Acessibilidade / Reportar erro

Morphologic and genetic variation within a relict Andean catfish, Hatcheria macraei , and its relationship with Trichomycterus areolatus and Bullockia maldonadoi (Siluriformes: Trichomycteridae)

Abstract

The South American siluriform fishes are found primarily in the Neotropical region, north and east of the Colorado River of Argentina, with a few relict species distributed southward and westward on both sides of the Andes Mountains. Three of these, the closely related trichomycterids Hatcheria macraei, Trichomycterus areolatus and Bullockia maldonadoi, have been subject to historical taxonomic and nomenclatural arrangements. Here, we amplify a 652-bp fragment of COI mtDNA from 55 H. macraei individuals and use publicly available Cytb mtDNA sequences of the three taxa to assess their relationship, genetic variation and haplotype distribution in relation to hydrographic basins. In addition, we extend a recent morphometric study on H. macraei by analyzing body shape in 447 individuals collected from 24 populations across their entire cis-Andean distribution. We identified some lineages previously assigned to T. areolatus that show a closer relationship to either B. maldonadoi or H. macraei, revealing new boundaries to their currently known trans-Andean distribution. We found a great morphologic variation among H. macraei populations and a high genetic variation in H. macraei, T. areolatus and B. maldonadoi associated with river basins. We highlight further integrative studies are needed to enhance our knowledge of the southern Andean trichomycterid diversity.

Key words
Andes mountains; catfishes; genetic variation; haplotype networks; morphologic variation; phylogeny

INTRODUCTION

The catfishes (Order Siluriformes) are a particular diverse group of more than 3700 species within the superorder Ostariophysi (Fricke et al. 2021FRICKE R, ESCHMEYER WN & VAN DER LAAN R. 2021. (http://researcharchive.calacademy.org/research/ichthyology/catalog/fishcatmain.asp). Electronic version accessed 7 July 2021.
http://researcharchive.calacademy.org/re...
). Trichomycteridae is the second most diverse family of the order, comprising 349 valid species in nine subfamilies of which Trichomycterinae, shown to be monophyletic firstly by a morphological phylogenetic study (Datovo & Bockmann 2010DATOVO A & BOCKMANN FA. 2010. Dorsolateral head muscles of the catfish families Nematogenyidae and Trichomycteridae (Siluriformes: Loricarioidei): comparative anatomy and phylogenetic analysis. Neotrop Ichthyol 8: 193-246.) and more recently by molecular phylogenetic studies (Ochoa et al. 2017OCHOA LE, ROXO FF, DONASCIMIENTO C, SABAJ MH, DATOVO A, ALFARO M & OLIVEIRA C. 2017. Multilocus analysis of the catfish family Trichomycteridae (Teleostei: Ostariophysi: Siluriformes) supporting a monophyletic Trichomycterinae. Mol Phylogenet Evol 115: 71-81., 2020OCHOA LE ET AL. 2020. Phylogenomic analysis of trichomycterid catfishes (Teleostei: Siluriformes) inferred from ultraconserved elements. Sci Rep 10: 1-15., Fernandez et al. 2021FERNANDEZ L, ARROYAVE J & SCHAEFER SA. 2021. Emerging patterns in phylogenetic studies of trichomycterid catfishes (Teleostei, Siluriformes) and the contribution of Andean diversity. Zoologica Scripta 50: 318-336.), is the most diverse with 253 valid species (Fricke et al. 2021FRICKE R, ESCHMEYER WN & VAN DER LAAN R. 2021. (http://researcharchive.calacademy.org/research/ichthyology/catalog/fishcatmain.asp). Electronic version accessed 7 July 2021.
http://researcharchive.calacademy.org/re...
). Trichomycterus Valenciennes, 1832 is the major genus with more than 160 species recognized (Froese & Pauly 2021FROESE R & PAULY D. 2021. FishBase (www.fishbase.org). Electronic version accessed 7 July 2021.
www.fishbase.org...
), whereas the Andean genera Hatcheria Eigenmann, 1909 and Bullockia Arratia, Chang, Menu-Marque & Rojas, 1978 are monotypic.

The taxonomic relationships between Trichomycterus and Hatcheria have long been discussed (Arratia & Menu-Marque 1981ARRATIA G & MENU-MARQUE S. 1981. Revision of the freshwater catfishes of the genus Hatcheria (Siluriformes, Trichomycteridae) with comentaries on ecology and biogeography. Zool Anz Jena 207: 88-111., Fernandez et al. 2021FERNANDEZ L, ARROYAVE J & SCHAEFER SA. 2021. Emerging patterns in phylogenetic studies of trichomycterid catfishes (Teleostei, Siluriformes) and the contribution of Andean diversity. Zoologica Scripta 50: 318-336.). The genus Trichomycterus was erected in 1832 by Valenciennes (Cuvier & Valenciennes 1832CUVIER G & VALENCIENNES M. 1832. Histoire naturelle des poissons. Vol 3, Paris, 500 p.), who in a following work described T. areolatus from the Maipo River basin, west of the Andes (Cuvier & Valenciennes 1846CUVIER G & VALENCIENNES M. 1846. Histoire naturelle des poissons. Vol 18. Paris, 505 p.). In 1855, Girard described T. macraei collected near Uspallata, east of the Andes (Girard 1855GIRARD C. 1855. Fishes. In: The U.S. Naval astronomical expedition to the southern hemisphere, during the years 1849-1852. Vol II. Washington DC, USA, p. 230-253.), and then in 1909 Eigenmann created the genus Hatcheria and placed the specific epithet macraei within it (Eigenmann 1909EIGENMANN CH. 1909. The fresh water fishes of Patagonia and an examination of the archiplata-archelenis theory. In: Reports of the Princeton University Expeditions to Patagonia, 1896-1899. Princeton, NJ, USA, p. 225-374.). There were two other Hatcheria species described from Chilean waters, H. maldonadoi Eigenmann, 1927 and H. bullocki Fowler, 1940, before Tchernavin (1944)TCHERNAVIN VV. 1944. A revision of some Trichomycterinae based on material preserved in the British Museum (Natural History). Proc Zool Soc Lon 114: 234-275. put Hatcheria in synonymy with Trichomycterus, followed by De Buen who proposed it as a subgenus of Trichomycterus (De Buen 1958DE BUEN F. 1958. Ictiología. La familia Ictaluridae nueva para la fauna aclimatada de Chile y algunas consideraciones sobre los Siluroidei indígenas. Inv Zool Chilenas 4: 46-158.). Later, Ringuelet et al. (1967)RINGUELET RA, ARÁMBURU RH & DE ARÁMBURU AA. 1967. Los peces argentinos de agua dulce. Comisión de Investigaciones Científicas de la Provincia de Buenos Aires, Buenos Aires, 602 p. mentioned five species of Hatcheria in Argentina; and Arratia et al. (1978)ARRATIA G, MENU-MARQUE S & ROJAS G. 1978. About Bullockia gen. nov., Trichomycterus mendozensis n. sp. and revision of the family Trichomycteridae (Pisces, Siluriformes). Stud Neotrop Fauna Environ 13: 157-194. synonymized H. bullocki to H. maldonadoi including it in a new, monotypic genus for Chilean waters, Bullockia. Finally, Arratia & Menu-Marque (1981)ARRATIA G & MENU-MARQUE S. 1981. Revision of the freshwater catfishes of the genus Hatcheria (Siluriformes, Trichomycteridae) with comentaries on ecology and biogeography. Zool Anz Jena 207: 88-111. synonymized all Argentinean Hatcheria species to H. macraei, leaving it as a monotypic genus.

The above taxonomic and nomenclatural arrangements have affected the identification of the geographic range for each of these species (Fig. 1a). The currently recognized distribution for T. areolatus is trans-Andean rivers in Central Chile (Froese & Pauly 2021FROESE R & PAULY D. 2021. FishBase (www.fishbase.org). Electronic version accessed 7 July 2021.
www.fishbase.org...
) southward from Limarí River (Unmack et al. 2009aUNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892.), whereas the distribution for H. macraei is cis-Andean rivers south latitude 29° S (Froese & Pauly 2021FROESE R & PAULY D. 2021. FishBase (www.fishbase.org). Electronic version accessed 7 July 2021.
www.fishbase.org...
). However, their southernmost distribution as well as their presence at one or both sides of the Andes have varied according to different authors. For T. areolatus, its southernmost distribution has been described at Puerto Varas (Pardo 2002PARDO R. 2002. Morphologic differentiation of Trichomycterus areolatus Valenciennes 1846 (Pisces: Siluriformes: Trichomycteridae) from Chile. Gayana 66: 203-205.), Abtao (Manriquez et al. 1988MANRIQUEZ A, HUAQUÍN L, ARELLANO M & ARRATIA G. 1988. Aspectos reproductivos de Trichomycterus areolatus Valenciennes, 1846 (Pisces: Teleostei: Siluriformes) en Rio Angostura, Chile. Stud Neotrop Fauna Environ 23: 89-102.) and Chiloé Island at the X Región de los Lagos (Arratia 1981ARRATIA G. 1981. Géneros de peces de aguas continentales de Chile. Mus Nac Hist Nat 34: 3-108., Dyer 2000DYER BS. 2000. Systematic review and biogeography of the freshwater fishes of Chile. Estud Oceanol 19: 77-98.), as well as further south at General Carrera/Buenos Aires Lake in the XI Región de Aysén (Arratia et al. 1983ARRATIA G, PEÑAFORT MB & MENU-MARQUE S. 1983. Peces de la región sureste de los Andes y sus probables relaciones biogeográficas actuales. Deserta 7: 48-107.). Moreover, it has also been identified in the eastern slope in the Cuyo region (Arratia & Menu-Marque 1981ARRATIA G & MENU-MARQUE S. 1981. Revision of the freshwater catfishes of the genus Hatcheria (Siluriformes, Trichomycteridae) with comentaries on ecology and biogeography. Zool Anz Jena 207: 88-111., Arratia et al. 1983ARRATIA G, PEÑAFORT MB & MENU-MARQUE S. 1983. Peces de la región sureste de los Andes y sus probables relaciones biogeográficas actuales. Deserta 7: 48-107.). For H. macraei, its southernmost cis-Andean distribution is described at Blanco River (Unmack et al. 2012UNMACK PJ, BARRIGA JP, BATTINI MA, HABIT EM & JOHNSON JB. 2012. Phylogeography of the catfish Hatcheria macraei reveals a negligible role of drainage divides in structuring populations. Mol Ecol 21: 942-959.), as well as in the trans-Andean Baker and Aysén River basins (Arratia & Menu-Marque 1981ARRATIA G & MENU-MARQUE S. 1981. Revision of the freshwater catfishes of the genus Hatcheria (Siluriformes, Trichomycteridae) with comentaries on ecology and biogeography. Zool Anz Jena 207: 88-111., Campos et al. 1998CAMPOS H et al. 1998. Categorías de conservación de peces nativos de aguas continentales de Chile. Bol Mus Nac Hist Nat 47: 101-122., Unmack et al. 2012UNMACK PJ, BARRIGA JP, BATTINI MA, HABIT EM & JOHNSON JB. 2012. Phylogeography of the catfish Hatcheria macraei reveals a negligible role of drainage divides in structuring populations. Mol Ecol 21: 942-959.). Furthermore, recent records extended its trans-Andean northern distribution to the X Región de Los Lagos, where it was found in sympatry with T. areolatus in the Valdivia and Bueno River basins (Unmack et al. 2009aUNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892., 2012).

Figure 1
a) Geographic distribution of Hatcheria macraei (black dots), Trichomycterus areolatus (white dots) and Bullockia maldonadoi (grey dots) following Unmack et al. (2009a, 2012). White stars indicate localities that were considered with presence of T. areolatus in cis-Andean rivers by previous authors: 1. Arratia & Menu-Marque (1981)ARRATIA G & MENU-MARQUE S. 1981. Revision of the freshwater catfishes of the genus Hatcheria (Siluriformes, Trichomycteridae) with comentaries on ecology and biogeography. Zool Anz Jena 207: 88-111.; 2. Arratia et al. (1983)ARRATIA G, PEÑAFORT MB & MENU-MARQUE S. 1983. Peces de la región sureste de los Andes y sus probables relaciones biogeográficas actuales. Deserta 7: 48-107.; 3 and 4: Baigún & Ferriz (2003)BAIGÚN C & FERRIZ R. 2003. Distribution patterns of native freshwater fishes in Patagonia (Argentina). Org Divers Evol 3: 151-159.. b) Capture sites of H. macraei from which individuals were employed for both COI amplification and geometric morphometric analysis (grey dots), COI amplification only (white dot) and morphometric analysis only (black dots). Numbers correspond to locality data presented in Table I.

