Supplementary material for:
How acoustic signals scale with body size: common trends in static allometry across diverse taxa
Rafael L. Rodríguez, Marcelo Araya–Salas, David A. Gray, Michael S. Reichert, Laurel B. Symes, Matthew R. Wilkins, Rebecca J. Safran, and Gerlinde Höbel
Appendix 1: Case studies
Dendropsophus ebraccatus treefrogs (Anura: Hylidae)
This Neotropical frog is found in lowland rainforests throughout much of Central America. Males produce advertisement and aggressive calls, both of which appear to play a role in both male-male competition and female attraction (Wells and Schwartz 1984a,b; Reichert 2011).
Males produce a complex advertisement call consisting of a pulsed introductory note followed by a series of from 0-4 click-note appendages that primarily functions in female attraction (Wells and Schwartz 1984a). Ten advertisement calls per male were recorded from a total of 139 males in Gamboa, Panama, and averages were obtained for each call characteristic for each male. Afterwards males were captured, their snout–vent length (SVL, the standard measure of body size in anurans) was measured to the nearest 0.01 mm with calipers, as well as their mass to the nearest 0.01 g with an electronic balance. The assessed call traits were the duration of the introductory note, the number of pulses in the introductory note, pulse rate, call period, duty cycle, number of click note appendages and the dominant frequency peak. Females prefer lower frequencies (Wollerman 1998), a greater number of click notes (Wells and Schwartz 1984a), shorter call periods (and therefore presumably greater duty cycles; Wells and Bard 1987), and pulse rates centered around the population mean (Wollerman 1998). Female preferences for introductory note duration and pulse number have not been tested.
Aggressive signals are similar to advertisement signals but have a greater pulse rate in the introductory note (Wells and Schwartz 1984b). At least 5 aggressive calls per male were recorded from 71 males and averages were obtained for each aggressive-call trait. The assessed aggressive-call traits were the duration of the introductory note, the number of pulses in the introductory note, pulse rate, call period, duty cycle, number of click note appendages and the dominant frequency peak.
Hirundo rustica erythrogaster barn swallows (Passeriformes: Hirundinidae)
The barn swallow (Hirundo rustica) is an insectivorous, semi-colonial, migratory songbird comprising six described subspecies distributed across the Holarctic. The subspecies H. r. erythrogaster breeds throughout North America. During the breeding season, males produce songs primarily composed of a warbling series of syllables, followed by a rising, frequency modulated syllable (the “p-syllable”), and a terminal trill, (termed the “rattle”) (Galeotti et al. 1997). Songs are utilized for both attraction of mates and competition among rivals, and are therefore considered advertisement signals. Morphological and acoustic data were collected for 62 males at breeding sites in Boulder County, Colorado, USA. Between 3 and 27 songs (11.16 mean ± 0.67 SE) were recorded between 5am and 1pm over the following dates: May 15-July 19, 2009; May 6-May 31, 2011, and May 1-August 21, 2012. Songs were recorded in 16-bit WAV format, with 48kHz sampling rate using a Marantz PMD 660 digital recorder, paired with an Audiotechnica AT815B shotgun microphone (2009), a Marantz PMD 660 paired with a Sennheiser MKH 20 and Telinga parabola (2011), or a Marantz PMD 661 paired with a Sennheiser ME62/k6 microphone and Telinga parabola (2012). The following song traits were extracted using Avisoft SAS-Lab Pro version 5.2 (www.avisoft.com, R. Specht): warble length, p-syllable length, trill length, number of trill pulses, trill rate, peak frequency of the warble, peak frequency of the rattle, bandwidth of the rattle, and Wiener entropy of the warble. Right wing length was used as a metric of body size, measured to the nearest 0.5 mm using a wing rule. Mass (for assessment of condition-dependence) was measured to the nearest 0.5 g using a Pesola spring scale. For both right wing length and mass, we took the average of three measurements to account for measurement error.
Hyla cinerea treefrogs (Anura: Hylidae)
This species ranges from southern Texas to Delaware, USA. Males produce acoustic signals to attract females (advertisement calls) and to mediate disputes over calling sites (aggressive calls).
