A new species of Utiaritichthys Miranda Ribeiro ( Characiformes : Serrasalmidae ) from the Serra dos Parecis , Tapajós drainage

Utiaritichthys esguiceroi is described from the upper portion of the rio Juruena, rio Tapajós drainage, Mato Grosso State, Central Brazil. The new species distinguished from its two congeners, U. sennaebragai Miranda Ribeiro and U. longidorsalis Jégu, Tito de Morais & Santos, by having 99 to 101 perforated scales on lateral line (vs. 69 to 83), presence of 17 to19 prepelvic spines (vs. 9-13 in U. sennaebragai and 28-31 in U. longidorsalis), 20 to 21 postpelvic spines (vs. 15 to 19 in U. sennaebragai, and 14 in U. longidorsalis), and 23 to 25 circumpeduncular scales (vs. 30-48 in U. sennaebragai, and 33-35 in U. longidorsalis). Furthermore, the new species differs from U. longidorsalis by having larger interdorsal width, and adipose-fin base length (11.8-15.6 vs. 7.1-7.9% of SL, and 4.2-5.8 vs. 3.7-3.8% of SL, respectively).

The real taxonomic level of the group remains unsolved.Machado-Allison (1983, 1986) proposed the monophyly of the serrasalmids, based on 27 morphological characters.Subsequently, in molecular phylogenies, Orti et al. (1996) and Calcagnotto et al. (2005) corroborated the monophyly of the group.Calcagnotto et al. (2005) based on their molecular tree, raised the group to the family level, disagreeing with Machado-Allison (1982, 1983, 1985, 1986) and Jégu (2003), that considered the clade in question as a subfamily of Characidae.
Utiaritichthys sennaebragai was described by Miranda Ribeiro (1937) from the rio Papagaio, rio Tapajós drainage, upriver from Salto de Utiariti (waterfall), Mato Grosso State, Central Brazil.Jégu et al. (1992) redescribed the genus and described an additional species, U. longidorsalis, from the rio Aripuanã, rio Madeira drainage, also in Central Brazil.Utiaritichthys sennaebragai is recorded in the rios Tapajós, Xingu, Tocantins-Araguaia and Trombetas, in the Amazon drainage, and also in the río Orinoco, in Venezuela; and U. longidorsalis Jégu, Tito de Moraes & Santos, 1992, is known only from the rio Madeira.A new species of Utiaritichthys was discovered during recent collections made in the rio Juruena, rio Tapajós drainage, in the region of the Chapada dos Parecis, Mato Grosso State, Brazil, and it is described herein.

Material and Methods
The counts and measurements followed Jégu et al. (1992).Measurements were taken point to point with the aid of calipers on the left side of the specimens, whenever possible.All measurements are expressed as percentages of standard length (SL), except for subunits of the head, which are expressed as percentages of the head length (HL).Vertebrae counts were made based mostly on 69 X-rayed specimens and two cleared and stained (c&s) specimens prepared according to Taylor & Van Dyke (1985).Skeletal meristic data, numbers of supraneurals, gill-rakers in the first gill arch, teeth, and procurrent caudal-fin rays, were also taken from cleared and stained material.Examined specimens are deposited in the ichthyological collection of the Laboratório de Ictiologia de Ribeirão Preto, Faculdade de Filosofia, Ciência e Letras da Universidade de São Paulo; Ribeirão Preto (LIRP); Museu de Zoologia da Universidade de São Paulo, São Paulo (MZUSP), and Museu de Ciências e Tecnologia da Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre (MCP).Asterisk indicates holotype data and numbers in parentheses indicate data from paratypes plus non-types specimens examined.All the non-type material listed in the description was formerly utilized in an independent study of stomach content analysis and sexual determination, having had their visceral cavities evacuated and general body shape and proportions distorted, as a result.

Description
Morphometric data presented in Table 1.Body deeply compressed, with rounded dorsal and ventral profiles.Greatest body depth at dorsal fin origin.Dorsal profile of head distinctly rounded from upper lip to vertical through anterior nostrils, convex from latter point to tip of supraoccipital spine.Predorsal profile slightly convex from tip of supraoccipital spine to dorsal-fin origin; posteroventrally inclined along dorsal-fin base; slightly convex from posterior dorsal-fin base to adipose-fin origin and very slightly concave along caudal peduncle.Anterior portion of ventral profile of head almost straight and posteroventrally inclined; posterior portion of head and body convex, strongly inclined posterodorsally along anal-fin base and slightly concave along caudal peduncle.
Mouth terminal, with spatulate teeth.Maxilla toothless.Posterior portion of maxilla wider than anterior.Premaxilla with two rows of teeth: labial row with three teeth; first two teeth tricuspid and third truncated: lingual row with four slightly incisform teeth, all with base wider than crown.Dentary teeth five: first four teeth tricuspid, and last tooth truncated: teeth decreasing gradually in size in outer tooth row and a single blunt conical, symphyseal tooth on inner row (Fig. 2): gill-rakers 12/1/18* (holotype + 2 c&s).

