Accepted by M. R. de Carvalho: 23 Jul. 2012; published: 5 Sept. 2012
ISSN 1175-5326 (print edition)
ISSN 1175-5334 (online edition)
Copyright © 2012 · Magnolia Press
Zootaxa 3453: 69–83 (2012)
Origin of species diversity in the catfish genus Hypostomus (Siluriformes: Lori-
cariidae) inhabiting the Paraná river basin, with the description of a new species
YA MI L A P. C A RD O S O1*, ADRIANA ALMIRÓN2, JORGE CASCIOTTA2, DANILO AICHINO3, MARTA S.
LIZARRALDE1 & JUAN I. MONTOYA-BURGOS4.
1 Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Centro Regional de Estudios Genónicos, UNLP, Av.Cal-
chaquí 23,5km, C. C. 1888, Fcio. Varela, Buenos Aires, Argentina.
2 División Zoología Vertebrados, Museo de La Plata, UNLP, Paseo del Bosque, C. C. 1900, La Plata, Argentina.
3 Facultad de Ciencias Exactas Químicas y Naturales, UNAM, Felix de Azara, C.C. 1552, Misiones, Argentina.
4 Department of Genetics and Evolution, University of Geneva, 30 quai Ernest Ansermet, 1211, Geneva 4, Switzerland.
Within the Loricariidae, the genus Hypostomus is one of the most diversified freshwater catfish groups. Using new se-
quence data from the mitochondrial Control Region (D-loop) we examined the phylogeny of this genus. Our phylogenetic
analyses suggest that, in the Paraná river basin, species diversity in the genus Hypostomus has been shaped by two proc-
esses: 1) by inter-basin diversification, generating groups of species that inhabit different basins, as a result of dispersal
events; and 2) via intra-basin speciation as a result of basin fragmentation due to past marine transgressions, which pro-
duced groups of species within a basin. Using the D-loop as a molecular clock, each event of diversification was dated
and linked with documented hydrological events or sea level changes. We also assessed the possible dispersal routes be-
tween the Paraná and Uruguay rivers, in addition to the obvious dispersal route via the Río de la Plata estuary. Finally, we
describe a new species of Hypostomus inhabiting Middle Paraná river, Hypostomus arecuta n. sp. This species can be sep-
arated from all other Hypostomus by having light roundish dots on a darker background and by number of premaxillary/
Key words: Armored catfish; Control Region; phylogeny; Paraná river.
In South America, the Loricariidae is the most species-rich endemic family of freshwater fishes. This family of
suckermouth-armored catfishes comprises 818 species (Eschmeyer and Fricke, 2011) and new species are
frequently discovered and described (e.g. Hollanda Carvalho et al., 2010; Zawadzki et al., 2010; Rodriguez et al.,
2011; Cardoso et al., in preparation). Within the Loricariidae, the genus Hypostomus constitutes a rich assemblage
of species, with approximately 130 recognized species (Weber, 2003; Ferraris 2007; Zawadzki et al., 2010,
Hollanda Carvalho et al., 2010). Representatives of Hypostomus are bottom-dwelling fishes widely distributed
throughout tropical and subtropical South America, occurring in a variety of freshwater ecosystems such as
mountain streams and large lowland rivers and ponds. Species delineation and diagnosis in Hypostomus is difficult,
in particular due to the diversity and widespread distribution of the genus, to elevated intra-specific morphological
variability, and because some older descriptions are too short or incomplete.
Numerous species of Hypostomus inhabit the La Plata basin, which comprises the Paraguay, Paraná, and
Uruguay rivers and the Río de la Plata (López and Miquelarena, 1991). Understanding the diversification history of
Hypostomus as a "model" genus might allow the development of a comprehensive view of the processes that
shaped the rich ichthyological diversity in the Paraná river basin.
According to the reconstruction of paleo basins in South America, from about 60 to10 million years ago (Ma),
the paleo Amazon–Orinoco system was a large watershed with waters flowing northward toward the Caribbean
Sea, while the La Plata basin was already oriented as present (Lundberg, 1998). This author suggested that at 12–10
CARDOSO ET AL.
70 · Zootaxa 3453 © 2012 Magnolia Press
Ma, the boundary between the paleo Amazon–Orinoco and the La Plata basins underwent a final and important
shift southward to its current location. This boundary displacement must have occasioned exchanges of water and
fishes between the two main basins 12 to 10 Ma, as proposed by Montoya-Burgos (2003). However, the boundary
displacement might have been more progressive, covering the last 10 Ma (Rasanen et al., 1995; Lundberg et al.,
1998), opening occasional dispersal routes between these two river systems.
During the Miocene (24-5 Ma), two main components of the La Plata basin, the upper plus middle Paraná river
on one hand, and Uruguay river on the other, which were forming a single large river flowing southward into the
Rio de la Plata estuary, disconnected from one another resulting in the modern configuration (Beurlen, 1970).
According to Bonetto (1994), geological changes caused this disconnection by modifying the course of the middle
Paraná river that subsequently reached the course of the Paraguay river. Today, the Paraná and Uruguay rivers are
connected exclusively via the Río de la Plata estuary.
Furthermore, in the second half of the Miocene, 15-5 Ma, marine transgressions occurred at least twice along
distinct paleogeographic corridors. The first maximum flooding event occurred between 15 and 13 Ma and the
second between 10 and 5 Ma. The two marine transgressions covered most of the Paraná river basin (Hernández et
al., 2005). It is likely that the diversity of strictly freshwater organisms might have been seriously impacted by
these marine transgressions. For example, the museum hypothesis of diversification (Nores, 1999) states that the
Miocene marine incursions have been major diversifying events via the fragmentation of emerged land resulting in
The goals of the present work are: (1) to expand the phylogeny of the genus Hypostomus that was previously
proposed by Montoya-Burgos (2003) using new sequence data from the mitochondrial Control Region, (2) to infer
the origin of the diversity of Hypostomus species in the Paraná river basin, (3) to assess the possible dispersal
routes between the Paraná and Uruguay river, in addition to the obvious dispersal route via the Río de la Plata
estuary, and (4) to describe a new species of Hypostomus inhabiting the Middle Paraná river basin.
The use of genetic markers is a powerful tool to estimate the extent of hidden biodiversity. For example, the
mitochondrial D-loop region is frequently used for answering a broad range of biological questions relative to
population processes, phylogeography (e.g. Cardoso and Montoya-Burgos, 2009) and species identification (e.g.
Cardoso et al., 2011). Here we used this molecular tool in order to infer the phylogenetic relationships among
Hypostomus species and to analyse the origin of species diversity in the río Paraná basin.
