Hyalinobatrachium mondolfii, Senaris and Ayarzaguena, 2001

Chávez, Germán, Pradel, Renzo & Catenazzi, Alessandro, 2019, Integrative taxonomy reveals first country record of Hyalinobatrachium mondolfii Señaris and Ayarzagüena 2001, and distribution range extensions for Cochranella nola Harvey 1996, and Rulyrana spiculata Duellman 1976 (Anura: Centrolenidae) in Peru, Zootaxa 4691 (5), pp. 541-560: 543-548

publication ID

https://doi.org/10.11646/zootaxa.4691.5.7

publication LSID

lsid:zoobank.org:pub:42A57F4C-7F4D-45BB-B9C1-DA1E0A481907

DOI

http://doi.org/10.5281/zenodo.4327819

persistent identifier

http://treatment.plazi.org/id/FF3087E1-FFF8-FFA0-FF6C-F88E5559F849

treatment provided by

Felipe

scientific name

Hyalinobatrachium mondolfii
status

 

Hyalinobatrachium mondolfii   .

On 20 January 2017, we collected three male specimens of Hyalinobatrachium   ( CORBIDI 18211, 18221–22) at Las Piedras Amazon Center ( LPAC) (12°4’12”S, 69°29’37.2”W; 207 m a.s.l), Tambopata Province , Madre de Dios department, Peru ( Fig. 1 View FIGURE 1 ). We recorded the advertisement call of one of the males ( CORBIDI 18211, n = 15 calls, T air = 24 °C) ( Fig. 4 View FIGURE 4 ). The external appearence and coloration of specimens of Hyalinobatrachium   collected at LPAC, Madre de Dios ( Table 2, Fig. 3 View FIGURE 3 ) matched the descriptions provided by previous authors ( Señaris and Ayarzagüena 2001; Castroviejo-Fisher et al. 2011a,b) for Hyalinobatrachium mondolfii   . However , we also observed some differences. Our specimens have truncate snout from dorsal and lateral view (vs. rounded in other populations, Castroviejo-Fisher et al. 2011b) and approximately 1/3 to 1/2 of the hepatic peritoneum unpigmented ( Fig. 3B,D View FIGURE 3 ; vs. completely white peritoneum). Additionally, we report small inter population differences in morphometric variables of H. mondolfii   ( Table 1 View TABLE 1 ) GoogleMaps   .

Despite the considerable variation in meristic and coloration traits reported above, our specimens share a number of diagnostic traits with Hyalinobatrachium mondolfii   that are sufficient to differentiate them from specimens of similar species such as H. kawense Castroviejo-Fisher, Vilà, Ayarzagüena, Blanc, and Ernst 2011   (which has white fingers and toes, vs. yellow in H. mondolfii   ) and H. yaku Guayasamin, Cisneros-Heredia, Maynard, Lynch, Culebras, and Hamilton 2017   (which has dark green spots on dorsum and a completely exposed heart, vs. no dark green spots on anterior half of dorsum, and a heart covered by white pericardium in H. mondolfii   ). However, the morphologically similar H. carlesvilai Castroviejo-Fisher, Padial, Chaparro, Aguayo and de la Riva 2009   , which inhabits the Andean foothills of Bolivia and Peru, and H. munozorum (Lynch and Duellman, 1973)   from the Amazonian lowlands of Ecuador, broadly overlap with the character variation of H. mondolfii   and our specimens.

