Dissertation

DISSERTATION

Titel der Dissertation Multimodal signals in anurans: The role of acoustic and visual signals in the communication of foot-flagging frogs

(Multimodale Signale bei Anuren: Die Rolle von akustischen und visuellen Signalen in der Kommunikation von Winkerfröschen)

Verfasserin Mag. Doris Preininger

angestrebter akademischer Grad Doktorin der Naturwissenschaften (Dr.rer.nat.)

Wien, 5. Oktober 2012

Studienkennzahl lt. Studienblatt: A 091 439 Dissertationsgebiet lt. Studienblatt: Dr.-Studium der Naturwissenschaften, Zoologie Betreuerin / Betreuer: Ao. Univ.-Prof. Dr. Walter Hödl

Acknowledgements

There are many people who contributed to this thesis and I am grateful for the help, cooperation and collaboration of each one of them, but it all started with the idea and enthusi- asm of one person, who saw 15 years ago, on one of his many excursions, something very simple and extremely striking – a frog that foot flags. The frog was from the genus Staurois, the attentive observer was Walter Hödl, the latter I deeply thank for his guidance of every aspect of my work and for putting me in charge of his research idea without thinking twice.

I particularly thank Markus Böckle for so many things that it would fill several pages. We started out as colleagues during our combined master studies became close friends and to this day he supports every scientific and not so scientific decision of mine. I am extremely grateful for the help, patience and friendship of Marc Sztatecsny and Iris Starnberger, they have not only enriched this project but also my life.

I thank my collaborators Ulmar Grafe from Brunei and K.V. Gururaja “Guru”, S.P. Vijaya- kumar “Vijay” from India for their endless help and support and for making my field trips an exceptional experience. I thank Dagmar Schratter, Anton Weissenbacher, Regina Riegler,

Evi Karell and several zoo-keepers for devoting their help and time to a new breeding and research project in the Vienna Zoo. I am especially grateful for the devotion of Thomas

Wampula to this project, for his assistance in the field, for his friendship and for continuously ensuring me that everything is going to work out.

I thank my parents for their encouragement when I excitedly told them: I am going to study frogs that wave their legs to communicate in tropical rainforests at the other end of the world. And although I am not sure they were as excited as I was, mainly about the “other end of the world”-part, they have supported me on every step of the way. I thank my husband for believing in me, for listening attentively for hours and hours to stories about frogs and for sometimes reminding me to turn off my computer.

Since so many people have contributed to my thesis I believe it is only fair to address most sections of my introduction and conclusion as our work rather than my work!

Table of content

Summary ...... 1 Zusammenfassung ……………………………………………………………………...... 2

Preface …………………………………………………………………………...…….…..... 3

Synopsis of the publications …...... 9 Chapter 1. Multimodal communication in a noisy environment: A case study of the Bornean rock frog Staurois parvus …………………………………...... 13 2. Multimodal signaling in the Small Torrent Frog (Micrixalus saxicola) in a complex acoustic environment …………………………………………………… 23 3. Divergent receiver responses to components of multimodal signals in foot-flagging frog species offer clues to visual signal evolution …………...... 47 4. The conservation breeding of two foot-flagging frog species from Borneo, Staurois parvus and Staurois guttatus ……………………………...... 73

Concluding Discussion ………………………………………………………...……...... 85

References ………………………………………………………………………………...... 91

Appendix A. Micrixalus saxicola a foot-flagging frog from India: Acoustic and visual signaling behavior during male-male agonistic interactions……………...... 97 B. Females do have a say in the matter: Female and male vocalizations and laryngeal structures in the foot-flagging frog species Staurois guttatus ...... 103 C. Don’t get the blues: conspicuous nuptial colouration of male moor frogs (Rana arvalis) supports visual mate recognition during scramble competition in large breeding aggregations ……………...... …...... 113

Curriculum Vitae …………………………………………………………………………...... 121

Summary

The predominant communication mode of anuran amphibians are vocalizations, however intraspecific communication may involve multimodal (acoustic and visual) cues or signals in many more species than previously thought. Visual signals may act as an additional or a complementary mode of communication in noisy environments. Foot flagging as a striking form of visual signaling behavior has evolved in at least in 16 species from 5 different families mainly living along fast-flowing streams, which generate continuous broadband background noise. To better understand the function of foot flagging as multimodal signal component, we studied three Asian species from two families and performed cue-isolation experiments in the field.

The Bornean species Staurois parvus and S. guttatus avoid acoustic interference of ambient stream noise by using call frequencies less masked by the background and utilize ac- companying visual signals to announce the readiness to defend calling sites, interestingly this behavior can be observed already in juveniles bred in the Vienna Zoo.

Micrixalus saxicola from the Western Ghats (India) occurs along less noisy streams and acoustic signals are rather masked by chorus noise from conspecifics than by abiotic noise.

Males use a variety of visual signals including foot-flagging and tapping during agonistic behavior. Receiver responses from acoustic- vs. multimodal playback presentations provide evidence that the vocal sac acts as an additional visual cue. Comparisons of signal response to acoustic, visual and multimodal stimuli and analysis of signal brightness of foot webbings in M. saxicola and S. parvus further highlight differences in the magnitude and significance of signaling behavior in the respective species. Together these investigations allow us to draw conclusions on signal efficacy and function and help to better understand the evolution of multimodal communication in anuran.

Zusammenfassung

Frösche kommunizieren vorwiegend mit akustischen Signalen, allerdings verwenden ei- nige Arten multimodale (akustische und visuelle) Signale während der intra-spezifischen

Kommunikation. Visuelle Signale können in geräuschvoller Umgebung eine zusätzliche oder ergänzende Kommunikationsweise ermöglichen. Das Beinwinken (foot flagging) eine auffäl- lige Form von visuellem Signalverhalten hat sich in mindestens 16 Arten aus 5 verschiede- nen Familien entwickelt. Die Vielzahl dieser Arten lebt und reproduziert entlang schnell- fliessender Bergbäche, die ein kontinuierliches Hintergrundrauschen generieren. Um die

Funktion des Beinwinkens als multimodale Signalkomponente besser zu verstehen, haben wir drei asiatische Arten aus zwei Familien untersucht und die akustischen und visuellen

Signale in Feldexperimenten getestet.

Die auf Borneo heimischen Arten Staurois parvus und S. guttatus verwenden hoch- frequente Rufe und reduzieren dadurch akustische Interferenzen vom umgebenden Bach- rauschen und benutzen zusätzliche visuelle Signale, um die Verteidigungsbereitschaft eines

Rufstandort anzuzeigen. Interessanter Weise kann dieses Verhalten bereits bei juvenilen

Tieren, die im Wiener Tiergarten Schönbrunn gezüchtet wurden, beobachtet werden.

Micrixalus saxicola aus den Western Ghats (Indien) bewohnt weniger geräuschvolle Bäche, welche die akustische Kommunikation nicht einschränken. Die akustischen Signale können jedoch von arteigenen Rufchören maskiert werden. Die Männchen verwenden visuelle Sig- nale, wie Beinwinken und Beinheben während agonistischen Interaktionen. Reaktionen auf akustische oder multimodale Signalstimuli zeigen, dass die Schallblase als zusätzliches vi- suelles Hinweissignal fungiert. Gegenüberstellungen von Reaktionen auf akustische, visuelle und multimodale Stimuli und Analysen der Signalhelligkeit der Schwimmhäute von M. saxicola und S. parvus, verdeutlichen die Unterschiede in Ausmaß und Stellenwert des Signalver- haltens der jeweiligen Art. Zusammengefasst ermöglichen diese Untersuchungen Schluss- folgerungen über die Funktion und Wirksamkeit der visuellen Signale in der Kommunikation von Fröschen. 3

Preface

In animal communication a signal from one organism stimulates the sensory system and influences a behavioral change of a perceiver, and generally signaler and perceiver benefit from this exchange (Rendall et al. 2009; Ruxton & Schaefer 2011). On the signaler's side, signals (traits evolved for communication) are produced and correlate with an attribute of the signaler, whereas on the perceiver's side signals have to be correctly received, processed and decided upon. The production and reception of a signal is additionally influenced by the environment through which a signal travels (Endler 1992; Endler 1993b). A change in one process (production, transmission through certain environmental conditions or reception) will eventually alter the other processes and lead to evolutionary changes in signaling (Endler

2000).

To understand selection pressures acting on signals and to further explain signal evolution, we ask two elementary questions: Why & How? Why is the receiver influenced by the signal (signal content)? How efficiently is a signal transmitted through the environment and how effectively does a signal influence the behavior of a receiver (signal efficacy). In addition to focusing on the causal and functional aspect of the signal or its communicative function we apply the same “Why & How” questions to the actual signaling behavior displayed by the signaler with regard to a current behavior or the behavioral development over time

(Tinbergen 1951) and ask: How and why is the behavior performed? How and why did the behavior develop?

Investigating hypotheses on the basis of signaling behavior and communicative signal function will eventually broaden our understanding of certain animal communication systems.

However, answering “Why & How” questions is not an easy task as animal signals consist of multiple components in a single sensory modality (e.g. acoustic, visual or chemical) or in multiple sensory modalities (multimodal communication) with components being presented either simultaneously or in succession (Partan & Marler 1999; Partan & Marler 2005). 4

During the course of my study „Multimodal signals in anurans: The role of acoustic and visual signals in the communication of foot-flagging frogs” I explored anuran signaling behavior from different angels by means of the above proposed questions. I examined potential environmental constraints and investigated acoustic, visual and multimodal signaling behavior within and across species and draw conclusions on its development.

Anuran amphibians are excellent model systems to study communications systems.

The predominant communication mode of anurans is vocalization. The male advertisement call attracts conspecific females and signals the readiness to defend territories against conspecific males (Duellman & Trueb 1986); hence calling behavior plays a vital role in reproductive success and is essential for sexual selection. In numerous species spectral and temporal features of calls and their function as static or dynamic signal properties, shaped by preferences of receivers over evolutionary time, were investigated during the last decades

(Gerhardt & Huber 2002). Several studies demonstrated that certain call characteristics correlate with interspecific (e.g. Ryan 1988) and intraspecific (e.g. Robertson 1990) body size and mass, signal the species identity (e.g. Hödl 1977), to a certain amount match the tuning of the auditory system of the receiver (e.g. Gerhardt & Schwartz 2001), regulate male spacing (e.g. Brenowitz 1989) and increase the male’s attractiveness to females (e.g. Ryan &

Keddy-Hector 1992). For a call to fulfill the proposed functions the acoustic signal has to be detected and discriminated in the environment by receivers. Biotic environmental background noise (chorusing conspecifics and heterospecifics) may mask calls and hamper accurate detection and discrimination. However frogs generally evolved calls or signal strategies to enhance transmission in their respective acoustic environment. To improve the signaling ef- fectiveness anuran species temporally organize calling periods (Klump & Gerhardt 1992), adjust their calls to random intercall intervals (Zelick & Narins 1985) or inhibit calling heterospecifics (Schwartz & Wells 1983). In dense breeding aggregations concurrently chorusing conspecifics can achieve release from masking interference by spatial separation from the biotic sound source (Bee 2008). Other constraints to signal detection and discrimination 5

are abiotic noise sources. Continuous background noise of torrential streams and waterfalls could impair acoustic communication of stream associated anurans. High-frequency calls opposed to the mainly low-frequency dominated stream noise enhance the signal-to-noise ratio and were suggested to have evolved as an adaptive strategy in some species (Feng &

Narins 2008; Arch et al. 2008; Boeckle et al. 2009). Opposed to these findings our understanding of similar aspects in visual signals in anurans is lagging far behind.

In the widely used textbook “Biology of Amphibians” published in 1986 by W. Duellmann and L. Trueb the authors describe visual signals for newts and salamanders but state that in anurans visual displays “seem to be unimportant to preamplectic courtship”. Up-right pos- tures and throat displays occurring during territorial defense and jumps as well as toe movements during courtship are mentioned for few species. A larger section of the book discusses visual cues used as antipredator mechanisms such as aposematic colorations and the so called Unkenreflex. In the last 26 years scientists provided numerous new findings in several anuran species with regard to visual displays used as signals during conspecific communication. Brightly colored vocal sacs, which are inevitably moved while calling, received much attention. The vocal sac recycles air while a male is calling. The highly elastic structure minimizes the loss of sound energy, increases the call rate and distributes sound waves omnidirectionally (Bucher et al. 1982; Rand & Dudley 1993; Pauly et al. 2006). The visual component of the vocal sac increases detection through movement and coloration (Rosenthal et al. 2004; Taylor et al. 2008) and is part of the advertisement signal in some species, thereby enhancing the call attractiveness to females and aggression during territorial male-male interactions. Females of the túngara frog (Physalaemus pustulosus) prefer advertisement calls in addition to the visual cue of a pulsating vocal sac over the call alone under low sound pressure levels. However, when the visual stimulus is presented with a less attractive slow call rate, females rather choose the attractive unimodal call, which emphasizes that vocalizations are necessary for mate attraction. The visual cue of the vocal sac was suggested to mainly facilitate detection and localization in noisy choruses (Rosenthal et al. 2004; Taylor et

al. 2008, 2011). First evidence of the function of the visual cue of a pulsating vocal sac during male territorial defense comes from studies in a dart-poison frog (Allobates femoralis). No stimulus alone (call or pulsating vocal sac) was able to elicit territorial aggression in the opponent male, only temporally overlapping dynamic bimodal cues evoked fighting behavior

(Narins et al. 2003; Narins et al. 2005). Hence, conspecific vocalizations in A. femoralis elicit phonotactic response and antiphonal calling but are not sufficient to evoke physical aggression. In an East-African stream frog, Phrynobatrachus kreffti, agonistic male-male interactions are dominated by inflations of a bright yellow vocal sac without vocalization

(Hirschmann & Hödl 2006), suggesting that the visual component alone is a functional signal in the communication of this species. These examples highlight the differences in the importance of uni- and bimodal signals and demonstrate differential emphasis of signal modalities in the respective receivers and species. Disentangling influences of isolated components of multimodal signals on receivers, especially those linked to the same organ, remain a difficult task as an adaption in one modality will most likely also affect the other modality.

Visual displays presented independently of acoustic signals were observed in several anuran species. Limb movements are used in addition to acoustic signals during courtship and male-male interaction (reviewed in Hödl & Amézquita 2001; Hartmann et al. 2005). The most conspicuous signaling behavior to the human observer is foot-flagging behavior, a display during which the hind leg is raised, the toes are spread and saliently colored interdigital webbings are displayed. Foot flagging was reported in 16 anuran species from five different families (Hödl & Amézquita 2001; Vasudevan 2001; Hartmann et al. 2005; Krishna & Krishna

2006; Grafe & Wanger 2007). The signaling behavior is mainly known from stream dwelling species and is displayed predominantly during male-male interaction or territorial encounters.

In previous studies of two species of the Bornean genus Staurois, foot-flagging behavior is suggested to function as an additional or alternative mode of communication in noisy stream environments. In S. guttatus and S. latopalmatus vocalizations are suggested to alert receiv- 7

ers and direct their attention to the subsequent visual signal (Grafe & Wanger 2007;

Preininger et al. 2009).

Foot-flagging displays as a visual communicative signal certainly did not develop from calling behavior and do not form fixed-composite signals (sensu Partan and Marler 2005) with vocalizations. The functional separation of the acoustic and visual signal allowed us to study influences of signal components on receivers without a determining linkage of signal modality and/or adaptation.

To understand multimodal communication in foot-flagging frogs, we investigated several

“Why & How”- questions related to signaling behavior and selection pressures. Due to the advances in the conceptional framework of multimodal signals (Hebets & Papaj 2005; Partan

& Marler 2005; Otovic & Partan 2009) we were able to explore hypotheses of signal efficacy, inter-signal interaction and, to a minor extent, signal content. Overall we examined and compared the function of multimodal signal components as well as possible environmental constraints that potentially influenced signal design of two species from two anuran families

(Ranidae and Micrixalidae). In both species, the Bornean Rock Frog (Staurois parvus) and the Small Torrent Frog (Micrixalus saxicola) from the Western Ghats of India, males announce the readiness to defend signaling sites in fast-flowing streams with acoustic and visual displays.

Vocalizations in stream-associated species may be masked by ambient environmental noise. We thus investigated how efficient the acoustic signal is transmitted through the environment by analyzing its signal-to-noise ratio at receiver distance (S. parvus – Chapter 1; M. saxicola – Chapter 2). To determine why & how foot-flagging behavior is performed we examined behavioral sequences and acoustic playback responses (S. parvus – Chapter 1; M. saxicola – Chapter 2, Appendix A). To investigate whether call and foot-flagging signals form a functional unit we analyzed the timing intervals between acoustic and visual signals (S. parvus – Chapter 1; M. saxicola – Appendix A). To gain insight why the receiver is influenced by a signal, the influence of independent and combined signal components on receivers was 8

compared within and across species responses. (S. parvus and M. saxicola – Chapter 3). A further cue on why foot-flagging displays are performed derives from observations that juveniles already perform the signal (S. parvus – Chapter 4).

Many studies have focused on signal function and transmission in acoustic communication but only little is known of parallel issues in visual or multimodal communication. In order to discuss the evolution of foot flagging it is fundamental to understand the efficacy of signals, signaling behavior and influence on receivers. The across-species studies on communicative signal function and signaling behavior will provide an insight in the convergent development of foot flagging in anuran species. 9

Synopsis of publications

Chapter 1. Multimodal communication in a noisy environment: A case study of the Bornean rock frog Staurois parvus. Grafe T.U, Preininger D, Sztatecsny M, Kasah R, Dehling M, Proksch S and W Hödl. 2012. Publication released: PlosOne 7(5): e37965. doi:10.1371/journal.pone.0037965 Contribution: Conceived and designed the experiments: TUG, DP, MS. Performed the experiments: TUG, DP, MS, RK, MD, SP. Analyzed the data: TUG, DP, MS, RK, MD, SP. Con- tributed materials/analysis tools: TUG, DP, MS, WH. Wrote the paper: TUG, DP, MS, WH.

Chapter 2. Multimodal signaling in the Small Torrent Frog (Micrixalus saxicola) in a complex acoustic environment. Preininger D, Boeckle M, Freudmann A, Starnberger I, Sztatecsny M, and W Hödl. 2012. Manuscript accepted: 20-08-2012 Behavioral Ecology and Sociobiology Contribution: Conceived and designed the experiments: DP, MB, WH. Performed the experiments: DP, MB. Analyzed the data: DP, MB, AF. Contributed materials/analysis tools: DP, MS, WH. Wrote the paper: DP, IS, MS, WH.

Chapter 3. Divergent receiver responses to components of multimodal signals in foot- flagging frog species offer clues to visual signal evolution Preininger D, Boeckle M, Sztatecsny M, and W Hödl. Manuscript submitted: 24-09-2012 PlosOne Contribution: Conceived and designed the experiments: DP, MB, WH. Performed the experiments: DP, MB. Analyzed the data: DP, MB, AF. Contributed materials/analysis tools: DP, MS, WH. Wrote the paper: DP, IS, MS, WH.

Chapter 4. The conservation breeding of two foot-flagging frog species from Borneo, Staurois parvus and Staurois guttatus. Preininger D, Weissenbacher A, Wampula T and W Hödl. 2012. Publication released: Amphibian and Reptile Conservation 5(3):45-56(e51) Contribution: Conceived and designed the experiments: DP, AW, TW, WH. Performed the experiments: DP, TW. Analyzed the data: DP. Contributed materials/analysis tools: DP, AW, TW, WH. Wrote the paper: DP, WH. 10 11

Chapter 1

MULITMODAL COMMUNICATION IN A NOISY ENVIRONMENT: A CASE STUDY OF THE BORNEAN ROCK FROG STAUROIS PARVUS

ULMAR T. GRAFE, DORIS PREININGER, MARC SZTATECSNY, ROSLI KASAH,

MAXIMILIAN DEHLING, SEBASTIAN PROKSCH AND WALTER HÖDL

PlosOne 7(5): e37965. doi:10.1371/journal.pone.0037965 Received 16 March 2012; Accepted 29 April 2012; Published 24 May 2012

CHAPTER 1 13

Multimodal Communication in a Noisy Environment: A Case Study of the Bornean Rock Frog Staurois parvus

T. Ulmar Grafe1*, Doris Preininger2, Marc Sztatecsny2, Rosli Kasah1, J. Maximilian Dehling3, Sebastian Proksch3, Walter Ho¨ dl2 1 Department of Biology, Universiti Brunei Darussalam, Gadong, Brunei Darussalam, 2 Department of Evolutionary Biology, University of Vienna, Vienna, Austria, 3 Department of Animal Ecology and Tropical Biology, University of Wu¨rzburg, Theodor-Boveri-Institut, Biozentrum, Wu¨rzburg, Germany

Abstract High background noise is an impediment to signal detection and perception. We report the use of multiple solutions to improve signal perception in the acoustic and visual modality by the Bornean rock frog, Staurois parvus. We discovered that vocal communication was not impaired by continuous abiotic background noise characterised by fast-flowing water. Males modified amplitude, pitch, repetition rate and duration of notes within their advertisement call. The difference in sound pressure between advertisement calls and background noise at the call dominant frequency of 5578 Hz was 8 dB, a difference sufficient for receiver detection. In addition, males used several visual signals to communicate with conspecifics with foot flagging and foot flashing being the most common and conspicuous visual displays, followed by arm waving, upright posture, crouching, and an open-mouth display. We used acoustic playback experiments to test the efficacy-based alerting signal hypothesis of multimodal communication. In support of the alerting hypothesis, we found that acoustic signals and foot flagging are functionally linked with advertisement calling preceding foot flagging. We conclude that S. parvus has solved the problem of continuous broadband low-frequency noise by both modifying its advertisement call in multiple ways and by using numerous visual signals. This is the first example of a frog using multiple acoustic and visual solutions to communicate in an environment characterised by continuous noise.

Citation: Grafe TU, Preininger D, Sztatecsny M, Kasah R, Dehling JM, et al. (2012) Multimodal Communication in a Noisy Environment: A Case Study of the Bornean Rock Frog Staurois parvus. PLoS ONE 7(5): e37965. doi:10.1371/journal.pone.0037965 Editor: Jochen Zeil, The Australian National University, Australia Received March 16, 2012; Accepted April 29, 2012; Published May 24, 2012 Copyright: ß 2012 Grafe et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This study was supported by grants from the Austrian Science Fund (FWF): P22069-B17 (www.fwf.ac.at/en/index.asp), the German Academic Exchange Programme (DAAD) (www.daad.org/) and a research grant from the Universiti Brunei Darussalam (RG 58) (http://www.ubd.edu.bn/). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected]

Introduction interference from conspecifics and heterospecifics, individuals alter spectral or temporal call characteristics to avoid overlap In any message, signals need to be successfully processed [11],[17],[18], and use spatial release of masking chorus noise through either single or multiple channels to effectively convey for species recognition [19]. Another strategy to reduce masking is information from senders to receivers [1]. Clear reception is a to utilize multiple signal modalities, where each modality increases minimum requirement for a successful communication system [2]. efficacy under specific conditions [14],[15],[20–22]. Visual signals Signal detectability depends on signal design, conditions of the may act as a complementary mode of communication in noisy environment, and the receiver’s sensory system [2], [3]. Additional habitats. For example, foot-flagging displays are conspicuous visual sensory stimulation in the environment can cause information to signals observed in tropical anuran species inhabiting fast flowing be lost. In the case of acoustic communication, noise and streams [15],[17],[22–25] or areas with heavy rains and noise transmission properties of the environment may shape the spectral produced by conspecifics [26]. and temporal structure of signals [4–6] as well as emphasize the Several non-mutually exclusive hypotheses have been proposed role of signal efficacy in the evolution of animal signals [7]. to explain the function of multimodal signals. Signals could be Senders can increase signal efficacy by either avoiding areas of redundant and act independently as a back-up for increased high noise [8], overriding environmental noise [9], adjusting their accuracy of information transfer [27] or could contain multiple signal timing [10–12] or by using frequencies less masked by messages with each signal conveying a different message [7]. In background noise [13],[14]. Furthermore, signallers may use contrast, the efficacy-based hypotheses address the factors affecting additional modes of communication to facilitate transmission the transmission and reception of multimodal signals, with the [15],[16]. efficacy-based alerting signal hypothesis suggesting that one signal Anurans are excellent model systems to investigate acoustic alters the response to a subsequent signal [28]. In this study, we communication during high levels of background noise and the advantages gained by the concomitant use of visual signals. Male test the efficacy-based alerting signal hypothesis to explain the advertisement calls are the principal mediators of sexual behaviour function of multimodal signals. For example, if signals of two that attract females and serve to announce the readiness to defend modalities are emitted sequentially, the hypothesis predicts, among calling sites and territories. To reduce certain patterns of acoustic others, that the signal in one modality consistently precedes the

PLoS ONE | www.plosone.org 1 May 2012 | Volume 7 | Issue 5 | e37965 14 CHAPTER 1

Mulitmodal Communication in a Bornean Frog signal of the other modality. Thus, a signal in one modality can function to alert the receiver to a subsequent signal in a different modality that might be more informative or, as is the case of visual signals, needs the receiver to look into the direction of the signaller. For example, in sticklebacks, male olfactory cues act as long distance messages that alert females to the following visual cue [29] while in the Bornean ranid frog Staurois guttatus vocalizations alert receivers to the subsequent foot flag [25]. In the present study, our aims are to (1) examine how the Bornean rock frog Staurois parvus communicates in noisy environments (2) characterize foot-flagging behaviour and other visual displays (3) record the key characteristics of their vocalizations, (4) determine the signal-to-noise ratio at a fast flowing stream in which males call, and (5) use acoustic playback experiments to test the efficacy-based alerting signal hypothesis of multimodal signalling. Figure 1. Male Staurois parvus foot-flagging in close proximity Materials and Methods to a rival male. doi:10.1371/journal.pone.0037965.g001 Ethics statement This was an observational study of free-ranging animals. The observations of signalling behaviour were recorded of male tactile experimental protocol adhered to the Animal Behaviour Society behaviours and female vocal and visual signalling. guidelines for the use of animals in research and was approved by If not stated otherwise, means and SD are given as descriptive the Universiti Brunei Darussalam Research Committee (UBD/ statistics and analyses were run using BIAS (v.8.2; epsilon-Verlag PNC2/2/RG/1(58)). GbR 1989–2006). All tests are two-tailed.

Study site and species Acoustic recordings th th We studied a population of S. parvus from 18 August–26 After locating a vocalizing male, stereo recordings of the multi- st September 2005, June 2006 - January 2007 and again from 1 note advertisement call were made from a distance of 1 m, using th March 2010–13 April 2010 in the Ulu Temburong National directional (sound left) and omni-directional microphones (Senn- Park, Brunei Darussalam, Borneo. The study site was at a narrow, heiser Me 66, Me 62, Sennheiser electronic GmbH & Co. KG, rocky (black shale) section of the Sungai Mata Ikan, a small Germany) and a digital recorder (Zoom HN4, Zoom Co., Japan; freshwater stream that merges into the Belalong River close to the settings: 44.1 kHz, 16-bit resolution). Microphones were placed Kuala Belalong Field Studies Centre (115u099E, 4u339N). Daily 50 cm apart from each other directed at the calling individual. temperatures varied between 24 and 27uC. Annual precipitation Peak sound pressure levels (SPLs) were measured with a sound at the site ranges between 2500 and 4000 mm. level meter (Voltkraft SL-100, Germany: settings: fast/max) during Staurois parvus is a ranid frog, endemic to Borneo, recently each sound recording at a distance of 1 m to the focal individual. resurrected from synonymy with S. tuberilinguis [30]. The separate The A-filter frequency weighting was used because it is species status has been verified using molecular markers [31]. The approximately flat from 1 to 8 kHz, which comprises the call snout-urostyle length and weight of the investigated population of range of S. parvus. male S. parvus averaged 21.560.5 mm (SD; range 20.7–22.7; Recordings with the directional microphone were used to n = 13) and 0.760.05 g (SD; range 0.65–0.80; n = 13). Males are measure call duration, note duration (each call was composed of diurnal and perch on rocks along fast-flowing forest streams. Their many notes), mean-, minimum- and maximum frequency. In white chest and white webbing between toes of hind legs strongly addition, the dependency of frequency and note duration on note contrast to their cryptic dark grey, brown dorsal body (Fig. 1). number was analysed. A period of 7 s of omni-directional Males display a conspicuous visual signal termed foot flagging recordings was selected after each call to analyse the ambient during agonistic male-male encounters in which the conspicuous noise. The sound pressure levels and energy spectra of advertise- webbings of the hind feet are exposed [32]. Male advertisement ment calls and noise were compared from omni-directional calls have not been previously described [33]. microphone recordings. Furthermore, the dependency of sound pressure on note number was analysed. Behavioural observations The acoustic features of stereo recordings were extracted and Behavioural sequences of acoustic and visual signals exhibited measured using custom built programs in PRAAT 5.1.25 DSP by males were recorded using continuous focal sampling [34]. package [35] that automatically logged these variables in an Focal individual males (n = 31) were observed between 1–20 min output file. To analyse single call notes the voiced intervals of the and their activities recorded on video (Sony HC 32E PAL cam call were extracted and note duration in seconds was measured. recorder; Sony Co., Japan). 40 hours of video recordings were Call duration in seconds was calculated from note start and end digitized, stored on DVD and analysed. times. For call frequency analysis a cross-correlation algorithm was To determine whether the vocalizations and visual foot-flagging used to produce a time-varying numerical representation of the displays function in concert or as separate entities, we determined fundamental frequency (F0) for each call. A time step of 0.375 ms the timing intervals between the advertisement calls and foot flags was applied over a range of 3500–6500 Hz according to the F0 from video recordings and tested for differences using a Wilcoxon observed on the spectrogram. From the F0, the parameters’ mean, 2 matched pairs test. Chi -tests were used to test for any associations minimum, and maximum F0 in Hertz were extracted. The mean between the three most common behaviours: advertisement calls, frequency value 6300 Hz was used to apply a filter before foot flags of the left foot, and foot flags of the right foot. Further measuring sound pressure. To extract parameters from noise files,

PLoS ONE | www.plosone.org 2 May 2012 | Volume 7 | Issue 5 | e37965 CHAPTER 1 15

Mulitmodal Communication in a Bornean Frog a similar analysis was applied except to measure maximum external battery amplified speaker (SME-AFS, Saul Mineroff frequency of the 7 s noise file, a long-term average spectrum was Electronics Inc., USA; flat 62 dB from 100 Hz–12 kHz) placed computed with a bandwidth of 50 Hz. To obtain sound pressure between 40–80 cm from the focal male without disturbing it. The (SP) values of ambient noise within the frequency range of the speaker could not be placed at a predetermined distance in the advertisement call, we applied a band-pass filter to the spectrum rough terrain and the distance between frog and speaker was for frequencies from 5300–5900 Hz. The extracted relative SP therefore measured after the experiment to determine the sound values for call and noise were transformed into absolute SP (Pa) by pressure level of each playback. The sound pressure level (SPL) of defining the most intensive SP of the complete sound file (SP the playback at a frog’s position varied between 72–82 dB (re absolute = SP relative6SP measured/SP most intensive). ‘‘SP 20 mPa; Realistic sound level meter with a flat-weighted and fast- measured’’ corresponds to the maximum sound pressure recorded response setting). The effect of SPL on males’ responses was tested in the field. using least squares linear regressions. To test for individual To test the hypothesis that S. parvus uses frequencies that differences in response to the playback treatment, we used the enhance the signal-to-noise ratio we compared maximum sound Wilcoxon matched pairs test. pressure values of ambient noise, advertisement calls and noise with a frequency filter in the range of the call frequency (labelled Results noise at call frequency) using Linear Mixed Models (LMMs). The statistical assumptions for LMM analysis were met (Kolmogorov- Behavioural Displays Smirnov test) and non-normal data were square-root transformed Male S. parvus showed a large repertoire of visual displays. to meet the criteria. LMMs were chosen to investigate differences Common displays were foot flagging and foot flashing. Less in sound pressure within differing number of calls per male and common were arm waving, upright posture, crouched posture and varying pressure values for notes per call. The sound pressure an open-mouth display. All displays were also seen on a regular values of noise, noise filter and call, with every call consisting of 35 basis outside the period of focal sampling. Males displayed from values for every note, were entered as a dependent variable, with the black shale within the stream bed often immediately adjacent the relationship of noise, noise filter and call as predictor variables. to running water. To correct for differences between male individuals, number of Foot flagging was the most common and conspicuous dynamic calls per male and number of notes per call were entered as nested visual signal produced by males (Fig. 1). It was given in both an random variables. For post-hoc tests we used the Student’s t intra- and intersexual context. Foot flags were produced by raising Statistic with the post-hoc sequential Bonferroni correction for either the left or right hind limb off the substrate and then rotating alpha because of repeated pairwise comparisons. it outward and backward in an arc during which the whitish To compare call and noise dominant frequencies the values of webbing between the toes was spread and exposed. The duration these parameters were entered as dependent variables with call of foot flags (time between the raising of the hind limb from the and noise as predictor variables. A nested term was included for substrate until it is returned to the substrate) averaged 1.560.24 s the identities of male (call) and call (note) as random variables to (n = 116). correct for differences between male individuals, number of calls Foot flashing was similar to foot flagging, however, it lacked the per male and number of notes per call. phase in which the hind limb was raised and the limb was not To test if note duration, frequency, and sound pressure are rotated but stretched outwards and retracted immediately. The dependent on the note number of an advertisement call of S. duration of a foot flash was shorter than that of a foot flag and parvus, the model was rerun entering either note duration, averaged 0.8360.15 s (n = 8). Foot flashing was only observed frequency, or transformed sound pressure values as dependent immediately following an advertisement call. variables, with note number as the predictor variable. The Arm waving, upright posture and crouching were observed identities of males (calls) were entered as nested random variables. during close-range male-male encounters. Open-mouth displays All analyses were run using SPSS version 19 (SPSS Inc., Chicago, involved elevating the head while exposing the whitish inner IL, USA). surface of the mouth. One female was seen to foot flag in an aggregation of males. As Acoustic playback experiments in the male display the foot was rotated in an upward, backward To determine whether the advertisement call is used to alert arc exposing the whitish webbing between the toes. Within a three other males to the subsequent visual signal, we conducted acoustic min period, the same female also gave several upright displays, an playback experiments with seven males in the field. To avoid open mouth display, and vocalized twice. The call was a feeble, pseudoreplication, a synthetic call based on the average call single note that could be heard by the observer, but could not be properties of five males was generated using Goldwave version extracted from the video because of the background noise. The 5.06 (Goldwave Inc., St. John’s, Canada). Average call parameters context in which these signals were given appears to have been matched those of a subsequent larger sample of calls verifying that intersexual. All visual signals by both males and females were this initial sample was representative of the population. The call dynamic visual signals that can be turned on and off by the consisted of 16 notes of 18 ms duration each. Each note was signaller. separated by an interval of 100 ms and had a 2-ms rise time and a A ‘‘leg-snout touch’’ tactile display was observed between a 2-ms fall time. The dominant frequency of each note was set at male and a female on one occasion. After having been approached 5770 Hz. by a female, the male turned his back on her and extended his After suitable males were located in the field, they were right leg toward her until his toes touched her snout. Seven presented with a five-minute silent pre-playback control prior to seconds later the right leg was retracted and the left leg extended each five-minute advertisement call playback period. We video in the same fashion. After 12 s the procedure was repeated. The recorded the activities exhibited by males using a digital video cam male then gave an advertisement call and jumped out of view. recorder (Sony HC 32E PAL, Sony Co., Japan) set on a tripod. Further interactions between the two individuals could not be seen The playback stimulus was presented from a portable Hi-MD and it remains unclear if the male and female went into amplexus player (Sony MZ-RH10, Sony Co., Japan) connected to an as might be expected.

