The first survey on harvestmen in Brazilian artificial cavities , with notes on distribution and natural history

Several species of harvestmen occur in natural subterranean cavities using these habitats as shelters and sites of oviposition. Many species have evolved in these environments, thus becoming cave-dwellers. In a few cases harvestmen have been reported in artificial cavities (mines), but without details about their distribution or natural history. Based on faunal inventories carried out in 111 artificial cavities in the state of Minas Gerais, Brazil, this work aimed to register species of harvestmen that are associated with these artificial cavities. Seventeen species were found in 12 municipalities, in addition to new occurrences for 14 of these species and some new behavioral notes. Two undescribed species were also recorded. This study highlights that artificial cavities may be used as shelters by harvestmen, mainly in human-modified

The first survey on harvestmen in Brazilian artificial cavities, with notes on distribution and natural history Introduction Subterranean cavities can be formed by the action of environmental agents (e.g.water action and eruptions) (Allred andAllred 1997, Culver andPipan 2009) or anthropogenic activities (Peck 1988, Bernardi et al. 2010, 2011, Isaia et al. 2011).Some natural cavities (paleo-burrows) were also made by the activity of large mammals extinct during the Pleistocene (Buchmann et al. 2009).
Natural subterranean cavities (karstic caves) are originated mainly by the dissolution of rock through water action, thus forming galleries (Ford and Williams 2007).In contrast, artificial subterranean cavities are generated by human action with the intention of extracting minerals or metals of economic value, for the construction of structures such as access tunnels or even for military purposes (Peck 1988, Bernardi et al. 2010, 2011, Isaia et al. 2011).
Although natural and artificial cavities present different genesis, these habitats may have similar environmental characteristics, which are primarily determined by the permanent absence of light, limited food, constantly high relative humidity and thermic stability (Ferreira 2004, Culver and Pipan 2009, Bernardi et al. 2010, 2011, Isaia et al. 2011).These environmental similarities generally enable a similar faunistic composition in these two systems (natural and artificial cavities) (Peck 1988, Ferreira 2004, Bernardi et al. 2010, 2011) in the same region.In this context, several taxa are common to both habitats, such as amphibians, arachnids, bats, crickets, moths and a large number of other invertebrates (Peck 1988, Gnaspini and Trajano 1994, Ferreira 2004, Bernardi et al. 2010, 2011, Isaia et al. 2011, Rosa and Penado 2013, Kurta and Smith 2014, Ueti et al. 2015).Among these groups, the members of the order Opiliones are also a common component inside natural and artificial cavities.
The order Opiliones Sundevall, 1833 is divided into four extant suborders, comprising 49 families and 6519 species worldwide (Kury 2011), representing the third most diverse group in the class Arachnida.In Brazil, there are 855 described species (Kury 2003).Epigean species of this order can be found in the soil, moss, leaf litter, under rocks, tree barks, stone walls, under the vegetation and in caves (Machado et al. 2007).The vast majority of species is nocturnal and need high levels of humidity to survive (Machado et al. 2007).This last condition is generally found both in natural (Culver and Pipan 2009) and artificial cavities (Peck 1988, Ferreira 2004, Bernardi et al. 2010, 2011, Isaia et al. 2011).Many harvestmen species use caves as shelters, where they may lay eggs and raise offsprings, and are often found in large aggregations on the cave walls.These species are classified as trogloxenes (e.g.Mitogoniella indistincta Mello-Leitão, 1936, Serracutisoma spelaeum (Mello-Leitão, 1933)).There are also troglophile species that may remain inside the cavities to complete their life cycle (e.g."Daguerreia inermis" Soares & Soares, 1947, Pararezendesius luridus H. Soares, 1972).The two previously conditions may be facultative, while the troglobitic species are exclusively found in caves and generally present morphological modifications, such as eye reduction, depigmentation and elongation of legs, which are considered adaptations to subterranean conditions (e.g.Spinopilar moria Kury & González, 2008, Giupponia chagasi González & Kury, 2002).
Considering the small number of faunal surveys performed in artificial subterranean cavities, the present study aimed to: (i) register harvestmen species associated with artificial cavities in Minas Gerais state, Brazil and (ii) present data regarding the species distribution, new occurrences and presence of aggregations in these environments.
Among the sampled municipalities, Alagoa, Ataléia, Caeté, Caraí, Mariana, Mateus Leme, Medina, Nova Lima, Novo Oriente de Minas, Ouro Preto, Padre Paraíso and São José da Safira are in the Atlantic Forest biome; Caeté and Nova Lima are situated in the transition between Cerrado and Atlantic Forest, while Vazante is located in the Cerrado (IBGE 2004) (Figure 1).However, most of the cavities were located in modified landscapes of these biomes, which were mainly surrounded by pastures.All collections were performed between April 2008 and November 2009.
The original names of the artificial cavities, most of them given by local residents, were maintained since there is no agency in Brazil responsible for the registration of these habitats.Besides, many names refer to the location of the cavities, what might be a helpful tool to find the locality.Since all the artificial cavities of this study are mines we will use further the term mines.
The specimens found in the mines were sampled with the aid of tweezers, after a detailed and extensive search.In order to assess the position of the species in the cavities, each observed individual was plotted in a schematic map of the mine according to the methodology proposed by Ferreira (2004).After the survey, information about the abundance and distribution of each species were obtained.In addition, the presence of loose aggregations was recorded, which were represented by groups of three or more individuals with their bodies between 0 to 5 cm apart from each other with their legs overlapping, such as proposed by Machado et al. (2000) and revised by Machado and Macías-Ordóñez (2007).
Individuals were identified whenever possible.In cases of nymphs, females and undescribed species, the individuals were just separated into morphospecies.However these morphospecies were theoretically considered as species.When there was more than one morphospecies in the same genus, they were separated using correlative numbers.
To determine the extension of the aphotic zone, a luximeter was positioned 1.2 m above the ground with its reception cell directed to the cavity entrance.The device was then conducted from the entrance to the interior of the subterranean system.The location where the luximeter presented luminosity equal to zero was considered the transition between the aphotic and the photic zone.
Specimens were deposited in the Collection of Subterranean Invertebrates of Lavras (ISLA) at the Federal University of Lavras, Lavras, Minas Gerais, and in the National Museum of the Federal University of Rio de Janeiro (MNRJ), Rio de Janeiro, both in Brazil.
The occurrences of harvestmen species are indicated in alphabetical order in the results with the abundance of each species in each locality indicated in parentheses.