Color in alcohol.
Overall body background color light brown.Dorsal portion of head and predorsal region dark brown.Dark brown patches on first to fourth and sixth infraorbitals.Yellow pale background on fifth infraorbital.Dark vertically elongated spot on posterior portion of opercle.A single, very inconspicuous humeral blotch.Flanks with a dark brown narrow stripe extending from posterior border of opercle to caudal fin.Five to six large vertical bands along dorsal portion of flank; first through third more conspicuous.Ventral portion of body pale yellow and white.Dorsal and anal fins hyaline with ray tips with dark chromatophores.Yellowish white  Color in life.Based on LIRP 8157, 193.0 mm SL. Background body color greenish silver, lightly golden over facial bones.Ventral portion of body white.All fins green; dorsal fin dark green, with hyaline ray tips forming a transparent narrow margin.Anterior anal-fin rays dark, becoming progressively light towards posterior.Anterior borders of pectoral and pelvic fins darker than posterior borders.Caudal-fin lobes and base darker than their respective margins.Iris golden with black dorsal and ventral areas.
Sexual dimorphism.Juveniles specimens lacking secondary sexual characters.Adult specimens (188.0-228.3mm SL) with distinct secondary sexual characters on analfin.Males specimens with anteriormost nine rays longer than additional rays forming a distinct anterior lobe, and elongation of 11 th to 28 th branched rays forming second bell-shaped lobe.This condition contrast with that of females in which the anteriormost 12 anal-fin rays are distinctly longer than other posteriormost rays, forming a distinct lobe.It remains unknown at this point whether such modifications are permanent or restricted to reproductive seasons.
Ecological notes.Esguicero (pers.comm.), based on the examination of the stomach contents of approximately 50 specimens (non-type material), found that juveniles of Utiaritichthys esguiceroi fed on aquatic and terrestrial insects and particulate organic matter, whether the adults fed almost exclusively on Podostemaceae macrophytes and filamentous algae, both typical of riffles and rapids.Furthermore, juveniles inhabit the calm portions of the rio Juruena -usually near the margins-, whereas the adults swim in the main river channel, mostly in riffles and rapids.It is noteworthy that Utiaritichthys esguiceroi diet and mesohabitat preference ontogenetic shifts are similar to the same shifts observed by Jégu et al. (1989) in Mylesinus paraschomburgkii Jégu, Santos & Ferreira.
Etymology.The specific epithet is named after André L. H. Esguícero, collector of the holotype and paratypes of the new species.Remarks.Notwithstanding the unresolved monophyly and taxonomic definition of Utiaritichthys, all the species of Utiaritichthys (including the one described herein) present non-parallel premaxillary teeth rows (see Fig. 2), as already pointed by Jégu et al. (1992) and Géry 1972 -and also lack, as the species presently allocated in Myloplus, the unique characters possessed by the Serrasalminae genus Myleus, Tometes, Mylesinus, and Ossubtus cited by Jégu et al. (1992).
Considering the all problems cited before, our decision to allocate the new species described herein in the genus Utiaritichthys is based on the fact that all our examined specimens share all the Utiaritichthys characters listed in the genus redescription, differing from the Myloplus only by a comparatively longer body (see Jégu et al., 1992).Furthermore, since an unequivocal phylogenetic definition of Myloplus is currently lacking, we believe that is most prudent to allocate the new species in the genus Utiaritichthys.
Specimens of Utiaritichthys esguiceroi and U. sennaebragai were both collected in the upper rio Tapajós drainage, although both species seem to be geographically separated by a major waterfall, at least 70 m high.
Although Utiaritichthys esguiceroi is morphometricallly very similar to Utiaritichthys sp.n. of Jégu et al. (1992: 116, table II) based on a single specimen deposited in the Museum National d'histoire naturelle, Paris (MNHN 1991-704) and collected in Fleuve Sinnamary, Guyana, the absence of mandibular simphyseal teeth, the lower number of perforated scales (83) on the lateral line, and the higher numbers of prepelvic (28) and postpelvic (14) spines, clearly separate Utiaritichthys esguiceroi from that known but undescribed species.
The apparent endemicity of Utiaritichthys esguiceroi, together with the respective endemicities of Jupiaba paranatinga Netto-Ferreira, Zanata, Birindelli & Sousa, Astyanax ajuricaba Marinho & Lima and Hyphessobrycon hexastichos Bertaco & Carvalho reinforces the recommendation of Carvalho & Albert (2011) to scientifically explore the ichthyofauna of the highly endemic area of the huge upper rio Tapajós basin.This exploration must done very quickly, due to the major and crescent anthropic impacts being inflicted to that area.All the fast flowing water species of the Serrasalmidae representatives (i.e., Acnodon, Mylesinus, Ossubtus, and Utiaritichthys) are most likely seriously threatened by at least ten hydroelectric dams, either in construction, or planned for the upriver and fast flowing sections of the major tributaries of the rio Amazonas right margin (Brasil -MME, 2011), like the rios Madeira, Tapajos, and Xingu (see Jégu et al., 1992;Zuanon & Jégu, 2008;Vieira et al., 2008).The 50 non-type specimens examined on "Ecological Notes" were dissected for stomach content analysis and, consequently, had their respective visceral cavities emptied, strongly altered body proportions making them unsuitable for taking morphometric data.Nonetheless, no significant differences were found in their meristic and coloration in comparison with type-specimens.
Range, Mean and Standard Deviation (SD) include values of the holotype (H); n = number of specimens.pectoralfin rays, becoming hyaline toward tips; yellowish white pelvic-fin rays.Yellowish white caudal-fin rays, with mostly brown tips.