New Species of the Genus Hypanus (Dasyatidae) from the Eastern Tropical Pacific Ocean

THE genus Hypanus is composed of ten valid species, one on the west coast of Africa (Hypanus rudis), two in the Eastern Tropical Pacific Ocean (ETP; Hypanus longus and H. dipterurus), and seven in the western Atlantic Ocean (H. americanus, H. berthalutzae, H. geijskesi, H. guttatus, H. marianae, H. sabinus, and H. say).

Species of Hypanus are broadly distributed along the shorelines of their habitats. Hypanus americanus is the most widely distributed species in the Atlantic, ranging from New Jersey (USA) to French Guiana. In contrast, H. geijskesi has the most restricted known range in this marine area, found from the Orinoco delta south to Fortaleza in the north of Brazil. In the ETP, the distribution of H. dipterurus is greater than that of H. longus, going from California to northern Peru (Last et al., 2016). These species exhibit benthic and coastal behaviors, frequently inhabiting sandy bottoms, seagrass beds, lagoons, and reefs, with particularly high abundances in bays and estuaries (Michael, 1993). According to many authors, species of Hypanus may have specific habitat requirements, which means that their distribution ranges can be altered by changes in salinity, sea temperature, currents, and other factors (Pyron and Burbrink, 2010; Last et al., 2016; Petean et al., 2020).

Most of these species face high levels of threat from anthropogenic activities such as unsustainable fisheries, habitat degradation, and pollution (Dulvy et al., 2014). Many of these species are regularly caught directly and as bycatch. This is documented in numerous studies for the ETP (López-Garro and Zanella, 2015; Navia and Mejía-Falla, 2016; Morales-Aguilar and Ortiz-Aldana, 2022) as well as in the Caribbean and Atlantic (Mazzoleni and Schwingel, 1999; Costa and Chaves, 2006; Tagliafico et al., 2013; Schmidt et al., 2015; Barrios-Garrido et al., 2017; Hernández-Fernández et al., 2021).

The above-mentioned anthropogenic pressures have led the International Union for the Conservation of Nature (IUCN) Red List of Threatened Species to assess H. geijskesi (Pollom et al., 2020a) and H. rudis as Critically Endangered (Jabado et al., 2021), H. marianae as Endangered (Pollom et al., 2020b), and H. longus, H. dipterurus, and H. berthalutzae as Vulnerable (Pollom et al., 2020c, 2020d; Charvet et al., 2020). The remaining species were considered Near Threatened (H. americanus, H. guttatus, and H. say) and Least Concern (H. sabinus) but with declining population trends (Carlson et al., 2020a, 2020b, 2020c, 2020d).

In the Pacific coast of Colombia, the confirmed presence of H. longus and H. dipterurus has been previously documented (Mejía-Falla and Navia, 2019). However, since 2003, a species with similar characteristics, but not matching either of these, has been observed. Given that species in Hypanus are used as a food source by local fishing communities, the lack of clear taxonomic definition for these species has led to inaccuracies in estimating catches and assessing the risk levels of the genus in the Colombian Pacific coast. Hence, the primary objective of this study is to describe a new ray species found in the Colombian Pacific coast, supported by both anatomical and molecular evidence. Additionally, this study aims to provide preliminary information on the biology, ecology, and potential distribution of this newly described species.

MATERIALS AND METHODS

A total of 90 individuals (47 females, 40 males, 3 unsexed) were detected between 2003 and 2023, during landing and on-board monitoring in artisanal fisheries in the Colombian Pacific, at specific localities between Pizarro, Chocó (4.97833°N, 77.40217°W) and Mallorquín, Valle del Cauca (3.60837°N, 77.21998°W; Fig. 1).

View Figure

• Download as Powerpoint
• Download Figure

Animals were measured and weighed. The lengths recorded were total length (TL) and disc width (DW) in cm, and total weight in g. Given that this species is primarily used for local consumption or sale, most specimens were unable to be preserved intact upon arrival at port. Instead, they were eviscerated or had their tails removed, thereby precluding their use as reference material. Holotype and additional tissue samples for genetic analysis were collected on a research cruise in the marine-coastal zone of the Uramba Bahía Málaga National Natural Park, Department of Valle del Cauca, within the area of influence of the Juanchaco and Ladrilleros settlements. Additional samples included in this study came from fishing landings off Charambirá, Chocó region, in the northern part of the Colombian Pacific.