Results
From all the 111 sampled mines 52 contained harvestmen distributed in 12 municipalities (Figure 1).Overall 6,066 individuals were observed, which belong to 17 species, 10 genera, 7 subfamilies and 4 families.We found different distribution and abundance patterns: from species with just one individual (e.g.Paratricommatus sp. 1, Goniosoma carum, Pseudopucrolia mutica and Eusarcus sp. 2) to broadly distributed and abundant species (e.g.Mitogoniella indistincta) presenting more than 500 individuals in 4 municipalities and 21 mines (Table 1).
Table 1.Distribution of harvestmen species in the municipalities and mines with information about their total abundance and presence (P)/absence (A) in aphotic and/or photic zones.*The abbreviated name of the species will be used to refer to each species in Table 2.The symbol "-" means the absence of data.N = abundance.This species was observed forming aggregations of 8, 19 and 26 individuals at a maximum distance of 5 meters from the entrance representing the first record of an aggregation behavior for this species.In other cavities, sparsely distributed individuals were observed, but always located in the photic zone no more than 10 meters from the entrance.These records represent a new occurrence of this species for this municipality, thus expanding its distribution with 250 km from the southwest of Juçari, Bahia state.This species was observed forming aggregations of 28, 45, 49, 103, 200, 400 and 2200 individuals at the entrance of seven cavities at a maximum distance of 5 m from the entrance.This was the first record of an aggregation behavior for this species.In other cavities, sparsely distributed individuals were also recorded, but always located in the photic zone at a maximum distance of 10 m from the entrance.During one of the fieldtrips, three aggregations with about 1000 individuals were observed.These aggregations were found in the epigean environment, next to the mine Túnel do Hotel Entre Vales I in holes and cracks in a ravine of exposed soil.Such shelters had a longitudinal form of ~ 80 cm in depth, 20 cm width and 1.5 cm height.The three municipalities represent new occurrences for this species, expanding its distribution ~ 1000 km from the south of Pernambuco state.

