Identity of Hypostomus plecostomus (Linnaeus, 1758), with an overview of Hypostomus species from the Guianas (Teleostei: Siluriformes: Loricariidae) by Claude Weber (1), raphaël Covain (1, 2, 3) & Sonia FiSCh-Muller* (1) ABSTRACT. - Twenty-one nominal species of Hypostomus are recorded in fresh and brackish waters of the Guianas from the oyapock to the essequibo rivers. one of those, Hypostomus plecostomus, was the irst described representative of the genus. however this name was often erroneously applied to numerous other species. The identity of H. plecostomus was here investigated based on a dual approach allying classical morphometric and allozyme analyses of several populations representing respectively 15 and ive of the Guianese species. Morphometric analyses revealed four morphological groups among the different populations, namely the H. plecostomus, H. watwata, H. gymnorhynchus, and boeseman’s species groups. The discriminant model used for a posteriori assignment of the linnean type specimens of H. plecostomus to an extant species showed that the type series was heterogenous and comprised two species: H. plecostomus and H. watwata. a lectotype was designated to stabilize the taxonomy of H. plecostomus and its junior synonyms H. guacari and H. lava. Hypostomus plecostomus was also recognized based on ixed allelic differences with other Hypostomus species analysed. its expected heterozygosity (0.091) was relatively high, with little differentiation between populations. both analyses highlighted that H. plecostomus was a widespread species mostly distributed in the lower part of Guianese rivers. a complementary description of the species was provided. our analysis of the Hypostomus material from the oyapock to the essequibo river basins revealed 15 potentially valid species, and new synonymies for H. plecostomus, H. gymnorhynchus and H. corantijni were proposed. The current distribution of the recognized species was speciied. RéSumé. - identité de Hypostomus plecostomus (linnaeus, 1758), avec une revue des espèces d’Hypostomus des Guyanes (Teleostei: Siluriformes: loricariidae). vingt-et-une espèces nominales d’Hypostomus sont connues des eaux douces et saumâtres des Guyanes du leuve Oyapock à l’essequibo. l’une d’elles, Hypostomus plecostomus, fût le premier représentant du genre à être décrit, cependant ce nom est souvent appliqué à de nombreuses autres espèces. l’identité de H. plecostomus a été ici examinée par une double approche alliant morphométrie classique et analyse des allozymes pour différentes populations représentant, respectivement, 15 et 5 des espèces guyanaises. les analyses morphométriques ont révélé que les différentes populations formaient quatre groupes morphologiques distincts, nommés groupes H. plecostomus, H. watwata, H. gymnorhynchus, et “boeseman”. le modèle discriminant utilisé pour l’assignation à posteriori des spécimens type linnéens de H. plecostomus à une espèce actuelle a montré que la série type était hétérogène et comprenait deux espèces : H. plecostomus et H. watwata. un lectotype a été désigné ain de stabiliser la taxonomie de H. plecostomus et de ses synonymes juniors H. guacari et H. lava. Hypostomus plecostomus a été également caractérisé sur la base de différences alléliques ixes au regard des autres espèces d’Hypostomus analysées. le taux d’hétérozygotie attendue (0,091) s’est révélé relativement haut, avec peu de différence entre populations. les deux analyses ont révélé que H. plecostomus était une espèce largement distribuée dans la partie basse des rivières guyanaises. une description complémentaire de l’espèce est fournie. notre analyse du matériel d’Hypostomus des bassins de l’oyapock à l’essequibo a révélé 15 espèces potentiellement valides et de nouvelles synonymies sont proposées chez H. plecostomus, H. gymnorhynchus et H. corantijni. la distribution actuelle des espèces reconnues a été précisée. Key words. - Catish - Morphometry - Allozyme electrophoresis - Genetic variation - Neotropics - French Guiana - Suriname - Guyana. use of the name Hypostomus plecostomus The name Hypostomus plecostomus (linnaeus, 1758) has often been applied to any of several distinct species of South and Central American mailed catishes included in the genus Hypostomus lacepède, 1803 and even to other hypostominae genera. it occurs inappropriately in reports (e.g., benson et al., 2004; anderson et al., 2005), check-lists of invasive species (e.g., McCann et al., 1996), oficial documents (e.g., Gutiérrez and reaser, 2005), and within anatomical (e.g., Fernandes and Perna, 1995; Mattias et al., 1996), parasitological (e.g., Kritsky and boeger, 1991), physiological and ecological studies (Mazzoni et al., 2010) and recently in a (1) Muséum d’histoire naturelle, Département d’herpétologie et d’ichtyologie, route de Malagnou 1, Case postale 6434, Ch-1211 Genève 6, Suisse. [Claude.Weber@ville-ge.ch] [raphael.Covain@ville-ge.ch] [Sonia.Fisch-Muller@ville-ge.ch] (2) université de Genève, Département de génétique et évolution, Sciences iii, Quai e. ansermet 30, Ch-1211 Genève 4, Suisse. (3) université lyon 1, laboratoire de biométrie et de biologie évolutive, CnrS, uMr 5558, 43 bd du 11 novembre 1918, 69622villeurbanne cedex, France. * Corresponding author [Sonia.Fisch-Muller@ville-ge.ch] Cybium 2012, 36(1): 195-227. hypostomus species from the Guianas study of the trophic ecology of invasive suckermouth catish in Texas (Pound et al., 2011). The main reasons for the extensive use of this name are that Hypostomus plecostomus was not only described since the beginning of the linnean system, but also Hypostomus is widely distributed, and one of the most species-rich genera of loricariidae, with 127 species (Ferraris, 2007; Jerep et al., 2007; birindelli et al., 2007; Zawadzki et al., 2008a, 2008b, 2010; hollanda Carvalho et al., 2010) including numerous species with unclear taxonomic status. Historical account This species irst appeared in a pre-Linnean description of Acipenser indicus by linnaeus (1754). Four years later, linnaeus (1758) established the new taxon Acipenser plecostomus, making references to his earlier publication and to the also pre-linnean description of a species identified as Plecostomus by Gronovius (1754), specifying its habitat as “Surinami”. The genus Hypostomus was later described by lacepède (1803), including a single species described as Hypostomus guacari based on numerous literature sources, including the account of Acipenser plecostomus linnaeus, 1758. Hypostomus guacari lacepède, 1803, is therefore type species of Hypostomus by monotypy. one year later, Shaw (1804) described Loricaria lava, again based on different works including that of linnaeus. Type specimens of Hypostomus plecostomus are thus also type specimens of H. guacari and of L. lava (Ferraris, 2007), two species since long placed in the synonymy of H. plecostomus. The nomenclatural imbroglio involving that ish continued with Kner (1854), bleeker (1862-1863, 1864), and later authors. During the 19th and the irst half of the 20th century, while ichthyologists like Steindachner (1877a, 1877b, 1878, 1881), regan (1904, 1908a, 1908b, 1912) and Fowler (e.g., 1915, 1941, 1943) described numerous new species of Hypostomus from the neotropical freshwaters, Fowler (1954), van der Stigchel (1946), and other authors synonymised numerous species with H. plecostomus. unfortunately, the latter works were widely regarded as authoritative references, so that H. plecostomus was mentioned from numerous neotropical rivers and places, as in Fowler (1954: 191-192): “amazônia, rio Tocantins, rio Tapajós, rio Paranaiba, São Paulo, Paraná, rio Grande do Sul, Paraguay, bolívia, Peru, Colombia, venezuela, Guianas, Trinidad”. as a consequence, the species is also listed in large ish fauna’s revisions, like that of Panama by Meek and hildebrand (1916), and of venezuela by Schultz (1944). eigenmann (1922) himself described a new subspecies, Plecostomus plecostomus panamensis. none of these works was based on the examination of type specimens of Hypostomus plecostomus, which were regarded as lost (see boeseman, 1968: 11). 196 Weber et al. Search for the identity of Hypostomus plecostomus The first serious attempt to clear up the problematic identity of Hypostomus plecostomus was that of boeseman (1968), who confirmed the validity of Hypostomus lacepède, 1803 and of Hypostomus plecostomus (linnaeus, 1758). He also designated a neotype in order to ix the species and revise its identity. however, this neotype designation was done despite being aware of the possible existence of Linnean syntypes, which became effectively conirmed by Fernholm and Wheeler (1983) in an annotated catalogue of the Linnean ish specimens deposited in the Naturhistoriska riksmuseet in Stockholm (nrM). These authors found three putative syntypes in a jar (nrM 32) and, based on their examination in regard to linnaeus works, they argued that at least the smallest was to consider as a type. examination of this material was thus a prerequisite to the present study. We had recently the opportunity to go to Stockholm for examination of the external morphology and photographs of putative linnaeus’ type specimens of loricariids, including H. plecostomus. Allozyme analysis as molecular tool for inferring species limits external morphology, that is susceptible to environmental inluences, may distinguish morphotypes within a single species. Conversely, external morphology is often hardly able to discriminate sibling species. Identiication of species boundaries in such cases has beneitted from molecular techniques (hillis et al., 1996), and more particularly, numerous species of fish and other organisms have been confirmed using allozyme electrophoresis (larson, 1989; lessios et al., 1995; emberton et al., 1995; allibone, 1996). analysis of allozymes (Prakash et al., 1969), or the different forms of an enzyme encoded at a same locus, allows the identiication of alleles, the calculation of allelic and genotypic frequencies of populations, and the estimation of genetic low between populations. In case of sympatry, a single ixed allozymic difference between populations is suficient to both recognize and characterize two coexisting species, as it shows that interbreeding is not occurring (richardson et al., 1986). in case of allopatry, the extent of allozymic divergence between populations has to be assessed taking into account their geographical variation, a conservative approach being to consider only fixed allozymic differences, rather than measurements based on allele frequency differences, and to ally other data to determine speciic status (Richardson et al., 1986; Thorpe and Solé-Cava, 1994). Systematics of several loricariid taxa including Hypostomus has been investigated with this tool leading to the discrimination of congeneric species (Zawadzki et al., 1999, 2000, 2001; de Paiva et al., 2005; renesto et al., 2007), to the discovery of cryptic ones (Fisch-Muller et al., 2001), and to the inference of phyloCybium 2012, 36(1) hypostomus species from the Guianas Weber et al. genetic relationships (Zawadzki et al., 2005; renesto et al., 2007). Objectives of present work in the present work, we integrated genetics with classical morphometry in order to redeine the enigmatic Hypostomus plecostomus. The three rediscovered syntypes were compared to hundreds of Hypostomus specimens from the Guianas representing 15 nominal species by means of classical multivariate morphometric analyses. Most of this material was collected by the authors and collaborators during the past 15 years (see le bail et al., 2012; Mol et al., 2012). a signiicant part of the specimens was speciically sampled for diverse genetic studies, including an allozyme analysis that was conducted at the MhnG. The taxonomy of H. plecostomus is here accordingly redeined and a complementary description provided. The taxonomic status of several of the 21 nominal Guianese species of Hypostomus occurring from the oyapock to the essequibo rivers is also revised to establish a list of presumably valid species of the region. mATeRIAL And meTHOdS morphometry Specimens examined in this work represent 15 nominal species from the Guianas, and are detailed in the list of Material examined. institutional acronyms follow Fricke and eschmeyer (2010). Specimens were measured with a digital calliper to the nearest 0.01 mm following boeseman (1968). Counts followed Weber (1985). replacement teeth were examined following Muller and Weber (1992). Morphometric analyses were based on 15 continuous morphometric variables (listed in table i, with abbreviations in squared brackets) of 287 examined specimens, with the addition of 236 specimens examined by boeseman (1968: tables 1-20), leading to a total of 523 specimens representing 43 samples of populations for 15 nominal species. Specimens smaller than 30 mm were excluded from the analyses to minimize the bias introduced by allometric growth, and because morphometric data are highly correlated between them, missing data were estimated using the least squares method with the standard length (Sl) used as explanatory variable. To estimate the absence of signiicant differences between measurements provided by boeseman and ours, a Wilcoxon matched pairs signed rank test was performed on all common measurements recorded from the holotype of H. gymnorhynchus. in order to highlight the morphological structure of the species and populations under study, and to construct the model used for the assignment of the linnean types to an extent species, the data were submitted to multivariate analyses. To prevent artiicial groupings, the different populations collected in different basins for a given Cybium 2012, 36(1) species were considered independently. Prior to the analyses, all measurements were standardized by the Sl and log transformed to control for size effect. The inal table included thus 523 rows representing 43 populations and species of Hypostomus, and 14 columns. This table was then submitted to a principal components analysis (PCa) using the correlation matrix to reveal its structuring. The PCa was secondarily submitted to a linear discriminant analysis (Da) using the 43 populations and species as grouping factor. Prior to the Da, a non-parametric version of the Pillai test was conducted using 9,999 random permutations to test against the null hypothesis of equality of means between groups. The constructed discriminant functions were subsequently used for the assignment of supplementary rows (herein nrM type specimens). The robustness of the model was estimated by cross validation to evaluate the rate of wrong assignments of individuals to their respective class. Multivariate analyses were performed with the MaSS 7.3-7 (venables and ripley, 2002), ade4 1.4-16 (Dray and Dufour, 2007) and ade4TkGui 0.2-5 (Thioulouse and Dray, 2007) packages in r 2.12.1 (r Development Core Team, 2009). Populations and species’ scores onto discriminant axes were secondarily converted into a distance matrix. This transformation allowed a hierarchical classiication of the different populations and species of Hypostomus represented in a dendrogram. euclidian distances were measured between the centres of gravity of each class of individuals and analysed using the Weighted PairGroup Method using arithmetic means (WPGMa) algorithm (McQuitty, 1966). The WPGMa was favoured since the size of the different clusters was suspected to be greatly uneven due to the numerous species described by boeseman (1968, 1969). robustness of the results was estimated using a multiscale bootstrap resampling procedure (Shimodaira, 2002, 2004). For each cluster in the hierarchical clustering, p-values were calculated to indicate how strong the cluster was supported by the data. Two types of p-values were computed for each node: approximately unbiased (au) (Shimodaira, 2002; 2004) and bootstrap Probability (bP) (efron, 1979; Felsenstein, 1985). Cluster analysis and multiscale bootstraps were performed using 1,000 pseudoreplicates with the pvclust 1.2-1 package (Suzuki and