Specimens used for description and comparisons were measured following Last et al. (2016) and Manjaji (2004). External morphological descriptions followed Manjaji (2004) and Last and Stevens (2009); clasper descriptions were in accordance with de Moreira et al. (2018). A length–weight relationship was estimated using the power function TW = a × S^b and the logarithmic form logTW = loga + b × logDW, where TW is the total weight of the individual, DW is the disc width, a is the model intercept, and b is the slope. The type of growth (isometric or allometric) was assessed using a student’s t test (H0: b = 3; Steel and Torrie, 1980). Clasper condition was assessed by considering the degree of calcification and classified as 0 (noncalcified), 1 (partially calcified), or 2 (calcified), with 2 corresponding to mature males (Clark and von Schmidt, 1965; Braccini and Chiaramonte, 2002). Measures of the reproductive structures in males (inner clasper) were plotted against DW to evaluate onset of maturity. Vertebrae were extracted and processed following Mejía-Falla et al. (2014) for a preliminary age estimation of the specimen.

Tissue samples were taken from the pectoral fin of the holotype and one additional specimen (caught and released) stored in 96% ethanol. DNA was extracted following a standard salting-out protocol (Sambrook et al., 1989). PCR amplifications of the mitochondrial gene cytochrome oxidase I (COI) using the FishF1 (5′–TTCTCAACCACA AGACATTGG-3′) and FishR1 (5′–TAGACTTCT GGGGCCAATCA-3′) primers (Ward et al., 2005) were conducted. Amplifications were performed in 13 μl reactions containing 0.5 μl of DNA, 0.625 μl of each primer (forward–reverse), and 11.25 μl of Thermo Scientific 1.1× PCR master mix (2.5 mM MgCl₂). After an initial denaturation of 2 min at 95°C, 30–35 cycles at 94°C for 45 seconds (s), followed by 45 s at an annealing temperature of 54°C, and 60 s at 72°C with a final extension of 5 min at 72°C were conducted. Sequencing was performed in one direction on an ABI 3100 automated sequencer (Applied Biosystems, Foster City, CA, USA). COI sequences were available on GenBank for all species of the genus Hypanus (only H. marianae was unavailable). As generic limits within this family are still uncertain, another member of Dasyatidae, Pteroplatytrygon violacea, and a distantly related rhinid, Rhynchobatus djiddensis, were used as outgroups for molecular comparisons and a preliminary phylogenetic reconstruction. GenBank accession numbers are as follows: H. americanus KT075327, KF461168; H. berthalutzae MT326604, MT326605; H. dipterurus GU440302; H. geijskesi MN105749; H. guttatus MN105793, JX034000; H. longus MT326628, MT326629; H. rubioi, new species, PP133089; H. rudis MT326632, MT326634; H. sabinus KF461169, JN025277; H. say OP056996; P. violacea MN105751, MK085554; and R. djiddensis JX263419. All COI sequences were cleaned and trimmed with Geneious 9.1.8 (https://www.geneious.com), and Muscle (Edgar, 2004) was used as the alignment algorithm with the default parameters. COI corrected genetic distances were calculated among all taxa included. Phylogenetic relationships were assessed using Bayesian inference. Two independent runs with a random starting tree using GTR+Gamma model of nucleotide substitution were performed with BEAST 2.75 (Bouckaert et al., 2019). Estimated sample sizes, mixture, and convergence of chains were evaluated in Tracer (Rambaut et al., 2018). Tree annotator was used to combine all Bayesian trees into the maximum clade credibility tree, discarding the initial 25% of the trees as the burn in phase.

