We describe a new species of tree frog from the middle Ucayali River, Peru. Scinax pyroinguinis sp. nov. is known from two males found in a patch of Amazon rainforest at 160 masl. Externally, the new species is similar to those in the Scinax rostratus group but it differs from all members by having a rounded head from dorsal view without proboscis or pointed tubercle on the tip of the snout, large conical tubercles on upper eyelids and heels, and bright orange blotches and spots on groins, posterior surface of thighs, and shanks. Genetic analysis supports our morphological study and confirms S. pyroinguinis sp nov as a new species, being tentatively the most basal member of the S. rostratus group. The new species is only known from the type locality that is currently threatened by habitat loss caused by wildfires.

Amphibia, new species, Ucayali River, Scinax pyroinguinis, threatened, wildfires

Currently containing 129 recognized species (Frost 2023), Scinax treefrogs are among the most diverse clades of Neotropical amphibians (Ferrão et al. 2016). Over the last decade, the taxonomy of Scinax has been studied comprehensively, resulting in the grouping of its species into two clades: the Scinax catharinae clade (containing members of the S. catharinae and S. perpusillus species groups), and the S. ruber clade (which gathers members of the S. rostratus and S. ruber species groups) (Faivovich 2002; Faivovich et al. 2005). Regarding the current taxonomy of the Scinax ruber clade, those with a combination of a snout projecting beyond the margin of the jaw, a pointed tubercle on the heels, and a dark triangular cephalic marking with the base lying between eyes are placed into the Scinax rostratus group (Duellman 1972; Faivovich et al. 2005; Hoogmoed and Avila-Pires 2011).

Of the ten species of Scinax known to occur in Peru (AmphibiaWeb 2023), two of them belong to the Scinax rostratus group: S. garbei and S. pedromedinae. Whereas S. garbei has a wide distribution range in the Amazon (Lopes et al. 2020), S. pedromedinae is restricted to the western Amazonian region in central and southern Peru, Brazil and northern Bolivia (Moravec and Aparicio 2004; Hoogmoed and Avila-Pires 2011; Melo-Sampaio and De Souza 2015).

During October and November of 2022, GC and WA conducted a field survey in a remnant forest surrounded by burnt areas and pastures in the Ucayali River Basin, Amazon Lowlands of central Peru (Fig. 1). The expedition resulted in a small collection of amphibians and reptiles. Two specimens of Scinax caught our attention because of their particular external appearance similar to members of the Scinax rostratus group. Following morphological and genetic analyses, we concluded that these specimens of Scinax correspond to an unknown taxon which we described below.

Figure 1.

Map showing type locality of Scinax pyroinguinis sp. nov. (yellow star), Ucayali department, central Peru.

We collected specimens at night while conducting visual transect surveys. We euthanized specimens with benzocaine 20%, preserved them in 10% formalin, and stored them in 70% alcohol. We extracted tissue samples from the tongue of the specimens and stored them in 96% alcohol. We have deposited all specimens in the herpetological collection of the Centro de Ornitologia y Biodiversidad (CORBIDI), Lima, Peru. This research was performed under the permit RDG N° D000010-2021-MIDAGRI-SERFOR-DGGSPFFS issued by Ministerio de Agricultura de Peru.

We follow Duellman (2001) for morphological characters terminology, Lima et al. (2004) for format description, Faivovich et al. (2005) and Duellman et al. (2016) for systematics of Scinax. We used the standard webbing formulae of Savage and Heyer (1967), and Myers and Duellman (1982). For comparisons we consider morphological data for Scinax acuminatus from Cope (1862), for S. boulengeri from Cope (1887) and Duellman (1972); for S. constrictus from Lima et al. (2004); S. garbei from Miranda-Ribeiro (1926), Duellman (1972), Duellman and Wiens (1973), and Duellman (2005); S. jolyi from Lescure and Marty (2000); for S. kennedyi from Pyburn (1973); for S. nebulosus from Spix (1824), Hoogmoed and Gruber (1983), Duellman (1972) and Lescure and Marty (2000); for S. pedromedinai from Henle (1991), Duellman and Wiens (1973), and Duellman (2005); for S. proboscideus from Brongersma (1933), Duellman (1972), and Lescure and Marty (2000); and for S. sugillatus from Duellman and Wiens (1973). Also, the examined specimens are listed in Appendix 1.

For morphological measurements, we follow Duellman (2001) measuring the following variables to the nearest 0.1 mm with digital callipers under a stereomicroscope: snout–vent length (SVL), head length (HL), head width (HW), eye diameter (ED), eye–nostril distance (END), interorbital distance (IOD), internarial distance (IND), nostril snout distance (NSD), tympanum diameter (TD), thigh length (THL), tibia length (TL), and foot length (FL). We numbered fingers and toes preaxially to postaxially from I–IV and I–V respectively. We determined comparative lengths of toes III and V by adpressing both toes against Toe IV; length of fingers I and II were determined by adpressing the fingers against each other. We observed external sexual characteristics (e.g., the presence of vocal sacs in males) to sex specimens. We used photographs of live animals taken by GC in the field for a description of coloration in life.

