Content uploaded by Sergio C. Gonzalez
Author content
All content in this area was uploaded by Sergio C. Gonzalez on Feb 17, 2022
Content may be subject to copyright.
Amphibia-Reptilia (2022) DOI:10.1163/15685381-bja10084 brill.com/amre
The effects of environmental cues on chorusing onset in a tropical
frog assemblage
Sergio C. Gonzalez, Venetia S. Briggs-Gonzalez∗
Department of Wildlife Ecology and Conservation, Fort Lauderdale Research and Education Center,
University of Florida, Fort Lauderdale, FL 33314, USA
*Corresponding author; e-mail: vsbriggs@ufl.edu
ORCID iD: Briggs-Gonzalez: 0000-0001-7748-7986
Received 21 September 2021; final revision received 28 January 2022; accepted 30 January 2022;
published online 14 February 2022
Associate Editor: Julian Glos
Abstract. There are extensive studies on frog calling behaviours, including the effects of environmental variables, however,
there are no known studies to explore the specific proximate cues that stimulate the onset of calling in an individual on a
given night. The aim of this study is to identify and quantify the species-specific set of environmental variables that stimulate
males to produce mating calls under natural conditions. Call surveys were conducted at an active breeding pond on the edge
of Parque Nacional Soberanía, Panama, during the breeding seasons of 2009 and 2010. Observations were made on 20 anuran
species at the study site and we examined the onset of calling in nine species that were active and most consistently present
during breeding seasons. We used logistic and linear regression models to investigate environmental conditions that affect
calling for each species. The initiation of chorusing differed by species and key factors included ambient light, rainfall, and
lunar cycle. Our data define the margins of a behavioural-environmental envelope that is species-specific and is not related to
calling behaviour itself but is rather defined by physiological constraints related to environmental exposure.
Keywords: abiotic cues, anuran, frog calls, mating behaviour, sexual selection.
Introduction
General relationships between environmental
factors and activity patterns of most amphib-
ian and reptilian species are well documented.
For example, the effects of rainfall and tem-
perature on the intensity of breeding activ-
ity of frogs have been documented in sev-
eral studies (Brooke, Alford and Schwarzkopf,
2000; Oseen and Wassersug, 2002; Gottsberger
and Gruber, 2004; Steelman and Dorcas, 2010;
Schalk and Saenz, 2016). Daily rainfall pat-
terns have also been correlated with varying
levels of breeding in the salamander Plethodon
hubrichti (Reichenbach and Sattler, 2007). Sim-
ilarly, canopy cover has been shown to affect
the spatial distribution of breeding amphib-
ians (Werner and Glennemeier, 1999). Steel-
man and Dorcas (2010) observed the effects
of environmental factors on chorus intensity
with the goal of determining when best to con-
duct call surveys for frog monitoring programs.
Despite these observations, relatively little is
known about which cues might trigger specific
behaviours at the organismal level and thus be
relevant to individual reproductive success.
Activity patterns of other nocturnal ectother-
mic taxa have been linked to temperature,
light, and moonphase (Lillywhite, 1971; Jaeger
and Hailman, 1981; Brischoux and Lillywhite,
2011; Weaver, 2011; Sperry, Ward and Weath-
erhead, 2013). In mammals, a similar pattern is
manifested as circadian rhythms (Heldmaier et
al., 1982; Gachon et al., 2004). Calling in spot-
ted owls increased in intensity during the last
quarter through new moon phases [which is at
reduced ambient light levels, and lessen chances
of being visually perceived by prey (Ganey,
1990)] and calling also increased on calm nights
with no precipitation. Courtship behavior in
©Koninklijke Brill NV, Leiden, 2022. DOI:10.1163/15685381-bja10084
2S.C. Gonzalez, V.S. Briggs-Gonzalez
guppies differs under variable ambient light
conditions (Endler, 1991). These responses can
often be linked to environmental influence on an
organism’s metabolic rate, exposure to preda-
tion risk, or risk of detection by prey. For exam-
ple, wire-tailed manakins displayed mainly in
the shade because this affected their visibil-
ity and conspicuousness (Heindl and Winkler,
2003). Because anurans are generally nocturnal,
small in size, and possess semipermeable skin,
their behaviour may be even more strongly dic-
tated by immediate environmental conditions
(Oseen and Wassersug, 2002).