Such taxonomic arrangements and variation in described range distributions seem to be due, at least in part, to the high morphological variability exhibited within these species, as well as similarities among them. Intraspecific differences in H. macraei were noted by Ringuelet et al. (1967)RINGUELET RA, ARÁMBURU RH & DE ARÁMBURU AA. 1967. Los peces argentinos de agua dulce. Comisión de Investigaciones Científicas de la Provincia de Buenos Aires, Buenos Aires, 602 p. in his taxonomic key allowing the recognition of distinct populations, which were found to differ in body shape at different latitudes (Chiarello-Sosa et al. 2018CHIARELLO-SOSA JM, BATTINI MA & BARRIGA JP. 2018. Latitudinal phenotypic variation in the southernmost trichomycterid, the catfish Hatcheria macraei: an amalgam of population divergence and environmental factors. Biol J Linn Soc 124: 718-731.). Likewise, Arratia et al. (1978)ARRATIA G, MENU-MARQUE S & ROJAS G. 1978. About Bullockia gen. nov., Trichomycterus mendozensis n. sp. and revision of the family Trichomycteridae (Pisces, Siluriformes). Stud Neotrop Fauna Environ 13: 157-194. found meristic and morphological characters, and particularly those considering body relations, highly variable in B. maldonadoi. For T. areolatus, differences have been found between its northern (Choapa River) and central (Bío-bío River) populations 500 km apart (Pardo 2002PARDO R. 2002. Morphologic differentiation of Trichomycterus areolatus Valenciennes 1846 (Pisces: Siluriformes: Trichomycteridae) from Chile. Gayana 66: 203-205.), as well as between the latter and a southern population (Bueno River) separated by a similar distance (Colihueque et al. 2017COLIHUEQUE N, CORRALES O & YÁÑEZ M. 2017. Morphological analysis of Trichomycterus areolatus Valenciennes, 1846 from southern Chilean rivers using a truss-based system (Siluriformes, Trichomycteridae). ZooKeys 695: 135-152.).

Among the morphological characters used to distinguish between H. macraei and T. areolatus are caudal peduncle depth, dorsal-ray counts, position of anus, dorsal-fin length, and dorsal-fin shape (Arratia et al. 1978ARRATIA G, MENU-MARQUE S & ROJAS G. 1978. About Bullockia gen. nov., Trichomycterus mendozensis n. sp. and revision of the family Trichomycteridae (Pisces, Siluriformes). Stud Neotrop Fauna Environ 13: 157-194., Unmack et al. 2009bUNMACK PJ, HABIT EM & JOHNSON JB. 2009b. New records of Hatcheria macraei (Siluriformes, Trichomycteridae) from Chilean province. Gayana 73: 102-110.). However, these characters do not completely discriminate the species. For instance, Unmack et al. (2009b)UNMACK PJ, HABIT EM & JOHNSON JB. 2009b. New records of Hatcheria macraei (Siluriformes, Trichomycteridae) from Chilean province. Gayana 73: 102-110. found significant variation in caudal peduncle depth and position of anus in H. macraei. These authors used dorsal-fin ray count as a nearly diagnostic character, with overall shape of the dorsal-fin as the only external character to discriminate both species.

The cause of such morphological variability is likely due to the aquatic environments (Senay et al. 2015SENAY C, BOISCLAIR D & PERES-NETO PR. 2015. Habitat-based polymorphism is common in stream fishes. J Anim Ecol 84: 219-227.). All three species occur in the rithronic zone of streams and rivers in areas with loose pebbles, gravel or sandy bottoms, substrates that allow individuals to bury themselves and avoid predators (Arratia 1983ARRATIA G. 1983. Preferencias de hábitat de peces siluriformes de aguas continentales de Chile (Fam. Diplomystidae and Trichomycteridae). Stud Neotrop Fauna Environ 18: 217-237.). Hatcheria macraei and T. areolatus prefer dark substrates, whereas B. maldonadoi prefer clearer bottoms (Arratia 1983ARRATIA G. 1983. Preferencias de hábitat de peces siluriformes de aguas continentales de Chile (Fam. Diplomystidae and Trichomycteridae). Stud Neotrop Fauna Environ 18: 217-237.). Bullockia maldonadoi is present in a few rivers with more restricted characteristics that are mainly pluvial and do not have tributaries (Arratia et al. 1978ARRATIA G, MENU-MARQUE S & ROJAS G. 1978. About Bullockia gen. nov., Trichomycterus mendozensis n. sp. and revision of the family Trichomycteridae (Pisces, Siluriformes). Stud Neotrop Fauna Environ 13: 157-194.). In contrast, H. macraei and T. areolatus have broad environmental tolerances with a widespread presence in small headwater streams to low elevation rivers on a variety of substrates, especially gravel and rocks (Arratia et al. 1983ARRATIA G, PEÑAFORT MB & MENU-MARQUE S. 1983. Peces de la región sureste de los Andes y sus probables relaciones biogeográficas actuales. Deserta 7: 48-107., Unmack et al. 2009aUNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892., 2012). Moreover, it appears that habitat preference changes with age and can be somewhat flexible, allowing individuals to respond to seasonal river flow fluctuations. This may in part explain their widespread distribution (Arratia 1983ARRATIA G. 1983. Preferencias de hábitat de peces siluriformes de aguas continentales de Chile (Fam. Diplomystidae and Trichomycteridae). Stud Neotrop Fauna Environ 18: 217-237.). Adults of both species inhabit the benthic part of the rhitronal region of rivers and streams (Arratia 1983ARRATIA G. 1983. Preferencias de hábitat de peces siluriformes de aguas continentales de Chile (Fam. Diplomystidae and Trichomycteridae). Stud Neotrop Fauna Environ 18: 217-237.) and have been described as negatively phototactic and orienting themselves against the current (Ringuelet et al. 1967RINGUELET RA, ARÁMBURU RH & DE ARÁMBURU AA. 1967. Los peces argentinos de agua dulce. Comisión de Investigaciones Científicas de la Provincia de Buenos Aires, Buenos Aires, 602 p., Ringuelet 1975RINGUELET RA. 1975. Zoogeografía y ecología de los peces de aguas continentales de la Argentina y consideraciones sobre las áreas ictiológicas de América del Sur. Ecosur 2: 1-122., Arratia 1976ARRATIA G. 1976. Variaciones de las hipurapófisis en algunos peces siluriformes (Familia Trichomycteridae). An Mus Hist Nat Valparaíso Chile 9: 105-114., Arratia & Menu-Marque 1981ARRATIA G & MENU-MARQUE S. 1981. Revision of the freshwater catfishes of the genus Hatcheria (Siluriformes, Trichomycteridae) with comentaries on ecology and biogeography. Zool Anz Jena 207: 88-111., Habit et al. 2005HABIT E, VICTORIANO P & CAMPOS H. 2005. Ecología trófica y aspectos reproductivos de Trichomycterus areolatus (Pisces, Trichomycteridae) en ambientes lóticos artificiales. Rev Biol Trop 53: 195-210., Barriga et al. 2013BARRIGA JP, ESPINÓS NA, CHIARELLO-SOSA JM & BATTINI MA. 2013. The importance of substrate size and interstitial space in the microhabitat selection by the stream-dwelling catfish Hatcheria macraei (Actinopterygii, Trichomycteridae). Hydrobiologia 705: 191-206.). An important size-related habitat shift has been reported for H. macraei with positive allometric larval growth affecting mainly the head region, locomotion structures, the trunk region, and body robustness. Changes from positive allometric to isometric growth in H. macraei reflects the larva–juvenile transition (Barriga & Battini 2009BARRIGA JP & BATTINI MA. 2009. Ecological significances of ontogenetic shifts in the stream-dwelling catfish, Hatcheria macraei (Siluriformes, Trichomycteridae), in a Patagonian river. Ecol Freshw Fish 18: 395-405.). Juvenile and adult isometric growth has been described in T. areolatus (Habit et al. 2005HABIT E, VICTORIANO P & CAMPOS H. 2005. Ecología trófica y aspectos reproductivos de Trichomycterus areolatus (Pisces, Trichomycteridae) en ambientes lóticos artificiales. Rev Biol Trop 53: 195-210.).

Andean streams and rivers form many isolated basins that can promote population differences in low vagility organisms such as fishes (Hillman et al. 2014HILLMAN SS, DREWES RC, HEDRICK MS & HANCOCK TV. 2014. Physiological vagility and its relationship to dispersal and neutral genetic heterogeneity in vertebrates. J Exp Biol 217: 3356-3364., Hancock & Hedrick 2018HANCOCK TV & HEDRICK MS. 2018. Physiological vagility affects population genetic structure and dispersal and enables migratory capacity in vertebrates. Comp Biochem Physiol Part A Mol & Integr Physiol 223: 42-51.). Intrinsic riverine differences may prevail depending on the river basin (Pardo et al. 2005PARDO R, SCOTT S & VILA I. 2005. Shape analysis in Chilean species of Trichomycterus (Osteichthyes: Siluriformes) using geometric morphometry. Gayana 69: 180-183.), which may be accentuated in large altitudinal and latitudinal ranges such as the geographical distribution of Andean trichomycterids. For instance, the northern distribution range of T. areolatus is characterized by short streams and rivers located about 4000 m above sea level (asl) with low flow and relatively high gradients, whereas southern rivers are relatively longer and at lower elevation, with relatively higher flows (Pardo et al. 2005PARDO R, SCOTT S & VILA I. 2005. Shape analysis in Chilean species of Trichomycterus (Osteichthyes: Siluriformes) using geometric morphometry. Gayana 69: 180-183.). Such differences in riverine characteristics, including water velocity, have been shown to be associated with body shape differences in H. macraei (Chiarello-Sosa et al. 2018CHIARELLO-SOSA JM, BATTINI MA & BARRIGA JP. 2018. Latitudinal phenotypic variation in the southernmost trichomycterid, the catfish Hatcheria macraei: an amalgam of population divergence and environmental factors. Biol J Linn Soc 124: 718-731.).

Even though morphological differentiation may or may not be present in closely related species, genetic divergence is expected as a result of genetic drift among species inhabiting isolated, environmentally different basins (Keeley et al. 2007KEELEY ER, PARKINSON EA & TAYLOR EB. 2007. The origins of ecotypic variation of rainbow trout: a test of environmental vs. genetically based differences in morphology. J Evol Biol 20: 725-736.). For instance, cryptic species sensu Mayr (i.e. species with poor morphological differentiation but that represent separate evolutionary lineages) are frequently observed in closely related taxa with broad distributions (Melo et al. 2016MELO BF, OCHOA LE, VARI RP & OLIVEIRA C. 2016. Cryptic species in the Neotropical fish genus Curimatopsis (Teleostei, Characiformes). Zool Scripta 45: 650-658.). In addition to phylogenetic inference, an analysis of geographic haplotype distributions may provide hints on the population structure and diversification between taxa. Moreover, species that have diverged very recently may prove difficult to distinguish by nuclear genes due to their relatively slow mutation rate (Hare 2001HARE MP. 2001. Prospects for nuclear gene phylogeography. Trens Ecol & Evol 16: 700-706.). On the contrary, the higher mutation rate of mitochondrial DNA makes it better fitted to resolve species-level and genus-level phylogenetic relationships among lineages (Rubinoff & Holland 2005RUBINOFF D & HOLLAND BS. 2005. Between two extremes: mitochondrial DNA is neither the panacea nor the nemesis of phylogenetic and taxonomic inference. Syst Biol 54: 952-961.). Being haploid and maternally inherited, mtDNA has one-quarter of the effective population size of nuclear genes, thus a mitochondrial haplotype tree can better resolve short internodes (Avise et al. 1987AVISE JC, ARNOLD J, BALL RM, BERMINGHAM E, LAMB T, NEIGEL JE, REEB CA & SAUNDERS NC. 1987. Intraspecific phylogeography: the mitochondrial DNA bridge between population genetics and systematics. Ann Rev Ecol Syst 18: 489-522., Moore 1995MOORE WS. 1995. Inferring phylogenies from mtDNA variation: mitochondrial-gene trees versus nuclear-gene trees. Evolution 49: 718-726.) than other markers.

