Uncovering the diversity in the Amazophrynella minuta complex: integrative taxonomy reveals a new species of Amazophrynella (Anura, Bufonidae) from southern Peru

Abstract A new species of the genus Amazophrynella (Anura, Bufonidae) is described from the departments of Madre de Dios, Cusco and Junin in Peru. An integrative taxonomy approach is used. A morphological diagnosis, morphometrics comparisons, description of the advertisement call, and the phylogenetic relationships of the new species are provided. Amazophrynella javierbustamantei sp. n. differs from other species of Amazophrynella by: intermediate body-size (snout-vent length 14.9 mm in males, n = 26 and 19.6 mm in females, n = 20), tuberculate skin texture of body, greatest hand length of the Amazophrynella spp. (3.6 mm in males, n = 26 and 4.6 mm in females, n = 20), venter coloration yellowish, tiny rounded black points covering the venter, and thirteen molecular autapomorphies in the 16S RNA gene. Its distribution varies from 215 to 708 m a.s.l. This discovery highlights the importance of the remnant forest in preserving the biodiversity in Peru, and increase in seven the species formally described in the genus Amazophrynella.


Introduction
Until 2012, two species of Amazophrynella were placed in the genus Dendrophryniscus Jimenez de la Espada, 1868. Fouquet et al. (2012a) recognized that species of Dendrophryniscus from the Amazon and Atlantic rainforests represented morphologically and genetically deeply divergent lineages, and thus the authors proposed a new genus: Amazophrynella Fouquet, Recorder, Texeira, Cassimiro, Amaro, Camacho, Damaceno, Carnaval, Moritz & Rodrigues, 2012 for the Amazonian species A. minuta and A. bokermanni.
Using a phylogenetic analysis based on mitochondrial and nuclear genes (Fouquet et al. , 2012a, the existence of three independent evolutionary lineages was discovered within the nominal species A. minuta from Ecuador and French Guianas; subsequently, other independent evolutionary lineages were discovered from Brazil and Peru (Rojas et al. , 2015. The difficulties in delimiting species within the A. minuta species complex resides in the relatively generalized diagnosis (see Melin 1941) and the poor geographic sampling. For these reasons, historically, the name A. minuta has been used for individuals distributed throughout the Amazonian biome (e.g. Duellman 1978;Zimmerman and Rodrigues 1990;Magnusson and Hero 1991;Rodrigues and Duellman 1993;Duellman and Mendelson 1995;Fouquet et al. 2012a). Thus, taxonomy and systematics of populations that are currently part of the A. minuta complex remains largely unresolved , in turn limiting the knowledge of the true taxonomic diversity of the genus (Ávila et al. 2012;Rojas et al. 2014Rojas et al. , 2015.
Given this scenario, herein is described an additional new species of Amazophrynella from the departments of Madre de Dios, Cusco and Junin, Peru, founded on the principles of integrative taxonomy. Morphological, morphometric, bioacoustic and phylogenetic relationships are provided as evidence for the existence of the new taxon.

Morphology
Forty eight specimens previously identified as Amazophrynella minuta (Melin, 1941), deposited at the Museo de Historia Natural del Cusco, Universidad Nacional de San Antonio Abad del Cusco (MHNC) and Museo de Historia Natural de la Universidad Nacional Mayor de San Marcos (MHNSM) were analyzed. This material was compared with twenty preserved specimens of A. minuta from the type locality (Taracuá mission, on the right bank of the Uaupés River, municipality of São Gabriel da Cachoeira, Brazil), deposited in the Collection of Amphibians and Reptiles of the Instituto Nacional de Pesquisas da Amazônia-INPA, Brazil (INPA-H). Further comparisons were made with three syntypes deposited at the Naturhistoriska Museet, Göteborg, Sweden (NHMG), and the original description of the species (Melin 1941).
