Female mountain log skinks are more likely to mate with males that court more, not males that are dominant


Stapley, Jessica

Animal Behaviour 75(Part 2): 529-538

2008


To understand the evolution and exaggeration of male traits, we need to clarify the combined and separate contributions of male competition and mate choice to male reproductive success. Here, I tested whether female skinks discriminate between males based on morphological or behavioural traits in a sequential mate choice experiment to identify whether females prefer dominant males. A total of 48 females were tested of which, 26 mated with at least one male. Some females mated multiply with the same male or with both males. Females were more likely to mate with males that courted more and did not prefer males with orange ventral colour, which is indicative of male dominance. Females may use courtship as a cue to male quality because it was positively correlated with the duration of the mating grasp, during which time the male has to carry the female, and mating grasp is also positively correlated with mating duration. Interestingly, males courted smaller females more than larger females, despite the fact that larger females are more fecund. The results of this study suggest that the two processes of sexual selection may be favouring different male traits, resulting in male competition favouring male orange colour and mate choice favouring male courtship.

ANIMAL
BEHAVIOUR,
2008,
75,
529-538
doi:1
0.1
01
6/j.anbehay.2007.05.01
7
Available
online
at
www.sciencedirect.com
**:'
ScienceDirect
ELSEVIER
Female
mountain
log
skinks
are
more
likely
to
mate
with
males
that
court
more,
not
males
that
are
dominant
JESSICA
STAPLEY
School
of
Botany
and
Zoology,
Australian
National
University
(Received
14
November
2006;
initial
acceptance
21
December
2006;
final
acceptance
20
May
2007;
published
online
22
October
2007;
MS.
number:
9174R)
To
understand
the
evolution
and
exaggeration
of
male
traits,
we
need
to
clarify
the
combined
and
separate
contributions
of
male
competition
and
mate
choice
to
male
reproductive
success.
Here,
I
tested
whether
female
skinks
discriminate
between
males
based
on
morphological
or
behavioural
traits
in
a
sequential
mate
choice
experiment
to
identify
whether
females
prefer
dominant
males.
A
total
of
48
females
were
tested
of
which,
26
mated
with
at
least
one
male.
Some
females
mated
multiply
with
the
same
male
or
with
both
males.
Females
were
more
likely
to
mate
with
males
that
courted
more
and
did
not
prefer
males
with
orange
ventral
colour,
which
is
indicative
of
male
dominance.
Females
may
use
courtship
as
a
cue
to
male
quality
because
it
was
positively
correlated
with
the
duration
of
the
mating
grasp,
during
which
time
the
male
has
to
carry
the
female,
and
mating
grasp
is
also
positively
correlated
with
mating
duration.
Interestingly,
males
courted
smaller
females
more
than
larger
females,
despite
the
fact
that
larger
females
are
more
fecund.
The
results
of
this
study
suggest
that
the
two
processes
of
sexual
selection
may
be
favour-
ing
different
male
traits,
resulting
in
male
competition
favouring
male
orange
colour
and
mate
choice
favouring
male
courtship.
©
2007
The
Association
for
the
Study
of
Animal
Behaviour.
Published
by
Elsevier
Ltd.
All
rights
reserved.
Keywords:
body
size;
female
resistance;
information
theoretic;
male
coloration;
male
traits;
mate
preference;
mountain
log
skink;
Pseudemoia
entrecasteauxii
Identifying
the
factors
that
underlie
variation
in
male
reproductive
success
is
essential
to
explaining
the
evolu-
tion
and
exaggeration
of
male
traits.
In
natural
situations,
both
male
competitive
ability
and
female
mate
choice
influence
the
skew
in
male
reproductive
success
(Kokko
et
al.
2003).
These
two
processes
of
sexual
selection
have
each
been
the
focus
of
much
empirical
and
theoretical
research,
however,
few
studies
have
quantified
the
relative
contributions
of
each
to
male
reproductive
success
(for
review
see
Wong
&
Candolin
2005).
Understanding
the
evolution
of
male
sexual
traits
requires
that
we
disentan-
gle
the
multiple
factors
underlying
male
reproductive
success.
Dominant
males
are
often
more
successful
at
securing
paternity
and
in
many
cases
dominant
males
are
consid-
ered
better
quality
mates.
Females
that
mate
with
Correspondence:
I.
Stapley,
Department
of
Animal
and
Plant
Science,
University
of
Sheffield,
Alfred
Denny
Building,
Western
Bank,
Sheffield
S10
2TN,
U.K.
(email:
j.stapley@sheffield.ac.uk).
dominant
males
may
gain
direct
benefits
from
male-
defended
resources
(Candolin
&
Voigt
2001)
or
indirect
benefits
if
competitive
ability
is
heritable
and
her
sons
enjoy
similar
competitive
ability
(e.g.
Moore
1990).
How-
ever,
in
several
cases
females
prefer
traits
unrelated
to
dominance
or
discriminate
against
dominant
males
(Qvarnstrom
&
Forsgren
1998).
Dominance
related
traits
such
as
aggression
and
fighting
ability
could
have
negative
effects
on
female
fitness
and
reproductive
success.
These
manifest
as
direct
costs
to
the
female
such
as
reduced
parental
care
(Forsgren
1997;
Wong
&
Jennions
2003)
or
increased
risk
of
injury
(Le
Boeuf
&
Mesnick
1991).
They
can
also
be
the
result
of
indirect
costs,
for
example,
traits
that
increase
fitness
of
offspring
of
one
sex
but
not
the
other
(Pai
&
Yan
2002).
Behavioural
trade-offs
may
also
re-
duce
a
dominant
male's
ability
to
secure
matings.
In
cases
where
courtship
displays
are
needed
to
attract
females
(Knapp
&
Kovach
1991;
Wong
2004)
or
to
overcome
fema-
le
resistance
(Holland
&
Rice
1998),
dominant/aggressive
males
may
court
females
less,
resulting
in
female
prefer-
ence
for
subordinate
males
(Sih
et
al.
2004).
529
0003-3472/08/$34.00/0
©
2007
The
Association
for
the
Study
of
Animal
Behaviour.
Published
by
Elsevier
Ltd.
All
rights
reserved.
530
ANIMAL
BEHAVIOUR,
75,
2
In
the
sexually
dichromatic
lizard
Pseudemoia
entrecas-
teauxii
the
presence
of
orange
ventral
colour,
not
body
size,
was
the
best
predictor
of
male
contest
success
(Stap-
ley
2006).
Orange
males
are
more
aggressive
and
domi-
nant
over
larger
males
without
orange
venters,
whereas
larger
males
court
more
(Stapley
2006).
In
seminatural
enclosures
large
males
with
orange
venters
sired
the
most
offspring
(Stapley
&
Keogh
2006).
If
females
were
simply
mating
with
the
most
dominant
males,
then
we
would
expect
that
both
small
and
large
males
with
orange
venters
would
have
similar
reproductive
success.
The
fact
that
small
males
with
orange
venters
sired
fewer
offspring
than
their
larger
counterparts
suggests
that
there
is
a
repro-
ductive
advantage
of
body
size
that
could
be
the
result
of
a
female
mating
bias
towards
either
larger
males
or
males
that
court
more.
The
aim
of
this
study
was
to
test
whether
females
discriminate
between
mates
based
on
morpholog-
ical
traits
related
to
dominance
(orange
ventral
colour),
morphological
traits
unrelated
to
dominance
(body
size
and
condition)
or
male
courtship
behaviour.
To
identify
whether
female
mate
choice
and
male
competition
act
in
unison
to
favour
the
exaggeration
of
male
traits.
Traditionally,
female
mate
choice
has
been
tested
in
simultaneous
two-choice
tests.
Females
can
view,
without
physical
contact,
two
males
who
can
in
turn
see
the
female
but
not
the
other
male.
This
design
controls
for
male—male
aggression,
which
can
interfere
with
female
choice,
however,
there
are
several
potential
limitations
to
this
method.
