Phylogeographic analysis of genus Herichthys (Perciformes: Cichlidae), with descriptions of Nosferatu new genus and H. tepehua n. sp.

Patricia Ornelas García; Pronatura Noreste; María Elena García Ramírez; Mauricio De la Maza Benignos

The genus Herichthys is widely considered to be the monophyletic representative of Cichlidae in northeastern Mexico and southern Texas. It is also the northernmost distributed genus of Neotropical Cichlids. Its distribution stretches over an area that is characterized by an intricate geologic and climatic history that affected its temporal and spatial diversification north of the Trans- Mexican Volcanic-Belt. We access the evolutionary history of the genus Herichthys based on a phylogenetic reconstruction using a mitochondrial fragment of gene Cox1. We evaluate its morphological variation, its correspondence with molecular differentiation and suggest a biogeographical scenario based on a molecular clock and demographic history. Furthermore, we describe Nosferatu new genus, composed of Nosferatu pame (assigned as type species),N. molango,N. pratinus, N. bartoni, N. labridens, N. pantostictus, and N. steindachneri. Genus is characterized by a transition to prolongation in the size of the symphysial pair of teeth relative to that of the other teeth in the outer row of the upper jaw; breeding pigmentation that consists of darkening of ventral area extending over nostrils, opercular series, or pectoral fins; depressed dorsal fin rarely expands beyond anterior third of caudal fin; and an elongated, elastic, smooth caecum adhered to a saccular stomach. We also describe Herichthys tepehua n. sp. found in the Pantepec, Cazones, Tenixtepec, Tecolutla, and Solteros rivers, in Veracruz, Mexico. Moreover, we provide re-descriptions for some of the species in Herichthys and propose a biogeographic hypothesis for both genera, based on available information on the geological and climate history of the area of study, associated to dating retrieved in our phylogenetic analysis.

Phylogeographic analysis of genus Herichthys (Perciformes: Cichlidae), with descriptions of Nosferatu new genus and H. tepehua n. sp. Mauricio De la Maza-Benignos, Claudia Patricia Ornelas-García, María de Lourdes Lozano-Vilano, María Elena García-Ramírez, et al. Hydrobiologia The International Journal of Aquatic Sciences ISSN 0018-8158 Hydrobiologia DOI 10.1007/s10750-014-1891-8 1 23 Your article is protected by copyright and all rights are held exclusively by Springer International Publishing Switzerland. This eoffprint is for personal use only and shall not be self-archived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”. 1 23 Author's personal copy Hydrobiologia DOI 10.1007/s10750-014-1891-8 ADVANCES IN CICHLID RESEARCH Phylogeographic analysis of genus Herichthys (Perciformes: Cichlidae), with descriptions of Nosferatu new genus and H. tepehua n. sp. Mauricio De la Maza-Benignos • Claudia Patricia Ornelas-Garcı́a • Marı́a de Lourdes Lozano-Vilano • Marı́a Elena Garcı́a-Ramı́rez • Ignacio Doadrio Received: 30 December 2013 / Accepted: 13 April 2014 Ó Springer International Publishing Switzerland 2014 Abstract The genus Herichthys is widely considered to be the monophyletic representative of Cichlidae in northeastern Mexico and southern Texas. It is also the northernmost distributed genus of Neotropical Cichlids. Its distribution stretches over an area that is characterized by an intricate geologic and climatic history that affected its temporal and spatial diversification north of the TransMexican Volcanic-Belt. We access the evolutionary history of the genus Herichthys based on a phylogenetic reconstruction using a mitochondrial fragment of gene Cox1. We evaluate its morphological variation, its correspondence with molecular differentiation and suggest a biogeographical scenario based on a molecular clock and demographic history. Furthermore, we describe Nosferatu new genus, composed of Nosferatu Guest editors: S. Koblmüller, R. C. Albertson, M. J. Genner, K. M. Sefc & T. Takahashi / Advances in Cichlid Research: Behavior, Ecology and Evolutionary Biology Electronic supplementary material The online version of this article (doi:10.1007/s10750-014-1891-8) contains supplementary material, which is available to authorized users. M. De la Maza-Benignos Pronatura Noreste, A.C., Loma Grande 2623, Col. Loma Larga, 64710 Monterrey, N.L, Mexico C. P. Ornelas-Garcı́a (&) Departamento de Zoologı́a, Universidad Nacional Autónoma de México, Mexico, D.F, Mexico e-mail: patriciaornelasg@gmail.com pame (assigned as type species), N. molango, N. pratinus, N. bartoni, N. labridens, N. pantostictus, and N. steindachneri. Genus is characterized by a transition to prolongation in the size of the symphysial pair of teeth relative to that of the other teeth in the outer row of the upper jaw; breeding pigmentation that consists of darkening of ventral area extending over nostrils, opercular series, or pectoral fins; depressed dorsal fin rarely expands beyond anterior third of caudal fin; and an elongated, elastic, smooth caecum adhered to a saccular stomach. We also describe Herichthys tepehua n. sp. found in the Pantepec, Cazones, Tenixtepec, Tecolutla, and Solteros rivers, in Veracruz, Mexico. Moreover, we provide re-descriptions for some of the species in Herichthys and propose a biogeographic hypothesis for both genera, based on available information on the geological and climate history of the area of study, associated to dating retrieved in our phylogenetic analysis. Keywords Nosferatu Herichthys Cox1 Phylogeny Phylogeography M. L. Lozano-Vilano M. E. Garcı́a-Ramı́rez Laboratorio de Ictiologı́a, Facultad de Ciencias Biológicas, UANL, Ap. Postal 425 San Nicolás de los Garza, 66450 Mexico, N.L, Mexico I. Doadrio Departamento de Biodiversidad y Biologı́a Evolutiva, Museo Nacional de Ciencias Naturales, CSIC, José Gutiérrez Abascal 2, 28006 Madrid, Spain 123 Author's personal copy Hydrobiologia Introduction The tribe Heroini forms the second largest monophyletic group of Neotropical cichlids, with distributions ranging from the province of Buenos Aires in Argentina, South America to the Rio Grande Basin in North America (Kullander, 1998; Concheiro-Pérez et al., 2006; De la Maza-Benignos & Lozano-Vilano, 2013). This tribe includes the Herichthyines, which are represented by the genera Paraneetroplus and Vieja as reviewed by McMahan et al. (2010), Herichthys, Herotilapia, Paratheraps, Theraps, Tomocichla, and Thorichthys, together with other ‘‘unnamed genera’’ (Concheiro-Pérez et al., 2006; Řı́čan et al., 2008, 2012; De la Maza-Benignos & Lozano-Vilano, 2013). During the last two decades, significant progress in molecular phylogenetic studies of the Neotropical Cichlids has been made (e.g., Roe et al., 1997; Martin & Bermingham, 1998; Farias et al., 1999, 2000, 2001; Sides & Lydeard, 2000; Hulsey et al., 2004; LópezFernández et al., 2005a, b; Concheiro-Pérez et al., 2006 ;Chakrabarty, 2007; Musilová et al., 2008; Řı́čan et al., 2008, 2012; López-Fernández et al., 2010); still, their intrageneric relationships remain only partially understood. Studies with few exceptions (e.g., Schmitter-Soto, 2007; Kullander et al., 2010; McMahan et al., 2010) have largely addressed intergeneric relationships and evolutionary history at higher taxonomic levels. Some of the factors that have so far hindered both phylogenetic and morphological analysis at the intrageneric level include high taxon diversity, wide distribution ranges, phenotypic plasticity, and morphological homoplasies, as well as low morphometric variation and absence of genetic resolution across some of the taxa that ‘‘make character sets inadequate for revealing relationships’’ (Martin & Bermingham, 1998; López-Fernández et al., 2010; De la MazaBenignos & Lozano-Vilano, 2013). The genus Herichthys is widely considered as the monophyletic representative of Cichlidae in northeastern Mexico and southern Texas (Hulsey et al., 2004; Concheiro-Pérez et al., 2006; Řı́čan et al., 2008; De la Maza-Benignos & Lozano-Vilano, 2013). It is also the northernmost distributed genus of Neotropical Cichlids. Herichthys are found from the lower tributaries of the Rio Grande, excluding the Rio Conchos Basin in 123 Chihuahua, to the Misantla River Basin in Veracruz, Mexico (De la Maza-Benignos & Lozano-Vilano, 2013), along the structural Burgos and Tampico– Misantla basins, over an area that is characterized by an intricate geologic (Byerly, 1991; Ferrari et al., 1999) and climatic history (Metcalfe et al., 2000; Ravelo et al., 2004; Metcalfe, 2006). Its distribution range also corresponds to the inland portion of the Eastern Mexico Continental Shelf and Slope Geological Province, which extends from 26° to 20°N (Antoine, 1972) and whose southernmost land point is known as Punta del Morro (PDM). PDM in turn is the easternmost extension where the Trans-Mexican Volcanic-Belt (TMVB) meets the Gulf of Mexico (Contreras-Balderas et al., 1996). The origin of the TMVB is related to Neogene subduction of the Cocos and Rivera plates beneath the southwestern margin of the North American plate on the Pacific coast (Avto et al., 2007). Climate change in the last 4 million years includes the end of the warm period (Ravelo et al., 2004), with higher winter precipitation, cooler and wetter conditions than today, during the Pleistocene, and early Holocene in northern Mexico (Metcalfe et al., 2000; Metcalfe, 2006). In this paper, we propose that both neo-volcanism and climate change played major roles in the temporal and spatial diversification of the cichlid fishes found north of PDM. In a previous taxonomic hypothesis, De la MazaBenignos & Lozano-Vilano (2013) suggested the separation of Herichthys into two distinct groups, the ‘‘labridens assemblage’’ (described herein as Nosferatu new genus) and the ‘‘cyanoguttatus assemblage’’ (herein referred to as true Herichthys). The separation is congruent with phylogenetic hypothesis developed by Hulsey et al. (2004), Concheiro-Pérez et al. (2006), and López-Fernández et al. (2010). The aim of this study was to review the evolutionary history of the genus Herichthys by accessing its phylogenetic relationships, as well as temporal and biogeographic patterns. To achieve this goal, we characterized its morphological variations as well as its molecular structure using the mitochondrial gene, Cox1. We reviewed the taxonomy of the group, described a new genus, described a new species, and re-described some of the species of Herichthys. Moreover, we propose a biogeographic hypothesis for the group. Author's personal copy Hydrobiologia Methods Specimens examined in the present study were collected throughout the historical range of Herichthys (Kullander, 2003; Miller et al., 2005; De la MazaBenignos & Lozano-Vilano, 2013) (Fig. 1) by angling and cast nets, and include representatives of all the nominal species that have been assigned to the genus Herichthys at various times. Samples were limited to 20 individuals per locality for each sampling event. Coordinates for the localities are given in decimal degrees. This sample size was based on the current conservation status of the species as determined by NOM-059 (SEMARNAT, 2010). The collected specimens were fixed in 10% formalin, then transferred to 50% isopropanol, and finally deposited in the Colección Ictiológica de la Facultad de Ciencias Biológicas de la Universidad Autónoma de Nuevo León (UANL). Tissue samples were taken from pectoral fin clips. Fin clips were preserved in 95% ethanol and deposited in the Museo Nacional de Ciencias Naturales of Madrid, Spain (MNCN). Morphometry and meristics Thirteen counts and 42 standard body measurements were obtained from 279 mature representatives of all the known nominal species assigned to Herichthys (Kullander, 2003), except H. minckleyi, plus 95 mature specimens of Nosferatu new genus (7 species), except N. steindachneri and N. bartoni, as interpreted by De la Maza-Benignos & Lozano-Vilano (2013) (Supplementary Material 1). Measurements (in millimeters) in the diagnosis and descriptions are expressed as percent of the standard length (SL) or head length (HL) (Álvarez, 1970; Taylor & Miller, 1983). Three specimens of each population, except for N. steindachneri (1); N. bartoni (1 pharyngeal plate), and H. minckleyi (0), were dissected to remove the digestive tracts and lower pharyngeal plates. The number of teeth along the posterior margin and median axis of the occlusal surface was counted. Counts followed the criteria of Taylor & Miller (1983), Snoeks (1994), and Chakrabarty (2007). Data on ecology and distribution were collected during our field observations and collection trips (2005–2007). Principal components analysis (PCA) followed by discriminant function analysis (DFA) were performed on the morphometric dataset using the software StatistiXL vers. 1.7. Measurements were subject to standardization in order to remove the allometry by effect of size using the equation Ms = Mo (Ls/Lo)b, where Ms is the standardized measurement, Mo is the measured character length (mm), Ls is the overall (arithmetic) mean standard component (SL or HL) for all individuals from all populations of each taxon, Lo is the standard component of specimen, and ‘‘b’’ for each character was estimated using the non-linear equation, M = aLb as the slope of the regression of log Mo on log Lo (Elliott et al., 1995; Ruiz-Campos et al., 2003). In addition, we examined (a) the morphology of observable characters (i.e., dentition of the oral jaws and pharyngeal plate, body contour, and fins shapes, among others); (b) breeding pigmentation [as observed by Kullander (1996)]; (c) anatomy of the digestive tract, and (d) ecology. DNA extraction, PCR amplification, and sequencing Genomic DNA was isolated from fin clips using standard proteinase K and phenol/chloroform extraction methods (Sambrook et al., 1989) and stored at 4°C. The DNA from 62 fish, representing all sampling sites (Table 1 of Supplementary Material 2), was amplified for the cytochrome c oxidase subunit I gene (Cox1, 585 bp) via polymerase chain reaction (PCR). The primers used for Cox1 were FISHF1-F 50 TCAAC CAACCACAAAGACATTGGCAC and FISHR1-R 30 TAGACTTCTGGGTGGCCAAAGAATCA (Ward et al., 2005). The amplification process was conducted under the following temperature conditions: 95°C (5 min), 35 cycles at 94°C (45 s), 54°C (1 min), 72°C (90 s), and 72°C (5 min). PCRs were performed in 10-ll reactions containing 0.4 mM of each primer, 0.2 mM of each dNTP, 2 mM MgCl2, 1 unit of Taq DNA polymerase (Invitrogen), and 10 ng of template DNA. PCR products were run on 1.0% agarose gels to confirm amplification and purified with the EXOSAP-IT PCR Product Clean-Up (Usb) kit. Both strands were sequenced on an ABI PRISM 3700 DNA automated sequencer (Applied Biosystems). Data analysis Chromatograms and alignments were visually checked and verified in MEGA5 (Tamura et al., 2011). All our phylogenetic inferences were carried 123 123 Author's personal copy Fig. 1 Distribution and sampling localities for Herichthys and Nosferatu new genus Hydrobiologia Author's personal copy Hydrobiologia out with the Cox1 data fragment, using a separate bestfit model for each codon position by gene. The evolutionary model of nucleotide substitution was estimated using an Akaike’s corrected information criterion (AICc) and a Bayesian information criterion (BIC), as implemented in the program jModeltest (Posada, 2008). Haplotypes from the complete dataset composed of 149 individuals, including 146 ingroup individuals (69 from this study ? 