To estimate the potential distribution area occupied by the species, we used 33 occurrence data points (Supplemental Material, Table 1; see Data Accessibility) applied to a species distribution model (SDM). Based on the known distribution of other species of the family Dasyatidae; the coastal region of the ETP was hypothesized as a possible area M (historically accessible area for the species; Soberón and Peterson, 2005). For model calibration, a 200 km buffer was used according to Barve et al. (2011) and taking as reference the marine ecoregions determined by Spalding et al. (2007). To characterize the species’ range, 20 benthic environmental variables from the Bio-Oracle 2.0 repository (Assis et al., 2018) were considered: current velocity, bottom light, primary productivity, salinity, and temperature and for each of them: Lt. min, Lt. max, mean, and range (Lt. refers to the average of the minimum and maximum record per year). Once the environmental variables were obtained, collinearity between them was evaluated to reduce overestimation of the models and bias in statistical inferences (Dormann et al., 2012) by means of a Pearson correlation test with the ‘ellipsenm’ package (https://github.com/marlonecobos/ellipsenm) in R. Since values greater than 0.7 were considered high collinearity, the variables used were the mean current velocity, bottom light, primary productivity, and temperature, and maximum Lt. of salinity.

The model was built with the Maxent 3.4.3 software (Phillips et al., 2006) using the R package ‘KUENM’ (Cobos et al., 2019). To determine the appropriate parameterization of the models, the kuenm_cal function was used to test the regularization multipliers between 0.1 and 5 (every 0.5), and the linear ‘l,’ quadratic ‘q,’ and hinge ‘h’ features, with their respective combinations. We used 50% of the data for training and 50% for evaluation, and 10,000 background points were used (Warren and Seifert, 2011; Merow et al., 2013; Cobos et al., 2019). Based on the aforementioned combinations, the evaluation and selection of the best models was performed using the cal_eval function. For this purpose, 1) the statistical significance of the candidate models was evaluated using the average area under the curve (AUC) of the partial receiver operating characteristic (partial ROC), which is based on a traditional ROC (Peterson et al., 2008), with 500 iterations and 50% of the data used for bootstrapping. The AUC varies between 1 and 2, where 1 represents random models and 2 represents better-than-random performance. 2) The models selected in the previous step were evaluated for their performance, selecting only those models whose omission rate was less than 5% (Anderson et al., 2003). 3) The complexity of the previously selected models was assessed, using the Akaike’s Corrected AICc (AICc), and retaining those models with AICc values less than 2 (Warren and Seifert, 2011). Subsequently, using the kuenm_mod function, the final models were created using the logistic output (Elith et al., 2011) and ten replicates of the models were created by Bootstrap to add variability to the prediction (Kass et al., 2018; Cobos et al., 2019). In addition, Maxent’s jacknife function was used to evaluate the importance of each environmental variable (Phillips et al., 2006). The best candidate models were projected to the previously determined area M using the clamping option (Owens et al., 2013). In order to obtain a single distribution map and uncertainty map, the mean and standard deviation of the final models were calculated with the kuenm_modstats function, and a complementary mobility-oriented parity (MOP) analysis was performed, which measures the distance between the two models (MOP) as the distance between the different points of the calibration area and the projected area, thereby finding the areas of strict extrapolation, which are values that may be outside the range sampled during model training, given some individual variable (Owens et al., 2013).

Hypanus rubioi, new species

urn:lsid:zoobank.org:act:1C4BAFAA-5E61-4830-A88B-8C1620712E96

Longnose Pacific Stingray

Spanish name: Raya picuda

Figures 2, 3, 4

Holotype.—

CIRUV 021336, 1,897 mm TL, 608 mm DW, Bahía Málaga, Colombia, 3.96375°N, 77.431583°W, immature male collected with hook and line at 8 m depth, 28 November 2021.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Paratypes.—

CIRUV 024008, 1,101.7 mm TL, 326 mm DW, Bahía Málaga, Colombia, 3.96375°N, 77.431583°W, immature male collected with hook and line at 0.5 m depth, 9 December 2021; CIRUV 024073, 504 mm DW, landed in Pueblo Nuevo market, Buenaventura, Colombia, 3.882887°N, 77.075239°W, immature male, without information on location of capture or fishing gear, December 2008; CIRUV 024074, 1,485 mm TL, 405 DW, landed in Pueblo Nuevo market, Buenaventura, Colombia, 3.882887°N, 77.075239°W, female, without information on location of capture or fishing gear, December 2008; CIRUV 024075, 1745 TL, 480 DW, landed in Pueblo Nuevo market, Buenaventura, Colombia, 3.882887°N, 77.075239°W, female without information on location of capture or fishing gear, March 2005.