PLoS ONE | www.plosone.org 3 May 2012 | Volume 7 | Issue 5 | e37965 16 CHAPTER 1

Mulitmodal Communication in a Bornean Frog

Call characteristics We recorded a total number of 141 advertisement calls of 14 males of S. parvus (all results 6 SE). The energy of the call was concentrated in a narrow frequency band and consisted of on average 3563 short pulsed notes with a dominant frequency of 5578653 Hz (range 5295–5854 Hz; Fig. 2a). The maximum sound pressure of calls of 11 recorded individuals was 0.023 Pa60.002 (SPL = 62 dB; range 0.001–0.126 Pa) at a distance of 1 m. The maximum sound pressure of the ambient background noise averaged 0.082 Pa60.001 (SPL = 72 dB; n = 11) and within the call frequency 0.010 Pa60.001 (SPL = 54 dB; n = 11; Fig. 2b). Thus, the difference in sound pressure between advertisement calls and background noise at 5578 Hz was 8 dB. Overall, the sound pressure between advertisement calls and noise differed significantly (LMM: F2,242.8 = 1560.732, P,0.001). The pairwise comparison of sound pressure between call and noise Figure 3. Maximum sound pressure (square-root transformed indicated that the maximum amplitude of the call had less energy values + S.E.) of noise, advertisement call, and noise within a than the ambient noise (call - noise: ß = 20.136; S.E. = 0.004; frequency filter in the range of the calls of 11 Staurois parvus df = 195; t = 232.464; P,0.001; Fig. 3) but significantly more males (Student’s t-test: ***P,0.001). energy than the noise at its dominant frequency (call – noise at call doi:10.1371/journal.pone.0037965.g003 frequency: ß = 0.052; S.E. = 0.004; df = 195; t = 12.409; P,0.001; 2 Fig. 3). The dominant frequency of the noise was lower than the number: ß = 0.0012; S.E. = 6610 5; df = 1652; t = 21.889; dominant advertisement call frequency (noise - call: ß = 25097; P,0.001) increased with note number (Fig. 4). S.E. = 23; df = 186; t = 2221.593; P,0.001). 2 Call duration (duration - note number: ß =2610 4; Patterns of signalling activity 6 26 , S.E. = 5 10 ; df = 3417; t = 44.452; P 0.001), call frequency In general, foot-flagging was accompanied by advertisement (frequency - note number: ß = 6.19; S.E. = 0.421; df = 2018; calling throughout all periods of the day. A representative t = 14.721; P,0.001) and sound pressure (sound pressure - note sequence of signalling behaviours of one male over a period of 10 minutes was CRLRLCLRLRCRLRLRLCLRRRL where C denotes an advertisement call, R denotes a right foot flag, and L denotes a left foot flag. There was a high degree of association 2 between the three behaviours (X 4 = 169.6, P,0.01). In particular, a left foot flag was strongly associated with a right foot flag and vice versa (Fig. 5). A male giving a right foot flag will follow it with a left foot flag 63% of the time. Likewise, a left foot flag is followed with a right foot flag 74% of the time suggesting that males usually alternate between left and right foot flag. There was also a high transition probability between advertisement call and foot flag. An advertisement call was followed by a foot flag 88% (R or L: 40% or 48%) of the time while a foot flag was followed by an advertisement call only 9–12% of the time. This suggests that advertisement calls are more likely to be followed by foot flagging than foot flagging by advertisement calling. The transition probabilities also indicate that both foot flag of the same leg and advertisement call will unlikely follow itself in the behavioural sequence.

Timing relationship between calls and foot-flags The timing relationship between advertisement calls and foot flags was measured for 19 males for which at least ten observations of foot flags were available. The average delay between an advertisement call and a foot flag was 0.5761.2 s (range 0.0–5.1 s, n = 19). In contrast, the average delay between a foot flag and a subsequent advertisement call was 11.067.6 s (range 1.2–24.8 s, n = 19). The time delay between advertisement call and foot flag Figure 2. Characteristics of the advertisement call of Staurois was significantly shorter than between foot flag and advertisement parvus and its acoustic environment. (A) Oscillogram and call (Wilcoxon matched pairs, Z = 3.82, P#0.001, n = 19; Fig. 6). spectrogram of a representative advertisement call with 34 notes. (B) Power spectrum of the same recording showing the energy contained Acoustic playback experiments in the ambient noise produced by the fast-flowing stream at which males called. The peak at 5500 Hz represents the advertisement call of Variation in sound pressure level of the playback had no S. parvus. significant effect on the number of advertisement calls or foot flags doi:10.1371/journal.pone.0037965.g002 given by males (least squares linear regression, r2 = 0.03, n.s. and

PLoS ONE | www.plosone.org 4 May 2012 | Volume 7 | Issue 5 | e37965 CHAPTER 1 17

Mulitmodal Communication in a Bornean Frog

Figure 5. Transitional frequency matrix between three signalling behaviours (two visual and one acoustic) shown by Staurois parvus. C, L, and R stand for advertisement call, left foot flag and right foot flag, respectively. Width of arrows and their direction show the probability of one behaviour occurring after another behaviour was shown and the sequence of those behaviours. Numbers next to arrows designate the transitional probabilities. doi:10.1371/journal.pone.0037965.g005

Discussion This study reinforces the findings that acoustic and visual displays are functionally linked in the genus Staurois. Grafe & Wanger [25] documented that the advertisement calls and foot flags of S. guttatus form a functional unit as a multicomponent and multimodal display. Their results suggested that the advertisement calls have an alerting function by drawing the attention of the receiver to the subsequent dynamic foot flag. Likewise, S. latopalmatus males use short calls in conjunction with foot flags for intra- and interspecific communication with short calls preceding foot flags [22] and S. tuberilinguis often give foot flags right after calling [36].

Figure 4. Scatterplots of the first 35 notes of the advertisement call of Staurois parvus of (A) mean note duration (n = 14), (B) mean frequency (n = 14) and (C) maximum sound pressure (n = 11). Plots show means of the original data (not estimates of the LMMs) for illustration that do not correspond directly with the statistical results. doi:10.1371/journal.pone.0037965.g004 r2 = 0.07, n.s., respectively). Males produced both advertisement calls and foot flags in response to synthetic advertisement calls. Significantly more foot flags (9.2566.8) were given during the playback period then during the pre-playback period (Wilcoxon matched pairs, Z = 2.20, P,0.05, n = 7; Fig. 6). Although an increase was also shown in the number of advertisement calls given Figure 6. Comparison of timing relationships between adver- in response to the playback, this increase was not significant tisement call and foot flagging display of 19 Staurois parvus (Wilcoxon matched pairs, Z = 1.83, n.s., n = 7; Fig. 7). During the males. Box plots show the median response with interquartile range playback period, males produced significantly more foot flags than and 10th and 90th percentile. calls (Wilcoxon matched pairs, Z = 2.37, P,0.05, n = 7). doi:10.1371/journal.pone.0037965.g006

PLoS ONE | www.plosone.org 5 May 2012 | Volume 7 | Issue 5 | e37965 18 CHAPTER 1

Mulitmodal Communication in a Bornean Frog

night near running water in the same habitat [25]. An additional correlate of visual signalling appears to be diurnality albeit with exceptions [26]. Our results indicate that vocal communication in S. parvus is not impaired by abiotic background noise. The high-frequency advertisement call does not overlap with dominant frequencies of the stream. Two major evolutionary trajectories seem to have been followed by male anurans in their need to avoid broadband low-frequency-dominated masking noise. First, to increase call dominant frequency above the background noise [14],[21],[42]. Such spectral shifts have been documented most notably in Odorrana tormota and Huia cavitympanum in which males call in the ultrasonic range [14],[42]. However, morphological constraints of body size and the inherent transmission limitations caused by the high rate of attenuation and degradation of high frequency sounds may limit widespread use of this solution. Secondly, males that switch to the use of visual signals as the prime mode of communication will be at an advantage, since continuous, chronic noise found along fast flowing streams will favour the evolution of Figure 7. Responses of seven male Staurois parvus to silent signalling in modalities less affected by noise [15]. control (pre-playback) and playback of synthetic advertise- Correlations of body size and call frequency of ranid frogs ment calls. Box plots show the median response with interquartile indicate that all investigated species of the genus Staurois display range and 10th and 90th percentile. *P,0.05, n.s. = non-significant. doi:10.1371/journal.pone.0037965.g007 calls with higher frequencies than expected from their body size [21]. These shifts in signal frequency clearly facilitate communication in the presence of high-intensity background noise as Males of S. parvus used advertisement calls and foot flags closely observed in this study. Likewise, other frog and bird species are combined throughout the day under varying light conditions. The able to increase the pitch of their calls or songs while vocalizing in timing relationship between the acoustic and visual signal supports areas of high ambient noise [12],[13],[43–45]. the alerting signal hypothesis as an explanation for multimodal Additional features of the advertisement call of S. parvus facilitate communication [28]. The latency between foot flags and calls was communication under continuous background noise and distin- significantly higher than between calls and foot flags. In addition, guish it from S. guttatus. Staurois parvus produces an advertisement the playback experiments suggest that one function of the call that varies in note number, ranging from 23 to 54 notes per advertisement call is to alert receivers to the subsequent visual call. We observed a continuous increase in frequency, sound foot flag. The acoustic playback elicited both acoustic and visual pressure and duration with increasing note number. We interpret signalling not just advertisement calling or foot flagging as would the production of very repetitive notes as a redundant feature of be expected if acoustic and visual signals were not linked. the calling behaviour of S. parvus that facilitates communication by Furthermore, males gave significantly more foot flags than calls enhancing the contrast with high levels of continuous background during advertisement call playback suggesting that the visual noise. Increased song duration, and/or increased call or note rate display may be the more informative signal with calls used has been shown to be a response by a wide range of animals to predominantly to gain a receivers attention. increases in background noise [9],[46–48], and females are known In addition to foot flags, male and female S. parvus show to prefer calls with greater intensity, higher call rate and duration numerous, less frequently observed visual displays that need to be [49],[50]. The additional increase in sound pressure and note explored further. Similar to foot flagging, the much faster foot duration with increasing note number could be interpreted as an flashing was also seen to be closely synchronized with advertise- attempt to increase signal range in a graded manner. Instead of ment calling suggesting a similar function in territorial intra- and producing a short long-range signal, males produce longer calls intersexual signalling, but possibly given when males are more that increase communication range with increasing note number. excited when approached by a female. The other visual and tactile Thus, receivers at close range will be targeted immediately while signals appear to be used for close range communication. In those further away will be reached only with later notes, particular, the leg-snout touch tactile display between a male and presumably saving energy. This suggests that males can adjust female, not previously reported in the genus Staurois, suggests that note number depending on proximity of receivers and background mate choice occurs after females approach a male. Similar tactile noise levels and resembles that of graded aggressive calling in other displays have been shown to occur in Hyla ehrhardti, a frog that also anurans [51]. Finally, the energy of the advertisement call of S. uses foot flagging as a visual display and in which males lead parvus is also concentrated in a narrow frequency band. Such females to oviposition sites [37]. narrowly tuned calls presumably facilitate communication in noisy The visual display in S. parvus typically closely follows acoustic environments [52],[53]. signalling. In contrast, multimodal signals in many other anurans Several studies have demonstrated that chorus noise produced are often simultaneous displays given most notably when the vocal by conspecifics and anthropogenic noise can interfere with female sacs are inflated during calling [38–41]. Foot flagging allows for call detection and perception [44],[54],[55]. A threshold for more flexibility as the visual and acoustic signals can be uncoupled detection of at least +1.5–3.0 dB seems to be critical for females to and used to different degrees as the ecological and social be able to detect males. Staurois parvus males generally produce no environments change. overlapping calls or choruses but communicate during constant As in S. gutattus, background noise may be necessary but not background noise. A sound pressure difference of 8 dB, as shown sufficient in explaining foot-flagging in S. parvus because such noise in this study, should be a more than sufficient threshold for female has not led to foot-flagging behaviour in other anurans that call at detection. Female anurans have been shown to discriminate

PLoS ONE | www.plosone.org 6 May 2012 | Volume 7 | Issue 5 | e37965 CHAPTER 1 19

Mulitmodal Communication in a Bornean Frog sounds even under lower signal-to-noise ratios [56–58]. In our context-dependent dynamic selection regimes are recently gaining study area, females are rarely seen with the exception of pairs in wider attention [62],[63] and enhance our understanding of the amplexus and thus their phonotactic behaviour remains unknown. flexibility seen in the use of multimodal signals in S. parvus. It should be noted that S. parvus males communicate in an We conclude that S. parvus has solved the problem of continuous environment of near continuous noise created by running water broadband low-frequency noise by modifying the amplitude, pitch, and thus solutions used by other animals to communicate in repetition rate and duration of notes within their advertisement environments with fluctuating noise levels may not be appropriate. call in addition to using numerous visual signals, foot-flagging Noise generated by social aggregations usually fluctuates in time being the most conspicuous. Such a multi-pronged approach has and thus receivers may adapt by evolving mechanisms that exploit not been documented before in amphibians. It seems likely that such fluctuations [10],[59]. Release from masking can occur by background noise has driven the evolution of multimodal receivers listening in the gaps or dips of fluctuating noise, a communication. Indeed, foot-flagging has evolved independently solution to the cocktail party effect encountered by human mainly in anuran species that communicate along fast-flowing listeners [60],[61]. In addition, spatial release from masking [19] is streams [15]. Playback experiments using visual foot-flagging difficult to achieve because males call in close proximity to running signals would be particularly useful to further our understanding of water from the stone surface of the waterfalls. Thus, both gap or the communication system of frogs in the genus Staurois. dip listening and spatial release from masking may not be viable alternatives for receivers, increasing the selective pressure on male Acknowledgments S. parvus to use visual signals to communicate. Although background noise in the environment of S. parvus is We sincerely thank the Universiti Brunei Darussalam and the staff of the nearly continuous over a time period of minutes to hours, it will Kuala Belalong Field Studies Centre (KBFSC) for logistical support. We vary strongly depending on rainfall. Especially in smaller streams also thank M. Boeckle for his assistance in data analyses and A. Wahab, with small catchment areas that are typical habitats of S. parvus, and O. Konopik for their assistance in the field. background noise levels will vary considerably between days and between dry and wet seasons. Multimodal signalling will be Author Contributions favoured under such fluctuating ecological environments if each Conceived and designed the experiments: TUG DP MS. Performed the modality is favoured under different conditions. Acoustic signalling experiments: TUG DP MS RK JMD SP. Analyzed the data: TUG DP MS will be at an advantage under more quiet conditions and low light RK JMD SP. Contributed reagents/materials/analysis tools: TUG DP MS levels, whereas visual signals will prevail when the noise of rushing WH. Wrote the paper: TUG DP MS WH. water is high and light levels provide the best contrast. Such

References 1. Shannon CE (1948) A mathematical theory of communication. Bell System 19. Bee MA (2008) Finding a mate at a cocktail party: spatial release from masking Technical Journal 27: 379–423. improves acoustic mate recognition in grey treefrogs. Animal Behaviour 75: 2. Endler JA (1993) Some general comments on the evolution and design of animal 1781–1791. communication systems. Philosophical Transactions Royal Society London B 20. Hebets EA, Uetz GW (1999) Female responses to isolated signals from 340: 215–225. multimodal male courtship displays in the wolf spider genus Schizocosa (Araneae: 3. Endler JA (1992) Signals, signal conditions, and the direction of evolution. The Lycosidae). Animal Behaviour 57: 865–872. American Naturalist 139: 125–153. 21. Boeckle M, Preininger D, Ho¨dl W (2009) Communication in noisy environments 4. Brenowitz EA (1982) The active space of red-winged blackbird song. Journal of I: acoustic signals of Staurois latopalmatus Boulenger 1887. Herpetologica 65: Comparative Physiology A 147: 511–522. 154–165. 5. Ryan MJ, Brenowitz EA (1985) The role of body size, phylogeny, and ambient 22. Preininger D, Boeckle M, Ho¨dl W (2009) Communication in noisy environments noise in the evolution of bird song. The American Naturalist 126: 87–100. II: Visual signaling behavior of male foot-flagging frogs Staurois latopalmatus. 6. Slabbekoorn H, Smith TB (2002) Habitat-dependent song divergence in the Herpetologica 65: 166–173. little greenbul: an analysis of environmental selection pressures on acoustic 23. Ho¨dl W, Rodrigues MT, Accacio dM, Lara PH, Pavan D, et al. (1997) Foot- signals. Evolution 56: 1849–1858. flagging display in the Brazilian stream-breeding frog Hylodes asper (Leptodacty- 7. Guilford T, Dawkins MS (1991) Receiver psychology and the evolution of lidae). Austria: Austrian Federal Institut of Scientific Film (O¨ WF).[web animal signals. Animal Behaviour 42: 1–14. application]. 2012. Berkeley, California: AmphibiaWeb Available: http:// 8. Schaub A, Ostwald J, Siemers BM (2008) Foraging bats avoid noise. Journal of amphibiaweb.org/. Experimental Biology 211: 3174–3180. 24. Haddad CFB, Giaretta AA (1999) Visual and acoustic communication in the 9. Diaza M, Parrab A, Gallardob B (2011) Serins respond to anthropogenic noise Brazilian torrent frog, Hylodes asper (Anura: Leptodactylidae). Herpetologica 55: by increasing vocal activity. Behavioral Ecology 22: 332–336. 324–333. 10. Zelick RD, Narins PM (1983) Intensity discrimination and the precision of call 25. Grafe TU, Wanger TC (2007) Multimodal signaling in male and female foot- timing in two species of Neotropical treefrogs. Journal of Comparative flagging frogs Staurois guttatus (Ranidae): an alerting function of calling. Ethology Physiology A 153: 403–412. 113: 772–781. 11. Grafe TU (1996) The function of call alternation in the African reed frog 26. Ame´zquita A, Ho¨dl W (2004) How, when, and where to perform visual displays? (Hyperolius marmoratus): precise call timing prevents auditory masking. Behavioral The case of the Amazonian frog Hyla parviceps. Herpetologica 60: 20–29. Ecology and Sociobiology 38: 149–158. 27. Møller AP, Pomiankowski A (1993) Why have birds got multiple sexual 12. Brumm H (2006) City birds have changed their tune. Current Biology 16: ornaments? Behavioral Ecology and Sociobiology 32: 167–176. R1003–1004. 28. Hebets EA, Papaj DR (2005) Complex signal function: developing a framework 13. Slabbekoorn H, Peet M (2003) Birds sing at a higher pitch in urban noise. of testable hypotheses. Behavioral Ecology and Sociobiology 57: 197–214. Nature 424: 267. 29. McLennan DA (2003) The importance of olfactory signals in the gasterosteid 14. Arch VS, Grafe TU, Narins PM (2008) Ultrasonic signalling by a Bornean frog. mating system: sticklebacks go multimodal. Biological Journal of the Linnaean Biology Letters 4: 19–22. Society 80: 555–572. 15. Ho¨dl W, Ame´zquita A (2001) Visual signaling in anuran amphibians. In 30. Matsui M, Mohamed M, Shimada T, Sudin A (2007) Resurrection of Staurois Ryan MJ, ed. Anuran Communication. Washington, DC: Smithsonian parvus from S. tuberilinguis from Borneo (Amphibia: Ranidae). Zoological Science Institution Press. pp 121–141. 24: 101–106. 16. Partan SR, Marler P (2005) Issues in the classification of multimodal 31. Arifin U, Iskandar DT, Bickford DP, Brown RM, Meier R, et al. (2011) communication signals. American Naturalist 166: 231–245. Phylogenetic relationships within the genus Staurois (Anura, Ranidae) based on 17. Ho¨dl W (1977) Call differences and calling site segregation in anuran species 16S rRNA sequences. Zootaxa 2744: 39–52. from central Amazonian floating meadows. Oecologia 28: 351–363. 32. Harding KA (1982) Courtship display in a Bornean frog. Proceedings of the 18. Schwartz J, Buchanan B, Gerhardt HC (2002) Acoustic interactions among male Biological Society of Washington 95: 621–624. gray treefrogs, Hyla versicolor, in a chorus setting. Behavioral Ecology and 33. Sukumaran J, Das I, Haas A (2006) A synopsis of bioacoustic studies of anuran Sociobiology 53: 9–19. amphibians of Borneo. Herpetological Review 37: 288–293.

PLoS ONE | www.plosone.org 7 May 2012 | Volume 7 | Issue 5 | e37965 20 CHAPTER 1

Mulitmodal Communication in a Bornean Frog

34. Martin P, Bateson P (1994) Measuring Behaviour: An Introductory Guide. 49. Gerhardt HC, Huber F (2002) Acoustic Communication in Insects and Anurans: Cambridge: Cambridge University Press. Common Problems and Diverse Solutions. Chicago: Chicago University Press. 35. Boersma P, Weenink D (2010) Praat: doing phonetics by computer [Computer 50. Bradbury J, Vehrencamp SL (2011) Principles of Animal Communication. program] Version 5.1.43, Available: http://www.praat.org/. Sunderland: Sinauer Associates. 36. Malkmus R, Manthey U, Vogel G, Hoffmann P, Kosuch J (2002) Amphibians 51. Wells KD (2007) The Ecology and Behavior of Amphibians. Chicago: and Reptiles of Mount Kinabalu. Rugell: Gantner Verlag. University of Chicago Press. 37. Hartmann MT, Hartmann PA, Haddad CFB (2004) Visual signaling and 52. Greer BJ, Wells KD (1980) Territorial and reproductive behavior of the tropical reproductive biology in a nocturnal treefrog, genus Hyla (Anura: Hylidae). American frog Centrolenella fleischmanni. Herpetologica 36: 318–326. Amphibia-Reptilia 25: 395–406. 53. Feng AS, Narins PM, Xu C-H (2002) Vocal acrobatics in a Chinese frog, Amolops 38. Narins PM, Grabul DS, Soma K, Gaucher P, Ho¨dl W (2005) Cross-modal tormotus. Naturwissenscahften 89: 352–356. integration in a dart-poison frog. Proceedings of the National Academy of 54. Schwartz JJ, Gerhardt HC (1989) Spatially mediated release from auditory Sciences 102: 2425–2429. masking in an anuran amphibian. Journal of Comparative Physiology A 166: 39. Narins PM, Ho¨dl W, Grabul DS (2003) Bimodal signal requisite for agonistic 37–41. behavior in a dart-poison frog, Epipedobates femoralis. Proceedings of the National 55. Wollerman L, Wiley RH (2002) Background noise from a natural chorus alters Academy of Sciences 100: 577–580. female discrimination of male calls in a Neotropical frog. Animal Behaviour 63: 40. Hirschman W, Ho¨dl W (2006) Visual signaling in Phrynobatrachus krefftii 15–22. Boulenger, 1909 (Anura: Ranidae). Herpetologica 62: 18–27. 56. Narins PM (1982) Effects of masking noise on evoked calling in the Puerto Rican 41. Taylor RC, Klein BA, Stein J, Ryan MJ (2011) Multimodal signal variation in coqui (Anura: Leptodactylidae). Journal of Comparative Physiology A 147: space and time: how important is matching a signal with its signaler? The 439–446. Journal of Experimental Biology 214: 815–820. 57. Schwartz JJ, Gerhardt HC (1998) The neuroethology of frequency preferences 42. Feng AS, Narins PM, Xu C-H, Lin WY, Yu ZL, et al. (2006) Ultrasonic in the spring peeper. Animal Behaviour 56: 55–69. communication in frogs. Nature 440: 333–336. 58. Penna M, Pottstock H, Velasquez N (2005) Effect of natural and synthetic noise 43. Dubois A, Martens J (1984) A case of possible vocal convergences between frogs on evoked vocal responses in a frog of the temperate austral forest. Animal and a bird in Himalyan torrents. Journal fu¨r Ornithologie 125: 455–463. Behaviour 70: 639–651. 44. Bee MA, Swanson EM (2007) Auditory masking of anuran advertisement calls 59. Grafe TU (2003) Synchronized interdigitated calling in the Kuvangu running by road traffic noise. Animal Behaviour 74: 1765–1776. frog, Kassina kuvangensis. Animal Behaviour 66: 127–136. 45. Parris KM, Velik-Lord M, North JMA (2009) Frogs call at higher pitch in traffic 60. Hulse SH (2002) Auditory scene analysis in animal communication. Advances in noise. Ecology and Society 14: 25. the Study of Behavior 31: 163–200. 46. Parks SE, Clark CW, Tyack PL (2007) Short- and long-term changes in right 61. Velez A, Bee MA (2011) Dip listening and the cocktail party problem in grey whale calling behavior: the potential effects of noise on acoustic communication. treefrogs: signal recognition in temporally fluctuating noise. Animal Behaviour The Journal of the Acoustical Society of America 122: 3725–3731. 82: 1319–1327. 47. Kaiser K, Hammers JL (2009) The effect of anthropogenic noise on male 62. Bro-Jørgensen J (2010) Dynamics of multiple signalling systems: animal advertisement call rate in the neotropical treefrog, Dendropsophus triangulum. communication in a world of flux. Trends in Ecology and Evolution 25: Behaviour 146: 1053–1069. 292–300. 48. Penna M, Meier A (2011) Vocal strategies in confronting interfering sounds by a 63. Wilgers DJ, Hebets EA (2011) Complex courtship displays facilitate male frog from the southern temperate forest, Batrachyla antartandica. Ethology 117: reproductive success and plasticity in signaling across variable environments. 1147–1157. Current Zoology 57: 175–186.

PLoS ONE | www.plosone.org 8 May 2012 | Volume 7 | Issue 5 | e37965 21

Chapter 2

MULITMODAL IN THE SMALL TORRENT FROG (MICRIXALUS SAXICOLA) IN A COMPLEX ACOUSTIC ENVIRONMENT

DORIS PREININGER, MARKUS BOECKLE, ANITA FREUDMANN, IRIS STARNBERGER,

MARC SZTATECSNY AND WALTER HÖDL

Behavioral Ecology and Sociobiology Special Issue on Multimodal Communication Received 28 June 2012; Accepted 20 August 2012

Changes made to the originally accepted manuscript to fit the format of this thesis: Line numbers were deleted and figures were inserted above the respective figure caption. 22

CHAPTER 2 23

Multimodal signaling in the Small Torrent Frog (Micrixalus saxicola) in a complex acoustic environment

Doris Preininger, Markus Boeckle, Anita Freudmann, Iris Starnberger, Marc Sztatecsny &

Walter Hödl

Corresponding author: Doris Preininger, Department of Evolutionary Biology, University of

Vienna, Althanstraße 14, A-1090 Vienna, Austria, (Tel.: +43 1 4277 54523; Fax: +43 1 4277

9544). Email address: [email protected].

Doris Preiningera,*, Markus Boeckleb, Anita Freudmannc, Iris Starnbergera, Marc Sztatecsnya

& Walter Hödla aDepartment of Evolutionary Biology bDepartment of Cognitive Biology cDepartment of Animal Biodiversity

University of Vienna, Althanstraße 14, A-1090 Vienna, Austria email: [email protected].

Abstract

Many animals use multimodal (both visual and acoustic) components in courtship signals.

The acoustic communication of anuran amphibians can be masked by the presence of environmental background noise, and multimodal displays may enhance receiver detection in complex acoustic environments. In the present study we measured sound pressure levels of concurrently calling males of the Small Torrent Frog (Micrixalus saxicola), and used acoustic playbacks and an inflatable balloon mimicking a vocal sac to investigate male responses to controlled unimodal (acoustic) and multimodal (acoustic and visual) dynamic stimuli in the frogs’ natural habitat. Our results suggest that abiotic noise of the stream does not constrain 24 CHAPTER 2 signal detection, but males are faced with acoustic interference and masking from conspecific chorus noise. Multimodal stimuli elicited greater response from males and triggered significantly more visual signal responses than unimodal stimuli. We suggest that the vocal sac acts as a visual cue and improves detection and discrimination of acoustic signals by making them more salient to receivers amidst complex biotic background noise.

Keywords: Anura, acoustic signal, background noise, multimodal communication, visual cue, vocal sac

Introduction

To explain evolutionary patterns in animal communication it is critical to understand the mechanisms of signal production, the conditions under which signals are produced, and how signals are perceived by receivers (Bradbury and Vehrencamp 1998; Brumm and

Slabbekoorn 2005; Miller and Bee 2012). During the last decade, it has become clear that communication signals in many taxa are more complex than previously thought (Hebets and

Papaj 2005). Complex signals can consist of multiple components in a single modality (e.g. acoustic, visual or chemical) or in multiple sensory modalities (multimodal communication) with components being presented either together or independently (Partan and Marler 1999;

Partan and Marler 2005). In so called fixed-composite signals (sensu Smith 1977; Partan and

Marler 2005), signal components occur always together. Based on their assumed information content, multimodal signals have been classified as redundant (all signal components elicit an equivalent response in the receiver) or non-redundant (signal components elicit a different response in the receiver). Hebets and Papaj (2005) suggested that multiple signal components may evolve when they increase the signal content (content-based hypothesis), CHAPTER 2 25 facilitate the perception of each other (inter-signal interaction hypothesis), or enhance signal transmission for instance in noisy environments (efficacy-based hypothesis).