Additionally five preserved specimens of Amazophrynella bokermanni (Izecksohn, 1993) from near the type locality (Juruti, 30 Km from type locality), the holotype and paratypes of A. manaos deposited in the Collection of Amphibians and Reptiles of the Instituto Nacional de Pesquisas da Amazônia-INPA, Manaus, Amazonas, Brazil (INPA-H), the holotype of Amazophrynella vote, deposited in the Coleção Zoológica de Ver-tebrados of the Universidade Federal de Mato Grosso-UFMT, Cuiabá, Mato Grosso, Brazil (UFMT-A), seventeen paratypes deposited in the Collection of Amphibians and Reptiles of the Instituto Nacional de Pesquisas da Amazônia-INPA, Manaus, Amazonas, Brazil (INPA-H), and the holotype and paratypes of A. amazonicola and A. matses, deposited at the Museo de Zoologia de la Universidad Nacional de la Amazonia Peruana (MZUNAP) were analyzed (see Appendix 1 listing all the revised specimens).
Morphological character analyses were carried out according to Cruz and Fussinato (2008) and Fouquet et al. (2012a). Sex was determined by gonad analysis.
Measurements were carried out with a digital caliper following Kok and Kalamandeen (2008) and Duellman (1978). SVL (snout-vent length) from the tip of the snout to the posterior edge of the cloaca; HL (head length) from the posterior edge of the jaw to the tip of the snout; HW (head width), the greatest width of the head, usually at the level of the posterior edges of the tympanum; ED (eye diameter); IND (internarinal distance), the distance between the edges of the nares; SL (snout length) from the anterior edge of the eye to the tip of the snout; HAL (hand length) from the proximal edge of the palmar tubercle to the tip of Finger III; UAL (upper arm length) from the edge of the body insertion to the tip of the elbow; THL (thigh length) from the vent to the posterior edge of the knee; TL (tibia length) from the outer edge of the knee to the tip of the heel; TAL (tarsal length) from the heel to the proximal edge of the inner metatarsal tubercle; FL (foot length) from the proximal edge of the inner metatarsal tubercle to the tip of Toe IV. Diagnosis of characters follow Chaparro et al. 2015.
Statistical analysis. We used a total of 80 adult males of the Amazophrynella minuta species complex (numbers of individuals and populations of origin in parentheses): A. minuta sensu stricto (n = 23, from Taracuá), A. amazonicola (n = 15, from Puerto Almendras and Fazenda Zamora); A. matses (n = 13, from Nuevo Salvador) and the new species of Amazophrynella (n = 29, from Tambopata, Nuevo Arequipa, Candamo and Inambari).
All morphometric measures were log10 transformed to conform to requirements of normality (Hayek et al. 2001). The effect of size was removed from all variables by regressing them against SVL and using the residuals of each variable in a Principal Component Analysis (PCA). Significance of morphometric differences was tested with Multivariate Analysis of Variance (MANOVA) with the two first principal components being treated as dependent variables and species as independent variables. The first two principal components were used since they explained the majority of observed variation in shape. A Discriminant Function Analysis (DFA) was performed to test classification of individuals in predicted groups. All the statistical analysis were performed in R (R Development Core Team 2011) adopting a 5% significance cut-off. PCA was used to detect groups representing putative cryptic species and DFA was subsequently applied to identify the set of characters that best diagnose those groups (Padial and De la Riva 2009). Additionally we noted large size in the HAL of the new species of Amazophrynella, and we used an Analysis of Variance (ANOVA) of the original data (from A. minuta, A. matses, A. amazonicola and the new species) to statistically support this hypothesis.
Sequences were aligned using the Clustal W algorithm (Thompson et al. 1996) implemented in BioEdit (Hall 1999 and alignment was adjusted as necessary against the secondary structure of the 16S rDNA. The existence of lineages in a phylogenetic tree-based context (Baum and Donoghue 1995) was performed using Maximum Likelihood analysis  in the program Treefinder (Jobb 2008) using the GTR+I+G model of substitution, selected via Akaike information criterion as implemented in Modeltest 3.7 (Posada 2006). Phylogenetic support was assessed via 10 000 non-parametric bootstrap (Felsenstein 1985). Additionally uncorrected pairwise genetic distances between linages identified by phylogenetic inference of Amazophrynella were calculated in MEGA 5.05 (Tamura et al. 2007).