First,
females
may
encounter
males
sequen-
tially
and
as
a
result
cannot
make
simultaneous
contrasts
(Pitcher
et
al.
2003;
Bissell
&
Martins
2006).
Second,
phys-
ical
barriers
such
as
glass
can
preclude
chemical
and
tactile
cues,
which
may
be
important
during
mate
choice
and/or
it
may
alter
male
courtship
behaviour
(Schafer
&
Uhl
2005;
Shackleton
et
al.
2005).
Another
problem
that
arises
in
the
simultaneous
two-choice
design
is
that
female
association,
rather
than
mate
choice,
is
usually
measured.
Few
studies
have
verified
that
female
association
behav-
iour
is
a
good
proxy
for
female
mate
choice
(but
see
Clayton
1990;
Wong
2004)
and
this
has
never
been
confirmed
in
studies
on
lizards.
For
these
reasons,
I
have
used
a
sequential
mate
choice
design
and
allowed
pairs
to
mate,
to
unequivocally
test
mate
choice.
METHODS
Study
Species
and
Animal
Maintenance
The
mountain
log
skink
is
a
small,
diurnal,
viviparous
skink
that
inhabits
cool
temperate
forest
in
southeastern
Australia
(Cogger
1996).
The
skink
is
sexually
dimorphic
for
body
size
and
colour.
Females
attain
longer
snout
vent
lengths
than
males
(mean
snout
vent
length
(SVL):
female
=
51.7
mm;
male
=
49.1
mm)
and
some
males
develop
orange
breeding
coloration
on
their
ventral
surface.
Males
mature
at
an
SVL
of
38
mm
and
females
at
42-mm
SVL
(Pengilley
1972;
Cogger
1996)
and
this
takes
2-3
years
in
cool
alpine
habitats
(S.
Hudson,
unpub-
lished
data).
Age
of
P.
entrecasteauxii
is
positively
correlated
with
SVL
until
maturity,
at
which
point
the
growth
rate
slows
and
age
cannot
be
accurately
estimated
from
SVL
(S.
Hudson,
unpublished
data)
similar
to
other
reptiles
(Halliday
&
Verrell
1988).
Larger
females
have
larger
clutch
sizes;
of
41
field
caught
females
that
gave
birth
in
the
laboratory
dutch
size
ranged
from
one
to
six
and
was
related
to
female
SVL:
x1
„40
=
12.96,
P
=
0.001
(un-
published
data).
The
species
is
not
territorial
and
there
is
a
lot
of
home
range
overlap
between
males
and
females
(Pengilley
1972).
Males
do,
however,
engage
in
aggressive
behaviours
and
this
results
in
the
formation
of
dominance
hierarchies.
Observations
of
P.
entrecasteauxii
in
seminatu-
ral
enclosures
and
in
the
laboratory
found
that
most
inter-
actions
involved
one
male
approaching
the
other
with
an
aggressive
display
(tail
wave)
and
the
other
male
flees
(Stapley
2006;
Stapley
&
Keogh
2006).
During
these
dom-
inance
interactions
males
with
orange
venters
were
more
aggressive
and
dominated,
and
small
males
with
orange
venters
dominated
over
large
males
without
orange
ven-
tral
colour
(Stapley
2006;
Stapley
&
Keogh
2006).
I
collected
lizards
in
January
2003
from
Namadgi
National
Park
by
hand
and
transferred
them
to
individual
cloth
bags
for
transport
to
the
laboratory.
In
the
labora-
tory,
lizards
were
housed
individually
in
plastic
containers
(42
x
32
cm
and
22
cm
high)
in
a
temperature-controlled
room
maintained
at
18-20°C.
The
floor
of
the
container
was
covered
in
paper
and
a
folded
cardboard
roll
was
provided
as
a
retreat
site.
Between
0900
and
1700
hours
a
temperature
gradient
was
established
by
placing
heating
tape
under
one
half
of
the
container
to
allow
animals
to
thermoregulate
freely.
The
room
was
under
a
12:12
h
light:dark
cycle
and
water
was
provided
ad
libitum.
The
lizards
were
fed
with
live
food
(crickets
and
mealworms)
dusted
with
vitamin
powder
every
2
days.
Live
food
was
used
to
reduce
any
acclimation
of
lizards
to
unnatural
food
items,
which
could
reduce
their
survival
when
released.
Animals
were
maintained
in
captivity
for
2
months
and
released
at
their
point
of
capture
at
the
end
of
the
study.
Before
experiments,
I
measured
lizard
SVL
to
the
nearest
millimetre
using
a
ruler,
and
weight
to
the
nearest
0.01
g
using
a
digital
balance.
Condition
was
cal-
culated
using
the
residuals
from
a
linear
regression
of
weight
on
SVL.
I
identified
male
ventral
colour
by
visual
comparison
to
Natural
Colour
Scheme
(NCS)
colour
chart
at
normal
incidence
(full,
morning
sunlight).
Most
male's
ventral
colour
was
identified
as
orange
or
white
(lacking
in
orange
pigment),
with
fewer
males
having
a
ventral
colour
in
between
these
two
colours
(Stapley
&
Keogh
2006).
Because
humans
cannot
see
within
the
ultraviolet
range,
I
measured
the
spectral
reflectance
of
the
venters
of
a
subset
of
males.
There
was
no
evidence
of
reflectance
within
the
UV
range
(320-400
nm)
and
the
orange
col-
our
of
the
male's
ventral
surface
appeared
to
have
peak
reflectance
at
550-650
nm
(J.
Zeil
&
J.
Stapley,
unpub-
lished
data).
Unfortunately,
it
was
not
possible
to
mea-
sure
all
the
males
before
the
experiment.
In
the
absence
of
an
accurate
unbiased
measure
of
male
colora-
tion
it
seems
reasonable
to
discriminate
based
on
the
apparent
dichotomy
that
exists
in
the
species,
which
is
the
presence
or
absence
of
ventral
colour,
as
described
in
previous
study
of
245
field
caught
males
(Stapley
&
Keogh
2006).
STAPLEY:
FEMALES
MATE
WITH
MALES
THAT
COURT
MORE
531
Females
are
receptive
following
parturition
in
January—
February.
I
tested
the
preferences
of
48
postpartum
females
(SVL
range
=
43-61
mm)
in
a
sequential
mate
choice
trial
in
February
2003.
The
experiments
took
place
in
plastic
enclosures
with
the
same
dimensions
and
condi-
tions
as
the
home
enclosures.
I
randomly
assigned
29
adult
males
to
pairs
(SVL
range
=
42-56
mm,
16
white
ventral
colour,
13
orange
ventral
colour).
Males
were
used
repeat-
edly
but
never
in
the
same
pair
and
were
allowed
a
mini-
mum
of
4
days
rest
between
trials.
To
begin
a
trial,
one
randomly
chosen
male
of
each
pair
was
placed
into
a
new
enclosure
with
a
female
at
0800
hours
and
removed
at
1800
hours
on
the
following
day
(34
h
later).
On
the
third
day
at
0800
hours
the
other
male
of
the
pair
was
placed
into
a
new
enclosure
with
the
female
for
34
h
and
both
lizards
were
removed
at
1800
hours
on
the
fourth
day
and
placed
back
into
their
home
enclosures.
Thus,
each
female
spent
34
h
with
each
male.
Lizards
were
provided
with
ad
libitum
water
and
food
during
the
trials.
Enclosures
were
monitored
from
0800
to
1800
hours
by
a
video
camera
mounted
above
and
recorded
with
a
video
recorder
onto
VHS
tape.
Lizard
activity
was
strongly
influenced
by
their
ability
to
obtain
preferred
body
temperature
and
they
are
strictly
diurnal
(Pengilley
1972;
Cogger
1996;
Stapley
2006).
In
all
cases
the
lizards
had
retreated
and
were
no
longer
active
when
cameras
were
turned
off
at
1800
hours.
In
a
few
cases
one
lizard
(usually
the
male)
was
outside
the
refuge
in
the
morning
before
0800
hours,
but
two
lizards
were
never
observed
active
before
cameras
were
turned
on.
