See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/236885902

A new species of Bokermannohyla (Anura:

Hylidae) from highlands of Central Brazil

Article in Zootaxa · January 2012

CITATION READS

1 250

8 authors, including:

Reuber Brandao Rafael Félix de Magalhães

University of Brasília Federal University of Minas Gerais
73 PUBLICATIONS 538 CITATIONS 7 PUBLICATIONS 9 CITATIONS

SEE PROFILE SEE PROFILE

Leandro Campos Natan Medeiros Maciel

University of Brasília Universidade Federal de Goiás
14 PUBLICATIONS 88 CITATIONS 33 PUBLICATIONS 258 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Projeto Amplexo View project

REVISÃO TAXONÔMICA DE RHAEBO GUTTATUS (SCHNEIDER, 1799) (ANURA: BUFONIDAE) View project

All content following this page was uploaded by Reuber Brandao on 08 July 2017.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
 Zootaxa 3527: 28–42 (2012) ISSN 1175-5326 (print edition)
www.mapress.com / zootaxa/
ZOOTAXA
Copyright © 2012 · Magnolia Press
Article ISSN 1175-5334 (online edition)

urn:lsid:zoobank.org:pub:81909153-1E90-4348-8B23-9A02D31B7789

A new species of Bokermannohyla (Anura: Hylidae) from highlands of Central Brazil

REUBER ALBUQUERQUE BRANDÃO1*, RAFAEL FÉLIX DE MAGALHÃES1,2, ADRIAN ANTONIO

GARDA3, LEANDRO ABRÓSIO CAMPOS4, ANTONIO SEBBEN4 & NATAN MEDEIROS MACIEL2
1
Laboratório de Fauna e Unidades de Conservação, Departamento de Engenharia Florestal, Universidade de Brasília, 70910-900,
Brasília, Distrito Federal, Brasil.
2
Laboratório de Herpetologia e Comportamento Animal. Departamento de Ecologia, Instituto de Ciências Biológicas, Universidade
Federal de Goiás, 74001-970, C.P. 131 - Goiânia, Goiás, Brasil.
3
Laboratório de Anfíbios e Répteis, Departamento de Botânica, Ecologia e Zoologia, Universidade Federal do Rio Grande do Norte,
Campus Universitário Lagoa Nova, 59078-900, Natal, Rio Grande do Norte, Brasil.
4
Laboratório de Anatomia Comparativa de Vertebrados, Departamento de Ciências Fisiológicas, Instituto de Ciências Biológicas,
Universidade de Brasília, 70910-900, Brasília, Distrito Federal, Brasil.

Abstract

We describe a new species of hylid frog, Bokermannohyla sapiranga sp. nov., from Central Brazil (15°55’49” S,
47°52’59” W, 1110m asl). The species is characterized by medium size (males snout-vent length 45.6±4.7mm, N=13; fe-
males 46.9±6.2, N=4) for the B. pseudopseudis group and by iris color varying from orange to reddish. The advertisement
call is similar in some aspects to calls of B. pseudopseudis, but differs mainly by structure and spectral features, but also
temporal ones such as call length, note duration, and number of notes per call. The new species uses streams with muddy
beds in gallery forests as well as rocky bed streams in the highlands of Goiás State and Distrito Federal. The new species
is tentatively included in the B. pseudopseudis group based on shared morphological features, ecology, and behavior. Mor-
phometric analysis of size-independent variables indicated that B. sapiranga differs from B. pseudopseudis mostly in head
length and width, interorbital distance, and diameter of the fourth toe disc.

Key words: Advertisement call, Bokermannohyla pseudopseudis group, Bokermannohyla sapiranga sp. nov., Brazilian Cerrado

Resumo

Uma nova espécie de hilídeo, Bokermannohyla sapiranga, é descrita para o Brasil Central. A espécie é caracterizada pelo
tamanho médio (comprimento rostro-cloacal dos machos 45,6±4,7 mm, N=13; fêmeas 46,9±6,2, N=4) para as espécies de
Bokermannohyla gr. pseudopseudis e íris avermelhada. O canto de anúncio é semelhante ao de B. pseudopseudis diferindo
principalmente na estrutura e parâmetros espectrais, mas também em parâmetros temporais como duração do canto e
número de notas. A nova espécie ocupa riachos com leito argiloso em matas de galeria, bem como riachos de leito rochoso
nos planaltos de Goiás e do Distrito Federal. A nova espécie é preliminarmente incluída no grupo de espécies de B. pseu-
dopseudis com base em características ecológicas, morfológicas e comportamentais. Uma análise morfométrica de
variáveis controladas para o tamanho do corpo mostrou que B. sapiranga difere de B. pseudopseudis principalmente na
largura e comprimento da cabeça, distância interorbital e diâmetro do disco do quarto dedo.

Palavras-chave: Bokermannohyla gr. pseudopseudis, Bokermannohyla sapiranga sp. nov., canto de anúncio, Cerrado brasileiro

Introduction

Bokermannohyla Faivovich, Haddad, Garcia, Frost, Campbell and Wheeler, 2005 is a Brazilian treefrog genus with
31 species described (Leite et al. 2012), encompassing four groups: B. circumdata, B. claresignata, B. martinsi, and
B. pseudopseudis (Faivovich et al. 2005). The B. pseudopseudis group (sensu Faivovich et al. 2005) is supported
only by molecular characters (mitochondrial and nuclear genes fragments), and includes the following species: B.

28 Accepted by S. Castroviejo-Fisher: 4 Sept. 2012; published: 26 Oct. 2012

pseudopseudis (Miranda-Ribeiro 1937), B. alvarengai (Bokermann 1956), B. saxicola (Bokermann 1964); B.
ibitiguara (Cardoso 1983), B. itapoty Lugli and Haddad 2006, B. oxente Lugli and Haddad 2006, B. sagarana Leite,
Pezzuti and Drummond 2011, and B. flavopicta Leite, Pezzuti and Garcia 2012 (Faivovich et al. 2009; Leite et al.
2012). Although no formal morphological synapomorphies have been proposed for the group, a combination of
features characterizes its species: robust body; a well-developed prepollex with a distal element ending in a simple
spine; forearm highly hypertrophied; large tadpoles with high number of posterior denticle rows and oral disc
completely surrounded by papillae; low ability for perching on vegetation, presence of basking behavior, and males
usually larger than females (Miranda-Ribeiro 1937; Bokermann 1964; Cardoso 1983; Pombal Jr. & Caramaschi,
1995; Brandão et al. 2005; Lugli & Haddad 2006a; Tattersall et al. 2006; Leite & Eterovick, 2010). These species
are usually associated with fast-flowing highland streams, where tadpoles develop (Eterovick & Brandão 2001;
Leite & Eterovick 2010). The species of the B. pseudopseudis group are typically found in open fields or savannas
on rocky areas of highlands, mainly in the Cerrado biome. Males are found calling next to waterfalls. Individuals
present a distinctive dorsal color pattern, composed of gray, dark gray or grayish brown backgrounds, with round
darker blotches. This color pattern provides effective camouflage on rocky surfaces backgrounds, and some species
have dorsal patterns similar to lichens encrusted on rocks (Leite et al. 2012).
Most of the species in the Bokermannohyla pseudopseudis group occur in mountain ranges in the States of
Minas Gerais and Bahia (Serra da Canastra and Serra do Espinhaço mountain ranges), but one species (B.
pseudopseudis), is found in the highlands in the States of Goiás and Distrito Federal.
Recent comparisons showed that some populations previously treated as B. pseudopseudis (e.g. Pombal Jr. &
Caramaschi 1995; Brandão & Araújo 2001; Eterovick & Brandão 2001) could be, in fact, a distinct species. Herein,
we describe those populations as a new species, including its advertisement call. Based on shared features
(morphology and behavior), we tentatively include the new species in the B. pseudopseudis group.