Molecular species delimitation. Evolutionary lineages are diagnosed by discontinuities in character variation among lineages, and correspond to phylogenetic species. The existence of lineages is therefore a necessary and sufficient prerequisite for inferring the existence of a species under the different conceptualizations of the Phylogenetic Spe-cies Concept (PSC) (Cracraft 1983;Baum and Donoghue 1995;De Queiroz 2007). The existence of lineages in a non-tree-based context (Cracraft 1983) was inferred using Population Aggregation Analysis performed at the level of an individual (Davis and Nixon 1992;Rach et al. 2008) using the dataset with the Amazophrynella minuta species complex: A. matses, A. minuta, A. amazonicola and the new species. The analyses were performed in the program R (R Development Core Team 2011).

Bioacoustics
We analyzed one advertisement call obtained from the CD of Frogs of Tambopata, Peru (Macauly Library of Natural Songs and Cornell Laboratory of Ornithology) by the authors Cocroft et al. (2001) from the Natural Reserve of Tambopata, a locality of occurrence of the new species. The call was edit with the software Audacity 1.2.2 for Windows (Free Software Foundation Inc. 1991). The spectral and temporal parameters of the recording were analyzed in the software Raven Pro. 1.3 for Windows (Cornell Laboratory of Ornithology). The advertisement call was obtained from one male in a temperature 25 °C (Crocoft et al. 2001). We measured the following quantitative parameters: call duration (seconds); pulses per call; length of silence between calls (seconds); dominant frequency (kHz); fundamental frequency (kHz) and time to peak at maximum frequency (seconds).

Phylogenetic analysis and systematics
In the resulting phylogeny, the six nominal species of Amazophrynella were recognized as monophyletic (Fig. 1). In the genus we can distinguish two monophyletic groups: One clade (bootstrap support = 100) formed by the species: A. manaos, A. bokermanni and A. vote and another represented by the species of the A. minuta "species complex" (bootstrap support = 98): A. minuta, A. amazonicola, A. matses and the new species described herein.
In the first clade the Amazophrynella species: A. manaos is sister taxon of the possible new specie from the Guiana Shield: A. sp. aff. manaos (bootstrap support= 91), and both are sister to A. bokermanni (bootstrap support= 98). Amazophrynella vote is sister of A. bokermanni + (A. manaos + A. sp. aff. manaos) with a bootstrap support of 81.
The second clade corresponding to the A. minuta "species complex", A. amazonicola is sister of A. minuta + A. sp. aff. minuta from western Amazonia (bootstrap sup-port= 99). Our analysis further highlighted the occurrence of a new monophyletic lineage (A. javierbustamantei sp. n.) showing sister relationship with A. matses (bootstrap support = 96), both being in turn sister group of A. amazonicola + (A. minuta + A. sp. aff. minuta) with a bootstrap support of 99. Smallest uncorrected 16S rDNA p-distances estimated between phylogenetic linages was observed between A. minuta and A. sp. aff. minuta (= 3%). Greatest interspecific distance (= 14%) was observed between Amazophrynella javierbustamantei sp. n. and A. bokermanni and was comparable to divergence observed between A. manaos and A. minuta. Within the "A. minuta" species complex, the new species shows a high degree of genetic divergence from A. minuta (= 7%), A. amazonicola (= 9%) and minor genetic distance with their sister taxon A. matses (= 3%) (see all pairwise genetic distance values summarized in Table 3). According to the Population Aggregation Analysis, the newly identified lineage was also diagnosable by thirteen molecular autapomorphic characters (Table 4) leading us to the conclusion that this lineage corresponds to a new species.