77 sequences from GenBank) and 9 outgroup sequences from GenBank were estimated using DnaSP 5.0 (Librado & Rozas, 2009). The haplotypes were constructed excluding sites with missing data for a total of 352 bp. A median joining network illustrating haplotype frequencies was generated using NETWORK v 4.5.1.6 (Bandelt et al., 1999). A phylogenetic hypothesis was constructed under maximum likelihood (ML) and implemented with RAXML, which uses a rapid hill-climbing algorithm (Stamatakis, 2006). The program performs a heuristic search with a general time-reversible (GTR) model that allows data partitioning and produces likelihood values using GTRCAT; a GTR approximation with optimization of individual per-site substitution rates and classification of those individual rates into a certain number of rate categories. To reconstruct the ML tree, we selected GTRMIX as a nucleotide substitution model that makes RAXML perform a tree inference (search for a good topology) under the GTRCAT model. When the analysis was completed in the GTRMIX model, RAXML switched to GTRGAMMA, and evaluated the final tree topology that yielded stable likelihood values. We set three codon partitions within Cox1 gene, performed 100 inferences in each analysis, and found the best ML tree by comparing final likelihoods among them. To evaluate the robustness of the internal branches of the ML tree, 100 bootstrap replications for the dataset were calculated. The Bayesian inference (BI) was carried out using MrBayes version 3.1.2 (Huelsenbeck & Ronquist, 2001) using the best-fit model in BIC by codon partition. The BI runs were performed using eight Markov chain Monte Carlo (MCMC), 10 million generations, sampling every thousands of steps. The first 1,000 trees were discarded as burn-in. We used the program Tracer v1. 4 (Rambaut & Drummond, 2007) to assess run convergence and determine burn-in. Divergence times among the main mitochondrial lineages were estimated using a Bayesian-coalescence approach as implemented in BEAST 1.6.1 (Drummond & Rambaut, 2007). For the analysis, we considered the mtDNA gene (Cox1) and used the matrix of 31 haplotypes and 585 bp (Table 2 of Supplementary Material 2). We applied an uncorrelated lognormal-relaxed molecular clock, using the SRD06 model of nucleotide substitution partitioning the nucleotide data by codon position and allowing third codon positions to differ from the other two in transition bias, substitution rate, and shape of the gamma distribution of rate heterogeneity (Shapiro et al., 2006). Due to the absence of fossil records or geological data, the age estimates were calibrated using an uniform prior distribution for the mean rate parameter, with a mean mutation rate of 0.8%/Mya and lower and upper values of 0.5–1.2%/Mya with a lower and upper values of 0.5–1.2% per million years based on what has been reported in other freshwater fish fauna for mitochondrial loci (Murphy et al., 1999; Mateos et al., 2002; Perdices et al., 2002, 2005; Doadrio & Dominguez, 2004; Doadrio & Perdices, 2005; Concheiro-Pérez et al., 2006; Hrbek et al., 2007; Ornelas-Garcı́a et al., 2008). A MCMC test was run for 30 million generations to optimize the scale factors of the priori function. The final MCMC chain was run twice for 20 million generations sampled every 2,000 generations. We checked for burn-in, convergence, and stationarity of the different analyses in Tracer 1.5. Measures of effective sample sizes (ESS) were used to determine the statistical significance of each parameter, where in most cases, it was determined to be higher than 200. Finally, we used the combined results in the BEAST module Log Combiner 1.6.1 after burn-in. Historical demography Patterns of historical demography were inferred from estimates of the effective population size over time using the Bayesian Skyline Plot (BSP) method, as implemented in BEAST v. 1.5.4 (Drummond & Rambaut, 2007). This method estimates a distribution of effective population sizes through time via MCMC procedures, by moving backward until the time of the most recent common ancestor is reached. We applied ten grouped coalescent intervals (m), and priors for the phylogenetic model. We used the HKY ? C model rate heterogeneity across all branches, partitioning by codon positions, separating third from first and second 123 Author's personal copy Hydrobiologia 6.000 PC2 12.45% 4.000 Herichthys Nosferatu 2.000 0.000 -2.000 -4.000 -6.000 -8.000 -10.000 -5.000 0.000 5.000 10.000 PC1 41.02% Fig. 2 Plot of scores on the first morphometric principal component and the second morphometric principal component of pooled material of Herichthys and Nosferatu new genus positions, assuming a strict molecular clock, and using an uniform rate with an initial value of 0.8% mutation rate per My, with lower and upper intervals of 0.5–1.2%, respectively (Doadrio & Perdices, 2005; Concheiro-Pérez et al., 2006; Ornelas-Garcı́a et al., 2008). Markov chains were run for 10 million generations and were sampled every 1,000 generations, with 10% of the initial samples discarded as burn-in. Results Morphological analysis In the pooled PC analysis of 42 standardized morphometric variables obtained of 279 specimens of true Herichthys, plus 95 specimens of Nosferatu new genus excluding N. bartoni, N. steindachneri, and H. minckleyi, the plot of scores on PC1 against PC2 (which explains 44.34% of the total variation in shape among the specimens) provided good separation between the two clusters of points, each corresponding to specimens of Herichthys and of Nosferatu new genus (Fig. 2; Supplementary Material 1). Pelvic fin base, pectoral fin base, body depth, and HL were the variables that most contributed to the variance in PC1. Jaw length, distance from dorsal fin origin to anal fin origin, distance from anal fin origin to hypural base, and anal fin base most contributed to the variance in PC2 (Table 1). Furthermore DFA of the morphometric database correctly classified 99% of Herichthys and 100% of Nosferatu 123 new genus (Table 2), clearly supporting the morphological separation of the two. In the pooled PC analysis of the standardized morphometric dataset of H. cyanoguttatus, H. teporatus, and H. carpintis, the plot of scores on PC1 against PC2 (which explains 41.21% of the total variation in shape among the specimens) provided rough separation between H. cyanoguttatus and H. carpintis, but H. teporatus overlapped with both species (Fig. 3). Snout width, distance from dorsal fin origin to anal fin origin, distance from rostral tip to anal fin origin, and predorsal distance were the variables that most contributed to the variance in PC1. Interorbital width, snout length, distance from rostral tip to pectoral fin origin and body depth most contributed to the variance in PC2 (Table 3). In the pooled PC analysis of the standardized morphometric dataset of H. carpintis, H. tepehua n. sp., and H. deppii, the plot of scores on PC1 against PC2 (which explains 47.88% of the total variation in shape among the specimens) provided good separation between H. carpintis and H. deppii, but H. tepehua n. sp. formed a cluster that overlapped with the other two (Fig. 4). Distance from rostral tip to anal fin origin, eye diameter, distance from dorsal fin origin to anal fin origin, and cheek depth were the variables that most contributed to the variance in PC1. Body depth, distance from rostral tip to pectoral fin origin, anal fin base, and distance from anal fin origin to hypural base most contributed to the variance in PC2 (Table 4). Further, DFA for the pooled morphometric database of H. deppii, H. tepehua n. sp., H. carpintis, H. tamasopoensis, H. teporatus, and H. cyanoguttatus provided good separation between H. deppii and H. tepehua n. sp., placing them phenetically closer to each other in the upper-right quadrant of the discriminant plot, and more distant from H. carpintis and H. tamasopoensis (grouped separate in the left quadrants of the discriminant plot) and from H. cyanoguttatus and H. teporatus (grouped separate in the lower quadrants of the discriminant plot) (Fig. 5). Morphological results also supported the separation of true Herichthys into three distinct geomorphological groups: the solid-colored Herichthys found south of Sierra Tantima (i.e., H. deppii and H. tepehua n. sp.); the iridescent pearl marked Herichthys found in the Pánuco Basin (i.e., H. carpintis and H. tamasopoensis); and the iridescent pearl marked Herichthys found north of Sierra de Tamaulipas (i.e., H. teporatus and H. cyanoguttatus) (Fig. 5). Author's personal copy Hydrobiologia Table 1 Component loadings on the first seven principal components of the pooled morphometric data for Herichthys (n = 279) and Nosferatu (n = 95) Variable PC 1 Body depth -0.178 PC 2 0.293 PC 3 0.135 PC 4 PC 5 PC 6 PC 7 -0.104 -0.093 0.022 0.033 0.014 HL 0.204 0.116 0.370 0.001 -0.086 0.003 Dorsal fin base 0.067 0.412 -0.139 -0.019 0.023 0.049 0.074 -0.163 0.153 -0.176 0.334 -0.178 0.030 -0.013 Anal fin base Predorsal distance 0.205 0.183 0.230 -0.049 0.056 0.143 0.013 Rostral tip–anal fin origin 0.241 0.232 0.002 -0.112 0.120 0.124 0.058 Rostral tip–pectoral fin origin 0.047 0.063 0.523 0.051 -0.029 0.032 -0.242 Rostral tip–ventral fin origin 0.206 0.227 0.208 0.079 -0.036 0.193 0.011 Caudal peduncle length 0.107 -0.078 0.150 0.208 0.314 -0.407 0.090 Caudal peduncle depth -0.151 0.144 0.125 0.238 -0.035 0.020 0.035 Postdorsal distance -0.024 0.041 0.271 0.182 0.100 0.084 0.380 Dorsal fin origin–anal fin origin -0.274 0.090 0.114 -0.080 -0.079 -0.059 0.035 Post dorsal fin base–anal fin origin -0.185 0.199 -0.058 0.232 -0.118 -0.051 0.140 Dorsl fin origin–post anal fin base -0.198 0.222 0.003 0.064 -0.123 -0.008 0.039 Dorsal fin origin–post anal fin base Post-dorsal fin origin–post anal fin base -0.169 -0.187 0.186 0.200 0.081 0.191 0.175 -0.122 0.057 -0.093 -0.173 0.002 0.074 0.032 Postdorsal fin base–hypural base 0.082 0.015 0.083 0.182 0.390 -0.431 -0.049 Anal fin origin–hypural base 0.134 0.263 -0.185 0.342 0.028 -0.047 0.123 Anal fin origin–pelvic fin origin -0.038 0.075 0.010 -0.425 0.210 0.012 -0.065 Pelvic fin origin–pectoral fin origin -0.157 0.238 0.063 -0.203 -0.030 -0.039 -0.252 0.161 0.238 -0.263 -0.082 0.219 0.088 0.173 Interorbital width -0.220 0.094 -0.134 -0.144 0.035 0.013 0.024 Snout length -0.122 -0.136 -0.114 0.185 0.218 0.283 -0.037 0.069 -0.082 -0.030 0.233 0.172 0.448 0.012 Premaxillary pedicel length -0.057 -0.022 0.103 0.169 0.244 0.412 -0.332 Cheek depth -0.229 -0.122 0.123 0.113 0.018 0.133 -0.021 Eye diameter 0.175 0.207 -0.041 -0.137 0.310 0.017 -0.060 Lachrymal depth -0.198 -0.067 0.104 0.103 0.132 -0.074 -0.252 Snout width -0.251 -0.052 0.081 -0.043 0.177 0.105 0.187 Preorbital width Snout width across the lachrymal -0.259 -0.180 0.013 -0.032 -0.098 0.104 0.036 0.032 -0.046 0.069 -0.136 0.258 0.090 0.459 Lower jaw width -0.208 -0.057 0.119 -0.119 0.295 -0.006 0.121 Pectoral fin base -0.125 0.172 -0.106 0.082 0.229 -0.138 -0.385 Pelvic fin base -0.117 0.201 -0.182 0.017 0.172 0.040 -0.209 Cum. % 31.622 44.340 52.107 59.390 65.332 69.792 72.942 Head width Lower jaw length Phylogenetic analysis and divergence times We obtained 31 different haplotypes (24 from the ingroup and 7 from the outgroup) from a total of 149 Cox1 (585 bp) sequences analyzed (Table 2 of Supplementary Material 2). Significant molecular differences were retrieved between Nosferatu new genus and true Herichthys in the phylogenetic reconstruction, including 7% mtDNA uncorrected divergence (Table 5) between the two taxa (Fig. 6). Moreover, we found similar topologies between the ML and BI trees. However, a few inconsistencies were detected in true Herichthys. While in the BI analysis, the most basal node was haplotype 4, shared by H. deppii and 123 Author's personal copy Hydrobiologia Table 2 Classification rate (CDA) with holdout showing the split between Herichthys and Nosferatu new genus Act. group Pred. group (std) Herichthys Herichthys Nosferatu Nosferatu 4 0.99 0 95 1.000 0.99 95 1.000 0.99 4.000 2.000 PC2 11.09% 275 3 0 6.000 0.000 -2.000 -4.000 -6.000 H. cyanoguttatus H. teporatus -8.000 H. carpintis -5.000 0.000 5.000 10.000 PC1 38.48% Fig. 3 Plot of scores on the first morphometric principal component and the second morphometric principal component of pooled material of Herichthys cyanoguttatus, H. teporatus, and H. carpintis H. teporatus, in the ML topology, the most basal node corresponded to haplotype 8, shared by H. tepehua n. sp., H. tamasopoensis, and one specimen of N. panctostictus (possibly a hybrid) (for hybrids N. pantostictus 9 H. carpintis see Supplementary Material 3). Still, a different relationship was found using the Bayesian-coalescence approach (implemented in BEAST), which resulted in H. carpintis as the most basal species in true Herichthys. Despite these inconsistencies, we retrieved highly consistent biogeographic and taxonomic patterns in our phylogenetic and demographic reconstructions. The estimated age for the analyzed ingroup (Herichthys ? Nosferatu new genus) is about 7 Mya (*5 to 11 Mya 95% HPD, Fig. 7). This initial divergence was followed by the separation between both genera about 5 Mya (3–8 Mya 95% HPD). 