I
made
scan
samples
of
the
video
footage
every
15
min
to
quantify
behaviour.
To
record
matings,
I
watched
the
video
in
fast
forward
to
identify
the
exact
timing
and
duration
of
mating.
During
the
scan
samples,
I
recorded
the
lizard's
be-
haviour
(within
a
1-min
window)
and
position.
I
recorded
whether
either
lizard
was
in
the
refuge,
whether
they
were
next
to
each
other
(<50
mm)
and
whether
the
male
was
courting
the
female.
During
courtship
the
male
waves
his
tail
while
approaching
the
female
indirectly
in
an
arc.
The
female
responds
by
either
behaving
aggressively
and
rejecting
the
male
or
accepting
the
courtship
and
mating.
During
rejection
the
female
waves
her
tail,
moves
away
and/or
bites
the
male.
When
the
female
accepts
the
court-
ship
she
allows
him
to
grasp
her
neck
and
they
mate
(sim-
ilar
to
behaviours
described
for
Lampropholis
guichenoti;
Torr
&
Shine
1994).
In
the
approach
and
courtship
bout
be-
fore
mating,
when
a
female
accepts
a
male,
females
never
tail
waved.
There
was
no
other
female
behaviours
observed
in
this
study
and
female
choice
was
determined
by
whether
she
mated
with
the
male
or
not.
Courtship
index
was
used
as
a
measure
of
male
court-
ship
effort
and
was
calculated
using
the
following
func-
tion:
courtship
intensity/activity.
Where
courtship
intensity
is
the
number
of
scans
the
male
was
observed
courting
the
female
minus
the
number
of
scans
he
was
next
to
her
(within
50
mm).
Activity
is
the
total
number
of
observations
minus
the
number
of
observations
one
or
more
lizard
was
in
the
retreat
site.
This
provided
a
mea-
sure
of
courtship
effort
as
a
function
of
opportunity.
Males
only
court
when
in
close
proximity
(<50
mm),
some
males
court
at
every
opportunity
(every
time
they
are
near
the
female)
whereas
other
males
court
at
a
much
lower
frequency
despite
ample
opportunity,
that
is,
next
to
the
female.
This
measure
is
highly
correlated
with
proportion
of
time
spent
courting
but
unlike
proportion
of
time
spent
courting,
courtship
index
could
be
trans-
formed
(square-root
transformation)
to
meet
normality.
Mating
involves
a
premating
grasp,
where
the
male
bites
the
female
holding
her
across
the
neck/thoracic
region
and
carries
her
around
before
mating
(Pengilley
1972).
Females
normally
go
limp
during
the
premating
grasp
with
their
limbs
held
close
to
their
body,
not
making
contact
with
the
ground
(personal
observation).
Mating
occurs
when
the
male
inserts
one
of
his
hemipenes
and
releases
his
grasp
on
the
female.
It
is
unlikely
that
forced
copulation
is
common
in
this
species
because
females
are
larger
than
males
(Pengilley
1972;
Cogger
1996)
and
in
one
case,
I
observed
a
female
struggle
and
free
herself
of
a
premating
grasp
before
intromission.
During
this
study,
all
females
that
mate
appeared
to
cooperate
with
the
male
(i.e.
she
was
limp
and
held
her
forelegs
flush
against
the
sides
of
her
body).
The
premating
grasp
results
in
a
mating
scar
and
the
number
of
mating
scars
is
a
reli-
able
indicator
of
mating
history
(Pengilley
1972).
At
the
end
of
each
trial,
I
recorded
the
presence
of
a
mating
scar
to
confirm
whether
a
mating
had
taken
place.
This
was
cross-referenced
with
the
analysis
of
videotapes
to
ensure
that
no
mating
had
taken
place
at
night
or
very
early
in
the
morning
before
the
cameras
were
turned
on.
The
research
described
in
this
paper
was
approved
by
the
Animal
Experimentation
and
Ethics
Committee
of
the
Australian
National
University
(Protocol
number:
F.BTZ.17.00)
and
is
in
accordance
with
the
ASAB
Guide-
lines
for
the
use
of
Animals
in
Research.
Data
Analysis
The
continuous
variables,
male
and
female
SVLs,
male
courtship
index
and
male
and
female
condition
(residuals
of
a
linear
regression
of
weight
on
SVL),
were
standardized
by
subtracting
the
mean
from
each
observation
and
then
dividing
this
by
the
standard
deviation.
Data
analysis
was
carried
out
using
information
theoretic
(IT)
model
selection
approach
based
on
Akiake's
Information
Crite-
rion
(AIC)
as
outlined
in
Burnham
&
Anderson
(2002).
Al-
though,
in
some
cases
it
may
be
beneficial
to
use
the
global
model
containing
all
explanatory
variables,
the
in-
clusion
of
nuisance
variables
often
results
in
imprecise
pa-
rameter
estimates
(Burnham
&
Anderson
2002).
For
this
reason,
I
constructed
a
set
of
candidate
models
for
each
response
term
and
compared
the
fit
of
each
of
these
models.
Each
model
represents
a
different
hypothesis
and
these
models
were
compared
in
their
entirety
(Burnham
&
Anderson
2002;
Whittingham
et
al.
2006).
In
all
the
cases,
AIC
was
corrected
for
small
sample
size
(also
called
the
second-order
criterion
AIC,),
this
is
neces-
sary
when
the
ratio
of
the
number
of
observations
(N)
to
the
number
of
parameters
(K)
is
small
(<40;
Burnham
&
Anderson
2002)
and
was
calculated
as:
AIC
=
AIC
+
((2K(K
+
1))1
(N
K
1))
(1)
532
ANIMAL
BEHAVIOUR,
75,
2
Model
AIC
was
calculated
and
compared
between
the
candidate
models,
the
model
with
the
lowest
AIC
(AIC„,
th
)
is
the
best
fitting
model
and
the
relative
change
in
AIC
(6.,)
between
models
was
calculated
using:
Ai
=
AIC,
AIGnin
(2)
To
compare
between
R
set
of
models
the
Akaike
weights
(w,)
are
calculated
as
follows:
exP
w,
R
(
3
)
E
exp
(-00
r=1
These
weights
provide
a
useful
way
to
compare
between
models,
as
w,
is
a
relative
estimate
of
the
weight
of
evidence
in
favour
of
i
being
the
best
model
(Burnham
&
Anderson
2002).
Models
with
high
A
i
(>10)
have
low
co
b
and
subsequently
little
evidence
that
this
is
the
best
model.
In
cases
where
there
is
a
clear
best
model
(w
>
0.90),
this
model's
parameter
estimates
are
used
(Burnham
&
Ander-
son
2002;
Whittingham
et
al.
2006).
The
direction
and
magnitude
of
the
explanatory
variable's
effect
on
the
re-
sponse
term
is
based
on
the
parameter
estimate
and
its
95%
confidence
interval
(CI).
Confidence
intervals
were
es-
timated
using
Markov
chain
Monte
Carlo
sampling
method
and
Bayesian
highest
posterior
density
(Baayen
et
al.,
submitted
for
publication).
If
the
CIs
do
not
overlap
with
zero
then
the
variable
in
question
is
said
to
be
having
a
strong
effect
on
the
response
term
(Baayen
et
al.,
submit-
ted
for
publication).
Analyses
were
carried
out
using
the
statistical
package
R
2.4.2
and
the
models
were
fitted
with
the
package
lme4,
with
specific
details
given
below
for
each
model
(R
Development
Core
Team
2006).
To
investigate
the
influence
of
male
traits
on
female
mate
choice,
I
constructed
generalized
linear
mixed
models
(GLMMs)
to
test
the
following
hypothesis:
Model
1:
mate
traits
interact
to
influence
mate
choice
(model
contained
all
morphological
and
behavioural
traits
measured:
SVL,
colour,
condition
and
courtship
index
and
their
second-order
interactions);
Model
2:
females
mate
with
dominant
males
(model
contained
male
colour;
Stapley
2006;
Stapley
&
Keogh
2006);
Model
3:
females
discriminate
between
males
based
on
morphological
traits
unrelated
to
dominance
(model
contained
SVL,
condition
and
their
interaction);
Model
4:
females
discriminate
between
males
based
on
their
courtship
(model
contained
courtship
index);
Model
5:
males
and
females
mate
size
assortatively
(model
contained
male
and
female
body
size);
and
Model
6:
female
body
size
influences
her
proba-
bility
of
mating
(model
contained
female
body
size).