Shimodaira, 2004) in r. Allozyme electrophoresis A total of 188 specimens (identiied by the mention of individual field numbers in the list of Material examined) representing 18 samples of populations originally attributed to ive Hypostomus species (H. plecostomus, H. ventromaculatus, H. watwata, H. gymnorhynchus, and H. tapanahoniensis) were analysed for allozymes. Two additional representatives of Hypostomus watwata (one from Mahury and a second from Kourou rivers) and a unique sample of H. corantijni were also stained for allozymes. These samples were not 197 Species Former identiication locality & acronym of population n Morphometry Standard length (Sl) Percents of Sl: Predorsal length [Predors_l] head length [head_l] Dorsal-in spine [Dspine_l] interdorsal length [interdor_l] Caudal-peduncule length [Peduncl_l] Caudal-peduncule depth [Peduncl_d] Percent of head length (hl): Cleithral width [Cleitr_w] head depth [Depth_head] Snout length [Snouth_l] orbital diameter [orbit_d] interorbital length [interorb_l] other percentages: Caudal-peduncule depth % of interdorsal length Caudal-peduncule depth % of peduncule length Dentary length % of interorbital length oyapock (voya) 28 range mean ± sd Kourou (vKou) 16 range mean ± sd 165.40 - 249.80 209.31 ± 21.08 145.50 - 248.30 213.07 ± 25.65 158.40 - 192.80 176.75 ± 14.71 132.20 - 212.40 177.58 ± 20.14 37.45 30.73 28.58 15.24 30.07 9.78 - 40.76 33.31 36.93 20.81 34.89 11.16 39.07 31.90 34.11 18.23 32.81 10.34 ± ± ± ± ± ± 0.87 0.76 2.07 1.23 1.11 0.37 37.54 31.35 29.19 15.78 29.59 9.88 - 42.51 34.48 37.18 21.13 34.93 11.30 1.27 1.06 2.51 1.52 1.58 0.42 38.07 30.76 31.64 17.65 31.75 9.40 - 39.76 - 32.51 - 37.18 - 18.83 - 33.04 - 10.13 38.97 31.57 34.03 18.23 32.44 9.84 ± ± ± ± ± ± 0.86 0.74 2.33 0.59 0.58 0.34 37.77 30.34 28.94 16.27 24.88 9.39 - 40.02 33.21 35.93 19.78 36.76 10.55 38.68 31.51 32.29 18.01 31.99 10.14 ± ± ± ± ± ± 0.75 0.86 2.00 1.07 2.42 0.30 87.07 57.32 57.40 14.16 40.30 - 104.40 - 68.64 - 61.56 - 16.67 - 44.56 99.50 62.46 59.65 15.23 42.37 ± ± ± ± ± 3.21 3.17 0.93 0.67 1.10 98.74 55.65 57.06 14.23 40.06 - 105.43 101.96 ± 2.09 75.97 68.02 ± 5.45 60.35 59.30 ± 0.80 16.36 15.22 ± 0.70 44.42 42.64 ± 1.44 93.20 59.38 56.70 16.58 41.71 - 96.46 - 61.84 - 58.68 - 17.71 - 43.51 94.83 61.08 57.57 17.23 42.19 ± ± ± ± ± 1.34 1.14 0.82 0.49 0.88 90.89 58.18 56.54 15.77 40.15 - 101.99 - 68.43 - 60.46 - 18.68 - 43.87 96.31 63.00 58.97 17.08 42.55 ± ± ± ± ± 2.93 2.85 0.93 0.84 1.00 49.54 - 70.13 28.04 - 35.42 28.57 - 37.33 19 22 19 22 26 26 2 8 6 7 14 - 39 42 39 41 27 27 3 9 8 8 15 57.03 ± 5.05 31.56 ± 1.67 31.46 ± 2.07 32 32 30 30 27 27 3 9 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 4 4 5 5 0 0 1 0 1 0 0 49.45 - 71.58 28.89 - 37.73 25.82 - 33.02 24 24 25 27 26 26 3 8 6 7 13 - 36 39 38 40 28 28 3 9 8 9 15 40.13 33.01 33.52 18.25 31.70 10.40 ± ± ± ± ± ± 57.46 ± 6.29 32.93 ± 2.64 29.36 ± 2.04 31 31 31 30 27 27 3 9 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 3 4 4 4 0 0 0 1 1 1 1 50.44 - 57.38 29.60 - 30.83 28.87 - 36.05 29 29 28 29 27 27 3 8 6 7 14 - 45 - 41 - 41 - 39 - 27 - 27 -3 -9 -7 -8 - 15 54.09 ± 3.48 30.34 ± 0.54 31.09 ± 3.35 36 34 34 33 27 27 3 9 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 7 5 5 4 0 0 0 1 1 1 1 47.46 - 64.04 28.92 - 35.45 23.89 - 37.80 23 23 22 21 26 26 3 7 6 7 13 - 40 41 38 41 27 27 3 9 8 9 15 56.70 ± 4.09 31.87 ± 1.60 29.21 ± 3.55 32 31 31 32 27 27 3 8 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 5 7 5 7 0 0 0 1 1 0 0 Weber et al. Cybium 2012, 36(1) Counts number of premaxillary teeth (l) number of premaxillary teeth (r) number of dentary teeth (l) number of dentary teeth (r) number of plates in the lateral series (l) number of plates in the lateral series (r) number of predorsal rows of plates number of dorsal plates Number of plates between dorsal and adipose ins Number of plates between adipose and caudal ins Number of plates between anal and caudal ins Hypostomus plecostomus H. cf. ventromaculatus Kaw (vKaw) Mahury (vCay) 16 4 range mean ± sd range mean ± sd hypostomus species from the Guianas 198 Table i. - Descriptive morphometrics and meristics for 21 samples of Hypostomus populations from French Guiana and Suriname, and of type material of H. plecostomus and H. watwata, with their abbreviations in brackets. asterisks indicate data from boeseman (1868). abbreviations for morphometric variables used in morphometric analyses are provided in square brackets. n: number of specimens. Weber et al. Cybium 2012, 36(1) Table i. - Continued. Species Former identiication locality & acronym of population n Morphometry Standard length (Sl) Percents of Sl: Predorsal length [Predors_l] head length [head_l] Dorsal-in spine [Dspine_l] interdorsal length [interdor_l] Caudal-peduncule length [Peduncl_l] Caudal-peduncule depth [Peduncl_d] Percent of head length (hl): Cleithral width [Cleitr_w] head depth [Depth_head] Snout length [Snouth_l] orbital diameter [orbit_d] interorbital length [interorb_l] other percentages: Caudal-peduncule depth % of interdorsal length Caudal-peduncule depth % of peduncule length Dentary length % of interorbital length Mana + Maroni (vMar) 26 range mean ± sd Suriname (vCom) + (vMak)* 13 range mean ± sd holotype 119.60 - 212.40 167.40 ± 25.68 72.00 - 107.00 84.33 ± 19.66 135.20 205.00 161.73 ± 23.10 56.00 - 150.00 92.08 ± 31.61 150.00 38.47 31.06 28.41 15.15 28.34 10.11 - 43.28 34.62 35.70 19.40 33.24 11.52 40.16 32.69 32.47 17.89 31.15 10.79 ± ± ± ± ± ± 1.09 0.89 1.90 1.02 1.28 0.32 40.37 33.64 36.45 16.62 29.46 9.86 - 42.57 37.30 37.36 17.29 32.08 10.37 41.81 36.01 36.90 17.00 30.48 10.17 ± ± ± ± ± ± 1.25 2.05 0.65 0.34 1.40 0.27 38.00 30.39 32.54 16.12 27.59 10.18 ± ± ± ± ± ± 1.45 1.34 3.00 1.40 1.48 0.39 40.07 32.00 32.46 15.43 28.08 9.82 - 42.32 36.79 40.62 17.92 29.23 10.77 41.33 35.18 36.42 16.63 28.79 10.28 ± ± ± ± ± ± 0.76 1.32 2.12 0.69 0.31 0.34 40.07 32.00 34.47 16.53 29.00 10.33 94.92 62.28 57.00 14.74 39.42 - 106.61 - 68.61 - 60.99 - 19.31 - 44.97 99.93 65.64 58.96 17.04 42.05 ± ± ± ± ± 2.79 1.88 1.08 1.41 1.36 84.64 51.31 53.93 19.93 38.77 - 93.06 56.94 55.56 20.60 41.67 87.86 54.68 54.97 20.36 39.80 ± ± ± ± ± 4.54 2.98 0.90 0.38 1.62 96.69 64.19 58.85 15.88 41.44 ± 107.38 101.10 ± 3.17 ± 69.85 66.62 ± 1.95 ± 61.67 60.17 ± 0.80 ± 17.78 16.99 ± 0.57 ± 46.90 43.52 ± 1.79 83.90 51.46 52.91 17.92 37.91 - 99.38 61.71 61.86 22.33 42.37 90.25 56.33 56.71 20.34 40.71 ± ± ± ± ± 5.66 3.46 2.91 1.38 1.23 99.38 61.67 60.00 17.92 41.04 61.88 ± 2.56 35.71 ± 1.17 - 62.50 35.63 - 52.92 - 70.77 31.27 - 39.40 27.65 - 34.91 23 22 20 24 27 27 2 8 6 7 13 - 42 45 41 41 28 28 3 9 8 8 15 60.56 ± 4.20 34.72 ± 1.94 31.09 ± 2.32 31 31 30 31 27 27 3 9 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 5 6 5 5 0 0 0 0 1 0 0 59.35 - 60.16 32.03 - 34.69 - - - 59.84 ± 0.43 33.40 ± 1.33 - - - 42.90 35.28 41.86 20.67 32.49 11.76 51.81 ± 72.94 31.35 ± 42.63 23.20 ± 30.53 22 22 19 21 26 26 3 8 6 7 13 ± ± ± ± ± ± ± ± ± ± ± 28 29 27 29 27 27 3 9 7 8 15 40.56 32.98 35.49 18.04 31.28 10.79 ± ± ± ± ± ± 60.26 ± 6.33 34.62 ± 2.95 26.44 ± 2.36 25 25 23 25 27 27 3 8 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 2 2 2 2 1 0 0 0 0 0 1 57.89 - 66.40 34.16 - 37.89 - - - - - - 199 hypostomus species from the Guianas Counts number of premaxillary teeth (l) number of premaxillary teeth (r) number of dentary teeth (l) number of dentary teeth (r) number of plates in the lateral series (l) number of plates in the lateral series (r) number of predorsal rows of plates number of dorsal plates Number of plates between dorsal and adipose ins Number of plates between adipose and caudal ins Number of plates between anal and caudal ins Hypostomus plecostomus H. ventromaculatus Maroni (vMarb)* Suriname. Paramaribo (vSur) 3 11 range mean ± sd range mean ± sd Species Former identiication locality & acronym of population n Morphometry Standard length (Sl) Percents of Sl: Predorsal length [Predors_l] head length [head_l] Dorsal-in spine [Dspine_l] interdorsal length [interdor_l] Caudal-peduncule length [Peduncl_l] Caudal-peduncule depth [Peduncl_d] Percent of head length (hl): Cleithral width [Cleitr_w] head depth [Depth_head] Snout length [Snouth_l] orbital diameter [orbit_d] interorbital length [interorb_l] other percentages: Caudal-peduncule depth % of interdorsal length Caudal-peduncule depth % of peduncule length Dentary length % of interorbital length H. hemiurus essequibo (heme) 4 range mean ± sd (Psyn1) lectotype 92.70 - 141.60 117.33 ± 16.57 161.80 201.90 175.39 ± 11.44 100.00 - 115.00 107.25 ± ± 39.15 - 41.99 40.47 ± 0.96 39.35 - 41.94 40.76 ± 0.85 38.14 - 42.70 40.94 ± 31.00 - 34.48 33.38 ± 1.23 31.71 - 33.76 32.93 ± 0.65 33.04 - 36.20 34.87 ± 27.80 - 38.05 34.07 ± 4.43 31.20 - 37.22 35.41 ± 1.75 35.44 - 38.94 36.83 ± 17.23 - 19.14 18.35 ± 0.64 16.55 - 21.16 18.93 ± 1.20 16.00 - 18.35 17.11 ± 29.52 - 32.45 31.21 ± 1.16 29.60 - 32.60 31.15 ± 1.03 27.62 - 31.20 29.50 ± 9.54 - 10.77 10.36 ± 0.39 10.27 - 11.67 10.94 ± 0.36 10.27 - 10.98 10.68 ± 6.09 57.80 98.25 - 141.10 116.24 ± 21.07 1.11 0.91 1.22 0.85 0.94 0.24 43.08 37.46 17.13 31.49 9.78 38.41 32.13 29.88 17.99 29.66 9.24 - 39.20 34.23 33.01 18.82 30.72 10.17 38.88 32.91 30.92 18.34 30.13 9.62 ± ± ± ± ± ± 0.38 0.95 1.41 0.36 0.52 0.40 92.08 - 102.96 59.94 - 71.53 57.74 - 61.72 18.10 - 20.54 41.15 - 45.48 95.74 64.27 59.19 19.22 43.08 2.20 1.92 1.07 0.93 1.85 81.06 64.29 54.04 21.48 39.26 87.27 56.82 54.92 15.50 38.42 - 97.44 62.62 57.91 16.94 41.34 92.37 60.89 56.23 16.16 39.37 ± ± ± ± ± 4.32 2.74 1.52 0.61 1.33 51.88 - 60.41 30.39 - 35.89 33.33 - 41.91 56.48 ± 2.64 33.21 ± 1.62 35.95 ± 3.02 62.54 ± 3.10 36.23 ± 1.47 - 57.07 31.04 27.06 50.05 - 56.54 30.40 - 34.22 27.11 - 31.38 21 20 20 18 26 26 3 7 7 5 12 - 27 27 24 26 27 28 3 8 8 8 14 24 23 22 21 27 27 3 8 8 6 13 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 3.45 3.61 1.43 0.78 1.43 2 3 2 3 0 1 0 0 1 1 1 98.23 63.00 58.51 14.97 40.60 - 107.73 103.42 ± 2.52 - 70.02 66.61 ± 1.83 - 62.68 61.00 ± 1.12 - 17.12 16.09 ± 0.67 - 47.10 45.48 ± 1.65 50.66 - 67.87 32.14 - 39.10 25.27 - 29.88 21 21 16 17 26 26 3 8 6 7 13 - 29 31 25 27 27 27 3 9 7 8 15 58.04 ± 4.60 35.16 ± 1.87 27.12 ± 1.38 25 25 22 23 26 26 3 9 6 8 14 ± ± ± ± ± ± ± ± ± ± ± 2 2 3 3 1 1 0 1 0 0 0 94.32 57.36 56.90 16.41 41.09 - 100.79 - 63.61 - 60.53 - 20.05 - 47.64 58.29 - 66.67 34.38 - 38.97 - - - 97.47 60.66 58.02 18.50 43.34 - ± ± ± ± ± - 27 27 3 7 7 - 24 24 24 23 25 27 3 8 7 8 14 - 26 26 26 28 27 27 3 8 8 8 15 52.47 ± 2.89 31.94 ± 1.62 29.57 ± 1.82 25 25 25 25 27 27 3 8 7 8 4 ± ± ± ± ± ± ± ± ± ± ± 1 1 1 1 1 0 0 0 1 0 1 Weber et al. Cybium 2012, 36(1) Counts number of premaxillary teeth (l) number of premaxillary teeth (r) number of dentary teeth (l) number of dentary teeth (r) number of plates in the lateral series (l) number of plates in the lateral series (r) number of predorsal rows of plates number of dorsal plates Number of plates between dorsal and adipose ins Number of plates between adipose and caudal ins Number of plates between anal and caudal ins Commewijne (PMap) 8 range mean ± sd Hypostomus plecostomus H. plecostomus Suriname. Paulus Creek (PPaK) Saramacca (PSarb)* 17 12 range mean ± sd range mean ± sd hypostomus species from the Guianas 200 Table i. - Continued. Weber et al. Cybium 2012, 36(1) Table i. - Continued. Species Former identiication locality & acronym of population n Hypostomus plecostomus all forms & populations mean ± sd Kourou (WKou) 10 range mean ± sd 56.00 - 249.80 159.65 ± 49.44 179.50 - 283.90 245.70 ± 48.73 258.30 - 390.00 324.31 ± 35.52 134.00 - 352.00 231.14 ± 71.26 37.45 30.34 27.80 15.15 24.88 9.24 - 44.71 41.39 41.86 21.16 36.76 11.76 40.23 33.22 34.29 17.87 31.18 10.46 ± ± ± ± ± ± 1.46 1.83 2.62 1.28 1.73 0.47 36.25 29.27 30.45 19.41 34.43 8.90 - 39.11 32.03 37.21 21.90 36.63 9.47 37.58 30.58 32.81 20.64 35.22 9.15 ± ± ± ± ± ± 1.24 1.20 3.17 1.16 1.04 0.27 34.58 27.72 0.00 19.55 33.37 7.54 - 39.58 32.49 32.29 22.90 38.05 8.85 36.77 30.03 27.48 21.25 35.54 8.36 ± ± ± ± ± ± 1.12 1.26 6.77 0.89 1.35 0.35 35.57 29.46 29.05 16.67 33.77 8.15 39.93 33.06 34.84 22.78 37.81 - 9.27 37.58 30.93 31.24 20.69 35.42 8.58 ± ± ± ± ± ± 1.20 1.28 2.15 1.80 1.28 0.31 76.51 - 107.73 49.75 - 75.97 46.98 - 62.68 14.16 - 24.16 37.91 - 47.64 97.78 63.19 58.75 17.17 42.51 ± ± ± ± ± 5.72 4.77 2.32 2.07 1.92 94.26 59.59 55.48 12.15 45.61 - 98.81 63.90 57.03 14.78 47.69 96.56 61.30 56.23 13.24 46.71 ± ± ± ± ± 2.46 2.08 0.64 1.11 1.03 84.69 55.63 50.80 11.26 41.27 - 99.63 65.54 57.38 14.16 47.88 93.07 61.16 55.69 13.04 44.64 ± ± ± ± ± 3.48 2.31 1.61 0.77 1.75 89.62 60.88 55.35 13.02 44.71 - 93.99 62.04 56.98 15.27 46.37 ± ± ± ± ± 2.10 1.49 0.83 1.60 1.17 47.46 - 72.94 28.04 - 42.63 23.20 - 41.91 58.81 ± 4.93 33.64 ± 2.46 29.95 ± 3.23 19 20 16 17 25 26 2 7 6 5 12 - 45 45 41 41 28 28 3 9 8 9 15 30 29 28 29 27 27 3 9 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 5 5 6 6 0 0 0 1 1 1 1 40.65 - 47.49 24.42 - 27.51 26.74 - 28.76 24 24 22 23 26 26 3 7 6 8 14 - 30 30 31 32 26 26 3 9 7 9 14 44.43 ± 3.01 25.99 ± 1.27 27.46 ± 0.90 27 27 26 27 26 26 3 8 7 9 14 ± ± ± ± ± ± ± ± ± ± ± 3 4 5 5 0 0 0 1 1 1 0 32.92 - 43.87 19.81 - 25.75 24.40 - 34.15 23 21 24 20 26 26 3 7 6 7 13 - 29 30 30 32 27 27 3 9 8 9 14 39.42 ± 2.61 23.56 ± 1.44 28.01 ± 2.20 27 26 28 27 27 27 3 8 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 2 3 2 4 1 1 0 1 1 1 1 - 96.27 65.70 57.84 17.60 48.50 35.79 - 55.64 21.56 - 27.46 25.07 - 30.91 20 21 16 22 26 26 3 7 6 8 13 - 29 28 31 29 27 27 3 8 7 9 14 41.92 ± 5.56 24.28 ± 1.57 28.01 ± 1.54 26 26 27 27 27 27 3 8 7 9 14 ± ± ± ± ± ± ± ± ± ± ± 3 3 5 2 1 1 0 1 1 1 1 201 hypostomus species from the Guianas Counts number of premaxillary teeth (l) number of premaxillary teeth (r) number of dentary teeth (l) number of dentary teeth (r) number of plates in the lateral series (l) number of plates in the lateral series (r) number of predorsal rows of plates number of dorsal plates Number of plates between dorsal and adipose ins Number of plates between adipose and caudal ins Number of plates between anal and caudal ins oyapok (Woya) 4 range mean ± sd 158 range Morphometry Standard length (Sl) Percents of Sl: Predorsal length [Predors_l] head length [head_l] Dorsal-in spine [Dspine_l] interdorsal length [interdor_l] Caudal-peduncule length [Peduncl_l] Caudal-peduncule depth [Peduncl_d] Percent of head length (hl): Cleithral width [Cleitr_w] head depth [Depth_head] Snout length [Snouth_l] orbital diameter [orbit_d] interorbital length [interorb_l] other percentages: Caudal-peduncule depth % of interdorsal length Caudal-peduncule depth % of peduncule length Dentary length % of interorbital length Hypostomus watwata H. watwata Cayenne (WCay) 20 range mean ± sd Species Former identiication locality & acronym of population n Morphometry Standard length (Sl) Percents of Sl: Predorsal length [Predors_l] head length [head_l] Dorsal-in spine [Dspine_l] interdorsal length [interdor_l] Caudal-peduncule length [Peduncl_l] Caudal-peduncule depth [Peduncl_d] Percent of head length (hl): Cleithral width [Cleitr_w] head depth [Depth_head] Snout length [Snouth_l] orbital diameter [orbit_d] interorbital length [interorb_l] other percentages: Caudal-peduncule depth % of interdorsal length Caudal-peduncule depth % of peduncule length Dentary length % of interorbital length 185.90 - 295.00 241.14 ± 39.38 120.10 372.00 232.71 Suriname (WPuK) 14 range mean ± sd 74.47 130.00 - 253.00 170.75 ± 56.23 107.90 - 141.10 126.70 ± 11.94 35.80 28.37 29.75 19.63 33.00 8.09 - 38.33 30.93 33.54 23.63 35.66 8.76 37.18 29.75 31.51 21.81 34.73 8.38 ± ± ± ± ± ± 1.04 0.96 1.29 1.26 0.81 0.24 35.76 29.24 27.80 17.82 33.14 8.80 - 39.88 33.14 36.25 22.77 36.14 9.53 38.46 31.81 32.42 20.63 34.51 9.15 ± ± ± ± ± ± 1.33 1.07 2.52 1.22 1.13 0.25 37.39 30.00 31.07 19.00 31.50 8.42 - 40.92 33.23 33.00 20.50 33.99 9.36 39.57 32.11 32.44 19.62 33.08 8.83 ± ± ± ± ± ± 1.54 1.44 0.92 0.64 1.14 0.43 39.26 32.25 32.33 18.14 32.96 9.22 - 40.59 34.57 34.07 19.37 33.95 9.78 39.94 33.43 33.20 18.71 33.35 9.56 ± ± ± ± ± ± 0.54 0.89 1.23 0.47 0.41 0.23 91.29 56.91 55.75 13.50 43.44 - 99.28 64.28 58.67 17.60 46.44 94.30 61.14 57.38 15.21 45.00 ± ± ± ± ± 