On the Taxonomy of Acanthoscurria Ausserer from Southeastern Brazil with Data on the Natural History of A . gomesiana Mello-Leitão ( Araneae , Mygalomorphae , Theraphosidae )

The study of type material and specimens of A. gomesiana of several Brazilian spider collections offered us the possibility to redescribe this species and consider synonyms of the first one of the following: A. violacea, A. pugnax, and A. aurita. Acanthoscurria cunhae and A. melanotheria are considered Species inquirendae, the types were not located, and the descriptions are not enough for identification of similar ones. We considered as valid only two species from the southeast of Brazil: A. paulensis and A. gomesiana. The distribution range of A. gomesiana is enhanced in the state of Minas Gerais, Brazil. Some data on natural history and phenology are presented.


Introduction
The genus Acanthoscurria Ausserer, 1871 currently comprises 34 species and is distributed mainly in South America [1].The genus is characterized by the presence of stridulating bristles on the retrolateral face of the trochanter of the palp, presence of only one tibial apophysis and a retrolateral nodule on palpal tibia of males and seminal receptacle of the female with a common base and two lobes, more or less evident [2,3].Nine species from the southeastern Brazil were described, the majority by Mello-Leitão [4].In 1923, he described six species: Acanthoscurria cunhae, A. chiracantha,A. violacea,A. paulensis,A. gomesiana,and A. melanotheria. Vellard [5] described A. pugnax, and Piza [6,7] described two other species, A. aurita and A. guaxupe.The study of specimens from several localities of southeastern Brazil, from expressive Brazilian collections, and the type material of A. gomesiana enables us to redescribe and consider synonyms of the following species: A. violacea, A. pugnax, and A. aurita.The other two species, A. cunhae and A. melanotheria, were considered Species inquirendae because the types are lost and the descriptions are too resumed to identify the species.Acanthoscurria guaxupe was considered a junior synonymy of A. paulensis [8] and A. chiracantha of A. natalensis Chamberlin, 1917 [9].We concluded that A. paulensis and A. gomesiana are the only valid species for the southeastern Brazil.The distribution range of A. gomesiana is enhanced.Some data of natural history, based on the field observations of the species A. gomesiana, were given.Also some phenological data through the analysis of the specimens received at the laboratory are presented.
Spine notation follows Petrunkevitch [10].Terminology of male palpal bulb follows Bertani [11].All measurements are in millimeters.Female seminal receptacles were dissected and cleared in lactic acid for observation of internal structures.The length of leg segments was measured between joints in dorsal view.Length and width of carapace, eye tubercle, labium, and sternum represent maximum values.Total body length excludes pedicel and spinnerets.Pictures were taken with a Leica DFC500 digital attached to a Leica MZ16A stereoscopic microscope.
The natural history data were obtained from 715 specimens deposited in the IBSP collection.For studies on the phenology were used data of the specimens received at laboratory during the last 64 years , from the states of São Paulo and Minas Gerais, Brazil.

Acanthoscurria violacea
Variation (N = 20).In some specimens, the sternum of A. gomesiana is very convex, and in others, it is slightly convex.Total length: male 21.2-36.2and female 31.2-48.3.The general color of this species could be deep dark until lighter brown (Figures 1(a)-1(d)).
Remark 1. Mello-Leitão [4] described A. gomesiana based on several specimens from two districts of São Paulo city: Ipiranga and Butantã, São Paulo, Brazil.Part of the type material was deposited in the collection of the MZSP and an other in MNRJ, and he did not choose a holotype.Schiapelli & Gerschman de Pikelin [3] examined the type material from both collections and erroneously cited a holotype, a male, instead of a lectotype and paralectotypes.From all type materials, only the male MNRJ 49 was located, the specimen has attached a label, handwritten by Mello-Leitão: typus from São Paulo, and so we elected the same as lectotype.
Based on the study of the type material of A. gomesiana, descriptions of the species, redescription and illustrations of the sexual organs, male palpal bulb and seminal receptacle of female given by Schiapelli & Gerschman de Pikelin [3] of A. gomesiana and A. violacea, and exam of specimens of several collections, A. violacea is considered a junior synonym of A. gomesiana.The option for A. gomesiana, although A. violacea is described some pages before, in the same paper, is due to the fact that A. gomesiana is the only type material available and is a common cited species, and A. violacea never more was cited since 1964 [3].
Acanthoscurria pugnax is considered a junior synonym of A. gomesiana based on the description, measurements, and spinulation of the legs, and figures of the male palpal bulb, tibial apophysis, and general dorsal aspect are given by Vellard [5], which agree with A. gomesiana.Also, the study of specimens of the type locality, Belo Horizonte, Minas Gerais, Brazil confirms the synonymy.The exam of the holotype of A. aurita confirms the junior synonymy with A. gomesiana which shares the same general aspect, mainly the morphology of the sexual organ.