Materials and methods
Taxon sampling and morphological analyses
Fish specimens were collected in 15 different localities in the Paraná river basin, Argentina (Fig. 1). Most of them
were sampled in the middle and lower section of the Paraná river basin. We also used available data from the upper
section of this basin taken from GenBank. Fishes were caught using gill nets, cast nets, hand nets, and seine. Tissue
samples for genetic studies were preserved in ethanol 96 % and frozen at -20 °C, the vouchers specimens were
fixed in formalin 10 % for morphological studies and deposited at MHNG, IPLA, and MACN according to the
institutional abbreviations are as listed in Ferraris (2007). Table I has more information about the specimens
All measurements were taken point to point with digital calipers to the nearest 0.01 mm, under a dissecting
microscope when necessary. Measurements and counts of bilaterally symmetrical features were taken from the left
side of the body whenever possible; if a feature was missing or broken on the left side, it was examined on the right
side. Counts and measurements follow Boeseman (1968), Weber (1985), and Reis et al. (1990). Body plate counts
and nomenclature follow Oyakawa et al. (2005). The oral disk width was measured at point of insertion of the
DNA amplification and sequencing
The genomic DNA was extracted using the salt-extraction protocol (Aljanabi and Martinez, 1997). The PCR
amplification of the Control Region (D-loop) of the mitochondrial DNA was performed using the following primers:
DLA-III 5’-TATT TAAAGRCATAATC TCTTGAC-3’ and HygDL-R 5’–WTGCKARTATGTGCCGYYTG–3’. The
amplifications were performed in a total volume of 50 l, containing 5 l of 10x reaction buffer, 1 l of
deoxyribonucleoside triphosphate (dNTP) mix at 10 mM each, 1 l of each primer at 10 M, 0.2 l of Taq DNA
Zootaxa 3453 © 2012 Magnolia Press · 71
ORIGIN OF SPECIES DIVERSITY IN HYPOSTOMUS.
Polymerase equivalent to 1 unit of Polymerase per tube, and 1 l of DNA. Cycles of amplification were programmed
as follows: (1) 3 min. at 94°C (initial denaturing), (2) 30 sec. at 94°C, (3) 30 sec. at 55–57°C, (4) 1 min. at 74°C, and
(5) 5 min. at 74°C (final elongation). Steps 2 to 4 were repeated 42 times. The PCR products were purified and
sequenced by the company MAGROGEN (Korea). Sequences were deposited in GenBank (Table I).
TAB LE I. Details of the specimens used in the molecular phylogeny with GenBank accession numbers.
Species GenBank Field number Locality
H. arecuta JF290442 AG09-163 Yahapé (27º22'12.1"S-57º39'14.6"W)
H. arecuta JF290441 AG09-181 Ituzaingó (27º 29'32''S-56º 39'38''W)
H. arecuta JF290445 AG09-198 Ituzaingó (27º 29'32''S-56º 39'38''W)
H. arecuta JF290446 PR-031 Santa Fé, Santa Fé, Argentina
H. arecuta JF290443 TAE01 Ituzaingó (27º 29'32''S-56º 39'38''W)
H. arecuta JF290444 TAE02 Ituzaingó (27º 29'32''S-56º 39'38''W)
H. derby JF290447 YC10-316 Uruzú (25°55'38.25''S-53°56,031'W)
H. paranensis JF290449 YC-025 Suquia (31°24'11,9''S-64°12'11,4''W)
H. paranensis JF290448 YC-026 Suquia (31°24'11,9''S-64°12'11,4''W)
H. commersoni JF290450 AG09-129 Tabay (26º59'56.3"S-55º10'44.9"W)
H. commersoni JF290451 YC09-124 Manucho (31°15'S- 60°53' W)
H. commersoni JF290452 YC-957 El Bosque (34°54'37.55''S-57°56'15.65''W)
H. commersoni JF290453 YC-607 Corrientes (29°48,574’S-59°23.600’W)
H. commersoni JF290454 14802 P. N. Pre-Delta (32º08'08.8"S- 60º37'26.2"W).
H. commersoni JF290455 AG09-013 Ituzaingó (25º29'54.5"S- 56º42'47.0"W)
H. commersoni JF290456 Reg02 Ensenada (34°50'23,96''S-57°55'13,04''W)
H. commersoni JF290457 AG09-077 Garupa (27º29'10.2"S-55º44'23.1"W)
H. commersoni JF290458 YC-926 El Pescado (34°57,790'S- 57°46,696'W)
H. cochliodon JF290476 AG09-016 Ituzaingó (27º 29'32''S-56º 39'38''W)
H.luteomaculatus JF290471 YC-162 Antequera (27°27'43.43" S-58°52'0.03" W)
H. luteomaculatus JF290469 UR004 Pedra Fortaleza, Itapiranga, Brazil
H. luteomaculatus JF290470 UR002 Pedra Fortaleza, Itapiranga, Brazil
H. luteomaculatus JF290468 AG09-157 Ituzaingó (27º 29'32''S-56º 39'38''W)
H. luteomaculatus JF290467 AG09-200 Ituzaingó (27º 29'32''S-56º 39'38''W)
H. luteomaculatus JF290459 AG09-012 Ituzaingó (27º 29'32''S-56º 39'38''W)
H. luteomaculatus JF290466 CIA283 Candelaria (27º26'92"S-55º44'50"W)
H. microstomus JF290461 AG09-015 Ituzaingó (27º 29'32''S-56º 39'38''W)
H. myersi JF290472 YC10-256 Deseado (25°47'1.30'' S-54°2'21.40'' W)
H.myersi JF290474 AG09-123 Tabay (26º59'56.3"S-55º10'44.9"W)
H. myersi JF290473 AG09-124 Tabay (26º59'56.3"S-55º10'44.9"W)
H. myersi JF290475 AG09-131 Tabay (26º59'56.3"S-55º10'44.9"W)
H. regani JF290460 Reg.06 Rio Mogi Guaçu, Brazil
H. ternetzi JF290462 YC-164 Antequera (27°27'43.43" S-58°52'0.03" W)
H. ternetzi JF290463 AG09-160 Yahapé (27º22'12.1"S-57º39'14.6"W)
H. uruguayensis JF290464 AG09-159 Yahapé (27º22'12.1"S-57º39'14.6"W)
H. uruguayensis JF290465 14731 P. N. Pre-Delta (32º07'18.0"S- 60º40'12.0"W)
CARDOSO ET AL.
72 · Zootaxa 3453 © 2012 Magnolia Press
Sequence alignment, phylogenetic reconstruction, and molecular clock calibration
The mitochondrial D-loop sequences were obtained for 36 individuals from Argentina (for more details see Fig.1
and Table I). We also used sequences of different species of Hypostomus deposited in GenBank and nine others
species of the family Loricariidae as outgroups, as in Montoya-Burgos (2003). The editing of the new sequences
and the alignment were performed using BioEdit 7.0.1 (Hall, 1999). Prior to phylogenetic reconstruction,
appropriate substitution models were estimated with the Akaike information criterion (AIC) as implemented in
MrAIC (Nylander, 2004). We us ed a total o f 7 4 sequences o f Hypostomus to reconstruct the phylogeny. Two
phylogenetic reconstruction methods were used. First, maximum likelihood (ML) phylogenetic reconstruction was
performed using TreeFinder (Jobb et al., 2004). Confidence values for the edges of the ML tree were computed by
bootstrapping (Felsenstein, 1985), with 1000 replications. Second, Bayesian Inference analysis (BI) was conducted
using MrBayes 3.1.2 (Huelsenbeck and Ronquist, 2001; Ronquist and Huelsenbeck, 2003). Four chains were run
simultaneously (three heated, one cold) for 20,000,000 generations, with tree space sampled every 100th
generation. After a graphical analysis of the evolution of the likelihood scores, the first 300, 000 generations were
discarded as burn-in. The remaining trees were used to calculate the final consensus tree.