Anuran vocalization has been the focus of
extensive studies in the contexts of sexual selec-
tion and male-male competition (Ryan, 1988;
Briggs, 2008, 2010), predation risk tradeoffs
(da Silva Nunes, 1988; Bernal, Rand and Ryan,
2007), and even speciation (Rand, 1985; Ryan,
Cocroft and Wilczynski, 1990; Richardson et
al., 2008; Bonachea and Ryan, 2011). Envi-
ronmental conditions that maximize breeding
activity (and therefore detection or capture by
field herpetologists) on a nightly basis are rel-
atively understood for some species (Oseen
and Wassersug, 2002; Steelman and Dorcas,
2010; Kusano et al., 2015). Other works have
examined the effects of seasonal variation in
environmental factors that influence breeding
activity (Gottsberger and Gruber, 2004; Schalk
and Saenz, 2016). Much of the work relating
environmental conditions with anuran breed-
ing calls, however, have measured activity as
whether or not a species is calling during a
given time period, often by automated means
(as in Oseen and Wassersug, 2002; Steelman
and Dorcas, 2010). Environmental variables
such as air and water temperature, rainfall, rel-
ative humidity, photoperiod, and moon phase
(Hatano, Rocha, and Van Sluys, 2002; Oseen
and Wassersug, 2002; Gottsberger and Gruber,
2004; Steelman and Dorcas, 2010; Kusano et
al., 2015) have been identified as important to
promoting chorus intensity (number of calling
individuals). The identification of these impor-
tant variables and relating remote weather sta-
tion data to a categorical assignment of call-
ing intensity characterizes the weather which is
favorable for breeding. Much attention has also
been given to the physiology of anuran vocaliza-
tion (Ryan, 1988; Wells, 2001), weather affect-
ing chorus intensity (Henzi et al., 1995; Brooke,
Alford and Schwarzkopf, 2000), and the role
of call traits on sexual selection and niche dif-
ferentiation (Briggs, 2010; Bignotte-Giró and
López-Iborra, 2019), however no studies have
attempted to identify the environmental cues
that stimulate calling in an individual on a given
night.
This approach assumes a highly mechanis-
tic view of behaviour as it implies the existence
of hardwired acute responses to specific abiotic
stimuli, which we know exists of feeding behav-
ior (Wassersug, Naitoh and Yamashita, 1999)
and predator avoidance (Yamashita, Naitoh and
Wassersug, 2000). Excluding the social aspect
of male-male interference calls or breeding calls
produced in response to hearing other males,
there likely exists specific external stimuli that
prompt the first male to initiate the chorus.
For example, Henderson and Nickerson (1976),
demonstrated that manipulation of light lev-
els to specific thresholds could stimulate spe-
cific activity level responses in three species of
blunt-headed tree snakes (Imantodes sp.). Pho-
totactic behavior has also been examined in
the laboratory in several frog species, includ-
ing some tropical species that we present calling
onset for (see Jaegar and Hailman, 1971, 1973,
1976), and results predicted that species’ activ-
ity would be limited to a narrow range of ambi-
ent light, presumably within which their eyes
are best adapted (Jaeger and Hailman, 1981).
Jaeger and Hailman (1981) also determined
species specific thresholds of ambient light
which prompted feeding activity. Beyond these
studies, proximate environmental triggers for
specific behaviors have not been documented
Cueing in the chorus 3
in other vertebrates. Demonstrating such a rela-
tionship may alter how we perceive reproduc-
tive behaviour and specifically the decision to
initiate breeding activity in anurans.
The aim of this study is to identify the
species-specific set of environmental variables
that stimulates the initiation of mating calls
under natural conditions. This study operates
on a fine scale where relatively rapid environ-
mental changes occur during the crepuscular
hour when focal species begin breeding activ-
ity. The first call of the night is the target of our
study because calling after that may be a social
response to the first male’s call (Richardson et
al., 2008; Höbel, 2017). We investigate whether
frog species respond to different thresholds of
ambient light and explore the effects of other
environmental variables on the initiation of call-
ing. We contextualize our findings with species’
life histories and discuss how differences may
reflect breeding niche differentiation that allow
dozens of anuran species to partition spatio-
temporal space in a tropical forest.
Materials and methods
Study site
Call surveys (N=71) were conducted at Experimental
Pond at the Smithsonian Tropical Research Institute facil-
ities in Gamboa, Panama between May and August 2009
(N=38), and between June and July 2010 (N=33). The
pond is at an elevation of 50 m and is on the edge of Parque
Nacional Soberanía (N 9°07.286, W 79°42.182). Experi-
mental Pond is a man-made, concrete pond surrounded by
potted and naturally growing vegetation constructed specifi-
cally to facilitate the study of anuran ecology and behaviour.
The pond measures approximately 5 m ×10 m with a
maximum depth of 1.5 m. Egg clutches of Dendropsophus
ebraccatus (Cope, 1874) and Agalychnis callidryas (Cope,
1862) are frequently collected from vegetation around the
pond for experiments during the rainy season. Numerous
species have colonized and use the pond and the immedi-
ate terrestrial area for breeding. Forest canopy cover around
the pond is between 62% on the south edge and 85% on the
north edge, as measured by a spherical convex densitome-
ter (Lemmon, 1956). Leaf litter and debris fall into the pond
naturally and emergent vegetation grows from the shallow
edges.
Survey methods
Pilot surveys showed that many of the nocturnal frog species
in the area began calling as the sun was setting. Sur-
veys were conducted before dusk, from about 1800 h until
1900 h, just after sunset. When the opportunity arose, sur-
veys started earlier. Surveys were not conducted during
heavy rains which impaired the surveyor’s hearing, as based
on North American Amphibian Monitoring Program pro-
tocol (Weir and Mossman, 2005). Surveys were conducted
from the same spot on the eastern edge of the pond using
a red headlamp to lessen effects of surveyor’s presence on
anuran behaviour. Start time, Julian calendar day, day of
moon phase (0-28 or 0-30), ambient temperature, percent
(%) relative humidity, ambient light levels (foot candles, ft
c. measured using a Fisher Scientific Enviro-meter), and sky
conditions were registered at the beginning of each survey.