Material and methods

Specimens used in the description and for comparisons were obtained from Coleção Herpetológica da Universidade
de Brasília (CHUNB), Coleção Célio Fernando Baptista Haddad—Universidade Estadual de Paulista (CFBH),
Coleção Antonio Sebben—Universidade de Brasília (ASUnB), Coleção Ariovaldo Antônio Giaretta—Universidade
Federal de Uberlândia (AAG-UFU), and Coleção Zoológica da Universidade Federal de Goiás (ZUFG).
We obtained adults, juveniles, and recorded advertisement calls of the new species in Brasília (Distrito Federal,
Brazil), and Pirenópolis municipality (state of Goiás, Brazil) from October 1994 to December 2009. Other
localities are presented below. We measured adult specimens (males and females) with digital calipers (to nearest
0.1mm), following Napoli and Caramaschi (1998): snout-vent length (SVL), head length (HL), head width (HW),
eye diameter (ED), interorbital distance (IOD), eye-nostril distance (END), internarial distance (IND), nostril-
snout distance (NSD), tympanum diameter (TD), third finger disc diameter (FDD), hand length (HD), forearm
length (FL), arm length (AL), thigh length (TGL), tibia length (TBL), foot length (FTL), and fourth toe disc
diameter (TDD). We also measured the prepollex length (PPL) and forearm width (FW). We follow Heyer et al.
(1990) for snout dorsal outline and snout profile standards, and Savage and Heyer (1967) for web formula. We
determined sex by direct observation of secondary sexual characters, such as developed forearm and prepollex in
males, and by direct observation of gonads. We produced differential diagnoses by measuring and examining all
species belonging to the B. pseudopseudis group, but B. sagarana and B. flavopicta. Information regarding B.
sagarana and B. flavopicta was obtained from the diagnoses presented in Leite et al. (2011) and Leite et al. (2012),
respectively.
Morphometric Analysis. We performed a model selection using Akaike Information Criterion to find the best
model to distinguish B. sapiranga sp. nov. and B. pseudopseudis, the morphologically most similar species. We
used two different approaches to search for differences in shape and size between species. We performed model
selection using log10-transformed morphometric variables for allometric differences between species, and using
size-adjusted morphometric variables to search for isometric differences between the species.
We removed the effects of body size from log10-transformed variables using the method described by Burnaby
(1966), where the n X p matrix of log10-transformed data is post-multiplied by a p X p matrix L defined as:
L = I p −V (V T V )−1V T

A NEW SPECIES OF BOKERMANNOHYLA FROM CENTRAL BRAZIL Zootaxa 3527 © 2012 Magnolia Press · 29
 where Ip is a p X p identity matrix, V is an isometric eigenvector and VT its transpose matrix (Rohlf &
Bookstein 1987). This procedure corrects for the effects of body size, and we refer to the resulting variables as
Shape Variables (Colli et al. 2003).
Morphological variables are dependent on allometric body grow, thus producing statistical correlations among
them (Bookstein 1996). We decided then to conduct an exploratory selection of regression models by using Akaike
information criteria (AIC) (Burnham & Anderson 2002; Burnham & Anderson 2004; Johnson & Omland 2004).
The AIC model selection was conducted in SAM software 4.0 (Rangel et al. 2010). All possible combinations of
variables (524,287 models) were tested to compare Bokermannohyla pseudopseudis against the new species. By
selecting among the 19 morphometric variables, we expected to find those with the best predictive power to
discriminate among species, decreasing the number of variables potentially correlated.
The analysis was conducted in two steps: (1) a selection of models was performed by using the log10-
transformed adjusted morphometric variables (with no size effect) as predictors; (2) the same procedure was
conducted with the log-transformed morphometric variables without allometric transformation. The selection of
the best models, in both cases, was based on the delta Akaike corrected criteria (ΔAICc), where values of ΔAIC <
2 were considered substantially supported (Burnham & Anderson 2002; Burnham & Anderson 2004).
The first criterion used to select the best model was AICw, computed as e-1/2(ΔAICc)/∑ΔAICc (Burnham &
Anderson 2002). High AICw values indicate greater chances that the model is a good predictor of reality. The
second criterion used was the value that best explained the regression model (r²). Higher values of r² and AICw
indicate more support for variables selected by the model as good predictors regarding differences between species.
At last, we used all variables of the minimum model to perform a principal component analysis (PCA) (correlation
matrix), because the minimum model has a higher importance compared to others. In addition, it had a r² higher
than the average model (see Results).
Advertisement Call. We recorded advertisement calls with a Marantz PMD 660 digital recorder coupled to a
Sennheiser ME66 microphone. Audiospectrograms were analyzed in SoundRuler (Gridi-Papp 2007), Avisoft
Light-SAS 4.53, and Adobe Audition 3.0. We used a Fast Fourier Transformation (FFT) at 256 points, a sampling
frequency of 22 kHz, and a 16 bit resolution. We recorded calls from one specimen at Centro de Instrução e
Adestramento de Brasília (CIAB), Brasília, Distrito Federal on December 11, 2004 21:00h (air temperature,
19.9ºC; air humidity, 75%) and in the surroundings of Parque Estadual da Serra dos Pirineus, Pirenópolis
Municipality, Goiás State on March 13 2009, 20:09h (air temperature, 26ºC; air humidity, 74%) (one specimen).
We compared results with calls described for other species in the Bokermannohyla pseudopseudis group
(Bokermann 1964; Cardoso 1983; Eterovick & Brandão 2001; Guimarães et al. 2001; Lugli & Haddad 2006b),
mainly using Lugli and Haddad (2006b). The terminology for advertisement calls follows Heyer et al. (1990).

Results

Bokermannohyla sapiranga sp. nov.