All
GLMMs
contained
three
random
terms,
male
identity,
female
identity
and
male
number
(first
or
second),
were
fitted
with
a
Binomial
error
structure
with
a
logit
link
function
using
the
Laplace
method
(which
is
an
approxi-
mation
of
the
Maximum
Likelihood).
To
investigate
factors
influencing
male
courtship
index,
I
constructed
linear
mixed
models
(LMMs)
to
test
the
following
hypothesis:
Model
1:
courtship
index
was
related
to
male
morphological
traits
(model
contained
all
male
traits
measured:
SVL,
colour
and
condition
and
their
second-order
interactions);
Model
2:
courtship
index
is
related
to
male
dominance
(model
contained
male
colour);
Model
3:
courtship
index
is
related
to
male
traits
unrelated
to
dominance
(model
contained
SVL,
condition
and
the
interaction);
Model
4:
courtship
index
is
related
to
male
body
size
because
this
has
previously
been
shown
to
be
correlated
with
courtship
index
(Stapley
2006;
model
con-
tained
male
SVL);
and
Model
5:
males
court
larger,
more
fecund
females
(model
contained
female
SVL).
All
models
contained
three
random
terms:
male
identity,
female
iden-
tity
and
male
number
(first
or
second),
the
LMMs
were
fit-
ted
using
the
Maximum
Likelihood
method.
For
each
mating,
the
grasp
latency
(from
the
beginning
of
the
trial
until
the
male
grasps
the
female),
the
grasp
duration
(time
from
when
the
male
grasps
the
female
to
intromission)
and
mating
duration
(time
from
intromis-
sion
to
separation)
were
recorded
and
were
log
transformed
to
meet
normality.
Any
relationship
between
grasp
latency,
grasp
duration
and
mate
duration
was
tested
with
Pearson
correlations.
In
the
case
where
a
pair
was
observed
mating
more
than
once
the
mean
for
that
pair
was
used.
To
investigate
how
male
and
female
traits
were
related
to
grasp
latency,
grasp
duration
and
mating
duration
I
used
the
IT
approach
outlined
above
and
constructed
LMMs
for
each
response
variable
and
included
the
random
terms,
male
and
female
identity.
These
models
were
fitted
using
the
Maximum
Likelihood
method.
For
grasp
latency,
I
constructed
LMMs
to
test
the
follow-
ing
hypothesis:
Model
1:
grasp
latency
is
related
to
female
traits
(model
contained
female
SVL,
female
condition
and
the
interaction
term);
Model
2:
grasp
latency
is
related
to
male
behaviour
(model
contained
courtship
index);
Model
3:
grasp
latency
is
related
to
male
traits
that
confer
dominance
(male
colour);
and
Model
4:
grasp
latency
is
related
to
male
morphological
traits
unrelated
to
dominance
(model
contained
male
SVL,
condition
and
the
interaction).
For
grasp
duration,
I
constructed
LMMs
testing
the
following
hypotheses:
Model
1:
the
difference
between
male
and
female
body
size
will
effect
grasp
duration
because
smaller
males
will
have
difficulty
in
carrying
larger
females
(model
contained
male
SVL—female
SVL);
Model
2:
traits
that
confer
dominance
will
influence
grasp
duration
(model
contained
male
colour);
Model
3:
grasp
duration
is
related
to
male
morphological
traits
unrelated
to
dominance
(model
contained
male
SVL,
condition
and
the
interaction);
and
Model
4:
grasp
duration
is
related
to
male
behaviour
(model
contained
courtship
index).
For
mating
duration,
I
constructed
LMMs
to
test
the
following
hypotheses:
Model
1:
mating
duration
is
related
to
female
traits
(model
contained
female
SVL,
female
condition
and
the
interaction
term);
Model
2:
mating
duration
is
related
to
male
traits
that
confer
dominance
(model
contained
male
colour);
Model
3:
mating
duration
is
related
to
male
morphological
traits
unrelated
to
dom-
inance
(model
contained
male
SVL,
condition
and
the
interaction);
Model
4:
mating
duration
is
related
to
behaviour
(model
contained
courtship
index);
and
Model
5:
mating
duration
is
longer
if
the
female
has
previously
mated
(model
contained
mating
history,
mated/unmated).
As
each
male
was
observed
with
a
female
for
2
days
and
males
were
used
repeatedly
with
different
females,
I
could
test
how
repeatable
male
courtship
behaviour
was.
STAPLEY:
FEMALES
MATE
WITH
MALES
THAT
COURT
MORE
533
Repeatability
was
calculated
using
equations
(2)-(5)
from
Lessells
&
Boag
(1987).
Two
measures
of
repeatability
were
calculated,
the
first
one
measured
how
repeatable
male
behaviour
was
between
the
first
and
second
day
he
was
paired
with
the
same
female
(repeatability
was
calculated
for
each
male
only
once
N
=
28).
The
second
repeatability
estimate
was
calculated
for
each
male
between
trials
with
a
different
female
(N
=
28).
I
used
the
mean
courtship
index
(across
day
1
and
day
2)
for
each
male
with
the
first
female
compared
to
the
mean
courtship
index
for
the
second
female
he
was
observed
with.
This
gives
two
repeatability
estimates,
one
quantifies
repeatability
across
days
with
the
same
female,
and
the
other
quantifies
repeatability
across
different
females.
RESULTS
In
total
26
(54%)
females
mated,
five
females
mated
more
than
once
and
a
total
of
34
matings
were
recorded.
Of
those
females
to
mate
more
than
once,
three
females
mated
with
both
males,
one
female
mated
twice
with
the
same
male
once
with
the
other
male,
and
one
female
mated
three
times
with
the
one
male
and
once
with
the
other
male.
Excluding
females
that
mated
more
than
once,
females
were
not
more
likely
to
mate
with
the
first
male
(12)
than
the
second
(9;
Exact
binomial
test:
P
=
0.66),
but
more
fe-
males
mated
on
the
first
day
she
was
paired
with
the
male
(24)
than
on
the
second
day
(10;
P
=
0.02).
Female
snout
vent
length
was
not
related
to
the
number
of
times
she
mated
(generalized
linear
model,
GLM,
with
Poisson
error:
XT,23
=
0.79,
P
=
0.37)
or
the
likelihood
that
she
mated
during
the
trials
(Table
1).
In
a
total
of
18
cases,
the
male's
body
size
was
equal
to
or
larger
than
the
female,
compared
to
11
when
the
male's
body
size
was
less
than
the
female,
but
there
was
no
evidence
of
size-assortative
mating
be-
tween
pairs
(Table
1).
Comparing
between
models
constructed
to
explain
female
mate
choice,
the
best
model
(W
i
>
0.90)
for
ex-
plaining
variation
in
the
probability
of
mating
was
the
model
that
contained
courtship
index
(Table
1).
The
slope
estimate
(95%
CI)
of
courtship
index
is
1.13
(0.35-1.20)
suggesting
that
courtship
index
is
having
a
positive
effect
on
a
male's
probability
of
mating
(Fig.
1).
Male
colour
a
trait
that
predicts
male
dominance
performed
poorly
as
an
explanatory
variable,
suggesting
that
females
do
not
prefer
to
mate
with
dominant
males.
When
considering
variables
that
influenced
male
courtship
index
(Table
2),
the
model
containing
female
body
size
was
the
best
model
(W
i
>
0.90).
But,
contrary
my
predic-
tion
the
relationship
was
negative
with
an
estimated
slope
of
-0.34
(95%
CI:
-0.54
to
-0.13),
suggesting
that
male
courtship
index
was
higher
for
smaller
females
(Fig.
