Simulated signal in S. virgatus lizard






Signals and behavioural responses are not coupled
in males: aggression affected by replacement
of an evolutionarily lost colour signal
Vanessa S. Quinn* and Diana K. Hews
Department of Life Sciences, Indiana State University,Terre Haute, IN
47809, USA

Male Sceloporus virgatus lack the blue abdominal patches which are used
during aggressive encounters in
other Sceloporus lizards. Herein we report that, despite having lost
this signal, males have retained a
behavioural response to experimentally restored blue abdominal
patches.We tested two adaptive hypotheses:
selection acted primarily upon signallers or selection acted upon both
signallers and receivers. The
¢rst predicts that only the signal is lost and that male interactions
should be a¡ected by the restoration of
blue patches. The latter predicts that both the signal and behavioural
response are lost and the
display of the restored blue patches should have no e¡ect on male^male
interactions. We compared the
behaviour of receivers in paired encounters where one male (signaller)
had blue-painted abdominal
patches to a set of trials where both males had white-painted abdomens,
unmanipulated abdomens or a
novel-painted pattern. The receivers of the blue-painted signal were
more likely to display submissive
behaviour. The receivers in either the unmanipulated, white-painted or
novel-painted signal trials were
more likely to display neutral behaviour. These results support the
hypothesis that receivers have retained
a behavioural response and selection has acted primarily on the
signaller.We believe this is the ¢rst documentation
of males responding to an evolutionarily lost signal in conspeci¢c
males.
Keywords: animal communication; colour; evolutionary loss; receiver;
Sceloporus virgatus; signal

1. INTRODUCTION
Communication is an important component of animal
social interactions and diverse types of information can
be transmitted, including reproductive condition, aggressive
status and species identity (Bradbury & Vehrencamp
1998). Many signals are sexually selected and studies
regarding the origin and maintenance of conspicuous
secondary sex characteristics have focused primarily on
female responses to male signals (Andersson 1994). These
studies suggest that signallers and receivers have
coevolved and that changes in the signal parallel changes
in the receiver (Butlin & Ritchie 1989). However, studies
of pre-existing bias have revealed that females of a given
species may respond to signals which have not yet evolved
in that species (Ryan 1990; Basolo 1996). The degree of
coevolution can be in£uenced by many factors, including
the sex of the signaller and receiver and selective forces
which may act di¡erently on signallers and receivers
(Morris & Ryan 1996).

While there has been much study of signals which are
present or signals which have not yet evolved, relatively
little empirical work has examined taxa with evolutionary
loss of male signals. The evolutionary loss of male
signals may be more common than gains (Wiens 1999)
and occurs in many lineages, e.g. mammals (Wright
1993), birds (Peterson 1996), ¢shes (Basolo 1996;
McLennan 1996) and reptiles (Wiens 1999). Nonadaptive
hypotheses such as genetic drift have been
generated in order to explain the loss of male traits (Hill
1994; Peterson 1996), as have adaptive hypotheses
(Endler 1980).

To examine the coevolution of signallers and receivers
in a communication system with male loss of a sexually
dimorphic trait, we recorded behaviour during male^
male encounters in the phyronosomatid lizard Sceloporus
virgatus. Most male Sceloporus have a pair of sexuallydimorphic
blue abdominal patches which are displayed
during male^male and male^female interactions. Using
well-corroborated phylogenies based on molecular and
morphological traits (Reeder & Wiens 1996; Wiens &
Reeder 1997), phylogentic analyses have indicated that at
least nine independent male losses of this trait have
occurred in Sceloporus (Wiens 1999). Some species have a
derived state in which neither sex has blue abdominal
patches such as the study species S. virgatus, while others
have an alternative derived state in which both sexes have
patches.