(Figs. 1–2)

Holotype. Brazil, Distrito Federal, Reserva Ecológica do Roncador (15°55’49” S, 47°52’59” W, 1110m asl),
CHUNB 62384, adult male, R.A. Brandão, R.D. Françoso, T. Marques, R.M. Japiassu, and L.R.A. Braga col., 10
December 2009.
Paratypes. Brazil, Distrito Federal, Brasília (15°52’25” S, 47°51’59” W, 1070m asl), CFBH 2248–2249, adult
males, A. Sebben col., no date. CHUNB 14383–14384, adult females, A. Sebben col., 01 April 1986. CHUNB
38832, adult male, unknown collector, 01 May 1993. CHUNB 14377–14379, juveniles, B. A. Duar col., 04
November 1994. Distrito Federal, Poço Azul (15°34’36’’ S, 48°02’30’’ W, 1090m als), CHUNB 40854–40855,
adult males, R. A. Brandão col., 05 October 1994. CHUNB 14369–14376, juveniles, R. A. Brandão col., 23
October 1994. Distrito Federal, Fazenda Água Limpa (15°56’44’’ S, 47°54’34’’ W, 1085m asl), CHUNB 25080,
adult male, unknown collector, 23 November 2002. CHUNB 49652, adult male, P. H. Valdujo col., 14 February
2007. CHUNB 5079550796, adult males, G. R. Colli col., 27 November 2007. State of Goiás, Cristalina
Municipality (16°44’30’’ S, 47°41’36’’ W, 1050m asl), CHUNB 8369, adult male, G. R. Colli col., no date.
CHUNB 14387, adult male, R. A. Brandão, col., 22 May 1998. State of Goiás, Pirenópolis Municipality

30 · Zootaxa 3527 © 2012 Magnolia Press BRANDÃO ET AL.

(15°49’08’’ S, 48°59’08’’ W, 880m asl), CHUNB 14386, adult male, R. A. Brandão col., 22 June 1998. CHUNB
14380, adult male, R. A. Brandão col., 01 September 1992. Reserva Particular do Patrimônio Natural Vaga Fogo
(15°49’17’’ S, 48°59’44’’ W, 810m asl), CHUNB 14390, adult male, M. G. Zatz col., 03 January 1999. Cocalzinho
Municipality, Parque Estadual dos Pireneus (15°48’16’’ S, 48°50’05’’ W, 1270m asl), CHUNB 14388, adult male,
R. A. Brandão col., 28 December 1998. Novo Gama Municipality (16°03’24’’ S, 48°01’21’’ W, 1080m asl),
CHUNB 14381, adult male, A. Sebben col., 10 November 1987.

FIGURE 1. (A) Holotype of Bokermannohyla sapiranga (CHUNB 62384). (B) Adult male of Bokermannohyla pseudopseudis
from the Parque Nacional da Chapada dos Veadeiros Goiás State, Brazil. Photos by Reuber Brandão.

A NEW SPECIES OF BOKERMANNOHYLA FROM CENTRAL BRAZIL Zootaxa 3527 © 2012 Magnolia Press · 31
 Diagnosis. The new species is diagnosed by the following combination of features: Medium size for the
Bokermannohyla pseudopseudis group; dorsal coloration brownish ochre with darker scattered blotches; large
eyes, forelimbs robust and short, with hypertrophied forearms in males; orange to reddish iris (bronze in some
individuals); snout rounded in lateral and dorsal views (slightly acuminated in lateral view in some individuals);
poorly-developed tibial fold, call frequency ranging from 0.5 to 0.7 kHz.
Comparison with other species of the Bokermannohyla pseudopseudis species group. Bokermannohyla
sapiranga differs from B. alvarengai (in parentheses) by having: 1) snout rounded in dorsal and lateral views
(truncate snout in dorsal view and vertical in lateral view); 2) head slightly longer than wide (head wider than
long); 3) tympanic fold present (absent); 4) tibial fold less conspicuous, not extending from the metacarpal external
tubercle to the elbow (very conspicuous tibial fold, from the elbow to the basis of the external metacarpal tubercle);
5) dorsum with several scattered blotches (dorsum without pattern or large dark patches); 6) dorsum blotches
extending to the inguinal region (inguinal region without blotches or pattern); 7) thigh bars poorly defined
(conspicuous); 8) hidden surfaces of thighs with bars (hidden surfaces thighs without blotches, bars or patterns); 9)
vent tubercles small, inconspicuous, and white (vent tubercles conspicuous); 10) gular region with small circular
and darker blotches (gular region immaculate). Bokermannohyla sapiranga differs from B. flavopicta (in
parentheses) by having: 1) dorsal background brownish with dark brown blotches (dorsal pattern lichenous-like,
composed of small dark brown blotches in a light gray to light brown background); 2) similar head width and head
length (heads wider than longer); 3) glandular tissues on mental region not evident (evident glandular tissues on
mental region); 4) absence of yellow dots on the body in live individuals (small yellow dots present on lips, eyelids,
loreal and gular regions, supratympanic fold, fore and hind limbs, flanks, and anal flap in live individuals); 5)
smaller females [39.8–54.5 (N=14) in B. sapiranga; 60.2–61.6 (N=2) in B. flavopicta]. The new species differs
from B. ibitiguara (in parentheses) by having: 1) snout rounded in dorsal view (truncate in dorsal view); 2)
presence of several darker blotches on the gular region, that sometimes reach the chest (gular region immaculate
white); 3) bars on legs poorly defined, becoming paler on hidden surfaces of thighs, and resembling diffuse
blotches; 4) same pattern on the flanks and inguinal regions (transversal bars conspicuous, narrower, and regularly
spaced in legs, flanks, and inguinal region); 5) dorsal coloration brown, with regular and darker blotches (dorsal
color tan without blotches); 6) males with conspicuously hypertrophied forearms (males with forearm less
hypertrophied). Bokermannohyla sapiranga differs from B. itapoty (in parentheses) by having: 1) tarsal fold poorly
developed, from the metacarpal external tubercle to 4/5 of forearm length (tarsal fold more evident, from the
metacarpal external tubercle to the elbow); 2) snout rounded in lateral and dorsal views (snout truncate in lateral
and dorsal views); 3) tympanum oval-shaped (tympanum round); 4) absence of white spots on the dorsum, canthus
rostralis, and loreal region (presence of white spots in some individuals); 5) presence of blotches on the hidden
surface of thighs (thighs dark brown, without any bars or spots); 6) evident hypertrophied forearms in males
(discrete hypertrophy in forearms of males). Bokermannohyla sapiranga differs from B. oxente (in parentheses) by
having: 1) forearms of adult males more robust and hypertrophied (forearm width 3.65±0.35, range: 3.08–3.97,
N=9 for B. oxente and 5.83±0.70, range 5.02–7.68, N=13 for B. sapiranga); 2) large dark blotches on dorsum (not
present); 3) developed prepollex (less developed prepollex); 4) hidden surface of thighs with blotches (hidden thigh
surface brownish, without blotches); 5) cloacal flap without white supracloacal stripe (white supracloacal stripe
present). Bokermannohyla sapiranga can be distinguished from B. pseudopseudis (in parentheses) by having: 1)
rounded snout in dorsal view (snout truncate in dorsal view) and rounded or slightly acuminated in lateral view
(vertical); 2) iris red, orange, or bronze (iris yellowish or golden); 3) less conspicuous tarsal fold (more
conspicuous); 4) dorsal surface of the fingers and toes discs grayish brown (whitish). Bokermannohyla sapiranga
differs from B. sagarana (in parentheses) by having: 1) snout profile truncate in dorsal and lateral views (rounded
in dorsal view, truncate in lateral view); 2) dorsal background brownish with dark brown blotches (lichenous-like
dorsal color pattern: light gray with darker and lighter blotches of irregular shape); 3) bars on dorsal thigh surface
brownish, poorly defined (dark gray perpendicular bars covering dorsal surface of thighs, highly contrasted with
the background color). Bokermannohyla sapiranga differs from B. saxicola (in parentheses) by having: 1) snout
profile rounded in dorsal and lateral view (rounded in dorsal view, truncate in lateral view); 2) brownish dorsum
(dorsum light gray with darker blotches); 3) bars on thighs reaching the hidden outer surface, but poorly defined
(bars on hidden thigh surface narrow, well defined); 4) head slender (head robust); conspicuous arm hypertrophy in
males (discrete hypertrophy in males forearm); 5) absence of light stripe on the cloacal flap (light stripe on cloacal
flap present); 6) forelimbs robust (forelimbs width: 5.83±0.70, range 5.02–7.68, N=17 in B. sapiranga; forelimbs
width 3.78±0.51, range 2.96–4.60, N=7 in B. saxicola); 8) prepollex well developed (prepollex less developed).