FIGURE 1. Map showing the sampling localities. The abbreviations mean: DE (arroyo Deseado, Iguazú river, Misiones); UR
(arroyo Uruzú, Paraná river, Misiones); GA (arroyo Garupá, Paraná river, Misiones); TA (arroyo Tabay, Paraná river,
Misiones); CA (Candelaria, Paraná river, Misiones); IT (Ituzaingó, Paraná river, Corrientes); YA (Yahapé, Paraná river,
Corrientes); CO (Corrientes river, Corrientes); AN (Antequera, Paraná river, Chaco); SU (Suquia river, Cordoba); MA
(Manucho river, Paraná river, Santa Fé); SF (Paraná river, Santa Fé); PN (Parque Nacional Pre-Delta, Paraná river, Entre Ríos);
EN (Ensenada, Río de La Plata, Buenos Aires); EP (arroyo El Pescado, Río de la Plata, Buenos Aires) and BO (lago del Paseo
del Bosque, La Plata).
Additionally, we performed molecular clock tests with HyPhy (Kosakovsky Pond et al., 2005) using the HKY85
model. The null hypothesis of constant molecular clock was tested for the ingroup taxa using the log likelihood ratio
test (Huelsenbeck and Crandall, 1997). Prior to these analyses, the data set was pruned to include only one
representative of each species. In addition, the sequence corresponding to H. fonchii and H. sp. Tib1 – used in
Montoya-Burgos (2003) – were discarded because it showed a particularly long branch in the phylogenetic tree.
Zootaxa 3453 © 2012 Magnolia Press · 73
ORIGIN OF SPECIES DIVERSITY IN HYPOSTOMUS.
TAB LE II. Morphometric data and counts of holotype and 23 paratypes of Hypostomus arecuta n. sp.
In order to evaluate the temporal diversification of species in the genus Hypostomus in the Paraná river basin,
the rate of evolution of the D-loop region needed to be calibrated. To do so, we used the same calibration point as
in Montoya-Burgos (2003), which is based on the following reasoning: the phylogeny of Hypostomus shows that
H. hondae, distributed only in the Lago Maracaibo and Magdalena basins, is the closest relative to H.
plecostomoides, which is known only from the Orinoco basin and some localities of upper Amazon. Because these
distribution patterns match the vicariant episode that separated Lago Maracaibo system from Amazon and Orinoco
basins 8 Ma (Hoorn, 1993), it is reasonable to attribute this age to the speciation event that gave rise to H. hondae
and H. plecostomoides. This geological event has also been used for calibrating other Neotropical fish phylogenies
(e.g. Lovejoy et al., 2000, Sivasundar et al., 2001).
Holotype range Mean / SD
Standard length (mm) 185.5 127.0–268.1
Percents of SL
Predorsal length 38.6 37.4–44.0 39.1 ± 1.57
Head length 32.1 29.5–34.4 31.3 ± 1.32
Cleithral width 32.3 29.2–32.7 30.9 ± 0.89
Head depth 19.5 17.2–19.9 18.9 ± 0.70
Dorsal-fin spine length 33.1 26.8–34.5 31.6 ± 2.28
Dorsal-fin base length 27.2 25.1–30.0 27.2 ± 1.20
Dorsal-adipose distance 16.3 15.1–16.9 16.1 ± 0.48
Thoracic length 24.0 19.8–26.2 23.3 ± 1.56
Pectoral-fin spine length 33.4 29.7–35.6 32.0 ± 1.23
Abdominal length 25.6 22.1–25.6 23.9 ± 0.84
Pelvic-fin spine length 24.9 22.3–25.8 24.1 ± 0.95
Caudal-peduncle length 27.4 27.4–33.3 30.6 ± 1.52
Caudal-peduncle depth 11.7 10.4–12.0 11.3 ± 0.45
Adipose-fin spine length 10.3 8.0–10.9 9.6 ± 0.86
Anal width 12.6 10.0–13.5 11.7 ± 0.79
Upper caudal-fin ray length 33.0 27.4–35.3 30.7 ± 2.06
Lower caudal-fin ray length 36.2 28.1–36.3 31.9 ± 2.44
Percents of head length
Head depth 60.8 57.7–63.6 60.7 ± 1.91
Snout length 61.3 61.2–66.8 63.0 ± 1.75
Orbital diameter 18.9 16.0–19.0 18.1 ± 0.94
Interorbital with 38.4 33.6–38.9 37.3 ± 1.47
Maxillary barbel length 12.2 9.3–15.2 12.1 ± 1.66
Mandibulary ramus length 24.5 21.7–25.4 23.3 ± 1.23
Median plates series 27/27 26/28 27
Predorsal plates 3 3–3 3
Dorsal plates below dorsal-fin base 9 9–10 9
Plates between dorsal and adipose fin 6 5–6 6
Plates between adipose and caudal fin 4 4–5 5
Plates between anal and caudal fin 14 12–14 14
Premaxillary teeth 74/79 66–85 77
Dentary teeth 71/72 63–84 80
CARDOSO ET AL.
74 · Zootaxa 3453 © 2012 Magnolia Press
FIGURE 2. Maximum likelihood Hypostomus phylogenetic tree based on D-loop haplotypes (-lnL = 5192.34998). The ML tree
was derived using the HKY + G model of sequence evolution. Numbers next to branches are Bayesian posterior probabilities
followed by bootstrap values when these are above 50%, respectively. These support values are showed only for the relevant
relationships of this work. Bold letters are abbreviations used for naming clades (see text). The specimens from Paraná river
basin and Amazon system (Amazon basin, French Guyana and Northeastern South America coastal rivers) are indicated. The
three species marked with * inhabit the Paraná and Uruguay rivers, but not on the Río de La Plata. Also, we show the estimated
ages of dispersal events between basins (black arrow) and for vicariance events inside the La Plata basin (white arrow).
Zootaxa 3453 © 2012 Magnolia Press · 75
ORIGIN OF SPECIES DIVERSITY IN HYPOSTOMUS.
A molecular phylogenetic approach was used to investigate the diversity of Hypostomus species from the Paraná
river basin. The sequence alignment comprised 592 positions, from which 179 were variable within the ingroup.
Base composition and structural characteristics of Hypostomus D-loop sequences are described elsewhere
(Montoya-Burgos et al., 2002). The model of sequence evolution that fit the best our sequence data set is HKY +
gamma, according to MrAIC (Nylander, 2004). The ML and Bayesian phylogenetic trees obtained have similar
topologies. The ML tree is shown in Fig. 2.
The evolutionary relationships of the outgroup species is the same as found in Montoya-Burgos (2003). All
Hypostomus species form a monophyletic clade named Clade D (Fig. 2). This clade can be organized into four
monophyletic groups, D1, D2, D3, and D4. Clade D1 clusters together with D2 and Clade D3 with D4. The
relationships among these clades are supported by high Bayesian posterior probabilities but relatively low
bootstrap values. Hypostomus gymnorhynchus is placed as the sister species to clades D1 and D2 and thus forms a
distinct monospecific lineage. The Clade D1 (Fig. 2) forms the H. cochliodon group including H. cochliodon from
the Paraná river basin, H. plecostomoides from Orinoco basin, H. hondae from Lago Maracaibo and Magdalena
basins, H. fonchii and Hypostomus sp. 1013 from the Amazon basin (see Montoya-Burgos (2003) for the non
described species metioned in this work).