We asked whether selection has acted upon signallers
and receivers in a similar manner. The loss of abdominal
patches in Sceloporus lizards is associated with a grounddwelling
habit (Wiens 1999), suggesting that natural
selection may be involved in the signal loss (e.g. Endler
1980). However, some major qualitative features of the
agonistic behavioural patterns of Sceloporus lizards are
conserved throughout the genus whereas other qualitative
and quantitative aspects are highly variable
(Martins 1993;Wiens 2000). We propose two hypotheses
for examining how selection has altered this communication
system. The one-loss hypothesis states that selection
has acted more strongly on signallers than on receivers
and predicts that receivers should retain an ability to
respond to abdominal patches. The two-loss hypothesis
states that selection has been relatively strong on both
signallers and receivers and predicts that the signal and
the male response to abdominal patches will be absent.
To test these predictions we manipulated the abdominal
colour of male S. virgatus by restoring lost abdominal
patches to the levels seen in the sister species Sceloporus undulatus.
We then determined whether this replacement altered
aggressive interactions between pairs of males in staged
encounters.

2. MATERIAL AND METHODS

We studied a population of S. virgatus in the Chiricahua
Mountains of south-eastern Arizona near the Southwestern
Research Station of the American Museum of Natural History,
which is located at the northern edge of the species' range
(elevation ca. 1650m). We captured adult (greater than 50mm
snout^vent length (SVL)) (Vinegar 1975) males and females
and measured their SVL and body mass. The lizards were
captured in the afternoon, maintained in an environment which
simulated overnight temperatures and used in trials the
following morning.

We conducted two trials per day in large outdoor arenas
(6m£10m) from 18 May to 11 June 1998 and from 16 May to
6 June 1999 between 08.00 and 12.00 h. During a trial, we
recorded lizard behaviour viewed through a one-way glass
window. Before the start of each trial, we placed three still-cool
lizards on a board (20 cm£60 cm) in the arena in the following
positions: male, female, male. To avoid a side bias, the positions
of the males were alternated.We placed three opaque containers
over the lizards, raised the containers remotely and began the
trial at the ¢rst lizard movement. To decrease the variation in
competitive ability between paired males, we matched males
with similar SVL (within 2mm measured using a transparent
ruler), tail status (broken or intact), mass (within 1g measured
using a Pesola spring scale) and date of capture. Females were
included in all sets of trials in order to enhance the likelihood of
male^male interactions (Thompson & Moore 1991).
We manipulated their abdominal signals using a non-toxic
latex paint (Tulip Pearl Fabric Paint, Tulip, Clovis, CA, USA)
which is £exible when dried, allowing painted males to extend
their ventral surface fully when engaged in aggressive postural
signals (see the behaviours described below).We assigned one of
four treatments to a randomly selected member (hereafter the
signaller) of each pair of males: unmanipulated, white painted,
blue painted or novel painted. Signallers in the unmanipulated
group were not painted on the abdomen, signallers in the whitepainted
group were painted with `snow white' and signallers in
the blue-painted group were painted with `blue'. Signallers in the
novel group were painted with a series of black dots on the
abdomen. As an internal paint control, the other male (hereafter
the receiver) was painted with snow white in a pattern
which matched the abdominal paint of the signaller of that pair
(i.e. a patch or dots). In unmanipulated trials both the signaller
and receiver were unpainted. Males given patches had two
stripes (ca. 5mm£20mm) of the appropriate paint colour
applied to the abdomen. Since paint could alter a possible UV
signal or alter their ability to perform a full show, a whitepainted
treatment group was used. Males given black dots had a
series of ten to 16 dots (the number varied with the size of the
animal), each 1mm in diameter, painted on the abdomen. For
purposes of identi¢cation during the trial, we also applied a dot
of white paint on the head of one male and on the tail base of
the other male (randomly assigned). In 1998 we conducted 19
unmanipulated and 18 blue-painted trials and in 1999 we ran
nine white-painted, eight additional blue-painted and eight
novel-painted trials.