32 · Zootaxa 3527 © 2012 Magnolia Press BRANDÃO ET AL.

FIGURE 2. Holotype of Bokermannohyla sapiranga (CHUNB 62384): Dorsal body view (A); head profile (B); ventral surface
of hand and forearm (C); ventral body view (D); ventral surface of foot and shank (E). Photos by Leandro Campos.

Description of holotype. Body robust (Fig. 1 and 2A); head slightly longer than wider; snout rounded in
lateral (Fig. 2B) and in dorsal views; nostrils slightly protuberant, nostril aperture oval; eyes large, slightly
protruding laterally; tympanum large, rounded; maximum eye length 1.5 times the maximum tympanum length;
supratympanic fold evident, well-developed, extending from the posterior eye margin to the shoulder, at the level
of the inferior tympanum margin. Vocal sac not externally expanded; vocal slits present; tongue large, wider than
longer, almost cordiform; vomerine teeth disposed in two continuous rows, placed posterior to the choanae,
forming an open angle (ten on the right and 9 on the left side); choanae large, almost circular, apart from each
other; hypertrophied forearms (Fig. 1 and Fig. 2 B–C); external forearm fold evident, extending from the elbow to
the wrist; well-developed prepollex, with a sharp spine, not exposed; slender fingers; subarticular tubercles simple,

A NEW SPECIES OF BOKERMANNOHYLA FROM CENTRAL BRAZIL Zootaxa 3527 © 2012 Magnolia Press · 33
round, and well-developed; few round and discrete palmar tubercles; a large external metacarpal tubercle, fan-
shaped, partially divided in three parts (united at the basis) by two grooves on the internal margin; finger pads
ovoid, single, well-developed; relative finger lengths IV>II>III>V; webbing hand formula II–III2-–3-IV3–2+V (Fig.
2C); nuptial excrescence small, discrete, round, placed on the dorsal surface of finger II basis, at the prepollex
junction; legs slender, thigh longer than tibia (Fig. 2D); tarsal folds evident; internal tarsal fold from the posterior
margin of the internal metatarsal tubercle to the heel (Fig. 2D); external metatarsal fold less developed, extending
two thirds the tarsal length from the first phalanx of toe V; internal metatarsal tubercle well developed, ovoid;
external metatarsal tubercle small and round; subarticular tubercles simple and round; plantar tubercles small,
discrete, round; relative toe length IV>V>III>II>I; web foot formula I1+–2II1+–1III1+–2IV2–1V (Fig. 2E). Toe
discs slightly less developed than finger discs; dorsal texture moderately granular; ventral region, chest, lateral
flanks, and ventral thigh surface densely granular (Fig. 2D); ventral tibia and ventral mandible surfaces smooth;
vent region dark; an evident fold delimits the dorsum and the cloacal flap.
Measurements of holotype (in mm). SVL: 46.98, HL: 16.25, HW: 16.86, ED: 5.51, IOD: 7.86, END: 4.46,
IND: 4.53, NSD: 3.28, TD: 2.95, PPL: 4.68, FDD: 1.90, TGL: 22.91, TBL: 25.52, FTL: 23.47, TDD: 1.83, FL:
7.48, FW: 5.75, HD: 15.72, AL: 6.56, TAL: 12.93.
Color in preservative. Dorsum brownish gray, with large dark brown or gray blotches; large dorsal dark
brown blotches similar in size; flanks light gray with large dark brown or gray blotches; venter immaculate,
whitish; gular region with several pale, brownish blotches; ventral tarsal surface brownish gray; dorsal surface of
limbs brownish gray; ventral surfaces of toe discs white; hidden surfaces of the thighs dark brown.
Color in life. Dorsum and limbs brownish, with large darker brown blotches; iris reddish (Fig. 1A), tympanum
and foot webbing brownish; venter light ochre.

FIGURE 3. Distribution of Bokermannohyla pseudopseudis and Bokermannohyla sapiranga in the two first principal
components using morphometric variables selected by Akaike Selection Criterion (B. pseudopseudis: male (■), female (□), and
young (half square); B. sapiranga: male (●), female (○), and young (half circle).

Morphometric analyses. Measurements of Bokermannohyla sapiranga are shown in Table 1.

To evaluate morphometric differences between B. sapiranga and B. pseudopseudis we used the model
selection with shape variables. In this case, eight models were selected, all combining more than nine variables
(Table 2). Just two models presented AICw value greater than 15.0. However, all the models selected share several
variables. The variables snout-vent length, eye diameter, and tibia length were less important to explain the average
model (Table 3). These variables were not recovered in any of the eight models selected by the AIC criterion (Table
2). The average model explained less than half of the differences between B. pseudopseudis and B. sapiranga (r² =

34 · Zootaxa 3527 © 2012 Magnolia Press BRANDÃO ET AL.

0.413; r² adjusted = 0.291). The variables head length, interorbital distance, head width, and fourth toe disc
diameter were the four most important selected by the average model (Table 3), suggesting that the differences
between B. pseudopseudis and the new species were related to the head proportions and size of the toe discs.

TABLE 1. Means, standard deviations, and ranges (in parentheses) of Bokermannohyla sapiranga and B. pseudopseudis.
All measurements are in millimeters.