Species already calling upon arrival were noted (but
not included in the analysis). During surveys, the time,
ambient temperature, % relative humidity, and ambient light
levels at first call were registered for each species. When
applicable, the time at which a species stopped calling and
the corresponding environmental data were also registered.
Data points were documented if a time gap greater than 10
minutes during which no new species joined the chorus.
This helped maintain a record of how ambient conditions
changed as the sun set during each survey and further
served to provide null data points for logistic models. Daily
rainfall data were obtained from weather station data kept in
Gamboa, Panama.
Data analysis
We explored the variability of light level at first call for
each species and tested whether data adhered to assump-
tions of normality. We tested the hypothesis that differ-
ent species respond to different light thresholds using an
analysis of variance (ANOVA) and Tukey’s HSD post-hoc
tests to locate the origin of differences. For species with
small variance in light levels at first call, logistic regressions
that included light as an independent variable were used to
model the environmental cues that stimulate calling (i.e., we
modeled binomial calling with our environmental data). We
conducted linear regressions to model light level (a proxy
for time) at onset of calling for species that displayed a
high degree of variability in light levels at first call and that
engaged in sporadic calling during daytime showers outside
of our survey period. Independent variables included in the
global models were % relative humidity, ambient light lev-
els (ft.c.), days until full moon (Full =0), air temperature,
daily total precipitation (mm), total precipitation over the
preceding 48 h period (mm), number of consecutive days
of rainfall prior to calling activity, as well as number of
consecutive days without rainfall prior to calling activity.
We used a Pearson’s correlation coefficient to test for cor-
relations between rain and humidity variables (r<±0.2).
We used backwards stepwise regression to build optimized
models for each species. For Craugastor fitzingeri we were
able to model both the onset and cessation of calling.
4S.C. Gonzalez, V.S. Briggs-Gonzalez
Results
Species-specific observations
Frog species observed or heard near our sur-
vey site included Agalychnis callidryas (Cope,
1862), Rhinella horribilis (Wiegmann, 1833),
Rhinella alata (Thominot, 1884), Chiasmo-
cleis panamensis (Dunn et al., 1948), Den-
dropsophus ebraccatus,Dendropsophus micro-
cephalus (Cope, 1886), Diasporus diastema
(Cope, 1875), Craugastor fitzingeri (Schmidt,
1857), Pristimantis taeniatus,Boana rosen-
bergi (Boulanger, 1898), Boana rufitelus (Fou-
quette, 1961), Leptodactylus insularum (Bar-
bour, 1906), Leptodactylus fragilis (Brocchi,
1877), Engystomops pustulosus (Cope, 1864),
Scinax boulengeri (Cope, 1887), Scinax ruber
(Lorenti, 1768), and Smilisca sila (Duellman
and Trueb, 1966). Other species observed in
Gamboa, Panama but not included in this
study were Dendropsophus phlebodes (Stej-
neger, 1906), Leptodactylus pentadactylus
(Lorenti, 1768), and Trachycephalus venulosus
(Laurenti, 1768). Scinax ruber and B. rufitelus
were not heard at the survey site during the 2009
field season but were both heard and observed
in 2010 with a single male present for approx-
imately 5 days, after which, it was no longer
heard. This species is known from other parts
of the Parque Nacional Soberanía but had not
been documented at this site (Ibáñez, Rand and
Jaramillo, 1999).
Males of C. fitzingeri,Diasporus diastema,
and Dendropsophus ebraccatus were usually
the first to begin calling and would sometimes
begin calling prior to the start of the survey
(constraining what can be deduced from our
models for these species). The effects of rain-
fall, and humidity were most noticeable in these
three species when the onset of calling was
delayed on days without rainfall. Pristiman-
tis taeniatus would only begin calling when
light levels were below 1.0 ft.c., and usually
at 0.0 ft.c. Boana rosenbergi would also con-
sistently begin calling when light levels were
below 4.0 ft.c., with few exceptions. Rhinella
horribilis was never heard calling during the
survey period, nor was it heard before 2000 h
in the Gamboa area. The onset of breeding sea-
sons of R. alata and S. boulengeri were docu-
mented during both years, as well as a pause in
S. boulengeri’s breeding activity in 2010.
Agalychnis callidryas responded negatively
(reduced activity) to rain during the survey
period and to night-time rainfall, but posi-
tively to diurnal rain. This contrasts with what
has been observed in some other parts of its
range (Briggs, 2008). In general, chorus inten-
sity was noticeably diminished on days without
any precipitation for all species. We observed
20 species during the study period of which
approximately half were consistently present at
the study site, and we collected data on nine
species to generate optimized models to evalu-
ate the cues associated with onset of calling. We
list these nine focal species with environmen-
tal data for ambient light, relative humidity, and
temperature along with pertinent natural history
details in table 1.
