A new species of Hyphessobrycon ( Teleostei : Characidae ) from the upper rio Tocantins drainage , with bony hooks on fins

* Laboratório de Ictiologia, Museu de Ciências e Tecnologia, Pontifícia Universidade Católica do Rio Grande do Sul, Av. Ipiranga 6681, 90619-900 Porto Alegre, RS, Brazil. e-mail: ubertaco@pucrs.br ** Departamento de Zoologia, IB, Universidade Federal do Rio Grande do Sul, Av. Bento Gonçalves, 9500, 91501-970 Porto Alegre, RS, Brazil. e-mail: malabarb@pucrs.br A new species of Hyphessobrycon (Teleostei: Characidae) from the upper rio Tocantins drainage, with bony hooks on fins


Introduction
Hyphessobrycon is a speciose genus of the family Characidae including more than 100 valid species (Lima et al., 2003;Lima & Moreira, 2003;Lucena, 2003) distributed from southern Mexico to the río de La Plata in Argentina.The genus was proposed by Durbin in Eigenmann (1908:100) as a subgenus of Hemigrammus Gill, differing from the later by the absence of scales on the caudal-fin.Hyphessobrycon was extensively revised by Eigenmann (1918Eigenmann ( , 1921) ) whose accounts still constitute the single comprehensive review of the then known members of the genus.Publications on Hyphessobrycon subsequent to Eigenmann's revision usually involve only the description of new species occurring in restricted geographic areas of South and Central America.The recognition of groups of species in the genus is based primarily on similarities of color patterns (e.g.Géry, 1961;Géry, 1966;Géry, 1977).Weitzman & Palmer (1997) recently hypothesized the existence of a monophyletic group of Hyphessobrycon species, based on the coloration pattern and the shape of the dorsal and anal fins of males that they termed the rosy tetra clade.Those authors did not, however, present a hypothesis for the resolution of the relationships of the remaining species in the genus.We herein describe a new species of Hyphessobrycon from the upper rio Tocantins basin.

Methods and Materials
Counts were taken as described by Fink & Weitzman (1974) with the exception of the number of scale rows below lateral line that were counted from the scale row ventral of the lateral line to the scale row closest to the first pelvic-fin ray.Vertebral counts, supraneurals, gill-rakers, teeth and procurrent caudal-fin-ray counts were taken from cleared and stained specimens (c&s) prepared according to the method of Taylor & Van Dyke (1985).Vertebral counts include the four vertebra integrated in the Weberian apparatus with the terminal centrum counted as one vertebra.Teeth and bone SEM (scanning electronic microscope) photos were taken from dissected cleared and stained specimens.
All measurements other than SL are expressed as a percentage of SL except subunits of the head that are recorded as a percentage of head length (HL).Statistical tests of differences between the sexes were performed, but no differences were found.

Hyphessobrycon hamatus, new species
Figs.  Premaxilla with two tooth rows; outer row with 3-4, rarely 5, tricuspid teeth with central cusp larger; inner row teeth 4-5, gradually decreasing in length from first to fourth teeth, last tooth considerably smaller with 3-5 cusps and central cusp longer and broader than other cusps.Two, rarely 3 maxillary teeth, with 3-5 cusps, central cusp slightly longer.Three or 4 anterior most dentary teeth larger, with 4-5 cusps, followed by medium sized tooth with 3-5 cusps, and 5-6 smaller teeth with 1-3 cusps or conical in shape; central cusp in all teeth two to three times longer and broader than remaining cusps.Cusp tips slightly curved posteriorly and towards inside of mouth.
Dorsal-fin rays ii, 9 (three specimens with ii, 8; n = 28); first unbranched ray approximately one-half length of second ray.Males with bony hooks in distal one-third of first to fifth branched rays.Dorsal-fin origin located posterior to middle of SL and posterior to vertical through pelvic-fin origin.Adipose-fin located approximately at vertical through insertion of last anal-fin ray.
Color in alcohol.Dorsal and dorsolateral portions of head and body dark brown.Infraorbital and opercular areas covered with scattered, dark chromatophores.Scales on lateral and ventral surface of body bordered with dark brown chromatophores.Anterior humeral spot very discrete, vertically-elongate, and narrowing ventrally, located over second to third lateral-line scales and extending over 5-6 horizontal series of scales including lateral line.Second humeral spot diffuse, located nearly above 7th-8th lateral line scale and at anterior end of midlateral stripe, and extending usually over 2-3 horizontal series of scales above lateral line.Faint midlateral body stripe extending from second humeral spot to tip of median caudal-fin rays, becoming gradually wider and more intensely pigmented towards caudal peduncle.Midlateral body stripe slightly expanded dorsally and ventrally along caudal peduncle and caudal-fin base and forming small spot (Fig. 1).
Sexual dimorphism.Males of H. hamatus are easily recog-