Males Females
Variables sapiranga pseudopseudis sapiranga pseudopseudis
(N = 13) (N = 30) (N = 14) (N = 17)
45.64±4.68 52.51±6.19 46.95±6.21 51.30±4.81
Snout-vent length
(40.14–57.30) (43.13–68.72) (39.79–54.49) (44.40–59.63)
17.73±1.32 19.69±1.96 18.82±3.82 19.45±1.63
Head length
(15.56–20.04) (16.99–24.12) (14.79–23.63) (17.09–23.75)
17.58±1.32 21.12±2.64 17.51±2.21 20.23±1.88
Head width
(15.67–21.10) (16.97–28.57) (14.51–19.56) (17.91–24.89)
4.92±0.57 5.71±0.57 5.34±0.98 5.81±0.7
Eye Diameter
(4.06–6.21) (4.55–6.92) (4.19–6.46) (4.60–7.02)
5.05±0.76 6.14±0.86 5.57±1.25 5.61±0.85
Interorbital distance
(4.11–6.56) (4.85–8.39) (4.63–7.30) (4.33–7.69)
4.79±0.55 5.25±0.62 4.90±0.56 5.16±0.44
Eye-nostril distance
(4.29–6.56) (4.39–6.67) (4.38–5.39) (4.57–6.33)
4.29±0.39 4.45±0.62 4.06±0.38 4.34±0.46
Internarial distance
(3.66–4.89) (2.88–5.87) (3.71–4.45) (3.41–5.29)
1.97±0.32 1.84±0.44 2.03±0.36 1.67±0.40
Nostril–snout distance
(1.34–2.42) (1.16–2.77) (1.55–2.43) (1.00–2.72)
2.90±0.39 3.35±0.49 2.76±0.74 3.50±0.53
Tympanum diameter
(2.43–3.71) (2.49–4.67) (2.23–3.83) (2.68–4.63)
5.05±0.49 6.01±0.80 4.87±0.59 5.26±0.76
Prepollex length
(4.45–6.19) (4.74–8.09) (4.33–5.55) (4.67–7.39)
1.87±0.23 2.44±0.47 1.98±0.38 2.28±0.25
Third finger disc diameter
(1.53–2.30) (1.24–3.50) (1.52–2.38) (1.69–2.66)
22.98±1.63 27.70±3.5 24.43±2.91 26.89±3.23
Thigh length
(20.59–26.22) (20.65–37.66) (21.84–27.75) (21.91–35.63)
23.99±2.18 27.47±3.29 25.03±2.50 27.10±2.50
Tibia length
(21.04–28.90) (21.89–35.71) (22.57–28.29) (23.86–33.13)
31.62±2.47 36.43±5.15 33.38±2.25 34.76±3.61
Foot length
(27.61–35.91) (23.49–49.78) (30.49–35.65) (30.17–44.75)
1.67±0.22 2.15±0.39 1.6±0.19 2.07±0.35
Fourth toe disc diameter
(1.37–2.13) (1.36–3.21) (1.39–1.84) (1.23–2.53)
8.37±1.2 10.09±1.97 9.03±1.46 9.40±0.94
Forearm length
(6.91–11.05) (7.39–16.67) (7.25–10.32) (7.82–11.80)
5.83±0.7 7.19±1.32 3.96±0.74 4.34±0.67
Forearm width
(5.02–7.68) (5.26–10.62) (2.86–4.45) (3.23–5.61)
14.64±1.21 17.47±1.95 14.08±1.96 15.58±2.09
Hand length
(12.8–17.68) (12.31–23.56) (11.35–16.02) (12.59–21.48)
7.59±1.31 9.87±1.41 8.66±2.12 8.86±1.11
Arm length
(5.47–11.25) (6.79–12.93) (6.89–11.37) (7.29–11.06)

A NEW SPECIES OF BOKERMANNOHYLA FROM CENTRAL BRAZIL Zootaxa 3527 © 2012 Magnolia Press · 35
TABLE 2. The eight most likely models selected using AIC using morphometric variables with size effect for B.
pseudopseudis and B. sapiranga. r² = regression model explanation; AICc = corrected AIC values; ΔAICc = delta AIC;
L(gi|x) = likelihood of model; AICw = Akaike weights. Variables number: 1 = SVL; 2 = HL; 3 = HW; 4 = ED; 5 = IOD;
6 = END; 7 = IND; 8 = NSD; 9 = TD; 10 = PPL; 11 = FDD; 12 = TGL; 13 = TBL; 14 = FTL; 15 = TDD; 16 = FL; 17 =
FW; 18 = HD; 19 = AL.

Variables r² AICc Delta AICc L(gi|x) AICcw

2, 3, 5, 8, 9, 11, 14, 15, 17, 18 0.609 44.855 0 1 22.24622
2, 3, 5, 9, 11, 14, 15, 17, 18 0.596 45.175 0.32 0.852 18.95698
2, 3, 5, 9, 10, 11, 14, 15, 17, 18 0.603 46.218 1.362 0.506 11.25905
2, 3, 5, 9, 11, 14, 15, 16, 17, 18 0.602 46.341 1.486 0.476 10.58219
3, 5, 8, 9, 11, 13, 14, 15, 17, 18 0.614 46.369 1.514 0.469 10.43507
2, 3, 5, 7, 9, 11, 14, 15, 17, 18 0.601 46.555 1.7 0.427 9.508366
2, 3, 5, 6, 9, 11, 14, 15, 17, 18 0.601 46.725 1.87 0.393 8.733551
3, 5, 6, 9, 11, 14, 15, 16, 17, 18 0.612 46.832 1.977 0.372 8.278585

TABLE 3. Parameter estimates of average model.

95%
Variables and Constant Importance Coeff. Std Coeff. Std Error t 95% Upper
Lower
Constant – 1.617 0 1.508 1.073 –1.338 4.573
Head Length 1 9.195 0.464 2.678 3.434 3.947 14.443
Interorbital Distance 0.984 2.385 0.328 1.905 1.252 –1.349 6.12
Head Width 0.957 –5.891 –0.347 2.717 –2.168 –11.215 –0.566
Fourth Toe Disc Diameter 0.928 1.741 0.246 1.771 0.983 –1.731 5.213
Arm Length 0.85 1.343 0.158 1.767 0.76 –2.12 4.806
Tympanum Diameter 0.835 0.63 0.123 1.606 0.392 –2.518 3.778
Forearm Length 0.681 1.152 0.097 1.786 0.645 –2.348 4.653
Hand Length 0.647 1.949 0.122 2.014 0.968 –1.997 5.896
Third Finger Disc Diameter 0.591 –0.14 –0.028 1.59 –0.088 –3.257 2.977
Eye-nostril Distance 0.545 –3.147 –0.286 1.726 –1.823 –6.531 0.236
Forearm Width 0.539 –2.135 –0.645 1.519 –1.406 –5.113 0.842
Nostril-snout Distance 0.409 –1.711 –0.437 1.501 –1.14 –4.654 1.231
Internarial Distance 0.332 –2.538 –0.324 1.524 –1.665 –5.526 0.449
Snout-vent Length 0.315 –4.14 –0.196 1.78 –2.325 –7.629 –0.65
Prepollex Length 0.3 –2.979 –0.347 1.525 –1.953 –5.968 0.011
Eye Diameter 0.292 –3.083 –0.282 1.526 –2.021 –6.074 –0.093
Foot Length 0.262 –2.253 –0.146 1.584 –1.422 –5.358 0.851
Tibia Length 0.253 –3.251 –0.222 1.608 –2.022 –6.403 –0.099

The PCA suggests that Bokermannohyla pseudopseudis and B. sapiranga are slightly different on general
morphology (Fig. 3). The first two scores of the PCA account for 61.34% of the variance. The first component
accounts for 40.48% of the variance and showed higher positive values for nostril-snout distance and interorbital
distance, and higher negative values for third finger disc diameter, and fourth toe disc diameter (Table 4). The
second component accounts for 20.86% of the variance, presenting higher positive value for third finger disc
diameter, and higher negative value for tympanum diameter.

36 · Zootaxa 3527 © 2012 Magnolia Press BRANDÃO ET AL.

TABLE 4. PCA factor loadings for morphometric variables of Bokermannohyla pseudopseudis and Bokermannohyla
sapiranga and percentage of variance explained.