Family Cryptogeobiidae Kury, 2014
Paratricommatus sp. 1 Previously known distribution of the genus.Alto Paraná (Paraguai), Espírito Santo, Rio de Janeiro and São Paulo (Mello-Leitão 1940, Piza Jr 1943, H. Soares 1945, 1966, Soares and Soares 1946, 1954, 1985, H. Soares 1966, Kury 2014)  Sparse individuals were registered in the photic and aphotic zones of the mines at a maximum distance of 45 meters from the entrance.Nova Lima corresponds to a new occurrence for this species, expanding its distribution with about 40 km from the east of Santa Bárbara, Minas Gerais.However, the Mina de Capão Xavier was destroyed by the expansion of mining activities, and its fauna was translocated to another artificial mine constructed in the same region of the original one (Ferreira, R.L. pers.observation).Godinho XII (51), 12/VII/2009; 1 male (ISLA 16056) from Túnel do Milto Godinho XIII (71), 12/VII/2009 (Ázara et al. 2013).

Mitogoniella indistincta Mello-Leitão, 1936
In seven cavities, this species was observed forming aggregations of 10, 15, 18, 22, 41 and 43 individuals up to 5 meters from the entrance.In other cavities, sparsely distributed individuals were also recorded both in the photic and aphotic zones at a maximum distance of 60 m from the entrance.In some cases, individuals were found guarding their eggs and nymphs.This species also occurs in natural cavities in Minas Gerais State (Machado 2002, DaSilva and Gnaspini 2010, Ázara et al. 2013) Individuals of this species were recorded sparsely distributed inside the mines both in the photic and aphotic zones at a maximum distance of 36 m from the entrance.In some mines, specimens were found guarding their eggs and nymphs.This species also occurs in natural cavities in Minas Gerais state (Ázara et al. 2013).Examined material.Novo Oriente de Minas: 1 male (MNRJ 2338) from Mina do João Bufinha (4), 05/I/2009, new record for locality.

Liogonyleptoides inermis (Mello
Individuals were sparsely distributed inside the mines, both in the photic and aphotic zones at a maximum distance of 10 m from the entrance.Novo Oriente de Minas corresponds to a new occurrence for this species, expanding its distribution with about 700 km from the northeast of Petrópolis, Rio de Janeiro state.
Only one individual was recorded in the photic zone of the cavity, at less than 10 m from the entrance.The record at Caraí expands the distribution of this species with about 750 km from the southwest of Salvador, Bahia.Individuals of this species were sparsely distributed in the photic and aphotic zones of the mine at a maximum distance of 30 m from the entrance.The records at Caeté and Mariana expand its distribution with about 60 km from the northwest of Ouro Preto, Minas Gerais state.
Examined material.Alagoa: 1 male (MNRJ 2374) from Mina da Companhia (7), 01/XI/2008; 1 male (MNRJ 2342) from Túnel do Garrafão (15), 31/X/2008, new record for state and locality.Individuals of this morphospecies were sparsely distributed in the photic and aphotic zones at a maximum distance of 12 m from the entrance.The record at Alagoa expands the distribution of the genus with about 150 km from the west of Teresópolis, Rio de Janeiro state.
Individuals of this species were observed sparsely distributed inside the mine.They were located in the photic and aphotic zones at a maximum distance of 500 m from the entrance.The record at Vazante expands its distribution with about 200 km from the south of Unaí, Minas Gerais.This species also occurs in natural cavities in Bahia, Goiás and Minas Gerais state (Hara and Pinto-da-Rocha 2010).

Eusarcus sp. 1
Previously known distribution of the genus.From the northeast to the southeast of Brazil, northeast of Argentina, east of Paraguai and Uruguai (Hara and Pinto-da-Rocha 2010) Individuals were observed sparsely distributed inside the cavities both in the photic and aphotic zones, at a maximum distance of 22 m from the entrance.Individuals were also observed next to the touristic path.The records at Mariana and São José da Safira correspond to new occurrences for the genus.