Diagnosis.—

A large-sized species of Hypanus (attaining about 1,250 mm DW; Fig. 2) with the following combination of characters: front-side margins of disc concave; snout pointed protruding sharply, rostrum angle 93–105°; disc equal or wider than long, DL 1.0–1.1 times in DW; long slender tail, 2.4 to 3.1 times disk length; tail width 0.6–1.1 times its height, at pelvic fin posterior end; ventral fold low and slender, its length 1.6–2.7 times DW, 13–25 times its depth below pelvic fin posterior end; preoral length 28–32% DW; distance between nostrils 9–11% DW; distance between first gill slits 17–19% DW; distance from anterior cloaca to sting 0.4–0.8 times in precloacal length (Table 1); dorsal surface rough with a row of enlarged denticles, flanked on each side by a short row of irregular mid-scapular blunt thorns; midline denticles extending from behind the head back into the tail to caudal sting.

Table 1.Measurements of five specimens of Hypanus rubioi, new species. All measurements are presented in mm.

View Fullscreen

Hypanus rubioi is distinguished from its two eastern Pacific sympatric species H. longus and H. dipterurus by having a long (29 to 32% in DW), pointed rostrum compared to 17 to 20% in H. longus, and a very long tail (2.4 to 3.1 times its disc length) compared to 2.4 in H. longus and 1.2 to 1.6 in H. dipterurus (Fig. 3). Hypanus rubioi shares with H. longus a small ventral caudal fin fold fitting 0.4 to 0.7 times the length of the disc and 0.55 to 0.62 times, respectively, while H. dipterurus has a larger fold where its length fits 0.33 times the length of the disc. Hypanus rubioi differs from H. dipterurus in the number of times the mouth width fits in the preoral length. While in the new species the mouth fits 3.1 to 3.7 times, in H. dipterurus it fits only 2.5 to 3 times. The size of the eye also allows separating the species. In H. rubioi the eye is smaller (∼1.7 to 3.4% of the disc width, while in H. longus it is 3.1 to 5.6%). The ratio between internostril width and pre-nasal length is smaller in H. longus (1.3–1.8) compared to values of 2.2 to 2.8 times in H. rubioi. Finally, the eye of H. rubioi fits 4 to 9 times in the interorbital distance, while in H. longus it fits 2 to 4 times and in H. dipterurus it fits 2.5 times.

Description.—

Disc spade-shaped; equal or slightly broader than long, 1.0–1.1 times its length in its width; maximum depth behind spiracles; snout pointed and long; rostrum particularly sharp and pronounced anteriorly; apical lobe acute and triangular; snout angle 93–105°; anterior margin of disc concave, especially closer to the snout; pectoral fin apex rounded; posterior margin convex, increasing toward the rear-most part. Pelvic fins subtriangular, its length from 13 to 18% DW, width 0.6 to 1.1 times in length, slightly convex posterior margin. Clasper rather large, robust, and cylindrical, with significant increase in size from 7% to 15% in DW from juvenile (326 mm DW) to immature males (608 mm DW). Tail long and slender, rounded but tapering toward the end, postcloacal tail 3.0 to 3.4 times precloacal length, tail width 0.6–1.1 times height, at pelvic fin posterior end; sting origin 13.1–16.7% of tail length; surface rough with small blunt thorns and denticles limited to compressed dorsal area reaching but not crossing the sting. Snout long and sharp, preoral length 3.1 to 3.8 times mouth width. Eye diameter 1.8 to 3.4 times in spiracle length. Interorbital distance 11.3–15.5% DW. Spiracles oval, 1.7 to 2.3 times longer than tall. Internasal distance 2.22 to 2.77 times in prenasal distance. Lower jaw subtriangular, symphysial teeth visible ventrally. Gill openings S-shaped, distance between first gill slits 17.8–19.2% DW, distance between the first and the fifth gill slit 1.43 to 1.54 times the distance between first gill slits (Table 1). Dorsal surface covered in rough denticles, becoming progressively sparse toward lateral margin of disc where only a few rows of concentrical denticles remain; behind head a central row of enlarged denticles, flanked on each side by a short row of irregular mid-scapular blunt thorns; this row extends into the tail and ends at caudal sting. Length of sting becoming larger in larger specimens 14% DW in paratype and 25% DW in the holotype (Table 1).