2).
When
paired
with
different
females,
an
individual
male's
courtship
index
was
repeatable
across
the
two
observational
days
(r
=
0.95),
in
other
words
individual
males
that
courted
more
on
day
1
also
courted
more
on
day
2.
Males
showed
similar
repeatability
in
individual
courtship
rates
when
paired
with
different
females
(r
=
0.93).
As
a
result,
individual
males
that
had
a
high
courtship
index
when
paired
with
one
female
also
showed
relatively
high
courtship
rates
with
a
different
female
(Fig.
3).
Similar
repeatability
in
male
behaviour
was
also
found
in
a
previous
study
(Stapley
2006).
The
average
time
to
mating
(grasp
latency)
was
over
5
h
(mean
±
SE:
397.25
±
147.8
min).
The
mean
number
(±SE)
of
times
a
male
was
observed
courting
a
female
be-
fore
the
grasp
was
3.59
±
0.61.
Males
carried
females
in
the
premating
grasp
for
4-33
min
(mean
=
13.72
±
1.39)
and
mating
took
between
1
and
41
min
(mean
=
9.92
±
1.71).
Considering
the
three
mating
behaviours,
(grasp
latency,
grasp
duration
and
mating
duration),
a
positive
correlation
was
found
between
grasp
duration
and
mating
duration
(Table
3,
Fig.
4),
but
no
other
correlations
were
observed.
Of
the
candidate
models
constructed
to
explain
variation
in
grasp
latency
and
mating
duration,
no
single
best
model
could
be
identified
(W
i
>
0.90;
Table
4).
This
suggests
that
the
variation
in
male
and
female
traits
measured
in
this
study
did
not
influence
grasp
latency
or
mating
duration.
In
contrast,
a
single
best
model
was
identified
to
explain
variation
in
grasp
duration
(Table
4).
Males
with
high
courtship
index
also
had
longer
grasp
durations
(Fig.
5).
DISCUSSION
Female
mating
decisions
were
influenced
by
male
behav-
iour,
but
not
by
the
male's
morphological
traits.
Females
were
more
likely
to
mate
with
males
that
courted
more.
Male
body
size,
condition
and/or
ventral
colour
did
not
influence
the
probability
of
mating.
In
this
species,
males
with
orange
ventral
colour
are
more
aggressive
and
dominate
males
with
white
venters
irrespective
of
male
body
size
(Stapley
2006).
These
results
suggest
that
Table
1.
Generalized
linear
mixed
models*
constructed
to
explain
female
mate
choice
Model
Explanatory
variables
logLik
K
AIC
Wi
4
Male
behaviour
-51.86
4
114.13
0
0.99
6
Female
body
size
-57.22
4
124.83
9.91
0.006
5
Size-assortative
mating
-56.44
6
127.84
13.71
0.001
3
Male
traits
unrelated
to
dominance
(size*condition)
-56.54
6
128.04
13.91
0.00091
2
Dominance
related
trait
(male
colour)
-59.01
4
128.43
14.3
0.00078
All
male
traits
and
second
order
interactions
-46.14
13
134.3
20.07
0.00004
Model
in
bold
is
the
best
fit
w,
>
0.90
(Burnham
&
Anderson
2002).
*Binomial
errors
structure,
logit
link
function,
Laplace
estimation
method,
random
terms:
male
identity;
female
identity;
male
number
(first
or
second).
534
ANIMAL
BEHAVIOUR,
75,
2
1
0.5
0.4
Nonpreferred
males
Preferred
males
Figure
1.
Mean
square-root
transformed
(SRT)
courtship
index
of
nonpreferred
and
preferred
males.
females
do
not
prefer
dominant
males
and
it
is
likely
that
female
mate
choice
and
male
competition
favour
the
ex-
aggeration
of
different
male
traits.
Dominant
orange
males
are
presumably
able
to
restrict
subordinate
male's
access
to
females
particularly
when
lizards
are
congregated
around
basking
sites.
After
reaching
preferred
body
temperatures,
however,
lizards
disperse
and
actively
forage
in
the
leaf
litter
(Pengilley
1972).
It
is
then
that
subordi-
nate
males
may
gain
access
to
females
and
through
greater
courtship
effort
secure
matings.
This
is
supported
by
the
paternity
data,
which
showed
that
subordinate
males
lacking
orange
colour
were
not
completely
excluded
from
siring
offspring
(Stapley
&
Keogh
2006).
These
results
confirm
theoretical
predictions
that
multiple
male
traits
may
be
maintained
by
a
combination
of
differ-
ent
selective
mechanisms
(Wong
&
Candolin
2005).
Results
of
experimental
mate
choice
studies
such
as
this,
where
male
competition
can
be
excluded,
enable
us
to
dis-
entangle
the
multiple
factors
underlying
male
reproduc-
tive
success.
Male
variation
in
courtship
rate
may
be
indicative
of
variation
in
male
quality.
In
other
species,
courtship
index
is
indicative
of
male
quality,
relating
to
fat
reserves
(Knapp
&
Kovach
1991)
or
male
parental
ability
(Ostlund
&
Ahnesjo
1998;
Wong
2004).
The
courtship
index
of
male
P.
entrecasteauxii
was
positively
correlated
with
grasp
dura-
tion,
which
may
be
related
to
male
quality.
The
act
of
carrying
the
female
in
the
mating
grasp
may
require
considerable
strength
endurance
and/or
energy
expendi-
ture.
Males
that
carried
the
female
for
longer
in
the
8
1
••
.a.
••
-
g
0.6-
8
It
0
.
U
.
••
"
8
II
0
I
I
40
45
50
55
60
65
Female
SVL
(mm)
Figure
2.
Observed
relationship
between
male
SRT
courtship
index
and
female
SVL.
premating
grasp
also
had
longer
mating
times,
which
may
result
in
greater
ejaculate
volume
and/or
better
fertilization
success
(Simmons
et
al.
1996;
Vermette
&
Fairbairn
2002).
Ejaculate
quantity
or
quality
may
be
very
important
in
this
species
because
females
store
sperm
over
a
6-month
winter
hibernation
period
and
do
not
fertilize
their
eggs
un-
til
the
following
spring
(Pengilley
1972;
Cogger
1996).
In
a
previous
study,
males
with
a
greater
courtship
index
were
larger
and
maintained
higher
preferred
body
tempera-
tures
(Stapley
2006).
In
the
current
study,
male
body
size
explained
little
variation
in
courtship
index.
Greater
court-
ship
index
and
preferred
body
temperature
of
some
males
observed
in
the
previous
study
were
thought
to
be
related
to
a
shy-bold
continuum,
males
at
the
upper
end
of
this
continuum
are
bolder
and
more
likely
to
engage
in
social
in-
teractions
(Stapley
2006).
In
the
previous
study,
males
were
only
observed
for
30
min
and
it
is
possible
that
this
may
account
for
the
disparity
between
studies.
Over
a
short
period
of
time
in
a
novel
environment,
as
in
the
previous
study,
differences
between
bold
and
shy
individuals
may
be
more
pronounced.
Over
a
longer
period
of
time
other
factors
such
as
female
body
size
and
male
acclimation
may
override
the
shy-bold
effect.
It
is
possible
that
females
provided
males
with
subtle
cues
to
encourage
courtship,
and
as
a
result,
females
I
1.2-
.5,
0.8-
0.4-
0.2-
1.4
•••
Table
2.
Linear
mixed
models*
constructed
to
explain
the
variation
in
male
courtship
index
Model
Explanatory
variables
logLik
K
AIC
Wi
5
Female
body
size
-132.8
4
275.95
0
0.958
4
Male
size
-136.5
4
283.43
7.48
0.022
2
Dominance
related
trait
(male
colour)
-137.2
4
284.93
8.98
0.010
3
Male
traits
unrelated
to
dominance
(size*condition)
-135.5
6
285.84
9.89
0.006
1
All
male
traits
and
second
order
interactions
-134.7
9
291.19
15.24
0.004
Model
in
bold
is
the
best
fit
w,>
0.90
(Burnham
&
Anderson
2002).