Here, we performed an integrative analysis to gain a more comprehensive knowledge of the limits of distribution of H. macraei, T. areolatus and B. maldonadoi. The goals of our work are: 1) to assess their relationship as well as their genetic variation and haplotype distribution in relation to hydrographic basins based on COI haplotypes -amplified for H. macraei in this study- and publicly available Cytb haplotypes for the three taxa generated by Unmack et al. (2009a, 2012); and 2) to analyze variation of body shape in H. macraei throughout its entire cis-Andean distribution in relation to Cytb genetic clades and river basins.

MATERIALS AND METHODS

Collection of H. macraei specimens

We collected a total of 480 H. macraei individuals using seine net or electrofishing (Smith-Root backpack 24V, 600–900 V, 60 Hz, 6 ms standard pulse) in littorals of streams, rivers and lakes from 25 Argentinean localities in San Juan, Mendoza, Neuquén, Río Negro, Chubut, and Santa Cruz provinces, covering the whole cis-Andean distribution area (Fig. 1b, Table I). We sacrificed fish with an overdose of anesthesia (benzocaine 1:10000), and took digital images of the left side of each fish immediately to minimize parallax error for morphometric analysis. We preserved the samples in 96% ethanol to avoid the toxic formaldehyde, as well as to better preserve the DNA for molecular analysis. We sampled muscle tissue from 65 specimens (Supplementary Material - Table SI) collected from three localities at the Colorado River basin and one locality each at Negro and Yelcho River basins (Fig. 1b, Table I), before depositing them in Museo de La Plata (MLP), Facultad de Ciencias Naturales y Museo, Universidad Nacional de La Plata. We deposited the remainder specimens in the Department of Zoology at Centro Regional Universitario Bariloche, Universidad Nacional del Comahue, Argentina. We provide specimens deposit information in Table SI. Fish were caught and euthanized according to institutional guidelines for animal welfare and regulations detailed in Argentinean National Law No. 14346. Samplings were performed with the allowance of Parques Nacionales (Argentina, https://sib.gob.ar/?#!/buscar/Cussac) and of the Provinces of San Juan, Mendoza, La Pampa, Rio Negro, Chubut, and Santa Cruz, in the frame of the project NSF-PIRE (OISE 0530267, USA) through the following institutions: Brigham Young University, Centro Nacional Patagónico, Dalhousie University, Instituto de Botánica Darwinion, Universidad Austral de Chile, Universidad Nacional del Comahue, Universidad de Concepción and University of Nebraska.

Table I
Sampling localities for Hatcheria macraei with abbreviation codes employed in this study, their geographic coordinates and drainage basins. Number of individuals captured at each sampling site (NS), number of samples successfully amplified for COI (NCOI) and number of specimens included in the morphometric analysis (NM).

Phylogenetic assessment and genetic variability in H. macraei, T. areolatus and B. maldonadoi

We first tested the efectiveness of a nuclear gene to discriminate distinct evolutionary lineages on the three closely related trichomycterids here studied by running a Bayesian analysis for the RAG-1 nuclear gene utilizing 200 sequences available from GenBank (acc. n. JN186409-608). Thereafter, we reassessed their relationships based on COI and Cytb mtDNA haplotypes.

We extracted DNA from the 65 H. macraei muscle samples using the AccuPrep Genomic DNA Extraction kit (Bioneer, South Korea) and successfully amplified a 652-bp fragment of COI mtDNA gene from 55 of them (Table SI) by polymerase chain reaction (PCR) at the International Barcode of Life (iBOL) Argentinean reference Barcode Laboratory at the Museo Argentino de Ciencias Naturales (MACN) in Buenos Aires, Argentina. Amplifications were performed in a total volume of 12.5 μl consisting of 6.25 μl of 10% trehalose, 1.25 μl of 10X PCR buffer, 0.625 μl MgCl2 (50 mM) 0.0625 μl of each dNTP (10 mM), 0.0625 μl of Taq DNA Polymerase (New England Biolabs), 2 μl of molecular grade water, 2 μl of DNA template, and 0.125 μl of each primer from the primer cocktail C_VF1LFt1-C_VR1LRt1 (Ivanova et al. 2007IVANOVA NV, ZEMLAK TS, HANNER RH & HEBERT PDN. 2007. Universal primer cocktails for fish DNA barcoding. Mol Ecol Notes 7: 544-548.). Cycling conditions consisted of an initial denaturation step of 94 °C for 2 min, followed by 35 cycles of 94 °C for 30 s, 52 °C for 40 s and 72 °C for 1 min, and a final extension at 72 °C for 10 min. We checked amplicons on 1.2% agarose gels before sending to sequencing on an ABI 3730XL capillary sequencer at the Canadian Centre for DNA Barcoding (CCDB) in Guelph, ON, Canada. We manually edited sequences with BioEdit v. 7.0.5.2 (Hall 1999HALL T. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl Acids Symposium Series 41: 95-98.) and made sequences available in the Barcode of Life Data Systems website (http://www.boldsystems.org/). We identified non-redundant COI haplotypes using DnaSP v. 5.10 (Librado & Rozas 2009LIBRADO P & ROZAS J. 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25: 1451-1452.). We used MrBayes 3.1.2 (Ronquist & Huelsenbeck 2003RONQUIST F & HUELSENBECK JP. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572-1574.) to construct by Bayesian inference the phylogenetic relationship of the H. macraei haplotypes together with sequences from T. areolatus and B. maldonadoi (GenBank acc. n. KY857926, KY857963, and KY857964) previously published by Ochoa et al. (2017)OCHOA LE, ROXO FF, DONASCIMIENTO C, SABAJ MH, DATOVO A, ALFARO M & OLIVEIRA C. 2017. Multilocus analysis of the catfish family Trichomycteridae (Teleostei: Ostariophysi: Siluriformes) supporting a monophyletic Trichomycterinae. Mol Phylogenet Evol 115: 71-81.. We searched for COI sequences of all other southern Andean species that clustered together with H. macraei, T. areolatus and B. maldonadoi in clade E of Fernandez et al. 2021FERNANDEZ L, ARROYAVE J & SCHAEFER SA. 2021. Emerging patterns in phylogenetic studies of trichomycterid catfishes (Teleostei, Siluriformes) and the contribution of Andean diversity. Zoologica Scripta 50: 318-336., however we found no sequences available. We then used sequences of Ituglanis parkoi (Miranda Ribeiro, 1944) (GenBank acc. n. KY857937) which clustered in sister clade D, and of T. brasiliensis Lütken, 1874 (GenBank acc. n. KY857993) that clustered in a more distant clade M from these authors’s work, to root the tree. We identified the best-fit substitution model to be TIM1+I, nst = 6 and rates = equal, using the Akaike Information Criterion (AIC) with jModelTest 2.1.10 (Posada 2008POSADA D. 2008. jModelTest: Phylogenetic model averaging. Mol Biol Evol 25: 1253-1256.). We executed two independent runs totaling four chains, for 50 million MCMC generations, sampling every 1000 trees. We discarded the initial 25% of the resulting trees as burn-in, and constructed a 50% majority-rule consensus tree. We calculated pairwise genetic divergences among H. macraei haplotypes and T. areolatus, B. maldonadoi, I. parkoi and T. brasiliensis with PAUP* 4.0a (build 166) (Swofford 2003SWOFFORD DL. 2003. PAUP*. Phylogenetic analysis using parsimony and other methods. Version 4.0a. Sinauer. Sunderland, MA.) using the above parameter estimates on the CIPRES Science Gateway (Miller et al. 2010MILLER MA, PFEIFFER W & SCHWARTZ T. 2010. “Creating the CIPRES Science Gateway for inference of large phylogenetic trees” in Proceedings of the Gateway Computing Environments Workshop (GCE), 14 Nov 2010, New Orleans, LA p 1-8.). Finally, we constructed a haplotype network using the median joining algorithm developed by Bandelt et al. (1999)BANDELT H-J, FORSTER P & RÖHL A. 1999. Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol 16: 37-48., implemented in PopART v. 1.7 (Leigh & Bryant 2015LEIGH JW & BRYANT D. 2015. POPART: full-feature software for haplotype network construction. Methods Ecol Evol 6: 1110-1116.), and edited it using the vector graphics editor Inkscape (https://www.inkscape.org).

For Cytb, we utilized the following haplotypes available from GenBank: H. macraei, n=73, acc. n. FJ772216–25, JN186346–408; T. areolatus, n=125, acc. n. FJ772091–215; and B. maldonadoi, n=12; acc. n. FJ772226–37. These haplotypes, generated by Unmack et al. (2009a, 2012), covered the entire geographic known distribution of each species. Unlike Unmack et al.’s previous analyses, we ran a single Bayesian analysis with all 210 haplotypes on MrBayes 3.1.2 (Ronquist & Huelsenbeck 2003RONQUIST F & HUELSENBECK JP. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572-1574.). In addition, we placed B. maldonadoi as part of the ingroup instead of the outgroup, and searched for Cytb sequences of all other southern Andean trichomycterids from clade E of Fernandez et al. (2021)FERNANDEZ L, ARROYAVE J & SCHAEFER SA. 2021. Emerging patterns in phylogenetic studies of trichomycterid catfishes (Teleostei, Siluriformes) and the contribution of Andean diversity. Zoologica Scripta 50: 318-336. but found none available. Subsequently, we also included as outgroups sequences of Ituglanis parkoi (GenBank acc. n. KY858018/026) and T. brasiliensis (GenBank acc. n. KY858062). We identified the best-fit substitution model to be TIM3+I+G, nst = 6 and rates = gamma. We then ran a Bayesian inference analysis to calculate pairwise genetic divergences and constructed a haplotype network as described above using all the Cytb haplotypes.

Morphological variability in H. macraei

We carried out the morphological analysis on digital images from 447 individuals captured at 24 localities of Argentina (Fig. 1b, Table SI). We collected a total of 20 landmarks, four links, and five sliders (Fig. 2) on individuals with the software TpsDig version 2.05 (F.J. RohlfROHLF FJ. 1996. Morphometric spaces, shape components, and the effects of linear transformations. In: Marcus LF et al. (Eds), Advances in Morphometrics. Proceedings of the 1993 NATO Advanced Studies Institute on Morphometrics in Il Ciocco. New York, Plenum Publishing Corp., State University of New York, Stony Brook, NY). We aligned, rotated, translated and scaled landmarks through a Generalized Procrustes Analysis (GPA) using a consensus configuration (Rohlf & Slice 1990ROHLF FJ & SLICE DE. 1990. Extensions of the Procrustes method for the optimal superimposition of landmarks. Syst Zool 39: 40-59., Rohlf & Marcus 1993ROHLF FJ & MARCUS LF. 1993. A revolution in morphometrics. Trends Ecol Evol 8: 129-132.). We calculated partial and relative warps using Tpsrelw version 1.35, and visualized deformation grids using Relative Warps Analysis (RWA), thus using the consensus configuration as an average body shape (Cadrin 2000CADRIN SX. 2000. Advances in morphometric identification of fishery stock. Rev Fish Biol Fish 10: 91-112.). We performed Discriminant Analysis (DA, Norusis 1986NORUSIS MJ. 1986. SpSS/PC+ advanced statistics. SPSS Inc., Chicago, Illinois, 204 p.) employing partial warps and uniform coordinates (Rohlf 1996) in order to find body shape differences, considering capture sites, river basins that contain them, and Cytb clades obtained by Bayesian inference. We conducted further DA within each genetic clade considering capture sites as the grouping variable. Basins were inappropriate as such due to the coexistence of two clades in some basins.

Figure 2
Landmarks (20, red circles), links (4, black lines), and sliders (5, black arrows) collected on the 447 Hatcheria macraei individuals included in the morphometric analysis.