We confirmed the taxonomic position of the new species within Scinax based on the sequence fragment of the non-coding 16S rRNA mitochondrial gene. To compare our newly generated sequence with congeners, we considered sequences of the 16 species of Scinax most similar to the new species (on the basis of BLAST results across Genbank sequences), as well as the outgroup species Scarthyla ostinodactyla and Sphaenorhynchus lacteus, and the new sequences of S. pyroinguinis sp. nov. from central Peru (sequences for CORBIDI 24669 and 24670 were identical, so only CORBIDI 24669 was used for analyses), and of S. garbei (CORBIDI 17064) from Amazonian Peru, for a total of 20 terminals (Appendix 2). Extraction, amplification, and sequencing of DNA followed standard protocols (Hedges et al. 2008; Catenazzi and Ttito 2016). We used the 16Sar (forward) primer (5’-3’ sequence: CGCCTGTTTATCAAAAACAT) and the 16Sbr (reverse) primer (5’-3’ sequence: CCGGTCTGAACTCAGATCACGT) (Palumbi and Benzie 1991) with the following thermocycling conditions (PCR) on a Proflex PCR system (Applied Biosystems): one cycle of 96 °C/3 min; 35 cycles of 95 °C/30 s, 55 °C/45 s, 72 °C/1.5 min; one cycle 72 °C/7 min. We purified PCR products with Exosap-IT (Affymetrix, Santa Clara, CA) and obtained sequences from MCLAB (San Francisco, CA). We used Geneious, version 11.1.5 (Biomatters, http://www.geneious.com/) to align sequences with the MAFFT v7.017 alignment program (Katoh and Standley 2013), and trimmed sequences to a length of 590 bp without removing any variable fragment. We calculated p-distances using MEGA v. 7 (Kumar et al. 2016). We inferred phylogenetic relationships with a Maximum Likelihood (ML) inference approach on IQ-TREE (Nguyen et al. 2015) using the online server (http://iqtree.cibiv.univie.ac.at/), default options (ultrafast bootstrap method, 1000 bootstrap alignments), and the GTR+G+I model of evolution.

In spite of having only two specimens, genetic and morphological evidence support our description and confirm Scinax pyroinguinis sp. nov. as a new taxon. Our phylogenetic tree (Fig. 2) shows that S. pyroinguinis forms a well-supported (91%) single lineage, sister to a greater clade that contains members of the S. rostratus group plus S. acuminatus. Our analysis of the external morphology of the new species also supports its inclusion in the S. rostratus group. Additionally, the phylogenetic tree (Fig. 2) suggests that S. pyroinguinis might be the most basal member of the group, but further studies increasing the terminals are needed to confirm its position in the S. rostratus group.

In the past, S. acuminatus has been suggested as closely related to the Scinax rostratus group (Faivovich 2002). Here we can confirm with high genetic support that the population of S. acuminatus we used in our analysis (GenBank accession number KJ004189 from Ñeembucu, southwestern Paraguay) does belong to the S. rostratus group. However, since its type locality was not mentioned in the original description and its geographic range includes eastern Bolivia, southern Brazil, northern Argentina and Paraguay (Boulenger 1889; Köhler and Böhme 1996; Magrini et al. 2011; Weiler et al. 2013), we are not able to assign S. acuminatus sensu lato to the S. rostratus group. Thus, increasing the genetic samples for S. acuminatus covering its entire distribution range is necessary to complete this picture.

A potential hidden diversity within Scinax garbei has been previously noticed (Lopes et al. 2020). Our genetic analysis found that samples of S. garbei from lower Madre de Dios river (WED 57696) and lower Kosñipata Valley (CORBIDI 17064), both localities in the Amazon rainforest at southern Peru, form a different lineage, sister to the samples of S. garbei from Ecuadorian Amazon (QCAZ 43755, 46403, 48844; TWQ 008, KU 202764), far northern than the aforementioned Peruvian samples. This finding does confirm what was suggested by Gomes-Lopes et al. (2020) that the populations from Ecuador might be considered taxonomically different from those of southern Peru. Also, Lopes et al. (2020) noticed that one of the available names for populations of S. garbei in southern Peru is S. epacrorhyna, a species described by Duellman (1972) and later synonymized with S. garbei by Duellman and Wiens (1973). Duellman and Wiens (1973) considered the morphological characters of S. epacrorhyna as a geographic variation of S. garbei. Although our results support the hypothesis that populations from southern Peru might be considered a different taxon, an integrative approach (acoustic data, updated morphological revisions and genetic samples from type specimens) is needed to assign a name and clarify the taxonomic status, Ecuadorian and Peruvian populations.

The type locality of S. pyroinguinis is on the western side of the middle basin of the Ucayali river, one of the largest tributaries of the Amazon river (Fig. 1). The middle Ucayali river has been mentioned as a geographic barrier for birds (Harvey et al. 2014), yet the dispersal of amphibians across river sides (Marin da Fonte et al. 2019) could play a role in the distribution of the new species. Even though the presence of S. pyroinguinis on the eastern side of the Ucayali river has not been confirmed, more fieldwork is needed to confirm its occurrence on both sides of the Ucayali river.