O F S
nized by the presence of bony hooks on the dorsal-, pectoral-, anal-and pelvic-fin rays.
Distribution.Hyphessobrycon hamatus is known only from two tributaries of the rio Vermelho, in the headwaters of the rio Tocantins basin, Mambaí, Goiás, Brazil.
Etymology.The specific epithet hamatus is from the Latin, meaning hooked, referring to the presence in males of hooks on all fins, except the caudal fin.

Discussion
The new species is described as a Hyphessobrycon in keeping with the current definition of the genus, proposed by Durbin in Eigenmann (1908) and further elaborated by Eigenmann (1918), and which is still in use (Reis et al., 2003): premaxillary teeth in two rows; maxillary teeth absent or reduced in number and restricted to the upper portion of its free margin; lateral line incomplete; adipose fin present, and caudal fin naked (not covered with scales).
As discussed by Weitzman & Malabarba (1998) and Lucena (2003:93-94), there is no evidence that Hyphessobrycon is monophyletic and recognition of monophyletic groups among Hyphessobrycon species is complicated by the difficulty in finding characters useful for hypothesis of relationships among the species.Weitzman & Palmer (1997) essentially attempted to define one group of Hyphessobrycon that they called the rosy tetras, but their paper well demonstrates the fuzzy boundaries of such a group.The question remains unanswered regarding what parts of the rosy tetras are historically real from a phylogenetic point of view and what parts are simply convergent.Weitzman & Malabarba (1998) pointed out that large characid genera such as Hyphessobrycon in some of their sections often have minimal morphological divergence useful for cladistic analyses of their species.Much of the information used to distinguish species are probably labile features such as small variations in color patterns, scales, finray, tooth, and vertebral counts, that are, at least in part, uninformative about phylogenetic relationships.On the other hand the three new species of Hyphessobrycon described in a single paper by Lima & Moreira (2003) represent some of the great external morphological diversity, found in the genus.However, that divergence does not render characters useful in the establishment of phylogenetic relationships, rather it suggests the phylogenetic unreliability of the traditional characters used to define Hyphesobrycon.
As discussed by Weitzman & Malabarba (1998) and Lucena (2003:93-94), there is no evidence that Hyphessobrycon in monophyletic, and recognition of monophyletic groups among Hyphessobrycon species is complicated by the difficulty in finding characters useful for hypothesis of relationships among the species.Weitzman & Malabarba (1998) pointed out that large characid genera such as Hyphessobrycon often have minimal morphological divergence useful for cladistic analyses of their species.Much of the information used to distinguish species are probably labile features such as small variations in color patterns, scales, fin-ray, tooth, and vertebral counts, that are, at least in part, uninformative about phylogenetic relationships.Curiously and apparently contrasting Weitzman & Malabarba (1998), three new species of Hyphessobrycon described in a single paper by Lima & Moreira (2003) present a great external morphological diversity, but that has not rendered characters useful in the establishment of phylogenetic relationships.
Mature males of H. hamatus possess bony hooks in rays in all fins, except in the caudal fin.The presence of hooks on the anal-and pelvic-fin rays and sometime caudal-fin rays of males is often found in several genera and subfamilies of the Characidae (Azpelicueta & Garcia, 2000;Malabarba & Weitzman, 2003), and usually represent a secondary sexual character.The presence of bony hooks on the dorsal-fin rays is known to occur only in two Hyphessobrycon species: H.