Eusarcus sp. 2
Previoulys known distribution of the genus.From the northeast to the southeast of Brazil, northeast of Argentina, east of Paraguai and Uruguai.(Hara and Pinto-da-Rocha 2010).
Only one female was found and the record of Mateus Leme corresponds to a new occurrence for this genus.
Only one male was found and the record of Alagoa corresponds to a new occurrence for this genus.

Discussion
This study presented the occurrence of two undescribed species (Eusarcus sp. 1 and Mitobatula sp. 1) and new records for states and localities in the case of nine species and four genera of harvestmen.The results obtained demonstrate the importance of observations in unusual environments, like mines for the potential of finding new species, provide more information about the occurrence of species and register new behavioral patterns, thus improving the knowledge about the group.
Although there are studies in Brazil and worldwide about the fauna associated to mines (Peck 1988, Gnaspini and Trajano 1994, Ferreira 2004, Bernardi et al. 2010, 2011, Isaia et al. 2011), most of them are focused on community ecology.Until present only few groups received special attention in studies of mines: anurans (Rosa and Penado 2013), bats (Kurta and Smith 2014) and salamanders (Manenti et al. 2009, Manenti and Ficetola 2013, Balogová and Uhrin 2014, Herrero and Hinckley 2014).
The age of the majority of the mines in our study varied from less than five years to more than 100 years (Table 2).Twenty of the 52 cavities containing harvestmen have only pastures surrounding them (30 and 250 m radius).The other cavities present different phytophysiognomies arround them, such as Atlantic Forest, Caatinga and Cerrado fragments, Eucalyptus plantations and pastures.
As shown by Bragagnolo et al. (2007), richness and abundance of harvestmen species in the Atlantic Forest are negatively affected by fragmentation, reduction of fragment size and forest quality (stratification and leaf density).The replacement of natural environments by areas with mineral exploitation can destroy the habitats.However, mines apparently might ensure the permanence for a number of species that otherwise might get locally extinct through the destruction of epigean habitats (Ferreira 2004).However, it is still unknown how populations inside mines surrounded by pastures and small fragments can maintain.
The results of this work show that harvestmen can use mines as shelters during the day and for reproduction.These mines are important because many of the studied regions are fragmented, with forest areas replaced by pastures and agricultural land.Future studies can demonstrate whether harvestmen feed in the pastures or use the organic material inside the mines.
The presence of Eusarcus cavernicola individuals, a species commonly found in caves (Hara and Pinto-da-Rocha 2010), inside of the mine Mina da Vmetais (500 m from the entrance) suggests that this habitat is a permanent shelter for the species, due the presence of the food resource in the cavity.
Further studies should be concentrated on understanding the dynamics of species inside mines and the taxonomic identification.Artificial cavities can contribute to the conservation of some harvestmen species in many areas where the original vegetation was removed and replaced by pastures, monocultures or constructions.

Figure 1 .
Figure 1.Map of the state of Minas Gerais, Brazil, showing the biomes and municipalities in which the artificial cavities are situated: a Alagoa b Ataléia c Caeté d Caraí e Mariana f Mateus Leme g Medina h Nova Lima i Novo Oriente de Minas j Ouro Preto k Padre Paraíso l São José da Safira m Vazante.

Figure 2 .
Figure 2. Artificial cavities in: A, B, C Alagoa D and E Padre Paraíso F Caraí G Mariana H São José da Safira I Caeté J Mariana.

Figure 3 .
Figure 3. Gryne perlata A Individual inside the mine Túnel dos Meninos II, Padre Paraíso B aggregation of individuals in the same cavity.

Figure 5 .
Figure 5. Mitogoniella mucuri inside the Túnel do Garrafão (Alagoa) A Female guarding eggs on the wall of the mine B Female guarding immatures on the wall of the mine.

.
Examined material.Caeté: 1 female (MNRJ 2266) from Mina do Morro Vermelho II (1), 19/VI/2009, new record for state and locality.Only one female was found.This is a new occurrence in the municipality, expanding the genus distribution with about 300 km from the southwest of Colatina, Espírito Santo state.