We were interested in the receivers' behavioural responses to
restored abdominal patches (the signal). For abdominal patches
to be viewed by receivers, senders must perform one of two
postural displays: the challenge display and the face-o¡ display
(Carpenter 1978). In the challenge display, signallers are
perpendicularly orientated to receivers at close range (one to
three body lengths). In the face-o¡ display, signallers and
receivers are in parallel body positions but orientated head to
tail at close range. The body postures in both challenge and
face-o¡ displays are similar: signallers expand their dewlap
(throat fan) ventrally, arch their back dorsally and compress
their trunk ventrally and laterally. This alteration in body
posture exposes abdominal patches (Carpenter 1978; V. S.
Quinn, personal obsvervation).We combined these displays into
one category (full show), since they di¡er only in the relative
positions of opponents and not the body posture of signallers. If
abdominal patches in S. virgatus are perceived as an aggressive
signal, the receiver's response immediately following the
signaller's
¢rst full show is the best measure of the response to abdominal
patches. Therefore, we analysed the ¢rst behaviour of
receivers immediately following the signaller's ¢rst full show
(hereafter post-full show behaviour). Behavioural responses were
categorized as follows: (i) aggressive, which included moves
towards, approaches and bites, (ii) submissive, which included
moves away and retreats, or (iii) neutral, which included nose
tapping or licking the substrate and head bobs. Moves towards
and approaches involve movement towards the signaller. Moves
away and retreats involve movement away from the signaller.
Nose tapping and licking the substrate are chemoreception behaviours
which occur in both social and non-social contexts, as is
the head bob display (Carpenter 1978).

If blue abdominal patches are an aggressive visual signal and
S. virgatus has retained the ability to respond to this signal, we
would expect the post-full show behaviour to di¡er between the
blue-painted trials and both the unmanipulated and whitepainted
trials. An alternative hypothesis to explain why males
may respond to abdominal patches di¡erentially is that males
display a general fear reaction to a novel signal. This hypothesis
predicts that males perceiving a novel signal would also respond
di¡erently than males perceiving a white-painted abdomen.
Thus, if blue abdominal patches are simply being responded to
as a novel signal and S. virgatus responds to all novel signals in
the same way, this novel signal hypothesis predicts that post-full
show behavioural responses will not di¡er between novelpainted
(black dot) and blue-painted trials, but that these
responses would di¡er from unmanipulated and white-painted
trials. Di¡erences in the frequencies of trials having neutral,
submissive or aggressive behavioural responses to the ¢rst full
show were compared using log-linear G-tests (Norusis 1990).

3. RESULTS
The number of trials in which a full show occurred in
1998 (32 out of 37) did not di¡er from the number in
1999 (23 out of 25) (w2 ? 0.0716, d.f. ? 1 and p4 0.5);
thus, we combined these data. The one-loss hypothesis
states that abdominal patches have been lost in S. virgatus
and that a behavioural response to this signal remains.
This hypothesis predicts that the ¢rst post-full show
behaviour should di¡er among receivers exposed to the
blue signal and those not exposed. Our data support this
hypothesis (¢gure 1). There was a signi¢cant association
between treatment and post-full show behavioural
756 V. S. Quinn and D. K. Hews Response to an evolutionarily lost
signal
Proc. R. Soc. Lond. B (2000)
categories (w2 ? 13.407, d.f. ? 6 and p ? 0.0403). Neutral
behaviours were the most frequent response category to
the ¢rst full show in the unmanipulated, white-painted
and novel-painted trials and submissive behaviours were
the most frequent response in blue-painted trials. Aggressive
responses occurred at the lowest frequency in all four
trial types.

4. DISCUSSION
The results of our manipulative paint study clearly
demonstrate that a response to male abdominal patches is
present in S. virgatus, even though this signal has been
lost.We did not measure the spectral re£ectance of either
the blue paint or the natural abdominal patches. The
colour vision abilities of Sceloporus lizards is also not
known, but the fact that the lizards responded to the blue
paint in a comparable manner to natural abdominal
patches suggests that this is not a problem. The frequency
of trials in which the receiver displayed submissive
behaviours after the ¢rst full show was greater in the
blue-painted trials and more trials with neutral responses
post-full show occurred in the control or novel-painted
trials. Our data do not support the hypothesis that male
S. virgatus perceive the blue abdominal patches as an
unknown signal and avoid this stimulus, because the
post-full show behaviour by signallers with the novelpainted
pattern did not di¡er from the response to
control males. These results support the conclusion that
blue-painted patches were retreated from because of a
speci¢c response to the patches. These patches are aggressive
visual signals in other Sceloporus lizards (Cooper &
Burns 1987) and the response to this restored signal in
S. virgatus was submissive, suggesting that receivers
perceived signallers as aggressive and retreated (e.g.
Rowher 1985).