Clade D2 is subdivided into two monophyletic groups: the first contains species from French Guyana (H.
watwata and H. plecostomus), Amazon basin (represented with Hypostomus spp.: 36, 49), and Northeastern South
America coastal rivers (Hypostomus spp.: 177, 219, and 270 from Gurupí, Itapicurú, and Parnaíba rivers,
respectively). The second clade includes species from Eastern South America coastal rivers (H. puntactus from
Ubatiba), and the La Plata basin (H. commersoni, H. derbyi, H. paranensis, H. boulengeri, and two Hypostomus
spp.: Tib13 and. 1211).
Clade D3 is also subdivided into two groups, one clade including species from the Amazon basin (H. asperatus
and three Hypostomus spp.: 906, 1100, and 1026). The other clade contains species from La Plata basin and São
Francisco river (H. regani; H. luteomaculatus; H. microstomus; H. myersi, H. nigromaculatus, and three
Hypostomus spp.: 678, 699, and 751).
Finally, clade D4 includes species inhabiting the La Plata basin: Hypostomus arecuta n. sp. (described below),
H. ternetzi, H. uruguayensis, H. aspilogaster, H. luteus, H. isbrueckeri, H. latifrons, H. latirostris, and H.
albopunctactus. This clade comprises also H. luetkeni from Eastern South America coastal rivers (Paraíba river)
and H. johnii from an northeastern South America coastal river (Parnaíba river).
Our results show that at least one species inhabiting Paraná river basin is present in each of the four main
Hypostomus clades (i.e. H. cochliodon in clade D1; H. commersoni, H. derbyi, H. paranensis, and Hypostomus
spp: Tib13 and 1211 in clade D2; H. luteomaculatus; H. microstomus, H. myersi, H. regani, H. nigromaculatus, and
H. sp 699 in clade D3; finally, the new species Hypostomus arecuta n. sp., H. ternetzi, H. uruguayensis, and H.
albopunctatus in clade D4).
Phylogenetic tree calibration.
When analysing the ingroup taxa, with the exclusion of H. fonchi which has a particularly long terminal branch
(clade D in Fig. 2), the log-likelihood ratio test of homogeneous evolutionary rate showed no significant
differences between the likelihood scores obtained when enforcing or not the molecular clock (X2 = 56.03; d.f. =
43; P = 0.087). This result indicates that the sequences of the ingroup representatives are evolving at a
homogeneous rate. With the aim of evaluate the temporal diversification of the genus Hypostomus in the Paraná
river basin, we calibrated the D-loop region. We found that the splitting between the Hypostomus from the Amazon
system (comprising the Amazon basin, French Guyana and North-eastern South America coastal rivers) and La
Plata basin is estimated to 6.5 Ma in clade D1, 11.3 Ma in clade D2, 11.8 Ma in clade D3 and 9.3 Ma in clade D4.
Moreover, the origin of temporal diversification among the lineages inhabiting the La Plata basin is dated to 7.45
Ma in clade D2, 9.35 Ma in clade D3 and 13.5 Ma in clade D4.
CARDOSO ET AL.
76 · Zootaxa 3453 © 2012 Magnolia Press
FIGURE 3. Hypostomus arecuta n. sp., Holotype. MACN-ict 9677 (198), 185.5 mm SL. Dorsal, lateral, and ventral views.
Photos by Yamila P. Cardoso.
Zootaxa 3453 © 2012 Magnolia Press · 77
ORIGIN OF SPECIES DIVERSITY IN HYPOSTOMUS.
Hypostomus arecuta n. sp.
Here, we describe a new species that inhabit the Paraná river basin and which contributes to the understanding of
the origin of the species diversity in this basin (see discussion).
Holotype: MACN-Ict 9677 (198), 185.5 mm SL, Argentina, Corrientes province, Ituzaingó, Paraná River
(27°29’54.5''S - 56°42’47.0''W). Col: Gonzalez et al., November, 2009.
Paratypes: MACN-ict 9678 (163, 166), 2 ex., 174.0–243.3 mm SL, Argentina, Corrientes province, Yahapé,
Paraná river, (27°22’12.1''S-57°39’14.6''W). Col: Gonzalez et al., November, 2009. MACN-ict 9679 (181, 191,
192, 193, 195, 196, 197, 199), 9 ex., 156.5–268.1 mm SL, same data as holotype. MACN-ict 9680 (CIA 284–285),
2 ex., 127.0–134.7 mm SL, Argentina, Misiones province, Candelaria city, Paraná river (27º26’92’’S-
55º44’50’’W). Col: Aichino and Capli , November, 2009.
Hypostomus arecuta n. sp. is distinguished from its congeners by the following combination of characters:
dorsum of head and body and all fins dark grey covered by numerous rounded cream dots. Ventral surface of head
and belly a plain cream color. This color pattern distinguishes H. arecuta n. sp. from Hypostomus species that have
dark roundish dots on a lighter background (such: H. ancistroides, H. brevis, H. commersoni, H. fluviatilis, H.
hermanni, H. iheringii, H. nigromaculatus, H. paulinus, H. topavae, H. isbruekeri, H. aspilogaster, H.
uruguayensis, H. latifrons, and H. latirostris). Among the species that have light roundish dots or irregular light
marks on a darker background, H. arecuta n. sp. is distinguished by number of premaxillary/dentary teeth (66–85/
63–84) as compared to H. albopuctatus (26–32/22–26), H. luteus (22–38/26–40), H. regani (63–107/63–104), H.
luetkeni (30–69/38–68), H. strigaticeps (about 60), H. multidens (115–260/122–267) and H. microstomus (7–11/7–
13). Hypostomus arecuta n. sp. is distinguished from its sister species H. ternetzi by the colour pattern of dorsum of
head and body, and all fins dark grey covered by numerous rounded cream dots vs. dorsum homogeneously dark,
and greater number of scutes at dorsal-fin base (9–10 vs. 8). Also, some morphometric characters differentiate H.
arecuta n. sp. from H. ternetzi: cleithral width (3.0–3.4 vs. 2.8–2.9 in SL), abdominal length (3.9–4.5 vs. 4.6–5.4 in
SL), eye diameter (5.2–6.2 vs. 6.2–6.9 in HL), pelvic fin-spine length (3.9–4.5 vs. 3.1–3.8 in SL), caudal penduncle
depth (8.3–9.6 vs. 7.5–8.1 in SL), upper caudal-ray length (2.8–3.6 vs. 2.0–2.3 in SL), lower caudal-ray length
(2.4–3.5 vs. 1.8–2.0 in SL), and right mandibular ramus (3.9–4.6 vs. 4.8–5.6 in HL). Besides, H. arecuta n. sp.
differs from H. luteus by short dorsal spine length (mean 31.6 % vs. 34.4 % of SL), the length of right mandibular
ramus (3.9–4.6 vs. 4.8–6.1 in HL), abdominal length (3.9–4.5 vs. 4.4–5.0 in SL), head depth (1.6–1.7 vs. 1.7–1.9 in
HL), and interorbital width (2.6–3.0 vs. 2.9–3.6 in HL). Also some ratios distinguish H. arecuta n. sp. from H.
luetkeni: predorsal length (2.3–2.6 vs. 2.5–3.0 in SL), cleithral width (3.0–3.4 vs. 3.3–4.0 in SL), pectoral-fin spine
length (2.8–3.3 vs. 3.1–3.8 in SL), and caudal peduncle length (3.0–3.6 vs. 2.8–3.1). Finally, short dorsal spine
length separates H. arecuta n. sp. from H. luteomaculatus (mean 31.6 % vs. 40% of SL).