Color in life.—

Dorsal surface of disc and tail uniformly brown, with pearl to beige spines and thorns. Ventral area uniformly whitish to pale, with outer margins narrowly dusky, resembling dorsal color. Tail paler near its base but darkening as it moves back to the skin fold (Figs. 2, 3, 4).

Distribution.—

Hypanus rubioi is found on the shallow shelf of the eastern Pacific coast of Colombia. The known range is between Pizarro, department of Chocó (4.97833°N, 77.40217°W) and Mallorquín, department of Valle del Cauca (3.60837°N, 77.21998°W; Fig. 1), although we suspect this species occurs over a larger range in the ETP. The specimens that were monitored at landings came mainly from the localities of Charambirá (Chocó) and Buenaventura (Valle del Cauca). Fishermen report capturing them exclusively in soft bottom habitats.

Habitat.—

Collection depths range from 0.5 to at least 15 m in shallow waters in soft bottoms or mud habitats.

Etymology.—

The name rubioi comes from the professor Efrain Rubio, who worked for many decades as ichthyology teacher at Universidad del Valle. Dr. Rubio was one of the first researchers interested in the fish fauna of the Colombian Pacific. During all these years at Universidad del Valle, he started the ichthyology collection (CIRUV) in addition to training many generations of young ichthyologists interested in the Pacific fishes of Colombia.

Bioecology.—

Using the 47 females and 40 males recorded in landings and onboard monitoring in artisanal fisheries in the Colombian Pacific, it was found that females presented disc widths between 450 and 1,380 mm (being more frequent between 800 and 900 mm DW; χ = 818 mm DW), while males presented disc widths between 280 and 970 mm DW with greater representation between 600 and 800 mm DW (χ = 692 mm DW; Fig. 5A). Significant differences in length between sexes were identified (U = 6565; Z = 2.140; P = 0.032). For both males and females, high correspondence was observed between disc length and disc width (females y = –3.8568 + 0.9894*x, r² = 0.955; males y = 4.0793 + 0.8558*x, r² = 0.8180; Fig. 5B). Males considered immature, with uncalcified or partially calcified claspers, presented sizes between 280 and 660 mm DW, while mature males with calcified claspers presented sizes between 660 and 970 mm (Fig. 5C). A 905 mm female, captured in March 2019, was in a pregnant state, with the presence of two pups just beginning their development (not measurable). Vertebral band counts in a 1,040 mm DW specimen suggest an age of at least 7 years (Fig. 5D).

View Figure

• Download as Powerpoint
• Download Figure

Distribution probability.—

Once the presence database was processed, we had a total of 33 georeferenced records of Hypanus rubioi located in the central Pacific area of Colombia. A total of 70 candidate models were constructed, all of them with average AUC values greater than 1.9. With respect to the omission rate, 11 models presented values lower than 5%, and of these, only three presented AICc values lower than 2. The three final models were the linear model ‘l,’ the quadratic model ‘q,’ and the linear-quadratic model ‘lq’ with regularization multipliers 0.1, 0.1, and 0.6, respectively. According to the jacknife analysis, the variable with the highest percentage contribution for the three final models was the average value of background light (between 71.1% and 76.0%), followed by the maximum Lt. of salinity (between 9.7% and 10.7%). The variable with the lowest contribution value for the three models was the average current velocity (between 0.5% and 1.0%). The habitat suitability map (Fig. 6) shows that the highest values (estimated area occupied by the species) are closely associated with coastal sections of the ETP (considered as area M), especially some small areas located on the coasts of central México (Fig. 6A), southern México, Guatemala, and El Salvador, and northern Costa Rica (Fig. 6B), and in larger areas along the coasts of Honduras, Nicaragua, and Panamá (Fig. 6C). In addition, habitats with high suitability are found in the central and southern Colombian Pacific (Fig. 6D) and along the entire coast of Ecuador and northern Perú (Fig. 6E). However, it should be noted that the standard deviation map and the MOP analysis show large areas of uncertainty, represented by high values of standard deviation and areas with risk of strict extrapolation, i.e., similarities of 0 or close to 0, so the probability of presence of the species in these areas should be taken with caution (Supplemental Material, Fig. 1; see Data Accessibility).