RESULTS

Phylogenetic assessment

Not surprisingly, we found the RAG-1 nuclear gene to be uninformative due to a lack of resolution (data not shown). On the contrary, mitochondrial genes were much more resolutive. We identified eight non-redundant haplotypes among the 55 COI sequences of H. macraei showing 19 variable sites (Table II). The COI Bayesian tree (Fig. 3) showed a well-supported subclade situating T. areolatus as the sister species of B. maldonadoi (posterior probability [PP] = 0.92). With the exception of haplotype 7 (Rosario Lake, Yelcho River basin, a Pacific drainage), relationships among the H. macraei COI haplotypes were shallow, although they did reveal a geographic pattern clustering southernmost haplotypes 6, 7, and 8 along T. areolatus and B. maldonadoi (PP = 0.94). The Colorado River basin haplotypes 1-5 remained basal to that subclade (Fig. 3).

Figure 3
Bayesian phylogenetic tree based on eight Hatcheria macraei haplotypes obtained by amplification of the COI mtDNA region, as well as two haplotypes from Trichomycterus areolatus and one from Bullockia maldonadoi available at GenBank. Haplotypes from T. brasiliensis and Ituglanis parkoi were used as outgroups. GenBank acc. n. of sequences for B. maldonadoi, T. areolatus, and outgroup taxa are indicated in their respective terminals. Bayesian posterior probabilities are shown on nodes.
Table II
Haplotypes and nucleotide position of variable sites found in COI sequences amplified from 55 individuals of Hatcheria macraei.

The Cytb Bayesian tree resulted in four well-supported major clades, all in a polytomy (Fig. 4). The “Bullockia spp. Clade” presented three subclades. The first one clustered the northernmost haplotypes from the Limarí, Choapa, Petorca and Aconcagua River basins (previously assigned to T. areolatus clade “A” by Unmack et al. 2009aUNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892.). The second subclade “B. maldonadoi”, clustered all B. maldonadoi haplotypes including those from its type locality (i.e. Andalién River; Arratia et al. 1978ARRATIA G, MENU-MARQUE S & ROJAS G. 1978. About Bullockia gen. nov., Trichomycterus mendozensis n. sp. and revision of the family Trichomycteridae (Pisces, Siluriformes). Stud Neotrop Fauna Environ 13: 157-194.). The third subclade clustered three haplotypes from the Reloca River along two haplotypes from the Maipo River (previously assigned to T. areolatus clade “B” by Unmack et al. 2009aUNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892.). In light of the pairwise genetic distances (see below), we labelled the subclades closely-related to the one of B. maldonadoi as “B. m. ssp 1” and “B. m. ssp 2” until further studies are conducted to resolve their taxonomic relationship to B. maldonadoi. The second, “T. areolatus Clade”, clustered many of T. areolatus haplotypes from Imperial and Toltén River basins in central Chile, along most of those from northern freshwater systems up to the Maipo River. It was composed of three subclades corresponding to clades “C”, “D”, and “E” sensu Unmack et al. (2009a)UNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892.. The third clade contained a minor subset of T. areolatus haplotypes from three central rivers: Imperial, Toltén and Valdivia (clade “F” sensu Unmack et al. 2009aUNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892.). We here refer to it as “T. areolatus ssp. Clade” to indicate it as a distinct evolutionary lineage from T. areolatus until further evidence is collected. The fourth, “H. macraei Clade”, presented three subclades. The major one clustered not only H. macraei haplotypes from basins that drain to both Atlantic and Pacific Oceans (“Big Clade” sensu Unmack et al. 2012UNMACK PJ, BARRIGA JP, BATTINI MA, HABIT EM & JOHNSON JB. 2012. Phylogeography of the catfish Hatcheria macraei reveals a negligible role of drainage divides in structuring populations. Mol Ecol 21: 942-959.), but also all haplotypes previously assigned to clade “G” of T. areolatus in Unmack et al. 2009aUNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892. (i.e., haplotypes from Bueno, Butalcura, Llico, and Maullín River basins). Considering the drainage of its basins, we named this major subclade “Atlantic-Pacific”. As the other two smaller subclades each clustered haplotypes from basins draining to either ocean, we named these “Atlantic” and “Pacific”, which correspond to subclades “Colorado” and “Baker/Cholila” sensu Unmack et al. (2012)UNMACK PJ, BARRIGA JP, BATTINI MA, HABIT EM & JOHNSON JB. 2012. Phylogeography of the catfish Hatcheria macraei reveals a negligible role of drainage divides in structuring populations. Mol Ecol 21: 942-959., respectively.

Figure 4
Bayesian phylogenetic tree based on 73, 125, and 12 publicly available haplotypes of the Cytb mtDNA region for Hatcheria macraei, Trichomycterus areolatus and Bullockia maldonadoi, respectively. Haplotypes from T. brasiliensis and Ituglanis parkoi were used as outgroups. Bayesian posterior probabilities are shown on major nodes. Terminals are represented by haplotypes with their corresponding GenBank accession numbers and the river basins where are present (see Fig. 5 for color coding). Clades and subclades are indicated to the right of the tree. Letters in brackets A-G correspond to the seven major clades described for T. areolatus by Unmack et al. (2009a). Abbreviations in brackets (BC: Big Clade; B-Ch: Baker-Cholila; Co: Colorado) correspond to Cytb clades described for H. macraei by Unmack et al. (2012).
Figure 4
Continuation

Genetic variability and haplotype networks

Pairwise COI genetic distances for H. macraei haplotypes, T. areolatus, B. maldonadoi and two outgroups are shown in Table III. The genetic divergence of H. macraei from T. areolatus haplotypes KY857963 and KY857964 were 3% (2.8-3.2%) and 2.7% (2.1-3.2%), respectively. The genetic divergence of H. macraei from B. maldonadoi was 4.6% (3.7-5.0%), whereas the latter from T. areolatus was 4.7% (4.6-4.8%). The intraspecific genetic divergence among the eight H. macraei COI haplotypes was 1.1% (0.2-2.4%).

Table III
Pairwise genetic distances corrected under the best-fit GTR substitution model calculated for COI, comparing the eight Hatcheria macraei haplotypes described in this study and haplotypes available from GenBank for Trichomycterus areolatus and Bullockia maldonadoi. Comparisons with the more distantly related species Ituglanis parkoi and T. brasiliensis are included.

Genetic distances among major Cytb clades found Bullockia spp. as the most divergent, with distances of 5.2% (3.9-7.0%), 5.4% (4.5-6.5%) and 6.1% (4.9-7.5%) from T. areolatus, T. areolatus ssp. and H. macraei, respectively (Table IVa). The latter presented distances of 3.5% (2.7-4.4%) and 3.7% (2.9-6.5%) from T. areolatus and T. areolatus ssp., respectively; whereas these two showed a divergence of 2.8% (2.2-3.4%) (Table IVa). Pairwise distances within major Cytb clades were more variable, with the lowest of 0.3% (0.1-0.7%) for T. areolatus ssp. and the highest of 4.1% (3.0-5.5%) found for the “Bullockia spp. Clade” (Table IVa). For the latter, the genetic distance between “B. maldonadoi ssp. 1” and “B. maldonadoi ssp. 2” subclades was 3.8% (3.0-4.6%), while each of them presented similar divergence (4.2%) from the “B. maldonadoi” subclade, with minimum-maximum distances of 3.3-5.5% and 3.4-5.0%, respectively (Table IVb). A pairwise distance of 2.2% was found among haplotypes of the “B. maldonadoi” subclade, with a minimum of 0.1% and a remarkably maximum of 4.3% (Table IVb). Divergences among T. areolatus subclades ranged between 1.4% and 2.2%, whereas within each subclade it ranged from 0.4% and 0.6% (Table IVc). Finally, the divergence among H. macraei subclades ranged 1.4-1.8%, with within-clade divergences between 0.2% and 0.6% (Table IVd).

Table IV
Pairwise genetic distances corrected under the best-fit GTR substitution model calculated for Cytb comparing: a) major Cytb clades; b) subclades within the Bullockia spp. clade; c) subclades within the Trichomycterus areolatus clade; and d) subclades within the Hatcheria macraei clade. Mean genetic distances are indicated in bold, with corresponding ranges in parentheses below.

The haplotype network for the Cytb mtDNA region (Fig. 5) exhibited high resolution as expected, as the haplotypes employed were representative of the whole distribution range of the southern Andean trichomycterids here considered. This network showed a deep genetic structure with a likely ancestral haplogroup common to all present haplogroups. In particular for H. macraei, the Cytb haplotype network showed three haplogroups coincident with the Bayesian tree for this marker: Atlantic, Pacific and Atlantic-Pacific haplogroups. On the contrary, the COI haplotype network for H. macraei presented a poor resolution as the sampling was limited to a few populations in their cis-Andean range. There must be undetected haplotypes that if included would improve the outcome of this network.

Figure 5
Median joining networks constructed with eight haplotypes of the COI mtDNA region found in Hatcheria macraei individuals in the present study (upper right); and with 73, 125, and 12 publicly available Cytb haplotypes from H. macraei, Trichomycterus areolatus and Bullockia maldonadoi, respectively (left side). Nodes represent unique haplotypes scaled according to their frequency, with colors matching with the basin where they are present (for location of rivers see also Fig. 1a). Black nodes represent inferred unsampled or ancestral haplotypes. Branch lengths are proportional to the number of nucleotide mutations between nodes. Single nucleotide mutations between nodes are indicated with bars crossing the branches if mutations <= 5, or with numbers in boxes if mutations > 5. Letters in brackets A-G correspond to the seven major Cytb clades described for T. areolatus by Unmack et al. (2009a)UNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892.. Abbreviations in brackets (BC: Big Clade; B-Ch: Baker-Cholila; Co: Colorado) correspond to Cytb clades described for H. macraei by Unmack et al. (2012)UNMACK PJ, BARRIGA JP, BATTINI MA, HABIT EM & JOHNSON JB. 2012. Phylogeography of the catfish Hatcheria macraei reveals a negligible role of drainage divides in structuring populations. Mol Ecol 21: 942-959..

Geometric morphometrics

The first two RWs of the GMA explained 31.9% of the variance (RW1= 22.5% and RW2= 9.4%), indicating a mostly isodiametric shape of the morphological data cloud in the hyperspace. The body shape variation explained by RW1 ranges between individuals with a bigger head, shorter trunk, a shorter basis of the dorsal-fin and a higher caudal peduncle in the positive semi-axis, and individuals with a smaller head, longer trunk, a longer base of the dorsal-fin and a lower caudal peduncle, in the negative semi-axis. The body shape variation explained by RW2, ranges between individuals with a bigger head, a shorter base of the dorsal-fin, and higher caudal peduncle in the negative semi-axis, and individuals with a smaller head, a longer basis of the dorsal-fin and lower caudal peduncle in the positive semi-axis (Fig. 6). At odds with the RWs, the DA of the partial warps and uniform coordinates, grouped data successively by capture sites, rivers basins, and Cytb genetic clades for H. macraei, and showed the latter to give better discrimination with two significant discriminating functions explaining 100% of the variance, and 95.7% of original grouped cases correctly classified (Fig. 7, Table V).

Figure 6
Geometric morphometrics of Hatcheria macraei. Relative warp 1 and 2 (RW1, RW2) values (n= 447) with deformation grids at the extreme of semi-axes.
Figure 7
Discriminant analyses (Discriminant functions 1 and 2, DF1, DF2) of the partial warps and uniform coordinates obtained from geometric morphometrics analysis on Hatcheria macraei individuals with cis-Andean geographic distribution (n= 447), by capture sites (n=24), basins (n=8), and genetic clades (n=3).
Table V
Discriminant analyses of the partial warps and uniform coordinates obtained from geometric morphometrics analysis for Hatcheria macraei (N= 447), by capture sites, river basins, and Cytb genetic clades.