Morphometric analysis
Comparative analysis of quantitative morphological data allowed us to distinguish Amazophrynella sp. n. from the other members of the A. minuta "species complex". The first two principal components extracted by the PCA account for 48.56% of the variation found in the dataset. The first component (PC1) explained 24.93% of total variation. In the first principal component axis, A. amazonicola is distinguished from the other species due to its larger size (SVL = 14.9 ± 0.7 mm, see Table 1), sharing relative size with A. minuta sensu stricto (SVL = 13.5 ± 0.6 mm, see Table 1), the species A. matses is distinguish by having the smallest size of the genus (SVL range= 12.1 ±0.6 mm, see Table 1), and shares this characteristic with Amazophrynella sp. n. (SVL = 14.9 ± 0.9 mm, see Table 1) (Fig. 2). The second component explains 23.63% of the variation. This axis represents a shape variation vector; in this axis Amazophrynella javierbustamantei sp. n. is well distinguished from the three formally described species, sharing more similarity with A. matses (Table 5).
All the species of the group are significantly different in shape (MANOVA, F 24.3 , Pillae´s trace < 0.001). The discriminate function analysis (DFA) found specimens correctly classified in 56.6% of cases and a moderate prior probabilities of groups (A. minuta = 28.75%, A. amazonicola = 18.75%, A. matses = 16.25% and A. javierbustamantei sp. n. = 36.25%). The variables that contributed most to the classification were HAL, SVL and TAL (Table 6). The differences in HAL were significant (ANOVA, F 45.27 , P < 0.001) among all the species of A. minuta "species complex" (see Fig. 1), and reveals Amazophrynella javierbustamantei sp. n. as the species with the largest HAL (Fig. 3).
Comparison with other species. Amazophrynella javierbustamantei sp. n. (Figs 4, 5, 6) differs in the following character states (states of other species in parentheses). From A. minuta (Fig. 6A) by having body skin texture tuberculate (roughly granular); relative abundance of spiny granules on the forelimbs (prickly warty skin on axillary region of the forelimbs); absence of large warts on dorsum (presence of large warts); throat and chest cream-grayish (light brown); posterior side of belly color pale orange yellowish with tiny rounded black or dark brown spots (throat and the whole belly intensely orange yellowish); tiny rounded black spots covering the belly (irregular black ocelli or blotches); metatarsal tubercle rounded (oval). From A. bokermanni (Fig. 6B) relative size of fingers, with finger I shorter than II (I>II); snout vent length smaller in males (15.8 mm) and females (22.25 mm) (A. bokermanni with maximum 22 mm SVL in males and 28 mm SVL in females, see Izecksohn 1993); smaller snout in males, with 2.2 mm SL, n = 26 (2.7 mm SL, n = 5; see Table 1); posterior side of belly color pale orange yellowish with tiny rounded black or dark brown spots (white coloration with small black dots). From to A. vote (Fig. 6C) snout subacuminated in dorsal view (rounded); posterior side of belly color pale orange yellowish with tiny rounded black or dark brown spots (ventral color pattern reddish brown, with presence of small white dots). From A. manaos (Fig. 6D) snout subacuminated (snout truncate); dorsal skin finely granular (dorsal surfaces granular); throat and chest grayish (dark coloration); posterior side of belly color pale orange yellowish with tiny rounded black spots (venter cream with black spots or stripes). From to A. matses (Fig. 6E) snout subacuminated (snout slightly truncate), edges of nasal protrusion not dilated (dilated in ventral view); shape of palmar tubercle rounded (palmar tubercles elliptical); finger tips unexpanded (expanded), rounded tiny black spots covering the belly (medium-sized black ocelli or streaks); coloration of the belly pale yellow (belly completely yellow). From A. amazonicola (Fig. 6F) by the absence of small triangular protrusion on the tip of the snout in both dorsal and ventral views (presence); body surface granular (finely granular), dorsum uncovered with medium-sized granules scattered irregularly (covered with medium-sized granules scattered irregularly); posterior side of belly color pale orange yellowish with tiny rounded black or dark brown spots (orange yellowish with dark red and brown blotches).