Variables 1 2 3
Head Length 0.028 –0.054 –0.125
Head Width –0.051 –0.029 –0.051
Interocular Distance 0.245 0.111 –0.252
Nostril-Snout Distance 0.809 0.072 0.502
Tympanum Diameter –0.180 –0.763 0.489
Third finger Disc Diameter –0.341 0.605 0.606
Fourth toe Disc Diameter –0.340 0.101 0.088
Forearm Length 0.068 0.101 –0.067
Hand Length –0.052 –0.046 –0.096
Arm Length 0.101 0.087 –0.196
Variance explained 40.48% 20.86% 14.38%

Advertisement call. The advertisement call of Bokermannohyla sapiranga from the type locality (Brasília
municipality) consists of series of 5 to 7 notes with no harmonic structure (Fig. 4). Some notes show a pulsed
structure (Fig. 4A). Mean call duration was 997±16.4 ms (range = 759–1202 ms, N=5, Fig. 4A). Mean note
duration was 98.4±9.3 ms (range = 85–116 ms), and mean internote interval was 110±16.5 ms (range = 92–132
ms, Fig. 4A). Dominant frequency did not vary (645.9 Hz, Fig. 4B). Temporal call parameters from Brasília
were slightly longer when compared to calls from Pirenópolis, in number of notes (3–5), mean call duration
520±14.2 ms (range = 380–790 ms) (N=13) and duration of notes (mean note duration = 67.5±6.2 ms, ranging
75–85 ms). Mean dominant frequency was also slightly shorter in Pirenópolis 559.9±49.7 Hz (range =
516.8–602.9 Hz).

FIGURE 4. Call of Bokermannohyla sapiranga from Brasília Municipality, Distrito Federal (type locality): waveform (A);
audiospectogam (B); and powerspectrum (C). Recording by Rubens Hisanari Matsushita (December 11, 2004; 21:00h; air
temperature, 19.9ºC; air humidity, 75%).

A NEW SPECIES OF BOKERMANNOHYLA FROM CENTRAL BRAZIL Zootaxa 3527 © 2012 Magnolia Press · 37
 The call of B. sapiranga is similar to calls of other species in the B. pseudopseudis group (Table 5). It is more
similar to B. pseudopseudis, but readily separated from it by lacking the harmonic structure showed by B.
pseudopseudis (pulsionated in B. sapiranga), along with differences in spectral features, number and duration of
notes, and call length (Table 5). The maximum frequency in B. sapiranga is under 1.0 kHz, whereas in B.
pseudopseudis it may reach 2.4 kHz (Eterovick & Brandão, 2001; Lugli & Haddad, 2006b). Mean call duration of
B. pseudopseudis was 2.6s (Eterovick & Brandão 2001), while ranging 0.4–1.2s in B. sapiranga.

TABLE 5. Comparisons among calls of members of the Bokermannohyla pseudopseudis species group.

Number Large Short Intervals between Range

Species of notes notes notes two consecutive frequenc Harmonics Note characteristics
(range) range (ms) range (ms) notes (range, ms) y (kHz)
B. ibitiguara1 19–21 43.8–56.3 12.5–25 1502 1.1–2.6 not present pulsed
pure notes, without
B. oxente2 10–60 20.5–57.6 17.4–36.8 71.5–124.53 1.0–1.9 not present
visible pulses
pure notes, without
B. pseudopseudis3 5–10 64–84 56.4–71.8 0.4–2.4 up to five
visible pulses
most notes are
B. sapiranga4 5–7 85–116 92–132 0.5–0.7 not present
pulsed
Three main pure notes, without
B. saxicola5 25–25 22–441 602 0.9–4.5
harmonics visible pulses

Source for comparisons: 1Cardoso (1983); 2Lugli & Haddad (2006b); 3Eterovick & Brandão (2001); 4Present work (call
parameters based in the specimen recorded from Brasília municipality, type locality); 5Bokermann (1964) and Eterovick &
Brandão (2001).

Etymology. In Tupi indigenous language, sapiranga means red eye, an allusion to the reddish iris in the most
individuals of the species. The specific epipeth is also homage to Marco (Sapiranga) Freitas, for his continuous
efforts to popularize the Brazilian herpetofauna.
Distribution. Bokermannohyla sapiranga is known from the type locality (Brasília, Distrito Federal), and the
Municipalities of Cristalina, Novo Gama, Catalão, and Pirenópolis, State of Goiás (Fig. 5).
Ecology and natural history notes. Bokermannohyla sapiranga and B. pseudopseudis are species associated
with streams and rivulets with fast water flow in highlands of Central Brazil. Basking behavior was observed for
both species, in Brasília (B. sapiranga) (Fig. 6A) and Parque Nacional da Chapada dos Veadeiros (B.
pseudopseudis). However, they present some ecological differences. Bokermannohyla pseudopseudis uses streams
in open areas and rivulets with rocky beds, while B. sapiranga besides using similar habitats in some localities (e.g.
Pirenópolis, Cristalina, and Brasília Municipalities) also occurs in dense wet gallery forest, calling near small
waterfalls formed by roots and logs along rivulet (Brasília Municipality). In these habitats, tadpoles are found in
pools and backwaters with mud beds. We never found B. pseudopseudis using such kind of habitat. As other
species of the B. pseudopseudis group, B. sapiranga shows low ability to perch on the vegetation. Most individuals
were observed on rocks, soil, or dead logs. In Brasília Municipality, Ameerega flavopicta, Aplastodiscus aff.
perviridis, Barycholos ternetzi, Hypsiboas lundii, Hypsiboas goianus, Rhinella cerradensis, Rhinella rubescens,
and Scinax skaios can be found in the same habitats where B. sapiranga occurs.

Discussion

The taxonomy of the Bokermannohyla pseudopseudis species group is still complex and poorly understood.
Pombal Jr. and Caramaschi (1995) created the group to accommodate B. pseudopseudis and B. saxicola, mainly
because of their distinctively wider heads and developed forearms and prepollexes. These authors considered the
group related to the B. circumdata species group, but distinct from it because of color pattern (mainly by lacking
narrow transversal bars on thighs and flanks). Later, Eterovick and Brandão (2001) diagnosed the tadpoles of this
group by the presence of short, lateral irregular tooth rows and by having from six to eight posterior rows of

38 · Zootaxa 3527 © 2012 Magnolia Press BRANDÃO ET AL.

denticles, more than those of the B. circumdata group, which presents a maximum of two anterior and five
posterior rows of denticles. Faivovich et al. (2005) had access to tissues of only two species belonging to the B.
pseudopseudis species group. Due to this, the monophyly of the B. pseudopseudis group could not be completely
tested and some species are only tentatively assigned to the group based on external morphology, behavior, and
natural history.

FIGURE 5. Type locality (black star) and geographical records (black circles) of Bokermannohyla sapiranga and
Bokermannohyla pseudopseudis (dot center circles). Map by Renata Françoso-Brandão.