We subsequently processed the original database separately within each of the three Cytb genetic clades, and visualized the different morphologies by Cytb clades (Fig. 8) and significantly different morphologies by capture sites based on DA applied to partial warps within clades. We obtained 10 significant discriminant functions (DF, P< 0.001) among the 18 capture sites in the Atlantic-Pacific subclade that explained 94.1% of the total variance with 87.2% of original grouped cases correctly classified. Within the Pacific subclade, we obtained one significant DF (P< 0.001) between capture sites, i.e. Cholila Lake and Blanco River, explaining 100% of total variance with 94.3% of original grouped cases correctly classified. Within the Atlantic subclade, we obtained two significant DFs (P< 0.001) between capture sites, i.e. Claro Stream, Nihuil and Ullum Reservoirs, explaining 100% of total variance with 100% of original grouped cases correctly classified. However, a significant cubic regression (P< 0.0001) of the latter two DF1 and DF2 with body size (total length) for the latter clade implies that these differences must be considered with caution. All files and photographs employed in the morphometric analysis are available at https://ri.conicet.gov.ar/handle/11336/151708.

Figure 8
Grids at RW1= 0, RW2= 0 coordinates of each Cytb genetic clade for Hatcheria macraei. Double arrows indicate the most conspicuous differences.

DISCUSSION

In this integrative study, we analyzed the morphologic variation of H. macraei and the genetic variation of H. macraei, T. areolatus and B. maldonadoi, in relation to hydrographic basins to help further our knowledge of their limits of distribution. Our results unveiled new geographic boundaries between B. maldonadoi and the northernmost T. areolatus populations, as well as between H. macraei and the southernmost T. areolatus populations. Recent advances in knowledge of catfishes in South America (Katz et al. 2018KATZ AM, BARBOSA MA, OLIVEIRA MATTOS JL & MOREIRA DA COSTA WJE. 2018. Multigene analysis of the catfish genus Trichomycterus and description of a new South American trichomycterine genus (Siluriformes, Trichomycteridae). Zoosyst Evol 94: 557-566., Ochoa et al. 2017OCHOA LE, ROXO FF, DONASCIMIENTO C, SABAJ MH, DATOVO A, ALFARO M & OLIVEIRA C. 2017. Multilocus analysis of the catfish family Trichomycteridae (Teleostei: Ostariophysi: Siluriformes) supporting a monophyletic Trichomycterinae. Mol Phylogenet Evol 115: 71-81., 2020, Costa et al. 2020COSTA WJ, HENSCHEL E & KATZ AM. 2020. Multigene phylogeny reveals convergent evolution in small interstitial catfishes from the Amazon and Atlantic forests (Siluriformes: Trichomycteridae). Zool Scripta 49: 159-173., Fernandez et al. 2021FERNANDEZ L, ARROYAVE J & SCHAEFER SA. 2021. Emerging patterns in phylogenetic studies of trichomycterid catfishes (Teleostei, Siluriformes) and the contribution of Andean diversity. Zoologica Scripta 50: 318-336.) allowed us to examine the relationships among lineages of these three southern Andean trichomycterids. By using all Cytb haplotypes available for them in a single phylogenetic inference, our results revealed some lineages previously assigned to T. areolatus actually have a closer relationship to either B. maldonadoi or H. macraei. Moreover, our results confirmed the monophyly of Hatcheria, which clustered all of its lineages together, and of Bullockia, by which it was also confirmed the monophyly of B. maldonadoi and its distinctiveness by presenting the relatively longest branch (Fig. 4), coincident with results in Fernandez et al. (2021)FERNANDEZ L, ARROYAVE J & SCHAEFER SA. 2021. Emerging patterns in phylogenetic studies of trichomycterid catfishes (Teleostei, Siluriformes) and the contribution of Andean diversity. Zoologica Scripta 50: 318-336.. Our analyses resolved Bullockia as the sister group of T. areolatus, contrary to Fernandez et al. (2021)FERNANDEZ L, ARROYAVE J & SCHAEFER SA. 2021. Emerging patterns in phylogenetic studies of trichomycterid catfishes (Teleostei, Siluriformes) and the contribution of Andean diversity. Zoologica Scripta 50: 318-336. who found it to be H. macraei. First, our Bayesian inference tree for the COI marker showed B. maldonadoi and T. areolatus clustering together with high support (Fig. 3, PP=0.92). Second, the Cytb pairwise genetic distance of T. areolatus from Bullockia spp. was lower than that from H. macraei, being 5.2% and 6.1% respectively (Table IVa). Third, the Cytb haplotype network showed Bullockia haplogroups more closely related to Trichomycterus than to Hatcheria (Fig. 4). Finally, and contrary to the current knowledge, we found that the geographic distribution of populations belonging to Bullockia spp. and T. areolatus each span an extent of about 6° of the meridian, with about half of that range overlapped.

Our results amount to the emerging evidence for geographically circumscribed subclades of Trichomycterus species as described by Fernandez et al. (2021)FERNANDEZ L, ARROYAVE J & SCHAEFER SA. 2021. Emerging patterns in phylogenetic studies of trichomycterid catfishes (Teleostei, Siluriformes) and the contribution of Andean diversity. Zoologica Scripta 50: 318-336., and provide hints this may also occur in Bullockia, which its fragmented distribution is reflected by the high divergence among its lineages (4.1%; Table IVa). Across the vast range of Bullockia, from the Limarí River at 30°40’S south to the Ramadillas River at 37°18’S (Fig. 5), B. maldonadoi showed a disjunct distribution in the very south coincident with was previously described by Arratia et al. (1978)ARRATIA G, MENU-MARQUE S & ROJAS G. 1978. About Bullockia gen. nov., Trichomycterus mendozensis n. sp. and revision of the family Trichomycteridae (Pisces, Siluriformes). Stud Neotrop Fauna Environ 13: 157-194., ranging from the Itata River at 37°09’S to the Ramadillas River (Fig. 5). Both B. maldonadoi ssp. 1 and B. maldonadoi ssp. 2 presented relatively high genetic distances of 4.2% from B. maldonadoi, with even a significant distance of 3.8% between themselves (Table IVb). This brings attention to the current status of Bullockia as a monotypic genus, as it can be hypothesized B. maldonadoi may have sister species north from the Itata River. We could argue that the northern Bullockia populations in Limarí, Choapa, Petorca and Aconcagua River basins belong to a cryptic species, or to a subspecies of B. maldonadoi, in the same sense we could argue they are Operational Taxonomic Units (OTUs). Any of these alternatives are just pragmatic definitions to group individuals by similarity (Sokal & Crovello 1970SOKAL RR & CROVELLO TJ. 1970. The Biological Species Concept: a critical evaluation. Am Nat 104: 127-153.; Mayr 1982MAYR E. 1982. Speciation and macroevolution. Evolution 36: 1119-1132.). Therefore, we highlight the necessity of further investigation to describe patterns of character distributions of Bullockia spp. subclades.

To date, the geographic distribution of T. areolatus was considered to be between 30°40’S (Limarí River) and 42º 22’S in Chiloé Island (Pardo 2002PARDO R. 2002. Morphologic differentiation of Trichomycterus areolatus Valenciennes 1846 (Pisces: Siluriformes: Trichomycteridae) from Chile. Gayana 66: 203-205., Unmack et al. 2009aUNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892., b). However, our results showed it to range about half of that, between 33°S in Maipo River basin – which contains its type locality (Girard 1855GIRARD C. 1855. Fishes. In: The U.S. Naval astronomical expedition to the southern hemisphere, during the years 1849-1852. Vol II. Washington DC, USA, p. 230-253.) – and 39°S at the Toltén River. This river is considered the northern limit of a geological province, the Patagonian Cordillera, characterized by a batholitic belt that extends southward to Cape Horn Islands (Ramos & Gighlione 2008). Besides its peculiar geological significance, the Toltén River also holds a strong climatological meaning as it is located at the estimated maximum extent of the major ice sheet during the last glacial maximum (LGM) (Rabassa et al. 2011RABASSA J, CORONATO AM & MARTÍNEZ O. 2011. Late Cenozoic glaciations in Patagonia and Tierra del Fuego: an updated review. Biol J Linn Soc 103: 316-335.). Moreover, our results showing Toltén River as the actual southern boundary of T. areolatus distribution agrees with the geographic limit Dyer (2000)DYER BS. 2000. Systematic review and biogeography of the freshwater fishes of Chile. Estud Oceanol 19: 77-98. established between the South-Central and the Southern ichthyogeographic subprovinces.

The great morphological variation among H. macraei populations was consistent with the intra-specific variation described in studies by Arratia & Menu-Marque (1981)ARRATIA G & MENU-MARQUE S. 1981. Revision of the freshwater catfishes of the genus Hatcheria (Siluriformes, Trichomycteridae) with comentaries on ecology and biogeography. Zool Anz Jena 207: 88-111. and Chiarello-Sosa et al. (2018)CHIARELLO-SOSA JM, BATTINI MA & BARRIGA JP. 2018. Latitudinal phenotypic variation in the southernmost trichomycterid, the catfish Hatcheria macraei: an amalgam of population divergence and environmental factors. Biol J Linn Soc 124: 718-731.. This was reflected mainly in the head size, trunk length, caudal peduncle height, and dorsal-fin length. Different microhabitat river conditions (e.g. water velocity) can affect body shape, mostly body height and fin size, as shown by the work of Chiarello-Sosa et al. (2018)CHIARELLO-SOSA JM, BATTINI MA & BARRIGA JP. 2018. Latitudinal phenotypic variation in the southernmost trichomycterid, the catfish Hatcheria macraei: an amalgam of population divergence and environmental factors. Biol J Linn Soc 124: 718-731.. These authors found a relationship between stream flow and H. macraei body shape, in which the faster the water flow, the narrow the caudal peduncle is as well as the longer the base of the dorsal-fin. In this sense, the morphological variation we described in H. macraei was congruent with river basins and Cytb genetic clades (Figs. 7 and 8; Table V), resulting in three well defined morphological groups. One group was formed by the northern Colorado River basin within the Atlantic Clade, another one was extended southward from the Negro River basin within the Atlantic-Pacific Clade, and the other group was restricted to two populations separated by more than 5 latitudinal degrees within the Pacific Clade (Lake Cholila in the head waters of the Yelcho River; and Blanco River - its southernmost record, presently a closed basin).

Immediately southward from the Toltén River, the identified southern boundary of T. areolatus, we confirmed the Valdivia River as the northern boundary of the trans-Andean distribution of H. macraei as previously reported by Unmack et al. (2009a, 2012). Moreover, our data showed that the remaining trichomycterid populations analysed to the south in Bueno, Llico and Maullín Rivers, as well as those from Butalcura River in Chiloé Island, which were previously assigned to T. areolatus in Unmack et al. (2009a)UNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892., are actually best aligned with H. macraei, thus expanding the southern boundary of the trans-Andean distribution of this species (Fig. 5). Furthermore, the position of haplotypes from these four populations at the base of the Atlantic-Pacific subclade of H. macraei (Fig. 4), suggests that these rivers may have been refugia during the later glacial periods. Evidence of glacial refugia has been found in Coastal Chile and southern flanks of the Andes (Premoli et al. 2000PREMOLI AC, KITZBERGER T & VEBLEN TT. 2000. Isozyme variation and recent biogeographical history of the long-lived conifer Fitzroya cupressoides. J Biogeogr 27: 251-260.), and populations located outside ice limits seem to have been isolated during the glacial period (Premoli et al. 2002PREMOLI AC, SOUTO CP, ROVERE AE, ALLNUT TR & NEWTON AC. 2002. Patterns of isozyme variation as indicators of biogeographic history in Pilgerodendron uviferum (D. Don) Florín. Div Distr 8: 57-66.). Following recent post-glacial periods, the subsequent deglaciation and drainage reversals (Unmack et al. 2012UNMACK PJ, BARRIGA JP, BATTINI MA, HABIT EM & JOHNSON JB. 2012. Phylogeography of the catfish Hatcheria macraei reveals a negligible role of drainage divides in structuring populations. Mol Ecol 21: 942-959.) may have allowed H. macraei to colonize freshwater systems at the eastern slope of the Andes. This scenario can also be interpreted from the shallow nodes that characterize the Negro, Chubut and Deseado River basins in the Atlantic-Pacific subclade (Fig. 4), as well as their star-like shape in the Cytb haplotype network, showing them as recent expansions (Fig. 5).