Description of the holotype. Body slender, head triangular, slightly longer than wide; head length 35.5% of SVL, head width 30.9% of SVL. Snout long, subacuminate in dorsal view, protruding in lateral view; canthus rostralis straight and loreal region vertical; without papilla; snout length 39.0% of head length; tympanic membrane and tympanic annulus not apparent through the skin, skin of the tympanic area covered by round sub-conical warts; vocal sac externally not visible, vocal slits absent; eyes prominent 23.8% of head length; upper eyelid covered with small tubercles; those close to the external margin aligned in a more or less distinct row; nostril closer to snout than  to eyes; internarial distance smaller than eye diameter; presence of a line of small spiny granules from the outer edge of the mouth to upper arm, choanas small and circular.
Dorsal skin finely tuberculate with several large tubercles scattered sub-conical tubercles on upper arm; texture of ventral skin granular, covered by rounded granules. Dorsolateral surfaces, granular, with presence of large rounded tubercles. Forelimbs slender, upper arm length 29.6% of SVL; edges of lower arm and upper arm finely tuberculate with several large sub-conical and spiny granules; hand length 76.5% of upper arm length; fingers slender, tips not expanded; relative length of fingers I<II<IV<III; supernumerary tubercles and accessory palmar tubercles present, palmar large and rounded, supernumerary tubercles low, small rounded; subarticular tubercles rounded and small, one tubercle on fingers I, II and IV and two on finger III; fingers I and II basally webbed; indistinct nuptial pad. Hind limbs slender; ventral skin from thigh to tarsus covered by spiny tubercles, foot length 66% of thigh length; relative length of toes I<II<V<III<IV: inner metatarsal tubercle oval, 2.5× larger than outer; outer metatarsal tubercles small, rounded; subarticular tubercles present, rounded, present one on fingers I, II, and two on fingers III, V and three on finger IV; and tip of toes not expanded.
Coloration of the holotype. In life: dorsum of the holotype mostly light brown with dark brown in the dorsum; dorsolaterally creamish-brown with scattered black blotches; dorsal surfaces of hands and feet creamish-brown, and gray on arms and legs; belly creamish-gray with black dots, and the throat gray; fingers, toes and plantar surfaces reddish-black; groin with orange marks; iris with a bronze ring; cloaca with orange flap, black pupil and bronze iris. In alcohol: dorsum brownish-grey; venter cream with black and brown dots; orange surfaces turned cream, with a white longitudinal stripe on upper jaw extending from nostril to forearm.
Variation. The new species is phenotypically variable. In some individuals (e.g. MHNC 8245 and MHNC 11002, see Fig. 5) patterning on the dorsum varies, with these specimens presenting brown chevrons extending from the head to the vent. Some individuals showed a white line extending from the tip of the nose to the upper arm. Another specimen (MHNC 9739, see Fig. 5) presented a yellow pale coloration in the axillary region (in ventral view). In some individuals, the coloration of the throat extended onto the chest (e.g. MHNC 11002, MHNC 9739 and MHNC 8245, see Fig. 5). The pale yellow coloration of the belly surface may extend from thighs to the chest or just to the middle of the belly (e.g. MHNC 8362, see Fig. 5 and Fig. 7B). In some individuals, the thighs are abundantly covered by rounded tiny spots extending to the shank (Fig. 7B). In preserved specimens the dorsum becomes light brown and the belly coloration vary from white to yellow pale (e.g. MHNSM 31255 and MHNSM 17993, see Fig. 5). The color of the finger becomes pale red and in other individuals the red coloration of the fingers became brown or orange (Fig. 5). Bioacoustics.
