pholcidae species collected at Porto Urucu.

Pholcidae comprises over 1000 species of spiders in about 85 genera (PLATNICK 2009) and its species occur in a variety of habitats worldwide, from rain forests to deserts, and from sea level to over 3500 m (HUBER 2000).The taxonomic and phylogenetic knowledge of these animals has increased considerably over the past few years (e.g.HUBER 2000, 2001, 2005a, b, BRUVO-MADARIC et al. 2005, ASTRIN et al. 2006, 2007).The New World genera were subject to a recent comprehensive revision (HUBER 2000) and further taxonomical additions (HUBER & BRESCOVIT 2003, HUBER 2005a, c, HUBER & WUNDERLICH 2006, HUBER et al. 2005a, b, c, HUBER et al. 2010, HUBER & ASTRIN 2009, ASTRIN et al. 2006, MACHADO et al. 2007a, b, c).Nevertheless, a large number of taxa are yet undescribed, many of them available in South American collections (HUBER 2000).This is the case for Metagonia Simon, 1893, which occurs from Mexico to Central Argentina and is one of the most species-rich pholcid genus in the neotropics, with 81 described species (PLATNICK 2009).The taxonomic knowledge of Pholcidae gathered to date contrasts with the sparse ecological and biological data available for most of the New World species, although advancements have already been made (e.g.BRICEÑO 1985, EBERHARD 1992a, b, EBERHARD & BRICEÑO 1983, 1985, HUBER 1994, 1997a, c, d, 1998a, b, HUBER & EBERHARD 1997, JAPYASSÚ & MACAGNAN 2004, MACHADO et al. 2007a, c, PERETTI et al. 2006, SEWLAL & STARR 2008).
In the present paper, the unknown female of Metagonia taruma Huber, 2000 is described and new records of four pholcid species in the Brazilian Amazon are reported.Data on the community ecology, the abundance of pholcid spiders in two different forest gap types, sex ratios, and natural history of pholcid species from Porto Urucu (Urucu River Basin, Coari, Amazonas, Brazil) are presented.

MATERIAL AND METHODS
Porto Urucu is a 54000 ha petroleum and natural gas production facility on the right margin of the Urucu River, Solimões River basin, Coari municipality, state of Amazonas, Brazil (04°53'S, 65°20'W, Fig. 1), where several structured inventory initiatives have been carried out to evaluate the environmental impact of such production activities on the forest matrix fauna.The regional climate is classified as "Afi" in the Köppen system, with the rainy season between September and April, and the dry season between May and August (RADAM-BRASIL 1978).The main forest formation in Porto Urucu is "terra firme" (upland forest) associated with lowlands and plateaus irrigated by small creeks ("igarapés"), tributaries of the Urucu River (for a description of an area with similar topography, see RIBEIRO et al. 1999) ZOOLOGIA 27 (3): 431-439, June, 2010 "I-good", age > 5 years) and seventeen poorly regenerated forest gaps (Class "II-poor", age < 5 years) were sampled during a three month expedition (between June and November, 2006), using Winkler extractors, beating trays, and nocturnal hand collecting.Beating trays were used to access shrub-dwelling spiders during the day.Each tray was composed of a 0.8 m 2 white cotton sheet.The nocturnal hand collecting method was modified by combining the "looking up -looking down" methods proposed by CODDINGTON et al. (1991).The samples, combined for each methodology, were the results of one square meter of litter (sifted and placed in Winkler extractors for two days), one hour using beating trays per collector and one hour of nocturnal hand collecting per collector, within an area of 300 m 2 .To complete the species list, a few individuals gathered on occasional collections were also computed.Kruskal-Wallis tests were used to test if the differences between the abundances by samples for each of the four most abundant species were statistically significant, and a Dunn test was performed to search for the source of statistical differences.Chi-squared tests were used to verify the sex ratio (always expressed as male/female) of each species and its frequency of occurrence (number of samples with a given species divided by the total number of samples with pholcid species).To test statistical differences between the species abundance by the forest gap classes, a Mann-Whitney test was performed, using all males and females sampled.All analyses were performed using BioEstat 5.
Diagnosis.Females of M. taruma differ from the remaining congeneric species, except from Metagonia samiria (Huber, 2000: figs 238-243), by the epigynum with a slightly projecting broad scape with a distal pocket; they differ from M. samiria by the larger pore plates (Figs 4-5).
Male.Described by HUBER (2000: 61).Males collected at Porto Urucu differ slightly with respect to the relative sizes of the distal elements of the procursus, and the clypeus apophysis (the lateral "wings" are more prominent), as described by HUBER (2000: 63-64).