123 Pred. group (holdout) 276 Overall correct class. rate -10.000 -10.000 Correctly classified Herichthys Nosferatu Correctly classified 0.99 However, despite their ancient origins, intra-diversification processes within both genera were recovered as more recent. By comparison with true Herichthys, we found high levels of intrageneric divergence and structure within Nosferatu new genus, from which the bartoni clade (conformed by N. bartoni ? N. labridens) was the first group to diverge during the Miocene (*3 Mya, 1.5–4.7 Mya, 95% HPD, Fig. 7). However, we also recovered the timing of cladogenetic events within the bartoni clade as very recent (\1 Mya). This recent divergence is evidenced by the low within-species divergence values (0.24 ± 0.1 uncorrected p distances) in N. bartoni, sharing haplotypes with sympatric N. labridens (i.e., haplotype 14, Supplementary Material 4). We retrieved two additional clades within Nosferatu new genus: the pantostictus and the steindachneri clades. The monophyletic pantostictus clade is composed of one nominal species, H. pantostictus, which has both the widest distribution range and the highest number of haplotypes (5) within the genus. (Table 3 of Supplementary Material 2; Supplementary Materials 4 and 5). Divergence between pantostictus and steindachneri clades occurred shortly after the separation of the bartoni clade (*2 Mya, 1–3 Mya, 95% HPD, Fig. 7). The steindachneri clade is the most morphologically diverse within Nosferatu new genus. It is composed of three nominal species: N. pame, N. pratinus, and N. steindachneri. The clade also exhibited the highest internal levels of differentiation (Fig. 6) (i.e., Dp = *4.5, *4.12, and *2.83%) between N. pratinus and N. bartoni, N. labridens, and N. pantostictus, respectively (Table 5). True Herichthys presented a lower ingroup divergence compared to Nosferatu new genus (Dp = 0.75 ± 0.74 vs. 2.01 ± 1.86, respectively, Table 6); phylogenetic reconstructions exhibited poor resolution, and no species were recovered as monophyletic (Fig. 6). Author's personal copy Hydrobiologia Table 3 Component loadings on the first eight principal components of the pooled morphometric data for Herichthys cyanoguttatus (n = 74), H. teporatus (n = 35), and H. carpintis (n = 67) Variable PC 1 Body depth -0.182 0.302 0.052 -0.057 0.253 HL 0.196 0.270 -0.199 -0.190 Dorsal fin base 0.202 0.256 0.134 0.082 Anal fin base PC 2 PC 3 PC 4 PC 5 PC 6 PC 7 PC 8 0.030 0.069 0.017 0.001 -0.041 -0.018 0.157 0.074 -0.004 0.134 -0.138 0.109 -0.122 0.131 0.406 0.053 -0.078 -0.046 -0.001 Predorsal distance 0.184 0.272 -0.111 -0.037 0.222 -0.042 0.050 0.242 Rostral tip–anal fin origin 0.290 0.134 -0.086 0.052 0.183 -0.036 0.073 -0.010 Rostral tip–pectoral fin origin 0.053 0.296 -0.254 -0.156 -0.215 -0.095 -0.039 0.236 Rostral tip–ventral fin origin 0.218 0.283 -0.044 -0.078 0.097 -0.121 -0.039 0.080 Caudal peduncle length 0.028 0.117 -0.178 0.239 -0.246 0.264 -0.289 -0.169 Caudal peduncle depth -0.093 0.254 0.074 0.056 -0.082 0.014 0.043 0.160 Postdorsal distance -0.067 0.209 -0.168 -0.100 -0.094 -0.161 -0.338 -0.221 Dorsal fin origin–anal fin origin -0.275 0.172 -0.045 -0.088 0.026 0.124 0.078 -0.034 Post dorsal fin base–anal fin origin -0.132 0.232 0.249 0.024 0.000 0.003 -0.051 -0.154 Dorsal fin origin–post anal fin base -0.111 0.226 0.124 -0.123 -0.146 -0.102 0.329 -0.374 Post-dorsal fin origin–post anal fin base Dorsal fin origin–pectoral fin origin -0.087 -0.211 0.248 0.272 0.118 0.017 0.173 -0.111 -0.039 0.190 0.163 0.084 -0.070 0.039 -0.132 0.065 Postdorsal fin base–hypural base 0.084 0.159 -0.121 0.295 -0.195 0.313 -0.290 0.060 Anal fin origin–hypural base 0.161 0.219 0.319 0.168 0.022 -0.011 -0.193 -0.041 Anal fin origin–pelvic fin origin 0.067 0.020 -0.224 0.183 0.133 0.194 0.475 -0.119 -0.088 0.072 -0.057 -0.108 0.051 0.375 0.233 -0.094 0.237 -0.001 0.087 0.277 0.281 -0.005 0.007 -0.101 Interorbital width -0.164 -0.062 0.060 0.140 0.388 0.194 -0.190 0.402 Snout length -0.160 -0.010 0.075 0.242 0.181 -0.191 -0.021 0.254 0.038 0.003 -0.043 0.191 0.070 -0.550 0.070 -0.052 Premaxillary pedicel length -0.069 0.102 -0.210 0.170 -0.092 -0.282 -0.004 0.009 Cheek depth -0.265 0.043 -0.006 -0.037 0.009 -0.231 -0.043 -0.001 Eye diameter 0.257 -0.007 -0.200 0.179 0.037 0.030 0.136 -0.003 Lachrymal depth -0.185 0.029 -0.084 0.103 -0.316 -0.087 0.088 0.256 Snout width -0.254 0.023 -0.202 0.129 0.112 -0.047 0.027 -0.116 Preorbital width Snout width across the lachrymal -0.265 -0.171 0.000 0.022 -0.083 -0.285 0.093 0.072 0.155 0.244 0.050 -0.034 0.051 -0.116 0.032 -0.236 Lower jaw width -0.185 -0.002 -0.320 0.162 0.070 -0.036 -0.065 -0.079 Pelvic fin origin–pectoral fin origin Head width Lower jaw length Pectoral fin base -0.064 0.067 0.004 0.354 -0.339 0.026 0.387 0.290 Pelvic fin base -0.012 0.003 0.138 0.418 0.001 -0.105 -0.030 -0.202 Cum. % 27.543 41.212 50.121 56.240 60.875 65.120 69.096 72.362 Moreover, we recovered true Herichthys as young in comparison to Nosferatu new genus (*1 Mya, 0.5–2 Mya, 95% HPD, Fig. 7). Despite low phylogenetic resolutions, the highest levels of differentiation corresponded to H. deppii, which had the highest levels of divergence with respect to H. carpintis, H. tamasopoensis, and H. cyanoguttatus (Dp = 1.65 ± 1.11, 1.60 ± 0.98, and 1.43 ± 0.86, respectively, Table 7). Haplotype distribution exhibited a gradual transition in haplotype frequencies in true Herichthys. H. deppii, and H. tepehua n. sp. shared the least number of haplotypes with the other species (Supplementary Materials 4 and 5). Nonetheless, similarly to 123 Author's personal copy PC2 10.64% Hydrobiologia 6.000 H. carpintis 5.000 H. tepehua 4.000 H. deppii 3.000 2.000 1.000 0.000 -1.000 -2.000 -3.000 -4.000 -5.000 -10.000 -5.000 0.000 5.000 10.000 PC1 45.08% Fig. 4 Plot of scores on the first morphometric principal component and the second morphometric principal component of pooled material of Herichthys carpintis, H. tepehua, and H. deppii Nosferatu new genus, the highest haplotype diversity was found in the Rı́o Pánuco Basin. Evaluation of demographic history within true Herichthys detected a pattern of contraction during the lower Pleistocene (Supplementary Material 6). Systematic section The list of nominal species currently included in Herichthys is as follows: In the ‘‘labridens assemblage’’ = Nosferatu new genus (De la Maza-Benignos & Lozano-Vilano, 2013), Nosferatu bartoni (Bean, 1892), N. steindachneri (Jordan & Snyder, 1899), N. pantostictus (Taylor & Miller, 1983), N. labridens (Pellegrin, 1903), N. pame (De la Maza-Benignos & Lozano-Vilano, 2013), N. pratinus (De la Maza-Benignos & Lozano-Vilano, 2013), and N. molango (De la Maza-Benignos & Lozano-Vilano, 2013); and in the ‘‘cyanoguttatus assemblage’’ = Herichthys [we follow Kullander (2003) and Eschmeyer (2013)]: H. deppii (Heckel, 1840), H. carpintis (Jordan & Snyder, 1899), H. tamasopoensis, ArtigasAzas, 1993, H. cyanoguttatus, Baird & Girard, 1854, and H. minckleyi (Kornfield & Taylor, 1983). Nosferatu new genus (Tables 4 and 5 of Supplementary Material 2; Figs. 8, 9, and 10) Type species Nosferatu pame, by original designation. 123 Diagnosis Differs from Herichthys by the following measurements: shallower body (mean 41%, SD 2% vs. mean 45%, SD 2%); shorter dorsal fin base (mean 55%, SD 2% vs. mean 58%, SD 2%), shorter anal fin base (mean 22%, SD 2% vs. mean 24%, SD 2%); shorter dorsal fin origin to anal fin origin (mean 51%, SD 3% vs. 55%, SD 2%); shorter post dorsal fin base to anal fin origin (mean 33%, SD 3% vs. 36%, SD 3%); shorter dorsal fin origin to post anal fin base (mean 61%, SD 3% vs. mean 65%, SD 3%); shorter dorsal fin origin to pectoral fin origin (mean 27%, SD 2% vs. mean 29%, SD 2%); shorter anal fin origin to pelvic fin origin (mean 28%, SD 2% vs. mean 30%, SD 2%), and shorter pelvic fin origin to pectoral fin origin (mean 16%, SD 1% vs. mean 28%, SD 3%), all in SL. Narrower interorbital width (mean 26%, SD 3% vs. mean 29%, SD 3%); longer lower jaw (mean 33%, SD 2% vs. mean 31%, SD 2%); narrower snout (mean 32%, SD 3% vs. mean 35%, SD 3%); narrower width at preorbital (mean 27%, SD 3% vs. mean 31%, SD 3%); shorter pectoral fin base (mean 21%, SD 2% vs. mean 23%, SD 2%); and shorter pelvic fin base (mean 11%, SD 1% vs. mean 13%, SD 2%), all in HL. Depressed dorsal fin rarely expands beyond the anterior third of the caudal fin. An elongated, elastic, smooth caecum (not present in Herichthys) is adhered to a saccular stomach. Genus is distinguished from most other Heroine genera by the following synapomorphies: Breeding pigmentation that consists of darkening of the ventral area, extending over nostrils, opercular series, or pectoral fins. All have red or purple marks in the axil of the pectoral fin, except for N. bartoni. Anterior teeth regularly set, well-spaced, conic, unicuspid, strongly recurved, and pointed, with erect implantation; with transition to prolongation in the size of the symphysial pair of teeth relative to that of the other teeth in the outer row of the upper jaw, reminiscent of those in the vampire Nosferatu (herein = nosferatuform teeth); and a less developed pair in the lower jaw (Figs. 9, 10). Posterior teeth small and pointed, none or few posterior rows of diminutive teeth in the upper jaw; teeth in the jaws are conical, recurved, well-spaced, and pointed; frontal anterior row regularly set and lateroposterior anterior row irregularly set, recurved, and pointed (frequently worn in older specimens). Description Body elongated and slender, depth 36–45% (mean 41%, SD 2%); scales ctenoid. dorsal fin base short 48–60% (mean 55%, SD 2%); anal fin Author's personal copy Hydrobiologia Table 4 Component loadings on the first seven principal components of the pooled morphometric data for Herichthys carpintis (n = 67), H. tepehua n. sp. (n = 60), and H. deppii (n = 27) Variable Body depth PC 1 PC 2 PC 3 PC 4 PC 5 PC 6 0.150 PC 7 0.133 -0.081 0.378 0.247 0.003 0.096 HL -0.200 -0.155 0.333 -0.069 -0.048 0.104 0.007 Dorsal fin base -0.190 0.223 0.159 0.255 0.044 -0.039 -0.116 -0.022 0.158 0.327 0.025 0.003 -0.238 0.136 Predorsal distance -0.235 -0.075 0.132 0.098 -0.019 0.088 0.140 Rostral tip–anal fin origin -0.262 0.034 0.111 0.166 0.108 -0.068 -0.046 Rostral tip–pectoral fin origin -0.016 -0.251 0.342 -0.304 0.096 0.089 0.046 Rostral tip–ventral fin origin -0.224 0.051 0.279 0.033 0.066 0.128 0.076 Caudal peduncle length 0.060 0.116 0.069 -0.288 -0.109 -0.463 0.193 Caudal peduncle depth 0.183 0.130 0.299 0.084 -0.002 -0.075 -0.034 Postdorsal distance 0.065 0.067 0.114 -0.261 0.216 0.052 -0.079 Dorsal fin origin–anal fin origin 0.246 -0.138 0.139 0.094 0.055 0.032 -0.107 Post dorsal fin base–anal fin origin 0.214 0.226 0.125 0.120 -0.227 0.083 -0.011 Dorsal fin origin–post anal fin base 0.213 0.160 0.189 0.191 -0.035 0.059 -0.177 Post-dorsal fin origin–post anal fin base Dorsal fin origin–pectoral fin origin 0.205 0.143 0.116 -0.182 0.263 0.229 0.100 0.076 -0.063 0.038 -0.143 0.011 -0.141 0.124 0.092 Anal fin base 0.032 0.172 0.118 -0.207 0.097 -0.487 Anal fin origin–hypural base Postdorsal fin base–hypural base -0.087 0.416 0.059 0.034 -0.214 -0.011 0.146 Anal fin origin–pelvic fin origin -0.100 -0.230 0.002 0.269 0.280 -0.230 -0.325 Pelvic fin origin–pectoral fin origin Head width Interorbital width Snout length Lower jaw length 0.041 -0.090 0.195 -0.073 -0.227 -0.165 0.262 -0.218 0.178 -0.046 0.227 0.094 -0.148 0.057 0.182 -0.055 -0.121 0.354 0.066 -0.057 0.049 0.178 0.102 -0.292 0.108 0.040 0.134 0.164 -0.034 0.190 0.123 -0.068 0.390 0.279 0.197 Premaxillary pedicel length 0.075 0.138 -0.022 -0.091 0.380 0.163 0.349 Cheek depth 0.226 0.012 -0.010 -0.154 0.039 0.217 -0.009 Eye diameter -0.215 0.083 0.015 0.074 0.135 -0.192 0.169 Lachrymal depth 0.195 0.037 -0.056 -0.242 0.089 0.021 -0.252 Snout width 0.236 -0.096 0.020 0.070 0.053 -0.044 0.162 Preorbital width Snout width across the lachrymal 0.221 0.152 -0.075 -0.188 -0.043 0.044 0.145 0.161 0.151 -0.047 -0.131 -0.219 0.003 0.211 Lower jaw width 0.190 -0.179 -0.072 0.047 0.177 -0.125 0.259 Pectoral fin base 0.095 0.217 0.099 -0.145 0.345 -0.152 -0.397 Pelvic fin base 0.060 0.250 -0.032 0.140 0.304 -0.018 Cum. % 36.17 47.88 base short 19–27% (mean 22%, SD 2%); origin of dorsal fin base to origin of anal fin base 45–55% (mean 51%, SD 3%), all of SL; preorbital width slender 18–34% (mean 27%, SD 3%) of HL. Dorsal fin XV–XVIII (mode XVI, freq 60%), 9–12 (mode 11, freq 48%); anal fin III–VII (mode V, freq 62%), 8–10 (mode 8, freq 62%); pectoral fin rays 13–16 (mode 15, freq 65%); scales ctenoid, longitudinal series 28–34 54.87 61.36 66.19 70.43 0.199 73.77 (mode 30, freq 37%). Markings consist of irregular variegating patterns along the flanks that range from a faint discontinuous horizontal stripe to 3–5 irregularly constituted blotches. Breeding pigmentation dark on anterioventral half over nostrils, opercular series and pectoral fins, as well as big portions of posterior half, almost imperceptible in N. stendachneri; anterior teeth well-spaced conic-unicuspid, acutely pointed, slightly 123 Author's personal copy Hydrobiologia Fig. 5 Graphical representation of canonical discriminant functions analysis using grouping discriminant analysis (GDA) of species in Herichthys on the basis of 42 morphometric characters Discriminant Plot H. carpintis H. cyanoguttatus H. deppii South of Sierra de Tantima Geomorphological Group 5 H. tamasopoensis H. tepehua 4 H. teporatus 3 Río Pánuco Geomorphological Group Funct 2 (24.1%) 2 1 0 -1 -2 -3 -4 North of Sierra de Tamaulipas Geomorphological Group -5 -10 -5 0 5 10 Funct 1 (47.4%) Table 5 Inter-generic COI–I genetic divergence (p = non-corrected distances) between Nosferatu and Herichthys Herichthys Nosferatu Paraneetroplus Herichthys 000.75 ± 000.74 Nosferatus 007.04 ± 001.13 002.01 ± 001.86 Paraneetroplus 008.64 ± 001.06 008.78 ± 000.44 006.97 ± 002.67 Vieja 009.09 ± 001.05 009.32 ± 000.56 005.32 ± 003.40 to strongly recurved, differentiated in length in upper and lower jaws: a short symphysial pair of teeth in the