The following values are presented as: min-max (average ± SD, number of notes). The call is a trill type call issued during continuous and regular intervals (Fig. 8). Each note had a duration of between 0.03 to 0.08 seconds (0.05 ± 0.01 seconds, n = 20). The number of pulses varied between 8 to 18 pulses per note (10.4 ± 2.6 pulses/note, n = 20). The silence between notes varied from 0.4 to 1.6 seconds (0.8 ± 0.3 seconds, n = 20). The dominant frequency varied from 3962.1 to 3789.8 kHz (3927.6 ± 70.7 kHz,n = 20), and coincides with the fundamental frequency. Time to peak amplitude was around 0.014 to 0.04 seconds (0.02 ± 0.01 seconds, n = 20). Distribution, ecology and conservation. Amazophrynella javierbustamantei sp. n. is known from the Department of Cusco, in the lower Urubamba river basin and Department of Madre de Dios (Inambari, Candamo and Nueva Arequipa) in Peru (Fig.  9). Its distribution can vary from 215 m a.s.l. to 708 m a.s.l. Additional specimens were recorded at Los Amigos Biological Station , Tapir Lodge, and Explorers Inn, in Tambopata National Reserve. Individuals were active during the day, jumping on leaf litter, at night they were sleeping on leaves around 30 cm above ground. This species breeds close to the edges of permanent oxbow lakes, males call during the day while perched above streams in tangles (Cocroft et al. 2001). Three of the localities, km 105, 107 and 117 of the highway Puerto Maldonado-Cusco, Department Madre de Dios, show evidence of serious environmental impacts due to illegal gold mining activities, with forest and soil removed, and environmental pollution via organic and inorganic chemicals and heavy metal (specially mercury) poisoning. In addition, the new species is distributed inside of territories where oil companies are operating. On the other hand, the species is present in two protected areas, the Tambopata Natural Reserve and Machiguenga Communal Reserve. The conservation status of this species remains unknow, but was listed in 2008 as Least Concern on the IUCN red list (2015), because it was confused with Amazophrynella minuta, and because Amazophrynella minuta s.l. had wide distribution at that time, apparent tolerance of a certain degree of habitat modification, presumed large population, and because it is unlikely to be declining, and thus did not qualify for listing in a more threatened category. With recent studies the genus, the species complex of Amazophrynella minuta, was split in five species, three of them are now formally described for Peru (Amazophrynella matses, A. amazonicola and A. javierbustamantei sp. n.). The recognition of these new species will require the reevaluation of the conservation status of these species. It should also act as an impetus for additional field and laboratory studies of Peruvian amphibians, in order to understand the real conservation status of this fauna.
Etymology. The species is named after Dr. Javier Bustamante, a Peruvian residing in United States, to whom we dedicate this species in recognition of his friendship and support of herpetological taxonomy and systematics research and amphibian conservation in Peru.

Discussion
Taxonomic reviews of Amazonian amphibians suggests that morphological characters are too conservative to permit delimiting species since closely related species share similar morphologies, and amphibians in general are morphologically conservative (e.g., Elmer et al. 2007;Fouquet et al. 2007c;Funk et al. 2011;Padial et al. 2009). Thus, the use of integrative techniques in taxonomy is revolutionizing the identification and delimitation of species based on independent lines of evolutionary evidence (Dayrat 2005; Padial and De la Riva 2009). The use of an integrative approach not only allows for the discovery and delimitation of new species, it also helps us to understand the mechanism of species formation. Thus, integrative taxonomy allows us to have a better understanding of the true scope of anuran diversity in the Amazon, and it allows us to better understand the processes that generated this biodiversity.
The taxonomic ambiguity surrounding the name A. minuta and to a lesser extent A. bokermanni resulted in a severe underestimation of the taxonomic diversity of this genus. Since the descriptions of A. minuta in 1941 and A. bokermanni in 1993, the taxonomy of the genus has not been revised, leading to misdiagnoses of other species as either A. minuta or A. bokermanni due to the relatively generalized descriptions of these taxa. Three publications since 2012 (Ávila et al. 2012;Rojas et al. 2014Rojas et al. , 2015 described four new species, increasing the taxonomic diversity of the genus by 200%. All four species were previously classified as populations of a single species with a large distribution (A. minuta sensu lato). Although striking, the severe underestimation of taxonomic diversity observed in Amazophrynella and the existence of multiple lineages in Amazophrynella minuta is nothing particular to this group. Examples of other Amazonian species complexes include Rhinella margaritifera and Scinax ruber, Pristimantis ockendeni, Pristimantis fenestratus, Engystomops petersi, Hybsiboas fasciatus, Dendropsophus minutus and Osteocephalus taurinus Elmer and Canatella 2008;Padial et al. 2009;Funk et al. 2011;Caminer and Ron 2014;Gehara et al. 2014, Jungfer et al. 2013.