Hypostomus arecuta can be differentiated from H. boulengeri, H. commersoni and H. cochliodon by the colour
pattern. Also, H. commersoni has strong lateral keels which are absent in H. arecuta, H. cohliodon bears fewer
premaxillary and dentary teeth than H. arecuta n. sp (8/9 vs. 66–85/63–84, respectively). Hypostomus arecuta
shares with H. luetomaculatus and H. microstomus a similar dorsal colour pattern, however H. luteomaculatus and
H. microstomus have dark ventral surface of head and body with light vermiculated dots vs. head and belly plain
cream in H. arecuta. Some counts distinguish H. arecuta from H. luteomaculatus: scutes along lateral line 26–28
(mode 27) vs 28–29 (mode 29), scutes between end of dorsal fin to adipose fin 5–6 (mode 6) vs 6–7 (mode 7);
scutes from adipose to caudal fins 3–5 (mode 5) vs. 5–8 (mode 6), and scutes from anal to caudal fins 12–14 (mode
14) vs. 14–16 (mode 16), respectively. Finally H. arecuta n. sp. has a greater number of teeth than H. microstomus
(66–85/63–84 vs. 7–11/7–13).
Meristic and morphometric data are presented in Table I. Dorsal profile slightly convex from snout tip to
anterior margin of eyes, straight at interorbital area, convex from interorbital area to dorsal-fin origin, and almost
straight from dorsal-fin origin to end of adipose fin. Body width at cleithral region larger than head depth. Head
broad and shallow dorsally covered with plates, except for a quadrangular naked area on snout tip.
CARDOSO ET AL.
78 · Zootaxa 3453 © 2012 Magnolia Press
Supraoccipital bone with a shallow median ridge, and with a relatively well developed posterior process
bordered by a wide nuchal plate. A shallow ridge originating laterally to the nares, passing through supraorbital,
and extending to median portion of pterotic-supracleithrum. Opercle small, with odontodes more developed
Oral disk ovoid, lower lip covered with numerous papillae decreasing in size posteriorly. Maxillary barbels
moderately developed, about as long as orbital diameter. Sixty-six to 85 (holotype 74) teeth in premaxilla, 63 to 84
(holotype 72) in dentary. Teeth bicuspid, curved inward distally, mesial cusp two or three times longer than lateral
cusp, distal margin of mesial cusp rounded in replacement teeth and straight in functional ones. Body covered with
five rows of moderately spinulose scutes. Tip of snout mostly naked even in large specimens, bearing two lateral
vertical patches of odontodes.
Ve nt r a l s u rf a ce o f h e ad n a ke d , w i th s m al l o r l ar g e p a tc h o f p l at e l et s b e fo r e b r an c h ia l o p en i ng . A bd o me n
covered with minute platelets, with exception of small area around pectoral fin and small or large area around
pelvic-fin insertions, and small area at urogenital opening. Preanal plate absent. Caudal peduncle laterally
compressed, rather ovoid in cross section.
Twe nty-one to 23 (mode 22) dors al plate s, 25–2 6 (mod e 24) mid-d ors al pla tes , 24– 25 (mo de 25) median
plates, 26–28 (mode 27) mid-ventral plates, 21–22 (mode 22) ventral plates. Three predorsal plates, 9–10 (mode 9)
plates below dorsal fin, 5–6 (mode 6) preadipose plates, 4–5 (mode 5) plates between adipose fin and caudal fin,
12–14 (mode 14) plates between anal fin and caudal fin.
Dorsal-fin II,7, its origin placed at vertical closer to pelvic-fin origin than pectoral-fin origin. Dorsal-fin
margin straight. Adipose-fin spine compressed and curved backward. Pectoral fin I,6, its posterior border straight.
Pectoral-fin spine slightly curved inward, covered with weakly developed odontodes, slightly more developed on
its distal portion in larger specimens. Tip of pectoral fin reaching one-third pelvic-fin spine length. Pelvic-fin I,5,
its posterior border slightly roundish. Pelvic-fin spine surpassing anal-fin origin. Anal fin I,4, its tip reaching the
6th plate after its origin, 2ed and 3rd branched rays longer. Caudal-fin margin concave, I,14,I, with inferior lobe
longer than superior one.
Phylogenetic position of Hypostomus arecuta
The new species described above, Hypostomus arecuta, is distinguished from others species of the genus by a
combination of morphological and molecular features. Hypostomus arecuta is apparently endemic to the Paraná
river in Argentina. According to our phylogenetic tree, H. arecuta, together with H. ternetzi, H. isbruekeri, H.
aspilogaster, H. uruguayensis, H. latifrons, H. latirostris, H. luteus, H. albopuntactus, H. johnii and H. luetkeni
form the clade D4 (see Fig. 2). Although our results show that the node that clusters clades D4 and D3 shows low
statistical support in the ML analysis, other data support this relationship: following Muller and Weber (1992), the
Hypostomus species of clade D4 shares with species of clade D3 the presence of white spots on the body, wide
mandible, and long-crowned teeth (defining the so called Hypostomus regani group). Species belonging to other
clades (D1 and D2, Fig. 2) display black widespread spots on the body, medium-sized mandible, and short-
crowned teeth (forming the so called Hypostomus plecostomus group). Moreover, karyological studies show that
some species of the Hypostomus regani group have a fundamental chromosome number of near 72 and some
species of the Hypostomus plecostomus group have a fundamental number near 68 (Zawadzki et al., 2004).
Therefore, these morphological, colour pattern and karyological data support our phylogenetic analyses showing a
division of Hypostomus species into two principal clades, clade D1+D2 (Hypostomus plecostomus group) and
clade D3+D4 (Hypostomus regani group). Thus, the new species H. arecuta. is considered as a member of the
Hypostomus regani group.
Colour in alcohol
Overall ground colour of body and fins dark grey. Overall ground color of ventral area a plain, lighter, cream
color in some specimens. Dorsal surface of head, body, and fins entirely covered by numerous rounded cream dots,
smaller and closer on head. Dorsal, pectoral, and pelvic- fins with dots regularly or irregularly arranged in rows
along their rays. Adipose fin with rounded, cream dots. Caudal fin with scattered, rounded cream dots on rays and
Zootaxa 3453 © 2012 Magnolia Press · 79
ORIGIN OF SPECIES DIVERSITY IN HYPOSTOMUS.
Hypostomus arecuta is known from the Paraná river at Yahapé and Ituzaingó (Corrientes province), Candelaria
(Misiones province), and Santa Fé (Santa Fé province), Argentina. Hypostomus arecuta is sympatric with H.
commersoni, H. cochliodon, H. uruguayensis, H. latifrons, H. ternetzi, H. luteomaculatus, H. microstomus, and H.
The specific epithet arecuta is a Guaraní word arecutá that means loricariid fish.
The specimens of Hypostomus arecuta were collected in coastal areas of the Paraná river main channel. The
bottom was composed mostly by large boulders of sandstone with patches of sand and pebbles. The species was
found in well oxygenated waters having moderate current speed, about 0.60 m s-1. Water transparency was within
the most frequent range registered in the river (1.50–2.40 m). Conductivity was generally low and typical for the
river (50.9–59.6 µS cm-1). The pH was slightly acidic to neutral (6.8–7.1).