Another interpretation of these results is that the
receivers of abdominal patches perceived the signaller as
a heterospeci¢c and avoided this animal due to the e¡ects
of interspeci¢c competition or predation. However, available
evidence suggests that heterospeci¢c avoidance is not
the most parsimonious explanation. Heterospeci¢c lizards
with blue abdominal coloration overlap with S. virgatus at
this study site, but di¡er in body size, shape and dorsal
coloration. Therefore, features other than the presence or
absence of abdominal patches are present and more
readily available for species identi¢cation. The mountain
spiny lizard Sceloporus jarrovii (both males and females
have blue abdomens) overlaps with S. virgatus and these
two species inhabit di¡erent microclimates, suggesting
that these species may not often interact (G. Middendorf,
personal communication). Another species co-occurring
at this study site is the common tree lizard Urosaurus
ornatus (males have blue abdomens and females have
white abdomens). This lizard does not compete with
S. virgatus (Smith 1981) suggesting that these similar-sized
species also may not interact aggressively.
Interestingly, display morphology tends to be more
labile than display behaviour in Sceloporus lizards.
Multiple, independent losses of abdominal patches and
the maintenance of a variety of aspects of display behaviour
have occurred in Sceloporus (Martins 1993; Wiens
2000). Morphological traits and behaviour may not be
tightly linked because other traits can convey the same
information as the lost morphological trait (e.g. Prum
1990). Redundant signals decrease the likelihood of
coupled evolution between signallers and receivers.
The conservation of some components of display
behaviour in Sceloporus is consistent with our results that
selection has acted primarily on the signaller, favouring
the loss of the male signal (Martins 1993; Wiens 2000).
Selection against male signals may have been greater in
terrestrial than arboreal habitats, possibly due to predation
(Wiens 1999). Increased natural selection against
male signals due to predation in some habitats has been
shown to favour the loss of sexually selected signals
(Endler 1980). Omland (1997) suggested a decrease in
female preference as a possible explanation for male
signal loss.

Traditionally, studies examining the roles of sexually
selected signals have focused on female responses to
signals which are present in males. However, sexually
selected signals can have dual functions, including the
deterrence of rival males and attraction of females.
Because the sexes often face di¡erent selective pressures
(Andersson 1994), selection could act di¡erently on male
versus female receivers.

A powerful method for studying the evolution of
communication systems and the signallers and receivers
of these systems is the use of phylogenetic comparisons.
Understanding the evolutionary relationships between
closely related species as well as the pattern of losses and
gains of morphological and behavioural traits within
these phylogenies may lead to more speci¢c hypotheses of
the function of male signalling traits and the selective
forces leading to the gains and losses of these traits.
We thank M. Galloway, C. Johnson, M. Westphal and R.
Worthington for their assistance in the ¢eld. In addition,
A. Abell, S. Lima, G. Middendorf, N. Rattenborg, P. Scott,
P. Zollner and two anonymous referees gave helpful reviews of
this manuscript and thoughtful discussion. This study was
supported by the School of Graduate Studies of Indiana State
University, Indiana Academy of Science, Southwestern Research
Station's Graduate Student Fund and Theodore Roosevelt
Graduate Student Fund of the American Museum of Natural
History (all to V.S.Q.) and United States National Science Foundation
IBN-9629783 (to D.K.H.).