New records
We also collected individuals of four described species, at Porto Urucu River Basin.Below, the records of these species are listed, including our findings and the remaining individuals present in the spider collection of MPEG.

Ecology and natural history
Combining the material from the two forest gap classes with additional occasional collections, 241 adult pholcid spiders representing eleven species were collected.This represents the highest known pholcid diversity on a single locality in the Brazilian Amazon, surpassing the nine and eight species recorded from Floresta Nacional de Caxiuanã (BONALDO et al. 2009) and Reserva Florestal Adolpho Ducke (HÖFER & BRESCOVIT 2001), respectively.On the other hand, for the entire Amazon Basin, the highest Pholcidae species richness is recorded for Pakitza (Madre de Dios, Peru), with twelve species (SILVA & CODDINGTON 1996).Four pholcid species collected at Porto Urucu were singletons belonging to Ibotyporanga Mello-Leitão, 1944 (collected with Winkler extractor), Otavaloa Huber, 2000 species (collected during nocturnal hand searches), and two species of undetermined genera (Pholcidae sp. 1 and sp.2; both collected with beating trays).Other infrequent species were L. dimona (n = 9), M. beni (n = 9) and an undetermined species of Litoporus Simon, 1893 (n = 2).The remaining four species, all collected with beating trays and nocturnal hand searches, were abundant at Porto Urucu: M. aurantiacus (n = 29), C. ocaina (n = 34); Metagonia taruma (n = 98) and an undescribed species of Mesabolivar González-Sponga, 1998 (n = 56, B.A. Huber & E.O. Machado, pers. comm.).Most M. aurantiacus and Mesabolivar sp.specimens were collected in nocturnal hand searches, and for C. ocaina the abundances using both collecting methods were similar.
Species of Metagonia are known to inhabit leaf litter, foliage and caves (HUBER 1997a, b, 1998a, PÉREZ-GONZÁLEZ & HUBER 1999, HUBER et al. 2005c, HUBER & SCHÜTTE 2009), and most of the forest species are adapted to a cryptic life on the underside of living leaves, with pale greenish bodies, very thin and long legs, elongated abdomens, and sometimes even elongated eggsacs (HUBER et al. 2005c).Metagonia taruma is a foliage-dwelling species with all of these cryptic morphological features, although its egg-sac shape has never been described and neither was it observed in the present study.This species was most efficiently collected with beating trays, a method that easily accesses the herb and shrub layers (CODDINGTON et al. 1991), and can also be found in the forest canopy or sub-canopy, as ZOOLOGIA 27 (3): 431-439, June, 2010 attested by fogging samples at Urucu River Basin, when the shrub layer was removed.
Metagonia taruma was significantly the most abundant (Kruskal-Wallis: H = 73.4694;d.f.= 3; p < 0.0001) and frequent ( 2 = 27.346,d.f.= 3, p < 0.0001) Pholcidae species from Porto Urucu.The abundance pattern of Pholcidae species on different collecting localities throughout the Amazon Basin is variable; at Floresta Nacional de Caxiuanã, the most abundant pholcid is Carapoia fowleri Huber, 2000(BONALDO et al. 2009).This variation could be the result of differences in the sampling effort and collecting methods and/or real differences between the local biota.A summary of the data (number of individuals, collecting method used and frequency of occurrence) for all pholcid species from Porto Urucu is shown in table I.The remaining Pholcidae species had similar abundances by samples, according to the Dunn test results.
The sex ratio in C. ocaina was 1.45 ( 2 = 0.926, d.f.= 1, p = 0.4414; with Yates correction), in M. aurantiacus 1.08 ( 2 = 0.037, d.f.= 1, p = 1, with Yates correction) showing that the ratios were fairly 1:1 and the differences found are likely due to stochastic variation.In M. taruma the sex ratio was 0.63 ( 2 = 4.742, d.f.= 1, p = 0.0381, with Yates correction) and for Mesabolivar sp. was 0.35 ( 2 = 9.524, d.f.= 1, p = 0.0034, with Yates correction).Table II shows a summary of the data used in the chi-square tests.For C. ocaina and M. aurantiacus the statistical tests did not reveal differences in the sex ratio, but the populations of M. taruma and Mesabolivar sp. at Urucu River Basin appear to be female biased.MACHADO et al. (2007a) measured the sex ratio of two Mesabolivar species, and concluded that Mesabolivar forceps Machado et al., 2007 was male biased, which was explained by the higher activity of males than females, making them easier to capture in pit-fall traps (ÁLVARES et al. 2004), while Mesabolivar mairyara Machado et al., 2007 had a homogeneous sex ratio, which was explained only by the low number of adult specimens collected.Our findings should not be explained by the low sampling intensity, neither by the use of pit-fall traps, as all of the four most abundant Pholcidae species were collected with active methods (beating tray and nocturnal hand searches).Thus, this sex ratio pattern can represent the real sex ratio of these species, because all of the visualized individuals were sampled independently of their sex class, although further studies are also required to understand the whole process of sexual ratio in Pholcidae.At least for species of Metagonia, female biased populations have already been reported by HUBER & SCHÜTTE (2009) who captured more than twice as many females than males -Metagonia osa Gertsch, 1986: 8 males, 18 females; Metagonia uvita Huber, 1997: 6 males, 20 females -, with active diurnal hand collecting.
The abundances of C. ocaina and M. aurantiacus were similar in both forest gap types of different regeneration levels, while M. taruma, Mesabolivar sp. and all the pholcid species sampled at the Urucu River Basin together were statistically higher in the poorly regenerated gap type forests (Tab.III).This data shows that pholcid abundance at Porto Urucu can be higher in poorly regenerated areas, where a myriad of effects could be underway: the availability of microhabitats could increase, certain predators might be absent, individuals might find better conditions for establishing their webs, or food availability could be higher.Thus, Pholcidae species can be considered potential ecological indicators of habitat quality, as the species respond differently to environmental conditions, such as the regeneration stage.This response has already been reported: MACHADO et al. (2007a) argued that M. forceps seems to present some degree of tolerance to habitats influenced by 0 software (AYRES et al. 2007).The statistical analyses were performed at 0.05 levels and only included data for species with abundances higher than ten individuals.The material examined is deposited in the collection of the Museu Paraense Emílio Goeldi, Pará, Brazil (MPEG, curator A.B. Bonaldo).The description style follows HUBER (2000).All measurements are in millimeters.The epigynum was dissected and examined in clove oil (LEVI 1965), after digestion of soft tissues with pancreatin/borax (ÁLVAREZ-PADILLA & HORMIGA 2008).The length/width (L/d) ratio of tibia 1 is a measure of the leg robustness and has a precision of about ± 2. The abbreviations used in the text are PME (posterior median eyes) and ALE (anterior median eyes).