The southern Andean trichomycterids here studied diversified in drainage networks characterized by strong flow rates that formed during the Andes uplift, which generated the vast extension of gravel mantels that cover most of Patagonia (Martínez & Kutschker 2011MARTÍNEZ OA & KUTSCHKER A. 2011. The ‘Rodados Patagónicos’ (Patagonian shingle formation) of eastern Patagonia: environmental conditions of gravel sedimentation. Biol J Linn Soc 103: 336-345.). Lineages within each species have likely evolved in allopatry as a consequence of the history of glacial cycles in southern South America during the Cenozoic (Rabassa et al. 2011RABASSA J, CORONATO AM & MARTÍNEZ O. 2011. Late Cenozoic glaciations in Patagonia and Tierra del Fuego: an updated review. Biol J Linn Soc 103: 316-335.). Therefore, the finding of small lineages such as the T. areolatus ssp. over an LGM boundary is not surprising since this type of area, which likely harbored refugia during glacial cycles, may indeed concentrate many endemisms (Dyer 2000DYER BS. 2000. Systematic review and biogeography of the freshwater fishes of Chile. Estud Oceanol 19: 77-98.). This lineage presented a genetic distance of 2.8% from the major T. areolatus Clade, doubling the within-clade distance of the latter (Table IVa). Taking into consideration that T. areolatus ssp. also presented a low within-group divergence of 0.3% (Table IVa), the fact that they live in sympatry with T. areolatus in Imperial and Toltén River basins, and with H. macraei in the Valdivia River, it would be prudent to consider this small lineage as a subspecies of T. areolatus until further evidence is collected. We hypothesize this divergent lineage may have originated as a consequence of isolation in a glacial refugium in the Valdivia River. The star-like shape of the haplotype network for this species, with a high number of haplotypes, and its most common haplotype the one of a few present in the Imperial and Toltén River basins in the north (Fig. 5), seem to support this. Fossil evidence suggests that Siluriforms were widely distributed in southern South America until at least the late Miocene, with a warmer climate than today (Cione et al. 2005CIONE AL, AZPELICUETA MM, CASCIOTTA JR & DOZO MT. 2005. Tropical freshwater teleosts from Miocene beds of eastern Patagonia, southern Argentina. Geobios 38: 29-42., Dozo et al. 2010DOZO MT, BOUZA P, MONTI A, PALAZZESI L, BARREDA V, MASSAFERRO G, SCASSO RA & TAMBUSSI CP. 2010. Late Miocene continental biota in Northeastern Patagonia (Península Valdés, Chubut, Argentina). Palaeogeogr Palaeoclim Palaeoecol 297: 100-109., Azpelicueta & Cione 2016AZPELICUETA MM & CIONE AL. 2016. A southern species of the tropical catfish genus Phractocephalus (Teleostei: Siluriformes) in the Miocene of South America. J South Am Earth Sci 67: 221-230.). Starting in early Pliocene, a regional uplift of the Andes promoted the eastward aridization of Patagonia (Ramos & Ghiglione 2008RAMOS VA & GHIGLIONE MC. 2008. Tectonic Evolution of the Patagonian Andes. In: Developments in Quaternary Sciences. Elsevier, p. 57-71.). Although warm climate could have extended until late Pliocene (Cione et al. 2005CIONE AL, AZPELICUETA MM, CASCIOTTA JR & DOZO MT. 2005. Tropical freshwater teleosts from Miocene beds of eastern Patagonia, southern Argentina. Geobios 38: 29-42.), glaciation cycles (Rabassa et al. 2011RABASSA J, CORONATO AM & MARTÍNEZ O. 2011. Late Cenozoic glaciations in Patagonia and Tierra del Fuego: an updated review. Biol J Linn Soc 103: 316-335.) likely fragmented further their distribution. Several putative refugia for plant and vertebrates seem to have existed in southern South America during the climate oscillations that characterized the Pleistocene (Sérsic et al. 2011SÉRSIC AN, COSACOV A, COCUCCI AA, JOHNSON LA, POZNER R, AVILA LJ, SITES JW JR & MORANDO M. 2011. Emerging phylogeographic patterns of plants and terrestrial vertebrates from Patagonia. Biol J Linn Soc 103: 475-494.), which may have isolated to and reunited populations several times. Therefore, the Andean uplift followed by glacial cycles may have promoted the origin of many endemisms in trichomycterids, which may not necessarily involve morphological change. Morphologically static cladogenetic events (Bickford et al. 2007BICKFORD D, LOHMAN DJ, SODHI NS, NG PKL, MEIER R, WINKER K, INGRAM KK & DAS I. 2007. Cryptic species as a window on diversity and conservation. Trends Ecol Evol 22: 148-155.) can result in morphologically indistinguishable species which are taxonomically classified as being one, i.e. cryptic species (Mayr 1942MAYR E. 1942. Systematics and the origin of species. NY, USA.). There are many instances in recent years of cryptic species described in Neotropical fishes (e.g. Martin & Bermingham 2000MARTIN AP & BERMINGHAM W. 2000. Regional endemism and cryptic species revealed by molecular and morphological analysis of a widespread species of Neotropical catfish. Proc R Soc Lond B 267: 1135-1141., García-Dávila et al. 2013GARCÍA-DÁVILA C, DUPONCHELLE F, CASTRO-RUIZ D, VILLACORTA J, QUÉROUIL S, CHOTA-MACUYAMA W, NÚÑEZ J, RÖMER U, CARVAJAL-VALLEJOS F & RENNO J-F. 2013. Molecular identification of a cryptic species in the Amazonian predatory catfish genus Pseudoplatystoma (Bleeker, 1962) from Peru. Genetica 141: 347-358., Gomes et al. 2015GOMES LC, PESSALI TC, SALES NG, POMPEU PS & CARVALHO DC. 2015. Integrative taxonomy detects cryptic and overlooked fish species in a neotropical river basin. Genetica 143: 581-588., Melo et al. 2016MELO BF, OCHOA LE, VARI RP & OLIVEIRA C. 2016. Cryptic species in the Neotropical fish genus Curimatopsis (Teleostei, Characiformes). Zool Scripta 45: 650-658., Guimarães et al. 2020GUIMARÃES EC, DE BRITO PS, BRAGANCA PHN, SANTOS JP, KATZ AM, CARVALHO COSTA LF & OTTONI FP. 2020. Integrative taxonomy reveals two new cryptic species of Hyphessobrycon Durbin, 1908 (Teleostei: Characidae) from the Maracaçumé and middle Tocantins River basins, Eastern Amazon region. Eur J Taxonomy 723: 77-107., Pereira et al. 2021PEREIRA LHG, CASTRO JRC, VARGAS PMH, GOMEZ JAM & OLIVEIRA C. 2021. The use of an integrative approach to improve accuracy of species identification and detection of new species in studies of stream fish diversity. Genetica 149: 103-116.), as well as in other regions of the world (e.g. Africa: Jirsová et al. 2019JIRSOVÁ, D, ŠTEFKA J, BLAžEK R, MALALA JO, LOTULIAKOU DE, MAHMOUD ZN & JIRKU M. 2019. From taxonomic deflation to newly detected cryptic species: Hidden diversity in a widespread African squeaker catfish. Sci Rep 9: 15748., Asia: Shao et al. 2021SHAO W-H, CHENG J-L & ZHANG E. 2021. Eight in one: hidden diversity of the bagrid catfish Tachysurus albomarginatus s.l. (Rendhal, 1928) widespread in lowlands of South China. Front Genetics 12: 713793.). Looking at the population structure of marine-descendant species, that settled well after the presence of the Siluriformes in Patagonia, can allow us to compare groups of taxa at the present time. The coincidence observed here between morphological grouping and genetic clades at the level of mitochondrial markers in H. macraei, also demonstrated in populations of T. areolatus (Pardo et al. 2005PARDO R, SCOTT S & VILA I. 2005. Shape analysis in Chilean species of Trichomycterus (Osteichthyes: Siluriformes) using geometric morphometry. Gayana 69: 180-183.), does not occur in marine related Patagonian fishes. Both Percichthys trucha (Valenciennes, 1833) and Odontesthes hatcheri (Eigenmann, 1909EIGENMANN CH. 1927. The fresh-water fishes of Chile. Mem Nat Acad Sci 22: 1-63.) show a major degree of morphological variation (Crichigno et al. 2012CRICHIGNO SA, BATTINI MA & CUSSAC VE. 2012. Early morphological variation and induction of phenotypic plasticity in Patagonian pejerrey. Neotrop Ichthyol 10: 341-348., 2013CRICHIGNO S, CONTE-GRAND C, BATTINI M & CUSSAC V. 2013. Cephalic morphological variation in freshwater silversides Odontesthes hatcheri and Odontesthes bonariensis in Patagonia: introgression and ecological relationships. J Fish Biol 83: 542-559., 2016CRICHIGNO SA, HATTORI RS, STRÜSSMANN CA & CUSSAC VE. 2016. Morphological comparison of wild, farmed and hybrid specimens of two South American silversides, Odontesthes bonariensis and Odontesthes hatcheri. Aquac Res 47: 1797-1808., Conte-Grand et al. 2015CONTE-GRAND C, SOMMER J, ORTÍ G & CUSSAC V. 2015. Populations of Odonthestes (Teleostei: Atheriniformes) in the Andean region of Southern South America: body shape and hybrid individuals. Neotrop Ichthyol 13: 1-14.). However, mitochondrial markers failed to reveal genetic structuring between populations for these species (Ruzzante et al. 2006RUZZANTE DE, WALDE SJ, CUSSAC VE, DALEBOUT ML, SEIBERT J, ORTUBAY S & HABIT E. 2006. Phylogeography of the Percichthyidae (Pisces) in Patagonia: roles of orogeny, glaciation, and volcanism. Mol Ecol 15: 2949-2968., 2011RUZZANTE DE, WALDE SJ, MACCHI PJ, ALONSO M & BARRIGA JP. 2011. Phylogeography and phenotypic diversification in the Patagonian fish Percichthys trucha: the roles of Quaternary glacial cycles and natural selection. Biol J Linn Soc 103: 514-529., Conte-Grand et al. 2015CONTE-GRAND C, SOMMER J, ORTÍ G & CUSSAC V. 2015. Populations of Odonthestes (Teleostei: Atheriniformes) in the Andean region of Southern South America: body shape and hybrid individuals. Neotrop Ichthyol 13: 1-14., Rueda et al. 2017RUEDA EC, MULLANEY KA, CONTE-GRAND C, HABIT EM, CUSSAC V & ORTÍ G. 2017. Displacement of native Patagonian freshwater silverside populations (Odontesthes hatcheri, Atherinopsidae) by introgressive hybridization with introduced O. bonariensis. Biol Invasions 19: 971-988.). Therefore, we suggest the older cladogenetic events that shaped freshwater assemblage in the southern Andean trichomycterids (Barber et al. 2011BARBER BR, UNMACK PJ, PÉREZ-LOSADA M, JOHNSON JB & CRANDALL KA. 2011. Different processes lead to similar patterns: a test of codivergence and the role of sea level and climate change in shaping a southern temperate freshwater assemblege. BMC Evol Biol 11: 343.) prevailed over the events that homogenized the populations of P. trucha and O. hatcheri (Ruzzante et al. 2006RUZZANTE DE, WALDE SJ, CUSSAC VE, DALEBOUT ML, SEIBERT J, ORTUBAY S & HABIT E. 2006. Phylogeography of the Percichthyidae (Pisces) in Patagonia: roles of orogeny, glaciation, and volcanism. Mol Ecol 15: 2949-2968., 2011, Crichigno et al. 2014CRICHIGNO SA, BATTINI MA & CUSSAC VE. 2014. Diet induces phenotypic plasticity of Percichthys trucha (Valenciennes, 1833) (Perciformes, Percichthyidae) in Patagonia. Zool Anzeiger 253: 192-202., Conte-Grand et al. 2015CONTE-GRAND C, SOMMER J, ORTÍ G & CUSSAC V. 2015. Populations of Odonthestes (Teleostei: Atheriniformes) in the Andean region of Southern South America: body shape and hybrid individuals. Neotrop Ichthyol 13: 1-14., Rueda et al. 2017RUEDA EC, MULLANEY KA, CONTE-GRAND C, HABIT EM, CUSSAC V & ORTÍ G. 2017. Displacement of native Patagonian freshwater silverside populations (Odontesthes hatcheri, Atherinopsidae) by introgressive hybridization with introduced O. bonariensis. Biol Invasions 19: 971-988.). Not surprisingly, another Patagonian species that showed an appreciable genetic structure at mitochondrial level, and evidenced refugia into the glaciated area, is Galaxias platei Steindachner, 1898 (Zemlak et al. 2008ZEMLAK TS, HABIT EM, WALDE SJ, BATTINI MA, ADAMS EDM & RUZZANTE DE. 2008. Across the southern Andes on fin: glacial refugia, drainage reversals and a secondary contact zone revealed by the phylogeographical signal of Galaxias platei in Patagonia. Mol Ecol 17: 5049-5061.). Its presence in South America has been dated not just before the Andes upraise in the Miocene (Ruzzante et al. 2008RUZZANTE DE, WALDE SJ, GOSSE JC, CUSSAC VE, HABIT E, ZEMLAK TS & ADAMS EDM. 2008. Climate Control on Ancestral Population Dynamics: Insight from Patagonian Fish Phylogeography. Mol Ecol 17: 2234-2244.; Zemlak et al. 2008ZEMLAK TS, HABIT EM, WALDE SJ, BATTINI MA, ADAMS EDM & RUZZANTE DE. 2008. Across the southern Andes on fin: glacial refugia, drainage reversals and a secondary contact zone revealed by the phylogeographical signal of Galaxias platei in Patagonia. Mol Ecol 17: 5049-5061.), but even before the separation of Australia and South America, in the Oligocene (Burridge et al. 2012BURRIDGE CP, MCDOWALL RM, CRAW D, WILSON MVH & WATERS JM. 2012. Marine dispersal as a pre-requisite for Gondwanan vicariance among elements of the galaxiid fish fauna. J Biogeogr 39: 306-321.). In light of our results, the historical taxonomic and nomenclatural arrangements of the relict southern Andean trichomycterids here investigated seem to have originated by their great intra-specific morphological variation, while at the same time by their high inter-specific morphological similarity. Both can be interpreted as consequences of sharing a recent common ancestor, as well as the great variety of microhabitat where these species live in lotic environments, from mountain rocky streams with rapid waters to slow-habitats with sandy bottoms (Chiarello-Sosa et al. 2018CHIARELLO-SOSA JM, BATTINI MA & BARRIGA JP. 2018. Latitudinal phenotypic variation in the southernmost trichomycterid, the catfish Hatcheria macraei: an amalgam of population divergence and environmental factors. Biol J Linn Soc 124: 718-731.). All in all, we confirmed both the great morphologic variation of H. macraei and the high genetic variation of H. macraei, T. areolatus and B. maldonadoi, to be congruent with hydrographic basins. We also revealed new boundaries to the currently known trans-Andean distribution of these relict trichomycterids and highlight the need of further integrative studies to continue improving our knowledge of the southern Andean trichomycterid diversity.