Species responses to light levels
Results of the ANOVA indicated that focal
species responded differently to light levels
(F=39.399,569,P<0.001). Based on results
of Tukey’s HSD, we sorted species into groups
that responded similarly to light levels (i.e., not
significantly different from each other within a
group; fig. 1). Group A is less sensitive to light
as demonstrated by the high ambient light levels
at which they would begin calling and with high
variance in the data (fig. 1). Species in Group
B began calling at lower ambient light levels
with less variance (fig. 1). The species in Group
C were most sensitive to light levels (as shown
by much lower variance) requiring darkness to
begin calling (fig. 1).
Environmental cues that induce calling
Light levels had significant negative effects
on the initiation of calling in all species,
except Dendropsophus ebraccatus. Days until
Cueing in the chorus 5
Tab le 1. Means of environmental variables measured at first call for focal anuran species, alongside average size (as snout-vent length – SVL according to Ibáñez et al., 1999), natural
history details, and light-sensitivity grouping. Group A =low light sensitivity, group B =medium light sensitivity, Group C =very light sensitive. Light =foot candles; RH % =percent
relative humidity, Temp =temperature (F).
Light sensitivity group Species Light RH % Temp Mean SVL (cm)
male/female
Forest level Clutch deposition site
Mean SD Mean SD Mean SD
ADendropsophus ebraccatus 29.54 23.98 0.82 0.05 82.03 1.91 2.8/3.7 Midstory, call from emergent
vegetation
Emergent vegetation, in or out
of water
ADiasporus diastema 29.19 22.19 0.83 0.03 81.83 1.84 2.1/2.5 Midstory to canopy Protected site in tree, undergo
direct development
ACraugastor fitzingeri (start) 40.07 30.93 0.82 0.03 82.26 1.98 3.5/5.3 Understory, low bushes On leaf litter, undergo direct
development
CC. fitzingeri (stop) 0.17 0.43 0.85 0.03 80.57 1.77
BRhinella alata 7.67 12.93 0.84 0.02 81.18 1.54 4.3/5.5 Terestrial In water
BDendropsophus microcephalus 7.52 10.11 0.82 0.04 81.35 1.84 2.5/3.1 Midstory, call from emergent
vegetation
In water
BScinax boulengeri 14.4 16.47 0.83 0.03 79.97 10.57 4.9/5.3 Midstory, low bushes In water
CAgalychnis callidryas 3.26 5.46 0.84 0.03 80.91 1.75 5.6 /7.1 Arboreal, Midstory-canopy Vegetation overhanging pond
CBoana rosenbergi 1.87 0.84 0.84 0.03 80.82 1.88 9.2/9.5 Arboreal, males call on ground Mud basin near water
CPristimantis taeniatus 0.37 0.55 0.83 0.02 80.82 1.76 2.5/3.2 Terestrial, low bushes Unknown
6S.C. Gonzalez, V.S. Briggs-Gonzalez
Figure 1. Ambient light level at first call* by tropical frog species, Gamboa, Panama. Groupings A, B, C are based on
results from Tukey’s HSD test with similarly behaving species grouped together. Box plots represent quartile ranges for each
species. *CF1 refers to onset of Craugaster fitzingeri, CF0 refers to cessation of calling by C. fitzingeri.DE–Dendropsophus
ebraccatus,DD–Diasporus diastema,SB–S. boulengeri,DM–Dendropsophus microcephalus,RA–R. alata,BR–B.
rosenbergi,PT–P. taeniatus,AC–Agalychnis callidryas.
full moon was a significant predictor of call
initiation in four of the ten logistic models
included (table 2). Logistic models for D. ebrac-
catus and Diadophus diastema do not include
sporadic daytime calling. Similarly, the logis-
tic model for C. fitzingeri may not be accurate
because the onset of calling could not always be
captured (table 3). These three species fall into
Group A, for which light level may not be a sig-
nificant stimulus to calling behavior and other
cues may be more important in stimulating the
first call.
Onset of calling for A. callidryas was nega-
tively affected by ambient light, and days from
full moon tended to have a negative effect. The
model for B. rosenbergi included a negative
effect of ambient light and a strong positive
effect of relative humidity. For Pristimantis tae-
niatus, the onset of calling was only predicted
by light levels. The first call of Rhinella alata
was predicted by number of consecutive days
with rain and ambient light tended to have a
negative effect on start of calling. The model
for D. microcephalus included only the effects
of ambient light on first call. The model for S.
boulengeri included ambient light levels, num-
ber of days to full moon, total precipitation
(mm) in the preceding 48 h and tended to be
affected by number of consecutive days with
rain (table 2).
Because the breeding activity of C. fitzin-
geri is strictly crepuscular, we modeled both
the onset and cessation of calling with logis-
tic models (table 3). Cessation of calling was
only dependent on ambient light levels, whereas
onset was more complex. The model predict-
ing onset of calling was based on data col-
lected on days in which the species began call-
ing later (after the start of this survey). Variables
included in this model were ambient light, con-
secutive days without rain, and number of days
to full moon (not significant). The model shows
that a period without rain delays the initiation of
calling in this species.