Genetic and bioacoustic data support our identification of specimens from Madre de Dios as Hyalinobatrachium mondolfii   ( Table 3, Fig. 4 View FIGURE 4 , Appendix 2). The 16S sequence of CORBIDI 18221 collected at LPAC is identical (and CORBIDI 18222 is nearly identical) to the 16S sequence (Genbank accession code JF266569 View Materials ) of voucher CBF 6453 identified as H. mondolfii   by Castroviejo-Fisher et al. (2011a). Other specimens identified as H. mondolfii   in Gen- Bank (including those from Delta Amacuro, Venezuela, where the type locality is located) also exhibit low genetic distances with our specimens (≤ 1.75 %). Meanwhile , the genetic distances with type specimens of H. carlesvilai   is ≥ 3.8 %. The genetic distances of our specimens with specimens of H. munozorum   are modest (2.1–3.9 %) but still larger than among specimens of H. mondolfii   . Furthermore , the two species are reciprocally monophyletic (Castro- viejo-Fisher et al. 2011a). Our analysis of the advertisement call of male CORBIDI 18211 at LPAC confirms the match with the known variation of the call of H. mondolfii   . The advertisement call of CORBIDI 18211 was pulsed, composed of a single note lacking frequency modulation, with highest amplitude at the beginning of the call ( Fig. 4 View FIGURE 4 ). The dominant frequency of all calls was 4593 Hz , with first harmonic at 9186 Hz ; call duration averaged 345 ms ± 15 ms (309–512 ms), and intervals between calls 4.7 s ± 1.5 s (0.5– 13.2 s). Although the dominant frequency is slightly lower than reported previously ( Señaris and Ayarzagüena 2001, Castroviejo-Fisher et al. 2011a,b; Venâncio et al. 2014), the spectrogram is very similar to calls recorded in Acre, Brazil, Pando, Bolivia, and at the type locality in Delta Amacuro, Venezuela. On the other hand, the call is clearly different from those of H. kawense   and H. carlesvilai   because these species have a dual structured call (pulsed+tonal) instead of a single note as in H. mondolfii   . Additionally the call in H. kawense   and H. carlesvilai   reaches the highest amplitude at the middle and at the second third respectively, while the amplitude is highest at the beginning of the call in H. mondolfii   . Unfortunately there are no call records from the type locality of H. munozorum   , and calls from a Bolivian population of H. munozorum   (consisting of five tonal notes) could not be assigned with certainty to collected vouchers ( Castroviejo-Fisher et al. 2011a). Considering the new variation in head shape and hepatic peritoneum reported herein, as well as the results of our genetic and acoustic data, we suggest that future records related with H. carlesvilai   , H. kawense   , H. mondolfii   , and H. munozorum   should be supported by both genetic and acoustic data. For example, H. carlesvilai   has been reported from Mato Grosso state, Brazil ( Cisneros-Heredia et al. 2010) on the basis of morphological comparisons. Description of head shape and eyes coloration of H. carlesvilai   from Mato Grosso are similar to those we observed for H. mondolfii   in Madre de Dios (i.e., having a truncate snout from lateral view). Although Cisneros-Heredia et al. (2010) did not include ventral images of specimens of H. carlesvilai   from Mato Grosso, they reported that iridophores are absent on the ventral parietal peritoneum but covering all parts of the visceral peritonea. On the basis of the new variation in snout shape reported in our specimens, the relative proximity with records of H. mondolfii   ( Venâncio et a l. 2014, this work), and that all other records of H. carlesvilai   are restricted to the hills of the Andes , while H. mondolfii   is known from the Amazonian lowlands, we suggest that the presence of H. carlesvilai   in Brazil should be re-evaluated through acoustic and molecular analysis.