Despite having collected only two individuals, our field sampling and comparisons with populations of sympatric species suggest that Scinax pyroinguinis is an uncommon species in the area. Indeed, we have obtained these two individuals after having completed 108 person-days of herpetological survey between 2021–2022. Also, the patch of forest where S. pyroinguinis lives is under pressure from wildfires (Fig. 6) caused by farmers who try to expand their pastures or clean the ground for agriculture. These fires occur mostly in the dry season and are illegal in Peruvian territory. However, environmental policies from the government have been largely ineffective and have not prevented the loss of forest territories in the Amazon (Eufemia et al. 2022). For instance, an update of the levels of deforestation caused by wildfires around the type locality gives a critical scenario where 273 fire alerts have been displayed over the last year (2022–2023, Watch GF 2015) and where the deforested area between 2011–2021 is 3130 km² (Watch GF 2015). Furthermore, as Fig. 1 shows, the new species inhabits a very fragmented landscape where the forest covers about 655 km² of the land. Finally, since cases of microendemism in Scinax frogs are not known, we believe that the presence of S. pyroinguinis on the eastern side of the Ucayali River is probable. Thus, following the IUCN criteria, based on the presence of the new species in a very fragmented habitat that faces increasing deforestation every year, and the absence of data on its distribution, ecology and population status, we suggest the new species be placed into the Near Threatened (NT) category of the IUCN red list.

We thank the two reviewers who helped us improve this manuscript. Our genetic analyses were performed thanks to the support of the Global Genome Initiative Awards Program of the Global Genome Biodiversity Network (GGBN). Our fieldwork would not have been possible without the support of Bosques Amazonicos del Peru (BAM). GC, WA, and LAGA thank CORBIDI and PASKAY S.A.C. for the logistic help in the field. Finally, we want to thank Oscar Castillo who helped us greatly during our fieldwork.

Specimens examined

Scinax garbei. ECUADOR: SUCUMBIOS: Provincia Sucumbios, Cuyabeno, CORBIDI 0008. PERU: CUSCO: Provincia La Convención, Comunidad Nativa Chokoriari, CORBIDI 8040-43, Comunidad Nativa Camisea CORBIDI 11181, 11224; Provincia Paucartambo, Kosñipata, Pillcopata, AMNH 85382, CORBIDI 17064, KU 139242–43, 139245; Rio Entoro, CN Qeros, MHNG 2607.28. HUÁNUCO: Provincia Puerto Inca, Serranía del Sira, CORBIDI 14421. LORETO: Provincia Datem del Marañon, Comunidad Nativa Nuevo Andoas, CORBIDI 1037, 6453, 6455, 6483, 6487; Provincia Maynas, Redondococha, CORBIDI 0056, 0059; Provincia Requena, Parque Nacional Sierra del Divisor, CORBIDI 2135, 2144. MADRE DE DIOS: Provincia Manu, Los Amigos Conservation Concession, MUSM 24281–82, Concesion Forestal Inambari, CORBIDI 13430; Provincia Tambopata, Alrededores de Puerto Maldonado CORBIDI 5169.

Scinax pedromedinae. PERU: LORETO: Provincia Requena, Parque Nacional Sierra del Divisor, CORBIDI 2142. MADRE DE DIOS: Provincia Manu, Los Amigos Conservation Concession, MUSM 24322, 24325; Concesion Forestal Inambari, CORBIDI 13435; Provincia Tambopata, Lago Tres Chimbadas, CORBIDI 5211.

Scinax ruber. ECUDADOR: Sucumbios: Provincia Sucumbios, Cuyabeno CORBIDI 0136. PERU: CUSCO: Provincia La Convención, Comunidad Nativa Kotsiri, CORBIDI 335-36, Comunidad Nativa Alto Shima CORBIDI 9701, 9707, Comunidad Nativa Puyantimari, CORBIDI 9761, Comunidad Nativa Camisea, CORBIDI 10007-09, 11221; Provincia Paucartambo, Kosñipata, Atalaya, KU 154729, 154731–32; Pillcopata, KU 139241; Radiochaylloc, MUSM 21080, MHNG 2607.14. HUÁNUCO: Provincia Leoncio Prado, Parque Nacional Tingo Maria, CORBIDI 15589. JUNÍN: Provincia Chanchamayo, Fundo La Génova, CORBIDI 9877. LORETO: Provincia Datem del Marañon, Comunidad Nativa Nuevo Andoas, CORBIDI 1052, 5041; Provincia Maynas, Aguas Negras CORBIDI 244; Provincia Requena, Parque Nacional Sierra del Divisor, CORBIDI 2125, 2130. MADRE DE DIOS: Provincia Manu, Los Amigos Conservation Concession, MUSM 24298–99, 24330; Concesion Forestal Inambari, CORBIDI 13449, 13468.