A linear model using light levels at first call
as the response variable included days to full
moon and a negative effect of consecutive days
of rain. The seemingly conflicting effects of
days of rain and no rain between the linear and
logistic models on calling activity can be recon-
ciled as such: periods of no rain were relatively
few during the sample period and usually lasted
for one or two days, with only one three-day and
a four-day dry spell across both sampling sea-
sons. The duration of consecutive rainfall was
much more variable, spanning up to 25 days.
The logistic model captures the high sensitivity
of C. fitzingeri to rare dry spells which reduce
the instantaneous probability of the onset of
chorusing at any given time. The linear model
Cueing in the chorus 7
explaining the timing (as light levels) of the
call shows that a delay in onset of calling can
occur during prolonged wet periods, while indi-
cating that calling behavior is hindered by the
full moon.
Both Dendropsophus ebraccatus and Diado-
phus diastema were heard calling during day-
light hours during light showers. Therefore,
because ambient light did not appear to be a
significant driver of their behavior, as also evi-
denced by the greater variance in light level at
first call for these species (fig. 1), the better
question to ask was which suite of variables pre-
dicts when they would begin calling. The logis-
tic model for D. ebraccatus included days to
full moon, total daily precipitation, and consec-
utive days with rain (table 2). The linear model
for D. ebraccatus included temperature, days of
no precipitation, and relative humidity (table 3).
The positive coefficients of the variables in the
logistic model indicate that rainfall correlates
with the likelihood of calling. The linear model
is in agreement, predicting a delayed response
in the absence of precipitation and earlier call-
ing brought on by higher humidity.
The logistic model of D. diastema included
ambient light, days to full moon, and relative
humidity (table 2). Ambient light was a signifi-
cant predictor of calling onset in this species and
relative humidity was not a significant predictor.
This lends support to our rationale for using lin-
ear models to understand the behavior of these
two species. The linear model for D. diastema
appears more consistent with our observations
and included days without precipitation and rel-
ative humidity (table 3).
Discussion
Our results indicate that the initiation of call-
ing behaviour in male frogs can be trig-
gered by environmental stimuli. Specific light-
level thresholds that stimulate calling behaviour
could be identified in several species, similar
to how specific light thresholds have previously
been demonstrated to induce foraging activity
in tropical tree snakes (Henderson and Nicker-
son, 1976) and anurans (Jaeger and Hailman,
1981). Jaeger and Hailman (1981) hypothesized
that photic cues could allow for sympatric anu-
ran species to partition the foraging environ-
ment both temporally and by microhabitat; and
may be the main factor explaining the sea-
sonal variation of anuran activity (Both et al.,
2008; Canavero and Arim, 2009). While many
of the species we examined appear to respond to
photic cues, sensitivity to those cues appears to
be species specific and modified by other envi-
ronmental conditions, behavioral context, and
life history.
Focal species could be grouped by sensitiv-
ity to light. Group A had the least sensitivity
to light levels and (fig. 1) includes the smallest
species, Group B had more sensitivity to light
levels and included intermediate sized species,
and Group C was the most sensitive to light
levels and included the largest species. The size-
sensitivity correlation is independent of calling
or egg deposition strategy (table 1). Simultane-
ously, the fact that species that fell into Group
A are sometimes heard calling in broad daylight
during rain showers should also be considered.
We suggest that, while predation risk is a trade-
off associated with many types of behaviours
(especially those requiring relocation or reposi-
tioning), male advertisement calls make individ-
uals especially conspicuous (Ryan, Tuttle and
Rand, 1982; Bernal, Rand and Ryan, 2007).
Large species are more vulnerable to detection
by visual predators and may only initiate breed-
ing activity at very low light levels, when the
chance of successfully breeding outweighs the
risk of detection by a predator.
Conversely, as light levels become less impor-
tant in the decision to initiate calling among
smaller species, other environmental factors
appear more important. Here, we begin to see
how breeding strategy dictates which environ-
mental cues are more important stimuli. For
instance, R. alata and S. boulengeri (fig.1,
Group B) are explosive breeders and deposit
eggs directly in the water, likely requiring cer-
8S.C. Gonzalez, V.S. Briggs-Gonzalez
Tab le 2. Logistic regression results for predicting the onset of calling in tropical frog species, Gamboa, Panama.
Species Variable Estimate SE P
Agalychnis callidryas Intercept 0.271 0.035 <0.001∗
Ambient light −0.004 0.001 <0.001∗
Days from full moon −0.005 0.004 0.156
Boana rosenbergi Intercept −0.656 0.324 0.043∗
Ambient light −0.002 0.001 <0.001∗
Relative humidity 0.967 0.386 0.013∗
Pristimantis taeniatus Intercept 0.144 0.015 <0.001∗
Ambient light −0.003 0.001 <0.001∗
Rhinella margaritifera Intercept 0.135 0.032 <0.001∗
Ambient light −0.002 0.001 0.077
Consec. days with rain 0.031 0.009 <0.001∗
Dendropsophus microcephalus Intercept 0.229 0.023 <0.001∗
Ambient light −0.003 0.001 <0.001∗
Smilax boulengeri Intercept 0.204 0.043 <0.001∗
Ambient light −0.001 0.001 0.014∗
Days from full moon −0.009 0.004 0.021∗
48 h precipitation (mm) 0.002 0.001 0.003∗
Consec. days with rain −0.097 0.006 0.0862
Craugastor fitzingeri (start) Intercept 0.956 0.101 <0.001∗
Ambient light −0.004 0.001 <0.001∗
Days from full moon 0.015 0.009 0.098
Consec. days without rain −0.149 0.041 <0.001∗
Caugastor fitzingeri (stop) Intercept 0.123 0.015 <0.001∗
Ambient light −0.003 0.007 <0.001∗
Dendropsophus ebraccatus Intercept 0.153 0.072 0.034∗
Days from full moon 0.020 0.009 0.022∗
24 h precipitation (mm) 0.006 0.003 0.064
Consecutive days with rain 0.048 0.015 0.002∗
Diasporus diastema Intercept −0.786 0.811 0.333
Ambient light −0.003 0.001 0.002∗
Days from full moon 0.014 0.033 <0.001∗
Relative humidity 1.682 0.968 0.084
∗Indicates significance at α=0.05 level.