The amplitude envelope of the advertisement call is shaped like one long pulse with a short pulsatile beginning. Data for advertisement calls are from a population at the Jasper Fish Hatchery, Jasper Co., Texas) (n=136 males). Descriptions of calls are based on the analysis of 10 calls per call type per male; averages per male were calculated, and data was temperature corrected to a common temperature of 25C. After recording, the males were captured, their SVL (snout–vent length) measured to the nearest 0.1mm with calipers, and their weight to the nearest 0.1mg with a Pesola spring scale. The assessed call traits were call duration, call period, low (dominant) frequency peak, and high frequency peak. Females show strongest preferences for call frequency. Preference functions for call frequency were described with non–parametric cubic splines using data from (Höbel and Gerhardt 2003). Females also show preferences for shorter call periods (i.e., higher repetition rates) and longer calls (Gerhardt 1987).
The amplitude envelope of the aggressive call is entirely pulsed. Data comparing advertisement and aggressive calls are from a population from Bowens Mill Fish Hatchery, Ben Hill Co., Georgia (n = 15 males). The assessed call traits were call duration, call period, call rise time, % rise time, duration pulsed section, # pulses, pulse period, amplitude modulation depth, low (dominant) frequency peak, and high frequency peak.
Hyla versicolor treefrogs (Anura: Hylidae)
This species is widely distributed in the eastern half of the USA. Males produce a trilled advertisement call to attract females. Data comparing the allometry of advertisement calls were collected from two locations: Byers Pond, Ozaukee Co., WI, and Ashland, MO, USA. Because these populations represent independent polyploid lineages (Holloway et al. 2006), we count them as two separate species in our analysis.
Different methods were used to record advertisement calls in the two populations. For the Wisconsin population, descriptions of calls are based on the analysis of 10 calls per male, averages per male were calculated, and data were temperature corrected to a common temperature of 20C. A total of 54 males were recorded; afterwards the males were captured, their SVL (snout–vent length) measured to the nearest 0.1 mm with calipers, and their weight to the nearest 0.1 g with a Pesola spring scale. The assessed call traits were call duration, % call rise time, call period, # pulses/call, pulse rate, low frequency peak, high (dominant) frequency peak, and amplitude difference between frequency peaks. For the Missouri population, calls were recorded during staged aggressive interactions between two males (Reichert and Gerhardt 2011). Ten advertisement calls per male were recorded from males interacting at a distance of 1.8 m from a competitor, averages per male were calculated and the data were temperature corrected to a common temperature of 23.4C. A total of 363 males were recorded; afterwards males were captured, their SVL measured to the nearest 0.01 mm with calipers and mass measured to the nearest 0.01 g with an electronic balance. The assessed advertisement-call traits were call duration, call period, number of pulses per call, call duty cycle, pulse rate and the low and high (dominant) frequency peak. Females prefer longer calls , higher call rate
s , and frequencies and pulse rates centered around the population mean .
Aggressive calls are shorter in duration than advertisement calls and lack pronounced amplitude modulation (Reichert 2013). Aggressive calls are generally given in distinct bouts, with long pauses between bouts. Data comparing advertisement and aggressive calls were recorded from males from the Missouri population during staged aggressive interactions. At least 5 aggressive calls per male were recorded from 210 males interacting at a distance of 10 cm or less with a competitor; only call duration of aggressive calls was temperature-corrected to 23.4C because the other characteristics do not covary with temperature (Reichert 2013). The assessed traits for aggressive calls were call duration, call period, duty cycle, calls per aggressive-calling bout, duty cycle within bouts, and the low and high (dominant) frequency peak.
Hypsiboas rosenbergi treefrogs (Anura: Hylidae)
This neotropical species is distributed in the lowland rainforests from Costa Rica to Ecuador.
Males produce a two–note advertisement call. Data comparing the allometry of advertisement calls were collected from La Gamba Field Station, Puntarenas province, Costa Rica. Descriptions of calls are based on the analysis of 10 calls per male, averages per male were calculated, and data was temperature corrected to a common temperature of 25C. A total of 48 males were recorded; afterwards the males were captured, their SVL (snout–vent length) measured to the nearest 0.1mm with calipers, and their weight to the nearest 0.1g with a Pesola spring scale. The assessed call traits were call duration, call period, note duration, inter–note interval, inter–call interval, dominant frequency and second frequency peak. Females prefer longer calls (calls with more notes/call) and shorter periods; frequency preferences have not been tested (Höbel, unpubl.). Females were collected in amplexus to assure receptivity to sexual signals. Trials were conducted in a portable playback setup in the field. Preferences for duration and period were described using two–choice playback trials that tested a standard call (populations average) against one lower and one higher alternative. In each trial, 5–12 females were tested.