Acoustic signal detection and discrimination can be constrained by abiotic and/or biotic sources such as waterfalls or vocalizing hetero- or conspecifics, thus favoring the evolution of complex signaling strategies (Gerhardt and Klump 1988; Schwartz and Gerhardt 1989;

Brumm and Slabbekoorn 2005; Gordon and Uetz 2012), which could facilitate faster and more accurate detection by receivers (Rowe 1999; Otovic and Partan 2009). However, the selection pressures influencing signaling strategies may differ when environmental noise originates primarily from conspecifics compared with other types of noise, because conspecific noise contains a high degree of frequency and temporal overlap between the signals and noise (Gerhardt and Huber 2002).

Anuran amphibians are excellent model systems to study multimodal communication in noisy environments. In many anurans, males produce loud advertisement calls that mediate both female choice and male spacing (Ryan 2001; Gerhardt and Huber 2002). Frog communication may take place in dense breeding choruses (Bee and Micheyl 2008) and/or noisy settings like streams that produce broadband ambient background noise (Boeckle et al.

2009). While advertising, males inflate and deflate the vocal sac which has a primary evolutionary function of recycling air during vocalization, thereby increasing the call rate and distributing sound waves omnidirectionally (Rand and Dudley 1993; Pauly et al. 2006). Since the vocal sac is inevitably moving while a male is calling , it can send a fixed-composite signal (see Hirschmann and Hödl 2006 for exception) imparting increased detectability due to movement and coloration (Rosenthal et al. 2004; Taylor et al. 2008). For example, in the dart-poison frog Allobates femoralis, simultaneous acoustic and visual signals are necessary to evoke an aggressive reaction in males defending a territory (Narins et al. 2003; Narins et al. 2005). For females of the squirrel tree frog (Hyla squirella), the availability of the vocal sac as a visual cue makes an unattractive male call more appealing, whereas additional visual information is assessed from lateral body stripes when male calls are equally attractive

(Taylor et al. 2007; Taylor et al. 2011b). Despite these examples of a preference for 26 CHAPTER 2 multimodal over unimodal signals, there seem to be vast differences in the importance of signal components and the responses they elicit in the receiver, even in species facing similar ecological problems. Since ecological settings to which animals are exposed can be complex (e.g., many calling individuals, various abiotic sources of noise etc.) we wanted to test receiver responses in a frog’s natural habitat by using an experimental model set-up.

Robotic models present 3-dimensional visual stimuli that can be detected from a wide angle of view making their use advantageous when the position of receivers can not be controlled before starting an experiment. The successful use of robots for testing isolated or combined signal components has been demonstrated in studies on a variety of animals, including frogs

(Narins et al. 2003; Taylor et al. 2008; Krause et al. 2011).

The Small Torrent Frog (Micrixalus saxicola) occurs along tropical streams, and communicates in large social aggregations. Signal detection and discrimination in M. saxicola could therefore be constrained by both conspecific chorus noise and ambient stream noise. Males display a bright white vocal sac during vocalizing (Fig.1) and perform additional visual signals (e.g. foot-flagging) in male-male agonistic interactions (Krishna and

Krishna 2006). Given the acoustically complex environment in which the frogs occur and their signaling behavior, we investigated whether stream noise and/or chorus noise constrains the acoustic signal component in male-male agonistic behaviors (signal efficacy approach, Hebets and Papaj 2005). To do so, we first characterized acoustic features of the male advertisement call and measured sound pressure levels of calls and background noise during the breeding season. Further, we investigated the sensory components in the male display by providing controlled and naturally occurring stimuli (call alone and call with synchronously presented artificial vocal sac) via an experimental set-up, and examined male responses. These experiments allowed us to test the signal-interaction hypothesis predicting that multimodal composites amplify signal detection and discrimination compared to the unimodal acoustic component (Hebets and Papaj 2005).

CHAPTER 2 27

Methods

Study site and animals

The Small Torrent Frog (M. saxicola) is endemic to the Western Ghats in India (Daniels

2005) and occurs exclusively along small, fast-flowing streams within the evergreen forests

(Chandran et al. 2010). Individuals are diurnal and inhabit perennial streams characterized by low water, air and soil temperature (Reddy et al. 2002). Males produce calls with a series of pulses from exposed sites on rocks in shallow areas of the stream to advertise for females and defend breeding grounds in relatively crowded aggregations (Krishna and Krishna 2006).

We studied a population of M. saxicola located at the Kathalekan Myristica swamp forest

(14.27414°N, 74.74704°E) in the central Western Ghats at the end of the monsoon season

(September and October 2010). Males in our study population had a mean snout-urostyle length (SUL) of 23.6 mm, and a mean mass of 1.1 g, (n = 13). Inter- individual distance between calling males was measured to determine average receiver distance. Median distance between advertising individuals in the study population was 0.98 m (range: 0.38 -

2.69 m, n = 15).

The frogs were captured with permission of the Centre for Ecological Sciences, Indian

Institute of Science, Bangalore (permission number: D.WL.CR-27/2008-09), and released immediately after taking body measurements. All behavioral experiments were performed without physical contact with the study animals.

Acoustic recordings

After locating a vocalizing male, we recorded advertisement calls from a distance of approx. 1 m, using an omnidirectional microphone (Sennheiser Me 62) and a digital recorder

(Zoom HN4; settings: 44.1 kHz, 16-bit resolution). We measured peak sound pressure levels with a sound level meter (Voltkraft SL-100; settings: fast/max, C-weighted) from distance of 1 28 CHAPTER 2 m, which equaled the measured the median male inter-individual distance of 0.98 m. During recordings the focal male was closer to the microphone than calling neighbors. One second after each advertisement call, a period of 3 s was selected from omni-directional call recordings to analyze environmental background noise comprising chorusing conspecifics

(termed “chorus noise”). We additionally recorded ambient stream noise without male calls

(simplifyingly termed “abiotic noise” despite occasional comprising insect signals) before and after frog choruses from the same recording position as calls. The stream was regarded as noise field in which ambient noise intensity was considered almost unchanging within the measured distance of 1m. For the call analyses we discarded recordings with overlapping calls from chorusing males. We measured SUL and body mass of each focal individual after the sound recordings with a sliding caliper to the nearest 0.1 mm and a digital mini scale to the neatest 0.01 g.

Acoustic features of recordings were extracted and measured using custom built programs in PRAAT 5.2.22 DSP package (Boersma and Weenik 2011) that automatically logged relevant variables in an output file. To analyze single call notes we extracted the voiced intervals of the call and measured note duration in seconds. Call duration in seconds was calculated with note start and end times. The spectral structure of calls was investigated using spectrograms (fast Fourier transform (FFT) method; window length: 0.01; time step:

1000; frequency step: 500; Gaussian window; and dynamic range: 50 dB). For call frequency analysis a cross-correlation pitch extraction algorithm was used to produce time-varying numerical representation of the fundamental frequency (F0) -contour for each call. We applied a time step of 0.5 ms over a range of 3000–6000 Hz according to the F0 observed in the spectrogram and extracted the parameters mean, minimum, and maximum F0 from the

F0-contour. The mean frequency value ± 500 Hz was used to apply a filter before measuring sound pressure. To extract parameters from noise files we applied a similar analysis and computed a long-term average spectrum with a bandwidth of 50 Hz to measure maximum frequency. To obtain sound pressure (SP) values of chorus and abiotic noise within the frequency range of the advertisement call we applied a pass Hann-band filter to the spectrum CHAPTER 2 29 for frequencies from 4300–5300 Hz. The extracted relative SP values for call and noise were transformed into absolute SP (Pa) by defining the most intensive SP of the complete sound file (SP absolute = SP relative x SP measured/SP most intensive). “SP measured” corresponds to the maximum sound pressure recorded in the field.

Playback experiments

Unimodal (acoustic) and multimodal (acoustic and visual) stimuli were presented on a platform made out of two plastic containers (Fig. 2). The larger container (7 x 18 x 11 cm) was filled with pebbles and placed in the stream where it served as an anchor for the attached smaller container (6 cm x 10 cm x 11 cm) and the loudspeaker (Sony SRS-M 30) connected to an MP3-Player (Odys Pax). To test if the vocal sac is the primary visual signal component that makes the display more salient to receivers we presented males exclusively with an artificial inflatable vocal sac. We did not use a stationary model frog as additional visual stimulus or further identification feature. The tip of a white latex glove (inflated diameter: 1 cm) on top of the small container mimicked the vocal sac, which could be inflated by the experimenter by gently blowing air through a 2.5 m long hose. We tested 10 males with unimodal playbacks and 10 males with multimodal playback presentations resulting in a total of 20 tested individuals (identifiable through photos). Playback stimuli were presented from the experimental set-up, placed 50 cm from the focal individual. From a distance of 1 m from the focal male, the experimenter operated the MP3 player and in multimodal presentation manually inflated the artificial vocal sac synchronously with each call. The pre- recorded advertisement call was generated by using averaged call values from the studied population (call duration: 2.6 s; note number: 21; mean frequency: 4.6 kHz; intercall interval

7.4 s). The acoustic stimulus consisted of 3 advertisement calls with an average intensity of

75 dB at 50 cm. Experimental play-back presentations were undertaken only when the focal individual showed no signaling behavior in the prior 60 s. Either unimodal or multimodal stimuli were presented for a period of 30 s, followed by a 90 s control phase and a 30 CHAPTER 2 subsequent second same stimulus playback and control phase. All trials were video recorded with a waterproof camera (Sanyo Xacti WH1) positioned on a tripod. We analyzed frequencies and durations of the behavior categories “calling”, “tapping”, “foot-flagging” and

“position change” during presentation and control phases with the behavioral coding software

Solomon Coder (Péter 2011). “Tapping” constitutes the lifting of either the right or left leg without stretching it, whereas “foot-flagging" labels the behavior of completely extending the leg above and back in an arc and bringing it back to the body side (Hödl and Amézquita

2001). Behaviors termed “position change” included approach, moving away and turn. Digital photographs of dorsal patterns and colorations allowed individual recognition, and ensured that we were able to avoid repeat testing of the same individuals.

Data Analysis

To test the hypothesis that M. saxicola advertisement calls are masked by noise, we analyzed 112 calls and respective noise recordings from 13 individuals, in turn comprising measurements at 13 positions in the stream. We compared maximum sound pressure values of the acoustic factors: advertisement calls, abiotic noise, abiotic noise in the frequency range of the call (filtered abiotic noise) and chorus noise in the frequency range of the call

(filtered chorus noise) using a Linear Mixed Model (LMM). The LMM allows repeated measurements of the same individual to be fitted in the model as random variables, thus controlling for differing number of calls per male and notes per call. The statistical assumptions for LMM analysis were met (Kolmogorov-Smirnov test).

The sound pressure values [Pa] of all acoustic factors were entered as dependent variables, with the acoustic factors as predictor variables. We entered the identities of male

(call) and call (note) as nested random variables, to correct for differences between male individuals, number of calls per male and number of notes per call. For post-hoc tests we used Student’s t Statistic with sequential Bonferroni correction for alpha because of repeated pairwise comparisons. CHAPTER 2 31

A second LMM was conducted to evaluate the differences between dominant frequencies of call and background noise. To compare frequencies of call and noise the dominant frequencies of these parameters were entered as dependent variables with call and noise as predictor variables. The identities of male (call) and call (note) were entered as nested random variables. To test if male SUL and body mass influence mean dominant call frequency we performed a linear regression analysis.

To investigate the hypothesis that the inflating vocal sac acts as an additional visual cue we compared behavioral responses to unimodal and multimodal playbacks using a two-tailed

Mann–Whitney U test for independent samples. To analyze differences in call duration in response to playback presentations we used a LMM to correct for differing numbers of calls per individual. Call duration was entered as a dependant variable, with modality (unimodal vs. multimodal) as the predictor variable. A nested term was included for the identity of male

(call) as a random variable. All analyses were undertaken using SPSS version 19 (SPSS

Inc., Chicago, IL, USA).

Results

Frog calls and environmental noise

Advertisement calls of Micrixalus saxicola (Fig. 3) had an average duration of 2.0 ± 0.1 s

(all results ± SE and N = 13 in all cases) and comprised a series of 21 ± 1 notes with an average duration of 0.021 ± 0.001 s (Fig. 3a, b). Single pulsed notes were produced at the beginning and end of the call with an inter-note interval of 0.136 s ± 0.005, whereas grouped notes in the middle of the call had multiple pulses and an inter-grouped notes interval of 0.03 s ± 0.001 (Fig. 3c). The frequency of the advertisement calls averaged 4771 Hz ± 29 (range:

4574 – 4969 Hz, Fig. 3d) and was negatively influenced by SUL (linear regression: N = 13, r 32 CHAPTER 2

2= 0.37, P = 0.016), but not affected by body weight (linear regression: N = 13, r 2= 0.08, P =

0.176).

The call frequency showed clear differences to the low-frequency dominated stream noise

(LMM: pairwise comparison: ß = 4168; SE = 22; t = 188.087, P < 0.001). The maximum SP of the call averaged 0.056 Pa (69 dB) at a distance of 1 m. Overall SP comparisons of call and noise differed significantly (LMM: F3,2357 = 39.806, P < 0.001). At 1 m distance the call had a higher SP than abiotic noise values (LMM: pairwise comparison: P < 0.001, Fig. 4), but did not differ from the SP of conspecific chorus noise filtered in the frequency range of the call (LMM: pairwise comparison: ß = 0.010; SE = 0.005; t = 1.829, P = 0.068). The estimated maximum SP of chorus noise averaged 0.046 Pa (67 dB) resulting in a difference of -2 dB relative to the analyzed frog calls.

Playback experiments

When presented with multimodal stimuli all tested males increases the number of calls they produced (Mann-Whitney U test = 12, N 1 = N 2 = 10, P = 0.004) and tapping behaviors

(Mann-Whitney U test = 12, N 1 = N 2 = 10, P = 0.002) and performed more position changes

(Mann-Whitney U test = 17, N 1 = N 2 = 10, P = 0.007) in comparison to unimodal trials. Most interestingly, foot-flagging behavior could only be elicited by multimodal playbacks (Fig. 5) and the mean call duration of 2.2 s ± 0.7 during unimodal playbacks expanded to 5.0 s ±

0.36 during multimodal presentations (LMM: F1,165 = 12.519, P = 0.001, Fig. 6).

Discussion

Our results show that sound pressure levels of M. saxicola male calls significantly exceeded the ambient abiotic noise level in the frogs’ habitat. Stream noise had less energy than frog calls across the entire human audible frequency range. Contrary to species of the CHAPTER 2 33

Bornean Splash Frogs (Staurois) who inhabit low-frequency dominated, torrential streams and waterfalls (Boeckle et al. 2009; Grafe et al. 2012), continuous stream noise alone was unlikely to constrain acoustic signal detection in M. saxicola. Chorus noise, however, appeared more likely to hamper individual call detection as the measured SPL differences between male M. saxicola calls and chorus noise were small (2 dB). Unfortunately, the perceptual capabilities of M. saxicola’s auditory system are unknown but studies on other frog species suggests that the substantial frequency overlap of conspecific noise observed in the present study is likely to interfere with acoustic signal detection (Wilczynski et al. 1993;

Schwartz and Gerhardt 1998; Wollerman and Wiley 2002; Bee 2008; also see Schwartz and

Gerhardt 1998 for improved call detection in the presence of noise). Chorus noise can vary based on seasonal and population density thereby creating a fluctuating environment, which has been suggested to favor the evolution of multimodal signals (Bro-Jørgensen 2010).

In the presence of noise several different strategies can increase the probability of signal recognition and detection. One strategy to provide release from masking is spatial distribution. In Cope´s grey tree frog (Hyla chrysoscelis) improvements in signal detection and mate recognition are obtained when a signal is spatially separated from a masker, particularly at a signal-to-noise ratio (SNR) of -3 dB (Bee 2008). However if calls are not spatially separated from background noise, females of a neotropical treefrog (Hyla ebraccata) located calls with +3 dB but not +1.5 dB SNR (Wollerman 1999). In several bird species spectral shifts and/or amplitude adjustments have been reported in areas with noise

(Slabbekoorn and Peet 2003; Nemeth and Brumm 2010), likewise adaptive strategies such as high frequency calls have evolved in anurans (Feng and Narins 2008; Boeckle et al. 2009) to increase the signal-to-noise ratio in unimodal signals. There are also an increasing number of examples of the use of visual displays to enhance signal efficacy in noisy environments including in crabs (Uca mjoebergi) and lizards (Anolis cristatellus, A. gundlachi and Amphibolurus muricatus) (Peters and Evans 2003; Ord et al. 2007; Milner et al. 2008).

The observed +2 dB SNR in M. saxicola could be a sufficient detection threshold, but we also have to consider that the results can be explained by differences in distance to the 34 CHAPTER 2 microphone between the focal male and the more distant, and thereby degraded, neighbor calls. In dense aggregations and close-range interactions we would expect the SNR to be less or even negative depending on the position of the receiver. Hence, spatial segregation of opponent males could reduce masking, and interacting signal components could be beneficial for early detection and localization of conspecifics.

In our behavioral experiments, multimodal stimuli significantly increased the frequency of response behaviors compared to unimodal acoustic stimuli, and foot-flagging behavior could only be elicited by multimodal stimuli. We suggest that the visual component acts as an amplifier to the acoustic component supporting the inter-signal interaction hypothesis. The advertisement call may serve as long-range signal (Bee 2007), and integration of a pulsating vocal sac could facilitate localization in dense aggregations of concurrently calling individuals

(Gomez et al. 2011; Taylor et al. 2011a; Taylor et al. 2011b). The localization and detection of a caller is more difficult when masked by conspecific calls with a high-degree of spectral overlap (Marshall et al. 2006), making the visual epiphenomenon particularly advantageous in large choruses. The visual component as part of the acoustic signal has been suggested to modulate male reactions including attacks (Narins et al. 2003; de Luna et al. 2010), therefore multimodal signals displayed in close proximity could trigger a more intense response. Across species quantification of multimodal signals in spiders (Hebets 2008), fish

(Van Staaden and Smith 2011) and anurans (Taylor et al. 2011a) provide evidence for response variations and highlight differences in signal dominance and receiver perception. It remains difficult to draw assumptions on signal function, but quantification of receiver responses under different environmental conditions will help to explain the processes acting on complex signals.

Hödl & Amézquita (2001) discussed ecological conditions favoring the evolution of visual signals in anurans such as displays at elevated perches, diurnality and ambient noise which all apply to the study species M. saxicola. Although males can be observed advertising around the year, aggregation density is greatest during the presumable main breeding period at the end of the monsoon season (Gururaja pers. comm.). During this period large CHAPTER 2 35 aggregations form in certain parts of the stream that provide favorable conditions for reproduction, such as shallow water riffle areas where males perch on rocks and display and females dig underwater oviposition cavities (Gururaja 2010).

When attending conspecific choruses, males increase the probability of attracting a mate

(Gerhardt and Huber 2002), but have to face continuous noise levels and limited options to deal with masking interference. However, an increase in signal duration as observed in M. saxicola males during multimodal stimuli presentations could not only indicate fighting ability to the opponent and facilitate spacing but enhance a male’s detectability for females in a chorus. Additionally, playback experiments in numerous anuran species have demonstrated that females prefer long call durations usually associated with energetic costs (Gerhardt and

Huber 2002) and possible weight loss in males during the breeding season (Murphy 1994).

Calling activity and mating tactics could be related to body mass in M. saxicola, whereas spectral features of the call, determined mostly by larynx size (Gerhard and Huber 2002), are not expected to be affected by the weight of an individual. Snout-urostyle length, however, showed a negative influence on call frequency, which could be a reliable cue to body size for receivers when detecting a call. Lower frequency calls would indicate larger body size usually preferred by female conspecifics (Ryan and Keddy-Hector 1992).

We conclude that limited shallow water areas in the stream used for reproduction by M. saxicola lead to strong competition between males, and dense breeding choruses create constant background noise levels. Our results indicate that multimodal signals are necessary to evoke agonistic behavior in this species. Thus, we suggest that the acoustic signal component modulated by the visual component makes the display more salient and facilitates localization and detection of nearby opponent individuals.

The vocal sac in anurans did not evolve as a visual cue but as an organ to improve calling ability, yet its role in communication has been demonstrated in a number of studies. Due to its evolutionary background the vocal sac’s secondary function as a visual cue or signal component is inevitably linked to the acoustic component. Accordingly, it seems not surprising that the present and several other studies found interaction between the acoustic 36 CHAPTER 2 and visual signal components (Narins et al. 2003; Rosenthal et al. 2004; Gomez et al. 2011;

Zeyl and Laberge 2011). In the majority of anuran species the vocal sac is a multimodal fixed-composite signal, but demonstrating whether the visual component adds additional signal content not included in the call, and assessing signal information content, remain difficult tasks. In M. saxicola the visual displays (e.g. foot-flagging) presented independently of calls further add to the complexity in communication behavior, but may also allow for sophisticated behavioral experiments. We suggest a research approach focusing on receiver detection sensitivity and receiver response to visual signaling behaviors that can be performed independently of the auditory signal (e.g. foot-flagging) to explain how selection on senders and receivers promotes complex displays under different acoustic and environmental conditions. Moreover, further across species comparisons of how single and combined signal components influence receivers are essential to draw conclusions on signal function.

Acknowledgements

We thank K. V. Gururaja, S. P. Vijayakumar and V. Torsekar for their logistic and professional help at the study site and K. Shanker for his scientific collaboration. B.

Weissinger helped analyzing video recordings. A. T. Hedge and his family provided us a pleasant stay. We thank V. Arch, R.C. Taylor and an anonymous reviewer for very helpful comments on the manuscript. The study was supported by the Austrian science fund FWF-

P22069 and the University of Vienna FS 100/2012.

Ethical standards

All experiments reported in this article comply with the current laws of the country in which they were performed.

Conflict of interest

The authors declare that they have no conflict of interest. CHAPTER 2 37

References

Bee MA (2007) Sound source segregation in grey treefrogs: Spatial release from masking by the sound of a chorus. Anim Behav 74:549-558 Bee MA (2008) Finding a mate at a cocktail party: spatial release from masking improves acoustic mate recognition in grey treefrogs. Anim Behav 75:1781-1791 Bee MA, Micheyl C (2008) The cocktail party problem: What is it? How can it be solved? And why should animal behaviorists study it? J Comp Psychol 122:235-251 Boeckle M, Preininger D, Hödl W (2009) Communication in noisy environments I: Acoustic signals of Staurois latopalmatus Boulenger 1887. Herpetologica 65:154-165 Boersma P, Weenik D (2011) PRAAT: Doing phonetics by computer (version 5.2.22) [computer program]. http://www.praat.org Bradbury JW, Vehrencamp SL (1998) Principles of Animal Communication. Sinauer Press, Sunderland Bro-Jørgensen J (2010) Dynamics of multiple signalling systems: animal communication in a world in flux. Trends Ecol Evol 25:292-300 Brumm H, Slabbekoorn H (2005) Acoustic communication in noise. Adv Stud Behav 35:151- 209 Chandran MDS, Rao GR, Gururaja KV, Ramachandra TV (2010) Ecology of the swampy relic forests of Kathalekan from Central Western Ghats. Bioremed Biodiv Bioavail 4:54-68 Daniels RJR (2005) Amphibians of Peninsular India. University Press (India) Private Limited, Hyderabad de Luna AG, Hödl W, Amézquita A (2010) Colour, size and movement as visual subcomponents in multimodal communication by the frog Allobates femoralis. Anim Behav 79:739-745 Feng AS, Narins PM (2008) Ultrasonic communication in concave-eared torrent frogs (Amolops tormotus). J Comp Physiol A 194:159-67 Gerhardt HC, Huber F (2002) Acoustic communication in Insects and Anurans: Common problems and diverse solutions. University of Chicago Press, Chicago Gerhardt HC, Klump GM (1988) Masking of acoustic signals by the chorus background noise in the green tree frog: A limitation on mate choice. Anim Behav 36:1247-1249 Gomez D, Théry M, Gauthier A-L, Lengagne T (2011) Costly help of audiovisual bimodality for female mate choice in a nocturnal anuran (Hyla arborea). Behav Ecol 22:889-898 Gordon SD, Uetz GW (2012) Environmental interference: impact of acoustic noise on seismic communication and mating success. Behav Ecol 23:707-714 Grafe TU, Preininger D, Sztatecsny M, Kasah R, Dehling JM, Proksch S, Hödl W (2012) Multimodal Communication in a Noisy Environment: A Case Study of the Bornean Rock Frog Staurois parvus. PLoS One 7:e37965 Gururaja KV (2010) Novel reproductive mode in a torrent frog Micrixalus saxicola (Jerdon) from the Western Ghats, India. Zootaxa 2642:45-52 Hebets EA (2008) Seismic signal dominance in the multimodal courtship display of the wolf spider Schizocosa stridulans Stratton 1991. Behav Ecol 19:1250-1257 Hebets EA, Papaj DR (2005) Complex signal function: developing a framework of testable hypotheses. Behav Ecol Sociobiol 57:197-214 Hirschmann W, Hödl W (2006) Visual signaling in Phrynobatrachus krefftii Boulenger, 1909 (Anura: Ranidae). Herpetologica 62:18-27 Hödl W, Amézquita A (2001) Visual signaling in anuran amphibians. In: Ryan MJ (ed) Anuran Communication. Smithsonian Institution Press, Washington, DC, U.S.A, pp 121-141 Krause J, Winfield AFT, Deneubourg J-L (2011) Interactive robots in experimental biology. Trends Ecol Evol 26:369-375 Krishna SN, Krishna SB (2006) Visual and acoustic communication in an endemic stream frog, Micrixalus saxicolus in the Western Ghats, India. Amphibia-Reptilia 27:143-147 38 CHAPTER 2

Marshall VT, Schwartz JJ, Gerhardt HC (2006) Effects of heterospecific call overlap on the phonotactic behaviour of grey treefrogs. Anim Behav 72:449-459 Miller CT, Bee MA (2012) Receiver psychology turns 20: is it time for a broader approach? Anim Behav 83:331-343 Milner RNC, D. JM, Backwell PRY (2008) Does the environmental context of a signalling male influence his attractiveness? Anim Behav 76:1565-1570 Murphy CG (1994) Determinants of chorus tenure in barking treefrogs (Hyla gratiosa). Behav Ecol Sociobiol 34:285-294 Narins PM, Grabul DS, Soma KK, Gaucher P, Hödl W (2005) Cross-modal integration in a dart-poison frog. P Natl Acad Sci USA 102:2425-2429 Narins PM, Hödl W, Grabul DS (2003) Bimodal signal requisite for agonistic behavior in a dart-poison frog, Epipedobates femoralis. P Natl Acad Sci USA 100:577-580 Nemeth E, Brumm H (2010) Birds and anthropogenic noise: are urban songs adaptive? Am Nat 176:465-475 Ord TJ, Peters RA, Clucas B, Stamps JA (2007) Lizards speed up visual displays in noisy motion habitats. P R Soc Lond B 274:1057-1062 Otovic P, Partan S (2009) Multimodal signaling in animals. In: Encyclopedia of Neuroscience (Ed. by L. R. Squire). Oxford: Academic Press, pp 1095-1105 Pauly GB, Bernal XE, Rand AS, Ryan MJ (2006) The vocal sac increases call rate in the Túngara Frog Physalaemus pustulosus. Physiol Biochem Zool 79:708-719 Péter A (2011) Solomon Coder (version beta 11.01.22): A simple solution for behavior coding. http://solomoncoder.com/ Peters RA, Evans CS (2003) Design of the Jacky dragon visual display: signal and noise characteristics in a complex moving environment. J Comp Physiol A 189:447-459 Rand AS, Dudley R (1993) Frogs in helium: The anuran vocal sac is not a cavity resonator. Physiol Zool 66:793-806 Reddy AH, Gururaja KV, Krishnamurthy SV (2002) Habitat features of an endemic anuran Micrixalus saxicola Jerdon, 1853 (Amphibia: Ranidae) in central Western Ghats, India. Amphibia-Reptilia 23:370-374 Rosenthal GG, Rand AS, Ryan MJ (2004) The vocal sac as a visual cue in anuran communication: an experimental analysis using video playback. Anim Behav 68:55- 58 Rowe C (1999) Receiver psychology and the evolution of multicomponent signals. Anim Behav 58:921-931 Ryan MJ, Keddy-Hector A (1992) Directional patterns of female mate choice and the role of sensory biases. Am Nat 139:4-35 Ryan MJE (2001) Anuran Communication. Smithsonian Institute Press, Washingtion, DC. Schwartz JJ, Gerhardt HC (1989) Spatially mediated release from auditory masking in an anuran amphibian. J Comp Physiol A 166:37-41 Schwartz JJ, Gerhardt HC (1998) The neuroethology of frequency preferences in the spring peeper. Anim Behav 56:55-69 Slabbekoorn H, Peet M (2003) Birds sing at a higher pitch in urban noise. Nature 424:267 Taylor RC, Buchanan BW, Doherty JL (2007) Sexual selection in the squirrel treefrog Hyla squirella: the role of multimodal cue assessment in female choice. Anim Behav 74:1753-1763 Taylor RC, Klein BA, Ryan MJ (2011a) Inter-signal interaction and uncertain information in anuran multimodal signals. Curr Zool 57:153-161 Taylor RC, Klein BA, Stein J, Ryan MJ (2008) Faux frogs: multimodal signalling and the value of robotics in animal behaviour. Anim Behav 76:1089 Taylor RC, Klein BA, Stein J, Ryan MJ (2011b) Multimodal signal variation in space and time: how important is matching a signal with its signaler? J Exp Biol 214:815-820 Van Staaden MJ, Smith AT (2011) Cutting the Gordian knot: Complex signaling in African cichlids is more than multimodal. Curr Zool 57:237-251 Wilczynski W, McClelland BE, Rand AS (1993) Acoustic, auditory, and morphological divergence in three species of neotropical frog. J Comp Physiol A 172:425-438 CHAPTER 2 39

Wollerman L (1999) Acoustic interference limits call detection in a Neotropical frog Hyla ebraccata. Anim Behav 57:529-536 Wollerman L, Wiley RH (2002) Background noise from a natural chorus alters female discrimination of male calls in a Neotropical frog. Anim Behav 63:15-22 Zeyl JN, Laberge F (2011) Multisensory signals trigger approach behaviour in the fire-bellied toad Bombina orientalis: sex differences and call specificity. Zoology 114:369-377

40 CHAPTER 2

Fig. 1 Micrixalus saxicola male displaying a bright whitish vocal sac during advertising and foot-flagging behavior

Fig. 2 Schematic diagram of the experimental set up positioned 50 cm from the focal individual. In the stream the lower box (1) serves as anchor to the upper set-up and a loudspeaker (2) connected to an MP3-player (3). Silicon hose (4) operated by the experimenter inserted through the upper box (5) to the artificial vocal sac: tip of a latex glove

(6) CHAPTER 2 41

Fig. 3 Multi-note advertisement call of Micrixalus saxicola (a–d). The spectrogram (a) of a single call (FFT method; window length: 0.005s; time step: 1000; frequency step: 1000;

Gaussian window; dynamic range: 30dB), (b) the corresponding waveform, (c) a close-up of the three indicated notes from the same male, (d) power spectrum showing the peak of the call at 4.9 kHz relative to the ambient stream noise

42 CHAPTER 2

Fig. 4 Comparison of sound pressure of advertisement calls of Micrixalus saxicola and the background noise. Shown here are estimated means (points), standard error (boxes) and

95% confidence intervals (whiskers) of the call (N = 13), abiotic noise (N = 13), abiotic noise filtered in the frequency range (4.3–5.3 kHz) of the call (N = 13) and chorus noise filtered in the same frequency range as abiotic noise. All pairwise comparisons apart from call and chorus noise filtered indicate significant differences (P < 0.001) CHAPTER 2 43

Fig. 5 Behavioral responses of Micrixalus saxicola to unimodal and multimodal playback stimuli. Statistical significant differences between responses are denoted by asterisk (** P <

0.01; * P < 0.05; Mann-Whitney U test; for each stimulus presentation N = 10)

Fig. 6 Differences in duration of Micrixalus saxicola calls in response to acoustic- and multimodal playback presentations (P = 0.001; LMM)

21 44 45

Chapter 3

DIVERGENT RECEIVER RESPONSES TO COMPONENTS OF MULTIMODAL SIGNALS IN FOOT-FLAGGING FROG SPECIES OFFER CLUES TO VISUAL SIGNAL EVOLUTION

DORIS PREININGER, MARKUS BOECKLE, MARC SZTATECSNY AND WALTER HÖDL

Plos One Received September 24 2012

Changes made to the originally submitted manuscript to fit the format of this thesis: Line numbers were deleted and figures were inserted above the respective figure caption. 46 CHAPTER 3 47

Divergent receiver responses to components of multimodal signals in foot-flagging frog species offer clues to visual signal evolution

Doris Preininger1,*, Markus Boeckle2, Marc Sztatecsny1 & Walter Hödl1

1 Department of Evolutionary Biology

2 Department of Cognitive Biology

University of Vienna, Althanstraße 14, A-1090 Vienna, Austria

* author for correspondence email: [email protected].