2.75 2.26 1.03 1.18 0.97 89.08 58.69 51.54 11.36 40.19 - 99.59 66.53 58.16 18.09 47.95 95.38 61.83 56.34 14.26 45.95 ± ± ± ± ± 2.49 2.13 1.65 1.99 1.98 90.38 58.10 55.32 13.18 43.67 - 98.55 61.53 57.18 18.06 45.60 94.80 59.64 56.32 16.00 44.70 ± ± ± ± ± 3.44 1.56 0.96 2.07 0.82 92.23 59.95 55.22 14.29 45.09 - 96.96 62.18 57.58 14.95 47.91 95.22 60.78 56.07 14.62 46.98 ± ± ± ± ± 1.94 0.84 0.94 0.25 1.17 36.28 - 41.64 23.46 - 24.84 28.57 - 32.90 24 22 20 23 25 26 3 7 6 8 13 - 34 35 33 33 26 26 3 9 7 9 14 38.53 ± 2.05 24.14 ± 0.49 30.10 ± 1.38 30 31 29 29 26 26 3 8 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 3 4 4 3 0 0 0 1 0 0 1 39.43 - 53.48 24.24 - 28.64 26.83 - 31.06 24 23 22 25 25 25 3 7 6 7 13 - 33 32 34 32 26 26 3 8 7 8 14 44.52 ± 3.41 26.25 ± 1.41 28.60 ± 1.20 298 28 27 28 26 26 3 8 6 8 14 ± ± ± ± ± ± ± ± ± ± ± 3 3 3 2 0 0 0 0 1 0 0 43.47 - 45.88 24.77 - 29.71 - - - 45.02 ± 1.09 26.75 ± 2.11 - - - 49.58 - 53.13 27.61 - 29.67 23.03 - 26.94 11 19 18 18 26 26 3 7 6 8 14 - 26 27 27 26 26 26 3 8 7 8 14 51.11 ± 1.80 28.66 ± 0.95 25.13 ± 1.78 20 22 22 21 26 26 3 8 6 8 14 ± ± ± ± ± ± ± ± ± ± ± 6 3 3 3 0 0 0 0 0 0 0 Weber et al. Cybium 2012, 36(1) Counts number of premaxillary teeth (l) number of premaxillary teeth (r) number of dentary teeth (l) number of dentary teeth (r) number of plates in the lateral series (l) number of plates in the lateral series (r) number of predorsal rows of plates number of dorsal plates Number of plates between dorsal and adipose ins Number of plates between adipose and caudal ins Number of plates between anal and caudal ins Sinnamary (WSin) 9 range mean ± sd Hypostomus watwata H. watwata Mana + Maroni (WMar) Maroni (WMarb)* 14 4 range mean ± sd range mean ± sd hypostomus species from the Guianas 202 Table i. - Continued. Weber et al. Cybium 2012, 36(1) Table i. - end. Species Former identiication locality & acronym of population Mouth of Suriname river * n 4 range Morphometry Standard length (Sl) Percents of Sl: Predorsal length [Predors_l] head length [head_l] Dorsal-in spine [Dspine_l] interdorsal length [interdor_l] Caudal-peduncule length [Peduncl_l] Caudal-peduncule depth [Peduncl_d] Percent of head length (hl): Cleithral width [Cleitr_w] head depth [Depth_head] Snout length [Snouth_l] orbital diameter [orbit_d] interorbital length [interorb_l] other percentages: Caudal-peduncule depth % of interdorsal length Caudal-peduncule depth % of peduncule length Dentary length % of interorbital length all forms & populations 86 range mean ± sd 113.00 - 217.00 162.25 ± ± 37.56 - 41.26 39.94 ± 31.34 - 34.07 33.02 ± 31.34 - 35.93 32.79 ± 19.71 - 20.99 20.48 ± 32.70 - 35.48 33.53 ± 8.61 - 9.51 9.03 ± 46.37 138.00 - 202.00 176.75 ± 30.41 260.00 81.40 82.00 107.90 - 390.00 240.90 ± 77.67 1.65 1.25 2.12 0.55 1.32 0.37 39.41 31.63 32.97 18.44 32.99 9.06 - 40.72 33.91 34.73 21.25 34.41 9.50 40.00 32.92 34.07 19.92 33.46 9.25 ± ± ± ± ± ± 0.57 1.00 0.77 1.48 0.64 0.19 37.04 30.00 31.54 20.19 35.92 8.65 44.10 37.71 34.15 19.10 34.83 8.97 41.59 35.24 18.90 35.00 8.41 34.58 27.72 25.90 16.67 31.50 7.54 44.10 37.71 37.21 23.63 38.05 - 9.78 37.96 31.13 31.33 20.71 34.78 8.70 ± ± ± ± ± ± 1.59 1.62 2.55 1.37 1.38 0.50 90.91 52.99 54.03 13.97 42.67 3.44 5.11 2.27 1.80 2.02 91.23 57.05 54.70 14.09 45.08 - 96.78 64.95 59.42 16.03 47.10 94.16 61.50 57.08 15.02 46.12 ± ± ± ± ± 2.70 3.98 1.93 0.92 0.89 98.97 62.31 56.54 13.72 44.87 85.34 61.38 52.28 17.92 43.81 87.20 59.58 53.63 17.99 44.64 84.69 52.99 50.80 11.26 40.19 - 94.28 61.16 56.39 14.34 45.52 ± ± ± ± ± 2.91 2.43 1.47 1.61 1.71 46.61 ± 3.18 27.65 ± 0.87 - 42.86 24.09 - 46.95 25.75 26.77 44.52 24.04 - 32.92 - 55.64 19.81 - 29.71 23.03 - 34.15 23 25 25 - 23 25 25 - - - 97.65 63.99 58.63 17.66 47.57 43.59 - 45.29 25.32 - 28.64 - - - 93.97 58.64 56.42 15.71 45.39 ± ± ± ± ± 44.10 ± 0.80 26.97 ± 1.44 - - - 43.33 - 50.32 26.62 - 28.57 - - - - - 11 19 16 18 25 25 3 7 6 7 13 - - 99.63 66.53 59.42 18.09 48.50 34 35 34 33 27 27 3 9 8 9 14 42.61 ± 4.60 25.29 ± 2.07 28.18 ± 2.01 27 27 27 27 26 26 3 8 7 8 14 ± ± ± ± ± ± ± ± ± ± ± 4 4 3 3 0 0 0 1 1 0 0 203 hypostomus species from the Guianas Counts number of premaxillary teeth (l) number of premaxillary teeth (r) number of dentary teeth (l) number of dentary teeth (r) number of plates in the lateral series (l) number of plates in the lateral series (r) number of predorsal rows of plates number of dorsal plates Number of plates between dorsal and adipose ins Number of plates between adipose and caudal ins Number of plates between anal and caudal ins mean ± sd Hypostomus watwata H. watwata H. plecostomus Coppename + Saramacca (Wneo)* (Psyn2) (Psyn3) (WCop)* 4 1 2 range mean ± sd neotype paralectotypes hypostomus species from the Guianas included in the analyses due to the small samples size and, for H. corantijni to a poor electrophoretic resolution that may lead to misinterpretation of the electrophoregrams. These samples were regarded only qualitatively, providing additional information to the species boundaries. White skeletal muscle samples were taken from each specimen and frozen in liquid nitrogen immediately after collection and later stored at –80°C. liver samples of some specimens were also preserved in the objective of electrophoregram interpretation. Preparation of protein extracts and electrophoretic procedures are detailed in Fisch-Muller et al. (2001). Staining for eight enzyme systems (Tab. ii) follow Pasteur et al. (1987). The mobility of the enzymes was determined in separate side-by-side electrophoretic runs, and electromorphs of various loci were considered homologous if they exhibited the same mobility. The individual genotypes were inferred on the basis of the allozyme phenotypes following Pasteur et al. (1987) and buth (1990). enzyme abbreviations and locus nomenclature follow Shaklee et al. (1990). alleles were designated alphabetically based on decreasing electrophoretic mobility. Data were analysed using TFPGa 1.3 (Miller, 1997). Genotype proportions at each polymorphic locus within a population were tested for deviation from hardy-Weinberg equilibrium using haldane’s exact test (haldane, 1954). a multiple testing correction procedure was applied to control for the False Discovery rate (FDr) (benjamini and hochberg, 1995). Parameters used to estimate genetic variation of populations and species are: percentage of polymorphic loci at 95% (P0.95) and 99% (P0.99) criteria, and observed (ho) and unbiased estimates of average heterozygosity (he) (nei, 1978). exact tests for population differentiation (raymond and rousset, 1995) were conducted using a contingency table approach and the Markov chain procedure on allelic frequencies of each locus at population and species levels, and the Fisher’s Combined Probability test was computed as a global test over loci to determine the overall signiicance. Intraspeciic and interspeciic genetic distances were calculated using Nei’s unbiased formulae adapted for small samples (nei, 1978), which can give negative values. The genetic distances matrix was then Weber et al. submitted to an agglomerative hierarchical clustering analysis using the unweighed Pair-Group Method with arithmetic mean (uPGMa) (Sneath and Sokal, 1973) algorithm. The resulting tree was used as a graphical representation of the genetic distances to infer phenetic relationships between populations. To estimate robustness of the clusters, a nonparametric bootstrap analysis (efron, 1979) was performed following Felsenstein’s (1985) methodology using 10’000 pseudoreplicates. Only ixed allozymic differences at isozyme loci, which are hypothesized from observed allelic frequencies of the samples of populations, provide unequivocal identiication of specimens and have been used as diagnostic markers of species according to Davis and nixon (1992). in case of allopatry and in the absence of evidence from other type of data, a minimum of two ixed allozymic differences was necessary to recognize distinct species. electrophoretic gels are preserved in MhnG together with photographs, and are available upon request. ReSuLTS morphometry Characteristics of the data set no significant differences were observed between our measurements and those from boeseman since the Wilcoxon matched pairs signed rank test failed to reject the null hypothesis (h0: MCW – Mmb = 0 vs. h1: MCW – Mmb ≠ 0; v = 54, p-value = 0.6441). all available data were accordingly gathered into a single data set and submitted to the multivariate analyses. The between class inertia recorded by the Da represented 37.02% of the total inertia (sum of eigenvalues / matrix rank: 5.1835 / 14 = 0.3702). The nonparametric version of the Pillai test (Fig. 1D) was highly signiicant with none of the null hypothesis sampling distribution of randomized values greater than the observed value of between class inertia (Xobs = 0.3702; pXrand ≥ pXobs = 0.0001). Signiicant differences between groups were thus present in the data, and these differences were not due to chance. Table ii. - enzymes names, abbreviations and enzyme Commission numbers. enzymes e.C. numbers Creatine kinase (CK) 2.7.3.2 Glycerol-3-phosphate dehydrogenase (naD+) (G3PDh) 1.1.1.8 Glucose-6-phosphate isomerase (GPi) 5.3.1.9 l-lactate dehydrogenase (lDh) 1.1.1.27 Malate dehydrogenase (MDh) 1.1.1.37 Malic enzyme (naDP+) (MeP) 1.1.1.40 Mannose-6-phosphate isomerase (MPi) 5.3.1.8 Phosphoglucomutase (PGM) 5.4.2.2 204 Discriminant analysis of the different populations and species Morphometric data were mainly structured on the first two axes of Da (Fig. 1C) that explained 35.84% of the total between-class inertia (18.46% for axis 1 and 17.38% for axis 2). The irst axis split the different populations and species of Hypostomus into four main groups (Fig. 1a). on the negative side, the first one corresponded to a group mixing five Surinamese species described by boeseman, followed by a second group comprising different populations of H. gymnorhynchus and their close morphological relatives. On the positive side, a irst group comCybium 2012, 36(1) Weber et al. hypostomus species from the Guianas Figure 1. - linear Discriminant analysis of the different populations and species representing most of the Guianese Hypostomus. A: Projection of 523 specimens distributed in 43 groups, and a posteriori group assignment of the three Linnean syntypes onto the irst factorial plane of the Da; populations and species labelled as in Material examined and table i. B: Projection of the discriminant factors onto the irst factorial plane of the DA; variables labelled as in table I. C: eigenvalues of the Da. d: non parametric Pillai test. prised different populations of H. plecostomus and H. ventromaculatus, followed by a second group comprising all populations of H. watwata. The second axis split H. emarginatus from the groups comprising the different populations of H. gymnorhynchus and relatives plus all populations of H. watwata on positive values, and the groups comprising the ive different species of Boeseman and the different populations of H. plecostomus and H. ventromaculatus on negatives values. Four species groups are accordingly recognized and named: boeseman (for five species described by boeseman), gymnorhynchus, plecostomus, and watwata groups. The species H. emarginatus was perfectly distinct from all other Guianese species, whereas the two populations of H. hemiurus appeared intermediary between the boeseman and gymnorhynchus groups. on the negative side of axis 1, the boeseman group corresponded to high values for maximum orbital diameter, and snout length (Fig. 1b). It comprised ive representatives of the Surinamese species of Hypostomus described by boeseman (inc. type material): H. macrophthalmus, H. pseudohemiurus, H. crassicauda, H. paucimaculatus, H. saramaccensis, and two populations of H. micromaculatus (brokopondo and Gran rio). on the positive side of axis 1, the watwata group corresponded to high values for the interorbital width (Fig. 1b). it consisted in the Guianese populations of the nominal species: Cayenne, Kourou, Sinnamary, Maroni, Suriname, oyapock, berbice (boeseman’s neotype of Acipenser plecostomus), and Coppename rivers, plus the population from Pulp Creek Cybium 2012, 36(1) (lower Suriname river). on the positive side of axis 2, H. emarginatus and the gymnorhynchus group corresponded to high values for the interdorsal and caudal peduncle lengths (Fig. 1b). The gymnorhynchus group comprised different populations of the nominal species (approuague, Sinnamary, Comté, Mana and upper Maroni rivers), plus the representatives (inc. type material) of H. tapanahoniensis, H. corantijni, H. nickeriensis, H. occidentalis, H. surinamensis, and H. sipaliwinii. on the negative side of axis 2 the plecostomus group corresponded to high values for the caudal peduncle depth, and in less important contributions to the cleithral width (Fig. 1b). The plecostomus group comprised a mix of different populations of H. plecostomus (Paulus Creek-lower Suriname river, Saramacca river, Marshall Creek-lower Suriname river, Suriname river, and Mapana Creek-lower Commewijn river), of H. ventromaculatus including H. cf. ventromaculatus (sensu le bail, 2000) [oyapock, Kaw, Kourou, Cayenne, Maroni, Commewijn, and Suriname (type material) rivers], and a population identiied as H. hemiurus from lower essequibo river. The two other populations of H. hemiurus were poorly characterized in negative values on the irst axis. Discriminant model and a posteriori assignment of Linnean types The cross validation procedure used to validate the discriminant functions, and accordingly their predictive abilities, recovered a rate of false assignment of 36% considering the 205 hypostomus species from the Guianas grouping by population. by grouping the different populations of a given species into species groups, this rate dropped to 8% of false assignment. These wrong assignments concerned however mainly the boeseman group. The model was thus considered valid at the speciic level, and cautiously at the population level. The a posteriori class assignment predicted the irst putative syntype of H. plecostomus (Psyn1: Sl = 57.8 mm) as a member of H. plecostomus Mapana Creek, lower Commewijn river using either all discriminant functions (n = 14) or only the two irst ones. The second putative syntype of H. plecostomus (Psyn2: Sl = 81.40 mm) was assigned to H. watwata, Maroni river, using either 14 or 2 discriminant functions. The third putative syntype of H. plecostomus (Psyn3: Sl = 82.00 mm) was assigned to H. watwata, Kourou river, using the 14 discriminant functions, and to H. watwata, Maroni river, using the discriminant functions 1 and 2. Since these two syntypes (Psyn2 and Psyn3) belonged to the same lot, it was very unlikely that one was collected in Maroni river and the second in Kourou River. The two irst discriminant functions appeared thus as the best predictors, and class’ scores onto the two irst axes of the Da were consequently used for the cluster analysis. Cluster analysis of the different populations and species The clustering structure of the morphometric dataset using WPGMA was high (agglomerative coeficient = 0.91). Weber et al. The WPGMa tree (Fig. 2) recovered five clusters corresponding to the four morphological groups previously deined plus H. emarginatus. The irst cluster only comprised H. emarginatus. The second cluster grouped all populations of H. watwata plus syntypes 2 and 3 of H. plecostomus. This cluster was statistically highly supported by both au and bP p-values (99% au / 99% bootstrap), as well as all internal positions including the nested positions of these two syntypes within the watwata group (95% au / 94% bootstrap for syntype 2, and 90% au / 78% bootstrap for syntype 3). The morphological group watwata was thus morphologically homogeneous and well structured. The next cluster comprised all species and populations constituting the gymnorhynchus group. This cluster was also highly supported (100% au / 100% bootstrap) as well as most of its internal relationships. like the watwata group, the gymnorhynchus group was morphologically