Rumikiru, n. gen. (Scorpiones: Bothriuridae), a New Scorpion Genus from the Atacama Desert

ABSTRACT Rumikiru, n. gen., a new bothriurid scorpion genus from the coastal Atacama Desert, Chile, is described. This is the first scorpion genus endemic to northern Chile. It is most closely related to Pachakutej Ochoa, 2004, from the inter-Andean valleys of Peru. Orobothriurus lourencoi Ojanguren-Affilastro, 2003, is transferred to the new genus and redescribed, creating Rumikiru lourencoi (Ojanguren-Affilastro, 2003), n. comb., and a second species of the genus, Rumikiru atacama, n. sp., is described.


INTRODUCTION
The scorpion fauna of Chile is largely isolated from that of the rest of South America by the high altitudes of the Andes to the south and the extreme desert of the Pacific coast to the north. This isolation has created a high level of endemism in the country, especially in southern Chile, where there are three endemic scorpion genera in the family Bothriuridae Simon, 1880: Centromachetes Lonnberg, 1897, Phoniocercus Pocock, 1893, and Tehuankea Cekalovic, 1973 These genera occur almost exclusively in the cold to temperate forests of the area, and are related to the endemic Australian bothriurid genus Cercophonius Peters, 1861 (Prendini, 2000. Another endemic Chilean scorpion genus, Caraboctonus Pocock, 1893 (family Iuridae Thorell, 1876), has a widespread distribution in the central part of the country (Louren^o, 1995;Agusto et al., 2006). Until now, however, the scorpion fauna in the arid zone of northern Chile appeared to be similar in generic composition to that in the arid zone of southern Peru, and also to the Andean scorpion fauna, no endemic genera having been recorded there (Maury, 1975;Ochoa and Acosta, 2002;Ochoa, 2005;Ochoa and Ojanguren-Affilastro, 2007;Ojanguren-Affilastro and Ramirez, 2009;. The absence from Chile of indigenous scorpions in the family Buthidae C.L. Koch, 1837, which occur in all other Neotropical countries, is of particular interest. The high altitudes of the Andes, together with the aridity of the Atacama Desert, apparently form an effective barrier to the family at these latitudes. A buthid, Tityus chilensis Louren^o, 2005, was described from the northeastern extreme of Chile (an area above 4000 m, in the Andes Altiplano Region), based on old material from the Museum National d'Histoire Naturelle, Paris, ambiguously labelled "Bolivia, West of Charana" (Louren<;o, 2005) and apparently originating from an area disputed by Bolivia, Chile, and Peru since the Pacific War (1879-1883). We have extensively surveyed this area, however, and confirm that no buthids occur either in the Altiplano or the high Andes of Bolivia and Chile, an area of exceptional aridity. The highest record for Tityus C.L. Koch, 1836, that we know of in this region is 3170 m, on the eastern slopes of the Andes (pers. obs.). The material described by Louren^o (2005) probably originates from the more humid, eastern slopes of the Andes in Bolivia.
We conducted several expeditions to Chile during the past decade, resulting in the collec¬ tion and description of various new bothriurid taxa, especially from the northern part of the country (Ojanguren-Affilastro, 2003bMattoni and Acosta, 2006;Ojanguren-Affilastro and Mattoni, 2006;Ojanguren-Affilastro et al., 2007a, 2007b. Among these was Orobothriurus lourencoi Ojanguren-Affilastro, 2003, from the Atacama Desert of Chile, a distinctive species, possessing several morphological characters unique among Bothriuridae. At the time of its description, the systematics of Orobothriurus Maury, 1976, was poorly understood, and it was divided into two species groups, the alticola group and the inca group (Maury, 1976;Ochoa, 2000, 2001). Based on the definition of Orobothriurus at the time, and without the support of phylogenetic analysis, O. lourencoi was placed in Orobothriurus and tentatively assigned to the inca group, although Ojanguren-Affilastro (2003a: 118, 119) noted that it pos¬ sessed several characters that did not fit the diagnosis. Ochoa (2004) revised Orobothriurus and created a new genus, Pachakutej Ochoa, 2004, to accommodate the species of the inca group of Orobothriurus. Ochoa (2004) did not include O. lourencoi in his study, however, so this species remained in Orobothriurus by default. A recent revision of Orobothriurus (Ochoa et al., 2011), together with a reanalysis of the phylogeny of Orobothriurus and Pachakutej (Mattoni et al., in press), and the discovery of a new species, sister to O. lourencoi, demonstrated that the two species do not belong in either genus, and required the creation of a new genus, Rumikiru, n. gen., to accommodate them. Orobothriurus lourencoi and its sister species exhibit several synapomorphies with Pachakutej, but also several unique synapomorphies ( fig. 1, appendix 1, and Diagnosis and Relationships below) that justify their placement in a new genus.