Tab le 3. Linear regression results predicting light level (ft.c.) at onset of calling in focal anuran species, Gamboa, Panama.
Species Variable Estimate SE P
Dendropsophus ebraccatus Intercept −508.688 238.901 0.037∗
Temperature 5.108 2.134 0.019∗
Consec. days without rain −6.585 3.587 0.071
Relative humidity 149.473 102.52 0.149
Diasporus diastema Intercept 182.414 55.444 0.001∗
Consec. days without rain −6.465 2.224 0.004∗
Relative humidity −172.101 67.329 0.011∗
Craugastor fintzingeri (start) Intercept 63.119 10.043 <0.001∗
Days from full moon −2.143 0.990 0.034∗
Consec. days with rain −3.040 1.460 0.043∗
tain thresholds of precipitation to fall before
investing energy into reproduction. Our results
indicated that, in addition to light, rainfall was
important for the onset of calling in these
species (table 2). Conversely, during night-
timerainshowers,A. callidryas, had reduced
breeding activity unlike its explosive breeding
behaviour in the first rains in Belize (Briggs,
Cueing in the chorus 9
2008) and may be related to availability of veg-
etation in Gamboa being too wet to deposit egg
masses during rain or perhaps more likely, as
social interactions with heterospecifics.
The onset of calling in the smaller species of
Group A were related to several other param-
eters including rainfall, relative humidity, and
moon phase. The higher sensitivity to moisture
in any form than to ambient light, may have
more to do with increased desiccation risks for
small bodied species with high surface area to
mass ratios and relatively exposed eggs laid out-
side of the water (table 1).
Days from full moon, was included in the
logistic models in 5 of the 9 species, and it was
a significant variable (p-value <0.05) in 3 of
them (table 2) in predicting the onset of calling.
It was also included in the linear model of light
level at onset of calling for C. fitzingeri. Unfor-
tunately, as with much of the previous literature
pertaining to moon phase and animal behav-
ior, we cannot offer much more quantitative and
conclusive evidence as to how the moon affects
the initiation of breeding calls in anurans. Anec-
dotally, as the lunar cycle approached full moon,
both A. callidryas and B. rosenbergi noticeably
delayed calling. B. rosenbergi did not call on 3
of the 4 full moons captured in our data (across
both sampling periods). This phenomenon is
not reflected in our statistical results, however.
We can only suggest that moonlight negatively
affects the decision to begin calling.
We also found that ambient light levels
induce the cessation of breeding behavior in C.
fitzingeri. While other species breed late into
the night, until females are no longer avail-
able or individuals are exhausted from calling,
this species’ calling is strictly crepuscular and
ended with the exhaustion of twilight. Crau-
gaster fitzingeri would usually cease calling
when or just after ambient light levels reached
0.0 ft.c. Only on three occasions during 2010,
one or two individuals could still be heard call-
ing at a rate of approximately 3 calls per hour at
the pond at roughly 2200 h. Interestingly, how-
ever, on at least five occasions over the course of
both years, individuals were heard calling well
after dark from vegetation directly under or near
streetlamps in Gamboa. Similarly, the noctur-
nal E. pustulosus would often be heard calling
from the darkness within gutters and sewers in
the daytime.
Given the anthropogenic causes of amphib-
ian declines, which include habitat loss, the
spread of disease, impacts of invasive species,
and warming temperatures (Carey, Cohen and
Rollins-Smith, 1999; Kiesecker, Blaustein and
Belden, 2001), our findings could be benefi-
cial in assessing the impacts of environmental
changes on amphibian reproduction. Within a
few hundred meters of the site of this study
lies the Panama Canal which has been recently
widened resulting in the elimination of large
swaths of tropical forest. Considering that the
edge-to-area ratio of remaining habitat is being
increased, a greater proportion of remaining
forest is being impacted by edge effects. The
associated changes in temperature, humidity,
and increased light pollution can negatively
impact amphibian reproduction (Demaynadier
and Hunter, 1998).