homogeneous and well structured. The fourth cluster grouped all populations and species constituting the plecostomus group, including the syntype 1. This cluster was also strongly supported (91% au / 71% bootstrap) but fell out of the bootstrap 95% conidence interval. This cluster comprised more variability, and only subsamples were comprised within the conidence interval such as H. ventromaculatus from Kaw river and H. plecostomus from Paulus Creek-lower Suriname river (100% au / 100% bootstrap), H. plecostomus from Marshall Creek, and the Figure 2. - WPGMa tree performed on the resulting scores of the Da with multiscale bootstraps analyses. Populations and species labelled as in Material examined and table I; agglomerative coeficient = 0.91; cluster supports are reported above nodes: left approximately unbiased (au) and right bootstrap probability (bp); squares represent conidence intervals at 95% bp. 206 Cybium 2012, 36(1) Hypostomus plecostomus H. ventromaculatus H. plecostomus Kourou Maroni Suriname 1 Suriname 2 Saramacca 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.350 0.143 0.650 0.821 1.000 1.000 1.000 0.036 1.000 1.000 1.000 0.971 0.833 0.029 0.167 1.000 1.000 1.000 1.000 1.000 0.850 1.000 1.000 0.971 1.000 0.150 0.029 1.000 1.000 1.000 1.000 1.000 1.000 0.750 1.000 1.000 1.000 0.250 0.550 0.607 0.546 0.647 0.333 0.450 0.393 0.455 0.353 0.667 1.000 1.000 1.000 1.000 1.000 10.000 12.818 10.900 16.900 2.900 30.000 27.273 10.000 10.000 20.000 30.000 27.273 10.000 30.000 20.000 0.150 0.110 0.036 0.071 0.100 0.127 0.109 0.052 0.059 0.087 Total 1.000 1.000 1.000 0.056 0.939 0.005 0.986 0.014 1.000 0.980 0.015 0.005 1.000 0.830 0.170 0.576 0.424 0.020 0.980 84.182 27.273 54.546 0.077 0.091 Hypostomus watwata H. watwata oyapock Cayenne Sinnamary 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.583 1.000 1.000 0.417 0.083 1.000 1.000 0.917 0.250 0.333 0.417 0.750 0.667 0.583 0.375 0.056 0.500 0.625 0.944 0.500 4.000 8.100 6.000 22.222 20.000 44.444 22.222 20.000 44.444 0.083 0.056 0.241 0.107 0.058 0.197 hypostomus species from the Guianas 207 Species Former identiication H. cf. ventromaculatus loci alleles oyapock Kaw Mahury Ck a 1.000 1.000 1.000 b G3pdh a b 1.000 1.000 1.000 Gpi-a a b c d Gpi-b a b c d 1.000 1.000 1.000 e ldh-1 a b 1.000 1.000 1.000 c ldh-2 a 1.000 1.000 1.000 b Mdh a 1.000 1.000 1.000 b c Mep-1 a b 1.000 1.000 1.000 Mep-2 a 0.605 0.607 1.000 b 0.395 0.393 c Mpi a 0.575 0.546 0.583 b 0.425 0.455 0.417 0.143 Pgm a b 1.000 0.857 1.000 n 17.500 12.500 7.800 P0.95 20.000 30.000 10.000 P0.99 20.000 30.000 10.000 ho 0.061 0.105 0.050 he 0.099 0.127 0.053 Weber et al. Cybium 2012, 36(1) Table iii. - observed allelic frequencies and measures of genetic diversity at population and species (bold) levels, for 18 samples of populations of Hypostomus from French Guiana and Suriname. H. tapa: H. tapanahoniensis. n: average sample size; P0.95, P0.99: percentage of polymorphic loci at 95% and 99% criterions; ho: observed heterozygosity; he: expected heterozygosity (nei, 1978). Dark grey areas indicate species’s diagnostic alleles. light grey area indicates no locus expression detected for the species. hypostomus species from the Guianas Weber et al. Table iii. - Continued. Species Former identiication loci alleles Ck a b G3pdh a b Gpi-a a b c d Gpi-b a b c d e ldh-1 a b c ldh-2 a b Mdh a b c Mep-1 a b Mep-2 a b c Mpi a b Pgm a b n P0.95 P0.99 ho he Hypostomus watwata H. watwata Mana Maroni Suriname 1.000 1.000 1.000 1.000 1.000 1.000 0.143 0.857 0.125 1.000 1.000 0.875 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.625 0.550 0.375 0.375 0.450 0.625 0.500 0.350 0.250 0.500 0.650 0.750 3.889 10.000 7.900 22.222 22.222 40.000 22.222 22.222 40.000 0.083 0.133 0.154 0.123 0.111 0.140 Hypostomus gymnorhynchus H. gymnorhynchus H. tapa Total approuague Mahury Sinnamary Maroni 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.111 0.889 1.000 0.929 1.000 0.857 0.071 0.143 0.071 1.000 1.000 1.000 0.929 0.024 0.976 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.088 0.913 1.000 1.000 1.000 1.000 0.013 0.988 1.000 1.000 1.000 1.000 0.427 1.000 1.000 1.000 1.000 0.573 0.305 1.000 1.000 0.695 1.000 1.000 37.500 8.909 7.000 6.000 6.727 40.000 0.000 11.111 0.000 18.182 60.000 0.000 11.111 0.000 18.182 0.135 0.000 0.016 0.000 0.039 0.137 0.000 0.016 0.000 0.037 type series of H. ventromaculatus (100% au / 100% bootstrap), and the populations of H. ventromaculatus from Cayenne, Kourou, and oyapock rivers, the population of H. plecostomus from Mapana Creek-lower Commewijn river, and the population of H. hemiurus from lower essequibo river (98% au / 97% bootstrap). The plecostomus group appeared thus morphologically less well structured than the previous groups. The last cluster comprised two highly supported groups, one made of two populations of H. hemiurus (100% au / 99% bootstrap), and a second comprising the species of the boeseman group (100% au / 100% bootstrap). The 208 Total 1.000 1.000 0.948 0.052 0.017 0.983 1.000 1.000 1.000 1.000 1.000 1.000 0.552 0.448 27.364 18.182 27.273 0.013 0.058 boeseman group displayed high supports in its internal relationships and the two populations of H. micromaculatus clustered together reinforcing the conclusion of a group morphologically well structured. Allozyme electrophoresis Enzymatic polymorphism The staining of eight enzyme systems (Tab. ii) in 18 populations of Hypostomus allowed the scoring of 30 alleles encoded by 11 putative gene loci. The surveyed loci and the observed allelic frequencies for each sample of population Cybium 2012, 36(1) Weber et al. hypostomus species from the Guianas are presented in table iii. no allele was detected for following cases: locus Gpi-A in H. plecostomus group except one specimen from Maroni, and in H. watwata group except two specimens from Cayenne, and locus Ldh-1 in all H. watwata group. none of the 11 studied loci was found invariant, as they had from two (Ck, Mep-1, Mpi and Pgm) to ive (GpiB) alternative alleles in the total sampling. none of the polymorphic loci for each population showed deviation to hardy-Weinberg equilibrium expectations except Mep-2, that showed a signiicant deviation (p = 0.004; cp(FDr) = 0.152) for the population of H. ventromaculatus from oyapock (H. plecostomus group), with an excess of homozygotes. however, after correction for False Discovery rate, this test could have been signiicant only by chance. In addition, this population appeared in hardy-Weinberg equilibrium at its second polymorphic locus, Mpi. Fixed allelic differences and species distinction Comparison of allelic composition at each locus indicated that all populations identiied as H. plecostomus group with morphology were ixed, in regards to H. gymnorhynchus and H. watwata groups, for distinctive and diagnostic alleles at three loci: G3pdh, Ldh-1 and Ldh-2 (Tab. iii). however, at the species group level, no ixed allelic difference was found between populations preliminary identiied as Hypostomus cf. ventromaculatus, H. ventromaculatus and H. plecostomus (Tab. iii), ranging from the oyapock to the Saramacca. The absence of any ixed allelic difference provided additional evidence that all these populations represent a single species: H. plecostomus. all analysed populations of H. watwata (H. watwata group), from the oyapock to the Suriname basins, showed also a unique and exclusive allele at locus Ck, conirming the speciic identity of H. watwata. Finally, Hypostomus gymnorhynchus group, including three populations of H. gymnorhynchus and one of H. tapanahoniensis, had ixed allelic differences with both other species (H. plecostomus and H. watwata) at no fewer than four loci: Gpi-A, Gpi-B, Ldh-1, and Mep-2. in addition alleles at these four loci are unique in our sampling. at H. gymnorhynchus group level, one fixed allelic difference was observed for locus Pgm between two pairs of populations: H. gymnorhynchus from approuague and Mahury, both with allele Pgm-a, and H. gymnorhynchus from Sinnamary plus H. tapanahoniensis from Maroni, both showing allele Pgm-b. These clusters did not correspond to the two putative morphological species of the H. gymnorhynchus group. in the clear absence of morphologic evidence, a single ixed allelic difference is not suficient to consider these two groups of allopatric populations as distinct species. in addition, the two alleles alternatively present were not unique in the sampling, as one of them was also found in H. plecostomus (Pgm-a), and both in Cybium 2012, 36(1) Figure 3. - uPGMa dendrogram of unbiased nei’s genetic distances (nei, 1978) between 18 samples of populations of Hypostomus from French Guiana and Suriname. negative branches not allowed. Agglomerative coeficient = 0.99. Only bootstrap values above 50 are reported. H. watwata, showing that they cannot constitute diagnostic characters. Thus, in the 18 populations analysed, a total of eight loci can be considered as diagnostic markers for three distinct species, H. plecostomus (including H. ventromaculatus and H. cf. ventromaculatus), H. watwata, and H. gymnorhynchus (including H. tapanahoniensis). exact tests for population differentiation between each pair of the three species conirmed overall signiicant differences in allele frequencies (p = 0.000 for each analysis). nei’s unbiased genetic distances between all pairs of populations ranged from -0.008 to 1.057. They are graphically summarized by an uPGMa dendrogram (Fig. 3), showing a clear grouping of the different populations into three clusters corresponding to the three species. Genetic distances were 0.558 between H. plecostomus and H. watwata, 0.933 between H. plecostomus and H. gymnorhynchus, and 0.624 between H. watwata and H. gymnorhynchus. Genetic identities between the species took the following respective values: 0.572, 0.393, and 0.536. besides allelic divergences, a difference in enzyme expression was observed for all populations of H. watwata, for which no product corresponding to Ldh-1 activity was revealed in contrast to each population of H. plecostomus and H. gymnorhynchus. The two single specimens from Mahury and from Kourou (not included in population analyses) showed same allelic characteristics as all other H. watwata for the different loci surveyed, conirming their identiication. 209 hypostomus species from the Guianas Intraspeciic variation Five rare alleles that appear exclusive to ive populations were found: Gpi-A-a in H. watwata from Suriname, Gpi-B-e in H. plecostomus from Maroni, Mdh-B-b in H. plecostomus from Kourou, Mdh-B-c in H. plecostomus from Suriname 2 (Paulus Creek), and Mep-1-a in H. watwata from Sinnamary (Tab. iii). Parameters estimating genetic variation are indicated in table iii. at population level, proportion of polymorphic loci (whatever the criteria) ranged from zero in two populations of H. gymnorhynchus (approuague and Sinnamary) to 44.4% in H. watwata from Sinnamary, and observed and unbiased estimates of average heterozygosity ranged respectively from zero to 0.241 and to 0.197 in the same populations. The hypothesis of an absence of population differentiation between each of the eight populations of Hypostomus plecostomus was rejected in all cases except two: oyapock and Kourou (p = 0.006), and Kaw and Kourou (p = 0.014), two pairs of populations that appeared differentiated on the base of their allelic frequencies. Genetic distances varied from -0.003 to 0.031, the most divergent populations being Kaw and Kourou. Percentage of polymorphic loci ranged from 10 to 30%, the most polymorph populations according to 95% criterion being Kaw and Kourou. expected heterozygosity (he) ranged from 0.052 for Suriname 1 (Paramaribo) to 0.127 for Kaw and Kourou populations. at species level, expected heterozygosity was 0.091. no overall significant population differentiation was found between the six populations of H. watwata (0.087 ≤ p ≤ 1.000), the species being thus genetically totally homogeneous. Genetic distances varied from -0.008 (Mana and Maroni) to 0.058 (Cayenne and Sinnamary). Percentage of polymorphic loci ranged from 20 to 44% and expected heterozygosity ranged from 0.058 to 0.197, Cayenne being the less variable and Sinnamary the most variable populations according to these parameters. Despite a fixed allelic difference between two pairs of populations (see above), H. gymnorhynchus showed no signiicant population differentiation (0.242 ≤ p ≤ 1.000). Genetic distances between populations varied from 0.000 (approuague and Mahury) to 0.120 (Mahury and Maroni). out of the four populations analysed, two showed no polymorphic locus (approuague and Sinnamary). The most polymorph population was that of Maroni (P0.95 = 18.18; he = 0.037). Hypostomus plecostomus (Linnaeus, 1758) (Figs 4, 5, 7a, 8a-b, 10; Tabs i, iii) Acipenser plecostomus linnaeus, 1758: 238. Type locality: Surinami (=Suriname river). lectotype (designated here): nrM 32, 57.8 mm Sl (illustrated in linnaeus, 1754: pl. 28, ig. 4). 210 Weber et al. Hypostomus guacari lacepède, 1803: 144, 145, pl. 4 (Fig. 2). Type locality: les rivières de l’amérique méridionale. lectotype (designated here): nrM 32, 57.8 mm Sl (illustrated in Linnaeus, 1754: pl. 28, ig. 4). Loricaria lava Shaw, 1804: 38, pl. 101. Type locality: indian Seas. lectotype (designated here): nrM 32, 57.8 mm SL (illustrated in Linnaeus, 1754: pl. 28, ig. 4). Plecostomus bicirrosus Gronow in Gray, 1854: 158. Type locality: in Americes Meridionalis luminibus. Holotype or syntype: specimen illustrated in Gronovius (1754: 24, pl. 3, igs 1-2) (Ferraris, 2007: 259). Locality stated by Gronovius: “Patria ejus Surinamae Fluvius est” (=Suriname river). Hypostomus ventromaculatus boeseman, 1968: 65, pl. 15 (ig. 1). Type locality: Suriname River between Afobaka and brokopondo, Suriname. holotype: rMnh 25507. nov. syn. The specimen nrM 32 (Fig. 5, Tab. i, Psyn1 in morphometric analyses), 57.8 mm Sl, is designated here as lectotype for A. plecostomus, H. guacari and L. lava to ix the identity of Hypostomus plecostomus and its junior synonyms, in a concern of nomenclatorial stability. The choice of the lectotype is based on literature, on the illustration of linnaeus (1954: pl. 28, fig. 4), and on examination of the probably still only existing putative syntypes. Description of H. plecostomus was based on references to pre-linnean descriptions of “Acipenser indicus” by linnaeus (1754: 55) and of a “Plecostomus” by Gronovius (1754: 24). The specimens described by Gronovius are very likely no more in the Copenhagen’s collection (Wheeler, 1989). according to Fernholm and Wheeler (1983), three specimens registered as nrM 32 may be putative type specimens of H. plecostomus, but very probably the smallest one is the specimen originally described by linnaeus in 1754. The authors list several arguments in favour of this conclusion, including characteristics matching with description (small size) and illustration (absence of caudal ilaments). We studied these specimens, and can conirm that the smallest one is the only one itting with the description and illustration given by Linnaeus, leading to its designation for name-bearing type of the species. The neotype designation for A. plecostomus proposed by boeseman (1968: 11) was invalid due to the existence of this name-bearing type specimen, following article 75.8 of iCZn. The descriptions of Hypostomus guacari lacepède, 1803 and Loricaria lava Shaw, 1804 being based on numerous literature sources including account of Acipenser plecostomus linnaeus, 1758, syntypes of the latter are also syntypes of the former (Ferraris, 2007: 258). The specimen nrM 32 (57.8 mm Sl) being also a syntype of the nominal species H. guacari and L. lava, its designation as lectotype for these two species as well deinitely conirms their status of objective junior synonyms of H. plecostomus. Cybium 2012, 36(1) Weber et al. hypostomus species from the Guianas Figure 4. - Hypostomus plecostomus, Suriname, Mapana Creek, tributary of Commewijne river, MhnG 2708.047 (121.4 mm Sl). as pointed out by Fernhom and Wheeler, the two other specimens nrM 32 (81.4 