*Random
terms:
male
identity;
female
identity;
male
number
(first
or
second).
0.
STAPLEY:
FEMALES
MATE
WITH
MALES
THAT
COURT
MORE
535
1.4
1.2-
u
tu
0.8-
'15
4
0
0
6
SO
.2
0.4-
o
0.2-
0
—0.2
0
0.2
0.4
0.6
0.8
1
1.2
1
4
Courtship
index
first
female
Figure
3.
Relationship
of
individual
male's
SRT
courtship
index
of
the
first
female
with
the
SRT
courtship
index
of
a
second
female.
Each
point
represents
a
single
male.
encouraged
preferred
males
to
court
more
rather
than
directly
preferring
males
that
courted
more.
The
consis-
tency
and
repeatability
of
male
courtship
behaviour,
however,
would
suggest
that
male
courtship
was
mostly
male
driven
and
less
influenced
by
female
behaviour.
In
line
with
the
findings
of
a
previous
study
(Stapley
2006),
there
was
large
individual
variation
in
courtship
index
and
repeatability
in
individual
male
courtship
index,
both
between
days
when
paired
with
the
same
female
and
between
observations
when
a
male
was
paired
with
a
different
female.
Together,
this
suggests
that
the
interin-
dividual
variation
in
male
courtship
index
represents
actual
differences
between
males
in
their
propensity
to
court
rather
than
differences
in
the
female's
receptivity
to
the
courting
male.
Repeatability
in
behaviour
is
often
considered
to
set
the
upper
bounds
of
heritability
(Boake
1989),
behaviours
that
are
repeatable
are
more
likely
to
have
higher
heritability
and
respond
to
natural
selection
faster
than
behaviours
with
low
repeatability
(Brodie
&
Russell
1999).
Dominant
males
are
often
considered
better
quality
mates
because
they
may
give
females
access
to
Table
3.
Pearson
correlations
between
mating
behaviours
Grasp
Grasp
Mating
latency*
duration* duration*
Grasp
latency*
0.04
0.18
Grasp
0.82
0.67
duration*
Mating
0.33
<0.001
duration*
Correlations
in
bold
are
significant
at
P
<
0.05
after
Bonferroni
cor-
rection.
The
correlation
coefficients
are
in
the
top
diagonal
of
the
matrix
and
P
values
in
bottom
half.
*Data
were
square-root
transformed.
4
3.5
-
3-
C
0
cts
2.5
-
2
-
bp
1.5
-
0.5
1
5
2
2.5
3
3.5
4
Log
grasp
duration
Figure
4.
Mating
duration
was
positively
correlated
with
grasp
dura-
tion.
Points
represent
observed
relationship
between
the
log
of
mat-
ing
duration
and
the
log
of
grasp
duration.
defended
resources
and/or
if
dominance
is
inherited
her
sons
will
enjoy
a
similar
competitive
ability.
The
results
of
this
study
suggest
that
dominant
males
are
not
necessar-
ily
better
quality
mates
and
courtship
may
be
a
consistent,
and
therefore,
reliable
indicator
of
other
aspects
of
quality.
What
is
surprising
is
that
males
courted
smaller
females
more.
In
many
taxa
female
body
size
is
positively
correlated
with
dutch
size
(Honek
1993;
Brana
1996;
Bon-
duriansky
2001;
Prado
&
Haddad
2005).
This
pattern
is
also
present
in
P.
entrecasteauxii
(unpublished
data).
In
dwarf
chameleons,
males
court
smaller
females
more
be-
cause
of
the
risk
of
injury
from
large
aggressive
females
(Stuart-Fox
&
Whiting
2005).
In
P.
entrecasteauxii,
similar
to
dwarf
chameleons,
females
can
attain
longer
snout
vent
lengths
than
males,
but
little
aggressive
behaviour
was
observed
during
this
study.
One
possible
explanation
for
increased
courtship
of
smaller
females
relates
to
the
premating
grasps.
Males
may
not
be
physically
capable
of
biting/carrying
females
larger
than
themselves.
Cer-
tainly,
matings
were
more
common
when
the
male
was
of
similar
or
larger
size
than
the
female.
Alternatively,
there
may
be
differences
between
small
and
large
females
in
their
choosiness
or
mating
frequency.
If
smaller
females
mated
more
frequently
then
it
could
be
beneficial
for
males
to
allocate
more
effort
to
courting
smaller
females.
Females
that
mated
three
and
four
times
were
two
of
the
smallest
females
in
this
study
(45
mm),
but
there
was
no
evidence
of
a
relationship
between
mating
number
and
female
snout
vent
length,
or
a
relationship
between
the
latency
to
mating
and
female
snout
vent
length.
The
fact
that
male
courtship
varied
with
female
body
size
and
that
courtship
index
was
a
strong
predictor
of
male
mating
success
suggests
that
male
and
female
mate
choices
may
interact
to
influence
a
male's
mating
success.
Mutual
mate
choice,
although
considered
rare
(Kokko
&
Johnstone
2002),
is
likely
to
be
influencing
male
41
536
ANIMAL
BEHAVIOUR,
75,
2
Table
4.
Linear
mixed
models*
constructed
to
explain
the
variation
in
(a)
grasp
latency,
(b)
grass
duration
and
(c)
mating
duration
Model
Explanatory
variables
logLik
K
AIC
Wi
(a)
Grasp
latency
1
Female
traits
(size*condition)
-42.17
5
97.13
0
0.57
2
Courtship
-44.86
3
99.08
1.95
0.21
3
Male
colour
-44.96
3
99.28
2.15
0.19
4
Male
traits
(size*condition)
-42.84
3
106.81
9.68
0.004
(b)
Grasp
duration
4
Courtship
index
8.76
3
-8.71
0
0.90
1
Size
difference
6.00
3
-3.10
5.61
0.05
2
Male
colour
5.61
3
-2.42
6.29
0.03
3
Male
traits
(size*condition)
5.83
5
2.47
11.18
0.003
(c)
Mating
duration
4
Courtship
index
-15.69
3
40.19
0
0.45
2
Male
colour
-15.94
3
40.67
0.48
0.35
1
Female
traits
(size*condition)
-14.70
5
43.53
3.34
0.08
5
Mated/unmated
-18.01
3
44.01
3.82
0.06
3
Male
traits
(size*condition)
-15.82
4
45.87
5.38
0.03
Model
in
bold
is
the
best
fit
w,
>
0.90
(Burnham
&
Anderson
2002).
*Random
terms:
male
identity;
female
identity.
reproductive
success.
Smaller
females
are
courted
intensely
by
small
and
large
males,
but
larger
females
are
only
courted
by
larger
males.
This
does
not
result
in
size-
assortative
mating,
but
it
may
limit
the
number
of
poten-
tial
mates
available
for
larger
females,
thereby
constraining
larger
female
reproductive
options.
An
important
goal
of
evolutionary
biology
is
to
understand
what
factors
may
constrain
female
mate
choice
(Jennions
&
Petrie
1997;
Kokko
et
al.
2003).
Male
competition
can
limit
female
reproductive
options
if
dominant
males
exclude
subordi-
nate
males
from
access
to
females
(Jennions
&
Petrie
1997;
Stapley
&
Keogh
2005).
The
results
of
this
study
suggest
that
male
preference
for
smaller
females
may
sim-
ilarly
constrain
female
reproductive
options.
1.4
1.2
1
0.8
^,74
0
0.6
0.4
0.2
0
6
0.8
1
1.2
1.4
16
Log
grasp
duration
Figure
5.
Observed
relationship
between
log
of
grasp
duration
and
SRT
male
courtship
index.
There
has
been
little
success
in
identifying
female
mate
choice
for
male
traits
in
lizards
compared
to
other
taxa
such
as
insects
and
birds.
Of
17
published
studies
experimentally
testing
mate
choice
in
lizards
(Greenberg
&
Noble
1944;
Crews
1975;
Sigmund
1983;
Andrews
1985;
Olsson
&
Mad-
sen
1995;
Baird
et
al.