The origin of species diversity of Hypostomus in the Paraná river basin
According to the results presented here and to Montoya-Burgos (2002, 2003), the phylogenetic tree of the
genus Hypostomus can be divided into four principal clades (Fig. 2). Since each clade includes at least one species
from the Paraná river basin and at least one from the large Amazon system, it can be deduced that old inter-basin
allopatric speciation has participated in the diversification of Hypostomus in the Paraná river basin. In addition,
lineages with multiple species inhabiting the Paraná river basin are found in clades D2, D3 and D4. This indicates
that speciation within the basin also shaped the diversity of Hypostomus there.
The biogeographic analysis of the inter-basin relationships in clade D1 shows that H. cochliodon, from the
Paraná river, clusters with Hypostomus sp. 1013, from the Amazon basin (Fig. 2), and our calibrations indicate that
the splitting event can be dated to 6.5 Ma. To explain this age, we would have to invoke temporal connections
between the upper Paraguay and Southern tributaries of the Amazon posterior to the inferred age of the boundary
displacements and water interchange between the Northern paleo Amazon-Orinoco basin and La Plata basin (11.8–
10 Ma) (Lundberg et al., 1998). When these temporal interconnections ceased, the isolation of populations in both
basins would have resulted in the allopatric speciation that gave rise to H. cochliodon and H. sp. 1013.
The clade D2 shows that species inhabiting the Amazon system cluster with species from La Plata basin +
Eastern South America coastal rivers (Fig. 2). According to the D-loop molecular clock, the splitting event between
these two groups can be dated to 11.3 Ma. This result is in accordance with the estimated date for this clade in
Montoya-Burgos (2003). This inferred age matches with the documented boundary displacements between the
Northern paleo Amazon-Orinoco system and the La Plata basin that occurred at about 11.8–10 Ma (Lundberg et al.,
1998). Once the headwater exchanges due to the boundary displacement were finished, the isolation of populations
in both basins would have resulted in speciation, giving rise to the two lineages of clade D2.
Within clade D3, the species inhabiting the Amazon basin and those from La Plata basin + São Francisco river
(see Fig.2) form two distinct lineages that originated about 11.8 Ma according to the D-loop data. In Montoya-
Burgos (2003), this event had a slightly more recent date (10.2–10.1 Ma.). However, both estimations correspond
well with the date estimated for the last major water interchange between the paleo Amazon-Orinoco and La Plata
basin reported above for clade D2.
In clade D4, the splitting between species inhabiting North-eastern South America coastal rivers (being part of
the Amazon system), represented by H. johnii, and species from the Paraná river + Eastern South America coastal
rivers, represented by H. albopunctatus + H. luetkeni, was dated to 9.3 Ma. Differing from what was proposed in
Montoya-Burgos (2003), this estimated age does not correspond to the boundary displacements between the
Northern paleo Amazon-Orinoco system and La Plata basin (11.8–10 Ma). This difference can be explained by the
exclusion of the sequence of Hypostomus sp. (Tib 1), used in Montoya-Burgos (2003), from our analysis and also
because in this work we used a smaller segment of D-loop marker than in Montoya-Burgos (2003). According to
our results, the diversification episode within clade D4 would be another case of more recent water interchange via
a temporal connection between the two main basins, as was already the case for clade D1. In addition, the
CARDOSO ET AL.
80 · Zootaxa 3453 © 2012 Magnolia Press
evolutionary position of H. johnii within the clade D4 (fig. 2) would allow us to suggest that the direction of the
dispersal event was from the La Plata basin towards the Amazon system.
In summary, our results suggest that at least four independent allopatric speciation episodes occurred between
the Amazon system and the Paraná basin river + São Francisco + Eastern South America coastal rivers. In clade D2
and D3, these allopatric speciations may be explained by the boundary displacement between the Northern paleo
Amazon–Orinoco and the Southern La Plata basin which occurred between 11.8–10 Ma (Lundberg et al., 1998).
As indicated in clades D1 and D4, two more recent allopatric speciations (6.5 Ma for clade D1 and 9.3 Ma for clade
D4) occurred by headwater exchanges and subsequent isolation between the northern and southern river systems,
involving probably the upper Paraguay and Southern tributaries of the Amazon (Lundberg et al., 1998).
Accordingly, more recent population splitting between the Amazon and Paraná river basins has been reported
(between 2.3 and 4.1 Ma) for the migratory fish Prochilodus (Sivasundar et al., 2001). Moreover, other recent
dispersal events between these two basins are further exemplified by the distribution ranges of Pygocentrus
nattereri (Hubert et al., 2007) and Pseudotylosurus augusticeps (Lovejoy and De Araújo, 2000). These data and
our results suggest that the temporary connections between the Amazon and Paraná river basins would be more
frequent than previously thought.
The second origin of the diversity of Hypostomus species in the Paraná river basin is shaped by intra-basin
speciation and occurs within the La Plata basin. In clade D2, the node including all species present in La Plata basin
(seven species) was estimated to 7.45 Ma according to our molecular clock. This date coincides with the second
reported event of maximum flooding of marine transgression during the Miocene (10-5 Ma) (Hernández et al.,
2005). The extensive marine incursion onto the Paraná river basin could have isolated the tributaries of this basin,
generating several allopatric speciations in different and strictly freshwater organisms that habited the La Plata
basin. Once the sea retreated, the newly formed species would have dispersed throughout the current La Plata
basin, enriching its biodiversity. It is important to note that H. puntactus from Eastern South American coastal
rivers emerges within the group inhabiting the La Plata basin. This fact was explained by Montoya-Burgos (2003)
as a dispersal event from the La Plata basin towards the Eastern South American coastal rivers.
In clade D3, the calculated age for the origin of the species occupying the La Plata basin is 9.35 Ma. (eight
species). This estimated age also coincides with the event of the marine transgression that occurred 10-5 Ma
(Hernández et al., 2005). Also, within this clade, there are species of the São Francisco river, which reveal,
according to Montoya-Burgos (2003), a colonization event of the São Francisco river from the La Plata basin.
In clade D4, the node that contains the species inhabiting the La Plata basin is dated to 13.5 Ma; since then at
least 10 species were formed. This estimated age is in accordance with the first event of maximum flooding of the
marine transgression that occurred 15-13 Ma (Hernández et al., 2005). As previously mentioned, within this clade
emerges H. johnii from Northern South America coastal rivers and H. luetkeni from Eastern South America coastal
rivers. This result demonstrates two dispersal events; both are posterior to the origin of the species diversity of the
La Plata basin.
It is important to note that intra-basin diversification increases with the age of the origin of the lineage. In clade
D2 the intra-basin diversification started 7.45 Ma and generated seven species; in clade D3 it started 9.35 Ma and
gave rise to eight species; in clade D4 it started 13.5 Ma and resulted in ten species. These species numbers are
underestimates as new species might be discovered and others might have become extinct. In this respect, the new
species described here, H. arecuta, contributes importantly to our understanding that clade D4 is the most ancient
and diverse Hypostomus lineage inhabiting almost exclusively the La Plata basin. This high diversity is the result
of intra-basin speciations.