and 82.0 mm Sl; Fig. 6; Psyn2 and Psyn3 in morphometric analyses) are more faded than the lectotype, clearly indicating that they either did not have the same origin or not the same conditions of preservation. as it cannot be excluded that linnaeus also examined them, they are considered possible paralectotypes. nevertheless, as already pointed out by the morphometric analyses, their morphological characters, and in particular the numerous plates bordering the supraoccipital and the elongated caudalin spines (see Fig. 6) clearly show that they area not H. plecostomus but H. watwata hancock, 1828. diagnosis Hypostomus plecostomus differs from species of the H. Cybium 2012, 36(1) cochliodon group (H. macushi and H. taphorni in the Guianas) by viliform teeth versus large spoon-shaped teeth. it differs from other Guianese species of Hypostomus by the teeth bearing a short roughly rounded crown, with lateral cusp about half the length of the main cusp (Fig. 7a), versus an elongated crown (Fig. 7b), with a small lateral cusp (H. coppenamensis, H. corantijni, H. crassicauda, H. gymnorhynchus, H. nematopterus, and H. watwata), or sometimes with an enlarged lateral cusp (observed in H. gymnorhynchus from upper Maroni river basin only, H. micromaculatus and H. paucimaculatus). Hypsotomus plecostomus is also distinguished from long-crown toothed species except H. watwata by the presence of a preanal plate, versus absence. H. plecostomus further differs from H. watwata in having an azygous plate (sometimes divided into two to three secondary plate211 hypostomus species from the Guianas Weber et al. Figure 5. - Hypostomus plecostomus, lectotype, nrM 32, 57.8 mm Sl. lets) bordering the posterior part of the supraoccipital and directly contacting the pterotic-supracleithrum, laterally bordered by two plates that contact only the pterotic-supracleithrum (Fig. 5), versus an azygous plate (often divided anteriorly in small secondary platelets) not connecting the pterotic-supracleithrum, and with two lateral areas divided in multiple platelets (4 to 13) bordering the postero-lateral part of the supraoccipital and the posterior part of the pterotic-supracleithrum “(e.g. Fig. 6). description Measurements and counts in table i. body relatively 212 stocky, width at cleithrum almost equal to head length (76.5107.7, mean 97.8 ± 5.7% in HL). Dorsal proile slightly convex rising at an angle around 30o from snout tip to posterior end of supraoccipital process, then at an angle around 15 o up to dorsal-in origin, and decreasing gently from that point to end of caudal peduncle, with a slightly concave area at last dorsal-in ray level. Ventral proile almost straight from snout tip to caudal in. Caudal peduncle laterally compressed, elliptic in cross section. head broad and deep (49.8-76.0, mean 63.2 ± 4.8% in hl), covered dorsally with dermal ossiications, except for small naked area at snout tip. Median longitudinal bulge associated with mesethmoid terminating Cybium 2012, 36(1) Weber et al. hypostomus species from the Guianas Figure 6. - Hypostomus watwata, possible paralectotype of H. plecostomus, nrM 32, 82.0 mm Sl. coequally with transversal through nares. eye medium sized (14.2-24.2, mean 17.2 ± 2.1% of hl), dorsolaterally placed. Conspicuous ridge originating laterally to nares, passing through supraorbital, and extending to posterior portion of pterotic-supracleithrum. Supraoccipital bone with moderately to highly developed median ridge, and with relatively well developed posterior process bordered by wide predorsal plate. Predorsal region with two almost parallel ridges, area between ridges lat. Oral disk circular, its width similar to interorbital distance. Medium-sized lips, lower lip not reachCybium 2012, 36(1) ing pectoral girdle, its inner surface covered with numerous small papillae. Patches of odontodes present on anterior pigmented surface of upper lip. Maxillary barbel length moderate, about one third of mouth width. Dentaries slightly curved and acutely angled, approximately 80º between left and right dentaries. Tooth sigmoid in proile, slender, with a short crown characteristic of H. plecostomus group, formed by short more or less rounded crown and smaller lateral cusp (external side) separated by v-shaped notch (Fig. 7a). Sides of body covered with five rows of moderately 213 hypostomus species from the Guianas Weber et al. Figure 7. - left dentary replacement teeth and tooth crowns. A: Hypostomus plecostomus from Suriname (1), oyapock (2), Kourou (3), Commewijne (4), Suriname (5, 6), and berbice river basins (7). B: other Guinanese species of Hypostomus: H. nematopterus (1), H. gymnorhynchus (2), H. crassicauda (3), H. corantijni (4), H. hemiurus (5), H. micromaculatus (6), and H. paucimaculatus (7). Scale bar 0.2 mm for teeth and 0.1 mm for tooth crowns. spinulose plates, all covered by small posteriorly directed odontodes. Dorsal series of plates starting at vertical through dorsal-in spinelet and posteriorly to third predorsal plate, with weak longitudinal keel from beginning to about adipose-fin origin. Mid-dorsal series of lateral plates keeled until about end of adipose in, forming a continued and longitudinally oriented keel on irst three plates, then straight lines of elongated odontodes slightly upward oriented on each plate. Median series bearing lateral line, not keeled but bearing few larger odontodes dorsally to pores of lateral line, up to vertical of adipose in. Mid-ventral series forming strong latero-ventral edge along irst ive or six plates, covered by fan-shaped lines of larger odontodes, gently decreas214 ing in size posteriorly. Plates in ventral series not keeled along corner of ventral and lateral surfaces. ventral surfaces of head plated laterally, variably plated areas posterior to lower lip, cleithrum plated, abdomen co vered with minute diamond shaped platelets in large specimens, except small areas around base of paired ins and urogenital opening. Single azygous plate anterior to anal-in origin (irst anal-in pterygiophore). Dorsal fin with anterior spinelet, laterally compressed spine, and seven branched rays. Dorsal in large; posterior most rays reaching adipose-in origin to adipose-in spine, dorsal-fin margin convex. adipose fin formed by single elevated platelet, and laterally compressed and curved spine Cybium 2012, 36(1) Weber et al. with membrane extending over two plates. Pectoral fin with spine, and six branched rays, irst ray equal in length to spine. Pectoral-in spine compressed and entirely covered with odontodes, aside with small leshy extremity. Pelvic in with one rounded spine and ive branched rays; distal border straight to slightly convex. Anal in with one lexible spine and four branched rays. Caudal in concave, with 14 inner branched rays and elongate unbranched spine dorsally and ventrally; ventral lobe longer than dorsal lobe. The osteology of H. plecostomus has been described by Schaefer (1987) based on 19 specimens from Demerara and Suriname rivers basins, in Guyana and Suriname respectively. Colouration living specimen beige-brown to brown with dark brown to blackish spots (Fig. 8a, b). in alcohol (Fig. 4), overall ground colour of body and ins brownish, lighter on ventral part. Dorsal surface of head and body entirely covered by numerous dark spots, very small and closely-set on head, larger and less closely-set along body, largest ones on caudal peduncle. Spots occasionally elongated, often coalescent, and forming more or less irregular rows along body. angular part of lateroventral series never spotted. Pattern of colouration of ventral surface of body very variable (Fig. 10), from hardly spotted, to partly covered faded blotches and to nearly fully covered with dark spots. These spots also very variable in size and shape, and sometimes coalescent. Dorsal with rounded spots on anterior part of spine and mainly along interradial membrane, forming rows. Pectoral-fin with rounded spots, usually larger on distal part; pectoralin spine with small to large dark spots dorsally, and large spots ventrally. Pelvic in with larger and wide-set spots, few paler or no spot on ventral side of spine. rounded blotches or stripes on anal in. Caudal in with several vertical rows of dark spots often forming transversal stripes. Geographic distribution in the Guianas, Hypostomus plecostomus is found in the coastal region from the oyapock river basin to the essequibo river basin (Fig. 11). The species inhabits the lower part of rivers still under tide inluence, characterized by a quiet lowing water and muddy substrate, and swamps. Synonymy of Hypostomus ventromaculatus Boeseman, 1968 Two main characters were given by boeseman (1968: 66, 33) to distinguish his new species, H. ventromaculatus, from H. plecostomus. The irst one is a “delated irst dorsal in almost or just reaching base of spine of second dorsal in” versus “delated irst dorsal in usually distinctly overlapping base of spine of second dorsal in”, and the second concerns ventral coloration, with “body and peduncle, including the Cybium 2012, 36(1) hypostomus species from the Guianas ventral surface, covered with rather large, round, vague, dark spots” versus “body and peduncle covered with large and rather vague dark spots or blotches, lacking only on belly”. This second character appeared in a key to Suriname species of Hypostomus, distinguishing H. ventromaculatus from four species including H. plecostomus. however examination of material shows a high variability of the ventral colouration pattern (see Fig. 10), as well as of the size of dorsal in respective to adipose in. For example in a single lot of 16 specimens from lower Corantijn river (MhnG 2671.065), dorsal in does not reach base of adipose in in a specimen while it largely overlaps it in another, coupled with a spotted belly. in a sample from the approuague river in French Guiana (MHNG 2713.028), previously identiied as H. cf. ventromaculatus, dorsal fin almost or just reaches base of adipose fin in six specimens while it overlaps adipose-fin base in four others. These observations conirm the results of both morphological and allozyme analyses, that showed no signiicant difference between the populations previously identiied as distinct species, and we conclude that H. ventromaculatus has to be considered as a junior synonym of H. plecostomus. Hypostomus species in the Guianas based on the examination of material, on the literature, on morphometric analyses, and on allozyme electrophoresis, the validity of several species is here conirmed, some synonymies are maintained, and new synonymies are proposed. awaiting additional work including the examination of type specimens, four species (marked below with an asterisk) still have an uncertain taxonomic status. Two of them are known only from their type specimens that are small sized and may represent juvenile forms of other described species. We nevertheless prefer to take a conservative position and include them in the following list of presumably valid species occurring from the oyapock to the essequibo rivers: - Hypostomus plecostomus (linnaeus, 1758) (synonyms: H. guacari; L. lava; H. bicirrhosus; H. ventromaculatus, nov. syn.; see above) - Hypostomus watwata hancock, 1828 (synonym: H. verres valenciennes in Cuvier & valenciennes, 1840) - Hypostomus hemiurus (eigenmann, 1912) - Hypostomus gymnorhynchus (norman, 1926) (synonyms: H. gymnorhynchus occidentalis boeseman, 1968, nov. syn.; H. surinamensis boeseman, 1968, nov. syn.; H. gymnorhynchus tapanahoniensis boeseman, 1969, nov. syn.) (H. gymnorhynchus species group; see below) - Hypostomus corantijni boeseman, 1968 (synonyms: H. nickeriensis boeseman, 1969, nov. syn; H. sipaliwinii boeseman, 1968, nov. syn.) (H. gymnorhynchus species group; see below) - Hypostomus crassicauda boeseman, 1968 215 hypostomus species from the Guianas Weber et al. Figure 8. - live-color photographs of Guianese species of Hypostomus. A: Hypostomus plecostomus (Mapana Creek, Commenwijne river basin). B: H. plecostomus (berbice river). C: H. corantijni (Sipliwini river). d: H. crassicauda (Sipaliwini river). e: H. taphorni (Wonotobo Falls, Corantijn river). F: H. gymnorhynchus (Crique Grillon, orapu river). Pictures: r. Covain. - Hypostomus macrophthalmus boeseman, 1968 * (paratype: 72 mm Sl, largest specimen) - Hypostomus micromaculatus boeseman, 1968 216 - Hypostomus paucimaculatus boeseman, 1968 - Hypostomus pseudohemiurus boeseman, 1968 * (holotype: 62 mm Sl, largest specimen) Cybium 2012, 36(1) Weber et al. hypostomus species from the Guianas Figure 9. - live-color photographs of Guianese species of Hypostomus. A: Hypostomus hemiurus (Kurupukari Cross, essequibo river). B: H. macushi (Kurupukari Cross, essequibo river). C: H. micromaculatus (Kossindo, Gran rio river). d: H. nematopterus (Camopi, oyapock river). e: H. paucimaculatus (Kossindo, Gran rio river). F: H. watwata (ouanary river, oyapock basin). Pictures: r. Covain, except F: J.i. Montoya-burgos. - Hypostomus saramaccensis boeseman, 1968 * - Hypostomus coppenamensis boeseman, 1969 * Cybium 2012, 36(1) - Hypostomus nematopterus isbrücker and nijssen, 1984 - Hypostomus taphorni (lilyestrom, 1984) 217 hypostomus species from the Guianas Weber et al. ple for describing H. g. tapanahoniensis, boeseman (1969: 129) after listing arguments in favour of the presence of H. gymnorhynchus in Suriname, argued that “13 out of 14 species of the now known Surinam inland forms are restriced [sic] to a single river basin or (one species) to two adjacent river systems, and the identiication of the single exception, the Tapanahoni H. gymnorhynchus, has thereby become even less acceptable”. he added that “Whether these forms really represent subspecies or, more likely, good species, remains a problem only to be solved after adequate additional material becomes available.” based on our analyses of recently collected material, no difference supporting the validity of these species was found. in particular, allozyme analysis showed no allelic distinction of the population previously identiied as H. tapanahoniensis versus the three populations of H. gymFigure 10. - ventral colour patterns of 4 Hypostomus plecostomus from Suriname, Mapana norhynchus. in addition, these four Creek, tributary of Commewijne river, MhnG 2708.047. H. gymnorhynchus-group populations showed fixed allelic differences with - Hypostomus macushi armbruster and de Souza, 2005 populations representing H. plecostomus and H. watwata, Pictures of living specimens representing all Hypostomus and share unique and diagnostic alleles at these loci, leadspecies from the Guianas but those of uncertain taxonomic ing to the conclusion that this sampling represented a single status are presented in igures 8 and 9. Geographic distribu- valid species, namely H. gymnorhynchus. tions of the 15 presumably valid species, based on examined Considering Hypostomus surinamensis, no unique charmaterial and on literature, are indicated in igure 11. acter was originally given to diagnose the species which was included in boeseman’s watwata group (as well as all other new synonymies in the Hypostomus gymnorhynchus species that form our gymnorhynchus group), and morphospecies group metric characters appeared very variable in this group (see based on morphology, seven of the 15 examined nomi- boeseman, 1969: diagrams 1 to 19). Specimens previously nal species of Guianese Hypostomus were here included in identified as H. surinamensis were here only investigated the H. gymnorhynchus group. The group was found very through morphology, but their data did not distinguish them homogeneous, and clearly distinct from the other groups, from H. gymnorhynchus. We therefore consider the species however no within group difference that may distinguish as a junior synonym of H. gymnorhynchus. any of the seven nominal species was supported by the morH. nickeriensis and H. sipaliwinii are regarded as synophometric analyses. except H. gymnorhynchus, originally nyms of H. corantijni because no unique diagnostic charknown from a single specimen from the approuague river