View Figure

• Download as Powerpoint
• Download Figure

Threats.—

The species is strongly associated with artisanal fisheries in the southern Colombian Pacific, where all the 90 individuals were recorded in fishing landings. The species is sought after because it provides a good amount of protein per specimen—an adult specimen can weigh more than 25 kg (Fig. 7)—and the liver is also used for medicinal purposes, as is done with many other elasmobranch species (Mejía-Falla et al., 2017; Puentes et al., 2022). The level of threat from fishing is different between the south-central Colombian Pacific zone (mouth of the San Juan River to Tumaco) where rays are frequently consumed, and the northern part of Colombia, where consumption is not as intense. In Charambirá, Chocó, in September 2022, 23 specimens of this species were caught in the same fishing gear. All of these came out alive, and 22 were returned live to the sea (J. Caicedo, pers. comm.). The retained specimen was used for family consumption.

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

The description of a third stingray species in the ETP is of great interest, and its marked morphological differences have surprisingly gone unnoticed. The main differences of this new species include the shape of its disc, which features a pointed, prominent, and concave snout, as well as rounded posterior disc margins, in contrast to the typical rhomboidal discs (Fig. 3). Hypanus rubioi is the second largest species of Hypanus, reaching a disc width of 125 cm compared to 180 cm for H. longus in the ETP (Table 2; Robertson and Allen, 2024). However, the average DW of H. rubioi in small-scale fisheries (χ = 78.4 cm DW) is higher than available data for H. dipterurus and H. longus in Mexico (n = 636, χ = 48.0 cm DW; n = 54, χ = 63.5 cm DW; González-González et al., 2020) and for H. longus in Colombia (n = 289, χ = 70.0 cm DW; López-García et al., 2012).

Table 2.Measurements of the five specimens of Hypanus longus caught on the Colombian Pacific coast. All measurements are presented in mm.

View Fullscreen

The present study describes this new species, using molecular, morphological, and ecological information for a group of animals that given their large size are difficult to keep in museum fish collections. In Colombia, H. rubioi occurs together with both H. dipterurus and H. longus, but it can be distinguished by its more prominent snout (29–30% DW), an acutely angular (93–105°) tip, and distinctly convex anterior margins. It also features a unique combination of characters: disc width 1.0 to 1.1 times its length, a broad interorbital space 3.2 to 8.9 times orbit length, and a dorsum with three rows of very prominent denticles, with only the central row extending from the back of the head to the sting on the tail. The two lateral rows are short, blunt, and confined to the middle scapula. Its long and slender tail measures 2.2–2.8 times the DW. This new species has probably been misidentified previously with H. longus in the past due to similarities in tail size. Hypanus rubioi is the third species of the genus Hypanus described this century and the first in the ETP. The last species of this taxon described in the region was H. dipterurus (Jordan and Gilbert, 1880), nearly a century and a half ago (143 years).