On the female of Metagonia taruma (Araneae: Pholcidae), ecology of the pholcid spiders in the Urucu River Basin, Amazonas, Brazil and new records from Brazilian Amazonia Leonardo S. Carvalho 1, 3 ; Sidclay C. Dias 2 ; David F. Candiani 2 & Alexandre B. Bonaldo 2
Huber, 2000 andMesabolivar aurantiacus (Mello-Leitão, 1930)ogy and natural history of the sampled species.The female of M. taruma is similar to that of M. samiria(Huber, 2000)by having an epigynum with a slightly projecting broad scape with a distal pocket; it differs by the larger pore plates.We collected twelve Pholcidae species at Porto Urucu and M. taruma was the most frequent and abundant.The populations of Carapoia ocainaHuber, 2000 andMesabolivar aurantiacus (Mello-Leitão, 1930)present homogeneous sex ratios, while M. taruma and Mesabolivar sp. were female biased.Only two species (M.taruma and Mesabolivar sp. ) exhibited differences in abundance in each forest gap type, being higher at the poorly regenerated gaps.Thus, the use of Pholcidae species as ecological indicators is promising.We also present new records, throughout the Amazon Basin, for the Pholcidae species collected at Porto Urucu.

Table I .
Summary of the abundance and capture method of the Pholcidae species found at Porto Urucu, Coari, Amazonas, Brazil.(BT)Beatingtray,(N) nocturnal hand collecting, (WIN) Winkler extractor, (CF) canopy fogging, (O) occasional sampling, (FO) frequency of occurrence.*"Forestgaptype"does not certainly represent the number of individuals collected in the whole basin.Mesabolivar cantharusMachado et al., 2007, M. camussi Machado  et al., 2007, M. cavicelatus Machado et al., 2007were collected in slightly higher numbers in Pinus than in native areas, although the opposite occurred for Tupigea cantareiraMachado  et al., 2007.

Table II .
Abundances of the four most common Pholcidae species collected at Porto Urucu, Coari, Amazonas, Brazil, by sex class, forest gap types and the entire Amazon Basin.

Table III .
Mann-Whitney test results to verify abundance by forest gap regeneration type for C. ocaina, M. aurantiacus, M. taruma, Mesabolivar sp. and the combined data for all Pholcidae species collected at Porto Urucu, Coari, Amazonas.