Nevertheless, this group is composed of morphologically and ecologically distinctive species, where B.
pseudopseudis, B. sapiranga, B. oxente, and B. saxicola are similar in size, degree of forearm hypertrophy, tadpole
morphology, and dorsal color pattern, and contrast with B. alvarengai, B. flavopicta, B. itapoty, and B. sagarana,
which are probably more related to each other. Lugli and Haddad (2006a) proposed the name B. alvarengai group
to accommodate these species, but Faivovich et al. (2009) and Leite et al. (2011, 2012) did not follow this

A NEW SPECIES OF BOKERMANNOHYLA FROM CENTRAL BRAZIL Zootaxa 3527 © 2012 Magnolia Press · 39
arrangement. Due to the number, conspicuousness, and distribution of the thigh and flank bars, B. ibitiguara was
originally designated as related to the B. circumdata species group (Cardoso 1983), and this was followed by
subsequent authors (Heyer 1985; Caramaschi & Feio 1990; Eterovick & Brandão 2001), but later transferred to the
B. pseudopseudis species group (Caramaschi et al. 2001).

FIGURE 6. Basking behavior of Bokermannohyla sapiranga juvenile in a pebble beach of Barão rivulet, Poço Azul, Brasília,
Distrito Federal in 23 October 1994, 16:00hs.

Pombal Jr. and Caramaschi (1995) reported differences on dorsum and iris colorations between populations of
B. pseudopseudis. Based on photos of live individuals from Brasília (CFBH 2248) and from Alto Paraíso de Goiás
(WCAB 47646), these authors noticed that the specimen from Brasília presented the dorsum brownish and the iris
orange, whereas the individual from Alto Paraíso de Goiás presented the dorsum gray-yellowish and the iris
yellowish and gray (Fig. 1). However, they did not suggest that these differences could indicate the existence of an
undescribed species. Brandão and Araújo (2001) reported differences in habitat use between populations from
Chapada dos Veadeiros (Alto Paraíso de Goiás) and Brasília Municipalities, where the population from Brasília
(i.e. Bokermannohyla sapiranga) was considered a gallery forest dependent species. Eterovick and Brandão (2001)
did not distinguish the tadpoles from Brasília and Chapada dos Veadeiros, composing a mixed series, and the
tadpole of the new species still lacks a formal description.
Our morphometric analysis was able to separate juveniles, males, and females of both species (Fig. 3). These
species are morphologically similar (Table 2), and although morphometric analyses do not necessarily provide
diagnostic characters, they can provide solid grounds to hypothesize independently evolving lineages (e.g. Garda et
al. 2010). Our analyses recovered head width (and all other measurements related to head proportions) as one of
the most important variables separating B. sapiranga and B. pseudopseudis, giving further support to this character
as a diagnostic difference between them. More individuals are needed to further improve such analyses, especially
in face of the rate of species descriptions in the group.
The Bokermannohyla pseudopseudis species group is rapidly growing, with the recognition of several new
species in the last few years. To fully understand species limits, biogeography, and evolution of these highland
treefrogs in Brazil, a deeper taxonomic and phylogenetic revision is needed. This is also the case for most of
highland frog species of Central Brazil.

40 · Zootaxa 3527 © 2012 Magnolia Press BRANDÃO ET AL.

Acknowledgments

We are in debt with Rubens Hisanari Matsushita for providing advertisement call recordings from CIAB, Brasília
Municipality, Distrito Federal and for fieldwork help. Hussam Zaher (MZUSP), Ariovaldo Giaretta (AAG-UFU),
Guarino Colli (CHUNB), Célio Haddad (CFBH), and Rogério Bastos (ZUFG) loaned specimens from the
collections under their care. We thank Santiago Castroviejo-Fisher and two anonymous reviewers for suggestions
that greatly improved the manuscript. We thank Rogério Pereira Bastos and Alessandro Ribeiro de Morais for
helping in the call analysis and interpretation. We are also grateful to José Alexandre F. Diniz Filho, Núbia Carla S.
Marques, and Fausto Nomura for helping with statistical analysis. We thank Julian Faivovich for checking several
individuals at CFBH with us. Daniel Rios helped us during fieldwork in the Parque Nacional da Chapada dos
Veadeiros. Carolina Mello (MZUSP) and Marcela Brasil (CHUNB) greatly facilitated our access to data for species
comparisons, and Renata Françoso constructed the distribution map. Collecting permits were issued by Instituto
Chico Mendes de Conservação da Biodiversidade (License IBAMA/ICMBio/SISBIO—17283-1).

Literature cited

Bokermann, W.C.A. (1956) Sôbre uma espécie de Hyla do Estado de Minas Gerais, Brasil. Papéis Avulsos de Zoologia, 12,
357–362.
Bokermann, W.C.A. (1964) Dos nuevas especies de Hyla de Minas Gerais y notas sobre Hyla alvarengai Bok. (Amphibia,
Salientia, Hylidae). Neotropica, 10, 67–76.
Bookstein, F.L. (1996) Biometrics, biomathematics and the morphometric synthesis. Bulletin of Mathematical Biology, 58,
313–365.
Brandão, R.A. & Araujo, A.F.B. (2001) A herpetofauna associada às matas de galeria do Distrito Federal. In: Ribeiro, J.F.,
Fonseca, C.E.L., & Sousa-Silva, J.C. (Eds). Cerrado: Caracterização e Recuperação de Matas de Galeria. Embrapa
Cerrados, Planaltina, pp. 560–604.
Brandão, R.A., Sebben, A. & Zerbini, G.J. (2005) A Herpetofauna da APA de Cafuringa. In: Netto, P.B.; Mecenas, V.V. &
Cardoso, E.S. (Eds). APA de Cafuringa: A Última Fronteira Natural do DF. Secretaria de Meio Ambiente do Distrito
Federal, Brasília, pp. 241–248.
Burnham, K.P. & Anderson, D.R. (2002) Model selection and multimodel inference: a practical information-theoretical
approach, Springer-Verlag New York, 488 pp.
Burnham, K.P. & Anderson, D.R. (2004) Multimodel inference: understanding AIC and BIC in model selection. Sociological
Methods & Research, 33, 261–304.
Burnaby, T.P. (1966) Growth-invariant discriminant functions and generalized distances. Biometrics, 22, 96–110.
Caramaschi, U. & Feio, R.N. (1990) A new species of Hyla (Anura, Hylidae) from southern Minas Gerais, Brazil. Copeia,
1990, 542–546.
Caramaschi, U., Napoli, M.F. & Bernardes, A.T. (2001) Nova espécie do grupo de Hyla circumdata (Cope, 1870) do Estado de
Minas Gerais, Brasil (Amphibia, Anura, Hylidae). Boletim do Museu Nacional, Nova Série, Zoologia, 457, 1–11.
Cardoso, A.J. (1983) Descrição e biologia de uma nova espécie de Hyla Laurenti, 1768 (Amphibia, Anura, Hylidae). Iheringia,
Série Zoologia, 62, 37–45.
Colli, G.R., Mesquita, D.O., Rodrigues, P.V.V. & Kitayama, K. (2003) Ecology of the gecko Gymnodactylus geckoides amarali
in a neotropical savanna. Journal of Herpetology, 37, 694–706.
Eterovick, P.C. & Brandão, R.A. (2001) A description of the tadpoles and advertisement calls of members of the Hyla
pseudopseudis group. Journal of Herpetology, 35, 442–450.
Faivovich, J., Haddad, C.F.B., Garcia, P.C.A., Frost, D.R., Campbell, J.A. & Wheeler, W.C. (2005) Systematic review of the
frog family Hylidae, with special reference to Hylinae: phylogenetic analysis and taxonomic revision. Bulletin of the
American Museum of Natural History, 294, 1–240.
Faivovich, J., Lugli, L., Lourenço, A.C.C. & Haddad, C.F.B. (2009) A new species of the Bokermannohyla martinsi group from
Central Bahia, Brazil with comments on Bokermannohyla (Anura: Hylidae). Herpetologica, 65, 303–310.
Garda, A.A., Santana, D.J. & São-Pedro, V.A. (2010) Taxonomic characterization of Paradoxical frogs (Anura, Hylidae,
Pseudae): geographic distribution, external morphology, and morphometry. Zootaxa, 2666, 1–28.
Gridi-Papp, M. (Ed.). (2003–2007) SoundRuler: Acoustic Analysis for Research and Teaching. http://soundruler.sourceforge.net.
Guimarães, L.D., Lima, L.P., Juliano, R. F. & Bastos, R.P. (2001) Vocalizações de espécies de anuros (Amphibia) no Brasil
Central. Boletim do Museu Nacional, Nova Série, Zoologia, 474, 1–14.
Heyer, W.R. (1985) New species of frogs from Boracéia, São Paulo, Brazil. Proceedings of the Biological Society of
Washington, 98, 657–671.
Heyer, W.R., Rand, A.S., Cruz, C.A.G., Peixoto, O.L. & Nelson, C.E. (1990) Frogs of Boracéia. Arquivos de Zoologia, 31,
231–410.