Uncorrected genetic distance based on the 578 bp alignment of the mitochondrial COI between H. rubioi and its congeners ranges from 2.27–2.94% (H. guttatus) to 14.18% (H. sabinus). The greatest intraspecific variation found was that of H. guttatus (0.759%), which is even greater than the distance between valid recognized species H. berthalutzae–H. rudis, 0.519–0.692%. The distance of the new species with the Sharpsnout Stingray H. geijskesi, recently included in the genus, is around 4.5%. Other elasmobranch species, as those of the genus Rhizoprionodon are separated by COI genetic distances of 1.2% to 1.5% (Mendonça et al., 2011), and species from the bonnethead complex S. tiburo–S. vespertina ranged from 1.3 to 1.8% (Aroca et al., 2022). Our molecular species separation values are consistent with those found in other stingray groups (Ward et al., 2008). The COI mitochondrial phylogenetic reconstruction (Fig. 8), which includes all valid species of the genus except the Brazilian Large-eyed Stingray H. marianae, resulted in a topology with H. rubioi nested deep within the genus Hypanus and recovered more closely related (posterior probability of 1) to the western Atlantic species H. guttatus which it resembles morphologically (Fig. 8). The Sharpsnout Stingray, H. geijskesi, was also recovered at the base of this three-species clade of snout-nosed rays. The other two ETP species are allocated in different clades. Hypanus longus was recovered sister to the lineage including H. rudis + H. berthalutzae and H. americanus. While H. dipterurus is sister to H. say, Hypanus sabinus was the basal branch of the genus. Pteroplatytrygon violacea was nested within the genus Hypanus, challenging its monophyly, yet its placement needs to be tested with more data. These results, although robust (all species and most of the clades exhibited 1.0 posterior probability support; Fig. 8), should be treated as preliminary as they are derived exclusively from a single mitochondrial marker.

View Figure

• Download as Powerpoint
• Download Figure

Hypanus rubioi is commonly captured by artisanal fishermen along the Colombian Pacific coast. Despite Decree 281 of 2021 that prohibits the fishing and commercialization of sharks and rays in Colombia, this and other shark and ray species continue to be caught as bycatch in most of the country’s fisheries. As already discussed by Castellanos et al. (2021) and Puentes et al. (2022), the threat status of these species is now even more uncertain since no records of catches are being kept. Fishing communities no longer facilitate the recording of catches of these species, yet they continue to use them. Additionally, it has been documented that since the issuance of the prohibition decree, a black market for shark fins has emerged in the country (Colombian general prosecutor’s office, pers. comm.).

Photographic records suggest the presence of H. rubioi in Ecuador, although the identity of these specimens must be confirmed (https://www.inaturalist.org/observations/163346941, https://www.inaturalist.org/observations/148447824, https://www.inaturalist.org/observations/108184540, https://www.inaturalist.org/observations/85374335).

These records would indicate this species is also being caught in artisanal fisheries in the area of the Gulf of Guayaquil, mainly with long-lines. In this regard, Aguilar (2010) recorded 138 specimens (between 2006 to 2007) and identified them as Dasyatis acutirostra (now provisionally under Telatrygon [Last et al., 2016]), but the image presented in this document looks much more like H. rubioi because the rostrum is shorter, 28–32% DW compared to the 37–44% of the Sharpnose Ray, T. acutirostra. Furthermore, the natural distribution of T. acutirostra (Nishida and Nakaya, 1988) is the northwestern Pacific: China and southern Japan (Nishida and Nakaya, 1990; Fricke et al., 2024), suggesting that the identification of these specimens by Aguilar (2010) can be misleading. However, Aguilar (2010) indicated that the specimens were frequent in artisanal fisheries landings and had always been important in this activity. This suggests that this species could be facing strong fishing pressure throughout its range, but due to lack of knowledge this has not been evaluated.

The morphometric data used herein to describe H. rubioi (five specimens) is limited by sample size given the actual restrictions imposed by the Colombian Decree 281 of 2021. Thus, our data do not fully cover the variability of the species, yet some characters appear to be allometric. Nevertheless, morphology along with molecular (COI) information differed greatly from others in Hypanus, including both ETP sympatric species, and are categorical in supporting the new species.

DATA ACCESSIBILITY

Supplemental material is available at https://www.ichthyologyandherpetology.org/i2024010. Unless an alternative copyright or statement noting that a figure is reprinted from a previous source is noted in a figure caption, the published images and illustrations in this article are licensed by the American Society of Ichthyologists and Herpetologists for use if the use includes a citation to the original source (American Society of Ichthyologists and Herpetologists, the DOI of the Ichthyology & Herpetology article, and any individual image credits listed in the figure aption) in accordance with the Creative Commons Attribution CC BY License. ZooBank publication urn:lsid:zoobank.org:pub:DF0C2E6C-F841-426D-ABE5-1C3A6B6DC9B1.

AI STATEMENT

The authors declare that no AI-assisted technologies were used in the design and generation of this article.