While previous papers target calling fre-
quency, chorus intensity, or time periods
of preferred calling, none describe a direct
behavioural response prompted by specific envi-
ronmental cues. Our data quantify the environ-
mental conditions that stimulate the first call
in male frogs in a diverse tropical assemblage;
thus, defining the margins of a behavioural-
environmental envelope. Interestingly, the envi-
ronmental envelope for calling is not related
to the behaviour itself (in a sender/receiver
context), but is defined by species’ specific
physiological constraints and tradeoffs related
to environmental exposure (i.e., desiccation
risk, oviposition strategy, and predation risk).
While our operating assumption of mechanistic
responses to abiotic stimuli appears to be con-
firmed, the environmental stimuli driving the
behavioral response in question are mediated
by an organism’s physiology and social inter-
actions.
10 S.C. Gonzalez, V.S. Briggs-Gonzalez
Acknowledgements. The research was made possible by
Dr. M.A. Nickerson, the Reptile and Amphibian Conser-
vation Corps, the Florida Museum of Natural History, the
University of Florida Center for Latin American Studies
Panama Canal Research Grant, the University of Florida
University Scholars Undergraduate Research Program, Drs.
K. Warkentin and J.R. Vonesh, and the Smithsonian Tropi-
cal Research Institute.
References
Bernal, X.E., Rand, S.A., Ryan, M.J. (2007): Sexual
differences in the behavioral response of túngara
frogs, Physalaemus pustulosus, to cues associated with
increased predation risk. Ethology 113: 755-763.
Bignotte-Giró, I., López-Iborra, G.M. (2019): Acoustic
niche partitioning in five Cuban frogs of the genus
Eleutherodactylus. Amphibia-Reptilia 40: 1-11.
Bonachea, L.A., Ryan, M.J. (2011): Localization error
and search costs during mate choice in tungara frogs,
Physalaemus pustulosus. Ethology 117: 56-62.
Both, C., Kaefer, I.L., Santos, T.G., Cechin, S.T.Z. (2008):
An austral anuran assemblage in the neotropics: seasonal
occurrence correlated with photoperiod. J. Nat. Hist. 42:
205-222.
Briggs, V.S. (2008): Mating patterns of red-eyed treefrogs,
Agalychnis callidryas and A. moreletii. Ethology 114:
489-498.
Briggs, V.S. (2010): Call trait variation in Morelett’s tree
frog, Agalychnis moreletii, of Belize. Herpetologica 66:
241-249.
Brischoux, F., Lillywhite, H.B. (2011): Light- and flotsam-
dependent float-and-wait foraging by pelagic sea snakes
(Pelamis platurus). Mar. Biol. 158: 2343-2347.
Brooke, P.N., Alford, R.A., Schwarzkopf, L. (2000): Envi-
ronmental and social factors influence chorusing behav-
ior in a tropical frog: examining various temporal and
spatial scales. Behav. Ecol. Sociobiol 49: 79-87.
Canavero, A., Arim, M. (2009): Clues supporting photope-
riod as the main determinant of seasonal variation in
amphibian activity. J. Nat. Hist. 43: 2975-2984.
Carey, C., Cohen, N., Rollins-Smith, L. (1999): Amphibian
declines: an immunological perspective. Dev. Comp.
Immun. 23: 459-472.
da Silva Nunes, V. (1988): Vocalization of treefrogs
(Smilisca sila) in response to bat predation. Herpetolog-
ica 44: 8-10.
Demaynadier, P.G., Hunter, M.L. (1998): Effects of sylvi-
cultural edges on the distribution and abundance of
amphibians in Maine. Conserv. Biol. 12: 340-352.
Endler, J.A. (1991): Variation in the appearance of guppy
color patterns to guppies and their predators under dif-
ferent visual conditions. Vision Research 31: 587-608.
Gachon, F., Nagoshi, E., Brown, S.A., Ripperger, J., Schi-
bler, U. (2004): The mammalian circadian timing sys-
tem: from gene expression to physiology. Chromosoma
113: 103-112.
Ganey, J.L. (1990): Calling behavior of spotted owls in
northern Arizona. The Condor 92: 485-490.
Gottsberger, B., Gruber, E. (2004): Temporal partitioning
of reproductive activity in a Neotropical anuran commu-
nity. J. Trop. Ecol. 20: 271-280.
Hatano, F.H., Rocha, C.F.D., Van Sluys, M. (2002): Envi-
ronmental factors affecting calling activity of a tropi-
cal diurnal frog (Hylodes phyllodes: Leptodactylidae).
J. Herpetol. 36: 314-318.
Heindl, M., Winkler, H. (2003): Interacting effects of ambi-
ent light and plumage color patterns in displaying wire-
tailed manakins (Aves, Pipridae). Behav. Ecol. Socio-
biol. 53: 153-162.
Heldmaier, G., Steinlechner, S., Rafael, J., Latteier, B.
(1982): Photoperiod and ambient temperature as envi-
ronmental cues for seasonal thermogenic adaptation in
the Djungarian hamster, Phodopus sungorus.Int’l.J.
Biometeor. 26: 339-345.
Henderson, R.W., Nickerson, M.A. (1976): Observations on
the behavioral ecology of three species of imantodes
(Reptilia, Serpentes, Colubridae). J. Herpetol. 10: 205-
210.