Gryllus field crickets (Orthoptera, Gryllidae)
Data are included from three sister species: G. lineaticeps from the Santa Monica Mountains National Recreation Area, Los Angeles Co., California, USA; G. personatus from Winslow, Navajo Co., Arizona, USA; and a currently undescribed species ‘G15’ from the Agua Fria National Monument, Yavapai Co., Arizona, USA. To help provide continuity in the literature, we note that G15 will be named “G. staccato” in an upcoming revision of Gryllus (D. B. Weissman and D. A. Gray, unpublished) [in accordance with ICZN the name G. staccato is here disclaimed as “unavailable”]; the G15 studied here is the same G15 as has been studied previously in the field (Sakaguchi and Gray 2011). For each species, wild caught adults were used to initiate laboratory cultures. Crickets were reared at 27 C with a 14:10 L:D photoperiod with dried cat food, water in cotton-plugged vials, and moist vermiculite/peat-moss mix for oviposition. Adult males were individually isolated in 500 ml containers with food, water, and egg carton shelter and were recorded digitally at 44.1 kHz 16 bit sampling directly to laptop using CoolEdit 2000 (Syntrillium Software). Following recording, males were weighed to the nearest 1 mg and body size was measured as pronotal width using digital calipers to 0.01 mm. Pronotal width is a common measure of body size in crickets (Gray 1997; Gray et al. 2001); as it is fixed at the adult molt, pronotal width reflects structural body size not short-term condition. CoolEdit 2000 was used to analyze calls for four song traits: numbers of pulses per chirp, pulse rate, peak frequency, and chirp rate; ten calls were analyzed per male, and the average used in subsequent analyses with pulse rate, peak frequency, and chirp rate corrected to 22 C.
Female preferences for song traits have been studied in detail using a kugel trackball system (R. M. Hennig, T. Blankers, and D. A. Gray, unpublished), see (Hennig 2009; Meckenhäuser et al. 2013) for trackball system details. Results were consistent with previous work with G. lineaticeps (Wagner and Reiser 2000; Beckers and Wagner 2011) and with G15 (D. A. Gray, unpublished). Preference functions were closed for pulse rate and peak frequency in all species and were generally concordant with male song means for pulse rate, whereas preferences for peak frequency showed some mismatch with male trait values with modest preferences for higher frequencies. Female show moderately strong or weakly directional preferences for more pulses and faster rates in all species; the preferences are categorized here as ‘open’ preferences reflecting selection within the natural range of variation, however the preference functions exhibit some threshold effects and/or decreased response at extreme values.
Oecanthus nigricornis black-horned tree crickets (Orthoptera, Gryllidae).
Oecanthus nigricornis is found in secondary growth fields in the northeastern United States. Males call to attract females, producing continuous trills that can last for minutes to hours without interruption (Sismondo 1979; Toms 1993; Mhatre 2012).
The insects used in this study were collected in Rennselaerville, NY, USA (42° 30' 57.49", 74° 9' 17.05"). Males were captured as adults and recorded with a Marantz 661 solid state recorder and ME 62 Sennheiser microphone (see Symes 2014 for details of male maintenance and call recording). Females were collected as nymphs and reared to adulthood.
Call characteristics were measured using Raven Pro Version 1.4 (Cornell Lab of Ornithology). Recordings are archived in the Macaulay Library of Natural Sound (Cornell University). The relationship between call characteristics and temperature is approximately linear over the biologically relevant range (Walker 1962, Walker 1963). Consequently, all calls were standardized to 25 C using population–specific regressions. The number of pulses produced per second was calculated by using the Raven interactive detector feature to detect the number of pulses in a two second interval and dividing this value in half. Pulse duration was measured on the waveform and had a mean duration of 12 ms at 25 C. Before making frequency measurements, the spectrogram parameters were manipulated to maximize the frequency resolution of the spectrogram (Hann spectrogram with 1024 samples, 50 percent overlap, and DFT size of 4096 samples, resulting in frequency resolution of 23.4 Hz). The mean dominant frequency of the male call was 4.05 Hz at 25 C.
Female response was assessed by playing synthetic stimuli to females. For details on male pulse rates and the synthesis and presentation of pulse rate stimuli see Symes 2014. Male frequency and pulse duration and female response to these traits are reported for the first time here. Frequency and pulse duration responses were assessed by playing a single stimulus and recording time until the female reached the speaker, up to a maximum of 120 seconds. The pulse duration series included seven pulse durations: 2, 4, 7, 10, 12, and 16 ms, as well as a continuous tone. The frequency series included seven frequencies: 3.4, 3.6, 3.8, 4.0, 4.2, 4.4, and 4.6 kHz. Each female experienced all stimuli in series in random order. Each testing session contained up to three stimuli presentations with two testing sessions per day until all stimuli had been presented. See Symes in review for detailed description of the methods for collecting analogous data on a closely related Oecanthus species.