lower jaw, whereas an enlarged symphysial pair of fangs on the premaxillae flanked by shorter caniniform teeth on each side; posterior teeth diminutive; few posterior rows or none in upper jaw. Depressed dorsal fin short; point rarely expands beyond anterior third of caudal fin. Robust-walled stomach is saccular rugged with longitudinal folds walls and adhered at its anterodorsal section to an elongated elastic smooth caecum. Etymology Masculine, proper name. The name refers to the pair of well-developed recurved fangs in 123 Vieja 004.59 ± 003.33 Fig. 6 Phylogenetic tree derived from the ML and BI analysis of a fragment of Cox1 (585 bp) mitochondrial data of Herichthys and Nosferatu new genus (scientific names of the outgroup c species are reported as they appear in genebank) the upper jaw present in all species of the genus, reminiscent of those in Marnau’s vampire Nosferatu. Geographical distribution Atlantic slope in Veracruz, Hidalgo, Querétaro, and Tamaulipas in the Pánuco-Tamesı́ River Basin; lagoon systems of San Andrés, including the Rı́o Tigre and Tamiahua system and its tributaries, the Cucharas and Naranjos rivers. H1. H. tepehua (Pantepec/Tenixtepec) H. cyanoguttatus (San Juan) H3. H. deppii (Nautla) H. tepehua (Pantepec/Tenixtepec/Tecolutla/Cazones) 77/88/0.99 -/-/1 Hydrobiologia H2. H. deppii (Nautla-Zanjas de arena) H. tepehua (Solteros) 54/60/- H4. H. deppii (Nautla/Misantla) H. teporatus (Soto la Marina) H16 H. carpintis (Lower Pánuco) H10. H. carpintis (El Salto) H. teporatus (Soto la Marina) H. tepehua (Tenixtepec) H. cyanoguttatus (Lampazos/San Marcos) H11. H. carpintis (Tamiahua) -/57/- true Herichthys genus H9. H. carpintis (Guayalejo/El Salto/Media Luna/Tamiahua) H. tamasopoensis (Tamasopo) H. tepehua (Tenixtepec) H. cyanoguttatus (San Fernando/San Juan) 99/99/- H17. H. carpintis (Guayalejo) H15. H. carpintis (Guayalejo) H6. N. pame (Tamasopo) N. steindachneri (Tamasopo) N. pratinus (El Salto) Clade Steindachneri 99/99/1 H23. N. steindachneri (Tamasopo) 61/62/0.83 95/100/1 82/87/1 H5. N. pratinus (El Salto) H24. N. steindachneri (Tamasopo) 62/-/0.95 H7. N. pame (Tamasopo) H18. N. pantostictus (Huazalingo/Axtla/Jaumave/Guayalejo/Mante/Tamiahua) 89/93/1 Clade Pantostictus Nosferatu new genus 9 6 /93/ 1 H20. N. pantostictus (Guayalejo) H21. N. pantostictus (Guayalejo) 9 6/96/1 H19. N. pantostictus (Axtla/Tamiahua) H22. N. pantostictus (Mante) Clade Bartoni 60/-/- 100/100/1 H14. N. bartoni (Los Anteojos) N. labridens (Media Luna) H12. N. bartoni (Los Anteojos) H13. N. bartoni (Los Anteojos) Hap26. Vieja heterospila Paraneetroplus maculicauda 100/-/H31. Paraneetroplus fenestratus H27. Vieja bifasciata 50/-/0.99 H30. Paraneetroplus sysnpilus H28. Vieja synspila Vieja melanura H25. Vieja regani 123 H29. Vieja argentea PhyML/RaxML/BI 0.02 Author's personal copy H8. H. tepehua (Tenixtepec) H. tamasopoensis (Tamasopo) H. pantostictus* (Mante) Author's personal copy Hydrobiologia 123 Author's personal copy Hydrobiologia b Fig. 7 Ultrametric tree based on a fragment of Cox1 mitochondrial gene, uncorrelated lognormal relaxed molecular clock, using the SRD06 model of nucleotide substitution partitioning the nucleotide data by codon position. Mean of divergence time with highest posterior density intervals [in brackets 95%] for each divergence time is annotated in the main nodes. Scale is in millions of years both nuclear and mitochondrial markers are required. We detected 14 Cox1 haplotypes exclusive to the genus: H5, H6, H7, H12, H13, H14, H18, H19, H20, H21, H22, H23, H24, and H26 (Table 3 of Supplementary Material 2, and Supplementary Material 7). Herichthys Baird & Girard, 1854 Species composition Seven species: Nosferatu molango, N. pame, N. pratinus, N. bartoni, N. labridens, N. pantostictus, and N. steindachneri. Remarks Our morphologic analysis supports Nosferatu as different from Herichthys. Our mitochondrial DNA analysis also recovered Nosferatu as monophyletic, and sister to Herichthys; average Dp = 7.0%, thus separating both genera in evolutionary terms, which is slightly higher than the divergence (5.32%) recovered between Paraneetroplus and Vieja. Our results concur with mtcytb analyses by Hulsey et al. (2004), Concheiro-Pérez et al. (2006), and LópezFernández et al. (2010) that registered 7.0% average non-parametric rate smoothing (NPRS) divergence between Nosferatu new genus and true Herichthys. Our findings are also in agreement with De la MazaBenignos & Lozano-Vilano (2013) whereby Nosferatu new genus corresponds to the ‘‘labridens assemblage.’’ In addition, our analysis recovered three monophyletic sister groups within Nosferatu new genus: the ‘‘bartoni clade’’ conformed by N. labridens and N. bartoni of the Media Luna and the upper Rı́o Verde; the ‘‘pantostictus clade’’ conformed by the polymorphic N. pantostictus; and the ‘‘steindachneri clade’’ conformed by N. steindachneri, N. pame, and N. pratinus. In the preliminary analysis, we recovered haplotypes from N. molango (1) within the true Herichthys group. We hypothesize that our preliminary finding could correspond to a pattern of recent introgression. The Lago Azteca is remote and isolated, and N. molango is the only native cichlid found therein. Further studies in order to clarify the evolutionary history of N molango using a combination of (Tables 4 and 5 of Supplementary Material 2; Figs. 8, 9, and 10) Type species Herichthys cyanoguttatus Baird & Girard, 1854. Diagnosis Differs from Nosferatu new genus in that the red/purple mark on the axil of the pectoral fin is absent, and depressed dorsal fin reaches past the frontal third of caudal fin (for comparative morphometrics, see Nosferatu new genus in the previous section). Genus is distinguished from most other Heroine genera by the following synapomorphies: Six to seven vertical flank bars bearing a series of dark blotches below the lateral line that make up the principal markings. Breeding pigmentation consists of darkening of posterior half and anterioventral areas that do not extend over the nostrils, opercular series, and pectoral fins. Anterior teeth are closely spaced, spatulate, chisel-like, bicuspid or weakly bicuspid, or a mixture of bicuspid and bluntly pointed conical, with straight curvature, undifferentiated in length in upper and lower jaws (Figs. 9, 10), except in H. minckleyi. Description Body depth 40–51% (mean 45%, SD 2%), scales ctenoid. Dorsal fin XIV–XVII (mode XVI, freq 15%) 9–12 (mode 11, freq 15%); anal fin IV–VII (mode VI, freq 19%); 6–10 (mode 8, freq 16%); pectoral rays 13–15 (mode 14, freq 26%). Teeth spatulate, bicuspid or weakly bicuspid, undifferentiated in length in upper and lower jaw. Soft dorsal fin and anal fins pointed extending past the frontal third of the caudal fin. Geographical distribution Rivers in the Atlantic slope of Mexico and Texas, north of PDM, including Santa Table 6 Intergeneric COI–I genetic divergence (p = non-corrected distances) between Nosferatu and Herichthys Herichthys Nosferatu Paraneetroplus Herichthys 000.75 ± 000.74 Nosferatu 007.04 ± 001.13 002.01 ± 001.86 Paraneetroplus 008.64 ± 001.06 008.78 ± 000.44 006.97 ± 002.67 Vieja 009.09 ± 001.05 009.32 ± 000.56 005.32 ± 003.40 Vieja 004.59 ± 003.33 123 Author's personal copy Hydrobiologia 123 Author's personal copy 000.81 ± 000.27 000.93 ± 000.15 000.69 ± 000.92 000.90 ± 000.70 000.77 ± 000.56 000.42 ± 000.51 001.65 ± 001.11 000.70 ± 000.83 000.75 ± 000.68 000.73 ± 000.32 H. deppii H. tamasopoensis H. tepehua n. sp. H. teporatus 001.08 ± 000.73 000.89 ± 000.92 000.46 ± 000.42 001.43 ± 000.86 000.31 ± 000.38 000.57 ± 000.43 H. carpintis H. cyanoguttatus 001.60 ± 000.98 000.84 ± 000.70 000.59 ± 000.54 H. tepehua n. sp. H. tamasopoensis H. deppii H. cyanoguttatus Herichthys deppii (Heckel, 1840). (Tables 6 and 7 of Supplementary Material 2; Figs. 8, 10, 11, and 12) Synonymy Herichthys geddesi (Regan, 1905) and Heros montezumae (Heckel, 1840). Holotype Lost (Kullander, 2003; Eschmeyer, 2013). Neotype Designated due to holotype loss and for the purpose of clarifying the taxonomic status and type locality: UANL 20300 (1: 131 mm male) La Palmilla (Rı́o Bobos), Tlapacoyan, Ver. Lat. 20.0148167, Long. -97.1418333, masl 137, De la Maza-Benignos, March 20, 2006. Diagnosis Differs from H. carpintis, H. cyanoguttatus, and H. teporatus in having a longer anal fin base (mean 27%, SD 2% vs. mean 22, 24 and 25%; SD 1, 2, and 1%, respectively) and distance from the anal fin origin to the hypural base (mean 42%, SD 1% vs. mean 36, 38, and 39%; SD 1, 2, and 2%, respectively). Also differs from other Herichthys in the following autapomorphies: ground color brownish with head area sometimes bluish-gray when alive and head marked H. carpintis South of Sierra de Tantima geomorphological group Table 7 Intrageneric COI–I genetic divergence (p = non corrected distances) in Herichthys Ana, Misantla, Nautla, Solteros, Tecolutla, Tenixtepec, Cazones, Pantepec, Pánuco, Soto la Marina, San Fernando, Lower Rı́o Grande; absent in the Rı́o Conchos, Cuatro Ciénegas, and the Rı́o Nueces in Texas. Species composition Seven species: H. deppii, H. tepehua n. sp., H. carpintis, H, tamasopoensis, H. cyanoguttatus, H. teporatus, and H. minckleyi. Remarks Our phylogenetic reconstructions exhibited poor resolution among species. None of the recognized Herichthys species were recovered as monophyletic. We detected 11 Cox1 haplotypes that were exclusive to the genus: H1, H2, H3, H4, H8, H9, H10, H11, H15, H16, and H17. 001.13 ± 000.46 b Fig. 8 H. teporatus A Herichthys deppii (Rı́o Bobos); B H. deppii (Zanjas de Arena); C H. tepehua n. sp. (Rı́o Solteros); D H. tepehua n. sp. (Rı́o Tecolutla); E H. tepehua n. sp. (Rı́o Cazones); F H. tepehua n. sp. (Rı́o Pantepec); G H. tamasopoensis; H H. carpintis (Rı́o El Salto); I H. carpintis (Tamiahua); J H. teporatus; K H. cyanoguttatus (Rı́o San Fernando); L H. cyanoguttatus (Rı́o Bravo); M H. minckleyi (Cuatro Ciénegas); N Nosferatu bartoni (Media Luna); O N. labridens (Media Luna); P N. pratinus (Rı́o El Salto); Q N. pame; R N steindachneri; S N. pantostictus (Laguna La Puerta); T N. molango (Laguna Azteca); and U N. pantostictus (Arroyo el Tigre) 000.47 ± 000.68 Hydrobiologia 123 Author's personal copy Hydrobiologia Fig. 9 The drawings by Aslam Narváez-Parra depict the anterior teeth of Nosferatu pratinus, N. labridens, N. bartoni, in comparison to Herichthys carpintis with conspicuous big round 3–5 mm orange/brown freckles. Dorsal and ventral contour symmetric and convex gives fish an elongated, elliptic appearance. Description Description is based on sexually mature specimens over 63.6 mm SL. Morphometric and meristic data are summarized in Tables 6 and 7 of Supplementary Material 2. Anal fin base 24–31% (mean 27%, SD 2%); anal fin origin–hypural base distance 29–44% (mean 42%, SD 1%). Dorsal fin XVI–XVII (Mode XVII, freq 67%) 9–12 (Mode 11, freq 39%), anal fin VI–VII (Mode VI, freq 72%) 7–10 (Mode 8, freq 44%); dorsal and ventral contour symmetric and convex gives fish an elongated, elliptic appearance. Nuchal hump absent in mature males. Teeth spatulate, chisel-like, bicuspid or weakly bicuspid, undifferentiated in length 123 in upper and lower jaws. Lower pharyngeal plate is stout and broad; two rows of 9–10 unpigmented molars increasing caudally in size flank midline; 20–22 conical, progressively compressed teeth along posterior margin. Coloration in preservative Light rust-brown; ventral area whitish in some individuals. Fins the same color as body, except paler, with speckles over soft areas of dorsal and anal fins. Live colors Ground color brown, sometimes splashed with blue-gray over dorsal, caudal, and central areas of the body. Head dotted with conspicuous, big, round 3–5 mm orange/brown freckles that vary in number and size. Freckles extend over gill covers, head, and pectoral fin base. Two to three rows of freckles form a diagonal with the horizontal that Author's personal copy Hydrobiologia Fig. 10 The drawings by Aslam Narváez-Parra depict the occlusal surface and caudal aspect of the lower pharyngeal plates in Nosferatu bartoni and N. pratinus, and the occlusal surface of the lower pharyngeal plates in Herichthys tamasopoensis, H. carpintis, H. cyanoguttatus, H. tepehua n. sp., and H. deppii Fig. 11 Neotype UANL 20300, Herichthys deppii (Heckel 1840) collected in La Palmilla (Rı́o Bobos), Tlapacoyan, Ver. (131 mm SL) extend from the lip fold to the orbit of the eye. Fins the same color as body with small brown streaks or spots on the soft ray regions and the caudal fin; pectoral and pelvic fins opaque. Geographical distribution Upper Nautla and Misantla Rivers and their tributaries, in the municipalities of Tlapacoyan and Misantla in the state of Veracruz, Mexico. Habitat and associates Herichthys deppii is found in clear water with current, over sand substrate with big boulders, between rapids of the upper Nautla and Misantla Rivers. It shares habitat with Xiphophorus cf. helleri, Poecilia mexicana, Astyanax mexicanus, Gobiomorus dormitor, Agonostomus monticola, and Awaous tajasica. Little is known of its behavior in the wild. It nests underneath big boulders where it protects itself and its fry from local fishermen that capture them year-round using homemade spears. Vernacular names Nautla cichlid. Remarks Until the late 1990s, both H. deppii and H. geddesi were treated as incertae sedis with type localities in ‘‘Mexico’’ and ‘‘Southern Mexico,’’ respectively. In 2005, Miller et al. noted that Stawikowski & Werner (1998) had retrieved the locality of H. deppii as Rı́o Misantla, Veracruz. Kullander (2003) noted that the validity of H. geddesi Regan, 1905 needs further research and its generic allocation was 123 Author's personal copy Hydrobiologia Fig. 13 UANL 20297 Herichthys tepehua n. sp. holotype collected by Mauricio De la Maza-Benignos in the Rı́o Pantepec (138.17 mm SL) Fig. 12 a Top Juvenile of Herichthys deppii collected and photographed in La Palmilla (Rı́o Bobos), Tlapacoyan, Ver. (*70 mm TL), Veracruz. b Bottom Drawing of H. geddesi from (Regan, Pisces, 1906–1908) uncertain. According to the Natural History Museum of the London Fish Collection Database (21/12/2004), syntypes BMNH 1880.4.7.40-45 (6) were collected by P. Geddes in 1880, in former ‘‘Hacienda del Hobo [sic] between Veracruz and Tampico, Mexico.’’ The first step in establishing the identity of H. geddesi was to determine the exact type locality of the syntypes. Former ‘‘Hacienda del Jobo’’ is located in the municipality of Tlapacoyan, Veracruz in the Nautla River Basin. The main compound of Hacienda del Jobo or San Joaquı́n del Jobo is located in the state of Veracruz, ‘‘between the city of Veracruz and Tampico.’’ The Hacienda was bought in 1825 by General Guadalupe Victoria (1786–1843), the first President of Mexico, sold by his son in 1857 to Rafael Martı́nez de la Torre, and sold again to Juan B. Diez, in 1878. Hence, the (6) H. geddesi were collected in the Nautla River System. Regan’s account (Regan 1905, 1906–1908) mentions its brownish color with the 123 principal markings composed of ‘‘7 or 8 cross-bars bearing a series of blackish blotches below the lateral line; vertical fins spotted.’’ His description matches almost all species in Herichthys. However, the drawing of H. geddesi shown in Plate III, Fig. 4, p. 188 of Regan (1906–1908) is almost certainly H. deppii (Fig. 11), hence in our opinion is synonymous to Heros deppii = Cichlasoma deppii = Herichthys deppi (Heckel, 1840). Average mitochondrial divergence retrieved between H. deppii and H. carpintis was Dp = 1.65%. We detected 7 Cox1 haplotypes among the specimens surveyed, including haplotypes H1 (*17%), uncommon in H. cyanoguttatus (*6%), H4 (*22%), uncommon in H. teporatus (*6%), H8 (*4%), very common in H. tamasopoensis, H9 (*4%), very common in H. carpintis, and also shared with H. tamasopoensis and H. cyanoguttatus, and H10 (*4%), very common in H. cyanoguttatus and less so in H. carpintis. We also detected haplotypes H2 (3%) and H3 (9%), representing divergent lineages exclusive to the south of Sierra de Tantima geomorphological group (i.e., H. deppii and H. tepehua n. sp.) (Supplementary Materials 5 and 7). Author's personal copy Hydrobiologia Herichthys tepehua n. sp. (Tables 6 and 7 of Supplementary Material 2; Figs. 8, 10, and 13) Holotype UANL 20297, adult male, 138.17 mm SL, Rı́o Pantepec, Salsipuedes, Francisco Z Mena, Puebla, Mexico, Long. 20.6750833, Lat. -97.8764667, masl 125, M. De La Maza-Benignos (MMB), June 25, 2006. Paratypes: Pantepec River Basin (54) 47.5–138.17 mm SL: UANL 17490 (17: 73.66–138.17 mm LP) same data as holotype; UANL 17468 (7: 47.5–97.6 mm LP), Potrero del Llano, Temapache, Veracruz, Lat. 21.0870410, Long. -97.7548250, 92 masl, Mauricio De la Maza-Benignos, March 22, 2006; UANL 17403 (10: 53.1–106.6 mm LP), Llano de En medio, Francisco Z Mena, Puebla, Lat. 20.7663290, Long. -97.8624350, 308 masl, MMB, 20 August, 2005; UANL 17489 (5: 55.1–67.7 mm LP), Llano de En medio, Francisco Z Mena, Puebla, Lat. 20.7663290, Long. -97.8624350, 308 masl, MMB, June 25, 2006; UANL 17490 (15: 73.66–132.44 mm LP), Salsipuedes, Francisco Z Mena, Puebla Long. 20.6750833, Lat. -97.8764667, 125 masl, MMB, June 25, 2006. Cazones River Basin (10) 62.44– 130.55 mm SL: UANL 17409 (9: 62.44–130.55 mm SL) Zanatepec, Venustiano Carranza, Puebla, Mexico, Lat. 20.4702330, Long. -97.7584210, masl 389, 1 m depth, MMB, 27 September, 2005; UANL 9326 (1: 85.85 mm SL), Buena Vista River ±1 km North of Chicoaloque, Veracruz, H. Obregón (HO), February 10, 1988; Rı́o Solteros (5) 58.5–119.5 mm SL: UANL 17499 (3: 58.5–78.85 mm SL) Solteros River, Veracruz, Mexico, Lat. 20.2647580, Long. -97.0546300 masl 19, MMB, June, 2006; UANL 17454 (2: 102.7– 119.5 mm SL), Solteros River, Veracruz, Mexico Lat. 20.2647580, Long. -97.0546300 masl 19, MMB, March, 2006; Tecolutla River Basin (9) 63.15– 160 mm SL: UANL 9895 (2: 63.15–138.1 mm SL), Rı́o Coyutla at the Port, HO, 12 February, 1988; UANL 9865 (5: 70.1–109.8 mm SL), Necaxa River at El Frijolillo bridge, Veracruz, Mexico, HO, 12 February, 1988; UANL 9797 (1: 106.35 mm SL), Chichicatzapan River, under the bridge found 2.2 km at the ‘‘5 de Mayo’’ junction—hnos. Valdez, Veracruz, Mexico, HO, February 9, 1988; UANL 17450 (1:160 mm SL), Estero Tlahuanapa, Veracruz, Mexico, lat. 20.3716540, long. -97.2987560, masl 42, MMB, March 18, 2006. Rı́o Tenixtepec (15) 51.9–102.09 mm SL: UANL 17447 (15: 51.9–102.09 mm SL), Arroyo Sta. Agueda, Papantla, Veracruz, Mexico, Lat. 20.71566667, Long. -98.04983333, masl 57, 1 m depth, MMB, March 18, 2006. Diagnosis Differs from H. deppii in that it has a longer head (mean 37%, SD 2% vs. mean 35%, SD 1%); shorter distance from the anal fin origin to the hypural base (mean 39%, SD 2% vs. 42%, SD 1%), all in SL, a larger eye (mean 26%, SD 3% vs. mean 24%, SD 2%) in HL. Differs from H. carpintis in that it has a longer anal fin origin–hypural base distance (mean 39%, SD 2% vs. mean 38%, SD 2%) in SL; deeper cheeks (mean 32%, SD 4% vs. mean 28%, SD 4%) and longer snout (mean 39%, SD 4% vs. mean 37%, SD 3%) all in HL. Also differs from other Herichthys in the following autapomorphies: Two conspicuous blue/ green parallel markings on the sides of the cheeks extending from the lip fold to the orbit of the eye. Ground color is blue-green. Shares with H. deppii the absence of iridescent spots/pearls on the body. Description Description is based on sexually mature specimens over 47.5 mm SL of the Rı́o Pantepec. Morphometric and meristic data are summarized in Tables 6 and 7 of Supplementary Material 2. HL 34–40% (mean 37%, SD 2%); anal fin origin– hypural base 36–42% (mean 39%, SD 2%). Eyes large 21–32% (mean 25%, SD 3%) in HL. Dorsal fin XV– XVII (mode XVI, freq 59%) 10–12 (mode 11, freq 48%). Anal fin IV–VI (mode V, freq 67%) 8–10 (mode 8, freq 44%). Predorsal profile curved; frontal profile concave before the eyes. Teeth spatulate, bicuspid or weakly bicuspid, undifferentiated in length in upper and lower jaw. Lower pharyngeal tooth plate moderately stout and broad, two rows of 10–11 pigmented molars increasing caudally in size and molarization flank midline; 22–24 conic progressively compressed teeth along posterior margin. Stomach is saccular with strong rugged walls (length 22.5% of SL). Coloration in preservative Color is reddish-brown, darker above the lateral line; ventral area between lower jaw and pelvic fin base darker in some individuals. Six vertical flank bars bearing a series of dark blotches below the lateral line make the principal markings. Live colors Ground color olive-green with Persianblue/green; dorsal and anal fins same color as body. Dorsal, anal, and caudal fins with teal spots or streaks randomly distributed. Tips of the dorsal fin rays red in some individuals; a conspicuous black blotch over the middle of the dorsal fin may be present in both males and females. Pectoral fins are opaque; pelvic fins bear 123 Author's personal copy Hydrobiologia longitudinal teal streaks over the ventral portion. Some have salmon pigmentation on gill covers. Two conspicuous, distinct blue/green parallel markings on the sides of the cheeks extend from the lip fold to the orbit of the eye. Breeding pigmentation In the aquarium, ground color changes to pale pink with golden brown markings on the nuchal area. Six vertical dark bars contrast pale frontal half; color between bars and tips of the dorsal rays changes to golden brown. Pelvic fins dark. Geographical distribution H. tepehua is found in the Tuxpan/Pantepec, Cazones, Tecolutla, Tenixtepec, and the Solteros river systems. Etymology From the Nahuatl language meaning ‘‘the ones who possess the mountains.’’ The name refers to the remaining, as of the 1990 Mexico census, 10,573 members of the indigenous Tepehua ethnic group and their language, spoken in eastern Mexico in the states of Veracruz, Hidalgo, and Puebla. Allopatric forms Forms of the Cazones, Tenixtepec, Solteros, and Tecolutla basins differ from the Pantepec in having ground color aquamarine, except for Tenixtepec, which is brownish and its ventral area whitish. Dorsal and anal fins are the same color as the body with mint-green color; variegated markings over the spiny sections; and golden yellow or mint-green spots and streaks over the soft rays of the dorsal, anal, and caudal fins. Pectoral fins are opaque; pelvic fins bear longitudinal turquoise dots or streaks. Head is marked with conspicuous variegated wavy, golden yellow patterns, and freckles over the lachrymal, less conspicuous than in H. deppii that extend onto and around the orbit of the eye. Dorsal view of head is solidmedium sea green over the premaxillae. Conspicuous golden yellow markings on the dorsa extend caudally from the interorbital space onto the dorsal fin base. Scales over the flanks, especially along the ventral half are pigmented in the center with light golden yellow spots, fringed by diamond-shaped aquamarine/medium sea green outline, thicker over the dorsal half, giving flanks a reticulate appearance. Remarks Miller et al. (2005) considered the forms herein described as H. tepehua as an undescribed species. Mitochondrial DNA analysis recovered H. deppii and H. tepehua as paraphyletic, with haplotypes H2 and H3 exclusive to both species. Average Dp = 1.08%, thus separating both species. H. tepehua has become extremely rare and is seldom 123 captured in the Cazones, Tecolutla, and Solteros Rivers. Remaining populations appear to be confined to the most isolated tributaries with turbid water where visibility limits the use of diving masks and homemade spears used by local fishermen. These lineages appear to be critically endangered. Rı́o Pánuco geomorphological group Herichthys carpintis (Fowler, 1903). (Tables 8 and 9 of Supplementary Material 2; Fig. 8, 9 and 10) Synonyms Cichlosoma laurae Regan, 1908. Holotype SU 6162; Laguna del Carpintero, near Tampico, Tamaulipas, Mexico; J. O. Snyder, January 15, 1899. Paratypes BMNH 1900.9.29.172-175 (4), FMNH (1), SU 6201 (29). Diagnosis Differs from H. cyanoguttatus in that adult males develop nuchal humps; from H. teporatus and H. cyanoguttatus in that it has a longer head (mean 38%, SD 1% vs. mean 35 and 35%, SD 2 and 1%, respectively); longer distance from the rostral tip to the pectoral fin origin (mean 37%, SD 1% vs. mean 35 and 35%, SD 2 and 1%, respectively); shorter snout (mean 37%, SD 3% vs. mean 39 and 40%, SD 3 and 3%, respectively); and larger eyes (mean 26%, SD 2% vs. mean 23 and 22%, SD 1 and 2%, respectively). Also differs from other Herichthys in the following autapomorphy: In live specimens, big [1.5 mm, iridescent, thick pearls giving the fish a beveled appearance. Description Description is based on sexually mature specimens over 53.4 mm SL. Morphometric and meristic data are summarized in Tables 8 and 9 of Supplementary Material 2. Body depth 40–50% (mean 44%, SD 2%); HL 35–40% (mean 38%, SD 1%); distance from the rostral tip to the pectoral fin origin length 35–40% (mean 37%, SD 1%); snout short 30–43% (mean 37%, SD 3%); eyes large 22–30% (mean 26%, SD 2%). Dorsal fin XV–XVII (mode XVI, freq 90%) 9–12 (mode 10, freq 56%); anal fin IV–VI (mode V, freq 86%) 7–10 (mode 8, freq 81%); pectoral rays 13–15 (mode 14, freq 82%). Pre-dorsal contour high, steep, and flattened at the front, forming a concavity before the eyes. Adult males develop nuchal humps. Teeth spatulate, chisel-like, bicuspid, weakly bicuspid or a mixture of bicuspid and bluntly pointed conical, undifferentiated in length in the upper and lower jaws, implantation erect, curvature straight, Author's personal copy Hydrobiologia neck wide.; teeth in the outer series of the premaxillae (22–28); tooth-rows in the inner series of the premaxillae (4–5); in lower jaw (4–5). One inner row anteriorly and three rows posteriorly in the upper jaw. Lower and upper arcades rounded. Lower pharyngeal tooth plate moderately stout and broad, two rows of 10–12, pigmented in some populations (e.g., Tamiahua), unpigmented in others (e.g., El Salto), enlarged teeth increasing caudally in size and molarization flank midline, 20–23 conic, progressively compressed teeth along posterior margin. Teeth range from enlarged-molariform to conical in some populations (e.g., Tamiahua). In live fish, ground color olive-green with big [1.5 mm iridescent, thick, Persian-green colored pearls that give the fish a beveled appearance. Dorsal and anal fins the same color as the body, splashed with pearls over the dorsal, anal, and frontal third of the caudal fins. Head marked with round and elongate series of 10–12 spots over lachrymal and cheeks that extend from the lip fold onto the orbit of the eye. Dorsal view of head solid olive-green over the premaxillae. Seven inconspicuous bars are visible in non-breeding coloration. Breeding pigmentation is typical of the genus. Geographic distribution Pánuco River Basin below 1,200 masl, except in the Rı́o Gallinas and its tributaries in Tamasopo; Laguna de Tamiahua and its tributaries in Veracruz; and San Andres Lagoon and its tributaries in Tamaulipas. Remarks This species was described as Neetroplus carpintis by Jordan & Snyder (1899) from type locality Laguna del Carpintero, Tamaulipas; and a number of very small specimens in the Rı́o Verde near Rascón, San Luis Potosı́. Heros laurae was later described by Regan (1908), from type locality Tampico, hence is Junior synonym of H. carpintis. Heros teporatus was described by Fowler in 1903 from type locality Victoria, on the Victoria River, a tributary of the Rı́o Soto la Marina, Tamps., Mexico. Álvarez (1970) considered H. carpintis, a subspecies of H. cyanoguttatus, separate from H. cyanoguttatus teporatus and H. cyanoguttatus cyanoguttatus. In 2003, Kullander synonymized Heros teporatus with H. carpintis. Observations of Álvarez (1970) generally coincide with ours: H. carpintis and H. cyanoguttatus are species too weakly differentiated. Low average Dp = 0.73 and 0.57%, separate H. carpintis from H. teporatus; and H. carpintis from H. cyanoguttatus, respectively. This is in agreement with López- Fernández et al. (2010). These authors noted the ‘‘zero to none’’ 0.1% divergence between Herichthys cyanoguttatus, H. carpintis, and H. tamasopoensis. In our analysis, H. carpintis was the species with the highest haplotype richness within true Herichthys. We detected six Cox1 haplotypes among the specimens surveyed, including haplotypes H9 (*79%), shared with H. cyanoguttatus and H. tamasopoensis; and H10 (*5%), uncommon in H. deppii and very common in H. cyanoguttatus. We also detected