acter was described for these species, and no evidence of a in French Guiana, all these species were originally described clear difference was found according to all data provided by by boeseman based on material originating from Suriname. boeseman (see tables 2 and 11 in boeseman, 1968, and table Hypostomus occidentalis and later H. tapanahoniensis ii in boeseman, 1969). Moreover examination of a large were both tentatively described as subspecies of H. gym- number of specimens from Corantijn and nickerie rivers norhynchus. They were extensively discussed (boeseman, did not provide evidences for the presence of three distinct 1968, 1969) and placed in what boeseman (1969: 128) sympatric species within these basins. We thus tentatively named the “gymnorhynchus-complex”. in the absence of identiied specimens representing both species in our samdiagnostic characters, his argumentation for these subspe- pling based on their geographic origin and morphological cies descriptions appeared nevertheless weak. as an exam- congruence with original descriptions. however, morpho218 Cybium 2012, 36(1) hypostomus species from the Guianas Weber et al. metric analyses were unable to distinguish these putative species within the group, as well as all other nominal species of the H. gymnorhynchus group. The single H. corantijni sample available for allozyme analysis was not included in present work due to insuficient electrophetic resolution, but it showed at least clearly distinctive alleles from those observed for H. gymnorhynchus at two loci (Gpi-b and Mep2), and also apparently a different allelic expression at Ldh-1 locus. These results, even when partial, sustain the validity of H. corantijni. a genetic delineation is thus observed between eastern and Western Suriname. Taxonomic status of Hypostomus brasiliensis (Bleeker, 1862, in Bleeker 1862-1863) Hypostomus brasiliensis was since today considered a synonym of H. plecostomus, mainly due to the fact that a complementary description of the species by bleeker (1864: 7) was considered to be the original description. however, Ferraris (2007: 259) rightly pointed out the correct date of availability of the name: Plecostomus brasiliensis was described by bleeker in 1862 (in bleeker 1862-1863: 2). This original description was solely based on the description of Hypostomus plecostomus val. (not linnaeus), and no type locality was stated. however, valenciennes in Cuvier and valenciennes (1840) explicitly based its description on “des individus de douze à quinze pouces, pris par feu M. Plée dans la lagune de Maracaïbo” (specimens of twelve to ifteen inches, collected by the late M. Plée in the Maracaïbo lagoon), thus on specimens collected in Maracaïbo lake in Western venezuela. These specimens, housed in Mnhn, are the syntypes of Plecostomus brasiliensis bleeker, 1862. Designation of rMnh 3102, a specimen from Suriname that was used in bleeker' complementary description, as lectotype by boeseman (1968: 38) is therefore invalid. The taxonomic status of the species, that is not part of the Guianese fauna, has now to be revised. dISCuSSIOn Hypostomus species of the Guianas initially, 21 species of Hypostomus were recorded in fresh and brackish waters of the Guianas from the oyapock to the essequibo rivers following boeseman (1968, 1969), le bail et al. (2000), Weber (2003), Ferraris (2007), and vari et al. (2009). The results of the present study sustain the validity of 15 species, of which four are regarded as doubtful and still deserve further investigations. Two additional species previously placed in Squaliforma isbrücker & Michels, 2001 (in isbrücker et al., 2001; Weber, 2003): Squaliforma squalina (Jardine in Schomburgk, 1841) and S. tenuis (boeseman, 1968) are listed as Hypostomus by armbruster Cybium 2012, 36(1) (2004) and vari et al. (2009). however, based on molecular evidences that Squaliforma constitutes the sister genus of Aphanotolurus (see Montoya-Burgos, 2003), we conirm their placement in the former. Squaliforma representatives are indeed known from the Magdalena, amazon and essequibo river basins, with S. squalina in branco, negro and essequibo rivers, and doubtfully from venezuela. Squaliforma tenuis description was actually based on a unique specimen said to be from Paramaribo surroundings. With its elongated body and deeply forked caudal in, inclusion of the specimen in Squaliforma makes no doubt, however no congeneric specimen was ever found again in Suriname despite the intense collecting effort. The type locality appears thus highly doubtful, and Squaliforma tenuis should better not be regarded as a Surinamese species. Two of the 15 Guianese species of Hypostomus belong to the Hypostomus cochliodon group (armbruster, 2003), or formerly the genus Cochliodon Kner, 1854 which was placed in synonymy of Hypostomus by Weber and Montoya-burgos (2002): H. taphorni (lilyestrom, 1984) and H. macushi armbruster and de Souza, 2005. These species have large spoon-shaped teeth in contrast to the viliform teeth observed in most other Hypostomus species including H. plecostomus. Identity of Hypostomus plecostomus based on morphometric approach The morphometric model used in the present study demonstrates unambiguously that the linnean type series of H. plecostomus is heterogeneous and comprises two distinct species: H. plecostomus and H. watwata. The watwata group corresponded to different populations of the nominal species, H. watwata, which occurs in brackish waters along the Guianese coast. This species displayed signiicant morphometric structures as attested by the high statistical support observed. Hypostomus watwata is mostly characterized by its slender appearance with long caudal peduncle. in addition this species possesses numerous platelets just bordering the posterior portion of the supraoccipital and the pteroticsupracleithrum. This feature is perfectly visible on both possible paralectotypes of H. plecostomus but absent on the lectotype. The heterogeneity among the different populations of H. plecostomus (including those previously identified H. ventromaculatus and H. cf. ventromaculatus), and the lack of statistical support recorded within the plecostomus group made dificult the strict assignment of the lectotype of H. plecostomus to a given population. however, the low statistical support relects the close morphological relatedness of all these populations. The morphometric approach remained thus insuficient to solve accurately the identity of H. plecostomus in regards to H. ventromaculatus, and complementary data were essential. 219 hypostomus species from the Guianas Weber et al. Figure 11. - Geographic distribution of Hypostomus species in the Guianas, based on examined material (see list) and literature (eigenmann, 1912; boeseman, 1968, 1969; le bail et al., 2000; armbruster and de Souza, 2005). one symbol may represent several localities of collection. Genetic contribution to the delineation of species Species limits were eficiently assessed using the allozyme electrophoresis approach. in the Hypostomus plecostomus group, the only significant genetic differentiations were found between the population of Kourou river and those from oyapock and Kaw rivers. While all three populations were initially identiied as H. cf. ventromaculatus, no difference was found between all other populations of the H. plecostomus group, including H. ventromaculatus, H. cf. ventromaculatus and H. plecostomus. in addition, all H. plecostomus group populations are distinguished from other Hypostomus species with fixed allelic differences, which constitute the absolute isozyme characters to distinguish species according to Davis and nixon (1992) and Davis (1996). They were thus regarded as a single species, Hypostomus plecostomus, of which H. ventromaculatus is synonym. like Hypostomus plecostomus, H. watwata and H. gymnorhynchus were recognized based on fixed allelic differences. Genetic isolation between H. plecostomus and H. wat220 wata is easily demonstrated as they were found in sympatry (ouanary on the oyapock river, Mahury river, Coswine river in Maroni basin, and Paulus Creek in Suriname basin), implying their biological species status. Hypostomus gymnorhynchus (and its synonym H. tapanahoniensis) was not found in sympatry of the two other species, but its high divergence (ixed allelic differences at four loci), in addition to morphological difference, makes no doubt about its validity. in addition to the diagnostic loci allowing unambiguous identification, each of these species was shown to have unique alleles in our sampling, in particular H. gymnorhynchus that showed six alleles not found in other species. With 36% of ixed allelic differences between H. plecostomus and H. watwata, 45% between H. watwata and H. gymnorhynchus, and 55% between H. plecostomus and H. gymnorhynchus, genetic divergence between these species is very high considering that populations differing at only 10% of their loci usually represent distinct species (avise and aquadro, 1982; richardson et al., 1986). nei’s genetic distances Cybium 2012, 36(1) Weber et al. values, varying from 0.558 between H. plecostomus and H. watwata to 0.933 between H. plecostomus and H. gymnorhynchus, can be compared to those reported for loricariids, that range from 0.112 up to 1.304 for other species of Hypostomus (de Paiva et al., 2005; Zawadzki et al., 2005; renesto et al., 2007), and equal to 0.51 for two species of Ancistrus (Fisch-Muller et al., 2001) and to 0.314 for two species of Neoplecostomus (Zawadzki et al., 2004). Genetic diversity of H. plecostomus, H. gymnorhynchus and H. watwata Hypostomus plecostomus analysed for allozymes ranged from the oyapock river in French Guiana to the Saramacca river in Suriname. Genetic diversity between the different populations exists but it is restricted, with statistical differences found only for two cases, and with nei’s genetic distances varying from null to 0.031. Populations of Kourou and Kaw rivers are the most divergent. They are also the most polymorphic according to P0.95 criterion (30% of polymorphic loci), and to expected heterozygosity (respectively, 0.127 and 0.109). at the species level heterozygosity (he = 0.091) is higher than the mean value indicated by Ward (1992) for ish of several orders (0.051). However, heterozygosity was indeed found very variable for numerous species of Hypostomus from eastern brazil (revised in de Paiva et al., 2005; Zawadzki et al., 2005; renesto et al., 2007), with values ranging from 0.000 to 0.144. it was shown that heterozygosity not only depends on population size and on taxonomic groups, but also on enzymes whose polymorphic levels diverge (Ward et al., 1992; avise, 1994). based on the same enzyme systems studied here, values ranging from 0.040 to 0.151 were found for ten species of Ancistrus (Fisch-Muller, 1999), the average value (0.084) being similar to the one obtained for H. plecostomus. Hypostomus gymnorhynchus also showed differentiated populations according to nei’s genetic distances, due essentially to a ixed allelic difference between Approuague and Mahury river’s populations on one hand and Sinnamary and Maroni on the other; however this cannot be statistically conirmed. In addition, the four populations studied are the less polymorph of the entire Hypostomus sampling, two of them (approuague and Sinnamary) showing no genetic variability at all. Such results generally indicate small population size, restricted geographical distribution and reduced gene low between populations (Kimura and Ohta, 1971; nei and Graur, 1984; hamrick and Godt, 1996). on the contrary Hypostomus watwata is characterised by the highest genetic diversity according to all parameters examined, at both population and species levels. The high value of expected heterozygosity found in H. watwata (0.137) is similar to the highest value observed for Hypostomus and for Ancistrus species (see above). it clearly corresponds to species occupying a broad ecological niche (nei and Graur, Cybium 2012, 36(1) hypostomus species from the Guianas 1984; Avise, 1994), and might ind an explanation in the fact that H. watwata is a brackish-water species able to disperse widely between lower parts of rivers through coastal waters. This result is perfectly congruent with the high morphological structure recovered within H. watwata. Hypostomus species zonation in the Guianas Hypostomus representatives are found from headwaters to estuaries; however none of the species is distributed all along the course of rivers. Three main zones can be recognized according to boujard and rojas-beltran (1988): the stream area of the high lands, the rivers and swamps area of coastal plain, and the estuarial and littoral area. The irst zone counts the largest part of freshwaters, from headwaters to upstream of the lower falls. This biotope is mostly characterized by fast lowing waters, and a rocky substrate consisting in boulders, stones, gravels, and sand. The exposed wet rocks are also often covered by the Podostemaceae Mourera luviatilis and Apinagia richardiana. among other species inhabiting these rheophilic biotopes, Hypostomus corantijni, H. crassicauda, H. gymnorhynchus, H. macushi, H. micromaculatus, and H. paucimaculatus are mostly characteristic from that zone, as well as H. nematopterus. Known only from two type specimens collected about 40 years ago in the oyapock river basin, the latter was regarded as a form of H. gymnorhynchus by le bail et al. (2000), but was later considered valid (Weber, 2003; Ferraris, 2007; vari et al., 2009). This species differs from all Guianese Hypostomus in having an astonishing elongation of the dorsal-in spine, and to a less extent of the caudal-in spines. The original illustration of the holotype shows a remarkable bony callus on the dorsal-in spine, showing that it was once broken and had grown again, possibly longer than originally. nevertheless, not only the paratype (examined) has the same elongated dorsal-in spine, but also an additional specimen that we recently collected near the type locality conirms the validity of H. nematopterus. The second zone comprises the areas from the lower falls to the estuary. This area is still under the tide inluence and is characterized by quiet waters and a muddy substrate. riparian vegetation often consists in the aracea Montrichardia arborescens. Hypostomus plecostomus occupies this area. The third zone corresponds to the estuarial and littoral area. This area is characterized by the direct inluence of the sea with different levels of salinity. Hypostomus watwata is the only Guianese species of Hypostomus that is a permanent resident of this zone. Acknowledgments. - We are grateful to Sven Kullander and erik Åhlander (nrM), Philippe Keith, romain Causse and Patrice Pruvost (Mnhn), Martien van oijen and Koos van egmond (rMnh), ronald vonk and hielke Praagman (ZMa), and Melanie Stiassny (aMnh) for the welcome in collections and loan of material; Juan i. Montoya-burgos (uniGe), Philippe Gaucher (CnrS Guyane), 221 hypostomus species from the Guianas régis vigouroux and Philippe Cerdan (hydreco Guyane), Michel Jégu and bernard de Mérona (irD), Pierre-Yves le bail (inra), François Meunier and Yves Fermon (Mnhn), Claudie bidaud (Terres de Guyane), richard Commergnat (French Guiana), Jan Mol (CeloS), Paul ouboter (nZCS), Kenneth Wan Tong You (Paramaribo), alexandre Fort, Pedro hollanda Carvalho, and alain Merguin (MhnG), Calvin bernard (CSbD), vincent Piron and Colin niel (Parc de la Guyane), the north rupununi District Development board (nrDDb), the association arataï, and the iwokrama organization for their ield and logistic assistance; the G. and A. Claraz Foundation for their inancial support for the missions in Suriname in 2001, 2005, 2007 and 2008, and in French Guiana in 2006; the académie Suisse des Sciences naturelles (Scnat) for their inancial support for the missions in Guyana 2004 and French Guiana 2006; alf Stalsberg (Cichlid Power) for the gift of specimens from Suriname; andré huser (Sallmann-Fehr aG) for the gift of gill nets for