Rumikiru, n. gen., is the first bothriurid genus endemic to northern Chile ( fig. 2). It is most closely related to Pachakutej from the inter-Andean valleys of Peru ( fig. 3). The presence of a new genus in this area together with the discovery of additional new species from the Coquimbo Region, which probably belong to another new genus (Mattoni et al., in prep.), suggest higher levels of diversity and endemism in the scorpion fauna of northern Chile than previ¬ ously recognized. In the present contribution, we describe Rumikiru, n. gen., transfer O. lou¬ rencoi to it, creating Rumikiru lourencoi (Ojanguren-Affilastro, 2003), n. comb., and describe a second species of the genus, Rumikiru atacama, n. sp. Measurements, taken using an ocular micrometer, were recorded in millimeters. Descrip¬ tive terminology follows Mattoni and Acosta (2005) for hemispermatophores; Vachon (1974) for trichobothria;   Etymology: The indigenous peoples of the Chilean Atacama originally spoke Kakan or Kunza, depending on the region and, after the Inca invasion, added Quechua and Aymara to their languages. Rumikiru is formed from two Quechua words, rumi meaning "stone" and kiru meaning "tooth," and refers to the large denticles on the movable fingers of the pedipalp chelae in this genus, unique in the family Bothriuridae, and to the habitat of its two species, both of which appear to be restricted to rocky slopes. Diagnosis: Species of Rumikiru, n. gen., are easily recognized by the enlarged basal den¬ ticle of the median denticle row of the pedipalp chela movable finger (figs. 18 A, 19 A, 24C, 25 A, C), which is approximately three times larger than and replaces the first five or six median denticles. Such a hypertrophied basal denticle is unique among bothriurids. An enlarged basal denticle on the movable finger is exhibited by some bothriurids, e.g., Brachistosternus ehrenbergii (Gervais, 1841), but replaces no more than two or three denticles (Ochoa and Ojan¬ guren-Affilastro, 2007). The submedial position of the apophysis on the internal surface of the pedipalp chela manus in the adult male of Rumikiru,n. gen. (figs. 18C,D,22 A,C,D,24A,B,25 A,B), is also unique in the family. The apophysis is situated in the distal third of the surface in all other bothriurid taxa in which it occurs, and is absent from the pedipalp chela manus of basal bothriurids, e.g., Lisposoma Lawrence, 1928, and Thestylus Simon, 1880(Prendini, 2000. The shape of the apophysis and the position of trichobothium ib distal to it are also unique to Rumikiru, n. gen. The new genus may be further separated by a twist in the medial to distal third of the dentate margin of the pedipalp chela fingers, which abruptly alters the orientation of the median denticle row (more conspicuously on the movable finger), in the male (figs. 18A, 22A), that is absent in other bothriurid genera. Additionally, the carapace of Rumikiru, n. gen., is dorsoventrally compressed ( fig. 12C, D), unlike many other bothriurids, in which it is more convex.

METHODS
Rumikiru, n. gen., is most closely related to Pachakutej, based mainly on characters of the hemispermatophore ( fig. 10). The two genera share the presence of one sclerotized apophysis on the internal fold of the internal lobe, and a papillose fold in the basal lobe (figs. 10C, H). Species   of Pachakutej, however, possess a unique synapomorphy, a spatulate terminal process on the basal lobe of the hemispermatophore (Ochoa, 2004) that is absent in Rumikiru, n. gen., and are mark¬ edly more pigmented. The pedipalp carinae are more pronounced, especially on the femur and patella, in Rumikiru,n. gen. (figs. 16,17,20,21), than in Pachakutej. The two genera may also be separated by the trichobothrial pattern of the pedipalp chela manus. Trichobothrium V2 is situ¬ ated in the same axis as Vx and V3 in Rumikiru, n. gen. Ecology. In addition to its unusual mor¬ phology, Rumikiru, n. gen., occupies an unusual habitat among bothriurid scorpions.