euknemos euknemos euknemos granulosa euknemos

Cochranella nola   (MN 509214) Cochranella   (EU 663008) Cochranella mache   (EU 663013) Cochranella nola   (EU 663015) Cochranella nola   (EU 663016) Cochranella   sp. (EU 663021) Cochranella   (FJ 784377) Cochranella   (FJ 784396) Cochranella   (FJ 784455) Cochranella   (FJ 784458) Espadarana   prosoble- pon (JX 564857) Centrolene durrellorum   (KF 534356) Cochranella nola   (MN509214) Cochranella euknemos   (EU663008) 4.25 Cochranella mache   (EU663013) 4.30 2.77 Cochranella nola   (EU663015) 1.45 4.81 4.90 Cochranella nola   (EU663016) 1.45 5.18 4.89 0.36 Cochranella   sp. (EU663021) 1.82 5.55 5.26 2.54 2.54 Cochranella euknemos   (FJ784377) 3.90 0.19 2.53 4.94 4.94 5.31 Cochranella euknemos   (FJ784396) 4.61 2.24 2.89 5.31 5.30 6.04 1.95 Cochranella granulosa   (FJ784455) 4.79 3.73 3.80 4.76 5.12 5.31 3.90 4.43 Cochranella euknemos   (FJ784458) 4.08 0.00 2.71 4.76 5.12 5.50 0.18 2.13 3.72 Espadarana prosoblepon   (JX564857) 4.82 6.64 5.72 5.25 5.24 5.44 6.19 5.66 6.02 6.37 Centrolene durrellorum   (KF534356) 4.77 5.92 5.32 5.14 5.14 6.06 5.74 5.19 5.93 5.93 3.66 Cochranella erminea   (KF534360) 4.69 4.05 3.42 4.88 4.87 5.61 3.82 4.36 4.00 4.00 5.41 5.67 Cochranella   sp. (KF534362) 4.86 4.42 3.93 4.88 4.87 5.97 4.06 4.59 4.06 4.24 6.07 6.58 Cochranella guayasamini   (KM068259) 4.70 4.06 3.58 4.89 4.88 5.98 3.89 4.25 3.72 4.07 6.25 6.04 Cochranella guayasamini   (KM068265) 4.77 4.10 3.67 4.95 4.95 6.06 3.85 4.22 3.67 4.04 6.41 6.10 Espadarana andina   (KP149354) 4.91 6.90 6.15 5.31 5.30 5.68 6.38 6.38 5.67 6.56 3.33 4.44 Cochranella euknemos   (KR863135) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.36 4.43 1.77 5.97 5.56 Cochranella euknemos   (KR863136) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.36 4.43 1.77 5.97 5.56 Cochranella euknemos   (KR863137) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.36 4.43 1.77 5.97 5.56 Cochranella euknemos   (KR863138) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.71 4.43 1.77 5.97 5.56 Cochranella euknemos   (KR863139) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.36 4.43 1.77 5.97 5.56 Cochranella euknemos   (KR863140) 4.57 2.24 2.89 5.31 5.30 6.04 1.95 0.00 4.43 2.13 5.61 5.19 Cochranella euknemos   (KR863141) 4.57 2.24 2.89 5.31 5.30 6.04 1.95 0.00 4.43 2.13 5.61 5.19

Our specimens were found in primary tropical floodplain forest ( Fig. 5 View FIGURE 5 ). We heard calling males during only one night out of our 10-day survey (survey effort 280 hours-person). Calling males were perched on the underside of leaves at night between 22:00–02:00 hrs at 25–26°C. We observed several of these males under leaves of a fig tree ( Ficus   sp.), at a height of 3 m above ground, at the side of a temporary pond. Two of the collected calling males were nearby egg masses, each of which contained 23– 24 eggs ( Fig. 5 View FIGURE 5 ). Hylinobatrachium mondolfii   was recorded in sympatry with the treefrogs ( Hylidae   ) Boana   aff. alfaroi and Boana punctata   . We did not observe other centrolenid frogs at this site, but we found males of Teratohyla midas   at a nearby floodplain site (~ 1 km away).

The site where we observed and recorded Hylinobatrachium mondolfii   was unexpected for a centrolenid because the temporary pond had a foul smell indicative of stagnant water with decomposing matter and low dissolved oxygen, in contrast with lotic and highly oxygenated habitats typical for glassfrogs. As previously noted ( Castroviejo-Fisher et al. 2011b), the pink coloration of tadpoles could indicate an adaptation to life in such lentic habitats ( Figure 5 View FIGURE 5 ), because the coloration is probably due to the high concentrations of hemoglobin in blood, which should facilitate oxygen uptake and transport.

Our record from the Las Piedras River in the Madre de Dios basin is the first report of Hyalinobatrachium mondolfii   for Peru ( Fig. 1 View FIGURE 1 ), and extends the species’ known distribution range ~ 90 km by airline SW of the closest locality in Pando, Bolivia ( Castroviejo-Fisher et al. 2011a). Our new site is ~ 2,400 km by airline from the type locality, the farthest and westernmost of any other known locality ( Fig. 1 View FIGURE 1 ). H. mondolfii   had previously been reported from Pará ( Brazil), Guyana, southern Surinam, Caño Acoima and Caño Jotajana (Delta Amacuro, Venezuela) and along the Río Guarapiche (Monagas, Venezuela), and the Amazonian regions of Leticia ( Colombia), and Acre ( Brazil). Thus , the presence of this species in Amazonian Peru was very likely, and our report fills this gap in the known distribution range of the species. Future surveys might contribute to fill other gaps in the seemingly disjoint distribution range of H. mondolfii   throughout the Orinoco and Amazon basins   .