Abstract

Multimodal communication of acoustic and visual signals serves a vital role in the mating system of anuran amphibians. To understand signal evolution and function in multimodal signal design it is critical to test receiver responses to unimodal signal components versus multimodal composite signals. We investigated two anuran species displaying a conspicuous foot-flagging behavior in addition to or combination with advertisement calls while announcing their signaling sites to conspecifics. To investigate the conspicuousness of the foot-flagging signals we measured and compared spectral reflectance of foot webbings of

Micrixalus saxicola and Staurois parvus using a spectrophotometer. We performed behavioral field experiments using a robotic frog including an extendable leg combined with acoustic playbacks to test receiver responses to acoustic, visual and combined audio-visual stimuli. Our results indicated that the foot webbings of S. parvus were almost five times brighter than those of M. saxicola. The main response to all experimental stimuli in S. parvus was foot flagging, whereas M. saxicola responded primarily with calls but never foot flagged.

Together these across-species differences suggest that in S. parvus foot-flagging behavior is applied as a salient and frequently used communicative signal during agonistic behavior, whereas we propose it constitutes an evolutionary nascent state in ritualization of the current fighting behavior in M. saxicola. 48 CHAPTER 3

Introduction

In order to understand the evolution of multimodal signals it is fundamental to investigate receiver responses to individual signal components. The individual components and their interaction with one another can have varying effects on receivers [1,2,3]. Three primary hypotheses have been suggested for explaining the evolution of multimodal signals and providing a signal classification framework: the content-based hypothesis, the efficacy-based hypothesis and the inter-signal interaction hypothesis [4,5,6,7,8]. The content-based hypothesis relates to the message of signal components and the response they elicit in receivers and classifies the function of signal components as “redundant” (“back-up”) or

“non-redundant” (“multiple”) messages. The efficacy-based hypothesis addresses signal efficacy related to the environment, e.g. signals can either solve different transmission problems or act as a backup in varying environmental conditions. Hypothesis from inter- signal interaction assumes that the signals do not always act independently, but the presence of one signal component alters the receiver’s response to the second component and for example increases detection and discrimination.

Recent studies on multimodal signaling have focused on the role of signals in female mate choice decisions. In particular behavioral experiments in wolf spiders using visual and seismic signals during courtship have provided profound insights into the evolution and function of multimodal signaling across species [1,3,9,10]. Very little is known about multimodal signaling in male-male competition and agonistic interaction; especially how isolated signal components influence receivers remains poorly understood. Male territoriality or spacing behaviors often involve long distance signals [11], that are less suitable to experimental manipulation than signals involved in close range mate attraction. Another problem in understanding receiver response to multimodal signals comes from the fact that similar signal components have differing functions across species [2,3]. Comparing responses to multimodal signal components across species may therefore allow more general conclusions about signal function and evolution to be drawn. CHAPTER 3 49

Anuran amphibians are excellent model systems to study multimodal communication, since all anuran species performing visual displays also use acoustic signals [12]. In particular the vocal sac has been shown to simultaneously serve acoustic as well as visual roles in mate attraction or territoriality [e.g. 13,14,15,16]. The linkage of acoustic and visual signal modes to the same organ, however, makes it difficult to study the two channels independently as any change in one channel will most likely affect the other. However, some frog species perform visual displays with their feet independently of sound production known as foot flagging [12,17,18,19], making it possible to disentangle receiver response to the two signal components. How the isolated visual signal component influences male agonistic behavior has not been studied, but it was suggested that the call alerts the receiver to the subsequent foot-flagging signal in the genus Staurois [18,19,20]. Foot-flagging displays have been reported from 16 anuran species [12,19,21,22,23]. The behavior probably evolved convergently in 5 anuran families mostly inhabiting fast-flowing streams [12]. The Bornean

Rock Frog (Ranidae: Staurois parvus) and the Small Torrent Frog (Micrixalidae: Micrixalus saxicola) from the Western Ghats of India belong to different anuran families. Males of both species use a complex signaling repertoire consisting of high pitched calls, foot flagging, and tapping (foot lifting) to signal the readiness to defend perching sites against other males

[20,21,24]. Acoustic communication in the two species is not impaired by ambient low- frequency dominated stream noise, but concurrently chorusing conspecifics are suggested to constrain vocal communication in M. saxicola [24]. The conspicuously white colored foot webbings of S. parvus pose a strong contrast to the dark body coloration whereas the feet of

M. saxicola do not differ from the general body coloration as judged by the human eye.

Previous studies have demonstrated that both species respond to acoustic playbacks, however, M. saxicola only displayed foot-flags if the acoustic signal was accompanied by a visual cue of a pulsating vocal sac (Preininger et al. 2012). Additionally, M. saxicola males repeatedly attack each other with leg kicks (Preininger unpublished data) a behavior that has not been observed in S. parvus. 50 CHAPTER 3

The aim of our study was to test how isolated unimodal signal components and their multimodal interactions influence male response in M. saxicola and S. parvus. Since visual signals may not be obvious to the human eye, for instance due to our lack of sensitivity to UV light, we first measured spectral reflectance of foot webbings in both species using a spectrophotometer. We then performed behavioral field experiments for which we employed a robotic frog with an extendable leg combined with acoustic playbacks to present acoustic, visual and audio-visual multimodal stimuli to the frogs. As the tapping behavior was too complex to be performed by a robot we restricted the visual stimulus to foot flagging.

Attaching a white or a dark grey foot to the robot’s leg enabled us to manipulate the visual signal’s conspicuousness and to explore the role of signal efficacy in receiver response. By comparatively describing the visual signal components as well as the response behavior, we discuss across species differences and hypothesize that foot flagging in M. saxicola presents a nascent state in evolution of multimodal signaling.

Methods

Ethics statement

The behavioral experiments were performed without physical contact with the study animals. The experimental protocol adhered to the Animal Behaviour Society guidelines for the use of animals in research and all necessary permits were obtained for the described field studies and approved by Universiti Brunei Darussalam Research Committee, the authority responsible for the Ulu Temburong National Park (permission number:

UBD/PNC2/2/RG/1(58)) and the Centre for Ecological Sciences, Indian Institute of Science,

Bangalore and Principal Chief Conservator of Forest (Wildlife), Karnataka State Forest

Department, Government of Karnataka, the relevant regulatory bodies concerned with protection of wildlife for the Kathalekan swamp forest. (permission number: D.WL.CR-

27/2008-09).

Study sites and species CHAPTER 3 51

Staurois parvus

The Bornean Rock Frog (Staurois parvus) is a ranid frog, endemic to Borneo, recently resurrected from synonymy with S. tuberilinguis [25,26]. We studied a population of S. parvus from March 2010 - April 2010 in the Ulu Temburong National Park, Brunei

Darussalam, Borneo. The study site was at a narrow, rocky (black shale) section of the

Sungai Mata Ikan, a small freshwater stream that merges into the Belalong River close to the

Kuala Belalong Field Studies Centre (115°09´E, 4°33´N). The snout-urostyle length (SUL) and weight of the investigated population of male S. parvus averaged 21.5 mm and 0.7 g respectively [20]. Males are diurnal and perch on rocks along fast-flowing forest streams.

Their white chest and white webbing between toes of hind legs strongly contrast to their cryptic dark grey, brown dorsal body. The acoustic and visual displays are functionally linked in the genus Staurois as the call is suggested to alert the receiver to the subsequent foot- flagging signal [18,19,20].

Micrixalus saxicola

The second study species belongs to the family Micrixalidae and is endemic to the

Western Ghats in India [27]. A population of the Small Torrent Frog (Micrixalus saxicola) was investigated at the end of the monsoon season (September 2010- October 2010). Micrixalus saxicola occurs exclusively along small, fast-flowing streams within the evergreen forests

[28]. Individuals are diurnal and inhabit perennial streams characterized by low water, air and soil temperature in which they produce advertisement calls from exposed sites on rocks. Our study population was located at the Kathalekan Myristica swamp forest (14.27414°N,

74.74704°E) in the central Western Ghats, which is considered a relict forest. Males of the study population have an average SUL of 23.6 mm and a mean mass of 1.1 g and display a bright white vocal sac during vocalizing and besides foot flagging they also kick other males in male-male agonistic interactions [24].

Spectral reflectance measurements 52 CHAPTER 3

We captured 13 S. parvus and 13 M. saxicola respectively during nightly census while they were resting on leaves or rocks along the stream banks and kept them in terraria until the next day. Catching the very agile and shy frogs in streams and waterfalls is almost impossible during the day while they are active. To avoid possible color changes occurring at night, we measured spectral reflectance using an Ocean Optics Jaz spectrometer (Ocean

Optics, Dunedin, FL, USA) with integrated pulsed xenon light source (Jaz-PX) the following morning. We took 3 spectral reflectance measurements (300-700 nm) relative to a white reflectance standard (WS-1 Diffuse Reflectance Standard, Ocean Optics) of two body parts: the back as a proxy of the frog’s general body coloration and of the foot webbings of each foot. To compare the conspicuousness of the foot webbings to the general body coloration within a species and the foot webbings between the two species, we used total brightness or the intensity of the reflectance spectrum (calculated as the area under the spectral curve).

Model frog experiments

Experimental design

The experimental set up (Fig. 1A) consisted of two containers that formed a platform for stimuli presentation, a model frog, an extendable artificial leg and a loudspeaker. The larger container (7 x 18 x 11 cm) was filled with pebbles and placed in the stream where it served as an anchor for the attached smaller container (6 cm x 10 cm x 11 cm) and the loudspeaker

(Sony SRS-M 30) connected to an MP3-Player (Odys Pax). On the smaller container we placed a stationary model frog as additional visual stimulus. To make the model frog, we created a silicone cast from a preserved specimen of S. parvus and filled it with Polyurethan resin (Neukadur MultiCast 1, Altropol, Stocklsdorf, Germany). Since S. parvus and M. saxicola males have similar body size, we used identical models for all experiments but painted them with acrylics according to previously taken photographs from the respective species. Finally, a clear coat was sprayed over the models to protect the paint from water and add a realistic sheen. Under the smaller container an extendable artificial leg made of sheet metal (0.25 mm thick) was affixed. The upper part of the leg including the CHAPTER 3 53 exchangeable foot could be extended via a string by the experimenter and was pulled back automatically by a rubber band (Fig. 1A enlarged image).

Experiments and play-back stimuli

The experimental set-up was placed 50 cm from a focal male individual in the stream and the experimenter operated play-backs from a distance of 1.5 m. Experimental presentations started when the focal individual showed no signaling behavior for a period of 60 s. We presented each individual with 3 stimuli counterbalanced between individuals: 2 unimodal stimuli (acoustic/visual) consisting of either 3 calls or 3 foot flags and one multimodal stimulus (combined acoustic and visual) consisting of 3 calls and 3 foot flags. Each stimulus presentation lasted 30 s followed by a 90 s control period (Fig. 1B) followed by the next stimulus adding up to a total duration of 420 s (incl. 60 s baseline) for one experiment. The 3 advertisement calls for the acoustic stimulus had an average intensity of 75 dB at 50 cm for

M. saxicola and 70 dB at 50cm for S. parvus. The pre-recorded advertisement calls were generated using averaged call values from the studied population (M. saxicola: call duration:

2.6 s, note number: 21, mean dominant frequency: 4.6 kHz, intercall interval 7.4 s; S. parvus: call duration: 6.1 s, note number: 35, mean dominant frequency: 5.5 kHz, intercall interval 3.9 s). Each foot-flag lasted 2 s (time between raising and retracting the artificial leg; inter-signal interval 8 s). For the combined multimodal stimulus we presented an advertisement call immediately followed by a foot-flag (inter-signal interval 2 s between multimodal stimuli).

Average call parameters and foot-flagging durations were representative of our two study populations [20,24].

To test if the brightness of interdigital webbings has an influence on response frequencies, we conducted experimental presentations with individuals of both species using a white (100% reflection of light compared to the white standard) and a dark-grey (10% light reflection) artificial foot during visual and multimodal stimuli (Fig. 2). Commercial paints absorb in the UV and we added Barium sulfate (ReagentPlus, 99%, Sigma Aldrich Germany) to our acrylics to boost the UV component and achieve a more even reflection between 300 54 CHAPTER 3

(UV) and 700 nm (red). As BaSO4 increases reflection in all wavelengths, we could only add so much as to adjust the overall brightness to 100% and 10% compared to our white reflectance standard respectively.

Data collection and analysis

All trials were video recorded with a waterproof camera (Sanyo Xacti WH1) positioned on a tripod. Dorsal patterns of frogs allowed individual recognition in order to avoid multiple testing of the same individual. We analyzed frequencies of the behavior categories “calling”,

“tapping”, “foot-flagging” during stimuli presentation and control periods with the behavioral coding software Solomon Coder [29]. “Tapping” constitutes the lifting of either the right or left leg without stretching it, whereas “foot-flagging" describes the behavior of completely extending the leg above and back in an arc and bringing it back to the body side [12]. For statistical analysis we only used data from recordings in which the focal individual could be observed for the complete experimental presentation. We analyzed responses of 16 M. saxicola males; 8 playback presentations were conducted with the white foot and 8 with the dark-grey foot. In S. parvus 31 males were tested, and 14 experiments were performed with the white foot and 17 with a dark-grey foot.

To test whether the frequency of responses is dependent on species (M. saxicola and S. parvus), brightness of foot (white and dark) and/or stimuli (acoustic, visual and multimodal), we calculated zero inflated Generalized Linear Mixed Models (GLMMs) with a poisson distribution and a log link function. We used the glmmADMB package [30] within the R statistical software [31]. The glmmADMB package allows for the simultaneous modeling of random effects and the overabundance of zeros in count data (i.e. zero inflation). The response variables “call”, “tap”, “foot flag” and the sum of responses (“call”, “tap” and “foot flag”) were modeled in four model sets for dependence on predictor variables using a backward step-wise selection procedure. The global model consisted of all predictor variables (species, brightness and stimuli) and their two-way interactions (Tab. 1). We started with the global model and excluded each predictor with a significance value P > 0.1. CHAPTER 3 55

In case we encountered significant interactions between the predictor variables we split the data accordingly into subsets in order to calculate significant effects within the subset. Terms were only regarded as being significant if P < 0.05. To correct for the differences between individuals we included the nested term species (individual) as random variable for all models with the exception of models performed for the response variable foot-flag. Micrixalus saxicola displayed no foot-flagging behavior during playback presentation and only responses of the subset S. parvus were corrected with the random effect (individual). From the log likelihood of each model we calculated the small sample Akaike’s Information

Criterion (AICc) to rank the models [the model with the lowest AICc value is the best supported by the data, 32]. Models with ∆AICc ≤ 2 (difference in AICc to the best model) can be considered to have substantial support for interpretation [32]. We also calculated Akaike weights (ω) that are data-dependent, posterior model probabilities [32] and can be used to evaluate how much better a model is. In the best model predictors and interactions remained regardless of their significance and the results of pair-wise comparisons of this final modal are presented.

Results

Foot webbings of both species were significantly brighter than their backs (total brightness 12532.2 ± 717.7 compared to 630.6 ± 83.4 in S. parvus, t = 17.63, d.f. = 12, P <

0.001, and 2538.5 ± 165.5 compared to 1509.5 ± 130.9 in M. saxicola, t = -5.190, d.f. = 12, P

< 0.001). The feet of S. parvus reflected 22.8 times more light than its back compared to 1.8 times in M. saxicola and the foot webbings in S. parvus reflected almost 5 times more light than those of M. saxicola (t = -13.61, d.f. = 12, P < 0.001).

During all playback experiments, the 16 tested M. saxicola males responded by performing a total of 125 calls (79%), 34 taps (21%) but no foot flags. The 31 S. parvus males tested displayed 21 calls, 19 taps and 83 foot flags, thus a playback was predominantly responded to with foot-flagging signals (68%) rather than calls (17%) or taps 56 CHAPTER 3

(15%). Four individuals of S. parvus responded five times with combined displays which could be regarded as multimodal signal (call and simultaneously preformed foot flag). Due to the low occurrence of multimodal responses they were not analyzed separately but included to the respective response category call or foot flag.

The frequency of the sum of behavioral responses was smaller in S. parvus compared to

M. saxicola (GLMM: pair-wise comparison: ß = -0.840, SE = 0.257; z = 7.92, P = 0.001), and both species displayed more overall responses to acoustic stimuli than to visual stimuli

(GLMM: pair-wise comparison: ß = 0.640, SE = 0.177; z = 3.61, P < 0.001) and multimodal stimuli (GLMM: pair-wise comparison: ß = 0.645, SE = 0.160; z = 4.04, P < 0.001; Fig. 3A).

The frequency of tap responses did not differ between the tested species (Tab. 2). Both species responded with less tapping behavior to visual stimuli compared to acoustic stimuli

(GLMM: pair-wise comparison: ß = -0.951, SE = 0.476; z = -2.00, P = 0.046) and the acoustic stimuli tended to be more frequently answered than the multimodal stimuli (GLMM: pair-wise comparison: ß = 1.019, SE = 0.598; z = 1.71, P = 0.088; Fig. 3B).

In the global model calculated for call responses the predictors brightness and stimuli showed significant interactions and the model was split into the subsets white and dark.

Response frequency between differing stimuli in the subset showed no significant differences

(Tab. 2).

Staurois parvus performed fewer foot flags in response to visual stimuli compared with acoustic stimuli (GLMM: pair-wise comparison: ß = -0.859, SE = 0.323; z = -2.66, P = 0.007) and tended to respond more to acoustic stimuli than multimodal stimuli (GLMM: pair-wise comparison: ß = 0.583, SE = 0.302; z = 1.93, P = 0.054; Fig. 4).

Discussion

Micrixalus saxicola and S. parvus are not closely related but share a similar breeding habitat and use similar, convergently evolved multimodal signals to communicate during male-male agonistic interaction. Despite similar ecological constrains, our study showed that a number of differences exist in visual signal conspicuousness and the response to identical CHAPTER 3 57 signal stimuli between the tested species. The foot webbings of S. parvus were almost five times brighter than those of M. saxicola. The main response to all experimental stimuli in S. parvus was foot flagging, whereas Micrixalus saxicola responded more actively to all tested stimuli conditions than S. parvus and responded primarily with calls (79% of all responses) but never foot flagged.

While across-species differences are distinct, the minor within species differences in receiver behavior to the presented stimuli are not easy to interpret. Neither species responded to the stimuli with explicitly aggressive behavior such as attacking the robotic model as found in the territorial dart poison frog Allobates femoralis [13,33], or in studies on female mate choice in the túngara frog or wolf spiders [2,3]. Instead our study species responded to all stimuli with a complex set of audio-visual signals. As response behavior can become more variable with increasing signal complexity [2] , testing specific hypotheses related to multimodal signaling becomes a challenging task. Focusing on the primary response to stimuli types in the two investigated species, the call response frequency in M. saxicola did not differ between the three types of playback stimuli suggesting that the acoustic, visual and multimodal displays are of equal significance and might act as redundant signal components. However S. parvus displayed a higher number of foot flags during acoustic stimuli than visual presentations and signal frequency also tended to be less during multimodal stimuli. The primary use of foot flagging and the differences in response to the stimulus types indicate that acoustic and visual signals may be non-redundant. Previous studies on the genus Staurois suggested that the call acts as an alerting signal to the subsequent foot-flagging display which supports our findings [18,19,20]. Whereas signal response in both species showed no per se qualitative difference (no response opposed to response) as discussed for 11 taxa performing composite acoustic and visual signals

[reviewed in 34] and drawing assumptions on signal message needs further experiments.

The low levels of response to the multimodal stimulus and the lack of differences in response to the white and the grey robot foot were unexpected results. Multimodal signals are assumed to elicit equal or enhanced response in receivers compared to their unimodal 58 CHAPTER 3 components [7,8,35]. For instance female house crickets and wolf spiders were more attracted towards multimodal than unimodal male signals in mate choice experiments [3,36].

During agonistic male-male interactions receiver response to threat signals should depend on the distance between sender and receiver as well as the fighting technique of a species

[37]. Aggressive kicking behavior in M. saxicola is only effective at close range (lengths of the hind legs) to the opponent and the distance between robot and actual frogs might have been too large to elicit aggressive response. The reduced frequency of foot-flagging in S. parvus to the visual and multimodal stimuli may indicate that the robot was not always received as a threat. No or reduced response does not necessarily suggest that the visual stimulus was not perceived (independently of the stimulus coloration) but the response may be graded. As we chose a sender to receiver distance typical for our study populations, the low response could at least partly be a consequence of insufficient visual stimulus quality.

The stimulus coloration did not exactly match the color of the frogs’ feet and foot-flagging behavior of actual frogs is complex with the robot’s leg movement perhaps being too simplified to elicit a natural response. Alternatively, the increased signal directionality and localizability of the visual stimulus [20,24], may have caused the receiver to retreat rather than signal back [38,39].

Signals used during aggressive or agonistic encounters reflect the species´ fighting technique [40]. During the breeding season, male M. saxicola occur in higher densities than

S. parvus and engage in numerous close-range agonistic interactions with individuals performing acoustic and visual signals [21]. Male-male signaling is often preceded by physical attacks during which individuals kick opponents off the rocks with their hind legs

(Preininger unpublished data). We therefore suggest that foot flagging may have evolved via ritualization from aggressive kicking behavior during male combat [sensu 12]. Ritualization is predicted to be the most common process for the evolution of animal signals [41] during which cues are thought to be modified to enhance their efficacy (Bradbury and Vehrencamp

2011, Scott et al 2010). Ritualized communication signals are expected to show increased conspicuousness, redundancy, and stereotypy compared to the original cue and additional CHAPTER 3 59 alerting signal components may occur [11,40]. A signal displayed during agonistic interactions should improve communicative thereby reducing energy costs [42] and lead to lower rates of attacks and injury as shown in jumping spiders (Phidippus clarus) [43]. We never observed any kicking behavior in S. parvus and the foot-flagging signal was more salient than in M. saxicola, was displayed more frequently, and appears to be preceded by an alerting call [18,19,20]. Formal testing of the ritualization hypotheses requires a phylogenetic comparison across species with homologous behavior [44], which is lacking in foot-flagging frogs. However, given the observed differences in our two study species we propose that the kicking and foot-flagging behavior in M. saxicola constitutes an evolutionary nascent state in ritualization of a communicative foot-flagging signal.

Male-male competition and agonistic interactions have rarely been considered a significant influence on multimodal signal evolution [but see 45,46,47,48,49]. In particular hypotheses on multimodal signal function were specifically set up in the context courtship behavior and female mate choice [4,5]. Testing content or efficacy based signal hypotheses in male aggressive signals in particular when receiver response involves complex signaling behavior might be more difficult than previously thought. The option to manipulate the distance between signaler and receiver in the study of aggressive signals appears important to draw conclusions on signal content. Signal characteristics could covary with physical parameters of the sender, as described for spectral and temporal properties of the advertisement call of several anuran species [50], visual signals in lizards [51], or vibratory signals in jumping spiders [47]. Future studies should investigate visual signal-to-noise ratios of the foot-flagging display and the respective environment as well as signal characteristics in regard to size and age and explore signal function in regard to female responses. We believe that primarily a comparative across species approach will help to explain multimodal signal evolution and will promote our understanding of how environmental selection pressures and sexual selection have influenced the evolution of the currently observed signal forms and functions.

60 CHAPTER 3

Acknowledgements

We thank K. V. Gururaja, S. P. Vijayakumar and V. Torsekar for their logistic and professional help at the study site in India and K. Shanker for his scientific collaboration. We are grateful for the support and collaboration of U. T. Grafe in Brunei. A. T. Hedge and his family provided us a pleasant stay in India and the KBFCS staff in Brunei. D. Murschenhofer constructed the artificial leg. I. Rubin painted the model frogs. M. Mayerhofer, C. Nebel and

M. Schlumpp helped to analyze video data. We thank L. Highfill for very helpful comments on the manuscript.

Author Contributions

Conceived and designed the experiments: DP, MS, WH. Performed the experiments: DP.

Analyzed the data: DP, MB, MS. Wrote the paper: DP, MS.

References

1. Wilgers DJ, Hebets EA (2012) Seismic Signaling is Crucial for Female Mate Choice in a Multimodal Signaling Wolf Spider. Ethology 118: 387-397. 2. Taylor RC, Klein BA, Ryan MJ (2011) Inter-signal interaction and uncertain information in anuran multimodal signals. Current Zoology 57: 153-161. 3. Hebets EA (2008) Seismic signal dominance in the multimodal courtship display of the wolf spider Schizocosa stridulans Stratton 1991. Behavioral Ecology 19: 1250-1257. 4. Candolin U (2003) The use of multiple cues in mate choice. Biological Reviews 78: 575- 595. 5. Hebets EA, Papaj DR (2005) Complex signal function: developing a framework of testable hypotheses. Behavioral Ecology and Sociobiology 57: 197-214. 6. Johnstone RA (1996) Multiple displays in animal communication: “backup signals” and “multiple messages.”. Philosophical Transactions of the Royal Society of London B Biological Sciences 351: 329-338. 7. Partan S, Marler P (1999) BEHAVIOR:Communication Goes Multimodal. Science 283: 1272-1273. 8. Partan SR, Marler P (2005) Issues in the Classification of Multimodal Communication Signals. The American Naturalist 166: 231-245. 9. Hebets EA, Uetz GW (2000) Leg ornamentation and the efficacy of courtship display in four species of wolf spider (Araneae : Lycosidae). Behavioral Ecology and Sociobiology 47: 280-286. 10. Wilgers DJ, Hebets EA (2011) Complex courtship displays facilitate male reproductive success and plasticity in signaling across variable environments. Current Zoology 57: 175-186. 11. Krebs JR, Dawkins R (1984) Animal signals: Mindreading and manipulation. In: Krebs JR, Davies NB, editors. Behavioural ecology: An evolutionary approach. 2 ed. Sunderland, MA: Sinauer. pp. 380-402. CHAPTER 3 61

12. Hödl W, Amézquita A (2001) Visual signaling in anuran amphibians. In: Ryan MJ, editor. Anuran Communication. Washington, DC, U.S.A: Smithsonian Institution Press. pp. 121-141. 13. Narins PM, Hödl W, Grabul DS (2003) Bimodal signal requisite for agonistic behavior in a dart-poison frog, Epipedobates femoralis. Proceedings of the National Academy of Sciences of the United States of America 100 577-580. 14. Rosenthal GG, Rand AS, Ryan MJ (2004) The vocal sac as a visual cue in anuran communication: an experimental analysis using video playback. Animal Behaviour 68: 55-58. 15. Taylor RC, Klein BA, Stein J, Ryan MJ (2008) Faux frogs: multimodal signalling and the value of robotics in animal behaviour. Animal Behaviour 76: 1089. 16. Gomez D, Richardson C, Lengagne T, Plenet S, Joly P, et al. (2009) The role of nocturnal vision in mate choice: females prefer conspicuous males in the European tree frog (Hyla arborea). Proceedings of the Royal Society Biological Sciences Series B 276: 2351-2358. 17. Amézquita A, Hödl W (2004) How, when, and where to perform visual displays: The case of the Amazonian frog Hyla parviceps. Herpetologica 60: 420-429. 18. Preininger D, Boeckle M, Hödl W (2009) Communication in noisy environments II: Visual signaling behavior of male foot-flagging frogs Staurois latopalmatus. Herpetologica 65: 166-173. 19. Grafe TU, Wanger TC (2007) Multimodal Signaling in Male and Female Foot-Flagging Frogs Staurois guttatus (Ranidae): An Alerting Function of Calling. Ethology 113: 772-781. 20. Grafe TU, Preininger D, Sztatecsny M, Kasah R, Dehling JM, et al. (2012) Multimodal Communication in a Noisy Environment: A Case Study of the Bornean Rock Frog Staurois parvus. PLoS One 7: e37965. 21. Krishna SN, Krishna SB (2006) Visual and acoustic communication in an endemic stream frog, Micrixalus saxicolus in the Western Ghats, India. Amphibia-Reptilia 27: 143-147. 22. Hartmann MT, Giasson LOM, Hartmann PA, Haddad CFB (2005) Visual communication in Brazilian species of anurans from the Atlantic forest. Journal of Natural History 39: 1675-1685. 23. Vasudevan K (2001) A foot flagging frog from the Western Ghats. Cobra 44: 25-29. 24. Preininger D, Boeckle M, Freudmann A, Starnberger I, Sztatecsny M, et al. (2012) Multimodal signaling in the Small Torrent Frog (Micrixalus saxicola) in a complex acoustic environment Behavioral Ecology and Sociobiology: In press. 25. Matsui M, Mohamed M, Shimada T, Sudin A (2007) Resurrection of Staurois parvus from S. tuberilinguis from Borneo (Amphibia, Ranidae). Zoological Science 24: 101-106. 26. Arifin U, Iskandar DT, Bickford DP, Brown RM, Meier R, et al. (2011) Phylogenetic relationships within the genus Staurois (Anura, Ranidae) based on 16S rRNA sequences. Zootaxa 2744: 39-52. 27. Daniels RJR (2005) Amphibians of Peninsular India. Hyderabad: University Press (India) Private Limited. 268 p. 28. Chandran MDS, Rao GR, Gururaja KV, Ramachandra TV (2010) Ecology of the swampy relic forests of Kathalekan from Central Western Ghats. Bioremediation, Biodiversity and Bioavailability 4: 54-68. 29. Péter A (2011) Solomon Coder (version beta 11.01.22): A simple solution for behavior coding. 30. Fournier DA, Skaug HJ, Ancheta J, Ianelli J, Magnusson A, et al. (2012) AD Model Builder: using automatic differentiation for statistical inference of highly parameterized complex nonlinear models. Optimization Methods and Software 27: 233-249. 31. R Core Team (2012) R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. 32. Burnham KP, Anderson DR (2010) Model Selection and Multi-Model Inference: A Practical Information-Theoretic Approach: Springer London, Limited. 62 CHAPTER 3

33. Narins PM, Grabul DS, Soma KK, Gaucher P, Hödl W (2005) Cross-modal integration in a dart-poison frog. Proceedings of the National Academy of Sciences of the United States of America 102: 2425-2429. 34. Otovic P, Partan S (2009) Multimodal signaling in animals. Encyclopedia of Neuroscience (Ed by L R Squire): Oxford: Academic Press. pp. 1095-1105. 35. Rowe C (1999) Receiver psychology and the evolution of multicomponent signals. Animal Behaviour 58: 921-931. 36. Stoffer B, Walker SE (2012) The use of multimodal communication in mate choice decisions by female house crickets, Acheta domesticus. Animal Behaviour. 37. Szamado S (2008) How threat displays work: species-specific fighting techniques, weaponry and proximity risk. Animal Behaviour 76: 1455-1463. 38. Holt MM, Southall BL, Insley SJ, Schusterman RJ (2010) Call directionality and its behavioural significance in male northern elephant seals, Mirounga angustirostris. Animal Behaviour 80: 351-361. 39. Van Dyk DA, Evans CS (2008) Opponent assessment in lizards: examining the effect of aggressive and submissive signals. Behavioral Ecology 19: 895-901. 40. Bradbury JW, Vehrencamp SL (2011) Principles of Animal Communication, Second Edition. Sunderland: Sinauer Press. 41. Scott-Phillips TC, Blythe RA, Gardner A, West SA (2012) How do communication systems emerge? Proceedings of the Royal Society Biological Sciences Series B 279: 1943-1949. 42. Viera VM, Viblanc VA, Filippi-Codaccioni O, Cote SD, Groscolas R (2011) Active territory defence at a low energy cost in a colonial seabird. Animal Behaviour 82: 69-76. 43. Elias DO, Botero CA, Andrade MCB, Mason AC, Kasumovic MM (2010) High resource valuation fuels “desperado” fighting tactics in female jumping spiders. Behavioral Ecology 21: 868-875. 44. Scott JL, Kawahara AY, Skevington JH, Yen S-H, Sami A, et al. (2010) The evolutionary origins of ritualized acoustic signals in caterpillars. Nature Communications 1. 45. Hughes M (1996) The function of concurrent signals: visual and chemical communication in snapping shrimp. Animal Behaviour 52: 247-257. 46. Partan S (2004) Multisensory animal communication. In: Calvert G, Spence C, Stein BE, editors. The handbook of multisensory processes. Cambridge, MA: MIT Press. pp. 225–240. 47. Elias DO, Kasumovic MM, Punzalan D, Andrade MCB, Mason AC (2008) Assessment during aggressive contests between male jumping spiders. Animal Behaviour 76: 901-910. 48. Morris MR, Ryan MJ (1996) Sexual difference in signal-receiver coevolution. Animal Behaviour 52: 1017-1024. 49. Borgia G, Coleman SW (2000) Co-option of male courtship signals from aggressive display in bowerbirds. Proceedings of the Royal Society Biological Sciences Series B 267: 1735-1740. 50. Gerhardt HC, Huber F (2002) Acoustic communication in Insects and Anurans: Common problems and diverse solutions. Chicago: University of Chicago Press. 51. Lappin AK, Brandt Y, Husak JF, Macedonia JM, Kemp DJ (2006) Gaping displays reveal and amplify a mechanically based index of weapon performance. American Naturalist 168: 100-113.