Therefore, we see that Hypostomus species diversity in the Paraná river, and in consequence in the La Plata
basin, is moulded by two processes. One is the inter-basin diversification, which generated groups of species
inhabiting different basins as result of dispersal events, as proposed by the hydrogeological hypothesis (Montoya-
Burgos, 2003). In this context, paleo hydrogeological changes during the Miocene have promoted vicariance and
dispersal routes yielding a high degree of diversification of species of fishes in the Neotropical region. The other
process that shaped species diversity of Hypostomus is intra-basin speciation, which produced groups of species
inside a basin due to habitat fragmentation. We see that the origin of species diversity inside the La Plata basin
temporally matches with the marine transgression in the Paraná river basin. As proposed by the museum hypothesis
(Nores, 1999), this marine incursion could have fragmented the La Plata basin resulting in several allopatric
speciation events. The historical biogeography of Hypostomus argues that several documented hydrological and
sea level changes deeply influenced the cladogenetic events observed in the phylogeny of this genus.
Zootaxa 3453 © 2012 Magnolia Press · 81
ORIGIN OF SPECIES DIVERSITY IN HYPOSTOMUS.
Possible connection between the Paraná and Uruguay rivers
The hydrographical patterns of the Paraná and Uruguay rivers indicate that they can be considered as
belonging to the same basin (La Plata basin) as the Lower Paraná is connected to the Lower Uruguay via the Río de
la Plata estuary. This configuration has been maintained almost unchanged for the last 10 Ma (Lundberg et al.,
1998). Sivasundar et al. (2001) mentioned that several conspecific populations are currently distributed along the
Paraná and Uruguay rivers, dispersing probably through the Río de la Plata estuary. This dispersal route, which
allows gene flow between the two rivers, may explain why representatives of H. commersoni are genetically
similar in those two rivers as well as in the Río de la Plata estuary. However, according to the distribution range of
some Hypostomus species in the La Plata basin, we can propose other possible ancient dispersal routes between the
Paraná and Uruguay rivers. The examined material in this work and the bibliography about Hypostomus
luteomaculatus, H. uruguayensis, and H. ternetzi show that these species are distributed in the Paraná and Uruguay
rivers. Contrary to H. commersoni, these species have never been reported from the Río de la Plata estuary. These
three species could have either gone extinct or have left the Río de la Plata estuary. Alternatively, they might have
dispersed through temporal connections between Northern tributaries of the Paraná and Uruguay rivers.
During the Miocene (24-5 Ma), the Paraná and Uruguay rivers became disconnected from one another
resulting in the present configuration (Beurlen, 1970). However, the topology and proximity between some
tributaries of these two rivers allows us to hypothesize that water pathways between the Paraná and Uruguay rivers
could have existed during flood periods. Weber (1987) suggested that the Aguapey river can be a connection
between the Paraná and Uruguay rivers. Later on, Casciotta et al. (2005) mentioned that it is probable that at
present ichthyofaunal exchanges can take place between the Laguna Iberá (Paraná river basin) and the Miriñay
river (Uruguay river basin) during flood periods. An exhaustive population genetic analysis could be useful to
understand the dispersal routes used by some Hypostomus species to maintain gene flow between populations
inhabiting Paraná and Uruguay rivers basin.
We ar e grateful to F. Br ancolini, E. Ru eda, A. Pa racampo, F. Vargas, and members of the INICNE (UNNE) for
helping us in the field trips and in the sampling process. We thank R. Corvain and S. Fisch-Muller (MHNG) for
sharing samples with us. We thank Convenio IBOL-CONICET Argentina, CONICET, and Comisión de
Investigaciones Científicas de la Provincia de Buenos Aires, Claraz Foundation, and the Canton de Genève for
their financial support for field trips and laboratory equipment. We also thank the PUCRS for sharing with us
specimens of H. luteomaculatus and H. luteus, and James Maclaine of NHM for the images of syntypes of H.
strigaticeps. Wa are grateful to M. Rodrigues de Carvalho and two anonymous reviewers for their valuable
comments and suggestions.
Additional specimens examined
Argentina. Hypostomus boulengeri, MACN-ict 9644, 1 ex., 121.8 mm, Corrientes Province, Paraná river at
Ituzaingó. Hypostomus commersoni: ILPLA1907, 8ex, El Pescado, La Plata, Buenos Aires. Hypostomus
paranensis: ILPLA1914, 2 ex, Córdoba, Córdoba Capital, Suquia river. Hypostomus ternetzi, MACN-ict 9645, 1
ex. 150 mm Corrientes Province, Paraná river at Yahape. Hypostomus uruguayensis, MACN-ict 9651, 1 ex., 159.0
mm, Corrientes Province, Paraná river at Yahape. Brazil. Hypostomus luteus: MCP19991, 1ex., Santa Catalina,
Uruguay river basin, Uruguai river, proximo a pedra da Fortaleza. MCP20751, 1ex., Santa Catalina, Uruguay river
basin, Uruguai river, proximo a pedra da Fortaleza. Hypostomus regani: MCP19989, 1 ex., Santa Catalina,
Uruguay river basin, Uruguai river, proximo a pedra da Fortaleza. MCP28628, 1 ex., Rio Grande do Sul, Uruguay
river basin, Uruguai river, no Remanso da Timbaúva, cerca de 1500m do início do Salto do Yucuma. MHNG
2547.017, 3ex. , 141, 86–162,09 mm, Sao Pablo, Mogui Guazu Cach.
CARDOSO ET AL.
82 · Zootaxa 3453 © 2012 Magnolia Press
Aljanabi, S.M., Martinez, I. (1997) Universal and rapid salt-extraction of high quality genomic DNA for PCR-based
techniques. Nucleic Acids Research, 25, 4692–4693.
Beurlen, K. (1970) Geologie Von Brasilien. Gebrüder Borntraeger, Berlin/Stuttgart.
Boeseman, M. (1968) The genus Hypostomus Lacepede, 1803, and its Surinam representatives (Siluriformes, Loricañidae).
Zoologische Verhandelingen, 99, 389.
Bonetto, A. (1994) Austral Rivers of South America. R. Margalef (Ed.): Limnology Now: A Paradigm of Planetary Problems.
Elsevier Science B.V. Amsterdam, The Netherlands.
Cardoso, Y.P., Montoya-Burgos J.I. (2009) Unexpected diversity in the catfish Pseudancistrus brevispinis reveals dispersal
routes in a Neotropical center of endemism: the Guyanas Region. Molecular Ecology, 18, 947–964.
Cardoso, Y.P., Brancolini, F., Protogino, L., Lizarralde, M. (2011) Actinopterygii, Siluriformes, Loricariidae, Hypostomus
aspilogaster (Cope, 1894) Distribution extension and first record for Argentina.Check List, 7, 596–598.
Cardoso, Y.P., Brancolini F., Paracampo A., Lizarralde M., Montoya-Burgos J.I. A new species of Hypostomus (Siluriformes:
Loricariidae) from the upper Paraguay river basin, Argentina. In preparation.
Casciotta, J., Almirón, A., Bechara, J. (2005) Peces del Iberá - Hábitat y diversidad. Fundacion Ecos.1st. edition, ISBN 987-
05-0375-6. 244 pp.
Eschmeyer, W.N., Fricke, R. (eds.). (14 July 2011) Catalog of Fishes electronic version. Available from: http://
Felsenstein, J. (1985) Confidence limits on phylogenies: An approach using the bootstrap. Evolution, 39, 783–791.