According to the IUCN Red List , the distribution of Hyalinobatrachium mondolfii   (classified as Least Concern) is restricted to the Orinoco river Delta. As previously reported, and as confirmed by our work, this geographic range is incorrect and should be extended to include known localities in the Guyana shield, and the Amazon rainforest in Colombia, Brazil, Bolivia, and Peru ( Avila-Pires et al. 2010, Castroviejo-Fisher et al. 2011a, 2011b; Oliveira et al 2015, this paper). Although our morphological observations, genetic data and advertisement calls of Peruvian specimens show close similarity with the type series and specimens identified as H. mondolfii   , we suggest future research should investigate genetic diversity across the wide distribution range of the species. For the time being, and for purposes such as assessing the conservation status for the IUCN Red List, we recommend all known populations in both the Amazon and Orinoco basins be considered as H. mondolfii   sensu lato   .

TABLE 1. Uncorrected genetic p-distances estimated from the 16S rRNA mitochondrial fragment between Cochranella nola from Chontachaca, Peru (in boldface) and a subset of related taxa of the genus Cochranella (see Appendix 1 for full list of specimens considered).

Cochranella nola   (MN 509214) Cochranella   (EU 663008) Cochranella mache   (EU 663013) Cochranella nola   (EU 663015) Cochranella nola   (EU 663016) Cochranella   sp. (EU 663021) Cochranella   (FJ 784377) Cochranella   (FJ 784396) Cochranella   (FJ 784455) Cochranella   (FJ 784458) Espadarana   prosoble- pon (JX 564857) Centrolene durrellorum   (KF 534356)
Cochranella nola   (MN509214)
Cochranella euknemos   (EU663008) 4.25                      
Cochranella mache   (EU663013) 4.30 2.77                    
Cochranella nola   (EU663015) 1.45 4.81 4.90                  
Cochranella nola   (EU663016) 1.45 5.18 4.89 0.36                
Cochranella   sp. (EU663021) 1.82 5.55 5.26 2.54 2.54              
Cochranella euknemos   (FJ784377) 3.90 0.19 2.53 4.94 4.94 5.31            
Cochranella euknemos   (FJ784396) 4.61 2.24 2.89 5.31 5.30 6.04 1.95          
Cochranella granulosa   (FJ784455) 4.79 3.73 3.80 4.76 5.12 5.31 3.90 4.43        
Cochranella euknemos   (FJ784458) 4.08 0.00 2.71 4.76 5.12 5.50 0.18 2.13 3.72      
Espadarana prosoblepon   (JX564857) 4.82 6.64 5.72 5.25 5.24 5.44 6.19 5.66 6.02 6.37    
Centrolene durrellorum   (KF534356) 4.77 5.92 5.32 5.14 5.14 6.06 5.74 5.19 5.93 5.93 3.66  
Cochranella erminea   (KF534360) 4.69 4.05 3.42 4.88 4.87 5.61 3.82 4.36 4.00 4.00 5.41 5.67
Cochranella   sp. (KF534362) 4.86 4.42 3.93 4.88 4.87 5.97 4.06 4.59 4.06 4.24 6.07 6.58
Cochranella guayasamini   (KM068259) 4.70 4.06 3.58 4.89 4.88 5.98 3.89 4.25 3.72 4.07 6.25 6.04
Cochranella guayasamini   (KM068265) 4.77 4.10 3.67 4.95 4.95 6.06 3.85 4.22 3.67 4.04 6.41 6.10
Espadarana andina   (KP149354) 4.91 6.90 6.15 5.31 5.30 5.68 6.38 6.38 5.67 6.56 3.33 4.44
Cochranella euknemos   (KR863135) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.36 4.43 1.77 5.97 5.56
Cochranella euknemos   (KR863136) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.36 4.43 1.77 5.97 5.56
Cochranella euknemos   (KR863137) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.36 4.43 1.77 5.97 5.56
Cochranella euknemos   (KR863138) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.71 4.43 1.77 5.97 5.56
Cochranella euknemos   (KR863139) 4.57 1.87 2.89 5.31 5.30 6.04 1.60 0.36 4.43 1.77 5.97 5.56
Cochranella euknemos   (KR863140) 4.57 2.24 2.89 5.31 5.30 6.04 1.95 0.00 4.43 2.13 5.61 5.19
Cochranella euknemos   (KR863141) 4.57 2.24 2.89 5.31 5.30 6.04 1.95 0.00 4.43 2.13 5.61 5.19

......continued on the next page

CORBIDI

Centro de Ornitologia y Biodiversidad

T

Tavera, Department of Geology and Geophysics

CBF

Coleccion Boliviana de Fauna