1997;
Smith
&
Zucker
1997;
Martin
&
Lopez
2000;
Lebas
&
Marshall
2001;
Olsson
2001;
Kwiat-
kowski
&
Sullivan
2002;
Lopez
et
al.
2002;
Tokarz
2002;
Ols-
son
et
al.
2003;
Hamilton
&
Sullivan
2005;
Bissell
&
Martins
2006),
only
one
study
has
convincingly
showed
female
mate
choice
(as
measured
by
actual
mating)
and
this
used
a
sequential
mate
choice
design
(Cooper
&
Vitt
1993).
There
have
been
several
arguments
made
as
to
why
mate
choice
is
rare
in
lizards
(Olsson
&
Madsen
1995;
Tokarz
1995;
Olsson
&
Madsen
1998),
but
I
propose
that
experimental
design
has
been
partly
at
fault.
Most
importantly,
studies
have
failed
to
verify
that
female
association
behaviour
accurately
represents
mate
choice.
It
is
probably
that
female
lizards
do
not
necessarily
search
for,
or
associate
with,
preferred
mates
but
discriminate
between
males
that
court
them.
As
such
si-
multaneous
two-choice
tests
where
mating
is
prohibited
are
likely
to
be
unsuccessful.
These
fundamental
differ-
ences
in
how
female
lizards
discriminate
between
mates
may
provide
a
base
for
novel
advances
in
our
understand-
ing
of
the
evolution
of
mate
choice
and
male
traits.
Acknowledgments
I
thank
Scott
Keogh,
Michael
Jennions,
Bob
Wong,
Devi
Stuart-Fox
and
the
anonymous
referees
for
comments
on
the
manuscript.
Funding
was
provided
by
the
Australian
Geographic
Society
and
the
Australian
Federation
of
University
Women.
References
Andrews,
R.
M.
1985.
Mate
choice
by
females
of
the
lizards,
Anolis
carolinensis.
Journal
of
Herpetology,
19,
284-289.
••
••
STAPLEY:
FEMALES
MATE
WITH
MALES
THAT
COURT
MORE
537
Baayen,
R.
H.,
Davidson,
D.
J.
&
Bates,
D.
M.
Submitted
for
publication.
Mixed-effects
modeling
with
crossed
random
effects
for
subjects
and
items
(http://www.mpi.nl/world/persons/private/
baayen/publications.html).
Baird,
T.
A.,
Fox,
S.
F.
&
McCoy,
J.
K.
1997.
Population
differences
in
the
roles
of
size
and
coloration
in
intra-
and
intersexual
selection
in
the
collared
lizard,
Crotaphytus
collaris:
influence
of
habitat
and
social
organization.
Behavioral
Ecology,
8,
506-517.
Bissell,
A.
N.
&
Martins,
E.
P.
2006.
Male
approach
and
female
avoidance
as
mechanisms
of
population
discrimination
in
sage-
bush
lizards.
Behavioral
Ecology
and
Sociobiology,
60,
655-662.
Boake,
C.
R.
B.
1989.
Repeatability:
its
role
in
evolutionary
studies
of
mating
behavior.
Evolutionary
Ecology,
3,
173-182.
Bonduriansky,
R.
2001.
The
evolution
of
male
mate
choice
in
insects:
a
synthesis
of
ideas
and
evidence.
Biological
Reviews,
76,
305-339.
Brana,
F.
1996.
Sexual
dimorphism
in
lacertid
lizards:
male
head
increase
vs
female
abdomen
increase?
Oikos,
75,
511-523.
Brodie,
E.
D.
&
Russell,
N.
H.
1999.
The
consistency
of
individual
differences
in
behaviour:
temperature
effects
on
antipredator
behaviour
in
garter
snakes.
Animal
Behaviour,
57,
445-451.
Burnham,
K.
P.
&
Anderson,
D.
R.
2002.
Model
Selection
and
Infer-
ence:
a
Practical
Information-theoretic
Approach.
2nd
edn.
New
York:
Springer-Verlag.
Candolin,
U.
&
Voigt,
H.-R.
2001.
Correlation
between
male
size
and
territory
quality:
consequence
of
male
competition
or
preda-
tion
susceptibility.
Oikos,
95,
225-230.
Clayton,
N.
S.
1990.
Assortative
mating
in
zebra
finch
subspecies,
Taeniopygia
guttata guttata
and
T.
g.
castanotis.
Philosophical
Transactions
of
the
Royal
Society
of
London,
Series
B,
330,
351-370.
Cogger,
H.
G.
1996.
Reptiles
and
Amphibians
of
Australia.
Sydney:
Reed.
Cooper,
W.
E.
J.
&
Vitt,
L.
J.
1993.
Female
mate
choice
of
large
broad-headed
skinks.
Animal
Behaviour,
45,
683-693.
Crews,
D.
1975.
Effects
of
different
components
of
male
courtship
behaviour
on
environmentally
induced
ovarian
recrudescence
and
mating
preferences
in
the
lizard,
Anolis
carolinensis.
Animal
Behaviour,
23,
349-356.
Forsgren,
E.
1997.
Female
sand
gobies
prefer
good
fathers
over
dominant
males.
Proceedings
of
the
Royal
Society
of
London,
Series
B,
264,
1283-1286.
Greenberg,
B.
&
Noble,
G.
K.
1944.
Social
behavior
of
the
Ameri-
can
chameleon
(Anolis
carolinensis).
Physiological
Zoology,
17,
392-439.
Halliday,
T.
R.
&
Verrell,
P.
A.
1988.
Body
size
and
age
in
amphib-
ians
and
reptiles.
Journal
of
Herpetology,
22,
253-265.
Hamilton,
P.
S.
&
Sullivan,
B.
K.
2005.
Female
mate
attraction
in
ornate
tree
lizards,
Urosaurus
omatus:
a
multivariate
analysis.
Animal
Behaviour,
69,
219-224.
Holland,
B.
&
Rice,
W.
R.
1998.
Chase-away
sexual
selection:
antag-
onistic
seduction
versus
resistance.
Evolution,
52,
1-7.
Honek,
A.
1993.
Intraspecific
variation
in
body
size
and
fecundity
in
insects:
a
general
relationship.
Oikos,
66,
483-492.
Jennions,
M.
D.
&
Petrie,
M.
1997.
Variation
in
mate
choice
and
mating
preferences:
a
review
of
causes
and
consequences.
(review)
Biological
Reviews
of
the
Cambridge
Philosophical
Society,
72,
283-327.
Knapp,
R.
A.
&
Kovach,
J.
T.
1991.
Courtship
as
an
honest
indicator
of
male
parental
quality
in
the
bicolor
damselfish,
Stegastes
parti-
tus.
Behavioral
Ecology,
2,
295-300.
Kokko,
H.
&
Johnstone,
R.
A.
2002.
Why
is
mutual
mate
choice
not
the
norm?
Operational
sex
ratios,
sex
roles
and
the
evolution
of
sexually
dimorphic
and
monomorphic
signalling.
Philosophical
Transactions
of
the
Royal
Society
of
London,
Series
B,
357,
319-330.
Kokko,
H.,
Brooks,
R.,
Jennions,
M.
D.
&
Morley,
J.
2003.
The
evolution
of
mate
choice
and
mating
biases.
Proceedings
of
the
Royal
Society
of
London,
Series
B,
270,
653-664.
Kwiatkowski,
M.
A.
&
Sullivan,
B.
K.
2002.
Geographic
variation
in
sexual
selection
among
populations
of
an
Iguanid
lizard,
Sauroma-
lus
obesus
(=ater).
Evolution,
56,
2039-3051.
Lebas,
N.
R.
&
Marshall,
N.
J.
2001.
No
evidence
of
female
choice
for
a
condition-dependent
trait
in
the
agamid
lizard,
Ctenophorus
ornatus.
Behaviour,
138,
965-980.
Le
Boeuf,
B.
J.
&
Mesnick,
S.
1991.
Sexual
behavior
of
male
north-
ern
elephant
seals:
1.