haplotypes H11 (*2%) in Tamiahua, H15 (*2%) in Rı́o Guayalejo, H16 (*9%) in the Lower Pánuco, and H17 (*2%) in the Rı́o Guayalejo, representing divergent lineages exclusive to this species (Supplementary Materials 5 and 7). Herichthys tamasopoensis Artigas-Azas, 1993 (Tables 8 and 9 of Supplementary Material 2; Figs. 8 and 9) Holotype UMMZ 221577; adult male, 85.5 mm SL ‘‘Las Cascadas’’ (99°230 4700 W Long., 21°560 4700 N. Lat.) in the Rı́o Tamasopo, J.M. Artigas-Azas. Paratypes UMMZ 221829 (6: 72.6–86.9 mm SL), same data as the holotype. Diagnosis Differs from H. carpintis in that it has a longer caudal peduncle (mean 17%, SD 1% vs. mean 15%, SD 1%), and shorter lower jaw (mean 29%, SD 2% vs. mean 31%, SD 2%). Differs from all other species in Herichthys in the following autapomorphies: Frontal teeth closely set, flattened, truncate, and unicuspid to weakly bicuspid in both jaws; lateral teeth bicuspid. Lower and upper arcades very rounded. Lower pharyngeal tooth plate stout, broad, and rugged; two rows of 10–12 unpigmented, compressed, very small, closely set, molars flank midline, 30–32 compressed teeth along posterior margin. Also differs from other Herichthys in having dorsal profile convex markedly curved between the nuchal area and the first dorsal spine, and in live ground color grayish yellowgreen to olive-green, with body covered with greenish yellow dots *1 mm in diameter. Description Description is based on sexually mature specimens over 57.25 mm SL. Morphometric and meristic data are summarized in Tables 8 and 9 of Supplementary Material 2. Dorsal fin XV–XI (mode XVI, freq 80%), 9–10 (mode 9, freq 50%); anal fin V (freq 100%), 7–8 (mode 8, freq 80%); pectoral rays 13–14 (mode 14, 50%); scales in the lateral line 123 Author's personal copy Hydrobiologia 28–31(mode 30, freq 60%). Body depth 41–47% (mean 44%, SD 1%). Pre-dorsal contour high, steep, and flattened at the front, convex before the eyes; nuchal hump present in older males. Dorsal contour strongly convex, markedly curved between nuchal area and first dorsal spine. Frontal teeth flattened, closely set, truncate, and unicuspid to weakly bicuspid in both jaws; lateral teeth bicuspid, implantation erect, curvature straight, neck wide, decreasing in size posteriorly. Teeth in the outer series of the premaxillae (18), tooth-rows in the inner series of the premaxillae (5), and lower jaw (6–7). One inner row anteriorly and 5–6 rows posteriorly in the upper jaw. Lower and upper arcades very rounded. Lower pharyngeal tooth plate stout, broad, and rugged; two rows of 10–12 unpigmented, compressed, small, closely set molars flank midline, 30–32 compressed teeth along posterior margin. Coloration in preservative Preserved specimens are light brown, darker in the head and dorsal area. Fins are the same color as the body, with dorsal, anal fins dotted with lighter spots. Seven inconspicuous vertical flank bars, bearing a series of dark blotches below the lateral line over the caudal, half make up the principal markings. Live colors Ground color grayish yellow-green to olive-green, darker dorsally; eyes and gill covers purple, body covered with vanilla color dots *1 mm in diameter. Dotting denser over dorsal half, and soft dorsal, anal, and caudal fins; absent on head and chest. Fins opaque. Seven dark blotches over the caudal half below the lateral line make the principal markings (central and caudal more conspicuous). Breeding pigmentation Typical of the genus, ground color changes to khaki. Five vertical dark flank bars contrast pale frontal half. The purple areas around the gill covers and cheeks, and the salmon pigmented areas of dorsal fin intensify. Geographic distribution Mainstream and tributaries of the Rı́o Gallinas, upriver from the Tamul Cascade (Artigas-Azas, 1993). Habitat and associates Habitat is characterized by hard, clear waters with pH 7.8–8.3, over rock substrate. Shares its habitat with Nosferatu pame, N. steindachneri, Xiphophorus montezumae, Gambusia panuco, and Astyanax mexicanus among other fish species. Vernacular name Mojarra de Tamasopo, Tamasopo Herichthys. 123 Remarks Hulsey et al. (2004) and Concheiro-Pérez et al. (2006) recovered H. tamasopoensis as sister to H. carpintis. We recovered a very low Cox1 average divergence (Dp = 0.70%) between H. tamasopoensis and H. carpintis. This is in agreement with LópezFernández et al. (2010). Nonetheless, our morphologic analysis supports the validity of H. tamasopoensis as a separate parapatric species. We detected two mitochondrial haplotypes among the specimens surveyed, including haplotypes H8 (*89%), uncommon in H. tepehua from the Tenixtepec River, and H9 (*11%), very common in H. carpintis and less so in H. cyanoguttatus. North of Sierra de Tamaulipas geomorphological group Herichthys cyanoguttatus Baird & Girard, 1854. (Tables 8 and 9 of Supplementary Material 2; Figs. 8 and 9) Synonyms Heros pavonaceus, Garman 1881; Parapetenia cyanostigma (Hernández-Rolón, 1990). Syntypes ANSP 9097 (1); MCZ 15415 [ex USNM 852]; UMMZ 92113 (1); USNM 851, 852 (now 4). Brownsville, Texas, USA. Collected by Capt. Van Vliet and John H. Clark. Diagnosis Differs from H. teporatus in that it does not develop prominent nuchal humps. Differs from H. carpintis in that it has a shorter head (mean 35%, SD 1% vs. mean 38%, SD 1%) and smaller eyes (mean 22%, SD 2% vs. mean 26%, SD 2%), all in HL; as well as shorter distance from rostral tip to pectoral fin origin (mean 35%, SD 1% vs. mean 37%, SD 1%) in SL. Differs from both H. carpintis and H. teporatus in that it has a longer snout (mean 40%, SD 3% vs. mean 37 and 39%, SD 3 and 3%, respectively). Also differs from other Herichthys in the following autapomorphy: In live fish, small \1 mm iridescent dots cover the flanks. Description Description is based on sexually mature specimens over 61.7 mm SL. Morphometric and meristic data are summarized in Tables 8 and 9 of Supplementary Material 2. Head short 31–38% (mean 35%, SD 1%); distance from rostral tip to pectoral fin origin short 31–37% (mean 35%, SD 1%); snout long 33–46% (mean 40%, SD 3%); eyes small 18–27% (mean 22%, SD 2%). In live fish, small \1 mm iridescent dots cover the flanks. Dorsal fin XV–XII (mode XVI, freq 60%) 5–12 (mode 11, freq 55%); anal Author's personal copy Hydrobiologia fin IV–VI (mode V, freq 75%) 8–10 (mode 9, freq 63%); pectoral rays 13–15 (mode 14, freq 88%); scales in the lateral line 28–32 (mode 29, freq 38%). Body depth 40–51% (mean 46%, SD 2%). Predorsal contour high, steep, and flattened at front, forming an angular depression before the eyes; nuchal hump absent in large males. Frontal teeth regularly set, decreasing in size posteriorly, spatulate, truncate and weakly bicuspid, implantation erect, curvature straight, neck wide, undifferentiated in length in upper and lower jaws. Teeth in the outer series of premaxillae (24); toothrows in the inner series of the premaxillae (2–3) and lower jaw (4); one inner row anteriorly and 1–2 rows posteriorly in the upper jaw. Lower and upper arcades rounded. Lower pharyngeal tooth plate moderately stout and broad, two rows of 11 pigmented, enlarged teeth increasing caudally in size and molarization flank midline, 24 conical, progressively compressed teeth along posterior margin. Ground color is grayish to asparagus-green with small \1 mm iridescent blue to olive-green dots scattered over body and fins. Dot diameter, density, patterns, and arrangements vary among lineages; from diminutive \0.5, densely and randomly scattered olive-green iridescent dots in the northernmost populations of the Salado-Nadadores and Ocampo sub-basins in Coahuila and Nuevo León, to larger 0.5–1 mm, horizontally aligned iridescent blue dots in the southernmost lineage of the San Fernando River Basin. Head is marked with series of 3–15 diagonal rows of dots over lachrymal and cheeks that extend from the edge of the lip fold onto the orbit of the eye. Dorsal view of the head is solid grayish to asparagus-green over the premaxillae. Breeding pigmentation Typical of the genus. Six vertical flank bars fuse darkening the entire posterior half of body, contrasting the pale frontal half of the fish. Geographic distribution Lower Rı́o Grande/Bravo basin in the US and Mexico, excluding the Conchos River in Chihuahua; and the San Fernando River Basin in Nuevo León and Tamaulipas, Mexico. Habitat and associates H. cyanoguttatus is found sympatric with Centrachid species (Lepomis sspp. and Micropterus salmoides) throughout its distribution range. Vernacular name Texas cichlid, guapota del rı́o Bravo. Remarks This species was described as Herichthys cyanoguttatus by Baird & Girard (1854) from type locality Brownsville, Texas, U.S. Heros pavonaceus, syntypes MCZ 24877 (5) and UMMZ 95837 (1) was described by Garman (1881) from a spring in the proximities of Monclova, Coahuila. Confusion always existed with respect to the type locality largely because type material was collected in an area of the Salado-Nadadores River Basin near the Cuatro Cienegas Valley. Meek’s (1904) account of H. pavonaceus is very general and matches about any species in Herichthys. Álvarez (1970) sets the distribution of H. pavonaceus as Coahuila and Nuevo León (in the Salado-Nadadores River Basin). Kullander (2003) considers H. pavonaceous a synonym of H. cyanoguttatus. Our analysis is in agreement with Kullander (2003). Some specimens within the allopatric San Fernando River lineage have in life different color patterns from H. cyanoguttatus of the Rı́o Grande Basin. The first have distinct, larger blue dots that are aligned in horizontal lines over the flanks. HernándezRolón (1990) described Parapetenia cyanostigma from type locality Playa Bruja, Tequesquitengo, Mexico. His description is based on an introduced population of H. cyanoguttatus. Álvarez (1970) considered H. carpintis and H. teporatus subspecies of H. cyanoguttatus. The observations of Álvarez (1970) generally coincide with our observations that H. carpintis and H. cyanoguttatus are weakly differentiated species. We recovered a low Cox1 average mitochondrial divergence (Dp = 0.57%) between H. carpintis and H. cyanoguttatus. This is in agreement with López-Fernández et al. (2010). We detected four mitochondrial haplotypes among the specimens surveyed, including haplotypes H1 (*6%) and H4 (*6%), shared with H. deppii; and H9 (*17%), shared with H. carpintis, H. tamasopoensis, and H. tepehua n. sp., as well as H10 (*72%), shared with H. carpintis and H. tepehua n. sp. (Supplementary Materials 5 and 7). Herichthys teporatus (Fowler, 1903) (Tables 8 and 9 of Supplementary Material 2; Fig. 8) Holotype ANSP 24242. Diagnosis Differs from H. cyanoguttatus in that adult males develop nuchal humps. Live H. teporatus also differs from H. cyanoguttatus in having irregularly shaped iridescent spots 1–1.5 mm covering the entire body versus iridescent dots \1 mm. Differs from H. carpintis in having shorter heads (mean 35%, 123 Author's personal copy Hydrobiologia SD 2% vs. mean 37%, SD 1%); smaller eyes (mean 23%, SD 1% vs. mean 26%, SD 2%); shorter distance from the rostral tip to the anal fin origin (mean 67%, SD 2% vs. mean 71%, SD 2%); and shorter distance from rostral tip to pectoral fin origin (mean 35%, SD 2% vs. mean 37%, SD 1%). Description Description is based on sexually mature specimens over 63.7 mm SL. Morphometric and meristic data are summarized in Tables 8 and 9 of Supplementary Material 2. Dorsal fin XV–XVI (mode XVI, freq 90%) 10–12 (mode 11, freq 70%); anal fin V–VI (mode VI, freq 50%) 9 (freq 100%); pectoral rays 14 (freq 100%); scales in the lateral line 28–33 (mode 31, freq 30%). Body depth 42–49% (mean 45%, SD 2%). Pre-dorsal contour steep, flattened at the front, forming a depression before the eyes. Adult males develop nuchal humps. Frontal teeth are regularly set, decreasing in size posteriorly, spatulate, truncate and weakly bicuspid, implantation erect, curvature straight, neck wide. Teeth in the outer series of the premaxillae (18); tooth-rows in the inner series of the premaxillae (4) and lower jaw (4); one inner row anteriorly and 3 rows posteriorly in the upper jaw. Lower and upper arcades rounded. Lower pharyngeal tooth plate moderately stout and broad, two rows of 10–12 lightly pigmented, enlarged teeth increasing caudally in size and molarization flank midline, 20 conical, progressively compressed teeth along posterior margin. Coloration in preservative Olive-brown with six to nine vertical flank bars bearing a series of dark blotches fading out below the lateral line make up the principal markings. Live colors Ground color is olive-green with a series of horizontal rows of asymmetric turquoise iridescent spots 1–1.5 mm in adults, scattered throughout the entire body. Tip of dorsal fin rays red in some individuals; a conspicuous dark blotch over the middle of the dorsal fin may be present in both males and females. Pectoral fins opaque; pelvic fins have longitudinal blue streaks. Seven inconspicuous bars are visible in non-breeding coloration. Breeding pigmentation Typical of the genus. Geographic distribution Soto La Marina River Basin, including the Rı́o Blanco in Aramberri, Nuevo León, as well as some of the smaller coastal rivers and creeks of the Villa Aldama Volcanic Complex, including Rı́o Carrizal and Arroyo Tepehuajes in El Panal, Tamaulipas. 