Female response to pulse rate was closed, with low response to pulse rates below the male mean, but continued response to pulse rates above the male mean (Symes 2014). Female response was elicited by a range of pulse durations. However, female response dropped dramatically when pulses became so long that they intergraded and produced a continuous tone. Female response to frequency was flat—females showed comparable response to all frequencies, including frequencies characteristic of heterospecifics.
All insects were weighed as adults using a Sartorius M–Prove scale. After death, a front leg and top right wing were removed from each insect, pasted on a transparency sheet, scanned, and measured in Image J.
Phaethornis longirostris Long-billed hermit (hummingbird)
This lekking hummingbird species is found in lowland rainforests from Southern Mexico to Northern Colombia. Males have repertoires of single song-types produced only on leks (Araya–Salas and Wright 2013). Song recordings and morphological measurements were taken during three breeding seasons (2010–2013) at La Selva Biological Station, Costa Rica (10°25' N; 84°00' W). Birds from four leks were mistnetted and marked with numbered bands plus a flag with unique color combinations on the back and chest or attached to leg bands. Eight morphological parameters were measured: mass, exposed culmen, bill base width, bill base height, mean tarsus length, unflattened wing length, central rectriz, external rectriz. We use a high precision digital caliper (±0.005mm) for bill measurements, a stopped wing ruler (±0.1mm) for wing a tail measurements and a digital (±0.001g) scale for body mass. Averages were used if the individuals were measured more than once. We calculated the ICC (intraclass correlation coefficient) to evaluate the repeatability of morphological measurements. Exposed culmen produced the highest repeatability (ICC=0.81) and smallest confidence interval (0.59-0.91) and was then used as an estimate of body size.
Recordings of singing lekking males were obtained from 5:00 to 11:00 in the morning and from 14:00 to 17:00 in the afternoon on a Marantz PMD 660 and a Sennheiser ME 67 microphone or a Sennheiser ME62 microphone on a parabolic antenna (53 cm diameter). Morphological and song parameters were taken in the same year for each individual. Spectrographic analyses were done using Seewave (Sueur et al. 2008) in the R environment (R Core Team 2012). We measured 11 acoustic parameters: duration, mean frequency (average frequency of the spectrum), skewness (asymmetry of the spectrum), kurtosis (peakness of the spectrum), spectral entropy (energy distribution; pure tone ~ 0; noisy ~ 1), dominant frequency (average of 10 measurements of dominant frequency equally spaced through the song), fundamental frequency (average of 10 measurements of fundamental frequency equally spaced through the song), peak frequency (frequency with highest amplitude), frequency range (difference between highest and lowest dominant frequency), modulation index (accumulated absolute difference between adjacent measurements of dominant frequency), and peak frequency time (ratio of the time at which the highest amplitude is reached to the duration of the song). Three songs were measured per recording and parameters were averaged.
Araya–Salas M, Wright T. 2013 Open-ended song learning in a hummingbird. Biol Lett 9: 20130625.
Beckers OM, Wagner WE. 2011. Mate sampling strategy in a field cricket: evidence for a fixed threshold strategy with last chance option. Anim Behav 81:519–527.
Bush SL, Gerhardt HC, Schul J. 2002. Pattern recognition and call preferences in treefrogs (Anura: Hylidae): a quantitative analysis using a no–choice paradigm. Anim Behav 63:7–14.
Galeotti P, Saino N, Sacchi R, Møller AP. 1997. Song correlates with social context, testosterone and body condition in male barn swallows. Anim Behav 53:687–700.
Gerhardt HC. 1987. Evolutionary and neurobiological implications of selective phonotaxis in the green treefrog (Hyla cinerea). Anim Behav 35:1479–1489.
Gerhardt HC. 1991. Female mate choice in treefrogs: static and dynamic acoustic criteria. Anim Behav 42:615–635.
Gerhardt HC. 2005. Advertisement–Call Preferences in Diploid–Tetraploid Treefrogs (Hyla chrysoscelis and Hyla versicolor): Implications for Mate Choice and The Evolution of Communication Systems. Evolution 59:395–408.
Gerhardt HC, Dyson ML, Tanner SD. 1996. Dynamic properties of the advertisement calls of gray tree frogs: patterns of variability and female choice. Behav Ecol 7:7–18.