CHAPTER 3 63

Figure 1. Schematic of the experimental set up and stimuli presentation. (A) The set up was positioned 50 cm from the focal individual. In the stream the lower box (1) serves as anchor to the upper set-up and a loudspeaker (2) connected to an MP3-player (3). A string (4) operated by the experimenter inserted through the upper box (5) stretched the artificial leg behind a model frog (6). A rubber band (7) automatically pulled back the leg and the attached foot (8). (B) After a 60s baseline of no response the stimuli (S; acoustic, visual and multimodal) were presented for 30s followed by a 90s control period. Stimuli conditions were counterbalanced between positions S1, S2 and S3.

64 CHAPTER 3

Figure 2. Brightness reflections of the artificial model feet and of foot webbings and backs of the study species. Artificial white foot (grey dotted line) and dark foot (black dotted line) used in the experimental playback presentations; Staurois parvus feet (grey solid line) and back

(grey dashed line); Micrixalus saxicola feet (black solid line) and back (black dashed line). N

= 13 in both species.

CHAPTER 3 65

Figure 3. Comparison of response frequency of signal behaviors of Micrixalus saxicola

(dashed lines) and Staurois parvus (solid lines) between acoustic, visual or multimodal stimuli. (A) sum (call, tap and foot flag) response of white and dark foot playback presentations; (B) tap response of white and dark foot playback presentations; call response of (C) white foot and (D) dark foot playback presentations. Statistical significant response frequency differences between species are denoted by asterisk (** P < 0.01; *** P < 0.001), between stimuli the values without the same superscript letter (a, b) differ significantly at P <

0.01 (for details see Table 2).

66 CHAPTER 3

Figure 4. Comparison of foot-flagging response of Staurois parvus between acoustic, visual and multimodal stimuli. Values without the same superscript letter (a, b) differ significantly at

P < 0.01.

20 CHAPTER 3 67

Table 1. Backward step-wise model selections obtained from Generalized Lineal Mixed

Models to explain the frequency of single response behaviors (call, tap, foot flag) and their sum as function of species (Micrixalus saxicola, Staurois parvus), artificial foot brightness

(dark, white), stimuli (acoustic, visual, multimodal) and their interactions. AICc based model rankings are shown. The final models are presented in bold. 68 CHAPTER 3

Variable Random Factor Subset Model AICc ∆AICc ω Sum (Species(Individual)) Species + Brightness + Stimuli + Species:Brightness + Species:Stimuli + Brightness:Stimuli (Full model) 517.41 10.00 0.0033 Species + Brightness + Stimuli + Species:Brightness + Species:Stimuli 513.24 5.83 0.0269 Species + Brightness + Stimuli + Species:Brightness + Brightness:Stimuli 514.95 7.55 0.0114 Species + Brightness + Stimuli + Species:Stimuli + Brightness:Stimuli 514.99 7.58 0.0112 Species + Brightness + Stimuli + Species:Brightness 511.92 4.51 0.0520 Species + Brightness + Stimuli + Species:Stimuli 510.89 3.48 0.0869 Species + Brightness + Stimuli + Brightness:Stimuli 512.60 5.19 0.0370 Species + Stimuli + Species:Stimuli 508.59 1.18 0.2752 Species + Stimuli 507.41 0 0.4961 Call (Species(Individual)) Species + Brightness + Stimuli + Species:Brightness + Species:Stimuli + Brightness:Stimuli (Full model) 323.59 Individual White Species + Stimuli + Species:Stimuli 147.78 -0.20 0.5248 Species + Stimuli 147.98 0.00 0.4752 Dark Species + Stimuli + Species:Stimuli 172.31 7.12 0.0217 Species + Stimuli 167.70 2.50 0.2177 Species 165.20 0 0.7606 Tap (Species(Individual)) Species + Brightness + Stimuli + Species:Brightness + Species:Stimuli + Brightness:Stimuli (Full model) 175.22 10.37 0.0029 Species + Brightness + Stimuli + Species:Brightness + Species:Stimuli 170.53 5.67 0.0300 Species + Brightness + Stimuli + Species:Brightness + Brightness:Stimuli 171.68 6.83 0.0169 Species + Brightness + Stimuli + Species:Stimuli + Brightness:Stimuli 172.96 8.11 0.0089 Species + Brightness + Stimuli + Species:Brightness 168.09 3.23 0.1017 Species + Brightness + Stimuli + Species:Stimuli 168.16 3.31 0.0979 Species + Brightness + Stimuli + Brightness:Stimuli 169.61 4.76 0.0474 Species + Brightness + Stimuli 216.73 51.88 0.0000 Species + Stimuli + Species:Stimuli 167.46 2.60 0.1394 Species + Brightness + Species:Brightness - - - Brightness + Stimuli + Brightness:Stimuli 169.81 4.96 0.0429 Species + Stimuli 164.85 0 0.5121 Brightness + Stimuli 211.60 46.75 0.000 Stimuli 215.77 50.91 0.000 Foot flag Individual S. parvus Brightness + Stimuli + Brightness:Stimuli (Full model) 245.08 4.84 0.0631 Brightness + Stimuli 242.53 0.29 0.2266 Stimuli 240.24 0 0.7103 CHAPTER 3 69

Table 2. Pair-wise comparisons of predictors and interactions of final models based on stepwise model selections (see Tab. 1). Estimates are given relative to the intercept.

Significant differences between species (Micrixalus saxicola (M.s.), Staurois parvus (S.p)) and/or stimuli (acoustic (A), visual (V) and multimodal (M)) in the frequency of single behavioral responses (call, tap, foot flag) or their sum are marked with asterix.

Variable Subset Coefficients (Reference level) Estimate SE z-Value P-Value Sum Intercept 1.653 0.209 7.92 2.3e-15 *** Species S.p. (M.s.) -0.840 0.257 -3.27 0.00108 ** Stimulus A (M) 0.645 0.160 4.04 5.3e-05 *** Stimulus V (A) -0.640 0.177 -3.61 0.00031 *** Stimulus M (V) -0.006 0.197 -0.03 0.9777 Call White Intercept 1.418 0.226 6.27 3.7e-10 *** Species S.p. (M.s.) -1.521 0.485 -3.14 0.0017 ** Stimulus A (M) 0.332 0.317 1.05 0.2953 Stimulus V (A) -0.727 0.533 -1.36 0.1731 Stimulus M (V) 0.395 0.533 0.74 0.4583 Species S.p. (M.s.): Stimulus A (M) 1.042 0.899 1.16 0.2462 Species S.p. (M.s.): Stimulus V (A) -12.326 238.290 -0.05 0.9587 Species S.p. (M.s.): Stimulus M (V) 11.284 238.290 0.05 0.9622 Dark Intercept 0.762 0.465 1.64 0.1 Species S.p. (M.s.) -3.045 0.731 -4.17 3.1e-05 *** Tap Intercept 0.953 0.858 1.11 0.267 Species S.p. (M.s.) -1.334 0.827 -1.61 0.107 Stimulus A (M) 1.019 0.598 1.71 0.088 . Stimulus V (A) -0.951 0.476 -2.00 0.046 * Stimulus M (V) -0.069 0.697 -0.10 0.922 Foot flag S. parvus Intercept 0.788 0.178 4.43 9.3e-06 *** Stimulus A (M) 0.583 0.302 1.93 0.0540 . Stimulus V (A) -0.859 0.323 -2.66 0.0079 ** Stimulus M (V) 0.277 0.360 0.77 0.442

70 71

Chapter 4

THE CONSERVATION BREEDING OF TWO FOOT-FLAGGING FROG SPECIES FROM BORNEO STAUROIS PARVUS AND STAUROIS GUTTATUS

DORIS PREININGER, ANTON WEISSENBACHER, THOMAS WAMPULA AND WALTER HÖDL

Amphibian and Reptile Conservation 5(3):45-56(e51) Received 12 May 2012; Accepted 20 June 2012; Published 07 September 2012

CHAPTER 4 73

Copyright: © 2012 Preininger et al. This is an open-access article distributed under the terms of the Creative Com- mons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided Amphibian and Reptile Conservation 5(3):45-56. the original author and source are credited.

7KHFRQVHUYDWLRQEUHHGLQJRIWZRIRRWÀDJJLQJIURJVSHFLHV from Borneo, Staurois parvus and Staurois guttatus

1,3Doris Preininger, 2Anton Weissenbacher, 27KRPDV:DPSXODDQG1:DOWHU+|GO

1Department of Evolutionary Biology, University of Vienna, Althanstraße 14 A-1090 Vienna, AUSTRIA 2Vienna Zoo, Maxingstraße 13B A-1130 Vienna, AUSTRIA

Abstract.—7KH%RUQHDQIURJVRIWKHJHQXVStauroisOLYHH[FOXVLYHO\DORQJIDVWÀRZLQJFOHDUZDWHU UDLQIRUHVWVWUHDPVDQGDUHIDPRXVIRUGLVSOD\LQJDYDULHW\RIYLVXDOVLJQDOVLQFOXGLQJIRRWÀDJJLQJ 7KHLUH[WUDRUGLQDU\EHKDYLRUDQGWKHFRQWLQXHGORVVRIWKHLUQDWXUDOKDELWDWGXHWRGHIRUHVWDWLRQDQG VXEVHTXHQWSROOXWLRQPDNHWKHPDJURXSRIWDUJHWVSHFLHVIRUFDSWLYHEUHHGLQJDVZHOODVEHKDY- LRUDOUHVHDUFK7KH9LHQQD=RRKDVSLRQHHUHGLQWKHGHYHORSPHQWRIDUHVHDUFKDQGFRQVHUYDWLRQ SURMHFWIRUS. parvus and S. guttatus:HLPSOHPHQWHGWZREUHHGLQJDQGUHVHDUFKDUHQDVRIIHU- LQJDQDUWL¿FLDOZDWHUIDOODQGGLIIHUHQWRSWLRQVIRUHJJGHSRVLWLRQLQDELRVHFXUHFRQWDLQHUIDFLOLW\ 7ZRPRQWKVDIWHULQWURGXFLQJWKHIURJVZHREVHUYHGDPSOHFWDQWSDLUVDQGWKH¿UVWWDGSROHVRIS. parvus and S. guttatus7KH9LHQQD=RRLVWKH¿UVW]RRZRUOGZLGHWKDWKDVVXFFHHGHGLQEUHHGLQJ IRRWÀDJJLQJIURJVSHFLHVDQGPHDQZKLOHKDVUHFRUGHGRYHUWDGSROHVDQGDWOHDVWMXYH- QLOHV2QHRIWKHPRVWVWULNLQJREVHUYDWLRQVKDVEHHQWKHXVHRIIRRWÀDJJLQJVLJQDOVLQUHFHQWO\ PHWDPRUSKRVHGS. parvus7KLVFRUURERUDWHVRXUDVVXPSWLRQWKDW³IRRWÀDJJLQJ´LVHPSOR\HGDV LQWUDVSHFL¿FVSDFLQJPHFKDQLVP7KHEUHHGLQJVXFFHVVRIWZRStauroisVSHFLHVDWWKH9LHQQD=RR FDQKHOSLQVSHFLHVFRQVHUYDWLRQDVLWLQFUHDVHVRXUNQRZOHGJHRQFRQGLWLRQVQHFHVVDU\WREUHHG WURSLFDOVWUHDPGZHOOLQJDQXUDQVSHFLHVIRXQGWREHSDUWLFXODUO\WKUHDWHQHGLQQDWXUH)XUWKHUPRUH WKHFDSWLYHFRORQ\SURYLGHVUHVHDUFKFRQGLWLRQVWREHWWHUXQGHUVWDQGWKHUROHRI³IRRWÀDJJLQJ´DV DYLVXDOVLJQDOFRPSRQHQWLQDQXUDQFRPPXQLFDWLRQ

Key words. Amphibia, anura, bio-secure management, conservation research, ex situ breeding

&LWDWLRQ3UHLQLQJHU':HLVVHQEDFKHU$:DPSXOD7+|GO:7KHFRQVHUYDWLRQEUHHGLQJRIWZRIRRWÀDJJLQJIURJVSHFLHVIURP%RUQHRStaurois parvus and Staurois guttatus. Amphibian and Reptile Conservation 5(3):45-56(e51).

,QWURGXFWLRQ species (Stuart et al. 2004). Deforestation of natural habitats increases siltation and chemical pollution in streams. Few stream-dwelling Bornean species are able to survive Amphibian species are declining in many parts of the LQKDELWDWVPRGL¿HGIRUKXPDQXVH ,QJHUDQG6WXHELQJ ZRUOG2QDYHUDJHRIDPSKLELDQVDUHFODVVL¿HGDV 2005). A recent study carried out in Brunei demonstrated Threatened on the International Union of Conservation that deforestation due to road construction enabled Lim- of Nature (IUCN) Red List. The extinction risk in South nonectes ingeri to migrate more than 500 m into primary East Asia still increases (Hoffmann et al. 2010). Only re- forest, which posed a potential threat to native amphibian cently an Amphibian Conservation Action Plan has been assemblages (Konopik 2010). Inger and Stuebing (2005) developed, which states important priorities for relevant mentioned an increase of the Giant river frog (Limno- amphibian research and conservation. Understanding the nectes leporinus) along silted streams of logged areas cause of decline, assessing changing diversity and im- and a simultaneous decrease in some species of Torrent plementing long-term conservation programs are some frogs (Meristogenys spp.). About half the frog species in of the immediate interventions necessary to conserve Southeast Asia are restricted to riparian habitats and de- amphibians (Gascon et al. 2007). Zoo-based amphibian velop in streams (Inger 1969; Zimmerman and Simberl- research and conservation breeding programs facilitat- off 1996). Most anuran stream-side communities in Bor- ing ex situ and in situ conservation of amphibian species neo are known to breed in clear, turbulent water and are have been established for a wide range of species over DEVHQWLQVWUHDPVZLWKVLOWERWWRPVWKDWDUHODFNLQJULIÀHV the last decades (Browne et al. 2011; Gagliardo et al. and torrents (Inger and Voris 1993). The heterogeneity of 2008; Lee et al. 2006; McFadden et al. 2008). riparian habitats in pristine rainforests results in reoccur- In South East Asia, habitat loss and destruction is one ULQJVWUHDPDVVHPEODJHVDQGKDELWDWVSHFL¿FDGDSWDWLRQV of the main causes for the rapid decline of amphibian (Keller et al. 2009).

&RUUHVSRQGHQFHEmail: [email protected]

Amphib. Reptile Conserv. | http://redlist-ARC.org 045 September 2012 | Volume 5 | Number 3 | e51 74 CHAPTER 4

Preininger et al.

Figure 1. Male and female Staurois guttatus in amplexus resting at a waterfall. Image by M. Böckle.

Many stream living anuran species in Borneo show unknown from S. parvus, though given the microhabitats morphological and behavioral adaptations to torrential of the adults tadpoles probably live in currents along the streams and waterfalls. For example, the tadpoles of stream. Staurois guttatus tadpoles, however, have been Huia cavitympanum and of all species of the genus Meri- found in leaf litter in side pools of streams (Haas and stogenys have large abdominal suckers specialized for a 'DV VLPLODUWRDQXQLGHQWL¿HG%RUQHDQWDGSROHRID life in currents (Haas and Das 2012). The adult males of ranid genus with slender body shape and nearly pigment- M. orphnocnemis use high frequency calls to communi- less skin resembling neotropical centrolenid larvae (In- cate in noisy stream environments (Boeckle et al. 2009; ger and Wassersug 1990). Staurois parvus has recently Preininger et al. 2007). An extraordinary spectral adap- been resurrected from the synonym with S. tuberilinguis tation to enhance the signal-to-noise ratio has also been $UL¿QHWDO0DWVXLHWDO 7KHWDGSROHVRI reported in Huia cavitympanum, in which males call in S. tuberilinguis, reported by Malkmus et al. (1999), ex- a band of ultrasonic frequencies (Arch et al. 2008). In hibit a fossorial life in leaf litter at the margins of forest WKHYLFLQLW\RIZDWHUIDOOVDQGIDVWÀRZLQJVWUHDPVVSH- streams. The IUCN Red List categorizes S. tuberilinguis cies of the genus Staurois display an exceptional behav- DV³1HDU7KUHDWHQHG´ZLWKDGHFUHDVLQJSRSXODWLRQWUHQG LRUWHUPHG³IRRWÀDJJLQJ´ *UDIHHWDO*UDIHDQG (Inger et al. 2004), and S. parvus and S. guttatus are listed Wanger 2007; Preininger et al. 2009). The conspicuous DV³'DWD'H¿FLHQW´ ,8&1 visual display mainly observed in tropical anuran spe- ,QLQOLJKWRIWKH³

Amphib. Reptile Conserv. | http://redlist-ARC.org046 September 2012 | Volume 5 | Number 3 | e51 CHAPTER 4 75

Conservation breeding success in Staurois parvus and Staurois guttatus

Figure 2. A male of Staurois parvusGLVSOD\LQJWKHZKLWHLQWHUGLJLWDOZHEELQJGXULQJIRRWÀDJJLQJEHKDYLRU7KHYLVXDOVLJQDOVDUH mainly employed during male-male agonistic interactions. Image by D. Preininger.

Methods that of females 50.1 ± 0.3 mm (n = 6) and 9.74 ± 0.2 g (n = 6) (Preininger et al., data not shown). Study species Individuals were collected with permission of the Brunei Museums Department. In May 2010 we collected 20 individuals (ten pairs) of the species S. parvus and S. guttatus in the Ulu Tembu- Ex situ breeding facility rong National Park, Brunei Darussalam, Borneo. Frogs were located at narrow, rocky (black shale) sections In the Vienna Zoo two connected bio-secure containers, of the Sg. Anak Apan and Sg. Mata Ikan (Fig. 3), two fully isolated from other facilities were implemented as small freshwater streams that merge into the Belalong the research complex for the animals (Fig. 4). The use River close to the Kuala Belalong Field Studies Centre of converted shipping containers for the ex situ breeding (115°09´E, 4°33´N). Staurois parvus is a ranid frog, en- and management of amphibians was pioneered by Gerry demic to Borneo. Males are diurnal and perch on rocks Marantelli at the Amphibian Research Centre (ARC) in DORQJIDVWÀRZLQJIRUHVWVWUHDPV7KHLUZKLWHFKHVWDQG Melbourne, Australia. The Vienna Zoo has tested speci- webbing between the toes of the hind legs strongly con- men (including S. parvus and S. guttatus) for infection trast to their cryptic dark grey, brown dorsal body. The with the chytrid fungus and no positives were detected. snout-urostyle length and weight of the investigated At the start of the project we kept individuals in pairs in population of male S. parvus averaged 21.5 ± 0.5 mm medium sized terraria (50 × 60 × 70 cm) in the container (n = 13) and 0.7 ± 0.05 g (n = 13) (Grafe et al. 2012) facility that contained some tree branches, plants, stones, and of females 29.5 ± 1.8 mm (n = 5) and 1.7 ± 0.2 g DQGÀRZLQJZDWHUZKLFKUDQRYHUSRWVKHUG:HDOVREXLOW (n = 5) (Preininger et al., data not shown). The closely a research arena (150 × 120 × 100 cm) for behavioral related species S. guttatus occurs throughout Borneo. It experiments that we converted into a breeding arena in was previously known as Staurois natator (Inger and Tan 2011 (Fig. 5) to improve space requirements because 1996), a name still used for populations in the Philip- neither of the species had reproduced in their original pines (Iskandar and Colijn 2000). Males of this diurnal terraria. We implemented a controllable waterfall with VSHFLHVSHUFKRQURFNVDQGEUDQFKHVDORQJIDVWÀRZLQJ VHYHUDO VPDOOHU FDVFDGHV FUHDWLQJ DUHDV RI ÀRZLQJ DQG mountain streams. Females were found 10-50 m away dripping water that additionally increased humidity lev- from the river under overhanging rock formations and els. Small burrows, ledges, and perching sites were built tree branches. The snout-urostyle length and weight ± SE out of foamed polystyrene. Similar to the smaller ter- of the investigated population of male S. guttatus aver- rariums we added plants with large leaves (Monstera sp., aged 33.6 ± 0.4 mm (n = 14) and 2.69 ± 0.07 g (n = 14), Philodendron sp., Spathiphyllum sp., Dieffenbachia sp.,

Amphib. Reptile Conserv. | http://redlist-ARC.org047 September 2012 | Volume 5 | Number 3 | e51 76 CHAPTER 4

reininger et al.

Figure 3. A waterfall habiat of Staurois guttatusDWWKH6XQJDL0DWD,NDQ ³)LVK(\H´5LYHU LQWKH7HPERURQJ'LVWULFWLQ%UXQHL Borneo. Image by D. Preininger.

Aglaonema sp., Scindapsus sp., and others) as nightly resembled the natural habitat temperature (Fig. 6). Rela- resting sites. We incorporated a self-built rain and mist- tive humidity ranged from 95% to 100%. For a period of ing system to simulate rainy and dry periods. The wa- 14 days we simulated a dry period with no rain and de- WHUDUHDZKLFKFRYHUHGWKHHQWLUHÀRRURIWKHWHUUDULXP creased water levels (10 cm), followed by a 14 day rainy ZDV¿OOHGZLWKJUDYHORIGLIIHUHQWJUDLQVL]HVDQGODUJHU period with four hours daily rainfall (7-8am and 5-8pm), pebbles that provided perching sites and interstitial spac- elevated water levels (15 cm) and an increased quantity es. We further installed two smaller glass containers (30 RIZDWHUÀRZLQJRYHUWKHZDWHUIDOO7KLVSURFHGXUHZDV × 30 × 30 cm), one placed directly under the waterfall repeated with the intervals between the dry and rainy pe- PLPLFNLQJDFRQVWDQWO\ÀXVKHGSRROZLWKODUJHVWRQHV riods reduced to seven days, and rain periods adjusted and the other containing sand, dead leaves, and standing to different times of day (e.g., 5-10pm and no morning water, as found in side ponds of waterfalls. A mixture UDLQ :H DOVR SOD\HG EDFN FRQVSHFL¿F DGYHUWLVHPHQW RI RVPRVLVSXUL¿HG ZDWHU DQG GULQNLQJ ZDWHU DYHUDJH FDOOVUHFRUGHGLQWKH¿HOGGXULQJSHDNDFWLYLW\SHULRGV conductivity = 9 S/cm, pH = 7.2) was discharged via (9-11am and 3-5pm). WKHZDWHUIDOODQGGUDLQHGLQWRDQH[WHUQDO¿OWHUUHVHUYRLU Adult frogs were fed with gut-loaded House crickets which created a slow current in the main water area. As (Acheta domesticus), Firebrat (Thermobia domestica), light source we used a metal-halide lamp (HIT-DE 70 DQGEORZÀLHV Lucilia sp.); tadpoles received algae tab- Watt [Daylight]) and placed several plastic boards on top OHWV¿VKIRRGÀDNHVDQG¿VK¿OHWWKHGLHWRIPHWDPRU- of the terrarium to mimic canopy coverage. Individuals phosed frogs consisted of Drosophila sp. and Collem- were housed under 12-hour light, 12-hour dark cycles. bola. All feeder insects were dusted with a vitamin and :HSODFHG¿YHSDLUVRIS. parvus into the arena. From mineral mixture (Vitakalk, Korvimin or Nekton MSA). then on individuals could only be counted at night when Tadpoles were photographed in petri-dishes on graph perching on leaves, while frogs rested in the many hiding paper and snout-vent length (SVL) and Gosner stage places during the day. (Gosner 1960) derived from the photos. We measured A similar facility (150 × 150 × 150 cm) was construct- SVL and body mass of juvenile S. parvus with a sliding ed for S. guttatus, however the water area did not contain caliper to the nearest 0.1 mm, and a digital mini scale to DGGLWLRQDODUWL¿FLDOSRROVRUSRQGVDQGWKHZDWHUIDOOZDV the nearest 0.01 g. Tadpole specimens of various stages amended with several tree branches. Temperature in both of S. parvus were deposited at the Austrian Natural His- facilities averaged 25 °C (range: 22-27 °C) and closely tory Museum (Staurois parvus larvae: NHMW 39337).

Amphib. Reptile Conserv. | http://redlist-ARC.org04 September 2012 | Volume 5 | Number 3 | e51 CHAPTER 4 77

Conservation breeding success in Staurois parvus and Staurois guttatus

Figure 4. The bio-secure container facility a modern Noah´s Ark, which houses Staurois guttatus and S. parvus at the Vienna Zoo Schönbrunn. Image by D. Preininger.

5HVXOWV

Staurois parvus raised for approximately 30 days and afterwards released DWDQDUWL¿FLDOZDWHUIDOOLQWKH5DLQIRUHVWKRXVHRIWKH]RR 2Q2FWREHUZHREVHUYHGWKH¿UVWWKUHHWDGSROHV (Fig. 9), where the establishment of a semi-wild popu- of S. parvus during an evening census of adult individu- lation is intended. The metamorphs have dark green or DOVLQWKHJUDYHORIWKHVORZÀRZLQJFXUUHQWDUHDRIWKH black spots and small tuberculi on the dorsal side, the WHUUDULXP:KHQDWDGSROHFRXOG¿UVWEHFDSWXUHGLWZDV latter eponymous for the closely related species S. tuberi- in Gosner stage 25 and measured 11.2 mm in total length linguis. They measured 11.8 mm (mean SVL, SD ± 0.8, n (SVL: 3.3 mm, n = 1) and was completely transparent = 20) and had a body mass of 0.12 g (SD ± 0.03, n = 20). (Fig. 7). Due to the transparency of the body, the organs Due to the high reproductive success we recently al- and blood vessels shined through the skin and the body lowed disturbance at the setup in order to search for egg- was of reddish appearance. The highly photophobe in- deposition sites. So far, we have discovered two clutches dividuals colonized the interstitial spaces of the gravel of eggs that were attached under big stones in the slow- area. More tadpoles staged 26-28, captured 24 days later, ÀRZLQJZDWHUFXUUHQW6XUSULVLQJO\ZLWKUHVSHFWWRWKH measured ca. 21 mm in total lengths (SVL: 6 mm, n = large tadpole numbers in the project, those two clutches 1) and the body and tail were covered with dorsal black contained only 14 and 26 eggs, respectively. The survival spots. After complete toe development (> stage 38) in- rate of 120 separated tadpoles (tank A: n = 40, tank B: dividuals showed a brown coloration with green irides- n = 80) was 87% (tank A: n = 34, 85%; tank B: n = 71, cence and a yellow iris, as seen in adults. At this stage, 70 88.8%). Presently, we house over 200 tadpoles, 6-10 ju- GD\VDIWHUWKH¿UVWVLJKWLQJLQGLYLGXDOOHQJWKZDVPP veniles and nine adults in the breeding facility. (SVL: 12 mm, n = 1). At the end of metamorphosis the Metamorphosed frogs were placed into separate ter- dorsal coloration of individuals turned into bright green raria, only hours after leaving the water, and were imme- (Fig. 8). GLDWHO\REVHUYHGWRGLVSOD\IRRWÀDJJLQJEHKDYLRU )LJ 7KH¿UVWPHWDPRUSKRVHGS. parvus left the water on 7KH\RXQJIURJVSHUIRUPHGFRPSOHWHIRRWÀDJVLQ 30 January 2012 (SVL: 13 mm, tail-length: 6 mm), 104 which the leg is raised and the toes are spread as observed GD\V DIWHU ZH REVHUYHG WKH ¿UVW WDGSROHV 7R GDWH ZH in adult individuals. Interdigital webbings were colored house 285 froglets in separate terraria in the bio-secure transparent grey and did not exhibit the contrasting white container, over 600 tadpoles and 180 juveniles have been coloration as seen in adults.

Amphib. Reptile Conserv. | http://redlist-ARC.org04 September 2012 | Volume 5 | Number 3 | e51 78 CHAPTER 4

reininger et al.

amphibians have gained global support and resulted in increased conservation efforts for many threatened species (Browne et al. 2011). Information on natural history, reproduction modes, and behavior of anurans is important to determine and protect key-habitats. The tadpoles of S. guttatus and S. parvus colonized the K\SRUKHLF LQWHUVWLWLDO LQ WKH VORZÀRZLQJ FXUUHQW DUHDV in the breeding facility, which supports our assumption that the larvae develop in fresh water streams or adjacent SRROV RI IDVWÀRZLQJ PRXQWDLQ VWUHDPV DQG ZDWHUIDOOV On two occasions we found eggs of S. parvus in underwater gaps between larger rocks and the subjacent grav- HO RI RXU EUHHGLQJ WHUUDULXP 1HLWKHU LQ WKH DUWL¿FLDOO\ ÀXVKHGSRROZLWKODUJHSHEEOHVQRULQWKHVDQGDQGOHDI ¿OOHGFRQWDLQHUPLPLFNLQJDVLGHSRRORIWKHZDWHUIDOO tadpoles or eggs could be observed. In a stream-dwelling, IRRWÀDJJLQJVSHFLHVIURP%UD]LO Hylodes dactylocinus) males dig underwater chambers prior to courtship and eggs are deposited on the sandy bottom between rocks along streams (Narvaes and Rodrigues 2005). Another diurnal species (Micrixalidae: Micrixalus saxicola) dis- SOD\V IRRWÀDJJLQJ VLJQDOV DQG OLYHV DORQJ SHUHQQLDO streams in the Western Ghats, India. Females of M. saxicola dig under-water cavities with the hind legs in gravel DUHDVRIÀRZLQJVWUHDPVZKLOHLQDPSOH[XVZLWKDPDOH or before courtship (Gururaja 2010; D. Preininger, pers. observ.). Although we did not observe S. parvus males or females digging under-water chambers, we assume that VXI¿FLHQWJDSVEHWZHHQURFNVFRXOGSURYLGHVLPLODUSUR- Figure 5. Ex situ breeding facility designed to offer different tection from predators. We observed amplectant pairs at egg deposition sites (described in detail in the Methods sec- the study site in Brunei to repeatedly move up the stream tion). Image by D. Preininger. only to dive back into pools at the bottom of cascades and Staurois guttatus smaller waterfalls over a period of 1-2 days. This behavior could indicate either the search for suitable deposition 7KH ¿UVW WDGSROHV RI S. guttatus were observed on 20 sites or the deposition of several clutches. March 2012, approximately 11 days after observing a pair in amplexus. In the estimated development stage 23-24, 36 days after discovery, the tadpoles had a mean length of 30 mm (8 mm SVL, range: 7-9 mm; 22 mm tail length, range: 21-24 mm, n = 5). At this stage, the dorsal body and tail was a light brown color and the body was transparent with a grey iridescence (Fig. 11). A darker dorsal line ran from the top of the head to the tip of the tail and a ventral line could be observed on both sides of the tail. So far we have moved 76 tadpoles to a separate aquarium and approximately 50 are housed in the breeding facility.