Ferraris, C.Jr. (2007) Checklist of catfishes, recent and fossil (Osteichthyes: Siluriformes), and catalogue of siluriform primary
types. Zootaxa, 1418, 1–628.
Hall, T. (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT.
Nucleic Acid Symposium Series, 41, 95–98.
Hernández, R.M., Jordan T.E., Dalenz Farjat A., Echavarría L., Idleman B.D., Reynolds J.H. (2005) Age, distribution,
tectoniccs, and eustatic controls of the Paranense and Caribean marine transgressions in the south Bolivia and Argentina.
South American Earth Sciences, 19, 495–512.
Hollanda Carvalho, P., Lima, F.C.T., Zawadzki, C.H. (2010) Two new species of the Hypostomus cochliodon group
(Siluriformes: Loricariidae) from the rio Negro basin in Brazil. Neotropical Ichthyology, 8, 39–48.
Hoorn, C. (1993) Marine incursions and the influence of Andean tectonics on the Miocene depositional history of northwestern
Amazonia: results of a palynostratigraphic study. Palaeogeography, Palaeoclimatology, Palaeoecology 105, 267–309.
Hubert, N., Duponchelle, F., Nuñez, J., Garcia-Davila, C., Paugy, D., Renno, J.F. (2007) Phylogeography of the piranha
Serrasalmus and Pygocentrus: implications for the diversifications of the Neotropical ichthyofauna. Molecular Ecology,
Huelsenbeck, J.P., Crandall, K.A. (1997) Phylogeny estimation and hypothesis testing using maximum likelihood. Annual
Review of Ecology and Systematic, 28, 437–466.
Huelsenbeck, J.P., Ronquist, F. (2001) MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics, 17, 754–755.
Jobb, G., von Haeseler, A., Strimmer, K. (2004) TreeFinder: a powerful graphical analysis environment for molecular
phylogenetics. BMC Evolutionary Biology, 4, 18.
Kosakovsky Pond, S.L., Muse, S.V., Frost, S.D.W. (2004) Hyphy package: hypothesis testing using phylogenies. Available
López, H.L., Miquelarena, A. (1991) Los Hypostominae (Pisces: Loricariidae) de Argentina. Vo l 4 0 , P i s ce s . Fa s c ic u l o 2
Hypostominae. Profadu (CONICET). Programa de Fauna de Agua Dulce Museo de La Plata. La Plata. Argentina.
Lovejoy, N.R., De Araújo, L.G. (2000) Molecular systematics, biogeography and population structure of Neotropical
freshwater needlefishes of the genus Potamorrhaphis Molecular Ecology, 9, 259–268.
Lundberg, J.G. (1998) The temporal context for the diversification of Neotropical fishes. In Phylogeny and classification of
Neotropical fishes, edited by L. R. MALABARBA, R. E. REIS, R. P. VARI, Z. M. LUCENA and C. A. S. LUCENA
(Edipucrs, Porto Alegre, Brasil), pp. 49–68.
Lundberg, J.G., Marshall, L.G., Guerrero, J. (1998) The stage for Neotropical fish diversification: a history of Tropical South
American rivers. In Phylogeny and classification of Neotropical fishes, edited by L. R. MALABARBA, R. E. REIS, R. P.
VA R I , Z . M . L U C E N A a n d C . A . S . L U C E N A ( E d i p u c r s , P o r t o A l e g r e , B r a s i l ) , p p . 1 3 – 4 8 .
Montoya-Burgos, J.I., Weber, C., Le Bail, P.Y. (2002) Phylogenetic relationships within Hypostomus (Siluriformes:
Loricariidae) and related genera based on mitochondrial D-loop sequences. Revue Suisse de Zoologie, 109, 369–382.
Montoya-Burgos, J.I. (2003) Historical biogeography of the catfish genus Hypostomus (Siluriformes: Loricariidae), with
implications on the diversification of Neotropical ichthyofauna. Molecular Ecology, 12, 1855–1867.
Muller, S., Weber, C. (1992) Les dents des sous-familles Hypostominae et Ancistrinae (Pisces, Siluriformes, Loricariidae) et
leur valeur taxonomique. Revue Suisse de Zoologie, Annales de la Société zoologique suisse et du Muséum d'Histoire
naturelle de Genève, 99, 747–754.
Nores, M. (1999) An alternative hypothesis for the origin of Amazonian bird diversity. Journal of Biogeography, 26, 475–485.
Nylander, J.A.A. (2004) MrAIC.pl. Program distributed by the author. Evolutionary Biology Centre, Uppsala University.
Oyakawa, O.T., Akama, A., Zanata, A.M. (2005) Review of the genus Hypostomus Lacépède, 1803 from rio Ribeira de Iguape
Zootaxa 3453 © 2012 Magnolia Press · 83
ORIGIN OF SPECIES DIVERSITY IN HYPOSTOMUS.
basin, with description of a new species (Pisces, Siluriformes, Loricariidae). Zootaxa, 921, 1–27.
Rasanen, M.E., Lima, A., Santos, J., Negri, F. (1995) Late Miocene Tidal Deposits in the Amazonian Foreland Basin. Science,
Reis, R.E., Weber, C., Malabarba, L.R. (1990) Review of the genus Hypostomus Lacepede,1803 from Southern Brazil, with
descriptions of the three new species (Pisces, Siluriformes, Loricariidae). Revue Suisse de Zoologie, 97, 729–766.
Rodriguez, M.S., Ortega, H., Covain, R. (2011) Intergeneric phylogenetic relationships in catfishes of the Loricariinae
(Siluriformes: Loricariidae), with the description of Fonchiiloricaria nanodon: a new genus and species from Peru.
Journal of Fish Biology, 79, 875–895.
Ronquist, F., Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19,
Sivasundar, A., Bermingham, E., Ortí, G. (2001) Population structure and biogeography of migratory freshwater fishes
(Prochilodus: Characiformes) in major South American rivers. Molecular Ecology, 10, 407–417.
Web er, C. ( 19 85) Hypostomus dlouhyi, nouvelle espèce, de poison chat cuirasse du Paraguay (Pisces, Siluriformes,
Loricariidae). Revue Suisse de Zoologie, 92, 955–968.
Web er, C. (19 87 ) Hypostomus microstomus sp. nov. et autres poissons-chats cuirasées du Rio Parana (Pisces, Siluriformes,
Loricariidae). Revue Suisse de Zoologie, 40, 273–284.
Web er, C. (20 03 ) Hy po sto mi nae . In: Reis, R.E., Kullander, S.O. and Ferraris, C.J., Jr. (Eds.), Check list of the Freshwater
Fishes of South and Central America. Edipucrs, Porto Alegre, Brazil, pp. 351–372.
Zawadzki, C.H., Renesto, E., de Paiva, S., Lara-Kamei, M.C.S. (2004) Allozyme differentiation of four populations of
Hypostomus (Teleostei: Loricariidae) from Ribeirão Keller, a small stream in the upper Rio Paraná basin, Brazil. Genetica,
Zawadzki C.H., Weber, C., Simone Pavanelli, C. (2010) A new dark-saddled species of Hypostomus (Siluriformes:
Loricariidae) from the upper rio Paraguay basin. Neotropical Ichthyology, 8, 719–725.