O F S
socolofi and H. erythrostigma (Weitzman, 1977).Hyphessobrycon hamatus can be easily distinguished from these species by a lower number of anal-fin rays (iii-v, 16-18, = 17, n = 28) versus (iv, 28-30 and iv, 27-29, respectively), and by the absence of a black spot on the dorsal fin.
Seven Hyphessobrycon species were previously described from the rio Araguaia and rio Tocantins drainages: H. haraldschultzi Travassos, H. stegemanni Géry, H. amandae Géry & Uj, H. moniliger Moreira, Lima & Costa, H. eilyos, H. langeanii, and H. weitzmanorum Lima & Moreira.Remarks on Hyphessobrycon balbus.Myers (1927) noted the presence of some specimens with a complete lateral line in his description of H. balbus.We examined a series of paratypes of the species and all specimens have incomplete lateral line.Marilyn Weitzman found two of the 9 paratypes from CAS 118069 with a complete lateral line, but noticed that they are apparently a species of Astyanax and not H. balbus (S.H. Weitzman, pers. comun.).
The type locality of H. balbus (lagoa Fervedeira, Planaltina, Goiás, Brazil) has presented problems.Looking at the sequence of the localities of Ternetz collections recorded by N. E. Pearson as listed in an unpublished list of Ternetz localities for the years 1923-1925, it appears that lagoa Fervedeira lies near Planaltina.A Millionth Map (S.D-23 for Carinhanha), published in 1933, lists Planaltina as Alta Mira and shows a little nearby lake at the headwaters of the rio São Bartolomeu which ultimately flows into the rio Paraná.This little lake (now appearing on more recent maps as a swamp close to Planaltina) was then called lagoa Fervedeira (S. H. Weitzman, pers. comm.).
posteroventrally slanted at dorsal-fin base; straight from last dorsal-fin ray to adipose-fin origin.Ventral profile of head convex.Ventral body profile convex from pectoral-fin origin to anal-fin origin.Body profile along anal-fin base straight and posterodorsally slanted.Caudal peduncle elongate, nearly straight to slightly concave along both dorsal and ventral margins.Snout convex from margin of upper lip to vertical through anterior nostrils, straight from that point to tip of supraoccipital spine.Head small.Mouth terminal.Maxilla extending anteriorly to under middle of orbit, slightly curved, and aligned at approximately 45 degrees angle relative to longitudinal axis of body.Anterodorsal border of maxilla concave, posterodorsal border slightly convex, anteroventral border concave, and posteroventral border convex.Maxilla slightly widened posteriorly.
apparent in cleared and stained specimens.Anal-fin profile slightly concave in females and males.Longest branched rays of depressed anal fin reaching origin of last anal-fin ray.Anal-fin origin located posterior to vertical through base of last dorsal-fin ray.Anal-fin rays of males bearing one pair of small, narrow, bony hooks along posterolateral border of each segment of lepidotrichia, usually along last unbranched two rays and ten anterior branched rays; hooks more numerous along second through fifth branched rays, rarely present on eleventh ray.Hooks usually located along posterior most branch and distal 1/2 to 2/ 3 of each ray.Pectoral-fin rays i, 9-12, rarely 10 or 13 ( = 10.8, n = 28).Males with bony hooks on distal portion of unbranched and first to sixth branched rays.Pelvic-fin rays i, 6-7 (usually i, 7; n = 28).Pelvic-fin origin located anterior to vertical through dorsal-fin origin.Pelvic fin of males usually bearing 1 bony hook per segment of lepidotrichia along ventromedial border of first to fifth branched rays.
notes.All specimens of H. hamatus were caught in lentic stretches of shallow streams with sand and scattered stones on the bottom, and moderate amount of riparian vegetation.Species collected with H. hamatus were Ancistrus sp., Astyanax sp., Cetopsorhamdia sp., Characidium sp., Knodus sp., and Trichomycterus sp.