Natural History
This species occurs in the Atlantic Forest (Figures 3(a by Pérez-Miles et al. [12] in A. suina Pocock, 1903, where the subcircular entrance was followed by an abrupt profile (step-shaped) vertical tube, continued by a horizontal great chamber.A female of A. gomesiana living in an anthropized area of the PEJ (Figures 3(b), 4(c), and 4(b)) was observed during the day digging its burrow with its palps and legs I-II, as occurs in some other Theraphosidae [12,13]; granular soil was found near the entrance (Figure 4(d)) as occurs in the entrances of A. suina and Eupalaestrus weijenberghi Thorell, 1894 [12].
The observations were made with three females in the field and seven in captivity.The period in which they made the egg sacs was from December to February, and the hatching of spiderlings occurred from February to March.We did not observe the construction of egg sacs to determine if the spider deposited urticating setae on it, as reported by Melchers [14] and Marshall & Uetz [15], but the silk used to construct it was examined and had many type I setae throughout the egg sac structure.Females took care of egg sacs during 50-70 days until the spiderlings hatch (Figures 5(a) and 5(b)).The number of spiderlings ranged from 300 to 570 (Figures 5(c)-5(e)), which represents an average number if compared to other theraphosids species [16].The measurement of oval egg sac shape was 6.5×5.0 cm (n = 470 spiderlings and about 40 nonhatched eggs).The mother care of the spiderlings was observed until the second instar was completed.
These spiders are aggressive and, when disturbed, they bite and easily throw urticating setae, and they can also display defensive behavior by raising the anterior legs, palp, and opening the chelicerae, to expose the red ventral setae (Figure 5(f)).Also they can release poison droplets or give false attacks, mainly males, like A. suina [12].It is the most commonly found Theraphosidae species in the metropolitan region of São Paulo, inhabiting peridomiciliary areas [17].Due to these factors it is the Mygalomorphae species which causes the greatest number of accidents in humans in the state of São Paulo, and all accidents presented no complications.The accidents caused only local pain [18].
For phenology were included data from 715 specimens of A. gomesiana from 646 municipalities from the state of São Paulo and 69 from the state of Minas Gerais.From the state of São Paulo, 249 specimens were obtained from metropolitan region of the municipality of São Paulo (Figure 7(b)).
Males presented the reproductive period in March, April, and May, in late Summer and early Fall, with a peak in April (Figures 7(a

Species Inquirendae
The descriptions of the two species were based on color, eyes distance, and other characters and nowadays not considered so important in taxonomy, without figures of the sexual structures, and does not allow an identification of the two species.So we considered the two, A. cunhae and A. melanotheria, Species inquirendae.Mello-Leitão, 1923.Mello-Leitão, 1923: 282, figures 55-59, 168.Type material.Holotype male from Belo Horizonte, Minas Gerais, Brazil.Almeida Cunha leg.should be deposited in the MNRJ 1397, not located, considered lost.Mello-Leitão, 1923.Acanthoscurria melanotheria Mello-Leitão, 1923: 310.Type material.Holotype female from Minas Gerais, Brazil.Handwritten note on the collection card, Goias, Brazil, should be deposited in the MNR 1398, not located, considered lost.