REFERENCES
Andersson, M. 1994 Sexual selection. Princeton University Press.
Basolo, A. L. 1996 The phylogenetic distribution of female
preference. Syst. Biol. 45, 290^317.
Bradbury, J. W. & Vehrencamp, S. L. 1998 Principles of animal
communication. Sunderland, MA: Sinauer Associates.
Butlin, R. K. & Ritchie, M. G. 1989 Genetic coupling in mate
recognition systems: what is the evidence? Biol. J. Linn. Soc.
37, 237^246.
Carpenter, C. C. 1978 Comparative display behavior in the
genus Sceloporus (Iguanidae). Milw. Publ. Mus. Contrib. Biol.
Geol. 18, 1^71.
Cooper Jr,W. E. & Burns, N. 1987 Social signi¢cance of ventrolateral
coloration in the fence lizard, Sceloporus undulatus. Anim.
Behav. 35, 526^532.
Endler, J. A. 1980 Natural selection on color patterns in Poecilia
reticulata. Evolution 34, 76^91.
Hill, G. E. 1994 Geographic variation in male ornamentation
and female mate preference in the house ¢nch: a comparative
test of models of sexual selection. Behav. Ecol. 5, 64^73.
McKinnon, J. S. & McPhail, J. D. 1996 Male aggression and
colour in divergent populations of the threespine stickleback:
experiments with animations. Can. J. Zool. 74, 1727^1733.
McLennan, D. A. 1996 Integrating phylogenetic and experimental
analyses: the evolution of male and female nuptial
coloration in the stickleback ¢shes (Gasterosteidae). Syst. Biol.
45, 261^277.
Martins, E. P. 1993 A comparative study of the evolution of
Sceloporus pushup displays. Am. Nat. 142, 994^1018.
Morris, M. R. & Ryan, M. J. 1996 Sexual di¡erence in signal^
receiver coevolution. Anim. Behav. 52, 1017^1024.
Norusis, M. J. 1990 SPSS/PC + advanced statistics 4.0. Chicago,
IL: SPSS, Inc.
Omland, K. E. 1997 Examining two standard assumptions of
ancestral reconstructions: repeated loss of dichromatism in
dabbling ducks (Anatini). Evolution 51, 1636^1646.
Peterson, A. T. 1996 Geographic variation in sexual dichromatism
in birds. Bull. Br. Ornithol. Club 116, 156^172.
Prum, R. O. 1990 Phylogenetic analysis of the evolution of
display behavior in the neotropical manakins (Aves:
Pripridae). Ethology 84, 202^231.
Reeder, T.W. & Wiens, J. J. 1996 Evolution of the lizard family
Phrynosomatidae as inferred from diverse types of data.
Herpetol.Monogr. 10, 43^84.
Rowher, S. 1985 Dyed birds achieve higher social status than
controls in Harris' sparrows. Anim. Behav. 33, 1325^1331.
Ryan, M. J. 1990 Sexual selection, sensory systems and sensory
exploitation. Oxford Surv. Evol. Biol. 7, 157^195.
Ryan, M. J. & Rand, A. S. 1998 Evoked vocal response in male
tungara frogs: pre-existing biases in male responses? Anim.
Behav. 56, 1509^1516.
Ryan, M. J. & Wagner Jr, W. E. 1987 Asymmetries in mating
preferences between species: female swordtails prefer heterospeci¢
c males. Science 236, 595^597.
Searcy, W. A. & Brenowitz, E. A. 1988 Sexual di¡erences in
species recognition of avian song. Nature 332, 152^154.
Smith, D. C. 1981 Competitive interactions of the striped
plateau lizard (Sceloporus virgatus) and the tree lizard
(Urosaurus ornatus). Ecology 62, 679^687.
Thompson, C. W. & Moore, M. C. 1991 Throat colour reliably
signals status in male tree lizards, Urosaurus ornatus. Anim.
Behav. 42, 745^753.
Vinegar, M. B. 1975 Comparative aggression in Sceloporus
virgatus, S. undulatus consobrinus, and S. u. tristichus (Sauria:
Iguanidae). Anim. Behav. 23, 279^286.
Wiens, J. J. 1999 Phylogenetic evidence for multiple losses of a
sexually selected character in phrynosomatid lizards. Proc. R.
Soc. Lond. B 266, 1529^1536.
Wiens, J. J. 2000 Decoupled evolution of display morphology
and display behaviour in phrynosomatid lizards. Biol. J. Linn.
Soc. (In the press.)
Wiens, J. J. & Reeder, T.W. 1997 Phylogeny of the spiny lizards
(Sceloporus) based on molecular and morphological evidence.
Herpetol.Monogr. 11, 1^101.
Wright, D. B. 1993 Evolution of sexual dimorphic characters in
peccaries (Mammalia,Tayassuidae). Paleobiology 19, 52^70.



.

Quantcast