'LVFXVVLRQ

The combined efforts of members of the Vienna Zoo, Figure 6. Comparison of temperatures and relative humid- University of Vienna, and the Universiti of Brunei Da- ity measured for a period of three weeks in the natural habitat russalam have established a research and conservation in Brunei (2010) and the breeding facility in the Vienna Zoo SURMHFW WKDW VXFFHHGHG WR EUHHG WKH IRRWÀDJJLQJ IURJV (2012). Solid lines represent air temperature, dotted line water Staurois guttatus and S. parvus ex situ. Zoo-based re- temperature, and dashed lines denote relative humidity in the search and conservation breeding programs focusing on respective habitat.

Amphib. Reptile Conserv. | http://redlist-ARC.org050 September 2012 | Volume 5 | Number 3 | e51 CHAPTER 4 79

Conservation breeding success in Staurois parvus and Staurois guttatus

7KHGLYHUVHO\VWUXFWXUHGDUWL¿FLDOKDELWDWLQWKHEUHHG- ing tank offered individuals similar conditions as observed in the natural habitat. Earlier studies that kept adults of S. parvus LQ WHUUDULXPV RI VLPSOHU GHVLJQ QR ÀRZLQJ water) showed that individuals did not display acoustic or visual signals under such conditions (R. Kasah, pers. comm.). At the beginning of our project we kept individuals pair-wise in simpler terraria with a small water area containing no gravel and only larger pebbles, some tree EUDQFKHVÀRZLQJZDWHUYLDDSXPSDQGWHPSHUDWXUHVRI 23-25 °C. Under these conditions individuals performed DGYHUWLVHPHQWFDOOVDQGIRRWÀDJJLQJEHKDYLRUEXWQRUH- productive behavior could be observed. Especially in S. guttatusIHPDOHVGLVSOD\HGWHUULWRULDOFDOOVDQGIRRWÀDJV if males approached, a behavior that was interpreted as a spacing mechanism (Preininger et al., data not shown). After transferring all individuals in the considerably larger and diversely structured breeding tank, calling activity LQWHQVL¿HGDQGSDLUVLQDPSOH[XVFRXOGEHREVHUYHGDIWHU DIHZZHHNV+HQFHZHVXJJHVWWKDW¿UVWDQGIRUHPRVW WKHJUDYHOFRQWDLQLQJÀRZLQJZDWHUDUHDZDVFUXFLDOIRU reproduction, but also the simulated dry and rainy season might have had an effect. It is now essential to alter or exclude single environmental conditions or habitat structures to determine factors necessary for reproduction. So Figure 7. Tadpoles of Staurois parvus. Image by N. Potensky. IDUZHKDYHUHPRYHGWKHDUWL¿FLDOVLGHSRRODQGÀXVKHG

Figure 8. Juvenile Staurois parvus. Image by D. Zupanc.

Amphib. Reptile Conserv. | http://redlist-ARC.org051 September 2012 | Volume 5 | Number 3 | e51 80 CHAPTER 4

reininger et al.

tats and monitoring. Nevertheless, to identify habitats necessary for survival of a species and subsequent immediate protection requires extensive research and conservations efforts. Captive breeding programs however should be extremely cautious to avoid disease transmission, hence in our project only individuals from the bio- secure container facility will be considered for transport to other institutions. Ex situ conservation and research programs not only can prevent extinction through captive management and re-introduction to the wild, but offer opportunities for research to identify and, thus, protect key habitats (Zippel et al. 2011).

&RQFOXVLRQ

The species of the genus Staurois live and breed along IDVWÀRZLQJVWUHDPVDQGZDWHUIDOOV)RUWKH¿UVWWLPHLW was possible to ex situEUHHGWZRIRRWÀDJJLQJVSHFLHV in captivity and demonstrate the importance of fresh water streams and adjacent gravel pools for reproduction. We suggest that to successfully breed stream dwelling anurans with territorial males/females (also immature juveniles as mentioned previously) performing spacing EHKDYLRUV HJIRRWÀDJJLQJ ODUJHDQGGLYHUVHO\VWUXF- tured terraria, including a waterfall and several options for egg deposition should promise the best success rate for future breeding programs. Further, we emphasize, that zoo-based conservations and research programs help Figure 9. $UWL¿FLDOZDWHUIDOOKDELWDWDWWKH%RUQHR5DLQIRUHVW to identify ecological factors that are necessary for the house in the Vienna Zoo. Image by N. Potensky. survival of threatened species, and also raise awareness to the ongoing amphibian decline. Public awareness of pool from the S. parvus breeding terrarium and still ob- the conservation needs of threatened amphibian species serve freshly hatched tadpoles. through zoo-based conservation breeding programs may )UHVKZDWHUVWUHDPVDQGDGMDFHQWÀRZQWKURXJKSRROV then be translated into in-range conservation initiatives with gravel areas seem to be important to secure the by regional governments and local stakeholders who are VXUYLYDORIWKHIRRWÀDJJLQJVSHFLHVLQWKHJHQXVStau- also concerned with the ex situ conservation of these two rois. However, deforestation and subsequent siltation of species. streams and rivers are the major threats to most stream- living and breeding anuran species in Borneo. Inger and Acknowledgments.—Export and import permission Voris (1993) found that a stream with a silt bottom com- were obtained from the Brunei Museums Department pletely lacked all the species known to breed along clear (Reference: 14/JMB/209/68/2) and the Austrian Federal DQGIDVWÀRZLQJVWUHDPV6HOHFWLYHORJJLQJFKDQJHVWKH Ministry of Health, respectively. We thank U. Grafe for water chemistry considerably in nearby streams and sedi- his continuous professional and logistic help. We are PHQW\LHOGVRIVWUHDPVDUHWLPHVKLJKHUIRUXSWR¿YH grateful for the dedication and support of R. Riegler, E. months after logging (Douglas et al. 1993; Douglas et Karell, and all other zoo-keepers that are involved in this al. 1992). So far, it is not well-understood how habitat project. We thank M. Boeckle, N. Potensky, and D. Zu- loss or alternations will affects riparian anurans on Bor- panc for providing their photographs. We also thank the neo, but considering the dramatic decline of this group reviewers for valuable comments on the manuscript. The of vertebrates it is expected that biodiversity will decline study was supported by the Austrian Science Fund FWF- considerably if ecosystems continue to degrade. P22069 and the Society of Friends of the Vienna Zoo. For some species ex situ programs may be the only option to avoid extinction (e.g., the Kihansi spray toad, Author Contributions.—DP carried out the study, Nectophrynoides asperginis [Krajick 2006] or the Pana- analyzed pictures and available data and wrote the man- manian golden frog, Atelopus zeteki [Zippel 2002]). Spe- uscript. AW participated in the design of the study and cies that are not considered Critically Endangered should coordinated its implementations at the Vienna Zoo. TW be preserved in the wild through protection of key habi- designed and build the breeding facility, carried out the

Amphib. Reptile Conserv. | http://redlist-ARC.org052 September 2012 | Volume 5 | Number 3 | e51 CHAPTER 4 81

Conservation breeding success in Staurois parvus and Staurois guttatus

Figure 10. Juvenile Staurois parvusSHUIRUPLQJDIRRWÀDJJLQJEHKDYLRU,QWHUGLJLWDOZHEELQJDUHWUDQVSDUHQWJUH\DQGQRWZKLWHDV observed in adults (see also Fig. 2). Image by N. Potensky.

Figure 11. Tadpoles of Staurois guttatus. Image by N. Potensky.

Amphib. Reptile Conserv. | http://redlist-ARC.org053 September 2012 | Volume 5 | Number 3 | e51 82 CHAPTER 4

reininger et al. import of the species, and participated in all decision Germany. [Online]. Available: http://www.frogsof- processes. WH conceived and coordinated the study. All borneo.org [Accessed: 08 May 2012]. DXWKRUVUHDGDQGDSSURYHGWKH¿QDOPDQXVFULSW Hödl W, Amézquita A. 2001. Visual signaling in anuran amphibians. In: Anuran Communication. Editor, Ryan MJ. Smithsonian Institution Press, Washington, DC, LLWHUDWXUHFLWHG USA. 121-141. Hoffmann M, Hilton-Taylor C, Angulo A, Böhm M, Arch VS, Grafe TU, Narins PM. 2008. Ultrasonic signal- Brooks TM, Butchart SHM, Carpenter KE, Chanson ling by a Bornean frog. Biological Letters 4(1):19-22. J, Collen B, Cox NA, Darwall WRT, Dulvy NK, Har- $UL¿Q8,VNDQGDU'7%LFNIRUG'3%URZQ500HLHU rison LR, Katariya V, Pollock CM, Quader S, Rich- R, Kutty SN. 2011. Phylogenetic relationships within man NI, Rodrigues ASL, Tognelli MF, Vié J-C, Agu- the genus Staurois (Anura, Ranidae) based on 16S iar JM, Allen DJ, Allen GR, Amori G, Ananjeva NB, rRNA sequences. Zootaxa 2744:39-52. Andreone F, Andrew P, Ortiz ALA, Baillie JEM, Baldi Boeckle M, Preininger D, Hödl W. 2009. Communica- R et al. 2010. The impact of conservation on the status tion in noisy environments I: Acoustic signals of Stau- of the world’s vertebrates. Science 330(6010):1503- rois latopalmatus Boulenger 1887. Herpetologica 1509. 65(2):154-165. Inger RF. 1969. Organization of communities of frogs Browne RK, Wolfram K, Garciá G, Bagaturov MF, Pere- along small rain forest streams in Sarawak. Journal of boom ZJJM. 2011. Zoo-based amphibian research and Animal Ecology 38(1):123-148. conservation breeding programs. Amphibian and Rep- Inger R, Iskandar D, Das I, Stuebing R, Lakim M, Yam- tile Conservation 5(3):1-14. bun P. 2004. IUCN Red List of Threatened Species. Douglas I, Greer T, Bidin K, Spilsbury M. 1993. Impacts [Online]. Available: www.iucnredlist.org [Accessed: of rainforest logging on river systems and communi- 16 April 2012]. ties in Malaysia and Kalimantan. Global Ecology and Inger RF, Stuebing RB. 2005. A Field Guide to the Frogs Biogeography Letters 3(4/6):245-252. of Borneo. (Second Edition). Natural History Publica- Douglas I, Spencer T, Greer T, Bidin K, Sinun W, Meng tions (Borneo) Sdn. Bhd. Kota Kinabalu, Sabah, Ma- WW. 1992. The impact of selective commercial laysia. 201p. logging on stream hydrology, chemistry and sedi- Inger RF, Tan FL. 1996. Checklist of the frogs of Borneo. ment loads in the Ulu Segama Rain Forest, Sabah, 5DIÀHV%XOOHWLQRI=RRORJ\ 44(2):551-574. Malaysia. Philosophical Transactions of the Royal Inger RF, Voris HK. 1993. A comparison of amphibian Society of London, Series B: Biological Sciences communities through time and from place to place 335(1275):397-406. in Bornean forests. Journal of Tropical Ecology *DJOLDUGR 5 &UXPS 3*ULI¿WK ( 0HQGHOVRQ - 5RVV 9(04):409-433. H, Zippel K. 2008. The principles of rapid response Inger RF, Wassersug RJ. 1990. A Centrolenid-Like An- for amphibian conservation, using the programmes in uran Larva from Southeast Asia. Zoological Science Panama as an example. International Zoo Yearbook 7(3):557-561. 42(1):125-135. Iskandar DT, Colijn E. 2000. Preliminary checklist of Gascon C, Collins JP, Moore RD, Church DR, McKay Southeast Asian and New Guinean amphibians. Treu- JE, Mendelson JRI. 2007. Amphibian Conservation bia 31:1-134. Action Plan. The World Conservation Union (IUCN)/ IUCN. 2011. IUCN Red List of Threatened Species. SSC Amphibian Specialist Group, Gland, Switzerland [Online]. Available: www.iucnredlist.org [Accessed: and Cambridge, United Kingdom. 64 p. 16 April 2012]. Grafe TU, Preininger D, Sztatecsny M, Kasah R, Deh- Keller A, Rödel M-O, Linsenmair KE, Grafe TU. 2009. ling JM, Proksch S, Hödl W. 2012. Multimodal com- The importance of environmental heterogeneity for munication in a noisy environment: A case study of species diversity and assemblage structure in Bornean the Bornean rock frog Staurois parvus. PLoS One stream frogs. Journal of Animal Ecology 78(2):305- 7(5):e37965. 314. Grafe TU, Wanger TC. 2007. Multimodal signaling in Konopik O. 2010. Movement patterns, habitat use and PDOHDQGIHPDOHIRRWÀDJJLQJIURJVStaurois guttatus diet of tropical ranid frogs: A comparison between pi- (Ranidae): An alerting function of calling. Ethology oneer and native anurans in Borneo. M.S. Thesis, Uni- 113(8):772-781. versity of Wuerzburg, Wuerzburg, Germany, Depart- Gururaja KV. 2010. Novel reproductive mode in a torrent ment of Animal Ecology and Tropical Biology. 96 p. frog Micrixalus saxicola (Jerdon) from the Western Krajick K. 2006. The lost world of the Kihansi toad. Sci- Ghats, India. Zootaxa 2642:45-52. ence 311(5765):1230-1232. Haas A, Das I. 2012. Frogs of Borneo — The Frogs of Lee S, Zippel K, Ramos L, Searle J. 2006. Captive- East Malaysia and their Larval Forms: An Online breeding programme for the Kihansi spray toad Nec- Photographic Guide. Zoological Museum Hamburg, tophrynoides asperginis at the Wildlife Conservation

Amphib. Reptile Conserv. | http://redlist-ARC.org054 September 2012 | Volume 5 | Number 3 | e51 CHAPTER 4 83

Conservation breeding success in Staurois parvus and Staurois guttatus

Society, Bronx, New York. International Zoo Year- Preininger D, Boeckle M, Hödl W. 2009. Communica- book 40(1):241-253. tion in noisy environments II: Visual signaling behav- Malkmus R, Kosuch J, Kreutz J. 1999. Die larve von LRURIPDOHIRRWÀDJJLQJIURJVStaurois latopalmatus. Staurois tuberilinguis Boulenger, 1918. Eine neue- Herpetologica 65(2):166-173. centroleniden Kaulquappe aus Borneo (Anura: Rani- Stuart SN, Chanson JS, Cox NA, Young BE, Rodrigues dae). Herpetozoa 12(1/2):17-22. ASL, Fischman DL, Waller RW. 2004. Status and Matsui M, Mohamed M, Shimada T, Sudin A. 2007. trends of amphibian declines and extinctions world- Resurrection of Staurois parvus from S. tuberilinguis wide. Science 306(5702):1783-1786. from Borneo (Amphibia, Ranidae). Zoological Sci- Zimmerman B, Simberloff D. 1996. An historical inter- ence 24(1):101-106. pretation of habitat use by frogs in a Central Ama- McFadden M, Duffy S, Harlow P, Hobcroft D, Webb C, zonian Forest. Journal of Biogeography 23(1):27-46. G W-F. 2008. A review of the green and golden bell Zippel K. 2002. Conserving the Panamanian golden frog Litoria aurea breeding program at Taronga Zoo. frog: Proyecto Rana Dorada. Herpetological Review Australia Zoologist 34(3):291-296. 33(1):11-12. Narvaes P, Rodrigues MT. 2005. Visual communication, Zippel K, Johnson K, Gagliardo R, Gibson R, McFad- reproductive behavior, and home range of Hylodes den M, Browne R, Martinez C, Townsend E. 2011. dactylocinus (Anura, Leptodactylidae). Phyllomedusa The Amphibian Ark: A global community for ex situ 4(2):147-158. conservation of amphibians. Herpetological Conser- Preininger D, Boeckle M, Hödl W. 2007. Comparison of vation and Biology 6(3):340-352. anuran acoustic communities of two habitat types in the Danum Valley Conservation Area, Sabah, Malay- Received: 12 May 2012 sia. Salamandra 43(3):129-138. Accepted: 26 June 2012 Published: 7 September 2012

Doris Preininger KDVDOUHDG\ZRUNHGZLWKIRRWÀDJJLQJIURJVLQKHUXQGHUJUDGXDWHVWXGLHV,QKHU graduation thesis she addresses the multimodal (acoustic and visual) communication in anurans and tries to explain how selection on senders and receivers promotes complex displays under different acoustic and environmental conditions. She is currently completing her dissertation at the Department RI(YROXWLRQDU\%LRORJ\8QLYHUVLW\9LHQQD+HUUHVHDUFKLQFOXGHVIRRWÀDJJLQJVSHFLHVIURP%RUQHR and India and focuses on a bio-acoustic and experimental approach in the natural habitat of the respective species. In several visits to Borneo it became quite obvious to her that agricultural demands gradu- ally degrade the primary forest and that every conservation effort possible should be immediately taken to conserve and protect the biodiversity of the rainforest.

Anton Weissenbacher is Zoological Curator at Vienna Zoo, committee member of the European As- sociation of Aquariums and coordinator of the European StudBook (ESB) of Brachylophus fasciatus. At Vienna Zoo he is responsible for the zoological and technical management of the aquarium, the ³'HVHUWKRXVH´WKH³5DLQIRUHVWKRXVH´DQGPRQLWRUVDOO]RRLVVXHVFRQFHUQLQJ¿VKHVDPSKLELDQV reptiles, and invertebrates. Under his zoological guidance, the zoo has recently registered several ex- FHSWLRQDOEUHHGLQJVXFFHVVHVVXFKDVWKHZRUOG¶V¿UVW1RUWKHUQULYHUWHUUDSLQBatagur baska, hatched in captivity. Together with his team he manages the world’s largest Aphanius species breeding group. He KDVVXSHUYLVHGYDULRXVVFLHQWL¿FSXEOLFDWLRQVDQGKDVLQLWLDWHGVHYHUDOFRQVHUYDWLRQSURMHFWVLQFOXGLQJ Project Batagur baska.

Thomas Wampula has worked since 1996 at the Vienna Zoo Schönbrunn. He started as Animal Care 7DNHUDWWKH$TXDULXPKRXVHDQGODWHUWUDQVIHUUHGWRWKH³5DLQIRUHVWKRXVH´ZKHUHKLV¿UVWDQGIRUH- PRVWLQWHUHVWVZHUHDPSKLELDQVUHSWLOHVDQG¿VK+LVGXWLHVDQGUHVSRQVLELOLWLHVLQFOXGHGWKHDUUDQJH- ment and design of terraria and the maintenance of facilities. In 2007 he became a member of the Department of Technology and Project Development at the zoo and now is engaged in planning, design, DQGGHYHORSPHQWRIYLYLDULDLQWKHHQWLUH9LHQQD=RR7KHIRRWÀDJJLQJIURJSURMHFWKDVUHSHDWHGO\OHG KLPWR%RUQHRZKHUHKHDVVLVWHGLQ¿HOGZRUNFDSWXUHWUDQVSRUWDQGFDUHRIIURJVDQGDWWKH]RRKH managed the construction of the breeding facility.

Amphib. Reptile Conserv. | http://redlist-ARC.org055 September 2012 | Volume 5 | Number 3 | e51 84 CHAPTER 4

reininger et al.

Walter Hödl has an international record in a wide range of topics in amphibian ecology and behavior. Since 1997 he has worked as an Associate Professor at the Institute of Zoology, University of Vienna. 'XULQJWKHODVW\HDUVKHKDVVWXGLHGQXPHURXVIRRWÀDJJLQJIURJVSHFLHVLQ$VLD$XVWUDOLDDQG6RXWK America and has established the South-East Asian frog genus Staurois spp. as a research model. Pre- ZRUNRQYLVXDOVLJQDOLQJIURJVSHFLHVEHJDQPRUHWKDQ\HDUVDJRZKHQKHGRFXPHQWHGIRUWKH¿UVW WLPHLQDVFLHQWL¿F¿OP1²WRJHWKHUZLWK%UD]LOLDQFROOHDJXHV²DQXUDQIRRWÀDJJLQJEHKDYLRUDQGODWHU compared visual signal repertoires of anuran species worldwide. He discovered the use of the vocal sac as a visual signal independently of sound production in Phrynobatrachus kreffti, and set off a study on color change in the explosively breeding anuran species Rana arvalis. In the neotropics, his so called ³KDQG\IHOORZ´Allobates femoralis has been his research focus over the past 30 years and has led to numerous research and teaching visits to Brazil (Universities at Belém, São Luís João Pessoa, Manaus, São Paulo, Rio Branco, Ribeirão Preto, Feira da Santana, and at MPEG Belem, INPA Manaus) and 3HUXDQG)UHQFK*XLDQDHQDEOLQJKLPWRVSHQGRYHUHLJKW\HDUVRI¿HOGZRUNLQ$PD]RQLD$PRQJ PDQ\IXQFWLRQVKHLVDPHPEHURIWKHVFLHQWL¿FFRPPLWWHHRI::)$XVWULDDQGWKHKHDGRIWKHQDWXUH conservation society of lower Austria and continuously establishes cooperation around the globe to promote anuran research and conservation.

1+|GO:5RGULJXHV07$FFDFLRGH0/DUD3+3DYDQ'6FKLHVDUL/&6NXN*)RRWÀDJJLQJGLVSOD\LQWKH%UD]LOLDQVWUHDPEUHHGLQJIURJHy- lodes asper /HSWRGDFW\OLGDH $XVWULDQ)HGHUDO,QVWLWXWHRI6FLHQWL¿F)LOP )LOP&7Ig:):LHQ >ZHEDSSOLFDWLRQ@$PSKLELD:HE%HUNHOH\ California. [Online]. Available: http://amphibiaweb.org/cgi/amphib_query?where-genus=Hylodes&where-species=asper [Accessed: 02 July 2012].

Amphib. Reptile Conserv. | http://redlist-ARC.org05 September 2012 | Volume 5 | Number 3 | e51 85

Concluding Discussion

Foot-flagging signals were observed mainly during agonistic male-male interactions

(Hödl & Amézquita 2001; Hartmann et al. 2005). In the current studies we interpret the display as an agonistic signal ritualized from physical attacks. Aggressive or threat signals usually reflect a former fighting strategy or posture movements before the initial attack

(Bradbury & Vehrencamp 2011). In physical fights M. saxicola males position themselves close to opponents and competitors repeatedly kick with their hind legs until one male is thrown off the perching site (pers. observation). Exaggerations of limb stretching or lifting and stereotype movement patterns causing a behavioral change in opponents could have triggered the ritualization of foot-lifting and foot-flagging behavior. Individuals displaying this behavior might have had an enhanced fitness, because they were able to secure resources necessary for reproductive success without getting physically attacked. We suggest that a receiver response to incomplete kicks or to kicks, which were performed from a distance, initially were the source of selection in the evolution of foot-flagging signals. Hartmann et al.

(2005) describe “leg kicking” as a visual signal in Brazilian Atlantic forest frogs, as a laterally backward air-kick suggested to be performed with the back turned to the rival during agonistic interactions. Two nocturnal species from the family Hylidae perform kicking signals and leg lifts but do not display foot-flagging behavior, whereas Hyla ehrhardti uses leg lifts and foot flags as visual signals but does not perform kicks (Hartmann et al. 2005). In the diurnal genus Hylodes foot-flagging and foot-lifting behavior but no leg kicks were observed

(Hödl et al. 1997; Haddad and Giaretta 1999; Narvaes & Rodrigues 2005). Hylodes asper and H. dactylocinus display conspicuous foot colorations during foot-flagging signals (Hödl et al. 1997; Narvaes & Rodrigues 2005). The availability of light may have favored the evolution of contrasting foot colorations in diurnal species. In M. saxicola the described kick is a physical attack against the opponent and does not constitute a visual signal. In the diurnal genus Staurois all species display foot-flagging signals and the hind feet interdigital webbings are brightly colored (Grafe & Wanger 2007; Preininger et al. 2009). Additionally, high transition probabilities between alternating right and left foot flags (and vice versa) in S. 86 parvus, indicate that the movement pattern is repetitive and stereotyped. We suggest that the behavioral differences represent a continuum in the evolution of foot displays resulting from ecological divergence and sexual selection.

We further suggest that foot flagging in M. saxicola represents a nascent state in the evolution of visual communication. Across species comparisons of M. saxicola and S. parvus demonstrate that the latter do not attack each other with kicking behavior and even more interestingly the visual signal in S. parvus shows increased conspicuousness and is the predominant response to all conducted playback presentations.

Visual signal conspicuousness can be enhanced by different types of contrast with the optical background: brightness, color, size, shape and movement contrasts (Endler 1978;

Bradbury & Vehrencamp 2011). The ambient light affects the conspicuousness of a color signal (Endler 1990; Endler 1993a; Leal & Fleishman 2002 ; Leal & Fleishman 2004). Light conditions in different microhabitats, such as various degrees of shading in a tropical forest create a complex system with varying demands for optimum signal efficacy (Endler 1993a).

To measure signal brightness we focused on contrasts of inter-digital webbing against the frog’s body coloration. The foot webbing of S. parvus is almost five times brighter than that of M. saxicola and poses a strong contrast to the body's brightness (Chapter 3). The environmental signal-to-noise ratio remains to be tested, but we predict a similar signal contrast as observed to body brightness since both species are well camouflaged in their respective environment unless they display foot-flagging signals.

The chapters 1 – 3 demonstrate the difference in the predominant communication mode in response to playback presentations in the two study species. When presented with purely acoustic conspecific stimuli S. parvus displayed more foot-flagging signals during playbacks than during pre-playback periods whereas the number of advertisement calls did not differ (Chapter 1). Micrixalus saxicola mainly responded with vocalizations but no foot- flagging signals could be recorded in response to acoustic playbacks that lack the visual cue of a pulsating vocal sac (Chapter 2). Additionally, the main response to acoustic, visual and multimodal experimental stimuli in S. parvus was foot flagging, whereas Micrixalus saxicola

87 responded primarily with calls and never foot flagged (Chapter 3). Although S. parvus displayed more foot flags during acoustic stimuli than during visual presentations, the visual stimulus by itself evoked foot flagging response in receivers. The results highlight that foot- flagging behavior in S. parvus constitutes a communicative signal during conspecific signal presentations. The signal is efficiently transmitted due to its visual signal-to-noise ratio and by itself effectively influences the behavior of a receiver. Assessing receiver responses to signal components and multimodal signals also allow categorizing signal components and drawing assumptions on the potential multimodal signal content. As explained in chapter 3, within species differences in response were small and hence our assumptions on signal content are based on the predominant receiver response in the respective species. Call responses in M. saxicola did not differ between isolated unimodal stimuli or the presented multimodal signal, indicating that the acoustic and visual displays equivalently influence receivers to emit calls, suggesting signal redundancy. Redundant displays can function to increase the accuracy of response or act as a backup to enhance signal efficacy (reviewed in Candolin 2003; Hebets & Papaj 2005). In S. parvus an acoustic stimulus triggered foot- flag responses in the receiver when a visual stimulus was absent, whereas an isolated visual stimulus decreased foot-flag responses similar to the multimodal stimulus. Our results indicate that the acoustic signal might be of less importance when both signals are perceived, indicating a non-redundant influence on receivers. The short latency times between advertisement call and subsequent foot-flagging behavior in the genus Staurois

(Grafe & Wanger 2007; Preininger et al. 2009; Chapter 1) suggests that the call is used as an alert signal that directs the receiver's attention to the informative visual signal, according to the "alert and attention altering hypothesis". However, present analyses of foot brightness and body size or weight show no correlations of signal reflectance and attributes in S. parvus (Sztatecsny, unpublished data). In M. saxicola the delay between call and foot flagging and vice versa did not differ (Appendix A), which leads to the assumption that the signals do not form a functional unit in this species. Yet, results of experiments on receiver responses to acoustic playback and additional visual cues of a pulsating vocal sac in M.

88 saxicola (Chapter 2) show that the conspecific advertisement calls elicits calls and foot lifts

(taps). When receivers were presented with simultaneous visual cues, a detectable source to the localization of the competitor, they also displayed foot-flagging signals. Hence, foot lifting could represent a moderate threat to males advertising close by, whereas when opponents are detected and localized a more pronounced signal is displayed, which could predict aggressive escalation. We suggest that foot flagging and foot lifting might predict the aggressive motivation of an opponent rather than his fighting ability.

Regarding a communicative behavior or a trait in respect to its evolutionary origin and development, the fitness benefits from receiver responses have to outweigh the costs of signaling (Searcy & Nowicki 2005). We already argued that foot-flagging signals could decrease physical attacks by male opponents during agonistic encounters and might increase mating opportunities by securing signaling sites where resources (e.g. shallow water areas or pools with gravel) are available for reproduction. Most diurnal foot-flagging species face similar ecological constraints: They live and breed along fast-flowing streams that produce continuous background noise (Hödl & Amézquita 2001). Signaling constraints imposed on the acoustic display might have influenced the evolution of visual signals in anurans. Our results demonstrate that acoustic communication in the two study species is not impaired by low-frequency dominated ambient stream noise. However, in the genus

Staurois high-frequency calls were suggested as an adaptation to increase the signal-to- noise ratio (Boeckle et al. 2009). In several frog species the dominant call frequency is inversely related to body size (Gingras et al. 2012). In most species the spectral characteristics, in particular the dominant frequency, is influenced by the size, tension and mass of the vocal cords, which in turn correlate to body size (Martin 1971; McClelland et al.

1996). Staurois spp. call at higher frequencies relative to size than other ranids (Boeckle et al. 2009), hence ecological selection pressures imposed by noise might have been greater than morphological constraints.

Advertisement calls of twenty frog species occurring in the same habitat as Staurois spp. attract frog-biting midges (Diptera: Corethrellidae) (Grafe et al. 2012). The flies parasitize

89 frogs with call frequencies below 4 kHz, suggesting further selective pressure on vocalization and an additional benefit of a shift to high-frequency calls. In S. guttatus the vocalizations of males and females have a similar dominant frequency, the female call however has less energy and can be masked by stream noise (Appendix B). Females also display foot-flagging signals (Grafe & Wanger 2007) when approached by males, probably signaling their unreceptive status or their motivation to defend resting sites (Preininger, unpublished data). In M. saxicola advertisement calls are subject to masking from chorus noise and signals in another sensory modality could enhance detection and localization of conspecifics. In summary we suggest that noise is a driving force for the evolution of visual signals and in particular of foot-flagging behavior.

Foot-flagging signals which pose a strong contrast to the overall body coloration as observed in S. parvus, might not only facilitate better detection and localization to intended receivers but also to eavesdropping predators. Potential predators of S. parvus are snakes

(e.g. Xenochrophis triangulifera), skinks (e.g. Mabuya sp.), lizards (e.g. Gonocephalus grandis), birds (e.g. Kittacincla macroura) and most notably other anuran species (e.g.

Limnonectes sp.), disregarding nocturnal spiders, centipedes and bats. Bright webbings are only displayed for a very short period but are highly conspicuous to attentive observers.

Nevertheless, I have never observed predator attacks on S. parvus and my own attempts to catch the very agile frogs during the daytime were hardly ever successful as they escape by jumping down steep cascades or into waterfall pools. We suggest that tadpoles are subject to higher predation risk than adult individuals. In chapter 4 we showed that tadpoles inhabit the gravel area of the study arena and we found egg deposition sites under larger rocks. In the fresh-water stream pools in the natural habitat of S. parvus we observed several potential predators (insect larvae, shrimps, crabs and aquatic bugs) that could dig below rocks or in the gravel and decimate a population. Interestingly, already juvenile S. parvus perform foot-flagging signals, but hind feet interdigital webbings are transparent grey.

Inconspicuous web coloration could minimize the predation risk but could also be associated as signal characteristic, which indicates sexual maturity.