123 Habitat and associates In the Rı́o Blanco and Rı́o Purificación at elevations above 1,000 m, it shares habitat with Xiphophorus xiphidium, whereas in the lower reaches it shares habitat with Poecilia mexicana, Poecilia formosa, and Fundulus sp. nov (under description by one coauthor) among other fish species. Vernacular name Green Texas cichlid, guapota del Soto la Marina. Remarks Heros teporatus was described by Fowler (1903) from type locality Victoria, on the Victoria River, a tributary of the Rı́o Soto la Marina, Tamps., Mexico. In 1904, Meek synonymized H. teporatus with H. cyanoguttatus. Álvarez (1970) considered H. teporatus a subspecies of H. cyanoguttatus and used the trinomen H. cyanoguttatus teporatus. In 2003, Kullander again synonymized Heros teporatus, this time with Herichthys carpintis, but noted the need for further analysis. Miller et al. (2005) mentioned that H. cyanoguttatus and H. carpintis possibly hybridize within sections of the Soto la Marina River Basin. Our morphometric analysis places H. teporatus phenetically closer to H. cyanoguttatus than H. carpintis. We recovered a low Cox1 mitochondrial divergence average (Dp = 0.73 and 0.42%) separate H. teporatus from H. carpintis and H. cyanoguttatus, respectively. Morphometric proportions between H. teporatus and H. cyanoguttatus are indistinguishable. Counts are indistinguishable between the three. We detected two mitochondrial haplotypes among the specimens surveyed, including haplotypes H4 (*17%), shared with H. deppii, and the more common H10 (*80%), very common in H. cyanoguttatus, less so in H. carpintis, and uncommon in H. tepehua n. sp. Herichthys minckleyi (Kornfield & Taylor, 1983) (Figure 8) Holotype UMMZ 209434, 93.4 mm SL male, Deep bodied form, papilliform morph, Posos[sic] de la Becerra, 15.7 km by road SSW of Cuatro Ciénegas de Carranza, Coahuila, Mexico, R.R. Miller and family, C.L. Hubbs, W.L. Minckley, D.R. Tindall, and J.E. Craddock, April 6, 1961. Remarks Kornfield & Taylor (1983) hypothesized that H. minckleyi appears to be more closely related to the cichlids described herein as Nosferatu new genus, specifically N. labridens, N. bartoni, N. pame, and N. steindachneri than to H. cyanoguttatus on the basis of jaw dentition. Hulsey et al. (2004) recovered Author's personal copy Hydrobiologia H. minckleyi as sister taxa to a clade shared with H. cyanoguttatus, H. carpintis, and H. tamasopoensis. Discussion Taxonomic implications Our molecular and morphologic analysis supports Nosferatu new genus as distinct from true Herichthys. Our results concur with De la Maza-Benignos & Lozano-Vilano’s (2013) interpretation of the labridens assemblage = Nosferatu new genus conformed with N. steindachneri (restricted to Rı́o Tamasopo), N. labridens (restricted to Media Luna and its surroundings), N. pratinus (restricted to Rı́o el Salto, above Micos cascades), N. pame (restricted to the Rı́o Tamasopo), the polymorphic N. pantostictus that includes a number of parapatric lotic and lentic forms inhabiting the lower reaches of the Pánuco-Tamesi River Basin and the Tamiahua and San Andres coastal lagoon systems, and N. molango (restricted to Laguna Azteca). This last species could correspond to a phenomenon of secondary contact (Nosil et al. 2009) between both genera, as N. molango exhibited mitochondrial DNA affinity to true Herichthys. Furthermore, phylogenetic studies, particularly using nuclear DNA loci, are needed to clarify its taxonomic status. It has been hypothesized, on the basis of morphological similarities between both species (Kornfield & Taylor, 1983; Artigas-Azas, 2006), that N. pame and N. steindachneri evolved sympatrically, and it was inferred that these were analogous to the polymorphic H. minckleyi in the Cuatro Ciénegas Basin (Kornfield & Taylor, 1983). Furthermore, Artigas-Azas (2006) stated on the basis of distributions of the two species that N. pame is found upriver of the waterfalls of Rı́o Tamasopo, whereas N. steindachneri is not, that N. pame is the ancestral species of the two. However, our results suggest both species together with allopatric N. pratinus form a paraphyletic group. This hypothesis is supported by the presence of shared haplotypes between N. steindachneri and N. pratinus and not shared with N. pame, which could be the result of incomplete lineage sorting in the marker used (Nosil & Sandoval, 2008; Takahashi et al., 2001) or secondary contact (Nosil et al., 2009) from H. pratinus into the Tamasopo River Basin. This latter hypothesis is supported by lower divergence between allopatric species than between sympatric (average Dp = 0.32 ± 000.15 and Dp = 000.87 ± 000.97, respectively). Our analysis supports that N. labridens and N. bartoni recently diverged from a common ancestor, on the basis of all haplotypes being shared by the two. Further analysis at the phylogenetic level may help resolve their evolutionary history. None of the species within true Herichthys were recovered as monophyletic. Nonetheless, our analysis supports H. deppii ? H. tepehua n. sp. (i.e., south of the Sierra de Tantima geomorphological group) as distinct from the rest of the genus. This is on the basis of high levels of divergence and the presence of exclusive haplotypes to this group (Supplementary Materials 5 and 7). Our molecular analysis, however, does not support H. tamasopoensis, H. cyanoguttatus, H. teporatus, and H. carpintis as distinct from each other. These species exhibited very low (Dp \ 0.75%) divergence values, and no differences in haplotypes with the more genetically diverse H. carpintis. Our findings concur with those of López-Fernández et al. (2010), who reported divergence values of *0.1% among the three. Nonetheless, our morphometric analysis exhibited the three species as clearly divergent and diagnosable from H. carpintis ecologically, chromatically, and morphologically. True Herichthys currently includes seven morphologically distinct diagnosable species: H. deppii (restricted to the Nautla-Misantla River basins) and the allopatric H. tepehua n. sp. (restricted to the Pantepec, Cazones, Tenixtepec, Tecolutla, and Solteros river systems); H. carpintis that includes a number of parapatric lotic and lentic forms inhabiting the Pánuco-Tamesı́ River Basin, except in the Rı́o Tamasopo, and the Tamiahua and San Andres coastal lagoon systems; H. tamasopoensis (restricted to the Rı́o Tamasopo); H. teporatus restricted to the Rı́o Soto La Marina; H. cyanoguttatus that includes a number of allopatric forms inhabiting the Rı́o San Fernando, the Rı́o Grande, and adjacent rivers in southeast Texas; and the polymorphic H. minckleyi (endemic to the Cuatro Ciénegas Valley). Biogeographical patterns Our estimated divergence time across the TMVB at *7 Mya (*5–11 Mya 95% HPD) of the stem group (Herichthys ? Nosferatu) from its sister clade corresponds with the formation of the Chiconquiaco-Palma 123 Author's personal copy Hydrobiologia Sola Massif. The Massif is located between the Gulf of Mexico and the Sierra Madre Oriental (SMO). It is part of the Eastern Alkaline Province (EAP), a volcanic-belt that stretches 2,000 km in NNW–SSE direction from northern Coahuila to Palma Sola, Veracruz, along the GOM coastal plains (Ferrari et al., 2005; Avto et al., 2007) and intersects the TMVB at the ChiconquiacoPalma Sola Massif (6.9–3.2 Mya) in central Veracruz (Vasconcelos-Fernández & Ramı́rez-Fernández, 2004; Ferrari et al., 2005; Avto et al., 2007). Moreover, divergence time estimated for the split between both genera was *5 Mya (3–8 Mya 95% HPD). Our timing coincides with intense volcanism in the Miocene-Pliocene that led to the formation of the sedimentary Rı́o Verde Basin (Planer-Friedrich, 2000). This is also reflected in other groups with similar divergence times (Ornelas-Garcı́a et al., 2008). Sudden subsidence of the graben structure in the basin during the Quaternary created a drain-less depression filled by shallow lakes that were subject to intensive evaporation in a semiarid environment (Planer-Friedrich, 2000). It was in some of these lakes that Nosferatu new genus diverged into the bartoni (*3Mya), the steindachneri, and the pantostictus clades (*2 Mya). Regional faulting in the WNW–ESE direction during the course of the Pleisteocene (*1.8 Mya) rejoined the Rı́o Verde with the Rı́o Pánuco, draining most of the lakes, as the Rı́o Verde cut its way through the valley (Planer-Friedrich, 2000; CONAGUA, 2002), allowing reinvasion of Nosferatu new genus into the Rı́o Pánuco, this time with evolved mechanisms of reproductive isolation that allowed sympatry of the two genera. In the coastal plains, true Herichthys did not begin speciation until *1 Mya, despite the significant volcanic activity that occurred over the Tampico– Misantla Structural Basin during the Miocene–Quaternary Period. Volcanic activity in the area is well represented by the Sierra de Tantima and the Alamo volcanic field (7.6–6.6 Mya); the Tlanchinol flows (7.3–5.7 Ma), Molango (7.4–6.5 Mya) (Ferrari et al., 2005); the Huejutla lava fields (7.3–2.87 Mya); the lava fields around Palma Sola (6.9–3.2 Mya) (Avto et al., 2007); the Huautla flow (*2.82 Mya); Metlaltoyuca (1.6–1.3 Mya) (Ferrari et al., 2005); and the Poza Rica lavas (1.3–1.62 Mya) (Avto et al., 2007). These events filled paleovalleys and redefined catchment delimitations in the Pánuco, Cazones, Pantepec, Tecolutla, and Solteros river systems. Moreover, there were repeated raising and lowering of sea level during 123 the Pleistocene, which would have allowed ancestral Herichthys to bypass any terrestrial boundary to dispersal (Hulsey et al. 2004). North of the Rı́o Pánuco Basin, over the Burgos Structural Basin, the EAP is represented by the CandelaMonclova belt (3.4–1.8 Mya) (Aranda-Gómez et al. 2005), the Villa Aldama volcanic complex (1.8–0.250 Mya) (Camacho, 1993; Vasconcelos-Fernández & Ramı́rez-Fernández, 2004; Gary & Sharp, 2006), and the Plio-quaternary sections of the Sierra de San Carlos-Cruillas and Sierra de Tamaulipas (Demant & Robin, 1975; Bryant et al. 1991). Similar to the Sierra de Tantima, the Sierra de Tamaulipas has functioned as a biogeographic barrier to the dispersal of N. pantostictus, and consequently to the modern H. carpintis, into the Pantepec and the Soto la Marina basins, respectively. Climate change in the last 4 million years includes the end of the warm period (5–3 Mya) and significant intensification of Northern Hemisphere glaciations *2.75 Mya (Ravelo et al., 2004). Higher winter precipitation with significantly cooler and wetter conditions than today prevailed in northern Mexico during the Pleistocene and early Holocene. Quaternary glacial sequences in central Mexico indicated that there were at least five glacial advances in the late Pleistocene and in the Holocene (Metcalfe et al., 2000; Metcalfe, 2006). We hypothesize that these events devastated cichlid populations north of parallel 24°N and possibly farther South, except for localized refuges with warm springs such as Cuatro Ciénegas. In a recent study, Řičan et al., (2012) dated to 5.6 Mya the cladogenetic event that created H. minckleyi. Our demographic history analysis of all the other species of true Herichthys exhibited a contraction during the lower Pleistocene that coincides with some of the glacial advances in North America during that period. These advances would have forced the range contraction of Herichthys (except H. minckleyi, which remained in the warm springs of Cuatro Ciénegas) to the Rı́o Pánuco Basin as the climate deteriorated. In more recent times, around 65,000 years ago, some populations of Herichthys would have expanded once more from the Rı́o Pánuco and reinvaded the rivers to the north. Acknowledgments The authors wish to thank Marco Arroyo, Hazzaed Ochoa, Alejandro Espinosa, Emanuel Pimentel, Jesús Maria Leza Hernández, and Anarbol Leal for their most valuable and enthusiastic help throughout the collection period. We thank Carlos Pedraza for his helpful suggestions Author's personal copy Hydrobiologia on an early version of the manuscript, and Aslam Narváez-Parra for the technical drawings of the lower pharyngeal plates and frontal teeth in Figs. 9 and 10, and two anonymous reviewers which advice helped improving crucial parts of the manuscript. We also thank the funding support derived from the projects CGL2006-1235-BOS and CGL2010-15231-BOS. Samples were collected under permit number DGOPA/13323/007015/.-6943 issued to MLLV and vouchered at Colección Ictiológica de la Facultad de Ciencias Biológicas de la UANL. References Álvarez, J., 1970. Peces Mexicanos (Claves). Servicio de Investigaciones Pesqueras Nacional. Instituto Nacional de Investigaciones Biológicas, Mexico: 166 pp. Aranda-Gómez, J. J., J. F. Luhr, T. B. Housh, G. Valdez-Moreno & G. Chávez-Cabello, 2005. El volcanismo tipo intraplaca del Cenozoico tardı́o en el centro y norte de Mexico: una revisión. Boletı́n de la Sociedad Geológica Mexicana, Volumen Conmemorativo del Centenario, Temas Selectos de la Geologı́a Mexicana 3: 187–225. Antoine, J. W., 1972. Structure of the Gulf of Mexico. In Rezak, R., & V. J. Henry (eds), Contributions on the Geological and Geophysical Oceanography of the Gulf of Mexico, Texas A&M University Oceanographic Studies, Vol. 3. Gulf Publishing Co., Houston, TX: 1–34. Artigas-Azas, J. M., 1993. Herichthys tamasopoensis n. sp., a new cichlid from Mexico (Pisces, Cichlidae). In Konings, A. (ed.), Cichlids Yearbook 3. Cichlid Press, El Paso, TX: 65–70. Artigas-Azas, J. M., 2006. Herichthys steindachneri. Cichlid News 15: 15–22. Avto, G., A. M. Petronille, B. Henry, L. A. Valdivia, J. Morales & J. Urrutia-Fucugauchi, 2007. Paleomagnetism of the Eastern Alkaline Province (Mexico): contribution to the time-averaged field global database and geomagnetic instability time scale. Earth Planets Space 59: 775–783. Bandelt, H. J., P. Forster & A. Rohl, 1999. Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution 16: 37–48. Baird, S. F. & C. F. Girard, 1854. Descriptions of new species of fishes collected in Texas, New Mexico and Sonora, by Mr. John H. Clark, on the U. S. and Mexican Boundary Survey, and in Texas by Capt. Stewart Van Vliet, USA. Proceedings of the Academy of Natural Sciences of Philadelphia 7: 24–29. Bean, T. H., 1892. Notes on fishes collected in Mexico by Prof. Alfredo Dugès, with descriptions of new species. Proceeding of the U. S. National Museum: 283–287. Bryant, W. R., J. Lugo, C. Córdova & A. Salvador, 1991. Physiometry and Bathymetry. The Geology of North America, Vol. J, The Gulf of Mexico Basin. The Geological Society of America, Boulder, CO: 13–30. Byerly, G. R., 1991. Igneous activity. In Salvador, A. (ed.), The Gulf of Mexico Basin: The Geology of North America. The Geological Society of America, Boulder, CO: 91–108. Camacho, A. F., 1993. Compilación geológica de la Vertiente del Golfo de México, Area 1. Comisión Federal de Electricidad, Subdirección Técnica, Gerencia de Estudios de Ingenierı́a Civil, Subgerencia de Estudios Geológicos: Departamento de Geologı́a 43: 123–130. Chakrabarty, P., 2007. A Morphological Phylogenetic Analysis of Middle American Cichlids with Special Emphasis on the Section ‘Nandopsis’ sensu Regan, Vol. 198. Miscellaneous Publications, Museum of Zoology, University of Michigan, Ann Arbor, MI: 1–31. CONAGUA, 2002. Determinación de la disponibilidad de agua en el Acuı́fero Rı́o Verde, estado de San Luis Potosı́, Comisión Nacional del Agua Subdirección General Técnica, Gerencia de Aguas Subterráneas, Subgerencia de Evaluación y Modelación Hidrogeológica. Retrieved 20 December, 2011. http://www.conagua.gob.mx/CONAGUA07/Noticias/ DR_2415.pdf. Concheiro-Pérez, G. A., I. Oldlich, G. Ortı́, E. Bermingham, I. Doadrio & R. Zardoya, 2006. Phylogeny and biogeography of 91 species of heroine cichlids (Teleostei: Cytochrome b gene). Molecular Phylogenetics and Evolution 43: 91–110. Contreras-Balderas, S., H. Obregón & M. L. Lozano-Vilano, 1996. Punta del Morro, una interesante barrera de los patrones de distribución de los peces continentales en el norte y centro de Veracruz, México. Acta Biológica 16: 37–42. De la Maza-Benignos, M. & M. L. Lozano-Vilano, 2013. Description of three new species of the genus Herichthys (Perciformes: Cichlidae) from eastern Mexico, with redescription of H. labridens, H. steindachneri, and H. pantostictus. Zootaxa 3734: 101–129. Demant, A. & C. Robin, 1975. Las fases del vulcanismo en México: una sı́ntesis en relación con la evolución geodinámica desde el Cretácico. Revista del Instituto de Geologı́a de la UNAM 1: 66–79. Doadrio, I. & O. Dominguez, 2004. Phylogenetic relationships within the fish family Goodeidae based on cytochrome b sequence data. Molecular Phylogenetics and Evolution 31: 416–430. Doadrio, I. & A. Perdices, 2005. Phylogenetic relationships among the Ibero-African cobitids (Cobitis, cobitidae) based on cytochrome b sequence data. Molecular Phylogenetics and Evolution 37: 484–493. Drummond, A. & A. Rambaut, 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology 7: 214. Elliott, N. G., K. Haskard & J. A. Koslow, 1995. Morphometric analysis of orange roughy (Hoplostethus atlanticus) off the continental slope of southern Australia. Journal of Fish Biology 46: 202–220. Eschmeyer, W. N. (ed). 2013 Catalog of fishes: genera, species. (http://research.calacademy.org/research/ichthyology/ catalog/fishcatmain.asp). Electronic version accessed 26 December, 2013 . Farias, I. P., G. Ortı́, I. Sampaio, H. Schneider & A. Meyer, 1999. Mitochondrial DNA phylogeny of the family Cichlidae: monophyly and fast molecular evolution of the Neotropical assemblage. Journal of Molecular Evolution 48: 703–711. Farias, I. P., G. Ortı́ & A. Meyer, 2000. Total evidence, molecules, morphology, and the phylogenetics of cichlid fishes. Journal of Experimental Zoology 288: 76–92. Farias, I. P., G. Ortı́, I. Sampaio, H. Schneider & A. Meyer, 2001. The Cytochrome b gene as a phylogenetic marker: 123 Author's personal copy Hydrobiologia the limits of resolution for analyzing relationships among cichlid fishes. Journal of Molecular Evolution 53: 89–103. Ferrari, L., M. López-Martinez, G. Aguirre-Dı́az & G. CarrascoNuñez, 1999. Space–time patterns of Cenozoic arc volcanism in central Mexico. From the Sierra Madre Occidental to the Mexican Volcanic Belt. Geology 27: 303–306. Ferrari, L. T., M. Tagami, M. Eguchib & M. T. Orozco-Esquivela, 2005. Geology, geochronology and tectonic setting of late Cenozoic volcanism along the southwestern Gulf of Mexico: The Eastern Alkaline Province revisited. Journal of Volcanology and Geothermal Research 146: 284–306. Fowler, H. W., 1903. Life colors of Poecilia limantouri and description of a new Heros from Mexico. Proceedings of the Academy of Natural Sciences of Philadelphia 63: 493–520. Garman, S., 1881. New and little-known reptiles and fishes in the museum collections. Bulletin of the Museum of Comparative Zoology 8: 85–93. Gary, M. O. & J. M. J. Sharp, 2006. Volcanogenic karstification of Sistema Zacatón, Mexico. Geological Society of America Special Papers 404: 79–89. Heckel, J. J., 1840. Johann Natterer’s neue Flussfische Brasilien’s nach den Beobachtungen und Mittheilungen des Entdeckers beschrieben (Erste Abtheilung, Die Labroiden). Annalen des Wiener Museums der Naturgeschichte 2: 325–471. Hernández-Rolón, A., 1990. Un nouveau cichlidé du système du Rio Balsas, Mexique (Pisces, Teleostei). Revue Française des Cichlidophiles 11: 4–13. Hrbek, T., J. Seckinger & A. Meyer, 2007. A phylogenetic and biogeographic perspective on the evolution of poeciliid fishes. Molecular Phylogenetics and Evolution 43: 986–998. Huelsenbeck, J. P. & F. Ronquist, 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17: 754–755. Hulsey, C. D., F. J. Garcı́a de León & Y. Sánchez-Jo, 2004. Temporal diversification of Mesoamerican cichlid fishes across a major biogeographic boundary. Molecular Phylogenetics and Evolution 31: 754–764. Jordan, D. S. & J. O. Snyder, 1899. Notes on a Collection of Fishes from the Rivers of Mexico, with Description of Twenty New Species. Proceedings of the Academy of Natural Sciences of Philadelphia 19: 115–147. Kornfield, I. & J. N. Taylor, 1983. A new species of polymorphic fish Cichlasoma minckleyi, from Cuatro Ciénegas, Mexico (Teleostei: Cichlidae). Proceedings of the Biological Society of Washington 96: 253–269. Kullander, S. O., 1996. Heroina isonycterina, a new genus and species of cichlid fish from Western Amazonia, with comments on Cichlasomine systematics. Ichthyological exploration of freshwaters 7: 149–172. Kullander, S. O., 1998. A phylogeny and classification of the South American Cichlidae (Teleostei: Perciformes). In Malabarba, L. R., R. E. Reis, R. P. Vari, Z. M. Lucena & C. A. S. Lucena (eds), Phylogeny and Classification of Neotropical Fishes. EdiPUCRS, Porto Alegre: 461–498. Kullander, S. O., 2003. Check List of the Freshwater Fishes of South and Central America. In Reis, E., S. O. Kullander & C. J. Ferraris (eds), Check List of the Freshwater Fishes of South and Central America. EdiPUCRS, Porto Alegre: 635–636. 123 Kullander, S. O., M. Norén, G. B. Friðriksson & C. A. Santos de Lucena, 2010. Phylogenetic relationships of species of Crenicichla (Teleostei: Cichlidae) from southern South America based on the mitochondrial cytochrome b gene. Journal of Zoological Systematics and Evolutionary Research 48: 248–258. Librado, P. & J. Rozas, 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25: 1451–1452. López-Fernández, H., K. O. Winemiller & R. L. Honeycutt, 2010. Multilocus phylogeny and rapid radiations in Neotropical cichlid fishes (Perciformes: Cichlidae). Molecular Phylogenetics and Evolution 55: 1070–1086. López-Fernández, H., R. L. Honeycutt & K. O. Winemiller, 2005a. Molecular phylogeny and evidence for an adaptive radiation of geophagine cichlids from South America (Perciformes: Labroidei). Molecular Phylogenetics and Evolution 34: 227–244. López-Fernández, H., L. R. Honeycutt, M. L. J. Stiassny & K. O. Winemiller, 2005b. Morphology, molecules, and character congruence in the phylogeny of South American geophagine cichlids (Perciformes, Labroidei). Zoologica Scripta 34: 627–651. Martin, A. P. & E. Bermingham, 1998. Systematics and evolution of Lower Central American Cichlids inferred from analysis of Cytochrome b gene sequences. Molecular Phylogenetics and Evolution 9: 192–203. Mateos, M., O. I. Sanjur & R. C. Vrijenhoeck, 2002. Historical biogeography of the live bearing fish genus Poeciliopsis (Poeciliidae: Cyprinodontiformes). Evolution 56: 972–984. McMahan, C. D., A. D. Geheber & K. R. Piller, 2010. Molecular systematics of the enigmatic Middle American genus Vieja (Teleostei: Cichlidae). Molecular Phylogenetics and Evolution 57: 293–300. Meek, S. E., 1904. The Freshwater Fishes of Mexico North of the Isthmus of Tehuantepec. Zoological Series, Field Columbian Museum, Chicago, IL: 210–212. Metcalfe, S. E., 2006. Late Quaternary environments of the northern deserts and Central Transvolcanic Belt of Mexico. Annals of the Missouri Botanical Garden 93: 258–273. Metcalfe, S. E., S. L. O’Hara & M. Caballero, 2000. Late Quaternary climate change in Mexico. Quaternary Science Reviews 19: 699–721. Miller, R. R., W. L. Minckley & S. M. Norris, 2005. Freshwater fishes of Mexico. Museum of Zoology, University of Michigan/University of Chicago Press, Ann Arbor, MI/ Chicago, IL: 490 pp. Murphy, W. J., J. E. Thomerson & G. E. Collier, 1999. Phylogeny of the Neotropical killifish family Rivulidae (Cyprinodontiformes, Aplocheiloidei) inferred from mitochondrial DNA sequences. Molecular Phylogenetics and Evolution 13: 289–301. Musilová, Z., O. Řičan, K. Janko & J. Novák, 2008. Molecular phylogeny and biogeography of the Neotropical cichlid fish tribe Cichlasomatini (Teleostei: Cichlidae: Cichlasomatinae). Molecular Phylogenetics and Evolution 46: 659–672. Nosil, P. & C. P. Sandoval, 2008. Ecological niche dimensionality and the evolutionary diversification of stick insects. PLoS One 3: 1907. Author's personal copy Hydrobiologia Nosil, P., L. Harmon & O. Seehausen, 2009. Ecological explanations for (incomplete) speciation. Trends in Ecology & Evolution 24: 145–156. Ornelas-Garcı́a, C. P., O. Dominguez-Dominguez & I. Doadrio, 2008. Evolutionary history of the fish genus Astyanax Baird & Girard (1854) (Actinopterygii, Characidae) in Mesoamerica reveals multiple morphological homoplasies. BMC Evolutionary Biology 8: 340. Pellegrin, J., 1903. Description de Cichlidés nouveaux de la collection du Muséum. Bulletin du Musée National d’Histoire Naturelle: 120–125. Perdices, A., E. A. M. Bermingham & I. Doadrio, 2002. Evolutionary history of the genus Rhamdia (Teleostei: Pimelodidae) in Central America. Molecular Phylogenetics and Evolution 25: 172–189. Perdices, A., I. Doadrio & E. A. M. Bermingham, 2005. Evolutionary history of the synbranchid eels (Teleostei: Synbranchidae) in Central America and the Caribbean islands inferred from their molecular phylogeny. Molecular Phylogenetics and Evolution 37: 460–473. Planer-Friedrich, B., 2000. Hydrogeological and hydrochemical investigations in the Rio Verde Basin, Mexico. Freiberg Online Geoscience 3: 1–145. Posada, D., 2008. jModelTest: phylogenetic model averaging. Molecular Biology and Evolution 7: 1253–1256. Rambaut, A. & A. Drummond, 2007. Tracer [computer program]. http://tree.bio.ed.ac.uk/software/tracer. Ravelo, A. C., D. H. Andreasen, M. Lyle, A. Olivarez-Lyle & M. W. Wara, 2004. Regional climate shifts caused by gradual global cooling in the Pliocene epoch. Nature 429: 263–267. Regan, C. T., 1905. A revision of the fishes of the American cichlid genus Cichlosoma [sic] and of the allied genera. Annals and Magazine of Natural History (Series 7): 436. Regan, C. T., 1906–1908. Pisces. In Godman, F. D., O. Salvin (eds), Biologia Centrali Americana, London: 203 pp. Řı́čan, O., R. Zardoya & I. Doadrio, 2008. Phylogenetic relationships of Middle American cichlids (Cichlidae, Heroini) based on combined evidence from nuclear genes, mtDNA, and morphology. Molecular Phylogenetics and Evolution 49: 941–957. Řı́čan, O., L. Piálek, R. Zardoya, I. Doadrio & J. Zrzavy, 2012. Biogeography of the Mesoamerican Cichlidae (Teleostei: Heroini): colonization through the GAARlandia land bridge and early diversification. Journal of Biogeography: 1–15. Roe, K. J., D. Conkel & C. Lydeard, 1997. Molecular systematics of Middle American Cichlid fishes and the evolution of trophic-types in ‘Cichlasoma (Amphilophus)’ and ‘C. (Thorichthys)’. Molecular Phylogenetics and Evolution 7: 366–376. Ruiz-Campos, G., F. Camarena-Rosales, A. Varela-Romero, S. Sánchez-Gonzáles & J. De La Rosa-Vélez, 2003. Morphometric variation of wild trout populations from northwestern Mexico (Pisces: Salmonidae). Reviews in Fish Biology and Fisheries 13: 91–110. Sambrook, J., E. Fritsch & T. Maniatis, 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Laboratory, New York. SEMARNAT 2010 (Secretarı́a del Medio Ambiente y Recursos Naturales). Norma Oficial Mexicana NOM–059–ECOL– 2001, protección ambiental–especies nativas de Mexico y de flora y fauna silvestres–categorı́as de riesgo y especificaciones para su inclusión, exclusión o cambio–lista de especies en riesgo. Shapiro, B., A. Rambaut & A. J. Drummond, 2006. Choosing appropriate substitution models for the phylogenetic analysis of protein-coding sequences. Molecular Biology and Evolution 23: 7–9. Schmitter-Soto, J. J., 2007. Phylogeny of species formerly assigned to the genus Archocentrus (Perciformes: Cichlidae). Zootaxa 1618: 1–50. Sides, J. & C. Lydeard, 2000. Phylogentic utility of the tyrosine kinase gene X-src for assessing relationships among representative cichlid fishes. Molecular Phylogenetics and Evolution 41: 51–74. Snoeks, J., 1994. The Haplochromines (Teleostei, Cichlidae) of Lake Kivu (East Africa). Annales-Musée Royale de l’Afrique Centrale. Sciences Zoologiques 270: 1–221. Stamatakis, A., 2006. RAxML-VI-HPC: maximum likelihoodbased phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22: 2688–2690. Stawikowski, R. & U. Werner, 1998. Die Buntbarsche Amerikas, Vol. 1. Eugen Ulmer, Stuttgart: 540. Takahashi, K., Y. Terai, M. Nishida & N. Okada, 2001. Phylogenetic relationships and ancient incomplete lineage sorting among Cichlid fishes in Lake Tanganyika as revealed by analysis of the insertion of retroposons. Molecular Biology and Evolution 18: 2057–2066. Tamura, K., D. Peterson, N. Peterson, G. Stecher, M. Nei & S. Kumar, 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolution 28: 2731–2739. Taylor, J. N. & R. R. Miller, 1983. Cichlid fishes (Genus Cichlasoma) of the Rio Panuco Basin, Eastern Mexico, with description of a new species. Museum of Natural History, University of Kansas, Kansas 104: 1–24. Vasconcelos-Fernández, J. M. & J. A. Ramı́rez-Fernández, 2004. Geologı́a y petrologı́a del complejo volcánico de Villa Aldama, Tamaulipas. Ciencia 7: 40–44. Ward, R. D., T. S. Zemlak, B. H. Innes, P. R. Last & P. D. N. Hebert, 2005. DNA barcoding Australia’s fish species. Philosophical Transactions of the Royal Society of London Series B, Biological Sciences 360: 1847–1857. 123

RELATED PAPERS

RELATED TOPICS

Log In

Phylogeographic analysis of genus Herichthys (Perciformes: Cichlidae), with descriptions of Nosferatu new genus and H. tepehua n. sp.

Phylogeographic analysis of genus Herichthys (Perciformes: Cichlidae), with descriptions of Nosferatu new genus and H. tepehua n. sp.

Related Papers

RELATED PAPERS

RELATED TOPICS