Gerhardt HC, Tanner SD, Corrigan CM, Walton HC. 2000. Female preference functions based on call duration in the gray tree frog (Hyla versicolor). Behav Ecol 11:663–669.
Gray DA. 1997. Female house crickets, Acheta domesticus, prefer the chirps of large males. Anim Behav 54:1553–1562.
Gray DA, Walker TJ, Conley BE, Cade WH. 2001. A morphological means of distinguishing females of the cryptic field cricket species, Gryllus rubens and G. texensis (Orthoptera: Gryllidae). Fla Entomol 84:314–315.
Hennig RM. 2009. Walking in Fourier's space: algorithms for the computation of periodicities in song patterns by the cricket Gryllus bimaculatus. J Comp Physiol A 195:971–987.
Höbel G, Gerhardt HC. 2003. Reproductive character displacement in the acoustic communication system of green tree frogs (Hyla cinerea). Evolution 57:894–904.
Holloway AK, Cannatella DC, Gerhardt HC, Hillis DM. 2006. Polyploids with different origins and ancestors form a single sexual polyploid species. Am Nat 167:E88–E101.
Klump GM, Gerhardt HC. 1987. Use of non–arbitrary acoustic criteria in mate choice by female gray tree frogs. Nature 326:286–288.
Meckenhäuser G, Hennig RM, Nawrot MP. 2013. Critical song features for auditory pattern recognition in crickets. PLoS ONE 8:e55349.
Mhatre N, Montealegre–Z F, Balakrishnan R, Robert D. 2012. Changing resonator geometry to boost sound power decouples size and song frequency in a small insect. Proc Natl Acad Sci USA 109:1444–1452.
R Core Team. 2012. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. See http://www.R-project.org.
Reichert MS, Gerhardt HC. 2011. The role of body size on the outcome, escalation and duration of contests in the grey treefrog, Hyla versicolor. Anim Behav 82:1357–1366.
Reichert MS. 2011. Aggressive calls improve leading callers’ attractiveness in the treefrog Dendropsophus ebraccatus. Behav Ecol 22:951–959.
Reichert MS. 2013. Sources of variability in advertisement and aggressive calling in competitive interactions in the grey treefrog, Hyla versicolor. Bioacoustics 22:195–214.
Sakaguchi KM, Gray DA. 2011. Host song selection by an acoustically-orienting parasitoid fly exploiting a multi-species assemblage of cricket hosts. Anim Behav 81:851–858.
Sismondo, E. 1979. Stridulation and tegminal resonance in the tree cricket Oecanthus nigricornis (Orthoptera: Gryllidae: Oecanthinae). J Comp Physiol A 129:269–279.
Sueur J, Aubin T, Simonis C. 2008. Equipment review: seewave, a free modular tool for sound analysis and synthesis. Bioacoustics 18:213–226.
Symes LB. 2014. Species composition affects the shape of mate response functions. Evolution (in press).
Symes LB. Male signals and female mate recognition in the tree cricket Oecanthus forbesi. In review.
Toms RB. 1993. Incidental effects and evolution of sound-producing organs in tree crickets (Orthoptera: Oecanthidae). Int J Insect Morphol Embryol 22:207–216.
Wagner Jr WE, Reiser MG. 2000. The importance of calling song and courtship song in female mate choice in the variable field cricket. Anim Behav 59:1219–1226.
Walker TJ. 1962. The taxonomy and calling songs of United States tree crickets (Orthoptera: Gryllidae: Oecaiithinae). I. The Genus Neoxabea and the niveus and varicornis groups of the genus Oecanthus. Ann Entomol Soc Am 55:303–322.
Walker TJ. 1963. The taxonomy and calling songs of United States tree crickets (Orthoptera: Gryllidae: Oecanthinae). II. The nigricornis group of the genus Oecanthus. Ann Entomol Soc Am 56:772–789.
Wells KD, Bard KM. 1987. Vocal communication in a Neotropical treefrog, Hyla ebraccata: Responses of females to advertisement and aggressive calls. Behaviour 101:199–210.
Wells KD, Schwartz JJ. 1984a. Vocal communication in a Neotropical treefrog, Hyla ebraccata: advertisement calls. Anim Behav 32:405–420.
Wells KD, Schwartz JJ. 1984b. Vocal communication in a Neotropical treefrog, Hyla ebraccata: aggressive calls. Behaviour 91:128–145.
Wollerman L. 1998. Stabilizing and directional preferences of female Hyla ebraccata for calls differing in static properties. Anim Behav 55:1619–1630.