A NEW SPECIES OF BOKERMANNOHYLA FROM CENTRAL BRAZIL Zootaxa 3527 © 2012 Magnolia Press · 41
 Johnson, J.B. & Omland, K.S. (2004) Model selection in ecology and evolution. Trends in Ecology and Evolution, 19,
101–108.
Leite, F.S.F. & Eterovick, P.C. (2010) Description of the tadpole of Bokermannohyla martinsi (Anura: Hylidae), morphological
and ecological comparison with related Bokermannohyla tadpoles. Journal of Herpetology, 44, 431–440.
Leite, F.S.F, Pezzuti, T.L. & Drummond, L.O (2011) A new species of Bokermannohyla from the Espinhaço Range, State of
Minas Gerais, Southeastern Brazil. Herpetologica, 67, 440–448.
Leite, F.S.F., Pezzuti, T.L. & Garcia, P.C.A. (2012) A new species of the Bokermannohyla pseudopseudis group from the
Espinhaço Range, Central Bahia, Brazil (Anura:Hylidae). Herpetologica, 68, 401–409.
Lugli, L. & Haddad, C.F.B. (2006a) A new species of Bokermannohyla (Anura, Hylidae) from Central Bahia, Brazil. Journal of
Herpetology, 40, 7–15.
Lugli, L. & Haddad, C.F.B. (2006b) A new species of the Bokermannohyla pseudopseudis group from Central Bahia, Brazil
(Amphibia, Hylidae). Herpetologica, 62, 453–465.
Miranda-Ribeiro, A. (1937) Alguns batrachios novos das collecções do Museu Nacional. O Campo, 8, 66–74.
Napoli, M.F. & Caramaschi, U. (1998) Duas novas espécies de Hyla Laurenti, 1768 do Brasil Central afins de H. tritaeniata
Bokermann, 1965 (Amphibia, Anura, Hylidae). Boletim do Museu Nacional, Nova Série, Zoologia, 391, 1–12.
Pombal Jr, J.P. & Caramaschi, U. (1995) Posição taxonômica de Hyla pseudopseudis Miranda-Ribeiro, 1937 e Hyla saxicola
Bokermann, 1964 (Anura, Hylidae). Boletim do Museu Nacional, Nova Série, Zoologia, 363, 1–7.
Rangel, T.F.L.V.B, Diniz-Filho, J.A.F & Bini, L.M. (2010) SAM: a comprehensive application for spatial analysis in
macroecology. Ecography, 33, 46–50.
Rohlf, F.J. & Bookstein, F.L. (1987) A comment on shearing as a method for size correction. Systematic Zoology, 36, 356–367.
São Pedro, V.A., Drummond, L.O., Costa, H.C., Fernandes, V.D. & Feio, R.N. (2008) Predação de embriões por girinos de
Bokermannohyla alvarengai (Anura, Hylidae) em riacho temporário na Serra de Ouro Branco, Minas Gerais, Brasil.
Boletim do Museu de Biologia Mello Leitão (N.S.) 24, 111–118.
Savage, J.M. & Heyer, W.R. (1967) Variation and distribution in the tree-frog genus Phyllomedusa in Costa Rica, Central
America. Beiträge zur Neotropischen Fauna, 5, 111–131.
Sazima, I. & Bokermann, W.C.A. (1977) Anfíbios da Serra do Cipó, Minas Gerais, Brasil, 3: Observações sobre a biologia de
Hyla alvarengai Bok. (Anura: Hylidae). Revista Brasileira de Biologia, 57, 413–417.
Tattersall, G.J., Eterovick, P.C. & Andrade, D.V. (2006) Tribute to R. G. Boutilier: Skin colour and body temperature changes in
basking Bokermannohyla alvarengai (Bokermann 1965). The Journal of Experimental Biology, 209, 1185–1196.

Specimens examined

Bokermannohyla alvarengai: BRAZIL: MINAS GERAIS: Santana do Riacho: CFBH 5629; Grão Mogol: CFBH 10226; Cardeal
Mota: CFBH 16682; Jaboticatubas: CHUNB 57654, CHUNB 57656, CHUNB 57659; ASUnB 676.
Bokermannohyla ibitiguara: BRAZIL: MINAS GERAIS: Alpinópolis: CFBH 0020; São Roque de Minas: CFBH 8010–8012,
CHUNB 51425; Furnas: CFBH 17318, CFBH 17321, CFBH 17323, CFBH 17325, CFBH 17344, Capitólio; AAG-UFU
4726. No locality data: AAG-UFU 2142.
Bokermannohyla itapoty: BRAZIL: BAHIA: Lençóis: CFBH 5643 (holotype), CFBH 5644–5645, 5648–5649 (paratypes);
Mucugê: CFBH 5646 (paratype); Palmeiras: CFBH 5650–5651 (paratypes); Rio de Contas: ASUnB 949.
Bokermannohyla oxente: BRAZIL: BAHIA: Lençóis: CFBH 5633 (holotype), CFBH 5636, 5638, 5642 (paratypes); Mucugê:
CFBH 5635 (paratype); Palmeiras: CFBH 5634, 5639–5641 (paratypes).
Bokermannohyla pseudopseudis: BRAZIL: GOIÁS: Alto Paraíso: CFBH 6800–6802, CHUNB 14065, 14382, 14385, 14389,
17526–17532, 28957, 32637, 32619, 38795, 42522–42523, 47516, 49511, 43650, 51383, 51388, 58873–58879,
58789–58794, 59085–59093.
Bokermannohyla saxicola: BRAZIL: MINAS GERAIS: Santana do Riacho: CFBH 799; CFBH 5630–5632; Conceição do Mato
Dentro: CFBH 1731217313; ASUnB 949.

42 · Zootaxa 3527 © 2012 Magnolia Press BRANDÃO ET AL.

View publication stats