SUPPLEMENTARY MATERIAL

Table SI.

REFERENCES

  • ARRATIA G. 1976. Variaciones de las hipurapófisis en algunos peces siluriformes (Familia Trichomycteridae). An Mus Hist Nat Valparaíso Chile 9: 105-114.
  • ARRATIA G. 1981. Géneros de peces de aguas continentales de Chile. Mus Nac Hist Nat 34: 3-108.
  • ARRATIA G. 1983. Preferencias de hábitat de peces siluriformes de aguas continentales de Chile (Fam. Diplomystidae and Trichomycteridae). Stud Neotrop Fauna Environ 18: 217-237.
  • ARRATIA G & MENU-MARQUE S. 1981. Revision of the freshwater catfishes of the genus Hatcheria (Siluriformes, Trichomycteridae) with comentaries on ecology and biogeography. Zool Anz Jena 207: 88-111.
  • ARRATIA G, MENU-MARQUE S & ROJAS G. 1978. About Bullockia gen. nov., Trichomycterus mendozensis n. sp. and revision of the family Trichomycteridae (Pisces, Siluriformes). Stud Neotrop Fauna Environ 13: 157-194.
  • ARRATIA G, PEÑAFORT MB & MENU-MARQUE S. 1983. Peces de la región sureste de los Andes y sus probables relaciones biogeográficas actuales. Deserta 7: 48-107.
  • ARRATIA G & QUEZADA-ROMEGIALLI C. 2017. Understanding morphological variability in a taxonomic context in Chilean diplomystids (Teleostei: Siluriformes), including the description of a new species. PeerJ 5: e2991.
  • AVISE JC, ARNOLD J, BALL RM, BERMINGHAM E, LAMB T, NEIGEL JE, REEB CA & SAUNDERS NC. 1987. Intraspecific phylogeography: the mitochondrial DNA bridge between population genetics and systematics. Ann Rev Ecol Syst 18: 489-522.
  • AZPELICUETA MM & CIONE AL. 2016. A southern species of the tropical catfish genus Phractocephalus (Teleostei: Siluriformes) in the Miocene of South America. J South Am Earth Sci 67: 221-230.
  • BAIGÚN C & FERRIZ R. 2003. Distribution patterns of native freshwater fishes in Patagonia (Argentina). Org Divers Evol 3: 151-159.
  • BANDELT H-J, FORSTER P & RÖHL A. 1999. Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol 16: 37-48.
  • BARBER BR, UNMACK PJ, PÉREZ-LOSADA M, JOHNSON JB & CRANDALL KA. 2011. Different processes lead to similar patterns: a test of codivergence and the role of sea level and climate change in shaping a southern temperate freshwater assemblege. BMC Evol Biol 11: 343.
  • BARRIGA JP & BATTINI MA. 2009. Ecological significances of ontogenetic shifts in the stream-dwelling catfish, Hatcheria macraei (Siluriformes, Trichomycteridae), in a Patagonian river. Ecol Freshw Fish 18: 395-405.
  • BARRIGA JP, ESPINÓS NA, CHIARELLO-SOSA JM & BATTINI MA. 2013. The importance of substrate size and interstitial space in the microhabitat selection by the stream-dwelling catfish Hatcheria macraei (Actinopterygii, Trichomycteridae). Hydrobiologia 705: 191-206.
  • BICKFORD D, LOHMAN DJ, SODHI NS, NG PKL, MEIER R, WINKER K, INGRAM KK & DAS I. 2007. Cryptic species as a window on diversity and conservation. Trends Ecol Evol 22: 148-155.
  • BURRIDGE CP, MCDOWALL RM, CRAW D, WILSON MVH & WATERS JM. 2012. Marine dispersal as a pre-requisite for Gondwanan vicariance among elements of the galaxiid fish fauna. J Biogeogr 39: 306-321.
  • CADRIN SX. 2000. Advances in morphometric identification of fishery stock. Rev Fish Biol Fish 10: 91-112.
  • CAMPOS H et al. 1998. Categorías de conservación de peces nativos de aguas continentales de Chile. Bol Mus Nac Hist Nat 47: 101-122.
  • CHIARELLO-SOSA JM, BATTINI MA & BARRIGA JP. 2018. Latitudinal phenotypic variation in the southernmost trichomycterid, the catfish Hatcheria macraei: an amalgam of population divergence and environmental factors. Biol J Linn Soc 124: 718-731.
  • CIONE AL, AZPELICUETA MM, CASCIOTTA JR & DOZO MT. 2005. Tropical freshwater teleosts from Miocene beds of eastern Patagonia, southern Argentina. Geobios 38: 29-42.
  • COLIHUEQUE N, CORRALES O & YÁÑEZ M. 2017. Morphological analysis of Trichomycterus areolatus Valenciennes, 1846 from southern Chilean rivers using a truss-based system (Siluriformes, Trichomycteridae). ZooKeys 695: 135-152.
  • CONTE-GRAND C, SOMMER J, ORTÍ G & CUSSAC V. 2015. Populations of Odonthestes (Teleostei: Atheriniformes) in the Andean region of Southern South America: body shape and hybrid individuals. Neotrop Ichthyol 13: 1-14.
  • COSTA WJ, HENSCHEL E & KATZ AM. 2020. Multigene phylogeny reveals convergent evolution in small interstitial catfishes from the Amazon and Atlantic forests (Siluriformes: Trichomycteridae). Zool Scripta 49: 159-173.
  • CRICHIGNO SA, BATTINI MA & CUSSAC VE. 2012. Early morphological variation and induction of phenotypic plasticity in Patagonian pejerrey. Neotrop Ichthyol 10: 341-348.
  • CRICHIGNO S, CONTE-GRAND C, BATTINI M & CUSSAC V. 2013. Cephalic morphological variation in freshwater silversides Odontesthes hatcheri and Odontesthes bonariensis in Patagonia: introgression and ecological relationships. J Fish Biol 83: 542-559.
  • CRICHIGNO SA, BATTINI MA & CUSSAC VE. 2014. Diet induces phenotypic plasticity of Percichthys trucha (Valenciennes, 1833) (Perciformes, Percichthyidae) in Patagonia. Zool Anzeiger 253: 192-202.
  • CRICHIGNO SA, HATTORI RS, STRÜSSMANN CA & CUSSAC VE. 2016. Morphological comparison of wild, farmed and hybrid specimens of two South American silversides, Odontesthes bonariensis and Odontesthes hatcheri. Aquac Res 47: 1797-1808.
  • CUVIER G & VALENCIENNES M. 1832. Histoire naturelle des poissons. Vol 3, Paris, 500 p.
  • CUVIER G & VALENCIENNES M. 1846. Histoire naturelle des poissons. Vol 18. Paris, 505 p.
  • DATOVO A & BOCKMANN FA. 2010. Dorsolateral head muscles of the catfish families Nematogenyidae and Trichomycteridae (Siluriformes: Loricarioidei): comparative anatomy and phylogenetic analysis. Neotrop Ichthyol 8: 193-246.
  • DOZO MT, BOUZA P, MONTI A, PALAZZESI L, BARREDA V, MASSAFERRO G, SCASSO RA & TAMBUSSI CP. 2010. Late Miocene continental biota in Northeastern Patagonia (Península Valdés, Chubut, Argentina). Palaeogeogr Palaeoclim Palaeoecol 297: 100-109.
  • DE BUEN F. 1958. Ictiología. La familia Ictaluridae nueva para la fauna aclimatada de Chile y algunas consideraciones sobre los Siluroidei indígenas. Inv Zool Chilenas 4: 46-158.
  • DYER BS. 2000. Systematic review and biogeography of the freshwater fishes of Chile. Estud Oceanol 19: 77-98.
  • EIGENMANN CH. 1909. The fresh water fishes of Patagonia and an examination of the archiplata-archelenis theory. In: Reports of the Princeton University Expeditions to Patagonia, 1896-1899. Princeton, NJ, USA, p. 225-374.
  • EIGENMANN CH. 1927. The fresh-water fishes of Chile. Mem Nat Acad Sci 22: 1-63.
  • FERNANDEZ L, ARROYAVE J & SCHAEFER SA. 2021. Emerging patterns in phylogenetic studies of trichomycterid catfishes (Teleostei, Siluriformes) and the contribution of Andean diversity. Zoologica Scripta 50: 318-336.
  • FRICKE R, ESCHMEYER WN & VAN DER LAAN R. 2021. (http://researcharchive.calacademy.org/research/ichthyology/catalog/fishcatmain.asp). Electronic version accessed 7 July 2021.
    » http://researcharchive.calacademy.org/research/ichthyology/catalog/fishcatmain.asp
  • FROESE R & PAULY D. 2021. FishBase (www.fishbase.org). Electronic version accessed 7 July 2021.
    » www.fishbase.org
  • GARCÍA-DÁVILA C, DUPONCHELLE F, CASTRO-RUIZ D, VILLACORTA J, QUÉROUIL S, CHOTA-MACUYAMA W, NÚÑEZ J, RÖMER U, CARVAJAL-VALLEJOS F & RENNO J-F. 2013. Molecular identification of a cryptic species in the Amazonian predatory catfish genus Pseudoplatystoma (Bleeker, 1962) from Peru. Genetica 141: 347-358.
  • GIRARD C. 1855. Fishes. In: The U.S. Naval astronomical expedition to the southern hemisphere, during the years 1849-1852. Vol II. Washington DC, USA, p. 230-253.
  • GOMES LC, PESSALI TC, SALES NG, POMPEU PS & CARVALHO DC. 2015. Integrative taxonomy detects cryptic and overlooked fish species in a neotropical river basin. Genetica 143: 581-588.
  • GUIMARÃES EC, DE BRITO PS, BRAGANCA PHN, SANTOS JP, KATZ AM, CARVALHO COSTA LF & OTTONI FP. 2020. Integrative taxonomy reveals two new cryptic species of Hyphessobrycon Durbin, 1908 (Teleostei: Characidae) from the Maracaçumé and middle Tocantins River basins, Eastern Amazon region. Eur J Taxonomy 723: 77-107.
  • HABIT E, VICTORIANO P & CAMPOS H. 2005. Ecología trófica y aspectos reproductivos de Trichomycterus areolatus (Pisces, Trichomycteridae) en ambientes lóticos artificiales. Rev Biol Trop 53: 195-210.
  • HALL T. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl Acids Symposium Series 41: 95-98.
  • HARE MP. 2001. Prospects for nuclear gene phylogeography. Trens Ecol & Evol 16: 700-706.
  • HANCOCK TV & HEDRICK MS. 2018. Physiological vagility affects population genetic structure and dispersal and enables migratory capacity in vertebrates. Comp Biochem Physiol Part A Mol & Integr Physiol 223: 42-51.