The above assumptions or hypotheses of how and why visual- and multimodal signals have developed, focused on the role of signals during male-male agonistic interactions, which have received little attention in recent literature on anuran multimodal signal evolution

(for other taxa see e.g. Hughes 1996; Morris & Ryan 1996; Elias et al. 2008). Additional evidence of the importance of visual cues in anuran male-male competition aside from foot- flagging signals comes from nuptial colorations of the moor frog (Rana arvalis) (Appendix C).

The study demonstrates that the conspicuous coloration acts as a visual signal, which facilitates mate recognition and reduces unwanted male-male interaction in dense aggregations of explosive breeders.

Understanding the processes of signal production, transmission and reception will eventually help to explain mechanisms that have influenced the evolution of visual and multimodal signals. I believe the combined studies in my thesis provide a first step in understanding the development of the complex signal repertoire of audio-visual displays in foot-flagging frog species.

References (of Preface and Concluding Discussion)

Arch, V. S., Grafe, T. U. & Narins, P. M. 2008. Ultrasonic signalling by a Bornean frog. Biological Letters, 4, 19-22.

Bee, M. A. 2008. Finding a mate at a cocktail party: spatial release from masking improves acoustic mate recognition in grey treefrogs. Animal Behaviour, 75, 1781-1791.

Boeckle, M., Preininger, D. & Hödl, W. 2009. Communication in noisy environments I: Acoustic signals of Staurois latopalmatus Boulenger 1887. Herpetologica, 65, 154-165.

Bradbury, J. W. & Vehrencamp, S. L. 2011. Principles of Animal Communication, Second Edition. Sunderland: Sinauer Press.

Brenowitz, E. A. 1989. Neighbor call amplitude influences aggressive behavior and intermale spacing in choruses of the Pacific Treefrog (Hyla regilla). Ethology, 83, 69-79.

Bucher, T. L., Ryan, M. J. & Bartholomew, G. A. 1982. Oxygen consumption during resting, calling, and nest building in the frog Physalaemus pustulosus. Physiological Zoology, 55, 10-22.

Candolin, U. 2003. The use of multiple cues in mate choice. Biological Reviews, 78, 575- 595.

Duellman, W. E. & Trueb, L. 1986. Biology of Amphibians. New York: McGraw-Hill Publishing Company.

Elias, D. O., Kasumovic, M. M., Punzalan, D., Andrade, M. C. B. & Mason, A. C. 2008. Assessment during aggressive contests between male jumping spiders. Animal Behaviour, 76, 901-910.

Endler, J. A. 1978. A predator's view of animal color patterns. Evolutionary Biology (New York), 11, 319-364.

Endler, J. A. 1990. On the measurement and classification of colour in studies of animal colour patterns. Biological Journal of the Linnean Society, 41, 315-352.

Endler, J. A. 1992. Signals, signal conditions, and the direction of evolution. American Naturalist, 139, 125–153.

Endler, J. A. 1993a. The color of light in forests and its implications. Ecological Monographs, 36, 1-27.

Endler, J. A. 1993b. Some general comments on the evolution and design of animal communication systems. Philosophical Transactions of the Royal Society of London B Biological Sciences, 340, 215-225.

Endler, J. A. 2000. Evolutionary implications of the interaction between animal signals and the environment. In: Animal Signals Signalling and Signal Design in Animal Communication (Ed. by Y. Espmark, T. Amundsen & G. Rosenqvist), pp. 11-46. Trondheim, Norway: Tapir Academic Press.

Feng, A. S. & Narins, P. M. 2008. Ultrasonic communication in concave-eared torrent frogs (Amolops tormotus). Journal of Comparativ Physiolology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, 194, 159-167.

Gerhardt, H. C. & Huber, F. 2002. Acoustic communication in Insects and Anurans: Common problems and diverse solutions. Chicago: University of Chicago Press.

Gerhardt, H. C. & Schwartz, J. 2001. Auditory tuning and frequency preferences in anurans. In: Anuran communication (Ed. by M. J. Ryan), pp. 73-85. Washington: Smithonian Institution Press.

Gingras, B., Boeckle, M., Herbst, C. T. & Fitch, W. T. 2012. Call acoustics reflect body size across four clades of anurans. Journal Of Zoology, In press.

Grafe, T. U. & Wanger, T. C. 2007. Multimodal signaling in male and female foot-flagging frogs Staurois guttatus (Ranidae): An alerting function of calling. Ethology, 113, 772-781.

Grafe, U., Konopik, O., Ahmad, N. & Art, B. 2012. Calling in the face of eavesdroppers: selective pressures on anuran advertisement calling in the presence of frog-biting midges. In: World Congress of Herpetology. Vancouver.

Haddad, C. F. B. & Giaretta, A. A. 1999. Visual and acoustic communication in the Brazilian torrent frog, Hylodes asper (Anura: Leptodactylidae). Herpetologica, 55, 324-333.

Hartmann, M. T., Giasson, L. O. M., Hartmann, P. A. & Haddad, C. F. B. 2005. Visual communication in Brazilian species of anurans from the Atlantic forest. Journal of Natural History, 39, 1675-1685.

Hebets, E. A. & Papaj, D. R. 2005. Complex signal function: developing a framework of testable hypotheses. Behavioral Ecology and Sociobiology, 57, 197-214.

Hirschmann, W. & Hödl, W. 2006. Visual signaling in Phrynobatrachus krefftii Boulenger, 1909 (Anura: Ranidae). Herpetologica, 62, 18-27.

Hödl, W. 1977. Call differences and calling site segregation in anuran species from central Amazonian floating meadows. Oecologia, 28, 351-363.

Hödl, W. & Amézquita, A. 2001. Visual signaling in anuran amphibians. In: Anuran Communication (Ed. by M. J. Ryan), pp. 121-141. Washington, DC, U.S.A: Smithsonian Institution Press.

Hödl, W., Rodrigues, M. T., Accacio, d. M., Lara, P. H., Pavan, D., Schiesari, L. C. & Skuk, G. 1997. Foot-flagging display in the Brazilian stream-breeding frog Hylodes asper (Leptodactylidae). Austria: Austrian Federal Institut of Scientific Film (ÖWF).[web application]. 2012. Berkeley, California: AmphibiaWeb. Available: http://amphibiaweb.org/.

Hughes, M. 1996. The function of concurrent signals: visual and chemical communication in snapping shrimp. Animal Behaviour, 52, 247-257.

Klump, G. M. & Gerhardt, H. C. 1992. Mechanisms and function of call-timing in male- male interactions in frogs. In: Playback and Studies of Animal Communication (Ed. by P. K. McGregor), pp. 153-174. New York: Plenum Press.

Krishna, S. N. & Krishna, S. B. 2006. Visual and acoustic communication in an endemic stream frog, Micrixalus saxicolus in the Western Ghats, India. Amphibia-Reptilia, 27, 143- 147.

Leal, M. & Fleishman, L. J. 2002 Evidence for habitat partitioning based on adaptation to environmental light in a pair of sympatric lizard species. Proceedings Of The Royal Society B: Biological Sciences, 269, 351–359.

Leal, M. & Fleishman, L. J. 2004. Differences in visual signal design and detectability between allopatric populations of Anolis lizards. American Naturalist, 163, 26-39.

Martin, W. F. 1971. Mechanics of sound production in toads of the genus Bufo: Passive elements. Journal of Experimental Zoology, 176, 273-293.

McClelland, B., Wilczynski, W. & Ryan, M. 1996. Correlations between call characteristics and morphology in male cricket frogs (Acris crepitans). Journal of Experimental Biology, 199, 1907-1919.

Morris, M. R. & Ryan, M. J. 1996. Sexual difference in signal-receiver coevolution. Animal Behaviour, 52, 1017-1024.

Narins, P. M., Grabul, D. S., Soma, K. K., Gaucher, P. & Hödl, W. 2005. Cross-modal integration in a dart-poison frog. Proceedings of the National Academy of Sciences of the United States of America, 102, 2425-2429.

Narins, P. M., Hödl, W. & Grabul, D. S. 2003. Bimodal signal requisite for agonistic behavior in a dart-poison frog, Epipedobates femoralis. Proceedings of the National Academy of Sciences of the United States of America, 100 577-580.

Narvaes, P. & Rodrigues, M. T. 2005. Visual communication, reproductive behavior, and home range of Hylodes dactylocinus (Anura, Leptodactylidae). Phyllomedusa, 4, 147-158.

Otovic, P. & Partan, S. 2009. Multimodal signaling in animals. In: Encyclopedia of Neuroscience (Ed. by L. R. Squire), pp. 1095-1105: Oxford: Academic Press.

Partan, S. & Marler, P. 1999. Communication Goes Multimodal. Science, 283, 1272-1273.

Partan, S. & Marler, P. 2005. Issues in the classification of multimodal communication signals. American Naturalist, 166, 231-245.

Pauly, G. B., Bernal, X. E., Rand, A. S. & Ryan, M. J. 2006. The vocal sac increases call rate in the Túngara Frog Physalaemus pustulosus. Physiological and Biochemical Zoology, 79, 708-719.

Preininger, D., Boeckle, M. & Hödl, W. 2009. Communication in noisy environments II: Visual signaling behavior of male foot-flagging frogs Staurois latopalmatus. Herpetologica, 65, 166-173.

Rand, A. S. & Dudley, R. 1993. Frogs in helium: The anuran vocal sac is not a cavity resonator. Physiological Zoology, 66, 793-806.

Rendall, D., Owren, M. J. & Ryan, M. J. 2009. What do animal signals mean? Animal Behaviour, 78, 233-240.

Robertson, J. G. M. 1990. Female choice increases fertilization success in the Australian frog, Uperoleia laevigata. Animal Behaviour, 39, 639-645.

Rosenthal, G. G., Rand, A. S. & Ryan, M. J. 2004. The vocal sac as a visual cue in anuran communication: an experimental analysis using video playback. Animal Behaviour, 68, 55- 58.

Ruxton, G. D. & Schaefer, H. M. 2011. Resolving current disagreements and ambiguities in the terminology of animal communication. Journal of Evolutionary Biology, 24, 2574–2585.

Ryan, M. J. 1988. Constrains and patterns in the evolution of anuran acoustic communication. In: The evolution of the amphibian auditory system (Ed. by B. Fritzsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington & W. Walkowiak), pp. 637-677. New York: John Wiley and Sons.

Ryan, M. J. & Keddy-Hector, A. 1992. Directional patterns of female mate choice and the role of sensory biases. American Naturalist, 139, 4-35.

Schwartz, J. J. & Wells, K. D. 1983. An experimental study of acoustic interference between two species of neotropical treefrogs. Animal Behaviour, 31, 181-190.

Searcy, W. A. & Nowicki, S. 2005. The Evolution of Animal Communication: Reliability and Deception in Signaling Systems. Princeton: Princeton University Press.

Taylor, R. C., Klein, B. A., Stein, J. & Ryan, M. J. 2008. Faux frogs: multimodal signalling and the value of robotics in animal behaviour. Animal Behaviour, 76, 1089.

Taylor, R. C., Klein, B. A., Stein, J. & Ryan, M. J. 2011. Multimodal signal variation in space and time: how important is matching a signal with its signaler? Journal of Experimental Biology, 214, 815-820.

Tinbergen, N. 1951. The Study of Instinct. Oxford: Oxford University Press.

Vasudevan, K. 2001. A foot flagging frog from the Western Ghats. Cobra, 44, 25-29.

Zelick, R. & Narins, P. M. 1985. Characterization of the advertisement call oscillator in the frog Eleutherodactylus coqui. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, 156, 223-229.

Appendix:

A. Micrixalus saxicola a foot-flagging frog from India: Acoustic and visual signaling behavior during agonistic interactions. Stiegler M., Preininger D., Gururaja K.V., Vijayakumar S.P. and W. Hödl Manuscript currently being processed Contribution: Conceived and designed the study: DP, WH. Performed the experiments: DP, KVG, SPV, WH. Analyzed the data: MS, DP. Contributed materials/analysis tools: DP, WH. Wrote the paper: MS, DP, WH.

B. Females do have a say in the matter: Female and male vocalizations and laryngeal structures in the foot-flagging frog species Staurois guttatus Preininger D., Handschuh S., Sztatecsny M. and W. Hödl. Manuscript currently being processed Contribution: Conceived and designed the study: DP, MS, WH. Performed the experiments: DP, SH. Analyzed the data: DP, SH. Contributed materials/analysis tools: DP, SH, MS WH. Wrote the paper: DP, SH, MS, WH.

C. Don’t get the blues: conspicuous nuptial colouration of male moor frogs (Rana arvalis) supports visual mate recognition in large breeding aggregations. Sztatecsny M., Preininger D., Freudmann A., Loretto M-C., Maier F. and W. Hödl. 2012. Publication released: Behavioral Ecology and Sociobiology; DOI: 10.1007/s00265-012-1412- 6Online First™Open Access Contribution: Conceived and designed the experiments: MS, WH. Performed the experiments: MS, DP, AF, MCL, FM, WH. Analyzed the data: MS. Contributed materials/analysis tools: DP, MS, WH. Wrote the paper: MS. 96 APPENDIX A 97

Micrixalus saxicola a foot-flagging frog from India: Acoustic and visual signaling behavior during male-male agonistic interactions (working title)

Michael J. Stiegler1, Doris Preininger1, K.V. Gururaja2, S.P.Vijayakumar3 and Walter Hödl1

1Department of Evolutionary Biology, University of Vienna, Althanstrasse 14, A-1090 Vienna,

Austria

2Centre for Infrastructure, Sustainable Transportation and Urban Planning (CiSTUP), Indian

Institute of Science, Bangalore - 560 012, Karnataka, India

3Centre for Ecological Sciences, Indian Institute of Science, Bangalore - 560 012, Karnataka,

India

Abstract

Several anuran species use acoustic and visual signals for inter- and intraspecific communication in diverse social contexts. Our study describes acoustic and visual behaviors of the Small Torrent Frog (Micrixalus saxicola) a diurnal ranid frog endemic to the Western

Ghats of India. During agonistic interactions males display advertisement calls, foot-flagging and taping (foot-lifting) behaviors to signal the readiness to defend perching sites in perennial streams. Our results from a quantitative video analysis of 10 male-male interactions indicate that foot-flagging displays were used as directional signals toward the opponent male, but were less abundant than calls. The acoustic and visual signals were not functionally linked as reported for other foot-flagging frogs. Analysis of behavioral transitions revealed that kicking behaviors (physical attacks) significantly elicited kicks from interacting males. We suggest that foot-flagging displays ritualized from the frequently observed fighting technique (kicking) to reduce physical attacks. Together the results of this study support the assumption that foot-flagging behavior in M. saxicola represents a nascent state in the evolution of visual signaling frogs. 98 APPENDIX A

Table 1: Side preferences of foot-flagging behavior of the small torrent frog (Micrixalus saxicola).

Position of interacting individual Right Snout Vent Left Total Right 15 11 7 7 42 Foot-flags Left 3 13 7 9 30 Total 18 24 14 16 72

Table 2. Dyadic matrix of behavioral inter-individual transitions during male-male agonistic interactions of Micrixalus saxicola. Asterisks show transitions that occurred at frequencies higher than expected (P < 0.01) according to chi-square tests.

Successive behavioral unit Call Tap Foot flagging Kick Location Turn Total Call Count 75 76 6 15 54 18 244 Expected Count 85 62 6 31 47 14 244 Tap Count 38 34 3 6 12 3 96 Expected Count 33 24 2 12 18 5 96 Foot flagging Count 12 8 0 3 4 0 27 Expected Count 9 7 1 4 5 2 27 Kick Count 13 2 1 27** 18 4 65 Expected Count 23 16 2 8 12 4 65 Location Count 30 8 3 16 12 4 73 Expected Count 25 19 2 9 14 4 73 Turn Count 19 8 0 2 3 1 33 Expected Count 12 8 1 4 6 2 33 Total Count 187 136 13 69 103 30 538 % 35 25 2 13 19 6 100

APPENDIX A 99

Figure 1. Daily individual signaling activity of Micrixalus saxicola of advertisement calls, taping (foot-lifting) and foot-flagging displays. Bars show means + SE per individual and hour

(n=4).

Figure 2. Frequency of behaviors displayed by two Micrixalus saxicola males during an agonistic interaction (n=10). Box plots show the median response with interquartile range and 10th and 90th percentile. 100 APPENDIX A

Figure 3: Frequency of behaviors displayed by two Micrixalus saxicola males during an agonistic interaction (n=10). Behavioral frequencies are separated according the status of the male individual at (a) the beginning (resident or intruder) and (b) end (winner and loser) of an agonistic interaction. Box plots show the median response with interquartile range and

10th and 90th percentile. APPENDIX A 101

Figure 4. Comparison of timing relationships between advertisement call and foot-flagging display of 19 Micrixalus saxicola males. Box plots show the median response with interquartile range and 10th and 90th percentile and minimum and maximum values.

102 APPENDIX B 103

Females do have a say in the matter: Female and male vocalizations and laryngeal structures in the foot-flagging frog species Staurois guttatus (working title)

Doris Preininger1, Stephan Handschuh2, Marc Sztatecsny1 and Walter Hödl1

1Department of Evolutionary Biology, University of Vienna, Althanstrasse 14, A-1090 Vienna,

Austria

2VetCore Facility for Research, Imaging Unit, University of Veterinary Medicine,

Veterinärplatz 1, A-1210 Vienna, Austria

Abstract

In the majority of anurans males produce species-specific advertisement calls necessary for courtship and mating, whereas females generally are silent apart from release calls. We reviewed spectral and temporal characteristics of female mating and encounter calls described for 19 species. Additionally, we describe for the first time the encounter call of female Staurois guttatus and compare call characteristics to conspecific male vocalizations in light of ambient environmental noise. Sound pressure levels of noise in the frequency range of the calls had less energy than male calls but similar energy than female calls.

These results indicate that female calls could be masked by ambient stream noise. Males and females showed no differences in mean, minimum and maximum frequency, but we found clear differences in temporal characteristics. Call characteristics of males in other species were shown to correlate with morphological characteristics and in turn with laryngeal structures and a sexual dimorphism in larynx size was reported. We tested predictions that the structural elements of female larynx, compared to males are smaller regardless of their bigger body size. The anatomy of laryngeal structures was analyzed from microCT-scans and muscle volumes were extracted based on voxel segmentation. Our results show that male laryngeal muscles are approximately one third smaller than those of females; a reverse dimorphism was reported for other frogs with silent females. We discuss the functional differences in calling behavior and environmental constraints potentionally favoring the selection of laryngeal structures to increase the signal-to-noise ratio. 104 APPENDIX B

Table 1. Comparison of spectral and temporal call characteristics of male and female

Staurois guttatus. Values represent estimated means, standard errors (SE) and p-values of

Generalized Linear Mixed Models (GLMM).

Female (n = 6) Male (n = 5) GLMM results

Mean frequency [Hz] 4234 (SE 34) 4195 (SE 50) F1/761 =0.813; p=0.367 Minimum frequency [Hz] 3661 (SE 29) 3699 (SE 45) F1/761 =0.884; p=0.347 Maximum frequency [Hz] 4807 (SE 41) 4747 (SE 62) F1/761 =1.279; p=0.258 Call duration [s] 3.06 (SE 0.19) 0.22 (SE 0.28) F1/106 =68.443; p<0.001 Note number/call 12.1 (SE 0.7) 1.8 (SE 0.9) F1/106 =73.943; p<0.001 Note duration [s] 0.033 (SE 0.001) 0.041 (SE 0.001) F1/765 =75.769; p<0.001

Table 2. Absolute and mean values of female and male Staurois guttatus morphological characteristics.

Female Male Characteristics 1 2 3 mean (±SD) 1 2 3 mean (±SD)

Size and weight Snout-vent-length (mm) 50.4 50.5 49.3 50.1 (0.7) 33.3 34.5 34.7 34.2 (0.8) Snout-urostyl-length (mm) 49.1 47 48.2 48.1 (1.1) 32.6 33.4 33.3 33.1 (0.4) Head width (mm) 15.2 14.2 13.2 14.2 (1.0) 9.5 9.2 9.6 9.4 (0.2) Weight (g) 9.91 8.17 7.07 8.38 (1.43) 2.17 2.29 2.56 2.34 (0.20)

Laryngeal measurements Dilator muscle volume (mm3) 2.664 2.601 2.960 2.741 (0.192) 1.794 1.182 1.851 1.609 (0.371) Constrictor muscle volume (mm3) - 1.263 1.549 1.406 (0.202) 0.836 0.848 - 0.842 (0.008)

APPENDIX B 105

Table 3. Distribution of female mating and encounter vocalization among anuran species. Former species names as used by the authors in parentheses. Call types include (1) advertisement-, (2) courtship-, (3) reciprocal-, (4) duet-, (5) aggressive-, (6) territorial calls. Mean dominant frequency, call duration, snout-vent length (SVL) and respective standard deviation (SD) are presented if not indicated other wise. NA = information not available, SE = standard error.

call mean dominant frequency mean call duration Family Species mean SVL Reference type [kHz] (± SD) [ms] (± SD) [ms] (± SD) Mating call Alytidae Alytes cisternasii female 3 1.40 (0.05, n=19) 144 (173, n=13) range = 34-43 Bosch & Márquez (2001) male 1 1.44 (0.04, n=14) 149.4 (12.4, n=14) range = 33-39 Márquez & Verrell (1991) Alytes muletensis female 2,3 1.70 (0.16, n=11) 62 (15, n=11) 38 a Bush (1997) male 1 1.80 (0.14, n=28) 102 (17, n=28) 30.6 (2.4, n=28 ) Alytes obstetricans female 3 1.38 (n=1) 119 (n=1) 47 (n=1) Heinzmann (1970) male 1 1.34 (n=1) 162 (n=1) 45 (n=1) Craugastoridae Craugastor podiciferus (Eleutherodactylus podiciferus) female 3 3.10 (n=3) 57,7 (n=3) 24.1 (n=1) Schlaepfer & Fieroa-Sandí (1998) male 1 2.7 (n=2) 43,7 (n=2) 15.9 (n=1) Ceratobatrachidae Platymantis vitiensis female 2 0.92 (n=1) 22100 (n=1) 56.5 (n=1) Boistel & Sueur (1997) male 1 2.10 (n=1) 17.4 (n=1) 35.7 (n=1) 106 APPENDIX B

Table 3. (cont.) Dicroglossidae Euphlyctis cyanophlyctis (Rana cyanophlyctis) female 3 0.74 (0.04, n=12) b 20 (4, n=12) b NA Roy et al.(1995) male 1 1.65 (0.04, n=34) b 615 (155, n=34) b 69 Daniels (2005) Fejervarya limnocharis (Rana limnocharis) female 3 1.53 (0.20, n=14) b 61 (27, n=14) b 60 c Roy et al.(1995) male 1 2.14 (1.25, n=40) b 503 (101, n=40) b 39-43 a Eleutherodactylidae Eleutherodactylus guanahacabibes female 2 2.03 (0.14, n=1) 56.4 (n=1) NA Diaz & Estrada (2000) male 1 2.39 (0.52, n=5) NA NA Pelobatidae Pelobates cultripes female 4 0.58 (n=5) 68.76 (n=5) 74.6 (5.7, n=66) Lizana et al. (1994) male 4 0.54 (n=5) 69 (n=5) 71.9 (6.0, n=76) Pelobates fuscus female 2 d 5.93 (n=8) e NA 58.1 (2.7, n=8) Andreone & Piazza (1990) male 1 4.47 (n=6) e NA 47.3 (3.5, n=6) Pipidae Xenopus laevis female 2 0.20 500 (300, n=8) 110 a Tobias et al. (1998) male 1 1.80 (SE 0.03) 83 a Wetzel & Kelly (1983) Ranidae Babina daunchina female 4 1.30 (n=2) 3195 (777, n=2) 45-50 f Cui et al. (2010) male 4 NA NA 45-50 f APPENDIX B 107

Table 3. (cont.) Clinotarsus curtipes (Rana curtipes) female 2 0.93 (0.25, n=13) 60 (10, n=13) 59.2 (4.2, n=38) Krishna & Krishna (2005) male 1 1.22 (0.49, n=22) 1009 (457, n=21) 46.2 (2.3, n=40) Hylarana erythraea (Rana erythraea) female 3 1.05 (0.11, n=14) e 32 (9, n=14) e 78 a Roy et al.(1995) male 1 2.46 (0.04, n=15) e 224 (4, n=15) e 48 a Lithobates virgatipes (Rana virgatipes) female 3 0.72 (n=2) NA 55 (n=2) Given (1987) male 1 0.46-0.72 (n=2) NA 52 (n=2) Odorrana tormota female 2 7.2 - 9.8 < 150 56 Shen et al. (2008) male 1 7.24 (0.96, n=43) ≤ 40 (n=48) 32.5 Feng (2002) Encounter call Eleutherodactylidae Eleutherodactylus coqui female 5 1.10-1.50 (n=6) 1050 (120, n=6) 44 (n=25) Stewart & Rand (1991) male 5 1.40-1.60 (n=6) 1140 (120, n=6) 34 (n=35) Ranidae Lithobates catesbeianus (Rana catesbeiana) female 6 0.3-0.5 1400-1800 125 a Capranica (1967) male 6 0.5-0.8 400-600 95-110 a Dendrobatide Colostethus inguinalis female 5 2.5 g NA 27 (n=141) Wells (1980) male 1 3.20-4.55 (n=6) NA 25 (n=90) a estimates retrieved from “amphibiaweb.org”; b n = number of calls, not number of individuals recorded; c estimates retrieved from “frogsofborneo.org”; d also duet call data available; e maximum frequency; f estimates for the species; g approximation of the author.

108 APPENDIX B

Table 4. Female vocalizations mentioned without available data on call characteristics.

Family Species call type Reference Bombinatoridae Bombina variegata courtship Savage (1932) Ceratobatrachidae Ceratobatrachus guentheri courtship Yoshimi et al. (1996) Dicroglossidae Limnonectes leporinus (Rana blythi) courtship Emerson (1992) Limnonectes poilani (Rana blythi)* courtship Orlov (1997) Eleutherodactylidae Eleutherodactylus angustidigitorum reciprocal Dixon (1957) Hyperoliidae Afrixalus fornasini agressive Stewart (1967) Hyperolius marmoratus marginatus agressive Stewart (1967) Ranidae Pelophylax esculentus (Rana esculenta) reciprocal Wahl Pelophylax ridibundus (Rana ridibunda) reciprocal Fraser (1983) Rhacophoridae Polypedates leucomystax agressive Roy (1997)

APPENDIX B 109

Figure 1. Multi-note calls of female (a-b) and male (c) Staurois guttatus. Waveform (± 0.5 amplitude relative 20 μPa) and spectrogram of a female territorial call (a) and a close-up of the two indicated notes (b). A male advertisement call (c) recorded at the stream.

Spectrogram settings: FFT method; window length: 0.005; time step: 1000; frequency step:

1000; Gaussian window; dynamic range: 40 dB (a-b), 20 dB (c).

110 APPENDIX B

Figure 2. Comparison of sound pressure of female and male calls of Staurois guttatus and the background noise. Shown here are estimated means (points), standard errors (boxes) and 95% confidence intervals (whiskers) of female territorial calls, male advertisement calls, background noise and noise filtered in the frequency range of female and male calls. Values without the same superscript letter (a, b, c) differ significantly at p < 0.001.

APPENDIX B 111

Figure 3. Cross sections of the laryngeal structures of female and male Staurois guttatus retrieved from microCT-scans. (For absolute values of dilator and constrictor muscle volume see Tab. 2) 112

APPENDIX C 113 Behav Ecol Sociobiol DOI 10.1007/s00265-012-1412-6

ORIGINAL PAPER

Don’t get the blues: conspicuous nuptial colouration of male moor frogs (Rana arvalis) supports visual mate recognition during scramble competition in large breeding aggregations

Marc Sztatecsny & Doris Preininger & Anita Freudmann & Matthias-Claudio Loretto & Franziska Maier & Walter Hödl

Received: 16 June 2012 /Revised: 3 September 2012 /Accepted: 5 September 2012 # The Author(s) 2012. This article is published with open access at Springerlink.com

Abstract Conspicuous male colouration is expected to have male moor frogs spent significantly more time in contact and evolved primarily through selection by female choice. In what in amplexus with the brown model than with the blue model. way conspicuous colours could be advantageous to males Our results suggest that the nuptial colouration in moor frogs scrambling for mates remains largely unknown. The moor can act as a new type of visual signal in anurans evolved to frog (Rana arvalis) belongs to the so-called explosive promote instantaneous mate recognition allowing males to breeders in which spawning period is short; intrasexual com- quickly move between rivals while scrambling for females. petition is strong, and males actively search and scramble for females. During breeding, male body colouration changes Keywords Anuran amphibians . Colour dimorphism . from a dull brown (similar to females) to a conspicuous blue, Explosive breeders . Intrasexual selection . Mate and we wanted to test if male blueness influences mating recognition . Scramble competition . Visual signals success or facilitates male mate recognition. To do so, we first measured the colour of mated and non-mated males using a spectrophotometer. In an experiment, we then analysed interactions of actual male moor frogs in natural spawning aggre- Introduction gations with a brown (resembling a female or a non-breeding male) and a blue model frog. Mated and non-mated males did Bright colouration and ornaments presented by males during not differ in colouration, suggesting that female choice based thematingseasonareacommonphenomenoninmany on colour traits was unlikely. In our behavioural experiment, animal taxa and are expected to evolve under sexual selection (Andersson and Iwasa 1996). Sexual selection might act Communicated by K. Warkentin through various mechanisms, but female mate choice is Electronic supplementary material The online version of this article most commonly invoked in explaining the evolution of (doi:10.1007/s00265-012-1412-6) contains supplementary material, conspicuous male characters. Well-known examples include which is available to authorized users. colour patches in the house finch (Hill 1991), red nuptial M. Sztatecsny (*) : D. Preininger : F. Maier : W. Hödl colouration in sticklebacks (Milinski and Bakker 1990) and Department of Evolutionary Biology, University of Vienna, long tail feathers in widowbirds (Andersson 1982). Intrasex- Althanstrasse 14, ual selection through male contests has been assumed to 1090 Vienna, Austria e-mail: [email protected] favour the evolution of weapons, such as horns and antlers or signals indicating male condition or fighting ability A. Freudmann (Bradbury and Vehrencamp 2011). Examples of male orna- Department of Tropical Ecology and Animal Biodiversity, ments not related to weapons but rather to enhanced conspic- University of Vienna, Rennweg 14, uousness seem less common but have been observed in wolf 1030 Vienna, Austria spiders (Hebets and Uetz 2000)andAnolis lizards (Charles and Ord 2012). Another, less-studied mechanism of sexual M.-C. Loretto selection is scramble competition in which males attempt to Department of Cognitive Biology, University of Vienna, Althanstrasse 14, find a mate before rivals do. Successful males are expected to 1090 Vienna, Austria have well-developed sensory organs and show high mobility 114 APPENDIX C Behav Ecol Sociobiol