Henzi, S.P., Dyson, M.L., Piper, S.E., Passmore, N.E.,
Bishop, P. (1995): Chorus attendance by male and
female painted Reed frogs (Hyperolius marmoratus):
environmental factors and selection pressures. Funct.
Ecol.: 485-491.
Höbel, G. (2017): Social facilitation is a better predictor
of frog reproductive activity than environmental factors.
Biotropica 49: 372-381.
Ibáñez, R.A., Rand, S., Jaramillo, C.A. (1999): Los anfibios
del Monumento Natural De Barro Colorado, Parque
Nacional Soberanía y areas adyacentes. The amphib-
ians of Barro Colorado Nature Monument, Soberania
National Park and adjacent areas. Fundación Natura,
Círculo Herpetológico de Panamá, Smithsonian Tropi-
cal Research Institute.
Jaeger, R.G., Hailman, J.P. (1971): Two types of phototactic
behaviour in anuran amphibians. Nature, Lond. 230:
189-190.
Jaeger, R.G., Hailman, J.P. (1973): Effects of intensity on
the phototactic responses of adult anuran amphibians: a
comparative survey. Z. Tierpsychol. 33: 352-407.
Jaeger, R.G., Hailman, J.P. (1976): Phototaxis in anu-
rans: relation between intensity and spectral preferences.
Copeia 1976: 92-98.
Jaeger, R.G., Hailman, J.P. (1981): Activity of Neotropical
frogs in relation to ambient light. Biotropica 13: 59-65.
Kiesecker, J.M., Blaustein, A.R., Belden, L.K. (2001):
Complex causes of amphibian population declines.
Nature 410: 681-684.
Kusano, T., Miura, T., Terui, S., Maruyama, K. (2015):
Factors affecting the breeding activity of the Japanese
common toad, Bufo japonicus formosus (Amphibia:
Bufonidae) with special reference to the lunar cycle.
Current Herpetol. 34: 101-111.
Lemmon, P.E. (1956): A spherical densiometer for estimat-
ing forest overstory density. Forest Sci. 2: 314-320.
Lillywhite, H.B. (1971): Temperature selection by the bull-
frog, Rana catesbeiana. Comp. Biochem. Physiol. – Part
A: Physiol. 40: 213-227.
Cueing in the chorus 11
Oseen, K., Wassersug, R. (2002): Environmental factors
influencing calling in sympatric anurans. Oecologia 133:
616-625.
Rand, S.A. (1985): Tradeoffs in the evolution of frog calls.
Proceedings: Animal Sci. 94: 623-637.
Reichenbach, N., Sattler, P. (2007): Effects of timbering on
Plethodon hubrichti over twelve years. J. Herpetol. 41:
622-629.
Richardson, C., Lena, J.P., Joly, P., Lengagne, T. (2008): Are
leaders good mates? A study of call timing and male
quality in a chorus situation. Animal Behav. 76: 1487-
1495.
Ryan, M.J. (1988): Energy, calling, and selection. Am. Zool.
28: 885-898.
Ryan, M.J., Tuttle, M.D., Rand, A.S. (1982): Bat predation
and sexual advertisement in a Neotropical anuran. The
Am. Nat. 119: 136-139.
Ryan, M.J., Cocroft, R.B., Wilczynski, W. (1990): The role
of environmental selection in intraspecific divergence
of mate recognition signals in the cricket frog, Acris
crepitans.Evol.44: 1869-1872.
Schalk, C.M., Saenz, D. (2016): Environmental drivers
of anuran calling phenology in a seasonal Neotropical
ecosystem. Austral Ecol. 41: 16-27.
Sperry, J.H., Ward, M.P., Weatherhead, P.J. (2013): Effects
of temperature, moon phase, and prey on nocturnal
activity in ratsnakes: an automated telemetry study.
J. Herpetol. 47: 105-111.
Steelman, C.K., Dorcas, M.E. (2010): Anuran calling sur-
vey optimization: developing and testing predictive
models of anuran calling activity. J. Herpetol. 44: 61-68.
Wassersug, R.J., Naitoh, T., Yamashita, M. (1999): Turning
bias in tadpoles. J. Herpetol. 33: 543-548.
Weaver, R.E. (2011): Effects of simulated moonlight
on activity in the desert nightsnake (Hypsiglena
chlorophaea). NW Sci. 85: 497-500.
Weir, L.A., Mossman, M.J. (2005): North American
Amphibian Monitoring Program (NAAMP). In:
Amphibian Declines: Conservation Status of United
States Amphibians, p. 307-313. Lannoo, M., Ed.,
University of California Press, Berkeley, California,
USA.
Wells, K.D. (2001): The energetics of calling in frogs.
In: Anuran Communication, p. 45-60. Ryan, M.J., Ed.,
Smithsonian Institution Press, Washington, DC.
Werner, E.E., Glennemeier, K.S. (1999): Influence of forest
canopy cover on the breeding pond distributiongs of
several amiphibian species. Copeia 1999: 1-12.
Yamashita, M., Naitoh, T., Wassersug, R.J. (2000): Startle
response and turning bias in Microhyla tadpoles. Zool.
Sci. 17: 185-189.