the mission in Suriname in 2005 and 2007; the Guyana environmental Protection agency, and Ministry of amerindian affairs; the French Guiana Diren, and Préfecture; and the Surinamese Ministry of agriculture, animal husbandry and Fisheries for the necessary authorizations and collecting permits. The pictures and igures were inalized by Philippe Wagneur and Florence Marteau (MhnG). We are grateful to Stéphane Dray (CnrS) for statistical assistance, to volker Mahnert (MhnG) for help with nomenclature, and to Carl Ferraris (Portland), Pedro hollanda Carvalho (MZuSP), and Claudio Zawadszki (ueM-nupélia) for their constructive comments reviewing the manuscript. mATeRIAL eXAmIned. - Material is listed by species (present and former identiications, including new synonymies), and within each species by countries and river basins from east to west. abbreviations for populations included in morphometric analyses are in brackets. Individual ield numbers indicate specimens included in allozyme analysis. Hypostomus plecostomus (Linnaeus, 1758) Hypostomus cf. ventromaculatus sensu Le Bail et al., 2000 Guyane française: Oyapock River Basin (voya): oyapock river, ouanary village, close to the landing stage: MhnG 2650.049 (28 ex.), GF99-146-157, 165-172. ouanary river: MhnG 2661.005 (6 ex); MhnG 2661.005 (5 ex). oyapock river: MhnG 2605.056 (1 ex.). Taparabou islands: MhnG 2724.080 (1ex.). Mouth of Taparabou Creek, tributary of oyapock river: MhnG 2724.078 (1ex.). Approuague River Basin: approuague river, régina: MhnG 2713.028 (10 ex.); Guisanbourg: MhnG 2724.081 (1ex.). Kaw River Basin (vKaw): Kaw river, in the vicinity of Kaw: MhnG 1223.019 (1 ex.); MhnG 1223.020 (1 ex.); MhnG 2645.012 (1 ex.), GF99-010; MhnG 2645.013 (1 ex.), GF99-014; MhnG 2645.014 (8 ex.), GF99-013, 016, 020025; MhnG 2645.015 (1 ex.), GF99-018; MhnG 2645.016 (4 ex.), GF99-030, 032-033 ; MhnG 2713.014 (8 ex.); MhnG 2713.015 (1 ex.); MhnG 2713.016 (4 ex.); MhnG 2724.079 (1ex.); MhnG 2724.085 (2 ex.). Mahury River Basin (vCay): Mahury river, downstream bridge of road to roura in front of Degrad des Cannes: MhnG 2724.076 (4 ex.), GF99-140-141, 143-144; Comté river, Saut bief (fish farm): MhnG 2724.077 (6 ex.), GF99-071-076. rivière des Cascades, tributary of Cayenne river: MhnG 2724.067 (3 ex.). Kourou River Basin (vKou): Kourou river, Soukoumou: 222 Weber et al. MhnG 2595.039 (2 ex.), GF98-001, 010; MhnG 2595.042 (8 ex.), GF98-002, 006, 008-009, MhnG 2595.043 (3 ex.), GF 98-030032. Kourou river, near Kourou: MhnG 2595.028 (2 ex); MhnG 2595.044 (1 ex), GF98-064; MhnG 2596.009 (1 ex.). Crique Couy, left bank tributary of the Kourou river: MhnG 2594.094(1 ex.). Crique des Pères: MhnG 2713.026 (1ex.). Malmanouri Creek Basin: Malmanoury Creek: MhnG 2713.025 (1 ex.). Hypostomus ventromaculatus Boeseman, 1968 Guyane française: Iracoubo River Basin: iracoubo river, Police camp “Carbet de la Gendarmerie” MhnG 2695.069 (1 ex.); iracoubo river, Degrad Florient; MhnG 2604.097 (10 ex.). Maroni River Basin (VMar): Mana River, rice ields: MHNG 2724.082. (1 ex.); MhnG 2724.083 (1 ex.); MhnG 2724.084 (2 ex.); Mana river, awala Yalimapo: MhnG 2596.008 (1 ex.); Tupo Creek, tributary of Coswine Creek, right bank tributary of lower Maroni river, near ayawande: MhnG 2595.034 (2 ex.), GF98-158; MhnG 2595.036 (3ex.), GF98-151, 161; MhnG 2595.037 (2 ex.); MhnG 2595.038 (10 ex.), GF98-147, 153, 156, 163, 165-167, 170, 182; MhnG 2595.041 (2 ex.), GF98-154, 164. Suriname: Suriname River Basin (vSur): Paramaribo, Central Market: MhnG 2621.022 (16 ex.), Su01-010-020. Acipenser plecostomus Linnaeus, 1758 Suriname: Suriname River Basin (Psyn1): Surinami (=Suriname river), nrM 32 (1/3 ex, lectotype of H. plecostomus). Hypostomus plecostomus sensu Boeseman, 1968 Suriname: Maroni River Basin: Maroni river, close to James between albina and apatou: MhnG 2680.090 (1 ex.); albina: MhnG 2604.019 (2 ex.). Commewijne river basin (PMap): Mapana Creek, tributary of Commewijne river: MhnG 2708.046 (4 ex.); MhnG 2708.047 (8 ex.). Suriname River Basin (PPaK): Paulus Creek at mouth, right bank tributary of lower Suriname river: MhnG 2621.023 (17 ex.), Su01- 021-037). Klass Creek, brokopondo: MhnG 2708.042 (1ex.). Saramacca River basin: Mindrineti Creek, tributary of Saramacca river, near mouth of Maikaboeka Creek, Gros rosebel Mining, MhnG 2621.032 (3ex.), Su01-046, 048-049. Coppename River Basin: unnamed tributary of Coppename river near raleigh vallen: MhnG 2690.019 (1 ex.). Coppename river, Witagron: MhnG 2724.075 (2 ex.). Corantijn River Basin: lower Corantijn river at Matapi: MhnG 2671.065 (16 ex.). Guyana: Berbice River Basin: berbice river, Dubulay ranch: MhnG 2651.066 (2 ex.). Cambo Cambo Creek, small tributary of Wuruni river, Wuruni: MhnG 2651.077 (1 ex.). Hypostomus hemiurus AMNH catalogue identiication Guyana: essequibo river basin (heme): essequibo river, Kartabo: aMnh 220361 (4 ex.). Hypostomus watwata Hancock, 1828 Guyane française: Oyapock River Basin (Woya): ouanary river, tributary of lower oyapock river, ouanary: MhnG 2645.009 (4 ex.), GF99-161-164. Mahury River Basin: Mahury river: Mnhn 1903-0055; Mnhn 1904-404-405. Mahury river, downstream bridge on road to roura, in front of Degrad des Cannes: MhnG 2645.008 (1 ex.), GF99-103. Cayenne River Basin Cybium 2012, 36(1) Weber et al. (WCay): Cayenne river, Cayenne: Mnhn a.8919 (1 ex.); Mnhn a.9450 (1 ex.); Mnhn a.9451 (1 ex.). along sea shore beyond Pointe Macouria, Cayenne river estuary: MhnG 2650.050 (19 ex.), GF00-001-002, 004-011. Kourou River Basin (WKou): Kourou river, near Kourou, under bridge of rn1: MhnG 2595.029 (1 ex.), GF98-063; MhnG 2595.030 (9 ex.). Sinnamary River Basin (WSin): Sinnamary river, downstream Sinnamary: MhnG 2595.031 (9 ex.), GF98-098, 101, 103-104, 116-117. Mana River Basin (WMar): Mana river, awala Yalimapo: MhnG 2595.024 (1 ex.), GF98-195. MhnG 2595.027 (4 ex.), GF98-184-186; MhnG 2595.033 (2 ex.). Maroni River Basin (WMar): Coswine Creek, near ayawande: MhnG 2595.025 (5 ex.), GF98-146, 150, 152, 159, 162; MhnG 2595.026 (2 ex.), GF98-173, 183; MhnG 2595.032 (2 ex.), GF98-144, 155; MhnG 2595.045 (1 ex.), GF98160; MhnG 2595.046 (1 ex.). Suriname: Suriname River Basin (WPuK): Pulp Creek, Paramaribo, just downstream of canal lock to Suriname river: MhnG 2621.021 (9 ex.): Su01-001-009. Acipenser plecostomus Linnaeus, 1758 Suriname: Suriname River Basin (Psyn2, Psyn3): Surinami (=Suriname river), nrM 32 (2/3 ex, possible paralectotypes of H. plecostomus). Hypostomus hemiurus (eigenmann, 1912) Guyana: essequibo river basin: essequibo river, Kurupukari Cross: MhnG 2651.070 (1 ex.); MhnG 2650.086 (6 ex.). Siparuni river downstream Georges Creek: MhnG 2650.084 (4 ex.). rupununi river, Pregogo: MhnG 2651.018 (4 ex.); Dadanawa ranch: MhnG 2651.037 (4 ex.). arakwai river, tributary of rupununi river: MhnG 2651.086 (2 ex.). venezuela: Cuiuni River Basin (hemC): Parapapoy river, tributary of lower Cuiuni river: uF 96192 (1 ex.). Hypostomus gymnorhynchus (norman, 1926) Guyane française: Oyapock River Basin: oyapock river, Saut Fourmi, upstream of Saint Georges: MhnG 2713.023 (1 ex.). Approuague River Basin (Gapp): Crique ipoussing: bMnh 1926.3.2.74 (holotype). arataye river: Mnhn i989-0048 (1 ex.). arataye river, Saut Japigny: Mnhn i989-0037 (1 ex.). approuague river, rapids of Saut Mapaou: MhnG 2621.098 (9ex.), Su01-143-144, 146, 151-155, 160. Mahury River Basin (GCom): Crique bagot, tributary of Comté river: 2671.005 (1 ex.). Comté river, Saut bief: MhnG 2645.010 (7 ex.), GF99-062-068. Comté river, lysis camp: MhnG 2671.004 (1 ex.). Crique Grillon, tributary of orapu river, onF camp: MhnG 2682.060 (1 ex.). Sinnamary River Basin (GSin): Sinnamary river, upstream Sinnamary: MhnG 2595.047 (2 ex.), GF98-091-092, MhnG 2595.048 (4 ex.), GF98-099-100,102. Sinnamary river, Saut vata: MhnG 2724.091 (3 ex.); Saut Parasol: irD uncat. (1 ex.). Saut Takari Tanté: MhnG 2724.092 (1 ex.). Sinnamary river, Pointe Combi: MhnG 2687.039 (1 ex.), GF98-066; MhnG 2595.035 (1 ex.). Mana River Basin (GMan): Mana river, Saut Fracas: Mnhn 1998-1714 (9 ex.); MhnG 2724.090 (2 ex.); Saut Dalles: MhnG 2713.027 (4 Cybium 2012, 36(1) hypostomus species from the Guianas ex); Saut ananas: MhnG 2724.087. (2 ex.); Saut Capiaie: MhnG 2724.088. (2 ex.); lezard Creek, Citron: MhnG 2724.089. Hypostomus tapanahoniensis Boeseman, 1969 Guyane française: Maroni River Basin (ThMar): Maroni river, Saut Gostou: MhnG 2724.093 (1 ex.). voltaire Creek, downstream voltaire Falls, voltaire Camp: MhnG 2683.025 (2 ex.); voltaire Falls: MhnG 2683.040 (1 ex). Grand inini river: MhnG 2724.096 (10 ex.); blax island: MhnG 2593.078 (3 ex.); MhnG 2593.080-082 (3 ex.); Mnhn uncat. (2ex.); near Maripasoula: Mnhn uncat. (5 ex.); Saut lobo, Petit and Grand inini conluence: MHNG 2724.094 (2 ex.); Saut Nicole: MHNG 2724.095 (2 ex.). litani river, Saut Tetombé: Mnhn 1998.1591 (2ex.); in the vicinity of antecume Pata: Mnhn 2000-5759 (5 ex.); 20005736 (2 ex.); Mnhn 2000-5736 (2 ex.) ; Mnhn 2000-5745 (1 ex.); Mnhn 2000-5731 (1 ex.). Marouini river, in the vicinity of antecume Pata: Mnhn 2000-5778 (1ex.): MhnG 2725.001 (2 ex.); MhnG 2725.002 (2 ex.); MhnG 2725.003 (1 ex.), GF00089; MhnG 2725.004 (3 ex.); MhnG 2725.005 (1 ex.); MhnG uncat. (5 ex.), GF00 106-108. Saut Tula lapata: MhnG 2724.098 (1 ex.), GF00-070; MhnG 2724.099 (3 ex.), GF00-075,078-079. aweimë enï: MhnG 2724.100 (3 ex.), GF00-091, 093, 095. Tampoc river: Mnhn 1998-1785 (9 ex.), Mnhn 1998-1786 (7 ex.); Saut Kwata: MhnG 2724.097 (1ex.); Saut Pierkuru: Mnhn 20005751 (15 ex.). Suriname: Maroni River Basin: Tapanahoni river: Kumaru Konde Sula: MhnG 2717.011 (9 ex.); MhnG 2717.012 (10 ex.); Tapanahoni river, Palomeu: MhnG 2717.020 (2 ex.); Palomeu river, tributary of Tapanahoni river, Wayu Camp: MhnG 2717.039 (9 ex.). Hypostomus surinamensis Boeseman, 1968 Suriname: Suriname River Basin: Suriname river at Jemongo: MhnG 2724.005 (1 ex.). awaradam, rapids in Gran rio river: MhnG 2674.009 (3 ex.). Gran rio river, Kossindo: MhnG 2674.027 (2 ex.); MhnG 2674.044 (1 ex.). Gran rio, Cajana Creek, left bank of Gran rio river; MhnG 2673.045 (1 ex.); MhnG 2673.047 (6 ex.); MhnG 2673.024 (12 ex.). Gran rio river, assigon: MhnG 2673.077 (2 ex.). Hypostomus corantijni Boeseman, 1968 Suriname: Corantijn River Basin: oema Creek, left bank tributary of Corantijn river: MhnG 2672.022 (2 ex.). Hypostomus sipaliwinii Boeseman, 1968 Suriname: Corantijn River Basin: Sipaliwini river: MhnG 2708.017 (2 ex.). Sipaliwini river landing stage of Sipaliwini villlage: MhnG 2708.022 (3 ex.). Forest creek, Witoto ecu, tributary of Sipaliwini river: MhnG 2708.025 (2 ex.). Sipaliwini river 15 minutes by boat downstream of Sipaliwini: MhnG 2708.034 (4 ex.). Manicouni river, upstream of mouth in Sipaliwini river: MhnG 2708.035 (3 ex.). Paikali river rapids: MhnG 2708.016 (6 ex.). Hypostomus nickeriensis Boeseman, 1969 Suriname: Nickerie River Basin: nickerie river, blanche Marie Falls West, pool at base of fall: MhnG 2621.068(3 ex.), Su01-114116, 124, 126, 133. 223 hypostomus species from the Guianas Hypostomus crassicauda Boeseman, 1968 Suriname: Corantijn River Basin: Corantijn river, Wonotobo Falls: MhnG 2672.063 (2 ex.). Sipaliwini river, Paikali river rapids: MhnG 2708.023 (12 ex.). Sipaliwini river, Sipaliwini village, landing stage: MhnG 2708.024 (4 ex.); MhnG 2708.026 (1 ex.). Sipaliwini river, rapids: MhnG 2708.033 (1 ex.), MhnG 2708.031 (1 ex.). unnamed creek, left bank tributary of Sipaliwini river, about half an hour downstream of Sipaliwini village: MhnG 2708.037 (2 ex.). Hypostomus micromaculatus Boeseman, 1968 Suriname: Suriname River Basin: Gran rio river, Kossindo: MhnG 2673.066 (1ex.), MhnG 2674.028 (1 ex.). Cajana Creek, left bank tributary of Gran rio river: MhnG 2673.025 (2 ex.). Gran rio river, assigon: MhnG 2673.076 (2 ex.). awaradam, rapids in Gran rio river: MhnG 2674.012 (4ex.). Hypostomus paucimaculatus Boeseman, 1968 Suriname: Suriname River Basin: Gran rio river, Kossindo: MhnG 2673.067 (1 ex.); MhnG 2674.043 (1 ex.). Hypostomus coppenamensis Boeseman, 1969 Suriname: Coppename River Basin: Coppename river, Witagron: MhnG 2604.010 (1 ex.). Hypostomus nematopterus Isbrücker & nijssen, 1984 Guyane Française: Oyapock River Basin: Trois Sauts rapids of oyapock river, 2º15’ n, 52º53’ W, ZMa 107804 (paratype). oyapock river, Camopi: MhnG 2682.046 (1 ex.). Hypostomus taphorni (Lilyestrom, 1984) Suriname: Corantijn River Basin: Corantijn river: Matapi: MhnG 2671.064 (7 ex.); Downstream Wonotobo Falls: MhnG 2671.075 (2 ex.); MhnG 2672.065 (1 ex.) Hypostomus macushi Armbruster and de Souza, 2005 Weber et al. arMbruSTer J.W., 2004. - Phylogenetic relationships of the suckermouth armoured catishes (Loricariidae) with emphasis on the hypostominae and ancistrinae. Zool. J. Linn. Soc., 141: 1-80. arMbruSTer J.W. & De SouZa l.S., 2005. - Hypostomus macushi, a new species of the Hypostomus cochliodon group (Siluriformes: loricariidae) from Guyana. Zootaxa, 920: 1-12. aviSe J.C., 1994. - Molecular Markers, natural history and evolution. 510 p. new York: Chapman & hall. aviSe J.C. & aQuaDro C.F., 1982. - a comparative summary of genetic distances in the vertebrates: Patterns and correlations. Evol. Biol., 15: 151-188. benJaMini Y. & hoChberG Y., 1995. - Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B, 57: 289-300. benSon a.J., JaCono C.C., Fuller P.l., MCKerCher e.r. & riCherSon M.M., 2004. - Summary report of nonindigenous aquatic Species in u.S. Fish and Wildlife Service region 5. 142 p. arlington, va, uSa: u.S. Fish and Wildlife Service. birinDelli J.l.o., ZanaTa a.M. & liMa F.C.T., 2007. Hypostomus chrysostiktos, a new species of armored catfish (Siluriformes: loricariidae) from rio Paraguaçu, bahia State, brazil. Neotrop. Ichthyol., 5(3): 271-278. bleeKer P., 1862-1863. - atlas ichtyologique des indes orientales néêrlandaises, publié sous les auspices du gouvernement colonial néêrlandais. Tome ii : Siluroïdes, Characoïdes et hétérobranchoïdes. 112 p. amsterdam: Fr. Müller. bleeKer P., 1864. - Description des espèces de Silures du Suriname conservées aux Musées de leide et amsterdam. Mémoire publié par la Société hollandaise des Sciences à harlem. 104 p. harlem: les héritiers loosjes. boeSeMan M., 1968. - The genus Hypostomus lacepède, 1803, and its Surinam representatives (Siluriformes, loricariidae). Zool. Verh., Leiden, 99: 1-89. boeSeMan M., 1969. - additional new species of Hypostomus lacepède, 1803, from Surinam; with remarks on the apparent “gymnorhynchus-complex” (Siluriformes, loricariidae). Beaufortia, 16: 119-136. bouJarD T. & roJaS-belTran r., 1988. - Zonation longitudinale du peuplement ichtyque du leuve Sinnamary (Guyane française). Rev. Hydrobiol. Trop., 21(1): 47-61. buTh D.G., 1990. - Chapter 1. Genetic principles and the interpretation of electrophoretic data. In: electrophoretic and isoelectric Focusing Techniques in Fisheries Management (Whitmore D.h., ed.), pp. 1-21. boca raton, California: CrC Press. Guyana: Essequibo River Basin: essequibo river, Kurukupari Cross: MhnG 2651.073 (1 ex.). Amazon drainage, Branco River Basin: Sarawab river, tributary of Takutu river, Sawarab bridge: MhnG 2651.025 (3 ex.). Cuvier G. & valenCienneS a., 1840. - histoire naturelle des poissons. Tome quinzième. Siluroïdes. 540 p. Paris: Ch. Pitois. ReFeRenCeS DaviS J.i., 1996. - Phylogenetics, molecular variation, and species concepts. Bioscience, 46(7): 502-511. allibone r.M., CroWl T.a., holMeS J.M., KinG T.M., MCDoWall r.M., ToWSenD C.r. & WalliS G.P., 1996. - isozyme analysis of Galaxias species (Teleostei: Galaxiidae) from the Taieri river, South island, new Zealand: a species complex revealed. Biol. J. Linn. Soc., 57(2): 107-127. anDerSon D.h., bouSQuin S.G., WilliaMS G.e. & COLANGELO D.J. (eds), 2005. - Deining Success: Expectations for restoration of the Kissimmee river. South Florida Water Management. District, West Palm beach, Florida, uSa. 141 p.: Technical Publication era, 433. arMbruSTer J.W., 2003. - The species of the Hypostomus cochliodon group (Siluriformes: loricariidae). Zootaxa, 249: 1-60. 