The descriptions by Rojas et al. (2014Rojas et al. ( , 2015 were based, in part, on diagnostic characters observed in the 16S rDNA. This gene is widely used as a DNA barcode for amphibians, for reliable species identification (Vences et al. 2005, for evaluating monophyly of species and for discovering divergent lineages (Pa-dial et al. 2009, Crawford et al. 2010Padial et al. 2010 andPadial et al. 2012). Based on 16S rDNA analyses, we also have evidence that A. bokermanni and A. vote represent species complexes (RRRZ, personal observation). This observation is in addition to the existence of the two candidate species of Amazophrynella already observed in previous analyses: one from the Guiana Shield (A. sp. aff. manaos), sister taxon of A. manaos, and another from Ecuador (A. sp. aff. minuta), sister taxon of A. minuta sensu stricto (Fig. 1). Although the taxonomic status of these candidate species will need to be confirmed using morphological and bioacoustics data, it is clear that even with the recent descriptions, the taxonomic diversity of the genus remains underestimated.
While part of our evidence for the existence of the new species as well as those described previously by Rojas et al. (2014Rojas et al. ( , 2015 comes from the use of molecular data, the descriptions make use of other data types and non-molecular diagnoses. Thus these undiscovered lineages were not truly cryptic (morphologically cryptic), but rather the result of poor taxonomic knowledge of the group. In this respect, the genus Amazophrynella again is not the exception, but rather the norm.
The new species A. javierbustamantei sp. n. is clearly differentiated in multivariate morphometric space from the other members of the Amazophrynella minuta "species group" (A. minuta, A. amazonicola and A. matses). Together with the description of Amazophrynella javierbustamantei sp. n. we also provide advertisement call. Amazophrynella javierbustamantei sp. n. is only the second species of the genus for which an advertisement call is known and recorded (see Duellman 1978). Acoustics can provide evidence of potentially new species with behavioral or premating isolating mechanisms (e.g. De la Riva et al. 1997;Gerhardt 1998;Simões et al. 2008, Padial andDe la Riva 2009;Padial et al. 2012), thus providing evidence of evolutionary mechanisms that contributed to the species diversity of the genus Amazophrynella.
The threats to the biological conservation of A. javierbustamantei sp. n. are evident, with uncontrolled exploration for gold, illegal mining and the destruction of habitat in the Departments of Madre de Dios and Cusco, probably causing a significant reduction in the population sizes of the species and fragmenting its distribution. For these reasons is necessary to analyze the current population status and trends of this and another amphibian species in this Department of southern Peru. quisas da Amazonônia -INPA), Göran Nilson (Naturhistoriska Museet, Göteborg -NHMG), Marcos A. Carvalho (Universidade Federal de Mato Grosso -UFMT) and Juan Carlos Cusi (Museo de Historia Natural de la Universidad Nacional Mayor de San Marcos -MUSM), Percy Yanque and Rocio Orellana (Museo de Historia Natural del Cusco -MHNC). We are grateful to Asociación para la Conservación de la Cuenca Amazonica (ACCA) which partially funded the expedition to Madre de Dios, under the project Manu-Tambopata Conservation Corridor,that resulted in the discovery of the new species. VTC was supported by a fellowship from Fundação de Amparo a Pesquisa do Estado do Amazonas (FAPEAM), IF and TH were supported by research productivity fellowships from the Conselho Nacional de Desenvolvimento Cientifico e Tecnológico (CNPq). Collecting permits in Peru were granted by Dirección General