Lethal
injuries
to
adult
females.
Behaviour,
116,
143-162.
Lessells,
C.
M.
&
Boag,
P.
T.
1987.
Unrepeatable
repeatabilities:
a
common
mistake.
Auk,
104,
116-121.
Lopez,
P.,
Munoz,
A.
&
Martin,
J.
2002.
Symmetry,
male
domi-
nance
and
female
mate
preference
in
the
Iberian
rock
lizard,
Lac-
erta
monticola.
Behavioral
Ecology
and
Sociobiology,
52,
342-347.
Martin,
J.
&
Lopez,
P.
2000.
Chemoreception,
symmetry
and
mate
choice
in
lizards.
Proceedings
of
the
Royal
Society
of
London,
Series
B,
267,
1265-1269.
Moore,
A.
J.
1990.
The
inheritance
of
social
dominance,
mating
be-
haviour
and
attractiveness
to
mates
in
male
Nauphoeta
cinerea.
Animal
Behaviour,
39,
388-397.
Olsson,
M.
2001.
No
female
choice
in
the
Mallee
dragon
lizards,
Ctenophorus
fordi.
Evolutionary
Ecology,
15,
129-141.
Olsson,
M.
&
Madsen,
T.
1995.
Female
choice
on
male
quantitative
traits
in
lizards:
why
is
it
so
rare.
Behavioral
Ecology
and
Sociobiol-
ogy,
36,
179-184.
Olsson,
M.
&
Madsen,
T.
1998.
Sexual
selection
and
sperm
compe-
tition
in
reptiles.
In:
Sperm
Competition
and
Sexual
Selection
(Ed.
by
T.
R.
Birkhead
&
A.
P.
Moller),
pp.
503-564.
Cambridge:
Aca-
demic
Press.
Olsson,
M.,
Madsen,
T.,
Nordby,
J.,
Wapstra,
E.,
Ujvari,
B.
&
Witt-
sell,
H.
2003.
Major
histocompatability
complex
and
mate
choice
in
sand
lizards.
Proceedings
of
the
Royal
Society
of
London,
Series
B,
270
(Supplement),
S254-S256.
Ostlund,
S.
&
Ahnesjo,
I.
1998.
Female
fifteen-spined
sticklebacks
prefer
better
fathers.
Animal
Behaviour,
56,
1177-1183.
Pai,
A.
T.
&
Yan,
G.
Y.
2002.
Polyandry
produces
sexy
sons
at
the
cost
of
daughters
in
red
flour
beetles.
Proceedings
of
the
Royal
Society
of
London,
Series
B,
269,
361-368.
Pengilley,
R.
1972.
Systematic
relationships
and
ecology
of
some
lizards
from
southeastern
Australia.
Ph.D.
thesis,
Australian
National
University.
Pitcher,
T.
E.,
Neff,
B.
D.,
Rodd,
F.
H.
&
Rowe,
L.
2003.
Multiple
mat-
ing
and
sequential
mate
choice
in
guppies:
females
trade
up.
Pro-
ceedings
of
the
Royal
Society
of
London,
Series
B,
270,
1623-1629.
Prado,
C.
P.
A.
&
Haddad,
C.
F.
B.
2005.
Size-fecundity
relation-
ships
and
reproductive
investment
in
female
frogs
in
the
Pantanal,
south-western
Brazil.
Herpetological
Journal,
15,
181-189.
Qvarnstrom,
A.
&
Forsgren,
E.
1998.
Should
females
prefer
domi-
nant
males?
Trends
in
Ecology
&
Evolution,
13,
498-501.
R
Development
Core
Team.
2006.
R:
a
Language
and
Environment
for
Statistical
Computing.
Vienna,
Austria
(http://cran.r-project.org/).
Schafer,
M.
A.
&
Uhl,
G.
2005.
Sequential
mate
encounters:
female
but
not
male
body
size
influences
female
remating
behavior.
Behavioral
Ecology,
16,
461-466.
Shackleton,
M.,
Jennions,
M.
D.
&
Hunt,
J.
2005.
Fighting
success
and
attractiveness
as
predictors
of
male
mating
success
in
the
black
field
cricket,
Teleogryllus
commodus:
the
effectiveness
of
no-choice
tests.
Behavioral
Ecology
and
Sociobiology,
58,
1-8.
Sigmund,
W.
R.
1983.
Female
preference
for
Anolis
carolinensis
males
as
a
function
of
dewlap
colour
and
background
coloration.
Journal
of
Herpetology,
17,
137-143.
538
ANIMAL
BEHAVIOUR,
75,
2
Sih,
A.,
Bell,
A.
M.
&
Johnson,
J.
C.
2004.
Behavioural
syndromes:
an
ecological
and
evolutionary
overview.
Trends
in
Ecology
&
Evolution,
19,
372-378.
Simmons,
L.
W.,
Stockley,
P.,
Jackson,
R.
L.
&
Parker,
G.
A.
1996.
Sperm
competition
or
sperm
selection:
no
evidence
for
female
influence
over
paternity
in
yellow
dung
flies
Scatophaga
stercora-
ria.
Behavioral
Ecology
and
Sociobiology,
38,
199-206.
Smith,
J.
M.
&
Zucker,
N.
1997.
Do
female
tree
lizards,
Urosaurus
ornatus,
exhibit
mate
choice?
journal
of
Herpetology,
31,
179-186.
Stapley,
J.
2006.
Individual
variation
in
preferred
body
temperature
covaries
with
social
behaviours
and
colour
in
male
lizards.
Journal
of
Thermal
Biology,
31,
362-369.
Stapley,
J.
&
Keogh,
J.
S.
2005.
Behavioural
syndromes
influence
mating
systems:
floater
pairs
of
a
lizard
have
heavier
offspring.
Behavioral
Ecology,
16,
514-520.
Stapley,
J.
&
Keogh,
J.
S.
2006.
Experimental
and
molecular
evidence
that
body
size
and
ventral
color
interact
to
influence
male
reproductive
success
in
a
lizard.
Ethology
Ecology
and
Evolu-
tion,
18,
275-280.
Stuart-Fox,
D.
&
Whiting,
M.
J.
2005.
Male
dwarf
chameleons
assess
risk
of
courting
large,
aggressive
females.
Proceedings
of
the
Royal
Society
of
London,
Series
B,
1
(Supplement),
231-234.
Tokarz,
R.
R.
1995.
Mate
choice
in
lizards:
a
review.
Herpetological
Monographs,
9,
17-40.
Tokarz,
R.
R.
2002.
An
experimental
test
of
the
importance
of
the
dewlap
in
male
mating
success
in
the
lizard
Anolis
sagrei.
Herpeto-
logica,
58,
87-94.
Torr,
G.
A.
&
Shine,
R.
1994.
An
ethogram
for
the
small
lizard
Lamp-
ropholis
guichenoti.
Amphibia-Reptilia,
15,
21-34.
Vermette,
R.
&
Fairbairn,
D.
J.
2002.
How
well
do
mating
frequency
and
duration
predict
paternity
success
in
the
polygy-
nandrous
water
strider
Aquarius
remigis?
Evolution,
56,
1808-
1820.
Whittingham,
M.
J.,
Stephens,
P.
A.,
Bradbury,
R.
B.
&
Freckle-
ton,
R.
P.
2006.
Why
do
we
still
use
stepwise
modelling
in
ecology
and
behaviour?
Journal
of
Animal
Ecology,
75,
1182-1189.
Wong,
B.
M.
2004.
Superior
fighters
make
mediocre
fathers
in
the
Pacific
blue-eye
fish.
Animal
Behaviour,
67,
583-590.
Wong,
B.
B.
M.
&
Candolin,
U.
2005.
How
is
female
mate
choice
affected
by
male
competition?
Biological
Reviews,
80,
559-571.
Wong,
B.
M.
&
Jennions,
M.
D.
2003.
Costs
influence
male
mate
choice
in
a
freshwater
fish.
Proceedings
of
the
Royal
Society
of
London,
Series
B,
270
(Supplement),
S36—S38.