224 DaviS J.i. & niXon K.C., 1992. - Populations, genetic variation, and the delimitation of phylogenetic species. Syst. Biol., 41(4): 421-435. De Paiva S., reneSTo e. & ZaWaDZKi C.h., 2005. - Genetic variability of Hypostomus (Teleostei, loricariidae) from the ribeirão Maringá, a stream of the upper rio Paraná basin, brazil. Genetic. Mol. Biol., 28 (3): 370-375. DraY S. & DuFour a.b., 2007. - The ade4 package: implementing the duality diagram for ecologists. J. Stat. Soft, 22 (4): 1-20. eFron b., 1979. - bootstrap methods: another look at the jackknife. Ann. Statist., 7: 1-26. Cybium 2012, 36(1) Weber et al. eiGenMann C.h., 1912. - The freshwater fishes of british Guiana, including a study of the ecological grouping of species and the relation of the fauna of the plateau to that of the lowland. Mem. Carneg. Mus., 5: 1-578. EIGENMANN C.H., 1922. - The ishes of western South America, part 1. The fresh-water ishes of Northwestern South America, including Columbia, Panama, and the Paciic slopes of Ecuador and Peru, together with an appendix upon the ishes of the Rio Meta in Columbia. Mem. Carneg. Mus., 9(1): 1-346. eMberTon K.C., 1995. - Cryptic, genetically extremely divergent, polytypic, convergent, and polymorphic taxa in Madagascan Tropidophora (Gastropoda: Pomatiasidae). Biol. J. Linn. Soc., 55: 183-208. FELSENSTEIN J., 1985. - Conidence limits on phylogenies: an approach using the bootstrap. Evolution, 39: 783-791. FernanDeS M.n. & Perna S.a., 1995. - internal morphology of the gill of a loricariid ish, Hypostomus plecostomus: arterioarterial vasculature and muscle organization. Can. J. Fish. Aquat. Sci., 73: 2259-2265. FERNHOLM B. & WHEELER A., 1983. - Linnean ish specimens in the Swedish Museum of natural history, Stockholm. Zool. J. Linn. Soc., 78(3): 199-286. FERRARIS C.J. Jr, 2007. - Checklist of catishes, recent and fossil (osteichthyes: Siluriformes), and catalogue of siluriform primary types. Zootaxa, 1418: 1-628. FiSCh-Muller S., 1999. - Systématique du genre Ancistrus Kner (Teleostei, loricariidae): approches morphologique et génétique. unpubl. PhD thesis, 300 p. univ. of Geneva, Switzerland. FiSCh-Muller S., MaZZoni r. & Weber C., 2001. - Genetic and morphological evidences for two new sibling species of Ancistrus (Siluriformes: loricariidae) in upper rio Tocantins drainage, brazil. Ichthyol. Explor. Freshw., 12(4): 289-304. FOWLER H.W., 1915. - Notes on Nematognathous ishes. Proc. Acad. Nat. Sci. Philad., 67: 203-243. FOWLER H.W., 1941. - A collection of fresh-water ishes obtained in eastern brazil by Dr. rodolph von ihering. Proc. Acad. Nat. Sci. Philad., 93: 123-199. FoWler h.W., 1943. - a collection of fresh-water fishes from Colombia, obtained chiely by Brother Nicéforo Maria. Proc. Acad. Nat. Sci. Philad., 95: 223-266. FoWler h.W., 1954. - os peixes de agua doce do brasil. 4a entraga. Archos Zool. Est. Sao Paulo, 9(2): 1-400. FriCKe r. & eSChMeYer W.n., 2010. - a guide to Fish Collections in the Catalogue of Fishes database. on-line version of 15 January 2010. GraY J.e., 1854. - Catalogue of Fish Collected and Described by laurence Theodore Gronow, now in the british Museum Trustees british Museum (natural history), london. 196 p. london: british Museum. hypostomus species from the Guianas halDane J.b.S., 1954. - an exact test for randomness of mating. J. Genet., 52: 631-635. haMriCK J.l. & GoDT M.J.W., 1996. - Conservation genetics of endemic plant species. In: Conservation Genetics: Case histories from nature (avise J.C. & hamrick J.l., eds), pp. 281304. new-York: Chapman & hall. hilliS D.M., MoriTZ C. & Mable b.K. (eds), 1996. - Molecular Systematics. 2nd edit., 655 p. Sunderland, Mass., uSa: Sinauer associates, inc. 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. Neotrop. Ichthyol., 8 (1): 39-48. iSbrÜCKer i.J.h. & niJSSen h., 1984. - Hypostomus nematoperus, a new species of mailed catish from the Oyapock river system, French Guiana (Pisces, Siluriformes, loricariidae). Bull. Zool. Mus. Amsterdam, 10(2): 9-14. iSbrÜCKer i.J.h., SeiDel i., MiChelS J.P., SChraMl e. & Werner a., 2001. - Diagnose vierzehn neuer Gatungen der Familie Loricariidae Rainesque, 1815 (Teleostei, Ostariophysi). In: harnischwelse 2 (Stawikowski r., ed.), pp. 17-24. Stuttgart: Die aquarien- und Terrarienzeitschrift, Sonderheft, eugen ulmer. JereP F.C., ShibaTTa o.a. & ZaWaDZKi C.h., 2007. - a new species of Hypostomus lacepède, 1803 (Siluriformes: loricariidae) from the upper rio Paraná basin, Southern brazil. Neotrop. Ichthyol., 5(4): 435-442. KiMura M. & ohTa T., 1971. - Theoretical aspects of Population Genetics. 172 p. new York: Princeton univ. Press. Kner r., 1854. - Die hypostomiden. Zweite hauptgruppe der Familie der Panzerische (Loricata vel Goniodontes). Denkschr. Akad. Wiss., Wien, Math.-Nat. Kl., 7: 251-286, 255 pls. KriTSKY D.C. & boeGer W.a., 1991. - neotropical Monogenea 16. new species of oviparous Gyrodactylidea with proposal of Nothogyrodactylus gen. n. (ooegyrodactylidae). J. Helm. Soc. Wash., 58: 7-15. laCePèDe b.G.e., 1803. - histoire naturelle des Poissons. lxviii + 803 p. + index, 821 pls. Paris: Plassan. larSon a., 1989. - The relationship between speciation and morphological evolution. In: Speciation and its Consequences (otte D. & endler J.a., eds), pp. 579-598. Sunderland, Massachussetts: Sinauer ass. le bail P.Y., KeiTh P. & PlanQueTTe P., 2000. - atlas des Poissons d’eau douce de Guyane. Tome 2, fascicule ii: Siluriformes. 307 p. Patrimoines Nat., 43(2). Paris: Mnhn/SPn. le bail P.Y., Covain r., JÉGu M., FiSCh-Muller S., viGourouX r. & KeiTh P., 2012. - updated checklist of the freshwater and estuarine fishes of French Guiana. Cybium, 36(1): 293-319. leSSioS h.a., allen G.r., WellonGTon G.M. & berMinGhaM e., 1995. - Genetic and morphological evidence that the eastern Pacific damselfish Abudefduf declivifrons is distinct from A. concolor (Pomacentridae). Copeia, 1995(2): 277-288. GronoviuS l.T., 1754. - Museum ichthyologicum, Sistens Piscium indigenorum & nonnullorum exoticorum, quorum Maximum pars in Museo laurenti Theodori Gronovii, J. u. D. adservantar, Descriptiones, ordine Systematico; accedunt nonnullorum exoticorum Piscium icones, aeri incisae. 70 p. lugduni batavorum: Theodorum haak. lilYeSTroM C.G., 1984. - Consideraciones sobre la taxonomía de las especies del genero Cochliodon heckel en venezuela (Pisces, loricariidae). Rev. Unellez Cienc. Tecnol., 2(2): 41-53. GuTiÉrreZ a.T. & reaSer J.K., 2005. - linkages between Development assistance and invasive alien Species in Freshwater Systems of Southeast asia. 87 p. Washington, DC: u.S. agency for international Development asia and near east bureau. linnaeuS C., 1754. - hans Maj:ts adolf Frideriks vàr allernàdigste Konungs naturalie Samling innehàllande Sällsynte och Främmande Djur, som bevaras pà Kongl. lust-slottet ulriksdahl beskrefne och afritade Samt pà nàdig befallning utgifne af Carl linnaeus. xxx + 96 + 97 p. holmiae. Cybium 2012, 36(1) 225 hypostomus species from the Guianas linnaeuS C., 1758. - Systema naturae Sive regna Tria naturae, Secundum Classes, ordines, Genera, Species, cum Characteribus, Differentiis, Synonymis, locis.Tomus i. ii + 824 p. holmiae: editio decima, reformata. MaTTiaS a.T., Perna S.a., Moron S.e., roDriGueS J.a.o. & FernanDeZ M.n., 1996. - Comparação entre a estrutura morfológica e a morfometria das brânquias de cascudo, Hypostomus regani e Hypostomus plecostomus (loricariidae) Anais VII Sem. Reg. Ecol.: 223-236. MaZZoni r., MoraeS M., reZenDe C.F. & MiranDa J.C., 2010. - alimentação e padrões das espécies de peixes de riacho do alto rio Tocantins, Goiás, brasil. Iheringia, 100(2): 162168. McCann J.a., arKin l.n. & WilliaMS J.D., 1996. - nonindigenous aquatic and Selected Terrestrial Species of Florida. Status, Pathways, Dates of introduction, range Distributions, and Significant ecological and economic effects. 301 p. Gainesville: univ. of Florida Press. Weber et al. REGAN C.T., 1904. - A monograph of the ishes of the family Loricariidae. Trans. Zool. Soc. Lond., 17(3): 191-350, pls 199-121. REGAN C.T., 1908a. - Description of a new loricariid ish of the genus Plecostomus from argentina. Ann. Mag. Nat. Hist., 2: 358 (ser. 358). REGAN C.T., 1908b. - Descriptions of new loricariid ishes from South america. Proc. Zool. Soc. Lond., 1907: 795-800, pls 747749. reGan C.T., 1912. - Descriptions of new species of the family loricariidae in the british Museum collection. Proc. Zool. Soc. Lond., 1912: 666-670, pls 675-677. reneSTo e., ZaWaDZKi C.h. & De Paiva S., 2007. - allozyme differentiation and relationships within Hypostomus lacepède, 1803 (osteichthyes: loricariidae) from the upper Paraguay river basin, brazil. Biochem. Syst. Ecol., 35: 869876. McQuiTTY l.l., 1966. - Similarity analysis by reciprocal pairs for discrete and continuous data. Educ. Psych. Meas., 26: 825831. riCharDSon b.J., baverSToCK P.r. & aDaMS M., 1986. allozyme electrophoresis. a handbook for animal Systematics and Population Studies. 410 p. San Diego: academic Press, inc. MEEK S.E. & HILDEBRAND S.F., 1916. - The ishes of the fresh waters of Panama. Publ. Field Mus. Nat. Hist. (Zool. Ser.), 10(15): 217-374. SChaeFer S.a., 1987. - osteology of Hypostomus plecostomus (linnaeus), with a phylogenetic analysis of the loricariid subfamilies (Pisces: Siluroidei). Contr. Sci., 394: 1-31. Mol J.h., vari r.P., Covain r., WillinK P.W. & FiSChMULLER S., 2012. - Annotated checklist if the freshwater ishes of Suriname. Cybium, 36(1): 263-292. SChoMburGK r.h., 1841. - The natural history of Fishes of Guiana. Part i. In: The naturalists’ library, vol. 3 (Jardine W., ed.), 263 p., pls. 261-230. edinburgh: W.h. lizars. MonToYa-burGoS J.i., 2003. - historical biogeography of the catfish genus Hypostomus (Siluriformes: loricariidae), with implications on the diversiication of Neotropical ichthyofauna. Mol. Ecol., 12: 1855-1867. Muller S. & Weber C., 1992. - les dents des sous-familles hypostominae et ancistrinae (Pisces, Siluriformes, loricariidae) et leur valeur taxonomique. Rev. Suisse Zool., 99 (4): 747754. nei M., 1978. - estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics, 89: 583590 nei M. & Graur D., 1984. - extent of protein polymorphism and the neutral mutation theory. Evol. Biol., 17: 73-118. NORMAN J.R., 1926. - Description of nine new fresh-water ishes from French Guiana and brazil. Ann. Mag. Nat. Hist., 18: 91-97. PaSTeur n., PaSTeur G., bonhoMMe F., CaTalan J. & briTTon-DaviDian J., 1987. - Manuel technique de Génétique par Électrophorèse des Protéines. 217 p. Paris: Technique et Documentation, lavoisier. PounD K.l., noWlin W.h., huFFMan D.G. & bonner T.h., 2011. - Trophic ecology of a nonnative population of suckermouth catfish (Hypostomus plecostomus) in a central Texas spring-fed stream. Environ. Biol. Fish., 90: 277-285. PraKaSh S., leWonTin r.C. & hubbY J.l., 1969. - a molecular approach to the study of genic heterozygosity in natural populations. iv. Patterns of genic variation in central, marginal and isolated populations of Drosophila pseudoobscura. Genetics, 61: 841-858. r DeveloPMenT Core TeaM, 2009. - r: a language and environment for statistical computing. r Foundation for Statistical Computing, vienna, austria. http://www.r-project.org. raYMonD M. & rouSSeT F., 1995. - an exact test for population differentiation. Evolution, 49(6): 1280-1283. 226 SCHULTZ L.P., 1944. - The catishes of Venezuela, with description of thirty-eight new forms. Proc. U.S. Natl. Mus., 94: 173338. ShaW G., 1804. - General Zoology or Systematic natural history. vol. 5. Pisces, in two Parts. [(pt 1): i-v + 1-250, pls. 293-132, 243+, 265+, 256+, 274+ and (pt 2): i-vi + 251-463, pls. 132182, 158+.]. london: G. Kearsley. ShiMoDaira h., 2002. - an approximately unbiased test of phylogenetic tree selection. Syst. Biol., 51: 492-508. ShiMoDaira h., 2004. - approximately unbiased tests of regions using multistep-multiscale bootstrap resampling. Ann. Statist., 32: 2616-2641. SneaTh P.h. & SoKal r.r., 1973. - numerical Taxonomy. The Principles and Practice of Numerical Classiication. 573 p. San Francisco: W. h. Freeman. STEINDACHNER F., 1877a. - Die Süsswasserische des südöstlichen brasilien, iii. Sber. Akad. Wiss. Wien, Math.-Nat. Kl., 74: 559-694, pls. 551-513. STEINDACHNER F., 1877b. - Die Süsswasserische des südöstlichen brasilien, iv. Sber. Akad. Wiss. Wien, Math.-Nat. Kl., 76: 217-230, pls. 211-212 STeinDaChner F., 1878. - ichthyologische beiträge, vi. Sber. Akad. Wiss. Wien, Math.-Nat. Kl., 77(1): 379-392, pls. 371373. STEINDACHNER F., 1881. - Beiträge zur Kenntniss der Flussische Südamerica’s, ii. Denkschr. Akad. Wiss., Wien, Math.-Nat. Kl., 43: 103-146, pls. 101-107. van Der STiGChel J.W.b., 1946. - South american nematognathi. 204 p. leiden: e. J. brill. SuZuKi r. & ShiMoDaira h., 2004. - an application of multiscale bootstrap resampling to hierarchical clustering of microarray data: how accurate are these clusters? In: The Fifteenth international Conference on Genome informatics, P034. Cybium 2012, 36(1) Weber et al. ThioulouSe J. & DraY S., 2007. - interactive multivariate data analysis in r with the ade4 and ade4TkGui packages. J. Stat. Soft., 22(5): 1-14. ThorPe J.P. & SolÉ-Cava a.M., 1994. - The use of allozyme electrophoresis in invertebrate systematics. Zool. Script., 23(1): 3-18. vari r.P., FerrariS Jr C.J., raDoSavlJeviC a. & FunK v.a. (eds), 2009. - Checklist of the Freshwater Fishes of the Guiana Shield. Bull. Biol. Soc. Wash., 17: 1-95. venableS W.n. & riPleY b.D., 2002. - Modern applied Statistics with S. Fourth edition. 512 p. new York: Springer. WarD r.D., SKibinSKi D.o.F. & WooDWarK M., 1992. Protein heterozygosity, protein structure, and taxonomic differentiation. Evol. Biol., 26: 73-159. Weber C., 1985. - Hypostomus dlouhyi, nouvelle espèce de poisson-chat cuirassé du Paraguay (Pisces, Siluriformes, loricariidae). Rev. Suisse Zool., 94(4): 955-968. WEBER C., 2003. - Subfamily Hypostominae (Armored catishes). In: Check list of the Freshwater Fishes of South and Central america (CloFFSCa) (Kullander S.o., Ferraris C.J.& reis r.e., eds), pp. 351-372. Porto alegre: edipucrs. Weber C. & MonToYa-burGoS J.i., 2002. - Hypostomus fonchii sp. n. (Siluriformes: loricariidae) from Peru, a key species suggesting the synonymy of Cochliodon with Hypostomus. Rev. Suisse Zool., 109(2): 355-368. WHEELER A.W., 1989. - Further notes on the ishes from the collection of laurens Theodore Gronovius (1730-1777). Zool. J. Linn. Soc., 95: 205-218. ZaWaDZKi C.h., reneSTo e. & bini l.M., 1999. - Genetic and morphometric analysis of three species of the genus Hypostomus lacepède, 1803 (osteichthyes: loricariidae) from the rio iguaçu basin (brazil). Rev. Suisse Zool., 106 (1): 91-105. Cybium 2012, 36(1) hypostomus species from the Guianas ZaWaDZKi C.h., reiS r.e. & reneSTo e., 2000. - allozyme discrimination of three species of Loricariichthys (Siluriformes: loricariidae) from Southern brazil. Rev. Suisse Zool., 107(4): 663-674. ZaWaDZKi C.h., MaChaDo M.F.P.S. & reneSTo e., 2001. Differential expression for tissue-specific isozymes in three species of Hypostomus lacepède, 1803 (Teleostei: loricariidae). Biochem. Syst. Ecol., 29: 911-922. ZaWaDZKi C.h., alveS a.l., reneSTo e. & oliveira C., 2004. - biochemical evidence of a possible new species of Neoplecostomus (Teleostei: loricariidae) from the upper rio Paraná basin brazil. Biochem. Syst. Ecol., 32: 573-582. ZaWaDZKi C.h., reneSTo e., reiS r.e., Moura M.o. & MaTeuS r.P., 2005. - allozyme relationships in hypostomines (Teleostei: loricariidae) from the itaipu reservoir, uper rio Paraná basin, brazil. Genetica, 123: 271-283. ZaWaDZKi C.h., birinDelli J.l.o. & liMa F.C.T., 2008a. a new pale-spotted species of Hypostomus lacepède (Siluriformes: loricariidae) from the rio Tocantins and rio Xingu basins in Central brazil. Neotrop. Ichthyol., 6(3): 395-402. ZaWaDZKi C.h., Weber C. & Pavanelli C.S., 2008b. - Two new species of Hypostomus lacepède (Teleostei: loricariidae) from the upper rio Paraná basin, Central brazil. Neotrop. Ichthyol., 6(3): 403-412. ZaWaDZKi C.h., Weber C. & Pavanelli C.S., 2010. - a new dark-saddled species of Hypostomus (Siluriformes: loricariidae) from the upper rio Paraguay basin. Neotrop. Ichthyol., 8(4): 719-725. 227