(Andersson and Iwasa 1996), but very little is known about in the common toad (Bufo bufo), clasp onto any animate or colour displays in scrambling species. inanimate object including conspecific males, other amphib- Anuran amphibians are well-established model organisms ian species, dead fish or even beer cans (Reading 1984; regarding their acoustic communication and role of call fea- Marco and Lizana 2002; MS personal observation). Males tures in female choice or male–male agonistic behaviour clasped by other males perform release calls or vibrations (Gerhardt and Huber 2002). Understanding parallel issues in to be released (Wells 2007). Nonetheless, ten or even more colour displays, such as the vocal sac or foot webbings of foot- common toad males may cling to one female (Verrell and flagging frog species, has increased during the last decade McCabe 1986; Sztatecsny et al. 2006), and after breeding, (Hödl and Amezquita 2001). In the European tree frog, many drowned females can be found at breeding sites (MS females have been shown to prefer males with bright orange personal observation). vocal sacs (Gomez et al. 2009), and among competing males The European moor frog (Rana arvalis)isatypical of Micrixalus saxicola and the Bornean foot-flagging frogs of explosive breeder but differs from many similar species by the genus Staurois, the vocal sac and conspicuous foot the spectacular blue and dynamically changing nuptial col- webbings respectively appear to facilitate the detection ouration of males, which is maintained during the breeding of a displaying individual (Grafe et al. 2012; Preininger et al. period (Ries et al. 2008a;Fig.1). This temporal colour accepted). Visual cues in anurans with scramble competition dimorphism has been known for more than 100 years belonging to the so-called explosive breeders (Wells 2007) (Brehm 1893), but the question of its communicative sig- have so far gained less attention, despite the striking colour nificance has remained largely unresolved. Male blueness changes occurring in some species (Ries et al. 2008a;Doucet has been reported to be more intense in mated compared to and Mennill 2010; Sztatecsny et al. 2010). In explosive non-mated males (Hedengren 1987; Hettyey et al. 2009c) breeders, all sexually receptive individuals arrive almost syn- and to influence offspring survival (Sheldon et al. 2003) but chronously at the spawning site, and breeding takes place over could not be linked to male fertility (Hettyey et al. 2009b). a period of only a few days to weeks. The operational sex ratio Ries et al. (2008a) measured male colour using a spectro- (OSR) at spawning sites is generally male biassed, (approxi- photometer and found no effect of male body condition or mately 50–95 % males, depending on species and population; the presence of females on colouration. Considering previ- Wells 2007), and in dense spawning aggregations, males do ous results and moor frog breeding behaviour, we wanted to not call to attract individual females but, instead, search and know if male colouration affects (1) mating success, (2) scramble for access to mates. As females are frequently co- mate recognition by males or (3) both. To test these hypoth- erced by males, their options for active choice are limited (but eses, we quantified colouration of mated (i.e. in amplexus see Hettyey et al. 2009a; Sherman et al. 2010). Still, only with a female) and non-mated male moor frogs using spec- about 5 % of the males may breed successfully (Wells 2007). trophotometry. To experimentally investigate whether body In the struggle to achieve mating success, males, as observed colouration influenced male mating and harassment

Fig. 1 A breeding aggregation of male moor frogs (R. arvalis) (a) and a pair of moor frogs in amplexus (b) APPENDIX C 115 Behav Ecol Sociobiol behaviour, we used a blue and a brown model frog resem- amplexus with a female) males with a dip net from two bling a breeding male and a female or non-breeding male. In spawning aggregations and immediately obtained spectral support of the first hypothesis, we expected to detect differ- data of each individual to avoid any colour change due to ences in colouration between mated and non-mated males, handling. An Ocean Optics Jaz spectrometer with integrated respectively. In support of the mate-finding hypothesis, we pulsed xenon light source (Jaz-PX) was used to measure expected to find no colour differences but differences in the spectral reflectance (300–700 nm) at two spots (three meas- response of male moor frogs towards the model frogs, whereas urements per spot): the tympanic membrane (because it is differences in both cases would back up our third hypothesis. clearly defined) and the flank (because it shows the highest variability in colouration; Ries et al. 2008a). We then measured body mass of each male to the nearest 0.1 g using an Methods electronic miniscale and snout-urostyle length (SUL) to the nearest 0.1 mm using vernier callipers and released all indi- Study species viduals immediately after data collection 100 m away from the spot of capture. From the obtained body measurements, we b The moor frog occurs from Central and Eastern Europe computed the scaled mass index Mi (Peig and Green 2009)as through Scandinavia eastwards to Siberia and China an index of body condition (IBC): (AmphibiaWeb 2012). Our study population belongs to the bSMA southern cluster formerly recognised as the subspecies R. b L0 Mi ¼ Mi arvalis wolterstorffi (Babik and Rafinski 2000). The dorsal Li body colouration in both sexes of the moor frog outside the where M and L are the body mass and the SUL of breeding period is generally beige brown, grey and rufous i i individual i, respectively; b is the scaling exponent with darker and lighter stripes running along the back. SMA estimated by the standardised major axis (SMA) regression During their migration to the breeding pond, male colour- of M on L; L is the mean SUL for our population. The ation changes from beige to an almost uniformly dark grey- 0 SMA regression was performed with the freeware program ish brown (some males appearing almost pinkish before SMATR (Falster et al. 2006). Male SUL and IBC have turning grey). Males turn blue when they enter the pond been shown to influence male mating success in explosive (Fig. 1; Ries et al. 2008a, b). Upon arriving at the breeding breeders and could be indicative of male quality (Davies and pond, males establish large assemblages of several hundred Halliday 1977; Höglund and Säterberg 1989; Vásquez and individuals around suitable communal spawning sites Pfennig 2007; Hettyey et al. 2009d). (Fig. 1). The low-intensity advertisement calls (males lack an external vocal sac) are assumed to stimulate females to Model frog experiments join the aggregations when they are ready to spawn (Glandt 2006). Male blueness, which is maintained for a few days, Model frog design and calling activity increase with increasing temperature (Glandt 2006; Hettyey et al. 2009c), and in our study pop- To make our model frogs, we created a silicone cast from a ulation, both reached the highest intensity during sunny preserved male moor frog specimen and filled it with poly- afternoons (MS personal observation). We frequently ob- urethane resin (Neukadur MultiCast 1, Altropol, Stocklsdorf, served male moor frogs actively searching for approaching Germany). The resulting cast frogs were fitted with artificial females, attempting to take over mated females scrambling glass eyes and airbrush painted with acrylics in either brown among each other. Mating balls of several individuals cling- ing to a female (Verrell and McCabe 1986; Sztatecsny et al. 2006) seem to be absent (Ries et al. 2008a).

Frog sampling and spectral reflectance measurements

We collected data for a population of moor frogs at a pond in eastern Austria (47°10′ N, 16°5′ E). As estimated by a spawn clump census, the number of breeding adults exceeded 5,000 individuals in 2011, and spawning took place in dense assemblages where male densities reached 20 individuals m−2. In both study seasons (2010 and 2011), spawning activity lasted for 3 days from March 25 to 27. On March 26, 2011, Fig. 2 Blue and brown model frogs presented to actual male moor we randomly captured 19 non-mated and 20 mated (in frogs 116 APPENDIX C Behav Ecol Sociobiol or blue resembling the base colouration of non-breeding males models into the water and remained still for 5 min for the or females and that of breeding males as assessed by the easily startled frogs to resume natural behaviour. The han- human eye (Fig. 2). Commercial paints absorb in the UV dler rhythmically moved the T-shaped handle and the at- light, resulting in a maximum reflection of the blue model tached model frogs a few centimetres up and down (this was frog at longer wavelengths than in the actual moor frogs necessary because moor frogs would not respond to still (Fig. 3). Finally, a clear coat was sprayed over the models to models). Interactions between frogs and models were video protect the paint from water and add a realistic sheen. To recorded for 5 min. Subsequently, the position of the model present the live moor frogs with the models, we attached a frogs was changed to test different males, and after 2 min of blue and a brown model 70 cm apart from each other on motionlessness (again for the frogs to resume normal activity), fishing rods (1.2 m long) that were connected to the two ends the next trial began. We recorded nine trials on 26 March 2010 of the cross section of a T-shaped handle. The handle allowed and six and five trials, respectively, on 26 and 27 March 2011. keeping a constant distance between the model frogs and The videos were analysed with the program Solomon coder moving both models synchronously with the handler keeping (Péter 2011) at a time resolution of 0.30 s. We recoded the a distance of approximately 1.5 m from models and live frogs. time (as the number of 0.30 s intervals) during which there was an active physical contact between frogs and either the Experiments and data collection blue or the brown model (i.e. pushing with snout and/or fore limbs) or a frog attempted to clasp (amplex) a model. We set up a waterproof camcorder (Panasonic SDR-SW20) on a tripod in spawning aggregations with high densities of Spectral data and statistical analysis male moor frogs. The model frog handler lowered the We used the program Avicol v.6 (Gomez 2006) to extract the 50 colour parameters total brightness, hue and UV-blue chroma mated males Tympanum from our reflectance spectra. Brightness refers to the intensity non-mated males 40 brown model frog of the reflectance spectrum (calculated as the area under the blue model frog spectral curve); hue corresponds to the everyday notion of colour (calculated as the wavelength of maximum reflec- 30 tance), and chroma, to its saturation or how concentrated the reflectance is around a wavelength (calculated as the reflec- 20 tance in the interval 300–450 nm divided by total brightness). As we found no differences between mated and non-mated

10 males captured at the spawning assemblages in any of the parameters (i.e. all had turned blue almost equally), we refrained from applying any vision models. To analyse a 0 possible relationship between colour variables and body mass, 300 400 500 600 700 SUL and IBC of male frogs, we applied Spearman’scorrela- 40 tions. We used a nonparametric test for the experimental data Flank because the variables ‘contact’ and ‘amplexus’ with the

mean reflectance [%] blue model were not normally distributed (Shapiro–Wilk 30 test, P<0.05 in both cases). All statistical analyses were performed with Stata/SE 11.0 (StataCorp 2009).

Results 10 Mated male moor frogs differed neither significantly in the colouration variables brightness, hue and chroma (Table 1, 0 Fig. 3) nor in SUL, body mass or IBC from their non-mated 300 400 500 600 700 rivals. Average SUL ± standard error (SE) of the 19 non- wavelength [nm] mated and the 20 mated males was 56.96±1.13 and 56.57± 0.8 mm; mean body mass ± SE was 27.28±1.77 and 25.26± Fig. 3 Reflectance spectra of the brown and the blue model frogs and b mean reflectance spectra ± SE of the tympanic membrane and the flank 1.49 g, and mean Mi ± SE was 27.15±1.67 and 25.23±1.42, of mated (n020) and non-mated (n019) male moor frogs respectively (Wilcoxon signed-rank test, P>0.1 for all APPENDIX C 117 Behav Ecol Sociobiol

Table 1 Comparison of body colouration parameters of non-mated female mate choice. By increasing a male’s conspicuousness and mated male moor frogs and its contrast to females, the blue colouration can act as a Non-mated males Mated males ZP new type of visual signal directed at other males evolved to Mean ± SE Mean ± SE promote swift mate recognition and mate finding in dense aggregations. Among competing individuals, a signal has been Flank suggested to remain reliable when competitors at least partially Total brightness 6,241.35±632.78 5,602.49±739.43 0.87 0.38 share an interest in common (Lachmann and Bergstrom 1998; Hue 418.84±21.72 414.5±21.93 1.28 0.20 Szamado 2011). All male moor frogs are rivals while trying to − Blue chroma 0.43±0.01 0.43±0.01 0.08 0.93 find a female, but it should be in the interest of all males not to Eardrum miss a mating opportunity and to reduce mate-searching effort Total brightness 8,354.4±718.06 7,089.03±476.24 1.64 0.10 because females are limited (Johnstone et al. 1996; Loman and Hue 404.16±8.2 391.2±2.98 0.66 0.51 Madsen 2010). Mate-searching behaviour might be energeti- Blue chroma 0.42±0.02 0.44±0.003 0.16 0.88 cally demanding for males of explosive breeders, as they can lose up to 1 % of their total body mass/day during the reproductive period (Arak 1983;Ryser1989). A conspicuous visual comparisons). We also found no significant correlation be- signal making breeding males distinguishable from females tween colour variables and male frogs’ body mass, SUL and and non-breeding males can reduce mate-searching time by (1) b Mi (Spearman correlation, P>0.07inallcases).Inthe facilitating the visual detection of females, (2) minimising behavioural experiments, male moor frogs spent almost four unwanted clasping attempts with rivals and (3) reducing ha- times as much time in contact and in amplexus with the brown rassment by rivals. A male not expressing the blue body as with the blue frog model with the differences being highly colouration would be subject to frequent mating attempts by significant (Wilcoxon signed-rank test; contact, z03.008, n0 other males and would only be released after emitting a release 20, P<0.001; amplexus, z03.17, n020, P<0.002; Fig. 4). call (Liao and Lu 2009; Chen and Lu 2011). Release calls emitted by male moor frogs have a significantly lower mean sound pressure level (34 dB, n09 males) at 1-m distance 0 Discussion compared to calls of the common toad (46 dB, n 12 males), in which males do not differ from females in body colour 0 0 Our study showed that mating success in male moor frogs was (t test, t 6.78, DF 19, P<0.001, Preininger unpublished not significantly influenced by colouration. Males, however, data). A visual signal may function more rapidly than calls were able to discriminate model frogs and favoured the brown allowing individuals to communicate instantaneously and model, similar in colour to a female or a non-breeding male therefore move more quickly among rivals while scram- compared to the blue model, resembling a rival breeding male. bling for mates. Colour displays reducing male mating We therefore argue that the nuptial colouration of male moor attempts are known from female lizards signalling non- frogs is unlikely to indicate male condition or to be subject to receptivity (Chan et al. 2009) and female damselflies mimicking male body colouration in order to reduce male

100 harassment in species with strong sexual conflict (Van Gossum et al. 2011;XuandFincke2011).

*** Surprisingly, mated males in our study population did not 80 blue differ significantly from non-mated males in body size, mass or IBC, traits that should be advantageous during direct male–

60 male competition (Arak 1983). A number of previous studies brown on explosive breeders found large male advantage (e.g. Davies and Halliday 1977; Hedengren 1987;Wells2007 and referen- 40 ces therein), whereas others failed to do so (e.g. Höglund and mean time [s]

*** Robertson 1987;Wells2007 and references therein; Greene 20 and Funk 2009; Hettyey et al. 2009c). The observed discrep- ancies in study outcomes indicate that mating systems of explosive breeders are dynamic, with mating behaviour likely 0 contact amplexus varying with population size, individual density and time available for competition (Höglund and Robertson 1988; Fig. 4 Mean time per trial that male moor frogs spend in active contact Höglund 1989; Wells 2007). It also has been suggested that or in amplexus with a blue model frog resembling the colour of a breeding male and a brown model frog resembling a non-breeding mate-searching tactics in small males differ from those of large male (and also the colouration of females) males, with large males being more successful in achieving 118 APPENDIX C Behav Ecol Sociobiol takeovers of females in direct competition with already mated Conflict of interest None. males (Arak 1983; Höglund and Robertson 1988). Small males Ethical standards All experiments reported in this article comply with may profit from higher mobility compared to large males as the current laws of Austria, the country in which they were performed. indicated in males of scrambling insect species (Moya-Larano et al. 2007; Kelly et al. 2008) and attempt to find approaching Open Access This article is distributed under the terms of the Crea- females on the edge of breeding aggregations. tive Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and The actual process of colour change in male moor frogs has the source are credited. not yet been investigated, but it is assumed that blue colour in amphibians derives from incoherent light scattering from reflecting platelets in specific chromatophores, the iridophores (Bagnara et al. 2007). Shifts in the adjacent chromatophore References layers of yellow xanthophores and dark melanophores lead to rapid colour change. The dependence of blueness on temper- AmphibiaWeb: Information on amphibian biology and conservation. ature (Hettyey et al. 2009c), which should be linked to a (2012) Available at: http://amphibiaweb.org. Accessed: April 2012 general level of activity in ectotherms and the ability to change Andersson M (1982) Female choice selects for extreme tail length in a widowbird. Nature 299:818–820 colouration, suggests that male blueness could be adjusted to Andersson M, Iwasa Y (1996) Sexual selection. Trends Ecol Evol the intensity of intrasexual competition. Traits involved in 11:53–58 male–male competition have been shown to be influenced by Arak A (1983) Male-male competition and mate choice in anuran population density and OSR (Knell 2009; Bretman et al. 2011). amphibians. In: Bateson P (ed) Mate choice. Cambridge University Press, Cambridge, pp 181–210 For instance in dung beetles, males from species with low- Babik W, Rafinski J (2000) Morphometric differentiation of the moor frog density populations and a male-biassed OSR develop horns (Rana arvalis Nilss.) in Central Europe. J Zool Syst Evol Res used as weapons for intrasexual aggressive behaviour (Pomfret 38:239–247 and Knell 2008). In species with high population density and Bagnara JT, Fernandez PJ, Fujii R (2007) On the blue coloration of vertebrates. Pigm Cell Res 20:14–26 female excess, males scramble for matings and lack horns Bradbury JW, Vehrencamp SL (2011) Principles of animal communi- enhancing their mobility and ability to find females. The social cation, 2nd edn. Sinauer, Sunderland environment males that are confronted with may not only vary Brehm AE (1893) Brehms Thierleben: allgemeine Kunde des Thierreichs, between but also within species, and males may respond with Kriechthiere, Lurche und Fische: 1. Die Kriechthiere und Lurche. Bibliographisches Institut Leipzig und Wien behavioural plasticity by choosing alternative mating strategies Bretman A, Gage MJG, Chapman T (2011) Quick-change artists: male (Bretman et al. 2011). For instance, male common toads have plastic behavioural responses to rivals. Trends Ecol Evol 26:467–473 been shown to switch from chorusing behaviour at low densi- Chan R, Stuart-Fox D, Jessop TS (2009) Why are females orna- ties to active searching and scrambling when the number of mented? A test of the courtship stimulation and courtship rejec- tion hypotheses. Behav Ecol 20:1334–1342 breeding individuals increases (Höglund and Robertson 1988). Charles GK, Ord TJ (2012) Factors leading to the evolution and Blueness of male moor frogs might also vary in response to maintenance of a male ornament in territorial species. Behav Ecol the social environment and would be predicted to be most Sociobiol 66:231–239 intense in large, high-density populations such as our study Chen W, Lu X (2011) Sex recognition and mate choice in male Rana kukunoris. Herpetol J 21:141–144 population. Davies NB, Halliday TR (1977) Optimal mate selection in the toad Sexual dimorphism in colour or pattern is known to Bufo bufo. Nature 269:56–58 occur in only 25 anuran species (Hoffman and Blouin Doucet SM, Mennill DJ (2010) Dynamic sexual dichromatism in an – 2000), but male nuptial colour change in explosive explosively breeding neotropical toad. Biol Lett 6:63 66 Falster DS, Warton DI, Wright J (2006) SMATR: standardised major breeders has been reported for at least ten species from axis tests & routines version 2.0. Computer program available at six families (Supplementary Material), emphasising the http://bio.mq.edu.au/ecology/SMATR/index.html importance of body colouration as a social signal in this Gerhardt HC, Huber F (2002) Acoustic communication in insects and group. Explosive breeding species showing temporal col- anurans: common problems and diverse solutions. University of Chicago Press, Chicago our dimorphism could therefore be an ideal study system Glandt D (2006) Der Moorfrosch—Einheit und Vielfalt einer Braun- to better understand flexible sex roles and signal plastic- froschart. Laurenti, Bielefeld ity in variable ecological and social environments. Gomez D (2006) AVICOL, a program to analyse spectrometric data. Free program available at http://sites.google.com/site/avicolprogram Gomez D, Richardson C, Lengagne T, Plenet S, Joly P, Léna J-P, Théry M Acknowledgments We thank R. Illek for his help with making the (2009) The role of nocturnal vision in mate choice: females prefer model frogs, S. Számadó and two anonymous reviewers for their valuable conspicuous males in the European tree frog (Hyla arborea). Proc R comments on the manuscript, F. Kottulinsky for the access to his property, Soc Lond B 276:2351–2358 and J. Cornils, H. Krebs, M. Pluch, T. Schernhammer, I. Starnberger and Grafe TU, Preininger D, Sztatecsny M, Kasah R, Dehling JM, Proksch K. Veitschegger for their assistance in behavioural experiments. Field S, Hödl W (2012) Multimodal communication in a noisy envi- work was conducted under permit number FA13C-53S-7/2011-89, and ronment: a case study of the Bornean rock frog Staurois parvus. the study was supported by the Austrian Science Fund FWF-P22069. PLoS One 7:e37965 APPENDIX C 119 Behav Ecol Sociobiol

Greene AE, Funk WC (2009) Sexual selection on morphology in an Moya-Larano J, El-Sayyid MET, Fox CW (2007) Smaller beetles are explosive breeding amphibian, the Columbia spotted frog (Rana better scramble competitors at cooler temperatures. Biol Lett luteiventris). J Herpetol 43:244–251 3:475–478 Hebets EA, Uetz GW (2000) Leg ornamentation and the efficacy of Peig J, Green AJ (2009) New perspectives for estimating body condi- courtship display in four species of wolf spider (Araneae: Lyco- tion from mass/length data: the scaled mass index as an alternative sidae). Behav Ecol Sociobiol 47:280–286 method. Oikos 118:1883–1891 Hedengren I (1987) Selection of body size, arm length and colour in Péter A (2011) Solomon Coder (version beta 11.01.22): a simple male and female moor frogs (Rana arvalis). MSc Thesis. Univer- solution for behavior coding. Computer programm available at sity of Stockholm, Stockholm, Sweden http://solomoncoder.com/ Hettyey A, Baksay S, Vagi B, Hoi H (2009a) Counterstrategies by Pomfret JC, Knell RJ (2008) Crowding, sex ratio and horn evolution in female frogs to sexual coercion by heterospecifics. Anim Behav a South African beetle community. Proc R Soc Lond B 275:315– 78:1365–1372 321 Hettyey A, Herczeg G, Hoi H (2009b) Testing the phenotype-linked Preininger D, Boeckle M, Freudmann A, Starnberger I, Sztatecsny M, fertility hypothesis in male moor frogs (Rana arvalis) exhibiting a Hödl W (accepted) Multimodal signaling in the Small torrent frog conspicuous nuptial colouration. Amphibia-Reptilia 30:581–586 (Micrixalus saxicola) in a complex acoustic environment Behav Hettyey A, Herczeg G, Laurila A, Crochet P-A, Merila J (2009c) Body Ecol Sociobiol temperature, size, nuptial colouration and mating success in male Reading CJ (1984) Interspecific spawning between common frogs moor frogs (Rana arvalis). Amphibia-Reptilia 30:37–43 (Rana temporaria) and common toads (Bufo bufo). J Zool Hettyey A, Vagi B, Hevizi G, Toeroek J (2009d) Changes in sperm 203:95–101 stores, ejaculate size, fertilization success, and sexual motivation Ries C, Spaethe J, Sztatecsny M, Strondl C, Hödl W (2008a) Turning over repeated matings in the common toad, Bufo bufo (Anura: blue and ultraviolet: sex-specific colour change during mating Bufonidae). Biol J Linn Soc 96:361–371 season in the Balkan moor frog. J Zool 276:229–236 Hill GE (1991) Plumage coloration is a sexually selected indicator of Ries C, Sztatecsny M, Hödl W (2008b) Geschlechtsspezifischer male quality. Nature 350:337–339 Farbwechsel beim Moorfrosch (Rana arvalis) während der Hödl W, Amezquita A (2001) Visual signaling in anuran amphibians. Paarungszeit. In: Glandt D, Jehle R (eds) Der Moorfrosch/The Moor In: Ryan MJ (ed) Anuran communication. Smithsonian Institution Frog. Z Feldherpetologie Suppl. 13, Laurenti, Bielefeld, pp 127–134 Press, Washington, DC, pp 121–141 Ryser J (1989) Weight loss, reproductive output, and the cost of Hoffman EA, Blouin MS (2000) A review of colour and pattern poly- reproduction in the common frog, Rana temporaria. Oecologia morphisms in anurans. Biol J Linn Soc 70:633–665 78:264–268 Höglund J (1989) Pairing and spawning patterns in the common toad, Sheldon BC, Arponen H, Laurila A, Crochet PA, Merilä J (2003) Sire Bufo bufo: the effects of sex ratios and the time available for male coloration influences offspring survival under predation risk in the male competition. Anim Behav 38:423–429 moorfrog. J Evol Biol 16:1288–1295 Höglund J, Robertson JGM (1987) Random mating by size in a popula- Sherman CDH, Sagvik J, Olsson M (2010) Female choice for males tion of common toads Bufo bufo. Amphibia-Reptilia 8:321–330 with greater fertilization success in the swedish Moor frog, Rana Höglund J, Robertson JGM (1988) Chorusing behavior, a density- arvalis. PLoS One 5:e13634 dependent alternative mating strategy in male common toads StataCorp (2009) Stata Statistical software: release 11. StataCorp LP, (Bufo bufo). Ethology 79:324–332 College Station Höglund J, Säterberg L (1989) Sexual selection in common toads: Szamado S (2011) The cost of honesty and the fallacy of the handicap correlates with age and body size. J Evol Biol 2:367–372 principle. Anim Behav 81:3–10 Johnstone RA, Reynolds JD, Deutsch JC (1996) Mutual mate choice Sztatecsny M, Jehle R, Burke T, Hödl W (2006) Female polyandry and sex differences in choosiness. Evolution 50:1382–1391 under male harassment: the case of the common toad (Bufo bufo). Kelly CD, Bussiere LF, Gwynne DT (2008) Sexual selection for male J Zool 270:517–522 mobility in a giant insect with female-biased size dimorphism. Sztatecsny M, Strondl C, Baierl A, Ries C, Hödl W (2010) Chin up: are Am Nat 172:417–423 the bright throats of male common frogs a condition-independent Knell RJ (2009) Population density and the evolution of male aggres- visual cue? Anim Behav 79:779–786 sion. J Zool 278:83–90 Van Gossum H, Bots J, Van Heusden J, Hammers M, Huyghe K, Lachmann M, Bergstrom CT (1998) Signalling among relatives II. Morehouse NI (2011) Reflectance spectra and mating patterns Beyond the tower of babel. Theor Popul Biol 54:146–160 support intraspecific mimicry in the colour polymorphic damselfly Liao WB, Lu X (2009) Sex recognition by male Andrew’s toad Bufo Ischnura elegans. Evol Ecol 25:139–154 andrewsi in a subtropical montane region. Behav Process 82:100– Vásquez T, Pfennig K (2007) Looking on the bright side: females 103 prefer coloration indicative of male size and condition in the Loman J, Madsen T (2010) Sex ratio of breeding common toads (Bufo sexually dichromatic spadefoot toad, Scaphiopus couchii. Behav bufo)—influence of survival and skipped breeding. Amphibia- Ecol Sociobiol 62:127 Reptilia 31:509–524 Verrell PA, McCabe N (1986) Mating balls in the common toad, Bufo Marco A, Lizana M (2002) The absence of species and sex recognition bufo. Brit Herpetol Soc Bull 16:28–29 during mate search by male Common Toads Bufo bufo. Ethol Ecol Wells KD (2007) The ecology and behavior of amphibians. University Evol 14:1–8 of Chicago Press, Chicago Milinski M, Bakker TCM (1990) Female sticklebacks use male color- Xu M, Fincke OM (2011) Tests of the harassment-reduction function and ation in mate choice and hence avoid parasitized males. Nature frequency-dependent maintenance of a female-specific color poly- 344:330–333 morphism in a damselfly. Behav Ecol Sociobiol 65:1215–1227 120 121

Curriculum Vitae

Name: Doris Preininger Private address: Paletzgasse 3-9 3/2/9 1160 Vienna, Austria Work address: Department of Evolutionary Biology, University of Vienna, Althanstr. 14, 1090 Vienna, Austria Phone: +43 (0)650 420 2955 email: [email protected]

2001-2007: Studies in Zoology, University of Vienna 2007: MSc., Supervisor Walter Hödl, University Vienna: Communication in noisy environments: Signaling behavior of male foot-flagging frogs Staurois latopalmatus. 2008: pre-study for FWF grant proposal, implementation of Zoo-project 2009-today: PhD program, Supervisors Walter Hödl 06/2011: teaching competence (Centre for teaching and learning, University Vienna)

Teachings SS 2011 and SS 2012 Zoological research course WS 2011 and WS 2012 ‘Biology of tropical amphibians’ (Seminar)

Researcher in the following projects 2004: Searching for the mechanisms of colony formation in Neoplamprologus caudopunctatus. Konrad Lorenz Institute for Ethology. 2005/05: Measuring anuran species abundance and diversity between a disturbed and non-disturbed habitat in Kibale Forest National Park. Tropical Biology Association, Uganda. 2005: Investigate sex- and context-specific differences in sound characteristics of croaking gouramis (Trichopsis vittata). University of Vienna. F. Ladich. 2005/06: Study of anuran communication in noisy environments. Danum Valley, Sabah; Royal Society, South East Asia Rainforest Research Program. 2007/08: Pre-Study Multimodal communication, Kuala Belalong, Brunei, Borneo. 2008 - today: Ex-situ breeding program of Staurois guttatus and Staurois parvus. Project development and implementation in the Vienna Zoo Schönbrunn 2011/04 Nuptial coloration in Rana arvalis, Neudau, Austria 2010 – today: Multimodal signals in anurans 122

Publications (chronologic) Preininger D., Boeckle M., and W. Hödl. 2007. Comparison of anuran acoustic communities of two habitat types in the Danum Valley Conservation Area, Sabah, Malaysia. Salamandra 43, 129-138. Preininger D., Boeckle M., and W. Hödl. 2009. Communication in Noisy Environments II: Visual Signaling Behavior of Male Foot-flagging Frogs staurois Latopalmatus. Herpetologica 65, 166-173. Boeckle M., Preininger D. and W. Hödl. 2009. Communication in Noisy Environments I: Acoustic Signals of Staurois latopalmatus Boulenger 1887. Herpetologica 65, 154- 165. Grafe T.U., Preininger D., Sztatecsny M., Kasah R., Dehling M., Proksch S. and W. Hödl. 2012. Multimodal communication in a noisy environment: a case study of the Bornean rock frog Staurois parvus. PlosOne 7(5): e37965. doi:10.1371/journal.pone.0037965 Preininger D., Weissenbacher A., Wampula T. and W. Hödl. 2012. The conservation breeding of two foot-flagging frog species from Borneo, Staurois parvus and Staurois guttatus. Amphibian and Reptile Conservation 5(3):45-56(e51) Sztatecsny M., Preininger D., Freudmann A., Loretto M-C., Maier F. and W. Hödl. 2012. Don’t get the blues: conspicuous nuptial colouration of male moor frogs (Rana arvalis) supports visual mate recognition in large breeding aggregations. Behavioral Ecology and Sociobiology; DOI: 10.1007/s00265-012-1412-6Online First™Open Access Preininger D., Boeckle M., Freudmann A., Starnberger I., Sztatecsny M., and W. Hödl. 2012. Multimodal signaling in the Small Torrent Frog (Micrixalus saxicola) in a complex acoustic environment. Behavioral Ecology and Sociobiology (in press) Preininger D., Boeckle M., Sztatecsny M., and W. Hödl. 2012. Divergent receiver responses to components of multimodal signals in foot-flagging frog species offer clues to visual signal evolution. PlosOne (submitted)

Presentations 03.-07.10.2007 – oral presentation “Acoustic signaling behaviour in noisy environments” congress of the German Herpetological Society (DGHT) (Hallein, Austria) 17.-22.08.2008 – oral presentation in the symposium “Sensory ecology of anuran communication” 6th world congress of Herpetology (Manaus, Brazil) 16.-18.01.2009 – oral presentation “the foot-flagging frogs of Borneo in the Vienna Zoo” 20th annual meeting of the Austrian Herpetological Society (ÖGH) 16. 11. 2011 – oral presentation “Looking for foot-flagging frogs in Borneo” Austrian Herpetological Society (ÖGH) (Vienna, Austria) 123

22.-25.05.2012 – oral presentation “First breeding success in the foot-flagging frog (Staurois parvus) from Borneo” 'Love And Loss': European Association of Zoos and Aquaria (EAZA) Conservation Forum (Vienna, Austria) 08.-14.08.2012 – oral presentation in the symposium “Stream Amphibians around the Globe: Adaptations, Ecology, and Conservation” 7th world congress of Herpetology (Vancouver, Canada) 09.-14.09.2012 – poster presentation “First breeding success in the foot-flagging frog Staurois parvus” 8th International Aquarium Congress (IAC) (Cape Town, South Africa)

Funding 2005 Short-term grant abroad (KWA) (University Vienna) 2005 British Ecological Society Scholarship (Tropical Biology Association) 2006 Wilhelm-Peters Grant of the German Herpetological Society (DGHT) 2008 Austrian Research Association (ÖFG) “international communication” 2008 Project-grant of the “Oesterreichischen Nationalbank” and Friends of the Vienna Zoo 2010-2013 Austrian Science Foundation (FWF) Grant for the project “Multimodal signals in anurans” (co-author)

Public relations

05.-09.03.2012 Radioshow Ö1“life of nature“: Visual communication of frogs

Preininger D. and T. Wampula. 2012. Die Winkerfrösche von Borneo. Aquaristik Fachmagazin 44(3), 100-104.

I confirm that I have written the Preface, Concluding Discussion, Abstract and substantially contributed to every manuscript listed in this thesis

Wien, am 5. Oktober 2012 ………………………………………