Most bothriurids are fossorial, requiring exposed soil to construct burrows. However, all personally collected specimens of Rumikiru, n. gen., were found in rocky habi¬ tats, with almost no exposed soil. At Pan de Azucar National Park, the type locality of R. lourencoi, n. comb., several specimens were captured on vegetationless scree slopes com¬ prising piles of sharp, loose stones accumu¬ lated below steep cliff faces. At Llanos de Challe National Park, specimens of R. atacama, n. sp., were collected in a slightly more vegetated environment, but this species was common only on scree slopes, cliff faces, and exposed rocky outcrops ( fig. 4A).
The environment inhabited by Rumikiru, n. gen., differs markedly from that of its sister genus, Pachakutej, which occurs under stones in more humid habitats, in inter-Andean valleys and montane rainforests on the eastern slopes of the Andes in Peru (Ochoa, 2004).
Rumikiru, n. gen., displays an unusual defensive behavior, not reported among other scor¬ pions, which typically curve the metasoma and strike out with the aculeus and, in some species, the pedipalp chelae, when threatened: males spread the pedipalps wide open, making short pulses or vibrations with them, without curving the metasoma ( fig. 4B).
Rumikiru atacama, n. sp.    Etymology: The specific epithet, atacama, is a noun in apposition, referring to the Chilean Region III, to which this species appears to be endemic. This region forms part of the Atacama Desert, which extends from southern Peru to northern Chile. Diagnosis: Rumikiru atacama, n. sp., can be separated from the only other known species of the genus, R. lourencoi, n. comb., by several morphological characters. The distal lamina of  fig. 16D) or comprising scattered granules (9). Patella, intercarinal surfaces densely granular (d) or smooth (9); DI, VI, and VE carinae granular, extending entire length of segment, DI carina especially pronounced and coarsely granular (d; figs. 17A, D) or obsolete, reduced to few scattered granules (9); external margin undulated ( fig. 17C), DE and EM carinae obsolete, reduced to few granules (d; fig. 17A, B) or DE carina absent, EM carina obsolete, reduced to slight curvature of surface along entire length of segment (9); IM carina granular, well developed along entire length in some d ( fig. 17D) but reduced to scattered granules in others, and absent in 9. Chela manus prism shaped, more robust in d (figs. 18, 19, 24), length/width ratio 2. 35-2.78   Ecology: The area in which this species was collected falls within the "Desierto Costero del Huasco" subregion of the "Desierto" botanical region (Gajardo, 1993). This area is extremely arid, with sparse shrubs and cacti, except in areas with greater exposure to sea fog, where a "Lomas" habitat, comprising more abundant vegetation, occurs. Rumikiru atacama, n. sp., has been collected close the seashore but not in "Lomas" habitat. Most personally collected speci¬ mens were taken from scree slopes, cliff faces, and exposed rocky outcrops, in areas with little or no vegetation ( fig. 4A). None of these species shares the same microhabitat as R. atacama, n. sp., however.
Ecology: The area where this species was collected falls within the "Desierto Costero de Tal-Tal" subregion of the "Desierto" botanical region (Gajardo, 1993). This area is extremely arid, with sparse shrubs and cacti, except in areas with greater exposure to sea fog, where a "Lomas" habitat, comprising more abundant vegetation, occurs. This species was collected in the rocky scree slopes of the "Quebrada Pan de Azucar," 8-10 km inland from the coast, in areas with almost no vegetation.

ACKNOWLEDGMENTS
We are grateful to Ivan Benoit (Corporacion Nacional Forestal del Gobierno de Chile, CONAF) for assistance with obtaining permits to collect scorpions in Chilean National Parks, and to the CONAF authorities for issuing the permits; Fermin Alfaro-Kong, Cristian Grismado, Paula Korob, and Jaime Pizarro-Araya for participating in the expeditions to Chile; the following curators and collections managers for donating, loaning and/or granting access to material from