  • HILLMAN SS, DREWES RC, HEDRICK MS & HANCOCK TV. 2014. Physiological vagility and its relationship to dispersal and neutral genetic heterogeneity in vertebrates. J Exp Biol 217: 3356-3364.
  • IVANOVA NV, ZEMLAK TS, HANNER RH & HEBERT PDN. 2007. Universal primer cocktails for fish DNA barcoding. Mol Ecol Notes 7: 544-548.
  • JIRSOVÁ, D, ŠTEFKA J, BLAžEK R, MALALA JO, LOTULIAKOU DE, MAHMOUD ZN & JIRKU M. 2019. From taxonomic deflation to newly detected cryptic species: Hidden diversity in a widespread African squeaker catfish. Sci Rep 9: 15748.
  • KATZ AM, BARBOSA MA, OLIVEIRA MATTOS JL & MOREIRA DA COSTA WJE. 2018. Multigene analysis of the catfish genus Trichomycterus and description of a new South American trichomycterine genus (Siluriformes, Trichomycteridae). Zoosyst Evol 94: 557-566.
  • KEELEY ER, PARKINSON EA & TAYLOR EB. 2007. The origins of ecotypic variation of rainbow trout: a test of environmental vs. genetically based differences in morphology. J Evol Biol 20: 725-736.
  • LEIGH JW & BRYANT D. 2015. POPART: full-feature software for haplotype network construction. Methods Ecol Evol 6: 1110-1116.
  • LIBRADO P & ROZAS J. 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25: 1451-1452.
  • MANRIQUEZ A, HUAQUÍN L, ARELLANO M & ARRATIA G. 1988. Aspectos reproductivos de Trichomycterus areolatus Valenciennes, 1846 (Pisces: Teleostei: Siluriformes) en Rio Angostura, Chile. Stud Neotrop Fauna Environ 23: 89-102.
  • MARTIN AP & BERMINGHAM W. 2000. Regional endemism and cryptic species revealed by molecular and morphological analysis of a widespread species of Neotropical catfish. Proc R Soc Lond B 267: 1135-1141.
  • MARTÍNEZ OA & KUTSCHKER A. 2011. The ‘Rodados Patagónicos’ (Patagonian shingle formation) of eastern Patagonia: environmental conditions of gravel sedimentation. Biol J Linn Soc 103: 336-345.
  • MAYR E. 1942. Systematics and the origin of species. NY, USA.
  • MAYR E. 1982. Speciation and macroevolution. Evolution 36: 1119-1132.
  • MELO BF, OCHOA LE, VARI RP & OLIVEIRA C. 2016. Cryptic species in the Neotropical fish genus Curimatopsis (Teleostei, Characiformes). Zool Scripta 45: 650-658.
  • MILLER MA, PFEIFFER W & SCHWARTZ T. 2010. “Creating the CIPRES Science Gateway for inference of large phylogenetic trees” in Proceedings of the Gateway Computing Environments Workshop (GCE), 14 Nov 2010, New Orleans, LA p 1-8.
  • MOORE WS. 1995. Inferring phylogenies from mtDNA variation: mitochondrial-gene trees versus nuclear-gene trees. Evolution 49: 718-726.
  • NORUSIS MJ. 1986. SpSS/PC+ advanced statistics. SPSS Inc., Chicago, Illinois, 204 p.
  • OCHOA LE, ROXO FF, DONASCIMIENTO C, SABAJ MH, DATOVO A, ALFARO M & OLIVEIRA C. 2017. Multilocus analysis of the catfish family Trichomycteridae (Teleostei: Ostariophysi: Siluriformes) supporting a monophyletic Trichomycterinae. Mol Phylogenet Evol 115: 71-81.
  • OCHOA LE ET AL. 2020. Phylogenomic analysis of trichomycterid catfishes (Teleostei: Siluriformes) inferred from ultraconserved elements. Sci Rep 10: 1-15.
  • PARDO R. 2002. Morphologic differentiation of Trichomycterus areolatus Valenciennes 1846 (Pisces: Siluriformes: Trichomycteridae) from Chile. Gayana 66: 203-205.
  • PARDO R, SCOTT S & VILA I. 2005. Shape analysis in Chilean species of Trichomycterus (Osteichthyes: Siluriformes) using geometric morphometry. Gayana 69: 180-183.
  • PEREIRA LHG, CASTRO JRC, VARGAS PMH, GOMEZ JAM & OLIVEIRA C. 2021. The use of an integrative approach to improve accuracy of species identification and detection of new species in studies of stream fish diversity. Genetica 149: 103-116.
  • POSADA D. 2008. jModelTest: Phylogenetic model averaging. Mol Biol Evol 25: 1253-1256.
  • PREMOLI AC, KITZBERGER T & VEBLEN TT. 2000. Isozyme variation and recent biogeographical history of the long-lived conifer Fitzroya cupressoides. J Biogeogr 27: 251-260.
  • PREMOLI AC, SOUTO CP, ROVERE AE, ALLNUT TR & NEWTON AC. 2002. Patterns of isozyme variation as indicators of biogeographic history in Pilgerodendron uviferum (D. Don) Florín. Div Distr 8: 57-66.
  • RABASSA J, CORONATO AM & MARTÍNEZ O. 2011. Late Cenozoic glaciations in Patagonia and Tierra del Fuego: an updated review. Biol J Linn Soc 103: 316-335.
  • RAMOS VA & GHIGLIONE MC. 2008. Tectonic Evolution of the Patagonian Andes. In: Developments in Quaternary Sciences. Elsevier, p. 57-71.
  • RINGUELET RA. 1975. Zoogeografía y ecología de los peces de aguas continentales de la Argentina y consideraciones sobre las áreas ictiológicas de América del Sur. Ecosur 2: 1-122.
  • RINGUELET RA, ARÁMBURU RH & DE ARÁMBURU AA. 1967. Los peces argentinos de agua dulce. Comisión de Investigaciones Científicas de la Provincia de Buenos Aires, Buenos Aires, 602 p.
  • ROHLF FJ. 1996. Morphometric spaces, shape components, and the effects of linear transformations. In: Marcus LF et al. (Eds), Advances in Morphometrics. Proceedings of the 1993 NATO Advanced Studies Institute on Morphometrics in Il Ciocco. New York, Plenum Publishing Corp.
  • ROHLF FJ & MARCUS LF. 1993. A revolution in morphometrics. Trends Ecol Evol 8: 129-132.
  • ROHLF FJ & SLICE DE. 1990. Extensions of the Procrustes method for the optimal superimposition of landmarks. Syst Zool 39: 40-59.
  • RONQUIST F & HUELSENBECK JP. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572-1574.
  • RUBINOFF D & HOLLAND BS. 2005. Between two extremes: mitochondrial DNA is neither the panacea nor the nemesis of phylogenetic and taxonomic inference. Syst Biol 54: 952-961.
  • RUEDA EC, MULLANEY KA, CONTE-GRAND C, HABIT EM, CUSSAC V & ORTÍ G. 2017. Displacement of native Patagonian freshwater silverside populations (Odontesthes hatcheri, Atherinopsidae) by introgressive hybridization with introduced O. bonariensis. Biol Invasions 19: 971-988.
  • RUZZANTE DE, WALDE SJ, CUSSAC VE, DALEBOUT ML, SEIBERT J, ORTUBAY S & HABIT E. 2006. Phylogeography of the Percichthyidae (Pisces) in Patagonia: roles of orogeny, glaciation, and volcanism. Mol Ecol 15: 2949-2968.
  • RUZZANTE DE, WALDE SJ, GOSSE JC, CUSSAC VE, HABIT E, ZEMLAK TS & ADAMS EDM. 2008. Climate Control on Ancestral Population Dynamics: Insight from Patagonian Fish Phylogeography. Mol Ecol 17: 2234-2244.
  • RUZZANTE DE, WALDE SJ, MACCHI PJ, ALONSO M & BARRIGA JP. 2011. Phylogeography and phenotypic diversification in the Patagonian fish Percichthys trucha: the roles of Quaternary glacial cycles and natural selection. Biol J Linn Soc 103: 514-529.
  • SENAY C, BOISCLAIR D & PERES-NETO PR. 2015. Habitat-based polymorphism is common in stream fishes. J Anim Ecol 84: 219-227.
  • SHAO W-H, CHENG J-L & ZHANG E. 2021. Eight in one: hidden diversity of the bagrid catfish Tachysurus albomarginatus s.l. (Rendhal, 1928) widespread in lowlands of South China. Front Genetics 12: 713793.
  • SÉRSIC AN, COSACOV A, COCUCCI AA, JOHNSON LA, POZNER R, AVILA LJ, SITES JW JR & MORANDO M. 2011. Emerging phylogeographic patterns of plants and terrestrial vertebrates from Patagonia. Biol J Linn Soc 103: 475-494.
  • SOKAL RR & CROVELLO TJ. 1970. The Biological Species Concept: a critical evaluation. Am Nat 104: 127-153.
  • SWOFFORD DL. 2003. PAUP*. Phylogenetic analysis using parsimony and other methods. Version 4.0a. Sinauer. Sunderland, MA.
  • TCHERNAVIN VV. 1944. A revision of some Trichomycterinae based on material preserved in the British Museum (Natural History). Proc Zool Soc Lon 114: 234-275.
  • UNMACK PJ, BENNIN AP, HABIT EM, VICTORIANO PF & JOHNSON JB. 2009a. Impact of ocean barriers, topography, and glaciation on the phylogeography of the catfish Trichomycterus areolatus (Teleostei: Trichomycteridae) in Chile. Biol J Linn Soc 97: 876-892.
  • UNMACK PJ, HABIT EM & JOHNSON JB. 2009b. New records of Hatcheria macraei (Siluriformes, Trichomycteridae) from Chilean province. Gayana 73: 102-110.
  • UNMACK PJ, BARRIGA JP, BATTINI MA, HABIT EM & JOHNSON JB. 2012. Phylogeography of the catfish Hatcheria macraei reveals a negligible role of drainage divides in structuring populations. Mol Ecol 21: 942-959.
  • ZEMLAK TS, HABIT EM, WALDE SJ, BATTINI MA, ADAMS EDM & RUZZANTE DE. 2008. Across the southern Andes on fin: glacial refugia, drainage reversals and a secondary contact zone revealed by the phylogeographical signal of Galaxias platei in Patagonia. Mol Ecol 17: 5049-5061.

Publication Dates

  • Publication in this collection
    08 May 2023
  • Date of issue
    2023

History

  • Received
    19 July 2021
  • Accepted
    15 Apr 2022
Academia Brasileira de Ciências Rua Anfilófio de Carvalho, 29, 3º andar, 20030-060 Rio de Janeiro RJ Brasil, Tel: +55 21 3907-8100 - Rio de Janeiro - RJ - Brazil
E-mail: aabc@abc.org.br